From d8f0dd163d1dd80f4d899ebe8ea4ae65f6bf8638 Mon Sep 17 00:00:00 2001 From: Gabe Smithline <63967324+gsmithline@users.noreply.github.com> Date: Tue, 30 Apr 2024 09:57:11 -0400 Subject: [PATCH 1/5] commit to checkout --- experiments.ipynb | 120142 ++++++++++++++++++++++++++++++++++++++++++- main.py | 2 +- 2 files changed, 120099 insertions(+), 45 deletions(-) diff --git a/experiments.ipynb b/experiments.ipynb index 96e9679..4972ff3 100644 --- a/experiments.ipynb +++ b/experiments.ipynb @@ -2,51 +2,14 @@ "cells": [ { "cell_type": "code", - "execution_count": 7, + "execution_count": 1, "metadata": {}, "outputs": [ { - "name": "stdout", + "name": "stderr", "output_type": "stream", "text": [ - "Collecting mtl\n", - " Downloading mtl-0.0.1.tar.gz (8.1 kB)\n", - " Preparing metadata (setup.py) ... \u001b[?25ldone\n", - "\u001b[?25hRequirement already satisfied: numpy in /Users/gabesmithline/miniconda3/lib/python3.10/site-packages (from mtl) (1.25.1)\n", - "Collecting argparse\n", - " Downloading argparse-1.4.0-py2.py3-none-any.whl (23 kB)\n", - "Building wheels for collected packages: mtl\n", - " Building wheel for mtl (setup.py) ... \u001b[?25ldone\n", - "\u001b[?25h Created wheel for mtl: filename=mtl-0.0.1-py3-none-any.whl size=8626 sha256=31679db8a56d5dde4f8b2c3819d835d13194a9a1b3ab9ac3c6aa9d7e13938cbd\n", - " Stored in directory: /Users/gabesmithline/Library/Caches/pip/wheels/0e/25/c4/6444d096f035b7e1d623cd136080e1a74c7f4bce951a3a21b8\n", - "Successfully built mtl\n", - "Installing collected packages: argparse, mtl\n", - "Successfully installed argparse-1.4.0 mtl-0.0.1\n" - ] - } - ], - "source": [ - "import sys\n", - "print(sys.executable)" - ] - }, - { - "cell_type": "code", - "execution_count": 8, - "metadata": {}, - "outputs": [ - { - "ename": "ImportError", - "evalue": "cannot import name 'readtps' from partially initialized module 'mtl' (most likely due to a circular import) (/Users/gabesmithline/miniconda3/lib/python3.10/site-packages/mtl/__init__.py)", - "output_type": "error", - "traceback": [ - "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m", - "\u001b[0;31mImportError\u001b[0m Traceback (most recent call last)", - "Cell \u001b[0;32mIn[8], line 6\u001b[0m\n\u001b[1;32m 4\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mrandom\u001b[39;00m\n\u001b[1;32m 5\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mpandas\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mpd\u001b[39;00m\n\u001b[0;32m----> 6\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mma_gym\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01menvs\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtraffic_junction\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m TrafficJunction \n\u001b[1;32m 7\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mcollections\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m defaultdict\n\u001b[1;32m 8\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mnashpy\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mnash\u001b[39;00m\n", - "File \u001b[0;32m~/Desktop/semester6/RL/project/ma-gym/ma_gym/envs/traffic_junction/__init__.py:1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtraffic_junction\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m TrafficJunction\n", - "File \u001b[0;32m~/Desktop/semester6/RL/project/ma-gym/ma_gym/envs/traffic_junction/traffic_junction.py:11\u001b[0m\n\u001b[1;32m 9\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mgym\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m spaces\n\u001b[1;32m 10\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mgym\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutils\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m seeding\n\u001b[0;32m---> 11\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mmtl\u001b[39;00m \n\u001b[1;32m 12\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutils\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01maction_space\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m MultiAgentActionSpace\n\u001b[1;32m 13\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutils\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mdraw\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m draw_grid, fill_cell, write_cell_text\n", - "File \u001b[0;32m~/miniconda3/lib/python3.10/site-packages/mtl/__init__.py:1\u001b[0m\n\u001b[0;32m----> 1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mmtl\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m readtps, ocontour, align\n\u001b[1;32m 3\u001b[0m \u001b[38;5;66;03m# calculation parts not ready for distribution yet thus not exported\u001b[39;00m\n", - "\u001b[0;31mImportError\u001b[0m: cannot import name 'readtps' from partially initialized module 'mtl' (most likely due to a circular import) (/Users/gabesmithline/miniconda3/lib/python3.10/site-packages/mtl/__init__.py)" + "Matplotlib created a temporary cache directory at /var/folders/fh/fwc37qhn04d8sxp65hwv1kxm0000gn/T/matplotlib-yjwp3tvd because the default path (/Users/gabesmithline/.matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n" ] } ], @@ -123,5380 +86,125471 @@ "text": [ "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: inf - 109ms/epoch - 2ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2569.0779 - 95ms/epoch - 2ms/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: inf - 119ms/epoch - 2ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2532.4629 - 111ms/epoch - 2ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: inf - 127ms/epoch - 2ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2540.1357 - 128ms/epoch - 2ms/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2564.6882 - 638us/epoch - 10us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2508.8125 - 676us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2844.0115 - 630us/epoch - 10us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2536.7380 - 561us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: inf - 147ms/epoch - 2ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1918 - 139ms/epoch - 2ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1336.1031 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3431.6448 - 603us/epoch - 10us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2608.4219 - 700us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3396.9351 - 747us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.0466 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3627.0986 - 768us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3754.2871 - 600us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2595.5183 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1629.4144 - 646us/epoch - 10us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2608.5198 - 739us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2559.9006 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.7803 - 584us/epoch - 9us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3502.4580 - 697us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.6543 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2964.1685 - 697us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1173.5975 - 591us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1738.0775 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3610.9917 - 676us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3022.4600 - 670us/epoch - 11us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2541.1553 - 781us/epoch - 13us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3318.0762 - 677us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3284.0339 - 602us/epoch - 10us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2088.2500 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1998.0671 - 625us/epoch - 10us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3117.5410 - 621us/epoch - 10us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1895.8032 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3850.7961 - 677us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2490.9741 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3077.7517 - 639us/epoch - 10us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2638.3818 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2660.7781 - 635us/epoch - 10us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3072.9041 - 776us/epoch - 13us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2249.6199 - 667us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.5048 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2642.8306 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3797.6006 - 784us/epoch - 13us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2821.7910 - 694us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2932.0625 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2753.1692 - 675us/epoch - 11us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.8900 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2313.2498 - 580us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3263.5166 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2380.8372 - 648us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2787.0159 - 791us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3911.7996 - 585us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3351.1040 - 661us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2488.5012 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2537.0168 - 612us/epoch - 10us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2828.7993 - 697us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2529.4326 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3416.2368 - 630us/epoch - 10us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2783.1772 - 731us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3393.5042 - 650us/epoch - 10us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3616.9033 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1439.4982 - 710us/epoch - 11us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2509.0349 - 651us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3578.3057 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2446.6648 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2595.5789 - 710us/epoch - 11us/sample\n", "Generation 2/300\n", "Solving for Nash Equilibrium in Generation 2/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.4628 - 757us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2470.6238 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2519.2363 - 786us/epoch - 13us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3019.5149 - 716us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2491.9749 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3688.4624 - 606us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2538.4207 - 720us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3044.7808 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2568.1826 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.2016 - 593us/epoch - 10us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2495.8093 - 651us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2520.9050 - 574us/epoch - 9us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3007.5017 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3624.3040 - 551us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2243.7122 - 624us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3727.5229 - 618us/epoch - 10us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3166.2935 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.7164 - 709us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2800.0806 - 662us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2505.8030 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2616.0405 - 679us/epoch - 11us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2515.1406 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2497.7563 - 620us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2754.2720 - 648us/epoch - 10us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2686.9392 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.5239 - 769us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2522.2812 - 750us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2785.7673 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3159.2405 - 674us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2113.0369 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3309.3545 - 542us/epoch - 9us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.6515 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3026.0376 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2538.0554 - 643us/epoch - 10us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3217.6975 - 618us/epoch - 10us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3681.9351 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3696.0085 - 681us/epoch - 11us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3857.9556 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2535.7878 - 618us/epoch - 10us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3033.9043 - 746us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2480.6975 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3616.8730 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2496.3127 - 664us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2532.3960 - 679us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2899.3359 - 684us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2498.7561 - 618us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1367.7980 - 610us/epoch - 10us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2476.5159 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.9336 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2521.2119 - 651us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2540.1570 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3574.3191 - 727us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2526.6824 - 606us/epoch - 10us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3114.0020 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2330.4331 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2918.3049 - 671us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3418.5684 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2760.8428 - 789us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2460.7632 - 777us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2399.3645 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3291.5227 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3546.9153 - 822us/epoch - 13us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3795.8440 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2433.6028 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3273.1816 - 827us/epoch - 13us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2493.9182 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1960.0703 - 742us/epoch - 12us/sample\n", "Episode 42/50\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2464.1658 - 581us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2420.0422 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2600.7593 - 681us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2447.9277 - 657us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2495.1167 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2631.4233 - 677us/epoch - 11us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1866.7495 - 794us/epoch - 13us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1316.6486 - 669us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2569.7881 - 808us/epoch - 13us/sample\n", "Generation 3/300\n", "Solving for Nash Equilibrium in Generation 3/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2898.4888 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2933.4998 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2559.3940 - 607us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3609.7009 - 617us/epoch - 10us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2404.7588 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2449.9458 - 609us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2556.6875 - 639us/epoch - 10us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2575.6841 - 554us/epoch - 9us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2568.5757 - 605us/epoch - 10us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3736.9658 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2609.1665 - 633us/epoch - 10us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2904.8271 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2247.6484 - 702us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2470.9092 - 737us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2618.2241 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2192.5872 - 806us/epoch - 13us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 12ms/epoch - 196us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2446.2849 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 184.8915 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3168.5833 - 728us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2759.7307 - 588us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2532.2783 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2735.8738 - 670us/epoch - 11us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3249.3606 - 722us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2898.7158 - 705us/epoch - 11us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3600.1641 - 665us/epoch - 11us/sample\n", "Episode 18/50\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2400.0298 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1439.0319 - 570us/epoch - 9us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2000.9689 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.6455 - 589us/epoch - 9us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3119.1851 - 706us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2401.1265 - 579us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2436.4670 - 715us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3908.0676 - 521us/epoch - 8us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2446.7031 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1452.4510 - 700us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2331.4673 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2454.9631 - 664us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2456.1252 - 686us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1623.3060 - 630us/epoch - 10us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2373.3125 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2468.1604 - 622us/epoch - 10us/sample\n", "Episode 31/50\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1996.4701 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2457.4502 - 613us/epoch - 10us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2448.9502 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2377.0437 - 726us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2459.4504 - 669us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2412.2612 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 800.5355 - 653us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2449.3040 - 624us/epoch - 10us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2444.9446 - 664us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2792.7327 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2479.1541 - 737us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3754.3948 - 621us/epoch - 10us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2447.1470 - 679us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2679.1011 - 747us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2214.3083 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2684.9487 - 576us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3220.1970 - 646us/epoch - 10us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3745.5269 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2411.6279 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2452.5239 - 600us/epoch - 10us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2077.0903 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2955.6790 - 655us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2438.0796 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2439.2095 - 714us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2394.9783 - 619us/epoch - 10us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2420.9360 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2414.0767 - 682us/epoch - 11us/sample\n", "Generation 4/300\n", "Solving for Nash Equilibrium in Generation 4/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3717.3901 - 644us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.2871 - 634us/epoch - 10us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2469.5906 - 639us/epoch - 10us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2933.8621 - 613us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2280.3936 - 666us/epoch - 11us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2729.8774 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3614.9480 - 695us/epoch - 11us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2943.7051 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2310.6699 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2522.1785 - 2ms/epoch - 31us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2540.1462 - 661us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 172.7332 - 744us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2290.3093 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3024.3374 - 666us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2417.1177 - 757us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2326.9548 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2485.6130 - 671us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3833.9607 - 706us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2027.9860 - 600us/epoch - 10us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2400.9248 - 701us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2420.5967 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2322.7458 - 607us/epoch - 10us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1781.1594 - 652us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2419.1448 - 800us/epoch - 13us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1247.4805 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2489.3247 - 677us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2312.6584 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2568.2085 - 671us/epoch - 11us/sample\n", "Episode 23/50\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2876.0190 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2506.5867 - 648us/epoch - 10us/sample\n", "Episode 26/50\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2450.4873 - 631us/epoch - 10us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1625.9358 - 747us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3313.2727 - 693us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.8114 - 814us/epoch - 13us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3215.8503 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2282.4583 - 565us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2310.4380 - 816us/epoch - 13us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2482.1165 - 723us/epoch - 12us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2206.9504 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2385.0581 - 573us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2490.7324 - 694us/epoch - 11us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2313.8955 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2380.4595 - 705us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2839.9746 - 622us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2436.3137 - 603us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2343.1875 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2418.9070 - 693us/epoch - 11us/sample\n", "Episode 41/50\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2386.5271 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2324.8823 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2436.2332 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1723.8506 - 631us/epoch - 10us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2349.0964 - 568us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1331.7965 - 717us/epoch - 12us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2484.0757 - 594us/epoch - 10us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2404.8818 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3337.8506 - 640us/epoch - 10us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2247.5376 - 612us/epoch - 10us/sample\n", "Generation 5/300\n", "Solving for Nash Equilibrium in Generation 5/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2998.8015 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1622.8932 - 893us/epoch - 14us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2320.9812 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 158.2192 - 613us/epoch - 10us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3106.6814 - 535us/epoch - 9us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2317.9839 - 540us/epoch - 9us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2363.7021 - 575us/epoch - 9us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2356.4646 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2590.2593 - 565us/epoch - 9us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3465.1841 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3076.0125 - 730us/epoch - 12us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2419.9622 - 600us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2278.3730 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2241.9109 - 1ms/epoch - 17us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3486.2456 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 174.1858 - 712us/epoch - 11us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2344.8486 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2345.2798 - 637us/epoch - 10us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2323.5981 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2570.3115 - 623us/epoch - 10us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2178.8247 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2364.9844 - 702us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.0714 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3066.2415 - 615us/epoch - 10us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2875.4426 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1421.3352 - 713us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3661.0242 - 769us/epoch - 12us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2270.3406 - 559us/epoch - 9us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3552.9385 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2396.5945 - 658us/epoch - 11us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2500.4490 - 697us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2237.2698 - 544us/epoch - 9us/sample\n", "Episode 27/50\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 190.1410 - 561us/epoch - 9us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2243.1426 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2713.7524 - 640us/epoch - 10us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 973.0861 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2393.7234 - 603us/epoch - 10us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3275.0283 - 539us/epoch - 9us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.8300 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1681.4667 - 571us/epoch - 9us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3138.4080 - 641us/epoch - 10us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2151.2029 - 686us/epoch - 11us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2163.7253 - 618us/epoch - 10us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2320.3838 - 603us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2394.1360 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.2371 - 683us/epoch - 11us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2260.0388 - 629us/epoch - 10us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2222.1562 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2254.7417 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3489.0798 - 701us/epoch - 11us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2432.0063 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2498.5076 - 699us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2216.6052 - 580us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2285.0913 - 740us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2786.8293 - 638us/epoch - 10us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2253.9194 - 590us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2729.1792 - 660us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1035.3180 - 695us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2303.6335 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2491.1426 - 721us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3358.8533 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2655.9998 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2295.7776 - 718us/epoch - 12us/sample\n", "Generation 6/300\n", "Solving for Nash Equilibrium in Generation 6/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3317.1028 - 818us/epoch - 13us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2180.3472 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2921.7329 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2321.9780 - 667us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3664.0156 - 566us/epoch - 9us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2130.2427 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2819.5679 - 594us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3285.8760 - 647us/epoch - 10us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2335.6682 - 567us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1267.3511 - 638us/epoch - 10us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2935.6497 - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1818.1240 - 650us/epoch - 10us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2248.1433 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2232.7583 - 672us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2109.0596 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2589.4160 - 849us/epoch - 14us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2465.7341 - 783us/epoch - 13us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2950.1355 - 798us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2173.7446 - 830us/epoch - 13us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123.5444 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2006.1877 - 727us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2222.8889 - 1ms/epoch - 18us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2568.7900 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2349.6108 - 614us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3735.0720 - 575us/epoch - 9us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1216.6033 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1547.0938 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3264.2249 - 587us/epoch - 9us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2132.8528 - 712us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2445.8472 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2166.3708 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2338.6533 - 633us/epoch - 10us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2357.0415 - 721us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2888.6880 - 692us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1136.6003 - 713us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1990.4104 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1146.1591 - 607us/epoch - 10us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2223.6057 - 784us/epoch - 13us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3226.7715 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2303.8347 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1983.9683 - 629us/epoch - 10us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2120.9412 - 649us/epoch - 10us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1583.1246 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2387.6106 - 750us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2348.4426 - 583us/epoch - 9us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.4306 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1384.5736 - 907us/epoch - 15us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2106.4531 - 684us/epoch - 11us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2740.5940 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2032.5532 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2079.9836 - 669us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2215.8396 - 19ms/epoch - 313us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.4141 - 749us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2293.0481 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2765.0222 - 691us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2307.2710 - 793us/epoch - 13us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1389.2594 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1509.3656 - 730us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2081.4500 - 798us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2164.9260 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2426.0835 - 767us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1728.0859 - 598us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2803.1343 - 760us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2064.1799 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3606.3284 - 765us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2123.3457 - 711us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1826.9340 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2659.7021 - 702us/epoch - 11us/sample\n", "Episode 50/50\n", "Generation 7/300\n", "Solving for Nash Equilibrium in Generation 7/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2016.4447 - 809us/epoch - 13us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2010.6832 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2215.2957 - 854us/epoch - 14us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3599.6797 - 857us/epoch - 14us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2034.8990 - 652us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2146.4048 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1347.7422 - 840us/epoch - 14us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1948.8575 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2231.9084 - 784us/epoch - 13us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1936.0421 - 614us/epoch - 10us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2190.7439 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2267.4497 - 721us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2202.4363 - 721us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2389.8831 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2427.9333 - 714us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 861.8051 - 774us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1965.6436 - 639us/epoch - 10us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2126.1074 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1875.5002 - 9ms/epoch - 141us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2038.7483 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1570.3910 - 652us/epoch - 11us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6768 - 782us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3103.7578 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1722.1884 - 805us/epoch - 13us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2113.1074 - 644us/epoch - 10us/sample\n", "Episode 21/50\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1714.1952 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.0025 - 709us/epoch - 11us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2612.7244 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1814.5153 - 812us/epoch - 13us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2193.0156 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2091.2092 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1990.5004 - 1ms/epoch - 18us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2709.3247 - 845us/epoch - 14us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 200.2999 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1684.5149 - 632us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2104.7258 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1531.3281 - 713us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2271.8191 - 687us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2669.2922 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2095.9282 - 672us/epoch - 11us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1811.4756 - 710us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2074.9963 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2500.8958 - 886us/epoch - 14us/sample\n", "Episode 36/50\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2067.0759 - 2ms/epoch - 28us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1256.5189 - 643us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2064.3088 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2053.7458 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1342.9572 - 607us/epoch - 10us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2177.4617 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2491.3547 - 617us/epoch - 10us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2257.1580 - 632us/epoch - 10us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2468.4065 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2846.4985 - 690us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2017.0245 - 816us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3086.3040 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2695.3298 - 608us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2748.8691 - 650us/epoch - 10us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2689.9172 - 624us/epoch - 10us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1668.4871 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.6132 - 683us/epoch - 11us/sample\n", "Episode 50/50\n", "Generation 8/300\n", "Solving for Nash Equilibrium in Generation 8/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1877.2645 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1154.0856 - 911us/epoch - 15us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2200.8167 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2805.6345 - 748us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1641.5806 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2755.2847 - 950us/epoch - 15us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2853.1016 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 141.3896 - 764us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1615.3602 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2026.3420 - 810us/epoch - 13us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1596.4679 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1813.7853 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1327.0469 - 638us/epoch - 10us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2572.8569 - 677us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1714.1885 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2083.7815 - 705us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2080.5559 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2427.2380 - 633us/epoch - 10us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1882.6489 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1207.5758 - 799us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2269.7495 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2989.3105 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1350.0051 - 715us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1881.5299 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2926.5171 - 730us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2019.3571 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2961.0120 - 752us/epoch - 12us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1874.9927 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1847.6949 - 710us/epoch - 11us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2275.9595 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1986.4886 - 685us/epoch - 11us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1791.4541 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1810.2899 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1663.3748 - 687us/epoch - 11us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.1326 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1883.0367 - 713us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1783.9437 - 4ms/epoch - 71us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 807.6732 - 774us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1526.8342 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.7853 - 640us/epoch - 10us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.8311 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2548.3657 - 631us/epoch - 10us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1590.0807 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2006.9464 - 18ms/epoch - 292us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1955.1022 - 1ms/epoch - 22us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2397.6848 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1777.2877 - 996us/epoch - 16us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1888.6241 - 957us/epoch - 15us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1534.8053 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2522.9780 - 737us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2878.6418 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1261.1814 - 672us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2534.9045 - 751us/epoch - 12us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2460.6455 - 705us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1791.9156 - 832us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1516.1909 - 770us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1720.4570 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2052.0579 - 807us/epoch - 13us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1424.7843 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2708.7649 - 801us/epoch - 13us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 63.8494 - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1697.0962 - 659us/epoch - 11us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1507.4375 - 746us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1740.7936 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 930.8878 - 829us/epoch - 13us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1732.3575 - 672us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2005.8407 - 764us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1755.9482 - 671us/epoch - 11us/sample\n", "Generation 9/300\n", "Solving for Nash Equilibrium in Generation 9/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1444.1110 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1663.7880 - 744us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1942.2819 - 773us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1735.1694 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1337.1058 - 872us/epoch - 14us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1651.9548 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1925.4703 - 647us/epoch - 10us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1844.7233 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1346.1940 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.5536 - 715us/epoch - 12us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1752.9167 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2431.1912 - 614us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1326.1410 - 728us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2194.0579 - 722us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2709.4722 - 756us/epoch - 12us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1466.7798 - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2029.2058 - 771us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1714.5985 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.4322 - 659us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1863.5374 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1332.1559 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2122.3406 - 612us/epoch - 10us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1682.1655 - 619us/epoch - 10us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1588.6060 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1528.4603 - 628us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1694.8994 - 993us/epoch - 16us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1683.3605 - 732us/epoch - 12us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2763.7834 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1399.1837 - 800us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1461.8864 - 675us/epoch - 11us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1386.3567 - 635us/epoch - 10us/sample\n", "Episode 26/50\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53.5462 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1682.0435 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1364.7755 - 744us/epoch - 12us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1626.9119 - 620us/epoch - 10us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50.0700 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1581.2374 - 633us/epoch - 10us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1022.2961 - 696us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1722.4321 - 715us/epoch - 12us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1670.1494 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1558.8938 - 638us/epoch - 10us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1660.5696 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1389.3364 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1101.5093 - 749us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1301.4297 - 873us/epoch - 14us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2293.2373 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3102 - 713us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1704.4875 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1556.1294 - 744us/epoch - 12us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1822.7128 - 700us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1569.1796 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1100.0308 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2370.8486 - 629us/epoch - 10us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1293.1383 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1995.9165 - 676us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1404.7155 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2282.9985 - 734us/epoch - 12us/sample\n", "Episode 45/50\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1225.1578 - 684us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1381.0773 - 559us/epoch - 9us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1392.5398 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.2613 - 913us/epoch - 15us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1144.5631 - 3ms/epoch - 44us/sample\n", "Generation 10/300\n", "Solving for Nash Equilibrium in Generation 10/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1488.1641 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1528.7758 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1338.9448 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1508.1129 - 710us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1760.7423 - 628us/epoch - 10us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1354.1676 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1038.9125 - 799us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2427.9053 - 890us/epoch - 14us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1670.0974 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1180.4232 - 771us/epoch - 12us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1328.1674 - 657us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1074.8451 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1612.6948 - 787us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1087.4449 - 795us/epoch - 13us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1506.6312 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1104.9314 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1398.2264 - 978us/epoch - 16us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1682.5522 - 606us/epoch - 10us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1168.7439 - 721us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1377.5088 - 838us/epoch - 14us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1867.9667 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1273.8805 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.5730 - 820us/epoch - 13us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1705.0156 - 643us/epoch - 10us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1208.7131 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1519.6208 - 644us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1528.4696 - 703us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1333.4364 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.8251 - 666us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1438.5372 - 713us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1847.1625 - 874us/epoch - 14us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1415.1923 - 735us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1051.5508 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98.7717 - 800us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1226.0920 - 728us/epoch - 12us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1204.6523 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 919.3829 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1816.7126 - 680us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1301.1599 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1187.1975 - 596us/epoch - 10us/sample\n", "Episode 28/50\n", "Episode 29/50\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1128.3881 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1468.2445 - 782us/epoch - 13us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1400.1134 - 5ms/epoch - 75us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2020.6725 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1155.7479 - 702us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1381.5498 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.0231 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 873.1114 - 672us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 989.5721 - 659us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1329.0399 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1338.3203 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1325.0127 - 620us/epoch - 10us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1360.5985 - 678us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.8474 - 721us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2116.0535 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1603.0239 - 617us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1081.7985 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1208.7010 - 808us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 945.1712 - 707us/epoch - 11us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1320.2925 - 691us/epoch - 11us/sample\n", "Episode 44/50\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 830.2342 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.2838 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1204.0608 - 807us/epoch - 13us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1798.4790 - 724us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52.3240 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1004.1401 - 620us/epoch - 10us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 937.0167 - 810us/epoch - 13us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1547.5134 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 906.9800 - 681us/epoch - 11us/sample\n", "Generation 11/300\n", "Solving for Nash Equilibrium in Generation 11/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1318.5568 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1860.8071 - 722us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1233.1376 - 690us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1392.7327 - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1226.2312 - 975us/epoch - 16us/sample\n", "Episode 4/50\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 953.5366 - 724us/epoch - 12us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1002.9103 - 604us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 957.4492 - 722us/epoch - 12us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 985.8682 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55.1374 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 992.5709 - 619us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1180.2684 - 663us/epoch - 11us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1190.3129 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 922.4528 - 739us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1153.5708 - 589us/epoch - 9us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 912.5236 - 625us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1158.0322 - 739us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1150.0515 - 692us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1139.2554 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1173.3680 - 637us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1141.0634 - 694us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1442.7692 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.9594 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.2197 - 728us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74.3508 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1105.6030 - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1388.8923 - 774us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1189.4850 - 882us/epoch - 14us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1705.7314 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1268.5037 - 729us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.7512 - 887us/epoch - 14us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 888.8225 - 817us/epoch - 13us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 893.4948 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.5130 - 2ms/epoch - 37us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 973.7333 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1105.0291 - 796us/epoch - 13us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1401.6102 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1178.5151 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.5281 - 697us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 903.5540 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1235.9263 - 704us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.8459 - 864us/epoch - 14us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1458.4257 - 694us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 905.0035 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1689.9581 - 754us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 932.1085 - 787us/epoch - 13us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1649.8711 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1209.4811 - 760us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 171.4737 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.6570 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 970.0877 - 711us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1142.9095 - 749us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1022.4727 - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1472.3466 - 851us/epoch - 14us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1373.7610 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1055.8000 - 2ms/epoch - 25us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 993.2573 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1267.1067 - 779us/epoch - 13us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.0685 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.9687 - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1418.4464 - 615us/epoch - 10us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.3925 - 2ms/epoch - 26us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.7791 - 780us/epoch - 13us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.3624 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.1479 - 1ms/epoch - 20us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 901.8265 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 895.2854 - 741us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.7885 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 948.9839 - 1ms/epoch - 17us/sample\n", "Episode 49/50\n", "Episode 50/50\n", "Generation 12/300\n", "Solving for Nash Equilibrium in Generation 12/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428.3302 - 1ms/epoch - 18us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.7402 - 1ms/epoch - 20us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90.8661 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1032.3804 - 783us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 867.1731 - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 886.0140 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 864.0155 - 744us/epoch - 12us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 921.4827 - 796us/epoch - 13us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 938.4952 - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1335.7720 - 940us/epoch - 15us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.9528 - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1040.1392 - 758us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 883.7370 - 708us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1549.1786 - 714us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.1589 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.4707 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 951.7719 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.9697 - 876us/epoch - 14us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 815.9955 - 836us/epoch - 13us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1128.2080 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 947.7083 - 638us/epoch - 10us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.8585 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1225.3082 - 637us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 928.1237 - 855us/epoch - 14us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 835.1740 - 747us/epoch - 12us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.3677 - 701us/epoch - 11us/sample\n", "Episode 18/50\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 831.5716 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 851.7261 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.6685 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.7067 - 776us/epoch - 13us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 991.6804 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 836.5216 - 787us/epoch - 13us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 967.6738 - 870us/epoch - 14us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1235.0885 - 701us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.6028 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.2877 - 688us/epoch - 11us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.5589 - 678us/epoch - 11us/sample\n", "Episode 29/50\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.6566 - 669us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55.1775 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.5353 - 757us/epoch - 12us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 968.1429 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.0449 - 682us/epoch - 11us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1145.8915 - 646us/epoch - 10us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 948.8263 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.0732 - 731us/epoch - 12us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 799.0230 - 709us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106.4936 - 663us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1039.8558 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 847.2338 - 714us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 203.5296 - 595us/epoch - 10us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 945.7206 - 1ms/epoch - 16us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 976.4711 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 786.1976 - 656us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.6149 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1257.8988 - 769us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1253.6268 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.5565 - 703us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1093.7847 - 812us/epoch - 13us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.1720 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.4515 - 661us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1093.7809 - 721us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1136.2817 - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 794.4174 - 735us/epoch - 12us/sample\n", "Generation 13/300\n", "Solving for Nash Equilibrium in Generation 13/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 853.9046 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 174.4582 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.9196 - 792us/epoch - 13us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.7096 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.6184 - 888us/epoch - 14us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 846.6786 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.4061 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 811.6002 - 1ms/epoch - 20us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 954.0018 - 1ms/epoch - 17us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1364.7836 - 790us/epoch - 13us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1179.6210 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.6406 - 2ms/epoch - 26us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1101.6945 - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.3058 - 859us/epoch - 14us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2385 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1167.0771 - 678us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.5408 - 661us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 805.0297 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.0640 - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 950.7153 - 859us/epoch - 14us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 917.2291 - 727us/epoch - 12us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 955.2093 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1265.5870 - 843us/epoch - 14us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 971.7255 - 967us/epoch - 16us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1089.9340 - 818us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 169.2400 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.7758 - 807us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.8978 - 1ms/epoch - 23us/sample\n", "Episode 20/50\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428.0284 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.7991 - 786us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 966.1448 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1759 - 725us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.6234 - 763us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1012.2357 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812.8607 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1096.2327 - 770us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 969.6362 - 869us/epoch - 14us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 843.8448 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.2449 - 756us/epoch - 12us/sample\n", "Episode 28/50\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.8160 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 224.5738 - 618us/epoch - 10us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.6476 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 933.5377 - 779us/epoch - 13us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.3189 - 811us/epoch - 13us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.6441 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1100.3959 - 700us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.1986 - 835us/epoch - 13us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.8201 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123.9901 - 849us/epoch - 14us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.2048 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.0818 - 764us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 972.1793 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 892.5856 - 1ms/epoch - 16us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.5199 - 713us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 954.8962 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.6143 - 923us/epoch - 15us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1127.6730 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.4322 - 784us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.7900 - 790us/epoch - 13us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.6039 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.8927 - 720us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 917.8094 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.4471 - 882us/epoch - 14us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.8177 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.8109 - 1ms/epoch - 20us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 891.9839 - 1ms/epoch - 20us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 848.8757 - 848us/epoch - 14us/sample\n", "Episode 50/50\n", "Generation 14/300\n", "Solving for Nash Equilibrium in Generation 14/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.3411 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.4752 - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.1339 - 812us/epoch - 13us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.3210 - 879us/epoch - 14us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.4917 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.2217 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.1846 - 737us/epoch - 12us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.7927 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 870.6910 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1845 - 720us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.4570 - 799us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.1933 - 770us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 912.4351 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 932.8345 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.0621 - 708us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.7205 - 1ms/epoch - 18us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.3939 - 2ms/epoch - 25us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.8242 - 709us/epoch - 11us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 185.9349 - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 854.2513 - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.3107 - 747us/epoch - 12us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.7693 - 633us/epoch - 10us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 947.5613 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1060.8981 - 693us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.2375 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 810.5726 - 1ms/epoch - 19us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 845.1282 - 644us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.8303 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.9175 - 2ms/epoch - 35us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.3726 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.0598 - 1ms/epoch - 17us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2051 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.0992 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 986.5731 - 839us/epoch - 14us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.9371 - 643us/epoch - 10us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 161us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.0179 - 1ms/epoch - 19us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1153.3967 - 1ms/epoch - 17us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 872.1046 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.1436 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.3723 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.6965 - 850us/epoch - 14us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 953.3495 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.2240 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.4199 - 762us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812.4578 - 684us/epoch - 11us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3250 - 715us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.1843 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.0791 - 701us/epoch - 11us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6446 - 760us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.6976 - 1ms/epoch - 24us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.1050 - 755us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.0919 - 931us/epoch - 15us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.7863 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.3780 - 719us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3162 - 772us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.3642 - 795us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812.1896 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.4042 - 687us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.5727 - 961us/epoch - 16us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.3265 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 820.4691 - 778us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 226.0774 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.1102 - 720us/epoch - 12us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.1723 - 599us/epoch - 10us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.5289 - 750us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.5251 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.0891 - 834us/epoch - 13us/sample\n", "Generation 15/300\n", "Solving for Nash Equilibrium in Generation 15/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.9008 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.5798 - 761us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 857.3987 - 719us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 898.1478 - 800us/epoch - 13us/sample\n", "Episode 4/50\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.1155 - 678us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7537 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.2739 - 2ms/epoch - 27us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.7801 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.0445 - 797us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.6730 - 674us/epoch - 11us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 800.8383 - 575us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.6452 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2529 - 707us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.7823 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 889.4247 - 672us/epoch - 11us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.7322 - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 853.3275 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 901.5270 - 810us/epoch - 13us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.5952 - 668us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.5703 - 912us/epoch - 15us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.8745 - 755us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.4819 - 1ms/epoch - 23us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.1667 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.2323 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 911.8853 - 1ms/epoch - 16us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406.5918 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.4520 - 797us/epoch - 13us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 909.2622 - 1ms/epoch - 18us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.6273 - 3ms/epoch - 50us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 946.4532 - 2ms/epoch - 27us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 815.8097 - 1ms/epoch - 19us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 853.3657 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.6422 - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.1100 - 935us/epoch - 15us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.2420 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 838.3089 - 1ms/epoch - 17us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.9288 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.5436 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.8225 - 727us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.5319 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.1857 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 995.9291 - 760us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.7185 - 688us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.9861 - 662us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 925.6586 - 669us/epoch - 11us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.4747 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.7272 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.0706 - 943us/epoch - 15us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.1219 - 700us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 848.6343 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1200 - 722us/epoch - 12us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 794.3254 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.7700 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.6497 - 633us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.3997 - 613us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.6119 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.0551 - 694us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.4708 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3420 - 655us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 923.1844 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.6030 - 616us/epoch - 10us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.6404 - 3ms/epoch - 54us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 877.9717 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.0537 - 709us/epoch - 11us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.2234 - 805us/epoch - 13us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 876.2302 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9004 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.0860 - 600us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.1210 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.6880 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 804.4704 - 716us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7249 - 1ms/epoch - 21us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 791.1951 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.8009 - 643us/epoch - 10us/sample\n", "Generation 16/300\n", "Solving for Nash Equilibrium in Generation 16/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 814.2706 - 605us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.5635 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.8528 - 822us/epoch - 13us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.9088 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 937.0598 - 678us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.7883 - 986us/epoch - 16us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 853.1812 - 648us/epoch - 10us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.6848 - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.6217 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.4656 - 661us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 928.2423 - 822us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7712 - 647us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.1421 - 787us/epoch - 13us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.3487 - 949us/epoch - 15us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.7564 - 792us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.1857 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.5110 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.0343 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1043.4404 - 733us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.2720 - 703us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 789.0522 - 808us/epoch - 13us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 810.0982 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.2458 - 667us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.1512 - 759us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.5697 - 759us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.8817 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.0806 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.4230 - 643us/epoch - 10us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 816.0189 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.2246 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.1937 - 613us/epoch - 10us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.8142 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.7120 - 2ms/epoch - 32us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.5208 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.3719 - 766us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.9064 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.7532 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.6676 - 783us/epoch - 13us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.2375 - 671us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.6639 - 673us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.3994 - 674us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.5339 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 819.1837 - 683us/epoch - 11us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.7051 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406.5911 - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.1678 - 685us/epoch - 11us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.7419 - 818us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.3657 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.9656 - 3ms/epoch - 42us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 933.1473 - 609us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.5738 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.1220 - 1ms/epoch - 20us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.1565 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.9066 - 790us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.4918 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.9516 - 643us/epoch - 10us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.4099 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.4265 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.7194 - 660us/epoch - 11us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.6857 - 813us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 833.9788 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.3291 - 831us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.9753 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.7504 - 2ms/epoch - 25us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 876.1246 - 994us/epoch - 16us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.6367 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.2751 - 607us/epoch - 10us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.6760 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.2287 - 1ms/epoch - 23us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.9285 - 699us/epoch - 11us/sample\n", "Generation 17/300\n", "Solving for Nash Equilibrium in Generation 17/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 891.6695 - 705us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.3749 - 621us/epoch - 10us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.9496 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.9087 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.6566 - 753us/epoch - 12us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.1110 - 1ms/epoch - 24us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.4335 - 791us/epoch - 13us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1043.2926 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.9628 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.1297 - 870us/epoch - 14us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 869.7555 - 857us/epoch - 14us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.1505 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.6598 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.9904 - 743us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7895 - 772us/epoch - 12us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.8676 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.6780 - 720us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 863.3710 - 628us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.4832 - 744us/epoch - 12us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.1591 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.8286 - 772us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.5457 - 789us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.1548 - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.1126 - 731us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.2132 - 717us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 790.2787 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.2305 - 686us/epoch - 11us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.6166 - 756us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.8098 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.7268 - 589us/epoch - 10us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.0862 - 2ms/epoch - 26us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 811.3502 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 823.9801 - 743us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.6853 - 669us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.0250 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.5223 - 620us/epoch - 10us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.2405 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.5421 - 686us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 933.9442 - 584us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 962.3926 - 742us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.7352 - 747us/epoch - 12us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6715 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.7740 - 734us/epoch - 12us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 847.4261 - 1ms/epoch - 23us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.6030 - 744us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.7930 - 1ms/epoch - 18us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.9839 - 811us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.5621 - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.0846 - 714us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.1384 - 1ms/epoch - 16us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.4783 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.8470 - 733us/epoch - 12us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2167 - 843us/epoch - 14us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.5695 - 698us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.7079 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.1876 - 820us/epoch - 13us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.0816 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.6551 - 695us/epoch - 11us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 833.0254 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.4555 - 653us/epoch - 11us/sample\n", "Generation 18/300\n", "Solving for Nash Equilibrium in Generation 18/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.2392 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.5006 - 839us/epoch - 14us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.9071 - 712us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.2542 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 981.8884 - 639us/epoch - 10us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.4191 - 625us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.5971 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.5138 - 694us/epoch - 11us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.7889 - 727us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.0965 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 905.7379 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.7254 - 717us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.8391 - 591us/epoch - 10us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.7047 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.0916 - 671us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.8695 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.5110 - 814us/epoch - 13us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.6777 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.6774 - 739us/epoch - 12us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.0491 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.5993 - 683us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.2590 - 691us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.8383 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.4968 - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 871.8938 - 941us/epoch - 15us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.1621 - 688us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6171 - 718us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6286 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.7130 - 630us/epoch - 10us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.0738 - 682us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.8935 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.4130 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.8080 - 632us/epoch - 10us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 836.3929 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.1871 - 644us/epoch - 10us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 833.2205 - 692us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8123 - 797us/epoch - 13us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 999.4918 - 826us/epoch - 13us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.3016 - 3ms/epoch - 48us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1681 - 667us/epoch - 11us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.3105 - 716us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.0224 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.7185 - 663us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.7006 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.2519 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.5412 - 697us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.8839 - 680us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.8807 - 687us/epoch - 11us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.3591 - 655us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.8650 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.2792 - 674us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.0633 - 697us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.0098 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.1376 - 567us/epoch - 9us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 823.1695 - 717us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.0961 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.5357 - 1ms/epoch - 17us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.7893 - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.7953 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.1794 - 784us/epoch - 13us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.0813 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.0488 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 824.5212 - 716us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.1879 - 872us/epoch - 14us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5359 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.6367 - 799us/epoch - 13us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.4841 - 2ms/epoch - 31us/sample\n", "Generation 19/300\n", "Solving for Nash Equilibrium in Generation 19/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.5304 - 734us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.4978 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.5125 - 665us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7254 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 977.6703 - 638us/epoch - 10us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6091 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.6678 - 588us/epoch - 9us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.0518 - 666us/epoch - 11us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.5087 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.7660 - 669us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.1932 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.1064 - 1ms/epoch - 16us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 799.1137 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.2906 - 717us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.9884 - 763us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1414 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.6538 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.8611 - 900us/epoch - 15us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 866.3765 - 796us/epoch - 13us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.3727 - 872us/epoch - 14us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.9512 - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.0286 - 642us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.5406 - 627us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.4628 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.9200 - 2ms/epoch - 25us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7250 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.9634 - 861us/epoch - 14us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.6052 - 764us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.8569 - 780us/epoch - 13us/sample\n", "Episode 19/50\n", "Episode 20/50\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.5679 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.8799 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.9474 - 804us/epoch - 13us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.0856 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.8833 - 749us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 826.8857 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.8868 - 634us/epoch - 10us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.6950 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.9958 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 926.0417 - 817us/epoch - 13us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.5710 - 590us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 796.3991 - 686us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 829.9436 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.7862 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4864 - 656us/epoch - 11us/sample\n", "Episode 28/50\n", "Episode 29/50\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.0727 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.9338 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.5759 - 701us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9882 - 694us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.7665 - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.5048 - 740us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 897.4641 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 829.2654 - 703us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 208.4666 - 580us/epoch - 9us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 834.6647 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.7889 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 903.8250 - 610us/epoch - 10us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.0028 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.4542 - 783us/epoch - 13us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 826.5320 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.1517 - 808us/epoch - 13us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.9827 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.6644 - 690us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.6276 - 614us/epoch - 10us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.9225 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.7628 - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.9252 - 619us/epoch - 10us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.8045 - 632us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.6452 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.6465 - 652us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.8560 - 731us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.0288 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.7119 - 749us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.3105 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 907.8189 - 3ms/epoch - 56us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.2692 - 1ms/epoch - 17us/sample\n", "Generation 20/300\n", "Solving for Nash Equilibrium in Generation 20/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.5185 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 791.4483 - 587us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7225 - 599us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.3753 - 741us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.0704 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.0262 - 761us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.6334 - 639us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 832.1275 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.8316 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9179 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.7083 - 767us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.9524 - 877us/epoch - 14us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.1434 - 779us/epoch - 13us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.6727 - 634us/epoch - 10us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3074 - 792us/epoch - 13us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.1525 - 906us/epoch - 15us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 846.1521 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.4386 - 660us/epoch - 11us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.0206 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.0499 - 803us/epoch - 13us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.5135 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.5369 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.5987 - 764us/epoch - 12us/sample\n", "Episode 16/50\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.5759 - 750us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.5281 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.8844 - 688us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.0055 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.4661 - 891us/epoch - 14us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.1033 - 712us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.3586 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.2665 - 824us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.7131 - 829us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1100 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.9022 - 891us/epoch - 14us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9831 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.6528 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.5039 - 756us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.3333 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.7362 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.3125 - 666us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 842.6253 - 772us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.7145 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3351 - 768us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.9456 - 638us/epoch - 10us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.5513 - 700us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.1625 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.9174 - 690us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.7009 - 773us/epoch - 12us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.8776 - 835us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9548 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.4924 - 808us/epoch - 13us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.7318 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.6915 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.1011 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 978.6025 - 895us/epoch - 14us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.6055 - 1ms/epoch - 17us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.0327 - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.2708 - 8ms/epoch - 122us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.2369 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.6486 - 908us/epoch - 15us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.1624 - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.8582 - 2ms/epoch - 30us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.4317 - 762us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.2665 - 715us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.5714 - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 824.6289 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.1541 - 10ms/epoch - 164us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7345 - 794us/epoch - 13us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.2513 - 2ms/epoch - 28us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.3904 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.2393 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 793.3152 - 754us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.8430 - 713us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 860.2107 - 708us/epoch - 11us/sample\n", "Generation 21/300\n", "Solving for Nash Equilibrium in Generation 21/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1092 - 5ms/epoch - 84us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.1334 - 659us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.3838 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.1839 - 648us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.0596 - 1ms/epoch - 17us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 793.4982 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.9316 - 831us/epoch - 13us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 870.8404 - 1ms/epoch - 18us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.3134 - 709us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.3406 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.3654 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.9260 - 711us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.4083 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.6705 - 649us/epoch - 10us/sample\n", "Episode 11/50\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 842.8892 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 949.1506 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.3533 - 819us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.2745 - 619us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.3150 - 591us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.1741 - 647us/epoch - 10us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.3289 - 779us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.5706 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.7770 - 611us/epoch - 10us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.2021 - 565us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.7269 - 621us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.9390 - 696us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.1166 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.4936 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.3999 - 2ms/epoch - 32us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3768 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.6177 - 822us/epoch - 13us/sample\n", "Episode 24/50\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.0379 - 874us/epoch - 14us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.7968 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.8398 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.5602 - 1ms/epoch - 23us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.7953 - 609us/epoch - 10us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.3516 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6329 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0911 - 558us/epoch - 9us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.7117 - 746us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.3438 - 674us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.6437 - 578us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.9707 - 3ms/epoch - 51us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.0026 - 607us/epoch - 10us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.7631 - 647us/epoch - 10us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7092 - 676us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8904 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7595 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.0757 - 722us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 109.3285 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7095 - 2ms/epoch - 29us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812.6287 - 695us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.2621 - 800us/epoch - 13us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 881.3873 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.4699 - 679us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 877.0595 - 708us/epoch - 11us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.2641 - 750us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4761 - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.0016 - 677us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.8562 - 632us/epoch - 10us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.6996 - 693us/epoch - 11us/sample\n", "Generation 22/300\n", "Solving for Nash Equilibrium in Generation 22/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.4206 - 713us/epoch - 11us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.2108 - 681us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.5258 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.3254 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9467 - 617us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 210.7064 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.8137 - 753us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2040 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.9061 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.6093 - 735us/epoch - 12us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.0677 - 586us/epoch - 9us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.0300 - 765us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.4175 - 898us/epoch - 14us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.2399 - 796us/epoch - 13us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 851.7114 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.6049 - 694us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.1586 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.3597 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.7812 - 647us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 917.8206 - 698us/epoch - 11us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.7109 - 571us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.0714 - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.1757 - 822us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.0055 - 626us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.9260 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.2601 - 585us/epoch - 9us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.7573 - 731us/epoch - 12us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.6856 - 609us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.5885 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.3913 - 615us/epoch - 10us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.1324 - 754us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 885.5309 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.3116 - 756us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 775.4496 - 571us/epoch - 9us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.6564 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.1151 - 616us/epoch - 10us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.3143 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.1227 - 733us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.7815 - 692us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 858.2267 - 895us/epoch - 14us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.2151 - 711us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 875.1071 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.7219 - 690us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.3782 - 718us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.1788 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.4290 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5516 - 1ms/epoch - 18us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.4341 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.9407 - 824us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.3745 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.4398 - 737us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.2140 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.6061 - 1ms/epoch - 19us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 775.4077 - 602us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 882.7852 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.8658 - 733us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5150 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.2585 - 767us/epoch - 12us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.1156 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.9599 - 629us/epoch - 10us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.6786 - 637us/epoch - 10us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.3313 - 926us/epoch - 15us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104.2243 - 701us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.2881 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.4747 - 666us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.5318 - 620us/epoch - 10us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 809.6251 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.8281 - 765us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 896.7700 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 797.4894 - 745us/epoch - 12us/sample\n", "Generation 23/300\n", "Solving for Nash Equilibrium in Generation 23/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 867.2643 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.7620 - 637us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.6164 - 670us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.2623 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.8062 - 2ms/epoch - 32us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.2162 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.2227 - 951us/epoch - 15us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.0619 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.2095 - 774us/epoch - 12us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.3547 - 701us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.3381 - 826us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.3896 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8591 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.2569 - 638us/epoch - 10us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1004 - 711us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 785.0443 - 596us/epoch - 10us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 29ms/epoch - 476us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.6084 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.7888 - 891us/epoch - 14us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.4835 - 624us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.4777 - 1ms/epoch - 22us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.3918 - 648us/epoch - 10us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.5884 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.1873 - 796us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.9564 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245.7663 - 2ms/epoch - 25us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1005.0622 - 675us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.6936 - 721us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.6223 - 743us/epoch - 12us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.8376 - 755us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.7544 - 611us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812.9517 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.3571 - 746us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.9694 - 641us/epoch - 10us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.4929 - 776us/epoch - 13us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 802.5786 - 671us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.3747 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.2462 - 691us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4333 - 651us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 836.8125 - 601us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4786 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.5515 - 787us/epoch - 13us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.0864 - 647us/epoch - 10us/sample\n", "Episode 32/50\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1118 - 675us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101.5416 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7986 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.7516 - 629us/epoch - 10us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.1282 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.2521 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.2295 - 628us/epoch - 10us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.7464 - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.6748 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.6023 - 747us/epoch - 12us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.7600 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.5985 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.5583 - 882us/epoch - 14us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.5991 - 958us/epoch - 15us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 806.1213 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.6220 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.1204 - 669us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 823.5064 - 722us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.0729 - 748us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.4070 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.6953 - 658us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.7442 - 665us/epoch - 11us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.9554 - 3ms/epoch - 46us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.9884 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.4706 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.4248 - 842us/epoch - 14us/sample\n", "Generation 24/300\n", "Solving for Nash Equilibrium in Generation 24/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7176 - 662us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.2786 - 643us/epoch - 10us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.3065 - 781us/epoch - 13us/sample\n", "Episode 4/50\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 73.1860 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.1816 - 5ms/epoch - 76us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.6006 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4268 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.1466 - 601us/epoch - 10us/sample\n", "Episode 9/50\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.6431 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.3524 - 651us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.5752 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.4915 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.1776 - 660us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.8109 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.4536 - 836us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0291 - 662us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.6941 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.7571 - 645us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.2521 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.6086 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 910.4091 - 34ms/epoch - 546us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.3655 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 848.6627 - 587us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.2812 - 901us/epoch - 15us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.9052 - 637us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.9529 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.6101 - 748us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.1564 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.7192 - 739us/epoch - 12us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.4379 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.0193 - 1ms/epoch - 16us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2557 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.5421 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.6667 - 630us/epoch - 10us/sample\n", "Episode 24/50\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 830.0555 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.8319 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.8878 - 910us/epoch - 15us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.0025 - 811us/epoch - 13us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.3037 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 793.3210 - 641us/epoch - 10us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.6588 - 588us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.0217 - 611us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.3879 - 608us/epoch - 10us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.3316 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.5778 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.8109 - 761us/epoch - 12us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.9152 - 696us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.9431 - 760us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.5699 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.7749 - 564us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.4636 - 698us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.9139 - 803us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.4104 - 674us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.9654 - 666us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.7636 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.2626 - 678us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.4263 - 2ms/epoch - 37us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.3059 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.7510 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.0126 - 646us/epoch - 10us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.8626 - 678us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.6661 - 769us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.2449 - 1ms/epoch - 21us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.9279 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.1993 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4420 - 641us/epoch - 10us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.4495 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.0676 - 698us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.3554 - 721us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.7967 - 856us/epoch - 14us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9806 - 822us/epoch - 13us/sample\n", "Generation 25/300\n", "Solving for Nash Equilibrium in Generation 25/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.0325 - 764us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4520 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.8609 - 650us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6752 - 638us/epoch - 10us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.0364 - 735us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.3949 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2718 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 802.8340 - 592us/epoch - 10us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.1073 - 719us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.2310 - 676us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7468 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.6236 - 631us/epoch - 10us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.7468 - 1ms/epoch - 18us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.8588 - 586us/epoch - 9us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.9366 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.9756 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.9431 - 597us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.4224 - 669us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.2349 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.0255 - 738us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.1765 - 702us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.4164 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 789.0500 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3466 - 907us/epoch - 15us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.9220 - 700us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.8203 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.8649 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.1003 - 707us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.7207 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.7791 - 693us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.1757 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4152 - 598us/epoch - 10us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.6140 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.9648 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.6621 - 669us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.7784 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.8389 - 697us/epoch - 11us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.1771 - 726us/epoch - 12us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 827.7884 - 616us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.5509 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.4561 - 567us/epoch - 9us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.9476 - 642us/epoch - 10us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.6476 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.7816 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.9254 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 794.6068 - 607us/epoch - 10us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.2294 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.3426 - 592us/epoch - 10us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.8893 - 854us/epoch - 14us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 783.4317 - 794us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.3345 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.9798 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6835 - 3ms/epoch - 44us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.5544 - 709us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.5391 - 712us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.5913 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.9948 - 642us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.7568 - 626us/epoch - 10us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.6989 - 790us/epoch - 13us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.2975 - 657us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.1792 - 3ms/epoch - 52us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.0314 - 731us/epoch - 12us/sample\n", "Episode 46/50\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.3546 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.1424 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.0835 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.0619 - 672us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.0758 - 2ms/epoch - 35us/sample\n", "Episode 50/50\n", "Generation 26/300\n", "Solving for Nash Equilibrium in Generation 26/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.9490 - 621us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7125 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.9387 - 728us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.1771 - 783us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.1937 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.0211 - 836us/epoch - 13us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.1019 - 623us/epoch - 10us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.2077 - 814us/epoch - 13us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.6799 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.1293 - 2ms/epoch - 25us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.1085 - 661us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.1573 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.6436 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.4340 - 696us/epoch - 11us/sample\n", "Episode 11/50\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.7747 - 632us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.4969 - 586us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.3079 - 632us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.3079 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.3074 - 618us/epoch - 10us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.2722 - 743us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 222.8002 - 717us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.8542 - 632us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.5952 - 649us/epoch - 10us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.4203 - 687us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.2957 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.6332 - 901us/epoch - 15us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.7708 - 766us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.2719 - 597us/epoch - 10us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.8604 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6293 - 696us/epoch - 11us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.1345 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74.8409 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4233 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.6376 - 741us/epoch - 12us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.8101 - 822us/epoch - 13us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.1991 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.3814 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.8420 - 629us/epoch - 10us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.3272 - 657us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6325 - 823us/epoch - 13us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.1320 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.5174 - 596us/epoch - 10us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.2164 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.7656 - 600us/epoch - 10us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.2720 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.6168 - 612us/epoch - 10us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.6741 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9586 - 692us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.4660 - 591us/epoch - 10us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.4823 - 608us/epoch - 10us/sample\n", "Episode 44/50\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.4105 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.3907 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.7455 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 128.8477 - 664us/epoch - 11us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.3009 - 760us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.1923 - 702us/epoch - 11us/sample\n", "Episode 50/50\n", "Generation 27/300\n", "Solving for Nash Equilibrium in Generation 27/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2747 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.1038 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.7316 - 664us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.9353 - 773us/epoch - 12us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.5250 - 601us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.1436 - 609us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.9279 - 645us/epoch - 10us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 796.4508 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.5245 - 688us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.9584 - 594us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.5406 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 432.6605 - 764us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9647 - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4601 - 638us/epoch - 10us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.0665 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.7908 - 697us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 818.9455 - 691us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.2684 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.1957 - 598us/epoch - 10us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2758 - 891us/epoch - 14us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.5913 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.5788 - 643us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.4999 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.0210 - 823us/epoch - 13us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.9453 - 740us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.8519 - 692us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3928 - 3ms/epoch - 56us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 830.7388 - 644us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.6996 - 668us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.1248 - 661us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.9598 - 765us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.8478 - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.3444 - 643us/epoch - 10us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.0331 - 803us/epoch - 13us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4849 - 759us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.0933 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.5593 - 723us/epoch - 12us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.5991 - 644us/epoch - 10us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.7242 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.9240 - 3ms/epoch - 51us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3219 - 632us/epoch - 10us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.8894 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.9046 - 593us/epoch - 10us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.4260 - 673us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.8929 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.0319 - 654us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.8754 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.1795 - 864us/epoch - 14us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.2479 - 660us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.0905 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.5397 - 683us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.9866 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.5729 - 1ms/epoch - 19us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.2053 - 904us/epoch - 15us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 783.9890 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5469 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6820 - 590us/epoch - 10us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8192 - 710us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.0422 - 599us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.9013 - 636us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.5278 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.7328 - 739us/epoch - 12us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.5574 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4766 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.8178 - 678us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.2084 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.2704 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.9202 - 761us/epoch - 12us/sample\n", "Episode 50/50\n", "Generation 28/300\n", "Solving for Nash Equilibrium in Generation 28/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.5784 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 841.1703 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 785.3357 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.5641 - 639us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.4668 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.9216 - 798us/epoch - 13us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.6144 - 669us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1934 - 649us/epoch - 10us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 837.7757 - 704us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.4023 - 684us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.6306 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.5759 - 791us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.4504 - 674us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 432.0298 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.5783 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.2834 - 744us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.8420 - 759us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.6851 - 701us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2327 - 653us/epoch - 11us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.1274 - 770us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.8230 - 717us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.2559 - 639us/epoch - 10us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.9651 - 827us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 826.2289 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.0477 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.6351 - 593us/epoch - 10us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9980 - 773us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.0731 - 829us/epoch - 13us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.7658 - 700us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 775.5190 - 699us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.5369 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.9242 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.7582 - 3ms/epoch - 44us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.5862 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.5203 - 677us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.0028 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.2700 - 761us/epoch - 12us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 172.3462 - 672us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 815.6365 - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.7672 - 4ms/epoch - 62us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.7243 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.4386 - 643us/epoch - 10us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.3424 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.6059 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.3287 - 717us/epoch - 12us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6368 - 642us/epoch - 10us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.2824 - 664us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.1533 - 722us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.0078 - 756us/epoch - 12us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.5732 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.9353 - 627us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6433 - 643us/epoch - 10us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 789.3856 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.3983 - 724us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 76.7227 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.4493 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.6775 - 728us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.3255 - 752us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.4222 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.8552 - 657us/epoch - 11us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.5098 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.1962 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.9072 - 665us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.6478 - 685us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.3705 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.8827 - 604us/epoch - 10us/sample\n", "Episode 50/50\n", "Generation 29/300\n", "Solving for Nash Equilibrium in Generation 29/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.4760 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.8646 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.6011 - 760us/epoch - 12us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.3799 - 874us/epoch - 14us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1589 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.3144 - 729us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 837.4536 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.0399 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.4754 - 587us/epoch - 9us/sample\n", "Episode 7/50\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.5425 - 619us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.8282 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 148us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.2557 - 2ms/epoch - 37us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.3113 - 732us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.5134 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.4052 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.8832 - 844us/epoch - 14us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.1665 - 745us/epoch - 12us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2765 - 616us/epoch - 10us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.4831 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 809.0490 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.2617 - 677us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.7894 - 630us/epoch - 10us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.0590 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.9319 - 4ms/epoch - 59us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95.3412 - 1ms/epoch - 18us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4998 - 1ms/epoch - 16us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.4072 - 728us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2231 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.8976 - 850us/epoch - 14us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.1277 - 827us/epoch - 13us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.2918 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.2855 - 685us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.1891 - 741us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.3314 - 717us/epoch - 12us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 806.1520 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.0939 - 836us/epoch - 13us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.3758 - 1ms/epoch - 17us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.6979 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.3572 - 754us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.2790 - 899us/epoch - 15us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3691 - 710us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6722 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.2509 - 812us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.6547 - 694us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.4828 - 680us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.7864 - 826us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6354 - 795us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.8269 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.5270 - 672us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.5492 - 887us/epoch - 14us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.2210 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.2430 - 795us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.2794 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.2430 - 709us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.9839 - 885us/epoch - 14us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3316 - 716us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.1807 - 664us/epoch - 11us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.8405 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.2318 - 900us/epoch - 15us/sample\n", "Generation 30/300\n", "Solving for Nash Equilibrium in Generation 30/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.9518 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.8579 - 944us/epoch - 15us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.4554 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.2797 - 644us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.7635 - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8669 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4693 - 1ms/epoch - 19us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 846.4285 - 768us/epoch - 12us/sample\n", "Episode 7/50\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.5589 - 718us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.2272 - 3ms/epoch - 46us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.0126 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.4130 - 773us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 785.8382 - 721us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.0260 - 1ms/epoch - 17us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.7683 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.5891 - 871us/epoch - 14us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.8502 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.5187 - 703us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.6375 - 1ms/epoch - 17us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 156us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.7149 - 1ms/epoch - 18us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.0289 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2744 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6372 - 907us/epoch - 15us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2430 - 753us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1571 - 1ms/epoch - 18us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.8248 - 977us/epoch - 16us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.3035 - 698us/epoch - 11us/sample\n", "Episode 25/50\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.2783 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.1398 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.9171 - 777us/epoch - 13us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9927 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 823.1089 - 768us/epoch - 12us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.7502 - 11ms/epoch - 177us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.7910 - 875us/epoch - 14us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.0375 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.2390 - 873us/epoch - 14us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.4899 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9413 - 784us/epoch - 13us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.7838 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46.0992 - 4ms/epoch - 62us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.6636 - 726us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5957 - 851us/epoch - 14us/sample\n", "Episode 38/50\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.3867 - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.0663 - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.5768 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.7844 - 777us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.1418 - 884us/epoch - 14us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.2056 - 1ms/epoch - 17us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.1910 - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.4784 - 938us/epoch - 15us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.8001 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.3508 - 1ms/epoch - 16us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.7989 - 2ms/epoch - 39us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 62.6186 - 791us/epoch - 13us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.4434 - 1ms/epoch - 17us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.6528 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.8322 - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.7476 - 744us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.7485 - 1ms/epoch - 17us/sample\n", "Generation 31/300\n", "Solving for Nash Equilibrium in Generation 31/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.3472 - 900us/epoch - 15us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.9790 - 794us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.1399 - 987us/epoch - 16us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.5632 - 819us/epoch - 13us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1437 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.9509 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.2901 - 794us/epoch - 13us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1180 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.9981 - 1ms/epoch - 18us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3024 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6838 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.0279 - 742us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.1147 - 760us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 822.5970 - 1ms/epoch - 17us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.1841 - 816us/epoch - 13us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.0277 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.3572 - 730us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240.6720 - 1ms/epoch - 18us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.7654 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.0275 - 721us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.0795 - 856us/epoch - 14us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.1847 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.7747 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6331 - 890us/epoch - 14us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.3999 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7323 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.2054 - 849us/epoch - 14us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.8583 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.5480 - 890us/epoch - 14us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.0251 - 3ms/epoch - 43us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.3389 - 727us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.1065 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66.6993 - 781us/epoch - 13us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0021 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0625 - 704us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.5370 - 1ms/epoch - 17us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.5835 - 3ms/epoch - 55us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.3463 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5103 - 823us/epoch - 13us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.9042 - 837us/epoch - 14us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3456 - 1ms/epoch - 18us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.0027 - 882us/epoch - 14us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.2540 - 1ms/epoch - 18us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.8965 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2580 - 822us/epoch - 13us/sample\n", "Episode 35/50\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.1248 - 913us/epoch - 15us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.8102 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48.6456 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1173 - 1ms/epoch - 17us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.6872 - 837us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.1021 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.4271 - 1ms/epoch - 20us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.3262 - 809us/epoch - 13us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.1184 - 857us/epoch - 14us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.1971 - 1ms/epoch - 16us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 119us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2928 - 5ms/epoch - 77us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.6586 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.6250 - 810us/epoch - 13us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.6620 - 965us/epoch - 16us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.1369 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1873 - 749us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 830.0240 - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.8028 - 688us/epoch - 11us/sample\n", "Generation 32/300\n", "Solving for Nash Equilibrium in Generation 32/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.6118 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9612 - 868us/epoch - 14us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 849.5948 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.5548 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.1648 - 816us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.4097 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2734 - 1ms/epoch - 21us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1967 - 833us/epoch - 13us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.4877 - 1ms/epoch - 16us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.5140 - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.6794 - 681us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.2759 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.9100 - 5ms/epoch - 82us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.5869 - 1ms/epoch - 19us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.0933 - 731us/epoch - 12us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.5595 - 867us/epoch - 14us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 803.1712 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.5107 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.4868 - 1ms/epoch - 18us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.3708 - 820us/epoch - 13us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.5699 - 1ms/epoch - 18us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.9738 - 1ms/epoch - 21us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9079 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.8693 - 901us/epoch - 15us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4816 - 964us/epoch - 16us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4558 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.7355 - 1ms/epoch - 21us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7493 - 800us/epoch - 13us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5134 - 4ms/epoch - 57us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 791.4697 - 1ms/epoch - 18us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.4594 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.0174 - 813us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.9264 - 688us/epoch - 11us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.0626 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.3380 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4702 - 811us/epoch - 13us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.6116 - 746us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5266 - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.2922 - 825us/epoch - 13us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.4382 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.3179 - 749us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.3898 - 2ms/epoch - 25us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.1827 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.9412 - 767us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.1035 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.4489 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.7871 - 720us/epoch - 12us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.8201 - 828us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.7261 - 10ms/epoch - 161us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.7542 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.3752 - 1ms/epoch - 17us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 118.1699 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.6237 - 864us/epoch - 14us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.2294 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.8049 - 1ms/epoch - 20us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.4311 - 1ms/epoch - 16us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.7265 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.7769 - 1ms/epoch - 18us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.4019 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.3934 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 823.4738 - 888us/epoch - 14us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.9998 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.9035 - 689us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.9012 - 731us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.6385 - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.3594 - 1ms/epoch - 16us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.4783 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.2863 - 1ms/epoch - 17us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1944 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.3043 - 787us/epoch - 13us/sample\n", "Generation 33/300\n", "Solving for Nash Equilibrium in Generation 33/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.8710 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8874 - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7554 - 770us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.9572 - 891us/epoch - 14us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.2471 - 925us/epoch - 15us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9003 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.3004 - 1ms/epoch - 19us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.1243 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.0894 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5529 - 1ms/epoch - 17us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.5659 - 734us/epoch - 12us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.9122 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.3307 - 1ms/epoch - 18us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.1525 - 810us/epoch - 13us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.1559 - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.8100 - 731us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.4223 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.9925 - 843us/epoch - 14us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.6730 - 966us/epoch - 16us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.8804 - 776us/epoch - 13us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4400 - 798us/epoch - 13us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8264 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.9290 - 768us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.9258 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.0281 - 1ms/epoch - 18us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.7745 - 795us/epoch - 13us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.0206 - 904us/epoch - 15us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.7188 - 930us/epoch - 15us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.5848 - 1ms/epoch - 21us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.6391 - 757us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.1620 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.4153 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.3173 - 6ms/epoch - 93us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.1577 - 995us/epoch - 16us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.4657 - 947us/epoch - 15us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 821.9069 - 873us/epoch - 14us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4987 - 982us/epoch - 16us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2725 - 768us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.8804 - 651us/epoch - 10us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1868 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.8300 - 4ms/epoch - 59us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6459 - 1ms/epoch - 19us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.0338 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.9798 - 635us/epoch - 10us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812.1151 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.3943 - 795us/epoch - 13us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.4291 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 14ms/epoch - 229us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.1698 - 1ms/epoch - 21us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.5779 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.5787 - 1ms/epoch - 23us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.5468 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.3346 - 973us/epoch - 16us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.0386 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.4157 - 892us/epoch - 14us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.8260 - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.0313 - 689us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.1779 - 1ms/epoch - 23us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.4323 - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.8510 - 956us/epoch - 15us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.3319 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.3715 - 622us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.3860 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.1560 - 724us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.0110 - 998us/epoch - 16us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.0450 - 871us/epoch - 14us/sample\n", "Generation 34/300\n", "Solving for Nash Equilibrium in Generation 34/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.0923 - 869us/epoch - 14us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.6523 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.8642 - 2ms/epoch - 29us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.1134 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 799.9348 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4323 - 772us/epoch - 12us/sample\n", "Episode 4/50\n", "Episode 5/50\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.5114 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.2306 - 1ms/epoch - 18us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.3057 - 790us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.0334 - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6964 - 740us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.9204 - 852us/epoch - 14us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.3102 - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.9551 - 1ms/epoch - 16us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0749 - 902us/epoch - 15us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.8814 - 861us/epoch - 14us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 137us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.2472 - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.0279 - 918us/epoch - 15us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.4599 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9152 - 732us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.7728 - 703us/epoch - 11us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 823.5938 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3064 - 820us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.6623 - 628us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.8422 - 1ms/epoch - 17us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1960 - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.1561 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.6812 - 3ms/epoch - 40us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.3954 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.6249 - 677us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.4636 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.0975 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.1224 - 737us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4141 - 3ms/epoch - 51us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.2297 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.4815 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.2200 - 798us/epoch - 13us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.8653 - 818us/epoch - 13us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.2389 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.3754 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.5542 - 691us/epoch - 11us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1121 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.1129 - 756us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.8642 - 1ms/epoch - 18us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.4836 - 7ms/epoch - 110us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.6826 - 915us/epoch - 15us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.5497 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.1055 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.2300 - 858us/epoch - 14us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 785.0622 - 813us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.4453 - 1ms/epoch - 16us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.0998 - 919us/epoch - 15us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.7331 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.2266 - 1ms/epoch - 18us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 838.2698 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6057 - 900us/epoch - 15us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.1551 - 983us/epoch - 16us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.5580 - 797us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.4512 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.7308 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.4724 - 749us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.5484 - 849us/epoch - 14us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.2434 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.5257 - 2ms/epoch - 37us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 821.6152 - 1ms/epoch - 16us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.9367 - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.7204 - 645us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.4630 - 826us/epoch - 13us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.7379 - 721us/epoch - 12us/sample\n", "Episode 50/50\n", "Generation 35/300\n", "Solving for Nash Equilibrium in Generation 35/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.0272 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.9149 - 2ms/epoch - 30us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.0466 - 800us/epoch - 13us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.7285 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.6157 - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.4919 - 792us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5528 - 613us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.2687 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.4579 - 682us/epoch - 11us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.9315 - 783us/epoch - 13us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.8867 - 678us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.0909 - 749us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 778.6383 - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9833 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.1342 - 676us/epoch - 11us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.0138 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.6540 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.3987 - 717us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.8209 - 926us/epoch - 15us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4950 - 944us/epoch - 15us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 824.9781 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.1346 - 851us/epoch - 14us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.1362 - 752us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.4550 - 598us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9406 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.7731 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.7368 - 740us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5956 - 618us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.3294 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.1716 - 765us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.3172 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.5730 - 837us/epoch - 13us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.6787 - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5692 - 761us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.0477 - 636us/epoch - 10us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7894 - 971us/epoch - 16us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.5295 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.7076 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2962 - 636us/epoch - 10us/sample\n", "Episode 26/50\n", "Episode 27/50\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 806.7463 - 760us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2520 - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6293 - 718us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.1974 - 741us/epoch - 12us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.4426 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.3609 - 668us/epoch - 11us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2737 - 838us/epoch - 14us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5472 - 683us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.1513 - 674us/epoch - 11us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.9164 - 716us/epoch - 12us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 103.8409 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.4869 - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.0777 - 993us/epoch - 16us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3244 - 919us/epoch - 15us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.2394 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.3575 - 718us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.3364 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.1631 - 814us/epoch - 13us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.4024 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.0793 - 664us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.3976 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.7424 - 915us/epoch - 15us/sample\n", "Episode 46/50\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.7034 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.4349 - 938us/epoch - 15us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.9994 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.9017 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.9119 - 1ms/epoch - 18us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.8325 - 628us/epoch - 10us/sample\n", "Generation 36/300\n", "Solving for Nash Equilibrium in Generation 36/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.1505 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.5124 - 629us/epoch - 10us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.8528 - 622us/epoch - 10us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.2729 - 730us/epoch - 12us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.1447 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.4085 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.0334 - 748us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.9186 - 784us/epoch - 13us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.8337 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2842 - 859us/epoch - 14us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.2553 - 936us/epoch - 15us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.0126 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.7560 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.3995 - 742us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.8307 - 8ms/epoch - 133us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 789.1339 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.5359 - 737us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 802.6755 - 1ms/epoch - 16us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.8428 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0477 - 1ms/epoch - 17us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.6482 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.5208 - 727us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.1874 - 968us/epoch - 16us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 805.2270 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 816.8431 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.5551 - 664us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.2597 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.1076 - 1ms/epoch - 17us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.6634 - 856us/epoch - 14us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.3568 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.5677 - 743us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.8294 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.3066 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.8793 - 664us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.5516 - 709us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.9647 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.2926 - 815us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3586 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.0297 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.8083 - 760us/epoch - 12us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.7910 - 623us/epoch - 10us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.8242 - 652us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.8723 - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.4333 - 727us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.7303 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8256 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.8320 - 714us/epoch - 12us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 814.8046 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.6252 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.7053 - 825us/epoch - 13us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.5344 - 949us/epoch - 15us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 213.6798 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.8424 - 917us/epoch - 15us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6750 - 900us/epoch - 15us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2739 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.7410 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.1535 - 878us/epoch - 14us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.3129 - 847us/epoch - 14us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.8744 - 805us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.5645 - 1ms/epoch - 18us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4857 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.2661 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.0521 - 974us/epoch - 16us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.4439 - 732us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.8855 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.0640 - 2ms/epoch - 30us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.8527 - 684us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.7380 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.1398 - 868us/epoch - 14us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.0881 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.8353 - 696us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.9892 - 736us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 819.5104 - 699us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.7440 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.1232 - 764us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.2711 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.4253 - 606us/epoch - 10us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.1232 - 823us/epoch - 13us/sample\n", "Generation 37/300\n", "Solving for Nash Equilibrium in Generation 37/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.4139 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2708 - 779us/epoch - 13us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9548 - 860us/epoch - 14us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.1929 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0803 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.5277 - 694us/epoch - 11us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.9999 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.3452 - 739us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.3963 - 2ms/epoch - 36us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.2261 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.3905 - 682us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.6423 - 684us/epoch - 11us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.3480 - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.1441 - 665us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.3347 - 690us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.2545 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0958 - 755us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 805.3423 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 797.3027 - 649us/epoch - 10us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.5104 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.8977 - 642us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.3806 - 867us/epoch - 14us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.7595 - 759us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.6281 - 753us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.0048 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.3560 - 848us/epoch - 14us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.1889 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.5931 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2968 - 692us/epoch - 11us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.3820 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5161 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.2214 - 741us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.1426 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.2263 - 1ms/epoch - 16us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6385 - 3ms/epoch - 42us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.6130 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.9549 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.6060 - 759us/epoch - 12us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.5360 - 735us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4199 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.7746 - 2ms/epoch - 26us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9207 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.7779 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.4675 - 740us/epoch - 12us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9313 - 795us/epoch - 13us/sample\n", "Episode 33/50\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 819.8244 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9331 - 843us/epoch - 14us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.6187 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.5598 - 678us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7375 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.1546 - 956us/epoch - 15us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.7190 - 812us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2319 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1503 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.1249 - 920us/epoch - 15us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.0193 - 636us/epoch - 10us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.2270 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.5057 - 678us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.4163 - 675us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.8536 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6880 - 687us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.8961 - 2ms/epoch - 29us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.5901 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.3147 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.5380 - 686us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.5616 - 928us/epoch - 15us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.1074 - 672us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2104 - 783us/epoch - 13us/sample\n", "Generation 38/300\n", "Solving for Nash Equilibrium in Generation 38/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 815.9902 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.2611 - 771us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.3605 - 760us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 800.8053 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.8397 - 706us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0631 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.3922 - 811us/epoch - 13us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.3941 - 621us/epoch - 10us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.9400 - 830us/epoch - 13us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.2183 - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.6188 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.5198 - 939us/epoch - 15us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0572 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.4656 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9377 - 736us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.5760 - 1ms/epoch - 20us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.9185 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.3432 - 902us/epoch - 15us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.2094 - 670us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.2105 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.8994 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.6803 - 913us/epoch - 15us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.7845 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.6327 - 1ms/epoch - 22us/sample\n", "Episode 16/50\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.1836 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.3981 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.1105 - 954us/epoch - 15us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.5869 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.8580 - 820us/epoch - 13us/sample\n", "Episode 20/50\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9747 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1302 - 711us/epoch - 11us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.2663 - 796us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.1367 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.2841 - 852us/epoch - 14us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7501 - 960us/epoch - 15us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.0499 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.5260 - 1ms/epoch - 24us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.4315 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.0402 - 732us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.2585 - 648us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.9758 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.3481 - 3ms/epoch - 51us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.4142 - 824us/epoch - 13us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.2108 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.1463 - 655us/epoch - 11us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.1303 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.3589 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.5580 - 766us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.1178 - 603us/epoch - 10us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.0777 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.3183 - 890us/epoch - 14us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.4346 - 665us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 793.1721 - 810us/epoch - 13us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.5967 - 591us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0650 - 686us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.8806 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.4974 - 695us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.2534 - 789us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9991 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9048 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.4693 - 703us/epoch - 11us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6191 - 823us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6696 - 2ms/epoch - 32us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.9523 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.1407 - 794us/epoch - 13us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.8406 - 885us/epoch - 14us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.1663 - 7ms/epoch - 109us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.4529 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.7634 - 939us/epoch - 15us/sample\n", "Generation 39/300\n", "Solving for Nash Equilibrium in Generation 39/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.9304 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.4349 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.8868 - 952us/epoch - 15us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.1198 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7640 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.8882 - 850us/epoch - 14us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.9949 - 640us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3611 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5425 - 779us/epoch - 13us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4232 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.2281 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.4679 - 726us/epoch - 12us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.4382 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2581 - 927us/epoch - 15us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.8674 - 863us/epoch - 14us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.0526 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.6456 - 900us/epoch - 15us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.4449 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.1614 - 737us/epoch - 12us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.3253 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.4872 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.4496 - 906us/epoch - 15us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 805.1606 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.4706 - 2ms/epoch - 33us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.7670 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.3898 - 692us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.5520 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.6171 - 716us/epoch - 12us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.1328 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2098 - 3ms/epoch - 51us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.3185 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1437 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.5008 - 801us/epoch - 13us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.7545 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.5604 - 884us/epoch - 14us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.6847 - 749us/epoch - 12us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3860 - 793us/epoch - 13us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6282 - 612us/epoch - 10us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.4598 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.6329 - 877us/epoch - 14us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.3733 - 869us/epoch - 14us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.0676 - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8790 - 698us/epoch - 11us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.1904 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.3596 - 650us/epoch - 10us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.2985 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.5275 - 588us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.3719 - 775us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.2532 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.4902 - 687us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.5413 - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.4212 - 883us/epoch - 14us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.1371 - 779us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.3584 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428.2027 - 713us/epoch - 12us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.3864 - 828us/epoch - 13us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 815.9323 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.5952 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.5222 - 2ms/epoch - 33us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 811.6292 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.1122 - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.5416 - 965us/epoch - 16us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.7159 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6801 - 737us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.7189 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.5753 - 889us/epoch - 14us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.8894 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.2509 - 827us/epoch - 13us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.3300 - 885us/epoch - 14us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.9401 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.1257 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.0262 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.4933 - 898us/epoch - 14us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.3489 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.1732 - 2ms/epoch - 39us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.4813 - 1ms/epoch - 22us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.2352 - 850us/epoch - 14us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.0219 - 787us/epoch - 13us/sample\n", "Generation 40/300\n", "Solving for Nash Equilibrium in Generation 40/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.9142 - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.6254 - 854us/epoch - 14us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.8302 - 3ms/epoch - 44us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.8445 - 738us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.0084 - 657us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89.0137 - 776us/epoch - 13us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.9386 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.2379 - 676us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2330 - 743us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9715 - 743us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.7849 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1147 - 2ms/epoch - 34us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.9515 - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.6161 - 2ms/epoch - 28us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.0672 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.6247 - 776us/epoch - 13us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4642 - 640us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7139 - 1ms/epoch - 17us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.0508 - 853us/epoch - 14us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.8736 - 852us/epoch - 14us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4296 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.1292 - 1ms/epoch - 18us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.5314 - 833us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9518 - 812us/epoch - 13us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.2555 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.6239 - 865us/epoch - 14us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.5860 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.8357 - 982us/epoch - 16us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4106 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.9434 - 831us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 807.0721 - 767us/epoch - 12us/sample\n", "Episode 24/50\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787.4158 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.2607 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5086 - 746us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5904 - 771us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.1466 - 693us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5024 - 916us/epoch - 15us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.0238 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.7720 - 1ms/epoch - 23us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.1203 - 1ms/epoch - 23us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.6615 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.7784 - 616us/epoch - 10us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.5926 - 812us/epoch - 13us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.5303 - 776us/epoch - 13us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9866 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2395 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.3912 - 3ms/epoch - 49us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.1563 - 798us/epoch - 13us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0289 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.9410 - 1ms/epoch - 19us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.1666 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.4901 - 873us/epoch - 14us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.4772 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.7279 - 968us/epoch - 16us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7458 - 878us/epoch - 14us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7553 - 778us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.4811 - 849us/epoch - 14us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 820.1048 - 1ms/epoch - 21us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.1553 - 697us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.5462 - 898us/epoch - 14us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.3645 - 849us/epoch - 14us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.3396 - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.8030 - 968us/epoch - 16us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 813.6647 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.3881 - 3ms/epoch - 45us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.4549 - 847us/epoch - 14us/sample\n", "Generation 41/300\n", "Solving for Nash Equilibrium in Generation 41/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9006 - 751us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.2443 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4828 - 893us/epoch - 14us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.0214 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.0130 - 821us/epoch - 13us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.1617 - 918us/epoch - 15us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.0623 - 855us/epoch - 14us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.4944 - 742us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7463 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.3913 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.2534 - 746us/epoch - 12us/sample\n", "Episode 10/50\n", "Episode 11/50\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.6929 - 3ms/epoch - 44us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.4557 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3619 - 1ms/epoch - 17us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.7499 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.0515 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.0358 - 692us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.4536 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.4257 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.4362 - 939us/epoch - 15us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.0182 - 857us/epoch - 14us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.0811 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.0761 - 2ms/epoch - 32us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.8275 - 826us/epoch - 13us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.0907 - 785us/epoch - 13us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.0833 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.6169 - 653us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.6376 - 883us/epoch - 14us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.8739 - 924us/epoch - 15us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.3141 - 909us/epoch - 15us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.0581 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.8146 - 925us/epoch - 15us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 134us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.1746 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.4429 - 981us/epoch - 16us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.6656 - 939us/epoch - 15us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.7200 - 969us/epoch - 16us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.5476 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.0914 - 891us/epoch - 14us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.1504 - 885us/epoch - 14us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.2520 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6665 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6713 - 913us/epoch - 15us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.3280 - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9368 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1707 - 907us/epoch - 15us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.7283 - 1ms/epoch - 19us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3963 - 879us/epoch - 14us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.2535 - 821us/epoch - 13us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.0225 - 1ms/epoch - 17us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.4462 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.3251 - 766us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4355 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.5134 - 1ms/epoch - 16us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9391 - 653us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3392 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.1918 - 717us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 803.9815 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.2137 - 645us/epoch - 10us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 829.1531 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.3561 - 1ms/epoch - 18us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.6495 - 711us/epoch - 11us/sample\n", "Episode 50/50\n", "Generation 42/300\n", "Solving for Nash Equilibrium in Generation 42/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.3116 - 676us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.4398 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4946 - 755us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.6758 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7094 - 801us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.9525 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 778.7455 - 702us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.6886 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5789 - 619us/epoch - 10us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.9854 - 721us/epoch - 12us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.5354 - 874us/epoch - 14us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.4349 - 750us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.8558 - 1ms/epoch - 17us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3381 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.8232 - 752us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.4919 - 602us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.1371 - 641us/epoch - 10us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.5187 - 899us/epoch - 14us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5926 - 733us/epoch - 12us/sample\n", "Episode 14/50\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.0957 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4803 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.1302 - 693us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.4413 - 2ms/epoch - 25us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.0165 - 982us/epoch - 16us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.5139 - 1ms/epoch - 16us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 804.0028 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.0601 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.0654 - 767us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.5762 - 820us/epoch - 13us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9120 - 802us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.5123 - 923us/epoch - 15us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.5078 - 729us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.1435 - 844us/epoch - 14us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 813.4850 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.7328 - 733us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.9801 - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.0367 - 1ms/epoch - 17us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.4948 - 727us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.1926 - 1ms/epoch - 19us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7790 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.3304 - 695us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.8668 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.9298 - 697us/epoch - 11us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.0646 - 580us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.4454 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.2599 - 917us/epoch - 15us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.0840 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7355 - 875us/epoch - 14us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3961 - 926us/epoch - 15us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 822.9355 - 789us/epoch - 13us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5559 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.8694 - 628us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.9922 - 702us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6249 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7914 - 635us/epoch - 10us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4980 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0711 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.9807 - 658us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0076 - 679us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.5483 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.6046 - 4ms/epoch - 58us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.9802 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.9210 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5574 - 621us/epoch - 10us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.4461 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7347 - 781us/epoch - 13us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0735 - 774us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4777 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.1392 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.9724 - 596us/epoch - 10us/sample\n", "Episode 50/50\n", "Generation 43/300\n", "Solving for Nash Equilibrium in Generation 43/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.5085 - 696us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.8729 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.8400 - 1ms/epoch - 24us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.9640 - 726us/epoch - 12us/sample\n", "Episode 4/50\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.1554 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 794.9537 - 868us/epoch - 14us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.2570 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.8865 - 1ms/epoch - 17us/sample\n", "Episode 8/50\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0330 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.1183 - 626us/epoch - 10us/sample\n", "Episode 11/50\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.5048 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.2454 - 603us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.3768 - 750us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7448 - 602us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.3535 - 711us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.4788 - 777us/epoch - 13us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.4719 - 788us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.9646 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.4816 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.7709 - 813us/epoch - 13us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.8611 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.8447 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9548 - 848us/epoch - 14us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8056 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6268 - 4ms/epoch - 57us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9255 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.0115 - 764us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.0538 - 647us/epoch - 10us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.4790 - 1ms/epoch - 19us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3918 - 769us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812.5150 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9121 - 730us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.5217 - 722us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.7387 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.2982 - 744us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.4957 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.6723 - 1ms/epoch - 20us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8440 - 1ms/epoch - 19us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.3040 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.8344 - 668us/epoch - 11us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.7374 - 672us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.2749 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.4135 - 707us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.3159 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.8573 - 669us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.3238 - 685us/epoch - 11us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.9446 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 805.4814 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.4819 - 692us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1229 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.7361 - 671us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8321 - 583us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.8184 - 640us/epoch - 10us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.5918 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.4203 - 660us/epoch - 11us/sample\n", "Episode 43/50\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.0915 - 756us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0167 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0436 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.8499 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5086 - 662us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.3614 - 1ms/epoch - 18us/sample\n", "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.6815 - 658us/epoch - 11us/sample\n", "Generation 44/300\n", "Solving for Nash Equilibrium in Generation 44/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2469 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8039 - 643us/epoch - 10us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.1589 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 81.1768 - 573us/epoch - 9us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.1380 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.9791 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.4743 - 634us/epoch - 10us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.8242 - 527us/epoch - 9us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.8985 - 640us/epoch - 10us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.0492 - 846us/epoch - 14us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.7388 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.8110 - 709us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2083 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.6717 - 641us/epoch - 10us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.0728 - 634us/epoch - 10us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 817.6127 - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.6241 - 806us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.8184 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.4698 - 685us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.8112 - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5275 - 606us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9891 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.4420 - 564us/epoch - 9us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2150 - 700us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.7543 - 1ms/epoch - 17us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.8768 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.8839 - 745us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2372 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.5816 - 665us/epoch - 11us/sample\n", "Episode 22/50\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.4395 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.0312 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.8279 - 703us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.0743 - 635us/epoch - 10us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0430 - 2ms/epoch - 31us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.3895 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8200 - 762us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7565 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.2241 - 702us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.1103 - 862us/epoch - 14us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.8403 - 667us/epoch - 11us/sample\n", "Episode 31/50\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5788 - 1ms/epoch - 17us/sample\n", "Episode 34/50\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.7895 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.6375 - 851us/epoch - 14us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.0435 - 676us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54.7539 - 2ms/epoch - 37us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.2667 - 702us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 152.5252 - 721us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.2927 - 694us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0446 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.7018 - 845us/epoch - 14us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.6146 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.3552 - 705us/epoch - 11us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.7661 - 727us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.2933 - 576us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.2697 - 547us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.0638 - 722us/epoch - 12us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.8303 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.8839 - 786us/epoch - 13us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.0600 - 758us/epoch - 12us/sample\n", "Generation 45/300\n", "Solving for Nash Equilibrium in Generation 45/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.1562 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.7775 - 800us/epoch - 13us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7396 - 699us/epoch - 11us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.2020 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 820.6994 - 658us/epoch - 11us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9092 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.9162 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.9825 - 676us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9481 - 748us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.4504 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.5330 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.5511 - 637us/epoch - 10us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.5990 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 30ms/epoch - 483us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6344 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.0532 - 762us/epoch - 12us/sample\n", "Episode 12/50\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.2016 - 712us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.8466 - 744us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.2087 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.8137 - 757us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.1196 - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1779 - 1ms/epoch - 22us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.8714 - 1ms/epoch - 16us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8284 - 1ms/epoch - 17us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.0009 - 923us/epoch - 15us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.5702 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.3589 - 687us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.3461 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.6995 - 732us/epoch - 12us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.0312 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9309 - 900us/epoch - 15us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.0703 - 655us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 154.0551 - 783us/epoch - 13us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0720 - 765us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.2655 - 853us/epoch - 14us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.5202 - 681us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4480 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.8226 - 1ms/epoch - 17us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8890 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.2483 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.2717 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.4368 - 671us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2570 - 702us/epoch - 11us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7823 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.5445 - 784us/epoch - 13us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.2532 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.3171 - 700us/epoch - 11us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.0250 - 769us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.4216 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5377 - 768us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.3934 - 822us/epoch - 13us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.7764 - 645us/epoch - 10us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.9986 - 739us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.3192 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.7599 - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.6053 - 683us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.0228 - 827us/epoch - 13us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6447 - 829us/epoch - 13us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2941 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361.9182 - 796us/epoch - 13us/sample\n", "Generation 46/300\n", "Solving for Nash Equilibrium in Generation 46/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.4490 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.9223 - 3ms/epoch - 50us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 807.7184 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.0984 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.0580 - 706us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.7996 - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.8565 - 609us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.1974 - 690us/epoch - 11us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787.1188 - 667us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5684 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2360 - 2ms/epoch - 30us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.6502 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.8550 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7269 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.3528 - 916us/epoch - 15us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.5707 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.2926 - 676us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.3995 - 680us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.1447 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.7486 - 1ms/epoch - 17us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.5595 - 776us/epoch - 13us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.3887 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.4518 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.0308 - 872us/epoch - 14us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.3876 - 657us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.9285 - 725us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.4310 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1124 - 664us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.9488 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.8973 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.2758 - 601us/epoch - 10us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.2274 - 616us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6077 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.4594 - 624us/epoch - 10us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.7365 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.0132 - 844us/epoch - 14us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 805.1492 - 617us/epoch - 10us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.6450 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.1978 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 786.1868 - 660us/epoch - 11us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.9506 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6909 - 743us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.3774 - 861us/epoch - 14us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.9367 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7682 - 732us/epoch - 12us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2847 - 974us/epoch - 16us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3774 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.0822 - 951us/epoch - 15us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.1057 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.3110 - 786us/epoch - 13us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7707 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.2303 - 743us/epoch - 12us/sample\n", "Episode 36/50\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2329 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9510 - 839us/epoch - 14us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.1551 - 782us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.2661 - 653us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4388 - 678us/epoch - 11us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.1255 - 924us/epoch - 15us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.2373 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.8599 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.6096 - 936us/epoch - 15us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.5482 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.1542 - 675us/epoch - 11us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.1572 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.9763 - 818us/epoch - 13us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 814.8184 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.1969 - 667us/epoch - 11us/sample\n", "Generation 47/300\n", "Solving for Nash Equilibrium in Generation 47/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.5905 - 943us/epoch - 15us/sample\n", "Episode 2/50\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.9753 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8120 - 639us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.1210 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.5966 - 776us/epoch - 13us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.7370 - 839us/epoch - 14us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7928 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5237 - 669us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1444 - 639us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.0742 - 599us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.5466 - 2ms/epoch - 27us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.7741 - 850us/epoch - 14us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.4335 - 644us/epoch - 10us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8912 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.0966 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.7458 - 1ms/epoch - 19us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.6396 - 639us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.4781 - 833us/epoch - 13us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.1500 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 796.6785 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.0604 - 686us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.7697 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.6737 - 738us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5831 - 755us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.5988 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.2690 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6281 - 767us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.6406 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.4214 - 740us/epoch - 12us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.7021 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.4864 - 687us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.1752 - 2ms/epoch - 26us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5038 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0495 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.4484 - 712us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.4390 - 803us/epoch - 13us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.9413 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.9765 - 723us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.5164 - 860us/epoch - 14us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7958 - 745us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.3737 - 646us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.1951 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.7702 - 683us/epoch - 11us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.0950 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7297 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.9104 - 684us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.1127 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2723 - 807us/epoch - 13us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.6652 - 869us/epoch - 14us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.9120 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.0179 - 765us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.3758 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9409 - 746us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.8665 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2244 - 971us/epoch - 16us/sample\n", "Episode 38/50\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.3672 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.3159 - 757us/epoch - 12us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 151.8131 - 948us/epoch - 15us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.1194 - 723us/epoch - 12us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1503 - 680us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.7982 - 702us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.1398 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.3770 - 614us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 91us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.9750 - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.9471 - 810us/epoch - 13us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48.4399 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.0441 - 669us/epoch - 11us/sample\n", "Episode 50/50\n", "Generation 48/300\n", "Solving for Nash Equilibrium in Generation 48/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.7611 - 3ms/epoch - 52us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.4036 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.4910 - 2ms/epoch - 32us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.5862 - 832us/epoch - 13us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.7497 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 811.6418 - 735us/epoch - 12us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.3797 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.8339 - 828us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.9727 - 654us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.2390 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.0939 - 664us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.3591 - 781us/epoch - 13us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.0953 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.9490 - 878us/epoch - 14us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.5684 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.1002 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.8932 - 724us/epoch - 12us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8831 - 654us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.5854 - 755us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 778.9514 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.5710 - 2ms/epoch - 38us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6815 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.7728 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0092 - 714us/epoch - 12us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9210 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.7471 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7421 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.5047 - 692us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.8282 - 3ms/epoch - 49us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.7316 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9448 - 533us/epoch - 9us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.8123 - 652us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.1284 - 691us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1764 - 713us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8972 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2318 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.8979 - 667us/epoch - 11us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.5294 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.2098 - 727us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.4802 - 675us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4468 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 811.3873 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.2817 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.1641 - 598us/epoch - 10us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.7502 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.7644 - 749us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2688 - 670us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7315 - 670us/epoch - 11us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2720 - 694us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.1546 - 641us/epoch - 10us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.7982 - 789us/epoch - 13us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.8757 - 743us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.4034 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5123 - 744us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50.8987 - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.1901 - 649us/epoch - 10us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5363 - 733us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 791.1877 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.3909 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.8394 - 729us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.0320 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.3116 - 709us/epoch - 11us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.7191 - 750us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5365 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.3251 - 2ms/epoch - 28us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.9846 - 647us/epoch - 10us/sample\n", "Generation 49/300\n", "Solving for Nash Equilibrium in Generation 49/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.0556 - 770us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5811 - 740us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7562 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.9304 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0843 - 709us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.3207 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.4667 - 675us/epoch - 11us/sample\n", "Episode 5/50\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1812 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.4328 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.5106 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.2899 - 646us/epoch - 10us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.7670 - 646us/epoch - 10us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4379 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.2330 - 847us/epoch - 14us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.5657 - 731us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.8477 - 672us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3366 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.2890 - 677us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5299 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4635 - 726us/epoch - 12us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.3752 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.8580 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.9468 - 657us/epoch - 11us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9031 - 649us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.3375 - 712us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.5812 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8351 - 645us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.7430 - 888us/epoch - 14us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.8945 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.5997 - 2ms/epoch - 28us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.9092 - 792us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.7213 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.9875 - 748us/epoch - 12us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2687 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.5674 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406.7148 - 581us/epoch - 9us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.0726 - 778us/epoch - 13us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5372 - 923us/epoch - 15us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.3135 - 725us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.2079 - 739us/epoch - 12us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.5861 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.5458 - 768us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.8453 - 751us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.5178 - 2ms/epoch - 37us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.2649 - 844us/epoch - 14us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.7042 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.3187 - 802us/epoch - 13us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.2011 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.5770 - 608us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.6389 - 693us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.5937 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.0391 - 682us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2198 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.7300 - 952us/epoch - 15us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.4293 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.2849 - 730us/epoch - 12us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.5990 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.7192 - 758us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.7735 - 592us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.6432 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.5743 - 677us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.1954 - 604us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4691 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.0586 - 769us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.8067 - 643us/epoch - 10us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.5569 - 4ms/epoch - 70us/sample\n", "Generation 50/300\n", "Solving for Nash Equilibrium in Generation 50/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.4771 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.1574 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.6902 - 645us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8481 - 743us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.7487 - 665us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8406 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.5982 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6149 - 731us/epoch - 12us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.5103 - 742us/epoch - 12us/sample\n", "Episode 6/50\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.1195 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.7175 - 550us/epoch - 9us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.0057 - 652us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.5714 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.7596 - 774us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.6869 - 673us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.3558 - 693us/epoch - 11us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6931 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.8921 - 594us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.5679 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.8112 - 1ms/epoch - 19us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.0806 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4686 - 586us/epoch - 9us/sample\n", "Episode 18/50\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2585 - 1ms/epoch - 20us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.3574 - 783us/epoch - 13us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2183 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.0652 - 662us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.1949 - 677us/epoch - 11us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.9554 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.8453 - 747us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.3403 - 731us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 802.6401 - 607us/epoch - 10us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.2496 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.7108 - 605us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.8015 - 714us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.2936 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.8375 - 1ms/epoch - 18us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.1169 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.2001 - 584us/epoch - 9us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.2680 - 594us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.2622 - 2ms/epoch - 40us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.6928 - 655us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 786.7005 - 761us/epoch - 12us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.7311 - 697us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8659 - 734us/epoch - 12us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4182 - 693us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.9516 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.4926 - 774us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4908 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.1245 - 625us/epoch - 10us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.1649 - 677us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.1348 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.6815 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.7420 - 747us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.0444 - 639us/epoch - 10us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.7347 - 3ms/epoch - 51us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.3544 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.7786 - 667us/epoch - 11us/sample\n", "Episode 47/50\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.2330 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.2555 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5450 - 1ms/epoch - 19us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 785.7797 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.2702 - 2ms/epoch - 32us/sample\n", "Generation 51/300\n", "Solving for Nash Equilibrium in Generation 51/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.2652 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.0501 - 2ms/epoch - 32us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7206 - 726us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.8244 - 726us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.9998 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.6321 - 850us/epoch - 14us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.8647 - 665us/epoch - 11us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.0804 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.1649 - 694us/epoch - 11us/sample\n", "Episode 7/50\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.0398 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.2292 - 748us/epoch - 12us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.2179 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.1612 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.5256 - 1ms/epoch - 16us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.3625 - 824us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.1439 - 908us/epoch - 15us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 81.9174 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.6880 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.7359 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.2245 - 733us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.8032 - 729us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.2007 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.7767 - 702us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6876 - 727us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.3220 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.3261 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.3112 - 668us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787.3357 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.8377 - 962us/epoch - 16us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7650 - 812us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2107 - 672us/epoch - 11us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.2991 - 830us/epoch - 13us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.1106 - 740us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9749 - 700us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.9238 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.4260 - 6ms/epoch - 101us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.9252 - 829us/epoch - 13us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.6855 - 873us/epoch - 14us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.0290 - 816us/epoch - 13us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 23ms/epoch - 364us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.5208 - 3ms/epoch - 44us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.8180 - 729us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.1285 - 666us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.6555 - 908us/epoch - 15us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.7765 - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.3137 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.0692 - 822us/epoch - 13us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5648 - 777us/epoch - 13us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.5640 - 820us/epoch - 13us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.0248 - 782us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40.2993 - 8ms/epoch - 125us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.9231 - 799us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4785 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.6859 - 672us/epoch - 11us/sample\n", "Episode 44/50\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3188 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0189 - 791us/epoch - 13us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.8787 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.7057 - 762us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1760 - 1ms/epoch - 21us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.1298 - 710us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7702 - 811us/epoch - 13us/sample\n", "Generation 52/300\n", "Solving for Nash Equilibrium in Generation 52/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.3285 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.9171 - 691us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.2007 - 707us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1048 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.2732 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.1927 - 2ms/epoch - 29us/sample\n", "Episode 4/50\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.2610 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.2559 - 767us/epoch - 12us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.6089 - 895us/epoch - 14us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8161 - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.8073 - 590us/epoch - 10us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7294 - 840us/epoch - 14us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8428 - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4565 - 820us/epoch - 13us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4443 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5043 - 692us/epoch - 11us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9788 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.2494 - 687us/epoch - 11us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 147.8437 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.1002 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3615 - 709us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2986 - 706us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 432.0403 - 790us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.7193 - 1ms/epoch - 17us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.4045 - 1ms/epoch - 16us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.4769 - 2ms/epoch - 30us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4208 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9235 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.8815 - 843us/epoch - 14us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.2465 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.7843 - 1ms/epoch - 21us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.5665 - 702us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.1195 - 661us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.5856 - 845us/epoch - 14us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8561 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6088 - 903us/epoch - 15us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.2636 - 838us/epoch - 14us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.4052 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0538 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2440 - 916us/epoch - 15us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.6063 - 813us/epoch - 13us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.8329 - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6578 - 671us/epoch - 11us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.2491 - 2ms/epoch - 24us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.6273 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.1324 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5185 - 809us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.4676 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.6337 - 721us/epoch - 12us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.2900 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.4853 - 5ms/epoch - 80us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5188 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.9283 - 960us/epoch - 15us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.4473 - 850us/epoch - 14us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0501 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.3620 - 1ms/epoch - 16us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.3225 - 735us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.2644 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3298 - 50ms/epoch - 813us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.8497 - 824us/epoch - 13us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.8089 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.6673 - 702us/epoch - 11us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 808.9374 - 2ms/epoch - 35us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.9753 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.5424 - 637us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1478 - 727us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.7311 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.0797 - 722us/epoch - 12us/sample\n", "Episode 50/50\n", "Generation 53/300\n", "Solving for Nash Equilibrium in Generation 53/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.0011 - 893us/epoch - 14us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 775.8773 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.3488 - 697us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.1674 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.7246 - 769us/epoch - 12us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5842 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.4275 - 709us/epoch - 11us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7036 - 599us/epoch - 10us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 790.8014 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.1742 - 676us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812.0405 - 670us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.1815 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.7264 - 950us/epoch - 15us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.3176 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3366 - 594us/epoch - 10us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7218 - 708us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.6219 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.3856 - 733us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.9759 - 2ms/epoch - 28us/sample\n", "Episode 14/50\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.1049 - 777us/epoch - 13us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52.5832 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9960 - 716us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 815.9133 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.0942 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1148 - 672us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.5151 - 1ms/epoch - 17us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.8867 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.7939 - 680us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.2514 - 627us/epoch - 10us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4186 - 651us/epoch - 10us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7759 - 785us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.8181 - 695us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.2645 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.3172 - 628us/epoch - 10us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.2368 - 940us/epoch - 15us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.0919 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5098 - 852us/epoch - 14us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6597 - 639us/epoch - 10us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773.9127 - 657us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.2122 - 673us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.3373 - 735us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.1980 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.9496 - 800us/epoch - 13us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1133 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0094 - 701us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.7923 - 3ms/epoch - 56us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812.3064 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.5527 - 738us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 807.2683 - 664us/epoch - 11us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.3781 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.3535 - 651us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.5545 - 2ms/epoch - 27us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.3087 - 603us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.7098 - 728us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.2834 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.0709 - 977us/epoch - 16us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8356 - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.6121 - 735us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.6549 - 872us/epoch - 14us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 162.8274 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.3781 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.8158 - 652us/epoch - 11us/sample\n", "Generation 54/300\n", "Solving for Nash Equilibrium in Generation 54/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7081 - 706us/epoch - 11us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3193 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.8649 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5463 - 4ms/epoch - 61us/sample\n", "Episode 4/50\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.0610 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7102 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.4943 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.8373 - 715us/epoch - 12us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4743 - 629us/epoch - 10us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8788 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7385 - 743us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.1146 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.3749 - 745us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.5325 - 701us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.6620 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.8470 - 946us/epoch - 15us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.7990 - 718us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.9041 - 661us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7728 - 623us/epoch - 10us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.9075 - 648us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7053 - 794us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.1484 - 1ms/epoch - 23us/sample\n", "Episode 20/50\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.2252 - 636us/epoch - 10us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.5968 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.7309 - 694us/epoch - 11us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.1791 - 703us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9531 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.5959 - 669us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.2767 - 694us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.1878 - 697us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0237 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.5280 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.8666 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0901 - 689us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.3477 - 753us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.3464 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.6664 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.9952 - 668us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.5477 - 786us/epoch - 13us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5111 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.2186 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4864 - 613us/epoch - 10us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.9398 - 860us/epoch - 14us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2175 - 852us/epoch - 14us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.6775 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.1484 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.0643 - 828us/epoch - 13us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 807.3423 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.4465 - 650us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.2511 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0200 - 844us/epoch - 14us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.0669 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6559 - 597us/epoch - 10us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1165 - 846us/epoch - 14us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.1266 - 654us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.8590 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.3698 - 729us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.5058 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5048 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.3199 - 833us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5620 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.0184 - 656us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.1279 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.5897 - 1ms/epoch - 17us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.5006 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.1630 - 626us/epoch - 10us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2475 - 784us/epoch - 13us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.0652 - 700us/epoch - 11us/sample\n", "Generation 55/300\n", "Solving for Nash Equilibrium in Generation 55/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.9857 - 731us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.5321 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.2422 - 2ms/epoch - 35us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47.0590 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.9553 - 575us/epoch - 9us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.0480 - 699us/epoch - 11us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.8031 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.1413 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.1881 - 752us/epoch - 12us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.7185 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.3525 - 846us/epoch - 14us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.1225 - 628us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.5033 - 704us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.1641 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.5275 - 2ms/epoch - 25us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.6245 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.8701 - 870us/epoch - 14us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.8429 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.4005 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.8170 - 716us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.9678 - 604us/epoch - 10us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.2442 - 1ms/epoch - 18us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.7177 - 632us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6121 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.6419 - 1ms/epoch - 17us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.0657 - 711us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.2694 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.1321 - 2ms/epoch - 25us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.9422 - 631us/epoch - 10us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.0065 - 700us/epoch - 11us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.6989 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.4557 - 5ms/epoch - 75us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.7766 - 905us/epoch - 15us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95.6406 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37.7196 - 761us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 804.6666 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.2902 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.2085 - 674us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.3518 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.4251 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2062 - 594us/epoch - 10us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.2475 - 904us/epoch - 15us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.4081 - 1ms/epoch - 16us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.9841 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7559 - 723us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.6521 - 990us/epoch - 16us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.6872 - 730us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.1783 - 711us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 215.5303 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.6263 - 807us/epoch - 13us/sample\n", "Episode 38/50\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4326 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.4046 - 2ms/epoch - 32us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 228.7124 - 784us/epoch - 13us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.3254 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.2877 - 755us/epoch - 12us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.1215 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.5042 - 739us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 15ms/epoch - 238us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9184 - 2ms/epoch - 26us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.4021 - 815us/epoch - 13us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3763 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.8245 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3645 - 1ms/epoch - 20us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.5009 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.4551 - 1ms/epoch - 16us/sample\n", "Generation 56/300\n", "Solving for Nash Equilibrium in Generation 56/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.4112 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.1407 - 739us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.9786 - 777us/epoch - 13us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.2698 - 840us/epoch - 14us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.6193 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8042 - 2ms/epoch - 33us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3707 - 954us/epoch - 15us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.8278 - 665us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.5063 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7506 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.2720 - 800us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2045 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9294 - 956us/epoch - 15us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.2542 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.1821 - 1ms/epoch - 22us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.5451 - 3ms/epoch - 42us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.4526 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.0074 - 1ms/epoch - 18us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8049 - 723us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5992 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.4690 - 877us/epoch - 14us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8273 - 833us/epoch - 13us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 808.3710 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.2133 - 607us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.9940 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.2065 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.5951 - 603us/epoch - 10us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9227 - 650us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0980 - 794us/epoch - 13us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.5413 - 759us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4358 - 704us/epoch - 11us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.8907 - 817us/epoch - 13us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.5239 - 1ms/epoch - 22us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.4864 - 770us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.4349 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.8403 - 616us/epoch - 10us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79.0136 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7844 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3017 - 578us/epoch - 9us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 786.8329 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.7816 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.4106 - 786us/epoch - 13us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.6580 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.4130 - 707us/epoch - 11us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1615 - 824us/epoch - 13us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0122 - 640us/epoch - 10us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.7953 - 2ms/epoch - 33us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.2398 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6555 - 629us/epoch - 10us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 810.7908 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.3917 - 713us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7491 - 721us/epoch - 12us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.1510 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6116 - 616us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7418 - 648us/epoch - 10us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.3174 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.3098 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6229 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.2104 - 866us/epoch - 14us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3970 - 715us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9692 - 617us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2784 - 798us/epoch - 13us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8396 - 638us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.2607 - 618us/epoch - 10us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.7507 - 750us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.6857 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.9617 - 659us/epoch - 11us/sample\n", "Generation 57/300\n", "Solving for Nash Equilibrium in Generation 57/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9100 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.3098 - 2ms/epoch - 36us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.9852 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.3435 - 617us/epoch - 10us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3432 - 1ms/epoch - 16us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2778 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.2134 - 779us/epoch - 13us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.9424 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.3688 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.1690 - 720us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.6070 - 746us/epoch - 12us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.9255 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5609 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9400 - 876us/epoch - 14us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.8877 - 661us/epoch - 11us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8033 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.6007 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4615 - 678us/epoch - 11us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.6500 - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6296 - 1ms/epoch - 23us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.2194 - 876us/epoch - 14us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.5216 - 725us/epoch - 12us/sample\n", "Episode 18/50\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.3659 - 896us/epoch - 14us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.5667 - 860us/epoch - 14us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 807.3879 - 4ms/epoch - 59us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.7298 - 915us/epoch - 15us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1703 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.6115 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.6635 - 708us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6868 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.5873 - 804us/epoch - 13us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.0616 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9073 - 840us/epoch - 14us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.9446 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.9332 - 996us/epoch - 16us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4061 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.1226 - 721us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.7963 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0031 - 823us/epoch - 13us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6935 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 807.0300 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.5897 - 827us/epoch - 13us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.2226 - 728us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.6176 - 865us/epoch - 14us/sample\n", "Episode 35/50\n", "Episode 36/50\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5704 - 602us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.8007 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.9477 - 686us/epoch - 11us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.2819 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.8423 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.6022 - 741us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.1427 - 712us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.8832 - 707us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.6189 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.2020 - 743us/epoch - 12us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.1687 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.6384 - 783us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.5304 - 691us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.6473 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.8103 - 631us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.2726 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5933 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.8349 - 824us/epoch - 13us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 793.9821 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.0389 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.9293 - 926us/epoch - 15us/sample\n", "Generation 58/300\n", "Solving for Nash Equilibrium in Generation 58/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.1500 - 613us/epoch - 10us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.7938 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.1365 - 686us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0504 - 690us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6926 - 785us/epoch - 13us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.1070 - 850us/epoch - 14us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.4188 - 770us/epoch - 12us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.0367 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.2542 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 789.0785 - 909us/epoch - 15us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.5716 - 710us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.5470 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.2444 - 755us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.6566 - 680us/epoch - 11us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6381 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6342 - 752us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.2625 - 3ms/epoch - 41us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.3897 - 2ms/epoch - 38us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.3531 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4878 - 980us/epoch - 16us/sample\n", "Episode 18/50\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 182.9919 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.1251 - 672us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5533 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.2834 - 693us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.9518 - 702us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.6893 - 627us/epoch - 10us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.7686 - 2ms/epoch - 38us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.9080 - 747us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.0689 - 1ms/epoch - 22us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9138 - 672us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.5871 - 808us/epoch - 13us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8997 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.5966 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9991 - 617us/epoch - 10us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.7317 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.4266 - 3ms/epoch - 40us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.4068 - 730us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3642 - 618us/epoch - 10us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0895 - 739us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.1663 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.9879 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6213 - 1ms/epoch - 17us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.0868 - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9562 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.4962 - 712us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.1445 - 707us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.4513 - 740us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.9327 - 691us/epoch - 11us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.4781 - 744us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7179 - 607us/epoch - 10us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 804.6009 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 807.7545 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.5089 - 796us/epoch - 13us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.3021 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.9892 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6328 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.0723 - 2ms/epoch - 29us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0939 - 636us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.0924 - 759us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.3780 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3519 - 996us/epoch - 16us/sample\n", "Episode 50/50\n", "Generation 59/300\n", "Solving for Nash Equilibrium in Generation 59/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.5131 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5907 - 2ms/epoch - 35us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.0864 - 801us/epoch - 13us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.7268 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.2415 - 733us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8264 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.5594 - 803us/epoch - 13us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.7899 - 658us/epoch - 11us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.2181 - 731us/epoch - 12us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.9117 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0776 - 738us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.8718 - 679us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.9130 - 701us/epoch - 11us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.9760 - 687us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.4361 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.5684 - 734us/epoch - 12us/sample\n", "Episode 14/50\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.0126 - 739us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8608 - 746us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.8531 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.9275 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.9401 - 996us/epoch - 16us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.0888 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4246 - 743us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.9953 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.3747 - 638us/epoch - 10us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.1567 - 731us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7922 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.4205 - 920us/epoch - 15us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1790 - 576us/epoch - 9us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.1247 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0798 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.7524 - 626us/epoch - 10us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.1499 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.9423 - 735us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.9561 - 875us/epoch - 14us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.7455 - 763us/epoch - 12us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.7125 - 724us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.9865 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.0724 - 3ms/epoch - 53us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.8376 - 759us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.0220 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.1039 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.8362 - 732us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.0882 - 597us/epoch - 10us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4970 - 728us/epoch - 12us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.8777 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.8535 - 647us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.0272 - 680us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2853 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6870 - 643us/epoch - 10us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.6331 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.9462 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6578 - 810us/epoch - 13us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.8478 - 644us/epoch - 10us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0054 - 696us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.7913 - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8944 - 720us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.3781 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.4154 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.0782 - 810us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2570 - 615us/epoch - 10us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.1175 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1938 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.5600 - 684us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.0685 - 698us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.2177 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7156 - 745us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.3800 - 734us/epoch - 12us/sample\n", "Generation 60/300\n", "Solving for Nash Equilibrium in Generation 60/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.6757 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.3060 - 701us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.0305 - 660us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8404 - 743us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.7472 - 678us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.0017 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.6283 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.2873 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9687 - 696us/epoch - 11us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5409 - 756us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.9315 - 769us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.7784 - 618us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6849 - 641us/epoch - 10us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.5955 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.1235 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.2994 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.6956 - 788us/epoch - 13us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8890 - 699us/epoch - 11us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6507 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4409 - 833us/epoch - 13us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.4504 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3089 - 718us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.1061 - 839us/epoch - 14us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.8264 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.0430 - 601us/epoch - 10us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.1313 - 635us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6059 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.5966 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2227 - 755us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.5138 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.3504 - 719us/epoch - 12us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9402 - 1ms/epoch - 18us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7834 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.6085 - 663us/epoch - 11us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3464 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.6224 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.7613 - 667us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.4650 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7260 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.4409 - 1ms/epoch - 17us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.1946 - 726us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9704 - 603us/epoch - 10us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.6341 - 608us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773.5679 - 990us/epoch - 16us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.4769 - 721us/epoch - 12us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3903 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6096 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.9958 - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.1532 - 925us/epoch - 15us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2065 - 858us/epoch - 14us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.8108 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3556 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8783 - 644us/epoch - 10us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.7318 - 716us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6166 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.7011 - 814us/epoch - 13us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8531 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6550 - 741us/epoch - 12us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.5761 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.7528 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.7641 - 590us/epoch - 10us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.4518 - 6ms/epoch - 104us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9800 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.9774 - 594us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.2538 - 600us/epoch - 10us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8292 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.8572 - 656us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.2528 - 642us/epoch - 10us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.2105 - 839us/epoch - 14us/sample\n", "Generation 61/300\n", "Solving for Nash Equilibrium in Generation 61/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.7251 - 541us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.1810 - 3ms/epoch - 44us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.2023 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.7323 - 714us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.7943 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4670 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.5269 - 906us/epoch - 15us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.2735 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.5995 - 872us/epoch - 14us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.7250 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.0680 - 1ms/epoch - 23us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.3471 - 713us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9235 - 740us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.4024 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8537 - 635us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.6068 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.4308 - 695us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.1138 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.6276 - 682us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3879 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.2841 - 2ms/epoch - 38us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.5661 - 798us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3647 - 669us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.3416 - 610us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.6175 - 589us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.3093 - 691us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.9371 - 603us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.6797 - 684us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5643 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4836 - 716us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.9807 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.6284 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.3297 - 793us/epoch - 13us/sample\n", "Episode 21/50\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.8131 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2550 - 664us/epoch - 11us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.1014 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.7250 - 707us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.9880 - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.1934 - 902us/epoch - 15us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.9202 - 753us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1190 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.3007 - 757us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.4879 - 706us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6617 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.5047 - 663us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.5679 - 754us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5977 - 689us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8434 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.0037 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 140us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.4532 - 3ms/epoch - 47us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.9954 - 980us/epoch - 16us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.8610 - 617us/epoch - 10us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.1562 - 613us/epoch - 10us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.1514 - 1ms/epoch - 18us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.0481 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5223 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.3845 - 716us/epoch - 12us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6472 - 654us/epoch - 11us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.2889 - 987us/epoch - 16us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.6881 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.0677 - 664us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.1433 - 786us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.3292 - 2ms/epoch - 33us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.4373 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6526 - 781us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.5012 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4056 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.9613 - 695us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.8322 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7823 - 742us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5410 - 746us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.0482 - 748us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.5266 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.5385 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.0778 - 690us/epoch - 11us/sample\n", "Episode 50/50\n", "Generation 62/300\n", "Solving for Nash Equilibrium in Generation 62/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.7554 - 625us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.6479 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 799.7974 - 740us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.8608 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.5041 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.0787 - 960us/epoch - 15us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.1552 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.6820 - 2ms/epoch - 31us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.8202 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.9957 - 679us/epoch - 11us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.8352 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.9331 - 649us/epoch - 10us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.8988 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.7204 - 761us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9255 - 659us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.1948 - 806us/epoch - 13us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.3081 - 755us/epoch - 12us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6645 - 641us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.1309 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.6594 - 750us/epoch - 12us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3035 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.2286 - 629us/epoch - 10us/sample\n", "Episode 18/50\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.5324 - 837us/epoch - 13us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 220.9937 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.8511 - 1ms/epoch - 20us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.4542 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.9780 - 703us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.2650 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.5219 - 786us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.0637 - 2ms/epoch - 26us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.8575 - 776us/epoch - 13us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.1391 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.3874 - 861us/epoch - 14us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.0344 - 668us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.7752 - 762us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1100 - 2ms/epoch - 38us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.1487 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.9847 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.2246 - 1ms/epoch - 18us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.4290 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8651 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.8530 - 689us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8689 - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.9985 - 826us/epoch - 13us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.0700 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1310 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.9058 - 685us/epoch - 11us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.3054 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.3806 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.7195 - 1ms/epoch - 18us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.4751 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.0672 - 822us/epoch - 13us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7327 - 713us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.6273 - 658us/epoch - 11us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.5767 - 728us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.3164 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.9849 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9075 - 734us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.2679 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.3714 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3593 - 693us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.2051 - 807us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.7416 - 826us/epoch - 13us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.6604 - 682us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0896 - 795us/epoch - 13us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4950 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.1592 - 1ms/epoch - 19us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0034 - 738us/epoch - 12us/sample\n", "Generation 63/300\n", "Solving for Nash Equilibrium in Generation 63/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 802.5117 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.1006 - 731us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.7072 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7330 - 599us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.6096 - 803us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.6157 - 618us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.1640 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.9666 - 1ms/epoch - 19us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.7425 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.5564 - 705us/epoch - 11us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9682 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.8477 - 610us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.6345 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.3244 - 708us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7457 - 764us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.4036 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5513 - 730us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6212 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8348 - 602us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.6331 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.6602 - 716us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.4398 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.9180 - 672us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.0989 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2953 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.9130 - 889us/epoch - 14us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.0343 - 790us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.0361 - 852us/epoch - 14us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9805 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.9151 - 703us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46.0870 - 1ms/epoch - 21us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.5705 - 627us/epoch - 10us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6523 - 615us/epoch - 10us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.9136 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.8909 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.8953 - 2ms/epoch - 36us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.1008 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6516 - 672us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.8891 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.2758 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.3293 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.4014 - 717us/epoch - 12us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0168 - 764us/epoch - 12us/sample\n", "Episode 32/50\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.7087 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.3239 - 863us/epoch - 14us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.9660 - 627us/epoch - 10us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40.4836 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.1378 - 806us/epoch - 13us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.4153 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.3253 - 663us/epoch - 11us/sample\n", "Episode 39/50\n", "Episode 40/50\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.0573 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.2132 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.1150 - 693us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.8189 - 1ms/epoch - 21us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.3574 - 718us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.7000 - 711us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.7475 - 827us/epoch - 13us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.2325 - 683us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.9664 - 4ms/epoch - 71us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.9780 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.8726 - 877us/epoch - 14us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.5656 - 1ms/epoch - 18us/sample\n", "Generation 64/300\n", "Solving for Nash Equilibrium in Generation 64/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.0295 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.6749 - 882us/epoch - 14us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.8115 - 2ms/epoch - 29us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2146 - 790us/epoch - 13us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.2930 - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9389 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.9182 - 1ms/epoch - 16us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37.8259 - 764us/epoch - 12us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9921 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.2180 - 801us/epoch - 13us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.5466 - 741us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.5222 - 871us/epoch - 14us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2661 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5339 - 707us/epoch - 11us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1447 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2593 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.8569 - 690us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.9239 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787.6362 - 715us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.8709 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.4231 - 644us/epoch - 10us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.7897 - 793us/epoch - 13us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.3124 - 731us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9694 - 696us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.0721 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.2693 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.8114 - 3ms/epoch - 47us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0233 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0452 - 956us/epoch - 15us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2521 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.9451 - 1ms/epoch - 16us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.8198 - 711us/epoch - 11us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.2471 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.0418 - 743us/epoch - 12us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0341 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.7424 - 792us/epoch - 13us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.5196 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2174 - 872us/epoch - 14us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.3352 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.2997 - 599us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.7678 - 671us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.7222 - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.8535 - 2ms/epoch - 35us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.8158 - 793us/epoch - 13us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 794.9575 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.9971 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3257 - 737us/epoch - 12us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2285 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4554 - 809us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2497 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3333 - 957us/epoch - 15us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.6058 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.9669 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8562 - 779us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.9593 - 775us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7062 - 940us/epoch - 15us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.3000 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.7031 - 824us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.3538 - 817us/epoch - 13us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.5091 - 861us/epoch - 14us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69.2074 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4684 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.1437 - 758us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3979 - 725us/epoch - 12us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.1000 - 800us/epoch - 13us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6873 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.5808 - 590us/epoch - 10us/sample\n", "Generation 65/300\n", "Solving for Nash Equilibrium in Generation 65/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.2402 - 757us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71.8484 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.1602 - 1ms/epoch - 18us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.3596 - 2ms/epoch - 33us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0782 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.4333 - 845us/epoch - 14us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.1288 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2651 - 802us/epoch - 13us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.6649 - 918us/epoch - 15us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.1548 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.5735 - 680us/epoch - 11us/sample\n", "Episode 8/50\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.9758 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.6795 - 798us/epoch - 13us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9955 - 752us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2669 - 831us/epoch - 13us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2437 - 6ms/epoch - 97us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 212.4832 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.1641 - 776us/epoch - 13us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.3124 - 823us/epoch - 13us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.2852 - 2ms/epoch - 24us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.1216 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8245 - 1ms/epoch - 17us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.6360 - 892us/epoch - 14us/sample\n", "Episode 20/50\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.8804 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.6728 - 607us/epoch - 10us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.7578 - 640us/epoch - 10us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.7130 - 589us/epoch - 10us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.7159 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.5715 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.9200 - 718us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.0706 - 729us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.4265 - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.7908 - 2ms/epoch - 35us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.5115 - 854us/epoch - 14us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.7897 - 697us/epoch - 11us/sample\n", "Episode 30/50\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.8876 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47.3134 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.4888 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.5062 - 590us/epoch - 10us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.2958 - 728us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3038 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.9363 - 2ms/epoch - 30us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9899 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.0220 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.3607 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.2912 - 811us/epoch - 13us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.7061 - 735us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0530 - 779us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3010 - 591us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 801.7679 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7468 - 655us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.6034 - 687us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.9872 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6693 - 802us/epoch - 13us/sample\n", "Episode 43/50\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.3762 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6598 - 684us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.7908 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9838 - 892us/epoch - 14us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6026 - 679us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.5119 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.8152 - 585us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.6450 - 702us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.9058 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3786 - 652us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.2703 - 610us/epoch - 10us/sample\n", "Generation 66/300\n", "Solving for Nash Equilibrium in Generation 66/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.5114 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.4311 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.8403 - 780us/epoch - 13us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.8781 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7324 - 2ms/epoch - 25us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.4633 - 904us/epoch - 15us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0307 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5612 - 613us/epoch - 10us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2991 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0490 - 615us/epoch - 10us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2268 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.6479 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.1537 - 802us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.2244 - 654us/epoch - 11us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.2924 - 719us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 207.3873 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.1725 - 822us/epoch - 13us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.8580 - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.2113 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.0280 - 642us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1725 - 621us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231.5875 - 694us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.0150 - 800us/epoch - 13us/sample\n", "Episode 18/50\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.1550 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.0507 - 1ms/epoch - 17us/sample\n", "Episode 20/50\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.4995 - 687us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.6403 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5681 - 876us/epoch - 14us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7815 - 623us/epoch - 10us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4786 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6904 - 696us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49.3799 - 694us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3295 - 676us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.0578 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.0023 - 632us/epoch - 10us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.3809 - 695us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7524 - 625us/epoch - 10us/sample\n", "Episode 31/50\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.2490 - 1ms/epoch - 21us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.4185 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.0362 - 641us/epoch - 10us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.4513 - 901us/epoch - 15us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.8498 - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.0962 - 824us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.5267 - 699us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7496 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.0643 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7366 - 596us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0107 - 769us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.0036 - 725us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6263 - 565us/epoch - 9us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4583 - 766us/epoch - 12us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.3962 - 968us/epoch - 16us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.7256 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.5136 - 817us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.2961 - 759us/epoch - 12us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0129 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.6416 - 846us/epoch - 14us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2170 - 1ms/epoch - 19us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2162 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6307 - 691us/epoch - 11us/sample\n", "Generation 67/300\n", "Solving for Nash Equilibrium in Generation 67/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.5282 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1906 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8990 - 670us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.6146 - 681us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.3524 - 672us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9954 - 699us/epoch - 11us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.5682 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0036 - 893us/epoch - 14us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5703 - 774us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.0388 - 650us/epoch - 10us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.3853 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6551 - 672us/epoch - 11us/sample\n", "Episode 10/50\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.5921 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5415 - 8ms/epoch - 122us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8099 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4253 - 1ms/epoch - 17us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8807 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9260 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4942 - 4ms/epoch - 59us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.9557 - 828us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8212 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.8737 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4595 - 764us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.4752 - 853us/epoch - 14us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0630 - 636us/epoch - 10us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.5901 - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.8362 - 1ms/epoch - 17us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.4966 - 872us/epoch - 14us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.0627 - 805us/epoch - 13us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.0209 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.9828 - 913us/epoch - 15us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.5270 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8613 - 637us/epoch - 10us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.6197 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5488 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.5212 - 778us/epoch - 13us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.4560 - 949us/epoch - 15us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5876 - 714us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2417 - 882us/epoch - 14us/sample\n", "Episode 31/50\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5278 - 716us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.8551 - 746us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3337 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.6678 - 979us/epoch - 16us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5604 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.6168 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7804 - 864us/epoch - 14us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.1343 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0620 - 679us/epoch - 11us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.0866 - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2556 - 748us/epoch - 12us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1137 - 870us/epoch - 14us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.5232 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.4701 - 1ms/epoch - 18us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.6460 - 814us/epoch - 13us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.6245 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.9498 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.6430 - 864us/epoch - 14us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.4061 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.8134 - 748us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4308 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.4542 - 814us/epoch - 13us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6368 - 734us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7438 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.2383 - 11ms/epoch - 183us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.6031 - 853us/epoch - 14us/sample\n", "Generation 68/300\n", "Solving for Nash Equilibrium in Generation 68/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.8250 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.0350 - 774us/epoch - 12us/sample\n", "Episode 2/50\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.4780 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.1358 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.1474 - 587us/epoch - 9us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.1301 - 679us/epoch - 11us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.4761 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.9233 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.2745 - 716us/epoch - 12us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3417 - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5966 - 910us/epoch - 15us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.8047 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.4119 - 826us/epoch - 13us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.4598 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7578 - 822us/epoch - 13us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.4666 - 765us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.2145 - 718us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.5242 - 2ms/epoch - 25us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.7475 - 658us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5820 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.2590 - 636us/epoch - 10us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.9385 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.4938 - 768us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.0448 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.4565 - 626us/epoch - 10us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8645 - 2ms/epoch - 37us/sample\n", "Episode 20/50\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.9808 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4623 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.0452 - 1ms/epoch - 23us/sample\n", "Episode 22/50\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.9641 - 1ms/epoch - 16us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8383 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.7718 - 1ms/epoch - 23us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.6703 - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.1434 - 732us/epoch - 12us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.0189 - 1ms/epoch - 18us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.6354 - 805us/epoch - 13us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.0648 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.8480 - 2ms/epoch - 25us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.6755 - 2ms/epoch - 36us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.8312 - 1ms/epoch - 16us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.1890 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2435 - 870us/epoch - 14us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.5899 - 774us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8706 - 1ms/epoch - 21us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 800.8709 - 630us/epoch - 10us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.8259 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.3193 - 630us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.1672 - 650us/epoch - 10us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5009 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.1349 - 577us/epoch - 9us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.0771 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.5123 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.3786 - 657us/epoch - 11us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.3374 - 594us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.0894 - 4ms/epoch - 67us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.0167 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.2909 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.0758 - 1ms/epoch - 20us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.1093 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.1813 - 914us/epoch - 15us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9904 - 817us/epoch - 13us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.7030 - 643us/epoch - 10us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1362 - 605us/epoch - 10us/sample\n", "Generation 69/300\n", "Solving for Nash Equilibrium in Generation 69/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.1128 - 882us/epoch - 14us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9399 - 727us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.3757 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9544 - 750us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41.7694 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.1816 - 643us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.6245 - 655us/epoch - 11us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9869 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.9749 - 920us/epoch - 15us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.3346 - 821us/epoch - 13us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.2771 - 716us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.0699 - 717us/epoch - 12us/sample\n", "Episode 11/50\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126.3991 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.8879 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3710 - 591us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.3228 - 678us/epoch - 11us/sample\n", "Episode 15/50\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.4251 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.1003 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6071 - 863us/epoch - 14us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.2558 - 752us/epoch - 12us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.2299 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0468 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.0482 - 840us/epoch - 14us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4508 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.5591 - 673us/epoch - 11us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.0963 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0669 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.1008 - 2ms/epoch - 30us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5688 - 665us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.5284 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.0725 - 629us/epoch - 10us/sample\n", "Episode 28/50\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56.1549 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.7449 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9911 - 812us/epoch - 13us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.3850 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9618 - 778us/epoch - 13us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.4670 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5490 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6272 - 2ms/epoch - 39us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8854 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.5868 - 721us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.2914 - 729us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.7276 - 705us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.2319 - 773us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.1328 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.8786 - 848us/epoch - 14us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.2905 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.4934 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.2101 - 841us/epoch - 14us/sample\n", "Episode 41/50\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5549 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.8348 - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.1443 - 650us/epoch - 10us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.6163 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.7079 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8452 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.2083 - 784us/epoch - 13us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.7711 - 688us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6600 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.7374 - 27ms/epoch - 431us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.0656 - 1ms/epoch - 22us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.4540 - 1ms/epoch - 18us/sample\n", "Generation 70/300\n", "Solving for Nash Equilibrium in Generation 70/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.9741 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.3814 - 911us/epoch - 15us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.0554 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.0723 - 742us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.2196 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.6604 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.4261 - 1ms/epoch - 21us/sample\n", "Episode 5/50\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 54ms/epoch - 866us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.7111 - 2ms/epoch - 34us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3920 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4545 - 756us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.0933 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7814 - 774us/epoch - 12us/sample\n", "Episode 10/50\n", "Episode 11/50\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4870 - 904us/epoch - 15us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.5032 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.6508 - 1ms/epoch - 17us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.8209 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0283 - 996us/epoch - 16us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.5032 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.5457 - 2ms/epoch - 40us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.9409 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.7580 - 757us/epoch - 12us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.1947 - 927us/epoch - 15us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5984 - 8ms/epoch - 129us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.6339 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.9779 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.0443 - 812us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7806 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.8979 - 4ms/epoch - 64us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.9884 - 767us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.2705 - 3ms/epoch - 56us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.5014 - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 803.7645 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 799.7155 - 909us/epoch - 15us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.7016 - 810us/epoch - 13us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3284 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.5872 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.9142 - 825us/epoch - 13us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.9450 - 2ms/epoch - 32us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.8015 - 887us/epoch - 14us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 18ms/epoch - 293us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.1164 - 5ms/epoch - 78us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.7359 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9363 - 917us/epoch - 15us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3127 - 1ms/epoch - 19us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406.0668 - 1ms/epoch - 19us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.8085 - 2ms/epoch - 30us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.6993 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.6807 - 1ms/epoch - 17us/sample\n", "Episode 38/50\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6731 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0592 - 6ms/epoch - 96us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.1508 - 724us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.0376 - 6ms/epoch - 92us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8214 - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.7108 - 1ms/epoch - 18us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.7008 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.8746 - 1ms/epoch - 21us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.7926 - 1ms/epoch - 21us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773.9818 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1892 - 1ms/epoch - 16us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.9480 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.8678 - 1ms/epoch - 21us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.0556 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.8638 - 6ms/epoch - 104us/sample\n", "Episode 49/50\n", "Episode 50/50\n", "Generation 71/300\n", "Solving for Nash Equilibrium in Generation 71/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6305 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.7406 - 1ms/epoch - 21us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.6961 - 909us/epoch - 15us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.4698 - 1ms/epoch - 18us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5854 - 1ms/epoch - 20us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0524 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8482 - 1ms/epoch - 19us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0466 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.4611 - 728us/epoch - 12us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.1215 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.2608 - 755us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.8028 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.4905 - 928us/epoch - 15us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.6478 - 955us/epoch - 15us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.8094 - 848us/epoch - 14us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1895 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.2137 - 706us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.9355 - 693us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.7914 - 866us/epoch - 14us/sample\n", "Episode 18/50\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4783 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.9876 - 807us/epoch - 13us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3670 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.5050 - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.2856 - 668us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3950 - 758us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 806.3345 - 820us/epoch - 13us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7782 - 846us/epoch - 14us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.6641 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0669 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.3205 - 662us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.0758 - 683us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6502 - 739us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.9573 - 868us/epoch - 14us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4468 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.3314 - 688us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9697 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.9341 - 791us/epoch - 13us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.7229 - 695us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.6786 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.3323 - 743us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5822 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7498 - 666us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8239 - 692us/epoch - 11us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.9978 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.8005 - 1ms/epoch - 20us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.3096 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.3129 - 1ms/epoch - 19us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.1725 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.3698 - 979us/epoch - 16us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.7833 - 942us/epoch - 15us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.0181 - 1ms/epoch - 22us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.5685 - 778us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.1626 - 1ms/epoch - 16us/sample\n", "Episode 46/50\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7565 - 918us/epoch - 15us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.4909 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.3178 - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.8432 - 782us/epoch - 13us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.5750 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.8746 - 855us/epoch - 14us/sample\n", "Generation 72/300\n", "Solving for Nash Equilibrium in Generation 72/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.5053 - 759us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.5567 - 1ms/epoch - 20us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.2887 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5159 - 849us/epoch - 14us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1779 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.0967 - 662us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.5081 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9100 - 728us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.3262 - 915us/epoch - 15us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6288 - 925us/epoch - 15us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.6467 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9185 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.4797 - 684us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.9837 - 712us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.6099 - 711us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4443 - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.2914 - 779us/epoch - 13us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.7394 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.6957 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.6807 - 693us/epoch - 11us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9393 - 2ms/epoch - 36us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8401 - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.6498 - 693us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 804.6340 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.9417 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.3043 - 742us/epoch - 12us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.6852 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.3791 - 639us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.2238 - 636us/epoch - 10us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 790.1090 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7563 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2245 - 707us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.0284 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.6176 - 801us/epoch - 13us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.5309 - 659us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.7189 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0806 - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.2440 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0335 - 633us/epoch - 10us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.7769 - 1ms/epoch - 23us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9538 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.5827 - 655us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.7420 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.1366 - 724us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.2197 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.4673 - 650us/epoch - 10us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.3118 - 743us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8501 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5037 - 738us/epoch - 12us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7878 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.9489 - 783us/epoch - 13us/sample\n", "Episode 32/50\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.9772 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.0781 - 815us/epoch - 13us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9788 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0171 - 602us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9391 - 779us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.6984 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.2481 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.0085 - 600us/epoch - 10us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.3718 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.2974 - 709us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.6495 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.8334 - 2ms/epoch - 37us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6469 - 770us/epoch - 12us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.8560 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.6281 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.7849 - 833us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8863 - 674us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 791.3621 - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.7067 - 746us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.4893 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.9431 - 801us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.1368 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.1479 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.2766 - 686us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.5396 - 648us/epoch - 10us/sample\n", "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.2990 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4063 - 686us/epoch - 11us/sample\n", "Generation 73/300\n", "Solving for Nash Equilibrium in Generation 73/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.2094 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5066 - 921us/epoch - 15us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.5592 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.4949 - 1ms/epoch - 21us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.8654 - 728us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.7436 - 2ms/epoch - 26us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.6358 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.1981 - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.1535 - 2ms/epoch - 29us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.9330 - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.7718 - 821us/epoch - 13us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.1041 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6986 - 683us/epoch - 11us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.8472 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.0294 - 696us/epoch - 11us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8116 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1620 - 645us/epoch - 10us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.8947 - 898us/epoch - 14us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.2112 - 796us/epoch - 13us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.0482 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1763 - 830us/epoch - 13us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.2845 - 1ms/epoch - 18us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.7066 - 686us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.2841 - 760us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.7783 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.3752 - 721us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.1413 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.6613 - 631us/epoch - 10us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.3959 - 790us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3855 - 770us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4558 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.0145 - 902us/epoch - 15us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6727 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.0286 - 719us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.1208 - 3ms/epoch - 42us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.0004 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.8607 - 756us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.5568 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.0533 - 749us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5554 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.4935 - 2ms/epoch - 38us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.9092 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.3028 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.3864 - 682us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.1240 - 675us/epoch - 11us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.9954 - 740us/epoch - 12us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.3563 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3462 - 707us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6881 - 677us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.0093 - 683us/epoch - 11us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.5904 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.4748 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.9175 - 848us/epoch - 14us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.6976 - 2ms/epoch - 34us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.1230 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.6363 - 811us/epoch - 13us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.1661 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.2361 - 1ms/epoch - 21us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1703 - 709us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.5640 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.3378 - 2ms/epoch - 35us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.0715 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7499 - 946us/epoch - 15us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.5826 - 750us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9673 - 1ms/epoch - 21us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7618 - 709us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.0556 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.2436 - 754us/epoch - 12us/sample\n", "Generation 74/300\n", "Solving for Nash Equilibrium in Generation 74/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48.0491 - 760us/epoch - 12us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.1284 - 798us/epoch - 13us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.2248 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.6570 - 4ms/epoch - 68us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 783.4833 - 861us/epoch - 14us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2489 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5002 - 961us/epoch - 16us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.8469 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.3217 - 580us/epoch - 9us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.4223 - 698us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0698 - 807us/epoch - 13us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0614 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4437 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.8572 - 574us/epoch - 9us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7516 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.9236 - 598us/epoch - 10us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.8011 - 783us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.9297 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.7622 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.2981 - 659us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.9069 - 665us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 811.1662 - 686us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.8369 - 761us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.9968 - 896us/epoch - 14us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.1556 - 746us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.4764 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.4064 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4979 - 651us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.4341 - 699us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5915 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.5341 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4550 - 670us/epoch - 11us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0749 - 691us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.3070 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.2536 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.8247 - 819us/epoch - 13us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.7612 - 683us/epoch - 11us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.8873 - 563us/epoch - 9us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6604 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6432 - 645us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8819 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.1456 - 820us/epoch - 13us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.8811 - 816us/epoch - 13us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.3269 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1019 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.1635 - 744us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.8911 - 691us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.1199 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.9750 - 611us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.6437 - 753us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7410 - 718us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.9041 - 570us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.6608 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6478 - 875us/epoch - 14us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.3818 - 675us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.0755 - 775us/epoch - 13us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.4553 - 738us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.9329 - 603us/epoch - 10us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3015 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.9797 - 647us/epoch - 10us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.2042 - 756us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8534 - 785us/epoch - 13us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.8481 - 810us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.2841 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2957 - 952us/epoch - 15us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.0197 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.1630 - 677us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.1687 - 656us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.0962 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.8346 - 1ms/epoch - 19us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.4413 - 713us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.3781 - 822us/epoch - 13us/sample\n", "Generation 75/300\n", "Solving for Nash Equilibrium in Generation 75/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.2628 - 671us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2708 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.8020 - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9324 - 719us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.9388 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6281 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.1135 - 763us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.1994 - 820us/epoch - 13us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.8116 - 685us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.8471 - 753us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.0015 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.4683 - 725us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.7115 - 4ms/epoch - 68us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5573 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3725 - 649us/epoch - 10us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.6275 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8914 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.3625 - 830us/epoch - 13us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.5027 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8388 - 755us/epoch - 12us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.6441 - 800us/epoch - 13us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.9155 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.8867 - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.2001 - 732us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.6640 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.0334 - 807us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.0670 - 830us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.5926 - 669us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.9399 - 709us/epoch - 11us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.6152 - 738us/epoch - 12us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.6783 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.2541 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.5641 - 685us/epoch - 11us/sample\n", "Episode 25/50\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.1795 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.5128 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2183 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.7728 - 750us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.0217 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.6608 - 822us/epoch - 13us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7468 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.7562 - 725us/epoch - 12us/sample\n", "Episode 30/50\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.0609 - 660us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.4414 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5686 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.6334 - 653us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0825 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.0711 - 685us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4880 - 812us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.9589 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.7266 - 765us/epoch - 12us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.2651 - 2ms/epoch - 30us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.3802 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.3124 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.2159 - 906us/epoch - 15us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3375 - 818us/epoch - 13us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.2030 - 688us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.9839 - 847us/epoch - 14us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6901 - 1ms/epoch - 24us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.6227 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.4609 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.4918 - 626us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.0080 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6849 - 696us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2719 - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.8518 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5783 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.9722 - 756us/epoch - 12us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.7801 - 569us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2341 - 687us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0903 - 784us/epoch - 13us/sample\n", "Generation 76/300\n", "Solving for Nash Equilibrium in Generation 76/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.4407 - 686us/epoch - 11us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3347 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3293 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6467 - 746us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.8574 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.5826 - 674us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.3859 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3055 - 753us/epoch - 12us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.9584 - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.0901 - 653us/epoch - 11us/sample\n", "Episode 7/50\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9819 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.4787 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.9698 - 649us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.8061 - 772us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.5392 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.6412 - 616us/epoch - 10us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.1870 - 653us/epoch - 11us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.2006 - 653us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.3890 - 718us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.9097 - 747us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.5861 - 700us/epoch - 11us/sample\n", "Episode 18/50\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.5155 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6136 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.0120 - 703us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.7364 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.8190 - 702us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.5174 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.7531 - 643us/epoch - 10us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.7288 - 648us/epoch - 10us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.3954 - 1ms/epoch - 20us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3903 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 435.9763 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.5252 - 748us/epoch - 12us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.6418 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.2446 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0526 - 788us/epoch - 13us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.8890 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.8549 - 2ms/epoch - 27us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.1913 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.5025 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.8699 - 738us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0690 - 841us/epoch - 14us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.7586 - 689us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.9360 - 869us/epoch - 14us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.7614 - 800us/epoch - 13us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7154 - 856us/epoch - 14us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.6530 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.9869 - 900us/epoch - 15us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.5621 - 731us/epoch - 12us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.1891 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9515 - 751us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.4523 - 700us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1835 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.2552 - 764us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.9663 - 743us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.9693 - 691us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.9843 - 672us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3596 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7147 - 2ms/epoch - 35us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0475 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.2766 - 766us/epoch - 12us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2550 - 597us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7541 - 2ms/epoch - 24us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3126 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5491 - 1ms/epoch - 16us/sample\n", "Episode 50/50\n", "Generation 77/300\n", "Solving for Nash Equilibrium in Generation 77/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8134 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9963 - 1ms/epoch - 23us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0107 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.5839 - 980us/epoch - 16us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9765 - 1ms/epoch - 17us/sample\n", "Episode 4/50\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.0083 - 715us/epoch - 12us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77.3197 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.0084 - 686us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.7835 - 4ms/epoch - 64us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9498 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2799 - 719us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.0419 - 720us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.2688 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.2560 - 616us/epoch - 10us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.0174 - 731us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.8618 - 771us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.0303 - 903us/epoch - 15us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.7554 - 729us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2084 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1752 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4308 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.2365 - 692us/epoch - 11us/sample\n", "Episode 18/50\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.5826 - 685us/epoch - 11us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.1326 - 3ms/epoch - 55us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.4071 - 676us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.0930 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.8046 - 583us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.1161 - 617us/epoch - 10us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8201 - 687us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.0074 - 681us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.6536 - 740us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.7934 - 590us/epoch - 10us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.0760 - 2ms/epoch - 27us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.5867 - 643us/epoch - 10us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.4037 - 756us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56.1389 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.0069 - 744us/epoch - 12us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.8519 - 673us/epoch - 11us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.8445 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.0262 - 728us/epoch - 12us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.3881 - 711us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.4524 - 1ms/epoch - 24us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.4133 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.5421 - 919us/epoch - 15us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.7108 - 780us/epoch - 13us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4157 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.4206 - 3ms/epoch - 55us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0332 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5782 - 1ms/epoch - 16us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9806 - 722us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2944 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4782 - 665us/epoch - 11us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.4537 - 4ms/epoch - 70us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5320 - 781us/epoch - 13us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9700 - 715us/epoch - 12us/sample\n", "Generation 78/300\n", "Solving for Nash Equilibrium in Generation 78/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.3754 - 1ms/epoch - 16us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4613 - 706us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.0143 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3081 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.0164 - 638us/epoch - 10us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.5447 - 674us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 807.3364 - 746us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.8557 - 2ms/epoch - 25us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.5640 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 820.1954 - 529us/epoch - 9us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.8632 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.7536 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.1880 - 754us/epoch - 12us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.0292 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3768 - 853us/epoch - 14us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2750 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.5072 - 2ms/epoch - 37us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.9825 - 718us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8600 - 732us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.2904 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.6808 - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4630 - 1ms/epoch - 20us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.4016 - 723us/epoch - 12us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2106 - 959us/epoch - 15us/sample\n", "Episode 16/50\n", "Episode 17/50\n", "Episode 18/50\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.4232 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.0284 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.8828 - 715us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.5212 - 658us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 64.1619 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.5105 - 6ms/epoch - 92us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.4575 - 1ms/epoch - 18us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.5687 - 714us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6216 - 607us/epoch - 10us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.5619 - 831us/epoch - 13us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6298 - 1ms/epoch - 17us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46.8875 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.2455 - 680us/epoch - 11us/sample\n", "Episode 29/50\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.2096 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0563 - 1ms/epoch - 24us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.1860 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8238 - 732us/epoch - 12us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.9307 - 635us/epoch - 10us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8992 - 4ms/epoch - 64us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8195 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.1174 - 673us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8364 - 838us/epoch - 14us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9897 - 794us/epoch - 13us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.3766 - 699us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.1649 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.1590 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.8605 - 780us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.3306 - 794us/epoch - 13us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5603 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9167 - 839us/epoch - 14us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.1497 - 9ms/epoch - 138us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.6263 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.9615 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.1713 - 724us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.3064 - 925us/epoch - 15us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1835 - 4ms/epoch - 66us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.4405 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4911 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.9481 - 829us/epoch - 13us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.7599 - 758us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.9841 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.5905 - 791us/epoch - 13us/sample\n", "Generation 79/300\n", "Solving for Nash Equilibrium in Generation 79/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.7911 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.1807 - 668us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.0712 - 781us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.7535 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.6098 - 867us/epoch - 14us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.0101 - 919us/epoch - 15us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 793.4785 - 711us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.2177 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.9320 - 992us/epoch - 16us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.5237 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1207 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7657 - 860us/epoch - 14us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.2800 - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0162 - 738us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.1030 - 731us/epoch - 12us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.7849 - 642us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.2972 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.6828 - 711us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6104 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.4760 - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.1774 - 725us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.9536 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.6476 - 1ms/epoch - 16us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.3865 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.6132 - 908us/epoch - 15us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.7670 - 861us/epoch - 14us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9314 - 1ms/epoch - 17us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.3925 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.3716 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.0220 - 690us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.2542 - 704us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.2252 - 824us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.6958 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.4832 - 674us/epoch - 11us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.6575 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7066 - 747us/epoch - 12us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.4593 - 954us/epoch - 15us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.5096 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7399 - 803us/epoch - 13us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.8434 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.6713 - 836us/epoch - 13us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.9949 - 704us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.0459 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1381 - 647us/epoch - 10us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.7845 - 731us/epoch - 12us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.8719 - 805us/epoch - 13us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.2849 - 808us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.4359 - 911us/epoch - 15us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.7353 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.6585 - 812us/epoch - 13us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0198 - 860us/epoch - 14us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.9441 - 852us/epoch - 14us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.7346 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2554 - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.0179 - 983us/epoch - 16us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.4445 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.8527 - 778us/epoch - 13us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9525 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.7543 - 1ms/epoch - 24us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1095 - 710us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9885 - 708us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.9234 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.6479 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.5327 - 1ms/epoch - 17us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.0651 - 1ms/epoch - 17us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.7172 - 738us/epoch - 12us/sample\n", "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.5421 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.4652 - 1ms/epoch - 17us/sample\n", "Generation 80/300\n", "Solving for Nash Equilibrium in Generation 80/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.6369 - 932us/epoch - 15us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.3639 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.6609 - 708us/epoch - 11us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.4705 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.7817 - 1ms/epoch - 21us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.2330 - 672us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.2034 - 866us/epoch - 14us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.7938 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.5814 - 932us/epoch - 15us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.7811 - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.1128 - 962us/epoch - 16us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6827 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.9689 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.2749 - 879us/epoch - 14us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1092 - 994us/epoch - 16us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8893 - 1ms/epoch - 16us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.0071 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4455 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4329 - 1ms/epoch - 18us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2465 - 798us/epoch - 13us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6986 - 844us/epoch - 14us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.4536 - 1ms/epoch - 21us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.3499 - 887us/epoch - 14us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.5086 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.0668 - 738us/epoch - 12us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.5803 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.6933 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2265 - 898us/epoch - 14us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9626 - 704us/epoch - 11us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.7313 - 828us/epoch - 13us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.7437 - 788us/epoch - 13us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.1374 - 759us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.7303 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.8606 - 598us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.6579 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.0576 - 753us/epoch - 12us/sample\n", "Episode 31/50\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 786.5776 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.9987 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.1889 - 1ms/epoch - 16us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.0806 - 999us/epoch - 16us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6168 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.2256 - 700us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.8377 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.4033 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.2678 - 3ms/epoch - 46us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.8253 - 663us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.0526 - 833us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.3547 - 666us/epoch - 11us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6246 - 667us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.1467 - 863us/epoch - 14us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.6768 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.8337 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.6703 - 701us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.2823 - 622us/epoch - 10us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 778.4406 - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.6027 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.1909 - 703us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.3891 - 755us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3055 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.1607 - 1ms/epoch - 17us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.8698 - 919us/epoch - 15us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.7971 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.8215 - 673us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.5561 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.4595 - 666us/epoch - 11us/sample\n", "Generation 81/300\n", "Solving for Nash Equilibrium in Generation 81/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.2418 - 631us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 785.4047 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.5913 - 760us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.3676 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.6848 - 709us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.5157 - 577us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2776 - 606us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1517 - 747us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.4199 - 851us/epoch - 14us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 145.0096 - 2ms/epoch - 40us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.6051 - 680us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.2248 - 2ms/epoch - 29us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2911 - 629us/epoch - 10us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.7797 - 683us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 785.3810 - 725us/epoch - 12us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3345 - 787us/epoch - 13us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.3134 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.4858 - 976us/epoch - 16us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.5034 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.6912 - 793us/epoch - 13us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.5916 - 589us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9349 - 587us/epoch - 9us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3542 - 779us/epoch - 13us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.1683 - 738us/epoch - 12us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.0356 - 690us/epoch - 11us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7327 - 668us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.8155 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.7905 - 655us/epoch - 11us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.6375 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.3457 - 856us/epoch - 14us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3061 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.6628 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.7371 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.3802 - 863us/epoch - 14us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1723 - 758us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.6351 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.4305 - 689us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.9387 - 760us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.9774 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.5417 - 735us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.3517 - 777us/epoch - 13us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6940 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5888 - 984us/epoch - 16us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.1985 - 609us/epoch - 10us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1999 - 749us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.4590 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.1125 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.8247 - 796us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.1700 - 762us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1958 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.1837 - 1ms/epoch - 24us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.0169 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.4282 - 821us/epoch - 13us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.0920 - 1ms/epoch - 24us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.2581 - 703us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.0576 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.0564 - 711us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.9568 - 5ms/epoch - 82us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.7172 - 712us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.9211 - 790us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.2206 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.3688 - 726us/epoch - 12us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.0544 - 700us/epoch - 11us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.5009 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.1606 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.6251 - 666us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79.7647 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.0040 - 1ms/epoch - 18us/sample\n", "Generation 82/300\n", "Solving for Nash Equilibrium in Generation 82/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9232 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4390 - 1ms/epoch - 17us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.8266 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.4656 - 758us/epoch - 12us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.8236 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.9765 - 615us/epoch - 10us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.3498 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.8890 - 620us/epoch - 10us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.1973 - 843us/epoch - 14us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.6341 - 721us/epoch - 12us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8191 - 762us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.8331 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.5062 - 875us/epoch - 14us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.4467 - 680us/epoch - 11us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.0230 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.9822 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3693 - 619us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.1105 - 686us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 797.4335 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7538 - 668us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.2763 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.6161 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.8027 - 710us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.2278 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.1310 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2691 - 732us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.0572 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.8770 - 857us/epoch - 14us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.0272 - 4ms/epoch - 57us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.7332 - 725us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.6906 - 796us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.1145 - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5983 - 704us/epoch - 11us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.0451 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4608 - 635us/epoch - 10us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.2203 - 827us/epoch - 13us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2310 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3712 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.3925 - 712us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5444 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.5480 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.1675 - 1ms/epoch - 19us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5911 - 994us/epoch - 16us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.3296 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.2162 - 768us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.6779 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.4553 - 789us/epoch - 13us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.7206 - 1ms/epoch - 17us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.7502 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.0411 - 647us/epoch - 10us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.1879 - 744us/epoch - 12us/sample\n", "Episode 38/50\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 789.9816 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7031 - 2ms/epoch - 32us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.5120 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.8569 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.7863 - 712us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.7277 - 775us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.2536 - 823us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4573 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9789 - 809us/epoch - 13us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.0938 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.5657 - 732us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.6698 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.2244 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7070 - 737us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.9500 - 772us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.8460 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.1102 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6539 - 599us/epoch - 10us/sample\n", "Generation 83/300\n", "Solving for Nash Equilibrium in Generation 83/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.6379 - 800us/epoch - 13us/sample\n", "Episode 2/50\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8861 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7258 - 2ms/epoch - 37us/sample\n", "Episode 5/50\n", "Episode 6/50\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.5572 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.4685 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3713 - 1ms/epoch - 16us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.4870 - 751us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.7439 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2332 - 674us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.3048 - 2ms/epoch - 25us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.2553 - 598us/epoch - 10us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.2668 - 630us/epoch - 10us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.4616 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3224 - 1ms/epoch - 24us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.2386 - 780us/epoch - 13us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.0238 - 975us/epoch - 16us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.1027 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4952 - 878us/epoch - 14us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4901 - 768us/epoch - 12us/sample\n", "Episode 20/50\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6964 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.3248 - 694us/epoch - 11us/sample\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.5714 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.2843 - 2ms/epoch - 27us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.6004 - 2ms/epoch - 29us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.7697 - 768us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.0326 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.8219 - 1ms/epoch - 19us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2159 - 6ms/epoch - 101us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.1151 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.3943 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.8362 - 3ms/epoch - 47us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.2721 - 862us/epoch - 14us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.1674 - 2ms/epoch - 26us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.0806 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.0991 - 678us/epoch - 11us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.7957 - 643us/epoch - 10us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.8950 - 2ms/epoch - 32us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9591 - 3ms/epoch - 42us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5023 - 664us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.0699 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5898 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.5411 - 755us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4192 - 690us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.5291 - 682us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1699 - 626us/epoch - 10us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.2170 - 1ms/epoch - 19us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.3255 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.4283 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1147 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7868 - 666us/epoch - 11us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9377 - 655us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 804.8691 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.0511 - 708us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.8044 - 661us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.7445 - 641us/epoch - 10us/sample\n", "Generation 84/300\n", "Solving for Nash Equilibrium in Generation 84/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.3036 - 790us/epoch - 13us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.8325 - 699us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.0711 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.8739 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5418 - 715us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.8934 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.5634 - 5ms/epoch - 77us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3469 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.8757 - 580us/epoch - 9us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.0635 - 729us/epoch - 12us/sample\n", "Episode 7/50\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.2381 - 758us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.1756 - 662us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7614 - 654us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.9677 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.5900 - 981us/epoch - 16us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.7361 - 747us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.6705 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.1593 - 600us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.0547 - 754us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.5889 - 768us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.8029 - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.8929 - 898us/epoch - 14us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.8825 - 707us/epoch - 11us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.0085 - 712us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.6207 - 547us/epoch - 9us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 216.8188 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9184 - 844us/epoch - 14us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2800 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2750 - 812us/epoch - 13us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.2169 - 2ms/epoch - 37us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1326 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.0797 - 633us/epoch - 10us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.2963 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.7892 - 704us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.5269 - 653us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.0995 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6254 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.1033 - 729us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.0606 - 878us/epoch - 14us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.6631 - 709us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.6970 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.7748 - 697us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.9002 - 730us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.3195 - 732us/epoch - 12us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.4822 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.8961 - 754us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.5448 - 776us/epoch - 13us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3965 - 880us/epoch - 14us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8342 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.6078 - 633us/epoch - 10us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.8306 - 699us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.0736 - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.7354 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.6924 - 656us/epoch - 11us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.3112 - 2ms/epoch - 34us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.3859 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.6894 - 695us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.1961 - 698us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.9293 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.1611 - 664us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 107.8777 - 2ms/epoch - 27us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.5772 - 1ms/epoch - 16us/sample\n", "Generation 85/300\n", "Solving for Nash Equilibrium in Generation 85/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 778.0698 - 786us/epoch - 13us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.8644 - 680us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.1277 - 2ms/epoch - 35us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9058 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.4942 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.4069 - 698us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5458 - 2ms/epoch - 32us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.5666 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.7498 - 683us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.0082 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0796 - 856us/epoch - 14us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.3990 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2314 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9012 - 735us/epoch - 12us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.6483 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2879 - 2ms/epoch - 35us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0876 - 753us/epoch - 12us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5910 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7112 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.6614 - 863us/epoch - 14us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.2305 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8433 - 841us/epoch - 14us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.6627 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.3287 - 682us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.6839 - 840us/epoch - 14us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.7189 - 911us/epoch - 15us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1400 - 665us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.4215 - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.2875 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.3615 - 724us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.2939 - 2ms/epoch - 37us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.5826 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.8707 - 851us/epoch - 14us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.6337 - 938us/epoch - 15us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.4911 - 849us/epoch - 14us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.8907 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.3787 - 842us/epoch - 14us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.1787 - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.2474 - 1ms/epoch - 16us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.1699 - 818us/epoch - 13us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.5216 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.3242 - 1ms/epoch - 17us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1592 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.3275 - 3ms/epoch - 53us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.3967 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7387 - 857us/epoch - 14us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.3577 - 710us/epoch - 11us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4391 - 899us/epoch - 14us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.5179 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.8578 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4734 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.4277 - 797us/epoch - 13us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.9078 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4485 - 709us/epoch - 11us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.0820 - 658us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.4917 - 705us/epoch - 11us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.2471 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7847 - 714us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0003 - 723us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.9324 - 713us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4415 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.2925 - 691us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5847 - 798us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.9108 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.7223 - 577us/epoch - 9us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9958 - 654us/epoch - 11us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 181.2410 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9581 - 823us/epoch - 13us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.8107 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.6546 - 728us/epoch - 12us/sample\n", "Generation 86/300\n", "Solving for Nash Equilibrium in Generation 86/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.8854 - 844us/epoch - 14us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.1115 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.7382 - 621us/epoch - 10us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.9825 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.4593 - 770us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.8664 - 867us/epoch - 14us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.9647 - 730us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.9633 - 1ms/epoch - 16us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5206 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.6646 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.4205 - 919us/epoch - 15us/sample\n", "Episode 8/50\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.9996 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51.7039 - 2ms/epoch - 35us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.2820 - 763us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4577 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.8362 - 749us/epoch - 12us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.4361 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.2792 - 698us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5414 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.5196 - 807us/epoch - 13us/sample\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.7828 - 653us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0759 - 729us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4491 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.6190 - 6ms/epoch - 89us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.4137 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6815 - 679us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.4062 - 715us/epoch - 12us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.6201 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.1746 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.7876 - 5ms/epoch - 84us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.1403 - 728us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.2211 - 1ms/epoch - 21us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.8796 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.6723 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.0668 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.5255 - 933us/epoch - 15us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0436 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.7970 - 640us/epoch - 10us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.6924 - 756us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.2459 - 894us/epoch - 14us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4579 - 928us/epoch - 15us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.4552 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 803.8552 - 11ms/epoch - 180us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.3696 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 155.4028 - 686us/epoch - 11us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3559 - 762us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.3660 - 694us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.4436 - 2ms/epoch - 27us/sample\n", "Episode 37/50\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8481 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.0222 - 723us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.5033 - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 140us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43.8886 - 906us/epoch - 15us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8752 - 853us/epoch - 14us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.6192 - 2ms/epoch - 33us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.7369 - 773us/epoch - 12us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.5495 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.2571 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.4838 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9599 - 727us/epoch - 12us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2601 - 1ms/epoch - 17us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.7820 - 2ms/epoch - 30us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.7787 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.5383 - 846us/epoch - 14us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.4558 - 694us/epoch - 11us/sample\n", "Generation 87/300\n", "Solving for Nash Equilibrium in Generation 87/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.1461 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.2396 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.7639 - 807us/epoch - 13us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.5026 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7428 - 708us/epoch - 11us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.0194 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.5501 - 801us/epoch - 13us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9898 - 5ms/epoch - 86us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.3523 - 883us/epoch - 14us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.6483 - 704us/epoch - 11us/sample\n", "Episode 8/50\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.0012 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.4374 - 898us/epoch - 14us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4943 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.8784 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.7501 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2017 - 737us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.6815 - 1ms/epoch - 21us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.8480 - 1ms/epoch - 17us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.9142 - 5ms/epoch - 75us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.3544 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.8726 - 703us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9478 - 746us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.8585 - 3ms/epoch - 55us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.7018 - 979us/epoch - 16us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.4203 - 746us/epoch - 12us/sample\n", "Episode 20/50\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.7453 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45.0331 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.1468 - 9ms/epoch - 148us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.6739 - 842us/epoch - 14us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3998 - 802us/epoch - 13us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.4898 - 1ms/epoch - 21us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4180 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.7061 - 670us/epoch - 11us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.2545 - 747us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.0126 - 580us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.5326 - 531us/epoch - 9us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.1847 - 748us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.6299 - 721us/epoch - 12us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.7099 - 1ms/epoch - 22us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.3029 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0408 - 665us/epoch - 11us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.0421 - 885us/epoch - 14us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.4677 - 814us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.4474 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.3986 - 981us/epoch - 16us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.9322 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.4332 - 888us/epoch - 14us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 775.9080 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2349 - 969us/epoch - 16us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6888 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.9343 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.3618 - 891us/epoch - 14us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.9877 - 782us/epoch - 13us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2620 - 1ms/epoch - 19us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.1005 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0975 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.7851 - 802us/epoch - 13us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9092 - 838us/epoch - 14us/sample\n", "Episode 45/50\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.2357 - 1ms/epoch - 20us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2893 - 644us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.2992 - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.3620 - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5554 - 993us/epoch - 16us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.1066 - 688us/epoch - 11us/sample\n", "Generation 88/300\n", "Solving for Nash Equilibrium in Generation 88/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7896 - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.2282 - 873us/epoch - 14us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.3480 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.0752 - 678us/epoch - 11us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.5047 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0437 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.8031 - 609us/epoch - 10us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.9480 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.8693 - 2ms/epoch - 34us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.8435 - 745us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.1048 - 971us/epoch - 16us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.1937 - 675us/epoch - 11us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.6854 - 779us/epoch - 13us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.1524 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.8070 - 797us/epoch - 13us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.9272 - 3ms/epoch - 51us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.1679 - 774us/epoch - 12us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.1688 - 888us/epoch - 14us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2874 - 2ms/epoch - 29us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.5419 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9554 - 726us/epoch - 12us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.2263 - 765us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.2041 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.1679 - 1ms/epoch - 19us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4828 - 749us/epoch - 12us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.2579 - 823us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5469 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6464 - 835us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.1528 - 889us/epoch - 14us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.0952 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.6895 - 905us/epoch - 15us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8944 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.1813 - 679us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.9650 - 5ms/epoch - 73us/sample\n", "Episode 30/50\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.9484 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3987 - 905us/epoch - 15us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.5224 - 1ms/epoch - 23us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.3363 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.8269 - 596us/epoch - 10us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.7442 - 672us/epoch - 11us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6825 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.0339 - 780us/epoch - 13us/sample\n", "Episode 37/50\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.5924 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6686 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9683 - 1ms/epoch - 18us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.3966 - 833us/epoch - 13us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.2491 - 772us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.5039 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.5630 - 1ms/epoch - 18us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.2077 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.3390 - 727us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.6256 - 658us/epoch - 11us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.5887 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.5625 - 792us/epoch - 13us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.2973 - 733us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.5847 - 783us/epoch - 13us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.1721 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7122 - 1ms/epoch - 20us/sample\n", "Generation 89/300\n", "Solving for Nash Equilibrium in Generation 89/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 205.3856 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.5930 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.6157 - 600us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.9741 - 729us/epoch - 12us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.5326 - 1ms/epoch - 21us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.6600 - 818us/epoch - 13us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.5002 - 599us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.9408 - 620us/epoch - 10us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.3809 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.7310 - 739us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.1140 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.4036 - 613us/epoch - 10us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 796.6478 - 592us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.8484 - 2ms/epoch - 36us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7007 - 811us/epoch - 13us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.6884 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7372 - 679us/epoch - 11us/sample\n", "Episode 13/50\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7576 - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 158us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.2310 - 1ms/epoch - 20us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1315 - 572us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.0820 - 635us/epoch - 10us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2966 - 656us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.4865 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0989 - 926us/epoch - 15us/sample\n", "Episode 20/50\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2026 - 618us/epoch - 10us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.6908 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.0979 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.1909 - 2ms/epoch - 40us/sample\n", "Episode 23/50\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6354 - 667us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.4059 - 626us/epoch - 10us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7374 - 691us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.8027 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.4650 - 2ms/epoch - 35us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9755 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.8143 - 686us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.9261 - 671us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.9425 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.7938 - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.5786 - 690us/epoch - 11us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.4695 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 791.6884 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.2037 - 836us/epoch - 13us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.7535 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.3571 - 1ms/epoch - 24us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4415 - 807us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.3093 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5152 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.8707 - 689us/epoch - 11us/sample\n", "Episode 37/50\n", "Episode 38/50\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.5518 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.9531 - 2ms/epoch - 37us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.6394 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.3102 - 720us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.3780 - 687us/epoch - 11us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.0544 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.3148 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7177 - 692us/epoch - 11us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.4629 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.2304 - 737us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.9542 - 705us/epoch - 11us/sample\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.4479 - 625us/epoch - 10us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.7604 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.4487 - 8ms/epoch - 133us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.9587 - 717us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.1223 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.4784 - 646us/epoch - 10us/sample\n", "Generation 90/300\n", "Solving for Nash Equilibrium in Generation 90/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9319 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4401 - 777us/epoch - 13us/sample\n", "Episode 2/50\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7755 - 668us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8564 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.5789 - 800us/epoch - 13us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 176.5195 - 713us/epoch - 11us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.5161 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6722 - 675us/epoch - 11us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.4984 - 964us/epoch - 16us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.7126 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.2902 - 844us/epoch - 14us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.5632 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.7293 - 700us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.4103 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.7618 - 629us/epoch - 10us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9500 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.6710 - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.5739 - 676us/epoch - 11us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2415 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.7497 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.1511 - 683us/epoch - 11us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.2869 - 587us/epoch - 9us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.9392 - 613us/epoch - 10us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.2699 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.3329 - 617us/epoch - 10us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.5757 - 668us/epoch - 11us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.7667 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.6549 - 627us/epoch - 10us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5399 - 1ms/epoch - 18us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 790.2098 - 774us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.3233 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.9139 - 667us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.3586 - 563us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.1661 - 1ms/epoch - 19us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.4713 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6915 - 2ms/epoch - 34us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.1902 - 842us/epoch - 14us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.4721 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.2952 - 660us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.1717 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.0054 - 1ms/epoch - 17us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3086 - 782us/epoch - 13us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.7695 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9999 - 1ms/epoch - 16us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3997 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.8093 - 625us/epoch - 10us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.1873 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.6009 - 742us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.3053 - 911us/epoch - 15us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.2513 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.3942 - 616us/epoch - 10us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 796.7064 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5513 - 4ms/epoch - 66us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.5421 - 587us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.8304 - 722us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.4197 - 775us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.3934 - 616us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.9463 - 672us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3602 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.2700 - 728us/epoch - 12us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 786.6141 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.9262 - 757us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.2415 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8215 - 777us/epoch - 13us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.3377 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2599 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4739 - 4ms/epoch - 66us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.1713 - 694us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.6302 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.1359 - 708us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.1063 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.2332 - 931us/epoch - 15us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0375 - 652us/epoch - 11us/sample\n", "Generation 91/300\n", "Solving for Nash Equilibrium in Generation 91/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.1832 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.3686 - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9030 - 635us/epoch - 10us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.6682 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.3998 - 711us/epoch - 11us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.5312 - 688us/epoch - 11us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9960 - 671us/epoch - 11us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.1035 - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.5132 - 885us/epoch - 14us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.1894 - 673us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.8098 - 621us/epoch - 10us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.6653 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.1736 - 637us/epoch - 10us/sample\n", "Episode 11/50\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.2258 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8642 - 912us/epoch - 15us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.7748 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5213 - 764us/epoch - 12us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.6023 - 1ms/epoch - 23us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 210.3867 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.7599 - 751us/epoch - 12us/sample\n", "Episode 18/50\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.7472 - 3ms/epoch - 47us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.7852 - 1ms/epoch - 21us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.9541 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.4104 - 658us/epoch - 11us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.7878 - 724us/epoch - 12us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8000 - 989us/epoch - 16us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2599 - 867us/epoch - 14us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.7158 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.7144 - 781us/epoch - 13us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.1973 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7855 - 753us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.6274 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.0409 - 655us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.5191 - 683us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.2996 - 705us/epoch - 11us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.3826 - 8ms/epoch - 128us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 102us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 783.6500 - 923us/epoch - 15us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.7910 - 638us/epoch - 10us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.1433 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.2049 - 687us/epoch - 11us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.0184 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6890 - 672us/epoch - 11us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0465 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7389 - 760us/epoch - 12us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.4373 - 714us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.9083 - 753us/epoch - 12us/sample\n", "Episode 40/50\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4706 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.9067 - 758us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.5269 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0943 - 656us/epoch - 11us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.5042 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.7442 - 685us/epoch - 11us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6675 - 3ms/epoch - 51us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.4240 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.6246 - 673us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.6296 - 878us/epoch - 14us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.8097 - 765us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.1162 - 838us/epoch - 14us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.1185 - 694us/epoch - 11us/sample\n", "Generation 92/300\n", "Solving for Nash Equilibrium in Generation 92/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.6689 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.9375 - 977us/epoch - 16us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.9713 - 2ms/epoch - 35us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.3826 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5385 - 671us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9566 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.7576 - 714us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.1972 - 911us/epoch - 15us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.6062 - 662us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.7649 - 737us/epoch - 12us/sample\n", "Episode 10/50\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.0629 - 1ms/epoch - 16us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.3001 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.0464 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.0715 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.0261 - 2ms/epoch - 25us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.4282 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.1475 - 1ms/epoch - 20us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9543 - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.9552 - 798us/epoch - 13us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.8892 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.2273 - 693us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.1186 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.6967 - 738us/epoch - 12us/sample\n", "Episode 18/50\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 794.6401 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.7671 - 1ms/epoch - 16us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1612 - 2ms/epoch - 28us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.0547 - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.3660 - 1ms/epoch - 19us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.2143 - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.9712 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.1185 - 934us/epoch - 15us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.1108 - 798us/epoch - 13us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.0938 - 972us/epoch - 16us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6155 - 667us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.2812 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.4193 - 706us/epoch - 11us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.5975 - 1ms/epoch - 22us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.4503 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.8441 - 737us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.8255 - 742us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.9084 - 936us/epoch - 15us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5286 - 788us/epoch - 13us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.5529 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.6103 - 745us/epoch - 12us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.5148 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.3745 - 963us/epoch - 16us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7151 - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7105 - 991us/epoch - 16us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7430 - 683us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.3958 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.9121 - 706us/epoch - 11us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.1988 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.1009 - 570us/epoch - 9us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.5278 - 747us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.0583 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.1633 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6316 - 723us/epoch - 12us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.0327 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.5231 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.5032 - 814us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.5380 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8161 - 4ms/epoch - 58us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.6544 - 1ms/epoch - 19us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.5823 - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.2918 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5005 - 782us/epoch - 13us/sample\n", "Episode 46/50\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.1789 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6259 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.4698 - 2ms/epoch - 26us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.6270 - 922us/epoch - 15us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.2418 - 901us/epoch - 15us/sample\n", "Generation 93/300\n", "Solving for Nash Equilibrium in Generation 93/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.8918 - 826us/epoch - 13us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.2105 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.7562 - 1ms/epoch - 16us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7026 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.3527 - 748us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1140 - 696us/epoch - 11us/sample\n", "Episode 5/50\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.7734 - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.0341 - 767us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.1499 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.8409 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.0887 - 711us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.4677 - 735us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.6075 - 856us/epoch - 14us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9211 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.8017 - 4ms/epoch - 57us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8977 - 569us/epoch - 9us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1603 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.6884 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.1555 - 591us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.5367 - 566us/epoch - 9us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9150 - 631us/epoch - 10us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.6524 - 797us/epoch - 13us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8644 - 706us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 218.8145 - 752us/epoch - 12us/sample\n", "Episode 19/50\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9286 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.2698 - 907us/epoch - 15us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.7927 - 827us/epoch - 13us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.7579 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.7296 - 747us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.5052 - 707us/epoch - 11us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.6406 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.0652 - 706us/epoch - 11us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.1589 - 725us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2244 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4320 - 740us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.6903 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.5626 - 958us/epoch - 15us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 93.4718 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2953 - 953us/epoch - 15us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 106us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.3805 - 858us/epoch - 14us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.4640 - 909us/epoch - 15us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.4486 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.4075 - 796us/epoch - 13us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.3138 - 795us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1780 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.7946 - 781us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.3087 - 686us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.0838 - 617us/epoch - 10us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3275 - 5ms/epoch - 76us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.9237 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.8885 - 802us/epoch - 13us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.8685 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0912 - 672us/epoch - 11us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.7558 - 1ms/epoch - 18us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4435 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2900 - 634us/epoch - 10us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.6316 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.0373 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.2574 - 926us/epoch - 15us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.3237 - 606us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0680 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.3724 - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.1798 - 708us/epoch - 11us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.5864 - 666us/epoch - 11us/sample\n", "Episode 49/50\n", "Episode 50/50\n", "Generation 94/300\n", "Solving for Nash Equilibrium in Generation 94/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1884 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.2758 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.9695 - 561us/epoch - 9us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.7332 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.7609 - 3ms/epoch - 48us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.3738 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.3889 - 801us/epoch - 13us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9476 - 638us/epoch - 10us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7819 - 761us/epoch - 12us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.4700 - 956us/epoch - 15us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9225 - 710us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.0402 - 745us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7208 - 693us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.7530 - 689us/epoch - 11us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2863 - 620us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.8106 - 630us/epoch - 10us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.3122 - 756us/epoch - 12us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3690 - 711us/epoch - 11us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.7027 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2261 - 916us/epoch - 15us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4387 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.1780 - 1ms/epoch - 17us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79.2105 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1053 - 828us/epoch - 13us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.0246 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5364 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9402 - 1ms/epoch - 24us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6693 - 842us/epoch - 14us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0261 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.1581 - 849us/epoch - 14us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2850 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.0939 - 1ms/epoch - 23us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6972 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.0371 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.9488 - 811us/epoch - 13us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.3305 - 5ms/epoch - 84us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.5085 - 868us/epoch - 14us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.8820 - 707us/epoch - 11us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.1523 - 1ms/epoch - 24us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.6571 - 673us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.6593 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.8834 - 764us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.6287 - 716us/epoch - 12us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.2214 - 1ms/epoch - 18us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.1719 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.2533 - 771us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5266 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.0871 - 893us/epoch - 14us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.8178 - 3ms/epoch - 51us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.8087 - 2ms/epoch - 37us/sample\n", "Episode 39/50\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5914 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0306 - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.1246 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1528 - 2ms/epoch - 26us/sample\n", "Episode 41/50\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4998 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.3041 - 716us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.1136 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.2755 - 904us/epoch - 15us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.2291 - 9ms/epoch - 147us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.8107 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.1841 - 1ms/epoch - 19us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.5818 - 693us/epoch - 11us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2296 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.8373 - 654us/epoch - 11us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.6655 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1550 - 800us/epoch - 13us/sample\n", "Episode 50/50\n", "Generation 95/300\n", "Solving for Nash Equilibrium in Generation 95/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.9388 - 873us/epoch - 14us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.3132 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.9375 - 766us/epoch - 12us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.0120 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0555 - 976us/epoch - 16us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.8224 - 717us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.6489 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.2571 - 6ms/epoch - 96us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.9315 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.2018 - 686us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.9388 - 818us/epoch - 13us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 149us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.8515 - 36ms/epoch - 588us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4953 - 787us/epoch - 13us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1652 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.2726 - 678us/epoch - 11us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.3784 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.4789 - 740us/epoch - 12us/sample\n", "Episode 13/50\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.7608 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2554 - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.8724 - 815us/epoch - 13us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.8105 - 884us/epoch - 14us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.8967 - 590us/epoch - 10us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.5568 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.6276 - 1ms/epoch - 16us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.0272 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.1853 - 712us/epoch - 11us/sample\n", "Episode 19/50\n", "Episode 20/50\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.1125 - 1ms/epoch - 19us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.7237 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.1291 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.8687 - 2ms/epoch - 30us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.7726 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.0305 - 789us/epoch - 13us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.6966 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9161 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1600 - 936us/epoch - 15us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.4467 - 627us/epoch - 10us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.7691 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.1164 - 663us/epoch - 11us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.9489 - 744us/epoch - 12us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.6761 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.3906 - 2ms/epoch - 32us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.2578 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.9962 - 1ms/epoch - 17us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3238 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6173 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.5079 - 725us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.5714 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.9832 - 712us/epoch - 11us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.1727 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2158 - 893us/epoch - 14us/sample\n", "Episode 36/50\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.9710 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.1917 - 1ms/epoch - 20us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6175 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.0627 - 822us/epoch - 13us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.4526 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.6509 - 856us/epoch - 14us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.9099 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6532 - 719us/epoch - 12us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.8937 - 735us/epoch - 12us/sample\n", "Episode 42/50\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3747 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.5709 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.7060 - 2ms/epoch - 34us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.5550 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.6476 - 922us/epoch - 15us/sample\n", "Episode 45/50\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6802 - 2ms/epoch - 26us/sample\n", "Episode 48/50\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.2156 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.5545 - 720us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45.6697 - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.1671 - 1ms/epoch - 20us/sample\n", "Generation 96/300\n", "Solving for Nash Equilibrium in Generation 96/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.6288 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9156 - 3ms/epoch - 46us/sample\n", "Episode 3/50\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.3981 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.6205 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.8307 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.5913 - 778us/epoch - 13us/sample\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4174 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9952 - 968us/epoch - 16us/sample\n", "Episode 7/50\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 141us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.3354 - 7ms/epoch - 115us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.3668 - 800us/epoch - 13us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.1057 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.7010 - 660us/epoch - 11us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.7663 - 1ms/epoch - 23us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7706 - 637us/epoch - 10us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7457 - 669us/epoch - 11us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.5272 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.5442 - 712us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.3025 - 793us/epoch - 13us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.0504 - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8816 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.9789 - 770us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.8925 - 5ms/epoch - 74us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.9432 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7377 - 605us/epoch - 10us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.4266 - 859us/epoch - 14us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.0818 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.6600 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.0238 - 832us/epoch - 13us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.1658 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3948 - 2ms/epoch - 39us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.5723 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.7943 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.0905 - 727us/epoch - 12us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.5440 - 754us/epoch - 12us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.1251 - 714us/epoch - 12us/sample\n", "Episode 27/50\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8436 - 826us/epoch - 13us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.0958 - 912us/epoch - 15us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.1610 - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.2544 - 811us/epoch - 13us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.7931 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.1595 - 770us/epoch - 12us/sample\n", "Episode 32/50\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.1204 - 804us/epoch - 13us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.4285 - 759us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 786.9603 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4287 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.1297 - 721us/epoch - 12us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9556 - 755us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.0674 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.6699 - 928us/epoch - 15us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9467 - 805us/epoch - 13us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.8162 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.9400 - 1ms/epoch - 19us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.3597 - 1ms/epoch - 16us/sample\n", "Episode 42/50\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8208 - 2ms/epoch - 32us/sample\n", "Episode 45/50\n", "Episode 46/50\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.0402 - 1ms/epoch - 18us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.9083 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.3168 - 741us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.3983 - 780us/epoch - 13us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7846 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5024 - 4ms/epoch - 57us/sample\n", "Generation 97/300\n", "Solving for Nash Equilibrium in Generation 97/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.2451 - 904us/epoch - 15us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.0699 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.1550 - 1ms/epoch - 17us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.7378 - 936us/epoch - 15us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.1048 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5744 - 842us/epoch - 14us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.2601 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.6283 - 819us/epoch - 13us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.1581 - 709us/epoch - 11us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.0431 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.6729 - 1ms/epoch - 21us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.4988 - 748us/epoch - 12us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.8655 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.2415 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3785 - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.2078 - 1ms/epoch - 16us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.2573 - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.8137 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.9688 - 945us/epoch - 15us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1541 - 2ms/epoch - 30us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1163 - 938us/epoch - 15us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.2387 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.7151 - 997us/epoch - 16us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 124us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.9203 - 5ms/epoch - 76us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.9941 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.2844 - 683us/epoch - 11us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.5048 - 779us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.6639 - 621us/epoch - 10us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.2855 - 3ms/epoch - 45us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.7391 - 3ms/epoch - 52us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.5569 - 962us/epoch - 16us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.0684 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.7595 - 1ms/epoch - 17us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.1823 - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.1209 - 1ms/epoch - 16us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.6106 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.8292 - 2ms/epoch - 39us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1880 - 811us/epoch - 13us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.6295 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4332 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.9192 - 751us/epoch - 12us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.5451 - 1ms/epoch - 20us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.3140 - 831us/epoch - 13us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5076 - 724us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.3622 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 147us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2016 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.0583 - 837us/epoch - 14us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.4896 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.7007 - 772us/epoch - 12us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0003 - 811us/epoch - 13us/sample\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9227 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.2003 - 751us/epoch - 12us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.1256 - 778us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1492 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6828 - 2ms/epoch - 38us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.6700 - 1ms/epoch - 21us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.1119 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5925 - 833us/epoch - 13us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4098 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.5529 - 705us/epoch - 11us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.7369 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.8275 - 945us/epoch - 15us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9051 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.3770 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7040 - 789us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3820 - 805us/epoch - 13us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.3236 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.7373 - 1ms/epoch - 21us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.8013 - 938us/epoch - 15us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.3300 - 911us/epoch - 15us/sample\n", "Episode 47/50\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.1207 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.7645 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.7975 - 1ms/epoch - 20us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.7039 - 665us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3873 - 722us/epoch - 12us/sample\n", "Generation 98/300\n", "Solving for Nash Equilibrium in Generation 98/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7580 - 1ms/epoch - 17us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.6706 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.9193 - 680us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.0425 - 688us/epoch - 11us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.2045 - 823us/epoch - 13us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.6218 - 924us/epoch - 15us/sample\n", "Episode 6/50\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.5827 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.0942 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6718 - 736us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.7237 - 708us/epoch - 11us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0802 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.4315 - 4ms/epoch - 62us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.8043 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.5288 - 955us/epoch - 15us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.9836 - 723us/epoch - 12us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.2288 - 935us/epoch - 15us/sample\n", "Episode 13/50\n", "Episode 14/50\n", "Episode 15/50\n", "Episode 16/50\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.8263 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.3135 - 761us/epoch - 12us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.7255 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.8748 - 4ms/epoch - 71us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.4285 - 733us/epoch - 12us/sample\n", "Episode 20/50\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.9509 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.1129 - 1ms/epoch - 17us/sample\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3220 - 757us/epoch - 12us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.7401 - 775us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.9552 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5240 - 633us/epoch - 10us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.9808 - 683us/epoch - 11us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.9261 - 13ms/epoch - 210us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4045 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5090 - 624us/epoch - 10us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.5914 - 752us/epoch - 12us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.0291 - 616us/epoch - 10us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.0236 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.8571 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.5756 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3816 - 2ms/epoch - 33us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9122 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.8312 - 721us/epoch - 12us/sample\n", "Episode 32/50\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.7694 - 725us/epoch - 12us/sample\n", "Episode 35/50\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5043 - 783us/epoch - 13us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.8534 - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.8885 - 710us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.1651 - 728us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5471 - 975us/epoch - 16us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.6413 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9242 - 1ms/epoch - 23us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.8465 - 908us/epoch - 15us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.9890 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.1710 - 826us/epoch - 13us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.7200 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0743 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.0704 - 1ms/epoch - 22us/sample\n", "Episode 44/50\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.1133 - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.4740 - 763us/epoch - 12us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.5359 - 710us/epoch - 11us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.5781 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.6350 - 5ms/epoch - 77us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.5717 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9930 - 4ms/epoch - 64us/sample\n", "Episode 49/50\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.3303 - 3ms/epoch - 50us/sample\n", "Generation 99/300\n", "Solving for Nash Equilibrium in Generation 99/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0357 - 2ms/epoch - 32us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.0870 - 721us/epoch - 12us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.0858 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.8785 - 700us/epoch - 11us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.4334 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6858 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7208 - 778us/epoch - 13us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.9635 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.1022 - 1ms/epoch - 19us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.9018 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.6339 - 2ms/epoch - 30us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.8085 - 799us/epoch - 13us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.6821 - 2ms/epoch - 31us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.0075 - 903us/epoch - 15us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.1449 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.0579 - 1ms/epoch - 18us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5397 - 4ms/epoch - 72us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2039 - 1ms/epoch - 19us/sample\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5219 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.9453 - 2ms/epoch - 32us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.0876 - 822us/epoch - 13us/sample\n", "Episode 17/50\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 811.1934 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3407 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.7012 - 774us/epoch - 12us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.3068 - 2ms/epoch - 37us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.3944 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.6214 - 680us/epoch - 11us/sample\n", "Episode 21/50\n", "Episode 22/50\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5705 - 780us/epoch - 13us/sample\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9931 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5162 - 672us/epoch - 11us/sample\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 110us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.6638 - 6ms/epoch - 93us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.7023 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 50ms/epoch - 799us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.3178 - 6ms/epoch - 93us/sample\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.1374 - 3ms/epoch - 44us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.0694 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.5187 - 841us/epoch - 14us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9249 - 968us/epoch - 16us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.4977 - 1ms/epoch - 17us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.5717 - 741us/epoch - 12us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2339 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.3560 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.9825 - 694us/epoch - 11us/sample\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.4860 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.1744 - 1ms/epoch - 17us/sample\n", "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.0817 - 858us/epoch - 14us/sample\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8616 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9027 - 802us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.9522 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.7214 - 911us/epoch - 15us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.3133 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.6310 - 7ms/epoch - 115us/sample\n", "Episode 38/50\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.3015 - 737us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.8494 - 819us/epoch - 13us/sample\n", "Episode 41/50\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.7198 - 773us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.2708 - 4ms/epoch - 61us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.1547 - 788us/epoch - 13us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.7126 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4929 - 1ms/epoch - 16us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.2434 - 3ms/epoch - 48us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.8680 - 5ms/epoch - 78us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.4476 - 878us/epoch - 14us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7539 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.2589 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.9754 - 710us/epoch - 11us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.7673 - 852us/epoch - 14us/sample\n", "Generation 100/300\n", "Solving for Nash Equilibrium in Generation 100/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773.0003 - 876us/epoch - 14us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.6750 - 705us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9077 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0980 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.7260 - 666us/epoch - 11us/sample\n", "Episode 4/50\n", "Episode 5/50\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.8546 - 717us/epoch - 12us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.2525 - 883us/epoch - 14us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.8802 - 2ms/epoch - 28us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.7472 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.4769 - 1ms/epoch - 20us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.1782 - 846us/epoch - 14us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.8672 - 833us/epoch - 13us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.4771 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.0780 - 798us/epoch - 13us/sample\n", "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.6487 - 1ms/epoch - 19us/sample\n", "Episode 15/50\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.2472 - 935us/epoch - 15us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37.3893 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.9653 - 1ms/epoch - 24us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.2811 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.7595 - 801us/epoch - 13us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.9155 - 842us/epoch - 14us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.6816 - 4ms/epoch - 70us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.0275 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.3914 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.7861 - 1ms/epoch - 20us/sample\n", "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.8749 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.8231 - 1ms/epoch - 19us/sample\n", "Episode 24/50\n", "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.5049 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7943 - 707us/epoch - 11us/sample\n", "Episode 26/50\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5106 - 1ms/epoch - 16us/sample\n", "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.8623 - 812us/epoch - 13us/sample\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.7116 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.4702 - 842us/epoch - 14us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.3771 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7841 - 1ms/epoch - 23us/sample\n", "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.7874 - 1ms/epoch - 17us/sample\n", "Episode 32/50\n", "Episode 33/50\n", "Episode 34/50\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.2660 - 831us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.1907 - 750us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.0654 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.7666 - 824us/epoch - 13us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.4095 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1546 - 839us/epoch - 14us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.2608 - 765us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.6136 - 1ms/epoch - 17us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.1734 - 1ms/epoch - 19us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.7849 - 752us/epoch - 12us/sample\n", "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.9722 - 724us/epoch - 12us/sample\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.4845 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.5574 - 742us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.0403 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0533 - 642us/epoch - 10us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.4118 - 617us/epoch - 10us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.6780 - 737us/epoch - 12us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.9891 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4001 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.8241 - 5ms/epoch - 76us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.6408 - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3098 - 2ms/epoch - 35us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7341 - 8ms/epoch - 134us/sample\n", "Generation 101/300\n", "Solving for Nash Equilibrium in Generation 101/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 85us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.4199 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7795 - 963us/epoch - 16us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.6527 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8766 - 6ms/epoch - 99us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.4426 - 922us/epoch - 15us/sample\n", "Episode 4/50\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1285 - 727us/epoch - 12us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.6216 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.0105 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 13ms/epoch - 213us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1741 - 1ms/epoch - 18us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0865 - 994us/epoch - 16us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2370 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.9802 - 991us/epoch - 16us/sample\n", "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.4953 - 2ms/epoch - 25us/sample\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.3695 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.3972 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.7442 - 1ms/epoch - 18us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.9921 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 12ms/epoch - 196us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.0490 - 3ms/epoch - 56us/sample\n", "Episode 12/50\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.4258 - 12ms/epoch - 194us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.6510 - 1ms/epoch - 20us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.3730 - 1ms/epoch - 18us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.1881 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 13ms/epoch - 205us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.0825 - 2ms/epoch - 39us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.8966 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.4781 - 689us/epoch - 11us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.9439 - 2ms/epoch - 27us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.4532 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.3015 - 723us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.9616 - 897us/epoch - 14us/sample\n", "Episode 21/50\n", "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.8124 - 817us/epoch - 13us/sample\n", "Episode 23/50\n", "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.6133 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.2834 - 774us/epoch - 12us/sample\n", "Episode 25/50\n", "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.2939 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.5409 - 2ms/epoch - 37us/sample\n", "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.2019 - 2ms/epoch - 37us/sample\n", "Episode 28/50\n", "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.1867 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 206.6608 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36.2528 - 838us/epoch - 14us/sample\n", "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.5160 - 673us/epoch - 11us/sample\n", "Episode 31/50\n", "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.5508 - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6605 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9556 - 810us/epoch - 13us/sample\n", "Episode 33/50\n", "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.9474 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.7241 - 802us/epoch - 13us/sample\n", "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.4573 - 795us/epoch - 13us/sample\n", "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180.6887 - 743us/epoch - 12us/sample\n", "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.0326 - 683us/epoch - 11us/sample\n", "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.8354 - 752us/epoch - 12us/sample\n", "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.2429 - 744us/epoch - 12us/sample\n", "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.5433 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.5415 - 2ms/epoch - 28us/sample\n", "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.8591 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.7425 - 783us/epoch - 13us/sample\n", "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.0898 - 1ms/epoch - 21us/sample\n", "Episode 43/50\n", "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.3130 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.5238 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2584 - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.5239 - 773us/epoch - 12us/sample\n", "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.5448 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.2736 - 844us/epoch - 14us/sample\n", "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.6414 - 960us/epoch - 15us/sample\n", "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.4928 - 795us/epoch - 13us/sample\n", "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776.4285 - 771us/epoch - 12us/sample\n", "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4692 - 723us/epoch - 12us/sample\n", "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8814 - 780us/epoch - 13us/sample\n", "Generation 102/300\n", "Solving for Nash Equilibrium in Generation 102/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6230 - 785us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.7059 - 956us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3318 - 724us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7446 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.3717 - 2ms/epoch - 25us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3331 - 735us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.5710 - 722us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5428 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.4087 - 7ms/epoch - 113us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.2112 - 819us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.4205 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.9549 - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6433 - 12ms/epoch - 201us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 802.6586 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1462 - 1ms/epoch - 18us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.7841 - 1ms/epoch - 21us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3535 - 753us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.5632 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.5965 - 9ms/epoch - 146us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9866 - 1ms/epoch - 20us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8853 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8802 - 741us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.9518 - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.4606 - 631us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.2871 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.8306 - 733us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.4286 - 633us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2476 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 41ms/epoch - 660us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.1704 - 107ms/epoch - 2ms/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.4428 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.3263 - 2ms/epoch - 28us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.9937 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.7910 - 1ms/epoch - 17us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.9433 - 835us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9531 - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.1246 - 791us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.5513 - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7558 - 2ms/epoch - 34us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.5109 - 725us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.3742 - 1ms/epoch - 18us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2783 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.7599 - 1ms/epoch - 22us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.3014 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.5876 - 937us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.6680 - 843us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.3770 - 578us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.2737 - 993us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2729 - 716us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.3194 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.1113 - 2ms/epoch - 27us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.3602 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.2952 - 829us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.5811 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7020 - 4ms/epoch - 66us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9429 - 802us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.9072 - 730us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.7852 - 758us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.8093 - 2ms/epoch - 31us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.6030 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.9393 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 130us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.7179 - 11ms/epoch - 179us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8004 - 738us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.8115 - 7ms/epoch - 121us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.3875 - 675us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.3292 - 776us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.4735 - 905us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.9515 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52.0646 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.6808 - 740us/epoch - 12us/sample\n", + "Generation 103/300\n", + "Solving for Nash Equilibrium in Generation 103/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.0296 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.9489 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.3143 - 1ms/epoch - 19us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.1709 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.2070 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 116us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.6029 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.7717 - 1ms/epoch - 16us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.0553 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.8262 - 676us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6915 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9147 - 1ms/epoch - 17us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.9412 - 789us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4509 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6434 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4601 - 711us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.5248 - 1ms/epoch - 20us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5653 - 1ms/epoch - 16us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8239 - 1ms/epoch - 18us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.1451 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.4007 - 1ms/epoch - 23us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.5212 - 9ms/epoch - 148us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.6647 - 863us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.5944 - 677us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.8547 - 891us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.9397 - 1ms/epoch - 20us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.2897 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2350 - 726us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.1813 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.9427 - 781us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0790 - 981us/epoch - 16us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7151 - 831us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.7378 - 1ms/epoch - 22us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9501 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.5956 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8293 - 615us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.7720 - 2ms/epoch - 34us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.5551 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.2307 - 672us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.5230 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.5316 - 720us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.6586 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.1924 - 1ms/epoch - 19us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.0982 - 889us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5421 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2165 - 2ms/epoch - 39us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.3120 - 876us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.2380 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.3627 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.6021 - 652us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 803.3009 - 678us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.2513 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.8358 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.1473 - 749us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.4952 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7115 - 949us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9169 - 14ms/epoch - 218us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.7985 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9357 - 852us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2869 - 875us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.8569 - 827us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.0367 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.9554 - 786us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9242 - 720us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0107 - 641us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.7863 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.1873 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.2209 - 686us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.7428 - 4ms/epoch - 65us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.4576 - 2ms/epoch - 32us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.7684 - 5ms/epoch - 74us/sample\n", + "Generation 104/300\n", + "Solving for Nash Equilibrium in Generation 104/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7192 - 2ms/epoch - 25us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.8978 - 926us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.4041 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.1342 - 898us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4271 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.3711 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1656 - 1ms/epoch - 21us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.7529 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4828 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7491 - 722us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0369 - 833us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.6736 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.3597 - 685us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.2346 - 760us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.5059 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.2620 - 627us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.8235 - 832us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.4047 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3005 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.1837 - 680us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.8339 - 1ms/epoch - 17us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.8776 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.4298 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.9788 - 721us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.4964 - 872us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.3091 - 753us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.1949 - 692us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.0836 - 811us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 217.3339 - 727us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48.9433 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.2958 - 6ms/epoch - 99us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9547 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.7622 - 853us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.7703 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.4144 - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.0663 - 1ms/epoch - 20us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.6916 - 686us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.3931 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.5844 - 763us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.7145 - 777us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6799 - 764us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6591 - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.7042 - 650us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214.8423 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.3915 - 755us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.8312 - 908us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.1996 - 763us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4337 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.7453 - 736us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.0810 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5426 - 672us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.9053 - 694us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6196 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.2859 - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.7309 - 617us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.3505 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.7132 - 7ms/epoch - 105us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8870 - 733us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8207 - 903us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.5968 - 764us/epoch - 12us/sample\n", + "Generation 105/300\n", + "Solving for Nash Equilibrium in Generation 105/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.4309 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9388 - 683us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.0881 - 645us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.3385 - 791us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.0699 - 5ms/epoch - 87us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7090 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7180 - 1ms/epoch - 20us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.0978 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.8134 - 736us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.5416 - 833us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7842 - 1ms/epoch - 21us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9637 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.4027 - 2ms/epoch - 39us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.4836 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8411 - 787us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8457 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.9149 - 713us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.6671 - 727us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0508 - 670us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.2622 - 957us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6661 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.6471 - 730us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.7154 - 679us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.3215 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9757 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.5696 - 783us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.9221 - 5ms/epoch - 80us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.7315 - 748us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.8433 - 594us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.6065 - 700us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.9054 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.2860 - 601us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.4057 - 7ms/epoch - 115us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.5814 - 797us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.4498 - 971us/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.8935 - 892us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.0089 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2074 - 630us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.8843 - 695us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 146.0850 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.2359 - 871us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.8026 - 1ms/epoch - 23us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.1922 - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.4028 - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4039 - 775us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.2740 - 650us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.1881 - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.7578 - 12ms/epoch - 199us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 105.5213 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8042 - 748us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.7555 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.1659 - 704us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.0118 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.9922 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.7838 - 603us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5031 - 4ms/epoch - 71us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.2499 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.7668 - 1ms/epoch - 18us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7307 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.7606 - 816us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.0641 - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.5168 - 704us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.1979 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.1245 - 796us/epoch - 13us/sample\n", + "Generation 106/300\n", + "Solving for Nash Equilibrium in Generation 106/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.2748 - 5ms/epoch - 85us/sample\n", "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.8625 - 688us/epoch - 11us/sample\n", "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.0194 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.7749 - 619us/epoch - 10us/sample\n", "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8712 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.8400 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.0724 - 2ms/epoch - 32us/sample\n", "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4660 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.9952 - 841us/epoch - 14us/sample\n", "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.3899 - 742us/epoch - 12us/sample\n", "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.9961 - 728us/epoch - 12us/sample\n", "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 17ms/epoch - 270us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 775.0373 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9000 - 701us/epoch - 11us/sample\n", "Episode 9/50\n", "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.6483 - 987us/epoch - 16us/sample\n", "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.7905 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.0897 - 2ms/epoch - 25us/sample\n", "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.5753 - 873us/epoch - 14us/sample\n", "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.9371 - 740us/epoch - 12us/sample\n", "Episode 14/50\n", "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180.2505 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.3679 - 681us/epoch - 11us/sample\n", "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.2896 - 754us/epoch - 12us/sample\n", "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.3027 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.0585 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.4931 - 1ms/epoch - 19us/sample\n", "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.2214 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.4053 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.2189 - 677us/epoch - 11us/sample\n", "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.1039 - 768us/epoch - 12us/sample\n", "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.6917 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3184 - 919us/epoch - 15us/sample\n", "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.4797 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.4899 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.7695 - 4ms/epoch - 66us/sample\n", "Episode 22/50\n", "Episode 23/50\n", - "Episode 24/50\n" + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1196 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.3264 - 829us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.2156 - 740us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.0168 - 866us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.4623 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.7440 - 724us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.1193 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3546 - 2ms/epoch - 25us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.8060 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.2022 - 861us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.7609 - 2ms/epoch - 31us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 806.8109 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.1522 - 643us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.0358 - 936us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.8425 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.2919 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.8922 - 792us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9523 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2253 - 975us/epoch - 16us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.5129 - 784us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7625 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.2072 - 796us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3097 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.2036 - 883us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 791.0524 - 733us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9435 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.7098 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.4292 - 829us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3052 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.4602 - 795us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.3313 - 7ms/epoch - 109us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2769 - 762us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.9534 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.2773 - 695us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.3296 - 698us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.1324 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.7770 - 937us/epoch - 15us/sample\n", + "Generation 107/300\n", + "Solving for Nash Equilibrium in Generation 107/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8803 - 757us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.5851 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.5543 - 817us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.6044 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.8488 - 5ms/epoch - 75us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1739 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8146 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.8590 - 563us/epoch - 9us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.7499 - 1ms/epoch - 18us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9984 - 3ms/epoch - 47us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.3303 - 961us/epoch - 16us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0149 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.8197 - 588us/epoch - 9us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.1290 - 783us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.9954 - 707us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.1359 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3458 - 706us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9737 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.0505 - 724us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3528 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9525 - 810us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7219 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.2727 - 3ms/epoch - 42us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.1406 - 770us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.7275 - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.4888 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.0862 - 820us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.4003 - 1ms/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.3644 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8209 - 3ms/epoch - 45us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.0732 - 890us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.2141 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.7703 - 861us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37.4143 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.8355 - 3ms/epoch - 45us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.0888 - 1ms/epoch - 19us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.8340 - 713us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.6768 - 765us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 11ms/epoch - 179us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.6648 - 5ms/epoch - 83us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.7518 - 1ms/epoch - 17us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.9206 - 761us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.4791 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.8972 - 852us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.2479 - 834us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.8751 - 1ms/epoch - 20us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.2948 - 817us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.7674 - 2ms/epoch - 40us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.9976 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.8995 - 793us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.0474 - 859us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.9611 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 92.7818 - 790us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.6352 - 651us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0813 - 768us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.3866 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.2504 - 859us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5399 - 821us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.2997 - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.6461 - 834us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9523 - 1ms/epoch - 16us/sample\n", + "Generation 108/300\n", + "Solving for Nash Equilibrium in Generation 108/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.8310 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.7446 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9644 - 1ms/epoch - 18us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.0468 - 11ms/epoch - 179us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.3205 - 809us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2418 - 975us/epoch - 16us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.4020 - 802us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.7290 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.9640 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.1167 - 707us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.0509 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.5297 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.6588 - 1ms/epoch - 19us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779.1460 - 709us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.3596 - 4ms/epoch - 60us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.0752 - 737us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.4690 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.1266 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.2757 - 730us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.6472 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.0919 - 775us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.5493 - 706us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.0748 - 756us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.4089 - 684us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.0897 - 600us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.3804 - 955us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.9674 - 11ms/epoch - 172us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.1070 - 1ms/epoch - 22us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.0593 - 1ms/epoch - 17us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.6429 - 796us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.9434 - 749us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.7216 - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8651 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.0285 - 861us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.2225 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.9390 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3537 - 2ms/epoch - 32us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7513 - 705us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.7853 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0709 - 806us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9780 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5847 - 731us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.6506 - 771us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.9462 - 962us/epoch - 16us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.9170 - 717us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.1106 - 788us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.8656 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.6796 - 984us/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.1156 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.6873 - 623us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.5192 - 734us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.6438 - 741us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8392 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.7352 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7944 - 1ms/epoch - 17us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.5138 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.1287 - 850us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7070 - 821us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.7147 - 1ms/epoch - 19us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3160 - 743us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0344 - 2ms/epoch - 30us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.2758 - 792us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.3042 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.6768 - 788us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.2771 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9151 - 3ms/epoch - 43us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7157 - 812us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2994 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3845 - 668us/epoch - 11us/sample\n", + "Generation 109/300\n", + "Solving for Nash Equilibrium in Generation 109/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.5585 - 1ms/epoch - 19us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.9183 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.3391 - 874us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8683 - 709us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.2668 - 660us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.7973 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3515 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.0442 - 1ms/epoch - 17us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.0211 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8987 - 763us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3256 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2221 - 863us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.4419 - 865us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.4037 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.0636 - 860us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2123 - 728us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.8970 - 896us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.2292 - 723us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.5101 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.9678 - 718us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.7981 - 699us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.0035 - 627us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4971 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.0834 - 780us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.2214 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.3253 - 1ms/epoch - 18us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.5139 - 742us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.5378 - 802us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.1642 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.3300 - 758us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.6332 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.7313 - 637us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.9318 - 699us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 795.8763 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3163 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.5753 - 845us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.7343 - 799us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2900 - 3ms/epoch - 44us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.8299 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.8445 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3812 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.8265 - 9ms/epoch - 145us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.8511 - 710us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1063 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.8189 - 803us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.0781 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.9044 - 830us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.6461 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.7054 - 945us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5428 - 830us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.0888 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.2681 - 959us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.2191 - 848us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.6846 - 799us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.6326 - 1ms/epoch - 17us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.2882 - 887us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214.6260 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.5444 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.3255 - 819us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.1916 - 620us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.6414 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6596 - 855us/epoch - 14us/sample\n", + "Generation 110/300\n", + "Solving for Nash Equilibrium in Generation 110/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.6469 - 737us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.2618 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7882 - 742us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9929 - 762us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.3698 - 796us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.7393 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.3399 - 797us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9497 - 696us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8156 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.3587 - 613us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.8147 - 735us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.2838 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.1880 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.3980 - 829us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.0344 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1974 - 678us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.3608 - 684us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.1855 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.8254 - 805us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232.7924 - 651us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.9890 - 809us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.4232 - 8ms/epoch - 128us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.7143 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.6877 - 621us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.5980 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.0027 - 681us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.8138 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.8779 - 669us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0201 - 686us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773.7896 - 764us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.6370 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.8138 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.2529 - 767us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5444 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.3737 - 853us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.1192 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.1337 - 765us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.8399 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.3877 - 1ms/epoch - 20us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.5414 - 664us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.8219 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.9124 - 794us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5427 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.5869 - 744us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.0487 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.0861 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.8887 - 972us/epoch - 16us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.3614 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.0863 - 634us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.6683 - 613us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.3306 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7831 - 652us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.3419 - 662us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.1912 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.6322 - 707us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.7361 - 2ms/epoch - 32us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.3734 - 796us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.6689 - 681us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.9877 - 1ms/epoch - 18us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0120 - 731us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2929 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 107.0361 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.6287 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.3632 - 2ms/epoch - 28us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.1597 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.5988 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5030 - 763us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 111/300\n", + "Solving for Nash Equilibrium in Generation 111/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4722 - 918us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.6487 - 671us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.9807 - 5ms/epoch - 74us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 223.2108 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7954 - 660us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.4296 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.4771 - 829us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.0887 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.5981 - 580us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.6022 - 554us/epoch - 9us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4778 - 768us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.1661 - 856us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.5416 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.6518 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4406 - 578us/epoch - 9us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.5352 - 679us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.9304 - 713us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.2625 - 3ms/epoch - 41us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.5869 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.4572 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.8785 - 754us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.9526 - 769us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.5837 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.3953 - 689us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0291 - 586us/epoch - 9us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.3602 - 751us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.0879 - 694us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.4901 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.6135 - 719us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.3494 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.8950 - 731us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.6873 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4235 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.7982 - 626us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0446 - 654us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 34ms/epoch - 552us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.5554 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.0041 - 1ms/epoch - 22us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.7315 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.8637 - 779us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.7719 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.7371 - 735us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.2939 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5109 - 1ms/epoch - 18us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.9019 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.4158 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.7379 - 7ms/epoch - 116us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3326 - 900us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6088 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.6406 - 660us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5841 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.9141 - 729us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1175 - 1ms/epoch - 20us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9734 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.9661 - 1ms/epoch - 22us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.2216 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.0916 - 709us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.7599 - 2ms/epoch - 29us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.2158 - 808us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9823 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.5390 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9711 - 669us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.0941 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.5015 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.3757 - 934us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 799.9415 - 613us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.9556 - 751us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48.3458 - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.3989 - 682us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3149 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.3977 - 1ms/epoch - 18us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.4295 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.7336 - 752us/epoch - 12us/sample\n", + "Generation 112/300\n", + "Solving for Nash Equilibrium in Generation 112/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.1861 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.0221 - 615us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.5091 - 708us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1274 - 783us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.6454 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.7444 - 804us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.8890 - 896us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8328 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3167 - 636us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8332 - 721us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 96.8238 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.1559 - 3ms/epoch - 45us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4349 - 949us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.2263 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.2404 - 9ms/epoch - 144us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3292 - 731us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.3271 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.3307 - 863us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.7789 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.3920 - 1ms/epoch - 21us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.5887 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.1815 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6837 - 877us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.7592 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.3527 - 2ms/epoch - 30us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.1888 - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7697 - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.1849 - 863us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.4695 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.3238 - 798us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.0964 - 1ms/epoch - 18us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4692 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.1270 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.0574 - 711us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0112 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8017 - 738us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.5147 - 751us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3851 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.4676 - 677us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.5522 - 696us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5729 - 2ms/epoch - 32us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9802 - 902us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1436 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.5917 - 783us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.2416 - 603us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.4121 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.9785 - 680us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.1593 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.5711 - 880us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.9745 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.0351 - 678us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1523 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6305 - 791us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.1733 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.4246 - 2ms/epoch - 39us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.3801 - 705us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.7457 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.1446 - 2ms/epoch - 29us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.4294 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.2812 - 773us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.5566 - 763us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.1167 - 611us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5471 - 773us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9135 - 980us/epoch - 16us/sample\n", + "Generation 113/300\n", + "Solving for Nash Equilibrium in Generation 113/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.7367 - 710us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.7855 - 730us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8738 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.1663 - 798us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8931 - 891us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2346 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.1979 - 566us/epoch - 9us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4800 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.0645 - 677us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4212 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.4365 - 835us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.5211 - 744us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.8763 - 854us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.9642 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9929 - 5ms/epoch - 83us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.8883 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2783 - 1ms/epoch - 20us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.2459 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.1783 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.5957 - 703us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.7646 - 740us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9980 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.9996 - 662us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.4270 - 728us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.5541 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.9842 - 758us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.2235 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.5163 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.9720 - 3ms/epoch - 45us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.0522 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5632 - 3ms/epoch - 55us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.8323 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.8855 - 680us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.4399 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.1602 - 604us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.9166 - 805us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8409 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.2990 - 7ms/epoch - 109us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.2734 - 846us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.0302 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7801 - 4ms/epoch - 71us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.2684 - 997us/epoch - 16us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8859 - 728us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 85us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.5876 - 4ms/epoch - 64us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.4811 - 716us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.1102 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.7719 - 770us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2436 - 755us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230.7028 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.0615 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.7056 - 578us/epoch - 9us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.3280 - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3623 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.1288 - 815us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.9163 - 701us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.9720 - 749us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.2978 - 1ms/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.3477 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.1861 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1107 - 634us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.0038 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.7241 - 566us/epoch - 9us/sample\n", + "Generation 114/300\n", + "Solving for Nash Equilibrium in Generation 114/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2377 - 819us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.5331 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.9310 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5237 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.4661 - 796us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6702 - 694us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4649 - 1ms/epoch - 22us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.3182 - 898us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 135.7350 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5913 - 778us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.9219 - 1ms/epoch - 17us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.9033 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.3289 - 855us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.9643 - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.7550 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.4496 - 800us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 146.1369 - 728us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.8826 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.6211 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.2601 - 841us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.2898 - 710us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.1401 - 746us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.3236 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.3560 - 591us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.5200 - 787us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.9945 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.0454 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.5340 - 742us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.4532 - 746us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.1826 - 600us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.5168 - 593us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.3120 - 676us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.9939 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.0776 - 606us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0974 - 831us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.8704 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.1805 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5657 - 781us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.9173 - 612us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.0209 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.3646 - 600us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.4493 - 724us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.1364 - 697us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.1578 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 99.0529 - 842us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.6680 - 628us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.3992 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.7758 - 716us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.3481 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.7909 - 594us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.3922 - 697us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5778 - 774us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.6118 - 683us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.3043 - 774us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52.7847 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.6893 - 881us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5551 - 647us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.5977 - 783us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.5555 - 2ms/epoch - 30us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.5189 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.1345 - 689us/epoch - 11us/sample\n", + "Generation 115/300\n", + "Solving for Nash Equilibrium in Generation 115/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.9851 - 1ms/epoch - 20us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.4770 - 738us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.5078 - 590us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 229.1320 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.5757 - 630us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.6851 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.3954 - 670us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.8379 - 792us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.0414 - 1ms/epoch - 17us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.4756 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.3109 - 731us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 803.7352 - 721us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.8433 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.4940 - 745us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.8520 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1211 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.0605 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.2007 - 816us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8605 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.6797 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.8437 - 657us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.1770 - 782us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.9993 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.5891 - 618us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.7044 - 701us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.5695 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4301 - 703us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.2276 - 699us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.9335 - 755us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.2706 - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.6303 - 743us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.0262 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.4630 - 2ms/epoch - 25us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.8365 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.2089 - 673us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.6016 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.5208 - 795us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8918 - 817us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4228 - 755us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.8739 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5918 - 731us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.3267 - 1ms/epoch - 19us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.0350 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.8181 - 691us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.3957 - 635us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.4409 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7346 - 745us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.0004 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7283 - 820us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 16ms/epoch - 261us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.3038 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.1602 - 792us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.3372 - 810us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.2312 - 868us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7922 - 912us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.2634 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.1479 - 725us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.5659 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.0247 - 2ms/epoch - 36us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.9658 - 688us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.0306 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.2755 - 673us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.1016 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6627 - 851us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.2768 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.3284 - 1ms/epoch - 20us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.8378 - 692us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.5921 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.9643 - 698us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.0595 - 857us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.9652 - 1ms/epoch - 21us/sample\n", + "Generation 116/300\n", + "Solving for Nash Equilibrium in Generation 116/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.1507 - 651us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.1890 - 660us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3828 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9760 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4979 - 959us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.3997 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.0117 - 765us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.0688 - 838us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.1981 - 969us/epoch - 16us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.0567 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8182 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.1946 - 1ms/epoch - 17us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9215 - 745us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.1207 - 640us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.2808 - 786us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.7938 - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.1752 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.1608 - 854us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.5847 - 770us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.8991 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.6755 - 844us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.3481 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.5884 - 561us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.2328 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.6217 - 916us/epoch - 15us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.8185 - 631us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.8613 - 738us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6057 - 572us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.0577 - 812us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.5020 - 939us/epoch - 15us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.5826 - 705us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6833 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7308 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.9420 - 591us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.7239 - 679us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.2846 - 663us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.2345 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.9670 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6119 - 865us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.2852 - 622us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.2623 - 992us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.2303 - 725us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.6474 - 648us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.6014 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.3951 - 874us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.5507 - 777us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.6209 - 740us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.0854 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1917 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.2789 - 4ms/epoch - 64us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.9120 - 734us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.8079 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.0764 - 728us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.1727 - 885us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.6077 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.9926 - 693us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.5457 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.3552 - 655us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.7545 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.1555 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.1386 - 2ms/epoch - 26us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.4656 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.1447 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5099 - 711us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.0159 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.7151 - 759us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.5762 - 687us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1888 - 727us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.6070 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.5772 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.3810 - 682us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.3298 - 724us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8336 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.6147 - 1ms/epoch - 21us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.0717 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.0627 - 631us/epoch - 10us/sample\n", + "Generation 117/300\n", + "Solving for Nash Equilibrium in Generation 117/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.9106 - 726us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.3608 - 686us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4361 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.4095 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.4070 - 698us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8068 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.9515 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.8683 - 632us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.0064 - 669us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5626 - 756us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8076 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6287 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.9122 - 652us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.1423 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.2422 - 607us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3364 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.0413 - 852us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4369 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.3585 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.3635 - 669us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.6312 - 869us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.0502 - 619us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.7484 - 906us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9124 - 763us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 798.7646 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773.8197 - 666us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 168.3531 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.3806 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.4009 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.7727 - 835us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.5668 - 980us/epoch - 16us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 76ms/epoch - 1ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.6862 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.3846 - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.4648 - 646us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.8976 - 804us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.5275 - 1ms/epoch - 19us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.8080 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.7885 - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7974 - 2ms/epoch - 29us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.9579 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.6605 - 993us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.3723 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.8739 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.4457 - 872us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.4252 - 955us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.9042 - 7ms/epoch - 117us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.8937 - 987us/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0489 - 952us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.5092 - 902us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.9439 - 656us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.2229 - 5ms/epoch - 77us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.3224 - 750us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.9198 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.4865 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2643 - 2ms/epoch - 30us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.5729 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.8508 - 577us/epoch - 9us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.2628 - 661us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.9939 - 788us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.8264 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.2308 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.6354 - 703us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5248 - 928us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.7713 - 763us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.1924 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5427 - 844us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.0805 - 567us/epoch - 9us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.6190 - 953us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9738 - 1ms/epoch - 18us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.6382 - 773us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.2088 - 874us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9262 - 1ms/epoch - 17us/sample\n", + "Generation 118/300\n", + "Solving for Nash Equilibrium in Generation 118/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 796.1811 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.7766 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.1264 - 983us/epoch - 16us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.4392 - 732us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.7896 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3353 - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.1985 - 828us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5717 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.9391 - 784us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.0435 - 1ms/epoch - 17us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.5031 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.0920 - 712us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.9434 - 921us/epoch - 15us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.8801 - 2ms/epoch - 29us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.4243 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.9204 - 751us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6968 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.2017 - 1ms/epoch - 18us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.6190 - 1ms/epoch - 20us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7133 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.0543 - 766us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.0659 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.5677 - 1ms/epoch - 23us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.5764 - 771us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.8971 - 648us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8984 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3300 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.1581 - 777us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.3860 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.9628 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.8246 - 695us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.0757 - 727us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.3499 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0552 - 730us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.7216 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.3727 - 725us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.2862 - 617us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.7189 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.5226 - 2ms/epoch - 32us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.5356 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.5158 - 765us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8893 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.4258 - 722us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7689 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.4700 - 795us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.8336 - 665us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.3831 - 668us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.5526 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.0274 - 8ms/epoch - 122us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.7892 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5621 - 651us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7299 - 683us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.4820 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6540 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5697 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361.5848 - 661us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.7728 - 1ms/epoch - 23us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.8035 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.6302 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.0730 - 656us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3939 - 743us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.2222 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0766 - 701us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4890 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.8157 - 675us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6033 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.7429 - 763us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1939 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7897 - 2ms/epoch - 28us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.1328 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.1542 - 649us/epoch - 10us/sample\n", + "Generation 119/300\n", + "Solving for Nash Equilibrium in Generation 119/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.3926 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7168 - 1ms/epoch - 16us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.1930 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.0483 - 665us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9655 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.6103 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.6202 - 888us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7309 - 4ms/epoch - 69us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.1198 - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.8142 - 1ms/epoch - 20us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.9632 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5821 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.3516 - 684us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.3566 - 978us/epoch - 16us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.9244 - 797us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.1609 - 1ms/epoch - 22us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.2105 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9685 - 959us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.5644 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6855 - 763us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.5170 - 751us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5071 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.9078 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.4844 - 772us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.2368 - 998us/epoch - 16us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.3391 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.1341 - 957us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.8677 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.1688 - 972us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.6635 - 934us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.2101 - 848us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.4466 - 2ms/epoch - 28us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.9409 - 767us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.8264 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.5096 - 840us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235.7522 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.6658 - 6ms/epoch - 104us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.8823 - 1ms/epoch - 18us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1237 - 699us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.7241 - 4ms/epoch - 62us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 132.0032 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.1928 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1627 - 636us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9507 - 730us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.1603 - 597us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1418 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.7221 - 1ms/epoch - 20us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.9874 - 575us/epoch - 9us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.8106 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.8245 - 1ms/epoch - 16us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.9545 - 870us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1652 - 684us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.1285 - 837us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.3308 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.1209 - 850us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.1041 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.8220 - 984us/epoch - 16us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8131 - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1689 - 719us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.5980 - 744us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4565 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.0217 - 992us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.4595 - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2248 - 866us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.3763 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1337 - 880us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.1303 - 787us/epoch - 13us/sample\n", + "Generation 120/300\n", + "Solving for Nash Equilibrium in Generation 120/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.7676 - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 133us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8825 - 2ms/epoch - 28us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.9085 - 784us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.2524 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8810 - 590us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4383 - 812us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.9266 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6576 - 6ms/epoch - 101us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.8093 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9792 - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.6995 - 704us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.2306 - 1000us/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.5069 - 60ms/epoch - 963us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.9240 - 728us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.5417 - 655us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3414 - 734us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.5605 - 753us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.1591 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.3215 - 763us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3680 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0698 - 969us/epoch - 16us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.9171 - 628us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.3824 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.8689 - 709us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.6354 - 686us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.2266 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4263 - 4ms/epoch - 66us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.3064 - 753us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6718 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.0166 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.8566 - 676us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.9417 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 157us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.9921 - 875us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.9249 - 832us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.3029 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.3648 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.4611 - 2ms/epoch - 30us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.5625 - 993us/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0122 - 1ms/epoch - 21us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.8555 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.7133 - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.6642 - 1ms/epoch - 18us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.6736 - 993us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.9050 - 848us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2653 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.9231 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.0800 - 737us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1173 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.4893 - 3ms/epoch - 48us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.7939 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.7633 - 679us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2742 - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.2475 - 733us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2560 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7928 - 719us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.9365 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.5033 - 662us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8594 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.2557 - 956us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0328 - 761us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.1046 - 898us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.1514 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3073 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.8483 - 830us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.4196 - 716us/epoch - 12us/sample\n", + "Generation 121/300\n", + "Solving for Nash Equilibrium in Generation 121/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9961 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.0888 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.4294 - 1ms/epoch - 17us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.3593 - 1ms/epoch - 20us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.4672 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.4435 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7118 - 881us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.6024 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.8191 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.6279 - 3ms/epoch - 54us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.7693 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4507 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.6013 - 3ms/epoch - 40us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.6047 - 680us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0692 - 890us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.2338 - 707us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1306 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.7040 - 897us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.3298 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.4911 - 828us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.7288 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.9930 - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.7249 - 899us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5925 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.6063 - 2ms/epoch - 39us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.7257 - 787us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.6859 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.7990 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.2999 - 762us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.8120 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.1999 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.3528 - 670us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.2045 - 996us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4460 - 859us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.1159 - 912us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2492 - 1ms/epoch - 18us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.1345 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2217 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.5236 - 670us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7487 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.6962 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.7876 - 765us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.3882 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.7091 - 881us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4579 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2172 - 1ms/epoch - 18us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.7599 - 1ms/epoch - 19us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1319 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 178.8918 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.4771 - 840us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.9727 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.3748 - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.4945 - 1ms/epoch - 18us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8409 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126.6989 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.8853 - 1ms/epoch - 19us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.1756 - 1ms/epoch - 17us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4643 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.0022 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.3923 - 2ms/epoch - 27us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.3462 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.5876 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.9801 - 1ms/epoch - 19us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.4459 - 989us/epoch - 16us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.6053 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.5991 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.0482 - 579us/epoch - 9us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.7505 - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.6699 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.6061 - 656us/epoch - 11us/sample\n", + "Generation 122/300\n", + "Solving for Nash Equilibrium in Generation 122/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.0844 - 709us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4023 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.5440 - 792us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.8976 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4695 - 690us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.2960 - 1ms/epoch - 17us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4634 - 723us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.8036 - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.4796 - 6ms/epoch - 91us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.0108 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.9655 - 1ms/epoch - 17us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.5335 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.1624 - 1ms/epoch - 16us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.5126 - 715us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.6309 - 788us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4728 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.3632 - 2ms/epoch - 30us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8103 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.0898 - 832us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.7835 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.9782 - 822us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3709 - 859us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6988 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.0729 - 833us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.3225 - 838us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.2091 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1840 - 1ms/epoch - 22us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.3441 - 1ms/epoch - 17us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.1417 - 1ms/epoch - 21us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3560 - 831us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46.5374 - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1837 - 923us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.4290 - 661us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.6245 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7363 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.3062 - 2ms/epoch - 34us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.4501 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.6691 - 2ms/epoch - 36us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.6834 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 102us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1661 - 3ms/epoch - 56us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.6106 - 734us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.0079 - 9ms/epoch - 146us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6252 - 870us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.4576 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.7325 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.7321 - 908us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428.7917 - 929us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.3435 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.6696 - 21ms/epoch - 332us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.5953 - 955us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9691 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.3298 - 718us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.4352 - 604us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.2906 - 755us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.8398 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.2849 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0768 - 5ms/epoch - 80us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.0505 - 757us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0510 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.2587 - 694us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1679 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.0558 - 881us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.4615 - 683us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.9786 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2692 - 586us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1561 - 2ms/epoch - 25us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.5601 - 639us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.6973 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5711 - 784us/epoch - 13us/sample\n", + "Generation 123/300\n", + "Solving for Nash Equilibrium in Generation 123/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.2176 - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.3386 - 923us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.7286 - 875us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7812 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5139 - 776us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.8606 - 748us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.5059 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2471 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8529 - 660us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.5326 - 882us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.4965 - 780us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.0042 - 3ms/epoch - 55us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.3778 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.2742 - 979us/epoch - 16us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7494 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.5725 - 821us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.8525 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.2281 - 3ms/epoch - 55us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.2135 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.1367 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.0728 - 2ms/epoch - 30us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.4750 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.9665 - 1ms/epoch - 17us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 11ms/epoch - 180us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.0372 - 14ms/epoch - 233us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8303 - 2ms/epoch - 32us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6673 - 687us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0807 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.9910 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.7252 - 634us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0990 - 589us/epoch - 9us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.1390 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.8737 - 1ms/epoch - 23us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 14ms/epoch - 220us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.5388 - 11ms/epoch - 184us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.7798 - 827us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.2752 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.4042 - 811us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.1573 - 727us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.0201 - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.8373 - 880us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.2302 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.5280 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5353 - 1ms/epoch - 16us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.4750 - 861us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.9637 - 3ms/epoch - 54us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.3464 - 760us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.5754 - 875us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.4175 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 14ms/epoch - 229us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2346 - 2ms/epoch - 26us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.6959 - 840us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4819 - 1ms/epoch - 18us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.3533 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3253 - 2ms/epoch - 27us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.0114 - 1ms/epoch - 17us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.8198 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.0518 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.2169 - 844us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.8306 - 1ms/epoch - 21us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.0994 - 832us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.1834 - 642us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.7723 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.3609 - 789us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250.3674 - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.6840 - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.3301 - 758us/epoch - 12us/sample\n", + "Generation 124/300\n", + "Solving for Nash Equilibrium in Generation 124/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.9019 - 1ms/epoch - 21us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.7618 - 822us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.0422 - 869us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.2077 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.5876 - 1ms/epoch - 17us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.0695 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 151us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.6569 - 2ms/epoch - 26us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.5121 - 1ms/epoch - 20us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.7442 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9600 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9930 - 1ms/epoch - 22us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.5753 - 1ms/epoch - 18us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.6340 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6735 - 935us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.2815 - 805us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2176 - 3ms/epoch - 45us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2638 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.3144 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.9471 - 1ms/epoch - 19us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.0016 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9764 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1614 - 701us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.6942 - 888us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.0963 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.4888 - 1ms/epoch - 19us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3502 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.2013 - 11ms/epoch - 180us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5891 - 2ms/epoch - 24us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.9335 - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.7270 - 1ms/epoch - 22us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.3027 - 3ms/epoch - 54us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.5297 - 1ms/epoch - 18us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.9761 - 2ms/epoch - 30us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.7717 - 11ms/epoch - 176us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.0663 - 717us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.6774 - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.8066 - 930us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.8859 - 826us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6863 - 842us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.5464 - 783us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.4854 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.6515 - 728us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.6593 - 972us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.1354 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.9590 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.4045 - 661us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.8112 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9067 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.3633 - 851us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.3603 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 25ms/epoch - 410us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.2212 - 10ms/epoch - 161us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4996 - 846us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.0937 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.6434 - 768us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.7547 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8101 - 761us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1443 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.2034 - 745us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.8660 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.8201 - 1ms/epoch - 20us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.4666 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.3723 - 1ms/epoch - 19us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.1148 - 985us/epoch - 16us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.3856 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4942 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4220 - 831us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.3979 - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.1859 - 765us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7695 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.7828 - 988us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.3490 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.9238 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 137us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.5816 - 953us/epoch - 15us/sample\n", + "Generation 125/300\n", + "Solving for Nash Equilibrium in Generation 125/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9638 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3301 - 2ms/epoch - 25us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7794 - 760us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0040 - 995us/epoch - 16us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.8342 - 687us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.3125 - 1ms/epoch - 18us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.3897 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8452 - 839us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.7023 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.9873 - 672us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.0653 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3359 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.7335 - 764us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8712 - 762us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.6558 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7825 - 838us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.7698 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.4514 - 728us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.1214 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.8820 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8378 - 847us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.5090 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.6903 - 734us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.9645 - 779us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5778 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.9677 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1395 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.5531 - 779us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9201 - 740us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.6703 - 787us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.0199 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.4608 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.7658 - 820us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.8315 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.3081 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7819 - 2ms/epoch - 24us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.8264 - 785us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.5327 - 750us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.8199 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.6349 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9534 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8369 - 637us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.1930 - 663us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8144 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0944 - 2ms/epoch - 28us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.0232 - 926us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4353 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.0685 - 919us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.2548 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0172 - 771us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.2461 - 749us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.7863 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.4344 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8688 - 2ms/epoch - 33us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.9095 - 1ms/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9974 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.9880 - 711us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4744 - 969us/epoch - 16us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.0939 - 540us/epoch - 9us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.9290 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.7370 - 865us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6340 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.3846 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.8200 - 888us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.4442 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.8300 - 643us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2654 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 18ms/epoch - 285us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.9731 - 7ms/epoch - 107us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0355 - 1ms/epoch - 24us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.9040 - 962us/epoch - 16us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6150 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.0476 - 967us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.3999 - 746us/epoch - 12us/sample\n", + "Generation 126/300\n", + "Solving for Nash Equilibrium in Generation 126/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.4554 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.9140 - 849us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.9053 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.0850 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.0101 - 928us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.4034 - 716us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1364 - 4ms/epoch - 69us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.2672 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.0165 - 909us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.0406 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.6174 - 792us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 98us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.7105 - 3ms/epoch - 48us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.3211 - 880us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.3331 - 3ms/epoch - 50us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79.8626 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.3706 - 744us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.2975 - 965us/epoch - 16us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6421 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.2408 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.3536 - 699us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.1750 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.7196 - 739us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1386 - 861us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.0912 - 875us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.5981 - 788us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0441 - 2ms/epoch - 25us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.1656 - 1ms/epoch - 16us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.8810 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7104 - 637us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.6589 - 870us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.1736 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.7785 - 828us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.3575 - 1ms/epoch - 21us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.6530 - 799us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.6176 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.1391 - 951us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.4094 - 922us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.1270 - 896us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.0781 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.8052 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.1280 - 1ms/epoch - 21us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.1462 - 918us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.8725 - 1ms/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.7993 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.4466 - 832us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.2596 - 719us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5345 - 1ms/epoch - 21us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.2417 - 1ms/epoch - 20us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.6840 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.9196 - 2ms/epoch - 25us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.9160 - 1ms/epoch - 18us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.3474 - 1ms/epoch - 22us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.0517 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8687 - 958us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83.8214 - 2ms/epoch - 25us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.0243 - 996us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.0687 - 758us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2267 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.0695 - 2ms/epoch - 28us/sample\n", + "Generation 127/300\n", + "Solving for Nash Equilibrium in Generation 127/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.3200 - 2ms/epoch - 30us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.1611 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.3777 - 5ms/epoch - 85us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3866 - 754us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.4473 - 624us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.9489 - 1ms/epoch - 17us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1277 - 2ms/epoch - 40us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.8397 - 777us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.1944 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 127us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7741 - 2ms/epoch - 26us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.2333 - 1ms/epoch - 18us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.5374 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2645 - 935us/epoch - 15us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7799 - 948us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.0151 - 799us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.0909 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.0107 - 693us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.7141 - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.2625 - 931us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.9875 - 819us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.7974 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.9371 - 745us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.1737 - 1ms/epoch - 20us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.6446 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.1913 - 717us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.1586 - 3ms/epoch - 53us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0048 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 148us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.6174 - 3ms/epoch - 47us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.3661 - 911us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.2848 - 2ms/epoch - 39us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.7581 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.8256 - 5ms/epoch - 77us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.9559 - 845us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.9415 - 889us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2947 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.3110 - 840us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2341 - 849us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 127us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8224 - 5ms/epoch - 79us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.4315 - 1ms/epoch - 18us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.0091 - 685us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6195 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7249 - 920us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32.7069 - 1ms/epoch - 17us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.1443 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.1906 - 700us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.5828 - 957us/epoch - 15us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.2891 - 2ms/epoch - 37us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.7650 - 2ms/epoch - 27us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.3474 - 13ms/epoch - 216us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9990 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.3503 - 759us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.2547 - 1ms/epoch - 17us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.8037 - 916us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.3514 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.9185 - 767us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.6776 - 1ms/epoch - 16us/sample\n", + "Episode 50/50\n", + "Generation 128/300\n", + "Solving for Nash Equilibrium in Generation 128/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 219.9710 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.4754 - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.5207 - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.1203 - 8ms/epoch - 132us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.6729 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.3645 - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.6643 - 785us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0042 - 669us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4380 - 768us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.7167 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.6850 - 6ms/epoch - 101us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.4045 - 1ms/epoch - 23us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.3497 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.6653 - 1ms/epoch - 16us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.0806 - 939us/epoch - 15us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3740 - 1ms/epoch - 22us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.6141 - 859us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.0565 - 1ms/epoch - 24us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.6266 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.9178 - 753us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.5890 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 104us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.0914 - 5ms/epoch - 80us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.0264 - 813us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.6447 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.6109 - 933us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.3390 - 825us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.6354 - 1ms/epoch - 18us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.2944 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.9445 - 1ms/epoch - 17us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2891 - 6ms/epoch - 99us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.6973 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.6664 - 665us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.6998 - 844us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.0311 - 1ms/epoch - 20us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.3699 - 1ms/epoch - 17us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2467 - 853us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.1214 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.8911 - 729us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1563 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 140us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.2992 - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4529 - 911us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.4365 - 755us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.8958 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2252 - 811us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0726 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4595 - 781us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.0347 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.3041 - 1ms/epoch - 18us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.9081 - 713us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.8507 - 2ms/epoch - 39us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0949 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.1368 - 709us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.0817 - 940us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.4348 - 749us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0136 - 540us/epoch - 9us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.2326 - 668us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.3286 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.5690 - 681us/epoch - 11us/sample\n", + "Generation 129/300\n", + "Solving for Nash Equilibrium in Generation 129/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.6604 - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.6052 - 825us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.7184 - 714us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.0661 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.5395 - 703us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.1243 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.0334 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.7549 - 719us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3220 - 1ms/epoch - 17us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.6609 - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.4939 - 787us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.0092 - 714us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.9318 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2248 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5933 - 825us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.8136 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 793.7245 - 982us/epoch - 16us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6144 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.4288 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.0729 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4271 - 879us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.8401 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9542 - 848us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.6396 - 779us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.7069 - 768us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4561 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.7717 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.5637 - 1ms/epoch - 17us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8852 - 850us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.6615 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.7805 - 813us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.2494 - 721us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.2746 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.3677 - 1ms/epoch - 17us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.5538 - 957us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.4901 - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.4747 - 3ms/epoch - 55us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.2444 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.6976 - 787us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1917 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.9131 - 955us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.4957 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.8226 - 807us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.8596 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8852 - 767us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.4963 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.7317 - 1ms/epoch - 17us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.7558 - 827us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.7212 - 2ms/epoch - 30us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.3717 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0134 - 697us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46.3504 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.0219 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4084 - 880us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9599 - 1ms/epoch - 18us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0452 - 1ms/epoch - 21us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.3783 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.0715 - 3ms/epoch - 47us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.2354 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1440 - 1ms/epoch - 17us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.7821 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.7483 - 786us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6547 - 1ms/epoch - 17us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.2423 - 2ms/epoch - 26us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0552 - 811us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6013 - 713us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.3245 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.9402 - 10ms/epoch - 157us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8692 - 796us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50.4874 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.5340 - 875us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.5931 - 762us/epoch - 12us/sample\n", + "Generation 130/300\n", + "Solving for Nash Equilibrium in Generation 130/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.4229 - 1ms/epoch - 18us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.3743 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.9114 - 1ms/epoch - 22us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6365 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.5671 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 797.4105 - 892us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.0606 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.4206 - 939us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.7346 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.9093 - 814us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8085 - 861us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.7058 - 827us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.0325 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.3399 - 2ms/epoch - 27us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9909 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.1814 - 641us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5204 - 809us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.2154 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0634 - 774us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9048 - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.0469 - 1ms/epoch - 17us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.5345 - 1ms/epoch - 22us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.0659 - 5ms/epoch - 84us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1049 - 788us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.9669 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.0074 - 1ms/epoch - 19us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.7638 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.7393 - 1ms/epoch - 18us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8269 - 3ms/epoch - 51us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.7150 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.5598 - 2ms/epoch - 31us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.1212 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.8420 - 932us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.0763 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6746 - 770us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9240 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.4904 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.9603 - 791us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.0291 - 1ms/epoch - 19us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.5658 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.1246 - 839us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.0375 - 779us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.9464 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3928 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 362.7666 - 782us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.9214 - 778us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.2761 - 704us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.6716 - 891us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.6236 - 1ms/epoch - 19us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.5523 - 790us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.6963 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2297 - 760us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5754 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6995 - 1ms/epoch - 23us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.1711 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.7468 - 650us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.8036 - 1ms/epoch - 20us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.9115 - 988us/epoch - 16us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.4000 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.8188 - 1ms/epoch - 19us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.1658 - 982us/epoch - 16us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.3806 - 10ms/epoch - 165us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.6073 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.0631 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.5761 - 4ms/epoch - 62us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3398 - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.6953 - 6ms/epoch - 103us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.8883 - 1ms/epoch - 23us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.1699 - 6ms/epoch - 91us/sample\n", + "Generation 131/300\n", + "Solving for Nash Equilibrium in Generation 131/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.5092 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.2791 - 986us/epoch - 16us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5612 - 767us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3042 - 592us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.1079 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.7260 - 737us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5592 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.3958 - 1ms/epoch - 17us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.6977 - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9989 - 873us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.8196 - 1ms/epoch - 21us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.9748 - 713us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.5217 - 800us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.1999 - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0448 - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4268 - 800us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.7292 - 4ms/epoch - 58us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.8841 - 835us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.8022 - 669us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.8154 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.9925 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.5012 - 614us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.1864 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.5787 - 854us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0080 - 690us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.4032 - 771us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 783.0167 - 2ms/epoch - 37us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.6190 - 901us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.5207 - 754us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6811 - 7ms/epoch - 119us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.3564 - 2ms/epoch - 27us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.5277 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.3037 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.8537 - 782us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.7006 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.7701 - 807us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.4023 - 596us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.7070 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.7353 - 598us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.9105 - 1ms/epoch - 23us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.9040 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.9803 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.2426 - 745us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.5092 - 885us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.9257 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.8366 - 1ms/epoch - 18us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.5622 - 867us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.5837 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.2002 - 948us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.8980 - 928us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8664 - 727us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.0561 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2518 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7501 - 739us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.1411 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.8668 - 756us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.9055 - 850us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.6779 - 614us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.7836 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.8458 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0517 - 796us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.6135 - 751us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.9620 - 709us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2559 - 731us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.0137 - 2ms/epoch - 36us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.6291 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8531 - 869us/epoch - 14us/sample\n", + "Generation 132/300\n", + "Solving for Nash Equilibrium in Generation 132/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4533 - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.2601 - 638us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.3241 - 653us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.4168 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.1185 - 759us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.3213 - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.1443 - 680us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8856 - 711us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.9941 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5383 - 711us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240.6311 - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.9011 - 706us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 152.8723 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.7397 - 966us/epoch - 16us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1481 - 910us/epoch - 15us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.6408 - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.4539 - 682us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.0264 - 711us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.2447 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.7138 - 690us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.5248 - 824us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.3704 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.1667 - 627us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6855 - 712us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2719 - 915us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8620 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 175.2509 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7908 - 1ms/epoch - 22us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.4896 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.3148 - 2ms/epoch - 40us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.1640 - 668us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.6214 - 888us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4450 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.5958 - 1ms/epoch - 18us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.6238 - 832us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.0504 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2340 - 767us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.2161 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.9381 - 1ms/epoch - 21us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.5864 - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.3694 - 840us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.6360 - 839us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.3707 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.4306 - 759us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.0431 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.0316 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.6306 - 866us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.7645 - 819us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.0552 - 869us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5758 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3787 - 6ms/epoch - 100us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5366 - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6755 - 2ms/epoch - 25us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.8210 - 1ms/epoch - 21us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.7814 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.8334 - 12ms/epoch - 186us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.2327 - 1ms/epoch - 22us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.0278 - 952us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.4923 - 11ms/epoch - 174us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6139 - 1ms/epoch - 22us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.8047 - 970us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.2148 - 1ms/epoch - 24us/sample\n", + "Generation 133/300\n", + "Solving for Nash Equilibrium in Generation 133/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8716 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.6401 - 8ms/epoch - 131us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46.2513 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.6859 - 2ms/epoch - 32us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.0486 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 16ms/epoch - 261us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.5096 - 1ms/epoch - 17us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8210 - 931us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.8470 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.9194 - 831us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.7670 - 1ms/epoch - 17us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.7451 - 2ms/epoch - 37us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.8251 - 2ms/epoch - 32us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428.7275 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.4172 - 718us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.7769 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.1085 - 2ms/epoch - 39us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2732 - 2ms/epoch - 32us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.2430 - 1ms/epoch - 20us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3317 - 1ms/epoch - 20us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7072 - 1ms/epoch - 17us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.9831 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.4302 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.0023 - 760us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.2881 - 977us/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.5455 - 1ms/epoch - 23us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.5663 - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.6594 - 7ms/epoch - 110us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8220 - 1ms/epoch - 17us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.1471 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.5247 - 3ms/epoch - 42us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.3347 - 834us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.0178 - 743us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.7140 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.5381 - 1ms/epoch - 17us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.2744 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8985 - 793us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.9954 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6136 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.5330 - 2ms/epoch - 28us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2765 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.1852 - 2ms/epoch - 28us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.0554 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.9112 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5911 - 963us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.5902 - 938us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.7399 - 857us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7321 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.8502 - 788us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.7090 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.9918 - 1ms/epoch - 17us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.7351 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.8192 - 1ms/epoch - 17us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.6791 - 1ms/epoch - 18us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.7676 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.1359 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.4144 - 854us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.1469 - 959us/epoch - 15us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.3013 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.1307 - 753us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 789.9073 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5991 - 1ms/epoch - 21us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.5401 - 1ms/epoch - 23us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 362.8751 - 1ms/epoch - 19us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.3949 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2516 - 716us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.6973 - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.9847 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.5296 - 776us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.8063 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.4747 - 740us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 134/300\n", + "Solving for Nash Equilibrium in Generation 134/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.1617 - 914us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.8047 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.0626 - 2ms/epoch - 24us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.6422 - 2ms/epoch - 24us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.4166 - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.7516 - 786us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7185 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.1277 - 645us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.6721 - 843us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.2066 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3242 - 1ms/epoch - 19us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.9881 - 789us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.1102 - 787us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245.9561 - 802us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.8389 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.3275 - 776us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.0588 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.3724 - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.8926 - 822us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4774 - 693us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.9239 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9666 - 533us/epoch - 9us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.6332 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8605 - 734us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0109 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.3422 - 977us/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0923 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.1061 - 648us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.1528 - 746us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.1705 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2504 - 784us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.3004 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.4792 - 768us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.9229 - 808us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.4171 - 659us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.9382 - 1ms/epoch - 16us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.4383 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.4525 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.7820 - 4ms/epoch - 63us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.5891 - 1ms/epoch - 17us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.1338 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.7145 - 3ms/epoch - 49us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.6305 - 728us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.2848 - 762us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.7495 - 796us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.3958 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.2054 - 710us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.9916 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.6587 - 701us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.1281 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.1473 - 1ms/epoch - 20us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3841 - 784us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.4669 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.2713 - 736us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1500 - 692us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.6766 - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.6956 - 872us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 182.0716 - 711us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9976 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.5294 - 670us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2437 - 802us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.2717 - 856us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.2566 - 943us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1089 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.3887 - 717us/epoch - 12us/sample\n", + "Generation 135/300\n", + "Solving for Nash Equilibrium in Generation 135/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.9699 - 570us/epoch - 9us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.0246 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 134us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.4729 - 881us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.7426 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.4947 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.4866 - 904us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5988 - 985us/epoch - 16us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 115us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4586 - 1ms/epoch - 22us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4274 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.0743 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.2413 - 923us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8622 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2617 - 885us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.2532 - 871us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.0402 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6700 - 1ms/epoch - 20us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.0981 - 817us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0644 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1524 - 2ms/epoch - 28us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48.6835 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.4691 - 746us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6761 - 1ms/epoch - 20us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.4354 - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.1726 - 785us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.3842 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.6240 - 1ms/epoch - 24us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4251 - 829us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.1183 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.6133 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.1097 - 1ms/epoch - 17us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.8567 - 2ms/epoch - 29us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.1196 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.6059 - 1ms/epoch - 16us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.1941 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.4382 - 826us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.0823 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.7142 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.3563 - 746us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.2388 - 1ms/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.0754 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.2593 - 919us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.8195 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.5124 - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.5457 - 681us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.3191 - 1ms/epoch - 20us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.8891 - 864us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.2367 - 783us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.8186 - 44ms/epoch - 717us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.1879 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.9286 - 2ms/epoch - 36us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.0410 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.5878 - 2ms/epoch - 26us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.0249 - 1ms/epoch - 22us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.2992 - 941us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.3930 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.0321 - 874us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7227 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.7142 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.3459 - 911us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.0717 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.6990 - 1ms/epoch - 21us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.0323 - 2ms/epoch - 33us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.8997 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.3238 - 1ms/epoch - 19us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.0080 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.1196 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 123us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.1182 - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.5574 - 964us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.1592 - 898us/epoch - 14us/sample\n", + "Generation 136/300\n", + "Solving for Nash Equilibrium in Generation 136/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.7095 - 759us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.1772 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4719 - 2ms/epoch - 40us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2189 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.8658 - 1ms/epoch - 21us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6989 - 960us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.4529 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.9523 - 749us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.4251 - 1ms/epoch - 24us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 790.3416 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8565 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.9110 - 8ms/epoch - 125us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.5643 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.5870 - 806us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.6965 - 1ms/epoch - 21us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.0417 - 942us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3744 - 912us/epoch - 15us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.1149 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245.1871 - 2ms/epoch - 33us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.4863 - 801us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8322 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.4378 - 641us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.4115 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5950 - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.2928 - 770us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.1465 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.7658 - 3ms/epoch - 54us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.2091 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0079 - 991us/epoch - 16us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2684 - 861us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787.2533 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.5048 - 1ms/epoch - 19us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.6622 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.6159 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5540 - 2ms/epoch - 28us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.0522 - 872us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9148 - 795us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.2090 - 9ms/epoch - 148us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.0112 - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7168 - 963us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1163 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.0336 - 1ms/epoch - 18us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.6682 - 963us/epoch - 16us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361.6968 - 999us/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.4733 - 704us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.4932 - 915us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.2191 - 939us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.0334 - 5ms/epoch - 82us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.7943 - 987us/epoch - 16us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.6959 - 853us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.3256 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.6340 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.3291 - 797us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.9603 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.8189 - 1ms/epoch - 18us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.1902 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.5254 - 767us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.2348 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.6969 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.7891 - 896us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.6334 - 985us/epoch - 16us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4773 - 9ms/epoch - 140us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.3849 - 930us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.2049 - 2ms/epoch - 25us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9540 - 828us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.1489 - 719us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6407 - 745us/epoch - 12us/sample\n", + "Generation 137/300\n", + "Solving for Nash Equilibrium in Generation 137/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.2015 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2985 - 678us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.2837 - 1ms/epoch - 18us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.2967 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.9714 - 825us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.1451 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.0283 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.5836 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.7008 - 621us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.6351 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.4008 - 883us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.2939 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1755 - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.9622 - 673us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.0427 - 727us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428.2193 - 3ms/epoch - 41us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0443 - 903us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.7764 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 793.2068 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.0001 - 658us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.5724 - 851us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.1451 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.7054 - 917us/epoch - 15us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.2909 - 4ms/epoch - 64us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.3375 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6074 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.3228 - 826us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2026 - 643us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.2223 - 2ms/epoch - 26us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3511 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.5129 - 695us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7390 - 781us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.2462 - 684us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.6536 - 903us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1786 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.6612 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.5871 - 817us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.7802 - 695us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.6622 - 756us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.6646 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.6580 - 707us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.5656 - 797us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 12ms/epoch - 197us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.6153 - 2ms/epoch - 28us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.9023 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.0356 - 782us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.0854 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.6122 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.6875 - 742us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2146 - 705us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.3582 - 1ms/epoch - 20us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.7811 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1719 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.4380 - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.3958 - 775us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.2889 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.2196 - 803us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.9850 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.2390 - 718us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.8333 - 904us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.9883 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.1549 - 2ms/epoch - 29us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.0004 - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.5731 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.8811 - 759us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.5174 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.3925 - 1ms/epoch - 18us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.1590 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1270 - 862us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.9517 - 1ms/epoch - 24us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3791 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.3558 - 1ms/epoch - 20us/sample\n", + "Episode 50/50\n", + "Generation 138/300\n", + "Solving for Nash Equilibrium in Generation 138/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0279 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.8937 - 892us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.3103 - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.3373 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.7223 - 1ms/epoch - 18us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8891 - 2ms/epoch - 37us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.2653 - 1ms/epoch - 19us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.5193 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.3722 - 776us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.0494 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 11ms/epoch - 182us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.0762 - 7ms/epoch - 114us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.5009 - 810us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.5787 - 762us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5157 - 850us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.6545 - 886us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.5804 - 790us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.0681 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.6417 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.9726 - 826us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98.3670 - 1ms/epoch - 17us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.3894 - 806us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.3675 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.1008 - 774us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.0960 - 1ms/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0604 - 1ms/epoch - 19us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.1676 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.9904 - 3ms/epoch - 49us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8199 - 2ms/epoch - 32us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.2008 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.3152 - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.1776 - 1ms/epoch - 16us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.4785 - 792us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.3627 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.1829 - 1ms/epoch - 20us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.8594 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.8089 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.4301 - 2ms/epoch - 28us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.1477 - 4ms/epoch - 72us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8657 - 664us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.0283 - 703us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.2075 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4448 - 754us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.8026 - 972us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.8557 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8290 - 666us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.6871 - 1ms/epoch - 23us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.2301 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.1406 - 575us/epoch - 9us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.5905 - 836us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.0609 - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4131 - 857us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 84us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.5691 - 8ms/epoch - 132us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.4156 - 901us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.6722 - 841us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6822 - 760us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5496 - 708us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4088 - 865us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.7578 - 755us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0591 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.5218 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.1249 - 3ms/epoch - 48us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.6592 - 931us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.8213 - 672us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6299 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.6635 - 679us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.6483 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.2480 - 660us/epoch - 11us/sample\n", + "Generation 139/300\n", + "Solving for Nash Equilibrium in Generation 139/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.6475 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1278 - 1ms/epoch - 17us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.7032 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.2280 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.1917 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.8727 - 6ms/epoch - 90us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7946 - 885us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0422 - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.5195 - 868us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.7316 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.5952 - 807us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.9235 - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.2678 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5688 - 783us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.0025 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8074 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.8436 - 3ms/epoch - 42us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.5781 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.0562 - 8ms/epoch - 125us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.4958 - 860us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9385 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6973 - 773us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.2135 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.9217 - 2ms/epoch - 25us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.5245 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.8064 - 819us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 124us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.9185 - 2ms/epoch - 36us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.8871 - 2ms/epoch - 35us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4448 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.1188 - 1ms/epoch - 18us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.0885 - 856us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.7567 - 5ms/epoch - 87us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3204 - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.0560 - 2ms/epoch - 25us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.7547 - 952us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2703 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.6066 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.1216 - 1ms/epoch - 18us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2150 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.0066 - 1ms/epoch - 24us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9199 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.5962 - 2ms/epoch - 27us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8511 - 2ms/epoch - 26us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.8992 - 32ms/epoch - 511us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.6391 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.2952 - 3ms/epoch - 44us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.5235 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.4543 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.3341 - 2ms/epoch - 36us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9814 - 1ms/epoch - 24us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.5225 - 1ms/epoch - 23us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.9919 - 1ms/epoch - 18us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.5910 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.7474 - 1ms/epoch - 21us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.8766 - 1ms/epoch - 23us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 116us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.3400 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.4508 - 817us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.5555 - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.6019 - 2ms/epoch - 25us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.1725 - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.6096 - 785us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.7748 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7535 - 861us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.7030 - 842us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4025 - 754us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.3969 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9223 - 4ms/epoch - 62us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.0345 - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7645 - 885us/epoch - 14us/sample\n", + "Generation 140/300\n", + "Solving for Nash Equilibrium in Generation 140/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.5894 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.7388 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.4791 - 855us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.2272 - 672us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.1806 - 7ms/epoch - 108us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.9573 - 1ms/epoch - 18us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.4059 - 2ms/epoch - 26us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.1876 - 841us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.8187 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.0221 - 769us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.7708 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.7508 - 3ms/epoch - 47us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.4650 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4281 - 1ms/epoch - 17us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.9592 - 811us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.6674 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6685 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.1885 - 709us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9161 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.9757 - 609us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.5359 - 793us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4990 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.0990 - 677us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.3679 - 730us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.3134 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.4218 - 2ms/epoch - 40us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 13ms/epoch - 205us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.4795 - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.0280 - 829us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.7866 - 2ms/epoch - 26us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.0048 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.1055 - 2ms/epoch - 26us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.6197 - 851us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.9773 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.4247 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.3345 - 5ms/epoch - 73us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.0369 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2755 - 3ms/epoch - 42us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.7469 - 2ms/epoch - 36us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.3952 - 980us/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.4845 - 895us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9622 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.3926 - 793us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9873 - 1ms/epoch - 18us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.6566 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.1174 - 767us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.4685 - 897us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.3066 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9606 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.7109 - 1ms/epoch - 17us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.1309 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4264 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.8707 - 1ms/epoch - 20us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0521 - 723us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.7025 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9508 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3833 - 774us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.4559 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.0110 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.8735 - 944us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8178 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.5367 - 806us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.3547 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8111 - 621us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.4278 - 590us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.5140 - 980us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.6699 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.8632 - 672us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.3975 - 1ms/epoch - 19us/sample\n", + "Generation 141/300\n", + "Solving for Nash Equilibrium in Generation 141/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.2427 - 756us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 166us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.4906 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.4082 - 740us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37.9880 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1566 - 1ms/epoch - 19us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.1519 - 816us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.8704 - 747us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.3531 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.7624 - 2ms/epoch - 37us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232.1009 - 772us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8889 - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1775 - 701us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.0211 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.1847 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.5897 - 734us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.2105 - 698us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.0167 - 805us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.4465 - 4ms/epoch - 71us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59.4248 - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.9950 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.3610 - 5ms/epoch - 88us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.4004 - 829us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1245 - 1ms/epoch - 20us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.7417 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.5690 - 841us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.4319 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.4773 - 1ms/epoch - 24us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.5725 - 943us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7559 - 948us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.4360 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.6341 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.0658 - 792us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.0878 - 810us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7177 - 827us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9309 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9680 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.9832 - 599us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.6083 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.4207 - 815us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.0050 - 666us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.3997 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.7016 - 601us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3696 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4969 - 653us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.8716 - 743us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.3628 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.2178 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.3046 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.7548 - 871us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6498 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.8762 - 774us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.6077 - 795us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.4580 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.9313 - 2ms/epoch - 37us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.9885 - 45ms/epoch - 733us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7490 - 709us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.5807 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.9164 - 7ms/epoch - 117us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.0840 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.0466 - 649us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6451 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.2648 - 1ms/epoch - 18us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.0427 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0118 - 657us/epoch - 11us/sample\n", + "Generation 142/300\n", + "Solving for Nash Equilibrium in Generation 142/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.8543 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5923 - 701us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2047 - 3ms/epoch - 50us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.6134 - 680us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7200 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.3694 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.3468 - 641us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.2160 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.1115 - 848us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6683 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.2856 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.8137 - 865us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.7773 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.8301 - 854us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.9823 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.1399 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.8266 - 719us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.4288 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8177 - 814us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.8791 - 689us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.7114 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.5156 - 740us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.3045 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 132us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.3777 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.9608 - 1ms/epoch - 20us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0784 - 2ms/epoch - 36us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.4602 - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.3268 - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.6331 - 1ms/epoch - 22us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.4330 - 6ms/epoch - 103us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8435 - 1ms/epoch - 20us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.2189 - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7951 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2003 - 1ms/epoch - 17us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 14ms/epoch - 224us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.9742 - 3ms/epoch - 45us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.3815 - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.4972 - 1ms/epoch - 21us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3706 - 2ms/epoch - 31us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.9108 - 1ms/epoch - 21us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.2390 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8860 - 900us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.6356 - 1ms/epoch - 16us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.9916 - 1ms/epoch - 19us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.8372 - 768us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.2847 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.7126 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.7864 - 1ms/epoch - 19us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.3804 - 1ms/epoch - 18us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.0858 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7081 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.6815 - 44ms/epoch - 708us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.4130 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.3679 - 936us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.9853 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7731 - 896us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4241 - 727us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.4693 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4917 - 765us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.8226 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.7390 - 766us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.0263 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.7756 - 807us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.2950 - 838us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.6283 - 806us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.9142 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.1635 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.3884 - 882us/epoch - 14us/sample\n", + "Generation 143/300\n", + "Solving for Nash Equilibrium in Generation 143/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.2635 - 758us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.2510 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.4144 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.0010 - 719us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.6130 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.4594 - 750us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.5996 - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.5589 - 1ms/epoch - 21us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.7512 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.7543 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.6464 - 775us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6935 - 739us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.8809 - 837us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.0720 - 701us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.6700 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.9291 - 1ms/epoch - 18us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.9470 - 1ms/epoch - 17us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2855 - 951us/epoch - 15us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.9166 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.1226 - 988us/epoch - 16us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.4410 - 958us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.2473 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.3084 - 1ms/epoch - 23us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.8197 - 743us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.0516 - 696us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.5080 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.9487 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.0139 - 1ms/epoch - 22us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6769 - 657us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0055 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.1817 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.0948 - 15ms/epoch - 236us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.1807 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.5256 - 2ms/epoch - 37us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.0372 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.2571 - 900us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.3510 - 759us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9747 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6776 - 2ms/epoch - 32us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.2422 - 868us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.5835 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.7851 - 1ms/epoch - 19us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.4082 - 696us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.1031 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4642 - 1ms/epoch - 18us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.1671 - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.0159 - 771us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.2623 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.9008 - 1ms/epoch - 22us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.0098 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.8666 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.1242 - 648us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.5339 - 1ms/epoch - 17us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.8065 - 903us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.8867 - 761us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8736 - 721us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.0331 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 219.7492 - 644us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6277 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.1223 - 707us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.1244 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.2700 - 659us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9240 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9075 - 683us/epoch - 11us/sample\n", + "Generation 144/300\n", + "Solving for Nash Equilibrium in Generation 144/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.4684 - 3ms/epoch - 42us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.4553 - 988us/epoch - 16us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 188.0264 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.5891 - 1ms/epoch - 22us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.4811 - 4ms/epoch - 62us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.3604 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 362.7502 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.1779 - 749us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.6000 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.8223 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9926 - 1ms/epoch - 17us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.5453 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.3276 - 1ms/epoch - 20us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1176 - 786us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.3624 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4504 - 10ms/epoch - 154us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.4564 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.2006 - 989us/epoch - 16us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.3386 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6545 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6685 - 940us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.2168 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.6199 - 2ms/epoch - 36us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5756 - 820us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8773 - 2ms/epoch - 29us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.8100 - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.4023 - 939us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.4506 - 926us/epoch - 15us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.7226 - 1ms/epoch - 20us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.9669 - 1ms/epoch - 17us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.0616 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 157us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.3082 - 3ms/epoch - 48us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.3277 - 1ms/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.0204 - 890us/epoch - 14us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 151us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46.9262 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.1090 - 1ms/epoch - 24us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9095 - 1ms/epoch - 16us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.6463 - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.2032 - 2ms/epoch - 34us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.3857 - 800us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.6143 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.0798 - 678us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 137us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.3648 - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.9773 - 814us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.1902 - 959us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.9066 - 998us/epoch - 16us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.2880 - 868us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.8458 - 11ms/epoch - 182us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7429 - 872us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.5822 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.8487 - 3ms/epoch - 52us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.7159 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 93us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.6991 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0125 - 1ms/epoch - 19us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.4620 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.2509 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.5903 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.9532 - 811us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.6649 - 724us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.3117 - 691us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.7629 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.3370 - 685us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.2004 - 1ms/epoch - 20us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.9768 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.5127 - 822us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.6745 - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2422 - 1ms/epoch - 18us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.6108 - 739us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.0115 - 653us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.0305 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8499 - 1ms/epoch - 16us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.7408 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.5146 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.1038 - 683us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.7345 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4404 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3673 - 3ms/epoch - 51us/sample\n", + "Episode 50/50\n", + "Generation 145/300\n", + "Solving for Nash Equilibrium in Generation 145/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5576 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.7472 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.5173 - 709us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.8644 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.8342 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.4667 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.1782 - 781us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.3932 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5708 - 863us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.3657 - 904us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3917 - 944us/epoch - 15us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.5069 - 707us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.9942 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.7420 - 824us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.0059 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6509 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.3676 - 797us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.4141 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.0409 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.7820 - 845us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.3687 - 784us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.2100 - 766us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.0400 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.8204 - 8ms/epoch - 128us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3427 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.8236 - 3ms/epoch - 50us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.0121 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7517 - 848us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.6115 - 996us/epoch - 16us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9334 - 2ms/epoch - 24us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6739 - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.8128 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.3657 - 3ms/epoch - 43us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2442 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.0202 - 3ms/epoch - 42us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1247 - 2ms/epoch - 25us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.9451 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5942 - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5959 - 1ms/epoch - 18us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.4929 - 3ms/epoch - 40us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.7708 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.1838 - 2ms/epoch - 34us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.6516 - 6ms/epoch - 100us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.1951 - 2ms/epoch - 36us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.8813 - 1ms/epoch - 19us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5421 - 851us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.9406 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.1664 - 2ms/epoch - 34us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3250 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.7659 - 1ms/epoch - 18us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.2148 - 16ms/epoch - 253us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.7161 - 1ms/epoch - 16us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5723 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.1542 - 959us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.5876 - 1ms/epoch - 24us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.8886 - 900us/epoch - 15us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.1219 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4525 - 677us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.0027 - 1ms/epoch - 18us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4761 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 362.6784 - 907us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 790.1742 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.6962 - 766us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.0990 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.8868 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.0327 - 629us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1871 - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6783 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.1466 - 1ms/epoch - 20us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.9391 - 1ms/epoch - 19us/sample\n", + "Generation 146/300\n", + "Solving for Nash Equilibrium in Generation 146/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.6608 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.8762 - 896us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.7101 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5909 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6161 - 918us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.5886 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2845 - 682us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.5206 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.1987 - 800us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.9907 - 761us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4955 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.5844 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.5338 - 6ms/epoch - 103us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0942 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.0984 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.3243 - 758us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.1830 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8192 - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3505 - 3ms/epoch - 41us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4321 - 668us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.6020 - 1ms/epoch - 22us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.1774 - 869us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.7638 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.7498 - 783us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.7228 - 791us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.9210 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.3253 - 4ms/epoch - 67us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.4402 - 740us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0551 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.8766 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.2888 - 809us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9940 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1957 - 615us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.0147 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.3459 - 746us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.9211 - 727us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.8252 - 777us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.7257 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4735 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9628 - 750us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.0634 - 752us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.8694 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.2878 - 688us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.1838 - 723us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.6831 - 643us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 128us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.0745 - 1ms/epoch - 23us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.2903 - 742us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4814 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.6694 - 697us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.1376 - 755us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.4684 - 650us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.4545 - 852us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.1719 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.6404 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.1319 - 805us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.5025 - 791us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8461 - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2343 - 741us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.5549 - 2ms/epoch - 24us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9281 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.4545 - 7ms/epoch - 118us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.6430 - 728us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5873 - 1ms/epoch - 23us/sample\n", + "Generation 147/300\n", + "Solving for Nash Equilibrium in Generation 147/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.5953 - 729us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3906 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.0357 - 1ms/epoch - 18us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4703 - 729us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.7191 - 768us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 145.8951 - 651us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.1450 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 757.9122 - 601us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.2058 - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.2307 - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4406 - 724us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8785 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.2549 - 727us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.8105 - 747us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.5068 - 710us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.9173 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.3655 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5018 - 779us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.5547 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8968 - 686us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.1608 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.4525 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.9203 - 763us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.1391 - 716us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0114 - 648us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.6384 - 794us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.7018 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.7546 - 4ms/epoch - 58us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.3304 - 803us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8551 - 813us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.3983 - 555us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.3900 - 759us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9396 - 772us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1351 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.4313 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5038 - 643us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.0016 - 1ms/epoch - 22us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.8148 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.0274 - 811us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2117 - 921us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.1732 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.6981 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.3705 - 1ms/epoch - 17us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.4577 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.8524 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.4126 - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8522 - 1ms/epoch - 16us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.6776 - 666us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.7656 - 723us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3746 - 832us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6144 - 657us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.2387 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.2039 - 717us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.0242 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.1515 - 8ms/epoch - 122us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9656 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.8832 - 693us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.4461 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.6324 - 644us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.8387 - 763us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8416 - 703us/epoch - 11us/sample\n", + "Generation 148/300\n", + "Solving for Nash Equilibrium in Generation 148/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.2719 - 1ms/epoch - 19us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.5056 - 1ms/epoch - 20us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9978 - 640us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5588 - 581us/epoch - 9us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.5921 - 812us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98.3324 - 606us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.5677 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.2157 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5858 - 755us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.7450 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.8388 - 5ms/epoch - 83us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.7896 - 625us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.9181 - 622us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.4498 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.3088 - 695us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.5252 - 759us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.9517 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7189 - 621us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 198.9570 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.1995 - 796us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.2213 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5748 - 626us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5742 - 637us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.3388 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.2194 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9490 - 679us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1808 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.7578 - 733us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.7894 - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.4047 - 567us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3772 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.1288 - 825us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8361 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.2561 - 632us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.3353 - 667us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.9185 - 2ms/epoch - 38us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.4774 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.7131 - 687us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0528 - 594us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.4067 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.7342 - 712us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 432.9210 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.3199 - 961us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.3543 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.4958 - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.8301 - 785us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.9940 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.5742 - 773us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.0320 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.2238 - 691us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.9925 - 2ms/epoch - 25us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.1788 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9443 - 710us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0171 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.1053 - 2ms/epoch - 25us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.3140 - 993us/epoch - 16us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.7658 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.4122 - 778us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.1083 - 738us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.7318 - 1ms/epoch - 23us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.9682 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5426 - 645us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.1892 - 1ms/epoch - 17us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.8000 - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.8666 - 620us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.4587 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.6382 - 742us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.1639 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.1621 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.6385 - 617us/epoch - 10us/sample\n", + "Generation 149/300\n", + "Solving for Nash Equilibrium in Generation 149/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6176 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.4128 - 849us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.8178 - 638us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.4238 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.3310 - 690us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.1285 - 755us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.4038 - 2ms/epoch - 35us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.2466 - 751us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3129 - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.7547 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.8245 - 1ms/epoch - 21us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9519 - 710us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.4081 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.7345 - 643us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.3257 - 700us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.1757 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.0847 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.8604 - 1ms/epoch - 17us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.4737 - 726us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.7443 - 760us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.4509 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.1142 - 764us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.2249 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2161 - 596us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.1456 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.8265 - 843us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.7147 - 856us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.8063 - 780us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8500 - 622us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.5719 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.3068 - 793us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.0410 - 11ms/epoch - 174us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.2854 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.0882 - 633us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.4362 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.0578 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.7403 - 772us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.3597 - 792us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.1273 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.9766 - 685us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.2581 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.1262 - 946us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.3606 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.0513 - 696us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.6497 - 712us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9669 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.0607 - 708us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.2701 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.8583 - 658us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6102 - 2ms/epoch - 33us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.9996 - 757us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2766 - 748us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.2619 - 2ms/epoch - 25us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.6660 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.4466 - 744us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.8414 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.2546 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.3785 - 747us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7822 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.3375 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.6981 - 753us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.9686 - 756us/epoch - 12us/sample\n", + "Generation 150/300\n", + "Solving for Nash Equilibrium in Generation 150/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.4954 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1638 - 658us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.8264 - 758us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.7472 - 802us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.2597 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.7169 - 873us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.5883 - 626us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.5194 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.7363 - 1ms/epoch - 19us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.4684 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5104 - 590us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.4457 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.3318 - 730us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.0297 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.1426 - 3ms/epoch - 46us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.2147 - 660us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.2796 - 648us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.1308 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.2921 - 756us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5547 - 616us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7822 - 2ms/epoch - 31us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.4260 - 688us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.0652 - 3ms/epoch - 48us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0986 - 831us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.4711 - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.2032 - 798us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.2689 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.5304 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.1892 - 860us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.4878 - 863us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.2024 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7072 - 761us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.6246 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.7151 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.3664 - 836us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0017 - 708us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.5917 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.7393 - 1ms/epoch - 19us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.6747 - 771us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5829 - 742us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.3893 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.2271 - 1ms/epoch - 18us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8085 - 716us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.6481 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.4601 - 681us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.0577 - 810us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.6450 - 668us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9787 - 5ms/epoch - 78us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.0573 - 587us/epoch - 9us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54.0342 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.2797 - 764us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 115us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.6386 - 5ms/epoch - 74us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3766 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.2454 - 730us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0687 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.2158 - 885us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3599 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.0015 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.4265 - 817us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.3072 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.6517 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.1353 - 726us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1620 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.4600 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.7435 - 746us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4654 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6031 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.6487 - 833us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.0661 - 781us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787.2401 - 876us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.0980 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.4136 - 743us/epoch - 12us/sample\n", + "Generation 151/300\n", + "Solving for Nash Equilibrium in Generation 151/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3680 - 615us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.6124 - 781us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.0346 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.7800 - 761us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.0867 - 886us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.2856 - 5ms/epoch - 85us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.2867 - 708us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.6943 - 976us/epoch - 16us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.3166 - 774us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.5040 - 2ms/epoch - 36us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6844 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.1954 - 842us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2880 - 723us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.6410 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.5528 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0761 - 723us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.6213 - 715us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1534 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.1497 - 1ms/epoch - 21us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 770.3696 - 729us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5245 - 675us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.8977 - 1ms/epoch - 18us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.6832 - 745us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.3879 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.1807 - 683us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.4240 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4603 - 893us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.2155 - 651us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.8395 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.3014 - 770us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4750 - 1ms/epoch - 17us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.3971 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1349 - 625us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.5203 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.7534 - 1ms/epoch - 22us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5002 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.5008 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.0686 - 6ms/epoch - 100us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.9070 - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.1658 - 705us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.6493 - 2ms/epoch - 30us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.9294 - 681us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1249 - 633us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.2053 - 684us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.4773 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.2346 - 659us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.4705 - 2ms/epoch - 33us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.1630 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.2629 - 629us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.4983 - 604us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.8865 - 1ms/epoch - 17us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5179 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.5686 - 651us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.3919 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2698 - 719us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8033 - 605us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.4056 - 670us/epoch - 11us/sample\n", + "Generation 152/300\n", + "Solving for Nash Equilibrium in Generation 152/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.8931 - 738us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.7615 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.2084 - 730us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6635 - 717us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7584 - 694us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.3051 - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6495 - 590us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.1809 - 699us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40.3186 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.3522 - 851us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.9933 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.0668 - 660us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.2263 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.9568 - 790us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.0676 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2025 - 964us/epoch - 16us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.2867 - 973us/epoch - 16us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.7449 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.9088 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.2809 - 719us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1052 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2626 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.7878 - 681us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2932 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3366 - 663us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.2162 - 623us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 104us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8146 - 1ms/epoch - 22us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3419 - 907us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.8421 - 845us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6268 - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9041 - 623us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.0905 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.3214 - 1ms/epoch - 18us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.9988 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.2497 - 795us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.0417 - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.1027 - 769us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9446 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7307 - 744us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.4785 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.2148 - 634us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.9236 - 6ms/epoch - 91us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.6432 - 914us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.5350 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.4381 - 895us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 159.6295 - 818us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.1436 - 855us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.0268 - 803us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.7961 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.5348 - 712us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.2342 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6144 - 3ms/epoch - 48us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2354 - 732us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.7648 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.9748 - 942us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.8885 - 884us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.1439 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.5566 - 678us/epoch - 11us/sample\n", + "Generation 153/300\n", + "Solving for Nash Equilibrium in Generation 153/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0454 - 805us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.4059 - 913us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.5360 - 788us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.1341 - 684us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.4997 - 792us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1238 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6301 - 778us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.2586 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4327 - 759us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.8246 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.3759 - 887us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0786 - 864us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6887 - 977us/epoch - 16us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.1715 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.5664 - 1ms/epoch - 19us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.7753 - 989us/epoch - 16us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 233.6729 - 843us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8045 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.5599 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.9176 - 743us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.5612 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.7980 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.9986 - 701us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.0044 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.4767 - 594us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.8387 - 774us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.7071 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.4543 - 708us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.9377 - 966us/epoch - 16us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.1329 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.9938 - 705us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.0510 - 786us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.5728 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.5678 - 736us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.8670 - 825us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.4178 - 764us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.3815 - 675us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.3972 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.9662 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.2717 - 574us/epoch - 9us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.1341 - 1ms/epoch - 16us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.1007 - 2ms/epoch - 29us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.3356 - 1ms/epoch - 19us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.9045 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.6403 - 764us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.5861 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4249 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.7036 - 1ms/epoch - 22us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2881 - 640us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.0484 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.8777 - 856us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.5983 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.4512 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.5610 - 1ms/epoch - 18us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.5163 - 779us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.7762 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.8592 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.7321 - 648us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.8679 - 785us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.0007 - 728us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.2269 - 688us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.1871 - 679us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.7913 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.2329 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4945 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.5445 - 746us/epoch - 12us/sample\n", + "Generation 154/300\n", + "Solving for Nash Equilibrium in Generation 154/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.3966 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.0601 - 642us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.0211 - 1ms/epoch - 19us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4413 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.3915 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2964 - 641us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.3145 - 740us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.8528 - 728us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.7472 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361.4197 - 649us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.0110 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8126 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.0092 - 588us/epoch - 9us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.9153 - 803us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.3794 - 838us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.5131 - 918us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406.4429 - 985us/epoch - 16us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.8569 - 801us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5938 - 770us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.0436 - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.6880 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.3741 - 760us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.4618 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2870 - 615us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.3689 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.1292 - 912us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.6513 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.6461 - 646us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.9610 - 917us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.2252 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9673 - 556us/epoch - 9us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.1043 - 1ms/epoch - 22us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.8084 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.5676 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1503 - 770us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.7024 - 591us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.9508 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.6492 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2131 - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.0234 - 807us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.7879 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.5571 - 741us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.4274 - 676us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.5895 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.0881 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8174 - 695us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.7541 - 667us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.8441 - 670us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.1362 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.0999 - 1ms/epoch - 16us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.3759 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.7589 - 694us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43.2621 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.9405 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5782 - 832us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.7841 - 690us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.0451 - 669us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.2950 - 661us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.4664 - 669us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.5468 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.1955 - 715us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 125.5406 - 791us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.6570 - 644us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4122 - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.4329 - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5411 - 745us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6389 - 599us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.4008 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.0961 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.1065 - 667us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4858 - 658us/epoch - 11us/sample\n", + "Generation 155/300\n", + "Solving for Nash Equilibrium in Generation 155/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.9146 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.4127 - 733us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.4125 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.3173 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0581 - 810us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.2773 - 969us/epoch - 16us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.1759 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.5747 - 691us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.5920 - 3ms/epoch - 47us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.9407 - 810us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.6650 - 1ms/epoch - 17us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40.3413 - 774us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3986 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.7953 - 1ms/epoch - 16us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.5438 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.7704 - 652us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9181 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4171 - 656us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7748 - 789us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.5436 - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.8195 - 697us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.7664 - 860us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4247 - 660us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.7839 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.8534 - 804us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.3817 - 741us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7344 - 990us/epoch - 16us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.8222 - 934us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.0988 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.6030 - 877us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.7297 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.5954 - 857us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.7382 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.8562 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1531 - 664us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787.2596 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.8551 - 748us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.2503 - 600us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.7242 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.3212 - 863us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.3003 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.2301 - 956us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2914 - 708us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.3668 - 938us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.0483 - 921us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.8002 - 2ms/epoch - 25us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2628 - 937us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.9036 - 794us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.6046 - 716us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.6749 - 842us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.6656 - 599us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2554 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112.3322 - 838us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.9015 - 680us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1486 - 2ms/epoch - 37us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.3163 - 714us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5770 - 676us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 156/300\n", + "Solving for Nash Equilibrium in Generation 156/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.0909 - 622us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9395 - 790us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.2136 - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 228.7115 - 721us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45.9925 - 638us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3793 - 809us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.1908 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232.8526 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8510 - 1ms/epoch - 21us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.9467 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.1950 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2911 - 696us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.2692 - 704us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.1880 - 689us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.5951 - 964us/epoch - 16us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.3101 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.8691 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7356 - 766us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.3910 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.9588 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1470 - 727us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 162.7045 - 1ms/epoch - 24us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.9755 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4777 - 678us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.3483 - 2ms/epoch - 40us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.6394 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.1757 - 5ms/epoch - 73us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.2585 - 916us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4411 - 720us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5049 - 733us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.6098 - 677us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.5818 - 672us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.7139 - 568us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.9597 - 646us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7785 - 803us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.7815 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.1760 - 672us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.0594 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.6263 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.5192 - 711us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9138 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2321 - 671us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 145us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.9452 - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2323 - 709us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9210 - 779us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.3576 - 8ms/epoch - 130us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.5067 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.1620 - 672us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.9724 - 1ms/epoch - 19us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2610 - 609us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.9910 - 3ms/epoch - 47us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.4441 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.8520 - 2ms/epoch - 29us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.2785 - 609us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.4110 - 836us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.5558 - 656us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.5674 - 13ms/epoch - 203us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.5487 - 2ms/epoch - 34us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3192 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.4686 - 660us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 157/300\n", + "Solving for Nash Equilibrium in Generation 157/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.9247 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.9869 - 2ms/epoch - 27us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47.8381 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.5892 - 651us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.8510 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.1556 - 1ms/epoch - 17us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.8672 - 645us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2128 - 691us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.5219 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.4634 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.8491 - 695us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.1640 - 801us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.9781 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8755 - 754us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.1842 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.4252 - 805us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.8507 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.9966 - 796us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.9036 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.3412 - 2ms/epoch - 27us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.9542 - 777us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.3293 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.1216 - 801us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.9067 - 748us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.3565 - 710us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3056 - 2ms/epoch - 35us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.0190 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.0199 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.7504 - 2ms/epoch - 39us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.1896 - 757us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.3310 - 884us/epoch - 14us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.2119 - 771us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.8718 - 5ms/epoch - 74us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.2072 - 746us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.5848 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7660 - 700us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.2587 - 795us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0424 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.7131 - 598us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.1979 - 626us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.2319 - 869us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1207 - 5ms/epoch - 76us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.9504 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5626 - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9365 - 694us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.5867 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3106 - 598us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.6470 - 606us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.9769 - 709us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.2136 - 594us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9326 - 765us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.6428 - 588us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.3522 - 576us/epoch - 9us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.1339 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5753 - 902us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.4605 - 829us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.7875 - 697us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.8089 - 1ms/epoch - 19us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.8527 - 742us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.6617 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0980 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.4800 - 738us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.1107 - 569us/epoch - 9us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.2545 - 3ms/epoch - 43us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9035 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.9134 - 746us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.6756 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.6994 - 894us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.8494 - 698us/epoch - 11us/sample\n", + "Generation 158/300\n", + "Solving for Nash Equilibrium in Generation 158/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.1603 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.1174 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8503 - 808us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.5385 - 604us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.9287 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.7219 - 9ms/epoch - 146us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.6071 - 609us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.9636 - 836us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37.5029 - 753us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.3983 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3477 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2020 - 617us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.8012 - 2ms/epoch - 37us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.0078 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.3008 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8948 - 683us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.3734 - 839us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6174 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.5619 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.0389 - 736us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.0885 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.8775 - 736us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.0018 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4783 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7620 - 684us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3464 - 845us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8294 - 642us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.9333 - 693us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9028 - 794us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.4955 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.5611 - 614us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.9291 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.3531 - 695us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0351 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8201 - 690us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.5533 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.6037 - 782us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.6944 - 683us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 772.3726 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.3404 - 706us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.3078 - 643us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.5110 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.7308 - 2ms/epoch - 38us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.3401 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.0662 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.9634 - 677us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.4475 - 660us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.4151 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1244 - 7ms/epoch - 112us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.4161 - 577us/epoch - 9us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.6862 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.2615 - 769us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.7479 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.4367 - 720us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 785.0875 - 691us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.8801 - 836us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.8010 - 873us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.6366 - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.7113 - 871us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1599 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.7281 - 2ms/epoch - 30us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30.9359 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3544 - 1ms/epoch - 21us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.0849 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 139us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8256 - 865us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.0561 - 1ms/epoch - 22us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 159/300\n", + "Solving for Nash Equilibrium in Generation 159/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787.9453 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.7502 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.8594 - 630us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.4952 - 706us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.6937 - 941us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.3813 - 906us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.6189 - 971us/epoch - 16us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.8773 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.1545 - 2ms/epoch - 37us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.5678 - 754us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.1616 - 700us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.5792 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.1536 - 795us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.3632 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.1062 - 909us/epoch - 15us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.1766 - 632us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.0509 - 632us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749.0114 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.7658 - 642us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.0673 - 705us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.3545 - 715us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4747 - 604us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.3871 - 742us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.3056 - 654us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.9081 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.3040 - 756us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.4582 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.1302 - 655us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.4575 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.3671 - 971us/epoch - 16us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.2160 - 683us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.5661 - 641us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7700 - 5ms/epoch - 84us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.0833 - 675us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.4783 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.9932 - 1ms/epoch - 17us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.1048 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 130us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.0758 - 3ms/epoch - 51us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.0374 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.3234 - 988us/epoch - 16us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.7684 - 7ms/epoch - 115us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.7818 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.4536 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.6069 - 2ms/epoch - 29us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.1373 - 835us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.8477 - 679us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5269 - 840us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.5535 - 638us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.2398 - 1ms/epoch - 18us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.4532 - 5ms/epoch - 79us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.7507 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.7224 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.5804 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.0127 - 648us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.6071 - 638us/epoch - 10us/sample\n", + "Generation 160/300\n", + "Solving for Nash Equilibrium in Generation 160/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.7263 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.5324 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.3362 - 4ms/epoch - 67us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8389 - 684us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.6490 - 738us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.8005 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.4948 - 621us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.5291 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.0321 - 536us/epoch - 9us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.0021 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6397 - 632us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.9123 - 2ms/epoch - 38us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3622 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.4623 - 667us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.5567 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3109 - 2ms/epoch - 37us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4864 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.9138 - 642us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.2880 - 2ms/epoch - 33us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.3542 - 807us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.2380 - 679us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.8947 - 4ms/epoch - 65us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6037 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9777 - 735us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8406 - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.8270 - 632us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.1373 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.3560 - 917us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43.0280 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.2302 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.6502 - 789us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4538 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.1660 - 621us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.1084 - 701us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5090 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.4954 - 676us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.4797 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9575 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.0344 - 582us/epoch - 9us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.1003 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3960 - 623us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.2043 - 3ms/epoch - 44us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65.8795 - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.9186 - 679us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3146 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5081 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.1646 - 584us/epoch - 9us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7800 - 4ms/epoch - 64us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.3822 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.7063 - 676us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.9744 - 567us/epoch - 9us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8009 - 605us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.1378 - 683us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.9714 - 701us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.5960 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8845 - 663us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.1038 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 12ms/epoch - 188us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.3117 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.8664 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.9884 - 2ms/epoch - 26us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.9928 - 743us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.0906 - 694us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.8516 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.6974 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1550 - 658us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784.9841 - 585us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.8575 - 826us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.9909 - 705us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.9728 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.8674 - 2ms/epoch - 38us/sample\n", + "Generation 161/300\n", + "Solving for Nash Equilibrium in Generation 161/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.5947 - 5ms/epoch - 80us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.3342 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.2039 - 741us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9705 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.8189 - 640us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.1234 - 629us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9875 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.8950 - 706us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.6541 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.3939 - 672us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.0596 - 876us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.4326 - 594us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.5615 - 794us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.6909 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3427 - 690us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.9603 - 742us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6873 - 593us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5794 - 745us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9315 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.5539 - 1ms/epoch - 22us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.7149 - 658us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248.6650 - 786us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.1041 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.3483 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.2847 - 3ms/epoch - 42us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.5798 - 3ms/epoch - 56us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5847 - 805us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.0370 - 631us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7738 - 942us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 220.3857 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.3681 - 25ms/epoch - 403us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.3555 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.9926 - 4ms/epoch - 69us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.2173 - 718us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7198 - 613us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.8752 - 637us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.8976 - 618us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.0818 - 612us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.0726 - 10ms/epoch - 169us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.1702 - 652us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7661 - 896us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9777 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.3365 - 684us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.2791 - 2ms/epoch - 25us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.9369 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9484 - 797us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.9774 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6282 - 617us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.0225 - 819us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.6317 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6818 - 2ms/epoch - 26us/sample\n", + "Episode 50/50\n", + "Generation 162/300\n", + "Solving for Nash Equilibrium in Generation 162/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.7555 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 435.3481 - 688us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3829 - 790us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3514 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.5815 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.8240 - 721us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6701 - 738us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.6993 - 689us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.1746 - 1ms/epoch - 18us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.3191 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.6053 - 713us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.5861 - 788us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.2204 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.8822 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.8268 - 728us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.0168 - 684us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.0278 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.7887 - 791us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.5972 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.6016 - 5ms/epoch - 78us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.4544 - 770us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5943 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.8655 - 715us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.5894 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.7461 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.5240 - 614us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.6750 - 707us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.9357 - 835us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.5931 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.7475 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2155 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.3714 - 759us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.3757 - 7ms/epoch - 112us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.8839 - 721us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.6852 - 1ms/epoch - 19us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.2433 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.9268 - 1ms/epoch - 18us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.4926 - 688us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 133us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.7987 - 1ms/epoch - 17us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.4161 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.5327 - 4ms/epoch - 59us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.3711 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.7198 - 956us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.6571 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.7637 - 854us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5182 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.3344 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.3773 - 620us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4124 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4232 - 558us/epoch - 9us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.2469 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.1297 - 773us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.0276 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.0878 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7364 - 730us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6391 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.9508 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.6982 - 819us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8351 - 694us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8657 - 714us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.9781 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3920 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.8071 - 750us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.5870 - 756us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.2815 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4073 - 767us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.6893 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.0230 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.3195 - 1ms/epoch - 21us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.3651 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.0082 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.3864 - 650us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.2826 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.2224 - 841us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.4948 - 3ms/epoch - 49us/sample\n", + "Generation 163/300\n", + "Solving for Nash Equilibrium in Generation 163/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.8997 - 907us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.6838 - 595us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.3619 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.4225 - 649us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.0614 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.5601 - 956us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7246 - 685us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.6946 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6262 - 627us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.6619 - 690us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.2537 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.4238 - 754us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.8942 - 639us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.6780 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.0582 - 673us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0963 - 724us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 130.7081 - 928us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.6713 - 807us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.6305 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.0016 - 667us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.5352 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.3533 - 897us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.8284 - 794us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.9664 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8383 - 742us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.4244 - 883us/epoch - 14us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.8159 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.8981 - 785us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.2531 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.1769 - 631us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.3610 - 737us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.8073 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.8532 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.5278 - 1ms/epoch - 16us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.1805 - 5ms/epoch - 77us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.6223 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1265 - 2ms/epoch - 40us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0031 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7115 - 713us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6899 - 771us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.7697 - 2ms/epoch - 30us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3922 - 2ms/epoch - 39us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.9301 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.7352 - 757us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.8832 - 797us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.9171 - 589us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6256 - 719us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48.4967 - 839us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.8204 - 1ms/epoch - 18us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.1733 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.7823 - 711us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2941 - 716us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.8793 - 767us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3019 - 694us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.0369 - 6ms/epoch - 101us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.2529 - 35ms/epoch - 563us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.1821 - 786us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.1243 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.5235 - 2ms/epoch - 33us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.4324 - 1ms/epoch - 16us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7763 - 824us/epoch - 13us/sample\n", + "Generation 164/300\n", + "Solving for Nash Equilibrium in Generation 164/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36.4285 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.0956 - 994us/epoch - 16us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.5877 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.3677 - 743us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.3677 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.6361 - 710us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.2153 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.4849 - 903us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6747 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0444 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.0619 - 673us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1605 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.9755 - 829us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9606 - 667us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.6465 - 902us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.4091 - 860us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6190 - 629us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5577 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49.0184 - 769us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.1208 - 725us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.8879 - 573us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9203 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.6599 - 8ms/epoch - 134us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.0901 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.6346 - 651us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9031 - 900us/epoch - 15us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.4642 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2072 - 839us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.5931 - 719us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.1008 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 148.4852 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.7935 - 753us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3533 - 1ms/epoch - 18us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8552 - 785us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.3702 - 7ms/epoch - 109us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.4913 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5194 - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9315 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.0896 - 2ms/epoch - 30us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.4596 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.6666 - 873us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.4394 - 777us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.8348 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.3685 - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.8414 - 849us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5114 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9077 - 2ms/epoch - 38us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3606 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.6302 - 792us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.5623 - 851us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.1393 - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.3921 - 834us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.6474 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.8621 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.9090 - 872us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.4915 - 844us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.9012 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.0555 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0665 - 2ms/epoch - 38us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.8987 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.6288 - 1ms/epoch - 17us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.2206 - 2ms/epoch - 29us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.6219 - 824us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0272 - 3ms/epoch - 42us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.5493 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.5325 - 727us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.7380 - 1ms/epoch - 17us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.7552 - 1ms/epoch - 19us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.8116 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.8294 - 866us/epoch - 14us/sample\n", + "Generation 165/300\n", + "Solving for Nash Equilibrium in Generation 165/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6505 - 822us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.2346 - 859us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2770 - 768us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.2600 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.3405 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0089 - 776us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.4822 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.2017 - 750us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.1710 - 690us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.3056 - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9753 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 17ms/epoch - 272us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6236 - 7ms/epoch - 111us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.8872 - 700us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.4994 - 1ms/epoch - 18us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.9639 - 2ms/epoch - 28us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.8549 - 841us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.4451 - 865us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.8946 - 672us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.4611 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2650 - 703us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.5420 - 749us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0035 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8877 - 2ms/epoch - 30us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.8351 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2094 - 841us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.6316 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9521 - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.6305 - 20ms/epoch - 324us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.1299 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.6616 - 695us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4144 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8660 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.0609 - 3ms/epoch - 48us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.8331 - 1ms/epoch - 18us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.0352 - 822us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.0219 - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.8022 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.1925 - 1ms/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.8323 - 859us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8497 - 776us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.9391 - 701us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.7214 - 1ms/epoch - 21us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3449 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.2263 - 3ms/epoch - 56us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.3097 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6600 - 726us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2016 - 800us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.2331 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6221 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1503 - 661us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.9664 - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0651 - 1ms/epoch - 19us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47.3913 - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.0397 - 4ms/epoch - 57us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.4613 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.3037 - 586us/epoch - 9us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.9454 - 885us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.7788 - 684us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.2874 - 663us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.5072 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.4757 - 785us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.6863 - 672us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755.9975 - 3ms/epoch - 56us/sample\n", + "Episode 50/50\n", + "Generation 166/300\n", + "Solving for Nash Equilibrium in Generation 166/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 233.2900 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.2348 - 549us/epoch - 9us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4190 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.1250 - 699us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6702 - 1ms/epoch - 24us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.2246 - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4431 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.4048 - 726us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.6623 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4256 - 640us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.3226 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.6152 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.6090 - 684us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.0837 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.0971 - 717us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2039 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.9293 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0917 - 1ms/epoch - 21us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.1152 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.7307 - 832us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3460 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.5759 - 20ms/epoch - 318us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.7596 - 665us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.3811 - 1ms/epoch - 17us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.1487 - 667us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.4180 - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.5580 - 4ms/epoch - 58us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.8875 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3903 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.1652 - 2ms/epoch - 26us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.7101 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1088 - 775us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.1346 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.1109 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.2484 - 2ms/epoch - 25us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.2805 - 4ms/epoch - 70us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.7875 - 699us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.4019 - 1ms/epoch - 20us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.2255 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.2592 - 851us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.5455 - 649us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.5902 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.3056 - 902us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.9433 - 888us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8049 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.3553 - 686us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.6244 - 855us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0072 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.3735 - 13ms/epoch - 206us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.2738 - 1ms/epoch - 24us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.0123 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.1770 - 2ms/epoch - 40us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 73.2788 - 3ms/epoch - 41us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.0856 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.7152 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 165us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.1910 - 1ms/epoch - 18us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.8156 - 2ms/epoch - 34us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8458 - 739us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406.0654 - 1ms/epoch - 17us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.7159 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7889 - 4ms/epoch - 65us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.4019 - 793us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8372 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.9169 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 125us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.3900 - 2ms/epoch - 37us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.6838 - 876us/epoch - 14us/sample\n", + "Generation 167/300\n", + "Solving for Nash Equilibrium in Generation 167/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.5646 - 841us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.5598 - 6ms/epoch - 92us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.4137 - 782us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.4740 - 689us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.1337 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 362.6317 - 635us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3084 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.7631 - 804us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 125us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.7793 - 929us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.3014 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5945 - 2ms/epoch - 25us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.4426 - 918us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 181.6683 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.9283 - 1ms/epoch - 16us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.7838 - 739us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.7072 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.2500 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.3353 - 659us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.3224 - 627us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.6746 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9189 - 829us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.5493 - 670us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6965 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.2034 - 643us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.6323 - 639us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0298 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.7758 - 654us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.2671 - 762us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9477 - 611us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.3553 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.7411 - 759us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.7099 - 1ms/epoch - 19us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.0724 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.4560 - 2ms/epoch - 24us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1351 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.5165 - 678us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.1568 - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.3441 - 837us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.5876 - 616us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.6959 - 1ms/epoch - 19us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787.2444 - 584us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.2872 - 733us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.8008 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.5660 - 604us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.9689 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8707 - 641us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.7073 - 698us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.7810 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.9179 - 798us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.6527 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.7834 - 1ms/epoch - 19us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.3806 - 975us/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.2970 - 760us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.3124 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6109 - 625us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.3564 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.1608 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.6605 - 881us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.3438 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.7218 - 595us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.0013 - 13ms/epoch - 208us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.8684 - 821us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.5255 - 746us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.6688 - 938us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8976 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1622 - 8ms/epoch - 123us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.4894 - 974us/epoch - 16us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6412 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.5098 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.7421 - 1ms/epoch - 20us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.5463 - 1ms/epoch - 20us/sample\n", + "Generation 168/300\n", + "Solving for Nash Equilibrium in Generation 168/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.1918 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.0211 - 902us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.0181 - 1ms/epoch - 19us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.9522 - 765us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.8220 - 810us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.6909 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.6823 - 715us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.7793 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.0154 - 640us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.0856 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.6515 - 802us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.0780 - 813us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 782.9211 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.7077 - 669us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.6896 - 807us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.3502 - 715us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.6883 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6561 - 732us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.1677 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.6389 - 1ms/epoch - 21us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.4035 - 754us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.6203 - 621us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.7598 - 702us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1389 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.8383 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.6990 - 689us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.1633 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7850 - 637us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.6961 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.7104 - 623us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2620 - 611us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.7277 - 831us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6180 - 833us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3907 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.7439 - 619us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.5559 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 225.3622 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.7338 - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.8555 - 669us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2457 - 668us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.4229 - 783us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.4876 - 774us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.7892 - 751us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.7654 - 708us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.6682 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.6432 - 892us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.9272 - 2ms/epoch - 26us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1281 - 699us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.2024 - 2ms/epoch - 32us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.0822 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.3895 - 1ms/epoch - 23us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.6799 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.8502 - 917us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.8599 - 623us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250.6322 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.8801 - 782us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.9943 - 919us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.5786 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.9308 - 577us/epoch - 9us/sample\n", + "Generation 169/300\n", + "Solving for Nash Equilibrium in Generation 169/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 14ms/epoch - 227us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.0345 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.6922 - 897us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.1196 - 2ms/epoch - 27us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7361 - 814us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.5696 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.4773 - 642us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.1626 - 3ms/epoch - 41us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.2613 - 621us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0681 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.8759 - 718us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 792.0992 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.3552 - 823us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.5089 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.4370 - 900us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.4197 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.6336 - 802us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.9283 - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.4846 - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.3509 - 653us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6938 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2580 - 644us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.7021 - 922us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.2196 - 791us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.7972 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.1484 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.5488 - 1ms/epoch - 18us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.6478 - 829us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9297 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1793 - 622us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.8832 - 608us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.5172 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.0121 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.1234 - 784us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.5645 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.4319 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.2884 - 669us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8081 - 702us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.9144 - 720us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.9040 - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.5201 - 1ms/epoch - 19us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.5889 - 3ms/epoch - 43us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.5784 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.9679 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.7667 - 649us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.4209 - 851us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.6587 - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6298 - 4ms/epoch - 65us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6297 - 835us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.9921 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.5126 - 1ms/epoch - 18us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.0963 - 771us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.0396 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.1741 - 681us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.2022 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.6708 - 1ms/epoch - 17us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.3112 - 619us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.5688 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.7629 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6285 - 2ms/epoch - 25us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.9294 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.5035 - 736us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.1745 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.4524 - 844us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.8131 - 1ms/epoch - 20us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.7819 - 606us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.6847 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.9054 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.8882 - 2ms/epoch - 28us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.9210 - 681us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1059 - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.9684 - 743us/epoch - 12us/sample\n", + "Generation 170/300\n", + "Solving for Nash Equilibrium in Generation 170/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.1469 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3491 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.4454 - 2ms/epoch - 25us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.2145 - 1ms/epoch - 19us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.0455 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.6705 - 865us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.5807 - 885us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.6174 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.4870 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.5527 - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2461 - 697us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.2982 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.8369 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1282 - 9ms/epoch - 138us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.7677 - 1ms/epoch - 17us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.6946 - 631us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5272 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6151 - 564us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.9121 - 769us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.3009 - 613us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8574 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 165.7563 - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.3118 - 717us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.2889 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.5350 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.3510 - 595us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.2874 - 610us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.5397 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.2343 - 806us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.3812 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.3785 - 713us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3742 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.2065 - 720us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.4837 - 866us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.3498 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.0896 - 747us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.5693 - 3ms/epoch - 51us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7157 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.2873 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.0343 - 765us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7759 - 621us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.6460 - 806us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.3638 - 770us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.6482 - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.7122 - 710us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.1128 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.7216 - 568us/epoch - 9us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.6465 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.1430 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.3020 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.8033 - 689us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.3755 - 646us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.9454 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.4212 - 1ms/epoch - 23us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.3374 - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6810 - 760us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.1735 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.7424 - 615us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3331 - 764us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35.5628 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2100 - 734us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.6870 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.4629 - 3ms/epoch - 51us/sample\n", + "Episode 50/50\n", + "Generation 171/300\n", + "Solving for Nash Equilibrium in Generation 171/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.8663 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6427 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.4648 - 781us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.9530 - 645us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.3278 - 585us/epoch - 9us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8643 - 742us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.3177 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.7785 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2883 - 652us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.2581 - 733us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3682 - 716us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.3716 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.5548 - 734us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.1682 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.0385 - 835us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.2087 - 806us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.1307 - 2ms/epoch - 32us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.2114 - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.0060 - 961us/epoch - 16us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.2385 - 784us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.9512 - 3ms/epoch - 44us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.3625 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.8826 - 1ms/epoch - 22us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.6402 - 654us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.9130 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.5570 - 2ms/epoch - 32us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.9730 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.2502 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.7225 - 597us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.5201 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.1042 - 751us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.2695 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7228 - 601us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.1051 - 747us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.7521 - 1ms/epoch - 20us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.2090 - 684us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1326 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7163 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3989 - 1ms/epoch - 17us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.2216 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.8348 - 683us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.7254 - 779us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.1466 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.1669 - 635us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.3698 - 643us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4769 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.2077 - 651us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.0551 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.6671 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.9828 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9159 - 7ms/epoch - 112us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4121 - 698us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9020 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.3592 - 954us/epoch - 15us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.8373 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3333 - 695us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.8118 - 2ms/epoch - 34us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.1534 - 1ms/epoch - 20us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.3812 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.6547 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.8163 - 834us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.9628 - 749us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.7546 - 918us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4901 - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245.5936 - 1ms/epoch - 23us/sample\n", + "Generation 172/300\n", + "Solving for Nash Equilibrium in Generation 172/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.7248 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.9143 - 666us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7731 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.1509 - 847us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.7634 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.7405 - 774us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2259 - 753us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.9543 - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.3721 - 669us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.7907 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.1556 - 750us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.8175 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.0771 - 1ms/epoch - 17us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.8401 - 2ms/epoch - 35us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.3712 - 760us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6242 - 756us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.0166 - 6ms/epoch - 98us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.7980 - 669us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1529 - 1ms/epoch - 16us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.0443 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.3752 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0954 - 716us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.9207 - 1ms/epoch - 17us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.1992 - 689us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3348 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.7523 - 3ms/epoch - 44us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.2155 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.2826 - 829us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.7003 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.4972 - 636us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.7365 - 668us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7768 - 954us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.6410 - 843us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.5795 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.4027 - 833us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.6062 - 682us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.9025 - 767us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.3429 - 781us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41.4946 - 819us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8521 - 709us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.7229 - 765us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.4684 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.0820 - 590us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 12ms/epoch - 194us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.3652 - 1ms/epoch - 18us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38.9872 - 654us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.2112 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.3961 - 871us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.9707 - 698us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.5479 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240.7143 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.4266 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2120 - 642us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.7867 - 730us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.2865 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3503 - 916us/epoch - 15us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.6257 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.0397 - 608us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.5801 - 4ms/epoch - 68us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.3136 - 842us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.7004 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.9037 - 710us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.3256 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.8614 - 723us/epoch - 12us/sample\n", + "Generation 173/300\n", + "Solving for Nash Equilibrium in Generation 173/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.5027 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.4538 - 3ms/epoch - 49us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.4194 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.0739 - 674us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.8592 - 870us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.0533 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.9102 - 529us/epoch - 9us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.8281 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.6771 - 3ms/epoch - 49us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.2456 - 709us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.6122 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4118 - 853us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7531 - 716us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.0674 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.3808 - 3ms/epoch - 46us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.0324 - 730us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.0047 - 4ms/epoch - 64us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0762 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 141us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.4617 - 4ms/epoch - 61us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.0233 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.1772 - 1ms/epoch - 21us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.3929 - 4ms/epoch - 67us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8774 - 975us/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.3944 - 689us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.9927 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.8418 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5368 - 950us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.6425 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2565 - 796us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.1058 - 960us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.2931 - 818us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.5973 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.7754 - 838us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5446 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.0632 - 749us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.1344 - 698us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.7192 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.4861 - 824us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.8886 - 800us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.7562 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1332 - 873us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.6839 - 2ms/epoch - 29us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6002 - 722us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.2257 - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2753 - 764us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.2224 - 803us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.7207 - 813us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45.7698 - 7ms/epoch - 113us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.0276 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.9498 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.5684 - 787us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.2584 - 711us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.7083 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.4358 - 3ms/epoch - 46us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.4165 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.0965 - 720us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.3906 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2268 - 2ms/epoch - 38us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.1987 - 620us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.1332 - 907us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.9791 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 100.4016 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.1229 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.4318 - 699us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.6569 - 635us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.4743 - 731us/epoch - 12us/sample\n", + "Generation 174/300\n", + "Solving for Nash Equilibrium in Generation 174/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.5834 - 923us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.9315 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.5096 - 881us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.8788 - 2ms/epoch - 38us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0103 - 718us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0677 - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.6864 - 916us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.3556 - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.2644 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.1348 - 2ms/epoch - 34us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.6255 - 673us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.1849 - 738us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0104 - 948us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.6333 - 564us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6234 - 661us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.0148 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.3657 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.0915 - 796us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1638 - 772us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.5180 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.2067 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6812 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.9169 - 712us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.3005 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.8407 - 709us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5078 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6698 - 1ms/epoch - 18us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.5604 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9427 - 716us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.2150 - 848us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.9620 - 662us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.1225 - 640us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.3302 - 578us/epoch - 9us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38.3670 - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.6187 - 789us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248.3686 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.0949 - 635us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.7338 - 765us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.6890 - 3ms/epoch - 54us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.2914 - 708us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.9088 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.0176 - 1ms/epoch - 19us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2574 - 749us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.8599 - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.7346 - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.9740 - 776us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6469 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.4437 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4611 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2999 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.9951 - 1ms/epoch - 18us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9775 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.9416 - 728us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.9310 - 6ms/epoch - 101us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.2365 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.0870 - 698us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0993 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.4276 - 697us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.1487 - 739us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.0360 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.8380 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.1211 - 825us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.1378 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.4758 - 1ms/epoch - 19us/sample\n", + "Generation 175/300\n", + "Solving for Nash Equilibrium in Generation 175/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.9374 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9708 - 2ms/epoch - 40us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9211 - 631us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.3155 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1083 - 625us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.9672 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.4637 - 762us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.7229 - 821us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.9062 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.7641 - 783us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.1761 - 625us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.7924 - 588us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.7211 - 854us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.3109 - 809us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.6122 - 14ms/epoch - 227us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6899 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.6305 - 611us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.0129 - 767us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.6484 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.6335 - 906us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5233 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.8078 - 692us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.6087 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.7928 - 4ms/epoch - 66us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.4432 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.8825 - 631us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.3132 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1072 - 1ms/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6377 - 765us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.0699 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.2913 - 867us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.5614 - 710us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.4052 - 818us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201.7755 - 688us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788.0151 - 644us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.4096 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.6769 - 613us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.6221 - 705us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.7538 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9673 - 757us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.2195 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.0063 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.8890 - 711us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2352 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.0912 - 599us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.7846 - 723us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.2750 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.6188 - 625us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.2412 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.3551 - 663us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.5898 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.4049 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.9594 - 648us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3841 - 578us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.2303 - 670us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.7741 - 660us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.9413 - 596us/epoch - 10us/sample\n", + "Generation 176/300\n", + "Solving for Nash Equilibrium in Generation 176/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.5719 - 611us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3743 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.1162 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7870 - 644us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.1042 - 4ms/epoch - 58us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.7799 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.6642 - 669us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7905 - 724us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.5228 - 796us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.6862 - 581us/epoch - 9us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.6324 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8247 - 670us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.0173 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.5710 - 787us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.0325 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.2578 - 677us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3475 - 2ms/epoch - 39us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.4267 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.1453 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123.1785 - 651us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.4894 - 851us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.2165 - 695us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.6820 - 704us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.6142 - 13ms/epoch - 206us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43.2157 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.3596 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.6336 - 606us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.1625 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.3285 - 736us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.8069 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.5912 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5141 - 875us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.3875 - 705us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.7511 - 859us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.4564 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.5519 - 703us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.0079 - 716us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.3678 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.9483 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2268 - 745us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5313 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.3698 - 709us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.0350 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.3383 - 1ms/epoch - 17us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1395 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.0084 - 647us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6615 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.3534 - 798us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773.3345 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.6277 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.6814 - 922us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.8812 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.2888 - 644us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.5767 - 770us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.9441 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.3497 - 2ms/epoch - 35us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.6865 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.7593 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2255 - 748us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.6229 - 604us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.4269 - 801us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.1586 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.0760 - 665us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.8274 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.7859 - 749us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.6746 - 643us/epoch - 10us/sample\n", + "Generation 177/300\n", + "Solving for Nash Equilibrium in Generation 177/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.2433 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 225.5968 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.1862 - 654us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4178 - 671us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.6357 - 1ms/epoch - 23us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.5682 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 121.9871 - 641us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.4746 - 661us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.6606 - 586us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.3914 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.0045 - 918us/epoch - 15us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1901 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 118us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248.6650 - 7ms/epoch - 113us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.5899 - 673us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.7878 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.8221 - 5ms/epoch - 83us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.5426 - 814us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.4153 - 641us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.0097 - 952us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 362.3039 - 850us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.7787 - 818us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.0416 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.5421 - 763us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.9444 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.5621 - 2ms/epoch - 40us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9426 - 2ms/epoch - 25us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.5039 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.9390 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.7806 - 773us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.8629 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2081 - 678us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.8836 - 905us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0096 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.8362 - 633us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.2875 - 5ms/epoch - 83us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.7048 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.6671 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0394 - 826us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.8816 - 774us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4105 - 692us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.2610 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.9965 - 718us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6758 - 730us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.3651 - 620us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.3936 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.8834 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.9281 - 771us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.9601 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.0744 - 654us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.6928 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.7029 - 718us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.5625 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.9940 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.9977 - 731us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.8235 - 756us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.4113 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.1532 - 2ms/epoch - 29us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.2661 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.7126 - 975us/epoch - 16us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.3441 - 842us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2377 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.9842 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.8691 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.7569 - 691us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 178/300\n", + "Solving for Nash Equilibrium in Generation 178/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.2618 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5768 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5893 - 795us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.8749 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4031 - 738us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.9872 - 888us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8110 - 4ms/epoch - 71us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.9358 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.2001 - 593us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.9807 - 1ms/epoch - 19us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5490 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.8951 - 1ms/epoch - 21us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.4937 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.7672 - 695us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.3120 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.2561 - 644us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.7913 - 823us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.9285 - 2ms/epoch - 27us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4567 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.9860 - 1ms/epoch - 18us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.2459 - 650us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.8496 - 785us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.1772 - 782us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8818 - 1ms/epoch - 18us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248.8298 - 689us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2043 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3516 - 680us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.2190 - 589us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.9280 - 745us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.6635 - 600us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.4342 - 728us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.8368 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.8254 - 715us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.7247 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.9266 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.6379 - 582us/epoch - 9us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.6875 - 671us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.2627 - 665us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1172 - 903us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9919 - 694us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5171 - 783us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.9608 - 642us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8716 - 708us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.0341 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1243 - 596us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.5616 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.3227 - 634us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.9406 - 964us/epoch - 16us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40.4391 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.4229 - 1ms/epoch - 17us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9350 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.8439 - 710us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.2469 - 744us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.0374 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.4961 - 2ms/epoch - 32us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.5965 - 726us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9161 - 580us/epoch - 9us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 124.5646 - 695us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.5912 - 605us/epoch - 10us/sample\n", + "Generation 179/300\n", + "Solving for Nash Equilibrium in Generation 179/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.8402 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.6042 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.8480 - 728us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3556 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.3444 - 709us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.0412 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.8143 - 572us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.4484 - 716us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.6003 - 898us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.9608 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.8962 - 760us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.1985 - 651us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.2965 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.1750 - 654us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.1398 - 1ms/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.2743 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.6284 - 720us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.3934 - 696us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.3188 - 759us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.1549 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.0963 - 1ms/epoch - 23us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.7125 - 734us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.5149 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.6608 - 813us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.8823 - 822us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.1589 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.7072 - 654us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.4136 - 780us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.5643 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.4311 - 752us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.1329 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.6107 - 688us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.7601 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3260 - 896us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.3780 - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.3342 - 975us/epoch - 16us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2430 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.7673 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.0079 - 2ms/epoch - 25us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.7357 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.9179 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0136 - 609us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.5613 - 2ms/epoch - 28us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.7713 - 574us/epoch - 9us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.6103 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.9610 - 658us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.6187 - 1ms/epoch - 17us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.5217 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.3104 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8535 - 1ms/epoch - 17us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6034 - 1ms/epoch - 16us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.5152 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.1559 - 838us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.0799 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.9083 - 746us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9810 - 1ms/epoch - 21us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46.1481 - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.1994 - 788us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2841 - 641us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.3905 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.5730 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0795 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.7043 - 915us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.8803 - 658us/epoch - 11us/sample\n", + "Generation 180/300\n", + "Solving for Nash Equilibrium in Generation 180/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.7281 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.8159 - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.0050 - 1ms/epoch - 20us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.5470 - 1ms/epoch - 20us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.8713 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9799 - 951us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.5474 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.6272 - 1ms/epoch - 17us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.0490 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.9832 - 550us/epoch - 9us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3417 - 678us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.2744 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.9201 - 634us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.1723 - 949us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.6246 - 589us/epoch - 9us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.8800 - 3ms/epoch - 41us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4229 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.2277 - 699us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8676 - 671us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8276 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.8507 - 694us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6902 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.8352 - 1ms/epoch - 16us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.4787 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.3112 - 635us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.3266 - 906us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.2313 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.9929 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.0013 - 635us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7214 - 737us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.7577 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.4388 - 771us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4627 - 727us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.4098 - 698us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.9835 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5836 - 6ms/epoch - 94us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.1573 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.3088 - 683us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8223 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1390 - 583us/epoch - 9us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.9525 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.7890 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.4117 - 728us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.8750 - 770us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7516 - 720us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.4197 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.3917 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2308 - 623us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.3811 - 733us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.9489 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.7665 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.0974 - 833us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9548 - 729us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.8889 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.1252 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.2657 - 635us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0685 - 734us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1605 - 581us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.3962 - 936us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.4169 - 622us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6276 - 719us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5206 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.0536 - 653us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.2518 - 592us/epoch - 10us/sample\n", + "Generation 181/300\n", + "Solving for Nash Equilibrium in Generation 181/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.9747 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.7207 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2950 - 629us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 137.3776 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0990 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.2234 - 813us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8060 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.5405 - 644us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.9084 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.9011 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.7406 - 604us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.8704 - 753us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.8464 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.4690 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.6483 - 828us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.4059 - 671us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.6182 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.5237 - 725us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.9626 - 625us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.1594 - 663us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.2419 - 666us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.2081 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.7305 - 662us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.3539 - 568us/epoch - 9us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.8972 - 6ms/epoch - 102us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5630 - 728us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.5487 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0077 - 788us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.2221 - 797us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.1380 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.5793 - 728us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.9891 - 613us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.6807 - 814us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.0663 - 1ms/epoch - 17us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.0457 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.9000 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.4130 - 677us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.6501 - 625us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.8835 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.8424 - 784us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.4264 - 779us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.1095 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8249 - 630us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.7428 - 1ms/epoch - 20us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.2298 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.3785 - 654us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.3008 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.2649 - 842us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.6330 - 955us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.8238 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.9346 - 655us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 148.4433 - 765us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.3776 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6155 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.3360 - 797us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.0139 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.2967 - 783us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.7648 - 843us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.9850 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.8517 - 2ms/epoch - 40us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.0981 - 719us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.2029 - 792us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 164us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.4905 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.3112 - 665us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.1004 - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.5797 - 5ms/epoch - 81us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.6459 - 710us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2411 - 688us/epoch - 11us/sample\n", + "Generation 182/300\n", + "Solving for Nash Equilibrium in Generation 182/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.2068 - 763us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.3850 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.7721 - 2ms/epoch - 25us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.9638 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.5796 - 2ms/epoch - 28us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.2003 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.8217 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.7556 - 683us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.6340 - 628us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.9609 - 676us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3689 - 712us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.7594 - 595us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.4681 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4275 - 551us/epoch - 9us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.4195 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.4433 - 774us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.6805 - 3ms/epoch - 47us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.6580 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.7191 - 713us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.6305 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.2973 - 648us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.3745 - 837us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.2597 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240.0808 - 1ms/epoch - 19us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3536 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.0367 - 608us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.2172 - 639us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.1385 - 765us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.9107 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.1814 - 3ms/epoch - 43us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.9146 - 1ms/epoch - 22us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.2095 - 978us/epoch - 16us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.4709 - 761us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.7856 - 683us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.8544 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.0119 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.7946 - 833us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3732 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.0231 - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3580 - 648us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4119 - 651us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.2275 - 661us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5641 - 726us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.9992 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.6258 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2827 - 604us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 362.3799 - 579us/epoch - 9us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8683 - 715us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.1728 - 2ms/epoch - 32us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.8918 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.7410 - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.7835 - 668us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8696 - 904us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8681 - 1ms/epoch - 21us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.8567 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5551 - 678us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4523 - 805us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.9663 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.4824 - 589us/epoch - 9us/sample\n", + "Generation 183/300\n", + "Solving for Nash Equilibrium in Generation 183/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.6195 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.0705 - 6ms/epoch - 100us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.7810 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4034 - 2ms/epoch - 32us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.9035 - 931us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.6466 - 866us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6366 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.5689 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.8279 - 781us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.0289 - 604us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0985 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.0904 - 1ms/epoch - 17us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.4316 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.0800 - 1ms/epoch - 21us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.0849 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5048 - 759us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0062 - 2ms/epoch - 35us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.1141 - 874us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.0273 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5989 - 1ms/epoch - 16us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.3444 - 887us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7885 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.3256 - 601us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2495 - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.5474 - 991us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.5834 - 7ms/epoch - 116us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4660 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0328 - 605us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9103 - 665us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.1527 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.8345 - 1ms/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.3312 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.3844 - 1ms/epoch - 17us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.5386 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.0012 - 792us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.0862 - 904us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5762 - 869us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.0426 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4191 - 1ms/epoch - 20us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.2421 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9213 - 942us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.3147 - 2ms/epoch - 28us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.3234 - 833us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.5405 - 1ms/epoch - 18us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.1472 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.1035 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.8782 - 917us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.2350 - 700us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.6684 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.4271 - 3ms/epoch - 50us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.3983 - 709us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.4890 - 660us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.4952 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.2318 - 738us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.9611 - 602us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.9572 - 900us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.7654 - 836us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.3532 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.7886 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240.6967 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.8891 - 1ms/epoch - 19us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8973 - 767us/epoch - 12us/sample\n", + "Generation 184/300\n", + "Solving for Nash Equilibrium in Generation 184/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.4091 - 817us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.9919 - 592us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.0054 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8290 - 2ms/epoch - 29us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1395 - 576us/epoch - 9us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.0112 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.3182 - 743us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3577 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.2247 - 690us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.7975 - 685us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.6046 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.7451 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.4535 - 718us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.2613 - 785us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.0634 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.0616 - 782us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.3936 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.3774 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.2379 - 628us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.1168 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4417 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9662 - 715us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.4338 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.7974 - 727us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.1665 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.4373 - 665us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6324 - 637us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.2512 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.1096 - 732us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7811 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.1259 - 628us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.2363 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.1347 - 758us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.1269 - 630us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.0978 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7652 - 669us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.6335 - 751us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.2944 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.0183 - 3ms/epoch - 50us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.0662 - 846us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.0760 - 1ms/epoch - 16us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.1090 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.0797 - 781us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.3660 - 907us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.7743 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.1266 - 725us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.7969 - 629us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.1998 - 6ms/epoch - 93us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.1918 - 742us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.7150 - 760us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6998 - 671us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.5849 - 635us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.8027 - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.7436 - 1ms/epoch - 23us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.3201 - 3ms/epoch - 53us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 185.9536 - 602us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.0153 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4016 - 600us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.0507 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.7767 - 625us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.7854 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1486 - 1ms/epoch - 18us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.1394 - 2ms/epoch - 27us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.9999 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.2631 - 755us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.0712 - 1ms/epoch - 18us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361.2486 - 601us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.0278 - 668us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.6901 - 4ms/epoch - 60us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 185/300\n", + "Solving for Nash Equilibrium in Generation 185/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.5666 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.8911 - 2ms/epoch - 25us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.0381 - 579us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.7318 - 685us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.5447 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.5480 - 6ms/epoch - 97us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.5654 - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.2653 - 850us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.1208 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.2847 - 584us/epoch - 9us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8141 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.8437 - 678us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.1684 - 742us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.0554 - 701us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.3962 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.3248 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.8339 - 615us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.9573 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.1292 - 530us/epoch - 9us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3207 - 703us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.5896 - 911us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.0982 - 564us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.6473 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.1261 - 1ms/epoch - 17us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.0526 - 678us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.6821 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.5997 - 701us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.6622 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.5656 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.0858 - 688us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.3818 - 720us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.6888 - 753us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.2803 - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.1372 - 720us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6913 - 936us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.6711 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4645 - 678us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.1389 - 663us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.0655 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.2579 - 48ms/epoch - 770us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.6414 - 852us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.2042 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.0892 - 762us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5097 - 627us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0848 - 782us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.1207 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.2665 - 612us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.0359 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6602 - 705us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 130.8583 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.6849 - 747us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.5759 - 793us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.8076 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.9609 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.6410 - 2ms/epoch - 40us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.1982 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.0613 - 874us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.8271 - 751us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.7567 - 684us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.8433 - 975us/epoch - 16us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.9313 - 613us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.1376 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.2653 - 716us/epoch - 12us/sample\n", + "Generation 186/300\n", + "Solving for Nash Equilibrium in Generation 186/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.0479 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5776 - 697us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 17ms/epoch - 277us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231.9506 - 7ms/epoch - 119us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.4057 - 723us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1094 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.6964 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.7715 - 662us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.2785 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.5854 - 927us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.0943 - 1ms/epoch - 21us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.4604 - 704us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.9015 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.0118 - 756us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.4332 - 806us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.3553 - 665us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5262 - 662us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8256 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.7358 - 840us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.2841 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1724 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.7805 - 728us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3480 - 1ms/epoch - 16us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.9494 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.6807 - 924us/epoch - 15us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.1028 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5372 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.9451 - 659us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.7015 - 863us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2209 - 819us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.9610 - 827us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.9162 - 738us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.0355 - 676us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9399 - 2ms/epoch - 30us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.4480 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.9852 - 757us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.5469 - 748us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.7029 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.6473 - 695us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8707 - 600us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.2249 - 721us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.2762 - 694us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3779 - 667us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.6886 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.6623 - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7073 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.5352 - 668us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.0865 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.6750 - 646us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.9161 - 1ms/epoch - 19us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2404 - 661us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.8353 - 743us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.4733 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3297 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.4243 - 663us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.7814 - 705us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.1491 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.0347 - 578us/epoch - 9us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.0468 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.7467 - 664us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.3928 - 742us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.9348 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.0780 - 656us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.6652 - 709us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0704 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.9590 - 3ms/epoch - 46us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.3117 - 801us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.0028 - 752us/epoch - 12us/sample\n", + "Generation 187/300\n", + "Solving for Nash Equilibrium in Generation 187/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.5236 - 551us/epoch - 9us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.3649 - 680us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.0965 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.7690 - 1ms/epoch - 19us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.6028 - 671us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.1287 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.7471 - 741us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.7816 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.1117 - 957us/epoch - 15us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.8951 - 738us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.5163 - 653us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.8655 - 768us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.1777 - 1ms/epoch - 20us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.2364 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 151us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1086 - 6ms/epoch - 94us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3913 - 853us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.1028 - 2ms/epoch - 27us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.5202 - 1ms/epoch - 18us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.3591 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3017 - 721us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.4119 - 893us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.5256 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 222.7367 - 588us/epoch - 9us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8928 - 592us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.6274 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.7776 - 773us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.6491 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8569 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.5958 - 725us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0536 - 5ms/epoch - 76us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.8889 - 675us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.0924 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.6509 - 646us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.4164 - 688us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.4623 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.8397 - 764us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.6315 - 645us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.7715 - 1ms/epoch - 21us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.2031 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.8276 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.8176 - 631us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.7589 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4579 - 819us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.0323 - 881us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7827 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.7245 - 2ms/epoch - 29us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.2884 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.7633 - 597us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.1394 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.5573 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.4691 - 788us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5413 - 3ms/epoch - 46us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43.2619 - 2ms/epoch - 30us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5443 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.9065 - 1ms/epoch - 17us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.6760 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1566 - 994us/epoch - 16us/sample\n", + "Generation 188/300\n", + "Solving for Nash Equilibrium in Generation 188/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 215.4801 - 3ms/epoch - 50us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.8233 - 716us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.7739 - 693us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.0170 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.7590 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.6206 - 707us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.7490 - 605us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.6984 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.2788 - 737us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.5048 - 616us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.9718 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.0250 - 576us/epoch - 9us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.2213 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8179 - 744us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.9609 - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8901 - 647us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.6661 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.0485 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.3124 - 4ms/epoch - 59us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.8513 - 3ms/epoch - 47us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.5383 - 663us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7771 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.6194 - 701us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.7900 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 766.6839 - 830us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2690 - 882us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.5441 - 883us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.9030 - 11ms/epoch - 181us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.5471 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.2958 - 718us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9983 - 679us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.0356 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.4069 - 752us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.6010 - 849us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.3429 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.6024 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.7280 - 2ms/epoch - 26us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.7603 - 741us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.8610 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.3754 - 774us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.2623 - 982us/epoch - 16us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 67.9814 - 677us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9026 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.2521 - 850us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.7109 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.3499 - 3ms/epoch - 47us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.2231 - 706us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.9048 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.4242 - 662us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.5395 - 868us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.9166 - 651us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 123us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.7875 - 1ms/epoch - 20us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.9440 - 5ms/epoch - 74us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45.1684 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.9229 - 824us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3246 - 818us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.1385 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.5148 - 605us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.7953 - 790us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.3122 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.2536 - 584us/epoch - 9us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.3041 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1511 - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.7371 - 678us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4654 - 669us/epoch - 11us/sample\n", + "Generation 189/300\n", + "Solving for Nash Equilibrium in Generation 189/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.6131 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.1822 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.8608 - 889us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.0700 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.3353 - 693us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.5738 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4352 - 771us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.3734 - 777us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.0118 - 762us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.6807 - 1ms/epoch - 18us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.2628 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.5388 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.8964 - 613us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.8701 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.2184 - 693us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.8950 - 2ms/epoch - 25us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.6294 - 900us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0412 - 849us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5522 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.9255 - 808us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.7034 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.4501 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.9884 - 789us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.7335 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.7134 - 932us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8984 - 747us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1792 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.1860 - 720us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1738 - 717us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2955 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.4371 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2236 - 1ms/epoch - 18us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2302 - 953us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.5710 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.4911 - 580us/epoch - 9us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.1456 - 773us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.7971 - 693us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3762 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5877 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.4850 - 586us/epoch - 9us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.0434 - 747us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0593 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.7697 - 666us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.6985 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7808 - 631us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.5637 - 1ms/epoch - 20us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.1485 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.3414 - 1ms/epoch - 20us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.2098 - 822us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.0120 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.4100 - 695us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.2544 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.7677 - 605us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.9990 - 712us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.6992 - 693us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7476 - 29ms/epoch - 475us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.1393 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.9672 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4198 - 979us/epoch - 16us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.5112 - 756us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.0304 - 711us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.1896 - 746us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.1295 - 2ms/epoch - 29us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.7342 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.5944 - 2ms/epoch - 26us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 198.8904 - 1ms/epoch - 16us/sample\n", + "Generation 190/300\n", + "Solving for Nash Equilibrium in Generation 190/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.2866 - 839us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.2741 - 719us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.5549 - 688us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2443 - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.7385 - 835us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.8260 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.6693 - 1ms/epoch - 17us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 435.8578 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.0001 - 1ms/epoch - 20us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.8154 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.3036 - 1ms/epoch - 18us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.8273 - 803us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.4442 - 1ms/epoch - 23us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.5723 - 1ms/epoch - 24us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.2830 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2455 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8347 - 2ms/epoch - 30us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.6444 - 818us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.5308 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.5668 - 848us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.7040 - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.2606 - 905us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.5096 - 810us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.3709 - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.8679 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.9075 - 826us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.5737 - 811us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.2076 - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3128 - 939us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 114us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.3080 - 976us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 122.9015 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6093 - 1ms/epoch - 19us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.9410 - 711us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.8804 - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.6939 - 852us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.7776 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3899 - 857us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5823 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.1761 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.0393 - 597us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.4541 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.1173 - 748us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9114 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.1754 - 656us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.4129 - 661us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.0627 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6742 - 940us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.6824 - 959us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.1280 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.9050 - 797us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2592 - 766us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.5233 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 114us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.7507 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.0993 - 805us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.2388 - 831us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.3677 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3160 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406.1631 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5176 - 746us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.7455 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.6562 - 1ms/epoch - 17us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.9243 - 689us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.2577 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9066 - 601us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.1890 - 837us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.7911 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.5526 - 935us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.6925 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.7051 - 704us/epoch - 11us/sample\n", + "Generation 191/300\n", + "Solving for Nash Equilibrium in Generation 191/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.3560 - 803us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.4176 - 832us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.8383 - 633us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0062 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.7434 - 1ms/epoch - 18us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.2168 - 859us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 777.1165 - 825us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.3665 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.1187 - 628us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.3236 - 19ms/epoch - 300us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.3139 - 4ms/epoch - 64us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.6129 - 1ms/epoch - 16us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.5628 - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.4905 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.7046 - 5ms/epoch - 83us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4517 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.8987 - 732us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.3833 - 791us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.4185 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7133 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2203 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.9954 - 772us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.6165 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.9505 - 716us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.2141 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2265 - 763us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.1243 - 2ms/epoch - 35us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.3133 - 842us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6730 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.7664 - 9ms/epoch - 151us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4103 - 745us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.7653 - 1ms/epoch - 18us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.6400 - 810us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.2709 - 895us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 85.2526 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.3033 - 793us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.7509 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.2971 - 714us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.9620 - 721us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.5686 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.8080 - 644us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6293 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1553 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.2862 - 726us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.5704 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.7078 - 551us/epoch - 9us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1353 - 770us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.7323 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.9708 - 795us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.3520 - 690us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.8743 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.8813 - 811us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.1691 - 813us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.5851 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.8802 - 707us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.9657 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.0734 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.8047 - 662us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.7268 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.4925 - 630us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0668 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.2957 - 607us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.6925 - 911us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.9949 - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.4855 - 799us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.3644 - 1ms/epoch - 23us/sample\n", + "Generation 192/300\n", + "Solving for Nash Equilibrium in Generation 192/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.5531 - 789us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.4251 - 909us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.1877 - 2ms/epoch - 26us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.5443 - 844us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.2794 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.5827 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.5906 - 662us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5291 - 1ms/epoch - 22us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.7432 - 659us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.9764 - 756us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8624 - 2ms/epoch - 40us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6356 - 677us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.4441 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.0718 - 917us/epoch - 15us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.9113 - 609us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.4638 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.6342 - 589us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.0198 - 727us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.8346 - 698us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.6321 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.1425 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.1362 - 1ms/epoch - 23us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.2050 - 628us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4648 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.2958 - 690us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7280 - 872us/epoch - 14us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.1434 - 614us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.1161 - 954us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.8444 - 735us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.2528 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.3135 - 916us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.5797 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1274 - 793us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.2715 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.8120 - 651us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.4666 - 588us/epoch - 9us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.4971 - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.2538 - 810us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.6026 - 3ms/epoch - 45us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.8133 - 693us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.7685 - 735us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.7830 - 685us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.1914 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.4458 - 1ms/epoch - 22us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.1696 - 782us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.9241 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.7654 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.2898 - 2ms/epoch - 25us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.8607 - 867us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.8133 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37.2730 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.6222 - 718us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.0849 - 1ms/epoch - 22us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.6284 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.8205 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.0952 - 679us/epoch - 11us/sample\n", + "Generation 193/300\n", + "Solving for Nash Equilibrium in Generation 193/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.4727 - 731us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43.7189 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.3695 - 921us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.3976 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.6310 - 713us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8134 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0641 - 2ms/epoch - 30us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.5536 - 2ms/epoch - 34us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.8852 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.4849 - 836us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.9836 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.2184 - 729us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.8502 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7630 - 687us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.6602 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.3975 - 767us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.5338 - 600us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5962 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.8687 - 579us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.4773 - 764us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.2430 - 651us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.4914 - 844us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.6445 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.3431 - 687us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5330 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.3605 - 937us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.4427 - 766us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.8888 - 5ms/epoch - 76us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.2894 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.2874 - 931us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.6407 - 829us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.9095 - 1ms/epoch - 20us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.5309 - 794us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.2510 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.2696 - 636us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46.5423 - 1ms/epoch - 24us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.4555 - 821us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.8221 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.4842 - 9ms/epoch - 145us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.8283 - 689us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.9169 - 2ms/epoch - 25us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.2304 - 1ms/epoch - 18us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7625 - 5ms/epoch - 81us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.8399 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.0446 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.5284 - 783us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.8699 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.7316 - 841us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5565 - 3ms/epoch - 53us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 13ms/epoch - 203us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.6217 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.6780 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.4543 - 3ms/epoch - 43us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.0171 - 2ms/epoch - 31us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.8608 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7654 - 2ms/epoch - 32us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.3166 - 733us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.4622 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.7164 - 629us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.9925 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1116 - 741us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5301 - 815us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6296 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.7231 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.1574 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.5172 - 980us/epoch - 16us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.6902 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.6515 - 740us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 194/300\n", + "Solving for Nash Equilibrium in Generation 194/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.0500 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.3569 - 686us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.2377 - 693us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.9440 - 764us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.8821 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.8102 - 571us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.7514 - 760us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.4304 - 650us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.0118 - 2ms/epoch - 28us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.6799 - 860us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.3809 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.4914 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.4400 - 727us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.4087 - 659us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.8206 - 843us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9642 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.7869 - 2ms/epoch - 30us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.9211 - 839us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.8300 - 569us/epoch - 9us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.4396 - 727us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.6595 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.6287 - 788us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.6780 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.3849 - 671us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.0067 - 850us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.9003 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7909 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.6520 - 910us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.0982 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.7464 - 2ms/epoch - 25us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.3939 - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.0689 - 881us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.4858 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0642 - 881us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.4236 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.9273 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.0174 - 740us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7333 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0389 - 998us/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.3372 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8552 - 948us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.5184 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.4102 - 767us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.4570 - 659us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.1617 - 635us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 101us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.5770 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.9095 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.4911 - 683us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.0122 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.4597 - 945us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.4524 - 758us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.1481 - 597us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4978 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9462 - 2ms/epoch - 27us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.1510 - 5ms/epoch - 73us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.9636 - 650us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54.9128 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9438 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.8027 - 670us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.8986 - 696us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 135us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.1894 - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.8479 - 722us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.1730 - 2ms/epoch - 34us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.4661 - 930us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.7988 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.2314 - 722us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7027 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.1893 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.9461 - 759us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.4835 - 839us/epoch - 14us/sample\n", + "Generation 195/300\n", + "Solving for Nash Equilibrium in Generation 195/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.4359 - 668us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.5271 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.5266 - 597us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.7647 - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.3777 - 840us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.4221 - 688us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1962 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8928 - 7ms/epoch - 113us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5789 - 941us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.8480 - 722us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.3193 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3375 - 642us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.8784 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.1944 - 703us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.0631 - 898us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.1819 - 751us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.4464 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8408 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.4884 - 715us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6599 - 675us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.1670 - 679us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.7838 - 553us/epoch - 9us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.8731 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.5515 - 767us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3723 - 616us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.0128 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.2379 - 662us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 116us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.8040 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.3232 - 662us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0897 - 1ms/epoch - 19us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.3439 - 643us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.9698 - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.6202 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5489 - 937us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.4309 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.4141 - 782us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.9176 - 1ms/epoch - 22us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 32ms/epoch - 518us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4648 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2303 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.1254 - 743us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.9916 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.8915 - 792us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.6027 - 1ms/epoch - 17us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.1581 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.7267 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.0218 - 684us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0405 - 644us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.3182 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.8812 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7339 - 927us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4009 - 747us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.9433 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.2858 - 744us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.8792 - 648us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.7173 - 710us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.9064 - 740us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.1893 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.5622 - 1ms/epoch - 17us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.2847 - 677us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.5041 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.4163 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.8324 - 1ms/epoch - 19us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.3222 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.0126 - 1ms/epoch - 17us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.1493 - 685us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.4770 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.4573 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8250 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.7285 - 653us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.4551 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.1931 - 1ms/epoch - 17us/sample\n", + "Episode 50/50\n", + "Generation 196/300\n", + "Solving for Nash Equilibrium in Generation 196/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0999 - 705us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.9318 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.0633 - 2ms/epoch - 36us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.3377 - 736us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.5841 - 738us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.7655 - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8019 - 756us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 217.7541 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.9274 - 7ms/epoch - 114us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.5826 - 729us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1447 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5587 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.6362 - 1ms/epoch - 23us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.6108 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.4242 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7662 - 2ms/epoch - 29us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8978 - 608us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.2988 - 790us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.5564 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.0596 - 786us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.9671 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.5597 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.8275 - 801us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.8934 - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.5728 - 712us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.0188 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.3696 - 651us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.5793 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.7546 - 624us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.2210 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.6502 - 642us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3610 - 943us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3363 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5916 - 1ms/epoch - 19us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.1791 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.8925 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.6857 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.7443 - 2ms/epoch - 28us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.8850 - 813us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4361 - 557us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.4937 - 750us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.7858 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.8099 - 730us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 128us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.1941 - 1ms/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.5840 - 2ms/epoch - 36us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.6467 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.8262 - 760us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.1173 - 802us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.0173 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.1069 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1127 - 619us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.2692 - 814us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.9802 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.1842 - 928us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 104us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.0582 - 3ms/epoch - 45us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.2235 - 726us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.5876 - 583us/epoch - 9us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.6044 - 600us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.4799 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8257 - 925us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.1157 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.6629 - 813us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.5852 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.7808 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250.0165 - 4ms/epoch - 70us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.8158 - 1ms/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.6830 - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9731 - 705us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.7988 - 699us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1394 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.0045 - 1ms/epoch - 19us/sample\n", + "Generation 197/300\n", + "Solving for Nash Equilibrium in Generation 197/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.1598 - 568us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.3160 - 658us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7871 - 864us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.3973 - 796us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.8399 - 696us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.5302 - 731us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.3659 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.5849 - 4ms/epoch - 59us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.3523 - 927us/epoch - 15us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.8976 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.6720 - 772us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.7366 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.6192 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.1404 - 921us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7902 - 650us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.5020 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.6547 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.6871 - 767us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.1612 - 826us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.8494 - 625us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.6967 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.3104 - 578us/epoch - 9us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.6818 - 662us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.6133 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.7195 - 677us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.9415 - 832us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.2016 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.3774 - 516us/epoch - 8us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.4842 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.2449 - 949us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.7534 - 6ms/epoch - 100us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.5772 - 1ms/epoch - 19us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.5690 - 855us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.8321 - 707us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.9971 - 741us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.4959 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.2195 - 1ms/epoch - 18us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.7864 - 653us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6095 - 776us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0776 - 803us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.4776 - 713us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.4675 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1240 - 1ms/epoch - 17us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231.3358 - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.8005 - 862us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8781 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2899 - 649us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.2930 - 774us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.2643 - 738us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.9425 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.7135 - 5ms/epoch - 74us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.0800 - 645us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.7488 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.4224 - 4ms/epoch - 65us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.9384 - 582us/epoch - 9us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.0325 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4348 - 858us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4098 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8763 - 802us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4218 - 691us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.1747 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.8496 - 702us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53.0066 - 900us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.8278 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.3232 - 607us/epoch - 10us/sample\n", + "Generation 198/300\n", + "Solving for Nash Equilibrium in Generation 198/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7708 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214.5357 - 5ms/epoch - 77us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0276 - 699us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.9274 - 899us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.5155 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.6163 - 610us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 179.4659 - 2ms/epoch - 37us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.6187 - 725us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.3900 - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.1164 - 665us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.9360 - 3ms/epoch - 44us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5857 - 995us/epoch - 16us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.8834 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.7036 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3975 - 763us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43.6487 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.8396 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.5230 - 777us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.1492 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4415 - 639us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.8702 - 10ms/epoch - 160us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.9665 - 575us/epoch - 9us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.9679 - 4ms/epoch - 69us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4124 - 4ms/epoch - 65us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.4633 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.0421 - 2ms/epoch - 25us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.5545 - 1ms/epoch - 17us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.7728 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.9578 - 680us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.4423 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.3594 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.7006 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.3710 - 681us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.4079 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.2522 - 2ms/epoch - 30us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.8741 - 5ms/epoch - 75us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.1794 - 662us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1677 - 867us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.7247 - 610us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6339 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.9821 - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.8043 - 696us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.4156 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.7228 - 3ms/epoch - 55us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.7570 - 775us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.0805 - 1ms/epoch - 22us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.9359 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.7653 - 726us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.5917 - 680us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.8189 - 604us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.4145 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.1519 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3944 - 632us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48.9149 - 685us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.0511 - 894us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.7841 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.8231 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.2931 - 608us/epoch - 10us/sample\n", + "Generation 199/300\n", + "Solving for Nash Equilibrium in Generation 199/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.3673 - 677us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4190 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.6685 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.7679 - 721us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.9833 - 725us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.1626 - 796us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.1796 - 625us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8413 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6122 - 711us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3741 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 147.7051 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7271 - 700us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0607 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.3477 - 576us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.9878 - 712us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.1872 - 768us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.7940 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.8711 - 683us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.8431 - 680us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.4998 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8125 - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3896 - 713us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.4759 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.3641 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.9078 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.0027 - 680us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.6793 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6273 - 662us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.2742 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.9772 - 706us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7076 - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.6712 - 2ms/epoch - 26us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8341 - 661us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.1655 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.0147 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.2789 - 1ms/epoch - 17us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.1297 - 960us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.0109 - 728us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8306 - 979us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.8322 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.9922 - 705us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.5652 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.9580 - 846us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.6837 - 665us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.2721 - 620us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.9599 - 2ms/epoch - 32us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.4176 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.7905 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.3123 - 879us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.7986 - 637us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.3991 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1757 - 604us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.1292 - 894us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.8469 - 703us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.5584 - 7ms/epoch - 108us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.2433 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.5400 - 941us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.2220 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.2253 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.5589 - 833us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.2593 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.3951 - 918us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.7814 - 897us/epoch - 14us/sample\n", + "Generation 200/300\n", + "Solving for Nash Equilibrium in Generation 200/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.8552 - 786us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.4359 - 778us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.8526 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.8425 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.6165 - 601us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.3714 - 721us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.2739 - 6ms/epoch - 94us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6241 - 759us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 432.1645 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.1356 - 725us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.9629 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.4180 - 926us/epoch - 15us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.4954 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.7963 - 607us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.9197 - 684us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.4691 - 774us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.9709 - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.6840 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.6773 - 612us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.5989 - 746us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.7281 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.1212 - 748us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.3714 - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8857 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4665 - 1ms/epoch - 19us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.1915 - 696us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.0322 - 6ms/epoch - 98us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.4880 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.0357 - 3ms/epoch - 49us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.2048 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.0585 - 747us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.1541 - 801us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428.8923 - 2ms/epoch - 26us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.3006 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0529 - 808us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1102 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.1962 - 760us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.6758 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.7161 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.1191 - 716us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.4572 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.8630 - 2ms/epoch - 25us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5117 - 741us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.2970 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.6159 - 731us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9988 - 617us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.5213 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.2296 - 818us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.9511 - 859us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4419 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.7467 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.3018 - 748us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0562 - 727us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.8350 - 826us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.3776 - 698us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.0828 - 693us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.8237 - 916us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.4994 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.9648 - 707us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2808 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.4318 - 903us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9297 - 770us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.7752 - 989us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Generation 201/300\n", + "Solving for Nash Equilibrium in Generation 201/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7478 - 735us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2224 - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.0023 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.6221 - 650us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6412 - 720us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.4081 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.8903 - 824us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5417 - 716us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.4725 - 827us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.0417 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.9926 - 2ms/epoch - 27us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.5060 - 784us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.3188 - 1ms/epoch - 17us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.3277 - 717us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248.4445 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.2599 - 849us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1444 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.9143 - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.5184 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.8597 - 812us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.0811 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.5094 - 743us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.2804 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.2059 - 618us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8455 - 713us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.9808 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8456 - 575us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7300 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3419 - 767us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.1948 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.9330 - 671us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.8746 - 657us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.8332 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.8648 - 3ms/epoch - 50us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.9673 - 651us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.8968 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.4161 - 662us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.5844 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.1307 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.0833 - 777us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7087 - 667us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.4940 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6269 - 4ms/epoch - 58us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41.5769 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0464 - 789us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.8753 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.6750 - 634us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.2993 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.5824 - 2ms/epoch - 28us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.4247 - 702us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.2303 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6954 - 744us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.0823 - 839us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.0548 - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5313 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.2557 - 736us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.3355 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.5217 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.5698 - 1ms/epoch - 16us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.8934 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.9446 - 645us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.5906 - 744us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.9320 - 693us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.0353 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.7311 - 710us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.6555 - 564us/epoch - 9us/sample\n", + "Generation 202/300\n", + "Solving for Nash Equilibrium in Generation 202/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.5555 - 973us/epoch - 16us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.3035 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5089 - 662us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.9286 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.6934 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7733 - 779us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.9335 - 691us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.5370 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.6078 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.6604 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.6163 - 734us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.8635 - 735us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.4188 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.0246 - 3ms/epoch - 41us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.5916 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4056 - 3ms/epoch - 46us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.0315 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.0611 - 773us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.5180 - 798us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5903 - 622us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.8340 - 557us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.6431 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.1550 - 4ms/epoch - 72us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.6416 - 618us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.2927 - 659us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.7685 - 636us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.4821 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.4008 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250.0850 - 2ms/epoch - 31us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.8940 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.4471 - 3ms/epoch - 48us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.3718 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.7960 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.6389 - 1ms/epoch - 20us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.7997 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.3179 - 634us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.7953 - 703us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.8393 - 747us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.1955 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.2455 - 652us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.6053 - 1ms/epoch - 18us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.1892 - 704us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.3525 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.4061 - 728us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773.0453 - 1ms/epoch - 22us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.3050 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.5660 - 4ms/epoch - 60us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.8852 - 863us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.0523 - 712us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4673 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.9352 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38.1205 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.3772 - 859us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.0374 - 760us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.2494 - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.9980 - 609us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 99us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.5477 - 7ms/epoch - 107us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0281 - 7ms/epoch - 109us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.1429 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.2717 - 672us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.9208 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.7934 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.6854 - 751us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3167 - 3ms/epoch - 42us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.4332 - 609us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.9443 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.8507 - 835us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.0056 - 695us/epoch - 11us/sample\n", + "Generation 203/300\n", + "Solving for Nash Equilibrium in Generation 203/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.2088 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.3038 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.4892 - 546us/epoch - 9us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.3844 - 1ms/epoch - 18us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7543 - 697us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.2231 - 1ms/epoch - 18us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.5567 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.9859 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.9397 - 712us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.6569 - 693us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.4051 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.8892 - 1ms/epoch - 18us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.7161 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.4508 - 902us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.7015 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.1102 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.9355 - 1ms/epoch - 16us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.9425 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.9021 - 625us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.8930 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.9684 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.4464 - 870us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.5392 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6447 - 754us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.7197 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.3887 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.1394 - 712us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.2708 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.7016 - 816us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.0267 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4590 - 2ms/epoch - 35us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.5257 - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.4731 - 751us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.5971 - 811us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.2351 - 763us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.0233 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.5386 - 2ms/epoch - 29us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.9389 - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.2069 - 1ms/epoch - 19us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5640 - 839us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.0097 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2054 - 1ms/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 234.0231 - 1ms/epoch - 18us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.3772 - 717us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.7218 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.5674 - 1ms/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6749 - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1095 - 8ms/epoch - 134us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.9599 - 3ms/epoch - 48us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.4723 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.9922 - 646us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.3193 - 658us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.2341 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.2208 - 3ms/epoch - 50us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4283 - 3ms/epoch - 46us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 229.3296 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.6372 - 611us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8987 - 795us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.8241 - 662us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.3650 - 1ms/epoch - 16us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9348 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.2657 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.0740 - 875us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9423 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.2683 - 567us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6738 - 619us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.3895 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.1941 - 734us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.1491 - 696us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.6271 - 796us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.6982 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.7190 - 2ms/epoch - 38us/sample\n", + "Generation 204/300\n", + "Solving for Nash Equilibrium in Generation 204/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.7791 - 704us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.1697 - 586us/epoch - 9us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.3842 - 865us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.4649 - 1ms/epoch - 23us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406.2440 - 985us/epoch - 16us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.9947 - 898us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.9412 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.1269 - 694us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201.0646 - 1ms/epoch - 24us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.9674 - 798us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1303 - 701us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.0022 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.5048 - 610us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7786 - 658us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.6423 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.6509 - 1ms/epoch - 16us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.2384 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.1249 - 643us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.4778 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.0048 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.7318 - 719us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 212.3229 - 759us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9978 - 630us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7479 - 643us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.4639 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.3033 - 834us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.0727 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.0961 - 978us/epoch - 16us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.3043 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.6010 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.3916 - 760us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.0846 - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.2502 - 627us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.4652 - 2ms/epoch - 28us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.3775 - 856us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.8669 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5334 - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.4375 - 801us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.7881 - 743us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.8373 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.8375 - 798us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.3739 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.1420 - 2ms/epoch - 25us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.0750 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6230 - 629us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.6342 - 651us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.6805 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1172 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7176 - 702us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.2404 - 887us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3271 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4585 - 3ms/epoch - 41us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.0335 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.4394 - 2ms/epoch - 31us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.3308 - 2ms/epoch - 35us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.5154 - 754us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.9348 - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.4500 - 772us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.9474 - 704us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.0740 - 698us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.8038 - 750us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7362 - 959us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.3880 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.9065 - 562us/epoch - 9us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.4633 - 889us/epoch - 14us/sample\n", + "Generation 205/300\n", + "Solving for Nash Equilibrium in Generation 205/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.1777 - 959us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.7234 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.0337 - 688us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.9369 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.1196 - 795us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.6036 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.1033 - 645us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.6366 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.2792 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8132 - 719us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.1649 - 864us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.9734 - 687us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.0198 - 815us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.4742 - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.8817 - 923us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.0872 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.3458 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.3345 - 634us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.0026 - 620us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.7915 - 569us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.2385 - 669us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.2587 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.4174 - 696us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.7578 - 607us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0719 - 695us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.5772 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.0639 - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.5977 - 960us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.4247 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4264 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.7621 - 536us/epoch - 9us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.4337 - 652us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.2595 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.1017 - 893us/epoch - 14us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.0823 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.5884 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.1293 - 716us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4225 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.7013 - 597us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.7037 - 772us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.8221 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.1273 - 786us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.4732 - 817us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.8102 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.0510 - 540us/epoch - 9us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.4253 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.4127 - 650us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.0754 - 694us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.3709 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7798 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.4228 - 641us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1770 - 738us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.9170 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.1597 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.8297 - 735us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.6543 - 6ms/epoch - 90us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.2788 - 726us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.0126 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0264 - 764us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9941 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.0546 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.7453 - 761us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.8417 - 658us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.7248 - 838us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.8386 - 741us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.4042 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.3046 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8026 - 648us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.2808 - 686us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.3756 - 819us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.7010 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9167 - 755us/epoch - 12us/sample\n", + "Generation 206/300\n", + "Solving for Nash Equilibrium in Generation 206/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.1316 - 779us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.3194 - 2ms/epoch - 33us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.2843 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.5652 - 2ms/epoch - 32us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.0536 - 780us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.9557 - 675us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.3594 - 806us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3829 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4930 - 725us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.1620 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.5351 - 906us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.8156 - 650us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.1585 - 848us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5894 - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.4454 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.1111 - 2ms/epoch - 30us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2363 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.7722 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.5147 - 1ms/epoch - 23us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.9692 - 750us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.4623 - 715us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.8532 - 879us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.1956 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.8716 - 748us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9446 - 673us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.0894 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.4602 - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.3516 - 830us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.3920 - 833us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.9196 - 607us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.0835 - 744us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.6183 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.0148 - 644us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.7545 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.5594 - 505us/epoch - 8us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3309 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3802 - 4ms/epoch - 67us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7406 - 824us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4846 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.8400 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9133 - 713us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9778 - 691us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2742 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.0853 - 607us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4299 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.1648 - 738us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.0238 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.8272 - 620us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.9862 - 721us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.7989 - 702us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.7959 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.0493 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2405 - 836us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.2542 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.7453 - 797us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1108 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3998 - 689us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4146 - 852us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.3580 - 702us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55.1111 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7159 - 632us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 207/300\n", + "Solving for Nash Equilibrium in Generation 207/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.3328 - 715us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.8762 - 7ms/epoch - 114us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.4904 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.5882 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.0797 - 705us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6645 - 715us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.5775 - 725us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.4194 - 905us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.4398 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6452 - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.1467 - 878us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.6257 - 748us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.6532 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3566 - 750us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.7785 - 918us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.7579 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230.0080 - 764us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.9809 - 713us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.3191 - 778us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.7591 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.0132 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.6425 - 877us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.6190 - 850us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.8278 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.9554 - 711us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6589 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9661 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.2780 - 797us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.0699 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.1909 - 605us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.9379 - 827us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.6338 - 646us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.4734 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.7399 - 556us/epoch - 9us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.2196 - 668us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4462 - 681us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.9269 - 712us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.6148 - 767us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.2913 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.1854 - 953us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.2979 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.7628 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.3868 - 663us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.4524 - 804us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.2249 - 1ms/epoch - 22us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.8719 - 594us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.2784 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.1261 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.3503 - 802us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.9919 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.3233 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.3686 - 571us/epoch - 9us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.2951 - 9ms/epoch - 146us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.6945 - 693us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.0575 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.6151 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.7105 - 998us/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.5112 - 733us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.1521 - 740us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.9935 - 16ms/epoch - 255us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.4454 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.6243 - 730us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.2436 - 890us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.0516 - 1ms/epoch - 18us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.3233 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.5355 - 683us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2845 - 672us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230.9521 - 585us/epoch - 9us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4060 - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.1840 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.8865 - 723us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.1824 - 753us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 80.4295 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.7003 - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.3514 - 759us/epoch - 12us/sample\n", + "Generation 208/300\n", + "Solving for Nash Equilibrium in Generation 208/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3301 - 689us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7049 - 626us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.0370 - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.2701 - 800us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.7926 - 700us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.1366 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.9905 - 1ms/epoch - 17us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.3195 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.9636 - 904us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2038 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.5384 - 712us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.1061 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.6708 - 635us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.9003 - 892us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.5393 - 644us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.7198 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.5593 - 682us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.6011 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.9310 - 766us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.7458 - 673us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.2868 - 664us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.1562 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.2514 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.4789 - 724us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.6526 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.9450 - 789us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.6937 - 4ms/epoch - 58us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3556 - 1ms/epoch - 23us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.4125 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.6185 - 6ms/epoch - 92us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.0891 - 705us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8161 - 569us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.1189 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.8542 - 834us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.2235 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.8255 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 761.8818 - 651us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2785 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.4459 - 721us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5733 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2189 - 691us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.8181 - 826us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.4429 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.1465 - 878us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.1315 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9716 - 1ms/epoch - 19us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.4327 - 804us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.2255 - 697us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.6607 - 656us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.6017 - 1ms/epoch - 17us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.9529 - 856us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.6112 - 765us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.2268 - 854us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.1592 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.9194 - 11ms/epoch - 173us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.0490 - 4ms/epoch - 60us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.6034 - 1ms/epoch - 17us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.5295 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.6812 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.9524 - 864us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.0937 - 658us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 181.4979 - 840us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.3103 - 757us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3287 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.0576 - 727us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.9785 - 819us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Generation 209/300\n", + "Solving for Nash Equilibrium in Generation 209/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.4917 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.9339 - 821us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2886 - 815us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.4610 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.1362 - 743us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.6412 - 913us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74.3516 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.2709 - 5ms/epoch - 88us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.1575 - 3ms/epoch - 42us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0820 - 764us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.0630 - 804us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.4213 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.8774 - 1ms/epoch - 17us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.6537 - 836us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.3616 - 689us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9761 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.8252 - 741us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.1157 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3210 - 2ms/epoch - 28us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.4350 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.4737 - 702us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 163.7416 - 705us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.0764 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.9125 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7657 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.8474 - 754us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.6951 - 681us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8176 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1443 - 890us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.4524 - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.2478 - 737us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 12ms/epoch - 194us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.2087 - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.9176 - 1ms/epoch - 17us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.4538 - 712us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.7507 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.5183 - 730us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.9160 - 5ms/epoch - 82us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.9753 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.1221 - 651us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.9384 - 806us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.3065 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.4034 - 1ms/epoch - 22us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.1455 - 668us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8476 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.7904 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.6260 - 14ms/epoch - 225us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7725 - 2ms/epoch - 25us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.7755 - 750us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.2194 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.3854 - 578us/epoch - 9us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.6061 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.8867 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.9821 - 1ms/epoch - 21us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214.0440 - 722us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.4451 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.1024 - 996us/epoch - 16us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.2492 - 3ms/epoch - 47us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.7337 - 2ms/epoch - 28us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2422 - 2ms/epoch - 25us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.9289 - 964us/epoch - 16us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.9542 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.9718 - 879us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.3182 - 2ms/epoch - 27us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.9538 - 724us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.3919 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.1707 - 1ms/epoch - 21us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.6229 - 2ms/epoch - 25us/sample\n", + "Generation 210/300\n", + "Solving for Nash Equilibrium in Generation 210/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.1683 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.0868 - 769us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.1107 - 656us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8519 - 765us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.5800 - 636us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 100us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 105.5790 - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.7520 - 1ms/epoch - 21us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.4691 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.8710 - 9ms/epoch - 142us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9107 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.9635 - 3ms/epoch - 50us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.5086 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.7214 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.0302 - 1ms/epoch - 18us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.5994 - 952us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.7839 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1169 - 714us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.9934 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.4913 - 855us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.9740 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.7475 - 814us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.2310 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0637 - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.8087 - 963us/epoch - 16us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.4158 - 737us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.1691 - 2ms/epoch - 31us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6907 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.7323 - 6ms/epoch - 90us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.5518 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.6039 - 621us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1449 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5623 - 890us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3857 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.6189 - 776us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.9066 - 5ms/epoch - 83us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.9052 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.8408 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.8565 - 6ms/epoch - 90us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.6849 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.9908 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.0754 - 937us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.4542 - 1ms/epoch - 18us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.6253 - 2ms/epoch - 39us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9526 - 847us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.4600 - 1ms/epoch - 16us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4047 - 1ms/epoch - 17us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.3136 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5437 - 759us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.2519 - 959us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.5847 - 779us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.4073 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.7263 - 764us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.9285 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.4165 - 837us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.3527 - 1ms/epoch - 23us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.9940 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.1793 - 16ms/epoch - 255us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 228.1307 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5965 - 680us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.2870 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.1778 - 691us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.2106 - 1ms/epoch - 17us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.4349 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 119us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.4615 - 3ms/epoch - 54us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.2981 - 751us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.4205 - 688us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.7660 - 3ms/epoch - 56us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51.6027 - 793us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 186.2168 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.8822 - 816us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.3827 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.0627 - 6ms/epoch - 91us/sample\n", + "Generation 211/300\n", + "Solving for Nash Equilibrium in Generation 211/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.3875 - 2ms/epoch - 25us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.9171 - 2ms/epoch - 39us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.1857 - 936us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.2646 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6773 - 7ms/epoch - 111us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.4822 - 688us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.0941 - 896us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.7312 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.2401 - 2ms/epoch - 24us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.5005 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.0995 - 796us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.9909 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.2904 - 578us/epoch - 9us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.0497 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.9984 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.3893 - 861us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.5269 - 786us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.6835 - 639us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.1694 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.1315 - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 234.2464 - 759us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.0350 - 2ms/epoch - 34us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.1573 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 13ms/epoch - 205us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.9896 - 1ms/epoch - 17us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.0526 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5129 - 664us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.1978 - 2ms/epoch - 31us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2502 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.7482 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.2990 - 936us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.4813 - 716us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9655 - 3ms/epoch - 45us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.3987 - 567us/epoch - 9us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.2141 - 3ms/epoch - 45us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.3492 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.0013 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.4534 - 924us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.0168 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.7293 - 661us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.0436 - 756us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.0069 - 842us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.9733 - 661us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.3475 - 695us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.8159 - 587us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.2913 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.7703 - 7ms/epoch - 109us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.4411 - 591us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1674 - 578us/epoch - 9us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.7672 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.3480 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.7952 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.1321 - 2ms/epoch - 31us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.9432 - 793us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.5268 - 3ms/epoch - 43us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.1951 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.6887 - 580us/epoch - 9us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.3754 - 767us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.7480 - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.9979 - 786us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.4205 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2128 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.8905 - 924us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45.1348 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.7486 - 1ms/epoch - 19us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6718 - 651us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.8875 - 786us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2010 - 765us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8635 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.2426 - 722us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.6672 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.4402 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.0706 - 588us/epoch - 9us/sample\n", + "Generation 212/300\n", + "Solving for Nash Equilibrium in Generation 212/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.2958 - 776us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.1237 - 717us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.3558 - 710us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.4721 - 3ms/epoch - 47us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.3207 - 763us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.9713 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9216 - 998us/epoch - 16us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.4315 - 678us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.6689 - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.2060 - 742us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.2628 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.2148 - 1ms/epoch - 18us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 229.7730 - 1ms/epoch - 21us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3463 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.4126 - 2ms/epoch - 33us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.0925 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.8986 - 1ms/epoch - 20us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.4222 - 777us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3793 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.8339 - 683us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.3604 - 638us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3690 - 1ms/epoch - 17us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.3162 - 801us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.4500 - 892us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 127us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.8300 - 2ms/epoch - 28us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.2248 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9759 - 875us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.1284 - 944us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.8035 - 2ms/epoch - 30us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.2078 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 21ms/epoch - 331us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.8988 - 995us/epoch - 16us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.8912 - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.5514 - 1ms/epoch - 17us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.0950 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.1961 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.7855 - 2ms/epoch - 37us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.7570 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.9122 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 91us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3679 - 5ms/epoch - 81us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.6299 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.8878 - 617us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.7574 - 837us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.6644 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.5030 - 1ms/epoch - 20us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.7150 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2369 - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.0816 - 735us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.3417 - 785us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.6923 - 915us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.5957 - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.1259 - 581us/epoch - 9us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.3906 - 2ms/epoch - 29us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.4570 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 432.8680 - 2ms/epoch - 28us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.5023 - 887us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.1386 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.9289 - 883us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7270 - 791us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.1497 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.6050 - 628us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.3098 - 769us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.5316 - 2ms/epoch - 26us/sample\n", + "Episode 50/50\n", + "Generation 213/300\n", + "Solving for Nash Equilibrium in Generation 213/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.0156 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.2772 - 686us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.0272 - 675us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 153.8040 - 2ms/epoch - 32us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.0865 - 819us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.9520 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.0978 - 789us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.7818 - 695us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.9368 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.1564 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.1959 - 3ms/epoch - 51us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.4672 - 586us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.8367 - 892us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.3221 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.0366 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.0593 - 899us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.3922 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6163 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3427 - 985us/epoch - 16us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.6343 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.0298 - 792us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1019 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1352 - 723us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.2501 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6323 - 711us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.6915 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.6456 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8977 - 878us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.6094 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.0261 - 824us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.1553 - 808us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.2933 - 686us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.3054 - 3ms/epoch - 44us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.9956 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.3644 - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.4500 - 984us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6432 - 1ms/epoch - 22us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.6995 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.0396 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.5083 - 814us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.6165 - 695us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6484 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2104 - 851us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.5370 - 879us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.8585 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.0323 - 2ms/epoch - 27us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.8911 - 677us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.6661 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.7561 - 617us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.6759 - 695us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4059 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.5079 - 680us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.1760 - 702us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5645 - 872us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.0404 - 710us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.9229 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.2079 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 125us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.7414 - 1ms/epoch - 21us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8644 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2419 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9296 - 3ms/epoch - 49us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2542 - 749us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.0128 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6502 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.5374 - 610us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Generation 214/300\n", + "Solving for Nash Equilibrium in Generation 214/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.5349 - 633us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.0974 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8079 - 752us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.9989 - 916us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.9904 - 716us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.1799 - 1ms/epoch - 21us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.5349 - 1ms/epoch - 19us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 125.9194 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7438 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.0084 - 884us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0229 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.0051 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.5065 - 2ms/epoch - 37us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.5360 - 2ms/epoch - 31us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.7724 - 2ms/epoch - 32us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.4351 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.0364 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.6513 - 5ms/epoch - 88us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.6739 - 3ms/epoch - 43us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.7475 - 791us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.5461 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.3010 - 727us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.6960 - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.5697 - 664us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.1207 - 599us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.7233 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.1684 - 905us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.0016 - 800us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4083 - 918us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8053 - 831us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.9155 - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.3070 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.6451 - 576us/epoch - 9us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.6213 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.9046 - 3ms/epoch - 53us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8922 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1809 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.0297 - 773us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.3032 - 727us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 144.8216 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0440 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.9608 - 718us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.0071 - 723us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9686 - 805us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.3962 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5297 - 2ms/epoch - 31us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.4929 - 843us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4251 - 652us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.0552 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.7537 - 720us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.4768 - 934us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.2463 - 649us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.1025 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231.9861 - 692us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.6665 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.6355 - 2ms/epoch - 33us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.8168 - 806us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7265 - 677us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.7146 - 836us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.7149 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.1972 - 586us/epoch - 9us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0238 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.0974 - 639us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.1104 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.7628 - 3ms/epoch - 45us/sample\n", + "Generation 215/300\n", + "Solving for Nash Equilibrium in Generation 215/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.8836 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.1216 - 1ms/epoch - 18us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.3112 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.0606 - 796us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 156us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.1297 - 7ms/epoch - 113us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.3773 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.5456 - 1ms/epoch - 23us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.2273 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.0648 - 801us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.1281 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.6126 - 781us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.8809 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.6555 - 645us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.0709 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.6817 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.8043 - 3ms/epoch - 45us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.7299 - 1ms/epoch - 17us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7055 - 751us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.0233 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.4372 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.7024 - 1ms/epoch - 19us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2799 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.1775 - 698us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4921 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.6552 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2704 - 868us/epoch - 14us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.1255 - 842us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.7249 - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729.2626 - 2ms/epoch - 29us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.0908 - 4ms/epoch - 66us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.2879 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.1704 - 2ms/epoch - 27us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.2136 - 951us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3189 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.9609 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.6427 - 887us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.6538 - 757us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1851 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.2786 - 819us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9016 - 777us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.6548 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.1608 - 868us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.7806 - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.4195 - 971us/epoch - 16us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.7839 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.4871 - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 84us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.3654 - 2ms/epoch - 38us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.1617 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.2221 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.9733 - 4ms/epoch - 65us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.2789 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 149us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7743 - 1ms/epoch - 19us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3376 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.2671 - 895us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.1682 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.1832 - 1ms/epoch - 21us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.8998 - 1ms/epoch - 20us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.9998 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.8689 - 743us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.4368 - 731us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.2632 - 862us/epoch - 14us/sample\n", + "Generation 216/300\n", + "Solving for Nash Equilibrium in Generation 216/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.9541 - 803us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.6959 - 1ms/epoch - 16us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.0576 - 1ms/epoch - 19us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.2902 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.5406 - 1ms/epoch - 18us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.3697 - 677us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.2604 - 2ms/epoch - 27us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.7634 - 838us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.2444 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.5298 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.1617 - 14ms/epoch - 233us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7769 - 788us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.7009 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.6695 - 741us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2032 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.3589 - 901us/epoch - 15us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.8334 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.7646 - 1ms/epoch - 18us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.4539 - 894us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.3680 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.2894 - 822us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.6682 - 727us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.9364 - 732us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2508 - 755us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1655 - 708us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.6844 - 2ms/epoch - 30us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4205 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.4130 - 2ms/epoch - 26us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.7226 - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.2367 - 6ms/epoch - 94us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.4137 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.9311 - 874us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.3656 - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.6945 - 769us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.5501 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4693 - 766us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3664 - 729us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.9960 - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.6795 - 937us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.9583 - 703us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.8931 - 647us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3798 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.2163 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.2240 - 603us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1472 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.9894 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9222 - 918us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.8619 - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.3098 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.9083 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.9329 - 796us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.9777 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.7551 - 571us/epoch - 9us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.2024 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.4319 - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.9243 - 992us/epoch - 16us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.6344 - 835us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.4782 - 735us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8246 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4513 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.5547 - 816us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.1258 - 2ms/epoch - 30us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.3531 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.2639 - 2ms/epoch - 32us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.8488 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6175 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2104 - 910us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.9248 - 717us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1526 - 741us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.0347 - 909us/epoch - 15us/sample\n", + "Generation 217/300\n", + "Solving for Nash Equilibrium in Generation 217/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0161 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232.3024 - 772us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9863 - 1ms/epoch - 22us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2036 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.3339 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.2179 - 718us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.8171 - 728us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4410 - 759us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.2351 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.8275 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.5759 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4177 - 896us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.9645 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.5756 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.3176 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.1236 - 1ms/epoch - 17us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.0592 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773.3005 - 661us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.4897 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.6696 - 970us/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.7859 - 755us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.2951 - 2ms/epoch - 28us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5524 - 741us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.6952 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.2438 - 2ms/epoch - 37us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.5325 - 689us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.9078 - 712us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.0788 - 754us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7842 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6339 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7429 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5921 - 657us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5426 - 833us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.4332 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.1814 - 703us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.5467 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4071 - 682us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.3079 - 708us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.7520 - 872us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.2511 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9300 - 896us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0820 - 911us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.4319 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.7653 - 715us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.9218 - 671us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.1491 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.9962 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.9308 - 674us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.1902 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.3065 - 1ms/epoch - 18us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.5150 - 640us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 118.9276 - 835us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.3950 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.5629 - 643us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.5806 - 678us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.1413 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4640 - 703us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.0122 - 699us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.8822 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.3931 - 655us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774.3532 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.4218 - 561us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.6895 - 724us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.9416 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.3567 - 8ms/epoch - 135us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.8807 - 797us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.3691 - 593us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.8378 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.4451 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.2941 - 978us/epoch - 16us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8884 - 995us/epoch - 16us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.5486 - 716us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250.8086 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.4767 - 15ms/epoch - 247us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.8955 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.6400 - 763us/epoch - 12us/sample\n", + "Generation 218/300\n", + "Solving for Nash Equilibrium in Generation 218/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.4742 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.0701 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.9796 - 724us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.9423 - 767us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.1719 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.3233 - 2ms/epoch - 26us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.4187 - 754us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.3607 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.7178 - 711us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0057 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2477 - 1ms/epoch - 18us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.4187 - 762us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.0507 - 775us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4049 - 689us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245.3589 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6329 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.2921 - 731us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6093 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4610 - 694us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.0879 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.8004 - 702us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.6618 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.3775 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.5397 - 784us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.5772 - 2ms/epoch - 39us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.4337 - 679us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.8236 - 686us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.5316 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.2609 - 670us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.3483 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.6649 - 735us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.4036 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.5766 - 849us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.5276 - 726us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.0809 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.8840 - 2ms/epoch - 38us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5015 - 791us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.8717 - 1ms/epoch - 17us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.7470 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 115us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.1426 - 2ms/epoch - 35us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.1154 - 2ms/epoch - 32us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.6559 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.3445 - 612us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1909 - 887us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.6520 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.4068 - 1ms/epoch - 16us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.8148 - 1ms/epoch - 23us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.9998 - 682us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.1490 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.9222 - 735us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.7932 - 734us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.9227 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 219.7552 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0834 - 843us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.5232 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.5964 - 592us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0385 - 632us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3463 - 733us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.4140 - 641us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.0695 - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.8912 - 753us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.3720 - 726us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.5492 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.5220 - 914us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.9796 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.5607 - 3ms/epoch - 42us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.3374 - 716us/epoch - 12us/sample\n", + "Generation 219/300\n", + "Solving for Nash Equilibrium in Generation 219/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6815 - 2ms/epoch - 30us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 743.8412 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.3347 - 624us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 219.9781 - 723us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.1174 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.2122 - 1ms/epoch - 17us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.9532 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.4438 - 905us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5145 - 739us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1743 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.1725 - 809us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.4023 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4457 - 2ms/epoch - 33us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.5315 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.3550 - 578us/epoch - 9us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 100us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.4512 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.6204 - 694us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.7028 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.8371 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.4641 - 651us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.0611 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.2819 - 636us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.5186 - 623us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.4842 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8147 - 763us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 143.2776 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.4010 - 931us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.6875 - 890us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5515 - 667us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6470 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3386 - 587us/epoch - 9us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.2029 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.6078 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.8992 - 960us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.9779 - 900us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.5115 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.2709 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.9622 - 634us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.2284 - 2ms/epoch - 30us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.1534 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.2305 - 700us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.0157 - 689us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.8104 - 548us/epoch - 9us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.8732 - 707us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.3149 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.5462 - 715us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.9088 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.9303 - 701us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.4446 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.8170 - 2ms/epoch - 30us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.0242 - 687us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.3386 - 600us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.3134 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.1900 - 733us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2770 - 793us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.2416 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.7483 - 625us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.8235 - 670us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.8907 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.8768 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.8079 - 670us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.6849 - 641us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.9787 - 667us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.7226 - 836us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.2875 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.6892 - 822us/epoch - 13us/sample\n", + "Generation 220/300\n", + "Solving for Nash Equilibrium in Generation 220/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.6744 - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 362.2310 - 2ms/epoch - 32us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.9239 - 747us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.2057 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.4726 - 932us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.0745 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.8076 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232.1271 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.1663 - 713us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.6568 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.1447 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.8477 - 634us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.9413 - 726us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.0336 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.6639 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.2510 - 797us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5492 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.5477 - 695us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.6330 - 702us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.9603 - 773us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.6808 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.9056 - 713us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.7601 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.9778 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759.8405 - 722us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.0260 - 777us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.9893 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.5752 - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.8688 - 3ms/epoch - 50us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1480 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4233 - 3ms/epoch - 45us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6877 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.8665 - 659us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.7713 - 586us/epoch - 9us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.4268 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.7967 - 916us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6723 - 727us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.9504 - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.5646 - 731us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714.1559 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6299 - 675us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.3480 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.1987 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8309 - 597us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.1935 - 653us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4431 - 3ms/epoch - 55us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.7298 - 703us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.0798 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.1107 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3997 - 775us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.6664 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.9452 - 835us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.3016 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.8393 - 687us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.3539 - 697us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.4848 - 592us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.5525 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.9927 - 670us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.5026 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.0450 - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.9677 - 1ms/epoch - 17us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.4053 - 794us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6688 - 715us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.6215 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762.9026 - 683us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.2083 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.8585 - 724us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 221/300\n", + "Solving for Nash Equilibrium in Generation 221/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.3435 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.4653 - 710us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.4922 - 909us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781.5418 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.4796 - 848us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.8122 - 656us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.8460 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.1014 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4381 - 678us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.6437 - 867us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.0213 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.9073 - 906us/epoch - 15us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.9510 - 667us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.2279 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.8997 - 778us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.1732 - 939us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7973 - 594us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.3539 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3476 - 684us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.0216 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.0865 - 601us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9269 - 756us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.3859 - 712us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.8516 - 776us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3621 - 648us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51.7496 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.9785 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.3967 - 729us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4930 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.5686 - 1ms/epoch - 16us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.0004 - 808us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.3243 - 2ms/epoch - 29us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 253.9309 - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.7790 - 606us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0297 - 673us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1215 - 654us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2245 - 1ms/epoch - 18us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.5661 - 588us/epoch - 9us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.0587 - 632us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3940 - 871us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50.0413 - 605us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1820 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5928 - 783us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.0327 - 750us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0851 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.9379 - 11ms/epoch - 182us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0892 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7504 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 159us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.3652 - 1ms/epoch - 20us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.3065 - 563us/epoch - 9us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.1971 - 731us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230.1911 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.9012 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9701 - 840us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.6817 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.9433 - 703us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.3308 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.5439 - 677us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.4089 - 789us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.3483 - 680us/epoch - 11us/sample\n", + "Generation 222/300\n", + "Solving for Nash Equilibrium in Generation 222/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.3640 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.3698 - 613us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.5299 - 716us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.7655 - 702us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.2150 - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.1600 - 1ms/epoch - 18us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.2369 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 780.2617 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.1709 - 768us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.4559 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.2358 - 778us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0043 - 756us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4674 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2301 - 2ms/epoch - 29us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.1950 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.4096 - 750us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.4576 - 744us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.3724 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.1975 - 727us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.8579 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.9765 - 598us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.2262 - 639us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.2844 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2998 - 750us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.4519 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.4923 - 676us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.1514 - 728us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.1350 - 779us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.4077 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.8079 - 708us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.5240 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.6606 - 612us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.1170 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.2406 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.0285 - 956us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.9499 - 719us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.8110 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.1122 - 780us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8634 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.6221 - 698us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4797 - 736us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.1992 - 714us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5851 - 1ms/epoch - 19us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6736 - 2ms/epoch - 29us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.9545 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.9429 - 865us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.7768 - 713us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.0437 - 2ms/epoch - 26us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.3403 - 6ms/epoch - 99us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.6639 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.1852 - 601us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.8058 - 622us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.3468 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.1406 - 744us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.1311 - 695us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.8396 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.6610 - 751us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7628 - 770us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.1464 - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.3901 - 3ms/epoch - 47us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.5693 - 815us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 135.2644 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.7382 - 562us/epoch - 9us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.5587 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.3307 - 672us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.6159 - 812us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0153 - 1ms/epoch - 18us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.9654 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.5262 - 763us/epoch - 12us/sample\n", + "Generation 223/300\n", + "Solving for Nash Equilibrium in Generation 223/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.5355 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.0336 - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.4988 - 994us/epoch - 16us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.6855 - 726us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.9724 - 802us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.3196 - 728us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.5446 - 2ms/epoch - 25us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 163.9953 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.5829 - 792us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214.2481 - 1ms/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.4641 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.2316 - 1ms/epoch - 22us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.1211 - 1ms/epoch - 18us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.1549 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 19ms/epoch - 305us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.4766 - 14ms/epoch - 220us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.8693 - 809us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 11ms/epoch - 174us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.4706 - 8ms/epoch - 128us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7769 - 829us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.2735 - 742us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.6470 - 761us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.2043 - 789us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.4200 - 937us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.0469 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.0311 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.7888 - 832us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235.1407 - 720us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.2096 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.0081 - 975us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.5748 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.0510 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.8143 - 687us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5369 - 679us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6301 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1195 - 590us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.7708 - 703us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.7369 - 622us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.2399 - 821us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.1922 - 2ms/epoch - 35us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.8502 - 785us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.6317 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.5800 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.8203 - 1ms/epoch - 21us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.3976 - 669us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6238 - 781us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1133 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.1294 - 646us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.6653 - 1ms/epoch - 19us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.3647 - 841us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.4495 - 692us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.2711 - 1ms/epoch - 23us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.3119 - 794us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2092 - 753us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8776 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.8005 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8715 - 1ms/epoch - 20us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.9280 - 700us/epoch - 11us/sample\n", + "Generation 224/300\n", + "Solving for Nash Equilibrium in Generation 224/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7582 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.7251 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.5159 - 4ms/epoch - 57us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.1850 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 217.8327 - 606us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.3586 - 798us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.1265 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.7954 - 894us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.1408 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.2592 - 1ms/epoch - 21us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1827 - 2ms/epoch - 31us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.3316 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.4057 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.5544 - 608us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.0776 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.2703 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.5690 - 866us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.1033 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.7882 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.8305 - 1ms/epoch - 24us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.5952 - 670us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3458 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0217 - 571us/epoch - 9us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.2703 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.4220 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.4015 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0958 - 1ms/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.8650 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0544 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.9566 - 743us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0001 - 655us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.6122 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3041 - 929us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.3546 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.3934 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.1442 - 972us/epoch - 16us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.0735 - 6ms/epoch - 93us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2374 - 993us/epoch - 16us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.3810 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.0309 - 1ms/epoch - 17us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.9266 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.2457 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0296 - 744us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.2288 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.0137 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.1995 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.7056 - 1ms/epoch - 21us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.9172 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.2179 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9316 - 705us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.8174 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4333 - 678us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6857 - 685us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2404 - 678us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9796 - 905us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7099 - 2ms/epoch - 38us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38.6979 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.8454 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.3432 - 775us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.8079 - 658us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.5895 - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.5969 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306.0262 - 630us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.7427 - 643us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6078 - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.3596 - 629us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.0534 - 812us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Generation 225/300\n", + "Solving for Nash Equilibrium in Generation 225/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.9925 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.8283 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.3761 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.0749 - 732us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.8559 - 3ms/epoch - 43us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.2785 - 686us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.0556 - 596us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.3635 - 960us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.2578 - 670us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.5051 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.7195 - 661us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5817 - 731us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.9589 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.0385 - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 229.9035 - 12ms/epoch - 197us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.3945 - 699us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.1480 - 4ms/epoch - 60us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.1384 - 705us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.8909 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9934 - 686us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.8945 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8954 - 592us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.3448 - 10ms/epoch - 162us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 138us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.8595 - 990us/epoch - 16us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.9904 - 706us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.4623 - 1ms/epoch - 16us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9068 - 648us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.1783 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.9386 - 999us/epoch - 16us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 91us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.9435 - 1ms/epoch - 17us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.1915 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 435.6305 - 606us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.4812 - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.3403 - 751us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.8151 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8821 - 6ms/epoch - 91us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.5993 - 971us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.9132 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.9662 - 636us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.2805 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.9402 - 840us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.0714 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737.6226 - 869us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.8799 - 949us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.5022 - 5ms/epoch - 84us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.2951 - 8ms/epoch - 136us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.0101 - 736us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40.6173 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.1499 - 701us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.5952 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.2843 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 74ms/epoch - 1ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.0615 - 8ms/epoch - 135us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.4177 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.1335 - 777us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7366 - 910us/epoch - 15us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.9217 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.8771 - 724us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6616 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2935 - 2ms/epoch - 28us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.6099 - 635us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.3317 - 1ms/epoch - 21us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.3122 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 203.2938 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.0167 - 745us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1110 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.4160 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.5939 - 859us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.3866 - 669us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.3415 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2128 - 3ms/epoch - 41us/sample\n", + "Generation 226/300\n", + "Solving for Nash Equilibrium in Generation 226/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.7172 - 701us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.9232 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1365 - 692us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.2149 - 2ms/epoch - 26us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5448 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361.8314 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.1208 - 730us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.3775 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.0977 - 770us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6968 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 14ms/epoch - 230us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.4572 - 1ms/epoch - 16us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.6718 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.4071 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.5764 - 745us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5795 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.7278 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.9177 - 742us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.6619 - 705us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.6102 - 824us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7185 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8993 - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.9633 - 2ms/epoch - 30us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0258 - 1ms/epoch - 23us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.5661 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 15ms/epoch - 234us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.0570 - 3ms/epoch - 49us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5152 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.6631 - 811us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.5726 - 697us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.4562 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.8000 - 4ms/epoch - 65us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.9081 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.3733 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.3284 - 681us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.1403 - 757us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.0693 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.9703 - 2ms/epoch - 28us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.5717 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9462 - 646us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.9016 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.3050 - 611us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.1809 - 805us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.9010 - 719us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.3854 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.1250 - 697us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.5477 - 728us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.6423 - 723us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.7989 - 1ms/epoch - 20us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.8569 - 707us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.9066 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.0960 - 883us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.0770 - 743us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.5512 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.7086 - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.0643 - 885us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.4116 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8070 - 611us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.9212 - 742us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.7664 - 688us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5093 - 663us/epoch - 11us/sample\n", + "Generation 227/300\n", + "Solving for Nash Equilibrium in Generation 227/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.0990 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 99us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.2278 - 2ms/epoch - 24us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.3044 - 891us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.0310 - 731us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 215.0756 - 695us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.8835 - 611us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.9490 - 1ms/epoch - 17us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.5638 - 557us/epoch - 9us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.9156 - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.5945 - 638us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.2571 - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.5919 - 865us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.0984 - 704us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.2894 - 3ms/epoch - 46us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8422 - 1ms/epoch - 23us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.5063 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.1983 - 4ms/epoch - 63us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.0387 - 722us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2247 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.8711 - 2ms/epoch - 28us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.6406 - 1ms/epoch - 18us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.6989 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.4375 - 898us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.9171 - 2ms/epoch - 25us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.2122 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9930 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.3984 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.0264 - 2ms/epoch - 26us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.5600 - 690us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.0105 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.5186 - 604us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.0320 - 556us/epoch - 9us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.0060 - 1ms/epoch - 19us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6666 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3195 - 699us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.2187 - 743us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.0735 - 691us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.7272 - 719us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 202.0417 - 2ms/epoch - 25us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0684 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.4480 - 609us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.9596 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.1837 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.3082 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.4149 - 901us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.4333 - 761us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2527 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.0021 - 728us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.6174 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.9978 - 704us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.2408 - 686us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.7104 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.0858 - 584us/epoch - 9us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4878 - 603us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.5732 - 690us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.9006 - 615us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.9738 - 5ms/epoch - 74us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.2800 - 600us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235.0412 - 718us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.5780 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.2808 - 749us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.6849 - 677us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2186 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.8387 - 668us/epoch - 11us/sample\n", + "Generation 228/300\n", + "Solving for Nash Equilibrium in Generation 228/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.8356 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.0685 - 744us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.4564 - 605us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5937 - 834us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.9158 - 739us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.0430 - 726us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.0141 - 849us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4682 - 810us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.5524 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.2631 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7998 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.4038 - 817us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.0708 - 731us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.7590 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.9083 - 1ms/epoch - 20us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4088 - 688us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9778 - 684us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2623 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3599 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.4100 - 1ms/epoch - 23us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3051 - 589us/epoch - 9us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.3829 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.7712 - 914us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.2118 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.6116 - 688us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.3016 - 680us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.0273 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.4008 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.9460 - 713us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.1655 - 8ms/epoch - 129us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.3332 - 735us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.5191 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.0414 - 833us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.8397 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.8624 - 988us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.2561 - 1ms/epoch - 21us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7924 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.5150 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.0328 - 2ms/epoch - 35us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.1953 - 770us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.1626 - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3910 - 684us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.1058 - 714us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.2831 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.5354 - 586us/epoch - 9us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.9544 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.1531 - 836us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.7233 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.6650 - 544us/epoch - 9us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9057 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.7376 - 751us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.4543 - 734us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8118 - 904us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.6774 - 692us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.0405 - 636us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.8812 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3952 - 688us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3715 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.6534 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.4003 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.6655 - 2ms/epoch - 26us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.0638 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.4720 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.4432 - 1ms/epoch - 21us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.9933 - 577us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9136 - 861us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2989 - 851us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4246 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.1940 - 732us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.6526 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.6074 - 754us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.3080 - 686us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.8859 - 699us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.7531 - 6ms/epoch - 93us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.0262 - 789us/epoch - 13us/sample\n", + "Generation 229/300\n", + "Solving for Nash Equilibrium in Generation 229/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.3849 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.9977 - 763us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.8192 - 826us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.5493 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.0053 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.4237 - 592us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.0233 - 625us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.4979 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.4613 - 668us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.2090 - 703us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.8953 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428.5670 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.3331 - 780us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.9200 - 757us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.9643 - 743us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.2376 - 719us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.7792 - 717us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.2576 - 890us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.8636 - 741us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.7761 - 787us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 73.0120 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.3809 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6100 - 569us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.5430 - 6ms/epoch - 90us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.9420 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.3988 - 824us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.3033 - 828us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.6151 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2726 - 2ms/epoch - 36us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.2621 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.4776 - 768us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1053 - 729us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.6095 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.8120 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6301 - 643us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.1885 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.4482 - 652us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3761 - 600us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.2841 - 855us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.3245 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.9883 - 753us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.9848 - 967us/epoch - 16us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.1363 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.3353 - 872us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738.5806 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4099 - 603us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.0854 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.0984 - 570us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230.2642 - 695us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7129 - 713us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.6465 - 697us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5895 - 5ms/epoch - 89us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.0527 - 982us/epoch - 16us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.1018 - 6ms/epoch - 99us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.9223 - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.8361 - 707us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.4445 - 831us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.2664 - 810us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.0025 - 1ms/epoch - 22us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.1952 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.1012 - 663us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.9056 - 698us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.5368 - 776us/epoch - 13us/sample\n", + "Generation 230/300\n", + "Solving for Nash Equilibrium in Generation 230/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.3174 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.6496 - 1ms/epoch - 17us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.1273 - 869us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 113.8790 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8121 - 787us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.9242 - 3ms/epoch - 44us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.7914 - 691us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.3129 - 551us/epoch - 9us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.4069 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2433 - 1ms/epoch - 16us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.5451 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.1590 - 2ms/epoch - 29us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.0756 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7188 - 753us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.8758 - 495us/epoch - 8us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2937 - 689us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.4287 - 1ms/epoch - 18us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.5328 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.8784 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.3546 - 762us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.2874 - 2ms/epoch - 28us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.2987 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232.8616 - 743us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.4708 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.0726 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.0443 - 768us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.2319 - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.5775 - 2ms/epoch - 39us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.8500 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3992 - 736us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.7553 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.3284 - 2ms/epoch - 33us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.5574 - 850us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.6743 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9457 - 721us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.8106 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.7383 - 624us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.8926 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.1121 - 743us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.2383 - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.5399 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5154 - 783us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.2621 - 975us/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.3701 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8178 - 725us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.8004 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.3030 - 805us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.2586 - 610us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5282 - 2ms/epoch - 29us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.9452 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.9660 - 2ms/epoch - 26us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.9662 - 624us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.3278 - 578us/epoch - 9us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.9264 - 2ms/epoch - 39us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.5422 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.5047 - 647us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.2118 - 645us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.5250 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1924 - 786us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.0747 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.1632 - 739us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.6576 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.3709 - 661us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.8227 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.6060 - 727us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.7906 - 763us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.0748 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.7815 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.9137 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8708 - 695us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.4581 - 823us/epoch - 13us/sample\n", + "Generation 231/300\n", + "Solving for Nash Equilibrium in Generation 231/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.3575 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.9444 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.9466 - 750us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6722 - 721us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.3608 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.6173 - 824us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.3161 - 768us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.5193 - 640us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.4864 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.2202 - 1ms/epoch - 18us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.8529 - 733us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.9200 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.8198 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.9917 - 857us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.8085 - 865us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7142 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.7062 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.7863 - 947us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.8733 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.7479 - 1ms/epoch - 17us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.1531 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.0114 - 810us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8292 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.8775 - 2ms/epoch - 25us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.7268 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.3631 - 725us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.4387 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.4026 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.1268 - 812us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.9128 - 780us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.1036 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.3931 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.5403 - 629us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.2644 - 966us/epoch - 16us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.3492 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.0188 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6184 - 713us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.8904 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.9174 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.6707 - 669us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.1727 - 2ms/epoch - 31us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.5154 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.0803 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.7108 - 631us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.4142 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.4758 - 720us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0046 - 790us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.5540 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.0923 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.9391 - 793us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.1807 - 852us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.8071 - 891us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.7963 - 916us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.9904 - 824us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.1737 - 792us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.4665 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.7246 - 724us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.2147 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.6304 - 760us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5466 - 698us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.6866 - 727us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.6126 - 698us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.8376 - 809us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.3200 - 767us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.3777 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.1246 - 919us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.9683 - 1ms/epoch - 23us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.4539 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.2907 - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.6050 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.0062 - 650us/epoch - 10us/sample\n", + "Generation 232/300\n", + "Solving for Nash Equilibrium in Generation 232/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.0319 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8707 - 909us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.2722 - 2ms/epoch - 29us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.0614 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0378 - 709us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.3036 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3970 - 611us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.6988 - 787us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.5076 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.6531 - 705us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.0399 - 825us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.3704 - 697us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.1462 - 706us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.8939 - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.1793 - 1ms/epoch - 21us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.7194 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.7048 - 737us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.2082 - 11ms/epoch - 183us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.2453 - 660us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.3021 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.7400 - 664us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.7628 - 705us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406.5177 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.8551 - 616us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.9310 - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.6265 - 813us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.0082 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.2253 - 951us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.2460 - 641us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.4860 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.7420 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.4570 - 787us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.7489 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.6430 - 609us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.7073 - 805us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.0080 - 916us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9957 - 586us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.7364 - 819us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.7606 - 937us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.2172 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.0500 - 683us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.9940 - 732us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.7455 - 1ms/epoch - 20us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.0605 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.3082 - 744us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8851 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.1321 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.5842 - 672us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.8828 - 754us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.6371 - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.9432 - 797us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.5524 - 905us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.2840 - 3ms/epoch - 48us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.2453 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.4657 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.2408 - 799us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.0318 - 699us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.7914 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.2314 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.3682 - 816us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Generation 233/300\n", + "Solving for Nash Equilibrium in Generation 233/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6258 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7282 - 714us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.0487 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.2802 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.6736 - 1ms/epoch - 19us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.7249 - 1ms/epoch - 18us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.8334 - 821us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.2711 - 909us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.2833 - 761us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.6378 - 1ms/epoch - 22us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.5899 - 1ms/epoch - 22us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.3279 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5891 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.6258 - 1ms/epoch - 17us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.3141 - 7ms/epoch - 120us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.5236 - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.7127 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.0977 - 1ms/epoch - 18us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.6731 - 776us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6825 - 873us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.8153 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.0634 - 738us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.8900 - 691us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.0151 - 1ms/epoch - 19us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.1834 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.3691 - 853us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.1773 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.2660 - 701us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.7677 - 1ms/epoch - 23us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.4272 - 784us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.6722 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5951 - 625us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.9031 - 594us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.2211 - 1ms/epoch - 17us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1730 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.0119 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.2853 - 873us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.5072 - 791us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.0869 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.0529 - 859us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6496 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.4555 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8847 - 863us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.6706 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.5735 - 759us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.2615 - 696us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.1227 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4592 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.3592 - 5ms/epoch - 74us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.2098 - 618us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.2013 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5417 - 742us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.9006 - 953us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.5884 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7260 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.8538 - 572us/epoch - 9us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.7328 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.7245 - 756us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.3953 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6822 - 725us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.7722 - 616us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.1874 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.0947 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.0754 - 934us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.1100 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1887 - 673us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.2133 - 744us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.9240 - 829us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.1151 - 732us/epoch - 12us/sample\n", + "Generation 234/300\n", + "Solving for Nash Equilibrium in Generation 234/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.6715 - 1ms/epoch - 19us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6385 - 599us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.3661 - 577us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.8967 - 891us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.4925 - 603us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.1774 - 900us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.4268 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.6092 - 918us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.7432 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.7621 - 596us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.1303 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.2764 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.5063 - 823us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.3340 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3754 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.5961 - 2ms/epoch - 30us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.0079 - 643us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.3577 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.2128 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.1855 - 801us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8161 - 701us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.8965 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.3655 - 676us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.3481 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.3668 - 822us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.6868 - 687us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.2259 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.7843 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3162 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.4415 - 837us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0126 - 660us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.9417 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.7606 - 755us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4337 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.3475 - 634us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.9073 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.5724 - 639us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.7390 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.8956 - 681us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.7734 - 662us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.2745 - 848us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.2399 - 2ms/epoch - 27us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.4402 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2845 - 712us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7154 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.4801 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.5036 - 686us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.0877 - 1ms/epoch - 20us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9079 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.9279 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.5721 - 823us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.1082 - 678us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.0372 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.6852 - 670us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.0262 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.1879 - 865us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.0808 - 1ms/epoch - 24us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.4350 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.8855 - 599us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.6536 - 1ms/epoch - 20us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.0484 - 644us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.2247 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763.7011 - 639us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 191.5304 - 695us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.0998 - 693us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.7865 - 927us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.6155 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.8490 - 654us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.6443 - 1ms/epoch - 17us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3521 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 209.1691 - 703us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.5974 - 725us/epoch - 12us/sample\n", + "Generation 235/300\n", + "Solving for Nash Equilibrium in Generation 235/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.1291 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.6591 - 749us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.2140 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.0255 - 811us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.6644 - 670us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.8072 - 696us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.6929 - 2ms/epoch - 39us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.9233 - 2ms/epoch - 31us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 161.6835 - 754us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.4081 - 3ms/epoch - 47us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.0842 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.5359 - 861us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.3809 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.3140 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 165us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.1934 - 5ms/epoch - 86us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.5028 - 729us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9829 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7057 - 3ms/epoch - 56us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.8107 - 878us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8360 - 1ms/epoch - 20us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.0518 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.8450 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2480 - 1ms/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232.6385 - 711us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.5917 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.7713 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.4280 - 680us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 139us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.6907 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3207 - 835us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.6906 - 4ms/epoch - 65us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4503 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.7573 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4568 - 893us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1175 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.2088 - 571us/epoch - 9us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9197 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.7838 - 832us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3109 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7595 - 660us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4194 - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.6873 - 691us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6012 - 2ms/epoch - 40us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2707 - 2ms/epoch - 24us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.7567 - 684us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.3174 - 10ms/epoch - 154us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 123us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.3635 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4335 - 707us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.6209 - 623us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.2112 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8765 - 5ms/epoch - 86us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.3740 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.6971 - 944us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.1766 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.1130 - 3ms/epoch - 47us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7547 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.7109 - 607us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.9865 - 885us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.0110 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.5692 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.1393 - 711us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.4752 - 643us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.8206 - 686us/epoch - 11us/sample\n", + "Generation 236/300\n", + "Solving for Nash Equilibrium in Generation 236/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2191 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41.1982 - 816us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.9669 - 636us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.4479 - 1ms/epoch - 20us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.0807 - 767us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.2157 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.3916 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1626 - 695us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.0815 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 253.0822 - 747us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.5885 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.0625 - 770us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1042 - 1ms/epoch - 18us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.1895 - 822us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.4370 - 1ms/epoch - 21us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9584 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.5195 - 585us/epoch - 9us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.1840 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.6629 - 723us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.0558 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8013 - 748us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.6752 - 542us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9787 - 2ms/epoch - 30us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.5223 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.4356 - 750us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.3924 - 828us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.4989 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.9800 - 785us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.1004 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.3073 - 762us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.6205 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.8781 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.8176 - 807us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0625 - 706us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.8518 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.2811 - 1ms/epoch - 23us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2828 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.5510 - 598us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.8187 - 611us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.4044 - 773us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.7676 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.5509 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1387 - 1ms/epoch - 17us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.4122 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.7800 - 1ms/epoch - 18us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.6249 - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.1336 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.7209 - 807us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.9691 - 556us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.2123 - 851us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.8672 - 739us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.5199 - 2ms/epoch - 31us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.6647 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.6924 - 715us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.7200 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.1544 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.0769 - 2ms/epoch - 28us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.0269 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3385 - 657us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.8052 - 7ms/epoch - 117us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6772 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.2179 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.3352 - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.4822 - 1ms/epoch - 17us/sample\n", + "Generation 237/300\n", + "Solving for Nash Equilibrium in Generation 237/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8630 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.6648 - 653us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.7838 - 919us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.5330 - 9ms/epoch - 143us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.0654 - 729us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43.9076 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.1698 - 1ms/epoch - 23us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.9337 - 779us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.9170 - 598us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.2838 - 721us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.5854 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.3236 - 5ms/epoch - 82us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116.5159 - 650us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.8674 - 746us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.2189 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5482 - 1ms/epoch - 16us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.4189 - 938us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.1873 - 586us/epoch - 9us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.2446 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.4321 - 653us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361.1407 - 7ms/epoch - 106us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.4442 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.0264 - 835us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.6592 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.5193 - 589us/epoch - 9us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.2079 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.3307 - 799us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.9839 - 645us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.1425 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9953 - 1ms/epoch - 17us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.1321 - 688us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.0054 - 717us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.8507 - 3ms/epoch - 40us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.6236 - 693us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 78.0856 - 622us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.3093 - 722us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.9106 - 686us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.3918 - 603us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 67.5230 - 577us/epoch - 9us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.2847 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.5311 - 638us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.1131 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.5579 - 698us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4739 - 752us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.2520 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6707 - 2ms/epoch - 31us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.7359 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.4250 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.9860 - 853us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.2585 - 799us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.0875 - 742us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 238/300\n", + "Solving for Nash Equilibrium in Generation 238/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.1899 - 713us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8641 - 4ms/epoch - 64us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.6049 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.8092 - 901us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.3216 - 699us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.3324 - 753us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.0214 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.3904 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.1156 - 662us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6296 - 641us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.7401 - 707us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.8596 - 802us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.6249 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.0090 - 643us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.0178 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.5956 - 648us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.9076 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.8007 - 765us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.5658 - 593us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.3209 - 839us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.5666 - 939us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.0283 - 1ms/epoch - 20us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.5712 - 590us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.7621 - 894us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.9240 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.4456 - 769us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.2220 - 759us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.8068 - 1ms/epoch - 21us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.0405 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.1730 - 648us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 177.1278 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0004 - 665us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.4656 - 779us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.9960 - 3ms/epoch - 45us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.3208 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.5265 - 625us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.4701 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.7384 - 651us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.0605 - 793us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.7289 - 2ms/epoch - 27us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.3459 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.3578 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.2504 - 783us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.0981 - 726us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.7357 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.0589 - 715us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8875 - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9720 - 2ms/epoch - 28us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.1251 - 725us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.5686 - 745us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.3606 - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.2822 - 1ms/epoch - 19us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2752 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.7086 - 675us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.3941 - 642us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.5496 - 782us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.2752 - 935us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.8485 - 3ms/epoch - 41us/sample\n", + "Episode 50/50\n", + "Generation 239/300\n", + "Solving for Nash Equilibrium in Generation 239/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.4462 - 593us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.9710 - 1ms/epoch - 21us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.4373 - 568us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235.5947 - 1ms/epoch - 24us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5851 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.5479 - 741us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.9768 - 723us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.9868 - 682us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.6898 - 785us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.5540 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.5827 - 1ms/epoch - 22us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.1910 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.9977 - 793us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.7478 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0632 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.4814 - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.6352 - 1ms/epoch - 16us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.8883 - 729us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.0023 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6528 - 2ms/epoch - 27us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.4714 - 703us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.1270 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.0906 - 1ms/epoch - 16us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.3094 - 874us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.7882 - 2ms/epoch - 25us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.4327 - 1ms/epoch - 16us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.2891 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.7009 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.4099 - 1ms/epoch - 17us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45.7589 - 794us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6777 - 751us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.2590 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.9169 - 901us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0042 - 658us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.0458 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.8986 - 4ms/epoch - 71us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.5248 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.5171 - 636us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.9084 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9161 - 5ms/epoch - 83us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.6376 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.8768 - 930us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.7827 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.4027 - 11ms/epoch - 175us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3251 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5452 - 678us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.9456 - 1ms/epoch - 18us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.0900 - 1ms/epoch - 18us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.3700 - 791us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.3884 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.6136 - 808us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.3225 - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 202.5912 - 6ms/epoch - 99us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.0437 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.5351 - 976us/epoch - 16us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.1814 - 936us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.2576 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.8318 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0678 - 624us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6658 - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2795 - 897us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.5208 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.1463 - 598us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.9698 - 593us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.0983 - 665us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7927 - 686us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2140 - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.6593 - 772us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6008 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.9116 - 879us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.5298 - 626us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.9136 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.2079 - 680us/epoch - 11us/sample\n", + "Generation 240/300\n", + "Solving for Nash Equilibrium in Generation 240/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.7404 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.1591 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.5809 - 613us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.8759 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.8890 - 592us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0042 - 813us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.2846 - 664us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.0887 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.5899 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7178 - 2ms/epoch - 27us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.0931 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.5257 - 774us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.0763 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8223 - 571us/epoch - 9us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.5201 - 713us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.7801 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.7913 - 727us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.3197 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 198.8990 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.5096 - 653us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.4655 - 741us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.9276 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.9496 - 618us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.4615 - 867us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.9247 - 796us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.0632 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.2981 - 636us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.3646 - 7ms/epoch - 119us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.2840 - 709us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2564 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.3312 - 788us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.1550 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.5228 - 597us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.7326 - 1ms/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.3240 - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2748 - 657us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.4545 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.4734 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.3116 - 714us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.6231 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.4667 - 683us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.3516 - 891us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4495 - 864us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.2801 - 816us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.2369 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.6387 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.3040 - 842us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 213.4010 - 725us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.7544 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.5426 - 707us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.3352 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.8937 - 623us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4187 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0448 - 681us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5623 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.1799 - 788us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.5840 - 908us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.4233 - 953us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6339 - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.8870 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.3729 - 775us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 435.3219 - 2ms/epoch - 26us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.8342 - 2ms/epoch - 37us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.9448 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.1371 - 631us/epoch - 10us/sample\n", + "Generation 241/300\n", + "Solving for Nash Equilibrium in Generation 241/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.1452 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.2964 - 569us/epoch - 9us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.5689 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4363 - 735us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.7347 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.8763 - 861us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41.4432 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.4616 - 741us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.7328 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5930 - 1ms/epoch - 21us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.4839 - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.5786 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.5383 - 805us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1422 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.2841 - 867us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 17ms/epoch - 281us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.3819 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.5027 - 1ms/epoch - 23us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.0483 - 1ms/epoch - 17us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.6010 - 1ms/epoch - 16us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.5926 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.6052 - 777us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.7787 - 798us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.3336 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.3850 - 17ms/epoch - 270us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.3913 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.0057 - 1ms/epoch - 18us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47.2147 - 918us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.9196 - 1ms/epoch - 17us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.7429 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.4513 - 910us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.8078 - 816us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.3615 - 878us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 152us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.5287 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.2726 - 28ms/epoch - 446us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.5164 - 598us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.0860 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.4867 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1945 - 1ms/epoch - 22us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6397 - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.3415 - 732us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.0518 - 3ms/epoch - 47us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7780 - 646us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3350 - 686us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.0708 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.8783 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 31ms/epoch - 499us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.9373 - 4ms/epoch - 62us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.6694 - 783us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.2328 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.0591 - 736us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.5323 - 2ms/epoch - 24us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.8392 - 649us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.1313 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.2177 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.6915 - 866us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.1420 - 921us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.4857 - 821us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.1868 - 14ms/epoch - 224us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.3378 - 891us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.5477 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.1602 - 995us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.3335 - 915us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248.8050 - 1ms/epoch - 16us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345.5826 - 798us/epoch - 13us/sample\n", + "Generation 242/300\n", + "Solving for Nash Equilibrium in Generation 242/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.0828 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.8723 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.3674 - 793us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.9272 - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6221 - 888us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.2244 - 692us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.3862 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6653 - 862us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325.8942 - 783us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.3174 - 659us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.3853 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.2168 - 1ms/epoch - 17us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.6888 - 820us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.7927 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.1993 - 684us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.1502 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.3220 - 2ms/epoch - 35us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.2368 - 811us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.2770 - 716us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.3076 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.1364 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.2117 - 19ms/epoch - 304us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.1037 - 836us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.9335 - 796us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8422 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.1709 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.4606 - 759us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8205 - 3ms/epoch - 53us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.4909 - 6ms/epoch - 101us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.9916 - 10ms/epoch - 162us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.7077 - 3ms/epoch - 43us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.4332 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.2819 - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3706 - 1ms/epoch - 17us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.5061 - 941us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.2071 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.9982 - 822us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.2583 - 804us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.7488 - 900us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.4178 - 785us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.9955 - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6733 - 730us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.6533 - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.1015 - 766us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.2336 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3641 - 1ms/epoch - 20us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.0463 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9711 - 672us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.8126 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.0960 - 930us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.5846 - 707us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.9405 - 773us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.5049 - 545us/epoch - 9us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.4515 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6903 - 888us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.0650 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.8469 - 748us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.2465 - 696us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4883 - 968us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Generation 243/300\n", + "Solving for Nash Equilibrium in Generation 243/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.0793 - 935us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.0211 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.6854 - 7ms/epoch - 108us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.8273 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.1997 - 744us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.3705 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.8910 - 767us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.5522 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.2651 - 652us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.3153 - 990us/epoch - 16us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.1820 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.0460 - 750us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0029 - 1ms/epoch - 21us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6956 - 622us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.3167 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.0445 - 846us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.3972 - 595us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.2859 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.2028 - 1ms/epoch - 18us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.2459 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.8512 - 1ms/epoch - 20us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.2360 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 12ms/epoch - 190us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6248 - 3ms/epoch - 54us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.2866 - 883us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 16ms/epoch - 250us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.7509 - 14ms/epoch - 219us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.7097 - 673us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.2116 - 772us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.2866 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.5103 - 3ms/epoch - 49us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8496 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.3338 - 2ms/epoch - 33us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.1625 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7411 - 637us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.5382 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9304 - 650us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.2275 - 2ms/epoch - 36us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2272 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.9365 - 3ms/epoch - 48us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.2988 - 824us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5958 - 620us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.9468 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.8158 - 1ms/epoch - 18us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3047 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.1323 - 571us/epoch - 9us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.4534 - 806us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.9927 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2740 - 620us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2724 - 808us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.8270 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.6977 - 660us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7480 - 6ms/epoch - 103us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 198.2825 - 773us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.0250 - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.9172 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.9323 - 736us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.9987 - 695us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.4113 - 941us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.1614 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.0205 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.3405 - 750us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.1277 - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.2809 - 2ms/epoch - 28us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.1563 - 740us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.5161 - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6587 - 640us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.1437 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.5249 - 661us/epoch - 11us/sample\n", + "Generation 244/300\n", + "Solving for Nash Equilibrium in Generation 244/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5369 - 588us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9810 - 640us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4080 - 665us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.2030 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7175 - 2ms/epoch - 38us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.9459 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391.8500 - 754us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2421 - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.7847 - 692us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.2487 - 706us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3058 - 697us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.6839 - 7ms/epoch - 114us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.7125 - 709us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.1691 - 818us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.0486 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.8477 - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.7289 - 769us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.9156 - 713us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.8606 - 815us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.1336 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.2934 - 744us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5953 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.6780 - 766us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.0852 - 1ms/epoch - 18us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.1712 - 738us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.1543 - 925us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.2548 - 907us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.7280 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79.8713 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.1700 - 4ms/epoch - 68us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.3885 - 740us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.8693 - 734us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.1683 - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.0331 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.6633 - 707us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.3795 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.5895 - 785us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.6398 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.9928 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.1520 - 831us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.0905 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742.7753 - 666us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.3331 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4785 - 1ms/epoch - 20us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.6953 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.8416 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.5388 - 678us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.6718 - 4ms/epoch - 72us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.3432 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.6269 - 592us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.0928 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.2217 - 776us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.6823 - 798us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.0548 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298.3696 - 704us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6450 - 859us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.5834 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.7775 - 1ms/epoch - 19us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.7045 - 748us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.0406 - 2ms/epoch - 28us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 157us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.3962 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.6038 - 969us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9183 - 1ms/epoch - 22us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.6403 - 655us/epoch - 11us/sample\n", + "Generation 245/300\n", + "Solving for Nash Equilibrium in Generation 245/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.2137 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.5085 - 744us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8776 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.5664 - 787us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2936 - 754us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.4221 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.7532 - 778us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.2706 - 2ms/epoch - 36us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2883 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235.8193 - 606us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.2722 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.2363 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2432 - 1ms/epoch - 19us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.3052 - 803us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.8287 - 651us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.3632 - 871us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.8166 - 2ms/epoch - 28us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 253.3056 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.1006 - 6ms/epoch - 98us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.0848 - 741us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45.5136 - 5ms/epoch - 85us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.5410 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2856 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.7156 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.8872 - 851us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.0594 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.6564 - 751us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.3600 - 704us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.3569 - 1ms/epoch - 23us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.7422 - 904us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.8571 - 695us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.5842 - 860us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.3630 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.3260 - 684us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6001 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.2258 - 663us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.6762 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8033 - 646us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.6751 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.0863 - 709us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.0604 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.2448 - 2ms/epoch - 27us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.0151 - 2ms/epoch - 25us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.9758 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5002 - 843us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.3440 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.8973 - 2ms/epoch - 25us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.1567 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.0096 - 2ms/epoch - 24us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1063 - 679us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.7390 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.3841 - 818us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8828 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.3506 - 657us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.9275 - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.9933 - 3ms/epoch - 52us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.5412 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1561 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1773 - 946us/epoch - 15us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.6897 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.4081 - 665us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8788 - 892us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.9886 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4096 - 772us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7318 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.6766 - 6ms/epoch - 91us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.4861 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6063 - 717us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.8798 - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.5391 - 685us/epoch - 11us/sample\n", + "Generation 246/300\n", + "Solving for Nash Equilibrium in Generation 246/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.4969 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.9902 - 644us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.6538 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.8268 - 701us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.0722 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.7767 - 698us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.4165 - 680us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557.6280 - 669us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.7576 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.4991 - 816us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.7657 - 817us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2466 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.6072 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.6726 - 4ms/epoch - 62us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.5417 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.9050 - 665us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5037 - 694us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6533 - 969us/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.2643 - 3ms/epoch - 45us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.7672 - 762us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3912 - 776us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.2399 - 836us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.7839 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.8134 - 509us/epoch - 8us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.9395 - 747us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.0773 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.9826 - 573us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.0540 - 954us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.9281 - 2ms/epoch - 40us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.9711 - 729us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.9371 - 2ms/epoch - 27us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.0878 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 13ms/epoch - 214us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.0825 - 2ms/epoch - 31us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.8694 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.6911 - 651us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7056 - 731us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.1913 - 1ms/epoch - 17us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.9478 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.6127 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.3218 - 865us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.5836 - 761us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.3298 - 609us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.8187 - 816us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.7450 - 3ms/epoch - 49us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9431 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7173 - 671us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.7076 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.7984 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4945 - 756us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.0244 - 1ms/epoch - 17us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.6758 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6259 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0837 - 995us/epoch - 16us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.1223 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.9306 - 2ms/epoch - 34us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.3482 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.8712 - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.3398 - 578us/epoch - 9us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.2713 - 843us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.2030 - 656us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.8908 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6411 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.1347 - 2ms/epoch - 39us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.7369 - 669us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.2001 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.2921 - 687us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.3802 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.1490 - 876us/epoch - 14us/sample\n", + "Generation 247/300\n", + "Solving for Nash Equilibrium in Generation 247/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.5238 - 705us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 93us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.3339 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.4785 - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.3039 - 2ms/epoch - 27us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.4142 - 857us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.7415 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.8199 - 1ms/epoch - 23us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.5836 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1448 - 678us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.9521 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.1875 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.7372 - 593us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.6149 - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8542 - 722us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.1434 - 629us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5367 - 737us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.6330 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2872 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.7246 - 748us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.9886 - 791us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1620 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.4585 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.4708 - 827us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.4951 - 693us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.2463 - 588us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.3613 - 771us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.3979 - 808us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.6830 - 730us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.1350 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.9256 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.6432 - 726us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.6360 - 734us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.2177 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.7364 - 667us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.2946 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.5031 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.6682 - 612us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.6073 - 3ms/epoch - 44us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.4187 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.8557 - 717us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.5998 - 698us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.4407 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.1595 - 664us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.0354 - 716us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.6912 - 760us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.4622 - 1ms/epoch - 17us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.0084 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.6870 - 654us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.9796 - 939us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.7829 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.0009 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.0269 - 772us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.5737 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.6395 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7319 - 767us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.9713 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.5015 - 707us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.1769 - 875us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7250 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.7870 - 1ms/epoch - 17us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.6618 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.4222 - 723us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.0321 - 702us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.7120 - 785us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.7805 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.7789 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245.4488 - 1ms/epoch - 19us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2272 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2352 - 781us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4570 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.0182 - 673us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4433 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5720 - 644us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.1331 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5905 - 664us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 248/300\n", + "Solving for Nash Equilibrium in Generation 248/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.7368 - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.3881 - 698us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.3411 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.2249 - 697us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.2533 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.1782 - 1ms/epoch - 18us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.0457 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9862 - 689us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.7607 - 8ms/epoch - 125us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8080 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.9299 - 690us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8466 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.5896 - 653us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.8148 - 649us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8810 - 663us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.5410 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.9886 - 671us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.8083 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.8012 - 741us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.2890 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1552 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.0916 - 2ms/epoch - 30us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.6789 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.1320 - 871us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.3436 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.5325 - 723us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.8709 - 737us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.0707 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771.8162 - 598us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.0604 - 656us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.8286 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.3312 - 1ms/epoch - 17us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.8868 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.5475 - 706us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.1021 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5818 - 580us/epoch - 9us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8469 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.0959 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.3782 - 943us/epoch - 15us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3584 - 746us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.4789 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.4969 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.1822 - 1ms/epoch - 19us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.0137 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.5165 - 1ms/epoch - 20us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.8503 - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.0890 - 14ms/epoch - 229us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.7968 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.9398 - 14ms/epoch - 227us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.8677 - 789us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.5762 - 696us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4426 - 2ms/epoch - 26us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.8173 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.2808 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.6973 - 1ms/epoch - 20us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.7969 - 1ms/epoch - 21us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.8335 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.8540 - 574us/epoch - 9us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.0272 - 701us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.0134 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.4528 - 771us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.0826 - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.4568 - 1ms/epoch - 21us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.4189 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.1685 - 961us/epoch - 16us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3621 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.8022 - 745us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.2899 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0685 - 749us/epoch - 12us/sample\n", + "Generation 249/300\n", + "Solving for Nash Equilibrium in Generation 249/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.7473 - 583us/epoch - 9us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.4275 - 605us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.0902 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.3483 - 722us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.2905 - 915us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.8607 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6954 - 709us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.4908 - 1ms/epoch - 16us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.1145 - 3ms/epoch - 45us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.7353 - 795us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.2033 - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.9474 - 703us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.9971 - 687us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.3452 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764.3792 - 626us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.3885 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.2518 - 679us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.3943 - 6ms/epoch - 96us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.5716 - 779us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.0628 - 875us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.5726 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.3486 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.9557 - 853us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.9067 - 671us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.0589 - 692us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.1701 - 948us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.2779 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.7861 - 663us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.9250 - 585us/epoch - 9us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.8004 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.5929 - 2ms/epoch - 37us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.4201 - 653us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.5450 - 752us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.7106 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.1943 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.3662 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.3975 - 2ms/epoch - 28us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.2097 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.1087 - 659us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.6399 - 724us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.0959 - 791us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.8393 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.2820 - 625us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.9437 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.1492 - 2ms/epoch - 36us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.4994 - 592us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.6111 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.4371 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2733 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.0555 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.8128 - 691us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.9798 - 634us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3777 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.6328 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.0679 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.2191 - 785us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5162 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.6643 - 715us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.8770 - 2ms/epoch - 34us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0198 - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5546 - 797us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.5837 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.1815 - 855us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 17ms/epoch - 278us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.5792 - 1ms/epoch - 19us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 223.7430 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.3059 - 601us/epoch - 10us/sample\n", + "Generation 250/300\n", + "Solving for Nash Equilibrium in Generation 250/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428.9723 - 602us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.1556 - 711us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.9976 - 764us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.1074 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.7115 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.5165 - 737us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.8774 - 615us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.5940 - 1ms/epoch - 20us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.0049 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.6909 - 653us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5447 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.3204 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.7119 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.6613 - 1ms/epoch - 22us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.9816 - 726us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.4592 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302.0621 - 755us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.0323 - 804us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.5781 - 746us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.0878 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.7578 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 14ms/epoch - 232us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.8275 - 991us/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.0930 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9819 - 705us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.3944 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.4814 - 7ms/epoch - 106us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1740 - 821us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.8909 - 10ms/epoch - 162us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.3261 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.0977 - 2ms/epoch - 29us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4072 - 2ms/epoch - 28us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.1758 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.5861 - 1ms/epoch - 19us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.1594 - 839us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8650 - 879us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.7316 - 1ms/epoch - 20us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8802 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.4811 - 1ms/epoch - 20us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4200 - 830us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.7692 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.2101 - 813us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.9473 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8503 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.5258 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.4411 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.7413 - 811us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.3041 - 816us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6369 - 686us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.1483 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.3764 - 2ms/epoch - 25us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.7871 - 772us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.7795 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 416.9670 - 843us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.2388 - 800us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.9521 - 1ms/epoch - 19us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.8427 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.2602 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.6109 - 947us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8150 - 872us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.2349 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.9796 - 2ms/epoch - 34us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.3881 - 750us/epoch - 12us/sample\n", + "Generation 251/300\n", + "Solving for Nash Equilibrium in Generation 251/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.7095 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.2834 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 133us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.4506 - 3ms/epoch - 44us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.0411 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.3840 - 917us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.2868 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.4420 - 1ms/epoch - 21us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.7574 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.4056 - 1ms/epoch - 17us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.1851 - 821us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.4338 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.1075 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.3348 - 3ms/epoch - 43us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9783 - 2ms/epoch - 26us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5802 - 725us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.3290 - 707us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.7214 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.9714 - 751us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6518 - 935us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.0192 - 732us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245.2198 - 781us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3198 - 667us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.9186 - 792us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.6859 - 755us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 181.2470 - 630us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.2865 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.1491 - 780us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.5348 - 617us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.2675 - 767us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.1018 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.8975 - 802us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.5392 - 1ms/epoch - 16us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.8491 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.0886 - 709us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.4839 - 767us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9293 - 1ms/epoch - 21us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2185 - 659us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.8102 - 968us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.4308 - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.1350 - 1ms/epoch - 21us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.8293 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.5911 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.2540 - 609us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.6425 - 676us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.8266 - 650us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.9012 - 847us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.7516 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.4963 - 732us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.6778 - 865us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.7966 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.7635 - 673us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.7809 - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.3224 - 764us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.8498 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4908 - 657us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.6332 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.0233 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9340 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.7160 - 753us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.6476 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.9356 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6703 - 735us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.1814 - 4ms/epoch - 64us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.2116 - 639us/epoch - 10us/sample\n", + "Generation 252/300\n", + "Solving for Nash Equilibrium in Generation 252/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.6537 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.8693 - 650us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.8499 - 4ms/epoch - 58us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.9767 - 764us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.9608 - 1ms/epoch - 16us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.1123 - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.8802 - 2ms/epoch - 25us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.8255 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.2870 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.2936 - 694us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.3778 - 675us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.0530 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.2045 - 964us/epoch - 16us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.8003 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4509 - 862us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.1060 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3287 - 700us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.2876 - 1ms/epoch - 22us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.8636 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 224.0306 - 813us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.5834 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4944 - 742us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38.7742 - 824us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.8170 - 830us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.3342 - 690us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.2795 - 2ms/epoch - 27us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.5067 - 850us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.9169 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.7103 - 767us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.1893 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.7684 - 658us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0629 - 803us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.8953 - 598us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.2284 - 933us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.8483 - 932us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.4557 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.6883 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.1586 - 768us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.7924 - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.2309 - 721us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.6933 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354.3239 - 1ms/epoch - 21us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.9218 - 777us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.3358 - 583us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.5524 - 803us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.3222 - 823us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.4185 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.6992 - 825us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.0533 - 3ms/epoch - 45us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.2346 - 835us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.4228 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.8093 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.4551 - 861us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.5411 - 717us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.5901 - 825us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.3466 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.5554 - 10ms/epoch - 157us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.3239 - 744us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.4149 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.5750 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.5098 - 1ms/epoch - 17us/sample\n", + "Generation 253/300\n", + "Solving for Nash Equilibrium in Generation 253/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.5759 - 645us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.9875 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.0377 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.4711 - 928us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.4559 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.3893 - 741us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.3434 - 3ms/epoch - 41us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7480 - 864us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.3285 - 668us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.7388 - 855us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.8141 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240.1512 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.6038 - 3ms/epoch - 43us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.3010 - 634us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.4981 - 901us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.8392 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.9470 - 580us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2562 - 854us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3370 - 637us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.6527 - 749us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.7762 - 2ms/epoch - 32us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.4862 - 969us/epoch - 16us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.1856 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.2310 - 754us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.0201 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.9916 - 834us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.7969 - 755us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.3037 - 790us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.0557 - 2ms/epoch - 38us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.1982 - 1ms/epoch - 18us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3792 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.4691 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.3529 - 4ms/epoch - 62us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.5148 - 891us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.8693 - 1ms/epoch - 17us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.9628 - 9ms/epoch - 144us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.6939 - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4050 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.2690 - 793us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.9310 - 890us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.3394 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.7352 - 8ms/epoch - 122us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.9351 - 2ms/epoch - 31us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.8226 - 698us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.2083 - 925us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.7938 - 819us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.0935 - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 14ms/epoch - 223us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.8491 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.2782 - 711us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.9715 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.1242 - 701us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.3036 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.4813 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2910 - 819us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.0788 - 738us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0432 - 901us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.6389 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.0974 - 2ms/epoch - 35us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.4796 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.5466 - 691us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.6418 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.6935 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276.7472 - 2ms/epoch - 28us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1709 - 726us/epoch - 12us/sample\n", + "Generation 254/300\n", + "Solving for Nash Equilibrium in Generation 254/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.6554 - 822us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7927 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.7867 - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.6823 - 721us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.3170 - 742us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.6604 - 768us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.8317 - 616us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.9756 - 2ms/epoch - 36us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.1170 - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4760 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.3201 - 758us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.9216 - 721us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4911 - 3ms/epoch - 41us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.6061 - 632us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.6830 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.1897 - 732us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.9136 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9900 - 5ms/epoch - 86us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.1948 - 951us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.5249 - 1ms/epoch - 17us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.4749 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6016 - 2ms/epoch - 30us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5669 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.7582 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.2113 - 4ms/epoch - 71us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.5657 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5155 - 3ms/epoch - 42us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.2177 - 729us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.5673 - 956us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.4332 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.6534 - 21ms/epoch - 345us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2855 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.6242 - 698us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.6629 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.9190 - 818us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.6898 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.0324 - 611us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.1920 - 690us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.3989 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.5359 - 933us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.4498 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.5255 - 842us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.2622 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.0833 - 597us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.6060 - 684us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.8725 - 647us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.2122 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.9630 - 632us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.1962 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.7911 - 626us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.0570 - 659us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.4551 - 626us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.7245 - 716us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.6071 - 708us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.0131 - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0712 - 666us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.3792 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.1349 - 747us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.4603 - 1ms/epoch - 20us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41.1252 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.8303 - 807us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.1193 - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.2294 - 960us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7134 - 1ms/epoch - 23us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.6024 - 869us/epoch - 14us/sample\n", + "Generation 255/300\n", + "Solving for Nash Equilibrium in Generation 255/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.6530 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.7956 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201.7095 - 1ms/epoch - 18us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.9658 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.2364 - 771us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.3063 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.3439 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.8318 - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.5968 - 795us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.1465 - 768us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.3290 - 754us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 143us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9406 - 1ms/epoch - 21us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.1163 - 613us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.4879 - 709us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.1897 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.0684 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.3664 - 695us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.5516 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2931 - 648us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.7970 - 602us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6548 - 553us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.6276 - 728us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.4389 - 2ms/epoch - 37us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.3617 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.9965 - 714us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.9680 - 645us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.2993 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.5134 - 789us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.7704 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.9062 - 626us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.6513 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3271 - 1ms/epoch - 21us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.0677 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8143 - 708us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.3636 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.7477 - 689us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.0978 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.5024 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.2214 - 622us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.5140 - 625us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.1659 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8009 - 607us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9991 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.6810 - 661us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.2083 - 1ms/epoch - 17us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.7661 - 684us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.3425 - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.9064 - 574us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.6479 - 790us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.6660 - 791us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.6711 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.1351 - 791us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.8106 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.5782 - 731us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.2445 - 734us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.7951 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.1802 - 667us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4897 - 721us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.6790 - 710us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.9426 - 711us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.2752 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.3097 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.2529 - 751us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.4178 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.8957 - 757us/epoch - 12us/sample\n", + "Generation 256/300\n", + "Solving for Nash Equilibrium in Generation 256/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.5967 - 695us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6112 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.6694 - 667us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 151us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.6094 - 927us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.4605 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.5921 - 2ms/epoch - 36us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.2664 - 712us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.3455 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2582 - 4ms/epoch - 56us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.5774 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.0748 - 613us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 155us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1564 - 1ms/epoch - 19us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.1802 - 4ms/epoch - 68us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.7675 - 636us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.7410 - 2ms/epoch - 39us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.6811 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.8146 - 626us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.4676 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.3997 - 683us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.0911 - 717us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.6775 - 614us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.6904 - 697us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.0432 - 710us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.2368 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.2623 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.7352 - 682us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.0301 - 795us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.3767 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4482 - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.4567 - 988us/epoch - 16us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.7135 - 1ms/epoch - 23us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.3529 - 859us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.1284 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.9705 - 2ms/epoch - 35us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.0985 - 1ms/epoch - 16us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.9024 - 777us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.6077 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.7107 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.4959 - 735us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.0419 - 4ms/epoch - 71us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.7871 - 6ms/epoch - 99us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.4949 - 2ms/epoch - 27us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.6443 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.1434 - 709us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.7780 - 723us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.3893 - 919us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.7243 - 820us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.0918 - 653us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.7568 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.9135 - 710us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.6064 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.7607 - 632us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.7383 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.3260 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.0657 - 1ms/epoch - 19us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7529 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9196 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.4853 - 580us/epoch - 9us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.1160 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.3181 - 688us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.0657 - 732us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.1362 - 707us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.9453 - 822us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.2706 - 686us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.8922 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.1185 - 2ms/epoch - 24us/sample\n", + "Generation 257/300\n", + "Solving for Nash Equilibrium in Generation 257/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.8531 - 773us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.0638 - 707us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.1789 - 950us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.1000 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.9164 - 708us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 87.8421 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.1370 - 525us/epoch - 8us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.5616 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6503 - 695us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.8189 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.8127 - 835us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.9535 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2537 - 720us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.9668 - 799us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7306 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.8080 - 713us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.1855 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.3729 - 712us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.1863 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.1090 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.6829 - 821us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.4539 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.0325 - 647us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.3046 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.5385 - 1ms/epoch - 20us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.5167 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.4185 - 587us/epoch - 9us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.0664 - 841us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.9971 - 770us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.1130 - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.4942 - 738us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.0828 - 709us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.9912 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.2605 - 908us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.4297 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.4362 - 792us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.9924 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.0992 - 684us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.7343 - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.1909 - 819us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.1122 - 917us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.2599 - 612us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.2540 - 642us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.2492 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 137us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.7188 - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.2901 - 709us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201.0032 - 730us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 169us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2731 - 2ms/epoch - 30us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.7058 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361.3831 - 7ms/epoch - 110us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.5441 - 651us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.9436 - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9622 - 894us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.6976 - 660us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.1822 - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.9514 - 657us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.7212 - 746us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7016 - 812us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.5138 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.9769 - 842us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.6211 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.9358 - 566us/epoch - 9us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.3642 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 99.4025 - 744us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 258/300\n", + "Solving for Nash Equilibrium in Generation 258/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.6172 - 727us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.0088 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.2963 - 822us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.2992 - 671us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 110us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.0961 - 828us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7944 - 619us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.8589 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.5713 - 630us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6693 - 652us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.9235 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.7474 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.6158 - 3ms/epoch - 50us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.6653 - 771us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.4478 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.7165 - 904us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.4329 - 674us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.3788 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.8370 - 726us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.1899 - 759us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 209.6717 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.2418 - 1ms/epoch - 22us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.2006 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4893 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.9727 - 786us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3776 - 747us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.9537 - 7ms/epoch - 105us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.8805 - 1ms/epoch - 22us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.6177 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3640 - 1ms/epoch - 21us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.3176 - 691us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.6163 - 852us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.9496 - 745us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.4698 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6310 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.6356 - 650us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.1554 - 4ms/epoch - 65us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.9199 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.8835 - 653us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.3185 - 642us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.7184 - 673us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 747.5931 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.6300 - 575us/epoch - 9us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.8192 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.4921 - 692us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.3560 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.5532 - 3ms/epoch - 46us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.8188 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389.5391 - 3ms/epoch - 47us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214.7083 - 957us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49.4337 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.0652 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4471 - 1ms/epoch - 21us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.1436 - 553us/epoch - 9us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.7984 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0497 - 696us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.4114 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.9435 - 569us/epoch - 9us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1060 - 2ms/epoch - 34us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.2192 - 761us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.8109 - 717us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.7693 - 4ms/epoch - 63us/sample\n", + "Episode 50/50\n", + "Generation 259/300\n", + "Solving for Nash Equilibrium in Generation 259/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.2777 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.6408 - 682us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.0056 - 663us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 139.4108 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.1370 - 6ms/epoch - 96us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2584 - 819us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.4242 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.8249 - 3ms/epoch - 41us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.8285 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.9195 - 583us/epoch - 9us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.8541 - 772us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.9681 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.9703 - 1ms/epoch - 18us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.9051 - 730us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0358 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.5674 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2213 - 708us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.5208 - 706us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.2404 - 734us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.3270 - 1ms/epoch - 18us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.3743 - 740us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.7663 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.1921 - 5ms/epoch - 74us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.1185 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.2675 - 795us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8171 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.6779 - 733us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.1116 - 747us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.5117 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.7394 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3536 - 789us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.5261 - 858us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.1147 - 808us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.3855 - 2ms/epoch - 28us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.2595 - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.7134 - 772us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.2980 - 611us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.2752 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319.0596 - 795us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.5091 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.4717 - 1ms/epoch - 20us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.1320 - 695us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.7344 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.1811 - 821us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.1897 - 757us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.4523 - 847us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.9916 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.2693 - 575us/epoch - 9us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 175.2186 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.4772 - 691us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.1550 - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.1892 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.7282 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.7804 - 728us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.0441 - 659us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.3610 - 873us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.6290 - 914us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.0739 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.9813 - 633us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.4423 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5332 - 707us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.1683 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 722.2518 - 756us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.9188 - 634us/epoch - 10us/sample\n", + "Generation 260/300\n", + "Solving for Nash Equilibrium in Generation 260/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.9726 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.7936 - 655us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.5260 - 1ms/epoch - 24us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.8787 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.1639 - 700us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2054 - 892us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.3928 - 632us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.9448 - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9177 - 756us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.5515 - 787us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.0126 - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.8452 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.5501 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.6664 - 705us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.3938 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321.2694 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2438 - 571us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.3746 - 640us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.8654 - 702us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.6135 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.9856 - 4ms/epoch - 63us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.6516 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.6490 - 809us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.5308 - 962us/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8867 - 661us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.4110 - 2ms/epoch - 30us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.0075 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.4566 - 770us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40.9194 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.2470 - 1ms/epoch - 20us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.7112 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.4043 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.6560 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.1000 - 761us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.7633 - 769us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.3881 - 1ms/epoch - 18us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.3763 - 720us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.4030 - 740us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.6096 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.1309 - 834us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.4572 - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.6238 - 762us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.2007 - 5ms/epoch - 76us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.1588 - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7604 - 1ms/epoch - 18us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.6972 - 785us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.9939 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.0363 - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.2805 - 651us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.4763 - 822us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.1957 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.8030 - 17ms/epoch - 276us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.7246 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2029 - 733us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.6570 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.8304 - 1ms/epoch - 18us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4631 - 565us/epoch - 9us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.4221 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.9165 - 7ms/epoch - 120us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.8906 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.0031 - 1ms/epoch - 18us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.3319 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9974 - 644us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.5830 - 992us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.6094 - 842us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.2136 - 811us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.6538 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.3061 - 743us/epoch - 12us/sample\n", + "Generation 261/300\n", + "Solving for Nash Equilibrium in Generation 261/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.2930 - 3ms/epoch - 41us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.9796 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.3962 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.7750 - 2ms/epoch - 38us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.0351 - 776us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.6434 - 1ms/epoch - 19us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.0651 - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.3052 - 803us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 730.5243 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0886 - 793us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.8339 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.2229 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7892 - 728us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 171.6415 - 950us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.6913 - 816us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.6559 - 843us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376.0885 - 678us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.7137 - 1ms/epoch - 24us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.1562 - 795us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.3728 - 785us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.4935 - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.2551 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.2259 - 722us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.1813 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.3897 - 749us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.8320 - 663us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.2506 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.4437 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.5892 - 970us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.7470 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.1957 - 2ms/epoch - 27us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.4370 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.9005 - 1ms/epoch - 17us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.0024 - 584us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.8616 - 632us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.6573 - 770us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.2378 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.2305 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.3071 - 844us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.6377 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7831 - 726us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.7714 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.5684 - 835us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.7605 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.4325 - 803us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.5168 - 1ms/epoch - 21us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.0646 - 1ms/epoch - 16us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8597 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.3786 - 562us/epoch - 9us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2021 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.5035 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2968 - 653us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.5450 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.1254 - 590us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.8068 - 855us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.4892 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.4510 - 692us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.0552 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.5359 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2552 - 794us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.3843 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.7236 - 5ms/epoch - 76us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.9646 - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8877 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.8318 - 2ms/epoch - 26us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.5178 - 1ms/epoch - 17us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.4861 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.4888 - 1ms/epoch - 18us/sample\n", + "Generation 262/300\n", + "Solving for Nash Equilibrium in Generation 262/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.5842 - 856us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.5314 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.4502 - 762us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.3708 - 648us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.8419 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.1929 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.1935 - 711us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753.7278 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.5236 - 967us/epoch - 16us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.7433 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9462 - 644us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.9114 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.6113 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.7409 - 2ms/epoch - 30us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5302 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6855 - 5ms/epoch - 85us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.9617 - 1ms/epoch - 17us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.6100 - 1ms/epoch - 18us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.6696 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.1940 - 1ms/epoch - 18us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769.0184 - 752us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.5574 - 857us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.2291 - 798us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6107 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.1985 - 765us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.4185 - 834us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.4996 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.9410 - 743us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.3151 - 625us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.7797 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.2746 - 955us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9508 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.7918 - 752us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.6215 - 842us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5371 - 1ms/epoch - 19us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.1598 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.5868 - 642us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8359 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.6431 - 802us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.5895 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.4696 - 786us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.4027 - 848us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.6849 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.5329 - 13ms/epoch - 203us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.8892 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.7966 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.5220 - 2ms/epoch - 30us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.9242 - 817us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.0609 - 795us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.2673 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4902 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.7806 - 835us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.8792 - 6ms/epoch - 91us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.5353 - 1ms/epoch - 20us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.1334 - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8595 - 959us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.3783 - 778us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.8584 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.9210 - 907us/epoch - 15us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.6397 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.0919 - 2ms/epoch - 33us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.0302 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.1833 - 874us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.7067 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.0497 - 715us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2892 - 860us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.8599 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231.6375 - 940us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8684 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.5499 - 788us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.0097 - 7ms/epoch - 116us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.2271 - 2ms/epoch - 26us/sample\n", + "Generation 263/300\n", + "Solving for Nash Equilibrium in Generation 263/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.0236 - 745us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 106us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.4088 - 2ms/epoch - 25us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.0179 - 799us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.7977 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.2308 - 8ms/epoch - 128us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.4035 - 698us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.0658 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.9214 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.0069 - 788us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.3309 - 889us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.9598 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5368 - 842us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.2172 - 759us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.1980 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.6431 - 603us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6924 - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.3120 - 821us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 612.3658 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.5414 - 664us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.6516 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1556 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0023 - 1ms/epoch - 22us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.3314 - 773us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.3994 - 5ms/epoch - 85us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5814 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.1376 - 967us/epoch - 16us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.4777 - 761us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.9481 - 941us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322.1032 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.0281 - 9ms/epoch - 145us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.2552 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9880 - 839us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.8642 - 829us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.7780 - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.8097 - 766us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.1200 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.5367 - 883us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.7352 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.9340 - 7ms/epoch - 105us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.8667 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.3864 - 2ms/epoch - 30us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.0000 - 1ms/epoch - 16us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.8848 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.0137 - 3ms/epoch - 43us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.8137 - 718us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.0670 - 777us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.5200 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.6457 - 634us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.8039 - 738us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.3279 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.4902 - 821us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.6954 - 1ms/epoch - 17us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.4872 - 692us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.6278 - 661us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.1559 - 635us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4529 - 14ms/epoch - 219us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.2681 - 782us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.6250 - 827us/epoch - 13us/sample\n", + "Generation 264/300\n", + "Solving for Nash Equilibrium in Generation 264/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.3741 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.7346 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.8083 - 810us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.4445 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.9568 - 4ms/epoch - 58us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.4558 - 762us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.4204 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.8602 - 787us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.2269 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9447 - 1ms/epoch - 17us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.4485 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245.0478 - 1ms/epoch - 22us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726.2261 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 92us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.6906 - 902us/epoch - 15us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.4270 - 778us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.2571 - 621us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.3650 - 645us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.0479 - 588us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.1048 - 723us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.2729 - 780us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.5879 - 729us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.8073 - 698us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.2663 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.3335 - 655us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.1068 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.0723 - 1ms/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.0392 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.3513 - 723us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1441 - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.2685 - 20ms/epoch - 320us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.9829 - 871us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.3771 - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.5562 - 836us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.2303 - 678us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.7875 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.2989 - 769us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 756.8454 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.7638 - 5ms/epoch - 81us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.3523 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.6296 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.5251 - 900us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.1182 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6750 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.9333 - 696us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.0040 - 830us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.0450 - 687us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.0075 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.8624 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.6682 - 616us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.5056 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.1360 - 3ms/epoch - 44us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.4738 - 2ms/epoch - 32us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.6106 - 899us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.3680 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.3855 - 769us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.9819 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.8155 - 622us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.4152 - 13ms/epoch - 203us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.8872 - 928us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.5691 - 1ms/epoch - 16us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8592 - 591us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1945 - 576us/epoch - 9us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.6209 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.6042 - 1ms/epoch - 23us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.8458 - 875us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.5270 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.8380 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.4287 - 915us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.6629 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.8130 - 1ms/epoch - 17us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.5239 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.4095 - 586us/epoch - 9us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.0225 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.4851 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.5556 - 820us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.7733 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.1223 - 2ms/epoch - 33us/sample\n", + "Episode 50/50\n", + "Generation 265/300\n", + "Solving for Nash Equilibrium in Generation 265/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.2094 - 889us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.7837 - 718us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278.2362 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.7891 - 1ms/epoch - 16us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47.6105 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.5933 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.2510 - 811us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.9805 - 5ms/epoch - 81us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7236 - 711us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.7889 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.8714 - 622us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.6856 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9282 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.8363 - 2ms/epoch - 30us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640.2745 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.4149 - 718us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.1713 - 662us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.8569 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.7384 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.1731 - 704us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.7016 - 671us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.4593 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.5708 - 785us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.9056 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.1904 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.6329 - 892us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.7430 - 716us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.3636 - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.8234 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414.2267 - 815us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.2346 - 780us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.4363 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.2672 - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4128 - 762us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.0793 - 913us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8941 - 755us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.0261 - 712us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.5563 - 713us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.1156 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.9724 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.3107 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 125us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.1081 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8412 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.1983 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.7574 - 746us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2334 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9096 - 2ms/epoch - 28us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.5818 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.1885 - 751us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.1431 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1412 - 796us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 751.2259 - 843us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.5129 - 802us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.6165 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.4340 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.6169 - 504us/epoch - 8us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.7911 - 791us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 155us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9485 - 789us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.9037 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.6885 - 720us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8453 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.8854 - 641us/epoch - 10us/sample\n", + "Generation 266/300\n", + "Solving for Nash Equilibrium in Generation 266/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.4348 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.2351 - 863us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.9733 - 817us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.0569 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.5936 - 714us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.7704 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.8151 - 904us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.1566 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.3008 - 640us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.5853 - 719us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.1671 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.6180 - 2ms/epoch - 32us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.1301 - 910us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.1162 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 643.7417 - 687us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.7630 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.1015 - 2ms/epoch - 26us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.6506 - 647us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.1704 - 686us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.3272 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.6906 - 711us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4297 - 785us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.7122 - 679us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.4091 - 1ms/epoch - 18us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.4987 - 585us/epoch - 9us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61.2052 - 675us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.0254 - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.5105 - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.4998 - 762us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.7012 - 677us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8090 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2355 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.3711 - 17ms/epoch - 271us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 574.8404 - 684us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9077 - 991us/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.1390 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.2325 - 1ms/epoch - 17us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.1246 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.8553 - 2ms/epoch - 27us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.3463 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250.0309 - 744us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7806 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.3205 - 648us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.6219 - 1ms/epoch - 23us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.7822 - 772us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.4903 - 2ms/epoch - 27us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.7658 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.6451 - 575us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.4482 - 934us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.6837 - 756us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.9688 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7745 - 820us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.1926 - 958us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.2557 - 669us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0007 - 3ms/epoch - 46us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.8204 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.3739 - 4ms/epoch - 59us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5765 - 701us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.6061 - 8ms/epoch - 130us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.1744 - 616us/epoch - 10us/sample\n", + "Generation 267/300\n", + "Solving for Nash Equilibrium in Generation 267/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5953 - 610us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.7961 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.1051 - 690us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.1576 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.7632 - 705us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.7206 - 730us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.8478 - 2ms/epoch - 35us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.1337 - 2ms/epoch - 30us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.4638 - 802us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.2795 - 640us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.2708 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6424 - 626us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.9523 - 5ms/epoch - 73us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.5949 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.5919 - 888us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.7800 - 871us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44.0811 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.9904 - 986us/epoch - 16us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.6219 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2487 - 613us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.7275 - 12ms/epoch - 191us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.8845 - 907us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.2939 - 598us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.2657 - 924us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.5390 - 756us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.2015 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.0416 - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.8860 - 860us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.7965 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.1407 - 804us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.0469 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.2472 - 2ms/epoch - 36us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.4656 - 670us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.4500 - 723us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.9049 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.2372 - 949us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.2031 - 5ms/epoch - 74us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.9430 - 734us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.8570 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.6563 - 621us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.7635 - 620us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9373 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.5988 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.7089 - 1ms/epoch - 17us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0825 - 918us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.0803 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.6498 - 714us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2939 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.6221 - 831us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6247 - 955us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.1597 - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.3990 - 801us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.3417 - 804us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.7031 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.5914 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9676 - 7ms/epoch - 107us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.3792 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.6445 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.0397 - 707us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.4616 - 623us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305.5118 - 1ms/epoch - 16us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0898 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.2099 - 789us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9886 - 3ms/epoch - 43us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.1638 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.0171 - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.4653 - 797us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6527 - 718us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.5294 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2905 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.6863 - 676us/epoch - 11us/sample\n", + "Generation 268/300\n", + "Solving for Nash Equilibrium in Generation 268/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.7026 - 922us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.9729 - 722us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.0155 - 767us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.1874 - 1ms/epoch - 18us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.6250 - 758us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.1721 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3893 - 604us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.7664 - 674us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.5934 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.8187 - 2ms/epoch - 28us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.7988 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.2028 - 709us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.7453 - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.1180 - 786us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.5568 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.3054 - 831us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.3221 - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.4749 - 9ms/epoch - 144us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.5582 - 1000us/epoch - 16us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.6064 - 4ms/epoch - 63us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.3599 - 601us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.5858 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.6086 - 901us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410.7718 - 991us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.8536 - 1ms/epoch - 22us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.3782 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.6744 - 773us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2000 - 786us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231.3515 - 870us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.9333 - 1ms/epoch - 17us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4425 - 847us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.8652 - 1ms/epoch - 17us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.3859 - 978us/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.6437 - 654us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.5919 - 720us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8171 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.1841 - 897us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2764 - 737us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.7078 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.8391 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.5575 - 826us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.4023 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.8382 - 633us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.9130 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.8067 - 813us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.6124 - 690us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.8940 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.6003 - 781us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.6146 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.6605 - 813us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9986 - 793us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.8793 - 2ms/epoch - 25us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.0788 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.0584 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.2194 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.1624 - 747us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.0743 - 712us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 269/300\n", + "Solving for Nash Equilibrium in Generation 269/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.3264 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37.9614 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.0964 - 697us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.8290 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.6805 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.5804 - 763us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.0092 - 731us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231.5872 - 733us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.6710 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.6617 - 845us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.7528 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.7216 - 2ms/epoch - 36us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.5184 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.4883 - 690us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 108.8872 - 848us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.2968 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.6492 - 889us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.2532 - 984us/epoch - 16us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.0704 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 141us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.2765 - 8ms/epoch - 129us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.2583 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.1113 - 638us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592.7692 - 587us/epoch - 9us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.7875 - 2ms/epoch - 29us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.4656 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.7431 - 801us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.0911 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.3966 - 3ms/epoch - 46us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1182 - 795us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.8160 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.0621 - 946us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346.8639 - 736us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 195.8988 - 1ms/epoch - 18us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41.9670 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.7453 - 781us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9388 - 899us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.5627 - 759us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.0880 - 877us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.7441 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.9144 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.4425 - 659us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.6282 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 157us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.9451 - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.4982 - 625us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.5321 - 2ms/epoch - 25us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0204 - 2ms/epoch - 33us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.6396 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.1001 - 744us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.5905 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0499 - 804us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.2347 - 897us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.6922 - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.3468 - 647us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 22ms/epoch - 349us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8378 - 1ms/epoch - 18us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.7087 - 2ms/epoch - 26us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235.0109 - 879us/epoch - 14us/sample\n", + "Generation 270/300\n", + "Solving for Nash Equilibrium in Generation 270/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.4372 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.3042 - 759us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4331 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.2583 - 954us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.0668 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.5081 - 787us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.6550 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417.0281 - 6ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.0650 - 1ms/epoch - 17us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.1744 - 950us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430.4241 - 776us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.9781 - 799us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.7519 - 1ms/epoch - 19us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.4924 - 964us/epoch - 16us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.0806 - 1ms/epoch - 17us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.2568 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.3990 - 780us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.9122 - 831us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.6002 - 733us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.3521 - 1ms/epoch - 21us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 122.9587 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.9059 - 797us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.6982 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.0718 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.2706 - 875us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.5823 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3508 - 1ms/epoch - 21us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.7591 - 788us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.0068 - 1ms/epoch - 21us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8387 - 712us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8147 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.3738 - 794us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230.2296 - 1ms/epoch - 18us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4719 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.0135 - 734us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.1373 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.8171 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7344 - 2ms/epoch - 25us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.1169 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.9664 - 732us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320.3976 - 853us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.3654 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.8285 - 652us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 159.9114 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.8371 - 713us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.1882 - 700us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.5796 - 589us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.7845 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.0138 - 720us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.7559 - 1ms/epoch - 21us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.1265 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.9222 - 2ms/epoch - 35us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1646 - 699us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 765.0519 - 735us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.8787 - 834us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.7075 - 3ms/epoch - 53us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.9028 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.3913 - 682us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.3782 - 876us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0224 - 813us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.5835 - 969us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2038 - 1ms/epoch - 17us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.7595 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.9907 - 2ms/epoch - 35us/sample\n", + "Generation 271/300\n", + "Solving for Nash Equilibrium in Generation 271/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7006 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9924 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.5653 - 618us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 189.5107 - 866us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.1640 - 1ms/epoch - 20us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.8247 - 860us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.6096 - 741us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.8029 - 768us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.4006 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.8976 - 764us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.5822 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.9753 - 622us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 218.5178 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.7283 - 582us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.9187 - 2ms/epoch - 35us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.8297 - 768us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 98us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.5619 - 1ms/epoch - 22us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.2586 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.6334 - 784us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.5581 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.9596 - 3ms/epoch - 42us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.9414 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.7033 - 694us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.1552 - 738us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.2376 - 727us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.5893 - 865us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3028 - 941us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.1342 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.0374 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.7730 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.5736 - 8ms/epoch - 127us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.3810 - 653us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.9114 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.1636 - 900us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.5402 - 654us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.1404 - 693us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.9133 - 757us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.9527 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.6493 - 607us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 57.7148 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.2004 - 832us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.7673 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.4072 - 570us/epoch - 9us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2896 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.5951 - 623us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.2302 - 744us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.0054 - 950us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7701 - 781us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.8611 - 829us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.8962 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.0691 - 583us/epoch - 9us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.6919 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.0436 - 697us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 146.6904 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.8798 - 720us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.5943 - 694us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.2321 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.7244 - 1ms/epoch - 20us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.6025 - 876us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.2733 - 776us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.0355 - 1ms/epoch - 18us/sample\n", + "Generation 272/300\n", + "Solving for Nash Equilibrium in Generation 272/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.3527 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.6118 - 677us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1962 - 866us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.4974 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7889 - 675us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.6219 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1555 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.9090 - 682us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.3060 - 975us/epoch - 16us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.6313 - 2ms/epoch - 30us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.0535 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.6016 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.7608 - 1ms/epoch - 19us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.4742 - 847us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355.5608 - 1ms/epoch - 23us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.6620 - 751us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.0160 - 749us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.6351 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.8536 - 1ms/epoch - 24us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.3470 - 783us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.6195 - 802us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.2388 - 627us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.0639 - 741us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.2806 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.1781 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.4128 - 721us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.6312 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.3889 - 878us/epoch - 14us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5160 - 1ms/epoch - 21us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.8315 - 3ms/epoch - 41us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9903 - 606us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.6957 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.0689 - 880us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3695 - 773us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.3710 - 673us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.0590 - 755us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.3816 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.4088 - 878us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.5928 - 775us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.5202 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.8513 - 959us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.1682 - 789us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.3627 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.7974 - 838us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.3571 - 614us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7183 - 768us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.4641 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.4109 - 687us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.2382 - 800us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.0584 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.3116 - 651us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.4766 - 861us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420.8700 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2063 - 713us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.4374 - 851us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.6136 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.1302 - 797us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.6163 - 632us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.6177 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.2107 - 660us/epoch - 11us/sample\n", + "Generation 273/300\n", + "Solving for Nash Equilibrium in Generation 273/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.2442 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.4078 - 717us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.3336 - 673us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.5424 - 725us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.8862 - 742us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.5949 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.8719 - 778us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.1205 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.4478 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.7436 - 731us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.1725 - 732us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.8656 - 1ms/epoch - 23us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.6184 - 754us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8494 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1735 - 723us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 11ms/epoch - 180us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.3568 - 4ms/epoch - 72us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.7924 - 959us/epoch - 15us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.9170 - 885us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8472 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.5403 - 565us/epoch - 9us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.9048 - 643us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.9776 - 691us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.7087 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.6295 - 728us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.0874 - 689us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.0516 - 963us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6156 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.2917 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.9701 - 783us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.2330 - 1ms/epoch - 18us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.8516 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.2938 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.7690 - 2ms/epoch - 28us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1527 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.2090 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.0557 - 600us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.7052 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.8321 - 765us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.2919 - 716us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.3923 - 811us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.3025 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.9562 - 820us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.8168 - 652us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.3150 - 754us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.6867 - 685us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.6773 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.1243 - 700us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0528 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.4283 - 728us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.5508 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.5137 - 782us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.1067 - 886us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.1603 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.1215 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.6717 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656.3896 - 1ms/epoch - 19us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.3579 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.1544 - 791us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.2678 - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425.1007 - 704us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.7440 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.0497 - 754us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.5446 - 1ms/epoch - 20us/sample\n", + "Generation 274/300\n", + "Solving for Nash Equilibrium in Generation 274/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.6630 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.1290 - 626us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 674.0487 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.5924 - 2ms/epoch - 29us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.4790 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.3011 - 950us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.6903 - 618us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.8609 - 628us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.4915 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.4728 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.8464 - 676us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.1008 - 722us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.2324 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.4629 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.5467 - 1ms/epoch - 24us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.1184 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 11ms/epoch - 170us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.2242 - 2ms/epoch - 26us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3389 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.4572 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.6953 - 1ms/epoch - 18us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.5703 - 686us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.6059 - 683us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.0329 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.5186 - 683us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.2747 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.4010 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 733.6992 - 1ms/epoch - 17us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.9974 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.8066 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.2619 - 602us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.1912 - 1ms/epoch - 21us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4665 - 642us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 92us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.5507 - 2ms/epoch - 29us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248.2089 - 675us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.0829 - 2ms/epoch - 26us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.5430 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4320 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.5913 - 694us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.1445 - 1ms/epoch - 21us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.9242 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1261 - 3ms/epoch - 41us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.9720 - 701us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.4606 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.7849 - 1ms/epoch - 20us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.1816 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.0304 - 725us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.3877 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.8765 - 1ms/epoch - 18us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.4183 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.8753 - 805us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.0460 - 566us/epoch - 9us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4684 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0341 - 2ms/epoch - 39us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.1303 - 701us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.8503 - 722us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6243 - 625us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2693 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0911 - 898us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.7458 - 732us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8441 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.1354 - 666us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768.4213 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.9537 - 643us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.8244 - 551us/epoch - 9us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.3890 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.3940 - 648us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.7649 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.8655 - 675us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.2249 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.4598 - 700us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.1219 - 727us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.4148 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408.5618 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5714 - 739us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8815 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.5006 - 793us/epoch - 13us/sample\n", + "Generation 275/300\n", + "Solving for Nash Equilibrium in Generation 275/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.2587 - 2ms/epoch - 27us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.1504 - 784us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.6663 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.3335 - 2ms/epoch - 24us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.4941 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.5593 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.9478 - 826us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.4952 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2431 - 691us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.5358 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.4871 - 703us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.2071 - 866us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4665 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.3433 - 1ms/epoch - 22us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.6107 - 2ms/epoch - 28us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.5321 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.9382 - 2ms/epoch - 34us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.7444 - 640us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.9270 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8202 - 688us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0023 - 800us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.9718 - 563us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.0402 - 702us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.6594 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.1052 - 607us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.5991 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.4541 - 656us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.2331 - 691us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377.2659 - 1ms/epoch - 18us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.4299 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.6627 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.9222 - 1ms/epoch - 19us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.0382 - 737us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.2202 - 1ms/epoch - 17us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.2735 - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.1951 - 847us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.3509 - 761us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.0519 - 892us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.6639 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.0157 - 1ms/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.0756 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8161 - 600us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.8539 - 659us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.6907 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.6183 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.0222 - 758us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.5226 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.3591 - 798us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.9573 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.5215 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6152 - 587us/epoch - 9us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.0720 - 878us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300.1183 - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.0316 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.6527 - 754us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.1076 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1234 - 687us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.7702 - 602us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.0312 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.9466 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6882 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 147us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.0678 - 3ms/epoch - 41us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.5833 - 635us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 711.1565 - 746us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255.8373 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314.9235 - 2ms/epoch - 35us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.2919 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.2862 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.9744 - 2ms/epoch - 38us/sample\n", + "Generation 276/300\n", + "Solving for Nash Equilibrium in Generation 276/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.9423 - 760us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49.1810 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289.0287 - 612us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.3680 - 716us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.3723 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.5318 - 1ms/epoch - 22us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.2391 - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.7903 - 2ms/epoch - 30us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.2162 - 697us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.6365 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.6411 - 699us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.0861 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.4236 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.1535 - 656us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.2963 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.5203 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.0499 - 695us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8162 - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.2298 - 969us/epoch - 16us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.6989 - 1ms/epoch - 16us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.7031 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 413.1872 - 820us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0790 - 662us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 432.3549 - 762us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.9300 - 660us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.3912 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.0024 - 1ms/epoch - 19us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.5075 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.7300 - 705us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.0015 - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.0130 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8665 - 640us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8789 - 949us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.1240 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.4208 - 701us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.2191 - 634us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9732 - 3ms/epoch - 41us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.5148 - 663us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.6691 - 670us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6002 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.5152 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.1089 - 2ms/epoch - 28us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.9288 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.6225 - 850us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.7845 - 752us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.8156 - 768us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441.5827 - 894us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.8864 - 926us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.0331 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.9258 - 628us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.5150 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.0405 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.9134 - 851us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.3390 - 760us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.4602 - 4ms/epoch - 65us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.5630 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.6909 - 829us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.4811 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.8343 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.8978 - 622us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.8368 - 687us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.2283 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.5816 - 703us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.0063 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9631 - 816us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.7006 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.6316 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.2134 - 2ms/epoch - 28us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.6124 - 811us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Generation 277/300\n", + "Solving for Nash Equilibrium in Generation 277/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.2502 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.1440 - 718us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.5952 - 816us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.9719 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.9772 - 767us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.0632 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 130us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.9546 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2649 - 985us/epoch - 16us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1996 - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.6027 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9944 - 769us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6255 - 970us/epoch - 16us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 148us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.3501 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.0583 - 713us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6534 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.3505 - 1ms/epoch - 22us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.9725 - 801us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.3859 - 901us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0541 - 712us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.6173 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.4652 - 840us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.6363 - 680us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 225.3580 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728.6049 - 721us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.3307 - 605us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.7872 - 786us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.1035 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371.7458 - 714us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1346 - 661us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8199 - 666us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.2768 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.7908 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4321 - 725us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.7225 - 1ms/epoch - 17us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.6045 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.4092 - 758us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.5436 - 5ms/epoch - 76us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.6404 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.0725 - 813us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.4359 - 724us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.6217 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.7226 - 581us/epoch - 9us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 207.0680 - 768us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.5312 - 686us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7214 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.5681 - 770us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.6292 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.0541 - 609us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.4431 - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.1116 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 599.9333 - 2ms/epoch - 35us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.5569 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.1056 - 788us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.5300 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.4863 - 766us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 118us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409.3542 - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.7322 - 722us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.7169 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.8785 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1688 - 686us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.5577 - 563us/epoch - 9us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347.7551 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.4625 - 610us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9032 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 659.6779 - 647us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.8203 - 968us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.3716 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328.6336 - 767us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366.9402 - 593us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.6763 - 2ms/epoch - 31us/sample\n", + "Generation 278/300\n", + "Solving for Nash Equilibrium in Generation 278/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.0381 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422.9046 - 2ms/epoch - 39us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.0818 - 695us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.8686 - 602us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4165 - 692us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 118us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.4949 - 2ms/epoch - 26us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.9452 - 713us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.7892 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.3042 - 879us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.7263 - 789us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.1010 - 803us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.7315 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.8268 - 694us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.9447 - 639us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.2369 - 708us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.0468 - 722us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.4942 - 656us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.2353 - 671us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2631 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4314 - 740us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.1506 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.9340 - 1ms/epoch - 24us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 239.5657 - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.9658 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.7054 - 1ms/epoch - 18us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3086 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.6275 - 1ms/epoch - 16us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.8611 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.0392 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4489 - 1ms/epoch - 17us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.3156 - 717us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.5290 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.1400 - 799us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.4229 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.5732 - 634us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358.4662 - 808us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.7905 - 722us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.6573 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.6128 - 850us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.7224 - 1ms/epoch - 17us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9893 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4717 - 1ms/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.6561 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558.8756 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.8207 - 921us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.3018 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.7938 - 867us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.1509 - 739us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.9520 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.0318 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.6044 - 906us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.3497 - 741us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348.0693 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.1368 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.4984 - 780us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.6703 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.6127 - 703us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1728 - 801us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.7057 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5079 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2137 - 595us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.0582 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.8995 - 590us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.9655 - 902us/epoch - 15us/sample\n", + "Generation 279/300\n", + "Solving for Nash Equilibrium in Generation 279/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465.8466 - 745us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 185.7159 - 2ms/epoch - 28us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.4020 - 2ms/epoch - 38us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.7288 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.6849 - 699us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.4896 - 687us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.4616 - 718us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.9906 - 2ms/epoch - 37us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.1481 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 379.5795 - 720us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700.1495 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.5641 - 730us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.4155 - 709us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.8448 - 698us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.8185 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.8251 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.3698 - 2ms/epoch - 32us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 234.7809 - 703us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.0239 - 823us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.5143 - 686us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.6559 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.0480 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.1771 - 642us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8354 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.6978 - 662us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.9085 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4720 - 743us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.4187 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.2740 - 720us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2783 - 688us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.6632 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6738 - 584us/epoch - 9us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.2410 - 898us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6246 - 674us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.2004 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.7410 - 770us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.1435 - 588us/epoch - 9us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.1151 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.8044 - 745us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.5391 - 634us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.5415 - 566us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1633 - 752us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.2867 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4000 - 705us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.7761 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.1733 - 696us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.5455 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2992 - 586us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.4691 - 665us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.8944 - 757us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.8580 - 971us/epoch - 16us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.3504 - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.8093 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.9802 - 1ms/epoch - 18us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3665 - 1ms/epoch - 17us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.3937 - 598us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.8464 - 761us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.3029 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.9200 - 2ms/epoch - 28us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440.7085 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.7076 - 552us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6800 - 848us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.7006 - 685us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.7311 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.5388 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.6259 - 916us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3446 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.7719 - 667us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.5806 - 913us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.9657 - 1ms/epoch - 16us/sample\n", + "Generation 280/300\n", + "Solving for Nash Equilibrium in Generation 280/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.2365 - 721us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.3355 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.0302 - 2ms/epoch - 30us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.4315 - 885us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.7108 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.1665 - 602us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.6071 - 812us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.0710 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622.3260 - 648us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.1620 - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 225.9471 - 940us/epoch - 15us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561.9802 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.1482 - 635us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.3000 - 686us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7894 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.5828 - 602us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.4739 - 682us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.6229 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.4530 - 721us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.0649 - 3ms/epoch - 56us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.5330 - 683us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.4049 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.0555 - 6ms/epoch - 102us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.1672 - 891us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.8513 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.1485 - 2ms/epoch - 24us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.2824 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2880 - 631us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.6917 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.8804 - 717us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2530 - 838us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.6107 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.2465 - 810us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.3915 - 2ms/epoch - 31us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.5144 - 618us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.6571 - 2ms/epoch - 29us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.5279 - 813us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.1921 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.9430 - 3ms/epoch - 46us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.7971 - 3ms/epoch - 44us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6842 - 783us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 723.1942 - 888us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.1983 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.4993 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.1358 - 673us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.5593 - 562us/epoch - 9us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2190 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.0821 - 895us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706.1249 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.3948 - 913us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.3425 - 718us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.4364 - 734us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8600 - 608us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750.0959 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.0939 - 558us/epoch - 9us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.8930 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.0510 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.6425 - 2ms/epoch - 30us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.8222 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.2509 - 562us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5515 - 1ms/epoch - 19us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4305 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.0275 - 2ms/epoch - 26us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.3388 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.2518 - 757us/epoch - 12us/sample\n", + "Generation 281/300\n", + "Solving for Nash Equilibrium in Generation 281/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.0204 - 582us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.2408 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 140us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.0246 - 997us/epoch - 16us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.1329 - 1ms/epoch - 20us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.4394 - 666us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3260 - 767us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.0928 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.0665 - 760us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.7980 - 720us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250.8750 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.7908 - 771us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.6379 - 1ms/epoch - 20us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.2731 - 640us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.6941 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.0069 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.7910 - 622us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.2582 - 693us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.6202 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.7440 - 661us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.2341 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.1540 - 3ms/epoch - 41us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5841 - 786us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.9190 - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.9883 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.5418 - 625us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240.9090 - 712us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 97.8924 - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.2205 - 761us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.5656 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.2283 - 995us/epoch - 16us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.2487 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5397 - 827us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.3968 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.3355 - 610us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.6743 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.0884 - 875us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.1606 - 599us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.1588 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.8628 - 598us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675.3964 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.7076 - 693us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.1324 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.9202 - 750us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442.8620 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.4461 - 587us/epoch - 9us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68.4759 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.2973 - 743us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.3364 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.5343 - 700us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.8137 - 3ms/epoch - 46us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.4183 - 632us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8565 - 652us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.3318 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.5284 - 758us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.2201 - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.6373 - 808us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.1043 - 821us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.4184 - 1ms/epoch - 17us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.3855 - 806us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.5980 - 2ms/epoch - 33us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.9914 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.9772 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.6143 - 1ms/epoch - 18us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 679.7869 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581.6960 - 1ms/epoch - 18us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.9598 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.2618 - 927us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.5518 - 2ms/epoch - 24us/sample\n", + "Generation 282/300\n", + "Solving for Nash Equilibrium in Generation 282/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201.4706 - 645us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2041 - 607us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.9050 - 830us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.7825 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460.4789 - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.4846 - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288.7498 - 1ms/epoch - 18us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.9601 - 1ms/epoch - 17us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.2314 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4799 - 583us/epoch - 9us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.4602 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.5015 - 691us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.3088 - 1ms/epoch - 17us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.0697 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.7833 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.6477 - 4ms/epoch - 70us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.7535 - 787us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.1902 - 704us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2243 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 598.2917 - 5ms/epoch - 80us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.9747 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.0179 - 784us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.4362 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.8029 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.8811 - 663us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418.4906 - 702us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.7935 - 706us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.1423 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.0682 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.5506 - 960us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.0815 - 682us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.6520 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668.8569 - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.3021 - 776us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.3281 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.9575 - 630us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.7984 - 1ms/epoch - 24us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.3210 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6459 - 8ms/epoch - 129us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 213.7166 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6862 - 923us/epoch - 15us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.2800 - 782us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.4184 - 714us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 697.4760 - 755us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.7698 - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.8522 - 846us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.8654 - 787us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.5568 - 829us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342.2226 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.9565 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8150 - 2ms/epoch - 30us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.9313 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.8219 - 766us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.0969 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.8861 - 772us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.0153 - 960us/epoch - 15us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.4423 - 815us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.7394 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.6324 - 629us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.9222 - 4ms/epoch - 73us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662.7023 - 741us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.3903 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.3624 - 714us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.7817 - 666us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.1241 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.0693 - 2ms/epoch - 36us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7415 - 863us/epoch - 14us/sample\n", + "Generation 283/300\n", + "Solving for Nash Equilibrium in Generation 283/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.9859 - 746us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.2003 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.9066 - 1ms/epoch - 19us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 707.3770 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.6597 - 1ms/epoch - 18us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 82.3016 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2496 - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.1255 - 685us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.3702 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681.4265 - 613us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676.3737 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9023 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2882 - 2ms/epoch - 26us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.0027 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.6994 - 761us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9529 - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.6990 - 1ms/epoch - 19us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.5968 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.9877 - 668us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.8571 - 618us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.6818 - 9ms/epoch - 146us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245.7813 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0275 - 1ms/epoch - 17us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.7501 - 649us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.1519 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.2490 - 6ms/epoch - 104us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.4868 - 799us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580.9151 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.4773 - 3ms/epoch - 45us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.2136 - 2ms/epoch - 30us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.6628 - 802us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.5949 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.5504 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.3024 - 709us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 716.0488 - 797us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554.0833 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.9995 - 768us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 128us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.5329 - 3ms/epoch - 56us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.2005 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.1924 - 683us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.1882 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.1805 - 659us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.7783 - 659us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.8396 - 3ms/epoch - 54us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 94.7427 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.0976 - 758us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.4802 - 564us/epoch - 9us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.2786 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.1889 - 750us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.9435 - 707us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.0371 - 594us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.9688 - 699us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.0107 - 632us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.9377 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.0451 - 651us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.4497 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.0789 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.5298 - 658us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.7786 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.4895 - 620us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.7077 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593.7160 - 776us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 614.5399 - 795us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.3220 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.3823 - 764us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.2122 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.3116 - 700us/epoch - 11us/sample\n", + "Generation 284/300\n", + "Solving for Nash Equilibrium in Generation 284/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.2204 - 759us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 128us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.2006 - 2ms/epoch - 26us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310.8250 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.9641 - 622us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 754.5701 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.9607 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 432.6996 - 582us/epoch - 9us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.7563 - 626us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3309 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.5884 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.9552 - 588us/epoch - 9us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.0010 - 849us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.8676 - 619us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.3185 - 892us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 541.2461 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.1728 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.1215 - 627us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.3323 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.7910 - 3ms/epoch - 42us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.3309 - 691us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3782 - 690us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.9021 - 792us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.9371 - 869us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.8715 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.7928 - 688us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.6202 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.5513 - 748us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.4816 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.4294 - 719us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339.2344 - 629us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335.6691 - 582us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316.8285 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.4460 - 3ms/epoch - 44us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.6146 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.8051 - 677us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3873 - 2ms/epoch - 28us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.0563 - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.1646 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.0621 - 637us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.4932 - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.4589 - 1ms/epoch - 19us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587.5678 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.9797 - 1ms/epoch - 24us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.2122 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3540 - 919us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.1796 - 803us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631.2349 - 907us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.3725 - 682us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.5362 - 1ms/epoch - 20us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.5310 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7499 - 703us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.8712 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.8204 - 841us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.9057 - 3ms/epoch - 49us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.9230 - 2ms/epoch - 30us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.6604 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8669 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8065 - 723us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496.5735 - 642us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.2034 - 775us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4702 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.6419 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727.4885 - 742us/epoch - 12us/sample\n", + "Generation 285/300\n", + "Solving for Nash Equilibrium in Generation 285/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.7795 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.3723 - 832us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37.5393 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470.0734 - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.9687 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.3064 - 627us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588.3138 - 581us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.0787 - 2ms/epoch - 34us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.9968 - 568us/epoch - 9us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.8088 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.5926 - 687us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.9296 - 823us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.4266 - 874us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.0742 - 855us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.3234 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.1371 - 745us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.2667 - 872us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.6350 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.8979 - 810us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.9398 - 792us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.7907 - 796us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.1096 - 718us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.0131 - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.6301 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.2850 - 2ms/epoch - 35us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.4650 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436.0374 - 2ms/epoch - 31us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.9651 - 848us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.0206 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264.2245 - 4ms/epoch - 58us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.3455 - 10ms/epoch - 154us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.8112 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.1714 - 671us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.4096 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.1087 - 758us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.3441 - 655us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.0941 - 894us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7205 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.9147 - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1508 - 5ms/epoch - 74us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.9905 - 926us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.3648 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.2946 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.8183 - 773us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.7015 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380.1765 - 851us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.2131 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.4330 - 970us/epoch - 16us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.6856 - 904us/epoch - 15us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35.6194 - 6ms/epoch - 102us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 253.3007 - 1ms/epoch - 19us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.0244 - 651us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.4968 - 656us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296.4969 - 830us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.6953 - 659us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.4158 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680.0135 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373.4787 - 865us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.2612 - 593us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.8085 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594.3720 - 670us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 286/300\n", + "Solving for Nash Equilibrium in Generation 286/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.8361 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.6907 - 873us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.1248 - 768us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.0240 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.6917 - 1ms/epoch - 18us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.3369 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.4708 - 939us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 717.5502 - 830us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.6219 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.4449 - 749us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.3088 - 740us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.9514 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.2338 - 727us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.2764 - 1ms/epoch - 23us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.9625 - 715us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 262.9796 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.9222 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.8080 - 803us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.2914 - 1ms/epoch - 17us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6996 - 675us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.0292 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.0283 - 908us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.2881 - 586us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.1909 - 2ms/epoch - 39us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.8235 - 892us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.9081 - 694us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667.0496 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.3505 - 694us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.9298 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.2690 - 618us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.8300 - 942us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.5469 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467.0268 - 658us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713.0562 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.1775 - 902us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.7064 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0974 - 689us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.3442 - 760us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.9362 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.4103 - 636us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265.6584 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0213 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399.8616 - 2ms/epoch - 36us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1867 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.2767 - 622us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2623 - 790us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360.7293 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.8245 - 579us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.6155 - 676us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250.6891 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 105us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.9980 - 2ms/epoch - 31us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.7117 - 693us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.0974 - 699us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337.9493 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.7991 - 954us/epoch - 15us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311.7764 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.4938 - 878us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6565 - 760us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.2229 - 667us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.0854 - 565us/epoch - 9us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.7926 - 7ms/epoch - 109us/sample\n", + "Generation 287/300\n", + "Solving for Nash Equilibrium in Generation 287/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.5097 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 403.6659 - 922us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.6785 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.6760 - 1ms/epoch - 17us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2218 - 973us/epoch - 16us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.1180 - 4ms/epoch - 66us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.9170 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.7748 - 1ms/epoch - 23us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.3571 - 872us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8097 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.2518 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.8825 - 696us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.4157 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.2126 - 622us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323.7906 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.8722 - 723us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.6446 - 708us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.6382 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.2049 - 692us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.3238 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439.0863 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6362 - 939us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.6790 - 619us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 113us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326.6340 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.3243 - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.4057 - 728us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.8008 - 1ms/epoch - 19us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.8477 - 744us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.4530 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.6425 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 233.3021 - 536us/epoch - 9us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.5502 - 701us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.6986 - 807us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.2926 - 1ms/epoch - 16us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 577.1547 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.7450 - 739us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.3241 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.8404 - 713us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381.7829 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4557 - 650us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260.8095 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.2944 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.7059 - 1ms/epoch - 22us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.7155 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.7339 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.4060 - 805us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.8608 - 675us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 223.2663 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 388.6570 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.2411 - 697us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 114us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.0356 - 11ms/epoch - 174us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.1270 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528.3103 - 555us/epoch - 9us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.9454 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 233.5658 - 18ms/epoch - 285us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.4057 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303.5672 - 667us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424.2547 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.7682 - 773us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47.0850 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.2404 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.1376 - 1ms/epoch - 22us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.2518 - 913us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.7343 - 933us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.5964 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.6367 - 1ms/epoch - 24us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.3511 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.8743 - 618us/epoch - 10us/sample\n", + "Generation 288/300\n", + "Solving for Nash Equilibrium in Generation 288/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.2918 - 767us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.5445 - 548us/epoch - 9us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395.5714 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.9531 - 651us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433.1161 - 751us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.8714 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.7135 - 914us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 153.3092 - 646us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.2621 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.0400 - 1ms/epoch - 19us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0692 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.5730 - 683us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.5786 - 636us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329.0647 - 7ms/epoch - 116us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5911 - 795us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.7611 - 6ms/epoch - 98us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.3175 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 12ms/epoch - 195us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.4806 - 2ms/epoch - 35us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.9090 - 802us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.5828 - 4ms/epoch - 58us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.7000 - 789us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.9417 - 821us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.2582 - 707us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.5400 - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.4623 - 593us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.8488 - 859us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 427.0891 - 771us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.1647 - 3ms/epoch - 53us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.8705 - 777us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.1851 - 950us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 641.0267 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.9295 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.9916 - 1ms/epoch - 20us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 164.1413 - 937us/epoch - 15us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.0120 - 908us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.3763 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.6865 - 927us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.1721 - 624us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.8241 - 552us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.3738 - 595us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.9499 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.3050 - 754us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.5970 - 711us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.1164 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.7445 - 877us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.7018 - 2ms/epoch - 32us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.1462 - 1ms/epoch - 18us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.0549 - 760us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.7245 - 882us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.0194 - 946us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649.2711 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.2019 - 1ms/epoch - 16us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9276 - 1ms/epoch - 20us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.5784 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.5984 - 1ms/epoch - 21us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.9722 - 1ms/epoch - 19us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.7821 - 971us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Generation 289/300\n", + "Solving for Nash Equilibrium in Generation 289/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 11ms/epoch - 178us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.8477 - 11ms/epoch - 179us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 144us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.9128 - 2ms/epoch - 33us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254.1751 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 435.6000 - 710us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.4126 - 732us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689.8875 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.6844 - 3ms/epoch - 43us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350.0612 - 2ms/epoch - 24us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363.5245 - 743us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.3629 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758.9636 - 700us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.4429 - 798us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 13ms/epoch - 214us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.1186 - 3ms/epoch - 42us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290.4043 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 16ms/epoch - 259us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568.2307 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 9ms/epoch - 143us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.5257 - 1ms/epoch - 16us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411.5776 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.1307 - 785us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.1276 - 730us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.2114 - 806us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.2535 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.0844 - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 11ms/epoch - 181us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.0556 - 9ms/epoch - 152us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.5245 - 906us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.6418 - 796us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3768 - 833us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.3746 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423.7080 - 681us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.0316 - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527.4798 - 2ms/epoch - 34us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39.3527 - 841us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.6074 - 1ms/epoch - 23us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.4208 - 776us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 121us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41.7300 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.1107 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 16ms/epoch - 255us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.3304 - 1ms/epoch - 18us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.0462 - 635us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438.2468 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.6100 - 1ms/epoch - 17us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.5074 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.2957 - 721us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.9427 - 713us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.2034 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572.6345 - 1ms/epoch - 23us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9390 - 1ms/epoch - 24us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.9017 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.1114 - 893us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.6143 - 729us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.0132 - 910us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.7827 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.6752 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.0298 - 753us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.4245 - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431.7836 - 678us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9829 - 1ms/epoch - 17us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.9091 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397.7200 - 2ms/epoch - 26us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291.5141 - 1ms/epoch - 20us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.4072 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.3575 - 721us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.7602 - 1ms/epoch - 17us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741.5078 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 576.8163 - 686us/epoch - 11us/sample\n", + "Generation 290/300\n", + "Solving for Nash Equilibrium in Generation 290/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.8231 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.3536 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 13ms/epoch - 209us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584.8383 - 1ms/epoch - 20us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.5381 - 699us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.0793 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269.9148 - 599us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456.1642 - 843us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.8895 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.9738 - 3ms/epoch - 56us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.1801 - 2ms/epoch - 32us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.1303 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.2778 - 3ms/epoch - 51us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.7219 - 698us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.1985 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497.0194 - 764us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.3450 - 731us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261.7315 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.4581 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 721.2116 - 2ms/epoch - 27us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.5577 - 1ms/epoch - 19us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8309 - 595us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.4868 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.4894 - 578us/epoch - 9us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.0050 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.0514 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.9166 - 668us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.1006 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.9066 - 3ms/epoch - 43us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.5697 - 696us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.5775 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386.8150 - 659us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 188.1524 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552.8127 - 815us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.7675 - 698us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.6620 - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.0385 - 3ms/epoch - 46us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.3338 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.4555 - 704us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8919 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.6172 - 624us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.7778 - 1ms/epoch - 17us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 648.9186 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.2574 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.6887 - 673us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.3021 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.2311 - 899us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365.8792 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.5899 - 4ms/epoch - 70us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565.0002 - 1ms/epoch - 23us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.8016 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.0854 - 612us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.4799 - 2ms/epoch - 27us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.8925 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4106 - 2ms/epoch - 40us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312.6138 - 1ms/epoch - 23us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.8250 - 713us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.2846 - 2ms/epoch - 28us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.7066 - 901us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443.8437 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 407.9537 - 4ms/epoch - 63us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402.7793 - 731us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.6028 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.4208 - 846us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.3568 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.2422 - 939us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.5140 - 627us/epoch - 10us/sample\n", + "Generation 291/300\n", + "Solving for Nash Equilibrium in Generation 291/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624.0083 - 745us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.5807 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6684 - 730us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.9236 - 857us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338.7137 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.1712 - 1ms/epoch - 20us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.1511 - 2ms/epoch - 28us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.8335 - 682us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.9061 - 34ms/epoch - 542us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.5449 - 761us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 174.6336 - 3ms/epoch - 43us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.5813 - 6ms/epoch - 94us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.6268 - 1ms/epoch - 19us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3134 - 859us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36.4086 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.3561 - 643us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.2975 - 659us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.6463 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.2993 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333.0321 - 667us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.6226 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 118us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.3955 - 1ms/epoch - 22us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.5162 - 745us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.8201 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.2771 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 652.2577 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.7290 - 828us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.7023 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.6740 - 4ms/epoch - 59us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.2821 - 2ms/epoch - 32us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.8915 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699.6909 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.8873 - 841us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5956 - 799us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.4402 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.1461 - 705us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.3546 - 719us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.8770 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.6966 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.7523 - 1ms/epoch - 17us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.7402 - 719us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 89us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353.2758 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.6119 - 1ms/epoch - 23us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4936 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.8611 - 1ms/epoch - 17us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.7912 - 583us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.4693 - 633us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616.1990 - 690us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.0938 - 743us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 141.1749 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.9155 - 1ms/epoch - 17us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 212.6171 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.8931 - 639us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.1862 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.3055 - 713us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.4992 - 625us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.0079 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.8364 - 599us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.1610 - 699us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 130us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.0697 - 2ms/epoch - 31us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.0097 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.0262 - 766us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.9665 - 696us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.2296 - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.0061 - 658us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.1688 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.7081 - 742us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.9906 - 678us/epoch - 11us/sample\n", + "Generation 292/300\n", + "Solving for Nash Equilibrium in Generation 292/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.5258 - 6ms/epoch - 89us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571.5147 - 652us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4734 - 638us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235.2899 - 643us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30.9745 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.8463 - 580us/epoch - 9us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.1866 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669.5080 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.0054 - 976us/epoch - 16us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.2941 - 3ms/epoch - 40us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.4011 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664.7517 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6320 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.3268 - 685us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651.6013 - 844us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.7520 - 696us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 127us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.5807 - 1ms/epoch - 20us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.8043 - 4ms/epoch - 65us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38.2333 - 753us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259.3655 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.5427 - 632us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2753 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.2399 - 658us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.1848 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 597.3925 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.4305 - 582us/epoch - 9us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.2086 - 1ms/epoch - 18us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396.8848 - 800us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.2969 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508.1218 - 718us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 570.4729 - 767us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767.1093 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.3754 - 619us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.4271 - 675us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352.7620 - 609us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.0295 - 712us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.6390 - 726us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.5846 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.9451 - 709us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.4512 - 753us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.0099 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.2758 - 1ms/epoch - 21us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.3042 - 696us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.3111 - 644us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42.1078 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.6677 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.6118 - 573us/epoch - 9us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.1998 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 183.5470 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.0670 - 5ms/epoch - 75us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.4204 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.6253 - 654us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3815 - 1ms/epoch - 20us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.9330 - 632us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 253.3247 - 563us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543.4738 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 210.7021 - 6ms/epoch - 94us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.3716 - 3ms/epoch - 48us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 421.0270 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304.7947 - 1ms/epoch - 17us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.3177 - 659us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 293/300\n", + "Solving for Nash Equilibrium in Generation 293/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471.4898 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.8752 - 750us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.1846 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.7201 - 669us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 744.1115 - 3ms/epoch - 53us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.6815 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.2242 - 674us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.0237 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.2277 - 715us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.8180 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.5428 - 2ms/epoch - 31us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.3773 - 743us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.2645 - 2ms/epoch - 29us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.9002 - 833us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.9886 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.4680 - 943us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445.0404 - 707us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.2441 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.9765 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 661.7610 - 692us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611.1238 - 841us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.2805 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629.2620 - 681us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401.5588 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.8919 - 614us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361.6769 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.3718 - 798us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.4478 - 791us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.2927 - 744us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356.7478 - 1ms/epoch - 17us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.5875 - 738us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.0086 - 694us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 685.1830 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.7636 - 600us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267.5010 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.5675 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.6502 - 957us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.7802 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.6475 - 4ms/epoch - 58us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.1804 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324.8357 - 565us/epoch - 9us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.7593 - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.4402 - 850us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.1429 - 789us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334.9698 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.6664 - 770us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.7586 - 922us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760.3425 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.0541 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.4116 - 1ms/epoch - 17us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.1331 - 659us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688.6259 - 543us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.4530 - 626us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8629 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.5890 - 758us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735.4236 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.8156 - 2ms/epoch - 39us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.2159 - 746us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.6316 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.8007 - 1ms/epoch - 18us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.9828 - 2ms/epoch - 29us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372.3074 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387.7089 - 810us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231.7866 - 672us/epoch - 11us/sample\n", + "Generation 294/300\n", + "Solving for Nash Equilibrium in Generation 294/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.5905 - 740us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364.8820 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.4068 - 690us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.1437 - 720us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.7797 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489.8765 - 807us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.7604 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 216.0269 - 1ms/epoch - 20us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.3766 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560.9428 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628.3181 - 655us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.0086 - 576us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.3416 - 2ms/epoch - 25us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 732.9797 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.1660 - 634us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.2104 - 602us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.3301 - 675us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368.0673 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.7419 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.8688 - 772us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.6304 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.4688 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.5844 - 635us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 498.8271 - 724us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.9493 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.9203 - 965us/epoch - 16us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702.5211 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.9767 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426.0484 - 726us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512.4863 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.6785 - 2ms/epoch - 25us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469.6764 - 726us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.3913 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.0568 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6326 - 732us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.4080 - 767us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718.7950 - 1ms/epoch - 19us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.4125 - 7ms/epoch - 119us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646.5082 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532.6970 - 693us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.8739 - 739us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317.7241 - 649us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.6428 - 921us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719.8817 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 636.4338 - 672us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.5770 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 429.8895 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516.2999 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.0007 - 688us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 567.2843 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.0417 - 715us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.5022 - 756us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.8085 - 696us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301.9022 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.7361 - 731us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.4035 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.6678 - 776us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4728 - 674us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246.9712 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.8622 - 10ms/epoch - 162us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.3574 - 634us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.4604 - 781us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5977 - 749us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258.7140 - 855us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270.4865 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.7164 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.5835 - 684us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639.5588 - 609us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.6852 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.2414 - 6ms/epoch - 102us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281.1775 - 664us/epoch - 11us/sample\n", + "Generation 295/300\n", + "Solving for Nash Equilibrium in Generation 295/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.9600 - 762us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287.5749 - 759us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.8226 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.6456 - 2ms/epoch - 26us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244.4549 - 2ms/epoch - 25us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.2560 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.6104 - 705us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.6135 - 616us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575.2053 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.5369 - 693us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.9526 - 1ms/epoch - 18us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.6084 - 849us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.9271 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235.4726 - 648us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251.3011 - 789us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 739.6884 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.0715 - 656us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.2066 - 8ms/epoch - 123us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5675 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559.5320 - 655us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.6140 - 1ms/epoch - 17us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.1931 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736.6569 - 610us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.8153 - 813us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.5793 - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375.7740 - 851us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692.8233 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273.0339 - 749us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.7148 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 404.7835 - 801us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.8996 - 993us/epoch - 16us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286.2021 - 607us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.9886 - 768us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.3571 - 682us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708.5719 - 574us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.0353 - 686us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268.5297 - 2ms/epoch - 31us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.8575 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.0513 - 651us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522.4388 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672.0671 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.1459 - 827us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9622 - 2ms/epoch - 28us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579.7276 - 2ms/epoch - 37us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392.4626 - 884us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.0853 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479.3046 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.0521 - 601us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.0789 - 3ms/epoch - 53us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.4117 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9483 - 616us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.7488 - 806us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.1673 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349.4977 - 3ms/epoch - 48us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.8100 - 730us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 653.7391 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695.9523 - 4ms/epoch - 64us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.8305 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243.3559 - 585us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357.7169 - 574us/epoch - 9us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.3589 - 704us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660.4737 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.6313 - 1ms/epoch - 21us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472.4028 - 876us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.4915 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.4894 - 659us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461.5875 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.8918 - 681us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607.9556 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724.2642 - 1ms/epoch - 18us/sample\n", + "Generation 296/300\n", + "Solving for Nash Equilibrium in Generation 296/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.8058 - 710us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336.2985 - 732us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.2100 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583.9872 - 537us/epoch - 9us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709.2961 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 715.4185 - 654us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476.0268 - 2ms/epoch - 28us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.6735 - 691us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2985 - 673us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.0174 - 826us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678.3450 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.1726 - 661us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.2383 - 653us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553.3019 - 797us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500.7365 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.5587 - 666us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504.1533 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.3024 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637.6946 - 4ms/epoch - 64us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533.5649 - 1ms/epoch - 22us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731.5056 - 597us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434.9289 - 608us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.6585 - 911us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6091 - 2ms/epoch - 25us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.7613 - 716us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.1459 - 646us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 590.4678 - 768us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492.1864 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 563.4466 - 657us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.5560 - 2ms/epoch - 33us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615.9888 - 855us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35.0815 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604.0304 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 573.4483 - 655us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.7570 - 3ms/epoch - 56us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.8692 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493.7519 - 724us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330.3864 - 1ms/epoch - 16us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617.3687 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 524.0757 - 725us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704.9401 - 871us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351.3992 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275.4955 - 1ms/epoch - 23us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 746.1452 - 713us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.0478 - 732us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.6642 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.0386 - 1ms/epoch - 19us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725.6550 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.0977 - 877us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 645.0825 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.0258 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8643 - 770us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.7189 - 956us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.6473 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9734 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.0518 - 935us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.7525 - 3ms/epoch - 51us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519.7664 - 746us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.5378 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.1991 - 880us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.2380 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 633.8723 - 2ms/epoch - 26us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8314 - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.3571 - 928us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451.3643 - 945us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486.0511 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.4740 - 3ms/epoch - 49us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.1324 - 1ms/epoch - 18us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.7947 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315.8719 - 908us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.9992 - 846us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Generation 297/300\n", + "Solving for Nash Equilibrium in Generation 297/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.6277 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 177.6342 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 84us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343.4575 - 4ms/epoch - 68us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 654.5011 - 851us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369.9113 - 853us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398.6039 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.3488 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283.4766 - 2ms/epoch - 36us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544.2060 - 1ms/epoch - 21us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383.6287 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.1168 - 873us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501.3577 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523.2547 - 761us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.2314 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.6981 - 1ms/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.6076 - 674us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.3563 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.6433 - 3ms/epoch - 51us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299.9958 - 1ms/epoch - 17us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8331 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 226.0677 - 631us/epoch - 10us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 457.2002 - 712us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566.8126 - 731us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502.9332 - 731us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280.0924 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.0102 - 1ms/epoch - 17us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696.2596 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.0577 - 720us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 432.8532 - 2ms/epoch - 34us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284.0345 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6240 - 603us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 529.6959 - 2ms/epoch - 30us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256.5952 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.0853 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720.1152 - 685us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.0751 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621.4149 - 784us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474.3958 - 614us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 92us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477.0042 - 957us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307.5851 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480.4870 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585.3235 - 4ms/epoch - 68us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.9692 - 706us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608.8765 - 7ms/epoch - 116us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.2404 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.6072 - 2ms/epoch - 31us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454.3183 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693.1923 - 850us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.6546 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 72ms/epoch - 1ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.7787 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.6541 - 2ms/epoch - 37us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.0737 - 643us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.2760 - 1ms/epoch - 23us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98.9337 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382.7783 - 858us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647.1547 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.3539 - 4ms/epoch - 66us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.5553 - 1ms/epoch - 22us/sample\n", + "Generation 298/300\n", + "Solving for Nash Equilibrium in Generation 298/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.8488 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545.2293 - 755us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.5003 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.5816 - 1ms/epoch - 18us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609.0994 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562.2281 - 764us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.6225 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.2127 - 5ms/epoch - 88us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.2520 - 735us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626.2302 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.0539 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535.4633 - 2ms/epoch - 40us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 582.7436 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 694.0559 - 627us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9267 - 656us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487.5367 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285.2971 - 709us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506.1842 - 769us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 740.4257 - 756us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 665.8144 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400.3322 - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510.5401 - 858us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.6709 - 2ms/epoch - 38us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525.6832 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.5038 - 1ms/epoch - 17us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.1329 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.4738 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453.1758 - 737us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505.9300 - 751us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.1714 - 757us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.0380 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.5346 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.0549 - 821us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670.5856 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.8002 - 679us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530.6573 - 877us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390.8352 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.1349 - 900us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.4963 - 1ms/epoch - 21us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.4464 - 775us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367.3054 - 10ms/epoch - 155us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.4842 - 740us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 106us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385.8281 - 2ms/epoch - 36us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444.0495 - 580us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618.7357 - 711us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491.9433 - 882us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344.6006 - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673.5385 - 740us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677.7379 - 6ms/epoch - 105us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503.1673 - 18ms/epoch - 293us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537.2356 - 1ms/epoch - 22us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542.5548 - 740us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 632.0587 - 794us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463.2847 - 829us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.9167 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.4291 - 924us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625.1128 - 691us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 253.8620 - 1ms/epoch - 20us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313.4278 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 526.1971 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295.1244 - 772us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.9443 - 702us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.5250 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520.1666 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.8104 - 767us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.9886 - 684us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 299/300\n", + "Solving for Nash Equilibrium in Generation 299/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555.9319 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237.4574 - 639us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449.5980 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 710.6464 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 605.8882 - 684us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.8586 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.6224 - 663us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462.1505 - 707us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540.0408 - 617us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.9184 - 2ms/epoch - 35us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 85us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71.6418 - 1ms/epoch - 22us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663.7889 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.3334 - 574us/epoch - 9us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9961 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468.5840 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515.6437 - 753us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.4315 - 777us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539.7919 - 788us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 705.6194 - 712us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65.4748 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606.3333 - 5ms/epoch - 80us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.3534 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.5088 - 558us/epoch - 9us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.8622 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.4404 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634.3533 - 727us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473.1196 - 787us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277.1144 - 1ms/epoch - 17us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 748.4926 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231.1246 - 653us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412.7679 - 964us/epoch - 16us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 591.4774 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.0672 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 596.6172 - 796us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.6138 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.0558 - 1ms/epoch - 18us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 687.8897 - 809us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331.9282 - 718us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.5106 - 579us/epoch - 9us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257.0688 - 652us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247.0909 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 745.9018 - 772us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703.5511 - 681us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635.3724 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.6617 - 7ms/epoch - 111us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455.6988 - 1ms/epoch - 19us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405.7538 - 623us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446.8160 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620.8641 - 762us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282.7329 - 6ms/epoch - 101us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279.5109 - 691us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.8193 - 649us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.8975 - 776us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517.5621 - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.5549 - 2ms/epoch - 38us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437.9832 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236.1839 - 689us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 551.5959 - 660us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252.5390 - 1ms/epoch - 16us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638.2511 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478.4591 - 606us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.1153 - 695us/epoch - 11us/sample\n", + "Generation 300/300\n", + "Solving for Nash Equilibrium in Generation 300/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569.1194 - 889us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292.7347 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.8500 - 1ms/epoch - 21us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595.9106 - 1ms/epoch - 21us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 447.2131 - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684.6676 - 7ms/epoch - 107us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 546.9655 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531.5606 - 673us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 341.9234 - 713us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658.2402 - 2ms/epoch - 26us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.4961 - 909us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271.3861 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475.3306 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 481.8119 - 704us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 701.7292 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.2359 - 549us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507.6337 - 1ms/epoch - 19us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490.8586 - 855us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274.6914 - 626us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.8013 - 690us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 690.1117 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 698.0403 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 253.3733 - 611us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691.4751 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650.6062 - 654us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 642.8953 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 509.9072 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 459.3140 - 3ms/epoch - 42us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293.7855 - 953us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514.0490 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538.4296 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 601.7597 - 766us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.6808 - 679us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.1360 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644.9742 - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374.8607 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602.9407 - 676us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610.3088 - 705us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294.6318 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.9164 - 738us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394.1826 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242.7150 - 920us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458.5810 - 4ms/epoch - 72us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308.0615 - 2ms/epoch - 27us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.5328 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603.1154 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448.1855 - 654us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 655.3441 - 866us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.1973 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452.1542 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 113us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485.9888 - 2ms/epoch - 30us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556.8896 - 2ms/epoch - 25us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263.8991 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484.2911 - 702us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482.4423 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521.1791 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.6540 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549.8536 - 606us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 547.8778 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 613.8491 - 845us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550.1566 - 664us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671.2555 - 6ms/epoch - 96us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534.7408 - 620us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249.6153 - 559us/epoch - 9us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511.0959 - 7ms/epoch - 120us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518.4940 - 700us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666.6506 - 780us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586.4875 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623.8998 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499.4393 - 942us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513.7951 - 689us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 734.5797 - 605us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 536.2237 - 960us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683.9996 - 713us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415.3363 - 723us/epoch - 12us/sample\n", + "Running Computed Policies\n", + "Episode 0 done\n", + "Episode 1 done\n", + "Episode 2 done\n", + "Episode 3 done\n", + "Episode 3 done\n", + "Total steps: 13\n", + "Episode 4 done\n", + "Episode 5 done\n", + "Episode 6 done\n", + "Episode 7 done\n", + "Episode 8 done\n", + "Episode 9 done\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Generation 1/300\n", + "Solving for Nash Equilibrium in Generation 1/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", + " updates=self.state_updates,\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: inf - 198ms/epoch - 3ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2559.8262 - 193ms/epoch - 3ms/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: inf - 212ms/epoch - 3ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.2522 - 204ms/epoch - 3ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: inf - 228ms/epoch - 4ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.8218 - 224ms/epoch - 4ms/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.2856 - 724us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: inf - 249ms/epoch - 4ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 220.1969 - 240ms/epoch - 4ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1334.5574 - 659us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2571.7368 - 739us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2513.5444 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2864.2744 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2641.7437 - 729us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.3757 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3448.6873 - 615us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.3159 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2899.1567 - 750us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3655.5291 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3505.3931 - 689us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1735.4155 - 712us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3564.7976 - 752us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682.4416 - 701us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3548.3989 - 700us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3447.5420 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2611.4851 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3623.7803 - 825us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.0566 - 663us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.7913 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2973.9500 - 638us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3326.7256 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2666.9424 - 701us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3099.0234 - 727us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340.7903 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2613.7993 - 805us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3099.5830 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3880.7854 - 732us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.2280 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3814.0354 - 675us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 808.6174 - 647us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2968.1973 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3748.9204 - 689us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2741.4839 - 701us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3087.1106 - 762us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2151.3992 - 641us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1360.0835 - 571us/epoch - 9us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2591.7031 - 721us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.3352 - 667us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2580.3630 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2627.1638 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2269.0669 - 757us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2417.3755 - 904us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3934.9248 - 685us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.9607 - 684us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2541.7795 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2687.2437 - 630us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2549.1108 - 2ms/epoch - 31us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2700.6287 - 738us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.6826 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2800.0359 - 698us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2326.1013 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.0104 - 734us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3340.3298 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3341.7261 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1458.1733 - 731us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3844.7366 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2636.8176 - 690us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3655.5906 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1433.9813 - 743us/epoch - 12us/sample\n", + "Generation 2/300\n", + "Solving for Nash Equilibrium in Generation 2/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.4709 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2641.8091 - 677us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.3076 - 665us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2647.0146 - 644us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2627.8816 - 781us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2540.3914 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3553.7378 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3013.8640 - 910us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3655.5439 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2630.8582 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3685.5403 - 803us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2611.5071 - 599us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3014.8447 - 633us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 222.0796 - 687us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2561.7087 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2669.7659 - 1ms/epoch - 22us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2563.8088 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3685.5383 - 879us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2546.2234 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2674.6592 - 663us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2960.2974 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2645.5449 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2567.3403 - 750us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3056.1060 - 658us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3098.8198 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2407.0168 - 679us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3217.2249 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3815.2805 - 922us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2694.5156 - 747us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2538.9343 - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2960.3167 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2094.5806 - 641us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1270.6653 - 825us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2558.8215 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2139.5356 - 635us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2287.1123 - 667us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2968.2510 - 805us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2410.7183 - 710us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2843.7803 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.7932 - 827us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2558.7764 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2421.0874 - 662us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1493.7343 - 2ms/epoch - 37us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2692.6277 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2558.3875 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.7534 - 798us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.8687 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2551.9607 - 702us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2831.1169 - 722us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.1738 - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3393.6936 - 658us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.4226 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2541.7021 - 643us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 589.7845 - 836us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1798.0844 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1462.8951 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3506.6809 - 932us/epoch - 15us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2666.7324 - 745us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 204.3427 - 761us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2179.7700 - 724us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2596.8171 - 816us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2556.6062 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1092.1888 - 699us/epoch - 11us/sample\n", + "Generation 3/300\n", + "Solving for Nash Equilibrium in Generation 3/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1336.3831 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2559.2778 - 730us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2537.0000 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2629.1294 - 756us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2973.9946 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3013.9172 - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1881.3251 - 673us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2620.1135 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3623.8083 - 654us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3209.7632 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2005.2843 - 650us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2663.3464 - 632us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3087.0850 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2625.0356 - 621us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327.9246 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3814.0532 - 647us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3098.7566 - 766us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2961.4075 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3814.0588 - 770us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3209.2783 - 774us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2968.3384 - 716us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2549.2966 - 643us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.6826 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2742.9192 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 934.4835 - 630us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.3413 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2741.4932 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2618.2134 - 661us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2350.2959 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3949.9822 - 644us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.0220 - 645us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2577.5686 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2831.1851 - 787us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3289.5618 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3393.7021 - 624us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3753.0701 - 819us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 219.5028 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359.6494 - 989us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.0706 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2831.0862 - 790us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1583.6630 - 708us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2549.3037 - 745us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2292.4272 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 619.0607 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1448.7162 - 671us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2538.4871 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2899.1934 - 735us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3651.5237 - 1ms/epoch - 16us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2608.6682 - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 830.4129 - 2ms/epoch - 31us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2168.9590 - 806us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2600.7119 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3749.0884 - 670us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2558.4199 - 671us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3623.8064 - 712us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1762.3201 - 1ms/epoch - 23us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1474.7971 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2663.7073 - 743us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3880.9399 - 935us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3334.7263 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3217.5410 - 768us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2836.8291 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2667.7458 - 716us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2690.2007 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3098.7920 - 759us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3143.7366 - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3814.0867 - 730us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2772.1033 - 769us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3880.9363 - 757us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 194.2561 - 708us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 4/300\n", + "Solving for Nash Equilibrium in Generation 4/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.7312 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2828.9712 - 721us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2667.0471 - 706us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630.4095 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2340.0388 - 684us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1173.0101 - 722us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.3030 - 699us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3934.5349 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2454.6130 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1158.4799 - 636us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2540.1150 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1250.2802 - 689us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.9688 - 633us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2633.8650 - 638us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2551.0159 - 2ms/epoch - 38us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3847.6934 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2407.2007 - 692us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3844.8018 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1434.2994 - 676us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.5427 - 797us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2684.8132 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2596.5322 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1448.4821 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2647.0308 - 711us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3651.4775 - 745us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2487.4641 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2565.2432 - 815us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3553.8499 - 621us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1662.8517 - 654us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1762.2454 - 698us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.4824 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2538.2000 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2672.0002 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2663.2820 - 677us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3218.7632 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1897.3783 - 802us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3334.8691 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2690.3665 - 719us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.1667 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3144.1865 - 640us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3209.2864 - 746us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2619.1697 - 710us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.7834 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2538.0232 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2554.6526 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3949.2214 - 646us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.4404 - 722us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2714.9973 - 826us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2539.4514 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3190.4668 - 816us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3949.3364 - 750us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.4097 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 934.2277 - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2693.0081 - 733us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2692.9893 - 791us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.0828 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.8979 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.0850 - 1ms/epoch - 17us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2831.1506 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3882.0442 - 1ms/epoch - 18us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3396.0127 - 1ms/epoch - 20us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2541.9402 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2843.8032 - 746us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.9292 - 776us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3751.4070 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.5356 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2558.1677 - 1ms/epoch - 17us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2538.7471 - 852us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.3762 - 785us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2548.1628 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2611.4409 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.8526 - 807us/epoch - 13us/sample\n", + "Generation 5/300\n", + "Solving for Nash Equilibrium in Generation 5/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.3271 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2973.9827 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2831.0952 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2599.1130 - 1ms/epoch - 20us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3393.7036 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3623.9565 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2407.3813 - 825us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2152.0059 - 950us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3753.8604 - 862us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2701.0237 - 809us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2663.1562 - 634us/epoch - 10us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.5669 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.3979 - 782us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2668.9722 - 741us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.5471 - 729us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2559.3850 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2417.8296 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2689.3347 - 952us/epoch - 15us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2556.4827 - 739us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2627.7871 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2680.4944 - 1000us/epoch - 16us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2534.9075 - 781us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3013.8792 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2686.7961 - 726us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.2373 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2741.5718 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3687.2322 - 1ms/epoch - 17us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2692.8044 - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2570.7017 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2668.6726 - 771us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2352.7107 - 707us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.1355 - 748us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2551.7908 - 650us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.9482 - 921us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3101.7896 - 811us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1240.7225 - 747us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3845.3218 - 765us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2505.4392 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.7354 - 707us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2611.4854 - 653us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.7239 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.8291 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2974.0823 - 726us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2557.2686 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.8152 - 725us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2769.8628 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2899.3103 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2601.7141 - 2ms/epoch - 35us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2549.5403 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3686.3740 - 725us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3288.9231 - 757us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2608.8760 - 651us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627.3325 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1544.9996 - 616us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2598.2625 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3814.0027 - 640us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1875.7864 - 639us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2968.2688 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2599.4573 - 682us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3624.2166 - 785us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.0298 - 656us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3623.9839 - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2663.8496 - 725us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2541.6870 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3394.8679 - 680us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3209.3430 - 663us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3750.5227 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1598.1118 - 719us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 6/300\n", + "Solving for Nash Equilibrium in Generation 6/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 873.1631 - 682us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3814.0691 - 594us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2611.0984 - 741us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241.5571 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3814.4709 - 737us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2960.1514 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266.5670 - 805us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2558.1980 - 782us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2799.6692 - 651us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 804.7482 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1081.1813 - 774us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3239.1421 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3340.4419 - 709us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2495.4844 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2628.6345 - 711us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3937.4373 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2645.6382 - 696us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1831.2330 - 716us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.8071 - 805us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.6072 - 754us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1933.6624 - 800us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.4199 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2569.1204 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3341.1675 - 708us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2596.4424 - 754us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2197.6038 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2935.6609 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3144.0564 - 747us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2540.8496 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2618.5693 - 662us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2599.2578 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2637.1506 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3950.1641 - 673us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.9080 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3623.8804 - 699us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2646.9121 - 736us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3750.9622 - 800us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2701.2698 - 589us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.5754 - 712us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3335.3445 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1554.2784 - 652us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2667.2104 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3098.7451 - 639us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2534.5850 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3547.8643 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2628.2312 - 756us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2568.9497 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1349.0000 - 697us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.4395 - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2571.7244 - 729us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2642.3420 - 703us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1983.4966 - 675us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3651.5151 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.1035 - 862us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2382.6101 - 888us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.3342 - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2546.1753 - 797us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2630.2043 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2770.1748 - 683us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2390.2712 - 761us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2564.1633 - 842us/epoch - 14us/sample\n", + "Generation 7/300\n", + "Solving for Nash Equilibrium in Generation 7/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2330.9854 - 801us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2447.7314 - 835us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.8774 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2690.1650 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464.9742 - 642us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 224.8822 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3340.3013 - 668us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2583.1504 - 640us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2573.2087 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2899.0706 - 779us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.3015 - 839us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1521.5287 - 750us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3190.8198 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2407.5771 - 786us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2899.4070 - 774us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2707.9539 - 721us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2623.7310 - 787us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.6262 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2217.6719 - 740us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2831.0923 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2974.2429 - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2627.1733 - 999us/epoch - 16us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3624.5864 - 905us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3934.3845 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2166.4714 - 880us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1335.4166 - 747us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3334.9263 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2294.7012 - 776us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2902.4902 - 973us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3505.4299 - 681us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2538.8435 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2628.1213 - 885us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2643.7183 - 1ms/epoch - 17us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3550.1155 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2564.8706 - 707us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3056.1365 - 832us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2647.8442 - 842us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3548.3811 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 197.3356 - 719us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2742.3413 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1764.0135 - 623us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.8379 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3238.8201 - 747us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1944.6090 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4019.4976 - 813us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3335.6467 - 646us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.6865 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.5576 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2667.3933 - 691us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.3376 - 806us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2584.2173 - 913us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.3530 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.0043 - 690us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2517.7573 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3238.7117 - 682us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4019.6760 - 730us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.5166 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2406.2969 - 924us/epoch - 15us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2611.2651 - 832us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2742.0759 - 644us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.3064 - 649us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2481.4207 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3239.1042 - 706us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.4631 - 775us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2554.8813 - 720us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.8892 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2557.7229 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2645.9622 - 663us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1894.7563 - 728us/epoch - 12us/sample\n", + "Generation 8/300\n", + "Solving for Nash Equilibrium in Generation 8/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2557.7224 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2558.4255 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2568.2012 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3291.0757 - 724us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.6045 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2407.3938 - 670us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2425.4810 - 875us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3935.6218 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2715.8862 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2693.3003 - 698us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.4460 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2571.5024 - 790us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2627.8860 - 667us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.9233 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3015.5000 - 672us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2382.6929 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2865.7007 - 700us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3451.2092 - 770us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2116.5129 - 690us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3652.7800 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3098.9043 - 742us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.6697 - 672us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2559.9355 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2613.5217 - 647us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2800.0139 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 686.0154 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3880.9688 - 673us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.5752 - 2ms/epoch - 27us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2152.4983 - 708us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2585.8931 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3056.3940 - 849us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1435.9576 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3751.4387 - 747us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3100.4524 - 764us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495.3725 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2588.2939 - 763us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3449.2422 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2268.6448 - 702us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2688.0400 - 4ms/epoch - 62us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3652.7776 - 2ms/epoch - 33us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.3052 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1750.1591 - 764us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2551.3694 - 677us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2597.7156 - 710us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.0513 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2548.2517 - 761us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2646.9653 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2602.5801 - 744us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1302.0691 - 838us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.8030 - 798us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2567.3064 - 786us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1996.6179 - 810us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.4915 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2569.0498 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1892.6125 - 752us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2546.6296 - 792us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2570.7324 - 728us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2549.4587 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2291.9851 - 690us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.8657 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2575.8752 - 687us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2899.3577 - 628us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600.4332 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3239.3694 - 792us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4019.2590 - 767us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2539.7454 - 701us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2248.8179 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2609.5894 - 734us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3565.0676 - 720us/epoch - 12us/sample\n", + "Generation 9/300\n", + "Solving for Nash Equilibrium in Generation 9/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2484.9783 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3441.3987 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483.4425 - 735us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.0427 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2646.9600 - 2ms/epoch - 27us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3396.5327 - 708us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.7141 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2569.6721 - 789us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.8481 - 754us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3099.5938 - 753us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2551.2009 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384.6263 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2613.7900 - 741us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1785.0149 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3144.0283 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3882.0667 - 734us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2541.3445 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2831.6733 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2620.0652 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3505.3093 - 674us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3555.6282 - 729us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1935.3884 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 578.2398 - 690us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.3845 - 901us/epoch - 15us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2577.7034 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3506.8167 - 821us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.1641 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3290.4500 - 898us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2568.9475 - 768us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494.0115 - 860us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3342.3518 - 762us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.3640 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2654.8059 - 764us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2692.2495 - 670us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3936.5767 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3846.3582 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3143.6895 - 681us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2556.9128 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1352.5739 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2658.4502 - 707us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2585.0125 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2622.2075 - 813us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3750.6426 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3336.6382 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3950.1885 - 797us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2670.2644 - 698us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2705.4016 - 793us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2533.0276 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2799.5098 - 760us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272.4679 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2570.1218 - 670us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2558.3250 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2584.2358 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2974.8228 - 755us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2573.5159 - 681us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2563.3579 - 759us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2402.8225 - 701us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2557.4146 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2769.8796 - 705us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2236.5598 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2061.1467 - 1ms/epoch - 17us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3290.6169 - 788us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2548.3904 - 756us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2612.8040 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332.0018 - 748us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3935.8250 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 809.9841 - 729us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2572.4675 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3749.2393 - 797us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2567.0481 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2865.7803 - 718us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2418.8406 - 923us/epoch - 15us/sample\n", + "Generation 10/300\n", + "Solving for Nash Equilibrium in Generation 10/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3091.2634 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2250.5928 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3015.3315 - 749us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3685.5823 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2289.8323 - 781us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2563.2693 - 743us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3222.4685 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2975.6487 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2692.2725 - 636us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2275.4944 - 972us/epoch - 16us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2704.3494 - 762us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2548.8733 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2835.1411 - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2802.7080 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2604.0549 - 782us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.9766 - 880us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.1309 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2136.2651 - 912us/epoch - 15us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3099.9287 - 730us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.2092 - 823us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2549.9114 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2632.6328 - 887us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2592.8538 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3882.5054 - 743us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3395.3423 - 973us/epoch - 16us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.5408 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2574.5266 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2691.1729 - 847us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3752.2356 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2867.8701 - 723us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2549.6348 - 774us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.4746 - 900us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.9561 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2706.6238 - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2252.9658 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2831.7671 - 706us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2937.0129 - 696us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.2688 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.8684 - 739us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2544.7656 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2590.5029 - 782us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.0093 - 3ms/epoch - 48us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2557.1379 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2182.2480 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2866.1899 - 689us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2974.5044 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3451.1609 - 795us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2254.1943 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2652.4626 - 823us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3559.4800 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 216.1114 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2604.3220 - 738us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2564.2913 - 633us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2564.9292 - 770us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2549.5557 - 712us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2648.7158 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2770.3831 - 678us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2489.3257 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2120.8647 - 819us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2602.1455 - 815us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2557.9834 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2455.2795 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2440.5554 - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3345.2327 - 757us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2539.4448 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2649.0034 - 893us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.8418 - 694us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3847.3250 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2567.6179 - 826us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2744.7791 - 859us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2558.4780 - 863us/epoch - 14us/sample\n", + "Generation 11/300\n", + "Solving for Nash Equilibrium in Generation 11/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3240.6140 - 720us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.4280 - 680us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.9861 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1679.5063 - 907us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2444.8733 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2835.8884 - 1ms/epoch - 19us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.6274 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3396.7485 - 829us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2559.2434 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3451.4116 - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2025.2251 - 911us/epoch - 15us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3759.5486 - 859us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3661.7195 - 806us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1484.2742 - 889us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2546.9082 - 890us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.5098 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2619.7168 - 1ms/epoch - 20us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2546.6646 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3950.7839 - 761us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2548.8999 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2549.4583 - 844us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2569.3240 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2713.6094 - 768us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.9358 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1650.4281 - 701us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2570.6753 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2646.5957 - 770us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3446.3376 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3061.6296 - 676us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2536.9106 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3754.5259 - 718us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2546.6230 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.0095 - 807us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3103.7021 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2617.1980 - 835us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.9785 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1893.8104 - 828us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2551.9741 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2584.3879 - 700us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2900.9890 - 730us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3508.5012 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2671.8394 - 824us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2493.6294 - 815us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2646.7175 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1091.2286 - 734us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3553.2649 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1492.7572 - 670us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.6826 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2448.6650 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2633.2771 - 714us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2692.4451 - 805us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2650.9172 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2007.6373 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 35ms/epoch - 569us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2485.5908 - 3ms/epoch - 50us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3149.8513 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3243.5667 - 856us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3882.2349 - 684us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1939.5271 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2627.0369 - 751us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712.6838 - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3058.8796 - 864us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2596.5881 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3291.3347 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3754.6304 - 646us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1611.8669 - 3ms/epoch - 42us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3938.4116 - 746us/epoch - 12us/sample\n", + "Generation 12/300\n", + "Solving for Nash Equilibrium in Generation 12/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3110.0518 - 762us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.2314 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2536.2119 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2309.7932 - 816us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2589.2446 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2904.5269 - 896us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2906.5012 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3510.4109 - 741us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2600.5271 - 775us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3549.1558 - 807us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2546.4956 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2215.8865 - 719us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2568.2415 - 771us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2511.2793 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560.3440 - 667us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2513.4187 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.1423 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2631.5220 - 1ms/epoch - 17us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2541.9663 - 637us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3018.8347 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1546.0367 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.2876 - 837us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2565.5261 - 840us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2565.7188 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2147.3340 - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 979.1265 - 816us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3107.7361 - 692us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3886.9509 - 806us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2662.9863 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3822.4771 - 836us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2680.9399 - 838us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1317.3486 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2663.8062 - 719us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2849.5266 - 668us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2771.0403 - 739us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2526.2332 - 834us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2002.7474 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1637.1458 - 836us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3245.0212 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3750.0825 - 761us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4027.8096 - 696us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2582.5994 - 858us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3096.9373 - 787us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2573.3218 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.8394 - 785us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2578.7373 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2611.6919 - 805us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2573.6375 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2977.1101 - 744us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2364.8501 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3626.2224 - 729us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2543.3997 - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2577.1467 - 755us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2865.8042 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1543.7000 - 1ms/epoch - 20us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3693.6074 - 743us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2612.8767 - 728us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3229.1877 - 863us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.1018 - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3507.4094 - 729us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2542.7507 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1097.2589 - 679us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3558.4675 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4020.6665 - 856us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1581.8795 - 1ms/epoch - 19us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2562.0625 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3556.2744 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2572.4382 - 794us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2571.1184 - 909us/epoch - 15us/sample\n", + "Generation 13/300\n", + "Solving for Nash Equilibrium in Generation 13/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2561.4810 - 719us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2564.1267 - 846us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2559.2935 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2556.8513 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2588.8052 - 870us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2600.5525 - 837us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2556.0090 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2584.3716 - 824us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2572.2085 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2942.8533 - 753us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2557.2178 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1257.5629 - 783us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553.5886 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2162.6584 - 716us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2649.9946 - 702us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2116.7993 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2587.9897 - 721us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3009.3921 - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1795.5042 - 813us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1907.1770 - 770us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2583.2578 - 758us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2693.1228 - 768us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2563.4285 - 661us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2576.2378 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2704.0608 - 762us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2808.6018 - 767us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3344.6438 - 733us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2171.5427 - 709us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238.2138 - 717us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2629.0330 - 768us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 657.3696 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3019.1013 - 725us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2576.9172 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2773.9907 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3690.5771 - 783us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3871.0254 - 717us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2540.3496 - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2595.3958 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1571.9596 - 620us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2909.6399 - 801us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2357.9392 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3516.2856 - 876us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.8340 - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2587.5886 - 997us/epoch - 16us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2590.8840 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1905.7301 - 897us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3554.2498 - 894us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2567.0967 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2570.1128 - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.1753 - 785us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2607.6660 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2648.2991 - 707us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2658.8342 - 851us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2474.5251 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1022.4626 - 757us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3068.5308 - 769us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2574.1279 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297.1777 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1969.4250 - 851us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2135.5916 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2940.8130 - 758us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2573.0737 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3566.6318 - 722us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 14/300\n", + "Solving for Nash Equilibrium in Generation 14/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2758.5212 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2582.7400 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2674.0618 - 804us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2474.8784 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3248.8308 - 908us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2583.7705 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2685.0847 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4022.7249 - 1ms/epoch - 17us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2617.3618 - 809us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2572.0386 - 743us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2596.8625 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2842.5786 - 825us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3413.2725 - 700us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2331.7727 - 746us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 935.2924 - 696us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2572.5146 - 813us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1832.6771 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2901.2043 - 730us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3507.5029 - 907us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2579.3762 - 1ms/epoch - 16us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2632.4004 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2653.4590 - 947us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2245.2671 - 705us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548.0293 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1689.2733 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2567.1599 - 880us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1794.2631 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3410.0146 - 783us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3630.7141 - 772us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2561.2678 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2674.8064 - 680us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1863.1108 - 864us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3513.6543 - 840us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2729.6145 - 842us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2649.3887 - 1ms/epoch - 18us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318.7264 - 749us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3556.0720 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2902.2451 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2686.5547 - 721us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1049.0190 - 2ms/epoch - 34us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3131.4829 - 736us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 872.7938 - 741us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2719.7195 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2629.7021 - 672us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3203.6118 - 686us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3962.7815 - 718us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2173.9233 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2097.5815 - 736us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 911.3597 - 776us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2712.2654 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2181.3240 - 829us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2590.6960 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2643.6741 - 659us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2592.6555 - 1ms/epoch - 19us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2995.5518 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370.1419 - 779us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2594.6008 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2859.0212 - 943us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1851.5032 - 3ms/epoch - 45us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2614.1809 - 775us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.4004 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2792.1194 - 752us/epoch - 12us/sample\n", + "Generation 15/300\n", + "Solving for Nash Equilibrium in Generation 15/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2332.8638 - 694us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2584.5715 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2912.2314 - 789us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3512.4033 - 716us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2586.3342 - 682us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2555.5212 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3553.2778 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3247.2322 - 770us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2598.6079 - 883us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2467.1882 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1755.3109 - 924us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2595.6997 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2835.6572 - 715us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2582.9126 - 669us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2887.5811 - 713us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378.1532 - 786us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2617.0103 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3088.3337 - 765us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2616.8652 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1569.3517 - 702us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3539.0557 - 987us/epoch - 16us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2706.7854 - 816us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3557.1824 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2314.4631 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3765.8655 - 792us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2610.5376 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2734.8669 - 754us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2580.6191 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1544.5144 - 1ms/epoch - 17us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.7212 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2726.3818 - 782us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2575.9424 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2579.0608 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2637.0706 - 803us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3219.8384 - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3957.8025 - 685us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2627.3213 - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2595.7507 - 821us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2718.5864 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1063.9207 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1397.0146 - 737us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3198.4392 - 745us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2651.9817 - 698us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3959.6238 - 696us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2708.3513 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2647.9485 - 1ms/epoch - 17us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3961.5835 - 845us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3789.0398 - 867us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2591.4463 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2745.6685 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2569.8530 - 703us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2597.6409 - 762us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2586.0000 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2628.7517 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1369.8842 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2041.9056 - 692us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2983.3411 - 813us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2721.3901 - 696us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2615.4910 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2601.0793 - 828us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2992.1897 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2563.2231 - 633us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2873.2949 - 895us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2630.3796 - 951us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1826.7679 - 737us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2088.6692 - 804us/epoch - 13us/sample\n", + "Generation 16/300\n", + "Solving for Nash Equilibrium in Generation 16/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2438.7104 - 781us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2578.2222 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2642.5857 - 785us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3457.9326 - 734us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2614.7107 - 919us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2649.0642 - 810us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 466.9705 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2752.4871 - 807us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2578.1765 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2695.0886 - 782us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3263.7664 - 781us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2636.6230 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3582.6519 - 684us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2576.2595 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2668.8723 - 834us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2641.2888 - 675us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2579.1128 - 750us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3057.5259 - 829us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2512.9521 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1825.1385 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1531.3002 - 597us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2855.4377 - 913us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3419.2234 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2694.5391 - 812us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2623.9419 - 1ms/epoch - 24us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2721.9656 - 743us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2746.2437 - 739us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1436.9806 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2209.6792 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309.2713 - 2ms/epoch - 27us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2736.6753 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1718.3245 - 856us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2583.6421 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2616.9792 - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3201.5654 - 720us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2667.7651 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2391.2080 - 851us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419.5385 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2444.2444 - 910us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2652.9246 - 956us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2631.3965 - 793us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2411.5776 - 886us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2626.7498 - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2746.9941 - 1ms/epoch - 20us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2567.5220 - 689us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2582.7949 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3202.1333 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1302.0908 - 1ms/epoch - 16us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2472.9233 - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3910.8457 - 755us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3863.7969 - 734us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2610.4116 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2745.4368 - 736us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2597.8081 - 1ms/epoch - 17us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2592.3423 - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2876.0625 - 842us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2751.3694 - 641us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393.2174 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2512.4678 - 793us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2610.0259 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2665.8198 - 893us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2847.7322 - 841us/epoch - 14us/sample\n", + "Generation 17/300\n", + "Solving for Nash Equilibrium in Generation 17/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2728.5371 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3427.4302 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3464.3359 - 803us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2644.3220 - 748us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3676.1152 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3035.4202 - 692us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2638.3950 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2620.0085 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3727.9849 - 721us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3009.9922 - 820us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2708.3147 - 857us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3292.6606 - 766us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2227.5083 - 843us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2554.4246 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2645.1772 - 955us/epoch - 15us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3093.3484 - 767us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2638.1128 - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2741.8462 - 739us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2608.1465 - 891us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1541.8387 - 1ms/epoch - 19us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2610.5547 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1298.4954 - 1ms/epoch - 18us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2669.2046 - 1ms/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.6558 - 1ms/epoch - 16us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2541.8984 - 782us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4061.1648 - 677us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2065.9280 - 877us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450.3197 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2011.7751 - 731us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 119us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2663.1138 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2730.0598 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3067.2849 - 1ms/epoch - 19us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3773.5376 - 882us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3243.4370 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3184.7854 - 1ms/epoch - 22us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3888.7188 - 782us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2666.6960 - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1603.8595 - 768us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2793.6265 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1545.1674 - 792us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3894.1628 - 1ms/epoch - 20us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2807.4771 - 843us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2626.7034 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752.6663 - 917us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3258.7285 - 4ms/epoch - 59us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3260.4558 - 756us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1814.4524 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4041.7827 - 809us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 811.6750 - 1ms/epoch - 18us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1768.4634 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2700.7102 - 1ms/epoch - 21us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3301.1501 - 723us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2370.0525 - 744us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2631.8323 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3158.0869 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2695.0830 - 1ms/epoch - 17us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1396.6804 - 712us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2551.9509 - 851us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3971.9880 - 701us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2601.7703 - 757us/epoch - 12us/sample\n", + "Generation 18/300\n", + "Solving for Nash Equilibrium in Generation 18/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2612.9407 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2177.2278 - 889us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2584.2400 - 1ms/epoch - 18us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3889.2400 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2731.1072 - 772us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2601.1335 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2615.1968 - 694us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2602.4333 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2618.7312 - 748us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2682.0801 - 629us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2603.0483 - 685us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2791.3804 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2618.9949 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1498.9857 - 708us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2636.3425 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4122.7046 - 790us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2624.4067 - 735us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2684.4126 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3117.2104 - 785us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1505.5822 - 690us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2723.7515 - 760us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3944.7849 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2683.0728 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2147.9275 - 749us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2177.5989 - 700us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2686.1436 - 785us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1105.3148 - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3201.1101 - 682us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2632.4971 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 942.7969 - 688us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2980.1804 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2653.2104 - 787us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2552.7456 - 1ms/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2664.7173 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2668.4429 - 775us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3971.3513 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2723.8186 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2779.8914 - 760us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2394.9946 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2578.6216 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3220.9094 - 739us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2876.9419 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3983.4077 - 740us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2781.9519 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2814.9880 - 792us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2674.1707 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2350.1843 - 801us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2608.1296 - 889us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2685.7766 - 918us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2615.9426 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2689.3057 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2725.3735 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2923.1438 - 1ms/epoch - 22us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3137.7107 - 936us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2958.0498 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3789.2703 - 700us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2635.7820 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2281.7756 - 720us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2949.0552 - 744us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2622.1616 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2670.7546 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3634.7605 - 742us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2743.4771 - 804us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2672.7861 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2734.5012 - 728us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2955.2209 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1721.3130 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3687.8250 - 688us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3232.4001 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3651.8420 - 748us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3367.7605 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2719.8560 - 718us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2885.4541 - 3ms/epoch - 44us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2248.8679 - 776us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2265.5127 - 638us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2791.8721 - 652us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3441.7480 - 780us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Generation 19/300\n", + "Solving for Nash Equilibrium in Generation 19/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2802.3179 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2755.2874 - 731us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3323.0801 - 927us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3948.8335 - 764us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2594.7986 - 756us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2683.1567 - 755us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2659.5415 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2647.2703 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2683.8748 - 748us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3025.9988 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2725.2332 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3076.8186 - 878us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1932.1135 - 819us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2704.7751 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3161.9504 - 809us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2690.8975 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3861.0481 - 808us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2675.6675 - 896us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2839.7078 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2640.5923 - 759us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3081.6555 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4035.6807 - 785us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3423.7905 - 834us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3835.5371 - 781us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2752.8308 - 912us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2771.7156 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2685.6003 - 729us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2681.5686 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2175.7864 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2836.0449 - 756us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3480.3337 - 770us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2699.0991 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3203.8557 - 702us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3432.3044 - 689us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2762.0461 - 679us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2682.3779 - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2261.7363 - 1ms/epoch - 22us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3115.1865 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3530.8765 - 884us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3867.4297 - 845us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 564.1490 - 692us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3830.4272 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3236.6064 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1991.2621 - 898us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2695.3054 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2696.6240 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3643.4138 - 782us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2983.5930 - 827us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2828.4954 - 1ms/epoch - 18us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3616.5115 - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3032.6960 - 951us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2728.4683 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 17ms/epoch - 274us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2691.7156 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3606.7688 - 1ms/epoch - 17us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3004.9585 - 1ms/epoch - 23us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3593.0330 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2760.4460 - 1ms/epoch - 17us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2715.6279 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2971.9370 - 844us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2696.4617 - 7ms/epoch - 107us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2707.0293 - 910us/epoch - 15us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2805.6475 - 1ms/epoch - 20us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2294.8059 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2803.6782 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2902.7097 - 1ms/epoch - 22us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3467.3936 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2157.8801 - 783us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2720.1121 - 3ms/epoch - 45us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2778.7659 - 4ms/epoch - 62us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3226.1306 - 1ms/epoch - 17us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3339.2458 - 977us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2563.9583 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2783.1472 - 2ms/epoch - 26us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4020.7827 - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2744.6221 - 750us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 20/300\n", + "Solving for Nash Equilibrium in Generation 20/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2609.2329 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2717.5598 - 7ms/epoch - 115us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2668.9160 - 1ms/epoch - 19us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4111.5195 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2671.3328 - 860us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2696.8811 - 794us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2963.9287 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2638.4053 - 872us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3226.6531 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2850.8572 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2882.7346 - 767us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2768.7778 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2889.4304 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3420.0393 - 760us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3098.5857 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1996.1057 - 887us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2849.9255 - 703us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1547.3052 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3566.4963 - 777us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2764.9512 - 9ms/epoch - 149us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3259.2441 - 944us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2604.3425 - 875us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2734.3367 - 887us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4075.9106 - 954us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2643.4819 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2664.6135 - 811us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2861.5159 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2838.1702 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1941.5302 - 767us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2903.1089 - 840us/epoch - 14us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2924.7004 - 966us/epoch - 16us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2696.7385 - 2ms/epoch - 29us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3291.8276 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1892.5530 - 714us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2321.0671 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3653.5481 - 924us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3635.6423 - 741us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2685.6084 - 750us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2743.9070 - 765us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3204.0017 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2998.7986 - 729us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1950.2697 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4021.6814 - 599us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3191.3865 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3322.0542 - 641us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3249.3740 - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3243.2825 - 1ms/epoch - 20us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3301.8113 - 680us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2777.3613 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2684.6672 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2781.7141 - 745us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2976.2373 - 761us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2830.1655 - 785us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2830.2317 - 1ms/epoch - 16us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2785.3174 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2659.8457 - 890us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3360.5042 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2727.1658 - 692us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2745.7236 - 639us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3083.6821 - 775us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3014.8738 - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3036.6311 - 912us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3303.3923 - 819us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1360.3999 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 943.8238 - 747us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2924.9490 - 705us/epoch - 11us/sample\n", + "Generation 21/300\n", + "Solving for Nash Equilibrium in Generation 21/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2957.9641 - 847us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3977.6008 - 4ms/epoch - 58us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3044.9971 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2858.8608 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2284.5120 - 792us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3137.7673 - 788us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3659.8057 - 794us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1611.1378 - 798us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3016.6272 - 861us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2982.6858 - 6ms/epoch - 102us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2356.1809 - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2828.8354 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2839.9971 - 1ms/epoch - 24us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3514.7803 - 930us/epoch - 15us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2741.6848 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2998.8843 - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3069.5081 - 773us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2669.8916 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3057.7610 - 869us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2825.1138 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2816.9360 - 688us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 919.0179 - 817us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2795.9783 - 1ms/epoch - 18us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2901.3826 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3070.5317 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3230.3616 - 653us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3003.6279 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3702.7798 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3795.5210 - 1ms/epoch - 24us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3189.9993 - 803us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3205.7693 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2764.9761 - 1ms/epoch - 17us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488.3637 - 2ms/epoch - 33us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2770.4734 - 776us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3920.6057 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2928.8013 - 1ms/epoch - 18us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2854.9448 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3218.7041 - 782us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2752.8254 - 995us/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3404.8235 - 1ms/epoch - 18us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2578.0557 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2413.7583 - 967us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2755.6611 - 895us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2757.8250 - 1ms/epoch - 19us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2934.3054 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2913.8787 - 1ms/epoch - 18us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2819.8145 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2678.1821 - 1ms/epoch - 21us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3135.6604 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3550.1252 - 989us/epoch - 16us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2545.9666 - 1ms/epoch - 17us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3862.5408 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1626.1544 - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3022.4893 - 1ms/epoch - 17us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3485.3430 - 790us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2958.9800 - 966us/epoch - 16us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3844.7480 - 1ms/epoch - 18us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2864.3196 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1352.3083 - 862us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1594.0837 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1889.1964 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3681.4910 - 999us/epoch - 16us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4176.0283 - 1ms/epoch - 18us/sample\n", + "Episode 50/50\n", + "Generation 22/300\n", + "Solving for Nash Equilibrium in Generation 22/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1979.3951 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2828.1165 - 854us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1933.6790 - 1ms/epoch - 16us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2901.6030 - 889us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1696.6385 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2347.8301 - 871us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2787.8098 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2322.1006 - 1ms/epoch - 22us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2978.0325 - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2882.4934 - 2ms/epoch - 28us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3684.7119 - 928us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 956.1506 - 851us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3156.0764 - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2656.5640 - 5ms/epoch - 74us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2745.8489 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2739.9756 - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2757.5161 - 1ms/epoch - 17us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3202.6072 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2995.7366 - 942us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3042.8494 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3798.8696 - 1ms/epoch - 17us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4098.0059 - 1ms/epoch - 20us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2691.2866 - 845us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3176.8918 - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3091.5010 - 844us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2920.6897 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3827.7910 - 1ms/epoch - 17us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2821.2734 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1751.1854 - 890us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3927.0769 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2880.2727 - 856us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3079.2957 - 766us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3864.1519 - 1ms/epoch - 18us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4042.6482 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3055.9780 - 1ms/epoch - 18us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3143.9153 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2630.5254 - 6ms/epoch - 100us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2902.7236 - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2936.2224 - 1ms/epoch - 20us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3450.7837 - 1ms/epoch - 19us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2934.1848 - 1ms/epoch - 20us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3861.2566 - 1ms/epoch - 17us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3824.5876 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2846.1753 - 5ms/epoch - 88us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2965.9468 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1883.8596 - 1ms/epoch - 18us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2804.1626 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2936.4338 - 1ms/epoch - 20us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3068.2405 - 966us/epoch - 16us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4013.3743 - 1ms/epoch - 17us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2721.7388 - 997us/epoch - 16us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3379.8394 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2628.9087 - 1ms/epoch - 20us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1954.1550 - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3079.8848 - 983us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3802.6902 - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3762.3318 - 1ms/epoch - 17us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3153.1804 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3139.1357 - 859us/epoch - 14us/sample\n", + "Generation 23/300\n", + "Solving for Nash Equilibrium in Generation 23/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1264.0729 - 962us/epoch - 16us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3454.5320 - 1ms/epoch - 22us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3437.4272 - 1ms/epoch - 20us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3038.9836 - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2846.2559 - 1ms/epoch - 17us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3453.3694 - 1ms/epoch - 19us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2467.0640 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2607.4468 - 1ms/epoch - 16us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4043.2620 - 1ms/epoch - 18us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2787.2322 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1689.0563 - 805us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2901.1948 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3195.0063 - 1ms/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2946.0032 - 5ms/epoch - 83us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3005.3813 - 1ms/epoch - 22us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2356.6389 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3196.5403 - 2ms/epoch - 30us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2866.2625 - 1ms/epoch - 17us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3640.8662 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3162.8779 - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2956.0835 - 1ms/epoch - 16us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2857.9568 - 1ms/epoch - 18us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2794.1111 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3409.7087 - 1ms/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4425.8125 - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3481.0986 - 1ms/epoch - 19us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3987.7678 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2930.8137 - 2ms/epoch - 30us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2955.7273 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2682.1377 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3136.1206 - 768us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3112.2251 - 1ms/epoch - 19us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3226.7290 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3165.5967 - 1ms/epoch - 24us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 900.9764 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2639.1748 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2933.2346 - 840us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3562.5896 - 778us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2809.3860 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2877.0374 - 819us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2713.2827 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4010.0283 - 991us/epoch - 16us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2920.5332 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2858.7454 - 972us/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2873.6582 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3011.2310 - 981us/epoch - 16us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1497.9470 - 867us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3199.9597 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3725.5105 - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3023.2605 - 761us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4159.9238 - 894us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2767.3335 - 1ms/epoch - 20us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2931.5950 - 821us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3197.5686 - 3ms/epoch - 53us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1541.5658 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3244.4365 - 771us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4141.4980 - 787us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3633.3059 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3086.8286 - 755us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2941.4280 - 738us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3313.3718 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2301.9231 - 921us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 24/300\n", + "Solving for Nash Equilibrium in Generation 24/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2588.9683 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3438.2432 - 736us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2828.6975 - 1ms/epoch - 23us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3450.3560 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3013.1448 - 892us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2692.6985 - 3ms/epoch - 42us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4205.7505 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4649.7241 - 670us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2053.3848 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1206.9696 - 683us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3251.6067 - 4ms/epoch - 57us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4333.3677 - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3424.6584 - 817us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2610.4399 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3690.3860 - 658us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2823.9663 - 790us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3924.7573 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3388.1479 - 830us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3562.2251 - 843us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3186.2898 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2112.1240 - 765us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4644.4053 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3424.3669 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3720.3350 - 643us/epoch - 10us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2547.4656 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4029.8677 - 749us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3988.6008 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3699.5713 - 666us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3270.0396 - 3ms/epoch - 42us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3022.9241 - 675us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4172.7236 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3106.7573 - 690us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3574.3738 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2123.3574 - 889us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2776.0596 - 1ms/epoch - 21us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4807.0029 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2340.6489 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3437.1628 - 751us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3331.2949 - 775us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3756.4485 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2735.6804 - 799us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2534.3716 - 903us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2540.3772 - 797us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4091.9558 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2883.0950 - 2ms/epoch - 39us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3986.5283 - 884us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3841.1372 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2673.2710 - 882us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1741.3698 - 659us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2834.8840 - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3614.8870 - 1ms/epoch - 18us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3535.3259 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2793.3110 - 946us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3268.9636 - 1ms/epoch - 18us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2576.8704 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4954.6812 - 746us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2895.3013 - 692us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4539.2026 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3051.5874 - 779us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3278.2966 - 694us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3402.7893 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2843.2261 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3741.2900 - 804us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Generation 25/300\n", + "Solving for Nash Equilibrium in Generation 25/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3091.8430 - 833us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4020.0288 - 4ms/epoch - 60us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1368.0266 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3407.3079 - 791us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3252.8367 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3746.7974 - 1ms/epoch - 24us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3122.5168 - 822us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3419.2327 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1777.9733 - 635us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3314.6501 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2870.2612 - 707us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3907.4231 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3990.4827 - 809us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4143.5298 - 962us/epoch - 16us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3703.6301 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2936.7944 - 825us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3206.5071 - 782us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4097.8467 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2881.7195 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3540.5002 - 666us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4264.4321 - 813us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3143.3052 - 729us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3383.6360 - 730us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2836.7932 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3546.6033 - 1ms/epoch - 17us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3958.5166 - 761us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3718.9460 - 915us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4108.4663 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3848.4993 - 770us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2885.1584 - 1ms/epoch - 22us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5817.1362 - 704us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3443.8247 - 982us/epoch - 16us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3030.7427 - 1ms/epoch - 17us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2642.7622 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2768.7261 - 707us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3640.3105 - 665us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3815.2939 - 837us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3605.4177 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3111.9592 - 796us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1960.5759 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3436.4436 - 1ms/epoch - 21us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3561.4270 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4072.0215 - 709us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3396.5771 - 736us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4869.4844 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3168.8582 - 813us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2812.1411 - 827us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3858.6492 - 796us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2646.2292 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2934.6333 - 960us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5333.9590 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2945.2065 - 3ms/epoch - 56us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3607.9622 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3607.7917 - 749us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1592.8463 - 797us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1627.1234 - 693us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4834.6567 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4333.5298 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4082.2026 - 652us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3115.7678 - 783us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3379.7637 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4645.6597 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3629.9045 - 583us/epoch - 9us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3091.5979 - 773us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4368.0103 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2915.0312 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3531.0869 - 638us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 26/300\n", + "Solving for Nash Equilibrium in Generation 26/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5871.7500 - 738us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2926.8430 - 756us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2988.9871 - 709us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3192.7146 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4509.2622 - 2ms/epoch - 34us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2737.9148 - 789us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4957.2026 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3035.6946 - 842us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3272.1294 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3894.1555 - 752us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3264.9058 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3243.6978 - 595us/epoch - 10us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3292.8159 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4152.8940 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3665.1753 - 688us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3974.0730 - 803us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5289.7920 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3439.3052 - 713us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2875.1208 - 4ms/epoch - 64us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2869.5339 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3275.6162 - 687us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3582.0208 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3979.2737 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3823.9597 - 706us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2883.4626 - 1ms/epoch - 17us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3446.7410 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4987.0137 - 720us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3868.3813 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2566.0911 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2292.6782 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3170.6904 - 873us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3348.0798 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4677.3184 - 616us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3449.7263 - 839us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3449.7332 - 809us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4755.3398 - 733us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3107.3245 - 633us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3214.3394 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3489.2935 - 688us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3099.2014 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4483.9546 - 722us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3069.1228 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3245.0085 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4374.2388 - 645us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4215.1304 - 759us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3288.8977 - 721us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3449.2832 - 707us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5137.5957 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3157.0171 - 755us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2378.8303 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4713.7871 - 757us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4705.6992 - 752us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3170.2637 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4027.2488 - 2ms/epoch - 28us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3655.1833 - 715us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3474.6838 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3864.2214 - 700us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6765.2910 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2771.8540 - 792us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4122.5337 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3238.0171 - 784us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6829.8154 - 770us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4679.1860 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4218.6509 - 1ms/epoch - 17us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4946.9966 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3581.7520 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4555.1592 - 833us/epoch - 13us/sample\n", + "Generation 27/300\n", + "Solving for Nash Equilibrium in Generation 27/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3847.6155 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3203.9980 - 971us/epoch - 16us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3706.1768 - 713us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4307.0454 - 780us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6556.4697 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2550.2292 - 789us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3190.7998 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3038.7480 - 995us/epoch - 16us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3221.4177 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3307.9167 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3685.7126 - 2ms/epoch - 27us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5539.0371 - 679us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3575.0403 - 694us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4537.7690 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3459.6741 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3005.9836 - 1ms/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4311.4595 - 669us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3680.5454 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3645.8486 - 903us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2711.2061 - 741us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2889.9360 - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5284.3389 - 3ms/epoch - 43us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4822.3618 - 745us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3965.3755 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3308.5281 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2908.5552 - 793us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3248.3975 - 5ms/epoch - 76us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3621.7034 - 674us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3983.7883 - 761us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4284.8569 - 735us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3668.6326 - 770us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3621.3782 - 1ms/epoch - 16us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3704.9224 - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4092.0466 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3328.9460 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5434.2935 - 915us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2791.7000 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4654.2485 - 923us/epoch - 15us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3277.3545 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3844.3042 - 895us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3360.6741 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3914.0654 - 819us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5519.3687 - 738us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2801.5142 - 828us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 115us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4628.5425 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3966.5979 - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4766.5845 - 892us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6698.7646 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4280.8877 - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3240.5574 - 706us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3349.2000 - 3ms/epoch - 47us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 33ms/epoch - 528us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5754.2480 - 11ms/epoch - 183us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4146.9902 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3551.3105 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4292.6553 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4637.1719 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2609.0691 - 2ms/epoch - 35us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4150.5439 - 727us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4104.6396 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3894.0378 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4169.9272 - 627us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2875.0093 - 866us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3714.5828 - 734us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3610.5525 - 778us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4824.8232 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9684.6357 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4012.5403 - 608us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3171.5994 - 722us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4702.5112 - 763us/epoch - 12us/sample\n", + "Generation 28/300\n", + "Solving for Nash Equilibrium in Generation 28/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3800.9150 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4502.2549 - 1ms/epoch - 22us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3505.5239 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3611.1531 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4532.8608 - 992us/epoch - 16us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4414.6699 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6472.6494 - 709us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4083.9199 - 2ms/epoch - 26us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8918.9395 - 1ms/epoch - 19us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4722.0986 - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3275.1616 - 886us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4892.1362 - 946us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3882.2056 - 988us/epoch - 16us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2776.3701 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3420.7190 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4598.3345 - 822us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4588.3354 - 1ms/epoch - 22us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6083.4775 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4453.5259 - 984us/epoch - 16us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5147.6860 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3629.0410 - 1ms/epoch - 19us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2862.7224 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7892.9072 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4253.3398 - 780us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3759.4038 - 838us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4429.4351 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3302.6025 - 1ms/epoch - 23us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6255.6880 - 871us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3101.4622 - 693us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4885.9839 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3974.5625 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2667.4949 - 1ms/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3976.1111 - 728us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9099.0869 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4204.6650 - 735us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3884.7988 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2718.8176 - 796us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5045.5679 - 744us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3004.0525 - 908us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5166.0586 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4023.3057 - 1ms/epoch - 18us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4220.5610 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3162.0996 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3289.5857 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3360.0569 - 2ms/epoch - 28us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4573.0645 - 1ms/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3885.6008 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4474.1382 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4479.7358 - 830us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3985.0769 - 734us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7044.6436 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4239.0400 - 734us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7892.2612 - 794us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4780.9248 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4354.4849 - 2ms/epoch - 26us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5020.1206 - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7961.6919 - 863us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4059.0293 - 748us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4173.6313 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7573.2344 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6606.7866 - 679us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3255.5317 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4275.7246 - 750us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6710.9180 - 768us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3470.8171 - 775us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3828.3301 - 743us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11227.7803 - 721us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5912.1948 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3423.3220 - 995us/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4518.1782 - 823us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4832.9287 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4456.6860 - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5192.9497 - 798us/epoch - 13us/sample\n", + "Generation 29/300\n", + "Solving for Nash Equilibrium in Generation 29/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3243.0022 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4897.6357 - 734us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5000.1094 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3022.2756 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5533.5864 - 855us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5749.1738 - 757us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4007.3574 - 902us/epoch - 15us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4192.2646 - 765us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4325.6616 - 1ms/epoch - 21us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11670.2441 - 782us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2938.0972 - 719us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3593.9268 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8726.6113 - 840us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6369.5439 - 765us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3001.2126 - 691us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4650.5225 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4616.0713 - 738us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8973.4980 - 973us/epoch - 16us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2788.8357 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4978.0967 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4514.0918 - 791us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5072.0366 - 2ms/epoch - 28us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2860.5625 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3726.8374 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3485.1313 - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6759.2778 - 736us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4308.5356 - 820us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3334.9338 - 983us/epoch - 16us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6764.0288 - 1ms/epoch - 20us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5621.1919 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3876.2393 - 4ms/epoch - 64us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7623.3916 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4118.6221 - 931us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3629.4465 - 754us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4332.1904 - 780us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7416.1221 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3018.9788 - 732us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3329.2036 - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3685.1187 - 992us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4980.5132 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11175.1162 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3855.3357 - 788us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4342.0947 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5107.4912 - 2ms/epoch - 36us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3848.5022 - 2ms/epoch - 32us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7849.6646 - 1ms/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4264.5669 - 956us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12283.7578 - 1ms/epoch - 18us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3202.2092 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3378.3120 - 830us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5522.6235 - 741us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3115.8840 - 893us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2353.5212 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7039.4810 - 946us/epoch - 15us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4048.6331 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4068.7842 - 799us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8918.2461 - 844us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4525.4497 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5138.9736 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5133.0161 - 938us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5039.4375 - 802us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5172.2808 - 1ms/epoch - 17us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8541.0439 - 1ms/epoch - 17us/sample\n", + "Generation 30/300\n", + "Solving for Nash Equilibrium in Generation 30/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5032.7031 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7095.5649 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3300.7795 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5977.8599 - 720us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5647.2573 - 754us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4425.8979 - 764us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8466.3301 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3942.9973 - 879us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3395.3538 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3655.9736 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5845.9751 - 1ms/epoch - 16us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11968.2021 - 2ms/epoch - 29us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6652.6367 - 859us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7068.8511 - 802us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5593.5029 - 739us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3218.1179 - 2ms/epoch - 25us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10960.9521 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4692.1201 - 2ms/epoch - 35us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3995.3149 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3483.5952 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5350.8931 - 850us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4924.6748 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4654.0156 - 685us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4528.8291 - 701us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4273.6445 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5170.4346 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3820.1785 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8054.3716 - 663us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3395.5464 - 814us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7438.1772 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3431.0432 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4207.1294 - 795us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6456.3438 - 668us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3855.6340 - 802us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6358.9946 - 663us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7563.9902 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4376.6030 - 814us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7299.8794 - 935us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6789.0415 - 1ms/epoch - 23us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4697.6313 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4041.0088 - 709us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6642.0479 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2791.9639 - 619us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3796.0798 - 668us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2807.6558 - 812us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6480.5664 - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5166.2290 - 735us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5456.7432 - 740us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5111.8110 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4180.6997 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3575.5464 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6710.1494 - 697us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7348.0161 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2924.0232 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4287.3418 - 869us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3989.6152 - 620us/epoch - 10us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5042.1348 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3818.1956 - 740us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4793.9668 - 720us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4774.6055 - 776us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10769.2148 - 841us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4496.8745 - 818us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17293.2031 - 3ms/epoch - 49us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4239.0210 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4790.1890 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7496.1245 - 748us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3688.8823 - 733us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4653.4268 - 763us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10091.3623 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4342.0898 - 721us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5273.3110 - 734us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 31/300\n", + "Solving for Nash Equilibrium in Generation 31/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16409.7422 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4327.2139 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2873.7761 - 734us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6250.8457 - 658us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10012.6660 - 681us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4898.6182 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6374.3706 - 725us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4599.6968 - 717us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3730.6382 - 713us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4627.2549 - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8042.9351 - 896us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4695.7607 - 700us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5180.0928 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4769.4653 - 733us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13034.6768 - 755us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3126.7444 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8311.3770 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3123.7322 - 4ms/epoch - 70us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4007.0276 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4550.6299 - 733us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4649.1055 - 720us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5318.2168 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4094.9966 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6368.8164 - 1ms/epoch - 17us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4507.4717 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5256.8560 - 773us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5892.1001 - 586us/epoch - 9us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4685.9160 - 620us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10613.3613 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4446.9229 - 578us/epoch - 9us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5803.9849 - 802us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4368.5415 - 717us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5386.7119 - 842us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7545.0562 - 743us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3931.8916 - 778us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6796.2178 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3821.7764 - 925us/epoch - 15us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5570.3145 - 676us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6809.6963 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3802.9314 - 669us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4376.6792 - 925us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15101.4121 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2864.8257 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4553.9170 - 800us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4932.7490 - 856us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4884.1631 - 745us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4438.6387 - 836us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4597.7417 - 1ms/epoch - 17us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5466.5698 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3337.2466 - 837us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5760.6543 - 609us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3569.7737 - 813us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4489.3271 - 715us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5175.0757 - 870us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3975.2507 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6722.9224 - 717us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3146.0737 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5737.4165 - 766us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4066.4197 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6968.6050 - 1ms/epoch - 17us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8351.7305 - 807us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4471.6592 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5497.9805 - 791us/epoch - 13us/sample\n", + "Generation 32/300\n", + "Solving for Nash Equilibrium in Generation 32/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5357.4546 - 706us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8234.8330 - 744us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2807.5642 - 636us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7286.3096 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5268.6279 - 694us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7172.5762 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5033.4282 - 705us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4818.4331 - 798us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13583.5527 - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5079.8018 - 762us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6745.4785 - 804us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5864.9771 - 713us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3679.9702 - 755us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6219.8813 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5126.5063 - 3ms/epoch - 42us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35513.1562 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5727.2598 - 794us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4294.5605 - 720us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16132.3330 - 796us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7228.6616 - 697us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6751.2007 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19476.8320 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4815.7134 - 823us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10848.4043 - 895us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5709.9365 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3377.8394 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9866.0254 - 789us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6018.2886 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17084.8945 - 746us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4120.5723 - 691us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4765.1753 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3986.5881 - 2ms/epoch - 29us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11899.6211 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4482.0664 - 732us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5377.2520 - 729us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3481.7520 - 783us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6843.4312 - 679us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7038.8164 - 744us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53302.5117 - 757us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4178.6270 - 672us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5255.9829 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3490.0693 - 929us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10253.8760 - 696us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5707.5811 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6891.9517 - 764us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6975.5332 - 779us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3541.1221 - 776us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16643.2852 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3551.3706 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5745.8208 - 787us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4517.1963 - 792us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9926.7012 - 735us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4705.1089 - 774us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6732.5059 - 765us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11353.8789 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3473.1533 - 675us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10327.8955 - 716us/epoch - 12us/sample\n", + "Generation 33/300\n", + "Solving for Nash Equilibrium in Generation 33/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5729.1602 - 687us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8128.4429 - 2ms/epoch - 29us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8663.7939 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10620.6016 - 674us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3035.1582 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8669.2754 - 770us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4348.3525 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5702.3662 - 3ms/epoch - 45us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5835.1562 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9296.5098 - 754us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4735.0537 - 703us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4206.0317 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6694.4268 - 710us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3759.5811 - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7702.6270 - 718us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5878.8389 - 703us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4059.1604 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5306.4258 - 635us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10199.3584 - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37473.5156 - 5ms/epoch - 88us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8511.6885 - 758us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5413.1226 - 810us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20837.4727 - 868us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5510.7231 - 991us/epoch - 16us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9170.4658 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9549.9238 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5599.5117 - 6ms/epoch - 90us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8678.6924 - 816us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6341.1152 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9273.4277 - 748us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4888.4536 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7180.7896 - 1ms/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12153.3506 - 636us/epoch - 10us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6531.0479 - 721us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5266.0801 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3707.0798 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29366.6328 - 646us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5059.5483 - 757us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5300.0386 - 911us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3057.3301 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4877.0522 - 996us/epoch - 16us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11996.2363 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10284.6367 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5272.4678 - 835us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8503.6543 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7757.1943 - 661us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5902.7236 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6482.3687 - 661us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26716.1250 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6688.4395 - 724us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6273.3105 - 718us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9964.7109 - 919us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3815.1975 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6950.9551 - 878us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9300.2109 - 954us/epoch - 15us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10313.4473 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6999.4702 - 837us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5128.5791 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3899.6047 - 789us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7050.8760 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8260.1855 - 598us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7558.8193 - 778us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4442.8721 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9837.7363 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5665.5010 - 840us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7788.7651 - 764us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19850.6309 - 887us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6671.1011 - 647us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18162.9824 - 846us/epoch - 14us/sample\n", + "Generation 34/300\n", + "Solving for Nash Equilibrium in Generation 34/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4764.8730 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9753.6182 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7516.7129 - 2ms/epoch - 39us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3005.4182 - 741us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9089.8545 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12353.4189 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3626.1357 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6470.8735 - 730us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4101.8726 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5068.7881 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10135.9414 - 756us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9615.1592 - 966us/epoch - 16us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8388.5273 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13121.3066 - 786us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4135.7480 - 2ms/epoch - 31us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8821.1250 - 849us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10973.3984 - 870us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8273.4209 - 695us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3654.9858 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3809.2830 - 1ms/epoch - 16us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3003.2168 - 1ms/epoch - 20us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7158.1191 - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7850.1055 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5147.8091 - 715us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5347.6650 - 750us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15361.1924 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8105.8677 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4424.2759 - 728us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6417.1123 - 792us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5920.8062 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5066.4971 - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13665.5234 - 1ms/epoch - 17us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5226.2720 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8980.5146 - 707us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5245.2559 - 796us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5674.1748 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20492.2754 - 644us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5939.6968 - 744us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4192.8975 - 677us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5444.9229 - 688us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5254.9951 - 770us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18613.9551 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6367.3076 - 897us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8166.7144 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4142.2612 - 2ms/epoch - 29us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6447.9414 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10173.4746 - 632us/epoch - 10us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10541.3525 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6959.6479 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11273.4912 - 655us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4777.2729 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6824.1665 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5305.3599 - 643us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8450.4785 - 989us/epoch - 16us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8295.0234 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9069.5352 - 3ms/epoch - 44us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2810.8093 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7099.7896 - 678us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6127.4917 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7816.5615 - 808us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20687.2656 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9476.9023 - 755us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4295.1738 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5255.5371 - 2ms/epoch - 25us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12196.4512 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7188.5723 - 804us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7398.6938 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5967.6226 - 708us/epoch - 11us/sample\n", + "Generation 35/300\n", + "Solving for Nash Equilibrium in Generation 35/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7421.5420 - 796us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9250.9697 - 721us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8833.8711 - 838us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30243.8770 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6861.5259 - 880us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5436.0195 - 699us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4872.4434 - 691us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6036.0054 - 1ms/epoch - 17us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4277.3481 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6047.2495 - 777us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8165.4336 - 819us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5852.4268 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3674.4517 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11142.2988 - 4ms/epoch - 71us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4547.9121 - 789us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14030.1738 - 859us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11600.4160 - 917us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12679.9316 - 765us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8428.4775 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6011.7964 - 760us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46970.3203 - 765us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11476.9805 - 942us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4250.7417 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12686.2617 - 787us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5526.2280 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4801.6353 - 739us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10419.7051 - 872us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4324.9395 - 754us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13142.9336 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4502.0806 - 681us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3975.8308 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10136.7998 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5843.8682 - 833us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12591.5811 - 814us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3700.0625 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6746.5454 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 113us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8017.5083 - 2ms/epoch - 29us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12659.1143 - 729us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5815.9497 - 773us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6609.2500 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10471.3672 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12268.9814 - 695us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4987.6660 - 751us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5457.1777 - 853us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4831.4648 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6913.6470 - 708us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4648.2397 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27494.9121 - 793us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13266.3730 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9760.2734 - 764us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4388.3320 - 720us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6097.8042 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4633.3115 - 756us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6805.7104 - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5240.7568 - 776us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3818.4573 - 775us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19298.1133 - 799us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9912.4033 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5567.1362 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13305.5254 - 806us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13194.7080 - 752us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8321.2207 - 886us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5858.3696 - 710us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 36/300\n", + "Solving for Nash Equilibrium in Generation 36/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8034.1636 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9392.5762 - 805us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4940.0059 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26791.7383 - 799us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16056.5068 - 743us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12519.8877 - 719us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23626.3340 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7067.2310 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3408.4419 - 1ms/epoch - 20us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7870.6914 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6222.1738 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5771.9277 - 810us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8272.5176 - 732us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5736.3149 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5566.7896 - 725us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10406.7920 - 978us/epoch - 16us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6146.0293 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3837.9460 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7568.3906 - 997us/epoch - 16us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7851.9805 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5790.3550 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10668.2617 - 818us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5236.6421 - 727us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7540.5713 - 874us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8713.4971 - 741us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5897.4521 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5191.8086 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32980.5547 - 633us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5339.8145 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6647.0669 - 942us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6283.2998 - 1ms/epoch - 21us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17978.3770 - 773us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11257.7236 - 736us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11715.6777 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8222.5127 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17875.1641 - 685us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8861.5518 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5676.1372 - 841us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5237.1499 - 891us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22896.2949 - 688us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13914.9453 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6846.8872 - 794us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8805.3633 - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6329.2915 - 764us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14082.8096 - 679us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10445.7578 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5381.1626 - 2ms/epoch - 25us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7126.3530 - 695us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6289.2407 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13684.1553 - 663us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49253.5820 - 687us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3779.0847 - 754us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11174.7002 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6078.6099 - 1ms/epoch - 17us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17492.1992 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9834.6240 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5062.6777 - 742us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7231.6973 - 755us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3872.1125 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10118.3320 - 692us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5572.4131 - 763us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33292.2617 - 869us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12336.6211 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8006.2495 - 765us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12382.3545 - 670us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4081.1099 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14721.8018 - 847us/epoch - 14us/sample\n", + "Generation 37/300\n", + "Solving for Nash Equilibrium in Generation 37/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15319.0186 - 721us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31583.5859 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9256.8516 - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6126.0161 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4871.1631 - 736us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7440.1465 - 696us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13377.0234 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6502.0947 - 820us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7707.8579 - 887us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4182.3809 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10827.9141 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6565.2085 - 752us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10949.2744 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5022.9072 - 755us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13131.0723 - 974us/epoch - 16us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11763.1328 - 809us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48923.1836 - 762us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9285.8408 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11118.5918 - 775us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7612.0298 - 680us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40514.7070 - 817us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23399.5742 - 878us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14343.4502 - 885us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10513.9238 - 820us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19103.9121 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8984.4141 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7258.7725 - 711us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13849.0215 - 745us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6467.6948 - 696us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16765.6035 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14430.7480 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16138.5459 - 694us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7565.5527 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10549.6055 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5597.9780 - 665us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11908.0146 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6223.5596 - 841us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20677.8008 - 798us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6096.9575 - 715us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9424.9883 - 2ms/epoch - 37us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6912.4434 - 774us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15694.6631 - 718us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10078.6455 - 856us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5395.6914 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10214.4727 - 866us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7448.9097 - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3009.3101 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8007.2026 - 726us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52117.6797 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6114.0708 - 730us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9893.5723 - 775us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4821.3745 - 795us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6596.4893 - 896us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12347.8516 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16211.4678 - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7309.2471 - 687us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6936.8027 - 833us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6886.0913 - 675us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7716.6157 - 687us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11435.8389 - 809us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14018.1641 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9852.1553 - 787us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5999.7686 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5671.8735 - 704us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89517.5547 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7013.3682 - 924us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Generation 38/300\n", + "Solving for Nash Equilibrium in Generation 38/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31178.6973 - 862us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9540.3438 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8043.1353 - 734us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15409.7207 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13642.8730 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7951.3618 - 681us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11278.4678 - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7106.8413 - 817us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10856.2900 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10671.3408 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34938.6641 - 976us/epoch - 16us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12011.0303 - 719us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40187.1992 - 881us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14770.7461 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13700.9434 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10729.2422 - 1ms/epoch - 20us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7987.0889 - 721us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6146.1401 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9233.6387 - 2ms/epoch - 36us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12886.3193 - 847us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9686.3340 - 610us/epoch - 10us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18352.1582 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4562.0088 - 639us/epoch - 10us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17181.8359 - 853us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10702.7422 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7843.7251 - 671us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8789.8828 - 675us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18112.6289 - 773us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4183.8032 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28453.5078 - 815us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13819.4502 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7370.2202 - 896us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6277.3457 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8646.9150 - 858us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8984.7988 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15169.9941 - 2ms/epoch - 25us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 62718.2227 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12683.2363 - 786us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9466.1992 - 783us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83596.9844 - 708us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9450.8623 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23193.6758 - 764us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17102.1973 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15354.7480 - 778us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9375.1895 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28829.7734 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18032.6914 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3126.8174 - 752us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14193.9609 - 766us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27585.9570 - 943us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12041.1475 - 751us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8685.8848 - 706us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47975.9297 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9188.3350 - 950us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8185.0854 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15180.0137 - 732us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42803.5977 - 778us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6179.9375 - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30236.6836 - 780us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8431.6475 - 650us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22033.1816 - 872us/epoch - 14us/sample\n", + "Generation 39/300\n", + "Solving for Nash Equilibrium in Generation 39/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7987.3247 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9977.2100 - 845us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18372.6016 - 788us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5156.4253 - 811us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14738.4863 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17518.3906 - 767us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28433.2832 - 720us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11092.3213 - 1ms/epoch - 19us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6809.7129 - 945us/epoch - 15us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22151.0488 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7456.7593 - 827us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7313.6670 - 795us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19377.4395 - 629us/epoch - 10us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9446.2100 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15308.0186 - 767us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18456.7930 - 757us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9050.5410 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7319.1162 - 1ms/epoch - 18us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24145.7715 - 774us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18896.1211 - 786us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7392.0215 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6584.9790 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16892.9238 - 927us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15886.1426 - 3ms/epoch - 56us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4703.2480 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8697.7715 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18037.8887 - 692us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9515.2686 - 790us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10323.6357 - 723us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23910.2480 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5839.8662 - 715us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7505.2905 - 845us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10747.3662 - 719us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59960.6836 - 708us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15384.6455 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7025.9429 - 596us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9137.1855 - 781us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18502.7324 - 691us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23479.3379 - 765us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7084.8755 - 824us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35813.2695 - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16178.6621 - 1ms/epoch - 20us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7601.9707 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27976.1328 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9863.6270 - 714us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13471.8389 - 745us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24984.2734 - 865us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22602.5156 - 1ms/epoch - 22us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11605.8389 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15814.3799 - 886us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10481.0615 - 684us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9551.5303 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11026.1035 - 723us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8556.2734 - 2ms/epoch - 35us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10525.3604 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8017.4819 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9042.8955 - 664us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50969.0859 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15684.8770 - 724us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7105.0669 - 657us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9574.5303 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11834.4688 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9210.0029 - 858us/epoch - 14us/sample\n", + "Generation 40/300\n", + "Solving for Nash Equilibrium in Generation 40/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17503.0605 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7491.8545 - 976us/epoch - 16us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24935.9609 - 687us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8230.6221 - 748us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16553.7402 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20547.2285 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9431.3662 - 647us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36335.4297 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19696.1504 - 862us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17017.7910 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9498.7324 - 725us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11055.6230 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13157.8564 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9993.2061 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10681.3643 - 698us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26522.1152 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17440.4980 - 733us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7569.4727 - 722us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18674.9824 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19276.0391 - 2ms/epoch - 26us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6674.6538 - 784us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19811.6426 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7454.3965 - 792us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5469.3770 - 2ms/epoch - 28us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8003.5063 - 768us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3456.7595 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41817.1211 - 742us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8161.9004 - 746us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6562.2544 - 828us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38719.9805 - 747us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7638.7910 - 4ms/epoch - 61us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20342.6055 - 669us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17529.3516 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13126.7354 - 713us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9299.5879 - 812us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27266.7617 - 781us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7102.1909 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30074.6094 - 712us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11331.5566 - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6268.3477 - 806us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17289.2500 - 710us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13946.9072 - 710us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13979.7529 - 669us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4910.6519 - 667us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9467.8398 - 736us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47554.8359 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11456.5986 - 763us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9871.2734 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9433.5684 - 1ms/epoch - 17us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46110.5312 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6656.2256 - 740us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10907.0479 - 753us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4755.2202 - 971us/epoch - 16us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21559.3555 - 711us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17001.0371 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10514.4326 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7414.1406 - 813us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13290.6084 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9572.0283 - 6ms/epoch - 95us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53431.2344 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5413.4912 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22593.8418 - 887us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7048.4438 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28666.0410 - 1ms/epoch - 17us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12470.4766 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 99us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7781.1782 - 901us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21257.2715 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11343.9199 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66264.6328 - 969us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24776.8789 - 825us/epoch - 13us/sample\n", + "Generation 41/300\n", + "Solving for Nash Equilibrium in Generation 41/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12769.9033 - 799us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6775.1582 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9638.5908 - 723us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9159.9980 - 703us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23423.5020 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6830.4131 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12175.2178 - 823us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6374.6152 - 734us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18455.7344 - 1ms/epoch - 18us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20033.3750 - 617us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18249.1895 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 96328.9844 - 729us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8942.9375 - 769us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30101.4473 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16503.2715 - 900us/epoch - 15us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19553.7539 - 837us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23581.7891 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15529.1523 - 834us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6493.6904 - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22274.5000 - 754us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37657.8047 - 766us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30231.3457 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5358.3594 - 2ms/epoch - 32us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54204.0742 - 899us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10357.3857 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31044.3496 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9647.7412 - 776us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23089.7949 - 861us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24053.1621 - 861us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13226.3340 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10184.3896 - 3ms/epoch - 47us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24946.2637 - 891us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11252.6064 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22573.0078 - 970us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8067.2754 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65271.0664 - 729us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12132.1191 - 2ms/epoch - 27us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33006.7148 - 767us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16098.7227 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11157.2773 - 858us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12930.7725 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11510.9209 - 859us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20253.7676 - 1ms/epoch - 21us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51147.9258 - 1ms/epoch - 19us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4661.2227 - 827us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 75554.4688 - 825us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42105.2773 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33098.5352 - 930us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33048.1836 - 678us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23751.4883 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34147.8516 - 784us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7339.9731 - 4ms/epoch - 64us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12251.4912 - 706us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27049.9941 - 743us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21487.8984 - 772us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10831.8965 - 838us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14374.4287 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22979.6758 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9472.3945 - 828us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74888.9297 - 1ms/epoch - 24us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12169.8047 - 719us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43867.4727 - 762us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14196.1260 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13783.5205 - 727us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44535.4492 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40520.5586 - 777us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12396.1045 - 810us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 70811.1797 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12672.0488 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27721.4980 - 731us/epoch - 12us/sample\n", + "Generation 42/300\n", + "Solving for Nash Equilibrium in Generation 42/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22497.4766 - 775us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6891.0151 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65587.2578 - 650us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16571.4219 - 672us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7016.4111 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36100.0938 - 712us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11924.0645 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40184.1797 - 732us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23500.5156 - 733us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10127.2480 - 727us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20070.5879 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17752.6387 - 782us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88226.0625 - 660us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25302.6816 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8081.6929 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37791.1211 - 759us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11013.8213 - 666us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40333.3047 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10753.9121 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53482.5938 - 689us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14335.5820 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8939.4277 - 693us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39089.3320 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13487.1963 - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5645.1060 - 1ms/epoch - 22us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32228.9590 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11398.5801 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6083.7065 - 764us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45665.5547 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6808.0923 - 745us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8570.3047 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24079.3281 - 772us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35491.6328 - 689us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8584.3545 - 971us/epoch - 16us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29596.0156 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11825.8750 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15843.7207 - 724us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56053.5938 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37319.7500 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11374.9863 - 712us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47604.0391 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25214.8594 - 701us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27702.9707 - 2ms/epoch - 30us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17812.0625 - 740us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5251.9399 - 668us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19446.8125 - 652us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47844.7617 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21917.7363 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16643.5820 - 684us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28283.4336 - 777us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10081.5264 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15247.2334 - 703us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7763.7964 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26660.7324 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 84318.4844 - 696us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28162.8535 - 619us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15118.0869 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17739.3496 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126650.4062 - 672us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48244.7461 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25903.6035 - 833us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3969.0500 - 711us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28354.1328 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46388.1172 - 841us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Generation 43/300\n", + "Solving for Nash Equilibrium in Generation 43/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11147.3428 - 662us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55722.5781 - 743us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13615.4238 - 806us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29609.5215 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14964.6250 - 2ms/epoch - 34us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13877.2559 - 697us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30347.1328 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23186.8320 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15555.6436 - 681us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51167.6680 - 2ms/epoch - 32us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36612.1836 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20927.7090 - 697us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41738.5195 - 880us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47018.5508 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54506.8438 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30796.6855 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17349.1660 - 3ms/epoch - 52us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 18ms/epoch - 285us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30513.7793 - 4ms/epoch - 58us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5476.1445 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11526.7617 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47990.4570 - 939us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7623.9341 - 848us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18150.0508 - 742us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19092.6895 - 733us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14185.6895 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21823.5195 - 2ms/epoch - 32us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61571.3555 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11428.5020 - 709us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10844.2314 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19509.3242 - 725us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15520.3086 - 982us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36317.2578 - 693us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29199.6895 - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50561.0195 - 885us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4697.4131 - 761us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13236.9600 - 665us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 190727.5625 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8892.9854 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32043.9434 - 837us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11585.9043 - 761us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112752.3594 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35503.1289 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19912.6895 - 795us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20006.2012 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32215.3594 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25436.3457 - 710us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14944.4912 - 722us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25243.5840 - 2ms/epoch - 27us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13462.1562 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50117.3750 - 779us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15878.3311 - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35950.0859 - 589us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6396.3564 - 755us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12011.7490 - 648us/epoch - 10us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12724.9609 - 908us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 225398.4219 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26264.1738 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6937.5464 - 1ms/epoch - 20us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 44/300\n", + "Solving for Nash Equilibrium in Generation 44/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3444.0354 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19736.2363 - 755us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9709.5352 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19080.4922 - 756us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 93463.6953 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7643.8931 - 798us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35681.7266 - 733us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52683.3633 - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21693.1289 - 984us/epoch - 16us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21876.4375 - 715us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 82342.1641 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11678.6650 - 2ms/epoch - 27us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8942.0977 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22081.1309 - 750us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20991.0332 - 796us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18297.2969 - 774us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 120308.5625 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17706.3457 - 783us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53172.8047 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10305.9355 - 636us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 86445.7422 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4214.4385 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18648.9414 - 652us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36468.9609 - 807us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36074.9805 - 732us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61707.3672 - 2ms/epoch - 28us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33585.6836 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15206.5576 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54878.5117 - 681us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9561.8555 - 723us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4959.1841 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18956.3496 - 828us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20672.6074 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16495.4727 - 974us/epoch - 16us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 210610.0156 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22293.4258 - 649us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22068.1992 - 852us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16362.2402 - 710us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36625.4844 - 725us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49192.7812 - 759us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20912.6719 - 668us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 115037.1484 - 731us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26282.0039 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45587.4570 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18499.4531 - 738us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49378.7227 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12437.4805 - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27699.1484 - 778us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6745.6694 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15480.5566 - 734us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5492.2090 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31073.9492 - 723us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31533.9668 - 659us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16289.1436 - 713us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11898.0303 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 57100.9805 - 1ms/epoch - 21us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16685.3613 - 700us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27922.7949 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14612.4150 - 720us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123261.5547 - 908us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23249.1328 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34982.0781 - 745us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 127022.3047 - 625us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26955.6348 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17471.3477 - 678us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 75299.9453 - 690us/epoch - 11us/sample\n", + "Generation 45/300\n", + "Solving for Nash Equilibrium in Generation 45/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19612.8672 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38943.3594 - 892us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20438.7363 - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25778.9492 - 742us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74064.5312 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35263.5273 - 746us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8731.6377 - 732us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37345.7266 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8036.5522 - 701us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17459.7695 - 2ms/epoch - 25us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20423.3125 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21164.3555 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15724.8623 - 2ms/epoch - 26us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27158.3027 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34458.9141 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20400.2734 - 764us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3572.1682 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23222.8672 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17185.4883 - 1ms/epoch - 17us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 60672.3086 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20558.1250 - 7ms/epoch - 111us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18473.1816 - 849us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7955.4478 - 648us/epoch - 10us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 206942.5781 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24811.1113 - 711us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39874.0234 - 667us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28311.5996 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19778.8105 - 756us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21037.5195 - 770us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42337.1094 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13337.7471 - 800us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38074.5195 - 738us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18551.8809 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13793.8086 - 815us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123729.3516 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24578.2344 - 792us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10327.7646 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41060.6016 - 601us/epoch - 10us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 75502.1719 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9757.0801 - 646us/epoch - 10us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66813.5859 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11637.2559 - 772us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22732.2832 - 959us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56322.3359 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34693.0430 - 777us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 199774.6562 - 699us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26060.9727 - 754us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 80249.9219 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40219.6094 - 591us/epoch - 10us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15468.0039 - 3ms/epoch - 41us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14861.5020 - 630us/epoch - 10us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59742.8242 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90777.8047 - 1000us/epoch - 16us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18727.4414 - 724us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40805.5078 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15511.1836 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16733.9648 - 638us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 76389.6953 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24922.9512 - 818us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52686.1133 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4817.9116 - 976us/epoch - 16us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 76995.8594 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41337.4609 - 804us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13539.9775 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10793.5381 - 885us/epoch - 14us/sample\n", + "Generation 46/300\n", + "Solving for Nash Equilibrium in Generation 46/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 80037.8281 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31768.5332 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17511.0254 - 645us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19548.0840 - 728us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45298.6758 - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36109.4805 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45856.6953 - 801us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13516.5322 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43778.0039 - 642us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15093.3213 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34937.5781 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11945.1445 - 645us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30929.3496 - 775us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19978.9766 - 743us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52580.4258 - 764us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 118706.9219 - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13034.4639 - 680us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31181.7734 - 738us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32723.3301 - 755us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33497.8984 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27564.4258 - 1ms/epoch - 16us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 154392.4062 - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34604.1641 - 802us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51476.8711 - 877us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14994.9795 - 725us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13035.8105 - 1ms/epoch - 24us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 81854.3594 - 801us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14761.1260 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46191.5547 - 1ms/epoch - 20us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33904.6250 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 120657.4141 - 1ms/epoch - 18us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16703.4844 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48101.8086 - 857us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 110338.8516 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20733.7930 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35038.2773 - 709us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20204.4805 - 615us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56202.1562 - 745us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18899.0586 - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15091.5762 - 839us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19086.2559 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26393.4082 - 761us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28598.2695 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16484.1836 - 730us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22731.5586 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 185423.7812 - 640us/epoch - 10us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9228.2744 - 991us/epoch - 16us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19571.7598 - 705us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22060.7109 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9737.1250 - 809us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46322.6836 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45459.6328 - 713us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23460.0078 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32219.2148 - 906us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13947.6475 - 943us/epoch - 15us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4857.5894 - 2ms/epoch - 28us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 82740.2188 - 3ms/epoch - 43us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48442.6641 - 694us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20718.3066 - 778us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27155.0781 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47293.5156 - 853us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9055.9688 - 2ms/epoch - 27us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17579.4453 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61148.4922 - 929us/epoch - 15us/sample\n", + "Generation 47/300\n", + "Solving for Nash Equilibrium in Generation 47/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59682.5352 - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21505.5605 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89297.8438 - 745us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24218.9434 - 738us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25444.6582 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22323.1016 - 860us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24352.8945 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21043.6602 - 739us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12851.0273 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47929.1914 - 830us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 124434.5000 - 761us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95915.9453 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22991.1094 - 709us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27733.0664 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25469.5039 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16943.6348 - 750us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 145379.0000 - 731us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16740.0020 - 863us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69312.5078 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18244.1172 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101886.1484 - 677us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77911.4922 - 913us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12790.1055 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54729.0000 - 843us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8703.0215 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79131.0156 - 3ms/epoch - 42us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38174.3945 - 687us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39147.0078 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18111.6758 - 729us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21979.9297 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51220.2930 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26499.9766 - 794us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48891.9492 - 713us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53419.8945 - 645us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26344.6582 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33014.5742 - 667us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26874.8535 - 816us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53076.6094 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 92us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41028.5312 - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 62365.9336 - 761us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23614.7344 - 714us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 150385.8906 - 662us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14769.0146 - 778us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40619.1445 - 745us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55149.8711 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38543.7656 - 692us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25316.2363 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88472.2109 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53482.7500 - 790us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17693.3848 - 744us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50669.9453 - 894us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18499.3770 - 670us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27325.2344 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9243.1982 - 880us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24186.2168 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56651.2266 - 629us/epoch - 10us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41128.5586 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14677.5811 - 715us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55622.3047 - 763us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23489.6133 - 718us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21289.4238 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29104.9805 - 956us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25548.4941 - 3ms/epoch - 52us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3460.2708 - 794us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23121.4023 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17296.5195 - 677us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Generation 48/300\n", + "Solving for Nash Equilibrium in Generation 48/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28361.2285 - 866us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16111.1436 - 761us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33165.3789 - 2ms/epoch - 39us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116523.3906 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19113.5234 - 714us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106122.6484 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25172.9336 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19121.3516 - 1ms/epoch - 17us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40786.5781 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42580.5703 - 723us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27084.1621 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21423.0762 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43498.6133 - 3ms/epoch - 49us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 84256.3203 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49635.4023 - 726us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24374.5254 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30943.4023 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 87209.4375 - 718us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29791.6406 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44448.5547 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26056.8242 - 692us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37896.7773 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45599.4688 - 756us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21162.2207 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 97883.5469 - 702us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19653.0645 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27640.0195 - 712us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54627.7305 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30367.1094 - 671us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4882.0537 - 740us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47647.7578 - 642us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 190140.3906 - 882us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24802.0254 - 3ms/epoch - 43us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18029.6602 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 167971.6094 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46681.3711 - 736us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19407.8789 - 546us/epoch - 9us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32014.9805 - 788us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21754.0547 - 722us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 169447.7031 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44963.4531 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26362.8145 - 773us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16523.2031 - 755us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 72414.8516 - 758us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 226386.0625 - 667us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44129.6914 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50510.1797 - 729us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48988.0391 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 82548.5000 - 766us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16181.6924 - 772us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59693.9180 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10199.2295 - 764us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19964.0508 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26089.6816 - 739us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25315.5215 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 93155.4922 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29042.9844 - 712us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17284.2773 - 660us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29315.1270 - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22236.8613 - 764us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46359.4727 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34035.3555 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50182.1914 - 739us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29469.4453 - 718us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65050.1328 - 2ms/epoch - 26us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 91036.2734 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25474.2480 - 674us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23430.7949 - 698us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112349.6953 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27795.2363 - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245843.2656 - 963us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13072.2617 - 854us/epoch - 14us/sample\n", + "Generation 49/300\n", + "Solving for Nash Equilibrium in Generation 49/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 115640.2031 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 154215.3750 - 700us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61483.3711 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38578.5234 - 3ms/epoch - 46us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28411.1836 - 725us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 62230.2266 - 758us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11121.7617 - 888us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5975.4858 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19093.1270 - 709us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30305.0762 - 760us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33153.5469 - 696us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55616.2656 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17175.6484 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304262.2812 - 675us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 63483.4102 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40848.8594 - 806us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44239.3438 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 100306.2656 - 798us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12838.0195 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79378.5156 - 779us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34001.7266 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 203179.8750 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31246.1230 - 933us/epoch - 15us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41728.7031 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17169.1289 - 976us/epoch - 16us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 60129.2734 - 804us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 73817.5547 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17630.1035 - 664us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4064.9980 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 218460.8125 - 727us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38379.5820 - 706us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55916.8984 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34447.7656 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36044.3633 - 787us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42562.4688 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66347.7031 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31676.0273 - 751us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65545.2188 - 706us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 162736.6250 - 703us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 63312.8711 - 862us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24989.7266 - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32627.9258 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56549.9922 - 736us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34822.8672 - 685us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98183.3594 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19591.5117 - 784us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30356.0898 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69464.3594 - 974us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 67266.4766 - 788us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37085.7773 - 851us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69861.6406 - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11638.6357 - 842us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20586.3496 - 753us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11921.1104 - 800us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18632.1484 - 1ms/epoch - 22us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31368.2969 - 832us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11627.7051 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34307.4570 - 802us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65877.3359 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48242.5312 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 149585.3438 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 129368.0703 - 716us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4667.2388 - 869us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 99915.8281 - 1ms/epoch - 18us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 130416.6172 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27669.5879 - 744us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33822.8594 - 657us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18753.9590 - 850us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32795.6758 - 866us/epoch - 14us/sample\n", + "Generation 50/300\n", + "Solving for Nash Equilibrium in Generation 50/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 57610.2070 - 707us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5045.0840 - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 213480.0156 - 1ms/epoch - 19us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71042.7188 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31736.6621 - 656us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95727.4688 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11230.2861 - 739us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39082.1602 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 173233.7344 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50700.2852 - 695us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14844.0684 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33489.8008 - 722us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25680.1211 - 3ms/epoch - 40us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 64228.3984 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28041.3887 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 57343.0938 - 665us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 210462.9062 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 92112.6641 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32472.8418 - 749us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40483.7734 - 807us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33137.2109 - 666us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83480.4531 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16439.2969 - 783us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20720.5781 - 845us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42912.1719 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 204656.7656 - 747us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46602.5352 - 684us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17349.2871 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101183.7031 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5603.8833 - 612us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33291.6641 - 812us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28835.2090 - 727us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 198087.4062 - 704us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37750.8594 - 3ms/epoch - 45us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 192721.1250 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74028.7031 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39572.2422 - 830us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74327.4688 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19150.0449 - 774us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 109699.0859 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68210.2188 - 871us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29436.5469 - 802us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19385.1855 - 707us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18571.6895 - 688us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 70073.1094 - 730us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357060.1250 - 749us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 80527.3906 - 695us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 192698.4531 - 702us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 202162.6250 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14424.3975 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10220.9424 - 749us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 225387.1406 - 706us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47903.7070 - 744us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45098.0156 - 689us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 35964.3359 - 688us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116303.3359 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68523.0234 - 757us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69714.4141 - 748us/epoch - 12us/sample\n", + "Generation 51/300\n", + "Solving for Nash Equilibrium in Generation 51/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25317.1523 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40102.0508 - 694us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322268.5312 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 75522.3203 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41220.8281 - 660us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90899.7734 - 706us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39552.0000 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38130.7344 - 797us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48682.1211 - 792us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68621.9453 - 718us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66798.6875 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41024.6484 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20104.2227 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256102.7656 - 762us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38204.2852 - 728us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46321.9102 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17441.9570 - 660us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 118612.4922 - 704us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27611.1777 - 762us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28419.1113 - 816us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54531.4805 - 934us/epoch - 15us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261465.0781 - 738us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42661.9570 - 916us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 120872.7656 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17740.0039 - 736us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 128352.2188 - 685us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46035.5781 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44104.5000 - 740us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 140429.0312 - 810us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271289.5312 - 1ms/epoch - 18us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355055.4375 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40531.8594 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20073.7422 - 649us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236884.7969 - 5ms/epoch - 81us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41226.4414 - 658us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 183941.7188 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20696.6133 - 738us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21246.5449 - 732us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 132530.4219 - 675us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55385.5352 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41939.1602 - 773us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29024.2344 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314181.7812 - 653us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27335.4199 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106925.9766 - 742us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 81269.5000 - 731us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28361.5527 - 695us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473928.7812 - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77343.5938 - 769us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 117728.8672 - 1ms/epoch - 18us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14483.3066 - 877us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 178345.5469 - 709us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 94662.9453 - 729us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45709.0391 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38976.9570 - 1ms/epoch - 22us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26386.6973 - 726us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 91351.4062 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37854.4961 - 2ms/epoch - 33us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52597.7812 - 716us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19141.1641 - 688us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267341.4375 - 719us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55953.6445 - 715us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24401.5137 - 682us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31022.6992 - 717us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29763.0410 - 760us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44387.7656 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24936.5996 - 656us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 110628.9453 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89032.0234 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22126.8652 - 732us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 52/300\n", + "Solving for Nash Equilibrium in Generation 52/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43651.6523 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 75889.6562 - 732us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8403.0352 - 664us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 76711.3281 - 2ms/epoch - 30us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 82785.6641 - 711us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14553.5332 - 801us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638370.3125 - 775us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90285.1328 - 662us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 195747.4688 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 85537.1641 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123310.0312 - 1ms/epoch - 18us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 86859.7188 - 837us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7419.1699 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95079.7969 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41836.0781 - 755us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20466.1406 - 836us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47010.2305 - 719us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34530.9297 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25402.5449 - 684us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98616.5781 - 706us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26077.5996 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26786.0215 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 82774.8672 - 719us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126990.2422 - 4ms/epoch - 65us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41986.7539 - 718us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54695.5508 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89931.5078 - 649us/epoch - 10us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23196.0059 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26746.5078 - 792us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 159901.9219 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26885.1250 - 707us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 122340.3828 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 145242.2188 - 932us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53725.9336 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166013.6875 - 839us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 142569.9531 - 1ms/epoch - 22us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37821.9453 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46735.6641 - 843us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31867.4727 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79114.4297 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 103081.6641 - 881us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 120661.5859 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32240.8613 - 875us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 80232.6484 - 752us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14711.6953 - 782us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32762.3652 - 723us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65062.0703 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27796.2305 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 187197.6719 - 924us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20351.3340 - 1ms/epoch - 22us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 139637.8438 - 3ms/epoch - 41us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 85493.8438 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401186.5312 - 724us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 63314.8789 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22063.0078 - 788us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47999.2227 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17429.7773 - 867us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 113743.7578 - 688us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 190951.9062 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23814.2695 - 810us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 119343.5859 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33821.2109 - 769us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43302.9180 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43484.6133 - 831us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11180.4072 - 6ms/epoch - 104us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15068.9102 - 715us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 133786.9375 - 632us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49620.5664 - 769us/epoch - 12us/sample\n", + "Generation 53/300\n", + "Solving for Nash Equilibrium in Generation 53/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30859.9648 - 5ms/epoch - 76us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 105688.7891 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 121711.4297 - 835us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 129108.5625 - 753us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243210.1562 - 603us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44399.8594 - 955us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49418.2852 - 641us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42749.2695 - 756us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83491.2422 - 723us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51149.6094 - 796us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45268.9219 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41983.9453 - 803us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46331.7617 - 1ms/epoch - 21us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30468.8301 - 731us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53495.2422 - 741us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61569.1797 - 1ms/epoch - 20us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15256.3936 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357373.9688 - 1ms/epoch - 22us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33196.3633 - 837us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33297.6836 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41032.1797 - 727us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 183764.6875 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248003.5938 - 741us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8667.3691 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301964.6250 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 141212.5469 - 3ms/epoch - 41us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25421.5137 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46599.6172 - 694us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112029.9844 - 824us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45052.0430 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26069.2402 - 713us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 198162.1406 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89439.1953 - 803us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42098.1211 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 234652.5781 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 138930.5000 - 1ms/epoch - 17us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101971.6484 - 803us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 119235.1797 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31305.9844 - 842us/epoch - 14us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55243.5117 - 665us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 107120.5391 - 849us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24450.2656 - 686us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9901.6299 - 782us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 87029.3672 - 776us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 99835.0938 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 64489.5703 - 931us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54577.0000 - 700us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65616.9141 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 103809.1484 - 786us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 154166.0781 - 694us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 93682.0625 - 721us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47389.0508 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 62060.3867 - 715us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15195.5693 - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106263.1953 - 750us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65313.1719 - 756us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31800.3594 - 2ms/epoch - 29us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 155976.7031 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 148478.4375 - 711us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95248.1875 - 739us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 120450.2422 - 775us/epoch - 12us/sample\n", + "Generation 54/300\n", + "Solving for Nash Equilibrium in Generation 54/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43828.2812 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 102959.9219 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56501.3359 - 868us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422271.8125 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 143701.0781 - 738us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 78479.3203 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49708.4648 - 669us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44442.6758 - 1ms/epoch - 16us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 99849.4766 - 705us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 62612.2500 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562431.2500 - 783us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 18858.6543 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49301.2305 - 791us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6901.9038 - 939us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31593.2676 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46074.8242 - 730us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52818.6602 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98731.9375 - 745us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39408.2773 - 631us/epoch - 10us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24550.7266 - 821us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 140747.1406 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16407.8770 - 740us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65249.1523 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66612.7031 - 717us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17230.1934 - 929us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 99346.7734 - 770us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274707.2188 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33817.1328 - 701us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166221.3906 - 679us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44766.6211 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48086.6992 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 217649.6719 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 167935.6719 - 713us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24356.4336 - 2ms/epoch - 30us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51779.8438 - 760us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344271.7188 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52917.9609 - 625us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25088.8203 - 839us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 118372.2891 - 732us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52941.5430 - 833us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284978.0625 - 870us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 422269.9375 - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38464.7891 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 105439.7031 - 661us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19222.6992 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 73712.1172 - 744us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235695.3594 - 718us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 70889.2188 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26355.5156 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104077.7656 - 753us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8558.0996 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83900.4375 - 752us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299630.7500 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 135036.7344 - 680us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276830.9688 - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 107612.3359 - 721us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21601.2070 - 678us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 191458.2812 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29654.1875 - 710us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 122297.5938 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106894.9844 - 652us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338941.7812 - 688us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22142.7383 - 4ms/epoch - 59us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106863.8516 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44515.7227 - 888us/epoch - 14us/sample\n", + "Generation 55/300\n", + "Solving for Nash Equilibrium in Generation 55/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13162.6377 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54966.8164 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236415.0625 - 798us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27014.1113 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30567.0078 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71090.7031 - 3ms/epoch - 41us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 57118.2695 - 774us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71017.9922 - 747us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116127.6875 - 2ms/epoch - 28us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 185228.0312 - 662us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44289.5859 - 630us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 111512.7188 - 788us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 169859.7969 - 669us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 184371.1406 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41719.6797 - 680us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244520.6562 - 758us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66331.6562 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36434.0547 - 715us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 129075.9453 - 722us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20367.6641 - 699us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 122110.8125 - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106438.5781 - 687us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 114280.2266 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17868.5156 - 798us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116145.5781 - 734us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69751.3359 - 788us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 76239.4219 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430135.2188 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32566.8242 - 2ms/epoch - 26us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 186801.8750 - 618us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38049.0977 - 717us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61978.5469 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 129554.2734 - 725us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27538.2832 - 817us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66743.6328 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 67487.4844 - 644us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 140910.5625 - 816us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352903.5625 - 767us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24518.4961 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98911.4453 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 57225.2812 - 2ms/epoch - 26us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 191931.7812 - 723us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14340.4951 - 1ms/epoch - 19us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29007.6348 - 734us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232585.0000 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37287.5078 - 671us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68978.8516 - 658us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83104.6953 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69298.8594 - 2ms/epoch - 31us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293330.7812 - 745us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 75472.5703 - 865us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 70587.1641 - 775us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 63443.0703 - 663us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 58628.8320 - 4ms/epoch - 63us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 122289.4297 - 711us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272520.0000 - 688us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 57875.0938 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49128.1094 - 720us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201374.4219 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335432.9375 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61900.4922 - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303214.9688 - 1ms/epoch - 18us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46432.2031 - 762us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 198081.5938 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59591.9180 - 2ms/epoch - 32us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59021.7891 - 756us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41647.8281 - 770us/epoch - 12us/sample\n", + "Generation 56/300\n", + "Solving for Nash Equilibrium in Generation 56/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 150890.8438 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71818.4453 - 710us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5863.6025 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 70627.7266 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 64126.7695 - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20182.4980 - 725us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 149832.1562 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 102524.8672 - 796us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 154019.8125 - 800us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56442.0273 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 179711.8594 - 765us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29929.2285 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 137958.1875 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 142131.5156 - 680us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 207220.2344 - 702us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 15856.2979 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 189689.8906 - 2ms/epoch - 35us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66021.0625 - 798us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214488.5312 - 758us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 119961.1172 - 639us/epoch - 10us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 184053.9219 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77965.3047 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 210707.1562 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 76072.9922 - 833us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272711.0938 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 53704.9766 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 152573.4375 - 844us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 75454.9062 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101650.0547 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61722.6016 - 752us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 186866.5625 - 611us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36594.1016 - 747us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 31351.3457 - 879us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16027.6006 - 2ms/epoch - 34us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66212.4531 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 216947.4688 - 705us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201911.1875 - 708us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55726.5586 - 811us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 37189.7188 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 78203.9922 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68016.2031 - 663us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 160063.7031 - 698us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 193900.7656 - 779us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 477469.0938 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 81556.0938 - 735us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38179.6211 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 203420.1250 - 741us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 163573.8125 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 157617.5938 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20377.8301 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 64029.2305 - 798us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65168.8281 - 717us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376703.7500 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98294.8828 - 902us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79591.6016 - 659us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69683.7344 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 81034.2266 - 755us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88415.6406 - 730us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 67684.0547 - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1095110.5000 - 636us/epoch - 10us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88231.1719 - 901us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 130418.2578 - 851us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 121831.4297 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68057.5078 - 731us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 205831.0312 - 701us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44728.9844 - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 93205.7734 - 802us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299229.8750 - 745us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 127935.8203 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95083.9375 - 2ms/epoch - 29us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 81521.7969 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166977.7188 - 656us/epoch - 11us/sample\n", + "Generation 57/300\n", + "Solving for Nash Equilibrium in Generation 57/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8650.5996 - 782us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 72378.1641 - 857us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 84783.0625 - 738us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45312.2852 - 716us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356468.6875 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 85555.4766 - 670us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68173.8828 - 667us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69184.7578 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55796.3750 - 693us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 60031.3008 - 1ms/epoch - 17us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 93533.2031 - 773us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 158179.3438 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 141473.7344 - 753us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33652.4180 - 659us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 96909.1406 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 73557.6094 - 746us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 284711.6875 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 221403.2812 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74266.0000 - 760us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 144992.1875 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 135328.9219 - 3ms/epoch - 45us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237379.9375 - 704us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 148222.4219 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 64864.7617 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34862.8047 - 664us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395131.0312 - 824us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 84090.9531 - 695us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316558.7500 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 111083.7656 - 753us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402257.2188 - 711us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 148538.4219 - 745us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 58664.7695 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29117.2637 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 162973.5938 - 2ms/epoch - 28us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 303873.2188 - 668us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 158135.7188 - 695us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90875.6719 - 679us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 165883.8594 - 730us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77451.8672 - 640us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 93739.4219 - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90267.2969 - 985us/epoch - 16us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 167796.8281 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22840.5664 - 570us/epoch - 9us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 150755.0625 - 684us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77011.3359 - 643us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104740.8047 - 703us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69023.0938 - 735us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49678.6797 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 154349.6719 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 189757.6094 - 741us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 162465.7812 - 630us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 121344.0156 - 854us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 198754.6875 - 784us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 156590.0625 - 2ms/epoch - 30us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 110531.6328 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232593.8594 - 763us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 516529.4062 - 578us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 96474.8984 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90654.9531 - 788us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 978481.3125 - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 96845.9219 - 783us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 87867.1797 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 153744.6094 - 771us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116507.3438 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36642.0820 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725693.5625 - 739us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 100793.5234 - 735us/epoch - 12us/sample\n", + "Generation 58/300\n", + "Solving for Nash Equilibrium in Generation 58/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373085.0000 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 161596.7031 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 67352.5625 - 725us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275818.0312 - 780us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 100891.5234 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47666.4531 - 641us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 173333.5781 - 966us/epoch - 16us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444967.0312 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83281.1016 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51312.8086 - 714us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609062.0000 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61832.3906 - 758us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306532.7500 - 719us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256427.2656 - 726us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25151.5293 - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 192257.0469 - 779us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258856.4844 - 963us/epoch - 16us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43489.4727 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 105387.9375 - 800us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 159727.1719 - 638us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44125.5547 - 741us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 82304.6797 - 591us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331911.5938 - 657us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83027.8359 - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180359.2188 - 767us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 165375.4531 - 1ms/epoch - 18us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 107855.5156 - 948us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358871.0000 - 713us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46068.9648 - 656us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268123.6562 - 668us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27267.7324 - 720us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56420.9023 - 688us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265815.8438 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 560409.4375 - 804us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46675.1172 - 708us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 181296.7812 - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 56843.1211 - 791us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 608580.4375 - 736us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66747.0625 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 85351.3906 - 1ms/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 183036.9688 - 618us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 81704.4844 - 814us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 58163.0977 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89916.8359 - 817us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13204.3555 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83153.5078 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719513.8750 - 879us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 167191.2031 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258052.9844 - 685us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40369.2969 - 716us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14430.5547 - 785us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 173516.6719 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54291.3203 - 963us/epoch - 16us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 178307.3281 - 758us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34309.6641 - 1ms/epoch - 17us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 24766.4727 - 727us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30696.4023 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257562.8438 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66369.5547 - 709us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277701.8438 - 2ms/epoch - 28us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 115494.8281 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 334794.9375 - 711us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214079.8750 - 722us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95052.6172 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 117838.3672 - 788us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89358.0234 - 717us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Generation 59/300\n", + "Solving for Nash Equilibrium in Generation 59/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 80581.1094 - 845us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 70741.5000 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1037665.2500 - 738us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83408.4062 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68251.1719 - 744us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 161276.2656 - 686us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 139124.2188 - 778us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326590.4688 - 4ms/epoch - 64us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369908.1562 - 1ms/epoch - 21us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41435.9648 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278944.7500 - 720us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89798.0000 - 763us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 181231.1875 - 818us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166134.3594 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 91947.2656 - 675us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 215911.6250 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 162603.4688 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 105437.3516 - 786us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106778.8594 - 774us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77849.9141 - 667us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112921.0312 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95408.9531 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 57869.1523 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88819.5859 - 922us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 58064.7461 - 818us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 120961.7344 - 848us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 209943.6406 - 755us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214176.8438 - 4ms/epoch - 60us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180827.2656 - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 131572.1094 - 770us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 60744.3164 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485251.2500 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166237.1406 - 2ms/epoch - 28us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166844.4531 - 841us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314236.3125 - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 87980.3516 - 720us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 145158.5156 - 1ms/epoch - 19us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71992.0547 - 684us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95293.7031 - 701us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475979.0000 - 878us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402945.8438 - 4ms/epoch - 62us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 175694.2344 - 888us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 129648.1797 - 929us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 159422.6719 - 883us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 199647.3125 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43808.4219 - 751us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259376.8750 - 664us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16138.6113 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90647.6484 - 733us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10265.2559 - 837us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 528823.3125 - 816us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 131661.5469 - 777us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 159392.8750 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21906.7246 - 1ms/epoch - 19us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104585.7422 - 692us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 138896.9375 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47209.3984 - 695us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201466.7812 - 664us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28665.5723 - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201235.6250 - 808us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 800103.0625 - 968us/epoch - 16us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59222.8555 - 734us/epoch - 12us/sample\n", + "Generation 60/300\n", + "Solving for Nash Equilibrium in Generation 60/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32427.5059 - 730us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 217949.4062 - 856us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 169785.2656 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116395.8438 - 709us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201716.9062 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 196051.9375 - 853us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 99909.7266 - 724us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 95699.3906 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339209.9688 - 679us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65002.7930 - 2ms/epoch - 31us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98101.0625 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54830.3750 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297424.1250 - 763us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71983.5000 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500412.0000 - 757us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 130156.9922 - 871us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 204687.5000 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558126.7500 - 946us/epoch - 15us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180508.9375 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 233899.3750 - 1ms/epoch - 18us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88002.6953 - 777us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 80582.8906 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 219183.5312 - 744us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 200073.1406 - 710us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 76838.5625 - 709us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544692.6875 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77969.5703 - 1ms/epoch - 19us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 185812.6250 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 203888.9531 - 1ms/epoch - 21us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 107093.4688 - 680us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 789592.7500 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 158261.9375 - 910us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 153409.4531 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 218794.1719 - 901us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 124637.1641 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 96797.6875 - 662us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272413.4062 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89726.3281 - 711us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534302.5000 - 736us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48137.0547 - 639us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123860.6719 - 910us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289255.2500 - 942us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 218160.1875 - 735us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 98000.1875 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 97720.6953 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 103301.9609 - 643us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 190927.7188 - 787us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36870.6992 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 100162.0000 - 833us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 221939.3750 - 771us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180580.2656 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66294.1484 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104797.5859 - 761us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106197.7734 - 869us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 211659.0000 - 829us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106839.7500 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 29772.3496 - 920us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104202.1328 - 779us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 96204.6953 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55219.7109 - 735us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 187876.7344 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126703.1875 - 699us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562934.2500 - 712us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 128889.1875 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33301.5312 - 787us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166774.5469 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 147475.5781 - 865us/epoch - 14us/sample\n", + "Generation 61/300\n", + "Solving for Nash Equilibrium in Generation 61/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32756.9043 - 697us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101576.1875 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83754.1328 - 4ms/epoch - 61us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246365.9688 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112576.3438 - 881us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 234471.4531 - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69777.7031 - 987us/epoch - 16us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 309375.2188 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 129557.5625 - 784us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286415.4688 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 110108.8594 - 738us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 96746.5391 - 703us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19240.1309 - 629us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672389.8125 - 718us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 339941.9375 - 1ms/epoch - 19us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 168758.3281 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 171612.5312 - 615us/epoch - 10us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301276.2188 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 221759.0312 - 744us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 91677.5156 - 700us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375743.1562 - 1ms/epoch - 22us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 58331.7109 - 762us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55939.4062 - 2ms/epoch - 28us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 227039.5469 - 831us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332264.0625 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112086.7891 - 659us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336374.1250 - 881us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 92173.6094 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 160091.5938 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 191778.5312 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338059.6250 - 779us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 50751.9141 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 137287.7812 - 670us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 150878.0312 - 833us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241820.1562 - 736us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 86225.0391 - 785us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1572941.3750 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270446.0625 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 36837.8516 - 728us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180009.7188 - 2ms/epoch - 37us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 60144.3555 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 91436.6094 - 860us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 252168.2344 - 1ms/epoch - 20us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 75131.5234 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 158152.1406 - 755us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33055.8086 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603363.6250 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 61539.5977 - 2ms/epoch - 25us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 486990.3125 - 730us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 154312.1250 - 769us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287040.6562 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 603497.1875 - 656us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 241644.1406 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 118386.5000 - 754us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180617.5156 - 739us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317506.0625 - 752us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 236721.9375 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 175377.1250 - 741us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55405.1992 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446564.9062 - 747us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 189339.5938 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20306.7793 - 768us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 119609.8203 - 709us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 134081.7031 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 150515.7812 - 843us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 41915.9688 - 785us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 119394.2109 - 752us/epoch - 12us/sample\n", + "Generation 62/300\n", + "Solving for Nash Equilibrium in Generation 62/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283527.5312 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372576.1875 - 959us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 155034.3125 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366214.1562 - 716us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52785.2734 - 792us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20889.6816 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 121118.4453 - 945us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257222.4062 - 1ms/epoch - 24us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 54783.7422 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 32062.6074 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 103517.6953 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 151973.1094 - 789us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693197.6250 - 720us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 46834.3555 - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 147837.0469 - 1ms/epoch - 16us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263042.1875 - 985us/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 80185.0000 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123704.0703 - 754us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 313633.0938 - 820us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 176884.7812 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 174250.4688 - 823us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77782.1875 - 652us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 153927.2812 - 878us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1053262.7500 - 855us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 172127.1719 - 784us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 173663.2812 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361792.3750 - 761us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 110089.1328 - 698us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276755.3750 - 658us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265072.3438 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402046.5938 - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 102146.0781 - 858us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504248.6562 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 115342.8594 - 712us/epoch - 11us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245366.0000 - 679us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247749.2656 - 892us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247634.7656 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278871.9688 - 739us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 84121.3906 - 665us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375484.8750 - 691us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66447.6953 - 672us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248347.0781 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34195.6484 - 922us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356873.3438 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264511.7188 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 151215.7969 - 722us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 131536.1719 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42231.7695 - 659us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230140.7031 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 224254.0312 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1373962.2500 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88817.5938 - 719us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 170307.9062 - 777us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 134371.0312 - 661us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 97024.2266 - 700us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273318.7812 - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316389.7188 - 841us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267175.3125 - 706us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 132136.2188 - 938us/epoch - 15us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 78573.3906 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384282.1250 - 746us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10333.9463 - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 164757.1562 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59941.2578 - 763us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666206.0000 - 784us/epoch - 13us/sample\n", + "Generation 63/300\n", + "Solving for Nash Equilibrium in Generation 63/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646118.5000 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 130408.4062 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68754.6484 - 739us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 108380.4219 - 638us/epoch - 10us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 139528.1875 - 792us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 517701.6250 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 131875.1719 - 761us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296097.0000 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285730.5938 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104657.9375 - 780us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366464.4688 - 709us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275854.5938 - 751us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684519.8750 - 802us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391455.0938 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 205501.0781 - 1ms/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 129736.8984 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269972.1562 - 2ms/epoch - 31us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394957.8125 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101155.0312 - 684us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387048.6875 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281411.8750 - 679us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 680091.5625 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 127541.8359 - 756us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1269293.3750 - 980us/epoch - 16us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 247361.9219 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 132755.7344 - 725us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 85720.3906 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408377.7500 - 829us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 85034.7422 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 119591.7578 - 700us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 52383.2969 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277536.1875 - 818us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250098.0469 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 63369.2188 - 1ms/epoch - 17us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244227.3438 - 751us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 134290.9844 - 754us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13017.1895 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326098.1250 - 824us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 253114.4375 - 1ms/epoch - 21us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123723.0312 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 118741.9141 - 673us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 63254.5977 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 184672.9219 - 1ms/epoch - 18us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 540343.2500 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 107489.9844 - 839us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 186807.8281 - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430791.7812 - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83486.2344 - 717us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257862.5625 - 651us/epoch - 10us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289080.8438 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 130302.2578 - 686us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 137709.1250 - 804us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288558.2188 - 921us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 66195.9844 - 1ms/epoch - 17us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 108750.7812 - 835us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436884.7188 - 945us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 152045.4531 - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 174202.7188 - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450137.1562 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45355.0391 - 847us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 65705.4766 - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306936.6875 - 815us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266710.8750 - 5ms/epoch - 75us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 225579.3594 - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 178859.4531 - 889us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298869.3750 - 821us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606702.4375 - 2ms/epoch - 24us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 107315.0156 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420991.8438 - 790us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 216578.5625 - 923us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1035741.9375 - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 199024.7344 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166516.5156 - 770us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 155473.9688 - 865us/epoch - 14us/sample\n", + "Generation 64/300\n", + "Solving for Nash Equilibrium in Generation 64/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 92606.3203 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42672.5859 - 743us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 187536.7031 - 623us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317085.0000 - 804us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88325.1172 - 797us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326109.8125 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610622.7500 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287987.2500 - 1ms/epoch - 17us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 165483.4375 - 798us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566042.5000 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 97787.6406 - 795us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51388.4922 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 164981.2812 - 752us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 364118.7188 - 852us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378212.0000 - 893us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1315665.2500 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69362.4609 - 828us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248143.0781 - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1053973.6250 - 762us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 156832.9531 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 39256.7227 - 725us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 105159.8047 - 708us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 738689.6250 - 701us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 137451.6719 - 794us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 527802.6250 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1531233.8750 - 1ms/epoch - 17us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44065.7422 - 1ms/epoch - 17us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261743.8125 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101988.9375 - 753us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90761.5312 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 89533.4531 - 770us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349772.6562 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 91926.9141 - 709us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325979.5312 - 693us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476441.0625 - 748us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 91980.1094 - 790us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 99490.5625 - 688us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1566064.7500 - 776us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 177182.5625 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 205308.7344 - 2ms/epoch - 26us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 213960.6875 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 329627.4062 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 205481.0938 - 2ms/epoch - 39us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 97117.3047 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 117942.6719 - 667us/epoch - 11us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 110022.9531 - 753us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 184075.3438 - 7ms/epoch - 120us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 129178.1562 - 754us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 786228.9375 - 694us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 210106.5312 - 1ms/epoch - 24us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251766.4531 - 733us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1016196.5625 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74013.6562 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 216653.0781 - 3ms/epoch - 49us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504314.4688 - 841us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 99033.0156 - 2ms/epoch - 26us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 249420.0469 - 750us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 221650.6875 - 775us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 131257.8281 - 639us/epoch - 10us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 125409.0391 - 736us/epoch - 12us/sample\n", + "Generation 65/300\n", + "Solving for Nash Equilibrium in Generation 65/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351824.0312 - 771us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318961.0000 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485845.3438 - 2ms/epoch - 33us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 217846.5156 - 710us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14545.6768 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 87110.1328 - 868us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 192555.5312 - 767us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 147483.7344 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 97454.0312 - 746us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 194217.4531 - 711us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104647.6172 - 1ms/epoch - 18us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83396.0703 - 673us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1632926.6250 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 124994.0703 - 1ms/epoch - 23us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 267775.3750 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502935.6562 - 2ms/epoch - 27us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44286.0898 - 947us/epoch - 15us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302095.0000 - 842us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 720960.9375 - 685us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287960.3125 - 800us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 174075.0000 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281627.6250 - 1ms/epoch - 16us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325913.9375 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348714.5625 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101378.6953 - 780us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 131376.6094 - 732us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366841.5625 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 435229.0312 - 788us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 434155.1250 - 702us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166626.3125 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352695.8125 - 1ms/epoch - 18us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 107600.1641 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 177726.8125 - 782us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 135276.1406 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 202291.6094 - 785us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 196107.5156 - 790us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243140.8125 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 200442.1719 - 3ms/epoch - 51us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351691.5625 - 784us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1520507.2500 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250696.1094 - 797us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230130.5938 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521774.1875 - 703us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 203761.0625 - 808us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265256.5625 - 751us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349742.0625 - 694us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383312.1250 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 128123.8281 - 778us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593700.7500 - 781us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180838.0312 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673045.3750 - 743us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365466.3438 - 693us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 23370.0449 - 853us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 515975.3438 - 2ms/epoch - 37us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514834.4062 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 774529.5625 - 804us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268553.6562 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 988972.0625 - 698us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 399819.7188 - 760us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232010.5000 - 701us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677034.3750 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 62240.3789 - 828us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 223338.6562 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381224.2500 - 753us/epoch - 12us/sample\n", + "Generation 66/300\n", + "Solving for Nash Equilibrium in Generation 66/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 138014.0312 - 866us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101710.4844 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 925528.9375 - 834us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370302.7812 - 648us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433012.2812 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 169032.7188 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 192393.0625 - 717us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344142.8750 - 693us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214311.9531 - 922us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43683.4219 - 752us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 473477.1250 - 849us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1108512.2500 - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351167.9375 - 741us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 87391.2109 - 890us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 44993.6719 - 735us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 423950.0312 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 139123.8750 - 690us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543896.7500 - 749us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 444556.1875 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 196679.4062 - 720us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 219485.2812 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 90067.1094 - 746us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240252.7969 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246674.3125 - 765us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297466.4062 - 770us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 198753.1875 - 657us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656905.5000 - 842us/epoch - 14us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1016787.5000 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304654.0625 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550188.4375 - 739us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534587.3750 - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 17ms/epoch - 282us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 117973.6094 - 1ms/epoch - 18us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460155.8438 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45770.2188 - 636us/epoch - 10us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378779.3750 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274092.1875 - 756us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365871.9062 - 754us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2095219.3750 - 732us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 161809.8438 - 797us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 581518.5625 - 723us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180317.4375 - 816us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 20452.1719 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 60076.3203 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411618.7812 - 747us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415841.2188 - 734us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 186383.3906 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272507.1562 - 800us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 183081.0156 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 223682.3438 - 890us/epoch - 14us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373771.3750 - 979us/epoch - 16us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 147033.2031 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380545.1562 - 759us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503654.4062 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 118447.1953 - 863us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 214726.8125 - 680us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549976.4375 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 454137.6875 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116834.5703 - 829us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488847.3125 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232943.8594 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 118237.9922 - 683us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700618.0625 - 2ms/epoch - 29us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 260764.0000 - 820us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 187499.8438 - 956us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 263439.3750 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 400130.4062 - 756us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497403.3438 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555635.2500 - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 148048.1562 - 2ms/epoch - 38us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 345108.8750 - 768us/epoch - 12us/sample\n", + "Generation 67/300\n", + "Solving for Nash Equilibrium in Generation 67/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 38058.7500 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261399.9531 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 858708.8750 - 770us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394549.3438 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 168721.6250 - 1ms/epoch - 17us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 127966.8672 - 822us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 150435.9844 - 799us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476441.0312 - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1370270.6250 - 3ms/epoch - 48us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635634.6875 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 161950.4844 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 193629.5156 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 170117.4375 - 695us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 172197.4375 - 847us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 110us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 62373.6992 - 2ms/epoch - 32us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265384.2812 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 885624.3750 - 803us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514333.0312 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101192.0156 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 155479.2656 - 755us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428565.1562 - 806us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 318644.5625 - 974us/epoch - 16us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 718163.2500 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 839503.5625 - 811us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10646.0225 - 732us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259713.8750 - 784us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510569.9062 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248165.9219 - 671us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281459.6562 - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1146587.3750 - 1ms/epoch - 17us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650649.8750 - 701us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1118855.0000 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 100518.0312 - 792us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 781106.8750 - 969us/epoch - 16us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 470481.2500 - 802us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2837780.2500 - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 160499.4688 - 916us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419725.5312 - 757us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 55409.6914 - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 113598.2969 - 592us/epoch - 10us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506631.1250 - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1465812.1250 - 738us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 179878.9688 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287020.9375 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289798.4688 - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 209307.9844 - 713us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166737.4219 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280383.9062 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 92049.7500 - 4ms/epoch - 62us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451616.4062 - 821us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 181017.5312 - 765us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1802986.8750 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 250657.6250 - 980us/epoch - 16us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 173764.4688 - 841us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 594419.8750 - 735us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 137158.4688 - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238325.4375 - 671us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 163722.6562 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 491461.5938 - 790us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1386113.0000 - 3ms/epoch - 45us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1520939.2500 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433006.0938 - 1ms/epoch - 21us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320070.3750 - 669us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 199270.0781 - 840us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 894505.3750 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 51707.2773 - 785us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 823979.6250 - 1ms/epoch - 17us/sample\n", + "Generation 68/300\n", + "Solving for Nash Equilibrium in Generation 68/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290752.9688 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 58086.8047 - 902us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1442656.3750 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 33710.7617 - 747us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 208130.2344 - 4ms/epoch - 58us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 867629.3125 - 774us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352615.0938 - 775us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1321215.2500 - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 155443.9688 - 730us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1005075.5000 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575678.7500 - 838us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511785.1875 - 720us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 213458.2344 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 96485.0547 - 722us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311907.1875 - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254079.1250 - 705us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 445590.1250 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1197141.5000 - 756us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350250.0938 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 639119.9375 - 1ms/epoch - 16us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 192062.4062 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112946.0703 - 1ms/epoch - 16us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 193573.5938 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 108517.8906 - 948us/epoch - 15us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 514707.5938 - 800us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27845.7012 - 977us/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 627909.6875 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 195533.3906 - 942us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297769.0312 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 281123.1562 - 928us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647338.0000 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297958.0625 - 901us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376835.4375 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323316.1875 - 815us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 216086.8125 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328631.3750 - 747us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479324.3750 - 824us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1625537.7500 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 109865.0078 - 795us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 896928.5625 - 771us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280088.6250 - 992us/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1875991.7500 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 220941.3125 - 719us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461894.0625 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1898578.8750 - 783us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 741416.1875 - 873us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68832.4453 - 830us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 136565.3750 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 220819.4688 - 900us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201562.8906 - 763us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235409.7031 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299478.8438 - 670us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1120087.7500 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279775.4688 - 919us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461225.0938 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 223915.7969 - 773us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401129.8125 - 1ms/epoch - 18us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 948157.8750 - 1ms/epoch - 17us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 194908.7656 - 850us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116629.2891 - 633us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275676.9375 - 761us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437205.1875 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 550929.2500 - 891us/epoch - 14us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9850.5596 - 788us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1314090.8750 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 220306.8438 - 743us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 196802.7656 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 187089.8906 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 557301.6250 - 758us/epoch - 12us/sample\n", + "Generation 69/300\n", + "Solving for Nash Equilibrium in Generation 69/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 365313.6250 - 762us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504725.4688 - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 140035.5000 - 695us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 839608.6875 - 1ms/epoch - 20us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 788710.5625 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 161677.0938 - 726us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 125521.1172 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 283217.8750 - 1ms/epoch - 17us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353215.0625 - 667us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 143640.0938 - 710us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 123928.0234 - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 180611.9062 - 1ms/epoch - 21us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240372.7031 - 770us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344495.8438 - 684us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 190148.7344 - 742us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1425373.3750 - 615us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 144902.8125 - 664us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1184865.7500 - 747us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325979.7500 - 1ms/epoch - 21us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 203582.3438 - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 145619.2812 - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482968.0625 - 878us/epoch - 14us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301197.3750 - 5ms/epoch - 86us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 518704.6875 - 793us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 850082.6250 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 163959.2188 - 806us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 291430.7812 - 757us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489224.4375 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 115375.3359 - 1ms/epoch - 17us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42325.8047 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 115359.7422 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294059.0938 - 998us/epoch - 16us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25693.9023 - 841us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 143429.4062 - 920us/epoch - 15us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387266.3750 - 815us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 153055.3438 - 1ms/epoch - 19us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604538.3750 - 2ms/epoch - 26us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261074.9688 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 374369.0938 - 817us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350877.5625 - 939us/epoch - 15us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2078110.7500 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240633.3750 - 1ms/epoch - 23us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 74494.8125 - 839us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244014.8281 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1626238.7500 - 880us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 156932.8438 - 2ms/epoch - 25us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 640572.3750 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288417.9375 - 1ms/epoch - 19us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 880263.7500 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112600.5859 - 836us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 45789.7891 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760183.9375 - 835us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 208345.4219 - 807us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338203.3438 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 487386.2188 - 966us/epoch - 16us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88922.9297 - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510785.7500 - 945us/epoch - 15us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1709885.8750 - 791us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 999824.3125 - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 79519.2812 - 6ms/epoch - 96us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592542.6875 - 752us/epoch - 12us/sample\n", + "Generation 70/300\n", + "Solving for Nash Equilibrium in Generation 70/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 949126.3125 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513276.0625 - 1ms/epoch - 16us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629386.3125 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 212631.2812 - 1ms/epoch - 16us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8752.6318 - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787741.6250 - 894us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 886845.1250 - 830us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 133366.9219 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448862.9062 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 441805.4375 - 839us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371289.6562 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 167223.0469 - 965us/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500396.9062 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315221.5625 - 850us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 22530.3535 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317777.0312 - 858us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545944.7500 - 828us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338125.1875 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 461886.3125 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 749305.9375 - 1ms/epoch - 20us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 229886.5781 - 828us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556659.4375 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 373312.0938 - 881us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 256180.8281 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 206160.5469 - 899us/epoch - 15us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368724.2812 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258303.9688 - 876us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337163.4375 - 757us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 324687.6562 - 895us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1458278.3750 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 174200.9062 - 660us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 355891.2500 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 151842.1406 - 834us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439262.8750 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 159302.0000 - 6ms/epoch - 90us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 220590.3125 - 915us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673121.6875 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276725.5938 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 203144.9531 - 806us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264816.2188 - 1ms/epoch - 18us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727486.1875 - 780us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 212712.3906 - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769532.8750 - 967us/epoch - 16us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1380897.0000 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 64845.4688 - 864us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 474922.4375 - 853us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 173238.3594 - 987us/epoch - 16us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 768061.3125 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2098843.7500 - 746us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126946.3672 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 311435.7188 - 983us/epoch - 16us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 335571.5625 - 808us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 622154.5625 - 809us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 544354.6875 - 725us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1581416.5000 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 154297.9844 - 792us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 175828.4844 - 753us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 171831.5781 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 86061.0781 - 1ms/epoch - 21us/sample\n", + "Generation 71/300\n", + "Solving for Nash Equilibrium in Generation 71/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295438.0625 - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2196770.5000 - 715us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 166009.3750 - 904us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1837016.8750 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 186080.0781 - 791us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376851.0000 - 852us/epoch - 14us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382234.2188 - 778us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1315239.7500 - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1940641.7500 - 810us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25805.7871 - 1ms/epoch - 18us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 595559.6875 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1329492.1250 - 799us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386593.3438 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277100.6875 - 1ms/epoch - 21us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1025526.3750 - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 136224.9844 - 822us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 128641.0703 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410308.3438 - 729us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325374.2188 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350375.9375 - 724us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 501465.3125 - 752us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 88295.3125 - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568882.5000 - 680us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 731608.3125 - 813us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1072536.0000 - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703387.6250 - 833us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 790911.1250 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 816049.3125 - 727us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 330447.4375 - 692us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 136us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 42886.5469 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386341.8750 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 333013.2188 - 673us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2541170.5000 - 633us/epoch - 10us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771535.0625 - 1ms/epoch - 17us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 372311.0312 - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337739.3438 - 797us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 72011.4844 - 685us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724206.7500 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264267.1250 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357988.8750 - 672us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 584545.6875 - 715us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493136.9688 - 932us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 285104.8125 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 592399.1875 - 8ms/epoch - 131us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279106.7500 - 884us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 346963.9375 - 762us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1152958.3750 - 906us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1724180.7500 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 800922.1250 - 812us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506868.7812 - 767us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667630.8750 - 994us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 912726.6250 - 1ms/epoch - 18us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 728841.5000 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 73609.3516 - 866us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 255713.4688 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 67443.7656 - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634777.3750 - 841us/epoch - 14us/sample\n", + "Generation 72/300\n", + "Solving for Nash Equilibrium in Generation 72/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 872700.1875 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 672861.1250 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68463.0781 - 670us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240287.1250 - 766us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1349703.7500 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376445.3750 - 670us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602554.3750 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71938.7031 - 765us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259999.9531 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651139.9375 - 3ms/epoch - 50us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649879.5625 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 363046.0625 - 814us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 390256.9375 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561592.6875 - 759us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688386.0625 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 176848.3594 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 616653.6250 - 851us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 130766.4688 - 767us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1293721.5000 - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 430602.7188 - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479474.7188 - 1ms/epoch - 18us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 442978.2500 - 2ms/epoch - 28us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 165489.6875 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623513.3125 - 845us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 127232.7344 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669362.8125 - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1430392.3750 - 878us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 27002.1250 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 909194.8125 - 864us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231124.0000 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 26212.4082 - 1ms/epoch - 21us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 289397.6562 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760635.4375 - 679us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 611830.6250 - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 452915.6562 - 761us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 352238.9062 - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254920.3906 - 832us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419426.9688 - 911us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384192.9688 - 1ms/epoch - 17us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1031539.2500 - 21ms/epoch - 331us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 375878.4688 - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 559788.8125 - 837us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378425.0000 - 1ms/epoch - 17us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2377585.5000 - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 664938.1875 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 85749.0312 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381475.6875 - 772us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 647710.0625 - 852us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302121.1250 - 941us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1038957.8750 - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 164911.0156 - 766us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437243.9062 - 775us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 948062.5625 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458655.2500 - 636us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 565705.7500 - 969us/epoch - 16us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438489.4375 - 2ms/epoch - 27us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2376377.7500 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1679480.7500 - 712us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302858.4375 - 684us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 189182.6719 - 871us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 216219.4219 - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656387.8750 - 884us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 40793.6680 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310566.1875 - 2ms/epoch - 34us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 150271.3906 - 773us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3790370.0000 - 763us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471254.8750 - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 959419.5625 - 717us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 703890.1875 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237701.0312 - 817us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Generation 73/300\n", + "Solving for Nash Equilibrium in Generation 73/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 426427.0938 - 989us/epoch - 16us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 776742.0000 - 752us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 21253.2207 - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 73393.6016 - 4ms/epoch - 58us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 415103.9375 - 728us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 867538.3125 - 902us/epoch - 15us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 237281.7812 - 887us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700684.1875 - 808us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438440.7500 - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414105.6250 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 195076.8750 - 896us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 762124.1875 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 75332.7656 - 880us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 343809.0312 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 142095.4531 - 756us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604777.7500 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 708447.8125 - 825us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300694.7188 - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412229.9375 - 802us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251601.2969 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 378686.9375 - 984us/epoch - 16us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248960.4219 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351619.4688 - 1ms/epoch - 24us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 424843.4062 - 898us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1096307.5000 - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 77796.3438 - 899us/epoch - 15us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1498340.6250 - 718us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 848656.7500 - 809us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736114.7500 - 747us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1249305.5000 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 93us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 588982.2500 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323811.0000 - 823us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773505.6875 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410864.8438 - 782us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408852.7812 - 660us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433085.4375 - 745us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 735957.2500 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2438295.7500 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 145303.8750 - 661us/epoch - 11us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1000756.2500 - 843us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 554915.6875 - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 57700.1055 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 86432.0781 - 985us/epoch - 16us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1263395.0000 - 762us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 83758.2891 - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 254502.0938 - 1ms/epoch - 18us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767977.1875 - 666us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 903469.9375 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 849618.1250 - 1ms/epoch - 17us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1476896.6250 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 449315.2500 - 673us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 93340.2500 - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2381541.0000 - 922us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 8ms/epoch - 121us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 476171.6562 - 1ms/epoch - 20us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 727774.3125 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508396.7812 - 1ms/epoch - 16us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1777093.8750 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 806363.8750 - 710us/epoch - 11us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 371014.6250 - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 135740.6406 - 832us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314834.0625 - 757us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 128270.4375 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 433222.8125 - 890us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1324508.0000 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 468434.9062 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 816605.1875 - 3ms/epoch - 41us/sample\n", + "Generation 74/300\n", + "Solving for Nash Equilibrium in Generation 74/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 160283.0938 - 835us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 625839.4375 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 480773.3438 - 683us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286465.6562 - 3ms/epoch - 43us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 827520.6875 - 763us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 224570.6406 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1314030.1250 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 435044.1875 - 601us/epoch - 10us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495011.1875 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 837795.2500 - 652us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724717.1875 - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1037990.6875 - 699us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 280205.4375 - 741us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298303.8750 - 849us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 30744.2637 - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663511.0625 - 3ms/epoch - 44us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 660178.6875 - 729us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583566.0000 - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 970726.8750 - 678us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 753994.0625 - 874us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 59892.9062 - 708us/epoch - 11us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 965290.1250 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620823.1875 - 851us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 272373.1875 - 964us/epoch - 16us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 49573.1133 - 1ms/epoch - 16us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482835.9062 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1176069.5000 - 732us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 892507.2500 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 628142.6250 - 836us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 977466.4375 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 278585.4375 - 1ms/epoch - 17us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 268446.9375 - 764us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1077020.6250 - 1ms/epoch - 18us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1259499.2500 - 977us/epoch - 16us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 892088.6875 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 634709.6875 - 934us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1431391.3750 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347264.8750 - 716us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 862338.6250 - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682752.5625 - 6ms/epoch - 96us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 877941.6250 - 52ms/epoch - 846us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784926.0000 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 85642.7734 - 768us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2696163.0000 - 947us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377497.1562 - 795us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312113.7188 - 722us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359817.8438 - 939us/epoch - 15us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1728405.0000 - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 287213.8125 - 654us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 451949.1250 - 2ms/epoch - 31us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3633113.5000 - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 274500.4688 - 745us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1237520.2500 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276724.4375 - 825us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 878529.1875 - 2ms/epoch - 28us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 410808.3125 - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 393640.8750 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295093.4688 - 810us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238603.1719 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 357556.8750 - 744us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 158037.3750 - 1ms/epoch - 21us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464056.9688 - 766us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 297330.5625 - 974us/epoch - 16us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301011.9062 - 889us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Generation 75/300\n", + "Solving for Nash Equilibrium in Generation 75/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 188736.7656 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618824.3125 - 822us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 94184.1719 - 1ms/epoch - 20us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1042887.9375 - 936us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1597027.2500 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 832992.1250 - 1ms/epoch - 20us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 92081.3984 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2222471.2500 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 823757.0625 - 789us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 161500.6562 - 918us/epoch - 15us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2270393.2500 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116015.6016 - 922us/epoch - 15us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1459217.6250 - 3ms/epoch - 47us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1187044.2500 - 62ms/epoch - 1ms/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 257128.8906 - 865us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 502877.7188 - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 224889.5469 - 737us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305613.3438 - 950us/epoch - 15us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 827693.6250 - 4ms/epoch - 59us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538980.3125 - 685us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488703.3750 - 912us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106221.2656 - 810us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 294332.7812 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232057.5156 - 705us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 817445.8750 - 751us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1097902.1250 - 667us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1680870.5000 - 816us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308870.9375 - 665us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1015099.8750 - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 818039.8125 - 797us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484706.3438 - 780us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 202532.6250 - 833us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2476748.5000 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 475348.9688 - 704us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1038875.8750 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 478319.8750 - 832us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240308.3906 - 1ms/epoch - 21us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 279621.6250 - 677us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 101483.0469 - 998us/epoch - 16us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126716.3672 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 197389.7188 - 714us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370778.0000 - 646us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1586611.6250 - 805us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 127870.9375 - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 816089.8125 - 1ms/epoch - 17us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1740538.1250 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 852852.5625 - 831us/epoch - 13us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 229915.7031 - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 131950.7188 - 835us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2872521.5000 - 914us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3440472.5000 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 796668.4375 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270371.4688 - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2142120.2500 - 765us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 861812.4375 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377210.9062 - 647us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 93683.0078 - 1ms/epoch - 21us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 556244.6250 - 691us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 181184.8438 - 786us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1239192.0000 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 846407.8125 - 787us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4037279.2500 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 823683.3125 - 732us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1247819.1250 - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2144874.5000 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 197823.5000 - 726us/epoch - 12us/sample\n", + "Generation 76/300\n", + "Solving for Nash Equilibrium in Generation 76/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307275.0938 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230995.4688 - 2ms/epoch - 32us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408190.0625 - 1ms/epoch - 21us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 233079.0156 - 779us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 203165.2188 - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1256903.8750 - 1ms/epoch - 17us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 259014.8438 - 732us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 369569.7500 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386423.1562 - 838us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1101048.0000 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 173127.5781 - 730us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 831758.8750 - 1ms/epoch - 17us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 771447.7500 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1230856.8750 - 778us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 328280.5938 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326192.4062 - 858us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2905254.2500 - 732us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1036909.8750 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462549.7188 - 897us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 215019.9062 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 315019.5938 - 727us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 238474.1094 - 796us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4020206.0000 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 644763.6250 - 734us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 917602.7500 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1138492.2500 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 562686.3750 - 715us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503198.0938 - 831us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353642.6875 - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1433276.3750 - 1ms/epoch - 16us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 344827.7188 - 690us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 396204.8125 - 709us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1709202.5000 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 58402.7578 - 4ms/epoch - 60us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568797.6250 - 951us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1066635.2500 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126920.0703 - 1ms/epoch - 21us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3342080.7500 - 4ms/epoch - 63us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 800365.2500 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 391127.0000 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 326353.9375 - 785us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 178848.3281 - 9ms/epoch - 151us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126200.9531 - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 251798.3281 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 508125.3438 - 4ms/epoch - 57us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 861922.3125 - 2ms/epoch - 26us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505407.3750 - 716us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 826264.5000 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397095.8438 - 754us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 993080.1875 - 824us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 888602.8750 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784486.1875 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 100634.5391 - 1ms/epoch - 18us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 142139.1875 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 248300.4062 - 1ms/epoch - 19us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4074158.7500 - 744us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2449224.0000 - 831us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 821586.1250 - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1070140.8750 - 837us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1987481.5000 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 525279.8750 - 879us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322415.8125 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 194065.7812 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 939728.0625 - 672us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104086.0234 - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 437132.0625 - 4ms/epoch - 64us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338708.2812 - 919us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411870.1875 - 829us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Generation 77/300\n", + "Solving for Nash Equilibrium in Generation 77/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 385692.0312 - 10ms/epoch - 154us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533815.9375 - 796us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 408248.8125 - 811us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270502.4688 - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519246.6562 - 772us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342579.5938 - 2ms/epoch - 26us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 847406.6875 - 941us/epoch - 15us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1630794.7500 - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 901055.8125 - 713us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 106823.5000 - 822us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2134298.5000 - 698us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2019227.1250 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 174603.5625 - 685us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 950406.2500 - 898us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 955500.5625 - 835us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1361871.8750 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602615.3125 - 723us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 548286.5625 - 916us/epoch - 15us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3345914.0000 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630635.2500 - 1ms/epoch - 20us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2083315.3750 - 844us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 60637.1719 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606492.9375 - 920us/epoch - 15us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 68226.3594 - 1ms/epoch - 17us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414679.2500 - 981us/epoch - 16us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1775708.2500 - 657us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 937085.6250 - 942us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1398471.1250 - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 380880.3438 - 717us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 763415.2500 - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 549197.5625 - 1ms/epoch - 24us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2276496.0000 - 681us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 472212.7500 - 636us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 552653.6250 - 995us/epoch - 16us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420581.0938 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 354074.7812 - 784us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 356654.9688 - 998us/epoch - 16us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 332357.0312 - 730us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 939258.5000 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 124629.0781 - 941us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558143.5625 - 876us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 191415.3906 - 715us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 296988.0625 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386994.7812 - 923us/epoch - 15us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 989688.6250 - 785us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6105605.0000 - 1ms/epoch - 16us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 980744.0625 - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3198150.0000 - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1448656.7500 - 847us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736538.1250 - 1ms/epoch - 18us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1469542.3750 - 933us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455889.2188 - 825us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 314820.2812 - 656us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 950032.3125 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1443686.6250 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 193630.5938 - 845us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 575868.1250 - 739us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 112309.6797 - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1010695.8125 - 1ms/epoch - 18us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 367971.9688 - 1ms/epoch - 19us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1472933.1250 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638630.4375 - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 368303.6250 - 2ms/epoch - 26us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545871.1250 - 796us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469466.4375 - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 957388.8750 - 1ms/epoch - 19us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1565286.8750 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1079960.5000 - 780us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 729156.8750 - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 406608.4688 - 949us/epoch - 15us/sample\n", + "Generation 78/300\n", + "Solving for Nash Equilibrium in Generation 78/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787438.0625 - 949us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414074.5312 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1338064.6250 - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1485125.2500 - 871us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555034.7500 - 932us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 266886.3750 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 327540.5938 - 825us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 113181.2734 - 824us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 114351.9844 - 928us/epoch - 15us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3920524.2500 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 678248.8125 - 833us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3366972.5000 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321598.9062 - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1170910.5000 - 755us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1054104.7500 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306922.7188 - 846us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 797181.7500 - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 212712.0469 - 848us/epoch - 14us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1694738.7500 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412120.3438 - 823us/epoch - 13us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646273.3750 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 666640.7500 - 939us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 600808.3750 - 716us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1112935.8750 - 675us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398366.3125 - 685us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1061803.1250 - 670us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 688284.3125 - 910us/epoch - 15us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675018.3125 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 992620.3750 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 620091.6250 - 722us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2338641.0000 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446955.0312 - 2ms/epoch - 35us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 381164.3438 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 435912.5625 - 646us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 116827.2578 - 728us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 542051.8750 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 117900.7188 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 571067.1875 - 1ms/epoch - 20us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2385748.2500 - 746us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 295573.0938 - 741us/epoch - 12us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 485729.2812 - 925us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1323507.7500 - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 966885.8125 - 758us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4001003.2500 - 714us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 242864.8125 - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 629902.6875 - 753us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1139632.2500 - 734us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700629.0625 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2672603.2500 - 2ms/epoch - 26us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387099.7188 - 672us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1062048.2500 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 681758.5625 - 771us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1091744.0000 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 202889.4219 - 803us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1480967.1250 - 842us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521412.6875 - 692us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1218802.1250 - 806us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1695325.3750 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 436680.8750 - 891us/epoch - 14us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405605.0312 - 788us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 702346.5000 - 714us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 15ms/epoch - 237us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1045147.6250 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1153001.2500 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 302361.1250 - 904us/epoch - 15us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585885.0625 - 801us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1982321.2500 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 448059.5938 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532371.2500 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 467125.6875 - 705us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 397418.3438 - 767us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 304043.8438 - 1ms/epoch - 16us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1232276.2500 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 417478.3438 - 762us/epoch - 12us/sample\n", + "Generation 79/300\n", + "Solving for Nash Equilibrium in Generation 79/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308259.3125 - 725us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 826956.7500 - 562us/epoch - 9us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3264664.2500 - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1476508.5000 - 766us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 847169.1250 - 681us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 769232.7500 - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 870452.1875 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 401659.9062 - 641us/epoch - 10us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240344.4062 - 2ms/epoch - 28us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604364.4375 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1748470.7500 - 836us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 228014.9531 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1142058.8750 - 2ms/epoch - 32us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1292014.5000 - 1ms/epoch - 23us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1088950.2500 - 813us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458869.6875 - 856us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2220035.5000 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725557.8750 - 1ms/epoch - 16us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2373583.0000 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 126684.6094 - 732us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1196028.7500 - 804us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545431.5000 - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1416082.5000 - 880us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1286366.1250 - 929us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 25573.7773 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1082702.1250 - 781us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 109913.5703 - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1389089.8750 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 162us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 662794.0625 - 3ms/epoch - 42us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1668589.6250 - 836us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671729.1875 - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258312.6406 - 947us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 150904.5312 - 801us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 201611.0469 - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719077.3750 - 834us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1710040.5000 - 781us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 376061.1562 - 1ms/epoch - 20us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658328.7500 - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1358960.5000 - 1ms/epoch - 17us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 135554.9219 - 789us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1254218.7500 - 908us/epoch - 15us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 231537.3438 - 856us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 361819.6562 - 959us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500825.2500 - 1ms/epoch - 16us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 497814.0938 - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1495707.0000 - 797us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450923.1875 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 469296.3438 - 1ms/epoch - 17us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2208971.7500 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 579377.3750 - 1ms/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1169584.2500 - 970us/epoch - 16us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6413393.5000 - 757us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1186338.2500 - 716us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425411.1875 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1145174.5000 - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 337519.0000 - 894us/epoch - 14us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1184001.8750 - 772us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3388257.2500 - 870us/epoch - 14us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2546566.2500 - 725us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 809759.5625 - 643us/epoch - 10us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 211887.0781 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1436015.7500 - 630us/epoch - 10us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758526.1875 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490493.0625 - 728us/epoch - 12us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1378938.3750 - 762us/epoch - 12us/sample\n", + "Generation 80/300\n", + "Solving for Nash Equilibrium in Generation 80/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 569690.9375 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 956330.2500 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 307046.5938 - 846us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 130526.2266 - 3ms/epoch - 56us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1201081.7500 - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656252.8750 - 874us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 692398.0000 - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1212450.3750 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5165210.0000 - 787us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 348382.0938 - 644us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1209924.7500 - 681us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 383160.9375 - 2ms/epoch - 27us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2303084.2500 - 704us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760454.7500 - 766us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737600.8750 - 692us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 901102.5625 - 728us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 78441.4922 - 688us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 35ms/epoch - 557us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1655315.0000 - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 411767.6875 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3705010.7500 - 5ms/epoch - 79us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456257.5312 - 755us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 136101.3594 - 780us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1777839.8750 - 716us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1841218.8750 - 868us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1098457.6250 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1372946.6250 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269216.2188 - 1ms/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1222391.1250 - 725us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 292048.0000 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1223637.2500 - 1ms/epoch - 18us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 19559.4746 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1377591.3750 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232313.1875 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532968.8750 - 795us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 179330.3438 - 1ms/epoch - 20us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 812193.5000 - 914us/epoch - 15us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759864.7500 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 704655.8750 - 1ms/epoch - 17us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1487845.0000 - 979us/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492674.9688 - 724us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3466198.7500 - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 48322.3086 - 1ms/epoch - 20us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1202611.0000 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462094.0312 - 1ms/epoch - 17us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1355707.6250 - 1ms/epoch - 16us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 132308.0781 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 533594.8750 - 930us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5784931.0000 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 796707.4375 - 838us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5633699.5000 - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530616.7500 - 754us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1856942.1250 - 745us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1913973.6250 - 1ms/epoch - 21us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 946422.8750 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276378.8438 - 739us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1331390.2500 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1861867.6250 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414486.7812 - 686us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1532397.5000 - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504085.5625 - 868us/epoch - 14us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 877264.3750 - 826us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 230878.4219 - 881us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 350484.8438 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2116978.5000 - 1ms/epoch - 18us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618200.4375 - 863us/epoch - 14us/sample\n", + "Generation 81/300\n", + "Solving for Nash Equilibrium in Generation 81/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1735380.7500 - 824us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1320605.8750 - 726us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1109002.7500 - 655us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1665423.0000 - 693us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 358164.3125 - 850us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 199904.0781 - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1262844.6250 - 806us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1403875.0000 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523733.2812 - 788us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 496339.7812 - 729us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1040305.3125 - 4ms/epoch - 72us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2310898.2500 - 713us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 500869.5312 - 688us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2802808.2500 - 778us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3034301.7500 - 2ms/epoch - 30us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3701363.5000 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450602.7812 - 815us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1604303.2500 - 779us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1477092.3750 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1328828.6250 - 861us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1371979.3750 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1392133.7500 - 812us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 43043.1016 - 663us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2148740.7500 - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384274.6875 - 1ms/epoch - 16us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 483819.9375 - 743us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545919.2500 - 734us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 637026.6250 - 959us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2725873.2500 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 353370.0938 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 750684.7500 - 742us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1348980.6250 - 798us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4523305.5000 - 752us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418831.2188 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 617713.3125 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 755210.5625 - 746us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3006433.2500 - 847us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2006144.8750 - 750us/epoch - 12us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 843236.8750 - 875us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1352844.6250 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668429.6875 - 752us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 276780.5625 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 495754.4688 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726326.7500 - 622us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504905.9375 - 674us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623384.1250 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1388327.1250 - 655us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3380832.5000 - 825us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 243787.0781 - 1ms/epoch - 21us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 537570.4375 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 816624.2500 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 494377.1562 - 3ms/epoch - 49us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1898549.3750 - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 273544.2812 - 749us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 851745.7500 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1830069.5000 - 723us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1028954.1250 - 777us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 864998.4375 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2059219.7500 - 748us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1419743.0000 - 700us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532482.1250 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 583090.5625 - 664us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 656823.9375 - 779us/epoch - 13us/sample\n", + "Generation 82/300\n", + "Solving for Nash Equilibrium in Generation 82/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2757958.0000 - 700us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 182138.8594 - 689us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606416.0000 - 682us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3703696.5000 - 631us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 610058.8125 - 953us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1644251.5000 - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545792.3750 - 785us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 87941.5000 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522309.7500 - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1480491.2500 - 833us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506585.2188 - 819us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 420585.5625 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3033489.5000 - 669us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1461862.3750 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1072343.0000 - 2ms/epoch - 24us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 555412.8750 - 876us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 227701.0625 - 3ms/epoch - 53us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649320.7500 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1735931.1250 - 810us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 638108.0000 - 2ms/epoch - 29us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5222414.0000 - 940us/epoch - 15us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2114003.2500 - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 463853.7188 - 809us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1746389.1250 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456332.0312 - 763us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1475031.6250 - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 852132.7500 - 2ms/epoch - 38us/sample\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7284080.5000 - 2ms/epoch - 31us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1491515.2500 - 675us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 670754.0625 - 1ms/epoch - 17us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700246.9375 - 4ms/epoch - 59us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 387915.7500 - 720us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1291028.8750 - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 505466.6875 - 777us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1379412.1250 - 695us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235160.5625 - 869us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6496600.5000 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 264011.2500 - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 962835.1250 - 2ms/epoch - 30us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1516370.8750 - 955us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 927385.5625 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 852320.3750 - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 511634.0000 - 903us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 122102.8906 - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1579527.0000 - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 535856.7500 - 702us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 156698.2969 - 858us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 932074.1875 - 1ms/epoch - 20us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 539870.3750 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 819517.3750 - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 941378.8750 - 787us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 258189.5312 - 810us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522859.9688 - 742us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1888395.3750 - 1ms/epoch - 21us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 689572.7500 - 740us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 586646.8750 - 1ms/epoch - 22us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7022400.5000 - 688us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482520.2500 - 942us/epoch - 15us/sample\n", + "Generation 83/300\n", + "Solving for Nash Equilibrium in Generation 83/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651293.0625 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1078367.3750 - 2ms/epoch - 25us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1244304.5000 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 157424.6562 - 798us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1583686.2500 - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2220445.0000 - 1ms/epoch - 20us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290136.0000 - 3ms/epoch - 45us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 932628.9375 - 821us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1026789.3125 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2225381.7500 - 650us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 265481.7500 - 735us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316761.4062 - 826us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 67533.4062 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 936441.4375 - 859us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 402731.7188 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 712907.6250 - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1815943.5000 - 1ms/epoch - 16us/sample\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5190492.5000 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 293217.6875 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 513421.1250 - 724us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 530900.5000 - 991us/epoch - 16us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1563698.8750 - 974us/epoch - 16us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 439183.0938 - 1ms/epoch - 23us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3385983.7500 - 786us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675639.5000 - 1ms/epoch - 20us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 175379.9688 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1205659.7500 - 1ms/epoch - 17us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1609050.3750 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428990.3438 - 932us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 160198.5000 - 725us/epoch - 12us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 386420.4375 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1796963.3750 - 3ms/epoch - 47us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2002581.8750 - 747us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1623568.6250 - 782us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1817460.1250 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 308033.7812 - 643us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 621013.5000 - 848us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 209679.6875 - 771us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 191194.2969 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1124238.7500 - 759us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 507784.1250 - 753us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1161462.7500 - 882us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 232328.8438 - 638us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5220921.0000 - 682us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349532.8125 - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3900957.0000 - 728us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 538324.5000 - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 558654.0000 - 809us/epoch - 13us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2152439.0000 - 676us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 986026.8125 - 961us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3997164.5000 - 725us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 676337.9375 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1653387.2500 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 669874.5000 - 5ms/epoch - 73us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2723165.5000 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 677776.0000 - 1ms/epoch - 17us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 848870.9375 - 856us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 34005.9805 - 910us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 668225.6875 - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2059500.8750 - 850us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 615596.1875 - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1076253.8750 - 1ms/epoch - 21us/sample\n", + "Generation 84/300\n", + "Solving for Nash Equilibrium in Generation 84/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 864917.5625 - 897us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4508818.0000 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2872692.7500 - 781us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1643596.5000 - 888us/epoch - 14us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 222909.5469 - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1650547.7500 - 974us/epoch - 16us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1519788.8750 - 864us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1601702.7500 - 771us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2495890.2500 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1609832.5000 - 1ms/epoch - 17us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 713299.8125 - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1066224.3750 - 696us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 938529.9375 - 963us/epoch - 16us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3301903.2500 - 2ms/epoch - 34us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1748244.5000 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 635206.6875 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1438060.6250 - 856us/epoch - 14us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2384957.5000 - 724us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 950176.1250 - 4ms/epoch - 57us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3445375.2500 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 787040.8750 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5234836.5000 - 772us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1046220.3750 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 316486.7500 - 1ms/epoch - 18us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 349743.5312 - 715us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 28233.9727 - 932us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 282536.5625 - 838us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1803847.3750 - 671us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2260663.0000 - 711us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1180829.1250 - 810us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4018825.2500 - 999us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7768191.5000 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1140342.7500 - 2ms/epoch - 26us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1670266.6250 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 887670.1250 - 854us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1669375.0000 - 834us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5539494.5000 - 792us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 319575.7500 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2438549.0000 - 980us/epoch - 16us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 418716.2188 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1727358.6250 - 3ms/epoch - 51us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2451135.0000 - 836us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440949.3125 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 288341.2500 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1737928.2500 - 831us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 170555.1406 - 877us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2307621.5000 - 1ms/epoch - 23us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1582403.7500 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 631188.0625 - 740us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5019664.5000 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462121.9062 - 1ms/epoch - 17us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1743716.3750 - 931us/epoch - 15us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 922092.7500 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1261163.8750 - 2ms/epoch - 39us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 651968.2500 - 749us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2216343.0000 - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 446543.3125 - 1ms/epoch - 22us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 286088.7500 - 829us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 431866.9688 - 2ms/epoch - 24us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2263064.5000 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 479938.8125 - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1891313.6250 - 1ms/epoch - 22us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2522769.7500 - 827us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 460417.9688 - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 602455.6875 - 842us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512462.4375 - 905us/epoch - 15us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3754142.7500 - 901us/epoch - 15us/sample\n", + "Generation 85/300\n", + "Solving for Nash Equilibrium in Generation 85/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 493487.0938 - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2138611.7500 - 807us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2848364.2500 - 990us/epoch - 16us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 523140.2812 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 604403.6250 - 2ms/epoch - 29us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 176914.0000 - 721us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1810465.5000 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 764010.6250 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 609621.7500 - 670us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1132749.7500 - 656us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1609025.8750 - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 419011.6562 - 840us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1770015.0000 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3341657.7500 - 878us/epoch - 14us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 172715.3906 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2575042.5000 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 737231.3750 - 1ms/epoch - 17us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5285649.5000 - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2247407.0000 - 755us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1824961.8750 - 848us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 982182.2500 - 700us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1027406.9375 - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2819090.7500 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1653150.5000 - 786us/epoch - 13us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3203701.7500 - 1ms/epoch - 21us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1258355.6250 - 692us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5238938.0000 - 658us/epoch - 11us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2716979.0000 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 331743.0000 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 671363.7500 - 831us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4567937.0000 - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 942113.1875 - 754us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 298111.8125 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1148406.6250 - 706us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3302907.2500 - 5ms/epoch - 73us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 783609.8750 - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 299911.8438 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 626116.6250 - 870us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 649299.0625 - 666us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 719483.6250 - 905us/epoch - 15us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3033806.7500 - 874us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 14258269.0000 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 245628.0000 - 816us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 271386.0938 - 837us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5640271.5000 - 779us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2918104.2500 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1668802.5000 - 955us/epoch - 15us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1884384.7500 - 637us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 456714.0312 - 768us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1365488.3750 - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1178160.1250 - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 184623.4844 - 899us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 336857.4688 - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 646607.5625 - 707us/epoch - 11us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4076676.7500 - 893us/epoch - 14us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 800709.2500 - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1069122.2500 - 2ms/epoch - 29us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1281545.2500 - 725us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630760.1250 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1287134.5000 - 823us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312006.9688 - 730us/epoch - 12us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696931.4375 - 2ms/epoch - 30us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 828758.0625 - 784us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2716276.2500 - 743us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 382898.2188 - 812us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 693424.7500 - 1ms/epoch - 17us/sample\n", + "Generation 86/300\n", + "Solving for Nash Equilibrium in Generation 86/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 414685.7188 - 813us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2695846.5000 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1088470.5000 - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 338995.7188 - 756us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1860278.1250 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340339.3125 - 927us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 706364.6875 - 928us/epoch - 15us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1867926.8750 - 939us/epoch - 15us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1061807.3750 - 774us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 607268.0625 - 664us/epoch - 11us/sample\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13084313.0000 - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 889355.5000 - 732us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1335940.1250 - 704us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 691930.7500 - 3ms/epoch - 54us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2766153.7500 - 757us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1996407.0000 - 667us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2888614.0000 - 2ms/epoch - 25us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1324954.2500 - 799us/epoch - 13us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 499120.8750 - 826us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3054408.5000 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 696524.7500 - 671us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8567835.0000 - 679us/epoch - 11us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 189305.5469 - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2811923.5000 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 752858.0000 - 653us/epoch - 11us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 340972.0312 - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 190498.7031 - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2829078.7500 - 875us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1972636.7500 - 715us/epoch - 12us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1975684.8750 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1981813.3750 - 769us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1760472.7500 - 727us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1990163.7500 - 671us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1226893.7500 - 943us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 531122.6875 - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684772.9375 - 1ms/epoch - 17us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 191980.1875 - 782us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3993468.7500 - 766us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 783415.1875 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 482212.4688 - 663us/epoch - 11us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1337935.6250 - 815us/epoch - 13us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2140636.5000 - 690us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 623098.7500 - 715us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606379.7500 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3434807.2500 - 714us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6774180.0000 - 704us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 322090.3438 - 737us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4590045.0000 - 801us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2662451.0000 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2601991.7500 - 649us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 306064.1875 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2025197.8750 - 668us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1121956.3750 - 799us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 488251.8438 - 675us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 864089.1875 - 941us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700120.0000 - 649us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1021294.7500 - 747us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 217279.7656 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 699600.2500 - 2ms/epoch - 26us/sample\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3778504.2500 - 837us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7146391.0000 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 758326.1250 - 2ms/epoch - 35us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5213141.0000 - 2ms/epoch - 29us/sample\n", + "Generation 87/300\n", + "Solving for Nash Equilibrium in Generation 87/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1974151.0000 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1292586.7500 - 892us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2063124.6250 - 663us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2268501.2500 - 15ms/epoch - 244us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 930269.0625 - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 725276.5000 - 751us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 458697.4688 - 1ms/epoch - 17us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1191375.8750 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2611873.0000 - 808us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1160312.2500 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 650163.0625 - 769us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 933509.2500 - 3ms/epoch - 41us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2019198.2500 - 739us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1289073.6250 - 1ms/epoch - 16us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1062608.6250 - 837us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 359778.2500 - 666us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1976174.3750 - 698us/epoch - 11us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 64738.8711 - 659us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1213372.6250 - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1111796.1250 - 688us/epoch - 11us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2694455.0000 - 737us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 325353.2188 - 886us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2033970.5000 - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779161.1875 - 2ms/epoch - 38us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6357619.0000 - 620us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1216329.0000 - 627us/epoch - 10us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3122778.5000 - 865us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1068523.8750 - 696us/epoch - 11us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3565270.7500 - 746us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4251239.0000 - 743us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1088658.8750 - 733us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 916282.8750 - 695us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1515918.2500 - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1235666.8750 - 711us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1074188.6250 - 2ms/epoch - 34us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2610926.5000 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1080903.2500 - 804us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4714422.0000 - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1899338.5000 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 244963.7031 - 2ms/epoch - 37us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1192808.5000 - 978us/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 440750.9062 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7314650.5000 - 760us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2720517.2500 - 717us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3727869.5000 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392893.0312 - 804us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 773671.3750 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1205906.3750 - 677us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 453156.7188 - 781us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 512142.7812 - 700us/epoch - 11us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1280491.1250 - 1ms/epoch - 20us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1206268.7500 - 2ms/epoch - 30us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1864850.2500 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2434852.2500 - 904us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2050203.3750 - 685us/epoch - 11us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 464475.3125 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1227106.1250 - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1411494.3750 - 702us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8193980.0000 - 721us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1268522.3750 - 2ms/epoch - 35us/sample\n", + "Generation 88/300\n", + "Solving for Nash Equilibrium in Generation 88/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 104245.7188 - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4327070.5000 - 784us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 830958.3750 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1247925.0000 - 5ms/epoch - 77us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2871979.5000 - 719us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675394.8125 - 731us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 553019.6875 - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2350787.5000 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2221347.0000 - 695us/epoch - 11us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 465071.8125 - 653us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1471637.0000 - 628us/epoch - 10us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2414423.2500 - 865us/epoch - 14us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 157732.3906 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2008707.3750 - 594us/epoch - 10us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2394592.0000 - 672us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 370816.7188 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1116711.0000 - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3175738.0000 - 707us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 810796.1875 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 840785.3750 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2049884.2500 - 880us/epoch - 14us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2133013.0000 - 806us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2321691.5000 - 789us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 675233.3125 - 705us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1689706.7500 - 6ms/epoch - 96us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6215913.5000 - 795us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 394102.1875 - 861us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5057317.0000 - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3761088.2500 - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 808278.6250 - 886us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1139342.7500 - 709us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736215.0625 - 3ms/epoch - 53us/sample\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 779395.1250 - 725us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1554735.8750 - 1ms/epoch - 23us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8563183.0000 - 659us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1302523.6250 - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5843632.0000 - 798us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301434.3438 - 747us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2213045.0000 - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2076698.3750 - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1291914.8750 - 1ms/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3240025.2500 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 791102.5000 - 3ms/epoch - 49us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 323767.1250 - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3254080.5000 - 680us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 217948.3594 - 792us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3561611.7500 - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3270531.7500 - 788us/epoch - 13us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3078999.0000 - 851us/epoch - 14us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1372118.8750 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3277045.2500 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3885772.7500 - 865us/epoch - 14us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2205872.5000 - 711us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 667412.5000 - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 543132.7500 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2199067.5000 - 2ms/epoch - 27us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 484712.8125 - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 821996.6875 - 904us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2167064.2500 - 746us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4877661.5000 - 1ms/epoch - 19us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 521968.5938 - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5569487.5000 - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 683968.0625 - 852us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 384716.2500 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 10ms/epoch - 161us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2396452.2500 - 8ms/epoch - 123us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1190845.3750 - 781us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 125738.2188 - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2217639.5000 - 1ms/epoch - 22us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1558743.8750 - 737us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2409067.0000 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606615.8125 - 696us/epoch - 11us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 714309.5000 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 503044.2500 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 398583.8750 - 919us/epoch - 15us/sample\n", + "Generation 89/300\n", + "Solving for Nash Equilibrium in Generation 89/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 275893.9062 - 843us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7876454.0000 - 698us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2098855.0000 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 545005.1875 - 843us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 742954.4375 - 724us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2245720.0000 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 833718.0000 - 5ms/epoch - 74us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 832035.3125 - 874us/epoch - 14us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 389381.0625 - 716us/epoch - 12us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 580661.3750 - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1996826.3750 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1599279.6250 - 703us/epoch - 11us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 958624.5000 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2231700.2500 - 1ms/epoch - 16us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1053897.7500 - 795us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3078223.2500 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4592649.0000 - 779us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 158436.0625 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5477830.5000 - 725us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 392252.9375 - 662us/epoch - 11us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1584116.8750 - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2689299.7500 - 714us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 226672.6562 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 843048.7500 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3506673.0000 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1234934.3750 - 688us/epoch - 11us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 227924.5312 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1358239.5000 - 3ms/epoch - 43us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2896337.5000 - 677us/epoch - 11us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 922095.4375 - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8249751.5000 - 710us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 726546.2500 - 983us/epoch - 16us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1712559.6250 - 891us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425324.4688 - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5121426.0000 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2277771.7500 - 2ms/epoch - 24us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270581.1875 - 678us/epoch - 11us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4233912.5000 - 645us/epoch - 10us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2528148.2500 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 269980.2188 - 849us/epoch - 14us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5129347.5000 - 751us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 853943.6250 - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1876580.5000 - 1ms/epoch - 16us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 736816.6250 - 820us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4228341.5000 - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1662922.2500 - 841us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4532341.5000 - 2ms/epoch - 30us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1274614.7500 - 750us/epoch - 12us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1253690.2500 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1002026.5000 - 955us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 534923.9375 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2873272.2500 - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2376051.2500 - 669us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 566517.0625 - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462892.5625 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2591409.5000 - 651us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2387509.5000 - 706us/epoch - 11us/sample\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 968764.1250 - 745us/epoch - 12us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2803202.0000 - 712us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9318936.0000 - 4ms/epoch - 67us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409213.0000 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3685015.0000 - 764us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 246946.8750 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1271287.5000 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2424260.2500 - 596us/epoch - 10us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1457590.1250 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2427344.2500 - 727us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3724707.5000 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 878591.6875 - 692us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3029883.0000 - 1ms/epoch - 16us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1537899.3750 - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2435339.5000 - 637us/epoch - 10us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2481170.2500 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 863312.3125 - 749us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 33ms/epoch - 538us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 234444.8594 - 2ms/epoch - 25us/sample\n", + "Generation 90/300\n", + "Solving for Nash Equilibrium in Generation 90/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1400194.8750 - 812us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2638203.2500 - 759us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 347703.1875 - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3068314.0000 - 805us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 618566.2500 - 717us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2836486.5000 - 776us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1737663.0000 - 782us/epoch - 13us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1347855.3750 - 752us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 351008.8750 - 904us/epoch - 15us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5052177.5000 - 918us/epoch - 15us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2508613.0000 - 1ms/epoch - 17us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4557486.5000 - 773us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 923106.2500 - 886us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 759824.0000 - 1ms/epoch - 21us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 819773.2500 - 744us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4515752.5000 - 928us/epoch - 15us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1463844.7500 - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3126569.7500 - 708us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2271678.5000 - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2447741.2500 - 838us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8872943.0000 - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2756271.0000 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 240040.4688 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2007984.3750 - 837us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1431064.1250 - 805us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1345315.6250 - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2650136.5000 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3557210.5000 - 934us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 986495.5000 - 962us/epoch - 16us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 945236.4375 - 875us/epoch - 14us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1734009.6250 - 757us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5071383.0000 - 925us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1490209.3750 - 966us/epoch - 16us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1274082.1250 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 301775.7500 - 976us/epoch - 16us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2779830.5000 - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 928061.9375 - 980us/epoch - 16us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1398688.1250 - 862us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 912827.5000 - 812us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1470449.2500 - 882us/epoch - 14us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10782498.0000 - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1847346.3750 - 933us/epoch - 15us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 700154.1250 - 844us/epoch - 14us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2622018.7500 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1346075.1250 - 801us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1520400.3750 - 647us/epoch - 10us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2824936.2500 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 951237.2500 - 832us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 438248.1875 - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 907513.5000 - 2ms/epoch - 38us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2124561.5000 - 915us/epoch - 15us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3845203.5000 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 510080.6250 - 762us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2005118.2500 - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1517141.6250 - 805us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3511645.0000 - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 838893.1875 - 675us/epoch - 11us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4893953.0000 - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1363903.0000 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 724334.0625 - 703us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 791164.4375 - 893us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1489761.8750 - 695us/epoch - 11us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 793706.5625 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1531737.7500 - 741us/epoch - 12us/sample\n", + "Generation 91/300\n", + "Solving for Nash Equilibrium in Generation 91/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1378150.8750 - 911us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3532740.2500 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6976729.0000 - 695us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3508055.7500 - 683us/epoch - 11us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506949.3750 - 706us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3999305.5000 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450342.7812 - 703us/epoch - 11us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 673207.3125 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443139.9688 - 855us/epoch - 14us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4056435.0000 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1576477.3750 - 2ms/epoch - 36us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 606440.9375 - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2711974.0000 - 655us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1317255.1250 - 801us/epoch - 13us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1602548.5000 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4189763.7500 - 797us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1508807.3750 - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1415595.7500 - 739us/epoch - 12us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3281247.7500 - 824us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 663814.4375 - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1136126.2500 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1652620.0000 - 821us/epoch - 13us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 965862.1875 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3005397.0000 - 783us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1259056.7500 - 1ms/epoch - 17us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7893916.5000 - 746us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1578635.2500 - 5ms/epoch - 83us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4598023.0000 - 724us/epoch - 12us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2886109.2500 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450447.7812 - 1ms/epoch - 20us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1537332.2500 - 772us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1600269.5000 - 647us/epoch - 10us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3099933.5000 - 3ms/epoch - 41us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1050609.6250 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 405842.2188 - 1ms/epoch - 22us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 945403.5000 - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 624626.0000 - 2ms/epoch - 27us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 982353.6875 - 816us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5761703.0000 - 966us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8890793.0000 - 783us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3600211.7500 - 2ms/epoch - 30us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2943232.0000 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 658073.8125 - 713us/epoch - 11us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3071318.0000 - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2900613.7500 - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 630807.5625 - 709us/epoch - 11us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1571670.5000 - 2ms/epoch - 30us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1869314.1250 - 6ms/epoch - 100us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1050018.2500 - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 7ms/epoch - 114us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 816767.3125 - 4ms/epoch - 62us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1050681.5000 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2929456.0000 - 776us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 409056.4688 - 1ms/epoch - 18us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1495846.7500 - 1ms/epoch - 19us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 277780.4688 - 947us/epoch - 15us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 695461.4375 - 3ms/epoch - 52us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1003679.0625 - 775us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2425221.7500 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1164473.5000 - 759us/epoch - 12us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1515473.0000 - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1626732.6250 - 1ms/epoch - 18us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1142694.1250 - 891us/epoch - 14us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5732936.5000 - 830us/epoch - 13us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2831058.0000 - 829us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2838483.2500 - 713us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489878.1875 - 3ms/epoch - 51us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1956323.5000 - 1ms/epoch - 18us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 261052.8281 - 865us/epoch - 14us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2151376.2500 - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1766170.7500 - 900us/epoch - 15us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 462903.9062 - 1ms/epoch - 17us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 847862.3125 - 994us/epoch - 16us/sample\n", + "Generation 92/300\n", + "Solving for Nash Equilibrium in Generation 92/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1842797.3750 - 880us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 816438.2500 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 979749.1250 - 893us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5048198.0000 - 781us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 994105.6875 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1075168.2500 - 1ms/epoch - 23us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5606321.5000 - 753us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1852235.0000 - 1ms/epoch - 23us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 519770.7500 - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4287321.0000 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2796009.7500 - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1652938.2500 - 799us/epoch - 13us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 412762.4688 - 3ms/epoch - 55us/sample\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684230.6875 - 995us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3345509.2500 - 753us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6432269.0000 - 878us/epoch - 14us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1641860.7500 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 471238.6250 - 841us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1088138.2500 - 737us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3048033.0000 - 746us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2119450.5000 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1679973.6250 - 740us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3682257.0000 - 717us/epoch - 12us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 572402.6875 - 961us/epoch - 16us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1050616.5000 - 755us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1740198.8750 - 701us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3343419.7500 - 746us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1706661.0000 - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17393354.0000 - 695us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 866167.3125 - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 5ms/epoch - 84us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1743112.8750 - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4784669.0000 - 2ms/epoch - 30us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3820003.5000 - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1572588.0000 - 812us/epoch - 13us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1595045.6250 - 799us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3508287.2500 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2869507.2500 - 778us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1884848.7500 - 819us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1262476.6250 - 790us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1033834.0000 - 739us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 270986.6250 - 735us/epoch - 12us/sample\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5774730.0000 - 777us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2917693.0000 - 866us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2892150.0000 - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1781032.7500 - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 682416.6875 - 1ms/epoch - 16us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6988473.0000 - 677us/epoch - 11us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5843132.5000 - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3126333.2500 - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1619035.5000 - 632us/epoch - 10us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1103769.2500 - 712us/epoch - 11us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4388027.0000 - 881us/epoch - 14us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1230859.1250 - 869us/epoch - 14us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4929866.5000 - 713us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 568728.4375 - 793us/epoch - 13us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2047660.8750 - 2ms/epoch - 32us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1160476.3750 - 828us/epoch - 13us/sample\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 425640.7812 - 912us/epoch - 15us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4069567.7500 - 716us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1068544.2500 - 4ms/epoch - 71us/sample\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 561392.6250 - 796us/epoch - 13us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1254589.5000 - 777us/epoch - 13us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 360848.1875 - 782us/epoch - 13us/sample\n", + "Generation 93/300\n", + "Solving for Nash Equilibrium in Generation 93/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1072710.0000 - 707us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 320863.0625 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2706849.7500 - 689us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3232360.0000 - 848us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1389186.1250 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 170567.6562 - 606us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 490867.3125 - 2ms/epoch - 34us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 492674.3125 - 813us/epoch - 13us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7202548.0000 - 9ms/epoch - 144us/sample\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 69587.0469 - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 11154826.0000 - 703us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 909980.9375 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1273956.0000 - 763us/epoch - 12us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1922558.2500 - 753us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1843395.5000 - 697us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2738549.7500 - 993us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4521010.0000 - 787us/epoch - 13us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3554431.0000 - 780us/epoch - 13us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4769371.0000 - 3ms/epoch - 45us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 489268.8438 - 980us/epoch - 16us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1664354.8750 - 734us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1214127.5000 - 6ms/epoch - 98us/sample\n", + "Episode 14/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 310050.0000 - 826us/epoch - 13us/sample\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3043267.7500 - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2017299.2500 - 719us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2676759.0000 - 697us/epoch - 11us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4710549.0000 - 735us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7882968.5000 - 690us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 428307.5625 - 8ms/epoch - 134us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3199946.5000 - 683us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1087930.6250 - 653us/epoch - 11us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1171771.1250 - 733us/epoch - 12us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1138291.0000 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 880535.6250 - 878us/epoch - 14us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1795311.0000 - 714us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1939102.8750 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3075805.5000 - 757us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5458575.0000 - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1103659.8750 - 736us/epoch - 12us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2003879.6250 - 786us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1329179.6250 - 776us/epoch - 13us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1777385.0000 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1802043.7500 - 2ms/epoch - 27us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3388935.5000 - 755us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3149479.5000 - 771us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 443688.3750 - 742us/epoch - 12us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506516.0938 - 832us/epoch - 13us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1196157.0000 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3126062.7500 - 934us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1972601.6250 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 321396.2188 - 775us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1556409.2500 - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4721376.5000 - 822us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4375911.0000 - 961us/epoch - 16us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 934227.6250 - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2080889.1250 - 835us/epoch - 13us/sample\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5522916.5000 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2522631.2500 - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 366265.5312 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2264183.7500 - 832us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 290466.6875 - 733us/epoch - 12us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5403199.0000 - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1815189.5000 - 797us/epoch - 13us/sample\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1209971.7500 - 1ms/epoch - 18us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12006691.0000 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1120702.7500 - 959us/epoch - 15us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1917050.3750 - 747us/epoch - 12us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2231620.2500 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1995588.0000 - 2ms/epoch - 27us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 522384.6562 - 1ms/epoch - 18us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4464878.0000 - 973us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1939093.8750 - 984us/epoch - 16us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5307339.5000 - 834us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1132748.2500 - 739us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 455057.1875 - 878us/epoch - 14us/sample\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 895529.5000 - 892us/epoch - 14us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3916636.5000 - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1847007.6250 - 756us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 709150.8750 - 841us/epoch - 14us/sample\n", + "Generation 94/300\n", + "Solving for Nash Equilibrium in Generation 94/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2016174.1250 - 807us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 47179.7383 - 923us/epoch - 15us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6215709.5000 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1175989.8750 - 794us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7365395.5000 - 5ms/epoch - 77us/sample\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 113273.2031 - 837us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1776806.6250 - 741us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 520736.3750 - 648us/epoch - 10us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1456807.1250 - 754us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7186700.5000 - 694us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 684148.3750 - 732us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6835306.0000 - 812us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 58146.2578 - 3ms/epoch - 47us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3203007.2500 - 763us/epoch - 12us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 920770.1250 - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1553979.1250 - 2ms/epoch - 30us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2069223.0000 - 774us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1186204.6250 - 681us/epoch - 11us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4931449.0000 - 719us/epoch - 12us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1951120.7500 - 751us/epoch - 12us/sample\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2608047.0000 - 732us/epoch - 12us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7893049.0000 - 664us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6542754.0000 - 795us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 300090.9375 - 721us/epoch - 12us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1169805.6250 - 781us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2705672.0000 - 798us/epoch - 13us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 847423.6250 - 747us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2028739.8750 - 773us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3518572.5000 - 865us/epoch - 14us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2637563.5000 - 619us/epoch - 10us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3234750.0000 - 803us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3525117.7500 - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1813333.0000 - 936us/epoch - 15us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 71459.3984 - 894us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1821241.8750 - 806us/epoch - 13us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1285356.6250 - 834us/epoch - 13us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3737057.2500 - 2ms/epoch - 32us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1786758.5000 - 811us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 938142.3125 - 685us/epoch - 11us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4498871.0000 - 12ms/epoch - 187us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3285001.5000 - 953us/epoch - 15us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7368901.0000 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1220559.0000 - 6ms/epoch - 101us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3301237.2500 - 736us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1189807.8750 - 753us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1212280.8750 - 780us/epoch - 13us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1964232.2500 - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1302228.8750 - 727us/epoch - 12us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2385866.2500 - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5599609.5000 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3596822.2500 - 2ms/epoch - 26us/sample\n", + "Episode 35/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 305702.2812 - 676us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3567725.7500 - 853us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3362272.5000 - 716us/epoch - 12us/sample\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2060412.3750 - 736us/epoch - 12us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3978377.2500 - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1826685.3750 - 804us/epoch - 13us/sample\n", + "Episode 38/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4109744.2500 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342984.9375 - 1ms/epoch - 19us/sample\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 940336.0625 - 750us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1965228.8750 - 2ms/epoch - 26us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2077803.5000 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1870109.6250 - 976us/epoch - 16us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2014719.7500 - 744us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3636530.7500 - 701us/epoch - 11us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 967819.9375 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 760278.9375 - 810us/epoch - 13us/sample\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5281079.5000 - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1956820.3750 - 2ms/epoch - 32us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 784526.6250 - 724us/epoch - 12us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3510739.2500 - 1ms/epoch - 22us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2274401.0000 - 745us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2305306.5000 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8019161.0000 - 731us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2707885.5000 - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 858160.1250 - 739us/epoch - 12us/sample\n", + "Generation 95/300\n", + "Solving for Nash Equilibrium in Generation 95/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 767454.5625 - 729us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6242885.0000 - 2ms/epoch - 27us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5224380.0000 - 2ms/epoch - 38us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 532415.8125 - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 312234.7188 - 827us/epoch - 13us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2327205.0000 - 719us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2141260.0000 - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1838247.1250 - 1000us/epoch - 16us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8949283.0000 - 666us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5271310.5000 - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1246608.5000 - 745us/epoch - 12us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2331874.7500 - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2040474.1250 - 1ms/epoch - 19us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3742262.2500 - 694us/epoch - 11us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2066764.7500 - 790us/epoch - 13us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2054553.1250 - 892us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1388513.6250 - 887us/epoch - 14us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2496103.2500 - 835us/epoch - 13us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 941143.3750 - 5ms/epoch - 74us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2130902.7500 - 846us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 785938.8750 - 727us/epoch - 12us/sample\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7514443.5000 - 794us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1343886.6250 - 800us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7314212.5000 - 908us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2147674.0000 - 710us/epoch - 11us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3719731.2500 - 4ms/epoch - 60us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1232073.2500 - 731us/epoch - 12us/sample\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1235474.1250 - 921us/epoch - 15us/sample\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2044110.6250 - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2877634.7500 - 815us/epoch - 13us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3507183.7500 - 743us/epoch - 12us/sample\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 864091.6250 - 864us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12613673.0000 - 830us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9582343.0000 - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1985513.7500 - 791us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 844910.8125 - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1246620.3750 - 771us/epoch - 12us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13303305.0000 - 754us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2553359.7500 - 667us/epoch - 11us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2578156.2500 - 741us/epoch - 12us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1408958.3750 - 1ms/epoch - 17us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8992591.0000 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4360431.0000 - 838us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1255649.0000 - 712us/epoch - 11us/sample\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450577.5938 - 755us/epoch - 12us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4408254.5000 - 687us/epoch - 11us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12872404.0000 - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2557510.5000 - 618us/epoch - 10us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1942549.6250 - 821us/epoch - 13us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2380300.7500 - 707us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 797985.7500 - 728us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10342834.0000 - 816us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3567996.7500 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3867475.7500 - 673us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1434583.0000 - 737us/epoch - 12us/sample\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4845750.5000 - 768us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2142583.2500 - 747us/epoch - 12us/sample\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3571691.7500 - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2253468.7500 - 704us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2088598.6250 - 725us/epoch - 12us/sample\n", + "Generation 96/300\n", + "Solving for Nash Equilibrium in Generation 96/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2499857.5000 - 734us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4050578.5000 - 883us/epoch - 14us/sample\n", + "Episode 4/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 121301.9609 - 749us/epoch - 12us/sample\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9982190.0000 - 1ms/epoch - 16us/sample\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2815522.0000 - 723us/epoch - 12us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2262914.5000 - 666us/epoch - 11us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1266560.0000 - 3ms/epoch - 48us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5542151.5000 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 450060.0625 - 699us/epoch - 11us/sample\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2572144.2500 - 902us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 235173.9062 - 728us/epoch - 12us/sample\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 13818174.0000 - 700us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3705027.0000 - 1ms/epoch - 21us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1204632.3750 - 672us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2250725.5000 - 758us/epoch - 12us/sample\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2121247.7500 - 958us/epoch - 15us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3679446.7500 - 810us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 317987.7188 - 847us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3660298.0000 - 935us/epoch - 15us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1296751.2500 - 966us/epoch - 16us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2497382.0000 - 798us/epoch - 13us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 8638864.0000 - 905us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 504004.3438 - 1ms/epoch - 19us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2392141.2500 - 849us/epoch - 14us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 10839182.0000 - 1ms/epoch - 19us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 377606.1562 - 1ms/epoch - 17us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2401782.2500 - 1ms/epoch - 17us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 7902905.5000 - 840us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4237547.5000 - 800us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2035633.2500 - 870us/epoch - 14us/sample\n", + "Episode 30/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 506494.5312 - 793us/epoch - 13us/sample\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1304937.3750 - 919us/epoch - 15us/sample\n", + "Episode 33/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2240468.7500 - 784us/epoch - 13us/sample\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1693302.8750 - 993us/epoch - 16us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2639029.2500 - 2ms/epoch - 32us/sample\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 191807.5781 - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2904265.2500 - 751us/epoch - 12us/sample\n", + "Episode 40/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 9457348.0000 - 1ms/epoch - 17us/sample\n", + "Episode 41/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2115359.5000 - 1ms/epoch - 18us/sample\n", + "Episode 42/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4305203.0000 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1387843.8750 - 3ms/epoch - 46us/sample\n", + "Episode 43/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3435792.7500 - 923us/epoch - 15us/sample\n", + "Episode 44/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3109019.2500 - 731us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5666734.0000 - 993us/epoch - 16us/sample\n", + "Episode 45/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1501818.6250 - 723us/epoch - 12us/sample\n", + "Episode 46/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 585814.7500 - 828us/epoch - 13us/sample\n", + "Episode 47/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4815975.0000 - 1ms/epoch - 18us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 395103.9688 - 1ms/epoch - 23us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5292211.0000 - 672us/epoch - 11us/sample\n", + "Episode 48/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 587455.8750 - 820us/epoch - 13us/sample\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5554949.0000 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3764589.0000 - 898us/epoch - 14us/sample\n", + "Generation 97/300\n", + "Solving for Nash Equilibrium in Generation 97/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5871947.5000 - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1474525.2500 - 749us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2445722.7500 - 741us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2560065.7500 - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 969684.0625 - 757us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 17110172.0000 - 752us/epoch - 12us/sample\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2415662.2500 - 3ms/epoch - 45us/sample\n", + "Episode 8/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5682156.5000 - 877us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4296715.0000 - 813us/epoch - 13us/sample\n", + "Episode 9/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1352037.5000 - 777us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1278497.5000 - 2ms/epoch - 26us/sample\n", + "Episode 10/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1512344.5000 - 745us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 593658.5625 - 5ms/epoch - 88us/sample\n", + "Episode 11/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2344798.7500 - 895us/epoch - 14us/sample\n", + "Episode 12/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4319715.5000 - 743us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2914878.0000 - 1ms/epoch - 23us/sample\n", + "Episode 13/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2145114.7500 - 869us/epoch - 14us/sample\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4132423.2500 - 739us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2331868.5000 - 632us/epoch - 10us/sample\n", + "Episode 16/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 342000.0000 - 775us/epoch - 13us/sample\n", + "Episode 17/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 6692617.0000 - 2ms/epoch - 29us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1887120.6250 - 869us/epoch - 14us/sample\n", + "Episode 18/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1671615.6250 - 3ms/epoch - 43us/sample\n", + "Episode 19/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 5032903.0000 - 858us/epoch - 14us/sample\n", + "Episode 20/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 883102.4375 - 877us/epoch - 14us/sample\n", + "Episode 21/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1375462.7500 - 796us/epoch - 13us/sample\n", + "Episode 22/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2173505.2500 - 749us/epoch - 12us/sample\n", + "Episode 23/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2388224.0000 - 946us/epoch - 15us/sample\n", + "Episode 24/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4283433.5000 - 855us/epoch - 14us/sample\n", + "Episode 25/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3562467.5000 - 759us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 12276266.0000 - 635us/epoch - 10us/sample\n", + "Episode 26/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1489430.7500 - 740us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3946604.7500 - 925us/epoch - 15us/sample\n", + "Episode 27/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1391791.0000 - 788us/epoch - 13us/sample\n", + "Episode 28/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1685692.3750 - 960us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1840143.2500 - 881us/epoch - 14us/sample\n", + "Episode 29/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 16017575.0000 - 1ms/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 3748236.5000 - 691us/epoch - 11us/sample\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 4269057.5000 - 803us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2517508.7500 - 869us/epoch - 14us/sample\n", + "Episode 32/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1750924.3750 - 840us/epoch - 14us/sample\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 200356.9844 - 955us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1181988.8750 - 779us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2370348.5000 - 645us/epoch - 10us/sample\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1327120.8750 - 972us/epoch - 16us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 2433162.0000 - 810us/epoch - 13us/sample\n", + "Episode 37/50\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", + "Train on 62 samples\n", + "62/62 - 0s - loss: 1022126.4375 - 744us/epoch - 12us/sample\n" ] }, { @@ -8485,7 +128539,7 @@ ], "metadata": { "kernelspec": { - "display_name": "Python 3 (ipykernel)", + "display_name": "Python 3", "language": "python", "name": "python3" }, @@ -8499,9 +128553,9 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.10.10" + "version": "3.11.5" } }, "nbformat": 4, - "nbformat_minor": 4 + "nbformat_minor": 2 } diff --git a/main.py b/main.py index a5cd0ec..93c22cd 100644 --- a/main.py +++ b/main.py @@ -14,7 +14,7 @@ info_df = pd.DataFrame(columns=['Actions', 'Observations', 'Information', 'Rewards', 'Done']) info_df_dqn_training = pd.DataFrame(columns=['Actions', 'Observations', 'Information', 'Rewards', 'Done']) episodes = 50 -generations = 300 +generations = 100 max_steps = 50 runs = 10 From 9a19573b631b4869ca258ef5d563985a343f720b Mon Sep 17 00:00:00 2001 From: Gabe Smithline <63967324+gsmithline@users.noreply.github.com> Date: Wed, 1 May 2024 23:15:16 -0400 Subject: [PATCH 2/5] added more --- __pycache__/main.cpython-311.pyc | Bin 18660 -> 18687 bytes __pycache__/ppo.cpython-311.pyc | Bin 10553 -> 9929 bytes experiments.ipynb | 226076 ++++++++++++---------------- main.py | 14 +- ppo.py | 203 +- 5 files changed, 98339 insertions(+), 127954 deletions(-) diff --git a/__pycache__/main.cpython-311.pyc b/__pycache__/main.cpython-311.pyc index b3673f1ef5dcd67d883210c2b135da082e33864e..b98127c614ea845460cff0543d117d10fc5577a2 100644 GIT binary patch delta 429 zcmaDdk@5dTM&9MTyj%=GaO<;CTJ=QU2drFCk|~@UKXEcL+HC&ASjfxBF*%Q4kMYdp zx%`(I6E+tLc(O2S*RW3x6c(F&SwxnZA&YggfWFLRQBg(Kgb7SN=Qdl5_AxQ`Z9XM# z%FJK9rRJiN^F;xdD*`ST1Y9-?OYe|kI>5O3gRKeUWJYs6tr9tq?GRAISi`W4m4RV3 z5JNx=69YpnS1or9^8$g%i~>rN6>JnIPcWawv{G=gp+yi=(Td3p78@D$CNHt6nQZUE zGkKR~8qZOa6Cx){F3N;lkqMbR$2JIP-{cRrHz$W%cam&4Oq#+#hf7X=;N${wyrT0$#`7S@ kZ?<<~VdOnpaw6nHaQH=;h$}J?n|)n@;mEc5jyne<00pyuDF6Tf delta 409 zcmexAk@3kyM&9MTyj%=G!1dN3En_0@16GbG$rP@QpEwyAxii_9$(suWJXx6aYS5_96kGl9d9JbuEG>w^*!W)S0}_re^Yb%Ob7=Mh8kR z%J^K7@tLe^6~wq@@Op_a}Zcd(U7QNZP-inbcMG$8Gbg<;)&rU9r oZJg&Zo`E`=XL7R353U16CrmEN1YD5`*qq@C3^&fr-`zPF0TR!F#Q*>R diff --git a/__pycache__/ppo.cpython-311.pyc b/__pycache__/ppo.cpython-311.pyc index f1cb8ee247510408100951e7601a39549ffaaf43..63dacf0eab82a1101c2f74301a2159dd47dea3ba 100644 GIT binary patch literal 9929 zcmdryTWl0pmR0?{{j#f#-S`DI*ajMm%^U1Y;&=ivAw04`l8u?9)pnIln|`IL9E|DC z@WJeMMxGtQdW6F;qO^(_BiIoitKEF;O5|fDR@#rQR%zWT$y!#c(TJ}#OFp8}YG=>6 z)m`1y#tg~d?CtJT_i^gnTlb!K^{)c~F9YQt{(jHIogl;fCpOB(S*kqyJE+`Xc!p== z%&2i?6Sh$sYt-!t$Ec%JcaFLM&lYzl$S5iGc}6|-UT?xT>Sq}nGs5ups|@cDY*T^q zWOwWga}Rz-8LgvOPQY@#0IQy2xdDsZvBCZDE8WlW>=`EPd4kQ8VODqY!pGCXcxjkk z%E-^Uq4}5afC4kh^313$$4sz1dzFmZdD~TH)RALm*ik2tv+K^`aXBT9o9l!-9q_~U zvn?#c{DQr~$Z#zexjYGUWEkrjm8!8aKEvFB>OHf`GP2ujFjF`eEg4v#m0E_0l+SC8 zd-&(w<0?~p{yaRJv8|fNT#@A-Ql`Y)?%VIc9^5nc0oLX*M;X{i$H^zn(DCFCP~dRv zd@K?6>P~4oEr=pMhE5`pSTZI@BDyV^)LqA8Ng*og_KES`2=?f%)U=$QmUZ`;)P$Uf z&ggE*sDvH5LlWW>x>u5;vJjDCvx4r6j?1xB(r9=hk@0v`k|Gfa_p`IJ^UCiJ^}j0# zqSQYb9TTKPOg4K7S5A&tOhw z+~g8y;AT?p{BNjOd2_1je4fvX8Pf9S=5!0SYWaY9jq=B+Wf-M;<>lKzZWNQV z%}%KN5qge{;|z0`JsEcCPL$d6xTy(b*Z@8|6QVCgtd) zAnBe|oR7e9k4d_dUL;%x@JnZ>xE}G{6DjdxRODv^#`%|;NLM-w08ZZ@oI7?+zAFGn^u??0lHpFc^rtErL6nDi}H27A;#p`_;~qVB!(Pg z^7sEg?aRMF`|_ws?I%T4$4~Gm!$d^QBZd*6UK0HX(8N}8J!N=1Wq7d>hbSxH9)cn? zn1SlFdSkaurPcx%(vJX?l-k5KUL|cBXG}Hn@oG2r8PX`iwTfkna2>8lM&<x(^7B>3AdFPJHNKvAUuD5pBj~H~9)M2X3(&>;0J?cUK*9$AdiXjJ z8BQnfymM-JQb@{AP?h}dHzWTs+xVOR8v5PMFR=W}(35T0{Xch~7EX-4GxTITuEYbN zTui0Y0QcVjLj(>^d5p=}poQB+OU4c@Q(Z7bh<_B5z$C5|IPK&4#|k8FlAmG?9qv-tl-W9TB78o$B7Ps62k2 z`lyBv>;$hhB~o>+yT+;Q)19fb982hqVLqDBz3He3J^*;8l3q`dBXk&CUQy?aJF0qI z2_caZFX^s$N|M08aSLfF7EdMhfSi(}@rWp(Gb8CPqXM3H8tTy`pMnV`LEx#b(tU|& z5`2sZ_JloEh6Eo>JcM8X0g9`t&OoTlZWis#<} z{R&fTMVGj~*aj9uBhAI8R%m+5y~f~jQ|rQ@a#?NKt2OP#;pI^K!l)YBriHelomuYa zzBRkpeSiDX_MyV|A$9w(wte{F`FzJIwd0i5aSAE7FL(BQ>0R7+|COa3!-XBg>W(+G z9dA7H<;(guX!(m9I&iSH7+OzfBi=dB&+C3#cTHAFJ5bI8 zaT3r-%L3TU4vlo=%U{FzRr|UqsHiTUJpzS-ibjFP>Vi?4Vyc|#aflM3zY>XAW4Wv| z<4~&6&HDGI=#s0}KV_;qORuT&GEVF2*1&REv1UP)S$neXj5}kWg_*O&Lc?53#tlzY zrTCCWO%FV~H3w0m=&?{lXN#5$$#|^0shBFSjDy!Z@VjSr_|RvwofLOKPuQcoCZqCs zkX6rgQiAk^0CJ1&fNmFN(z-pKn$(>TswUD@pmcj8med^yA({+(4YBJ)WH*Qy1h-dnwY+jPI>!G1M7poIt2)&Z3p)VM)#Ke(m^ zugYyU@O=wMl--lQ-iM~qaq6*F-o4ybDF{aAN zIIMdCsvVU31jHG4LETXI!&5TfZ^lzm`K*wXz_*|Q;3-&z?oOs6lVX$?U%{3$Ih}yy z)(F;JQE?J7Y+;w7b3Vd8PckZ3YNIT1@7>}KU298jT-?$4>Zhu_*rBw zvcj;QJroohn{OPuer#b{;TI39jk~qR-MJ$x_VO^rd5YNIn734JuBm1czYVUNZ&`R%4TiN~IOi(XH(kr->$~%$`x$zkUz6tH zUi;U*j^SPIulKS5Ex|2r5mi91{pkyU?K=v9?PUZ&G3_nvlv%7olC@{-pg<>K7&0I< zoIVfr3U7-sY$9Ka7Xavvm@Fj1jtV=A2^^zLi|PQG)?isLFey0XiF90;U02PyrOp$8 zE&Vfq@6W)4$X$cQ_RibKZXQ#n7x@Q=)%F3ceIR!N8Q2)6ILN?*$iP*7bf5;57`PVe zad6{Ou%{60QG;8x;MS#JZz0&L2K%&NALZLzKup1ti~}c{1=S1XPX+eHs-VFPnwVDA zQ*9t|zhP^EH_oi_R-&3t^_^Bl3QF5YNl}nt|BCq$tSxKH+Ov+VGwaH_vjluUGh_tS zXmSpO(y-Es51h4G@G2^_=B-86Zo#Y2t5vm*HSlV@+pKd9yc+*D>skZPUGp4nFi_R* zjJZLpp4hV{X6@WwFdIxISpYt=8+~OoJsI!Igz|_v3oFU`GQN!GrS1UwSwF1MOe&ZU z0u^wp{9xRG3oCC2vjM=aI_nC4CIC(t;sd|lLs_OJTL;f%CJ>Y=16szY_0z#X*C}RN zp@Icb!>YGi=qOZCH`5Q*cUd(Sz5|teD&tdTqQ;8%a{HI@)yPUctf!LTc)6#ofBw^2 z*F&GQLiweh6xL#MSH*{GuR7V@dRA z(&xj?5K9S3>RZ#Sgm@I@(OnWmR&hbM&&JXfv!Q#Br;?K;_vi87mda)FrQ8RK5 z`(H;Q9thbC$4Mt+(eVqIL4jn6ETn1jU3Z^2bL2R<;Nmde@&mhXGcvx4ix2Oj(8rf_Hln+Xtgvnjml|#v{53%?g@c_y3|?wsB8f(K zy04sufkjEk;b`rui(&a72b7LSFU{7QR?p}-3wKEW3;PZh&sUd|=Ps+hHqF?>_RWN=X@M&y3uyMZQ(sN*sTS-!QO>hZ(O*3 zLD`{(`m|7A&iA;XW#RCxV~d*>g}Y(3ecyx2O9zh?4jxkv9@h>YSKE)P4aaltVpIFV zC$~Sn`Kj8`uXXgRP5n9VV|Zcbz3yE&pbRcvDu=kmrj{EYUjI*;q!~Tp*iO(ybBnSYdT$us3*n=;n~x)~mJks${1|cIL^>Ql6{w zp2ieEUJvfiJR3_0y8B2vCZ%{mci=$S4FPvthUh$= zlBu-AyXTk;O2#!4hrbrYb=M{1>X6Q-E{0jE0O(7VlE>-8JE^1!(rlwCd{w;{VTt9p z%fvTFt~JgB(`|A8L?;>cWn01Zd2RSB_LQ3fN2{5D|Qa;I9GPW#|M{#fbgb zi(m`>w9uG87F z!kFMq*6Cb_!?nOzA}%Z(G(u)V|3A=|Od2}fXEbT5U3?YrM9do*!Ln~W6^|Ra7fD1e zHf&8W5tZ=_{r~{t1ePr_{wwrTWSa8UZ;@%sTfar7Az%I$nGJdCx5xyp&`*)+$XmZFjsOcU x{8xkA59i)`iv8bI1~>W?t`E`ftpXf@L}JWF8VXDVAbQaca;+VtUXF zF&(!gtb-60yn{Yc zW{cM)>IdtW>-~d%*3S9I0;R>U%sxoc6jy%-;(MiH$^>PHDb{g~Vx4zrsDXbi3p#JU;7>(z~VpO{chp-(8<2X&q6Nh08k2M zNZ|jIlO|kM#+0GLr8+7@Lk`NdkTFB9vJ-F?$c2U26sNi(QN$lD zV;zYk5)suIjmOeqJ^~f0b0{K4N0-|UO~m4C7<#05)v85qtt`sN#8_0dmEr`KNb!@Z zEuIntPEhS!T8PB~O+8{tjKsq{cPYZNf@;$eszu;9R;80^D2XK5R3hZSviLBM^u_}t zp!jtN8W1!BP#H3RIL@mMZBj_`>!C)#+h{L^A71SFxxn#4&q!p56B03TG#*QGJ%>19 zTuh~V1TMh|BFFFQdHq;Vnoo^!QL!hTPIadzRVEycC1YYZJmo84cz3A?Z$-EX;3`#Y z+K{u{s4X^c%2^bqp%`d_tg}?QTp4Iw3~XEoY%Gk*fgUB$Lz)De7K57?f}3X@a&Ws6 z+zy4VVz8kyTGtFYcd@k%$8IVH*O5v;wsAaSoHraF)qGepEzV5JOozgBNTu%yERBll zlmZ#VY(iz7Pl>S$ajinWn@FS~If43~e0ER6Vm!zpKn_ zfsu9E0(vsDmityCQ>`99Yki77R^ytg`v6Vpt72^{W~}aQ6LlJgv!sDXSw_qm+lcAD z-MH4QJ!3EE+y-x`6hxhdf#P-EGN$i+Oy?;@CHf@gb(2uO#r6+Ivc29gzZoxw8Qyar5y8_a>5Vee01QK78XncH%Zo}| zkMztoscySmw_T~*p0j`B-tw#d>CQq>a&M8{TNL+}*?{csQQSQ_Q*mXTPj+`H?k=eF z?*G&}%__l7bGs$)e%ZTU@$R3G$ljL}?@KxBx3EgCFgUkPYTf;y|3RKqIB+5I`=Jh9$e@=D4jefbsm&E&nunhB`&te@e3R;3E3rzdWk+v ze@9WnrdR1dX;IDtdB+jV0)!?2i?#loo9}{?GO#4PV8W7a!SVexmh{tD4z#Wi>@3Zi z?^hAQ3O0gF?Z^61$n$ zeH3)|0Vl2pU>a454a=;;46KA1@XtX%{&@u35Nt=#gJ36u7Xhf|1W4LFkP5k~30kLR zem^!xjgap{@DhT92yh1>TbX3}SFm&dfNH%MiBE9+tB@9uKWMb6QDzI|rvg=!sc8h= z*#2JttU;Qqk>dZI=o*0NA|Ep|Oy2jgYsR%qX!$p1CuT<^?=IQ9OY!b{N*Wu>d1Pi> zVa6q9yn=XisRVn%G$9^jO&?Gnm_DFyQaMwO&QZf=)^v@@nlom&I{Jix=J1s3;r~e; z(A=kHBh`&TnHDl2Ql{Y2Skrylh>w-ZK)L;OP`+1hKLrBIs6veCYDtD>P1kK@VLJtQ zT2%texdYq0r*A!LHmm>z2MLB!L^l#L^mPlaD`VDT<6eyP>(bO^g4H3gy&{)m0w5&f zs_Nu9b|MNo^F%`R5cPOyGOS@Gk&3S6`UDpNahL$TF)@({*;U(QSmY--jMLCWb)uYE zR;}$gPMTx+Lofs1j{xPBfLDYVe)Z~A4aL#WoNCiiyd>Yc_r+7uNL)D3jnZx}cKsUw z-vhQcHs5M3T%5E0!~656q{cqEu}^92`+etE&q*iWlm;)zCof1NqteSU`Q?}burZc9 z`Y6}{5~DCRw{7n9mtpyZLk}VkEnnBl18>T~x0K*pxxxNxCe+uX~Lt5C2@ zn?p7t>;mxp_pqTQ+Uz)QTB4o;Iw`L|=hB3Yu0O#z$N~wy6d-|CCTn8pU(#qxXdw&i z7p$;S7A-5C911$h>eY|{O$u6+ZAs9aDm2G6`j|8g5v)~P#-3qDs`NJ9Y9q$l<5MsV z9AhU$dn#R?wgEL{RN9 z*5kR{wVvSLhVvuDp)qP&(?U=c33*g=oJ;Z>u*F6IV61>Qfp5p8EjpTtMLA+xsCJ$s zmW^g|G-JIT`Ex7^UO2EzQeaa=_z^H^N*W2`yy{3sgi&DRKwiU`_h2$I6pMrNgA@37 zpsZRyQO!V=Q*NDkLyo91{|Rk@QM+n#6W90L*psI}?925P{f+sK{LA?bGq2_PAJzJ9 zOn!9b!z*%an^Fr7fTF8*ddJMZ1y`HoYWuc6FtdBHe*HrI`obY)!w$KAr&7Oj-ZCF~ zVEL!vxhqnAzvS-!#@~4J_$S8;9fi}k!?J&;;@_F;FSfMbI=a}>z0lGd!9 zx%aTrd-&h#zHX4-I4zwSRL%@buaC$_MwKI@Qcp~78Bt7;8)rRo%g)^KZv*RQ z-p*%dPl44U2cA~~&*uh;^^N)cvp%J{ORDdZ++8abew~i@{6Fa!=!4K79HJo*`p{{7 z75dPzzo1ic4P@Rg=a$oC`D^i1<~nv78(Nc z=OE%o5fCpKqA?#sKnP$AQ^a#N4ylqXMm**d2ne>~DIn|DI6vC-;Oa}H^>fBmbs9=Ss!bCHuB2zO6Z1u{MzN zX!uo$2ReSC!BizDDx|rQ(*RG?F$PZxj)q3hX{GsB=KIiEpRuxMHLg|OZDpUorQU&6 zX6dXsYsp%(wyZtNWEoNS1OqH%G^`{;%Nq_q&|&$x494s%w-~D{rN_{GC86Lir7B~hr|Gu|eo^D79ylNUW(&w4R|L>o+{A~A zZFSzjI_^8~0KdG4Jd=cM4R~4cSz<-N{#EnV!dPf3l29n4Ua1J;Os$b-i9J!qnzh#I z`jJ(7*IXssaMdb(Pq9khDy!6aVgcO|D}oWnmTer@iKyNUdH~{So(a75n-o5!opXay zOWt=Y)%cI;8y-*vX-x?v>L3axa4gf^&+{nY1iZ9mq4tAwUpyv==s`X(wO$u$JeQ6~ zCd(#txe*Ffp&h_gN^07xeX~3BI}5(T_JaS`atV~Lls`Xl3%mfc@VItf1W>tZ{vs4S z)p8TqvmF2|x2<3M=D=fEh|JOj_O?qeB+Dhr^3*eLpj*E^HQcVbz`^z1E{vuo;%xg6 z*S=+n-r7Fp3Tk96yv2xKOjE zl%RD=YBV+6CPYxDP|Z`ZG*M3A^^oTylRSFpG>zPO23>spJfGtERxBlo3=$W-F2LJ2 z2XB=YbeL!gkM2NJt?>RG7t|Ut6^?=xJd)yL;1uCe8B)!|F_C{6M{7ech-KC&K1YP= zmQ82!ys6rVi40Eypa6ksEWi^5c^kkt0k9Rc$4Xb(n1+`Ew5A2s6jp7-4J5$fpsl@} zv5R0naH2qbJ8JFnuwm#E9!-g=2MSVrcoYy5pGi^gf-Jgm)SBhj%u=1J^tuU^0y05V z*Nj0^-ry}2QF^Mj3dhl-A^aUsgr>BKH{O$7&5Elz&)!NcxSo?-&lTNuxq)xJ!5I&f zjnD45-zR(bD&D>DcyajX!$V)cCLMlDX8LosX?wo0aCr9Y{Kf~q2M7PhM%DRe8uFWD zUx(uB$k|X+YtRU0ZUADrX|Dp%w@>lygIZ|3CEqL8tyk*S=j`ZBdNu#H?BA^TH|HGH zwZVp&(foUIuuBPcLd z4?x&vaHe_Dw_(Azp>Rf5L4!>*&#lX76U64;Fo zQ)MU39+}ysFqE|)^XF}!wao_S);&1=RaiP7&bP^p7nH^exte0I>1N`S1mMIO zIk-a!?tmkx7-p}`Y*v`f60`Xmt+a80X_T2}g=v;b-%2iu`WOB``40d_Eq)E%8^lPZ z7tY`c)-67^%2Q-~i7spSY?U5X)ed<)382^fwRv@`a8=o7IWbjp3@Exi1)Q_}bSwcs z`{_@|1bBW_Em#4*V{(?o{2?!4j5kw0>;Cqi~2CwSBdc~ty(9|GiK8bD<(8NVXcDjB~b z<+)1!id36q{EE~olJP52j;rLaxN*x>`kGU6b;wlbH)O3c)u>R764kh58K6yo1FOUD MNco@Es3lnPznsF={Qv*} diff --git a/experiments.ipynb b/experiments.ipynb index 4972ff3..9f62641 100644 --- a/experiments.ipynb +++ b/experiments.ipynb @@ -9,7 +9,7 @@ "name": "stderr", "output_type": "stream", "text": [ - "Matplotlib created a temporary cache directory at /var/folders/fh/fwc37qhn04d8sxp65hwv1kxm0000gn/T/matplotlib-yjwp3tvd because the default path (/Users/gabesmithline/.matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n" + "Matplotlib created a temporary cache directory at /var/folders/fh/fwc37qhn04d8sxp65hwv1kxm0000gn/T/matplotlib-rvtvf6xk because the default path (/Users/gabesmithline/.matplotlib) is not a writable directory; it is highly recommended to set the MPLCONFIGDIR environment variable to a writable directory, in particular to speed up the import of Matplotlib and to better support multiprocessing.\n" ] } ], @@ -24,7 +24,9 @@ "import nashpy as nash\n", "from dqn import DQNAgent\n", "import matplotlib.pyplot as plt\n", - "from ppo import PPOAgent\n" + "from ppo import PPOAgent\n", + "import seaborn as sns\n", + "from scipy.stats import ttest_ind, kstest, gamma, lognorm" ] }, { @@ -34,7 +36,9 @@ "outputs": [], "source": [ "# Seeds for reproducibility\n", - "seeds = [0, 1, 2, 3, 4]\n", + "seeds = [42, 135, 15, 23, 99, 43]\n", + "window_size = 70\n", + "window_size2 = 20\n", "\n", "# Initialize dictionaries to store results\n", "\n" @@ -49,7 +53,7 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -69,126694 +73,65785 @@ "Generation 1/300\n", "Solving for Nash Equilibrium in Generation 1/300\n", "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ + "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: inf - 109ms/epoch - 2ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2569.0779 - 95ms/epoch - 2ms/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: inf - 119ms/epoch - 2ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2532.4629 - 111ms/epoch - 2ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: inf - 127ms/epoch - 2ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2540.1357 - 128ms/epoch - 2ms/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2564.6882 - 638us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2508.8125 - 676us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2844.0115 - 630us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2536.7380 - 561us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: inf - 147ms/epoch - 2ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1918 - 139ms/epoch - 2ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1336.1031 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3431.6448 - 603us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2608.4219 - 700us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3396.9351 - 747us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.0466 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3627.0986 - 768us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3754.2871 - 600us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2595.5183 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1629.4144 - 646us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2608.5198 - 739us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2559.9006 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.7803 - 584us/epoch - 9us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3502.4580 - 697us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.6543 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2964.1685 - 697us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1173.5975 - 591us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1738.0775 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3610.9917 - 676us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3022.4600 - 670us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2541.1553 - 781us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3318.0762 - 677us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3284.0339 - 602us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2088.2500 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1998.0671 - 625us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3117.5410 - 621us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1895.8032 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3850.7961 - 677us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2490.9741 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3077.7517 - 639us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2638.3818 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2660.7781 - 635us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3072.9041 - 776us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2249.6199 - 667us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.5048 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2642.8306 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3797.6006 - 784us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2821.7910 - 694us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2932.0625 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2753.1692 - 675us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.8900 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2313.2498 - 580us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3263.5166 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2380.8372 - 648us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2787.0159 - 791us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3911.7996 - 585us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3351.1040 - 661us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2488.5012 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2537.0168 - 612us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2828.7993 - 697us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2529.4326 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3416.2368 - 630us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2783.1772 - 731us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3393.5042 - 650us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3616.9033 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1439.4982 - 710us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2509.0349 - 651us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3578.3057 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2446.6648 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2595.5789 - 710us/epoch - 11us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 2/300\n", "Solving for Nash Equilibrium in Generation 2/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.4628 - 757us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2470.6238 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2519.2363 - 786us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3019.5149 - 716us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2491.9749 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3688.4624 - 606us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2538.4207 - 720us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3044.7808 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2568.1826 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.2016 - 593us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2495.8093 - 651us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2520.9050 - 574us/epoch - 9us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3007.5017 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3624.3040 - 551us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2243.7122 - 624us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3727.5229 - 618us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3166.2935 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.7164 - 709us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2800.0806 - 662us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2505.8030 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2616.0405 - 679us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2515.1406 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2497.7563 - 620us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2754.2720 - 648us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2686.9392 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.5239 - 769us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2522.2812 - 750us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2785.7673 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3159.2405 - 674us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2113.0369 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3309.3545 - 542us/epoch - 9us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.6515 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3026.0376 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2538.0554 - 643us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3217.6975 - 618us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3681.9351 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3696.0085 - 681us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3857.9556 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2535.7878 - 618us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3033.9043 - 746us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2480.6975 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3616.8730 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2496.3127 - 664us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2532.3960 - 679us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2899.3359 - 684us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2498.7561 - 618us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1367.7980 - 610us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2476.5159 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.9336 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2521.2119 - 651us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2540.1570 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3574.3191 - 727us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2526.6824 - 606us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3114.0020 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2330.4331 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2918.3049 - 671us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3418.5684 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2760.8428 - 789us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2460.7632 - 777us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2399.3645 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3291.5227 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3546.9153 - 822us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3795.8440 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2433.6028 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3273.1816 - 827us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2493.9182 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1960.0703 - 742us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2464.1658 - 581us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2420.0422 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2600.7593 - 681us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2447.9277 - 657us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2495.1167 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2631.4233 - 677us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1866.7495 - 794us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1316.6486 - 669us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2569.7881 - 808us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 3/300\n", "Solving for Nash Equilibrium in Generation 3/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2898.4888 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2933.4998 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2559.3940 - 607us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3609.7009 - 617us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2404.7588 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2449.9458 - 609us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2556.6875 - 639us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2575.6841 - 554us/epoch - 9us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2568.5757 - 605us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3736.9658 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2609.1665 - 633us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2904.8271 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2247.6484 - 702us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2470.9092 - 737us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2618.2241 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2192.5872 - 806us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 12ms/epoch - 196us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2446.2849 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 184.8915 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3168.5833 - 728us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2759.7307 - 588us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2532.2783 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2735.8738 - 670us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3249.3606 - 722us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2898.7158 - 705us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3600.1641 - 665us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2400.0298 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1439.0319 - 570us/epoch - 9us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2000.9689 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.6455 - 589us/epoch - 9us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3119.1851 - 706us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2401.1265 - 579us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2436.4670 - 715us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3908.0676 - 521us/epoch - 8us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2446.7031 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1452.4510 - 700us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2331.4673 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2454.9631 - 664us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2456.1252 - 686us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1623.3060 - 630us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2373.3125 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2468.1604 - 622us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1996.4701 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2457.4502 - 613us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2448.9502 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2377.0437 - 726us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2459.4504 - 669us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2412.2612 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 800.5355 - 653us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2449.3040 - 624us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2444.9446 - 664us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2792.7327 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2479.1541 - 737us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3754.3948 - 621us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2447.1470 - 679us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2679.1011 - 747us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2214.3083 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2684.9487 - 576us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3220.1970 - 646us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3745.5269 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2411.6279 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2452.5239 - 600us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2077.0903 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2955.6790 - 655us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2438.0796 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2439.2095 - 714us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2394.9783 - 619us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2420.9360 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2414.0767 - 682us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 4/300\n", "Solving for Nash Equilibrium in Generation 4/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3717.3901 - 644us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.2871 - 634us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2469.5906 - 639us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2933.8621 - 613us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2280.3936 - 666us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2729.8774 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3614.9480 - 695us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2943.7051 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2310.6699 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2522.1785 - 2ms/epoch - 31us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2540.1462 - 661us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 172.7332 - 744us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2290.3093 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3024.3374 - 666us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2417.1177 - 757us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2326.9548 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2485.6130 - 671us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3833.9607 - 706us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2027.9860 - 600us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2400.9248 - 701us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2420.5967 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2322.7458 - 607us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1781.1594 - 652us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2419.1448 - 800us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1247.4805 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2489.3247 - 677us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2312.6584 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2568.2085 - 671us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2876.0190 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2506.5867 - 648us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2450.4873 - 631us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1625.9358 - 747us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3313.2727 - 693us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.8114 - 814us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3215.8503 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2282.4583 - 565us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2310.4380 - 816us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2482.1165 - 723us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2206.9504 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2385.0581 - 573us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2490.7324 - 694us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2313.8955 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2380.4595 - 705us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2839.9746 - 622us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2436.3137 - 603us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2343.1875 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2418.9070 - 693us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2386.5271 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2324.8823 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2436.2332 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1723.8506 - 631us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2349.0964 - 568us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1331.7965 - 717us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2484.0757 - 594us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2404.8818 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3337.8506 - 640us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2247.5376 - 612us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 5/300\n", "Solving for Nash Equilibrium in Generation 5/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2998.8015 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1622.8932 - 893us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2320.9812 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 158.2192 - 613us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3106.6814 - 535us/epoch - 9us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2317.9839 - 540us/epoch - 9us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2363.7021 - 575us/epoch - 9us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2356.4646 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2590.2593 - 565us/epoch - 9us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3465.1841 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3076.0125 - 730us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2419.9622 - 600us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2278.3730 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2241.9109 - 1ms/epoch - 17us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3486.2456 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 174.1858 - 712us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2344.8486 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2345.2798 - 637us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2323.5981 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2570.3115 - 623us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2178.8247 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2364.9844 - 702us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.0714 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3066.2415 - 615us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2875.4426 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1421.3352 - 713us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3661.0242 - 769us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2270.3406 - 559us/epoch - 9us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3552.9385 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2396.5945 - 658us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2500.4490 - 697us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2237.2698 - 544us/epoch - 9us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 190.1410 - 561us/epoch - 9us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2243.1426 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2713.7524 - 640us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 973.0861 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2393.7234 - 603us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3275.0283 - 539us/epoch - 9us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.8300 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1681.4667 - 571us/epoch - 9us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3138.4080 - 641us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2151.2029 - 686us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2163.7253 - 618us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2320.3838 - 603us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2394.1360 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.2371 - 683us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2260.0388 - 629us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2222.1562 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2254.7417 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3489.0798 - 701us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2432.0063 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2498.5076 - 699us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2216.6052 - 580us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2285.0913 - 740us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2786.8293 - 638us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2253.9194 - 590us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2729.1792 - 660us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1035.3180 - 695us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2303.6335 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2491.1426 - 721us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3358.8533 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2655.9998 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2295.7776 - 718us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 6/300\n", "Solving for Nash Equilibrium in Generation 6/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3317.1028 - 818us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2180.3472 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2921.7329 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2321.9780 - 667us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3664.0156 - 566us/epoch - 9us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2130.2427 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2819.5679 - 594us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3285.8760 - 647us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2335.6682 - 567us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1267.3511 - 638us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2935.6497 - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1818.1240 - 650us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2248.1433 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2232.7583 - 672us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2109.0596 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2589.4160 - 849us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2465.7341 - 783us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2950.1355 - 798us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2173.7446 - 830us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123.5444 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2006.1877 - 727us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2222.8889 - 1ms/epoch - 18us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2568.7900 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2349.6108 - 614us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3735.0720 - 575us/epoch - 9us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1216.6033 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1547.0938 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3264.2249 - 587us/epoch - 9us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2132.8528 - 712us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2445.8472 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2166.3708 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2338.6533 - 633us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2357.0415 - 721us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2888.6880 - 692us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1136.6003 - 713us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1990.4104 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1146.1591 - 607us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2223.6057 - 784us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3226.7715 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2303.8347 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1983.9683 - 629us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2120.9412 - 649us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1583.1246 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2387.6106 - 750us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2348.4426 - 583us/epoch - 9us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.4306 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1384.5736 - 907us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2106.4531 - 684us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2740.5940 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2032.5532 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2079.9836 - 669us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2215.8396 - 19ms/epoch - 313us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.4141 - 749us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2293.0481 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2765.0222 - 691us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2307.2710 - 793us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1389.2594 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1509.3656 - 730us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2081.4500 - 798us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2164.9260 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2426.0835 - 767us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1728.0859 - 598us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2803.1343 - 760us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2064.1799 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3606.3284 - 765us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2123.3457 - 711us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1826.9340 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2659.7021 - 702us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", "Generation 7/300\n", "Solving for Nash Equilibrium in Generation 7/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2016.4447 - 809us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2010.6832 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2215.2957 - 854us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3599.6797 - 857us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2034.8990 - 652us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2146.4048 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1347.7422 - 840us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1948.8575 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2231.9084 - 784us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1936.0421 - 614us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2190.7439 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2267.4497 - 721us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2202.4363 - 721us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2389.8831 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2427.9333 - 714us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 861.8051 - 774us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1965.6436 - 639us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2126.1074 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1875.5002 - 9ms/epoch - 141us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2038.7483 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1570.3910 - 652us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6768 - 782us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3103.7578 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1722.1884 - 805us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2113.1074 - 644us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1714.1952 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.0025 - 709us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2612.7244 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1814.5153 - 812us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2193.0156 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2091.2092 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1990.5004 - 1ms/epoch - 18us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2709.3247 - 845us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 200.2999 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1684.5149 - 632us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2104.7258 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1531.3281 - 713us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2271.8191 - 687us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2669.2922 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2095.9282 - 672us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1811.4756 - 710us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2074.9963 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2500.8958 - 886us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2067.0759 - 2ms/epoch - 28us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1256.5189 - 643us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2064.3088 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2053.7458 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1342.9572 - 607us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2177.4617 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2491.3547 - 617us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2257.1580 - 632us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2468.4065 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2846.4985 - 690us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2017.0245 - 816us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3086.3040 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2695.3298 - 608us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2748.8691 - 650us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2689.9172 - 624us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1668.4871 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.6132 - 683us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", "Generation 8/300\n", "Solving for Nash Equilibrium in Generation 8/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1877.2645 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1154.0856 - 911us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2200.8167 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2805.6345 - 748us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1641.5806 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2755.2847 - 950us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2853.1016 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 141.3896 - 764us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1615.3602 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2026.3420 - 810us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1596.4679 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1813.7853 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1327.0469 - 638us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2572.8569 - 677us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1714.1885 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2083.7815 - 705us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2080.5559 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2427.2380 - 633us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1882.6489 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1207.5758 - 799us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2269.7495 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2989.3105 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1350.0051 - 715us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1881.5299 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2926.5171 - 730us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2019.3571 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2961.0120 - 752us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1874.9927 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1847.6949 - 710us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2275.9595 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1986.4886 - 685us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1791.4541 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1810.2899 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1663.3748 - 687us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.1326 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1883.0367 - 713us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1783.9437 - 4ms/epoch - 71us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 807.6732 - 774us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1526.8342 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.7853 - 640us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.8311 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2548.3657 - 631us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1590.0807 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2006.9464 - 18ms/epoch - 292us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1955.1022 - 1ms/epoch - 22us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2397.6848 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1777.2877 - 996us/epoch - 16us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1888.6241 - 957us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1534.8053 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2522.9780 - 737us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2878.6418 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1261.1814 - 672us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2534.9045 - 751us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2460.6455 - 705us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1791.9156 - 832us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1516.1909 - 770us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1720.4570 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2052.0579 - 807us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1424.7843 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2708.7649 - 801us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 63.8494 - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1697.0962 - 659us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1507.4375 - 746us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1740.7936 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 930.8878 - 829us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1732.3575 - 672us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2005.8407 - 764us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1755.9482 - 671us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 9/300\n", "Solving for Nash Equilibrium in Generation 9/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1444.1110 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1663.7880 - 744us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1942.2819 - 773us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1735.1694 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1337.1058 - 872us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1651.9548 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1925.4703 - 647us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1844.7233 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1346.1940 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.5536 - 715us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1752.9167 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2431.1912 - 614us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1326.1410 - 728us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2194.0579 - 722us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2709.4722 - 756us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1466.7798 - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2029.2058 - 771us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1714.5985 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.4322 - 659us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1863.5374 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1332.1559 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2122.3406 - 612us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1682.1655 - 619us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1588.6060 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1528.4603 - 628us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1694.8994 - 993us/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1683.3605 - 732us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2763.7834 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1399.1837 - 800us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1461.8864 - 675us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1386.3567 - 635us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53.5462 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1682.0435 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1364.7755 - 744us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1626.9119 - 620us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50.0700 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1581.2374 - 633us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1022.2961 - 696us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1722.4321 - 715us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1670.1494 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1558.8938 - 638us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1660.5696 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1389.3364 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1101.5093 - 749us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1301.4297 - 873us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2293.2373 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3102 - 713us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1704.4875 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1556.1294 - 744us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1822.7128 - 700us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1569.1796 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1100.0308 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2370.8486 - 629us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1293.1383 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1995.9165 - 676us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1404.7155 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2282.9985 - 734us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1225.1578 - 684us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1381.0773 - 559us/epoch - 9us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1392.5398 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.2613 - 913us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1144.5631 - 3ms/epoch - 44us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 10/300\n", "Solving for Nash Equilibrium in Generation 10/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1488.1641 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1528.7758 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1338.9448 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1508.1129 - 710us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1760.7423 - 628us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1354.1676 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1038.9125 - 799us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2427.9053 - 890us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1670.0974 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1180.4232 - 771us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1328.1674 - 657us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1074.8451 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1612.6948 - 787us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1087.4449 - 795us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1506.6312 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1104.9314 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1398.2264 - 978us/epoch - 16us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1682.5522 - 606us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1168.7439 - 721us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1377.5088 - 838us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1867.9667 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1273.8805 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.5730 - 820us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1705.0156 - 643us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1208.7131 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1519.6208 - 644us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1528.4696 - 703us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1333.4364 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.8251 - 666us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1438.5372 - 713us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1847.1625 - 874us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1415.1923 - 735us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1051.5508 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98.7717 - 800us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1226.0920 - 728us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1204.6523 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 919.3829 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1816.7126 - 680us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1301.1599 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1187.1975 - 596us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1128.3881 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1468.2445 - 782us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1400.1134 - 5ms/epoch - 75us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2020.6725 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1155.7479 - 702us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1381.5498 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.0231 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 873.1114 - 672us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 989.5721 - 659us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1329.0399 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1338.3203 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1325.0127 - 620us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1360.5985 - 678us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.8474 - 721us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2116.0535 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1603.0239 - 617us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1081.7985 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1208.7010 - 808us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 945.1712 - 707us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1320.2925 - 691us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 830.2342 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.2838 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1204.0608 - 807us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1798.4790 - 724us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52.3240 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1004.1401 - 620us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 937.0167 - 810us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1547.5134 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 906.9800 - 681us/epoch - 11us/sample\n", - "Generation 11/300\n", - "Solving for Nash Equilibrium in Generation 11/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1318.5568 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1860.8071 - 722us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1233.1376 - 690us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1392.7327 - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1226.2312 - 975us/epoch - 16us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 953.5366 - 724us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1002.9103 - 604us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 957.4492 - 722us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 985.8682 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55.1374 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 992.5709 - 619us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1180.2684 - 663us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1190.3129 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 922.4528 - 739us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1153.5708 - 589us/epoch - 9us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 912.5236 - 625us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1158.0322 - 739us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1150.0515 - 692us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1139.2554 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1173.3680 - 637us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1141.0634 - 694us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1442.7692 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.9594 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.2197 - 728us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74.3508 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1105.6030 - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1388.8923 - 774us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1189.4850 - 882us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1705.7314 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1268.5037 - 729us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.7512 - 887us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 888.8225 - 817us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 893.4948 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.5130 - 2ms/epoch - 37us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 973.7333 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1105.0291 - 796us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1401.6102 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1178.5151 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.5281 - 697us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 903.5540 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1235.9263 - 704us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.8459 - 864us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1458.4257 - 694us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 905.0035 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1689.9581 - 754us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 932.1085 - 787us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1649.8711 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1209.4811 - 760us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 171.4737 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.6570 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 970.0877 - 711us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1142.9095 - 749us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1022.4727 - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1472.3466 - 851us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1373.7610 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1055.8000 - 2ms/epoch - 25us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 993.2573 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1267.1067 - 779us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.0685 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.9687 - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1418.4464 - 615us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.3925 - 2ms/epoch - 26us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.7791 - 780us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.3624 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.1479 - 1ms/epoch - 20us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 901.8265 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 895.2854 - 741us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.7885 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 948.9839 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 12/300\n", - "Solving for Nash Equilibrium in Generation 12/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428.3302 - 1ms/epoch - 18us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.7402 - 1ms/epoch - 20us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90.8661 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1032.3804 - 783us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 867.1731 - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 886.0140 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 864.0155 - 744us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 921.4827 - 796us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 938.4952 - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1335.7720 - 940us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.9528 - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1040.1392 - 758us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 883.7370 - 708us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1549.1786 - 714us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.1589 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.4707 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 951.7719 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.9697 - 876us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 815.9955 - 836us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1128.2080 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 947.7083 - 638us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.8585 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1225.3082 - 637us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 928.1237 - 855us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 835.1740 - 747us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.3677 - 701us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 831.5716 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 851.7261 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.6685 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.7067 - 776us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 991.6804 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 836.5216 - 787us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 967.6738 - 870us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1235.0885 - 701us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.6028 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.2877 - 688us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.5589 - 678us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.6566 - 669us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55.1775 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.5353 - 757us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 968.1429 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.0449 - 682us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1145.8915 - 646us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 948.8263 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.0732 - 731us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 799.0230 - 709us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106.4936 - 663us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1039.8558 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 847.2338 - 714us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 203.5296 - 595us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 945.7206 - 1ms/epoch - 16us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 976.4711 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 786.1976 - 656us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.6149 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1257.8988 - 769us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1253.6268 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.5565 - 703us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1093.7847 - 812us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.1720 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.4515 - 661us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1093.7809 - 721us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1136.2817 - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 794.4174 - 735us/epoch - 12us/sample\n", - "Generation 13/300\n", - "Solving for Nash Equilibrium in Generation 13/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 853.9046 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 174.4582 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.9196 - 792us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.7096 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.6184 - 888us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 846.6786 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.4061 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 811.6002 - 1ms/epoch - 20us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 954.0018 - 1ms/epoch - 17us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1364.7836 - 790us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1179.6210 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.6406 - 2ms/epoch - 26us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1101.6945 - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.3058 - 859us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2385 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1167.0771 - 678us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.5408 - 661us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 805.0297 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.0640 - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 950.7153 - 859us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 917.2291 - 727us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 955.2093 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1265.5870 - 843us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 971.7255 - 967us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1089.9340 - 818us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 169.2400 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.7758 - 807us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.8978 - 1ms/epoch - 23us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428.0284 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.7991 - 786us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 966.1448 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1759 - 725us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.6234 - 763us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1012.2357 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812.8607 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1096.2327 - 770us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 969.6362 - 869us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 843.8448 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.2449 - 756us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.8160 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 224.5738 - 618us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.6476 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 933.5377 - 779us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.3189 - 811us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.6441 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1100.3959 - 700us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.1986 - 835us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.8201 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123.9901 - 849us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.2048 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.0818 - 764us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 972.1793 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 892.5856 - 1ms/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.5199 - 713us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 954.8962 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.6143 - 923us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1127.6730 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.4322 - 784us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.7900 - 790us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.6039 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.8927 - 720us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 917.8094 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.4471 - 882us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.8177 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.8109 - 1ms/epoch - 20us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 891.9839 - 1ms/epoch - 20us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 848.8757 - 848us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Generation 14/300\n", - "Solving for Nash Equilibrium in Generation 14/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.3411 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.4752 - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.1339 - 812us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.3210 - 879us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.4917 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.2217 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.1846 - 737us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.7927 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 870.6910 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1845 - 720us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.4570 - 799us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.1933 - 770us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 912.4351 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 932.8345 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.0621 - 708us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.7205 - 1ms/epoch - 18us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.3939 - 2ms/epoch - 25us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.8242 - 709us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 185.9349 - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 854.2513 - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.3107 - 747us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.7693 - 633us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 947.5613 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1060.8981 - 693us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.2375 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 810.5726 - 1ms/epoch - 19us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 845.1282 - 644us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.8303 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.9175 - 2ms/epoch - 35us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.3726 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.0598 - 1ms/epoch - 17us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2051 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.0992 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 986.5731 - 839us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.9371 - 643us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 161us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.0179 - 1ms/epoch - 19us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1153.3967 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 872.1046 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.1436 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.3723 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.6965 - 850us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 953.3495 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.2240 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.4199 - 762us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812.4578 - 684us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3250 - 715us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.1843 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.0791 - 701us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6446 - 760us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.6976 - 1ms/epoch - 24us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.1050 - 755us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.0919 - 931us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.7863 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.3780 - 719us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3162 - 772us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.3642 - 795us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812.1896 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.4042 - 687us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.5727 - 961us/epoch - 16us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.3265 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 820.4691 - 778us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 226.0774 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.1102 - 720us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.1723 - 599us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.5289 - 750us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.5251 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.0891 - 834us/epoch - 13us/sample\n", - "Generation 15/300\n", - "Solving for Nash Equilibrium in Generation 15/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.9008 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.5798 - 761us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 857.3987 - 719us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 898.1478 - 800us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.1155 - 678us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7537 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.2739 - 2ms/epoch - 27us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.7801 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.0445 - 797us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.6730 - 674us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 800.8383 - 575us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.6452 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2529 - 707us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.7823 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 889.4247 - 672us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.7322 - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 853.3275 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 901.5270 - 810us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.5952 - 668us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.5703 - 912us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.8745 - 755us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.4819 - 1ms/epoch - 23us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.1667 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.2323 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 911.8853 - 1ms/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406.5918 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.4520 - 797us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 909.2622 - 1ms/epoch - 18us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.6273 - 3ms/epoch - 50us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 946.4532 - 2ms/epoch - 27us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 815.8097 - 1ms/epoch - 19us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 853.3657 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.6422 - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.1100 - 935us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.2420 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 838.3089 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.9288 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.5436 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.8225 - 727us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.5319 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.1857 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 995.9291 - 760us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.7185 - 688us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.9861 - 662us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 925.6586 - 669us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.4747 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.7272 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.0706 - 943us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.1219 - 700us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 848.6343 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1200 - 722us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 794.3254 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.7700 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.6497 - 633us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.3997 - 613us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.6119 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.0551 - 694us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.4708 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3420 - 655us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 923.1844 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.6030 - 616us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.6404 - 3ms/epoch - 54us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 877.9717 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.0537 - 709us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.2234 - 805us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 876.2302 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9004 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.0860 - 600us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.1210 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.6880 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 804.4704 - 716us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7249 - 1ms/epoch - 21us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 791.1951 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.8009 - 643us/epoch - 10us/sample\n", - "Generation 16/300\n", - "Solving for Nash Equilibrium in Generation 16/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 814.2706 - 605us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.5635 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.8528 - 822us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.9088 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 937.0598 - 678us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.7883 - 986us/epoch - 16us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 853.1812 - 648us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.6848 - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.6217 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.4656 - 661us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 928.2423 - 822us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7712 - 647us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.1421 - 787us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.3487 - 949us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.7564 - 792us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.1857 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.5110 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.0343 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1043.4404 - 733us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.2720 - 703us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 789.0522 - 808us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 810.0982 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.2458 - 667us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.1512 - 759us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.5697 - 759us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.8817 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.0806 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.4230 - 643us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 816.0189 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.2246 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.1937 - 613us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.8142 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.7120 - 2ms/epoch - 32us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.5208 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.3719 - 766us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.9064 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.7532 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.6676 - 783us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.2375 - 671us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.6639 - 673us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.3994 - 674us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.5339 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 819.1837 - 683us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.7051 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406.5911 - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.1678 - 685us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.7419 - 818us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.3657 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.9656 - 3ms/epoch - 42us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 933.1473 - 609us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.5738 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.1220 - 1ms/epoch - 20us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.1565 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.9066 - 790us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.4918 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.9516 - 643us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.4099 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.4265 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.7194 - 660us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.6857 - 813us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 833.9788 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.3291 - 831us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.9753 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.7504 - 2ms/epoch - 25us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 876.1246 - 994us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.6367 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.2751 - 607us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.6760 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.2287 - 1ms/epoch - 23us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.9285 - 699us/epoch - 11us/sample\n", - "Generation 17/300\n", - "Solving for Nash Equilibrium in Generation 17/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 891.6695 - 705us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.3749 - 621us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.9496 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.9087 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.6566 - 753us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.1110 - 1ms/epoch - 24us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.4335 - 791us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1043.2926 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.9628 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.1297 - 870us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 869.7555 - 857us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.1505 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.6598 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.9904 - 743us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7895 - 772us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.8676 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.6780 - 720us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 863.3710 - 628us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.4832 - 744us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.1591 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.8286 - 772us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.5457 - 789us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.1548 - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.1126 - 731us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.2132 - 717us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 790.2787 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.2305 - 686us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.6166 - 756us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.8098 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.7268 - 589us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.0862 - 2ms/epoch - 26us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 811.3502 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 823.9801 - 743us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.6853 - 669us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.0250 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.5223 - 620us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.2405 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.5421 - 686us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 933.9442 - 584us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 962.3926 - 742us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.7352 - 747us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6715 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.7740 - 734us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 847.4261 - 1ms/epoch - 23us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.6030 - 744us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.7930 - 1ms/epoch - 18us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.9839 - 811us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.5621 - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.0846 - 714us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.1384 - 1ms/epoch - 16us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.4783 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.8470 - 733us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2167 - 843us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.5695 - 698us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.7079 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.1876 - 820us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.0816 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.6551 - 695us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 833.0254 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.4555 - 653us/epoch - 11us/sample\n", - "Generation 18/300\n", - "Solving for Nash Equilibrium in Generation 18/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.2392 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.5006 - 839us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.9071 - 712us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.2542 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 981.8884 - 639us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.4191 - 625us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.5971 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.5138 - 694us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.7889 - 727us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.0965 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 905.7379 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.7254 - 717us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.8391 - 591us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.7047 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.0916 - 671us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.8695 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.5110 - 814us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.6777 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.6774 - 739us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.0491 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.5993 - 683us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.2590 - 691us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.8383 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.4968 - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 871.8938 - 941us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.1621 - 688us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6171 - 718us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6286 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.7130 - 630us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.0738 - 682us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.8935 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.4130 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.8080 - 632us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 836.3929 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.1871 - 644us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 833.2205 - 692us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8123 - 797us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 999.4918 - 826us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.3016 - 3ms/epoch - 48us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1681 - 667us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.3105 - 716us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.0224 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.7185 - 663us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.7006 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.2519 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.5412 - 697us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.8839 - 680us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.8807 - 687us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.3591 - 655us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.8650 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.2792 - 674us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.0633 - 697us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.0098 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.1376 - 567us/epoch - 9us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 823.1695 - 717us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.0961 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.5357 - 1ms/epoch - 17us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.7893 - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.7953 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.1794 - 784us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.0813 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.0488 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 824.5212 - 716us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.1879 - 872us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5359 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.6367 - 799us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.4841 - 2ms/epoch - 31us/sample\n", - "Generation 19/300\n", - "Solving for Nash Equilibrium in Generation 19/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.5304 - 734us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.4978 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.5125 - 665us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7254 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 977.6703 - 638us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6091 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.6678 - 588us/epoch - 9us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.0518 - 666us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.5087 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.7660 - 669us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.1932 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.1064 - 1ms/epoch - 16us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 799.1137 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.2906 - 717us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.9884 - 763us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1414 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.6538 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.8611 - 900us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 866.3765 - 796us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.3727 - 872us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.9512 - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.0286 - 642us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.5406 - 627us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.4628 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.9200 - 2ms/epoch - 25us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7250 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.9634 - 861us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.6052 - 764us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.8569 - 780us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.5679 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.8799 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.9474 - 804us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.0856 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.8833 - 749us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 826.8857 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.8868 - 634us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.6950 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.9958 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 926.0417 - 817us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.5710 - 590us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 796.3991 - 686us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 829.9436 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.7862 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4864 - 656us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.0727 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.9338 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.5759 - 701us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9882 - 694us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.7665 - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.5048 - 740us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 897.4641 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 829.2654 - 703us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 208.4666 - 580us/epoch - 9us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 834.6647 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.7889 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 903.8250 - 610us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.0028 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.4542 - 783us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 826.5320 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.1517 - 808us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.9827 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.6644 - 690us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.6276 - 614us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.9225 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.7628 - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.9252 - 619us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.8045 - 632us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.6452 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.6465 - 652us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.8560 - 731us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.0288 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.7119 - 749us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.3105 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 907.8189 - 3ms/epoch - 56us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.2692 - 1ms/epoch - 17us/sample\n", - "Generation 20/300\n", - "Solving for Nash Equilibrium in Generation 20/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.5185 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 791.4483 - 587us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7225 - 599us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.3753 - 741us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.0704 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.0262 - 761us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.6334 - 639us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 832.1275 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.8316 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9179 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.7083 - 767us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.9524 - 877us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.1434 - 779us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.6727 - 634us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3074 - 792us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.1525 - 906us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 846.1521 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.4386 - 660us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.0206 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.0499 - 803us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.5135 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.5369 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.5987 - 764us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.5759 - 750us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.5281 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.8844 - 688us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.0055 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.4661 - 891us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.1033 - 712us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.3586 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.2665 - 824us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.7131 - 829us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1100 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.9022 - 891us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9831 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.6528 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.5039 - 756us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.3333 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.7362 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.3125 - 666us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 842.6253 - 772us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.7145 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3351 - 768us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.9456 - 638us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.5513 - 700us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.1625 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.9174 - 690us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.7009 - 773us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.8776 - 835us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9548 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.4924 - 808us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.7318 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.6915 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.1011 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 978.6025 - 895us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.6055 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.0327 - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.2708 - 8ms/epoch - 122us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.2369 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.6486 - 908us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.1624 - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.8582 - 2ms/epoch - 30us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.4317 - 762us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.2665 - 715us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.5714 - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 824.6289 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.1541 - 10ms/epoch - 164us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7345 - 794us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.2513 - 2ms/epoch - 28us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.3904 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.2393 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 793.3152 - 754us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.8430 - 713us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 860.2107 - 708us/epoch - 11us/sample\n", - "Generation 21/300\n", - "Solving for Nash Equilibrium in Generation 21/300\n", + "Generation 11/300\n", + "Solving for Nash Equilibrium in Generation 11/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1092 - 5ms/epoch - 84us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.1334 - 659us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.3838 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.1839 - 648us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.0596 - 1ms/epoch - 17us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 793.4982 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.9316 - 831us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 870.8404 - 1ms/epoch - 18us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.3134 - 709us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.3406 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.3654 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.9260 - 711us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.4083 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.6705 - 649us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 842.8892 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 949.1506 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.3533 - 819us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.2745 - 619us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.3150 - 591us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.1741 - 647us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.3289 - 779us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.5706 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.7770 - 611us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.2021 - 565us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.7269 - 621us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.9390 - 696us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.1166 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.4936 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.3999 - 2ms/epoch - 32us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3768 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.6177 - 822us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.0379 - 874us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.7968 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.8398 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.5602 - 1ms/epoch - 23us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.7953 - 609us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.3516 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6329 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0911 - 558us/epoch - 9us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.7117 - 746us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.3438 - 674us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.6437 - 578us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.9707 - 3ms/epoch - 51us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.0026 - 607us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.7631 - 647us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7092 - 676us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8904 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7595 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.0757 - 722us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 109.3285 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7095 - 2ms/epoch - 29us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812.6287 - 695us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.2621 - 800us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 881.3873 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.4699 - 679us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 877.0595 - 708us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.2641 - 750us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4761 - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.0016 - 677us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.8562 - 632us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.6996 - 693us/epoch - 11us/sample\n", - "Generation 22/300\n", - "Solving for Nash Equilibrium in Generation 22/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.4206 - 713us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.2108 - 681us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.5258 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.3254 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9467 - 617us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 210.7064 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.8137 - 753us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2040 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.9061 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.6093 - 735us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.0677 - 586us/epoch - 9us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.0300 - 765us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.4175 - 898us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.2399 - 796us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 851.7114 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.6049 - 694us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.1586 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.3597 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.7812 - 647us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 917.8206 - 698us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.7109 - 571us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.0714 - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.1757 - 822us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.0055 - 626us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.9260 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.2601 - 585us/epoch - 9us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.7573 - 731us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.6856 - 609us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.5885 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.3913 - 615us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.1324 - 754us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 885.5309 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.3116 - 756us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 775.4496 - 571us/epoch - 9us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.6564 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.1151 - 616us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.3143 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.1227 - 733us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.7815 - 692us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 858.2267 - 895us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.2151 - 711us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 875.1071 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.7219 - 690us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.3782 - 718us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.1788 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.4290 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5516 - 1ms/epoch - 18us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.4341 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.9407 - 824us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.3745 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.4398 - 737us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.2140 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.6061 - 1ms/epoch - 19us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 775.4077 - 602us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 882.7852 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.8658 - 733us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5150 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.2585 - 767us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.1156 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.9599 - 629us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.6786 - 637us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.3313 - 926us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104.2243 - 701us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.2881 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.4747 - 666us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.5318 - 620us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 809.6251 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.8281 - 765us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 896.7700 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 797.4894 - 745us/epoch - 12us/sample\n", - "Generation 23/300\n", - "Solving for Nash Equilibrium in Generation 23/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 867.2643 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.7620 - 637us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.6164 - 670us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.2623 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.8062 - 2ms/epoch - 32us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.2162 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.2227 - 951us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.0619 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.2095 - 774us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.3547 - 701us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.3381 - 826us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.3896 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8591 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.2569 - 638us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1004 - 711us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 785.0443 - 596us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 29ms/epoch - 476us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.6084 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.7888 - 891us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.4835 - 624us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.4777 - 1ms/epoch - 22us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.3918 - 648us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.5884 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.1873 - 796us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.9564 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245.7663 - 2ms/epoch - 25us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1005.0622 - 675us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.6936 - 721us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.6223 - 743us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.8376 - 755us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.7544 - 611us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812.9517 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.3571 - 746us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.9694 - 641us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.4929 - 776us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 802.5786 - 671us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.3747 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.2462 - 691us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4333 - 651us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 836.8125 - 601us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4786 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.5515 - 787us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.0864 - 647us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1118 - 675us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101.5416 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7986 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.7516 - 629us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.1282 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.2521 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.2295 - 628us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.7464 - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.6748 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.6023 - 747us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.7600 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.5985 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.5583 - 882us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.5991 - 958us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 806.1213 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.6220 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.1204 - 669us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 823.5064 - 722us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.0729 - 748us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.4070 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.6953 - 658us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.7442 - 665us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.9554 - 3ms/epoch - 46us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.9884 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.4706 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.4248 - 842us/epoch - 14us/sample\n", - "Generation 24/300\n", - "Solving for Nash Equilibrium in Generation 24/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7176 - 662us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.2786 - 643us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.3065 - 781us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 73.1860 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.1816 - 5ms/epoch - 76us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.6006 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4268 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.1466 - 601us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.6431 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.3524 - 651us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.5752 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.4915 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.1776 - 660us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.8109 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.4536 - 836us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0291 - 662us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.6941 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.7571 - 645us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.2521 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.6086 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 910.4091 - 34ms/epoch - 546us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.3655 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 848.6627 - 587us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.2812 - 901us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.9052 - 637us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.9529 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.6101 - 748us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.1564 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.7192 - 739us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.4379 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.0193 - 1ms/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2557 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.5421 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.6667 - 630us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 830.0555 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.8319 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.8878 - 910us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.0025 - 811us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.3037 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 793.3210 - 641us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.6588 - 588us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.0217 - 611us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.3879 - 608us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.3316 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.5778 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.8109 - 761us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.9152 - 696us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.9431 - 760us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.5699 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.7749 - 564us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.4636 - 698us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.9139 - 803us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.4104 - 674us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.9654 - 666us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.7636 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.2626 - 678us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.4263 - 2ms/epoch - 37us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.3059 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.7510 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.0126 - 646us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.8626 - 678us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.6661 - 769us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.2449 - 1ms/epoch - 21us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.9279 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.1993 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4420 - 641us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.4495 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.0676 - 698us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.3554 - 721us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.7967 - 856us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9806 - 822us/epoch - 13us/sample\n", - "Generation 25/300\n", - "Solving for Nash Equilibrium in Generation 25/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 12/300\n", + "Solving for Nash Equilibrium in Generation 12/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.0325 - 764us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4520 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.8609 - 650us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6752 - 638us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.0364 - 735us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.3949 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2718 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 802.8340 - 592us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.1073 - 719us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.2310 - 676us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7468 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.6236 - 631us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.7468 - 1ms/epoch - 18us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.8588 - 586us/epoch - 9us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.9366 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.9756 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.9431 - 597us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.4224 - 669us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.2349 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.0255 - 738us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.1765 - 702us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.4164 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 789.0500 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3466 - 907us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.9220 - 700us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.8203 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.8649 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.1003 - 707us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.7207 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.7791 - 693us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.1757 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4152 - 598us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.6140 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.9648 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.6621 - 669us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.7784 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.8389 - 697us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.1771 - 726us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 827.7884 - 616us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.5509 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.4561 - 567us/epoch - 9us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.9476 - 642us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.6476 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.7816 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.9254 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 794.6068 - 607us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.2294 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.3426 - 592us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.8893 - 854us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 783.4317 - 794us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.3345 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.9798 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6835 - 3ms/epoch - 44us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.5544 - 709us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.5391 - 712us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.5913 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.9948 - 642us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.7568 - 626us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.6989 - 790us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.2975 - 657us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.1792 - 3ms/epoch - 52us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.0314 - 731us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.3546 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.1424 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.0835 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.0619 - 672us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.0758 - 2ms/epoch - 35us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 26/300\n", - "Solving for Nash Equilibrium in Generation 26/300\n", + "Generation 13/300\n", + "Solving for Nash Equilibrium in Generation 13/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.9490 - 621us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7125 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.9387 - 728us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.1771 - 783us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.1937 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.0211 - 836us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.1019 - 623us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.2077 - 814us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.6799 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.1293 - 2ms/epoch - 25us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.1085 - 661us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.1573 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.6436 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.4340 - 696us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.7747 - 632us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.4969 - 586us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.3079 - 632us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.3079 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.3074 - 618us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.2722 - 743us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 222.8002 - 717us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.8542 - 632us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.5952 - 649us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.4203 - 687us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.2957 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.6332 - 901us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.7708 - 766us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.2719 - 597us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.8604 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6293 - 696us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.1345 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74.8409 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4233 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.6376 - 741us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.8101 - 822us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.1991 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.3814 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.8420 - 629us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.3272 - 657us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6325 - 823us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.1320 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.5174 - 596us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.2164 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.7656 - 600us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.2720 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.6168 - 612us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.6741 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9586 - 692us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.4660 - 591us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.4823 - 608us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.4105 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.3907 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.7455 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 128.8477 - 664us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.3009 - 760us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.1923 - 702us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 27/300\n", - "Solving for Nash Equilibrium in Generation 27/300\n", + "Generation 14/300\n", + "Solving for Nash Equilibrium in Generation 14/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2747 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.1038 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.7316 - 664us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.9353 - 773us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.5250 - 601us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.1436 - 609us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.9279 - 645us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 796.4508 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.5245 - 688us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.9584 - 594us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.5406 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 432.6605 - 764us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9647 - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4601 - 638us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.0665 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.7908 - 697us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 818.9455 - 691us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.2684 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.1957 - 598us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2758 - 891us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.5913 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.5788 - 643us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.4999 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.0210 - 823us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.9453 - 740us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.8519 - 692us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3928 - 3ms/epoch - 56us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 830.7388 - 644us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.6996 - 668us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.1248 - 661us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.9598 - 765us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.8478 - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.3444 - 643us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.0331 - 803us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4849 - 759us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.0933 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.5593 - 723us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.5991 - 644us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.7242 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.9240 - 3ms/epoch - 51us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3219 - 632us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.8894 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.9046 - 593us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.4260 - 673us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.8929 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.0319 - 654us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.8754 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.1795 - 864us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.2479 - 660us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.0905 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.5397 - 683us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.9866 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.5729 - 1ms/epoch - 19us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.2053 - 904us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 783.9890 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5469 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6820 - 590us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8192 - 710us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.0422 - 599us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.9013 - 636us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.5278 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.7328 - 739us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.5574 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4766 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.8178 - 678us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.2084 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.2704 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.9202 - 761us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 28/300\n", - "Solving for Nash Equilibrium in Generation 28/300\n", + "Generation 15/300\n", + "Solving for Nash Equilibrium in Generation 15/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.5784 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 841.1703 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 785.3357 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.5641 - 639us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.4668 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.9216 - 798us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.6144 - 669us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1934 - 649us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 837.7757 - 704us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.4023 - 684us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.6306 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.5759 - 791us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.4504 - 674us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 432.0298 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.5783 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.2834 - 744us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.8420 - 759us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.6851 - 701us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2327 - 653us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.1274 - 770us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.8230 - 717us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.2559 - 639us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.9651 - 827us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 826.2289 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.0477 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.6351 - 593us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9980 - 773us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.0731 - 829us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.7658 - 700us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 775.5190 - 699us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.5369 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.9242 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.7582 - 3ms/epoch - 44us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.5862 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.5203 - 677us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.0028 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.2700 - 761us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 172.3462 - 672us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 815.6365 - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.7672 - 4ms/epoch - 62us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.7243 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.4386 - 643us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.3424 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.6059 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.3287 - 717us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6368 - 642us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.2824 - 664us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.1533 - 722us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.0078 - 756us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.5732 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.9353 - 627us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6433 - 643us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 789.3856 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.3983 - 724us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 76.7227 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.4493 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.6775 - 728us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.3255 - 752us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.4222 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.8552 - 657us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.5098 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.1962 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.9072 - 665us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.6478 - 685us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.3705 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.8827 - 604us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 29/300\n", - "Solving for Nash Equilibrium in Generation 29/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.4760 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.8646 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.6011 - 760us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.3799 - 874us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1589 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.3144 - 729us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 837.4536 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.0399 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.4754 - 587us/epoch - 9us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.5425 - 619us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.8282 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 148us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.2557 - 2ms/epoch - 37us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.3113 - 732us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.5134 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.4052 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.8832 - 844us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.1665 - 745us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2765 - 616us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.4831 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 809.0490 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.2617 - 677us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.7894 - 630us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.0590 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.9319 - 4ms/epoch - 59us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95.3412 - 1ms/epoch - 18us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4998 - 1ms/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.4072 - 728us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2231 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.8976 - 850us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.1277 - 827us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.2918 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.2855 - 685us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.1891 - 741us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.3314 - 717us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 806.1520 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.0939 - 836us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.3758 - 1ms/epoch - 17us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.6979 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.3572 - 754us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.2790 - 899us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3691 - 710us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6722 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.2509 - 812us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.6547 - 694us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.4828 - 680us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.7864 - 826us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6354 - 795us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.8269 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.5270 - 672us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.5492 - 887us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.2210 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.2430 - 795us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.2794 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.2430 - 709us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.9839 - 885us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3316 - 716us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.1807 - 664us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.8405 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.2318 - 900us/epoch - 15us/sample\n", - "Generation 30/300\n", - "Solving for Nash Equilibrium in Generation 30/300\n", + "Generation 16/300\n", + "Solving for Nash Equilibrium in Generation 16/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.9518 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.8579 - 944us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.4554 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.2797 - 644us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.7635 - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8669 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4693 - 1ms/epoch - 19us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 846.4285 - 768us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.5589 - 718us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.2272 - 3ms/epoch - 46us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.0126 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.4130 - 773us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 785.8382 - 721us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.0260 - 1ms/epoch - 17us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.7683 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.5891 - 871us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.8502 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.5187 - 703us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.6375 - 1ms/epoch - 17us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 156us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.7149 - 1ms/epoch - 18us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.0289 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2744 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6372 - 907us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2430 - 753us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1571 - 1ms/epoch - 18us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.8248 - 977us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.3035 - 698us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.2783 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.1398 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.9171 - 777us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9927 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 823.1089 - 768us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.7502 - 11ms/epoch - 177us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.7910 - 875us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.0375 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.2390 - 873us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.4899 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9413 - 784us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.7838 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46.0992 - 4ms/epoch - 62us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.6636 - 726us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5957 - 851us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.3867 - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.0663 - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.5768 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.7844 - 777us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.1418 - 884us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.2056 - 1ms/epoch - 17us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.1910 - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.4784 - 938us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.8001 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.3508 - 1ms/epoch - 16us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.7989 - 2ms/epoch - 39us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 62.6186 - 791us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.4434 - 1ms/epoch - 17us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.6528 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.8322 - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.7476 - 744us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.7485 - 1ms/epoch - 17us/sample\n", - "Generation 31/300\n", - "Solving for Nash Equilibrium in Generation 31/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.3472 - 900us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.9790 - 794us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.1399 - 987us/epoch - 16us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.5632 - 819us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1437 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.9509 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.2901 - 794us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1180 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.9981 - 1ms/epoch - 18us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3024 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6838 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.0279 - 742us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.1147 - 760us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 822.5970 - 1ms/epoch - 17us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.1841 - 816us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.0277 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.3572 - 730us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240.6720 - 1ms/epoch - 18us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.7654 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.0275 - 721us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.0795 - 856us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.1847 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.7747 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6331 - 890us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.3999 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7323 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.2054 - 849us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.8583 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.5480 - 890us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.0251 - 3ms/epoch - 43us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.3389 - 727us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.1065 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66.6993 - 781us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0021 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0625 - 704us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.5370 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.5835 - 3ms/epoch - 55us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.3463 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5103 - 823us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.9042 - 837us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3456 - 1ms/epoch - 18us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.0027 - 882us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.2540 - 1ms/epoch - 18us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.8965 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2580 - 822us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.1248 - 913us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.8102 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48.6456 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1173 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.6872 - 837us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.1021 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.4271 - 1ms/epoch - 20us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.3262 - 809us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.1184 - 857us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.1971 - 1ms/epoch - 16us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 119us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2928 - 5ms/epoch - 77us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.6586 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.6250 - 810us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.6620 - 965us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.1369 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1873 - 749us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 830.0240 - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.8028 - 688us/epoch - 11us/sample\n", - "Generation 32/300\n", - "Solving for Nash Equilibrium in Generation 32/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.6118 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9612 - 868us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 849.5948 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.5548 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.1648 - 816us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.4097 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2734 - 1ms/epoch - 21us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1967 - 833us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.4877 - 1ms/epoch - 16us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.5140 - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.6794 - 681us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.2759 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.9100 - 5ms/epoch - 82us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.5869 - 1ms/epoch - 19us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.0933 - 731us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.5595 - 867us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 803.1712 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.5107 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.4868 - 1ms/epoch - 18us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.3708 - 820us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.5699 - 1ms/epoch - 18us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.9738 - 1ms/epoch - 21us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9079 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.8693 - 901us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4816 - 964us/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4558 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.7355 - 1ms/epoch - 21us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7493 - 800us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5134 - 4ms/epoch - 57us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 791.4697 - 1ms/epoch - 18us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.4594 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.0174 - 813us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.9264 - 688us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.0626 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.3380 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4702 - 811us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.6116 - 746us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5266 - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.2922 - 825us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.4382 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.3179 - 749us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.3898 - 2ms/epoch - 25us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.1827 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.9412 - 767us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.1035 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.4489 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.7871 - 720us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.8201 - 828us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.7261 - 10ms/epoch - 161us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.7542 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.3752 - 1ms/epoch - 17us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 118.1699 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.6237 - 864us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.2294 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.8049 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.4311 - 1ms/epoch - 16us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.7265 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.7769 - 1ms/epoch - 18us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.4019 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.3934 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 823.4738 - 888us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.9998 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.9035 - 689us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.9012 - 731us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.6385 - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.3594 - 1ms/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.4783 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.2863 - 1ms/epoch - 17us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1944 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.3043 - 787us/epoch - 13us/sample\n", - "Generation 33/300\n", - "Solving for Nash Equilibrium in Generation 33/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.8710 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8874 - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7554 - 770us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.9572 - 891us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.2471 - 925us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9003 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.3004 - 1ms/epoch - 19us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.1243 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.0894 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5529 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.5659 - 734us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.9122 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.3307 - 1ms/epoch - 18us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.1525 - 810us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.1559 - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.8100 - 731us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.4223 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.9925 - 843us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.6730 - 966us/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.8804 - 776us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4400 - 798us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8264 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.9290 - 768us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.9258 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.0281 - 1ms/epoch - 18us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.7745 - 795us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.0206 - 904us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.7188 - 930us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.5848 - 1ms/epoch - 21us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.6391 - 757us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.1620 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.4153 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.3173 - 6ms/epoch - 93us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.1577 - 995us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.4657 - 947us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 821.9069 - 873us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4987 - 982us/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2725 - 768us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.8804 - 651us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1868 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.8300 - 4ms/epoch - 59us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6459 - 1ms/epoch - 19us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.0338 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.9798 - 635us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812.1151 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.3943 - 795us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.4291 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 14ms/epoch - 229us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.1698 - 1ms/epoch - 21us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.5779 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.5787 - 1ms/epoch - 23us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.5468 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.3346 - 973us/epoch - 16us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.0386 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.4157 - 892us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.8260 - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.0313 - 689us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.1779 - 1ms/epoch - 23us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.4323 - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.8510 - 956us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.3319 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.3715 - 622us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.3860 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.1560 - 724us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.0110 - 998us/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.0450 - 871us/epoch - 14us/sample\n", - "Generation 34/300\n", - "Solving for Nash Equilibrium in Generation 34/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.0923 - 869us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.6523 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.8642 - 2ms/epoch - 29us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.1134 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 799.9348 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4323 - 772us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.5114 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.2306 - 1ms/epoch - 18us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.3057 - 790us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.0334 - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6964 - 740us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.9204 - 852us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.3102 - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.9551 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0749 - 902us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.8814 - 861us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 137us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.2472 - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.0279 - 918us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.4599 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9152 - 732us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.7728 - 703us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 823.5938 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3064 - 820us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.6623 - 628us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.8422 - 1ms/epoch - 17us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1960 - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.1561 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.6812 - 3ms/epoch - 40us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.3954 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.6249 - 677us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.4636 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.0975 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.1224 - 737us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4141 - 3ms/epoch - 51us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.2297 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.4815 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.2200 - 798us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.8653 - 818us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.2389 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.3754 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.5542 - 691us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1121 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.1129 - 756us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.8642 - 1ms/epoch - 18us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.4836 - 7ms/epoch - 110us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.6826 - 915us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.5497 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.1055 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.2300 - 858us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 785.0622 - 813us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.4453 - 1ms/epoch - 16us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.0998 - 919us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.7331 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.2266 - 1ms/epoch - 18us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 838.2698 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6057 - 900us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.1551 - 983us/epoch - 16us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.5580 - 797us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.4512 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.7308 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.4724 - 749us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.5484 - 849us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.2434 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.5257 - 2ms/epoch - 37us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 821.6152 - 1ms/epoch - 16us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.9367 - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.7204 - 645us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.4630 - 826us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.7379 - 721us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 35/300\n", - "Solving for Nash Equilibrium in Generation 35/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.0272 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.9149 - 2ms/epoch - 30us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.0466 - 800us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.7285 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.6157 - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.4919 - 792us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5528 - 613us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.2687 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.4579 - 682us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.9315 - 783us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.8867 - 678us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.0909 - 749us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 778.6383 - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9833 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.1342 - 676us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.0138 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.6540 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.3987 - 717us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.8209 - 926us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4950 - 944us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 824.9781 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.1346 - 851us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.1362 - 752us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.4550 - 598us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9406 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.7731 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.7368 - 740us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5956 - 618us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.3294 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.1716 - 765us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.3172 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.5730 - 837us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.6787 - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5692 - 761us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.0477 - 636us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7894 - 971us/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.5295 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.7076 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2962 - 636us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 806.7463 - 760us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2520 - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6293 - 718us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.1974 - 741us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.4426 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.3609 - 668us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2737 - 838us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5472 - 683us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.1513 - 674us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.9164 - 716us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 103.8409 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.4869 - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.0777 - 993us/epoch - 16us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3244 - 919us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.2394 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.3575 - 718us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.3364 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.1631 - 814us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.4024 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.0793 - 664us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.3976 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.7424 - 915us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.7034 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.4349 - 938us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.9994 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.9017 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.9119 - 1ms/epoch - 18us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.8325 - 628us/epoch - 10us/sample\n", - "Generation 36/300\n", - "Solving for Nash Equilibrium in Generation 36/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.1505 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.5124 - 629us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.8528 - 622us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.2729 - 730us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.1447 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.4085 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.0334 - 748us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.9186 - 784us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.8337 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2842 - 859us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.2553 - 936us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.0126 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.7560 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.3995 - 742us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.8307 - 8ms/epoch - 133us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 789.1339 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.5359 - 737us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 802.6755 - 1ms/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.8428 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0477 - 1ms/epoch - 17us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.6482 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.5208 - 727us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.1874 - 968us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 805.2270 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 816.8431 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.5551 - 664us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.2597 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.1076 - 1ms/epoch - 17us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.6634 - 856us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.3568 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.5677 - 743us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.8294 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.3066 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.8793 - 664us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.5516 - 709us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.9647 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.2926 - 815us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3586 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.0297 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.8083 - 760us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.7910 - 623us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.8242 - 652us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.8723 - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.4333 - 727us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.7303 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8256 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.8320 - 714us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 814.8046 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.6252 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.7053 - 825us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.5344 - 949us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 213.6798 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.8424 - 917us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6750 - 900us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2739 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.7410 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.1535 - 878us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.3129 - 847us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.8744 - 805us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.5645 - 1ms/epoch - 18us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4857 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.2661 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.0521 - 974us/epoch - 16us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.4439 - 732us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.8855 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.0640 - 2ms/epoch - 30us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.8527 - 684us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.7380 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.1398 - 868us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.0881 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.8353 - 696us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.9892 - 736us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 819.5104 - 699us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.7440 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.1232 - 764us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.2711 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.4253 - 606us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.1232 - 823us/epoch - 13us/sample\n", - "Generation 37/300\n", - "Solving for Nash Equilibrium in Generation 37/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.4139 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2708 - 779us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9548 - 860us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.1929 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0803 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.5277 - 694us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.9999 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.3452 - 739us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.3963 - 2ms/epoch - 36us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.2261 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.3905 - 682us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.6423 - 684us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.3480 - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.1441 - 665us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.3347 - 690us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.2545 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0958 - 755us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 805.3423 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 797.3027 - 649us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.5104 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.8977 - 642us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.3806 - 867us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.7595 - 759us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.6281 - 753us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.0048 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.3560 - 848us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.1889 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.5931 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2968 - 692us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.3820 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5161 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.2214 - 741us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.1426 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.2263 - 1ms/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6385 - 3ms/epoch - 42us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.6130 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.9549 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.6060 - 759us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.5360 - 735us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4199 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.7746 - 2ms/epoch - 26us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9207 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.7779 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.4675 - 740us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9313 - 795us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 819.8244 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9331 - 843us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.6187 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.5598 - 678us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7375 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.1546 - 956us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.7190 - 812us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2319 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1503 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.1249 - 920us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.0193 - 636us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.2270 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.5057 - 678us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.4163 - 675us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.8536 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6880 - 687us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.8961 - 2ms/epoch - 29us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.5901 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.3147 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.5380 - 686us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.5616 - 928us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.1074 - 672us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2104 - 783us/epoch - 13us/sample\n", - "Generation 38/300\n", - "Solving for Nash Equilibrium in Generation 38/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 815.9902 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.2611 - 771us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.3605 - 760us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 800.8053 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.8397 - 706us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0631 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.3922 - 811us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.3941 - 621us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.9400 - 830us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.2183 - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.6188 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.5198 - 939us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0572 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.4656 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9377 - 736us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.5760 - 1ms/epoch - 20us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.9185 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.3432 - 902us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.2094 - 670us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.2105 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.8994 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.6803 - 913us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.7845 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.6327 - 1ms/epoch - 22us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.1836 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.3981 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.1105 - 954us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.5869 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.8580 - 820us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9747 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1302 - 711us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.2663 - 796us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.1367 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.2841 - 852us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7501 - 960us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.0499 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.5260 - 1ms/epoch - 24us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.4315 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.0402 - 732us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.2585 - 648us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.9758 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.3481 - 3ms/epoch - 51us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.4142 - 824us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.2108 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.1463 - 655us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.1303 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.3589 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.5580 - 766us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.1178 - 603us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.0777 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.3183 - 890us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.4346 - 665us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 793.1721 - 810us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.5967 - 591us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0650 - 686us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.8806 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.4974 - 695us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.2534 - 789us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9991 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9048 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.4693 - 703us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6191 - 823us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6696 - 2ms/epoch - 32us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.9523 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.1407 - 794us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.8406 - 885us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.1663 - 7ms/epoch - 109us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.4529 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.7634 - 939us/epoch - 15us/sample\n", - "Generation 39/300\n", - "Solving for Nash Equilibrium in Generation 39/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.9304 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.4349 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.8868 - 952us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.1198 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7640 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.8882 - 850us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.9949 - 640us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3611 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5425 - 779us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4232 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.2281 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.4679 - 726us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.4382 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2581 - 927us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.8674 - 863us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.0526 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.6456 - 900us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.4449 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.1614 - 737us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.3253 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.4872 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.4496 - 906us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 805.1606 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.4706 - 2ms/epoch - 33us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.7670 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.3898 - 692us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.5520 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.6171 - 716us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.1328 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2098 - 3ms/epoch - 51us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.3185 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1437 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.5008 - 801us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.7545 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.5604 - 884us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.6847 - 749us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3860 - 793us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6282 - 612us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.4598 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.6329 - 877us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.3733 - 869us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.0676 - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8790 - 698us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.1904 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.3596 - 650us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.2985 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.5275 - 588us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.3719 - 775us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.2532 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.4902 - 687us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.5413 - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.4212 - 883us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.1371 - 779us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.3584 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428.2027 - 713us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.3864 - 828us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 815.9323 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.5952 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.5222 - 2ms/epoch - 33us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 811.6292 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.1122 - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.5416 - 965us/epoch - 16us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.7159 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6801 - 737us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.7189 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.5753 - 889us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.8894 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.2509 - 827us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.3300 - 885us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.9401 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.1257 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.0262 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.4933 - 898us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.3489 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.1732 - 2ms/epoch - 39us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.4813 - 1ms/epoch - 22us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.2352 - 850us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.0219 - 787us/epoch - 13us/sample\n", - "Generation 40/300\n", - "Solving for Nash Equilibrium in Generation 40/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.9142 - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.6254 - 854us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.8302 - 3ms/epoch - 44us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.8445 - 738us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.0084 - 657us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89.0137 - 776us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.9386 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.2379 - 676us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2330 - 743us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9715 - 743us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.7849 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1147 - 2ms/epoch - 34us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.9515 - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.6161 - 2ms/epoch - 28us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.0672 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.6247 - 776us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4642 - 640us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7139 - 1ms/epoch - 17us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.0508 - 853us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.8736 - 852us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4296 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.1292 - 1ms/epoch - 18us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.5314 - 833us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9518 - 812us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.2555 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.6239 - 865us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.5860 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.8357 - 982us/epoch - 16us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4106 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.9434 - 831us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 807.0721 - 767us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787.4158 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.2607 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5086 - 746us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5904 - 771us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.1466 - 693us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5024 - 916us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.0238 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.7720 - 1ms/epoch - 23us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.1203 - 1ms/epoch - 23us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.6615 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.7784 - 616us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.5926 - 812us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.5303 - 776us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9866 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2395 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.3912 - 3ms/epoch - 49us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.1563 - 798us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0289 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.9410 - 1ms/epoch - 19us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.1666 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.4901 - 873us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.4772 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.7279 - 968us/epoch - 16us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7458 - 878us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7553 - 778us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.4811 - 849us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 820.1048 - 1ms/epoch - 21us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.1553 - 697us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.5462 - 898us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.3645 - 849us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.3396 - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.8030 - 968us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 813.6647 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.3881 - 3ms/epoch - 45us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.4549 - 847us/epoch - 14us/sample\n", - "Generation 41/300\n", - "Solving for Nash Equilibrium in Generation 41/300\n", + "Generation 17/300\n", + "Solving for Nash Equilibrium in Generation 17/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9006 - 751us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.2443 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4828 - 893us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.0214 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.0130 - 821us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.1617 - 918us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.0623 - 855us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.4944 - 742us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7463 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.3913 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.2534 - 746us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.6929 - 3ms/epoch - 44us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.4557 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3619 - 1ms/epoch - 17us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.7499 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.0515 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.0358 - 692us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.4536 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.4257 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.4362 - 939us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.0182 - 857us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.0811 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.0761 - 2ms/epoch - 32us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.8275 - 826us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.0907 - 785us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.0833 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.6169 - 653us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.6376 - 883us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.8739 - 924us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.3141 - 909us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.0581 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.8146 - 925us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 134us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.1746 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.4429 - 981us/epoch - 16us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.6656 - 939us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.7200 - 969us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.5476 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.0914 - 891us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.1504 - 885us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.2520 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6665 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6713 - 913us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.3280 - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9368 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1707 - 907us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.7283 - 1ms/epoch - 19us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3963 - 879us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.2535 - 821us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.0225 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.4462 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.3251 - 766us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4355 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.5134 - 1ms/epoch - 16us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9391 - 653us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3392 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.1918 - 717us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 803.9815 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.2137 - 645us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 829.1531 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.3561 - 1ms/epoch - 18us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.6495 - 711us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 42/300\n", - "Solving for Nash Equilibrium in Generation 42/300\n", + "Generation 18/300\n", + "Solving for Nash Equilibrium in Generation 18/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.3116 - 676us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.4398 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4946 - 755us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.6758 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7094 - 801us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.9525 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 778.7455 - 702us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.6886 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5789 - 619us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.9854 - 721us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.5354 - 874us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.4349 - 750us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.8558 - 1ms/epoch - 17us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3381 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.8232 - 752us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.4919 - 602us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.1371 - 641us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.5187 - 899us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5926 - 733us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.0957 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4803 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.1302 - 693us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.4413 - 2ms/epoch - 25us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.0165 - 982us/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.5139 - 1ms/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 804.0028 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.0601 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.0654 - 767us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.5762 - 820us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9120 - 802us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.5123 - 923us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.5078 - 729us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.1435 - 844us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 813.4850 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.7328 - 733us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.9801 - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.0367 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.4948 - 727us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.1926 - 1ms/epoch - 19us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7790 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.3304 - 695us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.8668 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.9298 - 697us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.0646 - 580us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.4454 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.2599 - 917us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.0840 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7355 - 875us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3961 - 926us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 822.9355 - 789us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5559 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.8694 - 628us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.9922 - 702us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6249 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7914 - 635us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4980 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0711 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.9807 - 658us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0076 - 679us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.5483 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.6046 - 4ms/epoch - 58us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.9802 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.9210 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5574 - 621us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.4461 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7347 - 781us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0735 - 774us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4777 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.1392 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.9724 - 596us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 43/300\n", - "Solving for Nash Equilibrium in Generation 43/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.5085 - 696us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.8729 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.8400 - 1ms/epoch - 24us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.9640 - 726us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.1554 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 794.9537 - 868us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.2570 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.8865 - 1ms/epoch - 17us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0330 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.1183 - 626us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.5048 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.2454 - 603us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.3768 - 750us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7448 - 602us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.3535 - 711us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.4788 - 777us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.4719 - 788us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.9646 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.4816 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.7709 - 813us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.8611 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.8447 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9548 - 848us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8056 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6268 - 4ms/epoch - 57us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9255 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.0115 - 764us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.0538 - 647us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.4790 - 1ms/epoch - 19us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3918 - 769us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812.5150 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9121 - 730us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.5217 - 722us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.7387 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.2982 - 744us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.4957 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.6723 - 1ms/epoch - 20us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8440 - 1ms/epoch - 19us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.3040 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.8344 - 668us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.7374 - 672us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.2749 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.4135 - 707us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.3159 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.8573 - 669us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.3238 - 685us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.9446 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 805.4814 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.4819 - 692us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1229 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.7361 - 671us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8321 - 583us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.8184 - 640us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.5918 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.4203 - 660us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.0915 - 756us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0167 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0436 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.8499 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5086 - 662us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.3614 - 1ms/epoch - 18us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.6815 - 658us/epoch - 11us/sample\n", - "Generation 44/300\n", - "Solving for Nash Equilibrium in Generation 44/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2469 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8039 - 643us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.1589 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 81.1768 - 573us/epoch - 9us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.1380 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.9791 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.4743 - 634us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.8242 - 527us/epoch - 9us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.8985 - 640us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.0492 - 846us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.7388 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.8110 - 709us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2083 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.6717 - 641us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.0728 - 634us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 817.6127 - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.6241 - 806us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.8184 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.4698 - 685us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.8112 - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5275 - 606us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9891 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.4420 - 564us/epoch - 9us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2150 - 700us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.7543 - 1ms/epoch - 17us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.8768 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.8839 - 745us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2372 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.5816 - 665us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.4395 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.0312 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.8279 - 703us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.0743 - 635us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0430 - 2ms/epoch - 31us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.3895 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8200 - 762us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7565 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.2241 - 702us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.1103 - 862us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.8403 - 667us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5788 - 1ms/epoch - 17us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.7895 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.6375 - 851us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.0435 - 676us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54.7539 - 2ms/epoch - 37us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.2667 - 702us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 152.5252 - 721us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.2927 - 694us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0446 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.7018 - 845us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.6146 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.3552 - 705us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.7661 - 727us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.2933 - 576us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.2697 - 547us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.0638 - 722us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.8303 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.8839 - 786us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.0600 - 758us/epoch - 12us/sample\n", - "Generation 45/300\n", - "Solving for Nash Equilibrium in Generation 45/300\n", + "Generation 19/300\n", + "Solving for Nash Equilibrium in Generation 19/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.1562 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.7775 - 800us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7396 - 699us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.2020 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 820.6994 - 658us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9092 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.9162 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.9825 - 676us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9481 - 748us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.4504 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.5330 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.5511 - 637us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.5990 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 30ms/epoch - 483us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6344 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.0532 - 762us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.2016 - 712us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.8466 - 744us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.2087 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.8137 - 757us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.1196 - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1779 - 1ms/epoch - 22us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.8714 - 1ms/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8284 - 1ms/epoch - 17us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.0009 - 923us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.5702 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.3589 - 687us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.3461 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.6995 - 732us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.0312 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9309 - 900us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.0703 - 655us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 154.0551 - 783us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0720 - 765us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.2655 - 853us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.5202 - 681us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4480 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.8226 - 1ms/epoch - 17us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8890 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.2483 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.2717 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.4368 - 671us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2570 - 702us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7823 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.5445 - 784us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.2532 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.3171 - 700us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.0250 - 769us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.4216 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5377 - 768us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.3934 - 822us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.7764 - 645us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.9986 - 739us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.3192 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.7599 - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.6053 - 683us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.0228 - 827us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6447 - 829us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2941 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361.9182 - 796us/epoch - 13us/sample\n", - "Generation 46/300\n", - "Solving for Nash Equilibrium in Generation 46/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.4490 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.9223 - 3ms/epoch - 50us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 807.7184 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.0984 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.0580 - 706us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.7996 - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.8565 - 609us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.1974 - 690us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787.1188 - 667us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5684 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2360 - 2ms/epoch - 30us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.6502 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.8550 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7269 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.3528 - 916us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.5707 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.2926 - 676us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.3995 - 680us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.1447 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.7486 - 1ms/epoch - 17us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.5595 - 776us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.3887 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.4518 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.0308 - 872us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.3876 - 657us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.9285 - 725us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.4310 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1124 - 664us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.9488 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.8973 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.2758 - 601us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.2274 - 616us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6077 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.4594 - 624us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.7365 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.0132 - 844us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 805.1492 - 617us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.6450 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.1978 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 786.1868 - 660us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.9506 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6909 - 743us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.3774 - 861us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.9367 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7682 - 732us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2847 - 974us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3774 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.0822 - 951us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.1057 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.3110 - 786us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7707 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.2303 - 743us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2329 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9510 - 839us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.1551 - 782us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.2661 - 653us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4388 - 678us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.1255 - 924us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.2373 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.8599 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.6096 - 936us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.5482 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.1542 - 675us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.1572 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.9763 - 818us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 814.8184 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.1969 - 667us/epoch - 11us/sample\n", - "Generation 47/300\n", - "Solving for Nash Equilibrium in Generation 47/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.5905 - 943us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.9753 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8120 - 639us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.1210 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.5966 - 776us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.7370 - 839us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7928 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5237 - 669us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1444 - 639us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.0742 - 599us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.5466 - 2ms/epoch - 27us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.7741 - 850us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.4335 - 644us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8912 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.0966 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.7458 - 1ms/epoch - 19us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.6396 - 639us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.4781 - 833us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.1500 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 796.6785 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.0604 - 686us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.7697 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.6737 - 738us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5831 - 755us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.5988 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.2690 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6281 - 767us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.6406 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.4214 - 740us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.7021 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.4864 - 687us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.1752 - 2ms/epoch - 26us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5038 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0495 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.4484 - 712us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.4390 - 803us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.9413 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.9765 - 723us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.5164 - 860us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7958 - 745us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.3737 - 646us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.1951 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.7702 - 683us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.0950 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7297 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.9104 - 684us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.1127 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2723 - 807us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.6652 - 869us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.9120 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.0179 - 765us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.3758 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9409 - 746us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.8665 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2244 - 971us/epoch - 16us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.3672 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.3159 - 757us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 151.8131 - 948us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.1194 - 723us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1503 - 680us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.7982 - 702us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.1398 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.3770 - 614us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 91us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.9750 - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.9471 - 810us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48.4399 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.0441 - 669us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 48/300\n", - "Solving for Nash Equilibrium in Generation 48/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.7611 - 3ms/epoch - 52us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.4036 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.4910 - 2ms/epoch - 32us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.5862 - 832us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.7497 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 811.6418 - 735us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.3797 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.8339 - 828us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.9727 - 654us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.2390 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.0939 - 664us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.3591 - 781us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.0953 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.9490 - 878us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.5684 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.1002 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.8932 - 724us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8831 - 654us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.5854 - 755us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 778.9514 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.5710 - 2ms/epoch - 38us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6815 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.7728 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0092 - 714us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9210 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.7471 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7421 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.5047 - 692us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.8282 - 3ms/epoch - 49us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.7316 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9448 - 533us/epoch - 9us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.8123 - 652us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.1284 - 691us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1764 - 713us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8972 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2318 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.8979 - 667us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.5294 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.2098 - 727us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.4802 - 675us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4468 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 811.3873 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.2817 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.1641 - 598us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.7502 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.7644 - 749us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2688 - 670us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7315 - 670us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2720 - 694us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.1546 - 641us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.7982 - 789us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.8757 - 743us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.4034 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5123 - 744us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50.8987 - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.1901 - 649us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5363 - 733us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 791.1877 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.3909 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.8394 - 729us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.0320 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.3116 - 709us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.7191 - 750us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5365 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.3251 - 2ms/epoch - 28us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.9846 - 647us/epoch - 10us/sample\n", - "Generation 49/300\n", - "Solving for Nash Equilibrium in Generation 49/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.0556 - 770us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5811 - 740us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7562 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.9304 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0843 - 709us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.3207 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.4667 - 675us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1812 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.4328 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.5106 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.2899 - 646us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.7670 - 646us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4379 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.2330 - 847us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.5657 - 731us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.8477 - 672us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3366 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.2890 - 677us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5299 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4635 - 726us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.3752 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.8580 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.9468 - 657us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9031 - 649us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.3375 - 712us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.5812 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8351 - 645us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.7430 - 888us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.8945 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.5997 - 2ms/epoch - 28us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.9092 - 792us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.7213 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.9875 - 748us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2687 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.5674 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406.7148 - 581us/epoch - 9us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.0726 - 778us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5372 - 923us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.3135 - 725us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.2079 - 739us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.5861 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.5458 - 768us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.8453 - 751us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.5178 - 2ms/epoch - 37us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.2649 - 844us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.7042 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.3187 - 802us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.2011 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.5770 - 608us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.6389 - 693us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.5937 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.0391 - 682us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2198 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.7300 - 952us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.4293 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.2849 - 730us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.5990 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.7192 - 758us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.7735 - 592us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.6432 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.5743 - 677us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.1954 - 604us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4691 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.0586 - 769us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.8067 - 643us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.5569 - 4ms/epoch - 70us/sample\n", - "Generation 50/300\n", - "Solving for Nash Equilibrium in Generation 50/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.4771 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.1574 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.6902 - 645us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8481 - 743us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.7487 - 665us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8406 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.5982 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6149 - 731us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.5103 - 742us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.1195 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.7175 - 550us/epoch - 9us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.0057 - 652us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.5714 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.7596 - 774us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.6869 - 673us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.3558 - 693us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6931 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.8921 - 594us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.5679 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.8112 - 1ms/epoch - 19us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.0806 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4686 - 586us/epoch - 9us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2585 - 1ms/epoch - 20us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.3574 - 783us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2183 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.0652 - 662us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.1949 - 677us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.9554 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.8453 - 747us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.3403 - 731us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 802.6401 - 607us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.2496 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.7108 - 605us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.8015 - 714us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.2936 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.8375 - 1ms/epoch - 18us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.1169 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.2001 - 584us/epoch - 9us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.2680 - 594us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.2622 - 2ms/epoch - 40us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.6928 - 655us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 786.7005 - 761us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.7311 - 697us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8659 - 734us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4182 - 693us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.9516 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.4926 - 774us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4908 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.1245 - 625us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.1649 - 677us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.1348 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.6815 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.7420 - 747us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.0444 - 639us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.7347 - 3ms/epoch - 51us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.3544 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.7786 - 667us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.2330 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.2555 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5450 - 1ms/epoch - 19us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 785.7797 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.2702 - 2ms/epoch - 32us/sample\n", - "Generation 51/300\n", - "Solving for Nash Equilibrium in Generation 51/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.2652 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.0501 - 2ms/epoch - 32us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7206 - 726us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.8244 - 726us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.9998 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.6321 - 850us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.8647 - 665us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.0804 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.1649 - 694us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.0398 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.2292 - 748us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.2179 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.1612 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.5256 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.3625 - 824us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.1439 - 908us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 81.9174 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.6880 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.7359 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.2245 - 733us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.8032 - 729us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.2007 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.7767 - 702us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6876 - 727us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.3220 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.3261 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.3112 - 668us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787.3357 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.8377 - 962us/epoch - 16us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7650 - 812us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2107 - 672us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.2991 - 830us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.1106 - 740us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9749 - 700us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.9238 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.4260 - 6ms/epoch - 101us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.9252 - 829us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.6855 - 873us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.0290 - 816us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 23ms/epoch - 364us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.5208 - 3ms/epoch - 44us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.8180 - 729us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.1285 - 666us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.6555 - 908us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.7765 - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.3137 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.0692 - 822us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5648 - 777us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.5640 - 820us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.0248 - 782us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40.2993 - 8ms/epoch - 125us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.9231 - 799us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4785 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.6859 - 672us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3188 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0189 - 791us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.8787 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.7057 - 762us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1760 - 1ms/epoch - 21us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.1298 - 710us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7702 - 811us/epoch - 13us/sample\n", - "Generation 52/300\n", - "Solving for Nash Equilibrium in Generation 52/300\n", + "Generation 20/300\n", + "Solving for Nash Equilibrium in Generation 20/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.3285 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.9171 - 691us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.2007 - 707us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1048 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.2732 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.1927 - 2ms/epoch - 29us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.2610 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.2559 - 767us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.6089 - 895us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8161 - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.8073 - 590us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7294 - 840us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8428 - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4565 - 820us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4443 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5043 - 692us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9788 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.2494 - 687us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 147.8437 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.1002 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3615 - 709us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2986 - 706us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 432.0403 - 790us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.7193 - 1ms/epoch - 17us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.4045 - 1ms/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.4769 - 2ms/epoch - 30us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4208 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9235 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.8815 - 843us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.2465 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.7843 - 1ms/epoch - 21us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.5665 - 702us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.1195 - 661us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.5856 - 845us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8561 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6088 - 903us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.2636 - 838us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.4052 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0538 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2440 - 916us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.6063 - 813us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.8329 - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6578 - 671us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.2491 - 2ms/epoch - 24us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.6273 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.1324 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5185 - 809us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.4676 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.6337 - 721us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.2900 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.4853 - 5ms/epoch - 80us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5188 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.9283 - 960us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.4473 - 850us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0501 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.3620 - 1ms/epoch - 16us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.3225 - 735us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.2644 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3298 - 50ms/epoch - 813us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.8497 - 824us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.8089 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.6673 - 702us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 808.9374 - 2ms/epoch - 35us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.9753 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.5424 - 637us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1478 - 727us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.7311 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.0797 - 722us/epoch - 12us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 53/300\n", - "Solving for Nash Equilibrium in Generation 53/300\n", + "Generation 21/300\n", + "Solving for Nash Equilibrium in Generation 21/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.0011 - 893us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 775.8773 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.3488 - 697us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.1674 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.7246 - 769us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5842 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.4275 - 709us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7036 - 599us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 790.8014 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.1742 - 676us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812.0405 - 670us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.1815 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.7264 - 950us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.3176 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3366 - 594us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7218 - 708us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.6219 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.3856 - 733us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.9759 - 2ms/epoch - 28us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.1049 - 777us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52.5832 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9960 - 716us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 815.9133 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.0942 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1148 - 672us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.5151 - 1ms/epoch - 17us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.8867 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.7939 - 680us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.2514 - 627us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4186 - 651us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7759 - 785us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.8181 - 695us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.2645 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.3172 - 628us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.2368 - 940us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.0919 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5098 - 852us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6597 - 639us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773.9127 - 657us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.2122 - 673us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.3373 - 735us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.1980 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.9496 - 800us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1133 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0094 - 701us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.7923 - 3ms/epoch - 56us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812.3064 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.5527 - 738us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 807.2683 - 664us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.3781 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.3535 - 651us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.5545 - 2ms/epoch - 27us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.3087 - 603us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.7098 - 728us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.2834 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.0709 - 977us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8356 - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.6121 - 735us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.6549 - 872us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 162.8274 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.3781 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.8158 - 652us/epoch - 11us/sample\n", - "Generation 54/300\n", - "Solving for Nash Equilibrium in Generation 54/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7081 - 706us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3193 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.8649 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5463 - 4ms/epoch - 61us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.0610 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7102 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.4943 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.8373 - 715us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4743 - 629us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8788 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7385 - 743us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.1146 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.3749 - 745us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.5325 - 701us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.6620 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.8470 - 946us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.7990 - 718us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.9041 - 661us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7728 - 623us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.9075 - 648us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7053 - 794us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.1484 - 1ms/epoch - 23us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.2252 - 636us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.5968 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.7309 - 694us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.1791 - 703us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9531 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.5959 - 669us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.2767 - 694us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.1878 - 697us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0237 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.5280 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.8666 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0901 - 689us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.3477 - 753us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.3464 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.6664 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.9952 - 668us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.5477 - 786us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5111 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.2186 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4864 - 613us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.9398 - 860us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2175 - 852us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.6775 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.1484 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.0643 - 828us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 807.3423 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.4465 - 650us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.2511 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0200 - 844us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.0669 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6559 - 597us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1165 - 846us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.1266 - 654us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.8590 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.3698 - 729us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.5058 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5048 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.3199 - 833us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5620 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.0184 - 656us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.1279 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.5897 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.5006 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.1630 - 626us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2475 - 784us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.0652 - 700us/epoch - 11us/sample\n", - "Generation 55/300\n", - "Solving for Nash Equilibrium in Generation 55/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.9857 - 731us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.5321 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.2422 - 2ms/epoch - 35us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47.0590 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.9553 - 575us/epoch - 9us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.0480 - 699us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.8031 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.1413 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.1881 - 752us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.7185 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.3525 - 846us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.1225 - 628us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.5033 - 704us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.1641 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.5275 - 2ms/epoch - 25us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.6245 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.8701 - 870us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.8429 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.4005 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.8170 - 716us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.9678 - 604us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.2442 - 1ms/epoch - 18us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.7177 - 632us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6121 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.6419 - 1ms/epoch - 17us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.0657 - 711us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.2694 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.1321 - 2ms/epoch - 25us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.9422 - 631us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.0065 - 700us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.6989 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.4557 - 5ms/epoch - 75us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.7766 - 905us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95.6406 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37.7196 - 761us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 804.6666 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.2902 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.2085 - 674us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.3518 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.4251 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2062 - 594us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.2475 - 904us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.4081 - 1ms/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.9841 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7559 - 723us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.6521 - 990us/epoch - 16us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.6872 - 730us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.1783 - 711us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 215.5303 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.6263 - 807us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4326 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.4046 - 2ms/epoch - 32us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 228.7124 - 784us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.3254 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.2877 - 755us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.1215 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.5042 - 739us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 15ms/epoch - 238us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9184 - 2ms/epoch - 26us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.4021 - 815us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3763 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.8245 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3645 - 1ms/epoch - 20us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.5009 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.4551 - 1ms/epoch - 16us/sample\n", - "Generation 56/300\n", - "Solving for Nash Equilibrium in Generation 56/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.4112 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.1407 - 739us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.9786 - 777us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.2698 - 840us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.6193 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8042 - 2ms/epoch - 33us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3707 - 954us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.8278 - 665us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.5063 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7506 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.2720 - 800us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2045 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9294 - 956us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.2542 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.1821 - 1ms/epoch - 22us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.5451 - 3ms/epoch - 42us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.4526 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.0074 - 1ms/epoch - 18us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8049 - 723us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5992 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.4690 - 877us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8273 - 833us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 808.3710 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.2133 - 607us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.9940 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.2065 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.5951 - 603us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9227 - 650us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0980 - 794us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.5413 - 759us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4358 - 704us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.8907 - 817us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.5239 - 1ms/epoch - 22us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.4864 - 770us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.4349 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.8403 - 616us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79.0136 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7844 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3017 - 578us/epoch - 9us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 786.8329 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.7816 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.4106 - 786us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.6580 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.4130 - 707us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1615 - 824us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0122 - 640us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.7953 - 2ms/epoch - 33us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.2398 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6555 - 629us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 810.7908 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.3917 - 713us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7491 - 721us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.1510 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6116 - 616us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7418 - 648us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.3174 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.3098 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6229 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.2104 - 866us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3970 - 715us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9692 - 617us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2784 - 798us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8396 - 638us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.2607 - 618us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.7507 - 750us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.6857 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.9617 - 659us/epoch - 11us/sample\n", - "Generation 57/300\n", - "Solving for Nash Equilibrium in Generation 57/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 22/300\n", + "Solving for Nash Equilibrium in Generation 22/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9100 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.3098 - 2ms/epoch - 36us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.9852 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.3435 - 617us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3432 - 1ms/epoch - 16us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2778 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.2134 - 779us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.9424 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.3688 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.1690 - 720us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.6070 - 746us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.9255 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5609 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9400 - 876us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.8877 - 661us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8033 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.6007 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4615 - 678us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.6500 - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6296 - 1ms/epoch - 23us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.2194 - 876us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.5216 - 725us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.3659 - 896us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.5667 - 860us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 807.3879 - 4ms/epoch - 59us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.7298 - 915us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1703 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.6115 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.6635 - 708us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6868 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.5873 - 804us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.0616 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9073 - 840us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.9446 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.9332 - 996us/epoch - 16us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4061 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.1226 - 721us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.7963 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0031 - 823us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6935 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 807.0300 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.5897 - 827us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.2226 - 728us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.6176 - 865us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5704 - 602us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.8007 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.9477 - 686us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.2819 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.8423 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.6022 - 741us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.1427 - 712us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.8832 - 707us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.6189 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.2020 - 743us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.1687 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.6384 - 783us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.5304 - 691us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.6473 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.8103 - 631us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.2726 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5933 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.8349 - 824us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 793.9821 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.0389 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.9293 - 926us/epoch - 15us/sample\n", - "Generation 58/300\n", - "Solving for Nash Equilibrium in Generation 58/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.1500 - 613us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.7938 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.1365 - 686us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0504 - 690us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6926 - 785us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.1070 - 850us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.4188 - 770us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.0367 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.2542 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 789.0785 - 909us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.5716 - 710us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.5470 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.2444 - 755us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.6566 - 680us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6381 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6342 - 752us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.2625 - 3ms/epoch - 41us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.3897 - 2ms/epoch - 38us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.3531 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4878 - 980us/epoch - 16us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 182.9919 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.1251 - 672us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5533 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.2834 - 693us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.9518 - 702us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.6893 - 627us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.7686 - 2ms/epoch - 38us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.9080 - 747us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.0689 - 1ms/epoch - 22us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9138 - 672us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.5871 - 808us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8997 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.5966 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9991 - 617us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.7317 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.4266 - 3ms/epoch - 40us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.4068 - 730us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3642 - 618us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0895 - 739us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.1663 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.9879 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6213 - 1ms/epoch - 17us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.0868 - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9562 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.4962 - 712us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.1445 - 707us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.4513 - 740us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.9327 - 691us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.4781 - 744us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7179 - 607us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 804.6009 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 807.7545 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.5089 - 796us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.3021 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.9892 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6328 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.0723 - 2ms/epoch - 29us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0939 - 636us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.0924 - 759us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.3780 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3519 - 996us/epoch - 16us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 59/300\n", - "Solving for Nash Equilibrium in Generation 59/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.5131 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5907 - 2ms/epoch - 35us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.0864 - 801us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.7268 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.2415 - 733us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8264 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.5594 - 803us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.7899 - 658us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.2181 - 731us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.9117 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0776 - 738us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.8718 - 679us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.9130 - 701us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.9760 - 687us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.4361 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.5684 - 734us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.0126 - 739us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8608 - 746us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.8531 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.9275 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.9401 - 996us/epoch - 16us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.0888 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4246 - 743us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.9953 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.3747 - 638us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.1567 - 731us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7922 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.4205 - 920us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1790 - 576us/epoch - 9us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.1247 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0798 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.7524 - 626us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.1499 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.9423 - 735us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.9561 - 875us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.7455 - 763us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.7125 - 724us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.9865 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.0724 - 3ms/epoch - 53us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.8376 - 759us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.0220 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.1039 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.8362 - 732us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.0882 - 597us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4970 - 728us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.8777 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.8535 - 647us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.0272 - 680us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2853 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6870 - 643us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.6331 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.9462 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6578 - 810us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.8478 - 644us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0054 - 696us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.7913 - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8944 - 720us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.3781 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.4154 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.0782 - 810us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2570 - 615us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.1175 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1938 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.5600 - 684us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.0685 - 698us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.2177 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7156 - 745us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.3800 - 734us/epoch - 12us/sample\n", - "Generation 60/300\n", - "Solving for Nash Equilibrium in Generation 60/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.6757 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.3060 - 701us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.0305 - 660us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8404 - 743us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.7472 - 678us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.0017 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.6283 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.2873 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9687 - 696us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5409 - 756us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.9315 - 769us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.7784 - 618us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6849 - 641us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.5955 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.1235 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.2994 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.6956 - 788us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8890 - 699us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6507 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4409 - 833us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.4504 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3089 - 718us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.1061 - 839us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.8264 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.0430 - 601us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.1313 - 635us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6059 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.5966 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2227 - 755us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.5138 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.3504 - 719us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9402 - 1ms/epoch - 18us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7834 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.6085 - 663us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3464 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.6224 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.7613 - 667us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.4650 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7260 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.4409 - 1ms/epoch - 17us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.1946 - 726us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9704 - 603us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.6341 - 608us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773.5679 - 990us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.4769 - 721us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3903 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6096 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.9958 - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.1532 - 925us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2065 - 858us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.8108 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3556 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8783 - 644us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.7318 - 716us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6166 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.7011 - 814us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8531 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6550 - 741us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.5761 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.7528 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.7641 - 590us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.4518 - 6ms/epoch - 104us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9800 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.9774 - 594us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.2538 - 600us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8292 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.8572 - 656us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.2528 - 642us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.2105 - 839us/epoch - 14us/sample\n", - "Generation 61/300\n", - "Solving for Nash Equilibrium in Generation 61/300\n", + "Generation 23/300\n", + "Solving for Nash Equilibrium in Generation 23/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.7251 - 541us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.1810 - 3ms/epoch - 44us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.2023 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.7323 - 714us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.7943 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4670 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.5269 - 906us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.2735 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.5995 - 872us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.7250 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.0680 - 1ms/epoch - 23us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.3471 - 713us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9235 - 740us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.4024 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8537 - 635us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.6068 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.4308 - 695us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.1138 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.6276 - 682us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3879 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.2841 - 2ms/epoch - 38us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.5661 - 798us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3647 - 669us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.3416 - 610us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.6175 - 589us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.3093 - 691us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.9371 - 603us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.6797 - 684us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5643 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4836 - 716us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.9807 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.6284 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.3297 - 793us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.8131 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2550 - 664us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.1014 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.7250 - 707us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.9880 - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.1934 - 902us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.9202 - 753us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1190 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.3007 - 757us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.4879 - 706us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6617 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.5047 - 663us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.5679 - 754us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5977 - 689us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8434 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.0037 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 140us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.4532 - 3ms/epoch - 47us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.9954 - 980us/epoch - 16us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.8610 - 617us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.1562 - 613us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.1514 - 1ms/epoch - 18us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.0481 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5223 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.3845 - 716us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6472 - 654us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.2889 - 987us/epoch - 16us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.6881 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.0677 - 664us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.1433 - 786us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.3292 - 2ms/epoch - 33us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.4373 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6526 - 781us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.5012 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4056 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.9613 - 695us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.8322 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7823 - 742us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5410 - 746us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.0482 - 748us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.5266 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.5385 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.0778 - 690us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 62/300\n", - "Solving for Nash Equilibrium in Generation 62/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.7554 - 625us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.6479 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 799.7974 - 740us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.8608 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.5041 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.0787 - 960us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.1552 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.6820 - 2ms/epoch - 31us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.8202 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.9957 - 679us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.8352 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.9331 - 649us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.8988 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.7204 - 761us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9255 - 659us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.1948 - 806us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.3081 - 755us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6645 - 641us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.1309 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.6594 - 750us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3035 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.2286 - 629us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.5324 - 837us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 220.9937 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.8511 - 1ms/epoch - 20us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.4542 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.9780 - 703us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.2650 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.5219 - 786us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.0637 - 2ms/epoch - 26us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.8575 - 776us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.1391 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.3874 - 861us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.0344 - 668us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.7752 - 762us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1100 - 2ms/epoch - 38us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.1487 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.9847 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.2246 - 1ms/epoch - 18us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.4290 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8651 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.8530 - 689us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8689 - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.9985 - 826us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.0700 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1310 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.9058 - 685us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.3054 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.3806 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.7195 - 1ms/epoch - 18us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.4751 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.0672 - 822us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7327 - 713us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.6273 - 658us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.5767 - 728us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.3164 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.9849 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9075 - 734us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.2679 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.3714 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3593 - 693us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.2051 - 807us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.7416 - 826us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.6604 - 682us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0896 - 795us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4950 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.1592 - 1ms/epoch - 19us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0034 - 738us/epoch - 12us/sample\n", - "Generation 63/300\n", - "Solving for Nash Equilibrium in Generation 63/300\n", + "Generation 24/300\n", + "Solving for Nash Equilibrium in Generation 24/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 802.5117 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.1006 - 731us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.7072 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7330 - 599us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.6096 - 803us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.6157 - 618us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.1640 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.9666 - 1ms/epoch - 19us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.7425 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.5564 - 705us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9682 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.8477 - 610us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.6345 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.3244 - 708us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7457 - 764us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.4036 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5513 - 730us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6212 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8348 - 602us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.6331 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.6602 - 716us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.4398 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.9180 - 672us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.0989 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2953 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.9130 - 889us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.0343 - 790us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.0361 - 852us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9805 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.9151 - 703us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46.0870 - 1ms/epoch - 21us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.5705 - 627us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6523 - 615us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.9136 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.8909 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.8953 - 2ms/epoch - 36us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.1008 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6516 - 672us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.8891 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.2758 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.3293 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.4014 - 717us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0168 - 764us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.7087 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.3239 - 863us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.9660 - 627us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40.4836 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.1378 - 806us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.4153 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.3253 - 663us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.0573 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.2132 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.1150 - 693us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.8189 - 1ms/epoch - 21us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.3574 - 718us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.7000 - 711us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.7475 - 827us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.2325 - 683us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.9664 - 4ms/epoch - 71us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.9780 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.8726 - 877us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.5656 - 1ms/epoch - 18us/sample\n", - "Generation 64/300\n", - "Solving for Nash Equilibrium in Generation 64/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 25/300\n", + "Solving for Nash Equilibrium in Generation 25/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.0295 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.6749 - 882us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.8115 - 2ms/epoch - 29us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2146 - 790us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.2930 - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9389 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.9182 - 1ms/epoch - 16us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37.8259 - 764us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9921 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.2180 - 801us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.5466 - 741us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.5222 - 871us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2661 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5339 - 707us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1447 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2593 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.8569 - 690us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.9239 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787.6362 - 715us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.8709 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.4231 - 644us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.7897 - 793us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.3124 - 731us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9694 - 696us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.0721 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.2693 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.8114 - 3ms/epoch - 47us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0233 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0452 - 956us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2521 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.9451 - 1ms/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.8198 - 711us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.2471 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.0418 - 743us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0341 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.7424 - 792us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.5196 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2174 - 872us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.3352 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.2997 - 599us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.7678 - 671us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.7222 - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.8535 - 2ms/epoch - 35us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.8158 - 793us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 794.9575 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.9971 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3257 - 737us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2285 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4554 - 809us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2497 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3333 - 957us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.6058 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.9669 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8562 - 779us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.9593 - 775us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7062 - 940us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.3000 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.7031 - 824us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.3538 - 817us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.5091 - 861us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69.2074 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4684 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.1437 - 758us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3979 - 725us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.1000 - 800us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6873 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.5808 - 590us/epoch - 10us/sample\n", - "Generation 65/300\n", - "Solving for Nash Equilibrium in Generation 65/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.2402 - 757us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71.8484 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.1602 - 1ms/epoch - 18us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.3596 - 2ms/epoch - 33us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0782 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.4333 - 845us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.1288 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2651 - 802us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.6649 - 918us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.1548 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.5735 - 680us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.9758 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.6795 - 798us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9955 - 752us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2669 - 831us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2437 - 6ms/epoch - 97us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 212.4832 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.1641 - 776us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.3124 - 823us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.2852 - 2ms/epoch - 24us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.1216 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8245 - 1ms/epoch - 17us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.6360 - 892us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.8804 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.6728 - 607us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.7578 - 640us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.7130 - 589us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.7159 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.5715 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.9200 - 718us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.0706 - 729us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.4265 - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.7908 - 2ms/epoch - 35us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.5115 - 854us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.7897 - 697us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.8876 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47.3134 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.4888 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.5062 - 590us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.2958 - 728us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3038 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.9363 - 2ms/epoch - 30us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9899 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.0220 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.3607 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.2912 - 811us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.7061 - 735us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0530 - 779us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3010 - 591us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 801.7679 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7468 - 655us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.6034 - 687us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.9872 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6693 - 802us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.3762 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6598 - 684us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.7908 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9838 - 892us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6026 - 679us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.5119 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.8152 - 585us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.6450 - 702us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.9058 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3786 - 652us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.2703 - 610us/epoch - 10us/sample\n", - "Generation 66/300\n", - "Solving for Nash Equilibrium in Generation 66/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.5114 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.4311 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.8403 - 780us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.8781 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7324 - 2ms/epoch - 25us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.4633 - 904us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0307 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5612 - 613us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2991 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0490 - 615us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2268 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.6479 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.1537 - 802us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.2244 - 654us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.2924 - 719us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 207.3873 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.1725 - 822us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.8580 - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.2113 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.0280 - 642us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1725 - 621us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231.5875 - 694us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.0150 - 800us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.1550 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.0507 - 1ms/epoch - 17us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.4995 - 687us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.6403 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5681 - 876us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7815 - 623us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4786 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6904 - 696us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49.3799 - 694us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3295 - 676us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.0578 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.0023 - 632us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.3809 - 695us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7524 - 625us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.2490 - 1ms/epoch - 21us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.4185 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.0362 - 641us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.4513 - 901us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.8498 - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.0962 - 824us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.5267 - 699us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7496 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.0643 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7366 - 596us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0107 - 769us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.0036 - 725us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6263 - 565us/epoch - 9us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4583 - 766us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.3962 - 968us/epoch - 16us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.7256 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.5136 - 817us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.2961 - 759us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0129 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.6416 - 846us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2170 - 1ms/epoch - 19us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2162 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6307 - 691us/epoch - 11us/sample\n", - "Generation 67/300\n", - "Solving for Nash Equilibrium in Generation 67/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.5282 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1906 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8990 - 670us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.6146 - 681us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.3524 - 672us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9954 - 699us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.5682 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0036 - 893us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5703 - 774us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.0388 - 650us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.3853 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6551 - 672us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.5921 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5415 - 8ms/epoch - 122us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8099 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4253 - 1ms/epoch - 17us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8807 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9260 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4942 - 4ms/epoch - 59us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.9557 - 828us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8212 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.8737 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4595 - 764us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.4752 - 853us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0630 - 636us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.5901 - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.8362 - 1ms/epoch - 17us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.4966 - 872us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.0627 - 805us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.0209 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.9828 - 913us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.5270 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8613 - 637us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.6197 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5488 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.5212 - 778us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.4560 - 949us/epoch - 15us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5876 - 714us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2417 - 882us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5278 - 716us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.8551 - 746us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3337 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.6678 - 979us/epoch - 16us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5604 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.6168 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7804 - 864us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.1343 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0620 - 679us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.0866 - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2556 - 748us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1137 - 870us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.5232 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.4701 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.6460 - 814us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.6245 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.9498 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.6430 - 864us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.4061 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.8134 - 748us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4308 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.4542 - 814us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6368 - 734us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7438 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.2383 - 11ms/epoch - 183us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.6031 - 853us/epoch - 14us/sample\n", - "Generation 68/300\n", - "Solving for Nash Equilibrium in Generation 68/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.8250 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.0350 - 774us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.4780 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.1358 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.1474 - 587us/epoch - 9us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.1301 - 679us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.4761 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.9233 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.2745 - 716us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3417 - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5966 - 910us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.8047 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.4119 - 826us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.4598 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7578 - 822us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.4666 - 765us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.2145 - 718us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.5242 - 2ms/epoch - 25us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.7475 - 658us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5820 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.2590 - 636us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.9385 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.4938 - 768us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.0448 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.4565 - 626us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8645 - 2ms/epoch - 37us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.9808 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4623 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.0452 - 1ms/epoch - 23us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.9641 - 1ms/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8383 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.7718 - 1ms/epoch - 23us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.6703 - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.1434 - 732us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.0189 - 1ms/epoch - 18us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.6354 - 805us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.0648 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.8480 - 2ms/epoch - 25us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.6755 - 2ms/epoch - 36us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.8312 - 1ms/epoch - 16us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.1890 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2435 - 870us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.5899 - 774us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8706 - 1ms/epoch - 21us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 800.8709 - 630us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.8259 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.3193 - 630us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.1672 - 650us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5009 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.1349 - 577us/epoch - 9us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.0771 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.5123 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.3786 - 657us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.3374 - 594us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.0894 - 4ms/epoch - 67us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.0167 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.2909 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.0758 - 1ms/epoch - 20us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.1093 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.1813 - 914us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9904 - 817us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.7030 - 643us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1362 - 605us/epoch - 10us/sample\n", - "Generation 69/300\n", - "Solving for Nash Equilibrium in Generation 69/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.1128 - 882us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9399 - 727us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.3757 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9544 - 750us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41.7694 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.1816 - 643us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.6245 - 655us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9869 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.9749 - 920us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.3346 - 821us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.2771 - 716us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.0699 - 717us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126.3991 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.8879 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3710 - 591us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.3228 - 678us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.4251 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.1003 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6071 - 863us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.2558 - 752us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.2299 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0468 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.0482 - 840us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4508 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.5591 - 673us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.0963 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0669 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.1008 - 2ms/epoch - 30us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5688 - 665us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.5284 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.0725 - 629us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56.1549 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.7449 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9911 - 812us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.3850 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9618 - 778us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.4670 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5490 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6272 - 2ms/epoch - 39us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8854 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.5868 - 721us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.2914 - 729us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.7276 - 705us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.2319 - 773us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.1328 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.8786 - 848us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.2905 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.4934 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.2101 - 841us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5549 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.8348 - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.1443 - 650us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.6163 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.7079 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8452 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.2083 - 784us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.7711 - 688us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6600 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.7374 - 27ms/epoch - 431us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.0656 - 1ms/epoch - 22us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.4540 - 1ms/epoch - 18us/sample\n", - "Generation 70/300\n", - "Solving for Nash Equilibrium in Generation 70/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.9741 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.3814 - 911us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.0554 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.0723 - 742us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.2196 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.6604 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.4261 - 1ms/epoch - 21us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 54ms/epoch - 866us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.7111 - 2ms/epoch - 34us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3920 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4545 - 756us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.0933 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7814 - 774us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4870 - 904us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.5032 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.6508 - 1ms/epoch - 17us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.8209 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0283 - 996us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.5032 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.5457 - 2ms/epoch - 40us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.9409 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.7580 - 757us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.1947 - 927us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5984 - 8ms/epoch - 129us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.6339 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.9779 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.0443 - 812us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7806 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.8979 - 4ms/epoch - 64us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.9884 - 767us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.2705 - 3ms/epoch - 56us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.5014 - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 803.7645 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 799.7155 - 909us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.7016 - 810us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3284 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.5872 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.9142 - 825us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.9450 - 2ms/epoch - 32us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.8015 - 887us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 18ms/epoch - 293us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.1164 - 5ms/epoch - 78us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.7359 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9363 - 917us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3127 - 1ms/epoch - 19us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406.0668 - 1ms/epoch - 19us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.8085 - 2ms/epoch - 30us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.6993 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.6807 - 1ms/epoch - 17us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6731 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0592 - 6ms/epoch - 96us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.1508 - 724us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.0376 - 6ms/epoch - 92us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8214 - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.7108 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.7008 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.8746 - 1ms/epoch - 21us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.7926 - 1ms/epoch - 21us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773.9818 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1892 - 1ms/epoch - 16us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.9480 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.8678 - 1ms/epoch - 21us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.0556 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.8638 - 6ms/epoch - 104us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 71/300\n", - "Solving for Nash Equilibrium in Generation 71/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6305 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.7406 - 1ms/epoch - 21us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.6961 - 909us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.4698 - 1ms/epoch - 18us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5854 - 1ms/epoch - 20us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0524 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8482 - 1ms/epoch - 19us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0466 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.4611 - 728us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.1215 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.2608 - 755us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.8028 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.4905 - 928us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.6478 - 955us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.8094 - 848us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1895 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.2137 - 706us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.9355 - 693us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.7914 - 866us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4783 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.9876 - 807us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3670 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.5050 - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.2856 - 668us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3950 - 758us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 806.3345 - 820us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7782 - 846us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.6641 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0669 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.3205 - 662us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.0758 - 683us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6502 - 739us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.9573 - 868us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4468 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.3314 - 688us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9697 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.9341 - 791us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.7229 - 695us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.6786 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.3323 - 743us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5822 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7498 - 666us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8239 - 692us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.9978 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.8005 - 1ms/epoch - 20us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.3096 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.3129 - 1ms/epoch - 19us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.1725 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.3698 - 979us/epoch - 16us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.7833 - 942us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.0181 - 1ms/epoch - 22us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.5685 - 778us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.1626 - 1ms/epoch - 16us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7565 - 918us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.4909 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.3178 - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.8432 - 782us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.5750 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.8746 - 855us/epoch - 14us/sample\n", - "Generation 72/300\n", - "Solving for Nash Equilibrium in Generation 72/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.5053 - 759us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.5567 - 1ms/epoch - 20us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.2887 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5159 - 849us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1779 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.0967 - 662us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.5081 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9100 - 728us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.3262 - 915us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6288 - 925us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.6467 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9185 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.4797 - 684us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.9837 - 712us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.6099 - 711us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4443 - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.2914 - 779us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.7394 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.6957 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.6807 - 693us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9393 - 2ms/epoch - 36us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8401 - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.6498 - 693us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 804.6340 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.9417 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.3043 - 742us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.6852 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.3791 - 639us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.2238 - 636us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 790.1090 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7563 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2245 - 707us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.0284 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.6176 - 801us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.5309 - 659us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.7189 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0806 - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.2440 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0335 - 633us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.7769 - 1ms/epoch - 23us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9538 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.5827 - 655us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.7420 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.1366 - 724us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.2197 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.4673 - 650us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.3118 - 743us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8501 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5037 - 738us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7878 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.9489 - 783us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.9772 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.0781 - 815us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9788 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0171 - 602us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9391 - 779us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.6984 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.2481 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.0085 - 600us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.3718 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.2974 - 709us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.6495 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.8334 - 2ms/epoch - 37us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6469 - 770us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.8560 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.6281 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.7849 - 833us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8863 - 674us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 791.3621 - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.7067 - 746us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.4893 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.9431 - 801us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.1368 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.1479 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.2766 - 686us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.5396 - 648us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.2990 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4063 - 686us/epoch - 11us/sample\n", - "Generation 73/300\n", - "Solving for Nash Equilibrium in Generation 73/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.2094 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5066 - 921us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.5592 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.4949 - 1ms/epoch - 21us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.8654 - 728us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.7436 - 2ms/epoch - 26us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.6358 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.1981 - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.1535 - 2ms/epoch - 29us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.9330 - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.7718 - 821us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.1041 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6986 - 683us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.8472 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.0294 - 696us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8116 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1620 - 645us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.8947 - 898us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.2112 - 796us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.0482 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1763 - 830us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.2845 - 1ms/epoch - 18us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.7066 - 686us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.2841 - 760us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.7783 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.3752 - 721us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.1413 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.6613 - 631us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.3959 - 790us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3855 - 770us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4558 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.0145 - 902us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6727 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.0286 - 719us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.1208 - 3ms/epoch - 42us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.0004 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.8607 - 756us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.5568 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.0533 - 749us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5554 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.4935 - 2ms/epoch - 38us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.9092 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.3028 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.3864 - 682us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.1240 - 675us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.9954 - 740us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.3563 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3462 - 707us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6881 - 677us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.0093 - 683us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.5904 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.4748 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.9175 - 848us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.6976 - 2ms/epoch - 34us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.1230 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.6363 - 811us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.1661 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.2361 - 1ms/epoch - 21us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1703 - 709us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.5640 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.3378 - 2ms/epoch - 35us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.0715 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7499 - 946us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.5826 - 750us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9673 - 1ms/epoch - 21us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7618 - 709us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.0556 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.2436 - 754us/epoch - 12us/sample\n", - "Generation 74/300\n", - "Solving for Nash Equilibrium in Generation 74/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48.0491 - 760us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.1284 - 798us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.2248 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.6570 - 4ms/epoch - 68us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 783.4833 - 861us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2489 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5002 - 961us/epoch - 16us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.8469 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.3217 - 580us/epoch - 9us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.4223 - 698us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0698 - 807us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0614 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4437 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.8572 - 574us/epoch - 9us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7516 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.9236 - 598us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.8011 - 783us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.9297 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.7622 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.2981 - 659us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.9069 - 665us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 811.1662 - 686us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.8369 - 761us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.9968 - 896us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.1556 - 746us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.4764 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.4064 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4979 - 651us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.4341 - 699us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5915 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.5341 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4550 - 670us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0749 - 691us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.3070 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.2536 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.8247 - 819us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.7612 - 683us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.8873 - 563us/epoch - 9us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6604 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6432 - 645us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8819 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.1456 - 820us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.8811 - 816us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.3269 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1019 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.1635 - 744us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.8911 - 691us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.1199 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.9750 - 611us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.6437 - 753us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7410 - 718us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.9041 - 570us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.6608 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6478 - 875us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.3818 - 675us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.0755 - 775us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.4553 - 738us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.9329 - 603us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3015 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.9797 - 647us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.2042 - 756us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8534 - 785us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.8481 - 810us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.2841 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2957 - 952us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.0197 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.1630 - 677us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.1687 - 656us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.0962 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.8346 - 1ms/epoch - 19us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.4413 - 713us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.3781 - 822us/epoch - 13us/sample\n", - "Generation 75/300\n", - "Solving for Nash Equilibrium in Generation 75/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.2628 - 671us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2708 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.8020 - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9324 - 719us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.9388 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6281 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.1135 - 763us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.1994 - 820us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.8116 - 685us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.8471 - 753us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.0015 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.4683 - 725us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.7115 - 4ms/epoch - 68us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5573 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3725 - 649us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.6275 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8914 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.3625 - 830us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.5027 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8388 - 755us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.6441 - 800us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.9155 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.8867 - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.2001 - 732us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.6640 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.0334 - 807us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.0670 - 830us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.5926 - 669us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.9399 - 709us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.6152 - 738us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.6783 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.2541 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.5641 - 685us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.1795 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.5128 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2183 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.7728 - 750us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.0217 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.6608 - 822us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7468 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.7562 - 725us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.0609 - 660us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.4414 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5686 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.6334 - 653us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0825 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.0711 - 685us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4880 - 812us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.9589 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.7266 - 765us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.2651 - 2ms/epoch - 30us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.3802 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.3124 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.2159 - 906us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3375 - 818us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.2030 - 688us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.9839 - 847us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6901 - 1ms/epoch - 24us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.6227 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.4609 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.4918 - 626us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.0080 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6849 - 696us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2719 - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.8518 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5783 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.9722 - 756us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.7801 - 569us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2341 - 687us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0903 - 784us/epoch - 13us/sample\n", - "Generation 76/300\n", - "Solving for Nash Equilibrium in Generation 76/300\n", + "Generation 26/300\n", + "Solving for Nash Equilibrium in Generation 26/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.4407 - 686us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3347 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3293 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6467 - 746us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.8574 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.5826 - 674us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.3859 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3055 - 753us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.9584 - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.0901 - 653us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9819 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.4787 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.9698 - 649us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.8061 - 772us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.5392 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.6412 - 616us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.1870 - 653us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.2006 - 653us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.3890 - 718us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.9097 - 747us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.5861 - 700us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.5155 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6136 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.0120 - 703us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.7364 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.8190 - 702us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.5174 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.7531 - 643us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.7288 - 648us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.3954 - 1ms/epoch - 20us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3903 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 435.9763 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.5252 - 748us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.6418 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.2446 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0526 - 788us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.8890 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.8549 - 2ms/epoch - 27us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.1913 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.5025 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.8699 - 738us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0690 - 841us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.7586 - 689us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.9360 - 869us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.7614 - 800us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7154 - 856us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.6530 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.9869 - 900us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.5621 - 731us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.1891 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9515 - 751us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.4523 - 700us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1835 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.2552 - 764us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.9663 - 743us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.9693 - 691us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.9843 - 672us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3596 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7147 - 2ms/epoch - 35us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0475 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.2766 - 766us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2550 - 597us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7541 - 2ms/epoch - 24us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3126 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5491 - 1ms/epoch - 16us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 77/300\n", - "Solving for Nash Equilibrium in Generation 77/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8134 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9963 - 1ms/epoch - 23us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0107 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.5839 - 980us/epoch - 16us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9765 - 1ms/epoch - 17us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.0083 - 715us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77.3197 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.0084 - 686us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.7835 - 4ms/epoch - 64us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9498 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2799 - 719us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.0419 - 720us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.2688 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.2560 - 616us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.0174 - 731us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.8618 - 771us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.0303 - 903us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.7554 - 729us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2084 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1752 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4308 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.2365 - 692us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.5826 - 685us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.1326 - 3ms/epoch - 55us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.4071 - 676us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.0930 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.8046 - 583us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.1161 - 617us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8201 - 687us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.0074 - 681us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.6536 - 740us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.7934 - 590us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.0760 - 2ms/epoch - 27us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.5867 - 643us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.4037 - 756us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56.1389 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.0069 - 744us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.8519 - 673us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.8445 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.0262 - 728us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.3881 - 711us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.4524 - 1ms/epoch - 24us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.4133 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.5421 - 919us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.7108 - 780us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4157 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.4206 - 3ms/epoch - 55us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0332 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5782 - 1ms/epoch - 16us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9806 - 722us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2944 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4782 - 665us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.4537 - 4ms/epoch - 70us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5320 - 781us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9700 - 715us/epoch - 12us/sample\n", - "Generation 78/300\n", - "Solving for Nash Equilibrium in Generation 78/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.3754 - 1ms/epoch - 16us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4613 - 706us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.0143 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3081 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.0164 - 638us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.5447 - 674us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 807.3364 - 746us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.8557 - 2ms/epoch - 25us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.5640 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 820.1954 - 529us/epoch - 9us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.8632 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.7536 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.1880 - 754us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.0292 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3768 - 853us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2750 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.5072 - 2ms/epoch - 37us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.9825 - 718us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8600 - 732us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.2904 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.6808 - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4630 - 1ms/epoch - 20us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.4016 - 723us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2106 - 959us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.4232 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.0284 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.8828 - 715us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.5212 - 658us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 64.1619 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.5105 - 6ms/epoch - 92us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.4575 - 1ms/epoch - 18us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.5687 - 714us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6216 - 607us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.5619 - 831us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6298 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46.8875 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.2455 - 680us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.2096 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0563 - 1ms/epoch - 24us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.1860 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8238 - 732us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.9307 - 635us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8992 - 4ms/epoch - 64us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8195 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.1174 - 673us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8364 - 838us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9897 - 794us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.3766 - 699us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.1649 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.1590 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.8605 - 780us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.3306 - 794us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5603 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9167 - 839us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.1497 - 9ms/epoch - 138us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.6263 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.9615 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.1713 - 724us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.3064 - 925us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1835 - 4ms/epoch - 66us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.4405 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4911 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.9481 - 829us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.7599 - 758us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.9841 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.5905 - 791us/epoch - 13us/sample\n", - "Generation 79/300\n", - "Solving for Nash Equilibrium in Generation 79/300\n", + "Generation 27/300\n", + "Solving for Nash Equilibrium in Generation 27/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.7911 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.1807 - 668us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.0712 - 781us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.7535 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.6098 - 867us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.0101 - 919us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 793.4785 - 711us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.2177 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.9320 - 992us/epoch - 16us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.5237 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1207 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7657 - 860us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.2800 - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0162 - 738us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.1030 - 731us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.7849 - 642us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.2972 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.6828 - 711us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6104 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.4760 - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.1774 - 725us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.9536 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.6476 - 1ms/epoch - 16us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.3865 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.6132 - 908us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.7670 - 861us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9314 - 1ms/epoch - 17us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.3925 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.3716 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.0220 - 690us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.2542 - 704us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.2252 - 824us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.6958 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.4832 - 674us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.6575 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7066 - 747us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.4593 - 954us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.5096 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7399 - 803us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.8434 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.6713 - 836us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.9949 - 704us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.0459 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1381 - 647us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.7845 - 731us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.8719 - 805us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.2849 - 808us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.4359 - 911us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.7353 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.6585 - 812us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0198 - 860us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.9441 - 852us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.7346 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2554 - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.0179 - 983us/epoch - 16us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.4445 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.8527 - 778us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9525 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.7543 - 1ms/epoch - 24us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1095 - 710us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9885 - 708us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.9234 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.6479 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.5327 - 1ms/epoch - 17us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.0651 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.7172 - 738us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.5421 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.4652 - 1ms/epoch - 17us/sample\n", - "Generation 80/300\n", - "Solving for Nash Equilibrium in Generation 80/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.6369 - 932us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.3639 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.6609 - 708us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.4705 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.7817 - 1ms/epoch - 21us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.2330 - 672us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.2034 - 866us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.7938 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.5814 - 932us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.7811 - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.1128 - 962us/epoch - 16us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6827 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.9689 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.2749 - 879us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1092 - 994us/epoch - 16us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8893 - 1ms/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.0071 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4455 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4329 - 1ms/epoch - 18us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2465 - 798us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6986 - 844us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.4536 - 1ms/epoch - 21us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.3499 - 887us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.5086 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.0668 - 738us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.5803 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.6933 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2265 - 898us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9626 - 704us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.7313 - 828us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.7437 - 788us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.1374 - 759us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.7303 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.8606 - 598us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.6579 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.0576 - 753us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 786.5776 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.9987 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.1889 - 1ms/epoch - 16us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.0806 - 999us/epoch - 16us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6168 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.2256 - 700us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.8377 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.4033 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.2678 - 3ms/epoch - 46us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.8253 - 663us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.0526 - 833us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.3547 - 666us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6246 - 667us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.1467 - 863us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.6768 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.8337 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.6703 - 701us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.2823 - 622us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 778.4406 - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.6027 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.1909 - 703us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.3891 - 755us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3055 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.1607 - 1ms/epoch - 17us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.8698 - 919us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.7971 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.8215 - 673us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.5561 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.4595 - 666us/epoch - 11us/sample\n", - "Generation 81/300\n", - "Solving for Nash Equilibrium in Generation 81/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.2418 - 631us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 785.4047 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.5913 - 760us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.3676 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.6848 - 709us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.5157 - 577us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2776 - 606us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1517 - 747us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.4199 - 851us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 145.0096 - 2ms/epoch - 40us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.6051 - 680us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.2248 - 2ms/epoch - 29us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2911 - 629us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.7797 - 683us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 785.3810 - 725us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3345 - 787us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.3134 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.4858 - 976us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.5034 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.6912 - 793us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.5916 - 589us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9349 - 587us/epoch - 9us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3542 - 779us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.1683 - 738us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.0356 - 690us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7327 - 668us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.8155 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.7905 - 655us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.6375 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.3457 - 856us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3061 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.6628 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.7371 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.3802 - 863us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1723 - 758us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.6351 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.4305 - 689us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.9387 - 760us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.9774 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.5417 - 735us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.3517 - 777us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6940 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5888 - 984us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.1985 - 609us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1999 - 749us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.4590 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.1125 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.8247 - 796us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.1700 - 762us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1958 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.1837 - 1ms/epoch - 24us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.0169 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.4282 - 821us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.0920 - 1ms/epoch - 24us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.2581 - 703us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.0576 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.0564 - 711us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.9568 - 5ms/epoch - 82us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.7172 - 712us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.9211 - 790us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.2206 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.3688 - 726us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.0544 - 700us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.5009 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.1606 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.6251 - 666us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79.7647 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.0040 - 1ms/epoch - 18us/sample\n", - "Generation 82/300\n", - "Solving for Nash Equilibrium in Generation 82/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9232 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4390 - 1ms/epoch - 17us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.8266 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.4656 - 758us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.8236 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.9765 - 615us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.3498 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.8890 - 620us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.1973 - 843us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.6341 - 721us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8191 - 762us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.8331 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.5062 - 875us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.4467 - 680us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.0230 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.9822 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3693 - 619us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.1105 - 686us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 797.4335 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7538 - 668us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.2763 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.6161 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.8027 - 710us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.2278 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.1310 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2691 - 732us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.0572 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.8770 - 857us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.0272 - 4ms/epoch - 57us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.7332 - 725us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.6906 - 796us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.1145 - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5983 - 704us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.0451 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4608 - 635us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.2203 - 827us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2310 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3712 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.3925 - 712us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5444 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.5480 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.1675 - 1ms/epoch - 19us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5911 - 994us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.3296 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.2162 - 768us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.6779 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.4553 - 789us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.7206 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.7502 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.0411 - 647us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.1879 - 744us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 789.9816 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7031 - 2ms/epoch - 32us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.5120 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.8569 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.7863 - 712us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.7277 - 775us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.2536 - 823us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4573 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9789 - 809us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.0938 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.5657 - 732us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.6698 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.2244 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7070 - 737us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.9500 - 772us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.8460 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.1102 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6539 - 599us/epoch - 10us/sample\n", - "Generation 83/300\n", - "Solving for Nash Equilibrium in Generation 83/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.6379 - 800us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8861 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7258 - 2ms/epoch - 37us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.5572 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.4685 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3713 - 1ms/epoch - 16us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.4870 - 751us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.7439 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2332 - 674us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.3048 - 2ms/epoch - 25us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.2553 - 598us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.2668 - 630us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.4616 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3224 - 1ms/epoch - 24us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.2386 - 780us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.0238 - 975us/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.1027 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4952 - 878us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4901 - 768us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6964 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.3248 - 694us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.5714 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.2843 - 2ms/epoch - 27us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.6004 - 2ms/epoch - 29us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.7697 - 768us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.0326 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.8219 - 1ms/epoch - 19us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2159 - 6ms/epoch - 101us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.1151 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.3943 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.8362 - 3ms/epoch - 47us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.2721 - 862us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.1674 - 2ms/epoch - 26us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.0806 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.0991 - 678us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.7957 - 643us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.8950 - 2ms/epoch - 32us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9591 - 3ms/epoch - 42us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5023 - 664us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.0699 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5898 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.5411 - 755us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4192 - 690us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.5291 - 682us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1699 - 626us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.2170 - 1ms/epoch - 19us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.3255 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.4283 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1147 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7868 - 666us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9377 - 655us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 804.8691 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.0511 - 708us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.8044 - 661us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.7445 - 641us/epoch - 10us/sample\n", - "Generation 84/300\n", - "Solving for Nash Equilibrium in Generation 84/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.3036 - 790us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.8325 - 699us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.0711 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.8739 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5418 - 715us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.8934 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.5634 - 5ms/epoch - 77us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3469 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.8757 - 580us/epoch - 9us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.0635 - 729us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.2381 - 758us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.1756 - 662us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7614 - 654us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.9677 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.5900 - 981us/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.7361 - 747us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.6705 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.1593 - 600us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.0547 - 754us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.5889 - 768us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.8029 - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.8929 - 898us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.8825 - 707us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.0085 - 712us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.6207 - 547us/epoch - 9us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 216.8188 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9184 - 844us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2800 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2750 - 812us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.2169 - 2ms/epoch - 37us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1326 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.0797 - 633us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.2963 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.7892 - 704us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.5269 - 653us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.0995 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6254 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.1033 - 729us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.0606 - 878us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.6631 - 709us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.6970 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.7748 - 697us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.9002 - 730us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.3195 - 732us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.4822 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.8961 - 754us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.5448 - 776us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3965 - 880us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8342 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.6078 - 633us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.8306 - 699us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.0736 - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.7354 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.6924 - 656us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.3112 - 2ms/epoch - 34us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.3859 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.6894 - 695us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.1961 - 698us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.9293 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.1611 - 664us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 107.8777 - 2ms/epoch - 27us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.5772 - 1ms/epoch - 16us/sample\n", - "Generation 85/300\n", - "Solving for Nash Equilibrium in Generation 85/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 778.0698 - 786us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.8644 - 680us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.1277 - 2ms/epoch - 35us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9058 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.4942 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.4069 - 698us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5458 - 2ms/epoch - 32us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.5666 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.7498 - 683us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.0082 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0796 - 856us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.3990 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2314 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9012 - 735us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.6483 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2879 - 2ms/epoch - 35us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0876 - 753us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5910 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7112 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.6614 - 863us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.2305 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8433 - 841us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.6627 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.3287 - 682us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.6839 - 840us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.7189 - 911us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1400 - 665us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.4215 - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.2875 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.3615 - 724us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.2939 - 2ms/epoch - 37us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.5826 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.8707 - 851us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.6337 - 938us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.4911 - 849us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.8907 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.3787 - 842us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.1787 - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.2474 - 1ms/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.1699 - 818us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.5216 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.3242 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1592 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.3275 - 3ms/epoch - 53us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.3967 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7387 - 857us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.3577 - 710us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4391 - 899us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.5179 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.8578 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4734 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.4277 - 797us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.9078 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4485 - 709us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.0820 - 658us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.4917 - 705us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.2471 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7847 - 714us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0003 - 723us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.9324 - 713us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4415 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.2925 - 691us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5847 - 798us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.9108 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.7223 - 577us/epoch - 9us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9958 - 654us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 181.2410 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9581 - 823us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.8107 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.6546 - 728us/epoch - 12us/sample\n", - "Generation 86/300\n", - "Solving for Nash Equilibrium in Generation 86/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.8854 - 844us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.1115 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.7382 - 621us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.9825 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.4593 - 770us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.8664 - 867us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.9647 - 730us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.9633 - 1ms/epoch - 16us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5206 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.6646 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.4205 - 919us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.9996 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51.7039 - 2ms/epoch - 35us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.2820 - 763us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4577 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.8362 - 749us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.4361 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.2792 - 698us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5414 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.5196 - 807us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.7828 - 653us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0759 - 729us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4491 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.6190 - 6ms/epoch - 89us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.4137 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6815 - 679us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.4062 - 715us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.6201 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.1746 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.7876 - 5ms/epoch - 84us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.1403 - 728us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.2211 - 1ms/epoch - 21us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.8796 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.6723 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.0668 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.5255 - 933us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0436 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.7970 - 640us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.6924 - 756us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.2459 - 894us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4579 - 928us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.4552 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 803.8552 - 11ms/epoch - 180us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.3696 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 155.4028 - 686us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3559 - 762us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.3660 - 694us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.4436 - 2ms/epoch - 27us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8481 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.0222 - 723us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.5033 - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 140us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43.8886 - 906us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8752 - 853us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.6192 - 2ms/epoch - 33us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.7369 - 773us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.5495 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.2571 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.4838 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9599 - 727us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2601 - 1ms/epoch - 17us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.7820 - 2ms/epoch - 30us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.7787 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.5383 - 846us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.4558 - 694us/epoch - 11us/sample\n", - "Generation 87/300\n", - "Solving for Nash Equilibrium in Generation 87/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.1461 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.2396 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.7639 - 807us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.5026 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7428 - 708us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.0194 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.5501 - 801us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9898 - 5ms/epoch - 86us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.3523 - 883us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.6483 - 704us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.0012 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.4374 - 898us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4943 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.8784 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.7501 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2017 - 737us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.6815 - 1ms/epoch - 21us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.8480 - 1ms/epoch - 17us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.9142 - 5ms/epoch - 75us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.3544 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.8726 - 703us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9478 - 746us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.8585 - 3ms/epoch - 55us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.7018 - 979us/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.4203 - 746us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.7453 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45.0331 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.1468 - 9ms/epoch - 148us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.6739 - 842us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3998 - 802us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.4898 - 1ms/epoch - 21us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4180 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.7061 - 670us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.2545 - 747us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.0126 - 580us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.5326 - 531us/epoch - 9us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.1847 - 748us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.6299 - 721us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.7099 - 1ms/epoch - 22us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.3029 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0408 - 665us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.0421 - 885us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.4677 - 814us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.4474 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.3986 - 981us/epoch - 16us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.9322 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.4332 - 888us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 775.9080 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2349 - 969us/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6888 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.9343 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.3618 - 891us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.9877 - 782us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2620 - 1ms/epoch - 19us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.1005 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0975 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.7851 - 802us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9092 - 838us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.2357 - 1ms/epoch - 20us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2893 - 644us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.2992 - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.3620 - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5554 - 993us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.1066 - 688us/epoch - 11us/sample\n", - "Generation 88/300\n", - "Solving for Nash Equilibrium in Generation 88/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7896 - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.2282 - 873us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.3480 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.0752 - 678us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.5047 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0437 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.8031 - 609us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.9480 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.8693 - 2ms/epoch - 34us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.8435 - 745us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.1048 - 971us/epoch - 16us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.1937 - 675us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.6854 - 779us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.1524 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.8070 - 797us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.9272 - 3ms/epoch - 51us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.1679 - 774us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.1688 - 888us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2874 - 2ms/epoch - 29us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.5419 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9554 - 726us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.2263 - 765us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.2041 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.1679 - 1ms/epoch - 19us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4828 - 749us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.2579 - 823us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5469 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6464 - 835us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.1528 - 889us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.0952 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.6895 - 905us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8944 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.1813 - 679us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.9650 - 5ms/epoch - 73us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.9484 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3987 - 905us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.5224 - 1ms/epoch - 23us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.3363 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.8269 - 596us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.7442 - 672us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6825 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.0339 - 780us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.5924 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6686 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9683 - 1ms/epoch - 18us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.3966 - 833us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.2491 - 772us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.5039 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.5630 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.2077 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.3390 - 727us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.6256 - 658us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.5887 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.5625 - 792us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.2973 - 733us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.5847 - 783us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.1721 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7122 - 1ms/epoch - 20us/sample\n", - "Generation 89/300\n", - "Solving for Nash Equilibrium in Generation 89/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 205.3856 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.5930 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.6157 - 600us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.9741 - 729us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.5326 - 1ms/epoch - 21us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.6600 - 818us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.5002 - 599us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.9408 - 620us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.3809 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.7310 - 739us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.1140 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.4036 - 613us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 796.6478 - 592us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.8484 - 2ms/epoch - 36us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7007 - 811us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.6884 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7372 - 679us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7576 - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 158us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.2310 - 1ms/epoch - 20us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1315 - 572us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.0820 - 635us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2966 - 656us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.4865 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0989 - 926us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2026 - 618us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.6908 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.0979 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.1909 - 2ms/epoch - 40us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6354 - 667us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.4059 - 626us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7374 - 691us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.8027 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.4650 - 2ms/epoch - 35us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9755 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.8143 - 686us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.9261 - 671us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.9425 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.7938 - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.5786 - 690us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.4695 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 791.6884 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.2037 - 836us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.7535 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.3571 - 1ms/epoch - 24us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4415 - 807us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.3093 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5152 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.8707 - 689us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.5518 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.9531 - 2ms/epoch - 37us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.6394 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.3102 - 720us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.3780 - 687us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.0544 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.3148 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7177 - 692us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.4629 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.2304 - 737us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.9542 - 705us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.4479 - 625us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.7604 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.4487 - 8ms/epoch - 133us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.9587 - 717us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.1223 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.4784 - 646us/epoch - 10us/sample\n", - "Generation 90/300\n", - "Solving for Nash Equilibrium in Generation 90/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9319 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4401 - 777us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7755 - 668us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8564 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.5789 - 800us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 176.5195 - 713us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.5161 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6722 - 675us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.4984 - 964us/epoch - 16us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.7126 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.2902 - 844us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.5632 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.7293 - 700us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.4103 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.7618 - 629us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9500 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.6710 - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.5739 - 676us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2415 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.7497 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.1511 - 683us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.2869 - 587us/epoch - 9us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.9392 - 613us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.2699 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.3329 - 617us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.5757 - 668us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.7667 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.6549 - 627us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5399 - 1ms/epoch - 18us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 790.2098 - 774us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.3233 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.9139 - 667us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.3586 - 563us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.1661 - 1ms/epoch - 19us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.4713 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6915 - 2ms/epoch - 34us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.1902 - 842us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.4721 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.2952 - 660us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.1717 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.0054 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3086 - 782us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.7695 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9999 - 1ms/epoch - 16us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3997 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.8093 - 625us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.1873 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.6009 - 742us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.3053 - 911us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.2513 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.3942 - 616us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 796.7064 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5513 - 4ms/epoch - 66us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.5421 - 587us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.8304 - 722us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.4197 - 775us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.3934 - 616us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.9463 - 672us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3602 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.2700 - 728us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 786.6141 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.9262 - 757us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.2415 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8215 - 777us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.3377 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2599 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4739 - 4ms/epoch - 66us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.1713 - 694us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.6302 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.1359 - 708us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.1063 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.2332 - 931us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0375 - 652us/epoch - 11us/sample\n", - "Generation 91/300\n", - "Solving for Nash Equilibrium in Generation 91/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.1832 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.3686 - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9030 - 635us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.6682 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.3998 - 711us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.5312 - 688us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9960 - 671us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.1035 - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.5132 - 885us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.1894 - 673us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.8098 - 621us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.6653 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.1736 - 637us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.2258 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8642 - 912us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.7748 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5213 - 764us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.6023 - 1ms/epoch - 23us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 210.3867 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.7599 - 751us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.7472 - 3ms/epoch - 47us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.7852 - 1ms/epoch - 21us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.9541 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.4104 - 658us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.7878 - 724us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8000 - 989us/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2599 - 867us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.7158 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.7144 - 781us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.1973 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7855 - 753us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.6274 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.0409 - 655us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.5191 - 683us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.2996 - 705us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.3826 - 8ms/epoch - 128us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 102us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 783.6500 - 923us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.7910 - 638us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.1433 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.2049 - 687us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.0184 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6890 - 672us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0465 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7389 - 760us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.4373 - 714us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.9083 - 753us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4706 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.9067 - 758us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.5269 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0943 - 656us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.5042 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.7442 - 685us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6675 - 3ms/epoch - 51us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.4240 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.6246 - 673us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.6296 - 878us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.8097 - 765us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.1162 - 838us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.1185 - 694us/epoch - 11us/sample\n", - "Generation 92/300\n", - "Solving for Nash Equilibrium in Generation 92/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.6689 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.9375 - 977us/epoch - 16us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.9713 - 2ms/epoch - 35us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.3826 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5385 - 671us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9566 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.7576 - 714us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.1972 - 911us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.6062 - 662us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.7649 - 737us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.0629 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.3001 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.0464 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.0715 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.0261 - 2ms/epoch - 25us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.4282 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.1475 - 1ms/epoch - 20us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9543 - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.9552 - 798us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.8892 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.2273 - 693us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.1186 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.6967 - 738us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 794.6401 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.7671 - 1ms/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1612 - 2ms/epoch - 28us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.0547 - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.3660 - 1ms/epoch - 19us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.2143 - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.9712 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.1185 - 934us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.1108 - 798us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.0938 - 972us/epoch - 16us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6155 - 667us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.2812 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.4193 - 706us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.5975 - 1ms/epoch - 22us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.4503 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.8441 - 737us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.8255 - 742us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.9084 - 936us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5286 - 788us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.5529 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.6103 - 745us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.5148 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.3745 - 963us/epoch - 16us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7151 - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7105 - 991us/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7430 - 683us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.3958 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.9121 - 706us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.1988 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.1009 - 570us/epoch - 9us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.5278 - 747us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.0583 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.1633 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6316 - 723us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.0327 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.5231 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.5032 - 814us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.5380 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8161 - 4ms/epoch - 58us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.6544 - 1ms/epoch - 19us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.5823 - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.2918 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5005 - 782us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.1789 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6259 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.4698 - 2ms/epoch - 26us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.6270 - 922us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.2418 - 901us/epoch - 15us/sample\n", - "Generation 93/300\n", - "Solving for Nash Equilibrium in Generation 93/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.8918 - 826us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.2105 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.7562 - 1ms/epoch - 16us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7026 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.3527 - 748us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1140 - 696us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.7734 - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.0341 - 767us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.1499 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.8409 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.0887 - 711us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.4677 - 735us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.6075 - 856us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9211 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.8017 - 4ms/epoch - 57us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8977 - 569us/epoch - 9us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1603 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.6884 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.1555 - 591us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.5367 - 566us/epoch - 9us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9150 - 631us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.6524 - 797us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8644 - 706us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 218.8145 - 752us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9286 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.2698 - 907us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.7927 - 827us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.7579 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.7296 - 747us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.5052 - 707us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.6406 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.0652 - 706us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.1589 - 725us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2244 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4320 - 740us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.6903 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.5626 - 958us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 93.4718 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2953 - 953us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 106us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.3805 - 858us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.4640 - 909us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.4486 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.4075 - 796us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.3138 - 795us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1780 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.7946 - 781us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.3087 - 686us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.0838 - 617us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3275 - 5ms/epoch - 76us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.9237 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.8885 - 802us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.8685 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0912 - 672us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.7558 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4435 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2900 - 634us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.6316 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.0373 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.2574 - 926us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.3237 - 606us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0680 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.3724 - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.1798 - 708us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.5864 - 666us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 94/300\n", - "Solving for Nash Equilibrium in Generation 94/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1884 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.2758 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.9695 - 561us/epoch - 9us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.7332 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.7609 - 3ms/epoch - 48us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.3738 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.3889 - 801us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9476 - 638us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7819 - 761us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.4700 - 956us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9225 - 710us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.0402 - 745us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7208 - 693us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.7530 - 689us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2863 - 620us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.8106 - 630us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.3122 - 756us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3690 - 711us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.7027 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2261 - 916us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4387 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.1780 - 1ms/epoch - 17us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79.2105 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1053 - 828us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.0246 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5364 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9402 - 1ms/epoch - 24us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6693 - 842us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0261 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.1581 - 849us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2850 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.0939 - 1ms/epoch - 23us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6972 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.0371 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.9488 - 811us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.3305 - 5ms/epoch - 84us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.5085 - 868us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.8820 - 707us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.1523 - 1ms/epoch - 24us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.6571 - 673us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.6593 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.8834 - 764us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.6287 - 716us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.2214 - 1ms/epoch - 18us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.1719 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.2533 - 771us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5266 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.0871 - 893us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.8178 - 3ms/epoch - 51us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.8087 - 2ms/epoch - 37us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5914 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0306 - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.1246 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1528 - 2ms/epoch - 26us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4998 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.3041 - 716us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.1136 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.2755 - 904us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.2291 - 9ms/epoch - 147us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.8107 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.1841 - 1ms/epoch - 19us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.5818 - 693us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2296 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.8373 - 654us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.6655 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1550 - 800us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Generation 95/300\n", - "Solving for Nash Equilibrium in Generation 95/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.9388 - 873us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.3132 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.9375 - 766us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.0120 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0555 - 976us/epoch - 16us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.8224 - 717us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.6489 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.2571 - 6ms/epoch - 96us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.9315 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.2018 - 686us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.9388 - 818us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 149us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.8515 - 36ms/epoch - 588us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4953 - 787us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1652 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.2726 - 678us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.3784 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.4789 - 740us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.7608 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2554 - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.8724 - 815us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.8105 - 884us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.8967 - 590us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.5568 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.6276 - 1ms/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.0272 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.1853 - 712us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.1125 - 1ms/epoch - 19us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.7237 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.1291 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.8687 - 2ms/epoch - 30us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.7726 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.0305 - 789us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.6966 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9161 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1600 - 936us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.4467 - 627us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.7691 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.1164 - 663us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.9489 - 744us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.6761 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.3906 - 2ms/epoch - 32us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.2578 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.9962 - 1ms/epoch - 17us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3238 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6173 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.5079 - 725us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.5714 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.9832 - 712us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.1727 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2158 - 893us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.9710 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.1917 - 1ms/epoch - 20us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6175 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.0627 - 822us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.4526 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.6509 - 856us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.9099 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6532 - 719us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.8937 - 735us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3747 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.5709 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.7060 - 2ms/epoch - 34us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.5550 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.6476 - 922us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6802 - 2ms/epoch - 26us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.2156 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.5545 - 720us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45.6697 - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.1671 - 1ms/epoch - 20us/sample\n", - "Generation 96/300\n", - "Solving for Nash Equilibrium in Generation 96/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.6288 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9156 - 3ms/epoch - 46us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.3981 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.6205 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.8307 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.5913 - 778us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4174 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9952 - 968us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 141us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.3354 - 7ms/epoch - 115us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.3668 - 800us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.1057 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.7010 - 660us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.7663 - 1ms/epoch - 23us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7706 - 637us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7457 - 669us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.5272 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.5442 - 712us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.3025 - 793us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.0504 - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8816 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.9789 - 770us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.8925 - 5ms/epoch - 74us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.9432 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7377 - 605us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.4266 - 859us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.0818 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.6600 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.0238 - 832us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.1658 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3948 - 2ms/epoch - 39us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.5723 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.7943 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.0905 - 727us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.5440 - 754us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.1251 - 714us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8436 - 826us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.0958 - 912us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.1610 - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.2544 - 811us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.7931 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.1595 - 770us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.1204 - 804us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.4285 - 759us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 786.9603 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4287 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.1297 - 721us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9556 - 755us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.0674 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.6699 - 928us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9467 - 805us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.8162 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.9400 - 1ms/epoch - 19us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.3597 - 1ms/epoch - 16us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8208 - 2ms/epoch - 32us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.0402 - 1ms/epoch - 18us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.9083 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.3168 - 741us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.3983 - 780us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7846 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5024 - 4ms/epoch - 57us/sample\n", - "Generation 97/300\n", - "Solving for Nash Equilibrium in Generation 97/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.2451 - 904us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.0699 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.1550 - 1ms/epoch - 17us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.7378 - 936us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.1048 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5744 - 842us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.2601 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.6283 - 819us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.1581 - 709us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.0431 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.6729 - 1ms/epoch - 21us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.4988 - 748us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.8655 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.2415 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3785 - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.2078 - 1ms/epoch - 16us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.2573 - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.8137 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.9688 - 945us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1541 - 2ms/epoch - 30us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1163 - 938us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.2387 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.7151 - 997us/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 124us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.9203 - 5ms/epoch - 76us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.9941 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.2844 - 683us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.5048 - 779us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.6639 - 621us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.2855 - 3ms/epoch - 45us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.7391 - 3ms/epoch - 52us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.5569 - 962us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.0684 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.7595 - 1ms/epoch - 17us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.1823 - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.1209 - 1ms/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.6106 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.8292 - 2ms/epoch - 39us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1880 - 811us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.6295 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4332 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.9192 - 751us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.5451 - 1ms/epoch - 20us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.3140 - 831us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5076 - 724us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.3622 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 147us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2016 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.0583 - 837us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.4896 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.7007 - 772us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0003 - 811us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9227 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.2003 - 751us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.1256 - 778us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1492 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6828 - 2ms/epoch - 38us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.6700 - 1ms/epoch - 21us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.1119 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5925 - 833us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4098 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.5529 - 705us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.7369 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.8275 - 945us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9051 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.3770 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7040 - 789us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3820 - 805us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.3236 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.7373 - 1ms/epoch - 21us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.8013 - 938us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.3300 - 911us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.1207 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.7645 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.7975 - 1ms/epoch - 20us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.7039 - 665us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3873 - 722us/epoch - 12us/sample\n", - "Generation 98/300\n", - "Solving for Nash Equilibrium in Generation 98/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7580 - 1ms/epoch - 17us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.6706 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.9193 - 680us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.0425 - 688us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.2045 - 823us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.6218 - 924us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.5827 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.0942 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6718 - 736us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.7237 - 708us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0802 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.4315 - 4ms/epoch - 62us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.8043 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.5288 - 955us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.9836 - 723us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.2288 - 935us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.8263 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.3135 - 761us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.7255 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.8748 - 4ms/epoch - 71us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.4285 - 733us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.9509 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.1129 - 1ms/epoch - 17us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3220 - 757us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.7401 - 775us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.9552 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5240 - 633us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.9808 - 683us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.9261 - 13ms/epoch - 210us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4045 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5090 - 624us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.5914 - 752us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.0291 - 616us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.0236 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.8571 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.5756 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3816 - 2ms/epoch - 33us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9122 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.8312 - 721us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.7694 - 725us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5043 - 783us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.8534 - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.8885 - 710us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.1651 - 728us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5471 - 975us/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.6413 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9242 - 1ms/epoch - 23us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.8465 - 908us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.9890 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.1710 - 826us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.7200 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0743 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.0704 - 1ms/epoch - 22us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.1133 - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.4740 - 763us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.5359 - 710us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.5781 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.6350 - 5ms/epoch - 77us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.5717 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9930 - 4ms/epoch - 64us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.3303 - 3ms/epoch - 50us/sample\n", - "Generation 99/300\n", - "Solving for Nash Equilibrium in Generation 99/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0357 - 2ms/epoch - 32us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.0870 - 721us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.0858 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.8785 - 700us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.4334 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6858 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7208 - 778us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.9635 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.1022 - 1ms/epoch - 19us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.9018 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.6339 - 2ms/epoch - 30us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.8085 - 799us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.6821 - 2ms/epoch - 31us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.0075 - 903us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.1449 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.0579 - 1ms/epoch - 18us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5397 - 4ms/epoch - 72us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2039 - 1ms/epoch - 19us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5219 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.9453 - 2ms/epoch - 32us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.0876 - 822us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 811.1934 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3407 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.7012 - 774us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.3068 - 2ms/epoch - 37us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.3944 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.6214 - 680us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5705 - 780us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9931 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5162 - 672us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 110us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.6638 - 6ms/epoch - 93us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.7023 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 50ms/epoch - 799us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.3178 - 6ms/epoch - 93us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.1374 - 3ms/epoch - 44us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.0694 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.5187 - 841us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9249 - 968us/epoch - 16us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.4977 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.5717 - 741us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2339 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.3560 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.9825 - 694us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.4860 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.1744 - 1ms/epoch - 17us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.0817 - 858us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8616 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9027 - 802us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.9522 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.7214 - 911us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.3133 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.6310 - 7ms/epoch - 115us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.3015 - 737us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.8494 - 819us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.7198 - 773us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.2708 - 4ms/epoch - 61us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.1547 - 788us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.7126 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4929 - 1ms/epoch - 16us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.2434 - 3ms/epoch - 48us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.8680 - 5ms/epoch - 78us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.4476 - 878us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7539 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.2589 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.9754 - 710us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.7673 - 852us/epoch - 14us/sample\n", - "Generation 100/300\n", - "Solving for Nash Equilibrium in Generation 100/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773.0003 - 876us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.6750 - 705us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9077 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0980 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.7260 - 666us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.8546 - 717us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.2525 - 883us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.8802 - 2ms/epoch - 28us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.7472 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.4769 - 1ms/epoch - 20us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.1782 - 846us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.8672 - 833us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.4771 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.0780 - 798us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.6487 - 1ms/epoch - 19us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.2472 - 935us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37.3893 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.9653 - 1ms/epoch - 24us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.2811 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.7595 - 801us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.9155 - 842us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.6816 - 4ms/epoch - 70us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.0275 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.3914 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.7861 - 1ms/epoch - 20us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.8749 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.8231 - 1ms/epoch - 19us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.5049 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7943 - 707us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5106 - 1ms/epoch - 16us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.8623 - 812us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.7116 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.4702 - 842us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.3771 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7841 - 1ms/epoch - 23us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.7874 - 1ms/epoch - 17us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.2660 - 831us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.1907 - 750us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.0654 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.7666 - 824us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.4095 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1546 - 839us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.2608 - 765us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.6136 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.1734 - 1ms/epoch - 19us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.7849 - 752us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.9722 - 724us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.4845 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.5574 - 742us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.0403 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0533 - 642us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.4118 - 617us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.6780 - 737us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.9891 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4001 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.8241 - 5ms/epoch - 76us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.6408 - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3098 - 2ms/epoch - 35us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7341 - 8ms/epoch - 134us/sample\n", - "Generation 101/300\n", - "Solving for Nash Equilibrium in Generation 101/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 85us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.4199 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7795 - 963us/epoch - 16us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.6527 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8766 - 6ms/epoch - 99us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.4426 - 922us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1285 - 727us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.6216 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.0105 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 13ms/epoch - 213us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1741 - 1ms/epoch - 18us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0865 - 994us/epoch - 16us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2370 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.9802 - 991us/epoch - 16us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.4953 - 2ms/epoch - 25us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.3695 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.3972 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.7442 - 1ms/epoch - 18us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.9921 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 12ms/epoch - 196us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.0490 - 3ms/epoch - 56us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.4258 - 12ms/epoch - 194us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.6510 - 1ms/epoch - 20us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.3730 - 1ms/epoch - 18us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.1881 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 13ms/epoch - 205us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.0825 - 2ms/epoch - 39us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.8966 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.4781 - 689us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.9439 - 2ms/epoch - 27us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.4532 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.3015 - 723us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.9616 - 897us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.8124 - 817us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.6133 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.2834 - 774us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.2939 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.5409 - 2ms/epoch - 37us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.2019 - 2ms/epoch - 37us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.1867 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 206.6608 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36.2528 - 838us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.5160 - 673us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.5508 - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6605 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9556 - 810us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.9474 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.7241 - 802us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.4573 - 795us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180.6887 - 743us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.0326 - 683us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.8354 - 752us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.2429 - 744us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.5433 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.5415 - 2ms/epoch - 28us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.8591 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.7425 - 783us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.0898 - 1ms/epoch - 21us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.3130 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.5238 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2584 - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.5239 - 773us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.5448 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.2736 - 844us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.6414 - 960us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.4928 - 795us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776.4285 - 771us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4692 - 723us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8814 - 780us/epoch - 13us/sample\n", - "Generation 102/300\n", - "Solving for Nash Equilibrium in Generation 102/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6230 - 785us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.7059 - 956us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3318 - 724us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7446 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.3717 - 2ms/epoch - 25us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3331 - 735us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.5710 - 722us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5428 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.4087 - 7ms/epoch - 113us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.2112 - 819us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.4205 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.9549 - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6433 - 12ms/epoch - 201us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 802.6586 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1462 - 1ms/epoch - 18us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.7841 - 1ms/epoch - 21us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3535 - 753us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.5632 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.5965 - 9ms/epoch - 146us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9866 - 1ms/epoch - 20us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8853 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8802 - 741us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.9518 - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.4606 - 631us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.2871 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.8306 - 733us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.4286 - 633us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2476 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 41ms/epoch - 660us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.1704 - 107ms/epoch - 2ms/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.4428 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.3263 - 2ms/epoch - 28us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.9937 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.7910 - 1ms/epoch - 17us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.9433 - 835us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9531 - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.1246 - 791us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.5513 - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7558 - 2ms/epoch - 34us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.5109 - 725us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.3742 - 1ms/epoch - 18us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2783 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.7599 - 1ms/epoch - 22us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.3014 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.5876 - 937us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.6680 - 843us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.3770 - 578us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.2737 - 993us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2729 - 716us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.3194 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.1113 - 2ms/epoch - 27us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.3602 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.2952 - 829us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.5811 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7020 - 4ms/epoch - 66us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9429 - 802us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.9072 - 730us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.7852 - 758us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.8093 - 2ms/epoch - 31us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.6030 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.9393 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 130us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.7179 - 11ms/epoch - 179us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8004 - 738us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.8115 - 7ms/epoch - 121us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.3875 - 675us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.3292 - 776us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.4735 - 905us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.9515 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52.0646 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.6808 - 740us/epoch - 12us/sample\n", - "Generation 103/300\n", - "Solving for Nash Equilibrium in Generation 103/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.0296 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.9489 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.3143 - 1ms/epoch - 19us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.1709 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.2070 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 116us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.6029 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.7717 - 1ms/epoch - 16us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.0553 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.8262 - 676us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6915 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9147 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.9412 - 789us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4509 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6434 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4601 - 711us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.5248 - 1ms/epoch - 20us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5653 - 1ms/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8239 - 1ms/epoch - 18us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.1451 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.4007 - 1ms/epoch - 23us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.5212 - 9ms/epoch - 148us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.6647 - 863us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.5944 - 677us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.8547 - 891us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.9397 - 1ms/epoch - 20us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.2897 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2350 - 726us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.1813 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.9427 - 781us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0790 - 981us/epoch - 16us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7151 - 831us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.7378 - 1ms/epoch - 22us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9501 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.5956 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8293 - 615us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.7720 - 2ms/epoch - 34us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.5551 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.2307 - 672us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.5230 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.5316 - 720us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.6586 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.1924 - 1ms/epoch - 19us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.0982 - 889us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5421 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2165 - 2ms/epoch - 39us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.3120 - 876us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.2380 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.3627 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.6021 - 652us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 803.3009 - 678us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.2513 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.8358 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.1473 - 749us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.4952 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7115 - 949us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9169 - 14ms/epoch - 218us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.7985 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9357 - 852us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2869 - 875us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.8569 - 827us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.0367 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.9554 - 786us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9242 - 720us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0107 - 641us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.7863 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.1873 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.2209 - 686us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.7428 - 4ms/epoch - 65us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.4576 - 2ms/epoch - 32us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.7684 - 5ms/epoch - 74us/sample\n", - "Generation 104/300\n", - "Solving for Nash Equilibrium in Generation 104/300\n", + "Generation 28/300\n", + "Solving for Nash Equilibrium in Generation 28/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7192 - 2ms/epoch - 25us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.8978 - 926us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.4041 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.1342 - 898us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4271 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.3711 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1656 - 1ms/epoch - 21us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.7529 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4828 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7491 - 722us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0369 - 833us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.6736 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.3597 - 685us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.2346 - 760us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.5059 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.2620 - 627us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.8235 - 832us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.4047 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3005 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.1837 - 680us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.8339 - 1ms/epoch - 17us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.8776 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.4298 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.9788 - 721us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.4964 - 872us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.3091 - 753us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.1949 - 692us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.0836 - 811us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 217.3339 - 727us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48.9433 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.2958 - 6ms/epoch - 99us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9547 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.7622 - 853us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.7703 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.4144 - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.0663 - 1ms/epoch - 20us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.6916 - 686us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.3931 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.5844 - 763us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.7145 - 777us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6799 - 764us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6591 - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.7042 - 650us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214.8423 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.3915 - 755us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.8312 - 908us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.1996 - 763us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4337 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.7453 - 736us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.0810 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5426 - 672us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.9053 - 694us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6196 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.2859 - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.7309 - 617us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.3505 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.7132 - 7ms/epoch - 105us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8870 - 733us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8207 - 903us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.5968 - 764us/epoch - 12us/sample\n", - "Generation 105/300\n", - "Solving for Nash Equilibrium in Generation 105/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.4309 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9388 - 683us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.0881 - 645us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.3385 - 791us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.0699 - 5ms/epoch - 87us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7090 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7180 - 1ms/epoch - 20us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.0978 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.8134 - 736us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.5416 - 833us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7842 - 1ms/epoch - 21us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9637 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.4027 - 2ms/epoch - 39us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.4836 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8411 - 787us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8457 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.9149 - 713us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.6671 - 727us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0508 - 670us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.2622 - 957us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6661 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.6471 - 730us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.7154 - 679us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.3215 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9757 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.5696 - 783us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.9221 - 5ms/epoch - 80us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.7315 - 748us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.8433 - 594us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.6065 - 700us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.9054 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.2860 - 601us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.4057 - 7ms/epoch - 115us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.5814 - 797us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.4498 - 971us/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.8935 - 892us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.0089 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2074 - 630us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.8843 - 695us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 146.0850 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.2359 - 871us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.8026 - 1ms/epoch - 23us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.1922 - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.4028 - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4039 - 775us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.2740 - 650us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.1881 - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.7578 - 12ms/epoch - 199us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 105.5213 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8042 - 748us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.7555 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.1659 - 704us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.0118 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.9922 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.7838 - 603us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5031 - 4ms/epoch - 71us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.2499 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.7668 - 1ms/epoch - 18us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7307 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.7606 - 816us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.0641 - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.5168 - 704us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.1979 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.1245 - 796us/epoch - 13us/sample\n", - "Generation 106/300\n", - "Solving for Nash Equilibrium in Generation 106/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.2748 - 5ms/epoch - 85us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.8625 - 688us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.0194 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.7749 - 619us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8712 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.8400 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.0724 - 2ms/epoch - 32us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4660 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.9952 - 841us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.3899 - 742us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.9961 - 728us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 17ms/epoch - 270us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 775.0373 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9000 - 701us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.6483 - 987us/epoch - 16us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.7905 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.0897 - 2ms/epoch - 25us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.5753 - 873us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.9371 - 740us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180.2505 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.3679 - 681us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.2896 - 754us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.3027 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.0585 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.4931 - 1ms/epoch - 19us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.2214 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.4053 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.2189 - 677us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.1039 - 768us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.6917 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3184 - 919us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.4797 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.4899 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.7695 - 4ms/epoch - 66us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1196 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.3264 - 829us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.2156 - 740us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.0168 - 866us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.4623 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.7440 - 724us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.1193 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3546 - 2ms/epoch - 25us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.8060 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.2022 - 861us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.7609 - 2ms/epoch - 31us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 806.8109 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.1522 - 643us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.0358 - 936us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.8425 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.2919 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.8922 - 792us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9523 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2253 - 975us/epoch - 16us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.5129 - 784us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7625 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.2072 - 796us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3097 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.2036 - 883us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 791.0524 - 733us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9435 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.7098 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.4292 - 829us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3052 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.4602 - 795us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.3313 - 7ms/epoch - 109us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2769 - 762us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.9534 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.2773 - 695us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.3296 - 698us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.1324 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.7770 - 937us/epoch - 15us/sample\n", - "Generation 107/300\n", - "Solving for Nash Equilibrium in Generation 107/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8803 - 757us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.5851 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.5543 - 817us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.6044 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.8488 - 5ms/epoch - 75us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1739 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8146 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.8590 - 563us/epoch - 9us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.7499 - 1ms/epoch - 18us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9984 - 3ms/epoch - 47us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.3303 - 961us/epoch - 16us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0149 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.8197 - 588us/epoch - 9us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.1290 - 783us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.9954 - 707us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.1359 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3458 - 706us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9737 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.0505 - 724us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3528 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9525 - 810us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7219 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.2727 - 3ms/epoch - 42us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.1406 - 770us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.7275 - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.4888 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.0862 - 820us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.4003 - 1ms/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.3644 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8209 - 3ms/epoch - 45us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.0732 - 890us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.2141 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.7703 - 861us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37.4143 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.8355 - 3ms/epoch - 45us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.0888 - 1ms/epoch - 19us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.8340 - 713us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.6768 - 765us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 11ms/epoch - 179us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.6648 - 5ms/epoch - 83us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.7518 - 1ms/epoch - 17us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.9206 - 761us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.4791 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.8972 - 852us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.2479 - 834us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.8751 - 1ms/epoch - 20us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.2948 - 817us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.7674 - 2ms/epoch - 40us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.9976 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.8995 - 793us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.0474 - 859us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.9611 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 92.7818 - 790us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.6352 - 651us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0813 - 768us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.3866 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.2504 - 859us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5399 - 821us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.2997 - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.6461 - 834us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9523 - 1ms/epoch - 16us/sample\n", - "Generation 108/300\n", - "Solving for Nash Equilibrium in Generation 108/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.8310 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.7446 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9644 - 1ms/epoch - 18us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.0468 - 11ms/epoch - 179us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.3205 - 809us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2418 - 975us/epoch - 16us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.4020 - 802us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.7290 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.9640 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.1167 - 707us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.0509 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.5297 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.6588 - 1ms/epoch - 19us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779.1460 - 709us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.3596 - 4ms/epoch - 60us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.0752 - 737us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.4690 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.1266 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.2757 - 730us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.6472 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.0919 - 775us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.5493 - 706us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.0748 - 756us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.4089 - 684us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.0897 - 600us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.3804 - 955us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.9674 - 11ms/epoch - 172us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.1070 - 1ms/epoch - 22us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.0593 - 1ms/epoch - 17us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.6429 - 796us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.9434 - 749us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.7216 - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8651 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.0285 - 861us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.2225 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.9390 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3537 - 2ms/epoch - 32us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7513 - 705us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.7853 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0709 - 806us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9780 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5847 - 731us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.6506 - 771us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.9462 - 962us/epoch - 16us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.9170 - 717us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.1106 - 788us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.8656 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.6796 - 984us/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.1156 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.6873 - 623us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.5192 - 734us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.6438 - 741us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8392 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.7352 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7944 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.5138 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.1287 - 850us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7070 - 821us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.7147 - 1ms/epoch - 19us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3160 - 743us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0344 - 2ms/epoch - 30us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.2758 - 792us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.3042 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.6768 - 788us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.2771 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9151 - 3ms/epoch - 43us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7157 - 812us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2994 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3845 - 668us/epoch - 11us/sample\n", - "Generation 109/300\n", - "Solving for Nash Equilibrium in Generation 109/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.5585 - 1ms/epoch - 19us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.9183 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.3391 - 874us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8683 - 709us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.2668 - 660us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.7973 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3515 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.0442 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.0211 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8987 - 763us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3256 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2221 - 863us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.4419 - 865us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.4037 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.0636 - 860us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2123 - 728us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.8970 - 896us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.2292 - 723us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.5101 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.9678 - 718us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.7981 - 699us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.0035 - 627us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4971 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.0834 - 780us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.2214 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.3253 - 1ms/epoch - 18us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.5139 - 742us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.5378 - 802us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.1642 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.3300 - 758us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.6332 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.7313 - 637us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.9318 - 699us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 795.8763 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3163 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.5753 - 845us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.7343 - 799us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2900 - 3ms/epoch - 44us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.8299 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.8445 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3812 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.8265 - 9ms/epoch - 145us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.8511 - 710us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1063 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.8189 - 803us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.0781 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.9044 - 830us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.6461 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.7054 - 945us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5428 - 830us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.0888 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.2681 - 959us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.2191 - 848us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.6846 - 799us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.6326 - 1ms/epoch - 17us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.2882 - 887us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214.6260 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.5444 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.3255 - 819us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.1916 - 620us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.6414 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6596 - 855us/epoch - 14us/sample\n", - "Generation 110/300\n", - "Solving for Nash Equilibrium in Generation 110/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.6469 - 737us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.2618 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7882 - 742us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9929 - 762us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.3698 - 796us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.7393 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.3399 - 797us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9497 - 696us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8156 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.3587 - 613us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.8147 - 735us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.2838 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.1880 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.3980 - 829us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.0344 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1974 - 678us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.3608 - 684us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.1855 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.8254 - 805us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232.7924 - 651us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.9890 - 809us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.4232 - 8ms/epoch - 128us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.7143 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.6877 - 621us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.5980 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.0027 - 681us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.8138 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.8779 - 669us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0201 - 686us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773.7896 - 764us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.6370 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.8138 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.2529 - 767us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5444 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.3737 - 853us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.1192 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.1337 - 765us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.8399 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.3877 - 1ms/epoch - 20us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.5414 - 664us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.8219 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.9124 - 794us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5427 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.5869 - 744us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.0487 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.0861 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.8887 - 972us/epoch - 16us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.3614 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.0863 - 634us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.6683 - 613us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.3306 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7831 - 652us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.3419 - 662us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.1912 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.6322 - 707us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.7361 - 2ms/epoch - 32us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.3734 - 796us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.6689 - 681us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.9877 - 1ms/epoch - 18us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0120 - 731us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2929 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 107.0361 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.6287 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.3632 - 2ms/epoch - 28us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.1597 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.5988 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5030 - 763us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 111/300\n", - "Solving for Nash Equilibrium in Generation 111/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4722 - 918us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.6487 - 671us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.9807 - 5ms/epoch - 74us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 223.2108 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7954 - 660us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.4296 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.4771 - 829us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.0887 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.5981 - 580us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.6022 - 554us/epoch - 9us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4778 - 768us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.1661 - 856us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.5416 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.6518 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4406 - 578us/epoch - 9us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.5352 - 679us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.9304 - 713us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.2625 - 3ms/epoch - 41us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.5869 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.4572 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.8785 - 754us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.9526 - 769us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.5837 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.3953 - 689us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0291 - 586us/epoch - 9us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.3602 - 751us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.0879 - 694us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.4901 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.6135 - 719us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.3494 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.8950 - 731us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.6873 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4235 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.7982 - 626us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0446 - 654us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 34ms/epoch - 552us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.5554 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.0041 - 1ms/epoch - 22us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.7315 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.8637 - 779us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.7719 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.7371 - 735us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.2939 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5109 - 1ms/epoch - 18us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.9019 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.4158 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.7379 - 7ms/epoch - 116us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3326 - 900us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6088 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.6406 - 660us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5841 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.9141 - 729us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1175 - 1ms/epoch - 20us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9734 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.9661 - 1ms/epoch - 22us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.2216 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.0916 - 709us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.7599 - 2ms/epoch - 29us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.2158 - 808us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9823 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.5390 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9711 - 669us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.0941 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.5015 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.3757 - 934us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 799.9415 - 613us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.9556 - 751us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48.3458 - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.3989 - 682us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3149 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.3977 - 1ms/epoch - 18us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.4295 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.7336 - 752us/epoch - 12us/sample\n", - "Generation 112/300\n", - "Solving for Nash Equilibrium in Generation 112/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.1861 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.0221 - 615us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.5091 - 708us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1274 - 783us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.6454 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.7444 - 804us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.8890 - 896us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8328 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3167 - 636us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8332 - 721us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 96.8238 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.1559 - 3ms/epoch - 45us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4349 - 949us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.2263 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.2404 - 9ms/epoch - 144us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3292 - 731us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.3271 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.3307 - 863us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.7789 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.3920 - 1ms/epoch - 21us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.5887 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.1815 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6837 - 877us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.7592 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.3527 - 2ms/epoch - 30us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.1888 - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7697 - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.1849 - 863us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.4695 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.3238 - 798us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.0964 - 1ms/epoch - 18us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4692 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.1270 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.0574 - 711us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0112 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8017 - 738us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.5147 - 751us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3851 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.4676 - 677us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.5522 - 696us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5729 - 2ms/epoch - 32us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9802 - 902us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1436 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.5917 - 783us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.2416 - 603us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.4121 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.9785 - 680us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.1593 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.5711 - 880us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.9745 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.0351 - 678us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1523 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6305 - 791us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.1733 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.4246 - 2ms/epoch - 39us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.3801 - 705us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.7457 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.1446 - 2ms/epoch - 29us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.4294 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.2812 - 773us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.5566 - 763us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.1167 - 611us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5471 - 773us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9135 - 980us/epoch - 16us/sample\n", - "Generation 113/300\n", - "Solving for Nash Equilibrium in Generation 113/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.7367 - 710us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.7855 - 730us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8738 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.1663 - 798us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8931 - 891us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2346 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.1979 - 566us/epoch - 9us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4800 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.0645 - 677us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4212 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.4365 - 835us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.5211 - 744us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.8763 - 854us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.9642 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9929 - 5ms/epoch - 83us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.8883 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2783 - 1ms/epoch - 20us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.2459 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.1783 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.5957 - 703us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.7646 - 740us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9980 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.9996 - 662us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.4270 - 728us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.5541 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.9842 - 758us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.2235 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.5163 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.9720 - 3ms/epoch - 45us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.0522 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5632 - 3ms/epoch - 55us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.8323 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.8855 - 680us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.4399 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.1602 - 604us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.9166 - 805us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8409 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.2990 - 7ms/epoch - 109us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.2734 - 846us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.0302 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7801 - 4ms/epoch - 71us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.2684 - 997us/epoch - 16us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8859 - 728us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 85us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.5876 - 4ms/epoch - 64us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.4811 - 716us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.1102 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.7719 - 770us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2436 - 755us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230.7028 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.0615 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.7056 - 578us/epoch - 9us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.3280 - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3623 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.1288 - 815us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.9163 - 701us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.9720 - 749us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.2978 - 1ms/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.3477 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.1861 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1107 - 634us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.0038 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.7241 - 566us/epoch - 9us/sample\n", - "Generation 114/300\n", - "Solving for Nash Equilibrium in Generation 114/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2377 - 819us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.5331 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.9310 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5237 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.4661 - 796us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6702 - 694us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4649 - 1ms/epoch - 22us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.3182 - 898us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 135.7350 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5913 - 778us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.9219 - 1ms/epoch - 17us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.9033 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.3289 - 855us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.9643 - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.7550 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.4496 - 800us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 146.1369 - 728us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.8826 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.6211 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.2601 - 841us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.2898 - 710us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.1401 - 746us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.3236 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.3560 - 591us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.5200 - 787us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.9945 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.0454 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.5340 - 742us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.4532 - 746us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.1826 - 600us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.5168 - 593us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.3120 - 676us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.9939 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.0776 - 606us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0974 - 831us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.8704 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.1805 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5657 - 781us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.9173 - 612us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.0209 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.3646 - 600us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.4493 - 724us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.1364 - 697us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.1578 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 99.0529 - 842us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.6680 - 628us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.3992 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.7758 - 716us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.3481 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.7909 - 594us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.3922 - 697us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5778 - 774us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.6118 - 683us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.3043 - 774us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52.7847 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.6893 - 881us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5551 - 647us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.5977 - 783us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.5555 - 2ms/epoch - 30us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.5189 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.1345 - 689us/epoch - 11us/sample\n", - "Generation 115/300\n", - "Solving for Nash Equilibrium in Generation 115/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.9851 - 1ms/epoch - 20us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.4770 - 738us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.5078 - 590us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 229.1320 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.5757 - 630us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.6851 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.3954 - 670us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.8379 - 792us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.0414 - 1ms/epoch - 17us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.4756 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.3109 - 731us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 803.7352 - 721us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.8433 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.4940 - 745us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.8520 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1211 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.0605 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.2007 - 816us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8605 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.6797 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.8437 - 657us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.1770 - 782us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.9993 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.5891 - 618us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.7044 - 701us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.5695 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4301 - 703us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.2276 - 699us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.9335 - 755us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.2706 - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.6303 - 743us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.0262 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.4630 - 2ms/epoch - 25us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.8365 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.2089 - 673us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.6016 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.5208 - 795us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8918 - 817us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4228 - 755us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.8739 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5918 - 731us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.3267 - 1ms/epoch - 19us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.0350 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.8181 - 691us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.3957 - 635us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.4409 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7346 - 745us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.0004 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7283 - 820us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 16ms/epoch - 261us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.3038 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.1602 - 792us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.3372 - 810us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.2312 - 868us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7922 - 912us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.2634 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.1479 - 725us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.5659 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.0247 - 2ms/epoch - 36us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.9658 - 688us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.0306 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.2755 - 673us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.1016 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6627 - 851us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.2768 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.3284 - 1ms/epoch - 20us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.8378 - 692us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.5921 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.9643 - 698us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.0595 - 857us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.9652 - 1ms/epoch - 21us/sample\n", - "Generation 116/300\n", - "Solving for Nash Equilibrium in Generation 116/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.1507 - 651us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.1890 - 660us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3828 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9760 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4979 - 959us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.3997 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.0117 - 765us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.0688 - 838us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.1981 - 969us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.0567 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8182 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.1946 - 1ms/epoch - 17us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9215 - 745us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.1207 - 640us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.2808 - 786us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.7938 - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.1752 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.1608 - 854us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.5847 - 770us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.8991 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.6755 - 844us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.3481 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.5884 - 561us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.2328 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.6217 - 916us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.8185 - 631us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.8613 - 738us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6057 - 572us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.0577 - 812us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.5020 - 939us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.5826 - 705us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6833 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7308 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.9420 - 591us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.7239 - 679us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.2846 - 663us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.2345 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.9670 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6119 - 865us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.2852 - 622us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.2623 - 992us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.2303 - 725us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.6474 - 648us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.6014 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.3951 - 874us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.5507 - 777us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.6209 - 740us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.0854 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1917 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.2789 - 4ms/epoch - 64us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.9120 - 734us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.8079 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.0764 - 728us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.1727 - 885us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.6077 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.9926 - 693us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.5457 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.3552 - 655us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.7545 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.1555 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.1386 - 2ms/epoch - 26us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.4656 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.1447 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5099 - 711us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.0159 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.7151 - 759us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.5762 - 687us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1888 - 727us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.6070 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.5772 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.3810 - 682us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.3298 - 724us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8336 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.6147 - 1ms/epoch - 21us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.0717 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.0627 - 631us/epoch - 10us/sample\n", - "Generation 117/300\n", - "Solving for Nash Equilibrium in Generation 117/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.9106 - 726us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.3608 - 686us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4361 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.4095 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.4070 - 698us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8068 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.9515 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.8683 - 632us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.0064 - 669us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5626 - 756us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8076 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6287 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.9122 - 652us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.1423 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.2422 - 607us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3364 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.0413 - 852us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4369 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.3585 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.3635 - 669us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.6312 - 869us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.0502 - 619us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.7484 - 906us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9124 - 763us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 798.7646 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773.8197 - 666us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 168.3531 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.3806 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.4009 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.7727 - 835us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.5668 - 980us/epoch - 16us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 76ms/epoch - 1ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.6862 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.3846 - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.4648 - 646us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.8976 - 804us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.5275 - 1ms/epoch - 19us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.8080 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.7885 - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7974 - 2ms/epoch - 29us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.9579 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.6605 - 993us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.3723 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.8739 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.4457 - 872us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.4252 - 955us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.9042 - 7ms/epoch - 117us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.8937 - 987us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0489 - 952us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.5092 - 902us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.9439 - 656us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.2229 - 5ms/epoch - 77us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.3224 - 750us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.9198 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.4865 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2643 - 2ms/epoch - 30us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.5729 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.8508 - 577us/epoch - 9us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.2628 - 661us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.9939 - 788us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.8264 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.2308 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.6354 - 703us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5248 - 928us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.7713 - 763us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.1924 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5427 - 844us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.0805 - 567us/epoch - 9us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.6190 - 953us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9738 - 1ms/epoch - 18us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.6382 - 773us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.2088 - 874us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9262 - 1ms/epoch - 17us/sample\n", - "Generation 118/300\n", - "Solving for Nash Equilibrium in Generation 118/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 796.1811 - 1ms/epoch - 17us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.7766 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.1264 - 983us/epoch - 16us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.4392 - 732us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.7896 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3353 - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.1985 - 828us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5717 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.9391 - 784us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.0435 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.5031 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.0920 - 712us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.9434 - 921us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.8801 - 2ms/epoch - 29us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.4243 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.9204 - 751us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6968 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.2017 - 1ms/epoch - 18us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.6190 - 1ms/epoch - 20us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7133 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.0543 - 766us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.0659 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.5677 - 1ms/epoch - 23us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.5764 - 771us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.8971 - 648us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8984 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3300 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.1581 - 777us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.3860 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.9628 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.8246 - 695us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.0757 - 727us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.3499 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0552 - 730us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.7216 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.3727 - 725us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.2862 - 617us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.7189 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.5226 - 2ms/epoch - 32us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.5356 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.5158 - 765us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8893 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.4258 - 722us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7689 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.4700 - 795us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.8336 - 665us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.3831 - 668us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.5526 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.0274 - 8ms/epoch - 122us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.7892 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5621 - 651us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7299 - 683us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.4820 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6540 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5697 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361.5848 - 661us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.7728 - 1ms/epoch - 23us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.8035 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.6302 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.0730 - 656us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3939 - 743us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.2222 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0766 - 701us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4890 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.8157 - 675us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6033 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.7429 - 763us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1939 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7897 - 2ms/epoch - 28us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.1328 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.1542 - 649us/epoch - 10us/sample\n", - "Generation 119/300\n", - "Solving for Nash Equilibrium in Generation 119/300\n", + "Generation 29/300\n", + "Solving for Nash Equilibrium in Generation 29/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.3926 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7168 - 1ms/epoch - 16us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.1930 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.0483 - 665us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9655 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.6103 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.6202 - 888us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7309 - 4ms/epoch - 69us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.1198 - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.8142 - 1ms/epoch - 20us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.9632 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5821 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.3516 - 684us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.3566 - 978us/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.9244 - 797us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.1609 - 1ms/epoch - 22us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.2105 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9685 - 959us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.5644 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6855 - 763us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.5170 - 751us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5071 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.9078 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.4844 - 772us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.2368 - 998us/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.3391 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.1341 - 957us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.8677 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.1688 - 972us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.6635 - 934us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.2101 - 848us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.4466 - 2ms/epoch - 28us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.9409 - 767us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.8264 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.5096 - 840us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235.7522 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.6658 - 6ms/epoch - 104us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.8823 - 1ms/epoch - 18us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1237 - 699us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.7241 - 4ms/epoch - 62us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 132.0032 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.1928 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1627 - 636us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9507 - 730us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.1603 - 597us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1418 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.7221 - 1ms/epoch - 20us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.9874 - 575us/epoch - 9us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.8106 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.8245 - 1ms/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.9545 - 870us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1652 - 684us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.1285 - 837us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.3308 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.1209 - 850us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.1041 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.8220 - 984us/epoch - 16us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8131 - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1689 - 719us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.5980 - 744us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4565 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.0217 - 992us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.4595 - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2248 - 866us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.3763 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1337 - 880us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.1303 - 787us/epoch - 13us/sample\n", - "Generation 120/300\n", - "Solving for Nash Equilibrium in Generation 120/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.7676 - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 133us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8825 - 2ms/epoch - 28us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.9085 - 784us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.2524 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8810 - 590us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4383 - 812us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.9266 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6576 - 6ms/epoch - 101us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.8093 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9792 - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.6995 - 704us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.2306 - 1000us/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.5069 - 60ms/epoch - 963us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.9240 - 728us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.5417 - 655us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3414 - 734us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.5605 - 753us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.1591 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.3215 - 763us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3680 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0698 - 969us/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.9171 - 628us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.3824 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.8689 - 709us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.6354 - 686us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.2266 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4263 - 4ms/epoch - 66us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.3064 - 753us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6718 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.0166 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.8566 - 676us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.9417 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 157us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.9921 - 875us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.9249 - 832us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.3029 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.3648 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.4611 - 2ms/epoch - 30us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.5625 - 993us/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0122 - 1ms/epoch - 21us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.8555 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.7133 - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.6642 - 1ms/epoch - 18us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.6736 - 993us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.9050 - 848us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2653 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.9231 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.0800 - 737us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1173 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.4893 - 3ms/epoch - 48us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.7939 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.7633 - 679us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2742 - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.2475 - 733us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2560 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7928 - 719us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.9365 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.5033 - 662us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8594 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.2557 - 956us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0328 - 761us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.1046 - 898us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.1514 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3073 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.8483 - 830us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.4196 - 716us/epoch - 12us/sample\n", - "Generation 121/300\n", - "Solving for Nash Equilibrium in Generation 121/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 30/300\n", + "Solving for Nash Equilibrium in Generation 30/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9961 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.0888 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.4294 - 1ms/epoch - 17us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.3593 - 1ms/epoch - 20us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.4672 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.4435 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7118 - 881us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.6024 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.8191 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.6279 - 3ms/epoch - 54us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.7693 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4507 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.6013 - 3ms/epoch - 40us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.6047 - 680us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0692 - 890us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.2338 - 707us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1306 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.7040 - 897us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.3298 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.4911 - 828us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.7288 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.9930 - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.7249 - 899us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5925 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.6063 - 2ms/epoch - 39us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.7257 - 787us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.6859 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.7990 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.2999 - 762us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.8120 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.1999 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.3528 - 670us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.2045 - 996us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4460 - 859us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.1159 - 912us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2492 - 1ms/epoch - 18us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.1345 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2217 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.5236 - 670us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7487 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.6962 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.7876 - 765us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.3882 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.7091 - 881us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4579 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2172 - 1ms/epoch - 18us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.7599 - 1ms/epoch - 19us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1319 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 178.8918 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.4771 - 840us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.9727 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.3748 - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.4945 - 1ms/epoch - 18us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8409 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126.6989 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.8853 - 1ms/epoch - 19us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.1756 - 1ms/epoch - 17us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4643 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.0022 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.3923 - 2ms/epoch - 27us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.3462 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.5876 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.9801 - 1ms/epoch - 19us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.4459 - 989us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.6053 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.5991 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.0482 - 579us/epoch - 9us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.7505 - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.6699 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.6061 - 656us/epoch - 11us/sample\n", - "Generation 122/300\n", - "Solving for Nash Equilibrium in Generation 122/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.0844 - 709us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4023 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.5440 - 792us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.8976 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4695 - 690us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.2960 - 1ms/epoch - 17us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4634 - 723us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.8036 - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.4796 - 6ms/epoch - 91us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.0108 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.9655 - 1ms/epoch - 17us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.5335 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.1624 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.5126 - 715us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.6309 - 788us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4728 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.3632 - 2ms/epoch - 30us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8103 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.0898 - 832us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.7835 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.9782 - 822us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3709 - 859us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6988 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.0729 - 833us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.3225 - 838us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.2091 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1840 - 1ms/epoch - 22us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.3441 - 1ms/epoch - 17us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.1417 - 1ms/epoch - 21us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3560 - 831us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46.5374 - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1837 - 923us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.4290 - 661us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.6245 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7363 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.3062 - 2ms/epoch - 34us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.4501 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.6691 - 2ms/epoch - 36us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.6834 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 102us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1661 - 3ms/epoch - 56us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.6106 - 734us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.0079 - 9ms/epoch - 146us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6252 - 870us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.4576 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.7325 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.7321 - 908us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428.7917 - 929us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.3435 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.6696 - 21ms/epoch - 332us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.5953 - 955us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9691 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.3298 - 718us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.4352 - 604us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.2906 - 755us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.8398 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.2849 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0768 - 5ms/epoch - 80us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.0505 - 757us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0510 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.2587 - 694us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1679 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.0558 - 881us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.4615 - 683us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.9786 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2692 - 586us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1561 - 2ms/epoch - 25us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.5601 - 639us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.6973 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5711 - 784us/epoch - 13us/sample\n", - "Generation 123/300\n", - "Solving for Nash Equilibrium in Generation 123/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.2176 - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.3386 - 923us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.7286 - 875us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7812 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5139 - 776us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.8606 - 748us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.5059 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2471 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8529 - 660us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.5326 - 882us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.4965 - 780us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.0042 - 3ms/epoch - 55us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.3778 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.2742 - 979us/epoch - 16us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7494 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.5725 - 821us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.8525 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.2281 - 3ms/epoch - 55us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.2135 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.1367 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.0728 - 2ms/epoch - 30us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.4750 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.9665 - 1ms/epoch - 17us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 11ms/epoch - 180us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.0372 - 14ms/epoch - 233us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8303 - 2ms/epoch - 32us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6673 - 687us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0807 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.9910 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.7252 - 634us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0990 - 589us/epoch - 9us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.1390 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.8737 - 1ms/epoch - 23us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 14ms/epoch - 220us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.5388 - 11ms/epoch - 184us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.7798 - 827us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.2752 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.4042 - 811us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.1573 - 727us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.0201 - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.8373 - 880us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.2302 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.5280 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5353 - 1ms/epoch - 16us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.4750 - 861us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.9637 - 3ms/epoch - 54us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.3464 - 760us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.5754 - 875us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.4175 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 14ms/epoch - 229us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2346 - 2ms/epoch - 26us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.6959 - 840us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4819 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.3533 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3253 - 2ms/epoch - 27us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.0114 - 1ms/epoch - 17us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.8198 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.0518 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.2169 - 844us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.8306 - 1ms/epoch - 21us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.0994 - 832us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.1834 - 642us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.7723 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.3609 - 789us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250.3674 - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.6840 - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.3301 - 758us/epoch - 12us/sample\n", - "Generation 124/300\n", - "Solving for Nash Equilibrium in Generation 124/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.9019 - 1ms/epoch - 21us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.7618 - 822us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.0422 - 869us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.2077 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.5876 - 1ms/epoch - 17us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.0695 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 151us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.6569 - 2ms/epoch - 26us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.5121 - 1ms/epoch - 20us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.7442 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9600 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9930 - 1ms/epoch - 22us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.5753 - 1ms/epoch - 18us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.6340 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6735 - 935us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.2815 - 805us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2176 - 3ms/epoch - 45us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2638 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.3144 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.9471 - 1ms/epoch - 19us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.0016 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9764 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1614 - 701us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.6942 - 888us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.0963 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.4888 - 1ms/epoch - 19us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3502 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.2013 - 11ms/epoch - 180us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5891 - 2ms/epoch - 24us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.9335 - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.7270 - 1ms/epoch - 22us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.3027 - 3ms/epoch - 54us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.5297 - 1ms/epoch - 18us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.9761 - 2ms/epoch - 30us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.7717 - 11ms/epoch - 176us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.0663 - 717us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.6774 - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.8066 - 930us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.8859 - 826us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6863 - 842us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.5464 - 783us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.4854 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.6515 - 728us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.6593 - 972us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.1354 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.9590 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.4045 - 661us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.8112 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9067 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.3633 - 851us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.3603 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 25ms/epoch - 410us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.2212 - 10ms/epoch - 161us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4996 - 846us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.0937 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.6434 - 768us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.7547 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8101 - 761us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1443 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.2034 - 745us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.8660 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.8201 - 1ms/epoch - 20us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.4666 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.3723 - 1ms/epoch - 19us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.1148 - 985us/epoch - 16us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.3856 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4942 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4220 - 831us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.3979 - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.1859 - 765us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7695 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.7828 - 988us/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.3490 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.9238 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 137us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.5816 - 953us/epoch - 15us/sample\n", - "Generation 125/300\n", - "Solving for Nash Equilibrium in Generation 125/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9638 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3301 - 2ms/epoch - 25us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7794 - 760us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0040 - 995us/epoch - 16us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.8342 - 687us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.3125 - 1ms/epoch - 18us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.3897 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8452 - 839us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.7023 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.9873 - 672us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.0653 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3359 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.7335 - 764us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8712 - 762us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.6558 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7825 - 838us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.7698 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.4514 - 728us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.1214 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.8820 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8378 - 847us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.5090 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.6903 - 734us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.9645 - 779us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5778 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.9677 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1395 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.5531 - 779us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9201 - 740us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.6703 - 787us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.0199 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.4608 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.7658 - 820us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.8315 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.3081 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7819 - 2ms/epoch - 24us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.8264 - 785us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.5327 - 750us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.8199 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.6349 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9534 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8369 - 637us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.1930 - 663us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8144 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0944 - 2ms/epoch - 28us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.0232 - 926us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4353 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.0685 - 919us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.2548 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0172 - 771us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.2461 - 749us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.7863 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.4344 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8688 - 2ms/epoch - 33us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.9095 - 1ms/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9974 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.9880 - 711us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4744 - 969us/epoch - 16us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.0939 - 540us/epoch - 9us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.9290 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.7370 - 865us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6340 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.3846 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.8200 - 888us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.4442 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.8300 - 643us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2654 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 18ms/epoch - 285us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.9731 - 7ms/epoch - 107us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0355 - 1ms/epoch - 24us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.9040 - 962us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6150 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.0476 - 967us/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.3999 - 746us/epoch - 12us/sample\n", - "Generation 126/300\n", - "Solving for Nash Equilibrium in Generation 126/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.4554 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.9140 - 849us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.9053 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.0850 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.0101 - 928us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.4034 - 716us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1364 - 4ms/epoch - 69us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.2672 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.0165 - 909us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.0406 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.6174 - 792us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 98us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.7105 - 3ms/epoch - 48us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.3211 - 880us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.3331 - 3ms/epoch - 50us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79.8626 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.3706 - 744us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.2975 - 965us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6421 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.2408 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.3536 - 699us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.1750 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.7196 - 739us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1386 - 861us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.0912 - 875us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.5981 - 788us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0441 - 2ms/epoch - 25us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.1656 - 1ms/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.8810 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7104 - 637us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.6589 - 870us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.1736 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.7785 - 828us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.3575 - 1ms/epoch - 21us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.6530 - 799us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.6176 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.1391 - 951us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.4094 - 922us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.1270 - 896us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.0781 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.8052 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.1280 - 1ms/epoch - 21us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.1462 - 918us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.8725 - 1ms/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.7993 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.4466 - 832us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.2596 - 719us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5345 - 1ms/epoch - 21us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.2417 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.6840 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.9196 - 2ms/epoch - 25us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.9160 - 1ms/epoch - 18us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.3474 - 1ms/epoch - 22us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.0517 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8687 - 958us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83.8214 - 2ms/epoch - 25us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.0243 - 996us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.0687 - 758us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2267 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.0695 - 2ms/epoch - 28us/sample\n", - "Generation 127/300\n", - "Solving for Nash Equilibrium in Generation 127/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.3200 - 2ms/epoch - 30us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.1611 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.3777 - 5ms/epoch - 85us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3866 - 754us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.4473 - 624us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.9489 - 1ms/epoch - 17us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1277 - 2ms/epoch - 40us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.8397 - 777us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.1944 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 127us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7741 - 2ms/epoch - 26us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.2333 - 1ms/epoch - 18us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.5374 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2645 - 935us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7799 - 948us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.0151 - 799us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.0909 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.0107 - 693us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.7141 - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.2625 - 931us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.9875 - 819us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.7974 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.9371 - 745us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.1737 - 1ms/epoch - 20us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.6446 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.1913 - 717us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.1586 - 3ms/epoch - 53us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0048 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 148us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.6174 - 3ms/epoch - 47us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.3661 - 911us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.2848 - 2ms/epoch - 39us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.7581 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.8256 - 5ms/epoch - 77us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.9559 - 845us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.9415 - 889us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2947 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.3110 - 840us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2341 - 849us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 127us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8224 - 5ms/epoch - 79us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.4315 - 1ms/epoch - 18us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.0091 - 685us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6195 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7249 - 920us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32.7069 - 1ms/epoch - 17us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.1443 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.1906 - 700us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.5828 - 957us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.2891 - 2ms/epoch - 37us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.7650 - 2ms/epoch - 27us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.3474 - 13ms/epoch - 216us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9990 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.3503 - 759us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.2547 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.8037 - 916us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.3514 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.9185 - 767us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.6776 - 1ms/epoch - 16us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 128/300\n", - "Solving for Nash Equilibrium in Generation 128/300\n", + "Generation 31/300\n", + "Solving for Nash Equilibrium in Generation 31/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 219.9710 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.4754 - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.5207 - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.1203 - 8ms/epoch - 132us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.6729 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.3645 - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.6643 - 785us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0042 - 669us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4380 - 768us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.7167 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.6850 - 6ms/epoch - 101us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.4045 - 1ms/epoch - 23us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.3497 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.6653 - 1ms/epoch - 16us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.0806 - 939us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3740 - 1ms/epoch - 22us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.6141 - 859us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.0565 - 1ms/epoch - 24us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.6266 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.9178 - 753us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.5890 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 104us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.0914 - 5ms/epoch - 80us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.0264 - 813us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.6447 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.6109 - 933us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.3390 - 825us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.6354 - 1ms/epoch - 18us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.2944 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.9445 - 1ms/epoch - 17us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2891 - 6ms/epoch - 99us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.6973 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.6664 - 665us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.6998 - 844us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.0311 - 1ms/epoch - 20us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.3699 - 1ms/epoch - 17us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2467 - 853us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.1214 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.8911 - 729us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1563 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 140us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.2992 - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4529 - 911us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.4365 - 755us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.8958 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2252 - 811us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0726 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4595 - 781us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.0347 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.3041 - 1ms/epoch - 18us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.9081 - 713us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.8507 - 2ms/epoch - 39us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0949 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.1368 - 709us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.0817 - 940us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.4348 - 749us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0136 - 540us/epoch - 9us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.2326 - 668us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.3286 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.5690 - 681us/epoch - 11us/sample\n", - "Generation 129/300\n", - "Solving for Nash Equilibrium in Generation 129/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 32/300\n", + "Solving for Nash Equilibrium in Generation 32/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.6604 - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.6052 - 825us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.7184 - 714us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.0661 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.5395 - 703us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.1243 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.0334 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.7549 - 719us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3220 - 1ms/epoch - 17us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.6609 - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.4939 - 787us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.0092 - 714us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.9318 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2248 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5933 - 825us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.8136 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 793.7245 - 982us/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6144 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.4288 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.0729 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4271 - 879us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.8401 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9542 - 848us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.6396 - 779us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.7069 - 768us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4561 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.7717 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.5637 - 1ms/epoch - 17us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8852 - 850us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.6615 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.7805 - 813us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.2494 - 721us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.2746 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.3677 - 1ms/epoch - 17us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.5538 - 957us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.4901 - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.4747 - 3ms/epoch - 55us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.2444 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.6976 - 787us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1917 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.9131 - 955us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.4957 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.8226 - 807us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.8596 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8852 - 767us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.4963 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.7317 - 1ms/epoch - 17us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.7558 - 827us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.7212 - 2ms/epoch - 30us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.3717 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0134 - 697us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46.3504 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.0219 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4084 - 880us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9599 - 1ms/epoch - 18us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0452 - 1ms/epoch - 21us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.3783 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.0715 - 3ms/epoch - 47us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.2354 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1440 - 1ms/epoch - 17us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.7821 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.7483 - 786us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6547 - 1ms/epoch - 17us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.2423 - 2ms/epoch - 26us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0552 - 811us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6013 - 713us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.3245 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.9402 - 10ms/epoch - 157us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8692 - 796us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50.4874 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.5340 - 875us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.5931 - 762us/epoch - 12us/sample\n", - "Generation 130/300\n", - "Solving for Nash Equilibrium in Generation 130/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.4229 - 1ms/epoch - 18us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.3743 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.9114 - 1ms/epoch - 22us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6365 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.5671 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 797.4105 - 892us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.0606 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.4206 - 939us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.7346 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.9093 - 814us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8085 - 861us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.7058 - 827us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.0325 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.3399 - 2ms/epoch - 27us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9909 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.1814 - 641us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5204 - 809us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.2154 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0634 - 774us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9048 - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.0469 - 1ms/epoch - 17us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.5345 - 1ms/epoch - 22us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.0659 - 5ms/epoch - 84us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1049 - 788us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.9669 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.0074 - 1ms/epoch - 19us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.7638 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.7393 - 1ms/epoch - 18us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8269 - 3ms/epoch - 51us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.7150 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.5598 - 2ms/epoch - 31us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.1212 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.8420 - 932us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.0763 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6746 - 770us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9240 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.4904 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.9603 - 791us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.0291 - 1ms/epoch - 19us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.5658 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.1246 - 839us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.0375 - 779us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.9464 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3928 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 362.7666 - 782us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.9214 - 778us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.2761 - 704us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.6716 - 891us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.6236 - 1ms/epoch - 19us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.5523 - 790us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.6963 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2297 - 760us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5754 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6995 - 1ms/epoch - 23us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.1711 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.7468 - 650us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.8036 - 1ms/epoch - 20us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.9115 - 988us/epoch - 16us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.4000 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.8188 - 1ms/epoch - 19us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.1658 - 982us/epoch - 16us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.3806 - 10ms/epoch - 165us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.6073 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.0631 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.5761 - 4ms/epoch - 62us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3398 - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.6953 - 6ms/epoch - 103us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.8883 - 1ms/epoch - 23us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.1699 - 6ms/epoch - 91us/sample\n", - "Generation 131/300\n", - "Solving for Nash Equilibrium in Generation 131/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.5092 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.2791 - 986us/epoch - 16us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5612 - 767us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3042 - 592us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.1079 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.7260 - 737us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5592 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.3958 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.6977 - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9989 - 873us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.8196 - 1ms/epoch - 21us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.9748 - 713us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.5217 - 800us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.1999 - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0448 - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4268 - 800us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.7292 - 4ms/epoch - 58us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.8841 - 835us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.8022 - 669us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.8154 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.9925 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.5012 - 614us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.1864 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.5787 - 854us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0080 - 690us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.4032 - 771us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 783.0167 - 2ms/epoch - 37us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.6190 - 901us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.5207 - 754us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6811 - 7ms/epoch - 119us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.3564 - 2ms/epoch - 27us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.5277 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.3037 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.8537 - 782us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.7006 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.7701 - 807us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.4023 - 596us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.7070 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.7353 - 598us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.9105 - 1ms/epoch - 23us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.9040 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.9803 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.2426 - 745us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.5092 - 885us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.9257 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.8366 - 1ms/epoch - 18us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.5622 - 867us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.5837 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.2002 - 948us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.8980 - 928us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8664 - 727us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.0561 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2518 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7501 - 739us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.1411 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.8668 - 756us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.9055 - 850us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.6779 - 614us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.7836 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.8458 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0517 - 796us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.6135 - 751us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.9620 - 709us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2559 - 731us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.0137 - 2ms/epoch - 36us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.6291 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8531 - 869us/epoch - 14us/sample\n", - "Generation 132/300\n", - "Solving for Nash Equilibrium in Generation 132/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4533 - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.2601 - 638us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.3241 - 653us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.4168 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.1185 - 759us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.3213 - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.1443 - 680us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8856 - 711us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.9941 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5383 - 711us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240.6311 - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.9011 - 706us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 152.8723 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.7397 - 966us/epoch - 16us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1481 - 910us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.6408 - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.4539 - 682us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.0264 - 711us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.2447 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.7138 - 690us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.5248 - 824us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.3704 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.1667 - 627us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6855 - 712us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2719 - 915us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8620 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 175.2509 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7908 - 1ms/epoch - 22us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.4896 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.3148 - 2ms/epoch - 40us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.1640 - 668us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.6214 - 888us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4450 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.5958 - 1ms/epoch - 18us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.6238 - 832us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.0504 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2340 - 767us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.2161 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.9381 - 1ms/epoch - 21us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.5864 - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.3694 - 840us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.6360 - 839us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.3707 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.4306 - 759us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.0431 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.0316 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.6306 - 866us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.7645 - 819us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.0552 - 869us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5758 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3787 - 6ms/epoch - 100us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5366 - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6755 - 2ms/epoch - 25us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.8210 - 1ms/epoch - 21us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.7814 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.8334 - 12ms/epoch - 186us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.2327 - 1ms/epoch - 22us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.0278 - 952us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.4923 - 11ms/epoch - 174us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6139 - 1ms/epoch - 22us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.8047 - 970us/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.2148 - 1ms/epoch - 24us/sample\n", - "Generation 133/300\n", - "Solving for Nash Equilibrium in Generation 133/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8716 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.6401 - 8ms/epoch - 131us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46.2513 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.6859 - 2ms/epoch - 32us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.0486 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 16ms/epoch - 261us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.5096 - 1ms/epoch - 17us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8210 - 931us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.8470 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.9194 - 831us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.7670 - 1ms/epoch - 17us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.7451 - 2ms/epoch - 37us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.8251 - 2ms/epoch - 32us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428.7275 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.4172 - 718us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.7769 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.1085 - 2ms/epoch - 39us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2732 - 2ms/epoch - 32us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.2430 - 1ms/epoch - 20us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3317 - 1ms/epoch - 20us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7072 - 1ms/epoch - 17us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.9831 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.4302 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.0023 - 760us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.2881 - 977us/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.5455 - 1ms/epoch - 23us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.5663 - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.6594 - 7ms/epoch - 110us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8220 - 1ms/epoch - 17us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.1471 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.5247 - 3ms/epoch - 42us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.3347 - 834us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.0178 - 743us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.7140 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.5381 - 1ms/epoch - 17us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.2744 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8985 - 793us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.9954 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6136 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.5330 - 2ms/epoch - 28us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2765 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.1852 - 2ms/epoch - 28us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.0554 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.9112 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5911 - 963us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.5902 - 938us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.7399 - 857us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7321 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.8502 - 788us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.7090 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.9918 - 1ms/epoch - 17us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.7351 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.8192 - 1ms/epoch - 17us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.6791 - 1ms/epoch - 18us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.7676 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.1359 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.4144 - 854us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.1469 - 959us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.3013 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.1307 - 753us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 789.9073 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5991 - 1ms/epoch - 21us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.5401 - 1ms/epoch - 23us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 362.8751 - 1ms/epoch - 19us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.3949 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2516 - 716us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.6973 - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.9847 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.5296 - 776us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.8063 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.4747 - 740us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 134/300\n", - "Solving for Nash Equilibrium in Generation 134/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.1617 - 914us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.8047 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.0626 - 2ms/epoch - 24us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.6422 - 2ms/epoch - 24us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.4166 - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.7516 - 786us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7185 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.1277 - 645us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.6721 - 843us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.2066 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3242 - 1ms/epoch - 19us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.9881 - 789us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.1102 - 787us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245.9561 - 802us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.8389 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.3275 - 776us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.0588 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.3724 - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.8926 - 822us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4774 - 693us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.9239 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9666 - 533us/epoch - 9us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.6332 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8605 - 734us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0109 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.3422 - 977us/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0923 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.1061 - 648us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.1528 - 746us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.1705 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2504 - 784us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.3004 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.4792 - 768us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.9229 - 808us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.4171 - 659us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.9382 - 1ms/epoch - 16us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.4383 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.4525 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.7820 - 4ms/epoch - 63us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.5891 - 1ms/epoch - 17us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.1338 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.7145 - 3ms/epoch - 49us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.6305 - 728us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.2848 - 762us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.7495 - 796us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.3958 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.2054 - 710us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.9916 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.6587 - 701us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.1281 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.1473 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3841 - 784us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.4669 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.2713 - 736us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1500 - 692us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.6766 - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.6956 - 872us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 182.0716 - 711us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9976 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.5294 - 670us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2437 - 802us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.2717 - 856us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.2566 - 943us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1089 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.3887 - 717us/epoch - 12us/sample\n", - "Generation 135/300\n", - "Solving for Nash Equilibrium in Generation 135/300\n", + "Generation 33/300\n", + "Solving for Nash Equilibrium in Generation 33/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.9699 - 570us/epoch - 9us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.0246 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 134us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.4729 - 881us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.7426 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.4947 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.4866 - 904us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5988 - 985us/epoch - 16us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 115us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4586 - 1ms/epoch - 22us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4274 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.0743 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.2413 - 923us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8622 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2617 - 885us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.2532 - 871us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.0402 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6700 - 1ms/epoch - 20us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.0981 - 817us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0644 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1524 - 2ms/epoch - 28us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48.6835 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.4691 - 746us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6761 - 1ms/epoch - 20us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.4354 - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.1726 - 785us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.3842 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.6240 - 1ms/epoch - 24us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4251 - 829us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.1183 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.6133 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.1097 - 1ms/epoch - 17us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.8567 - 2ms/epoch - 29us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.1196 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.6059 - 1ms/epoch - 16us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.1941 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.4382 - 826us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.0823 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.7142 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.3563 - 746us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.2388 - 1ms/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.0754 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.2593 - 919us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.8195 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.5124 - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.5457 - 681us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.3191 - 1ms/epoch - 20us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.8891 - 864us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.2367 - 783us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.8186 - 44ms/epoch - 717us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.1879 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.9286 - 2ms/epoch - 36us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.0410 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.5878 - 2ms/epoch - 26us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.0249 - 1ms/epoch - 22us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.2992 - 941us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.3930 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.0321 - 874us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7227 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.7142 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.3459 - 911us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.0717 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.6990 - 1ms/epoch - 21us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.0323 - 2ms/epoch - 33us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.8997 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.3238 - 1ms/epoch - 19us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.0080 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.1196 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 123us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.1182 - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.5574 - 964us/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.1592 - 898us/epoch - 14us/sample\n", - "Generation 136/300\n", - "Solving for Nash Equilibrium in Generation 136/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.7095 - 759us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.1772 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4719 - 2ms/epoch - 40us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2189 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.8658 - 1ms/epoch - 21us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6989 - 960us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.4529 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.9523 - 749us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.4251 - 1ms/epoch - 24us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 790.3416 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8565 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.9110 - 8ms/epoch - 125us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.5643 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.5870 - 806us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.6965 - 1ms/epoch - 21us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.0417 - 942us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3744 - 912us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.1149 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245.1871 - 2ms/epoch - 33us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.4863 - 801us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8322 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.4378 - 641us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.4115 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5950 - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.2928 - 770us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.1465 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.7658 - 3ms/epoch - 54us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.2091 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0079 - 991us/epoch - 16us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2684 - 861us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787.2533 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.5048 - 1ms/epoch - 19us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.6622 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.6159 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5540 - 2ms/epoch - 28us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.0522 - 872us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9148 - 795us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.2090 - 9ms/epoch - 148us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.0112 - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7168 - 963us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1163 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.0336 - 1ms/epoch - 18us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.6682 - 963us/epoch - 16us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361.6968 - 999us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.4733 - 704us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.4932 - 915us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.2191 - 939us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.0334 - 5ms/epoch - 82us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.7943 - 987us/epoch - 16us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.6959 - 853us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.3256 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.6340 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.3291 - 797us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.9603 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.8189 - 1ms/epoch - 18us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.1902 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.5254 - 767us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.2348 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.6969 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.7891 - 896us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.6334 - 985us/epoch - 16us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4773 - 9ms/epoch - 140us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.3849 - 930us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.2049 - 2ms/epoch - 25us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9540 - 828us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.1489 - 719us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6407 - 745us/epoch - 12us/sample\n", - "Generation 137/300\n", - "Solving for Nash Equilibrium in Generation 137/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.2015 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2985 - 678us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.2837 - 1ms/epoch - 18us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.2967 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.9714 - 825us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.1451 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.0283 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.5836 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.7008 - 621us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.6351 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.4008 - 883us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.2939 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1755 - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.9622 - 673us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.0427 - 727us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428.2193 - 3ms/epoch - 41us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0443 - 903us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.7764 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 793.2068 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.0001 - 658us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.5724 - 851us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.1451 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.7054 - 917us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.2909 - 4ms/epoch - 64us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.3375 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6074 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.3228 - 826us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2026 - 643us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.2223 - 2ms/epoch - 26us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3511 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.5129 - 695us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7390 - 781us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.2462 - 684us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.6536 - 903us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1786 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.6612 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.5871 - 817us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.7802 - 695us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.6622 - 756us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.6646 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.6580 - 707us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.5656 - 797us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 12ms/epoch - 197us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.6153 - 2ms/epoch - 28us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.9023 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.0356 - 782us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.0854 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.6122 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.6875 - 742us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2146 - 705us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.3582 - 1ms/epoch - 20us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.7811 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1719 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.4380 - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.3958 - 775us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.2889 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.2196 - 803us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.9850 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.2390 - 718us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.8333 - 904us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.9883 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.1549 - 2ms/epoch - 29us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.0004 - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.5731 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.8811 - 759us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.5174 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.3925 - 1ms/epoch - 18us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.1590 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1270 - 862us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.9517 - 1ms/epoch - 24us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3791 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.3558 - 1ms/epoch - 20us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 138/300\n", - "Solving for Nash Equilibrium in Generation 138/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0279 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.8937 - 892us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.3103 - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.3373 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.7223 - 1ms/epoch - 18us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8891 - 2ms/epoch - 37us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.2653 - 1ms/epoch - 19us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.5193 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.3722 - 776us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.0494 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 11ms/epoch - 182us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.0762 - 7ms/epoch - 114us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.5009 - 810us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.5787 - 762us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5157 - 850us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.6545 - 886us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.5804 - 790us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.0681 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.6417 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.9726 - 826us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98.3670 - 1ms/epoch - 17us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.3894 - 806us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.3675 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.1008 - 774us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.0960 - 1ms/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0604 - 1ms/epoch - 19us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.1676 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.9904 - 3ms/epoch - 49us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8199 - 2ms/epoch - 32us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.2008 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.3152 - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.1776 - 1ms/epoch - 16us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.4785 - 792us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.3627 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.1829 - 1ms/epoch - 20us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.8594 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.8089 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.4301 - 2ms/epoch - 28us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.1477 - 4ms/epoch - 72us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8657 - 664us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.0283 - 703us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.2075 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4448 - 754us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.8026 - 972us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.8557 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8290 - 666us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.6871 - 1ms/epoch - 23us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.2301 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.1406 - 575us/epoch - 9us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.5905 - 836us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.0609 - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4131 - 857us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 84us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.5691 - 8ms/epoch - 132us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.4156 - 901us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.6722 - 841us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6822 - 760us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5496 - 708us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4088 - 865us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.7578 - 755us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0591 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.5218 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.1249 - 3ms/epoch - 48us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.6592 - 931us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.8213 - 672us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6299 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.6635 - 679us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.6483 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.2480 - 660us/epoch - 11us/sample\n", - "Generation 139/300\n", - "Solving for Nash Equilibrium in Generation 139/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.6475 - 1ms/epoch - 17us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1278 - 1ms/epoch - 17us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.7032 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.2280 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.1917 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.8727 - 6ms/epoch - 90us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7946 - 885us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0422 - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.5195 - 868us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.7316 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.5952 - 807us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.9235 - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.2678 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5688 - 783us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.0025 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8074 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.8436 - 3ms/epoch - 42us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.5781 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.0562 - 8ms/epoch - 125us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.4958 - 860us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9385 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6973 - 773us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.2135 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.9217 - 2ms/epoch - 25us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.5245 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.8064 - 819us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 124us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.9185 - 2ms/epoch - 36us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.8871 - 2ms/epoch - 35us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4448 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.1188 - 1ms/epoch - 18us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.0885 - 856us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.7567 - 5ms/epoch - 87us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3204 - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.0560 - 2ms/epoch - 25us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.7547 - 952us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2703 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.6066 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.1216 - 1ms/epoch - 18us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2150 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.0066 - 1ms/epoch - 24us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9199 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.5962 - 2ms/epoch - 27us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8511 - 2ms/epoch - 26us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.8992 - 32ms/epoch - 511us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.6391 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.2952 - 3ms/epoch - 44us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.5235 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.4543 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.3341 - 2ms/epoch - 36us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9814 - 1ms/epoch - 24us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.5225 - 1ms/epoch - 23us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.9919 - 1ms/epoch - 18us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.5910 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.7474 - 1ms/epoch - 21us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.8766 - 1ms/epoch - 23us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 116us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.3400 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.4508 - 817us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.5555 - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.6019 - 2ms/epoch - 25us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.1725 - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.6096 - 785us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.7748 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7535 - 861us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.7030 - 842us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4025 - 754us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.3969 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9223 - 4ms/epoch - 62us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.0345 - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7645 - 885us/epoch - 14us/sample\n", - "Generation 140/300\n", - "Solving for Nash Equilibrium in Generation 140/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.5894 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.7388 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.4791 - 855us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.2272 - 672us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.1806 - 7ms/epoch - 108us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.9573 - 1ms/epoch - 18us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.4059 - 2ms/epoch - 26us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.1876 - 841us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.8187 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.0221 - 769us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.7708 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.7508 - 3ms/epoch - 47us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.4650 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4281 - 1ms/epoch - 17us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.9592 - 811us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.6674 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6685 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.1885 - 709us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9161 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.9757 - 609us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.5359 - 793us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4990 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.0990 - 677us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.3679 - 730us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.3134 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.4218 - 2ms/epoch - 40us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 13ms/epoch - 205us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.4795 - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.0280 - 829us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.7866 - 2ms/epoch - 26us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.0048 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.1055 - 2ms/epoch - 26us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.6197 - 851us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.9773 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.4247 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.3345 - 5ms/epoch - 73us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.0369 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2755 - 3ms/epoch - 42us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.7469 - 2ms/epoch - 36us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.3952 - 980us/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.4845 - 895us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9622 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.3926 - 793us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9873 - 1ms/epoch - 18us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.6566 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.1174 - 767us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.4685 - 897us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.3066 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9606 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.7109 - 1ms/epoch - 17us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.1309 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4264 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.8707 - 1ms/epoch - 20us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0521 - 723us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.7025 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9508 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3833 - 774us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.4559 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.0110 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.8735 - 944us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8178 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.5367 - 806us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.3547 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8111 - 621us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.4278 - 590us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.5140 - 980us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.6699 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.8632 - 672us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.3975 - 1ms/epoch - 19us/sample\n", - "Generation 141/300\n", - "Solving for Nash Equilibrium in Generation 141/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.2427 - 756us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 166us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.4906 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.4082 - 740us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37.9880 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1566 - 1ms/epoch - 19us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.1519 - 816us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.8704 - 747us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.3531 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.7624 - 2ms/epoch - 37us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232.1009 - 772us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8889 - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1775 - 701us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.0211 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.1847 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.5897 - 734us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.2105 - 698us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.0167 - 805us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.4465 - 4ms/epoch - 71us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59.4248 - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.9950 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.3610 - 5ms/epoch - 88us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.4004 - 829us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1245 - 1ms/epoch - 20us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.7417 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.5690 - 841us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.4319 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.4773 - 1ms/epoch - 24us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.5725 - 943us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7559 - 948us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.4360 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.6341 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.0658 - 792us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.0878 - 810us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7177 - 827us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9309 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9680 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.9832 - 599us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.6083 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.4207 - 815us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.0050 - 666us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.3997 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.7016 - 601us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3696 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4969 - 653us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.8716 - 743us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.3628 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.2178 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.3046 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.7548 - 871us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6498 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.8762 - 774us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.6077 - 795us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.4580 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.9313 - 2ms/epoch - 37us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.9885 - 45ms/epoch - 733us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7490 - 709us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.5807 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.9164 - 7ms/epoch - 117us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.0840 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.0466 - 649us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6451 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.2648 - 1ms/epoch - 18us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.0427 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0118 - 657us/epoch - 11us/sample\n", - "Generation 142/300\n", - "Solving for Nash Equilibrium in Generation 142/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.8543 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5923 - 701us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2047 - 3ms/epoch - 50us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.6134 - 680us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7200 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.3694 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.3468 - 641us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.2160 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.1115 - 848us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6683 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.2856 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.8137 - 865us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.7773 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.8301 - 854us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.9823 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.1399 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.8266 - 719us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.4288 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8177 - 814us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.8791 - 689us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.7114 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.5156 - 740us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.3045 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 132us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.3777 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.9608 - 1ms/epoch - 20us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0784 - 2ms/epoch - 36us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.4602 - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.3268 - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.6331 - 1ms/epoch - 22us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.4330 - 6ms/epoch - 103us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8435 - 1ms/epoch - 20us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.2189 - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7951 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2003 - 1ms/epoch - 17us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 14ms/epoch - 224us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.9742 - 3ms/epoch - 45us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.3815 - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.4972 - 1ms/epoch - 21us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3706 - 2ms/epoch - 31us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.9108 - 1ms/epoch - 21us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.2390 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8860 - 900us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.6356 - 1ms/epoch - 16us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.9916 - 1ms/epoch - 19us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.8372 - 768us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.2847 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.7126 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.7864 - 1ms/epoch - 19us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.3804 - 1ms/epoch - 18us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.0858 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7081 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.6815 - 44ms/epoch - 708us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.4130 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.3679 - 936us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.9853 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7731 - 896us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4241 - 727us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.4693 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4917 - 765us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.8226 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.7390 - 766us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.0263 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.7756 - 807us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.2950 - 838us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.6283 - 806us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.9142 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.1635 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.3884 - 882us/epoch - 14us/sample\n", - "Generation 143/300\n", - "Solving for Nash Equilibrium in Generation 143/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.2635 - 758us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.2510 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.4144 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.0010 - 719us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.6130 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.4594 - 750us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.5996 - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.5589 - 1ms/epoch - 21us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.7512 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.7543 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.6464 - 775us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6935 - 739us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.8809 - 837us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.0720 - 701us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.6700 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.9291 - 1ms/epoch - 18us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.9470 - 1ms/epoch - 17us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2855 - 951us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.9166 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.1226 - 988us/epoch - 16us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.4410 - 958us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.2473 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.3084 - 1ms/epoch - 23us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.8197 - 743us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.0516 - 696us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.5080 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.9487 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.0139 - 1ms/epoch - 22us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6769 - 657us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0055 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.1817 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.0948 - 15ms/epoch - 236us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.1807 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.5256 - 2ms/epoch - 37us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.0372 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.2571 - 900us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.3510 - 759us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9747 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6776 - 2ms/epoch - 32us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.2422 - 868us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.5835 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.7851 - 1ms/epoch - 19us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.4082 - 696us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.1031 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4642 - 1ms/epoch - 18us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.1671 - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.0159 - 771us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.2623 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.9008 - 1ms/epoch - 22us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.0098 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.8666 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.1242 - 648us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.5339 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.8065 - 903us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.8867 - 761us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8736 - 721us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.0331 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 219.7492 - 644us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6277 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.1223 - 707us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.1244 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.2700 - 659us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9240 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9075 - 683us/epoch - 11us/sample\n", - "Generation 144/300\n", - "Solving for Nash Equilibrium in Generation 144/300\n", + "Generation 34/300\n", + "Solving for Nash Equilibrium in Generation 34/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.4684 - 3ms/epoch - 42us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.4553 - 988us/epoch - 16us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 188.0264 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.5891 - 1ms/epoch - 22us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.4811 - 4ms/epoch - 62us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.3604 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 362.7502 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.1779 - 749us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.6000 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.8223 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9926 - 1ms/epoch - 17us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.5453 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.3276 - 1ms/epoch - 20us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1176 - 786us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.3624 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4504 - 10ms/epoch - 154us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.4564 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.2006 - 989us/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.3386 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6545 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6685 - 940us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.2168 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.6199 - 2ms/epoch - 36us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5756 - 820us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8773 - 2ms/epoch - 29us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.8100 - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.4023 - 939us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.4506 - 926us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.7226 - 1ms/epoch - 20us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.9669 - 1ms/epoch - 17us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.0616 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 157us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.3082 - 3ms/epoch - 48us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.3277 - 1ms/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.0204 - 890us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 151us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46.9262 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.1090 - 1ms/epoch - 24us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9095 - 1ms/epoch - 16us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.6463 - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.2032 - 2ms/epoch - 34us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.3857 - 800us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.6143 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.0798 - 678us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 137us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.3648 - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.9773 - 814us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.1902 - 959us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.9066 - 998us/epoch - 16us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.2880 - 868us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.8458 - 11ms/epoch - 182us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7429 - 872us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.5822 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.8487 - 3ms/epoch - 52us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.7159 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 93us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.6991 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0125 - 1ms/epoch - 19us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.4620 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.2509 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.5903 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.9532 - 811us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.6649 - 724us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.3117 - 691us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.7629 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.3370 - 685us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.2004 - 1ms/epoch - 20us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.9768 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.5127 - 822us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.6745 - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2422 - 1ms/epoch - 18us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.6108 - 739us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.0115 - 653us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.0305 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8499 - 1ms/epoch - 16us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.7408 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.5146 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.1038 - 683us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.7345 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4404 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3673 - 3ms/epoch - 51us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 145/300\n", - "Solving for Nash Equilibrium in Generation 145/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5576 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.7472 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.5173 - 709us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.8644 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.8342 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.4667 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.1782 - 781us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.3932 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5708 - 863us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.3657 - 904us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3917 - 944us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.5069 - 707us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.9942 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.7420 - 824us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.0059 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6509 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.3676 - 797us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.4141 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.0409 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.7820 - 845us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.3687 - 784us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.2100 - 766us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.0400 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.8204 - 8ms/epoch - 128us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3427 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.8236 - 3ms/epoch - 50us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.0121 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7517 - 848us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.6115 - 996us/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9334 - 2ms/epoch - 24us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6739 - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.8128 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.3657 - 3ms/epoch - 43us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2442 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.0202 - 3ms/epoch - 42us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1247 - 2ms/epoch - 25us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.9451 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5942 - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5959 - 1ms/epoch - 18us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.4929 - 3ms/epoch - 40us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.7708 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.1838 - 2ms/epoch - 34us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.6516 - 6ms/epoch - 100us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.1951 - 2ms/epoch - 36us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.8813 - 1ms/epoch - 19us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5421 - 851us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.9406 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.1664 - 2ms/epoch - 34us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3250 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.7659 - 1ms/epoch - 18us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.2148 - 16ms/epoch - 253us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.7161 - 1ms/epoch - 16us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5723 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.1542 - 959us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.5876 - 1ms/epoch - 24us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.8886 - 900us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.1219 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4525 - 677us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.0027 - 1ms/epoch - 18us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4761 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 362.6784 - 907us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 790.1742 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.6962 - 766us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.0990 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.8868 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.0327 - 629us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1871 - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6783 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.1466 - 1ms/epoch - 20us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.9391 - 1ms/epoch - 19us/sample\n", - "Generation 146/300\n", - "Solving for Nash Equilibrium in Generation 146/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.6608 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.8762 - 896us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.7101 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5909 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6161 - 918us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.5886 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2845 - 682us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.5206 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.1987 - 800us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.9907 - 761us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4955 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.5844 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.5338 - 6ms/epoch - 103us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0942 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.0984 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.3243 - 758us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.1830 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8192 - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3505 - 3ms/epoch - 41us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4321 - 668us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.6020 - 1ms/epoch - 22us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.1774 - 869us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.7638 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.7498 - 783us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.7228 - 791us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.9210 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.3253 - 4ms/epoch - 67us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.4402 - 740us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0551 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.8766 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.2888 - 809us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9940 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1957 - 615us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.0147 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.3459 - 746us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.9211 - 727us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.8252 - 777us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.7257 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4735 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9628 - 750us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.0634 - 752us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.8694 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.2878 - 688us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.1838 - 723us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.6831 - 643us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 128us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.0745 - 1ms/epoch - 23us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.2903 - 742us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4814 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.6694 - 697us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.1376 - 755us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.4684 - 650us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.4545 - 852us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.1719 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.6404 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.1319 - 805us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.5025 - 791us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8461 - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2343 - 741us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.5549 - 2ms/epoch - 24us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9281 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.4545 - 7ms/epoch - 118us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.6430 - 728us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5873 - 1ms/epoch - 23us/sample\n", - "Generation 147/300\n", - "Solving for Nash Equilibrium in Generation 147/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.5953 - 729us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3906 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.0357 - 1ms/epoch - 18us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4703 - 729us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.7191 - 768us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 145.8951 - 651us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.1450 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 757.9122 - 601us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.2058 - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.2307 - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4406 - 724us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8785 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.2549 - 727us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.8105 - 747us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.5068 - 710us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.9173 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.3655 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5018 - 779us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.5547 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8968 - 686us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.1608 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.4525 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.9203 - 763us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.1391 - 716us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0114 - 648us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.6384 - 794us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.7018 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.7546 - 4ms/epoch - 58us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.3304 - 803us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8551 - 813us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.3983 - 555us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.3900 - 759us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9396 - 772us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1351 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.4313 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5038 - 643us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.0016 - 1ms/epoch - 22us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.8148 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.0274 - 811us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2117 - 921us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.1732 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.6981 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.3705 - 1ms/epoch - 17us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.4577 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.8524 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.4126 - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8522 - 1ms/epoch - 16us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.6776 - 666us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.7656 - 723us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3746 - 832us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6144 - 657us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.2387 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.2039 - 717us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.0242 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.1515 - 8ms/epoch - 122us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9656 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.8832 - 693us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.4461 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.6324 - 644us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.8387 - 763us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8416 - 703us/epoch - 11us/sample\n", - "Generation 148/300\n", - "Solving for Nash Equilibrium in Generation 148/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.2719 - 1ms/epoch - 19us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.5056 - 1ms/epoch - 20us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9978 - 640us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5588 - 581us/epoch - 9us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.5921 - 812us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98.3324 - 606us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.5677 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.2157 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5858 - 755us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.7450 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.8388 - 5ms/epoch - 83us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.7896 - 625us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.9181 - 622us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.4498 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.3088 - 695us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.5252 - 759us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.9517 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7189 - 621us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 198.9570 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.1995 - 796us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.2213 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5748 - 626us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5742 - 637us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.3388 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.2194 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9490 - 679us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1808 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.7578 - 733us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.7894 - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.4047 - 567us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3772 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.1288 - 825us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8361 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.2561 - 632us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.3353 - 667us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.9185 - 2ms/epoch - 38us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.4774 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.7131 - 687us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0528 - 594us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.4067 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.7342 - 712us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 432.9210 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.3199 - 961us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.3543 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.4958 - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.8301 - 785us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.9940 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.5742 - 773us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.0320 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.2238 - 691us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.9925 - 2ms/epoch - 25us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.1788 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9443 - 710us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0171 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.1053 - 2ms/epoch - 25us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.3140 - 993us/epoch - 16us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.7658 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.4122 - 778us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.1083 - 738us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.7318 - 1ms/epoch - 23us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.9682 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5426 - 645us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.1892 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.8000 - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.8666 - 620us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.4587 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.6382 - 742us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.1639 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.1621 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.6385 - 617us/epoch - 10us/sample\n", - "Generation 149/300\n", - "Solving for Nash Equilibrium in Generation 149/300\n", + "Generation 35/300\n", + "Solving for Nash Equilibrium in Generation 35/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6176 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.4128 - 849us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.8178 - 638us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.4238 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.3310 - 690us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.1285 - 755us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.4038 - 2ms/epoch - 35us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.2466 - 751us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3129 - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.7547 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.8245 - 1ms/epoch - 21us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9519 - 710us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.4081 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.7345 - 643us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.3257 - 700us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.1757 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.0847 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.8604 - 1ms/epoch - 17us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.4737 - 726us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.7443 - 760us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.4509 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.1142 - 764us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.2249 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2161 - 596us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.1456 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.8265 - 843us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.7147 - 856us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.8063 - 780us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8500 - 622us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.5719 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.3068 - 793us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.0410 - 11ms/epoch - 174us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.2854 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.0882 - 633us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.4362 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.0578 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.7403 - 772us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.3597 - 792us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.1273 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.9766 - 685us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.2581 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.1262 - 946us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.3606 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.0513 - 696us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.6497 - 712us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9669 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.0607 - 708us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.2701 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.8583 - 658us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6102 - 2ms/epoch - 33us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.9996 - 757us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2766 - 748us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.2619 - 2ms/epoch - 25us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.6660 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.4466 - 744us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.8414 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.2546 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.3785 - 747us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7822 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.3375 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.6981 - 753us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.9686 - 756us/epoch - 12us/sample\n", - "Generation 150/300\n", - "Solving for Nash Equilibrium in Generation 150/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.4954 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1638 - 658us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.8264 - 758us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.7472 - 802us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.2597 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.7169 - 873us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.5883 - 626us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.5194 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.7363 - 1ms/epoch - 19us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.4684 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5104 - 590us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.4457 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.3318 - 730us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.0297 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.1426 - 3ms/epoch - 46us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.2147 - 660us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.2796 - 648us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.1308 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.2921 - 756us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5547 - 616us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7822 - 2ms/epoch - 31us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.4260 - 688us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.0652 - 3ms/epoch - 48us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0986 - 831us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.4711 - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.2032 - 798us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.2689 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.5304 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.1892 - 860us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.4878 - 863us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.2024 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7072 - 761us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.6246 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.7151 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.3664 - 836us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0017 - 708us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.5917 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.7393 - 1ms/epoch - 19us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.6747 - 771us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5829 - 742us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.3893 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.2271 - 1ms/epoch - 18us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8085 - 716us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.6481 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.4601 - 681us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.0577 - 810us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.6450 - 668us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9787 - 5ms/epoch - 78us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.0573 - 587us/epoch - 9us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54.0342 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.2797 - 764us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 115us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.6386 - 5ms/epoch - 74us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3766 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.2454 - 730us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0687 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.2158 - 885us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3599 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.0015 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.4265 - 817us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.3072 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.6517 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.1353 - 726us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1620 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.4600 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.7435 - 746us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4654 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6031 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.6487 - 833us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.0661 - 781us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787.2401 - 876us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.0980 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.4136 - 743us/epoch - 12us/sample\n", - "Generation 151/300\n", - "Solving for Nash Equilibrium in Generation 151/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 36/300\n", + "Solving for Nash Equilibrium in Generation 36/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3680 - 615us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.6124 - 781us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.0346 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.7800 - 761us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.0867 - 886us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.2856 - 5ms/epoch - 85us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.2867 - 708us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.6943 - 976us/epoch - 16us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.3166 - 774us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.5040 - 2ms/epoch - 36us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6844 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.1954 - 842us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2880 - 723us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.6410 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.5528 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0761 - 723us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.6213 - 715us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1534 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.1497 - 1ms/epoch - 21us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 770.3696 - 729us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5245 - 675us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.8977 - 1ms/epoch - 18us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.6832 - 745us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.3879 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.1807 - 683us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.4240 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4603 - 893us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.2155 - 651us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.8395 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.3014 - 770us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4750 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.3971 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1349 - 625us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.5203 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.7534 - 1ms/epoch - 22us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5002 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.5008 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.0686 - 6ms/epoch - 100us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.9070 - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.1658 - 705us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.6493 - 2ms/epoch - 30us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.9294 - 681us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1249 - 633us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.2053 - 684us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.4773 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.2346 - 659us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.4705 - 2ms/epoch - 33us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.1630 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.2629 - 629us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.4983 - 604us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.8865 - 1ms/epoch - 17us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5179 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.5686 - 651us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.3919 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2698 - 719us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8033 - 605us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.4056 - 670us/epoch - 11us/sample\n", - "Generation 152/300\n", - "Solving for Nash Equilibrium in Generation 152/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.8931 - 738us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.7615 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.2084 - 730us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6635 - 717us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7584 - 694us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.3051 - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6495 - 590us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.1809 - 699us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40.3186 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.3522 - 851us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.9933 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.0668 - 660us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.2263 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.9568 - 790us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.0676 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2025 - 964us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.2867 - 973us/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.7449 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.9088 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.2809 - 719us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1052 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2626 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.7878 - 681us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2932 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3366 - 663us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.2162 - 623us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 104us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8146 - 1ms/epoch - 22us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3419 - 907us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.8421 - 845us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6268 - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9041 - 623us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.0905 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.3214 - 1ms/epoch - 18us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.9988 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.2497 - 795us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.0417 - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.1027 - 769us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9446 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7307 - 744us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.4785 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.2148 - 634us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.9236 - 6ms/epoch - 91us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.6432 - 914us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.5350 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.4381 - 895us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 159.6295 - 818us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.1436 - 855us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.0268 - 803us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.7961 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.5348 - 712us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.2342 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6144 - 3ms/epoch - 48us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2354 - 732us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.7648 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.9748 - 942us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.8885 - 884us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.1439 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.5566 - 678us/epoch - 11us/sample\n", - "Generation 153/300\n", - "Solving for Nash Equilibrium in Generation 153/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0454 - 805us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.4059 - 913us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.5360 - 788us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.1341 - 684us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.4997 - 792us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1238 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6301 - 778us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.2586 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4327 - 759us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.8246 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.3759 - 887us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0786 - 864us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6887 - 977us/epoch - 16us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.1715 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.5664 - 1ms/epoch - 19us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.7753 - 989us/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 233.6729 - 843us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8045 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.5599 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.9176 - 743us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.5612 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.7980 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.9986 - 701us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.0044 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.4767 - 594us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.8387 - 774us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.7071 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.4543 - 708us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.9377 - 966us/epoch - 16us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.1329 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.9938 - 705us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.0510 - 786us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.5728 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.5678 - 736us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.8670 - 825us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.4178 - 764us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.3815 - 675us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.3972 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.9662 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.2717 - 574us/epoch - 9us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.1341 - 1ms/epoch - 16us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.1007 - 2ms/epoch - 29us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.3356 - 1ms/epoch - 19us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.9045 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.6403 - 764us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.5861 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4249 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.7036 - 1ms/epoch - 22us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2881 - 640us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.0484 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.8777 - 856us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.5983 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.4512 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.5610 - 1ms/epoch - 18us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.5163 - 779us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.7762 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.8592 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.7321 - 648us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.8679 - 785us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.0007 - 728us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.2269 - 688us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.1871 - 679us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.7913 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.2329 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4945 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.5445 - 746us/epoch - 12us/sample\n", - "Generation 154/300\n", - "Solving for Nash Equilibrium in Generation 154/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.3966 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.0601 - 642us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.0211 - 1ms/epoch - 19us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4413 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.3915 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2964 - 641us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.3145 - 740us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.8528 - 728us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.7472 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361.4197 - 649us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.0110 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8126 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.0092 - 588us/epoch - 9us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.9153 - 803us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.3794 - 838us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.5131 - 918us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406.4429 - 985us/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.8569 - 801us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5938 - 770us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.0436 - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.6880 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.3741 - 760us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.4618 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2870 - 615us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.3689 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.1292 - 912us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.6513 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.6461 - 646us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.9610 - 917us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.2252 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9673 - 556us/epoch - 9us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.1043 - 1ms/epoch - 22us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.8084 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.5676 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1503 - 770us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.7024 - 591us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.9508 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.6492 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2131 - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.0234 - 807us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.7879 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.5571 - 741us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.4274 - 676us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.5895 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.0881 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8174 - 695us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.7541 - 667us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.8441 - 670us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.1362 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.0999 - 1ms/epoch - 16us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.3759 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.7589 - 694us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43.2621 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.9405 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5782 - 832us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.7841 - 690us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.0451 - 669us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.2950 - 661us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.4664 - 669us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.5468 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.1955 - 715us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 125.5406 - 791us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.6570 - 644us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4122 - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.4329 - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5411 - 745us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6389 - 599us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.4008 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.0961 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.1065 - 667us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4858 - 658us/epoch - 11us/sample\n", - "Generation 155/300\n", - "Solving for Nash Equilibrium in Generation 155/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.9146 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.4127 - 733us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.4125 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.3173 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0581 - 810us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.2773 - 969us/epoch - 16us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.1759 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.5747 - 691us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.5920 - 3ms/epoch - 47us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.9407 - 810us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.6650 - 1ms/epoch - 17us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40.3413 - 774us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3986 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.7953 - 1ms/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.5438 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.7704 - 652us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9181 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4171 - 656us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7748 - 789us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.5436 - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.8195 - 697us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.7664 - 860us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4247 - 660us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.7839 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.8534 - 804us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.3817 - 741us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7344 - 990us/epoch - 16us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.8222 - 934us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.0988 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.6030 - 877us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.7297 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.5954 - 857us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.7382 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.8562 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1531 - 664us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787.2596 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.8551 - 748us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.2503 - 600us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.7242 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.3212 - 863us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.3003 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.2301 - 956us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2914 - 708us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.3668 - 938us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.0483 - 921us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.8002 - 2ms/epoch - 25us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2628 - 937us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.9036 - 794us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.6046 - 716us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.6749 - 842us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.6656 - 599us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2554 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112.3322 - 838us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.9015 - 680us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1486 - 2ms/epoch - 37us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.3163 - 714us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5770 - 676us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 156/300\n", - "Solving for Nash Equilibrium in Generation 156/300\n", + "Generation 37/300\n", + "Solving for Nash Equilibrium in Generation 37/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.0909 - 622us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9395 - 790us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.2136 - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 228.7115 - 721us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45.9925 - 638us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3793 - 809us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.1908 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232.8526 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8510 - 1ms/epoch - 21us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.9467 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.1950 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2911 - 696us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.2692 - 704us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.1880 - 689us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.5951 - 964us/epoch - 16us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.3101 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.8691 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7356 - 766us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.3910 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.9588 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1470 - 727us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 162.7045 - 1ms/epoch - 24us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.9755 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4777 - 678us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.3483 - 2ms/epoch - 40us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.6394 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.1757 - 5ms/epoch - 73us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.2585 - 916us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4411 - 720us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5049 - 733us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.6098 - 677us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.5818 - 672us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.7139 - 568us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.9597 - 646us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7785 - 803us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.7815 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.1760 - 672us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.0594 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.6263 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.5192 - 711us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9138 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2321 - 671us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 145us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.9452 - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2323 - 709us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9210 - 779us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.3576 - 8ms/epoch - 130us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.5067 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.1620 - 672us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.9724 - 1ms/epoch - 19us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2610 - 609us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.9910 - 3ms/epoch - 47us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.4441 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.8520 - 2ms/epoch - 29us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.2785 - 609us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.4110 - 836us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.5558 - 656us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.5674 - 13ms/epoch - 203us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.5487 - 2ms/epoch - 34us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3192 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.4686 - 660us/epoch - 11us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 157/300\n", - "Solving for Nash Equilibrium in Generation 157/300\n", + "Generation 38/300\n", + "Solving for Nash Equilibrium in Generation 38/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.9247 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.9869 - 2ms/epoch - 27us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47.8381 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.5892 - 651us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.8510 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.1556 - 1ms/epoch - 17us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.8672 - 645us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2128 - 691us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.5219 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.4634 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.8491 - 695us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.1640 - 801us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.9781 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8755 - 754us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.1842 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.4252 - 805us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.8507 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.9966 - 796us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.9036 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.3412 - 2ms/epoch - 27us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.9542 - 777us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.3293 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.1216 - 801us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.9067 - 748us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.3565 - 710us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3056 - 2ms/epoch - 35us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.0190 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.0199 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.7504 - 2ms/epoch - 39us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.1896 - 757us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.3310 - 884us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.2119 - 771us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.8718 - 5ms/epoch - 74us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.2072 - 746us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.5848 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7660 - 700us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.2587 - 795us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0424 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.7131 - 598us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.1979 - 626us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.2319 - 869us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1207 - 5ms/epoch - 76us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.9504 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5626 - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9365 - 694us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.5867 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3106 - 598us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.6470 - 606us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.9769 - 709us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.2136 - 594us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9326 - 765us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.6428 - 588us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.3522 - 576us/epoch - 9us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.1339 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5753 - 902us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.4605 - 829us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.7875 - 697us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.8089 - 1ms/epoch - 19us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.8527 - 742us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.6617 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0980 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.4800 - 738us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.1107 - 569us/epoch - 9us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.2545 - 3ms/epoch - 43us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9035 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.9134 - 746us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.6756 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.6994 - 894us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.8494 - 698us/epoch - 11us/sample\n", - "Generation 158/300\n", - "Solving for Nash Equilibrium in Generation 158/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.1603 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.1174 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8503 - 808us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.5385 - 604us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.9287 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.7219 - 9ms/epoch - 146us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.6071 - 609us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.9636 - 836us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37.5029 - 753us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.3983 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3477 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2020 - 617us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.8012 - 2ms/epoch - 37us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.0078 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.3008 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8948 - 683us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.3734 - 839us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6174 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.5619 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.0389 - 736us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.0885 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.8775 - 736us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.0018 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4783 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7620 - 684us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3464 - 845us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8294 - 642us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.9333 - 693us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9028 - 794us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.4955 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.5611 - 614us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.9291 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.3531 - 695us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0351 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8201 - 690us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.5533 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.6037 - 782us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.6944 - 683us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 772.3726 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.3404 - 706us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.3078 - 643us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.5110 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.7308 - 2ms/epoch - 38us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.3401 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.0662 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.9634 - 677us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.4475 - 660us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.4151 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1244 - 7ms/epoch - 112us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.4161 - 577us/epoch - 9us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.6862 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.2615 - 769us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.7479 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.4367 - 720us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 785.0875 - 691us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.8801 - 836us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.8010 - 873us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.6366 - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.7113 - 871us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1599 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.7281 - 2ms/epoch - 30us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30.9359 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3544 - 1ms/epoch - 21us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.0849 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 139us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8256 - 865us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.0561 - 1ms/epoch - 22us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 159/300\n", - "Solving for Nash Equilibrium in Generation 159/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787.9453 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.7502 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.8594 - 630us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.4952 - 706us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.6937 - 941us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.3813 - 906us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.6189 - 971us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.8773 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.1545 - 2ms/epoch - 37us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.5678 - 754us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.1616 - 700us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.5792 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.1536 - 795us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.3632 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.1062 - 909us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.1766 - 632us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.0509 - 632us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749.0114 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.7658 - 642us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.0673 - 705us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.3545 - 715us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4747 - 604us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.3871 - 742us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.3056 - 654us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.9081 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.3040 - 756us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.4582 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.1302 - 655us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.4575 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.3671 - 971us/epoch - 16us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.2160 - 683us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.5661 - 641us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7700 - 5ms/epoch - 84us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.0833 - 675us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.4783 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.9932 - 1ms/epoch - 17us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.1048 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 130us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.0758 - 3ms/epoch - 51us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.0374 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.3234 - 988us/epoch - 16us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.7684 - 7ms/epoch - 115us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.7818 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.4536 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.6069 - 2ms/epoch - 29us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.1373 - 835us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.8477 - 679us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5269 - 840us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.5535 - 638us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.2398 - 1ms/epoch - 18us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.4532 - 5ms/epoch - 79us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.7507 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.7224 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.5804 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.0127 - 648us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.6071 - 638us/epoch - 10us/sample\n", - "Generation 160/300\n", - "Solving for Nash Equilibrium in Generation 160/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.7263 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.5324 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.3362 - 4ms/epoch - 67us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8389 - 684us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.6490 - 738us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.8005 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.4948 - 621us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.5291 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.0321 - 536us/epoch - 9us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.0021 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6397 - 632us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.9123 - 2ms/epoch - 38us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3622 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.4623 - 667us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.5567 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3109 - 2ms/epoch - 37us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4864 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.9138 - 642us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.2880 - 2ms/epoch - 33us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.3542 - 807us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.2380 - 679us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.8947 - 4ms/epoch - 65us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6037 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9777 - 735us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8406 - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.8270 - 632us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.1373 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.3560 - 917us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43.0280 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.2302 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.6502 - 789us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4538 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.1660 - 621us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.1084 - 701us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5090 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.4954 - 676us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.4797 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9575 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.0344 - 582us/epoch - 9us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.1003 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3960 - 623us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.2043 - 3ms/epoch - 44us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65.8795 - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.9186 - 679us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3146 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5081 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.1646 - 584us/epoch - 9us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7800 - 4ms/epoch - 64us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.3822 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.7063 - 676us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.9744 - 567us/epoch - 9us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8009 - 605us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.1378 - 683us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.9714 - 701us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.5960 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8845 - 663us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.1038 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 12ms/epoch - 188us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.3117 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.8664 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.9884 - 2ms/epoch - 26us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.9928 - 743us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.0906 - 694us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.8516 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.6974 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1550 - 658us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784.9841 - 585us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.8575 - 826us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.9909 - 705us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.9728 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.8674 - 2ms/epoch - 38us/sample\n", - "Generation 161/300\n", - "Solving for Nash Equilibrium in Generation 161/300\n", + "Generation 39/300\n", + "Solving for Nash Equilibrium in Generation 39/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.5947 - 5ms/epoch - 80us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.3342 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.2039 - 741us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9705 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.8189 - 640us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.1234 - 629us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9875 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.8950 - 706us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.6541 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.3939 - 672us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.0596 - 876us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.4326 - 594us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.5615 - 794us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.6909 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3427 - 690us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.9603 - 742us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6873 - 593us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5794 - 745us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9315 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.5539 - 1ms/epoch - 22us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.7149 - 658us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248.6650 - 786us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.1041 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.3483 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.2847 - 3ms/epoch - 42us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.5798 - 3ms/epoch - 56us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5847 - 805us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.0370 - 631us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7738 - 942us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 220.3857 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.3681 - 25ms/epoch - 403us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.3555 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.9926 - 4ms/epoch - 69us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.2173 - 718us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7198 - 613us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.8752 - 637us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.8976 - 618us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.0818 - 612us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.0726 - 10ms/epoch - 169us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.1702 - 652us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7661 - 896us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9777 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.3365 - 684us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.2791 - 2ms/epoch - 25us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.9369 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9484 - 797us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.9774 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6282 - 617us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.0225 - 819us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.6317 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6818 - 2ms/epoch - 26us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 162/300\n", - "Solving for Nash Equilibrium in Generation 162/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.7555 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 435.3481 - 688us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3829 - 790us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3514 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.5815 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.8240 - 721us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6701 - 738us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.6993 - 689us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.1746 - 1ms/epoch - 18us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.3191 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.6053 - 713us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.5861 - 788us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.2204 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.8822 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.8268 - 728us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.0168 - 684us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.0278 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.7887 - 791us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.5972 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.6016 - 5ms/epoch - 78us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.4544 - 770us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5943 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.8655 - 715us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.5894 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.7461 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.5240 - 614us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.6750 - 707us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.9357 - 835us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.5931 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.7475 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2155 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.3714 - 759us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.3757 - 7ms/epoch - 112us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.8839 - 721us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.6852 - 1ms/epoch - 19us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.2433 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.9268 - 1ms/epoch - 18us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.4926 - 688us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 133us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.7987 - 1ms/epoch - 17us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.4161 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.5327 - 4ms/epoch - 59us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.3711 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.7198 - 956us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.6571 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.7637 - 854us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5182 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.3344 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.3773 - 620us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4124 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4232 - 558us/epoch - 9us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.2469 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.1297 - 773us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.0276 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.0878 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7364 - 730us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6391 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.9508 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.6982 - 819us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8351 - 694us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8657 - 714us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.9781 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3920 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.8071 - 750us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.5870 - 756us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.2815 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4073 - 767us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.6893 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.0230 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.3195 - 1ms/epoch - 21us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.3651 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.0082 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.3864 - 650us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.2826 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.2224 - 841us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.4948 - 3ms/epoch - 49us/sample\n", - "Generation 163/300\n", - "Solving for Nash Equilibrium in Generation 163/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.8997 - 907us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.6838 - 595us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.3619 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.4225 - 649us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.0614 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.5601 - 956us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7246 - 685us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.6946 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6262 - 627us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.6619 - 690us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.2537 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.4238 - 754us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.8942 - 639us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.6780 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.0582 - 673us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0963 - 724us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 130.7081 - 928us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.6713 - 807us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.6305 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.0016 - 667us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.5352 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.3533 - 897us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.8284 - 794us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.9664 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8383 - 742us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.4244 - 883us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.8159 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.8981 - 785us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.2531 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.1769 - 631us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.3610 - 737us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.8073 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.8532 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.5278 - 1ms/epoch - 16us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.1805 - 5ms/epoch - 77us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.6223 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1265 - 2ms/epoch - 40us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0031 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7115 - 713us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6899 - 771us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.7697 - 2ms/epoch - 30us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3922 - 2ms/epoch - 39us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.9301 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.7352 - 757us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.8832 - 797us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.9171 - 589us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6256 - 719us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48.4967 - 839us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.8204 - 1ms/epoch - 18us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.1733 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.7823 - 711us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2941 - 716us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.8793 - 767us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3019 - 694us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.0369 - 6ms/epoch - 101us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.2529 - 35ms/epoch - 563us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.1821 - 786us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.1243 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.5235 - 2ms/epoch - 33us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.4324 - 1ms/epoch - 16us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7763 - 824us/epoch - 13us/sample\n", - "Generation 164/300\n", - "Solving for Nash Equilibrium in Generation 164/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36.4285 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.0956 - 994us/epoch - 16us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.5877 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.3677 - 743us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.3677 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.6361 - 710us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.2153 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.4849 - 903us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6747 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0444 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.0619 - 673us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1605 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.9755 - 829us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9606 - 667us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.6465 - 902us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.4091 - 860us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6190 - 629us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5577 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49.0184 - 769us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.1208 - 725us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.8879 - 573us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9203 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.6599 - 8ms/epoch - 134us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.0901 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.6346 - 651us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9031 - 900us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.4642 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2072 - 839us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.5931 - 719us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.1008 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 148.4852 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.7935 - 753us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3533 - 1ms/epoch - 18us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8552 - 785us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.3702 - 7ms/epoch - 109us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.4913 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5194 - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9315 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.0896 - 2ms/epoch - 30us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.4596 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.6666 - 873us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.4394 - 777us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.8348 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.3685 - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.8414 - 849us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5114 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9077 - 2ms/epoch - 38us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3606 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.6302 - 792us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.5623 - 851us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.1393 - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.3921 - 834us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.6474 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.8621 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.9090 - 872us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.4915 - 844us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.9012 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.0555 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0665 - 2ms/epoch - 38us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.8987 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.6288 - 1ms/epoch - 17us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.2206 - 2ms/epoch - 29us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.6219 - 824us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0272 - 3ms/epoch - 42us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.5493 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.5325 - 727us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.7380 - 1ms/epoch - 17us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.7552 - 1ms/epoch - 19us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.8116 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.8294 - 866us/epoch - 14us/sample\n", - "Generation 165/300\n", - "Solving for Nash Equilibrium in Generation 165/300\n", + "Generation 40/300\n", + "Solving for Nash Equilibrium in Generation 40/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6505 - 822us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.2346 - 859us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2770 - 768us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.2600 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.3405 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0089 - 776us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.4822 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.2017 - 750us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.1710 - 690us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.3056 - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9753 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 17ms/epoch - 272us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6236 - 7ms/epoch - 111us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.8872 - 700us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.4994 - 1ms/epoch - 18us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.9639 - 2ms/epoch - 28us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.8549 - 841us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.4451 - 865us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.8946 - 672us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.4611 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2650 - 703us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.5420 - 749us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0035 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8877 - 2ms/epoch - 30us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.8351 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2094 - 841us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.6316 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9521 - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.6305 - 20ms/epoch - 324us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.1299 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.6616 - 695us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4144 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8660 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.0609 - 3ms/epoch - 48us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.8331 - 1ms/epoch - 18us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.0352 - 822us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.0219 - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.8022 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.1925 - 1ms/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.8323 - 859us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8497 - 776us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.9391 - 701us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.7214 - 1ms/epoch - 21us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3449 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.2263 - 3ms/epoch - 56us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.3097 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6600 - 726us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2016 - 800us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.2331 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6221 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1503 - 661us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.9664 - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0651 - 1ms/epoch - 19us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47.3913 - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.0397 - 4ms/epoch - 57us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.4613 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.3037 - 586us/epoch - 9us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.9454 - 885us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.7788 - 684us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.2874 - 663us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.5072 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.4757 - 785us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.6863 - 672us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755.9975 - 3ms/epoch - 56us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 166/300\n", - "Solving for Nash Equilibrium in Generation 166/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 233.2900 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.2348 - 549us/epoch - 9us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4190 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.1250 - 699us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6702 - 1ms/epoch - 24us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.2246 - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4431 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.4048 - 726us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.6623 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4256 - 640us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.3226 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.6152 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.6090 - 684us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.0837 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.0971 - 717us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2039 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.9293 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0917 - 1ms/epoch - 21us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.1152 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.7307 - 832us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3460 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.5759 - 20ms/epoch - 318us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.7596 - 665us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.3811 - 1ms/epoch - 17us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.1487 - 667us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.4180 - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.5580 - 4ms/epoch - 58us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.8875 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3903 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.1652 - 2ms/epoch - 26us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.7101 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1088 - 775us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.1346 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.1109 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.2484 - 2ms/epoch - 25us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.2805 - 4ms/epoch - 70us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.7875 - 699us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.4019 - 1ms/epoch - 20us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.2255 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.2592 - 851us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.5455 - 649us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.5902 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.3056 - 902us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.9433 - 888us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8049 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.3553 - 686us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.6244 - 855us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0072 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.3735 - 13ms/epoch - 206us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.2738 - 1ms/epoch - 24us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.0123 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.1770 - 2ms/epoch - 40us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 73.2788 - 3ms/epoch - 41us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.0856 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.7152 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 165us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.1910 - 1ms/epoch - 18us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.8156 - 2ms/epoch - 34us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8458 - 739us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406.0654 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.7159 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7889 - 4ms/epoch - 65us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.4019 - 793us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8372 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.9169 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 125us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.3900 - 2ms/epoch - 37us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.6838 - 876us/epoch - 14us/sample\n", - "Generation 167/300\n", - "Solving for Nash Equilibrium in Generation 167/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.5646 - 841us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.5598 - 6ms/epoch - 92us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.4137 - 782us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.4740 - 689us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.1337 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 362.6317 - 635us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3084 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.7631 - 804us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 125us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.7793 - 929us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.3014 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5945 - 2ms/epoch - 25us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.4426 - 918us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 181.6683 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.9283 - 1ms/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.7838 - 739us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.7072 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.2500 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.3353 - 659us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.3224 - 627us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.6746 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9189 - 829us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.5493 - 670us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6965 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.2034 - 643us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.6323 - 639us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0298 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.7758 - 654us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.2671 - 762us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9477 - 611us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.3553 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.7411 - 759us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.7099 - 1ms/epoch - 19us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.0724 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.4560 - 2ms/epoch - 24us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1351 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.5165 - 678us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.1568 - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.3441 - 837us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.5876 - 616us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.6959 - 1ms/epoch - 19us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787.2444 - 584us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.2872 - 733us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.8008 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.5660 - 604us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.9689 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8707 - 641us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.7073 - 698us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.7810 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.9179 - 798us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.6527 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.7834 - 1ms/epoch - 19us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.3806 - 975us/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.2970 - 760us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.3124 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6109 - 625us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.3564 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.1608 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.6605 - 881us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.3438 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.7218 - 595us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.0013 - 13ms/epoch - 208us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.8684 - 821us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.5255 - 746us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.6688 - 938us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8976 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1622 - 8ms/epoch - 123us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.4894 - 974us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6412 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.5098 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.7421 - 1ms/epoch - 20us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.5463 - 1ms/epoch - 20us/sample\n", - "Generation 168/300\n", - "Solving for Nash Equilibrium in Generation 168/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.1918 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.0211 - 902us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.0181 - 1ms/epoch - 19us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.9522 - 765us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.8220 - 810us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.6909 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.6823 - 715us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.7793 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.0154 - 640us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.0856 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.6515 - 802us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.0780 - 813us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 782.9211 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.7077 - 669us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.6896 - 807us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.3502 - 715us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.6883 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6561 - 732us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.1677 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.6389 - 1ms/epoch - 21us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.4035 - 754us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.6203 - 621us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.7598 - 702us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1389 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.8383 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.6990 - 689us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.1633 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7850 - 637us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.6961 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.7104 - 623us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2620 - 611us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.7277 - 831us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6180 - 833us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3907 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.7439 - 619us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.5559 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 225.3622 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.7338 - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.8555 - 669us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2457 - 668us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.4229 - 783us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.4876 - 774us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.7892 - 751us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.7654 - 708us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.6682 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.6432 - 892us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.9272 - 2ms/epoch - 26us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1281 - 699us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.2024 - 2ms/epoch - 32us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.0822 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.3895 - 1ms/epoch - 23us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.6799 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.8502 - 917us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.8599 - 623us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250.6322 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.8801 - 782us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.9943 - 919us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.5786 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.9308 - 577us/epoch - 9us/sample\n", - "Generation 169/300\n", - "Solving for Nash Equilibrium in Generation 169/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 14ms/epoch - 227us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.0345 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.6922 - 897us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.1196 - 2ms/epoch - 27us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7361 - 814us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.5696 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.4773 - 642us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.1626 - 3ms/epoch - 41us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.2613 - 621us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0681 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.8759 - 718us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 792.0992 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.3552 - 823us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.5089 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.4370 - 900us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.4197 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.6336 - 802us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.9283 - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.4846 - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.3509 - 653us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6938 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2580 - 644us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.7021 - 922us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.2196 - 791us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.7972 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.1484 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.5488 - 1ms/epoch - 18us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.6478 - 829us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9297 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1793 - 622us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.8832 - 608us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.5172 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.0121 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.1234 - 784us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.5645 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.4319 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.2884 - 669us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8081 - 702us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.9144 - 720us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.9040 - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.5201 - 1ms/epoch - 19us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.5889 - 3ms/epoch - 43us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.5784 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.9679 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.7667 - 649us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.4209 - 851us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.6587 - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6298 - 4ms/epoch - 65us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6297 - 835us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.9921 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.5126 - 1ms/epoch - 18us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.0963 - 771us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.0396 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.1741 - 681us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.2022 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.6708 - 1ms/epoch - 17us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.3112 - 619us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.5688 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.7629 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6285 - 2ms/epoch - 25us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.9294 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.5035 - 736us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.1745 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.4524 - 844us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.8131 - 1ms/epoch - 20us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.7819 - 606us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.6847 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.9054 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.8882 - 2ms/epoch - 28us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.9210 - 681us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1059 - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.9684 - 743us/epoch - 12us/sample\n", - "Generation 170/300\n", - "Solving for Nash Equilibrium in Generation 170/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.1469 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3491 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.4454 - 2ms/epoch - 25us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.2145 - 1ms/epoch - 19us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.0455 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.6705 - 865us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.5807 - 885us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.6174 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.4870 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.5527 - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2461 - 697us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.2982 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.8369 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1282 - 9ms/epoch - 138us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.7677 - 1ms/epoch - 17us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.6946 - 631us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5272 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6151 - 564us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.9121 - 769us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.3009 - 613us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8574 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 165.7563 - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.3118 - 717us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.2889 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.5350 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.3510 - 595us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.2874 - 610us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.5397 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.2343 - 806us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.3812 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.3785 - 713us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3742 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.2065 - 720us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.4837 - 866us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.3498 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.0896 - 747us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.5693 - 3ms/epoch - 51us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7157 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.2873 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.0343 - 765us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7759 - 621us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.6460 - 806us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.3638 - 770us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.6482 - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.7122 - 710us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.1128 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.7216 - 568us/epoch - 9us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.6465 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.1430 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.3020 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.8033 - 689us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.3755 - 646us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.9454 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.4212 - 1ms/epoch - 23us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.3374 - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6810 - 760us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.1735 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.7424 - 615us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3331 - 764us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35.5628 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2100 - 734us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.6870 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.4629 - 3ms/epoch - 51us/sample\n", - "Episode 50/50\n", - "Generation 171/300\n", - "Solving for Nash Equilibrium in Generation 171/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.8663 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6427 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.4648 - 781us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.9530 - 645us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.3278 - 585us/epoch - 9us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8643 - 742us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.3177 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.7785 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2883 - 652us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.2581 - 733us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3682 - 716us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.3716 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.5548 - 734us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.1682 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.0385 - 835us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.2087 - 806us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.1307 - 2ms/epoch - 32us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.2114 - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.0060 - 961us/epoch - 16us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.2385 - 784us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.9512 - 3ms/epoch - 44us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.3625 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.8826 - 1ms/epoch - 22us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.6402 - 654us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.9130 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.5570 - 2ms/epoch - 32us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.9730 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.2502 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.7225 - 597us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.5201 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.1042 - 751us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.2695 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7228 - 601us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.1051 - 747us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.7521 - 1ms/epoch - 20us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.2090 - 684us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1326 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7163 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3989 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.2216 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.8348 - 683us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.7254 - 779us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.1466 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.1669 - 635us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.3698 - 643us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4769 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.2077 - 651us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.0551 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.6671 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.9828 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9159 - 7ms/epoch - 112us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4121 - 698us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9020 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.3592 - 954us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.8373 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3333 - 695us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.8118 - 2ms/epoch - 34us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.1534 - 1ms/epoch - 20us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.3812 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.6547 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.8163 - 834us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.9628 - 749us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.7546 - 918us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4901 - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245.5936 - 1ms/epoch - 23us/sample\n", - "Generation 172/300\n", - "Solving for Nash Equilibrium in Generation 172/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.7248 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.9143 - 666us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7731 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.1509 - 847us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.7634 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.7405 - 774us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2259 - 753us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.9543 - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.3721 - 669us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.7907 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.1556 - 750us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.8175 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.0771 - 1ms/epoch - 17us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.8401 - 2ms/epoch - 35us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.3712 - 760us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6242 - 756us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.0166 - 6ms/epoch - 98us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.7980 - 669us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1529 - 1ms/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.0443 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.3752 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0954 - 716us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.9207 - 1ms/epoch - 17us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.1992 - 689us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3348 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.7523 - 3ms/epoch - 44us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.2155 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.2826 - 829us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.7003 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.4972 - 636us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.7365 - 668us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7768 - 954us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.6410 - 843us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.5795 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.4027 - 833us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.6062 - 682us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.9025 - 767us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.3429 - 781us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41.4946 - 819us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8521 - 709us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.7229 - 765us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.4684 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.0820 - 590us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 12ms/epoch - 194us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.3652 - 1ms/epoch - 18us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38.9872 - 654us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.2112 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.3961 - 871us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.9707 - 698us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.5479 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240.7143 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.4266 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2120 - 642us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.7867 - 730us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.2865 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3503 - 916us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.6257 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.0397 - 608us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.5801 - 4ms/epoch - 68us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.3136 - 842us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.7004 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.9037 - 710us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.3256 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.8614 - 723us/epoch - 12us/sample\n", - "Generation 173/300\n", - "Solving for Nash Equilibrium in Generation 173/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.5027 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.4538 - 3ms/epoch - 49us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.4194 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.0739 - 674us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.8592 - 870us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.0533 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.9102 - 529us/epoch - 9us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.8281 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.6771 - 3ms/epoch - 49us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.2456 - 709us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.6122 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4118 - 853us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7531 - 716us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.0674 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.3808 - 3ms/epoch - 46us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.0324 - 730us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.0047 - 4ms/epoch - 64us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0762 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 141us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.4617 - 4ms/epoch - 61us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.0233 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.1772 - 1ms/epoch - 21us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.3929 - 4ms/epoch - 67us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8774 - 975us/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.3944 - 689us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.9927 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.8418 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5368 - 950us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.6425 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2565 - 796us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.1058 - 960us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.2931 - 818us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.5973 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.7754 - 838us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5446 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.0632 - 749us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.1344 - 698us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.7192 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.4861 - 824us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.8886 - 800us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.7562 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1332 - 873us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.6839 - 2ms/epoch - 29us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6002 - 722us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.2257 - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2753 - 764us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.2224 - 803us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.7207 - 813us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45.7698 - 7ms/epoch - 113us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.0276 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.9498 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.5684 - 787us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.2584 - 711us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.7083 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.4358 - 3ms/epoch - 46us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.4165 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.0965 - 720us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.3906 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2268 - 2ms/epoch - 38us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.1987 - 620us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.1332 - 907us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.9791 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 100.4016 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.1229 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.4318 - 699us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.6569 - 635us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.4743 - 731us/epoch - 12us/sample\n", - "Generation 174/300\n", - "Solving for Nash Equilibrium in Generation 174/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.5834 - 923us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.9315 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.5096 - 881us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.8788 - 2ms/epoch - 38us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0103 - 718us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0677 - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.6864 - 916us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.3556 - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.2644 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.1348 - 2ms/epoch - 34us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.6255 - 673us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.1849 - 738us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0104 - 948us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.6333 - 564us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6234 - 661us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.0148 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.3657 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.0915 - 796us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1638 - 772us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.5180 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.2067 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6812 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.9169 - 712us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.3005 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.8407 - 709us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5078 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6698 - 1ms/epoch - 18us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.5604 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9427 - 716us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.2150 - 848us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.9620 - 662us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.1225 - 640us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.3302 - 578us/epoch - 9us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38.3670 - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.6187 - 789us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248.3686 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.0949 - 635us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.7338 - 765us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.6890 - 3ms/epoch - 54us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.2914 - 708us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.9088 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.0176 - 1ms/epoch - 19us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2574 - 749us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.8599 - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.7346 - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.9740 - 776us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6469 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.4437 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4611 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2999 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.9951 - 1ms/epoch - 18us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9775 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.9416 - 728us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.9310 - 6ms/epoch - 101us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.2365 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.0870 - 698us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0993 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.4276 - 697us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.1487 - 739us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.0360 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.8380 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.1211 - 825us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.1378 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.4758 - 1ms/epoch - 19us/sample\n", - "Generation 175/300\n", - "Solving for Nash Equilibrium in Generation 175/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.9374 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9708 - 2ms/epoch - 40us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9211 - 631us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.3155 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1083 - 625us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.9672 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.4637 - 762us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.7229 - 821us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.9062 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.7641 - 783us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.1761 - 625us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.7924 - 588us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.7211 - 854us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.3109 - 809us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.6122 - 14ms/epoch - 227us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6899 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.6305 - 611us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.0129 - 767us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.6484 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.6335 - 906us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5233 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.8078 - 692us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.6087 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.7928 - 4ms/epoch - 66us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.4432 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.8825 - 631us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.3132 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1072 - 1ms/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6377 - 765us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.0699 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.2913 - 867us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.5614 - 710us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.4052 - 818us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201.7755 - 688us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788.0151 - 644us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.4096 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.6769 - 613us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.6221 - 705us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.7538 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9673 - 757us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.2195 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.0063 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.8890 - 711us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2352 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.0912 - 599us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.7846 - 723us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.2750 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.6188 - 625us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.2412 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.3551 - 663us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.5898 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.4049 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.9594 - 648us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3841 - 578us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.2303 - 670us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.7741 - 660us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.9413 - 596us/epoch - 10us/sample\n", - "Generation 176/300\n", - "Solving for Nash Equilibrium in Generation 176/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.5719 - 611us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3743 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.1162 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7870 - 644us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.1042 - 4ms/epoch - 58us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.7799 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.6642 - 669us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7905 - 724us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.5228 - 796us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.6862 - 581us/epoch - 9us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.6324 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8247 - 670us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.0173 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.5710 - 787us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.0325 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.2578 - 677us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3475 - 2ms/epoch - 39us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.4267 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.1453 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123.1785 - 651us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.4894 - 851us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.2165 - 695us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.6820 - 704us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.6142 - 13ms/epoch - 206us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43.2157 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.3596 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.6336 - 606us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.1625 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.3285 - 736us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.8069 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.5912 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5141 - 875us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.3875 - 705us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.7511 - 859us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.4564 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.5519 - 703us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.0079 - 716us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.3678 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.9483 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2268 - 745us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5313 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.3698 - 709us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.0350 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.3383 - 1ms/epoch - 17us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1395 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.0084 - 647us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6615 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.3534 - 798us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773.3345 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.6277 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.6814 - 922us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.8812 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.2888 - 644us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.5767 - 770us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.9441 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.3497 - 2ms/epoch - 35us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.6865 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.7593 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2255 - 748us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.6229 - 604us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.4269 - 801us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.1586 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.0760 - 665us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.8274 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.7859 - 749us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.6746 - 643us/epoch - 10us/sample\n", - "Generation 177/300\n", - "Solving for Nash Equilibrium in Generation 177/300\n", + "Generation 41/300\n", + "Solving for Nash Equilibrium in Generation 41/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.2433 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 225.5968 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.1862 - 654us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4178 - 671us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.6357 - 1ms/epoch - 23us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.5682 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 121.9871 - 641us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.4746 - 661us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.6606 - 586us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.3914 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.0045 - 918us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1901 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 118us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248.6650 - 7ms/epoch - 113us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.5899 - 673us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.7878 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.8221 - 5ms/epoch - 83us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.5426 - 814us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.4153 - 641us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.0097 - 952us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 362.3039 - 850us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.7787 - 818us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.0416 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.5421 - 763us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.9444 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.5621 - 2ms/epoch - 40us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9426 - 2ms/epoch - 25us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.5039 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.9390 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.7806 - 773us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.8629 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2081 - 678us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.8836 - 905us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0096 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.8362 - 633us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.2875 - 5ms/epoch - 83us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.7048 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.6671 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0394 - 826us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.8816 - 774us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4105 - 692us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.2610 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.9965 - 718us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6758 - 730us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.3651 - 620us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.3936 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.8834 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.9281 - 771us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.9601 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.0744 - 654us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.6928 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.7029 - 718us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.5625 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.9940 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.9977 - 731us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.8235 - 756us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.4113 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.1532 - 2ms/epoch - 29us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.2661 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.7126 - 975us/epoch - 16us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.3441 - 842us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2377 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.9842 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.8691 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.7569 - 691us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 178/300\n", - "Solving for Nash Equilibrium in Generation 178/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.2618 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5768 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5893 - 795us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.8749 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4031 - 738us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.9872 - 888us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8110 - 4ms/epoch - 71us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.9358 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.2001 - 593us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.9807 - 1ms/epoch - 19us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5490 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.8951 - 1ms/epoch - 21us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.4937 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.7672 - 695us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.3120 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.2561 - 644us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.7913 - 823us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.9285 - 2ms/epoch - 27us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4567 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.9860 - 1ms/epoch - 18us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.2459 - 650us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.8496 - 785us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.1772 - 782us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8818 - 1ms/epoch - 18us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248.8298 - 689us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2043 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3516 - 680us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.2190 - 589us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.9280 - 745us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.6635 - 600us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.4342 - 728us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.8368 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.8254 - 715us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.7247 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.9266 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.6379 - 582us/epoch - 9us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.6875 - 671us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.2627 - 665us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1172 - 903us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9919 - 694us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5171 - 783us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.9608 - 642us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8716 - 708us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.0341 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1243 - 596us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.5616 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.3227 - 634us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.9406 - 964us/epoch - 16us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40.4391 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.4229 - 1ms/epoch - 17us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9350 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.8439 - 710us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.2469 - 744us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.0374 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.4961 - 2ms/epoch - 32us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.5965 - 726us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9161 - 580us/epoch - 9us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 124.5646 - 695us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.5912 - 605us/epoch - 10us/sample\n", - "Generation 179/300\n", - "Solving for Nash Equilibrium in Generation 179/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.8402 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.6042 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.8480 - 728us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3556 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.3444 - 709us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.0412 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.8143 - 572us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.4484 - 716us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.6003 - 898us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.9608 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.8962 - 760us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.1985 - 651us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.2965 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.1750 - 654us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.1398 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.2743 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.6284 - 720us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.3934 - 696us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.3188 - 759us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.1549 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.0963 - 1ms/epoch - 23us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.7125 - 734us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.5149 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.6608 - 813us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.8823 - 822us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.1589 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.7072 - 654us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.4136 - 780us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.5643 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.4311 - 752us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.1329 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.6107 - 688us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.7601 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3260 - 896us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.3780 - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.3342 - 975us/epoch - 16us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2430 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.7673 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.0079 - 2ms/epoch - 25us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.7357 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.9179 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0136 - 609us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.5613 - 2ms/epoch - 28us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.7713 - 574us/epoch - 9us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.6103 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.9610 - 658us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.6187 - 1ms/epoch - 17us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.5217 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.3104 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8535 - 1ms/epoch - 17us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6034 - 1ms/epoch - 16us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.5152 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.1559 - 838us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.0799 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.9083 - 746us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9810 - 1ms/epoch - 21us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46.1481 - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.1994 - 788us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2841 - 641us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.3905 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.5730 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0795 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.7043 - 915us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.8803 - 658us/epoch - 11us/sample\n", - "Generation 180/300\n", - "Solving for Nash Equilibrium in Generation 180/300\n", + "Generation 42/300\n", + "Solving for Nash Equilibrium in Generation 42/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.7281 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.8159 - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.0050 - 1ms/epoch - 20us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.5470 - 1ms/epoch - 20us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.8713 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9799 - 951us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.5474 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.6272 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.0490 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.9832 - 550us/epoch - 9us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3417 - 678us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.2744 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.9201 - 634us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.1723 - 949us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.6246 - 589us/epoch - 9us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.8800 - 3ms/epoch - 41us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4229 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.2277 - 699us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8676 - 671us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8276 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.8507 - 694us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6902 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.8352 - 1ms/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.4787 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.3112 - 635us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.3266 - 906us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.2313 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.9929 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.0013 - 635us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7214 - 737us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.7577 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.4388 - 771us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4627 - 727us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.4098 - 698us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.9835 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5836 - 6ms/epoch - 94us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.1573 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.3088 - 683us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8223 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1390 - 583us/epoch - 9us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.9525 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.7890 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.4117 - 728us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.8750 - 770us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7516 - 720us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.4197 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.3917 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2308 - 623us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.3811 - 733us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.9489 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.7665 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.0974 - 833us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9548 - 729us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.8889 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.1252 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.2657 - 635us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0685 - 734us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1605 - 581us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.3962 - 936us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.4169 - 622us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6276 - 719us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5206 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.0536 - 653us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.2518 - 592us/epoch - 10us/sample\n", - "Generation 181/300\n", - "Solving for Nash Equilibrium in Generation 181/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.9747 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.7207 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2950 - 629us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 137.3776 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0990 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.2234 - 813us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8060 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.5405 - 644us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.9084 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.9011 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.7406 - 604us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.8704 - 753us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.8464 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.4690 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.6483 - 828us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.4059 - 671us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.6182 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.5237 - 725us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.9626 - 625us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.1594 - 663us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.2419 - 666us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.2081 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.7305 - 662us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.3539 - 568us/epoch - 9us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.8972 - 6ms/epoch - 102us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5630 - 728us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.5487 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0077 - 788us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.2221 - 797us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.1380 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.5793 - 728us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.9891 - 613us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.6807 - 814us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.0663 - 1ms/epoch - 17us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.0457 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.9000 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.4130 - 677us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.6501 - 625us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.8835 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.8424 - 784us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.4264 - 779us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.1095 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8249 - 630us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.7428 - 1ms/epoch - 20us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.2298 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.3785 - 654us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.3008 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.2649 - 842us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.6330 - 955us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.8238 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.9346 - 655us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 148.4433 - 765us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.3776 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6155 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.3360 - 797us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.0139 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.2967 - 783us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.7648 - 843us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.9850 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.8517 - 2ms/epoch - 40us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.0981 - 719us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.2029 - 792us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 164us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.4905 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.3112 - 665us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.1004 - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.5797 - 5ms/epoch - 81us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.6459 - 710us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2411 - 688us/epoch - 11us/sample\n", - "Generation 182/300\n", - "Solving for Nash Equilibrium in Generation 182/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.2068 - 763us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.3850 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.7721 - 2ms/epoch - 25us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.9638 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.5796 - 2ms/epoch - 28us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.2003 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.8217 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.7556 - 683us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.6340 - 628us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.9609 - 676us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3689 - 712us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.7594 - 595us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.4681 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4275 - 551us/epoch - 9us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.4195 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.4433 - 774us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.6805 - 3ms/epoch - 47us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.6580 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.7191 - 713us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.6305 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.2973 - 648us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.3745 - 837us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.2597 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240.0808 - 1ms/epoch - 19us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3536 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.0367 - 608us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.2172 - 639us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.1385 - 765us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.9107 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.1814 - 3ms/epoch - 43us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.9146 - 1ms/epoch - 22us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.2095 - 978us/epoch - 16us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.4709 - 761us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.7856 - 683us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.8544 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.0119 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.7946 - 833us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3732 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.0231 - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3580 - 648us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4119 - 651us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.2275 - 661us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5641 - 726us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.9992 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.6258 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2827 - 604us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 362.3799 - 579us/epoch - 9us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8683 - 715us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.1728 - 2ms/epoch - 32us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.8918 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.7410 - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.7835 - 668us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8696 - 904us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8681 - 1ms/epoch - 21us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.8567 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5551 - 678us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4523 - 805us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.9663 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.4824 - 589us/epoch - 9us/sample\n", - "Generation 183/300\n", - "Solving for Nash Equilibrium in Generation 183/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.6195 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.0705 - 6ms/epoch - 100us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.7810 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4034 - 2ms/epoch - 32us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.9035 - 931us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.6466 - 866us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6366 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.5689 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.8279 - 781us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.0289 - 604us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0985 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.0904 - 1ms/epoch - 17us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.4316 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.0800 - 1ms/epoch - 21us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.0849 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5048 - 759us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0062 - 2ms/epoch - 35us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.1141 - 874us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.0273 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5989 - 1ms/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.3444 - 887us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7885 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.3256 - 601us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2495 - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.5474 - 991us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.5834 - 7ms/epoch - 116us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4660 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0328 - 605us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9103 - 665us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.1527 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.8345 - 1ms/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.3312 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.3844 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.5386 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.0012 - 792us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.0862 - 904us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5762 - 869us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.0426 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4191 - 1ms/epoch - 20us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.2421 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9213 - 942us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.3147 - 2ms/epoch - 28us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.3234 - 833us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.5405 - 1ms/epoch - 18us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.1472 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.1035 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.8782 - 917us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.2350 - 700us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.6684 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.4271 - 3ms/epoch - 50us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.3983 - 709us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.4890 - 660us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.4952 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.2318 - 738us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.9611 - 602us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.9572 - 900us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.7654 - 836us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.3532 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.7886 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240.6967 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.8891 - 1ms/epoch - 19us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8973 - 767us/epoch - 12us/sample\n", - "Generation 184/300\n", - "Solving for Nash Equilibrium in Generation 184/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.4091 - 817us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.9919 - 592us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.0054 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8290 - 2ms/epoch - 29us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1395 - 576us/epoch - 9us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.0112 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.3182 - 743us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3577 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.2247 - 690us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.7975 - 685us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.6046 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.7451 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.4535 - 718us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.2613 - 785us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.0634 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.0616 - 782us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.3936 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.3774 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.2379 - 628us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.1168 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4417 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9662 - 715us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.4338 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.7974 - 727us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.1665 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.4373 - 665us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6324 - 637us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.2512 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.1096 - 732us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7811 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.1259 - 628us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.2363 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.1347 - 758us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.1269 - 630us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.0978 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7652 - 669us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.6335 - 751us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.2944 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.0183 - 3ms/epoch - 50us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.0662 - 846us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.0760 - 1ms/epoch - 16us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.1090 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.0797 - 781us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.3660 - 907us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.7743 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.1266 - 725us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.7969 - 629us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.1998 - 6ms/epoch - 93us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.1918 - 742us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.7150 - 760us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6998 - 671us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.5849 - 635us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.8027 - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.7436 - 1ms/epoch - 23us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.3201 - 3ms/epoch - 53us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 185.9536 - 602us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.0153 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4016 - 600us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.0507 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.7767 - 625us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.7854 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1486 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.1394 - 2ms/epoch - 27us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.9999 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.2631 - 755us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.0712 - 1ms/epoch - 18us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361.2486 - 601us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.0278 - 668us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.6901 - 4ms/epoch - 60us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 185/300\n", - "Solving for Nash Equilibrium in Generation 185/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.5666 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.8911 - 2ms/epoch - 25us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.0381 - 579us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.7318 - 685us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.5447 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.5480 - 6ms/epoch - 97us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.5654 - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.2653 - 850us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.1208 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.2847 - 584us/epoch - 9us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8141 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.8437 - 678us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.1684 - 742us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.0554 - 701us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.3962 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.3248 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.8339 - 615us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.9573 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.1292 - 530us/epoch - 9us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3207 - 703us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.5896 - 911us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.0982 - 564us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.6473 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.1261 - 1ms/epoch - 17us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.0526 - 678us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.6821 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.5997 - 701us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.6622 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.5656 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.0858 - 688us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.3818 - 720us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.6888 - 753us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.2803 - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.1372 - 720us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6913 - 936us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.6711 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4645 - 678us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.1389 - 663us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.0655 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.2579 - 48ms/epoch - 770us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.6414 - 852us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.2042 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.0892 - 762us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5097 - 627us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0848 - 782us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.1207 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.2665 - 612us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.0359 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6602 - 705us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 130.8583 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.6849 - 747us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.5759 - 793us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.8076 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.9609 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.6410 - 2ms/epoch - 40us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.1982 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.0613 - 874us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.8271 - 751us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.7567 - 684us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.8433 - 975us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.9313 - 613us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.1376 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.2653 - 716us/epoch - 12us/sample\n", - "Generation 186/300\n", - "Solving for Nash Equilibrium in Generation 186/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.0479 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5776 - 697us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 17ms/epoch - 277us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231.9506 - 7ms/epoch - 119us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.4057 - 723us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1094 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.6964 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.7715 - 662us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.2785 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.5854 - 927us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.0943 - 1ms/epoch - 21us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.4604 - 704us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.9015 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.0118 - 756us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.4332 - 806us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.3553 - 665us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5262 - 662us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8256 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.7358 - 840us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.2841 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1724 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.7805 - 728us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3480 - 1ms/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.9494 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.6807 - 924us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.1028 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5372 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.9451 - 659us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.7015 - 863us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2209 - 819us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.9610 - 827us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.9162 - 738us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.0355 - 676us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9399 - 2ms/epoch - 30us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.4480 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.9852 - 757us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.5469 - 748us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.7029 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.6473 - 695us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8707 - 600us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.2249 - 721us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.2762 - 694us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3779 - 667us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.6886 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.6623 - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7073 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.5352 - 668us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.0865 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.6750 - 646us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.9161 - 1ms/epoch - 19us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2404 - 661us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.8353 - 743us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.4733 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3297 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.4243 - 663us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.7814 - 705us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.1491 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.0347 - 578us/epoch - 9us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.0468 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.7467 - 664us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.3928 - 742us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.9348 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.0780 - 656us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.6652 - 709us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0704 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.9590 - 3ms/epoch - 46us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.3117 - 801us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.0028 - 752us/epoch - 12us/sample\n", - "Generation 187/300\n", - "Solving for Nash Equilibrium in Generation 187/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.5236 - 551us/epoch - 9us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.3649 - 680us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.0965 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.7690 - 1ms/epoch - 19us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.6028 - 671us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.1287 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.7471 - 741us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.7816 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.1117 - 957us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.8951 - 738us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.5163 - 653us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.8655 - 768us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.1777 - 1ms/epoch - 20us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.2364 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 151us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1086 - 6ms/epoch - 94us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3913 - 853us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.1028 - 2ms/epoch - 27us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.5202 - 1ms/epoch - 18us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.3591 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3017 - 721us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.4119 - 893us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.5256 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 222.7367 - 588us/epoch - 9us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8928 - 592us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.6274 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.7776 - 773us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.6491 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8569 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.5958 - 725us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0536 - 5ms/epoch - 76us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.8889 - 675us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.0924 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.6509 - 646us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.4164 - 688us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.4623 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.8397 - 764us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.6315 - 645us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.7715 - 1ms/epoch - 21us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.2031 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.8276 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.8176 - 631us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.7589 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4579 - 819us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.0323 - 881us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7827 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.7245 - 2ms/epoch - 29us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.2884 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.7633 - 597us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.1394 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.5573 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.4691 - 788us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5413 - 3ms/epoch - 46us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43.2619 - 2ms/epoch - 30us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5443 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.9065 - 1ms/epoch - 17us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.6760 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1566 - 994us/epoch - 16us/sample\n", - "Generation 188/300\n", - "Solving for Nash Equilibrium in Generation 188/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 215.4801 - 3ms/epoch - 50us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.8233 - 716us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.7739 - 693us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.0170 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.7590 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.6206 - 707us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.7490 - 605us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.6984 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.2788 - 737us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.5048 - 616us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.9718 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.0250 - 576us/epoch - 9us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.2213 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8179 - 744us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.9609 - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8901 - 647us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.6661 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.0485 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.3124 - 4ms/epoch - 59us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.8513 - 3ms/epoch - 47us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.5383 - 663us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7771 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.6194 - 701us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.7900 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 766.6839 - 830us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2690 - 882us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.5441 - 883us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.9030 - 11ms/epoch - 181us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.5471 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.2958 - 718us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9983 - 679us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.0356 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.4069 - 752us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.6010 - 849us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.3429 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.6024 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.7280 - 2ms/epoch - 26us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.7603 - 741us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.8610 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.3754 - 774us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.2623 - 982us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 67.9814 - 677us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9026 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.2521 - 850us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.7109 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.3499 - 3ms/epoch - 47us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.2231 - 706us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.9048 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.4242 - 662us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.5395 - 868us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.9166 - 651us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 123us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.7875 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.9440 - 5ms/epoch - 74us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45.1684 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.9229 - 824us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3246 - 818us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.1385 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.5148 - 605us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.7953 - 790us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.3122 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.2536 - 584us/epoch - 9us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.3041 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1511 - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.7371 - 678us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4654 - 669us/epoch - 11us/sample\n", - "Generation 189/300\n", - "Solving for Nash Equilibrium in Generation 189/300\n", + "Generation 43/300\n", + "Solving for Nash Equilibrium in Generation 43/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.6131 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.1822 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.8608 - 889us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.0700 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.3353 - 693us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.5738 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4352 - 771us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.3734 - 777us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.0118 - 762us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.6807 - 1ms/epoch - 18us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.2628 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.5388 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.8964 - 613us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.8701 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.2184 - 693us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.8950 - 2ms/epoch - 25us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.6294 - 900us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0412 - 849us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5522 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.9255 - 808us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.7034 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.4501 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.9884 - 789us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.7335 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.7134 - 932us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8984 - 747us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1792 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.1860 - 720us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1738 - 717us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2955 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.4371 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2236 - 1ms/epoch - 18us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2302 - 953us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.5710 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.4911 - 580us/epoch - 9us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.1456 - 773us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.7971 - 693us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3762 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5877 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.4850 - 586us/epoch - 9us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.0434 - 747us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0593 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.7697 - 666us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.6985 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7808 - 631us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.5637 - 1ms/epoch - 20us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.1485 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.3414 - 1ms/epoch - 20us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.2098 - 822us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.0120 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.4100 - 695us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.2544 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.7677 - 605us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.9990 - 712us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.6992 - 693us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7476 - 29ms/epoch - 475us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.1393 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.9672 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4198 - 979us/epoch - 16us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.5112 - 756us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.0304 - 711us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.1896 - 746us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.1295 - 2ms/epoch - 29us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.7342 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.5944 - 2ms/epoch - 26us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 198.8904 - 1ms/epoch - 16us/sample\n", - "Generation 190/300\n", - "Solving for Nash Equilibrium in Generation 190/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.2866 - 839us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.2741 - 719us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.5549 - 688us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2443 - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.7385 - 835us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.8260 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.6693 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 435.8578 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.0001 - 1ms/epoch - 20us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.8154 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.3036 - 1ms/epoch - 18us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.8273 - 803us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.4442 - 1ms/epoch - 23us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.5723 - 1ms/epoch - 24us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.2830 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2455 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8347 - 2ms/epoch - 30us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.6444 - 818us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.5308 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.5668 - 848us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.7040 - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.2606 - 905us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.5096 - 810us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.3709 - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.8679 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.9075 - 826us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.5737 - 811us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.2076 - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3128 - 939us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 114us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.3080 - 976us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 122.9015 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6093 - 1ms/epoch - 19us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.9410 - 711us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.8804 - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.6939 - 852us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.7776 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3899 - 857us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5823 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.1761 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.0393 - 597us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.4541 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.1173 - 748us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9114 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.1754 - 656us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.4129 - 661us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.0627 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6742 - 940us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.6824 - 959us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.1280 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.9050 - 797us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2592 - 766us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.5233 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 114us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.7507 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.0993 - 805us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.2388 - 831us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.3677 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3160 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406.1631 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5176 - 746us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.7455 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.6562 - 1ms/epoch - 17us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.9243 - 689us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.2577 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9066 - 601us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.1890 - 837us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.7911 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.5526 - 935us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.6925 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.7051 - 704us/epoch - 11us/sample\n", - "Generation 191/300\n", - "Solving for Nash Equilibrium in Generation 191/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 44/300\n", + "Solving for Nash Equilibrium in Generation 44/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.3560 - 803us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.4176 - 832us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.8383 - 633us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0062 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.7434 - 1ms/epoch - 18us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.2168 - 859us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 777.1165 - 825us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.3665 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.1187 - 628us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.3236 - 19ms/epoch - 300us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.3139 - 4ms/epoch - 64us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.6129 - 1ms/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.5628 - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.4905 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.7046 - 5ms/epoch - 83us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4517 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.8987 - 732us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.3833 - 791us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.4185 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7133 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2203 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.9954 - 772us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.6165 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.9505 - 716us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.2141 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2265 - 763us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.1243 - 2ms/epoch - 35us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.3133 - 842us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6730 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.7664 - 9ms/epoch - 151us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4103 - 745us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.7653 - 1ms/epoch - 18us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.6400 - 810us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.2709 - 895us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 85.2526 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.3033 - 793us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.7509 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.2971 - 714us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.9620 - 721us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.5686 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.8080 - 644us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6293 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1553 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.2862 - 726us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.5704 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.7078 - 551us/epoch - 9us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1353 - 770us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.7323 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.9708 - 795us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.3520 - 690us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.8743 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.8813 - 811us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.1691 - 813us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.5851 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.8802 - 707us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.9657 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.0734 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.8047 - 662us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.7268 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.4925 - 630us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0668 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.2957 - 607us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.6925 - 911us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.9949 - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.4855 - 799us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.3644 - 1ms/epoch - 23us/sample\n", - "Generation 192/300\n", - "Solving for Nash Equilibrium in Generation 192/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 45/300\n", + "Solving for Nash Equilibrium in Generation 45/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.5531 - 789us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.4251 - 909us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.1877 - 2ms/epoch - 26us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.5443 - 844us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.2794 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.5827 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.5906 - 662us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5291 - 1ms/epoch - 22us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.7432 - 659us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.9764 - 756us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8624 - 2ms/epoch - 40us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6356 - 677us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.4441 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.0718 - 917us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.9113 - 609us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.4638 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.6342 - 589us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.0198 - 727us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.8346 - 698us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.6321 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.1425 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.1362 - 1ms/epoch - 23us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.2050 - 628us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4648 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.2958 - 690us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7280 - 872us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.1434 - 614us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.1161 - 954us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.8444 - 735us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.2528 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.3135 - 916us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.5797 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1274 - 793us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.2715 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.8120 - 651us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.4666 - 588us/epoch - 9us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.4971 - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.2538 - 810us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.6026 - 3ms/epoch - 45us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.8133 - 693us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.7685 - 735us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.7830 - 685us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.1914 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.4458 - 1ms/epoch - 22us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.1696 - 782us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.9241 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.7654 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.2898 - 2ms/epoch - 25us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.8607 - 867us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.8133 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37.2730 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.6222 - 718us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.0849 - 1ms/epoch - 22us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.6284 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.8205 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.0952 - 679us/epoch - 11us/sample\n", - "Generation 193/300\n", - "Solving for Nash Equilibrium in Generation 193/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 46/300\n", + "Solving for Nash Equilibrium in Generation 46/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.4727 - 731us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43.7189 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.3695 - 921us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.3976 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.6310 - 713us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8134 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0641 - 2ms/epoch - 30us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.5536 - 2ms/epoch - 34us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.8852 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.4849 - 836us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.9836 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.2184 - 729us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.8502 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7630 - 687us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.6602 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.3975 - 767us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.5338 - 600us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5962 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.8687 - 579us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.4773 - 764us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.2430 - 651us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.4914 - 844us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.6445 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.3431 - 687us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5330 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.3605 - 937us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.4427 - 766us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.8888 - 5ms/epoch - 76us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.2894 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.2874 - 931us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.6407 - 829us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.9095 - 1ms/epoch - 20us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.5309 - 794us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.2510 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.2696 - 636us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46.5423 - 1ms/epoch - 24us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.4555 - 821us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.8221 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.4842 - 9ms/epoch - 145us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.8283 - 689us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.9169 - 2ms/epoch - 25us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.2304 - 1ms/epoch - 18us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7625 - 5ms/epoch - 81us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.8399 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.0446 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.5284 - 783us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.8699 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.7316 - 841us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5565 - 3ms/epoch - 53us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 13ms/epoch - 203us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.6217 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.6780 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.4543 - 3ms/epoch - 43us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.0171 - 2ms/epoch - 31us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.8608 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7654 - 2ms/epoch - 32us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.3166 - 733us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.4622 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.7164 - 629us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.9925 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1116 - 741us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5301 - 815us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6296 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.7231 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.1574 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.5172 - 980us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.6902 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.6515 - 740us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 194/300\n", - "Solving for Nash Equilibrium in Generation 194/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.0500 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.3569 - 686us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.2377 - 693us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.9440 - 764us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.8821 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.8102 - 571us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.7514 - 760us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.4304 - 650us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.0118 - 2ms/epoch - 28us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.6799 - 860us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.3809 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.4914 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.4400 - 727us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.4087 - 659us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.8206 - 843us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9642 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.7869 - 2ms/epoch - 30us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.9211 - 839us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.8300 - 569us/epoch - 9us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.4396 - 727us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.6595 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.6287 - 788us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.6780 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.3849 - 671us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.0067 - 850us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.9003 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7909 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.6520 - 910us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.0982 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.7464 - 2ms/epoch - 25us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.3939 - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.0689 - 881us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.4858 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0642 - 881us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.4236 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.9273 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.0174 - 740us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7333 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0389 - 998us/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.3372 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8552 - 948us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.5184 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.4102 - 767us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.4570 - 659us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.1617 - 635us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 101us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.5770 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.9095 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.4911 - 683us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.0122 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.4597 - 945us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.4524 - 758us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.1481 - 597us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4978 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9462 - 2ms/epoch - 27us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.1510 - 5ms/epoch - 73us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.9636 - 650us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54.9128 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9438 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.8027 - 670us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.8986 - 696us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 135us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.1894 - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.8479 - 722us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.1730 - 2ms/epoch - 34us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.4661 - 930us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.7988 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.2314 - 722us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7027 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.1893 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.9461 - 759us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.4835 - 839us/epoch - 14us/sample\n", - "Generation 195/300\n", - "Solving for Nash Equilibrium in Generation 195/300\n", + "Generation 47/300\n", + "Solving for Nash Equilibrium in Generation 47/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.4359 - 668us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.5271 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.5266 - 597us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.7647 - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.3777 - 840us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.4221 - 688us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1962 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8928 - 7ms/epoch - 113us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5789 - 941us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.8480 - 722us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.3193 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3375 - 642us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.8784 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.1944 - 703us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.0631 - 898us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.1819 - 751us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.4464 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8408 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.4884 - 715us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6599 - 675us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.1670 - 679us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.7838 - 553us/epoch - 9us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.8731 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.5515 - 767us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3723 - 616us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.0128 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.2379 - 662us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 116us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.8040 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.3232 - 662us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0897 - 1ms/epoch - 19us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.3439 - 643us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.9698 - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.6202 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5489 - 937us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.4309 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.4141 - 782us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.9176 - 1ms/epoch - 22us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 32ms/epoch - 518us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4648 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2303 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.1254 - 743us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.9916 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.8915 - 792us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.6027 - 1ms/epoch - 17us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.1581 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.7267 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.0218 - 684us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0405 - 644us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.3182 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.8812 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7339 - 927us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4009 - 747us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.9433 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.2858 - 744us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.8792 - 648us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.7173 - 710us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.9064 - 740us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.1893 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.5622 - 1ms/epoch - 17us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.2847 - 677us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.5041 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.4163 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.8324 - 1ms/epoch - 19us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.3222 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.0126 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.1493 - 685us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.4770 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.4573 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8250 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.7285 - 653us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.4551 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.1931 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 196/300\n", - "Solving for Nash Equilibrium in Generation 196/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0999 - 705us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.9318 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.0633 - 2ms/epoch - 36us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.3377 - 736us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.5841 - 738us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.7655 - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8019 - 756us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 217.7541 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.9274 - 7ms/epoch - 114us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.5826 - 729us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1447 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5587 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.6362 - 1ms/epoch - 23us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.6108 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.4242 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7662 - 2ms/epoch - 29us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8978 - 608us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.2988 - 790us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.5564 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.0596 - 786us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.9671 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.5597 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.8275 - 801us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.8934 - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.5728 - 712us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.0188 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.3696 - 651us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.5793 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.7546 - 624us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.2210 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.6502 - 642us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3610 - 943us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3363 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5916 - 1ms/epoch - 19us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.1791 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.8925 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.6857 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.7443 - 2ms/epoch - 28us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.8850 - 813us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4361 - 557us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.4937 - 750us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.7858 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.8099 - 730us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 128us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.1941 - 1ms/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.5840 - 2ms/epoch - 36us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.6467 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.8262 - 760us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.1173 - 802us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.0173 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.1069 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1127 - 619us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.2692 - 814us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.9802 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.1842 - 928us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 104us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.0582 - 3ms/epoch - 45us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.2235 - 726us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.5876 - 583us/epoch - 9us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.6044 - 600us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.4799 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8257 - 925us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.1157 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.6629 - 813us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.5852 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.7808 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250.0165 - 4ms/epoch - 70us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.8158 - 1ms/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.6830 - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9731 - 705us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.7988 - 699us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1394 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.0045 - 1ms/epoch - 19us/sample\n", - "Generation 197/300\n", - "Solving for Nash Equilibrium in Generation 197/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.1598 - 568us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.3160 - 658us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7871 - 864us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.3973 - 796us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.8399 - 696us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.5302 - 731us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.3659 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.5849 - 4ms/epoch - 59us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.3523 - 927us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.8976 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.6720 - 772us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.7366 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.6192 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.1404 - 921us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7902 - 650us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.5020 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.6547 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.6871 - 767us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.1612 - 826us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.8494 - 625us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.6967 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.3104 - 578us/epoch - 9us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.6818 - 662us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.6133 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.7195 - 677us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.9415 - 832us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.2016 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.3774 - 516us/epoch - 8us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.4842 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.2449 - 949us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.7534 - 6ms/epoch - 100us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.5772 - 1ms/epoch - 19us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.5690 - 855us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.8321 - 707us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.9971 - 741us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.4959 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.2195 - 1ms/epoch - 18us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.7864 - 653us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6095 - 776us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0776 - 803us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.4776 - 713us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.4675 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1240 - 1ms/epoch - 17us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231.3358 - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.8005 - 862us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8781 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2899 - 649us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 117us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.2930 - 774us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.2643 - 738us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.9425 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.7135 - 5ms/epoch - 74us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.0800 - 645us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.7488 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.4224 - 4ms/epoch - 65us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.9384 - 582us/epoch - 9us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.0325 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4348 - 858us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4098 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8763 - 802us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4218 - 691us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.1747 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.8496 - 702us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53.0066 - 900us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.8278 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.3232 - 607us/epoch - 10us/sample\n", - "Generation 198/300\n", - "Solving for Nash Equilibrium in Generation 198/300\n", + "Generation 48/300\n", + "Solving for Nash Equilibrium in Generation 48/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7708 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214.5357 - 5ms/epoch - 77us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0276 - 699us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.9274 - 899us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.5155 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.6163 - 610us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 179.4659 - 2ms/epoch - 37us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.6187 - 725us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.3900 - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.1164 - 665us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.9360 - 3ms/epoch - 44us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5857 - 995us/epoch - 16us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.8834 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.7036 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3975 - 763us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43.6487 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.8396 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.5230 - 777us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.1492 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4415 - 639us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.8702 - 10ms/epoch - 160us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.9665 - 575us/epoch - 9us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.9679 - 4ms/epoch - 69us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4124 - 4ms/epoch - 65us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.4633 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.0421 - 2ms/epoch - 25us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.5545 - 1ms/epoch - 17us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.7728 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.9578 - 680us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.4423 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.3594 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.7006 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.3710 - 681us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.4079 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.2522 - 2ms/epoch - 30us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.8741 - 5ms/epoch - 75us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.1794 - 662us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1677 - 867us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.7247 - 610us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6339 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.9821 - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.8043 - 696us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.4156 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.7228 - 3ms/epoch - 55us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.7570 - 775us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.0805 - 1ms/epoch - 22us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.9359 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.7653 - 726us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.5917 - 680us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.8189 - 604us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.4145 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.1519 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3944 - 632us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48.9149 - 685us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.0511 - 894us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.7841 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.8231 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.2931 - 608us/epoch - 10us/sample\n", - "Generation 199/300\n", - "Solving for Nash Equilibrium in Generation 199/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.3673 - 677us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4190 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.6685 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.7679 - 721us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.9833 - 725us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.1626 - 796us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.1796 - 625us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8413 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6122 - 711us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3741 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 147.7051 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7271 - 700us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0607 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.3477 - 576us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.9878 - 712us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.1872 - 768us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.7940 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.8711 - 683us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.8431 - 680us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.4998 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8125 - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3896 - 713us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.4759 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.3641 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.9078 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.0027 - 680us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.6793 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6273 - 662us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.2742 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.9772 - 706us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7076 - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.6712 - 2ms/epoch - 26us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8341 - 661us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.1655 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.0147 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.2789 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.1297 - 960us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.0109 - 728us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8306 - 979us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.8322 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.9922 - 705us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.5652 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.9580 - 846us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.6837 - 665us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.2721 - 620us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.9599 - 2ms/epoch - 32us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.4176 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.7905 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.3123 - 879us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.7986 - 637us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.3991 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1757 - 604us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.1292 - 894us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.8469 - 703us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.5584 - 7ms/epoch - 108us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.2433 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.5400 - 941us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.2220 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.2253 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.5589 - 833us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.2593 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.3951 - 918us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.7814 - 897us/epoch - 14us/sample\n", - "Generation 200/300\n", - "Solving for Nash Equilibrium in Generation 200/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.8552 - 786us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.4359 - 778us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.8526 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.8425 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.6165 - 601us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.3714 - 721us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.2739 - 6ms/epoch - 94us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6241 - 759us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 432.1645 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.1356 - 725us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.9629 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.4180 - 926us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.4954 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.7963 - 607us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.9197 - 684us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.4691 - 774us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.9709 - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.6840 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.6773 - 612us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.5989 - 746us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.7281 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.1212 - 748us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.3714 - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8857 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4665 - 1ms/epoch - 19us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.1915 - 696us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.0322 - 6ms/epoch - 98us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.4880 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.0357 - 3ms/epoch - 49us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.2048 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.0585 - 747us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.1541 - 801us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428.8923 - 2ms/epoch - 26us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.3006 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0529 - 808us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1102 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.1962 - 760us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.6758 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.7161 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.1191 - 716us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.4572 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.8630 - 2ms/epoch - 25us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5117 - 741us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.2970 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.6159 - 731us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9988 - 617us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.5213 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.2296 - 818us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.9511 - 859us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4419 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.7467 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.3018 - 748us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0562 - 727us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.8350 - 826us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.3776 - 698us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.0828 - 693us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.8237 - 916us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.4994 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.9648 - 707us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2808 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.4318 - 903us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9297 - 770us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.7752 - 989us/epoch - 16us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 201/300\n", - "Solving for Nash Equilibrium in Generation 201/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7478 - 735us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2224 - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.0023 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.6221 - 650us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6412 - 720us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.4081 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.8903 - 824us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5417 - 716us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.4725 - 827us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.0417 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.9926 - 2ms/epoch - 27us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.5060 - 784us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.3188 - 1ms/epoch - 17us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.3277 - 717us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248.4445 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.2599 - 849us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1444 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.9143 - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.5184 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.8597 - 812us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.0811 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.5094 - 743us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.2804 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.2059 - 618us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8455 - 713us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.9808 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8456 - 575us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7300 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3419 - 767us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.1948 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.9330 - 671us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.8746 - 657us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.8332 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.8648 - 3ms/epoch - 50us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.9673 - 651us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.8968 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.4161 - 662us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.5844 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.1307 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.0833 - 777us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7087 - 667us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.4940 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6269 - 4ms/epoch - 58us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41.5769 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0464 - 789us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.8753 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.6750 - 634us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.2993 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.5824 - 2ms/epoch - 28us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.4247 - 702us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.2303 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6954 - 744us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.0823 - 839us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.0548 - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5313 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.2557 - 736us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.3355 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.5217 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.5698 - 1ms/epoch - 16us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.8934 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.9446 - 645us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.5906 - 744us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.9320 - 693us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.0353 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.7311 - 710us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.6555 - 564us/epoch - 9us/sample\n", - "Generation 202/300\n", - "Solving for Nash Equilibrium in Generation 202/300\n", + "Generation 49/300\n", + "Solving for Nash Equilibrium in Generation 49/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.5555 - 973us/epoch - 16us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.3035 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5089 - 662us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.9286 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.6934 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7733 - 779us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.9335 - 691us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.5370 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.6078 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.6604 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.6163 - 734us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.8635 - 735us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.4188 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.0246 - 3ms/epoch - 41us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.5916 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4056 - 3ms/epoch - 46us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.0315 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.0611 - 773us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.5180 - 798us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5903 - 622us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.8340 - 557us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.6431 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.1550 - 4ms/epoch - 72us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.6416 - 618us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.2927 - 659us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.7685 - 636us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.4821 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.4008 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250.0850 - 2ms/epoch - 31us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.8940 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.4471 - 3ms/epoch - 48us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.3718 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.7960 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.6389 - 1ms/epoch - 20us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.7997 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.3179 - 634us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.7953 - 703us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.8393 - 747us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.1955 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.2455 - 652us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.6053 - 1ms/epoch - 18us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.1892 - 704us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.3525 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.4061 - 728us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773.0453 - 1ms/epoch - 22us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.3050 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.5660 - 4ms/epoch - 60us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.8852 - 863us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.0523 - 712us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4673 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.9352 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38.1205 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.3772 - 859us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.0374 - 760us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.2494 - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.9980 - 609us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 99us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.5477 - 7ms/epoch - 107us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0281 - 7ms/epoch - 109us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.1429 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.2717 - 672us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.9208 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.7934 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.6854 - 751us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3167 - 3ms/epoch - 42us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.4332 - 609us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.9443 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.8507 - 835us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.0056 - 695us/epoch - 11us/sample\n", - "Generation 203/300\n", - "Solving for Nash Equilibrium in Generation 203/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 50/300\n", + "Solving for Nash Equilibrium in Generation 50/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.2088 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.3038 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.4892 - 546us/epoch - 9us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.3844 - 1ms/epoch - 18us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7543 - 697us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.2231 - 1ms/epoch - 18us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.5567 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.9859 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.9397 - 712us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.6569 - 693us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.4051 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.8892 - 1ms/epoch - 18us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.7161 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.4508 - 902us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.7015 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.1102 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.9355 - 1ms/epoch - 16us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.9425 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.9021 - 625us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.8930 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.9684 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.4464 - 870us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.5392 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6447 - 754us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.7197 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.3887 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.1394 - 712us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.2708 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.7016 - 816us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.0267 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4590 - 2ms/epoch - 35us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.5257 - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.4731 - 751us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.5971 - 811us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.2351 - 763us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.0233 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.5386 - 2ms/epoch - 29us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.9389 - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.2069 - 1ms/epoch - 19us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5640 - 839us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.0097 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2054 - 1ms/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 234.0231 - 1ms/epoch - 18us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.3772 - 717us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.7218 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.5674 - 1ms/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6749 - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1095 - 8ms/epoch - 134us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.9599 - 3ms/epoch - 48us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.4723 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.9922 - 646us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.3193 - 658us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.2341 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.2208 - 3ms/epoch - 50us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4283 - 3ms/epoch - 46us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 229.3296 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.6372 - 611us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8987 - 795us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.8241 - 662us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.3650 - 1ms/epoch - 16us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9348 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.2657 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.0740 - 875us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9423 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.2683 - 567us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6738 - 619us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.3895 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.1941 - 734us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.1491 - 696us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.6271 - 796us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.6982 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.7190 - 2ms/epoch - 38us/sample\n", - "Generation 204/300\n", - "Solving for Nash Equilibrium in Generation 204/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 51/300\n", + "Solving for Nash Equilibrium in Generation 51/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.7791 - 704us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.1697 - 586us/epoch - 9us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.3842 - 865us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.4649 - 1ms/epoch - 23us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406.2440 - 985us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.9947 - 898us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.9412 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.1269 - 694us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201.0646 - 1ms/epoch - 24us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.9674 - 798us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1303 - 701us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.0022 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.5048 - 610us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7786 - 658us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.6423 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.6509 - 1ms/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.2384 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.1249 - 643us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.4778 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.0048 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.7318 - 719us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 212.3229 - 759us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9978 - 630us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7479 - 643us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.4639 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.3033 - 834us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.0727 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.0961 - 978us/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.3043 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.6010 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.3916 - 760us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.0846 - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.2502 - 627us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.4652 - 2ms/epoch - 28us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.3775 - 856us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.8669 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5334 - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.4375 - 801us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.7881 - 743us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.8373 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.8375 - 798us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.3739 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.1420 - 2ms/epoch - 25us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.0750 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6230 - 629us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.6342 - 651us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.6805 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1172 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7176 - 702us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.2404 - 887us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3271 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4585 - 3ms/epoch - 41us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.0335 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.4394 - 2ms/epoch - 31us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.3308 - 2ms/epoch - 35us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.5154 - 754us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.9348 - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.4500 - 772us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.9474 - 704us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.0740 - 698us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.8038 - 750us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7362 - 959us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.3880 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.9065 - 562us/epoch - 9us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.4633 - 889us/epoch - 14us/sample\n", - "Generation 205/300\n", - "Solving for Nash Equilibrium in Generation 205/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.1777 - 959us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.7234 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.0337 - 688us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.9369 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.1196 - 795us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.6036 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.1033 - 645us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.6366 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.2792 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8132 - 719us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.1649 - 864us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.9734 - 687us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.0198 - 815us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.4742 - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.8817 - 923us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.0872 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.3458 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.3345 - 634us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.0026 - 620us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.7915 - 569us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.2385 - 669us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.2587 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.4174 - 696us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.7578 - 607us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0719 - 695us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.5772 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.0639 - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.5977 - 960us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.4247 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4264 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.7621 - 536us/epoch - 9us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.4337 - 652us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.2595 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.1017 - 893us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.0823 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.5884 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.1293 - 716us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4225 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.7013 - 597us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.7037 - 772us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.8221 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.1273 - 786us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.4732 - 817us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.8102 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.0510 - 540us/epoch - 9us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.4253 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.4127 - 650us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.0754 - 694us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.3709 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7798 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.4228 - 641us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1770 - 738us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.9170 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.1597 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.8297 - 735us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.6543 - 6ms/epoch - 90us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.2788 - 726us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.0126 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0264 - 764us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9941 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.0546 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.7453 - 761us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.8417 - 658us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.7248 - 838us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.8386 - 741us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.4042 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.3046 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8026 - 648us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.2808 - 686us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.3756 - 819us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.7010 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9167 - 755us/epoch - 12us/sample\n", - "Generation 206/300\n", - "Solving for Nash Equilibrium in Generation 206/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.1316 - 779us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.3194 - 2ms/epoch - 33us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.2843 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.5652 - 2ms/epoch - 32us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.0536 - 780us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.9557 - 675us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.3594 - 806us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3829 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4930 - 725us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.1620 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.5351 - 906us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.8156 - 650us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.1585 - 848us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5894 - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.4454 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.1111 - 2ms/epoch - 30us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2363 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.7722 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.5147 - 1ms/epoch - 23us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.9692 - 750us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.4623 - 715us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.8532 - 879us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.1956 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.8716 - 748us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9446 - 673us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.0894 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.4602 - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.3516 - 830us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.3920 - 833us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.9196 - 607us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.0835 - 744us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.6183 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.0148 - 644us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.7545 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.5594 - 505us/epoch - 8us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3309 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3802 - 4ms/epoch - 67us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7406 - 824us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4846 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.8400 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9133 - 713us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9778 - 691us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2742 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.0853 - 607us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4299 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.1648 - 738us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.0238 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.8272 - 620us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.9862 - 721us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.7989 - 702us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.7959 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.0493 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2405 - 836us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.2542 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.7453 - 797us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1108 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3998 - 689us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4146 - 852us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.3580 - 702us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55.1111 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7159 - 632us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 207/300\n", - "Solving for Nash Equilibrium in Generation 207/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.3328 - 715us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.8762 - 7ms/epoch - 114us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.4904 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.5882 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.0797 - 705us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6645 - 715us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.5775 - 725us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.4194 - 905us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.4398 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6452 - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.1467 - 878us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.6257 - 748us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.6532 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3566 - 750us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.7785 - 918us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.7579 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230.0080 - 764us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.9809 - 713us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.3191 - 778us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.7591 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.0132 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.6425 - 877us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.6190 - 850us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.8278 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.9554 - 711us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6589 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9661 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.2780 - 797us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.0699 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.1909 - 605us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.9379 - 827us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.6338 - 646us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.4734 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.7399 - 556us/epoch - 9us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.2196 - 668us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4462 - 681us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.9269 - 712us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.6148 - 767us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.2913 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.1854 - 953us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.2979 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.7628 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.3868 - 663us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.4524 - 804us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.2249 - 1ms/epoch - 22us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.8719 - 594us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.2784 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.1261 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.3503 - 802us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.9919 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.3233 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.3686 - 571us/epoch - 9us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.2951 - 9ms/epoch - 146us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.6945 - 693us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.0575 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.6151 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.7105 - 998us/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.5112 - 733us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.1521 - 740us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.9935 - 16ms/epoch - 255us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.4454 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.6243 - 730us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.2436 - 890us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.0516 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.3233 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.5355 - 683us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2845 - 672us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230.9521 - 585us/epoch - 9us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4060 - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.1840 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.8865 - 723us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.1824 - 753us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 80.4295 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.7003 - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.3514 - 759us/epoch - 12us/sample\n", - "Generation 208/300\n", - "Solving for Nash Equilibrium in Generation 208/300\n", + "Generation 52/300\n", + "Solving for Nash Equilibrium in Generation 52/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3301 - 689us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7049 - 626us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.0370 - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.2701 - 800us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.7926 - 700us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.1366 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.9905 - 1ms/epoch - 17us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.3195 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.9636 - 904us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2038 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.5384 - 712us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.1061 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.6708 - 635us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.9003 - 892us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.5393 - 644us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.7198 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.5593 - 682us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.6011 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.9310 - 766us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.7458 - 673us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.2868 - 664us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.1562 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.2514 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.4789 - 724us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.6526 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.9450 - 789us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.6937 - 4ms/epoch - 58us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3556 - 1ms/epoch - 23us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.4125 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.6185 - 6ms/epoch - 92us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.0891 - 705us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8161 - 569us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.1189 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.8542 - 834us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.2235 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.8255 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 761.8818 - 651us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2785 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.4459 - 721us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5733 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2189 - 691us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.8181 - 826us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.4429 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.1465 - 878us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.1315 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9716 - 1ms/epoch - 19us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.4327 - 804us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.2255 - 697us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.6607 - 656us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.6017 - 1ms/epoch - 17us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.9529 - 856us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.6112 - 765us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.2268 - 854us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.1592 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.9194 - 11ms/epoch - 173us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.0490 - 4ms/epoch - 60us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.6034 - 1ms/epoch - 17us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.5295 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.6812 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.9524 - 864us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.0937 - 658us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 181.4979 - 840us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.3103 - 757us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3287 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.0576 - 727us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.9785 - 819us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 209/300\n", - "Solving for Nash Equilibrium in Generation 209/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.4917 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.9339 - 821us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2886 - 815us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.4610 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.1362 - 743us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.6412 - 913us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74.3516 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.2709 - 5ms/epoch - 88us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.1575 - 3ms/epoch - 42us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0820 - 764us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.0630 - 804us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.4213 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.8774 - 1ms/epoch - 17us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.6537 - 836us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.3616 - 689us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9761 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.8252 - 741us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.1157 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3210 - 2ms/epoch - 28us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.4350 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.4737 - 702us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 163.7416 - 705us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.0764 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.9125 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7657 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.8474 - 754us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.6951 - 681us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8176 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1443 - 890us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.4524 - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.2478 - 737us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 12ms/epoch - 194us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.2087 - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.9176 - 1ms/epoch - 17us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.4538 - 712us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.7507 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.5183 - 730us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.9160 - 5ms/epoch - 82us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.9753 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.1221 - 651us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.9384 - 806us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.3065 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.4034 - 1ms/epoch - 22us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.1455 - 668us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8476 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.7904 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.6260 - 14ms/epoch - 225us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7725 - 2ms/epoch - 25us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.7755 - 750us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.2194 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.3854 - 578us/epoch - 9us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.6061 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.8867 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.9821 - 1ms/epoch - 21us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214.0440 - 722us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.4451 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.1024 - 996us/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.2492 - 3ms/epoch - 47us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.7337 - 2ms/epoch - 28us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2422 - 2ms/epoch - 25us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.9289 - 964us/epoch - 16us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.9542 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.9718 - 879us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.3182 - 2ms/epoch - 27us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.9538 - 724us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.3919 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.1707 - 1ms/epoch - 21us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.6229 - 2ms/epoch - 25us/sample\n", - "Generation 210/300\n", - "Solving for Nash Equilibrium in Generation 210/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.1683 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.0868 - 769us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.1107 - 656us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8519 - 765us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.5800 - 636us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 100us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 105.5790 - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.7520 - 1ms/epoch - 21us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.4691 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.8710 - 9ms/epoch - 142us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9107 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.9635 - 3ms/epoch - 50us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.5086 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.7214 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.0302 - 1ms/epoch - 18us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.5994 - 952us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.7839 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1169 - 714us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.9934 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.4913 - 855us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.9740 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.7475 - 814us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.2310 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0637 - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.8087 - 963us/epoch - 16us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.4158 - 737us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.1691 - 2ms/epoch - 31us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6907 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.7323 - 6ms/epoch - 90us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.5518 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.6039 - 621us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1449 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5623 - 890us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3857 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.6189 - 776us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.9066 - 5ms/epoch - 83us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.9052 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.8408 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.8565 - 6ms/epoch - 90us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.6849 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.9908 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.0754 - 937us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.4542 - 1ms/epoch - 18us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.6253 - 2ms/epoch - 39us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9526 - 847us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.4600 - 1ms/epoch - 16us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4047 - 1ms/epoch - 17us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.3136 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5437 - 759us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.2519 - 959us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.5847 - 779us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.4073 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.7263 - 764us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.9285 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.4165 - 837us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.3527 - 1ms/epoch - 23us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.9940 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.1793 - 16ms/epoch - 255us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 228.1307 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5965 - 680us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.2870 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.1778 - 691us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.2106 - 1ms/epoch - 17us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.4349 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 119us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.4615 - 3ms/epoch - 54us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.2981 - 751us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.4205 - 688us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.7660 - 3ms/epoch - 56us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51.6027 - 793us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 186.2168 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.8822 - 816us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.3827 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.0627 - 6ms/epoch - 91us/sample\n", - "Generation 211/300\n", - "Solving for Nash Equilibrium in Generation 211/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.3875 - 2ms/epoch - 25us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.9171 - 2ms/epoch - 39us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.1857 - 936us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.2646 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6773 - 7ms/epoch - 111us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.4822 - 688us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.0941 - 896us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.7312 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.2401 - 2ms/epoch - 24us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.5005 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.0995 - 796us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.9909 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.2904 - 578us/epoch - 9us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.0497 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.9984 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.3893 - 861us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.5269 - 786us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.6835 - 639us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.1694 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.1315 - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 234.2464 - 759us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.0350 - 2ms/epoch - 34us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.1573 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 13ms/epoch - 205us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.9896 - 1ms/epoch - 17us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.0526 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5129 - 664us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.1978 - 2ms/epoch - 31us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2502 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.7482 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.2990 - 936us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.4813 - 716us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9655 - 3ms/epoch - 45us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.3987 - 567us/epoch - 9us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.2141 - 3ms/epoch - 45us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.3492 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.0013 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.4534 - 924us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.0168 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.7293 - 661us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.0436 - 756us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.0069 - 842us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.9733 - 661us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.3475 - 695us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.8159 - 587us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.2913 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.7703 - 7ms/epoch - 109us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.4411 - 591us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1674 - 578us/epoch - 9us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.7672 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.3480 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.7952 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.1321 - 2ms/epoch - 31us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.9432 - 793us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.5268 - 3ms/epoch - 43us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.1951 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.6887 - 580us/epoch - 9us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.3754 - 767us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.7480 - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.9979 - 786us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.4205 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2128 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.8905 - 924us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45.1348 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.7486 - 1ms/epoch - 19us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6718 - 651us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.8875 - 786us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2010 - 765us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8635 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.2426 - 722us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.6672 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.4402 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.0706 - 588us/epoch - 9us/sample\n", - "Generation 212/300\n", - "Solving for Nash Equilibrium in Generation 212/300\n", + "Generation 53/300\n", + "Solving for Nash Equilibrium in Generation 53/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.2958 - 776us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.1237 - 717us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.3558 - 710us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.4721 - 3ms/epoch - 47us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.3207 - 763us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.9713 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9216 - 998us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.4315 - 678us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.6689 - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.2060 - 742us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.2628 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.2148 - 1ms/epoch - 18us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 229.7730 - 1ms/epoch - 21us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3463 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.4126 - 2ms/epoch - 33us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.0925 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.8986 - 1ms/epoch - 20us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.4222 - 777us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3793 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.8339 - 683us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.3604 - 638us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3690 - 1ms/epoch - 17us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.3162 - 801us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.4500 - 892us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 127us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.8300 - 2ms/epoch - 28us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.2248 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9759 - 875us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.1284 - 944us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.8035 - 2ms/epoch - 30us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.2078 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 21ms/epoch - 331us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.8988 - 995us/epoch - 16us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.8912 - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.5514 - 1ms/epoch - 17us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.0950 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.1961 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.7855 - 2ms/epoch - 37us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.7570 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.9122 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 91us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3679 - 5ms/epoch - 81us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.6299 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.8878 - 617us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.7574 - 837us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.6644 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.5030 - 1ms/epoch - 20us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.7150 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2369 - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.0816 - 735us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.3417 - 785us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.6923 - 915us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.5957 - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.1259 - 581us/epoch - 9us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.3906 - 2ms/epoch - 29us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.4570 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 432.8680 - 2ms/epoch - 28us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.5023 - 887us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.1386 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.9289 - 883us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7270 - 791us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.1497 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.6050 - 628us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.3098 - 769us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.5316 - 2ms/epoch - 26us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 213/300\n", - "Solving for Nash Equilibrium in Generation 213/300\n", + "Generation 54/300\n", + "Solving for Nash Equilibrium in Generation 54/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.0156 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.2772 - 686us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.0272 - 675us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 153.8040 - 2ms/epoch - 32us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.0865 - 819us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.9520 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.0978 - 789us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.7818 - 695us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.9368 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.1564 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.1959 - 3ms/epoch - 51us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.4672 - 586us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.8367 - 892us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.3221 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.0366 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.0593 - 899us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.3922 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6163 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3427 - 985us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.6343 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.0298 - 792us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1019 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1352 - 723us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.2501 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6323 - 711us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.6915 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.6456 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8977 - 878us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.6094 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.0261 - 824us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.1553 - 808us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.2933 - 686us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.3054 - 3ms/epoch - 44us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.9956 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.3644 - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.4500 - 984us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6432 - 1ms/epoch - 22us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.6995 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.0396 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.5083 - 814us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.6165 - 695us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6484 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2104 - 851us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.5370 - 879us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.8585 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.0323 - 2ms/epoch - 27us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.8911 - 677us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.6661 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.7561 - 617us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.6759 - 695us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4059 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.5079 - 680us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.1760 - 702us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5645 - 872us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.0404 - 710us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.9229 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.2079 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 125us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.7414 - 1ms/epoch - 21us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8644 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2419 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9296 - 3ms/epoch - 49us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2542 - 749us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.0128 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6502 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.5374 - 610us/epoch - 10us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 214/300\n", - "Solving for Nash Equilibrium in Generation 214/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.5349 - 633us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.0974 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8079 - 752us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.9989 - 916us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.9904 - 716us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.1799 - 1ms/epoch - 21us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.5349 - 1ms/epoch - 19us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 125.9194 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7438 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.0084 - 884us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0229 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.0051 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.5065 - 2ms/epoch - 37us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.5360 - 2ms/epoch - 31us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.7724 - 2ms/epoch - 32us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.4351 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.0364 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.6513 - 5ms/epoch - 88us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.6739 - 3ms/epoch - 43us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.7475 - 791us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.5461 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.3010 - 727us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.6960 - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.5697 - 664us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.1207 - 599us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.7233 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.1684 - 905us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.0016 - 800us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4083 - 918us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8053 - 831us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.9155 - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.3070 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.6451 - 576us/epoch - 9us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.6213 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.9046 - 3ms/epoch - 53us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8922 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1809 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.0297 - 773us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.3032 - 727us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 144.8216 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0440 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.9608 - 718us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.0071 - 723us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9686 - 805us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.3962 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5297 - 2ms/epoch - 31us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.4929 - 843us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4251 - 652us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.0552 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.7537 - 720us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.4768 - 934us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.2463 - 649us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.1025 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231.9861 - 692us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.6665 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.6355 - 2ms/epoch - 33us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.8168 - 806us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7265 - 677us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.7146 - 836us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.7149 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.1972 - 586us/epoch - 9us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0238 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.0974 - 639us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.1104 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.7628 - 3ms/epoch - 45us/sample\n", - "Generation 215/300\n", - "Solving for Nash Equilibrium in Generation 215/300\n", + "Generation 55/300\n", + "Solving for Nash Equilibrium in Generation 55/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.8836 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.1216 - 1ms/epoch - 18us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.3112 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.0606 - 796us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 156us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.1297 - 7ms/epoch - 113us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.3773 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.5456 - 1ms/epoch - 23us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.2273 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.0648 - 801us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.1281 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.6126 - 781us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.8809 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.6555 - 645us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.0709 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.6817 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.8043 - 3ms/epoch - 45us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.7299 - 1ms/epoch - 17us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7055 - 751us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.0233 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.4372 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.7024 - 1ms/epoch - 19us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2799 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.1775 - 698us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4921 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.6552 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2704 - 868us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.1255 - 842us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.7249 - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729.2626 - 2ms/epoch - 29us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.0908 - 4ms/epoch - 66us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.2879 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.1704 - 2ms/epoch - 27us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.2136 - 951us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3189 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.9609 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.6427 - 887us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.6538 - 757us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1851 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.2786 - 819us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9016 - 777us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.6548 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.1608 - 868us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.7806 - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.4195 - 971us/epoch - 16us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.7839 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.4871 - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 84us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.3654 - 2ms/epoch - 38us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.1617 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.2221 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.9733 - 4ms/epoch - 65us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.2789 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 149us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7743 - 1ms/epoch - 19us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3376 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.2671 - 895us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.1682 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.1832 - 1ms/epoch - 21us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.8998 - 1ms/epoch - 20us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.9998 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.8689 - 743us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.4368 - 731us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.2632 - 862us/epoch - 14us/sample\n", - "Generation 216/300\n", - "Solving for Nash Equilibrium in Generation 216/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.9541 - 803us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.6959 - 1ms/epoch - 16us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.0576 - 1ms/epoch - 19us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.2902 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.5406 - 1ms/epoch - 18us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.3697 - 677us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.2604 - 2ms/epoch - 27us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.7634 - 838us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.2444 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.5298 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.1617 - 14ms/epoch - 233us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7769 - 788us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.7009 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.6695 - 741us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2032 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.3589 - 901us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.8334 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.7646 - 1ms/epoch - 18us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.4539 - 894us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.3680 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.2894 - 822us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.6682 - 727us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.9364 - 732us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2508 - 755us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1655 - 708us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.6844 - 2ms/epoch - 30us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4205 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.4130 - 2ms/epoch - 26us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.7226 - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.2367 - 6ms/epoch - 94us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.4137 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.9311 - 874us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.3656 - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.6945 - 769us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.5501 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4693 - 766us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3664 - 729us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.9960 - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.6795 - 937us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.9583 - 703us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.8931 - 647us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3798 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.2163 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.2240 - 603us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1472 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.9894 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9222 - 918us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.8619 - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.3098 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.9083 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.9329 - 796us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.9777 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.7551 - 571us/epoch - 9us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.2024 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.4319 - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.9243 - 992us/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.6344 - 835us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.4782 - 735us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8246 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4513 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.5547 - 816us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.1258 - 2ms/epoch - 30us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.3531 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.2639 - 2ms/epoch - 32us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.8488 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6175 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2104 - 910us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.9248 - 717us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1526 - 741us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.0347 - 909us/epoch - 15us/sample\n", - "Generation 217/300\n", - "Solving for Nash Equilibrium in Generation 217/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0161 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232.3024 - 772us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9863 - 1ms/epoch - 22us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2036 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.3339 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.2179 - 718us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.8171 - 728us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4410 - 759us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.2351 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.8275 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.5759 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4177 - 896us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.9645 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.5756 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.3176 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.1236 - 1ms/epoch - 17us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.0592 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773.3005 - 661us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.4897 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.6696 - 970us/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.7859 - 755us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.2951 - 2ms/epoch - 28us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5524 - 741us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.6952 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.2438 - 2ms/epoch - 37us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.5325 - 689us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.9078 - 712us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.0788 - 754us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7842 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6339 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7429 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5921 - 657us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5426 - 833us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.4332 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.1814 - 703us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.5467 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4071 - 682us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.3079 - 708us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.7520 - 872us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.2511 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9300 - 896us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0820 - 911us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.4319 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.7653 - 715us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.9218 - 671us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.1491 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.9962 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.9308 - 674us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.1902 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.3065 - 1ms/epoch - 18us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.5150 - 640us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 118.9276 - 835us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.3950 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.5629 - 643us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.5806 - 678us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.1413 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4640 - 703us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.0122 - 699us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.8822 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.3931 - 655us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774.3532 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.4218 - 561us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.6895 - 724us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.9416 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.3567 - 8ms/epoch - 135us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.8807 - 797us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.3691 - 593us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.8378 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.4451 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.2941 - 978us/epoch - 16us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8884 - 995us/epoch - 16us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.5486 - 716us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250.8086 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.4767 - 15ms/epoch - 247us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.8955 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.6400 - 763us/epoch - 12us/sample\n", - "Generation 218/300\n", - "Solving for Nash Equilibrium in Generation 218/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.4742 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.0701 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.9796 - 724us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.9423 - 767us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.1719 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.3233 - 2ms/epoch - 26us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.4187 - 754us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.3607 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.7178 - 711us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0057 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2477 - 1ms/epoch - 18us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.4187 - 762us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.0507 - 775us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4049 - 689us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245.3589 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6329 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.2921 - 731us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6093 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4610 - 694us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.0879 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.8004 - 702us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.6618 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.3775 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.5397 - 784us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.5772 - 2ms/epoch - 39us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.4337 - 679us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.8236 - 686us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.5316 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.2609 - 670us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.3483 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.6649 - 735us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.4036 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.5766 - 849us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.5276 - 726us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.0809 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.8840 - 2ms/epoch - 38us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5015 - 791us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.8717 - 1ms/epoch - 17us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.7470 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 115us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.1426 - 2ms/epoch - 35us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.1154 - 2ms/epoch - 32us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.6559 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.3445 - 612us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1909 - 887us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.6520 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.4068 - 1ms/epoch - 16us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.8148 - 1ms/epoch - 23us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.9998 - 682us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.1490 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.9222 - 735us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.7932 - 734us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.9227 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 219.7552 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0834 - 843us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.5232 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.5964 - 592us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0385 - 632us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3463 - 733us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.4140 - 641us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.0695 - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.8912 - 753us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.3720 - 726us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.5492 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.5220 - 914us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.9796 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.5607 - 3ms/epoch - 42us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.3374 - 716us/epoch - 12us/sample\n", - "Generation 219/300\n", - "Solving for Nash Equilibrium in Generation 219/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6815 - 2ms/epoch - 30us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 743.8412 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.3347 - 624us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 219.9781 - 723us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.1174 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.2122 - 1ms/epoch - 17us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.9532 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.4438 - 905us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5145 - 739us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1743 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.1725 - 809us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.4023 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4457 - 2ms/epoch - 33us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.5315 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.3550 - 578us/epoch - 9us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 100us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.4512 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.6204 - 694us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.7028 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.8371 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.4641 - 651us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.0611 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.2819 - 636us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.5186 - 623us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.4842 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8147 - 763us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 143.2776 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.4010 - 931us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.6875 - 890us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5515 - 667us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6470 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3386 - 587us/epoch - 9us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.2029 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.6078 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.8992 - 960us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.9779 - 900us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.5115 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.2709 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.9622 - 634us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.2284 - 2ms/epoch - 30us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.1534 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.2305 - 700us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.0157 - 689us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.8104 - 548us/epoch - 9us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.8732 - 707us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.3149 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.5462 - 715us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.9088 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.9303 - 701us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.4446 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.8170 - 2ms/epoch - 30us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.0242 - 687us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.3386 - 600us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.3134 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.1900 - 733us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2770 - 793us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.2416 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.7483 - 625us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.8235 - 670us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.8907 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.8768 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.8079 - 670us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.6849 - 641us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.9787 - 667us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.7226 - 836us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.2875 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.6892 - 822us/epoch - 13us/sample\n", - "Generation 220/300\n", - "Solving for Nash Equilibrium in Generation 220/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 56/300\n", + "Solving for Nash Equilibrium in Generation 56/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.6744 - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 362.2310 - 2ms/epoch - 32us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.9239 - 747us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.2057 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.4726 - 932us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.0745 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.8076 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232.1271 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.1663 - 713us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.6568 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.1447 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.8477 - 634us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.9413 - 726us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.0336 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.6639 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.2510 - 797us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5492 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.5477 - 695us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.6330 - 702us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.9603 - 773us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.6808 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.9056 - 713us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.7601 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.9778 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759.8405 - 722us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.0260 - 777us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.9893 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.5752 - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.8688 - 3ms/epoch - 50us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1480 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4233 - 3ms/epoch - 45us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6877 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.8665 - 659us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.7713 - 586us/epoch - 9us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.4268 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.7967 - 916us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6723 - 727us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.9504 - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.5646 - 731us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714.1559 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6299 - 675us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.3480 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.1987 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8309 - 597us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.1935 - 653us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4431 - 3ms/epoch - 55us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.7298 - 703us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.0798 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.1107 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3997 - 775us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.6664 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.9452 - 835us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.3016 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.8393 - 687us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.3539 - 697us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.4848 - 592us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.5525 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.9927 - 670us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.5026 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.0450 - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.9677 - 1ms/epoch - 17us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.4053 - 794us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6688 - 715us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.6215 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762.9026 - 683us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.2083 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.8585 - 724us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 221/300\n", - "Solving for Nash Equilibrium in Generation 221/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.3435 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.4653 - 710us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.4922 - 909us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781.5418 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.4796 - 848us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.8122 - 656us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.8460 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.1014 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4381 - 678us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.6437 - 867us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.0213 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.9073 - 906us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.9510 - 667us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.2279 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.8997 - 778us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.1732 - 939us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7973 - 594us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.3539 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3476 - 684us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.0216 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.0865 - 601us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9269 - 756us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.3859 - 712us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.8516 - 776us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3621 - 648us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51.7496 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.9785 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.3967 - 729us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4930 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.5686 - 1ms/epoch - 16us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.0004 - 808us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.3243 - 2ms/epoch - 29us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 253.9309 - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.7790 - 606us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0297 - 673us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1215 - 654us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2245 - 1ms/epoch - 18us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.5661 - 588us/epoch - 9us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.0587 - 632us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3940 - 871us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50.0413 - 605us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1820 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5928 - 783us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.0327 - 750us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0851 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.9379 - 11ms/epoch - 182us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0892 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7504 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 159us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.3652 - 1ms/epoch - 20us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.3065 - 563us/epoch - 9us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.1971 - 731us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230.1911 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.9012 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9701 - 840us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.6817 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.9433 - 703us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.3308 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.5439 - 677us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.4089 - 789us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.3483 - 680us/epoch - 11us/sample\n", - "Generation 222/300\n", - "Solving for Nash Equilibrium in Generation 222/300\n", + "Generation 57/300\n", + "Solving for Nash Equilibrium in Generation 57/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.3640 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.3698 - 613us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.5299 - 716us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.7655 - 702us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.2150 - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.1600 - 1ms/epoch - 18us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.2369 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 780.2617 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.1709 - 768us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.4559 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.2358 - 778us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0043 - 756us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4674 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2301 - 2ms/epoch - 29us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.1950 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.4096 - 750us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.4576 - 744us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.3724 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.1975 - 727us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.8579 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.9765 - 598us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.2262 - 639us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.2844 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2998 - 750us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.4519 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.4923 - 676us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.1514 - 728us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.1350 - 779us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.4077 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.8079 - 708us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.5240 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.6606 - 612us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.1170 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.2406 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.0285 - 956us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.9499 - 719us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.8110 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.1122 - 780us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8634 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.6221 - 698us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4797 - 736us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.1992 - 714us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5851 - 1ms/epoch - 19us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6736 - 2ms/epoch - 29us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.9545 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.9429 - 865us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.7768 - 713us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.0437 - 2ms/epoch - 26us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.3403 - 6ms/epoch - 99us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.6639 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.1852 - 601us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.8058 - 622us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.3468 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.1406 - 744us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.1311 - 695us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.8396 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.6610 - 751us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7628 - 770us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.1464 - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.3901 - 3ms/epoch - 47us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.5693 - 815us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 135.2644 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.7382 - 562us/epoch - 9us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.5587 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.3307 - 672us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.6159 - 812us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0153 - 1ms/epoch - 18us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.9654 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.5262 - 763us/epoch - 12us/sample\n", - "Generation 223/300\n", - "Solving for Nash Equilibrium in Generation 223/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.5355 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.0336 - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.4988 - 994us/epoch - 16us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.6855 - 726us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.9724 - 802us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.3196 - 728us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.5446 - 2ms/epoch - 25us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 163.9953 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.5829 - 792us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214.2481 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.4641 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.2316 - 1ms/epoch - 22us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.1211 - 1ms/epoch - 18us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.1549 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 19ms/epoch - 305us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.4766 - 14ms/epoch - 220us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.8693 - 809us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 11ms/epoch - 174us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.4706 - 8ms/epoch - 128us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7769 - 829us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.2735 - 742us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.6470 - 761us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.2043 - 789us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.4200 - 937us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.0469 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.0311 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.7888 - 832us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235.1407 - 720us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.2096 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.0081 - 975us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.5748 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.0510 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.8143 - 687us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5369 - 679us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6301 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1195 - 590us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.7708 - 703us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.7369 - 622us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.2399 - 821us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.1922 - 2ms/epoch - 35us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.8502 - 785us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.6317 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.5800 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.8203 - 1ms/epoch - 21us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.3976 - 669us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6238 - 781us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1133 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.1294 - 646us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.6653 - 1ms/epoch - 19us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.3647 - 841us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.4495 - 692us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.2711 - 1ms/epoch - 23us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.3119 - 794us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2092 - 753us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8776 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.8005 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8715 - 1ms/epoch - 20us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.9280 - 700us/epoch - 11us/sample\n", - "Generation 224/300\n", - "Solving for Nash Equilibrium in Generation 224/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 58/300\n", + "Solving for Nash Equilibrium in Generation 58/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7582 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.7251 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.5159 - 4ms/epoch - 57us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.1850 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 217.8327 - 606us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.3586 - 798us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.1265 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.7954 - 894us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.1408 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.2592 - 1ms/epoch - 21us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1827 - 2ms/epoch - 31us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.3316 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.4057 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.5544 - 608us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.0776 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.2703 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.5690 - 866us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.1033 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.7882 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.8305 - 1ms/epoch - 24us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.5952 - 670us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3458 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0217 - 571us/epoch - 9us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.2703 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.4220 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.4015 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0958 - 1ms/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.8650 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0544 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.9566 - 743us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0001 - 655us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.6122 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3041 - 929us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.3546 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.3934 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.1442 - 972us/epoch - 16us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.0735 - 6ms/epoch - 93us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2374 - 993us/epoch - 16us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.3810 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.0309 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.9266 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.2457 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0296 - 744us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.2288 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.0137 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.1995 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.7056 - 1ms/epoch - 21us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.9172 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.2179 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9316 - 705us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.8174 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4333 - 678us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6857 - 685us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2404 - 678us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9796 - 905us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7099 - 2ms/epoch - 38us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38.6979 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.8454 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.3432 - 775us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.8079 - 658us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.5895 - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.5969 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306.0262 - 630us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.7427 - 643us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6078 - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.3596 - 629us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.0534 - 812us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 225/300\n", - "Solving for Nash Equilibrium in Generation 225/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.9925 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.8283 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.3761 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.0749 - 732us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.8559 - 3ms/epoch - 43us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.2785 - 686us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.0556 - 596us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.3635 - 960us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.2578 - 670us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.5051 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.7195 - 661us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5817 - 731us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.9589 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.0385 - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 229.9035 - 12ms/epoch - 197us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.3945 - 699us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.1480 - 4ms/epoch - 60us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.1384 - 705us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.8909 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9934 - 686us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.8945 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8954 - 592us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.3448 - 10ms/epoch - 162us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 138us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.8595 - 990us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.9904 - 706us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.4623 - 1ms/epoch - 16us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9068 - 648us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.1783 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.9386 - 999us/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 91us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.9435 - 1ms/epoch - 17us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.1915 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 435.6305 - 606us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.4812 - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.3403 - 751us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.8151 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8821 - 6ms/epoch - 91us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.5993 - 971us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.9132 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.9662 - 636us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.2805 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.9402 - 840us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.0714 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737.6226 - 869us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.8799 - 949us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.5022 - 5ms/epoch - 84us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.2951 - 8ms/epoch - 136us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.0101 - 736us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40.6173 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.1499 - 701us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.5952 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.2843 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 74ms/epoch - 1ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.0615 - 8ms/epoch - 135us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.4177 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.1335 - 777us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7366 - 910us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.9217 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.8771 - 724us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6616 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2935 - 2ms/epoch - 28us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.6099 - 635us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.3317 - 1ms/epoch - 21us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.3122 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 203.2938 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.0167 - 745us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1110 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.4160 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.5939 - 859us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.3866 - 669us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.3415 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2128 - 3ms/epoch - 41us/sample\n", - "Generation 226/300\n", - "Solving for Nash Equilibrium in Generation 226/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.7172 - 701us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.9232 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1365 - 692us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.2149 - 2ms/epoch - 26us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5448 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361.8314 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.1208 - 730us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.3775 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.0977 - 770us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6968 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 14ms/epoch - 230us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.4572 - 1ms/epoch - 16us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.6718 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.4071 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.5764 - 745us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5795 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.7278 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.9177 - 742us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.6619 - 705us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.6102 - 824us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7185 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8993 - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.9633 - 2ms/epoch - 30us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0258 - 1ms/epoch - 23us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.5661 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 15ms/epoch - 234us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.0570 - 3ms/epoch - 49us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5152 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.6631 - 811us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.5726 - 697us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.4562 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.8000 - 4ms/epoch - 65us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.9081 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.3733 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.3284 - 681us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.1403 - 757us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.0693 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.9703 - 2ms/epoch - 28us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.5717 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9462 - 646us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.9016 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.3050 - 611us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.1809 - 805us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.9010 - 719us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.3854 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.1250 - 697us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.5477 - 728us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.6423 - 723us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.7989 - 1ms/epoch - 20us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.8569 - 707us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.9066 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.0960 - 883us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.0770 - 743us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.5512 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.7086 - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.0643 - 885us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.4116 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8070 - 611us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.9212 - 742us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.7664 - 688us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5093 - 663us/epoch - 11us/sample\n", - "Generation 227/300\n", - "Solving for Nash Equilibrium in Generation 227/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.0990 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 99us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.2278 - 2ms/epoch - 24us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.3044 - 891us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.0310 - 731us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 215.0756 - 695us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.8835 - 611us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.9490 - 1ms/epoch - 17us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.5638 - 557us/epoch - 9us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.9156 - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.5945 - 638us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.2571 - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.5919 - 865us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.0984 - 704us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.2894 - 3ms/epoch - 46us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8422 - 1ms/epoch - 23us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.5063 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.1983 - 4ms/epoch - 63us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.0387 - 722us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2247 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.8711 - 2ms/epoch - 28us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.6406 - 1ms/epoch - 18us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.6989 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.4375 - 898us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.9171 - 2ms/epoch - 25us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.2122 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9930 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.3984 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.0264 - 2ms/epoch - 26us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.5600 - 690us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.0105 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.5186 - 604us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.0320 - 556us/epoch - 9us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.0060 - 1ms/epoch - 19us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6666 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3195 - 699us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.2187 - 743us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.0735 - 691us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.7272 - 719us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 202.0417 - 2ms/epoch - 25us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0684 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.4480 - 609us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.9596 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.1837 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.3082 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.4149 - 901us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.4333 - 761us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2527 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.0021 - 728us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.6174 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.9978 - 704us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.2408 - 686us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.7104 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.0858 - 584us/epoch - 9us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4878 - 603us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.5732 - 690us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.9006 - 615us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.9738 - 5ms/epoch - 74us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.2800 - 600us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235.0412 - 718us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.5780 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.2808 - 749us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.6849 - 677us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2186 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.8387 - 668us/epoch - 11us/sample\n", - "Generation 228/300\n", - "Solving for Nash Equilibrium in Generation 228/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.8356 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.0685 - 744us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.4564 - 605us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5937 - 834us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.9158 - 739us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.0430 - 726us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.0141 - 849us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4682 - 810us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.5524 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.2631 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7998 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.4038 - 817us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.0708 - 731us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.7590 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.9083 - 1ms/epoch - 20us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4088 - 688us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9778 - 684us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2623 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3599 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.4100 - 1ms/epoch - 23us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3051 - 589us/epoch - 9us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.3829 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.7712 - 914us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.2118 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.6116 - 688us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.3016 - 680us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.0273 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.4008 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.9460 - 713us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.1655 - 8ms/epoch - 129us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.3332 - 735us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.5191 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.0414 - 833us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.8397 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.8624 - 988us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.2561 - 1ms/epoch - 21us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7924 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.5150 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.0328 - 2ms/epoch - 35us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.1953 - 770us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.1626 - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3910 - 684us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.1058 - 714us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.2831 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.5354 - 586us/epoch - 9us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.9544 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.1531 - 836us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.7233 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.6650 - 544us/epoch - 9us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9057 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.7376 - 751us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.4543 - 734us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8118 - 904us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.6774 - 692us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.0405 - 636us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.8812 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3952 - 688us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3715 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.6534 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.4003 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.6655 - 2ms/epoch - 26us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.0638 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.4720 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.4432 - 1ms/epoch - 21us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.9933 - 577us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9136 - 861us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2989 - 851us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4246 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.1940 - 732us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.6526 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.6074 - 754us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.3080 - 686us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.8859 - 699us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.7531 - 6ms/epoch - 93us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.0262 - 789us/epoch - 13us/sample\n", - "Generation 229/300\n", - "Solving for Nash Equilibrium in Generation 229/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.3849 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.9977 - 763us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.8192 - 826us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.5493 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.0053 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.4237 - 592us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.0233 - 625us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.4979 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.4613 - 668us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.2090 - 703us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.8953 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428.5670 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.3331 - 780us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.9200 - 757us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.9643 - 743us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.2376 - 719us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.7792 - 717us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.2576 - 890us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.8636 - 741us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.7761 - 787us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 73.0120 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.3809 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6100 - 569us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.5430 - 6ms/epoch - 90us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.9420 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.3988 - 824us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.3033 - 828us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.6151 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2726 - 2ms/epoch - 36us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.2621 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.4776 - 768us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1053 - 729us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.6095 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.8120 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6301 - 643us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.1885 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.4482 - 652us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3761 - 600us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.2841 - 855us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.3245 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.9883 - 753us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.9848 - 967us/epoch - 16us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.1363 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.3353 - 872us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738.5806 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4099 - 603us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.0854 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.0984 - 570us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230.2642 - 695us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7129 - 713us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.6465 - 697us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5895 - 5ms/epoch - 89us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.0527 - 982us/epoch - 16us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.1018 - 6ms/epoch - 99us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.9223 - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.8361 - 707us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.4445 - 831us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.2664 - 810us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.0025 - 1ms/epoch - 22us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.1952 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.1012 - 663us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.9056 - 698us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.5368 - 776us/epoch - 13us/sample\n", - "Generation 230/300\n", - "Solving for Nash Equilibrium in Generation 230/300\n", + "Generation 59/300\n", + "Solving for Nash Equilibrium in Generation 59/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.3174 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.6496 - 1ms/epoch - 17us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.1273 - 869us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 113.8790 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8121 - 787us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.9242 - 3ms/epoch - 44us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.7914 - 691us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.3129 - 551us/epoch - 9us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.4069 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2433 - 1ms/epoch - 16us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.5451 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.1590 - 2ms/epoch - 29us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.0756 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7188 - 753us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.8758 - 495us/epoch - 8us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2937 - 689us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.4287 - 1ms/epoch - 18us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.5328 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.8784 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.3546 - 762us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.2874 - 2ms/epoch - 28us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.2987 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232.8616 - 743us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.4708 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.0726 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.0443 - 768us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.2319 - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.5775 - 2ms/epoch - 39us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.8500 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3992 - 736us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.7553 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.3284 - 2ms/epoch - 33us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.5574 - 850us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.6743 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9457 - 721us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.8106 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.7383 - 624us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.8926 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.1121 - 743us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.2383 - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.5399 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5154 - 783us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.2621 - 975us/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.3701 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8178 - 725us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.8004 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.3030 - 805us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.2586 - 610us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5282 - 2ms/epoch - 29us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.9452 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.9660 - 2ms/epoch - 26us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.9662 - 624us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.3278 - 578us/epoch - 9us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.9264 - 2ms/epoch - 39us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.5422 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.5047 - 647us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.2118 - 645us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.5250 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1924 - 786us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.0747 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.1632 - 739us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.6576 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.3709 - 661us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.8227 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.6060 - 727us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.7906 - 763us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.0748 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.7815 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.9137 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8708 - 695us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.4581 - 823us/epoch - 13us/sample\n", - "Generation 231/300\n", - "Solving for Nash Equilibrium in Generation 231/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.3575 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.9444 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.9466 - 750us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6722 - 721us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.3608 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.6173 - 824us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.3161 - 768us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.5193 - 640us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.4864 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.2202 - 1ms/epoch - 18us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.8529 - 733us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.9200 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.8198 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.9917 - 857us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.8085 - 865us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7142 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.7062 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.7863 - 947us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.8733 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.7479 - 1ms/epoch - 17us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.1531 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.0114 - 810us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8292 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.8775 - 2ms/epoch - 25us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.7268 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.3631 - 725us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.4387 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.4026 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.1268 - 812us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.9128 - 780us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.1036 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.3931 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.5403 - 629us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.2644 - 966us/epoch - 16us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.3492 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.0188 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6184 - 713us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.8904 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.9174 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.6707 - 669us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.1727 - 2ms/epoch - 31us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.5154 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.0803 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.7108 - 631us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.4142 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.4758 - 720us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0046 - 790us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.5540 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.0923 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.9391 - 793us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.1807 - 852us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.8071 - 891us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.7963 - 916us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.9904 - 824us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.1737 - 792us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.4665 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.7246 - 724us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.2147 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.6304 - 760us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5466 - 698us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.6866 - 727us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.6126 - 698us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.8376 - 809us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.3200 - 767us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.3777 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.1246 - 919us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.9683 - 1ms/epoch - 23us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.4539 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.2907 - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.6050 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.0062 - 650us/epoch - 10us/sample\n", - "Generation 232/300\n", - "Solving for Nash Equilibrium in Generation 232/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.0319 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8707 - 909us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.2722 - 2ms/epoch - 29us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.0614 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0378 - 709us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.3036 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3970 - 611us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.6988 - 787us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.5076 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.6531 - 705us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.0399 - 825us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.3704 - 697us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.1462 - 706us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.8939 - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.1793 - 1ms/epoch - 21us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.7194 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.7048 - 737us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.2082 - 11ms/epoch - 183us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.2453 - 660us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.3021 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.7400 - 664us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.7628 - 705us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406.5177 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.8551 - 616us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.9310 - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.6265 - 813us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.0082 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.2253 - 951us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.2460 - 641us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.4860 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.7420 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.4570 - 787us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.7489 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.6430 - 609us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.7073 - 805us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.0080 - 916us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9957 - 586us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.7364 - 819us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.7606 - 937us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.2172 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.0500 - 683us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.9940 - 732us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.7455 - 1ms/epoch - 20us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.0605 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.3082 - 744us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8851 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.1321 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.5842 - 672us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.8828 - 754us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.6371 - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.9432 - 797us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.5524 - 905us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.2840 - 3ms/epoch - 48us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.2453 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.4657 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.2408 - 799us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.0318 - 699us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.7914 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.2314 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.3682 - 816us/epoch - 13us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 233/300\n", - "Solving for Nash Equilibrium in Generation 233/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6258 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7282 - 714us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.0487 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.2802 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.6736 - 1ms/epoch - 19us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.7249 - 1ms/epoch - 18us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.8334 - 821us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.2711 - 909us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.2833 - 761us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.6378 - 1ms/epoch - 22us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.5899 - 1ms/epoch - 22us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.3279 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5891 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.6258 - 1ms/epoch - 17us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.3141 - 7ms/epoch - 120us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.5236 - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.7127 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.0977 - 1ms/epoch - 18us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.6731 - 776us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6825 - 873us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.8153 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.0634 - 738us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.8900 - 691us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.0151 - 1ms/epoch - 19us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.1834 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.3691 - 853us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.1773 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.2660 - 701us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.7677 - 1ms/epoch - 23us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.4272 - 784us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.6722 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5951 - 625us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.9031 - 594us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.2211 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1730 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.0119 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.2853 - 873us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.5072 - 791us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.0869 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.0529 - 859us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6496 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.4555 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8847 - 863us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.6706 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.5735 - 759us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.2615 - 696us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.1227 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4592 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.3592 - 5ms/epoch - 74us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.2098 - 618us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.2013 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5417 - 742us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.9006 - 953us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.5884 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7260 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.8538 - 572us/epoch - 9us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.7328 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.7245 - 756us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.3953 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6822 - 725us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.7722 - 616us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.1874 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.0947 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.0754 - 934us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.1100 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1887 - 673us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.2133 - 744us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.9240 - 829us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.1151 - 732us/epoch - 12us/sample\n", - "Generation 234/300\n", - "Solving for Nash Equilibrium in Generation 234/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.6715 - 1ms/epoch - 19us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6385 - 599us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.3661 - 577us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.8967 - 891us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.4925 - 603us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.1774 - 900us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.4268 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.6092 - 918us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.7432 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.7621 - 596us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.1303 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.2764 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.5063 - 823us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.3340 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3754 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.5961 - 2ms/epoch - 30us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.0079 - 643us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.3577 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.2128 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.1855 - 801us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8161 - 701us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.8965 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.3655 - 676us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.3481 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.3668 - 822us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.6868 - 687us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.2259 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.7843 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3162 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.4415 - 837us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0126 - 660us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.9417 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.7606 - 755us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4337 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.3475 - 634us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.9073 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.5724 - 639us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.7390 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.8956 - 681us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.7734 - 662us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.2745 - 848us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.2399 - 2ms/epoch - 27us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.4402 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2845 - 712us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7154 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.4801 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.5036 - 686us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.0877 - 1ms/epoch - 20us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9079 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.9279 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.5721 - 823us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.1082 - 678us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.0372 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.6852 - 670us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.0262 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.1879 - 865us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.0808 - 1ms/epoch - 24us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.4350 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.8855 - 599us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.6536 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.0484 - 644us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.2247 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763.7011 - 639us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 191.5304 - 695us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.0998 - 693us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.7865 - 927us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.6155 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.8490 - 654us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.6443 - 1ms/epoch - 17us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3521 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 209.1691 - 703us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.5974 - 725us/epoch - 12us/sample\n", - "Generation 235/300\n", - "Solving for Nash Equilibrium in Generation 235/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.1291 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.6591 - 749us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.2140 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.0255 - 811us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.6644 - 670us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.8072 - 696us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.6929 - 2ms/epoch - 39us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.9233 - 2ms/epoch - 31us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 161.6835 - 754us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.4081 - 3ms/epoch - 47us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.0842 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.5359 - 861us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.3809 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.3140 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 165us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.1934 - 5ms/epoch - 86us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.5028 - 729us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9829 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7057 - 3ms/epoch - 56us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.8107 - 878us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8360 - 1ms/epoch - 20us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.0518 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.8450 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2480 - 1ms/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232.6385 - 711us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.5917 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.7713 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.4280 - 680us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 139us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.6907 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3207 - 835us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.6906 - 4ms/epoch - 65us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4503 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.7573 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4568 - 893us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1175 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.2088 - 571us/epoch - 9us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9197 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.7838 - 832us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3109 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7595 - 660us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4194 - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.6873 - 691us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6012 - 2ms/epoch - 40us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2707 - 2ms/epoch - 24us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.7567 - 684us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.3174 - 10ms/epoch - 154us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 123us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.3635 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4335 - 707us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.6209 - 623us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.2112 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8765 - 5ms/epoch - 86us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.3740 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.6971 - 944us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.1766 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.1130 - 3ms/epoch - 47us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7547 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.7109 - 607us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.9865 - 885us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.0110 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.5692 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.1393 - 711us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.4752 - 643us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.8206 - 686us/epoch - 11us/sample\n", - "Generation 236/300\n", - "Solving for Nash Equilibrium in Generation 236/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2191 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41.1982 - 816us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.9669 - 636us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.4479 - 1ms/epoch - 20us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.0807 - 767us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.2157 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.3916 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1626 - 695us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.0815 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 253.0822 - 747us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.5885 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.0625 - 770us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1042 - 1ms/epoch - 18us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.1895 - 822us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.4370 - 1ms/epoch - 21us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9584 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.5195 - 585us/epoch - 9us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.1840 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.6629 - 723us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.0558 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8013 - 748us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.6752 - 542us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9787 - 2ms/epoch - 30us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.5223 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.4356 - 750us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.3924 - 828us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.4989 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.9800 - 785us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.1004 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.3073 - 762us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.6205 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.8781 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.8176 - 807us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0625 - 706us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.8518 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.2811 - 1ms/epoch - 23us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2828 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.5510 - 598us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.8187 - 611us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.4044 - 773us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.7676 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.5509 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1387 - 1ms/epoch - 17us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.4122 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.7800 - 1ms/epoch - 18us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.6249 - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.1336 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.7209 - 807us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.9691 - 556us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.2123 - 851us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.8672 - 739us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.5199 - 2ms/epoch - 31us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.6647 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.6924 - 715us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.7200 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.1544 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.0769 - 2ms/epoch - 28us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.0269 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3385 - 657us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.8052 - 7ms/epoch - 117us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6772 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.2179 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.3352 - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.4822 - 1ms/epoch - 17us/sample\n", - "Generation 237/300\n", - "Solving for Nash Equilibrium in Generation 237/300\n", + "Generation 60/300\n", + "Solving for Nash Equilibrium in Generation 60/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8630 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.6648 - 653us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.7838 - 919us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.5330 - 9ms/epoch - 143us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.0654 - 729us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43.9076 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.1698 - 1ms/epoch - 23us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.9337 - 779us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.9170 - 598us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.2838 - 721us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.5854 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.3236 - 5ms/epoch - 82us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116.5159 - 650us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.8674 - 746us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.2189 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5482 - 1ms/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.4189 - 938us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.1873 - 586us/epoch - 9us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.2446 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.4321 - 653us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361.1407 - 7ms/epoch - 106us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.4442 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.0264 - 835us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.6592 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.5193 - 589us/epoch - 9us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.2079 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.3307 - 799us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.9839 - 645us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.1425 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9953 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.1321 - 688us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.0054 - 717us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.8507 - 3ms/epoch - 40us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.6236 - 693us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 78.0856 - 622us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.3093 - 722us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.9106 - 686us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.3918 - 603us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 67.5230 - 577us/epoch - 9us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.2847 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.5311 - 638us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.1131 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.5579 - 698us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4739 - 752us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.2520 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6707 - 2ms/epoch - 31us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.7359 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.4250 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.9860 - 853us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.2585 - 799us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.0875 - 742us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 238/300\n", - "Solving for Nash Equilibrium in Generation 238/300\n", + "Generation 61/300\n", + "Solving for Nash Equilibrium in Generation 61/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.1899 - 713us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8641 - 4ms/epoch - 64us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.6049 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.8092 - 901us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.3216 - 699us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.3324 - 753us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.0214 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.3904 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.1156 - 662us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6296 - 641us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.7401 - 707us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.8596 - 802us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.6249 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.0090 - 643us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.0178 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.5956 - 648us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.9076 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.8007 - 765us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.5658 - 593us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.3209 - 839us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.5666 - 939us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.0283 - 1ms/epoch - 20us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.5712 - 590us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.7621 - 894us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.9240 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.4456 - 769us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.2220 - 759us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.8068 - 1ms/epoch - 21us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.0405 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.1730 - 648us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 177.1278 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0004 - 665us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.4656 - 779us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.9960 - 3ms/epoch - 45us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.3208 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.5265 - 625us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.4701 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.7384 - 651us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.0605 - 793us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.7289 - 2ms/epoch - 27us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.3459 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.3578 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.2504 - 783us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.0981 - 726us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.7357 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.0589 - 715us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8875 - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9720 - 2ms/epoch - 28us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.1251 - 725us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.5686 - 745us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.3606 - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.2822 - 1ms/epoch - 19us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2752 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.7086 - 675us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.3941 - 642us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.5496 - 782us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.2752 - 935us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.8485 - 3ms/epoch - 41us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 239/300\n", - "Solving for Nash Equilibrium in Generation 239/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.4462 - 593us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.9710 - 1ms/epoch - 21us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.4373 - 568us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235.5947 - 1ms/epoch - 24us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5851 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.5479 - 741us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.9768 - 723us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.9868 - 682us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.6898 - 785us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.5540 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.5827 - 1ms/epoch - 22us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.1910 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.9977 - 793us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.7478 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0632 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.4814 - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.6352 - 1ms/epoch - 16us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.8883 - 729us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.0023 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6528 - 2ms/epoch - 27us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.4714 - 703us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.1270 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.0906 - 1ms/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.3094 - 874us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.7882 - 2ms/epoch - 25us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.4327 - 1ms/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.2891 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.7009 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.4099 - 1ms/epoch - 17us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45.7589 - 794us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6777 - 751us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.2590 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.9169 - 901us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0042 - 658us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.0458 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.8986 - 4ms/epoch - 71us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.5248 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.5171 - 636us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.9084 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9161 - 5ms/epoch - 83us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.6376 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.8768 - 930us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.7827 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.4027 - 11ms/epoch - 175us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3251 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5452 - 678us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.9456 - 1ms/epoch - 18us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.0900 - 1ms/epoch - 18us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.3700 - 791us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.3884 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.6136 - 808us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.3225 - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 202.5912 - 6ms/epoch - 99us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.0437 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.5351 - 976us/epoch - 16us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.1814 - 936us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.2576 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.8318 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0678 - 624us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6658 - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2795 - 897us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.5208 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.1463 - 598us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.9698 - 593us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.0983 - 665us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7927 - 686us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2140 - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.6593 - 772us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6008 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.9116 - 879us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.5298 - 626us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.9136 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.2079 - 680us/epoch - 11us/sample\n", - "Generation 240/300\n", - "Solving for Nash Equilibrium in Generation 240/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.7404 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.1591 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.5809 - 613us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.8759 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.8890 - 592us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0042 - 813us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.2846 - 664us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.0887 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.5899 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7178 - 2ms/epoch - 27us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.0931 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.5257 - 774us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.0763 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8223 - 571us/epoch - 9us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.5201 - 713us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.7801 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.7913 - 727us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.3197 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 198.8990 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.5096 - 653us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.4655 - 741us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.9276 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.9496 - 618us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.4615 - 867us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.9247 - 796us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.0632 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.2981 - 636us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.3646 - 7ms/epoch - 119us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.2840 - 709us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2564 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.3312 - 788us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.1550 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.5228 - 597us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.7326 - 1ms/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.3240 - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2748 - 657us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.4545 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.4734 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.3116 - 714us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.6231 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.4667 - 683us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.3516 - 891us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4495 - 864us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.2801 - 816us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.2369 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.6387 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.3040 - 842us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 213.4010 - 725us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.7544 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.5426 - 707us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.3352 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.8937 - 623us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4187 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0448 - 681us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5623 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.1799 - 788us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.5840 - 908us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.4233 - 953us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6339 - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.8870 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.3729 - 775us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 435.3219 - 2ms/epoch - 26us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.8342 - 2ms/epoch - 37us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.9448 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.1371 - 631us/epoch - 10us/sample\n", - "Generation 241/300\n", - "Solving for Nash Equilibrium in Generation 241/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.1452 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.2964 - 569us/epoch - 9us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.5689 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4363 - 735us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.7347 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.8763 - 861us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41.4432 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.4616 - 741us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.7328 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5930 - 1ms/epoch - 21us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.4839 - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.5786 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.5383 - 805us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1422 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.2841 - 867us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 17ms/epoch - 281us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.3819 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.5027 - 1ms/epoch - 23us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.0483 - 1ms/epoch - 17us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.6010 - 1ms/epoch - 16us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.5926 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.6052 - 777us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.7787 - 798us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.3336 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.3850 - 17ms/epoch - 270us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.3913 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.0057 - 1ms/epoch - 18us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47.2147 - 918us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.9196 - 1ms/epoch - 17us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.7429 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.4513 - 910us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.8078 - 816us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.3615 - 878us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 152us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.5287 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.2726 - 28ms/epoch - 446us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.5164 - 598us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.0860 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.4867 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1945 - 1ms/epoch - 22us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6397 - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.3415 - 732us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.0518 - 3ms/epoch - 47us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7780 - 646us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3350 - 686us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.0708 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.8783 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 31ms/epoch - 499us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.9373 - 4ms/epoch - 62us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.6694 - 783us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.2328 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.0591 - 736us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 87us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.5323 - 2ms/epoch - 24us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.8392 - 649us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.1313 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.2177 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.6915 - 866us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.1420 - 921us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.4857 - 821us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.1868 - 14ms/epoch - 224us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.3378 - 891us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.5477 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.1602 - 995us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.3335 - 915us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248.8050 - 1ms/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345.5826 - 798us/epoch - 13us/sample\n", - "Generation 242/300\n", - "Solving for Nash Equilibrium in Generation 242/300\n", + "Generation 62/300\n", + "Solving for Nash Equilibrium in Generation 62/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.0828 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.8723 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.3674 - 793us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.9272 - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6221 - 888us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.2244 - 692us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.3862 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6653 - 862us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325.8942 - 783us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.3174 - 659us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.3853 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.2168 - 1ms/epoch - 17us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.6888 - 820us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.7927 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.1993 - 684us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.1502 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.3220 - 2ms/epoch - 35us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.2368 - 811us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.2770 - 716us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.3076 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.1364 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.2117 - 19ms/epoch - 304us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.1037 - 836us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.9335 - 796us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8422 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.1709 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.4606 - 759us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8205 - 3ms/epoch - 53us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.4909 - 6ms/epoch - 101us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.9916 - 10ms/epoch - 162us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.7077 - 3ms/epoch - 43us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.4332 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.2819 - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3706 - 1ms/epoch - 17us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.5061 - 941us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.2071 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.9982 - 822us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.2583 - 804us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.7488 - 900us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.4178 - 785us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.9955 - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6733 - 730us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.6533 - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.1015 - 766us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.2336 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3641 - 1ms/epoch - 20us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.0463 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9711 - 672us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.8126 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.0960 - 930us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.5846 - 707us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.9405 - 773us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.5049 - 545us/epoch - 9us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.4515 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6903 - 888us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.0650 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.8469 - 748us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.2465 - 696us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4883 - 968us/epoch - 16us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 243/300\n", - "Solving for Nash Equilibrium in Generation 243/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.0793 - 935us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.0211 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.6854 - 7ms/epoch - 108us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.8273 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.1997 - 744us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.3705 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.8910 - 767us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.5522 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.2651 - 652us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.3153 - 990us/epoch - 16us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.1820 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.0460 - 750us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0029 - 1ms/epoch - 21us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6956 - 622us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.3167 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.0445 - 846us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.3972 - 595us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.2859 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.2028 - 1ms/epoch - 18us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.2459 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.8512 - 1ms/epoch - 20us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.2360 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 12ms/epoch - 190us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6248 - 3ms/epoch - 54us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.2866 - 883us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 16ms/epoch - 250us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.7509 - 14ms/epoch - 219us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.7097 - 673us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.2116 - 772us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.2866 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.5103 - 3ms/epoch - 49us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8496 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.3338 - 2ms/epoch - 33us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.1625 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7411 - 637us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.5382 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9304 - 650us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.2275 - 2ms/epoch - 36us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2272 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.9365 - 3ms/epoch - 48us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.2988 - 824us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5958 - 620us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.9468 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.8158 - 1ms/epoch - 18us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3047 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.1323 - 571us/epoch - 9us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.4534 - 806us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.9927 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2740 - 620us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2724 - 808us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.8270 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.6977 - 660us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7480 - 6ms/epoch - 103us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 198.2825 - 773us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.0250 - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.9172 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.9323 - 736us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.9987 - 695us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.4113 - 941us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.1614 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.0205 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.3405 - 750us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.1277 - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.2809 - 2ms/epoch - 28us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.1563 - 740us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.5161 - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6587 - 640us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.1437 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.5249 - 661us/epoch - 11us/sample\n", - "Generation 244/300\n", - "Solving for Nash Equilibrium in Generation 244/300\n", + "Generation 63/300\n", + "Solving for Nash Equilibrium in Generation 63/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5369 - 588us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9810 - 640us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4080 - 665us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.2030 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7175 - 2ms/epoch - 38us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.9459 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391.8500 - 754us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2421 - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.7847 - 692us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.2487 - 706us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3058 - 697us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.6839 - 7ms/epoch - 114us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.7125 - 709us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.1691 - 818us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.0486 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.8477 - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.7289 - 769us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.9156 - 713us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.8606 - 815us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.1336 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.2934 - 744us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5953 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.6780 - 766us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.0852 - 1ms/epoch - 18us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.1712 - 738us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.1543 - 925us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.2548 - 907us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.7280 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79.8713 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.1700 - 4ms/epoch - 68us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.3885 - 740us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.8693 - 734us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.1683 - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.0331 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.6633 - 707us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.3795 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.5895 - 785us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.6398 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.9928 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.1520 - 831us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.0905 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742.7753 - 666us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.3331 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4785 - 1ms/epoch - 20us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.6953 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.8416 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.5388 - 678us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.6718 - 4ms/epoch - 72us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.3432 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.6269 - 592us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.0928 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.2217 - 776us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.6823 - 798us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.0548 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298.3696 - 704us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6450 - 859us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.5834 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.7775 - 1ms/epoch - 19us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.7045 - 748us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.0406 - 2ms/epoch - 28us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 157us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.3962 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.6038 - 969us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9183 - 1ms/epoch - 22us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.6403 - 655us/epoch - 11us/sample\n", - "Generation 245/300\n", - "Solving for Nash Equilibrium in Generation 245/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.2137 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.5085 - 744us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8776 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.5664 - 787us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2936 - 754us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.4221 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.7532 - 778us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.2706 - 2ms/epoch - 36us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2883 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235.8193 - 606us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.2722 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.2363 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2432 - 1ms/epoch - 19us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.3052 - 803us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.8287 - 651us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.3632 - 871us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.8166 - 2ms/epoch - 28us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 253.3056 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.1006 - 6ms/epoch - 98us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.0848 - 741us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45.5136 - 5ms/epoch - 85us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.5410 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2856 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.7156 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.8872 - 851us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.0594 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.6564 - 751us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.3600 - 704us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.3569 - 1ms/epoch - 23us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.7422 - 904us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.8571 - 695us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.5842 - 860us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.3630 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.3260 - 684us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6001 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.2258 - 663us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.6762 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8033 - 646us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.6751 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.0863 - 709us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.0604 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.2448 - 2ms/epoch - 27us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.0151 - 2ms/epoch - 25us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.9758 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5002 - 843us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.3440 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.8973 - 2ms/epoch - 25us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.1567 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.0096 - 2ms/epoch - 24us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1063 - 679us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.7390 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.3841 - 818us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8828 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.3506 - 657us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.9275 - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.9933 - 3ms/epoch - 52us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.5412 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1561 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1773 - 946us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.6897 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.4081 - 665us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8788 - 892us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.9886 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4096 - 772us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7318 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.6766 - 6ms/epoch - 91us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.4861 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6063 - 717us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.8798 - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.5391 - 685us/epoch - 11us/sample\n", - "Generation 246/300\n", - "Solving for Nash Equilibrium in Generation 246/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.4969 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.9902 - 644us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.6538 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.8268 - 701us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.0722 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.7767 - 698us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.4165 - 680us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557.6280 - 669us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.7576 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.4991 - 816us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.7657 - 817us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2466 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.6072 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.6726 - 4ms/epoch - 62us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.5417 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.9050 - 665us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5037 - 694us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6533 - 969us/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.2643 - 3ms/epoch - 45us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.7672 - 762us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3912 - 776us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.2399 - 836us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.7839 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.8134 - 509us/epoch - 8us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.9395 - 747us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.0773 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.9826 - 573us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.0540 - 954us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.9281 - 2ms/epoch - 40us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.9711 - 729us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.9371 - 2ms/epoch - 27us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.0878 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 13ms/epoch - 214us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.0825 - 2ms/epoch - 31us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.8694 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.6911 - 651us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7056 - 731us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.1913 - 1ms/epoch - 17us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.9478 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.6127 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.3218 - 865us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.5836 - 761us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.3298 - 609us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.8187 - 816us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.7450 - 3ms/epoch - 49us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9431 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7173 - 671us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.7076 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.7984 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4945 - 756us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.0244 - 1ms/epoch - 17us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.6758 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6259 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0837 - 995us/epoch - 16us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.1223 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.9306 - 2ms/epoch - 34us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.3482 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.8712 - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.3398 - 578us/epoch - 9us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.2713 - 843us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.2030 - 656us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.8908 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6411 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.1347 - 2ms/epoch - 39us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.7369 - 669us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.2001 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.2921 - 687us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.3802 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.1490 - 876us/epoch - 14us/sample\n", - "Generation 247/300\n", - "Solving for Nash Equilibrium in Generation 247/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.5238 - 705us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 93us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.3339 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.4785 - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.3039 - 2ms/epoch - 27us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.4142 - 857us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.7415 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.8199 - 1ms/epoch - 23us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.5836 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1448 - 678us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.9521 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.1875 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.7372 - 593us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.6149 - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8542 - 722us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.1434 - 629us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5367 - 737us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.6330 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2872 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.7246 - 748us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.9886 - 791us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1620 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.4585 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.4708 - 827us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.4951 - 693us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.2463 - 588us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.3613 - 771us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.3979 - 808us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.6830 - 730us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.1350 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.9256 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.6432 - 726us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.6360 - 734us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.2177 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.7364 - 667us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.2946 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.5031 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.6682 - 612us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.6073 - 3ms/epoch - 44us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.4187 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.8557 - 717us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.5998 - 698us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.4407 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.1595 - 664us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.0354 - 716us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.6912 - 760us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.4622 - 1ms/epoch - 17us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.0084 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.6870 - 654us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.9796 - 939us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.7829 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.0009 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.0269 - 772us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.5737 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.6395 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7319 - 767us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.9713 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.5015 - 707us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.1769 - 875us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7250 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.7870 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.6618 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.4222 - 723us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.0321 - 702us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.7120 - 785us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.7805 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.7789 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245.4488 - 1ms/epoch - 19us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2272 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2352 - 781us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4570 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.0182 - 673us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4433 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5720 - 644us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.1331 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5905 - 664us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 248/300\n", - "Solving for Nash Equilibrium in Generation 248/300\n", + "Generation 64/300\n", + "Solving for Nash Equilibrium in Generation 64/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.7368 - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.3881 - 698us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.3411 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.2249 - 697us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.2533 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.1782 - 1ms/epoch - 18us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.0457 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9862 - 689us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.7607 - 8ms/epoch - 125us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8080 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.9299 - 690us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8466 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.5896 - 653us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.8148 - 649us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8810 - 663us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.5410 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.9886 - 671us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.8083 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.8012 - 741us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.2890 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1552 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.0916 - 2ms/epoch - 30us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.6789 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.1320 - 871us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.3436 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.5325 - 723us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.8709 - 737us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.0707 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771.8162 - 598us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.0604 - 656us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.8286 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.3312 - 1ms/epoch - 17us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.8868 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.5475 - 706us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.1021 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5818 - 580us/epoch - 9us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8469 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.0959 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.3782 - 943us/epoch - 15us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3584 - 746us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.4789 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.4969 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.1822 - 1ms/epoch - 19us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.0137 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.5165 - 1ms/epoch - 20us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.8503 - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.0890 - 14ms/epoch - 229us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.7968 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.9398 - 14ms/epoch - 227us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.8677 - 789us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.5762 - 696us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4426 - 2ms/epoch - 26us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.8173 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.2808 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.6973 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.7969 - 1ms/epoch - 21us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.8335 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.8540 - 574us/epoch - 9us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.0272 - 701us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.0134 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.4528 - 771us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.0826 - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.4568 - 1ms/epoch - 21us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.4189 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.1685 - 961us/epoch - 16us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3621 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.8022 - 745us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.2899 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0685 - 749us/epoch - 12us/sample\n", - "Generation 249/300\n", - "Solving for Nash Equilibrium in Generation 249/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.7473 - 583us/epoch - 9us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.4275 - 605us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.0902 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.3483 - 722us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.2905 - 915us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.8607 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6954 - 709us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.4908 - 1ms/epoch - 16us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.1145 - 3ms/epoch - 45us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.7353 - 795us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.2033 - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.9474 - 703us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.9971 - 687us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.3452 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764.3792 - 626us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.3885 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.2518 - 679us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.3943 - 6ms/epoch - 96us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.5716 - 779us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.0628 - 875us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.5726 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.3486 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.9557 - 853us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.9067 - 671us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.0589 - 692us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.1701 - 948us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.2779 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.7861 - 663us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.9250 - 585us/epoch - 9us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.8004 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.5929 - 2ms/epoch - 37us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.4201 - 653us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.5450 - 752us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.7106 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.1943 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.3662 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.3975 - 2ms/epoch - 28us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.2097 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.1087 - 659us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.6399 - 724us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.0959 - 791us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.8393 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.2820 - 625us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.9437 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.1492 - 2ms/epoch - 36us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.4994 - 592us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.6111 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.4371 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2733 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.0555 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.8128 - 691us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.9798 - 634us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3777 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.6328 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.0679 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.2191 - 785us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5162 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.6643 - 715us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.8770 - 2ms/epoch - 34us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0198 - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5546 - 797us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.5837 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.1815 - 855us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 17ms/epoch - 278us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.5792 - 1ms/epoch - 19us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 223.7430 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.3059 - 601us/epoch - 10us/sample\n", - "Generation 250/300\n", - "Solving for Nash Equilibrium in Generation 250/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428.9723 - 602us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.1556 - 711us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.9976 - 764us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.1074 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.7115 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.5165 - 737us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.8774 - 615us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.5940 - 1ms/epoch - 20us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.0049 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.6909 - 653us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5447 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.3204 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.7119 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.6613 - 1ms/epoch - 22us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.9816 - 726us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.4592 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302.0621 - 755us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.0323 - 804us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.5781 - 746us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.0878 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.7578 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 14ms/epoch - 232us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.8275 - 991us/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.0930 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9819 - 705us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.3944 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.4814 - 7ms/epoch - 106us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1740 - 821us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.8909 - 10ms/epoch - 162us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.3261 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.0977 - 2ms/epoch - 29us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4072 - 2ms/epoch - 28us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.1758 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.5861 - 1ms/epoch - 19us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.1594 - 839us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8650 - 879us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.7316 - 1ms/epoch - 20us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8802 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.4811 - 1ms/epoch - 20us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4200 - 830us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.7692 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.2101 - 813us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.9473 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8503 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.5258 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.4411 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.7413 - 811us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.3041 - 816us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6369 - 686us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.1483 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.3764 - 2ms/epoch - 25us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.7871 - 772us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.7795 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 416.9670 - 843us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.2388 - 800us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.9521 - 1ms/epoch - 19us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.8427 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.2602 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.6109 - 947us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8150 - 872us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.2349 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.9796 - 2ms/epoch - 34us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.3881 - 750us/epoch - 12us/sample\n", - "Generation 251/300\n", - "Solving for Nash Equilibrium in Generation 251/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 65/300\n", + "Solving for Nash Equilibrium in Generation 65/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.7095 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.2834 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 133us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.4506 - 3ms/epoch - 44us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.0411 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.3840 - 917us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.2868 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.4420 - 1ms/epoch - 21us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.7574 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.4056 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.1851 - 821us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.4338 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.1075 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.3348 - 3ms/epoch - 43us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9783 - 2ms/epoch - 26us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5802 - 725us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.3290 - 707us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.7214 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.9714 - 751us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6518 - 935us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.0192 - 732us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245.2198 - 781us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3198 - 667us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.9186 - 792us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.6859 - 755us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 181.2470 - 630us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.2865 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.1491 - 780us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.5348 - 617us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.2675 - 767us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.1018 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.8975 - 802us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.5392 - 1ms/epoch - 16us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.8491 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.0886 - 709us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.4839 - 767us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9293 - 1ms/epoch - 21us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2185 - 659us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.8102 - 968us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.4308 - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.1350 - 1ms/epoch - 21us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.8293 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.5911 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.2540 - 609us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.6425 - 676us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.8266 - 650us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.9012 - 847us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.7516 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.4963 - 732us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.6778 - 865us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.7966 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.7635 - 673us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.7809 - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.3224 - 764us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.8498 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4908 - 657us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.6332 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.0233 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9340 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.7160 - 753us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.6476 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.9356 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6703 - 735us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.1814 - 4ms/epoch - 64us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.2116 - 639us/epoch - 10us/sample\n", - "Generation 252/300\n", - "Solving for Nash Equilibrium in Generation 252/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.6537 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.8693 - 650us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.8499 - 4ms/epoch - 58us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.9767 - 764us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.9608 - 1ms/epoch - 16us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.1123 - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.8802 - 2ms/epoch - 25us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.8255 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.2870 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.2936 - 694us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.3778 - 675us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.0530 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.2045 - 964us/epoch - 16us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.8003 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4509 - 862us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.1060 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3287 - 700us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.2876 - 1ms/epoch - 22us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.8636 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 224.0306 - 813us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.5834 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4944 - 742us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38.7742 - 824us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.8170 - 830us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.3342 - 690us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.2795 - 2ms/epoch - 27us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.5067 - 850us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.9169 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.7103 - 767us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.1893 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.7684 - 658us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0629 - 803us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.8953 - 598us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.2284 - 933us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.8483 - 932us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.4557 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.6883 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.1586 - 768us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.7924 - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.2309 - 721us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.6933 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354.3239 - 1ms/epoch - 21us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.9218 - 777us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.3358 - 583us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.5524 - 803us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.3222 - 823us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.4185 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.6992 - 825us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.0533 - 3ms/epoch - 45us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.2346 - 835us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.4228 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.8093 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.4551 - 861us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.5411 - 717us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.5901 - 825us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.3466 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.5554 - 10ms/epoch - 157us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.3239 - 744us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.4149 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.5750 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.5098 - 1ms/epoch - 17us/sample\n", - "Generation 253/300\n", - "Solving for Nash Equilibrium in Generation 253/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.5759 - 645us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.9875 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.0377 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.4711 - 928us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.4559 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.3893 - 741us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.3434 - 3ms/epoch - 41us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7480 - 864us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.3285 - 668us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.7388 - 855us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.8141 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240.1512 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.6038 - 3ms/epoch - 43us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.3010 - 634us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.4981 - 901us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.8392 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.9470 - 580us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2562 - 854us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3370 - 637us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.6527 - 749us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.7762 - 2ms/epoch - 32us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.4862 - 969us/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.1856 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.2310 - 754us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.0201 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.9916 - 834us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.7969 - 755us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.3037 - 790us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.0557 - 2ms/epoch - 38us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.1982 - 1ms/epoch - 18us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3792 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.4691 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.3529 - 4ms/epoch - 62us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.5148 - 891us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.8693 - 1ms/epoch - 17us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.9628 - 9ms/epoch - 144us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.6939 - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4050 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.2690 - 793us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.9310 - 890us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.3394 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.7352 - 8ms/epoch - 122us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.9351 - 2ms/epoch - 31us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.8226 - 698us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.2083 - 925us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.7938 - 819us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.0935 - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 14ms/epoch - 223us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.8491 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.2782 - 711us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.9715 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.1242 - 701us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.3036 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.4813 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2910 - 819us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.0788 - 738us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0432 - 901us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.6389 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.0974 - 2ms/epoch - 35us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.4796 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.5466 - 691us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.6418 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.6935 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276.7472 - 2ms/epoch - 28us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1709 - 726us/epoch - 12us/sample\n", - "Generation 254/300\n", - "Solving for Nash Equilibrium in Generation 254/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.6554 - 822us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7927 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.7867 - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.6823 - 721us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.3170 - 742us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.6604 - 768us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.8317 - 616us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.9756 - 2ms/epoch - 36us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.1170 - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4760 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.3201 - 758us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.9216 - 721us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4911 - 3ms/epoch - 41us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.6061 - 632us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.6830 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.1897 - 732us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.9136 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9900 - 5ms/epoch - 86us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.1948 - 951us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.5249 - 1ms/epoch - 17us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.4749 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6016 - 2ms/epoch - 30us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5669 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.7582 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.2113 - 4ms/epoch - 71us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.5657 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5155 - 3ms/epoch - 42us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.2177 - 729us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.5673 - 956us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.4332 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.6534 - 21ms/epoch - 345us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2855 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.6242 - 698us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.6629 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.9190 - 818us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.6898 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.0324 - 611us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.1920 - 690us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.3989 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.5359 - 933us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.4498 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.5255 - 842us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.2622 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.0833 - 597us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.6060 - 684us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.8725 - 647us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.2122 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.9630 - 632us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.1962 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.7911 - 626us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.0570 - 659us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.4551 - 626us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.7245 - 716us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.6071 - 708us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.0131 - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0712 - 666us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.3792 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.1349 - 747us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.4603 - 1ms/epoch - 20us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41.1252 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.8303 - 807us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.1193 - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.2294 - 960us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7134 - 1ms/epoch - 23us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.6024 - 869us/epoch - 14us/sample\n", - "Generation 255/300\n", - "Solving for Nash Equilibrium in Generation 255/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 66/300\n", + "Solving for Nash Equilibrium in Generation 66/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.6530 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.7956 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201.7095 - 1ms/epoch - 18us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.9658 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.2364 - 771us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.3063 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.3439 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.8318 - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.5968 - 795us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.1465 - 768us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.3290 - 754us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 143us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9406 - 1ms/epoch - 21us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.1163 - 613us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.4879 - 709us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.1897 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.0684 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.3664 - 695us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.5516 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2931 - 648us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.7970 - 602us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6548 - 553us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.6276 - 728us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.4389 - 2ms/epoch - 37us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.3617 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.9965 - 714us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.9680 - 645us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.2993 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.5134 - 789us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.7704 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.9062 - 626us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.6513 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3271 - 1ms/epoch - 21us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.0677 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8143 - 708us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.3636 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.7477 - 689us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.0978 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.5024 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.2214 - 622us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.5140 - 625us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.1659 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8009 - 607us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9991 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.6810 - 661us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.2083 - 1ms/epoch - 17us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.7661 - 684us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.3425 - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.9064 - 574us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.6479 - 790us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.6660 - 791us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.6711 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.1351 - 791us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.8106 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.5782 - 731us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.2445 - 734us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.7951 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.1802 - 667us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4897 - 721us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.6790 - 710us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.9426 - 711us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.2752 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.3097 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.2529 - 751us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.4178 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.8957 - 757us/epoch - 12us/sample\n", - "Generation 256/300\n", - "Solving for Nash Equilibrium in Generation 256/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.5967 - 695us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6112 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.6694 - 667us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 151us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.6094 - 927us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.4605 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.5921 - 2ms/epoch - 36us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.2664 - 712us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.3455 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2582 - 4ms/epoch - 56us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.5774 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.0748 - 613us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 155us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1564 - 1ms/epoch - 19us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.1802 - 4ms/epoch - 68us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.7675 - 636us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.7410 - 2ms/epoch - 39us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.6811 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.8146 - 626us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.4676 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.3997 - 683us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.0911 - 717us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.6775 - 614us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.6904 - 697us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.0432 - 710us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.2368 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.2623 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.7352 - 682us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.0301 - 795us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.3767 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4482 - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.4567 - 988us/epoch - 16us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.7135 - 1ms/epoch - 23us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.3529 - 859us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.1284 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.9705 - 2ms/epoch - 35us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.0985 - 1ms/epoch - 16us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.9024 - 777us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.6077 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.7107 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.4959 - 735us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.0419 - 4ms/epoch - 71us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.7871 - 6ms/epoch - 99us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.4949 - 2ms/epoch - 27us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.6443 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.1434 - 709us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.7780 - 723us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.3893 - 919us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.7243 - 820us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.0918 - 653us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.7568 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.9135 - 710us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.6064 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.7607 - 632us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.7383 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.3260 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.0657 - 1ms/epoch - 19us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7529 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9196 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.4853 - 580us/epoch - 9us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.1160 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.3181 - 688us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.0657 - 732us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.1362 - 707us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.9453 - 822us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.2706 - 686us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.8922 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.1185 - 2ms/epoch - 24us/sample\n", - "Generation 257/300\n", - "Solving for Nash Equilibrium in Generation 257/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.8531 - 773us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.0638 - 707us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.1789 - 950us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.1000 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.9164 - 708us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 87.8421 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.1370 - 525us/epoch - 8us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.5616 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6503 - 695us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.8189 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.8127 - 835us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.9535 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2537 - 720us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.9668 - 799us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7306 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.8080 - 713us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.1855 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.3729 - 712us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.1863 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.1090 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.6829 - 821us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.4539 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.0325 - 647us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.3046 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.5385 - 1ms/epoch - 20us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.5167 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.4185 - 587us/epoch - 9us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.0664 - 841us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.9971 - 770us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.1130 - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.4942 - 738us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.0828 - 709us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.9912 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.2605 - 908us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.4297 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.4362 - 792us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.9924 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.0992 - 684us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.7343 - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.1909 - 819us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.1122 - 917us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.2599 - 612us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.2540 - 642us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.2492 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 137us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.7188 - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.2901 - 709us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201.0032 - 730us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 169us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2731 - 2ms/epoch - 30us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.7058 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361.3831 - 7ms/epoch - 110us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.5441 - 651us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.9436 - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9622 - 894us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.6976 - 660us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.1822 - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.9514 - 657us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.7212 - 746us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7016 - 812us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 121us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.5138 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.9769 - 842us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.6211 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.9358 - 566us/epoch - 9us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.3642 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 99.4025 - 744us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Generation 258/300\n", - "Solving for Nash Equilibrium in Generation 258/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.6172 - 727us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.0088 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.2963 - 822us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.2992 - 671us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 110us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.0961 - 828us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7944 - 619us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.8589 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.5713 - 630us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6693 - 652us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.9235 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.7474 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.6158 - 3ms/epoch - 50us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.6653 - 771us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.4478 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.7165 - 904us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.4329 - 674us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.3788 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.8370 - 726us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.1899 - 759us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 209.6717 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.2418 - 1ms/epoch - 22us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.2006 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4893 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.9727 - 786us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3776 - 747us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.9537 - 7ms/epoch - 105us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.8805 - 1ms/epoch - 22us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.6177 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3640 - 1ms/epoch - 21us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.3176 - 691us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.6163 - 852us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.9496 - 745us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.4698 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6310 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.6356 - 650us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.1554 - 4ms/epoch - 65us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.9199 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.8835 - 653us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.3185 - 642us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.7184 - 673us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 747.5931 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.6300 - 575us/epoch - 9us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.8192 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.4921 - 692us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.3560 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.5532 - 3ms/epoch - 46us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.8188 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389.5391 - 3ms/epoch - 47us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214.7083 - 957us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49.4337 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.0652 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4471 - 1ms/epoch - 21us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.1436 - 553us/epoch - 9us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.7984 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0497 - 696us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.4114 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.9435 - 569us/epoch - 9us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1060 - 2ms/epoch - 34us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.2192 - 761us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.8109 - 717us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.7693 - 4ms/epoch - 63us/sample\n", - "Episode 50/50\n", - "Generation 259/300\n", - "Solving for Nash Equilibrium in Generation 259/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.2777 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.6408 - 682us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.0056 - 663us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 139.4108 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.1370 - 6ms/epoch - 96us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2584 - 819us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.4242 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.8249 - 3ms/epoch - 41us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.8285 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.9195 - 583us/epoch - 9us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.8541 - 772us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.9681 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.9703 - 1ms/epoch - 18us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.9051 - 730us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0358 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.5674 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2213 - 708us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.5208 - 706us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.2404 - 734us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.3270 - 1ms/epoch - 18us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.3743 - 740us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.7663 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.1921 - 5ms/epoch - 74us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.1185 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.2675 - 795us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8171 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.6779 - 733us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.1116 - 747us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.5117 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.7394 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3536 - 789us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.5261 - 858us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.1147 - 808us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.3855 - 2ms/epoch - 28us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.2595 - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.7134 - 772us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.2980 - 611us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.2752 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319.0596 - 795us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.5091 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.4717 - 1ms/epoch - 20us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.1320 - 695us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.7344 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.1811 - 821us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.1897 - 757us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.4523 - 847us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.9916 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.2693 - 575us/epoch - 9us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 175.2186 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.4772 - 691us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.1550 - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.1892 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.7282 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.7804 - 728us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.0441 - 659us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.3610 - 873us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.6290 - 914us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.0739 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.9813 - 633us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.4423 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5332 - 707us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.1683 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 722.2518 - 756us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.9188 - 634us/epoch - 10us/sample\n", - "Generation 260/300\n", - "Solving for Nash Equilibrium in Generation 260/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.9726 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.7936 - 655us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.5260 - 1ms/epoch - 24us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.8787 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.1639 - 700us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2054 - 892us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.3928 - 632us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.9448 - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9177 - 756us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.5515 - 787us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.0126 - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.8452 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.5501 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.6664 - 705us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.3938 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321.2694 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2438 - 571us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.3746 - 640us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.8654 - 702us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.6135 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.9856 - 4ms/epoch - 63us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.6516 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.6490 - 809us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.5308 - 962us/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8867 - 661us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.4110 - 2ms/epoch - 30us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.0075 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.4566 - 770us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40.9194 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.2470 - 1ms/epoch - 20us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.7112 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.4043 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.6560 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.1000 - 761us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.7633 - 769us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.3881 - 1ms/epoch - 18us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.3763 - 720us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.4030 - 740us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.6096 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.1309 - 834us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.4572 - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.6238 - 762us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.2007 - 5ms/epoch - 76us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.1588 - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7604 - 1ms/epoch - 18us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.6972 - 785us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.9939 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.0363 - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.2805 - 651us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.4763 - 822us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.1957 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.8030 - 17ms/epoch - 276us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.7246 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2029 - 733us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.6570 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.8304 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4631 - 565us/epoch - 9us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.4221 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.9165 - 7ms/epoch - 120us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.8906 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.0031 - 1ms/epoch - 18us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.3319 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9974 - 644us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.5830 - 992us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.6094 - 842us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.2136 - 811us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.6538 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.3061 - 743us/epoch - 12us/sample\n", - "Generation 261/300\n", - "Solving for Nash Equilibrium in Generation 261/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.2930 - 3ms/epoch - 41us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.9796 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.3962 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.7750 - 2ms/epoch - 38us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.0351 - 776us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.6434 - 1ms/epoch - 19us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.0651 - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.3052 - 803us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 730.5243 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0886 - 793us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.8339 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.2229 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7892 - 728us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 171.6415 - 950us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.6913 - 816us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.6559 - 843us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376.0885 - 678us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.7137 - 1ms/epoch - 24us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.1562 - 795us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.3728 - 785us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.4935 - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.2551 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.2259 - 722us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.1813 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.3897 - 749us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.8320 - 663us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.2506 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.4437 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.5892 - 970us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.7470 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.1957 - 2ms/epoch - 27us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.4370 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.9005 - 1ms/epoch - 17us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.0024 - 584us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.8616 - 632us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.6573 - 770us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.2378 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.2305 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.3071 - 844us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.6377 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7831 - 726us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.7714 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.5684 - 835us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.7605 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.4325 - 803us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.5168 - 1ms/epoch - 21us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.0646 - 1ms/epoch - 16us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8597 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.3786 - 562us/epoch - 9us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2021 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.5035 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2968 - 653us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.5450 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.1254 - 590us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.8068 - 855us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.4892 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.4510 - 692us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.0552 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.5359 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2552 - 794us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.3843 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.7236 - 5ms/epoch - 76us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.9646 - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8877 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.8318 - 2ms/epoch - 26us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.5178 - 1ms/epoch - 17us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.4861 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.4888 - 1ms/epoch - 18us/sample\n", - "Generation 262/300\n", - "Solving for Nash Equilibrium in Generation 262/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.5842 - 856us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.5314 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.4502 - 762us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.3708 - 648us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.8419 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.1929 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.1935 - 711us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753.7278 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.5236 - 967us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.7433 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9462 - 644us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.9114 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.6113 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.7409 - 2ms/epoch - 30us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5302 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6855 - 5ms/epoch - 85us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.9617 - 1ms/epoch - 17us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.6100 - 1ms/epoch - 18us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.6696 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.1940 - 1ms/epoch - 18us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769.0184 - 752us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.5574 - 857us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.2291 - 798us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6107 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.1985 - 765us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.4185 - 834us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.4996 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.9410 - 743us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.3151 - 625us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.7797 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.2746 - 955us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9508 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.7918 - 752us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.6215 - 842us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5371 - 1ms/epoch - 19us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.1598 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.5868 - 642us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8359 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.6431 - 802us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.5895 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.4696 - 786us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.4027 - 848us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.6849 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.5329 - 13ms/epoch - 203us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.8892 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.7966 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.5220 - 2ms/epoch - 30us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.9242 - 817us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.0609 - 795us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.2673 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4902 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.7806 - 835us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.8792 - 6ms/epoch - 91us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.5353 - 1ms/epoch - 20us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.1334 - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8595 - 959us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.3783 - 778us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.8584 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.9210 - 907us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.6397 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.0919 - 2ms/epoch - 33us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.0302 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.1833 - 874us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.7067 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.0497 - 715us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2892 - 860us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.8599 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231.6375 - 940us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8684 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.5499 - 788us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.0097 - 7ms/epoch - 116us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.2271 - 2ms/epoch - 26us/sample\n", - "Generation 263/300\n", - "Solving for Nash Equilibrium in Generation 263/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.0236 - 745us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 106us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.4088 - 2ms/epoch - 25us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.0179 - 799us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.7977 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 95us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.2308 - 8ms/epoch - 128us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.4035 - 698us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.0658 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.9214 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.0069 - 788us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.3309 - 889us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.9598 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5368 - 842us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.2172 - 759us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.1980 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.6431 - 603us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6924 - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.3120 - 821us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 612.3658 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.5414 - 664us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.6516 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1556 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0023 - 1ms/epoch - 22us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.3314 - 773us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.3994 - 5ms/epoch - 85us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5814 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.1376 - 967us/epoch - 16us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.4777 - 761us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.9481 - 941us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322.1032 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.0281 - 9ms/epoch - 145us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.2552 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9880 - 839us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.8642 - 829us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.7780 - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.8097 - 766us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.1200 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.5367 - 883us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.7352 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.9340 - 7ms/epoch - 105us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.8667 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.3864 - 2ms/epoch - 30us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.0000 - 1ms/epoch - 16us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.8848 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.0137 - 3ms/epoch - 43us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.8137 - 718us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.0670 - 777us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.5200 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.6457 - 634us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.8039 - 738us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.3279 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.4902 - 821us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.6954 - 1ms/epoch - 17us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.4872 - 692us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.6278 - 661us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.1559 - 635us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 96us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4529 - 14ms/epoch - 219us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.2681 - 782us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.6250 - 827us/epoch - 13us/sample\n", - "Generation 264/300\n", - "Solving for Nash Equilibrium in Generation 264/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.3741 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.7346 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.8083 - 810us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.4445 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.9568 - 4ms/epoch - 58us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.4558 - 762us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.4204 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.8602 - 787us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.2269 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9447 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.4485 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245.0478 - 1ms/epoch - 22us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726.2261 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 92us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.6906 - 902us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.4270 - 778us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.2571 - 621us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.3650 - 645us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.0479 - 588us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.1048 - 723us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.2729 - 780us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.5879 - 729us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.8073 - 698us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.2663 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.3335 - 655us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.1068 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.0723 - 1ms/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.0392 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.3513 - 723us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1441 - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.2685 - 20ms/epoch - 320us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.9829 - 871us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.3771 - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.5562 - 836us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.2303 - 678us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.7875 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.2989 - 769us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 756.8454 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.7638 - 5ms/epoch - 81us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.3523 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.6296 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.5251 - 900us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.1182 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6750 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.9333 - 696us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.0040 - 830us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.0450 - 687us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.0075 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.8624 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.6682 - 616us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.5056 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 103us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.1360 - 3ms/epoch - 44us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.4738 - 2ms/epoch - 32us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.6106 - 899us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.3680 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.3855 - 769us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.9819 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.8155 - 622us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.4152 - 13ms/epoch - 203us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.8872 - 928us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.5691 - 1ms/epoch - 16us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8592 - 591us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1945 - 576us/epoch - 9us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.6209 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.6042 - 1ms/epoch - 23us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.8458 - 875us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.5270 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.8380 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.4287 - 915us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.6629 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.8130 - 1ms/epoch - 17us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.5239 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.4095 - 586us/epoch - 9us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.0225 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.4851 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.5556 - 820us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.7733 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.1223 - 2ms/epoch - 33us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 265/300\n", - "Solving for Nash Equilibrium in Generation 265/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.2094 - 889us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.7837 - 718us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278.2362 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.7891 - 1ms/epoch - 16us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47.6105 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.5933 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.2510 - 811us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.9805 - 5ms/epoch - 81us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7236 - 711us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.7889 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.8714 - 622us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.6856 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9282 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.8363 - 2ms/epoch - 30us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640.2745 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.4149 - 718us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.1713 - 662us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.8569 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.7384 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.1731 - 704us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.7016 - 671us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.4593 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.5708 - 785us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.9056 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.1904 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.6329 - 892us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.7430 - 716us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.3636 - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.8234 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414.2267 - 815us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.2346 - 780us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.4363 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.2672 - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4128 - 762us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.0793 - 913us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8941 - 755us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.0261 - 712us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.5563 - 713us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.1156 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.9724 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.3107 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 125us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.1081 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8412 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.1983 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.7574 - 746us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2334 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9096 - 2ms/epoch - 28us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.5818 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.1885 - 751us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.1431 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1412 - 796us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 751.2259 - 843us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.5129 - 802us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.6165 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.4340 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.6169 - 504us/epoch - 8us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.7911 - 791us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 155us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9485 - 789us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.9037 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.6885 - 720us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8453 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.8854 - 641us/epoch - 10us/sample\n", - "Generation 266/300\n", - "Solving for Nash Equilibrium in Generation 266/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.4348 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.2351 - 863us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.9733 - 817us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.0569 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.5936 - 714us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.7704 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.8151 - 904us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.1566 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.3008 - 640us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.5853 - 719us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.1671 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.6180 - 2ms/epoch - 32us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.1301 - 910us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.1162 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 643.7417 - 687us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.7630 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.1015 - 2ms/epoch - 26us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.6506 - 647us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.1704 - 686us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.3272 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.6906 - 711us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4297 - 785us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.7122 - 679us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.4091 - 1ms/epoch - 18us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.4987 - 585us/epoch - 9us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61.2052 - 675us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.0254 - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.5105 - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.4998 - 762us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.7012 - 677us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8090 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2355 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.3711 - 17ms/epoch - 271us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 574.8404 - 684us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9077 - 991us/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.1390 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.2325 - 1ms/epoch - 17us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.1246 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.8553 - 2ms/epoch - 27us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.3463 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250.0309 - 744us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7806 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.3205 - 648us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.6219 - 1ms/epoch - 23us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.7822 - 772us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.4903 - 2ms/epoch - 27us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.7658 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.6451 - 575us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.4482 - 934us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.6837 - 756us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.9688 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7745 - 820us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.1926 - 958us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.2557 - 669us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0007 - 3ms/epoch - 46us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.8204 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.3739 - 4ms/epoch - 59us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5765 - 701us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.6061 - 8ms/epoch - 130us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.1744 - 616us/epoch - 10us/sample\n", - "Generation 267/300\n", - "Solving for Nash Equilibrium in Generation 267/300\n", + "Generation 67/300\n", + "Solving for Nash Equilibrium in Generation 67/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5953 - 610us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.7961 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.1051 - 690us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.1576 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.7632 - 705us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.7206 - 730us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.8478 - 2ms/epoch - 35us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.1337 - 2ms/epoch - 30us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.4638 - 802us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.2795 - 640us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.2708 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6424 - 626us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.9523 - 5ms/epoch - 73us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.5949 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.5919 - 888us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.7800 - 871us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44.0811 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.9904 - 986us/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.6219 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2487 - 613us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.7275 - 12ms/epoch - 191us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.8845 - 907us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.2939 - 598us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.2657 - 924us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.5390 - 756us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.2015 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.0416 - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.8860 - 860us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.7965 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.1407 - 804us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.0469 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.2472 - 2ms/epoch - 36us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.4656 - 670us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.4500 - 723us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.9049 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.2372 - 949us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.2031 - 5ms/epoch - 74us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.9430 - 734us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.8570 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.6563 - 621us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.7635 - 620us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9373 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.5988 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.7089 - 1ms/epoch - 17us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0825 - 918us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.0803 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.6498 - 714us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2939 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.6221 - 831us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6247 - 955us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.1597 - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.3990 - 801us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.3417 - 804us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.7031 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.5914 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9676 - 7ms/epoch - 107us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.3792 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.6445 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.0397 - 707us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.4616 - 623us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305.5118 - 1ms/epoch - 16us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0898 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.2099 - 789us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9886 - 3ms/epoch - 43us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.1638 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.0171 - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.4653 - 797us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6527 - 718us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.5294 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2905 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.6863 - 676us/epoch - 11us/sample\n", - "Generation 268/300\n", - "Solving for Nash Equilibrium in Generation 268/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.7026 - 922us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.9729 - 722us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.0155 - 767us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.1874 - 1ms/epoch - 18us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.6250 - 758us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.1721 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3893 - 604us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.7664 - 674us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.5934 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.8187 - 2ms/epoch - 28us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.7988 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.2028 - 709us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.7453 - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.1180 - 786us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.5568 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.3054 - 831us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.3221 - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.4749 - 9ms/epoch - 144us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.5582 - 1000us/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.6064 - 4ms/epoch - 63us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.3599 - 601us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.5858 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.6086 - 901us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410.7718 - 991us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.8536 - 1ms/epoch - 22us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.3782 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.6744 - 773us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2000 - 786us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231.3515 - 870us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.9333 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4425 - 847us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.8652 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.3859 - 978us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.6437 - 654us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.5919 - 720us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8171 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.1841 - 897us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2764 - 737us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.7078 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.8391 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.5575 - 826us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.4023 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.8382 - 633us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.9130 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.8067 - 813us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.6124 - 690us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.8940 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.6003 - 781us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.6146 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.6605 - 813us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9986 - 793us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.8793 - 2ms/epoch - 25us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.0788 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.0584 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.2194 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.1624 - 747us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.0743 - 712us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 269/300\n", - "Solving for Nash Equilibrium in Generation 269/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.3264 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37.9614 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.0964 - 697us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.8290 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.6805 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.5804 - 763us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.0092 - 731us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231.5872 - 733us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.6710 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.6617 - 845us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.7528 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.7216 - 2ms/epoch - 36us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.5184 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.4883 - 690us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 108.8872 - 848us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.2968 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.6492 - 889us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.2532 - 984us/epoch - 16us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.0704 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 141us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.2765 - 8ms/epoch - 129us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.2583 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.1113 - 638us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592.7692 - 587us/epoch - 9us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.7875 - 2ms/epoch - 29us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.4656 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.7431 - 801us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.0911 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.3966 - 3ms/epoch - 46us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1182 - 795us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.8160 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.0621 - 946us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346.8639 - 736us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 195.8988 - 1ms/epoch - 18us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41.9670 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.7453 - 781us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9388 - 899us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.5627 - 759us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.0880 - 877us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.7441 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.9144 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.4425 - 659us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.6282 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 157us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.9451 - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.4982 - 625us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.5321 - 2ms/epoch - 25us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0204 - 2ms/epoch - 33us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.6396 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.1001 - 744us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.5905 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0499 - 804us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.2347 - 897us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.6922 - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.3468 - 647us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 22ms/epoch - 349us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8378 - 1ms/epoch - 18us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.7087 - 2ms/epoch - 26us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235.0109 - 879us/epoch - 14us/sample\n", - "Generation 270/300\n", - "Solving for Nash Equilibrium in Generation 270/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.4372 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.3042 - 759us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4331 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.2583 - 954us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.0668 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.5081 - 787us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.6550 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417.0281 - 6ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.0650 - 1ms/epoch - 17us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.1744 - 950us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430.4241 - 776us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.9781 - 799us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.7519 - 1ms/epoch - 19us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.4924 - 964us/epoch - 16us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.0806 - 1ms/epoch - 17us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.2568 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.3990 - 780us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.9122 - 831us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.6002 - 733us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.3521 - 1ms/epoch - 21us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 122.9587 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.9059 - 797us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.6982 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.0718 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.2706 - 875us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.5823 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3508 - 1ms/epoch - 21us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.7591 - 788us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.0068 - 1ms/epoch - 21us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8387 - 712us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8147 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.3738 - 794us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230.2296 - 1ms/epoch - 18us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4719 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.0135 - 734us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.1373 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.8171 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7344 - 2ms/epoch - 25us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.1169 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.9664 - 732us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320.3976 - 853us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.3654 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.8285 - 652us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 159.9114 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.8371 - 713us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.1882 - 700us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.5796 - 589us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.7845 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.0138 - 720us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.7559 - 1ms/epoch - 21us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.1265 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.9222 - 2ms/epoch - 35us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1646 - 699us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 765.0519 - 735us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.8787 - 834us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.7075 - 3ms/epoch - 53us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.9028 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.3913 - 682us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.3782 - 876us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0224 - 813us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.5835 - 969us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2038 - 1ms/epoch - 17us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.7595 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.9907 - 2ms/epoch - 35us/sample\n", - "Generation 271/300\n", - "Solving for Nash Equilibrium in Generation 271/300\n", + "Generation 68/300\n", + "Solving for Nash Equilibrium in Generation 68/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7006 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9924 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.5653 - 618us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 189.5107 - 866us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.1640 - 1ms/epoch - 20us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.8247 - 860us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.6096 - 741us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.8029 - 768us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.4006 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.8976 - 764us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.5822 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.9753 - 622us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 218.5178 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.7283 - 582us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.9187 - 2ms/epoch - 35us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.8297 - 768us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 98us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.5619 - 1ms/epoch - 22us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.2586 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.6334 - 784us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.5581 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.9596 - 3ms/epoch - 42us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.9414 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.7033 - 694us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.1552 - 738us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.2376 - 727us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.5893 - 865us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3028 - 941us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.1342 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.0374 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.7730 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.5736 - 8ms/epoch - 127us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.3810 - 653us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.9114 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.1636 - 900us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.5402 - 654us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.1404 - 693us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.9133 - 757us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.9527 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.6493 - 607us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 57.7148 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.2004 - 832us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.7673 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.4072 - 570us/epoch - 9us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2896 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.5951 - 623us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.2302 - 744us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.0054 - 950us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7701 - 781us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.8611 - 829us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.8962 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.0691 - 583us/epoch - 9us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.6919 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.0436 - 697us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 146.6904 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.8798 - 720us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.5943 - 694us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.2321 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.7244 - 1ms/epoch - 20us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.6025 - 876us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.2733 - 776us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.0355 - 1ms/epoch - 18us/sample\n", - "Generation 272/300\n", - "Solving for Nash Equilibrium in Generation 272/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.3527 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.6118 - 677us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1962 - 866us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.4974 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7889 - 675us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.6219 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1555 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.9090 - 682us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.3060 - 975us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.6313 - 2ms/epoch - 30us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.0535 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.6016 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.7608 - 1ms/epoch - 19us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.4742 - 847us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355.5608 - 1ms/epoch - 23us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.6620 - 751us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.0160 - 749us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.6351 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.8536 - 1ms/epoch - 24us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.3470 - 783us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.6195 - 802us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.2388 - 627us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.0639 - 741us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.2806 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.1781 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.4128 - 721us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.6312 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.3889 - 878us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5160 - 1ms/epoch - 21us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.8315 - 3ms/epoch - 41us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9903 - 606us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.6957 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.0689 - 880us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3695 - 773us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.3710 - 673us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.0590 - 755us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.3816 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.4088 - 878us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.5928 - 775us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.5202 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.8513 - 959us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.1682 - 789us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.3627 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.7974 - 838us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.3571 - 614us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7183 - 768us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.4641 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.4109 - 687us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.2382 - 800us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.0584 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.3116 - 651us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.4766 - 861us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420.8700 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2063 - 713us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.4374 - 851us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.6136 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.1302 - 797us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.6163 - 632us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.6177 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.2107 - 660us/epoch - 11us/sample\n", - "Generation 273/300\n", - "Solving for Nash Equilibrium in Generation 273/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.2442 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.4078 - 717us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.3336 - 673us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.5424 - 725us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.8862 - 742us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.5949 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.8719 - 778us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.1205 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.4478 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.7436 - 731us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.1725 - 732us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.8656 - 1ms/epoch - 23us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.6184 - 754us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8494 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1735 - 723us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 11ms/epoch - 180us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.3568 - 4ms/epoch - 72us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.7924 - 959us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.9170 - 885us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8472 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.5403 - 565us/epoch - 9us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.9048 - 643us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.9776 - 691us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.7087 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.6295 - 728us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.0874 - 689us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.0516 - 963us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6156 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.2917 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.9701 - 783us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.2330 - 1ms/epoch - 18us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.8516 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.2938 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 108us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.7690 - 2ms/epoch - 28us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1527 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.2090 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.0557 - 600us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.7052 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.8321 - 765us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.2919 - 716us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.3923 - 811us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.3025 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.9562 - 820us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.8168 - 652us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.3150 - 754us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.6867 - 685us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.6773 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.1243 - 700us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0528 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.4283 - 728us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.5508 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.5137 - 782us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.1067 - 886us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.1603 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.1215 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.6717 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656.3896 - 1ms/epoch - 19us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.3579 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.1544 - 791us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.2678 - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425.1007 - 704us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.7440 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.0497 - 754us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.5446 - 1ms/epoch - 20us/sample\n", - "Generation 274/300\n", - "Solving for Nash Equilibrium in Generation 274/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.6630 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.1290 - 626us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 674.0487 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.5924 - 2ms/epoch - 29us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.4790 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.3011 - 950us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.6903 - 618us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.8609 - 628us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.4915 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.4728 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.8464 - 676us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.1008 - 722us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.2324 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.4629 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.5467 - 1ms/epoch - 24us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.1184 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 11ms/epoch - 170us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.2242 - 2ms/epoch - 26us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3389 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.4572 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.6953 - 1ms/epoch - 18us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.5703 - 686us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.6059 - 683us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.0329 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.5186 - 683us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.2747 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.4010 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 733.6992 - 1ms/epoch - 17us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.9974 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.8066 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.2619 - 602us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.1912 - 1ms/epoch - 21us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4665 - 642us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 92us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.5507 - 2ms/epoch - 29us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248.2089 - 675us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.0829 - 2ms/epoch - 26us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.5430 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4320 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.5913 - 694us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.1445 - 1ms/epoch - 21us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.9242 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1261 - 3ms/epoch - 41us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.9720 - 701us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.4606 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.7849 - 1ms/epoch - 20us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.1816 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.0304 - 725us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.3877 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.8765 - 1ms/epoch - 18us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.4183 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.8753 - 805us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.0460 - 566us/epoch - 9us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4684 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0341 - 2ms/epoch - 39us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.1303 - 701us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.8503 - 722us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6243 - 625us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2693 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0911 - 898us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.7458 - 732us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8441 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.1354 - 666us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768.4213 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.9537 - 643us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.8244 - 551us/epoch - 9us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.3890 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.3940 - 648us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.7649 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.8655 - 675us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.2249 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.4598 - 700us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.1219 - 727us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.4148 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408.5618 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5714 - 739us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8815 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.5006 - 793us/epoch - 13us/sample\n", - "Generation 275/300\n", - "Solving for Nash Equilibrium in Generation 275/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.2587 - 2ms/epoch - 27us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.1504 - 784us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.6663 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.3335 - 2ms/epoch - 24us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.4941 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.5593 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.9478 - 826us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.4952 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2431 - 691us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.5358 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.4871 - 703us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.2071 - 866us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4665 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.3433 - 1ms/epoch - 22us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.6107 - 2ms/epoch - 28us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.5321 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.9382 - 2ms/epoch - 34us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.7444 - 640us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.9270 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8202 - 688us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0023 - 800us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.9718 - 563us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.0402 - 702us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.6594 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.1052 - 607us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.5991 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.4541 - 656us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.2331 - 691us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377.2659 - 1ms/epoch - 18us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.4299 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.6627 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.9222 - 1ms/epoch - 19us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.0382 - 737us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.2202 - 1ms/epoch - 17us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.2735 - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.1951 - 847us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.3509 - 761us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.0519 - 892us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.6639 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.0157 - 1ms/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.0756 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8161 - 600us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.8539 - 659us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.6907 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.6183 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.0222 - 758us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.5226 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.3591 - 798us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.9573 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.5215 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6152 - 587us/epoch - 9us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.0720 - 878us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300.1183 - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.0316 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.6527 - 754us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.1076 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1234 - 687us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.7702 - 602us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.0312 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.9466 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6882 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 147us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.0678 - 3ms/epoch - 41us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.5833 - 635us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 711.1565 - 746us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255.8373 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314.9235 - 2ms/epoch - 35us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.2919 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.2862 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.9744 - 2ms/epoch - 38us/sample\n", - "Generation 276/300\n", - "Solving for Nash Equilibrium in Generation 276/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.9423 - 760us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49.1810 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289.0287 - 612us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.3680 - 716us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.3723 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.5318 - 1ms/epoch - 22us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.2391 - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.7903 - 2ms/epoch - 30us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.2162 - 697us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.6365 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.6411 - 699us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.0861 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.4236 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.1535 - 656us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.2963 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.5203 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.0499 - 695us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8162 - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.2298 - 969us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.6989 - 1ms/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.7031 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 413.1872 - 820us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0790 - 662us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 432.3549 - 762us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.9300 - 660us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.3912 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.0024 - 1ms/epoch - 19us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.5075 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.7300 - 705us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.0015 - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.0130 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8665 - 640us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8789 - 949us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.1240 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.4208 - 701us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.2191 - 634us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9732 - 3ms/epoch - 41us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.5148 - 663us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.6691 - 670us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6002 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.5152 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.1089 - 2ms/epoch - 28us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.9288 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.6225 - 850us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.7845 - 752us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.8156 - 768us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441.5827 - 894us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.8864 - 926us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.0331 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.9258 - 628us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.5150 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.0405 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.9134 - 851us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.3390 - 760us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 129us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.4602 - 4ms/epoch - 65us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.5630 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.6909 - 829us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.4811 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.8343 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.8978 - 622us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.8368 - 687us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.2283 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.5816 - 703us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.0063 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9631 - 816us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.7006 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.6316 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.2134 - 2ms/epoch - 28us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.6124 - 811us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 277/300\n", - "Solving for Nash Equilibrium in Generation 277/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.2502 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.1440 - 718us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.5952 - 816us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.9719 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.9772 - 767us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.0632 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 130us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.9546 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2649 - 985us/epoch - 16us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1996 - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.6027 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9944 - 769us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6255 - 970us/epoch - 16us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 148us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.3501 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.0583 - 713us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6534 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.3505 - 1ms/epoch - 22us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.9725 - 801us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.3859 - 901us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0541 - 712us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.6173 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.4652 - 840us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.6363 - 680us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 225.3580 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728.6049 - 721us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.3307 - 605us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.7872 - 786us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.1035 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371.7458 - 714us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1346 - 661us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8199 - 666us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.2768 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.7908 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4321 - 725us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.7225 - 1ms/epoch - 17us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.6045 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.4092 - 758us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.5436 - 5ms/epoch - 76us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.6404 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.0725 - 813us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.4359 - 724us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.6217 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.7226 - 581us/epoch - 9us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 207.0680 - 768us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.5312 - 686us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7214 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.5681 - 770us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.6292 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.0541 - 609us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.4431 - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.1116 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 599.9333 - 2ms/epoch - 35us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.5569 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.1056 - 788us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.5300 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.4863 - 766us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 118us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409.3542 - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.7322 - 722us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.7169 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.8785 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1688 - 686us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.5577 - 563us/epoch - 9us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347.7551 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.4625 - 610us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9032 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 659.6779 - 647us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.8203 - 968us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.3716 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328.6336 - 767us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366.9402 - 593us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.6763 - 2ms/epoch - 31us/sample\n", - "Generation 278/300\n", - "Solving for Nash Equilibrium in Generation 278/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.0381 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422.9046 - 2ms/epoch - 39us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.0818 - 695us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.8686 - 602us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4165 - 692us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 118us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.4949 - 2ms/epoch - 26us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.9452 - 713us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.7892 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.3042 - 879us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.7263 - 789us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.1010 - 803us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.7315 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.8268 - 694us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.9447 - 639us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.2369 - 708us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.0468 - 722us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.4942 - 656us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.2353 - 671us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2631 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4314 - 740us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.1506 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.9340 - 1ms/epoch - 24us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 239.5657 - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.9658 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.7054 - 1ms/epoch - 18us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3086 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.6275 - 1ms/epoch - 16us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.8611 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.0392 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4489 - 1ms/epoch - 17us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.3156 - 717us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.5290 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.1400 - 799us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.4229 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.5732 - 634us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358.4662 - 808us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.7905 - 722us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.6573 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.6128 - 850us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.7224 - 1ms/epoch - 17us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9893 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4717 - 1ms/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.6561 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558.8756 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.8207 - 921us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.3018 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.7938 - 867us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.1509 - 739us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.9520 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.0318 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.6044 - 906us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.3497 - 741us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348.0693 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.1368 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.4984 - 780us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.6703 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.6127 - 703us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1728 - 801us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.7057 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5079 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2137 - 595us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.0582 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.8995 - 590us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.9655 - 902us/epoch - 15us/sample\n", - "Generation 279/300\n", - "Solving for Nash Equilibrium in Generation 279/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465.8466 - 745us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 185.7159 - 2ms/epoch - 28us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.4020 - 2ms/epoch - 38us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.7288 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.6849 - 699us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.4896 - 687us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.4616 - 718us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.9906 - 2ms/epoch - 37us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.1481 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 379.5795 - 720us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700.1495 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.5641 - 730us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.4155 - 709us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.8448 - 698us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.8185 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.8251 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.3698 - 2ms/epoch - 32us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 234.7809 - 703us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.0239 - 823us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.5143 - 686us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.6559 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.0480 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.1771 - 642us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8354 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.6978 - 662us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.9085 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4720 - 743us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.4187 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.2740 - 720us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2783 - 688us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.6632 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6738 - 584us/epoch - 9us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.2410 - 898us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6246 - 674us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.2004 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.7410 - 770us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.1435 - 588us/epoch - 9us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.1151 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.8044 - 745us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.5391 - 634us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.5415 - 566us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1633 - 752us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.2867 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4000 - 705us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.7761 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.1733 - 696us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.5455 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2992 - 586us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.4691 - 665us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.8944 - 757us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.8580 - 971us/epoch - 16us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.3504 - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.8093 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.9802 - 1ms/epoch - 18us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3665 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.3937 - 598us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.8464 - 761us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.3029 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.9200 - 2ms/epoch - 28us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440.7085 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.7076 - 552us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6800 - 848us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.7006 - 685us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.7311 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.5388 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.6259 - 916us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3446 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.7719 - 667us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.5806 - 913us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.9657 - 1ms/epoch - 16us/sample\n", - "Generation 280/300\n", - "Solving for Nash Equilibrium in Generation 280/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.2365 - 721us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.3355 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.0302 - 2ms/epoch - 30us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.4315 - 885us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.7108 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.1665 - 602us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.6071 - 812us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.0710 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622.3260 - 648us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.1620 - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 225.9471 - 940us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561.9802 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.1482 - 635us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.3000 - 686us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7894 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.5828 - 602us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.4739 - 682us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.6229 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.4530 - 721us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.0649 - 3ms/epoch - 56us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.5330 - 683us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.4049 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.0555 - 6ms/epoch - 102us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.1672 - 891us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.8513 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.1485 - 2ms/epoch - 24us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.2824 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2880 - 631us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.6917 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.8804 - 717us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2530 - 838us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.6107 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.2465 - 810us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.3915 - 2ms/epoch - 31us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.5144 - 618us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.6571 - 2ms/epoch - 29us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.5279 - 813us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.1921 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.9430 - 3ms/epoch - 46us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.7971 - 3ms/epoch - 44us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6842 - 783us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 723.1942 - 888us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.1983 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.4993 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.1358 - 673us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.5593 - 562us/epoch - 9us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2190 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.0821 - 895us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706.1249 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.3948 - 913us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.3425 - 718us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.4364 - 734us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8600 - 608us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750.0959 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.0939 - 558us/epoch - 9us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.8930 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.0510 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.6425 - 2ms/epoch - 30us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.8222 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.2509 - 562us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5515 - 1ms/epoch - 19us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4305 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.0275 - 2ms/epoch - 26us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.3388 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.2518 - 757us/epoch - 12us/sample\n", - "Generation 281/300\n", - "Solving for Nash Equilibrium in Generation 281/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.0204 - 582us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.2408 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 140us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.0246 - 997us/epoch - 16us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.1329 - 1ms/epoch - 20us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.4394 - 666us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3260 - 767us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.0928 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.0665 - 760us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.7980 - 720us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250.8750 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.7908 - 771us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.6379 - 1ms/epoch - 20us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.2731 - 640us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.6941 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.0069 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.7910 - 622us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.2582 - 693us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.6202 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.7440 - 661us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.2341 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.1540 - 3ms/epoch - 41us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5841 - 786us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.9190 - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.9883 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.5418 - 625us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240.9090 - 712us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 97.8924 - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.2205 - 761us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.5656 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.2283 - 995us/epoch - 16us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.2487 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5397 - 827us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.3968 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.3355 - 610us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.6743 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.0884 - 875us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.1606 - 599us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.1588 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.8628 - 598us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675.3964 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.7076 - 693us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.1324 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.9202 - 750us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442.8620 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.4461 - 587us/epoch - 9us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68.4759 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.2973 - 743us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.3364 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.5343 - 700us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.8137 - 3ms/epoch - 46us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.4183 - 632us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8565 - 652us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.3318 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.5284 - 758us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.2201 - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.6373 - 808us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.1043 - 821us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.4184 - 1ms/epoch - 17us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.3855 - 806us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.5980 - 2ms/epoch - 33us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.9914 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.9772 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.6143 - 1ms/epoch - 18us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 679.7869 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581.6960 - 1ms/epoch - 18us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.9598 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.2618 - 927us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.5518 - 2ms/epoch - 24us/sample\n", - "Generation 282/300\n", - "Solving for Nash Equilibrium in Generation 282/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201.4706 - 645us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2041 - 607us/epoch - 10us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.9050 - 830us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.7825 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460.4789 - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 142us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.4846 - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288.7498 - 1ms/epoch - 18us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.9601 - 1ms/epoch - 17us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.2314 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4799 - 583us/epoch - 9us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.4602 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.5015 - 691us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.3088 - 1ms/epoch - 17us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.0697 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.7833 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.6477 - 4ms/epoch - 70us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.7535 - 787us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.1902 - 704us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2243 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 598.2917 - 5ms/epoch - 80us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.9747 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.0179 - 784us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.4362 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.8029 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.8811 - 663us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418.4906 - 702us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.7935 - 706us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.1423 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.0682 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.5506 - 960us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.0815 - 682us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.6520 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668.8569 - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.3021 - 776us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.3281 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.9575 - 630us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.7984 - 1ms/epoch - 24us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.3210 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6459 - 8ms/epoch - 129us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 213.7166 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6862 - 923us/epoch - 15us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.2800 - 782us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.4184 - 714us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 697.4760 - 755us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.7698 - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.8522 - 846us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.8654 - 787us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.5568 - 829us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342.2226 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.9565 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8150 - 2ms/epoch - 30us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.9313 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.8219 - 766us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.0969 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.8861 - 772us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.0153 - 960us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.4423 - 815us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.7394 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.6324 - 629us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.9222 - 4ms/epoch - 73us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662.7023 - 741us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.3903 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.3624 - 714us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.7817 - 666us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.1241 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.0693 - 2ms/epoch - 36us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7415 - 863us/epoch - 14us/sample\n", - "Generation 283/300\n", - "Solving for Nash Equilibrium in Generation 283/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.9859 - 746us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.2003 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.9066 - 1ms/epoch - 19us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 707.3770 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.6597 - 1ms/epoch - 18us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 82.3016 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2496 - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.1255 - 685us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.3702 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681.4265 - 613us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676.3737 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9023 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2882 - 2ms/epoch - 26us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.0027 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.6994 - 761us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9529 - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.6990 - 1ms/epoch - 19us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.5968 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.9877 - 668us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.8571 - 618us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.6818 - 9ms/epoch - 146us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245.7813 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0275 - 1ms/epoch - 17us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.7501 - 649us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.1519 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.2490 - 6ms/epoch - 104us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.4868 - 799us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580.9151 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.4773 - 3ms/epoch - 45us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.2136 - 2ms/epoch - 30us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.6628 - 802us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.5949 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.5504 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.3024 - 709us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 716.0488 - 797us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554.0833 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.9995 - 768us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 128us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.5329 - 3ms/epoch - 56us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.2005 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.1924 - 683us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.1882 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.1805 - 659us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.7783 - 659us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.8396 - 3ms/epoch - 54us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 94.7427 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.0976 - 758us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.4802 - 564us/epoch - 9us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.2786 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.1889 - 750us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.9435 - 707us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.0371 - 594us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.9688 - 699us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.0107 - 632us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.9377 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.0451 - 651us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.4497 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.0789 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.5298 - 658us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.7786 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.4895 - 620us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.7077 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593.7160 - 776us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 614.5399 - 795us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.3220 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.3823 - 764us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 126us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.2122 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.3116 - 700us/epoch - 11us/sample\n", - "Generation 284/300\n", - "Solving for Nash Equilibrium in Generation 284/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.2204 - 759us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 128us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.2006 - 2ms/epoch - 26us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310.8250 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.9641 - 622us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 754.5701 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.9607 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 432.6996 - 582us/epoch - 9us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.7563 - 626us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3309 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.5884 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.9552 - 588us/epoch - 9us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.0010 - 849us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.8676 - 619us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.3185 - 892us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 541.2461 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.1728 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.1215 - 627us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.3323 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.7910 - 3ms/epoch - 42us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.3309 - 691us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3782 - 690us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.9021 - 792us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.9371 - 869us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.8715 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.7928 - 688us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.6202 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.5513 - 748us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.4816 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.4294 - 719us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339.2344 - 629us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335.6691 - 582us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316.8285 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.4460 - 3ms/epoch - 44us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.6146 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.8051 - 677us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3873 - 2ms/epoch - 28us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.0563 - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.1646 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.0621 - 637us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.4932 - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.4589 - 1ms/epoch - 19us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587.5678 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.9797 - 1ms/epoch - 24us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.2122 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3540 - 919us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.1796 - 803us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631.2349 - 907us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.3725 - 682us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.5362 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.5310 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7499 - 703us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.8712 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.8204 - 841us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.9057 - 3ms/epoch - 49us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.9230 - 2ms/epoch - 30us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.6604 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8669 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8065 - 723us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496.5735 - 642us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.2034 - 775us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4702 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.6419 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727.4885 - 742us/epoch - 12us/sample\n", - "Generation 285/300\n", - "Solving for Nash Equilibrium in Generation 285/300\n", + "Generation 69/300\n", + "Solving for Nash Equilibrium in Generation 69/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.7795 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.3723 - 832us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37.5393 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470.0734 - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.9687 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.3064 - 627us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588.3138 - 581us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.0787 - 2ms/epoch - 34us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.9968 - 568us/epoch - 9us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.8088 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.5926 - 687us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.9296 - 823us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.4266 - 874us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.0742 - 855us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.3234 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.1371 - 745us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.2667 - 872us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.6350 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.8979 - 810us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.9398 - 792us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.7907 - 796us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.1096 - 718us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.0131 - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.6301 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.2850 - 2ms/epoch - 35us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.4650 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436.0374 - 2ms/epoch - 31us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.9651 - 848us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.0206 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264.2245 - 4ms/epoch - 58us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.3455 - 10ms/epoch - 154us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.8112 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.1714 - 671us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.4096 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.1087 - 758us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.3441 - 655us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.0941 - 894us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7205 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.9147 - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 131us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1508 - 5ms/epoch - 74us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.9905 - 926us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.3648 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.2946 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.8183 - 773us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.7015 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380.1765 - 851us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.2131 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.4330 - 970us/epoch - 16us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.6856 - 904us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35.6194 - 6ms/epoch - 102us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 253.3007 - 1ms/epoch - 19us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.0244 - 651us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.4968 - 656us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296.4969 - 830us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.6953 - 659us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.4158 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680.0135 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373.4787 - 865us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.2612 - 593us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.8085 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594.3720 - 670us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 286/300\n", - "Solving for Nash Equilibrium in Generation 286/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.8361 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.6907 - 873us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.1248 - 768us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.0240 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.6917 - 1ms/epoch - 18us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.3369 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.4708 - 939us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 717.5502 - 830us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.6219 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.4449 - 749us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.3088 - 740us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.9514 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.2338 - 727us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.2764 - 1ms/epoch - 23us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.9625 - 715us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 262.9796 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.9222 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.8080 - 803us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.2914 - 1ms/epoch - 17us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6996 - 675us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.0292 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.0283 - 908us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.2881 - 586us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.1909 - 2ms/epoch - 39us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.8235 - 892us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.9081 - 694us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667.0496 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.3505 - 694us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.9298 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.2690 - 618us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.8300 - 942us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.5469 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467.0268 - 658us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713.0562 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.1775 - 902us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.7064 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0974 - 689us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.3442 - 760us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.9362 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.4103 - 636us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265.6584 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0213 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399.8616 - 2ms/epoch - 36us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1867 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.2767 - 622us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2623 - 790us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360.7293 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.8245 - 579us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.6155 - 676us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250.6891 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 105us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.9980 - 2ms/epoch - 31us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.7117 - 693us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.0974 - 699us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337.9493 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.7991 - 954us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311.7764 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.4938 - 878us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6565 - 760us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.2229 - 667us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.0854 - 565us/epoch - 9us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.7926 - 7ms/epoch - 109us/sample\n", - "Generation 287/300\n", - "Solving for Nash Equilibrium in Generation 287/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.5097 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 403.6659 - 922us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.6785 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.6760 - 1ms/epoch - 17us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2218 - 973us/epoch - 16us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.1180 - 4ms/epoch - 66us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.9170 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.7748 - 1ms/epoch - 23us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.3571 - 872us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8097 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.2518 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.8825 - 696us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.4157 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.2126 - 622us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323.7906 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.8722 - 723us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.6446 - 708us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.6382 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.2049 - 692us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.3238 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439.0863 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6362 - 939us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.6790 - 619us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 113us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326.6340 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.3243 - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.4057 - 728us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.8008 - 1ms/epoch - 19us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.8477 - 744us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.4530 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.6425 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 233.3021 - 536us/epoch - 9us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.5502 - 701us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.6986 - 807us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.2926 - 1ms/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 577.1547 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.7450 - 739us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.3241 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.8404 - 713us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381.7829 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4557 - 650us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260.8095 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.2944 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.7059 - 1ms/epoch - 22us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.7155 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.7339 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.4060 - 805us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.8608 - 675us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 223.2663 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 388.6570 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.2411 - 697us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 114us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.0356 - 11ms/epoch - 174us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.1270 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528.3103 - 555us/epoch - 9us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.9454 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 233.5658 - 18ms/epoch - 285us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.4057 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303.5672 - 667us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424.2547 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.7682 - 773us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47.0850 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.2404 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.1376 - 1ms/epoch - 22us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.2518 - 913us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.7343 - 933us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.5964 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.6367 - 1ms/epoch - 24us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.3511 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.8743 - 618us/epoch - 10us/sample\n", - "Generation 288/300\n", - "Solving for Nash Equilibrium in Generation 288/300\n", + "Generation 70/300\n", + "Solving for Nash Equilibrium in Generation 70/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.2918 - 767us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.5445 - 548us/epoch - 9us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395.5714 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.9531 - 651us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433.1161 - 751us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.8714 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.7135 - 914us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 153.3092 - 646us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.2621 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.0400 - 1ms/epoch - 19us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0692 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.5730 - 683us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.5786 - 636us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329.0647 - 7ms/epoch - 116us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5911 - 795us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.7611 - 6ms/epoch - 98us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.3175 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 12ms/epoch - 195us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.4806 - 2ms/epoch - 35us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.9090 - 802us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.5828 - 4ms/epoch - 58us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.7000 - 789us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.9417 - 821us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.2582 - 707us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.5400 - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.4623 - 593us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.8488 - 859us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 427.0891 - 771us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.1647 - 3ms/epoch - 53us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.8705 - 777us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.1851 - 950us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 641.0267 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.9295 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.9916 - 1ms/epoch - 20us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 164.1413 - 937us/epoch - 15us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.0120 - 908us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.3763 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.6865 - 927us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.1721 - 624us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.8241 - 552us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.3738 - 595us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.9499 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.3050 - 754us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.5970 - 711us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.1164 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.7445 - 877us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.7018 - 2ms/epoch - 32us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.1462 - 1ms/epoch - 18us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.0549 - 760us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.7245 - 882us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.0194 - 946us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649.2711 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.2019 - 1ms/epoch - 16us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9276 - 1ms/epoch - 20us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.5784 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.5984 - 1ms/epoch - 21us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.9722 - 1ms/epoch - 19us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.7821 - 971us/epoch - 16us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 289/300\n", - "Solving for Nash Equilibrium in Generation 289/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 11ms/epoch - 178us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.8477 - 11ms/epoch - 179us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 144us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.9128 - 2ms/epoch - 33us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254.1751 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 435.6000 - 710us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.4126 - 732us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689.8875 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.6844 - 3ms/epoch - 43us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350.0612 - 2ms/epoch - 24us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363.5245 - 743us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.3629 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758.9636 - 700us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.4429 - 798us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 13ms/epoch - 214us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.1186 - 3ms/epoch - 42us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290.4043 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 16ms/epoch - 259us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568.2307 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 9ms/epoch - 143us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.5257 - 1ms/epoch - 16us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411.5776 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.1307 - 785us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.1276 - 730us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.2114 - 806us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.2535 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.0844 - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 11ms/epoch - 181us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.0556 - 9ms/epoch - 152us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.5245 - 906us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.6418 - 796us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3768 - 833us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.3746 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423.7080 - 681us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.0316 - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527.4798 - 2ms/epoch - 34us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39.3527 - 841us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.6074 - 1ms/epoch - 23us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.4208 - 776us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 121us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41.7300 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.1107 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 16ms/epoch - 255us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.3304 - 1ms/epoch - 18us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.0462 - 635us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438.2468 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.6100 - 1ms/epoch - 17us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.5074 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.2957 - 721us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.9427 - 713us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.2034 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572.6345 - 1ms/epoch - 23us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9390 - 1ms/epoch - 24us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.9017 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.1114 - 893us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.6143 - 729us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.0132 - 910us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.7827 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.6752 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.0298 - 753us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.4245 - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431.7836 - 678us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9829 - 1ms/epoch - 17us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.9091 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397.7200 - 2ms/epoch - 26us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291.5141 - 1ms/epoch - 20us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.4072 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.3575 - 721us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.7602 - 1ms/epoch - 17us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741.5078 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 576.8163 - 686us/epoch - 11us/sample\n", - "Generation 290/300\n", - "Solving for Nash Equilibrium in Generation 290/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.8231 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.3536 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 13ms/epoch - 209us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584.8383 - 1ms/epoch - 20us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.5381 - 699us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.0793 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269.9148 - 599us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456.1642 - 843us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.8895 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.9738 - 3ms/epoch - 56us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.1801 - 2ms/epoch - 32us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.1303 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.2778 - 3ms/epoch - 51us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.7219 - 698us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.1985 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497.0194 - 764us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.3450 - 731us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261.7315 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.4581 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 721.2116 - 2ms/epoch - 27us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.5577 - 1ms/epoch - 19us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8309 - 595us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.4868 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.4894 - 578us/epoch - 9us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.0050 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.0514 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.9166 - 668us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.1006 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.9066 - 3ms/epoch - 43us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.5697 - 696us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.5775 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386.8150 - 659us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 188.1524 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552.8127 - 815us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.7675 - 698us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.6620 - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.0385 - 3ms/epoch - 46us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.3338 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.4555 - 704us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8919 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.6172 - 624us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.7778 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 648.9186 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.2574 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.6887 - 673us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.3021 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.2311 - 899us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365.8792 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.5899 - 4ms/epoch - 70us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565.0002 - 1ms/epoch - 23us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.8016 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.0854 - 612us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.4799 - 2ms/epoch - 27us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.8925 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4106 - 2ms/epoch - 40us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312.6138 - 1ms/epoch - 23us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.8250 - 713us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.2846 - 2ms/epoch - 28us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.7066 - 901us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443.8437 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 407.9537 - 4ms/epoch - 63us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402.7793 - 731us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.6028 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.4208 - 846us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.3568 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.2422 - 939us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.5140 - 627us/epoch - 10us/sample\n", - "Generation 291/300\n", - "Solving for Nash Equilibrium in Generation 291/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624.0083 - 745us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.5807 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6684 - 730us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.9236 - 857us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338.7137 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.1712 - 1ms/epoch - 20us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.1511 - 2ms/epoch - 28us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.8335 - 682us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.9061 - 34ms/epoch - 542us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.5449 - 761us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 174.6336 - 3ms/epoch - 43us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.5813 - 6ms/epoch - 94us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.6268 - 1ms/epoch - 19us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3134 - 859us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36.4086 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.3561 - 643us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.2975 - 659us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.6463 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.2993 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333.0321 - 667us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.6226 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 118us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.3955 - 1ms/epoch - 22us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.5162 - 745us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.8201 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.2771 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 652.2577 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.7290 - 828us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.7023 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.6740 - 4ms/epoch - 59us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.2821 - 2ms/epoch - 32us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.8915 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699.6909 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.8873 - 841us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5956 - 799us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.4402 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.1461 - 705us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.3546 - 719us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.8770 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.6966 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.7523 - 1ms/epoch - 17us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.7402 - 719us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 89us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353.2758 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.6119 - 1ms/epoch - 23us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4936 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.8611 - 1ms/epoch - 17us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.7912 - 583us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.4693 - 633us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616.1990 - 690us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.0938 - 743us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 141.1749 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.9155 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 212.6171 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.8931 - 639us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.1862 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.3055 - 713us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.4992 - 625us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.0079 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.8364 - 599us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.1610 - 699us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 130us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.0697 - 2ms/epoch - 31us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.0097 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.0262 - 766us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.9665 - 696us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.2296 - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.0061 - 658us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.1688 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.7081 - 742us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.9906 - 678us/epoch - 11us/sample\n", - "Generation 292/300\n", - "Solving for Nash Equilibrium in Generation 292/300\n", + "Generation 71/300\n", + "Solving for Nash Equilibrium in Generation 71/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.5258 - 6ms/epoch - 89us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571.5147 - 652us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4734 - 638us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235.2899 - 643us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30.9745 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.8463 - 580us/epoch - 9us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.1866 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669.5080 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.0054 - 976us/epoch - 16us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.2941 - 3ms/epoch - 40us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.4011 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664.7517 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6320 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.3268 - 685us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651.6013 - 844us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.7520 - 696us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 127us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.5807 - 1ms/epoch - 20us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.8043 - 4ms/epoch - 65us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38.2333 - 753us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259.3655 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.5427 - 632us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2753 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.2399 - 658us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.1848 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 597.3925 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.4305 - 582us/epoch - 9us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.2086 - 1ms/epoch - 18us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396.8848 - 800us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.2969 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508.1218 - 718us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 570.4729 - 767us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767.1093 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.3754 - 619us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.4271 - 675us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352.7620 - 609us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.0295 - 712us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.6390 - 726us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.5846 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.9451 - 709us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.4512 - 753us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.0099 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.2758 - 1ms/epoch - 21us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.3042 - 696us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.3111 - 644us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42.1078 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.6677 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.6118 - 573us/epoch - 9us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.1998 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 183.5470 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.0670 - 5ms/epoch - 75us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.4204 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.6253 - 654us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3815 - 1ms/epoch - 20us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.9330 - 632us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 253.3247 - 563us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543.4738 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 210.7021 - 6ms/epoch - 94us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.3716 - 3ms/epoch - 48us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 421.0270 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304.7947 - 1ms/epoch - 17us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.3177 - 659us/epoch - 11us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 293/300\n", - "Solving for Nash Equilibrium in Generation 293/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471.4898 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.8752 - 750us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.1846 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.7201 - 669us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 744.1115 - 3ms/epoch - 53us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.6815 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.2242 - 674us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.0237 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.2277 - 715us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.8180 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.5428 - 2ms/epoch - 31us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.3773 - 743us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.2645 - 2ms/epoch - 29us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.9002 - 833us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.9886 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.4680 - 943us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445.0404 - 707us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.2441 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.9765 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 661.7610 - 692us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611.1238 - 841us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.2805 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629.2620 - 681us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401.5588 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.8919 - 614us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361.6769 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.3718 - 798us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.4478 - 791us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.2927 - 744us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356.7478 - 1ms/epoch - 17us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.5875 - 738us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.0086 - 694us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 685.1830 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.7636 - 600us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267.5010 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.5675 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.6502 - 957us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.7802 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.6475 - 4ms/epoch - 58us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.1804 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324.8357 - 565us/epoch - 9us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.7593 - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.4402 - 850us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.1429 - 789us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334.9698 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.6664 - 770us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.7586 - 922us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760.3425 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.0541 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.4116 - 1ms/epoch - 17us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.1331 - 659us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688.6259 - 543us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.4530 - 626us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8629 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.5890 - 758us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735.4236 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.8156 - 2ms/epoch - 39us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.2159 - 746us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.6316 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.8007 - 1ms/epoch - 18us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.9828 - 2ms/epoch - 29us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372.3074 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387.7089 - 810us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231.7866 - 672us/epoch - 11us/sample\n", - "Generation 294/300\n", - "Solving for Nash Equilibrium in Generation 294/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.5905 - 740us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364.8820 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.4068 - 690us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.1437 - 720us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.7797 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489.8765 - 807us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.7604 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 216.0269 - 1ms/epoch - 20us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.3766 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560.9428 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628.3181 - 655us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.0086 - 576us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.3416 - 2ms/epoch - 25us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 732.9797 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.1660 - 634us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.2104 - 602us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.3301 - 675us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368.0673 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.7419 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.8688 - 772us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.6304 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.4688 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.5844 - 635us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 498.8271 - 724us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.9493 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.9203 - 965us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702.5211 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.9767 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426.0484 - 726us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512.4863 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.6785 - 2ms/epoch - 25us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469.6764 - 726us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.3913 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.0568 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6326 - 732us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.4080 - 767us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718.7950 - 1ms/epoch - 19us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.4125 - 7ms/epoch - 119us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646.5082 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532.6970 - 693us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.8739 - 739us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317.7241 - 649us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.6428 - 921us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719.8817 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 636.4338 - 672us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.5770 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 429.8895 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516.2999 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.0007 - 688us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 567.2843 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.0417 - 715us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.5022 - 756us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.8085 - 696us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301.9022 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.7361 - 731us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.4035 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.6678 - 776us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4728 - 674us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246.9712 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.8622 - 10ms/epoch - 162us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.3574 - 634us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.4604 - 781us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5977 - 749us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258.7140 - 855us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270.4865 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.7164 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.5835 - 684us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639.5588 - 609us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.6852 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.2414 - 6ms/epoch - 102us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281.1775 - 664us/epoch - 11us/sample\n", - "Generation 295/300\n", - "Solving for Nash Equilibrium in Generation 295/300\n", + "Generation 72/300\n", + "Solving for Nash Equilibrium in Generation 72/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.9600 - 762us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287.5749 - 759us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.8226 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.6456 - 2ms/epoch - 26us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244.4549 - 2ms/epoch - 25us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.2560 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.6104 - 705us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.6135 - 616us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575.2053 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.5369 - 693us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 83us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.9526 - 1ms/epoch - 18us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.6084 - 849us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.9271 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235.4726 - 648us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251.3011 - 789us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 739.6884 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.0715 - 656us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.2066 - 8ms/epoch - 123us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5675 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559.5320 - 655us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.6140 - 1ms/epoch - 17us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.1931 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736.6569 - 610us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.8153 - 813us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.5793 - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375.7740 - 851us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692.8233 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273.0339 - 749us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.7148 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 404.7835 - 801us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.8996 - 993us/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286.2021 - 607us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.9886 - 768us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.3571 - 682us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708.5719 - 574us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.0353 - 686us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268.5297 - 2ms/epoch - 31us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.8575 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.0513 - 651us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522.4388 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672.0671 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.1459 - 827us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9622 - 2ms/epoch - 28us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579.7276 - 2ms/epoch - 37us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392.4626 - 884us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.0853 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479.3046 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.0521 - 601us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.0789 - 3ms/epoch - 53us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.4117 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9483 - 616us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.7488 - 806us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.1673 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349.4977 - 3ms/epoch - 48us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.8100 - 730us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 653.7391 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695.9523 - 4ms/epoch - 64us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.8305 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243.3559 - 585us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357.7169 - 574us/epoch - 9us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.3589 - 704us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660.4737 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.6313 - 1ms/epoch - 21us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472.4028 - 876us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.4915 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.4894 - 659us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461.5875 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.8918 - 681us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607.9556 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724.2642 - 1ms/epoch - 18us/sample\n", - "Generation 296/300\n", - "Solving for Nash Equilibrium in Generation 296/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.8058 - 710us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336.2985 - 732us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.2100 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583.9872 - 537us/epoch - 9us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709.2961 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 715.4185 - 654us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476.0268 - 2ms/epoch - 28us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.6735 - 691us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2985 - 673us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.0174 - 826us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678.3450 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.1726 - 661us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.2383 - 653us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553.3019 - 797us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500.7365 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.5587 - 666us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504.1533 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.3024 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637.6946 - 4ms/epoch - 64us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533.5649 - 1ms/epoch - 22us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731.5056 - 597us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434.9289 - 608us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.6585 - 911us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6091 - 2ms/epoch - 25us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.7613 - 716us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.1459 - 646us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 590.4678 - 768us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492.1864 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 563.4466 - 657us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.5560 - 2ms/epoch - 33us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615.9888 - 855us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35.0815 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604.0304 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 573.4483 - 655us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.7570 - 3ms/epoch - 56us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.8692 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493.7519 - 724us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330.3864 - 1ms/epoch - 16us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617.3687 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 524.0757 - 725us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704.9401 - 871us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351.3992 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275.4955 - 1ms/epoch - 23us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 746.1452 - 713us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.0478 - 732us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.6642 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.0386 - 1ms/epoch - 19us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725.6550 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.0977 - 877us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 645.0825 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.0258 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8643 - 770us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.7189 - 956us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.6473 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9734 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.0518 - 935us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.7525 - 3ms/epoch - 51us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519.7664 - 746us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.5378 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.1991 - 880us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.2380 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 633.8723 - 2ms/epoch - 26us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8314 - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.3571 - 928us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451.3643 - 945us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486.0511 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.4740 - 3ms/epoch - 49us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.1324 - 1ms/epoch - 18us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.7947 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315.8719 - 908us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.9992 - 846us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 297/300\n", - "Solving for Nash Equilibrium in Generation 297/300\n", + "Generation 73/300\n", + "Solving for Nash Equilibrium in Generation 73/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.6277 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 177.6342 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 84us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343.4575 - 4ms/epoch - 68us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 654.5011 - 851us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369.9113 - 853us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398.6039 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.3488 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283.4766 - 2ms/epoch - 36us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544.2060 - 1ms/epoch - 21us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383.6287 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.1168 - 873us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501.3577 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523.2547 - 761us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.2314 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.6981 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.6076 - 674us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 111us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.3563 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.6433 - 3ms/epoch - 51us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299.9958 - 1ms/epoch - 17us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8331 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 226.0677 - 631us/epoch - 10us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 457.2002 - 712us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566.8126 - 731us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502.9332 - 731us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280.0924 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.0102 - 1ms/epoch - 17us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696.2596 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.0577 - 720us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 432.8532 - 2ms/epoch - 34us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284.0345 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6240 - 603us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 529.6959 - 2ms/epoch - 30us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256.5952 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.0853 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720.1152 - 685us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.0751 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621.4149 - 784us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474.3958 - 614us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 92us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477.0042 - 957us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307.5851 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480.4870 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585.3235 - 4ms/epoch - 68us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.9692 - 706us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 122us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608.8765 - 7ms/epoch - 116us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.2404 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.6072 - 2ms/epoch - 31us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454.3183 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693.1923 - 850us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.6546 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 72ms/epoch - 1ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.7787 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.6541 - 2ms/epoch - 37us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.0737 - 643us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.2760 - 1ms/epoch - 23us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98.9337 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382.7783 - 858us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647.1547 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.3539 - 4ms/epoch - 66us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.5553 - 1ms/epoch - 22us/sample\n", - "Generation 298/300\n", - "Solving for Nash Equilibrium in Generation 298/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.8488 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545.2293 - 755us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.5003 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.5816 - 1ms/epoch - 18us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609.0994 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562.2281 - 764us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.6225 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.2127 - 5ms/epoch - 88us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.2520 - 735us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626.2302 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.0539 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535.4633 - 2ms/epoch - 40us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 582.7436 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 694.0559 - 627us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9267 - 656us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487.5367 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285.2971 - 709us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506.1842 - 769us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 740.4257 - 756us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 665.8144 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400.3322 - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510.5401 - 858us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.6709 - 2ms/epoch - 38us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525.6832 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.5038 - 1ms/epoch - 17us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.1329 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.4738 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453.1758 - 737us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505.9300 - 751us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.1714 - 757us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.0380 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.5346 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.0549 - 821us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670.5856 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.8002 - 679us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530.6573 - 877us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390.8352 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.1349 - 900us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.4963 - 1ms/epoch - 21us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.4464 - 775us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367.3054 - 10ms/epoch - 155us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.4842 - 740us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 106us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385.8281 - 2ms/epoch - 36us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444.0495 - 580us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618.7357 - 711us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491.9433 - 882us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344.6006 - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673.5385 - 740us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677.7379 - 6ms/epoch - 105us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 107us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503.1673 - 18ms/epoch - 293us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537.2356 - 1ms/epoch - 22us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542.5548 - 740us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 632.0587 - 794us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463.2847 - 829us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.9167 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.4291 - 924us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625.1128 - 691us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 253.8620 - 1ms/epoch - 20us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313.4278 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 526.1971 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295.1244 - 772us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.9443 - 702us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.5250 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520.1666 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.8104 - 767us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.9886 - 684us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Generation 299/300\n", - "Solving for Nash Equilibrium in Generation 299/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555.9319 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237.4574 - 639us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449.5980 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 710.6464 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 605.8882 - 684us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.8586 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.6224 - 663us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462.1505 - 707us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540.0408 - 617us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.9184 - 2ms/epoch - 35us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 85us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71.6418 - 1ms/epoch - 22us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663.7889 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.3334 - 574us/epoch - 9us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9961 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468.5840 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515.6437 - 753us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.4315 - 777us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539.7919 - 788us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 705.6194 - 712us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65.4748 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606.3333 - 5ms/epoch - 80us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.3534 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.5088 - 558us/epoch - 9us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.8622 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.4404 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634.3533 - 727us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473.1196 - 787us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277.1144 - 1ms/epoch - 17us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 748.4926 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231.1246 - 653us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412.7679 - 964us/epoch - 16us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 591.4774 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.0672 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 596.6172 - 796us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.6138 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.0558 - 1ms/epoch - 18us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 687.8897 - 809us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331.9282 - 718us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.5106 - 579us/epoch - 9us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257.0688 - 652us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247.0909 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 745.9018 - 772us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703.5511 - 681us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635.3724 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.6617 - 7ms/epoch - 111us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455.6988 - 1ms/epoch - 19us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405.7538 - 623us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446.8160 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620.8641 - 762us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282.7329 - 6ms/epoch - 101us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279.5109 - 691us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.8193 - 649us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.8975 - 776us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517.5621 - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.5549 - 2ms/epoch - 38us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437.9832 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236.1839 - 689us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 551.5959 - 660us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252.5390 - 1ms/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638.2511 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478.4591 - 606us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.1153 - 695us/epoch - 11us/sample\n", - "Generation 300/300\n", - "Solving for Nash Equilibrium in Generation 300/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569.1194 - 889us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292.7347 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.8500 - 1ms/epoch - 21us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595.9106 - 1ms/epoch - 21us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 447.2131 - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684.6676 - 7ms/epoch - 107us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 546.9655 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531.5606 - 673us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 341.9234 - 713us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658.2402 - 2ms/epoch - 26us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.4961 - 909us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271.3861 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475.3306 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 481.8119 - 704us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 701.7292 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.2359 - 549us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507.6337 - 1ms/epoch - 19us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490.8586 - 855us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274.6914 - 626us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.8013 - 690us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 690.1117 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 698.0403 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 253.3733 - 611us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691.4751 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650.6062 - 654us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 642.8953 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 509.9072 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 459.3140 - 3ms/epoch - 42us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293.7855 - 953us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514.0490 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538.4296 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 601.7597 - 766us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.6808 - 679us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.1360 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644.9742 - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374.8607 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602.9407 - 676us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610.3088 - 705us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294.6318 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.9164 - 738us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394.1826 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242.7150 - 920us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458.5810 - 4ms/epoch - 72us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308.0615 - 2ms/epoch - 27us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.5328 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603.1154 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448.1855 - 654us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 655.3441 - 866us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.1973 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452.1542 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 113us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485.9888 - 2ms/epoch - 30us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556.8896 - 2ms/epoch - 25us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263.8991 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484.2911 - 702us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482.4423 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521.1791 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.6540 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549.8536 - 606us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 547.8778 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 613.8491 - 845us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550.1566 - 664us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671.2555 - 6ms/epoch - 96us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534.7408 - 620us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249.6153 - 559us/epoch - 9us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511.0959 - 7ms/epoch - 120us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518.4940 - 700us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666.6506 - 780us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586.4875 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623.8998 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499.4393 - 942us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513.7951 - 689us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 734.5797 - 605us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 536.2237 - 960us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683.9996 - 713us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415.3363 - 723us/epoch - 12us/sample\n", - "Running Computed Policies\n", - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 3 done\n", - "Episode 3 done\n", - "Total steps: 13\n", - "Episode 4 done\n", - "Episode 5 done\n", - "Episode 6 done\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 9 done\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/300\n", - "Solving for Nash Equilibrium in Generation 1/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: inf - 198ms/epoch - 3ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2559.8262 - 193ms/epoch - 3ms/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: inf - 212ms/epoch - 3ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.2522 - 204ms/epoch - 3ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: inf - 228ms/epoch - 4ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.8218 - 224ms/epoch - 4ms/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.2856 - 724us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: inf - 249ms/epoch - 4ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 220.1969 - 240ms/epoch - 4ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1334.5574 - 659us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2571.7368 - 739us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2513.5444 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2864.2744 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2641.7437 - 729us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.3757 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3448.6873 - 615us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.3159 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2899.1567 - 750us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3655.5291 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3505.3931 - 689us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1735.4155 - 712us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3564.7976 - 752us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682.4416 - 701us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3548.3989 - 700us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3447.5420 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2611.4851 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3623.7803 - 825us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.0566 - 663us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.7913 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2973.9500 - 638us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3326.7256 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2666.9424 - 701us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3099.0234 - 727us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340.7903 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2613.7993 - 805us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3099.5830 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3880.7854 - 732us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.2280 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3814.0354 - 675us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 808.6174 - 647us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2968.1973 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3748.9204 - 689us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2741.4839 - 701us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3087.1106 - 762us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2151.3992 - 641us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1360.0835 - 571us/epoch - 9us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2591.7031 - 721us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.3352 - 667us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2580.3630 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2627.1638 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2269.0669 - 757us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2417.3755 - 904us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3934.9248 - 685us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.9607 - 684us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2541.7795 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2687.2437 - 630us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2549.1108 - 2ms/epoch - 31us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2700.6287 - 738us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.6826 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2800.0359 - 698us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2326.1013 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.0104 - 734us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3340.3298 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3341.7261 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1458.1733 - 731us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3844.7366 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2636.8176 - 690us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3655.5906 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1433.9813 - 743us/epoch - 12us/sample\n", - "Generation 2/300\n", - "Solving for Nash Equilibrium in Generation 2/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.4709 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2641.8091 - 677us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.3076 - 665us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2647.0146 - 644us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2627.8816 - 781us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2540.3914 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3553.7378 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3013.8640 - 910us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3655.5439 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2630.8582 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3685.5403 - 803us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2611.5071 - 599us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3014.8447 - 633us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 222.0796 - 687us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2561.7087 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2669.7659 - 1ms/epoch - 22us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2563.8088 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3685.5383 - 879us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2546.2234 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2674.6592 - 663us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2960.2974 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2645.5449 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2567.3403 - 750us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3056.1060 - 658us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3098.8198 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2407.0168 - 679us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3217.2249 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3815.2805 - 922us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2694.5156 - 747us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2538.9343 - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2960.3167 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2094.5806 - 641us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1270.6653 - 825us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2558.8215 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2139.5356 - 635us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2287.1123 - 667us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2968.2510 - 805us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2410.7183 - 710us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2843.7803 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.7932 - 827us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2558.7764 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2421.0874 - 662us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1493.7343 - 2ms/epoch - 37us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2692.6277 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2558.3875 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.7534 - 798us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.8687 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2551.9607 - 702us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2831.1169 - 722us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.1738 - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3393.6936 - 658us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.4226 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2541.7021 - 643us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 589.7845 - 836us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1798.0844 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1462.8951 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3506.6809 - 932us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2666.7324 - 745us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 204.3427 - 761us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2179.7700 - 724us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2596.8171 - 816us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2556.6062 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1092.1888 - 699us/epoch - 11us/sample\n", - "Generation 3/300\n", - "Solving for Nash Equilibrium in Generation 3/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1336.3831 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2559.2778 - 730us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2537.0000 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2629.1294 - 756us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2973.9946 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3013.9172 - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1881.3251 - 673us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2620.1135 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3623.8083 - 654us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3209.7632 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2005.2843 - 650us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2663.3464 - 632us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3087.0850 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2625.0356 - 621us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327.9246 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3814.0532 - 647us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3098.7566 - 766us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2961.4075 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3814.0588 - 770us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3209.2783 - 774us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2968.3384 - 716us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2549.2966 - 643us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.6826 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2742.9192 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 934.4835 - 630us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.3413 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2741.4932 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2618.2134 - 661us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2350.2959 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3949.9822 - 644us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.0220 - 645us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2577.5686 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2831.1851 - 787us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3289.5618 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3393.7021 - 624us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3753.0701 - 819us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 219.5028 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359.6494 - 989us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.0706 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2831.0862 - 790us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1583.6630 - 708us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2549.3037 - 745us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2292.4272 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 619.0607 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1448.7162 - 671us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2538.4871 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2899.1934 - 735us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3651.5237 - 1ms/epoch - 16us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2608.6682 - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 830.4129 - 2ms/epoch - 31us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2168.9590 - 806us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2600.7119 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3749.0884 - 670us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2558.4199 - 671us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3623.8064 - 712us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1762.3201 - 1ms/epoch - 23us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1474.7971 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2663.7073 - 743us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3880.9399 - 935us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3334.7263 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3217.5410 - 768us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2836.8291 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2667.7458 - 716us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2690.2007 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3098.7920 - 759us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3143.7366 - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3814.0867 - 730us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2772.1033 - 769us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3880.9363 - 757us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 194.2561 - 708us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Generation 4/300\n", - "Solving for Nash Equilibrium in Generation 4/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.7312 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2828.9712 - 721us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2667.0471 - 706us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630.4095 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2340.0388 - 684us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1173.0101 - 722us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.3030 - 699us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3934.5349 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2454.6130 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1158.4799 - 636us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2540.1150 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1250.2802 - 689us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.9688 - 633us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2633.8650 - 638us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2551.0159 - 2ms/epoch - 38us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3847.6934 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2407.2007 - 692us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3844.8018 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1434.2994 - 676us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.5427 - 797us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2684.8132 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2596.5322 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1448.4821 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2647.0308 - 711us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3651.4775 - 745us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2487.4641 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2565.2432 - 815us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3553.8499 - 621us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1662.8517 - 654us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1762.2454 - 698us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.4824 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2538.2000 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2672.0002 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2663.2820 - 677us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3218.7632 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1897.3783 - 802us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3334.8691 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2690.3665 - 719us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.1667 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3144.1865 - 640us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3209.2864 - 746us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2619.1697 - 710us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.7834 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2538.0232 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2554.6526 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3949.2214 - 646us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.4404 - 722us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2714.9973 - 826us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2539.4514 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3190.4668 - 816us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3949.3364 - 750us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.4097 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 934.2277 - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2693.0081 - 733us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2692.9893 - 791us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.0828 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.8979 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.0850 - 1ms/epoch - 17us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2831.1506 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3882.0442 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3396.0127 - 1ms/epoch - 20us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2541.9402 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2843.8032 - 746us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.9292 - 776us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3751.4070 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.5356 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2558.1677 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2538.7471 - 852us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.3762 - 785us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2548.1628 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2611.4409 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.8526 - 807us/epoch - 13us/sample\n", - "Generation 5/300\n", - "Solving for Nash Equilibrium in Generation 5/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.3271 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2973.9827 - 1ms/epoch - 17us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2831.0952 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2599.1130 - 1ms/epoch - 20us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3393.7036 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3623.9565 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2407.3813 - 825us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2152.0059 - 950us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3753.8604 - 862us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2701.0237 - 809us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2663.1562 - 634us/epoch - 10us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.5669 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.3979 - 782us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2668.9722 - 741us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.5471 - 729us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2559.3850 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2417.8296 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2689.3347 - 952us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2556.4827 - 739us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2627.7871 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2680.4944 - 1000us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2534.9075 - 781us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3013.8792 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2686.7961 - 726us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.2373 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2741.5718 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3687.2322 - 1ms/epoch - 17us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2692.8044 - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2570.7017 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2668.6726 - 771us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2352.7107 - 707us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.1355 - 748us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2551.7908 - 650us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.9482 - 921us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3101.7896 - 811us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1240.7225 - 747us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3845.3218 - 765us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2505.4392 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.7354 - 707us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2611.4854 - 653us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.7239 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.8291 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2974.0823 - 726us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2557.2686 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.8152 - 725us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2769.8628 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2899.3103 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2601.7141 - 2ms/epoch - 35us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2549.5403 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3686.3740 - 725us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3288.9231 - 757us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2608.8760 - 651us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627.3325 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1544.9996 - 616us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2598.2625 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3814.0027 - 640us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1875.7864 - 639us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2968.2688 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2599.4573 - 682us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3624.2166 - 785us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.0298 - 656us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3623.9839 - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2663.8496 - 725us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2541.6870 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3394.8679 - 680us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3209.3430 - 663us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3750.5227 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1598.1118 - 719us/epoch - 12us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 6/300\n", - "Solving for Nash Equilibrium in Generation 6/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 873.1631 - 682us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3814.0691 - 594us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2611.0984 - 741us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241.5571 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3814.4709 - 737us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2960.1514 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266.5670 - 805us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2558.1980 - 782us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2799.6692 - 651us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 804.7482 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1081.1813 - 774us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3239.1421 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3340.4419 - 709us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2495.4844 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2628.6345 - 711us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3937.4373 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2645.6382 - 696us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1831.2330 - 716us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.8071 - 805us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.6072 - 754us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1933.6624 - 800us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.4199 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2569.1204 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3341.1675 - 708us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2596.4424 - 754us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2197.6038 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2935.6609 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3144.0564 - 747us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2540.8496 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2618.5693 - 662us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2599.2578 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2637.1506 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3950.1641 - 673us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.9080 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3623.8804 - 699us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2646.9121 - 736us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3750.9622 - 800us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2701.2698 - 589us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.5754 - 712us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3335.3445 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1554.2784 - 652us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2667.2104 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3098.7451 - 639us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2534.5850 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3547.8643 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2628.2312 - 756us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2568.9497 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1349.0000 - 697us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.4395 - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2571.7244 - 729us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2642.3420 - 703us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1983.4966 - 675us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3651.5151 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.1035 - 862us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2382.6101 - 888us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.3342 - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2546.1753 - 797us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2630.2043 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2770.1748 - 683us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2390.2712 - 761us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2564.1633 - 842us/epoch - 14us/sample\n", - "Generation 7/300\n", - "Solving for Nash Equilibrium in Generation 7/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2330.9854 - 801us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2447.7314 - 835us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.8774 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2690.1650 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464.9742 - 642us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 224.8822 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3340.3013 - 668us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2583.1504 - 640us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2573.2087 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2899.0706 - 779us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.3015 - 839us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1521.5287 - 750us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3190.8198 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2407.5771 - 786us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2899.4070 - 774us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2707.9539 - 721us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2623.7310 - 787us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.6262 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2217.6719 - 740us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2831.0923 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2974.2429 - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2627.1733 - 999us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3624.5864 - 905us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3934.3845 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2166.4714 - 880us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1335.4166 - 747us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3334.9263 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2294.7012 - 776us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2902.4902 - 973us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3505.4299 - 681us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2538.8435 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2628.1213 - 885us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2643.7183 - 1ms/epoch - 17us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3550.1155 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2564.8706 - 707us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3056.1365 - 832us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2647.8442 - 842us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3548.3811 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 197.3356 - 719us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2742.3413 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1764.0135 - 623us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.8379 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3238.8201 - 747us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1944.6090 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4019.4976 - 813us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3335.6467 - 646us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.6865 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.5576 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2667.3933 - 691us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.3376 - 806us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2584.2173 - 913us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.3530 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.0043 - 690us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2517.7573 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3238.7117 - 682us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4019.6760 - 730us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.5166 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2406.2969 - 924us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2611.2651 - 832us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2742.0759 - 644us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.3064 - 649us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2481.4207 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3239.1042 - 706us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.4631 - 775us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2554.8813 - 720us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.8892 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2557.7229 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2645.9622 - 663us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1894.7563 - 728us/epoch - 12us/sample\n", - "Generation 8/300\n", - "Solving for Nash Equilibrium in Generation 8/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2557.7224 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2558.4255 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2568.2012 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3291.0757 - 724us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.6045 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2407.3938 - 670us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2425.4810 - 875us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3935.6218 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2715.8862 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2693.3003 - 698us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.4460 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2571.5024 - 790us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2627.8860 - 667us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.9233 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3015.5000 - 672us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2382.6929 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2865.7007 - 700us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3451.2092 - 770us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2116.5129 - 690us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3652.7800 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3098.9043 - 742us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.6697 - 672us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2559.9355 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2613.5217 - 647us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2800.0139 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 686.0154 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3880.9688 - 673us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.5752 - 2ms/epoch - 27us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2152.4983 - 708us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2585.8931 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3056.3940 - 849us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1435.9576 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3751.4387 - 747us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3100.4524 - 764us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495.3725 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2588.2939 - 763us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3449.2422 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2268.6448 - 702us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2688.0400 - 4ms/epoch - 62us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3652.7776 - 2ms/epoch - 33us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.3052 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1750.1591 - 764us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2551.3694 - 677us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2597.7156 - 710us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.0513 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2548.2517 - 761us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2646.9653 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2602.5801 - 744us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1302.0691 - 838us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.8030 - 798us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2567.3064 - 786us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1996.6179 - 810us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.4915 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2569.0498 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1892.6125 - 752us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2546.6296 - 792us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2570.7324 - 728us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2549.4587 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2291.9851 - 690us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.8657 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2575.8752 - 687us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2899.3577 - 628us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600.4332 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3239.3694 - 792us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4019.2590 - 767us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2539.7454 - 701us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2248.8179 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2609.5894 - 734us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3565.0676 - 720us/epoch - 12us/sample\n", - "Generation 9/300\n", - "Solving for Nash Equilibrium in Generation 9/300\n", + "Generation 74/300\n", + "Solving for Nash Equilibrium in Generation 74/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2484.9783 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3441.3987 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483.4425 - 735us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.0427 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2646.9600 - 2ms/epoch - 27us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3396.5327 - 708us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.7141 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2569.6721 - 789us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.8481 - 754us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3099.5938 - 753us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2551.2009 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384.6263 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2613.7900 - 741us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1785.0149 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3144.0283 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3882.0667 - 734us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2541.3445 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2831.6733 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2620.0652 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3505.3093 - 674us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3555.6282 - 729us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1935.3884 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 578.2398 - 690us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.3845 - 901us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2577.7034 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3506.8167 - 821us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.1641 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3290.4500 - 898us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2568.9475 - 768us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494.0115 - 860us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3342.3518 - 762us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.3640 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2654.8059 - 764us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2692.2495 - 670us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3936.5767 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3846.3582 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3143.6895 - 681us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2556.9128 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1352.5739 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2658.4502 - 707us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2585.0125 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2622.2075 - 813us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3750.6426 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3336.6382 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3950.1885 - 797us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2670.2644 - 698us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2705.4016 - 793us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2533.0276 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2799.5098 - 760us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272.4679 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2570.1218 - 670us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2558.3250 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2584.2358 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2974.8228 - 755us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2573.5159 - 681us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2563.3579 - 759us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2402.8225 - 701us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2557.4146 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2769.8796 - 705us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2236.5598 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2061.1467 - 1ms/epoch - 17us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3290.6169 - 788us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2548.3904 - 756us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2612.8040 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332.0018 - 748us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3935.8250 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 809.9841 - 729us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2572.4675 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3749.2393 - 797us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2567.0481 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2865.7803 - 718us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2418.8406 - 923us/epoch - 15us/sample\n", - "Generation 10/300\n", - "Solving for Nash Equilibrium in Generation 10/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3091.2634 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2250.5928 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3015.3315 - 749us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3685.5823 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2289.8323 - 781us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2563.2693 - 743us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3222.4685 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2975.6487 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2692.2725 - 636us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2275.4944 - 972us/epoch - 16us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2704.3494 - 762us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2548.8733 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2835.1411 - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2802.7080 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2604.0549 - 782us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.9766 - 880us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.1309 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2136.2651 - 912us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3099.9287 - 730us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.2092 - 823us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2549.9114 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2632.6328 - 887us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2592.8538 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3882.5054 - 743us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3395.3423 - 973us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.5408 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2574.5266 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2691.1729 - 847us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3752.2356 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2867.8701 - 723us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2549.6348 - 774us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.4746 - 900us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.9561 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2706.6238 - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2252.9658 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2831.7671 - 706us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2937.0129 - 696us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.2688 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.8684 - 739us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2544.7656 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2590.5029 - 782us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.0093 - 3ms/epoch - 48us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2557.1379 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2182.2480 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2866.1899 - 689us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2974.5044 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3451.1609 - 795us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2254.1943 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2652.4626 - 823us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3559.4800 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 216.1114 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2604.3220 - 738us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2564.2913 - 633us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2564.9292 - 770us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2549.5557 - 712us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2648.7158 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2770.3831 - 678us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2489.3257 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2120.8647 - 819us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2602.1455 - 815us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2557.9834 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2455.2795 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2440.5554 - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3345.2327 - 757us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2539.4448 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2649.0034 - 893us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.8418 - 694us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3847.3250 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2567.6179 - 826us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2744.7791 - 859us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2558.4780 - 863us/epoch - 14us/sample\n", - "Generation 11/300\n", - "Solving for Nash Equilibrium in Generation 11/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3240.6140 - 720us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.4280 - 680us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.9861 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1679.5063 - 907us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2444.8733 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2835.8884 - 1ms/epoch - 19us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.6274 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3396.7485 - 829us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2559.2434 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3451.4116 - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2025.2251 - 911us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3759.5486 - 859us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3661.7195 - 806us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1484.2742 - 889us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2546.9082 - 890us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.5098 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2619.7168 - 1ms/epoch - 20us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2546.6646 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3950.7839 - 761us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2548.8999 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2549.4583 - 844us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2569.3240 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2713.6094 - 768us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.9358 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1650.4281 - 701us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2570.6753 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2646.5957 - 770us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3446.3376 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3061.6296 - 676us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2536.9106 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3754.5259 - 718us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2546.6230 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.0095 - 807us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3103.7021 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2617.1980 - 835us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.9785 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1893.8104 - 828us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2551.9741 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2584.3879 - 700us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2900.9890 - 730us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3508.5012 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2671.8394 - 824us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2493.6294 - 815us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2646.7175 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1091.2286 - 734us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3553.2649 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1492.7572 - 670us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.6826 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2448.6650 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2633.2771 - 714us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2692.4451 - 805us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2650.9172 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2007.6373 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 35ms/epoch - 569us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2485.5908 - 3ms/epoch - 50us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3149.8513 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3243.5667 - 856us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3882.2349 - 684us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1939.5271 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2627.0369 - 751us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712.6838 - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3058.8796 - 864us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2596.5881 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3291.3347 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3754.6304 - 646us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1611.8669 - 3ms/epoch - 42us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3938.4116 - 746us/epoch - 12us/sample\n", - "Generation 12/300\n", - "Solving for Nash Equilibrium in Generation 12/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3110.0518 - 762us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.2314 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2536.2119 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2309.7932 - 816us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2589.2446 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2904.5269 - 896us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2906.5012 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3510.4109 - 741us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2600.5271 - 775us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3549.1558 - 807us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2546.4956 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2215.8865 - 719us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2568.2415 - 771us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2511.2793 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560.3440 - 667us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2513.4187 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.1423 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2631.5220 - 1ms/epoch - 17us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2541.9663 - 637us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3018.8347 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1546.0367 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.2876 - 837us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2565.5261 - 840us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2565.7188 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2147.3340 - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 979.1265 - 816us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3107.7361 - 692us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3886.9509 - 806us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2662.9863 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3822.4771 - 836us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2680.9399 - 838us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1317.3486 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2663.8062 - 719us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2849.5266 - 668us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2771.0403 - 739us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2526.2332 - 834us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2002.7474 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1637.1458 - 836us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3245.0212 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3750.0825 - 761us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4027.8096 - 696us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2582.5994 - 858us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3096.9373 - 787us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2573.3218 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.8394 - 785us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2578.7373 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2611.6919 - 805us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2573.6375 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2977.1101 - 744us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2364.8501 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3626.2224 - 729us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2543.3997 - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2577.1467 - 755us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2865.8042 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1543.7000 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3693.6074 - 743us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2612.8767 - 728us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3229.1877 - 863us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.1018 - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3507.4094 - 729us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2542.7507 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1097.2589 - 679us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3558.4675 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4020.6665 - 856us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1581.8795 - 1ms/epoch - 19us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2562.0625 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3556.2744 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2572.4382 - 794us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2571.1184 - 909us/epoch - 15us/sample\n", - "Generation 13/300\n", - "Solving for Nash Equilibrium in Generation 13/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2561.4810 - 719us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2564.1267 - 846us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2559.2935 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2556.8513 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2588.8052 - 870us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2600.5525 - 837us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2556.0090 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2584.3716 - 824us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2572.2085 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2942.8533 - 753us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2557.2178 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1257.5629 - 783us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553.5886 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2162.6584 - 716us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2649.9946 - 702us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2116.7993 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2587.9897 - 721us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3009.3921 - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1795.5042 - 813us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1907.1770 - 770us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2583.2578 - 758us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2693.1228 - 768us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2563.4285 - 661us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2576.2378 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2704.0608 - 762us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2808.6018 - 767us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3344.6438 - 733us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2171.5427 - 709us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238.2138 - 717us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2629.0330 - 768us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 657.3696 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3019.1013 - 725us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2576.9172 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2773.9907 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3690.5771 - 783us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3871.0254 - 717us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2540.3496 - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2595.3958 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1571.9596 - 620us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2909.6399 - 801us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2357.9392 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3516.2856 - 876us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.8340 - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2587.5886 - 997us/epoch - 16us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2590.8840 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1905.7301 - 897us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3554.2498 - 894us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2567.0967 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2570.1128 - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.1753 - 785us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2607.6660 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2648.2991 - 707us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2658.8342 - 851us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2474.5251 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1022.4626 - 757us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3068.5308 - 769us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2574.1279 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297.1777 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1969.4250 - 851us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2135.5916 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2940.8130 - 758us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2573.0737 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3566.6318 - 722us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Generation 14/300\n", - "Solving for Nash Equilibrium in Generation 14/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2758.5212 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2582.7400 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2674.0618 - 804us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2474.8784 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3248.8308 - 908us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2583.7705 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2685.0847 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4022.7249 - 1ms/epoch - 17us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2617.3618 - 809us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2572.0386 - 743us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2596.8625 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2842.5786 - 825us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3413.2725 - 700us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2331.7727 - 746us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 935.2924 - 696us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2572.5146 - 813us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1832.6771 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2901.2043 - 730us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3507.5029 - 907us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2579.3762 - 1ms/epoch - 16us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2632.4004 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2653.4590 - 947us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2245.2671 - 705us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548.0293 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1689.2733 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2567.1599 - 880us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1794.2631 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3410.0146 - 783us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3630.7141 - 772us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2561.2678 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2674.8064 - 680us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1863.1108 - 864us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3513.6543 - 840us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2729.6145 - 842us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2649.3887 - 1ms/epoch - 18us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318.7264 - 749us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3556.0720 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2902.2451 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2686.5547 - 721us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1049.0190 - 2ms/epoch - 34us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3131.4829 - 736us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 872.7938 - 741us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2719.7195 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2629.7021 - 672us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3203.6118 - 686us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3962.7815 - 718us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2173.9233 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2097.5815 - 736us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 911.3597 - 776us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2712.2654 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2181.3240 - 829us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2590.6960 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2643.6741 - 659us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2592.6555 - 1ms/epoch - 19us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2995.5518 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370.1419 - 779us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2594.6008 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2859.0212 - 943us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1851.5032 - 3ms/epoch - 45us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2614.1809 - 775us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.4004 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2792.1194 - 752us/epoch - 12us/sample\n", - "Generation 15/300\n", - "Solving for Nash Equilibrium in Generation 15/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2332.8638 - 694us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2584.5715 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2912.2314 - 789us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3512.4033 - 716us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2586.3342 - 682us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2555.5212 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3553.2778 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3247.2322 - 770us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2598.6079 - 883us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2467.1882 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1755.3109 - 924us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2595.6997 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2835.6572 - 715us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2582.9126 - 669us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2887.5811 - 713us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378.1532 - 786us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2617.0103 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3088.3337 - 765us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2616.8652 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1569.3517 - 702us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3539.0557 - 987us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2706.7854 - 816us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3557.1824 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2314.4631 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3765.8655 - 792us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2610.5376 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2734.8669 - 754us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2580.6191 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1544.5144 - 1ms/epoch - 17us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.7212 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2726.3818 - 782us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2575.9424 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2579.0608 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2637.0706 - 803us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3219.8384 - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3957.8025 - 685us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 112us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2627.3213 - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2595.7507 - 821us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2718.5864 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1063.9207 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1397.0146 - 737us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3198.4392 - 745us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2651.9817 - 698us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3959.6238 - 696us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2708.3513 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2647.9485 - 1ms/epoch - 17us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3961.5835 - 845us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3789.0398 - 867us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2591.4463 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2745.6685 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2569.8530 - 703us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2597.6409 - 762us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2586.0000 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2628.7517 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1369.8842 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2041.9056 - 692us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2983.3411 - 813us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2721.3901 - 696us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2615.4910 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2601.0793 - 828us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2992.1897 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2563.2231 - 633us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2873.2949 - 895us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2630.3796 - 951us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1826.7679 - 737us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2088.6692 - 804us/epoch - 13us/sample\n", - "Generation 16/300\n", - "Solving for Nash Equilibrium in Generation 16/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2438.7104 - 781us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2578.2222 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2642.5857 - 785us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3457.9326 - 734us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2614.7107 - 919us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2649.0642 - 810us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 466.9705 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2752.4871 - 807us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2578.1765 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2695.0886 - 782us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3263.7664 - 781us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2636.6230 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3582.6519 - 684us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2576.2595 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2668.8723 - 834us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2641.2888 - 675us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2579.1128 - 750us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3057.5259 - 829us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2512.9521 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1825.1385 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1531.3002 - 597us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2855.4377 - 913us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3419.2234 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2694.5391 - 812us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2623.9419 - 1ms/epoch - 24us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2721.9656 - 743us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2746.2437 - 739us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1436.9806 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2209.6792 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309.2713 - 2ms/epoch - 27us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2736.6753 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1718.3245 - 856us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2583.6421 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2616.9792 - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3201.5654 - 720us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2667.7651 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2391.2080 - 851us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 77us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419.5385 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2444.2444 - 910us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2652.9246 - 956us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2631.3965 - 793us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2411.5776 - 886us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2626.7498 - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2746.9941 - 1ms/epoch - 20us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2567.5220 - 689us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2582.7949 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3202.1333 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1302.0908 - 1ms/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2472.9233 - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3910.8457 - 755us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3863.7969 - 734us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2610.4116 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2745.4368 - 736us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2597.8081 - 1ms/epoch - 17us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2592.3423 - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2876.0625 - 842us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2751.3694 - 641us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393.2174 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2512.4678 - 793us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2610.0259 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2665.8198 - 893us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2847.7322 - 841us/epoch - 14us/sample\n", - "Generation 17/300\n", - "Solving for Nash Equilibrium in Generation 17/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2728.5371 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3427.4302 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3464.3359 - 803us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2644.3220 - 748us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3676.1152 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3035.4202 - 692us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2638.3950 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2620.0085 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3727.9849 - 721us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3009.9922 - 820us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2708.3147 - 857us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3292.6606 - 766us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2227.5083 - 843us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2554.4246 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2645.1772 - 955us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3093.3484 - 767us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2638.1128 - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2741.8462 - 739us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2608.1465 - 891us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1541.8387 - 1ms/epoch - 19us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2610.5547 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1298.4954 - 1ms/epoch - 18us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2669.2046 - 1ms/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.6558 - 1ms/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2541.8984 - 782us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4061.1648 - 677us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2065.9280 - 877us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450.3197 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2011.7751 - 731us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 119us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2663.1138 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2730.0598 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3067.2849 - 1ms/epoch - 19us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3773.5376 - 882us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3243.4370 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3184.7854 - 1ms/epoch - 22us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3888.7188 - 782us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2666.6960 - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1603.8595 - 768us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2793.6265 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1545.1674 - 792us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3894.1628 - 1ms/epoch - 20us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2807.4771 - 843us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2626.7034 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752.6663 - 917us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3258.7285 - 4ms/epoch - 59us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3260.4558 - 756us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1814.4524 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4041.7827 - 809us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 811.6750 - 1ms/epoch - 18us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1768.4634 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2700.7102 - 1ms/epoch - 21us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3301.1501 - 723us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2370.0525 - 744us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2631.8323 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3158.0869 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2695.0830 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1396.6804 - 712us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2551.9509 - 851us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3971.9880 - 701us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2601.7703 - 757us/epoch - 12us/sample\n", - "Generation 18/300\n", - "Solving for Nash Equilibrium in Generation 18/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2612.9407 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2177.2278 - 889us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2584.2400 - 1ms/epoch - 18us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3889.2400 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2731.1072 - 772us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2601.1335 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2615.1968 - 694us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2602.4333 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2618.7312 - 748us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2682.0801 - 629us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2603.0483 - 685us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2791.3804 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2618.9949 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1498.9857 - 708us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2636.3425 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4122.7046 - 790us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2624.4067 - 735us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2684.4126 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3117.2104 - 785us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1505.5822 - 690us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2723.7515 - 760us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3944.7849 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2683.0728 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2147.9275 - 749us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2177.5989 - 700us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2686.1436 - 785us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1105.3148 - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3201.1101 - 682us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2632.4971 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 942.7969 - 688us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2980.1804 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2653.2104 - 787us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2552.7456 - 1ms/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2664.7173 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2668.4429 - 775us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3971.3513 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2723.8186 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2779.8914 - 760us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2394.9946 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2578.6216 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3220.9094 - 739us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2876.9419 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3983.4077 - 740us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2781.9519 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2814.9880 - 792us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2674.1707 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2350.1843 - 801us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2608.1296 - 889us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2685.7766 - 918us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2615.9426 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2689.3057 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2725.3735 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2923.1438 - 1ms/epoch - 22us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3137.7107 - 936us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2958.0498 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3789.2703 - 700us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2635.7820 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2281.7756 - 720us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2949.0552 - 744us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2622.1616 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2670.7546 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3634.7605 - 742us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2743.4771 - 804us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2672.7861 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2734.5012 - 728us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2955.2209 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1721.3130 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3687.8250 - 688us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3232.4001 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3651.8420 - 748us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3367.7605 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2719.8560 - 718us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2885.4541 - 3ms/epoch - 44us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2248.8679 - 776us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2265.5127 - 638us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2791.8721 - 652us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3441.7480 - 780us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Generation 19/300\n", - "Solving for Nash Equilibrium in Generation 19/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2802.3179 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2755.2874 - 731us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3323.0801 - 927us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3948.8335 - 764us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2594.7986 - 756us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2683.1567 - 755us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2659.5415 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2647.2703 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2683.8748 - 748us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3025.9988 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2725.2332 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3076.8186 - 878us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1932.1135 - 819us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2704.7751 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3161.9504 - 809us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2690.8975 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3861.0481 - 808us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2675.6675 - 896us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2839.7078 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2640.5923 - 759us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3081.6555 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4035.6807 - 785us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3423.7905 - 834us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3835.5371 - 781us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2752.8308 - 912us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2771.7156 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2685.6003 - 729us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2681.5686 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2175.7864 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2836.0449 - 756us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3480.3337 - 770us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2699.0991 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3203.8557 - 702us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3432.3044 - 689us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2762.0461 - 679us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2682.3779 - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2261.7363 - 1ms/epoch - 22us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3115.1865 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3530.8765 - 884us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3867.4297 - 845us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 564.1490 - 692us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3830.4272 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3236.6064 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1991.2621 - 898us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2695.3054 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2696.6240 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3643.4138 - 782us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2983.5930 - 827us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2828.4954 - 1ms/epoch - 18us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3616.5115 - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3032.6960 - 951us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2728.4683 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 17ms/epoch - 274us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2691.7156 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3606.7688 - 1ms/epoch - 17us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3004.9585 - 1ms/epoch - 23us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3593.0330 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2760.4460 - 1ms/epoch - 17us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2715.6279 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2971.9370 - 844us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2696.4617 - 7ms/epoch - 107us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2707.0293 - 910us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2805.6475 - 1ms/epoch - 20us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2294.8059 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2803.6782 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2902.7097 - 1ms/epoch - 22us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3467.3936 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2157.8801 - 783us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2720.1121 - 3ms/epoch - 45us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2778.7659 - 4ms/epoch - 62us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3226.1306 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3339.2458 - 977us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2563.9583 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2783.1472 - 2ms/epoch - 26us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4020.7827 - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2744.6221 - 750us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Generation 20/300\n", - "Solving for Nash Equilibrium in Generation 20/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2609.2329 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2717.5598 - 7ms/epoch - 115us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2668.9160 - 1ms/epoch - 19us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4111.5195 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2671.3328 - 860us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2696.8811 - 794us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2963.9287 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2638.4053 - 872us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3226.6531 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2850.8572 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2882.7346 - 767us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2768.7778 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2889.4304 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3420.0393 - 760us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3098.5857 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1996.1057 - 887us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2849.9255 - 703us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1547.3052 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3566.4963 - 777us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2764.9512 - 9ms/epoch - 149us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3259.2441 - 944us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2604.3425 - 875us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2734.3367 - 887us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4075.9106 - 954us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2643.4819 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2664.6135 - 811us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2861.5159 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2838.1702 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1941.5302 - 767us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2903.1089 - 840us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2924.7004 - 966us/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2696.7385 - 2ms/epoch - 29us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3291.8276 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1892.5530 - 714us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2321.0671 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3653.5481 - 924us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3635.6423 - 741us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2685.6084 - 750us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2743.9070 - 765us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3204.0017 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2998.7986 - 729us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1950.2697 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4021.6814 - 599us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3191.3865 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3322.0542 - 641us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3249.3740 - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3243.2825 - 1ms/epoch - 20us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3301.8113 - 680us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2777.3613 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2684.6672 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2781.7141 - 745us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2976.2373 - 761us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2830.1655 - 785us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2830.2317 - 1ms/epoch - 16us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2785.3174 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2659.8457 - 890us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3360.5042 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2727.1658 - 692us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2745.7236 - 639us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3083.6821 - 775us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3014.8738 - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3036.6311 - 912us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3303.3923 - 819us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1360.3999 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 943.8238 - 747us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2924.9490 - 705us/epoch - 11us/sample\n", - "Generation 21/300\n", - "Solving for Nash Equilibrium in Generation 21/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2957.9641 - 847us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3977.6008 - 4ms/epoch - 58us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3044.9971 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2858.8608 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2284.5120 - 792us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3137.7673 - 788us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3659.8057 - 794us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1611.1378 - 798us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3016.6272 - 861us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2982.6858 - 6ms/epoch - 102us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2356.1809 - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2828.8354 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2839.9971 - 1ms/epoch - 24us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3514.7803 - 930us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2741.6848 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2998.8843 - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3069.5081 - 773us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2669.8916 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3057.7610 - 869us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2825.1138 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2816.9360 - 688us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 919.0179 - 817us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2795.9783 - 1ms/epoch - 18us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2901.3826 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3070.5317 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3230.3616 - 653us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3003.6279 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3702.7798 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3795.5210 - 1ms/epoch - 24us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3189.9993 - 803us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3205.7693 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2764.9761 - 1ms/epoch - 17us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488.3637 - 2ms/epoch - 33us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2770.4734 - 776us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3920.6057 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2928.8013 - 1ms/epoch - 18us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2854.9448 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3218.7041 - 782us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2752.8254 - 995us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3404.8235 - 1ms/epoch - 18us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2578.0557 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2413.7583 - 967us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2755.6611 - 895us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2757.8250 - 1ms/epoch - 19us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2934.3054 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2913.8787 - 1ms/epoch - 18us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2819.8145 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2678.1821 - 1ms/epoch - 21us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3135.6604 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3550.1252 - 989us/epoch - 16us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2545.9666 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3862.5408 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1626.1544 - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3022.4893 - 1ms/epoch - 17us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3485.3430 - 790us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2958.9800 - 966us/epoch - 16us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3844.7480 - 1ms/epoch - 18us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2864.3196 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1352.3083 - 862us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1594.0837 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1889.1964 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3681.4910 - 999us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4176.0283 - 1ms/epoch - 18us/sample\n", - "Episode 50/50\n", - "Generation 22/300\n", - "Solving for Nash Equilibrium in Generation 22/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1979.3951 - 1ms/epoch - 17us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2828.1165 - 854us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1933.6790 - 1ms/epoch - 16us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2901.6030 - 889us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1696.6385 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2347.8301 - 871us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2787.8098 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2322.1006 - 1ms/epoch - 22us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2978.0325 - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2882.4934 - 2ms/epoch - 28us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3684.7119 - 928us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 956.1506 - 851us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3156.0764 - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2656.5640 - 5ms/epoch - 74us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2745.8489 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2739.9756 - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2757.5161 - 1ms/epoch - 17us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3202.6072 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2995.7366 - 942us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3042.8494 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3798.8696 - 1ms/epoch - 17us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4098.0059 - 1ms/epoch - 20us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2691.2866 - 845us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3176.8918 - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3091.5010 - 844us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2920.6897 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3827.7910 - 1ms/epoch - 17us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2821.2734 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1751.1854 - 890us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3927.0769 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2880.2727 - 856us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3079.2957 - 766us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3864.1519 - 1ms/epoch - 18us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4042.6482 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3055.9780 - 1ms/epoch - 18us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3143.9153 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2630.5254 - 6ms/epoch - 100us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2902.7236 - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2936.2224 - 1ms/epoch - 20us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3450.7837 - 1ms/epoch - 19us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2934.1848 - 1ms/epoch - 20us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3861.2566 - 1ms/epoch - 17us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3824.5876 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2846.1753 - 5ms/epoch - 88us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2965.9468 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1883.8596 - 1ms/epoch - 18us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2804.1626 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2936.4338 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3068.2405 - 966us/epoch - 16us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4013.3743 - 1ms/epoch - 17us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2721.7388 - 997us/epoch - 16us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3379.8394 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2628.9087 - 1ms/epoch - 20us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1954.1550 - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3079.8848 - 983us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3802.6902 - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3762.3318 - 1ms/epoch - 17us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3153.1804 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3139.1357 - 859us/epoch - 14us/sample\n", - "Generation 23/300\n", - "Solving for Nash Equilibrium in Generation 23/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1264.0729 - 962us/epoch - 16us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3454.5320 - 1ms/epoch - 22us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3437.4272 - 1ms/epoch - 20us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3038.9836 - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2846.2559 - 1ms/epoch - 17us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3453.3694 - 1ms/epoch - 19us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2467.0640 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2607.4468 - 1ms/epoch - 16us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4043.2620 - 1ms/epoch - 18us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2787.2322 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1689.0563 - 805us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2901.1948 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3195.0063 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2946.0032 - 5ms/epoch - 83us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3005.3813 - 1ms/epoch - 22us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2356.6389 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3196.5403 - 2ms/epoch - 30us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2866.2625 - 1ms/epoch - 17us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3640.8662 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3162.8779 - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2956.0835 - 1ms/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2857.9568 - 1ms/epoch - 18us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2794.1111 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3409.7087 - 1ms/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4425.8125 - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3481.0986 - 1ms/epoch - 19us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3987.7678 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2930.8137 - 2ms/epoch - 30us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2955.7273 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2682.1377 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3136.1206 - 768us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3112.2251 - 1ms/epoch - 19us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3226.7290 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3165.5967 - 1ms/epoch - 24us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 900.9764 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2639.1748 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2933.2346 - 840us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3562.5896 - 778us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2809.3860 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2877.0374 - 819us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2713.2827 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4010.0283 - 991us/epoch - 16us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2920.5332 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2858.7454 - 972us/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2873.6582 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3011.2310 - 981us/epoch - 16us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1497.9470 - 867us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3199.9597 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3725.5105 - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3023.2605 - 761us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4159.9238 - 894us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2767.3335 - 1ms/epoch - 20us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2931.5950 - 821us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3197.5686 - 3ms/epoch - 53us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1541.5658 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3244.4365 - 771us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4141.4980 - 787us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3633.3059 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3086.8286 - 755us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2941.4280 - 738us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3313.3718 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2301.9231 - 921us/epoch - 15us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 24/300\n", - "Solving for Nash Equilibrium in Generation 24/300\n", + "Generation 75/300\n", + "Solving for Nash Equilibrium in Generation 75/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2588.9683 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3438.2432 - 736us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2828.6975 - 1ms/epoch - 23us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3450.3560 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3013.1448 - 892us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2692.6985 - 3ms/epoch - 42us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4205.7505 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4649.7241 - 670us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2053.3848 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1206.9696 - 683us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3251.6067 - 4ms/epoch - 57us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4333.3677 - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3424.6584 - 817us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2610.4399 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3690.3860 - 658us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2823.9663 - 790us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3924.7573 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3388.1479 - 830us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3562.2251 - 843us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3186.2898 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2112.1240 - 765us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4644.4053 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3424.3669 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3720.3350 - 643us/epoch - 10us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2547.4656 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4029.8677 - 749us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3988.6008 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3699.5713 - 666us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3270.0396 - 3ms/epoch - 42us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3022.9241 - 675us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4172.7236 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3106.7573 - 690us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3574.3738 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2123.3574 - 889us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2776.0596 - 1ms/epoch - 21us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4807.0029 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2340.6489 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3437.1628 - 751us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3331.2949 - 775us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3756.4485 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2735.6804 - 799us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2534.3716 - 903us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2540.3772 - 797us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4091.9558 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2883.0950 - 2ms/epoch - 39us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3986.5283 - 884us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3841.1372 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2673.2710 - 882us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1741.3698 - 659us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2834.8840 - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3614.8870 - 1ms/epoch - 18us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3535.3259 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2793.3110 - 946us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3268.9636 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2576.8704 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4954.6812 - 746us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2895.3013 - 692us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4539.2026 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3051.5874 - 779us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3278.2966 - 694us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3402.7893 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2843.2261 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3741.2900 - 804us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Generation 25/300\n", - "Solving for Nash Equilibrium in Generation 25/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3091.8430 - 833us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4020.0288 - 4ms/epoch - 60us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1368.0266 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3407.3079 - 791us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3252.8367 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3746.7974 - 1ms/epoch - 24us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3122.5168 - 822us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3419.2327 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1777.9733 - 635us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3314.6501 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2870.2612 - 707us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3907.4231 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3990.4827 - 809us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4143.5298 - 962us/epoch - 16us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3703.6301 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2936.7944 - 825us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3206.5071 - 782us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4097.8467 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2881.7195 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3540.5002 - 666us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4264.4321 - 813us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3143.3052 - 729us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3383.6360 - 730us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2836.7932 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3546.6033 - 1ms/epoch - 17us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3958.5166 - 761us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3718.9460 - 915us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4108.4663 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3848.4993 - 770us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2885.1584 - 1ms/epoch - 22us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5817.1362 - 704us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3443.8247 - 982us/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3030.7427 - 1ms/epoch - 17us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2642.7622 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2768.7261 - 707us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3640.3105 - 665us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3815.2939 - 837us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3605.4177 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3111.9592 - 796us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1960.5759 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3436.4436 - 1ms/epoch - 21us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3561.4270 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4072.0215 - 709us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3396.5771 - 736us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4869.4844 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3168.8582 - 813us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2812.1411 - 827us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3858.6492 - 796us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2646.2292 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2934.6333 - 960us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5333.9590 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2945.2065 - 3ms/epoch - 56us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3607.9622 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3607.7917 - 749us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1592.8463 - 797us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1627.1234 - 693us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4834.6567 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4333.5298 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4082.2026 - 652us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3115.7678 - 783us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3379.7637 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4645.6597 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3629.9045 - 583us/epoch - 9us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3091.5979 - 773us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4368.0103 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2915.0312 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3531.0869 - 638us/epoch - 10us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 26/300\n", - "Solving for Nash Equilibrium in Generation 26/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5871.7500 - 738us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2926.8430 - 756us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2988.9871 - 709us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3192.7146 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4509.2622 - 2ms/epoch - 34us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2737.9148 - 789us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4957.2026 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3035.6946 - 842us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3272.1294 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3894.1555 - 752us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3264.9058 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3243.6978 - 595us/epoch - 10us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3292.8159 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4152.8940 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3665.1753 - 688us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3974.0730 - 803us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5289.7920 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3439.3052 - 713us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2875.1208 - 4ms/epoch - 64us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2869.5339 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3275.6162 - 687us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3582.0208 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3979.2737 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3823.9597 - 706us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2883.4626 - 1ms/epoch - 17us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3446.7410 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4987.0137 - 720us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3868.3813 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2566.0911 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2292.6782 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3170.6904 - 873us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3348.0798 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4677.3184 - 616us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3449.7263 - 839us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3449.7332 - 809us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4755.3398 - 733us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3107.3245 - 633us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3214.3394 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3489.2935 - 688us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3099.2014 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4483.9546 - 722us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3069.1228 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3245.0085 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4374.2388 - 645us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4215.1304 - 759us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3288.8977 - 721us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3449.2832 - 707us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5137.5957 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3157.0171 - 755us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2378.8303 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4713.7871 - 757us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4705.6992 - 752us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3170.2637 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4027.2488 - 2ms/epoch - 28us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3655.1833 - 715us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3474.6838 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3864.2214 - 700us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6765.2910 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2771.8540 - 792us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4122.5337 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3238.0171 - 784us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6829.8154 - 770us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4679.1860 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4218.6509 - 1ms/epoch - 17us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4946.9966 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3581.7520 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4555.1592 - 833us/epoch - 13us/sample\n", - "Generation 27/300\n", - "Solving for Nash Equilibrium in Generation 27/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3847.6155 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3203.9980 - 971us/epoch - 16us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3706.1768 - 713us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4307.0454 - 780us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6556.4697 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2550.2292 - 789us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3190.7998 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3038.7480 - 995us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3221.4177 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3307.9167 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3685.7126 - 2ms/epoch - 27us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5539.0371 - 679us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3575.0403 - 694us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4537.7690 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3459.6741 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3005.9836 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4311.4595 - 669us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3680.5454 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3645.8486 - 903us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2711.2061 - 741us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2889.9360 - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5284.3389 - 3ms/epoch - 43us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4822.3618 - 745us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3965.3755 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3308.5281 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2908.5552 - 793us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3248.3975 - 5ms/epoch - 76us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3621.7034 - 674us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3983.7883 - 761us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4284.8569 - 735us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3668.6326 - 770us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3621.3782 - 1ms/epoch - 16us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3704.9224 - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4092.0466 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3328.9460 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5434.2935 - 915us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2791.7000 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4654.2485 - 923us/epoch - 15us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3277.3545 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3844.3042 - 895us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3360.6741 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3914.0654 - 819us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5519.3687 - 738us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2801.5142 - 828us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 115us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4628.5425 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3966.5979 - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4766.5845 - 892us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6698.7646 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4280.8877 - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3240.5574 - 706us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3349.2000 - 3ms/epoch - 47us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 33ms/epoch - 528us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5754.2480 - 11ms/epoch - 183us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4146.9902 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3551.3105 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4292.6553 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4637.1719 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2609.0691 - 2ms/epoch - 35us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4150.5439 - 727us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4104.6396 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3894.0378 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4169.9272 - 627us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2875.0093 - 866us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3714.5828 - 734us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3610.5525 - 778us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4824.8232 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9684.6357 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4012.5403 - 608us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3171.5994 - 722us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4702.5112 - 763us/epoch - 12us/sample\n", - "Generation 28/300\n", - "Solving for Nash Equilibrium in Generation 28/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3800.9150 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4502.2549 - 1ms/epoch - 22us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3505.5239 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3611.1531 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4532.8608 - 992us/epoch - 16us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4414.6699 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6472.6494 - 709us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4083.9199 - 2ms/epoch - 26us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8918.9395 - 1ms/epoch - 19us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4722.0986 - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3275.1616 - 886us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4892.1362 - 946us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3882.2056 - 988us/epoch - 16us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2776.3701 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3420.7190 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4598.3345 - 822us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4588.3354 - 1ms/epoch - 22us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6083.4775 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4453.5259 - 984us/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5147.6860 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3629.0410 - 1ms/epoch - 19us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2862.7224 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7892.9072 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4253.3398 - 780us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3759.4038 - 838us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4429.4351 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3302.6025 - 1ms/epoch - 23us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6255.6880 - 871us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3101.4622 - 693us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4885.9839 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3974.5625 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2667.4949 - 1ms/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3976.1111 - 728us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9099.0869 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4204.6650 - 735us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3884.7988 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2718.8176 - 796us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5045.5679 - 744us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3004.0525 - 908us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5166.0586 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4023.3057 - 1ms/epoch - 18us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4220.5610 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3162.0996 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3289.5857 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3360.0569 - 2ms/epoch - 28us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4573.0645 - 1ms/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3885.6008 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4474.1382 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4479.7358 - 830us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3985.0769 - 734us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7044.6436 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4239.0400 - 734us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7892.2612 - 794us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4780.9248 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4354.4849 - 2ms/epoch - 26us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5020.1206 - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7961.6919 - 863us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4059.0293 - 748us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4173.6313 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7573.2344 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6606.7866 - 679us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3255.5317 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4275.7246 - 750us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6710.9180 - 768us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3470.8171 - 775us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3828.3301 - 743us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11227.7803 - 721us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5912.1948 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3423.3220 - 995us/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4518.1782 - 823us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4832.9287 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4456.6860 - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5192.9497 - 798us/epoch - 13us/sample\n", - "Generation 29/300\n", - "Solving for Nash Equilibrium in Generation 29/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3243.0022 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4897.6357 - 734us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5000.1094 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3022.2756 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5533.5864 - 855us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5749.1738 - 757us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4007.3574 - 902us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4192.2646 - 765us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4325.6616 - 1ms/epoch - 21us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11670.2441 - 782us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2938.0972 - 719us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3593.9268 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8726.6113 - 840us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6369.5439 - 765us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3001.2126 - 691us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4650.5225 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4616.0713 - 738us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8973.4980 - 973us/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2788.8357 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4978.0967 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4514.0918 - 791us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5072.0366 - 2ms/epoch - 28us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2860.5625 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3726.8374 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 97us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3485.1313 - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6759.2778 - 736us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4308.5356 - 820us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3334.9338 - 983us/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6764.0288 - 1ms/epoch - 20us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5621.1919 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3876.2393 - 4ms/epoch - 64us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7623.3916 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4118.6221 - 931us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3629.4465 - 754us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4332.1904 - 780us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7416.1221 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3018.9788 - 732us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3329.2036 - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3685.1187 - 992us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4980.5132 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11175.1162 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3855.3357 - 788us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4342.0947 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5107.4912 - 2ms/epoch - 36us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3848.5022 - 2ms/epoch - 32us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7849.6646 - 1ms/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4264.5669 - 956us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12283.7578 - 1ms/epoch - 18us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3202.2092 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3378.3120 - 830us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5522.6235 - 741us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3115.8840 - 893us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2353.5212 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7039.4810 - 946us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4048.6331 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4068.7842 - 799us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8918.2461 - 844us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4525.4497 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5138.9736 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5133.0161 - 938us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5039.4375 - 802us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5172.2808 - 1ms/epoch - 17us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8541.0439 - 1ms/epoch - 17us/sample\n", - "Generation 30/300\n", - "Solving for Nash Equilibrium in Generation 30/300\n", + "Generation 76/300\n", + "Solving for Nash Equilibrium in Generation 76/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5032.7031 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7095.5649 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3300.7795 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5977.8599 - 720us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5647.2573 - 754us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4425.8979 - 764us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8466.3301 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3942.9973 - 879us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3395.3538 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3655.9736 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5845.9751 - 1ms/epoch - 16us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11968.2021 - 2ms/epoch - 29us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6652.6367 - 859us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7068.8511 - 802us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5593.5029 - 739us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3218.1179 - 2ms/epoch - 25us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10960.9521 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4692.1201 - 2ms/epoch - 35us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3995.3149 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3483.5952 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5350.8931 - 850us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4924.6748 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4654.0156 - 685us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4528.8291 - 701us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4273.6445 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5170.4346 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3820.1785 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8054.3716 - 663us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3395.5464 - 814us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7438.1772 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 76us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3431.0432 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4207.1294 - 795us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6456.3438 - 668us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3855.6340 - 802us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6358.9946 - 663us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7563.9902 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4376.6030 - 814us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7299.8794 - 935us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6789.0415 - 1ms/epoch - 23us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4697.6313 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4041.0088 - 709us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6642.0479 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2791.9639 - 619us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3796.0798 - 668us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2807.6558 - 812us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6480.5664 - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5166.2290 - 735us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5456.7432 - 740us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5111.8110 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4180.6997 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3575.5464 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6710.1494 - 697us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7348.0161 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2924.0232 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4287.3418 - 869us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3989.6152 - 620us/epoch - 10us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5042.1348 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3818.1956 - 740us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4793.9668 - 720us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4774.6055 - 776us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10769.2148 - 841us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4496.8745 - 818us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17293.2031 - 3ms/epoch - 49us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4239.0210 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4790.1890 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7496.1245 - 748us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3688.8823 - 733us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4653.4268 - 763us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10091.3623 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4342.0898 - 721us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5273.3110 - 734us/epoch - 12us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 31/300\n", - "Solving for Nash Equilibrium in Generation 31/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16409.7422 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4327.2139 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2873.7761 - 734us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6250.8457 - 658us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10012.6660 - 681us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4898.6182 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6374.3706 - 725us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4599.6968 - 717us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3730.6382 - 713us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4627.2549 - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8042.9351 - 896us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4695.7607 - 700us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5180.0928 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4769.4653 - 733us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13034.6768 - 755us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3126.7444 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8311.3770 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3123.7322 - 4ms/epoch - 70us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4007.0276 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4550.6299 - 733us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4649.1055 - 720us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5318.2168 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4094.9966 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6368.8164 - 1ms/epoch - 17us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4507.4717 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5256.8560 - 773us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5892.1001 - 586us/epoch - 9us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4685.9160 - 620us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10613.3613 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4446.9229 - 578us/epoch - 9us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5803.9849 - 802us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4368.5415 - 717us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5386.7119 - 842us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7545.0562 - 743us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3931.8916 - 778us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6796.2178 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3821.7764 - 925us/epoch - 15us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5570.3145 - 676us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6809.6963 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3802.9314 - 669us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4376.6792 - 925us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15101.4121 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2864.8257 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4553.9170 - 800us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4932.7490 - 856us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4884.1631 - 745us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4438.6387 - 836us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4597.7417 - 1ms/epoch - 17us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5466.5698 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3337.2466 - 837us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5760.6543 - 609us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3569.7737 - 813us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4489.3271 - 715us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5175.0757 - 870us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3975.2507 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6722.9224 - 717us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3146.0737 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5737.4165 - 766us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4066.4197 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6968.6050 - 1ms/epoch - 17us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8351.7305 - 807us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4471.6592 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5497.9805 - 791us/epoch - 13us/sample\n", - "Generation 32/300\n", - "Solving for Nash Equilibrium in Generation 32/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5357.4546 - 706us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8234.8330 - 744us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2807.5642 - 636us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7286.3096 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5268.6279 - 694us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7172.5762 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5033.4282 - 705us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4818.4331 - 798us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13583.5527 - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5079.8018 - 762us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6745.4785 - 804us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5864.9771 - 713us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3679.9702 - 755us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6219.8813 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5126.5063 - 3ms/epoch - 42us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35513.1562 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5727.2598 - 794us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4294.5605 - 720us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16132.3330 - 796us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7228.6616 - 697us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6751.2007 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19476.8320 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4815.7134 - 823us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10848.4043 - 895us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5709.9365 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3377.8394 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9866.0254 - 789us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6018.2886 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17084.8945 - 746us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4120.5723 - 691us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4765.1753 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3986.5881 - 2ms/epoch - 29us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11899.6211 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4482.0664 - 732us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5377.2520 - 729us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3481.7520 - 783us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6843.4312 - 679us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7038.8164 - 744us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53302.5117 - 757us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4178.6270 - 672us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5255.9829 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3490.0693 - 929us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10253.8760 - 696us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5707.5811 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6891.9517 - 764us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6975.5332 - 779us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3541.1221 - 776us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16643.2852 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3551.3706 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5745.8208 - 787us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4517.1963 - 792us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9926.7012 - 735us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4705.1089 - 774us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6732.5059 - 765us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11353.8789 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3473.1533 - 675us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10327.8955 - 716us/epoch - 12us/sample\n", - "Generation 33/300\n", - "Solving for Nash Equilibrium in Generation 33/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5729.1602 - 687us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8128.4429 - 2ms/epoch - 29us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8663.7939 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10620.6016 - 674us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3035.1582 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8669.2754 - 770us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4348.3525 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5702.3662 - 3ms/epoch - 45us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5835.1562 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9296.5098 - 754us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4735.0537 - 703us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4206.0317 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6694.4268 - 710us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3759.5811 - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7702.6270 - 718us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5878.8389 - 703us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4059.1604 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5306.4258 - 635us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10199.3584 - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37473.5156 - 5ms/epoch - 88us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8511.6885 - 758us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5413.1226 - 810us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20837.4727 - 868us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5510.7231 - 991us/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9170.4658 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9549.9238 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5599.5117 - 6ms/epoch - 90us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8678.6924 - 816us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6341.1152 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9273.4277 - 748us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4888.4536 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7180.7896 - 1ms/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12153.3506 - 636us/epoch - 10us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6531.0479 - 721us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5266.0801 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3707.0798 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29366.6328 - 646us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5059.5483 - 757us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5300.0386 - 911us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3057.3301 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4877.0522 - 996us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11996.2363 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10284.6367 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5272.4678 - 835us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8503.6543 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7757.1943 - 661us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5902.7236 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6482.3687 - 661us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26716.1250 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6688.4395 - 724us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6273.3105 - 718us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9964.7109 - 919us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3815.1975 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6950.9551 - 878us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9300.2109 - 954us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10313.4473 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6999.4702 - 837us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5128.5791 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3899.6047 - 789us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7050.8760 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8260.1855 - 598us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7558.8193 - 778us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4442.8721 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9837.7363 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5665.5010 - 840us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7788.7651 - 764us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19850.6309 - 887us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6671.1011 - 647us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18162.9824 - 846us/epoch - 14us/sample\n", - "Generation 34/300\n", - "Solving for Nash Equilibrium in Generation 34/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4764.8730 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9753.6182 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7516.7129 - 2ms/epoch - 39us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3005.4182 - 741us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9089.8545 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12353.4189 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3626.1357 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6470.8735 - 730us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4101.8726 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5068.7881 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10135.9414 - 756us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9615.1592 - 966us/epoch - 16us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8388.5273 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13121.3066 - 786us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4135.7480 - 2ms/epoch - 31us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8821.1250 - 849us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10973.3984 - 870us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8273.4209 - 695us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3654.9858 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3809.2830 - 1ms/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3003.2168 - 1ms/epoch - 20us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7158.1191 - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7850.1055 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5147.8091 - 715us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5347.6650 - 750us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15361.1924 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8105.8677 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4424.2759 - 728us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6417.1123 - 792us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5920.8062 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5066.4971 - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13665.5234 - 1ms/epoch - 17us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5226.2720 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8980.5146 - 707us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5245.2559 - 796us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5674.1748 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20492.2754 - 644us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5939.6968 - 744us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4192.8975 - 677us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5444.9229 - 688us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5254.9951 - 770us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18613.9551 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6367.3076 - 897us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8166.7144 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4142.2612 - 2ms/epoch - 29us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6447.9414 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10173.4746 - 632us/epoch - 10us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10541.3525 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6959.6479 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11273.4912 - 655us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4777.2729 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6824.1665 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5305.3599 - 643us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8450.4785 - 989us/epoch - 16us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8295.0234 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9069.5352 - 3ms/epoch - 44us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2810.8093 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7099.7896 - 678us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6127.4917 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7816.5615 - 808us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20687.2656 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9476.9023 - 755us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4295.1738 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5255.5371 - 2ms/epoch - 25us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12196.4512 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7188.5723 - 804us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7398.6938 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5967.6226 - 708us/epoch - 11us/sample\n", - "Generation 35/300\n", - "Solving for Nash Equilibrium in Generation 35/300\n", + "Generation 77/300\n", + "Solving for Nash Equilibrium in Generation 77/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7421.5420 - 796us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9250.9697 - 721us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8833.8711 - 838us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30243.8770 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6861.5259 - 880us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5436.0195 - 699us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4872.4434 - 691us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6036.0054 - 1ms/epoch - 17us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4277.3481 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6047.2495 - 777us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8165.4336 - 819us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5852.4268 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3674.4517 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11142.2988 - 4ms/epoch - 71us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4547.9121 - 789us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14030.1738 - 859us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11600.4160 - 917us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12679.9316 - 765us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8428.4775 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6011.7964 - 760us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46970.3203 - 765us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11476.9805 - 942us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4250.7417 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12686.2617 - 787us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5526.2280 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4801.6353 - 739us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10419.7051 - 872us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4324.9395 - 754us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13142.9336 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4502.0806 - 681us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3975.8308 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10136.7998 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5843.8682 - 833us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12591.5811 - 814us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3700.0625 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6746.5454 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 113us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8017.5083 - 2ms/epoch - 29us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12659.1143 - 729us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5815.9497 - 773us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6609.2500 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10471.3672 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12268.9814 - 695us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4987.6660 - 751us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5457.1777 - 853us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4831.4648 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6913.6470 - 708us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4648.2397 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27494.9121 - 793us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13266.3730 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9760.2734 - 764us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4388.3320 - 720us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6097.8042 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4633.3115 - 756us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6805.7104 - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5240.7568 - 776us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3818.4573 - 775us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19298.1133 - 799us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9912.4033 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5567.1362 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13305.5254 - 806us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13194.7080 - 752us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8321.2207 - 886us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5858.3696 - 710us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Generation 36/300\n", - "Solving for Nash Equilibrium in Generation 36/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8034.1636 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9392.5762 - 805us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4940.0059 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26791.7383 - 799us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16056.5068 - 743us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12519.8877 - 719us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23626.3340 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7067.2310 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3408.4419 - 1ms/epoch - 20us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7870.6914 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6222.1738 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5771.9277 - 810us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8272.5176 - 732us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5736.3149 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5566.7896 - 725us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10406.7920 - 978us/epoch - 16us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6146.0293 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3837.9460 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7568.3906 - 997us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7851.9805 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5790.3550 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10668.2617 - 818us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5236.6421 - 727us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7540.5713 - 874us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8713.4971 - 741us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5897.4521 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5191.8086 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32980.5547 - 633us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5339.8145 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6647.0669 - 942us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6283.2998 - 1ms/epoch - 21us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17978.3770 - 773us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11257.7236 - 736us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11715.6777 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8222.5127 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17875.1641 - 685us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8861.5518 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5676.1372 - 841us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5237.1499 - 891us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22896.2949 - 688us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13914.9453 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6846.8872 - 794us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8805.3633 - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6329.2915 - 764us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14082.8096 - 679us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10445.7578 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5381.1626 - 2ms/epoch - 25us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7126.3530 - 695us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6289.2407 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13684.1553 - 663us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49253.5820 - 687us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3779.0847 - 754us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11174.7002 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6078.6099 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17492.1992 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9834.6240 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5062.6777 - 742us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7231.6973 - 755us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3872.1125 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10118.3320 - 692us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5572.4131 - 763us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33292.2617 - 869us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12336.6211 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8006.2495 - 765us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12382.3545 - 670us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4081.1099 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14721.8018 - 847us/epoch - 14us/sample\n", - "Generation 37/300\n", - "Solving for Nash Equilibrium in Generation 37/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15319.0186 - 721us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31583.5859 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9256.8516 - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6126.0161 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4871.1631 - 736us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7440.1465 - 696us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13377.0234 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6502.0947 - 820us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7707.8579 - 887us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4182.3809 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10827.9141 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6565.2085 - 752us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10949.2744 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5022.9072 - 755us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13131.0723 - 974us/epoch - 16us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11763.1328 - 809us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48923.1836 - 762us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9285.8408 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11118.5918 - 775us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7612.0298 - 680us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40514.7070 - 817us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23399.5742 - 878us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14343.4502 - 885us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10513.9238 - 820us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19103.9121 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8984.4141 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7258.7725 - 711us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13849.0215 - 745us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6467.6948 - 696us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16765.6035 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14430.7480 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16138.5459 - 694us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7565.5527 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10549.6055 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5597.9780 - 665us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11908.0146 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6223.5596 - 841us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20677.8008 - 798us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6096.9575 - 715us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9424.9883 - 2ms/epoch - 37us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6912.4434 - 774us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15694.6631 - 718us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10078.6455 - 856us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5395.6914 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10214.4727 - 866us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7448.9097 - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3009.3101 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8007.2026 - 726us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52117.6797 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6114.0708 - 730us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9893.5723 - 775us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4821.3745 - 795us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6596.4893 - 896us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12347.8516 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16211.4678 - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7309.2471 - 687us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6936.8027 - 833us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6886.0913 - 675us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7716.6157 - 687us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11435.8389 - 809us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14018.1641 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9852.1553 - 787us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5999.7686 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5671.8735 - 704us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89517.5547 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7013.3682 - 924us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Generation 38/300\n", - "Solving for Nash Equilibrium in Generation 38/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31178.6973 - 862us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9540.3438 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8043.1353 - 734us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15409.7207 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13642.8730 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7951.3618 - 681us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11278.4678 - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7106.8413 - 817us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10856.2900 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10671.3408 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34938.6641 - 976us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12011.0303 - 719us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40187.1992 - 881us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14770.7461 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13700.9434 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 86us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10729.2422 - 1ms/epoch - 20us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7987.0889 - 721us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6146.1401 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9233.6387 - 2ms/epoch - 36us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12886.3193 - 847us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9686.3340 - 610us/epoch - 10us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18352.1582 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4562.0088 - 639us/epoch - 10us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17181.8359 - 853us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10702.7422 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7843.7251 - 671us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8789.8828 - 675us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18112.6289 - 773us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4183.8032 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28453.5078 - 815us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13819.4502 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7370.2202 - 896us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6277.3457 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8646.9150 - 858us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8984.7988 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15169.9941 - 2ms/epoch - 25us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 62718.2227 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12683.2363 - 786us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9466.1992 - 783us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83596.9844 - 708us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9450.8623 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23193.6758 - 764us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17102.1973 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15354.7480 - 778us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9375.1895 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28829.7734 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18032.6914 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3126.8174 - 752us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14193.9609 - 766us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27585.9570 - 943us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12041.1475 - 751us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8685.8848 - 706us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47975.9297 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9188.3350 - 950us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8185.0854 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15180.0137 - 732us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42803.5977 - 778us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6179.9375 - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30236.6836 - 780us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8431.6475 - 650us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22033.1816 - 872us/epoch - 14us/sample\n", - "Generation 39/300\n", - "Solving for Nash Equilibrium in Generation 39/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7987.3247 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9977.2100 - 845us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18372.6016 - 788us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5156.4253 - 811us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14738.4863 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17518.3906 - 767us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28433.2832 - 720us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11092.3213 - 1ms/epoch - 19us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6809.7129 - 945us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22151.0488 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7456.7593 - 827us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7313.6670 - 795us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19377.4395 - 629us/epoch - 10us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9446.2100 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15308.0186 - 767us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18456.7930 - 757us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9050.5410 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7319.1162 - 1ms/epoch - 18us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24145.7715 - 774us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18896.1211 - 786us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7392.0215 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6584.9790 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16892.9238 - 927us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15886.1426 - 3ms/epoch - 56us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4703.2480 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8697.7715 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18037.8887 - 692us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9515.2686 - 790us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10323.6357 - 723us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23910.2480 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5839.8662 - 715us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7505.2905 - 845us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10747.3662 - 719us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59960.6836 - 708us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15384.6455 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7025.9429 - 596us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9137.1855 - 781us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18502.7324 - 691us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23479.3379 - 765us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7084.8755 - 824us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35813.2695 - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16178.6621 - 1ms/epoch - 20us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7601.9707 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27976.1328 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9863.6270 - 714us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13471.8389 - 745us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24984.2734 - 865us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22602.5156 - 1ms/epoch - 22us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11605.8389 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15814.3799 - 886us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10481.0615 - 684us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9551.5303 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11026.1035 - 723us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8556.2734 - 2ms/epoch - 35us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10525.3604 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8017.4819 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9042.8955 - 664us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50969.0859 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15684.8770 - 724us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7105.0669 - 657us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9574.5303 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11834.4688 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9210.0029 - 858us/epoch - 14us/sample\n", - "Generation 40/300\n", - "Solving for Nash Equilibrium in Generation 40/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17503.0605 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7491.8545 - 976us/epoch - 16us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24935.9609 - 687us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8230.6221 - 748us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16553.7402 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20547.2285 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9431.3662 - 647us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36335.4297 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19696.1504 - 862us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17017.7910 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9498.7324 - 725us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11055.6230 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13157.8564 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9993.2061 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10681.3643 - 698us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26522.1152 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17440.4980 - 733us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7569.4727 - 722us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18674.9824 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19276.0391 - 2ms/epoch - 26us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6674.6538 - 784us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19811.6426 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7454.3965 - 792us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5469.3770 - 2ms/epoch - 28us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8003.5063 - 768us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3456.7595 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41817.1211 - 742us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8161.9004 - 746us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6562.2544 - 828us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38719.9805 - 747us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7638.7910 - 4ms/epoch - 61us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20342.6055 - 669us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17529.3516 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13126.7354 - 713us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9299.5879 - 812us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27266.7617 - 781us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7102.1909 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30074.6094 - 712us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11331.5566 - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6268.3477 - 806us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17289.2500 - 710us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13946.9072 - 710us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13979.7529 - 669us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4910.6519 - 667us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9467.8398 - 736us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47554.8359 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11456.5986 - 763us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9871.2734 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9433.5684 - 1ms/epoch - 17us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46110.5312 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6656.2256 - 740us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10907.0479 - 753us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4755.2202 - 971us/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21559.3555 - 711us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17001.0371 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10514.4326 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7414.1406 - 813us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13290.6084 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9572.0283 - 6ms/epoch - 95us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53431.2344 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5413.4912 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22593.8418 - 887us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7048.4438 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28666.0410 - 1ms/epoch - 17us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12470.4766 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 99us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7781.1782 - 901us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21257.2715 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11343.9199 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66264.6328 - 969us/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24776.8789 - 825us/epoch - 13us/sample\n", - "Generation 41/300\n", - "Solving for Nash Equilibrium in Generation 41/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12769.9033 - 799us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6775.1582 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9638.5908 - 723us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9159.9980 - 703us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23423.5020 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6830.4131 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12175.2178 - 823us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6374.6152 - 734us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18455.7344 - 1ms/epoch - 18us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20033.3750 - 617us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18249.1895 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 96328.9844 - 729us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8942.9375 - 769us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30101.4473 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16503.2715 - 900us/epoch - 15us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19553.7539 - 837us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23581.7891 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15529.1523 - 834us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6493.6904 - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22274.5000 - 754us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37657.8047 - 766us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30231.3457 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5358.3594 - 2ms/epoch - 32us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54204.0742 - 899us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10357.3857 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31044.3496 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9647.7412 - 776us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23089.7949 - 861us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24053.1621 - 861us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13226.3340 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10184.3896 - 3ms/epoch - 47us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24946.2637 - 891us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11252.6064 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22573.0078 - 970us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8067.2754 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65271.0664 - 729us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12132.1191 - 2ms/epoch - 27us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33006.7148 - 767us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16098.7227 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11157.2773 - 858us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12930.7725 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11510.9209 - 859us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20253.7676 - 1ms/epoch - 21us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51147.9258 - 1ms/epoch - 19us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4661.2227 - 827us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 75554.4688 - 825us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42105.2773 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33098.5352 - 930us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33048.1836 - 678us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23751.4883 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34147.8516 - 784us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7339.9731 - 4ms/epoch - 64us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12251.4912 - 706us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27049.9941 - 743us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21487.8984 - 772us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10831.8965 - 838us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14374.4287 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22979.6758 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9472.3945 - 828us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74888.9297 - 1ms/epoch - 24us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12169.8047 - 719us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43867.4727 - 762us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14196.1260 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13783.5205 - 727us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44535.4492 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40520.5586 - 777us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12396.1045 - 810us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 70811.1797 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12672.0488 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27721.4980 - 731us/epoch - 12us/sample\n", - "Generation 42/300\n", - "Solving for Nash Equilibrium in Generation 42/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22497.4766 - 775us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6891.0151 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65587.2578 - 650us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16571.4219 - 672us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7016.4111 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36100.0938 - 712us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11924.0645 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40184.1797 - 732us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23500.5156 - 733us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10127.2480 - 727us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20070.5879 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17752.6387 - 782us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88226.0625 - 660us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25302.6816 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8081.6929 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37791.1211 - 759us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11013.8213 - 666us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40333.3047 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10753.9121 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53482.5938 - 689us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14335.5820 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8939.4277 - 693us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39089.3320 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13487.1963 - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5645.1060 - 1ms/epoch - 22us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32228.9590 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11398.5801 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6083.7065 - 764us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45665.5547 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6808.0923 - 745us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8570.3047 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24079.3281 - 772us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35491.6328 - 689us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8584.3545 - 971us/epoch - 16us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29596.0156 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11825.8750 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15843.7207 - 724us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56053.5938 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37319.7500 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11374.9863 - 712us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47604.0391 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25214.8594 - 701us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27702.9707 - 2ms/epoch - 30us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17812.0625 - 740us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5251.9399 - 668us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19446.8125 - 652us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47844.7617 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21917.7363 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16643.5820 - 684us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28283.4336 - 777us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10081.5264 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15247.2334 - 703us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7763.7964 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26660.7324 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 84318.4844 - 696us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28162.8535 - 619us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15118.0869 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17739.3496 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126650.4062 - 672us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48244.7461 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25903.6035 - 833us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3969.0500 - 711us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28354.1328 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46388.1172 - 841us/epoch - 14us/sample\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 43/300\n", - "Solving for Nash Equilibrium in Generation 43/300\n", + "Generation 78/300\n", + "Solving for Nash Equilibrium in Generation 78/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11147.3428 - 662us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55722.5781 - 743us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13615.4238 - 806us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29609.5215 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14964.6250 - 2ms/epoch - 34us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13877.2559 - 697us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30347.1328 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23186.8320 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15555.6436 - 681us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51167.6680 - 2ms/epoch - 32us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36612.1836 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20927.7090 - 697us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41738.5195 - 880us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47018.5508 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54506.8438 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30796.6855 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17349.1660 - 3ms/epoch - 52us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 18ms/epoch - 285us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30513.7793 - 4ms/epoch - 58us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5476.1445 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11526.7617 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47990.4570 - 939us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7623.9341 - 848us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18150.0508 - 742us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19092.6895 - 733us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14185.6895 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21823.5195 - 2ms/epoch - 32us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61571.3555 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11428.5020 - 709us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10844.2314 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19509.3242 - 725us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15520.3086 - 982us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36317.2578 - 693us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29199.6895 - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50561.0195 - 885us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4697.4131 - 761us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13236.9600 - 665us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 190727.5625 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8892.9854 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32043.9434 - 837us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11585.9043 - 761us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112752.3594 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35503.1289 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19912.6895 - 795us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20006.2012 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32215.3594 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25436.3457 - 710us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14944.4912 - 722us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25243.5840 - 2ms/epoch - 27us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13462.1562 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50117.3750 - 779us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15878.3311 - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35950.0859 - 589us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6396.3564 - 755us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12011.7490 - 648us/epoch - 10us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12724.9609 - 908us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 225398.4219 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26264.1738 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6937.5464 - 1ms/epoch - 20us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", "Episode 49/50\n", "Episode 50/50\n", - "Generation 44/300\n", - "Solving for Nash Equilibrium in Generation 44/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3444.0354 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19736.2363 - 755us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9709.5352 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19080.4922 - 756us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 93463.6953 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7643.8931 - 798us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35681.7266 - 733us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52683.3633 - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21693.1289 - 984us/epoch - 16us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21876.4375 - 715us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 82342.1641 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11678.6650 - 2ms/epoch - 27us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8942.0977 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22081.1309 - 750us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20991.0332 - 796us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18297.2969 - 774us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 120308.5625 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17706.3457 - 783us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53172.8047 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10305.9355 - 636us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 86445.7422 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4214.4385 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18648.9414 - 652us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36468.9609 - 807us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36074.9805 - 732us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61707.3672 - 2ms/epoch - 28us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33585.6836 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15206.5576 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54878.5117 - 681us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9561.8555 - 723us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4959.1841 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18956.3496 - 828us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20672.6074 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16495.4727 - 974us/epoch - 16us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 210610.0156 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22293.4258 - 649us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22068.1992 - 852us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16362.2402 - 710us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36625.4844 - 725us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49192.7812 - 759us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20912.6719 - 668us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 115037.1484 - 731us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26282.0039 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45587.4570 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18499.4531 - 738us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49378.7227 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12437.4805 - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27699.1484 - 778us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6745.6694 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15480.5566 - 734us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5492.2090 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31073.9492 - 723us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31533.9668 - 659us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16289.1436 - 713us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11898.0303 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 57100.9805 - 1ms/epoch - 21us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16685.3613 - 700us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27922.7949 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14612.4150 - 720us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123261.5547 - 908us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23249.1328 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34982.0781 - 745us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 127022.3047 - 625us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26955.6348 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17471.3477 - 678us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 75299.9453 - 690us/epoch - 11us/sample\n", - "Generation 45/300\n", - "Solving for Nash Equilibrium in Generation 45/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19612.8672 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38943.3594 - 892us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20438.7363 - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25778.9492 - 742us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74064.5312 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35263.5273 - 746us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8731.6377 - 732us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37345.7266 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8036.5522 - 701us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17459.7695 - 2ms/epoch - 25us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20423.3125 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21164.3555 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15724.8623 - 2ms/epoch - 26us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27158.3027 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34458.9141 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20400.2734 - 764us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3572.1682 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23222.8672 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17185.4883 - 1ms/epoch - 17us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 60672.3086 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20558.1250 - 7ms/epoch - 111us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18473.1816 - 849us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7955.4478 - 648us/epoch - 10us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 206942.5781 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24811.1113 - 711us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39874.0234 - 667us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28311.5996 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19778.8105 - 756us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21037.5195 - 770us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42337.1094 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13337.7471 - 800us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38074.5195 - 738us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18551.8809 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13793.8086 - 815us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123729.3516 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24578.2344 - 792us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10327.7646 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41060.6016 - 601us/epoch - 10us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 75502.1719 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9757.0801 - 646us/epoch - 10us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66813.5859 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11637.2559 - 772us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22732.2832 - 959us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56322.3359 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34693.0430 - 777us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 199774.6562 - 699us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26060.9727 - 754us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 80249.9219 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40219.6094 - 591us/epoch - 10us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15468.0039 - 3ms/epoch - 41us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14861.5020 - 630us/epoch - 10us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59742.8242 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90777.8047 - 1000us/epoch - 16us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18727.4414 - 724us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40805.5078 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15511.1836 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16733.9648 - 638us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 76389.6953 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24922.9512 - 818us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52686.1133 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4817.9116 - 976us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 76995.8594 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41337.4609 - 804us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13539.9775 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10793.5381 - 885us/epoch - 14us/sample\n", - "Generation 46/300\n", - "Solving for Nash Equilibrium in Generation 46/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 80037.8281 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31768.5332 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17511.0254 - 645us/epoch - 10us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19548.0840 - 728us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45298.6758 - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36109.4805 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45856.6953 - 801us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13516.5322 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43778.0039 - 642us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15093.3213 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34937.5781 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11945.1445 - 645us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30929.3496 - 775us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19978.9766 - 743us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52580.4258 - 764us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 118706.9219 - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13034.4639 - 680us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31181.7734 - 738us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32723.3301 - 755us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33497.8984 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27564.4258 - 1ms/epoch - 16us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 154392.4062 - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34604.1641 - 802us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51476.8711 - 877us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14994.9795 - 725us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13035.8105 - 1ms/epoch - 24us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 81854.3594 - 801us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14761.1260 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46191.5547 - 1ms/epoch - 20us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33904.6250 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 120657.4141 - 1ms/epoch - 18us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16703.4844 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48101.8086 - 857us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 110338.8516 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20733.7930 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35038.2773 - 709us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20204.4805 - 615us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56202.1562 - 745us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18899.0586 - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15091.5762 - 839us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19086.2559 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26393.4082 - 761us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28598.2695 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16484.1836 - 730us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22731.5586 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 185423.7812 - 640us/epoch - 10us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9228.2744 - 991us/epoch - 16us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19571.7598 - 705us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22060.7109 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9737.1250 - 809us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46322.6836 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45459.6328 - 713us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23460.0078 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32219.2148 - 906us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13947.6475 - 943us/epoch - 15us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4857.5894 - 2ms/epoch - 28us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 82740.2188 - 3ms/epoch - 43us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48442.6641 - 694us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20718.3066 - 778us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27155.0781 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47293.5156 - 853us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9055.9688 - 2ms/epoch - 27us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17579.4453 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61148.4922 - 929us/epoch - 15us/sample\n", - "Generation 47/300\n", - "Solving for Nash Equilibrium in Generation 47/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59682.5352 - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21505.5605 - 1ms/epoch - 17us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89297.8438 - 745us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24218.9434 - 738us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25444.6582 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22323.1016 - 860us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24352.8945 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21043.6602 - 739us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12851.0273 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47929.1914 - 830us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 124434.5000 - 761us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95915.9453 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22991.1094 - 709us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27733.0664 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25469.5039 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16943.6348 - 750us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 145379.0000 - 731us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16740.0020 - 863us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69312.5078 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18244.1172 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101886.1484 - 677us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77911.4922 - 913us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12790.1055 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54729.0000 - 843us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8703.0215 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79131.0156 - 3ms/epoch - 42us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38174.3945 - 687us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39147.0078 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18111.6758 - 729us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21979.9297 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51220.2930 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26499.9766 - 794us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48891.9492 - 713us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53419.8945 - 645us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26344.6582 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33014.5742 - 667us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26874.8535 - 816us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53076.6094 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 92us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41028.5312 - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 62365.9336 - 761us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23614.7344 - 714us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 150385.8906 - 662us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14769.0146 - 778us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40619.1445 - 745us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55149.8711 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38543.7656 - 692us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25316.2363 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88472.2109 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53482.7500 - 790us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17693.3848 - 744us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50669.9453 - 894us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18499.3770 - 670us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27325.2344 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9243.1982 - 880us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24186.2168 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56651.2266 - 629us/epoch - 10us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41128.5586 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14677.5811 - 715us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55622.3047 - 763us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23489.6133 - 718us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21289.4238 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29104.9805 - 956us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25548.4941 - 3ms/epoch - 52us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3460.2708 - 794us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23121.4023 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17296.5195 - 677us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Generation 48/300\n", - "Solving for Nash Equilibrium in Generation 48/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28361.2285 - 866us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16111.1436 - 761us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33165.3789 - 2ms/epoch - 39us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116523.3906 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19113.5234 - 714us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106122.6484 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25172.9336 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19121.3516 - 1ms/epoch - 17us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40786.5781 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42580.5703 - 723us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27084.1621 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21423.0762 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43498.6133 - 3ms/epoch - 49us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 84256.3203 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49635.4023 - 726us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24374.5254 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30943.4023 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 87209.4375 - 718us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29791.6406 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44448.5547 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26056.8242 - 692us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37896.7773 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45599.4688 - 756us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21162.2207 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 97883.5469 - 702us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19653.0645 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27640.0195 - 712us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54627.7305 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30367.1094 - 671us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4882.0537 - 740us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47647.7578 - 642us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 190140.3906 - 882us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24802.0254 - 3ms/epoch - 43us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18029.6602 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 167971.6094 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46681.3711 - 736us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19407.8789 - 546us/epoch - 9us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32014.9805 - 788us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21754.0547 - 722us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 169447.7031 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44963.4531 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26362.8145 - 773us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16523.2031 - 755us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 72414.8516 - 758us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 226386.0625 - 667us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44129.6914 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50510.1797 - 729us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48988.0391 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 82548.5000 - 766us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16181.6924 - 772us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59693.9180 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10199.2295 - 764us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19964.0508 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26089.6816 - 739us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25315.5215 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 93155.4922 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29042.9844 - 712us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17284.2773 - 660us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29315.1270 - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22236.8613 - 764us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46359.4727 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34035.3555 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50182.1914 - 739us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29469.4453 - 718us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65050.1328 - 2ms/epoch - 26us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 91036.2734 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25474.2480 - 674us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23430.7949 - 698us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112349.6953 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27795.2363 - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245843.2656 - 963us/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13072.2617 - 854us/epoch - 14us/sample\n", - "Generation 49/300\n", - "Solving for Nash Equilibrium in Generation 49/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 115640.2031 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 154215.3750 - 700us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61483.3711 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38578.5234 - 3ms/epoch - 46us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28411.1836 - 725us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 62230.2266 - 758us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11121.7617 - 888us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5975.4858 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19093.1270 - 709us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30305.0762 - 760us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33153.5469 - 696us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55616.2656 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17175.6484 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304262.2812 - 675us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 63483.4102 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40848.8594 - 806us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44239.3438 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 100306.2656 - 798us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12838.0195 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79378.5156 - 779us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34001.7266 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 203179.8750 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31246.1230 - 933us/epoch - 15us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41728.7031 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17169.1289 - 976us/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 60129.2734 - 804us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 73817.5547 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17630.1035 - 664us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4064.9980 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 218460.8125 - 727us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38379.5820 - 706us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55916.8984 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34447.7656 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36044.3633 - 787us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42562.4688 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66347.7031 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31676.0273 - 751us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65545.2188 - 706us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 162736.6250 - 703us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 63312.8711 - 862us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24989.7266 - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32627.9258 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56549.9922 - 736us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34822.8672 - 685us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98183.3594 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19591.5117 - 784us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30356.0898 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69464.3594 - 974us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 67266.4766 - 788us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37085.7773 - 851us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69861.6406 - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11638.6357 - 842us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20586.3496 - 753us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11921.1104 - 800us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18632.1484 - 1ms/epoch - 22us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31368.2969 - 832us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11627.7051 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34307.4570 - 802us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65877.3359 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48242.5312 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 149585.3438 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 129368.0703 - 716us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4667.2388 - 869us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 99915.8281 - 1ms/epoch - 18us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 130416.6172 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27669.5879 - 744us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33822.8594 - 657us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18753.9590 - 850us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32795.6758 - 866us/epoch - 14us/sample\n", - "Generation 50/300\n", - "Solving for Nash Equilibrium in Generation 50/300\n", + "Generation 79/300\n", + "Solving for Nash Equilibrium in Generation 79/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 57610.2070 - 707us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5045.0840 - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 213480.0156 - 1ms/epoch - 19us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71042.7188 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31736.6621 - 656us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95727.4688 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11230.2861 - 739us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39082.1602 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 173233.7344 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50700.2852 - 695us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14844.0684 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33489.8008 - 722us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25680.1211 - 3ms/epoch - 40us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 64228.3984 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28041.3887 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 57343.0938 - 665us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 210462.9062 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 92112.6641 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32472.8418 - 749us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40483.7734 - 807us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33137.2109 - 666us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83480.4531 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16439.2969 - 783us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20720.5781 - 845us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42912.1719 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 204656.7656 - 747us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46602.5352 - 684us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17349.2871 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101183.7031 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5603.8833 - 612us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33291.6641 - 812us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28835.2090 - 727us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 198087.4062 - 704us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37750.8594 - 3ms/epoch - 45us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 192721.1250 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74028.7031 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39572.2422 - 830us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74327.4688 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19150.0449 - 774us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 109699.0859 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68210.2188 - 871us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29436.5469 - 802us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19385.1855 - 707us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18571.6895 - 688us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 70073.1094 - 730us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357060.1250 - 749us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 80527.3906 - 695us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 192698.4531 - 702us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 202162.6250 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14424.3975 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10220.9424 - 749us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 225387.1406 - 706us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47903.7070 - 744us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45098.0156 - 689us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 35964.3359 - 688us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116303.3359 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68523.0234 - 757us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69714.4141 - 748us/epoch - 12us/sample\n", - "Generation 51/300\n", - "Solving for Nash Equilibrium in Generation 51/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25317.1523 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40102.0508 - 694us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322268.5312 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 75522.3203 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41220.8281 - 660us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90899.7734 - 706us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39552.0000 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38130.7344 - 797us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48682.1211 - 792us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68621.9453 - 718us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66798.6875 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41024.6484 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20104.2227 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256102.7656 - 762us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38204.2852 - 728us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46321.9102 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17441.9570 - 660us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 118612.4922 - 704us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27611.1777 - 762us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28419.1113 - 816us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54531.4805 - 934us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261465.0781 - 738us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42661.9570 - 916us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 120872.7656 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17740.0039 - 736us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 128352.2188 - 685us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46035.5781 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44104.5000 - 740us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 140429.0312 - 810us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271289.5312 - 1ms/epoch - 18us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355055.4375 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40531.8594 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20073.7422 - 649us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 68us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236884.7969 - 5ms/epoch - 81us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41226.4414 - 658us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 183941.7188 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20696.6133 - 738us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21246.5449 - 732us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 132530.4219 - 675us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55385.5352 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41939.1602 - 773us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29024.2344 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314181.7812 - 653us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27335.4199 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106925.9766 - 742us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 81269.5000 - 731us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28361.5527 - 695us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473928.7812 - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77343.5938 - 769us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 117728.8672 - 1ms/epoch - 18us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14483.3066 - 877us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 178345.5469 - 709us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 94662.9453 - 729us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45709.0391 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38976.9570 - 1ms/epoch - 22us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26386.6973 - 726us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 91351.4062 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37854.4961 - 2ms/epoch - 33us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52597.7812 - 716us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19141.1641 - 688us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267341.4375 - 719us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55953.6445 - 715us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24401.5137 - 682us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31022.6992 - 717us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29763.0410 - 760us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44387.7656 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24936.5996 - 656us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 110628.9453 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89032.0234 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22126.8652 - 732us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Generation 52/300\n", - "Solving for Nash Equilibrium in Generation 52/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43651.6523 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 75889.6562 - 732us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8403.0352 - 664us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 76711.3281 - 2ms/epoch - 30us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 82785.6641 - 711us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14553.5332 - 801us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638370.3125 - 775us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90285.1328 - 662us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 195747.4688 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 85537.1641 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123310.0312 - 1ms/epoch - 18us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 86859.7188 - 837us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7419.1699 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95079.7969 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41836.0781 - 755us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20466.1406 - 836us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47010.2305 - 719us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34530.9297 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25402.5449 - 684us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98616.5781 - 706us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26077.5996 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26786.0215 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 82774.8672 - 719us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126990.2422 - 4ms/epoch - 65us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41986.7539 - 718us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54695.5508 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89931.5078 - 649us/epoch - 10us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23196.0059 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26746.5078 - 792us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 159901.9219 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26885.1250 - 707us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 122340.3828 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 145242.2188 - 932us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53725.9336 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166013.6875 - 839us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 142569.9531 - 1ms/epoch - 22us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37821.9453 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46735.6641 - 843us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31867.4727 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79114.4297 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 103081.6641 - 881us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 120661.5859 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32240.8613 - 875us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 80232.6484 - 752us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14711.6953 - 782us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32762.3652 - 723us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65062.0703 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27796.2305 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 187197.6719 - 924us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20351.3340 - 1ms/epoch - 22us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 139637.8438 - 3ms/epoch - 41us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 85493.8438 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401186.5312 - 724us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 63314.8789 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22063.0078 - 788us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47999.2227 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17429.7773 - 867us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 113743.7578 - 688us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 190951.9062 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23814.2695 - 810us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 119343.5859 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33821.2109 - 769us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43302.9180 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43484.6133 - 831us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11180.4072 - 6ms/epoch - 104us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15068.9102 - 715us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 133786.9375 - 632us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49620.5664 - 769us/epoch - 12us/sample\n", - "Generation 53/300\n", - "Solving for Nash Equilibrium in Generation 53/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30859.9648 - 5ms/epoch - 76us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 105688.7891 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 121711.4297 - 835us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 129108.5625 - 753us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243210.1562 - 603us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44399.8594 - 955us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49418.2852 - 641us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42749.2695 - 756us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83491.2422 - 723us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51149.6094 - 796us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45268.9219 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41983.9453 - 803us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46331.7617 - 1ms/epoch - 21us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30468.8301 - 731us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53495.2422 - 741us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61569.1797 - 1ms/epoch - 20us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15256.3936 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357373.9688 - 1ms/epoch - 22us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33196.3633 - 837us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33297.6836 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41032.1797 - 727us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 183764.6875 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248003.5938 - 741us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8667.3691 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301964.6250 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 141212.5469 - 3ms/epoch - 41us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25421.5137 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46599.6172 - 694us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112029.9844 - 824us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45052.0430 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26069.2402 - 713us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 198162.1406 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89439.1953 - 803us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42098.1211 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 234652.5781 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 138930.5000 - 1ms/epoch - 17us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101971.6484 - 803us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 119235.1797 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31305.9844 - 842us/epoch - 14us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55243.5117 - 665us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 107120.5391 - 849us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24450.2656 - 686us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9901.6299 - 782us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 87029.3672 - 776us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 99835.0938 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 64489.5703 - 931us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54577.0000 - 700us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65616.9141 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 103809.1484 - 786us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 154166.0781 - 694us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 93682.0625 - 721us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47389.0508 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 62060.3867 - 715us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15195.5693 - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106263.1953 - 750us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65313.1719 - 756us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31800.3594 - 2ms/epoch - 29us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 155976.7031 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 148478.4375 - 711us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95248.1875 - 739us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 120450.2422 - 775us/epoch - 12us/sample\n", - "Generation 54/300\n", - "Solving for Nash Equilibrium in Generation 54/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43828.2812 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 102959.9219 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56501.3359 - 868us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422271.8125 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 143701.0781 - 738us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 78479.3203 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49708.4648 - 669us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44442.6758 - 1ms/epoch - 16us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 99849.4766 - 705us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 62612.2500 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562431.2500 - 783us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 18858.6543 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49301.2305 - 791us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6901.9038 - 939us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31593.2676 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46074.8242 - 730us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52818.6602 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98731.9375 - 745us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39408.2773 - 631us/epoch - 10us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24550.7266 - 821us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 140747.1406 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16407.8770 - 740us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65249.1523 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66612.7031 - 717us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17230.1934 - 929us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 99346.7734 - 770us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274707.2188 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33817.1328 - 701us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166221.3906 - 679us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44766.6211 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48086.6992 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 217649.6719 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 167935.6719 - 713us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24356.4336 - 2ms/epoch - 30us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51779.8438 - 760us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344271.7188 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52917.9609 - 625us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25088.8203 - 839us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 118372.2891 - 732us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52941.5430 - 833us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284978.0625 - 870us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 422269.9375 - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38464.7891 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 105439.7031 - 661us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19222.6992 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 73712.1172 - 744us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235695.3594 - 718us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 70889.2188 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26355.5156 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104077.7656 - 753us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8558.0996 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83900.4375 - 752us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299630.7500 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 135036.7344 - 680us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276830.9688 - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 107612.3359 - 721us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21601.2070 - 678us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 191458.2812 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29654.1875 - 710us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 122297.5938 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106894.9844 - 652us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338941.7812 - 688us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22142.7383 - 4ms/epoch - 59us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106863.8516 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44515.7227 - 888us/epoch - 14us/sample\n", - "Generation 55/300\n", - "Solving for Nash Equilibrium in Generation 55/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13162.6377 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54966.8164 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236415.0625 - 798us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27014.1113 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30567.0078 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71090.7031 - 3ms/epoch - 41us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 57118.2695 - 774us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71017.9922 - 747us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116127.6875 - 2ms/epoch - 28us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 185228.0312 - 662us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44289.5859 - 630us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 111512.7188 - 788us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 169859.7969 - 669us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 184371.1406 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41719.6797 - 680us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244520.6562 - 758us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66331.6562 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36434.0547 - 715us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 129075.9453 - 722us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20367.6641 - 699us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 122110.8125 - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106438.5781 - 687us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 114280.2266 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17868.5156 - 798us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116145.5781 - 734us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69751.3359 - 788us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 76239.4219 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430135.2188 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32566.8242 - 2ms/epoch - 26us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 186801.8750 - 618us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38049.0977 - 717us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61978.5469 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 129554.2734 - 725us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27538.2832 - 817us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66743.6328 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 67487.4844 - 644us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 140910.5625 - 816us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352903.5625 - 767us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24518.4961 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98911.4453 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 57225.2812 - 2ms/epoch - 26us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 191931.7812 - 723us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14340.4951 - 1ms/epoch - 19us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29007.6348 - 734us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232585.0000 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37287.5078 - 671us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68978.8516 - 658us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83104.6953 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69298.8594 - 2ms/epoch - 31us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293330.7812 - 745us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 75472.5703 - 865us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 70587.1641 - 775us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 63443.0703 - 663us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 58628.8320 - 4ms/epoch - 63us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 122289.4297 - 711us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272520.0000 - 688us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 57875.0938 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49128.1094 - 720us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201374.4219 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335432.9375 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61900.4922 - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303214.9688 - 1ms/epoch - 18us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46432.2031 - 762us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 198081.5938 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59591.9180 - 2ms/epoch - 32us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59021.7891 - 756us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41647.8281 - 770us/epoch - 12us/sample\n", - "Generation 56/300\n", - "Solving for Nash Equilibrium in Generation 56/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 150890.8438 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71818.4453 - 710us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5863.6025 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 70627.7266 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 64126.7695 - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20182.4980 - 725us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 149832.1562 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 102524.8672 - 796us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 154019.8125 - 800us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56442.0273 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 179711.8594 - 765us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29929.2285 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 137958.1875 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 142131.5156 - 680us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 207220.2344 - 702us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 15856.2979 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 189689.8906 - 2ms/epoch - 35us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66021.0625 - 798us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214488.5312 - 758us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 119961.1172 - 639us/epoch - 10us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 184053.9219 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77965.3047 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 210707.1562 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 76072.9922 - 833us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272711.0938 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 53704.9766 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 152573.4375 - 844us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 75454.9062 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101650.0547 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61722.6016 - 752us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 186866.5625 - 611us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36594.1016 - 747us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 31351.3457 - 879us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16027.6006 - 2ms/epoch - 34us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66212.4531 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 216947.4688 - 705us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201911.1875 - 708us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55726.5586 - 811us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 37189.7188 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 78203.9922 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68016.2031 - 663us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 160063.7031 - 698us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 193900.7656 - 779us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 477469.0938 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 81556.0938 - 735us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38179.6211 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 203420.1250 - 741us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 163573.8125 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 157617.5938 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20377.8301 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 64029.2305 - 798us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65168.8281 - 717us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376703.7500 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98294.8828 - 902us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79591.6016 - 659us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69683.7344 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 81034.2266 - 755us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88415.6406 - 730us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 67684.0547 - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1095110.5000 - 636us/epoch - 10us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88231.1719 - 901us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 130418.2578 - 851us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 121831.4297 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68057.5078 - 731us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 205831.0312 - 701us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44728.9844 - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 93205.7734 - 802us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299229.8750 - 745us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 127935.8203 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 82us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95083.9375 - 2ms/epoch - 29us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 81521.7969 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166977.7188 - 656us/epoch - 11us/sample\n", - "Generation 57/300\n", - "Solving for Nash Equilibrium in Generation 57/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8650.5996 - 782us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 72378.1641 - 857us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 84783.0625 - 738us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45312.2852 - 716us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356468.6875 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 85555.4766 - 670us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68173.8828 - 667us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69184.7578 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55796.3750 - 693us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 60031.3008 - 1ms/epoch - 17us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 93533.2031 - 773us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 158179.3438 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 141473.7344 - 753us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33652.4180 - 659us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 96909.1406 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 73557.6094 - 746us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 284711.6875 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 221403.2812 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74266.0000 - 760us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 144992.1875 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 135328.9219 - 3ms/epoch - 45us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237379.9375 - 704us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 148222.4219 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 64864.7617 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34862.8047 - 664us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395131.0312 - 824us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 84090.9531 - 695us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316558.7500 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 111083.7656 - 753us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402257.2188 - 711us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 148538.4219 - 745us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 58664.7695 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29117.2637 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 162973.5938 - 2ms/epoch - 28us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 303873.2188 - 668us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 158135.7188 - 695us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90875.6719 - 679us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 165883.8594 - 730us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77451.8672 - 640us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 93739.4219 - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90267.2969 - 985us/epoch - 16us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 167796.8281 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22840.5664 - 570us/epoch - 9us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 150755.0625 - 684us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77011.3359 - 643us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104740.8047 - 703us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69023.0938 - 735us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49678.6797 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 154349.6719 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 189757.6094 - 741us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 162465.7812 - 630us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 121344.0156 - 854us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 198754.6875 - 784us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 156590.0625 - 2ms/epoch - 30us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 110531.6328 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232593.8594 - 763us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 516529.4062 - 578us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 96474.8984 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90654.9531 - 788us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 978481.3125 - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 96845.9219 - 783us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 87867.1797 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 153744.6094 - 771us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116507.3438 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36642.0820 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725693.5625 - 739us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 100793.5234 - 735us/epoch - 12us/sample\n", - "Generation 58/300\n", - "Solving for Nash Equilibrium in Generation 58/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373085.0000 - 1ms/epoch - 17us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 161596.7031 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 67352.5625 - 725us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275818.0312 - 780us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 100891.5234 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47666.4531 - 641us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 173333.5781 - 966us/epoch - 16us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444967.0312 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83281.1016 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51312.8086 - 714us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609062.0000 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61832.3906 - 758us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306532.7500 - 719us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256427.2656 - 726us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25151.5293 - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 192257.0469 - 779us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258856.4844 - 963us/epoch - 16us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43489.4727 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 105387.9375 - 800us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 159727.1719 - 638us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44125.5547 - 741us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 82304.6797 - 591us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331911.5938 - 657us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83027.8359 - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180359.2188 - 767us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 165375.4531 - 1ms/epoch - 18us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 107855.5156 - 948us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358871.0000 - 713us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46068.9648 - 656us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268123.6562 - 668us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27267.7324 - 720us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56420.9023 - 688us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265815.8438 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 560409.4375 - 804us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46675.1172 - 708us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 181296.7812 - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 56843.1211 - 791us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 608580.4375 - 736us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66747.0625 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 85351.3906 - 1ms/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 183036.9688 - 618us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 81704.4844 - 814us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 58163.0977 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89916.8359 - 817us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13204.3555 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83153.5078 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719513.8750 - 879us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 167191.2031 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258052.9844 - 685us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40369.2969 - 716us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14430.5547 - 785us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 173516.6719 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54291.3203 - 963us/epoch - 16us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 178307.3281 - 758us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34309.6641 - 1ms/epoch - 17us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 24766.4727 - 727us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30696.4023 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257562.8438 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66369.5547 - 709us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277701.8438 - 2ms/epoch - 28us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 115494.8281 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 334794.9375 - 711us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214079.8750 - 722us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95052.6172 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 117838.3672 - 788us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89358.0234 - 717us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Generation 59/300\n", - "Solving for Nash Equilibrium in Generation 59/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 80581.1094 - 845us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 70741.5000 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1037665.2500 - 738us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83408.4062 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68251.1719 - 744us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 161276.2656 - 686us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 139124.2188 - 778us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326590.4688 - 4ms/epoch - 64us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369908.1562 - 1ms/epoch - 21us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41435.9648 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278944.7500 - 720us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89798.0000 - 763us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 181231.1875 - 818us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166134.3594 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 91947.2656 - 675us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 215911.6250 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 162603.4688 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 105437.3516 - 786us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106778.8594 - 774us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77849.9141 - 667us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112921.0312 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95408.9531 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 57869.1523 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88819.5859 - 922us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 58064.7461 - 818us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 120961.7344 - 848us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 209943.6406 - 755us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214176.8438 - 4ms/epoch - 60us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180827.2656 - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 131572.1094 - 770us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 60744.3164 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485251.2500 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166237.1406 - 2ms/epoch - 28us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166844.4531 - 841us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314236.3125 - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 87980.3516 - 720us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 145158.5156 - 1ms/epoch - 19us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71992.0547 - 684us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95293.7031 - 701us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475979.0000 - 878us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402945.8438 - 4ms/epoch - 62us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 175694.2344 - 888us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 129648.1797 - 929us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 159422.6719 - 883us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 199647.3125 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43808.4219 - 751us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259376.8750 - 664us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16138.6113 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90647.6484 - 733us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10265.2559 - 837us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 528823.3125 - 816us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 131661.5469 - 777us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 159392.8750 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21906.7246 - 1ms/epoch - 19us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104585.7422 - 692us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 138896.9375 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47209.3984 - 695us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201466.7812 - 664us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28665.5723 - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201235.6250 - 808us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 800103.0625 - 968us/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59222.8555 - 734us/epoch - 12us/sample\n", - "Generation 60/300\n", - "Solving for Nash Equilibrium in Generation 60/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32427.5059 - 730us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 217949.4062 - 856us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 169785.2656 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116395.8438 - 709us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201716.9062 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 196051.9375 - 853us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 99909.7266 - 724us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 95699.3906 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339209.9688 - 679us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65002.7930 - 2ms/epoch - 31us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98101.0625 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54830.3750 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297424.1250 - 763us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71983.5000 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500412.0000 - 757us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 130156.9922 - 871us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 204687.5000 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558126.7500 - 946us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180508.9375 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 233899.3750 - 1ms/epoch - 18us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88002.6953 - 777us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 80582.8906 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 219183.5312 - 744us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 200073.1406 - 710us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 76838.5625 - 709us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544692.6875 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77969.5703 - 1ms/epoch - 19us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 185812.6250 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 203888.9531 - 1ms/epoch - 21us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 107093.4688 - 680us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 789592.7500 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 158261.9375 - 910us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 153409.4531 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 218794.1719 - 901us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 124637.1641 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 96797.6875 - 662us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272413.4062 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89726.3281 - 711us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534302.5000 - 736us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48137.0547 - 639us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123860.6719 - 910us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289255.2500 - 942us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 218160.1875 - 735us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 98000.1875 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 97720.6953 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 103301.9609 - 643us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 190927.7188 - 787us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36870.6992 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 100162.0000 - 833us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 221939.3750 - 771us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180580.2656 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66294.1484 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104797.5859 - 761us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106197.7734 - 869us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 211659.0000 - 829us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106839.7500 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 29772.3496 - 920us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104202.1328 - 779us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 96204.6953 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55219.7109 - 735us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 187876.7344 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126703.1875 - 699us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562934.2500 - 712us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 128889.1875 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33301.5312 - 787us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166774.5469 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 147475.5781 - 865us/epoch - 14us/sample\n", - "Generation 61/300\n", - "Solving for Nash Equilibrium in Generation 61/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32756.9043 - 697us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101576.1875 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83754.1328 - 4ms/epoch - 61us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246365.9688 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112576.3438 - 881us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 234471.4531 - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69777.7031 - 987us/epoch - 16us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 309375.2188 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 129557.5625 - 784us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286415.4688 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 110108.8594 - 738us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 96746.5391 - 703us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19240.1309 - 629us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672389.8125 - 718us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 339941.9375 - 1ms/epoch - 19us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 168758.3281 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 171612.5312 - 615us/epoch - 10us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301276.2188 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 221759.0312 - 744us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 91677.5156 - 700us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375743.1562 - 1ms/epoch - 22us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 58331.7109 - 762us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55939.4062 - 2ms/epoch - 28us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 227039.5469 - 831us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332264.0625 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112086.7891 - 659us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336374.1250 - 881us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 92173.6094 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 160091.5938 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 191778.5312 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338059.6250 - 779us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 50751.9141 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 137287.7812 - 670us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 150878.0312 - 833us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241820.1562 - 736us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 86225.0391 - 785us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1572941.3750 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270446.0625 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 36837.8516 - 728us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180009.7188 - 2ms/epoch - 37us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 60144.3555 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 91436.6094 - 860us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 252168.2344 - 1ms/epoch - 20us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 75131.5234 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 158152.1406 - 755us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33055.8086 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603363.6250 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 61539.5977 - 2ms/epoch - 25us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 486990.3125 - 730us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 154312.1250 - 769us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287040.6562 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 603497.1875 - 656us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 241644.1406 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 118386.5000 - 754us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180617.5156 - 739us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317506.0625 - 752us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 236721.9375 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 175377.1250 - 741us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55405.1992 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446564.9062 - 747us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 189339.5938 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20306.7793 - 768us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 119609.8203 - 709us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 134081.7031 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 150515.7812 - 843us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 41915.9688 - 785us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 119394.2109 - 752us/epoch - 12us/sample\n", - "Generation 62/300\n", - "Solving for Nash Equilibrium in Generation 62/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283527.5312 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372576.1875 - 959us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 155034.3125 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366214.1562 - 716us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52785.2734 - 792us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20889.6816 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 121118.4453 - 945us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257222.4062 - 1ms/epoch - 24us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 54783.7422 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 32062.6074 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 103517.6953 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 151973.1094 - 789us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693197.6250 - 720us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 46834.3555 - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 147837.0469 - 1ms/epoch - 16us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263042.1875 - 985us/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 80185.0000 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123704.0703 - 754us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 313633.0938 - 820us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 176884.7812 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 174250.4688 - 823us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77782.1875 - 652us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 153927.2812 - 878us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1053262.7500 - 855us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 172127.1719 - 784us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 173663.2812 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361792.3750 - 761us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 110089.1328 - 698us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276755.3750 - 658us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265072.3438 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402046.5938 - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 102146.0781 - 858us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504248.6562 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 115342.8594 - 712us/epoch - 11us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245366.0000 - 679us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247749.2656 - 892us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247634.7656 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278871.9688 - 739us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 84121.3906 - 665us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375484.8750 - 691us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66447.6953 - 672us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248347.0781 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34195.6484 - 922us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356873.3438 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264511.7188 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 151215.7969 - 722us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 131536.1719 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42231.7695 - 659us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230140.7031 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 224254.0312 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1373962.2500 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88817.5938 - 719us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 170307.9062 - 777us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 134371.0312 - 661us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 97024.2266 - 700us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273318.7812 - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316389.7188 - 841us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267175.3125 - 706us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 132136.2188 - 938us/epoch - 15us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 78573.3906 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384282.1250 - 746us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10333.9463 - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 164757.1562 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59941.2578 - 763us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666206.0000 - 784us/epoch - 13us/sample\n", - "Generation 63/300\n", - "Solving for Nash Equilibrium in Generation 63/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646118.5000 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 130408.4062 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68754.6484 - 739us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 108380.4219 - 638us/epoch - 10us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 139528.1875 - 792us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 517701.6250 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 131875.1719 - 761us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296097.0000 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285730.5938 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104657.9375 - 780us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366464.4688 - 709us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275854.5938 - 751us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684519.8750 - 802us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391455.0938 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 205501.0781 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 129736.8984 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269972.1562 - 2ms/epoch - 31us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394957.8125 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101155.0312 - 684us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387048.6875 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281411.8750 - 679us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 680091.5625 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 127541.8359 - 756us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1269293.3750 - 980us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 247361.9219 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 132755.7344 - 725us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 85720.3906 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408377.7500 - 829us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 85034.7422 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 119591.7578 - 700us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 52383.2969 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277536.1875 - 818us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250098.0469 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 63369.2188 - 1ms/epoch - 17us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244227.3438 - 751us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 134290.9844 - 754us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13017.1895 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326098.1250 - 824us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 253114.4375 - 1ms/epoch - 21us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123723.0312 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 118741.9141 - 673us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 63254.5977 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 184672.9219 - 1ms/epoch - 18us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 540343.2500 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 107489.9844 - 839us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 186807.8281 - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430791.7812 - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83486.2344 - 717us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257862.5625 - 651us/epoch - 10us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289080.8438 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 130302.2578 - 686us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 137709.1250 - 804us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288558.2188 - 921us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 66195.9844 - 1ms/epoch - 17us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 108750.7812 - 835us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436884.7188 - 945us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 152045.4531 - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 174202.7188 - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450137.1562 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45355.0391 - 847us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 65705.4766 - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306936.6875 - 815us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266710.8750 - 5ms/epoch - 75us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 225579.3594 - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 178859.4531 - 889us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298869.3750 - 821us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606702.4375 - 2ms/epoch - 24us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 107315.0156 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420991.8438 - 790us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 216578.5625 - 923us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1035741.9375 - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 199024.7344 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166516.5156 - 770us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 155473.9688 - 865us/epoch - 14us/sample\n", - "Generation 64/300\n", - "Solving for Nash Equilibrium in Generation 64/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 92606.3203 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42672.5859 - 743us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 187536.7031 - 623us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317085.0000 - 804us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88325.1172 - 797us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326109.8125 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610622.7500 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287987.2500 - 1ms/epoch - 17us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 165483.4375 - 798us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566042.5000 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 97787.6406 - 795us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51388.4922 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 164981.2812 - 752us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 364118.7188 - 852us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378212.0000 - 893us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1315665.2500 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69362.4609 - 828us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248143.0781 - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1053973.6250 - 762us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 156832.9531 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 39256.7227 - 725us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 105159.8047 - 708us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 738689.6250 - 701us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 137451.6719 - 794us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 527802.6250 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1531233.8750 - 1ms/epoch - 17us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44065.7422 - 1ms/epoch - 17us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261743.8125 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101988.9375 - 753us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90761.5312 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 89533.4531 - 770us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349772.6562 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 91926.9141 - 709us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325979.5312 - 693us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476441.0625 - 748us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 91980.1094 - 790us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 99490.5625 - 688us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1566064.7500 - 776us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 177182.5625 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 205308.7344 - 2ms/epoch - 26us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 213960.6875 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 329627.4062 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 205481.0938 - 2ms/epoch - 39us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 97117.3047 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 117942.6719 - 667us/epoch - 11us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 110022.9531 - 753us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 184075.3438 - 7ms/epoch - 120us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 129178.1562 - 754us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 786228.9375 - 694us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 210106.5312 - 1ms/epoch - 24us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251766.4531 - 733us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1016196.5625 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74013.6562 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 216653.0781 - 3ms/epoch - 49us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504314.4688 - 841us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 99033.0156 - 2ms/epoch - 26us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 249420.0469 - 750us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 221650.6875 - 775us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 131257.8281 - 639us/epoch - 10us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 125409.0391 - 736us/epoch - 12us/sample\n", - "Generation 65/300\n", - "Solving for Nash Equilibrium in Generation 65/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351824.0312 - 771us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318961.0000 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485845.3438 - 2ms/epoch - 33us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 217846.5156 - 710us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14545.6768 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 87110.1328 - 868us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 192555.5312 - 767us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 147483.7344 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 97454.0312 - 746us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 194217.4531 - 711us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104647.6172 - 1ms/epoch - 18us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83396.0703 - 673us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1632926.6250 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 124994.0703 - 1ms/epoch - 23us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 267775.3750 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502935.6562 - 2ms/epoch - 27us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44286.0898 - 947us/epoch - 15us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302095.0000 - 842us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 720960.9375 - 685us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287960.3125 - 800us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 174075.0000 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281627.6250 - 1ms/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325913.9375 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348714.5625 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101378.6953 - 780us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 131376.6094 - 732us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366841.5625 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 435229.0312 - 788us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 434155.1250 - 702us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166626.3125 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352695.8125 - 1ms/epoch - 18us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 107600.1641 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 177726.8125 - 782us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 135276.1406 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 202291.6094 - 785us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 196107.5156 - 790us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243140.8125 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 200442.1719 - 3ms/epoch - 51us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351691.5625 - 784us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1520507.2500 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250696.1094 - 797us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230130.5938 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521774.1875 - 703us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 203761.0625 - 808us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265256.5625 - 751us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349742.0625 - 694us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383312.1250 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 695us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 128123.8281 - 778us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593700.7500 - 781us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180838.0312 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673045.3750 - 743us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365466.3438 - 693us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 23370.0449 - 853us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 515975.3438 - 2ms/epoch - 37us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514834.4062 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 774529.5625 - 804us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268553.6562 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 988972.0625 - 698us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 399819.7188 - 760us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232010.5000 - 701us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677034.3750 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 62240.3789 - 828us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 223338.6562 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381224.2500 - 753us/epoch - 12us/sample\n", - "Generation 66/300\n", - "Solving for Nash Equilibrium in Generation 66/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 138014.0312 - 866us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101710.4844 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 925528.9375 - 834us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370302.7812 - 648us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433012.2812 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 169032.7188 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 192393.0625 - 717us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344142.8750 - 693us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214311.9531 - 922us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43683.4219 - 752us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 473477.1250 - 849us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1108512.2500 - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351167.9375 - 741us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 87391.2109 - 890us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 44993.6719 - 735us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 423950.0312 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 139123.8750 - 690us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543896.7500 - 749us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 444556.1875 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 196679.4062 - 720us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 219485.2812 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 90067.1094 - 746us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240252.7969 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246674.3125 - 765us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297466.4062 - 770us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 198753.1875 - 657us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656905.5000 - 842us/epoch - 14us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1016787.5000 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304654.0625 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550188.4375 - 739us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534587.3750 - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 17ms/epoch - 282us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 117973.6094 - 1ms/epoch - 18us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460155.8438 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45770.2188 - 636us/epoch - 10us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378779.3750 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274092.1875 - 756us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365871.9062 - 754us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2095219.3750 - 732us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 161809.8438 - 797us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 581518.5625 - 723us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180317.4375 - 816us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 20452.1719 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 60076.3203 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411618.7812 - 747us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415841.2188 - 734us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 186383.3906 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272507.1562 - 800us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 183081.0156 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 223682.3438 - 890us/epoch - 14us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373771.3750 - 979us/epoch - 16us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 147033.2031 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380545.1562 - 759us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503654.4062 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 118447.1953 - 863us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 214726.8125 - 680us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549976.4375 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 454137.6875 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116834.5703 - 829us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488847.3125 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232943.8594 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 118237.9922 - 683us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700618.0625 - 2ms/epoch - 29us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 260764.0000 - 820us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 187499.8438 - 956us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 263439.3750 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 400130.4062 - 756us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497403.3438 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555635.2500 - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 80us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 148048.1562 - 2ms/epoch - 38us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 345108.8750 - 768us/epoch - 12us/sample\n", - "Generation 67/300\n", - "Solving for Nash Equilibrium in Generation 67/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 38058.7500 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261399.9531 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 858708.8750 - 770us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394549.3438 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 168721.6250 - 1ms/epoch - 17us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 127966.8672 - 822us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 150435.9844 - 799us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476441.0312 - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1370270.6250 - 3ms/epoch - 48us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635634.6875 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 161950.4844 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 193629.5156 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 170117.4375 - 695us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 172197.4375 - 847us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 110us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 62373.6992 - 2ms/epoch - 32us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265384.2812 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 885624.3750 - 803us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514333.0312 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101192.0156 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 155479.2656 - 755us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428565.1562 - 806us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 318644.5625 - 974us/epoch - 16us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 718163.2500 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 839503.5625 - 811us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10646.0225 - 732us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259713.8750 - 784us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510569.9062 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248165.9219 - 671us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281459.6562 - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1146587.3750 - 1ms/epoch - 17us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650649.8750 - 701us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1118855.0000 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 100518.0312 - 792us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 781106.8750 - 969us/epoch - 16us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 470481.2500 - 802us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2837780.2500 - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 160499.4688 - 916us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419725.5312 - 757us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 55409.6914 - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 113598.2969 - 592us/epoch - 10us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506631.1250 - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1465812.1250 - 738us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 179878.9688 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287020.9375 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289798.4688 - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 209307.9844 - 713us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166737.4219 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280383.9062 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 92049.7500 - 4ms/epoch - 62us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 48us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451616.4062 - 821us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 181017.5312 - 765us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1802986.8750 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 250657.6250 - 980us/epoch - 16us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 173764.4688 - 841us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 594419.8750 - 735us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 137158.4688 - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238325.4375 - 671us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 163722.6562 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 491461.5938 - 790us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1386113.0000 - 3ms/epoch - 45us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1520939.2500 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433006.0938 - 1ms/epoch - 21us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320070.3750 - 669us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 199270.0781 - 840us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 894505.3750 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 51707.2773 - 785us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 823979.6250 - 1ms/epoch - 17us/sample\n", - "Generation 68/300\n", - "Solving for Nash Equilibrium in Generation 68/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290752.9688 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 58086.8047 - 902us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1442656.3750 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 33710.7617 - 747us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 208130.2344 - 4ms/epoch - 58us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 867629.3125 - 774us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352615.0938 - 775us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1321215.2500 - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 155443.9688 - 730us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1005075.5000 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575678.7500 - 838us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511785.1875 - 720us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 213458.2344 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 96485.0547 - 722us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311907.1875 - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254079.1250 - 705us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 445590.1250 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1197141.5000 - 756us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350250.0938 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 639119.9375 - 1ms/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 192062.4062 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112946.0703 - 1ms/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 193573.5938 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 108517.8906 - 948us/epoch - 15us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 514707.5938 - 800us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27845.7012 - 977us/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 627909.6875 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 195533.3906 - 942us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297769.0312 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 281123.1562 - 928us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647338.0000 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297958.0625 - 901us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376835.4375 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323316.1875 - 815us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 216086.8125 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328631.3750 - 747us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479324.3750 - 824us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1625537.7500 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 109865.0078 - 795us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 896928.5625 - 771us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280088.6250 - 992us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1875991.7500 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 220941.3125 - 719us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461894.0625 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1898578.8750 - 783us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 741416.1875 - 873us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68832.4453 - 830us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 136565.3750 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 220819.4688 - 900us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201562.8906 - 763us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235409.7031 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299478.8438 - 670us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1120087.7500 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279775.4688 - 919us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461225.0938 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 223915.7969 - 773us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401129.8125 - 1ms/epoch - 18us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 948157.8750 - 1ms/epoch - 17us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 194908.7656 - 850us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116629.2891 - 633us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275676.9375 - 761us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437205.1875 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 550929.2500 - 891us/epoch - 14us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9850.5596 - 788us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1314090.8750 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 220306.8438 - 743us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 196802.7656 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 187089.8906 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 557301.6250 - 758us/epoch - 12us/sample\n", - "Generation 69/300\n", - "Solving for Nash Equilibrium in Generation 69/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 365313.6250 - 762us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504725.4688 - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 140035.5000 - 695us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 839608.6875 - 1ms/epoch - 20us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 788710.5625 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 161677.0938 - 726us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 125521.1172 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 283217.8750 - 1ms/epoch - 17us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353215.0625 - 667us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 143640.0938 - 710us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 123928.0234 - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 180611.9062 - 1ms/epoch - 21us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240372.7031 - 770us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344495.8438 - 684us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 190148.7344 - 742us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1425373.3750 - 615us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 144902.8125 - 664us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1184865.7500 - 747us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325979.7500 - 1ms/epoch - 21us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 203582.3438 - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 145619.2812 - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482968.0625 - 878us/epoch - 14us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301197.3750 - 5ms/epoch - 86us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 518704.6875 - 793us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 850082.6250 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 163959.2188 - 806us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 291430.7812 - 757us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489224.4375 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 115375.3359 - 1ms/epoch - 17us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42325.8047 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 115359.7422 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294059.0938 - 998us/epoch - 16us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25693.9023 - 841us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 143429.4062 - 920us/epoch - 15us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387266.3750 - 815us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 153055.3438 - 1ms/epoch - 19us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604538.3750 - 2ms/epoch - 26us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261074.9688 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 374369.0938 - 817us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350877.5625 - 939us/epoch - 15us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2078110.7500 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240633.3750 - 1ms/epoch - 23us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 74494.8125 - 839us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244014.8281 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1626238.7500 - 880us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 156932.8438 - 2ms/epoch - 25us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 640572.3750 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288417.9375 - 1ms/epoch - 19us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 880263.7500 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112600.5859 - 836us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 45789.7891 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760183.9375 - 835us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 208345.4219 - 807us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338203.3438 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 487386.2188 - 966us/epoch - 16us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88922.9297 - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510785.7500 - 945us/epoch - 15us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1709885.8750 - 791us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 999824.3125 - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 79519.2812 - 6ms/epoch - 96us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592542.6875 - 752us/epoch - 12us/sample\n", - "Generation 70/300\n", - "Solving for Nash Equilibrium in Generation 70/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 949126.3125 - 1ms/epoch - 17us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513276.0625 - 1ms/epoch - 16us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629386.3125 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 212631.2812 - 1ms/epoch - 16us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8752.6318 - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787741.6250 - 894us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 886845.1250 - 830us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 133366.9219 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448862.9062 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 441805.4375 - 839us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 987us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371289.6562 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 167223.0469 - 965us/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500396.9062 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315221.5625 - 850us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 22530.3535 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317777.0312 - 858us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545944.7500 - 828us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338125.1875 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 461886.3125 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 749305.9375 - 1ms/epoch - 20us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 229886.5781 - 828us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556659.4375 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 373312.0938 - 881us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 256180.8281 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 206160.5469 - 899us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368724.2812 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258303.9688 - 876us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337163.4375 - 757us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 324687.6562 - 895us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1458278.3750 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 174200.9062 - 660us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 355891.2500 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 151842.1406 - 834us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439262.8750 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 159302.0000 - 6ms/epoch - 90us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 220590.3125 - 915us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673121.6875 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276725.5938 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 203144.9531 - 806us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264816.2188 - 1ms/epoch - 18us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727486.1875 - 780us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 212712.3906 - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769532.8750 - 967us/epoch - 16us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1380897.0000 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 64845.4688 - 864us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 474922.4375 - 853us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 173238.3594 - 987us/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 768061.3125 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2098843.7500 - 746us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126946.3672 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 311435.7188 - 983us/epoch - 16us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 335571.5625 - 808us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 622154.5625 - 809us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 544354.6875 - 725us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1581416.5000 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 154297.9844 - 792us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 175828.4844 - 753us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 171831.5781 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 86061.0781 - 1ms/epoch - 21us/sample\n", - "Generation 71/300\n", - "Solving for Nash Equilibrium in Generation 71/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295438.0625 - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2196770.5000 - 715us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 166009.3750 - 904us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1837016.8750 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 186080.0781 - 791us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376851.0000 - 852us/epoch - 14us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382234.2188 - 778us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1315239.7500 - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1940641.7500 - 810us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25805.7871 - 1ms/epoch - 18us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 595559.6875 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1329492.1250 - 799us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386593.3438 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277100.6875 - 1ms/epoch - 21us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1025526.3750 - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 136224.9844 - 822us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 128641.0703 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410308.3438 - 729us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325374.2188 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350375.9375 - 724us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 501465.3125 - 752us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 88295.3125 - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568882.5000 - 680us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 731608.3125 - 813us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1072536.0000 - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703387.6250 - 833us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 790911.1250 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 816049.3125 - 727us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 330447.4375 - 692us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 136us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 42886.5469 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386341.8750 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 333013.2188 - 673us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2541170.5000 - 633us/epoch - 10us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771535.0625 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 372311.0312 - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337739.3438 - 797us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 72011.4844 - 685us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724206.7500 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264267.1250 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357988.8750 - 672us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 584545.6875 - 715us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493136.9688 - 932us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 285104.8125 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 592399.1875 - 8ms/epoch - 131us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279106.7500 - 884us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 346963.9375 - 762us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1152958.3750 - 906us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1724180.7500 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 800922.1250 - 812us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506868.7812 - 767us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667630.8750 - 994us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 912726.6250 - 1ms/epoch - 18us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 728841.5000 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 73609.3516 - 866us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 255713.4688 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 67443.7656 - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634777.3750 - 841us/epoch - 14us/sample\n", - "Generation 72/300\n", - "Solving for Nash Equilibrium in Generation 72/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 872700.1875 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 672861.1250 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68463.0781 - 670us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240287.1250 - 766us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1349703.7500 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376445.3750 - 670us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602554.3750 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71938.7031 - 765us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259999.9531 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651139.9375 - 3ms/epoch - 50us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649879.5625 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 363046.0625 - 814us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 390256.9375 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561592.6875 - 759us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688386.0625 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 176848.3594 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 616653.6250 - 851us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 130766.4688 - 767us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1293721.5000 - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 430602.7188 - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479474.7188 - 1ms/epoch - 18us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 442978.2500 - 2ms/epoch - 28us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 165489.6875 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623513.3125 - 845us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 127232.7344 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669362.8125 - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1430392.3750 - 878us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 27002.1250 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 909194.8125 - 864us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231124.0000 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 26212.4082 - 1ms/epoch - 21us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 289397.6562 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760635.4375 - 679us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 611830.6250 - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 452915.6562 - 761us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 352238.9062 - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254920.3906 - 832us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419426.9688 - 911us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384192.9688 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1031539.2500 - 21ms/epoch - 331us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 375878.4688 - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 559788.8125 - 837us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378425.0000 - 1ms/epoch - 17us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2377585.5000 - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 664938.1875 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 85749.0312 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381475.6875 - 772us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 647710.0625 - 852us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302121.1250 - 941us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1038957.8750 - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 164911.0156 - 766us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437243.9062 - 775us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 948062.5625 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458655.2500 - 636us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 565705.7500 - 969us/epoch - 16us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438489.4375 - 2ms/epoch - 27us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2376377.7500 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1679480.7500 - 712us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302858.4375 - 684us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 189182.6719 - 871us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 216219.4219 - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656387.8750 - 884us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 40793.6680 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310566.1875 - 2ms/epoch - 34us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 150271.3906 - 773us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3790370.0000 - 763us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471254.8750 - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 959419.5625 - 717us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 703890.1875 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237701.0312 - 817us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Generation 73/300\n", - "Solving for Nash Equilibrium in Generation 73/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 426427.0938 - 989us/epoch - 16us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 776742.0000 - 752us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 21253.2207 - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 73393.6016 - 4ms/epoch - 58us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 415103.9375 - 728us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 867538.3125 - 902us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 237281.7812 - 887us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700684.1875 - 808us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438440.7500 - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414105.6250 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 195076.8750 - 896us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 762124.1875 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 75332.7656 - 880us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 343809.0312 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 142095.4531 - 756us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604777.7500 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 708447.8125 - 825us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300694.7188 - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412229.9375 - 802us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251601.2969 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 378686.9375 - 984us/epoch - 16us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248960.4219 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351619.4688 - 1ms/epoch - 24us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 424843.4062 - 898us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1096307.5000 - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 77796.3438 - 899us/epoch - 15us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1498340.6250 - 718us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 848656.7500 - 809us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736114.7500 - 747us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1249305.5000 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 93us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 588982.2500 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323811.0000 - 823us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773505.6875 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410864.8438 - 782us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408852.7812 - 660us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433085.4375 - 745us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 735957.2500 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2438295.7500 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 145303.8750 - 661us/epoch - 11us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1000756.2500 - 843us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 554915.6875 - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 57700.1055 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 86432.0781 - 985us/epoch - 16us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1263395.0000 - 762us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 83758.2891 - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 254502.0938 - 1ms/epoch - 18us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767977.1875 - 666us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 903469.9375 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 849618.1250 - 1ms/epoch - 17us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1476896.6250 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 449315.2500 - 673us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 93340.2500 - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2381541.0000 - 922us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 8ms/epoch - 121us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 476171.6562 - 1ms/epoch - 20us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 727774.3125 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508396.7812 - 1ms/epoch - 16us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1777093.8750 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 806363.8750 - 710us/epoch - 11us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 371014.6250 - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 135740.6406 - 832us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314834.0625 - 757us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 128270.4375 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 433222.8125 - 890us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1324508.0000 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 468434.9062 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 816605.1875 - 3ms/epoch - 41us/sample\n", - "Generation 74/300\n", - "Solving for Nash Equilibrium in Generation 74/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 160283.0938 - 835us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 625839.4375 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 480773.3438 - 683us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286465.6562 - 3ms/epoch - 43us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 827520.6875 - 763us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 224570.6406 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1314030.1250 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 752us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 435044.1875 - 601us/epoch - 10us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495011.1875 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 837795.2500 - 652us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724717.1875 - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1037990.6875 - 699us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 280205.4375 - 741us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298303.8750 - 849us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 30744.2637 - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663511.0625 - 3ms/epoch - 44us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 660178.6875 - 729us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583566.0000 - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 970726.8750 - 678us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 753994.0625 - 874us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 59892.9062 - 708us/epoch - 11us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 965290.1250 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620823.1875 - 851us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 272373.1875 - 964us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 49573.1133 - 1ms/epoch - 16us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482835.9062 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1176069.5000 - 732us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 892507.2500 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 628142.6250 - 836us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 977466.4375 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 278585.4375 - 1ms/epoch - 17us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 268446.9375 - 764us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1077020.6250 - 1ms/epoch - 18us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1259499.2500 - 977us/epoch - 16us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 892088.6875 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 67us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 634709.6875 - 934us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1431391.3750 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347264.8750 - 716us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 862338.6250 - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682752.5625 - 6ms/epoch - 96us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 877941.6250 - 52ms/epoch - 846us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784926.0000 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 85642.7734 - 768us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2696163.0000 - 947us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377497.1562 - 795us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312113.7188 - 722us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359817.8438 - 939us/epoch - 15us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1728405.0000 - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 287213.8125 - 654us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 451949.1250 - 2ms/epoch - 31us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3633113.5000 - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 274500.4688 - 745us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1237520.2500 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276724.4375 - 825us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 878529.1875 - 2ms/epoch - 28us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 410808.3125 - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 393640.8750 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295093.4688 - 810us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238603.1719 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 357556.8750 - 744us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 158037.3750 - 1ms/epoch - 21us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464056.9688 - 766us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 297330.5625 - 974us/epoch - 16us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301011.9062 - 889us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Generation 75/300\n", - "Solving for Nash Equilibrium in Generation 75/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 188736.7656 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618824.3125 - 822us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 94184.1719 - 1ms/epoch - 20us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1042887.9375 - 936us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1597027.2500 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 832992.1250 - 1ms/epoch - 20us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 92081.3984 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 88us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2222471.2500 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 823757.0625 - 789us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 161500.6562 - 918us/epoch - 15us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2270393.2500 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116015.6016 - 922us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1459217.6250 - 3ms/epoch - 47us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1187044.2500 - 62ms/epoch - 1ms/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 257128.8906 - 865us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 502877.7188 - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 224889.5469 - 737us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305613.3438 - 950us/epoch - 15us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 827693.6250 - 4ms/epoch - 59us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538980.3125 - 685us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488703.3750 - 912us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106221.2656 - 810us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 294332.7812 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232057.5156 - 705us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 817445.8750 - 751us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1097902.1250 - 667us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1680870.5000 - 816us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308870.9375 - 665us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1015099.8750 - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 818039.8125 - 797us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484706.3438 - 780us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 202532.6250 - 833us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2476748.5000 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 475348.9688 - 704us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1038875.8750 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 478319.8750 - 832us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240308.3906 - 1ms/epoch - 21us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 709us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 279621.6250 - 677us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 101483.0469 - 998us/epoch - 16us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126716.3672 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 197389.7188 - 714us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370778.0000 - 646us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1586611.6250 - 805us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 120us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 127870.9375 - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 816089.8125 - 1ms/epoch - 17us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1740538.1250 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 852852.5625 - 831us/epoch - 13us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 229915.7031 - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 131950.7188 - 835us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2872521.5000 - 914us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3440472.5000 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 796668.4375 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270371.4688 - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2142120.2500 - 765us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 861812.4375 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377210.9062 - 647us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 93683.0078 - 1ms/epoch - 21us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 556244.6250 - 691us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 181184.8438 - 786us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1239192.0000 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 846407.8125 - 787us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4037279.2500 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 823683.3125 - 732us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1247819.1250 - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2144874.5000 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 197823.5000 - 726us/epoch - 12us/sample\n", - "Generation 76/300\n", - "Solving for Nash Equilibrium in Generation 76/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307275.0938 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230995.4688 - 2ms/epoch - 32us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408190.0625 - 1ms/epoch - 21us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 233079.0156 - 779us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 203165.2188 - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1256903.8750 - 1ms/epoch - 17us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 259014.8438 - 732us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 369569.7500 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386423.1562 - 838us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1101048.0000 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 173127.5781 - 730us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 831758.8750 - 1ms/epoch - 17us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 771447.7500 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1230856.8750 - 778us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 328280.5938 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326192.4062 - 858us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2905254.2500 - 732us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1036909.8750 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462549.7188 - 897us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 215019.9062 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 315019.5938 - 727us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 238474.1094 - 796us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4020206.0000 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 644763.6250 - 734us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 917602.7500 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1138492.2500 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 562686.3750 - 715us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503198.0938 - 831us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353642.6875 - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1433276.3750 - 1ms/epoch - 16us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 344827.7188 - 690us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 396204.8125 - 709us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1709202.5000 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 58402.7578 - 4ms/epoch - 60us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568797.6250 - 951us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1066635.2500 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126920.0703 - 1ms/epoch - 21us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3342080.7500 - 4ms/epoch - 63us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 800365.2500 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 391127.0000 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 326353.9375 - 785us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 178848.3281 - 9ms/epoch - 151us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126200.9531 - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 251798.3281 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 508125.3438 - 4ms/epoch - 57us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 861922.3125 - 2ms/epoch - 26us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505407.3750 - 716us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 826264.5000 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397095.8438 - 754us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 993080.1875 - 824us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 888602.8750 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784486.1875 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 100634.5391 - 1ms/epoch - 18us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 142139.1875 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 248300.4062 - 1ms/epoch - 19us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4074158.7500 - 744us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2449224.0000 - 831us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 821586.1250 - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1070140.8750 - 837us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1987481.5000 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 525279.8750 - 879us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322415.8125 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 194065.7812 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 939728.0625 - 672us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104086.0234 - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 437132.0625 - 4ms/epoch - 64us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338708.2812 - 919us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411870.1875 - 829us/epoch - 13us/sample\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", "Episode 50/50\n", - "Generation 77/300\n", - "Solving for Nash Equilibrium in Generation 77/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 385692.0312 - 10ms/epoch - 154us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533815.9375 - 796us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 408248.8125 - 811us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270502.4688 - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519246.6562 - 772us/epoch - 12us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342579.5938 - 2ms/epoch - 26us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 847406.6875 - 941us/epoch - 15us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1630794.7500 - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 901055.8125 - 713us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 106823.5000 - 822us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2134298.5000 - 698us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2019227.1250 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 174603.5625 - 685us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 950406.2500 - 898us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 955500.5625 - 835us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1361871.8750 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602615.3125 - 723us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 548286.5625 - 916us/epoch - 15us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3345914.0000 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630635.2500 - 1ms/epoch - 20us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2083315.3750 - 844us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 60637.1719 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606492.9375 - 920us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 68226.3594 - 1ms/epoch - 17us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414679.2500 - 981us/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1775708.2500 - 657us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 937085.6250 - 942us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1398471.1250 - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 380880.3438 - 717us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 763415.2500 - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 549197.5625 - 1ms/epoch - 24us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2276496.0000 - 681us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 472212.7500 - 636us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 552653.6250 - 995us/epoch - 16us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420581.0938 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 354074.7812 - 784us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 356654.9688 - 998us/epoch - 16us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 332357.0312 - 730us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 939258.5000 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 124629.0781 - 941us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558143.5625 - 876us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 191415.3906 - 715us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 296988.0625 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386994.7812 - 923us/epoch - 15us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 989688.6250 - 785us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6105605.0000 - 1ms/epoch - 16us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 980744.0625 - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3198150.0000 - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1448656.7500 - 847us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736538.1250 - 1ms/epoch - 18us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1469542.3750 - 933us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455889.2188 - 825us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 314820.2812 - 656us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 950032.3125 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1443686.6250 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 193630.5938 - 845us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 575868.1250 - 739us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 112309.6797 - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1010695.8125 - 1ms/epoch - 18us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 367971.9688 - 1ms/epoch - 19us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1472933.1250 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638630.4375 - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 368303.6250 - 2ms/epoch - 26us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545871.1250 - 796us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469466.4375 - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 57us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 957388.8750 - 1ms/epoch - 19us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1565286.8750 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1079960.5000 - 780us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 729156.8750 - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 406608.4688 - 949us/epoch - 15us/sample\n", - "Generation 78/300\n", - "Solving for Nash Equilibrium in Generation 78/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787438.0625 - 949us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414074.5312 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1338064.6250 - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1485125.2500 - 871us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555034.7500 - 932us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 266886.3750 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 327540.5938 - 825us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 113181.2734 - 824us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 114351.9844 - 928us/epoch - 15us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3920524.2500 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 678248.8125 - 833us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3366972.5000 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 997us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321598.9062 - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1170910.5000 - 755us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1054104.7500 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306922.7188 - 846us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 797181.7500 - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 212712.0469 - 848us/epoch - 14us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1694738.7500 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412120.3438 - 823us/epoch - 13us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646273.3750 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 666640.7500 - 939us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 600808.3750 - 716us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1112935.8750 - 675us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398366.3125 - 685us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1061803.1250 - 670us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 688284.3125 - 910us/epoch - 15us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675018.3125 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 992620.3750 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 620091.6250 - 722us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2338641.0000 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446955.0312 - 2ms/epoch - 35us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 381164.3438 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 435912.5625 - 646us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 116827.2578 - 728us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 542051.8750 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 117900.7188 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 571067.1875 - 1ms/epoch - 20us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2385748.2500 - 746us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 295573.0938 - 741us/epoch - 12us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 485729.2812 - 925us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1323507.7500 - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 966885.8125 - 758us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4001003.2500 - 714us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 242864.8125 - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 629902.6875 - 753us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1139632.2500 - 734us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700629.0625 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2672603.2500 - 2ms/epoch - 26us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387099.7188 - 672us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1062048.2500 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 681758.5625 - 771us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1091744.0000 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 202889.4219 - 803us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1480967.1250 - 842us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521412.6875 - 692us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1218802.1250 - 806us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1695325.3750 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 436680.8750 - 891us/epoch - 14us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405605.0312 - 788us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 702346.5000 - 714us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 15ms/epoch - 237us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1045147.6250 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1153001.2500 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 302361.1250 - 904us/epoch - 15us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585885.0625 - 801us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1982321.2500 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 448059.5938 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532371.2500 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 467125.6875 - 705us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 397418.3438 - 767us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 304043.8438 - 1ms/epoch - 16us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1232276.2500 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 417478.3438 - 762us/epoch - 12us/sample\n", - "Generation 79/300\n", - "Solving for Nash Equilibrium in Generation 79/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308259.3125 - 725us/epoch - 12us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 826956.7500 - 562us/epoch - 9us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3264664.2500 - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1476508.5000 - 766us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 847169.1250 - 681us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 769232.7500 - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 870452.1875 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 401659.9062 - 641us/epoch - 10us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240344.4062 - 2ms/epoch - 28us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604364.4375 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1748470.7500 - 836us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 228014.9531 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1142058.8750 - 2ms/epoch - 32us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1292014.5000 - 1ms/epoch - 23us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1088950.2500 - 813us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458869.6875 - 856us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2220035.5000 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725557.8750 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2373583.0000 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 126684.6094 - 732us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1196028.7500 - 804us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545431.5000 - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1416082.5000 - 880us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1286366.1250 - 929us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 25573.7773 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1082702.1250 - 781us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 109913.5703 - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1389089.8750 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 162us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 662794.0625 - 3ms/epoch - 42us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1668589.6250 - 836us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671729.1875 - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258312.6406 - 947us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 150904.5312 - 801us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 201611.0469 - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719077.3750 - 834us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1710040.5000 - 781us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 376061.1562 - 1ms/epoch - 20us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658328.7500 - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1358960.5000 - 1ms/epoch - 17us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 135554.9219 - 789us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1254218.7500 - 908us/epoch - 15us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 231537.3438 - 856us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 361819.6562 - 959us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500825.2500 - 1ms/epoch - 16us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 497814.0938 - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1495707.0000 - 797us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450923.1875 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 469296.3438 - 1ms/epoch - 17us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2208971.7500 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 579377.3750 - 1ms/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1169584.2500 - 970us/epoch - 16us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6413393.5000 - 757us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1186338.2500 - 716us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425411.1875 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1145174.5000 - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 337519.0000 - 894us/epoch - 14us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1184001.8750 - 772us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3388257.2500 - 870us/epoch - 14us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2546566.2500 - 725us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 809759.5625 - 643us/epoch - 10us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 211887.0781 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 954us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1436015.7500 - 630us/epoch - 10us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758526.1875 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490493.0625 - 728us/epoch - 12us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1378938.3750 - 762us/epoch - 12us/sample\n", "Generation 80/300\n", "Solving for Nash Equilibrium in Generation 80/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 569690.9375 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 956330.2500 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 307046.5938 - 846us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 130526.2266 - 3ms/epoch - 56us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1201081.7500 - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656252.8750 - 874us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 692398.0000 - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1212450.3750 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 952us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5165210.0000 - 787us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 348382.0938 - 644us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1209924.7500 - 681us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 383160.9375 - 2ms/epoch - 27us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2303084.2500 - 704us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760454.7500 - 766us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737600.8750 - 692us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 901102.5625 - 728us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 78441.4922 - 688us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 35ms/epoch - 557us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1655315.0000 - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 411767.6875 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3705010.7500 - 5ms/epoch - 79us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456257.5312 - 755us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 136101.3594 - 780us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1777839.8750 - 716us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1841218.8750 - 868us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1098457.6250 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1372946.6250 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269216.2188 - 1ms/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1222391.1250 - 725us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 292048.0000 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1223637.2500 - 1ms/epoch - 18us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 19559.4746 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1377591.3750 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232313.1875 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532968.8750 - 795us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 179330.3438 - 1ms/epoch - 20us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 812193.5000 - 914us/epoch - 15us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759864.7500 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 704655.8750 - 1ms/epoch - 17us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1487845.0000 - 979us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492674.9688 - 724us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3466198.7500 - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 48322.3086 - 1ms/epoch - 20us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1202611.0000 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462094.0312 - 1ms/epoch - 17us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1355707.6250 - 1ms/epoch - 16us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 132308.0781 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 533594.8750 - 930us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5784931.0000 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 796707.4375 - 838us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5633699.5000 - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530616.7500 - 754us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1856942.1250 - 745us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1913973.6250 - 1ms/epoch - 21us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 940us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 946422.8750 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276378.8438 - 739us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1331390.2500 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1861867.6250 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414486.7812 - 686us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1532397.5000 - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504085.5625 - 868us/epoch - 14us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 877264.3750 - 826us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 230878.4219 - 881us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 350484.8438 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2116978.5000 - 1ms/epoch - 18us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618200.4375 - 863us/epoch - 14us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 81/300\n", "Solving for Nash Equilibrium in Generation 81/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1735380.7500 - 824us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1320605.8750 - 726us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1109002.7500 - 655us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1665423.0000 - 693us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 358164.3125 - 850us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 199904.0781 - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1262844.6250 - 806us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1403875.0000 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523733.2812 - 788us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 496339.7812 - 729us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1040305.3125 - 4ms/epoch - 72us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2310898.2500 - 713us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 500869.5312 - 688us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2802808.2500 - 778us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3034301.7500 - 2ms/epoch - 30us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3701363.5000 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450602.7812 - 815us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1604303.2500 - 779us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1477092.3750 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1328828.6250 - 861us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1371979.3750 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1392133.7500 - 812us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 43043.1016 - 663us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2148740.7500 - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384274.6875 - 1ms/epoch - 16us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 483819.9375 - 743us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545919.2500 - 734us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 637026.6250 - 959us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2725873.2500 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 353370.0938 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 750684.7500 - 742us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1348980.6250 - 798us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4523305.5000 - 752us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418831.2188 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 617713.3125 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 755210.5625 - 746us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3006433.2500 - 847us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2006144.8750 - 750us/epoch - 12us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 843236.8750 - 875us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1352844.6250 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668429.6875 - 752us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 276780.5625 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 495754.4688 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726326.7500 - 622us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504905.9375 - 674us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623384.1250 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1388327.1250 - 655us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3380832.5000 - 825us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 243787.0781 - 1ms/epoch - 21us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 537570.4375 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 816624.2500 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 494377.1562 - 3ms/epoch - 49us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1898549.3750 - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 273544.2812 - 749us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 851745.7500 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1830069.5000 - 723us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1028954.1250 - 777us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 864998.4375 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2059219.7500 - 748us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1419743.0000 - 700us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532482.1250 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 583090.5625 - 664us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 656823.9375 - 779us/epoch - 13us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 82/300\n", "Solving for Nash Equilibrium in Generation 82/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2757958.0000 - 700us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 182138.8594 - 689us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606416.0000 - 682us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3703696.5000 - 631us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 610058.8125 - 953us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1644251.5000 - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545792.3750 - 785us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 87941.5000 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522309.7500 - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1480491.2500 - 833us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 944us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506585.2188 - 819us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 420585.5625 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3033489.5000 - 669us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1461862.3750 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1072343.0000 - 2ms/epoch - 24us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 555412.8750 - 876us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 227701.0625 - 3ms/epoch - 53us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649320.7500 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 962us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1735931.1250 - 810us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 109us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 638108.0000 - 2ms/epoch - 29us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5222414.0000 - 940us/epoch - 15us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2114003.2500 - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 463853.7188 - 809us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1746389.1250 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456332.0312 - 763us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1475031.6250 - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 852132.7500 - 2ms/epoch - 38us/sample\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7284080.5000 - 2ms/epoch - 31us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1491515.2500 - 675us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 670754.0625 - 1ms/epoch - 17us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700246.9375 - 4ms/epoch - 59us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 387915.7500 - 720us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1291028.8750 - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 909us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 505466.6875 - 777us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1379412.1250 - 695us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235160.5625 - 869us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6496600.5000 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 264011.2500 - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 962835.1250 - 2ms/epoch - 30us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1516370.8750 - 955us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 927385.5625 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 852320.3750 - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 511634.0000 - 903us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 122102.8906 - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1579527.0000 - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 535856.7500 - 702us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 156698.2969 - 858us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 932074.1875 - 1ms/epoch - 20us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 539870.3750 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 819517.3750 - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 941378.8750 - 787us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 258189.5312 - 810us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522859.9688 - 742us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1888395.3750 - 1ms/epoch - 21us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 689572.7500 - 740us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 586646.8750 - 1ms/epoch - 22us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7022400.5000 - 688us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482520.2500 - 942us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 83/300\n", "Solving for Nash Equilibrium in Generation 83/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651293.0625 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1078367.3750 - 2ms/epoch - 25us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1244304.5000 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 157424.6562 - 798us/epoch - 13us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1583686.2500 - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2220445.0000 - 1ms/epoch - 20us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290136.0000 - 3ms/epoch - 45us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 932628.9375 - 821us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1026789.3125 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2225381.7500 - 650us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 265481.7500 - 735us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316761.4062 - 826us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 67533.4062 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 936441.4375 - 859us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 402731.7188 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 712907.6250 - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1815943.5000 - 1ms/epoch - 16us/sample\n", - "Episode 12/50\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5190492.5000 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 293217.6875 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 513421.1250 - 724us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 530900.5000 - 991us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1563698.8750 - 974us/epoch - 16us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 439183.0938 - 1ms/epoch - 23us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3385983.7500 - 786us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675639.5000 - 1ms/epoch - 20us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 175379.9688 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1205659.7500 - 1ms/epoch - 17us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1609050.3750 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428990.3438 - 932us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 160198.5000 - 725us/epoch - 12us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 386420.4375 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1796963.3750 - 3ms/epoch - 47us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2002581.8750 - 747us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1623568.6250 - 782us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1817460.1250 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 308033.7812 - 643us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 621013.5000 - 848us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 209679.6875 - 771us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 191194.2969 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1124238.7500 - 759us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 507784.1250 - 753us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1161462.7500 - 882us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 232328.8438 - 638us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5220921.0000 - 682us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349532.8125 - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3900957.0000 - 728us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 538324.5000 - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 558654.0000 - 809us/epoch - 13us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2152439.0000 - 676us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 986026.8125 - 961us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3997164.5000 - 725us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 676337.9375 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1653387.2500 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 105us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 669874.5000 - 5ms/epoch - 73us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2723165.5000 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 677776.0000 - 1ms/epoch - 17us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 848870.9375 - 856us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 34005.9805 - 910us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 668225.6875 - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2059500.8750 - 850us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 976us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 615596.1875 - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 984us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1076253.8750 - 1ms/epoch - 21us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 84/300\n", "Solving for Nash Equilibrium in Generation 84/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 864917.5625 - 897us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4508818.0000 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2872692.7500 - 781us/epoch - 13us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1643596.5000 - 888us/epoch - 14us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 222909.5469 - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1650547.7500 - 974us/epoch - 16us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1519788.8750 - 864us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1601702.7500 - 771us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2495890.2500 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1609832.5000 - 1ms/epoch - 17us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 713299.8125 - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1066224.3750 - 696us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 938529.9375 - 963us/epoch - 16us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3301903.2500 - 2ms/epoch - 34us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1748244.5000 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 635206.6875 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1438060.6250 - 856us/epoch - 14us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2384957.5000 - 724us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 950176.1250 - 4ms/epoch - 57us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3445375.2500 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 787040.8750 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5234836.5000 - 772us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1046220.3750 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 316486.7500 - 1ms/epoch - 18us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 349743.5312 - 715us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 28233.9727 - 932us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 282536.5625 - 838us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1803847.3750 - 671us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2260663.0000 - 711us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1180829.1250 - 810us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4018825.2500 - 999us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7768191.5000 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1140342.7500 - 2ms/epoch - 26us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1670266.6250 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 887670.1250 - 854us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1669375.0000 - 834us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5539494.5000 - 792us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 319575.7500 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2438549.0000 - 980us/epoch - 16us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 418716.2188 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1727358.6250 - 3ms/epoch - 51us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2451135.0000 - 836us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440949.3125 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 288341.2500 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1737928.2500 - 831us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 170555.1406 - 877us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2307621.5000 - 1ms/epoch - 23us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1582403.7500 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 631188.0625 - 740us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5019664.5000 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 964us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462121.9062 - 1ms/epoch - 17us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1743716.3750 - 931us/epoch - 15us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 922092.7500 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1261163.8750 - 2ms/epoch - 39us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 651968.2500 - 749us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 986us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2216343.0000 - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 90us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 446543.3125 - 1ms/epoch - 22us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 286088.7500 - 829us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 431866.9688 - 2ms/epoch - 24us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2263064.5000 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 479938.8125 - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1891313.6250 - 1ms/epoch - 22us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2522769.7500 - 827us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 460417.9688 - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 602455.6875 - 842us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512462.4375 - 905us/epoch - 15us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3754142.7500 - 901us/epoch - 15us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 85/300\n", "Solving for Nash Equilibrium in Generation 85/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 493487.0938 - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2138611.7500 - 807us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2848364.2500 - 990us/epoch - 16us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 523140.2812 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 604403.6250 - 2ms/epoch - 29us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 176914.0000 - 721us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1810465.5000 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 764010.6250 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 609621.7500 - 670us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1132749.7500 - 656us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1609025.8750 - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 63us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 419011.6562 - 840us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1770015.0000 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3341657.7500 - 878us/epoch - 14us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 172715.3906 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 926us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2575042.5000 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 737231.3750 - 1ms/epoch - 17us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5285649.5000 - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2247407.0000 - 755us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1824961.8750 - 848us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 982182.2500 - 700us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1027406.9375 - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2819090.7500 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1653150.5000 - 786us/epoch - 13us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3203701.7500 - 1ms/epoch - 21us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1258355.6250 - 692us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5238938.0000 - 658us/epoch - 11us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2716979.0000 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 331743.0000 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 671363.7500 - 831us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4567937.0000 - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 942113.1875 - 754us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 298111.8125 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1148406.6250 - 706us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3302907.2500 - 5ms/epoch - 73us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 783609.8750 - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 299911.8438 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 626116.6250 - 870us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 649299.0625 - 666us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 719483.6250 - 905us/epoch - 15us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3033806.7500 - 874us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 14258269.0000 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 245628.0000 - 816us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 271386.0938 - 837us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5640271.5000 - 779us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2918104.2500 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1668802.5000 - 955us/epoch - 15us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 951us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1884384.7500 - 637us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 456714.0312 - 768us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1365488.3750 - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1178160.1250 - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 965us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 184623.4844 - 899us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 336857.4688 - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 646607.5625 - 707us/epoch - 11us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4076676.7500 - 893us/epoch - 14us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 800709.2500 - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1069122.2500 - 2ms/epoch - 29us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1281545.2500 - 725us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630760.1250 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1287134.5000 - 823us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312006.9688 - 730us/epoch - 12us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696931.4375 - 2ms/epoch - 30us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 828758.0625 - 784us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 841us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2716276.2500 - 743us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 382898.2188 - 812us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 693424.7500 - 1ms/epoch - 17us/sample\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", "Generation 86/300\n", "Solving for Nash Equilibrium in Generation 86/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 414685.7188 - 813us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2695846.5000 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1088470.5000 - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 338995.7188 - 756us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1860278.1250 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 78us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340339.3125 - 927us/epoch - 15us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 706364.6875 - 928us/epoch - 15us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1867926.8750 - 939us/epoch - 15us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1061807.3750 - 774us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 762us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 607268.0625 - 664us/epoch - 11us/sample\n", - "Episode 10/50\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13084313.0000 - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 889355.5000 - 732us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1335940.1250 - 704us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 691930.7500 - 3ms/epoch - 54us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2766153.7500 - 757us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1996407.0000 - 667us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2888614.0000 - 2ms/epoch - 25us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1324954.2500 - 799us/epoch - 13us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 499120.8750 - 826us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3054408.5000 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 696524.7500 - 671us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8567835.0000 - 679us/epoch - 11us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 189305.5469 - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 725us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2811923.5000 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 752858.0000 - 653us/epoch - 11us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 340972.0312 - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 190498.7031 - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2829078.7500 - 875us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1972636.7500 - 715us/epoch - 12us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1975684.8750 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1981813.3750 - 769us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1760472.7500 - 727us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1990163.7500 - 671us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1226893.7500 - 943us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 531122.6875 - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684772.9375 - 1ms/epoch - 17us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 191980.1875 - 782us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3993468.7500 - 766us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 783415.1875 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 482212.4688 - 663us/epoch - 11us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1337935.6250 - 815us/epoch - 13us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2140636.5000 - 690us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 623098.7500 - 715us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606379.7500 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3434807.2500 - 714us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6774180.0000 - 704us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 702us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 322090.3438 - 737us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4590045.0000 - 801us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2662451.0000 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 665us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2601991.7500 - 649us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 306064.1875 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2025197.8750 - 668us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1121956.3750 - 799us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 488251.8438 - 675us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 864089.1875 - 941us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700120.0000 - 649us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1021294.7500 - 747us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 217279.7656 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 924us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 699600.2500 - 2ms/epoch - 26us/sample\n", - "Episode 47/50\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3778504.2500 - 837us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7146391.0000 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 758326.1250 - 2ms/epoch - 35us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 977us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5213141.0000 - 2ms/epoch - 29us/sample\n", - "Generation 87/300\n", - "Solving for Nash Equilibrium in Generation 87/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", "Episode 2/50\n", "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1974151.0000 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1292586.7500 - 892us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2063124.6250 - 663us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2268501.2500 - 15ms/epoch - 244us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 930269.0625 - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 725276.5000 - 751us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 458697.4688 - 1ms/epoch - 17us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1191375.8750 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2611873.0000 - 808us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1160312.2500 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 650163.0625 - 769us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 933509.2500 - 3ms/epoch - 41us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2019198.2500 - 739us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 996us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1289073.6250 - 1ms/epoch - 16us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1062608.6250 - 837us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 359778.2500 - 666us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1976174.3750 - 698us/epoch - 11us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 64738.8711 - 659us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1213372.6250 - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1111796.1250 - 688us/epoch - 11us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2694455.0000 - 737us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 325353.2188 - 886us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2033970.5000 - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779161.1875 - 2ms/epoch - 38us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6357619.0000 - 620us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1216329.0000 - 627us/epoch - 10us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3122778.5000 - 865us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1068523.8750 - 696us/epoch - 11us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3565270.7500 - 746us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4251239.0000 - 743us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1088658.8750 - 733us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 916282.8750 - 695us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1515918.2500 - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1235666.8750 - 711us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1074188.6250 - 2ms/epoch - 34us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2610926.5000 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1080903.2500 - 804us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4714422.0000 - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1899338.5000 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 244963.7031 - 2ms/epoch - 37us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1192808.5000 - 978us/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 440750.9062 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 974us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7314650.5000 - 760us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2720517.2500 - 717us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3727869.5000 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392893.0312 - 804us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 773671.3750 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1205906.3750 - 677us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 453156.7188 - 781us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 512142.7812 - 700us/epoch - 11us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1280491.1250 - 1ms/epoch - 20us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1206268.7500 - 2ms/epoch - 30us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1864850.2500 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2434852.2500 - 904us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2050203.3750 - 685us/epoch - 11us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 464475.3125 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1227106.1250 - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1411494.3750 - 702us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8193980.0000 - 721us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 41us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1268522.3750 - 2ms/epoch - 35us/sample\n", - "Generation 88/300\n", - "Solving for Nash Equilibrium in Generation 88/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 104245.7188 - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4327070.5000 - 784us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 830958.3750 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1247925.0000 - 5ms/epoch - 77us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2871979.5000 - 719us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675394.8125 - 731us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 553019.6875 - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2350787.5000 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2221347.0000 - 695us/epoch - 11us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 465071.8125 - 653us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1471637.0000 - 628us/epoch - 10us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2414423.2500 - 865us/epoch - 14us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 157732.3906 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2008707.3750 - 594us/epoch - 10us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2394592.0000 - 672us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 370816.7188 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1116711.0000 - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3175738.0000 - 707us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 810796.1875 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 840785.3750 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2049884.2500 - 880us/epoch - 14us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2133013.0000 - 806us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2321691.5000 - 789us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 675233.3125 - 705us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 751us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1689706.7500 - 6ms/epoch - 96us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6215913.5000 - 795us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 394102.1875 - 861us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5057317.0000 - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3761088.2500 - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 808278.6250 - 886us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1139342.7500 - 709us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736215.0625 - 3ms/epoch - 53us/sample\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 779395.1250 - 725us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1554735.8750 - 1ms/epoch - 23us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8563183.0000 - 659us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1302523.6250 - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5843632.0000 - 798us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301434.3438 - 747us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2213045.0000 - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2076698.3750 - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 961us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1291914.8750 - 1ms/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 950us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3240025.2500 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 791102.5000 - 3ms/epoch - 49us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 323767.1250 - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3254080.5000 - 680us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 217948.3594 - 792us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3561611.7500 - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3270531.7500 - 788us/epoch - 13us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3078999.0000 - 851us/epoch - 14us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1372118.8750 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 42us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3277045.2500 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 948us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3885772.7500 - 865us/epoch - 14us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2205872.5000 - 711us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 667412.5000 - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 543132.7500 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 46us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2199067.5000 - 2ms/epoch - 27us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 484712.8125 - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 821996.6875 - 904us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2167064.2500 - 746us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4877661.5000 - 1ms/epoch - 19us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 521968.5938 - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5569487.5000 - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 683968.0625 - 852us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 384716.2500 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 10ms/epoch - 161us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2396452.2500 - 8ms/epoch - 123us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1190845.3750 - 781us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 125738.2188 - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2217639.5000 - 1ms/epoch - 22us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1558743.8750 - 737us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2409067.0000 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606615.8125 - 696us/epoch - 11us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 714309.5000 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 503044.2500 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 398583.8750 - 919us/epoch - 15us/sample\n", - "Generation 89/300\n", - "Solving for Nash Equilibrium in Generation 89/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 275893.9062 - 843us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7876454.0000 - 698us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2098855.0000 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 545005.1875 - 843us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 959us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 742954.4375 - 724us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2245720.0000 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 833718.0000 - 5ms/epoch - 74us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 832035.3125 - 874us/epoch - 14us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 389381.0625 - 716us/epoch - 12us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 580661.3750 - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1996826.3750 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1599279.6250 - 703us/epoch - 11us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 958624.5000 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2231700.2500 - 1ms/epoch - 16us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1053897.7500 - 795us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3078223.2500 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4592649.0000 - 779us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 158436.0625 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5477830.5000 - 725us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 392252.9375 - 662us/epoch - 11us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1584116.8750 - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2689299.7500 - 714us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 226672.6562 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 843048.7500 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3506673.0000 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1234934.3750 - 688us/epoch - 11us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 227924.5312 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1358239.5000 - 3ms/epoch - 43us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2896337.5000 - 677us/epoch - 11us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 922095.4375 - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8249751.5000 - 710us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 726546.2500 - 983us/epoch - 16us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1712559.6250 - 891us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425324.4688 - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 958us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5121426.0000 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2277771.7500 - 2ms/epoch - 24us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270581.1875 - 678us/epoch - 11us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4233912.5000 - 645us/epoch - 10us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2528148.2500 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 269980.2188 - 849us/epoch - 14us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5129347.5000 - 751us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 853943.6250 - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1876580.5000 - 1ms/epoch - 16us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 736816.6250 - 820us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 820us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4228341.5000 - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1662922.2500 - 841us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4532341.5000 - 2ms/epoch - 30us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 929us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1274614.7500 - 750us/epoch - 12us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 873us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1253690.2500 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1002026.5000 - 955us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 534923.9375 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2873272.2500 - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 928us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2376051.2500 - 669us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 566517.0625 - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 923us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462892.5625 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2591409.5000 - 651us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2387509.5000 - 706us/epoch - 11us/sample\n", - "Episode 39/50\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 968764.1250 - 745us/epoch - 12us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2803202.0000 - 712us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 61us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9318936.0000 - 4ms/epoch - 67us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409213.0000 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3685015.0000 - 764us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 246946.8750 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1271287.5000 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2424260.2500 - 596us/epoch - 10us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1457590.1250 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2427344.2500 - 727us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3724707.5000 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 859us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 878591.6875 - 692us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3029883.0000 - 1ms/epoch - 16us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1537899.3750 - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 775us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2435339.5000 - 637us/epoch - 10us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2481170.2500 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 830us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 863312.3125 - 749us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 33ms/epoch - 538us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 234444.8594 - 2ms/epoch - 25us/sample\n", - "Generation 90/300\n", - "Solving for Nash Equilibrium in Generation 90/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1400194.8750 - 812us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2638203.2500 - 759us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 347703.1875 - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 938us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3068314.0000 - 805us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 931us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 618566.2500 - 717us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2836486.5000 - 776us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1737663.0000 - 782us/epoch - 13us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1347855.3750 - 752us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 351008.8750 - 904us/epoch - 15us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5052177.5000 - 918us/epoch - 15us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2508613.0000 - 1ms/epoch - 17us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4557486.5000 - 773us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 923106.2500 - 886us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 969us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 759824.0000 - 1ms/epoch - 21us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 890us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 819773.2500 - 744us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4515752.5000 - 928us/epoch - 15us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1463844.7500 - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 898us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3126569.7500 - 708us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2271678.5000 - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2447741.2500 - 838us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8872943.0000 - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2756271.0000 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 240040.4688 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 878us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2007984.3750 - 837us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1431064.1250 - 805us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 69us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1345315.6250 - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2650136.5000 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3557210.5000 - 934us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 897us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 986495.5000 - 962us/epoch - 16us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 945236.4375 - 875us/epoch - 14us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1734009.6250 - 757us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 963us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5071383.0000 - 925us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 930us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1490209.3750 - 966us/epoch - 16us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1274082.1250 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 985us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 301775.7500 - 976us/epoch - 16us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2779830.5000 - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 928061.9375 - 980us/epoch - 16us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1398688.1250 - 862us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 912827.5000 - 812us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1470449.2500 - 882us/epoch - 14us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10782498.0000 - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1847346.3750 - 933us/epoch - 15us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 700154.1250 - 844us/epoch - 14us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2622018.7500 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 802us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1346075.1250 - 801us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1520400.3750 - 647us/epoch - 10us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2824936.2500 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 951237.2500 - 832us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 438248.1875 - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 79us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 907513.5000 - 2ms/epoch - 38us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2124561.5000 - 915us/epoch - 15us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3845203.5000 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 510080.6250 - 762us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2005118.2500 - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1517141.6250 - 805us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3511645.0000 - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 838893.1875 - 675us/epoch - 11us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 827us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4893953.0000 - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1363903.0000 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 724334.0625 - 703us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 791164.4375 - 893us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1489761.8750 - 695us/epoch - 11us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 793706.5625 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1531737.7500 - 741us/epoch - 12us/sample\n", - "Generation 91/300\n", - "Solving for Nash Equilibrium in Generation 91/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 939us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1378150.8750 - 911us/epoch - 15us/sample\n", - "Episode 2/50\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3532740.2500 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 738us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6976729.0000 - 695us/epoch - 11us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3508055.7500 - 683us/epoch - 11us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506949.3750 - 706us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3999305.5000 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450342.7812 - 703us/epoch - 11us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 673207.3125 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443139.9688 - 855us/epoch - 14us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4056435.0000 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1576477.3750 - 2ms/epoch - 36us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 606440.9375 - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2711974.0000 - 655us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 884us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1317255.1250 - 801us/epoch - 13us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1602548.5000 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 761us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4189763.7500 - 797us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1508807.3750 - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1415595.7500 - 739us/epoch - 12us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3281247.7500 - 824us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 663814.4375 - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1136126.2500 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 766us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1652620.0000 - 821us/epoch - 13us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 965862.1875 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3005397.0000 - 783us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1259056.7500 - 1ms/epoch - 17us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 723us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7893916.5000 - 746us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1578635.2500 - 5ms/epoch - 83us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4598023.0000 - 724us/epoch - 12us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2886109.2500 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 6ms/epoch - 94us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450447.7812 - 1ms/epoch - 20us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 914us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1537332.2500 - 772us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1600269.5000 - 647us/epoch - 10us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3099933.5000 - 3ms/epoch - 41us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1050609.6250 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 405842.2188 - 1ms/epoch - 22us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 992us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 945403.5000 - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 59us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 624626.0000 - 2ms/epoch - 27us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 888us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 982353.6875 - 816us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5761703.0000 - 966us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8890793.0000 - 783us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3600211.7500 - 2ms/epoch - 30us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2943232.0000 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 856us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 658073.8125 - 713us/epoch - 11us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3071318.0000 - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2900613.7500 - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 630807.5625 - 709us/epoch - 11us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1571670.5000 - 2ms/epoch - 30us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1869314.1250 - 6ms/epoch - 100us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1050018.2500 - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 7ms/epoch - 114us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 816767.3125 - 4ms/epoch - 62us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1050681.5000 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2929456.0000 - 776us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 409056.4688 - 1ms/epoch - 18us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1495846.7500 - 1ms/epoch - 19us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 277780.4688 - 947us/epoch - 15us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 695461.4375 - 3ms/epoch - 52us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1003679.0625 - 775us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2425221.7500 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 796us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1164473.5000 - 759us/epoch - 12us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1515473.0000 - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1626732.6250 - 1ms/epoch - 18us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1142694.1250 - 891us/epoch - 14us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5732936.5000 - 830us/epoch - 13us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2831058.0000 - 829us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2838483.2500 - 713us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489878.1875 - 3ms/epoch - 51us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1956323.5000 - 1ms/epoch - 18us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 261052.8281 - 865us/epoch - 14us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2151376.2500 - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1766170.7500 - 900us/epoch - 15us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 462903.9062 - 1ms/epoch - 17us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 847862.3125 - 994us/epoch - 16us/sample\n", - "Generation 92/300\n", - "Solving for Nash Equilibrium in Generation 92/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1842797.3750 - 880us/epoch - 14us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 816438.2500 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 872us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 979749.1250 - 893us/epoch - 14us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5048198.0000 - 781us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 994105.6875 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1075168.2500 - 1ms/epoch - 23us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 780us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5606321.5000 - 753us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1852235.0000 - 1ms/epoch - 23us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 787us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 519770.7500 - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4287321.0000 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2796009.7500 - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 767us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1652938.2500 - 799us/epoch - 13us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 412762.4688 - 3ms/epoch - 55us/sample\n", - "Episode 8/50\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 935us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684230.6875 - 995us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 967us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3345509.2500 - 753us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6432269.0000 - 878us/epoch - 14us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1641860.7500 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 471238.6250 - 841us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1088138.2500 - 737us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3048033.0000 - 746us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2119450.5000 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1679973.6250 - 740us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3682257.0000 - 717us/epoch - 12us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 572402.6875 - 961us/epoch - 16us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1050616.5000 - 755us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1740198.8750 - 701us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 828us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3343419.7500 - 746us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 912us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1706661.0000 - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17393354.0000 - 695us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 866167.3125 - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 5ms/epoch - 84us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1743112.8750 - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4784669.0000 - 2ms/epoch - 30us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3820003.5000 - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1572588.0000 - 812us/epoch - 13us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 769us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1595045.6250 - 799us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3508287.2500 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2869507.2500 - 778us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1884848.7500 - 819us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1262476.6250 - 790us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 808us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1033834.0000 - 739us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 270986.6250 - 735us/epoch - 12us/sample\n", - "Episode 29/50\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5774730.0000 - 777us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2917693.0000 - 866us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2892150.0000 - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 818us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1781032.7500 - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 682416.6875 - 1ms/epoch - 16us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6988473.0000 - 677us/epoch - 11us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5843132.5000 - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 726us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3126333.2500 - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1619035.5000 - 632us/epoch - 10us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1103769.2500 - 712us/epoch - 11us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4388027.0000 - 881us/epoch - 14us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1230859.1250 - 869us/epoch - 14us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4929866.5000 - 713us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 568728.4375 - 793us/epoch - 13us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2047660.8750 - 2ms/epoch - 32us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1160476.3750 - 828us/epoch - 13us/sample\n", - "Episode 43/50\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 425640.7812 - 912us/epoch - 15us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4069567.7500 - 716us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 899us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1068544.2500 - 4ms/epoch - 71us/sample\n", - "Episode 46/50\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 943us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 561392.6250 - 796us/epoch - 13us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1254589.5000 - 777us/epoch - 13us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 797us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 360848.1875 - 782us/epoch - 13us/sample\n", - "Generation 93/300\n", - "Solving for Nash Equilibrium in Generation 93/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1072710.0000 - 707us/epoch - 11us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 783us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 320863.0625 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 765us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2706849.7500 - 689us/epoch - 11us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3232360.0000 - 848us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1389186.1250 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 170567.6562 - 606us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 490867.3125 - 2ms/epoch - 34us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 492674.3125 - 813us/epoch - 13us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 39us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7202548.0000 - 9ms/epoch - 144us/sample\n", - "Episode 7/50\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 69587.0469 - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 921us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 11154826.0000 - 703us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 909980.9375 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1273956.0000 - 763us/epoch - 12us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 895us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1922558.2500 - 753us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1843395.5000 - 697us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2738549.7500 - 993us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4521010.0000 - 787us/epoch - 13us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 829us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3554431.0000 - 780us/epoch - 13us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4769371.0000 - 3ms/epoch - 45us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 489268.8438 - 980us/epoch - 16us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1664354.8750 - 734us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1214127.5000 - 6ms/epoch - 98us/sample\n", - "Episode 14/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 310050.0000 - 826us/epoch - 13us/sample\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 919us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3043267.7500 - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2017299.2500 - 719us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2676759.0000 - 697us/epoch - 11us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 817us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4710549.0000 - 735us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7882968.5000 - 690us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 428307.5625 - 8ms/epoch - 134us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3199946.5000 - 683us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1087930.6250 - 653us/epoch - 11us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 742us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1171771.1250 - 733us/epoch - 12us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 821us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1138291.0000 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 880535.6250 - 878us/epoch - 14us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1795311.0000 - 714us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1939102.8750 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 788us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3075805.5000 - 757us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5458575.0000 - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1103659.8750 - 736us/epoch - 12us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2003879.6250 - 786us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1329179.6250 - 776us/epoch - 13us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1777385.0000 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1802043.7500 - 2ms/epoch - 27us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 809us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3388935.5000 - 755us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3149479.5000 - 771us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1000us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 443688.3750 - 742us/epoch - 12us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506516.0938 - 832us/epoch - 13us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1196157.0000 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 983us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3126062.7500 - 934us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1972601.6250 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 825us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 321396.2188 - 775us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1556409.2500 - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4721376.5000 - 822us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 21us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4375911.0000 - 961us/epoch - 16us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 934227.6250 - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2080889.1250 - 835us/epoch - 13us/sample\n", - "Episode 36/50\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 38us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5522916.5000 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2522631.2500 - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 366265.5312 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2264183.7500 - 832us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 936us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 290466.6875 - 733us/epoch - 12us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 893us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5403199.0000 - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1815189.5000 - 797us/epoch - 13us/sample\n", - "Episode 40/50\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1209971.7500 - 1ms/epoch - 18us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 871us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12006691.0000 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1120702.7500 - 959us/epoch - 15us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1917050.3750 - 747us/epoch - 12us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 990us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2231620.2500 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 30us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1995588.0000 - 2ms/epoch - 27us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 857us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 522384.6562 - 1ms/epoch - 18us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 861us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4464878.0000 - 973us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1939093.8750 - 984us/epoch - 16us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5307339.5000 - 834us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1132748.2500 - 739us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 455057.1875 - 878us/epoch - 14us/sample\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 895529.5000 - 892us/epoch - 14us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3916636.5000 - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1847007.6250 - 756us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 880us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 709150.8750 - 841us/epoch - 14us/sample\n", - "Generation 94/300\n", - "Solving for Nash Equilibrium in Generation 94/300\n", - "Computing Nash Equilibrium for 16 matches\n", - "Episode 1/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2016174.1250 - 807us/epoch - 13us/sample\n", - "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 814us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 47179.7383 - 923us/epoch - 15us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 879us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6215709.5000 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1175989.8750 - 794us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7365395.5000 - 5ms/epoch - 77us/sample\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 113273.2031 - 837us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1776806.6250 - 741us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 721us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 520736.3750 - 648us/epoch - 10us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1456807.1250 - 754us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 847us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7186700.5000 - 694us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 837us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 684148.3750 - 732us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6835306.0000 - 812us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 58146.2578 - 3ms/epoch - 47us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 815us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3203007.2500 - 763us/epoch - 12us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 920770.1250 - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 34us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1553979.1250 - 2ms/epoch - 30us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2069223.0000 - 774us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1186204.6250 - 681us/epoch - 11us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4931449.0000 - 719us/epoch - 12us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1951120.7500 - 751us/epoch - 12us/sample\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2608047.0000 - 732us/epoch - 12us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7893049.0000 - 664us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 811us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6542754.0000 - 795us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 300090.9375 - 721us/epoch - 12us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1169805.6250 - 781us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 947us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2705672.0000 - 798us/epoch - 13us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 847423.6250 - 747us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2028739.8750 - 773us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 862us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3518572.5000 - 865us/epoch - 14us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 876us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2637563.5000 - 619us/epoch - 10us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3234750.0000 - 803us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3525117.7500 - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1813333.0000 - 936us/epoch - 15us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 945us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 71459.3984 - 894us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 989us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1821241.8750 - 806us/epoch - 13us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1285356.6250 - 834us/epoch - 13us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3737057.2500 - 2ms/epoch - 32us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 789us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1786758.5000 - 811us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 836us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 938142.3125 - 685us/epoch - 11us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4498871.0000 - 12ms/epoch - 187us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 883us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3285001.5000 - 953us/epoch - 15us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7368901.0000 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 43us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1220559.0000 - 6ms/epoch - 101us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3301237.2500 - 736us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1189807.8750 - 753us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1212280.8750 - 780us/epoch - 13us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1964232.2500 - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 850us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1302228.8750 - 727us/epoch - 12us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 907us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2385866.2500 - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5599609.5000 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3596822.2500 - 2ms/epoch - 26us/sample\n", - "Episode 35/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 782us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 305702.2812 - 676us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 937us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3567725.7500 - 853us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3362272.5000 - 716us/epoch - 12us/sample\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 851us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2060412.3750 - 736us/epoch - 12us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3978377.2500 - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 801us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1826685.3750 - 804us/epoch - 13us/sample\n", - "Episode 38/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 49us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4109744.2500 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342984.9375 - 1ms/epoch - 19us/sample\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 940336.0625 - 750us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1965228.8750 - 2ms/epoch - 26us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 778us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2077803.5000 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 843us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1870109.6250 - 976us/epoch - 16us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2014719.7500 - 744us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 776us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3636530.7500 - 701us/epoch - 11us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 967819.9375 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 760278.9375 - 810us/epoch - 13us/sample\n", - "Episode 44/50\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 889us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5281079.5000 - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 40us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1956820.3750 - 2ms/epoch - 32us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 875us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 784526.6250 - 724us/epoch - 12us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3510739.2500 - 1ms/epoch - 22us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2274401.0000 - 745us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 885us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2305306.5000 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8019161.0000 - 731us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2707885.5000 - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 773us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 858160.1250 - 739us/epoch - 12us/sample\n", - "Generation 95/300\n", - "Solving for Nash Equilibrium in Generation 95/300\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 87/300\n", + "Solving for Nash Equilibrium in Generation 87/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 767454.5625 - 729us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 917us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6242885.0000 - 2ms/epoch - 27us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 27us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5224380.0000 - 2ms/epoch - 38us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 532415.8125 - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 312234.7188 - 827us/epoch - 13us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 932us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2327205.0000 - 719us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2141260.0000 - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1838247.1250 - 1000us/epoch - 16us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 833us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8949283.0000 - 666us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 835us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5271310.5000 - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1246608.5000 - 745us/epoch - 12us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 52us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2331874.7500 - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 981us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2040474.1250 - 1ms/epoch - 19us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 790us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3742262.2500 - 694us/epoch - 11us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2066764.7500 - 790us/epoch - 13us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Episode 16/50\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 844us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2054553.1250 - 892us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1388513.6250 - 887us/epoch - 14us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2496103.2500 - 835us/epoch - 13us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 941143.3750 - 5ms/epoch - 74us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 953us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2130902.7500 - 846us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 910us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 785938.8750 - 727us/epoch - 12us/sample\n", - "Episode 22/50\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7514443.5000 - 794us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 832us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1343886.6250 - 800us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7314212.5000 - 908us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 764us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2147674.0000 - 710us/epoch - 11us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3719731.2500 - 4ms/epoch - 60us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 957us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1232073.2500 - 731us/epoch - 12us/sample\n", - "Episode 26/50\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1235474.1250 - 921us/epoch - 15us/sample\n", - "Episode 28/50\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2044110.6250 - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2877634.7500 - 815us/epoch - 13us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 842us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3507183.7500 - 743us/epoch - 12us/sample\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 982us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 864091.6250 - 864us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 720us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12613673.0000 - 830us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9582343.0000 - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1985513.7500 - 791us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 64us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 844910.8125 - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 758us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1246620.3750 - 771us/epoch - 12us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 882us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13303305.0000 - 754us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2553359.7500 - 667us/epoch - 11us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 918us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2578156.2500 - 741us/epoch - 12us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 804us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1408958.3750 - 1ms/epoch - 17us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8992591.0000 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 896us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4360431.0000 - 838us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1255649.0000 - 712us/epoch - 11us/sample\n", - "Episode 41/50\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 911us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450577.5938 - 755us/epoch - 12us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 949us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4408254.5000 - 687us/epoch - 11us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12872404.0000 - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 913us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2557510.5000 - 618us/epoch - 10us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1942549.6250 - 821us/epoch - 13us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2380300.7500 - 707us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 797985.7500 - 728us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 36us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10342834.0000 - 816us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3567996.7500 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 860us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3867475.7500 - 673us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1434583.0000 - 737us/epoch - 12us/sample\n", - "Episode 48/50\n", - "Episode 49/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 942us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4845750.5000 - 768us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 916us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2142583.2500 - 747us/epoch - 12us/sample\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3571691.7500 - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2253468.7500 - 704us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 770us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2088598.6250 - 725us/epoch - 12us/sample\n", - "Generation 96/300\n", - "Solving for Nash Equilibrium in Generation 96/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 88/300\n", + "Solving for Nash Equilibrium in Generation 88/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 849us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2499857.5000 - 734us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 846us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4050578.5000 - 883us/epoch - 14us/sample\n", - "Episode 4/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 915us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 121301.9609 - 749us/epoch - 12us/sample\n", - "Episode 5/50\n", - "Episode 6/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 870us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9982190.0000 - 1ms/epoch - 16us/sample\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2815522.0000 - 723us/epoch - 12us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 900us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2262914.5000 - 666us/epoch - 11us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 785us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1266560.0000 - 3ms/epoch - 48us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 784us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5542151.5000 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 869us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 450060.0625 - 699us/epoch - 11us/sample\n", - "Episode 11/50\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 904us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2572144.2500 - 902us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 760us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 235173.9062 - 728us/epoch - 12us/sample\n", - "Episode 13/50\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 822us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 13818174.0000 - 700us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 28us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3705027.0000 - 1ms/epoch - 21us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 819us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1204632.3750 - 672us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 845us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2250725.5000 - 758us/epoch - 12us/sample\n", - "Episode 17/50\n", - "Episode 18/50\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 806us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2121247.7500 - 958us/epoch - 15us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3679446.7500 - 810us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 317987.7188 - 847us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 25us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3660298.0000 - 935us/epoch - 15us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1296751.2500 - 966us/epoch - 16us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 891us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2497382.0000 - 798us/epoch - 13us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 823us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 8638864.0000 - 905us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 504004.3438 - 1ms/epoch - 19us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 864us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2392141.2500 - 849us/epoch - 14us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 988us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 10839182.0000 - 1ms/epoch - 19us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 816us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 377606.1562 - 1ms/epoch - 17us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 50us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2401782.2500 - 1ms/epoch - 17us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 855us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 7902905.5000 - 840us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 978us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4237547.5000 - 800us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2035633.2500 - 870us/epoch - 14us/sample\n", - "Episode 30/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 24us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 506494.5312 - 793us/epoch - 13us/sample\n", - "Episode 31/50\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1304937.3750 - 919us/epoch - 15us/sample\n", - "Episode 33/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 956us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2240468.7500 - 784us/epoch - 13us/sample\n", - "Episode 34/50\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 792us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1693302.8750 - 993us/epoch - 16us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 55us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2639029.2500 - 2ms/epoch - 32us/sample\n", - "Episode 38/50\n", - "Episode 39/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 839us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 191807.5781 - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 858us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2904265.2500 - 751us/epoch - 12us/sample\n", - "Episode 40/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 70us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 9457348.0000 - 1ms/epoch - 17us/sample\n", - "Episode 41/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 867us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2115359.5000 - 1ms/epoch - 18us/sample\n", - "Episode 42/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 35us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4305203.0000 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 65us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1387843.8750 - 3ms/epoch - 46us/sample\n", - "Episode 43/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 892us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3435792.7500 - 923us/epoch - 15us/sample\n", - "Episode 44/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3109019.2500 - 731us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 971us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5666734.0000 - 993us/epoch - 16us/sample\n", - "Episode 45/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1501818.6250 - 723us/epoch - 12us/sample\n", - "Episode 46/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 831us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 585814.7500 - 828us/epoch - 13us/sample\n", - "Episode 47/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 26us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4815975.0000 - 1ms/epoch - 18us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 4ms/epoch - 66us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 395103.9688 - 1ms/epoch - 23us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 979us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5292211.0000 - 672us/epoch - 11us/sample\n", - "Episode 48/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 998us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 587455.8750 - 820us/epoch - 13us/sample\n", - "Episode 49/50\n", - "Episode 50/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 20us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5554949.0000 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3764589.0000 - 898us/epoch - 14us/sample\n", - "Generation 97/300\n", - "Solving for Nash Equilibrium in Generation 97/300\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 89/300\n", + "Solving for Nash Equilibrium in Generation 89/300\n", "Computing Nash Equilibrium for 16 matches\n", "Episode 1/50\n", "Episode 2/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5871947.5000 - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 901us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1474525.2500 - 749us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 970us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2445722.7500 - 741us/epoch - 12us/sample\n", - "Episode 3/50\n", - "Episode 4/50\n", - "Episode 5/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2560065.7500 - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 980us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 969684.0625 - 757us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 887us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 17110172.0000 - 752us/epoch - 12us/sample\n", - "Episode 6/50\n", - "Episode 7/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 3ms/epoch - 44us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2415662.2500 - 3ms/epoch - 45us/sample\n", - "Episode 8/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 866us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5682156.5000 - 877us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 906us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4296715.0000 - 813us/epoch - 13us/sample\n", - "Episode 9/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 874us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1352037.5000 - 777us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 31us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1278497.5000 - 2ms/epoch - 26us/sample\n", - "Episode 10/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1512344.5000 - 745us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 593658.5625 - 5ms/epoch - 88us/sample\n", - "Episode 11/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 920us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2344798.7500 - 895us/epoch - 14us/sample\n", - "Episode 12/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 748us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4319715.5000 - 743us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 37us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2914878.0000 - 1ms/epoch - 23us/sample\n", - "Episode 13/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 922us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2145114.7500 - 869us/epoch - 14us/sample\n", - "Episode 14/50\n", - "Episode 15/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 799us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4132423.2500 - 739us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 881us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2331868.5000 - 632us/epoch - 10us/sample\n", - "Episode 16/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 793us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 342000.0000 - 775us/epoch - 13us/sample\n", - "Episode 17/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 6692617.0000 - 2ms/epoch - 29us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 865us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1887120.6250 - 869us/epoch - 14us/sample\n", - "Episode 18/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 2ms/epoch - 33us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1671615.6250 - 3ms/epoch - 43us/sample\n", - "Episode 19/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 946us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 5032903.0000 - 858us/epoch - 14us/sample\n", - "Episode 20/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 883102.4375 - 877us/epoch - 14us/sample\n", - "Episode 21/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 705us/epoch - 11us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1375462.7500 - 796us/epoch - 13us/sample\n", - "Episode 22/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 852us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2173505.2500 - 749us/epoch - 12us/sample\n", - "Episode 23/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 19us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2388224.0000 - 946us/epoch - 15us/sample\n", - "Episode 24/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4283433.5000 - 855us/epoch - 14us/sample\n", - "Episode 25/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 848us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3562467.5000 - 759us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 813us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 12276266.0000 - 635us/epoch - 10us/sample\n", - "Episode 26/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 807us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1489430.7500 - 740us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 868us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3946604.7500 - 925us/epoch - 15us/sample\n", - "Episode 27/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 826us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1391791.0000 - 788us/epoch - 13us/sample\n", - "Episode 28/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 975us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1685692.3750 - 960us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1840143.2500 - 881us/epoch - 14us/sample\n", - "Episode 29/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 16017575.0000 - 1ms/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 903us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 3748236.5000 - 691us/epoch - 11us/sample\n", - "Episode 30/50\n", - "Episode 31/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 968us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 4269057.5000 - 803us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 1ms/epoch - 17us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2517508.7500 - 869us/epoch - 14us/sample\n", - "Episode 32/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 763us/epoch - 12us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1750924.3750 - 840us/epoch - 14us/sample\n", - "Episode 33/50\n", - "Episode 34/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 934us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 200356.9844 - 955us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 927us/epoch - 15us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1181988.8750 - 779us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 991us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2370348.5000 - 645us/epoch - 10us/sample\n", - "Episode 35/50\n", - "Episode 36/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 824us/epoch - 13us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1327120.8750 - 972us/epoch - 16us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 854us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 2433162.0000 - 810us/epoch - 13us/sample\n", - "Episode 37/50\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: nan - 863us/epoch - 14us/sample\n", - "Train on 62 samples\n", - "62/62 - 0s - loss: 1022126.4375 - 744us/epoch - 12us/sample\n" - ] - }, - { - "ename": "", - "evalue": "", - "output_type": "error", - "traceback": [ - "\u001b[1;31mThe Kernel crashed while executing code in the current cell or a previous cell. \n", - "\u001b[1;31mPlease review the code in the cell(s) to identify a possible cause of the failure. \n", - "\u001b[1;31mClick here for more info. \n", - "\u001b[1;31mView Jupyter log for further details." - ] - } - ], - "source": [ - "from main import simulation_10_agents, simulation_4_agents, simulation_10_to_4_agents\n", - "import copy\n", - "\n", - "\n", - "avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f = None, None, None, None, None, None, None\n", - "for seed in seeds:\n", - " avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results= simulation_4_agents(seed, False) \n", - " if seed == seeds[0]:\n", - " avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f= avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results\n", - " first_gen_results_f = [list(item) for item in first_gen_results_f]\n", - " middle_gen_results_f = [list(item) for item in middle_gen_results_f]\n", - " last_gen_results_f = [list(item) for item in last_gen_results_f]\n", - " else:\n", - " avg_training_rewards_f = [sum(x) / 2 for x in zip(avg_training_rewards_f, avg_training_rewards)]\n", - " print(\"here\")\n", - " print(first_gen_results[0])\n", - " print(\"here1\")\n", - " print(first_gen_results[0][0])\n", - "\n", - "\n", - " first_gen_results_f[0][0] = [sum(x) / 2 for x in zip(first_gen_results_f[0][0], first_gen_results[0][0])]\n", - " middle_gen_results_f[0][0] = [sum(x) / 2 for x in zip(middle_gen_results_f[0][0], middle_gen_results[0][0])]\n", - " last_gen_results_f[0][0] = [sum(x) / 2 for x in zip(last_gen_results_f[0][0], last_gen_results[0][0])]\n", - "\n", - " avg_training_norm_violations_f = avg_training_norm_violations_f.add(avg_training_norm_violations).div(2)\n", - " first_gen_results_f[0][1] = first_gen_results_f[0][1].add(first_gen_results[0][1]).div(2)\n", - " middle_gen_results_f[0][1] = middle_gen_results_f[0][1].add(middle_gen_results[0][1]).div(2)\n", - " last_gen_results_f[0][1] = last_gen_results_f[0][1].add(last_gen_results[0][1]).div(2)\n", - "\n", - " sim_violated_f = sim_violated_f.add(sim_violated).div(2)\n", - " at_dest_f = [sum(x) / 2 for x in zip(at_dest_f, at_dest)]\n", - "\n", - "#plot results\n", - "plt.plot(avg_training_rewards_f, label='Average training reward')\n", - "plt.plot(first_gen_results_f[0][0], label='First generation reward')\n", - "plt.plot(middle_gen_results_f[0][0], label='Middle generation reward')\n", - "plt.plot(last_gen_results_f[0][0], label='Last generation reward')\n", - "plt.legend()\n", - "plt.title('Average training reward')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Average reward')\n", - "plt.show()\n", - "\n", - "\n", - "#training violations\n", - "costs_from_violations = copy.deepcopy(avg_training_norm_violations_f['total_violations_cost'])\n", - "avg_training_norm_violations_f.drop(columns=['seed'], inplace=True)\n", - "avg_training_norm_violations_f.drop(columns=['total_violations_cost'], inplace=True)\n", - "\n", - "#graph costs from violations\n", - "plt.plot(costs_from_violations)\n", - "plt.title('AVERAGE Costs from violations training')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "avg_training_norm_violations_f.plot(kind='bar', stacked=True)\n", - "plt.title('Average training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "plt.show()\n", - "\n", - "#FIRST GEN TRAINING VIOLATIONS \n", - "costs_from_violations = copy.deepcopy(first_gen_results_f[0][1]['total_violations_cost'])\n", - "first_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", - "first_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", - "\n", - "#graph costs from violations\n", - "plt.plot(costs_from_violations)\n", - "plt.title('First generation Costs from violations training')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "first_gen_results_f[0][1].plot(kind='bar', stacked=True)\n", - "plt.title('First generation training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "plt.show()\n", - "\n", - "#MIDDLE GEN TRAINING VIOLATIONS\n", - "costs_from_violations = copy.deepcopy(middle_gen_results_f[0][1]['total_violations_cost'])\n", - "middle_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", - "middle_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", - "plt.plot(costs_from_violations)\n", - "plt.title('Middle generation Costs from violations training')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "middle_gen_results_f[0][1].plot(kind='bar', stacked=True)\n", - "plt.title('Middle generation training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "\n", - "plt.show()\n", - "\n", - "#LAST GEN TRAINING VIOLATIONS\n", - "costs_from_violations = copy.deepcopy(last_gen_results_f[0][1]['total_violations_cost'])\n", - "last_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", - "last_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", - "plt.plot(costs_from_violations)\n", - "plt.title('Last generation Costs from violations training')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "last_gen_results_f[0][1].plot(kind='bar', stacked=True)\n", - "plt.title('Last generation training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "plt.show()\n", - "\n", - "\n", - "\n", - "\n", - "#training violations\n", - "costs_from_violations_f = copy.deepcopy(sim_violated_f['total_violations_cost'])\n", - "sim_violated_f.drop(columns=['seed'], inplace=True)\n", - "sim_violated_f.drop(columns=['total_violations_cost'], inplace=True)\n", - "\n", - "#graph costs from violations\n", - "plt.plot(costs_from_violations)\n", - "plt.title('Costs from violations simulation')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "\n", - "# Plot a stacked bar chart\n", - "sim_violated_f.plot(kind='bar', stacked=True)\n", - "plt.title('Average simulation norm violations')\n", - "plt.xlabel('Simulation Run')\n", - "plt.ylabel('Average norm violations')\n", - "plt.show()\n", - "\n", - "#simulation at destination, avg timeto destination\n", - "plt.plot(at_dest_f)\n", - "plt.title('Average Steps to destination')\n", - "plt.xlabel('Simulation Run')\n", - "plt.ylabel('Steps ')\n", - "plt.show()\n", - "\n", - "\n", - "\n", - "\n", - "\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 10 Agents" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [ - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n", - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 3 done\n", - "Episode 4 done\n", - "Episode 5 done\n", - "Episode 6 done\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 9 done\n", - "Episode 9 done\n", - "Total steps: 18\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n", - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 3 done\n", - "Episode 4 done\n", - "Episode 5 done\n", - "Episode 6 done\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 9 done\n", - "here\n", - "([550.1646499999999, 270.01365000000015, 513.1350000000007, 3593.441600000003, 6561.650350000002, 8492.149900000002, 9051.398650000005, 12018.088300000005, 14074.011400000003, 16089.473050000004], seed step_collisions not_on_track yield_violations \\\n", - "0 1.0 4.0 15.0 1.0 \n", - "1 1.0 5.0 26.0 3.0 \n", - "2 1.0 4.0 19.0 2.0 \n", - "3 1.0 4.0 15.0 1.0 \n", - "4 1.0 4.0 15.0 2.0 \n", - "5 1.0 6.0 28.0 1.0 \n", - "6 1.0 5.0 10.0 1.0 \n", - "7 1.0 5.0 17.0 1.0 \n", - "8 1.0 4.0 20.0 2.0 \n", - "9 1.0 4.0 20.0 2.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 653.0 \n", - "1 0.0 1175.0 \n", - "2 0.0 689.0 \n", - "3 0.0 678.0 \n", - "4 0.0 659.0 \n", - "5 0.0 953.0 \n", - "6 0.0 474.0 \n", - "7 0.0 626.0 \n", - "8 0.0 1032.0 \n", - "9 0.0 837.0 \n", - "\n", - " total_violations_cost \n", - "0 -189.65 \n", - "1 -294.75 \n", - "2 -213.45 \n", - "3 -190.90 \n", - "4 -191.95 \n", - "5 -309.65 \n", - "6 -175.70 \n", - "7 -218.30 \n", - "8 -235.60 \n", - "9 -225.85 )\n", - "here1\n", - "[550.1646499999999, 270.01365000000015, 513.1350000000007, 3593.441600000003, 6561.650350000002, 8492.149900000002, 9051.398650000005, 12018.088300000005, 14074.011400000003, 16089.473050000004]\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n", - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 3 done\n", - "Episode 4 done\n", - "Episode 5 done\n", - "Episode 6 done\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 8 done\n", - "Total steps: 17\n", - "Episode 9 done\n", - "here\n", - "([2730.790050000001, 4685.487450000001, 4491.37425, 4957.741300000003, 7941.8098000000045, 11802.204850000006, 12036.965599999945, 12382.346999999945, 14518.048149999948, 14159.171199999944], seed step_collisions not_on_track yield_violations \\\n", - "0 2.0 4.0 18.0 9.0 \n", - "1 2.0 5.0 27.0 2.0 \n", - "2 2.0 3.0 11.0 0.0 \n", - "3 2.0 5.0 21.0 1.0 \n", - "4 2.0 4.0 20.0 1.0 \n", - "5 2.0 3.0 21.0 0.0 \n", - "6 2.0 4.0 18.0 3.0 \n", - "7 2.0 4.0 9.0 0.0 \n", - "8 2.0 4.0 20.0 1.0 \n", - "9 2.0 9.0 7.0 4.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 1231.0 \n", - "1 0.0 979.0 \n", - "2 0.0 278.0 \n", - "3 0.0 848.0 \n", - "4 0.0 829.0 \n", - "5 0.0 670.0 \n", - "6 0.0 1061.0 \n", - "7 0.0 307.0 \n", - "8 0.0 837.0 \n", - "9 0.0 540.0 \n", - "\n", - " total_violations_cost \n", - "0 -249.55 \n", - "1 -287.95 \n", - "2 -128.90 \n", - "3 -249.40 \n", - "4 -223.45 \n", - "5 -198.50 \n", - "6 -229.05 \n", - "7 -140.35 \n", - "8 -223.85 \n", - "9 -250.00 )\n", - "here1\n", - "[2730.790050000001, 4685.487450000001, 4491.37425, 4957.741300000003, 7941.8098000000045, 11802.204850000006, 12036.965599999945, 12382.346999999945, 14518.048149999948, 14159.171199999944]\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n", - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 3 done\n", - "Episode 4 done\n", - "Episode 5 done\n", - "Episode 6 done\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 9 done\n", - "here\n", - "([2135.1220000000017, 3358.799500000001, 5155.691500000002, 6760.615500000004, 8742.117600000005, 9114.779150000004, 11010.966900000001, 13624.713550000004, 14993.061750000004, 83588.57138888891], seed step_collisions not_on_track yield_violations \\\n", - "0 3.0 4.0 4.0 2.0 \n", - "1 3.0 6.0 9.0 0.0 \n", - "2 3.0 6.0 15.0 2.0 \n", - "3 3.0 5.0 15.0 1.0 \n", - "4 3.0 4.0 14.0 2.0 \n", - "5 3.0 3.0 21.0 0.0 \n", - "6 3.0 5.0 21.0 0.0 \n", - "7 3.0 5.0 21.0 2.0 \n", - "8 3.0 6.0 23.0 2.0 \n", - "9 3.0 5.0 19.0 4.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 144.0 \n", - "1 0.0 334.0 \n", - "2 0.0 477.0 \n", - "3 0.0 477.0 \n", - "4 0.0 827.0 \n", - "5 0.0 678.0 \n", - "6 0.0 645.0 \n", - "7 0.0 1020.0 \n", - "8 0.0 968.0 \n", - "9 0.0 1041.0 \n", - "\n", - " total_violations_cost \n", - "0 -111.20 \n", - "1 -181.70 \n", - "2 -222.85 \n", - "3 -200.85 \n", - "4 -195.35 \n", - "5 -198.90 \n", - "6 -237.25 \n", - "7 -260.00 \n", - "8 -287.40 \n", - "9 -255.05 )\n", - "here1\n", - "[2135.1220000000017, 3358.799500000001, 5155.691500000002, 6760.615500000004, 8742.117600000005, 9114.779150000004, 11010.966900000001, 13624.713550000004, 14993.061750000004, 83588.57138888891]\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n", - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 3 done\n", - "Episode 4 done\n", - "Episode 5 done\n", - "Episode 6 done\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 9 done\n", - "here\n", - "([1081.7498000000003, 934.5116500000004, 3681.720250000002, 4149.717150000018, 7229.166200000019, 9554.48345000002, 9156.74115000002, 9157.009600000018, 9555.125149999987, 10880.996399999987], seed step_collisions not_on_track yield_violations \\\n", - "0 4.0 4.0 21.0 1.0 \n", - "1 4.0 4.0 23.0 0.0 \n", - "2 4.0 4.0 22.0 1.0 \n", - "3 4.0 5.0 24.0 4.0 \n", - "4 4.0 3.0 20.0 1.0 \n", - "5 4.0 5.0 10.0 1.0 \n", - "6 4.0 4.0 21.0 1.0 \n", - "7 4.0 4.0 25.0 2.0 \n", - "8 4.0 3.0 11.0 0.0 \n", - "9 4.0 4.0 20.0 1.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 828.0 \n", - "1 0.0 620.0 \n", - "2 0.0 631.0 \n", - "3 0.0 1399.0 \n", - "4 0.0 671.0 \n", - "5 0.0 312.0 \n", - "6 0.0 655.0 \n", - "7 0.0 837.0 \n", - "8 0.0 282.0 \n", - "9 0.0 835.0 \n", - "\n", - " total_violations_cost \n", - "0 -228.40 \n", - "1 -226.00 \n", - "2 -223.55 \n", - "3 -297.95 \n", - "4 -195.55 \n", - "5 -167.60 \n", - "6 -219.75 \n", - "7 -250.85 \n", - "8 -129.10 \n", - "9 -223.75 )\n", - "here1\n", - "[1081.7498000000003, 934.5116500000004, 3681.720250000002, 4149.717150000018, 7229.166200000019, 9554.48345000002, 9156.74115000002, 9157.009600000018, 9555.125149999987, 10880.996399999987]\n" - ] - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - }, + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 90/300\n", + "Solving for Nash Equilibrium in Generation 90/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 91/300\n", + "Solving for Nash Equilibrium in Generation 91/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 92/300\n", + "Solving for Nash Equilibrium in Generation 92/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 93/300\n", + "Solving for Nash Equilibrium in Generation 93/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 94/300\n", + "Solving for Nash Equilibrium in Generation 94/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 95/300\n", + "Solving for Nash Equilibrium in Generation 95/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 96/300\n", + "Solving for Nash Equilibrium in Generation 96/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 97/300\n", + "Solving for Nash Equilibrium in Generation 97/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 98/300\n", + "Solving for Nash Equilibrium in Generation 98/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 99/300\n", + "Solving for Nash Equilibrium in Generation 99/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 100/300\n", + "Solving for Nash Equilibrium in Generation 100/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 101/300\n", + "Solving for Nash Equilibrium in Generation 101/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 102/300\n", + "Solving for Nash Equilibrium in Generation 102/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 103/300\n", + "Solving for Nash Equilibrium in Generation 103/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 104/300\n", + "Solving for Nash Equilibrium in Generation 104/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 105/300\n", + "Solving for Nash Equilibrium in Generation 105/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 106/300\n", + "Solving for Nash Equilibrium in Generation 106/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 107/300\n", + "Solving for Nash Equilibrium in Generation 107/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 108/300\n", + "Solving for Nash Equilibrium in Generation 108/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 109/300\n", + "Solving for Nash Equilibrium in Generation 109/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 110/300\n", + "Solving for Nash Equilibrium in Generation 110/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 111/300\n", + "Solving for Nash Equilibrium in Generation 111/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 112/300\n", + "Solving for Nash Equilibrium in Generation 112/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 113/300\n", + "Solving for Nash Equilibrium in Generation 113/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 114/300\n", + "Solving for Nash Equilibrium in Generation 114/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 115/300\n", + "Solving for Nash Equilibrium in Generation 115/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 116/300\n", + "Solving for Nash Equilibrium in Generation 116/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 117/300\n", + "Solving for Nash Equilibrium in Generation 117/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 118/300\n", + "Solving for Nash Equilibrium in Generation 118/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 119/300\n", + "Solving for Nash Equilibrium in Generation 119/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 120/300\n", + "Solving for Nash Equilibrium in Generation 120/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 121/300\n", + "Solving for Nash Equilibrium in Generation 121/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 122/300\n", + "Solving for Nash Equilibrium in Generation 122/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 123/300\n", + "Solving for Nash Equilibrium in Generation 123/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 124/300\n", + "Solving for Nash Equilibrium in Generation 124/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 125/300\n", + "Solving for Nash Equilibrium in Generation 125/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 126/300\n", + "Solving for Nash Equilibrium in Generation 126/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 127/300\n", + "Solving for Nash Equilibrium in Generation 127/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 128/300\n", + "Solving for Nash Equilibrium in Generation 128/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 129/300\n", + "Solving for Nash Equilibrium in Generation 129/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 130/300\n", + "Solving for Nash Equilibrium in Generation 130/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 131/300\n", + "Solving for Nash Equilibrium in Generation 131/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 132/300\n", + "Solving for Nash Equilibrium in Generation 132/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 133/300\n", + "Solving for Nash Equilibrium in Generation 133/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 134/300\n", + "Solving for Nash Equilibrium in Generation 134/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 135/300\n", + "Solving for Nash Equilibrium in Generation 135/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 136/300\n", + "Solving for Nash Equilibrium in Generation 136/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 137/300\n", + "Solving for Nash Equilibrium in Generation 137/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 138/300\n", + "Solving for Nash Equilibrium in Generation 138/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 139/300\n", + "Solving for Nash Equilibrium in Generation 139/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 140/300\n", + "Solving for Nash Equilibrium in Generation 140/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 141/300\n", + "Solving for Nash Equilibrium in Generation 141/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 142/300\n", + "Solving for Nash Equilibrium in Generation 142/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 143/300\n", + "Solving for Nash Equilibrium in Generation 143/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 144/300\n", + "Solving for Nash Equilibrium in Generation 144/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 145/300\n", + "Solving for Nash Equilibrium in Generation 145/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 146/300\n", + "Solving for Nash Equilibrium in Generation 146/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 147/300\n", + "Solving for Nash Equilibrium in Generation 147/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 148/300\n", + "Solving for Nash Equilibrium in Generation 148/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 149/300\n", + "Solving for Nash Equilibrium in Generation 149/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 150/300\n", + "Solving for Nash Equilibrium in Generation 150/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 151/300\n", + "Solving for Nash Equilibrium in Generation 151/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 152/300\n", + "Solving for Nash Equilibrium in Generation 152/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 153/300\n", + "Solving for Nash Equilibrium in Generation 153/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 154/300\n", + "Solving for Nash Equilibrium in Generation 154/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 155/300\n", + "Solving for Nash Equilibrium in Generation 155/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 156/300\n", + "Solving for Nash Equilibrium in Generation 156/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 157/300\n", + "Solving for Nash Equilibrium in Generation 157/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 158/300\n", + "Solving for Nash Equilibrium in Generation 158/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 159/300\n", + "Solving for Nash Equilibrium in Generation 159/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 160/300\n", + "Solving for Nash Equilibrium in Generation 160/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 161/300\n", + "Solving for Nash Equilibrium in Generation 161/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 162/300\n", + "Solving for Nash Equilibrium in Generation 162/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 163/300\n", + "Solving for Nash Equilibrium in Generation 163/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 164/300\n", + "Solving for Nash Equilibrium in Generation 164/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 165/300\n", + "Solving for Nash Equilibrium in Generation 165/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 166/300\n", + "Solving for Nash Equilibrium in Generation 166/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 167/300\n", + "Solving for Nash Equilibrium in Generation 167/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 168/300\n", + "Solving for Nash Equilibrium in Generation 168/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 169/300\n", + "Solving for Nash Equilibrium in Generation 169/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 170/300\n", + "Solving for Nash Equilibrium in Generation 170/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 171/300\n", + "Solving for Nash Equilibrium in Generation 171/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 172/300\n", + "Solving for Nash Equilibrium in Generation 172/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 173/300\n", + "Solving for Nash Equilibrium in Generation 173/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 174/300\n", + "Solving for Nash Equilibrium in Generation 174/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 175/300\n", + "Solving for Nash Equilibrium in Generation 175/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 176/300\n", + "Solving for Nash Equilibrium in Generation 176/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 177/300\n", + "Solving for Nash Equilibrium in Generation 177/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 178/300\n", + "Solving for Nash Equilibrium in Generation 178/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 179/300\n", + "Solving for Nash Equilibrium in Generation 179/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 180/300\n", + "Solving for Nash Equilibrium in Generation 180/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 181/300\n", + "Solving for Nash Equilibrium in Generation 181/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 182/300\n", + "Solving for Nash Equilibrium in Generation 182/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 183/300\n", + "Solving for Nash Equilibrium in Generation 183/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 184/300\n", + "Solving for Nash Equilibrium in Generation 184/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 185/300\n", + "Solving for Nash Equilibrium in Generation 185/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 186/300\n", + "Solving for Nash Equilibrium in Generation 186/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 187/300\n", + "Solving for Nash Equilibrium in Generation 187/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 188/300\n", + "Solving for Nash Equilibrium in Generation 188/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 189/300\n", + "Solving for Nash Equilibrium in Generation 189/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 190/300\n", + "Solving for Nash Equilibrium in Generation 190/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 191/300\n", + "Solving for Nash Equilibrium in Generation 191/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 192/300\n", + "Solving for Nash Equilibrium in Generation 192/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 193/300\n", + "Solving for Nash Equilibrium in Generation 193/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 194/300\n", + "Solving for Nash Equilibrium in Generation 194/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 195/300\n", + "Solving for Nash Equilibrium in Generation 195/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 196/300\n", + "Solving for Nash Equilibrium in Generation 196/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 197/300\n", + "Solving for Nash Equilibrium in Generation 197/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 198/300\n", + "Solving for Nash Equilibrium in Generation 198/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 199/300\n", + "Solving for Nash Equilibrium in Generation 199/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 200/300\n", + "Solving for Nash Equilibrium in Generation 200/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 201/300\n", + "Solving for Nash Equilibrium in Generation 201/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 202/300\n", + "Solving for Nash Equilibrium in Generation 202/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 203/300\n", + "Solving for Nash Equilibrium in Generation 203/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 204/300\n", + "Solving for Nash Equilibrium in Generation 204/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 205/300\n", + "Solving for Nash Equilibrium in Generation 205/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 206/300\n", + "Solving for Nash Equilibrium in Generation 206/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 207/300\n", + "Solving for Nash Equilibrium in Generation 207/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 208/300\n", + "Solving for Nash Equilibrium in Generation 208/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 209/300\n", + "Solving for Nash Equilibrium in Generation 209/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 210/300\n", + "Solving for Nash Equilibrium in Generation 210/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 211/300\n", + "Solving for Nash Equilibrium in Generation 211/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 212/300\n", + "Solving for Nash Equilibrium in Generation 212/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 213/300\n", + "Solving for Nash Equilibrium in Generation 213/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 214/300\n", + "Solving for Nash Equilibrium in Generation 214/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 215/300\n", + "Solving for Nash Equilibrium in Generation 215/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 216/300\n", + "Solving for Nash Equilibrium in Generation 216/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 217/300\n", + "Solving for Nash Equilibrium in Generation 217/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 218/300\n", + "Solving for Nash Equilibrium in Generation 218/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 219/300\n", + "Solving for Nash Equilibrium in Generation 219/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 220/300\n", + "Solving for Nash Equilibrium in Generation 220/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 221/300\n", + "Solving for Nash Equilibrium in Generation 221/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 222/300\n", + "Solving for Nash Equilibrium in Generation 222/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 223/300\n", + "Solving for Nash Equilibrium in Generation 223/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 224/300\n", + "Solving for Nash Equilibrium in Generation 224/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 225/300\n", + "Solving for Nash Equilibrium in Generation 225/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 226/300\n", + "Solving for Nash Equilibrium in Generation 226/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 227/300\n", + "Solving for Nash Equilibrium in Generation 227/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 228/300\n", + "Solving for Nash Equilibrium in Generation 228/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 229/300\n", + "Solving for Nash Equilibrium in Generation 229/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 230/300\n", + "Solving for Nash Equilibrium in Generation 230/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 231/300\n", + "Solving for Nash Equilibrium in Generation 231/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 232/300\n", + "Solving for Nash Equilibrium in Generation 232/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 233/300\n", + "Solving for Nash Equilibrium in Generation 233/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 234/300\n", + "Solving for Nash Equilibrium in Generation 234/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 235/300\n", + "Solving for Nash Equilibrium in Generation 235/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 236/300\n", + "Solving for Nash Equilibrium in Generation 236/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 237/300\n", + "Solving for Nash Equilibrium in Generation 237/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 238/300\n", + "Solving for Nash Equilibrium in Generation 238/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 239/300\n", + "Solving for Nash Equilibrium in Generation 239/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 240/300\n", + "Solving for Nash Equilibrium in Generation 240/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 241/300\n", + "Solving for Nash Equilibrium in Generation 241/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 242/300\n", + "Solving for Nash Equilibrium in Generation 242/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 243/300\n", + "Solving for Nash Equilibrium in Generation 243/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 244/300\n", + "Solving for Nash Equilibrium in Generation 244/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 245/300\n", + "Solving for Nash Equilibrium in Generation 245/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 246/300\n", + "Solving for Nash Equilibrium in Generation 246/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 247/300\n", + "Solving for Nash Equilibrium in Generation 247/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 248/300\n", + "Solving for Nash Equilibrium in Generation 248/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 249/300\n", + "Solving for Nash Equilibrium in Generation 249/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 250/300\n", + "Solving for Nash Equilibrium in Generation 250/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 251/300\n", + "Solving for Nash Equilibrium in Generation 251/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 252/300\n", + "Solving for Nash Equilibrium in Generation 252/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 253/300\n", + "Solving for Nash Equilibrium in Generation 253/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 254/300\n", + "Solving for Nash Equilibrium in Generation 254/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 255/300\n", + "Solving for Nash Equilibrium in Generation 255/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 256/300\n", + "Solving for Nash Equilibrium in Generation 256/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 257/300\n", + "Solving for Nash Equilibrium in Generation 257/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 258/300\n", + "Solving for Nash Equilibrium in Generation 258/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 259/300\n", + "Solving for Nash Equilibrium in Generation 259/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 260/300\n", + "Solving for Nash Equilibrium in Generation 260/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 261/300\n", + "Solving for Nash Equilibrium in Generation 261/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 262/300\n", + "Solving for Nash Equilibrium in Generation 262/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 263/300\n", + "Solving for Nash Equilibrium in Generation 263/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 264/300\n", + "Solving for Nash Equilibrium in Generation 264/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 265/300\n", + "Solving for Nash Equilibrium in Generation 265/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 266/300\n", + "Solving for Nash Equilibrium in Generation 266/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 267/300\n", + "Solving for Nash Equilibrium in Generation 267/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 268/300\n", + "Solving for Nash Equilibrium in Generation 268/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 269/300\n", + "Solving for Nash Equilibrium in Generation 269/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 270/300\n", + "Solving for Nash Equilibrium in Generation 270/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 271/300\n", + "Solving for Nash Equilibrium in Generation 271/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 272/300\n", + "Solving for Nash Equilibrium in Generation 272/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 273/300\n", + "Solving for Nash Equilibrium in Generation 273/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 274/300\n", + "Solving for Nash Equilibrium in Generation 274/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 275/300\n", + "Solving for Nash Equilibrium in Generation 275/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 276/300\n", + "Solving for Nash Equilibrium in Generation 276/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 277/300\n", + "Solving for Nash Equilibrium in Generation 277/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 278/300\n", + "Solving for Nash Equilibrium in Generation 278/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 279/300\n", + "Solving for Nash Equilibrium in Generation 279/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 280/300\n", + "Solving for Nash Equilibrium in Generation 280/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 281/300\n", + "Solving for Nash Equilibrium in Generation 281/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 282/300\n", + "Solving for Nash Equilibrium in Generation 282/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 283/300\n", + "Solving for Nash Equilibrium in Generation 283/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 284/300\n", + "Solving for Nash Equilibrium in Generation 284/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 285/300\n", + "Solving for Nash Equilibrium in Generation 285/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 286/300\n", + "Solving for Nash Equilibrium in Generation 286/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 287/300\n", + "Solving for Nash Equilibrium in Generation 287/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 288/300\n", + "Solving for Nash Equilibrium in Generation 288/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 289/300\n", + "Solving for Nash Equilibrium in Generation 289/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 290/300\n", + "Solving for Nash Equilibrium in Generation 290/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 291/300\n", + "Solving for Nash Equilibrium in Generation 291/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 292/300\n", + "Solving for Nash Equilibrium in Generation 292/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 293/300\n", + "Solving for Nash Equilibrium in Generation 293/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 294/300\n", + "Solving for Nash Equilibrium in Generation 294/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 295/300\n", + "Solving for Nash Equilibrium in Generation 295/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 296/300\n", + "Solving for Nash Equilibrium in Generation 296/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 297/300\n", + "Solving for Nash Equilibrium in Generation 297/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 298/300\n", + "Solving for Nash Equilibrium in Generation 298/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 299/300\n", + "Solving for Nash Equilibrium in Generation 299/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 300/300\n", + "Solving for Nash Equilibrium in Generation 300/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Running Computed Policies\n", + "Episode 0 done\n", + "Episode 1 done\n", + "Episode 2 done\n", + "Episode 3 done\n", + "Episode 4 done\n", + "Episode 4 done\n", + "Total steps: 13\n", + "Episode 5 done\n", + "Episode 6 done\n", + "Episode 6 done\n", + "Total steps: 13\n", + "Episode 7 done\n", + "Episode 8 done\n", + "Episode 8 done\n", + "Total steps: 13\n", + "Episode 9 done\n", + "Episode 10 done\n", + "Episode 11 done\n", + "Episode 12 done\n", + "Episode 13 done\n", + "Episode 14 done\n", + "Episode 14 done\n", + "Total steps: 13\n", + "Episode 15 done\n", + "Episode 16 done\n", + "Episode 17 done\n", + "Episode 18 done\n", + "Episode 18 done\n", + "Total steps: 1\n", + "Episode 19 done\n", + "Episode 20 done\n", + "Episode 21 done\n", + "Episode 22 done\n", + "Episode 23 done\n", + "Episode 24 done\n", + "Episode 25 done\n", + "Episode 26 done\n", + "Episode 26 done\n", + "Total steps: 13\n", + "Episode 27 done\n", + "Episode 28 done\n", + "Episode 28 done\n", + "Total steps: 13\n", + "Episode 29 done\n", + "Episode 30 done\n", + "Episode 31 done\n", + "Episode 31 done\n", + "Total steps: 13\n", + "Episode 32 done\n", + "Episode 32 done\n", + "Total steps: 13\n", + "Episode 33 done\n", + "Episode 33 done\n", + "Total steps: 1\n", + "Episode 34 done\n", + "Episode 35 done\n", + "Episode 36 done\n", + "Episode 37 done\n", + "Episode 38 done\n", + "Episode 39 done\n", + "Episode 40 done\n", + "Episode 41 done\n", + "Episode 41 done\n", + "Total steps: 13\n", + "Episode 42 done\n", + "Episode 43 done\n", + "Episode 44 done\n", + "Episode 45 done\n", + "Episode 46 done\n", + "Episode 46 done\n", + "Total steps: 13\n", + "Episode 47 done\n", + "Episode 48 done\n", + "Episode 48 done\n", + "Total steps: 13\n", + "Episode 49 done\n", + "Episode 49 done\n", + "Total steps: 13\n", + "Episode 50 done\n", + "Episode 51 done\n", + "Episode 52 done\n", + "Episode 53 done\n", + "Episode 54 done\n", + "Episode 54 done\n", + "Total steps: 13\n", + "Episode 55 done\n", + "Episode 56 done\n", + "Episode 57 done\n", + "Episode 58 done\n", + "Episode 58 done\n", + "Total steps: 13\n", + "Episode 59 done\n", + "Episode 60 done\n", + "Episode 61 done\n", + "Episode 62 done\n", + "Episode 63 done\n", + "Episode 63 done\n", + "Total steps: 13\n", + "Episode 64 done\n", + "Episode 65 done\n", + "Episode 66 done\n", + "Episode 67 done\n", + "Episode 68 done\n", + "Episode 69 done\n", + "Episode 69 done\n", + "Total steps: 13\n", + "Episode 70 done\n", + "Episode 71 done\n", + "Episode 72 done\n", + "Episode 73 done\n", + "Episode 74 done\n", + "Episode 75 done\n", + "Episode 76 done\n", + "Episode 77 done\n", + "Episode 78 done\n", + "Episode 79 done\n", + "Episode 80 done\n", + "Episode 80 done\n", + "Total steps: 13\n", + "Episode 81 done\n", + "Episode 82 done\n", + "Episode 83 done\n", + "Episode 84 done\n", + "Episode 84 done\n", + "Total steps: 13\n", + "Episode 85 done\n", + "Episode 86 done\n", + "Episode 87 done\n", + "Episode 88 done\n", + "Episode 89 done\n", + "Episode 90 done\n", + "Episode 91 done\n", + "Episode 92 done\n", + "Episode 93 done\n", + "Episode 93 done\n", + "Total steps: 13\n", + "Episode 94 done\n", + "Episode 95 done\n", + "Episode 96 done\n", + "Episode 96 done\n", + "Total steps: 13\n", + "Episode 97 done\n", + "Episode 98 done\n", + "Episode 99 done\n", + "Episode 99 done\n", + "Total steps: 13\n", + "Episode 100 done\n", + "Episode 100 done\n", + "Total steps: 13\n", + "Episode 101 done\n", + "Episode 102 done\n", + "Episode 103 done\n", + "Episode 104 done\n", + "Episode 105 done\n", + "Episode 106 done\n", + "Episode 106 done\n", + "Total steps: 13\n", + "Episode 107 done\n", + "Episode 107 done\n", + "Total steps: 13\n", + "Episode 108 done\n", + "Episode 108 done\n", + "Total steps: 13\n", + "Episode 109 done\n", + "Episode 110 done\n", + "Episode 111 done\n", + "Episode 111 done\n", + "Total steps: 13\n", + "Episode 112 done\n", + "Episode 113 done\n", + "Episode 114 done\n", + "Episode 115 done\n", + "Episode 116 done\n", + "Episode 116 done\n", + "Total steps: 13\n", + "Episode 117 done\n", + "Episode 117 done\n", + "Total steps: 1\n", + "Episode 118 done\n", + "Episode 119 done\n", + "Episode 120 done\n", + "Episode 121 done\n", + "Episode 122 done\n", + "Episode 123 done\n", + "Episode 124 done\n", + "Episode 125 done\n", + "Episode 126 done\n", + "Episode 126 done\n", + "Total steps: 13\n", + "Episode 127 done\n", + "Episode 128 done\n", + "Episode 129 done\n", + "Episode 130 done\n", + "Episode 131 done\n", + "Episode 132 done\n", + "Episode 133 done\n", + "Episode 134 done\n", + "Episode 135 done\n", + "Episode 135 done\n", + "Total steps: 13\n", + "Episode 136 done\n", + "Episode 137 done\n", + "Episode 138 done\n", + "Episode 139 done\n", + "Episode 140 done\n", + "Episode 141 done\n", + "Episode 142 done\n", + "Episode 143 done\n", + "Episode 144 done\n", + "Episode 145 done\n", + "Episode 146 done\n", + "Episode 147 done\n", + "Episode 148 done\n", + "Episode 149 done\n", + "Episode 150 done\n", + "Episode 151 done\n", + "Episode 151 done\n", + "Total steps: 13\n", + "Episode 152 done\n", + "Episode 153 done\n", + "Episode 153 done\n", + "Total steps: 13\n", + "Episode 154 done\n", + "Episode 155 done\n", + "Episode 156 done\n", + "Episode 156 done\n", + "Total steps: 1\n", + "Episode 157 done\n", + "Episode 158 done\n", + "Episode 159 done\n", + "Episode 160 done\n", + "Episode 161 done\n", + "Episode 162 done\n", + "Episode 163 done\n", + "Episode 163 done\n", + "Total steps: 13\n", + "Episode 164 done\n", + "Episode 165 done\n", + "Episode 166 done\n", + "Episode 167 done\n", + "Episode 168 done\n", + "Episode 169 done\n", + "Episode 169 done\n", + "Total steps: 13\n", + "Episode 170 done\n", + "Episode 171 done\n", + "Episode 171 done\n", + "Total steps: 13\n", + "Episode 172 done\n", + "Episode 173 done\n", + "Episode 174 done\n", + "Episode 175 done\n", + "Episode 176 done\n", + "Episode 176 done\n", + "Total steps: 13\n", + "Episode 177 done\n", + "Episode 178 done\n", + "Episode 179 done\n", + "Episode 180 done\n", + "Episode 181 done\n", + "Episode 182 done\n", + "Episode 183 done\n", + "Episode 184 done\n", + "Episode 184 done\n", + "Total steps: 13\n", + "Episode 185 done\n", + "Episode 186 done\n", + "Episode 187 done\n", + "Episode 188 done\n", + "Episode 188 done\n", + "Total steps: 13\n", + "Episode 189 done\n", + "Episode 190 done\n", + "Episode 190 done\n", + "Total steps: 13\n", + "Episode 191 done\n", + "Episode 192 done\n", + "Episode 193 done\n", + "Episode 194 done\n", + "Episode 195 done\n", + "Episode 196 done\n", + "Episode 196 done\n", + "Total steps: 13\n", + "Episode 197 done\n", + "Episode 198 done\n", + "Episode 199 done\n", + "Episode 200 done\n", + "Episode 201 done\n", + "Episode 202 done\n", + "Episode 202 done\n", + "Total steps: 13\n", + "Episode 203 done\n", + "Episode 204 done\n", + "Episode 205 done\n", + "Episode 206 done\n", + "Episode 207 done\n", + "Episode 208 done\n", + "Episode 209 done\n", + "Episode 210 done\n", + "Episode 210 done\n", + "Total steps: 1\n", + "Episode 211 done\n", + "Episode 211 done\n", + "Total steps: 13\n", + "Episode 212 done\n", + "Episode 212 done\n", + "Total steps: 13\n", + "Episode 213 done\n", + "Episode 214 done\n", + "Episode 215 done\n", + "Episode 216 done\n", + "Episode 216 done\n", + "Total steps: 13\n", + "Episode 217 done\n", + "Episode 217 done\n", + "Total steps: 13\n", + "Episode 218 done\n", + "Episode 219 done\n", + "Episode 220 done\n", + "Episode 221 done\n", + "Episode 222 done\n", + "Episode 223 done\n", + "Episode 224 done\n", + "Episode 225 done\n", + "Episode 226 done\n", + "Episode 227 done\n", + "Episode 227 done\n", + "Total steps: 13\n", + "Episode 228 done\n", + "Episode 229 done\n", + "Episode 230 done\n", + "Episode 231 done\n", + "Episode 232 done\n", + "Episode 232 done\n", + "Total steps: 13\n", + "Episode 233 done\n", + "Episode 234 done\n", + "Episode 235 done\n", + "Episode 235 done\n", + "Total steps: 13\n", + "Episode 236 done\n", + "Episode 237 done\n", + "Episode 238 done\n", + "Episode 239 done\n", + "Episode 240 done\n", + "Episode 241 done\n", + "Episode 242 done\n", + "Episode 243 done\n", + "Episode 244 done\n", + "Episode 245 done\n", + "Episode 246 done\n", + "Episode 247 done\n", + "Episode 248 done\n", + "Episode 248 done\n", + "Total steps: 13\n", + "Episode 249 done\n", + "Episode 249 done\n", + "Total steps: 13\n", + "Episode 250 done\n", + "Episode 251 done\n", + "Episode 252 done\n", + "Episode 253 done\n", + "Episode 253 done\n", + "Total steps: 13\n", + "Episode 254 done\n", + "Episode 255 done\n", + "Episode 256 done\n", + "Episode 256 done\n", + "Total steps: 13\n", + "Episode 257 done\n", + "Episode 258 done\n", + "Episode 258 done\n", + "Total steps: 13\n", + "Episode 259 done\n", + "Episode 260 done\n", + "Episode 261 done\n", + "Episode 262 done\n", + "Episode 263 done\n", + "Episode 264 done\n", + "Episode 264 done\n", + "Total steps: 13\n", + "Episode 265 done\n", + "Episode 266 done\n", + "Episode 267 done\n", + "Episode 268 done\n", + "Episode 269 done\n", + "Episode 270 done\n", + "Episode 271 done\n", + "Episode 271 done\n", + "Total steps: 13\n", + "Episode 272 done\n", + "Episode 273 done\n", + "Episode 273 done\n", + "Total steps: 13\n", + "Episode 274 done\n", + "Episode 275 done\n", + "Episode 276 done\n", + "Episode 277 done\n", + "Episode 277 done\n", + "Total steps: 13\n", + "Episode 278 done\n", + "Episode 279 done\n", + "Episode 280 done\n", + "Episode 281 done\n", + "Episode 282 done\n", + "Episode 283 done\n", + "Episode 284 done\n", + "Episode 285 done\n", + "Episode 285 done\n", + "Total steps: 13\n", + "Episode 286 done\n", + "Episode 287 done\n", + "Episode 288 done\n", + "Episode 289 done\n", + "Episode 290 done\n", + "Episode 291 done\n", + "Episode 292 done\n", + "Episode 293 done\n", + "Episode 294 done\n", + "Episode 295 done\n", + "Episode 296 done\n", + "Episode 297 done\n", + "Episode 298 done\n", + "Episode 299 done\n", + "Generation 1/300\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Solving for Nash Equilibrium in Generation 1/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 2/300\n", + "Solving for Nash Equilibrium in Generation 2/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 3/300\n", + "Solving for Nash Equilibrium in Generation 3/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 4/300\n", + "Solving for Nash Equilibrium in Generation 4/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 5/300\n", + "Solving for Nash Equilibrium in Generation 5/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 6/300\n", + "Solving for Nash Equilibrium in Generation 6/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 7/300\n", + "Solving for Nash Equilibrium in Generation 7/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 8/300\n", + "Solving for Nash Equilibrium in Generation 8/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 9/300\n", + "Solving for Nash Equilibrium in Generation 9/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 10/300\n", + "Solving for Nash Equilibrium in Generation 10/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 11/300\n", + "Solving for Nash Equilibrium in Generation 11/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 12/300\n", + "Solving for Nash Equilibrium in Generation 12/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 13/300\n", + "Solving for Nash Equilibrium in Generation 13/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 14/300\n", + "Solving for Nash Equilibrium in Generation 14/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 15/300\n", + "Solving for Nash Equilibrium in Generation 15/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 16/300\n", + "Solving for Nash Equilibrium in Generation 16/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 17/300\n", + "Solving for Nash Equilibrium in Generation 17/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 18/300\n", + "Solving for Nash Equilibrium in Generation 18/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 19/300\n", + "Solving for Nash Equilibrium in Generation 19/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 20/300\n", + "Solving for Nash Equilibrium in Generation 20/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 21/300\n", + "Solving for Nash Equilibrium in Generation 21/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 22/300\n", + "Solving for Nash Equilibrium in Generation 22/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 23/300\n", + "Solving for Nash Equilibrium in Generation 23/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 24/300\n", + "Solving for Nash Equilibrium in Generation 24/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 25/300\n", + "Solving for Nash Equilibrium in Generation 25/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 26/300\n", + "Solving for Nash Equilibrium in Generation 26/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 27/300\n", + "Solving for Nash Equilibrium in Generation 27/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 28/300\n", + "Solving for Nash Equilibrium in Generation 28/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 29/300\n", + "Solving for Nash Equilibrium in Generation 29/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 30/300\n", + "Solving for Nash Equilibrium in Generation 30/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 31/300\n", + "Solving for Nash Equilibrium in Generation 31/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 32/300\n", + "Solving for Nash Equilibrium in Generation 32/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 33/300\n", + "Solving for Nash Equilibrium in Generation 33/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 34/300\n", + "Solving for Nash Equilibrium in Generation 34/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 35/300\n", + "Solving for Nash Equilibrium in Generation 35/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 36/300\n", + "Solving for Nash Equilibrium in Generation 36/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 37/300\n", + "Solving for Nash Equilibrium in Generation 37/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 38/300\n", + "Solving for Nash Equilibrium in Generation 38/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 39/300\n", + "Solving for Nash Equilibrium in Generation 39/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 40/300\n", + "Solving for Nash Equilibrium in Generation 40/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 41/300\n", + "Solving for Nash Equilibrium in Generation 41/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 42/300\n", + "Solving for Nash Equilibrium in Generation 42/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 43/300\n", + "Solving for Nash Equilibrium in Generation 43/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 44/300\n", + "Solving for Nash Equilibrium in Generation 44/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 45/300\n", + "Solving for Nash Equilibrium in Generation 45/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 46/300\n", + "Solving for Nash Equilibrium in Generation 46/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 47/300\n", + "Solving for Nash Equilibrium in Generation 47/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 48/300\n", + "Solving for Nash Equilibrium in Generation 48/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 49/300\n", + "Solving for Nash Equilibrium in Generation 49/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 50/300\n", + "Solving for Nash Equilibrium in Generation 50/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 51/300\n", + "Solving for Nash Equilibrium in Generation 51/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 52/300\n", + "Solving for Nash Equilibrium in Generation 52/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 53/300\n", + "Solving for Nash Equilibrium in Generation 53/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 54/300\n", + "Solving for Nash Equilibrium in Generation 54/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 55/300\n", + "Solving for Nash Equilibrium in Generation 55/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 56/300\n", + "Solving for Nash Equilibrium in Generation 56/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 57/300\n", + "Solving for Nash Equilibrium in Generation 57/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 58/300\n", + "Solving for Nash Equilibrium in Generation 58/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 59/300\n", + "Solving for Nash Equilibrium in Generation 59/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 60/300\n", + "Solving for Nash Equilibrium in Generation 60/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 61/300\n", + "Solving for Nash Equilibrium in Generation 61/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 62/300\n", + "Solving for Nash Equilibrium in Generation 62/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 63/300\n", + "Solving for Nash Equilibrium in Generation 63/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 64/300\n", + "Solving for Nash Equilibrium in Generation 64/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 65/300\n", + "Solving for Nash Equilibrium in Generation 65/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 66/300\n", + "Solving for Nash Equilibrium in Generation 66/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 67/300\n", + "Solving for Nash Equilibrium in Generation 67/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 68/300\n", + "Solving for Nash Equilibrium in Generation 68/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 69/300\n", + "Solving for Nash Equilibrium in Generation 69/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 70/300\n", + "Solving for Nash Equilibrium in Generation 70/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 71/300\n", + "Solving for Nash Equilibrium in Generation 71/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 72/300\n", + "Solving for Nash Equilibrium in Generation 72/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 73/300\n", + "Solving for Nash Equilibrium in Generation 73/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 74/300\n", + "Solving for Nash Equilibrium in Generation 74/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 75/300\n", + "Solving for Nash Equilibrium in Generation 75/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 76/300\n", + "Solving for Nash Equilibrium in Generation 76/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 77/300\n", + "Solving for Nash Equilibrium in Generation 77/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 78/300\n", + "Solving for Nash Equilibrium in Generation 78/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 79/300\n", + "Solving for Nash Equilibrium in Generation 79/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 80/300\n", + "Solving for Nash Equilibrium in Generation 80/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 81/300\n", + "Solving for Nash Equilibrium in Generation 81/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 82/300\n", + "Solving for Nash Equilibrium in Generation 82/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 83/300\n", + "Solving for Nash Equilibrium in Generation 83/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 84/300\n", + "Solving for Nash Equilibrium in Generation 84/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 85/300\n", + "Solving for Nash Equilibrium in Generation 85/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 86/300\n", + "Solving for Nash Equilibrium in Generation 86/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 87/300\n", + "Solving for Nash Equilibrium in Generation 87/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 88/300\n", + "Solving for Nash Equilibrium in Generation 88/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 89/300\n", + "Solving for Nash Equilibrium in Generation 89/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 90/300\n", + "Solving for Nash Equilibrium in Generation 90/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 91/300\n", + "Solving for Nash Equilibrium in Generation 91/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 92/300\n", + "Solving for Nash Equilibrium in Generation 92/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 93/300\n", + "Solving for Nash Equilibrium in Generation 93/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 94/300\n", + "Solving for Nash Equilibrium in Generation 94/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 95/300\n", + "Solving for Nash Equilibrium in Generation 95/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 96/300\n", + "Solving for Nash Equilibrium in Generation 96/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 97/300\n", + "Solving for Nash Equilibrium in Generation 97/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 98/300\n", + "Solving for Nash Equilibrium in Generation 98/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 99/300\n", + "Solving for Nash Equilibrium in Generation 99/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 100/300\n", + "Solving for Nash Equilibrium in Generation 100/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 101/300\n", + "Solving for Nash Equilibrium in Generation 101/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 102/300\n", + "Solving for Nash Equilibrium in Generation 102/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 103/300\n", + "Solving for Nash Equilibrium in Generation 103/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 104/300\n", + "Solving for Nash Equilibrium in Generation 104/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 105/300\n", + "Solving for Nash Equilibrium in Generation 105/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 106/300\n", + "Solving for Nash Equilibrium in Generation 106/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 107/300\n", + "Solving for Nash Equilibrium in Generation 107/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 108/300\n", + "Solving for Nash Equilibrium in Generation 108/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 109/300\n", + "Solving for Nash Equilibrium in Generation 109/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 110/300\n", + "Solving for Nash Equilibrium in Generation 110/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 111/300\n", + "Solving for Nash Equilibrium in Generation 111/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 112/300\n", + "Solving for Nash Equilibrium in Generation 112/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 113/300\n", + "Solving for Nash Equilibrium in Generation 113/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 114/300\n", + "Solving for Nash Equilibrium in Generation 114/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 115/300\n", + "Solving for Nash Equilibrium in Generation 115/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 116/300\n", + "Solving for Nash Equilibrium in Generation 116/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 117/300\n", + "Solving for Nash Equilibrium in Generation 117/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 118/300\n", + "Solving for Nash Equilibrium in Generation 118/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 119/300\n", + "Solving for Nash Equilibrium in Generation 119/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 120/300\n", + "Solving for Nash Equilibrium in Generation 120/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 121/300\n", + "Solving for Nash Equilibrium in Generation 121/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 122/300\n", + "Solving for Nash Equilibrium in Generation 122/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 123/300\n", + "Solving for Nash Equilibrium in Generation 123/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 124/300\n", + "Solving for Nash Equilibrium in Generation 124/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 125/300\n", + "Solving for Nash Equilibrium in Generation 125/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 126/300\n", + "Solving for Nash Equilibrium in Generation 126/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 127/300\n", + "Solving for Nash Equilibrium in Generation 127/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 128/300\n", + "Solving for Nash Equilibrium in Generation 128/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 129/300\n", + "Solving for Nash Equilibrium in Generation 129/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 130/300\n", + "Solving for Nash Equilibrium in Generation 130/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 131/300\n", + "Solving for Nash Equilibrium in Generation 131/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 132/300\n", + "Solving for Nash Equilibrium in Generation 132/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 133/300\n", + "Solving for Nash Equilibrium in Generation 133/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 134/300\n", + "Solving for Nash Equilibrium in Generation 134/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 135/300\n", + "Solving for Nash Equilibrium in Generation 135/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 136/300\n", + "Solving for Nash Equilibrium in Generation 136/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 137/300\n", + "Solving for Nash Equilibrium in Generation 137/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 138/300\n", + "Solving for Nash Equilibrium in Generation 138/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 139/300\n", + "Solving for Nash Equilibrium in Generation 139/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 140/300\n", + "Solving for Nash Equilibrium in Generation 140/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 141/300\n", + "Solving for Nash Equilibrium in Generation 141/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 142/300\n", + "Solving for Nash Equilibrium in Generation 142/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 143/300\n", + "Solving for Nash Equilibrium in Generation 143/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 144/300\n", + "Solving for Nash Equilibrium in Generation 144/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 145/300\n", + "Solving for Nash Equilibrium in Generation 145/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 146/300\n", + "Solving for Nash Equilibrium in Generation 146/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 147/300\n", + "Solving for Nash Equilibrium in Generation 147/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 148/300\n", + "Solving for Nash Equilibrium in Generation 148/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 149/300\n", + "Solving for Nash Equilibrium in Generation 149/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 150/300\n", + "Solving for Nash Equilibrium in Generation 150/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 151/300\n", + "Solving for Nash Equilibrium in Generation 151/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 152/300\n", + "Solving for Nash Equilibrium in Generation 152/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 153/300\n", + "Solving for Nash Equilibrium in Generation 153/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 154/300\n", + "Solving for Nash Equilibrium in Generation 154/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 155/300\n", + "Solving for Nash Equilibrium in Generation 155/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 156/300\n", + "Solving for Nash Equilibrium in Generation 156/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 157/300\n", + "Solving for Nash Equilibrium in Generation 157/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 158/300\n", + "Solving for Nash Equilibrium in Generation 158/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 159/300\n", + "Solving for Nash Equilibrium in Generation 159/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 160/300\n", + "Solving for Nash Equilibrium in Generation 160/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 161/300\n", + "Solving for Nash Equilibrium in Generation 161/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 162/300\n", + "Solving for Nash Equilibrium in Generation 162/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 163/300\n", + "Solving for Nash Equilibrium in Generation 163/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 164/300\n", + "Solving for Nash Equilibrium in Generation 164/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 165/300\n", + "Solving for Nash Equilibrium in Generation 165/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 166/300\n", + "Solving for Nash Equilibrium in Generation 166/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 167/300\n", + "Solving for Nash Equilibrium in Generation 167/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 168/300\n", + "Solving for Nash Equilibrium in Generation 168/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 169/300\n", + "Solving for Nash Equilibrium in Generation 169/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 170/300\n", + "Solving for Nash Equilibrium in Generation 170/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 171/300\n", + "Solving for Nash Equilibrium in Generation 171/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 172/300\n", + "Solving for Nash Equilibrium in Generation 172/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 173/300\n", + "Solving for Nash Equilibrium in Generation 173/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 174/300\n", + "Solving for Nash Equilibrium in Generation 174/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 175/300\n", + "Solving for Nash Equilibrium in Generation 175/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 176/300\n", + "Solving for Nash Equilibrium in Generation 176/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 177/300\n", + "Solving for Nash Equilibrium in Generation 177/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 178/300\n", + "Solving for Nash Equilibrium in Generation 178/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 179/300\n", + "Solving for Nash Equilibrium in Generation 179/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 180/300\n", + "Solving for Nash Equilibrium in Generation 180/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 181/300\n", + "Solving for Nash Equilibrium in Generation 181/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 182/300\n", + "Solving for Nash Equilibrium in Generation 182/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 183/300\n", + "Solving for Nash Equilibrium in Generation 183/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 184/300\n", + "Solving for Nash Equilibrium in Generation 184/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 185/300\n", + "Solving for Nash Equilibrium in Generation 185/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 186/300\n", + "Solving for Nash Equilibrium in Generation 186/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 187/300\n", + "Solving for Nash Equilibrium in Generation 187/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 188/300\n", + "Solving for Nash Equilibrium in Generation 188/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 189/300\n", + "Solving for Nash Equilibrium in Generation 189/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 190/300\n", + "Solving for Nash Equilibrium in Generation 190/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 191/300\n", + "Solving for Nash Equilibrium in Generation 191/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 192/300\n", + "Solving for Nash Equilibrium in Generation 192/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 193/300\n", + "Solving for Nash Equilibrium in Generation 193/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 194/300\n", + "Solving for Nash Equilibrium in Generation 194/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 195/300\n", + "Solving for Nash Equilibrium in Generation 195/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 196/300\n", + "Solving for Nash Equilibrium in Generation 196/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 197/300\n", + "Solving for Nash Equilibrium in Generation 197/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 198/300\n", + "Solving for Nash Equilibrium in Generation 198/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 199/300\n", + "Solving for Nash Equilibrium in Generation 199/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 200/300\n", + "Solving for Nash Equilibrium in Generation 200/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 201/300\n", + "Solving for Nash Equilibrium in Generation 201/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 202/300\n", + "Solving for Nash Equilibrium in Generation 202/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 203/300\n", + "Solving for Nash Equilibrium in Generation 203/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 204/300\n", + "Solving for Nash Equilibrium in Generation 204/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 205/300\n", + "Solving for Nash Equilibrium in Generation 205/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 206/300\n", + "Solving for Nash Equilibrium in Generation 206/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 207/300\n", + "Solving for Nash Equilibrium in Generation 207/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 208/300\n", + "Solving for Nash Equilibrium in Generation 208/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 209/300\n", + "Solving for Nash Equilibrium in Generation 209/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 210/300\n", + "Solving for Nash Equilibrium in Generation 210/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 211/300\n", + "Solving for Nash Equilibrium in Generation 211/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 212/300\n", + "Solving for Nash Equilibrium in Generation 212/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 213/300\n", + "Solving for Nash Equilibrium in Generation 213/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 214/300\n", + "Solving for Nash Equilibrium in Generation 214/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 215/300\n", + "Solving for Nash Equilibrium in Generation 215/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 216/300\n", + "Solving for Nash Equilibrium in Generation 216/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 217/300\n", + "Solving for Nash Equilibrium in Generation 217/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 218/300\n", + "Solving for Nash Equilibrium in Generation 218/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 219/300\n", + "Solving for Nash Equilibrium in Generation 219/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 220/300\n", + "Solving for Nash Equilibrium in Generation 220/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 221/300\n", + "Solving for Nash Equilibrium in Generation 221/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 222/300\n", + "Solving for Nash Equilibrium in Generation 222/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 223/300\n", + "Solving for Nash Equilibrium in Generation 223/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 224/300\n", + "Solving for Nash Equilibrium in Generation 224/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 225/300\n", + "Solving for Nash Equilibrium in Generation 225/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 226/300\n", + "Solving for Nash Equilibrium in Generation 226/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 227/300\n", + "Solving for Nash Equilibrium in Generation 227/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 228/300\n", + "Solving for Nash Equilibrium in Generation 228/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 229/300\n", + "Solving for Nash Equilibrium in Generation 229/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 230/300\n", + "Solving for Nash Equilibrium in Generation 230/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 231/300\n", + "Solving for Nash Equilibrium in Generation 231/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 232/300\n", + "Solving for Nash Equilibrium in Generation 232/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 233/300\n", + "Solving for Nash Equilibrium in Generation 233/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 234/300\n", + "Solving for Nash Equilibrium in Generation 234/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 235/300\n", + "Solving for Nash Equilibrium in Generation 235/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 236/300\n", + "Solving for Nash Equilibrium in Generation 236/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 237/300\n", + "Solving for Nash Equilibrium in Generation 237/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 238/300\n", + "Solving for Nash Equilibrium in Generation 238/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 239/300\n", + "Solving for Nash Equilibrium in Generation 239/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 240/300\n", + "Solving for Nash Equilibrium in Generation 240/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 241/300\n", + "Solving for Nash Equilibrium in Generation 241/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 242/300\n", + "Solving for Nash Equilibrium in Generation 242/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 243/300\n", + "Solving for Nash Equilibrium in Generation 243/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 244/300\n", + "Solving for Nash Equilibrium in Generation 244/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 245/300\n", + "Solving for Nash Equilibrium in Generation 245/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 246/300\n", + "Solving for Nash Equilibrium in Generation 246/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 247/300\n", + "Solving for Nash Equilibrium in Generation 247/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 248/300\n", + "Solving for Nash Equilibrium in Generation 248/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 249/300\n", + "Solving for Nash Equilibrium in Generation 249/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 250/300\n", + "Solving for Nash Equilibrium in Generation 250/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 251/300\n", + "Solving for Nash Equilibrium in Generation 251/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 252/300\n", + "Solving for Nash Equilibrium in Generation 252/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 253/300\n", + "Solving for Nash Equilibrium in Generation 253/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 254/300\n", + "Solving for Nash Equilibrium in Generation 254/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 255/300\n", + "Solving for Nash Equilibrium in Generation 255/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 256/300\n", + "Solving for Nash Equilibrium in Generation 256/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 257/300\n", + "Solving for Nash Equilibrium in Generation 257/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 258/300\n", + "Solving for Nash Equilibrium in Generation 258/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 259/300\n", + "Solving for Nash Equilibrium in Generation 259/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 260/300\n", + "Solving for Nash Equilibrium in Generation 260/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 261/300\n", + "Solving for Nash Equilibrium in Generation 261/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 262/300\n", + "Solving for Nash Equilibrium in Generation 262/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 263/300\n", + "Solving for Nash Equilibrium in Generation 263/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 264/300\n", + "Solving for Nash Equilibrium in Generation 264/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 265/300\n", + "Solving for Nash Equilibrium in Generation 265/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 266/300\n", + "Solving for Nash Equilibrium in Generation 266/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 267/300\n", + "Solving for Nash Equilibrium in Generation 267/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 268/300\n", + "Solving for Nash Equilibrium in Generation 268/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 269/300\n", + "Solving for Nash Equilibrium in Generation 269/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 270/300\n", + "Solving for Nash Equilibrium in Generation 270/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 271/300\n", + "Solving for Nash Equilibrium in Generation 271/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 272/300\n", + "Solving for Nash Equilibrium in Generation 272/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 273/300\n", + "Solving for Nash Equilibrium in Generation 273/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 274/300\n", + "Solving for Nash Equilibrium in Generation 274/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 275/300\n", + "Solving for Nash Equilibrium in Generation 275/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 276/300\n", + "Solving for Nash Equilibrium in Generation 276/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 277/300\n", + "Solving for Nash Equilibrium in Generation 277/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 278/300\n", + "Solving for Nash Equilibrium in Generation 278/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 279/300\n", + "Solving for Nash Equilibrium in Generation 279/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 280/300\n", + "Solving for Nash Equilibrium in Generation 280/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 281/300\n", + "Solving for Nash Equilibrium in Generation 281/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 282/300\n", + "Solving for Nash Equilibrium in Generation 282/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 283/300\n", + "Solving for Nash Equilibrium in Generation 283/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 284/300\n", + "Solving for Nash Equilibrium in Generation 284/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 285/300\n", + "Solving for Nash Equilibrium in Generation 285/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 286/300\n", + "Solving for Nash Equilibrium in Generation 286/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 287/300\n", + "Solving for Nash Equilibrium in Generation 287/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 288/300\n", + "Solving for Nash Equilibrium in Generation 288/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 289/300\n", + "Solving for Nash Equilibrium in Generation 289/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 290/300\n", + "Solving for Nash Equilibrium in Generation 290/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 291/300\n", + "Solving for Nash Equilibrium in Generation 291/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 292/300\n", + "Solving for Nash Equilibrium in Generation 292/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 293/300\n", + "Solving for Nash Equilibrium in Generation 293/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 294/300\n", + "Solving for Nash Equilibrium in Generation 294/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 295/300\n", + "Solving for Nash Equilibrium in Generation 295/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 296/300\n", + "Solving for Nash Equilibrium in Generation 296/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 297/300\n", + "Solving for Nash Equilibrium in Generation 297/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 298/300\n", + "Solving for Nash Equilibrium in Generation 298/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 299/300\n", + "Solving for Nash Equilibrium in Generation 299/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 300/300\n", + "Solving for Nash Equilibrium in Generation 300/300\n", + "Computing Nash Equilibrium for 16 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Running Computed Policies\n", + "Episode 0 done\n", + "Episode 1 done\n", + "Episode 2 done\n", + "Episode 3 done\n", + "Episode 4 done\n", + "Episode 5 done\n", + "Episode 6 done\n", + "Episode 6 done\n", + "Total steps: 13\n", + "Episode 7 done\n", + "Episode 7 done\n", + "Total steps: 13\n", + "Episode 8 done\n", + "Episode 8 done\n", + "Total steps: 13\n", + "Episode 9 done\n", + "Episode 10 done\n", + "Episode 11 done\n", + "Episode 12 done\n", + "Episode 13 done\n", + "Episode 14 done\n", + "Episode 15 done\n", + "Episode 16 done\n", + "Episode 17 done\n", + "Episode 17 done\n", + "Total steps: 13\n", + "Episode 18 done\n", + "Episode 19 done\n", + "Episode 20 done\n", + "Episode 20 done\n", + "Total steps: 13\n", + "Episode 21 done\n", + "Episode 22 done\n", + "Episode 23 done\n", + "Episode 23 done\n", + "Total steps: 13\n", + "Episode 24 done\n", + "Episode 25 done\n", + "Episode 26 done\n", + "Episode 26 done\n", + "Total steps: 13\n", + "Episode 27 done\n", + "Episode 27 done\n", + "Total steps: 1\n", + "Episode 28 done\n", + "Episode 29 done\n", + "Episode 30 done\n", + "Episode 31 done\n", + "Episode 31 done\n", + "Total steps: 13\n", + "Episode 32 done\n", + "Episode 32 done\n", + "Total steps: 13\n", + "Episode 33 done\n", + "Episode 34 done\n", + "Episode 35 done\n", + "Episode 36 done\n", + "Episode 37 done\n", + "Episode 38 done\n", + "Episode 39 done\n", + "Episode 40 done\n", + "Episode 41 done\n", + "Episode 42 done\n", + "Episode 43 done\n", + "Episode 44 done\n", + "Episode 45 done\n", + "Episode 46 done\n", + "Episode 47 done\n", + "Episode 48 done\n", + "Episode 48 done\n", + "Total steps: 13\n", + "Episode 49 done\n", + "Episode 50 done\n", + "Episode 51 done\n", + "Episode 52 done\n", + "Episode 52 done\n", + "Total steps: 13\n", + "Episode 53 done\n", + "Episode 53 done\n", + "Total steps: 13\n", + "Episode 54 done\n", + "Episode 55 done\n", + "Episode 56 done\n", + "Episode 57 done\n", + "Episode 58 done\n", + "Episode 59 done\n", + "Episode 60 done\n", + "Episode 61 done\n", + "Episode 62 done\n", + "Episode 63 done\n", + "Episode 64 done\n", + "Episode 65 done\n", + "Episode 66 done\n", + "Episode 67 done\n", + "Episode 68 done\n", + "Episode 69 done\n", + "Episode 70 done\n", + "Episode 71 done\n", + "Episode 72 done\n", + "Episode 73 done\n", + "Episode 74 done\n", + "Episode 75 done\n", + "Episode 76 done\n", + "Episode 77 done\n", + "Episode 78 done\n", + "Episode 79 done\n", + "Episode 80 done\n", + "Episode 81 done\n", + "Episode 81 done\n", + "Total steps: 13\n", + "Episode 82 done\n", + "Episode 83 done\n", + "Episode 84 done\n", + "Episode 85 done\n", + "Episode 86 done\n", + "Episode 87 done\n", + "Episode 88 done\n", + "Episode 89 done\n", + "Episode 90 done\n", + "Episode 91 done\n", + "Episode 92 done\n", + "Episode 93 done\n", + "Episode 94 done\n", + "Episode 94 done\n", + "Total steps: 13\n", + "Episode 95 done\n", + "Episode 95 done\n", + "Total steps: 13\n", + "Episode 96 done\n", + "Episode 97 done\n", + "Episode 97 done\n", + "Total steps: 13\n", + "Episode 98 done\n", + "Episode 99 done\n", + "Episode 99 done\n", + "Total steps: 13\n", + "Episode 100 done\n", + "Episode 101 done\n", + "Episode 102 done\n", + "Episode 103 done\n", + "Episode 104 done\n", + "Episode 105 done\n", + "Episode 106 done\n", + "Episode 107 done\n", + "Episode 108 done\n", + "Episode 109 done\n", + "Episode 110 done\n", + "Episode 110 done\n", + "Total steps: 13\n", + "Episode 111 done\n", + "Episode 112 done\n", + "Episode 113 done\n", + "Episode 114 done\n", + "Episode 114 done\n", + "Total steps: 13\n", + "Episode 115 done\n", + "Episode 116 done\n", + "Episode 117 done\n", + "Episode 117 done\n", + "Total steps: 13\n", + "Episode 118 done\n", + "Episode 119 done\n", + "Episode 119 done\n", + "Total steps: 13\n", + "Episode 120 done\n", + "Episode 121 done\n", + "Episode 121 done\n", + "Total steps: 13\n", + "Episode 122 done\n", + "Episode 123 done\n", + "Episode 124 done\n", + "Episode 125 done\n", + "Episode 126 done\n", + "Episode 127 done\n", + "Episode 128 done\n", + "Episode 129 done\n", + "Episode 130 done\n", + "Episode 131 done\n", + "Episode 132 done\n", + "Episode 133 done\n", + "Episode 133 done\n", + "Total steps: 13\n", + "Episode 134 done\n", + "Episode 135 done\n", + "Episode 135 done\n", + "Total steps: 13\n", + "Episode 136 done\n", + "Episode 137 done\n", + "Episode 137 done\n", + "Total steps: 13\n", + "Episode 138 done\n", + "Episode 139 done\n", + "Episode 140 done\n", + "Episode 141 done\n", + "Episode 142 done\n", + "Episode 142 done\n", + "Total steps: 13\n", + "Episode 143 done\n", + "Episode 144 done\n", + "Episode 145 done\n", + "Episode 146 done\n", + "Episode 147 done\n", + "Episode 148 done\n", + "Episode 148 done\n", + "Total steps: 13\n", + "Episode 149 done\n", + "Episode 150 done\n", + "Episode 150 done\n", + "Total steps: 13\n", + "Episode 151 done\n", + "Episode 152 done\n", + "Episode 153 done\n", + "Episode 153 done\n", + "Total steps: 13\n", + "Episode 154 done\n", + "Episode 155 done\n", + "Episode 156 done\n", + "Episode 157 done\n", + "Episode 158 done\n", + "Episode 159 done\n", + "Episode 160 done\n", + "Episode 161 done\n", + "Episode 162 done\n", + "Episode 163 done\n", + "Episode 164 done\n", + "Episode 164 done\n", + "Total steps: 13\n", + "Episode 165 done\n", + "Episode 166 done\n", + "Episode 167 done\n", + "Episode 168 done\n", + "Episode 168 done\n", + "Total steps: 13\n", + "Episode 169 done\n", + "Episode 170 done\n", + "Episode 171 done\n", + "Episode 172 done\n", + "Episode 173 done\n", + "Episode 174 done\n", + "Episode 175 done\n", + "Episode 176 done\n", + "Episode 176 done\n", + "Total steps: 13\n", + "Episode 177 done\n", + "Episode 178 done\n", + "Episode 179 done\n", + "Episode 180 done\n", + "Episode 181 done\n", + "Episode 182 done\n", + "Episode 183 done\n", + "Episode 183 done\n", + "Total steps: 13\n", + "Episode 184 done\n", + "Episode 185 done\n", + "Episode 186 done\n", + "Episode 187 done\n", + "Episode 188 done\n", + "Episode 188 done\n", + "Total steps: 13\n", + "Episode 189 done\n", + "Episode 190 done\n", + "Episode 191 done\n", + "Episode 191 done\n", + "Total steps: 13\n", + "Episode 192 done\n", + "Episode 193 done\n", + "Episode 194 done\n", + "Episode 195 done\n", + "Episode 195 done\n", + "Total steps: 13\n", + "Episode 196 done\n", + "Episode 196 done\n", + "Total steps: 13\n", + "Episode 197 done\n", + "Episode 198 done\n", + "Episode 199 done\n", + "Episode 200 done\n", + "Episode 201 done\n", + "Episode 202 done\n", + "Episode 203 done\n", + "Episode 204 done\n", + "Episode 204 done\n", + "Total steps: 13\n", + "Episode 205 done\n", + "Episode 206 done\n", + "Episode 207 done\n", + "Episode 208 done\n", + "Episode 209 done\n", + "Episode 210 done\n", + "Episode 211 done\n", + "Episode 212 done\n", + "Episode 213 done\n", + "Episode 214 done\n", + "Episode 215 done\n", + "Episode 216 done\n", + "Episode 216 done\n", + "Total steps: 13\n", + "Episode 217 done\n", + "Episode 218 done\n", + "Episode 219 done\n", + "Episode 219 done\n", + "Total steps: 13\n", + "Episode 220 done\n", + "Episode 221 done\n", + "Episode 222 done\n", + "Episode 223 done\n", + "Episode 224 done\n", + "Episode 225 done\n", + "Episode 226 done\n", + "Episode 227 done\n", + "Episode 227 done\n", + "Total steps: 13\n", + "Episode 228 done\n", + "Episode 229 done\n", + "Episode 230 done\n", + "Episode 231 done\n", + "Episode 232 done\n", + "Episode 233 done\n", + "Episode 234 done\n", + "Episode 235 done\n", + "Episode 236 done\n", + "Episode 236 done\n", + "Total steps: 1\n", + "Episode 237 done\n", + "Episode 238 done\n", + "Episode 239 done\n", + "Episode 239 done\n", + "Total steps: 1\n", + "Episode 240 done\n", + "Episode 241 done\n", + "Episode 242 done\n", + "Episode 243 done\n", + "Episode 244 done\n", + "Episode 245 done\n", + "Episode 246 done\n", + "Episode 247 done\n", + "Episode 247 done\n", + "Total steps: 13\n", + "Episode 248 done\n", + "Episode 249 done\n", + "Episode 250 done\n", + "Episode 251 done\n", + "Episode 252 done\n", + "Episode 253 done\n", + "Episode 254 done\n", + "Episode 255 done\n", + "Episode 256 done\n", + "Episode 257 done\n", + "Episode 258 done\n", + "Episode 259 done\n", + "Episode 260 done\n", + "Episode 260 done\n", + "Total steps: 1\n", + "Episode 261 done\n", + "Episode 262 done\n", + "Episode 263 done\n", + "Episode 264 done\n", + "Episode 265 done\n", + "Episode 265 done\n", + "Total steps: 13\n", + "Episode 266 done\n", + "Episode 266 done\n", + "Total steps: 13\n", + "Episode 267 done\n", + "Episode 268 done\n", + "Episode 269 done\n", + "Episode 270 done\n", + "Episode 271 done\n", + "Episode 271 done\n", + "Total steps: 13\n", + "Episode 272 done\n", + "Episode 273 done\n", + "Episode 274 done\n", + "Episode 275 done\n", + "Episode 276 done\n", + "Episode 277 done\n", + "Episode 278 done\n", + "Episode 279 done\n", + "Episode 280 done\n", + "Episode 281 done\n", + "Episode 282 done\n", + "Episode 283 done\n", + "Episode 284 done\n", + "Episode 285 done\n", + "Episode 286 done\n", + "Episode 287 done\n", + "Episode 288 done\n", + "Episode 289 done\n", + "Episode 290 done\n", + "Episode 291 done\n", + "Episode 291 done\n", + "Total steps: 13\n", + "Episode 292 done\n", + "Episode 293 done\n", + "Episode 294 done\n", + "Episode 295 done\n", + "Episode 296 done\n", + "Episode 297 done\n", + "Episode 297 done\n", + "Total steps: 13\n", + "Episode 298 done\n", + "Episode 299 done\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkQAAAHHCAYAAABeLEexAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAACQG0lEQVR4nO3dd3xT9foH8E/SNulMF510UCirjMqm7GUL1oHgFTcqDrCgoBeRexUR9aK4B8L1p9dyryJDxQECFlkCZe8CZRUKtGmBtkn3yvn9kZzTpDMtbU/aft6vV1/a5CT5JqTJc57v832+CkEQBBARERG1YUq5B0BEREQkNwZERERE1OYxICIiIqI2jwERERERtXkMiIiIiKjNY0BEREREbR4DIiIiImrzGBARERFRm8eAiIiIiNo8BkRk0y5dugSFQoH4+PgmfZwOHTrg8ccfb9LHIOudO3cO0dHRcHd3h0KhwM8//yz3kG5ZfHw8FAoFLl26VO/bPv744+jQoYPNjKelu5XXc+HChVAoFI07ILIJDIhIVuKHcnU/r7zyitzDq1ZxcTE+++wzDBs2DJ6enlCpVAgMDMTdd9+N77//HuXl5bKNTa/X44033kBkZCRcXV3h5OSEnj17Yt68eUhLS2uSx/ziiy8aPWCdOnUqTpw4gbfffhv/+9//0L9//0a9/7bkX//6V4sLKNPS0rBw4UIcPXpU7qFQG6LgXmYkp/j4eDzxxBNYtGgRwsLCLK7r2bMnIiMjUVxcDAcHB9jZ2TXZODp06IBRo0bV+cV+/fp1TJgwAYcOHUJMTAxuv/12eHl5QavVYsuWLdi6dSsWLVqE1157rcnGWpOLFy9i3LhxSE1Nxd/+9jcMGzYMKpUKx48fx/fffw8vLy+cPXu20R+3Z8+eaNeuHbZv394o91dYWAhnZ2f885//xFtvvdUo92kLysvLUVpaCrVaXe8Mw+OPP47t27c3KJvj6uqK++67r8p7+1bG09QOHjyIAQMG4JtvvmmSzG1paSkMBgPUanW9b1tWVoaysjI4Ojo2+rhIXvZyD4AIACZMmFBjFsCWPngeffRRHDlyBD/++CMmTZpkcd38+fNx8OBBJCcnN/u4ysrKMGnSJGRkZGD79u0YNmyYxfVvv/023n333WYfV0Ncv34dAODh4VHnsfn5+XBxcWniETUOOzu7Jg3q68vWxnMrCgoK4OzsbPXxDg4ODX4se3t72Nvzq7NVEohk9M033wgAhAMHDlR7fUpKigBA+Oabb6TLpk6dKri4uAhXr14V7rnnHsHFxUVo166d8NJLLwllZWUWt3/vvfeEqKgowcvLS3B0dBT69u0rrF27tsrjhIaGClOnTq11rHv27BEACNOnT6/XcywqKhIWLFggdOrUSVCpVEJQUJAwd+5coaioyOI4AEJcXJywbt06oUePHoJKpRIiIiKEjRs31vkYq1atEgAIb7/9ttXjWrNmjdC3b1/B0dFR8Pb2Fh5++GHh6tWrFsekp6cLjz/+uNC+fXtBpVIJ/v7+wt133y2kpKQIgmB83QBY/IwcOVIQBEEoKSkRFi5cKISHhwtqtVrw8vIShg4dKvzxxx81jun111+vcn+hoaEW1yUlJQkPPvig4OHhIdx2222CIAhCaWmpsGjRIqFjx46CSqUSQkNDhfnz51d5jUNDQ4XY2Fhh27ZtQr9+/QRHR0ehZ8+ewrZt2wRBEIQff/xR6Nmzp6BWq4W+ffsKhw8frvU1PHDggABAiI+Pr3Ldpk2bBADCb7/9JghCxXtdfO1ES5cuFSIiIgSVSiUEBAQIzz33nJCdnW1xzNSpU6XXQWTNe7vyawlAep/fynhGjhwp9OjRQ0hKShJGjRolODk5CYGBgcK7775b5XX49NNPhYiICMHJyUnw8PAQ+vXrJ3z33XfVv6CCIGzbtq3acYufAeJjHzx4UBg+fLjg5OQkvPDCC4IgCMLPP/8s3HHHHUJAQICgUqmEjh07CosWLaryuVD59RQ/Z9577z3h3//+t/Q+6t+/v7B//36L24rvw8qvs7V/u+J7T61WCx07dhSWL19e7X1S82OYSzZBp9Phxo0bFpe1a9euxuPLy8sRExODQYMG4f3338eWLVvwwQcfoFOnTpgxY4Z03CeffIK7774bDz/8MEpKSrBq1Sr87W9/w/r16xEbG1uvMf72228AgEceecTq2xgMBtx9993YtWsXnnnmGXTv3h0nTpzARx99hLNnz1ap7di1axd++uknPPfcc3Bzc8Onn36KyZMnIzU1Fd7e3jU+zq+//grAmMGyhjhVOWDAACxevBgZGRn45JNPsHv3bhw5ckTKzkyePBlJSUmYNWsWOnTogMzMTCQkJCA1NRUdOnTAxx9/jFmzZsHV1RX//Oc/AQB+fn4AjMWnixcvxlNPPYWBAwdCr9fj4MGDOHz4MG6//fZqxzVp0iR4eHhgzpw5ePDBB3HHHXfA1dXV4pi//e1v6Ny5M/71r39BMM34P/XUU1ixYgXuu+8+vPTSS9i3bx8WL16M06dPY926dRa3P3/+PB566CE8++yzeOSRR/D+++/jrrvuwvLly/GPf/wDzz33HABg8eLFuP/++5GcnAylsvpyy/79+6Njx45Ys2YNpk6danHd6tWr4enpiZiYmBr/HRYuXIg33ngD48aNw4wZM5CcnIxly5bhwIED2L17d62ZDGve2//73/+k1/+ZZ54BAHTq1KlRxpOdnY3x48dj0qRJuP/++/HDDz9g3rx56NWrFyZMmAAA+L//+z88//zzuO+++/DCCy+gqKgIx48fx759+/DQQw9VO4bu3btj0aJFWLBgAZ555hkMHz4cADBkyBDpmJs3b2LChAl44IEH8Mgjj0jvufj4eLi6uuLFF1+Eq6srtm7digULFkCv1+O9996r8XmLVq5cidzcXDz77LNQKBRYsmQJJk2ahIsXL9aZVbLmb/fIkSMYP348AgIC8MYbb6C8vByLFi2Cj49PnWOjZiB3REZtm3iWWt2PINScIQIgLFq0yOK++vTpI/Tr18/isoKCAovfS0pKhJ49ewpjxoyxuNyaDNG9994rABBycnIsLi8sLBSuX78u/ZifTf/vf/8TlEql8Ndff1ncZvny5QIAYffu3dJlAASVSiWcP39euuzYsWMCAOGzzz6rdWx9+vQR3N3daz1GVFJSIvj6+go9e/YUCgsLpcvXr18vABAWLFggCIIgZGdnS2fNtenRo4eUFTIXGRkpxMbGWjUmc+Zn6+bEs+gHH3zQ4vKjR48KAISnnnrK4vK///3vAgBh69at0mViRmvPnj3SZZs3bxYACE5OTsLly5ely//9738LAKTsUU3mz58vODg4CFlZWdJlxcXFgoeHh/Dkk09Kl1XOyGRmZgoqlUqIjo4WysvLpeM+//xzAYDwn//8R7qsugyRte9tFxeXat/btzKekSNHCgCE//73vxbP2d/fX5g8ebJ02T333CP06NGjymPXRcy8mf/dV37s5cuXV7mu8msiCILw7LPPCs7OzhbZwpoyRN7e3hb/jr/88otFlk8Qas4QWfO3e9dddwnOzs7CtWvXpMvOnTsn2NvbM0NkA7jKjGzC0qVLkZCQYPFTl+nTp1v8Pnz4cFy8eNHiMicnJ+n/s7OzodPpMHz4cBw+fLjeY9Tr9QBQJWOxfPly+Pj4SD/m9Ttr165F9+7d0a1bN9y4cUP6GTNmDABg27ZtFvc1btw4izP43r17Q6PRVHle1Y3Nzc3Nqudx8OBBZGZm4rnnnrOoz4qNjUW3bt2wYcMGAMbXTqVSYfv27cjOzrbqvs15eHggKSkJ586dq/dta1P53/33338HALz44osWl7/00ksAID0fUUREBKKioqTfBw0aBAAYM2YMQkJCqlxe12s/ZcoUlJaW4qeffpIu++OPP5CTk4MpU6bUeLstW7agpKQEs2fPtshAPf3009BoNFXGXVljvrcbMh5XV1eLbKlKpcLAgQMtXi8PDw9cvXoVBw4caNCYaqJWq/HEE09Uudz8NcnNzcWNGzcwfPhwFBQU4MyZM3Xe75QpU+Dp6Sn9Lman6noPAHX/7ZaXl2PLli2YOHEiAgMDpePCw8OljBrJiwER2YSBAwdi3LhxFj+1cXR0rJJm9vT0rPLFvX79egwePBiOjo7w8vKCj48Pli1bBp1OV+8xigFHXl6exeWTJ0+WgrjevXtbXHfu3DkkJSVZBEw+Pj7o0qULACAzM9PiePMv5NqeV2UajQa5ublWPY/Lly8DALp27Vrlum7duknXq9VqvPvuu9i4cSP8/PwwYsQILFmyBFqt1qrHWbRoEXJyctClSxf06tULc+fOxfHjx626bW0qr0a8fPkylEolwsPDLS739/eHh4eH9HxElV9jd3d3AEBwcHC1l9f12kdGRqJbt25YvXq1dNnq1avRrl07KfCtTk3/DiqVCh07dqwy7soa873dkPEEBQVVWZ1W+b06b948uLq6YuDAgejcuTPi4uKwe/fuBo3PXPv27aFSqapcnpSUhHvvvRfu7u7QaDTw8fGRgjZrXpfK7w0xOLLmhKCuv93MzEwUFhZWeZ8CqPYyan4MiKhFsmZ1zF9//YW7774bjo6O+OKLL/D7778jISEBDz30kFR7Uh/dunUDAJw8edLi8uDgYCmIMz+7BIw1RL169aqS/RJ/xHqVup5XXePt1q0bdDodrly5Ut+nVavZs2fj7NmzWLx4MRwdHfHaa6+he/fuOHLkSJ23HTFiBC5cuID//Oc/6NmzJ7766iv07dsXX3311S2NyTwLYM7apeM1vcYNfe0BY2Zh27ZtuHHjBoqLi/Hrr79i8uTJTbYaqbHf2w1hzevVvXt3JCcnY9WqVRg2bBh+/PFHDBs2DK+//votPXZ174GcnByMHDkSx44dw6JFi/Dbb78hISFBWl1pMBjqvN9beQ/cym3JNjAgolbrxx9/hKOjIzZv3ownn3wSEyZMqDPzVJs777wTAPDdd99ZfZtOnTohKysLY8eOrZIBGzduXLVZmoa46667AADffvttnceGhoYCQLXtAZKTk6XrRZ06dcJLL72EP/74AydPnkRJSQk++OAD6fraAhEvLy888cQT+P7773HlyhX07t0bCxcutOYpWS00NBQGg6HK1FxGRgZycnKqPJ+mMGXKFJSVleHHH3/Exo0bodfr8cADD9R6m5r+HUpKSpCSklLruOvz3rY2ULyV8dTGxcUFU6ZMwTfffIPU1FTExsbi7bffRlFRUY23aUhfpO3bt+PmzZuIj4/HCy+8gDvvvLPakxS5+Pr6wtHREefPn69yXXWXUfNjQEStlp2dHRQKhUXn6EuXLjW4a+/QoUNx++2348svv8Qvv/xS7TGVzwbvv/9+XLt2Df/3f/9X5djCwkLk5+c3aCyV3XfffejVqxfefvttJCYmVrk+NzdXWgXWv39/+Pr6Yvny5SguLpaO2bhxI06fPi2tUCooKKjypdWpUye4ublZ3M7FxQU5OTlVHvPmzZsWv7u6uiI8PNzito3hjjvuAAB8/PHHFpd/+OGHAFDv1YQN0b17d/Tq1QurV6/G6tWrERAQgBEjRtR6m3HjxkGlUuHTTz+1eN98/fXX0Ol0tY67Pu/tmv59GnM8Nan8HlCpVIiIiIAgCCgtLa3xdmJvKWvGLRIzNOZjLykpwRdffFGPETcdOzs7jBs3Dj///LNF1/jz589j48aNMo6MRFx2T61WbGwsPvzwQ4wfPx4PPfQQMjMzsXTpUoSHhze4luXbb7/F+PHjMXHiROms3NPTU+pUvXPnTosCyUcffRRr1qzB9OnTsW3bNgwdOhTl5eU4c+YM1qxZg82bNzfKthQODg746aefMG7cOIwYMQL3338/hg4dCgcHByQlJWHlypXw9PTE22+/DQcHB7z77rt44oknMHLkSDz44IPSsvsOHTpgzpw5AICzZ89i7NixuP/++xEREQF7e3usW7cOGRkZFtmPfv36YdmyZXjrrbcQHh4OX19fjBkzBhERERg1ahT69esHLy8vHDx4ED/88ANmzpx5y8/XXGRkJKZOnYovv/xSmjbZv38/VqxYgYkTJ2L06NGN+ng1mTJlChYsWABHR0dMmzatxqX6Ih8fH8yfPx9vvPEGxo8fj7vvvhvJycn44osvMGDAgFrbO9Tnvd2vXz9s2bIFH374IQIDAxEWFiYVjDfWeGoSHR0Nf39/DB06FH5+fjh9+jQ+//xzxMbG1roIoFOnTvDw8MDy5cvh5uYGFxcXDBo0qEr9mLkhQ4bA09MTU6dOxfPPPw+FQoH//e9/NjVltXDhQvzxxx8YOnQoZsyYgfLycnz++efo2bMntymxBbKsbSMyuZXGjJVVtxz266+/Fjp37iyo1WqhW7duwjfffFPtcdYsuxcVFhYKH3/8sRAVFSVoNBrB3t5e8Pf3F+68807hu+++q9IErqSkRHj33XeFHj16CGq1WvD09BT69esnvPHGG4JOp5OOg6m5W2X1GVt2drawYMECoVevXoKzs7PUeHD+/PlCenq6xbGrV68W+vTpIzVNrNyY8caNG0JcXJzQrVs3wcXFRXB3dxcGDRokrFmzxuJ+tFqtEBsbK7i5uVk0ZnzrrbeEgQMHCh4eHoKTk5PQrVs34e233xZKSkpqfQ51Lbu/fv16lduUlpYKb7zxhhAWFiY4ODgIwcHBtTZmrKy6176mcdTk3LlzUsuIXbt2Vbm+pkaIn3/+udCtWzfBwcFB8PPzE2bMmGFVY0Zr39tnzpwRRowYITg5OVnVmNGa8YjNESurPM5///vfwogRIwRvb29BrVYLnTp1EubOnWvxvq/JL7/8IkREREhL0is3ZqzO7t27hcGDB0uNIl9++WWprYJ5+4TaGjNWBkB4/fXXpd9ra8xYWXV/u3/++afQp08fQaVSCZ06dRK++uor4aWXXhIcHR1rf0GoyXEvMyIiIhlNnDixSVpUUP2whoiIiKiZFBYWWvx+7tw5/P777xg1apQ8AyIJM0RERETNJCAgAI8//rjU22nZsmUoLi7GkSNH0LlzZ7mH16axqJqIiKiZjB8/Ht9//z20Wi3UajWioqLwr3/9i8GQDWCGiIiIiNo81hARERFRm8eAiIiIiNo81hBZyWAwIC0tDW5ubg1qK09ERETNTxAE5ObmIjAwsNaGqQyIrJSWllZlN2wiIiJqGa5cuYKgoKAar2dAZCWxzfyVK1eg0WhkHg0RERFZQ6/XIzg4uNbtYgAGRFYTp8k0Gg0DIiIiohamrnIXFlUTERFRm8eAiIiIiNo8BkRERETU5jEgIiIiojaPARERERG1eQyIiIiIqM1jQERERERtHgMiIiIiavMYEBEREVGbx4CIiIiI2jwGRERERNTmMSAiIiKiNo+bu8osLacQ5QYBAe6OsLdjfEpERCQHfgPLbNR72zF8yTZczyuWeyhERERtFgMimSkUxv8aBHnHQURE1JbJGhAtW7YMvXv3hkajgUajQVRUFDZu3AgAyMrKwqxZs9C1a1c4OTkhJCQEzz//PHQ6ncV9pKamIjY2Fs7OzvD19cXcuXNRVlZmccz27dvRt29fqNVqhIeHIz4+vrmeYp2UpojIwIiIiIhINrLWEAUFBeGdd95B586dIQgCVqxYgXvuuQdHjhyBIAhIS0vD+++/j4iICFy+fBnTp09HWloafvjhBwBAeXk5YmNj4e/vjz179iA9PR2PPfYYHBwc8K9//QsAkJKSgtjYWEyfPh3fffcd/vzzTzz11FMICAhATEyMnE8fAKA0ZYgExkNERESyUQiCbX0Ve3l54b333sO0adOqXLd27Vo88sgjyM/Ph729PTZu3Ig777wTaWlp8PPzAwAsX74c8+bNw/Xr16FSqTBv3jxs2LABJ0+elO7ngQceQE5ODjZt2mT1uPR6Pdzd3aHT6aDRaG79iZr0en0zcovLsP3vo9ChnUuj3S8RERFZ//1tMzVE5eXlWLVqFfLz8xEVFVXtMeKTsbc3JrYSExPRq1cvKRgCgJiYGOj1eiQlJUnHjBs3zuJ+YmJikJiY2ETPpH4qaohsKi4lIiJqU2Rfdn/ixAlERUWhqKgIrq6uWLduHSIiIqocd+PGDbz55pt45plnpMu0Wq1FMARA+l2r1dZ6jF6vR2FhIZycnKodV3FxMYqLK1Z+6fX6hj3BOihNc2YsISIiIpKP7Bmirl274ujRo9i3bx9mzJiBqVOn4tSpUxbH6PV6xMbGIiIiAgsXLmyWcS1evBju7u7ST3BwcJM8jlhUbWMzl0RERG2K7AGRSqVCeHg4+vXrh8WLFyMyMhKffPKJdH1ubi7Gjx8PNzc3rFu3Dg4ODtJ1/v7+yMjIsLg/8Xd/f/9aj9FoNDVmhwBg/vz50Ol00s+VK1du+blWR8ll90RERLKTPSCqzGAwSFNVer0e0dHRUKlU+PXXX+Ho6GhxbFRUFE6cOIHMzEzpsoSEBGg0GmnaLSoqCn/++afF7RISEmqsUxKp1WqpHYD40xQU4rJ7ZoiIiIhkI2sN0fz58zFhwgSEhIQgNzcXK1euxPbt27F582YpGCooKMC3334LvV4v1fH4+PjAzs4O0dHRiIiIwKOPPoolS5ZAq9Xi1VdfRVxcHNRqNQBg+vTp+Pzzz/Hyyy/jySefxNatW7FmzRps2LBBzqcuUbKomoiISHayBkSZmZl47LHHkJ6eDnd3d/Tu3RubN2/G7bffju3bt2Pfvn0AgPDwcIvbpaSkoEOHDrCzs8P69esxY8YMREVFwcXFBVOnTsWiRYukY8PCwrBhwwbMmTMHn3zyCYKCgvDVV1/ZRA8iwLyGSOaBEBERtWE214fIVjVVH6Kh72zFtZxC/DpzKHoHeTTa/RIREVEL7EPUVnEvMyIiIvkxIJKZkkXVREREsmNAJLOKvcwYEBEREcmFAZHMKjJEMg+EiIioDWNAJDOphogRERERkWwYEMmMGSIiIiL5MSCSGfcyIyIikh8DIplx2T0REZH8GBDJjMvuiYiI5MeASGZK078AAyIiIiL5MCCSGfcyIyIikh8DIpkpOGVGREQkOwZEMlOyqJqIiEh2DIhkxqJqIiIi+TEgkhn3MiMiIpIfAyKZKdipmoiISHYMiGRWUUPEiIiIiEguDIhkxr3MiIiI5MeASGbcy4yIiEh+DIhkpuCUGRERkewYEMlMmjIzyDwQIiKiNowBkcxYVE1ERCQ/BkQy415mRERE8mNAJDPuZUZERCQ/BkQy415mRERE8mNAJDPuZUZERCQ/BkQyU5r+BdiHiIiISD4MiGTGvcyIiIjkx4BIZpwyIyIikh8DIpmxqJqIiEh+DIhkxr3MiIiI5MeASGbcy4yIiEh+DIhkpmRRNRERkewYEMmMe5kRERHJjwGRzLiXGRERkfwYEMlM6kPEOTMiIiLZMCCSGZfdExERyY8BkczYmJGIiEh+sgZEy5YtQ+/evaHRaKDRaBAVFYWNGzdK1xcVFSEuLg7e3t5wdXXF5MmTkZGRYXEfqampiI2NhbOzM3x9fTF37lyUlZVZHLN9+3b07dsXarUa4eHhiI+Pb46nZxUxQ8Q+RERERPKRNSAKCgrCO++8g0OHDuHgwYMYM2YM7rnnHiQlJQEA5syZg99++w1r167Fjh07kJaWhkmTJkm3Ly8vR2xsLEpKSrBnzx6sWLEC8fHxWLBggXRMSkoKYmNjMXr0aBw9ehSzZ8/GU089hc2bNzf7860O9zIjIiKSn0KwsdSEl5cX3nvvPdx3333w8fHBypUrcd999wEAzpw5g+7duyMxMRGDBw/Gxo0bceeddyItLQ1+fn4AgOXLl2PevHm4fv06VCoV5s2bhw0bNuDkyZPSYzzwwAPIycnBpk2brB6XXq+Hu7s7dDodNBpNoz3fRb+dwn92p+C5UZ3w8vhujXa/REREZP33t83UEJWXl2PVqlXIz89HVFQUDh06hNLSUowbN046plu3bggJCUFiYiIAIDExEb169ZKCIQCIiYmBXq+XskyJiYkW9yEeI95HTYqLi6HX6y1+mgKLqomIiOQne0B04sQJuLq6Qq1WY/r06Vi3bh0iIiKg1WqhUqng4eFhcbyfnx+0Wi0AQKvVWgRD4vXidbUdo9frUVhYWOO4Fi9eDHd3d+knODj4Vp9qtZRK7mVGREQkN9kDoq5du+Lo0aPYt28fZsyYgalTp+LUqVNyDwvz58+HTqeTfq5cudIkj8O9zIiIiORnL/cAVCoVwsPDAQD9+vXDgQMH8Mknn2DKlCkoKSlBTk6ORZYoIyMD/v7+AAB/f3/s37/f4v7EVWjmx1RemZaRkQGNRgMnJ6cax6VWq6FWq2/5+dWFe5kRERHJT/YMUWUGgwHFxcXo168fHBwc8Oeff0rXJScnIzU1FVFRUQCAqKgonDhxApmZmdIxCQkJ0Gg0iIiIkI4xvw/xGPE+5Ma9zIiIiOQna4Zo/vz5mDBhAkJCQpCbm4uVK1di+/bt2Lx5M9zd3TFt2jS8+OKL8PLygkajwaxZsxAVFYXBgwcDAKKjoxEREYFHH30US5YsgVarxauvvoq4uDgpuzN9+nR8/vnnePnll/Hkk09i69atWLNmDTZs2CDnU5dwLzMiIiL5yRoQZWZm4rHHHkN6ejrc3d3Ru3dvbN68GbfffjsA4KOPPoJSqcTkyZNRXFyMmJgYfPHFF9Lt7ezssH79esyYMQNRUVFwcXHB1KlTsWjRIumYsLAwbNiwAXPmzMEnn3yCoKAgfPXVV4iJiWn251sdBTtVExERyc7m+hDZqqbqQ/TxlrP4eMs5PDI4BG9N7NVo90tEREQtsA9RW8WiaiIiIvkxIJIZ9zIjIiKSHwMimUk1RAaZB0JERNSGMSCSmZJF1URERLJjQCQz7mVGREQkPwZEMqvoQ8SIiIiISC4MiGTGvcyIiIjkx4BIZlx2T0REJD8GRDLjXmZERETyY0AkM6WSe5kRERHJjQGRzLiXGRERkfwYEMmMU2ZERETyY0AkMxZVExERyY8Bkcy4lxkREZH8GBDJTMEMERERkewYEMmMe5kRERHJjwGRzLiXGRERkfwYEMmMe5kRERHJjwGRzLiXGRERkfwYEMlMqiEyyDwQIiKiNowBkcxYVE1ERCQ/BkQyq+hDJO84iIiI2jIGRDLjXmZERETyY0AkM+5lRkREJD8GRDLjXmZERETyY0AkM6XpX4B9iIiIiOTDgEhm3MuMiIhIfgyIZMZl90RERPJjQCQz7mVGREQkPwZEMuNeZkRERPJjQCQz7mVGREQkPwZEMuOyeyIiIvkxIJIZi6qJiIjkx4BIZtzLjIiISH4MiGTGvcyIiIjkx4BIZtzLjIiISH4MiGQm1RAZZB4IERFRGyZrQLR48WIMGDAAbm5u8PX1xcSJE5GcnGxxjFarxaOPPgp/f3+4uLigb9+++PHHHy2OycrKwsMPPwyNRgMPDw9MmzYNeXl5FsccP34cw4cPh6OjI4KDg7FkyZImf37WYB8iIiIi+ckaEO3YsQNxcXHYu3cvEhISUFpaiujoaOTn50vHPPbYY0hOTsavv/6KEydOYNKkSbj//vtx5MgR6ZiHH34YSUlJSEhIwPr167Fz504888wz0vV6vR7R0dEIDQ3FoUOH8N5772HhwoX48ssvm/X5VkfBTtVERETyE2xIZmamAEDYsWOHdJmLi4vw3//+1+I4Ly8v4f/+7/8EQRCEU6dOCQCEAwcOSNdv3LhRUCgUwrVr1wRBEIQvvvhC8PT0FIqLi6Vj5s2bJ3Tt2tXqsel0OgGAoNPpGvTcapJ0TSeEzlsv9H8roVHvl4iIiKz//rapGiKdTgcA8PLyki4bMmQIVq9ejaysLBgMBqxatQpFRUUYNWoUACAxMREeHh7o37+/dJtx48ZBqVRi37590jEjRoyASqWSjomJiUFycjKys7Ob4ZnVTGn6FxA4ZUZERCQbe7kHIDIYDJg9ezaGDh2Knj17SpevWbMGU6ZMgbe3N+zt7eHs7Ix169YhPDwcgLHGyNfX1+K+7O3t4eXlBa1WKx0TFhZmcYyfn590naenZ5XxFBcXo7i4WPpdr9c3zhOthJ2qiYiI5GczGaK4uDicPHkSq1atsrj8tddeQ05ODrZs2YKDBw/ixRdfxP33348TJ0406XgWL14Md3d36Sc4OLhJHofL7omIiORnExmimTNnSsXQQUFB0uUXLlzA559/jpMnT6JHjx4AgMjISPz1119YunQpli9fDn9/f2RmZlrcX1lZGbKysuDv7w8A8Pf3R0ZGhsUx4u/iMZXNnz8fL774ovS7Xq9vkqBIaszIFBEREZFsZM0QCYKAmTNnYt26ddi6dWuVaa2CggIAgFJpOUw7OzsYTI17oqKikJOTg0OHDknXb926FQaDAYMGDZKO2blzJ0pLS6VjEhIS0LVr12qnywBArVZDo9FY/DSFimX3TXL3REREZAVZA6K4uDh8++23WLlyJdzc3KDVaqHValFYWAgA6NatG8LDw/Hss89i//79uHDhAj744AMkJCRg4sSJAIDu3btj/PjxePrpp7F//37s3r0bM2fOxAMPPIDAwEAAwEMPPQSVSoVp06YhKSkJq1evxieffGKRAZKLHbfuICIikp2sAdGyZcug0+kwatQoBAQESD+rV68GADg4OOD333+Hj48P7rrrLvTu3Rv//e9/sWLFCtxxxx3S/Xz33Xfo1q0bxo4dizvuuAPDhg2z6DHk7u6OP/74AykpKejXrx9eeuklLFiwwKJXkVzYh4iIiEh+CoHrva2i1+vh7u4OnU7XqNNn13IKMfSdrVDbK5H81oRGu18iIiKy/vvbZlaZtVXiKjOGpURERPJhQCQzJWuIiIiIZMeASGYK9iEiIiKSHQMimbFTNRERkfwYEMlMDIgA7mdGREQkFwZEMlNWxEPMEhEREcmEAZHMFGYZItYRERERyYMBkcwsM0QMiIiIiOTAgEhmljVEMg6EiIioDWNAJDMlp8yIiIhkx4BIZgoWVRMREcmOAZHMmCEiIiKSHwMimZkXVQsG+cZBRETUljEgkhkzRERERPJjQCQzBZfdExERyY4BkcwUCoXZBq/yjoWIiKitYkBkA8RpM+5lRkREJA8GRDZAyQwRERGRrBgQ2QBxPzPWEBEREcmDAZENqMgQMSAiIiKSAwMiG1BRQyTzQIiIiNooBkQ2QMkpMyIiIlkxILIBXHZPREQkLwZENkDMEKVmFeBUml7m0RAREbU9DIhsgFhUPfU/+zFx6W7oCkrlHRAREVEbw4DIBpjvZ1ZSbsCN/GIZR0NERNT2MCCyAQrzDc0AlJRx23siIqLmxIDIBigt4yEGRERERM2MAZENUFbOEJUzICIiImpODIhsADNERERE8mJAZANYQ0RERCQvBkQ2QFnpX6GYAREREVGzYkBkA1hDREREJC8GRDagckBUXFou00iIiIjaJgZENkBRuaiaGSIiIqJmxYDIBlSZMmMNERERUbNiQGQDuOyeiIhIXgyIbAAzRERERPKSNSBavHgxBgwYADc3N/j6+mLixIlITk6uclxiYiLGjBkDFxcXaDQajBgxAoWFhdL1WVlZePjhh6HRaODh4YFp06YhLy/P4j6OHz+O4cOHw9HREcHBwViyZEmTPz9rVelDxBoiIiKiZiVrQLRjxw7ExcVh7969SEhIQGlpKaKjo5Gfny8dk5iYiPHjxyM6Ohr79+/HgQMHMHPmTCjNmvc8/PDDSEpKQkJCAtavX4+dO3fimWeeka7X6/WIjo5GaGgoDh06hPfeew8LFy7El19+2azPtyacMiMiIpKXQhAEQe5BiK5fvw5fX1/s2LEDI0aMAAAMHjwYt99+O958881qb3P69GlERETgwIED6N+/PwBg06ZNuOOOO3D16lUEBgZi2bJl+Oc//wmtVguVSgUAeOWVV/Dzzz/jzJkzVo1Nr9fD3d0dOp0OGo2mEZ5thbs+24UT13TS748P6YCFd/do1McgIiJqi6z9/rapGiKdzhgUeHl5AQAyMzOxb98++Pr6YsiQIfDz88PIkSOxa9cu6TaJiYnw8PCQgiEAGDduHJRKJfbt2ycdM2LECCkYAoCYmBgkJycjOzu72rEUFxdDr9db/DSVKhkiTpkRERE1K5sJiAwGA2bPno2hQ4eiZ8+eAICLFy8CABYuXIinn34amzZtQt++fTF27FicO3cOAKDVauHr62txX/b29vDy8oJWq5WO8fPzszhG/F08prLFixfD3d1d+gkODm68J1sJ9zIjIiKSl80ERHFxcTh58iRWrVolXWYwGAODZ599Fk888QT69OmDjz76CF27dsV//vOfJh3P/PnzodPppJ8rV6402WOxhoiIiEhe9nIPAABmzpwpFUMHBQVJlwcEBAAAIiIiLI7v3r07UlNTAQD+/v7IzMy0uL6srAxZWVnw9/eXjsnIyLA4RvxdPKYytVoNtVp9C8/Kelx2T0REJC9ZM0SCIGDmzJlYt24dtm7dirCwMIvrO3TogMDAwCpL8c+ePYvQ0FAAQFRUFHJycnDo0CHp+q1bt8JgMGDQoEHSMTt37kRpaal0TEJCArp27QpPT8+menpW4+auRERE8mpQQLRo0SIUFBRUubywsBCLFi2y+n7i4uLw7bffYuXKlXBzc4NWq4VWq5V6DCkUCsydOxeffvopfvjhB5w/fx6vvfYazpw5g2nTpgEwZovGjx+Pp59+Gvv378fu3bsxc+ZMPPDAAwgMDAQAPPTQQ1CpVJg2bRqSkpKwevVqfPLJJ3jxxRcb8vQbXZW9zJghIiIialYNWnZvZ2eH9PT0KsXMN2/ehK+vL8rLrdutvXIxseibb77B448/Lv3+zjvvYOnSpcjKykJkZCSWLFmCYcOGSddnZWVh5syZ+O2336BUKjF58mR8+umncHV1lY45fvw44uLicODAAbRr1w6zZs3CvHnzrH7OTbns/sEv9yLx4k3p9wEdPLF2+pBGfQwiIqK2yNrv7wbVEAmCUG0wc+zYMWnJvLX3Y41XXnkFr7zySo3Xe3l5YeXKlbXeR+/evfHXX39ZPbbmpKyUp2OGiIiIqHnVKyDy9PSEQqGAQqFAly5dLIKi8vJy5OXlYfr06Y0+yNaucg1RMQMiIiKiZlWvgOjjjz+GIAh48skn8cYbb8Dd3V26TqVSoUOHDoiKimr0QbZ23MuMiIhIXvUKiKZOnQoACAsLw9ChQ2FvbxOr9ls89iEiIiKSV4NWmbm5ueH06dPS77/88gsmTpyIf/zjHygpKWm0wbUV7ENEREQkrwYFRM8++yzOnj0LwLi9xpQpU+Ds7Iy1a9fi5ZdfbtQBtgXcy4yIiEheDQqIzp49i9tuuw0AsHbtWowcORIrV65EfHw8fvzxx8YcX5vAvcyIiIjk1aCASBAEaZ+xLVu24I477gAABAcH48aNG403ujaCNURERETyalBA1L9/f7z11lv43//+hx07diA2NhYAkJKSUmVXeapb5RqiMoMAg6He/TKJiIiogRoUEH388cc4fPgwZs6ciX/+858IDw8HAPzwww8YMoQdluurckAEsI6IiIioOTVo3Xzv3r1x4sSJKpe/9957sLOzu+VBtTXV7WBSXGaAowNfSyIiouZwS42EDh06JC2/j4iIQN++fRtlUG1NtRki1hERERE1mwYFRJmZmZgyZQp27NgBDw8PAEBOTg5Gjx6NVatWwcfHpzHH2OpVLqoGOGVGRETUnBpUQzRr1izk5eUhKSkJWVlZyMrKwsmTJ6HX6/H888839hhbveoyRMWl5TKMhIiIqG1qUIZo06ZN2LJlC7p37y5dFhERgaVLlyI6OrrRBtdWVO5DBDBDRERE1JwalCEyGAxwcHCocrmDg4PUn4isV+2UGWuIiIiImk2DAqIxY8bghRdeQFpamnTZtWvXMGfOHIwdO7bRBtdWmE+ZOToY/0kYEBERETWfBgVEn3/+OfR6PTp06IBOnTqhU6dOCAsLg16vx2effdbYY2z1BFQ0YXRVG2cxGRARERE1nwbVEAUHB+Pw4cPYsmULzpw5AwDo3r07xo0b16iDayvKDJYB0Y28EhSzhoiIiKjZ1CtDtHXrVkRERECv10OhUOD222/HrFmzMGvWLAwYMAA9evTAX3/91VRjbbXKys0CIkdmiIiIiJpbvQKijz/+GE8//TQ0Gk2V69zd3fHss8/iww8/bLTBtRXlZhkiZxUDIiIiouZWr4Do2LFjGD9+fI3XR0dH49ChQ7c8qLam1Gx6TG3PomoiIqLmVq+AKCMjo9rl9iJ7e3tcv379lgfV1phniKSAiDVEREREzaZeAVH79u1x8uTJGq8/fvw4AgICbnlQbU2pWUCkYoaIiIio2dUrILrjjjvw2muvoaioqMp1hYWFeP3113HnnXc22uDaijKzbJDKjgERERFRc6vXsvtXX30VP/30E7p06YKZM2eia9euAIAzZ85g6dKlKC8vxz//+c8mGWhrVlZdhohTZkRERM2mXgGRn58f9uzZgxkzZmD+/PkQBOMXuUKhQExMDJYuXQo/P78mGWhrZpEhMgVExcwQERERNZt6N2YMDQ3F77//juzsbJw/fx6CIKBz587w9PRsivG1CeZF1So7OwBAcRl3uyciImouDepUDQCenp4YMGBAY46lzSo1a8yo5l5mREREza5Be5lR4yozsKiaiIhITgyIbEC1RdUMiIiIiJoNAyIbYL6XmZQh4iozIiKiZsOAyAaYF1Xb2ykAWGaNiIiIqGkxILIB5nuZ2SuNAVF5OQMiIiKi5sKAyAaUWWSIlFUuIyIioqbFgMgGvDK+GwBgalQo7JTilBlriIiIiJpLg/sQUeO5f0AwhndpB3+NI34+eg2AZV0RERERNS0GRDYiwN0JAGCnNE2ZsYaIiIio2cg6ZbZ48WIMGDAAbm5u8PX1xcSJE5GcnFztsYIgYMKECVAoFPj5558trktNTUVsbCycnZ3h6+uLuXPnoqyszOKY7du3o2/fvlCr1QgPD0d8fHwTPatbY88pMyIiomYna0C0Y8cOxMXFYe/evUhISEBpaSmio6ORn59f5diPP/4YCoWiyuXl5eWIjY1FSUkJ9uzZgxUrViA+Ph4LFiyQjklJSUFsbCxGjx6No0ePYvbs2XjqqaewefPmJn1+DVEREDFDRERE1FxknTLbtGmTxe/x8fHw9fXFoUOHMGLECOnyo0eP4oMPPsDBgwcREBBgcZs//vgDp06dwpYtW+Dn54fbbrsNb775JubNm4eFCxdCpVJh+fLlCAsLwwcffAAA6N69O3bt2oWPPvoIMTExTf9E60HqQ8QpMyIiomZjU6vMdDodAMDLy0u6rKCgAA899BCWLl0Kf3//KrdJTExEr1694OfnJ10WExMDvV6PpKQk6Zhx48ZZ3C4mJgaJiYlN8TRuiVRDxAwRERFRs7GZomqDwYDZs2dj6NCh6Nmzp3T5nDlzMGTIENxzzz3V3k6r1VoEQwCk37Vaba3H6PV6FBYWwsnJqcr9FhcXo7i4WPpdr9c37InVk4PYmJE1RERERM3GZgKiuLg4nDx5Ert27ZIu+/XXX7F161YcOXKk2cezePFivPHGG83+uFIfIk6ZERERNRubmDKbOXMm1q9fj23btiEoKEi6fOvWrbhw4QI8PDxgb28Pe3tj/DZ58mSMGjUKAODv74+MjAyL+xN/F6fYajpGo9FUmx0CgPnz50On00k/V65caZTnWhd2qiYiImp+smaIBEHArFmzsG7dOmzfvh1hYWEW17/yyit46qmnLC7r1asXPvroI9x1110AgKioKLz99tvIzMyEr68vACAhIQEajQYRERHSMb///rvF/SQkJCAqKqrGsanVaqjV6lt+jvUlrTLjbvdERETNRtaAKC4uDitXrsQvv/wCNzc3qebH3d0dTk5O8Pf3r7aQOiQkRAqeoqOjERERgUcffRRLliyBVqvFq6++iri4OCmgmT59Oj7//HO8/PLLePLJJ7F161asWbMGGzZsaL4nayU7LrsnIiJqdrJOmS1btgw6nQ6jRo1CQECA9LN69Wqr78POzg7r16+HnZ0doqKi8Mgjj+Cxxx7DokWLpGPCwsKwYcMGJCQkIDIyEh988AG++uorm1tyDwAOpikzbt1BRETUfGSfMmuM24SGhlaZEqts1KhRshRn15eYISrllBkREVGzsYmiaqpgLy27Z4aIiIiouTAgsjFSp2oGRERERM2GAZGNsWenaiIiombHgMjGiBmicoPQoBorIiIiqj8GRDZGrCECmCUiIiJqLgyIbIydWUDEwmoiIqLmwYDIxoh9iABmiIiIiJoLAyIbY54h4vYdREREzYMBkY1hDREREVHzY0BkYxQKhZQlYg0RERFR82BAZIO4fQcREVHzYkBkgxyYISIiImpWDIhsUEWGiAERERFRc2BAZIPsTUvvmSEiIiJqHgyIbJC40qzMwBqiunyx/TyW77gg9zCIiCSHLmdje3Km3MOgemJAZIOkgIhTZrW6klWAJZuS8c7GM8gpKJF7OEREKCs34PFv9mPaioNI1xU2y2Mma3Mx5oPt+HjLWZQbBPx46CouXM9rlsduTezlHgBVZWcnZogYENXmwKUs6f/TdUXwcFbJOBoiIuDijXzkFpUBAE6l6RHg7tSkj5dXXIYZ3x7CxRv5+OTPc0jW5mLjSS2CvZyw9aVRFrsfUO34StkgByVriKxhHhBp9UUyjoSIyOh0ul76/zPa3CZ/vLfWn8LFG/kAAEEANp7UAgCuZBVi7cGrTf74rQkDIhtkJ02ZsYaoNvtTzAIiHQMiIpKfeRCU3MQBkcEg4PcT6QCAj6ZEws3ROOnTsZ0LAOCzredQVFrepGNoTRgQ2SBxlRmnzGp2M68YF67nS7+nt8GAKDu/BFvPZEAQ+D4hshXmGaKzGU0bEF26mQ99URnU9krc2TsQXz3WH3NjuuLnmUPhr3FEuq4IO89eb9IxtCYMiGwQV5nV7cClbIvfM9pgQPSPdSfwZPxBKUVORPI7k14RBF24nlfjjgPFZeXQFZTe0mMdu5oDAOjZ3h0OdkoM6uiNuNHh0Dg6YFjndgCApDR9LfdA5hgQ2SA7rjKr05ErxoDITW1MEae3sRqikjIDdpjO/MxrqYhIPtn5JVI9o6ODEqXlAlJu5Fd77GNf78dtb/6BR7/ehytZBQ16vGNXdACA24I9qlwXEaAB0LgB0frjabjzs79w+Wb1z6mlY0BkgxzsuHVHXa5mGZezDgjzAgBom2l5q604kpqNghJjbQDPAInkl5ZTiJ+OXAMABHs5SQFJdYXVRaXl2JeSBUEA/jp3A6/+fLJBj3nkSg4AILKagKhHoPHxzafwbtW3ey/j5DU9fjjUOou1GRDZIGnrDgZENbqWYwyA+oZ4AGi5NUR/nbuON9efQm5R/VLnu8/fkP7/dJqedUREMkrXFWLCJ3/hzfWnAADd/DXo6u8GADhTTUAifn6J9l68We/i5+Kycpw2nQzdFuRR5frupoDoWk5ho/Vpu2I6ET2SmtMo92drGBDZIAdp6w7WENUkTQyIQj0BALlFZcgvLpNzSA3y6s8n8fWuFMxde7xeQc0us4Aot7gMV7PbVoaMyFYYDAJeWnMMusJSeDg7IKydCx4aFII+wcbPph8OXUVhiWWwk3rTOEXWzd8N/hpHFJcZLFbNWuNMei5Kyg3wdHZAsFfVXkcaRweEeDkDMPZDulXFZeVIM2Xij17JaZUzGAyIbBBriKp34qoO0+IP4OQ1Ha7nFQMAuvi5SXVELa0X0ZWsAlw2fTBuStLiP7svWXW73KJSHLtqrB3wdVMD4LQZkVzWHrqCPRduwsnBDuueG4ptfx+F0V19cU+fQAR5OiEztxj/2Z1icZtUU81QiJczRnQxFj9XtxrsWk4hdpy9bhFQFZeVw2AQcDLN+BnQK8gDCoWi2rGJ03anGmHa7Fp2IcRztrzislbZCZsBkQ2qWGXGgMjc9wdS8eeZTHzwRzIEAVDZK+HtooKfuyMA4Nejadh38abMo6zb1jMZeP2Xk/jjVAYAY/ElAMTvSantZpKEUxkoNwjo2M4Fo7r6AABOmT4ciaj5CIKAb0wnMrPHdUaYqf8PAKjt7fD36K4AgOXbLyDPLINtHhAN72z8G/7rXEXWV/TEN/sx9T/70e+tBKw9eAVJaTr0ev0PvLvpjLSarXuAW43jE+uIGuOE6XKlwu8jqdk1HNlyMSCyQfZK9iGqTqYpA7TngjHoCXR3hEKhQIApIPrkz3N4+Kt9Nt+k8c31p7Ei8TLe3XQGAPDgwBAAxvn5m6bMV23WHLwCAJjYp32jngESUf0cu6rDGW0u1PZKTBkQXOX6uyMDEezlhNziMhwwmxITA6JQb2cMC28HhQJIzsi1WG12JasAZzOMWZiCknL8e+dF/HosDSXlBvxyNE1q+tjNv5aAqL3x8+HEtVs/Yaq8Eq411hExILJB0l5m7FRtITPXGCwUlxlfl0AP47y5xtFBOqbMIOCXo9eaf3BW0heVSstwS0zPY0LPAHTyMZ5ZHr9a+wdX6s0C7L2YBYUCmNwvCBGB7gCAk9cYEBE1t+/3pQIA7ugVUO1eikqlAkM6GqfE9pplr8XgItjLGZ4uKgztZDzmi+0XpGP2mQKojj4uUCqA85l5Uldqrb4IR00rzLr6aWocX6Sp2Pp8Zt4t9zwSp/dDvY11SbvO32iRdZu1YUBkg8Qps9ZYtHYrMirVCIkB0eBO3gCAIE/j7+uO2G5AdLpS6tpZZYfbgj2kZbPiMtqa/HDYuNx1WHg7tPdwQo9ADZQK4wek+Pqk3MjH0m3n2bKfqAmVGwSsP54GAHigmuyQaHAnY2uQvaYARxAEiykzAHh+bGcAwNqDV6RgSQygYnr4o2d744mPuMoLAErKDbBTKtDJt2KarjJvVzU6mAIYsXdbQ4lj/lu/IHg4O+BqdiGmf3sIxWWt53OGAZEN4pRZVeUGATfyLJeOBpqmyh4eGIJtfx+F9bOGQWWnxBltbqOsqmgK4lx+uK8r3BztcW+f9lDZK6XGasfqCIjEtPudvQMAAC5qe3TxM6bMxRT22xtO4b3Nyfhy58XGfwJEBAC4nluM/JJy2CkV6N/Bq8bjBoUZT9hOXtMhr7gMN/JKUFBSDoUCaG86iRsY5oXhnduhzCDgs63nAFQERIM7emOIKYNUWScfF6jt7WodZ98Q42q3w7c4xSWujOvZ3h3/eXwAnFV2+OvcDfxyNO2W7teWMCCyQfbc3LWKrPySKhkzMUOkVCoQ1s4FHs4qjO3uCwBYd8Q2G4eJAVFsrwAcfu12vH1vLwAVnWaPXc2Rlt+Xlhvw5vpT2JxUsTXHzXzjtGGQp7N0WR9TL6ajV3JgMAjStiYbjqc36XMhasvEJej+GkdpZXB1Aj2cEOzlhHKDgIOXsqRMS4DG0SKYmT2uCwDgx8PXsPv8DVzNLjQGW6GeGBZeERB5OFeUCHT1r3m6TNTH1JqkoUXQgiBAX1RqkdXqG+KJRweHAjAGeq0FAyIbZG/HVWaVVZ4uAyoCInP39mkPAPj5aJpNBpRJptVgPQI1Ur8pwNjITWWnRE5BxQfPppNafL0rBQt/TZKOu2nKknm5VNQriMHUkdRspNzMh67QWCuQnJGLc028uSQ1n5Qb+bjYCpc6N7UDl7IwLf5Ao283IfZCExd11GawKUv04+Fr2JdizPyEeDtbHNMv1BOjuvqg3CDg2f8dAgD0au8OF7U9+nfwhMre+HkxNaqDdJvaCqpFYvPao6nGE6b6+tfvp9F74R8oNE3BiydjnU2Z6XMZrec9yYDIBrGGqKrruVVXXwV6VP0gGtXVF57ODrieW4zdF5pnCf7Ja7o6C7nPaPX4eleK1Ma/h6kmQKSyVyLCtERWLJYU9ypL1xUhO78EBoOAbFPHWW9X84DIeAZ44poOByvta7aeWaJWoai0HPd8vgtjPtiB1QdS5R5Oi/Lt3sv480wmvt17uVHvNz3HeJJW3YlZZXdGBgIAfjuWhiWbkgFUTKWZm2PKEuUVl8HJwQ4vRRt/d3Sww7MjOqJ/qCeeHBYGb9MJkTUBUVc/Nzir7JBbXIazmfU7QcrKL8GKxIrXrUegRgrMuvi5AgDO1fM+bRkDIhtkZ6ohKmVjRklmrvHDp5cpkLBTKhDgXvWDSGWvxF2mD591hyumzXSFpfh867lqM023Iq+4DI98vQ8vrDoq1f8Ul5Xj3U1nLFLUz39/RGrr7+ZoL9U/mauoI9JBEAQpIAKA01o9cgpLIcbInmYrWsJ9XeGiskNBSTlWHzAuyW9v+pDeeJIBUWuQmlUAfZFxRc+8H09g9Pvb8e6mMzaZBbU1WfnGk4iDlxu3b464/YY1AdHILj747ME+Us+xZ0d0xKwx4VWOiwz2wHOjOiGqozfWxQ2RehQBwEvRXfHDjCFwd3LAwrt74JHBIRjRxafKfVRmb6fEQNOejyv2XLLmqUm+35+KkjIDerbXYOVTg/CfxwdI13XyMQZEN/JKpNe4pWNAZIMqNnflh50oQ2/MEEUEaPDOpF54Z1IvuJg6VFcmTpv9ciwND3yZiLMZuVix5xLe/+MsXlh1pF5bZLzxWxLu/3dijXuNrdhzCTmm5azHruYYH/doGpZtv4CZK4+gtNyA67nFUj8RAIiO8K+2s2xksLt0P6fTcy2yYmfSc6UeRe5ODhbTbXZKhbRKTSycnGn6sD2bkWfREI5aJvMd0+2UCqTcyMey7Rcw47vDXE1YB/HL+uQ1XZUtNG5Fuk4MiOqeMgOAuyIDkTBnJNbPGob5d3SHvV31X78vj++G758ZjG611AfdFRmItyb2svgcqM3M0cbPg9UHruB8pnVTXKXlBvzPlB16YkgYhoS3g5+m4rm6qO2llb1ntHqr79eWMSCyQdLmro2QITp5TYfVB1Jb/OafYobIV6PGAwND8Lf+NS9zvS3YA5P6tIcgAHsvZuHLnRel2p29F7Okxo51ySsuw4o9l7A/Java3Z3zisvw1V8VK7mSTL2AxEzRtZxC/H4iXZrG6urnhn3/GIt/TepZ7eOJPUNOXtPhz9MZFted0epxM7/qdJloZqWzzbHdfeFv+vCqbnNJalnE+pe7IgNx+NXb8f7fIqGyVyLhVAamrThQbT+YD/9Ixis/1m+PvMZQWFIuBQu2QDxhKS0XpJOWxpAmTplVk6muSbCXs7SEvjn17+CF2yP8YBCAj7acteo2J67poNUXwcPZAXdGBlR7TGdfY5boof/bh3Ef7rDpHnDWYEBkgxqzhujOz3Zh3o8nsPu87W9pUZtMU4ZI3LurNgqFAh9OuQ2fPtgHgDHAMM/QvLc52aoviSOp2dIU1f8SL1cpSPz9eDqyC0qlAFbsFm3eXPHLnRex3xQQDQzzgl+llSXmOni7QONoj+IyA+JNqe2ojsY6gzPaXKmg2tulakA0pFM7LH2oL5QKYxbN181RqkliF+uWL+WGsdC+g7cz3J0dcF+/IMQ/MQAuKjvsPn8T01YcsHh/FpWW47Nt57HqwBWL935zmPX9YYxYsg2HLtdvs9KmYj6dU7nG7lZIRdVWZojkFmfKEu08e92q4mrzDWhr+swSW36IvvrLuu2HbJWsAdHixYsxYMAAuLm5wdfXFxMnTkRycrJ0fVZWFmbNmoWuXbvCyckJISEheP7556HTWS7zS01NRWxsLJydneHr64u5c+eirMzyjGn79u3o27cv1Go1wsPDER8f3xxPsUHEVOqtrjIzT6VfzS6o5Ujbl2GaPvLVWP/h08+03PR8Zp50hq1QGIuWxfn/2hy8VFFzcPFGPnZfsNxrSAx07uhlPHtKzshFQUkZzmiNAYi9UoGkND3WHjRmlwaE1dyrBDC2DxCnvm7ml8BZZYe/xxj3QkrW5uK6KUvm7VJ9UBjbOwDb/z4aK58eBMBsY0cb7clE1hPfvx28K5rwDenUDt8+NQgqeyX2Xsyy2GzzWk7FRpzi+7E5FJSUYXvydZSWC3hn4xnZM9NFpeXS6igAUkuKxrhfMWPb3ooaIlvQM1ADRwclcovKcPFG3UGyuNo12NO5xmPCTRki0YlrOpyoo9u+LZM1INqxYwfi4uKwd+9eJCQkoLS0FNHR0cjPN/7xp6WlIS0tDe+//z5OnjyJ+Ph4bNq0CdOmTZPuo7y8HLGxsSgpKcGePXuwYsUKxMfHY8GCBdIxKSkpiI2NxejRo3H06FHMnj0bTz31FDZv3tzsz9kado3Uh8h8Tlft0LKTgddNxdDWZIhEge6O8HB2QJlBgEEAPJ0dpM6w17LrDogOmYowxYzMOxvPoKCkItAWzzbv7RMIN7U9SsoMWH88HaXlAjycHfCc6YxMrOEZWEvzNpE4bQYATw4NQ59gDzg52KG4zIBDpvogr2qmzEQh3s7SFgLMELV83+69jNUHUnHJVEPUoZ3ll1OfEE9EBom1ZxVfRFfN3t+n05tvFdChy9nSidyBS9kWCwPkIK7KFB2+nN0omfd0036JTg52cHdyqONo22Bvp0Rv0+eLNU0ar1Tqpl0d8wyReDK3cn/LXQUp67fkpk2b8Pjjj6NHjx6IjIxEfHw8UlNTceiQsQdDz5498eOPP+Kuu+5Cp06dMGbMGLz99tv47bffpAzQH3/8gVOnTuHbb7/FbbfdhgkTJuDNN9/E0qVLUVJi/GNYvnw5wsLC8MEHH6B79+6YOXMm7rvvPnz00UeyPffaVBRV39ofrrj5HwDkFrXcwlqDQcD1vPpniBQKhbTbM2D84xV7hqTXsQFsWblBWiW25L7e8HJRISlNj7+vPQZBEHA9txiXbhZAoQD6hXqhuykbI67y6tXeHbPGhEsfEsFeTvC3ol9J31Dj8RpHezw9oiOUSgW6mpbW7j5vzFBVN2VWHTFDlKzN5WqkFuhsRi5e/fkk5v14Ammm92uod9VtGsQvueNm9THmG3E2Z4Zo30XjSYLKlOWWewpFnC7zclHBRVx63gi9udJzKgqqq1sgYav6mD6PjtbSEX/98TQkpekqMkS1BES9g9wxbVgYFt4VgfkTugEwthYobaGfNzaVNhCnwry8aj6T1ul00Gg0sLc3rjBKTExEr1694OfnJx0TExMDvV6PpKQk6Zhx48ZZ3E9MTAwSExNrfJzi4mLo9XqLn+Zi10hbd5j3nGjJAZFWX4TScgFKBeDjan2GCKgICgCgq7+bVACZpivEmgNX8Nmf56q93RltLvJLyuGmtseorr5Y/kg/ONgp8PsJLU6l66X6iK5+bnB3cpCyMWJWKTLIAw52Snz2QB8MCvOSVnnUZVQXX8yN6Yp/P9pfOvMUgzrxw93agCjEyxkuKmN2yXyVErUMlffkc1PbV/tv37uaDNEVsynyM82YIRKbDj40KAQAcDJNJ+u0WXa+saDax1WNvqYp9FutIxIEQZqetGbJvS0Ru9rXtFP9vos3MXPlEcz49rCUZawtIFIoFHjtzgg8PjQMAzt4wdtFhbziMhxu5BYHzcVmAiKDwYDZs2dj6NCh6Nmz+lU4N27cwJtvvolnnnlGukyr1VoEQwCk37Vaba3H6PV6FBZWP3WyePFiuLu7Sz/BwTWvamps0tYdt7js/mwryRCJH7I927tLTcGs1SOwYkVHZz83qQDySlYBXv35JD5IOFvtclGxMPq2EA/YKRUYGOYl1SSdSc+VahH6dzBe1qvSypFepi+pEG9nrH42ClMGhFg1XqVSgbjR4YjqVNG0rfJ9e1kZFCqVCilz9eD/7cVdn+3CUysOWFU/RfIyGAT8UikgCm3nXH27BlOG6HSaHiVlxs8M8ykzrd7Y2LOpFZaU49gV49/NQ4NCoFQYV3iJ2d3q7L14s0lXpIlTZh7ODugfajzRvpU6onKDgPuWJ+K1X4wn2/VZYWYL+pj2NUvW6qtdmbjFtLo1NatA+pwI9rLuOSqVCgzrbNxiZOc5eadKG8pmAqK4uDicPHkSq1atqvZ6vV6P2NhYREREYOHChU0+nvnz50On00k/V65cafLHFElbd9zisnvz1SU19dFpCfaYVsiZBwnWijCbMuvq5yY1c0y8cBMlprTu6WpqbMQPUvO2/OJ8+dnMXKnJ2wBTXVBs7wDMHtcZo7r6ILZXAEZa0TDNWpWX6bazMkMEGDeBVSiMzdNOXNNhy+lM/DfxUqONjZrGvpQspOmK4OZYkRXqUM10GQCEejvD3ckBJeUGaZr8apblIooz2qbPEh1JzUZJuQH+Gkd09nWVpvfOaqsv4D16JQcPfLkX07893GRjEv+OvVxUGNCh7gzR6gOp+H5/RZuSE1d1+PvaY1Lmd+fZ6zh0OVtazfngIOtOdGyFn8YRAe6OMAgVJ5pl5QYs3XYeO89ex/Zky0DG0UFZr6z8CFMjyZ1nb9RxpG2qvrNdM5s5cybWr1+PnTt3IigoqMr1ubm5GD9+PNzc3LBu3To4OFQUsfn7+2P//v0Wx2dkZEjXif8VLzM/RqPRwMmp+uhXrVZDra7f9ExjqcgQNTwgyi0qtcgEtOQMkdg3qKYdn2vTsZ0LAtwdkV9chm4BbsgrNgaGl25a1liI3a1FelMA6eZY8V4Te24cv6JDkmlDQzFr5OhgJ23O2Ni6+LnBwU4h9aWqrai6sseHhuHePkFIuZmPP09n4LOt56UAk2zX1jPGz6sJPf3h6+aIz7edl87uK1MoFOgd5I6/zt3Asas56BXkjiumDFGQpxOuZhfijFbfoBOK+th5zvglOKSTNxQKBTr7uiLlRj7OZuRKmQNz+01fyMeu5CBTX1Sv+sCaFJaU467Pd6FjOxd8+Vh/aZrZ00UlZXvTdEW4llNYZXVYVn4J5v14wvhczl7HyC4+eHP9KeSXlOOXo9ew6J6eUn+wJ4eG4dU7I255vHKI6eGP+D2X8PaG0xga3g4/HrqG9zYnQ2WvlDKMomDP6rOSNRlu+nc+mabDzbxieNezxEFusmaIBEHAzJkzsW7dOmzduhVhYWFVjtHr9YiOjoZKpcKvv/4KR0fLP5qoqCicOHECmZmZ0mUJCQnQaDSIiIiQjvnzzz8tbpeQkICoqKgmeFa3rjFqiM5VmgZqqd2Kr5hSt/amXZ/ry95OiR9nDMH6WcOhcXSodruP6mosxADSzbHinEHczDDx4k2UGQQEuDs2y5Jblb1SKqwGLDd2tYa7swNuC/bAI+Lu1Gk65BS0jlb7tqQxa2XEov+u/hrMub0L1jwbhceiQms8XiyW/fnINeQXl0mBQHSE8aRwf0rT9wTaaVpRNrKrMUsgZlTN97oqKi3H/J+OY9PJdJy8VpGZ3XW+cTIKp9J1OJ+Zhz9OZeB6brE0Vejp7ABnlT16mjLGj369D59Wqh803/x140ktXvnpBPJLytHOVYXScgHzfzqBLaeN3zMtLTNkbs64LmjnqsaF6/lYsikZX2w/DwBSMKQy635dW/1QdXw1jujm7wZBaLx/0+Yka0AUFxeHb7/9FitXroSbmxu0Wi20Wq1U1yMGQ/n5+fj666+h1+ulY8rLjb0loqOjERERgUcffRTHjh3D5s2b8eqrryIuLk7K8EyfPh0XL17Eyy+/jDNnzuCLL77AmjVrMGfOHNmee20aY+uOyjs7t4Qps3KDgJfWHMM7G88AAL7bdxnzfjwOwNh9uqatOuoS6OEk7Sxd3Zx/ddMJeVJAVJEhqtyErF+oZ7OtMDGvI/Jyrl9AJPLTOCLc19XUwZtZosZUUFKG0e9vx/PfH2mU+xObcLZzVUk1bLVt0/DAwBA4Oihx8HI2/r3jAgDjFi+T+xm3sUk4lYHMRt7Hz1xmbhFOpeuhUADDwo1Zgs6mzT/Np+43nkzH9/uv4NWfk3DiWkUR+K5zjfPleSWrIit+6HIWsk1dqsW9/8Z1N9aSXryejw8TzlqsxBWzavZKBcb38EefEA9MjQrFrnljMDemK8Q/9aiO3tI+Xi2Ru7MDFt5tTBZ8vSsFV7ML4ensINVnPmQW7NW25L4mYpYosZk2125MsgZEy5Ytg06nw6hRoxAQECD9rF69GgBw+PBh7Nu3DydOnEB4eLjFMWJNj52dHdavXw87OztERUXhkUcewWOPPYZFixZJjxMWFoYNGzYgISEBkZGR+OCDD/DVV18hJiZGluddl8bYuiMrX/wgMH6ht4Qps9Ppevx4+CqW77iAw6nZ+Oe6k9J0mTWbGFpD42QPF5Vl19VrOYXQFVoGjGIA6WYWhHm5qNDObLpqgBV9hRqLWBzu4exQ4x5I1hC/rFri2ZstO3olB5duFuDXY2lV3ksNccNUiNzOyimHQA8nTB/ZCQDw6VbjGX+wlxN6BLqjf6gnygxCo/WHycwtwt+W78GDX+7F8h0XUFZuwF+mmpGege7SNIlUc5eRK2XP9qdkS8/PfOXjX+dvNEqGzbzdwIFL2RY1RIBxi5sNzw+TPk9WmNXTibe9OzIQyx/th3XPDcUb9/SEo4Md4kaH4z9TByA6wg8L7mqZU2Xm7uwdiDfu7iGVZzw7shOWTO6N4Z3bIW50OMLaGeu/xL3K6mOwqbv+vmbISjY2WWuI6voDGDVqlFV/JKGhofj999/rvK8jRxrn7K2p2ZumzG6lD5E4JRLs5YzsAl2LCIjMt7z43PSh3snHBU8N74h7bgus6Wb1olAoEODhJK0sE+fNz2bkWgQ41U2ZAUBnXzfcyDMGaeIKs+YwKMwLCgVu+cx0SCdvxO+5xDqiRma+AOLolZxbLqqvb0AEAM+O6IRfjqZJgYaYDX00KhQHL2dj5b5UxI0Ot3pD0JpsPKGVVmolXrwJO4UCR009kEZ0qagV6ujjAjulArlFZdDqixDg7iTVDYnauaqRV1yK67nFSM7IrXVDU3Nl5Qa8tzkZ4b6uFvsamq+uO3ApS/oMFTNExt5k7pgxshN2nr2OdYevYXLfIHQPcJNuG1RDVmR0N1+M7uZr1fhagqlDOqBviCeOXc3BAwOCYW+nxETTxthTo0Lxf3+l4PYIvzrupar+HbygVBg3JM7QF1lsCGvrbGaVGVVojKLqbLOACGgZU2YnruVI/7/1jHGufkLPADw4MATOqsaL3cWVYyp7JYaYCk0rb4CaW82UGQB0MU0DuKrtrf7wbgyd/dzwS9xQLH+k3y3dz0DT9iEXb+RLheN068yzQodusc9NWblBmuqpbiPfmjip7PDzc0MxpX8w7JQK6ct7Qs8AuDnaIzO3uFF6EokBl7gc+73NydhwPB0AMLZ7xReo2t4OHUxT1afT9biRV4wL1y2n8m8LdpdORPbWY4plc1IG/r3zIv657qTFXmXm/ZeS0vTSwhLPSnV3gzt6oaufGwpLyzF52R5M+OQvpGaZnlcDsiItVa8gdzwyOLRK1vnxoWHY/cqYahuB1sW8L5s1U/PXcgrx/uZkZDThlK61GBDZIDtp2X3Da4jED1RxDjivuEz2fYXqcryaPXAG1rH/V0OIZ84d27lU7PdVJSASV5lZBmI9TLU8Azp4SlObzaV3kAd86rF1SXU8nFUINAWEyc2wFLutMA8uD6XeWlM68QteqajIbFjL3dkB797XG6cXjceDA421ICp7JXqaplyrazFRX2JANH1kJ/QP9ZTaV8wY1Ql9K62EE3/fn5ItLXcPdHeU6nF6BLpLW9ocqEczv9UHjSUTJeUGiwaW5gFRuUGQdrqvXHenUCgw/45u6GTKYl2+WYADpum8+hYSU1WDw4wnmnsv1n1y8OWOC/h823kMeWdro2yrcisYENkgh0aYMtOZPgjEjfkMApBfUl7bTZqdrqAUZ7R6JKXpkFtUWqWlvp1SIS1rb0zivHgXPzepy2/lzq01TZlN6tMei+7pgUX3VN88tCXoZgoCK2fFqOHMM0RHUnNu6WTmRl5F3UtDg+7KDUy71xD4N8Ql04KNju1c8c7k3ogMcscLYzvjZdNGxOYGdRS/GG9K9UNjuvsiynT5oI5e6G8KiA5eyrLqpO1qdgH+Mmv8t+bAFQiCgLJyA9JzjFmGIZVaDHi4VN1vbFRXX/z50ijpWDGwY0B066Q6IisyRCdNm0+XGwT8Z5e8W73YRB8islRRVH0rGaKKxoJ2SgXKDQLyisrg2sCVWo2pyJSmTqpmF3YvFxXU9kqk64rQq717g1eW1ea+/kG4ml2IqUM6SEXSyRm5yC0qhZujAwwGAXkl1U+Z2dsp8VhUh0YfU3Pq5u+GrWcycZoZokZjHhAVlJTjjDa3SkNNazWkfqgujbXRb0mZQSo+7ujjAj+NI36ZOazG4weZMrwnrulwPdf4vAaGeWP2uC44m5GLIZ3aobCkHPZKBTL0xbiaXYhgL2fpZLC6gPCHQ1chCMbNRM+k65GckYsjV3Lg66ZGmUGAg50C7/8tErGf/iVlyt1q+RwZ0MELf5lWudkrFfBvQTUvtkqcBr14Ix85BSXShtPVcTZb5PLeH8kY3c0X4b7yrOJjhsgG2TfC5q5iqtjD2UHKcthKHdHu8zekYMjLRWXxBxHu6yp9iIpnGY0twN0J797XGxGBGvhqHBHs5QRBgLTtQF5JGcQT1coZotZAzBA1xvQJGekrrSyb9+NxJJzKwKaT6fjl6LUqbTBqczO/CQIis3/zW5k6v5JdAINg/BLztWL6NtjLGe09nFBuEHAtpxAezg4Y280X7VzVUqNVJ5WdFDwevJyFTH0R+r6ZgJkrq+9gLQYvDw8KQWzvAADAkk1npCX37T2cEOjhhD/mjMSorj54dkTHWttjmC+mCPRwavap8NbI3dlBqh+rrhTCXLZZT7QHBwQj0EO+gLT1fdq3Ao1ZVO3prIKboz1yCkqht5GVZuIH2kODQvCve3sh9WYBRry3DYDxjPKBgSFo7+mEZ0Z0apbx9A3xxJWsQhy6nI1hndtJ02UqOyUcHezquHXL093U5DFZmwuDQYCSXwC3TMwQTe4bhG3JmUhK0+Pp/x6UrvfXOGLPK2NqfK3XHrwCV7U9JvQKwI1c0ya+9Siorku4rysc7IwrvsQsTEOkmIqiO3i7WN2Da3BHb/x4+CoAYNrQsGqzvgM6eOLolRwcuJSNsnIBusJSbDypxbmMXKkhKmA8SRQD+T7BHojq6I0Nx9Ox92IWnByM/ZfE5+bjpkb8EwPrHN9twR5SJ3hr9+2iuvUO8sClmwU4fjUHI7r4YNe5G3jvj2QsuLM7+oVWBKHiBrxrp0c1ayuT6jBDZIPEZfcN3cusuKwcBaZ6IU9nFdzUxmkfW+lWLW78J+57E+LtjN+fH46pUaF4LKoD2ns4YW5MN2m396Ym1imJxbB5NdQPtRZh7VygsleioKTcogiVGk5faHzPDOvsjfWzhmF8D3/0CNSgf6gnVPZKaPVFSM6ofooyWZuLuT8cR9zKw8jQF+FGE2SIVPZKhPsaA4tbyQyK9UNinxprDO5o/JJzU9vjsSEdqj1GrCPad/EmjlzJkS7/fr/lHpKXbuajoKQcjg5KdPRxRbCXs9R/aZtpH64gz/oFe+YZquB63pZqJtZnHjNliD5MSMaxKzmYvCzR4rtIPJlozPd7QzEgskHS5q4NzBCJ02VKhfFL3bWJp8yKSsvxyFdVW+FX52p2AS5ez4edUmGxt1JEoAZv3NPzlldRNYS4EuZIajYMBqHGFWathb2dUmofcLoRlmFTxYe6u5MDAj2csPzRftjw/HD8MGOINAVcU4Hp+uNpAIwLH349mtYkGSIA6B5gDIhupY7o4o36B0R3RQbiwYHBeP/+yBpPcgaHeUOpAC5cz8eWUxX7Tv505CqKSisWg4hT7d38NdLU1vSRnaSVakDDmgnG9jJOvQ3qKG+GojXpHeQBADhu6lElfi8BwLum3QhKygxScOTRTCfAtWFAZIPEKbOGbt1RUT+kglKpgEYKiCwzRHnFZXj2fwfx27G0WxitsSZo1/kb+DDhbJ2ra8QW/bcFezRbBqgu3fzd4Kq2R25RGbafzZReJ9dWGhABkJZhi1MZdGvMA6LKauvcKwiC1MMHMP57NEUNEVDR7fxEHTUdtblkCog61CMgcnSww+JJvRHTw7/GY9ydHaRMbaap+NrT2QE5BaXYdiYTBoMAXUEpTpkCoh6BFT3AnFR2+P6ZwXg1tjuGhnvjTlNdUX08OTQMO+aOwsTb2tf7tlS9nu01UCqADH0xMvRFUlE9AKzcn4rcolLkFBqDf4UC0NjA9wEDIhsknvk0dMpMrB/yMG3bIa6Uqpwh+iNJi81JGVhu2vuooczrIs5fz6vlSGC96cN/eDW7X8vF3k4p7d/z3uaz0pebONXYGk0bFgY7pQIJpzKwp9I2HreyurGt2HomA49/sx/pOmMhb+0BkSlDlFJ1WfmpdD0u3siHyl4JlZ0SZ7S50klDu0bOEPUN8QAAHE7NblBhdXFZuZSh6eRT/4Z9dRnVtaILtL/GEfeYgpN9KVn41++n0fetBOmzSgzuRHZKBZ4a3hHfPTW4Qc0ElUoFQutRF0V1c1bZo7NpmnZ7ciZyi8ugUAB+GjXKDQIOXs6W2sO4OznYRDE7AyIbJLbWb/iUmSkgMn04i0vt84rKUG4Q8OOhq7h0I1+qJRCX+Voj9WZBlb2aCoorUtq1rSg4nJqNXedvwF6pwOS+QVY/ZnOYMbIT3NT2OJ2ux/emPZ9a65QZYOx8/YgpCJz7w3GsP56G9cfTMOXfiej66kasO8LMUW2W77iI7cnX8X87U2AwCFJjRo1j1YCoV3sPODookZVfgnOZlicMv5qys6O7+mBsd2NAIP7dN0WGSGWvRHZBqcU+YtbadiYTusJS+GscpemQxjSqa8V2J7cFe6CvKWN04FIW1h25ZrHq1jxDRLZLrCP6+YjxfR6gccRwU+3o/pSKzXdtYboMYEBkk6QMUQOnzCrv8Cx+seuLyrA5SYuX1h7DqPe3S/UjWfklVp0xpt4swNgPt2P6/w5ZXJ5XXBEgifPF1fnMVGM0qW97m2t+5umiwpPDwgBUTG1U7kHU2swe1wXtPZxwLacQM1cewcyVR7AvJQsGAVj466l6BcptzUXTaitxM1fxz6e6tL/KXilNB5ln44pKy7HmgLFo+N4+7fH82M5wsKs4S/Zu5IBIZa9EpOkL6lA9ukKLfjxs7Ag9sU/7JjmbjwjQSEv5bwvxQH/Ta5aUpsdNs+05AKCrv1uV25PtEd/3e0172IV6u1jU1FXMZjRuNrShGBDZoFtddl/5TVYxZVaGpLSKDI4YvJSWC1YtyT9xTYfScgGJF29KWSjxfqVjasgQXckqwLbk61AqgOdGhdfvCTWTyt1tW3OGCDAGgRtnD8ezIzsi1NsZkUHueHJoGLoHaKArLMW/NpyWe4g2SV9UKgWLN/KKseGEcRpYbV9zmwZxReWfpj36AGMBdXZBKdp7OGFcdz90D9Bg9rgu0vXeLo3/JSEuIDhcz+1FbuYVY5tp7JP7Nk2djUKhQNzocHTxc8VdkYEI9HCS9h0EjAGTp6mPUWtsh9EaiRtgiycMHdo5SzV1x6/qkC7uNedsGyefrfsTv4USN9oTBGDA21swN6Yr7jfb0bkmgiBgy+lMqfBRfJOZrzITm2UBsAiCsvJL6ixyTsup2El6f0oWok1FkvlmU2an03NRUmaosnXAGVNX5O4BmnoVZDYnsWGhSNPKAyLAOMUzf0J3zJ/QXbrscGo2Ji/bg5+OXEPvIHc8PjRMxhHanpRKG5TG77kEoPr6IdHtEX5YvPEM9l68CX1RKdzU9vjGdLtHoyo215wxshOKywzwcnZoki99cRqqvhmineeuo8wgoGd7jUVfoMY2dUgHTDVbmt8v1FOqO3xwUAj+1i8Ianuex7cUnXxc4ensIM1ahHq7IMjTCYHujkjTFWGrqVVCfffsayp8Z9kg83T09dxifLv3slW32372Op7+70GsOWis/xB3eDZfZVY59Sy6acX0yDWzgMh8xYz5lFlJuaHaTUPPZRovk6sluzXcnRzQ3qNiyW5rnzKrSd8QT/w92rgv1RvrT2FbcmYdt2hbLt4w1gGJJxznTXVBtQVEHX1c0cnHBaXlArYnX8f5zDycTtdDZa/EAwMqTnaUSgVevL1LkwWhYobobEZelVrA2qTrjHuEdfNv3tqd/mZ7GY7s7ANHBzsWPrcgCoXlfpQdvJ2hUCikTbt3njUGRO42kiFiQGSDzOsIAOMcekFJ3VNaRyqd9Ykf0GJx5o284hrrQsQNJWuTZhEQVfRUqdzwcW81/VbEL41wH9sNiICKTTCB1j9lVpvnRnXCgwODIQjA3LXHkV1DIN0WiRmi2yP84GK27Uxdy4ZvjzBmVBNOZWDPBePfyMAOXs1aP+Hjppb69CRds375vdgbqbmb5w3r7AN7pQI9AjUI8batukOyjnlXanEFYJ8Qy027mSGiGlUuWCw3CDhq1r21JuLqNJH4JvPTGD/EtPoiZNXwxVbT5ebSdBUB0ak0vbSyRqwhEpfi/nLsWpXbXjAFRJ39bDsgigiomA5oqxkiwHhm9/pdPRDu64obecVY8GsSAONyc/Gsrq1Zd+Qq7vl8F7acNmbMOvu6WXyw1zXlfHuEHwDjai2xliiqU9Ps11ebHg3Y6PW66USquRunhvu6Yv3zw/DNEwOa9XGp8Yh1RAAQagpqK298bCs1RAyIbJC4dYe5g5fqnvPXV+ozJNYO+Zp2bxb3MTInThHVNmV2LiMXOQUlSMsxps1VdkoYhIo6BDFDNGVAMOyVCpy8ppcyQoCxtknKENnwlBnADJE5Rwc7fPC3SNgpFfjtWBpe/fkEnow/iGkrDlR5r7UFL645hmNXdVIg0dHHRarJAeoOiPoEe6CDtzPyisukoFKegMj4ZST2FLLG9Vzj374cneS7+Wvg68Yd6Fuq24I9MK67Lx4eFAJnlfEzNSJAY3Hi784MEdWkuiWtB60oghQzNY4OSozs4iMtb3RT20s7yqdmGfeu+uBvkVg8qRfuvi0QAGqsLbpwPQ/jP/kLj39zQMoiDQgzfgmk3jTel7j3V4iXM0Z2Ma6m+eVoRZYoXVeE/JJy2Juan9ky84DIWcWVLJHBHphh2ivq273G/kyl5QKSrjV8+4eWyq5S7UpYOxeL+oi6AiKlUoGHB4VKv7uo7NCr0plycxAzROYrTusidhn2sYH9pqhlcbBT4qupA/D2vb2ky5xUduhsdnLMDBFZJdC07PTw5WyLxmTVEc/aXxnfDSueHCitUlEoFPDTWJ5hDevcDg8ODJGW9tYUECWl6S2m7NzUFd1HM/TGs0YxQ+SqdpACrJX7UnHxeh72XbwpFeWGejtXmdazNSFm/ZHa8pSZuefHdka3Sn1fTtaj/qS1qFw/E+zljNuCPaTfi8vq7ht2n9kqqQFhXrL8PUSYAqIL1/Mt9gmrjRQQyZAhotbJ/GSANURklXv6tIeLyg55xWW4WMe2GGKGqLriTrGOSCS+AcUP+ZqmzLQ6yym2QA8n+JruK0NvvI0YELmo7RDTwx/dAzS4mV+C2z/aiSlf7sU/150EACmQsmVKpQLfPD4Ab9zdg83fTFT2Snz5aH9MGxYmbXFyvB4B0dmMXNz12S78da5l1x5lm/XeiuroDQc7pUVWyLw3V008XVSYZOrSPq67X+MP0gr+Gkd4uahQbhCkdhi1KSotl1p0MENEjaVXUEVAZCv7WjIgsnHDw9tJXZ2v5hTWeqy+sObtA8wzRO5ODlKfIHFH7ZqKqsXltqJAD0f4mebzM3MtM0RujvZwdLDDf58ciI7tXFBuEGA++9fJ17any0Sju/la9EIhIMTbGa/dGYHxpt5T9ckQ/XT4Gk5c0+FTU6fyxlRSZmiWjtpFpeVSBmjrSyPx9eP9pev+cUc3uKjsMN00tViXhXdHYOXTg/DQwJAmGWtdFApFRWG1FXVE4uurslNC49S26+qo8ZjXk3o2QRPShuC720Ytf6QfruUUYkh4OwR5OuGMNhfXsusIiIoqApPK/M0CIvMOuF6m/69p2b22SkDkJAVXGfoiCIIg1RC5mjZD9XFT46fnhmDvxSwM7uiFdzaewW/H0jCmmzxnxNR4xDR3yo186ApLrTqzEzdAPXQ5Gzfziht1S4qn/3sQu8/fwDuTe+O+fk23P56YHbJXKhDWznIT0GdGdMK0YR2t3s5CbW+HIZ3k3dw4IlCDv87dwEkr6ojMp8vYA4gaS79QT4R4OcNVbW/RvkJODIhs1Pie/tL/iyvBrtWRIRJ3s69uyszXPCAy20VbnDLLLiiBwSBY7FwPVJchcqpYxq8rQnGZQdpixNUsEPNwVknP4Z3JvfH2vb1sYjdjujWeLioEeznhSlYhkq7pMCS86hf7m+tPQV9YiiX39YZCoUC6aXWiQQC2nsnE36zoum6tHabVWn9fewz2SgUm9mmabSWy802bUDqrqg0KWtp7u0+wsRh8XzU9wyoTA6J2rB+iRqS2t8OfL40EAJsJtDll1gIEmgKitDqnzKyrIfJ2qfh/sZao3CBU27lWzBCJEXwHbxcpuNIXlUkflgoF4FzLVgMt7QuDaiZmifaadSsX5RSU4OtdKVh76CoumVYhmvev2nI6o9HGIQiCxfvqX7+fRmm5AWczcpFfqVlodn4JHv5qL77fn9qgxxLrg2xlNcytiuroDaXCWFhdOQtcmdSDiPVD1Mgc7JQ2tdDGdkZCNWpv6ixb25RZUWk5SsqNNQ51TZl5mWWIVPZKaWuPm/mWtRhl5QapTuiLR/rh79FdENPDDxpHezg6GN86F0yF3q4q+yrZJWqdRnf1BQD8e8cFnM+0LMo9Z9Z/6uL1PBgMgrQaEQB2nr1h9cqmuuSXlEsrL13V9sjMLcaMbw8h+qOdmP/TCYtjfz2Wht3nbza4jimnUMwQtY6AyN3ZQWqOt+fCjVqPFbtU+7jZRp0HUVNhQNQCtLciQyQuuVcojMFJZeZF1e0qFbBVrDSzrCO6nlcMg2Csmxge3g4zx3SGvZ3SYhn/BdM2Bi5qzr62FZP7BmF453YoLjNgzupjEAQBS7edx4d/JONsRkWAdOF6Hm7kFaO03Fhc7+OmRmFpOQ7Xc2PRmogZTZW9Ek8O7QAAUhfpX4+lWRybaNoqI11XJNU01YdYQ9Sc22w0NbGOaff52qfNrueZmjIyQ0StHAOiFkAMiLT6IpSWV9/rRJwuc1VXn6nxNZ8yq/TBJgZEmbmWGSKxfshP41jlPsWVZlKGqI13dW5LlEoF3v9bJNT2Spy4psOu8zfw3uZkfLr1PDad1ErHXcjMR5rpPeTr5oihpq7M1e111xDiNJaHkwMeGhRaZVpWvN5gECz23jt8OacBj2UMvlrLlBkADA03/nvsuXADglBzjzP2IKK2ggFRC9DOVS1tl2E+/WBOKqiuoZmg2t5O+jD3qpQh8tFUHxCJtQUB7lXb5osBltgbyZUZojbFT+MoNSVctv2CdPlf5yqmXy7eyEO6KasZ4OEobVOR2EgBka6gYhrL390RcaPD0TfEA/amwCjZ1GMnOSMX2QUV9XGHU+ufocpphRmi/qFeUNkpka4rQsqN/BqPY0BEbQUDohZAqVQgwMMYlNRUR6SvpSmjqJNpp/kOlbbPqNxXSCRmiPyrCYgqT5m19X2/2qKBpq1hxJ3bK7twvSJDFOjuhMEdjQHR0Ss5KCy59Toiqa7HyRikvHh7F/z03FBp+xhx+k7MSDnYGQOlhgRE2QWtq4YIMG6f0DfUAwCwu4Z/Q6DixKi5d7onam4MiFqIupbeixmi2gKTj6bchq+n9rfoEApUZHsy9ZUzRKaz+2oDIuNtxLNHZojanv4dvGq9Piu/RGr8F+DuiBAvZwS6O6K0XMAbvyXhu32Xb+nxxWks90pBShdTh/FkU0AkZq3EPkVJ1/T1LuyuWGXWejJEADDUVEe053z1hdVaXRHSdEVQKipeV6LWigFRC1FXYbW05L6W/beCvZwxtprtAvykKbMiFJeV44ppA9iKDJFTNbexDJIYELU9fUM8LDqRq0zLZwPdHaX3667zxj5BAR5OUCgUGGyaNlt14Ar+ue4kzmXUvXVETXIKjUFK5eaQXf1MAZE2F1tOZWDrGWOh9WNRHdDOVYWScoO027y1pAyRjWwx0FjEPlKJF29Wu1eiWHsVEaip9bOFqDVgQNRCBNaRIdJLNUT1D0x83cTO08V4ac0xDF+yDS+uPoo/TSt2Qs02PBVVDoi4yqztcXN0QDd/4xYQnXxc0L+DsdlfZz83dPQxTsuK+92JmxTf3z8Yjg5KaeuYI6k5DX58XQ1BirgH3YFL2Zj343EAwFPDwtA9QIO7Io2bD8//6QRe+/kk4lYertKzqDqtsYYIACKD3OGqtkdOQamUzbt8Mx8lpm1K9pl6TQ0K85ZtjETNhQFRCyEWNFZeGi+qrUt1XaQMkb4IG02rhH46cg2FpeUY2cUHI7v6VLlNZJCHRT0Fa4japkEdjdNmgzt6Y4KpM3lUJ2+pXk0UYAroB3f0xpk3J+AJ0zL5I1dyGvzYOTXU9YjBGADczC9BN383zB3fFQAwb3w39Ag0bj78v72XseF4On4/kW71Y3m6tK4sib2dEoNMtWC7zt/At3svY+R72/HupjMAKjpZi8cQtWYMiFoIR1MX6JI6lt03JEPk41bReVpMm6vtlYjtFYB/P9qv2k6iTio7PDIoVPqdU2Zt0/NjOuOFsZ0x5/YueGRwKNbPGoZpw8JwV2SgNJXlYKdASKUsYx/TCrWjtxIQiVNmlbI2ans7RAQYM1cRARr898mBUNsb/34cHeyw7OF+0go5ADidXvu0nSAIVQq4W5MRpiL0L7afx6s/nwQAfL0rBddzi3Hhej4UiooCeqLWjN9iLYTaNMVQXFpDQCQVVdf/DFbsPF1kum8/jRp754+tc3+Zx4aE4vNt5wEABY2waohaHk8XFebc3kX6Xex+3C/UE/v+MRYJpzLg4exQpdXDbaa9tJK1ehSUlMG5mmaidcmppa7noym34djVHNzbp32VgD7E2xk/xw3FD4eu4u9rj+FUeu0bnJqfKLSmVWaiKQOCsf54Gg5cslx9t81Ue9XVz63VTRUSVYcZohZCCojKqg88cqVl9/X/YlEoFFIdEQCE+7patdmer5sj4kZ3gspeiTt6BdT7cal1c3Sww12RgRjeueqUq7+7I/w1jjAIwImrde+4Xh1dLdtpdPV3w/39g2vdJ0nMIp1K09fYmPCvc9fx2Nf7ABj/Bh1r2a+vpXJ0sMNXUwdgYJiXVAwPAF/+dREApP5RRK2drAHR4sWLMWDAALi5ucHX1xcTJ05EcnKyxTFFRUWIi4uDt7c3XF1dMXnyZGRkWG4QmZqaitjYWDg7O8PX1xdz585FWZlloeT27dvRt29fqNVqhIeHIz4+vqmfXqNSmz6Ii8tqmjJreIYIsNz8NbxS/Udt/h7dFafeiJEKWYmsFRlszCY1dNqsIkPUsOxFuK8rHOwU0BeV1bhY4fVfk3DMFLC15j487k4OWPNsFHbMHYXbI4wrUc+b9qUTp9SIWjtZA6IdO3YgLi4Oe/fuRUJCAkpLSxEdHY38/IquqXPmzMFvv/2GtWvXYseOHUhLS8OkSZOk68vLyxEbG4uSkhLs2bMHK1asQHx8PBYsWCAdk5KSgtjYWIwePRpHjx7F7Nmz8dRTT2Hz5s3N+nxvhZghqql/ipQhamBAVDlDZC2FQgF7G9qtmFoOcdrs2NWcet/WYBCkGqKGTmOp7JXo7GsM5MUVVubOZ+bh4vV8ONgpMLKLD2aNCW/Q47Qk9nZK9G5f0adMZa/EYK4wozZC1hqiTZs2WfweHx8PX19fHDp0CCNGjIBOp8PXX3+NlStXYsyYMQCAb775Bt27d8fevXsxePBg/PHHHzh16hS2bNkCPz8/3HbbbXjzzTcxb948LFy4ECqVCsuXL0dYWBg++OADAED37t2xa9cufPTRR4iJiWn2590QFVNm1WeIrGnMWBvzvc461SMgImoosbD5aGoOkrW5eO67Q3h+bGfcc1v7Wm/3wqoj2J58Xap5q9yYsT66B2hwKl2PL3deRIa+CA8PCpX27ducZFxxOaRTO6x4cmCDH6Ol6WnWuHVQmBecVK1vmpCoOjZ1aq/TGVPTXl7GFQ2HDh1CaWkpxo0bJx3TrVs3hISEIDExEQCQmJiIXr16wc+vouFgTEwM9Ho9kpKSpGPM70M8RryPlkBcJVNjQGTqpdLQTVYbmiEiaqjeQe5QKoA0XRE++CMZF67n45Mt52rdaBQAfjmaJtUP2SkVcLuFFY4RgcY6ooOXs/HaL0n4+eg16bo/Thmn5mN6+Df4/luiXmYZohHV1H8RtVY2ExAZDAbMnj0bQ4cORc+ePQEAWq0WKpUKHh4eFsf6+flBq9VKx5gHQ+L14nW1HaPX61FYWH3tQHFxMfR6vcWPnNQO4iqzqlNmgiBIzeUauvxdrCHSONrDpxXXSpDtcFHbS1NWYvBx8UZ+rcvgxdWUIncnB6sWANRkgKmZpCjBNA6trgjHruRAoQDGRfg2+P5bonauanTzd4ODnQJju7et505tm80ERHFxcTh58iRWrVol91AAGAu+3d3dpZ/g4GBZx+NYS1F1UakBYtf9hnaM7m5acTOgg9ctfcEQ1Yd5PyDR+uNpNR6fnmO5AbHYQbqhegd5YN1zQ/DvR/sBAHaevY6SMgMSLxr39urd3t0ie9pWxD8xEL/NGoaO9VhgQdTS2URANHPmTKxfvx7btm1DUFCQdLm/vz9KSkqQk5NjcXxGRgb8/f2lYyqvOhN/r+sYjUYDJ6eq+3QBwPz586HT6aSfK1eu3NJzvFXmNUSVpxRyiyvOmp0buCy4e4AGW14cgU8e7NPwQRLV020hHtL/O5qyoBtOpNc4bZaus8zoVrP9Vr31CfHE7d394OOmRn5JOfanZOGgqSfPgDo2sG2t/N0dpW1ZiNoKWQMiQRAwc+ZMrFu3Dlu3bkVYWJjF9f369YODgwP+/PNP6bLk5GSkpqYiKioKABAVFYUTJ04gMzNTOiYhIQEajQYRERHSMeb3IR4j3kd11Go1NBqNxY+cxIAIqNqtOr/YOI3mqraXCkIbItzXjR2nqVmZZ4ieHt4Rjg5KXL5ZgFPp1U9RixsOi8Ql4rdKqVRgTFfj9NCW0xk4dNkYEPUL9aztZkTUisgaEMXFxeHbb7/FypUr4ebmBq1WC61WK9X1uLu7Y9q0aXjxxRexbds2HDp0CE888QSioqIwePBgAEB0dDQiIiLw6KOP4tixY9i8eTNeffVVxMXFQa021sJMnz4dFy9exMsvv4wzZ87giy++wJo1azBnzhzZnnt9iUXVQNVpM7F+yEXN1SDUsnT2dZW2m5nQMwDDwo1FvFtPZ1Z7fLqpX9BDg0Lw3ycH4p1JvRptLGNM9TLrj6chOcNYx9SvAwMiorZC1oBo2bJl0Ol0GDVqFAICAqSf1atXS8d89NFHuPPOOzF58mSMGDEC/v7++Omnn6Tr7ezssH79etjZ2SEqKgqPPPIIHnvsMSxatEg6JiwsDBs2bEBCQgIiIyPxwQcf4KuvvmoxS+4B435QYmlP5e07xB5E3HGeWhp7OyX+77H++PyhPogI1EhFvH+eqSEgMmWIAt0dMaKLD7wbcQHA6K6+CHB3xI28EggCEOLl3Cbrh4jaKlm/QetaXgsAjo6OWLp0KZYuXVrjMaGhofj9999rvZ9Ro0bhyJEj9R6jrVAoFFDbG/cbq7x9h5ghupXlx0RyGdSxovHfaNO01bGrOUi8YNxp3XzrCDEgCnCvvvbvVqjslXhmREe88dspAJwuI2prbKKomqxTuRdRucG43D6/hBkiah383R3Rs70GggA8+H978dBXe3Hhep50fZqpqDrAvWkyNw8MCEE7V+NWIAyIiNoWBkQtSOXtO6atOIDB//oTV7IKADAgotZhTLeKQmlBAHafv2H6fwFaMUPk0fgZIgBwUtnh4yl98ODAEEzqW3vHbCJqXfgN2oJIzRlNS+/3XryJolIDjqTmAGh4U0YiW/L08DAUFJfhwKUsHLuqw96LN/FYVAfoC8tQUGI8GWiqDBEADOvcDsM6t2uy+yci28QMUQsiTZmVGqArLJX2cko1ZYgYEFFr4ObogFfvjMA/Y41tM/anZEEQBGm6zNPZQWpUSkTUWBgQtSAVzRnLLfqxpHLKjFqhyGB3qO2VuJFXggvX85CWI9YPNc10GRG1bfwGbUHMu1VrzQIiscjalX2IqBVR29uhT4gH9l7Mwvf7r2CraSl+J24+TERNgBmiFsR8P7PKHXsBTplR6zMozLjk/utdKUi5kY/2Hk6YG91V5lERUWvEb9AWRMoQlZZDq68aEHHKjFqb+wcEY+/Fm7ieV4wQL2csntSLU2ZE1CT4DdqCmPch0lba5BJghohan/YeTlj9bM17DhIRNRZOmbUg5svuq5syY4aIiIioYRgQtSDmq8y01dUQOTIgIiIiaggGRC2IeR+i6mqIOGVGRETUMAyIWhAxQ5SVXyLtcG+OU2ZEREQNw4CoBRFriC7dzAdg3N3eWVXRe8hVxYCIiIioIRgQtSDilNnlm8bO1P7ujvByUUnXu7AxIxERUYMwIGpBxCkzcasOf3dHeJsCIkcHJezt+M9JRETUEJxjaUHEgEjk46aGvVIBgAXVREREt4Lfoi2IutIO3z6uaijAgIiIiOhW8Vu0BXF0sMwQebuqIJj+nyvMiIiIGo7foi2IWFQtaueqhsEUETEgIiIiajh+i7YglWuIvF3VcDRNo/lrHOUYEhERUavAgKgFqZohUmFIJ298NCUSUR3byTQqIiKilo8BUQuirlRD1M5VDQc7Je7tEyTTiIiIiFoHNq5pQSpPmZk3ZSQiIqKGY0DUgphPmXk4O8CBjRiJiIgaBb9RWxDzDFE7V7WMIyEiImpdGBC1IOY1RN6cLiMiImo0DIhaEPMpM29XBkRERESNhQFRC2I+ZeamdpBxJERERK0LA6IWxDwgcnVkxwQiIqLGwoCoBbE3W1XmxoCIiIio0TAgaqG4uz0REVHjYUDUQnk6s6iaiIiosTAgamEeH9IBvdq7I7Z3gNxDISIiajU479LCLLy7h9xDICIianWYISIiIqI2jwERERERtXkMiIiIiKjNkzUg2rlzJ+666y4EBgZCoVDg559/trg+Ly8PM2fORFBQEJycnBAREYHly5dbHFNUVIS4uDh4e3vD1dUVkydPRkZGhsUxqampiI2NhbOzM3x9fTF37lyUlZU19dMjIiKiFkLWgCg/Px+RkZFYunRptde/+OKL2LRpE7799lucPn0as2fPxsyZM/Hrr79Kx8yZMwe//fYb1q5dix07diAtLQ2TJk2Sri8vL0dsbCxKSkqwZ88erFixAvHx8ViwYEGTPz8iIiJqGRSCIAhyDwIAFAoF1q1bh4kTJ0qX9ezZE1OmTMFrr70mXdavXz9MmDABb731FnQ6HXx8fLBy5Urcd999AIAzZ86ge/fuSExMxODBg7Fx40bceeedSEtLg5+fHwBg+fLlmDdvHq5fvw6Vyrp+Pnq9Hu7u7tDpdNBoNI33xImIiKjJWPv9bdM1REOGDMGvv/6Ka9euQRAEbNu2DWfPnkV0dDQA4NChQygtLcW4ceOk23Tr1g0hISFITEwEACQmJqJXr15SMAQAMTEx0Ov1SEpKqvGxi4uLodfrLX6IiIiodbLpgOizzz5DREQEgoKCoFKpMH78eCxduhQjRowAAGi1WqhUKnh4eFjczs/PD1qtVjrGPBgSrxevq8nixYvh7u4u/QQHBzfiMyMiIiJbYvMB0d69e/Hrr7/i0KFD+OCDDxAXF4ctW7Y0+WPPnz8fOp1O+rly5UqTPyYRERHJw2Y7VRcWFuIf//gH1q1bh9jYWABA7969cfToUbz//vsYN24c/P39UVJSgpycHIssUUZGBvz9/QEA/v7+2L9/v8V9i6vQxGOqo1aroVarG/lZERERkS2y2QxRaWkpSktLoVRaDtHOzg4GgwGAscDawcEBf/75p3R9cnIyUlNTERUVBQCIiorCiRMnkJmZKR2TkJAAjUaDiIiIZngmREREZOtkzRDl5eXh/Pnz0u8pKSk4evQovLy8EBISgpEjR2Lu3LlwcnJCaGgoduzYgf/+97/48MMPAQDu7u6YNm0aXnzxRXh5eUGj0WDWrFmIiorC4MGDAQDR0dGIiIjAo48+iiVLlkCr1eLVV19FXFwcM0BEREQEQOZl99u3b8fo0aOrXD516lTEx8dDq9Vi/vz5+OOPP5CVlYXQ0FA888wzmDNnDhQKBQBjY8aXXnoJ33//PYqLixETE4MvvvjCYjrs8uXLmDFjBrZv3w4XFxdMnToV77zzDuztrY8HueyeiIio5bH2+9tm+hDZOgZERERELY+13982W1Rta8S4kf2IiIiIWg7xe7uu/A8DIivl5uYCAPsRERERtUC5ublwd3ev8XpOmVnJYDAgLS0Nbm5uUv1SY9Dr9QgODsaVK1c4FVcHvlb1w9fLenyt6oevl/X4WlmvqV4rQRCQm5uLwMDAKivXzTFDZCWlUomgoKAmu3+NRsM/Fivxtaofvl7W42tVP3y9rMfXynpN8VrVlhkS2WwfIiIiIqLmwoCIiIiI2jwGRDJTq9V4/fXX2STSCnyt6oevl/X4WtUPXy/r8bWyntyvFYuqiYiIqM1jhoiIiIjaPAZERERE1OYxICIiIqI2jwERERERtXkMiGS2dOlSdOjQAY6Ojhg0aBD2798v95Bkt3DhQigUCoufbt26SdcXFRUhLi4O3t7ecHV1xeTJk5GRkSHjiJvPzp07cddddyEwMBAKhQI///yzxfWCIGDBggUICAiAk5MTxo0bh3Pnzlkck5WVhYcffhgajQYeHh6YNm0a8vLymvFZNJ+6Xq/HH3+8yntt/PjxFse0lddr8eLFGDBgANzc3ODr64uJEyciOTnZ4hhr/vZSU1MRGxsLZ2dn+Pr6Yu7cuSgrK2vOp9LkrHmtRo0aVeW9NX36dItj2sJrtWzZMvTu3VtqthgVFYWNGzdK19vSe4oBkYxWr16NF198Ea+//joOHz6MyMhIxMTEIDMzU+6hya5Hjx5IT0+Xfnbt2iVdN2fOHPz2229Yu3YtduzYgbS0NEyaNEnG0Taf/Px8REZGYunSpdVev2TJEnz66adYvnw59u3bBxcXF8TExKCoqEg65uGHH0ZSUhISEhKwfv167Ny5E88880xzPYVmVdfrBQDjx4+3eK99//33Fte3lddrx44diIuLw969e5GQkIDS0lJER0cjPz9fOqauv73y8nLExsaipKQEe/bswYoVKxAfH48FCxbI8ZSajDWvFQA8/fTTFu+tJUuWSNe1ldcqKCgI77zzDg4dOoSDBw9izJgxuOeee5CUlATAxt5TAslm4MCBQlxcnPR7eXm5EBgYKCxevFjGUcnv9ddfFyIjI6u9LicnR3BwcBDWrl0rXXb69GkBgJCYmNhMI7QNAIR169ZJvxsMBsHf31947733pMtycnIEtVotfP/994IgCMKpU6cEAMKBAwekYzZu3CgoFArh2rVrzTZ2OVR+vQRBEKZOnSrcc889Nd6mLb9emZmZAgBhx44dgiBY97f3+++/C0qlUtBqtdIxy5YtEzQajVBcXNy8T6AZVX6tBEEQRo4cKbzwwgs13qatvlaCIAienp7CV199ZXPvKWaIZFJSUoJDhw5h3Lhx0mVKpRLjxo1DYmKijCOzDefOnUNgYCA6duyIhx9+GKmpqQCAQ4cOobS01OJ169atG0JCQtr865aSkgKtVmvx2ri7u2PQoEHSa5OYmAgPDw/0799fOmbcuHFQKpXYt29fs4/ZFmzfvh2+vr7o2rUrZsyYgZs3b0rXteXXS6fTAQC8vLwAWPe3l5iYiF69esHPz086JiYmBnq9XsoItEaVXyvRd999h3bt2qFnz56YP38+CgoKpOva4mtVXl6OVatWIT8/H1FRUTb3nuLmrjK5ceMGysvLLf6RAcDPzw9nzpyRaVS2YdCgQYiPj0fXrl2Rnp6ON954A8OHD8fJkyeh1WqhUqng4eFhcRs/Pz9otVp5BmwjxOdf3XtKvE6r1cLX19fient7e3h5ebXJ12/8+PGYNGkSwsLCcOHCBfzjH//AhAkTkJiYCDs7uzb7ehkMBsyePRtDhw5Fz549AcCqvz2tVlvt+0+8rjWq7rUCgIceegihoaEIDAzE8ePHMW/ePCQnJ+Onn34C0LZeqxMnTiAqKgpFRUVwdXXFunXrEBERgaNHj9rUe4oBEdmcCRMmSP/fu3dvDBo0CKGhoVizZg2cnJxkHBm1Ng888ID0/7169ULv3r3RqVMnbN++HWPHjpVxZPKKi4vDyZMnLWr3qHo1vVbmdWa9evVCQEAAxo4diwsXLqBTp07NPUxZde3aFUePHoVOp8MPP/yAqVOnYseOHXIPqwpOmcmkXbt2sLOzq1JNn5GRAX9/f5lGZZs8PDzQpUsXnD9/Hv7+/igpKUFOTo7FMXzdID3/2t5T/v7+VYr2y8rKkJWV1eZfPwDo2LEj2rVrh/PnzwNom6/XzJkzsX79emzbtg1BQUHS5db87fn7+1f7/hOva21qeq2qM2jQIACweG+1lddKpVIhPDwc/fr1w+LFixEZGYlPPvnE5t5TDIhkolKp0K9fP/z555/SZQaDAX/++SeioqJkHJntycvLw4ULFxAQEIB+/frBwcHB4nVLTk5Gampqm3/dwsLC4O/vb/Ha6PV67Nu3T3ptoqKikJOTg0OHDknHbN26FQaDQfrAbsuuXr2KmzdvIiAgAEDber0EQcDMmTOxbt06bN26FWFhYRbXW/O3FxUVhRMnTlgEkQkJCdBoNIiIiGieJ9IM6nqtqnP06FEAsHhvtYXXqjoGgwHFxcW2955q1BJtqpdVq1YJarVaiI+PF06dOiU888wzgoeHh0U1fVv00ksvCdu3bxdSUlKE3bt3C+PGjRPatWsnZGZmCoIgCNOnTxdCQkKErVu3CgcPHhSioqKEqKgomUfdPHJzc4UjR44IR44cEQAIH374oXDkyBHh8uXLgiAIwjvvvCN4eHgIv/zyi3D8+HHhnnvuEcLCwoTCwkLpPsaPHy/06dNH2Ldvn7Br1y6hc+fOwoMPPijXU2pStb1eubm5wt///nchMTFRSElJEbZs2SL07dtX6Ny5s1BUVCTdR1t5vWbMmCG4u7sL27dvF9LT06WfgoIC6Zi6/vbKysqEnj17CtHR0cLRo0eFTZs2CT4+PsL8+fPleEpNpq7X6vz588KiRYuEgwcPCikpKcIvv/widOzYURgxYoR0H23ltXrllVeEHTt2CCkpKcLx48eFV155RVAoFMIff/whCIJtvacYEMnss88+E0JCQgSVSiUMHDhQ2Lt3r9xDkt2UKVOEgIAAQaVSCe3btxemTJkinD9/Xrq+sLBQeO655wRPT0/B2dlZuPfee4X09HQZR9x8tm3bJgCo8jN16lRBEIxL71977TXBz89PUKvVwtixY4Xk5GSL+7h586bw4IMPCq6uroJGoxGeeOIJITc3V4Zn0/Rqe70KCgqE6OhowcfHR3BwcBBCQ0OFp59+usoJSVt5vap7nQAI33zzjXSMNX97ly5dEiZMmCA4OTkJ7dq1E1566SWhtLS0mZ9N06rrtUpNTRVGjBgheHl5CWq1WggPDxfmzp0r6HQ6i/tpC6/Vk08+KYSGhgoqlUrw8fERxo4dKwVDgmBb7ymFIAhC4+aciIiIiFoW1hARERFRm8eAiIiIiNo8BkRERETU5jEgIiIiojaPARERERG1eQyIiIiIqM1jQERERERtHgMiImpVLl26BIVCIW2V0BQef/xxTJw4scnun4iaHwMiIrIpjz/+OBQKRZWf8ePHW3X74OBgpKeno2fPnk08UiJqTezlHgARUWXjx4/HN998Y3GZWq226rZ2dnatbrdwImp6zBARkc1Rq9Xw9/e3+PH09AQAKBQKLFu2DBMmTICTkxM6duyIH374Qbpt5Smz7OxsPPzww/Dx8YGTkxM6d+5sEWydOHECY8aMgZOTE7y9vfHMM88gLy9Pur68vBwvvvgiPDw84O3tjZdffhmVdzwyGAxYvHgxwsLC4OTkhMjISIsx1TUGIpIfAyIianFee+01TJ48GceOHcPDDz+MBx54AKdPn67x2FOnTmHjxo04ffo0li1bhnbt2gEA8vPzERMTA09PTxw4cABr167Fli1bMHPmTOn2H3zwAeLj4/Gf//wHu3btQlZWFtatW2fxGIsXL8Z///tfLF++HElJSZgzZw4eeeQR7Nixo84xEJGNaPTtYomIbsHUqVMFOzs7wcXFxeLn7bffFgTBuNP49OnTLW4zaNAgYcaMGYIgCEJKSooAQDhy5IggCIJw1113CU888US1j/Xll18Knp6eQl5ennTZhg0bBKVSKe16HxAQICxZskS6vrS0VAgKChLuueceQRAEoaioSHB2dhb27Nljcd/Tpk0THnzwwTrHQES2gTVERGRzRo8ejWXLlllc5uXlJf1/VFSUxXVRUVE1riqbMWMGJk+ejMOHDyM6OhoTJ07EkCFDAACnT59GZGQkXFxcpOOHDh0Kg8GA5ORkODo6Ij09HYMGDZKut7e3R//+/aVps/Pnz6OgoAC33367xeOWlJSgT58+dY6BiGwDAyIisjkuLi4IDw9vlPuaMGECLl++jN9//x0JCQkYO3Ys4uLi8P777zfK/Yv1Rhs2bED79u0trhMLwZt6DER061hDREQtzt69e6v83r179xqP9/HxwdSpU/Htt9/i448/xpdffgkA6N69O44dO4b8/Hzp2N27d0OpVKJr165wd3dHQEAA9u3bJ11fVlaGQ4cOSb9HRERArVYjNTUV4eHhFj/BwcF1joGIbAMzRERkc4qLi6HVai0us7e3lwqR165di/79+2PYsGH47rvvsH//fnz99dfV3teCBQvQr18/9OjRA8XFxVi/fr0UPD388MN4/fXXMXXqVCxcuBDXr1/HrFmz8Oijj8LPzw8A8MILL+Cdd95B586d0a1bN3z44YfIycmR7t/NzQ1///vfMWfOHBgMBgwbNgw6nQ67d++GRqPB1KlTax0DEdkGBkREZHM2bdqEgIAAi8u6du2KM2fOAADeeOMNrFq1Cs899xwCAgLw/fffIyIiotr7UqlUmD9/Pi5dugQnJycMHz4cq1atAgA4Oztj8+bNeOGFFzBgwAA4Oztj8uTJ+PDDD6Xbv/TSS0hPT8fUqVOhVCrx5JNP4t5774VOp5OOefPNN+Hj44PFixfj4sWL8PDwQN++ffGPf/yjzjEQkW1QCEKlhhpERDZMoVBg3bp13DqDiBoVa4iIiIiozWNARERERG0ea4iIqEXhLD8RNQVmiIiIiKjNY0BEREREbR4DIiIiImrzGBARERFRm8eAiIiIiNo8BkRERETU5jEgIiIiojaPARERERG1eQyIiIiIqM37fywe4508bKSmAAAAAElFTkSuQmCC", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Kolmogorov-Smirnov test: D = 0.04999115704615964, p-value = 0.42766368637858587\n" + ] + }, + { + "data": { + "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlYAAAGdCAYAAADQYj31AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAACCh0lEQVR4nO3deVwU9f/A8dcuNyggHhxKgomKFxgq4pGZFKaVlCUa5RFph5iKZVqm1s/STM1Mk7SUDk0zy/yqkYSapoSI4oGKF3kCHggIyiE7vz9WNldBRYHheD8fj3ns7sx7Zt4zu+y+mfnMZzSKoigIIYQQQoj7plU7ASGEEEKI6kIKKyGEEEKIMiKFlRBCCCFEGZHCSgghhBCijEhhJYQQQghRRqSwEkIIIYQoI1JYCSGEEEKUESmshBBCCCHKiKnaCVQ1Op2Os2fPUrt2bTQajdrpCCGEEOIuKIrC5cuXcXFxQastv+NKUliV0tmzZ3F1dVU7DSGEEELcg1OnTtGoUaNyW74UVqVUu3ZtQP/G2NraqpyNEEIIIe5GVlYWrq6uht/x8iKFVSkVnf6ztbWVwkoIIYSoYsq7GY80XhdCCCGEKCNSWAkhhBBClBEprIQQQgghyoi0sRLiJoqicO3aNQoLC9VORQghxF0yMTHB1NRU9a6QpLAS4gb5+fmkpKRw5coVtVMRQghRStbW1jg7O2Nubq5aDlJYCXGdTqcjOTkZExMTXFxcMDc3V/0/HyGEEHemKAr5+fmcP3+e5ORkPDw8yrUT0NuRwkqI6/Lz89HpdLi6umJtba12OkIIIUrBysoKMzMzTpw4QX5+PpaWlqrkIY3XhbiJWv/lCCGEuD+V4ftb/QyEEEIIIaoJKayEqCEiIiKwt7dXO40qQVEUhg8fjoODAxqNhoSEBLVTui+pqak89thj2NjYVKnPgJubG3PmzLnr+ClTpuDt7X3f6y2r5ZSnf//9t8I+mxqNhtWrV1ea5VR20sZKiDv4LOpwha5vzGPNShU/ZMgQMjIybvnC2rx5Mz169ODSpUvY29sTFBRE796972qZERERjB49moyMjFLlUl1ERkYSERHB5s2badKkCfXq1Ss2TlEUFi1axDfffENiYiKmpqY0bdqUF198keHDh5dJWz2NRsOvv/5KYGDgPS/js88+IyUlhYSEBOzs7O47p/uRlpZGo0aN+P777xkwYMAt00NCQti9eze7du0iLi4OGxubcs2nuP371ltvMXLkyHJd750kJyfz3nvvsXnzZtLT06lXrx4+Pj588skntGjRAldXV1JSUkr8bKppypQprF69+paiLyUlhTp16qiTVAWSI1ZC1BBWVlY0aNBA7TRukZ+fr3YKtzh27BjOzs507twZJycnTE2L/x/0pZdeYvTo0fTt25dNmzaRkJDA+++/z2+//caGDRsqOOuSHTt2DB8fHzw8PEr8DBQUFFRILo6OjvTp04fFixffMi0nJ4effvqJkJAQAOrXr6/KhSS1atWibt26Fb7eIgUFBTz22GNkZmbyyy+/kJSUxIoVK2jTpo3hnx0TE5PbfjYrIycnJywsLNROo9xJYSVEDXHzqcA9e/bQo0cPateuja2tLT4+PuzcuZPNmzczdOhQMjMz0Wg0aDQapkyZAsClS5cYNGgQderUwdramieeeIIjR44YrWfRokWGKyufeeYZZs+ebbTeotMsX3/9Ne7u7oYrdyIjI+natSv29vbUrVuXJ598kmPHjhnmKzr18dNPP9GtWzesrKzo0KEDhw8fJi4ujvbt21OrVi2eeOIJzp8/f9t98ddff9GxY0csLCxwdnZm/PjxXLt2DdAfARw5ciQnT55Eo9Hg5uZW7DJ++uknli5dyo8//si7775Lhw4dcHNzo2/fvmzcuJEePXoA+m48PvzwQxo1aoSFhQXe3t5ERkYalpOfn09oaCjOzs5YWlrSuHFjpk2bBmBY9zPPPGOUS0nvXXHc3NxYtWoV3333HRqNhiFDhgD6IzULFizg6aefxsbGho8++giABQsW8OCDD2Jubk7z5s35/vvvjZan0Wj46quvePLJJ7G2tsbT05OYmBiOHj3KI488go2NDZ07dzZ6724WEhJCdHQ0J0+eNBq/cuVKrl27RnBwsCH3G08Fnjx5kr59+1KrVi1sbW3p378/aWlpJa4nLi6Oxx57jHr16mFnZ0f37t3ZtWuX0b4pbv/efCrwTu9h0Wfzl19+oUePHlhbW+Pl5UVMTIwh5sSJEzz11FPUqVMHGxsbWrVqxfr164vNOzExkWPHjvHll1/SqVMnGjduTJcuXZg6dSqdOnUyWmfRUaHNmzej0Wj4448/aNeuHVZWVjz66KOcO3eO33//HU9PT2xtbXnhhReM+ukr7nSrt7e34W++OO+88w7NmjXD2tqaJk2a8P777xsK84iICD744AP27Nlj+P6IiIgAbj0VuG/fPh599FGsrKyoW7cuw4cPJzs72zB9yJAhBAYGMnPmTJydnalbty4jRoyosH8C7pUUVkLUUMHBwTRq1Ii4uDji4+MZP348ZmZmdO7cmTlz5mBra0tKSgopKSm89dZbgP6LbufOnaxZs4aYmBgURaF3796GL7pt27bx2muvMWrUKBISEnjssccMP9g3Onr0KKtWreKXX34x/DDk5OQQFhbGzp07iY6ORqvV8swzz6DT6YzmnTx5MhMnTmTXrl2YmprywgsvMG7cOD7//HO2bt3K0aNHmTRpUonbfebMGXr37k2HDh3Ys2cPCxYs4JtvvmHq1KkAfP7554Yf0ZSUFOLi4opdztKlS2nevDl9+/a9ZZpGozGccvv888+ZNWsWM2fOZO/evQQEBPD0008bCtK5c+eyZs0afvrpJ5KSkli6dKnhB75o3UuWLDHKpaT3rjhxcXH06tWL/v37k5KSwueff26YNmXKFJ555hn27dvHyy+/zK+//sqoUaMYO3Ys+/fv59VXX2Xo0KFs2rTJaJn/93//x6BBg0hISKBFixa88MILvPrqq0yYMIGdO3eiKAqhoaElvge9e/fG0dHR8INbZMmSJTz77LPFtgPT6XT07duX9PR0/vrrL6Kiojh+/DhBQUElrufy5csMHjyYv//+m3/++QcPDw969+7N5cuXb7t/b3an97DIe++9x1tvvUVCQgLNmjVj4MCBhoJ9xIgR5OXlsWXLFvbt28cnn3xCrVq1il1f/fr10Wq1/Pzzz6W+A8SUKVOYN28e27dv59SpU/Tv3585c+awbNky1q1bx4YNG/jiiy9Ktcyb1a5dm4iICA4cOMDnn3/OokWL+OyzzwAICgpi7NixtGrVyvD9Udx7lJOTQ0BAAHXq1CEuLo6VK1fy559/3vK52bRpE8eOHWPTpk18++23RERE3PK5qXQUUSqZmZkKoGRmZqqdiihjV69eVQ4cOKBcvXrVaPzsDUkVOpQkNfNqsUP/F15UTExMFGsbG6PB0tJSAZSkEymKoijKkiVLFDs7O8PyateurURERBS7rptjFUVRDh8+rADKtm3bDOMuXLigWFlZKT/99JOiKIoSFBSk9OnTx2i+4OBgo2VNnjxZMTMzU86dO1fitiqKopw/f14BlH379imKoijJyckKoHz99deGmB9//FEBlOjoaMO4adOmKc2bNy9xue+++67SvHlzRafTGcbNnz9fqVWrllJYWKgoiqJ89tlnSuPGjW+bn6enp/L000/fNkZRFMXFxUX56KOPjMZ16NBBeeONNxRFUZSRI0cqjz76qFE+NwKUX3/91Wjc7d674vTt21cZPHjwLcsdPXq00bjOnTsrw4YNMxr3/PPPK7179zaab+LEiYbXMTExCqB88803hnE//vijYmlpeducxo8fr7i7uxu2++jRo4pGo1H+/PNPQ0zjxo2Vzz77TFEURdmwYYNiYmKinDx50jA9MTFRAZQdO3YoiqL/bHl5eZW4zsLCQqV27drK//73P6PtuXn/3rycO72HxX02i3I7ePCgoiiK0qZNG2XKlCm33Sc3mjdvnmJtba3Url1b6dGjh/Lhhx8qx44dM0wvWufu3bsVRVGUTZs2KYDR/ps2bZoCGM336quvKgEBAYbXN+7jIl5eXsrkyZMNr4vbRzf69NNPFR8fH8Prkt6HG5ezcOFCpU6dOkp2drZh+rp16xStVqukpqYqiqIogwcPVho3bqxcu3bNEPP8888rQUFBJeZS0ve4olTc77ccsRKiGujSrTvRW2ONhllffHnbecLCwnjllVfw9/dn+vTptz11A3Dw4EFMTU3x9fU1jKtbty7Nmzfn4MGDACQlJdGxY0ej+W5+DdC4cWPq169vNO7IkSMMHDiQJk2aYGtrazhqc/PporZt2xqeOzo6AtCmTRujcefOnbvtdvj5+Rn1qt+lSxeys7M5ffp0ifPdTFGUO8ZkZWVx9uxZunTpYjS+S5cuhn02ZMgQEhISaN68OW+++eZdtc0q7XtXkvbt2xu9Pnjw4G1zLXI370Fubi5ZWVklrvvll18mOTnZcDRsyZIluLm58eijjxYbf/DgQVxdXXF1dTWMa9myJfb29rfkVyQtLY1hw4bh4eGBnZ0dtra2ZGdn3/KZup27eQ+L3LhfnJ2dAQyfxTfffJOpU6fSpUsXJk+ezN69e2+73hEjRpCamsrSpUvx8/Nj5cqVtGrViqioqNvOd/N7U3S67sZxt/v7uBsrVqygS5cuODk5UatWLSZOnFiqfQr699PLy8vo4oQuXbqg0+lISkoyjGvVqhUmJiaG187Ozvedf3mTwkqIasDaxhr3Bx80GpycG952nilTppCYmEifPn3YuHEjLVu25Ndff62QfIu70uupp54iPT2dRYsWERsbS2xsLHBr4/YbT3kVFUc3j7v59GF5aNasGYcOHbrv5Tz00EMkJyfzf//3f1y9epX+/fvz3HPP3Xaesnrv7vWKu7t5D4Dbvg8eHh5069aNJUuWoNPp+O677xg6dGiZ3kZq8ODBJCQk8Pnnn7N9+3YSEhKoW7duuV0wcbt98Morr3D8+HFeeukl9u3bR/v27e94Sq527do89dRTfPTRR+zZs4du3boZTlnfbQ43nyK++e9Dq9Xe8k/C7dowxcTEEBwcTO/evVm7di27d+/mvffeq5B9ChX3930/pLASogZr1qwZY8aMYcOGDTz77LMsWbIEAHNz81vadnh6enLt2jVDwQNw8eJFkpKSaNmyJQDNmze/pZ1KSe1WblS0nIkTJ9KzZ088PT25dOnS/W5esYoaW9/4Y7Jt2zZq165No0aN7no5L7zwAocPH+a33367ZZqiKGRmZmJra4uLiwvbtm0zmr5t2zbDPgOwtbUlKCiIRYsWsWLFClatWkV6ejqg/2Eprp1NSe/d/fD09LxjrmUpJCSEVatWsWrVKs6cOWNoWF9SbqdOneLUqVOGcQcOHCAjI6PE/LZt28abb75J7969adWqFRYWFly4cMEopqT9W+Ru38O74erqymuvvcYvv/zC2LFjWbRo0V3Pq9FoaNGiBTk5OaVa553Ur1+flJQUw+usrCySk5NLjN++fTuNGzfmvffeo3379nh4eHDixAmjmOK+P27m6enJnj17jLZn27ZtaLVamjdvfo9bUzlIYSVEDXT16lVCQ0PZvHkzJ06cYNu2bcTFxeHp6QnorxTKzs4mOjqaCxcucOXKFTw8POjbty/Dhg3j77//Zs+ePbz44os0bNjQ0IB75MiRrF+/ntmzZ3PkyBG++uorfv/99zsehahTpw5169Zl4cKFHD16lI0bNxIWFlYu2/7GG29w6tQpRo4cyaFDh/jtt9+YPHkyYWFhpbodRv/+/QkKCmLgwIF8/PHH7Ny5kxMnTrB27Vr8/f0Np7jefvttPvnkE1asWEFSUhLjx48nISGBUaNGATB79mx+/PFHDh06xOHDh1m5ciVOTk6GBtxubm5ER0eTmprKpUuX7vje3Y+3336biIgIFixYwJEjR5g9eza//PKL4eKFsvb8889jZmbGq6++yuOPP250mu9m/v7+tGnThuDgYHbt2sWOHTsYNGgQ3bt3v+WUZhEPDw++//57Dh48SGxsLMHBwVhZWRnF3Lx/i3On9/BujB49mj/++IPk5GR27drFpk2bSnzPEhIS6Nu3Lz///DMHDhzg6NGjfPPNNyxevLjYiyXux6OPPsr333/P1q1b2bdvH4MHDzY69XYzDw8PTp48yfLlyzl27Bhz58695Wipm5sbycnJJCQkcOHCBfLy8m5ZTnBwMJaWlgwePJj9+/ezadMmRo4cyUsvvWQ4vVxVSWElRA1kYmLCxYsXGTRoEM2aNaN///488cQTfPDBBwB07tyZ1157jaCgIOrXr8+MGTMAfTsYHx8fnnzySfz8/FAUhfXr1xsO13fp0oXw8HBmz56Nl5cXkZGRjBkz5o43Q9VqtSxfvpz4+Hhat27NmDFj+PTTT8tl2xs2bMj69evZsWMHXl5evPbaa4SEhDBx4sRSLUej0bBs2TJmz57N6tWr6d69O23btmXKlCn07duXgIAAQN+2JiwsjLFjx9KmTRsiIyNZs2YNHh4egP50z4wZM2jfvj0dOnTg33//Zf369YYib9asWURFReHq6kq7du3u+N7dj8DAQD7//HNmzpxJq1at+Oqrr1iyZAmPPPLIfS+7ONbW1gwYMIBLly7x8ssv3zZWo9Hw22+/UadOHR5++GH8/f1p0qQJK1asKHGeb775hkuXLvHQQw/x0ksv8eabb97Sj9fN+7c4d3oP70ZhYSEjRozA09OTXr160axZM778svh2kI0aNcLNzY0PPvgAX19fHnroIT7//HM++OAD3nvvvbte592YMGEC3bt358knn6RPnz4EBgby4IMPlhj/9NNPM2bMGEJDQ/H29mb79u28//77RjH9+vWjV69e9OjRg/r16/Pjjz/eshxra2v++OMP0tPT6dChA8899xw9e/Zk3rx5Zbp9atAod9MCUxhkZWVhZ2dnOMwvqo/c3FySk5ON+laqTNKycu95Xkdb9bZn2LBhHDp0iK1bt6qWgxCiZrjd93hF/X5XnS5bhRBVwsyZMw33pfv999/59ttvS/zPXAghqhsprIQQZWrHjh3MmDGDy5cv06RJE+bOncsrr7yidlpCCFEhpLASQpSpn376Se0UhBBCNdJ4XQghhBCijEhhJYQQQghRRqSwEkIIIYQoI1JYCSGEEEKUESmshBBCCCHKiBRWQgghhBBlRAorIUSN8O+//6LRaEhISFA7FSFENXZPhdX8+fNxc3PD0tISX19fduzYcdv4lStX0qJFCywtLWnTpg3r1683mq4oCpMmTcLZ2RkrKyv8/f05cuSIUUx6ejrBwcHY2tpib29PSEgI2dnZxa7v6NGj1K5d23AT09LkIkRV8+brwxjywvNqp1EtPPLII2g0GjQaDZaWlrRs2dKo1/iIiAjDdBMTE+rUqYOvry8ffvghmZmZRssaMmSIIfbG4ejRoxW9WUKIClTqwmrFihWEhYUxefJkdu3ahZeXFwEBAZw7d67Y+O3btzNw4EBCQkLYvXs3gYGBBAYGsn//fkPMjBkzmDt3LuHh4cTGxmJjY0NAQAC5uf/dGy04OJjExESioqJYu3YtW7ZsYfjw4besr6CggIEDB9KtW7d7ykUIoZ78/Hy1U2DYsGGkpKRw4MAB+vfvz4gRI4xuImtra0tKSgqnT59m+/btDB8+nO+++w5vb2/Onj1rtKxevXqRkpJiNLi7u1f0JgkhKpJSSh07dlRGjBhheF1YWKi4uLgo06ZNKza+f//+Sp8+fYzG+fr6Kq+++qqiKIqi0+kUJycn5dNPPzVMz8jIUCwsLJQff/xRURRFOXDggAIocXFxhpjff/9d0Wg0ypkzZ4yWPW7cOOXFF19UlixZotjZ2ZUql7uRmZmpAEpmZuZdzyOqhqtXryoHDhxQrl69qnYqxUrNvFrs0P+FF5VefZ4scfov6zYoHTp0UMzNzRUnJyflnXfeUQoKCgzLzcrKUl544QXF2tpacXJyUmbPnq10795dGTVq1G3zAZRFixYpgYGBipWVldK0aVPlt99+M4rZvHnzbdfdvXt3ZcSIEcqoUaOUunXrKo888oiyadMmBVAiIyMVb29vxdLSUunRo4eSlpamrF+/XmnRooVSu3ZtZeDAgUpOTo5hWb///rvSpUsXxc7OTnFwcFD69OmjHD161DA9OTlZAZTdu3eXuE3FbbeHh4cyYMAARVGUYr9XFEVR0tLSlHr16inBwcGGcYMHD1b69u17230ohChbt/ser6jf71IdscrPzyc+Ph5/f3/DOK1Wi7+/PzExMcXOExMTYxQPEBAQYIhPTk4mNTXVKMbOzg5fX19DTExMDPb29rRv394Q4+/vj1arJTY21jBu48aNrFy5kvnz599TLsXJy8sjKyvLaBA1hKJATo46g6KUySaknD1D8POBdOjQgT179rBgwQK++eYbpk6daogJCwtj27ZtrFmzhqioKLZu3cquXbvuavkffPAB/fv3Z+/evfTu3Zvg4GDS09MBOHPmDL17977tugG+/fZbzM3N2bZtG+Hh4YbxU6ZMYd68eWzfvp1Tp07Rv39/5syZw7Jly1i3bh0bNmzgiy++MMTn5OQQFhbGzp07iY6ORqvV8swzz6DT6e5nF2JlZXXHI2kNGjQgODiYNWvWUFhYeF/rE0JUbaW6V+CFCxcoLCzE0dHRaLyjoyOHDh0qdp7U1NRi41NTUw3Ti8bdLqZBgwbGiZua4uDgYIi5ePEiQ4YM4YcffsDW1vaecinOtGnT+OCDD0qcLqqxK1egVi111p2dDTY2972YiK8X0rBhI+bNm4dGo6FFixacPXuWd955h0mTJpGTk8O3337LsmXL6NmzJwBLlizBxcXlrpY/ZMgQBg4cCMDHH3/M3Llz2bFjB7169eLLL7/E1dW1xHVrtfr/6zw8PJgxY4ZhmSkpKQBMnTqVLl26ABASEsKECRM4duwYTZo0AeC5555j06ZNvPPOOwD069fPKLfFixdTv359Dhw4QOvWrUu97woLC/nxxx/Zu3dvsc0ObtaiRQsuX77MxYsXDd9Xa9eupdYNn6EnnniClStXljoXIUTVUW2uChw2bBgvvPACDz/8cJkud8KECWRmZhqGU6dOlenyhShPR5IO4dPRF41GYxjXpUsXsrOzOX36NMePH6egoICOHTsaptvZ2dG8eXPD648//phatWoZhpMnTxqmtW3b1vDcxsYGW1tbQ3vLgwcP4ufnV+K6i/j4+BSb+43LdnR0xNra2lBUFY27sW3nkSNHGDhwIE2aNMHW1hY3NzcAo3zvxpdffkmtWrWwsrJi2LBhjBkzhtdff/2O8ynXjzLeuL09evQgISHBMMydO7dUuQghqp5SHbGqV68eJiYmpKWlGY1PS0vDycmp2HmcnJxuG1/0mJaWhrOzs1GMt7e3IebmxvHXrl0jPT3dMP/GjRtZs2YNM2fOBPRfcjqdDlNTUxYuXMjLL798x1yKY2FhgYWFRYnTRTVmba0/cqTWuiuJ1157jf79+xte33g0y8zMzChWo9GU+tSbTQlH5m5ctkajueO6nnrqKRo3bsyiRYtwcXFBp9PRunXrUjeIDw4O5r333sPKygpnZ2fDkbU7OXjwILa2ttStW9cwzsbGhqZNm5Zq/UKIqq1UR6zMzc3x8fEhOjraME6n0xEdHY2fn1+x8/j5+RnFA0RFRRni3d3dcXJyMorJysoiNjbWEOPn50dGRgbx8fGGmI0bN6LT6fD19QX07adu/M/www8/pHbt2iQkJPDMM8/cVS5CGNFo9Kfj1BhuOOpxPzyatyB+R6zhaArAtm3bqF27No0aNaJJkyaYmZkRFxdnmJ6Zmcnhw4cNrx0cHGjatKlhMDW9u//HPD09iYmJKXHdZenixYskJSUxceJEevbsiaenJ5cuXbqnZdnZ2dG0aVMaNmx410XVuXPnWLZsGYGBgXc9jxCieirVESvQN3QdPHgw7du3p2PHjsyZM4ecnByGDh0KwKBBg2jYsCHTpk0DYNSoUXTv3p1Zs2bRp08fli9fzs6dO1m4cCGg/69z9OjRTJ06FQ8PD9zd3Xn//fdxcXEhMDAQ0H9B9+rVi2HDhhEeHk5BQQGhoaEMGDDA8N+zp6enUZ47d+5Eq9Uata24Uy5CVFVZWVns37vHaFwdBweGvDKchQvmMXLkSEJDQ0lKSmLy5MmEhYWh1WqpXbs2gwcP5u2338bBwYEGDRowefJktFqt0Smte/HGG28wZ86cEtddlurUqUPdunVZuHAhzs7OnDx5kvHjx5fpOoooikJqaiqKopCRkUFMTAwff/wxdnZ2TJ8+vVzWKYSoOkpdWAUFBXH+/HkmTZpEamoq3t7eREZGGhqFnzx50uhLs3PnzixbtoyJEyfy7rvv4uHhwerVq40KnnHjxpGTk8Pw4cPJyMiga9euREZGYmlpaYhZunQpoaGh9OzZE61WS79+/UrdXuFuchGiKtq+dQv+3ToZjXvhpSHMnreApStXM23Ke3h5eeHg4EBISAgTJ040xM2ePZvXXnuNJ598EltbW8aNG8epU6eM/v7uRcOGDVm/fj1vv/12iesuK1qtluXLl/Pmm2/SunVrmjdvzty5c3nkkUfKfF1ZWVk4Ozuj0WiwtbWlefPmDB48mFGjRpV44YwQoubQKEoZXdddQ2RlZWFnZ0dmZqZ8iVYzubm5JCcn4+7uft9FRXlIy8q9c1AJHG3vfntycnJo2LAhs2bNIiQk5J7XKYQQFe123+MV9ftd6iNWQojqZffu3Rw6dIiOHTuSmZnJhx9+CEDfvn1VzkwIIaoeKayEEMycOZOkpCTDBSpbt26lXr16aqclhBBVjhRWQtRw7dq1M7riVgghxL2T64KFEEIIIcqIFFZCCCGEEGVECishbiIXygohRNVUGb6/pbAS4rqiW6ZcuXJF5UyEEELci6Lv75tvgVWRpPG6ENeZmJhgb29vuC+ltbX1ffc+XpYK8vPued7ce+8CSwghKj1FUbhy5Qrnzp3D3t4eExMT1XKRwkqIGxTdkPvmm35XBllXC+553stW6v33JoQQFcXe3t7wPa4WKayEuIFGo8HZ2ZkGDRpQUHDvhUx5iNiWfM/zDmnpXoaZCCFE5WNmZqbqkaoiUlgJUQwTE5NK8Qd6o6u6e8+nMt6iRwghqiNpvC6EEEIIUUaksBJCCCGEKCNSWAkhhBBClBEprIQQQgghyogUVkIIIYQQZUSuChQ1Q2EhJCbCsWOQkgI6HVhagqsrNG8OjRtDJeoMVAghRNUkhZWovvLyYO1a+OEH2LgRsrJKjm3YEB57DF54AR59FCpZVwtCCCGqBimsRPWTlwdffw3TpsGZM/+Nr10bPD3BxQXMzCAnB06ehKQkfVxEhH544AF46y145RWwslJrK4QQQlRBUliJ6uXvvyEkBA4f1r92doZBg+C556Bdu+KPRF25AjExsGoVrFihL7befBM+/hg+/RSCg+U0oRBCiLsijddF9XDtGrzzDnTrpi+qHB1h/nxITobp06F9+5JP71lbQ8+e8OWX+iNXCxbo21ylpsJLL0H37nD0aMVujxBCiCpJCitR9aWnwxNPwIwZ+tcvvwwHD8Ibb4CFRemWZWkJr72mPz348cf6omvrVv3RrogIUJQyT18IIUT1IYWVqNpSUuDhh+HPP/VF0IoV8M03UKfO/S3XwgImTIADB/TLz86GoUP1RVteXtnkLoQQotqRNlZCNZ9FHb7necc81kzfFqpnT/1pOhcX+P13aNu2DDNEf0pw40b90bCJEyEigrM79rBm8nyu1qlbtusSQghR5ckRK1E1XbwIjz+uL6rc3PSn68q6qCpiYqI/evX77+Ta1MblwG4Gvvk8dmdPls/6hBBCVFlSWIkqxzT3Kjz1lL4dVKNGsGULNGlS/it+/HGWz/2JDJcHsEs7Q/+wF3A4IY3ahRBC/EcKK1G1KAqPz5qg7x6hTh344w997+kV5JJrE1bMXsYFt2bUSj9P/7HB1D96oMLWL4QQonKTwkpUKe1++Zbmf/2u7+Dzt9+gZcsKz+GKQ31+mvk9qc3bYJWVwbMTXsbh5LEKz0MIIUTlI4WVqDIa7tvJw4uud6kwe7a+zyqV5Nnas2r6EtI8WmGdeYlnxw/FNvW0avkIIYSoHKSwElWCec5lnpg2Fq2ukEM9noQRI9ROiXyb2vzy8ddcbNyU2hfS6PfOEKwvXVA7LSGEECq6p8Jq/vz5uLm5YWlpia+vLzt27Lht/MqVK2nRogWWlpa0adOG9evXG01XFIVJkybh7OyMlZUV/v7+HDlyxCgmPT2d4OBgbG1tsbe3JyQkhOzsbMP0pKQkevTogaOjI5aWljRp0oSJEydSUFBgiImIiECj0RgNlpaW97ILRAV7ZMHH1L6QSobLA/w5+sNKc4uZXDsHVk1fTIazK/Ypp3hqyghM8nLVTksIIYRKSl1YrVixgrCwMCZPnsyuXbvw8vIiICCAc+fOFRu/fft2Bg4cSEhICLt37yYwMJDAwED2799viJkxYwZz584lPDyc2NhYbGxsCAgIIDf3vx+o4OBgEhMTiYqKYu3atWzZsoXhw4cbppuZmTFo0CA2bNhAUlISc+bMYdGiRUyePNkoH1tbW1JSUgzDiRMnSrsLRAVrErORVht+QdFo+OOtaRRY2aidkpGcuo78+tEicmvb4XIwgYCZ40GnUzstIYQQKtAoSunu0eHr60uHDh2YN28eADqdDldXV0aOHMn48eNviQ8KCiInJ4e1a9caxnXq1Alvb2/Cw8NRFAUXFxfGjh3LW2+9BUBmZiaOjo5EREQwYMAADh48SMuWLYmLi6N9+/YAREZG0rt3b06fPo2Li0uxuYaFhREXF8fWrVsB/RGr0aNHk5GRUZpNNpKVlYWdnR2ZmZnY2tre83LE3XUQap5zmSEhT2CTfp6dz73M1uHvANc7CFXB7XJuuHcH/ca/jMm1AmIHvsb2oWMqMLPbU2t/CSFEZVFRv9+lOmKVn59PfHw8/v7+/y1Aq8Xf35+YmJhi54mJiTGKBwgICDDEJycnk5qaahRjZ2eHr6+vISYmJgZ7e3tDUQXg7++PVqslNja22PUePXqUyMhIunfvbjQ+Ozubxo0b4+rqSt++fUlMTCzFHhAVze+7L7BJP096Ize2Dxmtdjq3daZtR/1pSsD3x3Ca/bX+DnMIIYSobkpVWF24cIHCwkIcHR2Nxjs6OpKamlrsPKmpqbeNL3q8U0yDBg2MppuamuLg4HDLejt37oylpSUeHh5069aNDz/80DCtefPmLF68mN9++40ffvgBnU5H586dOX265Ku58vLyyMrKMhpExah3/BDev/0AwKY33qfQvJQ3VFbBgcefZefzIQA8Nus96YZBCCFqmGp3VeCKFSvYtWsXy5YtY926dcycOdMwzc/Pj0GDBuHt7U337t355ZdfqF+/Pl999VWJy5s2bRp2dnaGwbUCO6Os0RSFR+d9iFZXyOFuAZxs31XtjO7a3y+HcdLLF/PcKzz5QShmV7LvPJMQQohqoVSFVb169TAxMSEtLc1ofFpaGk5OTsXO4+TkdNv4osc7xdzcOP7atWukp6ffsl5XV1datmzJwIEDmT59OlOmTKGwsLDY3MzMzGjXrh1Hj5Z8W5IJEyaQmZlpGE6dOlVirCg7zf76nYb74ymwsOKvVyeonU6pKCam/P7ubC7Xc6TuqeM8PutdKF1TRiGEEFVUqQorc3NzfHx8iI6ONozT6XRER0fj5+dX7Dx+fn5G8QBRUVGGeHd3d5ycnIxisrKyiI2NNcT4+fmRkZFBfHy8IWbjxo3odDp8fX1LzFen01FQUICuhCu0CgsL2bdvH87OziUuw8LCAltbW6NBlC9tQT5dIj4DIC5oGNkNSn5/KqsrdeqxbuLnFJqa0WzrH3itWap2SkIIISqAaWlnCAsLY/DgwbRv356OHTsyZ84ccnJyGDp0KACDBg2iYcOGTJs2DYBRo0bRvXt3Zs2aRZ8+fVi+fDk7d+5k4cKFAGg0GkaPHs3UqVPx8PDA3d2d999/HxcXFwIDAwHw9PSkV69eDBs2jPDwcAoKCggNDWXAgAGGKwKXLl2KmZkZbdq0wcLCgp07dzJhwgSCgoIwMzMD4MMPP6RTp040bdqUjIwMPv30U06cOMErr7xy3ztSlJ3Wv6/E/uxJcurUY1e/IWqnc89SWrZj67C3eWTBxzy88BNOt+3IRXe5Ok8IIaqzUhdWQUFBnD9/nkmTJpGamoq3tzeRkZGGxucnT55Eq/3vQFjnzp1ZtmwZEydO5N1338XDw4PVq1fTunVrQ8y4cePIyclh+PDhZGRk0LVrVyIjI40671y6dCmhoaH07NkTrVZLv379mDt37n8bYmrKJ598wuHDh1EUhcaNGxMaGsqYMf9d8n7p0iWGDRtGamoqderUwcfHh+3bt9NShfvNieKZXc2h0w/zAfgn+I1K12dVae0OHETjnX/jHreF3tPGsuyLlRRaSKe0QghRXZW6H6uaTvqxKjvF9QnVcemXdPn2cy65NOa7r9ehMzUrdt7K2I9VSawuXeSlV5/CJuMiu/u+xOYRE8shs9uTfqyEEDVdpezHSojyZHYlG59fIgCIGfxmiUVVVXO1Tl3+eHs6AO1++x73fzapnJEQQojyIoWVqDS81v6I5eVM0hu5c/jhJ9ROp0yd6PAwu54ZDMBjn03EMuuSyhkJIYQoD1JYiUrBJC+Xh35eAkDcgFdRTExUzqjs/R0ylouNm2Jz6QI95k9VOx0hhBDlQAorUSm0+X0lNhkXyXRsyKFHn1Q7nXJRaG7BH2OnodNqabFpLQ/+HaV2SkIIIcqYFFZCddqCfNr/9DUAO4OGVZu2VcVJa9GWnc/ru/foOXcylpnpKmckhBCiLElhJVTX7K/fqX0hlWyH+iQ+/qza6ZS7f14ayYXGHthkXJRTgkIIUc1IYSXUpSg89Mu3AOx5OrhK3Gj5fhWam7Ph7WnotCa02LyOpn9vUDslIYQQZaTUHYQKcaN76dfpRi6J8TgeTeSauQX7egdVyHorQ59Oac3aEBc0DN8fw+kx//842c6PfJvaaqclhBDiPskRK6Gqdr9+B8DBnk9z1d5B5WwqVuwLr3PJpTG1Lp6jc8QctdMRQghRBqSwEqqpnXaGptv0V8btDhykcjYVr9DCkuhRHwDgvWYpTgf3qJyREEKI+yWFlVCN928/oNXpONGuc429OfGpdn4c8O+LRlHw//x9tNcK1E5JCCHEfZDCSqjCJD+PVn/8AkBC4EsqZ6OuLcPHc7W2PfWPJ9HuekN+IYQQVZMUVkIVTbdFYXU5g6z6ziR37K52Oqq6au/AluHjAPD7/gtsU0+rnJEQQoh7JYWVUEWb9SsASOz1XLW8fU1pHXj8WU55dcQsL5ce8/9P7XSEEELcIymsRIWzP52M654d6LRa9gf0UzudykGjIXrkFApNzWgSuxn3fzapnZEQQoh7IIWVqHBt1v8EwL8dHia7gbPK2VQelx54kF3PDgbgkfCPMcnPUzkjIYQQpSWFlahQJvn5tIz6FYB9vfurnE3lE/vC62TXbYD92ZP4/LxY7XSEEEKUkhRWokI9GPMn1pmXyK7boMY3Wi9OgXUttgx/B4COP4ZT+9xZlTMSQghRGlJYiQrlGbUagMTHn0UxkTsqFSfpkT6cbtMBs7xcHl74idrpCCGEKAUprESFsb50AbedfwNw0L+vytlUYhoNm0ZMRKfV0mxLJK67Y9TOSAghxF2SwkpUmOab16HVFZLSvC2XXJuonU6ldqFJC/Y8FQxAj/n/Jz2yCyFEFSGFlagwnn/+BsjRqrsVM/hNrtg5UPfkMbzWLFU7HSGEEHdBCitRIRxOHsPxSCKFJqYkPdJb7XSqhLxatmx7OQyATj/MxzLrksoZCSGEuBMprESFKDpa9W+HbuTaOaicTdWR+PiznGvSAsvsLDp9P1/tdIQQQtyBFFai/Ol0tNj4P0BOA5aWYmLCllfHA+D1v2XUOXlM5YyEEELcjhRWotw13L8T23NnybWpzfFOj6qdTpVzqp0fxzr1QKsr5OFFM9RORwghxG1IYSXKXfPN6wE42vVxCs0tVM6matoy/B0KTUxpEruZB+K3qZ2OEEKIEkhhJcqVpvAaTf/eAEBSd2m0fq8yGrmz5+kXAOj+1XQ0hddUzkgIIURxpLAS5arR3jhsMi5ytbY9p7191U6nSvsneAS5te2o9+9hWkeuUjsdIYQQxZDCSpQrj62RABzt+hg6UzOVs6na8mztiXkxFAC/bz/HPCdb5YyEEELcTAorUW40hdfw2Ko/DXj44SdUzqZ62PvUQNIbuWGTcZGOy8PVTkcIIcRN7qmwmj9/Pm5ublhaWuLr68uOHTtuG79y5UpatGiBpaUlbdq0Yf369UbTFUVh0qRJODs7Y2Vlhb+/P0eOHDGKSU9PJzg4GFtbW+zt7QkJCSE7+7//2JOSkujRoweOjo5YWlrSpEkTJk6cSEGB8a1A7pSLKDuN9sZhnZnOVVt7TslpwDKhMzVj67B3AGj3SwS1086onJEQQogblbqwWrFiBWFhYUyePJldu3bh5eVFQEAA586dKzZ++/btDBw4kJCQEHbv3k1gYCCBgYHs37/fEDNjxgzmzp1LeHg4sbGx2NjYEBAQQG5uriEmODiYxMREoqKiWLt2LVu2bGH48OGG6WZmZgwaNIgNGzaQlJTEnDlzWLRoEZMnTy5VLqLsNNvyOwBHuzyOYmKqcjbVx/FOPTjp5YtpQQGdv/1c7XSEEELcQKMoilKaGXx9fenQoQPz5s0DQKfT4erqysiRIxk/fvwt8UFBQeTk5LB27VrDuE6dOuHt7U14eDiKouDi4sLYsWN56623AMjMzMTR0ZGIiAgGDBjAwYMHadmyJXFxcbRv3x6AyMhIevfuzenTp3FxcSk217CwMOLi4ti6detd5XI3srKysLOzIzMzE1tb27uapzr7LOpwseM1hdcYPqAr1pmXWDVtMSd9ulRwZiUb81ize563pO2taI6H9/FC6HMoGg1Lv/yV8w963jb+frZZCCGqg4r6/S7VEav8/Hzi4+Px9/f/bwFaLf7+/sTExBQ7T0xMjFE8QEBAgCE+OTmZ1NRUoxg7Ozt8fX0NMTExMdjb2xuKKgB/f3+0Wi2xsbHFrvfo0aNERkbSvXv3u86lOHl5eWRlZRkN4s70pwEvyWnAcpLWrA2HHumDRlHo+vVMtdMRQghxXakKqwsXLlBYWIijo6PReEdHR1JTU4udJzU19bbxRY93imnQoIHRdFNTUxwcHG5Zb+fOnbG0tMTDw4Nu3brx4Ycf3nUuxZk2bRp2dnaGwdXVtcRY8Z+m26IAONrZX04DlpPtQ8dQaGqGW/zf0mmoEEJUEtXuqsAVK1awa9culi1bxrp165g58/7+m58wYQKZmZmG4dSpU2WUaTWmKDy4PRqAY5397xAs7lWmsyt7nhoIQNdvZoJOp3JGQgghSlVY1atXDxMTE9LS0ozGp6Wl4eTkVOw8Tk5Ot40verxTzM2N469du0Z6evot63V1daVly5YMHDiQ6dOnM2XKFAoLC+8ql+JYWFhga2trNIjba3AkkdoXUimwsOJkOz+106nWYl94nTzrWjgePUCLTWvvPIMQQohyVarCytzcHB8fH6Kjow3jdDod0dHR+PkV/wPq5+dnFA8QFRVliHd3d8fJyckoJisri9jYWEOMn58fGRkZxMfHG2I2btyITqfD17fk9js6nY6CggJ01/+Tv1Muomw8GKPfx/926EahhaXK2VRvuXYOxAXpr47tHDEHk/x8lTMSQoiardSNX8LCwhg8eDDt27enY8eOzJkzh5ycHIYOHQrAoEGDaNiwIdOmTQNg1KhRdO/enVmzZtGnTx+WL1/Ozp07WbhwIQAajYbRo0czdepUPDw8cHd35/3338fFxYXAwEAAPD096dWrF8OGDSM8PJyCggJCQ0MZMGCA4YrApUuXYmZmRps2bbCwsGDnzp1MmDCBoKAgzMzM7ioXUTaKCqtjfj1VzqRm2P3MILz+txS7tDN4rVnKrueGqp2SEELUWKUurIKCgjh//jyTJk0iNTUVb29vIiMjDY3CT548iVb734Gwzp07s2zZMiZOnMi7776Lh4cHq1evpnXr1oaYcePGkZOTw/Dhw8nIyKBr165ERkZiafnf0Y6lS5cSGhpKz5490Wq19OvXj7lz5/63IaamfPLJJxw+fBhFUWjcuDGhoaGMGTOmVLmI+2Obcor6x5PQaU1I9u1+5xnEfbtmaUXMoDd5fPZ7+P64gMSAZ8mrbad2WkIIUSOVuh+rmk76sTJ2c79O7X79lkcWfMypth35eeb3KmV1e9WhH6ubaQoLefG1vtQ7cYS4/q/w9ytvG02XfqyEEDVdpezHSog7eXD7n4CcBqxoiokJf4eMBaDdr99R61yKyhkJIUTNJIWVKDMWWRk03Ke/wOB4ZymsKlqy7yOcatsR04J8udWNEEKoRAorUWaa7NiMVlfIefdmZDpLR6oVTqNh6/VTgC3/XE3d5Mp52lIIIaozKaxEmXGP3QzA8U6PqptIDZbWoi1HugagURQ6R8xROx0hhKhxpLASZUJTeI3GO/8G4HinHipnU7NtGzIKnVZL05honA/sVjsdIYSoUaSwEmXC5UACljmXuWprT1qzNmqnU6NdeuBBEh9/FoAui2eDXPgrhBAVRgorUSbc4rYA8K9PVxQTE5WzEf+8GMo1M3Nc9+6gcfzfaqcjhBA1hhRWoky47fgLgH87SqeglUF2A2cS+r4IXD9qJTdoFkKICiGFlbhvNhfSaHD8EIpGw7/tu6qdjrgubsBwww2aWblS7XSEEKJGkMJK3Lei04CpzduSa+egcjaiSK5tHXY+H6J/MXEiFBSom5AQQtQAUliJ++Z+vbBK7viwypmIm+1+djA59nXh6FFYvFjtdIQQotqTwkrcF21BPg/s2gZI+6rKqMDKhh0vvK5/8cEHcOWKugkJIUQ1J4WVuC8uB3ZjcSWHHPu6pDVtpXY6ohj7egeBmxukpMC8eWqnI4QQ1ZoUVuK+uF+/GvBE+26glY9TZVRobg4ffqh/MW0aXLqkbkJCCFGNyS+huC9uO4raV8lpwErthRegdWvIyIBPP1U7GyGEqLZM1U5AVGGnT1PvxBF0Wi0nHuqsdjbidkxM4OOP4emnYc4cGDkSnJ3LfbWfRd37jaDHPNasDDMRQoiKIUesxL2LigIgrVkb8mzt1c1F3NmTT4KfH1y9Cv/3f2pnI4QQ1ZIUVuLeXS+s5GhVFaHRwPTp+ueLFum7YBBCCFGmpLAS90angz//BODkQ11UTkbctYcfhieegGvXYNIktbMRQohqRworcW/27oXz58m3tCbF00vtbERpfPyx/vHHHyEhQdVUhBCiupHCStyb66cBT3t1RGdmrnIyolS8vWHgQP3z995TNRUhhKhupLAS9+Z6YXVS2ldVTR9+CKamsH49bNmidjZCCFFtSGElSi83F7ZuBeCEtK+qmpo2hWHD9M8nTABFUTcfIYSoJqSwEqW3bZu+uHJxIf2BB9XORtyr998HKyvYvh3WrlU7GyGEqBaksBKld/00IP7++kv4RdXk7AyjRumfT5gAhYXq5iOEENWAFFai9IoKq8ceUzcPcf/eeQfq1IHERFi6VO1shBCiypPCSpTOhQuwe7f+ub+/urmI+2dvD+PH659PmgR5eaqmI4QQVZ0UVqJ0oqP1DZ3btAEnJ7WzEWVh5EhwcYETJyA8XO1shBCiSpPCSpSOnAasfqysYMoU/fOpUyErS9V0hBCiKpPCStw9RTHcxkYKq2pm6FBo1kx/qnf2bLWzEUKIKuueCqv58+fj5uaGpaUlvr6+7Nix47bxK1eupEWLFlhaWtKmTRvWr19vNF1RFCZNmoSzszNWVlb4+/tz5MgRo5j09HSCg4OxtbXF3t6ekJAQsrOzDdM3b95M3759cXZ2xsbGBm9vb5be1Bg3IiICjUZjNFhaWt7LLqiZkpP1p4tMTaFbN7WzEWXJ1BQ++kj/fNYsOHdO3XyEEKKKKnVhtWLFCsLCwpg8eTK7du3Cy8uLgIAAzpXwRbx9+3YGDhxISEgIu3fvJjAwkMDAQPbv32+ImTFjBnPnziU8PJzY2FhsbGwICAggNzfXEBMcHExiYiJRUVGsXbuWLVu2MHz4cKP1tG3bllWrVrF3716GDh3KoEGDWHtT/zy2trakpKQYhhMnTpR2F9RcmzfrHzt2BBsbVVMR5aBfP2jfHrKz/yuyhBBClIpGUUrX5bKvry8dOnRg3rx5AOh0OlxdXRk5ciTji64uukFQUBA5OTlGBU6nTp3w9vYmPDwcRVFwcXFh7NixvPXWWwBkZmbi6OhIREQEAwYM4ODBg7Rs2ZK4uDjat28PQGRkJL179+b06dO4uLgUm2ufPn1wdHRk8eLFgP6I1ejRo8nIyCjNJhvJysrCzs6OzMxMbG1t73k5VdJLL8EPP+jvLzd1KgCfRR1WOanSG/NYs3uetypuL5Rim6Oj9Vd7mplBUhK4u9/Xeu9nf93P+ySEEDerqN/vUh2xys/PJz4+Hv8bLrPXarX4+/sTExNT7DwxMTFG8QABAQGG+OTkZFJTU41i7Ozs8PX1NcTExMRgb29vKKoA/P390Wq1xMbGlphvZmYmDg4ORuOys7Np3Lgxrq6u9O3bl8TExNtuc15eHllZWUZDjaQosGmT/nmPHurmIspPz5769nMFBfruF4QQQpRKqQqrCxcuUFhYiKOjo9F4R0dHUlNTi50nNTX1tvFFj3eKadCggdF0U1NTHBwcSlzvTz/9RFxcHEOHDjWMa968OYsXL+a3337jhx9+QKfT0blzZ06fPl3iNk+bNg07OzvD4OrqWmJstXb0KJw5A+bm0FluvFytTZumf1y6FPbuVTcXIYSoYqrlVYGbNm1i6NChLFq0iFatWhnG+/n5MWjQILy9venevTu//PIL9evX56uvvipxWRMmTCAzM9MwnDp1qiI2ofIpOlrVqZP+8nxRffn4QP/++qOU776rdjZCCFGllKqwqlevHiYmJqSlpRmNT0tLw6mEziKdnJxuG1/0eKeYmxvHX7t2jfT09FvW+9dff/HUU0/x2WefMWjQoNtuj5mZGe3atePo0aMlxlhYWGBra2s01EhyGrBm+b//AxMTWLcOtm5VOxshhKgySlVYmZub4+PjQ3R0tGGcTqcjOjoaPz+/Yufx8/MzigeIiooyxLu7u+Pk5GQUk5WVRWxsrCHGz8+PjIwM4uPjDTEbN25Ep9Ph6+trGLd582b69OnDJ598YnTFYEkKCwvZt28fzs7Od7H1NZi0r6p5mjWDV17RPx8/Xv8ZEEIIcUelPhUYFhbGokWL+Pbbbzl48CCvv/46OTk5hrZMgwYNYsKECYb4UaNGERkZyaxZszh06BBTpkxh586dhIaGAqDRaBg9ejRTp05lzZo17Nu3j0GDBuHi4kJgYCAAnp6e9OrVi2HDhrFjxw62bdtGaGgoAwYMMFwRuGnTJvr06cObb75Jv379SE1NJTU1lfT0dEMuH374IRs2bOD48ePs2rWLF198kRMnTvBK0Q+IKF5SEqSlgYUF3FDIimpu0iT9ad/t2+GmbkuEEEIUr9SFVVBQEDNnzmTSpEl4e3uTkJBAZGSkofH5yZMnSUlJMcR37tyZZcuWsXDhQry8vPj5559ZvXo1rVu3NsSMGzeOkSNHMnz4cDp06EB2djaRkZFGnXcuXbqUFi1a0LNnT3r37k3Xrl1ZuHChYfq3337LlStXmDZtGs7Ozobh2WefNcRcunSJYcOG4enpSe/evcnKymL79u20bNmytLuhZik6WtW5M0iHqjWHiwuMGqV/PmECFBaqm48QQlQBpe7Hqqarkf1Y9e8PK1fChx/C++8bTaqK/TpJP1alcOkSNGkCGRnw7bdwh3aLN5N+rIQQlUWl7MdK1ECK8l+P69K+quapU0d/tAr0pwbz8tTNRwghKjkprMTtJSbC+fNgba2/lY2oeUJD9acFT5yA8HC1sxFCiEpNCitxe0Xtq7p00XcOKmoea2uYMkX/fOpUqKl3HxBCiLsghZW4PTkNKACGDtV3wXDhAsyerXY2QghRaUlhJUqm00lhJfRMTeGjj/TPZ82CmzrsFUIIoSeFlSjZ/v2Qng42NvrbnIiarV8/aN8esrP/K7KEEEIYkcJKlKzoViZduoCZmbq5CPVpNDB9uv75ggWQnKxuPkIIUQlJYSVK9vff+seuXdXNQ1QePXuCvz8UFMDEiWpnI4QQlY4UVqJ4ivLfESsprMSNPvlE/7hsGdxw/04hhBBSWImSnDwJZ87oGy3L/QHFjR56CF58Uf/87bflBs1CCHEDKaxE8YqOVvn46PsxEuJGU6fqb8q9aROsX692NkIIUWlIYSWKJ+2rxO00bvzfDZrHjYNr19TNRwghKgkprETxpLASdzJhAtStCwcOwJIlamcjhBCVghRW4lbp6fp7BIK+qwUhimNvD++/r38+aZK+fyshhKjhpLASt9q+Xf/YvDnUr69uLqJye/11ePBBSE3V98guhBA1nBRW4lZyGlDcLXNzmDZN//zTT/UFlhBC1GBSWIlbFRVW3bqpm4eoGp57Djp1gpwcmDxZ7WyEEEJVUlgJY7m5EBenfy5HrMTd0Ghg5kz986+/1jdmF0KIGkoKK2EsLg7y88HJCZo0UTsbUVV06QLPPAM6HbzzjtrZCCGEaqSwEsZubF+l0aibi6hapk/X99S/di1s3qx2NkIIoQoprIQxabgu7lWzZvDqq/rnb72lP3olhBA1jBRW4j86HWzbpn8uhZW4F5MnQ+3a+psz//ij2tkIIUSFk8JK/CcxETIzoVYt8PJSOxtRFdWvD+PH659PmIBJXq66+QghRAWTwkr8p+g0oJ+fvq2MEPdizBhwdYVTp/D5ebHa2QghRIWSX0/BZ1GHAei1cj2ewHbnFsReH1cdfVaNt61SsLKCGTNg4EA6Ll9IYq9+5NR1VDuru3Y/n48xjzUrw0yEEFWRHLESBg33xwNwtpWPypmIKi8oCPz8MMu7SpfFn6mdjRBCVBgprAQAtc+dxfZ8CjqtCSktpH2VuE8aDcyZA0CrqF9xPLxP3XyEEKKCSGElAHC5frTqnEdLrllZq5yNqBY6duSAf18Aui/4GBRF5YSEEKL8SWElgP9OA56R04CiDG0bGkaBhRUNE3fRbMvvaqcjhBDlTgorAYBL4vX2Va2lsBJlJ7u+E3FBrwDQ9euZ0v2CEKLak8JKYHE5k3r/HgHgbKuHVM5GVDfxz4VwuZ4TdmlneOiXCLXTEUKIcnVPhdX8+fNxc3PD0tISX19fduzYcdv4lStX0qJFCywtLWnTpg3r1683mq4oCpMmTcLZ2RkrKyv8/f05cuSIUUx6ejrBwcHY2tpib29PSEgI2dnZhumbN2+mb9++ODs7Y2Njg7e3N0uXLi11LjWR88EENIpCeiM3rtSpp3Y6opq5ZmnF1lfeAqDjj19hc/GcyhkJIUT5KXVhtWLFCsLCwpg8eTK7du3Cy8uLgIAAzp0r/sty+/btDBw4kJCQEHbv3k1gYCCBgYHs37/fEDNjxgzmzp1LeHg4sbGx2NjYEBAQQG7uf6cNgoODSUxMJCoqirVr17JlyxaGDx9utJ62bduyatUq9u7dy9ChQxk0aBBr164tVS41UcN9OwHpZkGUn6QeT3LW0xvz3Ct0jpDuF4QQ1ZdGUUp3qY6vry8dOnRg3rx5AOh0OlxdXRk5ciTji25lcYOgoCBycnKMCpxOnTrh7e1NeHg4iqLg4uLC2LFjeest/X+1mZmZODo6EhERwYABAzh48CAtW7YkLi6O9u3bAxAZGUnv3r05ffo0Li4uxebap08fHB0dWbx48V3lcjeysrKws7MjMzMTW1vbu5qnsjvTpj0N98fzx9iPORDQT+10RDlQq+PKGzvbdDq4h4Gj+qNoNCz74mfONWt923krQ86lJR2EClF5VdTvd6mOWOXn5xMfH4+/v/9/C9Bq8ff3JyYmpth5YmJijOIBAgICDPHJycmkpqYaxdjZ2eHr62uIiYmJwd7e3lBUAfj7+6PVaomNjS0x38zMTBwcHO46l+Lk5eWRlZVlNFQrubk4Ju0FpOG6KF+pnl4c7Pk0GkXhkQUfSfcLQohqqVSF1YULFygsLMTR0fj2FI6OjqSmphY7T2pq6m3jix7vFNOgQQOj6aampjg4OJS43p9++om4uDiGDh1617kUZ9q0adjZ2RkGV1fXEmOrpPh4TAsKyLGvS4ZLY7WzEdXc3y+PNXS/0CJ6jdrpCCFEmauWVwVu2rSJoUOHsmjRIlq1anVfy5owYQKZmZmG4dSpU2WUZSVx/cbLZ1v76HvLFqIcZdd3Ijb4DQAe/vpTzHOy7zCHEEJULaW6CXO9evUwMTEhLS3NaHxaWhpOTk7FzuPk5HTb+KLHtLQ0nJ2djWK8vb0NMTc3jr927Rrp6em3rPevv/7iqaee4rPPPmPQoEGlyqU4FhYWWFhYlDi9yrteWEnHoKKi7Hp2CK3+WEWdM//iu3Q+W4e/U+brkBttCyHUUqojVubm5vj4+BAdHW0Yp9PpiI6Oxs/Pr9h5/Pz8jOIBoqKiDPHu7u44OTkZxWRlZREbG2uI8fPzIyMjg/j4eEPMxo0b0el0+Pr6GsZt3ryZPn368MknnxhdMXi3udQ4Oh1s2wbA2TZSWImKUWhuzqY33gOg3a/f4XDiqMoZCSFE2Sn1qcCwsDAWLVrEt99+y8GDB3n99dfJyckxtGUaNGgQEyZMMMSPGjWKyMhIZs2axaFDh5gyZQo7d+4kNDQUAI1Gw+jRo5k6dSpr1qxh3759DBo0CBcXFwIDAwHw9PSkV69eDBs2jB07drBt2zZCQ0MZMGCA4YrATZs20adPH95880369etHamoqqamppKen33UuNc6BA3DpEvmW1px70FPtbEQNcqLDwxzzexSTwmv0+HKqNGQXQlQbpS6sgoKCmDlzJpMmTcLb25uEhAQiIyMNjcJPnjxJSkqKIb5z584sW7aMhQsX4uXlxc8//8zq1atp3fq/S63HjRvHyJEjGT58OB06dCA7O5vIyEgsLS0NMUuXLqVFixb07NmT3r1707VrVxYuXGiY/u2333LlyhWmTZuGs7OzYXj22WdLlUuNcv00YKqnF4pJqc4KC3Hf/nrtXa6ZmfPA7hg8tv6hdjpCCFEmSt2PVU1XrfqxevFFWLqUmBdD+WfQSLWzEeWosvYJ5ffdXDr9MJ+s+s58+/V6rllZG6bdT85qtbGSfqyEqLwqZT9Wopq58YpAIVQQ138YmY4NsT2fQsflX6mdjhBC3DcprGqqU6fgxAkwMSGlRVu1sxE11DVLK/56Vd8m0+fnb7A7c0LljIQQ4v5IYVVTXb8aEG9vCqxrqZuLqNGOdfHnX5+umBYU8Ej4x2qnI4QQ90UKq5rq+mlAunZVNw8hNBo2v/EehaZmNIndTJOY6DvPI4QQlZQUVjXV1q36x27d1M1DCOCSaxN29RsCQI/5UzG9ekXdhIQQ4h5JYVUTZWTAvn365126qJqKEEX+eeENfUP2c2fx+2Ge2ukIIcQ9kcKqJoqJ0XfI2LQp3OZ2PkJUpGtW1mwKnQTAQ6siYO9edRMSQoh7IL1CVhOl6ben8/dr8AUSm7Rlg9xTTVQiyb6PcKTr43j8vQFee03fFlAr//8JIaoO+caqgRru199z8Uyrh1TORIhbbX79PfKtrPVHVr/+Wu10hBCiVKSwqmFM8vNxStKfYpGOQUVllF3fie2DR+lfvPMOnDunbkJCCFEKUljVMA2OJmKan8cVOwcuNXJXOx0hipXQ90Vo105/ocXYsWqnI4QQd00Kqxqm4f6dAJxp7QMajcrZCFE8xcQUwsP1n9EffoCNG9VOSQgh7ooUVjWMy/X2VWdbyWlAUcl17AhvvKF//vrrkJenbj5CCHEXpLCqSXQ6XBJ3A9ePWAlR2X30kb5LkMOHYdo0tbMRQog7ksKqBnE4dRyryxkUWFhyvqmn2ukIcWd2dvD55/rnH38M+/erm48QQtyBFFY1SFE3CyktvNCZmqmcjRB36fnn4emnoaAAQkKgsFDtjIQQokRSWNUgLonX21fJaUBRlWg08OWXYGsLO3bA3LlqZySEECWSwqoG+a9jUCmsRBXTsCHMnKl/PnEiHD+ubj5CCFECKaxqCJsLadilnkan1ZLS0lvtdIQovVdegUcegStX4NVX9fe7FEKISkYKqxqi6GjV+SaeFFjXUjkbIe6BRgOLFoGlJfz5J0REqJ2REELcQgqrGuK/9lVyf0BRhTVtCh9+qH8eFgYpKermI4QQN5HCqoYwtK+ShuuiqhszBnx89Le7GTlS7WyEEMKIFFY1gHnOZeolJwFwtpUcsRJVnKkpfPON/nHVKvj5Z7UzEkIIAymsagDngwlodToynF3JqeuodjpC3D8vLxg/Xv/89dfh3Dl18xFCiOuksKoB5P6Aolp6/31o2xYuXIDXXpOrBIUQlYIUVjVAw0RpXyWqIXNz+O47MDODX3+FZcvUzkgIIaSwqu60Bfk4H9wDSGElqiEvL5g8Wf88NBTOnFE3HyFEjSeFVTXX4OgBTPPzuGprzyXXJmqnI0TZe+cd6NBBf5XgK6/IKUEhhKqksKrmjG5jo9GonI0Q5cDUFL79FiwsIDJSf8WgEEKoRAqras7QcF1OA4rqzNMTPvpI/3zMGGxTT6ubjxCixrqnwmr+/Pm4ublhaWmJr68vO3bsuG38ypUradGiBZaWlrRp04b169cbTVcUhUmTJuHs7IyVlRX+/v4cOXLEKCY9PZ3g4GBsbW2xt7cnJCSE7Oxsw/Tc3FyGDBlCmzZtMDU1JTAw8JY8Nm/ejEajuWVITU29l91Q+SmKNFwXNcfo0dC1K2Rn8/isCaDTqZ2REKIGKnVhtWLFCsLCwpg8eTK7du3Cy8uLgIAAzpXQj8z27dsZOHAgISEh7N69m8DAQAIDA9m/f78hZsaMGcydO5fw8HBiY2OxsbEhICCA3NxcQ0xwcDCJiYlERUWxdu1atmzZwvDhww3TCwsLsbKy4s0338Tf3/+225CUlERKSophaNCgQWl3Q5VQ59RxrLIyuGZuwbmmLdVOR4jyZWKiv3+gtTWue3bw0C/fqp2REKIGKnVhNXv2bIYNG8bQoUNp2bIl4eHhWFtbs3jx4mLjP//8c3r16sXbb7+Np6cn//d//8dDDz3EvHnzAP3Rqjlz5jBx4kT69u1L27Zt+e677zh79iyrV68G4ODBg0RGRvL111/j6+tL165d+eKLL1i+fDlnz54FwMbGhgULFjBs2DCcnJxuuw0NGjTAycnJMGi11fOMaFH7qpQWXujMzFXORogK8OCD8NlnAHRZMov6xw6qnJAQoqYxLU1wfn4+8fHxTJgwwTBOq9Xi7+9PTExMsfPExMQQFhZmNC4gIMBQNCUnJ5Oammp0lMnOzg5fX19iYmIYMGAAMTEx2Nvb0759e0OMv78/Wq2W2NhYnnnmmdJsBt7e3uTl5dG6dWumTJlCly5dSozNy8sjLy/P8DorK6tU6yqNz6IOl+nyGho6BpXb2IgaZNgwjkb8RNOYaJ6YNpZl81ZxzdJK7ayEEDVEqQ7VXLhwgcLCQhwdjW+L4ujoWGI7pdTU1NvGFz3eKebm03WmpqY4ODiUqn2Us7Mz4eHhrFq1ilWrVuHq6sojjzzCrl27Spxn2rRp2NnZGQZXV9e7Xp/aGu7fCcCZNu3vEClENaLREBU2lRyH+tQ9eYxuX3+qdkZCiBqkep4DK0Hz5s159dVX8fHxoXPnzixevJjOnTvz2fVTB8WZMGECmZmZhuHUqVMVmPG9s7mQhl3qaXRaLSme7dROR4gKlWvnQORb0wHwXrMU9382qZyREKKmKFVhVa9ePUxMTEhLSzMan5aWVmK7Jicnp9vGFz3eKebmxvHXrl0jPT39ju2p7qRjx44cPXq0xOkWFhbY2toaDVVBw336o1Xnm3iSb1NL5WyEqHgn23cl/tkhADw+612sL11QNyEhRI1QqsLK3NwcHx8foqOjDeN0Oh3R0dH4+fkVO4+fn59RPEBUVJQh3t3dHScnJ6OYrKwsYmNjDTF+fn5kZGQQHx9viNm4cSM6nQ5fX9/SbMItEhIScHZ2vq9lVEZFpwHPtpb2VaLm2vZyGOebNMc6M53HZ06QXtmFEOWuVI3XAcLCwhg8eDDt27enY8eOzJkzh5ycHIYOHQrAoEGDaNiwIdOmTQNg1KhRdO/enVmzZtGnTx+WL1/Ozp07WbhwIQAajYbRo0czdepUPDw8cHd35/3338fFxcXQF5Wnpye9evVi2LBhhIeHU1BQQGhoKAMGDMDFxcWQ24EDB8jPzyc9PZ3Lly+TkJAA6BurA8yZMwd3d3datWpFbm4uX3/9NRs3bmTDhg33uv8qrf/6r5L2VaLmKjS34Pfxs3ghtB/ucVvw/u0HEgJfUjstIUQ1VurCKigoiPPnzzNp0iRSU1Px9vYmMjLS0Pj85MmTRt0XdO7cmWXLljFx4kTeffddPDw8WL16Na1btzbEjBs3jpycHIYPH05GRgZdu3YlMjISS0tLQ8zSpUsJDQ2lZ8+eaLVa+vXrx9y5c41y6927NydOnDC8btdO37ZIuf5fan5+PmPHjuXMmTNYW1vTtm1b/vzzT3r06FHa3VCpWWRnUS9Zf4WhNFwXNd1FNw+2DBvHo/P/j26LZnCmtQ/npV83IUQ50SiKHBsvjaysLOzs7MjMzCzz9lZl1d2Ce+xmAt9/lUsN3YhY8keZLFNUbWMea6bKeu/nM30/Od+yXkXh6cmv8+A/m7jk0phl838pl7aHau1nIcSdlefv941q1FWBNYWhmwVpXyWEnkbDH29NJ6uBC3XOnqDn55OkvZUQolxIYVUNFd14WdpXCfGfPFt71r87m0ITU1psXkeb9T+pnZIQohqSwqqaMcnPw/HwPkAKKyFultKyHduGjgHgkS+nUu/YIZUzEkJUN1JYVTNOh/ZiWlBAjkN9Ml0eUDsdISqd+Ode5rjvI5gW5NPno1GYXclWOyUhRDVS6qsCReVmaF/V6iHQaFTORlQWZX0fyopQbjlrtfzx9nRefC0Qh9P/0nPuFCLf+VT+Xu6BWhcnCFGZyRGraua/hutyGlCIkuTa1mH9e7PRaU3w3Pg/2qxboXZKQohqQgqrakRTWIjzgd2A9F8lxJ2cbeXDtpf17a16fDkVp4N7VM5ICFEdSGFVjdRLTsLiSg551jZccG+udjpCVHo7n3+FI10DMLlWwJP/N1LuJyiEuG9SWFUjRTdeTmnZDsXEROVshKgCNBr+eOtjLro2ofaFNHp/PAZN4TW1sxJCVGFSWFUj0r5KiNIrsK7F/ybPI8/aBtc9O+j6zSy1UxJCVGFSWFUXikLDxF2AFFZClNalBx5kw1vTAWj/82KabV6vckZCiKpKCqtqwu7sSWzSz3PNzIzUFm3VTkeIKudo18fZETQcgMdnvUvd5KrXRYUQQn1SWFUTRacBz3m0ptDcQuVshKiatg8ZzYl2nTHLu8rTH4zAIitD7ZSEEFWMFFbVhKF9lXSzIMQ9U0xMWP/uLDIdG2J/9iRPTh2F9lqB2mkJIaoQKayqiUZ74wA43aaDypkIUbXl2jnw24cLyLey5oGEf+i+4GO1UxJCVCFSWFUDtc6lYJ9yCp1Wy9lWPmqnI0SVd9G9Ob+Pn4mi0eD9v2W0XbNU7ZSEEFWEFFbVQKO9OwA459GKfJtaKmcjRPVw3K8nf78cBkCPLz/CdXeMyhkJIaoCuQlzNVBUWJ1u21HlTISoXnb2H0a9f4/gGb2GJ6eO4se5P5HR0E3ttG4hN0MWovKQI1bVgKsUVkKUD42GqDFTOevpjeXlTPq+/xoW2VlqZyWEqMSksKriap1Pxf7sSXRaLWdaS/sqIcpaobkF/5s8j8v1nHA4ncxTH4zAJD9f7bSEEJWUFFZVXNFpwPMPepJvU1vlbISonq441Gf1/31luO3NY7PfBUVROy0hRCUkhVUVV9TNwikvX5UzEaJ6u/BgC9ZOnItOa4Lnxv/ROWKO2ikJISohKayqOGm4LkTFOdm+K1Fj/g8A3x/DabNuhcoZCSEqGymsqjCbi2nUOfMvikbDWWlfJUSFOBDQj5gXQwF49IsPcNvxl8oZCSEqEymsqrCi04DnHmxJXi1blbMRoub456VQEh97Bq2ukD5TR9Pg8H61UxJCVBJSWFVhjfZcPw3oJacBhahQGg1/jv6QE+06Y557hcCJw7E/86/aWQkhKgEprKqwRntjATjdVu4PKERF05mZs3bSXM496IlNxkWeHf8ynD6tdlpCCJVJYVVF2Vw8h8NpffuqM63bq52OEDVSvk1tfv34ay65NMYu7QwEBMDFi2qnJYRQkRRWVVTD6+2rzjdpQV5tO5WzEaLmulKnHr9MX8zleo5w4AD07g2XL6udlhBCJVJYVVGN9kk3C0JUFllOjfhl2mJwcIAdO+DZZyEvT+20hBAquKfCav78+bi5uWFpaYmvry87duy4bfzKlStp0aIFlpaWtGnThvXr1xtNVxSFSZMm4ezsjJWVFf7+/hw5csQoJj09neDgYGxtbbG3tyckJITs7GzD9NzcXIYMGUKbNm0wNTUlMDCw2Fw2b97MQw89hIWFBU2bNiUiIuJedoHqHtj9DwCnpOG6EJVCeuOm8PvvYGMDf/4JwcFw7ZraaQkhKlipC6sVK1YQFhbG5MmT2bVrF15eXgQEBHDu3Lli47dv387AgQMJCQlh9+7dBAYGEhgYyP79/12ePGPGDObOnUt4eDixsbHY2NgQEBBAbm6uISY4OJjExESioqJYu3YtW7ZsYfjw4YbphYWFWFlZ8eabb+Lv719sLsnJyfTp04cePXqQkJDA6NGjeeWVV/jjjz9KuxtUVetcCnXO/ItOq+W09LguROXRsSP89huYm8OqVTBkCBQWqp2VEKICaRSldDe88vX1pUOHDsybNw8AnU6Hq6srI0eOZPz48bfEBwUFkZOTw9q1aw3jOnXqhLe3N+Hh4SiKgouLC2PHjuWtt94CIDMzE0dHRyIiIhgwYAAHDx6kZcuWxMXF0b69vqF2ZGQkvXv35vTp07i4uBitc8iQIWRkZLB69Wqj8e+88w7r1q0zKuoGDBhARkYGkZGRd7X9WVlZ2NnZkZmZia1t2fYd9VnU4buKa7nhFwJmTiClhRfL5/5UpjkIIe7dmMea6Z/89hs895z+iNWQIfDNN6Atv5YXd/vdURxDzlVovULci/L8/b5Rqf7S8/PziY+PNzoipNVq8ff3JyYmpth5YmJibjmCFBAQYIhPTk4mNTXVKMbOzg5fX19DTExMDPb29oaiCsDf3x+tVktsbOxd53+nXKqKB3ZtB+BkOz+VMxFCFKtvX/jxRzAxgYgIePVV0OnUzkoIUQFKVVhduHCBwsJCHB0djcY7OjqSmppa7Dypqam3jS96vFNMgwYNjKabmpri4OBQ4npLk0tWVhZXr14tdp68vDyysrKMBlUpCg8k6NtXSWElRCX23HPwww/6I1Vffw0jRkDpThAIIaoguSrwDqZNm4adnZ1hcHV1VTWfuieOYpN+nmvmFqS0bKdqLkKIOxgwAL79FjQaCA+HN9+U4kqIaq5UhVW9evUwMTEhLS3NaHxaWhpOTk7FzuPk5HTb+KLHO8Xc3Dj+2rVrpKenl7je0uRia2uLlZVVsfNMmDCBzMxMw3Dq1Km7Xl95cN2tP215ppUPheYWquYihLgLL76ob2MFMG8ejBolxZUQ1VipCitzc3N8fHyIjo42jNPpdERHR+PnV/xpKT8/P6N4gKioKEO8u7s7Tk5ORjFZWVnExsYaYvz8/MjIyCA+Pt4Qs3HjRnQ6Hb6+d39V3J1yKY6FhQW2trZGg5oeSNAXVicfktOAQlQZQ4fCwoX65198oW9zJVcLClEtmZZ2hrCwMAYPHkz79u3p2LEjc+bMIScnh6FDhwIwaNAgGjZsyLRp0wAYNWoU3bt3Z9asWfTp04fly5ezc+dOFl7/ktFoNIwePZqpU6fi4eGBu7s777//Pi4uLoa+qDw9PenVqxfDhg0jPDycgoICQkNDGTBggNEVgQcOHCA/P5/09HQuX75MQkICAN7e3gC89tprzJs3j3HjxvHyyy+zceNGfvrpJ9atW3ev+69CaQqvGW68fEraVwlRtQwbpu+G4eWXYdEiuHoVliwB01J/DQshKrFS/0UHBQVx/vx5Jk2aRGpqKt7e3kRGRhoahZ88eRLtDZcVd+7cmWXLljFx4kTeffddPDw8WL16Na1btzbEjBs3jpycHIYPH05GRgZdu3YlMjISS0tLQ8zSpUsJDQ2lZ8+eaLVa+vXrx9y5c41y6927NydOnDC8btdO3wapqEcJd3d31q1bx5gxY/j8889p1KgRX3/9NQEBAaXdDapwPLwfiyvZ5Na249yDLdVORwhRWoMHg6WlvvPQH37QF1fLlukLLiFEtVDqfqxqOjX7seq4bAFdIuZwpOvjrJ30RZmuWwhx/+66b6bffoP+/SE/H/r0gZ9/1hdc90j6sRLiziplP1ZCXQ/svt5/lXcnlTMRQtyXvn1hzRp9MbVuHTz5pNy4WYhqQgqrKsI09yrOB3YDcPKhzipnI4S4bwEB+nsL1qoF0dHQoweUcGswIUTVIYVVFdFwXxymBQVcrudERkM3tdMRQpSFRx6BjRuhXj2Ij4cuXeD4cbWzEkLcBymsqgi3nX8D8G+HbvrOBoUQ1UOHDrBtG7i5wdGj0LkzXL+iWQhR9UhhVUU03rkVgH/bd1M5EyFEmWvWDLZvh7ZtIS0NHn4YNm1SOyshxD2QwqoKsE09Td1Tx9FpTaT/KiGqK2dn2LIFunfXN2Tv1QuWL1c7KyFEKUlhVQU0vn4aMKWlN3m11O35XQhRjuzsIDISnn1W3xXDwIEwdarcAkeIKkQKqyrATU4DClFzWFrCTz9BWJj+9fvv6zsWzctTNy8hxF2RwqqS0xbk43r9/oBSWAlRQ5iYwKxZEB6uf/799/DYY3DxotqZCSHuQAqrSs75YAIWV3K4YufAuaZyGxshapRXX9X3dWVrC1u3QqdOkJSkdlZCiNuQwqqSc4vTnwY80b4raOXtEqLGeewxiIn5rzuGTp30xZYQolKSX+pKTtpXCSFo2RJiY/V9XGVk6O8v+PHH0qhdiEpICqtKzDr9PA2OHUTRaDjh01XtdIQQamrQQN+31auv6guq996D55+H7Gy1MxNC3MBU7QREydzitgBwrmlLrto7qJyNEOJOPos6fM/zjnms2Z2DzM31Ddp9fGDECFi1Cg4dwv6t2XKrKyEqCTliVYk9GLMRgOOdHlU5EyFEpTJsGPz1l75T0cREBoY+R5Pr3xdCCHVJYVVJmeTn0Th+GwDHO/VQORshRKXj56e/cXPnzljmXKbv5Nd5+KvpaAvy1c5MiBpNCqtKyjXhH8zyrnK5nqN0syCEKJ6zM2zaxK5nBgPgs2oJ/ce+SO20MyonJkTNJYVVJdXkn6LTgD1Ao1E5GyFEpWVuzl+vv8uayfPJrWWL86E9vPh6IE22/6l2ZkLUSFJYVUaKQpN/9He2l/ZVQoi7cayLP0u//JWU5m2xzM6i75QRdF/wMSb5cmpQiIokhVUl1ODoAWpfSKPAwopT3p3UTkcIUUVkOTXip9lLie83FICHfv2WgW8+R91k6a1diIoihVUlVHS06oRPFwrNLVTORghRlejMzNny6nh++2ABV+wcqH88iRdC+/HQz0tAp1M7PSGqPSmsKqGiy6aP+clpQCHEvTnu9yjfL/wfx30fwbSggO4Lp9Nv/FBqnUtROzUhqjUprCqZWudTcTyaiKLRkNzxEbXTEUJUYVfq1OO3D8P5880PKLCw4oGEf3jptadpvvF/cjscIcqJFFaVTNO/NwBwtmU7rtapq3I2QogqT6Nh35MD+GHBr6Q2b4Nldha9p7/F01PewOZimtrZCVHtSGFVyXhs/QOAIw/3UjkTIUR1ktHInRWf/cj2QSMpNDXjwZiNDHqlD3zzjRy9EqIMSWFVidhcPEfDxHgAjnR5XOVshBDVjc7UjNgXQ1k6/xf90aucy/DKK/D445CcrHZ6QlQLUlhVIg9ui0KjKKS08CK7gbPa6QghqqmL7s1YPmc5W4aNA0tL+PNPaNMGPvsMrl1TOz0hqjQprCoRw2nAbgEqZyKEqO4UE1Pinw+BvXvh4YchJwfCwsDHB7ZvVzs9IaosKawqi/PnabQvDoAjXeU0oBCignh4wKZNsHAhODjoC60uXSAkBC5cUDs7IaocKawqi9Wr0ep0pHm0IsvZVe1shBA1iVYLw4ZBUpK+oAJYvBiaN9cXXNKxqBB3TQqryuLnnwE40k2uBhRCqKRePfj6a9i2Dby8ID0dXn0VOnSALVvUzk6IKuGeCqv58+fj5uaGpaUlvr6+7Nix47bxK1eupEWLFlhaWtKmTRvWr19vNF1RFCZNmoSzszNWVlb4+/tz5MgRo5j09HSCg4OxtbXF3t6ekJAQsrOzjWL27t1Lt27dsLS0xNXVlRkzZhhNj4iIQKPRGA2Wlpb3sgvKlqKAgwMFFpYc6SanAYUQKuvcGXbuhDlzwNYWdu2C7t2hXz84dkzt7ISo1EpdWK1YsYKwsDAmT57Mrl278PLyIiAggHPnzhUbv337dgYOHEhISAi7d+8mMDCQwMBA9u/fb4iZMWMGc+fOJTw8nNjYWGxsbAgICCA3N9cQExwcTGJiIlFRUaxdu5YtW7YwfPhww/SsrCwef/xxGjduTHx8PJ9++ilTpkxh4cKFRvnY2tqSkpJiGE6cOFHaXVD2NBr48UfCV/5DRkM3tbMRQggwNYVRo+DIEXjtNf3pwl9+gZYt4e23ISND7QyFqJQ0ilK6nuF8fX3p0KED8+bNA0Cn0+Hq6srIkSMZP378LfFBQUHk5OSwdu1aw7hOnTrh7e1NeHg4iqLg4uLC2LFjeeuttwDIzMzE0dGRiIgIBgwYwMGDB2nZsiVxcXG0b98egMjISHr37s3p06dxcXFhwYIFvPfee6SmpmJubg7A+PHjWb16NYcOHQL0R6xGjx5Nxn18IWRlZWFnZ0dmZia2trb3vJzifBZ1uEyXJ4SoOsY81uye572f7467Xu/+/TB2LGzQ3x2CevXY9Pyr7OszgMLr37nlsl4hykh5/n7fqFRHrPLz84mPj8ff3/+/BWi1+Pv7ExMTU+w8MTExRvEAAQEBhvjk5GRSU1ONYuzs7PD19TXExMTEYG9vbyiqAPz9/dFqtcTGxhpiHn74YUNRVbSepKQkLl26ZBiXnZ1N48aNcXV1pW/fviQmJt52m/Py8sjKyjIahBCixmndGiIjYd06aNECLlygx4KPGPJyAC3/WIWmUPq/EgJKWVhduHCBwsJCHB0djcY7OjqSmppa7Dypqam3jS96vFNMgwYNjKabmpri4OBgFFPcMm5cR/PmzVm8eDG//fYbP/zwAzqdjs6dO3P69OkSt3natGnY2dkZBldXuWJPCFFDaTTQu7e+S4bwcLLrNsD23FkCZr3LoGFP4rHld7mCUNR4NeqqQD8/PwYNGoS3tzfdu3fnl19+oX79+nz11VclzjNhwgQyMzMNw6lTpyowYyGEqITMzODVV1kcEcWWYeO4Wtseh9PJPDl1NC+EPod77Ga5/6CosUpVWNWrVw8TExPS0ozviJ6WloaTk1Ox8zg5Od02vujxTjE3N46/du0a6enpRjHFLePGddzMzMyMdu3acfTo0eI3GLCwsMDW1tZoEEIIAYUWlsQ/H8Li76KJeTGUfCtrHI8mEvj+q7ww4lma/r1BjmCJGqdUhZW5uTk+Pj5ER0cbxul0OqKjo/Hz8yt2Hj8/P6N4gKioKEO8u7s7Tk5ORjFZWVnExsYaYvz8/MjIyCA+Pt4Qs3HjRnQ6Hb6+voaYLVu2UFBQYLSe5s2bU6dOnWJzKywsZN++fTg7y335hBDiXuXb1OKfQSP55rtodj73MvmW1jgePcBTH45k0KtP0Xzj/6QNlqgxSn0qMCwsjEWLFvHtt99y8OBBXn/9dXJychg6dCgAgwYNYsKECYb4UaNGERkZyaxZszh06BBTpkxh586dhIaGAqDRaBg9ejRTp05lzZo17Nu3j0GDBuHi4kJgYCAAnp6e9OrVi2HDhrFjxw62bdtGaGgoAwYMwMXFBYAXXngBc3NzQkJCSExMZMWKFXz++eeEhYUZcvnwww/ZsGEDx48fZ9euXbz44oucOHGCV1555Z53oBBCCL1cOwe2Dn+Hb76P5p8XXifPuhZ1Txyl9/S3GBzSm1a/r8QkP1/tNIUoV6alnSEoKIjz588zadIkUlNT8fb2JjIy0tBQ/OTJk2i1/9VrnTt3ZtmyZUycOJF3330XDw8PVq9eTevWrQ0x48aNIycnh+HDh5ORkUHXrl2JjIw06rxz6dKlhIaG0rNnT7RaLf369WPu3LmG6XZ2dmzYsIERI0bg4+NDvXr1mDRpklFfV5cuXWLYsGGkpqZSp04dfHx82L59Oy1btiztbhBCCFGCXDsHYoaMZtdzL+O1ZikP/RJBnbMnePyziXSJmEPC08Hw0PtQt67aqQpR5krdj1VNJ/1YCSHKQ6Xvx+o+1mt2NYc261bw0C8R1L5wvS2slRUMHgxjxkAz6dNKlL9K2Y+VEEIIUVoFVjbseu5lFn8XzfrxM0lr2gquXoXwcP2Nnp9+GqKj5UpCUS1IYSWEEKJC6EzNSHr0KZbNXwWbN8NTT+kn/O9/4O+v73j0s8/ghk6dhahqpLASQghRsTQa/U2d16yBpCR44w2oXRsOH4awMHBxgaFDYccOOYolqhwprIQQQqinWTOYPx/OnNGfGvTygtxciIgAX19o3x4WLJCjWKLKkMJKCCGE+mrXhldfhd27Yft2eOklsLCAXbv0R7ScnWHAAP39CgsL1c5WiBJJYSWEEKLy0GjAzw+++w5On4bZs6FNG8jLgxUr4Ikn4IEHYPx4OHRI7WyFuIUUVkIIISqnevX03THs2aM/cjVyJDg4wNmz8Mkn4OkJPj7w6adw4oTa2QoBSGElhBCistNooF07mDtXX1StWqW/otDERF9wjRsHbm7QpQt88QWkpqqdsajBpLASQghRdVhYwLPP6q8oTEnRN2x/5BF98bV9O7z5pv6qwkcfhS+/1DeKF6ICSc/rpSQ9rwshqhO1eny/H8XmfPYsrFwJy5fDP/8YT+vQAQID9YOnp74Iuwdq9XBfFVXGfSU9rwshhBB3y8UFRo2CmBhIToYZM6BzZ30RFRcH770HrVrpe3p/+234+2+4dk3trEU1JIWVEEKI6sXNTV88bdumP5K1cCH07g3m5nDkCMycCd26Qf360L8/LF4spwxFmZHCSgghRPXl5ATDhsG6dXDhgv504Qsv6K8uzMjQvw4JgUaNoG1beOcd2LQJ8vPVzlxUUaZqJyCEEEJUiNq14bnn9ENhof4UYWSkftixA/bt0w8zZoC1NXTtqm8Y36OHvgd4Ie6CFFZCCCFqHhMT6NRJP0yZAhcvQlQU/P47/PEHpKXBhg36AaBWLQI9H+KUly+nvX0592BLFBMTVTdBVE5SWAkhhBB16+pvmTNgAOh0cOCA/pTgpk2weTNcuoR73Bbc47YAkGddi5SW3pxt2Y6Ulg+R0qItBda11N0GUSlIYSWEEELcSKuF1q31w8iR+kJrzx7+Cv8J1z2xNNwXh8WVbNx2/o3bzr8B0Gm1XHBvfr3QasfZlu3Icmp0z107iKpLCishhBDidrRaaNeOXc/ZsOu5oWgKC6l//BDOB3bjcmA3zgd2Y5d2hgbHDtLg2EG8/7cMgBz7uqQ1a825Zq0hN0B/+x0XF5U3RpQ3KayEEEKIUlBMTDjn0YpzHq3Y0/dFAGwupBmKLJfEXTQ4egCbjIs02fEXTXb8BT/M18/s7KxvCO/j89+jo6Mc2apGpLASQggh7lNOPUeOPNyLIw/3AsAkL5f6xw/heHg/jof30yrliL7dVkoK/O9/+qFI/frQpo2+u4eix5Yt9VcmiipHCishhBCijBVaWJLq6U2qpzcArR5rBjk5kJAA8fGwc6d+SEqC8+dh40b9UESrhaZN/yu2WreGFi3gwQf190sUlZYUVkIIIURFsLGBLl30Q5ErV/RHsvbtg717/xsuXIDDh/XDzz//F29iAk2a6IusG4fmzfVXNgrVSWElhBA1WE27+Xul215ra31bqxs7IFUUfT9aNxZbhw7BwYNw+bL+tjxHjhifTgSoV09fYD34oH5o0oTlF83IcH6Aq/YO0o6rgkhhJYQQQlQmGo3+VjxOTvDYY/+NVxRITdUXWTcPJ0/qj3JduKC/R+J1A64/5ltZk+nsSqaTKxkuD1x/3ojLjg25XN+JAiubit3GakwKKyGEEKIq0Gj0VxU6O+tvs3OjnBz9acOkJDh+XD8cO0ZWYhK1L6RifvUK9Y8nUf94UrGLzq1tR1Z9Zy43cOZy0WMDFy7Xd+JyAxey6zZAMZGS4W7IXhJCCCGqOhsbaNdOP9zgm6jDmOTnY5t2GruUU9ilnML+7EnsUk9hl3KaWudTsMy5jOXlTCwvZ9Lg+KFiF6/TaslxqK8f6ja4/lz/mF30um59rtjXrfEFWM3eeiGEEKKaKzQ355JrEy65Nil2unlONrXPp1D73Flqn0vRPz+fcsPzVEyuFVD7Qhq1L6Tddl06rZar9nXJvl6EXalTl6v2dbli58BVOweu2tXhqr0DV66PK7SwLI9NVpUUVkIIIUQNlm9Ti4s2Hlx08yg+QKfD5tIFal1IxSb9PDYXz2OTfp5aF8/pX6frH60vXUCr010fd/7u1m1pzVW7Olyxr8tVe33xRdSD8Oqr+gb4VZAUVkIIIYQomVarP/1Xt8FtwzSFhVhlphsKq1oXz2GVcRHrjHT9Y+YlrDLTDYNpQQHmuVcwz72CXdqZ/xa0AXj2WSmshBBCCFFzKSYmXHGozxWH+tzxeJWiYH4lG6uMdKyLiq2MdKwzL9HVTgdubhWQcfnQ3stM8+fPx83NDUtLS3x9fdmxY8dt41euXEmLFi2wtLSkTZs2rF+/3mi6oihMmjQJZ2dnrKys8Pf358iRI0Yx6enpBAcHY2tri729PSEhIWRnZxvF7N27l27dumFpaYmrqyszZswodS5CCCGEKGcaDfk2tcls2JiUlu047teTxCeeJ27AcJg9W9/VRBVV6sJqxYoVhIWFMXnyZHbt2oWXlxcBAQGcO3eu2Pjt27czcOBAQkJC2L17N4GBgQQGBrJ//35DzIwZM5g7dy7h4eHExsZiY2NDQEAAubm5hpjg4GASExOJiopi7dq1bNmyheHDhxumZ2Vl8fjjj9O4cWPi4+P59NNPmTJlCgsXLixVLkIIIYQQ90qjKIpSmhl8fX3p0KED8+bNA0Cn0+Hq6srIkSMZP378LfFBQUHk5OSwdu1aw7hOnTrh7e1NeHg4iqLg4uLC2LFjeeuttwDIzMzE0dGRiIgIBgwYwMGDB2nZsiVxcXG0v947bWRkJL179+b06dO4uLiwYMEC3nvvPVJTUzE3Nwdg/PjxrF69mkOHDt1VLncjKysLOzs7MjMzsbW1Lc2uu6NK1yOwEEJUQmMea3bP86r1PXs/Od+Pqvi7Ul77qjx/v29UqjZW+fn5xMfHM2HCBMM4rVaLv78/MTExxc4TExNDWFiY0biAgABWr14NQHJyMqmpqfj7+xum29nZ4evrS0xMDAMGDCAmJgZ7e3tDUQXg7++PVqslNjaWZ555hpiYGB5++GFDUVW0nk8++YRLly5Rp06dO+ZSnLy8PPLy8gyvMzMzAf0bVNZyc7LvHCSEEDXc/Xz/qvU9Wx6/GXejKv6ulNe+KlpuKY8nlVqpCqsLFy5QWFiIo6Oj0XhHR0fDUaGbpaamFhufmppqmF407nYxDRoYX41gamqKg4ODUYy7u/styyiaVqdOnTvmUpxp06bxwQcf3DLe1dW1xHmEEEKUn3fVTuAeVMWc1VLe++ry5cvY2dmV2/LlqsA7mDBhgtFRLp1OR3p6OnXr1kVTyhtaZmVl4erqyqlTp8r1MGRVIfvjVrJPjMn+uJXsE2OyP24l+8RY0f44efIkGo0GFxeXcl1fqQqrevXqYWJiQlqacc+raWlpOJXQgt/Jyem28UWPaWlpODs7G8V4e3sbYm5uHH/t2jXS09ONllPcem5cx51yKY6FhQUWFhZG4+zt7UuMvxu2trbyYb+B7I9byT4xJvvjVrJPjMn+uJXsE2N2dnYVsj9KdVWgubk5Pj4+REdHG8bpdDqio6Px8/Mrdh4/Pz+jeICoqChDvLu7O05OTkYxWVlZxMbGGmL8/PzIyMggPj7eELNx40Z0Oh2+vr6GmC1btlBQUGC0nubNm1OnTp27ykUIIYQQ4r4opbR8+XLFwsJCiYiIUA4cOKAMHz5csbe3V1JTUxVFUZSXXnpJGT9+vCF+27ZtiqmpqTJz5kzl4MGDyuTJkxUzMzNl3759hpjp06cr9vb2ym+//abs3btX6du3r+Lu7q5cvXrVENOrVy+lXbt2SmxsrPL3338rHh4eysCBAw3TMzIyFEdHR+Wll15S9u/fryxfvlyxtrZWvvrqq1LlUp4yMzMVQMnMzKyQ9VV2sj9uJfvEmOyPW8k+MSb741ayT4xV9P4odWGlKIryxRdfKA888IBibm6udOzYUfnnn38M07p3764MHjzYKP6nn35SmjVrppibmyutWrVS1q1bZzRdp9Mp77//vuLo6KhYWFgoPXv2VJKSkoxiLl68qAwcOFCpVauWYmtrqwwdOlS5fPmyUcyePXuUrl27KhYWFkrDhg2V6dOn35L7nXIpT7m5ucrkyZOV3NzcCltnZSb741ayT4zJ/riV7BNjsj9uJfvEWEXvj1L3YyWEEEIIIYp3T7e0EUIIIYQQt5LCSgghhBCijEhhJYQQQghRRqSwEkIIIYQoI1JYVZD58+fj5uaGpaUlvr6+7NixQ+2UysSWLVt46qmncHFxQaPR3HLfRUVRmDRpEs7OzlhZWeHv78+RI0eMYtLT0wkODsbW1hZ7e3tCQkLIzja+v9XevXvp1q0blpaWuLq6MmPGjPLetHsybdo0OnToQO3atWnQoAGBgYEkJSUZxeTm5jJixAjq1q1LrVq16Nev3y0d1548eZI+ffpgbW1NgwYNePvtt7l27ZpRzObNm3nooYewsLCgadOmRERElPfm3ZMFCxbQtm1bQ2eFfn5+/P7774bpNW1/3Gz69OloNBpGjx5tGFfT9smUKVPQaDRGQ4sWLQzTa9r+ADhz5gwvvvgidevWxcrKijZt2rBz507D9Jr23erm5nbLZ0Sj0TBixAigkn1GKuTawxpu+fLlirm5ubJ48WIlMTFRGTZsmGJvb6+kpaWpndp9W79+vfLee+8pv/zyiwIov/76q9H06dOnK3Z2dsrq1auVPXv2KE8//XSxfZR5eXkp//zzj7J161aladOmRn2UZWZmKo6OjkpwcLCyf/9+5ccff1SsrKyM+iirLAICApQlS5Yo+/fvVxISEpTevXsrDzzwgJKdnW2Iee211xRXV1clOjpa2blzp9KpUyelc+fOhunXrl1TWrdurfj7+yu7d+9W1q9fr9SrV0+ZMGGCIeb48eOKtbW1EhYWphw4cED54osvFBMTEyUyMrJCt/durFmzRlm3bp1y+PBhJSkpSXn33XcVMzMzZf/+/Yqi1Lz9caMdO3Yobm5uStu2bZVRo0YZxte0fTJ58mSlVatWSkpKimE4f/68YXpN2x/p6elK48aNlSFDhiixsbHK8ePHlT/++EM5evSoIaamfbeeO3fO6PMRFRWlAMqmTZsURalcnxEprCpAx44dlREjRhheFxYWKi4uLsq0adNUzKrs3VxY6XQ6xcnJSfn0008N4zIyMhQLCwvlxx9/VBRFUQ4cOKAASlxcnCHm999/VzQajXLmzBlFURTlyy+/VOrUqaPk5eUZYt555x2lefPm5bxF9+/cuXMKoPz111+Koui338zMTFm5cqUh5uDBgwqgxMTEKIqiL1a1Wq2h011FUZQFCxYotra2hn0wbtw4pVWrVkbrCgoKUgICAsp7k8pEnTp1lK+//rpG74/Lly8rHh4eSlRUlNK9e3dDYVUT98nkyZMVLy+vYqfVxP3xzjvvKF27di1xuny3KsqoUaOUBx98UNHpdJXuMyKnAstZfn4+8fHx+Pv7G8ZptVr8/f2JiYlRMbPyl5ycTGpqqtG229nZ4evra9j2mJgY7O3tad++vSHG398frVZLbGysIebhhx/G3NzcEBMQEEBSUhKXLl2qoK25N5mZmQA4ODgAEB8fT0FBgdE+adGiBQ888IDRPmnTpg2Ojo6GmICAALKyskhMTDTE3LiMopjK/pkqLCxk+fLl5OTk4OfnV6P3x4gRI+jTp88tedfUfXLkyBFcXFxo0qQJwcHBnDx5EqiZ+2PNmjW0b9+e559/ngYNGtCuXTsWLVpkmF7Tv1vz8/P54YcfePnll9FoNJXuMyKFVTm7cOEChYWFRm8mgKOjI6mpqSplVTGKtu92256amkqDBg2MppuamuLg4GAUU9wyblxHZaTT6Rg9ejRdunShdevWgD5fc3PzW27kffM+udP2lhSTlZXF1atXy2Nz7su+ffuoVasWFhYWvPbaa/z666+0bNmyxu6P5cuXs2vXLqZNm3bLtJq4T3x9fYmIiCAyMpIFCxaQnJxMt27duHz5co3cH8ePH2fBggV4eHjwxx9/8Prrr/Pmm2/y7bffAvLdunr1ajIyMhgyZAhQ+f5mTEuzMUKIuzdixAj279/P33//rXYqqmvevDkJCQlkZmby888/M3jwYP766y+101LFqVOnGDVqFFFRUVhaWqqdTqXwxBNPGJ63bdsWX19fGjduzE8//YSVlZWKmalDp9PRvn17Pv74YwDatWvH/v37CQ8PZ/DgwSpnp75vvvmGJ554AhcXF7VTKZYcsSpn9erVw8TE5JarE9LS0nByclIpq4pRtH2323YnJyfOnTtnNP3atWukp6cbxRS3jBvXUdmEhoaydu1aNm3aRKNGjQzjnZycyM/PJyMjwyj+5n1yp+0tKcbW1rZS/hCZm5vTtGlTfHx8mDZtGl5eXnz++ec1cn/Ex8dz7tw5HnroIUxNTTE1NeWvv/5i7ty5mJqa4ujoWOP2yc3s7e1p1qwZR48erZGfEWdnZ1q2bGk0ztPT03B6tCZ/t544cYI///yTV155xTCusn1GpLAqZ+bm5vj4+BAdHW0Yp9PpiI6Oxs/PT8XMyp+7uztOTk5G256VlUVsbKxh2/38/MjIyCA+Pt4Qs3HjRnQ6Hb6+voaYLVu2UFBQYIiJioqiefPm1KlTp4K25u4oikJoaCi//vorGzduxN3d3Wi6j48PZmZmRvskKSmJkydPGu2Tffv2GX0pRkVFYWtra/iy9fPzM1pGUUxV+UzpdDry8vJq5P7o2bMn+/btIyEhwTC0b9+e4OBgw/Oatk9ulp2dzbFjx3B2dq6Rn5EuXbrc0k3L4cOHady4MVAzv1uLLFmyhAYNGtCnTx/DuEr3GbnHBvmiFJYvX65YWFgoERERyoEDB5Thw4cr9vb2RlcnVFWXL19Wdu/erezevVsBlNmzZyu7d+9WTpw4oSiK/pJge3t75bffflP27t2r9O3bt9hLgtu1a6fExsYqf//9t+Lh4WF0SXBGRobi6OiovPTSS8r+/fuV5cuXK9bW1pXykuDXX39dsbOzUzZv3mx0afCVK1cMMa+99prywAMPKBs3blR27typ+Pn5KX5+fobpRZcFP/7440pCQoISGRmp1K9fv9jLgt9++23l4MGDyvz58yvtpePjx49X/vrrLyU5OVnZu3evMn78eEWj0SgbNmxQFKXm7Y/i3HhVoKLUvH0yduxYZfPmzUpycrKybds2xd/fX6lXr55y7tw5RVFq3v7YsWOHYmpqqnz00UfKkSNHlKVLlyrW1tbKDz/8YIipad+tiqK/ov6BBx5Q3nnnnVumVabPiBRWFeSLL75QHnjgAcXc3Fzp2LGj8s8//6idUpnYtGmTAtwyDB48WFEU/WXB77//vuLo6KhYWFgoPXv2VJKSkoyWcfHiRWXgwIFKrVq1FFtbW2Xo0KHK5cuXjWL27NmjdO3aVbGwsFAaNmyoTJ8+vaI2sVSK2xeAsmTJEkPM1atXlTfeeEOpU6eOYm1trTzzzDNKSkqK0XL+/fdf5YknnlCsrKyUevXqKWPHjlUKCgqMYjZt2qR4e3sr5ubmSpMmTYzWUZm8/PLLSuPGjRVzc3Olfv36Ss+ePQ1FlaLUvP1RnJsLq5q2T4KCghRnZ2fF3NxcadiwoRIUFGTUZ1NN2x+Koij/+9//lNatWysWFhZKixYtlIULFxpNr2nfrYqiKH/88YcC3LKdilK5PiMaRVGU0h3jEkIIIYQQxZE2VkIIIYQQZUQKKyGEEEKIMiKFlRBCCCFEGZHCSgghhBCijEhhJYQQQghRRqSwEkIIIYQoI1JYCSGEEEKUESmshBBCCCHKiBRWQgghhBBlRAorIYQQQogyIoWVEEIIIUQZkcJKCCGEEKKM/D8OK4neXb76wQAAAABJRU5ErkJggg==", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Kolmogorov-Smirnov test: D = 0.05600409953637353, p-value = 0.2925487122321031\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "The correlation between 'costs_from_violations_smooth' and 'at_dest_f' is 0.501214315800651\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "from main import simulation_10_agents, simulation_4_agents, simulation_10_to_4_agents\n", + "import copy\n", + "\n", + "\n", + "avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f = None, None, None, None, None, None, None\n", + "for seed in seeds:\n", + " avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results= simulation_4_agents(seed, False) \n", + " if seed == seeds[0]:\n", + " avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f= avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results\n", + " first_gen_results_f = [list(item) for item in first_gen_results_f]\n", + " middle_gen_results_f = [list(item) for item in middle_gen_results_f]\n", + " last_gen_results_f = [list(item) for item in last_gen_results_f]\n", + " else:\n", + " avg_training_rewards_f = [sum(x) / 2 for x in zip(avg_training_rewards_f, avg_training_rewards)]\n", + " \n", + "\n", + "\n", + " first_gen_results_f[0][0] = [sum(x) / 2 for x in zip(first_gen_results_f[0][0], first_gen_results[0][0])]\n", + " middle_gen_results_f[0][0] = [sum(x) / 2 for x in zip(middle_gen_results_f[0][0], middle_gen_results[0][0])]\n", + " last_gen_results_f[0][0] = [sum(x) / 2 for x in zip(last_gen_results_f[0][0], last_gen_results[0][0])]\n", + "\n", + " avg_training_norm_violations_f = avg_training_norm_violations_f.add(avg_training_norm_violations).div(2)\n", + " \n", + " first_gen_results_f[0][1] = first_gen_results_f[0][1].add(first_gen_results[0][1]).div(2)\n", + " middle_gen_results_f[0][1] = middle_gen_results_f[0][1].add(middle_gen_results[0][1]).div(2)\n", + " last_gen_results_f[0][1] = last_gen_results_f[0][1].add(last_gen_results[0][1]).div(2)\n", + "\n", + " sim_violated_f = sim_violated_f.add(sim_violated).div(2)\n", + " at_dest_f = [sum(x) / 2 for x in zip(at_dest_f, at_dest)]\n", + "\n", + "#plot results\n", + "plt.plot(avg_training_rewards_f, label='Average training reward')\n", + "plt.plot(first_gen_results_f[0][0], label='First generation reward')\n", + "plt.plot(middle_gen_results_f[0][0], label='Middle generation reward')\n", + "plt.plot(last_gen_results_f[0][0], label='Last generation reward')\n", + "plt.legend()\n", + "plt.title('Average training reward')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Average reward')\n", + "plt.show()\n", + "\n", + "\n", + "#training violations\n", + "costs_from_violations = copy.deepcopy(avg_training_norm_violations_f['total_violations_cost'])\n", + "\n", + "avg_training_norm_violations_f.drop(columns=['seed'], inplace=True)\n", + "avg_training_norm_violations_f.drop(columns=['total_violations_cost'], inplace=True)\n", + "\n", + "# Calculate the moving average\n", + " # Change this to the desired window size\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=200).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('AVERAGE Costs from violations training')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Cost')\n", + "plt.show()\n", + "\n", + "costs_from_violations = copy.deepcopy(first_gen_results_f[0][1]['total_violations_cost'])\n", + "first_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", + "first_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", + "\n", + "#graph costs from violations\n", + "####costs_from_violations = np.mean(costs_from_violations, axis=0)\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=200).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('First Gen Costs from violations training')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Cost')\n", + "plt.show()\n", + "\n", + "costs_from_violations = copy.deepcopy(middle_gen_results_f[0][1]['total_violations_cost'])\n", + "middle_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", + "middle_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", + "#costs_from_violations = np.mean(costs_from_violations, axis=0)\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=200).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('Middle Gen Costs from violations training')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Cost')\n", + "plt.show()\n", + "\n", + "costs_from_violations = copy.deepcopy(last_gen_results_f[0][1]['total_violations_cost'])\n", + "last_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", + "last_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", + "#cost_from_violations = np.mean(costs_from_violations, axis=0)\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=100).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('Final Gen Costs from violations training')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Cost')\n", + "plt.show()\n", + "\n", + "\n", + "#training violations\n", + "costs_from_violations_f = copy.deepcopy(sim_violated_f['total_violations_cost'])\n", + "sim_violated_f.drop(columns=['seed'], inplace=True)\n", + "sim_violated_f.drop(columns=['total_violations_cost'], inplace=True)\n", + "costs_from_violations = -1 * costs_from_violations_f\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations).rolling(window=window_size2).mean()\n", + "#costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "\n", + "sns.histplot(costs_from_violations, kde=True)\n", + "plt.title('Distribution of Costs from Violations Simulation')\n", + "plt.xlabel('Cost')\n", + "plt.ylabel('Density')\n", + "plt.show()\n", + "\n", + "#GAMMA\n", + "data = costs_from_violations\n", + "\n", + "# Fit the gamma distribution to your data\n", + "shape, loc, scale = gamma.fit(data)\n", + "\n", + "# Generate a range of values from the min to the max of your data\n", + "x = np.linspace(min(data), max(data), 10000)\n", + "\n", + "# Generate the gamma PDF with the parameters obtained from the fit\n", + "pdf = gamma.pdf(x, shape, loc, scale)\n", + "\n", + "# Plot the histogram of your data and the gamma PDF\n", + "plt.hist(data, bins=30, density=True, alpha=0.5, label='Histogram of Costs from Violations Simulation')\n", + "plt.plot(x, pdf, 'r-', label='Gamma PDF')\n", + "plt.legend()\n", + "plt.show()\n", + "\n", + "# Perform the Kolmogorov-Smirnov test\n", + "d, p_value = kstest(data, 'gamma', args=(shape, loc, scale))\n", + "print(f\"Kolmogorov-Smirnov test: D = {d}, p-value = {p_value}\")\n", + "\n", + "#LOG-NORMAL\n", + "# Fit the log-normal distribution to your data\n", + "shape, loc, scale = lognorm.fit(data)\n", + "\n", + "# Generate the log-normal PDF with the parameters obtained from the fit\n", + "pdf = lognorm.pdf(x, shape, loc, scale)\n", + "\n", + "# Plot the histogram of your data and the log-normal PDF\n", + "plt.hist(data, bins=30, density=True, alpha=0.5, label='Histogram of Costs from Violations Simulation')\n", + "plt.plot(x, pdf, 'r-', label='Log-normal PDF')\n", + "plt.legend()\n", + "plt.show()\n", + "\n", + "# Perform the Kolmogorov-Smirnov test\n", + "d, p_value = kstest(data, 'lognorm', args=(shape, loc, scale))\n", + "print(f\"Kolmogorov-Smirnov test: D = {d}, p-value = {p_value}\")\n", + "\n", + "\n", + "#simulation at destination, avg timeto destination\n", + "#simulation at destination, avg timeto destination\n", + "at_dest_f = np.array(at_dest_f)\n", + "sns.histplot(at_dest_f, kde=True)\n", + "plt.title('Distribution of Average Steps to Destination')\n", + "plt.xlabel('Steps')\n", + "plt.ylabel('Density')\n", + "plt.show()\n", + "\n", + "\n", + "\n", + "df = pd.DataFrame({\n", + " 'costs_from_violations_smooth': costs_from_violations,\n", + " 'at_dest_f': at_dest_f\n", + "})\n", + "\n", + "# Calculate the correlation\n", + "correlation = df['costs_from_violations_smooth'].corr(df['at_dest_f'])\n", + "\n", + "print(f\"The correlation between 'costs_from_violations_smooth' and 'at_dest_f' is {correlation}\")\n", + "violation_counts = sim_violated_f.sum()\n", + "\n", + "\n", + "plt.pie(violation_counts, autopct='%1.1f%%')\n", + "\n", + "plt.title('Violation Count')\n", + "\n", + "plt.legend(violation_counts.index, title=\"Violations\", loc=\"best\")\n", + "\n", + "plt.show()\n", + "\n" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 10 Agents to 4 Transfer" + ] + }, + { + "cell_type": "code", + "execution_count": 4, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Generation 1/300\n", + "Solving for Nash Equilibrium in Generation 1/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 2/300\n", + "Solving for Nash Equilibrium in Generation 2/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 3/300\n", + "Solving for Nash Equilibrium in Generation 3/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 4/300\n", + "Solving for Nash Equilibrium in Generation 4/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 5/300\n", + "Solving for Nash Equilibrium in Generation 5/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 6/300\n", + "Solving for Nash Equilibrium in Generation 6/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 7/300\n", + "Solving for Nash Equilibrium in Generation 7/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 8/300\n", + "Solving for Nash Equilibrium in Generation 8/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 9/300\n", + "Solving for Nash Equilibrium in Generation 9/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 10/300\n", + "Solving for Nash Equilibrium in Generation 10/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 11/300\n", + "Solving for Nash Equilibrium in Generation 11/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 12/300\n", + "Solving for Nash Equilibrium in Generation 12/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 13/300\n", + "Solving for Nash Equilibrium in Generation 13/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 14/300\n", + "Solving for Nash Equilibrium in Generation 14/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 15/300\n", + "Solving for Nash Equilibrium in Generation 15/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 16/300\n", + "Solving for Nash Equilibrium in Generation 16/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 17/300\n", + "Solving for Nash Equilibrium in Generation 17/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 18/300\n", + "Solving for Nash Equilibrium in Generation 18/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 19/300\n", + "Solving for Nash Equilibrium in Generation 19/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 20/300\n", + "Solving for Nash Equilibrium in Generation 20/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 21/300\n", + "Solving for Nash Equilibrium in Generation 21/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 22/300\n", + "Solving for Nash Equilibrium in Generation 22/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 23/300\n", + "Solving for Nash Equilibrium in Generation 23/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 24/300\n", + "Solving for Nash Equilibrium in Generation 24/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 25/300\n", + "Solving for Nash Equilibrium in Generation 25/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 26/300\n", + "Solving for Nash Equilibrium in Generation 26/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 27/300\n", + "Solving for Nash Equilibrium in Generation 27/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 28/300\n", + "Solving for Nash Equilibrium in Generation 28/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 29/300\n", + "Solving for Nash Equilibrium in Generation 29/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 30/300\n", + "Solving for Nash Equilibrium in Generation 30/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 31/300\n", + "Solving for Nash Equilibrium in Generation 31/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 32/300\n", + "Solving for Nash Equilibrium in Generation 32/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 33/300\n", + "Solving for Nash Equilibrium in Generation 33/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 34/300\n", + "Solving for Nash Equilibrium in Generation 34/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 35/300\n", + "Solving for Nash Equilibrium in Generation 35/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 36/300\n", + "Solving for Nash Equilibrium in Generation 36/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 37/300\n", + "Solving for Nash Equilibrium in Generation 37/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 38/300\n", + "Solving for Nash Equilibrium in Generation 38/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 39/300\n", + "Solving for Nash Equilibrium in Generation 39/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 40/300\n", + "Solving for Nash Equilibrium in Generation 40/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 41/300\n", + "Solving for Nash Equilibrium in Generation 41/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 42/300\n", + "Solving for Nash Equilibrium in Generation 42/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 43/300\n", + "Solving for Nash Equilibrium in Generation 43/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 44/300\n", + "Solving for Nash Equilibrium in Generation 44/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 45/300\n", + "Solving for Nash Equilibrium in Generation 45/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 46/300\n", + "Solving for Nash Equilibrium in Generation 46/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 47/300\n", + "Solving for Nash Equilibrium in Generation 47/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 48/300\n", + "Solving for Nash Equilibrium in Generation 48/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 49/300\n", + "Solving for Nash Equilibrium in Generation 49/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 50/300\n", + "Solving for Nash Equilibrium in Generation 50/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 51/300\n", + "Solving for Nash Equilibrium in Generation 51/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 52/300\n", + "Solving for Nash Equilibrium in Generation 52/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 53/300\n", + "Solving for Nash Equilibrium in Generation 53/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 54/300\n", + "Solving for Nash Equilibrium in Generation 54/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 55/300\n", + "Solving for Nash Equilibrium in Generation 55/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 56/300\n", + "Solving for Nash Equilibrium in Generation 56/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 57/300\n", + "Solving for Nash Equilibrium in Generation 57/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 58/300\n", + "Solving for Nash Equilibrium in Generation 58/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 59/300\n", + "Solving for Nash Equilibrium in Generation 59/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 60/300\n", + "Solving for Nash Equilibrium in Generation 60/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 61/300\n", + "Solving for Nash Equilibrium in Generation 61/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 62/300\n", + "Solving for Nash Equilibrium in Generation 62/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 63/300\n", + "Solving for Nash Equilibrium in Generation 63/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 64/300\n", + "Solving for Nash Equilibrium in Generation 64/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 65/300\n", + "Solving for Nash Equilibrium in Generation 65/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 66/300\n", + "Solving for Nash Equilibrium in Generation 66/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 67/300\n", + "Solving for Nash Equilibrium in Generation 67/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 68/300\n", + "Solving for Nash Equilibrium in Generation 68/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 69/300\n", + "Solving for Nash Equilibrium in Generation 69/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 70/300\n", + "Solving for Nash Equilibrium in Generation 70/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 71/300\n", + "Solving for Nash Equilibrium in Generation 71/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 72/300\n", + "Solving for Nash Equilibrium in Generation 72/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 73/300\n", + "Solving for Nash Equilibrium in Generation 73/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 74/300\n", + "Solving for Nash Equilibrium in Generation 74/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 75/300\n", + "Solving for Nash Equilibrium in Generation 75/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 76/300\n", + "Solving for Nash Equilibrium in Generation 76/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 77/300\n", + "Solving for Nash Equilibrium in Generation 77/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 78/300\n", + "Solving for Nash Equilibrium in Generation 78/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 79/300\n", + "Solving for Nash Equilibrium in Generation 79/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 80/300\n", + "Solving for Nash Equilibrium in Generation 80/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 81/300\n", + "Solving for Nash Equilibrium in Generation 81/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 82/300\n", + "Solving for Nash Equilibrium in Generation 82/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 83/300\n", + "Solving for Nash Equilibrium in Generation 83/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 84/300\n", + "Solving for Nash Equilibrium in Generation 84/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 85/300\n", + "Solving for Nash Equilibrium in Generation 85/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 86/300\n", + "Solving for Nash Equilibrium in Generation 86/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 87/300\n", + "Solving for Nash Equilibrium in Generation 87/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 88/300\n", + "Solving for Nash Equilibrium in Generation 88/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 89/300\n", + "Solving for Nash Equilibrium in Generation 89/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 90/300\n", + "Solving for Nash Equilibrium in Generation 90/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 91/300\n", + "Solving for Nash Equilibrium in Generation 91/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 92/300\n", + "Solving for Nash Equilibrium in Generation 92/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 93/300\n", + "Solving for Nash Equilibrium in Generation 93/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 94/300\n", + "Solving for Nash Equilibrium in Generation 94/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 95/300\n", + "Solving for Nash Equilibrium in Generation 95/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 96/300\n", + "Solving for Nash Equilibrium in Generation 96/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 97/300\n", + "Solving for Nash Equilibrium in Generation 97/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 98/300\n", + "Solving for Nash Equilibrium in Generation 98/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 99/300\n", + "Solving for Nash Equilibrium in Generation 99/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 100/300\n", + "Solving for Nash Equilibrium in Generation 100/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 101/300\n", + "Solving for Nash Equilibrium in Generation 101/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 102/300\n", + "Solving for Nash Equilibrium in Generation 102/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 103/300\n", + "Solving for Nash Equilibrium in Generation 103/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 104/300\n", + "Solving for Nash Equilibrium in Generation 104/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 105/300\n", + "Solving for Nash Equilibrium in Generation 105/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 106/300\n", + "Solving for Nash Equilibrium in Generation 106/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 107/300\n", + "Solving for Nash Equilibrium in Generation 107/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 108/300\n", + "Solving for Nash Equilibrium in Generation 108/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 109/300\n", + "Solving for Nash Equilibrium in Generation 109/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 110/300\n", + "Solving for Nash Equilibrium in Generation 110/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 111/300\n", + "Solving for Nash Equilibrium in Generation 111/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 112/300\n", + "Solving for Nash Equilibrium in Generation 112/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 113/300\n", + "Solving for Nash Equilibrium in Generation 113/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 114/300\n", + "Solving for Nash Equilibrium in Generation 114/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 115/300\n", + "Solving for Nash Equilibrium in Generation 115/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 116/300\n", + "Solving for Nash Equilibrium in Generation 116/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 117/300\n", + "Solving for Nash Equilibrium in Generation 117/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 118/300\n", + "Solving for Nash Equilibrium in Generation 118/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 119/300\n", + "Solving for Nash Equilibrium in Generation 119/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 120/300\n", + "Solving for Nash Equilibrium in Generation 120/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 121/300\n", + "Solving for Nash Equilibrium in Generation 121/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 122/300\n", + "Solving for Nash Equilibrium in Generation 122/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 123/300\n", + "Solving for Nash Equilibrium in Generation 123/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 124/300\n", + "Solving for Nash Equilibrium in Generation 124/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 125/300\n", + "Solving for Nash Equilibrium in Generation 125/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 126/300\n", + "Solving for Nash Equilibrium in Generation 126/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 127/300\n", + "Solving for Nash Equilibrium in Generation 127/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 128/300\n", + "Solving for Nash Equilibrium in Generation 128/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 129/300\n", + "Solving for Nash Equilibrium in Generation 129/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 130/300\n", + "Solving for Nash Equilibrium in Generation 130/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 131/300\n", + "Solving for Nash Equilibrium in Generation 131/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 132/300\n", + "Solving for Nash Equilibrium in Generation 132/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 133/300\n", + "Solving for Nash Equilibrium in Generation 133/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 134/300\n", + "Solving for Nash Equilibrium in Generation 134/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 135/300\n", + "Solving for Nash Equilibrium in Generation 135/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 136/300\n", + "Solving for Nash Equilibrium in Generation 136/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 137/300\n", + "Solving for Nash Equilibrium in Generation 137/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 138/300\n", + "Solving for Nash Equilibrium in Generation 138/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 139/300\n", + "Solving for Nash Equilibrium in Generation 139/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 140/300\n", + "Solving for Nash Equilibrium in Generation 140/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 141/300\n", + "Solving for Nash Equilibrium in Generation 141/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 142/300\n", + "Solving for Nash Equilibrium in Generation 142/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 143/300\n", + "Solving for Nash Equilibrium in Generation 143/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 144/300\n", + "Solving for Nash Equilibrium in Generation 144/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 145/300\n", + "Solving for Nash Equilibrium in Generation 145/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 146/300\n", + "Solving for Nash Equilibrium in Generation 146/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 147/300\n", + "Solving for Nash Equilibrium in Generation 147/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 148/300\n", + "Solving for Nash Equilibrium in Generation 148/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 149/300\n", + "Solving for Nash Equilibrium in Generation 149/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 150/300\n", + "Solving for Nash Equilibrium in Generation 150/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 151/300\n", + "Solving for Nash Equilibrium in Generation 151/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 152/300\n", + "Solving for Nash Equilibrium in Generation 152/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 153/300\n", + "Solving for Nash Equilibrium in Generation 153/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 154/300\n", + "Solving for Nash Equilibrium in Generation 154/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 155/300\n", + "Solving for Nash Equilibrium in Generation 155/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 156/300\n", + "Solving for Nash Equilibrium in Generation 156/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 157/300\n", + "Solving for Nash Equilibrium in Generation 157/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 158/300\n", + "Solving for Nash Equilibrium in Generation 158/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 159/300\n", + "Solving for Nash Equilibrium in Generation 159/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 160/300\n", + "Solving for Nash Equilibrium in Generation 160/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 161/300\n", + "Solving for Nash Equilibrium in Generation 161/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 162/300\n", + "Solving for Nash Equilibrium in Generation 162/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 163/300\n", + "Solving for Nash Equilibrium in Generation 163/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 164/300\n", + "Solving for Nash Equilibrium in Generation 164/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 165/300\n", + "Solving for Nash Equilibrium in Generation 165/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 166/300\n", + "Solving for Nash Equilibrium in Generation 166/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 167/300\n", + "Solving for Nash Equilibrium in Generation 167/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 168/300\n", + "Solving for Nash Equilibrium in Generation 168/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 169/300\n", + "Solving for Nash Equilibrium in Generation 169/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 170/300\n", + "Solving for Nash Equilibrium in Generation 170/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 171/300\n", + "Solving for Nash Equilibrium in Generation 171/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 172/300\n", + "Solving for Nash Equilibrium in Generation 172/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 173/300\n", + "Solving for Nash Equilibrium in Generation 173/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 174/300\n", + "Solving for Nash Equilibrium in Generation 174/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 175/300\n", + "Solving for Nash Equilibrium in Generation 175/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 176/300\n", + "Solving for Nash Equilibrium in Generation 176/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 177/300\n", + "Solving for Nash Equilibrium in Generation 177/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 178/300\n", + "Solving for Nash Equilibrium in Generation 178/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 179/300\n", + "Solving for Nash Equilibrium in Generation 179/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 180/300\n", + "Solving for Nash Equilibrium in Generation 180/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 181/300\n", + "Solving for Nash Equilibrium in Generation 181/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 182/300\n", + "Solving for Nash Equilibrium in Generation 182/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 183/300\n", + "Solving for Nash Equilibrium in Generation 183/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 184/300\n", + "Solving for Nash Equilibrium in Generation 184/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 185/300\n", + "Solving for Nash Equilibrium in Generation 185/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 186/300\n", + "Solving for Nash Equilibrium in Generation 186/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 187/300\n", + "Solving for Nash Equilibrium in Generation 187/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 188/300\n", + "Solving for Nash Equilibrium in Generation 188/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 189/300\n", + "Solving for Nash Equilibrium in Generation 189/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 190/300\n", + "Solving for Nash Equilibrium in Generation 190/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 191/300\n", + "Solving for Nash Equilibrium in Generation 191/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 192/300\n", + "Solving for Nash Equilibrium in Generation 192/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 193/300\n", + "Solving for Nash Equilibrium in Generation 193/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 194/300\n", + "Solving for Nash Equilibrium in Generation 194/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 195/300\n", + "Solving for Nash Equilibrium in Generation 195/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 196/300\n", + "Solving for Nash Equilibrium in Generation 196/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 197/300\n", + "Solving for Nash Equilibrium in Generation 197/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 198/300\n", + "Solving for Nash Equilibrium in Generation 198/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 199/300\n", + "Solving for Nash Equilibrium in Generation 199/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 200/300\n", + "Solving for Nash Equilibrium in Generation 200/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 201/300\n", + "Solving for Nash Equilibrium in Generation 201/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 202/300\n", + "Solving for Nash Equilibrium in Generation 202/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 203/300\n", + "Solving for Nash Equilibrium in Generation 203/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 204/300\n", + "Solving for Nash Equilibrium in Generation 204/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 205/300\n", + "Solving for Nash Equilibrium in Generation 205/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 206/300\n", + "Solving for Nash Equilibrium in Generation 206/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 207/300\n", + "Solving for Nash Equilibrium in Generation 207/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 208/300\n", + "Solving for Nash Equilibrium in Generation 208/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 209/300\n", + "Solving for Nash Equilibrium in Generation 209/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 210/300\n", + "Solving for Nash Equilibrium in Generation 210/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 211/300\n", + "Solving for Nash Equilibrium in Generation 211/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 212/300\n", + "Solving for Nash Equilibrium in Generation 212/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 213/300\n", + "Solving for Nash Equilibrium in Generation 213/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 214/300\n", + "Solving for Nash Equilibrium in Generation 214/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 215/300\n", + "Solving for Nash Equilibrium in Generation 215/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 216/300\n", + "Solving for Nash Equilibrium in Generation 216/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 217/300\n", + "Solving for Nash Equilibrium in Generation 217/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 218/300\n", + "Solving for Nash Equilibrium in Generation 218/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 219/300\n", + "Solving for Nash Equilibrium in Generation 219/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 220/300\n", + "Solving for Nash Equilibrium in Generation 220/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 221/300\n", + "Solving for Nash Equilibrium in Generation 221/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 222/300\n", + "Solving for Nash Equilibrium in Generation 222/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 223/300\n", + "Solving for Nash Equilibrium in Generation 223/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 224/300\n", + "Solving for Nash Equilibrium in Generation 224/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 225/300\n", + "Solving for Nash Equilibrium in Generation 225/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 226/300\n", + "Solving for Nash Equilibrium in Generation 226/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 227/300\n", + "Solving for Nash Equilibrium in Generation 227/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 228/300\n", + "Solving for Nash Equilibrium in Generation 228/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 229/300\n", + "Solving for Nash Equilibrium in Generation 229/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 230/300\n", + "Solving for Nash Equilibrium in Generation 230/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 231/300\n", + "Solving for Nash Equilibrium in Generation 231/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 232/300\n", + "Solving for Nash Equilibrium in Generation 232/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 233/300\n", + "Solving for Nash Equilibrium in Generation 233/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 234/300\n", + "Solving for Nash Equilibrium in Generation 234/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 235/300\n", + "Solving for Nash Equilibrium in Generation 235/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 236/300\n", + "Solving for Nash Equilibrium in Generation 236/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 237/300\n", + "Solving for Nash Equilibrium in Generation 237/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 238/300\n", + "Solving for Nash Equilibrium in Generation 238/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 239/300\n", + "Solving for Nash Equilibrium in Generation 239/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 240/300\n", + "Solving for Nash Equilibrium in Generation 240/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 241/300\n", + "Solving for Nash Equilibrium in Generation 241/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 242/300\n", + "Solving for Nash Equilibrium in Generation 242/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 243/300\n", + "Solving for Nash Equilibrium in Generation 243/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 244/300\n", + "Solving for Nash Equilibrium in Generation 244/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 245/300\n", + "Solving for Nash Equilibrium in Generation 245/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 246/300\n", + "Solving for Nash Equilibrium in Generation 246/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 247/300\n", + "Solving for Nash Equilibrium in Generation 247/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 248/300\n", + "Solving for Nash Equilibrium in Generation 248/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 249/300\n", + "Solving for Nash Equilibrium in Generation 249/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 250/300\n", + "Solving for Nash Equilibrium in Generation 250/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 251/300\n", + "Solving for Nash Equilibrium in Generation 251/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 252/300\n", + "Solving for Nash Equilibrium in Generation 252/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 253/300\n", + "Solving for Nash Equilibrium in Generation 253/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 254/300\n", + "Solving for Nash Equilibrium in Generation 254/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 255/300\n", + "Solving for Nash Equilibrium in Generation 255/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 256/300\n", + "Solving for Nash Equilibrium in Generation 256/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 257/300\n", + "Solving for Nash Equilibrium in Generation 257/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 258/300\n", + "Solving for Nash Equilibrium in Generation 258/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 259/300\n", + "Solving for Nash Equilibrium in Generation 259/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 260/300\n", + "Solving for Nash Equilibrium in Generation 260/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 261/300\n", + "Solving for Nash Equilibrium in Generation 261/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 262/300\n", + "Solving for Nash Equilibrium in Generation 262/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 263/300\n", + "Solving for Nash Equilibrium in Generation 263/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 264/300\n", + "Solving for Nash Equilibrium in Generation 264/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 265/300\n", + "Solving for Nash Equilibrium in Generation 265/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 266/300\n", + "Solving for Nash Equilibrium in Generation 266/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 267/300\n", + "Solving for Nash Equilibrium in Generation 267/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 268/300\n", + "Solving for Nash Equilibrium in Generation 268/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 269/300\n", + "Solving for Nash Equilibrium in Generation 269/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 270/300\n", + "Solving for Nash Equilibrium in Generation 270/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 271/300\n", + "Solving for Nash Equilibrium in Generation 271/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 272/300\n", + "Solving for Nash Equilibrium in Generation 272/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 273/300\n", + "Solving for Nash Equilibrium in Generation 273/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 274/300\n", + "Solving for Nash Equilibrium in Generation 274/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 275/300\n", + "Solving for Nash Equilibrium in Generation 275/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 276/300\n", + "Solving for Nash Equilibrium in Generation 276/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 277/300\n", + "Solving for Nash Equilibrium in Generation 277/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 278/300\n", + "Solving for Nash Equilibrium in Generation 278/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 279/300\n", + "Solving for Nash Equilibrium in Generation 279/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 280/300\n", + "Solving for Nash Equilibrium in Generation 280/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 281/300\n", + "Solving for Nash Equilibrium in Generation 281/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 282/300\n", + "Solving for Nash Equilibrium in Generation 282/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 283/300\n", + "Solving for Nash Equilibrium in Generation 283/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 284/300\n", + "Solving for Nash Equilibrium in Generation 284/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 285/300\n", + "Solving for Nash Equilibrium in Generation 285/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 286/300\n", + "Solving for Nash Equilibrium in Generation 286/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 287/300\n", + "Solving for Nash Equilibrium in Generation 287/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 288/300\n", + "Solving for Nash Equilibrium in Generation 288/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 289/300\n", + "Solving for Nash Equilibrium in Generation 289/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 290/300\n", + "Solving for Nash Equilibrium in Generation 290/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 291/300\n", + "Solving for Nash Equilibrium in Generation 291/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 292/300\n", + "Solving for Nash Equilibrium in Generation 292/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 293/300\n", + "Solving for Nash Equilibrium in Generation 293/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 294/300\n", + "Solving for Nash Equilibrium in Generation 294/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 295/300\n", + "Solving for Nash Equilibrium in Generation 295/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 296/300\n", + "Solving for Nash Equilibrium in Generation 296/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 297/300\n", + "Solving for Nash Equilibrium in Generation 297/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 298/300\n", + "Solving for Nash Equilibrium in Generation 298/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 299/300\n", + "Solving for Nash Equilibrium in Generation 299/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 300/300\n", + "Solving for Nash Equilibrium in Generation 300/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Running Computed Policies\n", + "Episode 0 done\n", + "Episode 1 done\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", + "/opt/homebrew/lib/python3.11/site-packages/numpy/core/fromnumeric.py:3504: RuntimeWarning: Mean of empty slice.\n", + " return _methods._mean(a, axis=axis, dtype=dtype,\n", + "/opt/homebrew/lib/python3.11/site-packages/numpy/core/_methods.py:129: RuntimeWarning: invalid value encountered in scalar divide\n", + " ret = ret.dtype.type(ret / rcount)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Episode 2 done\n", + "Episode 3 done\n", + "Episode 4 done\n", + "Episode 5 done\n", + "Episode 6 done\n", + "Episode 7 done\n", + "Episode 8 done\n", + "Episode 9 done\n", + "Episode 10 done\n", + "Episode 10 done\n", + "Total steps: 13\n", + "Episode 11 done\n", + "Episode 11 done\n", + "Total steps: 13\n", + "Episode 12 done\n", + "Episode 12 done\n", + "Total steps: 13\n", + "Episode 13 done\n", + "Episode 13 done\n", + "Total steps: 13\n", + "Episode 14 done\n", + "Episode 15 done\n", + "Episode 15 done\n", + "Total steps: 13\n", + "Episode 16 done\n", + "Episode 17 done\n", + "Episode 17 done\n", + "Total steps: 1\n", + "Episode 18 done\n", + "Episode 19 done\n", + "Episode 20 done\n", + "Episode 20 done\n", + "Total steps: 13\n", + "Episode 21 done\n", + "Episode 22 done\n", + "Episode 23 done\n", + "Episode 24 done\n", + "Episode 25 done\n", + "Episode 26 done\n", + "Episode 27 done\n", + "Episode 27 done\n", + "Total steps: 13\n", + "Episode 28 done\n", + "Episode 28 done\n", + "Total steps: 13\n", + "Episode 29 done\n", + "Episode 29 done\n", + "Total steps: 1\n", + "Episode 30 done\n", + "Episode 31 done\n", + "Episode 32 done\n", + "Episode 33 done\n", + "Episode 33 done\n", + "Total steps: 13\n", + "Episode 34 done\n", + "Episode 35 done\n", + "Episode 36 done\n", + "Episode 36 done\n", + "Total steps: 13\n", + "Episode 37 done\n", + "Episode 37 done\n", + "Total steps: 13\n", + "Episode 38 done\n", + "Episode 39 done\n", + "Episode 40 done\n", + "Episode 40 done\n", + "Total steps: 13\n", + "Episode 41 done\n", + "Episode 42 done\n", + "Episode 43 done\n", + "Episode 44 done\n", + "Episode 45 done\n", + "Episode 45 done\n", + "Total steps: 13\n", + "Episode 46 done\n", + "Episode 47 done\n", + "Episode 47 done\n", + "Total steps: 13\n", + "Episode 48 done\n", + "Episode 48 done\n", + "Total steps: 1\n", + "Episode 49 done\n", + "Episode 50 done\n", + "Episode 51 done\n", + "Episode 52 done\n", + "Episode 53 done\n", + "Episode 54 done\n", + "Episode 55 done\n", + "Episode 56 done\n", + "Episode 56 done\n", + "Total steps: 13\n", + "Episode 57 done\n", + "Episode 58 done\n", + "Episode 58 done\n", + "Total steps: 13\n", + "Episode 59 done\n", + "Episode 60 done\n", + "Episode 61 done\n", + "Episode 62 done\n", + "Episode 63 done\n", + "Episode 64 done\n", + "Episode 65 done\n", + "Episode 66 done\n", + "Episode 67 done\n", + "Episode 68 done\n", + "Episode 69 done\n", + "Episode 70 done\n", + "Episode 71 done\n", + "Episode 71 done\n", + "Total steps: 13\n", + "Episode 72 done\n", + "Episode 73 done\n", + "Episode 74 done\n", + "Episode 75 done\n", + "Episode 76 done\n", + "Episode 77 done\n", + "Episode 78 done\n", + "Episode 78 done\n", + "Total steps: 13\n", + "Episode 79 done\n", + "Episode 80 done\n", + "Episode 81 done\n", + "Episode 82 done\n", + "Episode 83 done\n", + "Episode 84 done\n", + "Episode 85 done\n", + "Episode 86 done\n", + "Episode 87 done\n", + "Episode 88 done\n", + "Episode 89 done\n", + "Episode 90 done\n", + "Episode 91 done\n", + "Episode 91 done\n", + "Total steps: 13\n", + "Episode 92 done\n", + "Episode 93 done\n", + "Episode 94 done\n", + "Episode 95 done\n", + "Episode 96 done\n", + "Episode 97 done\n", + "Episode 98 done\n", + "Episode 99 done\n", + "Episode 100 done\n", + "Episode 101 done\n", + "Episode 102 done\n", + "Episode 103 done\n", + "Episode 104 done\n", + "Episode 105 done\n", + "Episode 106 done\n", + "Episode 107 done\n", + "Episode 108 done\n", + "Episode 109 done\n", + "Episode 110 done\n", + "Episode 111 done\n", + "Episode 111 done\n", + "Total steps: 13\n", + "Episode 112 done\n", + "Episode 113 done\n", + "Episode 114 done\n", + "Episode 115 done\n", + "Episode 116 done\n", + "Episode 116 done\n", + "Total steps: 13\n", + "Episode 117 done\n", + "Episode 118 done\n", + "Episode 118 done\n", + "Total steps: 13\n", + "Episode 119 done\n", + "Episode 120 done\n", + "Episode 121 done\n", + "Episode 122 done\n", + "Episode 123 done\n", + "Episode 124 done\n", + "Episode 125 done\n", + "Episode 126 done\n", + "Episode 127 done\n", + "Episode 128 done\n", + "Episode 128 done\n", + "Total steps: 13\n", + "Episode 129 done\n", + "Episode 130 done\n", + "Episode 131 done\n", + "Episode 131 done\n", + "Total steps: 13\n", + "Episode 132 done\n", + "Episode 133 done\n", + "Episode 133 done\n", + "Total steps: 13\n", + "Episode 134 done\n", + "Episode 135 done\n", + "Episode 136 done\n", + "Episode 137 done\n", + "Episode 138 done\n", + "Episode 138 done\n", + "Total steps: 13\n", + "Episode 139 done\n", + "Episode 140 done\n", + "Episode 141 done\n", + "Episode 142 done\n", + "Episode 142 done\n", + "Total steps: 13\n", + "Episode 143 done\n", + "Episode 144 done\n", + "Episode 144 done\n", + "Total steps: 13\n", + "Episode 145 done\n", + "Episode 145 done\n", + "Total steps: 13\n", + "Episode 146 done\n", + "Episode 147 done\n", + "Episode 148 done\n", + "Episode 148 done\n", + "Total steps: 13\n", + "Episode 149 done\n", + "Episode 150 done\n", + "Episode 151 done\n", + "Episode 152 done\n", + "Episode 153 done\n", + "Episode 154 done\n", + "Episode 155 done\n", + "Episode 156 done\n", + "Episode 157 done\n", + "Episode 158 done\n", + "Episode 159 done\n", + "Episode 160 done\n", + "Episode 161 done\n", + "Episode 162 done\n", + "Episode 163 done\n", + "Episode 164 done\n", + "Episode 165 done\n", + "Episode 165 done\n", + "Total steps: 13\n", + "Episode 166 done\n", + "Episode 166 done\n", + "Total steps: 13\n", + "Episode 167 done\n", + "Episode 167 done\n", + "Total steps: 13\n", + "Episode 168 done\n", + "Episode 169 done\n", + "Episode 170 done\n", + "Episode 171 done\n", + "Episode 172 done\n", + "Episode 172 done\n", + "Total steps: 13\n", + "Episode 173 done\n", + "Episode 174 done\n", + "Episode 175 done\n", + "Episode 175 done\n", + "Total steps: 13\n", + "Episode 176 done\n", + "Episode 177 done\n", + "Episode 178 done\n", + "Episode 179 done\n", + "Episode 180 done\n", + "Episode 181 done\n", + "Episode 182 done\n", + "Episode 183 done\n", + "Episode 184 done\n", + "Episode 185 done\n", + "Episode 186 done\n", + "Episode 186 done\n", + "Total steps: 13\n", + "Episode 187 done\n", + "Episode 188 done\n", + "Episode 189 done\n", + "Episode 190 done\n", + "Episode 191 done\n", + "Episode 192 done\n", + "Episode 193 done\n", + "Episode 194 done\n", + "Episode 195 done\n", + "Episode 195 done\n", + "Total steps: 1\n", + "Episode 196 done\n", + "Episode 197 done\n", + "Episode 198 done\n", + "Episode 199 done\n", + "Episode 200 done\n", + "Episode 201 done\n", + "Episode 202 done\n", + "Episode 203 done\n", + "Episode 204 done\n", + "Episode 205 done\n", + "Episode 206 done\n", + "Episode 206 done\n", + "Total steps: 13\n", + "Episode 207 done\n", + "Episode 207 done\n", + "Total steps: 13\n", + "Episode 208 done\n", + "Episode 209 done\n", + "Episode 210 done\n", + "Episode 211 done\n", + "Episode 212 done\n", + "Episode 212 done\n", + "Total steps: 13\n", + "Episode 213 done\n", + "Episode 214 done\n", + "Episode 215 done\n", + "Episode 216 done\n", + "Episode 217 done\n", + "Episode 218 done\n", + "Episode 219 done\n", + "Episode 220 done\n", + "Episode 221 done\n", + "Episode 222 done\n", + "Episode 223 done\n", + "Episode 224 done\n", + "Episode 225 done\n", + "Episode 226 done\n", + "Episode 227 done\n", + "Episode 228 done\n", + "Episode 229 done\n", + "Episode 230 done\n", + "Episode 231 done\n", + "Episode 232 done\n", + "Episode 233 done\n", + "Episode 234 done\n", + "Episode 234 done\n", + "Total steps: 13\n", + "Episode 235 done\n", + "Episode 235 done\n", + "Total steps: 13\n", + "Episode 236 done\n", + "Episode 237 done\n", + "Episode 237 done\n", + "Total steps: 13\n", + "Episode 238 done\n", + "Episode 239 done\n", + "Episode 240 done\n", + "Episode 240 done\n", + "Total steps: 13\n", + "Episode 241 done\n", + "Episode 242 done\n", + "Episode 243 done\n", + "Episode 244 done\n", + "Episode 245 done\n", + "Episode 245 done\n", + "Total steps: 13\n", + "Episode 246 done\n", + "Episode 247 done\n", + "Episode 247 done\n", + "Total steps: 13\n", + "Episode 248 done\n", + "Episode 249 done\n", + "Episode 250 done\n", + "Episode 251 done\n", + "Episode 252 done\n", + "Episode 252 done\n", + "Total steps: 13\n", + "Episode 253 done\n", + "Episode 254 done\n", + "Episode 255 done\n", + "Episode 256 done\n", + "Episode 257 done\n", + "Episode 258 done\n", + "Episode 259 done\n", + "Episode 260 done\n", + "Episode 260 done\n", + "Total steps: 13\n", + "Episode 261 done\n", + "Episode 262 done\n", + "Episode 263 done\n", + "Episode 264 done\n", + "Episode 265 done\n", + "Episode 266 done\n", + "Episode 267 done\n", + "Episode 268 done\n", + "Episode 269 done\n", + "Episode 269 done\n", + "Total steps: 13\n", + "Episode 270 done\n", + "Episode 271 done\n", + "Episode 272 done\n", + "Episode 273 done\n", + "Episode 274 done\n", + "Episode 275 done\n", + "Episode 276 done\n", + "Episode 277 done\n", + "Episode 278 done\n", + "Episode 279 done\n", + "Episode 280 done\n", + "Episode 281 done\n", + "Episode 282 done\n", + "Episode 283 done\n", + "Episode 284 done\n", + "Episode 284 done\n", + "Total steps: 13\n", + "Episode 285 done\n", + "Episode 286 done\n", + "Episode 287 done\n", + "Episode 288 done\n", + "Episode 288 done\n", + "Total steps: 13\n", + "Episode 289 done\n", + "Episode 289 done\n", + "Total steps: 1\n", + "Episode 290 done\n", + "Episode 291 done\n", + "Episode 292 done\n", + "Episode 293 done\n", + "Episode 294 done\n", + "Episode 295 done\n", + "Episode 296 done\n", + "Episode 297 done\n", + "Episode 298 done\n", + "Episode 299 done\n", + " seed step_collisions not_on_track yield_violations \\\n", + "0 42.0 0.0 19.0 0.0 \n", + "1 42.0 0.0 9.0 0.0 \n", + "2 42.0 0.0 13.0 0.0 \n", + "3 42.0 0.0 13.0 0.0 \n", + "4 42.0 0.0 8.0 0.0 \n", + ".. ... ... ... ... \n", + "295 42.0 0.0 18.0 0.0 \n", + "296 42.0 0.0 18.0 0.0 \n", + "297 42.0 0.0 13.0 0.0 \n", + "298 42.0 0.0 13.0 0.0 \n", + "299 42.0 0.0 18.0 0.0 \n", + "\n", + " unncessary_brake_violations efficient_crossing_violations \\\n", + "0 0.0 342.0 \n", + "1 0.0 114.0 \n", + "2 0.0 252.0 \n", + "3 0.0 252.0 \n", + "4 0.0 138.0 \n", + ".. ... ... \n", + "295 0.0 366.0 \n", + "296 0.0 366.0 \n", + "297 0.0 252.0 \n", + "298 0.0 252.0 \n", + "299 0.0 366.0 \n", + "\n", + " total_violations_cost \n", + "0 -5524.5 \n", + "1 -640.5 \n", + "2 -2951.0 \n", + "3 -2951.0 \n", + "4 -923.0 \n", + ".. ... \n", + "295 -6196.0 \n", + "296 -6196.0 \n", + "297 -2951.0 \n", + "298 -2951.0 \n", + "299 -6196.0 \n", + "\n", + "[300 rows x 7 columns]\n", + "Generation 1/300\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Solving for Nash Equilibrium in Generation 1/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 2/300\n", + "Solving for Nash Equilibrium in Generation 2/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 3/300\n", + "Solving for Nash Equilibrium in Generation 3/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 4/300\n", + "Solving for Nash Equilibrium in Generation 4/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 5/300\n", + "Solving for Nash Equilibrium in Generation 5/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 6/300\n", + "Solving for Nash Equilibrium in Generation 6/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 7/300\n", + "Solving for Nash Equilibrium in Generation 7/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 8/300\n", + "Solving for Nash Equilibrium in Generation 8/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 9/300\n", + "Solving for Nash Equilibrium in Generation 9/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 10/300\n", + "Solving for Nash Equilibrium in Generation 10/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 11/300\n", + "Solving for Nash Equilibrium in Generation 11/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 12/300\n", + "Solving for Nash Equilibrium in Generation 12/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 13/300\n", + "Solving for Nash Equilibrium in Generation 13/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 14/300\n", + "Solving for Nash Equilibrium in Generation 14/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 15/300\n", + "Solving for Nash Equilibrium in Generation 15/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 16/300\n", + "Solving for Nash Equilibrium in Generation 16/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 17/300\n", + "Solving for Nash Equilibrium in Generation 17/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 18/300\n", + "Solving for Nash Equilibrium in Generation 18/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 19/300\n", + "Solving for Nash Equilibrium in Generation 19/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 20/300\n", + "Solving for Nash Equilibrium in Generation 20/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 21/300\n", + "Solving for Nash Equilibrium in Generation 21/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 22/300\n", + "Solving for Nash Equilibrium in Generation 22/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 23/300\n", + "Solving for Nash Equilibrium in Generation 23/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 24/300\n", + "Solving for Nash Equilibrium in Generation 24/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 25/300\n", + "Solving for Nash Equilibrium in Generation 25/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 26/300\n", + "Solving for Nash Equilibrium in Generation 26/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 27/300\n", + "Solving for Nash Equilibrium in Generation 27/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 28/300\n", + "Solving for Nash Equilibrium in Generation 28/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 29/300\n", + "Solving for Nash Equilibrium in Generation 29/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 30/300\n", + "Solving for Nash Equilibrium in Generation 30/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 31/300\n", + "Solving for Nash Equilibrium in Generation 31/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 32/300\n", + "Solving for Nash Equilibrium in Generation 32/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 33/300\n", + "Solving for Nash Equilibrium in Generation 33/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 34/300\n", + "Solving for Nash Equilibrium in Generation 34/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 35/300\n", + "Solving for Nash Equilibrium in Generation 35/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 36/300\n", + "Solving for Nash Equilibrium in Generation 36/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 37/300\n", + "Solving for Nash Equilibrium in Generation 37/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 38/300\n", + "Solving for Nash Equilibrium in Generation 38/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 39/300\n", + "Solving for Nash Equilibrium in Generation 39/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 40/300\n", + "Solving for Nash Equilibrium in Generation 40/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 41/300\n", + "Solving for Nash Equilibrium in Generation 41/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 42/300\n", + "Solving for Nash Equilibrium in Generation 42/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 43/300\n", + "Solving for Nash Equilibrium in Generation 43/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 44/300\n", + "Solving for Nash Equilibrium in Generation 44/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 45/300\n", + "Solving for Nash Equilibrium in Generation 45/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 46/300\n", + "Solving for Nash Equilibrium in Generation 46/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 47/300\n", + "Solving for Nash Equilibrium in Generation 47/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 48/300\n", + "Solving for Nash Equilibrium in Generation 48/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 49/300\n", + "Solving for Nash Equilibrium in Generation 49/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 50/300\n", + "Solving for Nash Equilibrium in Generation 50/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 51/300\n", + "Solving for Nash Equilibrium in Generation 51/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 52/300\n", + "Solving for Nash Equilibrium in Generation 52/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 53/300\n", + "Solving for Nash Equilibrium in Generation 53/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 54/300\n", + "Solving for Nash Equilibrium in Generation 54/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 55/300\n", + "Solving for Nash Equilibrium in Generation 55/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 56/300\n", + "Solving for Nash Equilibrium in Generation 56/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 57/300\n", + "Solving for Nash Equilibrium in Generation 57/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 58/300\n", + "Solving for Nash Equilibrium in Generation 58/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 59/300\n", + "Solving for Nash Equilibrium in Generation 59/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 60/300\n", + "Solving for Nash Equilibrium in Generation 60/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 61/300\n", + "Solving for Nash Equilibrium in Generation 61/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 62/300\n", + "Solving for Nash Equilibrium in Generation 62/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 63/300\n", + "Solving for Nash Equilibrium in Generation 63/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 64/300\n", + "Solving for Nash Equilibrium in Generation 64/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 65/300\n", + "Solving for Nash Equilibrium in Generation 65/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 66/300\n", + "Solving for Nash Equilibrium in Generation 66/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 67/300\n", + "Solving for Nash Equilibrium in Generation 67/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 68/300\n", + "Solving for Nash Equilibrium in Generation 68/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 69/300\n", + "Solving for Nash Equilibrium in Generation 69/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 70/300\n", + "Solving for Nash Equilibrium in Generation 70/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 71/300\n", + "Solving for Nash Equilibrium in Generation 71/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 72/300\n", + "Solving for Nash Equilibrium in Generation 72/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 73/300\n", + "Solving for Nash Equilibrium in Generation 73/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 74/300\n", + "Solving for Nash Equilibrium in Generation 74/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 75/300\n", + "Solving for Nash Equilibrium in Generation 75/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 76/300\n", + "Solving for Nash Equilibrium in Generation 76/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 77/300\n", + "Solving for Nash Equilibrium in Generation 77/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 78/300\n", + "Solving for Nash Equilibrium in Generation 78/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 79/300\n", + "Solving for Nash Equilibrium in Generation 79/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 80/300\n", + "Solving for Nash Equilibrium in Generation 80/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 81/300\n", + "Solving for Nash Equilibrium in Generation 81/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 82/300\n", + "Solving for Nash Equilibrium in Generation 82/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 83/300\n", + "Solving for Nash Equilibrium in Generation 83/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 84/300\n", + "Solving for Nash Equilibrium in Generation 84/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 85/300\n", + "Solving for Nash Equilibrium in Generation 85/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 86/300\n", + "Solving for Nash Equilibrium in Generation 86/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 87/300\n", + "Solving for Nash Equilibrium in Generation 87/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 88/300\n", + "Solving for Nash Equilibrium in Generation 88/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 89/300\n", + "Solving for Nash Equilibrium in Generation 89/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 90/300\n", + "Solving for Nash Equilibrium in Generation 90/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 91/300\n", + "Solving for Nash Equilibrium in Generation 91/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 92/300\n", + "Solving for Nash Equilibrium in Generation 92/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 93/300\n", + "Solving for Nash Equilibrium in Generation 93/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 94/300\n", + "Solving for Nash Equilibrium in Generation 94/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 95/300\n", + "Solving for Nash Equilibrium in Generation 95/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 96/300\n", + "Solving for Nash Equilibrium in Generation 96/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 97/300\n", + "Solving for Nash Equilibrium in Generation 97/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 98/300\n", + "Solving for Nash Equilibrium in Generation 98/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 99/300\n", + "Solving for Nash Equilibrium in Generation 99/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 100/300\n", + "Solving for Nash Equilibrium in Generation 100/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 101/300\n", + "Solving for Nash Equilibrium in Generation 101/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 102/300\n", + "Solving for Nash Equilibrium in Generation 102/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 103/300\n", + "Solving for Nash Equilibrium in Generation 103/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 104/300\n", + "Solving for Nash Equilibrium in Generation 104/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 105/300\n", + "Solving for Nash Equilibrium in Generation 105/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 106/300\n", + "Solving for Nash Equilibrium in Generation 106/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 107/300\n", + "Solving for Nash Equilibrium in Generation 107/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 108/300\n", + "Solving for Nash Equilibrium in Generation 108/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 109/300\n", + "Solving for Nash Equilibrium in Generation 109/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 110/300\n", + "Solving for Nash Equilibrium in Generation 110/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 111/300\n", + "Solving for Nash Equilibrium in Generation 111/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 112/300\n", + "Solving for Nash Equilibrium in Generation 112/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 113/300\n", + "Solving for Nash Equilibrium in Generation 113/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 114/300\n", + "Solving for Nash Equilibrium in Generation 114/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 115/300\n", + "Solving for Nash Equilibrium in Generation 115/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 116/300\n", + "Solving for Nash Equilibrium in Generation 116/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 117/300\n", + "Solving for Nash Equilibrium in Generation 117/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 118/300\n", + "Solving for Nash Equilibrium in Generation 118/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 119/300\n", + "Solving for Nash Equilibrium in Generation 119/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 120/300\n", + "Solving for Nash Equilibrium in Generation 120/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 121/300\n", + "Solving for Nash Equilibrium in Generation 121/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 122/300\n", + "Solving for Nash Equilibrium in Generation 122/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 123/300\n", + "Solving for Nash Equilibrium in Generation 123/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 124/300\n", + "Solving for Nash Equilibrium in Generation 124/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 125/300\n", + "Solving for Nash Equilibrium in Generation 125/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 126/300\n", + "Solving for Nash Equilibrium in Generation 126/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 127/300\n", + "Solving for Nash Equilibrium in Generation 127/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 128/300\n", + "Solving for Nash Equilibrium in Generation 128/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 129/300\n", + "Solving for Nash Equilibrium in Generation 129/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 130/300\n", + "Solving for Nash Equilibrium in Generation 130/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 131/300\n", + "Solving for Nash Equilibrium in Generation 131/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 132/300\n", + "Solving for Nash Equilibrium in Generation 132/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 133/300\n", + "Solving for Nash Equilibrium in Generation 133/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 134/300\n", + "Solving for Nash Equilibrium in Generation 134/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 135/300\n", + "Solving for Nash Equilibrium in Generation 135/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 136/300\n", + "Solving for Nash Equilibrium in Generation 136/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 137/300\n", + "Solving for Nash Equilibrium in Generation 137/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 138/300\n", + "Solving for Nash Equilibrium in Generation 138/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 139/300\n", + "Solving for Nash Equilibrium in Generation 139/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 140/300\n", + "Solving for Nash Equilibrium in Generation 140/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 141/300\n", + "Solving for Nash Equilibrium in Generation 141/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 142/300\n", + "Solving for Nash Equilibrium in Generation 142/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 143/300\n", + "Solving for Nash Equilibrium in Generation 143/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 144/300\n", + "Solving for Nash Equilibrium in Generation 144/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 145/300\n", + "Solving for Nash Equilibrium in Generation 145/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 146/300\n", + "Solving for Nash Equilibrium in Generation 146/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 147/300\n", + "Solving for Nash Equilibrium in Generation 147/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 148/300\n", + "Solving for Nash Equilibrium in Generation 148/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 149/300\n", + "Solving for Nash Equilibrium in Generation 149/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 150/300\n", + "Solving for Nash Equilibrium in Generation 150/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 151/300\n", + "Solving for Nash Equilibrium in Generation 151/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 152/300\n", + "Solving for Nash Equilibrium in Generation 152/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 153/300\n", + "Solving for Nash Equilibrium in Generation 153/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 154/300\n", + "Solving for Nash Equilibrium in Generation 154/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 155/300\n", + "Solving for Nash Equilibrium in Generation 155/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 156/300\n", + "Solving for Nash Equilibrium in Generation 156/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 157/300\n", + "Solving for Nash Equilibrium in Generation 157/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 158/300\n", + "Solving for Nash Equilibrium in Generation 158/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 159/300\n", + "Solving for Nash Equilibrium in Generation 159/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 160/300\n", + "Solving for Nash Equilibrium in Generation 160/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 161/300\n", + "Solving for Nash Equilibrium in Generation 161/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 162/300\n", + "Solving for Nash Equilibrium in Generation 162/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 163/300\n", + "Solving for Nash Equilibrium in Generation 163/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 164/300\n", + "Solving for Nash Equilibrium in Generation 164/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 165/300\n", + "Solving for Nash Equilibrium in Generation 165/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 166/300\n", + "Solving for Nash Equilibrium in Generation 166/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 167/300\n", + "Solving for Nash Equilibrium in Generation 167/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 168/300\n", + "Solving for Nash Equilibrium in Generation 168/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 169/300\n", + "Solving for Nash Equilibrium in Generation 169/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 170/300\n", + "Solving for Nash Equilibrium in Generation 170/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 171/300\n", + "Solving for Nash Equilibrium in Generation 171/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 172/300\n", + "Solving for Nash Equilibrium in Generation 172/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 173/300\n", + "Solving for Nash Equilibrium in Generation 173/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 174/300\n", + "Solving for Nash Equilibrium in Generation 174/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 175/300\n", + "Solving for Nash Equilibrium in Generation 175/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 176/300\n", + "Solving for Nash Equilibrium in Generation 176/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 177/300\n", + "Solving for Nash Equilibrium in Generation 177/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 178/300\n", + "Solving for Nash Equilibrium in Generation 178/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 179/300\n", + "Solving for Nash Equilibrium in Generation 179/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 180/300\n", + "Solving for Nash Equilibrium in Generation 180/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 181/300\n", + "Solving for Nash Equilibrium in Generation 181/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 182/300\n", + "Solving for Nash Equilibrium in Generation 182/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 183/300\n", + "Solving for Nash Equilibrium in Generation 183/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 184/300\n", + "Solving for Nash Equilibrium in Generation 184/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 185/300\n", + "Solving for Nash Equilibrium in Generation 185/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 186/300\n", + "Solving for Nash Equilibrium in Generation 186/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 187/300\n", + "Solving for Nash Equilibrium in Generation 187/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 188/300\n", + "Solving for Nash Equilibrium in Generation 188/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 189/300\n", + "Solving for Nash Equilibrium in Generation 189/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 190/300\n", + "Solving for Nash Equilibrium in Generation 190/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 191/300\n", + "Solving for Nash Equilibrium in Generation 191/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 192/300\n", + "Solving for Nash Equilibrium in Generation 192/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 193/300\n", + "Solving for Nash Equilibrium in Generation 193/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 194/300\n", + "Solving for Nash Equilibrium in Generation 194/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 195/300\n", + "Solving for Nash Equilibrium in Generation 195/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 196/300\n", + "Solving for Nash Equilibrium in Generation 196/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 197/300\n", + "Solving for Nash Equilibrium in Generation 197/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 198/300\n", + "Solving for Nash Equilibrium in Generation 198/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 199/300\n", + "Solving for Nash Equilibrium in Generation 199/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 200/300\n", + "Solving for Nash Equilibrium in Generation 200/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 201/300\n", + "Solving for Nash Equilibrium in Generation 201/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 202/300\n", + "Solving for Nash Equilibrium in Generation 202/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 203/300\n", + "Solving for Nash Equilibrium in Generation 203/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 204/300\n", + "Solving for Nash Equilibrium in Generation 204/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 205/300\n", + "Solving for Nash Equilibrium in Generation 205/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 206/300\n", + "Solving for Nash Equilibrium in Generation 206/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 207/300\n", + "Solving for Nash Equilibrium in Generation 207/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 208/300\n", + "Solving for Nash Equilibrium in Generation 208/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 209/300\n", + "Solving for Nash Equilibrium in Generation 209/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 210/300\n", + "Solving for Nash Equilibrium in Generation 210/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 211/300\n", + "Solving for Nash Equilibrium in Generation 211/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 212/300\n", + "Solving for Nash Equilibrium in Generation 212/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 213/300\n", + "Solving for Nash Equilibrium in Generation 213/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 214/300\n", + "Solving for Nash Equilibrium in Generation 214/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 215/300\n", + "Solving for Nash Equilibrium in Generation 215/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 216/300\n", + "Solving for Nash Equilibrium in Generation 216/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 217/300\n", + "Solving for Nash Equilibrium in Generation 217/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 218/300\n", + "Solving for Nash Equilibrium in Generation 218/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 219/300\n", + "Solving for Nash Equilibrium in Generation 219/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 220/300\n", + "Solving for Nash Equilibrium in Generation 220/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 221/300\n", + "Solving for Nash Equilibrium in Generation 221/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 222/300\n", + "Solving for Nash Equilibrium in Generation 222/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 223/300\n", + "Solving for Nash Equilibrium in Generation 223/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 224/300\n", + "Solving for Nash Equilibrium in Generation 224/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 225/300\n", + "Solving for Nash Equilibrium in Generation 225/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 226/300\n", + "Solving for Nash Equilibrium in Generation 226/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 227/300\n", + "Solving for Nash Equilibrium in Generation 227/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 228/300\n", + "Solving for Nash Equilibrium in Generation 228/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 229/300\n", + "Solving for Nash Equilibrium in Generation 229/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 230/300\n", + "Solving for Nash Equilibrium in Generation 230/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 231/300\n", + "Solving for Nash Equilibrium in Generation 231/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 232/300\n", + "Solving for Nash Equilibrium in Generation 232/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 233/300\n", + "Solving for Nash Equilibrium in Generation 233/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 234/300\n", + "Solving for Nash Equilibrium in Generation 234/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 235/300\n", + "Solving for Nash Equilibrium in Generation 235/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 236/300\n", + "Solving for Nash Equilibrium in Generation 236/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 237/300\n", + "Solving for Nash Equilibrium in Generation 237/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 238/300\n", + "Solving for Nash Equilibrium in Generation 238/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 239/300\n", + "Solving for Nash Equilibrium in Generation 239/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 240/300\n", + "Solving for Nash Equilibrium in Generation 240/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 241/300\n", + "Solving for Nash Equilibrium in Generation 241/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 242/300\n", + "Solving for Nash Equilibrium in Generation 242/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 243/300\n", + "Solving for Nash Equilibrium in Generation 243/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 244/300\n", + "Solving for Nash Equilibrium in Generation 244/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 245/300\n", + "Solving for Nash Equilibrium in Generation 245/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 246/300\n", + "Solving for Nash Equilibrium in Generation 246/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 247/300\n", + "Solving for Nash Equilibrium in Generation 247/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 248/300\n", + "Solving for Nash Equilibrium in Generation 248/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 249/300\n", + "Solving for Nash Equilibrium in Generation 249/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 250/300\n", + "Solving for Nash Equilibrium in Generation 250/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 251/300\n", + "Solving for Nash Equilibrium in Generation 251/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 252/300\n", + "Solving for Nash Equilibrium in Generation 252/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 253/300\n", + "Solving for Nash Equilibrium in Generation 253/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 254/300\n", + "Solving for Nash Equilibrium in Generation 254/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 255/300\n", + "Solving for Nash Equilibrium in Generation 255/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 256/300\n", + "Solving for Nash Equilibrium in Generation 256/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 257/300\n", + "Solving for Nash Equilibrium in Generation 257/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 258/300\n", + "Solving for Nash Equilibrium in Generation 258/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 259/300\n", + "Solving for Nash Equilibrium in Generation 259/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 260/300\n", + "Solving for Nash Equilibrium in Generation 260/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 261/300\n", + "Solving for Nash Equilibrium in Generation 261/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 262/300\n", + "Solving for Nash Equilibrium in Generation 262/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 263/300\n", + "Solving for Nash Equilibrium in Generation 263/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 264/300\n", + "Solving for Nash Equilibrium in Generation 264/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 265/300\n", + "Solving for Nash Equilibrium in Generation 265/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 266/300\n", + "Solving for Nash Equilibrium in Generation 266/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 267/300\n", + "Solving for Nash Equilibrium in Generation 267/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 268/300\n", + "Solving for Nash Equilibrium in Generation 268/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 269/300\n", + "Solving for Nash Equilibrium in Generation 269/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 270/300\n", + "Solving for Nash Equilibrium in Generation 270/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 271/300\n", + "Solving for Nash Equilibrium in Generation 271/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 272/300\n", + "Solving for Nash Equilibrium in Generation 272/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 273/300\n", + "Solving for Nash Equilibrium in Generation 273/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 274/300\n", + "Solving for Nash Equilibrium in Generation 274/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 275/300\n", + "Solving for Nash Equilibrium in Generation 275/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 276/300\n", + "Solving for Nash Equilibrium in Generation 276/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 277/300\n", + "Solving for Nash Equilibrium in Generation 277/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 278/300\n", + "Solving for Nash Equilibrium in Generation 278/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 279/300\n", + "Solving for Nash Equilibrium in Generation 279/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 280/300\n", + "Solving for Nash Equilibrium in Generation 280/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 281/300\n", + "Solving for Nash Equilibrium in Generation 281/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 282/300\n", + "Solving for Nash Equilibrium in Generation 282/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 283/300\n", + "Solving for Nash Equilibrium in Generation 283/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 284/300\n", + "Solving for Nash Equilibrium in Generation 284/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 285/300\n", + "Solving for Nash Equilibrium in Generation 285/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 286/300\n", + "Solving for Nash Equilibrium in Generation 286/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 287/300\n", + "Solving for Nash Equilibrium in Generation 287/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 288/300\n", + "Solving for Nash Equilibrium in Generation 288/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 289/300\n", + "Solving for Nash Equilibrium in Generation 289/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 290/300\n", + "Solving for Nash Equilibrium in Generation 290/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 291/300\n", + "Solving for Nash Equilibrium in Generation 291/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 292/300\n", + "Solving for Nash Equilibrium in Generation 292/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 293/300\n", + "Solving for Nash Equilibrium in Generation 293/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 294/300\n", + "Solving for Nash Equilibrium in Generation 294/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 295/300\n", + "Solving for Nash Equilibrium in Generation 295/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 296/300\n", + "Solving for Nash Equilibrium in Generation 296/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 297/300\n", + "Solving for Nash Equilibrium in Generation 297/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 298/300\n", + "Solving for Nash Equilibrium in Generation 298/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 299/300\n", + "Solving for Nash Equilibrium in Generation 299/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 300/300\n", + "Solving for Nash Equilibrium in Generation 300/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Running Computed Policies\n", + "Episode 0 done\n", + "Episode 1 done\n", + "Episode 1 done\n", + "Total steps: 13\n", + "Episode 2 done\n", + "Episode 2 done\n", + "Total steps: 13\n", + "Episode 3 done\n", + "Episode 3 done\n", + "Total steps: 13\n", + "Episode 4 done\n", + "Episode 4 done\n", + "Total steps: 1\n", + "Episode 5 done\n", + "Episode 6 done\n", + "Episode 6 done\n", + "Total steps: 13\n", + "Episode 7 done\n", + "Episode 8 done\n", + "Episode 9 done\n", + "Episode 10 done\n", + "Episode 11 done\n", + "Episode 11 done\n", + "Total steps: 13\n" + ] + }, + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", + "/opt/homebrew/lib/python3.11/site-packages/numpy/core/fromnumeric.py:3504: RuntimeWarning: Mean of empty slice.\n", + " return _methods._mean(a, axis=axis, dtype=dtype,\n", + "/opt/homebrew/lib/python3.11/site-packages/numpy/core/_methods.py:129: RuntimeWarning: invalid value encountered in scalar divide\n", + " ret = ret.dtype.type(ret / rcount)\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Episode 12 done\n", + "Episode 13 done\n", + "Episode 14 done\n", + "Episode 15 done\n", + "Episode 16 done\n", + "Episode 17 done\n", + "Episode 18 done\n", + "Episode 19 done\n", + "Episode 19 done\n", + "Total steps: 13\n", + "Episode 20 done\n", + "Episode 21 done\n", + "Episode 22 done\n", + "Episode 23 done\n", + "Episode 23 done\n", + "Total steps: 13\n", + "Episode 24 done\n", + "Episode 25 done\n", + "Episode 26 done\n", + "Episode 27 done\n", + "Episode 28 done\n", + "Episode 29 done\n", + "Episode 29 done\n", + "Total steps: 13\n", + "Episode 30 done\n", + "Episode 31 done\n", + "Episode 31 done\n", + "Total steps: 13\n", + "Episode 32 done\n", + "Episode 32 done\n", + "Total steps: 13\n", + "Episode 33 done\n", + "Episode 34 done\n", + "Episode 34 done\n", + "Total steps: 13\n", + "Episode 35 done\n", + "Episode 35 done\n", + "Total steps: 13\n", + "Episode 36 done\n", + "Episode 37 done\n", + "Episode 38 done\n", + "Episode 39 done\n", + "Episode 40 done\n", + "Episode 40 done\n", + "Total steps: 13\n", + "Episode 41 done\n", + "Episode 42 done\n", + "Episode 43 done\n", + "Episode 44 done\n", + "Episode 45 done\n", + "Episode 46 done\n", + "Episode 47 done\n", + "Episode 48 done\n", + "Episode 49 done\n", + "Episode 50 done\n", + "Episode 51 done\n", + "Episode 52 done\n", + "Episode 53 done\n", + "Episode 53 done\n", + "Total steps: 13\n", + "Episode 54 done\n", + "Episode 55 done\n", + "Episode 56 done\n", + "Episode 57 done\n", + "Episode 58 done\n", + "Episode 59 done\n", + "Episode 60 done\n", + "Episode 60 done\n", + "Total steps: 1\n", + "Episode 61 done\n", + "Episode 62 done\n", + "Episode 63 done\n", + "Episode 64 done\n", + "Episode 65 done\n", + "Episode 66 done\n", + "Episode 67 done\n", + "Episode 67 done\n", + "Total steps: 13\n", + "Episode 68 done\n", + "Episode 69 done\n", + "Episode 69 done\n", + "Total steps: 13\n", + "Episode 70 done\n", + "Episode 71 done\n", + "Episode 72 done\n", + "Episode 73 done\n", + "Episode 74 done\n", + "Episode 75 done\n", + "Episode 76 done\n", + "Episode 77 done\n", + "Episode 77 done\n", + "Total steps: 1\n", + "Episode 78 done\n", + "Episode 79 done\n", + "Episode 80 done\n", + "Episode 81 done\n", + "Episode 82 done\n", + "Episode 83 done\n", + "Episode 84 done\n", + "Episode 84 done\n", + "Total steps: 13\n", + "Episode 85 done\n", + "Episode 86 done\n", + "Episode 87 done\n", + "Episode 88 done\n", + "Episode 89 done\n", + "Episode 90 done\n", + "Episode 91 done\n", + "Episode 91 done\n", + "Total steps: 13\n", + "Episode 92 done\n", + "Episode 92 done\n", + "Total steps: 13\n", + "Episode 93 done\n", + "Episode 94 done\n", + "Episode 95 done\n", + "Episode 96 done\n", + "Episode 97 done\n", + "Episode 98 done\n", + "Episode 99 done\n", + "Episode 100 done\n", + "Episode 101 done\n", + "Episode 102 done\n", + "Episode 103 done\n", + "Episode 104 done\n", + "Episode 105 done\n", + "Episode 106 done\n", + "Episode 107 done\n", + "Episode 107 done\n", + "Total steps: 13\n", + "Episode 108 done\n", + "Episode 109 done\n", + "Episode 109 done\n", + "Total steps: 13\n", + "Episode 110 done\n", + "Episode 111 done\n", + "Episode 112 done\n", + "Episode 113 done\n", + "Episode 114 done\n", + "Episode 115 done\n", + "Episode 116 done\n", + "Episode 116 done\n", + "Total steps: 13\n", + "Episode 117 done\n", + "Episode 118 done\n", + "Episode 119 done\n", + "Episode 120 done\n", + "Episode 121 done\n", + "Episode 122 done\n", + "Episode 123 done\n", + "Episode 124 done\n", + "Episode 125 done\n", + "Episode 126 done\n", + "Episode 126 done\n", + "Total steps: 13\n", + "Episode 127 done\n", + "Episode 128 done\n", + "Episode 129 done\n", + "Episode 129 done\n", + "Total steps: 13\n", + "Episode 130 done\n", + "Episode 130 done\n", + "Total steps: 13\n", + "Episode 131 done\n", + "Episode 131 done\n", + "Total steps: 13\n", + "Episode 132 done\n", + "Episode 133 done\n", + "Episode 134 done\n", + "Episode 135 done\n", + "Episode 136 done\n", + "Episode 136 done\n", + "Total steps: 13\n", + "Episode 137 done\n", + "Episode 138 done\n", + "Episode 139 done\n", + "Episode 140 done\n", + "Episode 141 done\n", + "Episode 142 done\n", + "Episode 143 done\n", + "Episode 144 done\n", + "Episode 145 done\n", + "Episode 146 done\n", + "Episode 146 done\n", + "Total steps: 13\n", + "Episode 147 done\n", + "Episode 148 done\n", + "Episode 148 done\n", + "Total steps: 13\n", + "Episode 149 done\n", + "Episode 150 done\n", + "Episode 151 done\n", + "Episode 152 done\n", + "Episode 153 done\n", + "Episode 154 done\n", + "Episode 155 done\n", + "Episode 156 done\n", + "Episode 157 done\n", + "Episode 158 done\n", + "Episode 159 done\n", + "Episode 160 done\n", + "Episode 161 done\n", + "Episode 162 done\n", + "Episode 163 done\n", + "Episode 163 done\n", + "Total steps: 1\n", + "Episode 164 done\n", + "Episode 165 done\n", + "Episode 166 done\n", + "Episode 167 done\n", + "Episode 168 done\n", + "Episode 169 done\n", + "Episode 170 done\n", + "Episode 171 done\n", + "Episode 172 done\n", + "Episode 173 done\n", + "Episode 174 done\n", + "Episode 174 done\n", + "Total steps: 13\n", + "Episode 175 done\n", + "Episode 176 done\n", + "Episode 177 done\n", + "Episode 178 done\n", + "Episode 178 done\n", + "Total steps: 13\n", + "Episode 179 done\n", + "Episode 180 done\n", + "Episode 181 done\n", + "Episode 182 done\n", + "Episode 183 done\n", + "Episode 184 done\n", + "Episode 184 done\n", + "Total steps: 13\n", + "Episode 185 done\n", + "Episode 186 done\n", + "Episode 186 done\n", + "Total steps: 13\n", + "Episode 187 done\n", + "Episode 187 done\n", + "Total steps: 13\n", + "Episode 188 done\n", + "Episode 188 done\n", + "Total steps: 13\n", + "Episode 189 done\n", + "Episode 190 done\n", + "Episode 191 done\n", + "Episode 191 done\n", + "Total steps: 13\n", + "Episode 192 done\n", + "Episode 193 done\n", + "Episode 194 done\n", + "Episode 195 done\n", + "Episode 196 done\n", + "Episode 196 done\n", + "Total steps: 13\n", + "Episode 197 done\n", + "Episode 198 done\n", + "Episode 199 done\n", + "Episode 199 done\n", + "Total steps: 13\n", + "Episode 200 done\n", + "Episode 201 done\n", + "Episode 202 done\n", + "Episode 203 done\n", + "Episode 204 done\n", + "Episode 205 done\n", + "Episode 205 done\n", + "Total steps: 13\n", + "Episode 206 done\n", + "Episode 206 done\n", + "Total steps: 13\n", + "Episode 207 done\n", + "Episode 208 done\n", + "Episode 209 done\n", + "Episode 210 done\n", + "Episode 211 done\n", + "Episode 212 done\n", + "Episode 212 done\n", + "Total steps: 13\n", + "Episode 213 done\n", + "Episode 213 done\n", + "Total steps: 13\n", + "Episode 214 done\n", + "Episode 215 done\n", + "Episode 215 done\n", + "Total steps: 13\n", + "Episode 216 done\n", + "Episode 216 done\n", + "Total steps: 13\n", + "Episode 217 done\n", + "Episode 218 done\n", + "Episode 219 done\n", + "Episode 220 done\n", + "Episode 220 done\n", + "Total steps: 13\n", + "Episode 221 done\n", + "Episode 221 done\n", + "Total steps: 13\n", + "Episode 222 done\n", + "Episode 222 done\n", + "Total steps: 13\n", + "Episode 223 done\n", + "Episode 224 done\n", + "Episode 225 done\n", + "Episode 225 done\n", + "Total steps: 13\n", + "Episode 226 done\n", + "Episode 227 done\n", + "Episode 228 done\n", + "Episode 229 done\n", + "Episode 230 done\n", + "Episode 230 done\n", + "Total steps: 13\n", + "Episode 231 done\n", + "Episode 232 done\n", + "Episode 233 done\n", + "Episode 234 done\n", + "Episode 235 done\n", + "Episode 236 done\n", + "Episode 236 done\n", + "Total steps: 13\n", + "Episode 237 done\n", + "Episode 238 done\n", + "Episode 239 done\n", + "Episode 240 done\n", + "Episode 240 done\n", + "Total steps: 13\n", + "Episode 241 done\n", + "Episode 241 done\n", + "Total steps: 13\n", + "Episode 242 done\n", + "Episode 243 done\n", + "Episode 244 done\n", + "Episode 245 done\n", + "Episode 246 done\n", + "Episode 247 done\n", + "Episode 248 done\n", + "Episode 249 done\n", + "Episode 250 done\n", + "Episode 251 done\n", + "Episode 252 done\n", + "Episode 252 done\n", + "Total steps: 13\n", + "Episode 253 done\n", + "Episode 253 done\n", + "Total steps: 13\n", + "Episode 254 done\n", + "Episode 254 done\n", + "Total steps: 13\n", + "Episode 255 done\n", + "Episode 256 done\n", + "Episode 257 done\n", + "Episode 258 done\n", + "Episode 259 done\n", + "Episode 260 done\n", + "Episode 261 done\n", + "Episode 262 done\n", + "Episode 263 done\n", + "Episode 264 done\n", + "Episode 265 done\n", + "Episode 266 done\n", + "Episode 267 done\n", + "Episode 267 done\n", + "Total steps: 13\n", + "Episode 268 done\n", + "Episode 269 done\n", + "Episode 270 done\n", + "Episode 271 done\n", + "Episode 272 done\n", + "Episode 273 done\n", + "Episode 273 done\n", + "Total steps: 13\n", + "Episode 274 done\n", + "Episode 275 done\n", + "Episode 276 done\n", + "Episode 276 done\n", + "Total steps: 13\n", + "Episode 277 done\n", + "Episode 277 done\n", + "Total steps: 1\n", + "Episode 278 done\n", + "Episode 279 done\n", + "Episode 280 done\n", + "Episode 281 done\n", + "Episode 282 done\n", + "Episode 283 done\n", + "Episode 284 done\n", + "Episode 284 done\n", + "Total steps: 13\n", + "Episode 285 done\n", + "Episode 286 done\n", + "Episode 287 done\n", + "Episode 288 done\n", + "Episode 289 done\n", + "Episode 290 done\n", + "Episode 291 done\n", + "Episode 291 done\n", + "Total steps: 13\n", + "Episode 292 done\n", + "Episode 293 done\n", + "Episode 294 done\n", + "Episode 295 done\n", + "Episode 295 done\n", + "Total steps: 13\n", + "Episode 296 done\n", + "Episode 297 done\n", + "Episode 298 done\n", + "Episode 299 done\n", + " seed step_collisions not_on_track yield_violations \\\n", + "0 135.0 0.0 8.0 0.0 \n", + "1 135.0 1.0 2.0 0.0 \n", + "2 135.0 0.0 3.0 0.0 \n", + "3 135.0 4.0 0.0 0.0 \n", + "4 135.0 0.0 4.0 0.0 \n", + ".. ... ... ... ... \n", + "295 135.0 0.0 2.0 0.0 \n", + "296 135.0 0.0 24.0 0.0 \n", + "297 135.0 0.0 18.0 0.0 \n", + "298 135.0 0.0 19.0 0.0 \n", + "299 135.0 1.0 7.0 0.0 \n", + "\n", + " unncessary_brake_violations efficient_crossing_violations \\\n", + "0 0.0 138.0 \n", + "1 0.0 48.0 \n", + "2 0.0 24.0 \n", + "3 0.0 96.0 \n", + "4 0.0 0.0 \n", + ".. ... ... \n", + "295 0.0 48.0 \n", + "296 0.0 456.0 \n", + "297 0.0 366.0 \n", + "298 0.0 342.0 \n", + "299 0.0 162.0 \n", + "\n", + " total_violations_cost \n", + "0 -923.0 \n", + "1 -363.0 \n", + "2 -112.0 \n", + "3 -1368.0 \n", + "4 -24.0 \n", + ".. ... \n", + "295 -356.0 \n", + "296 -9792.0 \n", + "297 -6196.0 \n", + "298 -5524.5 \n", + "299 -1372.0 \n", + "\n", + "[300 rows x 7 columns]\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkQAAAHCCAYAAAAO4dYCAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAABzVUlEQVR4nO3deVyN6f8/8Ncp7atopVSEMpERJg1ZmskyiEaWhrKOIfsyzKDsywdjX8eU0Nj3JRIxQ0gmGmWPGkq2ytaic//+8Ov+OloU51Q6r+fjcR7j3Nd1rvt9nXOa3t3XdV+XRBAEAURERERKTKW8AyAiIiIqb0yIiIiISOkxISIiIiKlx4SIiIiIlB4TIiIiIlJ6TIiIiIhI6TEhIiIiIqXHhIiIiIiUHhMiIiIiUnpMiIioTAUHB0MikeDu3bulfm1kZCQkEgkiIyPlHhf9n9atW6N169Yf9VqJRILAwMAKEw9RSTEhIqW1atUqSCQSNG/evLxDqXDmzJmDvXv3lncYpETi4+MRGBj4UYkykTxIuJcZKStXV1c8ePAAd+/exc2bN1GnTp3yDqnC0NXVxffff4/g4GC5t52Xl4fc3FxoaGhAIpGU6rVSqRQ5OTlQV1eHigr/nlOUnJwcAIC6unqpXyuRSBAQEFDqq0Q7d+5Ejx49cPLkyQJXgz4lHqKS4v9RSCklJibi7NmzWLx4MYyNjbFly5Yyj0EqlSIrK6vMzytvL1++LFV9VVVVaGpqljoZAgAVFRVoampW2mRIEAS8fv26vMOAurp6hUo+Klo8VDlVzv+rEH3Ali1bULVqVXTq1Anff/+9TEKUm5sLIyMj9O/fv8DrMjMzoampifHjx4vHsrOzERAQgDp16kBDQwOWlpaYOHEisrOzZV4rkUjg7++PLVu2oEGDBtDQ0EBYWBgAYOHChWjRogWqVasGLS0tNGnSBDt37ixw/tevX2PkyJGoXr069PT00KVLF9y/f7/QeRv379/HgAEDYGpqCg0NDTRo0AB//PHHB98biUSCly9fYuPGjZBIJJBIJPDz8wMABAYGQiKRID4+Hn369EHVqlXx9ddfAwCuXLkCPz8/2NraQlNTE2ZmZhgwYACePHki035hc4isra3x3Xff4e+//0azZs2gqakJW1tbhISEyLy2sDlErVu3xhdffIH4+Hi0adMG2traqFGjBhYsWFCgb/fu3UOXLl2go6MDExMTjBkzBkePHi3RvKT8vt+6dQt+fn4wNDSEgYEB+vfvj1evXsnUffPmDWbOnInatWtDQ0MD1tbW+OWXXwp8J/L7ffToUTg7O0NLSwtr164V+7l9+3ZMnz4dNWrUgJ6eHr7//ntkZGQgOzsbo0ePhomJCXR1ddG/f/8Cbb/P398furq6BWIFgN69e8PMzAx5eXnie/r+VZq0tDQMHDgQpqam0NTURKNGjbBx48Zizwm8fc+HDRuGevXqQUtLC9WqVUOPHj1kPv/g4GD06NEDANCmTRvxe5f/mXxsPHfv3oVEIsHChQuxbt068fNo2rQpoqOjZeqmpqaif//+qFmzJjQ0NGBubo6uXbtyCE+JVCnvAIjKw5YtW9C9e3eoq6ujd+/eWL16NaKjo9G0aVOoqamhW7du2L17N9auXSvzl+nevXuRnZ2NXr16AXh7ladLly74+++/MWTIENjb2yMuLg6//fYbbty4UWAezokTJ7B9+3b4+/ujevXqsLa2BgAsXboUXbp0gY+PD3JycrB161b06NEDBw8eRKdOncTX+/n5Yfv27ejbty+++uornDp1SqY838OHD/HVV1+JSZixsTGOHDmCgQMHIjMzE6NHjy7yvdm0aRMGDRqEZs2aYciQIQCA2rVry9Tp0aMH7OzsMGfOHOSPuoeHh+POnTvo378/zMzMcPXqVaxbtw5Xr17FuXPnPnhF6NatW/j+++8xcOBA+Pr64o8//oCfnx+aNGmCBg0aFPvaZ8+eoX379ujevTu8vb2xc+dO/Pzzz3B0dESHDh0AvL2S1bZtW6SkpGDUqFEwMzNDaGgoTp48WWzb7/P29oaNjQ3mzp2LS5cu4ffff4eJiQnmz58v1hk0aBA2btyI77//HuPGjcP58+cxd+5cJCQkYM+ePTLtXb9+Hb1798aPP/6IwYMHo169emLZ3LlzoaWlhUmTJuHWrVtYvnw51NTUoKKigmfPniEwMBDnzp1DcHAwbGxsMG3atCLj7tmzJ1auXIlDhw6JyQcAvHr1CgcOHICfnx9UVVULfe3r16/RunVr3Lp1C/7+/rCxscGOHTvg5+eH9PR0jBo1qsjzRkdH4+zZs+jVqxdq1qyJu3fvYvXq1WjdujXi4+Ohra2NVq1aYeTIkVi2bBl++eUX2NvbA4D430+NJzQ0FM+fP8ePP/4IiUSCBQsWoHv37rhz5w7U1NQAAF5eXrh69SpGjBgBa2trpKWlITw8HElJSeLPKVVyApGSuXjxogBACA8PFwRBEKRSqVCzZk1h1KhRYp2jR48KAIQDBw7IvLZjx46Cra2t+HzTpk2CioqK8Ndff8nUW7NmjQBAOHPmjHgMgKCioiJcvXq1QEyvXr2SeZ6TkyN88cUXQtu2bcVjMTExAgBh9OjRMnX9/PwEAEJAQIB4bODAgYK5ubnw+PFjmbq9evUSDAwMCpzvfTo6OoKvr2+B4wEBAQIAoXfv3h/sgyAIwp9//ikAEE6fPi0eCwoKEgAIiYmJ4rFatWoVqJeWliZoaGgI48aNE4+dPHlSACCcPHlSPObm5iYAEEJCQsRj2dnZgpmZmeDl5SUeW7RokQBA2Lt3r3js9evXQv369Qu0WZj8vg8YMEDmeLdu3YRq1aqJz2NjYwUAwqBBg2TqjR8/XgAgnDhxokC/w8LCZOrm9/OLL74QcnJyxOO9e/cWJBKJ0KFDB5n6Li4uQq1atYqNXyqVCjVq1JB5TwRBELZv317gvXdzcxPc3NzE50uWLBEACJs3bxaP5eTkCC4uLoKurq6QmZkpHn//u1jY9yIqKqrAZ7Zjx44iP4ePjScxMVEAIFSrVk14+vSpWHffvn0yP9/Pnj0TAAj/+9//CpyblAeHzEjpbNmyBaampmjTpg2At0NEPXv2xNatW8Uhg7Zt26J69erYtm2b+Lpnz54hPDwcPXv2FI/t2LED9vb2qF+/Ph4/fiw+2rZtCwAFrj64ubnBwcGhQExaWloy58nIyEDLli1x6dIl8Xj+8NqwYcNkXjtixAiZ54IgYNeuXejcuTMEQZCJy8PDAxkZGTLtfoyhQ4cW24esrCw8fvwYX331FQCU6HwODg5o2bKl+NzY2Bj16tXDnTt3PvhaXV1d/PDDD+JzdXV1NGvWTOa1YWFhqFGjBrp06SIe09TUxODBgz/Y/rve73vLli3x5MkTZGZmAgAOHz4MABg7dqxMvXHjxgEADh06JHPcxsYGHh4ehZ6rX79+4hUMAGjevDkEQcCAAQNk6jVv3hzJycl48+ZNkXFLJBL06NEDhw8fxosXL8Tj27ZtQ40aNcShz8IcPnwYZmZm6N27t3hMTU0NI0eOxIsXL3Dq1KkiX/vu9yI3NxdPnjxBnTp1YGho+NHfw9LG07NnT1StWlV8nv89y/9+aGlpQV1dHZGRkXj27NlHxUSfPyZEpFTy8vKwdetWtGnTBomJibh16xZu3bqF5s2b4+HDh4iIiAAAVKlSBV5eXti3b584N2P37t3Izc2VSYhu3ryJq1evwtjYWOZRt25dAG/nObzLxsam0LgOHjyIr776CpqamjAyMoKxsTFWr16NjIwMsc69e/egoqJSoI3374579OgR0tPTsW7dugJx5c+Lej+u0iqsH0+fPsWoUaNgamoKLS0tGBsbi/Xe7UdRrKysChyrWrVqiX5B1axZs8CQ3PuvvXfvHmrXrl2gXmnvLnw/zvxftPnnyv+c3m/XzMwMhoaGuHfvnszxor4ThZ3LwMAAAGBpaVnguFQq/eD73LNnT7x+/Rr79+8HALx48QKHDx9Gjx49ih3SvHfvHuzs7ApMZs8f0nq/T+96/fo1pk2bBktLS2hoaKB69eowNjZGenp6ib4X8ojnQ5+ZhoYG5s+fjyNHjsDU1BStWrXCggULkJqa+lHx0eeJc4hIqZw4cQIpKSnYunUrtm7dWqB8y5Yt+PbbbwEAvXr1wtq1a3HkyBF4enpi+/btqF+/Pho1aiTWl0qlcHR0xOLFiws93/u/uN79aznfX3/9hS5duqBVq1ZYtWoVzM3NoaamhqCgIISGhpa6j1KpFADwww8/wNfXt9A6DRs2LHW77yqsH97e3jh79iwmTJgAJycn6OrqQiqVon379mJMxSlq/opQgpVBPuW1pVXSc5X0LrrC3ssPnetj+/vVV1/B2toa27dvR58+fXDgwAG8fv1aJsmXtxEjRiAoKAijR4+Gi4sLDAwMIJFI0KtXrxJ9L+ShJO/X6NGj0blzZ+zduxdHjx7F1KlTMXfuXJw4cQKNGzcukzipfDEhIqWyZcsWmJiYYOXKlQXKdu/ejT179mDNmjXQ0tJCq1atYG5ujm3btuHrr7/GiRMn8Ouvv8q8pnbt2rh8+TLatWv3UbeRA8CuXbugqamJo0ePQkNDQzweFBQkU69WrVqQSqVITEyEnZ2dePzWrVsy9YyNjaGnp4e8vDy4u7t/VEyl7cuzZ88QERGB6dOny0zsvXnz5kedXxFq1aqF+Ph4CIIg07/33z95nEcqleLmzZsyk4IfPnyI9PR01KpVS67nKy1vb28sXboUmZmZ2LZtG6ytrcWhzaLUqlULV65cgVQqlbkqc+3aNbG8KDt37oSvry8WLVokHsvKykJ6erpMvdJ85z4lnuLUrl0b48aNw7hx43Dz5k04OTlh0aJF2Lx580e1R58XDpmR0nj9+jV2796N7777Dt9//32Bh7+/P54/fy4OJ6ioqOD777/HgQMHsGnTJrx586bAX9Le3t64f/8+1q9fX+j5SrJGj6qqKiQSiTh/CXh7u/D7d6jlzzNZtWqVzPHly5cXaM/Lywu7du3Cv//+W+B8jx49+mBMOjo6BX5hFSf/L/D3r1AsWbKkxG0omoeHB+7fvy9+vsDbX8yFfXafomPHjgAK9j3/KmJhdwWWpZ49eyI7OxsbN25EWFgYvL29P/iajh07IjU1VWZO3Zs3b7B8+XLo6urCzc2tyNeqqqoW+F4sX75c5vsOvP3OASjR9+5T4inMq1evCqwJVrt2bejp6X1wOQOqPHiFiJTG/v378fz5c5lJte/66quvxEUa8xOfnj17Yvny5QgICICjo2OB24D79u2L7du3Y+jQoTh58iRcXV2Rl5eHa9euYfv27eL6MsXp1KkTFi9ejPbt26NPnz5IS0vDypUrUadOHVy5ckWs16RJE3h5eWHJkiV48uSJeNv9jRs3AMj+hT1v3jycPHkSzZs3x+DBg+Hg4ICnT5/i0qVLOH78OJ4+fVpsTE2aNMHx48exePFiWFhYwMbGptgtTvT19cV5F7m5uahRowaOHTuGxMTEYs9Tln788UesWLECvXv3xqhRo2Bubo4tW7ZAU1MTQOmvihWlUaNG8PX1xbp165Ceng43NzdcuHABGzduhKenpziZv7x8+eWXqFOnDn799VdkZ2eXaLhsyJAhWLt2Lfz8/BATEwNra2vs3LkTZ86cwZIlS6Cnp1fka7/77jts2rQJBgYGcHBwQFRUFI4fP45q1arJ1HNycoKqqirmz5+PjIwMaGhooG3btjAxMZFrPIW5ceMG2rVrB29vbzg4OKBKlSrYs2cPHj58KC6xQZUfEyJSGvm//L755ptCy1VUVNCpUyds2bIFT548QbVq1dCiRQtYWloiOTm50F8cKioq2Lt3L3777TeEhIRgz5490NbWhq2tLUaNGiVOri5O27ZtsWHDBsybNw+jR4+GjY0N5s+fj7t378okRAAQEhICMzMz/Pnnn9izZw/c3d2xbds21KtXT/zFDgCmpqa4cOECZsyYgd27d2PVqlWoVq0aGjRoILNeTlEWL16MIUOGYMqUKXj9+jV8fX0/uOdbaGgoRowYgZUrV0IQBHz77bc4cuQILCwsPni+sqCrq4sTJ05gxIgRWLp0KXR1ddGvXz+0aNECXl5eMu/fp/r9999ha2uL4OBg7NmzB2ZmZpg8eTICAgLkdo5P0bNnT8yePRt16tTBl19++cH6WlpaiIyMxKRJk7Bx40ZkZmaiXr16CAoKEhftLMrSpUuhqqqKLVu2ICsrC66urjh+/HiBO+vMzMywZs0azJ07FwMHDkReXh5OnjxZaEL0KfEUxtLSEr1790ZERAQ2bdqEKlWqoH79+ti+fTu8vLxK3R59nriXGdFnLjY2Fo0bN8bmzZvh4+NT3uF8dpYsWYIxY8bgv//+Q40aNco7HCIqJ5xDRPQZKWyfqyVLlkBFRQWtWrUqh4g+L++/f1lZWVi7di3s7OyYDBEpOQ6ZEX1GFixYgJiYGLRp0wZVqlTBkSNHcOTIEQwZMqTALf5UUPfu3WFlZQUnJydkZGRg8+bNuHbtWrls7ktEFQuHzIg+I+Hh4Zg+fTri4+Px4sULWFlZoW/fvvj1119RpQr/vvmQJUuW4Pfff8fdu3eRl5cHBwcHTJw4UaHr8BDR54EJERERESk9ziEiIiIipceEiIiIiJQeJx2UgFQqxYMHD6Cnpye3xduIiIhIsQRBwPPnz2FhYVFgM+D3MSEqgQcPHvAOHiIios9UcnIyatasWWwdJkQlkL8MfHJyMvT19cs5GiIiIiqJzMxMWFpalmg7FyZEJZA/TKavr8+EiIiI6DNTkukunFRNRERESo8JERERESk9JkRERESk9DiHSI7y8vKQm5tb3mEQkRypqalBVVW1vMMgIgVjQiQHgiAgNTUV6enp5R0KESmAoaEhzMzMuA4ZUSXGhEgO8pMhExMTaGtr83+aRJWEIAh49eoV0tLSAADm5ublHBERKQoTok+Ul5cnJkPVqlUr73CISM60tLQAAGlpaTAxMeHwGVElxUnVnyh/zpC2tnY5R0JEipL/8805gkSVFxMiOeEwGVHlxZ9vosqPCREREREpPSZEVGn4+fnB09NTfN66dWuMHj1afG5tbY0lS5aUqK3S1CUios8fJ1UrkPWkQ2V6vrvzOsmlHT8/P6Snp2Pv3r1yaa+iiI6Oho6OjtzrEhHR548JESkNY2NjhdQlIqLPH4fMlNjOnTvh6OgILS0tVKtWDe7u7pgwYQI2btyIffv2QSKRQCKRIDIyEgCQnJwMb29vGBoawsjICF27dsXdu3fF9vKHrKZPnw5jY2Po6+tj6NChyMnJKVE8UqkUCxYsQJ06daChoQErKyvMnj1bLI+Li0Pbtm3FeIcMGYIXL16UuL/vDoMJgoDAwEBYWVlBQ0MDFhYWGDlyZKF1ASApKQldu3aFrq4u9PX14e3tjYcPH4rlgYGBcHJywqZNm2BtbQ0DAwP06tULz58/L/b9fvnyZYnjJyIixWFCpKRSUlLQu3dvDBgwAAkJCYiMjET37t0REBAAb29vtG/fHikpKUhJSUGLFi2Qm5sLDw8P6Onp4a+//sKZM2egq6uL9u3byyQ8ERERYnt//vkndu/ejenTp5copsmTJ2PevHmYOnUq4uPjERoaClNTUwDAy5cv4eHhgapVqyI6Oho7duzA8ePH4e/v/1H937VrF3777TesXbsWN2/exN69e+Ho6FhoXalUiq5du+Lp06c4deoUwsPDcefOHfTs2VOm3u3bt7F3714cPHgQBw8exKlTpzBv3jwARb/fgiB8VPxERCRfHDJTUikpKXjz5g26d++OWrVqAYCYEGhpaSE7OxtmZmZi/c2bN0MqleL3338Xb0EOCgqCoaEhIiMj8e233wIA1NXV8ccff0BbWxsNGjTAjBkzMGHCBMycORMqKkXn38+fP8fSpUuxYsUK+Pr6AgBq166Nr7/+GgAQGhqKrKwshISEiHN7VqxYgc6dO2P+/Pli4lRSSUlJMDMzg7u7O9TU1GBlZYVmzZoVWjciIgJxcXFITEyEpaUlACAkJAQNGjRAdHQ0mjZtCuBt4hQcHAw9PT0AQN++fREREYHZs2cX+36XRtq9zFK/pjRMaukrtH0iooqKV4iUVKNGjdCuXTs4OjqiR48eWL9+PZ49e1Zk/cuXL+PWrVvQ09ODrq4udHV1YWRkhKysLNy+fVum3XcXqXRxccGLFy+QnJxcbDwJCQnIzs5Gu3btiixv1KiRzERnV1dXSKVSXL9+vaTdFvXo0QOvX7+Gra0tBg8ejD179uDNmzdFntvS0lJMhgDAwcEBhoaGSEhIEI9ZW1uLyRDwdpuH/C0fSvt+ExFR2WJCpKRUVVURHh6OI0eOwMHBAcuXL0e9evWQmJhYaP0XL16gSZMmiI2NlXncuHEDffr0+eR48rdHKCuWlpa4fv06Vq1aBS0tLQwbNgytWrX6pJWI1dTUZJ5LJBJIpVIApX+/iYiobDEhUmISiQSurq6YPn06/vnnH6irq2PPnj1QV1dHXl6eTN0vv/wSN2/ehImJCerUqSPzMDAwEOtdvnwZr1+/Fp+fO3cOurq6MldXCmNnZwctLS1EREQUWm5vb4/Lly/LTEI+c+YMVFRUUK9evY/pPrS0tNC5c2csW7YMkZGRiIqKQlxcXKHnTk5OlrnKFR8fj/T0dDg4OJT4fEW930REVP7KNSE6ffo0OnfuDAsLC0gkkgLr3giCgGnTpsHc3BxaWlpwd3fHzZs3Zeo8ffoUPj4+0NfXh6GhIQYOHFjgzqMrV66gZcuW0NTUhKWlJRYsWKDorlV458+fx5w5c3Dx4kUkJSVh9+7dePToEezt7WFtbY0rV67g+vXrePz4MXJzc+Hj44Pq1auja9eu+Ouvv5CYmIjIyEiMHDkS//33n9huTk4OBg4ciPj4eBw+fBgBAQHw9/cvdv4QAGhqauLnn3/GxIkTERISgtu3b+PcuXPYsGEDAMDHxweamprw9fXFv//+i5MnT2LEiBHo27dvqecPAUBwcDA2bNiAf//9F3fu3MHmzZuhpaUlzu95l7u7OxwdHeHj44NLly7hwoUL6NevH9zc3ODs7Fyi8xX3fhMRUfkr14To5cuXaNSoEVauXFlo+YIFC7Bs2TKsWbMG58+fh46ODjw8PJCVlSXW8fHxwdWrVxEeHo6DBw/i9OnTGDJkiFiemZmJb7/9FrVq1UJMTAz+97//ITAwEOvWrVN4/yoyfX19nD59Gh07dkTdunUxZcoULFq0CB06dMDgwYNRr149ODs7w9jYGGfOnIG2tjZOnz4NKysrdO/eHfb29hg4cCCysrKgr/9/E3HbtWsHOzs7tGrVCj179kSXLl0QGBhYopimTp2KcePGYdq0abC3t0fPnj3FOTja2to4evQonj59iqZNm+L7779Hu3btsGLFio/qv6GhIdavXw9XV1c0bNgQx48fx4EDB1CtWrUCdSUSCfbt24eqVauiVatWcHd3h62tLbZt21bi8xX3fhMRUfmTCBXkvl+JRII9e/aIWy8IggALCwuMGzcO48ePBwBkZGTA1NQUwcHB6NWrFxISEuDg4IDo6GjxL/WwsDB07NgR//33HywsLLB69Wr8+uuvSE1Nhbq6OgBg0qRJ2Lt3L65du1ai2DIzM2FgYICMjAyZX/4AkJWVhcTERNjY2EBTU1NO78bnqbKucF2R8C6z8sGfc6LPU3G/v99XYecQJSYmIjU1Fe7u7uIxAwMDNG/eHFFRUQCAqKgoGBoaygxbuLu7Q0VFBefPnxfrtGrVSkyGAMDDwwPXr18v8i6f7OxsZGZmyjyIiIio8qqwCVFqaioAFJgfYmpqKpalpqbCxMREprxKlSowMjKSqVNYG++e431z586FgYGB+PjQhGD6sKSkJPF2/cIeSUlJ5R0iEREpMS7MWIjJkydj7Nix4vPMzEwmRSUQHBxcZJmFhQViY2OLLSciIiovFTYhyl8l+eHDhzA3NxePP3z4EE5OTmKd/Em3+d68eYOnT5+KrzczM5PZcyq/jXfP8T4NDQ1oaGjIpR/0VpUqVVCnTp3yDoOIiKhQFXbIzMbGBmZmZjLr0mRmZuL8+fNwcXEB8HYV5PT0dMTExIh1Tpw4AalUiubNm4t1Tp8+LbPgXnh4OOrVq4eqVauWUW+IiIioIivXhOjFixfiisfA24nUsbGxSEpKgkQiwejRozFr1izs378fcXFx6NevHywsLMQ70ezt7dG+fXsMHjwYFy5cwJkzZ+Dv749evXqJQzB9+vSBuro6Bg4ciKtXr2Lbtm1YunSpzJAYERERKbdyHTK7ePEi2rRpIz7PT1J8fX0RHByMiRMn4uXLlxgyZAjS09Px9ddfIywsTOa21y1btsDf3x/t2rWDiooKvLy8sGzZMrHcwMAAx44dw/Dhw9GkSRNUr14d06ZNk1mriIiIiJRbhVmHqCLjOkRUUXAdovLBn3Oiz1OlWIeIiIiIqKwwISL6SJGRkZBIJEhPTy/vUIiI6BNV2NvuK4VAgw/Xkev5Msr2dIGB2Lt3b7HrC5W11q1bw8nJCUuWLCnvUIiI6DPCK0SkdARBwJs3b8o7DCIiqkCYECmx1q1bY+TIkZg4cSKMjIxgZmYmszN9UlISunbtCl1dXejr68Pb21tc1DI4OBjTp0/H5cuXIZFIIJFIil2puiRtAm+vOjk5OWHTpk2wtraGgYEBevXqhefPn3+wbT8/P5w6dQpLly4VY7p79644tHXkyBE0adIEGhoa+Pvvv3H79m107doVpqam0NXVRdOmTXH8+HGZNrOzs/Hzzz/D0tISGhoaqFOnDjZs2FDo+V+9eoUOHTrA1dWVw2hERJ8ZJkRKbuPGjdDR0cH58+exYMECzJgxA+Hh4ZBKpejatSuePn2KU6dOITw8HHfu3EHPnj0BAD179sS4cePQoEEDpKSkICUlRSwryofazHf79m3s3bsXBw8exMGDB3Hq1CnMmzfvg31ZunQpXFxcMHjwYDGmd7dcmTRpEubNm4eEhAQ0bNgQL168QMeOHREREYF//vkH7du3R+fOnWX2VevXrx/+/PNPLFu2DAkJCVi7di10dXULnDs9PR3ffPMNpFIpwsPDYWho+MF4iYio4uAcIiXXsGFDBAQEAADs7OywYsUKcXXwuLg4JCYmiklFSEgIGjRogOjoaDRt2hS6urqoUqVKkVugvC8iIuKDbQJvE6fg4GDo6ekBAPr27YuIiAjMnj272PYNDAygrq4ObW3tQmOaMWMGvvnmG/G5kZERGjVqJD6fOXMm9uzZg/3798Pf3x83btzA9u3bER4eDnd3dwCAra1tgXZTU1PRs2dP2NnZITQ0FOrq6iV6P4iIqOLgFSIl17BhQ5nn5ubmSEtLQ0JCAiwtLWWusDg4OMDQ0BAJCQkfda6StmltbS0mQ+/G9KmcnZ1lnr948QLjx4+Hvb09DA0Noauri4SEBPEKUWxsLFRVVeHm5lZsu9988w3q1KmDbdu2MRkiIvpMMSFScmpqajLPJRIJpFJpOUXzlqJi0tHRkXk+fvx47NmzB3PmzMFff/2F2NhYODo6IicnBwCgpaVVonY7deqE06dPIz4+/pNjJCKi8sGEiAplb2+P5ORkJCcni8fi4+ORnp4OBwcHAIC6ujry8vLk2uanKk1MZ86cgZ+fH7p16wZHR0eYmZnh7t27YrmjoyOkUilOnTpVbDvz5s2Dr68v2rVrx6SIiOgzxYSICuXu7g5HR0f4+Pjg0qVLuHDhAvr16wc3Nzdx6Mna2lrckPfx48fIzs7+5DY/lbW1Nc6fP4+7d+/i8ePHxV5ZsrOzw+7duxEbG4vLly+jT58+MvWtra3h6+uLAQMGYO/evUhMTERkZCS2b99eoK2FCxfCx8cHbdu2xbVr1+TSFyIiKjtMiKhQEokE+/btQ9WqVdGqVSu4u7vD1tYW27ZtE+t4eXmhffv2aNOmDYyNjfHnn39+cpufavz48VBVVYWDgwOMjY1l7hh73+LFi1G1alW0aNECnTt3hoeHB7788kuZOqtXr8b333+PYcOGoX79+hg8eDBevnxZaHu//fYbvL290bZtW9y4cUNufSIiIsXj5q4lwM1dqaLg5q7lgz/nRJ8nbu5KREREVApMiEhutmzZAl1d3UIfDRo0+OT2k5KSimxfV1e32OExIiKi4nBhRpKbLl26oHnz5oWWvX8r/cewsLAodiNZCwuLTz4HEREpJyZEJDd6enoyCyrKW5UqVVCnTh2FtU9ERMqLQ2ZERESk9JgQERERkdJjQkRERERKjwkRERERKT0mRERERKT0mBBRoQIDA+Hk5FTi+nfv3oVEIin2tvjIyEhIJBKkp6d/cnwlOd/7WrdujdGjR3/yueXVDhERVRy87V6BHDc6lun54nzj5NbW+PHjMWLECLm1J2+WlpZISUlB9erVFXaOyMhItGnTBs+ePYOhoaF4fPfu3XJZV4mIiCoOJkRUqPzVnysqVVVVmJmZlcu5jYyMyuW8RESkOBwyU1IhISGoVq0asrOzZY57enqib9++hQ6Z/f7777C3t4empibq16+PVatWFXuOw4cPo27dutDS0kKbNm1w9+7dEsWWmZkJLS0tHDlyROb4nj17oKenh1evXhU6ZHbq1Ck0a9YMGhoaMDc3x6RJk/DmzZsiz7Np0yY4OztDT08PZmZm6NOnD9LS0gC8HZJr06YNAKBq1aqQSCTw8/MDUHDI7NmzZ+jXrx+qVq0KbW1tdOjQATdv3hTLg4ODYWhoiKNHj8Le3h66urpo3749UlJSxDqRkZFo1qwZdHR0YGhoCFdXV9y7d69E7xcREX06JkRKqkePHsjLy8P+/fvFY2lpaTh06BAGDBhQoP6WLVswbdo0zJ49GwkJCZgzZw6mTp2KjRs3Ftp+cnIyunfvjs6dOyM2NhaDBg3CpEmTShSbvr4+vvvuO4SGhhaIwdPTE9ra2gVec//+fXTs2BFNmzbF5cuXsXr1amzYsAGzZs0q8jy5ubmYOXMmLl++jL179+Lu3bti0mNpaYldu3YBAK5fv46UlBQsXbq00Hb8/Pxw8eJF7N+/H1FRURAEAR07dkRubq5Y59WrV1i4cCE2bdqE06dPIykpCePHjwcAvHnzBp6ennBzc8OVK1cQFRWFIUOGQCKRlOj9IiKiT8chMyWlpaWFPn36ICgoCD169AAAbN68GVZWVmjdujVOnTolUz8gIACLFi1C9+7dAQA2NjaIj4/H2rVr4evrW6D91atXo3bt2li0aBEAoF69eoiLi8P8+fNLFJ+Pjw/69u2LV69eQVtbG5mZmTh06BD27NlTaP1Vq1bB0tISK1asgEQiQf369fHgwQP8/PPPmDZtGlRUCub+7yZ+tra2WLZsGZo2bYoXL15AV1dXHBozMTGRmUP0rps3b2L//v04c+YMWrRoAeBt4mZpaYm9e/eK721ubi7WrFmD2rVrAwD8/f0xY8YMAG+viGVkZOC7774Ty+3t7Uv0PhERkXzwCpESGzx4MI4dO4b79+8DeDu04+fnV+DKxMuXL3H79m0MHDhQZnf5WbNm4fbt24W2nZCQUGCjVxcXlxLH1rFjR6ipqYlXsHbt2gV9fX24u7sXeT4XFxeZ2F1dXfHixQv8999/hb4mJiYGnTt3hpWVFfT09ODm5gYASEpKKnGcCQkJqFKlikxfq1Wrhnr16iEhIUE8pq2tLSY7AGBubi4OzxkZGcHPzw8eHh7o3Lkzli5dKjOcRkREiseESIk1btwYjRo1QkhICGJiYnD16lVxyOhdL168AACsX78esbGx4uPff//FuXPnFBKburo6vv/+e3HYLDQ0FD179kSVKvK5qPny5Ut4eHhAX18fW7ZsQXR0tHj1KScnRy7neNf7d6VJJBIIgiA+DwoKQlRUFFq0aIFt27ahbt26CntviYioICZESm7QoEEIDg5GUFAQ3N3dYWlpWaCOqakpLCwscOfOHdSpU0fmYWNjU2i79vb2uHDhgsyx0v6C9/HxQVhYGK5evYoTJ07Ax8enyLr29vbi/J18Z86cgZ6eHmrWrFmg/rVr1/DkyRPMmzcPLVu2RP369cUrNvnU1dUBAHl5ecWe982bNzh//rx47MmTJ7h+/TocHBxK3FfgbYI6efJknD17Fl988UWBOVRERKQ4TIiUXJ8+ffDff/9h/fr1hU6mzjd9+nTMnTsXy5Ytw40bNxAXF4egoCAsXry40PpDhw7FzZs3MWHCBFy/fh2hoaEIDg4uVWytWrWCmZkZfHx8YGNjU2AI7l3Dhg1DcnIyRowYgWvXrmHfvn0ICAjA2LFjC50/ZGVlBXV1dSxfvhx37tzB/v37MXPmTJk6tWrVgkQiwcGDB/Ho0SPxStm77Ozs0LVrVwwePBh///03Ll++jB9++AE1atRA165dS9TPxMRETJ48GVFRUbh37x6OHTuGmzdvch4REVEZYkKk5AwMDODl5QVdXV14enoWWW/QoEH4/fffERQUBEdHR7i5uSE4OLjIK0RWVlbYtWsX9u7di0aNGmHNmjWYM2dOqWKTSCTo3bs3Ll++XOzVIQCoUaMGDh8+jAsXLqBRo0YYOnQoBg4ciClTphRa39jYGMHBwdixYwccHBwwb948LFy4sECb06dPx6RJk2Bqagp/f/9C2woKCkKTJk3w3XffwcXFBYIg4PDhwyVevFFbWxvXrl2Dl5cX6tatiyFDhmD48OH48ccfS/R6IiL6dBLh3TEGKlRmZiYMDAyQkZEBfX19mbKsrCwkJibCxsYGmpqa5RThp2nXrh0aNGiAZcuWlXco9AFp9zIV2r5JLf0PV1JCleHnnEgZFff7+3287V6JPXv2DJGRkYiMjPzgIotERESVGYfMlFjjxo3h5+eH+fPno169emV67g4dOsjcwv/uo7RDa0RERJ+KV4iUWEm30lCE33//Ha9fvy60jHuFERFRWWNCROWiRo0a5R0CERGRiENmREREpPSYEBEREZHSY0JERERESo8JERERESk9JkRERESk9JgQEclRYGAgnJycFNJ2ZGQkTK0NkJGRrpD2P8Ta2hpLliwpcX15vReKfE+JiPLxtnsFSqhftptz2l9LKNPzkXKJjo6Gjo6OQs8hkUiwZ88emX31xo8fjxEjRij0vEQkHyuHnlD4OYavaauQdpkQEf1/OTk5UFdXV5rzlpaxsXG5nDd/BXMiIkXikJkSK2wIxMnJCYGBgQDe/rX++++/o1u3btDW1oadnR32798v1o2MjIREIkFERAScnZ2hra2NFi1a4Pr16zJtHjhwAE2bNoWmpiaqV6+Obt26iWXZ2dkYP348atSoAR0dHTRv3hyRkZFi+b1799C5c2dUrVoVOjo6aNCgAQ4fPgzg7V5sPj4+MDY2hpaWFuzs7BAUFCS+9ueff0bdunWhra0NW1tbTJ06Fbm5uWJ5/lDM77//Lm7aGRISgmrVqiE7O1umD56enujbt2+J39u1a9fC0tIS2tra8Pb2RkZGhljm5+cHT09PzJ49GxYWFuK2KZs2bYKzszP09PRgZmaGPn36IC0trchzvHr9Cr19vfCd17fiMNrmrRvxdbumsKprAte2zgjatL5E8Xbq/g1mzp0mc+zRo0dQU1PD6dOnART8viQlJaFr167Q1dWFvr4+vL298fDhwyLPER0djW+++QbVq1eHgYEB3NzccOnSJbHc2toaANCtWzdIJBLx+ftDZlKpFDNmzEDNmjWhoaEBJycnhIWFieV3796FRCLB7t270aZNG2hra6NRo0aIiooS6xT3vSIi5cSEiIo1ffp0eHt748qVK+jYsSN8fHzw9OlTmTq//vorFi1ahIsXL6JKlSoYMGCAWHbo0CF069YNHTt2xD///IOIiAg0a9ZMLPf390dUVBS2bt2KK1euoEePHmjfvj1u3rwJABg+fDiys7Nx+vRpxMXFYf78+eLVgqlTpyI+Ph5HjhxBQkICVq9ejerVq4tt6+npITg4GPHx8Vi6dCnWr1+P3377TSb2W7duYdeuXdi9ezdiY2PRo0cP5OXlySR+aWlpOHTokEy/inPr1i1s374dBw4cQFhYGP755x8MGzZMpk5ERASuX7+O8PBwHDx4EACQm5uLmTNn4vLly9i7dy/u3r0LPz+/Qs+RkZEO7x88IZVKsX3zXhgYGGLn3u1YsHgOJk+Yir8iLuCXidMwf9FsbNsZ+sGYvTy9sffgbgiCIB7btm0bLCws0LJlywL1pVIpunbtiqdPn+LUqVMIDw/HnTt30LNnzyLP8fz5c/j6+uLvv//GuXPnYGdnh44dO+L58+cA3iZMABAUFISUlBTx+fuWLl2KRYsWYeHChbhy5Qo8PDzQpUsX8TuT79dff8X48eMRGxuLunXronfv3njz5g2A4r9XRKScOGRGxfLz80Pv3r0BAHPmzMGyZctw4cIFtG/fXqwze/ZsuLm5AQAmTZqETp06ISsrC5qampg9ezZ69eqF6dOni/UbNWoE4O0VhqCgICQlJcHCwgLA2/kiYWFhCAoKwpw5c5CUlAQvLy84OjoCAGxtbcV2kpKS0LhxYzg7OwP4vysM+aZMmSL+29raGuPHj8fWrVsxceJE8XhOTg5CQkJkhoP69OmDoKAg9OjRAwCwefNmWFlZoXXr1iV6z7KyshASEiJuT7J8+XJ06tQJixYtgpmZGQBAR0cHv//+u8xQ2bsJl62tLZYtW4amTZvixYsXMr+s0x6lYYh/f9ja2GL10g1iG//7bQ4Cf52NTu27AABqWVrj+s3rCAkNQs/v+xQbc5dO3TB1xiT8/fffYgIUGhqK3r17QyKRFKgfERGBuLg4JCYmwtLSEgAQEhKCBg0aIDo6Gk2bNi3wmrZtZcf9161bB0NDQ5w6dQrfffed+BkYGhqK71NhFi5ciJ9//hm9evUCAMyfPx8nT57EkiVLsHLlSrHe+PHj0alTJwBvE/sGDRrg1q1bqF+/frHfKyJSTrxCRMVq2LCh+G8dHR3o6+sXGMZ5t465uTkAiHViY2PRrl27QtuOi4tDXl4e6tatK7Pb/alTp3D79m0AwMiRIzFr1iy4uroiICAAV65cEV//008/YevWrXBycsLEiRNx9uxZmfa3bdsGV1dXmJmZQVdXF1OmTEFSUpJMnVq1ahWYGzN48GAcO3YM9+/fBwAEBwfDz8+v0MSgMFZWVjJ7tbm4uEAqlcoMJTo6OhaYNxQTE4POnTvDysoKenp6YpL5fszefT1hY22DdSuCxTZevnqJu/cSMfZnf9g4WIiPJcv/h7tJiR+MuXq16mjdsi22bNkCAEhMTERUVBR8fHwKrZ+QkABLS0sxGQIABwcHGBoaIiGh8Mn9Dx8+xODBg2FnZwcDAwPo6+vjxYsXBfpXnMzMTDx48ACurq4yx11dXQuct7jvZXHfKyJSTkyIlJiKiorMEAkAmTk2AKCmpibzXCKRQCqVFlknP2nIr6OlpVXk+V+8eAFVVVXExMQgNjZWfCQkJGDp0qUAgEGDBuHOnTvo27cv4uLi4OzsjOXLlwMAOnTogHv37mHMmDF48OAB2rVrh/HjxwOA+Mu8Y8eOOHjwIP755x/8+uuvyMnJkYmhsLumGjdujEaNGiEkJAQxMTG4evVqkUNXH+v98758+RIeHh7Q19fHli1bEB0djT179gBAgZjd23yLcxfO4vrNazKvB4CF85bhxOG/xMepY1E4vOd4iWLq7umNnTt3Ijc3F6GhoXB0dBSvoMiDr68vYmNjsXTpUpw9exaxsbGoVq1agf7JS3Hfy+K+V0SknDhkpsSMjY2RkpIiPs/MzERi4oevJpRGw4YNERERgf79+xcoa9y4MfLy8pCWllboPJV8lpaWGDp0KIYOHYrJkydj/fr14m3YxsbG8PX1ha+vL1q2bIkJEyZg4cKFOHv2LGrVqoVff/1VbOfevXsljnvQoEFYsmQJ7t+/D3d3d5krIR+SlJSEBw8eiMOA586dg4qKijh5ujDXrl3DkydPMG/ePPFcFy9eLLTulJ8DoaOjg+99umDP1kOoZ1cfJsYmMDM1R1LSXXzv6V3iWN/V/puOmPDLKISFhSE0NBT9+vUrsq69vT2Sk5ORnJwsxhsfH4/09HQ4ODgU+pozZ85g1apV6NixIwAgOTkZjx8/lqmjpqaGvLy8Is+rr68PCwsLnDlzRryClt/2u3PTSqK47xV9Xj7nW72p4mBCpMTatm2L4OBgdO7cGYaGhpg2bRpUVVXleo6AgAC0a9cOtWvXRq9evfDmzRscPnxYvAPMx8cH/fr1w6JFi9C4cWM8evQIERERaNiwITp16oTRo0ejQ4cOqFu3Lp49e4aTJ0/C3v7t+k7Tpk1DkyZN0KBBA2RnZ+PgwYNimZ2dHZKSkrB161Y0bdoUhw4dEq+4lESfPn0wfvx4rF+/HiEhIaXqs6amJnx9fbFw4UJkZmZi5MiR8Pb2LnZejJWVFdTV1bF8+XIMHToU//77L2bOnFlk/cBfZyMvTwqv3p2xZ+sh2NWpiwljJmNK4M/Q09NHWzd3ZOdk4/KVf5CRmY6hg/w/GLeOtg48PT0xdepUJCQkiHPHCuPu7g5HR0f4+PhgyZIlePPmDYYNGwY3NzdxTtf77OzsxDvpMjMzMWHChAJXEK2trREREQFXV1doaGigatWqBdqZMGECAgICULt2bTg5OSEoKAixsbHicF9JFPe9IiLlxCEzJTZ58mS4ubnhu+++Q6dOneDp6YnatWvL9RytW7fGjh07sH//fjg5OaFt27a4cOGCWB4UFIR+/fph3LhxqFevHjw9PREdHQ0rKysAQF5eHoYPHw57e3u0b98edevWxapVqwAA6urqmDx5Mho2bIhWrVpBVVUVW7duBQB06dIFY8aMgb+/P5ycnHD27FlMnTq1xHEbGBjAy8sLurq6MosElkSdOnXQvXt3dOzYEd9++y0aNmwoxlwUY2NjBAcHY8eOHXBwcMC8efOwcOHCYl8zc9pcdPmuG7z6dMbtO7fwQy9fLJq/HFt3bEbr9i7o1rMjtu0MhVXNWiWO3cfHB5cvX0bLli3Fz6AwEokE+/btQ9WqVdGqVSu4u7vD1tYW27ZtK/I1GzZswLNnz/Dll1+ib9++GDlyJExMTGTqLFq0COHh4bC0tETjxo0LbWfkyJEYO3Ysxo0bB0dHR4SFhWH//v2ws7MrcT+L+14RkXKSCO9PIqECMjMzYWBggIyMDOjr68uUZWVlITExUVzHhiqPdu3aoUGDBli2bFl5hyJKu5ep0PZNaul/uJIS4s95xcYhs4qjon0Wxf3+fh+HzIje8+zZM0RGRiIyMpJXDYiIlAQTIqL3NG7cGM+ePcP8+fMLTIRu0KBBkZOz165dW+Rt6uVtycqFWLpycaFlXzV1wZ8bd5VxREREFUuFTojy8vIQGBiIzZs3IzU1FRYWFvDz88OUKVPE22gFQUBAQADWr1+P9PR0uLq6YvXq1TLzCZ4+fYoRI0bgwIEDUFFRgZeXF5YuXcqVaalQd+/eLbLs8OHDBZYmyGdqaqqgiD6dr88AdO3UrdAyTc2il0YgIlIWFTohmj9/PlavXo2NGzeiQYMGuHjxIvr37w8DAwOMHDkSALBgwQIsW7YMGzduhI2NDaZOnQoPDw/Ex8eLY/0+Pj5ISUlBeHg4cnNz0b9/fwwZMgShoR/e0oDoXbVqlXyCckVS1dAIVQ2NyjsMIqIKq0InRGfPnkXXrl3F5fetra3x559/incpCYKAJUuWYMqUKejatSuAt9sHmJqaYu/evejVqxcSEhIQFhaG6Oho8Xbg5cuXo2PHjli4cKG4VgwREREprwp9232LFi0QERGBGzduAAAuX76Mv//+Gx06dADwdnuB1NRUuLu7i68xMDBA8+bNxZ2to6KiYGhoKLM2iru7O1RUVHD+/Pky7A0RERFVVBX6CtGkSZOQmZmJ+vXrQ1VVFXl5eZg9e7Y4cTU1NRVAwbkbpqamYllqamqBtU6qVKkCIyMjsc77srOzkZ2dLT7PzFTsrc5ERERUvir0FaLt27djy5YtCA0NxaVLl7Bx40YsXLgQGzduVOh5586dCwMDA/FRmm0biIiI6PNToROiCRMmYNKkSejVqxccHR3Rt29fjBkzBnPnzgUAcSuEhw8fyrzu4cOHYpmZmVmB3dnfvHmDp0+fFrmVwuTJk5GRkSE+kpOT5d01IiIiqkAqdEL06tUrqKjIhqiqqiruWG1jYwMzMzNERESI5ZmZmTh//jxcXFwAAC4uLkhPT0dMTIxY58SJE5BKpWjevHmh59XQ0IC+vr7MQ1mdOXMGjo6OUFNTE7eweP9YZGQkJBIJ0tPTS9Rm69atMXr0aIXF/Dkq7XtYFgIDA+Hk5FTi+nfv3oVEIkFsbOwnnVde7RARlUaFnkPUuXNnzJ49G1ZWVmjQoAH++ecfLF68GAMGDADwdj+l0aNHY9asWbCzsxNvu7ewsBB/eefvVTR48GCsWbMGubm58Pf3R69evRR+h1lZLGH+LkUsLT927Fg4OTnhyJEj4rpN7x/T1tZGSkoKDAwMStTm7t27oaamJtc4/fz8kJ6ejr1798q13bLSokWLUr2HZWH8+PEK3/29sM/N0tISKSkpqF69ukLPTUT0rgqdEC1fvhxTp07FsGHDkJaWBgsLC/z444+YNm2aWGfixIl4+fIlhgwZgvT0dHz99dcICwuT2W9oy5Yt8Pf3R7t27cSFGSvS/lQV2e3btzF06FDUrFmz2GPF7eT+PiOjz289nJycHKirqyusfXV19VK9h2VBV1e3XBYvVVVVrXDvBRFVfhV6yExPTw9LlizBvXv38Pr1a9y+fRuzZs2S+cUkkUgwY8YMpKamIisrC8ePH0fdunVl2jEyMkJoaCieP3+OjIwM/PHHH1yl+v+TSqWYO3cubGxsoKWlhUaNGmHnzp3isMWTJ08wYMAASCQSBAcHF3qssOGeM2fOoHXr1tDW1kbVqlXh4eGBZ8+eASg4ZJadnY3x48ejRo0a0NHRQfPmzREZGSmWBwcHw9DQEEePHoW9vT10dXXRvn17pKSkAHg7tLNx40bs27cPEokEEolE5vVF+e+//9C7d28YGRlBR0cHzs7O4lIM+cNFv//+u8yGnklJSejatSt0dXWhr68Pb29vmTlsly9fRps2baCnpwd9fX00adIEFy9eBADcu3cPnTt3RtWqVaGjo4MGDRrg8OHDAAoOmRXV54dp/3dn5Js3b/BL4ETYOVqhvpM1Zs6dhhFjh8J3cJ8P9j0kNAgNm9UTh5/zde3aVbwC+/6QmVQqxYwZM1CzZk1oaGjAyckJYWFhRZ4jLy8PAwcOFL9b9erVw9KlS8Xyoj63wobMTp06hWbNmkFDQwPm5uaYNGkS3rx5I5a3bt0aI0eOxMSJE2FkZAQzMzMEBgaK5YIgIDAwEFZWVtDQ0ICFhYW4uCsREVDBEyJSvLlz5yIkJARr1qzB1atXMWbMGPzwww+4d+8eUlJSoK+vjyVLliAlJQU9evQocKxnz54F2oyNjUW7du3g4OCAqKgo/P333+jcuTPy8vIKjcHf3x9RUVHYunUrrly5gh49eqB9+/a4efOmWOfVq1dYuHAhNm3ahNOnTyMpKQnjx48H8HZox9vbW0ySUlJS0KJFi2L7/eLFC7i5ueH+/fvYv38/Ll++jIkTJ8okCLdu3cKuXbuwe/duxMbGQiqVomvXrnj69ClOnTqF8PBw3LlzR+Y98PHxQc2aNREdHY2YmBhMmjRJHB4cPnw4srOzcfr0acTFxWH+/PnFJuaF9Tlw9hSxfPma37B773Ys/d9KHNh5DM9fPMeR8EPF9jtfl06eeJb+FH9HnRaPPUt/irCwsCL3Y1u6dCkWLVqEhQsX4sqVK/Dw8ECXLl1kPqd3SaVS1KxZEzt27EB8fDymTZuGX375Bdu3bwdQ8s/t/v376NixI5o2bYrLly9j9erV2LBhA2bNmiVTb+PGjdDR0cH58+exYMECzJgxA+Hh4QCAXbt24bfffsPatWtx8+ZN7N27F46OjiV6r4hIOVToITNSrOzsbMyZMwfHjx8XJ6Hb2tri77//xtq1axEaGgqJRAIDAwNxCENHR6fAsfctWLAAzs7OMjvFN2jQoNC6SUlJCAoKQlJSkjina/z48QgLC0NQUBDmzJkDAMjNzcWaNWtQu3ZtAG+TqBkzZgB4O7SjpaWF7OzsEg+1hIaG4tGjR4iOjhaH8OrUqSNTJycnByEhITA2NgYAhIeHIy4uDomJieJSDCEhIWjQoAGio6PRtGlTJCUlYcKECahfvz4AyOypl5SUBC8vL/EXsa2tbbExFtbnwIDpYvmG4HUYOWwsOrbvDACYO2MhIiLDS9R/Q4OqaOv2DXbv24FWrq0BAAcO70P16tXRpk2bQl+zcOFC/Pzzz+jVqxeAt1vrnDx5EkuWLMHKlSsL1FdTU8P06f8Xr42NDaKiorB9+3Z4e3uX+HNbtWoVLC0tsWLFCkgkEtSvXx8PHjzAzz//jGnTpok3XjRs2BABAQEA3r7vK1asQEREBL755hskJSXBzMwM7u7uUFNTg5WVFZo1a1ai94qIlAOvECmxW7du4dWrV/jmm2/E+SK6uroICQnB7du3P7rd/CtEJREXF4e8vDzUrVtXJoZTp07JxKCtrS0mBgBgbm5eYDmF0sbYuHHjYucz1apVS0yGACAhIQGWlpYy61I5ODjA0NAQCQkJAN5OOB80aBDc3d0xb948mT6MHDkSs2bNgqurKwICAnDlypViYyysz4+fPAIAZGZm4NHjNDRu1EQsV1VVRcMvGpXwHQC8PHvg0JED4iKku/fuQK9evQrc2fn2fJl48OABXF1dZY67urqKfS/MypUr0aRJExgbG0NXVxfr1q1DUlJSiWME3r7vLi4u4obO+ed98eIF/vvvP/FYw4YNZV737nekR48eeP36NWxtbTF48GDs2bNHZsiNiIgJkRJ78eIFAODQoUOIjY0VH/Hx8di5c+dHt6ulVfLd01+8eAFVVVXExMTIxJCQkCAz3+T9u9IkEgkEQVBojDo6OqVuNzAwEFevXkWnTp1w4sQJODg4YM+ePQCAQYMG4c6dO+jbty/i4uLg7OyM5cuXF9mWvPv8vm/bdYAAAcdPHsX9B//hXPTZIofLPsbWrVsxfvx4DBw4EMeOHUNsbCz69++PnJwcuZ3jXYW9X/lDoJaWlrh+/TpWrVoFLS0tDBs2DK1atUJubq5CYiGizw8TIiXm4OAADQ0NJCUloU6dOjKPT1mdu2HDhjJrQxWncePGyMvLQ1paWoEYSnOnkbq6epFzlIqKMTY2Fk+fPi3xa+zt7ZGcnCyzUGd8fDzS09Ph4OAgHqtbty7GjBmDY8eOoXv37ggKChLLLC0tMXToUOzevRvjxo3D+vXrS3z+d+nrG8C4uglir1wSj+Xl5SHu3+KvOr1LU1MTnTw6Y9fe7dizfyfq2Nrhyy+/LOJ8+rCwsMCZM2dkjp85c0am7++XtWjRAsOGDUPjxo1Rp06dAlceS/K52dvbIyoqSiYZPHPmDPT09GTudPwQLS0tdO7cGcuWLUNkZCSioqIQFxdX4tcTUeXGhEiJ6enpYfz48RgzZgw2btyI27dv49KlS1i+fPknbY8yefJkREdHY9iwYbhy5QquXbuG1atX4/HjxwXq1q1bFz4+PujXrx92796NxMREXLhwAXPnzsWhQyWbIAwA1tbWuHLlCq5fv47Hjx9/8C//3r17w8zMDJ6enjhz5gzu3LmDXbt2iZsCF8bd3R2Ojo7w8fHBpUuXcOHCBfTr1w9ubm5wdnbG69ev4e/vj8jISNy7dw9nzpxBdHQ07O3tAQCjR4/G0aNHkZiYiEuXLuHkyZNi2ccY6DcEy1YtxpFjh3Dr9k1Mmf4z0jPTZYaWPqS7Zw8cP3EMf27fjO6e3sXWnTBhAubPn49t27bh+vXrmDRpEmJjYzFq1KhC69vZ2eHixYs4evQobty4galTpyI6OlqmTkk+t2HDhiE5ORkjRozAtWvXsG/fPgQEBGDs2LGFDu8VJjg4GBs2bMC///6LO3fuYPPmzdDS0kKtWrVK9HoiqvyYECm5mTNnYurUqZg7d664iOWhQ4dgY2Pz0W3WrVsXx44dw+XLl9GsWTO4uLhg3759qFKl8Dn8QUFB6NevH8aNG4d69erB09MT0dHRsLKyKvE5Bw8ejHr16sHZ2RnGxsYFrmS8T11dHceOHYOJiQk6duwIR0dHzJs3D6qqqkW+RiKRYN++fahatSpatWoFd3d32NraYtu2bQDezuF58uQJ+vXrh7p168Lb2xsdOnQQJxbn5eVh+PDh4vtct25dmYnnpTVi6Bh4dvkeI8YNRafu7tDW1kGbVm2hoaFR4jZatnCDoWFV3LpzE927fl9s3ZEjR2Ls2LEYN24cHB0dERYWhv3798tMHH/Xjz/+iO7du6Nnz55o3rw5njx5gmHDhsnUKcnnVqNGDRw+fBgXLlxAo0aNMHToUAwcOBBTpkwpULcohoaGWL9+PVxdXdGwYUMcP34cBw4cQLVq1UrcBhFVbhJBnpMSKqnMzEwYGBggIyOjwDYeWVlZSExMlFmrhkhR0u5lFlkmlUrxtXtTdOnUDZPGlTxZeJdJLeXdpqY4/Dmv2MpiVwBF7ARQGVW0z6K439/v4233RJ+p5P+SEPnXCbRo/jWyc7Lxx8Z1SEq+h+5de5R3aEREnx0mRFQpzZkzR1zD6H0tW7bEkSNHyjgi+VNRUcG2naGYPmcqBEFA/br22LF5H+rWqYf/7iej5TeFb14MAH+Fn0fNGh8/cZ6IqLJhQkSV0tChQ+HtXfgk4dIsC1CR1bCoiYO7jhVaZmZqjhOH/yrytWam5ooK64OKG/aTFw79lUxFG94gKk9MiKhSMjIy+iw3kZWXKlWqwMa69ocrEhERAN5lRkRERMSESF54sx5R5cWfb6LKjwnRJ8rfLuDVq1flHAkRKUr+z/f724MQUeXBOUSfSFVVFYaGhuImktra2qVaKZioNHLfKGYfsHxZWVkKbR9QfB8A+fVDEAS8evUKaWlpMDQ0LHbhTiL6vDEhkoP8Pbc+Zfd1opJ4/kSxCUtmjuIXHVR0HwD598PQ0LBUe+sR0eeHCZEcSCQSmJubw8TERGYvpi0B5xR+bp/pXyn8HFRxbAlW7HfKZ/rH761WUoruAyDffqipqfHKEJESYEIkR6qqqjL/48zKkCr8nNxGQLko+jtVFt8n/lwQUUXESdVERESk9EqdEF26dAlxcXHi83379sHT0xO//PILcnIUP1mSiIiISN5KnRD9+OOPuHHjBgDgzp076NWrF7S1tbFjxw5MnDhR7gESERERKVqpE6IbN27AyckJALBjxw60atUKoaGhCA4Oxq5du+QdHxEREZHClTohEgQBUunbSZHHjx9Hx44dAQCWlpZ4/PixfKMjIiIiKgOlToicnZ0xa9YsbNq0CadOnUKnTp0AAImJiTA1NZV7gERERESKVuqEaMmSJbh06RL8/f3x66+/ok6dOgCAnTt3okWLFnIPkIiIiEjRSr0OUcOGDWXuMsv3v//9j4uXERER0WfpoxdmzMnJQVpamjifKJ+VldUnB0VERERUlkqdEN24cQMDBw7E2bNnZY4LggCJRIK8vDy5BUdERERUFkqdEPXv3x9VqlTBwYMHYW5uzp3diYiI6LNX6oQoNjYWMTExqF+/viLiISIiIipzpb7LzMHBgesNERERUaVS6oRo/vz5mDhxIiIjI/HkyRNkZmbKPIiIiIg+N6UeMnN3dwcAtGvXTuY4J1UTERHR56rUCdHJkycVEQcRERFRuSl1QuTm5qaIOIiIiIjKzUctzJieno4NGzYgISEBANCgQQMMGDAABgYGcg2Oyt/KoScUfo7ha9oq/BxERETFKfWk6osXL6J27dr47bff8PTpUzx9+hSLFy9G7dq1cenSJUXESERERKRQpb5CNGbMGHTp0gXr169HlSpvX/7mzRsMGjQIo0ePxunTp+UeJBEREZEilTohunjxokwyBABVqlTBxIkT4ezsLNfgiIiIiMpCqYfM9PX1kZSUVOB4cnIy9PT05BIUERERUVkqdULUs2dPDBw4ENu2bUNycjKSk5OxdetWDBo0CL1791ZEjEREREQKVeohs4ULF0IikaBfv3548+YNAEBNTQ0//fQT5s2bJ/cAiYiIiBSt1AmRuro6li5dirlz5+L27dsAgNq1a0NbW1vuwRERERGVhY9ahwgAtLW14ejoKM9YiIiIiMpFiRKi7t27Izg4GPr6+ujevXuxdXfv3i2XwIiIiIjKSokSIgMDA0gkEgBv7zLL/zcRERFRZVCihCgoKEj8d3BwsKJiISIiIioXpb7tvm3btkhPTy9wPDMzE23bck8qIiIi+vyUOiGKjIxETk5OgeNZWVn466+/5BIUERERUVkq8V1mV65cEf8dHx+P1NRU8XleXh7CwsJQo0YN+UZHREREVAZKnBA5OTlBIpFAIpEUOjSmpaWF5cuXyzU4IiIiorJQ4oQoMTERgiDA1tYWFy5cgLGxsVimrq4OExMTqKqqKiRIIiIiIkUqcUJUq1YtAIBUKlVYMERERETl4aNXqo6Pj0dSUlKBCdZdunT55KCIiIiUzcqhJxTa/vA1vBO8OKVOiO7cuYNu3bohLi4OEokEgiAAgLhYY15ennwjJCIiIlKwUt92P2rUKNjY2CAtLQ3a2tq4evUqTp8+DWdnZ0RGRiogRCIiIiLFKvUVoqioKJw4cQLVq1eHiooKVFRU8PXXX2Pu3LkYOXIk/vnnH0XESURERKQwpb5ClJeXBz09PQBA9erV8eDBAwBvJ11fv35dvtEBuH//Pn744QdUq1YNWlpacHR0xMWLF8VyQRAwbdo0mJubQ0tLC+7u7rh586ZMG0+fPoWPjw/09fVhaGiIgQMH4sWLF3KPlYiIiD5PpU6IvvjiC1y+fBkA0Lx5cyxYsABnzpzBjBkzYGtrK9fgnj17BldXV6ipqeHIkSOIj4/HokWLULVqVbHOggULsGzZMqxZswbnz5+Hjo4OPDw8kJWVJdbx8fHB1atXER4ejoMHD+L06dMYMmSIXGMlIiKiz1eph8ymTJmCly9fAgBmzJiB7777Di1btkS1atWwbds2uQY3f/58WFpaymwua2NjI/5bEAQsWbIEU6ZMQdeuXQEAISEhMDU1xd69e9GrVy8kJCQgLCwM0dHRcHZ2BgAsX74cHTt2xMKFC2FhYSHXmImIiOjzU+orRB4eHujevTsAoE6dOrh27RoeP36MtLQ0uW/uun//fjg7O6NHjx4wMTFB48aNsX79erE8MTERqampcHd3F48ZGBigefPmiIqKAvB2zpOhoaGYDAGAu7s7VFRUcP78ebnGS0RERJ+nUidEhTEyMhJvu5enO3fuYPXq1bCzs8PRo0fx008/YeTIkdi4cSMAiPupmZqayrzO1NRULEtNTYWJiYlMeZUqVWBkZCSzH9u7srOzkZmZKfMgIiKiyqtEQ2b5V4RKYvfu3R8dzPukUimcnZ0xZ84cAEDjxo3x77//Ys2aNfD19ZXbed43d+5cTJ8+XWHtExERUcVSoitEBgYGJX7Ik7m5ORwcHGSO2dvbIykpCQBgZmYGAHj48KFMnYcPH4plZmZmSEtLkyl/8+YNnj59KtZ53+TJk5GRkSE+kpOT5dIfIiIiqphKdIXo3UnNZcnV1bXArfw3btwQ91WzsbGBmZkZIiIi4OTkBADIzMzE+fPn8dNPPwEAXFxckJ6ejpiYGDRp0gQAcOLECUilUjRv3rzQ82poaEBDQ0NBvSIiIqKK5qP3Mnv06JGYrNSrVw/GxsZyCyrfmDFj0KJFC8yZMwfe3t64cOEC1q1bh3Xr1gF4u13I6NGjMWvWLNjZ2cHGxgZTp06FhYUFPD09Aby9otS+fXsMHjwYa9asQW5uLvz9/dGrVy/eYUZEREQAPiIhevnyJUaMGIGQkBBIpVIAgKqqKvr164fly5dDW1tbbsE1bdoUe/bsweTJkzFjxgzY2NhgyZIl8PHxEetMnDgRL1++xJAhQ5Ceno6vv/4aYWFh0NTUFOts2bIF/v7+aNeuHVRUVODl5YVly5bJLU6q+BS9aSLAjROJiD5npU6Ixo4di1OnTuHAgQNwdXUFAPz9998YOXIkxo0bh9WrV8s1wO+++w7fffddkeUSiQQzZszAjBkziqxjZGSE0NBQucZFRERElUepE6Jdu3Zh586daN26tXisY8eO0NLSgre3t9wTIiIiIiJFK/U6RK9evSqw7g8AmJiY4NWrV3IJioiIiKgslTohcnFxQUBAgMxeYa9fv8b06dPh4uIi1+CIiIiIykKph8yWLl0KDw8P1KxZE40aNQIAXL58GZqamjh69KjcAyQiIiJStFInRF988QVu3ryJLVu24Nq1awCA3r17w8fHB1paWnIPkIiIiEjRPmodIm1tbQwePFjesRARERGVixIlRPv370eHDh2gpqaG/fv3F1u3S5cucgmMiIiIqKyUKCHy9PQUd43PXwG6MBKJBHl5efKKjYiIiKhMlCghyl+R+v1/ExEREVUGpb7tnju/ExERUWVT6oTI2toabm5uWL9+PZ49e6aImIiIiIjKVKkToosXL6JZs2aYMWMGzM3N4enpiZ07dyI7O1sR8REREREpXKkTosaNG+N///sfkpKScOTIERgbG2PIkCEwNTXFgAEDFBEjERERkUKVOiHKJ5FI0KZNG6xfvx7Hjx+HjY0NNm7cKM/YiIiIiMrERydE//33HxYsWAAnJyc0a9YMurq6WLlypTxjIyIiIioTpV6peu3atQgNDcWZM2dQv359+Pj4YN++fahVq5Yi4iMiIiJSuFInRLNmzULv3r2xbNkycXNXIiIios9ZqROipKQkSCQSRcRCREREVC5KPYeIyRARERFVNh89qZqIiIiosmBCREREREqvRAnR/v37kZubq+hYiIiIiMpFiRKibt26IT09HQCgqqqKtLQ0RcZEREREVKZKlBAZGxvj3LlzAABBEDixmoiIiCqVEt12P3ToUHTt2hUSiQQSiQRmZmZF1s3Ly5NbcERERERloUQJUWBgIHr16oVbt26hS5cuCAoKgqGhoYJDIyIiIiobJV6YsX79+qhfvz4CAgLQo0cPaGtrKzIuIqIKbeXQEwo/x/A1bRV+DiJ6q9QrVQcEBAAAHj16hOvXrwMA6tWrB2NjY/lGRkRERFRGSr0O0atXrzBgwABYWFigVatWaNWqFSwsLDBw4EC8evVKETESERERKVSpE6IxY8bg1KlT2L9/P9LT05Geno59+/bh1KlTGDdunCJiJCIiIlKoUg+Z7dq1Czt37kTr1q3FYx07doSWlha8vb2xevVqecZHREREpHAfNWRmampa4LiJiQmHzIiIiOizVOqEyMXFBQEBAcjKyhKPvX79GtOnT4eLi4tcgyMiIiIqC6UeMlu6dCk8PDxQs2ZNNGrUCABw+fJlaGpq4ujRo3IPkIiIiEjRSp0QffHFF7h58ya2bNmCa9euAQB69+4NHx8faGlpyT1AIiIiIkUrdUIEANra2hg8eLC8YyEiIiIqF6WeQ0RERERU2TAhIiIiIqXHhIiIiIiUHhMiIiIiUnqlTohsbW3x5MmTAsfT09Nha2srl6CIiIiIylKpE6K7d+8iLy+vwPHs7Gzcv39fLkERERERlaUS33a/f/9+8d9Hjx6FgYGB+DwvLw8RERGwtraWa3BEREREZaHECZGnpycAQCKRwNfXV6ZMTU0N1tbWWLRokVyDIyIiIioLJU6IpFIpAMDGxgbR0dGoXr26woIiIiIiKkulXqk6MTFREXEQERERlZuP2rojIiICERERSEtLE68c5fvjjz/kEhgRERFRWSl1QjR9+nTMmDEDzs7OMDc3h0QiUURcRERERGWm1AnRmjVrEBwcjL59+yoiHiIiIqIyV+p1iHJyctCiRQtFxEJERERULkqdEA0aNAihoaGKiIWIiIioXJR6yCwrKwvr1q3D8ePH0bBhQ6ipqcmUL168WG7BEREREZWFUidEV65cgZOTEwDg33//lSnjBGsiIiL6HJU6ITp58qQi4iAiIiIqN6WeQ0RERERU2ZT6ClGbNm2KHRo7ceLEJwVEREREVNZKnRDlzx/Kl5ubi9jYWPz7778FNn0lIiIi+hyUOiH67bffCj0eGBiIFy9efHJARERERGVNbnOIfvjhB4XvYzZv3jxIJBKMHj1aPJaVlYXhw4ejWrVq0NXVhZeXFx4+fCjzuqSkJHTq1Ana2towMTHBhAkT8ObNG4XGSkRERJ8PuSVEUVFR0NTUlFdzBURHR2Pt2rVo2LChzPExY8bgwIED2LFjB06dOoUHDx6ge/fuYnleXh46deqEnJwcnD17Fhs3bkRwcDCmTZumsFiJiIjo81LqIbN3kw0AEAQBKSkpuHjxIqZOnSq3wN714sUL+Pj4YP369Zg1a5Z4PCMjAxs2bEBoaCjatm0LAAgKCoK9vT3OnTuHr776CseOHUN8fDyOHz8OU1NTODk5YebMmfj5558RGBgIdXV1hcRMREREn49SXyEyMDCQeRgZGaF169Y4fPgwAgICFBEjhg8fjk6dOsHd3V3meExMDHJzc2WO169fH1ZWVoiKigLw9sqVo6MjTE1NxToeHh7IzMzE1atXCz1fdnY2MjMzZR5ERERUeZX6ClFQUJAi4ijS1q1bcenSJURHRxcoS01Nhbq6OgwNDWWOm5qaIjU1VazzbjKUX55fVpi5c+di+vTpcoieiIiIPgelTojyxcTEICEhAQDQoEEDNG7cWG5B5UtOTsaoUaMQHh6u0PlJ75s8eTLGjh0rPs/MzISlpWWZnZ+IiIjKVqkTorS0NPTq1QuRkZHilZn09HS0adMGW7duhbGxsdyCi4mJQVpaGr788kvxWF5eHk6fPo0VK1bg6NGjyMnJQXp6usxVoocPH8LMzAwAYGZmhgsXLsi0m38XWn6d92loaEBDQ0Nu/SAiIqKKrdRziEaMGIHnz5/j6tWrePr0KZ4+fYp///0XmZmZGDlypFyDa9euHeLi4hAbGys+nJ2d4ePjI/5bTU0NERER4muuX7+OpKQkuLi4AABcXFwQFxeHtLQ0sU54eDj09fXh4OAg13iJiIjo81TqK0RhYWE4fvw47O3txWMODg5YuXIlvv32W7kGp6enhy+++ELmmI6ODqpVqyYeHzhwIMaOHQsjIyPo6+tjxIgRcHFxwVdffQUA+Pbbb+Hg4IC+fftiwYIFSE1NxZQpUzB8+HBeBSIiIiIAH5EQSaVSqKmpFTiupqYGqVQql6BK47fffoOKigq8vLyQnZ0NDw8PrFq1SixXVVXFwYMH8dNPP8HFxQU6Ojrw9fXFjBkzyjxWIiIiqphKnRC1bdsWo0aNwp9//gkLCwsAwP379zFmzBi0a9dO7gG+LzIyUua5pqYmVq5ciZUrVxb5mlq1auHw4cMKjoyIiIg+V6WeQ7RixQpkZmbC2toatWvXRu3atWFjY4PMzEwsX75cETESERERKVSprxBZWlri0qVLOH78OK5duwYAsLe3L7BoIhEREdHn4qPWIZJIJPjmm2/wzTffyDseIiIiojJX4iGzEydOwMHBodBtLDIyMtCgQQP89ddfcg2OiIiIqCyUOCFasmQJBg8eDH19/QJlBgYG+PHHH7F48WK5BkdERERUFkqcEF2+fBnt27cvsvzbb79FTEyMXIIiIiIiKkslTogePnxY6PpD+apUqYJHjx7JJSgiIiKislTihKhGjRr4999/iyy/cuUKzM3N5RIUERERUVkqcULUsWNHTJ06FVlZWQXKXr9+jYCAAHz33XdyDY6IiIioLJT4tvspU6Zg9+7dqFu3Lvz9/VGvXj0AwLVr17By5Urk5eXh119/VVigRERERIpS4oTI1NQUZ8+exU8//YTJkydDEAQAb9ck8vDwwMqVK2FqaqqwQImIiIgUpVQLM+bvCfbs2TPcunULgiDAzs4OVatWVVR8RERERAr3UStVV61aFU2bNpV3LERERETlotSbuxIRERFVNkyIiIiISOkxISIiIiKlx4SIiIiIlB4TIiIiIlJ6TIiIiIhI6TEhIiIiIqXHhIiIiIiUHhMiIiIiUnpMiIiIiEjpMSEiIiIipceEiIiIiJQeEyIiIiJSekyIiIiISOkxISIiIiKlx4SIiIiIlB4TIiIiIlJ6TIiIiIhI6TEhIiIiIqXHhIiIiIiUHhMiIiIiUnpMiIiIiEjpMSEiIiIipceEiIiIiJQeEyIiIiJSekyIiIiISOkxISIiIiKlx4SIiIiIlB4TIiIiIlJ6TIiIiIhI6TEhIiIiIqXHhIiIiIiUHhMiIiIiUnpMiIiIiEjpMSEiIiIipceEiIiIiJQeEyIiIiJSekyIiIiISOkxISIiIiKlx4SIiIiIlB4TIiIiIlJ6TIiIiIhI6TEhIiIiIqXHhIiIiIiUHhMiIiIiUnoVOiGaO3cumjZtCj09PZiYmMDT0xPXr1+XqZOVlYXhw4ejWrVq0NXVhZeXFx4+fChTJykpCZ06dYK2tjZMTEwwYcIEvHnzpiy7QkRERBVYhU6ITp06heHDh+PcuXMIDw9Hbm4uvv32W7x8+VKsM2bMGBw4cAA7duzAqVOn8ODBA3Tv3l0sz8vLQ6dOnZCTk4OzZ89i48aNCA4OxrRp08qjS0RERFQBVSnvAIoTFhYm8zw4OBgmJiaIiYlBq1atkJGRgQ0bNiA0NBRt27YFAAQFBcHe3h7nzp3DV199hWPHjiE+Ph7Hjx+HqakpnJycMHPmTPz8888IDAyEurp6eXSNiIiIKpAKfYXofRkZGQAAIyMjAEBMTAxyc3Ph7u4u1qlfvz6srKwQFRUFAIiKioKjoyNMTU3FOh4eHsjMzMTVq1cLPU92djYyMzNlHkRERFR5fTYJkVQqxejRo+Hq6oovvvgCAJCamgp1dXUYGhrK1DU1NUVqaqpY591kKL88v6wwc+fOhYGBgfiwtLSUc2+IiIioIvlsEqLhw4fj33//xdatWxV+rsmTJyMjI0N8JCcnK/ycREREVH4q9ByifP7+/jh48CBOnz6NmjVrisfNzMyQk5OD9PR0matEDx8+hJmZmVjnwoULMu3l34WWX+d9Ghoa0NDQkHMviIiIqKKq0FeIBEGAv78/9uzZgxMnTsDGxkamvEmTJlBTU0NERIR47Pr160hKSoKLiwsAwMXFBXFxcUhLSxPrhIeHQ19fHw4ODmXTESIiIqrQKvQVouHDhyM0NBT79u2Dnp6eOOfHwMAAWlpaMDAwwMCBAzF27FgYGRlBX18fI0aMgIuLC7766isAwLfffgsHBwf07dsXCxYsQGpqKqZMmYLhw4fzKhAREREBqOAJ0erVqwEArVu3ljkeFBQEPz8/AMBvv/0GFRUVeHl5ITs7Gx4eHli1apVYV1VVFQcPHsRPP/0EFxcX6OjowNfXFzNmzCirbhAREVEFV6ETIkEQPlhHU1MTK1euxMqVK4usU6tWLRw+fFieoREREVElUqHnEBERERGVBSZEREREpPSYEBEREZHSY0JERERESo8JERERESk9JkRERESk9JgQERERkdJjQkRERERKjwkRERERKT0mRERERKT0mBARERGR0mNCREREREqPCREREREpPSZEREREpPSYEBEREZHSY0JERERESo8JERERESk9JkRERESk9JgQERERkdJjQkRERERKjwkRERERKT0mRERERKT0mBARERGR0mNCREREREqPCREREREpPSZEREREpPSYEBEREZHSY0JERERESo8JERERESk9JkRERESk9JgQERERkdJjQkRERERKjwkRERERKT0mRERERKT0mBARERGR0mNCREREREqPCREREREpPSZEREREpPSYEBEREZHSY0JERERESo8JERERESk9JkRERESk9JgQERERkdJjQkRERERKjwkRERERKT0mRERERKT0mBARERGR0mNCREREREqPCREREREpPSZEREREpPSYEBEREZHSY0JERERESq9KeQdARCW3xmWUQtsfjjiFtk+kCIr+uQD4s6EMmBBRsYabdSuDs2Qo/AyV5X+YcYlJCj8HKY/K8vPNn4uK43P+fy0TIgX6nL8YlQ3/h1lxVJZfwkRU0Of8/1omRAr0OX8xiKh4TOxI3hT/neL3qTicVE1ERERKT6kSopUrV8La2hqamppo3rw5Lly4UN4hERERUQWgNAnRtm3bMHbsWAQEBODSpUto1KgRPDw8kJaWVt6hERERUTlTmoRo8eLFGDx4MPr37w8HBwesWbMG2tra+OOPP8o7NCIiIipnSjGpOicnBzExMZg8ebJ4TEVFBe7u7oiKilLYea2zQhXWdr67Cm6/MvQBYD9K6q5CW69cKsN3qjL0AWA/SuquQlt/63P+LJQiIXr8+DHy8vJgamoqc9zU1BTXrl0rUD87OxvZ2dni84yMtzPzMzMzS3Veafarj4i2dEobU2lVhj4A7EdJlUUfvsj4XeHn+LcSfBYAf75Liv0omcrQB6B0/civKwjChysLSuD+/fsCAOHs2bMyxydMmCA0a9asQP2AgAABAB988MEHH3zwUQkeycnJH8wVlOIKUfXq1aGqqoqHDx/KHH/48CHMzMwK1J88eTLGjh0rPpdKpXj69CmqVasGiUSikBgzMzNhaWmJ5ORk6OvrK+QcZaEy9KMy9AFgPyqSytAHoHL0ozL0AWA/SkoQBDx//hwWFhYfrKsUCZG6ujqaNGmCiIgIeHp6Anib5ERERMDf379AfQ0NDWhoaMgcMzQ0LINIAX19/c/6y52vMvSjMvQBYD8qksrQB6By9KMy9AFgP0rCwMCgRPWUIiECgLFjx8LX1xfOzs5o1qwZlixZgpcvX6J///7lHRoRERGVM6VJiHr27IlHjx5h2rRpSE1NhZOTE8LCwgpMtCYiIiLlozQJEQD4+/sXOkRWEWhoaCAgIKDAUN3npjL0ozL0AWA/KpLK0AegcvSjMvQBYD8UQSIIJbkXjYiIiKjyUpqVqomIiIiKwoSIiIiIlB4TIiIiIlJ6TIiIqELjNEciKgtKdZdZRfL48WP88ccfiIqKQmpqKgDAzMwMLVq0gJ+fH4yNjcs5QqKKQUNDA5cvX4a9vX15h0JElRjvMisH0dHR8PDwgLa2Ntzd3cW1kB4+fIiIiAi8evUKR48ehbOzczlHqhxev36NmJgYGBkZwcHBQaYsKysL27dvR79+/copupJLSEjAuXPn4OLigvr16+PatWtYunQpsrOz8cMPP6Bt27blHWKx3t0u511Lly7FDz/8gGrVqgEAFi9eXJZhfbKXL19i+/btuHXrFszNzdG7d2+xL6R4I0aMgLe3N1q2bFneoSi9lJQUrF69Gn///TdSUlKgoqICW1tbeHp6ws/PD6qqquUboBz2TqVSat68uTBkyBBBKpUWKJNKpcKQIUOEr776qhwik6+kpCShf//+5R1Gsa5fvy7UqlVLkEgkgoqKitCqVSvhwYMHYnlqaqqgoqJSjhGWzJEjRwR1dXXByMhI0NTUFI4cOSIYGxsL7u7uQtu2bQVVVVUhIiKivMMslkQiEZycnITWrVvLPCQSidC0aVOhdevWQps2bco7zA+yt7cXnjx5IgjC258Ba2trwcDAQGjatKlgZGQkmJiYCHfu3CnnKD8sJiZGJs6QkBChRYsWQs2aNQVXV1fhzz//LMfoSi7/Z9vOzk6YN2+ekJKSUt4hfZTly5cLffv2Fd/3kJAQwd7eXqhXr54wefJkITc3t5wjLF50dLRgYGAgNGnSRPj6668FVVVVoW/fvkLPnj0FQ0NDoUWLFkJmZma5xsiEqBxoamoKCQkJRZYnJCQImpqaZRiRYsTGxlb4ZMLT01Po1KmT8OjRI+HmzZtCp06dBBsbG+HevXuCIHw+CZGLi4vw66+/CoIgCH/++adQtWpV4ZdffhHLJ02aJHzzzTflFV6JzJ07V7CxsSmQuFWpUkW4evVqOUVVehKJRHj48KEgCILg4+MjtGjRQkhPTxcEQRCeP38uuLu7C7179y7PEEukYcOGQnh4uCAIgrB+/XpBS0tLGDlypLB69Wph9OjRgq6urrBhw4ZyjvLDJBKJcPz4cWHUqFFC9erVBTU1NaFLly7CgQMHhLy8vPIOr0Rmzpwp6OnpCV5eXoKZmZkwb948oVq1asKsWbOEOXPmCMbGxsK0adPKO8xiubq6CoGBgeLzTZs2Cc2bNxcEQRCePn0qODk5CSNHjiyv8ARBYEJULqytrYWNGzcWWb5x40ahVq1aZRfQR9q3b1+xj99++63CJxMmJibClStXxOdSqVQYOnSoYGVlJdy+ffuzSYj09fWFmzdvCoIgCHl5eUKVKlWES5cuieVxcXGCqalpeYVXYhcuXBDq1q0rjBs3TsjJyREE4fNOiGxtbYVjx47JlJ85c0awtLQsj9BKRUtLS7h7964gCILQuHFjYd26dTLlW7ZsERwcHMojtFJ59/PIyckRtm3bJnh4eAiqqqqChYWF8Msvv4g/OxVV7dq1hV27dgmC8PYPTVVVVWHz5s1i+e7du4U6deqUV3gloqWlJdy+fVt8npeXJ6ipqQmpqamCIAjCsWPHBAsLi/IKTxAEQeCk6nIwfvx4DBkyBDExMWjXrl2BOUTr16/HwoULyznKD/P09IREIin2LiCJRFKGEZXe69evUaXK//0YSCQSrF69Gv7+/nBzc0NoaGg5Rlc6+e+1iooKNDU1ZXZ41tPTQ0ZGRnmFVmJNmzZFTEwMhg8fDmdnZ2zZsqXCf4cKkx9zVlYWzM3NZcpq1KiBR48elUdYpaKtrY3Hjx+jVq1auH//Ppo1ayZT3rx5cyQmJpZTdB9HTU0N3t7e8Pb2RlJSEv744w8EBwdj3rx5yMvLK+/wivTgwQNxTmmjRo2goqICJycnsfzLL7/EgwcPyim6kjExMUFKSgpsbW0BvP199+bNG3GHezs7Ozx9+rQ8Q+Rt9+Vh+PDh2LhxI86fPw8vLy+4uLjAxcUFXl5eOH/+PIKDgzFs2LDyDvODzM3NsXv3bkil0kIfly5dKu8QP6h+/fq4ePFigeMrVqxA165d0aVLl3KIqvSsra1x8+ZN8XlUVBSsrKzE50lJSQV+MVdUurq62LhxIyZPngx3d/cK/YuqKO3atcOXX36JzMxMXL9+Xabs3r17n8Wk6g4dOmD16tUAADc3N+zcuVOmfPv27ahTp055hCYXVlZWCAwMRGJiIsLCwso7nGKZmZkhPj4eAHDz5k3k5eWJzwHg6tWrMDExKa/wSsTT0xNDhw5FWFgYTp48CR8fH7i5uUFLSwsAcP36ddSoUaNcY+QVonLSs2dP9OzZE7m5uXj8+DEAoHr16lBTUyvnyEquSZMmiImJQdeuXQst/9DVo4qgW7du+PPPP9G3b98CZStWrIBUKsWaNWvKIbLS+emnn2QShy+++EKm/MiRIxX+LrP39erVC19//TViYmJQq1at8g6nxAICAmSe6+rqyjw/cODAZ3HH0/z58+Hq6go3Nzc4Oztj0aJFiIyMhL29Pa5fv45z585hz5495R3mB9WqVavYu5ckEgm++eabMoyo9Hx8fNCvXz907doVERERmDhxIsaPH48nT55AIpFg9uzZ+P7778s7zGLNmjULKSkp6Ny5M/Ly8uDi4oLNmzeL5RKJBHPnzi3HCHnbPX2Cv/76Cy9fvkT79u0LLX/58iUuXrwINze3Mo6MiOQhPT0d8+bNw4EDB3Dnzh1IpVKYm5vD1dUVY8aM4dIgZUQqlWLevHmIiopCixYtMGnSJGzbtg0TJ07Eq1ev0LlzZ6xYsQI6OjrlHeoHZWVl4c2bNwX+UKgImBARERGR0uMcIiIiIlJ6TIiIiIhI6TEhIiIiIqXHhIiIKpW7d+9CIpEgNjZWYefw8/ODp6enwtonorLHhIiIKhQ/Pz9IJJICj6LuZnyfpaUlUlJSCiw9QERUHK5DREQVTvv27REUFCRzTENDo0SvVVVVhZmZmSLCIqJKjFeIiKjC0dDQgJmZmcyjatWqAP5ve5UOHTpAS0sLtra2Mqsovz9k9uzZM/j4+MDY2BhaWlqws7OTSbbi4uLQtm1baGlpoVq1ahgyZAhevHghlufl5WHs2LEwNDREtWrVMHHixAILjkqlUsydOxc2NjbQ0tJCo0aNZGL6UAxEVP6YEBHRZ2fq1Knw8vLC5cuX4ePjg169eiEhIaHIuvHx8Thy5AgSEhKwevVqVK9eHcDbxUM9PDxQtWpVREdHY8eOHTh+/Dj8/f3F1y9atAjBwcH4448/8Pfff+Pp06cFVmieO3cuQkJCsGbNGly9ehVjxozBDz/8gFOnTn0wBiKqIMpxY1kiogJ8fX0FVVVVQUdHR+Yxe/ZsQRAEAYAwdOhQmdc0b95c+OmnnwRBEITExEQBgPDPP/8IgiAInTt3Fvr371/oudatWydUrVpVePHihXjs0KFDgoqKirgLt7m5ubBgwQKxPDc3V6hZs6bQtWtXQRAEISsrS9DW1hbOnj0r0/bAgQOF3r17fzAGIqoYOIeIiCqcNm3aiBuL5jMyMhL/7eLiIlPm4uJS5F1lP/30E7y8vHDp0iV8++238PT0RIsWLQAACQkJaNSokcyWB66urpBKpbh+/To0NTWRkpKC5s2bi+VVqlSBs7OzOGx269YtvHr1qsB+WDk5OWjcuPEHYyCiioEJERFVODo6OnLbSb1Dhw64d+8eDh8+jPDwcLRr1w7Dhw/HwoUL5dJ+/nyjQ4cOFditO38iuKJjIKJPxzlERPTZOXfuXIHn9vb2RdY3NjaGr68vNm/ejCVLlmDdunUAAHt7e1y+fBkvX74U6545cwYqKiqoV68eDAwMYG5ujvPnz4vlb968QUxMjPjcwcEBGhoaSEpKQp06dWQelpaWH4yBiCoGXiEiogonOzsbqampMseqVKkiTkTesWMHnJ2d8fXXX2PLli24cOECNmzYUGhb06ZNQ5MmTdCgQQNkZ2fj4MGDYvLk4+ODgIAA+Pr6IjAwEI8ePcKIESPQt29fmJqaAgBGjRqFefPmwc7ODvXr18fixYuRnp4utq+np4fx48djzJgxkEql+Prrr5GRkYEzZ85AX18fvr6+xcZARBUDEyIiqnDCwsJgbm4uc6xevXq4du0aAGD69OnYunUrhg0bBnNzc/z5559wcHAotC11dXVMnjwZd+/ehZaWFlq2bImtW7cCALS1tXH06FGMGjUKTZs2hba2Nry8vLB48WLx9ePGjUNKSgp8fX2hoqKCAQMGoFu3bsjIyBDrzJw5E8bGxpg7dy7u3LkDQ0NDfPnll/jll18+GAMRVQwSQXhvQQ0iogpMIpFgz5493DqDiOSKc4iIiIhI6TEhIiIiIqXHOURE9FnhKD8RKQKvEBEREZHSY0JERERESo8JERERESk9JkRERESk9JgQERERkdJjQkRERERKjwkRERERKT0mRERERKT0mBARERGR0vt/ktQcz5MYq7IAAAAASUVORK5CYII=", + "image/png": "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", "text/plain": [ "
" ] @@ -126766,7 +65861,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -126776,7 +65871,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -126786,7 +65881,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -126796,7 +65891,7 @@ }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAkQAAAHCCAYAAAAO4dYCAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAAB1LElEQVR4nO3deVyN6f8/8Ncp7atoJS3WaiKET0KWyC6MLA0xhrHE2JcZlH0ZjD3GmGJGY9+XLI0YhGRKQxpLZCghla313L8//Lq/jhYnzqk4r+fjcR4P931d932/r3NOendf131dEkEQBBARERGpMLXyDoCIiIiovDEhIiIiIpXHhIiIiIhUHhMiIiIiUnlMiIiIiEjlMSEiIiIilceEiIiIiFQeEyIiIiJSeUyIiIiISOUxIaIKQSKRIDAw8L31AgMDIZFIPuicISEhkEgkuHv37ocFSYVERERAIpEgIiKivENRqI9p1927dyGRSBASEqLwuOj/DB48GLa2th90rK2tLQYPHlxh4qGKgQkRKUxBwiGRSHD27NlC5YIgwNraGhKJBF27di2HCOlDrVu3rsL9gq+IMdHn7eHDhwgMDERMTEx5h0JKUKm8A6DPj7a2NkJDQ9GiRQuZ/adPn8Z///0HLS2tQse8fv0alSrx61hRrVu3DlWrVi30V3WrVq3w+vVraGpqVpiYFOFj2mVjY4PXr19DQ0ND4XHR/9m4cSOkUmmZXvPhw4eYPXs2bG1t4eLiUu7xkGLxDhEpXOfOnbFz507k5eXJ7A8NDUXjxo1hYWFR6BhtbW0mRGVEEAS8fv1aIedSU1ODtrY21NQq9n8lL1++LFX9j2mXRCKBtrY21NXVS33sp+LVq1flHQI0NDSK/OOqvFS0eKj0Kvb/YvRJ6t+/P54+fYoTJ06I+3JycrBr1y4MGDCgyGOKGkN09uxZNGnSBNra2qhZsyY2bNhQ5LHZ2dkYP348TE1NYWBggO7du+O///6TO96jR4+iZcuW0NPTg4GBAbp06YJr167JdezVq1fh4eEBHR0dVK9eHfPmzUNwcHCRY5Xkuc7gwYOhr6+PBw8ewNvbG/r6+jA1NcWkSZOQn58vU1cqlWLFihVwcnKCtrY2zM3N8e233+LZs2cy9WxtbdG1a1ccO3YMrq6u0NHREd/L4OBgtG3bFmZmZtDS0oKjoyOCgoIKHX/t2jWcPn1a7BJt3bo1gOLH2uzcuRONGzeGjo4Oqlatiq+++goPHjz44La+q6SYCrpuT58+jVGjRsHMzAzVq1cHANy7dw+jRo1C3bp1oaOjgypVqqBPnz6FPqui2tW6dWt88cUXuH79Otq0aQNdXV1Uq1YNS5YskTm2qDFEpWnr06dPMXDgQBgaGsLY2Bh+fn6IjY2Va1xSQdvPnTuHCRMmwNTUFHp6eujZsyceP35cqP66devg5OQELS0tWFlZYfTo0UhPT5epU9Du6OhotGrVCrq6uvj+++/Fdi5duhRr166Fvb09dHV10aFDB9y/fx+CIGDu3LmoXr06dHR00KNHD6SlpZUY/9KlSyGRSHDv3r1CZdOnT4empqb4/S5qzM7Lly8xceJEWFtbQ0tLC3Xr1sXSpUshCEKJ101LS8OkSZPg7OwMfX19GBoaolOnToiNjRXrREREoEmTJgCAIUOGiN+7gs/kY+KRSCTw9/fHvn378MUXX0BLSwtOTk4ICwuTqff8+XOMGzcOtra20NLSgpmZGdq3b48rV66U2D6SD/8kJ4WztbWFm5sb/vjjD3Tq1AnAm2QgIyMD/fr1w6pVq957jri4OHTo0AGmpqYIDAxEXl4eAgICYG5uXqjuN998g99//x0DBgxA8+bN8eeff6JLly5yxfrbb7/Bz88PXl5eWLx4MV69eoWgoCC0aNECf//9d4mDJB88eIA2bdpAIpFg+vTp0NPTwy+//FLkX4mluU5+fj68vLzQrFkzLF26FCdPnsSyZctQs2ZNjBw5Uqz37bffIiQkBEOGDMHYsWORmJiINWvW4O+//8a5c+dkumwSEhLQv39/fPvttxg2bBjq1q0LAAgKCoKTkxO6d++OSpUq4eDBgxg1ahSkUilGjx4NAFixYgXGjBkDfX19/PDDDwBQ5OdQoCCmJk2aYOHChXj06BFWrlyJc+fO4e+//4axsXGp2/oueWIaNWoUTE1NMWvWLPEOUVRUFM6fP49+/fqhevXquHv3LoKCgtC6dWtcv34durq6xV4TAJ49e4aOHTuiV69e8PHxwa5duzB16lQ4OzuL3/XiyNNWqVSKbt264dKlSxg5ciTq1auH/fv3w8/Pr8Rzv2vMmDGoXLkyAgICcPfuXaxYsQL+/v7Yvn27WCcwMBCzZ8+Gp6cnRo4ciYSEBAQFBSEqKqrQ9+fp06fo1KkT+vXrh6+++krmvd66dStycnIwZswYpKWlYcmSJfDx8UHbtm0RERGBqVOn4tatW1i9ejUmTZqEX3/9tdi4fXx8MGXKFOzYsQOTJ0+WKduxYwc6dOiAypUrF3msIAjo3r07Tp06haFDh8LFxQXHjh3D5MmT8eDBA/z000/FXvfOnTvYt28f+vTpAzs7Ozx69AgbNmyAh4cHrl+/DisrKzg4OGDOnDmYNWsWhg8fjpYtWwIAmjdvrpB4zp49iz179mDUqFEwMDDAqlWr0Lt3byQlJaFKlSoAgBEjRmDXrl3w9/eHo6Mjnj59irNnzyI+Ph6NGjUqtn0kJ4FIQYKDgwUAQlRUlLBmzRrBwMBAePXqlSAIgtCnTx+hTZs2giAIgo2NjdClSxeZYwEIAQEB4ra3t7egra0t3Lt3T9x3/fp1QV1dXXj7axsTEyMAEEaNGiVzvgEDBhQ6Z0F8iYmJgiAIwvPnzwVjY2Nh2LBhMsempKQIRkZGhfa/a8yYMYJEIhH+/vtvcd/Tp08FExOTD76On5+fAECYM2eOTN2GDRsKjRs3Frf/+usvAYCwdetWmXphYWGF9tvY2AgAhLCwsEJtKPh83ubl5SXY29vL7HNychI8PDwK1T116pQAQDh16pQgCIKQk5MjmJmZCV988YXw+vVrsd6hQ4cEAMKsWbNK3dbiFBdTwefcokULIS8vT6asqPZGRkYKAIQtW7YU2y5BEAQPD49C9bKzswULCwuhd+/e4r7ExEQBgBAcHFzqtu7evVsAIKxYsULcl5+fL7Rt27bQOYtS0HZPT09BKpWK+8ePHy+oq6sL6enpgiAIQmpqqqCpqSl06NBByM/PF+utWbNGACD8+uuvhdq9fv16mWsVtNPU1FQ8ryAIwvTp0wUAQoMGDYTc3Fxxf//+/QVNTU0hKyurxDa4ubkV+vwvXbpU6L338/MTbGxsxO19+/YJAIR58+bJHPvll18KEolEuHXrlrjPxsZG8PPzE7ezsrJk3oeC9mlpacl8ZlFRUcV+Dh8TDwBBU1NTZl9sbKwAQFi9erW4z8jISBg9enSha5NisMuMlMLHxwevX7/GoUOH8Pz5cxw6dKjY7rJ35efn49ixY/D29kaNGjXE/Q4ODvDy8pKpe+TIEQDA2LFjZfaPGzfuvdc5ceIE0tPT0b9/fzx58kR8qauro1mzZjh16lSJx4eFhcHNzU1mcKWJiQl8fX0/+jojRoyQ2W7ZsiXu3Lkjbu/cuRNGRkZo3769zDkbN24MfX39Que0s7Mr9N4BgI6OjvjvjIwMPHnyBB4eHrhz5w4yMjJKbH9RLl++jNTUVIwaNQra2tri/i5duqBevXo4fPhwqdv6oYYNG1ZoHM/b7c3NzcXTp09Rq1YtGBsby9XtoK+vj6+++krc1tTURNOmTeWO931tDQsLg4aGBoYNGybuU1NTE+/WyWv48OEy01O0bNkS+fn5YlfUyZMnkZOTg3HjxsmMkxo2bBgMDQ0LfU5aWloYMmRIkdfq06cPjIyMxO1mzZoBAL766iuZcYHNmjVDTk5Ooa7Td/Xt2xfR0dG4ffu2uG/79u3Q0tJCjx49ij3uyJEjUFdXL/R/wcSJEyEIAo4ePVrssVpaWuL7kJ+fj6dPn0JfXx9169b94O6o0sbj6emJmjVritv169eHoaGhzPfD2NgYFy9exMOHDz8oJioZu8xIKUxNTeHp6YnQ0FC8evUK+fn5+PLLL+U69vHjx3j9+jVq165dqKxu3bpiEgS8GROipqYm8x9JQb33uXnzJgCgbdu2RZYbGhqWePy9e/fg5uZWaH+tWrU+6jra2towNTWV2Ve5cmWZsUE3b95ERkYGzMzMijxnamqqzLadnV2R9c6dO4eAgABERkYWGiibkZEh84tOHgW/cIt6/+vVq1doOgZ52vqhimrz69evsXDhQgQHB+PBgwcyYznkSQCrV69eaB6sypUr4+rVq+89Vp623rt3D5aWloW67t79Tr3P239IFFwHgHit4j4nTU1N2NvbFxrDU61atWKfuHv3WgXfGWtr6yL3v++z7dOnDyZMmIDt27fj+++/hyAI2LlzJzp16lTiz+S9e/dgZWUFAwMDmf0ODg5ieXGkUilWrlyJdevWITExUWZcV0F3VWmVNp5330eg8PdjyZIl8PPzg7W1NRo3bozOnTtj0KBBsLe3/6AYSRYTIlKaAQMGYNiwYUhJSUGnTp1kxo5UBAWPyP72229FPvmmqKfeSnsdeZ5OkkqlMDMzw9atW4ssf/cX79t3Rgrcvn0b7dq1Q7169bB8+XJYW1tDU1MTR44cwU8//VQmjxAr80msoto8ZswYBAcHY9y4cXBzc4ORkREkEgn69esnV3uLi1d4z6Ddko5Vho+JsyhFvZfvu9aHxmBlZYWWLVtix44d+P7773HhwgUkJSVh8eLF8gdcSgsWLMDMmTPx9ddfY+7cuTAxMYGamhrGjRtXZo/Sy/N++fj4oGXLlti7dy+OHz+OH3/8EYsXL8aePXveO4aN3o8JESlNz5498e233+LChQsygznfx9TUFDo6OuKdlbclJCTIbNvY2EAqleL27dsyf+2+W68oBXeVzMzM4OnpKXd8b1/71q1bhfa/u+9jr1OUmjVr4uTJk3B3dy/xl1VJDh48iOzsbBw4cEDmr9OiuvDknR3cxsYGwJv3/907YgkJCWK5Isgb09t27doFPz8/LFu2TNyXlZVV6Mmq8mJjY4NTp07h1atXMneJivqefex1gDefydt3F3JycpCYmKiw7+mH6tu3L0aNGoWEhARs374durq66NatW4nH2NjY4OTJk3j+/LnMXZkbN26I5cXZtWsX2rRpg02bNsnsT09PR9WqVcXt0nznPiaeklhaWmLUqFEYNWoUUlNT0ahRI8yfP58JkQJwDBEpjb6+PoKCghAYGPje/8zepq6uDi8vL+zbtw9JSUni/vj4eBw7dkymbsF/Au8+ubZixYr3XsfLywuGhoZYsGABcnNzC5UX9Zjyu8dHRkbKzFqblpZW6K7Nx16nKD4+PsjPz8fcuXMLleXl5cn1C77gL9J3u42Cg4ML1dXT05PrnK6urjAzM8P69euRnZ0t7j969Cji4+PlfvpPHvLG9DZ1dfVCdyhWr1793sf8y4qXlxdyc3OxceNGcZ9UKsXatWsVeh1PT09oampi1apVMu/Hpk2bkJGRodDP6UP07t0b6urq+OOPP7Bz50507doVenp6JR7TuXNn5OfnY82aNTL7f/rpJ0gkkhIThqK+Fzt37iw03qkgBnm+dx8TT1Hy8/MLdeuamZnByspK5meNPhzvEJFSlfZx4QKzZ89GWFgYWrZsiVGjRiEvLw+rV6+Gk5OTzHgNFxcX9O/fH+vWrUNGRgaaN2+O8PBwuf6iNjQ0RFBQEAYOHIhGjRqhX79+MDU1RVJSEg4fPgx3d/dC/5m9bcqUKfj999/Rvn17jBkzRnzsvkaNGkhLSxP/mvzY6xTFw8MD3377LRYuXIiYmBh06NABGhoauHnzJnbu3ImVK1e+d8xWhw4doKmpiW7duuHbb7/FixcvsHHjRpiZmSE5OVmmbuPGjREUFIR58+ahVq1aMDMzK3JMlIaGBhYvXowhQ4bAw8MD/fv3Fx+7t7W1xfjx40vVzpLIG9Pbunbtit9++w1GRkZwdHREZGQkTp48+cHjRBTN29sbTZs2xcSJE3Hr1i3Uq1cPBw4cEOfv+ZC7YkUxNTXF9OnTMXv2bHTs2BHdu3dHQkIC1q1bhyZNmsgMHC8PZmZmaNOmDZYvX47nz5+jb9++7z2mW7duaNOmDX744QfcvXsXDRo0wPHjx7F//36MGzeu0DjDt3Xt2hVz5szBkCFD0Lx5c8TFxWHr1q2FxubUrFkTxsbGWL9+PQwMDKCnp4dmzZoVOV7tY+IpyvPnz1G9enV8+eWXaNCgAfT19XHy5ElERUXJ3PGkD8eEiCqk+vXr49ixY5gwYQJmzZqF6tWrY/bs2UhOTi40gPXXX3+Fqakptm7din379qFt27Y4fPhwoUGdRRkwYACsrKywaNEi/Pjjj8jOzka1atXQsmXLYp+qKWBtbY1Tp05h7NixWLBgAUxNTTF69Gjo6elh7NixMk9Zfcx1irN+/Xo0btwYGzZswPfff49KlSrB1tYWX331Fdzd3d97fN26dbFr1y7MmDEDkyZNgoWFBUaOHAlTU1N8/fXXMnVnzZqFe/fuYcmSJXj+/Dk8PDyKTT4GDx4MXV1dLFq0CFOnThUnBly8eLFCx5GVJqYCK1euhLq6OrZu3YqsrCy4u7vj5MmTRT6BVx7U1dVx+PBhfPfdd9i8eTPU1NTQs2dPBAQEwN3dXeY79bECAwNhamqKNWvWYPz48TAxMcHw4cOxYMGCCrHsSN++fXHy5EkYGBigc+fO762vpqaGAwcOYNasWdi+fTuCg4Nha2uLH3/8ERMnTizx2O+//x4vX75EaGgotm/fjkaNGuHw4cOYNm2aTD0NDQ1s3rwZ06dPx4gRI5CXl4fg4OAiE6KPiacourq6GDVqFI4fP449e/ZAKpWiVq1aWLduXYlzdpH8JMKHjrIjoiKNGzcOGzZswIsXLz7r5Ruo7Ozbtw89e/bE2bNn5Up2iaj0OIaI6CO8uybY06dP8dtvv6FFixZMhuiDvPudys/Px+rVq2FoaMjZiImUiF1mRB/Bzc0NrVu3hoODAx49eoRNmzYhMzMTM2fOLO/Q6BM1ZswYvH79Gm5ubsjOzsaePXtw/vx5LFiw4IOfKCSi92OXGdFH+P7777Fr1y78999/kEgkaNSoEQICAsr9sWX6dIWGhmLZsmW4desWsrKyUKtWLYwcORL+/v7lHRrRZ40JEREREak8jiEiIiIilceEiIiIiFQeB1XLQSqV4uHDhzAwMFDYxGhERESkXIIg4Pnz57CysoKaWsn3gJgQyeHhw4dyTfJHREREFc/9+/dRvXr1EuswIZJDwcJ89+/fh6GhYTlHQ0RERPLIzMyEtbW1zAK7xWFCJIe316RiQkRERPRpkWe4CwdVExERkcpjQkREREQqjwkRERERqTyOIVKg/Px85ObmlncYRKRAGhoaXKiXSAUwIVIAQRCQkpKC9PT08g6FiJTA2NgYFhYWnIeM6DPGhEgBCpIhMzMz6Orq8j9Nos+EIAh49eoVUlNTAQCWlpblHBERKQsToo+Un58vJkNVqlQp73CISMF0dHQAAKmpqTAzM2P3GdFnioOqP1LBmCFdXd1yjoSIlKXg55tjBIk+X0yIFITdZESfL/58E33+mBARERGRymNCRJ+NwYMHw9vbW9xu3bo1xo0bJ27b2tpixYoVcp2rNHWJiOjTx0HVSmQ77XCZXu/uoi4KOc/gwYORnp6Offv2KeR8FUVUVBT09PQUXpeIiD59TIhIZZiamiqlLhERffrYZabCdu3aBWdnZ+jo6KBKlSrw9PTE5MmTsXnzZuzfvx8SiQQSiQQREREAgPv378PHxwfGxsYwMTFBjx49cPfuXfF8BV1Ws2fPhqmpKQwNDTFixAjk5OTIFY9UKsWSJUtQq1YtaGlpoUaNGpg/f75YHhcXh7Zt24rxDh8+HC9evJC7vW93gwmCgMDAQNSoUQNaWlqwsrLC2LFji6wLAElJSejRowf09fVhaGgIHx8fPHr0SCwPDAyEi4sLfvvtN9ja2sLIyAj9+vXD8+fPS3y/X758KXf8RESkPEyIVFRycjL69++Pr7/+GvHx8YiIiECvXr0QEBAAHx8fdOzYEcnJyUhOTkbz5s2Rm5sLLy8vGBgY4K+//sK5c+egr6+Pjh07yiQ84eHh4vn++OMP7NmzB7Nnz5YrpunTp2PRokWYOXMmrl+/jtDQUJibmwMAXr58CS8vL1SuXBlRUVHYuXMnTp48CX9//w9q/+7du/HTTz9hw4YNuHnzJvbt2wdnZ+ci60qlUvTo0QNpaWk4ffo0Tpw4gTt37qBv374y9W7fvo19+/bh0KFDOHToEE6fPo1FixYBKP79FgThg+InIiLFYpeZikpOTkZeXh569eoFGxsbABATAh0dHWRnZ8PCwkKs//vvv0MqleKXX34RH0EODg6GsbExIiIi0KFDBwCApqYmfv31V+jq6sLJyQlz5szB5MmTMXfuXKipFZ9/P3/+HCtXrsSaNWvg5+cHAKhZsyZatGgBAAgNDUVWVha2bNkiju1Zs2YNunXrhsWLF4uJk7ySkpJgYWEBT09PaGhooEaNGmjatGmRdcPDwxEXF4fExERYW1sDALZs2QInJydERUWhSZMmAN4kTiEhITAwMAAADBw4EOHh4Zg/f36J73dppN7LLPUxpWFmY6jU8xMRVVS8Q6SiGjRogHbt2sHZ2Rl9+vTBxo0b8ezZs2Lrx8bG4tatWzAwMIC+vj709fVhYmKCrKws3L59W+a8b09S6ebmhhcvXuD+/fslxhMfH4/s7Gy0a9eu2PIGDRrIDHR2d3eHVCpFQkKCvM0W9enTB69fv4a9vT2GDRuGvXv3Ii8vr9hrW1tbi8kQADg6OsLY2Bjx8fHiPltbWzEZAt4s81Cw5ENp328iIipbTIhUlLq6Ok6cOIGjR4/C0dERq1evRt26dZGYmFhk/RcvXqBx48aIiYmRef37778YMGDAR8dTsDxCWbG2tkZCQgLWrVsHHR0djBo1Cq1atfqomYg1NDRktiUSCaRSKYDSv99ERFS2mBCpMIlEAnd3d8yePRt///03NDU1sXfvXmhqaiI/P1+mbqNGjXDz5k2YmZmhVq1aMi8jIyOxXmxsLF6/fi1uX7hwAfr6+jJ3V4pSu3Zt6OjoIDw8vMhyBwcHxMbGygxCPnfuHNTU1FC3bt0PaT50dHTQrVs3rFq1ChEREYiMjERcXFyR175//77MXa7r168jPT0djo6Ocl+vuPebiIjKX7kmRGfOnEG3bt1gZWUFiURSaN4bQRAwa9YsWFpaQkdHB56enrh586ZMnbS0NPj6+sLQ0BDGxsYYOnRooSePrl69ipYtW0JbWxvW1tZYsmSJsptW4V28eBELFizA5cuXkZSUhD179uDx48dwcHCAra0trl69ioSEBDx58gS5ubnw9fVF1apV0aNHD/z1119ITExEREQExo4di//++088b05ODoYOHYrr16/jyJEjCAgIgL+/f4njhwBAW1sbU6dOxZQpU7Blyxbcvn0bFy5cwKZNmwAAvr6+0NbWhp+fH/755x+cOnUKY8aMwcCBA0s9fggAQkJCsGnTJvzzzz+4c+cOfv/9d+jo6Ijje97m6ekJZ2dn+Pr64sqVK7h06RIGDRoEDw8PuLq6ynW9kt5vIiIqf+WaEL18+RINGjTA2rVriyxfsmQJVq1ahfXr1+PixYvQ09ODl5cXsrKyxDq+vr64du0aTpw4gUOHDuHMmTMYPny4WJ6ZmYkOHTrAxsYG0dHR+PHHHxEYGIiff/5Z6e2ryAwNDXHmzBl07twZderUwYwZM7Bs2TJ06tQJw4YNQ926deHq6gpTU1OcO3cOurq6OHPmDGrUqIFevXrBwcEBQ4cORVZWFgwN/28gbrt27VC7dm20atUKffv2Rffu3REYGChXTDNnzsTEiRMxa9YsODg4oG/fvuIYHF1dXRw7dgxpaWlo0qQJvvzyS7Rr1w5r1qz5oPYbGxtj48aNcHd3R/369XHy5EkcPHgQVapUKVRXIpFg//79qFy5Mlq1agVPT0/Y29tj+/btcl+vpPebiIjKn0SoIM/9SiQS7N27V1x6QRAEWFlZYeLEiZg0aRIAICMjA+bm5ggJCUG/fv0QHx8PR0dHREVFiX+ph4WFoXPnzvjvv/9gZWWFoKAg/PDDD0hJSYGmpiYAYNq0adi3bx9u3LghV2yZmZkwMjJCRkaGzC9/AMjKykJiYiLs7Oygra2toHfj0/S5znBdkfAps/LBn3OiT1NJv7/fVWHHECUmJiIlJQWenp7iPiMjIzRr1gyRkZEAgMjISBgbG8t0W3h6ekJNTQ0XL14U67Rq1UpMhgDAy8sLCQkJxT7lk52djczMTJkXERERfb4qbEKUkpICAIXGh5ibm4tlKSkpMDMzkymvVKkSTExMZOoUdY63r/GuhQsXwsjISHy9b0AwvV9SUpL4uH5Rr6SkpPIOkYiIVBgnZizC9OnTMWHCBHE7MzOTSZEcQkJCii2zsrJCTExMieVERETlpcImRAWzJD969AiWlpbi/kePHsHFxUWsUzDotkBeXh7S0tLE4y0sLGTWnCo4x9vXeJeWlha0tLQU0g56o1KlSqhVq1Z5h0FERFSkCttlZmdnBwsLC5l5aTIzM3Hx4kW4ubkBeDMLcnp6OqKjo8U6f/75J6RSKZo1aybWOXPmjMyEeydOnEDdunVRuXLlMmoNERERVWTlmhC9ePFCnPEYeDOQOiYmBklJSZBIJBg3bhzmzZuHAwcOIC4uDoMGDYKVlZX4JJqDgwM6duyIYcOG4dKlSzh37hz8/f3Rr18/sQtmwIAB0NTUxNChQ3Ht2jVs374dK1eulOkSIyIiItVWrl1mly9fRps2bcTtgiTFz88PISEhmDJlCl6+fInhw4cjPT0dLVq0QFhYmMxjr1u3boW/vz/atWsHNTU19O7dG6tWrRLLjYyMcPz4cYwePRqNGzdG1apVMWvWLJm5ioiIiEi1VZh5iCoyzkNEFQXnISof/Dkn+jR9FvMQEREREZUVJkREHygiIgISiQTp6enlHQoREX2kCvvY/Wch0Oj9dRR6vYyyvVxgIPbt21fi/EJlrXXr1nBxccGKFSvKOxQiIvqE8A4RqRxBEJCXl1feYRARUQXChEiFtW7dGmPHjsWUKVNgYmICCwsLmZXpk5KS0KNHD+jr68PQ0BA+Pj7ipJYhISGYPXs2YmNjIZFIIJFISpypWp5zAm/uOrm4uOC3336Dra0tjIyM0K9fPzx//vy95x48eDBOnz6NlStXijHdvXtX7No6evQoGjduDC0tLZw9exa3b99Gjx49YG5uDn19fTRp0gQnT56UOWd2djamTp0Ka2traGlpoVatWti0aVOR13/16hU6deoEd3d3dqMREX1imBCpuM2bN0NPTw8XL17EkiVLMGfOHJw4cQJSqRQ9evRAWloaTp8+jRMnTuDOnTvo27cvAKBv376YOHEinJyckJycjOTkZLGsOO87Z4Hbt29j3759OHToEA4dOoTTp09j0aJF723LypUr4ebmhmHDhokxvb3kyrRp07Bo0SLEx8ejfv36ePHiBTp37ozw8HD8/fff6NixI7p16yazrtqgQYPwxx9/YNWqVYiPj8eGDRugr69f6Nrp6elo3749pFIpTpw4AWNj4/fGS0REFQfHEKm4+vXrIyAgAABQu3ZtrFmzRpwdPC4uDomJiWJSsWXLFjg5OSEqKgpNmjSBvr4+KlWqVOwSKO8KDw9/7zmBN4lTSEgIDAwMAAADBw5EeHg45s+fX+L5jYyMoKmpCV1d3SJjmjNnDtq3by9um5iYoEGDBuL23LlzsXfvXhw4cAD+/v74999/sWPHDpw4cQKenp4AAHt7+0LnTUlJQd++fVG7dm2EhoZCU1NTrveDiIgqDt4hUnH169eX2ba0tERqairi4+NhbW0tc4fF0dERxsbGiI+P/6BryXtOW1tbMRl6O6aP5erqKrP94sULTJo0CQ4ODjA2Noa+vj7i4+PFO0QxMTFQV1eHh4dHiedt3749atWqhe3btzMZIiL6RDEhUnEaGhoy2xKJBFKptJyieUNZMenp6clsT5o0CXv37sWCBQvw119/ISYmBs7OzsjJyQEA6OjoyHXeLl264MyZM7h+/fpHx0hEROWDCREVycHBAffv38f9+/fFfdevX0d6ejocHR0BAJqamsjPz1foOT9WaWI6d+4cBg8ejJ49e8LZ2RkWFha4e/euWO7s7AypVIrTp0+XeJ5FixbBz88P7dq1Y1JERPSJYkJERfL09ISzszN8fX1x5coVXLp0CYMGDYKHh4fY9WRraysuyPvkyRNkZ2d/9Dk/lq2tLS5evIi7d+/iyZMnJd5Zql27Nvbs2YOYmBjExsZiwIABMvVtbW3h5+eHr7/+Gvv27UNiYiIiIiKwY8eOQudaunQpfH190bZtW9y4cUMhbSEiorLDhIiKJJFIsH//flSuXBmtWrWCp6cn7O3tsX37drFO79690bFjR7Rp0wampqb4448/PvqcH2vSpElQV1eHo6MjTE1NZZ4Ye9fy5ctRuXJlNG/eHN26dYOXlxcaNWokUycoKAhffvklRo0ahXr16mHYsGF4+fJlkef76aef4OPjg7Zt2+Lff/9VWJuIiEj5uLirHLi4K1UUXNy1fPDnnOjTxMVdiYiIiEqBCREpzNatW6Gvr1/ky8nJ6aPPn5SUVOz59fX1S+weIyIiKgknZiSF6d69O5o1a1Zk2buP0n8IKyurEheStbKy+uhrEBGRamJCRApjYGAgM6GiolWqVAm1atVS2vmJiEh1scuMiIiIVB4TIiIiIlJ5TIiIiIhI5TEhIiIiIpXHhIiIiIhUHhMiKlJgYCBcXFzkrn/37l1IJJISH4uPiIiARCJBenr6R8cnz/Xe1bp1a4wbN+6jr62o8xARUcXBx+6VyHmzc5leL84vTmHnmjRpEsaMGaOw8ymatbU1kpOTUbVqVaVdIyIiAm3atMGzZ89gbGws7t+zZ49C5lUiIqKKgwkRFalg9ueKSl1dHRYWFuVybRMTk3K5LhERKQ+7zFTUli1bUKVKFWRnZ8vs9/b2xsCBA4vsMvvll1/g4OAAbW1t1KtXD+vWrSvxGkeOHEGdOnWgo6ODNm3a4O7du3LFlpmZCR0dHRw9elRm/969e2FgYIBXr14V2WV2+vRpNG3aFFpaWrC0tMS0adOQl5dX7HV+++03uLq6wsDAABYWFhgwYABSU1MBvOmSa9OmDQCgcuXKkEgkGDx4MIDCXWbPnj3DoEGDULlyZejq6qJTp064efOmWB4SEgJjY2McO3YMDg4O0NfXR8eOHZGcnCzWiYiIQNOmTaGnpwdjY2O4u7vj3r17cr1fRET08ZgQqag+ffogPz8fBw4cEPelpqbi8OHD+PrrrwvV37p1K2bNmoX58+cjPj4eCxYswMyZM7F58+Yiz3///n306tUL3bp1Q0xMDL755htMmzZNrtgMDQ3RtWtXhIaGForB29sburq6hY558OABOnfujCZNmiA2NhZBQUHYtGkT5s2bV+x1cnNzMXfuXMTGxmLfvn24e/eumPRYW1tj9+7dAICEhAQkJydj5cqVRZ5n8ODBuHz5Mg4cOIDIyEgIgoDOnTsjNzdXrPPq1SssXboUv/32G86cOYOkpCRMmjQJAJCXlwdvb294eHjg6tWriIyMxPDhwyGRSOR6v4iI6OOxy0xF6ejoYMCAAQgODkafPn0AAL///jtq1KiB1q1b4/Tp0zL1AwICsGzZMvTq1QsAYGdnh+vXr2PDhg3w8/MrdP6goCDUrFkTy5YtAwDUrVsXcXFxWLx4sVzx+fr6YuDAgXj16hV0dXWRmZmJw4cPY+/evUXWX7duHaytrbFmzRpIJBLUq1cPDx8+xNSpUzFr1iyoqRXO/d9O/Ozt7bFq1So0adIEL168gL6+vtg1ZmZmJjOG6G03b97EgQMHcO7cOTRv3hzAm8TN2toa+/btE9/b3NxcrF+/HjVr1gQA+Pv7Y86cOQDe3BHLyMhA165dxXIHBwe53iciIlIM3iFSYcOGDcPx48fx4MEDAG+6dgYPHlzozsTLly9x+/ZtDB06VGZ1+Xnz5uH27dtFnjs+Pr7QQq9ubm5yx9a5c2doaGiId7B2794NQ0NDeHp6Fns9Nzc3mdjd3d3x4sUL/Pfff0UeEx0djW7duqFGjRowMDCAh4cHACApKUnuOOPj41GpUiWZtlapUgV169ZFfHy8uE9XV1dMdgDA0tJS7J4zMTHB4MGD4eXlhW7dumHlypUy3WlERKR8TIhUWMOGDdGgQQNs2bIF0dHRuHbtmthl9LYXL14AADZu3IiYmBjx9c8//+DChQtKiU1TUxNffvml2G0WGhqKvn37olIlxdzUfPnyJby8vGBoaIitW7ciKipKvPuUk5OjkGu87d2n0iQSCQRBELeDg4MRGRmJ5s2bY/v27ahTp47S3lsiIiqMCZGK++abbxASEoLg4GB4enrC2tq6UB1zc3NYWVnhzp07qFWrlszLzs6uyPM6ODjg0qVLMvtK+wve19cXYWFhuHbtGv7880/4+voWW9fBwUEcv1Pg3LlzMDAwQPXq1QvVv3HjBp4+fYpFixahZcuWqFevnnjHpoCmpiYAID8/v8Tr5uXl4eLFi+K+p0+fIiEhAY6OjnK3FXiToE6fPh3nz5/HF198UWgMFRERKQ8TIhU3YMAA/Pfff9i4cWORg6kLzJ49GwsXLsSqVavw77//Ii4uDsHBwVi+fHmR9UeMGIGbN29i8uTJSEhIQGhoKEJCQkoVW6tWrWBhYQFfX1/Y2dkV6oJ726hRo3D//n2MGTMGN27cwP79+xEQEIAJEyYUOX6oRo0a0NTUxOrVq3Hnzh0cOHAAc+fOlaljY2MDiUSCQ4cO4fHjx+KdsrfVrl0bPXr0wLBhw3D27FnExsbiq6++QrVq1dCjRw+52pmYmIjp06cjMjIS9+7dw/Hjx3Hz5k2OIyIiKkNMiFSckZERevfuDX19fXh7exdb75tvvsEvv/yC4OBgODs7w8PDAyEhIcXeIapRowZ2796Nffv2oUGDBli/fj0WLFhQqtgkEgn69++P2NjYEu8OAUC1atVw5MgRXLp0CQ0aNMCIESMwdOhQzJgxo8j6pqamCAkJwc6dO+Ho6IhFixZh6dKlhc45e/ZsTJs2Debm5vD39y/yXMHBwWjcuDG6du0KNzc3CIKAI0eOyD15o66uLm7cuIHevXujTp06GD58OEaPHo1vv/1WruOJiOjjSYS3+xioSJmZmTAyMkJGRgYMDQ1lyrKyspCYmAg7Oztoa2uXU4Qfp127dnBycsKqVavKOxR6j9R7mUo9v5mN4fsrqaDP4eecSBWV9Pv7XXzsXoU9e/YMERERiIiIeO8ki0RERJ8zdpmpsIYNG2Lw4MFYvHgx6tatW6bX7tSpk8wj/G+/Stu1RkRE9LF4h0iFybuUhjL88ssveP36dZFlXCuMiIjKGhMiKhfVqlUr7xCIiIhE7DIjIiIilceEiIiIiFQeEyIiIiJSeUyIiIiISOUxISIiIiKVx4SISIECAwPh4uKilHNHRETA3NYIGRnpSjn/+9ja2mLFihVy11fUe6HM95SIqAAfu1ei+Hpluzinw434Mr0eqZaoqCjo6ekp9RoSiQR79+6VWVdv0qRJGDNmjFKvS0SKsXbEn0q/xuj1bZVyXiZERP9fTk4ONDU1Vea6pWVqalou1y2YwZyISJnYZabCiuoCcXFxQWBgIIA3f63/8ssv6NmzJ3R1dVG7dm0cOHBArBsREQGJRILw8HC4urpCV1cXzZs3R0JCgsw5Dx48iCZNmkBbWxtVq1ZFz549xbLs7GxMmjQJ1apVg56eHpo1a4aIiAix/N69e+jWrRsqV64MPT09ODk54ciRIwDerMXm6+sLU1NT6OjooHbt2ggODhaPnTp1KurUqQNdXV3Y29tj5syZyM3NFcsLumJ++eUXcdHOLVu2oEqVKsjOzpZpg7e3NwYOHCj3e7thwwZYW1tDV1cXPj4+yMjIEMsGDx4Mb29vzJ8/H1ZWVuKyKb/99htcXV1hYGAACwsLDBgwAKmpqcVe49XrV+jv1xtde3cQu9F+37YZLdo1QY06ZnBv64rg3zbKFW+XXu0xd+EsmX2PHz+GhoYGzpw5A6Dw9yUpKQk9evSAvr4+DA0N4ePjg0ePHhV7jaioKLRv3x5Vq1aFkZERPDw8cOXKFbHc1tYWANCzZ09IJBJx+90uM6lUijlz5qB69erQ0tKCi4sLwsLCxPK7d+9CIpFgz549aNOmDXR1ddGgQQNERkaKdUr6XhGRamJCRCWaPXs2fHx8cPXqVXTu3Bm+vr5IS0uTqfPDDz9g2bJluHz5MipVqoSvv/5aLDt8+DB69uyJzp074++//0Z4eDiaNm0qlvv7+yMyMhLbtm3D1atX0adPH3Ts2BE3b94EAIwePRrZ2dk4c+YM4uLisHjxYvFuwcyZM3H9+nUcPXoU8fHxCAoKQtWqVcVzGxgYICQkBNevX8fKlSuxceNG/PTTTzKx37p1C7t378aePXsQExODPn36ID8/XybxS01NxeHDh2XaVZJbt25hx44dOHjwIMLCwvD3339j1KhRMnXCw8ORkJCAEydO4NChQwCA3NxczJ07F7Gxsdi3bx/u3r2LwYMHF3mNjIx0+HzlDalUih2/74ORkTF27duBJcsXYPrkmfgr/BK+nzILi5fNx/Zdoe+Nube3D/Yd2gNBEMR927dvh5WVFVq2bFmovlQqRY8ePZCWlobTp0/jxIkTuHPnDvr27VvsNZ4/fw4/Pz+cPXsWFy5cQO3atdG5c2c8f/4cwJuECQCCg4ORnJwsbr9r5cqVWLZsGZYuXYqrV6/Cy8sL3bt3F78zBX744QdMmjQJMTExqFOnDvr374+8vDwAJX+viEg1scuMSjR48GD0798fALBgwQKsWrUKly5dQseOHcU68+fPh4eHBwBg2rRp6NKlC7KysqCtrY358+ejX79+mD17tli/QYMGAN7cYQgODkZSUhKsrKwAvBkvEhYWhuDgYCxYsABJSUno3bs3nJ2dAQD29vbieZKSktCwYUO4uroC+L87DAVmzJgh/tvW1haTJk3Ctm3bMGXKFHF/Tk4OtmzZItMdNGDAAAQHB6NPnz4AgN9//x01atRA69at5XrPsrKysGXLFnF5ktWrV6NLly5YtmwZLCwsAAB6enr45ZdfZLrK3k647O3tsWrVKjRp0gQvXryQ+WWd+jgVw/2HwN7OHkErN4nn+PGnBQj8YT66dOwOALCxtkXCzQRsCQ1G3y8HlBhz9y49MXPONJw9e1ZMgEJDQ9G/f39IJJJC9cPDwxEXF4fExERYW1sDALZs2QInJydERUWhSZMmhY5p21a23//nn3+GsbExTp8+ja5du4qfgbGxsfg+FWXp0qWYOnUq+vXrBwBYvHgxTp06hRUrVmDt2rVivUmTJqFLly4A3iT2Tk5OuHXrFurVq1fi94qIVBPvEFGJ6tevL/5bT08PhoaGhbpx3q5jaWkJAGKdmJgYtGvXrshzx8XFIT8/H3Xq1JFZ7f706dO4ffs2AGDs2LGYN28e3N3dERAQgKtXr4rHjxw5Etu2bYOLiwumTJmC8+fPy5x/+/btcHd3h4WFBfT19TFjxgwkJSXJ1LGxsSk0NmbYsGE4fvw4Hjx4AAAICQnB4MGDi0wMilKjRg2Ztdrc3NwglUpluhKdnZ0LjRuKjo5Gt27dUKNGDRgYGIhJ5rsx+wz0hp2tHX5eEyKe4+Wrl7h7LxETpvrDztFKfK1Y/SPuJiW+N+aqVaqidcu22Lp1KwAgMTERkZGR8PX1LbJ+fHw8rK2txWQIABwdHWFsbIz4+KIH9z969AjDhg1D7dq1YWRkBENDQ7x48aJQ+0qSmZmJhw8fwt3dXWa/u7t7oeuW9L0s6XtFRKqJCZEKU1NTk+kiASAzxgYANDQ0ZLYlEgmkUmmxdQqShoI6Ojo6xV7/xYsXUFdXR3R0NGJiYsRXfHw8Vq5cCQD45ptvcOfOHQwcOBBxcXFwdXXF6tWrAQCdOnXCvXv3MH78eDx8+BDt2rXDpEmTAED8Zd65c2ccOnQIf//9N3744Qfk5OTIxFDUU1MNGzZEgwYNsGXLFkRHR+PatWvFdl19qHev+/LlS3h5ecHQ0BBbt25FVFQU9u7dCwCFYvZs0wEXLp1Hws0bMscDwNJFq/Dnkb/E1+njkTiy96RcMfXy9sGuXbuQm5uL0NBQODs7i3dQFMHPzw8xMTFYuXIlzp8/j5iYGFSpUqVQ+xSlpO9lSd8rIlJN7DJTYaampkhOTha3MzMzkZj4/rsJpVG/fn2Eh4djyJAhhcoaNmyI/Px8pKamFjlOpYC1tTVGjBiBESNGYPr06di4caP4GLapqSn8/Pzg5+eHli1bYvLkyVi6dCnOnz8PGxsb/PDDD+J57t27J3fc33zzDVasWIEHDx7A09NT5k7I+yQlJeHhw4diN+CFCxegpqYmDp4uyo0bN/D06VMsWrRIvNbly5eLrDtjaiD09PTwpW937N12GHVr14OZqRkszC2RlHQXX3r7yB3r2zq274zJ33+HsLAwhIaGYtCgQcXWdXBwwP3793H//n0x3uvXryM9PR2Ojo5FHnPu3DmsW7cOnTt3BgDcv38fT548kamjoaGB/Pz8Yq9raGgIKysrnDt3TryDVnDut8emyaOk7xV9Wj7lR72p4mBCpMLatm2LkJAQdOvWDcbGxpg1axbU1dUVeo2AgAC0a9cONWvWRL9+/ZCXl4cjR46IT4D5+vpi0KBBWLZsGRo2bIjHjx8jPDwc9evXR5cuXTBu3Dh06tQJderUwbNnz3Dq1Ck4OLyZ32nWrFlo3LgxnJyckJ2djUOHDolltWvXRlJSErZt24YmTZrg8OHD4h0XeQwYMACTJk3Cxo0bsWXLllK1WVtbG35+fli6dCkyMzMxduxY+Pj4lDgupkaNGtDU1MTq1asxYsQI/PPPP5g7d26x9QN/mI/8fCl69++GvdsOo3atOpg8fjpmBE6FgYEh2np4IjsnG7FX/0ZGZjpGfOP/3rj1dPXg7e2NmTNnIj4+Xhw7VhRPT084OzvD19cXK1asQF5eHkaNGgUPDw9xTNe7ateuLT5Jl5mZicmTJxe6g2hra4vw8HC4u7tDS0sLlStXLnSeyZMnIyAgADVr1oSLiwuCg4MRExMjdvfJo6TvFRGpJnaZqbDp06fDw8MDXbt2RZcuXeDt7Y2aNWsq9BqtW7fGzp07ceDAAbi4uKBt27a4dOmSWB4cHIxBgwZh4sSJqFu3Lry9vREVFYUaNWoAAPLz8zF69Gg4ODigY8eOqFOnDtatWwcA0NTUxPTp01G/fn20atUK6urq2LZtGwCge/fuGD9+PPz9/eHi4oLz589j5syZcsdtZGSE3r17Q19fX2aSQHnUqlULvXr1QufOndGhQwfUr19fjLk4pqamCAkJwc6dO+Ho6IhFixZh6dKlJR4zd9ZCdO/aE70HdMPtO7fwVT8/LFu8Gtt2/o7WHd3Qs29nbN8VihrVbeSO3dfXF7GxsWjZsqX4GRRFIpFg//79qFy5Mlq1agVPT0/Y29tj+/btxR6zadMmPHv2DI0aNcLAgQMxduxYmJmZydRZtmwZTpw4AWtrazRs2LDI84wdOxYTJkzAxIkT4ezsjLCwMBw4cAC1a9eWu50lfa+ISDVJhHcHkVAhmZmZMDIyQkZGBgwNDWXKsrKykJiYKM5jQ5+Pdu3awcnJCatWrSrvUESp9zKVen4zG8P3V1JB/Dmv2NhlVnFUtM+ipN/f72KXGdE7nj17hoiICERERPCuARGRimBCRPSOhg0b4tmzZ1i8eHGhgdBOTk7FDs7esGFDsY+pl7cVa5di5drlRZb9r4kb/ti8u4wjIiKqWCp0QpSfn4/AwED8/vvvSElJgZWVFQYPHowZM2aIj9EKgoCAgABs3LgR6enpcHd3R1BQkMx4grS0NIwZMwYHDx6EmpoaevfujZUrV3JmWirS3bt3iy07cuRIoakJCpibmyspoo/n5/s1enTpWWSZtnbxUyMQEamKCp0QLV68GEFBQdi8eTOcnJxw+fJlDBkyBEZGRhg7diwAYMmSJVi1ahU2b94MOzs7zJw5E15eXrh+/brY1+/r64vk5GScOHECubm5GDJkCIYPH47Q0PcvaUD0Nhsb+QcoVySVjU1Q2dikvMMgIqqwKnRCdP78efTo0UOcft/W1hZ//PGH+JSSIAhYsWIFZsyYgR49egB4s3yAubk59u3bh379+iE+Ph5hYWGIiooSHwdevXo1OnfujKVLl4pzxRAREZHqqtCP3Tdv3hzh4eH4999/AQCxsbE4e/YsOnXqBODN8gIpKSnw9PQUjzEyMkKzZs3Ela0jIyNhbGwsMzeKp6cn1NTUcPHixTJsDREREVVUFfoO0bRp05CZmYl69epBXV0d+fn5mD9/vjhwNSUlBUDhsRvm5uZiWUpKSqG5TipVqgQTExOxzruys7ORnZ0tbmdmKvdRZyIiIipfFfoO0Y4dO7B161aEhobiypUr2Lx5M5YuXYrNmzcr9boLFy6EkZGR+CrNsg1ERET06anQCdHkyZMxbdo09OvXD87Ozhg4cCDGjx+PhQsXAoC4FMKjR49kjnv06JFYZmFhUWh19ry8PKSlpRW7lML06dORkZEhvu7fv6/ophEREVEFUqETolevXkFNTTZEdXV1ccVqOzs7WFhYIDw8XCzPzMzExYsX4ebmBgBwc3NDeno6oqOjxTp//vknpFIpmjVrVuR1tbS0YGhoKPNSVefOnYOzszM0NDTEJSze3RcREQGJRIL09HS5ztm6dWuMGzdOaTF/ikr7HpaFwMBAuLi4yF3/7t27kEgkiImJ+ajrKuo8RESlUaHHEHXr1g3z589HjRo14OTkhL///hvLly/H119/DeDNekrjxo3DvHnzULt2bfGxeysrK/GXd8FaRcOGDcP69euRm5sLf39/9OvXT+lPmJXFFOZvU8bU8hMmTICLiwuOHj0qztv07j5dXV0kJyfDyMhIrnPu2bMHGhoaCo1z8ODBSE9Px759+xR63rLSvHnzUr2HZWHSpElKX/29qM/N2toaycnJqFq1qlKvTUT0tgqdEK1evRozZ87EqFGjkJqaCisrK3z77beYNWuWWGfKlCl4+fIlhg8fjvT0dLRo0QJhYWEy6w1t3boV/v7+aNeunTgxY0Van6oiu337NkaMGIHq1auXuK+kldzfZWLy6c2Hk5OTA01NTaWdX1NTs1TvYVnQ19cvl8lL1dXVK9x7QUSfvwrdZWZgYIAVK1bg3r17eP36NW7fvo158+bJ/GKSSCSYM2cOUlJSkJWVhZMnT6JOnToy5zExMUFoaCieP3+OjIwM/Prrr5yl+v+TSqVYuHAh7OzsoKOjgwYNGmDXrl1it8XTp0/x9ddfQyKRICQkpMh9RXX3nDt3Dq1bt4auri4qV64MLy8vPHv2DEDhLrPs7GxMmjQJ1apVg56eHpo1a4aIiAixPCQkBMbGxjh27BgcHBygr6+Pjh07Ijk5GcCbrp3Nmzdj//79kEgkkEgkMscX57///kP//v1hYmICPT09uLq6ilMxFHQX/fLLLzILeiYlJaFHjx7Q19eHoaEhfHx8ZMawxcbGok2bNjAwMIChoSEaN26My5cvAwDu3buHbt26oXLlytDT04OTkxOOHDkCoHCXWXFtfpT6f09G5uXl4fvAKajtXAP1XGwxd+EsjJkwAn7DBry37VtCg1G/aV2x+7lAjx49xDuw73aZSaVSzJkzB9WrV4eWlhZcXFwQFhZW7DXy8/MxdOhQ8btVt25drFy5Uiwv7nMrqsvs9OnTaNq0KbS0tGBpaYlp06YhLy9PLG/dujXGjh2LKVOmwMTEBBYWFggMDBTLBUFAYGAgatSoAS0tLVhZWYmTuxIRARU8ISLlW7hwIbZs2YL169fj2rVrGD9+PL766ivcu3cPycnJMDQ0xIoVK5CcnIw+ffoU2te3b99C54yJiUG7du3g6OiIyMhInD17Ft26dUN+fn6RMfj7+yMyMhLbtm3D1atX0adPH3Ts2BE3b94U67x69QpLly7Fb7/9hjNnziApKQmTJk0C8KZrx8fHR0ySkpOT0bx58xLb/eLFC3h4eODBgwc4cOAAYmNjMWXKFJkE4datW9i9ezf27NmDmJgYSKVS9OjRA2lpaTh9+jROnDiBO3fuyLwHvr6+qF69OqKiohAdHY1p06aJ3YOjR49GdnY2zpw5g7i4OCxevLjExLyoNgfOnyGWr17/E/bs24GVP67FwV3H8fzFcxw9cbjEdhfo3sUbz9LTcDbyjLjvWXoawsLCil2PbeXKlVi2bBmWLl2Kq1evwsvLC927d5f5nN4mlUpRvXp17Ny5E9evX8esWbPw/fffY8eOHQDk/9wePHiAzp07o0mTJoiNjUVQUBA2bdqEefPmydTbvHkz9PT0cPHiRSxZsgRz5szBiRMnAAC7d+/GTz/9hA0bNuDmzZvYt28fnJ2d5XqviEg1VOguM1Ku7OxsLFiwACdPnhQHodvb2+Ps2bPYsGEDQkNDIZFIYGRkJHZh6OnpFdr3riVLlsDV1VVmpXgnJ6ci6yYlJSE4OBhJSUnimK5JkyYhLCwMwcHBWLBgAQAgNzcX69evR82aNQG8SaLmzJkD4E3Xjo6ODrKzs+XuagkNDcXjx48RFRUlduHVqlVLpk5OTg62bNkCU1NTAMCJEycQFxeHxMREcSqGLVu2wMnJCVFRUWjSpAmSkpIwefJk1KtXDwBk1tRLSkpC7969xV/E9vb2JcZYVJsDA2aL5ZtCfsbYURPQuWM3AMDCOUsRHnFCrvYbG1VGW4/22LN/J1q5twYAHDyyH1WrVkWbNm2KPGbp0qWYOnUq+vXrB+DN0jqnTp3CihUrsHbt2kL1NTQ0MHv2/8VrZ2eHyMhI7NixAz4+PnJ/buvWrYO1tTXWrFkDiUSCevXq4eHDh5g6dSpmzZolPnhRv359BAQEAHjzvq9Zswbh4eFo3749kpKSYGFhAU9PT2hoaKBGjRpo2rSpXO8VEakG3iFSYbdu3cKrV6/Qvn17cbyIvr4+tmzZgtu3b3/weQvuEMkjLi4O+fn5qFOnjkwMp0+flolBV1dXTAwAwNLSstB0CqWNsWHDhiWOZ7KxsRGTIQCIj4+HtbW1zLxUjo6OMDY2Rnx8PIA3A86/+eYbeHp6YtGiRTJtGDt2LObNmwd3d3cEBATg6tWrJcZYVJufPH0MAMjMzMDjJ6lo2KCxWK6uro76XzSQ8x0Aenv3weGjB8VJSPfs24l+/foVerLzzfUy8fDhQ7i7u8vsd3d3F9telLVr16Jx48YwNTWFvr4+fv75ZyQlJckdI/DmfXdzcxMXdC647osXL/Dff/+J++rXry9z3NvfkT59+uD169ewt7fHsGHDsHfvXpkuNyIiJkQq7MWLFwCAw4cPIyYmRnxdv34du3bt+uDz6ujIv3r6ixcvoK6ujujoaJkY4uPjZcabvPtUmkQigSAISo1RT0+v1OcNDAzEtWvX0KVLF/z5559wdHTE3r17AQDffPMN7ty5g4EDByIuLg6urq5YvXp1sedSdJvf1aFdJwgQcPLUMTx4+B8uRJ0vtrvsQ2zbtg2TJk3C0KFDcfz4ccTExGDIkCHIyclR2DXeVtT7VdAFam1tjYSEBKxbtw46OjoYNWoUWrVqhdzcXKXEQkSfHiZEKszR0RFaWlpISkpCrVq1ZF4fMzt3/fr1ZeaGKknDhg2Rn5+P1NTUQjGU5kkjTU3NYscoFRdjTEwM0tLS5D7GwcEB9+/fl5mo8/r160hPT4ejo6O4r06dOhg/fjyOHz+OXr16ITg4WCyztrbGiBEjsGfPHkycOBEbN26U+/pvMzQ0gmlVM8RcvSLuy8/PR9w/Jd91epu2tja6eHXD7n07sPfALtSyr41GjRoVcz1DWFlZ4dy5czL7z507J9P2d8uaN2+OUaNGoWHDhqhVq1ahO4/yfG4ODg6IjIyUSQbPnTsHAwMDmScd30dHRwfdunXDqlWrEBERgcjISMTFxcl9PBF93pgQqTADAwNMmjQJ48ePx+bNm3H79m1cuXIFq1ev/qjlUaZPn46oqCiMGjUKV69exY0bNxAUFIQnT54UqlunTh34+vpi0KBB2LNnDxITE3Hp0iUsXLgQhw/LN0AYAGxtbXH16lUkJCTgyZMn7/3Lv3///rCwsIC3tzfOnTuHO3fuYPfu3eKiwEXx9PSEs7MzfH19ceXKFVy6dAmDBg2Ch4cHXF1d8fr1a/j7+yMiIgL37t3DuXPnEBUVBQcHBwDAuHHjcOzYMSQmJuLKlSs4deqUWPYhhg4ejlXrluPo8cO4dfsmZsyeivTMdJmupffp5d0HJ/88jj92/I5e3j4l1p08eTIWL16M7du3IyEhAdOmTUNMTAy+++67IuvXrl0bly9fxrFjx/Dvv/9i5syZiIqKkqkjz+c2atQo3L9/H2PGjMGNGzewf/9+BAQEYMKECUV27xUlJCQEmzZtwj///IM7d+7g999/h46ODmxsbOQ6nog+f0yIVNzcuXMxc+ZMLFy4UJzE8vDhw7Czs/vgc9apUwfHjx9HbGwsmjZtCjc3N+zfvx+VKhU9hj84OBiDBg3CxIkTUbduXXh7eyMqKgo1atSQ+5rDhg1D3bp14erqClNT00J3Mt6lqamJ48ePw8zMDJ07d4azszMWLVoEdXX1Yo+RSCTYv38/KleujFatWsHT0xP29vbYvn07gDdjeJ4+fYpBgwahTp068PHxQadOncSBxfn5+Rg9erT4PtepU0dm4HlpjRkxHt7dv8SYiSPQpZcndHX10KZVW2hpacl9jpbNPWBsXBm37txErx5fllh37NixmDBhAiZOnAhnZ2eEhYXhwIEDMgPH3/btt9+iV69e6Nu3L5o1a4anT59i1KhRMnXk+dyqVauGI0eO4NKlS2jQoAFGjBiBoUOHYsaMGYXqFsfY2BgbN26Eu7s76tevj5MnT+LgwYOoUqWK3Ocgos+bRFDkoITPVGZmJoyMjJCRkVFoGY+srCwkJibKzFVDpCyp9zKLLZNKpWjh2QTdu/TEtInyJwtvM7NR3WVqSsKf84qtLFYFUMZKAJ+jivZZlPT7+1187J7oE3X/vyRE/PUnmjdrgeycbPy6+Wck3b+HXj36lHdoRESfHCZE9FlasGCBOIfRu1q2bImjR4+WcUSKp6amhu27QjF7wUwIgoB6dRyw8/f9qFOrLv57cB8t2xe9eDEA/HXiIqpX+/CB80REnxsmRPRZGjFiBHx8ih4kXJppASqyalbVcWj38SLLLMwt8eeRv4o91sLcUllhvVdJ3X6Kwq4/+VS07g2i8sSEiD5LJiYmn+QisopSqVIl2NnWfH9FIiICwKfMiIiIiJgQKQof1iP6fPHnm+jzx4ToIxUsF/Dq1atyjoSIlKXg5/vd5UGI6PPBMUQfSV1dHcbGxuIikrq6uqWaKZioNHLzlLMOWIGsrCylnh9QfhsAxbVDEAS8evUKqampMDY2LnHiTiL6tDEhUoCCNbc+ZvV1Ink8f6rchCUzR/mTDiq7DYDi22FsbFyqtfWI6NPDhEgBJBIJLC0tYWZmJrMW09aAC0q/tu/s/yn9GlRxbA1R7nfKd/aHr60mL2W3AVBsOzQ0NHhniEgFMCFSIHV1dZn/OLMypEq/JpcRUC3K/k6VxfeJPxdEVBFxUDURERGpvFInRFeuXEFcXJy4vX//fnh7e+P7779HTo7yB0sSERERKVqpE6Jvv/0W//77LwDgzp076NevH3R1dbFz505MmTJF4QESERERKVupE6J///0XLi4uAICdO3eiVatWCA0NRUhICHbv3q3o+IiIiIiUrtQJkSAIkErfDIo8efIkOnfuDACwtrbGkydPFBsdERERURkodULk6uqKefPm4bfffsPp06fRpUsXAEBiYiLMzc0VHiARERGRspU6IVqxYgWuXLkCf39//PDDD6hVqxYAYNeuXWjevLnCAyQiIiJStlLPQ1S/fn2Zp8wK/Pjjj5y8jIiIiD5JHzwxY05ODlJTU8XxRAVq1Kjx0UERERERlaVSJ0T//vsvhg4divPnz8vsFwQBEokE+fn5CguOiIiIqCyUOiEaMmQIKlWqhEOHDsHS0pIruxMREdEnr9QJUUxMDKKjo1GvXj1lxENERERU5kr9lJmjoyPnGyIiIqLPSqkTosWLF2PKlCmIiIjA06dPkZmZKfMiIiIi+tSUusvM09MTANCuXTuZ/RxUTURERJ+qUidEp06dUkYcREREROWm1AmRh4eHMuIgIiIiKjcfNDFjeno6Nm3ahPj4eACAk5MTvv76axgZGSk0OCp/a0f8qfRrjF7fVunXICIiKkmpB1VfvnwZNWvWxE8//YS0tDSkpaVh+fLlqFmzJq5cuaKMGImIiIiUqtR3iMaPH4/u3btj48aNqFTpzeF5eXn45ptvMG7cOJw5c0bhQRIREREpU6kTosuXL8skQwBQqVIlTJkyBa6urgoNjoiIiKgslLrLzNDQEElJSYX2379/HwYGBgoJioiIiKgslToh6tu3L4YOHYrt27fj/v37uH//PrZt24ZvvvkG/fv3V0aMREREREpV6i6zpUuXQiKRYNCgQcjLywMAaGhoYOTIkVi0aJHCAyQiIiJStlInRJqamli5ciUWLlyI27dvAwBq1qwJXV1dhQdHREREVBY+aB4iANDV1YWzs7MiYyEiIiIqF3IlRL169UJISAgMDQ3Rq1evEuvu2bNHIYERERERlRW5EiIjIyNIJBIAb54yK/g3ERER0edAroQoODhY/HdISIiyYiEiIiIqF6V+7L5t27ZIT08vtD8zMxNt23JNKiIiIvr0lDohioiIQE5OTqH9WVlZ+OuvvxQSFBEREVFZkvsps6tXr4r/vn79OlJSUsTt/Px8hIWFoVq1aoqNjoiIiKgMyJ0Qubi4QCKRQCKRFNk1pqOjg9WrVys0OCIiIqKyIHdClJiYCEEQYG9vj0uXLsHU1FQs09TUhJmZGdTV1ZUSJBEREZEyyZ0Q2djYAACkUqnSgiEiIiIqDx88U/X169eRlJRUaIB19+7dPzooIiIiVbN2xJ9KPf/o9XwSvCSlToju3LmDnj17Ii4uDhKJBIIgAIA4WWN+fr5iIyQiIiJSslI/dv/dd9/Bzs4Oqamp0NXVxbVr13DmzBm4uroiIiJCCSESERERKVep7xBFRkbizz//RNWqVaGmpgY1NTW0aNECCxcuxNixY/H3338rI04iIiIipSn1HaL8/HwYGBgAAKpWrYqHDx8CeDPoOiEhQbHRAXjw4AG++uorVKlSBTo6OnB2dsbly5fFckEQMGvWLFhaWkJHRweenp64efOmzDnS0tLg6+sLQ0NDGBsbY+jQoXjx4oXCYyUiIqJPU6kToi+++AKxsbEAgGbNmmHJkiU4d+4c5syZA3t7e4UG9+zZM7i7u0NDQwNHjx7F9evXsWzZMlSuXFmss2TJEqxatQrr16/HxYsXoaenBy8vL2RlZYl1fH19ce3aNZw4cQKHDh3CmTNnMHz4cIXGSkRERJ+uUneZzZgxAy9fvgQAzJkzB127dkXLli1RpUoVbN++XaHBLV68GNbW1jKLy9rZ2Yn/FgQBK1aswIwZM9CjRw8AwJYtW2Bubo59+/ahX79+iI+PR1hYGKKiouDq6goAWL16NTp37oylS5fCyspKoTETERHRp6fUd4i8vLzQq1cvAECtWrVw48YNPHnyBKmpqQpf3PXAgQNwdXVFnz59YGZmhoYNG2Ljxo1ieWJiIlJSUuDp6SnuMzIyQrNmzRAZGQngzZgnY2NjMRkCAE9PT6ipqeHixYsKjZeIiIg+TaVOiIpiYmIiPnavSHfu3EFQUBBq166NY8eOYeTIkRg7diw2b94MAOJ6aubm5jLHmZubi2UpKSkwMzOTKa9UqRJMTExk1mN7W3Z2NjIzM2VeRERE9PmSq8us4I6QPPbs2fPBwbxLKpXC1dUVCxYsAAA0bNgQ//zzD9avXw8/Pz+FXeddCxcuxOzZs5V2fiIiIqpY5LpDZGRkJPdLkSwtLeHo6Cizz8HBAUlJSQAACwsLAMCjR49k6jx69Egss7CwQGpqqkx5Xl4e0tLSxDrvmj59OjIyMsTX/fv3FdIeIiIiqpjkukP09qDmsuTu7l7oUf5///1XXFfNzs4OFhYWCA8Ph4uLCwAgMzMTFy9exMiRIwEAbm5uSE9PR3R0NBo3bgwA+PPPPyGVStGsWbMir6ulpQUtLS0ltYqIiIgqmg9ey+zx48dislK3bl2YmpoqLKgC48ePR/PmzbFgwQL4+Pjg0qVL+Pnnn/Hzzz8DeLNcyLhx4zBv3jzUrl0bdnZ2mDlzJqysrODt7Q3gzR2ljh07YtiwYVi/fj1yc3Ph7++Pfv368QkzIiIiAvABCdHLly8xZswYbNmyBVKpFACgrq6OQYMGYfXq1dDV1VVYcE2aNMHevXsxffp0zJkzB3Z2dlixYgV8fX3FOlOmTMHLly8xfPhwpKeno0WLFggLC4O2trZYZ+vWrfD390e7du2gpqaG3r17Y9WqVQqLkyo+ZS+aCHDhRCKiT1mpE6IJEybg9OnTOHjwINzd3QEAZ8+exdixYzFx4kQEBQUpNMCuXbuia9euxZZLJBLMmTMHc+bMKbaOiYkJQkNDFRoXERERfT5KnRDt3r0bu3btQuvWrcV9nTt3ho6ODnx8fBSeEBEREREpW6nnIXr16lWheX8AwMzMDK9evVJIUERERERlqdQJkZubGwICAmTWCnv9+jVmz54NNzc3hQZHREREVBZK3WW2cuVKeHl5oXr16mjQoAEAIDY2Ftra2jh27JjCAyQiIiJStlInRF988QVu3ryJrVu34saNGwCA/v37w9fXFzo6OgoPkIiIiEjZPmgeIl1dXQwbNkzRsRARERGVC7kSogMHDqBTp07Q0NDAgQMHSqzbvXt3hQRGREREVFbkSoi8vb3FVeMLZoAuikQiQX5+vqJiIyIiIioTciVEBTNSv/tvIiIios9BqR+758rvRERE9LkpdUJka2sLDw8PbNy4Ec+ePVNGTERERERlqtQJ0eXLl9G0aVPMmTMHlpaW8Pb2xq5du5Cdna2M+IiIiIiUrtQJUcOGDfHjjz8iKSkJR48ehampKYYPHw5zc3N8/fXXyoiRiIiISKlKnRAVkEgkaNOmDTZu3IiTJ0/Czs4OmzdvVmRsRERERGXigxOi//77D0uWLIGLiwuaNm0KfX19rF27VpGxEREREZWJUs9UvWHDBoSGhuLcuXOoV68efH19sX//ftjY2CgjPiIiIiKlK3VCNG/ePPTv3x+rVq0SF3clIiIi+pSVOiFKSkqCRCJRRixERERE5aLUY4iYDBEREdHn5oMHVRMRERF9LpgQERERkcqTKyE6cOAAcnNzlR0LERERUbmQKyHq2bMn0tPTAQDq6upITU1VZkxEREREZUquhMjU1BQXLlwAAAiCwIHVRERE9FmR67H7ESNGoEePHpBIJJBIJLCwsCi2bn5+vsKCIyIiIioLciVEgYGB6NevH27duoXu3bsjODgYxsbGSg6NiIiIqGzIPTFjvXr1UK9ePQQEBKBPnz7Q1dVVZlxERBXa2hF/Kv0ao9e3Vfo1iOiNUs9UHRAQAAB4/PgxEhISAAB169aFqampYiMjIiIiKiOlnofo1atX+Prrr2FlZYVWrVqhVatWsLKywtChQ/Hq1StlxEhERESkVKVOiMaPH4/Tp0/jwIEDSE9PR3p6Ovbv34/Tp09j4sSJyoiRiIiISKlK3WW2e/du7Nq1C61btxb3de7cGTo6OvDx8UFQUJAi4yMiIiJSug/qMjM3Ny+038zMjF1mRERE9EkqdULk5uaGgIAAZGVliftev36N2bNnw83NTaHBEREREZWFUneZrVy5El5eXqhevToaNGgAAIiNjYW2tjaOHTum8ACJiIiIlK3UCdEXX3yBmzdvYuvWrbhx4wYAoH///vD19YWOjo7CAyQiIiJStlInRACgq6uLYcOGKToWIiIionJR6jFERERERJ8bJkRERESk8pgQERERkcpjQkREREQqr9QJkb29PZ4+fVpof3p6Ouzt7RUSFBEREVFZKnVCdPfuXeTn5xfan52djQcPHigkKCIiIqKyJPdj9wcOHBD/fezYMRgZGYnb+fn5CA8Ph62trUKDIyIiIioLcidE3t7eAACJRAI/Pz+ZMg0NDdja2mLZsmUKDY6IiIioLMidEEmlUgCAnZ0doqKiULVqVaUFRURERFSWSj1TdWJiojLiICIiIio3H7R0R3h4OMLDw5GamireOSrw66+/KiQwIiIiorJS6oRo9uzZmDNnDlxdXWFpaQmJRKKMuIiIiIjKTKkTovXr1yMkJAQDBw5URjxEREREZa7U8xDl5OSgefPmyoiFiIiIqFyUOiH65ptvEBoaqoxYiIiIiMpFqbvMsrKy8PPPP+PkyZOoX78+NDQ0ZMqXL1+usOCIiIiIykKpE6KrV6/CxcUFAPDPP//IlHGANREREX2KSp0QnTp1ShlxEBEREZWbUo8hIiIiIvrclPoOUZs2bUrsGvvzzz8/KiAiIiKislbqhKhg/FCB3NxcxMTE4J9//im06CsRERHRp6DUCdFPP/1U5P7AwEC8ePHiowMiIiIiKmsKG0P01VdfKX0ds0WLFkEikWDcuHHivqysLIwePRpVqlSBvr4+evfujUePHskcl5SUhC5dukBXVxdmZmaYPHky8vLylBorERERfToUlhBFRkZCW1tbUacrJCoqChs2bED9+vVl9o8fPx4HDx7Ezp07cfr0aTx8+BC9evUSy/Pz89GlSxfk5OTg/Pnz2Lx5M0JCQjBr1iylxUpERESfllJ3mb2dbACAIAhITk7G5cuXMXPmTIUF9rYXL17A19cXGzduxLx588T9GRkZ2LRpE0JDQ9G2bVsAQHBwMBwcHHDhwgX873//w/Hjx3H9+nWcPHkS5ubmcHFxwdy5czF16lQEBgZCU1NTKTETERHRp6PUd4iMjIxkXiYmJmjdujWOHDmCgIAAZcSI0aNHo0uXLvD09JTZHx0djdzcXJn99erVQ40aNRAZGQngzZ0rZ2dnmJubi3W8vLyQmZmJa9euFXm97OxsZGZmyryIiIjo81XqO0TBwcHKiKNY27Ztw5UrVxAVFVWoLCUlBZqamjA2NpbZb25ujpSUFLHO28lQQXlBWVEWLlyI2bNnKyB6IiIi+hSUOiEqEB0djfj4eACAk5MTGjZsqLCgCty/fx/fffcdTpw4odTxSe+aPn06JkyYIG5nZmbC2tq6zK5PREREZavUCVFqair69euHiIgI8c5Meno62rRpg23btsHU1FRhwUVHRyM1NRWNGjUS9+Xn5+PMmTNYs2YNjh07hpycHKSnp8vcJXr06BEsLCwAABYWFrh06ZLMeQueQiuo8y4tLS1oaWkprB1ERERUsZV6DNGYMWPw/PlzXLt2DWlpaUhLS8M///yDzMxMjB07VqHBtWvXDnFxcYiJiRFfrq6u8PX1Ff+toaGB8PBw8ZiEhAQkJSXBzc0NAODm5oa4uDikpqaKdU6cOAFDQ0M4OjoqNF4iIiL6NJX6DlFYWBhOnjwJBwcHcZ+joyPWrl2LDh06KDQ4AwMDfPHFFzL79PT0UKVKFXH/0KFDMWHCBJiYmMDQ0BBjxoyBm5sb/ve//wEAOnToAEdHRwwcOBBLlixBSkoKZsyYgdGjR/MuEBEREQH4gIRIKpVCQ0Oj0H4NDQ1IpVKFBFUaP/30E9TU1NC7d29kZ2fDy8sL69atE8vV1dVx6NAhjBw5Em5ubtDT04Ofnx/mzJlT5rESERFRxVTqhKht27b47rvv8Mcff8DKygoA8ODBA4wfPx7t2rVTeIDvioiIkNnW1tbG2rVrsXbt2mKPsbGxwZEjR5QcGREREX2qSj2GaM2aNcjMzIStrS1q1qyJmjVrws7ODpmZmVi9erUyYiQiIiJSqlLfIbK2tsaVK1dw8uRJ3LhxAwDg4OBQaNJEIiIiok/FB81DJJFI0L59e7Rv317R8RARERGVObm7zP788084OjoWuYxFRkYGnJyc8Ndffyk0OCIiIqKyIHdCtGLFCgwbNgyGhoaFyoyMjPDtt99i+fLlCg2OiIiIqCzInRDFxsaiY8eOxZZ36NAB0dHRCgmKiIiIqCzJnRA9evSoyPmHClSqVAmPHz9WSFBEREREZUnuhKhatWr4559/ii2/evUqLC0tFRIUERERUVmSOyHq3LkzZs6ciaysrEJlr1+/RkBAALp27arQ4IiIiIjKgtyP3c+YMQN79uxBnTp14O/vj7p16wIAbty4gbVr1yI/Px8//PCD0gIlIiIiUha5EyJzc3OcP38eI0eOxPTp0yEIAoA3cxJ5eXlh7dq1MDc3V1qgRERERMpSqokZC9YEe/bsGW7dugVBEFC7dm1UrlxZWfERERERKd0HzVRduXJlNGnSRNGxEBEREZWLUi/uSkRERPS5YUJEREREKo8JEREREak8JkRERESk8pgQERERkcpjQkREREQqjwkRERERqTwmRERERKTymBARERGRymNCRERERCqPCRERERGpPCZEREREpPKYEBEREZHKY0JEREREKo8JEREREak8JkRERESk8pgQERERkcpjQkREREQqjwkRERERqTwmRERERKTymBARERGRymNCRERERCqPCRERERGpPCZEREREpPKYEBEREZHKY0JEREREKo8JEREREak8JkRERESk8pgQERERkcpjQkREREQqjwkRERERqTwmRERERKTymBARERGRymNCRERERCqPCRERERGpPCZEREREpPKYEBEREZHKY0JEREREKo8JEREREak8JkRERESk8pgQERERkcpjQkREREQqjwkRERERqTwmRERERKTymBARERGRyqvQCdHChQvRpEkTGBgYwMzMDN7e3khISJCpk5WVhdGjR6NKlSrQ19dH79698ejRI5k6SUlJ6NKlC3R1dWFmZobJkycjLy+vLJtCREREFViFTohOnz6N0aNH48KFCzhx4gRyc3PRoUMHvHz5Uqwzfvx4HDx4EDt37sTp06fx8OFD9OrVSyzPz89Hly5dkJOTg/Pnz2Pz5s0ICQnBrFmzyqNJREREVAFVKu8AShIWFiazHRISAjMzM0RHR6NVq1bIyMjApk2bEBoairZt2wIAgoOD4eDggAsXLuB///sfjh8/juvXr+PkyZMwNzeHi4sL5s6di6lTpyIwMBCamprl0TQiIiKqQCr0HaJ3ZWRkAABMTEwAANHR0cjNzYWnp6dYp169eqhRowYiIyMBAJGRkXB2doa5ublYx8vLC5mZmbh27VqR18nOzkZmZqbMi4iIiD5fn0xCJJVKMW7cOLi7u+OLL74AAKSkpEBTUxPGxsYydc3NzZGSkiLWeTsZKigvKCvKwoULYWRkJL6sra0V3BoiIiKqSD6ZhGj06NH4559/sG3bNqVfa/r06cjIyBBf9+/fV/o1iYiIqPxU6DFEBfz9/XHo0CGcOXMG1atXF/dbWFggJycH6enpMneJHj16BAsLC7HOpUuXZM5X8BRaQZ13aWlpQUtLS8GtICIiooqqQt8hEgQB/v7+2Lt3L/7880/Y2dnJlDdu3BgaGhoIDw8X9yUkJCApKQlubm4AADc3N8TFxSE1NVWsc+LECRgaGsLR0bFsGkJEREQVWoW+QzR69GiEhoZi//79MDAwEMf8GBkZQUdHB0ZGRhg6dCgmTJgAExMTGBoaYsyYMXBzc8P//vc/AECHDh3g6OiIgQMHYsmSJUhJScGMGTMwevRo3gUiIiIiABU8IQoKCgIAtG7dWmZ/cHAwBg8eDAD46aefoKamht69eyM7OxteXl5Yt26dWFddXR2HDh3CyJEj4ebmBj09Pfj5+WHOnDll1QwiIiKq4Cp0QiQIwnvraGtrY+3atVi7dm2xdWxsbHDkyBFFhkZERESfkQo9hoiIiIioLDAhIiIiIpXHhIiIiIhUHhMiIiIiUnlMiIiIiEjlMSEiIiIilceEiIiIiFQeEyIiIiJSeUyIiIiISOUxISIiIiKVx4SIiIiIVB4TIiIiIlJ5TIiIiIhI5TEhIiIiIpXHhIiIiIhUHhMiIiIiUnlMiIiIiEjlMSEiIiIilceEiIiIiFQeEyIiIiJSeUyIiIiISOUxISIiIiKVx4SIiIiIVB4TIiIiIlJ5TIiIiIhI5TEhIiIiIpXHhIiIiIhUHhMiIiIiUnlMiIiIiEjlMSEiIiIilceEiIiIiFQeEyIiIiJSeUyIiIiISOUxISIiIiKVx4SIiIiIVB4TIiIiIlJ5TIiIiIhI5TEhIiIiIpXHhIiIiIhUHhMiIiIiUnlMiIiIiEjlMSEiIiIilceEiIiIiFQeEyIiIiJSeUyIiIiISOUxISIiIiKVx4SIiIiIVB4TIiIiIlJ5TIiIiIhI5TEhIiIiIpXHhIiIiIhUHhMiIiIiUnmVyjsAIpLferfvlHr+0YhT6vmJlEHZPxcAfzZUARMiKtFoi55lcJUMpV/hc/kPMy4xSenXINXxufx88+ei4viU/69lQqREn/IX43PD/zArjs/llzARFfYp/1/LhEiJPuUvBhGVjIkdKZryv1P8PpWEg6qJiIhI5alUQrR27VrY2tpCW1sbzZo1w6VLl8o7JCIiIqoAVCYh2r59OyZMmICAgABcuXIFDRo0gJeXF1JTU8s7NCIiIipnKpMQLV++HMOGDcOQIUPg6OiI9evXQ1dXF7/++mt5h0ZERETlTCUGVefk5CA6OhrTp08X96mpqcHT0xORkZFKu65tVqjSzl3grpLP/zm0AWA75HVXqWf/vHwO36nPoQ0A2yGvu0o9+xuf8mehEgnRkydPkJ+fD3Nzc5n95ubmuHHjRqH62dnZyM7OFrczMt6MzM/MzCzVdaXZrz4g2tIpbUyl9Tm0AWA75FUWbfgi4xelX+Ofz+CzAPjzLS+2Qz6fQxuA0rWjoK4gCO+vLKiABw8eCACE8+fPy+yfPHmy0LRp00L1AwICBAB88cUXX3zxxddn8Lp///57cwWVuENUtWpVqKur49GjRzL7Hz16BAsLi0L1p0+fjgkTJojbUqkUaWlpqFKlCiQSiVJizMzMhLW1Ne7fvw9DQ0OlXKMsfA7t+BzaALAdFcnn0Abg82jH59AGgO2QlyAIeP78OaysrN5bVyUSIk1NTTRu3Bjh4eHw9vYG8CbJCQ8Ph7+/f6H6Wlpa0NLSktlnbGxcBpEChoaGn/SXu8Dn0I7PoQ0A21GRfA5tAD6PdnwObQDYDnkYGRnJVU8lEiIAmDBhAvz8/ODq6oqmTZtixYoVePnyJYYMGVLeoREREVE5U5mEqG/fvnj8+DFmzZqFlJQUuLi4ICwsrNBAayIiIlI9KpMQAYC/v3+RXWQVgZaWFgICAgp11X1qPod2fA5tANiOiuRzaAPwebTjc2gDwHYog0QQ5HkWjYiIiOjzpTIzVRMREREVhwkRERERqTwmRERERKTymBARUYXGYY5EVBZU6imziuTJkyf49ddfERkZiZSUFACAhYUFmjdvjsGDB8PU1LScIySqGLS0tBAbGwsHB4fyDoWIPmN8yqwcREVFwcvLC7q6uvD09BTnQnr06BHCw8Px6tUrHDt2DK6uruUcqWp4/fo1oqOjYWJiAkdHR5myrKws7NixA4MGDSqn6OQXHx+PCxcuwM3NDfXq1cONGzewcuVKZGdn46uvvkLbtm3LO8QSvb1czttWrlyJr776ClWqVAEALF++vCzD+mgvX77Ejh07cOvWLVhaWqJ///5iW0j5xowZAx8fH7Rs2bK8Q1F5ycnJCAoKwtmzZ5GcnAw1NTXY29vD29sbgwcPhrq6evkGqIC1U6mUmjVrJgwfPlyQSqWFyqRSqTB8+HDhf//7XzlEplhJSUnCkCFDyjuMEiUkJAg2NjaCRCIR1NTUhFatWgkPHz4Uy1NSUgQ1NbVyjFA+R48eFTQ1NQUTExNBW1tbOHr0qGBqaip4enoKbdu2FdTV1YXw8PDyDrNEEolEcHFxEVq3bi3zkkgkQpMmTYTWrVsLbdq0Ke8w38vBwUF4+vSpIAhvfgZsbW0FIyMjoUmTJoKJiYlgZmYm3Llzp5yjfL/o6GiZOLds2SI0b95cqF69uuDu7i788ccf5Rid/Ap+tmvXri0sWrRISE5OLu+QPsjq1auFgQMHiu/7li1bBAcHB6Fu3brC9OnThdzc3HKOsGRRUVGCkZGR0LhxY6FFixaCurq6MHDgQKFv376CsbGx0Lx5cyEzM7NcY2RCVA60tbWF+Pj4Ysvj4+MFbW3tMoxIOWJiYip8MuHt7S106dJFePz4sXDz5k2hS5cugp2dnXDv3j1BED6dhMjNzU344YcfBEEQhD/++EOoXLmy8P3334vl06ZNE9q3b19e4cll4cKFgp2dXaHErVKlSsK1a9fKKarSk0gkwqNHjwRBEARfX1+hefPmQnp6uiAIgvD8+XPB09NT6N+/f3mGKJf69esLJ06cEARBEDZu3Cjo6OgIY8eOFYKCgoRx48YJ+vr6wqZNm8o5yveTSCTCyZMnhe+++06oWrWqoKGhIXTv3l04ePCgkJ+fX97hyWXu3LmCgYGB0Lt3b8HCwkJYtGiRUKVKFWHevHnCggULBFNTU2HWrFnlHWaJ3N3dhcDAQHH7t99+E5o1ayYIgiCkpaUJLi4uwtixY8srPEEQmBCVC1tbW2Hz5s3Flm/evFmwsbEpu4A+0P79+0t8/fTTTxU+mTAzMxOuXr0qbkulUmHEiBFCjRo1hNu3b38yCZGhoaFw8+ZNQRAEIT8/X6hUqZJw5coVsTwuLk4wNzcvr/DkdunSJaFOnTrCxIkThZycHEEQPu2EyN7eXjh+/LhM+blz5wRra+vyCK1UdHR0hLt37wqCIAgNGzYUfv75Z5nyrVu3Co6OjuURWqm8/Xnk5OQI27dvF7y8vAR1dXXByspK+P7778WfnYqqZs2awu7duwVBePOHprq6uvD777+L5Xv27BFq1apVXuHJRUdHR7h9+7a4nZ+fL2hoaAgpKSmCIAjC8ePHBSsrq/IKTxAEQeCg6nIwadIkDB8+HNHR0WjXrl2hMUQbN27E0qVLyznK9/P29oZEIinxKSCJRFKGEZXe69evUanS//0YSCQSBAUFwd/fHx4eHggNDS3H6Eqn4L1WU1ODtra2zArPBgYGyMjIKK/Q5NakSRNER0dj9OjRcHV1xdatWyv8d6goBTFnZWXB0tJSpqxatWp4/PhxeYRVKrq6unjy5AlsbGzw4MEDNG3aVKa8WbNmSExMLKfoPoyGhgZ8fHzg4+ODpKQk/PrrrwgJCcGiRYuQn59f3uEV6+HDh+KY0gYNGkBNTQ0uLi5ieaNGjfDw4cNyik4+ZmZmSE5Ohr29PYA3v+/y8vLEFe5r166NtLS08gyRj92Xh9GjR2Pz5s24ePEievfuDTc3N7i5uaF37964ePEiQkJCMGrUqPIO870sLS2xZ88eSKXSIl9Xrlwp7xDfq169erh8+XKh/WvWrEGPHj3QvXv3coiq9GxtbXHz5k1xOzIyEjVq1BC3k5KSCv1irqj09fWxefNmTJ8+HZ6enhX6F1Vx2rVrh0aNGiEzMxMJCQkyZffu3fskBlV36tQJQUFBAAAPDw/s2rVLpnzHjh2oVatWeYSmEDVq1EBgYCASExMRFhZW3uGUyMLCAtevXwcA3Lx5E/n5+eI2AFy7dg1mZmblFZ5cvL29MWLECISFheHUqVPw9fWFh4cHdHR0AAAJCQmoVq1aucbIO0TlpG/fvujbty9yc3Px5MkTAEDVqlWhoaFRzpHJr3HjxoiOjkaPHj2KLH/f3aOKoGfPnvjjjz8wcODAQmVr1qyBVCrF+vXryyGy0hk5cqRM4vDFF1/IlB89erTCP2X2rn79+qFFixaIjo6GjY1NeYcjt4CAAJltfX19me2DBw9+Ek88LV68GO7u7vDw8ICrqyuWLVuGiIgIODg4ICEhARcuXMDevXvLO8z3srGxKfHpJYlEgvbt25dhRKXn6+uLQYMGoUePHggPD8eUKVMwadIkPH36FBKJBPPnz8eXX35Z3mGWaN68eUhOTka3bt2Qn58PNzc3/P7772K5RCLBwoULyzFCPnZPH+Gvv/7Cy5cv0bFjxyLLX758icuXL8PDw6OMIyMiRUhPT8eiRYtw8OBB3LlzB1KpFJaWlnB3d8f48eM5NUgZkUqlWLRoESIjI9G8eXNMmzYN27dvx5QpU/Dq1St069YNa9asgZ6eXnmH+l5ZWVnIy8sr9IdCRcCEiIiIiFQexxARERGRymNCRERERCqPCRERERGpPCZERPRZuXv3LiQSCWJiYpR2jcGDB8Pb21tp5yeisseEiIgqlMGDB0MikRR6Ffc047usra2RnJxcaOoBIqKScB4iIqpwOnbsiODgYJl9Wlpach2rrq4OCwsLZYRFRJ8x3iEiogpHS0sLFhYWMq/KlSsD+L/lVTp16gQdHR3Y29vLzKL8bpfZs2fP4OvrC1NTU+jo6KB27doyyVZcXBzatm0LHR0dVKlSBcOHD8eLFy/E8vz8fEyYMAHGxsaoUqUKpkyZUmjCUalUioULF8LOzg46Ojpo0KCBTEzvi4GIyh8TIiL65MycORO9e/dGbGwsfH190a9fP8THxxdb9/r16zh69Cji4+MRFBSEqlWrAngzeaiXlxcqV66MqKgo7Ny5EydPnoS/v794/LJlyxASEoJff/0VZ8+eRVpaWqEZmhcuXIgtW7Zg/fr1uHbtGsaPH4+vvvoKp0+ffm8MRFRBlOPCskREhfj5+Qnq6uqCnp6ezGv+/PmCIAgCAGHEiBEyxzRr1kwYOXKkIAiCkJiYKAAQ/v77b0EQBKFbt27CkCFDirzWzz//LFSuXFl48eKFuO/w4cOCmpqauAq3paWlsGTJErE8NzdXqF69utCjRw9BEAQhKytL0NXVFc6fPy9z7qFDhwr9+/d/bwxEVDFwDBERVTht2rQRFxYtYGJiIv7bzc1NpszNza3Yp8pGjhyJ3r1748qVK+jQoQO8vb3RvHlzAEB8fDwaNGggs+SBu7s7pFIpEhISoK2tjeTkZDRr1kwsr1SpElxdXcVus1u3buHVq1eF1sPKyclBw4YN3xsDEVUMTIiIqMLR09NT2ErqnTp1wr1793DkyBGcOHEC7dq1w+jRo7F06VKFnL9gvNHhw4cLrdZdMBBc2TEQ0cfjGCIi+uRcuHCh0LaDg0Ox9U1NTeHn54fff/8dK1aswM8//wwAcHBwQGxsLF6+fCnWPXfuHNTU1FC3bl0YGRnB0tISFy9eFMvz8vIQHR0tbjs6OkJLSwtJSUmoVauWzMva2vq9MRBRxcA7RERU4WRnZyMlJUVmX6VKlcSByDt37oSrqytatGiBrVu34tKlS9i0aVOR55o1axYaN24MJycnZGdn49ChQ2Ly5Ovri4CAAPj5+SEwMBCPHz/GmDFjMHDgQJibmwMAvvvuOyxatAi1a9dGvXr1sHz5cqSnp4vnNzAwwKRJkzB+/HhIpVK0aNECGRkZOHfuHAwNDeHn51diDERUMTAhIqIKJywsDJaWljL76tatixs3bgAAZs+ejW3btmHUqFGwtLTEH3/8AUdHxyLPpampienTp+Pu3bvQ0dFBy5YtsW3bNgCArq4ujh07hu+++w5NmjSBrq4uevfujeXLl4vHT5w4EcnJyfDz84Oamhq+/vpr9OzZExkZGWKduXPnwtTUFAsXLsSdO3dgbGyMRo0a4fvvv39vDERUMUgE4Z0JNYiIKjCJRIK9e/dy6QwiUiiOISIiIiKVx4SIiIiIVB7HEBHRJ4W9/ESkDLxDRERERCqPCRERERGpPCZEREREpPKYEBEREZHKY0JEREREKo8JEREREak8JkRERESk8pgQERERkcpjQkREREQq7/8BoijHTbZ+GvEAAAAASUVORK5CYII=", + "image/png": "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", "text/plain": [ "
" ] @@ -126806,7 +65901,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -126814,9 +65909,16 @@ "metadata": {}, "output_type": "display_data" }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Kolmogorov-Smirnov test: D = 0.04116569526822145, p-value = 0.6737409862064982\n" + ] + }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -126824,9 +65926,16 @@ "metadata": {}, "output_type": "display_data" }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Kolmogorov-Smirnov test: D = 0.04817574274495584, p-value = 0.47467333600545725\n" + ] + }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -126834,179 +65943,16478 @@ "metadata": {}, "output_type": "display_data" }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "The correlation between 'costs_from_violations_smooth' and 'at_dest_f' is 0.5093685120369722\n" + ] + } + ], + "source": [ + "avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f = None, None, None, None, None, None, None\n", + "for seed in seeds:\n", + " avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results= simulation_10_to_4_agents(seed, False) \n", + " if seed == seeds[0]:\n", + " avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f= avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results\n", + " first_gen_results_f = [list(item) for item in first_gen_results_f]\n", + " middle_gen_results_f = [list(item) for item in middle_gen_results_f]\n", + " last_gen_results_f = [list(item) for item in last_gen_results_f]\n", + " else:\n", + " avg_training_rewards_f = [sum(x) / 2 for x in zip(avg_training_rewards_f, avg_training_rewards)]\n", + " \n", + "\n", + "\n", + " first_gen_results_f[0][0] = [sum(x) / 2 for x in zip(first_gen_results_f[0][0], first_gen_results[0][0])]\n", + " middle_gen_results_f[0][0] = [sum(x) / 2 for x in zip(middle_gen_results_f[0][0], middle_gen_results[0][0])]\n", + " last_gen_results_f[0][0] = [sum(x) / 2 for x in zip(last_gen_results_f[0][0], last_gen_results[0][0])]\n", + "\n", + " avg_training_norm_violations_f = avg_training_norm_violations_f.add(avg_training_norm_violations).div(2)\n", + " \n", + " first_gen_results_f[0][1] = first_gen_results_f[0][1].add(first_gen_results[0][1]).div(2)\n", + " middle_gen_results_f[0][1] = middle_gen_results_f[0][1].add(middle_gen_results[0][1]).div(2)\n", + " last_gen_results_f[0][1] = last_gen_results_f[0][1].add(last_gen_results[0][1]).div(2)\n", + "\n", + " sim_violated_f = sim_violated_f.add(sim_violated).div(2)\n", + " at_dest_f = [sum(x) / 2 for x in zip(at_dest_f, at_dest)]\n", + "\n", + "#plot results\n", + "plt.plot(avg_training_rewards_f, label='Average training reward')\n", + "plt.plot(first_gen_results_f[0][0], label='First generation reward')\n", + "plt.plot(middle_gen_results_f[0][0], label='Middle generation reward')\n", + "plt.plot(last_gen_results_f[0][0], label='Last generation reward')\n", + "plt.legend()\n", + "plt.title('Average training reward')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Average reward')\n", + "plt.show()\n", + "\n", + "\n", + "#training violations\n", + "costs_from_violations = copy.deepcopy(avg_training_norm_violations_f['total_violations_cost'])\n", + "\n", + "avg_training_norm_violations_f.drop(columns=['seed'], inplace=True)\n", + "avg_training_norm_violations_f.drop(columns=['total_violations_cost'], inplace=True)\n", + "\n", + "# Calculate the moving average\n", + " # Change this to the desired window size\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=200).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('AVERAGE Costs from violations training')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Cost')\n", + "plt.show()\n", + "\n", + "costs_from_violations = copy.deepcopy(first_gen_results_f[0][1]['total_violations_cost'])\n", + "first_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", + "first_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", + "\n", + "#graph costs from violations\n", + "####costs_from_violations = np.mean(costs_from_violations, axis=0)\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=200).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('First Gen Costs from violations training')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Cost')\n", + "plt.show()\n", + "\n", + "costs_from_violations = copy.deepcopy(middle_gen_results_f[0][1]['total_violations_cost'])\n", + "middle_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", + "middle_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", + "#costs_from_violations = np.mean(costs_from_violations, axis=0)\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=200).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('Middle Gen Costs from violations training')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Cost')\n", + "plt.show()\n", + "\n", + "costs_from_violations = copy.deepcopy(last_gen_results_f[0][1]['total_violations_cost'])\n", + "last_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", + "last_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", + "#cost_from_violations = np.mean(costs_from_violations, axis=0)\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=100).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('Final Gen Costs from violations training')\n", + "plt.xlabel('Episodes')\n", + "plt.ylabel('Cost')\n", + "plt.show()\n", + "\n", + "\n", + "#training violations\n", + "costs_from_violations_f = copy.deepcopy(sim_violated_f['total_violations_cost'])\n", + "sim_violated_f.drop(columns=['seed'], inplace=True)\n", + "sim_violated_f.drop(columns=['total_violations_cost'], inplace=True)\n", + "costs_from_violations = -1 * costs_from_violations_f\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations).rolling(window=window_size2).mean()\n", + "#costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "\n", + "sns.histplot(costs_from_violations, kde=True)\n", + "plt.title('Distribution of Costs from Violations Simulation')\n", + "plt.xlabel('Cost')\n", + "plt.ylabel('Density')\n", + "plt.show()\n", + "\n", + "#GAMMA\n", + "data = costs_from_violations\n", + "\n", + "# Fit the gamma distribution to your data\n", + "shape, loc, scale = gamma.fit(data)\n", + "\n", + "# Generate a range of values from the min to the max of your data\n", + "x = np.linspace(min(data), max(data), 10000)\n", + "\n", + "# Generate the gamma PDF with the parameters obtained from the fit\n", + "pdf = gamma.pdf(x, shape, loc, scale)\n", + "\n", + "# Plot the histogram of your data and the gamma PDF\n", + "plt.hist(data, bins=30, density=True, alpha=0.5, label='Histogram of Costs from Violations Simulation')\n", + "plt.plot(x, pdf, 'r-', label='Gamma PDF')\n", + "plt.legend()\n", + "plt.show()\n", + "\n", + "# Perform the Kolmogorov-Smirnov test\n", + "d, p_value = kstest(data, 'gamma', args=(shape, loc, scale))\n", + "print(f\"Kolmogorov-Smirnov test: D = {d}, p-value = {p_value}\")\n", + "\n", + "#LOG-NORMAL\n", + "# Fit the log-normal distribution to your data\n", + "shape, loc, scale = lognorm.fit(data)\n", + "\n", + "# Generate the log-normal PDF with the parameters obtained from the fit\n", + "pdf = lognorm.pdf(x, shape, loc, scale)\n", + "\n", + "# Plot the histogram of your data and the log-normal PDF\n", + "plt.hist(data, bins=30, density=True, alpha=0.5, label='Histogram of Costs from Violations Simulation')\n", + "plt.plot(x, pdf, 'r-', label='Log-normal PDF')\n", + "plt.legend()\n", + "plt.show()\n", + "\n", + "# Perform the Kolmogorov-Smirnov test\n", + "d, p_value = kstest(data, 'lognorm', args=(shape, loc, scale))\n", + "print(f\"Kolmogorov-Smirnov test: D = {d}, p-value = {p_value}\")\n", + "\n", + "\n", + "#simulation at destination, avg timeto destination\n", + "#simulation at destination, avg timeto destination\n", + "at_dest_f = np.array(at_dest_f)\n", + "sns.histplot(at_dest_f, kde=True)\n", + "plt.title('Distribution of Average Steps to Destination')\n", + "plt.xlabel('Steps')\n", + "plt.ylabel('Density')\n", + "plt.show()\n", + "\n", + "\n", + "\n", + "df = pd.DataFrame({\n", + " 'costs_from_violations_smooth': costs_from_violations,\n", + " 'at_dest_f': at_dest_f\n", + "})\n", + "\n", + "# Calculate the correlation\n", + "correlation = df['costs_from_violations_smooth'].corr(df['at_dest_f'])\n", + "\n", + "print(f\"The correlation between 'costs_from_violations_smooth' and 'at_dest_f' is {correlation}\")\n", + "violation_counts = sim_violated_f.sum()\n", + "\n", + "\n", + "plt.pie(violation_counts, autopct='%1.1f%%')\n", + "\n", + "plt.title('Violation Count')\n", + "\n", + "plt.legend(violation_counts.index, title=\"Violations\", loc=\"best\")\n", + "\n", + "plt.show()\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": {}, + "outputs": [ { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" + } + ], + "source": [ + "\n", + "at_dest_f = np.array(at_dest_f)\n", + "\n", + "plt.boxplot(at_dest_f)\n", + "plt.title('Average Steps to destination')\n", + "plt.xlabel('Simulation Run')\n", + "plt.ylabel('Steps ')\n", + "plt.show()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "## 10 Agents" + ] + }, + { + "cell_type": "code", + "execution_count": 7, + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", + "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", + " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Generation 1/300\n", + "Solving for Nash Equilibrium in Generation 1/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 2/300\n", + "Solving for Nash Equilibrium in Generation 2/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 3/300\n", + "Solving for Nash Equilibrium in Generation 3/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 4/300\n", + "Solving for Nash Equilibrium in Generation 4/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 5/300\n", + "Solving for Nash Equilibrium in Generation 5/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 6/300\n", + "Solving for Nash Equilibrium in Generation 6/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 7/300\n", + "Solving for Nash Equilibrium in Generation 7/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 8/300\n", + "Solving for Nash Equilibrium in Generation 8/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 9/300\n", + "Solving for Nash Equilibrium in Generation 9/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 10/300\n", + "Solving for Nash Equilibrium in Generation 10/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 11/300\n", + "Solving for Nash Equilibrium in Generation 11/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 12/300\n", + "Solving for Nash Equilibrium in Generation 12/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 13/300\n", + "Solving for Nash Equilibrium in Generation 13/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 14/300\n", + "Solving for Nash Equilibrium in Generation 14/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 15/300\n", + "Solving for Nash Equilibrium in Generation 15/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 16/300\n", + "Solving for Nash Equilibrium in Generation 16/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 17/300\n", + "Solving for Nash Equilibrium in Generation 17/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 18/300\n", + "Solving for Nash Equilibrium in Generation 18/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 19/300\n", + "Solving for Nash Equilibrium in Generation 19/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 20/300\n", + "Solving for Nash Equilibrium in Generation 20/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 21/300\n", + "Solving for Nash Equilibrium in Generation 21/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 22/300\n", + "Solving for Nash Equilibrium in Generation 22/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 23/300\n", + "Solving for Nash Equilibrium in Generation 23/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 24/300\n", + "Solving for Nash Equilibrium in Generation 24/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 25/300\n", + "Solving for Nash Equilibrium in Generation 25/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 26/300\n", + "Solving for Nash Equilibrium in Generation 26/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 27/300\n", + "Solving for Nash Equilibrium in Generation 27/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 28/300\n", + "Solving for Nash Equilibrium in Generation 28/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 29/300\n", + "Solving for Nash Equilibrium in Generation 29/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 30/300\n", + "Solving for Nash Equilibrium in Generation 30/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 31/300\n", + "Solving for Nash Equilibrium in Generation 31/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 32/300\n", + "Solving for Nash Equilibrium in Generation 32/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 33/300\n", + "Solving for Nash Equilibrium in Generation 33/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 34/300\n", + "Solving for Nash Equilibrium in Generation 34/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 35/300\n", + "Solving for Nash Equilibrium in Generation 35/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 36/300\n", + "Solving for Nash Equilibrium in Generation 36/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 37/300\n", + "Solving for Nash Equilibrium in Generation 37/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 38/300\n", + "Solving for Nash Equilibrium in Generation 38/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 39/300\n", + "Solving for Nash Equilibrium in Generation 39/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 40/300\n", + "Solving for Nash Equilibrium in Generation 40/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 41/300\n", + "Solving for Nash Equilibrium in Generation 41/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 42/300\n", + "Solving for Nash Equilibrium in Generation 42/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 43/300\n", + "Solving for Nash Equilibrium in Generation 43/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 44/300\n", + "Solving for Nash Equilibrium in Generation 44/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 45/300\n", + "Solving for Nash Equilibrium in Generation 45/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 46/300\n", + "Solving for Nash Equilibrium in Generation 46/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 47/300\n", + "Solving for Nash Equilibrium in Generation 47/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 48/300\n", + "Solving for Nash Equilibrium in Generation 48/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 49/300\n", + "Solving for Nash Equilibrium in Generation 49/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 50/300\n", + "Solving for Nash Equilibrium in Generation 50/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 51/300\n", + "Solving for Nash Equilibrium in Generation 51/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 52/300\n", + "Solving for Nash Equilibrium in Generation 52/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 53/300\n", + "Solving for Nash Equilibrium in Generation 53/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 54/300\n", + "Solving for Nash Equilibrium in Generation 54/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 55/300\n", + "Solving for Nash Equilibrium in Generation 55/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 56/300\n", + "Solving for Nash Equilibrium in Generation 56/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 57/300\n", + "Solving for Nash Equilibrium in Generation 57/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 58/300\n", + "Solving for Nash Equilibrium in Generation 58/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 59/300\n", + "Solving for Nash Equilibrium in Generation 59/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 60/300\n", + "Solving for Nash Equilibrium in Generation 60/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 61/300\n", + "Solving for Nash Equilibrium in Generation 61/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 62/300\n", + "Solving for Nash Equilibrium in Generation 62/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 63/300\n", + "Solving for Nash Equilibrium in Generation 63/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 64/300\n", + "Solving for Nash Equilibrium in Generation 64/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 65/300\n", + "Solving for Nash Equilibrium in Generation 65/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 66/300\n", + "Solving for Nash Equilibrium in Generation 66/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 67/300\n", + "Solving for Nash Equilibrium in Generation 67/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 68/300\n", + "Solving for Nash Equilibrium in Generation 68/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 69/300\n", + "Solving for Nash Equilibrium in Generation 69/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 70/300\n", + "Solving for Nash Equilibrium in Generation 70/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 71/300\n", + "Solving for Nash Equilibrium in Generation 71/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 72/300\n", + "Solving for Nash Equilibrium in Generation 72/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 73/300\n", + "Solving for Nash Equilibrium in Generation 73/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 74/300\n", + "Solving for Nash Equilibrium in Generation 74/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 75/300\n", + "Solving for Nash Equilibrium in Generation 75/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 76/300\n", + "Solving for Nash Equilibrium in Generation 76/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 77/300\n", + "Solving for Nash Equilibrium in Generation 77/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 78/300\n", + "Solving for Nash Equilibrium in Generation 78/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 79/300\n", + "Solving for Nash Equilibrium in Generation 79/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 80/300\n", + "Solving for Nash Equilibrium in Generation 80/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 81/300\n", + "Solving for Nash Equilibrium in Generation 81/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 82/300\n", + "Solving for Nash Equilibrium in Generation 82/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 83/300\n", + "Solving for Nash Equilibrium in Generation 83/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 84/300\n", + "Solving for Nash Equilibrium in Generation 84/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 85/300\n", + "Solving for Nash Equilibrium in Generation 85/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 86/300\n", + "Solving for Nash Equilibrium in Generation 86/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 87/300\n", + "Solving for Nash Equilibrium in Generation 87/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 88/300\n", + "Solving for Nash Equilibrium in Generation 88/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 89/300\n", + "Solving for Nash Equilibrium in Generation 89/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 90/300\n", + "Solving for Nash Equilibrium in Generation 90/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 91/300\n", + "Solving for Nash Equilibrium in Generation 91/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 92/300\n", + "Solving for Nash Equilibrium in Generation 92/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 93/300\n", + "Solving for Nash Equilibrium in Generation 93/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 94/300\n", + "Solving for Nash Equilibrium in Generation 94/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 95/300\n", + "Solving for Nash Equilibrium in Generation 95/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 96/300\n", + "Solving for Nash Equilibrium in Generation 96/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 97/300\n", + "Solving for Nash Equilibrium in Generation 97/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 98/300\n", + "Solving for Nash Equilibrium in Generation 98/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 99/300\n", + "Solving for Nash Equilibrium in Generation 99/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 100/300\n", + "Solving for Nash Equilibrium in Generation 100/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 101/300\n", + "Solving for Nash Equilibrium in Generation 101/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 102/300\n", + "Solving for Nash Equilibrium in Generation 102/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 103/300\n", + "Solving for Nash Equilibrium in Generation 103/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 104/300\n", + "Solving for Nash Equilibrium in Generation 104/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 105/300\n", + "Solving for Nash Equilibrium in Generation 105/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 106/300\n", + "Solving for Nash Equilibrium in Generation 106/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 107/300\n", + "Solving for Nash Equilibrium in Generation 107/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 108/300\n", + "Solving for Nash Equilibrium in Generation 108/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 109/300\n", + "Solving for Nash Equilibrium in Generation 109/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 110/300\n", + "Solving for Nash Equilibrium in Generation 110/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 111/300\n", + "Solving for Nash Equilibrium in Generation 111/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 112/300\n", + "Solving for Nash Equilibrium in Generation 112/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 113/300\n", + "Solving for Nash Equilibrium in Generation 113/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 114/300\n", + "Solving for Nash Equilibrium in Generation 114/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 115/300\n", + "Solving for Nash Equilibrium in Generation 115/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 116/300\n", + "Solving for Nash Equilibrium in Generation 116/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 117/300\n", + "Solving for Nash Equilibrium in Generation 117/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 118/300\n", + "Solving for Nash Equilibrium in Generation 118/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 119/300\n", + "Solving for Nash Equilibrium in Generation 119/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 120/300\n", + "Solving for Nash Equilibrium in Generation 120/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 121/300\n", + "Solving for Nash Equilibrium in Generation 121/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 122/300\n", + "Solving for Nash Equilibrium in Generation 122/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 123/300\n", + "Solving for Nash Equilibrium in Generation 123/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 124/300\n", + "Solving for Nash Equilibrium in Generation 124/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 125/300\n", + "Solving for Nash Equilibrium in Generation 125/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 126/300\n", + "Solving for Nash Equilibrium in Generation 126/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 127/300\n", + "Solving for Nash Equilibrium in Generation 127/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 128/300\n", + "Solving for Nash Equilibrium in Generation 128/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 129/300\n", + "Solving for Nash Equilibrium in Generation 129/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 130/300\n", + "Solving for Nash Equilibrium in Generation 130/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 131/300\n", + "Solving for Nash Equilibrium in Generation 131/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 132/300\n", + "Solving for Nash Equilibrium in Generation 132/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 133/300\n", + "Solving for Nash Equilibrium in Generation 133/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 134/300\n", + "Solving for Nash Equilibrium in Generation 134/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 135/300\n", + "Solving for Nash Equilibrium in Generation 135/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 136/300\n", + "Solving for Nash Equilibrium in Generation 136/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 137/300\n", + "Solving for Nash Equilibrium in Generation 137/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 138/300\n", + "Solving for Nash Equilibrium in Generation 138/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 139/300\n", + "Solving for Nash Equilibrium in Generation 139/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 140/300\n", + "Solving for Nash Equilibrium in Generation 140/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 141/300\n", + "Solving for Nash Equilibrium in Generation 141/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 142/300\n", + "Solving for Nash Equilibrium in Generation 142/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 143/300\n", + "Solving for Nash Equilibrium in Generation 143/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 144/300\n", + "Solving for Nash Equilibrium in Generation 144/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 145/300\n", + "Solving for Nash Equilibrium in Generation 145/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 146/300\n", + "Solving for Nash Equilibrium in Generation 146/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 147/300\n", + "Solving for Nash Equilibrium in Generation 147/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 148/300\n", + "Solving for Nash Equilibrium in Generation 148/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 149/300\n", + "Solving for Nash Equilibrium in Generation 149/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 150/300\n", + "Solving for Nash Equilibrium in Generation 150/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 151/300\n", + "Solving for Nash Equilibrium in Generation 151/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 152/300\n", + "Solving for Nash Equilibrium in Generation 152/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 153/300\n", + "Solving for Nash Equilibrium in Generation 153/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 154/300\n", + "Solving for Nash Equilibrium in Generation 154/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 155/300\n", + "Solving for Nash Equilibrium in Generation 155/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 156/300\n", + "Solving for Nash Equilibrium in Generation 156/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 157/300\n", + "Solving for Nash Equilibrium in Generation 157/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 158/300\n", + "Solving for Nash Equilibrium in Generation 158/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 159/300\n", + "Solving for Nash Equilibrium in Generation 159/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 160/300\n", + "Solving for Nash Equilibrium in Generation 160/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 161/300\n", + "Solving for Nash Equilibrium in Generation 161/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 162/300\n", + "Solving for Nash Equilibrium in Generation 162/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 163/300\n", + "Solving for Nash Equilibrium in Generation 163/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 164/300\n", + "Solving for Nash Equilibrium in Generation 164/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 165/300\n", + "Solving for Nash Equilibrium in Generation 165/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 166/300\n", + "Solving for Nash Equilibrium in Generation 166/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 167/300\n", + "Solving for Nash Equilibrium in Generation 167/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 168/300\n", + "Solving for Nash Equilibrium in Generation 168/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 169/300\n", + "Solving for Nash Equilibrium in Generation 169/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 170/300\n", + "Solving for Nash Equilibrium in Generation 170/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 171/300\n", + "Solving for Nash Equilibrium in Generation 171/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 172/300\n", + "Solving for Nash Equilibrium in Generation 172/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 173/300\n", + "Solving for Nash Equilibrium in Generation 173/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 174/300\n", + "Solving for Nash Equilibrium in Generation 174/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 175/300\n", + "Solving for Nash Equilibrium in Generation 175/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 176/300\n", + "Solving for Nash Equilibrium in Generation 176/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 177/300\n", + "Solving for Nash Equilibrium in Generation 177/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 178/300\n", + "Solving for Nash Equilibrium in Generation 178/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 179/300\n", + "Solving for Nash Equilibrium in Generation 179/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 180/300\n", + "Solving for Nash Equilibrium in Generation 180/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 181/300\n", + "Solving for Nash Equilibrium in Generation 181/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 182/300\n", + "Solving for Nash Equilibrium in Generation 182/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 183/300\n", + "Solving for Nash Equilibrium in Generation 183/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 184/300\n", + "Solving for Nash Equilibrium in Generation 184/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 185/300\n", + "Solving for Nash Equilibrium in Generation 185/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 186/300\n", + "Solving for Nash Equilibrium in Generation 186/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 187/300\n", + "Solving for Nash Equilibrium in Generation 187/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 188/300\n", + "Solving for Nash Equilibrium in Generation 188/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 189/300\n", + "Solving for Nash Equilibrium in Generation 189/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 190/300\n", + "Solving for Nash Equilibrium in Generation 190/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 191/300\n", + "Solving for Nash Equilibrium in Generation 191/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 192/300\n", + "Solving for Nash Equilibrium in Generation 192/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 193/300\n", + "Solving for Nash Equilibrium in Generation 193/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 194/300\n", + "Solving for Nash Equilibrium in Generation 194/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 195/300\n", + "Solving for Nash Equilibrium in Generation 195/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 196/300\n", + "Solving for Nash Equilibrium in Generation 196/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 197/300\n", + "Solving for Nash Equilibrium in Generation 197/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 198/300\n", + "Solving for Nash Equilibrium in Generation 198/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 199/300\n", + "Solving for Nash Equilibrium in Generation 199/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 200/300\n", + "Solving for Nash Equilibrium in Generation 200/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 201/300\n", + "Solving for Nash Equilibrium in Generation 201/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 202/300\n", + "Solving for Nash Equilibrium in Generation 202/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 203/300\n", + "Solving for Nash Equilibrium in Generation 203/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 204/300\n", + "Solving for Nash Equilibrium in Generation 204/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 205/300\n", + "Solving for Nash Equilibrium in Generation 205/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 206/300\n", + "Solving for Nash Equilibrium in Generation 206/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 207/300\n", + "Solving for Nash Equilibrium in Generation 207/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 208/300\n", + "Solving for Nash Equilibrium in Generation 208/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 209/300\n", + "Solving for Nash Equilibrium in Generation 209/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 210/300\n", + "Solving for Nash Equilibrium in Generation 210/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 211/300\n", + "Solving for Nash Equilibrium in Generation 211/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 212/300\n", + "Solving for Nash Equilibrium in Generation 212/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 213/300\n", + "Solving for Nash Equilibrium in Generation 213/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 214/300\n", + "Solving for Nash Equilibrium in Generation 214/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 215/300\n", + "Solving for Nash Equilibrium in Generation 215/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 216/300\n", + "Solving for Nash Equilibrium in Generation 216/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 217/300\n", + "Solving for Nash Equilibrium in Generation 217/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 218/300\n", + "Solving for Nash Equilibrium in Generation 218/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 219/300\n", + "Solving for Nash Equilibrium in Generation 219/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 220/300\n", + "Solving for Nash Equilibrium in Generation 220/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 221/300\n", + "Solving for Nash Equilibrium in Generation 221/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 222/300\n", + "Solving for Nash Equilibrium in Generation 222/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 223/300\n", + "Solving for Nash Equilibrium in Generation 223/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 224/300\n", + "Solving for Nash Equilibrium in Generation 224/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 225/300\n", + "Solving for Nash Equilibrium in Generation 225/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 226/300\n", + "Solving for Nash Equilibrium in Generation 226/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 227/300\n", + "Solving for Nash Equilibrium in Generation 227/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 228/300\n", + "Solving for Nash Equilibrium in Generation 228/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 229/300\n", + "Solving for Nash Equilibrium in Generation 229/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 230/300\n", + "Solving for Nash Equilibrium in Generation 230/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 231/300\n", + "Solving for Nash Equilibrium in Generation 231/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 232/300\n", + "Solving for Nash Equilibrium in Generation 232/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 233/300\n", + "Solving for Nash Equilibrium in Generation 233/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 234/300\n", + "Solving for Nash Equilibrium in Generation 234/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 235/300\n", + "Solving for Nash Equilibrium in Generation 235/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 236/300\n", + "Solving for Nash Equilibrium in Generation 236/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 237/300\n", + "Solving for Nash Equilibrium in Generation 237/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 238/300\n", + "Solving for Nash Equilibrium in Generation 238/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 239/300\n", + "Solving for Nash Equilibrium in Generation 239/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 240/300\n", + "Solving for Nash Equilibrium in Generation 240/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 241/300\n", + "Solving for Nash Equilibrium in Generation 241/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 242/300\n", + "Solving for Nash Equilibrium in Generation 242/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 243/300\n", + "Solving for Nash Equilibrium in Generation 243/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 244/300\n", + "Solving for Nash Equilibrium in Generation 244/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 245/300\n", + "Solving for Nash Equilibrium in Generation 245/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 246/300\n", + "Solving for Nash Equilibrium in Generation 246/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 247/300\n", + "Solving for Nash Equilibrium in Generation 247/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 248/300\n", + "Solving for Nash Equilibrium in Generation 248/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 249/300\n", + "Solving for Nash Equilibrium in Generation 249/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 250/300\n", + "Solving for Nash Equilibrium in Generation 250/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 251/300\n", + "Solving for Nash Equilibrium in Generation 251/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 252/300\n", + "Solving for Nash Equilibrium in Generation 252/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 253/300\n", + "Solving for Nash Equilibrium in Generation 253/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 254/300\n", + "Solving for Nash Equilibrium in Generation 254/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 255/300\n", + "Solving for Nash Equilibrium in Generation 255/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 256/300\n", + "Solving for Nash Equilibrium in Generation 256/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 257/300\n", + "Solving for Nash Equilibrium in Generation 257/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 258/300\n", + "Solving for Nash Equilibrium in Generation 258/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 259/300\n", + "Solving for Nash Equilibrium in Generation 259/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 260/300\n", + "Solving for Nash Equilibrium in Generation 260/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 261/300\n", + "Solving for Nash Equilibrium in Generation 261/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 262/300\n", + "Solving for Nash Equilibrium in Generation 262/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 263/300\n", + "Solving for Nash Equilibrium in Generation 263/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 264/300\n", + "Solving for Nash Equilibrium in Generation 264/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 265/300\n", + "Solving for Nash Equilibrium in Generation 265/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 266/300\n", + "Solving for Nash Equilibrium in Generation 266/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 267/300\n", + "Solving for Nash Equilibrium in Generation 267/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 268/300\n", + "Solving for Nash Equilibrium in Generation 268/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 269/300\n", + "Solving for Nash Equilibrium in Generation 269/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 270/300\n", + "Solving for Nash Equilibrium in Generation 270/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 271/300\n", + "Solving for Nash Equilibrium in Generation 271/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 272/300\n", + "Solving for Nash Equilibrium in Generation 272/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 273/300\n", + "Solving for Nash Equilibrium in Generation 273/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 274/300\n", + "Solving for Nash Equilibrium in Generation 274/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 275/300\n", + "Solving for Nash Equilibrium in Generation 275/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 276/300\n", + "Solving for Nash Equilibrium in Generation 276/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 277/300\n", + "Solving for Nash Equilibrium in Generation 277/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 278/300\n", + "Solving for Nash Equilibrium in Generation 278/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 279/300\n", + "Solving for Nash Equilibrium in Generation 279/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 280/300\n", + "Solving for Nash Equilibrium in Generation 280/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 281/300\n", + "Solving for Nash Equilibrium in Generation 281/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 282/300\n", + "Solving for Nash Equilibrium in Generation 282/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 283/300\n", + "Solving for Nash Equilibrium in Generation 283/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 284/300\n", + "Solving for Nash Equilibrium in Generation 284/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 285/300\n", + "Solving for Nash Equilibrium in Generation 285/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 286/300\n", + "Solving for Nash Equilibrium in Generation 286/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 287/300\n", + "Solving for Nash Equilibrium in Generation 287/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 288/300\n", + "Solving for Nash Equilibrium in Generation 288/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 289/300\n", + "Solving for Nash Equilibrium in Generation 289/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 290/300\n", + "Solving for Nash Equilibrium in Generation 290/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 291/300\n", + "Solving for Nash Equilibrium in Generation 291/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 292/300\n", + "Solving for Nash Equilibrium in Generation 292/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 293/300\n", + "Solving for Nash Equilibrium in Generation 293/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 294/300\n", + "Solving for Nash Equilibrium in Generation 294/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 295/300\n", + "Solving for Nash Equilibrium in Generation 295/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 296/300\n", + "Solving for Nash Equilibrium in Generation 296/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 297/300\n", + "Solving for Nash Equilibrium in Generation 297/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 298/300\n", + "Solving for Nash Equilibrium in Generation 298/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 299/300\n", + "Solving for Nash Equilibrium in Generation 299/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 300/300\n", + "Solving for Nash Equilibrium in Generation 300/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Running Computed Policies\n", + "Episode 0 done\n", + "Episode 1 done\n", + "Episode 2 done\n", + "Episode 3 done\n", + "Episode 4 done\n", + "Episode 5 done\n", + "Episode 6 done\n", + "Episode 7 done\n", + "Episode 8 done\n", + "Episode 9 done\n", + "Episode 10 done\n", + "Episode 11 done\n", + "Episode 12 done\n", + "Episode 13 done\n", + "Episode 14 done\n", + "Episode 15 done\n", + "Episode 16 done\n", + "Episode 17 done\n", + "Episode 18 done\n", + "Episode 19 done\n", + "Episode 20 done\n", + "Episode 21 done\n", + "Episode 22 done\n", + "Episode 23 done\n", + "Episode 24 done\n", + "Episode 25 done\n", + "Episode 26 done\n", + "Episode 27 done\n", + "Episode 28 done\n", + "Episode 29 done\n", + "Episode 30 done\n", + "Episode 31 done\n", + "Episode 32 done\n", + "Episode 33 done\n", + "Episode 34 done\n", + "Episode 34 done\n", + "Total steps: 19\n", + "Episode 35 done\n", + "Episode 36 done\n", + "Episode 36 done\n", + "Total steps: 17\n", + "Episode 37 done\n", + "Episode 38 done\n", + "Episode 39 done\n", + "Episode 40 done\n", + "Episode 41 done\n", + "Episode 42 done\n", + "Episode 43 done\n", + "Episode 44 done\n", + "Episode 45 done\n", + "Episode 46 done\n", + "Episode 47 done\n", + "Episode 48 done\n", + "Episode 49 done\n", + "Episode 50 done\n", + "Episode 51 done\n", + "Episode 52 done\n", + "Episode 53 done\n", + "Episode 54 done\n", + "Episode 55 done\n", + "Episode 56 done\n", + "Episode 57 done\n", + "Episode 58 done\n", + "Episode 59 done\n", + "Episode 60 done\n", + "Episode 61 done\n", + "Episode 62 done\n", + "Episode 63 done\n", + "Episode 64 done\n", + "Episode 65 done\n", + "Episode 66 done\n", + "Episode 67 done\n", + "Episode 68 done\n", + "Episode 69 done\n", + "Episode 70 done\n", + "Episode 71 done\n", + "Episode 72 done\n", + "Episode 73 done\n", + "Episode 74 done\n", + "Episode 75 done\n", + "Episode 76 done\n", + "Episode 77 done\n", + "Episode 78 done\n", + "Episode 79 done\n", + "Episode 80 done\n", + "Episode 81 done\n", + "Episode 82 done\n", + "Episode 83 done\n", + "Episode 84 done\n", + "Episode 85 done\n", + "Episode 86 done\n", + "Episode 87 done\n", + "Episode 88 done\n", + "Episode 89 done\n", + "Episode 90 done\n", + "Episode 91 done\n", + "Episode 92 done\n", + "Episode 93 done\n", + "Episode 94 done\n", + "Episode 95 done\n", + "Episode 96 done\n", + "Episode 97 done\n", + "Episode 98 done\n", + "Episode 99 done\n", + "Episode 100 done\n", + "Episode 101 done\n", + "Episode 102 done\n", + "Episode 103 done\n", + "Episode 104 done\n", + "Episode 105 done\n", + "Episode 106 done\n", + "Episode 107 done\n", + "Episode 108 done\n", + "Episode 109 done\n", + "Episode 110 done\n", + "Episode 111 done\n", + "Episode 112 done\n", + "Episode 113 done\n", + "Episode 114 done\n", + "Episode 115 done\n", + "Episode 116 done\n", + "Episode 117 done\n", + "Episode 118 done\n", + "Episode 119 done\n", + "Episode 120 done\n", + "Episode 121 done\n", + "Episode 122 done\n", + "Episode 123 done\n", + "Episode 124 done\n", + "Episode 125 done\n", + "Episode 126 done\n", + "Episode 127 done\n", + "Episode 128 done\n", + "Episode 129 done\n", + "Episode 130 done\n", + "Episode 131 done\n", + "Episode 132 done\n", + "Episode 133 done\n", + "Episode 134 done\n", + "Episode 135 done\n", + "Episode 136 done\n", + "Episode 137 done\n", + "Episode 138 done\n", + "Episode 139 done\n", + "Episode 140 done\n", + "Episode 141 done\n", + "Episode 142 done\n", + "Episode 143 done\n", + "Episode 144 done\n", + "Episode 145 done\n", + "Episode 146 done\n", + "Episode 147 done\n", + "Episode 148 done\n", + "Episode 149 done\n", + "Episode 150 done\n", + "Episode 151 done\n", + "Episode 152 done\n", + "Episode 153 done\n", + "Episode 154 done\n", + "Episode 155 done\n", + "Episode 156 done\n", + "Episode 157 done\n", + "Episode 158 done\n", + "Episode 159 done\n", + "Episode 160 done\n", + "Episode 161 done\n", + "Episode 162 done\n", + "Episode 163 done\n", + "Episode 164 done\n", + "Episode 165 done\n", + "Episode 166 done\n", + "Episode 167 done\n", + "Episode 168 done\n", + "Episode 169 done\n", + "Episode 170 done\n", + "Episode 171 done\n", + "Episode 172 done\n", + "Episode 173 done\n", + "Episode 173 done\n", + "Total steps: 14\n", + "Episode 174 done\n", + "Episode 175 done\n", + "Episode 176 done\n", + "Episode 177 done\n", + "Episode 178 done\n", + "Episode 179 done\n", + "Episode 180 done\n", + "Episode 180 done\n", + "Total steps: 18\n", + "Episode 181 done\n", + "Episode 182 done\n", + "Episode 183 done\n", + "Episode 184 done\n", + "Episode 185 done\n", + "Episode 186 done\n", + "Episode 187 done\n", + "Episode 188 done\n", + "Episode 189 done\n", + "Episode 190 done\n", + "Episode 191 done\n", + "Episode 192 done\n", + "Episode 193 done\n", + "Episode 194 done\n", + "Episode 195 done\n", + "Episode 196 done\n", + "Episode 197 done\n", + "Episode 198 done\n", + "Episode 199 done\n", + "Episode 200 done\n", + "Episode 201 done\n", + "Episode 202 done\n", + "Episode 203 done\n", + "Episode 204 done\n", + "Episode 205 done\n", + "Episode 206 done\n", + "Episode 207 done\n", + "Episode 208 done\n", + "Episode 209 done\n", + "Episode 210 done\n", + "Episode 211 done\n", + "Episode 212 done\n", + "Episode 213 done\n", + "Episode 214 done\n", + "Episode 215 done\n", + "Episode 216 done\n", + "Episode 217 done\n", + "Episode 218 done\n", + "Episode 219 done\n", + "Episode 220 done\n", + "Episode 221 done\n", + "Episode 221 done\n", + "Total steps: 17\n", + "Episode 222 done\n", + "Episode 223 done\n", + "Episode 224 done\n", + "Episode 225 done\n", + "Episode 226 done\n", + "Episode 227 done\n", + "Episode 228 done\n", + "Episode 229 done\n", + "Episode 230 done\n", + "Episode 231 done\n", + "Episode 232 done\n", + "Episode 233 done\n", + "Episode 234 done\n", + "Episode 235 done\n", + "Episode 236 done\n", + "Episode 237 done\n", + "Episode 238 done\n", + "Episode 239 done\n", + "Episode 240 done\n", + "Episode 241 done\n", + "Episode 242 done\n", + "Episode 243 done\n", + "Episode 244 done\n", + "Episode 245 done\n", + "Episode 245 done\n", + "Total steps: 18\n", + "Episode 246 done\n", + "Episode 247 done\n", + "Episode 248 done\n", + "Episode 249 done\n", + "Episode 250 done\n", + "Episode 251 done\n", + "Episode 252 done\n", + "Episode 253 done\n", + "Episode 254 done\n", + "Episode 255 done\n", + "Episode 256 done\n", + "Episode 257 done\n", + "Episode 258 done\n", + "Episode 259 done\n", + "Episode 260 done\n", + "Episode 261 done\n", + "Episode 262 done\n", + "Episode 263 done\n", + "Episode 264 done\n", + "Episode 265 done\n", + "Episode 266 done\n", + "Episode 267 done\n", + "Episode 268 done\n", + "Episode 269 done\n", + "Episode 270 done\n", + "Episode 271 done\n", + "Episode 272 done\n", + "Episode 273 done\n", + "Episode 274 done\n", + "Episode 275 done\n", + "Episode 276 done\n", + "Episode 277 done\n", + "Episode 278 done\n", + "Episode 279 done\n", + "Episode 280 done\n", + "Episode 281 done\n", + "Episode 282 done\n", + "Episode 283 done\n", + "Episode 283 done\n", + "Total steps: 19\n", + "Episode 284 done\n", + "Episode 285 done\n", + "Episode 286 done\n", + "Episode 287 done\n", + "Episode 288 done\n", + "Episode 289 done\n", + "Episode 290 done\n", + "Episode 291 done\n", + "Episode 292 done\n", + "Episode 293 done\n", + "Episode 294 done\n", + "Episode 295 done\n", + "Episode 296 done\n", + "Episode 297 done\n", + "Episode 298 done\n", + "Episode 299 done\n", + "Generation 1/300\n" + ] }, - { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f = None, None, None, None, None, None, None\n", - "for seed in seeds:\n", - " avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results= simulation_10_agents(seed, False) \n", - " if seed == seeds[0]:\n", - " avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f= avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results\n", - " first_gen_results_f = [list(item) for item in first_gen_results_f]\n", - " middle_gen_results_f = [list(item) for item in middle_gen_results_f]\n", - " last_gen_results_f = [list(item) for item in last_gen_results_f]\n", - " else:\n", - " avg_training_rewards_f = [sum(x) / 2 for x in zip(avg_training_rewards_f, avg_training_rewards)]\n", - " print(\"here\")\n", - " print(first_gen_results[0])\n", - " print(\"here1\")\n", - " print(first_gen_results[0][0])\n", - "\n", - "\n", - " first_gen_results_f[0][0] = [sum(x) / 2 for x in zip(first_gen_results_f[0][0], first_gen_results[0][0])]\n", - " middle_gen_results_f[0][0] = [sum(x) / 2 for x in zip(middle_gen_results_f[0][0], middle_gen_results[0][0])]\n", - " last_gen_results_f[0][0] = [sum(x) / 2 for x in zip(last_gen_results_f[0][0], last_gen_results[0][0])]\n", - "\n", - " avg_training_norm_violations_f = avg_training_norm_violations_f.add(avg_training_norm_violations).div(2)\n", - " first_gen_results_f[0][1] = first_gen_results_f[0][1].add(first_gen_results[0][1]).div(2)\n", - " middle_gen_results_f[0][1] = middle_gen_results_f[0][1].add(middle_gen_results[0][1]).div(2)\n", - " last_gen_results_f[0][1] = last_gen_results_f[0][1].add(last_gen_results[0][1]).div(2)\n", - "\n", - " sim_violated_f = sim_violated_f.add(sim_violated).div(2)\n", - " at_dest_f = [sum(x) / 2 for x in zip(at_dest_f, at_dest)]\n", - "\n", - "#plot results\n", - "plt.plot(avg_training_rewards_f, label='Average training reward')\n", - "plt.plot(first_gen_results_f[0][0], label='First generation reward')\n", - "plt.plot(middle_gen_results_f[0][0], label='Middle generation reward')\n", - "plt.plot(last_gen_results_f[0][0], label='Last generation reward')\n", - "plt.legend()\n", - "plt.title('Average training reward')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Average reward')\n", - "plt.show()\n", - "\n", - "\n", - "#training violations\n", - "costs_from_violations = copy.deepcopy(avg_training_norm_violations_f['total_violations_cost'])\n", - "avg_training_norm_violations_f.drop(columns=['seed'], inplace=True)\n", - "avg_training_norm_violations_f.drop(columns=['total_violations_cost'], inplace=True)\n", - "\n", - "#graph costs from violations\n", - "plt.plot(costs_from_violations)\n", - "plt.title('AVERAGE Costs from violations training')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "avg_training_norm_violations_f.plot(kind='bar', stacked=True)\n", - "plt.title('Average training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "plt.show()\n", - "\n", - "#FIRST GEN TRAINING VIOLATIONS \n", - "costs_from_violations = copy.deepcopy(first_gen_results_f[0][1]['total_violations_cost'])\n", - "first_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", - "first_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", - "\n", - "#graph costs from violations\n", - "plt.plot(costs_from_violations)\n", - "plt.title('First generation Costs from violations training')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "first_gen_results_f[0][1].plot(kind='bar', stacked=True)\n", - "plt.title('First generation training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "plt.show()\n", - "\n", - "#MIDDLE GEN TRAINING VIOLATIONS\n", - "costs_from_violations = copy.deepcopy(middle_gen_results_f[0][1]['total_violations_cost'])\n", - "middle_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", - "middle_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", - "plt.plot(costs_from_violations)\n", - "plt.title('Middle generation Costs from violations training')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "middle_gen_results_f[0][1].plot(kind='bar', stacked=True)\n", - "plt.title('Middle generation training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "\n", - "plt.show()\n", - "\n", - "#LAST GEN TRAINING VIOLATIONS\n", - "costs_from_violations = copy.deepcopy(last_gen_results_f[0][1]['total_violations_cost'])\n", - "last_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", - "last_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", - "plt.plot(costs_from_violations)\n", - "plt.title('Last generation Costs from violations training')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "last_gen_results_f[0][1].plot(kind='bar', stacked=True)\n", - "plt.title('Last generation training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "plt.show()\n", - "\n", - "\n", - "\n", - "\n", - "#training violations\n", - "costs_from_violations_f = copy.deepcopy(sim_violated_f['total_violations_cost'])\n", - "sim_violated_f.drop(columns=['seed'], inplace=True)\n", - "sim_violated_f.drop(columns=['total_violations_cost'], inplace=True)\n", - "\n", - "#graph costs from violations\n", - "plt.plot(costs_from_violations)\n", - "plt.title('Costs from violations simulation')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", - "plt.show()\n", - "\n", - "# Plot a stacked bar chart\n", - "sim_violated_f.plot(kind='bar', stacked=True)\n", - "plt.title('Average simulation norm violations')\n", - "plt.xlabel('Simulation Run')\n", - "plt.ylabel('Average norm violations')\n", - "plt.show()\n", - "\n", - "#simulation at destination, avg timeto destination\n", - "plt.plot(at_dest_f)\n", - "plt.title('Average Steps to destination')\n", - "plt.xlabel('Simulation Run')\n", - "plt.ylabel('Steps ')\n", - "plt.show()\n", - "\n", - "\n", - "\n", - "\n" - ] - }, - { - "cell_type": "markdown", - "metadata": {}, - "source": [ - "## 10 Agents to 4 Transfer" - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": {}, - "outputs": [ { "name": "stderr", "output_type": "stream", @@ -127021,1173 +82429,15906 @@ "name": "stdout", "output_type": "stream", "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", + "Solving for Nash Equilibrium in Generation 1/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 2/300\n", + "Solving for Nash Equilibrium in Generation 2/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 3/300\n", + "Solving for Nash Equilibrium in Generation 3/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 4/300\n", + "Solving for Nash Equilibrium in Generation 4/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 5/300\n", + "Solving for Nash Equilibrium in Generation 5/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 6/300\n", + "Solving for Nash Equilibrium in Generation 6/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 7/300\n", + "Solving for Nash Equilibrium in Generation 7/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 8/300\n", + "Solving for Nash Equilibrium in Generation 8/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 9/300\n", + "Solving for Nash Equilibrium in Generation 9/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 10/300\n", + "Solving for Nash Equilibrium in Generation 10/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 11/300\n", + "Solving for Nash Equilibrium in Generation 11/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 12/300\n", + "Solving for Nash Equilibrium in Generation 12/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 13/300\n", + "Solving for Nash Equilibrium in Generation 13/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 14/300\n", + "Solving for Nash Equilibrium in Generation 14/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 15/300\n", + "Solving for Nash Equilibrium in Generation 15/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 16/300\n", + "Solving for Nash Equilibrium in Generation 16/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 17/300\n", + "Solving for Nash Equilibrium in Generation 17/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 18/300\n", + "Solving for Nash Equilibrium in Generation 18/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 19/300\n", + "Solving for Nash Equilibrium in Generation 19/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 20/300\n", + "Solving for Nash Equilibrium in Generation 20/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 21/300\n", + "Solving for Nash Equilibrium in Generation 21/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 22/300\n", + "Solving for Nash Equilibrium in Generation 22/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 23/300\n", + "Solving for Nash Equilibrium in Generation 23/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 24/300\n", + "Solving for Nash Equilibrium in Generation 24/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 25/300\n", + "Solving for Nash Equilibrium in Generation 25/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 26/300\n", + "Solving for Nash Equilibrium in Generation 26/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 27/300\n", + "Solving for Nash Equilibrium in Generation 27/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 28/300\n", + "Solving for Nash Equilibrium in Generation 28/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 29/300\n", + "Solving for Nash Equilibrium in Generation 29/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 30/300\n", + "Solving for Nash Equilibrium in Generation 30/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 31/300\n", + "Solving for Nash Equilibrium in Generation 31/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 32/300\n", + "Solving for Nash Equilibrium in Generation 32/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 33/300\n", + "Solving for Nash Equilibrium in Generation 33/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 34/300\n", + "Solving for Nash Equilibrium in Generation 34/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 35/300\n", + "Solving for Nash Equilibrium in Generation 35/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 36/300\n", + "Solving for Nash Equilibrium in Generation 36/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 37/300\n", + "Solving for Nash Equilibrium in Generation 37/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 38/300\n", + "Solving for Nash Equilibrium in Generation 38/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 39/300\n", + "Solving for Nash Equilibrium in Generation 39/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 40/300\n", + "Solving for Nash Equilibrium in Generation 40/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 41/300\n", + "Solving for Nash Equilibrium in Generation 41/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 42/300\n", + "Solving for Nash Equilibrium in Generation 42/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 43/300\n", + "Solving for Nash Equilibrium in Generation 43/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 44/300\n", + "Solving for Nash Equilibrium in Generation 44/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 45/300\n", + "Solving for Nash Equilibrium in Generation 45/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 46/300\n", + "Solving for Nash Equilibrium in Generation 46/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 47/300\n", + "Solving for Nash Equilibrium in Generation 47/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 48/300\n", + "Solving for Nash Equilibrium in Generation 48/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 49/300\n", + "Solving for Nash Equilibrium in Generation 49/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 50/300\n", + "Solving for Nash Equilibrium in Generation 50/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 51/300\n", + "Solving for Nash Equilibrium in Generation 51/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 52/300\n", + "Solving for Nash Equilibrium in Generation 52/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 53/300\n", + "Solving for Nash Equilibrium in Generation 53/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 54/300\n", + "Solving for Nash Equilibrium in Generation 54/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 55/300\n", + "Solving for Nash Equilibrium in Generation 55/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 56/300\n", + "Solving for Nash Equilibrium in Generation 56/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 57/300\n", + "Solving for Nash Equilibrium in Generation 57/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 58/300\n", + "Solving for Nash Equilibrium in Generation 58/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 59/300\n", + "Solving for Nash Equilibrium in Generation 59/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 60/300\n", + "Solving for Nash Equilibrium in Generation 60/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 61/300\n", + "Solving for Nash Equilibrium in Generation 61/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 62/300\n", + "Solving for Nash Equilibrium in Generation 62/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 63/300\n", + "Solving for Nash Equilibrium in Generation 63/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 64/300\n", + "Solving for Nash Equilibrium in Generation 64/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 65/300\n", + "Solving for Nash Equilibrium in Generation 65/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 66/300\n", + "Solving for Nash Equilibrium in Generation 66/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 67/300\n", + "Solving for Nash Equilibrium in Generation 67/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 68/300\n", + "Solving for Nash Equilibrium in Generation 68/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 69/300\n", + "Solving for Nash Equilibrium in Generation 69/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 70/300\n", + "Solving for Nash Equilibrium in Generation 70/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 71/300\n", + "Solving for Nash Equilibrium in Generation 71/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 72/300\n", + "Solving for Nash Equilibrium in Generation 72/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 73/300\n", + "Solving for Nash Equilibrium in Generation 73/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 74/300\n", + "Solving for Nash Equilibrium in Generation 74/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 75/300\n", + "Solving for Nash Equilibrium in Generation 75/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 76/300\n", + "Solving for Nash Equilibrium in Generation 76/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 77/300\n", + "Solving for Nash Equilibrium in Generation 77/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 78/300\n", + "Solving for Nash Equilibrium in Generation 78/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 79/300\n", + "Solving for Nash Equilibrium in Generation 79/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 80/300\n", + "Solving for Nash Equilibrium in Generation 80/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 81/300\n", + "Solving for Nash Equilibrium in Generation 81/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 82/300\n", + "Solving for Nash Equilibrium in Generation 82/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 83/300\n", + "Solving for Nash Equilibrium in Generation 83/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 84/300\n", + "Solving for Nash Equilibrium in Generation 84/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 85/300\n", + "Solving for Nash Equilibrium in Generation 85/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 86/300\n", + "Solving for Nash Equilibrium in Generation 86/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 87/300\n", + "Solving for Nash Equilibrium in Generation 87/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 88/300\n", + "Solving for Nash Equilibrium in Generation 88/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 89/300\n", + "Solving for Nash Equilibrium in Generation 89/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 90/300\n", + "Solving for Nash Equilibrium in Generation 90/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 91/300\n", + "Solving for Nash Equilibrium in Generation 91/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 92/300\n", + "Solving for Nash Equilibrium in Generation 92/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 93/300\n", + "Solving for Nash Equilibrium in Generation 93/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 94/300\n", + "Solving for Nash Equilibrium in Generation 94/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 95/300\n", + "Solving for Nash Equilibrium in Generation 95/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 96/300\n", + "Solving for Nash Equilibrium in Generation 96/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 97/300\n", + "Solving for Nash Equilibrium in Generation 97/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 98/300\n", + "Solving for Nash Equilibrium in Generation 98/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 99/300\n", + "Solving for Nash Equilibrium in Generation 99/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 100/300\n", + "Solving for Nash Equilibrium in Generation 100/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 101/300\n", + "Solving for Nash Equilibrium in Generation 101/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 102/300\n", + "Solving for Nash Equilibrium in Generation 102/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 103/300\n", + "Solving for Nash Equilibrium in Generation 103/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 104/300\n", + "Solving for Nash Equilibrium in Generation 104/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 105/300\n", + "Solving for Nash Equilibrium in Generation 105/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 106/300\n", + "Solving for Nash Equilibrium in Generation 106/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 107/300\n", + "Solving for Nash Equilibrium in Generation 107/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 108/300\n", + "Solving for Nash Equilibrium in Generation 108/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 109/300\n", + "Solving for Nash Equilibrium in Generation 109/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 110/300\n", + "Solving for Nash Equilibrium in Generation 110/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 111/300\n", + "Solving for Nash Equilibrium in Generation 111/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 112/300\n", + "Solving for Nash Equilibrium in Generation 112/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 113/300\n", + "Solving for Nash Equilibrium in Generation 113/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 114/300\n", + "Solving for Nash Equilibrium in Generation 114/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 115/300\n", + "Solving for Nash Equilibrium in Generation 115/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 116/300\n", + "Solving for Nash Equilibrium in Generation 116/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 117/300\n", + "Solving for Nash Equilibrium in Generation 117/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 118/300\n", + "Solving for Nash Equilibrium in Generation 118/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 119/300\n", + "Solving for Nash Equilibrium in Generation 119/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 120/300\n", + "Solving for Nash Equilibrium in Generation 120/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 121/300\n", + "Solving for Nash Equilibrium in Generation 121/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 122/300\n", + "Solving for Nash Equilibrium in Generation 122/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 123/300\n", + "Solving for Nash Equilibrium in Generation 123/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 124/300\n", + "Solving for Nash Equilibrium in Generation 124/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 125/300\n", + "Solving for Nash Equilibrium in Generation 125/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 126/300\n", + "Solving for Nash Equilibrium in Generation 126/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 127/300\n", + "Solving for Nash Equilibrium in Generation 127/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 128/300\n", + "Solving for Nash Equilibrium in Generation 128/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 129/300\n", + "Solving for Nash Equilibrium in Generation 129/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 130/300\n", + "Solving for Nash Equilibrium in Generation 130/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 131/300\n", + "Solving for Nash Equilibrium in Generation 131/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 132/300\n", + "Solving for Nash Equilibrium in Generation 132/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 133/300\n", + "Solving for Nash Equilibrium in Generation 133/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 134/300\n", + "Solving for Nash Equilibrium in Generation 134/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 135/300\n", + "Solving for Nash Equilibrium in Generation 135/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 136/300\n", + "Solving for Nash Equilibrium in Generation 136/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 137/300\n", + "Solving for Nash Equilibrium in Generation 137/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 138/300\n", + "Solving for Nash Equilibrium in Generation 138/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 139/300\n", + "Solving for Nash Equilibrium in Generation 139/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 140/300\n", + "Solving for Nash Equilibrium in Generation 140/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 141/300\n", + "Solving for Nash Equilibrium in Generation 141/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 142/300\n", + "Solving for Nash Equilibrium in Generation 142/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 143/300\n", + "Solving for Nash Equilibrium in Generation 143/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 144/300\n", + "Solving for Nash Equilibrium in Generation 144/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 145/300\n", + "Solving for Nash Equilibrium in Generation 145/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 146/300\n", + "Solving for Nash Equilibrium in Generation 146/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 147/300\n", + "Solving for Nash Equilibrium in Generation 147/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 148/300\n", + "Solving for Nash Equilibrium in Generation 148/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 149/300\n", + "Solving for Nash Equilibrium in Generation 149/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 150/300\n", + "Solving for Nash Equilibrium in Generation 150/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 151/300\n", + "Solving for Nash Equilibrium in Generation 151/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 152/300\n", + "Solving for Nash Equilibrium in Generation 152/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 153/300\n", + "Solving for Nash Equilibrium in Generation 153/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 154/300\n", + "Solving for Nash Equilibrium in Generation 154/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 155/300\n", + "Solving for Nash Equilibrium in Generation 155/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 156/300\n", + "Solving for Nash Equilibrium in Generation 156/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 157/300\n", + "Solving for Nash Equilibrium in Generation 157/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 158/300\n", + "Solving for Nash Equilibrium in Generation 158/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 159/300\n", + "Solving for Nash Equilibrium in Generation 159/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 160/300\n", + "Solving for Nash Equilibrium in Generation 160/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 161/300\n", + "Solving for Nash Equilibrium in Generation 161/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 162/300\n", + "Solving for Nash Equilibrium in Generation 162/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 163/300\n", + "Solving for Nash Equilibrium in Generation 163/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 164/300\n", + "Solving for Nash Equilibrium in Generation 164/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 165/300\n", + "Solving for Nash Equilibrium in Generation 165/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 166/300\n", + "Solving for Nash Equilibrium in Generation 166/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 167/300\n", + "Solving for Nash Equilibrium in Generation 167/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 168/300\n", + "Solving for Nash Equilibrium in Generation 168/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 169/300\n", + "Solving for Nash Equilibrium in Generation 169/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 170/300\n", + "Solving for Nash Equilibrium in Generation 170/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 171/300\n", + "Solving for Nash Equilibrium in Generation 171/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 172/300\n", + "Solving for Nash Equilibrium in Generation 172/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 173/300\n", + "Solving for Nash Equilibrium in Generation 173/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 174/300\n", + "Solving for Nash Equilibrium in Generation 174/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 175/300\n", + "Solving for Nash Equilibrium in Generation 175/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 176/300\n", + "Solving for Nash Equilibrium in Generation 176/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 177/300\n", + "Solving for Nash Equilibrium in Generation 177/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 178/300\n", + "Solving for Nash Equilibrium in Generation 178/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 179/300\n", + "Solving for Nash Equilibrium in Generation 179/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 180/300\n", + "Solving for Nash Equilibrium in Generation 180/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 181/300\n", + "Solving for Nash Equilibrium in Generation 181/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 182/300\n", + "Solving for Nash Equilibrium in Generation 182/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 183/300\n", + "Solving for Nash Equilibrium in Generation 183/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 184/300\n", + "Solving for Nash Equilibrium in Generation 184/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 185/300\n", + "Solving for Nash Equilibrium in Generation 185/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 186/300\n", + "Solving for Nash Equilibrium in Generation 186/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 187/300\n", + "Solving for Nash Equilibrium in Generation 187/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 188/300\n", + "Solving for Nash Equilibrium in Generation 188/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 189/300\n", + "Solving for Nash Equilibrium in Generation 189/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 190/300\n", + "Solving for Nash Equilibrium in Generation 190/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 191/300\n", + "Solving for Nash Equilibrium in Generation 191/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 192/300\n", + "Solving for Nash Equilibrium in Generation 192/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 193/300\n", + "Solving for Nash Equilibrium in Generation 193/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 194/300\n", + "Solving for Nash Equilibrium in Generation 194/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 195/300\n", + "Solving for Nash Equilibrium in Generation 195/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 196/300\n", + "Solving for Nash Equilibrium in Generation 196/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 197/300\n", + "Solving for Nash Equilibrium in Generation 197/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 198/300\n", + "Solving for Nash Equilibrium in Generation 198/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 199/300\n", + "Solving for Nash Equilibrium in Generation 199/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 200/300\n", + "Solving for Nash Equilibrium in Generation 200/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 201/300\n", + "Solving for Nash Equilibrium in Generation 201/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 202/300\n", + "Solving for Nash Equilibrium in Generation 202/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 203/300\n", + "Solving for Nash Equilibrium in Generation 203/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 204/300\n", + "Solving for Nash Equilibrium in Generation 204/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 205/300\n", + "Solving for Nash Equilibrium in Generation 205/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 206/300\n", + "Solving for Nash Equilibrium in Generation 206/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 207/300\n", + "Solving for Nash Equilibrium in Generation 207/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 208/300\n", + "Solving for Nash Equilibrium in Generation 208/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 209/300\n", + "Solving for Nash Equilibrium in Generation 209/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 210/300\n", + "Solving for Nash Equilibrium in Generation 210/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 211/300\n", + "Solving for Nash Equilibrium in Generation 211/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 212/300\n", + "Solving for Nash Equilibrium in Generation 212/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 213/300\n", + "Solving for Nash Equilibrium in Generation 213/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 214/300\n", + "Solving for Nash Equilibrium in Generation 214/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 215/300\n", + "Solving for Nash Equilibrium in Generation 215/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 216/300\n", + "Solving for Nash Equilibrium in Generation 216/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 217/300\n", + "Solving for Nash Equilibrium in Generation 217/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 218/300\n", + "Solving for Nash Equilibrium in Generation 218/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 219/300\n", + "Solving for Nash Equilibrium in Generation 219/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 220/300\n", + "Solving for Nash Equilibrium in Generation 220/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 221/300\n", + "Solving for Nash Equilibrium in Generation 221/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 222/300\n", + "Solving for Nash Equilibrium in Generation 222/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 223/300\n", + "Solving for Nash Equilibrium in Generation 223/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 224/300\n", + "Solving for Nash Equilibrium in Generation 224/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 225/300\n", + "Solving for Nash Equilibrium in Generation 225/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 226/300\n", + "Solving for Nash Equilibrium in Generation 226/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 227/300\n", + "Solving for Nash Equilibrium in Generation 227/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 228/300\n", + "Solving for Nash Equilibrium in Generation 228/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 229/300\n", + "Solving for Nash Equilibrium in Generation 229/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 230/300\n", + "Solving for Nash Equilibrium in Generation 230/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 231/300\n", + "Solving for Nash Equilibrium in Generation 231/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 232/300\n", + "Solving for Nash Equilibrium in Generation 232/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 233/300\n", + "Solving for Nash Equilibrium in Generation 233/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 234/300\n", + "Solving for Nash Equilibrium in Generation 234/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 235/300\n", + "Solving for Nash Equilibrium in Generation 235/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 236/300\n", + "Solving for Nash Equilibrium in Generation 236/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 237/300\n", + "Solving for Nash Equilibrium in Generation 237/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 238/300\n", + "Solving for Nash Equilibrium in Generation 238/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 239/300\n", + "Solving for Nash Equilibrium in Generation 239/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 240/300\n", + "Solving for Nash Equilibrium in Generation 240/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 241/300\n", + "Solving for Nash Equilibrium in Generation 241/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 242/300\n", + "Solving for Nash Equilibrium in Generation 242/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 243/300\n", + "Solving for Nash Equilibrium in Generation 243/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 244/300\n", + "Solving for Nash Equilibrium in Generation 244/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 245/300\n", + "Solving for Nash Equilibrium in Generation 245/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 246/300\n", + "Solving for Nash Equilibrium in Generation 246/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 247/300\n", + "Solving for Nash Equilibrium in Generation 247/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 248/300\n", + "Solving for Nash Equilibrium in Generation 248/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 249/300\n", + "Solving for Nash Equilibrium in Generation 249/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 250/300\n", + "Solving for Nash Equilibrium in Generation 250/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 251/300\n", + "Solving for Nash Equilibrium in Generation 251/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 252/300\n", + "Solving for Nash Equilibrium in Generation 252/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 253/300\n", + "Solving for Nash Equilibrium in Generation 253/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 254/300\n", + "Solving for Nash Equilibrium in Generation 254/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 255/300\n", + "Solving for Nash Equilibrium in Generation 255/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 256/300\n", + "Solving for Nash Equilibrium in Generation 256/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 257/300\n", + "Solving for Nash Equilibrium in Generation 257/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 258/300\n", + "Solving for Nash Equilibrium in Generation 258/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 259/300\n", + "Solving for Nash Equilibrium in Generation 259/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 260/300\n", + "Solving for Nash Equilibrium in Generation 260/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 261/300\n", + "Solving for Nash Equilibrium in Generation 261/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 262/300\n", + "Solving for Nash Equilibrium in Generation 262/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 263/300\n", + "Solving for Nash Equilibrium in Generation 263/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 264/300\n", + "Solving for Nash Equilibrium in Generation 264/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 265/300\n", + "Solving for Nash Equilibrium in Generation 265/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 266/300\n", + "Solving for Nash Equilibrium in Generation 266/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 267/300\n", + "Solving for Nash Equilibrium in Generation 267/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 268/300\n", + "Solving for Nash Equilibrium in Generation 268/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 269/300\n", + "Solving for Nash Equilibrium in Generation 269/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 270/300\n", + "Solving for Nash Equilibrium in Generation 270/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 271/300\n", + "Solving for Nash Equilibrium in Generation 271/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 272/300\n", + "Solving for Nash Equilibrium in Generation 272/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 273/300\n", + "Solving for Nash Equilibrium in Generation 273/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 274/300\n", + "Solving for Nash Equilibrium in Generation 274/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 275/300\n", + "Solving for Nash Equilibrium in Generation 275/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 276/300\n", + "Solving for Nash Equilibrium in Generation 276/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 277/300\n", + "Solving for Nash Equilibrium in Generation 277/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 278/300\n", + "Solving for Nash Equilibrium in Generation 278/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 279/300\n", + "Solving for Nash Equilibrium in Generation 279/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 280/300\n", + "Solving for Nash Equilibrium in Generation 280/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 281/300\n", + "Solving for Nash Equilibrium in Generation 281/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 282/300\n", + "Solving for Nash Equilibrium in Generation 282/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 283/300\n", + "Solving for Nash Equilibrium in Generation 283/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 284/300\n", + "Solving for Nash Equilibrium in Generation 284/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 285/300\n", + "Solving for Nash Equilibrium in Generation 285/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 286/300\n", + "Solving for Nash Equilibrium in Generation 286/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 287/300\n", + "Solving for Nash Equilibrium in Generation 287/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 288/300\n", + "Solving for Nash Equilibrium in Generation 288/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 289/300\n", + "Solving for Nash Equilibrium in Generation 289/300\n", "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 3 done\n", - "Episode 4 done\n", - "Episode 4 done\n", - "Total steps: 1\n", - "Episode 5 done\n", - "Episode 6 done\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 9 done\n", - " seed step_collisions not_on_track yield_violations \\\n", - "0 42.0 0.0 14.0 0.0 \n", - "1 42.0 0.0 8.0 0.0 \n", - "2 42.0 0.0 8.0 0.0 \n", - "3 42.0 0.0 8.0 0.0 \n", - "4 42.0 0.0 4.0 0.0 \n", - "5 42.0 0.0 19.0 0.0 \n", - "6 42.0 0.0 8.0 0.0 \n", - "7 42.0 0.0 8.0 0.0 \n", - "8 42.0 1.0 12.0 0.0 \n", - "9 42.0 0.0 7.0 0.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 188.0 \n", - "1 0.0 118.0 \n", - "2 0.0 118.0 \n", - "3 0.0 118.0 \n", - "4 0.0 0.0 \n", - "5 0.0 282.0 \n", - "6 0.0 118.0 \n", - "7 0.0 118.0 \n", - "8 0.0 236.0 \n", - "9 0.0 142.0 \n", - "\n", - " total_violations_cost \n", - "0 -79.4 \n", - "1 -45.9 \n", - "2 -45.9 \n", - "3 -45.9 \n", - "4 -20.0 \n", - "5 -109.1 \n", - "6 -45.9 \n", - "7 -45.9 \n", - "8 -91.8 \n", - "9 -42.1 \n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 290/300\n", + "Solving for Nash Equilibrium in Generation 290/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 291/300\n", + "Solving for Nash Equilibrium in Generation 291/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 292/300\n", + "Solving for Nash Equilibrium in Generation 292/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 293/300\n", + "Solving for Nash Equilibrium in Generation 293/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 294/300\n", + "Solving for Nash Equilibrium in Generation 294/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 295/300\n", + "Solving for Nash Equilibrium in Generation 295/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 296/300\n", + "Solving for Nash Equilibrium in Generation 296/300\n", + "Computing Nash Equilibrium for 100 matches\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 297/300\n", + "Solving for Nash Equilibrium in Generation 297/300\n", "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 2 done\n", - "Total steps: 13\n", - "Episode 3 done\n", - "Episode 4 done\n", - "Episode 5 done\n", - "Episode 6 done\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 9 done\n", - " seed step_collisions not_on_track yield_violations \\\n", - "0 42.0 0.0 8.0 0.0 \n", - "1 42.0 0.0 7.0 0.0 \n", - "2 42.0 1.0 2.0 0.0 \n", - "3 42.0 0.0 14.0 0.0 \n", - "4 42.0 0.0 13.0 0.0 \n", - "5 42.0 0.0 24.0 0.0 \n", - "6 42.0 1.0 7.0 0.0 \n", - "7 42.0 0.0 18.0 0.0 \n", - "8 42.0 0.0 14.0 0.0 \n", - "9 42.0 0.0 7.0 0.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 118.0 \n", - "1 0.0 142.0 \n", - "2 0.0 48.0 \n", - "3 0.0 188.0 \n", - "4 0.0 212.0 \n", - "5 0.0 376.0 \n", - "6 0.0 142.0 \n", - "7 0.0 306.0 \n", - "8 0.0 188.0 \n", - "9 0.0 142.0 \n", - "\n", - " total_violations_cost \n", - "0 -45.9 \n", - "1 -42.1 \n", - "2 -32.4 \n", - "3 -79.4 \n", - "4 -75.6 \n", - "5 -138.8 \n", - "6 -62.1 \n", - "7 -105.3 \n", - "8 -79.4 \n", - "9 -42.1 \n", - "here\n", - "([3001.5952000000016, 4102.5579000000025, 6481.339750000003, 6701.127499999986, 8950.818599999984, 10192.108349999986, 11073.238349999985, 12979.160299999987, 15990.34934999999, 17832.420349999993], seed step_collisions not_on_track yield_violations \\\n", - "0 1.0 3.0 20.0 1.0 \n", - "1 1.0 5.0 8.0 1.0 \n", - "2 1.0 4.0 12.0 1.0 \n", - "3 1.0 4.0 19.0 1.0 \n", - "4 1.0 5.0 21.0 3.0 \n", - "5 1.0 4.0 15.0 3.0 \n", - "6 1.0 5.0 20.0 3.0 \n", - "7 1.0 6.0 13.0 1.0 \n", - "8 1.0 4.0 20.0 2.0 \n", - "9 1.0 5.0 13.0 3.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 675.0 \n", - "1 0.0 360.0 \n", - "2 0.0 436.0 \n", - "3 0.0 689.0 \n", - "4 0.0 1001.0 \n", - "5 0.0 647.0 \n", - "6 0.0 1013.0 \n", - "7 0.0 519.0 \n", - "8 0.0 669.0 \n", - "9 0.0 533.0 \n", - "\n", - " total_violations_cost \n", - "0 -195.75 \n", - "1 -160.00 \n", - "2 -163.80 \n", - "3 -211.45 \n", - "4 -261.05 \n", - "5 -193.35 \n", - "6 -256.65 \n", - "7 -212.95 \n", - "8 -217.45 \n", - "9 -197.65 )\n", - "here1\n", - "[3001.5952000000016, 4102.5579000000025, 6481.339750000003, 6701.127499999986, 8950.818599999984, 10192.108349999986, 11073.238349999985, 12979.160299999987, 15990.34934999999, 17832.420349999993]\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 298/300\n", + "Solving for Nash Equilibrium in Generation 298/300\n", "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 3 done\n", - "Episode 4 done\n", - "Episode 5 done\n", - "Episode 6 done\n", - "Episode 6 done\n", - "Total steps: 13\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 8 done\n", - "Total steps: 13\n", - "Episode 9 done\n", - "Episode 9 done\n", - "Total steps: 13\n", - " seed step_collisions not_on_track yield_violations \\\n", - "0 42.0 0.0 8.0 0.0 \n", - "1 42.0 0.0 13.0 0.0 \n", - "2 42.0 0.0 7.0 0.0 \n", - "3 42.0 1.0 6.0 0.0 \n", - "4 42.0 0.0 13.0 0.0 \n", - "5 42.0 0.0 8.0 0.0 \n", - "6 42.0 4.0 0.0 0.0 \n", - "7 42.0 0.0 13.0 0.0 \n", - "8 42.0 0.0 3.0 0.0 \n", - "9 42.0 0.0 1.0 0.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 118.0 \n", - "1 0.0 212.0 \n", - "2 0.0 142.0 \n", - "3 0.0 166.0 \n", - "4 0.0 212.0 \n", - "5 0.0 118.0 \n", - "6 0.0 96.0 \n", - "7 0.0 212.0 \n", - "8 0.0 24.0 \n", - "9 0.0 72.0 \n", - "\n", - " total_violations_cost \n", - "0 -45.9 \n", - "1 -75.6 \n", - "2 -42.1 \n", - "3 -58.3 \n", - "4 -75.6 \n", - "5 -45.9 \n", - "6 -84.8 \n", - "7 -75.6 \n", - "8 -16.2 \n", - "9 -8.6 \n", - "here\n", - "([333.0485499999994, 2578.5020000000004, 3026.334050000007, 17151.488611111163, 4213.272900000007, 24020.01555555559, 2817.4908500000074, 1725.2031500000078, 2816.349650000008, 5842.494050000009], seed step_collisions not_on_track yield_violations \\\n", - "0 2.0 5.0 18.0 0.0 \n", - "1 2.0 4.0 20.0 1.0 \n", - "2 2.0 7.0 12.0 4.0 \n", - "3 2.0 5.0 16.0 1.0 \n", - "4 2.0 4.0 22.0 1.0 \n", - "5 2.0 6.0 14.0 3.0 \n", - "6 2.0 3.0 10.0 0.0 \n", - "7 2.0 5.0 24.0 3.0 \n", - "8 2.0 7.0 9.0 2.0 \n", - "9 2.0 3.0 20.0 0.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 422.0 \n", - "1 0.0 663.0 \n", - "2 0.0 881.0 \n", - "3 0.0 642.0 \n", - "4 0.0 613.0 \n", - "5 0.0 679.0 \n", - "6 0.0 316.0 \n", - "7 0.0 891.0 \n", - "8 0.0 338.0 \n", - "9 0.0 675.0 \n", - "\n", - " total_violations_cost \n", - "0 -211.10 \n", - "1 -215.15 \n", - "2 -252.05 \n", - "3 -214.10 \n", - "4 -222.65 \n", - "5 -229.95 \n", - "6 -125.80 \n", - "7 -270.55 \n", - "8 -205.90 \n", - "9 -193.75 )\n", - "here1\n", - "[333.0485499999994, 2578.5020000000004, 3026.334050000007, 17151.488611111163, 4213.272900000007, 24020.01555555559, 2817.4908500000074, 1725.2031500000078, 2816.349650000008, 5842.494050000009]\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 299/300\n", + "Solving for Nash Equilibrium in Generation 299/300\n", "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 0 done\n", - "Episode 1 done\n", - "Episode 2 done\n", - "Episode 3 done\n", - "Episode 4 done\n", - "Episode 5 done\n", - "Episode 5 done\n", - "Total steps: 13\n", - "Episode 6 done\n", - "Episode 7 done\n", - "Episode 8 done\n", - "Episode 9 done\n", - "Episode 9 done\n", - "Total steps: 13\n", - " seed step_collisions not_on_track yield_violations \\\n", - "0 42.0 1.0 7.0 0.0 \n", - "1 42.0 1.0 6.0 0.0 \n", - "2 42.0 0.0 12.0 0.0 \n", - "3 42.0 0.0 12.0 0.0 \n", - "4 42.0 0.0 8.0 0.0 \n", - "5 42.0 1.0 2.0 0.0 \n", - "6 42.0 0.0 8.0 0.0 \n", - "7 42.0 0.0 12.0 0.0 \n", - "8 42.0 1.0 7.0 0.0 \n", - "9 42.0 1.0 2.0 0.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 142.0 \n", - "1 0.0 166.0 \n", - "2 0.0 236.0 \n", - "3 0.0 236.0 \n", - "4 0.0 118.0 \n", - "5 0.0 48.0 \n", - "6 0.0 118.0 \n", - "7 0.0 236.0 \n", - "8 0.0 142.0 \n", - "9 0.0 48.0 \n", - "\n", - " total_violations_cost \n", - "0 -62.1 \n", - "1 -58.3 \n", - "2 -71.8 \n", - "3 -71.8 \n", - "4 -45.9 \n", - "5 -32.4 \n", - "6 -45.9 \n", - "7 -71.8 \n", - "8 -62.1 \n", - "9 -32.4 \n", - "here\n", - "([2064.427600000002, 3184.9379000000017, 5206.963000000002, 6515.829050000002, 9333.194000000003, 11958.051350000002, 13805.627550000005, 15036.535450000005, 17981.56380000001, 20019.83765000001], seed step_collisions not_on_track yield_violations \\\n", - "0 3.0 4.0 15.0 2.0 \n", - "1 3.0 5.0 14.0 2.0 \n", - "2 3.0 6.0 10.0 1.0 \n", - "3 3.0 8.0 2.0 4.0 \n", - "4 3.0 6.0 14.0 3.0 \n", - "5 3.0 6.0 9.0 1.0 \n", - "6 3.0 5.0 24.0 4.0 \n", - "7 3.0 3.0 13.0 3.0 \n", - "8 3.0 4.0 9.0 2.0 \n", - "9 3.0 5.0 21.0 3.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 649.0 \n", - "1 0.0 679.0 \n", - "2 0.0 312.0 \n", - "3 0.0 192.0 \n", - "4 0.0 675.0 \n", - "5 0.0 498.0 \n", - "6 0.0 1405.0 \n", - "7 0.0 697.0 \n", - "8 0.0 512.0 \n", - "9 0.0 979.0 \n", - "\n", - " total_violations_cost \n", - "0 -191.45 \n", - "1 -207.95 \n", - "2 -187.60 \n", - "3 -187.60 \n", - "4 -229.75 \n", - "5 -191.90 \n", - "6 -298.25 \n", - "7 -165.85 \n", - "8 -154.60 \n", - "9 -259.95 )\n", - "here1\n", - "[2064.427600000002, 3184.9379000000017, 5206.963000000002, 6515.829050000002, 9333.194000000003, 11958.051350000002, 13805.627550000005, 15036.535450000005, 17981.56380000001, 20019.83765000001]\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Overriding environment TrafficJunction4-v0\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n", - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Generation 1/10\n", - "Solving for Nash Equilibrium in Generation 1/10\n", + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Generation 300/300\n", + "Solving for Nash Equilibrium in Generation 300/300\n", "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/keras/src/engine/training_v1.py:2359: UserWarning: `Model.state_updates` will be removed in a future version. This property should not be used in TensorFlow 2.0, as `updates` are applied automatically.\n", - " updates=self.state_updates,\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 2/10\n", - "Solving for Nash Equilibrium in Generation 2/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 3/10\n", - "Solving for Nash Equilibrium in Generation 3/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 4/10\n", - "Solving for Nash Equilibrium in Generation 4/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 5/10\n", - "Solving for Nash Equilibrium in Generation 5/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 6/10\n", - "Solving for Nash Equilibrium in Generation 6/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 7/10\n", - "Solving for Nash Equilibrium in Generation 7/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 8/10\n", - "Solving for Nash Equilibrium in Generation 8/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 9/10\n", - "Solving for Nash Equilibrium in Generation 9/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Generation 10/10\n", - "Solving for Nash Equilibrium in Generation 10/10\n", - "Computing Nash Equilibrium for 100 matches\n", - "Episode 1/10\n", - "Episode 2/10\n", - "Episode 3/10\n", - "Episode 4/10\n", - "Episode 5/10\n", - "Episode 6/10\n", - "Episode 7/10\n", - "Episode 8/10\n", - "Episode 9/10\n", - "Episode 10/10\n", - "Running Computed Policies\n" - ] - }, - { - "name": "stderr", - "output_type": "stream", - "text": [ - "/opt/homebrew/lib/python3.11/site-packages/gym/logger.py:34: UserWarning: \u001b[33mWARN: Box bound precision lowered by casting to float32\u001b[0m\n", - " warnings.warn(colorize(\"%s: %s\" % (\"WARN\", msg % args), \"yellow\"))\n" - ] - }, - { - "name": "stdout", - "output_type": "stream", - "text": [ + "Episode 1/50\n", + "Episode 2/50\n", + "Episode 3/50\n", + "Episode 4/50\n", + "Episode 5/50\n", + "Episode 6/50\n", + "Episode 7/50\n", + "Episode 8/50\n", + "Episode 9/50\n", + "Episode 10/50\n", + "Episode 11/50\n", + "Episode 12/50\n", + "Episode 13/50\n", + "Episode 14/50\n", + "Episode 15/50\n", + "Episode 16/50\n", + "Episode 17/50\n", + "Episode 18/50\n", + "Episode 19/50\n", + "Episode 20/50\n", + "Episode 21/50\n", + "Episode 22/50\n", + "Episode 23/50\n", + "Episode 24/50\n", + "Episode 25/50\n", + "Episode 26/50\n", + "Episode 27/50\n", + "Episode 28/50\n", + "Episode 29/50\n", + "Episode 30/50\n", + "Episode 31/50\n", + "Episode 32/50\n", + "Episode 33/50\n", + "Episode 34/50\n", + "Episode 35/50\n", + "Episode 36/50\n", + "Episode 37/50\n", + "Episode 38/50\n", + "Episode 39/50\n", + "Episode 40/50\n", + "Episode 41/50\n", + "Episode 42/50\n", + "Episode 43/50\n", + "Episode 44/50\n", + "Episode 45/50\n", + "Episode 46/50\n", + "Episode 47/50\n", + "Episode 48/50\n", + "Episode 49/50\n", + "Episode 50/50\n", + "Running Computed Policies\n", "Episode 0 done\n", "Episode 1 done\n", "Episode 2 done\n", @@ -128198,84 +98339,307 @@ "Episode 7 done\n", "Episode 8 done\n", "Episode 9 done\n", - " seed step_collisions not_on_track yield_violations \\\n", - "0 42.0 0.0 6.0 0.0 \n", - "1 42.0 0.0 12.0 0.0 \n", - "2 42.0 0.0 14.0 0.0 \n", - "3 42.0 0.0 13.0 0.0 \n", - "4 42.0 1.0 7.0 0.0 \n", - "5 42.0 1.0 7.0 0.0 \n", - "6 42.0 0.0 9.0 0.0 \n", - "7 42.0 0.0 8.0 0.0 \n", - "8 42.0 2.0 6.0 0.0 \n", - "9 42.0 0.0 8.0 0.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 166.0 \n", - "1 0.0 236.0 \n", - "2 0.0 188.0 \n", - "3 0.0 212.0 \n", - "4 0.0 142.0 \n", - "5 0.0 142.0 \n", - "6 0.0 94.0 \n", - "7 0.0 118.0 \n", - "8 0.0 166.0 \n", - "9 0.0 118.0 \n", - "\n", - " total_violations_cost \n", - "0 -38.3 \n", - "1 -71.8 \n", - "2 -79.4 \n", - "3 -75.6 \n", - "4 -62.1 \n", - "5 -62.1 \n", - "6 -49.7 \n", - "7 -45.9 \n", - "8 -78.3 \n", - "9 -45.9 \n", - "here\n", - "([1402.2654999999988, 2524.876699999998, 3971.5589999999966, 6464.942549999997, 9316.507349999996, 11153.124199999995, 12856.672999999992, 14546.66429999999, 16283.509499999991, 18864.406099999993], seed step_collisions not_on_track yield_violations \\\n", - "0 4.0 3.0 15.0 1.0 \n", - "1 4.0 4.0 17.0 0.0 \n", - "2 4.0 3.0 8.0 3.0 \n", - "3 4.0 5.0 20.0 3.0 \n", - "4 4.0 5.0 18.0 0.0 \n", - "5 4.0 15.0 15.0 0.0 \n", - "6 4.0 5.0 5.0 1.0 \n", - "7 4.0 2.0 19.0 2.0 \n", - "8 4.0 6.0 23.0 0.0 \n", - "9 4.0 4.0 26.0 1.0 \n", - "\n", - " unncessary_brake_violations efficient_crossing_violations \\\n", - "0 0.0 502.0 \n", - "1 0.0 448.0 \n", - "2 0.0 524.0 \n", - "3 0.0 1212.0 \n", - "4 0.0 428.0 \n", - "5 0.0 508.0 \n", - "6 0.0 120.0 \n", - "7 0.0 1054.0 \n", - "8 0.0 593.0 \n", - "9 0.0 1013.0 \n", - "\n", - " total_violations_cost \n", - "0 -162.10 \n", - "1 -187.40 \n", - "2 -132.20 \n", - "3 -266.60 \n", - "4 -211.40 \n", - "5 -400.40 \n", - "6 -133.00 \n", - "7 -191.70 \n", - "8 -264.65 \n", - "9 -262.65 )\n", - "here1\n", - "[1402.2654999999988, 2524.876699999998, 3971.5589999999966, 6464.942549999997, 9316.507349999996, 11153.124199999995, 12856.672999999992, 14546.66429999999, 16283.509499999991, 18864.406099999993]\n" + "Episode 10 done\n", + "Episode 11 done\n", + "Episode 12 done\n", + "Episode 13 done\n", + "Episode 14 done\n", + "Episode 15 done\n", + "Episode 16 done\n", + "Episode 17 done\n", + "Episode 18 done\n", + "Episode 19 done\n", + "Episode 20 done\n", + "Episode 21 done\n", + "Episode 22 done\n", + "Episode 23 done\n", + "Episode 24 done\n", + "Episode 25 done\n", + "Episode 26 done\n", + "Episode 27 done\n", + "Episode 28 done\n", + "Episode 29 done\n", + "Episode 30 done\n", + "Episode 31 done\n", + "Episode 32 done\n", + "Episode 33 done\n", + "Episode 34 done\n", + "Episode 35 done\n", + "Episode 36 done\n", + "Episode 37 done\n", + "Episode 38 done\n", + "Episode 39 done\n", + "Episode 40 done\n", + "Episode 41 done\n", + "Episode 42 done\n", + "Episode 43 done\n", + "Episode 44 done\n", + "Episode 45 done\n", + "Episode 46 done\n", + "Episode 47 done\n", + "Episode 48 done\n", + "Episode 49 done\n", + "Episode 50 done\n", + "Episode 51 done\n", + "Episode 52 done\n", + "Episode 53 done\n", + "Episode 54 done\n", + "Episode 55 done\n", + "Episode 56 done\n", + "Episode 57 done\n", + "Episode 58 done\n", + "Episode 59 done\n", + "Episode 60 done\n", + "Episode 61 done\n", + "Episode 62 done\n", + "Episode 63 done\n", + "Episode 64 done\n", + "Episode 65 done\n", + "Episode 66 done\n", + "Episode 67 done\n", + "Episode 68 done\n", + "Episode 69 done\n", + "Episode 70 done\n", + "Episode 71 done\n", + "Episode 72 done\n", + "Episode 73 done\n", + "Episode 74 done\n", + "Episode 75 done\n", + "Episode 76 done\n", + "Episode 77 done\n", + "Episode 78 done\n", + "Episode 79 done\n", + "Episode 80 done\n", + "Episode 81 done\n", + "Episode 82 done\n", + "Episode 83 done\n", + "Episode 84 done\n", + "Episode 85 done\n", + "Episode 86 done\n", + "Episode 87 done\n", + "Episode 88 done\n", + "Episode 89 done\n", + "Episode 90 done\n", + "Episode 91 done\n", + "Episode 92 done\n", + "Episode 93 done\n", + "Episode 94 done\n", + "Episode 95 done\n", + "Episode 95 done\n", + "Total steps: 17\n", + "Episode 96 done\n", + "Episode 97 done\n", + "Episode 98 done\n", + "Episode 99 done\n", + "Episode 100 done\n", + "Episode 101 done\n", + "Episode 102 done\n", + "Episode 103 done\n", + "Episode 104 done\n", + "Episode 105 done\n", + "Episode 106 done\n", + "Episode 107 done\n", + "Episode 108 done\n", + "Episode 109 done\n", + "Episode 110 done\n", + "Episode 111 done\n", + "Episode 112 done\n", + "Episode 113 done\n", + "Episode 114 done\n", + "Episode 115 done\n", + "Episode 116 done\n", + "Episode 117 done\n", + "Episode 118 done\n", + "Episode 119 done\n", + "Episode 120 done\n", + "Episode 121 done\n", + "Episode 122 done\n", + "Episode 123 done\n", + "Episode 124 done\n", + "Episode 125 done\n", + "Episode 126 done\n", + "Episode 127 done\n", + "Episode 127 done\n", + "Total steps: 19\n", + "Episode 128 done\n", + "Episode 129 done\n", + "Episode 130 done\n", + "Episode 131 done\n", + "Episode 132 done\n", + "Episode 133 done\n", + "Episode 134 done\n", + "Episode 135 done\n", + "Episode 136 done\n", + "Episode 137 done\n", + "Episode 138 done\n", + "Episode 139 done\n", + "Episode 140 done\n", + "Episode 141 done\n", + "Episode 142 done\n", + "Episode 143 done\n", + "Episode 144 done\n", + "Episode 145 done\n", + "Episode 146 done\n", + "Episode 147 done\n", + "Episode 148 done\n", + "Episode 149 done\n", + "Episode 150 done\n", + "Episode 151 done\n", + "Episode 152 done\n", + "Episode 153 done\n", + "Episode 154 done\n", + "Episode 155 done\n", + "Episode 156 done\n", + "Episode 157 done\n", + "Episode 158 done\n", + "Episode 159 done\n", + "Episode 160 done\n", + "Episode 161 done\n", + "Episode 162 done\n", + "Episode 163 done\n", + "Episode 164 done\n", + "Episode 165 done\n", + "Episode 166 done\n", + "Episode 167 done\n", + "Episode 168 done\n", + "Episode 169 done\n", + "Episode 170 done\n", + "Episode 171 done\n", + "Episode 172 done\n", + "Episode 173 done\n", + "Episode 174 done\n", + "Episode 175 done\n", + "Episode 176 done\n", + "Episode 177 done\n", + "Episode 178 done\n", + "Episode 179 done\n", + "Episode 180 done\n", + "Episode 181 done\n", + "Episode 182 done\n", + "Episode 183 done\n", + "Episode 184 done\n", + "Episode 185 done\n", + "Episode 186 done\n", + "Episode 187 done\n", + "Episode 188 done\n", + "Episode 189 done\n", + "Episode 190 done\n", + "Episode 191 done\n", + "Episode 192 done\n", + "Episode 193 done\n", + "Episode 194 done\n", + "Episode 195 done\n", + "Episode 196 done\n", + "Episode 197 done\n", + "Episode 198 done\n", + "Episode 199 done\n", + "Episode 200 done\n", + "Episode 201 done\n", + "Episode 202 done\n", + "Episode 203 done\n", + "Episode 204 done\n", + "Episode 205 done\n", + "Episode 206 done\n", + "Episode 207 done\n", + "Episode 208 done\n", + "Episode 209 done\n", + "Episode 210 done\n", + "Episode 211 done\n", + "Episode 212 done\n", + "Episode 213 done\n", + "Episode 214 done\n", + "Episode 215 done\n", + "Episode 216 done\n", + "Episode 217 done\n", + "Episode 218 done\n", + "Episode 219 done\n", + "Episode 220 done\n", + "Episode 221 done\n", + "Episode 222 done\n", + "Episode 223 done\n", + "Episode 224 done\n", + "Episode 225 done\n", + "Episode 226 done\n", + "Episode 227 done\n", + "Episode 228 done\n", + "Episode 229 done\n", + "Episode 230 done\n", + "Episode 231 done\n", + "Episode 232 done\n", + "Episode 233 done\n", + "Episode 234 done\n", + "Episode 235 done\n", + "Episode 236 done\n", + "Episode 237 done\n", + "Episode 238 done\n", + "Episode 239 done\n", + "Episode 240 done\n", + "Episode 241 done\n", + "Episode 242 done\n", + "Episode 243 done\n", + "Episode 244 done\n", + "Episode 245 done\n", + "Episode 246 done\n", + "Episode 247 done\n", + "Episode 248 done\n", + "Episode 249 done\n", + "Episode 249 done\n", + "Total steps: 18\n", + "Episode 250 done\n", + "Episode 251 done\n", + "Episode 252 done\n", + "Episode 253 done\n", + "Episode 254 done\n", + "Episode 255 done\n", + "Episode 256 done\n", + "Episode 257 done\n", + "Episode 258 done\n", + "Episode 259 done\n", + "Episode 260 done\n", + "Episode 261 done\n", + "Episode 262 done\n", + "Episode 263 done\n", + "Episode 264 done\n", + "Episode 265 done\n", + "Episode 266 done\n", + "Episode 267 done\n", + "Episode 268 done\n", + "Episode 269 done\n", + "Episode 270 done\n", + "Episode 271 done\n", + "Episode 272 done\n", + "Episode 273 done\n", + "Episode 274 done\n", + "Episode 275 done\n", + "Episode 276 done\n", + "Episode 277 done\n", + "Episode 278 done\n", + "Episode 279 done\n", + "Episode 280 done\n", + "Episode 281 done\n", + "Episode 282 done\n", + "Episode 283 done\n", + "Episode 284 done\n", + "Episode 285 done\n", + "Episode 286 done\n", + "Episode 287 done\n", + "Episode 288 done\n", + "Episode 289 done\n", + "Episode 290 done\n", + "Episode 291 done\n", + "Episode 292 done\n", + "Episode 293 done\n", + "Episode 294 done\n", + "Episode 295 done\n", + "Episode 296 done\n", + "Episode 297 done\n", + "Episode 298 done\n", + "Episode 299 done\n" ] }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAk0AAAHHCAYAAACiOWx7AAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAADYL0lEQVR4nOzdd1gUVxfA4d/Se1MUUMSODRV7BaxYY+8xajTG2LvxM7HEGntJMcZEjUGNxpLErij2rtgoNrCCqDTpsHu/PzZuXEGKAot63+fhkZm9M3NmWNjjnTvnKoQQAkmSJEmSJClTeroOQJIkSZIk6V0gkyZJkiRJkqRskEmTJEmSJElSNsikSZIkSZIkKRtk0iRJkiRJkpQNMmmSJEmSJEnKBpk0SZIkSZIkZYNMmiRJkiRJkrJBJk2SJEmSJEnZIJMmSZKkTKxduxaFQkFoaGiOt/Xz80OhUODn55frcX2I3uZnIUm5QSZNkvSO+OGHH1AoFNStW1fXoRQ4c+bMYceOHboOQ5Kk95xCzj0nSe+Ghg0b8ujRI0JDQ7l58yZly5bVdUgFhoWFBV27dmXt2rW5vm+lUklqairGxsYoFIocbatSqUhJScHIyAg9Pfl/1Le1du1aBgwYQEhICCVLltR1ONIHSP4WS9I7ICQkhJMnT7J48WLs7e3x8fHJ9xhUKhVJSUn5ftzcFh8fn6P2+vr6mJiY5DhhAtDT08PExKTAJEw5Pff89r68x6T3V8H4TZYkKVM+Pj7Y2trStm1bunbtqpU0paamYmdnx4ABA9JtFxsbi4mJCePHj9esS05OZtq0aZQtWxZjY2OcnZ2ZOHEiycnJWtsqFAqGDx+Oj48PlStXxtjYmL179wKwcOFCGjRoQKFChTA1NaVmzZr8+eef6Y6fmJjIyJEjKVy4MJaWlnz00Uc8fPgQhULB9OnTtdo+fPiQTz/9lKJFi2JsbEzlypX59ddfs7w2CoWC+Ph41q1bh0KhQKFQ0L9/fwCmT5+OQqEgICCA3r17Y2trS6NGjQC4cuUK/fv3p3Tp0piYmODg4MCnn37Ks2fPtPaf0TiakiVL0q5dO44fP06dOnUwMTGhdOnS/Pbbb1rbZjSmycvLiypVqhAQEECTJk0wMzOjWLFizJ8/P9253b17l48++ghzc3OKFCnCmDFj2LdvX7bGSWV27gC///47NWvWxNTUFDs7O3r27Mn9+/c1ry9fvhx9fX2io6M16xYtWoRCoWDs2LGadUqlEktLSyZNmqRZl933R2bvsevXr9O0aVNMTU0pXrw4s2bNQqVSZXrOkpTXDHQdgCRJWfPx8aFz584YGRnRq1cvfvzxR86dO0ft2rUxNDSkU6dObNu2jZ9++gkjIyPNdjt27CA5OZmePXsC6v/Jf/TRRxw/fpzBgwdTsWJFrl69ypIlS7hx40a6cUGHDh1i8+bNDB8+nMKFC2tuiSxbtoyPPvqIPn36kJKSwqZNm+jWrRs7d+6kbdu2mu379+/P5s2b6du3L/Xq1ePIkSNar7/w+PFj6tWrp/kQtbe3Z8+ePQwcOJDY2FhGjx792muzfv16Bg0aRJ06dRg8eDAAZcqU0WrTrVs3ypUrx5w5c3gxIuHAgQPcuXOHAQMG4ODgwPXr11m1ahXXr1/n9OnTWfYs3bp1i65duzJw4ED69evHr7/+Sv/+/alZsyaVK1fOdNuoqChatWpF586d6d69O3/++SeTJk3Czc2N1q1bA+peoaZNmxIWFsaoUaNwcHBgw4YNHD58ONN9vyqjc589ezZff/013bt3Z9CgQTx58oQVK1bg4eHBpUuXsLGxoXHjxqhUKo4fP067du0AOHbsGHp6ehw7dkyz/0uXLhEXF4eHh4dmXXbfH5Dxeyw8PJwmTZqQlpbGl19+ibm5OatWrcLU1DRH5y5JuU5IklSgnT9/XgDiwIEDQgghVCqVKF68uBg1apSmzb59+wQg/vnnH61t27RpI0qXLq1ZXr9+vdDT0xPHjh3Tardy5UoBiBMnTmjWAUJPT09cv349XUwJCQlayykpKaJKlSqiadOmmnUXLlwQgBg9erRW2/79+wtATJs2TbNu4MCBwtHRUTx9+lSrbc+ePYW1tXW6473K3Nxc9OvXL936adOmCUD06tUry3MQQoiNGzcKQBw9elSzbs2aNQIQISEhmnUuLi7p2kVERAhjY2Mxbtw4zbrDhw8LQBw+fFizztPTUwDit99+06xLTk4WDg4OokuXLpp1ixYtEoDYsWOHZl1iYqKoUKFCun1m5HXnHhoaKvT19cXs2bO11l+9elUYGBho1iuVSmFlZSUmTpwohFC/7woVKiS6desm9PX1xfPnz4UQQixevFjo6emJqKgozb6y8/4Q4vXvsdGjRwtAnDlzRrMuIiJCWFtbp/tZSFJ+krfnJKmA8/HxoWjRojRp0gRQ39Lo0aMHmzZtQqlUAtC0aVMKFy7MH3/8odkuKiqKAwcO0KNHD826LVu2ULFiRSpUqMDTp081X02bNgVI14vh6elJpUqV0sX08v/4o6KiiImJoXHjxly8eFGz/sVtlqFDh2ptO2LECK1lIQRbt26lffv2CCG04vL29iYmJkZrv29iyJAhmZ5DUlIST58+pV69egDZOl6lSpVo3LixZtne3h5XV1fu3LmT5bYWFhZ8/PHHmmUjIyPq1Kmjte3evXspVqwYH330kWadiYkJn332WZb7f9mr575t2zZUKhXdu3fXutYODg6UK1dO8x7Q09OjQYMGHD16FIDAwECePXvGl19+iRCCU6dOAerepypVqmBjY6M5RnbeHy9k9B7bvXs39erVo06dOpp19vb29OnTJ0fnLkm5Td6ek6QCTKlUsmnTJpo0aUJISIhmfd26dVm0aBG+vr60bNkSAwMDunTpwoYNG0hOTsbY2Jht27aRmpqqlTTdvHmTwMBA7O3tMzxeRESE1nKpUqUybLdz505mzZqFv7+/1liol29p3b17Fz09vXT7ePWpvydPnhAdHc2qVatYtWpVtuLKqYzOIzIykhkzZrBp06Z0+4+JiclynyVKlEi3ztbWlqioqCy3LV68eLrbf7a2tly5ckWzfPfuXcqUKZOuXU6fmnz13G/evIkQgnLlymXY3tDQUPN948aNmT59OomJiRw7dgxHR0dq1KhBtWrVOHbsGC1atOD48eN0795dax/ZeX+8Lj5Qn3tGpTVcXV0zP1lJymMyaZKkAuzQoUOEhYWxadMmNm3alO51Hx8fWrZsCUDPnj356aef2LNnDx07dmTz5s1UqFCBatWqadqrVCrc3NxYvHhxhsdzdnbWWs5oDMmxY8f46KOP8PDw4IcffsDR0RFDQ0PWrFnDhg0bcnyOLwb3fvzxx/Tr1y/DNlWrVs3xfl+W0Xl0796dkydPMmHCBKpXr46FhQUqlYpWrVpla8Cxvr5+hutFNqq4vM22OfXquatUKhQKBXv27MkwDgsLC833jRo1IjU1lVOnTnHs2DFNz1rjxo05duwYQUFBPHnyRKvHLafvDzlOSXqXyKRJkgowHx8fihQpwvfff5/utW3btrF9+3ZWrlyJqakpHh4eODo68scff9CoUSMOHTrElClTtLYpU6YMly9fplmzZm/0CD3A1q1bMTExYd++fRgbG2vWr1mzRqudi4sLKpWKkJAQrV6NW7duabWzt7fH0tISpVJJ8+bN3yimnJ5LVFQUvr6+zJgxg6lTp2rW37x5842OnxdcXFwICAhACKF1fq9ev5wqU6YMQghKlSpF+fLlM21bp04djIyMOHbsGMeOHWPChAkAeHh48PPPP+Pr66tZfiG774/MuLi4ZPizCA4OzvY+JCkvyDFNklRAJSYmsm3bNtq1a0fXrl3TfQ0fPpznz5/z999/A+oxKF27duWff/5h/fr1pKWlad2aA3XvysOHD/n5558zPF526vjo6+ujUCg046kAQkND0z155+3tDagrmb9sxYoV6fbXpUsXtm7dyrVr19Id78mTJ1nGZG5urvVofFZe9LC82rOzdOnSbO8jr3l7e/Pw4UPNzxfUY68y+tnlROfOndHX12fGjBnpzl8IoVVywcTEhNq1a7Nx40bu3bun1dOUmJjI8uXLKVOmDI6Ojpptsvv+yEybNm04ffo0Z8+e1ax78uSJTuqTSdLLZE+TJBVQf//9N8+fP9caCPyyevXqaQpdvkiOevTowYoVK5g2bRpubm5UrFhRa5u+ffuyefNmhgwZwuHDh2nYsCFKpZKgoCA2b97Mvn37qFWrVqZxtW3blsWLF9OqVSt69+5NREQE33//PWXLltUak1OzZk26dOnC0qVLefbsmabkwI0bNwDt3qF58+Zx+PBh6taty2effUalSpWIjIzk4sWLHDx4kMjIyExjqlmzJgcPHmTx4sU4OTlRqlSpTKebsbKywsPDg/nz55OamkqxYsXYv3+/1rgxXfv888/57rvv6NWrF6NGjcLR0REfHx9MTEyAnPeuvVCmTBlmzZrF5MmTCQ0NpWPHjlhaWhISEsL27dsZPHiwVl2vxo0bM2/ePKytrXFzcwOgSJEiuLq6EhwcrKmJ9UJ23x+ZmThxIuvXr6dVq1aMGjVKU3LAxcUl2/uQpDyho6f2JEnKQvv27YWJiYmIj49/bZv+/fsLQ0NDzaP6KpVKODs7C0DMmjUrw21SUlLEt99+KypXriyMjY2Fra2tqFmzppgxY4aIiYnRtAPEsGHDMtzHL7/8IsqVKyeMjY1FhQoVxJo1azSPuL8sPj5eDBs2TNjZ2QkLCwvRsWNHERwcLAAxb948rbaPHz8Ww4YNE87OzsLQ0FA4ODiIZs2aiVWrVmV5rYKCgoSHh4cwNTUVgKb8wIuYnjx5km6bBw8eiE6dOgkbGxthbW0tunXrJh49epSuHMLrSg60bds23T49PT2Fp6enZvl1JQcqV66cbtt+/foJFxcXrXV37twRbdu2FaampsLe3l6MGzdObN26VQDi9OnTmV6TzM5dCCG2bt0qGjVqJMzNzYW5ubmoUKGCGDZsmAgODtZqt2vXLgGI1q1ba60fNGiQAMQvv/ySbt/ZfX9k9h67cuWK8PT0FCYmJqJYsWJi5syZ4pdffpElBySdknPPSZKUr/z9/XF3d+f333+Xj5C/gaVLlzJmzBgePHhAsWLFdB2OJH1Q5JgmSZLyTGJiYrp1S5cuRU9PT2vwsJSxV69fUlISP/30E+XKlZMJkyTpgBzTJElSnpk/fz4XLlygSZMmGBgYsGfPHvbs2cPgwYPTlTeQ0uvcuTMlSpSgevXqxMTE8PvvvxMUFCQHREuSjsjbc5Ik5ZkDBw4wY8YMAgICiIuLo0SJEvTt25cpU6ZgYCD/z5aVpUuXsnr1akJDQ1EqlVSqVImJEyemeypSkqT8IZMmSZIkSZKkbJBjmiRJkiRJkrJBJk2SJEmSJEnZIAcV5BKVSsWjR4+wtLR846JzkiRJkiTlLyEEz58/x8nJCT29zPuSZNKUSx49eiSfBpIkSZKkd9T9+/cpXrx4pm1k0pRLLC0tAfVFt7Ky0nE0kiRJkiRlR2xsLM7OzprP8UzpsBq5mDNnjqhVq5awsLAQ9vb2okOHDiIoKEirTWJiohg6dKiws7MT5ubmonPnziI8PFyrzd27d0WbNm00Uw2MHz9epKamarU5fPiwcHd3F0ZGRqJMmTJizZo16eL57rvvhIuLizA2NhZ16tQRZ86cyfa5xMTECEBrGgpJkiRJkgq2nHx+63Qg+JEjRxg2bBinT5/mwIEDpKam0rJlS62Z1seMGcM///zDli1bOHLkCI8ePaJz586a15VKJW3btiUlJYWTJ0+ybt061q5dy9SpUzVtQkJCaNu2LU2aNMHf35/Ro0czaNAg9u3bp2nzxx9/MHbsWKZNm8bFixepVq0a3t7eRERE5M/FkCRJkiSpYMuHJC7bIiIiBCCOHDkihBAiOjpaGBoaii1btmjaBAYGCkCcOnVKCCHE7t27hZ6enlbv048//iisrKxEcnKyEEKIiRMnppsgs0ePHsLb21uzXKdOHa2JI5VKpXBychJz587NVuyyp0mSJEmS3j3vTE/Tq2JiYgCws7MD4MKFC6SmptK8eXNNmwoVKlCiRAlOnToFwKlTp3Bzc6No0aKaNt7e3sTGxnL9+nVNm5f38aLNi32kpKRw4cIFrTZ6eno0b95c00aSJEmSpA9bgRkIrlKpGD16NA0bNqRKlSoAhIeHY2RkhI2NjVbbokWLEh4ermnzcsL04vUXr2XWJjY2lsTERKKiolAqlRm2CQoKyjDe5ORkkpOTNcuxsbHZOk+lUklqamq22kqSlPsMDQ3R19fXdRiSJL2DCkzSNGzYMK5du8bx48d1HUq2zJ07lxkzZmS7vRCC8PBwoqOj8y4oSZKyxcbGBgcHB1lTTZKkHCkQSdPw4cPZuXMnR48e1aqR4ODgQEpKCtHR0Vq9TY8fP8bBwUHT5uzZs1r7e/z4sea1F/++WPdyGysrK0xNTdHX10dfXz/DNi/28arJkyczduxYzfKLRxZf50XCVKRIEczMzOQfa0nSASEECQkJmgc8HB0ddRyRJEnvEp0mTUIIRowYwfbt2/Hz86NUqVJar9esWRNDQ0N8fX3p0qULAMHBwdy7d4/69esDUL9+fWbPnk1ERARFihQB1DOrW1lZUalSJU2b3bt3a+37wIEDmn0YGRlRs2ZNfH196dixI6C+Xejr68vw4cMzjN3Y2BhjY+NsnadSqdQkTIUKFcrWNpIk5Q1TU1MAzd8MeatOkqTs0mnSNGzYMDZs2MBff/2FpaWlZgyStbU1pqamWFtbM3DgQMaOHYudnR1WVlaMGDGC+vXrU69ePQBatmxJpUqV6Nu3L/Pnzyc8PJyvvvqKYcOGaZKaIUOG8N133zFx4kQ+/fRTDh06xObNm9m1a5cmlrFjx9KvXz9q1apFnTp1WLp0KfHx8QwYMOCtz/PFGCYzM7O33pckSW/vxe9iamqqTJokScq+vH6ULzNAhl8vF558UdzS1tZWmJmZiU6dOomwsDCt/YSGhorWrVsLU1NTUbhwYTFu3LgMi1tWr15dGBkZidKlS2dY3HLFihWiRIkSwsjISNSpU0ecPn062+eS2SOLiYmJIiAgQCQmJmZ7f5Ik5R35OylJ0gs5KTmgEEII3aVs74/Y2Fisra2JiYlJN41KUlISISEhlCpVChMTEx1FKEnSC/J3UpKkFzL7/H5VgarTJEnvIz8/PxQKRY6enOzfv79mfJ30nze5lpIkSblFJk1Stpw6dQp9fX3atm2r61DyXGhoKAqFAn9//1zZX4MGDQgLC8Pa2jrb2yxbtoy1a9fmyvElSZKk3CGTJilbfvnlF0aMGMHRo0d59OhRnh5LCEFaWlqeHiM3pKSkZKudkZFRjmsCWVtbpyvqmh+ye055raDEIUlSweEb+BiVSrcjimTSJGUpLi6OP/74gy+++IK2bdtq9YD07t2bHj16aLVPTU2lcOHC/Pbbb4C6fMPcuXMpVaoUpqamVKtWjT///FPT/sUtlz179lCzZk2MjY05fvw4t2/fpkOHDhQtWhQLCwtq167NwYMHtY4VFhZG27ZtMTU1pVSpUmzYsIGSJUuydOlSTZvo6GgGDRqEvb09VlZWNG3alMuXL7/2fF+UvnB3d0ehUODl5QX8d8ts9uzZODk54erqCsD69eupVasWlpaWODg40Lt3b62Jnl+9pbR27VpsbGzYt28fFStWxMLCglatWhEWFqbZ5tXbc15eXowcOZKJEydiZ2eHg4MD06dP14o7KCiIRo0aYWJiQqVKlTh48CAKhYIdO3a89ly9vLwYPnw4o0ePpnDhwnh7ewNw7do1WrdujYWFBUWLFqVv3748ffoUgJ07d2JjY4NSqQTA398fhULBl19+qdnvoEGD+PjjjwF49uwZvXr1olixYpiZmeHm5sbGjRuzFcfu3bspX748pqamNGnShNDQ0NeeiyRJ76+NZ+8xcN15hvx+QaeJk0yadEQIQUJKWr5/vcm4/82bN1OhQgVcXV35+OOP+fXXXzX76dOnD//88w9xcXGa9vv27SMhIYFOnToB6urpv/32GytXruT69euMGTOGjz/+mCNHjmgd58svv2TevHkEBgZStWpV4uLiaNOmDb6+vly6dIlWrVrRvn177t27p9nmk08+4dGjR/j5+bF161ZWrVqllbAAdOvWjYiICPbs2cOFCxeoUaMGzZo1IzIyMsPzfVEs9eDBg4SFhbFt2zbNa76+vgQHB3PgwAF27twJqJPEmTNncvnyZXbs2EFoaCj9+/fP9JomJCSwcOFC1q9fz9GjR7l37x7jx4/PdJt169Zhbm7OmTNnmD9/Pt988w0HDhwA1LXAOnbsiJmZGWfOnGHVqlVMmTIl0/29vF8jIyNOnDjBypUriY6OpmnTpri7u3P+/Hn27t3L48eP6d69OwCNGzfm+fPnXLp0CYAjR45QuHBh/Pz8NPs8cuSIJtlMSkqiZs2a7Nq1i2vXrjF48GD69u2brijtq3Hcv3+fzp070759e/z9/Rk0aJBWYiZJ0ofh4r0opv2lnku2mrMNenq6Kw5dICqCf4gSU5VUmrov348b8I03ZkY5+7H/8ssvml6DVq1aERMTo/lQ9Pb2xtzcnO3bt9O3b18ANmzYwEcffYSlpSXJycnMmTOHgwcPaoqJli5dmuPHj/PTTz/h6empOc4333xDixYtNMt2dnZUq1ZNszxz5ky2b9/O33//zfDhwwkKCuLgwYOcO3eOWrVqAbB69WrKlSun2eb48eOcPXuWiIgITd2uhQsXsmPHDv78808GDx6c7nzt7e0BKFSoULqK8Obm5qxevRojIyPNuk8//VTzfenSpVm+fDm1a9cmLi4OCwuLDK9pamoqK1eupEyZMoC6Kv4333yT8Q/gX1WrVmXatGkAlCtXju+++w5fX19atGjBgQMHuH37Nn5+fpqYZ8+erXU9X6dcuXLMnz9fszxr1izc3d2ZM2eOZt2vv/6Ks7MzN27coHz58lSvXh0/Pz9q1aqFn58fY8aMYcaMGcTFxRETE8OtW7c0P9tixYppJYQjRoxg3759bN68mTp16rw2jv/973+UKVOGRYsWAeDq6srVq1f59ttvszwnSZLeDxHPk/ji9wukKFW0quzAUK8yOo1H9jRJmQoODubs2bP06tULAAMDA3r06MEvv/yiWe7evTs+Pj4AxMfH89dff9GnTx8Abt26RUJCAi1atMDCwkLz9dtvv3H79m2tY71IfF6Ii4tj/PjxVKxYERsbGywsLAgMDNT0NAUHB2NgYECNGjU025QtWxZbW1vN8uXLl4mLi6NQoUJaxw8JCUl3/Oxwc3PTSpgALly4QPv27SlRogSWlpaaZOHlHrFXmZmZaRImUE/n8WoP2auqVq2qtfzyNsHBwTg7O2sleS8nJJmpWbOm1vLly5c5fPiw1vWqUKECgOaaeXp64ufnhxCCY8eO0blzZypWrMjx48c5cuQITk5OmuRVqVQyc+ZM3NzcsLOzw8LCgn379qW7Pq/GERgYSN26dbXWvUi8JUl6/6WkqRj6+0UexyZTrogFC7tX0/kUZLKnSUdMDfUJ+MZbJ8fNiV9++YW0tDScnJw064QQGBsb891332FtbU2fPn3w9PQkIiKCAwcOYGpqSqtWrQA0t+127dpFsWLFtPb96jQ05ubmWsvjx4/nwIEDLFy4kLJly2JqakrXrl1zNEg4Li4OR0dHrVtHL7zJQOtXY4yPj8fb2xtvb298fHywt7fn3r17eHt7ZxqnoaGh1rJCocjy1mlG26hUqhyeQXqvnlNcXBzt27fPsEfnxVxtXl5e/Prrr1y+fBlDQ0MqVKiAl5cXfn5+REVFafUgLliwgGXLlrF06VLc3NwwNzdn9OjR6a7Pq3FIkvRh+2bndc7fjcLSxIBVn9TCwlj3KYvuI/hAKRSKHN8my29paWn89ttvLFq0iJYtW2q91rFjRzZu3MiQIUNo0KABzs7O/PHHH+zZs4du3bppPuArVaqEsbEx9+7d0/ogzY4TJ07Qv39/zdiouLg4rYHArq6upKWlcenSJU0vxa1bt4iKitK0qVGjBuHh4RgYGFCyZMlsHfdFT9KLgc6ZCQoK4tmzZ8ybN08zYfP58+ezdZzc5Orqyv3793n8+DFFixYF4Ny5c2+0rxo1arB161ZKliyJgUHG79EX45qWLFmi+bl6eXkxb948oqKiGDdunKbtiRMn6NChg+YWr0ql4saNG5q5IV+nYsWK/P3331rrTp8+/UbnJEnSu+WPc/f4/fQ9FApY1rM6pQoXjP9Uydtz0mvt3LmTqKgoBg4cSJUqVbS+unTporlFB+qn6FauXMmBAwc0t+YALC0tGT9+PGPGjGHdunXcvn2bixcvsmLFCtatW5fp8cuVK8e2bdvw9/fn8uXL9O7dW6tnpUKFCjRv3pzBgwdz9uxZLl26xODBgzE1NdV04TZv3pz69evTsWNH9u/fT2hoKCdPnmTKlCmvTW6KFCmCqampZgB0TEzMa2MsUaIERkZGrFixgjt37vD3338zc+bMbF3f3NSiRQvKlClDv379uHLlCidOnOCrr74CyHF39rBhw4iMjKRXr16cO3eO27dvs2/fPgYMGKBJJG1tbalatSo+Pj6aAd8eHh5cvHiRGzduaCXI5cqV48CBA5w8eZLAwEA+//xzHj9+nGUcQ4YM4ebNm0yYMIHg4GA2bNgga1dJ0gfg0r0ovt6hHvg9tnl5mlYoquOI/iOTJum1fvnlF5o3b55hUcYuXbpw/vx5rly5AqifogsICKBYsWI0bNhQq+3MmTP5+uuvmTt3LhUrVqRVq1bs2rVL82j/6yxevBhbW1saNGhA+/bt8fb21hq/BPDbb79RtGhRPDw86NSpE5999hmWlpaaqTEUCgW7d+/Gw8ODAQMGUL58eXr27Mndu3c1PTKvMjAwYPny5fz00084OTnRoUOH18Zob2/P2rVr2bJlC5UqVWLevHksXLgw0/PKC/r6+uzYsYO4uDhq167NoEGDNE/P5XSaECcnJ06cOIFSqaRly5a4ubkxevRobGxs0NP770+Gp6cnSqVSkzTZ2dlRqVIlHBwcNOUYAL766itq1KiBt7c3Xl5eODg4ZKvaeYkSJdi6dSs7duygWrVqrFy5UmtwuiRJ7x/1wO+LpChVeFcuyrAmZXUdkhY591wukXPPFQwPHjzA2dmZgwcP0qxZM12Ho1MnTpygUaNG3Lp1S2vQuSR/JyWpIEpJU9Fn9WnOhUZRtogFO4Y1zJdxTDmZe65gD6qRpCwcOnSIuLg43NzcCAsLY+LEiZQsWRIPDw9dh5bvtm/fjoWFBeXKlePWrVuMGjWKhg0byoRJkqR3wqxdAZwLjcLS2IBVfWsWiIHfryp4EUlSDqSmpvK///2PO3fuYGlpSYMGDfDx8Un3pNmH4Pnz50yaNIl79+5RuHBhmjdvrqlxJEmSVJBtPn+f307dRaGApT2rU9o+4xp3uiZvz+USeXtOkt4d8ndSkgoO//vRdF95ihSlirEtyjOyWbmsN8pFObk9JweCS5IkSZKkE0+eJzNkvbrid8tKRRlewAZ+v0omTZIkSZIk5btUpYphPhcJj02ijL05i7pX0+m8ctkhkyZJkiRJkvLdrJ0BnA2NxMJYXfHb0qTgj0WVSZMkSZIkSflqy/n7rDt1F4AlPapTpoAO/H6VTJokSZIkSco3l+9HM2XHNQBGNy9Hi0oFp+J3VmTSJEmSJElSvngal8yQ3y+QkqaiecWijGyav0/KvS2ZNElvzMvLi9GjR+s6jHda//79szWlyIdGXhdJev+8GPgdFpNEaXtzFvco+AO/XyWTJilT/fv3R6FQpPu6desW27Zte+vJaRUKBTt27MidYAuw0NBQFAoF/v7+WuuXLVsmJ6GVJOmDMHtXIGdC/h343bcWVu/AwO9XyYrgUpZatWrFmjVrtNbZ29ujr6+f6XYpKSkYGRnlZWg697bnmNFkyPkhNTW1QFRNLyhxSJKUt7ZeeMDak6EALO5ejbJF3o2B36+SPU1SloyNjXFwcND60tfXT3d7rmTJksycOZNPPvkEKysrBg8eTEpKCsOHD8fR0RETExNcXFyYO3eupj1Ap06dUCgUmuWMnDx5kurVq2NiYkKtWrXYsWNHup6ba9eu0bp1aywsLChatCh9+/bl6dOnmte9vLwYOXIkEydOxM7ODgcHB6ZPn651nOjoaAYNGoS9vT1WVlY0bdqUy5cva16fPn061atXZ/Xq1VrVpPfu3UujRo2wsbGhUKFCtGvXjtu3b2u2K1WqFADu7u4oFAq8vLyA9LehkpOTGTlyJEWKFMHExIRGjRpx7tw5zet+fn4oFAp8fX2pVasWZmZmNGjQgODg4Ndeuxe9XH/88Qeenp6YmJjg4+MDwOrVq6lYsSImJiZUqFCBH374QbNd165dGT58uGZ59OjRKBQKgoKCAHXCaG5uzsGDB7N1DV4Xh1KpZOzYsZrtJk6ciJyoQJLeH1cfxPC/7VcBGNmsHC0rO+g4ojcnkyZdEQJS4vP/K48/jBYuXEi1atW4dOkSX3/9NcuXL+fvv/9m8+bNBAcH4+Pjo0mOXiQDa9asISwsTCs5eFlsbCzt27fHzc2NixcvMnPmTCZNmqTVJjo6mqZNm+Lu7s758+fZu3cvjx8/pnv37lrt1q1bh7m5OWfOnGH+/Pl88803HDhwQPN6t27diIiIYM+ePVy4cIEaNWrQrFkzIiMjNW1u3brF1q1b2bZtmyZpi4+PZ+zYsZw/fx5fX1/09PTo1KkTKpUKgLNnzwJw8OBBwsLC2LZtW4bnOnHiRLZu3cq6deu4ePEiZcuWxdvbW+v4AFOmTGHRokWcP38eAwMDPv3008x+LAB8+eWXjBo1isDAQLy9vfHx8WHq1KnMnj2bwMBA5syZw9dff826desA8PT0xM/PT7P9kSNHKFy4sGbduXPnSE1NpUGDBtm6Bq+LY9GiRaxdu5Zff/2V48ePExkZyfbt27M8H0mSCr5nccl8vv48yWkqmlUowug3nCIlWZnMrNOzCIsLy+UIc0hIuSImJkYAIiYmJt1riYmJIiAgQCQmJv63MjlOiGlW+f+VHJej8+rXr5/Q19cX5ubmmq+uXbsKIYTw9PQUo0aN0rR1cXERHTt21Np+xIgRomnTpkKlUmW4f0Bs37490xh+/PFHUahQIa3r9/PPPwtAXLp0SQghxMyZM0XLli21trt//74ARHBwsCbeRo0aabWpXbu2mDRpkhBCiGPHjgkrKyuRlJSk1aZMmTLip59+EkIIMW3aNGFoaCgiIiIyjfnJkycCEFevXhVCCBESEqIV7wv9+vUTHTp0EEIIERcXJwwNDYWPj4/m9ZSUFOHk5CTmz58vhBDi8OHDAhAHDx7UtNm1a5cAtN9fL3lx7KVLl6Y7rw0bNmitmzlzpqhfv74QQogrV64IhUIhIiIiRGRkpDAyMhIzZ84UPXr0EEIIMWvWLNGgQYMcX4NX43B0dNScnxBCpKamiuLFi2uuS17I8HdSkqRclZKmFD1+OilcJu0UTRYcFjGJKW+0n6jEKNF3d19RZW0V0e3vbkKpUuZqnJl9fr9KjmmSstSkSRN+/PFHzbK5uflr29aqVUtruX///rRo0QJXV1datWpFu3btaNmyZY6OHxwcTNWqVbUmVq1Tp45Wm8uXL3P48GEsLNLfJ799+zbly5cHoGrVqlqvOTo6EhERodlHXFwchQoV0mqTmJiodZvJxcUFe3t7rTY3b95k6tSpnDlzhqdPn2p6V+7du0eVKlWydZ63b98mNTWVhg0batYZGhpSp04dAgMDtdq+fB6Ojo4AREREUKJEidfu/+WfTXx8PLdv32bgwIF89tlnmvVpaWmacVZVqlTBzs6OI0eOYGRkhLu7O+3ateP7778H1D1PL24z5uQavBxHTEwMYWFh1K1bV7POwMCAWrVqyVt0kvSOm7s7iNN3IjE30mfVJzXfaOD3g+cP+OLgF4TGhmJpaMn4WuPRU+juJplMmnTF0Az+90g3x80hc3NzypbN3iSKryZUNWrUICQkhD179nDw4EG6d+9O8+bN+fPPP3McR2bi4uJo37493377bbrXXiQVQLpBxwqFQvPhHhcXh6Ojo9YtqRdsbGw032eUNLZv3x4XFxd+/vlnnJycUKlUVKlShZSUlDc8o8y9fB4KhfqR3Vdvg73q5bjj4uIA+Pnnn7USFkAzwF+hUODh4YGfnx/GxsZ4eXlRtWpVkpOTuXbtGidPnmT8+PGa7bJ7DTJLuiVJej9su/iAX0+EALC4R3XKFrHM8T6uP73OUN+hRCZF4mDuwI/NfqSsrW4n9JVJk64oFGD0YXx4WFlZ0aNHD3r06EHXrl1p1aoVkZGR2NnZYWhoiFKpzHR7V1dXfv/9d5KTkzE2NgZIN/6pRo0abN26lZIlS2Jg8GZv6xo1ahAeHo6BgUGmg9Jf9ezZM4KDg/n5559p3LgxAMePH9dq8+IJu8zOtUyZMhgZGXHixAlcXFwA9dNl586dy/V6WEWLFsXJyYk7d+7Qp0+f17bz9PTk559/xtjYmNmzZ6Onp4eHhwcLFiwgOTlZ0yuWnWuQEWtraxwdHTlz5gweHh6AurfrxXgySZLePdcexjB5278Dv5uWxfsNBn4fuX+ECUcnkJiWiKutKz80/4EiZkVyO9QckwPBpTy1ePFiNm7cSFBQEDdu3GDLli04ODhoem5KliyJr68v4eHhREVFZbiP3r17o1KpGDx4MIGBgezbt4+FCxcC//WyDBs2jMjISHr16sW5c+e4ffs2+/btY8CAAVkmZS80b96c+vXr07FjR/bv309oaCgnT55kypQpnD9//rXb2draUqhQIVatWsWtW7c4dOgQY8eO1WpTpEgRTE1NNQPUY2Ji0u3H3NycL774ggkTJrB3714CAgL47LPPSEhIYODAgdk6h5yYMWMGc+fOZfny5dy4cYOrV6+yZs0aFi9erGnj5eVFQEAA169fp1GjRpp1Pj4+1KpVS9NrlJ1r8DqjRo1i3rx57Nixg6CgIIYOHUp0dHSun68kSXlPPfD7AslpKpq42jO6efkc72Nz8GZGHh5JYloiDZwasLbV2gKRMIFMmqQ8Zmlpyfz586lVqxa1a9cmNDSU3bt3o6enfustWrSIAwcO4OzsjLu7e4b7sLKy4p9//sHf35/q1aszZcoUpk6dCqAZ5+Tk5MSJEydQKpW0bNkSNzc3Ro8ejY2NjeZYWVEoFOzevRsPDw8GDBhA+fLl6dmzJ3fv3qVo0dfPjaSnp8emTZu4cOECVapUYcyYMSxYsECrjYGBAcuXL+enn37CycmJDh06ZLivefPm0aVLF/r27UuNGjW4desW+/btw9bWNlvnkBODBg1i9erVrFmzBjc3Nzw9PVm7dq2mPAKAm5sbNjY2VK9eXTNezMvLC6VSqTWeKTvX4HXGjRtH37596devH/Xr18fS0pJOnTrl6rlKkpT30pQqhm+4xMPoREoVNmdpT/ccVfwWQrDs4jJmnp6JSqjoWLYj3zX7DgujglPTSSHkaMtcERsbi7W1NTExMVhZWWm9lpSUREhIiFZdH+nt+Pj4MGDAAGJiYjA1NdV1ONI7Rv5OSlLum7kzgF+Oh2BupM+OYQ0pVzT745hSlal8ffJrdt3ZBcDQakMZUm2I5m5CXsrs8/tVckyT9E747bffKF26NMWKFePy5ctMmjSJ7t27y4RJkiSpANhx6SG/HFcP/F7UvVqOEqbYlFjGHB7D2fCzGCgMmFp/Kp3KFczeZpk0Se+E8PBwpk6dSnh4OI6OjnTr1o3Zs2frOixJkqQP3rWHMUzaegWA4U3K0qqKYxZb/Cc8PpwvDn7BrehbmBmYscRrCQ2KNcirUN+aTJqkd8LEiROZOHGirsOQJEmSXhIZn6IZ+O3las+YFtkf+B0cGczQg0OJSIzA3tSeH5r/QAW7CnkY7dvT6UDwo0eP0r59e5ycnDKc7V6hUGT49fIA05IlS6Z7fd68eVr7uXLlCo0bN8bExARnZ2fmz5+fLpYtW7ZQoUIFTExMcHNzY/fu3XlyzpIkSZL0PlAP/L7Iw+hEShYyY1lPd/SzOfD75KOT9Nvbj4jECMralMWnjU+BT5hAx0lTfHw81apV01QYflVYWJjW16+//opCoaBLly5a7b755hutdiNGjNC8FhsbS8uWLXFxceHChQssWLCA6dOns2rVKk2bkydP0qtXLwYOHMilS5fo2LEjHTt25Nq1a3lz4pIkSZL0jvt2bxAnbz/DzEifn/rWwto0exW/d9zawbCDw4hPjae2Q23WtV6Ho0X2b+npkk5vz7Vu3ZrWrVu/9nUHB+2CWH/99RdNmjShdOnSWustLS3TtX3Bx8eHlJQUfv31V4yMjKhcuTL+/v4sXryYwYMHA7Bs2TJatWrFhAkTAJg5cyYHDhzgu+++Y+XKlW9zipIkSZL03vnL/yE/H/t34He3arg6ZD3wWwjByisr+cH/BwDalGrDzIYzMdI3ytNYc9M7U6fp8ePH7Nq1K8Mif/PmzaNQoUK4u7uzYMEC0tLSNK+dOnUKDw8PTUVmAG9vb4KDgzXFFE+dOkXz5s219unt7c2pU6fy6GwkSZIk6d10/dF/A7+HepWhtVvWvUSpqlSmn5quSZgGVhnI3MZz36mECd6hgeDr1q3D0tKSzp07a60fOXIkNWrUwM7OjpMnTzJ58mTCwsI0VY3Dw8O1ivUBmkKF4eHh2NraEh4enq54YdGiRQkPD39tPMnJySQnJ2uWY2Nj3+r8JEmSJKmgi/p34HdSqgrP8vaMa+ma5TbxqfGM8xvHiUcn0FPo8b86/6NHhR75EG3ue2eSpl9//ZU+ffqkK0T38lQNVatWxcjIiM8//5y5c+dq5inLC3PnzmXGjBl5tn9JkiRJKkjSlCpGbLzEg6hEXAqZsTwbA7+fJDxhmO8wAiMDMTUwZb7HfLycvfIn4DzwTtyeO3bsGMHBwQwaNCjLtnXr1iUtLY3Q0FBAPS7q8ePHWm1eLL8YB/W6Nq8bJwUwefJkYmJiNF/379/PySm9t7y8vLKcXLZkyZIsXbo00zYvP00ZGhqKQqHA398/V2J832Tnen6I5HWRpNy1YF8wx289xcxIn1V9a2FtlvnA79vRt+mzuw+BkYHYmdjxq/ev73TCBO9I0vTLL79Qs2ZNqlWrlmVbf39/9PT0KFJEPblf/fr1OXr0KKmpqZo2Bw4cwNXVVTOfV/369fH19dXaz4EDB6hfv/5rj2NsbIyVlZXW1/uof//+KBQKhgwZku61YcOGoVAo6N+/v2bdtm3bmDlzZj5G+OFYu3atZqLjl507d07zUIMkSVJe+PvyI346egeABV2zHvh9LvwcfXf3JSw+jJJWJfm9ze9UKVwlP0LNUzpNmuLi4vD399f0IISEhODv78+9e/c0bWJjY9myZUuGvUynTp1i6dKlXL58mTt37uDj48OYMWP4+OOPNQlR7969MTIyYuDAgVy/fp0//viDZcuWad3WGzVqFHv37mXRokUEBQUxffp0zp8/z/Dhw/P2ArwjnJ2d2bRpE4mJiZp1SUlJbNiwgRIlSmi1tbOzw9Iy++XzJUhJSXmr7e3t7TEzM8ulaLJHqVSiUqny9ZgFOQ5Jep8FPIpl4p+XARjiWYa2VTMf+L0nZA+fH/ic56nPqW5fnfWt1+Ns6ZwfoeY5nSZN58+fx93dXTO7/dixY3F3d9fMYA+wadMmhBD06tUr3fbGxsZs2rQJT09PKleuzOzZsxkzZoxWDSZra2v2799PSEgINWvWZNy4cUydOlXrf+YNGjRgw4YNrFq1imrVqvHnn3+yY8cOqlR597Pi3FCjRg2cnZ3Ztm2bZt22bdsoUaKE5mf3wqu35yIiImjfvj2mpqaUKlUKHx+fdPu/efMmHh4emJiYUKlSJQ4cOJBlTNeuXaN169ZYWFhQtGhR+vbty9OnTzPd5ueff8bZ2RkzMzM6derE4sWL0/Xc/PXXX9SoUQMTExNKly7NjBkztJ7GVCgUrF69mk6dOmFmZka5cuX4+++/cxSbl5cXw4cPZ/To0RQuXBhvb28AFi9ejJubG+bm5jg7OzN06FDi4uIA8PPz00xQ/KKI6/Tp04H0t6Hu3btHhw4dsLCwwMrKiu7du2vdfp4+fTrVq1dn/fr1lCxZEmtra3r27Mnz589fe+1e9HL9/fffVKpUCWNjY+7du0dycjLjx4+nWLFimJubU7duXfz8/AD148X29vb8+eefmv1Ur14dR8f//uAeP34cY2NjEhISsrwGmcWRnfeZJEk5FxWfwue/nycpVUXjcoWZ4P36gd9CCH699isTj04kVZVKC5cW/NzyZ2xMbHInmIC/IC0563Z5SUi5IiYmRgAiJiYm3WuJiYkiICBAJCYmatapVCoRnxKf718qlSpH59WvXz/RoUMHsXjxYtGsWTPN+mbNmoklS5aIDh06iH79+mnWe3p6ilGjRmmWW7duLapVqyZOnTolzp8/Lxo0aCBMTU3FkiVLhBBCKJVKUaVKFdGsWTPh7+8vjhw5Itzd3QUgtm/fLoQQIiQkRADi0qVLQgghoqKihL29vZg8ebIIDAwUFy9eFC1atBBNmjR57XkcP35c6OnpiQULFojg4GDx/fffCzs7O2Ftba1pc/ToUWFlZSXWrl0rbt++Lfbv3y9Kliwppk+frmkDiOLFi4sNGzaImzdvipEjRwoLCwvx7NmzbMfm6ekpLCwsxIQJE0RQUJAICgoSQgixZMkScejQIRESEiJ8fX2Fq6ur+OKLL4QQQiQnJ4ulS5cKKysrERYWJsLCwsTz58+FEEK4uLhoXc/q1auLRo0aifPnz4vTp0+LmjVrCk9PT83xp02bJiwsLETnzp3F1atXxdGjR4WDg4P43//+99rrt2bNGmFoaCgaNGggTpw4IYKCgkR8fLwYNGiQaNCggTh69Ki4deuWWLBggTA2NhY3btwQQgjRuXNnMWzYMCGEEJGRkcLIyEhYW1uLwMBAIYQQs2bNEg0bNtQcJ7NrkFkcWb3PXpXR76QkSdrSlCrx8erTwmXSTtH420MiKj45k7ZpYuapmaLK2iqiytoqYt6ZeSJNmZY7gahUQhycIcQ0KyE29xdCqcyd/f4rs8/vV8mkKZfkNGmKT4nXvLny8ys+JT5H5/UiaYqIiBDGxsYiNDRUhIaGChMTE/HkyZNMk6bg4GABiLNnz2peDwwMFIDmw2zfvn3CwMBAPHz4UNNmz549mSZNM2fOFC1bttSK8/79+wIQwcHBGZ5Hjx49RNu2bbXW9enTRytpatasmZgzZ45Wm/Xr1wtHR0fNMiC++uorzXJcXJwAxJ49e7Idm6enp3B3d88wzpdt2bJFFCpUSLO8Zs0arXhfeDlp2r9/v9DX1xf37t3TvH79+nWtn8O0adOEmZmZiI2N1bSZMGGCqFu37mtjWbNmjQCEv7+/Zt3du3eFvr6+1s9OCPV1nDx5shBCiOXLl4vKlSsLIYTYsWOHqFu3rujQoYP48ccfhRBCNG/ePNNkLaNr8Goc2XmfvUomTZKUtTm7A4TLpJ2iwld7RMCj1ycUCakJYrjvcFFlbRXhttZN/Hb9t9wLIi1ViL+GqxOmaVZCHFmgTqJyUU6Spnem5ICkW/b29rRt25a1a9cihKBt27YULlw4020CAwMxMDCgZs2amnUVKlTQuiUWGBiIs7MzTk5OmnWZDcAHuHz5MocPH8bCwiLda7dv36Z8+fQTRgYHB9OpUyetdXXq1GHnzp1a+z1x4gSzZ8/WrFMqlSQlJZGQkKAZN1S1alXN6+bm5lhZWREREZGj2F6+Ji8cPHiQuXPnEhQURGxsLGlpaemOnZUX19PZ+b/xA5UqVcLGxobAwEBq164NqG/pvTz2zNHRUXMOr2NkZKR17levXkWpVKa73snJyRQqVAgAT09PRo0axZMnTzhy5AheXl44ODjg5+fHwIEDOXnypNZEzNm5Bq/GkZ33mSRJObPzyiN+OvLvwO9uVanomPHDTs8SnzHi0AiuPr2KkZ4R8zzm0cKlRe4EkZoIWwdB0E5Q6EG7pVCzX+7s+w3JpElHTA1MOdP7jE6O+6Y+/fRTzeD4180XmB/i4uJo37493377bbrXXh4v8yb7nTFjRroCqoBWfTBDQ+3HbBUKhWYwcnZjMzc313otNDSUdu3a8cUXXzB79mzs7Ow4fvw4AwcOJCUlJdcHemd2Dq9jamqKQvFfTZa4uDj09fW5cOEC+vr6Wm1fJI1ubm7Y2dlx5MgRjhw5wuzZs3FwcODbb7/l3LlzpKam0qBBAyD71+DVOCRJyl2BYbFM2KKu+P25Z2naVXXKsF1oTChfHPyCB3EPsDa2ZkXTFbgXcc+wbY4lRsPGXnDvJOgbQ9dfoWK73Nn3W5BJk44oFArMDPP3iae31apVK1JSUlAoFJrBy5mpUKECaWlpXLhwQdPDERwcTHR0tKZNxYoVuX//PmFhYZqk4vTp05nut0aNGmzdupWSJUtiYJC9t7Crqyvnzp3TWvfqco0aNQgODqZs2bLZ2mduxQZw4cIFVCoVixYtQk9P/XzG5s2btdoYGRmhVCoz3c+L63n//n1Nb1NAQADR0dFUqlQph2eTOXd3d5RKJRERETRu3DjDNgqFgsaNG/PXX39x/fp1GjVqhJmZGcnJyfz000/UqlVLk0Bm5xpkJDvvM0mSsic6QV3xOzFVSeNyhZnoXSHDdv4R/ow4NILo5GiKWRTjx+Y/Usq6VIZtc+x5OPzeBR5fA2Mr6LURSjbKnX2/pXeiTpNUMOjr6xMYGEhAQEC6noWMuLq60qpVKz7//HPOnDnDhQsXGDRoEKam//V2NW/enPLly9OvXz8uX77MsWPHmDJlSqb7HTZsGJGRkfTq1Ytz585x+/Zt9u3bx4ABA16bVIwYMYLdu3ezePFibt68yU8//cSePXu0eiymTp3Kb7/9xowZM7h+/TqBgYFs2rSJr776KptX6M1iAyhbtiypqamsWLGCO3fusH79+nSTRZcsWZK4uDh8fX15+vSp5omzlzVv3hw3Nzf69OnDxYsXOXv2LJ988gmenp7UqlUr2+eRHeXLl6dPnz588sknbNu2jZCQEM6ePcvcuXPZtWuXpp2XlxcbN26kevXqWFhYoKenh4eHBz4+Pnh6euboGmQkO+8zSZKyplQJRmy8xL3IBJztTFnRK+OK3753fRm0fxDRydFULlSZ39v8nnsJ07Pb8EsLdcJkURQG7C4wCRPIpEnKoZwW8lyzZg1OTk54enrSuXNnBg8erCk8CqCnp8f27dtJTEykTp06DBo0SGtMUUacnJw4ceIESqWSli1b4ubmxujRo7GxsdH0ULyqYcOGrFy5ksWLF1OtWjX27t3LmDFjtG67eXt7s3PnTvbv30/t2rWpV68eS5YswcXFJdvn+yaxAVSrVo3Fixfz7bffUqVKFXx8fJg7d65WmwYNGjBkyBB69OiBvb098+fPT7cfhULBX3/9ha2tLR4eHjRv3pzSpUvzxx9/ZPsccmLNmjV88sknjBs3DldXVzp27Mi5c+e06nd5enqiVCrx8vLSrPPy8kq3LjvXILM4MnufSZKUtYX7gzl28ykmhnr89HEtbMzST6brE+jDGL8xJCuT8Szuya/ev1LYNPPxrdn26BL80hKi74Fdafh0Hzi45c6+c4lCCCF0HcT7IDY2Fmtra2JiYtIlFUlJSYSEhFCqVKl0c+dJuvPZZ58RFBTEsWPHdB2KlM/k76Qkadt1JYxhGy4CsLyXOx9V0x7HpBIqFp5fyPqA9QD0cO3Bl3W+xEAvl0b53PGDTX0gJQ4cqsLHW8Eif/7jk9nn96vkmCbpg7Fw4UJatGiBubk5e/bsYd26dfzwww+6DkuSJEmngsJjGb9FXfF7sEfpdAlTsjKZ/x37H/vv7gdgVI1RDKwyMPceyLi+HbYNBmUKlPKAHj5gUjCnJpNJk/TBOHv2LPPnz+f58+eULl2a5cuXZ2sSaEmSpPdVTEKqZuB3w7KFmPhKxe+Y5BhGHhrJxYiLGOgZMKvhLNqWbpt7AZz9GXZPAARU6gidV4GBce7tP5fJpEn6YGTnSSxJkqQPhVIlGLnpEnefJVDc1pTvetXAQP+/sZcPnj/gi4NfEBobiqWhJUubLKWOY53cObgQ4DcXjvxbnqX2IGg9H/SyfshIl2TSJEmSJEkfoMUHgjly44l64Hffmtia/zfw+/rT6wz1HUpkUiQO5g782OxHytq+eTkWLSol7BoHF9aol70mg+ckeAfqr8mkSZIkSZI+MHuuhvH94dsAfNulKpWdrDWvHX1wlPFHxpOYloirrSs/NP+BIma5NCg7NQm2fQaBfwMKaLsIag/MnX3nA5k0SZIkSdIH5Mbj54z7d+D3oEal6FC9mOa1LTe2MOv0LFRCRQOnBizyXISFUfppod5IUixs6g2hx0DfCDr/DJU75s6+84lMmiRJkiTpAxGTkMrg386TkKKkQZlCfNlaXfFbCMHyS8tZfXU1AB3LdmRq/akY6hlmtrvsi4tQV/kOvwJGltDTB0p7Zr1dASOTJkmSJEn6AChVglF/XCL0WQLFbEz5rrd64HeqMpWvT37NrjvqSv5Dqw1lSLUhuVdSIDIE1neCqBAwt4c+f4JT9dzZdz6TSZMkSZIkfQCWHLiBX/ATjA3UA7/tzI2ITYllzOExnA0/i4HCgKn1p9KpXKfcO2jYFXUPU3wE2LhA3+1QqEzu7T+fyWlUJEmHSpYsydKlS3UdRoEjr4sk5a6918L47vAtQD3wu0oxa8Ljw+m3px9nw89iZmDG982+z92EKeQYrG2rTpiKusHA/e90wgQyaZKy0L9/fzp27Jgn+/bz80OhUHwQs9GvXbsWGxubdOvPnTvH4MGD8z8gSZI+GDcfP2fcZvXA74GNStHRvRjBkcH02dWHW9G3sDe1Z13rdTQo1iD3Dhrwt7qHKTkWXBrBgF1g6ZB7+9cReXtOkt5CSkoKRkbpJ7XMLnt7+1yMJnuUSiUKhSLTCYQ/pDgk6X0Wk5jK4PUXiE9RUr90ISa3rsDJRycZ6zeW+NR4ytqU5YdmP+Bo4Zh7Bz2/BnaNBaGCCu2gyy9g+H7M8Sj/WklvZfHixbi5uWFubo6zszNDhw4lLi5O8/rdu3dp3749tra2mJubU7lyZXbv3k1oaChNmjQBwNbWFoVCQf/+/V97nJ9//hlnZ2fMzMzo1KkTixcvTtdz89dff1GjRg1MTEwoXbo0M2bMIC0tTfO6QqFg9erVdOrUCTMzM8qVK8fff/+ttY9r167RunVrLCwsKFq0KH379uXp06ea1728vBg+fDijR4+mcOHCeHt7Z3kd/Pz8GDBgADExMSgUChQKBdOnTwfS34a6d+8eHTp0wMLCAisrK7p3787jx481r0+fPp3q1auzfv16SpYsibW1NT179uT58+evvXYvern+/vtvKlWqhLGxMffu3SM5OZnx48dTrFgxzM3NqVu3Ln5+foD6SRp7e3v+/PNPzX6qV6+Oo+N/f1iPHz+OsbExCQkJWV6DzOKIiIigffv2mJqaUqpUKXx8fF57LlLBFxoTysagjVx/dh2lSqnrcD5oAY9i6fj9CUKexv878NudnSF/M+zgMOJT46ntUJt1rdflXsIkBByZDztHqxOmGv2g+2/vTcIEMmnSGSEEqoSEfP8SQuTqeejp6bF8+XKuX7/OunXrOHToEBMnTtS8PmzYMJKTkzl69ChXr17l22+/xcLCAmdnZ7Zu3QpAcHAwYWFhLFu2LMNjnDhxgiFDhjBq1Cj8/f1p0aIFs2fP1mpz7NgxPvnkE0aNGkVAQAA//fQTa9euTdduxowZdO/enStXrtCmTRv69OlDZGQkANHR0TRt2hR3d3fOnz/P3r17efz4Md27d9fax7p16zAyMuLEiROsXLkyy+vQoEEDli5dipWVFWFhYYSFhTF+/Ph056lSqejQoQORkZEcOXKEAwcOcOfOHXr06KHV7vbt2+zYsYOdO3eyc+dOjhw5wrx58zL9OSUkJPDtt9+yevVqrl+/TpEiRRg+fDinTp1i06ZNXLlyhW7dutGqVStu3ryJQqHAw8NDk0RFRUURGBhIYmIiQUFBABw5coTatWtjZmaW5TXILI7+/ftz//59Dh8+zJ9//skPP/xAREREpucjFUwnHp6g566ezDkzh547e+LxhwejD49mQ+AG7kTfyfW/P1LGhBBsPHuPTj+oEyZHaxNW9a3J5tu/8vWJr0kTabQp1YaVzVdiZZRLE+OqVOo55A7/+zfXYwK0X1bgp0XJMSHlipiYGAGImJiYdK8lJiaKgIAAkZiYqFmnjI8XAa4V8v1LGR+fo/Pq16+f6NChQ7bbb9myRRQqVEiz7ObmJqZPn55h28OHDwtAREVFZbrPHj16iLZt22qt69Onj7C2ttYsN2vWTMyZM0erzfr164Wjo6NmGRBfffWVZjkuLk4AYs+ePUIIIWbOnClatmyptY/79+8LQAQHBwshhPD09BTu7u6ZxitE+uuwZs0arXhfcHFxEUuWLBFCCLF//36hr68v7t27p3n9+vXrAhBnz54VQggxbdo0YWZmJmJjYzVtJkyYIOrWrfvaWNasWSMA4e/vr1l39+5doa+vLx4+fKjVtlmzZmLy5MlCCCGWL18uKleuLIQQYseOHaJu3bqiQ4cO4scffxRCCNG8eXPxv//9L0fX4NU4goODtc5PCCECAwMFoLkueSGj30np7fwR9Ieotq6aqLK2imiztY2o51NPVFlbRevL6w8vMfHIRLHtxjbx4PkDXYf8XopLShWjN10SLpN2CpdJO0W/X8+Ix7FxYuqJqZqfw5LzS4RSpcy9g6YmCbG5vxDTrISYZi3E6ZW5t+98kNnn96vkmCbprRw8eJC5c+cSFBREbGwsaWlpJCUlkZCQgJmZGSNHjuSLL75g//79NG/enC5dulC1atUcHSM4OJhOnbSf6KhTpw47d+7ULF++fJkTJ05o9SwplUqtWACtY5ubm2NlZaXp1bh8+TKHDx/GwiJ99dvbt29Tvnx5AGrWrJnj65AdgYGBODs74+zsrFlXqVIlbGxsCAwMpHbt2oD6lp6lpaWmjaOjY5Y9M0ZGRlrnfvXqVZRKpeacXkhOTqZQoUIAeHp6MmrUKJ48ecKRI0fw8vLCwcEBPz8/Bg4cyMmTJ7V6krJzDV6NIzAwEAMDA61rWqFChQwHzUsFk1KlZMmFJawLWAfAR2U+Ynr96SgUCgKeBXA2/Cynw07jH+HP08Sn7A7Zze6Q3QAUsyhGPcd61HGoQx3HOhQ2LazLU3nnBYc/Z6jPBW4/iUdfT8Go5i54VE5j6plxnHh4Aj2FHlPqTqG7a/esd5Zdyc/hj4/hjh/oGUKnleDWNff2X8DIpElHFKamuF68oJPj5pbQ0FDatWvHF198wezZs7Gzs+P48eMMHDiQlJQUzMzMGDRoEN7e3uzatYv9+/czd+5cFi1axIgRI3ItDoC4uDhmzJhB586d071mYvLf/XRDQ+3qtgqFApVKpdlH+/bt+fbbb9Pt4+WxPObm5lqvZec65KbMzuF1TE1NtQrVxcXFoa+vz4ULF9DX1+4+f5E0urm5YWdnx5EjRzhy5AizZ8/GwcGBb7/9lnPnzpGamkqDBuqnbbJ7DV6NQ3q3JaQmMPnYZA7dPwTA8OrDGVx1sOZnXNW+KlXtqzLIbRDJymSuPLnC6bDTnA07y7Wn13gY95CtN7ey9ab6Vn1Zm7KaBKq2Q+3cu3X0AfA5e5NZBw6SZnAfmxKPcSzylF8f3OXn++pxZaYGpsz3mI+Xs1fuHTT+Kfh0hUeXwNAcev4OZZrm3v4LIJk06YhCoUCRyx+m+e3ChQuoVCoWLVqkeQJq8+bN6do5OzszZMgQhgwZwuTJk/n5558ZMWKE5qkzpTLzwaKurq6cO3dOa92ryzVq1CA4OJiyZd98Fu4aNWqwdetWSpYsiYFB9n81snMdjIyMsjzPihUrcv/+fe7fv6/pbQoICCA6OppKlSrl8Gwy5+7ujlKpJCIigsaNG2fYRqFQ0LhxY/766y+uX79Oo0aNMDMzIzk5mZ9++olatWppEsjsvhdeVaFCBdLS0rhw4YKmJy04OPiDKEPxrnuS8IThh4YT8CwAIz0jZjacSZvSbV7b3ljfmNoOtantUBvcIT41nouPL3Im7Axnw88SFBnErehb3Iq+xYagDegp9KhoV5E6jnWo51CP6kWqY2b4bv/NzC3xqfEERQYR8CyAq08COHr3EnGqRxgUFxgASuBBvLqtnYkdboXd+KLaF1QuXDn3goi6q67yHXkbzApBny1QLH0v/PtGJk1SlmJiYvD399daV6hQIcqWLUtqaiorVqygffv2WgOjXxg9ejStW7emfPnyREVFcfjwYSpWrAiAi4sLCoWCnTt30qZNG0xNTTO8NTZixAg8PDxYvHgx7du359ChQ+zZs0erx2Lq1Km0a9eOEiVK0LVrV/T09Lh8+TLXrl1j1qxZ2TrPYcOG8fPPP9OrVy8mTpyInZ0dt27dYtOmTaxevTpdj8wL2bkOJUuWJC4uDl9fX6pVq4aZmVm6HqjmzZvj5uZGnz59WLp0KWlpaQwdOhRPT09q1aqVrXPIrvLly9OnTx8++eQTFi1ahLu7O0+ePMHX15eqVavStm1bQP204Lhx46hVq5bmZ+Ph4YGPjw8TJkzI0TXIiKurK61ateLzzz/nxx9/xMDAgNGjR2Oaiz2iUu4Ljgxm+KHhhMeHY2tsy7Kmy3Av4p6jfZgbmtO4eGMaF1cn7dFJ0Zx7fE6TRIXEhHD92XWuP7vOmmtrMNAzoGrhqurbeY51qFq4Kob6uTQvWgH2POW5JkF68XU39i4C7UH1CgWY6dlSy9GNyoUrU9GuIpUKVaKIWZHc790Nv6auwRQXDtYloO82KFwud49RUOXDGKsPQk4Hgr8r+vXrJ4B0XwMHDhRCCLF48WLh6OgoTE1Nhbe3t/jtt9+0BncPHz5clClTRhgbGwt7e3vRt29f8fTpU83+v/nmG+Hg4CAUCoXo16/fa+NYtWqVKFasmDA1NRUdO3YUs2bNEg4ODlpt9u7dKxo0aCBMTU2FlZWVqFOnjli1apXmdUBs375daxtra2uxZs0azfKNGzdEp06dhI2NjTA1NRUVKlQQo0ePFiqVSgihHgg+atSodPFldR2EEGLIkCGiUKFCAhDTpk0TQmgPBBdCPUD7o48+Eubm5sLS0lJ069ZNhIeHa16fNm2aqFatmtaxlyxZIlxcXF577V43CD0lJUVMnTpVlCxZUhgaGgpHR0fRqVMnceXKFU2bS5cuCUBMmjRJ63iA2Lt3b46uweviCAsLE23bthXGxsaiRIkS4rfffkt3XXLbu/w7qWvHHhwTdX3qiiprq4h229qJezH3st7oDYTHhYu/b/0tvjr+lWixpUW6QeW1f68tPt//ufjl6i/i2pNrIk2Zlidx5KfopGhx6tEp8cvVX8R4v/GizdY26c77xVcDHy9R6bseotyC0cJ94VKxOyA4f4IMPSHEHGf1oO/v6wkR8yh/jpuHcjIQXCGEfAY0N8TGxmJtbU1MTAxWVtr34ZOSkggJCaFUqVJa42ukN/fZZ58RFBTEsWPHdB2K9A6Sv5Nv5o+gP5hzdg4qoaK2Q22WeC3B2tg6z48rhODB8wecCT+j6YmKTIrUamNpZEntorXVt/Mc61HaunSBHj8XlRRF4LNAAiL/60F6GPcww7ZO5k5UKlSJioUqUta6An+fVbDjQiwADcsWYmkPd+wtjfM+6KDd8OcASEuCEvWh10Ywtc374+axzD6/XyVvz0nvhIULF9KiRQvMzc3Zs2cP69at44cfftB1WJL0QVCqlCy6sIj1AesB6FCmA9PqT8u322MKhQJnK2ecrZzpWr4rQghuRd/iTNgZzoSf4Xz4eZ6nPOfQ/UOaQemFTAppEqg6DnUoblk8X2LNyNPEp+oE6VkAgZHqf8PiwzJsW9yiuCZBqlSoEpXsKmFjYgPA7SdxDPO5SFB4LAoFjGpWjhFNy6Gvlw/J4cX18M9IddHK8q2h2xowzN/b6MroaIRKhYGdXb4e92WypymXyJ6mvNW9e3f8/Px4/vw5pUuXZsSIEQwZMkTXYUnvKPk7mX0JqQlMOjYJv/t+AIx0H8kgt0EFqhcnTZVG4LNAzoSf4WzYWS5FXCJJmaTVpphFMeo61lU/nedQB3uzvJnCKCIhQpMgBTwLICAygIiEjEuCuFi5UMnuvwSpgl2F1/bc/eX/kP9tu0p8ipLCFkYs7eFOo3L5UKJBCDi+BHxnqJerf6wuWqmfP30uqoQEnh86TOyuXcQdP06h/v0pMm5srh4jJz1NMmnKJTJpkqR3h/ydzJ6IhAiG+w4nMDIQIz0jZjeaTatSrXQdVpZSlClcfnKZs+FnORN2hqtPrpIm0rTalLEuQx3HOtR1qEsth1o5vs0ohOBxwmOtAdqBkYE8TXyarq0CBSWtS2p6jioWqkhFu4pYGKV/8OVVSalKZu4MwOfMPQDqlbZjeU93iljlw/tWpYL9U+D0v736DUdD8+nqUed5SKSkEHfiBLE7d/H80CFEYqLmNYumTXH+4ftcPZ5MmnRAJk2S9O6Qv5NZC44MZpjvMB4nPMbW2JblTZdTvUh1XYf1RhJSE7gYoS5vcCbsDEGRQVpPnylQULFQReo61KWuY13ci7hrlTcQQvAo/pFWD1JgZGC6cVUAego9SluXVt9is/uvB+lNyiWEPo1nqM9FAsLUt+OGNynLqGblMNDPhxnQ0lLgr2Fw9d/SId5zoP6wPDucUKlIOH9enSjt24cyJkbzmmGJEli1bYN127YYv0VZmdeRY5oKKJmfSlLBIH8XM3f0wVEmHJlAQloCpaxL8X2z73G2dM56wwLKzNCMRsUa0ahYIwBikmM4F/5feYM7MXc0ydCa6/+VN3C1cyUkJoTAyEBikmPS7VdfoU8ZmzJaCZKrnSumBm8/1mfXlTAmbb1CXHIaduZGLOlRHc/yeXNLMZ2UePijL9z2BT0D6PADVOuR9XY5JIQgKSCA2J27iN29m7SXJifXty+MdZs2WLVti4mbW4G5HSyTpnzwooJzQkKCrD8jSQVAQkICkL66ugQbgzYy7+w8VEJFXYe6LPJalC9PyOUna2Nrmrs0p7lLc0BdqPPFeKgzYWd4FP+IixEXuRhxUbONgZ4B5WzKqW+x/Zsklbcrj7F+7j61lpymZM6uQNadugtA7ZK2rOhVAwfrfOoRjX8GG7rBwwtgaAbdf4NyLXL1EMkhIcTu2k3szp2khIZq1utZWmLp3RLrtm0xq1MHxWtq4+mSTJrygb6+PjY2Npr5wczMzApM1ixJHxIhBAkJCURERGBjY/PagqUfIqVKycLzC/k98HcAOpXtxNf1vv4gCkjam9nTrnQ72pVuB8D95/c5G6bugXoxFqmcTTmM9I3yNI57zxIYvvEiVx6oe7W+8CrDuBbl8+d2HED0fXWV72c31aUEem8B59q5suvUx4+J3b2H2J07Sbp+XbNeYWyMRdMmWLdti7mHB3pGeXuN35ZMmvKJg4MDQJYTq0qSlPdsbGw0v5PSv0/IHZ2E3wM/AEbVGMXAKgM/2P/cOVs65/vtyL3Xwpnw52WeJ6VhY2bIku7VaVKhSP4FEBEI6zvD80dgVVxd5dve9a12qYyOJnb/fmJ37iLh3Dn1k3gA+vqYN2yAdbt2WDRthr6FeeY7KkB0mjQdPXqUBQsWcOHCBcLCwti+fTsdO3bUvN6/f3/WrVuntY23tzd79+7VLEdGRjJixAj++ecf9PT06NKlC8uWLdOajuPKlSsMGzaMc+fOYW9vz4gRI7RmZwfYsmULX3/9NaGhoZQrV45vv/2WNm1eP49STikUChwdHSlSpAipqam5tl9JknLG0NBQ9jC95HH8Y0YcGvHfE3KNZ9OqZMF/Qu59kZKmYu6eQNacCAWgRgkbvutdAyebfBzKce8MbOgOSdFQ2FWdMFm/WV0rVUICzw8fJnanukQAL33emdasiXW7tlh6e+u01tLb0GnSFB8fT7Vq1fj0008znJ0eoFWrVqxZs0azbGysff+4T58+hIWFceDAAVJTUxkwYACDBw9mw4YNgHpUfMuWLWnevDkrV67k6tWrfPrpp9jY2DB48GAATp48Sa9evZg7dy7t2rVjw4YNdOzYkYsXL1KlSpVcPWd9fX35B1uSpAIhKDKIYb7DiEiIwM7EjuVNl1PNvpquw/pg3I9MYPjGS1y+Hw3AYI/STPB2xTC/bscB3NgHm/tBWiIUrw29N4NZzhIaTYmAXbvVJQL+HTMIYFyhAtbt2mLVpg2GTk65HX2+KzAlBxQKRYY9TdHR0ezYsSPDbQIDA6lUqRLnzp3TTGi6d+9e2rRpw4MHD3BycuLHH39kypQphIeHY/TvvdIvv/ySHTt2EBQUBECPHj2Ij49n586dmn3Xq1eP6tWrZ2vSUcjZI4uSJEm6dvTBUcYfGU9iWiKlrUvzfbPvdVo1+0NzIOAx4zb7E5uUhrWpIYu6VaN5paL5G4T/RnVZAaGEsi2g+zowyt6tMk2JgF27eb53r3aJAGdnrNq1zbMSAbntvSo54OfnR5EiRbC1taVp06bMmjWLQoUKAXDq1ClsbGy0ZoBv3rw5enp6nDlzhk6dOnHq1Ck8PDw0CROob/F9++23REVFYWtry6lTpxg7VrvCqLe392uTNUmSpHeZT6AP88/NVz8h51iXxV6LsTKS/9nLD6lKFQv2BbPq6B0Aqjnb8F0vd5ztcl7H6a2cWA4HvlZ/X7UndPgOshj0rykRsGu3ukRAeLjmNf3ChbFq0xrrdu0KVImA3Fagk6ZWrVrRuXNnSpUqxe3bt/nf//5H69atOXXqFPr6+oSHh1OkiPZAOQMDA+zs7Aj/94cZHh5OqVKltNoULVpU85qtrS3h4eGadS+3CX/pDfGq5ORkkpOTNcuxsbFvda6SJEl5TalSMv/cfDYEqYcvdC7Xma/qfYWh3vv/hFxB8Cg6keEbLnLxXjQAnzYsxZetK2BkkI+344SAA1Ph5HL1cv3h0GIm6L0+hpTQUGJ27SJ25y5SQkI06/UsLbFs2QLrdu0KbImA3Fagk6aePXtqvndzc6Nq1aqUKVMGPz8/mjVrpsPIYO7cucyYMUOnMUiSJGVXQmoCE49O5MiDIwCMrjGaT6t8+t72CBQ0h4MiGLvZn6iEVCxNDFjQtRqtquTzE5zKVPh7JFxWJ820+AYajsqwqaZEwK5dJF27plmvMDbGokkTrNu9GyUCcluBTppeVbp0aQoXLsytW7do1qwZDg4O6R7hT0tLIzIyUvM4sYODA49fqjIKaJazapPZI8mTJ0/WuqUXGxuLs/O7WzFXkqT3V3h8OCMOjSAoMghjfWPmNJpDy5ItdR3WByFNqWLh/husPHIbALdi1nzfuwYlCuXz7biUBNjSH27uA4U+fLQC3PtoNdGUCNi1m4SzZ9OXCGjbFotmzd+pEgG57Z1Kmh48eMCzZ89wdHQEoH79+kRHR3PhwgVq1qwJwKFDh1CpVNStW1fTZsqUKaSmpmqq/x44cABXV1dsbW01bXx9fRk9erTmWAcOHKB+/fqvjcXY2Djdk3ySJEkFTeCzQIb7DiciUf2E3IqmK6hqX1XXYX0QwmOSGLHxIudCowDoV9+F/7WtiLFBPt/GSoiEDT3gwVkwMIVua8FVXVYiqxIBVm3bYNWq1TtbIiC36TRpiouL49atW5rlkJAQ/P39sbOzw87OjhkzZtClSxccHBy4ffs2EydOpGzZsnh7ewNQsWJFWrVqxWeffcbKlStJTU1l+PDh9OzZE6d/H23s3bs3M2bMYODAgUyaNIlr166xbNkylixZojnuqFGj8PT0ZNGiRbRt25ZNmzZx/vx5Vq1alb8XRJIkKRf53fdj4tGJJKYlUsa6DN83/55iFsV0HdYH4ciNJ4z5w5/I+BQsjA34tktV2lZ1zP9Aou+DT1d4EgQm1tB7C8KxBnF+furJcTMoEWDVtg3WbdpgWEy+V16l05IDfn5+NGnSJN36fv368eOPP9KxY0cuXbpEdHQ0Tk5OtGzZkpkzZ2oN2o6MjGT48OFaxS2XL1/+2uKWhQsXZsSIEUyaNEnrmFu2bOGrr77SFLecP39+jopbypIDkiQVJC8/IVfPsR6LvBbJJ+TygVIlWHrwBt8dvoUQUMnRih/61KBkYR3c0rp5ELYNgsQohIUjiVVnEnPiOs/37UMZHa1p9q6VCMhtOfn8LjB1mt51MmmSJKkgSFOlMf/cfDYGbQSgS7kuTKk3RT4hlw8iYpMYuekSp+9EAtCnbgm+blcJE8N8vh2nUoLfXDi6kLQkBc/ulyb2vhlpEU81TTQlAtq2xaRq1Q/6gYD3qk6TJEmSlD3xqfFMODKBYw+PATC25lj6V+7/QX8g5pcTt54yatMlnsalYG6kz5zObnSoroPbW3ERsHUghBwlMdKQB2ecSYtJABL+KxHQti1mdet+ECUCcptMmiRJkt4D4fHhDPcdTnBUMMb6xsxtPJcWLi10HdZ7T6kSrDh0k2W+NxECKjhY8n2fGpSxt8h649wWegL+/BTiwol5YEPYGStEahJGJUtiP24sFh4e6MkHmN6KTJokSZLecQHPAhjuO5wniU8oZFKIFU1X4Gbvpuuw3ntPnicz+o9LnLj1DICetZ2Z/lFlHdyOU6mLVfp+g1AqibhZkshLKUAa5p4eFFuwAH05bCRXyKRJkiTpHXb43mEmHZtEYloiZW3K8n2z73GyePcnRi3oTt1+xshNl3jyPBlTQ33mdK5CJ3cdzN2XGAXbv4Abe1CmKHh4tTLxN9UlDgp9/jn2I0fI23C5SCZNkiRJ7yAhBL8H/s6CcwsQCBo4NWCh50IsjSx1Hdp7TaUSfH/4FksO3kAloHxRC37oU4OyRXRw3R9ehC39IPoeyXFm3D9XktTHUShMTXGaMxur1q3zP6b3nEyaJEmS3jFpqjTmnZ3HH8F/ANCtfDcm150sn5DLY8/ikhn9hz/HbqqfQutaszjfdKiMmVE+f5QKAed/gb2TQZnC8yhnHh01RJUYjaGTE8W//w6TihXzN6YPhEyaJEmS3iFxKXFMODqB4w+Po0DBuFrj+KTSJ/IJuTx2NiSSERsv8jg2GRNDPWZ2qEK3WjqYOis5Dv4ZBdf+RAh4+rgmT/3CACVmdepQbOkSWb07D8mkSZIk6R0RHh/OUN+h3Iy6iYm+CfM85tGshG4nL3/fqVSCn47eYeH+YJQqQRl7c37oUxNXBx3cjosIgs2fwNNglGkGhIXU4/mFOwDYfvwxRSdNRGEoexvzkkyaJEmS3gHXn11nhO8IniQ+obBpYb5r+h2VC1fWdVjvtaj4FMZu9udw8BMAOrkXY1bHKpgb6+Cj8/IfsHM0pCaQIhx5cK44yXfvoDA0xGH6NGy6dMn/mD5AMmmSJEkq4Hzv+TL52GQS0xIpZ1uO75t+j6OFDuYx+4CcD41k5MZLPIpJwthAjxkfVaZHbef8vw2amgR7J8GFtQDEK2rzYHcsqtiH6NsXpvjy5Zi5u+dvTB8wmTRJkiQVUEIIfgv4jUXnFyEQNCzWkIUeC7Ew0kHhxA9EmlLF94dvs8xX/XRc6cLmfN+nBhUddVDnKPIObO4H4VcQQkGksj0R2y6CSoVJ1aoUX7Ecw5fmYpXynkyaJEmSCqA0VRpzzsxhy40tAPRw7cGXdb7EQE/+2c4rD6ISGPOHP+dC1XWOOrsX45uOVbDQxe24wJ2wYygkx6AyKkT4/cbEHD4LgHWnTjhMnyare+uA/O2TJEkqYOJS4hh/ZDwnHp1AgYLxtcbTt1Jf+YRcHtp55RGTt13leVIalsYGzOpURTdzxylT4eB0OPUdAKk2NXlwxJKkwLOgr0/RSZOw7fuxfC/oiEyaJEmSCpBHcY8Y5juMW9G3MDUwZV7jeTQt0VTXYb234pPTmP73dbZceACAewkblvd0x9nOLP+DiX0EWwbA/dMAJDj04sH6ayif3UDf2ppiy5ZiXq9e/sclacikSZIkqYC49vQaw32H8yzpGfam9qxotoLKheQTcnnlyoNoRm3yJ+RpPHoKGN6kLCOblcNAXy//g7l9GLYOgoSnYGxFlNknhH//D6SmYuzqSvHvv8OouA6maZG0yKRJkiSpAPC778eEIxNIUiZR3rY83zf7HgdzB12H9V5SqQSrjt1h4b5g0lQCJ2sTlvSoTt3ShXQQjBKOLgC/eYBA2Ffh8cP6RK3bBoCltzdOc2ajZ26e/7FJ6cikSZIkSccexT1i0tFJJCmTaFSsEQs9F2JuKD8k80J4TBLjtvhz4tYzANq4OTC3U1WszXRQFDL+KWz7DG4fAiCtfC8e7npOwoVdoFBgP2okhT7/XI5fKkBk0iRJkqRDQgi+OfUNCWkJuBdxZ0XTFfIJuTyy/3o4k7ZeISohFVNDfWZ8VJlutYrrJim5dwa29Ifnj8DAlMTKE3mwfCdpYWHomZvjtGABlk2b5H9cUqbkb6YkSZIO/X37b048OoGRnhEzGsyQCVMeSExRMnt3AL+fvgdAlWJWLOvpThl7HdS7EgJOfQ8Hp4EqDQqVI8buM8JmrEQkJWHk4kLxH77HuEyZ/I9NypL87ZQkSdKRp4lPmX9uPgBfVP+CUtaldBzR+ycwLJaRGy9xMyIOgM89SjOupStGBjoY7J0Uo669FLQTAFGpM0/ulOPZ90sBMPdoTLGFC9G30kEhTSlbZNIkSZKkI3POzCE2JZaKdhXpX7m/rsN5rwghWHsylLl7gkhJU2Fvaczi7tVoXM5eNwGFXVFPthsVAnqGKBtP4+GGK8QfWw9Aoc8+w370KBT6+rqJT8oWmTRJkiTpwIG7Bzhw9wAGCgNmNpwpb8vloqdxyUzYclkz0W6zCkWY37UqhSx0UEFbCLi4DnZPBGUyWJcgue5sHsz4gZS7d1GYmOA4exbWbdvmf2xSjsnfUkmSpHwWkxzD7NOzARhQZQCudq46juj9ceTGE8ZtvszTuGSMDPT4qm1F+tZz0c1g75R42DUOLm9UL5dvxXPb3jwa8Q2q+HgMnBxx/u47TCpVyv/YpDcikyZJkqR8Nv/cfJ4lPaO0dWmGVBui63DeC8lpShbsDWb18RAAXItasryXO64OlroJ6MkN9e24J4Gg0EM0+ZqnV415+s2XAJjVqkWx5cswsLPTTXzSG5FJkyRJUj46/vA4f9/+GwUKZjSYgZG+ka5Deufdiohj5MZLBITFAtCvvguT21TExFBH44Ou/gn/jIKUOLAoiqrtDzz68R+e798PgG3v3hSd/CUKQx3UhpLeSraSps6dO2d7h9u2bXvjYCRJkt5n8anxfHPqGwD6VOxD9SLVdRvQO04IwaZz95nxz3WSUlXYmRuxoGtVmlUsqpuA0pJh3xQ497N6uWRjUurO5MGk6STfuAGGhjhM/Rrbbt10E5/01rKVNFlbW2u+F0Kwfft2rK2tqVWrFgAXLlwgOjo6R8mVJEnSh2bJhSWExYdR3KI4I9xH6Dqcd1p0Qgpfbr3K3uvhADQuV5hF3apRxMpENwFF3VUXq3x0Ub3ceBzxJk152P8LlDEx6NsXpviy5ZjVcNdNfFKuyFbStGbNGs33kyZNonv37qxcuRL9fx+NVCqVDB06FCtZW0KSJClD58PP80fwHwBMbzAdM0MzHUf07jp1+xlj/vAnPDYJQ30FE70rMLBRKfT0dDTdSPBe2P45JEWDqS2i409EnQ7n8fwhoFRi4uZG8e9WYFhURz1gUq5RCCFETjawt7fn+PHjuLpqP+0RHBxMgwYNePbsWa4G+K6IjY3F2tqamJgYmTxKkqQlKS2Jrv905W7sXbqU68L0BtN1HdI7KVWpYunBG/zgdxshoHRhc5b3cqdKMeusN84LyjQ4PAuOL1EvF6uJ6qOfCV+ympgdOwCw7tgRhxnT0TPWQbkDKVty8vmd44HgaWlpBAUFpUuagoKCUKlUOd2dJEnSe+8H/x+4G3uXIqZFGFdrnK7DeSfdfRbPyE3+XL4fDUCPWs5MbV8Jc2MdPc/0PBz+HAh3j6uX63xOarURPBgxnqQrV0Bfn6ITJ2D7ySdywt33SI7fbQMGDGDgwIHcvn2bOnXqAHDmzBnmzZvHgAEDcj1ASZKkd9n1p9dZF7AOgK/rf42lkY4egX+Hbbv4gK93XCM+RYmViQFzO1elbVVH3QUUclSdMMVHgJEldFhBQnJJHvTsjfLJU/StrSm2dAnm9evrLkYpT+Q4aVq4cCEODg4sWrSIsLAwABwdHZkwYQLjxsn/QUmSJL2Qqkzl65NfoxIqWpdqjZezl65DeqfEJqUydcc1dvg/AqBOSTuW9KxOMRtT3QSkUsHxxXB4NggVFKkM3X8j6vAlwr/pB6mpGJcrR/EfvsfI2Vk3MUp5KkdJU1paGhs2bKBfv35MnDiR2Fh1TQw5hkeSJCm91ddWczPqJrbGtnxZ50tdh/NOuXA3itF/XOJ+ZCL6egpGNyvH0CZl0dfVYO+ESPVg75vqWktU/xjRYg6PFy0nasMGACxbtsRp7hz0zM11E6OU53KUNBkYGDBkyBACAwMBmSxJkiS9zs2om6y6sgqAL+t8iZ2JrPycHUqV4IfDt1jqexOlSlDc1pRlPd2p6WKru6AenFeXE4i5DwYm0GYhaSXb8nDICBLOnQOg8MgRFB4yBIWenu7ilPJcjm/P1alTh0uXLuHi4pIX8UiSJL3zlCol005OI02VhldxL1qXap35BkLAji/UvRjVekHdIWDz4d3eeRidyJhN/pwNjQSgQ3UnZnasgpWJjipnCwFnV6kLVqpSwa40dP+NpCgD7nftStqjMPTMzXFaMB/Lpk11E6OUr3KcNA0dOpRx48bx4MEDatasifkr3ZBVq1bNteAkSZLeRb8H/s7Vp1exMLTgq3pfZf301PXt/03qeuo7OP0jVO4E9YdBsRp5H3ABsPtqGF9uvUJsUhrmRvrM6lSFTu7FdRdQQiTsGqv+2QBU6gAffUfsoeM8+t8URFISRi4uFP/+O4zLltVdnFK+ynE/Ys+ePQkJCWHkyJE0bNiQ6tWr4+7urvk3J44ePUr79u1xcnJCoVCw49+6FgCpqalMmjQJNzc3zM3NcXJy4pNPPuHRo0da+yhZsiQKhULra968eVptrly5QuPGjTExMcHZ2Zn58+eni2XLli1UqFABExMT3Nzc2L17d47ORZIkCeBe7D2+u/QdAONrjaeoeRYFDROjYM8k9fdVe0IpDxBKuPYn/NwE1rSBoN3qQcjvoYSUNCb9eYWhPheJTUqjmrMNu0c11l3C9PAi7BgGiyuqEyY9A2j1LaLzr0R8v5qHY8chkpIwb9yYkls2y4TpA5PjnqaQkJBcO3h8fDzVqlXj008/TTcFS0JCAhcvXuTrr7+mWrVqREVFMWrUKD766CPOnz+v1fabb77hs88+0yxbWv73SG9sbCwtW7akefPmrFy5kqtXr/Lpp59iY2PD4MGDATh58iS9evVi7ty5tGvXjg0bNtCxY0cuXrxIlSpVcu18JUl6v6mEiumnppOkTKKuQ106l8vG1FIHp6sfXS9cHj5aDgbGEHYZTn0P17bC3RPqr0Jlod5Q9e07o/ejmvi1hzGM3HiJO0/jUShgqFcZRjcvj6F+Po8LSk2Ea9vg3Or/pkEBKOoG7ZagtHbl4dChxB89BkChQQOxHzMGhb6OJgSWdCbHFcHzikKhYPv27XTs2PG1bc6dO0edOnW4e/cuJUqUANQ9TaNHj2b06NEZbvPjjz8yZcoUwsPDMTJSzyb+5ZdfsmPHDoKCggDo0aMH8fHx7Ny5U7NdvXr1qF69OitXrsxW/LIiuCRJm4M3M/P0TEwNTNn60VacLbMYl3T3JKz5d7zTgD3g0kD79dhHcOYnuLAGkmLU60ztoPZAqP0ZWL6b03KoVILVx++wYF8wqUqBg5UJS3pUp36ZQvkbyLPbcP5XuPS7egoUAH0jqNQRag8C5zok37nDg2HDSQkNRWFsjOPs2Vi3a5u/cUp5Kk8rgr8QEBDAvXv3SElJ0Vr/0UcfvekusxQTE4NCocDGxkZr/bx585g5cyYlSpSgd+/ejBkzBgMD9amdOnUKDw8PTcIE4O3tzbfffktUVBS2tracOnWKsWPHau3T29tb63ahJElSZsLjw1l8YTEAI9xHZJ0wpSXDP6PU39folz5hArByghYzwGOC+oP99A8QfReOLoATy6Bqd6g3DIpWyuWzyTsRsUmM23KZYzefAtCqsgPzurhhY2aUxZa5RKWEG/vUvUq3ff9bb10Cag0A975gYQ/A80OHeTRhAqr4eAwcHSn+3QpMK1fOnzilAinHSdOdO3fo1KkTV69eRaFQ8KKj6sVAR6VSmbsR/ispKYlJkybRq1cvrUxw5MiR1KhRAzs7O06ePMnkyZMJCwtj8WL1H6/w8HBKlSqlta+i/06aGB4ejq2tLeHh4Zp1L7cJDw9/bTzJyckkJydrll/UrJIk6cMjhOCbU98QnxpPNftq9K7QO+uNji+FpzfAvIg6McqMsQXUGwJ1PoOgnXDyO3hwVp1IXfodyjSDBsOhdBMowFN2+AY+ZsKfV4iMT8HEUI9p7SvTs7Zz/kwzEhcBF3+DC2vVpQMAUEDZ5upepXItQE99u00olTxduZKn330PQmBaqybFly3DoFA+94RJBU6Ok6ZRo0ZRqlQpfH19KVWqFGfPnuXZs2eMGzeOhQsX5kWMpKam0r17d4QQ/Pjjj1qvvdxDVLVqVYyMjPj888+ZO3cuxnk4QeLcuXOZMSOLP3SSJH0QdoXs4tjDYxjqGfJNg2/Q18tirMuTG3Ds37+Xrb8F02zWINLTVz/FVakD3D8LJ1eok6jbvuqvIpXVT9y5dVWPjSogklKVzN0dyLpTdwGo5GjF8l7ulC1ikbcHFgLunVb3KgX8pS4bAOrr7d5X3bNkV1prk7Rnz3g0YSLxJ08CYNOrJw6TJ6MwyqeeMKlAy3HSdOrUKQ4dOkThwoXR09NDT0+PRo0aMXfuXEaOHMmlS5dyNcAXCdPdu3c5dOhQlvcb69atS1paGqGhobi6uuLg4MDjx4+12rxYdnBw0PybUZsXr2dk8uTJWglbbGwszrJsviR9cJ4lPuPbs98CMKTaEErblM58A5VKfVtOmQLlWqpLC7wJ5zrQYz1EhsCZlXBxPURch7+Ggu8MqDMYan0KZrotqhkc/pyRGy8R/Pg5AIMalWJCK1eMDfJwEHXyc7iyGc79or4mLxSrpe5VqtwRDNNPxZJw/jwPx44jLSIChakpDtOmYpPJOFvpw5PjpEmpVGqeTitcuDCPHj3C1dUVFxcXgoODczW4FwnTzZs3OXz4MIWy0TXq7++Pnp4eRYoUAaB+/fpMmTKF1NRUDA3VBdIOHDiAq6srtra2mja+vr5ag8kPHDhA/UwmWzQ2Ns7TnixJkt4Nc8/OJTo5GldbVwZUycak5ZfWw72TYGgGbRe9/e00u1Lq3iqvL9W3ns78BM/D4NBMOLYIqvdWP3VXqMzbHSeHhBCsP32XWbsCSUlTUdjCmEXdq+FZ3j7vDhoRqE6ULm+CFHWShoGpuuet9kBwyrgsjlCpePbLLzxZugyUSozKlKH40iUYlyuXd7FK76QcJ01VqlTh8uXLlCpVirp16zJ//nyMjIxYtWoVpUtn8T+sV8TFxXHr1i3NckhICP7+/tjZ2eHo6EjXrl25ePEiO3fuRKlUasYY2dnZYWRkxKlTpzhz5gxNmjTB0tKSU6dOMWbMGD7++GNNQtS7d29mzJjBwIEDmTRpEteuXWPZsmUsWbJEc9xRo0bh6enJokWLaNu2LZs2beL8+fOsWrUqp5dHkqQPiO89X/aF7kNfoc83Db/BUC+LytVxEXDga/X3Tb8CmxK5F4ypLTQaox4Yfn07nFoB4VfVt6bO/QIV2qpv3ZWon+fjnp7FJTPxzyv4BkUA0MTVngXdqlHYIg/+o6lMVd+iPPcLhB77b71dGXWvUvVemd7+TIuK4tGXXxJ/5CgAVh+1x3HaNDl/nJQxkUN79+4VW7duFUIIcfPmTeHq6ioUCoUoXLiw8PX1zdG+Dh8+LIB0X/369RMhISEZvgaIw4cPCyGEuHDhgqhbt66wtrYWJiYmomLFimLOnDkiKSlJ6ziXL18WjRo1EsbGxqJYsWJi3rx56WLZvHmzKF++vDAyMhKVK1cWu3btytG5xMTECEDExMTkaDtJkt5N0UnRwusPL1FlbRWx5PyS7G20ZYAQ06yEWNlYiLTUPI1PqFRC3PYT4veu6mO++PrJS4irf+bJ8ZVKldh15ZGoNeuAcJm0U5SbslusOX5HqFSqXD+WiH4gxKHZQiwo99+5TbcRYmNvIW4dEkKpzHIXCZcuiRteTUSAawUR6FZVRG7enDexSgVaTj6/c6VOU2RkJLa2tvnzBEQBJes0SdKH5esTX7Pj1g5KWpXkz4/+xFg/i16UG/thQzdQ6MFnh8Gper7ECcCTYHWxzMubQPnvU7/WJdRP5Ln3BZO3+5ulUgn2B4Sz9OBNgsLVt8XKFbFgeS93Kjrm4t9DIeCOH5z/RV0lXfz7tLZ5EajZH2r2A+usK4kLIYj67TceL1gIaWkYupSg+NKlmFSsmHuxSu+MnHx+5zhpOnToEA0aNMDExOStgnzfyKRJkj4cJx+d5PMDn6NAwbrW63AvksUUUinx8H09iLkH9YeD9+z8CfRVcU/UCcfZnyFBXScJYyuo8ckbTRKcUbJkYWzAgIYlGepVFlOjXBrsnRgN/hvUsT/7b0gHLo3UY5UqtAOD7D3dpoyNJWzKFJ4fOAiAZatWOM6aib5FHj/JJxVYeZo0WVhYkJaWRu3atfHy8sLT05OGDRtiapr+SYQPiUyaJOnDkJCaQKe/OvEo/hG9KvTif3X/l/VG+6aoJ+K1LgHDToORjsfLpCbClT/UvU9Pb6jXKfSzPUmwSiXYdz2cZb7aydKnDUvyaaNSuVeo8pG/OlG6sgXSEtXrjCyhWk91slQkZz1Dideu83D0aFIfPABDQ4p+OQnb3r0/6LskUh4nTampqZw9e5YjR45w5MgRTp48SUpKCrVq1aJJkybMmjXrrYJ/V8mkSZI+DHPPzGVD0AaczJ3Y3mE7ZoZZzAP3yF898a5QQZ8/1UUUCwqVCm4dVA8aDzn633qXhuoesfKtQE/vpeb5kCylJkHADvUA9gfn/ltfpJJ6YHfV7mBs+drNMyKEIHrTJh7PmYtITcWwWDGKLV2CqZvb28crvfPyNGl61fXr11mwYAE+Pj6oVKo8qwhe0MmkSZLef5ciLtFvTz8Egp+a/0SDYhlMffIyZRqsbqqegLdKF+j6a/4E+iZeniRYlaZeZ1cG6g9FVbUX+27EaiVLlv/ehsu1ZCkyRD3H3sX1kBipXqdnqC7kWXsQlKj3Rk/9KePiCZ86ldjduwGwaNoUp7lz0Le2fvuYpfdCniZNN27cwM/PDz8/P44cOUJycjKNGzfGy8sLLy8vqlWr9lbBv6tk0iRJ77dkZTJd/+5KaGwoHct2ZGbDmVlvdPI72D8FTKxh+HmwKJL3gb6tDCYJjlFYsi61GevTWpJkXJgBjUoxsGEprM2yKLGQFZVS3dN1bjXcPID6AWnAqjjU6g/un7zVpMRJwTd4OGoUKaGhoK9PkXHjsBvQX96Ok7TkadKkp6eHvb09o0aNol27dri5uck3IDJpkqT33bKLy1h9dTWFTQuzo8MOrI2z6KmIugs/1IPUBPhohXqw9TtCpRIcuHyb2/tW0i5hByX0ngCQpjBEVaUbRo1GvN0kwfFP1UU+z/8K0ff+W1+m6b/zwHmD/hvPJ48Qgpht2wj/ZiYiORkDBweKLV6MWY0sBuxLH6Q8TZpGjx7N0aNHCQgIoEaNGpoepkaNGmFmlsW9/feYTJok6f0V8CyA3rt6oxRKlnotpZlLs8w3EAJ8usGtA+rxQf13FeiJdF9QqQR7r4ez7OBNzbQn1sZ6zKpwl9bP/8Tg0UtjjHI6SbAQ6jFK51ari28qU9TrTWzA/WP1lC+5ULVclZBA+DczidmxAwDzxo1xmv8tBrbZnN9P+uDky5im6Ohojh07phkQfv36ddzd3Tlx4sQbBf2uk0mTJL2fUlWp9N7Vm6DIIFq6tGSR16KsN7q2Ff78FPSNYMgJsC+f94G+BZVKsOdaOMt9/0uWLI0N+LRRKT59+Tbcy5MEC5V6XVaTBKfEw9Ut6mQp/Op/653cofZnUKVzhvPAvYnk27d5OHo0yTdvgZ4e9qNGUeizQSheGswuSa/Kyef3G/d/KpVKUlNTSU5OJikpieTk5Fyfe06SJEnX1lxbQ1BkENbG1kyuOznrDRKjYM8k9feNxxfohOlFsrTM9wY3HscBYGliwKcNX0mWXsjJJMFPbqjLBfhvhGT12CgMTNQD4msPhGI1c/VcYv75h7Bp0xEJCejbF6bYwkWY162Tq8eQpBz3NI0cORI/Pz8CAgKwtbXFw8MDT09PvLy8PujxTbKnSZLeP7ejb9Ptn26kqlKZ02gO7cu0z3qjv0fCxXVQuDwMOZ5x74uOvS5ZGtioFAMalsLaNJsDvBOjtCcJBvVExEUqwcPz/7WzLaVOlKr3USdUuXkuyck8nj2H6M2bATCrV49iCxdgULhwrh5Hen/laU9TWFgYgwcPxsvLiypVqrxxkJIkSQWZUqVk6smppKpSaVysMe1Kt8t6o9AT6oQJoP2yApcwqVSC3dfCWO578+2SpRdeN0nww/Pq6WLKt1InS6WbatV7yi0pd+/yYPQYkgMDQaGg8BdfUHjYUBT6uVSJXJJekeOkacuWLXkRhyRJUoGyMWgjV55cwdzQnKn1p2bdi56WDP+MUn9fsz+4ZFHDKR/lerL0KgMjqNZDXXgy9Jh6rrvy3mBTIheiz1js3r2ETfkKVXw8+nZ2OC2Yj0XDhnl2PEmCNxzTtH79elauXElISAinTp3CxcWFpUuXUqpUKTp06JDbMUqSJOWr+8/vs/zScgDG1hyLg7lD1hsdWwzPboJFUWg+I48jzJ4XydKygze5GfFfsjSoUWn6Nyz59snSqxQKKOWh/sojqpQUIuYvIOr33wEwrVmTYosXYVj0zes5SVJ25bi/9Mcff2Ts2LG0adOG6OhoTQVwGxsbli5dmtvxSZIk5SshBDNOzSAxLZFaRWvRtXzXrDd6EgzHF6u/bzUPTG3yNMasKFWCfy4/wnvpUYZvuMTNiDisTAwY07w8xyc1ZVTzcrmfMOWDlAcPudvnY03CVOizQbisWysTJinf5LinacWKFfz888907NiRefPmadbXqlWL8ePH52pwkiRJ+W3bzW2cCTuDib4JMxrMQE+Rxf8tVSr4Z7S67lA5b/WktzqiVAl2X1XfhnvRs2RlYsCgxuqeJSuTdy9ReuH5oUM8+nIyqthY9KytcZo3F8smTXQdlvSByXHSFBISgrt7+qqqxsbGxMfH50pQkiRJuvA4/jELzy8EYLj7cEpYZWNMzqXf4N5JMDSHtgt1UsTyfU6WRGoqEUuWEvmret4+k2pVKb54MYbFiuk4MulDlOOkqVSpUvj7++Pi4qK1fu/evVSsWDHXApMkScpPQghmnZ5FXGocboXd+Ljix1lv9Pwx7J+q/r7pV3k68DkjSpVg17/J0q33LFkCSA0P5+GYsSReugSAXb9PKDJuHAqjXJggWJLeQI6TprFjxzJs2DCSkpIQQnD27Fk2btzI3LlzWb16dV7EKEmSlOf2hu7F74EfBnoGzGgwA329bDy2vvdLdeFGx+pQ9/M8j/GF1yVLnzUuTb/3IFkCiDt2jEcTJqKMjkbPwgLHObOxatlS12FJH7gcJ02DBg3C1NSUr776ioSEBHr37o2TkxPLli2jZ8+eeRGjJElSnopMimTumbkADHYbTDnbcllvdGMfXN8GCn34aDlkJ8l6SxklS9amhgxqVOq9SZZEWhpPvvuOZz+tAiEwrlSR4kuXYlQif3vxJCkjOUqa0tLS2LBhA97e3vTp04eEhATi4uIoUqRIXsUnSZKU5+adnUdUchTlbMsxyG1Q1hskx8Gucerv6w8Fx2p5Gp9SJdh55RHLfW9y+4l67Oj7liwBpEZE8Gj8BBLOngXApldPin75JXrGBatIqPThylHSZGBgwJAhQwgMDATAzMwMMzOzPAlMkiQpP/jd92NPyB70FHp80+AbDPWzkYAcngMx98G6BHhlYz66N/S6ZOmzxqXo16Aklu9JsgQQf/o0D8dPQPn0KXpmZjh88w3W7drqOixJ0pLj23N16tTh0qVL6QaCS5IkvWtiU2KZeWomAP0q9aNK4WxMDfXoEpz5Uf19u8VgZJ7rcX1IyZJQqXi6ciVPv/seVCqMy5en2NKlGJcupevQJCmdHCdNQ4cOZdy4cTx48ICaNWtibq79B6Nq1aq5FpwkSVJeWnx+MRGJEZSwLMEX1b/IegNlmnpCXqGCKl2hXItcjSejZMnGzJDPGpfmk/ou71WyBJAWGcmjCROJP3ECAOsunXH46iv0TE11HJkkZSzHSdOLwd4jR47UrFMoFAghUCgUmgrhkiRJBdmZsDNsvbkVgBkNZmBqkI0P6jM/QvgVMLGBVnNzLRYhBHuvhbNgfzB3PoBkCSDhwgUejh1H2uPHKExMcJg6FZvOuisMKknZ8UbFLSVJkt5lCakJTDs5DYAerj2o5VAr642i7qrHMgG0nAkWufMAzP3IBKb+dY3DwU+A9z9ZEioVkb/+SsSSpaBUYlSqFMWWLcWkfHldhyZJWcpx0iTHMkmS9K5bcWkFD+Me4mDuwOgao7PeQAjYNRZSE8ClEbj3fesYUpUqfjkewtKDN0hKVWGkr8cQz9J85lH6vUyWAJTR0Tz6cjJxfn4AWLVrh+OM6eiZ5/64MEnKCzlOmiRJkt5l/hH++AT6ADC13lQsjCyy3ujaVrh1EPSNoP3St54q5cLdKKZsv0pQ+HMA6pW2Y3YnN8rYZyOWd1Ti5cs8GDOGtEdhKIyMKDplCjbdu6HQwbQzkvSmZNIkSdIHI0WZwrST/2/vvuNrPP8/jr/OOdl7yE5sYsVWYlVLpahRWl9qlmptovam9qr2p6VararRUtTWxi6i1J4xEkJkkXGyxzn3749D2tRKSCSRz/PxyEPOfa7rPp9bJHm77utc1xQUFNqWbUsTzybP7pQcY1j5G6DpKCiRg4UvnyA+JYN5u6+w9ngoigL2FsZMaFOFTrU9XtnwoCgKsT+tJnL+fMjIwLhkSTwXf45ZlSoFXZoQuSahSQhRbHxz7huC44NxMHNgdL3ROeu0ZwokRUMJb2g0/LleV1EUtp0LZ/q2S9xLTAPg/TqejGtdGQfLV3cfNV1CAuETJpLwxx8AWLdsidvMGWisrQu4MiGej4QmIUSxEBQTxPfnvwdgQv0J2JnZPbvTzcNwapXh87ZfgFHuA86t+0lM/O0Cf167B0A5J0tmvutDg7KOuT5XUZJy8SJh/iPICA0FY2NcRo/Gvnu3V3ZETRQPzxWa4uLi+PXXX7lx4wajRo3CwcGBU6dO4eLigoeHR17XKIQQLyRTn8mkI5PIVDJpUbIFLUvnYOPXjFTYNtzweZ0PoZRvrl4zPVPPt38G8+Xea6Rl6jExUjP4jfJ88npZTI3yf5+6gqIoCnG//ELkrNko6ekYu7vjsfhzzGUNP/EKyHVoOnfuHC1atMDW1pabN2/Sr18/HBwc2LRpE6GhoaxatSo/6hRCiOe28uJKLsdcxtrEmgkNJuSs0+FFcP8aWLlAi6m5er0TN2OYsPk8VyMNm+o2Ku/IjA4+lCnxar9LTJ+cTPiUqWi3bQPA6o03cJ89C42dXcEWJkQeyXVoGjFiBL1792bevHlY/+u+dOvWrfnggw/ytDghhHhRIfEhLD1j2PZkdL3RlDAv8exO0UHw5yLD563mgrldjl4rLjmdObuu8POJ2wA4Wpow8Z3KdKj56k70figtJISwocNIu3YNNBqcR4zAoc+Hr/x1i+Il16HpxIkTfPPNN48c9/DwICIiIk+KEkKIvKBX9Ew5OoV0fTqN3BvRvlz7HHTSw7ZhoM+Aim9DlQ7P7KIoCr+dCWPG9svcT0oHoEs9L8a2qoSdxas70fshbUAA4WPHoU9KQuNUAs/PP8eibg4WDBWiiMl1aDI1NUWr1T5y/OrVqzg5OeVJUUIIkRd+vvIzp6NOY2FkwWTfyTkb9Tj1I4QGgrEltF7wzDWZQu4lMfG38xy5fh+ACs5WzOroQ73SDnlxCYWakplJ9OLF3P9uBQDmdevgsWgRxs55s1q6EIVNrkNTu3btmD59OuvXrwcM+86FhoYyZswYOnXqlOcFCiHE8whLDGPxqcUADK8zHHcr92d3SoiAAMP2Krw5Eey8ntg0LVPHNweDWbL/OumZekyN1AxtXoF+TcpiYqTOgyso3DLv3SNsxKckHz8OgEPv3jh/OgKV8au5mrkQALn+zl64cCGJiYk4OzuTkpLC66+/Tvny5bG2tmbmzJm5OtehQ4do27Yt7u7uqFQqfvvtt2zPK4rC5MmTcXNzw9zcnBYtWnDt2rVsbWJiYujWrRs2NjbY2dnRt29fEhMTs7U5d+4cTZo0wczMDC8vL+bNm/dILRs2bKBSpUqYmZnh4+PDzp07c3UtQojCQ1EUph2dRkpmCrWda/M/7//lrOPusZAWD+61oP4nT2x2LPg+rb/4k0UBV0nP1NOkQgn+8G/KoDfKF4vAlHzqNCEdO5F8/DhqCws8Fn+Oy9gxEpjEKy/X3922trYEBASwbds2vvzySwYPHszOnTs5ePAglrncPygpKYkaNWrw1VdfPfb5efPm8eWXX7Js2TL++usvLC0t8fPzIzU1NatNt27duHjxIgEBAWzfvp1Dhw7x8ccfZz2v1Wpp2bIlpUqV4uTJk8yfP5+pU6eyfPnyrDZHjx6la9eu9O3bl9OnT9OhQwc6dOjAhQsXcvm3I4QoDH67/huB4YGYqE2Y2nAqalUOftQF7YaLm0GlMazJpH50WYDYpHRGbThLl+XHuBGdRAkrU77oUpNVfV6jlOOr/c44MITRmFU/catnTzKjojApV47SG9Zj8/bbBV2aEC+HUkgAyubNm7Me6/V6xdXVVZk/f37Wsbi4OMXU1FRZt26doiiKcunSJQVQTpw4kdVm165dikqlUsLCwhRFUZSvv/5asbe3V9LS0rLajBkzRvH29s563LlzZ6VNmzbZ6qlfv77yySef5Lj++Ph4BVDi4+Nz3EcIkfeikqIU37W+SrWV1ZTvzn2Xs06pCYqysIqiTLFRlN8nPvK0Xq9XNvx9W6k1/Q+l1JjtSqkx25Vxm84pcUnpeVx94aVLTFTujPhUueRdSbnkXUm5PXy4oktMLOiyhHhhufn9nes5TV9++eVjj6tUKszMzChfvjxNmzZFo3mxxdtCQkKIiIigRYsWWcdsbW2pX78+gYGBdOnShcDAQOzs7Kj7r3dptGjRArVazV9//cW7775LYGAgTZs2xcTkn3ew+Pn5MXfuXGJjY7G3tycwMJARI0Zke30/P79HbhcKIQo3RVGYcWwGCekJVHGsQq+qvXLWcf9M0N4Bu5LQbGy2p25EJzJh83mOBccA4O1izayO1ahT6tWf6P1QWnAId4YOIf36DTAywmXUSOx79pTlBESxk+vQ9PnnnxMdHU1ycjL29vYAxMbGYmFhgZWVFVFRUZQtW5b9+/fj5fXkSZTP8nD5AhcXl2zHXVxcsp6LiIjA+T/v0jAyMsLBwSFbmzJlyjxyjofP2dvbExER8dTXeZy0tDTS0tKyHj/uHYVCiJfrj1t/sO/2PoxURkxvOB0jdQ5+xIWdgr+WGT5v8zmYGG6zpWboWHrgBksP3CBdp8fMWM2w5hX5qEkZjDWv/rylh7S//0H4+PHok5IwcnLCY/HnWNSpU9BlCVEgcv2dP2vWLOrVq8e1a9e4f/8+9+/f5+rVq9SvX58vvviC0NBQXF1d8ff3z496C43Zs2dja2ub9fEiAVEI8eLiUuOY9dcsAPr69MXbwfvZnXSZsG0oKHrweR8qGEa2j964R+sv/uSLvddI1+l5vaITAf6vM6BZuWITmJTMTCLnzSds2DD0SUlY1K1LmU0bJTCJYi3XI00TJ05k48aNlCtXLutY+fLlWbBgAZ06dSI4OJh58+a98PIDrq6uAERGRuLm5pZ1PDIykpo1a2a1iYqKytYvMzOTmJiYrP6urq5ERkZma/Pw8bPaPHz+ccaNG5ftlp5Wq5XgJEQBmntiLjGpMZSzLcfH1T9+dgeAY19DxHkwswO/2dxPTGPmzstsOhUGgJO1KVPaVqGNj1uxuhWVGR1NmP8Ikv/+GwCHPn1w9h8u744TxV6u/8sUHh5OZmbmI8czMzOzbme5u7uTkJDwQoWVKVMGV1dX9u7dm3VMq9Xy119/4etr2DjT19eXuLg4Tp48mdVm37596PV66tevn9Xm0KFDZGRkZLUJCAjA29s76/air69vttd52Obh6zyOqakpNjY22T6EEAXj0J1DbA/ejgoV0xpNw0STg1W4Y2/CfsPIlNLyM9ZfTqX5ooNsOhWGSgU9GpRi76ev805192IVmJJPnjQsJ/D33w+WE1iMy+hREpiE4DlC0xtvvMEnn3zC6dOns46dPn2aAQMG8OabbwJw/vz5R+YRPU5iYiJnzpzhzJkzgGHy95kzZwgNDUWlUjF8+HBmzJjB1q1bOX/+PD179sTd3Z0OHToAULlyZd5++2369evH8ePHOXLkCIMHD6ZLly64uxsWsvvggw8wMTGhb9++XLx4kV9++YUvvvgi2yjRsGHD2L17NwsXLuTKlStMnTqVv//+m8GDB+f2r0cI8ZIlpicyPXA6AN2rdKeGU41nd1IU2PEpZKaQ7O7L/46XZ/TGc8QlZ1DJ1ZpNAxryWYdq2JgVn6CgKAoxP/7IrV69yYyOxqR8OUr/+is2b/sVdGlCFB65fWteeHi40qJFC0WlUikmJiaKiYmJolarlbfeekuJiIhQFEVR9u3bp/z+++/PPNf+/fsV4JGPXr16KYpieJvvpEmTFBcXF8XU1FRp3ry5EhQUlO0c9+/fV7p27apYWVkpNjY2yocffqgkJCRka3P27FmlcePGiqmpqeLh4aHMmTPnkVrWr1+vVKxYUTExMVGqVq2q7NixI1d/L7LkgBAvX3pmujL20Fil2spqytu/vq0kZyTnrOO5DYoyxUbJmFpCaTH+O6XUmO1KpYm7lG8OXlfSM3X5W3QhpEtMVO74+2ctJ3DHf4QsJyCKjdz8/lYpiqI8T9i6cuUKV69eBcDb2xtv7xxMunyFabVabG1tiY+Pl1t1QrwEQTFBTDg8gaDYIABWtFzBa26vPbtjcgzpX9TBJC2GhRnv8X+6jrxZyZnp7aviaW+Rz1UXPmnBwdwZMpT0Gw+WExg9Gvse3YvVLUlRvOXm93euJ4I/VKlSJSpVqvS83YUQ4rlk6DNYcX4F35z9hkwlEztTOyY1mJSjwBSdkMaN7wbRIC2Gq3oPNlm8x9J2NXm7mmuxDAna3b8blhNITsbI2dmwnEDt2gVdlhCF1nOFpjt37rB161ZCQ0NJT0/P9tyiRYvypDAhhPiva7HXmHB4ApdjLgPQvGRzJjaYSAnzEk/tp9cr/HziNnt2/cr3GPaV/LPSJHZ3bIF1MZq39JCSkUHUwkXErFwJgMVrr+GxaCFGJZ7+9yhEcZfr0LR3717atWtH2bJluXLlCtWqVePmzZsoikJt+R+KECIfZOozWXlxJV+f+ZoMfQY2JjaMrz+e1mVaP3OEKCgigfGbz3PhViQ7TZaDGu5V6k7fLl1fUvWFS0ZUFGEjRpDyt+Fdxw59++Ds74/K6LlvPAhRbOT6u2TcuHGMHDmSadOmYW1tzcaNG3F2dqZbt268LZs2CiHyWHBcMBMOT+DCfcMG2s08mzHZdzJOFk5P7ZeSruPLfdf49lAwmXqF0abbKKcKR7FyoUT7mS+j9EIn+e+/uePvjy76HmpLS9xmz8KmZcuCLkuIIiPXoeny5cusW7fO0NnIiJSUFKysrJg+fTrt27dnwIABeV6kEKL40el1rLq0iiWnl5CuT8fa2Jqx9cfStmzbZ44uHQiKYtKWC9yOSQGgZ/kUBtzdCnpQtZoH5nYv4QoKD+XBcgJR8xeATodphfJ4fPklpjlYGkYI8Y9chyZLS8useUxubm7cuHGDqlWrAnDv3r28rU4IUSyFxIcw6cgkzkafBaCxR2Om+k7FxdLlqf2itKlM336J7efCAXCzNWNq28r4/fUh6DOgYiuo0j7f6y9MdIlJhE+cSMLu3QDYvPMObtOnobYofu8UFOJF5To0NWjQgMOHD1O5cmVat27Np59+yvnz59m0aRMNGjTIjxqFEMWETq9jzeU1fHn6S9J0aVgZWzG63mg6lO/w1NElvV5hzfFQ5u2+QkJqJmoVfNioDP5vVcTq/Cq4fQxMrKDNAihG75JLu3HDsJxAcLBhOYGxY7Hv9kGxfKegEHkh16Fp0aJFJCYmAjBt2jQSExP55ZdfqFChgrxzTgjx3EK1oUw6MolTUacAaOjekGkNp+Fq+eQ9IAEuh2sZv/k8p0PjAPDxsGV2Rx+qediCNhwCphgavjkRbD3z8xIKFe2uXdydMBElORkjFxfDcgK1ahV0WUIUabkKTTqdjjt37lC9enXAcKtu2bJl+VKYEKJ40Ct61l1Zx+KTi0nVpWJhZMGoeqPoVKHTU0dE4lMy+DzgKj8du4VOr2BlasTIlhXp4VsajfpBv91jIE0L7rXgtRxu4lvEKRkZRC1YQMyPqwCwqF/fsJyAo2MBVyZE0Zer0KTRaGjZsiWXL1/Gzs4un0oSQhQXtxNuM/nIZP6O/BuA+q71md5oOu5W7k/so9cr/HrqDnN3XeF+kmF+ZatqrkxuWwU3W/N/GgbtgktbQKWBtl+CWpOv11IYZERFEeY/gpQHm5g79vsIp2HDZDkBIfJIrr+TqlWrRnBwcI425BVCiMfRK3o2BG1g4cmFpGSmYG5kzqd1PuV97/dRq568j/iFsHgmbbmQdSuurJMl09pVpUmF/yw/kJYAO0YaPvcdBG7V8+lKCo/kEye44z8C3b17qK2scJ8zG+sWLQq6LCFeKbkOTTNmzGDkyJF89tln1KlTB0tLy2zPy75rQoinuZt4l8lHJ/NX+F8A1HWpy/RG0/Gy9npin7jkdOb/HsTa46EoCliYaBjWvAIfNiqDidFjQta+maC9A3aloNnY/LqUQkFRFGJ+WEnUwoUPlhOogOf/fYlJ6dIFXZoQr5xcb9irVv/zA+rf8w0URUGlUqHT6fKuuiJENuwV4ukURWHjtY3MPzGf5MxkzDRmDK8znK6Vuj5xdEmnV/jlxG3m/36F2OQMANrVcGd868q42po9/oXCTsK3zQEFum+E8q/uaIsuMZHw8RNI+OMPAGzatcVt6lRZTkCIXMjXDXv379//3IUJIYqniKQIphydwtG7RwGo5VyLGY1mUNKm5BP7nLkdx+QtFzh3Jx6Aii5WTGtXDd9yT5nQrMuArcMABXw6v9KBKe3aNe4MHUZ6SAgYG+Mybiz2XbvKcgJC5KNch6bXX389P+oQQryCFEXht+u/Me/EPBIzEjHVmDK01lC6Ve6G5gkTs+8npjFvdxC//H0bAGtTI4a/VZGevqUw1jx5vhMAx76GyPNgbg9+s/L6cgqN+B07CJ802bCcgKsrnos/x7xmzYIuS4hX3nO9peLPP//km2++ITg4mA0bNuDh4cFPP/1EmTJlaNy4cV7XKIQogiKTIpkWOI0/w/4EoLpTdWY0mkEZ28e/iUSnV1jz1y0W/B6ENjUTgE61PRnTyhtn6yfcivu3mBDYP9vwecsZYPX0vemKIiU9ncj5C4j96ScALHwb4LFwIUYODgVcmRDFQ65D08aNG+nRowfdunXj1KlTpKWlARAfH8+sWbPYuXNnnhcphCg6FEVhW/A25hyfQ0J6AiZqEwbXGkzPKj2fOLr0980YJm+5yKVwLQBV3GyY3r4qdUvnMAwoCuwYAZkpULoJ1OyWV5dTaGRERhI23J+U06cBcPz4Y5yGDUWlefWXUhCisHiud88tW7aMnj178vPPP2cdb9SoETNmzMjT4oQQRUt0cjTTA6dz4M4BAKo5VmNm45mUtSv72PZRCanM2XWFTafCALAxM2KUnzcf1C/1zwKVOXH+V7ixDzSm8M7iV26rlKS/jhM2YgS6+/cNywnMnYN18+YFXZYQxU6uQ1NQUBBNmzZ95LitrS1xcXF5UZMQoohRFIWdITuZ9dcstOlajNXGDKw5kN5Ve2OkfvTHTIZOz6rAWywOuEpCWiYqFfyvrhej/LxxtDLN3Ysnx8DuB8sKNB0FJcrnwRUVDoqiEPP990Qt+tywnIC3N55ffoFJqVIFXZoQxVKuQ5OrqyvXr1+n9H/WADl8+DBlyz7+f5NCiFfXvZR7zDg2g72hewGo7FCZmY1nUsG+wmPbB964z9StFwmKTACguqct09tXo6aXXe5eODMNQv6EwCWQfA+cKkGjYS9yKYWKLjGR8HHjSQgIAMC2fTtcp05FbW7+jJ5CiPyS69DUr18/hg0bxvfff49KpeLu3bsEBgYycuRIJk2alB81CiEKqd03dzPz2Ezi0uIwUhvRv3p/+vj0wVht/EjbiPhUZu68zLazdwGwtzBm9NuV6FzXK+e34pJj4FoABO2A63sh3bB5OCo1tP0CjEzy6tIKVOrVq4QNGUr6rVtgbIzr+HHYdekiywkIUcByHZrGjh2LXq+nefPmJCcn07RpU0xNTRk5ciRDhgzJjxqFEIVMTGoMM4/N5I9bhkUVve29mdl4Jt4O3o+0Tc/U88OREL7ce42kdB0qFXSrX5KRLb2xs8hByIm9CVd2QtBOuHUUlH8toGvtBt6toMYH4FUvj66uYMVv20745MkoKSmG5QS+WIx5jRoFXZYQgudYEfyh9PR0rl+/TmJiIlWqVMHKyiqvaytSZEVwUVzsubWHz459RkxqDBqVhn7V+/Gxz8cYax4dXfrzWjRTtl4kODoJgNol7ZjevhrVPGyf/AJ6PYSffhCUdkHUxezPO1c1BKVKrcGtFqifsXZTEaGkpxM5bz6xq1cDYNnQF/cFC2Q5ASHyWb6uCL569Wo6duyIhYUFVapUee4ihRBFS1xqHLOOz2JXyC4AytuVZ2bjmVRxfPTnQFhcCjO2X2LXhQgASliZMLZVZTrW8kD9uFtxmWkQcsgwmhS0CxLC/3lOpYFSDcG7tSEsORTNzcKVzEwyo6LICAsj4+5d0sPCsj7PCLtLRng4ZBi2inHs/wlOQ4bIcgJCFDK5HmlycnIiJSWFdu3a0b17d/z8/NDIN7aMNIlX2v7Q/UwLnMb91PuoVWr6VutL/xr9MdFkv72Wlqnj20PBLNl/ndQMPRq1ip6+pRjeoiK25v8ZiUqOgWt/GILSv+cnAZhYQfnm4N0GKrwFFoV/tEXJyCAjMpKMOw+D0L9DURgZkZGQmfnUc2gcHHCbMQPrN994SVULIfJ1pCk8PJzdu3ezbt06OnfujIWFBe+//z7dunWjYcOGz120EKLwiU+LZ+7xuWwL3gZAOdtyzGg8g2olqj3Sdv+VKKZtu8jN+8kAvFbGgWntqlLZ7V8/hGJCDCNJT5uf5N0GyjQBo1wuPZDP9OnpZIaHP3GkKDMy0nBr8WmMjTF2c8PYwx1jDw+M3d0xefCnsYcHRs7OqIyea6MGIcRL8NxzmgCSk5PZvHkza9euZc+ePXh6enLjxo28rK/IkJEm8ao5dOcQ045OIyolCrVKTe+qvRlYcyCmmuxhJvR+MtO3X2LP5UgAnK1NmdCmMu1quKNSlH/NT9oJUZeyv4hzVcPcJO9WBT4/SZ+aSsbd8OyjQ/8aMcqMjjasPP4UKhOTrACU9aeHR1ZIMnJyQvWKzMES4lWRryNN/2ZhYYGfnx+xsbHcunWLy5cvv8jphBCFgDZdy/wT8/nt+m8AlLYpzYzGM6jhlP0dXKkZOpYeuMHSgzdIz9RjpFbRp3EZhjT1xDr8GGyfD1d3F5r5SfrkZEMI+s+tM8OI0V109+498xwqc/MHYcg9KxT9e6RI4+gooUiIV9hzhaaHI0xr1qxh7969eHl50bVrV3799de8rk8I8RIdCTvClKNTiEyORIWKnlV6MrjWYMyM/tkwV1EUAi5FMn37Je7EpgDQsqwxn1UOw+XuOvhyX4HMT9IlJj0IQoYQ9N+RIl1s7DPPobaw+Gd06DEjRRp7e1krSYhiLNehqUuXLmzfvh0LCws6d+7MpEmT8PX1zY/ahBAvSWJ6Igv+XsDGaxsBKGldkhmNZ1DLuVa2diH3kpi27SIHgqLxUkXib3mO7vYXcAg/heruy52fpCgKyYGBxPy0mpRTp9DFxz+zj9ra+j+hKPvcIrWtrYQiIcQT5To0aTQa1q9f/9h3zV24cIFq1R6dICqEKLwC7wYy5egUwpMMt9G6V+7O0NpDMTf6Z7uO5PRMvtp3lb8O76UZJxhvepKKqjugAx7e1cqan9Qa3Gvl26a5+tRU4rdtI3bVKtKuXc/2nMbW9jEjRf8EI43MNxRCvIAXmggOkJCQwLp16/juu+84efIkOp3u2Z1eQTIRXBQ1yRnJLDq5iF+CfgHA08qT6Y2mU8/1n5W1lYwU/t6/hbBjv+KrO4GLKu6fEzycn1SpjWFUyb50vtabERVF7Lp1xP38S9atNpWFBXbvvottx3cxKVUajZVlvtYghHj1vJSJ4IcOHWLFihVs3LgRd3d3OnbsyFdfffW8pxNCvEThieH0C+jHLe0tALp4d8G/jj8WxhaG9ZOu/k7iua0YheyjnpJKPQAVZBpZoKnYEpV365e2flLKhYvErPoR7a7dWYs/Gru7Y9+9O3bvdZLRIyHES5Or0BQREcHKlStZsWIFWq2Wzp07k5aWxm+//SargwtRRIQlhtH3976EJYbhaunKjEYzqG/qBH//AFd2ooQGolJ0PNwYKUJxINLtDSo3+x8m5Zu9lPWTFJ2OhD17iVm1ipSTJ7OOm9epg0PPnlg3f1PWMxJCvHQ5/qnTtm1bDh06RJs2bVi8eDFvv/02Go2GZcuW5Wd9Qog8dFt7m75/9CU8KZySFq6ssK6D66ZhEP3PciEq4LK+JAH62sSXbEmvjh2oUeLl3PbSJSQQ9+tGYlevJiMszHDQyAibVq1w6NkTcx+ZMymEKDg5Dk27du1i6NChDBgwgAoVKuRnTUKIfHBLe4s+v/chKjmK0qaOrLh6FueM4wAoKg0XjauxMak6Afo6aBxKM7VtVd6o5PxSaku/dYuYn1YTv2kT+mTDiuIaOzvsuvwP+64fYOzycuoQQoinyfEqbIcPHyYhIYE6depQv359lixZwr0cLAb3okqXLo1KpXrkY9CgQQA0a9bskef69++f7RyhoaG0adMGCwsLnJ2dGTVqFJn/2QPqwIED1K5dG1NTU8qXL8/KlSvz/dqEeFmC44P5cPeHRCVHUc7Ylh+uncU5I43MUk3ZVHoKddKW8Y52DOvUbejasgm/D2+a74FJURSSjv3F7YGDuPF2K2JXr0afnIxphfK4fjad8gf24zx8uAQmIUShkeORpgYNGtCgQQMWL17ML7/8wvfff8+IESPQ6/UEBATg5eWFtbV1nhd44sSJbO/Iu3DhAm+99Rbvv/9+1rF+/foxffr0rMcWFhZZn+t0Otq0aYOrqytHjx4lPDycnj17YmxszKxZswAICQmhTZs29O/fP2vBzo8++gg3Nzf8/Pzy/JqEeJluxN2g7+99uZ96nwpqC769fhFHvZ6rFT6iW3BLopMM/4FoVc2Vie9UwcPO/BlnfDH6tDS023cQs2oVaUFBWcctX2+KQ8+eWDZsKGslCSEKpRdaciAoKIgVK1bw008/ERcXx1tvvcXWrVvzsr5HDB8+nO3bt3Pt2jVUKhXNmjWjZs2aLF68+LHtd+3axTvvvMPdu3dxcXEBYNmyZYwZM4bo6GhMTEwYM2YMO3bs4MKFC1n9unTpQlxcHLt3785RXbLkgCiMrsZepd8f/YhJjaGSYszy0BDsFBWb3D/l0xs1ASjrZMm0dlVpUsEpX2vJjI4mdt3PxP78M7qYGMCwLYndux2w794D07Ivb0sVIYR4KDe/v19okyRvb2/mzZvHnTt3WLdu3YucKkfS09NZvXo1ffr0yfY/0TVr1lCiRAmqVavGuHHjSH4wJwIgMDAQHx+frMAE4Ofnh1ar5eLFi1ltWrRoke21/Pz8CAwMzOcrEiL/XIm5Qt/f+xKTGkOVTBXfhYZga2TJFOtpfHqjJmoV+LeoyO5hTfM1MKVevszdseO4/mZz7n39NbqYGIxcXXEe+SkVDuzHdfJkCUxCiCIhT96zq9Fo6NChAx06dMiL0z3Rb7/9RlxcHL1798469sEHH1CqVCnc3d05d+4cY8aMISgoiE2bNgGGZRL+HZiArMcRERFPbaPVaklJScHc/NHbFWlpaaSlpWU91mq1eXKNQuSFi/cv8vEfH6NN1+KTrmNZ+F1MzNx4P2UkJ7Wu2FsY82XXWvkWlhSdjsT9+4n5cRXJJ05kHTevWROHXj2xbtEClbFxvry2EELklyK10MmKFSto1aoV7u7uWcc+/vjjrM99fHxwc3OjefPm3Lhxg3LlyuVbLbNnz2batGn5dn4hntf56PN8EvAJCRkJ1EhLZ2l4JMlWlWlxbzBRih3VPW35ulttPO0tnn2yXNIlJhK/cSMxP60m484dw0GNBhs/Pxx69cS8Ro08f00hhHhZikxounXrFnv27MkaQXqS+vXrA3D9+nXKlSuHq6srx48fz9YmMjISAFdX16w/Hx77dxsbG5vHjjIBjBs3jhEjRmQ91mq1eHl55e6ihMhjZ6LOMGDPABIzEqmdmsrXEdEEmTfig+i+pGJKl3peTG1XFTNjzbNPlgvpt28Tu3o1cb9uRJ+UBIDa1hb7zp2x7/YBxg++14QQoigrMqHphx9+wNnZmTZt2jy13ZkzZwBwc3MDwNfXl5kzZxIVFYWzs+GtywEBAdjY2GStYu7r68vOnTuznScgIABfX98nvo6pqSmmpvm/MrIQOXUy8iQD9wwkOTOZeimpLImMZptRe8bFvIeRkRFz2lWly2sl8+z1FEUh5e+/uf/jjyTu3QcP3lNiUrYsDj17Ytu+Heon/KdDCCGKoiIRmvR6PT/88AO9evXC6F9bJ9y4cYO1a9fSunVrHB0dOXfuHP7+/jRt2pTq1asD0LJlS6pUqUKPHj2YN28eERERTJw4kUGDBmWFnv79+7NkyRJGjx5Nnz592LdvH+vXr2fHjh0Fcr1C5NaJiBMM2jOQFF0q9VNS+SIqhnm6vqxMeRMPO3OWdq9NdU+7PHktfXo62p07DUsGXPpnJXHLxo1x6NUTy0aNUKlf6D0mQghRKBWJ0LRnzx5CQ0Pp06dPtuMmJibs2bOHxYsXk5SUhJeXF506dWLixIlZbTQaDdu3b2fAgAH4+vpiaWlJr169sq3rVKZMGXbs2IG/vz9ffPEFnp6efPfdd7JGkygSjoUfY8ieQaTq02mUnMLce0kMTP2Ug/oaNKlQgi+61MLB0uSFXyfz/n1if/6Z2HU/o3uwsK3KzAzb9u1x6NEd0/LlX/g1hBCiMHuhdZrEP2SdJlEQjoQdYdi+IaTpM2ianMKEaIU+KSO5opRk0BvlGPGWNxr1iy0UmRoURMyPq9Bu346Sng6AkbMz9t26Ydf5fYzs7fPiUoQQokDk5vd3kRhpEkI86tCdQwzfN5QMRccbScl8FGVNh9SRpJg6sbxzDVpWff7J14peT+KBg8SsWkXysWNZx818fHDo1Qsbv5ayZIAQotiR0CREEbQ/dB8j9vuTiZ4WScn4RZTkf+mD8XIpwc/d61DWyeq5zqtLTCJ+82ZiVv9Exq1Qw0GNBuuWb+HQsyfmNWvKFidCiGJLQpMQRcyekN2MOjSaTBRaJibhHV6PgZndaVPDk7mdfLAwyf23dfqdMGLXrCHu11/RJyQAoLaxwe7993Do1g3jf62NJoQQxZWEJiGKkN1XNzE2cAo6oFViMiZ3WzNPac3EdyrzYaPSuRoFUhSFlFOniPlxFQl79oBeD4BJ6dLY9+yBXfv2qC0t8+lKhBCi6JHQJEQRseP8j4w/uQC9ClonpBB5txcXLRuy9oPavFbGIVfnSr91i7vjJ5By8mTWMcuGvtj37IlV06ayZIAQQjyGhCYhioBtJ75k4sXl6FUq3tamc+7uMCxL1mFHt9o425jl+DyKohC/aTMRM2eiJCejMjHBtn077Hv0wKxixXy8AiGEKPokNAlRyG0+OJkpIZtQVCpaxMOhu+Np1bAuE9pUxliT8xEhXVwc4VOmkvD77wBY1KuH+9w5Ml9JCCFySEKTEIWVorBh10CmRx8GlYomcaYcvDeeKV3q076mR65OlXTsL+6OGUNmZCQYGeE0dCiOffug0uTtHnRCCPEqk9AkRGGky2Tdpv8xK/kqALVj7bmSPom1g+pTyTXni6cq6elEffEFMd//AIqCSenSuM+fj7lPtfyqXAghXlkSmoQobFK1rPqlPfMxbFVSMcYTI/upbP5fbWzNc76gZFpwMGEjR2btD2fXuTMuY8egtrDIl7KFEOJVJ6FJiMIk7jYrfunAYjPDdiUe971pVmMWg96ogDqH26EoikLcL78QOWcuSmoqGjs73GZ8hnWLFvlZuRBCvPIkNAlRWNw9zbKNXfnKxjCa5BRXl3Ft5vK6t3OOT5EZE0P4hIkk7t8PgGXDhrjNno2xS87PIYQQ4vEkNAlRCCiXt7P0jyEstTNsf+KR8gbf9piDl0POb6Ul/vknd8eNR3fvHipjY5xHfop9jx6y5pIQQuQRCU1CFCRFIeXP/2PFqQV8Y28LQBWjd/mx7xTMjHP2zjZ9WhpRCxYS+9NPAJhWKI/7ggWYeXvnW9lCCFEcSWgSoqDoMond5M+qu1v47kFgau70IYtbj8jxKVKDrnJ35EjSrl0DwL57d5xHforaLOcLXgohhMgZCU1CFIS0BCK+78qa9POstDMEpu7lhzCm0cc56q7o9cSuXk3UgoUo6eloSpTAfdZMrJo2zc+qhRCiWJPQJMRLlhF7m3vLO7DK/B6r7QxrLg2rOZqPavTIWf+oKMLHTyDp8GEArJo1w23mDIwcHfOtZiGEEBKahHipYq6fQL/2f3xvr+dnG0Ngmlh/Mv+r9H6O+ifs3Uv4xEnoYmNRmZriMnYMdl26oFLlbDkCIYQQz09CkxAvybU/N+C2dzALHc351cYaFSqmNZzGuxXefWZffXIykXPnEffLLwCYVq6Mx4L5mJYrl99lCyGEeEBCkxD5TFEU/vp5DnWuzOUzJ3s2W1uhRs1njT+jXbl2z+yfcuEid0eNIj0kBACHvn1wGjYMtYlJfpcuhBDiXyQ0CZGPklPTOP7NAJrEbmSykwNbHwSmWU1m0aZsm6f2VXQ67n//PdFffAmZmRi5uOA+ZzaWvr4vqXohhBD/JqFJiHxy624U4d93o3HmccY7ObLTyhKNSsPcpnPxK+331L4Z4eHcHTOW5OPHAbBu2RLXaVMxsrd/GaULIYR4DAlNQuSDQyfPUWJbT+oQwmgnJwKszDFSGTH/9fm0KPX0PeC0u3cTPnkKeq0WlYUFrhPGY9uxo0z2FkKIAiahSYg8pNMrrN6yg7fODMFJFYO/sxsHLY0xUhux6PVFvFHyjSf3TUwicuZM4jdvBsCsenU85s3FpHTpl1S9EEKIp5HQJEQeiU1K59sfljMw+jNMVKkMdi3JUXMwVhuz+I3FNPV88sKTKWfOEDZqNBm3b4NajeMnH+M0cCAqY+OXeAVCCCGeRkKTEHng/J14dq+cyYiMb9Gp9Qz0rMBfRmmYakz54o0vaOTR6LH9lMxM7n3zDfe+Xgo6Hcbu7rjPm4tF3bov+QqEEEI8i4QmIV7Q+uO3SNg+jlHqHaSqVQwq48NxJR4zjRn/1/z/aODW4LH90u/c4e6o0aScPg2AzTvv4DplMhpr65dZvhBCiByS0CTEc0rL1DHzt1M0PDuOzpoTpKhUDPaux/G0CMyNzPmq+VfUc633SD9FUdBu20bEtOnok5JQW1nhOmUytm3bFsBVCCGEyCkJTUI8h7C4FMavCsD/3hRqaoLRqk0ZVrU+fyfexMLIgqUtllLbpfYj/XRaLRFTp6HduRMA89q1cZ83DxNPj5d9CUIIIXJJQpMQuXTk+j0Wr93C57pZeKrvEWdqz7DKtTgVfx0rYyuWtlhKTeeaj/RLPnGCsDFjyLwbDhoNToMH4divHyoj+TYUQoiiQH5aC5FDqRk6FgVc5fLhLawwXoyNKoUYu7IMLVuOs7FBWBtb881b3+Dj5JOtn5KRQfSSr7i/fDkoCsYlS+Ixfx7mNWoU0JUIIYR4HhKahMiBC2HxjFh/hvLRe/je+CuMVTriSjZgsJMt52MuY2Niw/KWy6nqWDVbv/SbNwkbNZrU8+cBsO3UEZdx49FYWRbEZQghhHgBEpqEeIpMnZ6vD9zgy73XaMshFph8gwY98VU78LFZMpdjLmNnase3Lb+lkkOlrH6KohC/cSMRM2ehpKSgtrXFbdo0bN5++vYpQgghCi8JTUI8wfWoRD5df4azd+LpotnHLOMVqFGIrdmFfup7BMVcxcHMgW9bfktF+4pZ/TJjY4mYPJmEgD0AWDRogPuc2Ri7uhbUpQghhMgDEpqE+A+9XuHHwJvM2XWFtEw9n5gFMI4fAIio3YMBShjXY6/jaObICr8VlLMrl9U36ehR7o4dR2ZUFBgb4zx8OA4f9kalVhfU5QghhMgjEpqE+Jc7scmM2nCOwOD7AMxx2UeXeENgulivB0OSLhGdEo2TuRPf+X1HWduyAOjT04n+fDExPxjampQti8eC+ZhVqVIwFyKEECLPFer//k6dOhWVSpXto1Klf+aNpKamMmjQIBwdHbGysqJTp05ERkZmO0doaCht2rTBwsICZ2dnRo0aRWZmZrY2Bw4coHbt2piamlK+fHlWrlz5Mi5PFCKKorD+79u8vfhPAoPvY26sZku1P+kS/x0Ae+p9QO/Yv4hOiaa8XXlWt16dFZjSrl/nZuf/ZQUmu65dKLPxVwlMQgjxiin0I01Vq1Zlz549WY+N/rWmjb+/Pzt27GDDhg3Y2toyePBgOnbsyJEjRwDQ6XS0adMGV1dXjh49Snh4OD179sTY2JhZs2YBEBISQps2bejfvz9r1qxh7969fPTRR7i5ueHnJ5N2i4PohDTGbTrPnsuGwF3by5YfvHZie2opCvBDnXf5/N5hABp7NGZ+0/lYmVihKAqx69YRNXceSloaGgcH3GbMwPrNNwrwaoQQQuQXlaIoSkEX8SRTp07lt99+48yZM488Fx8fj5OTE2vXruW9994D4MqVK1SuXJnAwEAaNGjArl27eOedd7h79y4uLi4ALFu2jDFjxhAdHY2JiQljxoxhx44dXLhwIevcXbp0IS4ujt27d+e4Vq1Wi62tLfHx8djY2LzYhYuXZtf5cCb8doGYpHSMNSr8W1Sgf8q3qI9/QwbwWY2WbNZeAaBrpa6MrjcaI7URmffuET5hIokHDwJg2aQJ7rNmYuTkVIBXI4QQIrdy8/u7UN+eA7h27Rru7u6ULVuWbt26ERoaCsDJkyfJyMigRYsWWW0rVapEyZIlCQwMBCAwMBAfH5+swATg5+eHVqvl4sWLWW3+fY6HbR6eQ7ya4lMy8P/lDAPWnCImKZ1KrtZsHdSQgYlLUB//hni1mv5VG7JZewW1Ss2418Yxvv54jNRGJB48SHD7DiQePIjKxASXCRPwWv6NBCYhhHjFFerbc/Xr12flypV4e3sTHh7OtGnTaNKkCRcuXCAiIgITExPs7Oyy9XFxcSEiIgKAiIiIbIHp4fMPn3taG61WS0pKCubm5o+tLS0tjbS0tKzHWq32ha5VvDx/Xotm9K/nCI9PRa2C/q+XY9ibZTDdMQzOruOWsQmDy1XhZvIdLI0tmdd0Hk09m6KkpxO16HNiHsx5M61YEfcF8zGrWPHpLyiEEOKVUKhDU6tWrbI+r169OvXr16dUqVKsX7/+iWHmZZk9ezbTpk0r0BpE7iSnZzJn1xVWBd4CoLSjBQs716SOpxVs6gcXN/O3mTnDPUsSnxaDm6UbS5ovoaJ9RdLvhBH26QhSz54DwL5nD5w//RS1qWlBXpIQQoiXqNDfnvs3Ozs7KlasyPXr13F1dSU9PZ24uLhsbSIjI3F9sIigq6vrI++me/j4WW1sbGyeGszGjRtHfHx81sft27df9PJEPjp5K5bWX/yZFZh6+pZi57Am1PGwgPW94OJmtljb0M/dlXhdCj4lfFjbZi0V7SuSsGcPIR07knr2HGobGzy/WoLr+PESmIQQopgpUqEpMTGRGzdu4ObmRp06dTA2Nmbv3r1ZzwcFBREaGoqvry8Avr6+nD9/nqioqKw2AQEB2NjYUOXB28F9fX2zneNhm4fneBJTU1NsbGyyfYjCJz1Tz7zdV3h/2VFu3k/G1caMn/q+xvT21bBQZcDPH6AP2sGXjo5MLGFHpqKjZamWfO/3PQ4aGyJmzuLO4CHotVrMa9Sg7OZNWDdvXtCXJYQQogAU6ttzI0eOpG3btpQqVYq7d+8yZcoUNBoNXbt2xdbWlr59+zJixAgcHBywsbFhyJAh+Pr60qBBAwBatmxJlSpV6NGjB/PmzSMiIoKJEycyaNAgTB+MEvTv358lS5YwevRo+vTpw759+1i/fj07duwoyEsXeeByuJYR689yOdww3+zdWh5MbVsVWwtjSEuEdV1IvXWY8S4uBFgY/j308+nH4FqDybx9h1v+I0h98IYBh759cB4+HJWxcYFdjxBCiIJVqEPTnTt36Nq1K/fv38fJyYnGjRtz7NgxnB68S+nzzz9HrVbTqVMn0tLS8PPz4+uvv87qr9Fo2L59OwMGDMDX1xdLS0t69erF9OnTs9qUKVOGHTt24O/vzxdffIGnpyffffedrNFUhOn0CssPBbMoIIgMnYK9hTGz3vWhlY+boUFqPKx5n3t3TzDU3Y3zJkYYqY2Y1nAa7cq1Q7t7N+ETJ6FPTERjZ4fbnNlYN2tWoNckhBCi4BXqdZqKElmnqXC4eS+JTzec5eStWABaVHZhdkcfnKwfzD9KjoHVHbkafYHBri6EG6mxNbVlcbPF1Lb3IXLOHOLW/QyAee3aeCxaKBvtCiHEKyw3v78L9UiTEDmlKAqrj91i1s4rpGTosDI1YkrbKrxXxxOVSmVolBgNP3XgkPY6o9xdSVarKG1Tmq+af4VrjJ6bXbqSdvkyAI4ff4zT0CGojORbRAghhIH8RhBFXnh8CqN/Pcef1+4B0KCsAwver4GnvcU/jbThsKo9a9LDmOfihF6l4jXX11jUbBEEHCZk8mT0ycloHBxwnzsXqyaNC+hqhBBCFFYSmkSRpSgKW87cZfKWC2hTMzE1UjPm7Ur0blgatVr1T8O422SuastcVRw/OzoA0LFCR8bXGEnMjPnEbdgAgEW9ergvWICxi3NBXI4QQohCTkKTKJJiktKZsPk8uy4YVnav4WnLws41Ke9s9Z+GwST+2J6RZikcsbBGhQr/Ov50NW9KWJdupF27BioVJQYMoMTAAXI7TgghxBPJbwhR5Oy5FMnYTee5l5iGkVrF0OYVGNisHEaa/yw7Fn2Vu6vbMchK4bqJOWYaU+Y0mUud0wncnPY+SkoKmhIl8Jg/D8tnrMslhBBCSGgSRUZCagafbb/E+r/vAFDB2YrP/1eTah62jzaOvMjZtR0YamtMjEaDk5kj/9doIfZfbSR882YALHwb4DFvnmy0K4QQIkckNIkiIfDGfUZuOEtYXAoqFXzUuAyftvTGzFjzaOO7Z9i94T0m2JmRrlZRya48i8uMJrX/FOKv3wC1mhKDB1Hik09QaR7TXwghhHgMCU2iUEvN0DFvdxDfHwkBwMvBnAXv1aB+WcfHtldCj7N8SzeW2Bn2DWzm5sukuDeI6TkQJTUVIycn3BcuwPK1117aNQghhHg1SGgShda5O3H4/3KGG9FJAHR9zYsJbapgZfr4f7bpwQeY+vvHbLMxA6B3yY78b0cS97cbVoC3bNwY97lzMHJ8fOASQgghnkZCkyh0MnR6/m/fdb7afx2dXsHJ2pR5narzRqUnLwUQe2Urww+O4pSFKRpgulNvqs7dQ8LNm6DR4DRsGI4f9UWlLlJ7VAshhChEJDSJQuVaZAL+689wIcywye471d34rH017C1Nntgn+MwqBv89m9tmJlgrKr5Mfh+rhatIT0/HyNUVj4ULsKhT52VdghBCiFeUhCZRKOj0Ct8fDmH+H0GkZ+qxNTfmsw7VaFfD/an9jh1bxIhLK0gwNqJsmjHzztZBv3ctCmD1+uu4zZmNkb39y7kIIYQQrzQJTaLA3Y5J5tMNZzkeEgPA6xWdmPdedVwezE16ko37xjEjdBuZGjV+UWZ8/IctutuHwcgIZ39/HD7sLbfjhBBC5BkJTaLAKIrCLydu89n2SySl67Aw0TCxTRW6vub1zya7j6FX9Cze+RE/3DsBwIgzFjTYk4IuIwwjdzc8Fy3CvGbNl3QVQgghigsJTaJARGlTGbvpPPuuRAFQr7Q9C96vQSlHy6f2S85IZtzWruxLDMYiVWH+7+Y4XTLMf7Jq3hz3mTPQ2Nnld/lCCCGKIQlN4qXbfu4uE3+7QFxyBiYaNSP9KtK3cVk06iePLgFEJkUyZFsXLqfdo1KYnknbjDGOTQRjY1xGjcS+R4+njlAJIYQQL0JCk3hp4pLTmbzlIlvP3gWgqrsNizrXxNvV+pl9L9+/zOBdvYnKTOK94zreP6hCpUvD2NMTj88XYe7jk9/lCyGEKOYkNImX4veLEUzecoFIbRoatYqBzcox5M0KmBg9e6L2/tB9jDkwAnVyJlO2ZVL1hgpQsG7ZErcZn6Gxscn/CxBCCFHsSWgS+SpSm8qULRfZfTECgLIlLFnYuQa1Sj57GQBFUVh18UcWnlxI+TCFMZszsUlQoTI2xnnsGOw/+EBuxwkhhHhpJDSJfKHXK/x84jazd10mITUTjVrFJ03LMrR5hcdvsvsfGfoMZh2bycarv/LOXwrdDupQ61UYlyxpuB1XtepLuAohhBDiHxKaRJ67EZ3IuI3nOX7TsO5SDU9bZnesThX3nN1G06ZrGbF/BJeCjzF6u446NwBU2LRuhev06WisrPKveCGEEOIJJDSJPJOeqWfZwRss2XeddJ0ec2MNI/286d2w9DPfGffQbe1tBu0diMnFEOZt0eGYACpjI1wmTsKu8/tyO04IIUSBkdAk8sSp0FjGbjzH1chEwLCq94wO1fBysMjxOU5HnWb4niE0PRjL/w7p0Shg4l4Cj6XfYebtnV+lCyGEEDkioUm8kMS0TBb8HsSPgTdRFHCwNGFK2yq0q+Geq1Gh7cHbWfDHJPpvSaNmiAKATbM6uC38BrXl0xe8FEIIIV4GCU3iue29HMmk3y5wNz4VgI61PZjYpgoOliY5PoeiKHx99msOblvKzC06HBJBpVFwHdwd2/4T5HacEEKIQkNCk8i16IQ0pm67yI5z4QB4OZgz610fmlRwytV50nRpTD40EfO1O5lyWI9aARNbHZ5zZ2DarHN+lC6EEEI8NwlNIscURWHD33eYufMy8SkZqFXQr0lZhreoiLnJs5cR+Lf7KfcZv2Ugb648T/WbhttxtuUzcF20EnXFxvlRvhBCCPFCJDSJHAm5l8T4TecJDL4PGLZAmdupOtU8bHN9ruux1/ni2770/iUKuyRQjBQ8fDOwnbYR3GvmceVCCCFE3pDQJJ4qQ6fn2z+D+WLPNdIy9ZgZq/FvUZG+jctgpHn2Fij/dST0Tw5/NoT+f6ahBlR2mZRtrmA6dBu4VMn7CxBCCCHyiIQm8URnb8cxdtN5LodrAWhcvgSz3vWhpGPOlxF4SFEUfg38FvXUxbQLNdyOMy+XTMmm5qj7bAWninlZuhBCCJHnJDSJRySnZ7Lwj6v8cCQEvQJ2FsZMalOFjrU9cv1uNr1ez9GAlVz/ZQVVT8VglQoZxuBRNxbH6k7Qaws4lM2nKxFCCCHyjoQmkc2BoCgmbL5AWFwKAO1rujPpnSqUsDLN1XnSQkM5v2YJqTt+x/FeOo4PjieVAJ/XIjEtXQp6bgU7rzy+AiGEECJ/SGgSANxPTOOz7Zf47cxdADzszJnxbjXe8HbO8Tl0cXHE79rFnV/XYHzxBpaAJZBmDPd9SlDVKZxKttGonCsaApONW/5cjBBCCJEPJDQVc4qisPl0GJ9tv0RssmEZgd4Ny/Bpy4pYmj77n4c+PZ3EAweI37qNhIMHUGVkYgzoVXCptAZd/VK0MA+mZso5QwcXH+ixGaxyt6aTEEIIUdAkNBVjofeTmfDbef68dg+ASq7WzO1UnRpedk/tpygKKadOEb91G9rdu9DHGyaKq4CbznC0uglO9b35IPocJbSHIQWwdofG/lC7Jxib5e+FCSGEEPlAQlMxlKnT8/2REBYFXCU1Q4+JkZrhLSrQr0lZjJ+yjEBaSAjabduI37qNjDt3so7ft4bDVVUE+pjSsHxlhl3/mxI3fjc8aeNhCEu1ekhYEkIIUaTlfqGdl2j27NnUq1cPa2trnJ2d6dChA0FBQdnaNGvWDJVKle2jf//+2dqEhobSpk0bLCwscHZ2ZtSoUWRmZmZrc+DAAWrXro2pqSnly5dn5cqV+X15BeJCWDwdvj7CrJ1XSM3Q06CsA78Pb8rAZuUfG5gyY2KIWb2GkM7/I7hVa+59vZSMO3dIN1VzwEfF9K5qhg82R+lQnW9JZMzpHZRIiARbL3jncxh6Gl7rJ4FJCCFEkVeoR5oOHjzIoEGDqFevHpmZmYwfP56WLVty6dIlLP+1832/fv2YPn161mMLi3/WEdLpdLRp0wZXV1eOHj1KeHg4PXv2xNjYmFmzZgEQEhJCmzZt6N+/P2vWrGHv3r189NFHuLm54efn9/IuOB+lpOtYvOcq3x0OQadXsDEzYkKbynSu6/XIMgL61FQS9+8nfstWEg8fhgcBU9GoCaloy7YK8ZyoqEIxNaWTVXkW3ziFS8guQ2e7ktDkU6jxARjlfONeIYQQorBTKYqiFHQRORUdHY2zszMHDx6kadOmgGGkqWbNmixevPixfXbt2sU777zD3bt3cXFxAWDZsmWMGTOG6OhoTExMGDNmDDt27ODChQtZ/bp06UJcXBy7d+/OUW1arRZbW1vi4+OxsbF5sQvNY4ev3WP85vOExiQD0Ka6G1PaVsHZ+p/RH0WvJ/nE38Rv3ULC73+gT0zMek7vXYZDVVWs8bxFvKUKI7URHS3L0S/4NK6JhvlQ2JeGJiOhRhfQGL/MyxNCCCGeW25+fxfqkab/io+PB8DBwSHb8TVr1rB69WpcXV1p27YtkyZNyhptCgwMxMfHJyswAfj5+TFgwAAuXrxIrVq1CAwMpEWLFtnO6efnx/Dhw/P3gvJZbFI6M3ZcZuMpw/wjN1szPmtfjRZV/vm7SLt+nfgtW4nfvp3M8PCs40bubqQ1b8CakrfZrj9jOKYy4j2rsvS7cRr3pABDQ/sy0HQUVO8sYUkIIcQrrciEJr1ez/Dhw2nUqBHVqlXLOv7BBx9QqlQp3N3dOXfuHGPGjCEoKIhNmzYBEBERkS0wAVmPIyIintpGq9WSkpKCubn5I/WkpaWRlpaW9Vir1ebNheYBRVHYevYu07dd4n5SOioV9GxQipF+3libGZMZHU38jh1ot24j9dKlrH5qa2ts3vYj5vXqfK7fx5/h20APGpWG9pZl6Bd8Gs/gvYbGjuUNYanae6ApMv+MhBBCiOdWZH7bDRo0iAsXLnD48OFsxz/++OOsz318fHBzc6N58+bcuHGDcuXK5Vs9s2fPZtq0afl2/ud1JzaZib9d4EBQNAAVXayY3bE6tZxMSQjYTejWbSQdOQJ6vaGDkRFWTZti264td6q7MvPydxy8Y5gfplFpeMeiFJ8En8UreZ+hvWMFeH00VOsEak1BXKIQQghRIIpEaBo8eDDbt2/n0KFDeHp6PrVt/fr1Abh+/TrlypXD1dWV48ePZ2sTGRkJgKura9afD4/9u42Njc1jR5kAxo0bx4gRI7Iea7VavLwKbksQnV7hx6M3WfBHEMnpOkw0agY3K0NP03skfz2XawEB6JOTs9qb16iBTbu22LRuzXUlkvlnvmZ/wH4A1Kh5x6Ikn4Sco2RyiKFDCW9DWKr6roQlIYQQxVKhDk2KojBkyBA2b97MgQMHKFOmzDP7nDlzBgA3N8MWHb6+vsycOZOoqCicnQ1bggQEBGBjY0OVKlWy2uzcuTPbeQICAvD19X3i65iammJqmrv92PLL5XAtYzed5+ztOADaWiYxWHcN9WdzCI+Kympn7OWFbdu22LZri0np0gTFBDH9zDT2hhpuualQ0drCi09CzlMm+aahk3MVw224Kh1AXahXqBBCCCHyVaF+99zAgQNZu3YtW7ZswdvbO+u4ra0t5ubm3Lhxg7Vr19K6dWscHR05d+4c/v7+eHp6cvDgQcCw5EDNmjVxd3dn3rx5RERE0KNHDz766KNsSw5Uq1aNQYMG0adPH/bt28fQoUPZsWNHjpccKIh3z6Vm6Phy7zWWHwrGNikOv4izvHf/HOZ3bma1UdvaYtPqbWzbtce8Vk1UKhXXYq+x9OxSAm4ZJnOrUPG2uSf9b56nbPKDuVku1QwjS5XaSlgSQgjxysrN7+9CHZr+u37QQz/88AO9e/fm9u3bdO/enQsXLpCUlISXlxfvvvsuEydOzHbht27dYsCAARw4cABLS0t69erFnDlzMDL6Z6DtwIED+Pv7c+nSJTw9PZk0aRK9e/fOca0vOzQF3rjPtF9O4HnhGG/ePkXN6OuoMHwpVcbGWDVrhm37dlg2bYraxLBe0o24Gyw9u5Q/bv6B8qCtn5kH/W9epHzKg7Dk6gOvjwHvNhKWhBBCvPJemdBUlLys0BSnTWHVkvUY7fsd3/ALmOkysp4zr1MH23btsHnbD42tbdbx4Phglp1dxu6Q3Vlh6S1TN/rfukTFlARDI7ca8PpY8G4FTwirQgghxKvmlV2nqbhSFIXUixc5+/062BfAW6kJWc9pSpbCvkM7bNu1w+Q/k+RvaW+x7OwydobsRK8Y3i3X3NSVAbcu4Z0SamjkXssQlir6SVgSQgghnkJCUyGXsGcPdxcsQn8zhIdjRwlmVpi+5UfFHp0x8/F55Dbmbe1tlp1bxvbg7VlhqZmJMwNDr1D5YVjyqAvNxkL5FhKWhBBCiByQ0FTI/XUzFo+bIaSrjfjLvSrmbdryfv9OmJk/ugHu7YTbLD+3nG03tqFTdAC8buLEgFtXqJr6ICx5vgbNxkC55hKWhBBCiFyQ0FTImTdqzKK9nUl8rTHTPmhARRfrR9qEJYbx7blv2XJ9C5mKYXPdxsaODLx9FZ+HI0slfQ0TvMs2k7AkhBBCPAcJTYVco8puKNOH0LCcI2p19rATnhjO8vPL+e3ab1lhqaGRAwNvX6XGw5GlUo0MYalMUwlLQgghxAuQ0FQENK5QItvjiKQIvjv/HRuvbSRTbwhLDYzsGXTnKjUfjiyVbmKYs1S68csuVwghhHglSWgqQqKSo/ju/Hf8evVXMvSGpQZeM7Jl4J0b1HkYlsq8bghLpRoWYKVCCCHEq0dCUxEQnRzNigsr2BC0gXR9OgB1NDYMunODeg9vw5V703AbrmSDAqxUCCGEeHVJaCrkNl7dyOzjs0nTpQFQW2PNwLBgXksJRQWGJQNeHwNerxVonUIIIcSrTkJTIVfOrhxpujRqqq0YeDeEBg/DUgU/Q1jyrFPQJQohhBDFgoSmQq5mcCBrwyKplv4gLFVsZdhI16N2QZcmhBBCFCsSmgq7Uo3wSU+HSu8YwpJbjYKuSAghhCiWJDQVdm7VYdgZsC9d0JUIIYQQxZq6oAsQOSCBSQghhChwEpqEEEIIIXJAQpMQQgghRA5IaBJCCCGEyAEJTUIIIYQQOSChSQghhBAiByQ0CSGEEELkgIQmIYQQQogckNAkhBBCCJEDEpqEEEIIIXJAQpMQQgghRA5IaBJCCCGEyAEJTUIIIYQQOSChSQghhBAiB4wKuoBXhaIoAGi12gKuRAghhBA59fD39sPf408joSmPJCQkAODl5VXAlQghhBAitxISErC1tX1qG5WSk2glnkmv13P37l2sra1RqVR5em6tVouXlxe3b9/GxsYmT88tck++HoWLfD0KF/l6FD7yNXk6RVFISEjA3d0dtfrps5ZkpCmPqNVqPD098/U1bGxs5B98ISJfj8JFvh6Fi3w9Ch/5mjzZs0aYHpKJ4EIIIYQQOSChSQghhBAiByQ0FQGmpqZMmTIFU1PTgi5FIF+Pwka+HoWLfD0KH/ma5B2ZCC6EEEIIkQMy0iSEEEIIkQMSmoQQQgghckBCkxBCCCFEDkhoEkIIIYTIAQlNhdxXX31F6dKlMTMzo379+hw/frygSyq2Zs+eTb169bC2tsbZ2ZkOHToQFBRU0GUJYM6cOahUKoYPH17QpRRrYWFhdO/eHUdHR8zNzfHx8eHvv/8u6LKKJZ1Ox6RJkyhTpgzm5uaUK1eOzz77LEf7q4knk9BUiP3yyy+MGDGCKVOmcOrUKWrUqIGfnx9RUVEFXVqxdPDgQQYNGsSxY8cICAggIyODli1bkpSUVNClFWsnTpzgm2++oXr16gVdSrEWGxtLo0aNMDY2ZteuXVy6dImFCxdib29f0KUVS3PnzmXp0qUsWbKEy5cvM3fuXObNm8f//d//FXRpRZosOVCI1a9fn3r16rFkyRLAsL+dl5cXQ4YMYezYsQVcnYiOjsbZ2ZmDBw/StGnTgi6nWEpMTKR27dp8/fXXzJgxg5o1a7J48eKCLqtYGjt2LEeOHOHPP/8s6FIE8M477+Di4sKKFSuyjnXq1Alzc3NWr15dgJUVbTLSVEilp6dz8uRJWrRokXVMrVbTokULAgMDC7Ay8VB8fDwADg4OBVxJ8TVo0CDatGmT7ftEFIytW7dSt25d3n//fZydnalVqxbffvttQZdVbDVs2JC9e/dy9epVAM6ePcvhw4dp1apVAVdWtMmGvYXUvXv30Ol0uLi4ZDvu4uLClStXCqgq8ZBer2f48OE0atSIatWqFXQ5xdLPP//MqVOnOHHiREGXIoDg4GCWLl3KiBEjGD9+PCdOnGDo0KGYmJjQq1evgi6v2Bk7dixarZZKlSqh0WjQ6XTMnDmTbt26FXRpRZqEJiGew6BBg7hw4QKHDx8u6FKKpdu3bzNs2DACAgIwMzMr6HIEhv9I1K1bl1mzZgFQq1YtLly4wLJlyyQ0FYD169ezZs0a1q5dS9WqVTlz5gzDhw/H3d1dvh4vQEJTIVWiRAk0Gg2RkZHZjkdGRuLq6lpAVQmAwYMHs337dg4dOoSnp2dBl1MsnTx5kqioKGrXrp11TKfTcejQIZYsWUJaWhoajaYAKyx+3NzcqFKlSrZjlStXZuPGjQVUUfE2atQoxo4dS5cuXQDw8fHh1q1bzJ49W0LTC5A5TYWUiYkJderUYe/evVnH9Ho9e/fuxdfXtwArK74URWHw4MFs3ryZffv2UaZMmYIuqdhq3rw558+f58yZM1kfdevWpVu3bpw5c0YCUwFo1KjRI0twXL16lVKlShVQRcVbcnIyanX2X/EajQa9Xl9AFb0aZKSpEBsxYgS9evWibt26vPbaayxevJikpCQ+/PDDgi6tWBo0aBBr165ly5YtWFtbExERAYCtrS3m5uYFXF3xYm1t/chcMktLSxwdHWWOWQHx9/enYcOGzJo1i86dO3P8+HGWL1/O8uXLC7q0Yqlt27bMnDmTkiVLUrVqVU6fPs2iRYvo06dPQZdWpMmSA4XckiVLmD9/PhEREdSsWZMvv/yS+vXrF3RZxZJKpXrs8R9++IHevXu/3GLEI5o1ayZLDhSw7du3M27cOK5du0aZMmUYMWIE/fr1K+iyiqWEhAQmTZrE5s2biYqKwt3dna5duzJ58mRMTEwKurwiS0KTEEIIIUQOyJwmIYQQQogckNAkhBBCCJEDEpqEEEIIIXJAQpMQQgghRA5IaBJCCCGEyAEJTUIIIYQQOSChSQghhBAiByQ0CSGKnZs3b6JSqThz5ky+vUbv3r3p0KFDvp1fCPHySWgSQhQ5vXv3RqVSPfLx9ttv56i/l5cX4eHhsuWKECJXZO85IUSR9Pbbb/PDDz9kO2ZqapqjvhqNBldX1/woSwjxCpORJiFEkWRqaoqrq2u2D3t7e8CwT+DSpUtp1aoV5ubmlC1bll9//TWr739vz8XGxtKtWzecnJwwNzenQoUK2QLZ+fPnefPNNzE3N8fR0ZGPP/6YxMTErOd1Oh0jRozAzs4OR0dHRo8ezX93qNLr9cyePZsyZcpgbm5OjRo1stX0rBqEEAVPQpMQ4pU0adIkOnXqxNmzZ+nWrRtdunTh8uXLT2x76dIldu3axeXLl1m6dCklSpQAICkpCT8/P+zt7Tlx4gQbNmxgz549DB48OKv/woULWblyJd9//z2HDx8mJiaGzZs3Z3uN2bNns2rVKpYtW8bFixfx9/ene/fuHDx48Jk1CCEKCUUIIYqYXr16KRqNRrG0tMz2MXPmTEVRFAVQ+vfvn61P/fr1lQEDBiiKoighISEKoJw+fVpRFEVp27at8uGHHz72tZYvX67Y29sriYmJWcd27NihqNVqJSIiQlEURXFzc1PmzZuX9XxGRobi6emptG/fXlEURUlNTVUsLCyUo0ePZjt33759la5duz6zBiFE4SBzmoQQRdIbb7zB0qVLsx1zcHDI+tzX1zfbc76+vk98t9yAAQPo1KkTp06domXLlnTo0IGGDRsCcPnyZWrUqIGlpWVW+0aNGqHX6wkKCsLMzIzw8HDq16+f9byRkRF169bNukV3/fp1kpOTeeutt7K9bnp6OrVq1XpmDUKIwkFCkxCiSLK0tKR8+fJ5cq5WrVpx69Ytdu7cSUBAAM2bN2fQoEEsWLAgT87/cP7Tjh078PDwyPbcw8nr+V2DEOLFyZwmIcQr6dixY488rly58hPbOzk50atXL1avXs3ixYtZvnw5AJUrV+bs2bMkJSVltT1y5AhqtRpvb29sbW1xc3Pjr7/+yno+MzOTkydPZj2uUqUKpqamhIaGUr58+WwfXl5ez6xBCFE4yEiTEKJISktLIyIiItsxIyOjrMnTGzZsoG7dujRu3Jg1a9Zw/PhxVqxY8dhzTZ48mTp16lC1alXS0tLYvn17VsDq1q0bU6ZMoVevXkydOpXo6GiGDBlCjx49cHFxAWDYsGHMmTOHChUqUKlSJRYtWkRcXFzW+a2trRk5ciT+/v7o9XoaN25MfHw8R44cwcbGhl69ej21BiFE4SChSQhRJO3evRs3N7dsx7y9vbly5QoA06ZN4+eff2bgwIG4ubmxbt06qlSp8thzmZiYMG7cOG7evIm5uTlNmjTh559/BsDCwoLff/+dYcOGUa9ePSwsLOjUqROLFi3K6v/pp58SHh5Or169UKvV9OnTh3fffZf4+PisNp999hlOTk7Mnj2b4OBg7OzsqF27NuPHj39mDUKIwkGlKP9ZTEQIIYo4lUrF5s2bZRsTIUSekjlNQgghhBA5IKFJCCGEECIHZE6TEOKVI7MOhBD5QUaahBBCCCFyQEKTEEIIIUQOSGgSQgghhMgBCU1CCCGEEDkgoUkIIYQQIgckNAkhhBBC5ICEJiGEEEKIHJDQJIQQQgiRAxKahBBCCCFy4P8BoO4EQHLiAVQAAAAASUVORK5CYII=", + "image/png": "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", "text/plain": [ "
" ] @@ -128285,7 +98649,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -128295,7 +98659,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -128305,7 +98669,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -128315,7 +98679,7 @@ }, { "data": { - "image/png": "iVBORw0KGgoAAAANSUhEUgAAAjsAAAHCCAYAAAAJowgXAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjguMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8g+/7EAAAACXBIWXMAAA9hAAAPYQGoP6dpAABxwElEQVR4nO3deVyN6f8/8Ndp30u00kooIiNMGrJkQpZobNNQ1rHEWLLNoKzhYx37YIqhse87ETOEZCLEWCJDdpWkUt2/P/y6v44W59RJOV7Px+M8Hs59X+e63td9znHe3dd1X7dEEAQBREREREpKpbwDICIiIipLTHaIiIhIqTHZISIiIqXGZIeIiIiUGpMdIiIiUmpMdoiIiEipMdkhIiIipcZkh4iIiJQakx0iIiJSakx26JO6e/cuJBIJwsPDyzuUL5ayvgel7ZdEIkFISIhCYyJpISEhkEgkJXptixYt0KJFiwoTD31emOyQQoWHh0MikRT6mDBhQpm0OWvWLOzatatM6v6cRUREYNGiReUdhpSKGBMpt4yMDISEhCAqKqq8Q6FyJOG9sUiRwsPD0bdvX0ybNg12dnZS++rWrYv69esjKysL6urqUFVVVUibenp6+O6775TuTEVpdejQAVeuXMHdu3eltguCoPD3oLQxKUJp+5WZmQk1NTWoqakpPDZ6JycnBzk5OdDS0pL7tflndeRNWp49ewYTExMEBwcXOHNXmnjo88JvNZWJdu3awdXVtdB9svzH8vr1a+jq6io6rM9aRkYGdHR0Sl2PRCL5LP5zz8zMhIaGBlRUZDsBXdp+fQ7HpDQqwneqoiWTFS0eKjscxqJPqrB5FQEBAdDT08Pt27fRvn176Ovrw8/PDwBw8+ZN+Pr6wtzcHFpaWqhWrRp69uyJ1NRUAO9+4F6/fo1169aJw2UBAQHFxnDv3j106tQJurq6MDU1xahRo3D48GFIJJICfzWeO3cObdu2haGhIXR0dODh4YHTp09Llckf97916xYCAgJgZGQEQ0ND9O3bFxkZGQXa37BhAxo2bAhtbW0YGxujZ8+euH//vlSZFi1aoG7duoiNjUXz5s2ho6ODn3/+GQCwe/dueHt7w9LSEpqamqhevTqmT5+O3Nxcqdfv378f9+7dE4+Lra1tke8BABw/fhzNmjWDrq4ujIyM0LlzZyQkJJSqrx/2qaiYoqKiIJFIsGnTJkyaNAlVq1aFjo4O0tLS8OLFCwQFBcHZ2Rl6enowMDBAu3btcOnSJan6i/tsPXjwAD4+PtDT04OJiQmCgoKkjhdQcM6OPH198+YNRowYgSpVqkBfXx+dOnXCgwcPZJoHlN/3LVu2YObMmahWrRq0tLTQunVr3Lp1q0D5rVu3ip+fKlWq4IcffsCDBw+kyhT3nZJIJAgMDMTWrVvh5OQEbW1tuLm5IT4+HgCwatUq1KhRA1paWmjRosVHz8Jt27YNEokEJ0+eLLBv1apVkEgkuHLlitQxfV9OTg6mT5+O6tWrQ1NTE7a2tvj555+RlZVVbLvZ2dmYMmUKGjZsCENDQ+jq6qJZs2Y4ceKEWObu3bswMTEBAEydOlX83OW/J6WJx9bWFh06dMDff/+Nxo0bQ0tLC/b29li/fr1Uubdv32Lq1KlwcHCAlpYWKleujG+++QZHjx4ttn+kWExpqUykpqbi2bNnUtuqVKlSZPmcnBx4eXnhm2++wbx586Cjo4Ps7Gx4eXkhKysLw4cPh7m5OR48eIB9+/YhJSUFhoaG+OOPPzBgwAA0btwYgwYNAgBUr169yHZev36NVq1aITk5GT/99BPMzc0REREh9R9kvuPHj6Ndu3Zo2LAhgoODoaKigrCwMLRq1Qp//fUXGjduLFW+e/fusLOzQ2hoKC5evIg1a9bA1NQUc+bMEcvMnDkTkydPRvfu3TFgwAA8ffoUS5YsQfPmzfHPP//AyMhILPv8+XO0a9cOPXv2xA8//AAzMzMA74YK9fT0MHr0aOjp6eH48eOYMmUK0tLS8L///Q8A8MsvvyA1NRX//fcfFi5cCODdcF9Rjh07hnbt2sHe3h4hISF48+YNlixZAnd3d1y8eFFMSuTp64dkiWn69OnQ0NBAUFAQsrKyoKGhgWvXrmHXrl3o1q0b7Ozs8PjxY6xatQoeHh64du0aLC0ti2wTAHJzc+Hl5YUmTZpg3rx5OHbsGObPn4/q1atjyJAhxb5W1r4GBARgy5Yt6N27N77++mucPHkS3t7eH637fbNnz4aKigqCgoKQmpqKuXPnws/PD+fOnRPL5A8TN2rUCKGhoXj8+DEWL16M06dPF/j8FPadyvfXX39hz549GDZsGAAgNDQUHTp0wLhx47B8+XIMHToUL1++xNy5c9GvXz8cP368yLi9vb2hp6eHLVu2wMPDQ2rf5s2bUadOHdStW7fI1w8YMADr1q3Dd999hzFjxuDcuXMIDQ1FQkICdu7cWeTr0tLSsGbNGvTq1QsDBw7Eq1evsHbtWnh5eeH8+fNwcXGBiYkJVqxYgSFDhqBLly7o2rUrAKBevXoKiefWrVv47rvv0L9/f/j7++P3339HQEAAGjZsiDp16gB4l1CFhoaK/0+lpaXhwoULuHjxItq0aVNkHKRgApEChYWFCQAKfQiCICQmJgoAhLCwMPE1/v7+AgBhwoQJUnX9888/AgBh69atxbapq6sr+Pv7yxTf/PnzBQDCrl27xG1v3rwRateuLQAQTpw4IQiCIOTl5QkODg6Cl5eXkJeXJ5bNyMgQ7OzshDZt2ojbgoODBQBCv379pNrq0qWLULlyZfH53bt3BVVVVWHmzJlS5eLj4wU1NTWp7R4eHgIAYeXKlQX6kJGRUWDbjz/+KOjo6AiZmZniNm9vb8HGxqZA2cLeAxcXF8HU1FR4/vy5uO3SpUuCioqK0KdPH7n7WpSiYjpx4oQAQLC3ty/Qv8zMTCE3N7dAHzQ1NYVp06YV26/8z9b75QRBEBo0aCA0bNhQahsAITg4WO6+xsbGCgCEkSNHSpULCAgoUGdh8vvu6OgoZGVlidsXL14sABDi4+MFQRCE7OxswdTUVKhbt67w5s0bsdy+ffsEAMKUKVMK9PvD71R+PzU1NYXExERx26pVqwQAgrm5uZCWliZunzhxogBAqmxhevXqJZiamgo5OTnituTkZEFFRUXq2Ocf03xxcXECAGHAgAFS9QUFBQkAhOPHj4vbPDw8BA8PD/F5Tk6O1PESBEF4+fKlYGZmJvWePX36tMj3oTTx2NjYCACEU6dOiduePHkiaGpqCmPGjBG31a9fX/D29i7QNn1aHMaiMrFs2TIcPXpU6vExH/6VbWhoCAA4fPjwR4dIZHXo0CFUrVoVnTp1ErdpaWlh4MCBUuXi4uJw8+ZNfP/993j+/DmePXuGZ8+e4fXr12jdujVOnTqFvLw8qdcMHjxY6nmzZs3w/PlzpKWlAQB27NiBvLw8dO/eXazv2bNnMDc3h4ODQ4GzS5qamujbt2+BPmhra4v/fvXqFZ49e4ZmzZohIyMD169fl/uYJCcnIy4uDgEBATA2Nha316tXD23atMGBAwcKvOZjfS0pf39/qf4B745D/ryd3NxcPH/+HHp6eqhVqxYuXrwoU72FxXvnzp0Sv/b9vh46dAgAMHToUKlyw4cPl6n+fH379oWGhoZUOwDEOC9cuIAnT55g6NChUvOLvL29Ubt2bezfv79AnUWduWrdurXU2bomTZoAAHx9faGvr19g+8eOVY8ePfDkyROpYeBt27YhLy8PPXr0KPJ1+Z+t0aNHS20fM2YMABTap3yqqqri8crLy8OLFy+Qk5MDV1dXmT8XpY3HyclJfJ8AwMTEBLVq1ZI6XkZGRrh69Spu3rxZophIMTiMRWWicePGRU5QLoyamhqqVasmtc3Ozg6jR4/GggULsHHjRjRr1gydOnXCDz/8ICZC8rp37x6qV69eYJy+Ro0aUs/z/2Py9/cvsq7U1FRUqlRJfG5tbS21P3/fy5cvYWBggJs3b0IQBDg4OBRan7q6utTzqlWrSv345bt69SomTZqE48ePF0gu8ucyyePevXsAgFq1ahXY5+joiMOHDxeY3PqxvpbUh1fwAe9+yBYvXozly5cjMTFRaq5N5cqVP1qnlpaWOG/j/XhfvnwpU0wf6+u9e/egoqJSIPYPP1OlaQco/n2qXbs2/v77b6lthX2nimor//tkZWVV6PaPHav8eW2bN29G69atAbwbwnJxcUHNmjWLfF3+sfvwWJmbm8PIyEjsc1HWrVuH+fPn4/r163j79q24vbDPkSzkjefD4wgU/GxNmzYNnTt3Rs2aNVG3bl20bdsWvXv3LnYojRSPyQ5VCO//9f6++fPnIyAgALt378aRI0cwYsQIhIaG4uzZs0X+R64I+Wdt/ve//8HFxaXQMh/ONynqcmfh/6/ukJeXB4lEgoMHDxZa9sP6PjzDAQApKSnw8PCAgYEBpk2bhurVq0NLSwsXL17E+PHjC5xtKisf62tJFdbnWbNmYfLkyejXrx+mT58OY2NjqKioYOTIkTL1t7SX15dVX8u6naK+U8W1VdIYNDU14ePjg507d2L58uV4/PgxTp8+jVmzZskUa0kW9tuwYQMCAgLg4+ODsWPHwtTUFKqqqggNDcXt27flrq8k8chyvJo3b47bt2+L/4etWbMGCxcuxMqVKzFgwIBSxUmyY7JDFZ6zszOcnZ0xadIknDlzBu7u7li5ciVmzJgBQL7/KG1sbHDt2jUIgiD1ug+vesmf5GxgYABPT08F9OJdnYIgwM7Orti/dosTFRWF58+fY8eOHWjevLm4PTExsUBZWY+LjY0NAODGjRsF9l2/fh1VqlRR2CXLJflR27ZtG1q2bIm1a9dKbU9JSSl20vunYmNjg7y8PCQmJkqdtSvsSqrStgO8e59atWolte/GjRvi/vLSo0cPrFu3DpGRkUhISIAgCMUOYQH/d+xu3rwJR0dHcfvjx4+RkpJSbJ+2bdsGe3t77NixQ+pzFRwcLFVO3v8fShpPcYyNjdG3b1/07dsX6enpaN68OUJCQpjsfEKcs0MVVlpaGnJycqS2OTs7Q0VFReoyUF1dXaSkpMhUp5eXFx48eIA9e/aI2zIzM7F69Wqpcg0bNkT16tUxb948pKenF6jn6dOncvTkna5du0JVVRVTp04t8JeyIAh4/vz5R+vI/0vy/ddnZ2dj+fLlBcrq6urKNKxlYWEBFxcXrFu3Tuo4XrlyBUeOHEH79u0/WoesZI3pfaqqqgWO19atWwtcbl1evLy8AKDAe7BkyRKFtuPq6gpTU1OsXLlS6vN/8OBBJCQkyH31l6J5enrC2NgYmzdvxubNm9G4ceOPDiflf7Y+XFV7wYIFAFBsnwr7Lpw7dw7R0dFS5fKvQpPl/4jSxFOUD7/Xenp6qFGjxkcvrSfF4pkdqrCOHz+OwMBAdOvWDTVr1kROTg7++OMPqKqqwtfXVyzXsGFDHDt2DAsWLIClpSXs7OzEiZUf+vHHH7F06VL06tULP/30EywsLLBx40Zxwmf+X4EqKipYs2YN2rVrhzp16qBv376oWrUqHjx4gBMnTsDAwAB79+6Vqz/Vq1fHjBkzMHHiRNy9exc+Pj7Q19dHYmIidu7ciUGDBiEoKKjYOpo2bYpKlSrB398fI0aMgEQiwR9//FHoMEPDhg2xefNmjB49Go0aNYKenh46duxYaL3/+9//0K5dO7i5uaF///7ipeeGhoYKvV+UPDHl69ChA6ZNm4a+ffuiadOmiI+Px8aNG2Fvb6+wuEqjYcOG8PX1xaJFi/D8+XPx0vN///0XQMnOZhVGXV0dc+bMQd++feHh4YFevXqJl57b2tpi1KhRCmmnNPF17doVmzZtwuvXrzFv3ryPvqZ+/frw9/fHb7/9Jg7Rnj9/HuvWrYOPjw9atmxZ5Gs7dOiAHTt2oEuXLvD29kZiYiJWrlwJJycnqT9QtLW14eTkhM2bN6NmzZowNjZG3bp1C70cvjTxFMXJyQktWrRAw4YNYWxsjAsXLmDbtm0IDAyUuy4qOSY7VGHVr18fXl5e2Lt3Lx48eAAdHR3Ur18fBw8exNdffy2WW7BgAQYNGoRJkybhzZs38Pf3LzLZyV+XZvjw4Vi8eDH09PTQp08fNG3aFL6+vlJXubRo0QLR0dGYPn06li5divT0dJibm6NJkyb48ccfS9SnCRMmoGbNmli4cCGmTp0K4N2k0G+//VbqCrGiVK5cGfv27cOYMWMwadIkVKpUCT/88ANat24tnmHIN3ToUMTFxSEsLAwLFy6EjY1NkYmFp6cnDh06hODgYEyZMgXq6urw8PDAnDlzSjzZszDyxJTv559/xuvXrxEREYHNmzfjq6++wv79+8vsXmslsX79epibm+PPP//Ezp074enpic2bN6NWrVoKXZk5ICAAOjo6mD17NsaPHw9dXV106dIFc+bMkVpjp7z06NEDa9asgUQiQffu3WV6zZo1a2Bvb4/w8HDs3LkT5ubmmDhxYoHhqA8FBATg0aNHWLVqFQ4fPgwnJyds2LABW7duLbA46Jo1azB8+HCMGjUK2dnZCA4OLnLtn5LGU5QRI0Zgz549OHLkCLKysmBjY4MZM2Zg7NixJaqPSob3xiLCu9PWo0aNwn///YeqVauWdzikBOLi4tCgQQNs2LBBXL2YiMoH5+zQF+fNmzdSzzMzM7Fq1So4ODgw0aES+fAzBbxLoFVUVKQmkhNR+eAwFn1xunbtCmtra7i4uCA1NRUbNmzA9evXsXHjxvIOjT5Tc+fORWxsLFq2bAk1NTUcPHgQBw8exKBBgwqsXUNEnx6HseiLs2jRIqxZswZ3795Fbm4unJycMG7cuI9eJktUlKNHj2Lq1Km4du0a0tPTYW1tjd69e+OXX37hXbWJKgAmO0RERKTUOGeHiIiIlBqTHSIiIlJqHEzGu3sWPXz4EPr6+gpbAIyIiIjKliAIePXqFSwtLYu8FxzAZAcA8PDhQ14xQURE9Jm6f/9+sTeHZrIDQF9fH8C7g2VgYFDO0RAREZEs0tLSYGVlJf6OF4XJDv7v3jUGBgZMdoiIiD4zH5uCwgnKREREpNSY7BAREZFSY7JDRERESo1zdmSUl5eH7Ozs8g6DiBRIXV0dqqqq5R0GEZUxJjsyyM7ORmJiIvLy8so7FCJSMCMjI5ibm3ONLSIlxmTnIwRBQHJyMlRVVWFlZVXsokVE9PkQBAEZGRl48uQJAMDCwqKcIyKissJk5yNycnKQkZEBS0tL6OjolHc4RKRA2traAIAnT57A1NSUQ1pESoqnKT4iNzcXAKChoVHOkRBRWcj/I+bt27flHAkRlRUmOzLieD6RcuJ3m0j5MdkhIiIipcZkhz4bAQEB8PHxEZ+3aNECI0eOFJ/b2tpi0aJFMtUlT1kiIvq8cYJyCdlO2P9J27s721thdQUEBCAlJQW7du1SWJ0VQUxMDHR1dRVeloiIPm9MdkhpmJiYlElZIiL6vHEYS4lt27YNzs7O0NbWRuXKleHp6YmxY8di3bp12L17NyQSCSQSCaKiogAA9+/fR/fu3WFkZARjY2N07twZd+/eFevLH0aaOnUqTExMYGBggMGDB8u8snReXh7mzp2LGjVqQFNTE9bW1pg5c6a4Pz4+Hq1atRLjHTRoENLT02Xu7/tDU4IgICQkBNbW1tDU1ISlpSVGjBhRaFkASEpKQufOnaGnpwcDAwN0794djx8/FveHhITAxcUFf/zxB2xtbWFoaIiePXvi1atXxR7v169fyxw/ERGVDSY7Sio5ORm9evVCv379kJCQgKioKHTt2hXBwcHo3r072rZti+TkZCQnJ6Np06Z4+/YtvLy8oK+vj7/++gunT5+Gnp4e2rZtK5XMREZGivX9+eef2LFjB6ZOnSpTTBMnTsTs2bMxefJkXLt2DRERETAzMwMAvH79Gl5eXqhUqRJiYmKwdetWHDt2DIGBgSXq//bt27Fw4UKsWrUKN2/exK5du+Ds7Fxo2by8PHTu3BkvXrzAyZMncfToUdy5cwc9evSQKnf79m3s2rUL+/btw759+3Dy5EnMnj0bQNHHWxCEEsVPRESKw2EsJZWcnIycnBx07doVNjY2ACD+2GtrayMrKwvm5uZi+Q0bNiAvLw9r1qwRL8UNCwuDkZERoqKi8O233wJ4t97Q77//Dh0dHdSpUwfTpk3D2LFjMX369GJXl3716hUWL16MpUuXwt/fHwBQvXp1fPPNNwCAiIgIZGZmYv369eJcmqVLl6Jjx46YM2eOmBTJKikpCebm5vD09IS6ujqsra3RuHHjQstGRkYiPj4eiYmJsLKyAgCsX78ederUQUxMDBo1agTgXVIUHh4OfX19AEDv3r0RGRmJmTNnFnu8iYiUwbLBx8u8jWErW5VJvTyzo6Tq16+P1q1bw9nZGd26dcPq1avx8uXLIstfunQJt27dgr6+PvT09KCnpwdjY2NkZmbi9u3bUvW+v5K0m5sb0tPTcf/+/WLjSUhIQFZWFlq3bl3k/vr160tNGnZ3d0deXh5u3Lgha7dF3bp1w5s3b2Bvb4+BAwdi586dyMnJKbJtKysrMdEBACcnJxgZGSEhIUHcZmtrKyY6wLvbC+TfakDe401ERJ8Okx0lpaqqiqNHj+LgwYNwcnLCkiVLUKtWLSQmJhZaPj09HQ0bNkRcXJzU499//8X3339f6njyl+X/VKysrHDjxg0sX74c2traGDp0KJo3b16qVXLV1dWlnkskEvHmsPIebyIi+nQ4jKXEJBIJ3N3d4e7ujilTpsDGxgY7d+6EhoaGeBuMfF999RU2b94MU1NTGBgYFFnnpUuX8ObNGzF5OXv2LPT09KTOihTGwcEB2traiIyMxIABAwrsd3R0RHh4OF6/fi2e3Tl9+jRUVFRQq1YtebsO4F2C1bFjR3Ts2BHDhg1D7dq1ER8fj6+++qpA2/fv38f9+/fFfly7dg0pKSlwcnKSub2ijvfo0aNLFD8Rfd5DJ1Rx8MyOkjp37hxmzZqFCxcuICkpCTt27MDTp0/h6OgIW1tbXL58GTdu3MCzZ8/w9u1b+Pn5oUqVKujcuTP++usvJCYmIioqCiNGjMB///0n1pudnY3+/fvj2rVrOHDgAIKDgxEYGPjRu8FraWlh/PjxGDduHNavX4/bt2/j7NmzWLt2LQDAz88PWlpa8Pf3x5UrV3DixAkMHz4cvXv3lnu+DgCEh4dj7dq1uHLlCu7cuYMNGzZAW1tbnE/zPk9PTzg7O8PPzw8XL17E+fPn0adPH3h4eMDV1VWm9oo73kREVL54ZkdJGRgY4NSpU1i0aBHS0tJgY2OD+fPno127dnB1dUVUVBRcXV2Rnp6OEydOoEWLFjh16hTGjx+Prl274tWrV6hatSpat24tdaandevWcHBwQPPmzZGVlYVevXohJCREppgmT54MNTU1TJkyBQ8fPoSFhQUGDx4M4N3NGA8fPoyffvoJjRo1go6ODnx9fbFgwYIS9d/IyAizZ8/G6NGjkZubC2dnZ+zduxeVK1cuUFYikWD37t0YPnw4mjdvDhUVFbRt2xZLliyRub3ijjcREZUvicBrY5GWlgZDQ0OkpqYWGMLJzMxEYmIi7OzsoKWlVU4RVgzKuvIyfdn4Ha/YOIxVcVTE96K43+/3cRiLiIiIlBqTHVKIpKQk8ZL1wh5JSUnlHSIREX2hOGeHZBYeHl7kPktLS8TFxRW7n4iIqDww2SGFUFNTQ40aNco7DCIiogI4jEVERERKjckOERERKTUmO0RERKTUmOwQERGRUmOyQ0REREqNyQ5RIaKioiCRSJCSklLeoRARUSnx0vOSCjH8xO2lftrmQkKwa9euYtfO+dRatGgBFxcXLFq0qLxDISKizwjP7JBSEQQBOTk55R0GERFVIEx2lFSLFi0wYsQIjBs3DsbGxjA3N5e6O3lSUhI6d+4MPT09GBgYoHv37nj8+DGAdyslT506FZcuXYJEIoFEIil29WRZ6gTenS1ycXHBH3/8AVtbWxgaGqJnz5549erVR+sOCAjAyZMnsXjxYjGmu3fvisNNBw8eRMOGDaGpqYm///4bt2/fRufOnWFmZgY9PT00atQIx44dk6ozKysL48ePh5WVFTQ1NVGjRg2sXbu20PYzMjLQrl07uLu7c2iLiOgzw2RHia1btw66uro4d+4c5s6di2nTpuHo0aPIy8tD586d8eLFC5w8eRJHjx7FnTt30KNHDwBAjx49MGbMGNSpUwfJyclITk4W9xXlY3Xmu337Nnbt2oV9+/Zh3759OHnyJGbPnv3RvixevBhubm4YOHCgGJOVlZW4f8KECZg9ezYSEhJQr149pKeno3379oiMjMQ///yDtm3bomPHjlL36OrTpw/+/PNP/Prrr0hISMCqVaugp6dXoO2UlBS0adMGeXl5OHr0KIyMjD4aLxERVRycs6PE6tWrh+DgYACAg4MDli5disjISABAfHw8EhMTxYRh/fr1qFOnDmJiYtCoUSPo6elBTU0N5ubmMrUVGRn50TqBd0lReHg49PX1AQC9e/dGZGQkZs6cWWz9hoaG0NDQgI6OTqExTZs2DW3atBGfGxsbo379+uLz6dOnY+fOndizZw8CAwPx77//YsuWLTh69Cg8PT0BAPb29gXqffToEXr06AEHBwdERERAQ0NDpuNBREQVB8/sKLF69epJPbewsMCTJ0+QkJAAKysrqTMjTk5OMDIyQkJCQonakrVOW1tbMdF5P6bScnV1lXqenp6OoKAgODo6wsjICHp6ekhISBDP7MTFxUFVVRUeHh7F1tumTRvUqFEDmzdvZqJDRPSZYrKjxNTV1aWeSyQS5OXllVM075RVTLq6ulLPg4KCsHPnTsyaNQt//fUX4uLi4OzsjOzsbACAtra2TPV6e3vj1KlTuHbtWqljJCKi8sFk5wvk6OiI+/fv4/79++K2a9euISUlBU5OTgAADQ0N5ObmKrTO0pInptOnTyMgIABdunSBs7MzzM3NcffuXXG/s7Mz8vLycPLkyWLrmT17Nvz9/dG6dWsmPEREnykmO18gT09PODs7w8/PDxcvXsT58+fRp08feHh4iMNBtra2SExMRFxcHJ49e4asrKxS11latra2OHfuHO7evYtnz54Ve0bIwcEBO3bsQFxcHC5duoTvv/9eqrytrS38/f3Rr18/7Nq1C4mJiYiKisKWLVsK1DVv3jz4+fmhVatWuH79ukL6QkREnw6TnS+QRCLB7t27UalSJTRv3hyenp6wt7fH5s2bxTK+vr5o27YtWrZsCRMTE/z555+lrrO0goKCoKqqCicnJ5iYmEhdWfWhBQsWoFKlSmjatCk6duwILy8vfPXVV1JlVqxYge+++w5Dhw5F7dq1MXDgQLx+/brQ+hYuXIju3bujVatW+PfffxXWJyIiKnsSQRCE8g6ivKWlpcHQ0BCpqakwMDCQ2peZmYnExETY2dlBS0urnCIkorLC73jFtmzw8TJvY9jKVmXehjKoiO9Fcb/f7+OZHSIiIlJqTHZIJhs3boSenl6hjzp16pS6/qSkpCLr19PTK3bIioiIqDhcVJBk0qlTJzRp0qTQfR9eTl4SlpaWxd501NLSstRtEBHRl4nJDslEX19fajFARVNTU0ONGjXKrH4iIvpycRiLiIiIlBqTHSIiIlJqTHaIiIhIqTHZISIiIqXGZIeIiIiUGpOdL1BISAhcXFxkLn/37l1IJJJiLw2PioqCRCJBSkpKqeOTpb0PtWjRAiNHjix124qqh4iIKg5eel5CzuucP2l78f7xCqsrKCgIw4cPV1h9imZlZYXk5GRUqVKlzNqIiopCy5Yt8fLlSxgZGYnbd+zYoZB1g4iIqOJgsvMFyl+VuKJSVVWFubl5ubRtbGxcLu0SEVHZ4TCWElq/fj0qV66MrKwsqe0+Pj7o3bt3ocNYa9asgaOjI7S0tFC7dm0sX7682DYOHDiAmjVrQltbGy1btsTdu3dlii0tLQ3a2to4ePCg1PadO3dCX18fGRkZhQ5jnTx5Eo0bN4ampiYsLCwwYcIE5OTkFNnOH3/8AVdXV+jr68Pc3Bzff/89njx5AuDdMFnLli0BAJUqVYJEIkFAQACAgsNYL1++RJ8+fVCpUiXo6OigXbt2uHnzprg/PDwcRkZGOHz4MBwdHaGnp4e2bdsiOTlZLBMVFYXGjRtDV1cXRkZGcHd3x7179wrE/OReWpk/iIi+REx2lFC3bt2Qm5uLPXv2iNuePHmC/fv3o1+/fgXKb9y4EVOmTMHMmTORkJCAWbNmYfLkyVi3bl2h9d+/fx9du3ZFx44dERcXhwEDBmDChAkyxWZgYIAOHTogIiKiQAw+Pj7Q0dEp8JoHDx6gffv2aNSoES5duoQVK1Zg7dq1mDFjRpHtvH37FtOnT8elS5ewa9cu3L17V0xorKyssH37dgDAjRs3kJycjMWLFxdaT0BAAC5cuIA9e/YgOjoagiCgffv2ePv2rVgmIyMD8+bNwx9//IFTp04hKSkJQUFBAICcnBz4+PjAw8MDly9fRnR0NAYNGgSJRCLT8SIiotLjMJYS0tbWxvfff4+wsDB069YNALBhwwZYW1ujRYsWOHnypFT54OBgzJ8/H127dgUA2NnZ4dq1a1i1ahX8/f0L1L9ixQpUr14d8+fPBwDUqlUL8fHxmDNnjkzx+fn5oXfv3sjIyICOjg7S0tKwf/9+7Ny5s9Dyy5cvh5WVFZYuXQqJRILatWvj4cOHGD9+PKZMmQIVlYI5+/tJnb29PX799Vc0atQI6enp0NPTE4erTE1NpebsvO/mzZvYs2cPTp8+jaZNmwJ4l5RZWVlh165d4rF9+/YtVq5cierVqwMAAgMDMW3aNADvzmSlpqaiQ4cO4n5HR0eZjhMRESkGz+woqYEDB+LIkSN48OABgHfDLQEBAQXOKLx+/Rq3b99G//79pe4yPmPGDNy+fbvQuhMSEgrcFNTNzU3m2Nq3bw91dXXxzNP27dthYGAAT0/PIttzc3OTit3d3R3p6en477//Cn1NbGwsOnbsCGtra+jr68PDwwMA5Lp7ekJCAtTU1KT6WrlyZdSqVQsJCQniNh0dHTGRAQALCwtxyMzY2BgBAQHw8vJCx44dsXjxYqkhLiIiKntMdpRUgwYNUL9+faxfvx6xsbG4evWqOIzzvvT0dADA6tWrERcXJz6uXLmCs2fPlklsGhoa+O6778ShrIiICPTo0QNqaoo50fj69Wt4eXnBwMAAGzduRExMjHjWKDs7WyFtvO/Dq7ckEgkEQRCfh4WFITo6Gk2bNsXmzZtRs2bNMju2RERUEJMdJTZgwACEh4cjLCwMnp6esLKyKlDGzMwMlpaWuHPnDmrUqCH1sLOzK7ReR0dHnD9/XmqbvD/efn5+OHToEK5evYrjx4/Dz8+vyLKOjo7ifJl8p0+fhr6+PqpVq1ag/PXr1/H8+XPMnj0bzZo1Q+3atcUzLfk0NDQAALm5ucW2m5OTg3Pnzonbnj9/jhs3bsDJyUnmvgLvks+JEyfizJkzqFu3boE5S0REVHaY7Cix77//Hv/99x9Wr15d6MTkfFOnTkVoaCh+/fVX/Pvvv4iPj0dYWBgWLFhQaPnBgwfj5s2bGDt2LG7cuIGIiAiEh4fLFVvz5s1hbm4OPz8/2NnZFRgWe9/QoUNx//59DB8+HNevX8fu3bsRHByM0aNHFzpfx9raGhoaGliyZAnu3LmDPXv2YPr06VJlbGxsIJFIsG/fPjx9+lQ8w/U+BwcHdO7cGQMHDsTff/+NS5cu4YcffkDVqlXRuXNnmfqZmJiIiRMnIjo6Gvfu3cORI0dw8+ZNztshIvqEmOwoMUNDQ/j6+kJPTw8+Pj5FlhswYADWrFmDsLAwODs7w8PDA+Hh4UWe2bG2tsb27duxa9cu1K9fHytXrsSsWbPkik0ikaBXr164dOlSsWd1AKBq1ao4cOAAzp8/j/r162Pw4MHo378/Jk2aVGh5ExMThIeHY+vWrXBycsLs2bMxb968AnVOnToVEyZMgJmZGQIDAwutKywsDA0bNkSHDh3g5uYGQRBw4MABmRce1NHRwfXr1+Hr64uaNWti0KBBGDZsGH788UeZXk9ERKUnEd4fG/jEcnNzERISgg0bNuDRo0ewtLREQEAAJk2aJE5GFQQBwcHBWL16NVJSUuDu7o4VK1bAwcFBrOfFixcYPnw49u7dCxUVFfj6+mLx4sUyL5yXlpYGQ0NDpKamwsDAQGpfZmYmEhMTYWdnBy0tLcV1/hNp3bo16tSpg19//bW8Q6GP+BTr4JjaGHy80Bfmc/+OK7tlg4+XeRvDVrYq8zaUQUV8L4r7/X5fuZ7ZmTNnDlasWIGlS5ciISEBc+bMwdy5c7FkyRKxzNy5c/Hrr79i5cqVOHfuHHR1deHl5YXMzEyxjJ+fH65evYqjR49i3759OHXqFAYNGlQeXaowXr58iZ07dyIqKgrDhg0r73CIiIjKTbmus3PmzBl07twZ3t7eAABbW1v8+eef4uRXQRCwaNEiTJo0SZwjsX79epiZmWHXrl3o2bMnEhIScOjQIcTExMDV1RUAsGTJErRv3x7z5s2DpaVl+XSunDVo0AAvX77EnDlzUKtWrU/adrt27fDXX38Vuu/nn3/Gzz///EnjISKiL1u5JjtNmzbFb7/9hn///Rc1a9bEpUuX8Pfff4sTYxMTE/Ho0SOp9VcMDQ3RpEkTREdHo2fPnoiOjoaRkZGY6ACAp6cnVFRUcO7cOXTp0qVAu1lZWVK3UkhLU75l9GW9fUNZWLNmDd68eVPoPt57ioiIPrVyTXYmTJiAtLQ01K5dG6qqqsjNzcXMmTPFCauPHj0C8O7y6PeZmZmJ+x49egRTU1Op/WpqajA2NhbLfCg0NBRTp05VdHfo/6tatWp5h0BERCQq1zk7W7ZswcaNGxEREYGLFy9i3bp1mDdvXpH3ZFKUiRMnIjU1VXzcv3+/TNsjIiKi8lOuZ3bGjh2LCRMmoGfPngAAZ2dn3Lt3D6GhofD394e5uTkA4PHjx7CwsBBf9/jxY/Gu3ebm5gUWjMvJycGLFy/E139IU1MTmpqaZdAjIiIiqmjK9cxORkZGgUXhVFVVkZeXB+DdDSnNzc0RGRkp7k9LS8O5c+fEezG5ubkhJSUFsbGxYpnjx48jLy+v2IXqiIiI6MtQrmd2OnbsiJkzZ8La2hp16tTBP//8gwULFoir/UokEowcORIzZsyAg4MD7OzsMHnyZFhaWoqL5Dk6OqJt27YYOHAgVq5cibdv3yIwMBA9e/b8Yq/EIiIiov9TrsnOkiVLMHnyZAwdOhRPnjyBpaUlfvzxR0yZMkUsM27cOLx+/RqDBg1CSkoKvvnmGxw6dEhq8a+NGzciMDAQrVu3FhcV5CJ6REREBJTzMJa+vj4WLVqEe/fu4c2bN7h9+zZmzJgh3qQReHd2Z9q0aXj06BEyMzNx7Ngx1KxZU6oeY2NjRERE4NWrV0hNTcXvv/8u8+rJRLIKCQkR54opWlRUFMxsDZGamlIm9X+Mra0tFi1aJHN5RR2LsjymRET5yvXMzucsofanvZGj4/WET9oefVliYmKgq6tbpm1IJBLs3LlT6j5tQUFBGD58eJm2S0TEZIe+CNnZ2VJnDJW9XXmZmJiUS7t6eno8C0tEZY53PVdShQ1LuLi4ICQkBMC7v7LXrFmDLl26QEdHBw4ODtizZ49YNioqChKJBJGRkXB1dYWOjg6aNm2KGzduSNW5d+9eNGrUCFpaWqhSpYrUitVZWVkICgpC1apVoauriyZNmiAqKkrcf+/ePXTs2BGVKlWCrq4u6tSpgwMHDgB4d28vPz8/mJiYQFtbGw4ODggLCxNfO378eNSsWRM6Ojqwt7fH5MmT8fbtW3F//vDImjVrxBs8rl+/HpUrV5ZaPRsAfHx80Lt3b5mP7apVq2BlZQUdHR10794dqamp4r6AgAD4+Phg5syZsLS0FG/V8ccff8DV1RX6+vowNzfH999/X2DJhPdlvMlAL39fdPD9Vhza2rBpHb5p3QjWNU3h3soVYX+slile765tMD10itS2p0+fQl1dHadOnQJQ8POSlJSEzp07Q09PDwYGBujevTseP35cZBsxMTFo06YNqlSpAkNDQ3h4eODixYvifltbWwBAly5dIJFIxOcfDmPl5eVh2rRpqFatGjQ1NeHi4oJDhw6J++/evQuJRIIdO3agZcuW0NHRQf369REdHS2WKe5zRURfJiY7X7CpU6eie/fuuHz5Mtq3bw8/Pz+8ePFCqswvv/yC+fPn48KFC1BTUxOvlAOA/fv3o0uXLmjfvj3++ecfREZGonHjxuL+wMBAREdHY9OmTbh8+TK6deuGtm3b4ubNmwCAYcOGISsrC6dOnUJ8fDzmzJkj/pU/efJkXLt2DQcPHkRCQgJWrFiBKlWqiHXr6+sjPDwc165dw+LFi7F69WosXLhQKvZbt25h+/bt2LFjB+Li4tCtWzfk5uZKJXVPnjzB/v37pfpVnFu3bmHLli3Yu3cvDh06hH/++QdDhw6VKhMZGYkbN26IN6YFgLdv32L69Om4dOkSdu3ahbt37yIgIKDQNlJTU9D9Bx/k5eVhy4ZdMDQ0wrZdWzB3wSxMHDsZf0Wex8/jpmDO/JnYvC3iozH7+nTHrn07IAiCuG3z5s2wtLREs2bNCpTPy8tD586d8eLFC5w8eRJHjx7FnTt30KNHjyLbePXqFfz9/fH333/j7NmzcHBwQPv27fHq1SsA75IhAAgLC0NycrL4/EOLFy/G/PnzMW/ePFy+fBleXl7o1KmT+JnJ98svvyAoKAhxcXGoWbMmevXqhZycHADFf66I6MvEYawvWEBAAHr16gUAmDVrFn799VecP38ebdu2FcvMnDkTHh4eAN7d3sPb2xuZmZnQ0tLCzJkz0bNnT6lbb9SvXx/AuzMDYWFhSEpKEpcACAoKwqFDhxAWFoZZs2YhKSkJvr6+cHZ2BgDY29uL9SQlJaFBgwbiPc/yzwTkmzRpkvhvW1tbBAUFYdOmTRg3bpy4PTs7G+vXr5caovn+++8RFhaGbt26AQA2bNgAa2trtGjRQqZjlpmZifXr14u3xFiyZAm8vb0xf/58cRFLXV1drFmzRmr46v1kyt7eHr/++isaNWqE9PR0qR/iJ0+fYFBgX9jb2WPF4rViHf9bOAshv8yEd9tOAAAbK1vcuHkD6yPC0OO774uNuZN3F0yeNgHnYqLRyfbdexsREYFevXpBIpEUKB8ZGYn4+HgkJibCysoKwLsb8NapUwcxMTFo1KhRgde0atVK6vlvv/0GIyMjnDx5Eh06dBDfAyMjoyIX+wSAefPmYfz48eJCo3PmzMGJEyewaNEiLFu2TCwXFBQk3kB46tSpqFOnDm7duoXatWsX+7kioi8Tz+x8werVqyf+W1dXFwYGBgWGVt4vk7+KdX6ZuLg4tG7dutC64+PjkZubi5o1a4rzMvT09HDy5Encvn0bADBixAjMmDED7u7uCA4OxuXLl8XXDxkyBJs2bYKLiwvGjRuHM2fOSNW/efNmuLu7w9zcHHp6epg0aRKSkpKkytjY2BSYizJw4EAcOXIEDx48AACEh4cjICCg0B/9wlhbW0vd+8vNzQ15eXlSw3vOzs4F5unExsaiY8eOsLa2hr6+vphAfhhz994+sLO1w29Lw8U6Xme8xt17iRg9PhB2TpbiY9GS/+FuUuJHY65SuQpaNGuF7bu2AHh3g93o6GjxHnQfSkhIgJWVlZjoAICTkxOMjIyQkFD4RPnHjx9j4MCBcHBwgKGhIQwMDJCenl6gf8VJS0vDw4cP4e7uLrXd3d29QLvFfS6L+1wR0ZeJyY6SUlFRkRq2ACA1pwUA1NXVpZ5LJBJx9erCyuQnBPlltLW1i2w/PT0dqqqqiI2NRVxcnPhISEjA4sWLAQADBgzAnTt30Lt3b8THx8PV1RVLliwBALRr1w737t3DqFGj8PDhQ7Ru3RpBQUEAIP5Qt2/fHvv27cM///yDX375BdnZ2VIxFHZ1UYMGDVC/fn2sX78esbGxuHr1apHDSSX1YbuvX7+Gl5cXDAwMsHHjRsTExGDnzp0AUCBmz5bf4uz5M7hx87rU6wFg3uxfcfzAX+Lj5JFoHNh5TKaYuvp0x76Du/H27VtERETA2dlZPPOhCP7+/oiLi8PixYtx5swZxMXFoXLlygX6pyjFfS6L+1wR0ZeJyY6SMjExQXJysvg8LS0NiYkfPwsgj3r16kndyuN9DRo0QG5uLp48eYIaNWpIPd4fxrCyssLgwYOxY8cOjBkzBqtX/9+kWxMTE/j7+2PDhg1YtGgRfvvtNwDAmTNnYGNjg19++QWurq5wcHDAvXv3ZI57wIABCA8PR1hYGDw9PaXOYHxMUlISHj58KD4/e/YsVFRUxInIhbl+/TqeP3+O2bNno1mzZqhdu3aRk5MnjQ9Bd99e+M6vk5jwmJqYwtzMAklJd2FnW13qYWNlK1Pcbdu0R2ZWFg4dOoSIiIgiz+oA71Ylv3//vtQNcq9du4aUlBQ4OTkV+prTp09jxIgRaN++PerUqQNNTU08e/ZMqoy6ujpyc3OLbNfAwACWlpY4ffp0gbqLarcoxX2uiOjLwzk7SqpVq1YIDw9Hx44dYWRkhClTpkBVVVWhbQQHB6N169aoXr06evbsiZycHBw4cEC8UsrPzw99+vTB/Pnz0aBBAzx9+hSRkZGoV68evL29MXLkSLRr1w41a9bEy5cvceLECTg6vlu/aMqUKWjYsCHq1KmDrKws7Nu3T9zn4OCApKQkbNq0CY0aNcL+/fvFMyWy+P777xEUFITVq1dj/fr1cvVZS0sL/v7+mDdvHtLS0jBixAh079692Hko1tbW0NDQwJIlSzB48GBcuXIF06dPL7J8yC8zkZubB99eHbFz03441KiJsaMmYlLIeOjrG6CVhyeysrNw6fI/SE1LweABgR+NW1dHF+2+9cbkyZORkJAgztUqjKenJ5ydneHn54dFixYhJycHQ4cOhYeHhziH6kMODg7iFWdpaWkYO3ZsgTN/tra2iIyMhLu7OzQ1NVGpUqUC9YwdOxbBwcGoXr06XFxcEBYWhri4OGzcuPGjfcxX3OeKiL5MPLOjpCZOnAgPDw906NAB3t7e8PHxQfXq1RXaRosWLbB161bs2bMHLi4uaNWqFc6fPy/uDwsLQ58+fTBmzBjUqlULPj4+iImJgbW1NQAgNzcXw4YNE+9vVrNmTSxfvhwAoKGhgYkTJ6JevXpo3rw5VFVVsWnTJgBAp06dMGrUKAQGBsLFxQVnzpzB5MmTZY7b0NAQvr6+0NPTk1rgThY1atRA165d0b59e3z77beoV6+eGHNRTExMEB4ejq1bt8LJyQmzZ8/GvHnzin3N9Cmh6NShC3y/74jbd27hh57+mD9nCTZt3YAWbd3QpUd7bN4WAetqNjLH7tu5Oy5duoRmzZqJ70FhJBIJdu/ejUqVKqF58+bw9PSEvb09Nm/eXORr1q5di5cvX+Krr75C7969MWLECJiamkqVmT9/Po4ePQorKys0aNCg0HpGjBiB0aNHY8yYMXB2dsahQ4ewZ88eODg4yNzP4j5XRPRlkggfTuz4AqWlpcHQ0BCpqakwMDCQ2peZmYnExERxrRZSDq1bt0adOnUq1D3UntxLK/M2TG0MPl7oC8PveMW2bPDxMm9j2MpWHy9EFfK9KO73+30cxqIvysuXLxEVFYWoqCj+tU9E9IVgskNflAYNGuDly5eYM2dOgUnFderUKXKi86pVq4qd1FueFi2bh8XLFhS67+tGbvhz3fZPHBERUcXCZIe+KHfv3i1y34EDBwpcnp/PzMysjCIqPX+/fujs3aXQfVpaRS8PQET0pWCyQ/T/2djIPtm3IqlkZIxKRsblHQYRUYXFq7GIiIhIqTHZISIiIqXGZIeIiIiUGpMdIiIiUmpMdoiIiEipMdn5gp0+fRrOzs5QV1cXb5vw4baoqChIJBKkpKTIVGeLFi0wcuTIMov5cyTvMfwUQkJC4OLiInP5u3fvQiKRIC4urlTtKqoeIiJ58NLzEvoUy2a/ryyWMx89ejRcXFxw8OBB6OnpFbpNR0cHycnJMDQ0lKnOHTt2QF1dXaFxBgQEICUlBbt27VJovZ9K06ZN5TqGn0JQUBCGDx9epm0U9r5ZWVkhOTkZVapUKdO2iYjexzM7X7Dbt2+jVatWqFatGoyMjArdpqGhAXNzc0gkEpnqNDY2hr6+fhlGrXjZ2dllWr+8x/BT0NPTQ+XKlT95u6qqqjA3N4eaGv/OIqJPh8mOEsvLy0NoaCjs7Oygra2N+vXrY9u2beJQwvPnz9GvXz9IJBKEh4cXuq2wIZjTp0+jRYsW0NHRQaVKleDl5YWXL18CKDiMlZWVhaCgIFStWhW6urpo0qQJoqKixP3h4eEwMjLC4cOH4ejoCD09PbRt2xbJyckA3g23rFu3Drt374ZEIoFEIpF6fVH+++8/9OrVC8bGxtDV1YWrqyvOnTsn1uni4oI1a9ZI3fwxKSkJnTt3hp6eHgwMDNC9e3c8fvxYrPPSpUto2bIl9PX1YWBggIYNG+LChQsAgHv37qFjx46oVKkSdHV1UadOHRw4cABAwWGsovr8+Mkjsa2cnBz8HDIODs7WqO1ii+mhUzB89GD4D/z+o31fHxGGeo1rIS8vT2p7nwG90K9fP6ljkC8vLw/Tpk1DtWrVoKmpCRcXFxw6dKjINnJzc9G/f3/xs1WrVi0sXrxY3F/U+1bYMNbJkyfRuHFjaGpqwsLCAhMmTEBOTo64v0WLFhgxYgTGjRsHY2NjmJubIyQkRNwvCAJCQkJgbW0NTU1NWFpaYsSIER89TkT05WCyo8RCQ0Oxfv16rFy5ElevXsWoUaPwww8/4N69e0hOToaBgQEWLVqE5ORkdOvWrcC2Hj16FKgzLi4OrVu3hpOTE6Kjo/H333+jY8eOyM3NLTSGwMBAREdHY9OmTbh8+TK6deuGtm3b4ubNm2KZjIwMzJs3D3/88QdOnTqFpKQkBAUFAXg33NK9e3cxAUpOTkbTpk2L7Xd6ejo8PDzw4MED7NmzB5cuXcK4ceOkfvxv3bqF7du3Y8eOHYiLi0NeXh46d+6MFy9e4OTJkzh69Cju3LkjdQz8/PxQrVo1xMTEIDY2FhMmTBCH7IYNG4asrCycOnUK8fHxmDNnjjg0WJjC+hwyc5K4f8nKhdixawsW/28Z9m47glfpr3Dw6P5i+52vk7cPXqa8wN/Rp8RtL1Ne4MSpY0Xe32vx4sWYP38+5s2bh8uXL8PLywudOnWSep/el5eXh2rVqmHr1q24du0apkyZgp9//hlbtmwBIPv79uDBA7Rv3x6NGjXCpUuXsGLFCqxduxYzZsyQKrdu3Tro6uri3LlzmDt3LqZNm4ajR48CALZv346FCxdi1apVuHnzJnbt2gVnZ2eZjhURfRl4LllJZWVlYdasWTh27Bjc3NwAAPb29vj777+xatUqREREQCKRwNDQEObm5gAAXV3dAts+NHfuXLi6ukrdMbxOnTqFlk1KSkJYWBiSkpJgaWkJ4N2P4KFDhxAWFoZZs2YBAN6+fYuVK1eievXqAN4lSNOmTQPwbrhFW1sbWVlZRcb0oYiICDx9+hQxMTEwNn53G4UaNWpIlcnOzsb69ethYmICADh69Cji4+ORmJgIKysrAMD69etRp04dxMTEoFGjRkhKSsLYsWNRu3ZtAICDg4NUX319fcUfWXt7+2JjLKzPIcFTxf1rw3/DiKGj0b5tRwBA6LR5iIw6KlP/jQwroZVHG+zYvRXN3VsAAPYe2A3jSpXRsmXLQl8zb948jB8/Hj179gQAzJkzBydOnMCiRYuwbNmyAuXV1dUxder/xWtnZ4fo6Ghs2bIF3bt3l/l9W758OaysrLB06VJIJBLUrl0bDx8+xPjx4zFlyhSoqLz7e6xevXoIDg4G8O64L126FJGRkWjTpg2SkpJgbm4OT09PqKurw9raGo0bN5bpWBHRl4FndpTUrVu3kJGRgTZt2kBPT098rF+/Hrdv3y5xvflndmQRHx+P3Nxc1KxZUyqGkydPSsWgo6Mj/ugDgIWFBZ48eVKqGBs0aCAmOoWxsbEREx0ASEhIgJWVlZjoAICTkxOMjIyQkJAA4N3k7QEDBsDT0xOzZ8+W6sOIESMwY8YMuLu7Izg4GJcvXy42xsL6/Oz5UwBAWloqnj57ggb1G4r7VVVVUa9ufRmPAODr0w37D+5FVlYWAGDHrq3w6egrJg/vS0tLw8OHD+Hu7i613d3dXex7YZYtW4aGDRvCxMQEenp6+O2335CUlCRzjMC74+7m5iY1n8nd3R3p6en477//xG316tWTet37n5Fu3brhzZs3sLe3x8CBA7Fz506pYTAiIiY7Sio9PR0AsH//fsTFxYmPa9euYdu2bSWuV1tb9rtop6enQ1VVFbGxsVIxJCQkSM3v+PDqLYlEAkEQyjRGXV1duesNCQnB1atX4e3tjePHj8PJyQk7d+4EAAwYMAB37txB7969ER8fD1dXVyxZsqTIuhTd5w9927odBAg4duIwHjz8D2djzsDXp5vC6t+0aROCgoLQv39/HDlyBHFxcejbt2+ZTfYu7HjlD0taWVnhxo0bWL58ObS1tTF06FA0b968yDvYE9GXh8mOknJycoKmpiaSkpJQo0YNqcf7Zy/kVa9ePURGRspUtkGDBsjNzcWTJ08KxCDrkBTw7mqmouYEFRVjXFwcXrx4IfNrHB0dcf/+fdy/f1/cdu3aNaSkpMDJyUncVrNmTYwaNQpHjhxB165dERYWJu6zsrLC4MGDsWPHDowZMwarV6+Wuf33GRgYwqSKKeIuXxS35ebmIv5K8WeL3qelpQVvr47YvmsLdu7Zhhr2DqhX16WI9gxgaWmJ06dPS20/ffq0VN8/3Ne0aVMMHToUDRo0QI0aNQqcMZTlfXN0dER0dLRUonf69Gno6+ujWrVqMvT0HW1tbXTs2BG//voroqKiEB0djfj4eJlfT0TKjcmOktLX10dQUBBGjRqFdevW4fbt27h48SKWLFmCdevWlbjeiRMnIiYmBkOHDsXly5dx/fp1rFixAs+ePStQtmbNmvDz80OfPn2wY8cOJCYm4vz58wgNDcX+/bJNtgUAW1tbXL58GTdu3MCzZ88++hd7r169YG5uDh8fH5w+fRp37tzB9u3bER0dXeRrPD094ezsDD8/P1y8eBHnz59Hnz594OHhAVdXV7x58waBgYGIiorCvXv3cPr0acTExMDR0REAMHLkSBw+fBiJiYm4ePEiTpw4Ie4rif4Bg/Dr8gU4eGQ/bt2+iUlTxyMlLUWuy9e7+nTDseNH8OeWDejq073YsmPHjsWcOXOwefNm3LhxAxMmTEBcXBx++umnQss7ODjgwoULOHz4MP79919MnjwZMTExUmVked+GDh2K+/fvY/jw4bh+/Tp2796N4OBgjB49utAht8KEh4dj7dq1uHLlCu7cuYMNGzZAW1sbNjY2Mr2eiJQfkx0lNn36dEyePBmhoaFwdHRE27ZtsX//ftjZ2ZW4zpo1a+LIkSO4dOkSGjduDDc3N+zevbvIdVPCwsLQp08fjBkzBrVq1YKPjw9iYmJgbW0tc5sDBw5ErVq14OrqChMTkwJnID6koaGBI0eOwNTUFO3bt4ezszNmz54NVVXVIl8jkUiwe/duVKpUCc2bN4enpyfs7e2xefNmAO/mzDx//hx9+vRBzZo10b17d7Rr106cpJubm4thw4aJx7lmzZpSk7jlNXzwKPh0+g7DxwyGd1dP6OjoomXzVtDU1JS5jmZNPWBkVAm37txE187fFVt2xIgRGD16NMaMGQNnZ2ccOnQIe/bskZqE/b4ff/wRXbt2RY8ePdCkSRM8f/4cQ4cOlSojy/tWtWpVHDhwAOfPn0f9+vUxePBg9O/fH5MmTSpQtihGRkZYvXo13N3dUa9ePRw7dgx79+4tl3WEiKhikgiKnCjwmUpLS4OhoSFSU1NhYGAgtS8zMxOJiYlS67EQlYUn99KK3JeXl4dvPBuhk3cXTBgjeyLwIVMbg48X+sLwO16xfYrV6stihXplVBHfi+J+v9/HS8+JKqD7/yUh6q/jaNrkG2RlZ+H3db8h6f49dO2suEnGRERfCiY79NmZNWuWuEbPh5o1a4aDBw9+4ogUT0VFBZu3RWDqrMkQBAG1azpi64bdqFmjFv57cB/N2jQp8rV/HT2HalVLPgmdiEjZMNmhz87gwYPRvXvhE27luTS+IqtqWQ37th8pdJ+5mQWOH/iryNeam1mUVVhERJ8lJjv02TE2Ni52wUBlp6amBjvb6h8vSEREAHg1FhERESk5Jjsy4kVrRMqJ320i5cdk5yPy12Ypq2Xwiah8ZWRkACh4SwoiUh6cs/MRampq0NHRwdOnT6Guri7zqq5E8nqbU/YJdWZmZpm38bkQBAEZGRl48uQJjIyMil10kog+b0x2PkIikcDCwgKJiYm4d+9eeYdDSuzV87JPRNKyuWjeh4yMjOS6VxsRfX6Y7MhAQ0MDDg4OHMqiMrUx/GyZt+E3teT361JG6urqPKND9AVgsiMjFRUVLiVPZSozNa/M2+BnmIi+RJyAQkREREpN7mTn4sWLiI+PF5/v3r0bPj4++PnnnznMQ0RERBWO3MnOjz/+iH///RcAcOfOHfTs2RM6OjrYunUrxo0bp/AAiYiIiEpD7mTn33//hYuLCwBg69ataN68OSIiIhAeHo7t27crOj4iIiKiUpE72REEAXl57yZSHjt2DO3btwcAWFlZ4dmzZ4qNjoiIiKiU5E52XF1dMWPGDPzxxx84efIkvL29AQCJiYkwMzNTeIBEREREpSF3srNo0SJcvHgRgYGB+OWXX1CjRg0AwLZt29C0aVOFB0hERERUGnKvs1OvXj2pq7Hy/e9//+PiXERERFThlHhRwezsbDx58kScv5PP2tq61EERERERKYrcyc6///6L/v3748yZM1LbBUGARCJBbm6uwoIjIiIiKi25k52+fftCTU0N+/btg4WFBSQSSVnERURERKQQcic7cXFxiI2NRe3atcsiHiIiIiKFkvtqLCcnJ66nQ0RERJ8NuZOdOXPmYNy4cYiKisLz58+RlpYm9SAiIiKqSOQexvL09AQAtG7dWmo7JygTERFRRSR3snPixImyiIOIiIioTMid7Hh4eJRFHERERERlokSLCqakpGDt2rVISEgAANSpUwf9+vWDoaGhQoMjIiIiKi25JyhfuHAB1atXx8KFC/HixQu8ePECCxYsQPXq1XHx4sWyiJGIiIioxOQ+szNq1Ch06tQJq1evhprau5fn5ORgwIABGDlyJE6dOqXwIImIiIhKSu5k58KFC1KJDgCoqalh3LhxcHV1VWhwRERERKUl9zCWgYEBkpKSCmy/f/8+9PX15Q7gwYMH+OGHH1C5cmVoa2vD2dkZFy5cEPcLgoApU6bAwsIC2tra8PT0xM2bN6XqePHiBfz8/GBgYAAjIyP0798f6enpcsdCREREykfuZKdHjx7o378/Nm/ejPv37+P+/fvYtGkTBgwYgF69eslV18uXL+Hu7g51dXUcPHgQ165dw/z581GpUiWxzNy5c/Hrr79i5cqVOHfuHHR1deHl5YXMzEyxjJ+fH65evYqjR49i3759OHXqFAYNGiRv14iIiEgJyT2MNW/ePEgkEvTp0wc5OTkAAHV1dQwZMgSzZ8+Wq645c+bAysoKYWFh4jY7Ozvx34IgYNGiRZg0aRI6d+4MAFi/fj3MzMywa9cu9OzZEwkJCTh06BBiYmLEYbQlS5agffv2mDdvHiwtLeXtIhERESkRuc/saGhoYPHixXj58iXi4uIQFxeHFy9eYOHChdDU1JSrrj179sDV1RXdunWDqakpGjRogNWrV4v7ExMT8ejRI3HVZgAwNDREkyZNEB0dDQCIjo6GkZGR1HwhT09PqKio4Ny5c4W2m5WVxdtcEBERfSHkTnby6ejowNnZGc7OztDR0SlRHXfu3MGKFSvg4OCAw4cPY8iQIRgxYgTWrVsHAHj06BEAwMzMTOp1ZmZm4r5Hjx7B1NRUar+amhqMjY3FMh8KDQ2FoaGh+LCysipR/ERERFTxyTSM1bVrV4SHh8PAwABdu3YttuyOHTtkbjwvLw+urq6YNWsWAKBBgwa4cuUKVq5cCX9/f5nrkdfEiRMxevRo8XlaWhoTHiIFWDb4eJm3MWxlqzJvg4iUi0zJjqGhISQSCYB3V2Pl/7u0LCws4OTkJLXN0dER27dvBwCYm5sDAB4/fgwLCwuxzOPHj+Hi4iKWefLkiVQdOTk5ePHihfj6D2lqaso95EZERFRS/EOgfMmU7Lw/gTg8PFxhjbu7u+PGjRtS2/7991/Y2NgAeDdZ2dzcHJGRkWJyk5aWhnPnzmHIkCEAADc3N6SkpCA2NhYNGzYEABw/fhx5eXlo0qSJwmIlIiKiz5Pcc3ZatWqFlJSUAtvT0tLQqpV8WeWoUaNw9uxZzJo1C7du3UJERAR+++03DBs2DAAgkUgwcuRIzJgxA3v27EF8fDz69OkDS0tL+Pj4AHh3Jqht27YYOHAgzp8/j9OnTyMwMBA9e/bklVhEREQk/6XnUVFRyM7OLrA9MzMTf/31l1x1NWrUCDt37sTEiRMxbdo02NnZYdGiRfDz8xPLjBs3Dq9fv8agQYOQkpKCb775BocOHYKWlpZYZuPGjQgMDETr1q2hoqICX19f/Prrr/J2jYiIiJSQzMnO5cuXxX9fu3ZN6kqn3NxcHDp0CFWrVpU7gA4dOqBDhw5F7pdIJJg2bRqmTZtWZBljY2NERETI3TYREREpP5mTHRcXF0gkEkgkkkKHq7S1tbFkyRKFBkdEVB44mZRIucic7CQmJkIQBNjb2+P8+fMwMTER92loaMDU1BSqqqplEiQRERFRScmc7ORfIZWXl1dmwRARkeKU9Rkqnp2iz4XcE5TzXbt2DUlJSQUmK3fq1KnUQREREREpitzJzp07d9ClSxfEx8dDIpFAEAQAEBcazM3NVWyEVKb4lx8RESk7uZOdn376CXZ2doiMjISdnR3Onz+P58+fY8yYMZg3b15ZxFghcQIjERHR50HuZCc6OhrHjx9HlSpVoKKiAhUVFXzzzTcIDQ3FiBEj8M8//5RFnEREREQlIvcKyrm5udDX1wcAVKlSBQ8fPgTwbgLzh7d+ICIiIipvcp/ZqVu3Li5dugQ7Ozs0adIEc+fOhYaGBn777TfY29uXRYxExeKQIhERFUfuZGfSpEl4/fo1AGDatGno0KEDmjVrhsqVK2Pz5s0KD5CIiIioNOROdry8vMR/16hRA9evX8eLFy9QqVIl8YosIiIiooqixOvsvM/Y2FgR1RAREREpnEzJTteuXWWucMeOHSUOhoiIiEjRZEp2DA0NyzoOIiIiojIhU7ITFhZW1nEQERERlYkSz9l5+vSpuK5OrVq1pO6CTkRERFRRyL2o4OvXr9GvXz9YWFigefPmaN68OSwtLdG/f39kZGSURYxEREREJSZ3sjN69GicPHkSe/fuRUpKClJSUrB7926cPHkSY8aMKYsYiYiIiEpM7mGs7du3Y9u2bWjRooW4rX379tDW1kb37t2xYsUKRcZHREREVCpyn9nJyMiAmZlZge2mpqYcxiIiIqIKR+5kx83NDcHBwcjMzBS3vXnzBlOnToWbm5tCgyMiIiIqLbmHsRYvXgwvLy9Uq1YN9evXBwBcunQJWlpaOHz4sMIDJCIiIiqNEt31/ObNm9i4cSOuX78OAOjVqxf8/Pygra2t8ACJiIiISqNE6+zo6Ohg4MCBio6FiIiISOFkSnb27NmDdu3aQV1dHXv27Cm2bKdOnRQSGBEREZEiyJTs+Pj44NGjRzA1NYWPj0+R5SQSCXJzcxUVGxEREVGpyZTs5OXlFfpvIiIioopO7kvP79+/XxZxEBEREZUJuZMdW1tbeHh4YPXq1Xj58mVZxERERESkMHInOxcuXEDjxo0xbdo0WFhYwMfHB9u2bUNWVlZZxEdERERUKnInOw0aNMD//vc/JCUl4eDBgzAxMcGgQYNgZmaGfv36lUWMRERERCUmd7KTTyKRoGXLlli9ejWOHTsGOzs7rFu3TpGxEREREZVaiZOd//77D3PnzoWLiwsaN24MPT09LFu2TJGxEREREZWa3Csor1q1ChERETh9+jRq164NPz8/7N69GzY2NmURHxEREVGpyJ3szJgxA7169cKvv/4q3giUiIiIqKKSO9lJSkqCRCIpi1iIiIiIFE7uOTtMdIiIiOhzUuIJykRERESfAyY7REREpNRkSnb27NmDt2/flnUsRERERAonU7LTpUsXpKSkAABUVVXx5MmTsoyJiIiISGFkSnZMTExw9uxZAIAgCJykTERERJ8NmS49Hzx4MDp37gyJRAKJRAJzc/Miy+bm5iosOCIiIqLSkinZCQkJQc+ePXHr1i106tQJYWFhMDIyKuPQiIiIiEpP5kUFa9eujdq1ayM4OBjdunWDjo5OWcZFREREpBByr6AcHBwMAHj69Clu3LgBAKhVqxZMTEwUGxkRERGRAsi9zk5GRgb69esHS0tLNG/eHM2bN4elpSX69++PjIyMsoiRiIiIqMTkTnZGjRqFkydPYs+ePUhJSUFKSgp2796NkydPYsyYMWURIxEREVGJyT2MtX37dmzbtg0tWrQQt7Vv3x7a2tro3r07VqxYocj4iIiIiEqlRMNYZmZmBbabmppyGIuIiIgqHLmTHTc3NwQHByMzM1Pc9ubNG0ydOhVubm4KDY6IiIiotOQexlq8eDG8vLxQrVo11K9fHwBw6dIlaGlp4fDhwwoPkIiIiKg05E526tati5s3b2Ljxo24fv06AKBXr17w8/ODtra2wgMkIiIiKg25kx0A0NHRwcCBAxUdCxEREZHCyT1nh4iIiOhzwmSHiIiIlBqTHSIiIlJqTHaIiIhIqcmd7Njb2+P58+cFtqekpMDe3l4hQREREREpitzJzt27d5Gbm1tge1ZWFh48eKCQoIiIiIgUReZLz/fs2SP++/DhwzA0NBSf5+bmIjIyEra2tgoNjoiIiKi0ZE52fHx8AAASiQT+/v5S+9TV1WFra4v58+crNDgiIiKi0pI52cnLywMA2NnZISYmBlWqVCmzoIiIiIgURe45O4mJiWWS6MyePRsSiQQjR44Ut2VmZmLYsGGoXLky9PT04Ovri8ePH0u9LikpCd7e3tDR0YGpqSnGjh2LnJwchcdHREREn6cS3S4iMjISkZGRePLkiXjGJ9/vv/8ud30xMTFYtWoV6tWrJ7V91KhR2L9/P7Zu3QpDQ0MEBgaia9euOH36NIB3c4W8vb1hbm6OM2fOIDk5GX369IG6ujpmzZpVkq4RERGRkpH7zM7UqVPx7bffIjIyEs+ePcPLly+lHvJKT0+Hn58fVq9ejUqVKonbU1NTsXbtWixYsACtWrVCw4YNERYWhjNnzuDs2bMAgCNHjuDatWvYsGEDXFxc0K5dO0yfPh3Lli1Ddna23LEQERGR8pH7zM7KlSsRHh6O3r17KySAYcOGwdvbG56enpgxY4a4PTY2Fm/fvoWnp6e4rXbt2rC2tkZ0dDS+/vprREdHw9nZGWZmZmIZLy8vDBkyBFevXkWDBg0UEiMRERF9vuROdrKzs9G0aVOFNL5p0yZcvHgRMTExBfY9evQIGhoaMDIyktpuZmaGR48eiWXeT3Ty9+fvK0pWVhaysrLE52lpaSXtAhEREVVwcg9jDRgwABEREaVu+P79+/jpp5+wceNGaGlplbo+eYSGhsLQ0FB8WFlZfdL2iYiI6NOR+8xOZmYmfvvtNxw7dgz16tWDurq61P4FCxbIVE9sbCyePHmCr776StyWm5uLU6dOYenSpTh8+DCys7ORkpIidXbn8ePHMDc3BwCYm5vj/PnzUvXmX62VX6YwEydOxOjRo8XnaWlpTHiIiIiUlNzJzuXLl+Hi4gIAuHLlitQ+iUQicz2tW7dGfHy81La+ffuidu3aGD9+PKysrKCuro7IyEj4+voCAG7cuIGkpCS4ubkBANzc3DBz5kw8efIEpqamAICjR4/CwMAATk5ORbatqakJTU1NmWMlIiKiz5fcyc6JEycU0rC+vj7q1q0rtU1XVxeVK1cWt/fv3x+jR4+GsbExDAwMMHz4cLi5ueHrr78GAHz77bdwcnJC7969MXfuXDx69AiTJk3CsGHDmMwQERERgBKus/OpLFy4ECoqKvD19UVWVha8vLywfPlycb+qqir27duHIUOGwM3NDbq6uvD398e0adPKMWoiIiKqSOROdlq2bFnscNXx48dLHExUVJTUcy0tLSxbtgzLli0r8jU2NjY4cOBAidskIiIi5SZ3spM/Xyff27dvERcXhytXrhS4QSgRERFReZM72Vm4cGGh20NCQpCenl7qgIiIiIgUSe51doryww8/lOi+WERERERlSWHJTnR09CdfHJCIiIjoY+QexuratavUc0EQkJycjAsXLmDy5MkKC4yIiIhIEeROdgwNDaWeq6iooFatWpg2bRq+/fZbhQVGREREpAhyJzthYWFlEQcRERFRmSjxooKxsbFISEgAANSpUwcNGjRQWFBEREREiiJ3svPkyRP07NkTUVFR4g06U1JS0LJlS2zatAkmJiaKjpGIiIioxOS+Gmv48OF49eoVrl69ihcvXuDFixe4cuUK0tLSMGLEiLKIkYiIiKjE5D6zc+jQIRw7dgyOjo7iNicnJyxbtowTlImIiKjCkfvMTl5eHtTV1QtsV1dXR15enkKCIiIiIlIUuZOdVq1a4aeffsLDhw/FbQ8ePMCoUaPQunVrhQZHREREVFpyJztLly5FWloabG1tUb16dVSvXh12dnZIS0vDkiVLyiJGIiIiohKTe86OlZUVLl68iGPHjuH69esAAEdHR3h6eio8OCIiIqLSKtE6OxKJBG3atEGbNm0UHQ8RERGRQsk8jHX8+HE4OTkhLS2twL7U1FTUqVMHf/31l0KDIyIiIiotmZOdRYsWYeDAgTAwMCiwz9DQED/++CMWLFig0OCIiIiISkvmZOfSpUto27Ztkfu//fZbxMbGKiQoIiIiIkWROdl5/Phxoevr5FNTU8PTp08VEhQRERGRosic7FStWhVXrlwpcv/ly5dhYWGhkKCIiIiIFEXmZKd9+/aYPHkyMjMzC+x78+YNgoOD0aFDB4UGR0RERFRaMl96PmnSJOzYsQM1a9ZEYGAgatWqBQC4fv06li1bhtzcXPzyyy9lFigRERFRScic7JiZmeHMmTMYMmQIJk6cCEEQALxbc8fLywvLli2DmZlZmQVKREREVBJyLSpoY2ODAwcO4OXLl7h16xYEQYCDgwMqVapUVvERERERlUqJVlCuVKkSGjVqpOhYiIiIiBRO7huBEhEREX1OmOwQERGRUmOyQ0REREqNyQ4REREpNSY7REREpNSY7BAREZFSY7JDRERESo3JDhERESk1JjtERESk1JjsEBERkVJjskNERERKjckOERERKTUmO0RERKTUmOwQERGRUmOyQ0REREqNyQ4REREpNSY7REREpNSY7BAREZFSY7JDRERESo3JDhERESk1JjtERESk1JjsEBERkVJjskNERERKjckOERERKTUmO0RERKTUmOwQERGRUmOyQ0REREqNyQ4REREpNSY7REREpNSY7BAREZFSY7JDRERESo3JDhERESk1JjtERESk1JjsEBERkVJjskNERERKjckOERERKTUmO0RERKTUmOwQERGRUivXZCc0NBSNGjWCvr4+TE1N4ePjgxs3bkiVyczMxLBhw1C5cmXo6enB19cXjx8/liqTlJQEb29v6OjowNTUFGPHjkVOTs6n7AoRERFVUOWa7Jw8eRLDhg3D2bNncfToUbx9+xbffvstXr9+LZYZNWoU9u7di61bt+LkyZN4+PAhunbtKu7Pzc2Ft7c3srOzcebMGaxbtw7h4eGYMmVKeXSJiIiIKhi18mz80KFDUs/Dw8NhamqK2NhYNG/eHKmpqVi7di0iIiLQqlUrAEBYWBgcHR1x9uxZfP311zhy5AiuXbuGY8eOwczMDC4uLpg+fTrGjx+PkJAQaGholEfXiIiIqIKoUHN2UlNTAQDGxsYAgNjYWLx9+xaenp5imdq1a8Pa2hrR0dEAgOjoaDg7O8PMzEws4+XlhbS0NFy9evUTRk9EREQVUbme2XlfXl4eRo4cCXd3d9StWxcA8OjRI2hoaMDIyEiqrJmZGR49eiSWeT/Ryd+fv68wWVlZyMrKEp+npaUpqhtERERUwVSYMzvDhg3DlStXsGnTpjJvKzQ0FIaGhuLDysqqzNskIiKi8lEhkp3AwEDs27cPJ06cQLVq1cTt5ubmyM7ORkpKilT5x48fw9zcXCzz4dVZ+c/zy3xo4sSJSE1NFR/3799XYG+IiIioIinXZEcQBAQGBmLnzp04fvw47OzspPY3bNgQ6urqiIyMFLfduHEDSUlJcHNzAwC4ubkhPj4eT548EcscPXoUBgYGcHJyKrRdTU1NGBgYSD2IiIhIOZXrnJ1hw4YhIiICu3fvhr6+vjjHxtDQENra2jA0NET//v0xevRoGBsbw8DAAMOHD4ebmxu+/vprAMC3334LJycn9O7dG3PnzsWjR48wadIkDBs2DJqamuXZPSIiIqoAyjXZWbFiBQCgRYsWUtvDwsIQEBAAAFi4cCFUVFTg6+uLrKwseHl5Yfny5WJZVVVV7Nu3D0OGDIGbmxt0dXXh7++PadOmfapuEBERUQVWrsmOIAgfLaOlpYVly5Zh2bJlRZaxsbHBgQMHFBkaERERKYkKMUGZiIiIqKww2SEiIiKlxmSHiIiIlBqTHSIiIlJqTHaIiIhIqTHZISIiIqXGZIeIiIiUGpMdIiIiUmpMdoiIiEipMdkhIiIipcZkh4iIiJQakx0iIiJSakx2iIiISKkx2SEiIiKlxmSHiIiIlBqTHSIiIlJqTHaIiIhIqTHZISIiIqXGZIeIiIiUGpMdIiIiUmpMdoiIiEipMdkhIiIipcZkh4iIiJQakx0iIiJSakx2iIiISKkx2SEiIiKlxmSHiIiIlBqTHSIiIlJqTHaIiIhIqTHZISIiIqXGZIeIiIiUGpMdIiIiUmpMdoiIiEipMdkhIiIipcZkh4iIiJQakx0iIiJSakx2iIiISKkx2SEiIiKlxmSHiIiIlBqTHSIiIlJqTHaIiIhIqTHZISIiIqXGZIeIiIiUGpMdIiIiUmpMdoiIiEipMdkhIiIipcZkh4iIiJQakx0iIiJSakx2iIiISKkx2SEiIiKlxmSHiIiIlBqTHSIiIlJqTHaIiIhIqTHZISIiIqXGZIeIiIiUGpMdIiIiUmpMdoiIiEipMdkhIiIipaZW3gEQERFRxbfS7acyb2MY4sukXiY7REQf+Jz/UycqK/GJSeUdQokx2SGqIJThB3aYeZcyrf+d1DJv4XP+T13ZKMP3AlCe78bnislOCSnLB7fs+8Evn6z4A0tUEL8XpAhMduizpyx/+REpWll/N/i9oM8Fkx367PEvP6LC8btB9I7SXHq+bNky2NraQktLC02aNMH58+fLOyQiIiKqAJTizM7mzZsxevRorFy5Ek2aNMGiRYvg5eWFGzduwNTUtLzDI6LPjG1mRJm3cbfMWyCifEpxZmfBggUYOHAg+vbtCycnJ6xcuRI6Ojr4/fffyzs0IiIiKmef/Zmd7OxsxMbGYuLEieI2FRUVeHp6Ijo6uszaVZa//Mq6H3fLtPZ3+F7I7m4Z168MfVAmyvD9VhbK8N34nPvw2Sc7z549Q25uLszMzKS2m5mZ4fr164W+JisrC1lZWeLz1NR3l0enpaXJ3G5eVkYJopWPPPGUVFn3Qxn6ALAfslKGPgDsh6w+RR/qpq4p8zauKMF7AXyZ3+/88oIgFF9Q+Mw9ePBAACCcOXNGavvYsWOFxo0bF/qa4OBgAQAffPDBBx988KEEj/v37xebK3z2Z3aqVKkCVVVVPH78WGr748ePYW5uXuhrJk6ciNGjR4vP8/Ly8OLFC1SuXBkSiUThMaalpcHKygr379+HgYGBwuv/VNiPikMZ+gAoRz+UoQ8A+1GRKEMfgE/TD0EQ8OrVK1haWhZb7rNPdjQ0NNCwYUNERkbCx8cHwLvkJTIyEoGBgYW+RlNTE5qamlLbjIyMyjhSwMDA4LP+4OZjPyoOZegDoBz9UIY+AOxHRaIMfQDKvh+GhoYfLfPZJzsAMHr0aPj7+8PV1RWNGzfGokWL8Pr1a/Tt27e8QyMiIqJyphTJTo8ePfD06VNMmTIFjx49gouLCw4dOlRg0jIRERF9eZQi2QGAwMDAIoetypumpiaCg4MLDJ19btiPikMZ+gAoRz+UoQ8A+1GRKEMfgIrVD4kgfOx6LSIiIqLPl1KsoExERERUFCY7REREpNSY7BAREZFSY7JDROWK0waJqKwpzdVYFcmzZ8/w+++/Izo6Go8ePQIAmJubo2nTpggICICJiUk5R0hUcWhqauLSpUtwdHQs71CISEnxaiwFi4mJgZeXF3R0dODp6Smu9fP48WNERkYiIyMDhw8fhqurazlH+uV48+YNYmNjYWxsDCcnJ6l9mZmZ2LJlC/r06VNO0ckmISEBZ8+ehZubG2rXro3r169j8eLFyMrKwg8//IBWrVqVd4gf9f4tWt63ePFi/PDDD6hcuTIAYMGCBZ8yrFJ7/fo1tmzZglu3bsHCwgK9evUS+0Jla/jw4ejevTuaNWtW3qEQgOTkZKxYsQJ///03kpOToaKiAnt7e/j4+CAgIACqqqrlF5wC7sVJ72nSpIkwaNAgIS8vr8C+vLw8YdCgQcLXX39dDpEpXlJSktC3b9/yDqNYN27cEGxsbASJRCKoqKgIzZs3Fx4+fCjuf/TokaCiolKOEX7cwYMHBQ0NDcHY2FjQ0tISDh48KJiYmAienp5Cq1atBFVVVSEyMrK8w/woiUQiuLi4CC1atJB6SCQSoVGjRkKLFi2Eli1blneYH+Xo6Cg8f/5cEIR33wFbW1vB0NBQaNSokWBsbCyYmpoKd+7cKecoixcbGysV4/r164WmTZsK1apVE9zd3YU///yzHKOTXf732sHBQZg9e7aQnJxc3iGVyJIlS4TevXuLx339+vWCo6OjUKtWLWHixInC27dvyznCj4uJiREMDQ2Fhg0bCt98842gqqoq9O7dW+jRo4dgZGQkNG3aVEhLSyu3+JjsKJiWlpaQkJBQ5P6EhARBS0vrE0ZUduLi4ip8ouDj4yN4e3sLT58+FW7evCl4e3sLdnZ2wr179wRB+DySHTc3N+GXX34RBEEQ/vzzT6FSpUrCzz//LO6fMGGC0KZNm/IKT2ahoaGCnZ1dgcRMTU1NuHr1ajlFJT+JRCI8fvxYEARB8PPzE5o2bSqkpKQIgiAIr169Ejw9PYVevXqVZ4gfVa9ePeHo0aOCIAjC6tWrBW1tbWHEiBHCihUrhJEjRwp6enrC2rVryznKj5NIJMKxY8eEn376SahSpYqgrq4udOrUSdi7d6+Qm5tb3uHJZPr06YK+vr7g6+srmJubC7NnzxYqV64szJgxQ5g1a5ZgYmIiTJkypbzD/Ch3d3chJCREfP7HH38ITZo0EQRBEF68eCG4uLgII0aMKK/wmOwomq2trbBu3boi969bt06wsbH5dAGVwu7du4t9LFy4sMInCqampsLly5fF53l5ecLgwYMFa2tr4fbt259FsmNgYCDcvHlTEARByM3NFdTU1ISLFy+K++Pj4wUzM7PyCk8u58+fF2rWrCmMGTNGyM7OFgTh80527O3thSNHjkjtP336tGBlZVUeoclMW1tbuHv3riAIgtCgQQPht99+k9q/ceNGwcnJqTxCk8v770V2drawefNmwcvLS1BVVRUsLS2Fn3/+WfzuVFTVq1cXtm/fLgjCuz8gVVVVhQ0bNoj7d+zYIdSoUaO8wpOZtra2cPv2bfF5bm6uoK6uLjx69EgQBEE4cuSIYGlpWV7hCZygrGBBQUEYNGgQYmNj0bp16wJzdlavXo158+aVc5Sy8fHxgUQiKfZqGYlE8gkjkt+bN2+gpvZ/H3OJRIIVK1YgMDAQHh4eiIiIKMfoZJd/nFVUVKClpSV1l199fX2kpqaWV2hyadSoEWJjYzFs2DC4urpi48aNFf4zVJj8mDMzM2FhYSG1r2rVqnj69Gl5hCUzHR0dPHv2DDY2Nnjw4AEaN24stb9JkyZITEwsp+hKRl1dHd27d0f37t2RlJSE33//HeHh4Zg9ezZyc3PLO7wiPXz4UJzDWb9+faioqMDFxUXc/9VXX+Hhw4flFJ3sTE1NkZycDHt7ewDvfvNycnLEu507ODjgxYsX5RYfLz1XsGHDhmHdunU4d+4cfH194ebmBjc3N/j6+uLcuXMIDw/H0KFDyztMmVhYWGDHjh3Iy8sr9HHx4sXyDvGjateujQsXLhTYvnTpUnTu3BmdOnUqh6jkY2tri5s3b4rPo6OjYW1tLT5PSkoq8INbkenp6WHdunWYOHEiPD09K/QPUVFat26Nr776Cmlpabhx44bUvnv37lX4Ccrt2rXDihUrAAAeHh7Ytm2b1P4tW7agRo0a5RGaQlhbWyMkJASJiYk4dOhQeYdTLHNzc1y7dg0AcPPmTeTm5orPAeDq1aswNTUtr/Bk5uPjg8GDB+PQoUM4ceIE/Pz84OHhAW1tbQDAjRs3ULVq1XKLj2d2ykCPHj3Qo0cPvH37Fs+ePQMAVKlSBerq6uUcmXwaNmyI2NhYdO7cudD9HzvrUxF06dIFf/75J3r37l1g39KlS5GXl4eVK1eWQ2SyGzJkiFRCULduXan9Bw8e/CyuxvpQz5498c033yA2NhY2NjblHY7MgoODpZ7r6elJPd+7d2+Fvzpozpw5cHd3h4eHB1xdXTF//nxERUXB0dERN27cwNmzZ7Fz587yDvOjbGxsir3CRyKRoE2bNp8wIvn5+fmhT58+6Ny5MyIjIzFu3DgEBQXh+fPnkEgkmDlzJr777rvyDvOjZsyYgeTkZHTs2BG5ublwc3PDhg0bxP0SiQShoaHlFh8vPaci/fXXX3j9+jXatm1b6P7Xr1/jwoUL8PDw+MSREVFppaSkYPbs2di7dy/u3LmDvLw8WFhYwN3dHaNGjeLyGJ9IXl4eZs+ejejoaDRt2hQTJkzA5s2bMW7cOGRkZKBjx45YunQpdHV1yztUmWRmZiInJ6fAHwHljckOERERKTXO2SEiIiKlxmSHiIiIlBqTHSIiIlJqTHaI6LNx9+5dSCQSxMXFlVkbAQEB8PHxKbP6iejTY7JDRJ9MQEAAJBJJgUdRV/x9yMrKCsnJyQUuvyciKg7X2SGiT6pt27YICwuT2qapqSnTa1VVVWFubl4WYRGREuOZHSL6pDQ1NWFubi71qFSpEoD/u51Hu3btoK2tDXt7e6nVfT8cxnr58iX8/PxgYmICbW1tODg4SCVS8fHxaNWqFbS1tVG5cmUMGjQI6enp4v7c3FyMHj0aRkZGqFy5MsaNG1dgocy8vDyEhobCzs4O2traqF+/vlRMH4uBiMofkx0iqlAmT54MX19fXLp0CX5+fujZsycSEhKKLHvt2jUcPHgQCQkJWLFiBapUqQLg3aKXXl5eqFSpEmJiYrB161YcO3YMgYGB4uvnz5+P8PBw/P777/j777/x4sWLAisHh4aGYv369Vi5ciWuXr2KUaNG4YcffsDJkyc/GgMRVRDldgtSIvri+Pv7C6qqqoKurq7UY+bMmYIgCAIAYfDgwVKvadKkiTBkyBBBEAQhMTFRACD8888/giAIQseOHYW+ffsW2tZvv/0mVKpUSUhPTxe37d+/X1BRURHvxGxhYSHMnTtX3P/27VuhWrVqQufOnQVBEITMzExBR0dHOHPmjFTd/fv3F3r16vXRGIioYuCcHSL6pFq2bCnehDKfsbGx+G83NzepfW5ubkVefTVkyBD4+vri4sWL+Pbbb+Hj44OmTZsCABISElC/fn2pZfbd3d2Rl5eHGzduQEtLC8nJyWjSpIm4X01NDa6uruJQ1q1bt5CRkVHg/krZ2dlo0KDBR2MgooqByQ4RfVK6uroKu6N2u3btcO/ePRw4cABHjx5F69atMWzYMMybN08h9efP79m/f3+BOzbnT6ou6xiIqPQ4Z4eIKpSzZ88WeO7o6FhkeRMTE/j7+2PDhg1YtGgRfvvtNwCAo6MjLl26hNevX4tlT58+DRUVFdSqVQuGhoawsLDAuXPnxP05OTmIjY0Vnzs5OUFTUxNJSUmoUaOG1MPKyuqjMRBRxcAzO0T0SWVlZeHRo0dS29TU1MRJvVu3boWrqyu++eYbbNy4EefPn8fatWsLrWvKlClo2LAh6tSpg6ysLOzbt09MjPz8/BAcHAx/f3+EhITg6dOnGD58OHr37g0zMzMAwE8//YTZs2fDwcEBtWvXxoIFC5CSkiLWr6+vj6CgIIwaNQp5eXn45ptvkJqaitOnT8PAwAD+/v7FxkBEFQOTHSL6pA4dOgQLCwupbbVq1cL169cBAFOnTsWmTZswdOhQWFhY4M8//4STk1OhdWloaGDixIm4e/cutLW10axZM2zatAkAoKOjg8OHD+Onn35Co0aNoKOjA19fXyxYsEB8/ZgxY5CcnAx/f3+oqKigX79+6NKlC1JTU8Uy06dPh4mJCUJDQ3Hnzh0YGRnhq6++ws8///zRGIioYpAIwgeLShARlROJRIKdO3fydg1EpFCcs0NERERKjckOERERKTXO2SGiCoOj6kRUFnhmh4iIiJQakx0iIiJSakx2iIiISKkx2SEiIiKlxmSHiIiIlBqTHSIiIlJqTHaIiIhIqTHZISIiIqXGZIeIiIiU2v8Db8+E2iee5wUAAAAASUVORK5CYII=", + "image/png": "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", "text/plain": [ "
" ] @@ -128325,7 +98689,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -128335,7 +98699,7 @@ }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -128344,18 +98708,15 @@ "output_type": "display_data" }, { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" + "name": "stdout", + "output_type": "stream", + "text": [ + "Kolmogorov-Smirnov test: D = 0.024356554511759887, p-value = 0.9924129380161606\n" + ] }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -128364,18 +98725,15 @@ "output_type": "display_data" }, { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" + "name": "stdout", + "output_type": "stream", + "text": [ + "Kolmogorov-Smirnov test: D = 0.029132953297616726, p-value = 0.954427253065449\n" + ] }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -128384,20 +98742,17 @@ "output_type": "display_data" }, { - "data": { - "image/png": "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", - "text/plain": [ - "
" - ] - }, - "metadata": {}, - "output_type": "display_data" + "name": "stdout", + "output_type": "stream", + "text": [ + "The correlation between 'costs_from_violations_smooth' and 'at_dest_f' is 0.1085294435239275\n" + ] } ], "source": [ "avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f = None, None, None, None, None, None, None\n", "for seed in seeds:\n", - " avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results= simulation_10_to_4_agents(seed, False) \n", + " avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results= simulation_10_agents(seed, False) \n", " if seed == seeds[0]:\n", " avg_training_rewards_f, avg_training_norm_violations_f, sim_violated_f, at_dest_f, first_gen_results_f, middle_gen_results_f, last_gen_results_f= avg_training_rewards, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results\n", " first_gen_results_f = [list(item) for item in first_gen_results_f]\n", @@ -128405,10 +98760,7 @@ " last_gen_results_f = [list(item) for item in last_gen_results_f]\n", " else:\n", " avg_training_rewards_f = [sum(x) / 2 for x in zip(avg_training_rewards_f, avg_training_rewards)]\n", - " print(\"here\")\n", - " print(first_gen_results[0])\n", - " print(\"here1\")\n", - " print(first_gen_results[0][0])\n", + " \n", "\n", "\n", " first_gen_results_f[0][0] = [sum(x) / 2 for x in zip(first_gen_results_f[0][0], first_gen_results[0][0])]\n", @@ -128416,6 +98768,7 @@ " last_gen_results_f[0][0] = [sum(x) / 2 for x in zip(last_gen_results_f[0][0], last_gen_results[0][0])]\n", "\n", " avg_training_norm_violations_f = avg_training_norm_violations_f.add(avg_training_norm_violations).div(2)\n", + " \n", " first_gen_results_f[0][1] = first_gen_results_f[0][1].add(first_gen_results[0][1]).div(2)\n", " middle_gen_results_f[0][1] = middle_gen_results_f[0][1].add(middle_gen_results[0][1]).div(2)\n", " last_gen_results_f[0][1] = last_gen_results_f[0][1].add(last_gen_results[0][1]).div(2)\n", @@ -128437,100 +98790,155 @@ "\n", "#training violations\n", "costs_from_violations = copy.deepcopy(avg_training_norm_violations_f['total_violations_cost'])\n", + "\n", "avg_training_norm_violations_f.drop(columns=['seed'], inplace=True)\n", "avg_training_norm_violations_f.drop(columns=['total_violations_cost'], inplace=True)\n", "\n", + "# Calculate the moving average\n", + " # Change this to the desired window size\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=200).mean()\n", + "\n", "#graph costs from violations\n", - "plt.plot(costs_from_violations)\n", + "plt.plot(costs_from_violations_smooth)\n", "plt.title('AVERAGE Costs from violations training')\n", "plt.xlabel('Episodes')\n", "plt.ylabel('Cost')\n", "plt.show()\n", - "avg_training_norm_violations_f.plot(kind='bar', stacked=True)\n", - "plt.title('Average training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "plt.show()\n", "\n", - "#FIRST GEN TRAINING VIOLATIONS \n", "costs_from_violations = copy.deepcopy(first_gen_results_f[0][1]['total_violations_cost'])\n", "first_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", "first_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", "\n", "#graph costs from violations\n", - "plt.plot(costs_from_violations)\n", - "plt.title('First generation Costs from violations training')\n", + "####costs_from_violations = np.mean(costs_from_violations, axis=0)\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=200).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('First Gen Costs from violations training')\n", "plt.xlabel('Episodes')\n", "plt.ylabel('Cost')\n", "plt.show()\n", - "first_gen_results_f[0][1].plot(kind='bar', stacked=True)\n", - "plt.title('First generation training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "plt.show()\n", "\n", - "#MIDDLE GEN TRAINING VIOLATIONS\n", "costs_from_violations = copy.deepcopy(middle_gen_results_f[0][1]['total_violations_cost'])\n", "middle_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", "middle_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", - "plt.plot(costs_from_violations)\n", - "plt.title('Middle generation Costs from violations training')\n", + "#costs_from_violations = np.mean(costs_from_violations, axis=0)\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=200).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('Middle Gen Costs from violations training')\n", "plt.xlabel('Episodes')\n", "plt.ylabel('Cost')\n", "plt.show()\n", - "middle_gen_results_f[0][1].plot(kind='bar', stacked=True)\n", - "plt.title('Middle generation training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "\n", - "plt.show()\n", "\n", - "#LAST GEN TRAINING VIOLATIONS\n", "costs_from_violations = copy.deepcopy(last_gen_results_f[0][1]['total_violations_cost'])\n", "last_gen_results_f[0][1].drop(columns=['seed'], inplace=True)\n", "last_gen_results_f[0][1].drop(columns=['total_violations_cost'], inplace=True)\n", - "plt.plot(costs_from_violations)\n", - "plt.title('Last generation Costs from violations training')\n", + "#cost_from_violations = np.mean(costs_from_violations, axis=0)\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations)\n", + "costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", + "costs_from_violations_smooth = costs_from_violations_smooth.ewm(span=100).mean()\n", + "\n", + "#graph costs from violations\n", + "plt.plot(costs_from_violations_smooth)\n", + "plt.title('Final Gen Costs from violations training')\n", "plt.xlabel('Episodes')\n", "plt.ylabel('Cost')\n", "plt.show()\n", - "last_gen_results_f[0][1].plot(kind='bar', stacked=True)\n", - "plt.title('Last generation training norm violations')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Count of violations')\n", - "plt.show()\n", - "\n", - "\n", "\n", "\n", "#training violations\n", "costs_from_violations_f = copy.deepcopy(sim_violated_f['total_violations_cost'])\n", "sim_violated_f.drop(columns=['seed'], inplace=True)\n", "sim_violated_f.drop(columns=['total_violations_cost'], inplace=True)\n", + "costs_from_violations = -1 * costs_from_violations_f\n", + "costs_from_violations_smooth = pd.Series(costs_from_violations).rolling(window=window_size2).mean()\n", + "#costs_from_violations_smooth = -1 * costs_from_violations_smooth\n", "\n", - "#graph costs from violations\n", - "plt.plot(costs_from_violations)\n", - "plt.title('Costs from violations simulation')\n", - "plt.xlabel('Episodes')\n", - "plt.ylabel('Cost')\n", + "sns.histplot(costs_from_violations, kde=True)\n", + "plt.title('Distribution of Costs from Violations Simulation')\n", + "plt.xlabel('Cost')\n", + "plt.ylabel('Density')\n", "plt.show()\n", "\n", - "# Plot a stacked bar chart\n", - "sim_violated_f.plot(kind='bar', stacked=True)\n", - "plt.title('Average simulation norm violations')\n", - "plt.xlabel('Simulation Run')\n", - "plt.ylabel('Average norm violations')\n", + "#GAMMA\n", + "data = costs_from_violations\n", + "\n", + "# Fit the gamma distribution to your data\n", + "shape, loc, scale = gamma.fit(data)\n", + "\n", + "# Generate a range of values from the min to the max of your data\n", + "x = np.linspace(min(data), max(data), 10000)\n", + "\n", + "# Generate the gamma PDF with the parameters obtained from the fit\n", + "pdf = gamma.pdf(x, shape, loc, scale)\n", + "\n", + "# Plot the histogram of your data and the gamma PDF\n", + "plt.hist(data, bins=30, density=True, alpha=0.5, label='Histogram of Costs from Violations Simulation')\n", + "plt.plot(x, pdf, 'r-', label='Gamma PDF')\n", + "plt.legend()\n", + "plt.show()\n", + "\n", + "# Perform the Kolmogorov-Smirnov test\n", + "d, p_value = kstest(data, 'gamma', args=(shape, loc, scale))\n", + "print(f\"Kolmogorov-Smirnov test: D = {d}, p-value = {p_value}\")\n", + "\n", + "#LOG-NORMAL\n", + "# Fit the log-normal distribution to your data\n", + "shape, loc, scale = lognorm.fit(data)\n", + "\n", + "# Generate the log-normal PDF with the parameters obtained from the fit\n", + "pdf = lognorm.pdf(x, shape, loc, scale)\n", + "\n", + "# Plot the histogram of your data and the log-normal PDF\n", + "plt.hist(data, bins=30, density=True, alpha=0.5, label='Histogram of Costs from Violations Simulation')\n", + "plt.plot(x, pdf, 'r-', label='Log-normal PDF')\n", + "plt.legend()\n", "plt.show()\n", "\n", + "# Perform the Kolmogorov-Smirnov test\n", + "d, p_value = kstest(data, 'lognorm', args=(shape, loc, scale))\n", + "print(f\"Kolmogorov-Smirnov test: D = {d}, p-value = {p_value}\")\n", + "\n", + "\n", "#simulation at destination, avg timeto destination\n", - "plt.plot(at_dest_f)\n", - "plt.title('Average Steps to destination')\n", - "plt.xlabel('Simulation Run')\n", - "plt.ylabel('Steps ')\n", + "#simulation at destination, avg timeto destination\n", + "at_dest_f = np.array(at_dest_f)\n", + "sns.histplot(at_dest_f, kde=True)\n", + "plt.title('Distribution of Average Steps to Destination')\n", + "plt.xlabel('Steps')\n", + "plt.ylabel('Density')\n", "plt.show()\n", "\n", "\n", "\n", + "df = pd.DataFrame({\n", + " 'costs_from_violations_smooth': costs_from_violations,\n", + " 'at_dest_f': at_dest_f\n", + "})\n", + "\n", + "# Calculate the correlation\n", + "correlation = df['costs_from_violations_smooth'].corr(df['at_dest_f'])\n", + "\n", + "print(f\"The correlation between 'costs_from_violations_smooth' and 'at_dest_f' is {correlation}\")\n", + "violation_counts = sim_violated_f.sum()\n", + "\n", + "\n", + "plt.pie(violation_counts, autopct='%1.1f%%')\n", + "\n", + "plt.title('Violation Count')\n", + "\n", + "plt.legend(violation_counts.index, title=\"Violations\", loc=\"best\")\n", + "\n", + "plt.show()\n", "\n", "\n", "\n" @@ -128553,7 +98961,7 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.11.5" + "version": "3.11.9" } }, "nbformat": 4, diff --git a/main.py b/main.py index 93c22cd..50b7db9 100644 --- a/main.py +++ b/main.py @@ -14,9 +14,9 @@ info_df = pd.DataFrame(columns=['Actions', 'Observations', 'Information', 'Rewards', 'Done']) info_df_dqn_training = pd.DataFrame(columns=['Actions', 'Observations', 'Information', 'Rewards', 'Done']) episodes = 50 -generations = 100 -max_steps = 50 -runs = 10 +generations = 300 +max_steps = 100 +runs = 300 first_gen = 1 last_gen = generations @@ -110,7 +110,7 @@ def train_agents(env, agents, episodes=2, policies=None, info_df=info_df, seed=4 states = next_states steps += 1 if all(done): - avg_reward.append(sum(episode_rewards)/steps/env.n_agents) + avg_reward.append(sum(episode_rewards)/env.n_agents) break @@ -248,7 +248,7 @@ def simulation_10_agents(seed=42, view = False): env = gym.make('TrafficJunction10-v0') agents, avg_rewards_training, avg_training_norm_violations, first_gen_results, middle_gen_results, last_gen_results = psro_simulation(env, generations, episodes, "Nash", seed, info_training) print("Running Computed Policies") - sim_violated, at_dest = run_computed_policies_dqn(env, agents, 10, view) #nash + sim_violated, at_dest = run_computed_policies_dqn(env, agents, runs, view, seed) #nash env.close() return avg_rewards_training, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results @@ -267,7 +267,7 @@ def simulation_4_agents(seed=42, view = False): env = gym.make('TrafficJunction4-v0') agents, avg_rewards_training, avg_training_norm_violations, first_gen_results, middle_gen_results, last_gen_results = psro_simulation(env, generations, episodes, "Nash", seed, info_training) print("Running Computed Policies") - sim_violated, at_dest = run_computed_policies_dqn(env, agents, 10, view) #nash + sim_violated, at_dest = run_computed_policies_dqn(env, agents, runs, view, seed) #nash env.close() return avg_rewards_training, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results @@ -300,7 +300,7 @@ def simulation_10_to_4_agents(seed=42, view = False): counter += 1 else: break - sim_violated, at_dest = run_computed_policies_dqn(env, new_agents, 10, view) #nash + sim_violated, at_dest = run_computed_policies_dqn(env, new_agents, runs, view, seed) #nash print(sim_violated) env.close() return avg_rewards_training, avg_training_norm_violations, sim_violated, at_dest, first_gen_results, middle_gen_results, last_gen_results diff --git a/ppo.py b/ppo.py index c2038d3..460afff 100644 --- a/ppo.py +++ b/ppo.py @@ -1,144 +1,119 @@ -import tensorflow as tf +import torch +import torch.nn as nn +import torch.optim as optim import numpy as np -from tensorflow.keras.models import Model -from tensorflow.keras.layers import Input, Dense, Flatten -from tensorflow.keras.optimizers import Adam from collections import deque import random import matplotlib.pyplot as plt -from tensorflow.python.framework.ops import disable_eager_execution -import os -os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3' -disable_eager_execution() +class Actor(nn.Module): + def __init__(self, state_size, action_size): + super(Actor, self).__init__() + self.fc1 = nn.Linear(state_size, 24) + self.fc2 = nn.Linear(24, 24) + self.output = nn.Linear(24, action_size) + self.softmax = nn.Softmax(dim=-1) + + + def forward(self, x, advantages, old_probs): + x = torch.relu(self.fc1(x)) + x = torch.relu(self.fc2(x)) + probs = self.softmax(self.output(x)) + return probs + +class Critic(nn.Module): + def __init__(self, state_size): + super(Critic, self).__init__() + self.fc1 = nn.Linear(state_size, 24) + self.fc2 = nn.Linear(24, 24) + self.output = nn.Linear(24, 1) + + + def forward(self, x): + x = torch.relu(self.fc1(x)) + x = torch.relu(self.fc2(x)) + value = self.output(x) + return value class PPOAgent: - def __init__(self, state_size, action_size=2, gamma=0.99, clip_ratio=0.2, batch_size=62, actor_lr=0.001, critic_lr=0.001, seed=42): + def __init__(self, state_size, action_size=2, gamma=0.99, clip_ratio=0.2, batch_size=40, actor_lr=0.001, critic_lr=0.001, seed=42): self.state_size = state_size self.action_size = action_size self.gamma = gamma self.clip_ratio = clip_ratio self.batch_size = batch_size - self.actor = self.build_actor(actor_lr) - self.critic = self.build_critic(critic_lr) - self.memory = [] # Memory to store trajectories for updating - self.losses = [] # To store losses for plotting - self.epsilon = 0.01 + self.actor = Actor(state_size, action_size) + self.critic = Critic(state_size) + self.actor_optimizer = optim.Adam(self.actor.parameters(), lr=actor_lr) + self.critic_optimizer = optim.Adam(self.critic.parameters(), lr=critic_lr) + self.memory = [] + self.losses = [] + self.epsilon = 0.01 self.total_rewards = [] self.reward = 0 - self.seed = seed - np.random.seed(self.seed) - - def build_actor(self, learning_rate): - input = Input(shape=(self.state_size,)) - advantages = Input(shape=(1,)) - Flatten()(input) - old_prb = Input(shape=(self.action_size,)) - - x = Dense(24, activation='relu')(input) - x = Dense(24, activation='relu')(x) - probs = Dense(self.action_size, activation='softmax')(x) - - model = Model(inputs=[input, advantages, old_prb], outputs=[probs]) - model.compile(optimizer = tf.keras.optimizers.legacy.Adam(learning_rate=learning_rate), loss=self.ppo_loss(advantages, old_prb)) - return model - - def build_critic(self, learning_rate): - input = Input(shape=(self.state_size,)) - Flatten()(input) - - x = Dense(24, activation='relu')(input) - x = Dense(24, activation='relu')(x) - value = Dense(1)(x) - - model = Model(inputs=[input], outputs=[value]) - model.compile(optimizer = tf.keras.optimizers.legacy.Adam(learning_rate=learning_rate), loss='mse') - return model - - def ppo_loss(self, advantages, old_prb): - def loss(y_true, y_pred): - prob_ratio = tf.reduce_sum(y_true * y_pred, axis=1) / tf.reduce_sum(old_prb * y_true, axis=1) - clipped = tf.clip_by_value(prob_ratio, 1-self.clip_ratio, 1+self.clip_ratio) - return -tf.reduce_mean(tf.minimum(prob_ratio * advantages, clipped * advantages)) - - return loss - + np.random.seed(seed) + torch.manual_seed(seed) + + def ppo_loss(self, probs, actions, advantages, old_probs): + m = probs.gather(1, actions.unsqueeze(1)) + old_m = old_probs.gather(1, actions.unsqueeze(1)) + ratio = torch.exp(torch.log(m) - torch.log(old_m)) + surr1 = ratio * advantages + surr2 = torch.clamp(ratio, 1-self.clip_ratio, 1+self.clip_ratio) * advantages + return -torch.min(surr1, surr2).mean() + def remember(self, state, action, prob, reward, next_state, done, direction, pos): - state = state[0:81] - self.reward = reward - self.total_rewards.append(reward) - self.memory.append([state, action, prob, reward, next_state, done, direction, pos]) - + self.memory.append((state[:self.state_size], action, prob, reward, next_state[:self.state_size], done, direction, pos)) + def act(self, state, nash_prob): - if len(state) > 81: - state = state[0:81] - rand = np.random.rand() - if rand <= self.epsilon: - - return np.random.choice(self.action_size), [rand, 1-rand] - else: - state = state.reshape(1, self.state_size) - probabilities = self.actor.predict([state, np.zeros((1, 1)), np.zeros((1, self.action_size))]) - probabilities = nash_prob * probabilities - #action = np.random.choice(self.action_size, p=probabilities[0]) - action = np.argmax(probabilities[0]) - return action, probabilities[0] - + state = torch.FloatTensor(state[:self.state_size]).unsqueeze(0) + with torch.no_grad(): + probs = self.actor(state, None, None) + probabilities = probs.numpy()[0] * nash_prob + #action = np.random.choice(self.action_size, p=probabilities) + action = np.argmax(probabilities) + return action, probabilities + def act_simple(self, state): - if len(state) > 81: - state = state[0:81] - state = state.reshape(1, self.state_size) - probabilities = self.actor.predict([state, np.zeros((1, 1)), np.zeros((1, self.action_size))]) - action = np.argmax(probabilities[0]) + state = torch.FloatTensor(state[:self.state_size]).unsqueeze(0) + with torch.no_grad(): + probabilities = self.actor(state, None, None) + action = probabilities.argmax().item() return action - + def replay(self): if len(self.memory) < self.batch_size: return batch = random.sample(self.memory, self.batch_size) - states, actions, old_probs, rewards, next_states, dones, _, _= zip(*self.memory) - - # Convert to numpy arrays and ensure all are correctly shaped - states = np.array(states) - states = np.array([state[0:81] for state in states]) - actions = np.array(actions) - old_probs = np.array(old_probs) - rewards = np.array(rewards) - next_states = np.array(next_states) - next_states = np.array([state[0:81] for state in next_states]) - dones = np.array(dones) - - # Ensure old_probs is correctly shaped - if old_probs.shape[1] != self.action_size: - raise ValueError(f"Shape of old_probs should be ({self.batch_size}, {self.action_size}), but got {old_probs.shape}") - - # Update critic - values = self.critic.predict(states) - next_values = self.critic.predict(next_states) - targets = rewards + self.gamma * (1 - dones) * next_values.squeeze() - advantages = targets - values.squeeze() - - # Clip advantages - advantages = np.clip(advantages, -1, 1).reshape(-1, 1) - - # Convert actions to one-hot encoding - actions_one_hot = tf.keras.utils.to_categorical(actions, num_classes=self.action_size) + states, actions, old_probs, rewards, next_states, dones, _, _ = zip(*batch) + states = torch.FloatTensor(states) + actions = torch.LongTensor(actions) + old_probs = torch.FloatTensor(old_probs) + rewards = torch.FloatTensor(rewards) + next_states = torch.FloatTensor(next_states) + dones = torch.FloatTensor(dones) - # Train the actor model - actor_history = self.actor.fit([states, advantages, old_probs], actions_one_hot, batch_size=self.batch_size, verbose=2) - actor_loss = self.ppo_loss(advantages, old_probs) - self.losses.append(actor_loss) + values = self.critic(states) + next_values = self.critic(next_states) - critic_history = self.critic.fit(states, targets, batch_size=self.batch_size, verbose=2) - critic_loss = self.ppo_loss(targets, values) - self.losses.append(critic_loss) - # Clear memory - self.memory.clear() + targets = rewards + self.gamma * (1 - dones) * next_values.squeeze() + advantages = targets - values.squeeze() + probs = self.actor(states, advantages, old_probs) + actor_loss = self.ppo_loss(probs, actions, advantages, old_probs) + self.actor_optimizer.zero_grad() + actor_loss.backward(retain_graph=True) # Retain the graph here + self.actor_optimizer.step() + critic_loss = nn.MSELoss()(values.squeeze(), targets) + self.critic_optimizer.zero_grad() + critic_loss.backward() # No need to retain graph here + self.critic_optimizer.step() + self.losses.append(actor_loss.item()) + self.memory = [] def plot_losses(self): @@ -146,4 +121,6 @@ def plot_losses(self): plt.title('Loss over Time') plt.xlabel('Episode') plt.ylabel('Loss') - plt.show() \ No newline at end of file + plt.show() + + From 8f6e683559ba521346cde3e4aca369b2949b9503 Mon Sep 17 00:00:00 2001 From: Gabe Smithline <63967324+gsmithline@users.noreply.github.com> Date: Wed, 22 May 2024 12:14:36 -0400 Subject: [PATCH 3/5] adding code --- .DS_Store | Bin 0 -> 8196 bytes Final_project.zip | Bin 0 -> 7059581 bytes README.MD | 31 + RL_Final_Project.pdf | Bin 0 -> 504488 bytes __pycache__/dqn.cpython-311.pyc | Bin 6522 -> 6521 bytes __pycache__/main.cpython-311.pyc | Bin 18687 -> 19914 bytes __pycache__/ppo.cpython-311.pyc | Bin 9929 -> 9936 bytes dqn.py | 12 +- experiments.ipynb | 215728 ++++++++------- main.py | 151 +- ppo.py | 4 +- ...naigym.video.0.67326.video000000.meta.json | 1 + .../openaigym.video.0.67326.video000000.mp4 | 0 ...naigym.video.1.67326.video000000.meta.json | 1 + .../openaigym.video.1.67326.video000000.mp4 | Bin 0 -> 12636 bytes 15 files changed, 122466 insertions(+), 93462 deletions(-) create mode 100644 .DS_Store create mode 100644 Final_project.zip create mode 100644 README.MD create mode 100644 RL_Final_Project.pdf create mode 100644 recordings1/openaigym.video.0.67326.video000000.meta.json create mode 100644 recordings1/openaigym.video.0.67326.video000000.mp4 create mode 100644 recordings1/openaigym.video.1.67326.video000000.meta.json create mode 100644 recordings1/openaigym.video.1.67326.video000000.mp4 diff --git a/.DS_Store b/.DS_Store new file mode 100644 index 0000000000000000000000000000000000000000..a0f02fb4f000f3a6d2c071f5743a419dc31b0935 GIT binary patch literal 8196 zcmeHMU5nE|6uq-sw<$%butE`pfUmW-EcF8)?AER#D9BdyK_w<_w;N3}wI5b3g*@wD z5q$NR_zV0WdhSeIk~V$tMO>YMxo48O=Vs<)lIcx}NVI0XI#HR3EL6r)5hkN>Ki9Rg z6`wf=6yOte$wfV;J~?aN=7dqeC}0#Y3K#{90)K-7*t5A<8}@zOYphYgDDYn@!0!he zm9ZIPq*1*(U{neK%;B^uXd@3$96QEljFCn~MNFMNh?pwk6hmk_`fYXxHe-x5YB~u` zClOy+#2JcER|n5l?j)KTjWr4w1=bbd+`Wiw-Jub^Pv!5<{XllTKzi`thw3{XD(8nP zD$E9!H=m4R)>m&7_>*a^_DyVU=g#cpt-MvVUUZJ-yfbyDvq{SxKjWjvQU>0yw(C9d z`*WwXdq;*-*AM$+MeuulxIBOAhdnuO$ywN&C~il0SOu%#l*)@mtGnlk81oR;#Ucs)yF6H6IvT)_qH<*0~rqB!R#IC(Hjb> z%IdA?^#%RagkI`0}ii+YSTkt6}G)H7M^*MQG{kZPYLWP4F*uIo@UX#wH2Dl`e#;X=ImBh=q|K=fYp** zIgg-Xa^;Mm=}}rs!K)#82bwtM6(4C-j}DC5g@I=Ohc2cV1x}L!+j_8N&i@ZKfB!#C zQ!-;U3K#``Qvs1{ciJtG)UKXE=3U-KeTmA2{UVKu0+ae7fa8-t4AHjX%A975kw%!0aY|21poj5 literal 0 HcmV?d00001 diff --git a/Final_project.zip b/Final_project.zip new file mode 100644 index 0000000000000000000000000000000000000000..2ac449d8f4b2ac7ce4265373b9b6bd64f5f890d5 GIT binary patch literal 7059581 zcmdSC30zWJxIXM@vQZmOnx%4@v@*4vaHdn4S*dkOtsKZo%n6)vgv!dw(#p~hmCDk} z6w8zph04^-DMwU998p9-MMOp5+u1qie)nED=Xd-6f8W;N0N3#kcQ~;^PCnn>0#*q86aYUVVaw!x zS-}5AP7c7z$)!0RM)Vo2Bnh4Jx?-N4&@zS#?pQhewkO_sK9P!DwfL>UoAzf1lb$BN zD;EnC_<09y+PAc9ylYH1H`b@$ffd~ASiS7yLDzukgU>e>p3Vz(P4x~5J-kcMKatcjf@6rZW~1l1s|@78H=4wL)Z4D({VZmp&S34cS3-Qs>GEd- zzPOk9kwNa;;L33=-UaXGAMNJ}?e`R%>D^O!2J}pP;0dvw`mj^|D=X?eER)7HbnKK!rR>KPa3h0Djo2iqE6n;M+vRv&kG+fdOjov-#S z;Et#GboQj)IGyy~b!n%y+l4(Yzt*1#Jk0-PWZ6-j+rPv-qD5~r);mYt;`dBWFM05W z@3z3fnoFg7CtoXyRMEzN%cMBAK~0+y7cr$Y^b&Q zL&_QZ({J6B;vW6#cw8gu4d49mQP$mIy^ZtIcad{meYza6mN;=VuJl#@&BdzA6-2w+ zcxRS9xN_)4>xAlx?0}b#w|fMq-4RgSKOX4%Tp#}%=(EGQ#i!Pmhx79~i~E zu6{ALZ@i#H@dD@s+BiA#NocEZUpOJSccAp}vT$&&mA}<=bf7LKQ~T({R-dpWUgj3M%DWJ-m9Lj0&RmB3SJCG z_y)_BuhFm_H_<-X{aN@fr6GNrpp0Lb_q3(#7n5}=iU+|>M|&A9avv+O2U`5zUqXM3 z(_;QQCW=s>^);B%Q|QACZ_e(VSiFhngJL=h%rp=-hNK6*_Sohdn-jtUek-Wyor0qF zt~&Op3dgYofuG-j zpZ_vXjnE>4HrPby4oT{>_I%#G`(8rI-nE<#_U(spo+Eul;X3Ldd%=!vtN&=SU2Py2 zN-*!?k{o_}2Gi3zK2($5H&&7H&^?jgt9`=5b#zPEv}hWAD?K$wR~KpT7B8NF9LRtm zC1mGuZwUuR(~b@-G=W0GQ71`aSac|n+~z_yrFOb;iQYj@?@n{=>+JX;5E^r_ffwXQC#0vzyp)U~DzV2C3T>HP%~8 z@k}fyqXR6`({+*b&Kgf>l*oZM0Cwu*(5WnC;0nKIyX7?FNpz&JoX=`t@!OefxJ8G+ z2TC?6E$vKz+Xe1Le25F8XzK(gdcdXrVyt-`W3;&Cu}stS+h=cA{l?Ik^g3_jmU znOYkncrJQQ&5tWvN|aa35l;{G7E^23;@*+T@!-glgSExZ)9hOJa%E8-+ZqebkS~jf zhz=X4utca*H8vcfkWe%{In@gznmBVql_>~&OCl-|dC(FvI|vab;OJ@+8Y48aQ0)IT zy%SPRMLM}Q@y@ijFtg!f+fJjRBSV_hO5ok_Cmn^<_qHW3vG4%OR}q&nOYOz78N=dJ zcExw8X(NOzM3+ly=?sC&$`?nWAdARkXEC`;SKZ=tWjkX$FnSxii_nRaA5``fhRNF@ zp`F~{1??FjX%b;VX)8wvPGO8i%4_$Z{8=$nkT?(U|bCxXb zXH1M#HxXDIqe;%N74kkkmq7Cp1nNbb3?mR|bDA4cSybKNN$3mU&DbYawSdqQRA{NH zXeJw;ty4!l?GKFz-fat+_Ku>Ph$la?jeHGeX5h8NL~&vk&8o27ZMJfT>1mJ3i!_<} zz{%Il0NpCa&v*()s(1}896v&LurO3GGCh>9$t}K1j}o;fde*Dw@^uMBNLQ481Wm1) z*I!&Jx-Plj*E(jal;$PVnI2o+2KP&yR?V zc(>I^SEFyNOl_LR7|%wb>Ar;-&=5jV16?F!TYJmH?QqlUAra)p0+xk-O4E*A;qfWf+cA}$MZkOGH1+!vb*Mfw)R@{X8u=QK{27c~_@-j7XaB%o*vy#ssHwXLT z$1T7b7P0*%twwp$b6dxZxKY^|0du65Oa~wENs9EAf&Qh}33o~6+$KbpJh>LdX zKdiTgBC%-lP+m7P-)?Ht5(%7 z7w&ETx+V6JP5kdz)fl5_$e$lf7WoA(@91Cmwrcl<#B*OYD$m?gjj;(_uEE551AkZT zez@iDQ{M|Isxe-H%QJyOCiK+3bGvNgKNy%SYU*EiByp~=(_~S8|GFCB%HjAAJ4NF_ z{ZoOmH3y-9 zVjgJ>@9%z~xXk_l#H_SrM7h+`LO4FUFHk6sttwPp4-r;NetXX)WO=g6l*~&zMkYmy z^CuyPgIfKlL*u~&^yGL?*X;uk;yafFOQ8;gJ+4SLIpLP-SIAP1HQ+8H?eRpEWX>9qFDdl=M8~g-q3C9Ft_&+6zBs*Rs%&mK`~|^wdsHE7b|1 z3&z{)LS4D)Iwf5e8@eXAGl>{wFqhLdE_pr&wDF8E-U5 zNM=a}$1-s!{$vX(GNM6J^C=ttxwx#;&o}R_gjwhWb@kOzWU!F~y&?*W$9XMu8jQu= zj0}2BV?#?_p-5Ic&5c8}c5_X|L37JIL?4zVPt+HZ%mRo#G`0|AHI&{r$yoxmg{G;a z+D5!2G4Og}eh>jQ83qv_l8DHlX>io0s6pJNqG1IG|BAxuvSXliH;vP;qE)>23& z9$vhm&)qvJk{1=6UsPgDJjqp8YbTb8N>%Ax*2!6`gwJ>*3ae~mxG_o*#{yBfZiwQH zQAJT4WpWZG$(tJG;kw{LH;hx3V{Cp}q zG+L}zFid_zVRDv}%&sD#zN9WPJ-0He8ZIeDM>C=Ek{;>;!N$k)!T!G1Hzl-8xA3ONNl@(PL?8Z+YC} zd`U?cscVGP*C1-k6_+F5o;niW)d1!y4rU{CJA_Tb+zIx$Qx?%F3r`ZC4Jtwvb5y!$ zXtHT(30p!$i%KM?S|~KXFdoB!tU?qg=GTTO@&+quK?$Whza<$oG&WvmgSmUNB9dAQ z57o>9zBVjgmSSa*B;*-;x``+;7k;j8M$_!@N`0}yC_+(8@tCzEo(&KRsk!>%^Dl|9?W3|x*W)PLAuiXJJXd`e&)Z{?6IN6zl zT4z>*q|=HO$r(uT%YA!1ABl)gP=qkZM;InlQ1(W8AoXh!o$xdCHxc|{dt?@!%|AtW zjte12@TX(r$j&owBmGw-i_gY1fGWC2Z8Mo=$e@P1X9Xy-PHw#pqF(rm*-kqS4cgln zXq;e{%JT?X7nh)#jjio@o*uF}L~@-ol#VAC5wp7_q?2H-d2>8Hz*gw%2T1El=iX|?u$bAW5Xg|v$D5zya)Ee^f2{G7{xQtJAj~EX@=D;gJzmFWCX{$s&!(9$aM zbTb7+QVb89rJ3CoWi^-`7k!*6eSu>3A+x$5L8I~g*C;?#k?|dp z#Dj*&`XtGr$EaN;28=P_x)9L>TvuWso^6QF44uSt&Y))~;YWHKRNMGs@CPCNNK^-H z2|is7HS$z4vPV4CfU9Jwt}m)N6svBT*BaKZ&$Pq7iN*0XpE<$8g(4&Q6Ijm- z<&4SExAcknX^=A5KEc$4n+%CV28ONa-C#304YG*pHVB)X{_1KFbST*Atbp9PMiaZp zj!yjeo}-w)6olpkQ@&okhnk#vIz1JX713M08`YJN>fd1DJHfP8_YaIfi46pIJT(WE zp%#9)^7^O=udfqjcVDR$o*BIe#%1BmL*x|eALyv3`%cLXru$9ZQA^iyFg0)b`iojb z`jkppayn7DLVlTo5BlAn{r}}bZkVdF^J@p#EqqQBM3e`4jhYIVaAs=< znC&ybViQ5J25Y997ZrX;8wnj7jYwa2DPcte$2-3ry{hDvqUC6NRcO#u04od0O+n9$ z*7BydiaxA#;o6bjwW*197^5$72_*A`N&bf$lTEpikz+HH9Qo3&;4EheiQ5yoM6oWN z9v!A^%)cy1yrld_pc@Ktg>rQ!x$)>U7C9%_n7AnPg8w>_F&lm7nb|X3La1j}vIx9c z0gdDJ*OrUD$b}!WrZXek__mS^IW=5etJ=X_uKH|LaL7!p&>k!H%#jx&>gZ8nPdL0R z+5=_3D*tl?fymv=yqwp-g3q^1yM;@r<1QNZe^=~WK2WQig^Ra`ffn-wuJsIH>Hf0+ zpjzEs_!`6vu4%SsG7*G2$*vb#tP?S^AZGFVn0=Gr4EigAsQ$hx>dhz!oa}@p05_o7 z_9$e?>(Z_$!(mDAa15~AsmT`5mzD~NF9n;&%gP>wI=4x#R@&n#GE5ElcINcz276+Q zWO^1t^c41s@J-K_Cu0e?i6_RTs%emtXB=<_5=&)2 zhdi0y&?UrXI zXyUlQUd6J9E@n@($8kr83aJ(L8vByPqC}*#0A*&08bIns;yKWxhoDnJ+mT^Kdny9z zldkD@YGyDxGJ;5T?Q~tKm~MEIE--Z;=E^Hm$J^QSwW8XlI18skc7+$yv$&U*tIS^@ zw^)8em-ogruaS;iZtP+sAe$QX(>Q;8rn$~Im>0W7`T8L}ztm#0`Vz$qB~B#63)a_| z#pL3wA+=A!5K43pgk!mV{*0Y{{ti@%gV>7%{H_tumx;WP3uj;9>di$wLbFBOek~6$WBjuE@i?{YI_1C6*C%^N{%x4&u7vdR<=YR5wj9 z1+Gena5fK+=R&5tHixE+MOWC_Q@kXTB0i@!!a4a}P;_KCrt|vdy+{(J`#!M|y6e#X zx`b5R$Dl!7%z(bSaPvj+;zy8qLkV&-|M5`_0?vbWh%fTd1jrfDP_^Xc=-RNWPXBSN z*fBaoNhQt0k1qX>Qxu`@%WL1q<-iTil%1hTV?%v#GUiWlO(1UgG9tQxf`U2gRrJ7JlmH z+_BzqbS-pjE%fkIM>J-_=l*@Z>+B9#9CZ4@e;hGi1pO3iC}hh8+pmtDHTc;tbH{qe zskusVvnlZ%-rjEtdts;nr@4cY_i?Vu>3&LZhvV@zgtcIv7I}Q`xVH9tl}+T#`Gk~| zPdV=}e>G+JYILrO(zIb(QA_g`czNxgZGFMJU!AK}@EOl+%gNM0YQY)b+j2s!(E47> z3T3N_Ff?p%4$%Lzt#O}|J#)3R$9+trvR~NHZ({%MS=rLLT0VE7P1DYPwqhw=J3QNw z$s=^6%t{TC-#}4ZaypSLy>qg0hH2!M)_#Qa&duf&YM`ZeZV*vmI4z@d=d+iu=W0*w)d04r1uR-3P6<1 zA5F?X41)Cbz9;3UfxVJC;6PHI@d&UVOiJxPFhzQMb4mGmU?sAWlEou1el#inFyzwP z`<|4a2KGcoQe1&KMd=5VqWup{k>1{1QhpxT8(B#K=9Hu#P0BwEx%BqFC*`Msy_A&{ zU`}!T!KCE;15>28HY_W6V2g9$o z^A^X=9UB~1D!H#xa<5)2yzu+QuPN(tFTt-^ot!(`@-ENSaX!!%ywXWO*QR&<(1Gul zPF`7|_1%NV6A=Bnz%w#{2H#p_NCnnD z4Kj3Kx&(kfZrV9__@4ZIu0z_Jd=qU3yHzP2x8%b4S_I?-8lE>oZ+!9Za1%r3@1GAA&q$ z>U8`;-CE@4?_JyZ@!Hz&!>IQW)>xGCGC23$hCka{W%TN|xmro9j1ucwq<8Ilo*xio zj*K8d+KX47mfc*(|G2riOcu({WP$X7o6DrEAc0K!{X-_RWeoQbQ_54RfmmHhJKtEO zAtREJjD^4K{U7`ExB}3`Ru=PQhVuUhOnTq;w|Qx#`N-me<>{5by_K5G?>&(2^FSNW z@&6UDS5o^15XGah2=;?Sk?wOYU$y^0l%EGCeNgweg#$!!lQC)h3sGvN`8Bea>|}=O2jjbHHq~f(rUlf~76@_O>acsC;aF8nhx<%Ccu} zN&%SkNZPt5t*#~Br?F^Jv6K~96VrwL?tceR=3+5!F%_6uWG(w98!pQ|mO-YUYHfPr zhnC$k7K@vFzb&9;ebfJCmH)G>O`V)&OpL%P{~wm!hSG00R=H~ecQ7_GR>^ZWg|e6O zGJMY6!X*B0fHg@CQfXCaip^k(^xL9ac?n7j7XQN?hjgEFcUp%3D`44DZ~0cadjdo` z^@Bu_?sJYP`BKN=d%k`e7*A@D08!EbqBzNzqyB{`(tUpKI9}=)%;oE+ffYQL8ssW> zWq>GFKSmT8KIb|fCv^|}Tk05m&)3fZ0~VNSd$bmt{r7)}J-nE@QQu{hN@cOsG009ATmouN z0ldy}AS{?vJyvL{(sTD6^93|>&xQvP7GBiJSvp7N@Ov~S?L?L#u6 zSSt!ORgdnn7^z2Ar@P?Xa&RT87VYeM3Y9GkF>mkW5clzbJx&Z#u`A9k`ZsBdm7p1{IGgO6VWFB)>P1H_48O z4CHYJ&d&r+#L^;!l}&!=10|}_p=Rn}yu<+GA{w(Mi(0B162SCZZU{KA*9lDJx0mPQ z-nqa@ZU|yNgHLQ|XL5GAwj&OF-^LM1zB~qOUsJhFB1K-N*+{CI0Qng#aty@!CT05_ z906&Io;*Jhb4=hMd5W_Lw=C``Wh6~;6Q1oO!ZRSm6V|}`LN-#nkU}a-P#D-j=X{%8 zZ01*v9rfe!sWL5>@zpM<*e|P;2X{mZzI2Z32}+(|-4aPGL#EluMAB75W@Ou*?VTTc z!u=wm1DFpj>_9yl(busPJ>E^RK97QHmK~epF?S-{7L;pbG4202QSC}dXpr>K{CUE9TMc$jH@E=B#eSA;?4 z_d=5T?Nbp3FlNKLZE67rY8%fPR~zq||6yWX9q{pRLeEZ%YdoOPbk5j5p~u`}J#XT| z^84{05E07<0nb32>FQvI_O+a3`H zNk9RJ;8E}=ulyqkJ$e?`swdXz=s&5s^XBvY>WRR7(6hth zTK2>`1)x!6M8Eu@%2&H8r*92bs(ek*Hv&CZJ4X+wnCM$RT&dEMpuZR7nP*XVA@WgZ zGsh!-Z>Xa~|AdCwhHBN4&KhsvWafDVH~66>tlQc;eBhc_mpYDR*QD;iU`{1sQVcPt z91}4xLrgeeE*N4ktw-6jD)oatYUpqG)#DK# zcdVM9gi$iWg!cdtwR+XDnk#RrYPMl(?gM5!w&rQz^UxZOhen1i%*gbF*)UU29`Pvj z6Y~yc)(F#%NW_SZFw=mM17lSD6EX9^7!$zMDYV3!EXH^NALYTA8-Q5?#=HQ`GR2no z_R<>F;;JK0ur)gYlYy-<2h0s@P3pUs z*wvB`+9PBO*5&grjc5+v$S^3L<8Ly zp!M=6jPws9^#3-GJB|_WOdB1AC0PPVF@s@Zk}=N|TP#sG3ou21r{`*)Rma~D7If4& z&+}twUozq)?ZING1dL4tRygU!5oWa`uEOH?aY>^fkKr~lOiw!<`%_Z6JeA=mW3sXelWT4NCK{2N{3_rgNk zVX$Ps|6)?wV3{}kVE@^WC@@B(AM7s!u!Ip@bz}{Y?i{_KU-j~QYu81UXgn03ndjFE)VpLM~ixHrkr#9^?|%&vwHrMs#RJk%kj^bDD8|t*)VZk z+uPCl)lmOaNqrR!O%f^VT>m0Ce0le9-Bbd+VfVRa=sbG;F_z9v?tu9){vOvW&JhG{ z*c}Tg9);ua^{qr2bCQ-#p*HDKJ`5C~c$t!cyDU+5J{v|LzC#ILXY2A2s(#cMbQ8=Z zYKc;gpTTWPYZ5*uO#yn#$_Lv8DjT}F^OSSb&b|UG<<2tn*nsiT_}toiejBv0VN?a> z3k@BM61>n40OU}VnafLhUt!K6f?xOY8#*1)l0(AoKStd2k-|LB?(LRdj(ug(6Xz)H znu+bRV|~|hAK9l6V`lIA^d*Ms#i=Xf_&9^6s^ADcv*9Sq)R!QW;}DFlvh7>Y{K4*S zafK(Pdul}*xQrE7cOCAMS=%Ryo|?u1TNp18rz5~Ku(#_%bc#!hyQ0OTs}#kI{Xr6C zP0k^S8hRs}0u6Bv3FwkEw$IFD$FUhRFC&>MB2;7tdASG=Y;Z+cXW^m@#jIJ|w-PZK zjqEHsF2U+IoTAF-jul$p zg^*jL*rX#fhKIh)=%Ci-m;a@4oKrgfJ`g$nr>^aSbY-5kpP}I?8Ol7r>=!qv(q)I2 zoNaA;rN|DyDf2qLz4yvc=8ZTWPb5kg9VP;b$|0#@MME`GOY_tJq3Ccc%1@>?GH!Ne zh!wlw|GMnZ(fKD)RQmiOHIUV0yT)nf#2?gzYRFKo_?V()vx9zlTZJy8=l8 zh|=_fNy+?&q)2aXE-60`%uxnY>HwzX|6o#J|Bw{v?ad|S=YhG&N(#Uf-VY{)@efIn z-rihNejeB{8A;Irm=gDcNs*e+Zq_;Phl%EIY zAR{Tc08_GmFe#4zkQC|d%_Zfhfn}qEP0q*{FG)J{{U}>HCxrC9afYa?e5?eppV$9IQy|0VoH|vOq!N%K{5-Hs z>3WGxBL<0eGN0`Ji%wRC&u?s&DX9d$#Xk?sQ3g@!0HWmoAW@|IoFfWM`WVdR>!*P+ zqzAca#2g@stS7twLKGQ3=Q=J+QV9^{=YbuQK@=T;C~-eX6zM+ah$2f;`R$AK)4-Bs z5k)Vtu0Yn4-G3p944-pEktL}Fi1PEm9ApqB7a)qPC%galt4F%eIifg9AA`Ak{T#5& zrlO7jTT^Mvy^X%|waeXQ$r$ID0!%so{V8+xVyxYuccmfIWxljMEzIEDP?!H5KtMqt zefy2o6+eJ5vNqEEi?FTbr&*iIR_FfHvRk?i5Lo2{drf3$BmZ@k|6{Vl|NCoGnOl@-lFFY7<|Vanaf{8q?Mae;TXdl9Uv#q4 zeF8JiKS(NpZ}HCqbC4?MyPlU15JlD=n|~vUbf0rXktM18mam@%7A}h@^-qIjZq4}@ zqDc4oohY&-l>kwG8rW%BGfEsl6j^(0{)H$qe9jR?mZTCO%FhGykUlrtg%8$X00 zUqjJ>K@pP!Oe&vK4!mw$R|kv$-jb~e=DOKs>RO}QatOeCvL`)1NqbFp7_WAS&zPJ6 z-s|d$a~+wvQ9Zz5@e9p-+{nNy-#H3R4@9sDMqp4e?yW#;9n?bgZo?3ZsM(e_fit(D zy1)ySC9hF4?TMh^a}zN%PvHH(>q4M>d*IAqMmCKK|G=DNbJ}?gp7vcixG0rrK%k`W5lU=MBFr~?8H(m8?ELu`4@Pwx1{2B z@&6soXfkffdEiLi%QAPU} zzmTTf-rJ;i;^B$$-l*p264!)QiYG0A3W`4g9_vg?b4!O?18=!>O~q%pszi67%}soV zZ6rNjMuT)M)EQ*ZVYrOEsb?K8*yx zECWo2hQ-DC&o)=Qut|taR4+5KxaP+Q*=%tQIA=t%O^Cv(PtR!{zbXHG-DkoIn0FCP zn;|v{vwN7=wNsxaP`2>);?$=vHSWgVwRB6I7)t@p?O{HwJqW_v%!JuCR{Eq^x;0IF zwg+HGP0x}uRWKeGlKw{gdBrwicBjR7{>10{#?6rLy2t$G4ow-m4<}UYVXoT+C>@jL zl#YMO9v=lSub*=WyUOe{JZcJ<4D)XWv)yK<^I=xPxOwz_WoAanLC}EJ%r?dQ@oj(F zQ|>d{bb+Yus!v-*^eZ#(LZ4crFyG=?MJzc2!UH|gIszK-mr*#)W=8haL?Kc!qqp_-Ndd4g;JTWw3QRy%-6j3Ld)S48nIDS6{o$KZu}c)2 z<1t(Y?CaIzL!GLo`ZwTu`9m7|x8Zs`z-)!<1@!{jN8lu2KKSYBcWUUD`{@M%#??=c z*P@~C4wULO^rwCG@&R+#S5Lo6L;r#5OSPZ~Smq-@j_8OnS&Aw0NyJsL{ysmw#99q~s;^!XVCcSjYw9%g`+W6|*L_pQZPaH15fyx?IkxWkH}j#o#tB<< zrJ$@}&qd$8e7 zk7`S_zQVfJUeM<*eT!j6V=-BTB1GsQiJB$~MRJ5RV1T2!rh-~lClVh-zpnmVE$C1` z=~5b|i!fLHoB5`;#DP8o3Q-~;wfx)n86X6m^YI_|u%A1?6xs%GV>TBMW=B|v=D?pX^-;-AIhFd|v5Pk#HpDO|N=XE2V(*FPC8}a+w zzc0@L0dXjefN2LZ@TbWnV3r5~b~J6(w+DS6>Cba%M9qmi72MkY?5%z&Lcd99^tS;j z=STjPobU)BrGUP*qo(HAcQ{GyVHaKGNpQm@C~TM*_6-UTpR$>yT?j61CwI7zI}rBu zB-PtpuCH|Kq0u3)i_B+l1~^GZ@?*2t3qaVak@MzD#kWB^7cvv71^iea@A%VA{nrf( z=u{(H62Z_m;Z)Oele4y(nz54fPKlt5?Em`-5_I*g%23$O!-_xN0O~s)bA(+)o9v{Vx;Xxy|9Yw72IMms|-z?KQJ^;)K$p#zYEAE zJ3jS=zB?M%Y83pdg0c@FqpT;0FN?Zbm0aHn#SdmbJZmeB;!L{Nzds(OhkK*PU5Afo=}LzBQp>O;*MEArDn5egNmTBxF*g!1cgYgj`l2uL^lT8 z>*H_}r(%(kltCKjwo`t1&>LURdh0~y`Boi!r8wp=hu%AVp{vG`_Gau!HrQ4lt`wJO zBbg4gOyB9K*++Wq$EXc7wZwIk!NcI}^mNs0=S^LAIyHLmEP?J=M&_is%Y??p8iYr`{XH=VMpXi2tC ziSTBkC<2|DE`XHpX0`4jnyJ=CnWh_MTai^jjtyz{*;Qx5!)D6QH*%rKr$TGwe0|n| z*)%> zNM-iKRlQ_7SwNy1)Ox&-k6_3IY+Z3srdfWvfS&Dqi)j09Lt&^n*RH_^8K7R5=|vm@ zhlFf1jDyo1s}l<)QAW;3iZ`4^8->PB`Jdp_#TQa9dq`g9twTq4g?KgxcdnQoN1YUU zG}9kEjXATZnBv>%HPmvN)mOV@TE31lGKR9E<%yYJCYS7; z+2?^Aw^xMLWbsggikdJpE>@$K7V>(;?gK1}ZUDWNEUJB88rk`Tn!}O1l?BTEQ0QFd zZREp|e{8m@GO>LGUP6p}V%De8lvvi6nFwpQ;Pjx7SrcKeA9kiaV!xr%y#x-B;c6?v zEw)H5)cPe+qe8IZ^>AJDSQrX4GwiqVMDKB_ujoN6oF+>9m-$@_G0o6mPxoHZ>aCj=+AzodRFucE2J+S!f1;mG)4 z#nfPI05aOXn0NpQ@lvE3U=xzr%(6l8{S;7= zc5#LotNuZM?K>np2#*ph4Qix6RPN(TRO_EO(Q`Y3HBS2OrDNZ7&^kCxL#J${0Aa#{#T zEasxeGbA*Tpb{b~<0nhHAl($nI29)veB&F_el{Z+e3jVDQWQ8%rv#sUy6TkkcpTAPS23>y<`<=`3C-;iey!#p#+nSj zv}``r6QG}>cjo3bkB??$jDgQ*S=+2r?j0KQ%@!muR$m(sO`XzGSE*bww!ie2K)2wb zRdMvX$R#liX6?^Tp5h)G@pa3CgRq{*^$n*>cAA;H zy}J{38Ybf_%j504`jF{8*JjwW$4g^}k2QO5-G9=VQ?1C(YyP}6X9HYq_G!_1`wU4sPX0I zOHl9Q+nU_<4=$fFeb}OhQhk25=k8@EiQUSQ?MXwlHR(m-E%slAdWA?-e29x zptKzBDXokqM%KU1G+0+}%gm`?(u9DNbD|lU`#psLm=?oXuIdFYJE~ggG}(r_otE(y zO;)GyQdaI==9^ium$XRF>CTNawAiEB+rk$2-XBQA4X}a^yT`*1l_n%#om@0tern&% zQ~HBVyELFhX@VVfHH?54PmXgOu;Kfg+|U2&b8xHUOw2ip!tLS5P8F2?F^v+YP8~4Z zRsCE)k!%UCYg{-svHz0zSJw6G`pXvGR4%JfiVGH*L(Y{jyV6=>84ta3PGtO%d#3il zgS+P!C1q}ZwQF&kJECBXWkH;k=B=+sZ!{QedbRJx3HqH6<0Ydsg8jmgjfbh{C)2#3 zxx=OX0dcI~W6HyY?hBSDcw16F9i`emkjDcP_iwHKub3B>@LN3^%f}X0XO!K?Zm;9p zWC&h=9EL~kgPg`oz)13OZ~oKML|)zC-SW|sp}Wkn;Y!mxAqxn#W3!z*HtPB1fit z+w=Pd-z?TzzsO)kkY&a3xIQaEBZzeGT5?9G-Ml*g^KUPo$g_XZp0TVHeU;x{e|(Mi z{8!`qJzBf6Sg+3+=Vsq{x3^I|+V_->NCxAQ*Ka*(v&aOye-sn!_hL)=(jeRMYnW5( z@RU`$qH{Pma+@(Z=F%;QZcTB(5Mt?(FY9b@+Y_VS9t^ft*gsGI2&wf!@5h)eF|%*| zv%%o=XXA1XEO&m&>tGWcg2zp^x!=|}Ei@g?bKH-wKC*Djph~T~+twQ}`Y+3SLN~UZ z*7XpKr{L14laY&%-Jj@B-5%V$X*Sl0`NR17C}|`rQvc%ou?$Yr-t5ku=MP+j5dO#= zOs}Q(Kz*w;%-?tv$hTDR3T~>~)t247l2CpB$ggLZLeI{tMpN96A%_v?DIVNtN}?_tahl)8bi&yY`u1F_W)^4QL0nW<9*bRp=e|PD^%WTm8}m+XqPsTT zJM`w@LDHR6-08=&$oaU-m#C3DTu1!RPi?$KNrEl3YD~Cq+z%4EWETHMzLQi=ke~qV5hgh&hNIi?S@TTQiawblmO+6$5pp3 zUZbAaT{Ljq{dLx*%7W4eXoM-V%Dp-2kBMI+V1Y`#UhcvhSR_MST03!2N#cDP44-uDI|&OW^gR|^OC$P3zSn-YxY+k*@nC1RB3OmLOAny^;r6i z4R_T{_UN|4mL)6R_ysS|S^4?)j$&4w?xh12i(J?t@bt8I-e!>NW{C=8%b-ijcA6Fb zk{PvQ&&Z4Q&(AaT?RP3EDpYPwj2bHvU$;HP)1o|0d=a^4=Lsl*ynmzCC!WAc|-70=&tLC z){7l2Ei2c_t^Q527djbybmiL>GaH746ME`l_jtzXM;{T6@6_L@z5mKD;&<0RrX1xv zZ2Q~xeHn3>MycgKkcGb@dTV@ z*SdYd^zr%rg%f)enxDsOd^&is&$xQ0UvN7p`$GKMM8>%5ILupr%b#9#*Lpo<9W6%`Cc|acS8hgItFl1c6+q;5c2)r*pxrYPo@q zduo(@p8AQ;wG(RIE>H@$b$a+H3Y+-bXBz^qK7r?ist!tOIJIv5%*DK4<GjFrX<>o>*~fber1n{B7O&*u48J2c3cA8Bw^^XL$Oi1sU?@=DPRH71}nT ziQ>aPxEvBHHOQmWcPY)XfAs8G=El+3J+~*hh*r1TrwUbH`8;ZA>Irf@XuER1Q)`p`a^p= zlZ|uGYdh|{*;0FH6teAU)e%$rxs9&FT&2?3O6&NG&eejdW*t7M^lVJ;1k=mrVb$da z{lRMe-JT(9LA!J$sghZ#RE! zzo^Q3alX~ZkyY3pt9un&x}yU5lPMM9lWJ9uwJ)y)-KI2ohj+L6tBtDaCM~l*w;@!R zzGFaC?B;q@TSfcT+lsZ`qGtx(<4t-B`$|-&ao0|NNnf$`x9tY>*iHB5p;4~CBY%}k z*?Z7bzEHJ-Q5*I9)lL3qiJnwiso*q5{DpPEE$i9}-eND!D_+YFW^6XIn_UL`4db#$ zyru8qJ%w_v{P+oTbW~5#Op2?P*{Ahqp0`J&Y43}Vc+(yHn_Agz!WNg6sk61N+F_&3 zA1SJpCacQI;FQC)h(&|5%Uf-C`QXJ%t%w&FhjkQ9?MZ}jZOyyEz4L}=ml-%+eEtv~ zNnc=WW#-Di6>p*(cyMLiHr;b_M(4KhhI$-Ko^cPTSb$P1oNw>Cu!#8lcU~k7#JB`0 zRAIbQ(aZLT^z}eA#f;5^LmCb*n_M>3cF}7>VFUz_i@+ zP4nv@BUz3Jy$9P27kA>&kIm=RKlNSoqH6HS&4EXOAWDyr%0I0$eeMs`W>8+u9sf64 zsMUAZrf>dwZM=Gm(2eyp`j?mCr+W4%z4s*UI^*lcXXJQL&0*tXX0-1kvr1j z-;#a9=qRT9MP1|JHHz!jFT3isT(#Vz>5{<`L(|gQOIG301NiWQovG2WD_s}Y`{`_x zb2!*~W+6|0fqU1=MJrrUzrbf)KUXe4n`SGRmv*>$1YmPJR#d@)D;77vGrqPn&QYNQo)VV=U)jcyf{ty#Dk9bu-(BQ~2sSbvfcy_Ojjs#$J^d%|9QN_?gHF7Gwggz#=7gj<@&vjR!Z-7C zq^tY^uHeaFc%;&)YRMro7GF2Us$R2rwgJ~-AMtaKP+KaN6K0h%)(-42JpAhf|Fsbn zRM%b+aJ+1_7xH$40~sADLEQTlsv`i;y2R2WheXVX7sD4NmO;^-iFkH9O~)0T|B%s|oXO<&GY(O7s;DuS1lA#Q zNtVeJd(y6ao6_PjP>I5-|OH5hX}kd5i7Uq zw5Z=;P2NA-?7;|A74Sp%bVd+he|~`)#q#rWWaOcDJkBGjn5(a&AT3*6IyxB+Rw7-{ zt^wHYkKzPQja(T6Vz86G$vM1$da3@88yo1olqEv z7eJuULdgOqbPkKopRN}?n4fFNLkhff^Kk>g&i%62^f+JcK+Y#rk2>Cr`K(y!e(!?H zJ;>X9%BA zP{0{}+3eavuBGAkMZDwtb@Lj~R3FWEN}C6NYR5vpyq-brynI#6%VItF+vUep$b03| z=XAeR(Vl_cWb5Y`jWKDu068@9?fpYBV)(AF_7REHR|Pl_4YM{(d`-%u5Y7h58yob* z=BN{Pob*V~6+6$Fd-?+)(H z&!m9tfj4%d_xfsgP>s-V$hh+^09v4#dVi?K$2L7da9nQyLbXSY8#A%V?cOu4Ka+)rjXeETR6Xn)^a zip{NXYMhpr-`}%re5V^hPq&=^)~fF`=nL=7ne2pZ zoZ)ebnivG>L`~HEw+{We%Z);ksv~D#+d0EyTvbt#(tp=b>|QP_=kxdaT^_=hkhE}> zI`!I~T?Tl58J?faLXiTU;8T(ePb#~v+i;m&O&)+0rhH$mlU-px%q@C~C@1)p8Ep4* z6+=hTj=Id1GNLlcj~>pV7Q^?_y|0dbj)+#S$QH&BNUd6OZ)&(5Aw_-+P!@Knm= zNzH6{h6H&Z;9~CbRsNpKi8BT9uE1EzrXg%x`fe|TsZ<; z-((}=qW6I*48a$VU4M56f9kKgD&@FItJZ*DU#k7shIw)8X!=5k^K4aE?d~et9XMkI zL}GLjNM=9*-A7VmG*K1FuKXD=N+%wtrcoi6AG%j`u$B^`mb(vR@}YydDfC81(z;rlu4jOkYz7$LV`5JZgO9LjXnj{z+ZZ!xj%WcB|9y`vZhbHJ zlF}gL8L-tnvxf(;EeYCWdO}>TFO;j-?s@&ZiG*J~rK0oPI()KDzM$s=$qy6o9t{EJ zvPYYyePuE1UAft4k3!tJ>`Pjp9e*$1y8d{5GmgUIy$6mLuOt@YX^*T>{Gl3}`q+9s zG2s4VO>x`W9?S-Q{p;RfJ-=u(V5fny9_9Hb-)}?kC@(e8%yOiQV}NT~Kp)cVv05Cq z_TQvom%N1Bps~r_D?1pNbdTQK>-cP^w`a8_%?&CKIVnR>R)ye&yIq65?l_7NEN{o3 zsk$T$Q-5AhX?sT(e(8mRn4G6?o}G(&`{tYL9NMm2x;DlSBE$o8)bhRtx&nHEX^0O%zp3LO4 z^RYD?9SqOcETmA`W#f~_2Q)PKe!Tr_`GdQOWabApVcdo~Uyl`lk_@?(ss~CFa0}IB zeWeH)0mzQoaGViL$PZ7hT*W(YdMqMak(!}$=aLZC%J|z?LG}7Y8`!BunrH~MiMn3w z*~fU$K&bdCUM6I^^usKlASi$kFg&6|78{3aca=WBbZ_i!O(^``vhnxf84Unn_f=3$ z=>~=p6~gGaQHJ={2&;#5Z{Qt+b#T~xBKmlXrr%2g;ckMWm3!p;Z}+W3HRaBgKI~%V z$sFzTnlzkPZn#$0r0|WNPZ8~X_-v&6FTji2_qBCI@}e$p+YD83K|6DPqkR^*eU9_w zAi(<}8ao2joB!q))$V`Bj}=~^8Ke1oYLu%Bx&OGG-{%K#Y305(RPR$hua}r=@8XSc zVt?5Cx^iqR1kC4m%Jy$w6ld)WI(=o0h1Lkh^gi$CkxL6Vv5gnayQt?_Q7oOW-mhrR zjRzgq0C zD!cwlJQDN=taz=ug97Wuo~J$YCBy|Hca z5gG#%w?z%6C4;UwT0cAli=C5AnBPyVuhd z@uhorqzk^%+yXjTFN2XB9H64;?)X8SCj_M7?|jm~j(ma`;f#YJxF98ROZxhiB73*b z4?8kU$Ze(7y@hnib!boC_HF7ETVS73q}Y%0<#D< z)2C;m-a*&D{?&_f>4NY(*Uk{E5RH#3syUvsYvnrD_k@{DzWT%Ye7~Ht@ld;OWvnHE?{=pC>f}p{Uce+<3gNVa>pOQ;G;zKW&-$v1i*+ z@3Q+<@k=CKOgdC8>LBvhYJ#)ESR4eyTlFc^?jPL2UHRUS9Df#g{zuLf7uSdxH@%Q1 zOMbiXzzt!XIp{1 z@c!-nKEv?%?2P;!Y&wth#{b*LL;HPOx@s=&v&chs6ok;w&)~2|gs0;?)go=jf{-?ONm=>=XEj z_~UfH1jAR(A8R5(`3nZ}^)o+e04!`@jG5_G%<#>Ym;TgGxpjV{65GEz z8>XLxd0p9Hq7qGqFb)0N3rzdQR3t##NYl+5cv#yP>%{(HqN2X~_>^VI>*$$9yo*1+ zZB>FWNi21GGeI@?%lniL=6HX`WNg@QOSNR3jem+={3J^(a0Rga_7Aw_00APDV6YTa zZ!smSIc^gtU(yz4{S#^cbpcpG;9pe#NBg$Vq}~*c|;byijfDg^XY}V5qeKNr>hdKVWW6W`(Kd8cf2tg=+@omfnE){^DLG zS)ovlj(v2&_3V~KgbD0ztD;9-xexNZobd})W>$h?rymo6qYQNEU)DxXmrtg?{Y*l!xty!*Hv=qo{9pMFTQwfOGo*ZJ|4&15?kMr8%J>>&ma}YMt}MYIz(GS;t?M3q0Hp_v1Tf35zkKP?&%u zW4$S8&S53Zp;VM=zJ=sp(m;wBdy2T{goU-^Lte|NCGm+(C}%3ky7Qn6Bg8+KFtB2V zXX2o27@I^uO>!yte%dci8x`}R$ZG%+)5FozYe z3#tJ(x0%*JZ_I(hjTOwsU){HIWFoPrhj?ohDnm8jE1GaNgPdfC8wMEXF6GDy)>Hhx zsCSloJ#enm71AaIG(+7#XO{$%fD}C1!2bz(ACO6bp_>2QA+j+W+GwI;OTsZ!^OSth zwmnsnH0kNM+5D)DdUxh6FTz3ejP|HqqtRx94OBDyrYpKIUzFR6WcmF5Q0`yrk4ev2VlThWR5$=%B&y*^063jG|zf_YdqHcbf{0EH%1x+xtEz-`pllh z5S+V3vf87aN9Q{S!{ihNPSru1fPk#5asE@D5Xt1e4Tx-;7 zmv6RadO7kN-xq5lQ!58RDs?y$3CU%gA5Sgd!zco8$fg8LhBRN0KbwAt^tRlitD`rQ zl+4eCD^?yu`^~TU@pe!R=_nF%XVUfHSr|5$HfIExg0&%BGA>bhd4BfyrzWDllm4fP zJM*^=pEt}Ls{6&AIuY=DpejN>qgwC})t2qbxS65V+zix3OmNk9=FDS)G$8}&SwkLK z#p7v$6$ouYFiIj;L`y0`nc*1Ltvvy9fwj1%k6!bWRH;O)9ye#am(D~3^Rk8^;P^q= z80$Y!f1DFIz9`Y0u!P{x*L0<^0`<1R1pHoNc#WvKcM~9i;&$+)K}r@F3)6!bTmgCw zNsI1t$enmpt7E{?>OauBg%m}=m8a>c3H%Fezb$h!E9_fV6n9Z}&u@Ualb_AE3~PfrJg0g+XQG&{}iJ=!{HQ|k*CMtTh^aoIiMkKmq7j4Z;uhM z_0|!t5!})uL0t6&zLb5)?+uPjG3F>puOUN~#iuTq<;ao_a zO3eqz=mlv9^@2RAE_uCg1l=&7)-$Txr|U?>xwD(V>2s=ObVP3_9)$7p@geKN*!f9Y zf_-rtb{f^a3$zUs{0aV~4-*!M2p(T=YsFx^$S`)NJih1%q>ga^!`t;@v2p^0GFo*9rn;{E>%vAmf37#~;zUY$8@W1$L1VpL%;>{vo zWBsQ3GCW^eNoVI@%Zq>S{oX2SjIS6=FC}Zb z*u$&yAo%LSe%v&i8pIf4VX_S`6F(hhdCW@-4UlO}o|$xkF?_p6H#tJyNnPb4QUUvu zFT1oF$<~0Y>i46r_zX6(i)!O~eWA+k+rNKS)1tHnZ>}z}*e~`mZeBL4w@3ea&gQ3Z z?C;{%SF~UH!$74Ap_)9%naC1zeG}wg({e&xmqZ!EwOL7~1k~^t*&~G2zv-uXvrnJ% zT@Dwz<^BDYPNcjj-sGjn&ycS7GvF}BVy=0)4w7mN+E9;CnRVMxC$AqVnY83{-q5`| zrU*%aGwGRhxR{!Us6jR2oik9*K0Ls_N7O`rJh&6{&3Tj?YrIeT?ecy-7VUdFnd>+R zUC}+(aGXAsJ_l4Vv8m~s#-sGnX4Mf(Pvvy z*(D;)obw`cB1D%O3Jg9&5%Tn(KV`4C}^yFMOVwq1mCnCe*je zywaj*VCuV&7qYDQ^E?2@cMG(==zy7>EDjJj*z{Z)I7%*X&|Nqz;u2CVYIo&CR9gV^W9jsYTD**gPf(dM zIPhYkA`@+}P@I5UsD?2FU{k<>gXJ069uHy+&lW!a;6VKMcp3bkvJ|Mqc^+I-?!~8}(_Sy^D!Eo*IYq73hs@yr4Q7vO-t%BZj${vvEAr zf3-63{!57JGhJ+!KQ=Fle$2+-^E`5>OBF=-=kAMl53~xs@yM8L0v_qH{;>aa7{opP z`~7%)P`b~W>^yH3bg&V{{c=5nm^6mvK>J5k&lcYFkZzJe-7?T?6J+TYG#lV}GT}rUlU~lDtVb$NXuSqSS}qPul49 zOYrYNbE&AGECExUuL1)I-fD%;3wSzTD|@B^BwYMf+4&L^2fJ|v#f+j4+h4r~8$rrm zsup343nF4;=SKMwp=LaGLQsDnPNIy%Q>@P*?rJCkCqk8+C}z^lo<}UzT4&g{>biQ$S)HJww>}A z2*KE0*R@yw_UYSp?;lPRed8L$O~(Z$=*Mrwt0^e~0oHe4=m7PO!@Yv{A6_)rb(GA>@ruo-K8iP;MkB2%2^U`{hS*CTlX z7M|5gCUSGI5(dLCT=8!LUifQ5GTiSai%rao2*ctg1bvK!FoT_Zhvx-Bp8BS{KONV( zyLa*82TI=dtl$2dy6!VyGL>z6h-~w z%RM1w`*I8LFr0?EfSlvodf;fc6T`NTJ^yy6BtaA_0dECkmfHUP>8a7~zgxNn4Us%J zjH5=8;7E{4IwWe1`W7tAC`QnqDlR2ap zdZabfL$`xHW$2dShvzVVtm{c?BgxOV%DVel5mlaK7(xf(P%101k@DDCnYR#h)SOIqv|u0MaHXCT^lu(xQc z0qY1=yj^habovSBXJj3HTg?VuylGmxZ_~>~^)p0i!NX_Ga`XY065~!N?=G%-eU8#X zzNA0Aup{;Tq63Lp<0%}zp_jf#gyRxPl$aGqOnu_S({7Y*4cNAu0#4D0FZuQMvGHW3 zdCZZC0+boJ8aC*ZPZwLw(t{o2`H^^t$Nbg^Ti zXWn23e`|1%Rh?~skCP{Eudv<$NLceL+r#4>&(EP(pLLO+gDA%fa({Sgyj*FD!}4sw zPXq%V^YTWu-ngs8&g?p^w-ny;hz%x3qDi4hzy`(jb94sqbM4kFJB(7#NhIi}AifKvEF6U8N@#;tqoi!-V z;7CqjBh-*a#F_aR)OUwkmk1*a9G49vdf^*niH>NJZ$2r4mMUHs9S=be42Oa|sVY*e zaqCzK2j4~_P67fWz*+7^a!ll!C~~)A?SG9Ib?4L^eKiRLy4$;bVd9;6&0hvW z2K6KIiMZK;U4COWx5RHd@ofO|>FWG4DQthIvjjPk;S9~>VSNHLI;BYCMbrLzNAXic z=S9|oWm)X+*jU`xd-}%qSHmo2XWq^*yp}J}WZUG271*4wu6GSvtY7zCYVOUIlnJDm zANoxhGpeI!LlOK?y^Dl;iSp*iR;h<*apjY#09<42FUckfY0SX$Kj^#(+{x*V4afVV zW6k$;2oH1~%)h=vzy!WkuV4AEj?omRWL}GBhmHfj9MK6xkT}8;mYa-fwW!?j(sijk zPHPH!bqBy#0egQDfV(Op8F6>Ic<-wt!pmjBFeLh1^MoC<=G}pd;wW97X^!XC3x35X z_pf6G!uUY_dNv{60>VQSmw5RL^2b5@`c6?;0I|PQDuZ~d9;o;K;n%ur%f7K^XNu87 z;2~)fil8~AR~Yhqd{B?~D6&G+0`?3o#TN;sl<;}9|0tNJ^l1c5gcTBp^l=jW77vWq zp)HbBHn;jv#+l~!7ix^QEF4Q4P)(L5#u2RX zbR+4t-u=q>$!ByP?NClaBLF*+Kv5qO?Hhs+=Zj}OsxyyTE_DXH{qLAK6X8@ORVLEd zf4#Z1G*cz=KIJ6?y9?lqeZtx=1ir&PzABn3RkX2~I7 zWcCzJKym&DVm}8Q2+b$MbWFM@U^%iD9`{13TU5~i>#c;&F`+Yu$Gj!NIguashO)&!`UPwy znF&CO_t+YxiH3v<@4b3`^MzHriAaQ}1l7>dA-CIgo09wM@#zQivwy0;nt^LplgsC}lgUi5dvEt|#`Nx}2^Gu7kz>C|=gZ^W zx!fgy6II-__Z-kIp>~a1cL`6ASu^?DlvYz^5Ne*U9B+EyU#{rz87}?O#|52R(i`i6 zG=gG(ul}_(%x%9G#JjqTeXLzb`%0{3HrD)zO{UHB8=v8o^ zr^U)+>%H`AHOU^@aFoBV=DwA8{4hdu&VEFD zsXnh+PX1FX`H|ecPi`G0l2j-$YC87&^SvE5Njyj?`xNrJKRR~dFOpKasAh5LT&+)k z(dc!cpwh0m8;`&}k;zq5ghvnB=|@z$P%!8Jouxh#rxfRF3!|7LBV@*tMpn5G(! zAFV#+S;2}^S!KV!9=cxz7H>U`SxpLdv+d^8_SatX^zLKr?(LL&zjH7}Y++E8T#`dw zIof!p0_A^azu(0#0oWivKjsV(UpY0F?9bbc?Kxhp>V|ziu)0}Z`0tfEx~)I`nJegg z%NPyu)Oi9T{ZJC}=?j-Fzq|a;!x|gX-Im;JgWf8*yPKOtINyHIx<0bRn7!}QZ8!zf+zRLT}r=6E;l|KRbyL#ez?Dbct z94(~3z9o+1iUGNWnrVpmEF}vqH_NzwZ`h~&tG{BFRO&l1&XUk`S1Mn?a+TZ!day2e-_&R)P-pNJ7IAmNSY^mL3iqB4q`_g)BUxdE1XbC5r~_rE@2`NBec?oQ?R=y*n}ZDk{v_#tAEwdY;lhO@a2#OL+&K z1XO@qV1cabuTtjr0L@fikEr@P2jBVZOZR~k6jv{8{K=y;>f5UeP*b`HugUxnYP^8< zh=M`*^8bG?lm6G^k9z5K??r8M8#O8L4zu4cX3$;G-oFrBJ5P;n3dAm8nt!_Khbw$u zPm$o}Uvvit0aFX(g|OKH?8LKRQK2e6Mt{&HK)k6Bla64Pq(QZ>4_=-ysmPZD__(=z zUa1IUBt<420wOCyW&cg3;rYQevf6I}vX~@whp^ zR1&?U2X$iCLea8r)mugW#*Xi-gPCtn~K3t_AuqOPe+gP&e1)wX`kGsqFk5gKR-91Yfj2!X31^@wo(iU!*wd>yE%TurMc# zf;=bNHlkfU&^2vBuu2@5;IF?*1tK;FrUlxmSmgc4u=(CwUelH&l z?0ZIyvZ%-5Bj=J%w=jkyM0=d9fqH61$G7eKMY^5%2ige#oKZgo7+0>qd2FGJHCpg| z&5vwh^7H$AX^wu6xRA79<-e##Gm78sJ;Pd1kjH${5wXtX_-jI##7wdAz`cozbSlaE zpMWtLsz)RHI5c)^mP3rrn^wK)p)+%r@A)muXEw|Gcl7WGx?SwhIo6!v6}!1RSgATv zw_1~dmPZvGQKz5yDW{mP2#UW;Tr+Nuz*L(v=p?E}`=F6kr|7JCZ@~nyOMv#0Jbauc zUf2tZdr{(BsUAxCI&`6C_EbCjk4rTwd)f@s#FOPe9-s{z9)z?$@aP>h&^cu1>i|_cNQS;*}lS z_es5i1Bxvg!_XP_%(}g8|KB+nY=sc%apZ}#x8{uG&Y+r8kNV2)-%GZJeSm5-O$okY zc89=SNb|zu#rJnXXEalj@a=N>MJBxa7yvO<%!V{3yxc=MwJ}u2MwjmFAa;>m5(qJ7x`quY)E zRe|aj*Y~9P2Z>{NG|w4+^A224E>#4MV@t!wP$b;x^j><@XUCnk8l4M8&$ZCtgo_02 zW&T}<3)9w3M-I~onkK$5u1XVk)*pZ^~r1LdchJeL#5|SdKTC@iY@2lPK_L}<# zhT~oo3fOHnC2hCbISE?Kdk+`o+67mt<@zb`m=8|5JDG9)v-fV*$R*|BWnwAXZCn5$ zjk-iG=zB3Jcz*e~ZFP`^I>k{t&N#qW zpw^xoTuv@b2l@_;U|+`Pn1Wyl*WhSZR(LdI^BK_h%c$n@3rU;0dQwy4dnMve@OEO7{(oB6>JO5*9sA3xX997d8rO`F8zeg zto?qG-hr{y>{K~Yrp^d*m6uN!OEIUu(7sIrVtQ3Q-E07S86_Nom+nyNX)zPQL4%fF~L{LBz~51Gha)~HHAF}-7oyQ zs&rEt+};vuV5!lzzxKzeZyOxt^$foIqpuYWzH@F-N| zsR?A^l6^n^SoII*mxE^}ee&|Xf8=L>Yh2Wa4@`N`!2 z7n^N=%2I8@DeHkR3=-(%lQd3`%Nu=XlIcY`&Km%O7Qz9fPmM?GD*7xSZu6p!c%@iw zWjknr6{IlC{lmG(O*f#ulpYCi~Qvb%sirAXHN<|3@ozAJ7e38>X2#)LJYt#xD7kh8+uNOeA z-R_6#Q``r`y{`2b4%{w=?Rd-s@f-5D8@?k0+6{fVPq@(BUl*o?>LmPa!f`#2xH_{_4GGuXN{E0PgI~9u*@I&Gj5QWVcw)uq3qi`B-W@`1-`Bv!se?2=(Ku zGdxxC=b;SvDTM0n!(;7(E^erHoCP+2~Wj=(r__AtwLWr>uGOG+|VzJBTf;j!d;7LxgZzxCs_wJE21#blDazIsp}Y z_vqXn%tV!l<)s_U*SYeP;Y}JGfLS!4@6gy${}>AxW#cp}L=|5AA7}5e^SIIk3qBAR zq(spzC6b~Sz1`7!BhmBeZ`If52jZ)f#nnWvHtk63A9>PSkS8S7gO1f0<+D8KD2WIZz|U@81hvA` zprnIj%;Veuh#N^S6A8{f3&2)q zPLI`S?n!k>P-L7#xTnMZUb{3k@v%C9AN|BUPc$I`3qdYnZEpctXrPhggBq%;LM7km z|95^9G>=uch;YXq8EUtq0@vN~2|O>yIS+6U7^iu_#jd|zRyV6B&Nx0uJMvwy zf2C;F|LJ1ZVSH#Y6J(iSIFf!M=z;WDLg6rIae%$>+<(1c`QXE&tJxGfSx?boNix_w zoAr;>1{n;cLAs}PImC>UZ^@Y*3yL5J(Q=$jaYC98S7 zE0!ty01vsK7s6+~SMSe#<}-a;&1@xnZp$&2-T*vYj@` zU%wlv&hJcqea?@}mUOdnyUr{50aDUb9hxM6ULd!KN8@hJ!?mZCCfeK6{9m0;d4;W= ztL_%%vu@$a96cqxC&7j?EGPrw$72bq2c8q6`xe4_wfMi@Pmt)emp5+)6bg9UY8B_T z3oEBBiK__SOE%L>S^DJBPfY{hWoa$ED(=t!)z+*m_GY0*?t$R6ui|*h05~#if!>LNJU++Z7$7A(JGGE-Kj_W7+IV5TO zHM@;Ik|kaC{^LCLDxXiMSV)AQz>ik7=FsDRHAu;{RJ|O!-PR z=9-X3bxIB`50lQilHRdI)|W1z)Fs(GWf6bnAm%-dI_7TZ%a~U~;(PNHK7rUh#L8&1 z(sfS-fL0kSg=@sVfofen4%db@^|P;B`6eWQivbMEDPBN2+%kCIvo(Qx zVvus7K&(e@1QUNe9@z|H^Z_<|&raU=_uB9a_CbCA&Y_=xN_0?@G#qsCp$L5jP>U|2 zyVW+*sgoc*Dw{Z5+T!znJr2Zpi|;Z>XDo<FP9W;E_XYD96mx08e>uzXt-X+$ahY@2Lxd;rj*Oh31$TjI(ctMmHSx|Gnr$ zU035PUKE)TyxdD$)V}&h5fEgqd#+VW`?FqycC#bdlv!^gWFa_lVv7C#NcA_HEC_PJ z^OUQ*Q?ClJfxaWR?oa;QPd}x$v|$DBQ2s#~hRPpL>t8+eQ@%_9e_36G64Cqp&P3Y` zOK)kU`4SIx6b~3{DQmt*(w`?A6&U~jYfL38{eSA1n&HnE69D8K`r}12;N3aGa63$S z&Z{i_u#Z{`p8Mxw`2eurxCT3eeaujM#2>C_xv%FF_H7`^tK18WT5d^_z`_J_0UWV1 zE0vuLNCSH-u>#j$@(`gDEHo-1EEg-ccz(cI4%0+;=M8mc`q1DiJfDQVOt0ofF<%Qj z%me`ZIe@+1@W>d%RuC~CJv^I*T|KbJ;Axf;`%l{(3RcfB2hqO4)@*hF`Y*JH&_DSg z$l$;_zAn{LHUF^Gq`OI3r)$fWTf0xmPgV*w_Ab5pq2lKoA%eevTot6rLewXb3%~^r3Ws%;{%$QvjK|rK z;T;LKcBFROoEsvMg3lRAeOWH}-2D5m0AMrr3lu?Z?@DpQ%QmCGVh^ml?wv&FpWh*D z`v3N~C$do#bqf-x@A>ooqnP6VT}w)@13}*WBs3*3aw82TAj*!n`FFXJN ze^#;yC%}A*3j2Hh^h9v^*x8@=U+Bxt7lC*l>YQ|ueL+2s-VXy#+OQ5k0kEilTqMA+ zqw;K6?ZA(p4DqepDnz!D;TQT|wz0kZY>X+2$xSj8fXM{_@#|`AOKyMG{&1Clz7kOL z1kZr}RDY8^I1AoY;TF_~+}8Fj{=8WPg!u&j+&G5RrPFLJ9VcP^4Qgd%{&fp3QurB< zzRIPaWP2I8sUiT4-Iu!AWQ^kWpc$20MgYI{`=Oq8@K+;pK2GsIWWKQ=Wac{EY00pk z#xvBOPgv)g{O?!bmb=PwiB5F4zWj@dS_DHHSqD#C~PUgBIgn;F|^!Mxusxf zRUF?h$pJg4iI+-?%7px>@w7i2{}w%cZ0*U6<)3YZDdNNEW4=nd6j0#vMq~0Avt_0a zR!L~J^;hRO=br*A=Ccp=4U&y*+^K;F;OgREzXeJm$eq?j=ga>{dV@ZzQ3_sF(AxVk z!#jLQifk8)fXzmEey(ZSXd2bpH+E`dvyP7iFV(7 zl_m0xugJU^Kf)FA>F@y0lHhpGTCOi|a;1-$Jd@Nn7s=wL3%|qEN+#>DCQhA#b4FZ8 zhgn`g&r(NHs!)L*USI2X%y=^1FdVl0kbhi^w?6_6>#YKpbJ?H~61YQE;%gBWmGuf!I%|?%ALs?-Z8`#`vr4XYhdX0oUnY zjH$f*&_aQ)H0OE3^E{o%y!HZt=1Fy_(^~wC2j8J}Cp_P^CP03&`igc0PZ*?!UVsqS zz_!J6Q92t|52)~qrQ4Jv?**O($gI|s&^za!Zsh5E0BtIv2%g<`W84AN`|YRn=`H>8hFDA`sa}ZR0+Ld?`|#U!5eCOXzIvv zi?3ig>H{uo2ei9)29V$XJdW5SU7@D_WB_I(@%RD(tj9pi_~%2UIs9c}i^^h*=0M3c ztO2Kk3kZPP6KxFriXW6}7`lie2CUKD6WEs?5}mk=HA|m_zsD*Srfh>0;%vKz%$oQ8y_===eBsP8_frF0HU9Y@SX(gNpdfUYOqplw zxW9ezWX3_{o{3B^%`E@9BW%!{&%!w9;7z1y0C({T)*`_O>Qp%2&i%nNV!>e$;}!bv zWzdgEaS(>xLkJ6A)eD}_&48RlMFMVT&?Ul<@9JmZ2PQUjeoch_5-Kl2M{W_icdP1-!NTH%qE_6E3E*W-hPF6XJ*3frVrJ=u@v^==3+-+ z@eMr3&=4PFHh8x@h{G~6q**ij!#-cqhCgt7obnhh#kifUStt z>0zn@II*9MD2E=n;IT1B7I+?lJ$++2_b}BYuB3YC8Qqt}Tag?dzdMKQWfJr5?%lXn zm}g_3dmx@t7?O`Qe`{WpO=yuE#`b34#uc@I4~2_=m*EKq*hfnQ;O_D*BZ+^S6^Q(4 zp>c0as5K_BBIC#2ZQpNS;U?QYemj*V-fkDCE1~H}-XivSY=HZ!tg9uJ_}^)N~+qsyw1-PQNro z|LJ(#kCjf)!kbSw4MgSGz0#YD0zO=D$A(r(F|4;p5r4JIDA8uPBwM!R4QJaS**IjU z?Mh6|!lw|+TM_|KYnBN}XmOGYejwvYC6M~;O`#ayS^>H)4t-2HbIe>beJX!QpcCi%7y#@=G{T^_4v zwFNa`A!kSGBVW^xv)rgU6MrGVeAZX#ovqrS$JKiR)D~iS9Iz(n$$FpveBGA8x@E5a zacm*zzVC7_7^Y|#q{D6E_&kzFexmP`N`uDR7f)ETE7|DR?H&xhvP`jNpg(VV1Jp}n zf@*=g{&mC)dT|wk;aN}qVH#QfQn`*FyLWl*TM9Bht+l`2wA!4F5v{(*j^$gs^P}*&S!9I$r1Mrg zkEc_`Lh`EYhVllJ=m~>pT<+;I@#MAUxSf4@B#I>e@AGK+(N99UBY?mI@K3?xk%c-wje5r)uvC`M>x`uT z`Kbl92z7K%a5dvMU3OZQxN~WYbpOBu7Q$}ThIw?lcL#_z)+G0q3fVdvT0%bR&s%vl zynci8Lf$4%aODFG6Fl z4?$cqpzudY=jto&;oOTF>|%?if`49ho9x~6(*PTzLDUDWa^A7)1^=%5?z$C{Ys=#? zc@uS~)=9psE%qZ*Hd~>=aLj4Ie^S<4Xjsy5IcIXr$%_KX#S|nf2M7G z_3iP4b0zboix~#+5mD{h!>jg}kv9D8C6O~t`Br0((a4GMNVrQNJH}zW&5>#B5Qn-O z_p{`17D@(KOrV00=Q75Hr@0JX#linxaNX#MQ=YH8KMZww|AqJ4S3k$2^M>@8+_>dG z?h@6KB`9x6@nj@{a@i*|o>A3IJpnwjVCR>qya2Aht#%OLWCfF@pOZ;HO-kHW6w@&b$Nh){&t#mCqGbobv(g~Iw(2`L)g#=!>jnVhA|&anaVN@I=p(fxdW1X|8sF<~Nm8c>iB+6sv@pBh>LCl8rdJJ#6;} zrz!956_1pEFYl5ff)A$+8C4A9(PXPRyX?ftnvS}BG#}w*`*(>ITHPDLWYgG-C;kD*0C2T5pL-T*&9z`w^PrWaoNvzp5D5=Sz-cgyYfZYtES5}2a=*nTxWA=C}8 zWmTV{r}X7RFR|hCm;-fBo_A(Re!+Dkm*9aIuvaMbt10mf^qxb{7{Q5VK?B!L7dAG@ zF~HOZtjW#1p9{OsX<`fcqTM-e9Q-M?4>-I7pC9xuucKA-Az zUEY(0*Y8cCwF)VS%sy$+R2e$Y_uiq^5W7BwwLZPIr2cxj_W`}`j6nVSJS;*#!Zx*F z&v1KXTas)&Xj=>DS6t*CeFoG0voty$&cuXKQfBzKf#rS6_|sUMwSZHZ+`OxkU}??> zvSad>x4s{|Y`J1xnG1hSH<8|g-&3fg{;pjGSNmwOR*}z^kGwu*LQ0@#q}g9v<7f1Y zYW-7YsE+gi6O`hnEk{b0H{6cS-IQ#Hz*tj=#`@!zrB|UyJ;YUPQWf@Z?iKqTZp1+y z4}wVSmy9vNt4`|~huVU-AaB>r=bDoA0e`6tU*gvxet$0z-2ctidN-^ktK4*N@Lr^L z&fIQ~X&U^sp2ASDJ-SrQ!NCXUPsLLUx%A&n_z9YrXt`#_h^cmM+d{tr$zm-R=j`9Bzdz`xdMF98`agsv{&}dr zUvjXMN|d3cE+Yu?lI>sdWY*4DBRmXh)z-+qgKJ)AQroae$_*N8zVBj^>d*7%G1;Dp z9rs_}e)r%$2ayhs92KcS{0@yWVVPiSu}8oMBHMy$97>h3?DKQ*QhuPrOquq%n%`Me^Py13uZrTOM6ba+lPsUc=NsvWtk1y*ml5E*#?pq ziV}AW#DpitcGh#+41>;VpDAd^pK^ROI0NG(WA*M~ZDM_FANOfaWv9s2Ay8?vgS}J) zSoQEs*Ny0mle&uXwzm4fbrzo*WlN+`e!4x3`crYgJ(sxXD`z%#Ag@mCi8Y@Fz@b9uCjzzMH-*TA2M4MDpP1=; zS`s|hb^))eqgya+D!}3M6XDFc@zS@y-$dCmG3a0U&>4f!ibm9hv`G z1w+;TiFh=+4MYF(?VBsT;o7VLUCmUg$ifx5E`)i|v?%?{9E5ppbHCp{o%*SK@UT%+9eTuC#7}SAGT#$=%K5h`azI$G zh;NJ_{Qw|vg;rRN3jJ%!H@j)tjwOXHvRXWjO(AB5zGO08aX3t+-vDOdwo zR>Y#YY_G|`uSw^8)1k%&r)p>=#uOqw8HoR+vw9@EsCWo1S}IL@7tj0rWD85%i{D8g1?bQe1i#7IgzyH$1!2P8Ww>-pb2XaGB8mJhJ&sbyaSQn&x+1cy0P7D+G$1 zNiMs(QXRw;ovhs)rPWqA=jhwt!XQunSF||x4Xl=*%=ocx8h!P?s7<%D`8;TW6}Z>HqQ!=Py(L(!|3C50Ln+wVup$*TE^>p59E63 zk5~Tt*ckw(cmAH1B1O4KR8p6*`a?4}N&xiBu@R*xbyD1d>R-H;#$k3lZrC5Bq z7b~vzJRC7VAzZn?vZBz0){czR;9nqDIc&6Tos6Fpc}9Ty{^EpNoWe}pjC{WdOA1Nh z+l;5jdAC_LuiJ7xZ*RBO>EPQ@hxx5hC#*C(g8yLfsIW#9Hxav}JcRVQZbMG~{nA5vcDd_$v_U!c6Hd z!d6rHPANt(hVc8@3puhm;I_SaXo@6uwDA^x9=UF}_npzPkf~)JA@iQ2gg!%*)iPj^9Jq;!!j#A&2I>Z%dC1kypfcKS2?YWFz((-`3 zy8LA9tk>Bx>sm`daMR?INt)|N>Zl5e79V7gQ=Kq$t~IbqaihqmOlh7%-H$g)Y)fy) zZN`UMHE_)+93y#ZxdinoV+ZI^9d=-^@t9n-L41A?%Ih+ji&(kq7nHQ(Z?bgk-HH+K z;-(w-yyDxQaX;Sh1Ygf${_`Wu??55Pq=@N+r-cDH3U__XbkDc&w4n$VZBEW<7BdAgvMFu^B@ts*KFc$9Va;yXD-I zYcS|+J!bn|*@sWD1H?3Oh3$p;iMsR$1fD>BK^?Fy3+4&-EMaH}iyRd{91EOt7n8xf z2?=)^&ryzse6wogxGTZl*MGSQR|x?{F_C=G>6834Nr)PR&7LH&8KiS(_-f`V5fSxg z?%%hn#d@>3StGlK1$gL~R%8*GK>vqgc$I&8qxAl(-j0wYq+*V8zPPJ$@ZmG~qG+a< zdmGMv8ziVvaxwfm?uF=9&g4k4A=oj?ZO(09;^ZAd;kuDHMZ4a9G#APn+8|+d4%sPf zO$5C)m`s)*4>Ps4jo@QM{4p~Axqecr@9~5d6bST& zgeq@ucoe<){?c=LpTX10S;wJ_A^I5b9d9olo-2j^3(oQCz65NqE`Ek-{&5O{I1oC7 z`<<@Z$7jn#)5yvOMJtp$WSK-w-DY35Rcz*$_6t;gk=x7Y>!e>J)RZND!INiex8@h1 zr=8@5MSi>{Ya5E|B9sq^J0XaBNE0vij3*4?GKu^06pdwvZvp4rR_bYXU$I0OT=9~N zy46_MwU6fkYhBtinHcv)3R26rAlL&C=&^NY0YJ!Wn2b1iOExS*KR&YxedefGn{m}(c} z2~8lJF9G|zdzDyq%0A{?jOKg(Sd*c8tTGF&42ClOQ~`@dZE$inZEb*3V+|;g{1f0JYSy8^xLrznRfBs zQ|(QYQ+TVyAG;bh?>nuAs~m-nmTvVcOML^(fT(?KR=VF?ueT*fWw}V zjYwQOL4W=4w>Sy+XW_41d%rE$t73ngHVtBRVF37s0+G)=<@CaY%yUf|GtD0%8DgG8 ze-t|*5Yc=s%DK$eh_Z8^8CSZ0IfFZYKb0VhNGFo`Zh*DQhS*Rl!aM4v?9jNUAxUJmoFFYr{W=q$XFfqI4h4SVD*O;+&?US81toWv#OPtLUsZR`@nWW006=Dq^rojU@v3{?&)u;M$rJq z&Rkhc!k$79oPO^$Fa}+`9-2N(WMX}9?$4Xf^cT@&I{E-jmVdo35EIywI|u9FOgrrJ z#bru?HBfFiNr1peniQVfYgw^rj2~@}C{m4P?KKQIE}Ar%u?~Wv-YJf05zkKVpDRY3eh7Et$&pEY`6Kx{l_md`KXhSM5Pyu{?}s)762Acv zryNf6tu_+jEM^OOTuo#)blZke%RM_~d7mHu=Y|^8&oK;_@ZXhvN~;qn z)IOp3eu?u#TF8rBzDT!M+-dR_{`sDsT4yB2=EF^EC8&LB^!sCeRqi)t<8iv>U>~_W z3Ua?S`uB%{dAc}_@;77&MPJ#08mjhG#y~w9_lFj#+vpCU}KcKGaF_0PeE!z z>^-k2m!iq++mSvfm4AbJ89dqW3GdKvKYe2x{PDd%5Vb}-U4xufCDOtw4EzdMEHYk; ze(_G~3xmy%b-dnUuxi6Pmv74BZ4J&ZbfT{mhFIK}!}qjJDNr+q}$ZKIP6cW#E&w$uCh6PklDU?0l&{k!4z_A>?)8D3I%6S72o z3-#--X33nz-;5A&PsO)*92!67RgxLzn%@yVyzAeJx<^ux^hmWEO=*|Wc^j1FtKJui zwdbEC2?@p3DvG>r#3U-Gj>1|En`kewIYtcAm=UMAoxdc9ItVAxdc3>0%ts!DrRu*% zJyxlFWzF&USMUC-^TPSHi29#qDhIr?5J--m35WRJ*b1z+-}~$7gRzyVUkrVJN8Nb% z0emd)Srun9UH`;82`ur8C9ur883$HtBn$-ou<6hV9pwg>;@vbOdQ z(V_DhkCxW&YCdg;4&lNJ<;qo*lj)14XlAuiR|i+YJnHWY_MU+Nvp*#^?5jx|oI^5s z?S+G1BJ4$`3vPcHkV?0DxH%*C*FT&(GDetl@7%|Oo2mRp(5}iDG@`ywWM8PlNCyVI z7#z-=yu?T=SuGI9TP_!Bz`RP1At%3H#<{zYkDb5BWS>{w<0dyHZN5+6)KyFFhwlUR zFb?uK?cWXK%+F@?(`reSIAg~e4IPENP-pH3CGUT;*kwzscWfhGcFw-z6|oqDFOnCa zR}vAZHpakSK&+E({|!eO1PZ;#Zfl2htc6*=21P?52t1u2ie%IT&Z(d*AI<>}ItxPz zxr83bEmKLbd?#Je^UBwc_vp0itKaG~vveM?Q>a z@+tQ}{asKPAUzN8%x(_gyohAJhdQ*@`QncgYBYpu44;23BGA?tgp0!DLuk_$bVPp1 z7}cS$kL5$q_izo89}j1nr^7r!?tthIfUXk9XHPAA58(~d+Rx~P2;fE*VB9(d7|Y|$ zKC@r!i_X@}_~_~|NY-*7E6rYTN4iYl;8{+q@XAY9q#u#Jwe}S*x!!M$N{k)`Z^!)l#Df! ze_s0HiEgR_Od-hPZszW;fw;EG_?HtI^@k-q$c$6-!TB25dS+C$cF%5HDoVM*k%KCK zCNuKahta<8??A!vTKN_*Mol1fI%^bogRBB0Ka@Y9xIYpASIh zrTZDbt*oeif^gnBxnt0xSLoT)=NCJ2ueNQ9Ui(~LlK8|2c{J1;VJP2qd0hc%lKK(K zfdM$C#k|E~HQyneNpBP3{D6HD&=ULz;iS|*pMWnp|DZ+|{j%lD7y zrb-%wgQ|@#YUY1@CtaSg{dwuavjol!CqjM6bN^-w%MBjJ}b!=EM7v!k*9G|`Z@cguG-8XiUYDpH7|_^;>Sisr+-+-eSe z5fD^YwM405B~>tjGddwg9}LDe_ruQnX!xZV_Vp~?TH{X5D;IYMEJgZO=Gt@x8NT*T zEJs=cZ`tid`9=_G&?Hx44FCEuCDs`wIZU17;;DQ0qcQT%E+@&I=1gs4YTi-hEKsia z!`E=8s{?z@qLKj1=exo4#P$U10DK6@EoNFTncA~%Km_db2tM6sJX6j)rdnb8$W0xk zv7IjYHg!DOfnS{?CQCkx@VW>uvsN97*+p~tc^0>BD`tPt3ipSJX1y#0>tiD$z%AUT z1z?=_2F53!N>FNTsM>K5`yeU|WWBM)SRn&PW-Q!n`} zY17$-CrgGE*_O2voZ%gi&<*ydbBF6Oi9x=edp>%6hn+9*i~?ibq-uC%Z|ccklmGM$ z_bX4M`ce!Ld7U2yklO=6#qfksW#`)TSQW}o-MA0D^of|-ajvjJbufk!LE5k14v z-mlq9*Qo2RQK%55w$3ht!$<5Lo=y9|Y;|D&fZIWlR|PUV<)jy_JZqNze7W=K^x|7u zQB>v-`WN<(*%&pcmhDRFGcz^famGF>yYX|7a1<)4ciY+8z&4I6a^)+w{8nk!CPW?@ zh#NnB|NZK+%~|4;%1%n^&G%1xJhHFuI3JBf8tC#;9@4X63&J=1S!|OmlP`}mCw2cA z{xWQ6y}BIRzMi3t3UKB3gPGQIG?}$m~?s? zdIBAA{_a6&tpmL#t3)`EsX^d>6JEE&{7T_a_xG#FlRC7T{|Lv2@yhe-xSi@opA`0K zSRa+3DUcW?yQPubtGp@hRqt2PxWwlG%ca>{egpD@=;*HC3rg5tLV8-E7lZ zN%`HPs!{}-XdfWYAzv$#5!LI6pLZoSwdcAfB{wm7=5{fqG&6|LV7*i*_cGRB>_YFj z?JD=@U+?Dc17(}5hn64~eBU;Uqd#|lH)u&DGJlcv@tlKP9mb51)}XYW(>E8b6v^RP zOZg=@<4fY*%Y08r@Z|>3#qL-)49*wD2w}6?Z?Jkb$XfE3 zpz$QM?>)d{CP*f=ZC}&P`7}(nK3Nb8(xASPFcu)(2eaiu+IrP&b(Q9w`u@DVtFCwF zCN>^T2M|Gl^O8^t2vtIVxstEa^vITvCqD=BR^#sl!M=4$aj(#9Z5uYkK?<`NM3p`oR&g2Lo3_v+bXK!vFj^ra7X3K_TS6^`ZOxH`jU(#DDfag z-7gyaHJqg6`F(!&CEmJ|>^@H5J{*`1y~|}dW5x@j&Yw9}RG5*PV|-;h39z<-<+!*i zn2Wv}{!ec(VsF(Z%%dnC{h+1bZk>0Z7EDYEUzIzj5A0m5Kk$?`gT9D9nT4k|D`#*+ z=0d#8IT2<2@vO_qlCK17&K~1flq1=7)cAC6-*4^jJa|l5IFPrw-!OayWVfXHO05@$ z>|d+QMjm#6@DScZBUm1bzY}Cg7f`MZ09eMog-M-)q$8BqBmV{cd>LkpuvnS_y+HEn>v;P8_s%7Wa zn+UQL8A;|NQlH9*6kh#TTBp(^y2QN${3AoNHqP@5=RuZ#|NfL?wZS^Q1CdbnjROR_ zV{zWdb_ase#@Qqe_xjo3gh`P{2V@4ipiXLwD;sy`CgaIxY|3j`rKj90dtA<`)OCL zIQlJp@@keo(ChiVqqo^9{K-p48%jW5_XlHSPqXI&kt9@b&fx&Jf|677ihDe?)p5@_~Tu<`)x@pqxQ2^wy<$B z?6tt=W_}H`>j)FBmds#yvx_Ttl#QZW8+VE34`FfjPy>K&0SZznuN;zFNC z^pH`@A$avosQ1THD1S$G0=;h3LJQ=_H^0b!%Lr?4lF* z&JjQ&=XRullAJnTtf8>{`ZIgM`4G9}Gj^}S-e>STty=m$488y7*P1uKOdpF8cWXmbE3zPm`xbwURoC?NWVQQv5Sp05fSDf@b`A(>VLR!n45$*EzSqD&32G zY8g|{2=&zmyjH5*@7fXz?AJV!Yy@0`%2tgmb2Hi!7yQ9cQ_$%X~Kp ziTzC^)X(oBXH-9|EW6}a2|+swx7i|5rVz?B7!E4=mLR6f2lpB3_xWgL=LhHew8G;j z9?1OKwzj~sD)L@?yr(Hz4)sts>L5(q7wPKIovQ;^m{f@1m`n5a$a&W|zkG4RNV8Q5MknLKz<)mBE-fXTF%}#(WlsQi{gwo9QdtcD<*Y zK@9nfeD*m|=ko*`)Cv7pX{yjhoWmI2O8jfZrz7OYkPVR^n&9FXSEO&L1A6~GdZ)7A z0GBNCaD6RQZH7U~^iHV+>9 z3#W}bO_$9zo&HKcZXOPARr1z9Z4BdV6P%4X!lo~P+;bv>eeMUjxWB&{nG~Fq*uYsg z61#64zK}Vbu?8~eo0XJO2mz+wDug+)r~Cs;p6!ok#syd$Rl2oW_m^e1-JxF!mEi)K zG;>4HH@=m7*__O8cl%QBJa6p_5Iigv_QYZ5SRu4TG_@{VlIHi|Z^2hS_(XE4eqHS@ zS?Zn2*0Y4d;CO|7`D6#<*WXMv5QDCTWpcb=ciOnLC8Q;PT3_t@clzC_*E*JK#MN zj}1RnFL@nmt5%20R*5@jKL5gfRiYd{0o3?WgBt*7G}w5{>qi$h6o*eBicw}J+9BA{->2c8@%nl~J%Q;h&L`9D7b{YoZ0uX|{i*8b za50MYDAMZ{MmS$Nh>?U3xSm`-dG7o6fA8OyTu1i2eItTK-~fWwe!jrvBZ)plE8d!4 zmVUqa10H7vNk^#`C;T>hQjgI3`lXHId#lxN_R*p#|B6OM=d2K2@8X5%$#l=AQC0Ob z)xI5p-G=ad#&J<@dq~0HG?YufUVEo)*XFgPAr3 zz~h-omMC9M#`9?$#VwHlL%m%zmi)~fRHVK-G^O0w;^&}vt!q^a)kwxW5qIQoACWeb z)533ZzY3sJAfti@);9r{E&GG}?BAFr_nzbSVGn}OLup-w3Mq$wZYmA1nEP&uZ%OD0 z;K{|`44+E!-)$88)0U*r4~4GH9gf=_z779*h544QLQ0SFO(iD(;PHIN=C-a{<%Akx zz(Wp!iLH+v*W>dju_T1O`BJnsUi0x#etUoj8iiR^IchZak_h;la~H<_Zb=Ft9hm+B zRK_b4zn|x-+lLSKclPN_82RT@1!7nEg%j)R z`7u|5%o8C#+$$DaXJ(dpZgek`{&mQ+~ZJXy&XDihb zv09vAt-W{_b?ac3Tk5%r?Yt{}(s>w?`h~ ztnv9>yk8XNyz=lUIKg>*fWpV~EDy=ckG@?ojbEI;zLxd5?BCXRTEEah5NDMf2~F(q z_%(?C+vC$4#S-@8^?Ik$l`ba++T_cmCga~(;k62IuWh}fL>@d@xA=K=vJ*GaLf&=q zi9LTM6`vcg$wF(P3C%Xht!TIAl(8a@^a%H<^Iz`HGy-aYY3zDE9)_CSXn0Y-NN4g= z$?|cJa{2ycAO3pYhRNi}wmHFCl>$W6jiW+r6BM-yvc99wU^TTGsC8uFCr!NuTCgCG z&b=ZvT~hk=RpcC3`uF~m3z3)AX9ebqug0It7p|VT!~q~XFvw>2o(QD{5>dJL=%cQq z2F;u`1=H^eq^0QB@aU@%GU|BNYkg`x=8&#Wxvrkkcr}%)OdLY0fBWRWHl)%XBcRsb zAb9T>@XJ^IWgOCx&PzrWaibTM3pg?KiG7Am+6z*7uTd4aZJXfgSDLhDG(WA9xRLX6 zCF|F6G}rs)ERTo4=~oEqqI?PLM~K~zmTsr|2>mTns!M=RI`Smv5f0#cw-Tp#QLEvr zm4oOZuUIDS2$Pk-J!}JYoOQ_ClE=)_eTqJ)O#c}HyR<%}*na>ff+~5%!w0j50uN_B#L&5NE$ULJOA0fzFVFOM^ zzBF7ATZOJt0wIDFpRCswL1QfbC(luon!{Z2e(Q&!L4IwDaX-h{=a{C*`8~)8?7gK@ zE3Gy&;kS9)fSfNZe(3SOk(P*yXKSc6O}iU%=Q1(^UVWqAx=lum%6`o|0P;`q)B{N2teO|@%bzFVTo!~jBmJMdbq+N?!PwVO@bO^K)puZzbr=UQnVZ>}?UXVb%u7`PzFv zq%ijuxID4cs6OLMVD6Mxhc)7%WaX9~pV7wfs_Q#o!0H*d!K73G%wlnUnx*>t>EFlzZt=5f2zBs@CbO zlLH~VY0=f&wJ7%zpZ z?iZYTzvB9Bj#|esxbj^htI@*|@#GBHBs1*O*uv5~0)gyij7 z)P7TYF2GVwm`STzg;)Ra!DkaN`Zg?wd$~O~BiVR`rk8~dfVvh0AB~#0L0QSHh;V`@ zxHso;reVB?i!?`!p7_Iccnk?$O$vjo(tmqTx(G3dr`M zK_wc63a>B25h?Jw*k@Zi{khOggTVF{5@_b41x z{w5>wIB*d7LV!9P2?RQ%^P5wU)tjVJkp%&j-qGspD}Nd4_KG);{Kl+oNawd_jRXsF zDGkzUe!wuu2{LM?|DS2rfC@P3XYRI4U>c6FWFi};IL!>w?;caW;YsV_aL+nvyT3fs z9ikZzcAWOD&?(0#V&;nWU$31YBTsMgk$DeB_`oH;JalA`m|)u*8Id5wx)TN?{j{<2ZSySY~1#*xBJRYk^U64 z@|F!+48QVRq)7rOl7F~qKkE77vpHB*ev#b0@z7|uk^*@>*F&~ z(%MQT3>BGW9-VeoV19|dZjs63?K|UE_8$42PtwTz+^7cEi)5WK%}gxLMy^CA9+Ux< zEE?q`%B1?{6sppDKSi6pB3@*N?{!1Vpw+DKx|1TJkb`1x4!)(^9UQ^y{Q~NIziEGN zPxBF)K7a1NzLnJR%P|H;%KCf;yE3JtPr>hZ)nqjZ_!X#hO^e!#zEI+j7s66Fv;ks! z8dHE`0*-;@JpjRhEs48-Q4P%V?H#|EyfdWx;_t>&mf5rcjDCFQmvvu>9{|)n&oOZq zR@xY_4#6QsN|duVHZrDTP4LDx^E>=%$*9aO0AHMBK8Etwg8k=JGA*PAdOG1y43Jx*=H?)Fw(7Df~IqX)%IESr<$XK03 zD)Up^$eV`^I?7ErmD}E1$afor*2`Cz^7j+!qv0R5La#STV}JcdW;XL>V=-+wvkBCPF&RR$TvD=`%Z%Dl* zsE$-SC6sM>gs!)rAYca}vpHcH#k$4H1yzW6tDTVo)&~At+rhVO-+vkGDnn(yW;LVo z>u3M*VIcqYV*euXxBGWL!O7AjJ4yDgW=RI3Omlc5PI+9) zZw6AV`;v}|dkqRWp0@>NKr^m*5AiRE=la$@FXpV8=51z~kFY{-LJ4cH@>YqdqengA z5`adcNw#qVHr1v4+5n$P(9iSHgRdoi{Kwvkt*-Cg=ePej;7D#uD0Zz>yX!*>u?2lK(BmHK8_=IM2ovxb(;HI*O=BHh9R2-; zTI!JEN5sDPT1Ic_z~31BOjg83_Y_?=MPsE$>-<&&W^1=sM9PuT+zO1>039%#-@S=N zVUN2#uon}n9rD+E?w3Oc#cFSZGfG0>uctQnyTjoyR{PiYR5P~U8P-G*4^G!Z{H@|K zbM5X#beY)pZghS`Egb zghu)qlHB*Ay*q0ew4;hJiA*+VM0cEQt!tj_-+1@G@&5H5NwwN23@p_~=GOyq^W#0( zgimdU3o?4P)ckz?&(|`ISh{#()iWxd;;%2<4;4Y2lLqxsq}STAI|BQRp{9yCenu;X0ShR(qgi-K$%u7;4|kY*+%G2& z=fb4mxJQe=KY_61;5?)^4ms3AA9iL&wk@2KIfZoWpfKU;Lid;`)6Y~6LTqT&By^6{ z76qv)zg_uYAsMz`>3Ro`qxNR^{la+ipfS(CcAxnw3IN{+95f~Bgvn-Kd0`n0Np5Hb ztExS9j%f>~N^nf~y3sT%f7mIWhw zPxdY0vrV@&ZMCRhna%d_^3b{b8C_P*$$nJxw8kw+LeWH*lX}+nJ%Z{oo2hId{XJxo1*&lB?rPV5ha6vS=@OGWbi6psf(tbe9KE%3oqV##k|a^Q1?7Fuy5SiE9kd z!;831)rVr2FIk&g2HUn&#Xqy8j^66!r$joC;L4{Y0utcUBz;1e8#{2K z7ia_Z`yRd-V=LPA{bH;KIN!#D=7s!N>Z73H(J)eG&N$D>4wsv+m&bivZhWn8gC7cm zFBryajPKMm4UBj9`mia#A2Z)&S zeH=*vp`)iZISHL!&gziSJm0Pd2@XRGkmgIfYadtTPY*%1RlbabI+?2^H^uen|8W-U zt45>-`4s%S06&mB^RDWl1ca3<@{HW3a241C068G z+b*;>RH|r4S)WA2_lVRN(8$A;zYJtfRkS*++?eJzL{g6+i<3*_i&xx-TU)G_&#l)$ z=d%}&ko?P2r=HOhSQPaiQgDU$uTQuvFK&geTG)NU7=ewGZYY(wROS2jYuuXTVgf*% zOiGhB`tZvZJUYXBK>^pP<=I$zb1pFo_?YGZ=HSFxXZfv^T5;dz(NJ=4JFOX-d<@;u zt@RBIfHrw+u*X*`0qkiLX{GTl4E0c5Z`?(c@pI^cgB|o z)<4sYXYXaK(~sf%?zNyKosBpEbX{jlE%WPXP26H9niIu3ISdL1#o(}a&hkzm+oH=Xt?_I<4m{StK)P-Y^gX3 zf8Vw!*@9E6jAR(6jzw%Lc>xLy1hI`IWO$e##2>24As-&6_Zx4SY^YHHx)&VNccT!> zZ#M|uf?<5*dhVll_CPbKFYit9aDY%&OoV-eF%YXev*U{(BIIoE?12UMv^WFOcTjVC zBVY3BP}TBko`%yZj7rngfs3t$RZ2yvgQ_dr27nhPesyfS6G{h4AKImq5uSS1idtVl z#98II=vCcBgRhM2?FSiNXJ&Wf!7*2-)W9D#-QW5mEeVKUQS{B&+CWWb(T zU8aR8K=^n=3*G1ZLFA!jh>Xl4m^*{wL)QVO$D6=mdOnC+2ku*e6w(VsltIR7f}tE9 zzS$rE974E3@3j_G5HHtQKicw)t`}AYAVk{#iAp;k{3`A1kj>Kg(G$EDSXV zu)ssnr|#YZ!LigM4gn8ReUfU`;8LVIE?$LSSS%mTi*Hn_enCz! zKC@p}lXmt5m*9fF|4`^k*CY8Z0zKsIz!GMH#tn;MlQ&>~J*=UIYLOHf11uMoqzI;a zecY25y2OBkJ|@;_Mx4FTd2!-&_!i$d;^*uPK!Mt!d|q3dX+tsUV>l8kVfyCp#qo2W z>%9(XZS|?G*pob{YC#=rCU_W#dP&15fwtQ_3r`(RpvaK}rrH{pHc4Vr1{7Pjq=*v> zi>!BJc2|~KQh3~O{Se8F&HL1Eng&?ND5Ke(K_Y1SMUT;$-jop_Ubn9ov2+t&m*|c0 z=u9GiZy`R}SR17=N$D!p=L>j!s9lm5=vvYIB9_N3sGl!uB65X6Up+2E%NOhO?GFsK zJyb9ZFYYU>k;#7H)&u=vbX~v6!B4vxix)7%`cB>>_IXqN*zxg0-HbGaFNNQYQ6gHX zvbrVSO&yc=eO_fc=w`yV+*kK1aYz{8?Hl$s7I71HnGhx>o%)edlN0@ZuYvCnUs3gW z`{Qo;EVa*WERiZWg8O4IiS+wI1b-?u`o00h*!?~AA7P^E_obCL?f1wEJ3MNoEb94; z=h?h!`z@@QtF3~{`UqgW7xn;ENo_e%bbXY&m>@#HtM2b`R0W0HFYXT9n0XT{q^IV$ zJEbkUA}QqdZW1GvG?nsv)tQC}Qg-i39FwPpCa@V*gC~IB1u)^ybAn33JlpUhD&K6dAR?szxAt}oR!Hn;bGBMGS1U5dr`sk7_=k2GI{l)jb@HT#1=lzi4urdC@ZTOU$ z8T~@XTFMr!ik9a0T$83Q{SIx#qeIWUzz|RYb{A=&J$0vrR77xF*Q-p%2-oI^Fdogb ztr#$FP-?)Y|6pO7KV^EBVRQA{V~4^CS*eCl1N3v{mkkOwA^p`fY!>kbXUNW%B3))` zdFa~JV2@Hu%TSHyU;jt)*OO5FMR#^#7tY*;H@_ZU``v!s4UPXYMVmxJqwlZJso#@$ zXoYIm8aowCT}1Ops0JzIsq2^tEI{rRc8PGQ6IxJ9=Awz@vgPW57)SAmScI%dDr|mx zhEHuz`1EbWlEf{T7|x#aV^ow&^w^EaJcBEvXV!%`&mw1sjvhO0PkiZNdUjC!8zXvE zC?BCxLrKJZo@Y5^Ie#-wEB#so>1Zub!0GLY$@L(+3@mvqJnBE5J6uR% zcRzXUod)-*^Ecim1whsjULG$n#9aoSr%(>R1{`L7LV8aHhPgk$+O7UFm%jx&V4FZc z(2LKBxzewFE#}a7cgSBCzP!H39el+=<fJz@^g8DHnB+#VxBn8GQ6e_Cc&6zbvTSAf0iR{%Lc#=j9C(en;R3A3?Vl$z8* za%J%(oYer8>rF}LK6LW0%V|Tmj8PgVLk3TPB(GHatY zdpu=9)n0j2)uedR18k!w$#)l}w4IP0Io}hi#)$htUIme%ndhJQXZKp?N>I3Cx%&M( zj(=yMS{C2;%i%oxRXHZ`l?D%0$Fo3)CZU>P(p=iIVpp1CPG8SAD(&fZk1Ue6wPN(J z@4iLTRf(r7;NSDYT$FPn=cuB7*NFqr(uR4gT__eYBC?)ha37qQl^2t~&iQNlYu5V{Y=&HVOo>~U9Zbwo7~YPceJTE`D|-hnBY@HzzL#{lhZY=6(2@U;_aTHElmLy#fEiw{;ipYr1l-=+v7pL zQGbwz0`@}&4PLZ$FZbh=d9`E;u7D!0PT~H2Tl+d|N5I#Uv%zy}TG?*zN?_SWz%}qz z4Vz*)qx_v|&2st@k~#L*XYBJnl3P5&FZF9_$NqZc378@XLZ|=kc~j3&+`3*D1@fTN z;O;w!c&oSjkEK|vSJ)8==AbK3fg?)5c%juIB`^N_Q$IQcnf)Q8z_-g;A9!gb5YBsy zm8eX6=DZ?)=$>fiYE1m8LVV|(km34l(?LrSF2!xLMY!%$D>9N$0guugdk^G<-8C=9 zq4OAV+a^Q}rCq(-2718hxAexoi{scy3{$(f`t5l8@#Ng!^ZU!yzSEU8AT`|3V0~Wn zW_1w1QR>%+w`m{>XuQ674Xn##S+GE1`U1SL<01>;4mNm<2hgVZfwCoc_+<4e7<@)p z+j|$Ji#FIaGYAd5o9=?|c>C=-Ubj_BQ8|gwG+OI;hsQWKYIfi&l~K}w(Dz9G)(`NB zcGFeQoG$RHApH&x6i}dtt0l!1hw8g+qAA+b`Pj>aD2UNPYeFo-eVjJZ{!g3 zbdz98M>Ib3hoZexLLn(k!xs0TY_Hclqxy(14^n{Q$rB(*P7Gwj^Nn-@1pj1`3_Y^H zQUhnp0=l+BCoUk} zWj0^vcYAc#Gqju=ZW9va`H962>zx{cT_v@OQScEoBRIZ1)rrfS|7_nPbyNYf1hM94 zzpA<+1N;LlJ=Zl~7V+I4r7b_2N%M1g$Kf|PhYQim*Nc6TA$|fZi3;2xl=OrbIwr=_ z)Jk>oS}W4Xt{KYT=MrsAgpH%iGkcJG7~0WaWC74cEry9%?s$K!DzA=|{88DGnu|ac zhM(WzE5x?4PmZ2s<#NG^E(DUZR=H6Z50tx``E7IikbU3-{L%^r)cIUDw|<`;2#<%u zfMHkDxY>B|tLhj4DBp<)^Qh|+)_k0gS4}kNf*S2C%ihW3|@RpJrtlQ zf(MtSGLU16N%f89oZc+Yt{3ZEX2n>xnTqoXw?d%SWpQM%WXONkW4}KxH%MMZWgo^b zK=6@Ow*J{vqPlxD*h_wdzjhv>2hWBBFDQzeqo|@JYiI3)AJAM~(E{gm#QgY_BLU54 zSfuy+2=swim&fTu&*ZUf5d~&5(jY9VG9X;+S6=(1i4I&P!{v^D@wEBUirP7`G;;^@ zXr;XZfoVpBd`Yx#KNG|U_|PQ+uSPd*$UDy%4}2z43U8o(SJT!iH2>5Ob#?0pN^X`A zkIBQu3QoNHb$B26w^sIu_f#8}&c+|Y(uOuP^Sef%@5|O^w^ugOx;s^d`LDTPy9dR{ z!@%aX%)dp23QR3K`&?0~eck;7b!)`7?V1<4JU<$;(Z0S`+(sD|GLVh$-9o|(=n{DB z?#{{xn_mY!6U@fZ#qt>+bs*O{`4%5Juy_jBl?o z(|%+ZxJf;u=QrR+gE&JHj66*cJK&GFSQfHUcVky;i?CW0W*!`}RASoW@YwEvxzcz( zv9I$?+-^qXXYOBgpLarxcyQBpR_%OKtYOu(+U~Vo=pyM_j1j;?4q0$H4tD-R__b2&33$kXCjY z)U{dNi)qkgH5n3^lWjcK@IyK*xz{gLmtO(y$HRe3++ zBV%6aV+EfXSkxZFZ*BEFPs1sfg_ecolCo90&#-5?yOI``JL0=g-?NhEL&k8{@`Cgp z$85S+8}ip+2((Oxd-S&yv;e0o-ldgz`PB+HV$&#WShuNvcYuU(Nv!VlMT|xnHO?Rk zIX|}W{6*~Z@JQf>^5>asH7&)B#ip6Lh~XEgEo~nq(XfvM*cBk2Kd&#xqhZmLh#*fl z{L3nSj9c-`k{EawI-oDnpg3no;H&IUqFV$7b$C}vrEj54TxIx@@-0(Ci=jF)UFqG# zYgP=gs)t9wv6t9MV)&prwD7WH{Q3r{g!D#6KDa=dUP8dKDt&#yXQ0iEfG!wHA8bYJ zC$;P(tPLSd9L&KMO^0zHcvPQ@+DCn=DW(I`7gWXY8avO3Q13m&n2iLst=a)(?5Zzv ze$k{RTHQ;u9+Xj;vE1)ca@`G-$tcP>G2o?)0G2CGoknc{wHYQc<@5Bj_g zDSt1IkADk@{dg*%FP<+!SH?1YSj5P`q&ul?Q-RixZ>i{4OBfQjq0$s(T+6@pe13>; z`4W%3p4lsSMMc|sffkFYyH0U(GV>J6*8ZUYF$3xv{!H@1M*KvLH6Xy$R~=&h+Q7hU zaYX2IMA*ftnA~({0@`NU9Wz&Rr|gJjyf2S@={5MbHX$L73@38)SnMd74M-Edb=fS; zHA$Go{q&N;5d9=d!cN!vksHjT2cmEVA64AY#&x9a;y5H-N^rOBPS|n6Ww5k>vlJ&bk4F;_Mr}0BdWqTSc`l?ZG5w3K@Y#MdD0!M)GufH_eTF*I zS-)Lz21)}Ey4o=t>ghvy18jQ5=(yyj?+3V#BSF(ypTea(bXc0=j$oDVkrSA+KAxXI z@Ft!u8futok&&<>8rSPx^n>A#OHxJBtJfn+s--@l*g_+KROF*KU*Gl$kU_nv9JRRcVKeyA!gIUHp;R z9+ME7dLM8H-?k$HHKz1Q#f+7!A3OdG5%kHWJb)=48 zR{GLRd81^Wi^rcjJ*3%=?&@zXDxQD`IM_q{M4vih`W&JmZ*K!mmd^$SxuL<@o@ono zq54-?&{GYUXq_4$QJ7Ox%HqDSDRn6n-=)3I1nR5P-vm}pHXp*eC}ubD!8vMM^A#;? z(jGK;9TXVn?wOKLpJ#dur;cs53AziZP^-nkbiTQ1zS7V4l9u=D(Gr{VyYgOIqGPPi zqqAVHw;q7$;+Yavr~C7xW*_S*Yl*272hivJ6YegtA{}%U5by!w@_h8}rQM5O6Af;; z1|hVRw!K5fpOkQ1)pz&)JV+J3v+v}S%#KfbIB0zd1&Aj$20rMG|ACC!EPptmO`n08 zRZp5O+v6UY~mo6*M5NeJ&%*JrJlYl8&da^;X|@h(0z#(;2|ZTU2zcGJUw(FJzv zv{AV#!*8jpgmUA(^w}~{OE3?2#Xi0Q3_E=?2eK7x4b|X*{PE0J$>2?dw|#{h-5mbQ zg=Mq??uVOz-8ph?udPaOxw0=KZQ`-ftaQ9bAL{Bsu!5zmUSj_Jj|Xw+qk{gC%b{q0 zdfIF7J+CAGAdL;^ha6o4)rUTmIwx^FDAI1L^Y=_WO)Zj-4-?;zw<8N|1bRvH(7WHV z7#M#(CR!b75BPqsCvjL9S`Q?A5z^bjQfGX$v~xEo2Kh2uVxL3i1QxuzTPH~qezlv= z9t>TBvS|%!WE+v{TMP?jWy%2Gk?dZLYg!Ec4O=~uI>ip|&de$K;p*LlRs_``En*SQ zpqLj$hASdg$Y*1xzj12dH8~trp<&gEEQ7cu)i=p>Up95&Bu%1V8nE(+BnjSW2^CJh zY!77VxuN^glXPze0;Fdn)@|B+$kkSezZJ7NkaMZz9q`t$U&8k(dyAu2Hhui9f97yy z^y+-c(3dHj=7nk8N6Jk66Fb$TIz$jeP5TJno&(XST^!&>1|90E-voDM>oH9fUeO8- z9NbZ9xO6BWQ2O2-wh1t(S1~u+4L*~9uZ^t~neC%@hPUgIIR#aLPCni}lFE$|ZxQs# z24d@a2N8ihuA=Cqy5fJ|FNxE8MVZDJD&CgFI7z;anenchWTJgS^8ErNA^ip=8 ze3#Sl;|+JDv)ce99e*bse{M0bo9(Z&nm-CqiO(O(lii!(8cYX>4Xj@y^OXdq3f z;wy+C4Hv0~MRy*M$uGxo%P^f#b< zTG+~|z;`RqHz!RyUZl#g@~iq!Ilg$eO>R=Q0hP>k}CCT&(Js7@k>O<8Q!|$ zYHHl@h`cu=I;IUXjMZU|!hs5cAV3WRn^cK)0p@u=Qu)3^B@hC=H9A-J3?bhU$f{(1 zi|5Y6!A#2Me$;LHVkM*iC6QcTDRj`?Mp+azUWHt)-h72KDJ}7hL;QSt`Xwn>p1_`VYVRrvr+J!F_aAygvFuzb1q);&8Z6nRL5Iyg`Y@UN_Z%%^S#UqOdn%6X+_> z=smM=Gj}CbDhJ141CVR*b>H`nCsM6nklGX{xf;gjm2R9Zi@W^Y+XVC)^Yu@_or!Oc z7!~>0D&Pt{P~#9RKCplXs+K%$I8cY4O>D%aj^QBZj9pg$eEDZYNJDw8tuuoh4GQ_9 zkq!s)0v?VH*(()*r5uX|tT#4>>=nY~hszc!{)rOxj)WR^uIL_TsM=5WhA)7Z0&Gus z_VZm5TRk>Ky5r|HmZh%mv)aq;FPj#g@bvnftwZawG_x%ePkx{T-M+Dgz_v=351{|H zWCsyuGp$-=;K!Qjc;G<+ybAeKhAA(5Q#D_$MYZYjX4;_<*FI-446QUFn7bACk0a3x zir(7SFS{yPF@ilaZoNoRagNrctQ=uo!u7lDekkc<2qvo`uv-)F+-JRY$4KCio^sc# zD?r=ObANeRUqt6wL5tmKX{9RqB_*IB(4NmH;7ZZ@1gsnq>IBOOKvBjomoM9E{Jm?# zR8hLDn{bDyC-h>WB<{8IBS8abtOHumfz*{7B7H4PnZY?l0-bE<7?J5T zev3dl^H+KM?2y#Jhc(_ShLZUj36HN8y7dcFJy_8H&b*=-y%Jx>1X;mCcTs)e6a#9a zXc%ddRhI*A@gszWlgoXYipZD%_VK`y>=CczUyrBO!ee7{zdchdN|t453jEH?=R#dV zH6_w>UA$b*bRCzo0{>jTn33BgI+F6r4N+3*8G8|bG7;~73g}u zW$wa1#cP)Cl>+LyBId)pa|~|L;7d;4X3^@l%BPx8f&@z1Vso?f({vn11VR+Mk=CS; zw2&SROFQ)j2*pwX01!U1az?0?8>?VK4`7@4n$y7SRzCp&DyDgYC-V4pKaN3EY46Tf zd%H(?gtw{ubApeeL@6bBe!m5!r4ip?mdr1YqPR%MC(#X^_(H#uOo^oNC^>kt7cS*| zf9zUNLu<;;1*L<`(zLj_q4O)CWgV14kTCgl#=d(>MpQ$uFX0+?$K2_KHdKB}>Iz>v05RGIx>DOO>chCZThKMZo--GrZWTqevkdXQhG~mp5 z9Y86wFF+nE0jTXw0q1Kta`Q$jZ&{4*i@(!pXLo;KXqR*dQK{EBeIt@m%fo5C{_V}7 z={ZkMn%yG8bD8c|J#2rDhhcTmx+}e=ZUCEBep51Xs8m!RvE$ zAM*Kg`^ND{o23R(LrGoDh7^{FQxCsnR%e`CIQ@oZNztTptnXbLH0VuV}X0&#>J%bv2`G3wqzr zMPZ}^JRM;I+Ita*i~gVqkf2l{*zWo1CEl(w*l`w#Cwy#O&m5pCO_ zSbA+_u_ngiegY}i#qMvG{u=8b^|AKkTBDEX+Bquq>u`e9O=^vI_Egger zxZiurE1@{p6?X2{5gV$72hk3nV(hCrw!GA_(Rn(}V(IpxP$9VN| zKy{5#aCXb>I3!_!N%%cy9u&a8QTjVFv$*=WN}9cFatqsrXGW4-*EjvSGWh{dY0mhbH{&S{fQS`yMBGqoXVmL8gtT4 zoexO+mCS~CL2y9p(UUDjJN-WA`TBuDn}Ir^fIuryItAX;ho5-7UYrB`3DHDS+UpJ` zxKvB!u+dmsY6ckECaR!ydX6t2Ej~X#=^TLO{px5f-#R|2?-Tz)#^*HVzhTWSc?J_U_DVd!135M8`O+26sxVNc8u< zd(^@_o@Im zS0LtWc_K#ei?oqM8UsE^gn?R~W0~!%Y@B_G_4oON%}7q9>V!v08u8<(AgW0pI=u-? zm15UeCam^2y1Oo~x+m0Ey22o~OiM6X7B+|V>FGt=6vh-E_5wX9)v8TQ>q`i#svKar zx{kb$au4^L+Am;TJDRupqanaBL*?08UU;x{@8DUH?Ds=myJ5FBjt90p5<`2CNF~)cv1=GnDlFlMi+-^$)F?ojspag- zsrJU;<$y_A$GA30_4jWJ^&jgZ#OiWv;=4*?vhN-%9%r8n@@ytJrui$ch z&rqL@2o$1kUoZaiZPp)RmVVnhCy_lF0V&sM<$1v-LfYkCO9z}(`PXLVX{jB@s(?Z0`X*d=Bypz03IP5h!e_akece@gJeRV}84`Kx zyFanpIt!1?u%=xO+3yE6futyb<)5_-du;gNNwWL<4p;deFqt9h4>gB$r!0t+{*W5L zjjC!XKw|X$3O>@$Bm3aY0D<$LiNkFCZC02Yv^a2EK|{8{wX~7C1~uYXza7c* zwI(P=xrEB=E7jstf5by~Y*NK!EK?UyYv1#q2><=`Amj{VNNhR?*G_naa2T?)NvXhF0)xa^Xbhb13~@SSqQxzZ$`7{9j3@-u;N z>OSp>Aocyc4@r|ME9tpQ=xcSN!0pR6xBk3c3HiJPo%PYUzv6+je0P2_ zy3_9p&~76epXpXdAC?LFCEuKApZ5)Tm)%${AkK;%h3d`CM3`fU?w+xyje`+{!YEpz zXWxNq9P6I@kA!@^H12cb!Amcd^XPHLr){PgN3u%{k6=HaTp44jE#1C&%@8!W^!`OQ zz4u{4!YAbCd6B&dKXS8J))Bl;Dj%3=3xr`XsZ*+aAZV}qc~_|+iUW=>jP1ZL$w!7t zH;l>n7}1IRs#cuat-B9@HvN3Q7vc61Ur|SYKMam+{__zr1?V;{Py*}Vv_d!^>r7hZ z`npZOwU-l7EuU3QP}P1KJ5iv)z&_sNo?O48 zupB%M@_GxS5%>H~-o%q&D|a3GkmW$kmgAH>So{UNqCzS4Ubv^(5EpTsK5c3b=#>UT`hYOMiIx)99O4i6bYF9B%DDL zA{AC^WWTN@6)q$#4j*_FFY_MAXmOmVP5C9^s*zs`{-JVp*xJf}1TO$aS;>5!)N?l* zq)u%I-;J%WaI-AGUHO%g@AT{4j#P{<=h$y=AG==soebc_9vP?+tXZ@nl3oqtA zmVtkoaD{F$ey{Ir>=}JrQB1R-;VJ!jR@fG51O7!H?xtkIM^>xHHe==H@~boBc;g3D zO#zYS(n))G8=0+yB%3Drz0%JELHvyMKPxS;;m7;Mm6x{Daf+i z)h)%?OOxX>Kq^wU-ne=nMb^q((gOmtNsWY;@jR~{ma-z`YIvjLkew1Ba9*RKNt!wG!DdYqGAQT~f53;V-|TWqh9=0;(GB&h zMWBau~L1jA7%GBt-VZ;!h{q9L^77sz{t0niLK-l z$-<0$Gh{%Aypw-`mNE}>SVOf zBNS_dm-68{tMZ{vXy)iV@;SyM&7uv}?Awb<2MpupS_KDI2+PIQ($%`ahPUbza8!n_ zxZaQWsCabDj?_B&T7GP??qu%?>j*EwH~=Du5dlr{dL_U;9oIV-R1OIvTr;8<=1O+S zyqF9wH+?45Gq)iOQ`RSGNd9xU-iz^3zNpc5D7--mE;b` zHBSUUgzlzJeegY+F^L)A*O7)Nmzc?tOv$P(Qgb5xrfdQ#I%gJPO;-`%EiHQCZLXNd&amz6eOy3AEm_z;_BwS&GJ}-oyJSw0uO1Hgy|>3+d0OHqlurjw8k&e?&MWX=jbf7k&999{KfY0E( z9v=78%Pde{QC9u}o+(P)`&i%$SgQ1CrV{tF)dhFLKY>)~`W4MJdCw3!cWz4!WOwvv zj$*bZfQm2X=_Wp@uS(K=j&!d&gav{j{ifvC{FMezOy0GHyMyLgkH9&6(O=k5TABCS z@;xUY*S}w!r&ur2TapC^W#G|?N`FaXvo59AS2u?*`W!ho zvSFOUN6^vgrc9LEe#1&!6~$&B9v%YsehRzAKGz6*=N&(0i(0#&@6?B@M<+ucA0W3s z#a=fSVI2TU+)qCv6oDnp^L|END&Xs4dKy3T&|YoIagr6A@CNpVI=x@0jb%^&ukRjOal#o=3eBJSWH z#9ix$KwvbOx%zoY&1Ji&Hu@m__KU`0wu6lKm$mAH3o5)!N1Ndrp~RBIuzg0@Q{Ici ziBny9gC)`U_N*e|tWP$E8cnM7bekr6Ip#;wA5-SMzcMBh`16UtwMV_T26FimUMDQ` zH@@-e=i3P>=sMEk@jYJt=V$Y8-ZGgNpC^8G{$zlyW^Gkm!*2WDcRUptz(ZG7xQW#; zR`7n2?`LyLY=())EZ8UfR?Hy~)| z%3t+ieN*AJF;OjWZUIqOhP}?##`0{W3Eh~>L`RmteL**uV#0BtlI42b$7C!&F5+HEXJ=%9QK)- zT`=fS2N3gtPdGIi@KA1b`pkU7R|?UNM~jr0a4T?!C)F2OI+`$ z`2N-hb#Hh!fiHO0d3AUox1~Zjm=rWOHPUhQrYh{DuqZa%OPOc~f4d;_XOzdt$Jd1G z<2Ycz%{J`C0|~6v{Ef`{n|%|#;8GOL$vJb&L~y3OmOG}G5F*>XSvr2lOFwh6_7MkUMap)^%3Pk z|22e1dDZ+o)q$s`a4kR>fkS`?NF8q}KyT0kmO*4nE@$LIMexe}1X7__x&c?_9P0TE zFr?^d0$1kQ0XP~;Ecq)DzinFoQGUPjcfFrM-IbGB6!{dZWuB0T_S(=ddJ^)ki@z9n z@AkM*eT}cvEd0;4UY@v^^gGV{eO^lkzL*I^#y{)YjruzB9g(0|2zs2^+EJw7c4p+msiez^cm9cu@Va^J}g#oK=Afq<4CN`=6FW2di{ir{<_a7 z2n(e-%?F74$6MofzgF%#G2$=FNB2Ia`JfMyIm#m}9i=_l9)@YQFs|buBZj}nAhZVY zb>_JEOEiK?`{!{lYI4B@_4m4LqE)}2a}4?$VTByug)Km+)mLVkrm8V3im?d-{IuE% zO_)tL40J?lYXKPld}+_0A3%Axuilv0&EI>d*v?Qv`ACeuZeJgBK&M+Ac+Gv9`|O`v zXT12B`?TBr`PW^9LHL{Be^sx4&NBJC_N%k;d+fp6_PzhDyk9@HsxtpP`IrCN_x}F- zwI!+5eU8STSO3}y$52ahWJ+u`fGSr(8~|Pp)tnU^4Pm1kt|W50sO4Iphv{-pZXiE* zTR&PpO>!xEXc6q})sNF6CpF+dp#Z*u)~~Kax^`e!hAYKaIaGIzr>5o@XXpwKg-02l zYuG~Jh==AHLeVUjhxX|L#qls7&zSZgW19PsUXNk*hS%d~%#@n&MxBpjt>>{h5e1Bq!B^p!c*FX)^tFDMmTm(<9HN4n@mspE9hhGKq(yRS+j#RZY3z1$k>P{OXiFS^X?e<-fkqm?2 z9GZFFU#)i1|9l@#yRPfz-t|OD|6FQOu~wSc4Vt2esUNBQj20vY7icZ7JHjETNZoE= zJq0>S*hL&=;VfQ#DrZze(Kuc-pUqN^YZc^Uo>LI?{sRVv3q$A1QXeV8bwE9H&^L zcW66y?E9<<)X0l{T~l;@nju%k*ftjij-s$@g`;}>or9Qu1cAD&Y)P~1tm}hd?`L7D ztxi}q>Bw#-KwD1YQPniNAs0<+5FCu$zK1bs7{ACJ`LcK(iMEU*H*6Iu{`GV#K z0SRL|_ddJ`y27QX2}paj_EA7A4Z~fM_oHl=D74P3;?L8$0v#sUD4P%WnaGEl@aZIMzOL0>44rF8!WNFcu%?xidKjP#zbx9eQl zvH2DOxd8kupC!XR070&|&iDbmXZCrXFECa@`JnQJ%HAG3vBk^)%{L`nHnPT-$hqSO zBm4c#ASaOoQFgje;rw^C$lTB z6G)_`ZE6rr5oGkFZ&%Z}1(HOv%@qwO@TH3KGx_H@C-=b@K>RM_ojqy-A$t)hAUyi* zS`uZ%=y23WLZx{|Fbby6J?onFsT8HrdiROOA5&fmFJK(nmhN6fclozoeJ&lk2% zw8fs8(N`vGcsK=r2Ot=2Ae?g@=vPxLG=qicjCWXQH+p!C*%=@Nz+5&#wzoTZ`sXW} zkI!?6s}T$XfkF4#Ikbk+9z0Y0==<%Yc<6q{3w_6sc|I~F$ntiA>X6dM=~b2hlP5eAz~wEK$?drd^+9@qfTM02XqO9H)E;~9n6&}8 zchx-?Ah+Q5TK7~hLo+D99RcaOtq3k~@fs=ysSL6o3)3tqp^ zag6;AXs08?e?+FYH(}iEFZQ4dK>VQceST|XkAw*;_j$kbFBP8E0EBra*Vc~(BAH;@ zk{S_H-yp*nM-qzJvzcX7%?lV|~EMGX)oS&60Ep3`zt(;GE>N_t=iigONRGn#SaY)ExpS z#i2QKg9FC5nkXVGk!qLe(VDnFKVc0dAE^HyYH3SKfOdX+(G*uhT!+#7`=dR~aXWxr zpZokMYP%S7s40o*w2!Rljx->z5am<0H|{fnW@skvbK=xmy{PQyt$@e#6F_*~Y3GIZWZ;A)n6@Iba}-;ft?o15NT}$RKPA&rmW7QQr;9z|MKHdpskJQOnnMYzflyg$AfL^6WP-hb7XE_; zlLWjrya0JK2~XAIVowF&!?~8l+4m89=DQ_S{#{Mta86$kiH_8qS;2TiQNZUxZ1gsN z&445KiR^t5W=>W zOGPS?5;P4tkJ2>KeA`yw@2_tQhNBdI7zV*$RIM`B?>Yo9KBIJa-CnRiB_YDH&g!DC z=y86K0}(S)8pUpKZ+V<#eOA_$LH2XO6l@$0a?2YsQQmGG$} z_0WLduh;Cb>(JBY@HU_(;=&g*85-8vSnf2W@pfH=Il@%@6avaVqLRw&5txXkQ@jlu zKfqzkKyJ^=?WV#Df)`r-!NA}dAQ?N}{dJ#P@5}sMu{6{1;yz98bpRZOFOe5r=Pjz; zUtA?w^?L+B{2b#tdmByyp8NdXcP&NX_J9mr*EMA9Gy?|#&T6+Hp8ImUwQIDTq3Ms4 z46ue>kMlf{C7_A)@c_hiMH{Zxx3-XK@%sEy{%_119N?kje&`c z!PWlTe}8k>)^S_^>w1nkc>mv7wSeDv{@1_H{g?mY>;LUPKL4K=_|HaGU0mky*q8UW zEAOdWKSbus;l3>m>Hb0AAJ7Af^99REun5U8?u^W*c5iD) zcos?cwi1P?O?z!jFyd0`a#O2c83Tk#1(i}xcKshet90xc0hWPa`g33V6+A^hP{Awq z3w{`($qxpVjgQA0;gR-ZdKS^^SC-+22BQDLWI2mD>q~5T>kto3etXk*b$odmp=yp| z{%fskGMiQgB6ZRR%1YP7zh`H%Q1qe-X!h<@X+T00L2sPgNH38Q2uRoq3l~&~B`C z{l(rBy!+{#Kb_>7n5~J;-~6L=f7O)k^=h-T-_qxwc}wtaX-BGQ)+c|eUom?RSqdd; z9xHv4*~2G9Vl9{|@^Gr$&Okff#&9ztWjIv)1!g0HzNhqQ@52OtFkc@~fny)Bu~lj- z^7NxZdAY$XIKI7zcq|U{QkMmGu=hUz*q>; zTLjbQ*jV|GZB~|Xf)w1uNX22OcnqH0*muQC0xp8!0up#tOynP20cw~^g8O;C&|l~C z<`n8!ZKRdLyf%l(+yZMz_+w-C1c?lx{N3}}(FA)`;2^+`1-CYHX|A$-0FYe8=C%76 z5a7`P!WNv&aeZ=%I)%h;* z+57(!8fLY|SGXnMk0F*hdp{`(f}?$5yE+4|3f!6c)~}KPw^Zc0u4>nK>#DeOV~zLZ z(X9fUW?LS3D|`Ws`7y9oQ>#5j8+LZ9>IxRB-CjfXC`NzZ>jhsj&-g7!R&}=kRY0o0 zCr3hHAD?b*Hh&H#y+V-EHzzJ&6n?4k zE&&yGGY7*sz3v_c^rY8cB*~4e{zahv#}?sPTvus;&JbB zT5eXwW__ULwhR5q69|R#T!AMufuwH_paU#PaK1nD)wSMTlOc0~|2@llPGnMJU5M`o ztfuy%cnoyUzT(b=ewycW``~W2$4@uib<3v?gl{>TCXwoB1}qnn&&1*@G6qyX9B3K~ z4YH?qlM=1iIX-G@!RblZubX@04ADn$w@P<)z=1E{0V2`mI>K^gK- zf4plxK3m|$$8!wV2|yk{`D{kGGUtgv9dq$^fqvq&$Va!?3{`%L8h{J8=0B_z=veEX z;YAqQqd|;&wk0-!OVhwY_uxlRpnqQEvUstOlN)=DkAcxEHb0&&(9yE%E@P?6HOQZB zzWc`Ix9gsNe#YPSB;#mLqWkYDQac-1cJO=Ij4C`}-2;W2HSCT7FMP)UBa*d5&3s04 z!9gjF)6{!@BU)I;y`PQ(>dTYB5G2+L!O7|pk67hj4?kOSBN!WdobxF(!dZ zpM7>faJ+p0S58G&pW`=AL||A306cKNU38cl0vZD|5OvrAIzlcVnxCjVk0y3A$x{oO zwJX2^JnRnNobkKhN%eieh2uDIu8C*ZS#6v6)DCiQn``l>0?+mV1`XsP&9Yvx?gjWh z2MD-zFB_M8hlI9=qBk<|h$xNQ2&;R*G8fD)9ICsdt)oB&W>woB-Sc%L_9Jpbu0d9y zz~^E3$|K-@h)I9f#J**YeHRY*hEV|g^`t|K=$iV6Yxg*H7s(n(EOBvm5~}}5wQawj zlT%ja;dZfI#YRp_*F#AduW4+H$`E*o|wPWY`*qv`x}oO zG~-^;Em2ucbfO&OS0!Sa=Ir3s!^IDkXDC0N5h1%74iHwhz~IZ#y!19QDmad_MUIhc zlHxQ?-z!f)^^S`Z7Snqk$LH2$!G+rD=ya2~wxhinCFQtVrpLoHOh7XXztQ(I;QF;t z0C7IAQbB3<5SkL%U9t?>2x0Z8-Py#Y#PD)c1DzOqn_4Rlp6}64?ia;&$ z010PVpO|kZ0~q69iW?3f#hVuHHQgiKJUIbzGM(E!+gTx#r?3~+e;&RS1%x1qjwk@B zz}rW-h;J|!i}SJgZxXZFcGr>{j_+L`?Osz7J^-JAA*Al@vGDPMUE8|tM^g0t=nCX$9lEgU9f3ckNsX6VQ0{G|K`IgOzbrQ@}}(7 z9YLC9kDQ6GLqs!Rnh+0wEKiOD789gjmJdC^#rZ_I>MR}b zAPGF~aKXoz*IH?3OOBTL2m#IXmKpuM`ZRg+is4s*R?q5N?DcW`@b6vYQh%3(%FnK> zj5tkD_Lujf4;ueSs5fR0(6$$=Kf z`?`L|Ab*&6^5S&>7rjNq`dfkax$D=5dffF`2`Fg4pcA=7>n|uaKtG9$inI*-+MiwFapmb0O5oRO0k&(ZVwIKy|4!gmFNCFgG(_5 z;5pa;xdmU~KbXlh&<3HdF9|&+skyHGp_PI;Pi`bt0VfY zz=bgQ?UxTA%IceUa-k)c0Enb*4rT`alSjm`>=#EmQ)`azFTvOv;I|Z|zJXLR03UAM zyL5@hcR+tC&bnZQp>;dxC;k2NAEj6+ubC@aRBj#M&{I4^kF*Dgd4hje-w#2}{;u6} z3IP@EXt<$~Za~k+IDY8Pjnj`eimzTgy0Sx4)Juh+t*q}eNbFnlS>+X#*F`S)8|XEm zG~T~dG4d8rlJ^3?XSQaQcImv38*TaVN9&vSoK)2G8-XAB%~_A_4P?K87aqLVdz%A` z3zTJLiCwZcfULNKWthhY;N#<|I-leDFh9&a1lTGelmQv1_bEn^bX<{nkU*ig)fqjy z)u)S&&9dTxo#C!ud9QbY9;3L&iwzcdo@rQLx&V*~xE9hH&Xp!zsnrQ1fSk*tdmm1_ z<@7zDWkUM7D9^{pY9mAS4_&^{>w@71M7daaRLiN4)ZV%-sDagWMrvdb<{ohmObq zGY0rRid);0TUVCOv`>N(u++o6*Ndg*48vz4%v^WfJTZEJ_R6 ztsvXRAKAHsW2kh`Sr4-W3?@ML9sw2!IE@KfG6z86*GUWH^@Qqy(!30qWe*1QYxt>E z<8y2BXeoY-@hX8ssiH-{0EFmWHb_k7DErk9=1`|g_kKP~e<)y@l$f2p*NXpia^j5$ zl7MUtKs(*xz8^0EQpi3EgqB(xm@FzYlY~A2Q1vZ$c@Ri7Z^N^l%#&^$-Og1fBCn_! z^%Q_(s8b3vNk)a&zFG%0-oRP#E zOCeZnpJ3Dia{4gu`m;`m00AnYwDz?)V0I{3*wL`GMA~d&S>tO@a>nMcN?=aBE|A+d zmoPzupCNaz$&bK#k(RxXs(=Fx3>qHqJ!lm%X!q_m2XH9veIJd3>VSt(|JH@>aPLz% z;Wdk6`@sP1iT>?-bKzn}Pg-eyiZvpKUEXUuBS4)_e(>+_LExnL#tMvCW{G0`v5kDc zy{9|>?vb^TI5XUz_<4H;&^?KZjidO!_nc5uUKFhY@=Mp!KEu5~Z7lWIcv$wg4c^y4 z=yHA6I+<^8Hf=zYV0$a+WL`Pn34N8^1g`s}1{mJIZ^gszxyYio-lP+Wofr{M-~$l% z-bc30ASib{$@IT9uDgv&=HmS00(NFi9{B{c3av1;KTdH!9R~}0w=Zc;OsnGwwCpuC zK*M)4!RVh!t0O&QD zckS&xN^=&G&*cM6H%naTI8ad5zt`fjwfWY3$w+@Ji&a}ckn63$>pj~+W z(pCh_2`e~@7U2Irlkq;%nALxwt9gqracfZQK?{t;7j6IcUEE`$+R}%?^vs9t%m^mW za#KOR7`VVZkV5aY=MES6XB3S=X5liY766NN4-6N)%<}2pa{{f$Lc0SbhgHb2+c_k) z03*V*;@wqav`|UB3*8m3)o<_s_i)aC`0^12-0R=VqG>PLcg{`?&;{ER1UPZTe+$7Q z_~ir}`B^`bXV!an=c2C9RpDAQt>k(vvo!smH(kf$_5o8B8GUzPy=Z@)M$mN_GkIth zJl&dnc%SlA@c#RJ8guqGU3ml?M=^O}3>cuh&xPXm{-sTr(RfVW#S9jGvoZ#Cv`Xv!7S@ zH_uysmxi~25c=DPmPilMN%~Lte*UOTYPp(dQJsUevvZTx{+s!^b~+93xR}{_poRlyB)&%ku8H zol(tUdHt4I0YAjbBPfqXdgr(g8@s(7+M)5+zxl{jS}X=A)w@6xJLxktFk!fl0qNfS zySuX8o!H!n0ZzFx_;DZXb;8An-uQSVT640(z~0+N2p!-MlURty%_&ryVko~$7TALO zP{oZ#fFLsNb@q-4&U;OeS-l`QJ`wMB;JN|9q-han-AP6odG`}h@~+!hf`uEmK4IRC zFFuySLoLs67+9DmJ)FTk-SZ_Am+zp}iG~4D)B40luNSj{;QPAMsJM~l5AI)WcpZ49 zq6ThM@H~{UxTp2;xiVbm7<&lLijDp-y}mEAkwhEW>U+N%u=IHjU&nDkkJM?u43uau z^QT7o>`vAXDx?Ys5F)vA_Nva-jLG2Z8IrSm|XM=-vcUHw@jFhu41$LtbCKa8$Jn7eKT$nASO;-WnV<75rX zHDDZR=wu~<#XkYDTRXlNFzeV=<8j}(dnjIi9n^ZoK(~{YKB6kXNzw)Xj^>5bc=6X1 z0a7-dgwZ{4%3W{hcrcCJgvOnR!S_@bwS)U)V4qEG{@CvMk%T9=m2E(5diY4?&7`~z zHAqlOOW5}(mL6LVXuTYi7ixfi?;>^!8t7UJi4ga@u1!~aGct3vbpaV7fA1rl0Eb>C z+hiZ>`N#wpEkpT*UH6%78E6Bo)yI!`0J9jDG4PpJoIa07$cx{6#)@EWL34j0;FsqL z=n8$AgDloRrhhJY(tTIDri>}XN8`>Xk5yB$Ug#XZbW8lcs>InzT=)E`qsp{l=-|soTI_YX4}zule(j`x(Gi4~k8Ayny8-9)UtKSbS;s zTHzXC;5|Biz90w05Hh#->_XU0n2{DCO|Ooy?rd&~t}fOg?g?9A^bDL^=JL0efyiG= z1PRSc&;DQ`BDGbjcS%3eqsLxF8k^6a2*wrXuLSszo-u^uk-m-V$KHGluzl<45K=14 z0ONLhDOD)nBTDt}i)?U=&bEH_I=~qSUlKyR`s;nJ(2jRMto=6jr;)*J_(&Uo@2uvb z6VO=a4a3s%v1G!go(0|=+N*jL$-P(bK5%N~XHTBlx9q$C0r#)_Ih>AVzofMFYTN_k zXpMyWdeYiSHsGcpkpsv^AGePA(|{ttw4YG(x_!^F{O$AbmAig+Y6wugt}IpET6*Q- zLa5)Equ}RR?SK2(@3C7jJkj_;h518PApED*Mp91jtL|OHi!KGDGeV zf7q(uPIiHP0L`cTI`+SU05Wei#wkillOx&Gmkl$>X;?F^2#?Wf}U&KiY%u?En+#Z#vCx z-V&aY&4%wG$#I>aOBs=L_t3p=-PCs5v$p#V4)$t@8>5}s%6e0o`EG;`X;N2P^*Z(d zRX2fD2Q686u!bD9SEw%Tnl8A!zSVnR`cl=p5Ga}z?DwnK1eBCVuns#Om~V`SG#XZ) z)9@ZXZgY&sKr43gLvzuClJ2+9#|mq+y>+w@58|}{x-x*nu7#x7bn4DI2*)M3l$4D+ z0q+u{;xM47+^AP6C-OJjVXv(!hh$jTCgnJ8ro8s>xN(_Nw7(ktyt|Tmp6Lu8IP72w z^LFfb>Dp`KY8{^6t-a2#>oDc$kw?W`YOe4K_WNPPxif)_;%I^EAG=>|Tlo-i3ZXG+ z$@up^b;wJz3zR1v6b4WcAarW1?Meh}@NqQ{u)@jr`k4R3hTj#@ zhsgTtmSV-w=|dnhk|;~ifH52d>i-DL1d!o zNx&G@_bt-xd*IO?ET}Xaeb18sk1rXyiGZEP@b^-pp{D$=@6&i%_wY0-Zm=i0jULii z=hNmRv&wkeZrErMCptxZ^jWm2=#tQj2iM=1n`3%{yvJtV_j$XY9BL;?4_$jEnqOL* z)8Ttk&i4%uG@5cyH{!XUa8n*e_H`y)0{L&qtfP`$q`HFXqCNl;L;*hK>OHn!~##c~$=HvqTWM_651BUHqdGOiW2~ z=DX`q&(kdiY`Axc4+HBMQeM|JJt?Czv~+LOZ$bNhj@!+~Iw`Ma(LHNo1nR!oY?hVk zBlusw!cA`%v}Q=AiuccRT<$iCg$ywX#zao>}H1 zFFU;+TSl3u)MMEqG5=Oay*! z#Jdf2`JT@dyk3tbbBn^$-EOhpHososFBVDl$}oc9_*34xpMXlrHR65ubIp&;z-38F z5KOi}=3m$M2s<+zY`zd|Ne9fie#x?deZ;o_VauZy2QJnU^rweLn@cNeWP1YO2wWFA zjB4M1RMOK*JI1b^S7j`kJX3kOwC-xJTgNRm7x%Bov7tYy7W?>4_rJuEh}_cD{lCLxSQHR9(>5z6_Zf$ z5R`DewbxlqsW*{PdA(iitUjEq`m(nHmO3Ux+;T# z`tH>i$~50Ut~Vh6ujhTd>?6X~U+1@$ZU`ZOJwxqY_gdv?sp38f79OIGX#L*r{JUbC zb1*8d2NpL5}y25R%z$nGEMjmRMh;P31nmuZ_J}QsSJYd7@vB-0{87_NTwOuQ&$=+=IGK?Uzwe zt5A4+)_1fnXZk_#hUBz_sDHb>*fBMkxD0`7{Q7Ce*_NNMIAT%!H2q@_Sl2_ntP6y2 z2ZZnlKfCJn`@@>*L)xHD?+Z`NS4iGC0SAYu2Du@qH&W`Ut7v^lEW=zNGZ~d|uTE9o z@l&6BlGBi(ak-tu3P^P(Bv$Ug%UE(6)l4&R;9p;T^G3}qJfLrwZIVwGJMRdEmZ1V1 zj>tCNI3FoHobs%8JOFB zpzOyRDWo4qTc(E+V*N7N&7Vv`JqM8b$M;6S=B&sBm;%WfUJ!hJ-FA;fq}lXDLDcT- zx@D}#R|dkWJ+>!h1niv$^5b#d?PQ>4EtW;ZJfPP!<>iMgV?X%(WOGFpr<<~$*+R47 zg)lRHDOZ(+%~d~zjLsuwxYz|Gf~$Pq%Xs?W?ghk?{{#Y&p{~5w=z&h6p`zkNUC80> zP=ZuFsrmyjQrtkoo{V>&!e-?3dzc-c9$LkhpTJJ#(r!H;_PK&T^I6EAfp5cm5ezIA zcd&OZ-wV+F?*jleP9ePq!DX=S444ZictSnA$NDLEdWs|j$tw{6(X3dm@0Gf0>dnra ztk$x+ac`c#mPFcF($YVK5v|R*ly$-$*>6?6ASE7f2sdxHYwVj9+{j8?4Iw+RysVt9Y z$cwM1R>q$SYbI?F-K>32@nYXbS3cXspAUdC?+IJ~a^P_euX;yHZrCquhu2FAI?34XCBR?%h@Wl%(7G0DUEL1#TEDZ{ zspU0J)qy;0Hv9b^eu9c84{I9K1-1M@6ndYcqAT!^`D(V5xkkl$6NAHiy2Zyxbo8ep zLWm7)&su<+cfGo7cAeyz5KA@o+L&_NkpK}=fAX*ocd>H#Oz$FR{o@-NfP(_8J-UrD zi}_v~$h{u!PrXQ3C)gO}vV3~Y?jeftzMawmXT1O| zuFQ_AI{uf-D3s6 zhc`y1Us|tg?d6jkZ;&R&pC%2W7Oa7IQxOE=MjV8Mu66~+Q?GXtHEs%^?l*P88K1jy zgscXb8UXi$&oWEgw9$vT0iSwL%qIU*;^V2jD|L6TA*f&xfTt32i4Q@7@Br!1reTZb&qwzh@f z-X3nQJ7?DRfm*^i-94@Wuc$4zgS;(q?yMe&y4B7)?0UYt$&csl#Hltpy+u(3WVl57 z`ru6Or=hG?j+h+x8n6{_8+)cqkBLuz#Ll?+X{wW#vulsPy5Z~HEU;`UoUwoEo0fTs zcsfUE7zwi}`y1~U+(|EfwKvN!iY~6}$n|)|meO{mLe9gqGWadG*L^uyKL-Kb!GE8_WaiA5|wuz z?^!rOFjSB9m8$1h=}G>e2-3)`CcDz4iGrV*xWt98a|%0 zqMt5@okmJuZJeg2HCxHdB07L)LIDm84fH37hxoHNkoQ!x<%r~5*d`ephusa(%j*fm zr^i{%Cojbv;EU7y_N)DFwvTWA=EnHc^7k5NkM(u3P*0oTlmu#=#XO>;qH?14E1KJF zk`rIx2MtmNoz6|;GH`+kIgBs!X+>Qm3)pW7vA22O^kke-6wa5)M|%nYPC5k^18I5H zjh`0>>W&oz@bnBffu8~JcU*5$wYiVmQ|f#&$i-vhCwhfa97+S0x7ou_j6_QB#3DFh zy$N+&dz|q=j{2#l;8MYg-j$Aa^t+>x_!MGjR`_WdY2eP`@cM?)wQ;X!)KdeW8cyi* zF#yXzE!BSeWnFVT=4z{wB@Sp4&S7_UT6<4Xt+CNmzb;pgmrgI>XN-pWGQ#Zoo`Y0z zRpAkEC6X#Um*J=T-jSUCa`&i*QD05ByuREBv>@2dkYns8+8zA*C^TT*D6*^#qJ9DuxNsqNz zOFWXq${HrvGMwzZ}%yva6Z+&TgvN-J+@d-B&6HF*+ga~N%wv?1MPrX z*P3IcOi-WSZz!YWeZ4$qmV402t*ISOC~A2A4JPP^{P7d*fOQHqX<&fS99_@#jGM$V zk9e`RcR7|z$F6>T$)p%A&n=-@dA;T}tM*%Jk$$f>fAc2fYJYQU9vYpQ@Tc}&!G zu;%6Xf>%tr11tmOb^G;ASUl6?kpLK=sLc%fp~9J}5-`%SkvlJLzNz`V%6jK&ck2M5 zG8$+$MJ-uKOn#rOuy)3<+Coe^&#UWZmhvGXXsmmj1(*YL%f5$C;syzoWG(?O@1G=J z>4AtEf1EdT%4c{|-Ofud@$tpoiHi_qh*7bhde`8c7!4VpkMbUt?oKY$^DX+&$&B-r z@rj1t#X)$FK08QlHA9^`RP+Hvpz4kg;6?TDbiXaUPiDK`(Pz%Afe;C|hRlC)-kRPI z`72|;{eZQ&6Y7r7fFX$lwvfRB2Cz@t*SUne1(%66?7@HR*fA+F_-+f%a^!WhZ`pt} zwysPLdhy)7&saxuu>NSpp?*?!c?!?#SUZAJS<0I{@2_-;OXxQzvDpyBvQapUTUczc z+iDjotu0Ljat-wVk=<0e;enupXodSo_c{+apcUZJUx%CSG&A`=!<8bgsAjT8U7WCe zK)C0s-Oc)G3-P0Oi+3qMPDFck?>GY^lndQ&ED{J;(Z=2lH4#ZPdgaZ>5b>ff$9GgQ z)AV4`+DR6iemvfq7@{vkWdFc>0(u3k=90?4dWB-pg7LZ`wzJ#tW5)mq^Ec{qRqUfk zwJSNXU3XV_0wm4@+3b_^rB8aVEs;;IU)J1Lp>+64eI{hcLtnsS!ohDmtnsv-Sv!yG zX**p4SYxGmW|iN3B+d{{E6cJ3cdA68PSXCt64oAPOSNPCIbj_ImZa96;+wEx1c5cl_4^{3;J%20DfEiTWz66NPu9dL z&C7*#d4;v~l50lbUoI9#Jl^6Y8t|7V@4KEROQ(VjbZy>AdTH)^PWL?YE2@F-t2Nxz zv=ZYuiDJ|D%w8hu;LaG!_tT@6(Rx2(aq@Wfj`f1`;+8kS)gzM=F1eNE^9Oiz5>Rt` zS;$<7AK(9Qdhw$;MYo}Ze>gIYb%#nsZwWj%X&n5|+pmsHXpWR~y=6F)nl|4E0_mxq zXjreGtbBT1d3&Na7+nuUkA5D4G0^r^>la@o*~lcSZhU zLrmFoVs+lhvIaEzw&70>J4DKB{KIa_L8SW+eNjXKxHh|aL%jX6_wWq+`|CKQWc68o zLXd-6#RGtl6+LoI0txV3M!FWM#E`uqKyJBp)O#8QzJ^ zioQe+F)WsnE9TEcA)ecBcyhbcRML(~M8I+zN{LR_0*K?pv7c$G_E=M3a*uvOmWS~F z_`;@vSB(9JIiENVuumKfh(2&$so;}l%O)poYEA5EVef)XVU1J=+*pS>^l#`ByJfdDyD*SD|)rwEv%tk5@kh9rlH!^lwkfLowFR_O;laAan4LPdK5}Bh*qs^4eC< zCO!h5jExIyphpKdpgsI>sN-!SZ>jhrbprGFNw10lj`;8>2Msy;Fi;WsW zeMbU+2Auf~aW;C{s3H1^v^lC_;~8hok(V+E?rsCIRmv9B z2JHZ;?RA*}@UX}AhS$Hmoq(sSx4U!}#fpSV0B5ha^W8qs#xtrTBE4IAQYvIR=Q)6D zgCigV%zXERHFU|iE8Sn-i|urvFt9?0Y4!1?PJ2+qk=-;GdV zPDj&{(@UH#YfB6I`&Ub4=#wfJvE4pY`2zgy3Y^TBL%<3vdW)>)cfNh~@*v9p@Fd`X zn;j6j@9Cksl--iW8MPn=cP5Y5UByCZD%G3#F0SaE$L2+CLy*0nBJQ#f>G?PFY-Yjl z8nq=mpvPXW<$u@wJski0YvA_S|G&%-4<_eYi4#vu%>u-M-3rMK@`s2VSw7rOx9;Nb z*N3sc)80uHDLN=N&*<>us?z=nDWL0=H^nywdJVvGfV3ST|HT@+EsQUs=R8V@9Fjr70lCTW)*`{>PcUJE@%?I0qze zaVP;X_bc(nC6WtRrS=17UorP95r27w)f4r?v{6TFme=zJXu_(e8^TeG72uSNUN~qV z_5^P8K3{2_zmM1R1&H0(P;LcqmzSgs{+f+J&+uO(*Q@`4Hd*U6}?f>++&Je%`;CJJGs}3hO3E52%G_G zZ-BC%3m2^bccF(eb`baVAML~p1@zXRR~}m zaN$jWZ60jNR0r%7l@KS|naI`hzxnF~IL;Jc`SdS#wDzYHEnbxwGMJT4*cUZUy}MHe zA;W%qal5)(nMsDPeBQa`GY4ExSWh-Q4ZY3ad`68kt8_I?46p`@=hAzSGBe+polU0Ee2N1qD;7t@r>Utq%L5f+KnuR6(Kd6w*wm6y54+;-_5P9`5;d+ z`gvp)1n=Pk+*&PE`HI(aC33UAGck}tsj30PhdHo~Y|X85F)AjW_1Eo8Y=lkYpPmEc zD^EvJqm15K^rfXV8a{V-z>&c(Ge#B8yR0fHlmIy;%H#n?Ve6cf2*2+I77+cSW9R&y zt)c#>H^sFr(-+8q-)&FvY^l9^?Xs_}B)ClDRh2pH+hx30s@y?vq|VwXt5CP=aTEFG zZKDwhEZUVg?YH#~AH3}A_HUly9gU3g&Z!>`Q-54P+oWSquk-D?t=6bLqIPa@a7Ixm zQB&~4T@~=mhw?>VHxl)Kn@^S!m&3~g#?X>^K0UYeCc)X1Uw;dAZ`u*(ppxL{cZSr= zRWF9#`Iq~IIX#-{g`=v370sh44P~90kcz{dt-8~Ko;-T}{)Xq!RN@Zy4ZvH9=2PSN zs-5jq?t$YFvFA~OnpShUw6`avJ#W!l$r#tFVYgmVNbW!Xu&59-2aCutmUth(d27oC zOggT~K=S*iNH~YvVT(RRk3C$=FnQV>7QcShD}RW`M+jlhOxtUW^)c@S`#<)Y3XRG%a_;*L@HJ*n4w{V` zfY_MCP0kq4k8LW>c!KD}ly7;9dzxbv#R}fvGlDdIeYA?BN7^d#K9y6NebDwL0q%H(o|XG+6tmYZhbgPolu|)@QC0CZjdLuT*?% zzTr2A@WWP@*P=_7!#Lw@bCMK3KD=TX`!}rFcT0bK&w`ECMFB;xV-*3lqY8_?P`fK> zwa5WVywH_1{IZErJ46Z*HqvkKL-W@@+nb6yd0w&k+uT$ON&LKg;1{%rVUql^JT`BE;LNQ4 z@I;=Eq|yjqQQ7_)DuZ3-(|d*ADc0Jn)c_~c!mx5*Q$KBG`(crwjvT_BA$6wbkKXMKJ-)>F@RzQhfO_y2c?aCo>fTRr4U|B_ENEfNGTL_O)_3_sS}%HKG!^|yZ@|6yHIyzk%J z$6dK^7(Puf;c1x_eAqk(fPV<_~ludjdD^iOYjM{PriysfOiW06)O14RZ&!?{t_C&QG88wcZ&Z z=4s+k_tX76IWhMpKA$|ext=5!~jDNmW6H|xt$L;1&PojY) zqtoq?z$axj*5UVClwG8O$I}5<3KTzhKYx<<7IJ;pdYyptYz@x1HOBabh2$K>F}VD6 zD$E=GBRD?xgVP?MiV+FLleP}eL8ju?J2?Y=*#q|2nVwd9zgb*ZH1oHc4V)9_Q(7=M zQb@%lWgwp3l=D}r_5|L;ao4&p8Q#lN4o`=-TD-hyRJPBd6Q3F#T*}c094=tPXisUa z_d>1iw1*fZb(3Gf^X3cmq)5%le}_9+KHzDQKBSrt0NlqlG{5Bu7SHbJW3>{VLKOPP z*Y|g3!0tO-qT(A~FM#dn_n#(w3WOeF#BB#5*r6Ix)zSVDwM6J`%Z-oVT8C*r^RiuZ zFWB(c{`?8nP4M0=r!4r?>f+uGH*;TebWW=_v*7xioF}o_Ka-(aRe{m;xqt7 z%zefD4xs=2<0@OU8{4O;v@rx>e>f~RzS(~{??x8hqkV+yJv=R!+P%(NgP0lh`c^@h zQ=q(WsAqCg*xif6u9AUu!3RI~F`o>mrD78XJ^vOt^X-3qnHs?O#EsJa`k|LyxxcB) z_3(TuPrB9ODy#wXZpJt4T4oi!X4Z|fL+M-ik8y3DHJQHVkT36}gTsg%n=#Y+z20-w z)HNm^;DS}t)9$# zQ4en{LCQs{+}Osy{6vqy-Iy|t*q&!xTuLH z&52r#=l5a%Jelzx{X`-x^qG|s8a*WoPP4xw#?(?I_A=y_$NSyz=AVWFvepB2!lwFa z`q$4-b2{IVb2kiJeZ8`5m>r%?V)w^lg#bomf=>HF^kHL~Va5y2E)>YG;?%5kA9&qi z9i9&nGENGB_et_wdGLj}z#c?!kqF=)=<{i}hr;Q~Y+45hG*l&JZ$G$DM1`I<(I*T4 zA1=BIWFA8-S_JeORp_*eI_=Rkod!AZ{eG`T+5(SgP*Yz1<_OU<{C0R=c321I0XBW8 zO2FDP)}=PW7QK}qMQ%*=tv>C6f-2;uk@i z=R*$mLMoAjyRdzu`A6`O8*5LcnwI*%8EEW#!8-3fGhxC!`OtncLwIEg?K!hz*tU_w zBmP5-_;gxEXcu}z06zb=SG+&3Xj&t18u#Q)3BQ4Cr;$br*r=(xAerdgw|?HrxXyw$ z5wNfQ(``k~CKq`6;<%J|LvO1E2li24@BYO1I+0m;3IycDr+jfyQy|n6!Kiz0T24-_ z%p04O963HQMe3~>XIBXx8ePZ34$pX#cTU0m<#!+MvOR^B{FG7GKxA-BJi1$P=r>%V zkFe=*N~Szm)MDuaVBL+m-+njC>f|{K(W(Q?DP`>uwfx7&@Y8kue&d&G`N4+y>nM(JecC>p;JsN85hZ9XqJ;X$(gl$HP z)8*DL=dJ}N*Q1xI`XUF2oD^Tx35w#oBcz8qF#%`tpVNLTBNh`AtVu^WH$QZnHnw%= zd;Oe~;4%JY{jW|VnBQh#q#QNS19>~$1`RcS=!vt(KEazCA!VSR%m)m>J`Kn9(Dak3 z879CuU!v|AMV$BiOvZYsFQJb5)MXQ}k1a5@=L0eNWF>LT)ij;V`awc2$1 z(6=KQV!mUsP9@Ideh0TpP&LNN55q*+Oze+-)-pr z+G2hBy=5h$Kbh71hZWLexE#DM)(p{8K3E{0IZ_9F@ACyBte+hHG^H#nUXD0Fq6^6) zO^ zHf^8CFFu#V+DK8CN}>|eQwOxGg5**Ir{7rUNjRi@g+wbEc+n1B?HH)65v16)Jz|eg z4^~z@tOC{?+B;*Rx0Y<4?s`aW_ANAXzf+c^J=C-J}fINaacx&G?TfN06a7s8K03-?erPy6UP1v4w7 z+!HYodM~wzvoVimO2eyfJh6?JVgq}LI;?dc>jxi2kTT>Njb}n$jeqA_2)B`m8e4Vy zFZPWaDeq0J)f}+jclvouP_=u4^{|FwsH3IDcLXNF9M?Nx;VkLLzwBhpDPeVd7Fp1L zCY>Oxo-bf|d@ndd25ij!VVfxHLjd9i-;gLGLQRct_eI48WouYd)6+{QKf zfg6c}?Hf+=Vu)Um4Xc4#(54=_mP^>3Nud{I;t-F`BnXLM1wR8g^!KDE z%i+Fk<8hJ%SIm!HCz$J>H^|(4!?V1c(_HKkg`tu@$YkRwbB%nUZw%(#7G3!{C zM#tV`v-5>)L@BZxLxjd7%nq;11>mKfLl7p#|LBSpjLC(wa_-`^GLS%C{PgDP;xQy$ zi$cQ=YY_CrMmilfh}YXR@4Y+hshb$iyruPF{V+Rk$#SN-FU@#)u&@}?S4bOk3Nc>F ztbAjA@}>#(Z+s+jx7TY`>E}7!Sr^)dG^f7KZyAraNEF`46x1>Yzq$kfrbWeheRm^2 zHmt#A4y-#VSW#SX>2l_uV#t*U)Lg6U)Z{||>Xx$aeAj&qL0{hZDdj$s zk^pwCf3tUx)C*1X-4pEckJg5t_!50k3)L;Zehc^W!MtE9jYW3R_ecNZ?+;Y?vCqj- zx&$AoeGLR@hrTCLX&uz$;IQljT$Vp(*Pl<opE3ljZ4b_Hdfs}My1Ec$@4?}21+5!17NfiK;OOZY zpkImcD5@ATd(+!Tc#JE4flzt9!uqeJ&l9NUzcqsmSG^stFX;r=*ay96(3R6w@$@sG zMc)NI#BPUF32ADmv+DHvUA@bj&khAe&4+c=!5~P1JPz0-u@9}rBGjf)&EZ6IZRzmIC^`T`oyf!cC5-$xhW(@KplRc;<{YB<@ z#?=8tUNTaV^bT{+I7$1ruJARJ<|q8CR zaG!VWxcgyL5Br=?VdPAl&8L%cRj~&0*y+GI&B@~|Y%th|Qa#&`=7m1v^f(Q|4(LR^ za5?sG?(pt`|aj}vmef+JZC(c#P+rjS>p=NQ$zgx zjDe4~J&J5PFP=`=7t~48TuF&hNec3O1=;P)MNw;?&o@>aue8<|MNbp&;&}l02|)6C zXj0EiP2ye{`2Qz zc2mjHmie969*Uo~#z1Y&OTDIlXJ~2`e!ZxSu<}81U(hRFIf!J2*!qyQ-9MTi1M8%( zkz&d~^EAVk(MRQFM`qDocencI`y{_M*zY{=oBl*hbul`>UmD(65@n~wp7i=usH*F0 z-2P@n4mcC?(BI$s=OusKVEy)bv2B_A*Mpw8|8srb{Xc&e|MO>0fIi$Aee2(Ii?!La z(RXH6{@rWDvUUW}Mx)_bCDwm`-pk-!<4~nP4=qzu|8zGeTSonn>kuf3l8s|@qH+rs z|NT1Nxk_aD_uuos|2_Z3+Wo#R0`d3Jk^<@K zZ!RXc;n6-0PwVH!|L-$6%xXcNH2>S6XO>m(u;2MM?ed@du*xH90&gZO;c>dx@tXhL zD^oo5`G4R4vH0KL^M8MRsN&yRvwpr`+k2N6|3{k@vE<*+SoZDM#7nh5GxYAaCOGzs z`l@sJzCN$a-?}N5=(kQv{`QyG_8Hc+3$IM#_kZ4>|9b}qzAk3(x1Mi%9u`O?qrdC# z=eECE_}3q*?eX91+odYrv)X?o$IG0)zr0qtxA(`3V_&g8|1eCKk1-Lv&Xu|(jzj}w z$MzI=Xs_(Vkw%|6ZGTl%l6gxkQ-2mZ|Jcz2Nb78W;WLzX#W~N`E!uK)$Wpxq{m3PU zm#{onI~x0tF;SwaMOq`E!C( zG<|LQixwC=zf!+vcmW@NrS;#uRsF~#xH$9`R!+(oqi1IP4&4ay{W~V63x%WhjI=xZ zkvUsicNW6StB`i#{xU2=ow@Vh-d*;I?HPxsatd1p&pUu{)u0D-U;5tuwdv2o!THWY z0QbXR$Uk4U9e2K`YxM@AeA@Z?$t$9*9G^ZN@WOO4oZSr&k4gfQpQDnmI(53G5~H-u zpC8?8qu8k_%UAPvo@jVLAA?(ir2`o?8CW~tpBQ%5!zX!N_uE3|PxI$(jEfoHTh%s? zwCaSDOkMH)XR=V3r<%t)YUklt^!`%5$hZF1Y)BlI&g>u8i@>tYaNlpX+(2CPUDr!N z!#vXaLFTo^kT^+j4YhFPcxb<6@j?w{=$!XXX;z5$*tl+&yNv#*E%sa6>)>9XS} z^9JzoEboMLq;0dHxa(nz_ZL{p!$4J!uf?BA%`QGmws3!EdRPj3Z@(n!#NOz;Tr0#z zfdw@#=QX3%oJBQLSAkJ0SHmUwSm*l+=$}Tm28v9lL9#E^d=@|h>THspH z#~Awq8%t3Xrns?;KDeah7)??;d&*GiFncIRcI{+6yH(X+Y+f&7zQ1E#aODBl%@6jj z?dU#0FXwM+a)0MW8j>XH)P0{DLdv9OQC?c-KFV8(W7pNMmwT=}?cwZ^kGTyF1mQeS z!nrQ;&o9TF^rb#pBv-5vjaoUn5`18QLohl^y1ZGsykB|edMnQc@8rB+YS{+z;l-qM zHW+pfeK~pG-#QC$o%M?}c@>kC_&HP1e}aTwhv4y4nd;D(kIRW!qAO`7`{HXtehh|7 z25%SE-Z}F4KTHrj7i+&VqDKBmUM#*-m{<-)r@x zmwB)kR;oU={5l91g%nTDS<1|RXF)A3yaU!jZ_q6%nGSj{d>{~GKv{y*c++{NH4~DpL!Mimpz=aPy+oYyz0cYNJSJqrx_}NP(#B0#rsa?%y z!qbdu=b!bO;d;emCeSlFGJdn!6!vWHFSiL@Uf4)8F#T-N*O^JBvs=v1{quErzZc#6 z<%5#soVG$uA?JqPGqYMWdHK= zWivo>h1_{cUAzzUSk!6fd^oA^Q1AwW-ejBw|L`e>EAG^OzhnRD#vQM8#Qh^hQLF~L zh+6*Q)A*d3eMn|j)Zkp!gC&eTYPe+7UdJ9Vgo}u}kMYJo#-jz$u8>X zd`pVEYo&)tGoK!vyH-{+dw6D0<~a2j2mMdio|=yka9c|JZq56l=@dkE?PcI zhO|QOKX^qztBvP%tZyy$%nPCx0m~TQArFq&Bt)aw51!Hc$qBDrs-Mr&f72fzoh))fAVbOWKmA~17eq)WkGj9fEUiVUHhK=TsNbxdTYKm#=E)NPylJiG#NS*? z6Rey#IRnk*LBJ5{J+n@6c8^Ua{9wPYsdB3>Va=RR=&SHb=#AA28v;Hhe_JCv^113M zXZiEwU;6Ue(+N{w)$<>wf9FL8dr{fWI?gNZWWaC6NZ>3jj;pMfL?AYU(-c<~eLUs+ z4*OoedipK=_*B{)wpaIdoO1<-Uxg;+!IQ^%oC>#J3=YvUX1z&x!A3#%;+ELk5w5MC)y9v3T7QMP0 zc37>yI0`GSb@qG=vD;euRZ#el%&X;eY4w7v$x3m$X;Syp?~Jh7v@Qehv>U5lf3v&d zl3TNs_|hcqM)OwPn-S1cL@W@a51qHuie+wElQ0`+%( zp1fh~L+#+KdgTDN24uK56+9bA{>@RdpWs3#!pVYXYwMNT#5kMyx!CkPTy6!)M~0(5 zzOoB1@Td1^@jO;0kytX57@YB)rTw*%BiLze>5~viEPgRQg1(|o?YJxLEqvT-S(EC_ z-;Os3$>>XcC$lUI##grgvYMk(hEbf_X0M&*KY0AH#X+Y#LxS9`JnGkoyRfy zas6f>IFe7VyT*mo{rrkKnj-@Ybq-H$hxdVV#z21V`E_L z-nU_$%)nV%5{pv!`M=r1D9!Vt{NDF6JuIn?wVH}89CcM$@wES)Gf#C=zVhnvg=fAt zL1Y9F8(TUy{J}^uumU#C*d4&aDWuKr64Jm;e=gJ5RVEc~5UJ{$@XCD^m*V1wvG* z&dfjFOOdh0Ws8gC((I=Jc8AOhA<&?gEVj$fJLUOJA&eGiFmy$oGOoMX%JZBP z1KkmrVLQ>x$hDt(yKHrAvF}yIrQdZu3~9qtcxD$idZ}tLH3(K5D9Ents)i9 z3Ul|R)n=y{-txlIhk0=B#wu+>vI-&W1716e=JgQ<+4-}19nm8_;*MC;!1FZ<)1zy6 zWat}!y`Z)@?YHGKT-0<)`htnmvv2|&e4Rtd$@8o?vZ6SUPifys+bVe6K8^7j9^P^v zum4~|6EPv?&pW*1GrB(s2Bt>?R0FFA9Sz0 z|9HT=zTxSk1+zc2agio%IB+)ZW&X^}Vv~WHWp3PW)Okzq(N+TnFh91)fgU#{HLk4wnDo~k1{-P(Q%C1eJANwv7M=U^L9jMSMBV)LoRjOYq5e)xU*A_+ zRDO1g+n>7{XEXX>wBXgMouU;@(Q#6gfμFct5`Os&u@9A*G+5NOF*MDbPSba|*8BLsUoo>y_Up z;l?lzzk&4dc#Ip;VuE5HE#P=Nt>_4ze#>d~yy#T)m>;i1kMm*NyKl7oU!SXcdP2c8 z3{erTy*oXg?tgll{-58A(VwB@u0EwniUN3CEsCV!A_JesAJ{Reh^(h!LSxBz&3LUx zCvWY4J^Dqs(P8t)1)ey4z4pQw&YPo{cWfNuvkNWKy;P{>rLXU=zeLjl%eZ=7=uee> zX;J>;@3Fvm+OsbxP#*c~%}u^6|4YWl!??43mg%|ZG^JTg#-H3zU2oqx{4aL2Gx%Mt z_IS2GyWsx(`&Y{|bP;fjynt}AYw7vU4dBgphnE(-oPRq(>0 z$;CIB;}bLtM7Kh-9`PFHtp9f|A4?JEnUg!79kSGQ-p^91e!ohoNS$Pa(;Aa<&6cbC z9>;Ul^4F(%HF*E6^SSesmo_4azxWLF9m@OMisv2OU|xNW-;Mi6a3InFTzCP8&Z2v~ zz3%#}%$Wo|b0eTwvD?S&f-w$Q$#|=vPf6#J2HzEq-g3eHsB1~|+>*Dy-cx)@Pwty} zf!)NCHAD*`$(Fpa#YscXPQES!@unDi!dWOqxe}gXcb$7mif7uqqu0lq@ z9+5wzXtL^2QZqXRts_2qdarf{&&uaKDbDirOF|F?z@&+LrLNIm|JA!?%7mL%s9%V` zAH$BSUpq4#scz^l5XfWUV9U$>z9nT2UO`b+zsz~Zj4llgpf|F!Q~fh9(m64?D%JjY zH=t@Q4Ihql!f`>skU`io;fP62HukwoGaM8NGF&(iHv@`5oDloRG6&o?PF8h_Q@SWF zM9*@TlEm3^xH&h3Yd4X=J0zhIqK(AX1|r#I#LI=e9ok_Mi}Bcr?|3VmnN5KRuPq50 zY;Y*g)ck>-5mstNIoZuwq}ndb&6m)Zfni{i0@j&7-zR_Y&NJWsH)A|kH|N<*BC!5e z!vTM590cJ{u!RFu#qnn50**J%1H9BX2DLgt!bw+c-m7*$zqj$GD1ZZAi^@b0v&(B6 z0DeJlpt*jmTE7u?$q{`77p^jsX{%1;!*4XTJ&geepwR*O*YSVYk%fHv+?TK%{z&qV|7#^LHc z819Za#m)GlgaQo3*yUXn_HsP6AiYRn<6;P^9g42r8=z3{T~4m<`XwD7d?psg0tL*Xz*-{>=?tEkX0;dYsuh(^e9W&E@^RJM52Vs8`uY&IfA+AyO0sP5gZk?FxzetMH79eTmpRiIY^+tsC-i1$gl;Pp-Ms z4fJ^t(GdxhVuFh{uXU(5A4%aL}haR!*^YO*eOycxGQ6BzgB z@j-ybFC;;@>er|j-~+gy`wcEz=Fl=xT#rHN?sUlli2fFmT&8PMHE56cwE$cL(}Y2j zW;^?vCreVA66=OS0QR@H?9Q=JXU6TM@AK>RHEJR!GX18wi?76kKSk5@3V-Cn4cY>v zP-lo3dr+&C6%U{Rkqqgm^4@Y=8&ZTKl z9gM@rN|t}zM;l|(RO9SVUlAfce}hu8g7?@O^A}V}+8w9o z+V77ZWA%*!50JES5VnuqY;srRnrv`svjyyWkcD7HQ_*Agp(QCgieA&##m-0#vJ)0k z*>wEl8Ez)Jf`w|&T-visoqhlLiAPA(x$WBlfIKa%AC{isylq1n+=cX%Ac$c04O)ox z7$`Sth&Xk=cA&cm9?~BvJZZ-!0bl2)zW4=Hi>;u;>P6JuHC`bTAcEZOZ84Q_nq+Jb zk3PO-mY{BK-dS@1D17g_JAD)qq?#c=s^G3@KdJXDmol8tKaGA5gfq?^@1O!!xuJed z%=`cf+ZdpcK-$Udm*sRFzdjDq8NLMLsyFONh0v#ePG0%$27y5D7ya`t8YfUlHcB5M z{j=!jnF%olb8yda3l9Yi;|WTF1k~-Uhki{Zo|vvEB53GYgNHVZ4ufl+Q_zgy@m*CV zx0GuJB9(kM016IxyUimv;^L8t9tB)!|0VK*W0x0{Y(k2NsIC4m?_{_OX)`z##fWed zBZZRe#yLfe<^1H1AK~$x{?od71Sl8Mw?94SORYpM`p}sRJJT3wsYkFbfqE>=!n+gI zSFV*e9?2(mBK2)4V?7YV%ESaQfn2(lG`<#6t2H z-l$+z=>hBeuwS<@HVs#tU!>->q)reW3EwnAaU~$Xae*Pb9Vii>`+(=L*HI2+8RJYw z`cmbbEEio`T5P4}9!5te``xu|1MdQL!*RS|%l*%{Ggk$+u^$K5heLNqdQ{7zrud*{ z(e!m1z)=?B2CY8&=rh!%5I`F&2L+RR^)hsLwr9+Qxt z!uGM61W4JNbIwqm#tL7nL#D|Nm$yqxo21C{m1$Rf>^H|fff^qmm(qU zBzHKhvy=8=!X>o>9TqKLBWm_%H~4mw8VYZluaP6VbRwVlZ7w}{onS`-&wc+jZQ$lE zYS^59l4}lzw`9g(uRAO1cuF}1l0;6`b4XHM(ffN+)v@07;dj2VDR$&%8T;REy=OnL z#-_RikT)lozM-^>Y{yu!cKQv#-&rkCUdNk|96J#E6A=;`=mp!Whjrm$Ob@muV`tL8 znq$(w5XfEBD#2ujgAf?e?dlfa>a&RUq6-;`M!-SbwUoNz;DZqMK*NP!(;2c5)7lJG z^6)LQ^lq+7`A+H|pLEwH#KDbL@r}(zqu|ZWu=QRd;-vXE!I*IA$Pjsk^)iN550d7yh+_sS-Sno@rs1m4{ibNsRM8|9BUd z4jR4RCB(Hf_4@#3T)N_w7!*FEpyHfl7iW@R(Tz9YSuRh&LOIC|+uc%jP6-Pfrbxgp zeOWV~GDiE}@D*&1RUbOJv}e0JBn1UM8t{9=A*^=L1E)2c^3b$DjozV!XLa2*3|-?y zy16B?<9tLqH>Zgc+Q`D>B4k5Nak#u#T1JQOV$NZp1_hsib*?SF=sA%@vqx?kN z-oA!Z_NY1!70KF_k@Lg@Ad_?F?}N(%Cj2>?po(<#Lm*xO(?7yzn|;ii#-2L&HoSr!PaX8ef$ihMO`09a&2fCP0hO;-@Y1j!y@$H@ z#k(X!7u4(R~bGO{K3L-KTK$KfzgmG zv@_A>nL{wTmi-bbbmk)yS$SEtpx+~6s+aWn(?hDOj48Pzt+P6R2oO4B@|j`N!cG2Y zDBFoc>{ZiAzfcX0;iU8l3?Un*N5&b_03Z|AMOgA1tJU2>USHrP2A>l<^A-G|uhLqU#yzIDBQO0I|~ zzdC(YSpdTM95(w72?o3}@L41-X2eq1)1A2QG)#{L#-lTuuY#ic-ihY)omVLop+SGE z+*LdA`%`l-sM$7;QtS5`Z;viFYpf~zI`&L8vp?z}H(kOyEP`qMmgBN5_krfUO1 zBJ#Tyh8=RuePLx{QRKAz(*60!CwD+C<9}s@n_@Su5Dx`D6jvN6 zsF&vORbLnwYv$Q`*YT0W)OP-S^L%+^B%)NCFFNpBlEA-+ytcmQr{h#}AcqNuev@gK z5AE3AMRM+Z>aBgq*o_|-D}_GOhM5d9$M z=ELRFKVI+ZEZ#KynV(v$^6`Q|HnPEaoB3ku@c7m~z6qILo8ReZCd*_FU z##7t%_ky;w+jDD%>W{uXFHX&G#^>Jp>03p2ektmmMjR~klBe}_o4dG^oWg>D}&bbnYm=19+2dwri7n7bI#0MW*aJv=?%&!L*s>ZCPKVC^9CgyGta z4P>tKk`Fl%K8RI0QoMZ*e0!03QI(kPy{INSqf;@!9K);#YZBhFRErPg?v6>~Sf4#2 ztsy*p9HaF$IL&)O+AErH9kaeJs`iHXk&D~Uz8mZDnEUh3eS8T1?px7hML+GFFy~E8 z;p*{?()ivk-#MVG{{m9#RDSRt8`$Ux1u`nb^}$a`{&*W5PI^Jy)6Fvga#VK2I$33J{{Uszb@)4eWeS<&QKe!5?r6}0S-V6kpgt~;g<6>ad zf@;n2y$64HNnjD6`+0w(F^Mp~x$-<**I^_WRR%0*try89&uq@E7cjU}jY7F`WK!2$ zeh<%v$s60THR+id$_=f!_M~YhWf_Mnjj^5EB_AZ>TL2m?H+Lh$K)&7t)3jbeaHZ7Fc ziljMH%|J7;LJHF?C$7uRq-HUCCMs>2`MDFO_uno9OT%!1acitc>)neGc><1k{uf7L zOF8&BPcbwE$q&-}$qiLo{CM`_4}MXASoCBgVV`-3VwZ~E#_`Rb0~uoXNo&5BLU>Q8mq3m>4_FD(iIXR)*bzCP^XdK%ww5z(`7aQIYpge?Ptrv5%Zj%s$OEl||3}$G+753YXXEpx^7Y zr;%;rqr8@1aBXI3!BX?Y-H&MWNFd_^p5G`h4D@fi3=3J=v{7EeYpf3Cj_}5ea8i?? z!J^p&N}xsF{`7PDwgk|2gI(cs*R@CQSejV8=Y`gZm0Ih;g{dE%XL+GMelSF4Kktz) zqJ7^BW~lMA{dksy3qlpooZ;~JpiLJ9HS_E5ZuZ#^w=j!;r8jZ9#78FqJYJ`JqqB3u zA48eaI+=Bs)?f7-qj=_icA$>mS^L@X{4wt`wm8-OOMahc4+hV!sM6feJ60d>l2qS( zC72LWYBrMEGKE1QmOuP`s2CsyfLgoYj5dsBDK_ITd;S zTb>oM*21^WpQSZ^*At}nr+4#ubQ?c3nBZqFUi|!NyC2}Mwxve0v>(OKm1w@cfyXpnYLNJlEvdPw>M<8W ziMx|`k-IM5pDtIPU3o^L>1YCj1ihl<&PK+I))$Z4(@cIVJMp@>(wvFbm*u-$ncBov zr&;J2$=ak_0M>ECU3}gY_E_CBhHki;DC5pVOu4!RT&>hZic@@eR1dC-bKOO6qiwxLzw>8S~VfB3vC}{I%3P z%AxPB%5gk^#DFga@xtvyhiB}=y-<-vU-HE8aM`ljW=%(22v9)N#rX3T9F}7-MBe>b%A1w&1 zmg4Gr$=j*>E_Qpzf^yC8UP3viEY2VR<_izw_$x8!J zw6X9zOCGpZyKBo{ib@ej7?|PsdvI} z!SK?9&w1|TW?#ufFGy<5F9jWe>W(23CODct@FhJbT0VY?1{}0u*~o?_bwi;AH4Pn6 zT)HEca|bR+;=UGUimKYp;E6@(#G47lep4aJyHwSE^bflCZGBjqYB*!?vSy)??A&5d z!{6))tw3i^NBpJ%zzMdHvZT?MYlUbN5MnA?7W*-#^bEd6N_Rg_RWG+!`AQN3T((Q! zh@#6_1tasRK&T7o!MR1EKy%D=CW0;9;_8=QF%y5`+Kj)vw!N3qeMP=CUJV@3MU>S@ zSct*W*JEUT&ztY*{)@r0wD&ndovoG^e&tPJA<5~9aE_DwlUJbo*Rjet(mHNpgYm*< z<~9F236eg<=Q?YN|{;Y-VPoJ@(y?bTG zCkW|U969;U*V@Al^OM|Oz)PRqCQg3}7cMR?rQL=;mu>uA4<{YzwaE{TQ}uFts+kV6 z%4KVDz)o8c^A*Ypj;zpx3D;>UO^dg@=Xl%insnYIrC&N_&Hi}4iw`;$f^~7d5fc8o z)%{ftm<^}x&DJ!UE?46*ZT7`=!b%ik?>y96>94n$OTfc<-0~GurUV0vhx=$PZAWjg ze5-6!FYv7*0e8;UVE;aI*dg%~`{Azli`QV(PciY|H4x;-r3`^|al<{y{Ci@k!mf19 z8$pcD$3QN6icjR77iiy@3$RDRP@P!Q2b`bgK7L?|;K}JAJFw&v z#0a$z#zorKk%njQq@o8 zhe*~{0q2zjj79UQYsp$Su>__;%^EP{No>>iQMX5mk$RYtmL6HB?D~B=pf^AzofOs? zv4ZM4Fb@1GVd^lC%K1RDMt-;1lUh%itA*sQ+)f@G5b|32C4Elw2tAmtFQAG}T_2k) z`_M0GUp>VL%a-pnS~uz8UCJlg7oW{1Py~<@Uoo1!b7eg$_gC`~Z3#m-ACRKJ1Nbi< zC{XCTz-zck*DqPd?sOR6bgPD#+&mk1G%n_i^pAE$L+XNixtUV)f6*111#5hiLdr%F#g*d%sKH!!cKLO%iwr4G6}EP z+27%Ib>gOOp%GSccEJr)QQHS5DJ32Bs50=;u8t0AtqR3xwzFpaq zxnm(TDq&HP36JuJ_8P6E3)s_lT4)OmR)Pv_^X96+?k1c!{`he{`l-KFr-boam|Ud-w^oo`~00>Eu=@J1l}I5uWC1;vW^%pj5TF~{{-KTLr4`zhG;zT zcWjbMhT&`4&(}dvIs8ji$ULT@>}^U=_8$%>YgZgBMt9_|%hNF~hHyK$*Lz=>gEz`3 zg!pht+qLlV3TjHsrPD=feo66q%!ryR(?8`Wp}bORh;~|`#NxkJzz|8_4J4K2;^uE7 zjDG@yJ3xHp`sj}8^v#jUc!hrc-Qu>JluK1FpeUdX%(SkPoUm?*t0)r_2rnb}m)Eaq zxpD?<3|p#1w4{-IrGmRUeaD%*@c;t7bt%mIVK5m;x&PLb=+~=9&VaesD5MImIlp@J_ z15ECd1CnnJ| z9bo}|TYCCFdrfLKhwLB{3iB8xIQxLZ;d%h+A&WYajs()8{25J)O-3c2Gtt8VT6+~e zR7W$FgjWYN^P3yJdTcHp?MjV~5mx_+&yF113O;E`N#FFz*Gk~?EUaxMk{D?sR|-5K zFI!O9(OP>(nf-DM^rnmeC2^KVo4s*8;CK_U_zv?m`@i$z4M()+qX;J%nOpP>L^sIg zA2W`A-I9YME{TT0q6Kc)@zfhi=$DIVlZnAOGF&SN=tY$`cRYW&LwR%US=meL#Pgp? z9Qm{pI9I6lm)&Lp05xZqo_jpStL4!RpWFyu8(@%R3RIpBH-teKTWy+np}W_71=v<=daF}=we^> zb0PVDjqxjKpT5eYH-JkrtDxjUz~=FE;C|r-Yoni?n*k{UyEt7wc!vTdk?Zg5UpzJp z%&yXV|K`NBE;<=^G~wlhMQJ}( zn>(>T06NJOQ8_dWFy=UcysT)ap&xwzV&(m4K2V_bpA{_k=k_2PSKMQqCfS!*@u60B zD9Mhf&O{pH#vZs+&p1r1;qY)Z7$lij@rJ3`kMql8Cd@uT^*Hl38N0lNhpkoJ?)NRw z8*{QUKm{dY0dNOPxK(PDh!Z(&fLwy`0HkETzl=eQty|RrTdx)Fxp^0b+XpGM#}X7B zaRa3rR4!y>xYG%12nP1Ep}ln6Q% z&@?mzd>R1CbHPsaIS1|Sp>7zGOh-j)#mLL~;a^%A5ezu^B(!)m@w@r^`1#Sf9#6=b zxGRi@TO1(oE%HHxA7K4Ja@eR;M2g}Y2hPzVd`t*m?pm&qx{HB^1SPUlgrCG`1@A+S zI&SqrI&IXZDC9A>^&Yh%4@Y!*%7JM|h+@Rr(hCy#{@JU*YwcJ11asAy1UPod@aSHT zn`(G|E(HXBrU_G4VcRWCW4#SqU6hbde5X;=vR96yX$vYCxf^qnsNjf-{GMJtf`a%> zW>E4ohNbVAReq?Q*gidTzY*R3m<)7Y$XCFe>8jA<3OYwZE5p`R&kP~0qR!Kg*$pBD z<5LP$BTkVS7lJkZqbm$j9(B)4Fz=V&8UeUN2npSRQ_$D=rGJmd_8WIP2=$hV0_LbJb?gT2 zqw5xALF#vC{VF6_bnAx8O{YnO84(*{sd&TRq@D(C(9yo&*WV4V93RS!)Zrq0ajGv1P+LxYtBoZ;ofCW-%u`*64=?uYingatm>(3}+gWFEX+kCfU zkf=y}vR-`IwDE?hRC%L!PzOKAU)Zu^XEQkOV>i3Ay)iVo0@B&gOpuZpijTlsee?^v zPUSAdw5m~8Q8~MbItjzUoeJo8UCS$YmwVI(Y@~AMmL%q{A8$nUW)E;xBp)ihD^)N4 zM1-wAow{eF-LK$SKpDP(l4>}W7x}p#B~$C!#?uzPxehdPap%f(0jX?{gF~D@{Wwo9 zq(e2{Us}9rk6aFOp)#9jR+y&0YkojEq(a72Y=ihE?(*7y9dzzu zQXu_awpFB^*V-C5Li))|5~AoqE~wFREQ!UCO#6BPfnY@d*L20kpd3jaU^G?D4&Bh- zuqYX4fv6)Z`S7`rch%tt5toM#dHtByddg9`NGgT5W1hJuLX%IAqRud|$@F8W+nqMM z06muxV8BnHzv)N%n^`ewOowb-jkN9l(gvgEonAES3h=%kgFkPmoqh34_nn!~zI5|V zIG$#Rh@T(MyWd#MVoC9et1M_n@_T3Lc#5#k<`Tu^c_myXKB(1S>$+y6_%0+{@HhDu zj^bh3;IBXLTyW4}|NA8GQUb)|d?k-O)?J<8#vWfk7u?@OtHmsB2d;|%%|HFDXiAFT z>wY_zot-i7xOSP-!fB=V!-?$qO$Q2uO{u@(e5u|nZO<|}q5RP5G|T`j`i{i(r{{6M zT^RMd%a?Jr0!Ig485e-{SA%*R)S_*2v zzG#dJ?X^@%i!LMHKx@I;9~#+8B}TE@NVmzK?gI=CYJt09D{_;arN6LE8wdu!9ffb$ zqAoouPNWraq&VHNWynWpD^EY`l}1vzY4J5`vzQz3P;Pb}^zZYu8!2JFbSSA=)Bp%5 zMlwarqWVQEiHbql&PalCii@KTlo@l_>6ar0jlv-;gTTAq_BcQD|K_L~GSy{n$w&#o zrof&q{(Jmh=|qRo#y9(TmqjS`K?Vl{us%_9PreJGgH&eZ{eF{LL20f++Nd_>mM72q z!kvF%YFXV2YmGWhGN9ypiM3s70rCt0A&+Em* z`3CteslGeSfZx;t5-eD5L|QwW0>}1*)EH@Bs|^hXmKunl@D=@frwK>NYZ>uo`@2)K zWdLQdUqTLOX$S0bCwlRz1eB=M%znWd^Ha+$#U66s)wOSE-}U#nB(b>8uo~(ccF1 z#&e(Xkb#qS@l|4n~)1u!;I@R$dMkv&e|odrQ1 zey4m;=_?0__?w+RSet4LS~zgro!&#!Yb8R75bcD8+xthFwpMJGPrreO-}E~a+ZR~Z zz#CmF8u`*f!GdH2qq(GcJXUOz96c^H3=*uHI+(QgGT)qQaVBT;;#gtE>~q*l6pQ)R`iM{G9K;#nXlE{HE45u{m!qe4v&-r2 zvD^>#hCC_Uo7F6c6>qlx>)KDFWaiMPp(E8DhPOVjBiURmQ4cajEh_4Xc(ssT7R<;) zGx2RUo5c?YV8gMfJIC;eVAUSuMq}y9z5Nz{gUh|M0Q|MGC20x6Tf7P0rT#++Z)7Ps z{r7C`SziMmHI8#7Ub2qvj<4Q9Vc$l+uyO+df)(asLL&YKS%k)a`@JzxEeUQIN`N}{ zO@sMdR*u4v>ucg~K-lmP0H_iDCic5r9)@x#RIiTPq=>VU^BV}SKA%b(p>>{(FlI+a z(}$%pa90Dp^gmiTeNfXmwX^IMSPBQvJ=9CkIR&nqnxBOOH zt_?Q&Br$7xpRcwYlS>l==mQxUfNwf`)#ENDKPBC9^_PqEt5)7Qd+0wpvA52-7yfcYxdp6!!(}jZsEei0 z#?{1<8;mnijsxq~FT%OHW_#CCx6KA--pG za_EP`or=WJ%(+M-7oxg^Bfc6V$-)-R{fqHnT zd!v?TK*0Job=xk=8y)>L?LGQ(naIjkXQ+yR`QhQXO^2q=MHKO+iq|*1LH8;+qs3Vr z$7GDh+Qg`_RQuxAd1F6FO5hGlxWQ=jBZU`KZWhj6q?0oJdxrT9PffAvMw$n`9yr;o zrHb`^ZQlv3ni+4Y?|iA2gx@st^Tk>&T;d^)jM^a@d@Rl*{cM2XfCFqDqc1r{D@ky6 zv6=+{B8VOF;KxaOQ+V5N7<#FeO(53{9K6X(y3iVHZ(Z22)w)Ap5;1P;X zq#D;^U2#eydXqMsz_t%j9mqK(41{fwDid~BI#}1MB2T~!SUs=GunAa-!9fTWQ4pH$ z;%b&(@xoKiPa4$-z6;gmt&qG6oGg>=kUGBSn6^Qp3lrRV3Z)P$`z3 zw9|f`Hz_gMX2(2q)i2T~SLf`X%W(&!j7kdQ-%zjUSl? zqfEyy0EN-xZfoEW5Qfs;!4*GyqNY7*`nbDLN<5Dcv|hb!BGX^S0aG@u!-rp;N!=!B z30yJ)0-oYh@x{JxR2J$=@xfcPcw=!Je%BzSoCvAdpKnI4mYXP}}fgUkpA|<+{0_FV!Z~R+O62Mk&W{$9pf}FX*W4jDC z7wm{<&o=_EjKPf|A$8h=S7F|!ACJUAc>xsvW`A7J7W3$tBl6AR0Y%L|;Lqjf^XM0j zpe~gJ-~zrXYIS&fALJwnr}S70xaoFFJ~MhbfF}jrFP238;QQ-rt7~Fq3fAPMWi$5L z_Qb>>&6P`!s3VMVDWMgbob%21&+`*RxEB4Ibqz98{>4Ab56t5R^TV`Mr(HpolLW8R zS`bIfFP`Guw=^l+5z^xpeVhNs(RpjP3Pe%#gDA+U6gi0`a!1ZN^YstELXDSg!OWd| z_C{jJsrsBuwofkf+tGiQ|Go_BD_?fC^*A<-4{G#1Qs#khM@ttp5A(U1}nvWJwMf}M%EwO7_m?{H! zg_})t^;<)d`JH|}x3x|UnGI3Gu4419r~I;H@EBV*f8d_eIg1Hl-4Sj1K{V92U~i`F zFFWnLH=`X%*Wh9=M>jB${^_f(H)Skv;@D9$FR$a5HYrkuk^~)SgG z=t(0M%;!9?$@mT#dc+{yslROtXaQyrj70>cQ~}WemGjDVylif_P>Yll>+4Xo@7bPg z|KVCbBM?@w6UKLbHY?r}**#SsU-03dBjyT^ zdW9TU_}?u)$(P|)38Mp2wl%K7Ss43KRCN;DD!P*uAHKg37dqBK-OU|7qHoMPSi#z; z@Z_{e8pPvk^Dz%2%q`SVJXiJ}EK2uf1Q`O@Gb8?`>7F4|~!pL!OH+Q5($_!}6Clv>WNFLQPrfF|o~H9sGgMJcx;c_Z)ZCb)4QpEsXl!5LzQpVyq2(nVt~3 z(BJ)@B#e-!SoUAYy{yfgD$N!YY#e({2cxBN=asCZ@6cXU-ZcEhM@WHE$Z>hzxea>n z>M$b&c>c@%_RT}@*azatki6P0(Nz!E6q4baOdgpgP)X}i=40OLv3sZ^LT3MBc1A^` zFEetgvU5Siw*e_#1Q|=3XR&W>YIvl>as=uR{CUQyzTp^Z{72QGJocAf0Z4FbNKqCI z!`sU5G`$VuH#heC&*Pume>P3?j3JDTI~E*E5YmOz1- z^h*I7uU|Js5&;%PK%O2U6&ia!Lvc9UfS^VBau4?_4m>*VtGlPMtCZwz+*2@Tyv}%i z>9rVw$Dt!c*ZfnBE}Qci+27-0Rf`nIrH!xrNPmu5MD+#4oS;eF z!lD=mKKTU9U|)h(O1b-Pe|N+od;iuEv_4y%{+SEK-m+x23(H(|eXlbCR>pBEKl_Zm zAh^83{lYzUB%2IMjT$`%JfbpKnlhTaAiq8Uy6Wp!e>T-^?CeW70isPySJuBCfQxZM zkp<`ale4kehJ^3DeLpzpq3cbJ7q9Z<=N-(3z5iHX?_V{8Yk0K>p#$O9;gg5Y9dt9t zMXdp&_?d{dG}M-)ROU2q@r$5%)3?Semi2)zoYf6CGx^F+7fnxK=$-!yh2j~8O=FH= zlbnWCcj3@aV*s@BClnsBo5F#WG$W#AS2Hq@{5T2-UT|XIy2u|o1CKnd_F$ohmgcJ853~-^g>u>Dj%|mQl!G0AIy3-KVCP7vd-y9T( z+zaLmp1*TO-FUe~v)%P4THn*4nr!oz5Snx)C*ikm1Nq)~2z-0;fL7M>mO!ay&CE;? z$sU1yn4@bm9E4FE)xp3rWJizE@N@G{rx z{3zQg#Gs&hm93>_T$-Z@y^)$R#rmP2M1DOV0$S2`=S%mIUPRjf8%2vrTT>|wr9+MA zp7j{y{U8r#UdTkYEW=<1!oQnu<%(?epU8 z@)KI}oTJB0@6@pN6!`7(g@{b<`+5BaXAZ!g#7v=nYe7DYKg!@G=8L3dnq)($ zItn(dCdNXHwe3c!w0c#`Jy;U;Dt>~*n`7;!TvNMmy(hZVQCesqXyJQzHP*ZX&NxgP zg`83u5|EP|2r6RG!|+AK_A~55(j-Q@2h!4cnf#`#Rbu7+j;RA1`)&R@hko3KshPg}rMtSG6{hRA(^RooA0q3xemf3Br(lJ=B85>+qhbp;er- z%CIfy*TnZ56MXHDJJymv5i5{Gy!Gj{lXmPe9(TUhn6LZIYO)M?tn;ZX2n(S^%rx7i zRgu0cg7oN$M%8D6xH{c+`H()W4Sf<`=Kg%)gnItdA1ktEZD^BGbVYLj_b6zAHp7z4JD{wTEL*AL(>ya zmDD?rWK94L1zI3Lk_2~cf%o0ww5Rrtje$~>AuI!C{(^Ng2wY>?5E7$Vj4g!YXor32 zptd+l+xL%vFW$EVss&D#nWDxfO2aSP0)@Ssm6dr#(IyS<1{IN+EvOmXVWQ;UCQ&0n z<#mU`;$ANmpu*Pog7eo$l@8Bw}o#YwWPO8-mp|W!Xs;|RClbWeiI84Lg)QU zM((nkAoUqlIuill3aN%F92f7O+*t!_>fd9uI^M1~QNGsiTTMl=5;lSINE@r;7h$G} znp>?-(+d2_h54vqgpywcLXWb~D*ZkA0bl{qoL;E_hPx46ehUoP<8G4BptN=euvm!ehalY2n;kw(kVFRg6|n03^=rVBuOO#BuY$s=-8JHZOLug#5liKN+)Tt zrtI>Prp&C6`77mCYPN?H;eb_`<^4%t$M*&UN1(>`m>%o;@)lZFX|N8IZvpwv>V)ls zTyN5bB%sJ#>fIM+;;aP8IKo#=@wcZa%ToSev6+F-x{GH-Mm!gaG>doD;m`267H?yc z&DP`h;K!7qHIQXq&OYd*BRZH075%oq=SaT?=PvAIrt;9B%Q8lhVr$BQ0Y9<;RIq;( z6BXI(LxJ)Ej;_0Q>DkVMMIKOWo?^qZU4&##o;h(6U=2gO{7{BzOzRg@CO;riJ1=Tr z5#JdJz4jCV9Cn7q6&Qx~*ZFYzGpFhM?1?^o-&|WyPks165LsXxL31 zKProLBzD4(OoO&f-pwf zbL4lq>5eyu70(-42^sC)_|tcLJQ1myAmP#LQ1gwoMyY- z(XRE_d}ydx@DVIfn|Y5d8`u-@E?Z58mq=9xF>Q)!cKyssyN>NSqM?^9rOzus6m0)lk!6+UPlKHt^G) z*=V*aRV(c9XFIQ=2Y~w-)R*wjnhkByi_C?WJ)Hj#zv-;5ve5Z$^-3m}2_(Kg- z+XToy?91ZQZMu?hea0wvxsQ|+GPey$P7oX46Is05=n)l(fMiY~v3St-E-yG&e4qB7hrniOfmD!0SJLOFF@zU%I z12xBS2hQN+>#DVUV3s1-=E4~qM6Rm`gL?zb6j9v?lWbRU;rUvyU!}JKKDvCXRXQ^r z!pfd>HjzeP%?I4#W-tZ&6r4$P4&u755?rJpRtQeqLO>*W`JT_k?G774zdaA#$t47bhq}c zBe9ghYlx4I>3^SW3)1T|j%$?VqXL!7Zxzf0%Gh_HJ(G0cThZGDoaIvMKm|F2)h{$ak9Yyy##|v5kQ6c-a!jtUaU$`F!oCpZB6!XMp?=z%EJ*V!hOb<6 ztv{sqXCA_D071QBzEJRKvafG~V7yr~anv}(KKu|z9%>5DmKKaN8r=J$oe{Ms=>r&H z{?tyN{wT7t26>N@UcebYzB$)TeZRB`-RHd|tC6rGr{pp8XcgdIT$3UQf$+V>P zR=b)u!au*XMAAkp^55b(^JlozFQ?r}vOV-J2t4AI{%^p!8Yd^=4gX0BS?H=J7B9TO+^<%SRE+;RBJ z@NEWRcFvlSQ(CMeKuNq|kXHQ7=!IHJ?_^k*yo(CKvp z!TK~|Djk(6dUUh=08CHISVR2E#Zb?sdEh9wRTZdGDL$2V)#yIWCI5vGI2 z5S=^5~mKs|$RU;-Ef- zGir3!Q zPj2SPlIFZj?DTWDIJSXr0N>gub?J(GMV8SQG-EFy#Pm;@XfHc6GA%cam8TtM<+>~#2b$iT* zGwVbJ{=OPHIPrhUI` zFUkvc08Q(23+Ab99O8nj_+Az5NsqTtpVa<5ts5_61_Szy{3SSwEA1?iocZRbl>Yp< zH9RV!_bRvVRUIi|nr2%flMmGf`;s!_fkx0@2;&fTMaGnQ=^277tUl){tn7E1J{gB1 zKv^ZG&%nmeh)yT;51{T*z0+jv)0TYgI<5x2Mi0>TcfIbXbXR+R{&@U44q>(ZAEjg< zR@(OG#uKrxsl)1X%Jz*X+MG^NSW+(oEpnp|J4y^P?MPjtY(WcuJb@Km@vdUPuv4;$ z78H1*GBtK9UDH0Qf|H0STl-K4iZm8N_XfBkTpnNDm-A@t{cWv2@71C1;jc#hlj{5? z66)CuKa|8TU-heuZn>QqAWoJGPd)|N_8T{O;}FK?elB_#P?MWz1FvWUv>&9jM z6lIS8wb25mOFH5)cagRGhplUMIWxgzoKNLRrN|GK06RB~+BbDqym#${u3jlLIyl@? z|4HZ(gDVHg+HX)7j<~Wi)y>35?tD!&Z$O8>7gm_^V;Fi~S&hkm?>QzwVFZ0z)O1=p zS~7#n0urYy)<2&V;+d*k16RJpo8VIe%6dItGc^lAIr4qb z3|M1^Lf9TNwis;Ya|2{L%E|uk^QmNs6(ZjTGbffG0lejjJlKl-=?U^c&9J^JeM1_s zX?glX2l>Tc>t>d39@n#OraYq6TmN0lFM8CF+)dcBnL*jW2o=Tg_(c@W@pNUs;_rNk zxvV)e2u#UlczP`Ek*2^rCr~G|aBX}+TNcBTLtjhcHWSG*t5%SZ&Y;*RTON~rby4i=^Obe|xH6TWh%7Oh>8Sw)@`edC zHJmjYKog2MF`jxQXUcWJOfi@S`WU<@roj)T&@pV3 zaIE`f^k!0(ky`pqc3pxeVrYk_X6_Bkhbt>C=94)&k)w8bybKQFa?=IkI4Ix`j(e$C zSsu1AvJKHSaO$j0pAqQs1~Lu#rp7_LjqH~<1kO}$-ZUeN@{X%~n@sB8_x#%nQw)Wb zP)pU|_W5hRKLUpKf^%mG`IqKYQ*%3u;78_E<8pzPd96V^CdVpAPmD?VoogNFTQleg zv1)H`AnmPy%^{Dh$p;_`Qtl=Ab~2M#&zxTUOfS8!2I8@i4%=1omYEn|8aQzvs!fS< z-$s2dnE2x~0~vZV#vtSXvq@)Gqo*kg=OKU%P*fZ3cl?<&WZ7To&F_-zb+VjrbxM_i z7HdeVu(mKi6=AJGxGrBs;?(Pj12T#vuomnqFMgY~=N6X!XfR146$D%ScO2J?9bT+s zdjCLjE|y*UCYCKjRl1#niG7VW%R<`nMZ1Q<&ebRQtFS!@DOwZ8pcAUE;d0pF#g#SZ z%-3o{@|iu$mwvdQUGU5ng{kwYbQQ@qTQNf(&ddH8(TL5R-bgLl0CIsGJeKV_e;-x` zjg@)OG<#z)3C$rth;*PgiP*PlAAzU5R^GkyF|5F64nL4;00BAJp0sCGBWCfVLV@G@ zWQg`VO6|NtN2j?h&ha*mvLf_zE(`Ko{xQMYTHYhZ!k@cw?XROK)KB3{zAeA(r?kPy zBba*85^=)FmXaX!SK!^>fG<)XGxx%EXq^bM;oruOO*<7+5Sls%!%4*wUgW3ehZQ6^cEmv5qW7>Dc@)eut&OU_fTc$i*{iv% z(9H!tU}8bpU!NJU^b`%PRXt|Hay1^4?Z4Fm9G?ajZ~_G9dhy|7@u`+j#o^mssh_>M zpAOEnyN}hRZ>z)x(`MP%w7(*bXjb*&7~2{TgGM&$>~*1I?irAjiTNPKflL z$e_eOKAcn9v;|Bn7Y=qy>Nu%c_P0Hk98h@XpbOStD0U8>G=8NKjo|;E>rqv_o%rLF zB=bdzw3A-3mwl>)WXN#rqs$h&jhP98ss)nSRwF@ncA=!4Bf3l@eNj870z+&QD+XLe zZu;JlMX~MADX4x@)zdF;ipsCt#E3Snk(#w{E>0E`Xw`WriNCr%umuz+*;)WBG%)Q2zUZRqHlGNhCaDwFM#*U-j$7_bWBH7ZQ44w&&OWT(Qx{Ddsq(!nbGf@S9N_mcD?Q{gqU@XkSlSY%&o(Bmuo6R_DJ^0 z=iY^^f&j0i7UK#*^AG~QQw^(4yg`(i>R{w;k z1Rh&n#$^_orS{WSj=gRUD)#sNUNu60ow^S4T*BpY%Revdau#om%9#PusE`PPZ3flj zblRNfs~I@YN5|Eb1%P~O(%k%wdnbRy#{$%(`9%rYuAz_2jW|vX8F&gVM(82q*wO#$OXw*03|4ZD&4_y3Rhd~8h%(Yw{>oPr*#-I*HTevm0;plZU?=lD^ zHU@0Z)M2^F*QZhOf??4e_y?IwP;y!X_Jhx@~dr!n`UhyPqlW#)$NU;@eS6C>DNJyw5{zcf!SZ?PheU z*>qFJLuRD;ekQrUhcIn&dA;5h7(=VuFn_da;+ir3Qh==Wkf{l!O;f1{%Os+X{p+>0 zf?g1QXTVwhjcI3I6v?c}^qo`0OLln3y#vIg9f?{0a>Mm1*5#;bfBGmOdRCll>rDF* zoG|;m=o)OJ@z$v3V41?y9Q64XM+ueu*%OQ^5>3p_Gg1++Z-f)Aj*d$avSTN{5ydaG zi(1E`Ov?wATV5GbE@Z9ZaOZ3W*hGr#BBBrQ?m;F|h79sYYxDRwv~ZVOuo&sT(*nk=qR$lpHF7qe9yD+UU@_Io#VFt+cgkhUA@k4CH5!^(I$@0 zGjNE7&a)_SE1UjILuS{T<(>#oG;%|F9o2}85QZ35IzUI|201bOQ+0 z?gGYj?j+pEoNwyW-?0VISx2ek;IQ7{Qj*vg^Jjq*BwN%u;{?GDm=hUPtzL@nt0x%H zYwGQOCJlJY8V*W;I&XldqGW8QF6e;m+K*{MxgFs}%(f}~-kMVhPhyQG$FuhfnIQwf zP&nmeAL3A05+E&{%S6Cd%dLZ>jGuTyw9=c6>vfw;M1*o z1%cZcmTn4mL3)V0_VxjSgP8LN943Gt6{G@n`>j&XOhO<3xg$`he33% zVI%=%wc?WBOF-0$T;WDjul_zp2Jo_cR{_jOburM>2!WT{Z3`h9vVP&R#(O@ zIpG&-1ldo4T%uErTY8VYdy0!YUYvK`TFe|8lkhk~1227e}ecZ#( zLfR-=LgSbeU3)mu=|pPy>hMCP+vw(0l&ZLZ-yt6!$LpXaTbS9*SNO`Kb3dVv`?nia z3v9h@NX68cK$76mL<(b0H>1w|?eF)MMEsZP0*;Xv#oMf5=^K0VtwKQA%l;~-YdFyRt{{8HP>)x=?ewK))stTQ+F(C#Q~B0<|`D-NDv4N z-2YwkD6Q-NwUo=^&q5a?dmCW!wNFL2>HL%wFwH|_KG^r0AmSe(H)Mlov^B!(>zygvHy zG#9c7FG12f*xHY#iXmlYdz{N_BwcP`zahnY?W{%f{)pW@-x{E$=TG)J$w4jVio`zn zpl*?I_vrkp96OULhZ1r;WYmCyMnaqk@Nsqurlp`+jy_v?HB^h!D$wiGP-8d)G!{Z|wI)nmLKc#CDbb zQy>81cb(T4LO19m{2OB?S#57aG_!@++0IU&$kqq*T(rY+PG+vaKWOE&9Lt=J-^2cG zU5zK)fMa1Q_YI~4&jcsOL@{w@@(;WkY64~E4~{SM#!kv-PuPj-SgfziHlC^W2vc+l zI-f~9J`|i5R?3#|5-X->eY3ZYDZLO?qp{+EJzAa{^cgu5!Bek`ZsPJFL*`Kt(F9Cd zU??J!)*p^Og+Cx7=~Jw@Asm?9^6bVC4REQiL+Qc$E%yC@zwyREoIW7g}^oQUKCuzIUUP!nM}lmHKWz^fv% zksjMT{PEnZ`jZonVqb=7E<&gZQV??>m$kI#Ly&t;g0lmFNc;+)rKr|1J(Z0G?vU%* zKb$(up^7a|r+);{<)MWJ56#mZyZW)&iSOJ9-Hl%45bsz4Uc@~Wbb7~(} zziRmxt0F!A01}KWqJp8Zx3RkR!t|hu`3|q@8%>j+<^?(a&@0_hcpSdryj95@J>?AO#wyu@dw2eDx@zFv#%5B-u#H>nJgeSc|(W$Ho|$KzyI0p6E* zg)orj24A4-@AcX~E76!|J(2{-@WQovR?0^17z~KTYjvZhhCDtpp`Woc9~Y?r&$QCk z!S6I3yj4|kYhrFdYb9El+DyLdE*wE+)S7taInFoJuszXGYS$35))kkynY}1cjjcAJ z^aJj)dH6 zHX+DxtM^IiEwZWW#MhmZp36j=EwYxo3YfKRH&6VXDz2u9flk^3@qzt`L+;5d6th9d zd1#JKvitx^>iH!hSP#QfkPvw1)+5Wc)tp(Mv9vz@vG9yN9|aOCuQBMX{^Jd;4_fHv zK9$qr1~uTgpMG?O$N~9GD)9Jxa<{LIf}&39l5v_7E`@y(3hB6RXOSKR`K89ced%>( zx05ZlAv6g7-UzHna7t_U&ll0}>x=r;8>46dJm?%u1(wlZp}fY@@~^NW*;$L zIQkaLoTRet2@4<+Y6Zt~AuU8%j;l17XFB_VH%a0w3HsFMT`|ZNqo(!LNZk7l;x9mL z>pX`-VhUjf*CL6#`{bS#R5%F9S2=_$7!psHS`o6^*^A6sZ@qa)#pv3BQ?-S9)9y+% z>uxAhj_d1re!QP9#q;lH1G*gp#uujC;P!{trOy0?g(XY&cKh_)w*p%;Sl_TIs5>+J zf>7{thgw?091*j;njv_In@CpEOr65BB(_n^@*izZ+%wS-InsM|Y&pyUxxs&%$dRih zJ1TL_AlggQjDsK-Q#8xxTWOz^iYfCq-=KksO{cg@x!EF(xA9$DwU|o!BY=Rgy%f^# z6tf8s`sCxxpHv2v#b^d7PQ_y>6CoLCD8YQ*q%RFq8_ivUOswm)E;CHHlir$z^_yTk zA)vU(*AnA2bc>GwO!-qF8^xcW`^SYSTzG3vdG? z6%Wg7(VL9ze&k2ni%_!6t!UWTR{lN0O#(Fw|~y0gj9yJCVF$X zVMDy}nl;Py_TF;eB>?z%eQ@awQN+@N+{!Oq-sunVQO$$$Wjv6lw>G;b=lxsWHjvLP z(Y!M$%q}C1+Zwdl%S&>2TDb2Ez)Yu$IQ*&gdMjCR+Y+jt!rE$+jrStUNAwl)wJ(n@ z`_`bY$b3B%m+nJZzIdk^m>S%*JECL@VO72CGB-*lcH`oBU8(Q!5VmD8$x|SsNm}pU zcweS^aHoE96+W|5Aj)z#iim-E*>GO6_!HUR2gpBl^(^ythtbh3y}6p;-J2;xEw?~* z(g+sOOK`kYR81L(ry=Ie@lNFjF*zL7!S#4`R6Q)<=?JPU=0lcDUx>F`w(p%V|*=9_mHC*UCE+2Y8yM zB{KiHKQ~70)l5GM_no<((8I+N8b;F8(Dd69)?t8JlaDvOFd>9 z*hta=#MpxNcOQ-OnUWnN)WwaLTjVHXO6>LaW~@$V3SjGR{Z&q~0i-m~)Ij@sC@7Uw z^}tP5ztRJ^U zy7EAC;ACTVO|x+j2&Q^F_TYMHYMy z)Qz3DB&NSL-GPXAUnp>sGG2akVYCUWo8|O3&2HJ=1`sKcCjV$`sQIcu^g1bS6jf_g znj1SxQx&2>i=cve!YQYD`9af?;Rh1}s5;v3k0I!`dG$*;6S*Iaz6wA&v)Yg6)RzIG z5P=<1RzUN_w{QE+mx@Zp;5wkV9aT738U>phY1hz_W)2> zm`tuaw9^y6r?AuGvj@tT(IZ$XcG#na+kP~%ij%}fH6XVQchIFOt%!CxL{<#qb4d8@wg$2O>K?4-3Ho(%0Rs= z)U=VfOps^(`+L@W#*L@gicOgLpAH;+`D?cgI024P**kRq3fKR+aKmfldQiJ*{1o-J zeYG2LMEuOA0yP%2Xk#!wM=U6!Ijbc6D{cJ;TVxa+RsokF(9DL_fSYd; zz1Yw|6S8VA|0w4mg`g~83y^3#!*-hL-HrFfBprm)sL=UQ}IT5|D+A<}< z-nKxUd)HK(K=dVS^PG<@@WK5X==ysqLrcV{`ol~;#twWDD62@1#+0`!>p$I_p*Dcx z4hbvh;7;|4bFAX~O8$3$!J$7^|4WWTW_~WuePFi7Xkhz)zlhDDh{61QtUww*)jcew zUbK^wKNv2|>IirKe7*2&-e|&b;XB&30q}QGzj+eEf$$U!Iqc@9`SxrpuDbEU@urgakVanP9nKhu)?lnU1OVg!p6`tJ>8Yn)tl)$eKZ ziLGj)w(Einp_aGrNE3ssL{U!<6&02Pg7t=L$wo`+pm^DLjp{kt`uReWP;(di^FvHc z{*a5ru=Dql7=dfAa0ZWkEDYD3F$gbTBYbvG;;k9gc8;B8`fcWqoBz&iU!dVUlt6`` zt!c>$g-Xb;&nl|GFL9Fy))vPR-A-n?f?XLP8PXNJM2~PtP+**f)buAiA`XQ^=lUH#X$kdtlbj#1=kYa{iAMcN!r73D?;mSF}8 z?TkXlUymCZKr-elzIR-sgFIm;7;`7GA1b-~%i72=E0>J{pdJFC7z#l(kmwUD3dSmT z4h0n1-$yEKaPD96@TO`&IT>~XRi#o+rCe$>4CS1&4FC6LoqMlg zG|=|Q=sh?`yS9oV`GvI>DU#Qw`-|_yuJT0`!|=X2*)ja$jTJ7DvI*BGP)tpE>DaTS zcW>xmXW$fTd@muC+-W{iUU^Yo5`MN%guTM*o#8wjmt^djLgA8P=t-#J_;JUC_D25J zhf%IsVuAL~_d$*e2F^{S=Xvdud_(npg@y5gp?meWxpi35$`^`sU>!5~Vc~=n;vPBo zYyzB4?zWSLb4smP!2oBOSJsn^Po_SJb$Aiw#XK^&c!pdY5GH$mx`C~L;z9kG^F4eZ zwAWYPXhfF;qF`U9K^49te7;*j7djp@l#LOw7Ucu(wxrQqeolR7IMmZZa&4DbN%Ia_ zp+)NjwJe5+rR|Ep7fr9b@1Eb-LOR6sQpx$P7ZyE( znqFlX?~e+_9@P);f*ZDAbxJW87Q$SthMqHID8EiKr8AYk9gm)XCsoUKLzrSVu!-PC zxZrO%k0R8H*WyBh0U`&9bF*A~bWbyRN4oMMQ)-_r=Ica8gAzdWu>gzp?5j&oarP+r zcy*#%mKW4C`8u*q?1=>4u_=lOdI26a3#JVcy2T%U{Z-SKsKC&izft*o1#L@}0%9FA zcNA_6%I$MV8DFYar$wIA`ew7`xqMo*e^7a`AgRkr)=JOeOh^0vS;8+)4-{tDbIvb< z+h0`DxX}7YS;|!mb8T?nC&>`Sw_f=RTpV%_hzAl7}~Ib`1M(g9`AVVRcHJ z>Mf3>jol^p&*-iaY*-Gn$D5pS^%Z|*xqQysv=Gx4Eb2uwsRcy81nO|JG)$Kx+ zOxWJSxrRlYPI!V$O`nNn%-vxCXzyZMH1K59Df8CMcfhNLy+GEvuu9=OPx)ny#aypV z7I8b#^n5t8#J2}TPPD7pevFIkP_gX@%fpY~)hAFHJ5bEtPATY7FVSij=RBe7ZF{v4 zsFexl#$BiB$&?bhsb;earXj+*XzQ&18mCb&dml;REJ zK9)bPhh6%=&2mdXa>Nho;PL&L;R>^|cTK*4H?pu2IM*i^Xm!U|$Zd26(*qT51B7jd z_>^3vQTYHNVWw%@eMQCM5tz8c2Qj6$7gcr|+MX=)IP2=fqmRl-PhY~ghA+ji0xZ9& zjO9tC;%gBabeIm73<%xga-w-IpLBY(fOeF&8M_mDXLs~|nT#bnOg^9^*YgZY7U&wq zdhu8V9v+Qn9X99oY&N!?4S5TWa~SLLjfc09#8OjxMyV1@R*Ef5-zsM9n5#0q+B{%B znqM#)d$cCW#YLW8^!;xOA&|`SWM)GN=qu`&2lxF|7Mcd*eVvX*LK0^G>wI zG$7!Fyd6+`nzDN>^;ffCWy%3rg)>Qf4b(xtm+TPX+#*74z4el5$4~OOJ+{wVHWoKw zRoW=~c#c(n7GNZwvy9gjZGQ5?H`wZ3x(Y5wRGdDbM87!U@<@!cOV*zGl6e9Zf_NgH?b_XyFspUQ`ApOTvQO0l=`!nS_N%pg@ir8ygkV5`Adxzo|%~q^I&C$2jnfkCVnntb1XZeoE;bm(&_4uRA@Ct{Xz6Ym323?(| zCG?K=M2$Lgq9Y_`EXZ%{u2x{=%ajy+SKC) z9OE1WLkz$1Xa)}a1ix5_m$7O7?XZ7E93%Iz7AQsu~AAl&I3xfoU2-rfKoz@S=nW z!VBV&6|>2SAx|lg=v1ZY^GW+hS}7ZX?!<%g(zg_GcgXm6G_QDK>F)PKvuOtxy!vFH zxI>ub8N^K!RdraQUgC=y(qkL+x zp6Se&K?2$Q)Sygecv8%f8}nm$czAY3c6L5~>!nrnv~<$7Bfp?h;i2VdB23pj^~G4lKTApsz_iFP5kf_QQT_#@EtGj zG%^sDdqBi%zFtlUKQs-;WP2XkXcjo=&;wYgfl_6wgfR2RBgke7%WY^LqPQT)-azBN z3u5ojTtU##Qr#X}i2T;=F;r^ELPCkJq_Hrh5V1Wip;24sc-8`?$PsL%_p-zy+NPqT+n z$&{+o0gmOL6y6`xFaYWP_0BJQiDi#zDnv7Z>&(Gds8RZjF%r?flghJo-u36Ii*Ny< ztrVg7Rcx?rq3}3U?eNDR^+}mx*L%!nUwJp_&Z_32#q?cu)}~*-saw0hoEQeOR8nME z-3f^E-fyB7IwA>PGxXKYZ36(z@G6>R<=mF;8(C)Xb%>-Q&YQA2Q3f%qAF^_R$O6D9 zmfhpPHi@k$MKP~{$4I&vRdz{ePZ&M3(uA)mpkEwF0$qlGGv|C4KE)e(+hYoS`b3RJ zs@)nf;}F-f`J1!{dqimdx37h=8_-H4rXNJF=y6Zf3w_rD!y;MuNDeK3pDA7i+Jgu1 z*Q<*q5J4zObzWI+ZG9bS{?1Z@8`S3umGAwtAp+y0;`*&j4uLn+0NIoe1azh ztIw%)>En*a_P|W_Ivo4+(s3GKiK%F8w($Y}HM#r9%;Q8Q%9vCr0|BY3sNkydZ&6e*j;Eki`TqYt?GI+z}2AEn0)>BC~-`XyeEz3H9yc@UE=nYuf(k+6D= zp?>4%*sN6*0h>n(_Gzyv(uxq!#$7%=7;)B_9KQvIW+(PJ)7ksD%2@4Z8VkL3=x|m$ zM~RZC=Fssg_F2=Kf@Q}2dB}SgfvP0XT6)3o%RE^v7 ze1fnqfR3^ell)PW#5bQe#GR}1am%)XKP4>1_<4n^_dCG+dH&iUFW0B3yd+Z%T;_ZP z>OKO-vdV&7ZV*~7o!_rS&gU^o9Bl+J8rqat)JrLfG)}aje0|=_R6#zB*_{q6BqTP` z$h$$OTwOnAm08yp;eTHp>E*&QD*LoCFVA*fp-mpO^I_QXKp%9qGCGMz>y)Zu6_(!9 zE;~M=6@9O$F?gN(IDQNoF}wk>S&&~i2+epUIAt_buZjNVzC0)FYxh7u&1pe?&`$%c zHZKqQ8n~ToiBcN!VG%|m!nc>~tE2Ct%U#nPC}T(7;ApAhL+^i5Jgpw3?w91Rv9LprLE70b71r*miaHK@~kHKCICtV z`%Gj>pg5z3U-YmN{TJ~y0;L+DdRV6LrQ@YR$J^S<>)v0u#v;8%D57pDVF#h?6pJ??ULsZd_+wNS(=+6v*tF~2lPB^^S4y73ePNCF*8Z;PgA`TE;R0xRK9q-miXT#= zQMMwKm(Ty+xBT$D1ZSXQi?2TJ&(|oG4ZbInsU{x$9H03(N##4(`3K&+C<0}@A^5}y(!wU;skrOv7LH{@Q>3zvzY(4a`Z z_&R+;@YiUT20r}8K{Y-}SGFr&9cpEApRg;>!uAd^-83v&aB!$Nt0Qk@6p(J;H)2Zg zlRcw;MS%klFpQZu(Dpz@N3A{Q`1{e2hO|MdfN!s=c$e9igdw4L0p^pkOELHO>o8b; zVIMukE;)oMwt3PtDVz|OAVEvL=k|KubGmReIX3}}D7?Tu2Q#eoM7<&H&SuB+x?sb5 z{9QjU`CD^E+AMenK*_x z2vdxTnMFyG)!@bm_cNLDU`o#U^!W9~g^)zGzTf&`fo#B7fLmMA{Y7F#EUD@y`~4t3 zZ2+<6AYdwboRd|5b=O!TR+cGDL$`%1&NRXL+K#`uPA{f5ZMcO-4 zTDJjz1Y0E(QxwcJD@>MeqNoCiPp!At#LUnlErQ^ZX#xrV>~%C_qSB6JZHDzjP?k6a0a|Y$3Zxp-q$eJ97?gy2qkc!Mr4& zp_cmRf)LS1gyJwTg-(#rH&>Pt5jOW`#^Z;5_B~3RA4wQ;riLI(uXvIXt;{hg@n8@x z9g3O6V(*-&{w?LF?g}M;Zec{Ijv_f54Ck*mg`uqMzLziDk z%?h1*2fs7pjB)EoY?k@Pq=9D-;+fvoK&(E+eWBGi)Nm@(SUsocmMblRWRV#?L&hjt zXcU~Aa01&K{669h)85T;KkbbSFn9}qz6mHQdbLOrCPWMb%$yAh(F_<)p4YVV=bFYY_RXtd4Bi#{?02)utfdl4GDRE|29S zc%BhA{V( z_;io){@Ma*5CJ)Y{O@cv13#)zCm;lJAj=dHs_e~P3lG|X=U4gVzQ-QY41vy8_x1qy z#!-~Ccb4VxmasTajVG92#)o_(Ss(#np=Kf%vC)2s+*PCvdu}Kn)z-^WzF4XWk!!zf zoVUA3DfQOU4P6GbVDg<~B8tvM2!r~re0N6s+lf1^MdgM`c|JtW8x9%3J4bjBBu8@q zz#HnIcc1QFO%QZaajfy-pl|u92M=?ehL6qYL`PH+oL`GK1nf18 z?-^HO)lCX*tEQZLJuRCX+-uC2$R5M$Ug)>VB|e!zS=?ipStHJ7G1^nd69YZC5|5oT zu*HX{pL{Y4=|Wdz$==b^ZLP1l^`G$-I^XBDrI~9{)c`3=B-iJeJ_ztUd~oXu^}AD# zva5jc{4?mVep7GvH-F?9RJQx}=4(Rhp+uRQ&E;BTj<1d$mTK02tx7XEcla0^!1%;$ z%jPXbrt}1cL?Dz}u#U=?v+l6$5JlalOHsT)E$*D0g804HRTj<6YHd0e)1?q%aB8oU zISjJRpOvK96aXGZJD|Kxf=UNou_;;Y-Gy1;O6v$|ip}3doNJUK^pP_pEf^z^U`Ae{lI!LS~cw8Bt70h>K<4lAWAg-Gvq;+@xGB73Q{`IScPG|Z6% zm4|OtR&UrGJ9s|g2bI(G%e)B!S~v9Slw)qT!0wd#N{PEW6QlPpD)(<4V>-Bp_wRc& zd1%|wNNuWsM1_vsO_})CT3#G+6qn7=H2o4`F^A*GJkoJz&rU;==W-XGJuWW@6RU!V zGdyV3^_CG=e)$Dsicc0Y-b$BC>k#`Kyh4h*K?Og_%5{6+-N+b^FAkTp3f;rnRwbBQ z2)`^conKyl-dBInLS!Wzc}>5K_F%)SbcHBC72KOA@{Z8aur2}qG>(LLt$0jJ?zIGm z29NZB(Bd$k8kQAl@3t^gSnnw{R7dN6^LR~7)8d26TP6)(0Q|D&n|+|7pHJxO(=JF8 zkf3jeJS|sLCbQMVcbXwpqHVqpsPsBKEB(*jy6)fapG({xQBalkpXO_fF3iEqL-sNU zqOtw%^XX1U6Gj?r@scu8oe|ijM$$IMVO><*Im!B{&89hLvhj-~ZgISSd*-rqI=pl9 zIQ2MP&MrrySND}=6;?!)0jskr6|WepvL52^O-;lZ%e`Rf_E)|=ZX4;D6zrHB+y>{F4?@^4zo2V{hQy3-XJr#Ey(54m_A&e z&4a~hHtxD;J+ahuZlZJ^*(w@Oibsx-Fa~lSqOdlW%zI1n=tkn$);_L~*%?@g87dZT zc&~Z-opTsJ^4KV=*8B2=iE)K|?+0IIzDuUt<2MID+^FyAk9IecOt4q~y|(Z}=x>rR zDuRm&fXZF&Gix-d`0J(fp2|$b<)xlhpT}Ft!1a?QQm?*Xj4#5uBw52|zh} z3b+;d82#E|EpS#zmL9GX`YKb&x{f_?S5lyPtup04*=|&TJUU>wd7Q#32GiGo(l)f|)F48jIdDE-!VBbJ(v z1wljqo!0FxkIiIXDSn@{nk3$G_S`AWEzOdyr98c--HW@v=b|{T-3RMDYcRzx!CGI& z$g>B@0_qVhb-G9*;33+CZ%06Ep zY3*)J3=kkzLhbnO>IEfzh=iK(cvOk&sW0zsA^A;6nn`T+00MMZDTrmuk@sj@ z!_;sU7%bM2Y)4V!4ja6?Nj48_I?qi_N2Ad@SUT(LaK+~5j?;vaGZ0?N-mvYh1hRhT zq|Zt7z0Ms!g}M!`kLz0v+jF2p)9%j?4O-vdzE%XMAa>~b;!o#mw0%O~t|!!Nv;_K~ z^BnRN_fa$LRQ}rPajAGlLOY>y0|0Wci>h(@-JX9;{ZLpi`iZiJJ<8A}Ff1 zy}y)>6%=K~S+#E5&1#gYw3M)W$7tK1-9CuKhM<>X)i)_v(&qOD99B?OetvteAJ6(& z5c$}p<6L~pgKwPNaVELHEbx!1TA?^bRkTIfk8w}-hAqD{S0F&!3rWanz2Fw%P6Z0m zoo{XcUhlwhMgf4IpxjyOb2v*7m~5uB^&-eIlH*IVQMth5zRcN$6p%uyg{E%dz{HEu9_m}VuKLT?i4fA4P-PU~#_#V#GU%l*VX{3}lCM)4TM=NPYs2dqG20|)h(+MH7(FJ2xd zt{J0nh`VdXCe+Yk=`?N)cpJ9Z#$$UD*sM^Z+P3@(Im&B9n7zZVq&aN$Idq9qJ`H4{ zaJA$RG+LcFY4`1cBoAqPu}i4)&x0;UC9jcbw;+6}U+=3KR<2(@pEqu*^O1cv|GPdU zYW1hLy48YXEqMnr#zk8vNEdVi@0!^%5$l+-;Cx<0YLKp!8;WVrP}4kXv;Qi=P~J2s=UN|EO4JXu%~Ce&|Q$yQ%l=b-eusS`B~ zE?<79YLbTcl^Wg?0G7a^{=mTST-ytPkBfl-|499FIKnU7hao48#aDP<<7afSE6T92 z8w+XZ!YL2n_tlZ|>lN#9AcVWQl1PoAjK#5%p=%9L1U z8_F2fwmUR=2Cj3NY?8X!d7!mF!1F zrw}82&3wZhYeF9t>^9NE40cR@Epv5hz0(4Ez>(q=R(5h0+Or@sQz-r{5<)06)|u_A zT+O1sKC{B?Q@U>FZ^fd}ONC0|#1igzsY*@pqcgbxC)pmbXxL|F{)rTkE+G}j((aSD z{1W8I`SyTA`>~hyzy<>%e0HNGb}y)(4(WWGo``qjXuX6yDv0}9A(czkVtzvR>*4D6 zCKq8sC?U$*V}$DLMaC%yjoF4%bRh@ah#o&a$)7aqS~oifrNn47+wa<=Rx( zJQ=r@ZB1@9@=MCwd(O^ZOyJKuq0MA!ZSrx^k1@_Gtuqy~|JLI*8qWpJwgIcg*9W0D z$}=$mOB>i)7q9v4A9%px??rWSeNdH}?CD@8{im#3c`^0}B+~Kt$etPLeG^M)k~*VY z^bt-%hk-d0W3#oko28D{ zXE79C_>)MYNWP%((UT`Zm7Q>_WHLsoYtspt8^AV6+_l+VBJ`WH?q@sZ0exkHH<8S) zHSeWi?;EMRCDMe%JPDv>0{zBmp6+ia&t*=#5nNHHh)e2jtB~)&2o?u@f~Qku&%;RN zcakU!g@7EGEl?1(A8>CUz8~$A*JaD=MUQ!`kU?v9mi(rV-bR&5<|7t@2VjecEyB-D zdCpmi56@md`ukpw#y5-aEUq9SdCK*s-r@J5r7=V7gOw$U-aS!I5;H3?jfhb zqQSHMcMT1$0XX~lKIYA-#2L{dN#LUIAZLC=qA#Ud%7i_VqFEpFV@!=cs!9e;^2hMw;wzIwsJZ;vNXd5d__ob@v^Y5sxjBGhVO3SOmoriQ+$^4)(^HOoL}Ul3&T+= z+cez_oUWhMv5O&c5GXNln@CP0uwS;_L9eZ~$Z*G{0`?&bD)TJ0Q}Nd-9>~%h7WS0X9o9?H5U|y`SzN4 zD8TErm#TLV;cJ}`S=J+(LlK|02;>p^y(;)jJ)|;Kowl*#@vJj=!4-(JCsYWi#gBU= zJ2nVKj~ZxIs#M6Ro=}PZV#nfsja)KesR@+aCuNg_=f?xjSdv8loaI&1%k)>Q8T$^? z$}zZV67Z*Fzsurlzn`X}*14JB`tj9#>HP-0>uUgxXB|u?KTDAJEWQi{43aaqMgh-y zLU2AXY-%|o*IwlViS_*j|K_-Wf)NHiQ|Ipz1by{&MA>7JI!#Cj+@3klLzJ`+HhABGX z*yNE3lc&kpk{)cR=oNL=K7CAO>Ke(j2_dgN6o~aj;<>+;yx)8OEq;MS5mwh#7a#t< zd7U!KOf(x4Ljfj!k%Qd)PmrVhN#Tpha?@|@+oo`luHbItT5hA0B+r2^xf0m)kjNA# zl*q(ls0YfpD#HX_`aYR7?(Jk!nt8dbs0wBxDAFJ&|4mPqKhpX~N5+&x7V; zz~8834U|_tqkLen*|c^7#06g&_>}lP1~l1!C$SgeirHQh`awx{LT_;;l9z6H_*aKV@h<~E9K}7O}pN|&p@9gO7vU# zeYt!y;a?750jS%1=$mtVRZ}dL5`x&3`>C4r@P7N8#_!saWnPlOvjSI8#$^)t_y6l2 zt@=xUkGw<|n*&mmo|CC|;D@uXiJMAyLFsMGZ4@vy$`=iNtmcq!ITb6f!<{Q1s_^p(8+ zRgNst<$7fe@@waBjNttQ_xyDY{vN+y%0G3NpHkc?;2VFg75RKbP4pXs`Kt}*!~XB> zv25XdR8{oXHF&l_zAD@P>$*xl0%(){=Nhze`Mk%j`s=!J0m%ar|q`!Xs+t%1`49-&iu{9>W z1IJ7V9LG-%0JFEK89pLdtZ#R`LhR;Hy{%z^!t)av4Zd0d(A;Mlq1em>96*}xH~;+n z#xXQOy_mh1u~cxL#Y5`kDoDq3XR@)O(yEzSV)KK5{xb5KQN&y#lqH+y-ZVi=W(FWy zh*Db8K_rIoAm^%3Xz-&J@8|FM14DnU{+%0rjnX`pCWa$~j(XpBy}ipJt_b7#sI38o zOpfCylxmp{^nVY*`?^EBGk&mVA-yKAYC$90CS5UaMOf7qIJtQ#z=(52Z?|5ZkMc@- zdCPp#)^(hN%{Y#c>#Iq;++1HvEPv~wLY>mPTSkBo2~zrZ-Z-6ZxQ4}=U;;czOu&Jb zQy6fB(ghCJ&*UO&z{@j0pOC1n$n@Id6(c&eqI7`sB%Y*^*TXo?!sfcNu)T?VJu!Lf zw~0n&{kRHnE{>>i!hp)&Ey+|?o||u8EQt)r(LCRW@%Q|%MkPjer`riyaitWyo$|*O zpFwz%pqCxzCQEj^!`%E$UmsNZ9f1wHw{yd6D>pY_tisQKvp~KRthdKU|JxW@-;#9o z)1~|n;Xy1PSqG?e>A)x89$1Mb>ACOY9=)o~&c@#vR?7sqkG7cPs>zjmX#6YGxlfPU}!Hy7@~j^p9jjrNx|&&O1IjqsQ_|?acQA)Sbc-qK#u z_ic;y`CcQsaDU&@>s>xTzBbr9#UosR_MU9Q*C%1fuw{;OOTXSp zxGGN;Y<7Pg;GV@C=h=6MAvgy7w%LC??gaM!T=?X(WK*H=EM$hE>rAa&le_^gIOxw& zlS{*q5w^SJNBID`G`c??J|L~_$#z3{ zc7|GG^VRtU!iaxi{bzUxz+K>u4AWe$ABU^Z=zd=LxoJ2-Rr@{93U;%A=dT8>bcxO$E5jKy;MqJ(nrV$r9qp>U~wL1+4hLBnMHX{W@4 z_}YT+;%kZ7-`}L$!=>`LvbauMPUpZ8keG#_S`36=3`_vil;Twmkw`%B8qb0EI|l^> zH|0e*#*s$Q4qg}sH-QQT+AU`XcmhH%zPBEX`nt4QA`I?Xkl)j%!{zV~45323L$~;N zVQu5Cn?fvsipEGbBUcWS44DBpW<~&9Sih+o5RZ-g`Wr)jI!*D(EnX%pPB#}aOX9ka zU1gYy9NZvipRZ4&xT?2d$|EsEo_igO;x0gbZ+AWC2h>hb;#v`jS~0)lH_1;d;Otjw zdJC1A(X#mJ^N|D=;)yiNdb!#ututQpMuV!*Gz`z>!DiqijLvodtkJYNJ^9O7UILW` zkxs>jf7A`*9$Y^ek}0opwUB<28MW1M-%BEPP)BGEcGfgX-87}LZW;2Z#%bwbyunMaILfYkcBUtw}Fx8m-i3fb1hq~mHvM}NB>JtFe6B&)u6QQ%!4#0 z#_T$qrm1PLGouiM^&SJX&q@AO2ivQ=?8m4*yA3*YS_75bvVigwK2vJ~>%WV^ul)p` zu{Zb-rgzX^h1{Tk}gY}~Sm<%dvBNzmr&U`nY$KGhUD0~^c?oMUVJ zGjQ=5yw}*>B&pLZ#U{b?7)0Qr)funLec`VSCs`F>Enses0y1j?gl+km(C?}sAZh5u zx%JLi3Ytjqw!!-k3RqMIRQts9L<*X4D`RU3*d4M8o zYeW`na@H&^Ks~p71wdk3wn=$^eCPlsmH<0eBk73)GqKB?+1Se)EKY8iKZghE_)ov(O*@IDtHK0{out7V=nz`2;t~zARqLc586PP*Mck40DT7-f0j z`t6|nlfwX8>O7P#fv~SC?s-5ET>=fcbZn8FzFR1 z{Q7(ldr`dFgZuf}hC3MXNwG%xoA72MJ=?_33p4<$esiS;*N}b^;W5i(3)^be`}1b^ z7(2CG{N;UtSs{@yaqCZ>+Hc}- zEF;FAH3u{(2y)x!OX&kJ&x_0m^t}zM2m{f6Od5yEfRilx5lW z$Jy6c@%d=V>sSF+M62aLbiq;HKig=5Gq^fQ(>G=}q)sx48)VrujxcvKhhE@nc>?F9 zS`TQJR>0#H>zr@}&Ke11er!O9G0-;6^^<|c*$f>CB)X*5Zo%pA5hJ#FxYFM?x(rBG zqnHTDUm9DfJYb`ye*oI#4X_7v3-E>1e83#gC(xald}R$x=D5$tLju4T$q<~+ZNKcs zJ^Xa{6NGe#IwP2b+C&pjeyB;nLbk12|vo*F`1~bbr|9nrw ze{Ba*3aZRO8PEkpsokcdZ86Xe@cU#kQSpop{Q31v0r&R%Z0Fj2&jUZre&c5K#(B1q z41QR>KR|aZDXaAxI1h9GJUq`i;|3d%uIJ4#d{P*9t{@;K7|KI*D@ZVjwMrpK0ptS$` z-#;9-b==nfvR-2bKL2-GE#QgQfBAFYfA}B1{$KyT|DOW>sbtkfMFx+3@%wh=Jq_ys z*|`{g4byG6e7gm>3K8&+5WgPZ-{*PRVDRt42rS+gEhol8B*l1u2T%PQSVt_dNG!CK zB!+F$>tl=&w^|pQRvkqoT1X;llzOr2|9G$3u~&F81q9RYU)itVDFMk2uGkX%V}3uv zU{dJ_`kUU6_WkiJqBrm~#ZR3);RlmIDduX;yJ@BG@M+%TR;_6O`fT)VBCB}a)7QwK z@~PO+FpO3&Q`>gr>0hdU9Ch`wuLj>ungCqQUb)YD#7vLD!JhMGmXRcWo12sO8AIOv zt}zO5XkG%!rYKaGU+Bk6c)vAYl%GghJ|Dh>wBieVcDl9wb|vZS2zL6rzpb?Y^*yyF zAq+(@k^Z4P+XOlEp}&a4*CHJwmuzNH3JqPQ4Al1jVedMY9mRs|2eH7rrG!8j;k|ZP z;l2O*xxDI_KbYNgL`T(og%IS;JSJ}Q-9s^C^@~z$=t1>+ z>a}J0*ubXpT;p!yjY+lN#?{ed_x*ww${!B-wsa#f7Ww^Y&+)6{$(m-HebS3N&Zh{R}$h38Sn9Rot-n;h9IzpQBbSw7Z0p=N${@TL z*nk45wzalnI_U&X#%&~JE`|oDNtPR5K!#(qDYG9NS%6ab*p3e2Aa3PDo$LWPtwr9& z%0^hjLuJQBmPK?~uDidp{Z>GwInf`$*v=y-yS(*PlaKc6+-mgqbHr#s0DPG z3~s;zHuU7CbJcMszkR5ZKRX5($0eCA8U~BQm7zbo~y@nZ&sV`8F zKqP3~^Y)ryn|4p)>5qs%<{ zR!|H$k?p9(9Qt76*4S)T$fC+Ewct6 z$XMtvvTOsmTvjP9b5R<*7k&!1OpeE0o@bR%rbleHG6{m?*Rw}B4uWBlo^c@eYP^EB zU>de`T;fTJ=j^3aA8{b$dVDQ`zHJztEC<}1GJjYPD5t><;m&=nErh9&3cRE{0m}yD#hm?oY;hKl`s=N+_=%8Gw4uyH$a= zG{b>34uJk=-8+$ov)L zio{dy;&y@j5eFZ{eje8#-qdrnyW!lx`lHUhc)j-b9`|?DD!)C{ntPADER}k4jf9x4 zssX6#`hy$k?|**(N9s0EdyQocH5NDRqGmphcR>rohVo$+pcT*V#>Kxny!)<$T`GNf zeCOK!J5Sgy(@2`L?|uIEoB6`^zH!U|bxtWJ@O$$;ZIro7i^s|j{M?WDLgy#yrTL#1 z;xj{Re1HpWiD_G`V1NS}*Z&B`j>9M+Q^ep>-@2vK8FM7T)o&}6MbDv32e1$;Qw{c- zZ5vG5UAL`zL$EM{xo5`n2W(&V$A2rJHoLa? zC+B^dc}J(g2Cl9LM}rVUNvO7_bbE}CfKiPBB*dWx0+YlseLP*^VKPvxg^Vrnc|}2$ z42;|FVE{XuG*^3L*_m9x5@c+32MsneyKhNv4u-@Aryn$vQ}I!7L5K{3r41r zvJ~gtEBL2RP--y?$p*R_6hH#Mex^4fk;+dKeQ_WUfRrc>2{zix(c%N>p*Pfdxbft+gpZFm zp;!zIQj-Ahr%@oAXWZzVACc<4J&n{dh-UQq0oClY`X0I}!1Lq!>$A&`!o&f>=p1m8 zRCVunfS21v1{@keL3zH|6H2g;pTs2#7To!LXnD}wdk^rZZ80diycNVi1RMrb@M6KfMA_ zJ$RD`;(zcH`e^Iey9WBc`P94GnqPz$;HzxB$G#A6JToC|6Vc+ zb&j(bM?$6hwse~z3%BlYAAtJu{4M+Uo@3W&t}=Khz3e1%S&zF7@m+Lagm29@s&7X> zr)*j8{*1dn8vtUED)6aj(b~eo#^tgj7Js*nDJ;QtBlWHDkG~!^M8;mj zyMw29PCsN5n(N)gZfzHnOdI@%UFF{UO>{puLDL(A>qM{_;`>dE04R0)p3|PlYHbz` zyGKJBZV9gSP>l*z9j~ zV-w3gVt*V~0N1s8gz)7YB+^8Mc%IN^4G8TJkXQ-m^7X0IpBl9LW@ProA*LK?g9LKp zC0h49v-KB4DwP_9wi$Y@ndo)kRS9eDmEbz7W;d7gi_kZplF)hJr`|zx4gqeMbThO7 z5q{&Z;R?%KJHQCWd02o0Ip7!O!~aalB)ZSfOxFgR{#+A(%0>xae9QZB=0#^`9-KYRzo+=JBT`SKt8 z{1#u&YCx&l2B-5>1OfVV#(~I#jj8}Eq%Kz)1kDZ0-3z6c`EjiANYw!U4`3nR`^I4L zg#hhk(9(e49^gzYB&cWYPv;fw4HY>eq88kdoH!e7c*`<`yhC0$V^v zO$D_M6j*-8%C_IDX1oAIh`{p8+sOxDhnA)N`CW!W|WjJ5avW$zf|0U882DwzD^arc$J z+TC*n{G7gL-_gsSGC z@A*Ja8Qd~d#n(lfYKfxxX9#*bOYb-t58%0i3F-NpMF0gnpN4~g^;<8wF1-&o*pFO1 z7;drp{{0ckHs0kHJ$ z=mD~sB4CW#MZW%?$9gi03(z0H-)VL-s^5xdW(bA*Rgb80t+(})_Ix`C0GuTdO{)Ad zkp`fwg&V+63C!VG1oM8}Ge|&#_UJZI&}LL+315IG7y(np1P5*?^;4LPO}4QCu1p~u z`{**Gp0iFRLFn7mLdaLMlAwVXs3TwK6UFXXCeKNS zZ)qHy<1ldH$DWD~zccy^@N@LAK-h3NUmtS?v?=_mM^||WKdgno)%)e`t#ZS$EE=4a z)q%%M|MqA6W{UDmvqsiuKwPGqm3~j)A*UzB`a=XbO(SQJ@n2bvyASQmtE+r9z-!21 zoM(k36$qg|_#eK!Ey2Bq3P7&dQxG%Y$g*Ahh+la1qC%F>ek`|&$T$#o`8mDs*#nrS zs1pX7OEd1dpq&M;gF36v%5Uy?PGjdqai2pP!^Ub-1+-b zxg%r}C5>zCeZ@6wbfBUg$3gG1bMt&FpZDH4X*9`MfLISf$pBWGg!VcbGr0LWx@URd z)KndOrq33qA+*>S=5Nkd#d3WXVK^VjY)eM}<83`Y;k1O!T2wv4yV>yc0fV)H+P>a@ ze0}R}was-=dWh3d_oJ~b%QYLh4o6FP&+de#XyICULXfRuk>)n4qXJD{gY)w(SPC?} z_Y>|ps&LM9F;7@CSVI~fimKxO<}dE&2?l^^#eC)l$^7sB1P_#<+uBeA@4yFG7fF4~ zpiRqK^f4qlzA8-P8_XW}?2xhHhU$gD?r41H(lC2S;@q$DZ;yl6iU39Yr zJEOibe4v2>W4jj5N0jT_XbQe<#Kr9~18lf{4a!22Bc1i_Cq4jAaw9tMk?rnTEs!w4 zJMZvzkO5E~LKf?Q>27VVEWJRkA01AXF}r(Y0z2Vp1pzAYxVwFq=mIB{Y*wkl@BC=v z%bcJfy%wMWt6Sd!lkE)Ev%Q`_DqRoh*7EZe#07vo(E%F4$S+jT3?^&4j+S8HF#HUXSbUQK`UYJi3kv=lFeK(bv1blXw>`lGb0phy1Dp*!+=Tg(CV}H76P$ zcb^5T(sl9%f9p;%ZQ1O~fyQdvD27?KSuq=ixI>G8#hBISPs{V%_plt)3YC@a*s18o z?DxLL&ia#wU`w_?PW94(hgIJ<#t-Xg{mQ}_ekWKfw7j1k0#@~q*oEm?QEIrxDWy>{ zN5f+D6Zkh7;K_I?^8$*D`db4KJSy@u6i={NV=$k9s@|4dMM%!k>OS!Srx3VVpFcwX zIdI!J1kBj@lQ06!p55b!#N|UY;QpD+a;XMt4Ew|uN=6<)EQw*5{rU(L(*i6Y4-iIx zIW}d|_ed&fSYP$UUiSyb+}`DT0&PJ3bc;5RyE$Vgw-JHeu?sL)je2kiWTVxdYSs-agMJDpk6g*I@n@|Yu z8lV%0_qR=2G-C5`6YJ~d_Cp0)pp7b{sWrUy8t&p z$iFlOGM@8^ljhyqRx7Q$Mhk}Fds5-?$Z;0F-{S!Js*^Pqr~OM~+vJ{8x*wwbvgIB| zM>vNvMyG?-*QP9$Qt5wyR|Dp(b3Wb=htJbaXZ%xI1CW*N473Dgc+pS;PxKukzX4p< zbwlgEpF0pyqEJOkG?(8se#0A1WBGS(UEoDNyr+EcP2R`rIb)R^EgvlEGSqT$ zbckUH(jFUgI1*ZSfFWE)^b3H1zUM-Epgkb&2Ndh~U(eFOX%f$XikFDEBOAU*DFn-^Ch-@k@c7DBkW_Qct7_ z!pxi2&ob9LlHh5l&X^NWUJrN7PAy8H^=?T|pQc1SptHpzrL&;z>_O^IkQ{JJf^3jA zqj>>6TbAoIQh$ADcWTaW7C6quddOlpR~nM{%;k|7FB5$Oii>lAx6MgfU5BH-;`w8v zd;*H}&u=o16Mda!cjp_R@;c3&=zCG1ZkX&N*c_y~&9r8_6GMF;&zI5@|cV@+V`mV8MKLC2OBjZY9lhu-j0LYMj zOuH{`_dNglrhvCYSYxtg9Xlpa?tR7TIMspDd0vuA@?A@WY3{95Q|7NIK9A6L29D$S z*&3ApKj#1Nm??H&y!wfdM!oEiG~~-2eza3Nnry3eVO*t&>R$sP`O9`I_4pQp6|r`Y z+LQY(+_5r8?WVlRQR29pSRF;Xkr6G_{E8YnCxK!2TA*Yy(c`u=mE|`j6w;3ow@Wtp zkivG4&RiH|F01Rq&aLgUyj~yY^^9bb@_b-!%7cFMuaIcsj{&8zA&jB9R4w^MpfgF8 z!d3c>2ls`JPfKv4JuRM`AUOzHv(?8=Pt}u))wm1sGVHkYx^zD(gX|`B zm@XIh1W`H^{m~zeR~%Wk{Y26FnhJlVVjHPxID?PaS;+;Z#T^GqDvRhT)<7Gc`xz4srwTdejWo%> zctS^#*;sa&X+xM*05}xU4mqjRyyY>y}srxf#_cv z0*8LY@EnMyJ5u8l@}_2#Rnq#Lw9M_w{a)bih(#r~d>&;Of}}0&%HkP659EoayF9L} zM&;6q{zy@H=ys}_Rd|`Ok@pQxiooP`J9%)TbzX8UEdUyrL@xg>2EgEmWj?9Xyz4zLS*AtV7 zIag1Qp#}e%6W6T3+6B$n%_cUR0lklZ*JyCyrovLJhtb_resQfUQ%=y?-0Y zS@kBymi}IqY1HL5FFmP5e&PpEW)4!drxMB)Mtwl`MC{`*yX8m2-h`mcEXr3JM?_eO zH3!z#=HvyH$c1}z{57vNSr;#i@x60I%X&bKsEd2+)MeWtOXCC!WSp{HLzQVvpH5~- z%m;Q_ad@vq^vWxo02`2`&)?VlRoo!yPBX2e+yw-y$8p6YHfnv9_|t^A)JE8=YK8?? zI{SQNnl}t_$sRIN@T6VC8mX48{h;Bu+fZd?nmll3^qc(PA>E8Z^>O#8;;+^Q*Q6@a z#FZ8nV6o9Y$`)6tJ*~PzZF2eQ8!F(o!dF^xae33a=TutlWIDCPo)NGS4VBHuTj;Sv zva$;?jl%wFR6tyR8O}c|#_aV2xiiQ3pEHWWwGay5(A65~5!Lh@eh?r&$f-<5s z9iN_q2UmzZ8H*NJm=851uyy4vETR}I<A(nkyy7OpgBladTX9+eNTbl#D-3IWuvgV zz%$YG9b4G$9bW?qeSAxgL8%$?N6uAii|%Ipx?}36!ZE)n_DHSujlC9|H=nj9la+qN zZwBaGMqk7*G=aQs2Kwl6B{|U~Ces9X@H&-ZcF?!yQ; zE-ziT@__t6`Nobi>74f+Bpwdquc0;ULawbR;1GY1-MTB~PhAsMBU90tj$4I)$XGuA za%0ZSoUEzIqXvl}q|%nb^?miIRo|Ps)1UsmaE4+(#+5>Hp}ya*vZF+@-A4Vg0sJZi zT2HzNT-jl9MR0|>|3hTchNMAV?O5TY6Zu^XyLguE6r7+)pMrueBFwAz1Ggc3uq5}s z*GE_x7X&+ZPJTg-nnhRL+ZkxL6HF%?yvu|7o%q8NKCx=J0nc~^z@QJ$LmRVOKkTZg zr+n$*K{`vJ%XctMbG3_~ z8kTBxXV0W@7H<)(6P7QkWcFzX{*L6e9d-Ire8GE!W^%Sq1IzMS=VhSvfR(KFV~1g0 zWPnW@3xQwll-p;<3&)6NNV|=wMKP_tne4ifzYv2QNTW5RfPQMqs?4k%+Wf*;?`;#T z{+Am^3SNRVGpx6Kr9P({$3DDyDDA~C&JXVovO#^1BXCfzBuV zVP*B<_rH>4UR!gOG)#`@e*Un(o49q)aWU{$LHFayk?+ri%UV*APdzQsC#ZcuLjYI4 z?B$fNbo>og`u5v5+Tk>~P9Cx#ic(updlq(0upqTKG6IcQyh@{xw;=~+n@RRBKYfro zmGAQb=SXXMVgvx^-n=5^pW*xr6ii0qVj-h_mD)*Ry{(f`wgjM-`X8rHJzCC77o#;@ zahWg6sJP3|;j8M#0l80?<1=d?=C)gvO0+=%Yor7gGBYh~x+bG?3sS5z=|BUHwZQM3 zW1}4|138ayO;?xRPq$A9Gph@$T1=C+QT0BmrvWZ1?5att0ltQ}Zb%-el;V3qY(4Nb z>SF-U1zpl+hE+10DT0!dwL@zB5i`rLJdlqBJNgDscZo4gYN;{s52FLFpR1pCF1P8+ ztj=MON5H?Gt(Z3LkyId3rs-VV@GWPi?#v8Xf<5EuW%Z{!AI*v_kJuCVj{J^GpEMx; z#m)0o>2*WwN4)AkAH&w!P7v=Tz`NtA({!gS`Gb#$1b7q9ES~E?3v(Q=$S$@ZSDz$E z!LdaRUUAJm#jQiwd&G^1{J>;z4&tFtA681HovBZaPkcktJ~G{WkKHH>S^qIVP`}@G z>aD)`(b@TYk4u=>fES+h^TM1~1E(8_30=uYXAaFzg}KbVBMM>&=F`xfjLP z%k?8LAXhrk9j|6vlZJ<4JX_7<{ok+n3CMI|Dsp~`@6#6KVdwNX#9MUn>c9KrekIXr zw*K*A@kea8R)M~@^=)6d-nRLNF!u$A7t1`q$>ci$&=Snr{mDGp_ac4Y3+pLd`W1Qi zFa2v(4UoZld#X(mKXrF%uq`@*-d?6{BOP<7cTP>}|F77?jM0vcMAE0|w?_13y`)?^jU)NdXVz;sP-hv!U6Id-(N-X&H* z+#mnXrEQ#RK8UFTv9t5RY*FAy{7RTrYYcG7wQu;fZEbxE}e)kSY=OO}wI z_}fQ56qM%6BN8JG-C#{k2XidYrVn!TdI!*MNw|Q1Zr{civLj~}+~3%63H_DTv!&SS z;|EI>!%GPOF0E<%F^=N?tQQyc=a;LpvRpGcrUg}R6jbi*eOWhQc?r+i-6`*6rSo!&LiAFFiZm`9 z8f0qFC8+S+w)y!*kw{s$$RJ%dk1O-71N)-u%C0F+;}1rho_0(kTL<6=@^ak(Z*nA9Z9COCgr7_XeUMRm#--2oms87-rjSF7gyc?k0|DBVyW-xp%JHJ^T#f3zibD zX=pXYrWRMn+qdGSZz9;Q_%NHwsg6xdPK(izSSST@vgzb)QcbXWzW$A1K4=|inHxT+ zppWZ_zgSP-G!s=E*Tg7lwe0vc^q^R|eK7HHR-|jtRR*E?iF#oH3+-jfHtVye(5uHN z{y}8{v4NV@zcx|=>X^74$0_|)2oWg{C_eK&%<7h&mjvdjPolzyhk*-mO zvR}6;v2)sd_cThCZ`D(tcE8izSH7hVEB{-UO$LHa0&5c6cnTd6|H1Hbbjyu{pX;7& zIsVap*gI$K#_kq!fR@}SQ66#lPWiTgOq1$0J{EL-;4pkhzE0(RhWq=o`y6nwe>!bZ zuIzoJv_{5rr;2`##`ek&9WZ+@hW@I3i!Jca({--*uval6PWvK;r(CqisW=f{rg=sv zVM{?%Jr~N;8{yy78Jvmf;);x);t&~iQHJJAaTYEybln~ee@dpG%;97E9&*Vw0jvtq zTGyoD!{UGli5#T^8dY=}KR5;vf9NC8YN`UlLa-MsPx2{t;E%VFF8X+LCmR(xZQzjj zdNPFVPVO>V)hC9VMwp1O(nS~*S#x}4__RH$M+|RWqIZ!J{Wt>yYa!_icVZ)ZS@BRu z-LZVXKX$SJ!4;JQAm%BfNn#Dtp;zmCpr{|%sqrSL_Bc?sDu3n|{}HlbRfC6+pTm}> zmjv@|+;@R%uLdQy(V(x!cWQ)&GgHa63mZWVxk4^9Iz~s?d}57jD|NKwY9P6~+9!{` zx=W9R)rw0>4bojU&nTfb2&c5?R&&ABFNa7Ce2#poaVacJRtk;&AV+#%&F zd96Wx@V|6S}sGam9nqcU!}{LU9wz7>>t+c?i7gN1l4`A;ttbmLm)cS9X2ke>YlL29)@|z8m*4 z4_H}Owy&SudvUO99L+H+Te)gpMEdX^IJHI_LBy1;6y(_Dp8)Z{Z_eI=d@f8uEugOK zF~oWU)O?A84At&Vi&Bx4AN%KGPIHg=&{Fhyek|X2N@2`7Z3q$VcnmS6|1LhOSgXcm zqf%s#s1N1P*hkO8ES-%lRh2OQZC>zknvb=_HEJo(nEYiwXE0ug5nMmaal6>wA1ps?eo>nUr~V^Y>-b13M!R^%8oP;PF<`EIFJd4B`>j(NX> zKwOT``uOwu4F0Qin!-mKF?Eu2`I-^*G*b`FjU#v#L5k{oN~$jew>o|Zc?D6rC+5aN z$Y^FRj?t(30?VtfscJjs!^K`r6Ub;UKCY4WJ~mcnCBpeXRAb-$Gz!ktJ~zOTm15F3 zcJai)d_m3bLiWM1Y6NQ%zX;u#DNmE_Z`+r~fh&?uE)0{Um%L`(SES?<1U!?W8hG*a%l@Tbu2WQTC6deBWMoxOM0`4=B1ADaxjwTwm* zvBsjDrBS|A3M}GVQJhqqZXHanA~Zozxi;{fD$qlXAG9$+46dzn2oMzyFlMpVAC|PW}7mAGzQv z{~TgcSPujEC9^3XRj9uGGx5?be8+FpnKicyM_oR=FDqJ0rHMl&OXR?N4V#R9XBIrc zc%J!lL*rdyir+F+_T0Smz!)@0bWEtlBveQ0e6FUde<`XCn~$HQ(NCt|TmbMF8O#$n zl7wh*?j^u6bwBWSD8E1_-+EDFbaiPJYK~OulP{KZ<1;BFD?)vQ{(dw(_M)Wsdm==H z9WCNGwA409ZL=P{cjp27n$dYd<3ZuBznPEU#RLA-uP+xTiq=cW>mO#E zmiLvs$51xXj1})34jqPj|+PTENLt$$oXvm6~Dk^Y|~aDZL^ZPKW6E>dZyR? zMP0@fdybpVB;j$*JE^+`60fs|`sFi<*pK```b2_K_%L^`<@vQ@1z57JXS>uvz9JoK zze9TpS_WL0ory~H%S&zc(_1Sa`1#SF^$DoKEMG&2+)^P%xWILQiKti zi`IC{LR9Wf+9hcs?vEQ^4`+N@OKMy%cqhg^X<)-?e;rRZxHhVf7DkW+;?&1N8r9o? z@vtINhQoqsINB;mo8w1QT*ws%o$LzNTIv9z-DWRwwj)=oYQ z1!aUIF0qVG9wlL2bks;x;IO41CA`E&5?hxX7it{ITi50bT(u0g4vzb@AqYlTW!wCH zOl#dd-Vb8M97+rJ<83FR@_TbUDSeIl33P%BvKv@IsSW_rGC&ypp6?^6js#x%x2DL4 zvXO7pJV4Z5sP4F;x*5203RWqk2$#@xamHr-jf$~JzHQjrY3A6Tt-vd%Wuw=NB%`?W zd$2X9FITD^+N@2SiuqiVmw#&oZv|pCDsyQ<*pN>a3uwmMU)*$HUA%JF^B$?CY)3#C zabfiCE7t4ry_7U=pE$tJ_x3=7GPu9L=ospz_vqu(tOkch`&hk~qs~jBiFDh+-4OzI zYQ;=)nmyqG^0H1mS9zl2x=1!r^T;c}^bLWQZ=MeZT~O4ad=`YMKDO=+=ZHSvua%tH zaLW?Wg=_d#;`LXvpt!UEZjFnL?bO`c?DtCT0aLD6ouPN^=dJPU-~AVPD%@76Vy#cOsH4hB-zn0NJ5p7N}JCDF_JdB8PmczaS#yS<=*aV+WC zacZ*Ck(`lUL|~ugVmG-6SEZtaVM{)=-`hVx&XU1fYGCCp)tgUZ-CdJESb^5j>7hEj zm5B>E!ziR~v#}v`NDquh6tNPHRa^K`E&rm6o!Ep1PWLhAEUYm28eaEjP&}kUt6uM^ za4W-p2QphivQ3XV;OWcK9vYgcM+pgu0vEL&JmPhTK`h{g7ZuWg`&18q?a|5a*%i-v%x4Ihc&?|M~nXPPT@%XUiKjf&M?dY`y z3BvGZ!E$R1I1MDbi$x|+gmY&7OUUsIsW=ilDnK4<4j$`8cf$;wHvpdLR!mwVMm>kbfPOpdzQzl?xPr=AYFZp~QvM5-b+tf#kLK!AVr!R(Of z7e7=`I+xW~^owSnS4io*#QflfNCNhq!tHxioeBOp1AmcJ-ci_i@x`I4lhY*BVdn!S zZL47+kJpiIMVEzkGrYO61JDWpPPeZC zpjJv@M>{E5;$6Dxh`zMM>6;dX@RW%{RJj^ zxa>C71QiQpq%5mWbu1JK304dOQziEMoNYL2$q{6X9c>O{Hu&$Ap>sF>T*icf52HS0 zx>vD4uJ8r*tO38S;aArKi@&d=b<)nuNWG=$IhE}5QHZrV&fN00qdZ&$;?=$yL?P?1 zTf23^{U};KknL5GZ+{g-M-+xrvHA`|CBp3U^-N1N(aa5m^G6CBq{5Qx9K`yxb7cZ| z@M}bGk^g$SV@PPW@(73Tf4y7L1vn=G$>5yK8_h(FH2bo=P-;X;GMUC;1_qJic_mD~ zkEye>V5E{-wC_(Q|KO=i{-%M%&>#8aI8n8YAmW*G7yt*vxj6VdBe}UH-{*G})F(2j z@ce@-75Vw4mgg7Mkpf8LSanCrmum+M8_R5+{!*e|oR+Gf` zcyUy5rSa#9encwepBX(ov~#g_e*Vy<|8d{-{l2O+maXBv3&$A*HEB-}$guZqXA(*t z`i=B|`)v2TP$A-KP4{$U9(UhgUyqbk{q>?<8epyM-k%mEr28EF>)nr2k&*7>aGd^u z_k1%wL97)eR7PS!sYfUu7&)EbdY%ywn(f=rZKx_mF4mMYv~S{_Qz(F=WGXD!mD_a0V~f6<-6@1fDZ<3SQ7>@<4y@s=JV>hAB~{=PqdX8HtF z{P;|3U=Y`F@7aiacp{f$^OyIxKjiD4g0sgt$nwP>byW7s({3nZC*m0K%w_om*ufdx zPjIDONjzZyy3vqrourI|_0i>^;i-**Q`LXb>vz9izZNbB%of8{fB8_uNmpKT8|ay00HhXy(ZY znx487d*Iq7X6c-;7e%Wcl%!Zl%K>WtE8~iTw*@*^3v;dW{Z)oY5$K{5V}4msq1he2 z@GgL*NM@R{%v>(srtl1%bV;rN8q5 z93Y}wgrw%c-?L(w*2U4I30UV#P<0e79Yf+!SV=ZsQ7im1m`FAFhAKWtUI$Q;#X$%6 zbbbXM1@(}u#ncK+C2HTZCa3(alO6K>-q*Ho7zFF^yl65_%JYBtx{hTxfo=Oi49ICH z2>}vOWSEh2&V2oCj^E(b9o%xc%Jz{Z)a+iN>E7!~AvjC*k&gAN*7pNS%}VA+k{r0V zD!_FJ|9m#OBK2Tb&&N02mK&3KsY!wouIylsfoA1p$r!_8BiRfeZjQcU;+n2@^G`KS zL%D{5=n=6~&rUoe`|Xju(si)a+3I{;z}mO1c^~exr&v{uDa}%1gVNIu+0!ke*-D6D zW;378PaYBSV?XVOC>_5b+>G#5IOjGc?3I&RrUS`@w+MM&94a~8Kxy6x;pY2 za!o|7SeD2b&>%FUm6+Wr`T>$M?NWlLX5$$arypuNp#+#gsMcDP9NLIrzcMoq)3w8* z3kAQtzg%b$M zgE(%RM~8f51N%H#SOp9Kq=W@41dDC*y2`Py!*5P%JhVr~rL50XTi%F=zU^-fCFTdG zKEdzHzFxCcda0NtZnrH`;Sc6uU8>iGSTaVe(!s`cU*lk&O(cy%D3u?mrsrq;+!M;4 zAX)C!jDC9xd$b4L(gD==j!ZT}mF4;U8YRtGm3#PL7!kZOM^q}l@q8WDMvifaTWLFuLXp6w!E5iPa|SlopZj7@Cm{f9tupJP|CJhY_WEYeO;kxX#kli_+VD;7PnVu9VEg}0wt zUr=%YK7Eqi1>4d~9KWNaU*q&D#XROprn#29E-e-4OG<_mjrMI7pritRAiR_x5Uwn0iDr>x04)4w&Ct{tH+nB}z)`;9aY>yB+ zGS{x z-wx0Qm3Cw5Snfr_*u%o=?Az1ur(KC6AD`Z5CHFbaDp#)sOa(qy*JR=|71wi2)hI}( z_Q7mB!QvZ)O|5Fo4D!u}lb3A9T*RGi$}0GKX_KFDM%mI@0|D(3y%?}~T-#W6x&Yq?h3z2WjGg>#?&PjLz)Mzv z;A0StGajArP;x3wi3H5s^Ni>HM1d;hnZhHfX&GqsLix*`P^Q1Oq0Vzmi+s6#RT^+9l3@7Mjasx9F+FyKi-w< zdS4y0c6`>Hz}##-kBqbdc==#2;q*%GNz9Ow z8;w;IFKr^Az$#$E+&L)U5ebFZutqUhN4;$dWS57WVxi3k@=gN~VTai|%EjcYSrrlBKEP)w+j+1Lhl>tnIB7u-1{I40#?QRlME zDb=a*9bu)%=LMOGF6HQ04fv?g;pB!qfa|%l_)j;W*(rJ?UdG+`I+*fO@m!;?FxvF! z6`2K?3vd~1zYMdbr4@e>P%uW{g#8=(W}RP~gln{k*JsaA$uG3O55S2q!BZE6urvo< z0+PH37h35}tpaQj&eWaTN?+r~-`shFbFZdSeW14siktcU$s3A&zj2`6Dr;fY#yVcv z`;C@|Jqj~{hzqzbu_^J8;jd2YfA0eNdTRfWi8P+YIwRx4CKKJvvB=_BdiM+`S-r;Iop>~EknJI+;1R zx4^{}i3QISUTrd7_uZu<_RMOTgp2ts<5=&e&-C5{oR(r;bmJJx^tjTopeH>s*D3C0 zQD$XrRe`IRr*^K)7IJX+b+ViWCGe4qq?3}G7h3M=i|8y8&TpzURU_)OQOf#D|ix(dB7XLLr6)~+6*W_WT|r_m3W|%YrUYfJ#Cmh>h^i8q&>nXu4 zp9m|ZO6N@NY7Ai&KmK?hE<=W|DSlB_hZ8#NQ$sddpF)7V0);*4QgAm zj3@aL;6>W!y1Vy92O3m1)UOh&)&V*_b1#3V!+A5`sN5qG`Pwc7xAss^@HaLJUr(!+ zMKyZL%@*ZJM0IDK_lln{XxryRgLlsoi3Td%DKSFjvCnIk9N2b)=;eGI@dQ46uYp#U0yGyvze_WWAB)(lNllsQ=l65n5o79j>v!NDUuN>?zWZVeOYO&P z@v2tw4ScEo;5p=FTDdTwW*>W!>y!n@dntxJ+yaU7e38_q;uy~TnQ`^V2cwc33Trg7 z8&Y(~^ZSSj`FH_pnq#wKQf{)P&)%e+=X1zc#~$UpCyEZ?C^{gS5yK(8{5m!ZLiu^2 zU%`a*9P14NV=50g6q!t&$fcw*N$Fqn7bQ0*T(bl^F`0qyiJlce)hxO4%dA%Vg)rV@-`+I`hYj^@=gGIh6L%Dr zf2vupLIt@E6(+=Q^lSvEEFqw=Q0;DnyRDx?=uw7E5QM72(HNQshfOwziKBBTCW%*x ziUP$7qv)NEE;ZQ|KtFm%3hC@%afs}&;Nk+7>mFw$n)_4dIE~h z?WCf@Ro%tEfZd4cov6QnG_LqVnOjZTAxaXz>#~FE+t>u;Ma@~0Y5|l z?yf9auzx+VJ-POODWtf_nRKFOB%Zzg>VW@S#~JyyQpeSs>U2V_FM71b{bmwSIIuCo zdO8;f#KO@m3x^-O<#P8W%Q%5j1Y6QXVWkv!6M)r6i12_9UcYoGCpvq|5F5p$HO?YN z_`S?}(0eS3<;9(B@~^IngSiI4X`Q4UkVek)fFA(?;cNuFeW2qwcfDA+Z@zipO6vjr zsHdD6p=4z}e`mCmXSng9QYq@D>xButz0$v&5{BpmkB`rAp#n};cn!&=59vP3O>owo z7}YWwJaE(O#AMh-)ySK7Wtr9@BR7wMO2SY}SHYd*U$#5r&$g%*BFJ7~BHpC-9-W z;Yr8&w$iKF&~gzOLf1k|bu1{R|4!i(iGO5pUJ zcrX^+XCHhxd%(@QWq-j{Y(3$1_CuwsHt407FiPkcu2Yh2F#KA)gRC0CpU*nf7sT&u zN}LcT2pe;1T|Jg4`>2&2Q`RHyPism$9pdj|4r zA5nu+vxs+Uk*oK8uZK1;lf|WNw0iE2$Cb_RdVlO@7#H}}4Fpy#%9r@q6Xsf|J2|6s zN7dbVjpiLccK_x0>(zB#s@0?8v}t)yW%@fJhTm$x;CJy+%P1F~pjf@@d)Z6luO3{i znpTzlYYQ-#wGEUSjDj+2Lsp8;Q-Oru17wk{R0^K`adq8=;>G&PZC-ypfbf}5Qs_zX z;u*qC^$kp=<{h^Em>7dg#uyGv%F3ef_fjrJwW1BFsJ3_jU+jJx?|USgxUH4Ygl+=; zc^zAsxbm3s8HpzDQA=b{3zb{NV?}UJIBD>by~@u9M>r$T`6j23BQU zbKijNlnRDm4o6!yD(%(c3$|81zu`SNk+BGDqv~J98U;gl>W4JA(w$|*sJtlRVvoE; zB(0D;L;O9lNu@l?JxQJgMHcAifi3pn?In}|bJ#p{ktJs`y=Gz=S~Jn$T9v|n8epI2 zpu1-;XVT_ed~cnKsa+U2j!1^jvap|8#~p35sw_>s*->@Ib4VxX%|#KGOa`;HQ&8`8 zr@C@8$#AdIn%7aO9Gp1-?$z5HY&}LgvjUs>b;ncnXGeliDltb;Bh=ZY>82*?h$mG; zbpecI+lBt<_lGHkNj*_cy4WuCt;6_gU5$_go09qN|74nlnC??bFWy|-SjVZvmtOhC z;VhK(NT>lXfQ9N?DUJL<=;W8EbZW9s9MR-Io=7TrHu!IZnhc zKApTQ0Fu@6&Ltwa0`@ODoi|OqkPj~!b$GGa55awm{mtG5)ua!QPItf`quK=IAr>vD1V9Eg--kjfjG7%^ zcl=QkPr=>2Hp{HiB$F?{{ggocw`rI>KEPtG{=Kewhz+c4*61zMjU7~S77_(!I;m}~ zO2lm8hm#-a@RdHti}#op(B28$J@;oGVZh@|^_1#Rr9{?O%uc?NZr`Wlcz4Fu0zvV5-9~z8${hyH_ zwcq>oI8s)2d0C=8V?C&-b<-$~+VY(~zjYw_R%W`UPs2ie4BI051b3Q9X)^e&)!kvq$BE!f<&EJ56Ucu@@NqGhb8f2Jcj!eNESL_mwC@fXmG;we{5ffCTv6JyqP&OkwF> z)5E^6+CSgnWDFg5t-tN1zH3+Pz7Ir{-Jh=8prGq9dB&d%xAz$+~?5_n>JQ}_1Ee_!EZpbOmMw~P3SMWD8g?X)Jwzn zvakp#S?S0coS43flPo@~JklQbj-7Z2Q)W~LrR%ioId$e7JP(2ijLGRy1mDDW75SL9 zoG|g=6OA^SEr!!&9hF~EgrUcDLKVB$s4d2Qjy^2!@kpkG56~5fu*X;q^5hHXOBblv zTqYA*?461$t1PX7$o>82u@${?xT=CJ~yY zlDnVm!+LgCu|v?BS5>i-c@6k$bWFv~G?D*}yh)^NTcDc9sCFV408{;v0P~X>JjV5Y z6pVdVgRI($B?0+{OGkwRFsXL9Jf)}eI$`dtAMg6RP!J9K%KdyZgWzBfuK`SAbzL3t;Zy;D{@f0%{=#4MNjL>BG` zbz5Gb>CD?_^vmf{9fwwMr^1Wt{@k(w9UR;Xh!&vpvTcBKWO4R%0}~xmrN_o^Z(o~j zG#MOwEUM)NXd%mBg1nnI2GE^9U8+U}#_?Uu(d&wq7On1#B2O;%+yGpabH5&mb)zR@ z(=I-*yLYH2N{zX;sXL;tAu-vp+JvC{g`8zZ=mUJ)?>4*B;SQ1FGHz3TRz_*fha0u> zI~8UeenbpThL^r)JgrU_LP4+?jaig6H7#UdoCVRLd(>yElwtQsV}h9wD}AVuYRkk* z(8Ug)zg*t`^_{+3tj)sp6smB+ZOSrce1Ooe-jAQk3YdUd{b|yH0Sq9mi!DqC(38d= zWQK&32B1?>I{hM%UFeayjaZ4^gpa|{?$1s|m6{5o#&3(#}iI7Q_eYR@DbR^L> z^MlO@lJI0itk=8)iEa?Ii!Z>!%D@;wi^@U`VyUVH@O+hf<6RtMy203~CjrYO8w`du zZRnA`PkjO{v)kV+KlT|;8_tuF$%A=`x8 z&-XZboq-N|C{U&mJRTmR z5pc;LQ^>L*y{=^9eCNl^=3Bc|%+tCuqT(d-U|f^1bdJ z<`kRa`kfDn(F-cy`C4IO_U{M?w4eOA#qizZpDP&X`XG^t6!M8<{R0I~C`IVy6nZ-^ z@63vl;viA;D&q0h!MyOOB!F6gKI*S;=}KS&uAoy3vFfi&3;29pY^&n~{9xoH9ZzEfz{wrP<2W z$tjhLCt_rbKYLh-y*kUT+{jt}fkIgX%!lF5qc;N4g`KV>Q>bPI8!&CTC2M-g^Y&;s z?-b61$0Ib`0id8*laJryg)NJ0;2TyC7|@xd_9`N#{TQVG-)qr@m+1^n-F?-TXP+3~ zIyXV_x2U`DoeJg~wRXDM2PKkzJhk&ftfmJE$T=l>AYqeAnE%%UTcRMQUKK{a7`Q4J zIO@i2X8GSfujwkxgYdvf4cC;ir|Z@T(3r&FELGA>#OfW<4@lIpS<({<6KqDAppX3N zQ=n7azkcXg?_OCqh|dkd_Qzxr6BAqF@O=ew#<&4Wr9;v(QMaZH;Sl)OU5U38IgGX^ zk+G6=UBV}Lug|U9Rn4ktT*h;+azb?^V*t&Gb-pWg1wXrUV{;KEi@h<*m$ZNhl&WG| zD8;zoFCYW|wtbyFY7VieIf9`;V!G>|80}RaFzvRb$>Y1$W3@pN_Ui^TmMQ=u<@0qNR{EVPwXOh5j zM&H?Q0-;Z`yES!uMVMAkz523Z^Yy-Tk4><=I@VV=P)HT+?weTS_R4&QHmyN_NT7y( zjh-IgZ3qNYQk{jEyHbP#tvpf=Nj8Z{hp#P|(2=~dAH-&WcM=LuXZs8Af7yvUy7jm- zuVB@ISD7#O-}@`5Y2Xt+w-fN>v4W1n`xZXQ7|y01kTrB6>$?4}jHn?K+}r8Qn3KgdXYl+*;>Tnhh8#oNAjoqO^Kcs;w-W5dY3E zPulVH?moilEAOR!iQFb$Jt@<5RmI!YXE7@I{e)-x?E>Tuvd4%!-*#)iviYta*Gy?_ zx_K>-)u`EHQ3fSU<1psCQ9q`9z6TnTx8s6kC>?Y--F(dzgXQ;KUL2y1&rH7pm-}2X z4)=-HN``esTOl^T?(FSxE~lNO%XG%)+8-*woyPZm9P;=~+9(b*BkxEEJzK+>XdFh8 zwIySV$1(ZSQulhS@50Ogr1<1DzpWq<{NhhE`6=FSZm)nX6J4bi^`<60TQI_z6vxqm zE?O~!{#I}JH zgRXfOp*$3kTdiebVQo%Nw)TYcj`OSod6vXr6TyHSH?1Ck&gsrqswv=$S=>yHYj zviT7RL5pl~gD>WW5wB%~*qV+Yy1DZemRDmQ@#D_;(nPU*a({i|vkj-g z6~bUVS##?S^6uOHsKILon@9lTg1@Rj1~8kq;X@5&dieuGu5J;+zJB=^+*7h4%z`7_(l!ly;m8I5fWUYxE$l}b zrM@s_2g7SL3f%ohxTBkgSRVNKYAv0Kb9Mc`Oe?O&gh$Y@h5WbkYGcH4`M@n^m5TULV-gQkpxo3}*>J7k3BKnwK?d(O)XOY419I(aLwA)z6p|n9ssB&Y~RPU2}v5S_q7;-2C;Erw=kdU(=f7k2664 zH`3g#>qp0}s-~M0w}%YpYkmfxceTuaz^>sFDJXB`vWAN(=!(!(_0yxE)~LdsJqaSJ zB^$3VoMijl38x=t**6^Uysm9+S4$Fk6t z!7$(hCJ|RRx(qX|wDA;kA$=;$EkDdd2lKY$7)XD| z2Fxnfdw=_CIJ*Wg?rMcVYno+j3C1C=N*miySdU`{3dSXO<>vTQ2f-r3q+aKOK;xB; zpnf>R%+<0VG`~%=Xv;~hdaJD3GI2J_@p&K5d{nyi$}Ycmu^DgTaTc-)k!iA^{ZU0F z(kNIUlf!3PlhdrzVcAymc>a_h(@|-_qo~T*Pd=G!>CLvU)sV+;9MwVMoj3 zJYFuDM$1oDpj;%x2gYt;qH{u^KkmxT56|> z(JaUSW1s+r;U%k9C(AO&xdwtgA)p$rhJ~u1d`tdo*vY3qE`F>Dso0Q6wBXVHc|E#M zGVMvuXx#J8=wmjTSm=V1hz|^=Vs-BU-}K3(t}eY`@sN8pouEwYu;f%sDaNmKC2;1_ zt9T-y#d`jQ(FE#->lK+}W?e{^947HkyaRnJ`IJ*}cC$?GsZ3L%Y2UMgbllJ>@sMn1 zT1K3d{HkAqEpYz@8MeXCusd6Xk?C(VhZG*lcrxfn>g?i_l~}QatI~}9k^Xph|KOZ? zR>UZm`*P<6H#C5(M4N0WNzTU}Ylljw-P~H=3z$bCkHzQy;eKcX>0F@&{^rvDorbvu zcg|&e`iiemN!sszEWP!o9&OHVG3bsC8BWg6Xyk>})qarIU3_A+GD0X|hS~WwW$hD3|#;(lQ51cU%$npX{Yr6z6|mMG+dXE#6-DKhYPKAy7YKL9 zQkc3jCcjbSk9ih>t8}End}B@=pl=#@J^d9hnRMZY419bnYqk$9IYA-A<21i|{h^1` zrsO9+TW>adU2pi#fX~{$&r1@cBB%!#*%rJ}j?B>8GISwFTO)ZSGf6eiX?PAQ|JiHN z;d5!QC3lY#bqm73K68JW0xCB)W|}<5MCg`IkASx-8r!n={mul)6D-Npucg$jWS#Nl z1+rcv0k}0{>iTeMl{1_AtN5)OQRmG2S+TnPh$uafAPm|XSBPA;>I);efY|thB&^p} zx1h*?)vBuUWq!ZL@f$lIaqr(f`YvQJevR@|)@&dvp+Knd5IzJ!g~pzXc}u>|Jbgtr zMPcFPVzp>?Iq7R@f8l=S>#|*1#sX)*%FSTSOSXd*<3g|KwKgv}YkTx1cx?qYa1PLb zon20Tr_D8uV{D?0g*2N1oe3L!_v58nfkYx!<7@L;QF%=zi&lY((UR&Hg5QCt9-C=j z->((-wVZlix{W&aa-vyI;6=r;6dq|s+}g)7B9V}W>Q*&C2}UE1rm!Qzn|j0 ze6GKa`ri1i+~Zjb?K}C~0|Cduot7@-MnGVyNUKXRJxxrVo20 zRQEX(?|khiG`Ul}fp=L{=46w#wrF-m20Q$`xt!37kNKVjI?>A4gwua~T@k791ZFR} zpJfMn1nf)%TuXfwGxzx2=kEZCPQ{*O`r|1WbuesQb@W(ue?M5~IF5=m7cg{ZZ+G6} z*F@$EBkql@lQAGn#*ANgWVu0gOC%h^%|%}7vlNLZULWDRfI%geR_Y3sckckx(f}H~ zdT%G2=qSsbU$bcOpMhL(qBuVmX#@CY-7mo$gZdGQDWQeoYNVT`qg7rkQ8@d>FJ% zQ<6-mDUD5uk}yokuN_gTO+HYE=;swzE^+w+h^Xxrt4^8Y0%|_ z(BSI>?xC+Ls~!-UpyaSLR;1F@aCU_!AH+KVyzK+j!YXrkbaw`^eQ&0@FRL{|1BFWD$8$FWTT1~yZ%eV3|bi9-lg4l!n4K`cC*~Nxa63o;`y7R!ydtcdK z1{=k51q<$nq28I#$7L|U6qSGxQ1FjmUgS`!0d=x2!swWr)&WVmYwNZlWat(H3}?1? z5Jcj$Qf*o}rRDOz|KcpxB!~u}TVaztynYPNcK600CVn=wF!kfMc$|k7pYQ{y{Q0f=w*^>@>!E7qzn+ zYvs+$A!gzcdj{Mas?d|~hC#CecpPbN92wFN_>C7G3xFS|O)p_#s(v??-g?K}dw3N! z(*RL3)MC7wt_&JM%dV`K`i7k`fR;_=`$2_rZ~j}iaEQsC+bEUHctP_OyHds*Z{e<2hc^cn7KA1%X!wx+4`Cy zD!Pu~w`t;pNG~*6MuRJMs>PW-p=lf!M^&iOJF821Kl8{O(RfoT?HX(Df*Ry&yDzB- z-FkTCoi!XkyZM@z6lNoD1^o!Um`%TIAgF+f!KNDw=E{S!M-BZwO>2 z!;M&I3qO510kS@P-qvY`L0GGeoC!V0N9e#dP>MR0V01#z(}r6^J<3M-wR%b@FDk}D z)zQ;5jg$NRzt#_No;4mdr`BO`1^^p~#n&Sf;EhyTs|k8EB&YOV8*mnJNZx~DeU4yF zk~Bro0gpzMx{o$EntiW-Np@H<5j_bUC^6~&T(civm9^*|xV~is1VYBEEA?e5AtRgV zjAtRJ_5~E5A3XVWBr>TK_+aBSoOF~Xyb_7QT-s#0Y*@zrsogE2s;dUleaw8)W%xA5aKchfgrG{mjEQF^PX=#rGjgx$o+ z_dZQ&@wfZfH7OmgYK%$@>njmvBp8mk&vlumX8hl4#vI3+;FZ_k3pgX^MaQ;*#9fe3 zdWqoHL+~_qUmfhVe$N&6`In7K4o)&3MQ_3ZjKYKsbc!#qJPafh(pd(4#XRMEtUGt6 zb}Vx{xZhAJh$3=b)xp41&PivWbiD0-Av|?z_JWSy`SEYpNUxEXUi*CM27L3^z1CIw z?e~lvApE5!56>yTbH{L*V1&288KX_mk}g`Su)B|3QJ+kr+|Sz$sVVpGE_f{<`-~oBi;!9`CUF)xr4f7L3M#o_NCq$@~t0HI`?|Y3BDixY=!J zTpzIBn`{Bc8T3gYErkO;s{LxYAtaJPj^9y{SM{`9TW!BJt+5B}AQM(=1eS&IbUOAz zj5W#1wUY)q{M_s_RbIHrw1B%~fnGlB&`!=#+(~!&MjVYi)5jWn=y_b(Qx9cXMN?;S-ed$ zS!duiO^wL(`n-aB>%yH|4PHCBGCrYy*@ps7)3g<9qBXs5el_rcTANzosxIzBaGNbK zH&Ebrmy`U8BaR#tSx83kpIB<<`jf!*o!s}G^v zXk|KF`zDmWC?fC6JvUOC6J6A_%R0l0%HoAqSGFQQjd}M@hQvAQQ`Sa&jxE-AK-o{WvqFcuX{80H`ft>G+PHnZ{z*^#`l^li_<8rVk4@}@KFM9 zq0OG!UKx4M|4=RWkI6fE-^iyt;AB(1_$f2FskF~}$~y!~4Gr!fmRky+idQ@r8hhX+ zP$a3?y5CYRKe-5zF`8zr$1sP3k4LcFXJLF;!*sMI)8c-{9LbA&E`+}z!GF#cvai^$LGe=+=4*2f*A{9D%K)n!w=GP&=`XC zCuif1>Kiqkl$Y98ZgpqSvf35Y%%xZ=fO#w#-6W(F45scAq<0?g$WuV20hGfsK-k#3 z@I7MzrCH=mW~~4Ne_)Pi*6l}TERy(Mdd(!s{C(b?pZd2iNoa;JyD$uc+VsxGEPv&v z{{j^l7%Ltu=tw3fGddEi0J){>Is^4~sQIcpyD^k$a`M11b_)M^nWhOMpUrpi@$da- zc(mXuvexW5L`!*4TR;Q~7VXryZp_TUFw7mt0Yy#J+)dM~3Dc=cYWpR5HO94Kk=nd` z#gVG#j8_G}abp4`&j*Rv@P(>B1Wa}2$6vGj3t=t4_)&NuQX9eej6d;SbT2w+kipP_ z?p@U{LhGO8uLUyHWyv@m&nq)A;NcjIkbcCgSHuiZKYl?vaZ0%XK96$l>{cXXkzGOv zT22R8WTp3{u{Ay&Bq|L~Nto~FJJiJiU(BV-aFPY}9T|7-WF#E3#0#Y%2dwH%NT`IG zXm2F5x&W|{aKyD12Bk6>f z^h&LX)|rR(+jI*}pbYrQ!+|rL&8*?yzLKVy-aEh}ji|z*%dhD|YGN!;Zy{Zpmz<~Z)HW^o|lT5VM zL`I%_GTw8y`6KSW=GuAF(41^qdC@ji6#=N&yiwaQA{Sj7AmF8Qks6a$mYBKo<468g z*kDqI6Qp+}8Rc=#2>Cr9tW7)^hexylH5X7$Hl(;l)^49zNZ`=#%TAtvo8)czJ_YjW z&4q$Ta%ZLvuJ-C{m#rLlMFR#@`2t;q--ndC{Kg);`o7HfoojsWJ9@oVYr_SDr<(M= z!wov3QKVv2bDt}LWmSc{EkyRaUDx-w%ZDuupwej=-Y`JijkydP`~irdaXm$D`FL^& zZ<6!w+tM0GK*c3fNZfsVzWS(U;>Wvlp`G<&s+j>b3C2x24V&6kYgMhP{jd%H-`^vxx-$mfo9y9WAGrP;Iir5B3N2JHv!%9} zTfd}PAAr2M72zE&McucWTPd(9=Ud`;j0mtLx}a;$zA4>pSS zeJ;-#yQQB8Y|``Mtf4b+xHXf#ynHk|x$>q7wIDc9H^mZ<2t$hjU<9iAVGO&mTV_To zwZDxrq;*<(ZiTS;2naq|F=O`^gZ!43*|vlVGBl=;xbI6u{-oRMNvnJ-8zfJ$a>&6zGxs`x8c)n?r^0W zhd6h0R=t`2p6MHLu<<@?#m453($%_`29~8zTK|r{Y=B|dViSsC!Flt6H8q|!9$*8o zudVS`f-cs9Ewayq!lR6175KN(aoxQJQPA|URE3ZxJ!&Jf3GlE0+Oe1wdg&f)yT_dG z$VhL1hc*mf&gwvaUoeOF=kqU|zm0V^#Wv>cy+ z|9s%7(Sf0)6P6|S;C(>dH5)GK5CHZ}3i!Zl5D(DQ;}$M8Jm>wP)CR1n2<$Chc>Fi) z3Ot7TPuIQ70RBox`tH>3SS5h7_`rTPif4^TVfdw46qNu6^fZC^$-Qob4hl;U0TNJ8iQ78Pw44a6O(8p~y7JHkS$9{|Ix9x=eH0-Sr<<>FCK|ch4AV$bnyh z$Lr?X;1Dw8KNQ%3CktZW)$EiYf=5gcnHQKi&ZFb&f!Xd~-wxQ{KkcIoL9YAQCCA#& zOwz=@zaI|}uM8Okf%3|c!U5Rq!Sw`B-0>I8zy7S@X-2Ds5x^$U5gt?ifw-;vAgMcI z3;2Z`U~vU-1-()oy@V2?N+(r@SO? z0C~~{HKqWPC4%PJ6G|2Z&-+m4Vo0ms`5dZ$+gV)h*Y~tm^Fg|GVQ%XP9Mnu#KaN4F zYPHB5MtWIXl!GZd$gx><=PbE~x?R!)yjhoa&bAF|gaa$$_f7Bs2YC4inXRzfN|Q0h zR>P&=(^H%&)}dEe`<;=#ju?RYf}^{%?Ik>4)?g{%x({G|Od(X=JcC^mVJsOkH${qO zuv170c+U?$@2OyIPcwH5GR1}Vd6MaOiU8k#9X!}4>AC$d zWTcq8;|2~V3$>ZRJ@}xL?-T>33HTwofM?$P?!7e15zg~XIQJ7gzrH__>h?Yo?zO{u z&D5vk_$;+136Eks5<}Ynr;nqExU$s%n>U&~#n;51&SS_weWnV+SNeUHOE{;_uX;eE zXrrJ5z>+K-+VS7d#h;SmwQ$hSTs^=kyKJJB(}y&o%PS!puIOoEJ*jUC`aFyJiBNY@ zxPp3D3C5O{!}7Q&L4hx<+;7KS0Cyl^E{M+99k(I4^_P)tFjgFE9QCO+2D`b_pY>;r za_!wF^2O2PNaDjEwMhql8CV`Kfs4S6M>sSw zcx->#B@nnG`V(FIcvuqbL$ceXH3ZmKL{~8N{fb-P8lMo4de)EUsit+fxCU;L$|X@C zl4Fz)8k;W!#i$7PvAIuXJ@8)Ayf!z&@%2b(@1&()C1uH`?Eh+$Hw=!gopm z{N8K?;2ijW^dVp;KTWL@2ilyseMS2cKnBvsl@RjU2jRbO5Q4!q-f+=WXRKAV&j0C; z$B7(C4gmU}eqIoTgH9*v%=xt}nAJGb{IdgI^7;M7K|??^7i+GR1zWpb0K6Bw;eDj{ zn?gh{*AqZHJj_=_F`Oe=WXTvKVq9#Jr}`xp?-LUsgfWN%bE>yC8J_z^FOOMX?6len z>^fUggLsnWm!iTLs2w#rfMKEj!FEW1hg_(4_jU334o#A1R%<8^_b&$v`|Fbm))iCw z|2Pk14|M$F3(`LFTn{kX<`FbYKh4_`YpQf_-7hV=jDWo4$p19qkQ;b&bnmkUT9}i< zfL!SWKfGWaX~4*8Bej8GtIC;i1a$$mmYEdbs1NXSLPH!8RxJ!U-Gv(Fanm&23xv(v z)Cm+8V9wu|Mv}bvcX%x+hm;^VLe3F^us!Nu!c@5mXADy59_IrvtFr(PZBbiM=>Sg} zN4T%MT?gj$KBhfIUEtPtgd_E;ERzC!W^$CSSa%J@pK~94&<^H5=)APGb-4pLu4C*< zynZ9eeg{(pcv0SY047f^xMB1g`c@=j#vjX}KE;)E6?`!Mr1H{1QdWWzUFqaD6THfN zV|NSyP2Vi}Pg=7+0ml6BA9C60DbpI>VT*wR>%-n)s*K?`IL9TkX8=#t_Wikgdujpp z2C(kQz*ZT6h0hnj0}#j7w;F19P*>M=>Y+D({U935o3z6(ubY}GPJp)+fG^AL344u) zE~rWh=9BhlE>;5Olb%aYn!$zgg^etu4P~eH2W!2jiL(}zyOz~&U9X-)WPm?hut7Fh z9=!|{xbr_ueKi+d!y2_DKGVB`>i7@VByOE?{`Fgs=8Q}IbAhQqU_by<**58XrSEkc z<^Dq8^`qdqkqDJLio^i|a&WA}E=n=KjnE9A1K_n?skYZwCpj)Js8R>t05}tHKNMge zAi{#Z?#M9y@{|?=51yTH|B@p~#xNcv4j>#j0Et--M}0VHR>8dGNM5J(u=TF>7^nG= z1zDEmHhqXCT6%=X-r;Jo6(6RsdP-^c!j}KTxNnP{3|m>sS0?~XCs&62K#q500A|?$=DHC&Nsp= z9d0r$~#n-IF}fj|&7sr2$^YFyxvZU~B+1rDuPy`<>x;#ufmv2qxH+ zJN3Y_>mB+8C<1$(TY&wlkwxh5dTBDUoXKQKOxeISkqol(ZiJ_? z>;5*&=6OzmH4Lqp?k0*L^BRqB>k-gYi=B?B+|su#PB*EZ+Y+s|%Ds+*PaK+hw-o|N zKNb&OY_BJ3+OYB5}MQg~d79TIC zNLp|JC@l)s(!h->iVS-Zj}MQc@7@bV!3P+Fel18h5so;Vm}$huh(9+8xb-Afw*h#+ z6nt5iAevzUqqbO!9xlFV_qtSEo0!jkzbJ9uQ8=R3}fiqPsHwmnl16FPAfp9L~Qbip5jiA za5@e?&ri4ZE_vhaumXj0mYLYZaZ&&qs8sE^33wklrQT5MPrWsw1mzMjej4O?qb6rQ z9|iEgRB(Qh#%<5c44Bfp`$%xh-n}kKs=_@iGBP~R>s#w4Vof6oPXo@kmNeyn`mGSA zRp7kr|2FU8NTjQ*b22aaO$hKG@#crLvb=9~UH^yw@UQ>;KaWA)cxQ^ zq*y_yXX=S|68rm0y2Y3$zy9a%H3$AoPp(E# zR)L=_#iH?yE$|cHfd9COr%f(Xe875t#;{ks$fw}BA7!+Mw9o0B#clklk$GF+@2xSn zZQ1spcQQ_DI(^jm)2H+O^U=#Fe9m#dJU?msiATJ{q$4~D|lA-{51_)YnfB6O6w8}^NCLMgu=t9jn z|9-;7*7|#&?fqB%PBx>WHfcDgj$&C94ez!lKk85#h!|oz{~p$jB?A}(bt;DRd6^W2 z8J0i97)he&UI4yQti2*CJ{CDtv0)hlc`R~Z;%ou<0pMvZes=tmKA!`v{awIp1AJ#~ z=FFTc+8s}2zt4_@5ydm=6U%>t@70&#QQ$S<4jKHDhe3*ef2y?hp1_|Pql*cv6{)UK z#4>JJIHVPiy~$V7=c7&(Q5GxADVBV#+i>IjBB}rmhKbW-y6Qw-(EF6rbM1- z#~0-T1FTIKM$!~1jTIzMu@*$YqmTNfq1G^hzbW5>QKx6F+W@k78-TRq&bh+GJu>G8 z?#1X$^UMeFF+Zi5`7W15+G~V4z|Tf?3iUy!)#-A7Hu0Nw0s*%JY))VDwH8W8ju}dX zo|39WpI6n3q-40ZO7=c9QH>M?IJK8_d<&J(;*Tg9;MTB8y1tYnO`!zqDMZVDhKJYR zG8NWIb^!0Zndd~9H#Sv<+w5S4s8ja2Y9uOzbZ=SkelnHYpPvff?u~bU*&7i&%V)^z z$w76K-{vz3-Kn?peLgL_HDIbQK<6gP5;b;aCG^rOZ$A2n6sY|jLB@F2iMK40IcxuZvgD_qNt=ax{1~v z@97SOk(l|xGwv*$0|4x+)ZjCJttSn6;hh$sBZ&TGjU8$OgfoBNnkiM$gFmsmC~7i~xPl=ItO~KU za&Scl`2b-8eu%PQt@!)0d?@wgOW*lD7Z?yD@OhZHy#}Vo9dlxg2J+zs;-V%?`}7$# zurEWCY0eR(8dg`H%iG&m^Pzw^%zt8J^P)2+T_4Wi`rOMgLBMAR4a>JX@hT<-yz7Sq z+yaP(cB|-(X$`<($^_=^H)z@H3>S_oEEhtPS~~aRJl# zI+`5XOxXeBu0!x|G+Z=g3rb@LIY4WP-DJeq%t`y=oz53va-XlL(FV)Otugz)6SpAMM*%vaq3@R_Lm>0)6nKmvQ$-eK>g>5d&aVRL8- z&YMs~@IG0e@%tNQZUR^!FeIb(df-{!lmqQSfB=qevtho0XFVAbR8_dc1$zQ#Ts8A* z+wK~@y?)*2L!&zxSCyBkUgC7#5=97XrmHM)`3+q@)!|kru*XMuARH$c;KMF&3N5{hNpb_*uQ{#x zUVp&4u#sl+{dv6JbQv}`$yb&i>4!uQen+c`Ce)^ctqF)2O~vxa2~g>4Os=cP7*S%1 z@~c73*Hc|z6YN>-{xm__+Jb$3(a3$nZwM>b1?evM{LgwqKMnP%5*L7_Y}Tbe>AC}4 z^eaDcn9XR?3Fb2(r;nZi6vr0(7wl0~?|U3+tLjJ0YWdUQ+!Ujxb0n+bYublTbJZK1 zZpxsK{-G%1;4|?O-oT)oznajdkAN&f9T4Dt@hG0(>AtWivVBvYm?kDJ$BsH(bU4vF z)J9#J5xux+PXwt==d$3u!2pUd>bs08AT9?H2#SW}K)z|q!2TPv zr5sq-Zc@!xU_!u<)%K`lc>OlP<~z*sULxe%0@ffi+}9MGPx4l+KsXh}>C!J#aSgGc z=ntn5gdnXan>!ZZZj24$7kjD*)BW+_8xJn5hcPWeS#eCCVPnf)*z8>8_kO6}{QIq1 z2-e%8#~dCkbEMx(Ck1fHcQmM~logrTNCzw@g$vd{nlL-S2nWSD<`XVqV^VD@ z=AbU|px%$%)!&alUH>9BJ1c_N!Psfcp64S^T)p1m;e^<$ITH3lfB`1URL15$Y?a*-h0pylQXI#jiVxMtTWq#~ zv6)?Ru|3P5j3$f2-=5Hb&*XNf1-ra2QMjauvnv;@2wRVAZb>=`Z1<7c!xT`6USrwn z8L;oG7hICDH8>_$6p-t+?%WuBE&=Me6971#WiaB|kt*Ptlpo3h*m|#o3L!vfPSEkB z16X^Kw)nMaFz1t?dV3V@o_N55*YoCrSSoPa(|r5_tKs7b13||qbqWOrkGE$kz}ydG z`@m8JPm*fNI$-X`)_|XISvD*@{=fT8q#-R!VhoJ&o{WE_)`vw1x`;#k0 z6*x0ig<1+l1g>D~$j`dp?jIN9bx-ubVt$B?RrSwZLOJS*Jofxoe7F$Jtyo{9&&PNy z+)t)au6$2r``Jj>C{L}sy<1s;&ja|eB&(U>GkZcE_IA+tqO_<43q0xT$^9-oYkUvp z+m4dHtm-ATy0XRYC<@Y1p*&I+7uc4iIshz7m4LsiYj=BeDB%TL6-m3iK8Gmet+j2h>)lU$!VnRs_ z1L|(9$2gMAY-KdK{#Og*oC^526~DeuS8v;spK;fk1jhvLL!X*jJ$~(aPc*64RW8wd=CJnjZbZ55E)-FOL=yp7!n5DWtIinIZ-LqX zgvAU41cn9L-4WA6H8RMoYF+c|a2)4we5($&`ce6Z!DQ;kkdJc$D6a9K5X3CWm9F`c z%x(nY`!IQ{BCmm-Z(8TiEyVKF=D6+w!Zpr>pQGUD75s_{J#V{;4bK6)Y7iI|g-O<` zJ4~z^qTuN#f|X>RDBvDu1Loh)5;2K)<_;Z?{swTLOU(V2j^~`_{QX!aBh)23zo&J% zJrk@X+xYiSWV~;W@MdHW^i_dp>f3Y)Hpsw3jXkD-*d7C0v15g(9pCc=5sgtk-@4*w@RAn`!k#j>`ffBa0v~anAj?!qWxGd*3YC!$b-BE$r#& zD+|G335lks2)PSvlvfrAetjkUWe-=4yyR6%Mz&UJ#olEkJwa_&KW*>W0H#WZ$<9xnm4ibqph0Kp1X*l2Q7DHQX<& z9q&~0LWBpQ?DrlZzoj$Vhc&Dw-Ci0_b6g&_1SA@k9O3?RTW0p$B;a$uvUbd=S9#CJ zo3;Qv_6$fTc6-S*$Qr(#bx9ZI;LiuPU9vFon+kE~E_u~aRVBjlur@8ik)_`USWA0H z-R@92o`kslp?vN;>`O!Rk*4zX$!2v{I74fkmZ@vVffn{&4x7)F5$_`5B1&LtZc73!X!REgPDAA9i?eV%TT2bi^-PrMD#u*PG{9=lb$FN?SYR z)`AbLc>wEt*R{|DF2Xm*A;4K1>UyW>AD%>0d}iO@QVh(eEbcQ}9B}@L7Xp9WX8>N4 z388k%@lB+=z0(M<27Vnk^iMq?l$`e#Y?N4nJ>KK^reVI@8*Z)9L7jk`|46Qanwf*y zfJ|*uL_?4cQB0tLNEXkt#41zbQFyLz!)HqB*R-8^qKdB zD;??uF8O1-NSYDNw{X7!i?A4+bC%qZM~qW|;kBV^JOn_Zw)3V3D?x#0JE7ga@7$86 zU#YL$-@s|cEPfXJUi&a-UCStf2+r^48h2Q?ly(8|iDbjgN<6u*3HCjm4}w_Bz&C6% zY*hw$-Ma~v9AK>NxfIzV1IEpRK*Jh<59~RHjqphv;}9C z1vpb5Fx!(uf6LibQ$G&(FVBcJ)$nyhNP); z&TZbv*Q$Yg-_Fc+z`EPE`M{=meF0pA#vOQ8)gLB;m_n7nLN5(~38^+9cj5*2_t$;o zdW0Xv`ukA?tf#e5iw4Jv71=mOk2=ikld@jm%D)RXg!}{gE|@FvNTvdgL4>LFWc^l- zE(q|x09LQ6sRo>W5ujcJYAvv56W})D9%N6Kr4DA#^c^@}6|l zuuKq7X|g@Y_j@;yI?Q=}NtUZ%{H#A^tnTvX@rkWVITSyP8>su(^a^k&VB{hQd@#mh zuNvkij|JdPRMs>C`Kznub@hgUJ6XZ=j#?_0bZ5K}cVSsZTEIH5-=N#ggjg4eTmhCM zQQn4u%>cKDCbca;FmHYF5COC#9gcR>Dd2*AJ1!?BMT%X1ZMgVg}@ zF=W8iWGi>aJ%Q`8C%`Ww@{c`{$`}Ecsq&*h==^2RnDngN@f(M8Wcm7mxOwSnf+v8z zcN#j0a>}!EMoR;HvE0-A`*h?tefoq)7FVu_Vbo-g_wD`|J{^6K7hysXlMeHMQA_7- z{CCfd_l*;%iI2?@K}~`fw%?w2h6Lkselty$c<1Tz%_|l`e=@L;F(hvgVT|%WKYb;5e_TBGLAV^^>VG;rfbNr4 zvh{9|f8;`{?BYV44*UPw3Po8RV=8yTU@oR62n0P~A04%PS(p39<8}D)1=g*i7zRQA z`Z!^2_#Je(9e410ayOo%=ap}w0x$`yKrOR$5Nn~^mq!2}x&a?Yf&dD9LFr00rr%&x zud`vk8=<37PH=nkz-OK)07-=DC6;z(#keN1*xu{R>V$p2f%tTZ6sHM)`~v!7SOa$} z+4m1l6lBP8C9p8m3);f04%ZFNBH0UpsjCOTw+pO!Yb1?SjVL8Hu+7SB*a_GN0T59i zC7UiZP^VKTtsyCAVmaDPCT%D!2ES7}IP(U(AE^b~4h8*!rwP6F=MTVC0@{6(K)u0V z55qrvIlib|IV*R;trJ%!)rQ=~`m?WBu@Mu%?$^+jKe-<+nz11XfErcdCNur?(3-() z4`bZBh+s6WMC zxBfh3i%6Zp9e^V#rAsK4VnG`8%*6Wr<}LFA62ELq&?8s_4( z0mRf2pgaICCJb=lxkzb)QG5n*K^083+>7yp6B5g{Ml;)q->Y zc|RIk#_o^w>AxWcdv>%H{N{0!*K|0?3o8slW#a{(96v7<(uf_bT;|b^iVx;mFP}6dUqO7Pf9@*64O+_i-Gv!zVbBl;MX43KcvZmMVAHY*)sO40H!w{Up9vI zEMJ@csr@Ma3_lb8l)>7~f8FmQPrW@9DC?q1P4<7VhbHVM#$jw=-gpL6IDlKzpN&0E z<-iI#i6p~AoNi(#oZ%{ng>GhXFPJ_RrRYy*!O`rk09750PI{~ z+xSE*j}Pjy$A{s9u!z72>bJPsB?x|0?HXbQfyy2doj$sQ3b-4Fb^)moHgHN z#C8WEi*RmQ%1L7ZUBE2DIsZ^t>pBUD;Y08N_xU&1t_w2$F$qvh#Xr{P zQGxa3xIga60UzsGSIV&p%m>+N+P?_Sy`A}3!unCmUT!P3^W?%9L3y>9K8q=Bq$X

&C8%fJc+NK}!GJ~Qt8gBA&|`w~ynem{|JxcpYH`~ zuHQ`b$$!p^ClBIa3c2o&upaIsgSBc6y-Uywgsq}l_gGwv9y8)>J*Q)O#HF5AgMKH~5J_$orm+Zcc;yFH(n zypfCf#l>V3=kz-A64UhoY2VC-8EFXsNF^T9_|k`sptr9^+=gDH_{#0RfYQ}{C=Ab2 zG{H(uk63yId(YgLVv+bZNv$w6`|zSrV!boB)}M6|_RSIFZ~Ti25V(#N(bt{!O+$)I zCJMUx^}q(e)o&f(6_8&|LSp(Z#D*>Px#U_o7h4RsSp_iDgXeMqxW;3e3>@owF!Jf% zbT%4qQ8C6*FBRy+h?wd4R9RyTq>`B4R%^nJeaeD_dnNhh>O4_e94Zztv>y;vGBAGN zIInfP55YWYM=!PapyNL8LVQzNSq{bZVid0$B0G92b^-G0!1)h)0x(`v8Gd^-ksKwlQs2iV zo_Wi(jK0~g^9m||nc>Wq6wPu!7TqC;j9i0!u-L5uR@+&SO&Q~x8s0@?_mZUu9{ zcLTr;$=!^2=Yv&PC40pCL|-VX%Fo}PVLSuOI|4jV`&}>ytv~(Vf9H#JzvktQ0rapJ z-zV!ok(EwNsC z`i_>lKy$1>J{O@qSYZga-`pFo1GqmI+zDRpcGR>VO?aa9=JU!qSB+cm36>~=oNZ>e zjDQ1=Rj3&_PSPIuU!{NOUi!{pxcY2po*v2Z!-5H1WvP@{Y^m4YU@fVsB&{CgN3z|% zdvYCQs*iMnS_PQX*@W1kX4K|J&jr2na4$%WQoujfk*$Q(zG}os%8TIy7&@G)cGYIN z*fMdNc8y;1ac%mhKR@6;ccJYsH)1ON>>2?);ww$HCn8e*S!bg%K(2<%AR*k;VgMk> z1o=+rW_`xX#^Klmzo|F1iv3(GC>Fn@`J@TZ|`sz3{ zehE>j^x%6C2>SM^dhVJdcke0SyAd*ImND}YR7_bz}K4D$!9IfEfC&GRM@04i!S zv_2TZwZmcmIEUoZ;_o2+j>;Z3clB{-BhHsjhf9$r!{+}W0Q2;9W^!{phUEaHycs2u zO_|mutWgSrp;Eq&SF?GT^A3qc@m=56j04jW+kc0*`lT@MvTSNzXQkRGI=JJ&m@gQ6 zPqDc&N*_vy;{}vqyMtS{N2ObDkdGiAbK!ipXLEWr(;Wv%@gc6^*$4m?3uENGdBW9& zW{Fb^?jajIjF}^K^Ylir1aQmZhERC}(1!b$&;s$@Y(DZ8+v+?K90L89Ps95ZrSV(v z`zN-(DSmo9q~c}Pf^A8aTb6I0arK!1lNGRtz9NgOAH7fBat|yQ0{i>t{kK|hp3XaF zH(M{(+=|kuetK*h4H}?+@1)jS-H*ILZm^p%j+D{*2aj47QQDrJ8Mz_DYN<^^lq3^_- zS{{)&M*Y`Cs=TGC*m`NlSZAg0h&IiL*pr;$^Nm{bzaM4C-XZr6pKsQ_e1*J4ut#ZJ z15G9n4dsgIWdx#i<5eR59se%g#Ns`KFL_t514vlwqGJV011z!?%hM4tzPd@5{Zyj- zfp5R1a4~et8tSu zIOiI^K^kdzj&Y5E_ajD(FxTYrJ|z_JU0h}nMP1%lqzrn@actF-< zK5w2RtZbYu(SC`%l0KSef)r|_>nt94qHvIr1=t{JR>xVGTx*O20zdn5?#n}vg}aw^ zR+LM467IGwmIjkm%tSsy>2<}lfV@uo$%aJoF#TcudT_5xP0X8YY5c{f_ESsF{#BEk z>&s8NJ5>=$;@>q3sUuO(>qZy&-B0`wtOk&qk7hf#7+`s!lCvp$SN`R`-G;53wa)iH zL74UWvVovSS$KsT9qRx3^wmhJKv<8RBCj$HxxYR05mOIh-PpU?f2OG=s|vLi!oDc{ z_POF$k41(jXLU4!Cgx-PJ(n^Lyd4Xg>f-3GZ~!CXHo(d6E;^j&hE}2!;br|cqxPgG zmy0av%LrZu_qTHmz?iyg`ecflL-^3^^Ut2kHl3WKH(&bbM1X4#0OV8f<6)GbZxQlM zmvCvWdl>ufG zVf~9van+j=+#sqiHURm#9C-xpi3L+|8FOW3aSo+e+^STj6T9p&a`x-2?ePI8=jr1yiKC??)Ix?P9~@@6t-0 z)6JgQuaLYFfPHoOg~vLDfa&*&NLicdu05}>K<23KJ2q?vaIih(pTzZH9m$Vp0a|md zj^BO>IPWO|xub0XK&gXAIe%`KqAj*yD`3qGtaqgzSJ!=MQ#2Ag)TtJ(mfu6|Rl;R> z=Vi#kp}_qqF2_j{DoLWOynM)~APC_4VXA5c=*b{}Tze;j5f?h^-H`GDzyX)@0IMj% zPj{C-Gu;IDWPkj8C15=nccodD&|wWs%0q4RwxXTvBlpa;je}gQAJ&#cey|{Yvo7N( ze3Kd7KP2f-qOpeYy<1RQ9cGl$`4-@rTk{TVd-{9--WT@ImexxeYFMiUkVDds8LTDu zN1GSmgs|HM?g73Yw{W=X8mwdACI-od_3yoUP4#m}t)U?U*T2Kx-kkY-&?wry675)~ z(q^HisQ(%IvE?@VtK8t0RK={L3H4w)dBDfn$IDHwgupoM_NFzBUydKt1&E5d@NkoK zy+7?6P4AcLbtx9Zf62czXsL!g10)ITk2^5;R=N78^XS3!immKK@jP;Vj8a6n=a zs8gpd4d#kLPJW%+8yRnHHe)kj8jSL~p{_^LV*lJx*Csca`RK>Kb|CMjATNHIBfUTF z9~vq2pC$$D3x!_C+!J~W`%owAr)A9D9h)H2>#sjylc5fUIr*C7H%EKugAH(%k8_Wv z!K`u?bMY(_w3{CSSVAq6_$YrR9>4GN|Mi~Gl5$29(vJTIP(4Cl!Av`kd$O%^_u_kw zoL;PTaxN_S*FJ<9&uQDC3I*@d0-L?+Y`sf0Q+wKfubF46-D|&df&@ZHd|n7aVQKOz zqm8i->*?-axf3EfMI^OAK1M1#U>1_UJn0VqKh@b^w~b)K>-rp7jfPr_vlZjh(;znM z%aJ+={4dHuHf_g;E6Y5^P$%4v^d;<@!M|dBsR^PijTo&l!n2T==c)PGrjDsqm*CmK9g7Dy& zY=?G-q)u^~W@~}1Rr)O=-1eZpYWJIKqw9?*{N!_C8ww23qVR9(vEbIy0d-e? z`zzINYYCRK3I^2S0a<5$YCqj|<)=3Y;?JXo`h74@yyJ~V9LMFdtSk}e6**Jc>KSC$ zwsLgRSrvg8evd-}6}Y#*%Zxp1WT|0d!(*sxvr@s8W>tDBEof8 z?!q}n0JFA$UgGw`r%nJUu$8Z5X4#(r%*>0Yz`9ZR-0u{Ld;*J58t-S4vm<6IGC0|! zasXpS99af=yrwc8#^3t$yng2Qa(N*|Yo4=Z?3v?VhBWSa+B4 zl;2Q8DAfz4Q1oC3yAK2RztIg`E43G{KQ1v_XccA4kD1BFySDXw5Jm=U)V_fH{@Isj zq(aS=tN?=?_yacH<`1JvvFRJbzJ6J47+2qH2GZf{8;`_Rn4|sD?W?=FsGMDKq#{>< zq05cU$hRo?`XDBuw+Zh{9Q&__EvH014NTF!$p!6s>|IN0y9+*8Tt>chBpyuj{2>FW zj!$PP9i};TCqgJy>a|WmdHH=}e9b!oGJlrDAT3E(zdf&-gUV>&aw8>*uEWNnv@KD2 zGN`AvOWq^nT7L{WfI~87Q&Mz`*6IK;n6iTP;r1M=CY_tO-sZ77B8qanED_$!kNxAz z`c*+jKE0>R#+=9P33490yiQHkqnG;B^~W5eNh1r`LxddVEa*j(&-2P3ZO=xzR;z%& zHFj>)^tg@2D<$N2GuLLzjRxV)<<|~B$CGa^9>Y!&mmOEp#-8<1zJug(dCd3Xb~z1@ zZ~(T9F1xuBg$_7t6yTS8%Th@Z5`S=XqY5vqExBi1*}w2Ax1a;0=_7@lBfh#;&FUio;w7 z6FosF-WsGLWSH2OBEQQzF^TlVFa4c#Z$>i}fUiEMhHQz1yf{T@vks2m!rO$tw%&Sw z=(-`9W^mj^3ealn`gf`xuJ_DUtWcuW<8M^E%RPBS#b%nLqPrK#J#lM&~dQ>LH=V8Wmf<5Pd?+onO%lkZ9dy&&~* z+??8Yw=F?Yaxk_JoJ79LCFfS#ZY{BR+QOqJ#5;>9V!gvF2J67wN01hBa zqVY3&@89L)q-3yyG^KXL2&cq**IyhRPvKaG!Cgp9Z<2b$X9-ooAGl-$7(;^vOxd<8 z_;!%E`P%F6Z{i-5A>miz2ydZ(*Wl-1L&@#pya!k(TKa<_LB$SEe>c16%BL`~l*DZ& zcE2fp6e{34rv?jX&3AndEr*-SxW>&CFA5?P&q7|7UEQ993=n978sp`J`w}MEDvepc z(2W*K`P_4`O{_qED*9) zW4C+;WLYbcKxmAO?C4vV+rdCVE&Y%kB`eso^y(QPmtT7%Ry}{iCa1H?>{f^yIsLQ-uY5u9?d*=qz@N+kkYoTnMC+5pV9J7n>>sYawke)UATj zuDAC)=DQ}nav+-Ym}anDFbjW|XQ{tse$@PcEb1 zREgfN@_{2rUm%_Ut~;w-B9e%PO@sRy^RH@-@XnAhz_}|ZT#Vg=!C$gpuY9;a&E=_U zBf`uSUgFW`dR%LMt2bOmI~Ly^)^i{Adq}Md>#!p05aLVJgVmDBYl0s zwU2HQqbS#u{w`R(bSrw(jFxk)v2c2qLJMC1AXija?=HX{DFN{H8wf!1w$Yg;z{2-8%}Dq5@Kdq% zIXR7|dGb_Gwf)L}g>a%*oCzcK7P`|n zK;*_RX+}1^by7LZ{zQCTB%Ch+%K|NRkA1VEQ<(gtS zAm4u{07sV2Q~xMSq(D+RKMgJ%Ey;l|!fuZf0~Dc~{=TfVuA|CGKU zcc3?@#k-TVNs@cJ1U4pKdh8WmZYTY`LGATIyIsZxEc4TB0IjfPr}aF)H%Y6i#WYSd zNmzqwZ~&DSxWc_DBDF4&-3tQe?sl#_qZLkokjki$| zm?e$s-rPl{=3f+5hzXS!d{ZN+KQN`AmDY~A_GAb0Bt^Q!R$-vI3@oXSYH*tX&LPvu zEAid|%v}z<>ay!k%OW5imEnf4flQDDC|f~oGFIEA-i~bfrE3P+?hlu-#Rx+?z=dvI zMK`HB7-reB3^5bey?;&mA3Y4z#Z-Y&)}R*3U>t&OAzQNA`&jpiE^~@c9}-5BxdfBz z=f5#T@GASNC`QB4~C<5caeY2`f zF?2|tq@iDo)h@h>y z5#uA}sj1Ho5h>v>M1)27y>>K}j#SShycKbIzTtF#Q%0&0NAg8zMNzNX`uyy{)H@6l zOab~eSaFI^%Z$9Y8BhK#W#8M=$Tw4j&p&j%JYNsy850&Oj>IYFpO-m0KmSA*O!=1C zXVk;jU2?2mY#m0z>DqrkD{2luOnt1_E56bIao$Jk(R8HopotHW3_K$qH)oSe>kFba zsizMTBUF@+&B1Lu9as1KH3mX$Mfi|zvs68%l#?}j)!k~ziM?pKuPf)XynGZP?CyLRs#gu5(ahj3|!ycwT;b z@SF8I<9MupShjg9%R64?hL~TGxg`(xIr6u?@Qa~7)4<`)D*dqYq3$&Jjj#_45M~BI z*VhofN|4=+8o0jm%Xpeyks})%uNoVj+BC3 zMQ>6^T$;cleK%Yo!xVhk`So}}UCI%cS?FH`<+&;k+L^fG-hy@2BMfmgMW_X=Dq3 z+I$h77cCAg2`CUbc2E@ZgCNN5F+a6^&(^>Qms|o7DC$)6BGQkz(8QrGgOo1WLrpGOCBp3Go({Xyi`nRU)vxumynWhOJQ&!TQ_fpVDFqk6u-4Nqht zMvL>q$WcRcKiaWfrgx&$92nzXHul<{JO%vRe+v(-B~PLT5KRzf%L)iY{HTgzh7a#i z-f;wb=5q`^Y{VU^?~bsiwqdW$@sS5S9d!a~wg2=epbk6Y|N4(>3UOsTUE&H%VfD!F zGW5st?N94ngo{*wND75Ol$y%h-4MhZ7+UWF0{LC;F-e1UwVD9^lM}6Q?^^>R@s#Q6 zv1emsqIz8)9F%Ff9P1#rxC8KR4CckM`yDngm+NRI1h^4L6613V>&ak!@FRa?0drec z_I;!8pBSyb9D1wxGXFZnI6QcC2AmV*-6*BkP>iER5t&85|6(A*@d#Mr)ILgYhD|pd z{W^E_-CO@w&!)CN-^W!S^Uz%NjinG+z1h&z&%dKj%~o3tNf5N}bEz=lb!n2{g}5z? za9c2?#6lI@)X+_$L2*Pu_;)pKRH$eCilEn9QKB^M_+C`nEc6+G&mv5KoSzr|aeE7y zt3ywty*v>V42B@z#D?C8=<}eTw2d{S%XOPfQs@oi%4IT%)I=^@9Ctm1@ZTZkrXjM% zmhhK*$y(9)dG+`VF6Lbot|+X-U8?r{=IGn*fy~=8&}XGfR%JtnS*WZd1G(b0pMXqy z5#`9Pb@p~^RhU{>gHIP=QgI&%0Mp`)Ot^S&*cU;*kU|o2=QikKxox+skf9dNb3u@{ z{6a%h#`%nPPWsr}KuuSo1~csp{`J1xN(-wxRFT@m`?log$A2X{n79qyK9Ex9ts7Pgcl9hts1GjV-l*OWzAQol>R5RMy-xG9 zEA)ZcmuI}ZZ#dLU34^dZKWz&Tkp%18p{W&QK+yvEuPhu+G<;1?8y-MJ%kM4`A3X*>*ju6rRAs8*WX?P~9NC8!YgHdEue zZOsr^k8p7VJQqZHV2XD-GD?$`B&P8>NLV_^BilXrZl-RWilY7Whnt`A@X{u+1wFG^ z4*LIbQve2FwP1gI3S!cxf4=GU*1rM)oF07MnnVs+uK1-{K}J%bo>6T&ZT}Tu%{hqJ zNd$e1@wood9F^O5x$-o1NN=BcGQY~K>GE+U*!+7>b$KBIJs5H216_#>=uOywb3oa{ zV?$&do|pH7C@knt))e3QX_9dmX@BN9Q1dK@F|@kcB&%Hoa`{|Nj-UilOM z{QyWnx4*M@^LKyde}1hVu%wSnME=m#BM5PqLO{5rY)vK_W_ZE%{PT85K=$zkD4zR! zrf*%lFEL_8p{Xh(dZv8S7M@39#=XKGcntip{&;=@3@}ho8~ybFcac|q-lFh0RsXq2 zkl7i~Ku0p~EopZx!G;b2y@}%iDxnfC#$2_1VZp3T!#B)Z5mJ5n=g2%OxuVxcoVpnF z+SS^Fx_T~3Kn*g4m%89UsMV^4J`%!WW@<$+z7LL3X4d|G?LRVkEd++dirvCLGIvew4k?;4SMd!Ty8alD9MgOF{Q^H9o|ovRF>M%k19_2bXHPWFKx7bzkA?kgwej}IA7nXR~eEh zLIJDjf+2IauXPeA6k7L3;Zh6Ecx-;iJZ1MDNJa2ilwMAhGPkzO7%@svoOFLfDb zA_vy8%oJ2XfIetv4D|*27xbw05Cy`y2ge62BmV+x7oyQ6!93Yc z-U93X4I;rbnxk3uS=sJM4tV{2^P!%?7FDE?+^Cjs&dF1`L$Fo22jld}_XznG66dFQdWkz6^hNm`k)+ z42WFspB5jh4%qoP=y`uA00J94pcH2geGZm6IiQxt1$rpPnp3*lv_(;Q5B93AtNme2 z1IaS9UtojTypm^`@z||j0JQ+6bKCFti@vi5#^%Qp3a zx%;6csqysQ2iWp({Cpcc&wHJPtG)of9m0fVD@p@+Qxv_e+4heL4&A??S7hiwm5UHd zpr#}2Q#&0S!c|2Ol|VTi1#@uRI6&!8u%@#;$!wLDG9? z@q0{dSviydMG({WyuJ+vjI#vCf|^R_2gD@=!+KSi4(L}SHJXCBmSs`#0`MVMz6`^T z@22X2#CZFlu%Ti@+GwP6&3yts^s&NGRidS%Rp>|13ql3`mW>{nA_gfy-sr=p{$*T? zk^R>RdtgO?djkCaEGbQ_3QJXSG}O#&LomeH7fyedFLFLj)FT0#_Ljf*G-Y~8@@Vn+ zS6WI~eEkGNT6DrL|8Una^5u4bnBb3}-)*;%?A!$~flv;)$YHMg8RAF(AkHOeS!?G;_np6F-!t0SkZ5I&znY@UZTqkpV zhygCQe||1cmZbqGl3#8_lV5BEAjpStRGT8e6E+#=^Ls*~kN@YIpUrxD0K|7P!K$Cv zFoM_ae%<9@;W4Ub3O&D}I+&A)zCAHMo*r{kf^bEw*|31GNmi$QL1U$R#T1`Q&eOCA zax;0=`YoO24GCTk=bNp*E7v=>T@}3K7qlh!BI1MM+fTfFqMjGu`f?%A(g!PJo0Cnf z1j=cr2KMX3<20TTz*D~-9mLUtwSoDV+uC0}1oWV*z~YlAs&px82tjsYL1cq5$OJLy zPY+nQeALJ71N}3uCt4o-=kZCR`32_K1U11B9r}{P+O;a-5Tt5<)``JDaSilU6RW3o zlnmCXSgs!p`5gCWkHT2?1S{%h>p>stpUhnm_BifghjtUg z0?J1Bns*PL8oo3**;St?tXF_7e*Pn&EbeSr2yzF#I%~)~XWjW$lrGm_VbQuEb6icn z9oedig^1(%2Jsu>L=n!2D?78+o8cM(Tm!V{+k7EVpr-e3k4u0+-yA)j!qHIg_wKDG z?1``z(tw0EG6ergKf_qx8QrdHe@sBI^7}5JS%oA)nhaC~_5*RBWqT4F<$Tu`dV;PX zMHzLErKJITogf${?jHK-z1{m$sgA=^y zbTpR_^i1CqE^}Cb8CCK=;u1)IDY99+T){d;z-oRydN>c2$Jl^>Cm6d=5Qv{bOT9)b$?uc zJjf0(@Vy-L%6`u;ZlZX8^KK23t)U4y$~B>4HM>KtUAnk`z=C+40vPxWE29aARFiVF z32~8$MFmINHk6ehU+K_xQ*!x=MrFoh>Ko07nQ?eeZ0yCB6T*j{fxZ%{x7ZBB8BZp*|MB`>W$1^jK@TjcCM7EzetJ>+wF8s{L`&H_~vlm&D&7kSyI1GII2Uxq^nMq{JlffvLy%t^`Xz- z&knhNFL~UdZqc#5Wg!o6wK-CKa~}B&?B$@aM-5ehA?-SfE*QAG@wz!GtlIQnJby1` z$3ZwFC6-C?v3$DgGy~3whq?cKQkkN1&>QZj1$G*T9jP>rdjnXS1o@EQx9D>~=qZW)lgTA;Xm=gfRMR@+KR0U-nKg*G5w7gYslnOjqiP+eY3X6l!3_J#dR1RFtJa-sfB#f=2~F0jY2 z6Zcsci5x`bV@?}bjAyhJ2tIHwYDKvLd|2x1ig0F(FV7^VS8JOH_O^gf*VHPq#1!|5h@y-g8@fc&I2m{{c7%AmrV=lMst1M(68YBJQt^Ujwf zIXIkb+$Qo#Jk&DDK#p0gp9h0O9Dnu|w4ap_{NJDF{NVQkS_%Ds{hFSly9Uzil*>MH zE~Zwg;(+q z*YG#9e}DR_pZe`T-9F9K_1j-dF)#o0m;Bd%`3w8c7)uj8LvR>RQy4{Z)PMYI_)lB< zUW-33GbBOdB*!xh$AEc?|L@&@AGH68hrTUom#0(ymt1U3x-@yT3m*9(!u< z#?T+{th)|Rg8Eu*wb~8H0J@K5y3JT8zx?BStq*@bPOd^wR)NQs?xGRjZSX(727kGU zr}ZwA4`7u|;_Iz=kxz9FnHSL>@+GHp*KEBa_1Wk~Z!}5MHPuKBTB;gp{JA2VPy70y zc10|*Z}b^rCc|OZkT|%RIm=HU_OW*{(hFrYQzPH2%8=@lbxQI$nZRT-Y3Hm7xwm%3 z6OS)Qu#)%6>5wCcwG$!S~O?@2S1j{QCP2EP~af9N%v zzR?27v{Aa8T;czeCRA_Ky!F_BdhH(?R z)o`~KLQb&sdK(C*zxqxdFY!?bbbpNiJUk;Pw5|Bi+QI)hdm?XklPlX_qcC#^dx(rb zJF>;+flZS*yEgtOBBuWwE#rTVmMWm8N`Q-!p>do1eF2bT-Mt=Ga`{!Fh)~tLL*KCj z9o`HSs^^N~+qeOOL>YPpyC%W_(j1UIGebJ&QVlH16&t0t0C8_2ix;azgRT^{r5$#R zHn;Hh5S?7hn!VOmUz7f%7dq?sUc5aNv)ABexX13QH5xxI!RnQ~A0l{pkE;*u%8LRrgLq zCq{&>_9Qgl)0AOQEeCO$wmpPG?^%F6f}cx0#@j(hvWyvxSSl!hSZ1%wEsPL$9tN`e z(jVRJT0-^S3kY@QJf-B@O#2L0Q;(MfppUBzclT(CX%=?!-3AnuR#!QH+497XEMK_~d48|Xd1hYR0j z)%9DDx6gRY>{b5LtpQfV%L<^=%M4C602mI+OfgHbS=_2BI#hXw#*gt>T9^uA?UyS| z%Q_9-^)VFPsum(}Sj8bv{(xam=1YxY6>67_-lEr&;N9@B1IJr?99G@l@Z;ii*b9oT z>BV+mTe#Z}hB@Ti06PE<^bZA$F%6hVQo~JWsShRi2X+ezyPQEA-kr+anvL;m_Js!A z=Xs(n+yQXb<*|_rpKJGW2b^lY2SYfJxH_C4!V+r%nva>@A<>W?AOFSV#i(#d=^Edj(V*DM>y|s+eTUsd6bi25jFA6} zy)RjMRco?)AO?Wwhx8!POQM@05S?g6&(nKvefKX{l_TsZPaKY@!HqcgNJkQoJ9Dk< zmeK*cxuO&q|TNz<<9^slxmL(8N57cP^}tJ8B!PB)@xwKgmeJb z1DrpVg`awd7GL)9s#Kf62W!q3u7xL6vk-gG++(x-v1*3lxMqSAv!*6}?u8xefFt4w zDnT{l-{vrlKv~lTmIwa+ZPOHW5vCWv5!RWp9ETJ3G(T>F7tI$ys@Hs#2&gh$G0R~> zB~x(O0KLL}L8?*~36(9#cUq2A0lv0^MK-7egnKOtm6RzPZZ!u#_4egh-x}a+d&CPo zE_RRd(nD!-t==LyWw$3FRM&lCFbt`;`}~5EyHWe=@_=du_P|-s7+pVfeh0)y6JE&6 zLBV^8FToV}nG!YzzSD-W={ZbapJs~2&)x+f8rFTA3`-pS`wEymf*!`#>7-VIhx9jP zh}8GNUD7W8m+I_6%F2b#EG3Rv#cSbX5R0d{o|nso>j{SBU@Txw-7pXdR(wb<%X0F~ zEYjfLO7Q0#CpxFal`dg0))8HF!ooQJVsro%wmTF2U~g8N$9W~WVEYEJTL!pJ)Pd(p z)Hu{Ffa%39qC65fN&NdRRx(c>81x)Ua8uIf+QzphmfNBJfvS;g-gO=82%D?@-m?JT zaH|Q~+zSu)p|X|Iyx|X zTy;+0KLssV!GU=i306i5V6wdj5O)%12$L>xfhT)ngU?l64EQ>Sp=qkPyu2a!DwVLJ z3@z^I8cuNYlYJxJBvb?BPnLCHwqff(k1Hdzr#Jjm3Ie0tbYS;;)ce?w6%b3n`%vMh z-BkFCv|nurx5}0GWN7$)P^{pFC(|Z?XQ^vYy*!0g#`1=&;0$n_|AZRD>Ehi`)X|Nf zl6zj0Zx!6qPe>OP6>)TIwS+g`0w!1PPueaA4nbcEtV;iN@jbu(r+)k<%1Q9ui18eK zi=^ob!CIE%T(tWD7`NR$$Dgu#n7`ldyY_lQgmEK_Uldd*msq&*C7WL2{@XGLC&;xo zRC}h{b2zT>KD$LFz)5?BnH%k5CcW|juTDf`}EEKFQ?NABnfuQTW!p;bX5OCm$;q>VO!ctByIWt(sC5lwXS$iGE zG^f4R5vQ{6&@{nqGR66X-5L`7eW_i$di+yN$@8@z==nZ8kp~Uqm_+;c$^*7{)6xtP zbL+ysf66Nnz|+bBK+8oEHWcpHF{Uv5`pfy6Hv~mdl3CaDI0$`au?BdpSb^KU&H>o+ z4(|OAtN78}ZZWS)`bSiM_ZrGOU2}rFseK*?ezRqHsBplM^p#yt$r8TSZQnUmaYhbS z(sTjDoKD(+)0m*zDLl(1fFQTwDAG^kJoXe0Gt$d^G?-!~NH?>e)J`0hzU7yzU8xI! zVcvm!tO2o5w2M`hTmYiKsjSy;k?wA$q(dwi+A!;WD@6T`=8Je&<%{7BA4bIH>!TfE zgHHkcMnxU%6S*kQn-1fhxJlK|M7{+B;+hUz*ExmYm{h7P%P>2nLCp4O?`)yVLKi?~ zgz~3oTHz!9x&}-LyY~cF*rBqp9=R^WQ+zK)ld4L%`m02X1z#8<<){Y0Qg@@G7B653 zo<&hbB$wCcI{W=x`wRE}2K|$M+lPrd3a0B*0E}VLCPvw-{Qc{u@qL_~rwDW1x}fcT z1m};6V*CI{n~-5O=L&FqG~HzJeNz;iamRPI^QH$?vp*(He{%o*gfX`g0M;$NgjpWf z4-u-_n@Id&&r%p^zYgc>*wORm06!eDOb1zyP}zH5uBwQGn-WK7gsd8#>7CPLTQi3Vsjx z?d^=rzp>lbY6H;1K3~}q<_fUc47KO=r0krbwCfFoVFeAy02dweXRN=3L^WxOX1J`+W z=ejz!4YP7hMxto>34hs8s2WZ)*gDn@t*LpQCo#l#jYWp@4vsL(z=|R43Y@(lZim|A zpn|zOHqUB#d}2|RsTO%b`Iy68jSu%_IdD&uC$KlMy)=F5XykatEOIM1Hdlrrz!z@S zYPlK<<)@+sJE*!qlga{X&eDWCL{Tf0X~E0c{tQLHC=QCz>Tj!%;Y$Q{Zwm>Yja0W48u6zUyw)Gb&ZnT+o6T>?SWqPgXk;n zvoLSjw6kdJyY6csQx16yGy8Q{tvNdh;QXNtFq26`a{A*J_HbVEq&^!T&c- zm0M5TpqNXwAjemC`0Jh2@I4D*m1g6n%h$ao-)BM)3}e7jY!j37@SOp_A?7ZuUiz5QG3C=DVYel1xC67A^V5pCK!_pP`zg2P z>$!Uva0ov(%dRZ5-AY#nPu}BUr}~n`al3+-4s*CY8@dB-p-3O{t{N}6N$*!3_Bf6c zWI9_m-N&;qG@8RpGV2+#Xt#jh*pm<7W%`$i!s;TN@+hh0b!n9OVS1HWsbdNKakQQ! z>lTg#&g~7)HpCH1PtLTB1J5;*+WQcir+oYr+Rp)F-i}o8e_NgbcoYGz!xC6cw6NcD zc7o^%2^P&7sV^QBcuT$#44VUxHv>nmD_`Ieq%Yv!?J{_qI(-w|`{E>*I|Pn1CCT1( z8>;d4H?g?S-34y5hWcP(YRjT$ZOpt;0N0);1>)90i$!nTxJM;GoE)(Lw|am(KvlB0 z`+(hkiUR(v3G)|g2^HA8yFtV4e5~De-Nn%A@}k!yAT7e@gj;wk4EVf4<{NPCs+M>G z3YvlsAaj&tq0nR(M6DII`w^!PyOUmr?+qzmz0aSgai395FQvan?}S?Iq_ybVRu_aMVAp2M!%?!tb>pIV_%?<7+m089L~muTM$ zqjY_v@&bJ3xszuZSs4{(J$=E26EFo-PQ;Mc^<>h1coWhz6zuzPMGyodu42@ZbG1k3 zPqlGJjtz$1Uw%qDQMQ0xvWNlRu>SITznoL0GYI%g1*b?p12(r42(nBo+-f9NmRegp zmF3URpACZh4H+huiEeQXzRe!fYS}=(_?vWqt)3PP#DJjG*f?K*YI{*T4Arg|o~;qI z#gl+67FYRr73U+*8<}yAKK%J3v*x$$SE>AMQpa8#nF5^S?9;Ta}x6}t1eKf>h`P7-rbL& z*zH5?@F=2ggJ*!)&S|haDEMYeeQzGaA_8Nizm1Bo7lSXk1*t{cpRHPLl~n=;C^IoBwjc=uH6)% zD+90wQi0nY!Zz_t7&?GrN_Psth8BU=e5i_WgNV zr(^$d;om*&K!{$T`WVWhIk6Lf-YeOP;C`0xu?v?mAt)5W)`|j_Y+yDIZ$D>Z@dalS zFNFaZ8>Y+GUgXag>i!Ay(%fw>^qhodEUyPZ7IL#iU00O8FS}cf zi+TWz?X}mcPe}yh+mJ)&Bo8jp{o^wvzp z%Xcc8wj3~bPK7vl3XRVlUe7%54OF-hBh!=ppog6~fW?Y(raU4f`~KgRtzey{_l*WY zBpsLxR8Tb(-pjg@Nj&!}|jQ zbh6FgL~SU5fMH%0CZ4Z&1HcybY`+YmNdU_RBT;Rzlmn-vM+;{A7kA+ka4s>JkTLA8 zz8Jrl!5w7!=$*~K558R9?h|lgTncMf0#vahtr_^1Kr6TGZ>okqkUXEN+r7EQgs`Vc zH$LvSDuvq)#&3zP=lLKWfJA@5qk<^{iHEV}bA5+b?1BOug}@DhFt5tHaCB~e&BXzm zT1AMr@A$Dmr9{AmUZw`%?6rEetmETe0Ql$CSh80Jc+Xl3EUFE#Qzx0fcbv_`pQgJF z>p~D~O0e!lzT?(w-fvr02m|Pw6_+!PXHeOyE$Go-ATOx)1HU^Y`T9J>SGs7j1Sg4T>psYTm;z6@ z*KgbTmfjSReqNw@1^|4dsWBAgz2SC{>I<%I`vRQDju`Y2)Wz9I6Ds+_oaxER0`C^q zd!{d?f_^2_~7*dn?Ye` zr{+T^6+WG60t^fI=toI&Vl8gdt?n10M#se& zZ}c&JZ`(V!+Lv{qm5`u0Uglrch=~zx*R7wRP%->e{x<%L?>yIU0Q2P)#22&$;ve0C zu+-z|!JSY&_*41Dpg!B{{VDJQz%Y6V-qZ^`uS@)?oYZH7-CzZrq5yw|qIQbfTD|GF z*y%QsS0%6i$kYsRwC8}|6Kb$y8iREt;>u-EdE&Rp4BGLbabIBP?qQzh?f{%k zXaiP-A8JV|m?OYC&$wd?;3v(MIZ=ud9GH{U)Dd<}4pzb~mv8<^U@Zl538>(Hn{8Tz z+w{sfMi9rCzcw)XmRh{~O`IbC%TGLrzGR^S&^Ic8uEkJ^xyMETKJ5A#>~WswPaozy z`QJzn+_*M8hR-Q10HFoQz`EV?WKeBp00prw*E*v8l1fLPN=Bu+_3wHwJSELjP5gjc zRGCv_$Nt>XBf0!A4{J*Twys)zM1Ui7!>l}e#eO1L;Cts)-FNCeKgIGEX6g6KY$8wpff;66+D(HFs~!3Rwalo9~l zagPCSaXjZ*nW8?{pR0#x_rUX>O25I;OWKAVU+z|vTKXY#r`TxN=Ysv}V#%c(RPU4M zbx)~q^DOVeiMY=XcXuGfz5%?gBg}%mbGPoyly?Vgart3)IAsMsMw2#)hi4P@AI9{( zV}|wyMjL*WZ@duh0DbQv9G_|&ND!FH5ylQq)(K;020UxPAH|2OGb0G&|HN+yVM|R3 z=B>c9N=&iAHU2KkpTU0U5g-#ypTLe}HvM!RE8Vg3W+(@;G}`n2%>K!Nq=vL{aKdQG z(&)!mx5#w>Uqu}0WJum7k~y+ruK^rd+!HRE#*Bbl4fhO~2xp_Yc7TUB1K*t5z_(^a zvJR8SrU1L`5&iAIRdXA52OO|d*~!lxoL4M?c{bQ#eub2^2}_#&f^aWN#b!c(^t_c} zN?6+#O|Jg9_Lv9^WSoo;@KyhQD6-&VVJOd^Sbgx=tl#N(H=v^34^*Ob6?d%H&!Hig zIpx%Cde~+yUv4kOYVv6RG}pi{Bj7ucn_(~%AE{zN*iToE{W$BQi(Hxyr~vd_3N*U_ zT(m@h?atEz-0k$KWGa6y49A%>lix6BSN2 zpzHT*aCLHYt^<{bRXCBD`3KCAd)RK)7Qjq`;)~3&G!XE}^>>U64~7sd@vbseCGBJw z$2J3=xeZf(fi(rU|2&QMIwn1grBEfDLG4VGWokNN9u^g-DsEFv+qV?Z`qNEfJ8An7 zawWK_H46MsX-d4?#%+xKZBPDfyo8mxB=Jx=53DeFb)DY4&oq)>zc$Nn!^CxFTo-t@Ne(N z`T(;qwOa#FYPR4BYr~t4@Jc}Fl}QuOl&!F&Gz8dU-T~1oyLb8%Wb9So z8RlQ!1>myd3e#j!UT~jPDOx={eE=#Mcw-|P@Qd#rSUV9p%te-J0!n-bgX&(|Xk+xEeUwvMm<+S2eRAV?4 zLnE{=USU7Z3;Uhy*&M0R;JWJurHjB1dzo#flW^MXZ<}bWDs|()$!56OIx}AsvJTZr zksOHC@-73!cjVH}W?I`)L*jl` z0jyxFOAcO5;52Uq`QBaToE-r;iKOAV{Y_Tw51L6bwesQuP)gdv=(=6vc$P$7Tk_9f zigoqzYT}Hx+X+>!tU!btzF}`rclpFtemzitDWwaX(o-VX19Q%D0qYEK8uyOC(Imho ziH!K=%2t#iP~;X`;Q2l*yza^d`p_e72e$9K_TFqpHpv!aD)kk(Vw;YsL7v06J5*$) z-=Qjzf?F4-4g8sIF%&v1Ku>nDD_S0B9R*4O3widN(b?p>j4!uHNANR0vB zy8cVK9>C{06KE6#9EUlNqndv;)v^HeV=*pJ=&cU{xBNm`zuW$L{zSE^=lZDiq;8Ks z6(ueJAidsip9uP)IaJ<_l5M_q0JkrU9S9npQYLMBJlZq^V6Oho{GG=v#TyyKsQ7MX z-V}C$g%UVX^O*5%X5Sk*Ygjc|#0d~Yw^W2j`nNalL) zcL8FXrOVfP0DkgHJd|d+0D=n#RJ;LW^V`+GNdmwPu;+5Pw=a%nRTsQ!ir_vkpC=#j zu;1}VazUgczBr>wIBi{TkIzM!UoQSbbDn%J;&ch99E<)MW22EHE@o1%?;WdAB~0_4 zxkYSioZ@)MQ1$2DwCIYqlO_Haa~uEt+W%*--R5B%)DeHL9?ahd!k832hP=jw=`3zg8V)%3wTM@9iEv|X!)Fs*TqwzpXH z`j#P73a4_UKU}|l4-W92!H=B;y}iQ13c8_g^1rbwz}wY5pnmv0P^8NeTi^8pAh zvV#5z{e2D2A4er$<#)(NPLXx}8Iv)*nb@s)TwyNM?fG6wzU2GyDf!Ri3J88Jw92Jm zW3%vLfMaPs@d2w~HrA1NSYy+pJKxi0Jpis>t#I0j1s;fgOEnxO2b}ccz`kI;B4_EK zq8U#1)e)K5o6f@=76~U5At7V|*neqyu&y}0r2reu6Jb>Q!w8-p+{jaHvk|NBuAu#U z6SVJSWa6_sOe9pO(vR+0(7&?ZL{WpHX%cqe48p}H3JyO?I_l<&xz@B1@0ZuA0Ki_2L`|%_kP)P*0|vo>{rfi z(Mv7A;yxwq#R(U}_&nfu)%K5S+|?0$8NeYm5uM5n!V^aQ){UZgRNA^Psvr2p<- z4C+2TWZ13(Fhc}Fo&Ck(wl}y^fOvZ>tnwfWCXkYgDdMn3iy^{^)G$ooGtytm() zd0(vAa54C7?bZU|mc*mOS*)fb-5 ziF<@)7&C7%?BVzDIB4d37^8s81OQWrGMt2z`74auR}klqE9`^iAqY$XhtKBUB@~W7h(l1oZ zy*KZV_Si|;0oaUvf9`;Z!5YVyAugPX#CesOl@U6~KC5QJ6na2J%Y|TQ%MP7Zmp+v0 z{mLX)F_vk|N4R(C->4U>_1|B;Z3g(cf;k|>kt7MxM!N5TFG&j6tMnXZz!jqsiWI=S zhx=3L9SDY`-H-6A|J`0Mz*i$<2Y{e{P%QEU)Byq;SxJ^qY%GG=DHFb`3re-e*WTUu z_xhSDxYbZHLib-d?#$yOh)#u8z_!M*3_{AaUJ@UP86%D-(L zN*@3H+P{BIO^1wE0y9puj=IwVRJ>(SRsUb+eEw}J@_)_%**N-`n3+|x>JZ$*5imLz zX5-GQR^6+L)N58p-hzSt zCR&BI5B7|tcDz8k2ah+&yXae}3MxD=%`^Yp3hK?GQC|1x(AVBm=kvVh6E5PhOaZ7+ zZ)KZ!3G5N^uZrZIeB~(){dO%E6@l9wDB+jT)r6vs#ZRa0@%Ko*YMw-2S;B(gm&t@n zXJVbspQo{ztb;gZOQ#=DEFkHi(Xh6> zsRk*?7qlO*BKEeQ_fg0lmhc9{|5{1zgM-OGFt)Ia$3Y(u*7h~PIODXFFXv-_2?CLW zw&g=Qo!!iRk((?PjJ~?J?#W4j$GG1f=K+0lLz{S4Y`iJ$ME~IKZmrdrK z6G5w#q)w-xy}bRc&@ApO=Ou+-IP2=BQ=o_>f2JPX>>r(C^_-ix-DmIcz+@hfM$>I} z;Fg@V5gyNrBnhpwgeJi(p)Dkt8Pz@DqzaT%bgK8P~e~Gy>mVa2TeelsUYT0?Dgf*G2hjAZj4$VV)yPCT{Z-rMNp<&TPcLE{= z?MWAkUG;A}-5e;Naw!3L)>h@thM8l_Hqj;VjP1C;6MW}bxS7%Kx?jEXRzO?!EzvkP zD%>K<&da__^Uwy-8STEE^m(+>;6oXqyDhEjF{9lRn2gSq=1s4xoal66%&JeaRQFEY z*~_?;&oyv$Q{!vkRIB%wcmcE-%q`&RNeyx8Ab)@teGN;%yP}P3sD`?U?Z2{P@C*{| zmLkIPf@$yqgowgqoDu&3-eZKxa|>#?&8&w#EPtwQVY9z&s1P5mzc$noaMr+-p122l zOkM(t5_CpCbsgq50C%>fnun{#04{-D)>VNW5pH=unPGSI@7wf+8$rPLmMF7LCx9R? z{c{H#?t{Hlp-;9M%~zbyK(r{~hNm)m0Zel55U(5!<1M`}3WGe3gQm+ZbEKE=BmDB{ zw&OtfVzV}`IyY`!*!qfq2QbJ>=eY8@BObYDzacM z95!1#;7m=)AglvP*zb>vh{B3kje-ui zaR#@NxRTw8&n*pkK3gympj|o;+HUgXv3@uz}E|rt%NZX62LkB62>pNQN1FXjQhpw2GVYxU)(pB zUNUF##B&|sQ*fK_t5H@CBlY9UsS#ZRXO((H`7|2meg3SepDh0M+~cv-Ljc4qvEEN} zWt$H&V&6Q3guMiqJpYfFoE8obI8L8+wEZ`_a@Eei9Ox**6@fR#cOLRCr`_#%lyINS z4BJSque)CB7o~~sfua}SvT{{`3Q4}=J>QQ=0Hm!XT^Z)XSj4^=uY^5AaF2NUwnO~+vAiY*$I~k=l35!Vda!Rihl;x}AMy21kxr4+hIyzc$hzTl z-seOw7mcKkw7dTDa7vT3N9JVvCs8Lu-N!gZBNH|ZY%tHxSTqMv8^BNr+;juCu_+Pz zww1@~z{O>e+6*@r3KUaKI3P&+*d9FlCIj1z)w^=pg!yP_yL|G}7vCJK-Av+Dh_%Pq zD2(5JJS3SPBrP?)h_36ju%zGP6*f~wize5<98en?P5ER_7NArLaH)#4;FjuFq z#dHj*@a6-mYYC6=5|2b{mQmQ2J*Rq;;5eI-=Lb%}8h|g`z(v0E^ z^vlG(d+Bajf!%@a9nHg1-rwIfzl1}$!}Tc^`R@A45O-4@vtZoiKq=z&g|HQcY8P&f z2Wb<_K2h#(*oh4}4d59^TI(*bo5p&m7jAmNJb$IP*H6|1C&Kfv4;Z$}Qdujza7G3= z4$=|KgD$SYN?i2qJXk_J<8V#0P<<@aQ9a_1xH|5gb;NS`qF#ijvs}tQ`CWD*fAdxH z9ry%*H|nKMmxmn~IAVWq+f)lZE*AWjJLP7nXC&(jhB8ZTsh|4~JKqrm;f-J4G|Vcb z`J;b^$Lx5Q>i>|h?1^aAi44o~e7)>&REbe8jQ{;mOI)E&y{poIjV9?QLGE-P-+mO- z{fS#Uic;KezEpk|Fu*!_m|xx4P!SRkxt`YxAoG`(Rq+sd{I;DCXmZkA`@F3lHNP`n z7andZa zJj%xhe6vxRLhYv$o1ulJ*35PPUgXChdBAv5Dh{%mMYsmRmXIIAbiCRSbWN(hkBb*+ zR>X~V_GVTjZmV_tfLLBoa&1W-r`3h~cCI4x%kFhhtt$YAb74}Q@ig^Aao81YlbO0t z5JzDzzD!rXbGoB}%MG-_c9S|7PbxeBu32r5Fg6@3N9s@r8)MW1YmP=}OM8sN?JcD6 zTJG>#qKQC|0^?;1`F8B#J2wkKBN6x$6<52Od8L>X1HM^ZNwL7?IgW>2B5_-DpZrzx znCCklZxc*iiMSO?>|(z#sH%#Uvt~00C$1%9hyAkV%&B|G=U$@80N+0u35g!A)J~8*qqI^OC^7qur<4<0A;Cip1;Emw6rT>#|Mh8TgECJA61ZBsj1Uz=umR z0q3A|8-M}FeowvcnvPL!Hohy|*%@Un1jeTCgrZ@>LG=q$?Uh=gKYXWQ)X-Hv0Ss;t zO%b*s$GHE>?shf#9(F@j9 z(k6piCBCX#o{{OFpfeNZ`PG>E8a}5Wv9_!|0Y~a;_qsQ2%+9d{->>xG{dWU5=zfN3 zqSnP!vD+4ZuCB~nIAHeh94P*C+toR8N955@ADXQRmUnS(#+&bwj%_xMb=wG&1OOrG zk)|HE?gY=m2(P1QOTbDCUN?hT0-VcG?~7MOfG4`9=8jF!W5^B3#81qw8RkPDV!<;c zf9U<`iadS6zhtIX`+)PN?3;ic=Yp?;d0uy1b* z94J{N>gQbM3u2!<1Q80pwZc=xi#8;4z>A2Fx_`@fS}*C4)(hou<)JCF=bID4)JPZ@ z)ScEKq?CU)D|n2=@`ZXQtf@Fn5USPnbLBT%e#E`g`x1|)61u$`awaav>v zu~6&Y-_FqQw<>{e*OPPUK1%jRKDvqOqTDT7cMG31@d+&8{-q_cFIR&2MnO}*mqXcr zjcM09Rr|7?Nwe(vk8K1>*S0aV7xa=cNTA*J3k}d9AsirPOrmxQx`&6l>D56>a$U{q zJUCN(^BwQWvnsd9)Kd!9c7dZME+s-MZ|Vl?DdlGe{$Z=zx6Ql4GB16R9-;5!U+LNx zZ=@jk-hL+!v~mM*O+kMO0RNtkC!(CWF9ZK`zi%8>(Ouj3ZxaLu_2Kg!_ekhhLoV3~ zZn5Y-gwae+N0|)Z0g?SnlyMf9KWntKTyiHq)h5J3_ao8YpiP zPbrbozSKEd`{R20t3=6i`(E{O0{nG$N9pNeO_;p$slF6XvrOxdg!^su_AK&beBKS0 zS-iJZ%i%5jZ97Xnmnm+2T0YW?p|;&8ekt9EJcj^RK&ZbA9f{`KX$8*P1OeDrxDJQI za4)Xt_WFYj*%kQ4BLcBh1qS*Si}kM0V-Q-B9Q%4;0QU4qa6QNCbnR!|{JU=j2JYKf z*j;viiaBo0-i@d4U<2{ua!F?F%E z!owKqM*lFM99f>^dmBn$KQj{@KCYjGlwtdzlo7bu!)F7a^ZFM%#ohs=`I}_kNc-UM z@1@_}Y{{b?rO45TbM*+M^9n%4srxa8G1J~&Eaa%mCPS&L9*6fWqW2|?a=nn5Pdj|; zFCm{x+Ky(fS~V(ZR8I9oNU#>_wB7piQEC%O;uyJKTH){!FR!C}*f0ji;Ly#2ah~<^ zvS>@=H!?pI7-odP-jYh4q6w8;K~GZVkzB(H!&t$?@?tCeVsmd4ENAz{SB;d1qzK!7 zZyBMU-~Kl1cf{_$=X-UW!m6SRaGGSL-^v8o+od{9TtFk26Zp^01Z!8y64DAYGhk%$-F?!Z45RJ*GFAN0^o0P_8Isl2igD-yNT^OHlP zWxB%lIsv0IUA$Iqn^?i3!d=z_`q9TKV2YJeR*my;ufWBwE`^lvlQ+U{>>DaQ`n>(L zJ-CeXr&}$x=WsC~Pw)K;lfk9aX#ly!?Ip0D94SyW`}e#4NnoK43VuhZ0kYo?XRJ4O zMmVVFIEAk%c>0R>>-U4ZDwx|j(a#Hn@+<1(%FUr@5JM z9=X&J=Rv7BoD1t41kD$?vnB)fVeXLotqyUyp!Ga+=nXlQ=Y7MLj(_rmcAbsTZ05)l z`xkUMpI8vU{)(U%|yHvdx)A{qOfesgwz{c4NhvP)pV;l=0z_lbzSd==eA>L?Np zfx^{jiWgH@?U9rW`%$>%jQi6!e5C#EQ{!9l_eUBMQTWdhNqb z4z&HM83SN>IWEi<$*KHXDGKDsaNPn{PX*jMP9jcGnDy;Av=3~@GqhCOxjgVM0Lfn~ zowEJ5tCOI)z7BT1xfi$-2lmK)KDx)l;dxmcd{G~7Ufs!J@Y5i#KDux@%za2Hl+ zyX=3+TWyi0E~ROmE>`Ln>+#aQ#*>|4&(>S1%`n`^Y{d7y~054$^b*$f?ou|gg{Zm|-H#$`mqZ0Fh z?0Kp=*xa)hz-^6#b>RH)p3sD21W*ktYU7V@M+oo@&GR;d_EW%oO;Twh^DOE!j8oD) z_&dAm5}vWghIfr#>$A>KE-YTZ)PE~pOP&q)*}>gP#LT>*k`D8}%-Bxy?#GgbJX(y3 z-ex8bfiLquk=l=9HPo*+Vf`uTKs5r~{^d-)(SkN%4rzXAaazR6NSU|Wf>9u7-D}|A zsHw-~-7K#BBU%*gFE@ETZ~48SZ(^atCe|H!S?w-|3Ymd(hlTKr!oC587Yf9Y%2sO5 zVfGMg^scZ^Z1dm_9p$80s)TzhZ0(EP%*4?)mWQ=~R@y1E{LPW%`TBta64?f~FC$Q` zqRuL|&Ik|x?aqPicIg{925&=hp{{ol)nH#rmQcD)tN`j<>xfc;Pyb?YB6FXjiu)>E z5Mi0UC&1?u;@{?XCbh#8$9-@EL$ib1l-c^8hk4*N{Nv=b8AQ&2B#~{*&HxX{DcH=L z($1OSrr=`#u@@@1Mnu`FlF*9Kb4M(l+RxH{^a-2|6W!14^O$hh9j`psW+j$yGT)aA zAg?#LEtyPB$06_E@mPN|pZG|iYnI#aH-2D$JKP3d@w4F)*+OL6T`u&8j9bGa28MkK z2%MNrGDkZq|6csc_i*>sbA&6JH!l1(hev#~mDp*XDw2C)zk1xDIOG(Xqr6X07^U`l zeCuH!z4WE=^ZrjpTXwuc?0m=8_j_sh=GD3TK=oA z&;EfQi(c|EQvL(pWFIt~6B#_p5%ms{#=2$pQwC*GVdUZ@Tc?N8}@!2E!p4gR4W z_Beb!V2<<&Z~>0JHfmBUk6Od$7#@LIs$K{46-A4oNlA_&iJOF6vCetAYSglJ@^TgW zO$fi_F~Eu^Mx?gZ&gKQPnDmqMaPPr6VHr3aIND<@@_9|aZY3eIy|tb>wGqE;D&2+V zbK%~>QVz_A;dhpc!sT%tcN-sXh*uA7gNf2i@O$gN448W_`UGb$+)Y=Ft0m5+_U(A*xKZyIuBozIfx$b z6)%#a9#5wI_OEL+@e(+J(n3JhAzpeW-_IB3gZ7h;=!Y|7_Z#-&TfC1|cJ`&Xy%v1H zL_B~-d?MBxon(iP{ZrSpuAKn5P5788nwE|v`@m&*H2V=oWQ8tbUiV>2rTlZ|TU9=e4(1+Gmlg#j*a5#<5x9{k&-RUlap3@MOkn_4|f4vv~Nc6j7q@TjseNqUiQVc`$7cjoGdQ}#vHI+ z9a)u52eY?T{XE+6{gAhxaO*ZJ7670vMlF0qOS5D0@p0; zMRU}r$N6i)C}@rQME3>Ue=Om9ElsIdNh(Q^0>BoEar;mCnb@8l*?4@?A;p6;{)Blt z-}|UKfj?7WoFCrGZ?g0b#=0_m*3N8tq9d0}u6!#9ruU5f0z5+vgr1~356iFZxaD~i z<)!FS7+SqK>zck_&aQ*cmBHr{ay_SbN12aF!xfF1goL1k)@;7f7 z7+eDGv#ubv>Ac@lsgJz^&n8@{vFX?g_|^Hs4lfZ85Dp+)!b*E*M@fc21aZ%Qxe`F+ z&c>Md+xQlH%(vRh_@19nyv1}OcP{v~lsI-nx1(B;4yNP%mpIMR>A#`{+-r)BDsQa- z2r9|->@S25LW4Hy~)>_vOgCxP%tZsK2# zVy^QMbtQw1U6~TQd6Mkasw8^A&_;Bi8>~3c_mqij>`7nR}Eqg}r^9P&9 zf_b^L4%W!q351n)$baoMxO z-)D;-8c^Pk5VwFveheS>e{DuQw~NR|uDM0zD^-B|q3=&Y>Pw%@wJ9Yrn+bt|F zBq_luh%m)9z$hIM530V@4OG-W4~j$-U|(9md*so-glIvH^uZ1Dx~-x6C9WYqP1vocbtz()? z&h@}TR1iSGBo2-UfJx&nEj=GWcmI9>yFQsiGLNmkbVGLwU19*3zpn7T&Gm#L^1c30 zh}P`BST+gN?|=T|rtVlCS7Rb#KjIuNAA+ z!v$@Oe$hv3056S3TrAUoj1ySJel#IDO8MBJX%__LS+DKYddigaqp; z6}XJYP@(?n3wMxK-MS=C8}(O3AKXhCpUdNOC|E3C1y=1ZU(*(+hyUq9o)?J@F+Kf%u=ZxX&MVt? z*t>tl#(AY}_T5DsfQ?8>Y{Z_FlE5IvUSi*hM*e%M^zG{$-`GY1WIvqqVXdWx)yz3Y z)tqB6{pQ|V~7IJ5u&J*q-63B|i@m&++Ym@mvpT4GILEe}_1ZUe+aD>io&#NkzyA34-OA5Km((zgu!@cFx3)hy5i33ef-2HAEVx0PAeuk}vJs z2r~J^H_RRYW0*LRwQ)S(bCkiAW_;7#mDeZ(-L_pdR(3g}0Zgf7Vd5>aO|y1?mNRmUVJ~QD5e5bKLGp6i`mM zoH-D~*aMoD;W&;?xNdplQwlzkV}%U;(sN!5)No$H{LMwB8$N=md$QUzBd9|^L)&++ zwRv~GfmIcLD6$jOCLaP=4~31PVZxT@E8qc6zeH?$fwl9{F{J6S)H>22vw}%w#fpn`WSXEyFsR0G-{E0eQ60Qd0FY3u=yU@|PQo1l9=c z^Q22qCwlOI4!R&-1OQRO+VB(Vs_cC8h}0L)722_C#nv^>3f)Mcj-kP**I!4=xDGUATFfa-NeI0!Nf0^8RKDD_)1j7rxia;sy=FBBfjqX1-P^x z+^H2Y=+r{ri%vNw{n2J!g@8eojLr+_yD=#s-J>1>2px72yLD8nYjn)1<|m2~nFn7~ z^qQ#94Ubx(Lu3li&Fm!{t?5Is6x0^u!j)rgN==AE;=L_6Wk(7q>viI6t89~jHI5(` z10`e=HIg;@IlG^45dwNkGqydJagXL9r`y(-_86N9E6$Y zC~povPMo(yH>cP79#F(De}rH8g1cZ5z1@;{`SO@Ncn~zS(7MOW(FBCmgYWb$piap9 z1w{{ArFs-H!PoRxYxymuNdl~$smGQ2|eTj4whI z_Ph+o4aE20v-ls35ykI|v~Ya2p-b+>FxM@6!OD|-2$-Uz?xNH6NaevjXLI$~P-|Ai zVzb8STV>;m^&pMbejKK<-`Ohu?Kx?9tq!kL{`EDAiB5U#`9C=u5-RA#D3lA;vjm&3}T3JJ?Du3*ay!=N&Bs{Yd(Ww-KAkMa`g_@e}y{W`WW?2)9he%RgUpjwSidK=~y{Jz%K!ad#_J{Hq;nS9Ydpr1;y zZEr80qRm0qVtZ&sW!X=D5x-CVF(c0azYG^HaLTgE4vqP*npj|``3$Rbn78V4 z9aPz5I9{H7Ul<3(zGlc9S(vQlhVI*M)-!}HsC?u`eUdv8;xVac-DiLfo8+qc%;j?q zglE#fJ^Yg`Hp=ta@U^3JcGIE4t;O0OF@sUy0b$x8_O=1+9tn!1H@}7*OR=q z9TVcsP~n+e0?{RJ?MkIU!ey6}ZKA2U!+IWpN6jaQ@Nel2uh)aMdDQ(0n*^%15$Sf{ zy0;>TT<)Pg2!s^ezV~AV!H!WPb(WVC=Z->iQR_-l4#Il%wb3g82imbMRDly#6?S!J zk=5K{2Sm7i76cckQH2N3J&PGlB)ey3empM>}M~aK^6|nZ7aa#yD;G zdVRo>7)Ek`rlHLl?D|EkHJ{VPKAY$1fF0jCy0aW>Y}xB;ofP9+H$j7PkIJxwA_s)PQ${O@>mK zAW5G*U%aR?hK6uSnf#*961k)6YZ-5$zQ{GDeC?cn-_ydMg(aL;uK#nkg+H2V-}8Nq zdsn!cF7BW7ZA__zJFwK=z9PI6?{C8sWWaDf#kVfUfycWWM){YIW*w0jG`>eELjH+* z+-%Va#H3dTYaTwi9wpxehh%3 zo*yK@n@YqKT*5dI`%5!CrHDmEG3?f>S04SkudT>78&fOoR?PtJwU@nL`gW!+GTNd@ z$k*pXB3CQgrJ7tw%HEXfC^gg2~yBh2M$p z<@Ad1JBNhk1p)JtG6dCQ@9N?X81N-lxZHB2t7;-8O+8d>%FggIcb_sikSaJi5zLU7 zZ-JyW7>Eay(l!sp-l)gd9K98D>T$k28htc+>?O_`=e`rSaS>S%aygDE1>_mYlvO!0 zbn$5tGSu$sT!hl=q*Nb`^vBmp7<~OIA;m*!4oU_*C40qJ`Te|0JbyYOldPL{$pd&+ zCS5Hz3olNi?9b*}>afwLguU$Yr=){P*d3M1O6+056O^Vazh6XT)h|7MK~jPBK*mI& zaWQvkAboRop>^gm+Wi-#xz-niHO}-)|9Ds>zTKyVm!l8MNhW^vn4#l2!~p#S`tdRi8C0dTdFQ#FgThN-Q$lwZ*}w#b$q6Z*#MAP3%J z@9l>QSQ3{kG~;xZR`Mm&96#OLd5EmyyEiMXyj z#|h_?Mc_w5bSVVC9W94CmwQ)4*lHC7rUp`#T3;~*-KbQRAEqh8 zj#3l1s4uqcSza83_1J0oSO3O_fYX~lJ;E4Z*j|~*#~nI44!HkB8hpg?jF)o|zhr>e zl!Z#kGA{Plry4ZmFkK_iV6Pn>QkGhZ{!j{?j}2Ng8PljsO!sish|fw~S)eozD4+C{jE(?u{84Go9xH9 zF}LqShTd*tC)QAF-K^)@y6 zQkiRpe8um9v(%0DgM+p}EN_281vgq;#=tr!7E@;i;l4|1zssCURITBoK#S*I1J8f!O>VZ$J-!}w*Q`t?wF~Zw~S|^I<+knvDNZ;ifPq?Hy z)%KSTyZYV@EoaNe3fWu(l~~EF>7D?tn?gFx*nx&bJOR8>Ze`+Z4^t{#u#B}*?oP;Ju8Y~75y_MhHJOvs zh&0I*?6ClP!70vSI5Us({^@t{U^0w?M*E)T*JZ50s6&Pq1hM52rpA(Y^SX|fGn#&o zR3!T}+Ls3i@k{zXpX0M_0L~D;r7pW(Ahqt1J6B;A5>pp5Z~E zZ6NG7htO+*i)9NC3JEN?du+5V^z4dH#p~n+q!i@YzsI}-u_$PIcE&DvP`W?yiBW*& z+V{#i9YJ97j}YFR_6CxlV56MCe^D7dkBWHo4>gBZ40$&VRsE=QssGVD{HeZgo9%he zQW8+@EKl)cdGhcT<%&_{idEj#O(@a@+joV%6X3^$u(X6|wHRjeGWtwu}S_`W&;-H4=r0FshP^^x+`7>TfWyrcRu z-wE(Wy0@Jsh8GWp$?OkFUQG=Jvu1apexZ+hm+7ynS#Ym0xv8ZyKLT8o2PjW#vPt}u z$!82B3{0ReJmC50?Vh$M&MM(sxKj5aBoh&1DnQ1Dy)`ePmQbmtKO>>$RusG#@D+)s zPdENH;VO&Gm!Yijw$bh7(85gU73%(-)nX!1vfx0z5uEbsXbgAr9-OVZZIU?e+d39+ z!NdnU$~r#)dHFIcHa!z~;o0Ncw_o6KXIf^>P9Igc7B3$DdAx6!p?{uT{kq(SJDC^v zB)Ivpc%16l$&EM8-!6$WP~S1xbsq4&lpy{z=c?M1m{n3B0S!^Zh{ z>gmJ!RCuR5f6TE27{Chdw$?}8lXysaitR0M1-L<8n_X zELUr$_~55SwGRb?^+rER958Fs^oNQ31nU^T?G;I?Y+x@8S zF9zz(sNZ6=8`Ft2I)IC~ToGu4ShX+XxPVp7*37y=tH(u&DS$KVh26^s#N@s~=R;!z zULE9kc&S#3`e;iSUek3Q@+Yq4-RzV7fxDYl000L)E z(#gsJDd>A5DCr`A)p09l@y(p9d0A}SNj~AN?nrW*vt%HMh8(s?yI;WqO@`_-n3ws= zklf*|VR;$C+>kIs8@M2_vvnASBVc>nVMz(@Q4djs9vzb_+YfR$2jlZ8i}Ytdol{Fv z(j+9Wbyy$ar%RK)YQi3>&KW+jBL&Ql6ab)oX(qn2IQ2=TH`eFPZC2c%l!hU` zXoA@Q0<4`}=4*{{XWFZUI+FcSkB^Y5wOxEK)h2vKy5v<@U-;|2gU;IfycrT~m!|S< z`l2<3*e>k(a#jp3qdm3C-B3nM`sr^J&_}Lf5ybx0fu<4h;u-H2tUDMiJ+&B4^{CYt zNr{C-NC)l@gwDjoq+D6UY2<^sLXpFtN=E{^-^AJ?Ml@)0e!|^*V>qC?P zYqlri%WcN#jG@_0%7{y)^yl32?p-9;Y=8cXk2sN!_g7Rt&*-QB$YIkPjO_+xl;mWI z=BfhpqyK{W9|)c2c~sja2lg97Jsjx0`al{743AM`wgLE$tGR>)d&q@Sl$^Iw{;Npi z&TH-k>hhb4rjf*jzPoSrXOk$l=dJ$vt5VRffQZJy+rMUWE0d`Gom-EJH*(#94NqfA zM>@!!%*Wm{)aHh}cJkdCjVsy_AQ&V3T3EPK$ojcFERRA_!$c^TLJ>Q?g`6=4uY8(a zTpBr=M^r|yy_I>Php!_Wo=vr-!X{qWN4mehPm>7`Zwx$G-dRg$8&@8@=eSDc<~4)v;`7Yy_p32Fc`M6USNmc0SNkF;{X zs@LXvUZ5V2K}B`Bg&eZB9lZb0#_uu(>oRz?tu>?seccATfO@@xCGxxb+V0=++@GO= zbM+Us(398Yg-37aXPn+&6gHBBHD%8m9{Ctv0W`pK&1*AhNym?b!{6ZqajkEUHW602;d z2+xuhSJc?;-_Un}NBTkm)59v~+X?`B`WT}WVP;`R36p?udcMmU9cicwPqMRgR(Z^D z`iKQ4QQUF4(+4G)-p?_qI}mN(iYvI}4b01z0`FQuUGh&%jcGTQ*4mM51wN2W2Ogmc zH#DTH#_|X%&rctDz~`J2VgdNpyokEjPk`Qefnqf&0l0p{5`6LK7Kvn9cqg*I6mg&k zt)hy6UZAMcWI+F}-+jnTf9ua=Q^?#;&!lc|^#k5M@OXN{^>U<7G^)pNe5VLi6(zsE z|9mMl9V~(_0BsvQ1NN?yyu}TEf}>X==T|e!u2^ORwX6aHJ(?X@s}Y z2z>)HtE(<+MYqPT3svGb`#DYR@iUCEM;ATOIe}P!wx? z*#E5uU<`E4FV^DTB{!NwQVANuEb0_`&|47$aSY%(WwN^SfJ))1bij209^K`9?Lby+ zjW=?tN+P}0Kd(RN5BylX9y}w07BiCu(C1WTGK5?G3wX78SLIh+ACK9snB zJivl$;8Y`bNtS6T{xF4qYySKFBHxAL;Jw#`%M0{PeCtmrTiZZywNc48ZJmPgVTtAb z2-o+g&oYb<)CT&3R%~eYiU={#E}+>$Ry&pL^}21aqZoVj%5{)VfajbRYL|TWmnI4A z^Zf*mz`gF8Bt9_$)~!oIc`YyFryQbCajvpJtIx66UK7;cXKv7PeYgBbdDEu<0P9H>u0`+79SjkN61&SixLLksH-M?{&7LmzvRh7yKpbh&0^#(}epabIdbaV0q za~+51wg{BuUa?m+PWT4JztY0R(f!hM3sOVRizm(Vw<#LVZQTru)579S&b*E|M1<%g z8xa%DuP5|Ld!!aY1y7E7U7)AO0o^+v_00Ot_V zK^{Nv9}8-g!W&SEi@e}-!#0^Jybdgm+XedIR1d^%&I&<|=CP3oIvWwL_z(d|zt65~8<4j<>7401)akA* zu(_#aA=t+ajMM3C*~L7C{c>3&LsYtxrhfO7;~)Z(5@7{2i+>r~ffX}hC^GcT^WU!O zg8b9@fIVWo-9-U)KUYI^=7_nU{(&NVxD?9;)2~pwqK0+q$P{$Ln2d|1#%Nw#F1K!_ zb?-gr$u6^t4J5~CsGx+oy=}8ed|uIKOiU=(4c;tuElJX#hsIlU zE&$eI%TjD49x{rfn;z7yzJPJ#_$-W}Nlvu%$o1hp?2cewnsvT;3g9*90kFkB^U9&V zA@`zK7WWgria+o^9dIP-wMM#SvVZ%WCNrV6BZq=K0w6u~4DPLpPL>H#Pq8eurmEZw zJJ!xXt9#AanHUzVHJRrY+-Uuc{cB&7Q9}Q5tLw zIYarf0;~n_pECjFs|20SH>VdMj&>q?o!)c?LYJp^GIR=V(7Dgbl<_C`e9LQ64YAiFBf?= zTAIpmlH_98BZ4O*Hy~f`b8}*rUvU|BGWAi%%o~!UGSiO}^86GS_{xm-RSad->`n#* zj3YzDbO7<6sXNUv*_&xcZ0y41{mpH$%hoQGVGJA!=s!g<&_|WZo4&sNd|DX(-v05- zJ5fY}L<4G%er5x(=-^gEmo~1jd+*tX?I*q6p$3^$QuePVY0#bbdE-#1pTwIo^bl~t zpql`nG|mTn8ITo+?3rn4x_E$pu@@zYszHUlpG{_WMUlVDZq&y$KB5ZjQe8O7Om;3Z zbn_V&u-jfzO*D_&h%a$;5cqd*^w9xtYcbiqd+-2n`Mpiu05bEg;76qZ8?j(7^-n`T z_<2BiL$H3*q+SU=CsQ2Z_+^-08raIgO%fWG<1Tj&`mp_U@~^7sJjkS=XEMc7Fn;kt zwClz^&iI4kG5*^Vq_qL=6ptZNo!3Y3Y{7ng-9dgT4nU1i1_NvV`!F+O`m@G0QGgyk z7e#VPJJxP*Z`R9=XeSelKRb(`| zgdcz-DOElw^XmA~q_mCs#d`fxF=YB&qN7X!kjXNh760f*-RfY8KA#!vn&AGy2@u~d zu*vR^kJ=UX#cJUD-N5MeBl*fL73L2BAc9cMxo!QT)I`|{`0n0t7pDYUU~-hUIPkGQ zqE1_IgrDf-5L<~Y2c&4(8JD1-S;VS*KV+p+9r^F$^9vi*#u0vh@ zw#l}ppJRhFa5DxY#DkdS2o!GxQ5t{9lr`LL!w6F34izL;;cOqTo~ri_@Qf?u?+S48h$jKqbk70%KCr|R z1gM>_vub9q91FHqOjujW@SgX6sv=aA?g}FOf7YU{p8%{p_JPX)t5(*T#Ytr*YT1+E z-w!ZGMwJNJaz~*RJBs`{s=!=;7JL)_U2Jr*_19g`6*}u4gqlwh|^P+~{Zjv-l zI!)xr694>{^zXhHK}h9BbElI(X~!w61Fpe2l0_)-!YD8cm3^_X^z-MTXCJUe)Z^xU zmCti{X8C0Td&0ZEZx=t9a0LCD^J$|^DZpOJFCY*&o!y=}=F=DxK0Ls|H;~}n1A4DQ z8^tSeEa{)VaFEUi!#`$-TcIv-*;Ev-!y3y+1XyyGr^+fqUvVt0j>oC+Naumc94QuI zi|vctwUb+$ibIW3x`LP%*DGu!T;ivP=No{aQP|!(5=EjX*lOX`RAs6Bni~##91TXV z``L&25X4nH%RClfyT!P`{@>NGES?Ks{8E;cOpgN4p+qtI7h%ax0N-GMH(%g=qk(a7tijy-sDP zE|UAXd2fYZBFX?*rZNQ=@W9QaK^{0r_<4qHkT2+S0d+sUn1#NV3S^_R0oI|9!90m+ zDM(gJEQ>SY1AZJ}c@KTA)${XMq7wE@&z<%_uo_W={kal7=>NY zJJWJhSE!W}=Ihb?go|}w`6GVpy^rW8eta3su>Al1cQ#-*L*CX;0BYJ^fd6?xcz1PL zdR(vn=^y|8-~ILb?;lHkEa>n_>h2%^^4I_N_fP-&$N!?#k|hD}_Fw+%5Bqf)*5yAh-ysJ-|3_BO z;1j?9`S*AK?w?-%U;h64E@^*%I4>D^=|BDlZ!<4bp0xk$ks_d|v+LFZr+k^cVIYF_ywG62mEu!Fi70@qho<@E^A1v*v&Q8pn8=mw|Jy&&y}xE?b2l&b>tFtR5BG2W z`0pqEm;b?o{=0vg+J7@O|2nt-`et>TlsP(dC7h8u6z}ei=^zEe6=!)tbqPxFuVx1S zqiK@x0RJyO)1iICKe)%xpwsnJ;)!L1@_jm@^@!a)Re3B`KlN36eqNn5qy%*uO}U!& zUMBD`N;4Vb%P;@@w?@lle)+LjYi3gmR52us#*$qood3ah{vO#MoGYF%60dvT)Lkc zM1@fGWN&E$KZyWjs%!8r=eMxg{!F&sBLb{RZ%jeZQ)h|iJdw2S-@5pobH{O$B;{u{ za8pd(IkMls!`thT^*^h(NYxUkfpx{Jih!q$p7r?u{5$5V@*GF8Z^;|tfq}NmIb%iB z2VoC*>~js?9<+-;$bm-7w||>ucZ`plJ+`l_X>YW&Sb&6{Hb*cNfN@pV9oMGNhAqq)*Lb21#LPWRropEdai z+o6U}k;b=HU8knx5JV6GH9tjoHPo;nKsfsINi#HkUX#Xo=JAJdZbFNq#WS@-cZg=D zuqnj7Vuf^YFAaWfVtpyorFmXae%DI*2Ec@Q(FoY~NvQgyKL9;?!gJ-%lCn9Z#*9yq ziqI?2N(%!&_fGJiY(HiqvnL-Q1m~oQVPl?HB44i!@+{&JcXS~##Ig*b0}A=GH9rJE z10=w3^SQfS-oL9fhze(NGCsE?s-dRNwlG}<=x$jG4?x=RBFAGu|7NdTjpe}uA`N~c zlva!8g@R5aG}TwY&jWvZ0Y1-jy?@AQ$rcpnYjcni<9fv~P+Jkrob!lOV|E6B_DS5)9;+u9LMfBk}D9h|V z&DMYU$r5b}YzI4DW2kchwE3)2o8(1DKIofe*8`Z{&w!HME!>LdxP*NGkf1(=&-4k1 zE;DdX`JvK6Y({FT{W;HG+#7$E^|dtMm6P_MIKHYCxmx8f{aW{Vx=O+ytK!$DvV~U$ zikM2o(m%a1IenUa$I?C<Ji;(;Zg0RErCp<%Lob}p4JaXCOV6h@M75h)saPStZ9 zO+#1nD-C!U#C=*xnX*6Zln`W^%i^$+Yp80lT;(@)zsm#IC6UTzJ?0r^+XM!Lp?7F^ zQ>X=se!IAx;qkBBK0nXRXSP8)i41dnbMXd|Xidz9^L3@Hsm_J&lB7l;}o*KZ6jnW)t1+q)wZXM1-_zm02lpLOBW=PS|60|*pmU+-#-o2@L z_hBo<;a+!1V6t_GQavFQgIB&wJdG!YdiCW3uU9L`hcyWUyVp{uz=JK7$t2~B^EJYi zP@o%lI(5Y@;`t5iJ)WRmfn~yyEO;Ep6e&M`qDgNLJIhcTEHZvMN~AEVJtJQ+JX5w* z_7%bR4Td)qD~KZ7?iy^u?PlZAJ5RP238tTnjuwwWz2-!VTt z!gZi_W!|-Bp4zhA7ga@iF9zA)bp6l;>H798jhWL7$-^)8x+=dIl>lDw^5JLBclrdI z>}A8bfKuI$S(QW(Yg-w%?K68u9P=bfHG8rBowMZC7@2x_pHHgawyly`UNfb7%P&h7F6yXvMC zcKUo02DJQ^O05}&XMuHPE+kgIc7=@bq7OcIT_fW;w}njdM*%ih5*CAoMwlX&Q4@|s z1R|8+{OHgC(sJMRsmXnc*DV3y|8>u=tajY#ZfY(5g5V%IeTi6v0s!&^7XROtER74@k-i0GL5o zJ)_4P#Hy(;-1cR6cmi7)VWpA|jVEe9FHoNUY2jxhp5U*FHBLw?rFZT8=pI*eT}N#+ zhk%bfUh(g>1Y2w!8RJY=fQa^6SDnnG=*XT^2s=7KBv@2FCyOj0(f3!pRIb$`Nv6N+ zs7V(6^II~Jq12UIexKJ;xXLWUNOpD~C$@l%h1C=6ZC3hni?P{7o+Aca8xeOVcf0NqB6&#p3bGv~bnpDXf(XpV3W5-pN@4O#F@@2tx^1EnP|dorW~Y2Lrxr z*)qj={7AC)JH#TH9oe?os*2pd=ZU${Ef=n_pUs=aBe%;pyEqp~$e~3O7sN5SR|+7! zSMhhX_|il`@&S?nqY|kycX}c!Ls5W`FI}fB0iiG4J}@;hH`$$|!Cu1N!nH~lf)3XQ z!*;p3qG$wF(RFXkLZy2o7D3HWy(7C@Aemh>ArLtf?ESUVLdoCAaV76EL;tSqggdN= zV*Qo?V}4WM9J-g|-l%&f!1c7nf3ENZ)9X5tPN=j)Qt<50A5Qu?f4{t-y-C7sT-bzI zx+#q}W?Ir7jQ@UY{sRH|haP-^w2lnq-dRM65}B@kzh$9Jc^I}d>!@bn#fMF6=AlKbO!g~ zZ@(@W^*~Sasu7&pZ)?jC3mX#Kw;nc}6?uew_JD3D_9N4WmvO1x`(zgg^~$aCN#tFA z9$)$bI6ybmIL@O}rkx?kq#H`CSi)mn ziU<22%{Z;(Y?<1hCiWoD8Cp35_!9h%yYHGm1U=fHW@GGNN1+UZlgG7m-;#5)(Q_{e z_BASM^SV!g{_@phom42PDg7**;R$9S!1{J7pTd!B#xcpE!<`6A>IOaUt|yciBpKe+ zs-1dGn~40bKBOiJmpzb6U9YdB*A2+Mst)h~w;d_mK_IfQ$^az9$Q{}Zkahr8mhZX%qNtB|zqPBKozoX%O|1rPCr{f?k0 z#Ykm;WATrWWkrzdf4q2wbw=*Y%EaqZMS|L7YZEaUUUBq#S$00yub45v>RoHQK6P}Cy$mED^@v7+$; zkBUjuqTy8xATQ;X6=!JWt^Zt{=rihag)vv@(TdIGFkjFr+U*lCp#Vp2ErI~v$q2Cg zV|Wv(aC7-GLi0`Rz=Bryg?(v)^5yd?5KN5)SQwzmwYOpt<*mnP)U=v)2}z`Mppv+ZB$Za?f|C9{YzSEJYvdh&2Uu81CMyOE8?`$!*( zgqzz*GH(vke29TOK-$*$aYwfk`IYoS?14h-KeA~WKii>=;Z1VhjDkOdgKj_H<5_~3 zL*O~HZMT9)f|1~O+uq*qrrPNex&*i}_+q&RuL>8XUm(E^bh81|x`a!H25PjSG=_tq z?~Q@qodfwRX(DJ)GY;$hH+hbbAv? zr2zowX0TG%(M6YHk|<8PGgxbO9JHk&bd8-mXw4)Fkn^zTSU?pNkQsSQhq8`GwpLD2 z*n)@x1b~*DCRc|5ri)KlQlksKk=E#Pxz@Kj)7VSf1*52b0q-6SaRz&ohq5*qg|5T- zDm0M7Xwu5zr>G7966#Y$^n+@1Bbj;1SCI!$MblN`^5+#*Fxpc7G}$NM!3sA zwakU!3VH^#o&-<4b}Gq!P%=AT(59${c0Iy0t(x_M(n6Ie?;L8$P=6!&xbgEfvd=f* zmrG)e$d{*(X_TNHEKy=@JEF!C()Ml%fg(82>K;l%ricM;$$DY}&py(xx&Zi*LG>&w zaCBr09xv%8NMzZaiQSqKbs?C{v33;QKJl_73QJHZ3f406lSGEs7a!}SpsNm)VPy_3 zrG9%(9F4Ih!OL)p=W!C2nN=*{tN4SC2>?6#@Ssx^gK$;|#8mmv~M46H_o(< zZJ!`#Ba;d}fkLeTh!Ok*URxD_-+^u8o&_?cMofb^OVi(%2A_ArKX487iqK~o--Ic= zJ?k*#0)V}l3-c(P3IF)hvI5WCGCkwRb1lYhT|Oy6G+YnM-lQp9$Z3CC1e)XG(ls~k zIb3-y8!K-^NkU}+M0a#N>d63Cd=N*tB1hA?d>}RyfFuM3q~Lk{;KcqAE_dx@Ojs(q z!@Ws>JXAFB_|zRPZ+{DUyy=;H1X9l}KW7$}LQ%GZS)xvmVo=d{|H2M9LnVNJylPls z0p&g%2u^$vXFh&rEBDqnRlp0GE0l4f%y-ve?hF_|Si5V7qhQY3IkmNNhjv0ufwCLQ z%dxV*;8|e>EuIM*mJ$78mp*>k0c63JdVr-`N*9>Ies*#)Y%u`2YBrhR!wcYe6KaDSB&|5T zvTZDrP@YyX7I?Vian~;O4-(w~)J63veVM zcF&)5IN!i>D0sG^erTG024h0^^y~Z3rE=Pl``y065=kle-2CcG3nh_e{RP@UZ;$ks zGJg(jtb#G&4N&b)JUPzp6SayAx4$^l>kjMR6%iz{K7^7lF4#f0H&cKJ#^-1jAy;Le z(Bgz`1h8}gB(UQUtsYCR3M{FI<(=ccL>YgF7Q{0#Iky**noeHayPwp{2GVH}f4g~x zo{jK%1YC|T5eP2AmjGMgSNi?45Qeo72+gx}F{bSjPADPg_wRAA@ZjiXc5>;$vkKFM z=lUoZo?_CR_evhm2I0Qto%g3Dqq97cwMUG5su#W~^5JhO$9g?;2hL)R;tpO3wcoJa?%S|g-A_)k^*GrX+JlV{? z$6NavnH%856!oMa@2`@3)040?K(uXJy?B>Fzm)QYAP{u&i*4&j*1f*&+5>(^M-&_l zWGn<#YQRAZeP0Z3o-GEX=;4;#x5$T{YeESlr3)(vhc{9fqu%s@-a%VBg5YZ?@qKNC zJ+PXDScg^fm6Lvd-F6s93qhd8`~}*TSGOn(AK_gLL6qWRjL9qomd;j?ctfjZ5%5{S zu>hbw*A?v&K40sg^>2MEl!EvxUxvDY?j3M|R}6MANR%}$eYCLWdDr*%kLGfjem|o5 z?AF=rG9pr|DfzBoj=8n#K>SD*mPe@Z!*224aUCZXWC=J;UskQ?c4+Px|2Z3l#m~R< zcvwF?gisEyosUl32>T|u>D^V`Lily?#RksI*qv1rSVJL7U|$w04Mmz_YzB5#0OQixXnWV&H{ z*TjPSJ}Sobdu{ef1n>Er_NyQgMs-~An+L!jm|G!I`Uk+!b?J;5=UU!Azpy)uVPAJj zSy6zGGn+X)?p58%&@6c#nd8mtUax0eNMxD(xx)(wFgVu|;IUY%=sb+5Iww%@IsvY& zak|D4w1jVND=f9iSZpv%^WNWQ6?$q0aZ$itLil?KfB{Ow7$4uBoW7SRn$207lLK8r zS?I6nzs9}Q9e3^x{7YSKVA!c7>s%7--~Bhx(7}P>a6eSM3oJ$1`3`X1gjlNbc$BUe z{9o*SNt3Hc7u^riLQG;z%3U{_F_b;f*64cDOgMU)8G;WZpdY zoSSz*Z4^$3c^G`{#Acm^h-scV{;`*N{zhBZy}=tfW``)kIZHxw0-GD1AFqc#8i>-D z24S73U><5=>*jdU8BC2gVC;@y_{KqgX`tY2nkaGtq6j`no*$HWm0^7Smr4|yBwfZ? zeCOf2s|Ro{FG54J_eB_PQzA4w*|-E<9tLGK^R{TPJ(MtfAFnda5;n)Z`!XBypg092(bEwXBWbc&Uz&MI}YU=dCD9Alx=QkuWNGvkv>x9=)Tzg1t~Ye_*Nt zhhd$^K$aCz6vjOE$YnVVP2%BIp2lXS+1;RRR$OmlcK|%z1ji1u5cfYTS-~+=?0ZbG zYyyr-{+_Rk2Te^dj7e{?$_Nd)*r@yo;7+)_QGW9pB9sAKPyeWo|?cm2lV=&<{g^-053g!Lr%ojr8H%0TGU9 zSXcR4*>8*>t!9EcR|Q`aSk{r6REF_s^+4X9SJ`8WQnAc@THtR(rm`D2Qan*yV|%(| zIeIV57MdE5KO;iHc_xL7t}tK}%R$dMV22Tyk_~kJdT_8T2l5>wvIkFnNKGmK#lUQD zLH7%=k4iQGtE}k|XjY@YI|1-H-2n0YoDKXpmECOgll;tdB%tIw1dLSVYyA%J8oXl| ztbjYYaH8oij&*`fP%OqrJ)jyT;`$lKyaECQ`U_d0FSZoF<=Q4q1IdgXEmsbIx$%1X=d3SN&3Rno3U=i4GHDP(15a4nJ z117PI1J1{dFXj(JM5hC5UaykBNL=j8`s`n@d!NO~(jRoD+wA^_eS-?)ityjW)|uZU z0T)^hv5N39G`!AK%uOoC7x~Zg1OmR{wKO5iVNUtF%i&Rb5KMJlUPTkW$MJ<&4E}Fj z(;be5_<#0$L*}fCP0+GvDD56YUIb&j@SG#Jb$bL_zu(mUfqN$`w+dd*3v4G`z6t0l zo<<+bg7>*ys{S{syMHoOw2cTXvHp~DmQf^k7rWR9H4z_oU3#9*fL(|4KBB%y{P2tg z?#do}q7+(wr>$kQsD?-tytF?URYF}3@Vmm%?;GG<5XL>QjVD|0lNVF4-0fn8QFL-Q z8VEH6-gVWmJ#T@KCd)l;@mdG-`?GE2bK3MSvApTbe}2-27+?SjVP{!}Nl*hcJJ11b zLZj(7`70r}9wKa|tOD$3SHM_cYg>5;NOUH0UK!3CQ=y0N!QUr(e6rfMX);Vrna?jw zb40oIn$Sd!R~MMS-M~^KO1JHJQ+V)~S8xwEh<(m*xxtiKQ^wP`OlWsl>Npw4GlK$v zxvIiM$}c#|eHlNR+&~hE{@2p?W%SG5@BIYMw0w$9>1n!o&aYr*c%TzXBxR~?{Ft)q z2OJ!4c;?t!c}~Cb$r!5~^?XAz4*oK_c6Pg5J(%DE-`^48wLL3`BhW=3eA*!CFeOu<_zaK>eK&@syF37G4Hq>Ve_OD6nrhLWoh>eo6Na=s|E59Z^x9B|FU zCtO&Ot?Vnqd^tWgj@WR+Fd%#XVpG6HD3*1()X+b{$?SuGQt$poDbODeODu3N4E#dC zloT;4OCV;E6vmi?Z9rfvV+BGWf75>Dqpxm%0=xO8Z30#NfBgLYU-!NHfau%_Q`iM~ z9RfAni|u$=rczj$Jt#aUJ|8b3H)Mfhvm0fc;$bM>8Zw^@`_+wF-o3lYQID2(h)8hk zseEc&bBA|A7hwOn4Rei_QLcbikW{auLcs;n>ix*pPDEY-;PQ{m zoU!e3yAOV{YdElKsBo|Z;Yul z&e0U%jh&p(!%^PgynM;KT$JB#k6B(8zO-e7cZOrKiRqUQ1u(D%j?Zk8A(eq+_kX!3 z546n&!UA$>gehHtK9ct0r9C%>MUbpvf ztei#-J*&z3};X?$BoV z*kW7jDD1I&C_;lf@cy1P_}q%@-~a#G8N*m_+u&4l{6<7cPc)5fy*tV_ULW=ajzfX> zy|}l)doO!8I2M-CbH7#w4`9Ri4)v%!%s+m67w{C2mLhR9o6IA~<(ETN|26)SCgX>I$n8d3H_ekF^rWGIS$Bl`&OBYs+8C8L&d`rWyD9ckbq2Y%k$6UJz*>!Y5v zRnND`_$KQf|H{OGriM9~>O*XPZSr4Y>wZ~yhYDccS)*)5j?sTddhI~Xv=Eo$8AIFW z40SvF{*^tCM_Dx-%WG)5BKH;hUvVb=zvE2ge>l@+;XGyZp2WraX}V8pkUgBO7YFI$ zCwatu)A#UANN=C}bkWpb!G?FlZ?}1!=L7>lk^@|W%Af1iV0vQEovh zq0gFu&pPOjaEbuJ7X=1LUDWj@2-KGFBI5zxy*%Oh$MpWZpH3IOR#h z{BjKZE^g;*crRD3YhfuY3!~$8H~0Iesw#rP@v&2&_M>h4+%IRcUZa$N`xPic0-F(# zV5|1JnP1<_8|zq+Q^BsAO3Ne|9O>fnPJ4exe>$#EnD|})Fewb}7$#Dwecxig9NBvk zGyl83H!ObfycHXoW<|mMbR1vIl9K*fZ@{+ZF@uo>w9k2Y4=@K`ZGg$+{J8-S{}ppZ z-C-Ma7WS8d?ap*KCIqFj_m2#U%*xHuSp7x?ByC{>^}rzw!AfeY3*- zD@{6B3y2Tko-Udsp(Ohv^YQO9in4Tm^B%xi&+QI4e?X)r#GL02Y9OtVPkB2KQ|lj; zL^9z^bdjPZ44%7KBt@OvPKfLz$x(9TS6jdU&sOEnf-!Hjd%>4RaBT96K3zjqx_3v? z?2~hcqbe80DIw2+?YqcH{^QQ$%Q2ma`rUtJCH}tLRIlwC4z{4WC~wNadVd<>MBC`` zl75S724QQqE5|y3$03)GArPnWH80gj*b(QrFg-O(dl%y*J`ko09K8wa1_adk(EvZh z$b~Uuc*nsyoZ3w;Y+u7+?^kLfSc{ZGyoTlk@6Q)EQGXlB3!8R1fZsM`#~j?Tvd0s# zUV$|aS`uh)`4b8Ag{Hd&xYplUMp%njU|K-VvaF#fPgglrvJN#k3`j!C!Sw zXy=bpLu?MhO9_m~XgP$E;CPvlLkhzK=4-kAt%y1r`85TS)0d0gJ3RZ|%ChtW=(E+< zK_nS0ts`jd=Z?UU!r1l!(u08)hwoJZoVc4u@G-P}XK1?a!=*p&u!Y~A10wLjDF!KA zKasX^0l0nnm1+LY=YQgpm<{|^8bN@r$a#*vD%JH$L@=&n799=O1swfUDLRg2fj+_kp0J|!t(Xu=16|r3nR_Q% zTmVIl6xy#t;D#;orK*X)yGuX@VxHyy=Xz%`Ug+=Y47{j|(KYzQ}TWztusV)hF8;gJ&wtSp(`&00WKDSp0D} zKY&;`f;ILknBVn*cZey5N7Kay`y#Rr5jC5%9ADUthBL;orT8qjb<`iInVkd75rCz? z1^o)#IdGjhc%aeg{KB!LJEK_m%jE&P|DCsQ%)nU2Z~QfEi-UVD%>B(RmGijVIBWF_yp+fu z&D6U{AVGe{CTd5dVHB-*@S>Ok@2m-^1OHlY(jWcrZiW^4C71z>7=WvQVTO}0OtzgV2$(}EW^U*__JHNY_UkhUZA3p0ij5#;YJ{117{3Z$3} z@4;lqn@{PDwZK%uunA#f9X(hiSgs#M`XYgU4&(JA4-g%zeZJn~)lu<9Y1~`nLSA3*j=F|AUmxQDmwAUJ%;7x1DS_%uE9AJW-f!lzQQrK+da`?% z);BVab>U&IEgWUa%kGfW>SV!`@JnU_9JYS{R#XBKaN{ik3|}uk(Fs^9&u)5s16r5- zwb^>Z6xHb=azA`6(dd!?d|z95(_xU%SAl0z#UH+b3O(nxtKo=52(jVYiudK~A>oJEjW2FhKQ8CeP5y4poP~*>f>d(MoV4f=S z-*{zt;A(yLcoMFMH0lm>Ay4V~Rhv{#_6EL*F; zP*M=rD~_ksTmJcrW?qmYh#`5$baNE06iFVJ+>X`|>UlUq+9@uVD%>dh=g3qge>06s zegXibsaZUa#Q8jlO|2h^zP(;34`|^d;EJD?5sOHz5?mI$dkx3$3<5as18~&pP)|IM zq=g~Fwr+#KHEeUcUTvA2U)Jx@T!1KGy#Z%@?=Xf%nH%n1XnCyp9cl$w1~0&z_Ad@| zfTJ1l9}pbS!etcjqMIVN_z3XYd2!s%0>9$89~-e&l27tMM?mqlC!y|6EyLLXm+^OxC z(=Vx%dSZzc1%56O@?GvoimujE9U5od$4<6?e?ynbKC zJoH7bN#hUV%s7`W+XvQ8?5cAYN$xQ}<|66=fb;JkiR;QCk=I_;uKv}iA3l#KPYQ5o&rPk%bJzdkn`jyb*L@}V zx{GD&ac89>b8ZD1z}NxrQm`f2Lu70dcM^t%t-c~afJs@Otf_7|e8syn6aLm2qOO0Do z=HBOaZW)SmLlZ3()em~byMP2PND$tgDO07sJ~&cv*j+R)-q1GLP}R?Qw&S=|<=O?t z#)9)e$%yPbL;867HHt<4!38h3D2hlr>Cw3tSahK#T>9Im)v4QcyJ_-AR#8PmRzxd< zjgiVC@B{da@NN%TqfRj%UGD{|@)klCd7TKcTCYC;{%kw`~DzCuo0SWu!40%k?Vy1 z?g*c@fQz_; z6STEp*_meyH5bP@WHiCFOcgMd5yb}v+UK_;TghuKwrjA6IV$|)bsD=F1@_yxwQdLt zCYX%^-m3Rr?=Sj!3y#lnENp+;SMdJCsbR}g^KgdyJDHc?zGcrxH(B|#_^XtrYWlp_ zq@vHRG31y#GUXCLuDg>QS6>J+g7G9)U=hYTkFoc)Ch)blyute!lRMF+ z?XtNdpX~Fez1OCb^TCpx)pX7*?*+~sg8X}G8iRjbbPcmMEVW=SrHv+3>+*(ii>aO; z>ohwtf5>}Y`2=5z!Rga`tRo{&(l5pR0l}aD?G}K zDPl#w_In=nIk+ZIn5UTmJG_vo)$!ha=b}LW_xth3)eHSGkuT(`?CoBJ>avhk$JQ1*&C>HK* z+=)B>7Z=nXS{BGp;(bp-d7NA-VBNo78Hx1)M`i~A^J)^Tv+QpUeQs;~vqr`kZEh2@ zlZ>_ixKqrEhAaJd4|x}jm*ofQ`55Hk@o1K$WTt%mYU6n^)vyyM;c0o=Jxaq!l zk`ivohouNR82;H8@yJD1AG4QN0L?tr1+C1d`p-Q`yqRepr&-@KGyrZ|_|&MM{H=xR z1q`uo-Uz?;`p?(;=W{QZ*R86!yah+WPc_Ij;P9`=KeeWd7eY?QdfAu*n6%6THwojn zjnzvIe2y6Ptb|#$-wo)1R#*xeUO1_)Xe zAq%^qtkZz4(>$P6Sg`bSoUZau-;h3j^IrrKQG|(_NOnCI_l4{k@EQY2z&2it3-AS{ z%#pfW!r~XFg>9Cd9?-ifwiS$GQ;2=mO9?vhPQVf~CXDSzR<(r$D1NeKHDL<`tr0Mh zi;`ya@mevRuN%7?0s)w3o44wrZ`(@2UXR_)iYDO~ zI}EsG%2r!?()i~rTcO=xya33m*eo_YZCrl|HrNvdjsis}n3oUxv$AvW;QxAU^spxd z{9695P~Z~@ktHl)nVS^&&?2jYUmodW`#P%he7hOeZ`qz_F;;->KFP+8iBM@SpNsO{ z^A`4qgN;`#9(Ivr`y>e3BG16Ha#myyBnR)Wot*~ftIG$X@88(MYZG5yguq4RD)dHg zBCU34-Kh@#+RK(Q*S*w7A_qO0W8Uz3rK!7B6>7Z#oyXryp8bqfRUX=xOJ2?#YmWy) zhn{e=rhU$MufLTF%!4nL7dZ{K(k}eI8vLcI-UtFdZo0|;Ci@9N*F*TjDPb}=mPc~E zl=?ykR3MDG*|@Y-q-i8m)0V)|w(@lXU`y}Y<$_`Hko=9|fYqTp%zHSc@dO5Ec_Rnw zr{H)17f@PRyNbX2j(h^#njKbQWz~CacH{@>1lximo((xOWZCT!LkLheg6&5ISQ9~% z%gomTL28x_>tPI87kSQs$$i)Bl_Frfu455xPrT6{m)dZexT6wSGqvY8vNm;aL;kDxFKwl>b?~*yRx2l^a z>)&Sn4&*8nv{vtHZIjZc%mesqx@?QGVfiaG)43$HmU94UNxhh9AB1;D z^f4a82YWS~B=sZlqV4NxncYN`Kski;nBw&aH*ql{4Gghi@Ubo?~v+HRrn})iVMl1I~wX!e{i6ID8 z|0?5?=X#SGQhM;Bu-q1KpgA+jRk=~$snZjaFC2%h<}F@KEx$&xlFaoWD^J;J?%AqF zSLQ3Yn{@|@*JTaXkP0=m`nArOOE^V}JALKnH!}gU9_YS)?&0*#n$lN!>%$4ATcXaj1iwv7FzF20Cr#DJ8J%z-HS~3;G-@ZnGr`LyN`VR1cTjA zK|_~Soj%_|ucU}nzMH17%0V`@i&1J%ofKN__NlKQXTqdD4Wy<_QUz#VSyaeT! zr+@jNPku|@A3^|n2+F3ia$oKDA=pT|$+NXjUU9oFJ@4E-_mxaY*QU7pey!Q7F;}8E z48VSR2~(V>7-BSK%FN)reQ7v-PYISR^b$WoE_F!l#S&w=5Du_Q*i$?$s@fkP9dFe! z%{?I5O5vRl1jf~6TYYCU^v)=mjnn;P-C~V3zQ5$SkiJ0wMAC~p#iV?+9?4Qy=}X9- z5VYmV`@?=2oSK#G%f4W;-Z?V##fSE&TBDUt;@fq}x8PJqv6&ihejH18O0`kCZomF^#-)dFm#33*%9u^)dP;}|d?q}j zt$GJC|F_4|hc1N}a;vMvH5W6T7|e6=WpUz8|eG(->dU%Uuk-c}Pvh6ty2OhITL#o*9&sew?(- zo229IU?bv@-|wh5@Uu5amQ9HSjBk-EgTB;pByJ<|@JC)Fu({h`Iu=d&=5jH7m-|dc zqa@o=?}vx}n6$)MHOgT=^@J*KcvdJ?dR1tV*=8ik2Z=qJho9PlHzym>z3we?UsS+q z8rqB+a_3n(6fcyCQGF{w8+4!RRs9mc$KotzIGcqwNo-M_KGx%DCW~$yoyf7HEBX~2VRwykNP)2d z9ur7y0<@Uc<`=-U*W^+;XVRp>L|~zfy08ZdZk594><;H{2Q%TwVYBPXGQ>~1pX8@s zW2A}a-(fbL#Qr{!@%~i-4HRkRj#s-8#pFsGwI#Ykke}s2n{MRZB#8YMK2fn{AgV3#9mPp9TRx1~gUekE+G=H|zH3-rHOuANjsok>Rldk2chW zn{Q$TIkQpEBQadtA&8VN*gAKpD)-Z1ep7l>whiwzXZH7-bnY5^(0%_p0Gqa&?eiH= zt6ThVaTjuGfMZ9%8c?fY?LeeS1&kl^1)aMn`ITqs;`Q~p#ve{U(5r9Y1!eM|x3{u{ z4YZ)WCml_(&MngPXDb!gYtCVBa@lzdXd40g*+deeYzvyJP{HS1-KMDSKs_pVXU8P9arv}G}QNL1~Z9~(nrGq6Ff})_j?85j>Q3yyn}CI zJ)%j)B0D2Xgaxj2?8Gs;zd!Ejtc1U?mOwE6k+*VHPu2CVp;2I)=DAxQYct!oW`VtVFVVUmeEk7;Cq%VJ>eVs(EEv2HXX)mMHaHiaM**1Sk z#EV!t7R#bx`HL0MMt?t^;Pb?0bY6{hhw*lIFcs}Q%PEta<0hieuI9Gjd9c}(qgq!S z(R`Xj`C**vvq;x3f3`gDFxDzl84JOv`c}45HRGDM^~L2jm>n@D z+<>Xt%bTwk&B0$KWEzVg&jxXwQ4Z?7Rfi>HaLfZ7FU6hzmf4*yQb;6y^=8!B9(SuV zPuTWL73ceWKQ4JNd(!O`Az|2*F`lf?cMP5w;moE|p3b!bii>A$%zoq#yPAbdI>*U& zyA|~99NV#$Zx}P|OfNq@-?K=|Z42E=+yNCjq(CzAuI>U@-#7_G zJeBnrpsWvV`T0hlro+;dz*3koIyWJHx#Mx>?{JIo%|KHoW6vBhUkF(^Fy0w?Eu5wX z^I!bp3dh#>4adc_ZAz5BJ-AnRWBZFm?oO3_-$IW?$X)Dt<{NJZk8A)e-j!Zk)z`pt z#!-P$H39#AtkbsNx;keq9%Ucs&05M5h*^LO?>+&{kA+DLwkN|LiwbNSF|^tH31cjG z$CtPrptvu6^Y(reN_Oy1?4tnJr4VMTmR;?vG82>qv5H$;N_S1V$~#SHnZK`;E|fUa z0Xnd)siIH)veOy4Xi~Bu#ej>^4A9v`EC;iWe`3TvdL#J&qlzxd-UEsfH-;pi!x~b# z1yiD6A{J++TNH0_1QEGx?{rhx2G*|?qV2speA&k1$qh=>xDT44*B+eEj7SmZY}O2K zb0C)evgpH>QR0XHG0?%*-@N}B|Pohpx<<(2iGV*|1JG+zYQOm7|O@W8#&Wz(gL&sUC6 zAj}w&%zpPtWPehyUd1t};zEhFV<-SY?s)bIgc=T#5a2L0--`d+{B{TJZ%tD3V>Nc1 zJG&{s>E$`*()(h4P&Vy9_d|XTU(3JM%lb(UxxkY=;b{84wL4CgJA>&_MS#oK=QBA4n@U0!Y-NRH3aw$~An^}M{e!dIvJOwg^J(&FObZ@TC` z)e}a~K^HgWb+NhFY0cc(cU-o45mo4Z7GjSmjFajGRfJKJ0%sBHxA}&-Gciq>4zR(6 zHHa;&-yamAsWD)Wrts({&F|ri!sah=D8?n&dfeKiJe*P*n<3e++Vw3OqA-v;|)37CO?P zq%us?!(!XJy@{(a7Ip7P%nje3_q+JMk&6&M)_VS4P<4KdH(c!BWy#gt((dng-Y*vm z5UlCr^D_0UwUVUu*lC8$ckxs`gqF#9)m1JRimy2$LxGZ#&nDe??Pf5BT7vy?apyC4 z?OK0ytQ)?!au!D-Qq_s6N%c&|hp3?h6D5$)sMV3|`lR)U#&GA*)%><#50*n|4&p6|pd!?** zx~tQdRB?R(UyND?ps4dvBHUmuJRWld?Oq123W9?Iwi8+3pd^Zmg20$mdY`vMd(?El zcrzCRrVY+@zA79gen!Xmn%6+;0pifkrBqL&Nj?bxs6DK5mfF#U-}A-nCQPb#%2iz3 z#E2-icl>*k{?!iagU;2$cE8BB2{-y9P^ztT;)I6Fi4kW%y;)7DTipHZ=*JsS<0AKx ztCT_CZ(^QqmU@Px47hxzOJJRfSj&M1a1-vcUrMNP_aZ*c{hRx_jnV~x<@+PAkt~#6 z_4TD<%|VJ3)C+#%A2jmW8M9Jy9$^#$DV9ojzxdn!mg01Ko@9?-@x8JS0js`XwYvSA z`RpAt4%!aE1|-%XnLrBAR^DN9^VnHa)59 zlkifzxVW`TC5hi7bNEe=D4MBHGVq;S6~vYk%%?&%M>BCAOTPqB<~~yB=BWzr6|}W} zYimfLdGXb}PSKYH@Z9SmDo36<-$LpGZc6w)6`j3VGQ94AepPTLbq~X9O(e|Ju>rm) zqzPbSh6Cd^boB_tzr1qhkN><*cF4W;!cAE$p+0iIv54WtTI$f2=uxgvqO}3Y2v$bX zs=1i{5CSY*WP}(ewn%Pf{d(~`z$6Nj#)Evm+b@hC3e(L^{g|QA@2Ea0!F%gDd1Q+E zVg-hf_x=$N$GFpFZIPTkH3LAqmvymYMW`71D5T*i1o0(b&8xm_?Irv$A*p%G&Ehmt zqvE1&)EY3IYJapgKFcB{;Y7;3yx}r1j+D6o-T>F%I!4R)Fl={}TW;y5@kiSa2-dggQ-esE|HR^5GHQSyKl7n}L-9613Nv`nD|8X-KEJuucN~bCm z0w~z@o`K}!1qn_ymGH;5NlaRP?Q$Nbe0wy3W=hY?Y7>6;-aie29p?1}D_bw+guiH4j`@tA>3=t;W5SmD9XGN+W(5cma>^WKvr)M?GY2e(&OzO^eWks@5Fm z{pHm4da=7l+f#a@;txT(m?Zt!WN{bw{pPaYwCJXF47zN-yp`lC?eVa{z7x|rKcgY3 z7rMY4%C<4bBpeohH>9$z_SY6dv5gSay+*5@Og8V_Lm?$u6nIhWDg^?NdqUt;=*|4n zzxd5Le}&7^W(1LSd(;z~v(HkalrVyR+W9B#e;Vj@R_Cnv6CUpOFzGkcHLnV{pgGIh zG`@0vnU&~9sLNEn>TIj|+XHD-N14eqz%OK}KiL7$3dD+>D4Lp^buVRGFViQ=_BcJ- zLT`Kl|CA^h@>W*?3oi6=Tv4`kRfNOch&)>z?M7|K{T&3W#gW2;UO(LmG}a>k{wt2n zuA{BpC-R2~J{r@efUj3gkZ`R6Wb3`A!EwuzQg&5dmHYBa$*=22- z6A&G-$ejC~xcj|8chA~7<0}JDVDScS&>b4Q+eNyiW(qWd9{fF%gYQ*2P0FNYy^?^k>(7pjtxF&1$4+=Iq(n%%4cd95W|jpsbu`cb z_BIf7d?$MGqel!oEkW42*xS&w7h*WL@%VnX^?XOk)E`V`j?Ud(#{zG(Y}`M>cCH_= zTVGB&4ttMBVxU^$RNh3GjA_{(g1jt)`zDggxTU(ow^V|C$)dGBoWBsTZbyF1$?^2p zv&x_^T+;`B3L zjFHvDZBg4?h3x&kw6oT#h~~#js=Q^=GW*>uLxt8VvKrz7`0*?BR|&8= zZn2~qZvkhaCwOJ?@vIV{-@db@o+|bwws_gel%%iW8g$k7X?hgFczSqcOmA6A&rINC z&vg3W*4v)w*V{LBFRsyo?S?5CGCfOdipZo7dgX?v`NfoT=4y(moXe&JfM}O%LZ#{AW#4e&9ke8{e@t?6xqPy6 zrhe`zr|-97-|10oWgv+xL#RR^H%>&}$Z*yp;x=EGiwyWA(5 zDpeL}SMVKJF`u$P`vSXlQsZ+$|vdK#qg^g7)C+tJT=!Ex` zE?himQlG7L@{QZvwMG}=BqIDh>nWP0U-nE4ArApYyEJXB0U$*^#PXgJ4xir#+d7mN zqA^~1SKBYk1HEDI5rezH4VEpzPIuE*k3Cl=Z}@2PND0b{YSQPtWNcD4`SmUhLKlQ8 zj6J{H>CHD`it_VZ^hNXa0U=j$APaNWUw?O~NzW%5$sfKUJaBf$w9$?=1FaWNQz#(I z%uCUCl;#Xzv4V&xcf3%Dmy|7R%)2?t}Uf3QySrRVH~@{4;95}LjRc;~p} zQ_j4yVb#l{R;`{ZkGBrkYU13e2as3c!k@ZqV#cG(WP<2p1DNYkuc*Va;j(0L9s}3I z?#JC=@LppKukZ-1cQWE!G1j5ZXV#GCWh4BTruW%ZigVPxs1CW6Vq7AbdY&qB;lQft z=i@50v;SD_TgqP4ir9-WkN6w8)uEZlxsht()GZ zY7r&yIsNf`d#VPELM9gf;EC0Rw)6~UJ*0Z6G{dMBS>vqjL6vY}h@PWwD6uRKQ`oiX zwI)K^R-P;K)Szy)#`=4RN0E=^qs$o zyUI|O@=hdVi8HKrRqT8g9U&!`yG@bPIta*SaHEQ}7Sx)ox?kKI5BN zdfh5#xZL&$F$MW)oAk-vOE0ICGvf41D~`$q==-NQ>JIV+j3CaRj=_e~5Kl48v6u;r zl)mb*f zY-Ur(H+597v3nwhbs2kFGWv`*hoy!=(p*o1y%;dc`1r4~2F#u&6l#5Xvs7I6n|eNP zVkKsrjTjf8Xf^R(x-H=uEWBq{pypulbzi>Hsm_b}3B1~`Uc^gjl%rVO>vZjo;ZZ-i zm$g4f5jXa?w8&L`-Wxp1wrDvEJ4T%8cgLY@k2}*Ld6~|drxM*GY~M+I4Wmy&2tF4& z@wi=GG8oiwnIo7(6V)(oI_j8v%?100NuOs-T-rjA&b?zos^xn$@BC-biFg|7=*5C_ zi&WkRSPJ{TSmTJ30Q)dlSh03kI&>!AbzYd7`dvn`3{W4(+3rEqst?4lP{zJ5)_G;FMKopIXdfVeC@d?26J z*{2%tHp?TI-?lIMN`eN`SGMQ7mwzNYecm29F}qz*8m)aC#pK_v_+wPXzTbV|{$!?S zfn^3oUuSz~moTM`=8#gJDmUZ6;pY7^~Wu3s_XUgR7tCO zZP`F7dUpi(ZsUez?|vyW^o+;v=!H2j_w;I`ZQV-A*QxE+sAg}J4!(a7%wcBtVBDI9hWkY*wQcWYbHMD#G>ztmBj84*0*B9r%q`cPQZjk#rsW7%?F zq@qoA!b#YFVeid)URSp5uy_B8gY(MdihXkt1Yl2!6h%>Nl>`R+Mq=MdH1gk5#kG%f z?BE7=;0Gh{0`^)PpY*AzX3jCH<{U$5=}c@}cfbwk;l-E2*`s@E-e(-k?pR)2Fbv?} z@CLYU@RK6E(gfGpCs&+j^vPE&b#bEtNtJ%H4O#eQwf0Q=)d^@v1@mvJOloFDMy*wL z<=d}MwV~kP&=__;U|iU)$ejO$ZM zbQ4R<)sU4Sia+^Ejm+%dUpgyU%1oq*pm{j2fX{e5@a3L=F1;&f-tpd92KHdwwQy8m zOJ3(T&>Qdc^Bu9$%>lb(3%I>E^oo zQt?)d!3m5}kY!fs#K<6%14=R=_VA!aW_2=@7E zmBTeFdnMt2A6j|{ul8Axues-|2eCg_f(L}>Qg97wRHYJ6Qx=9JO(_oe(l>)3GIbD5 zo-eOCwSBEV3U}o4aaDX{?GcOtu^647=) z+p#C?#={hc%r)=|(096YKL@pzdCd6nOMSKb@_B9S9sh@hilVpIIuVHv&RwJj_PLX) zI@3hzZlA%hw%9WxHdM*96o8u6>2#%%~<@v+JhcAO@%Vif*^%c;)zi6^Ctch zjzwS%Y|mXkzS#S|IC1XMr<#LL1y3}y>-jxW!s!o1kFMI3XMGU6%`i$`k4 zKK^@Oq`*&9cEI@(>jDz1?mw?$x_~?T=w4vdtD5G|zFUL-n*XmFYi^Z?=I890&V$bo z>C%HxpiiRY6Hgvn$yAi-3?Xl@;v8y%MOnQ^$;3=ocf#T=w24%{aF?CXY+H?+y6wQ9 zFdY-wDkJDm_E8);e%aklbuYY#@MWl>2Y?bzrBw#VHj)1Fxj(P+udpkej{$z$$<&T6 zV67F>kL~sMGpq6BwdI*8%TNzlK)KUdK8Tf#U}!Smf{|!`(1R(u6VVJcOv$|IYv|ne z-LfK1)@6XA_`%uZ1@-$Q)oOUyc24ezwNo(E(KOx6MI!WQe5Hxu%B)<9bgoc6`fyPt zp)4(u9VW`=eV4X=%T*qtIf#wq`*ra+3=o%|j^MRE-nt-^`ybzmq@~u}4t<|i0_;Ls zo%6wsNg2>z0zTbMgZs-mb(x?e`grEhr@WXq!WksQee~WvvOAQ9w(NC-G2AGki{Qdu zIH~Apgy_D2T~bQg8NcLHvpFD!1_%&VOT@HFNYI~he9bu22*23Jad6-{?(MF*qdXXo zw>&I!G|`s;?eW`9^FuDJ?qrzH%f3JYCO~n0}#H;rYJ$6;`6G5CwbS+ZtY^=Ldu@H z@I_;4&#F-0`Qd#G5o&Fc*0OCZz(*dE!2^*B@5 z?T;7vTL0P8m>M*j8>UH)e94Oy7L1B)o6!f=b>0(BKQIofC+yMN(vEyzHz@8_&y)D8 zys?u{tuRLrLvui%5B>oy)X^JpK<#$XYoFg6o><7baNF{wh*v;u&6nH_@aBk(kp!4% z+0nqLc^Mk`>^$OJ@&&=pgtayP)ZRicgs7oN-{;-74V^eDm!MxnaRuW#LC-o2Q68gw zdN!`zCzzD7g8R{dI=e9ue1A6pYRkb$)xo2SQ@?|7Yvs|aNKwr7tudeFHT1Xjm=iE4 zZ|d$6TLUGoq;KjeNw(r3|H}doHgWlNi1JS5n|EsallRNkKyQ?@M^NTZkFeJ>_^i`H zuoC~!Dh8xU^iren{h=@H9%uS{103wvDh082G?y>>)9W0ZOz+aEJVN+PYU@9_B=5~w zdvtbRJMg>}H7m~$pgs#0Zr1YjnEMwH7bPG+98x>cRx4_yvRqyyVyJ|HWP;01H!7D<)oH>EAYNsz` zXjVH9=G^f4@#6%DX&o8yQlIbEh{|>opij&D8u-=Y?OxYL{RK+lk1y87Lfpc7!~_9Y z2*9{>lM748NI;%nrpc^v-HY0(PafR@8%@TDg#Ie9)qTHpSJBJwnyknU{{;B%{>|=y z0bnoY0=B%WOHfx=f?gP~|1`hfE$hqYTF-#_tfThTUDJU>pzkgKpT|h>F_D2-9)y3^ zvy5GzF%-_O*r3H~Ty4v<&ZL|}bFH!0LoKrl$$*8Z!#WPyx_1(OZKALIkQJ$TWiLI~ zYLeM5!k)@tE#XQ-vUagGl>qaX)9&*$9`WLh6?i7JL3?Z-DF8yk03pmU(^h z#{ETu2DG_YI2LJ96h^TaK^Urm< zbC9PHD!@x76xZAewO!uGgf!&$+Wf5Sk%FC?!~P*Zmejl*dPCY-o5y?~td?7=tJ$3> zYKQ}4%K#LQ=%mL#>Wy5_!9Uo(XaHVd@gfU1mxp4g-!sNSbE7T%+`yt&ziiep9L+%u zjo*jp-s~L_3@LGmf>AFa)JxqTkpgw|+Siy|nY{S*!mR+v3;G%NS{4joLH35_j8+aE z%f@b(rkX)m8)^U%#p1i$?C~~3k~;^Ma#64d8{F;`7cAkQ{dAzL4KU&TM;HDuN7+h4 z{d;E)EbY|i0CO>NehoFgpkL(;7d{e7LN~(W)k+fv`0Pfo7m!z3X8PRDVyBzrqRjn1 z(lY}!M`|5EwR^6|fHQn)iu=l4oH^4XT(0=1OC-qT(}>zqb0U;zoLL1$nf%75l6~xK zu2mF`96W%^h1OJr%cS>xmISChU!7c~>u`o+?a7ZPk-*zt`b$SJ1kk7t{!<;nxY~bn z;Spx_<-gG1Zz4wQz~kiOk62v6SO)c>2$$%r^u^`lWqTt7i}kTm)Q7Of`{i<(gO zX$nL-VovR3)cAP)8;iMEe**ed2Y@`3AI(8?PFe3QZ39gBh4GZB$(t(a&6^5b5X^;# zu18P3j@}NGg^|U~?kYE&FgApyUdn|*F5^`L8A_eD1KI53_d`s$o@-dsNsh?nQe8RZ z68Ufi=ry{y59n8G+x9bOw}eojpUpP~I1u%t;9A1Yx+Aqb4C((Qde-U9 z>C5LRj`xXLtNA9-qv<(;wG~-7ot8;H_tUb(IJHh)^ZEFh3`6d!G2FV;;Cr#U_;@?e zb3#FV#R0$H+91}x@~nH}^ArY>>&~Rb9$8s<;*t#H3;2_r;#YWIIh{1*l zb&HvbUi(w1D4wvq8ql$@r(s~t1jpvn?|bm(FfH8P^X?&c<9S~iXE33$C@S*fZHUl& ztRjS8TInNLG-%AdZl|xtPiOAjx>p&}d4M*F9>Dl8#=XPl+Y<`({B%HsX;g}i0=3HT zs-mFJ#_4f+NdhdlkM9@tf9a$P22E20ak+#5|Dc)XWuc(29lf#b zf$gNuKOTJrb(gL%8bU#3u-F#UgVt_@EPOe(7t%*_3$tYkphqoAbD0f&tKqs45v?JA z_tqV-n0(~zm|xuV17dX)fE=Mep>~Dxiil5$(G$YDdOQ(~AhV;ruQC4t`K5=^5`(1L{h1!QW^H-Ef}X-9vV3)DSSrEJv{ zie)`N1>E8q6317_Vt{gNf*`mexwHh)^`?%-w;jCw>rik;sc%|?#;}e;C)*$_oT|f*?oDuUT0*Bh3BM6-7H4F*-)N>`{ zyeQsKb1TY{>g`~TpXbo(rhYg2Zh^^><ts#X+X=Y3t6<(9F53=zP2aAP#*?UGVgklgEaSoUT90o?cYzMh_0T$UBe zPp( zkB`D{@=xF4@3%be{`GGdP5cXr694q)fA9Y9Z~t84E{mt^Z&JS1@lyBTHz@E6Gzzd4 zM^YR{Gx$IM0#1P;81gC2yQTi*;Pp}ZU#Ne6#x2ik_{KPj{tNog4>14vul_%t@qFid zm(TOR{_TIp{{`hRhQetICrJhafBo|ljdfc3xdu;U!6*C|8YdYJ!$=xuSn#xe_fHe^ zpAGCkX67F+RJU=NUxu!P<5~N{-Put?g~Bg9Z#}(AhN9s7I|6^HnaIO`CE_sm3_jr= zI!d59Zi2ti3|D@yhhRmavnMKtChDiwwbv)@tRcb4!)(g+R_|cRJ7}uiX!FM{|NLCD z<$kXpTOt=mhOaG#f*ynbA@W=W{<5Qo+S_ynMOa(k`^XR9A-g+~<^Jw*1*@@Ex?;OQ z>*5&pEe|3`o!1Hoe?z4O*SIfC?0Iu+m%X{aQT*BF4P}4w_(hxJH(uGeepd0bFZ-@9 z!?MglbTLH2-l$8WnbbU5q-L;Pf6;7B=QUiHfACN>Vq zl@M5riA!EQ{pPGbO0i^yTiuIJ`Wt+;u@@5yM`|dP@_kUqsb zQZ+m?{+6-&d68=QRedpqBtrhCzTg6GzDzNGlO8C=qjVirqxh8NKVr3)+)iD<^TA?E zG`T(MpaKwr?w={ZGw?McJnMw-&5<$0d%)(p+Y#YauL^u_HY79{*zYe_i7UF0-Z4G6 z#@1g%FgN2yj^wfC+xL13iO1!iadFh1{gTbN2wwG}F4IL~W}Kdkp~ju3DKH#P0&w;t zAM0&{h6AS&U49#LNk^Oa7gGN~XlZ8J0P1nSH4j7sP?s}S^o(8~Q z5-Ics5Q>Wg4d#c#<3!RI%a`0Zh)J^fZNEtZ56zgtn2zs@fne-JYaNxN^vAGpCN1^5 zv=bJU9L$uV%f|(&IE`4xb3GiIC_Kf-6aJ?PWwhTWLj+)en*&vH*3uV842N!vtBGM4 zyP?G6BbOb{SZ#v^7qt=I@IY&stgER2TrKn;*;`93U-C69nt6tBXh%QdgShX_zB-SV}de^hN&U^U!Sj#%DMI|1Z@*h(?C>C?5dh*Vr3YXnP3M1=U>UkT zo<7tX*Kn?&0QENppYh8lDhW^!lL&bBd9f6GwOyQ(7bL+l^(fvcXYOFXSV8)J$4_`8Bq($|1MVM{;<$fr z{As$50+1C{xqZPeo1`?~R@%X?JB5a#V3|ayaL(d8G{uaHW+z=9v%J6sKzM)6XnZ6* zh?5kr4@$2^Fa7DYf+JpqKq;eS`H^KtbNPPbq4Nc=gXP|U_b7#j<_!!xNl*gWKFy>h zr$>3dz1A=CQiNfehlY|8dvEY1*f0r610eY}605}?6Gn3Gi=TgR-*O-xLr;3Zy1sLy z!Xg?pa&ORZIl7IW;Z^$65@H*@B+#(YYA<=C03z!6VC*Obqp2(kxJ`l=mtT0B1Ww!EZyDb6P z2#UzR!0Ws}Q|QfakU9Yor?$i*_hnG@N0&mP33L2fBVr6XuHllVrZpUN9 z(yfyQ?AdOaV){sTA?J@7St7?AP;~#XUh*wa>ZsJ=p|@O-*@Vxo_QmZ-FDpH2@B;kq zC|LVj^tI;FRTlskq$(yGG1{WJJJl0l(sMgpYBato@MkL*vEo5>Lbz05bS9H< zXI#+K@BAXU%n#F|0CDo5D2nS*I&2I_;ef)m7(nKVPPtH=at9$hwrdFe0f4o1&kGrC3_MxbJ9BV8q3ddW`nu(y% z`8&13$sZN_!jL>vk4Cc@aB)20=o!fj`*eR*k8~A4vX9yrL%pl?z?4MYXK1lGgEbkx ziHpfxSQ$2>Hx-98ZhKr?37X2nvx-6fVK~>lB4{-J5P2w`b`_(R9F}bZi<^0s--wW) zi!BQeRz#k)t1mBtl|W#Zp)Tx$yuG3=B%KC9z=<0=%FSM0T4{Q@RTJ6|A?LW!Vfpt{XIS zNQa%vzfiTGB1sY@EMDF8Wi!5PszCzp1kU?v7p)~Xcxc!lJN%=(-Z~dvEW7Mw!wuq5T#!qkd|t8VGyEE`G)@q`L9Jn1r7y?0 zCLr*@Q!IT4xRfmaaR^c|Igv5ikr!?+d$Fzc*yXalwabNJK%C07OogUsY`#7FuN^Xr zW#Cby+VFc5`c|q!Rj0mPjBS!oL0-{syaM8s6Jpm5S{l+ zi{=2tzKUdnd&Q-fW{>mTv={YT{n%;B`JjkUwpm7t%G+R|vmBO@e9FQN zPypw?1Wc;(Qc2&d2t<(BQdwL7@ePQvF-U8LTnln^C*qmfLXf)u7>8~41wp`PF9RZme!yzC45LMW-vDPm4ProU4OVT$@~wx$2k#0RmOpgQE}8pZ zAm$x>S$umA{0H7bC~yeLmu1^;c!N(oRw4uNu^~TFnS=3;L7?OEOCn z`4zHZdRF@x;-oq#l4&bsx+0D@KhOaGB*?{PK|F+a;5!SJumRrL3a)wg9HK3tDUcO~ z>ttq7KfAEDeE~~z0K{;k!C-xKz}Ba%qPEMnu4HZ2oIN)gh;vVSW^&Z5Uhgi3VzAs@ z)=S|u34@&lk*&Rdu<#dRQAx{)3^}6tVFTdiO<8)GQEqKXO$y%^%s{%I_f1|4 z&WXnh<>29M15eGNuUW&cm|JQ-*E!!bQ)T@vVlJ%Q46aRdgd9HklW?G%oe~q;FvTa* z1USca5A8tj(u2~Zdl!zCS(A6|!118aLC%$7F%k=6c|w5<^yZNG<@@^!>tOq(AAat; zb=TgI=e|DTkAcoQqfeD%^}iFo2~Gcn-|Vc*xE9Vv^I zQIC|L$K~SsD2ey0?br=;i1*0IW^|Og75rrHTsh&rLB+27ljk zC&8)_jQ7s2KK}OFwmQ(|+Hu9sY7&XqbqNazadRO|6c*IsAlFX)8h_nhlQ{nz+fu!E zAYb2T5cAXOK}H^qM(c;Q6a2dY?89>K;;tG1pZw|6-D!DxUfb=SuJ*lyk1w1#KqDjm z(rhe?Jg%sGDJ1$;nwWF75-5QD`eW{_HxEwRKn4L=8@CsI9 z&1EVyQTe$zjNOpuzkkIB3k*!@bbIq~B@}6GiqQoY{0>dX!V9EeULNm&@W+og6?Riq zQ0bPJ@_e143O+ffKTM&a=~Dc`KI7{wlB2-)!p-g&a1uaqoSNO8Y)2+9Zf&yA?3jND zHoq#+1xiu20*WKEH(AqY0gzYfG#zgM*-Ak^DPj~KLNxeIJ%}IX{V17jE7JTFKF&LL zI2DB0jb%!uBGBX3n1+>Jo|ej_!F@g2Ar7huWC_(ww1Y!5Sfs~hnnvIG^1BvUo>3F74siNc6=eU0eO`Db^}bX>Xnq#*n+!m0 z3p6~dpjn?P`J;c;ngSN?T^bC-xG$aQpU|XO)de(P7qKM?+d_cF(yI*ko(Nx$Oz#>M zGs8Ld=4)1;&mF0oJR9>hM?H4ha)Wm;W0apa4o(;Z+_?qI?CFy24mp8t>u`nqYbxZxH6{;mAz zjBml*kc!PeDp1Mi=zNh(B-s~ifu~GZIn3l|C0UD#iyi8ip!8@ui zSvnw_kIH!i>4L0&lN{8M!FSz{sJ+(ke42-57T+>a@f9=DrRx)C34{cmZ%H9>;i?40 z&FfBQj}c%hcFl%GaZp%cJ)7rwVtp3#O#P8D{`kVDJIb`@pjW!0fq-odv+>8jJMXO!gn^7`Fr8!~IYA+4JK`L;GrTXn;4@ zx5IJ`MW2WU+aOm^09fD+b(?-Hj$0)DagP~^Ny07MN1EOR`-74`>zgQO(I5|;r3ZBc(}%mIz%#sdTp!40tGRl_p{y-5lGuvlo;*lzywy}@E{WhLCnfgC>yuC=~8zcp{t zHL*zw@b2#&MJM?28BhB%>6DN2UT?46XE~|uqZQn3T349Fkd70w(HP7HGkylmi%KtkpkNQ z<+1CX7x^6WUAmc2uJC*?mN=7vAqlp8Ge6*!429}^@SVOV53h_C%ym_43`J5DMpUv< zq5SuPG#Cp-@6Yq+pX&aiXVcR*H2Ku}pbeTVG-d^*MW> z@@0XAFxtXAiFZ1+`^?n8JZBo-aGs zaQx*Q-FgRuFyp)d`tp%qunFb;ZO_X!dAPmG^Pa_U)#&`mf6ZgAZHi_8v|%ZWQMdg8 zv19E0O7F}?pAk>^_T}=?Wc;?X_=N`CVuzJrAf#xe#6yw~3u~s$q@b9cP#ISN>_I_~ z)#@zP_K-{AxYRxT@DEAcXC2*X|7oOP6I$XoLIH4Se3M4iS9$NR+{0fwWCu9#8&T;Z zI93pDBWdb}S%(S8@h;e#{gYbV7Gy}>yh}5)vyCp^0nSTpXq4a#^4r~PO;wqGm01up zdM`gdAO94zBvAlubmsouptY$+-C8#agtF8DN0jjdo|FCV{%n@4G&Z`FhLsQ5YC#uGh!N)=)R-Bn|SYa-LNNlhgB0 zuyGMstc6HINjw*uplSM2LGI~cmJ9DE$pY3gWpum|fZ%Kuy^f<*;J$$sBC$|8d-Su( zgMifq(ip#o^Zo}K{IRa~1r+uecBGzN-(WM0YoW4B)Jc*O=B$2QH5u_?uCObldo&1m^?}fgF ztLXHvb)9r>ftI)Y1K!?9SAc3I%zC0fDSFnPdnEM7t06^)M&pkyTGY1-udynnJzj%P zbA88fhnh8SJK{!RP-0Kw4cL&sU=xp56hzxYffq$h)3C zi}E<>;azv6a3&6{A*q*D=?xNhZ-whfBgkV&*D#boK5kifcGpT^3ATWru}gZRr8uAh z*Si>Q8R1q_XnBrYDu@ABZ6lJYZC%vC#2xu~F9Jm*Z7)ZB=Jc1ZOK4_@Nf(=^;fO42 z!Y^ISk)SI_$DPmXA@kQak~i=iVApwgPBye3IDt0I1!Mx&(QL;{u$jJ+0kAwgJ?!>P z@BcW95_#D06zfq3hP;HB4Nj8v0BAW=zqr~`tJ(11)K}QccpL{KF#z7ec`jJVKG2K1 z5dhzG&E-<00&l|es@PDTFhsywza>>fS%zNjY?c)Tt{ES({VX-oBkdGMXyc;SJ=f~HJ`UAsx)%FS5#lu*9&v88c6>=$pF@+4KTj%ue1=%v zOjrf!f5-wvkc0o;o*!(jqClz2<5ljg?Gg@bo`K%I(Y<-}iTIttI3o7OoW@&4Ls8VF zf~F-IM^VoAUUowamx2Ny?pf3h1xg)-%C!bi>3v&n`Sj=`yCuFL?pa__kKu}DE#Q9_ zume^NR$3WX-)MyL%flv!`imivL;D===GFh-QDap1TO$k8LR^h4 zBmE6fWp?qZ9!#2g#?`m{ZmS16FTwQSc>{O5;F7Mo3OIRIqoZ?0VB;Dzrj3-750GHY zA1HN~-rt)mivF_+t6mn?WMl#w#zn-Cq=Vv9jo(iW+k@p?m6iU~PV*UvTWTn(*-FLM z5odw+iRoGpiz8Lj3Wvd^&mEk^wTi?2*`e;-hS=aIky& zUWq6&Me`Go`}L=nI?u1>%0%oMn0b?L?#%ZJh&Lc%jTOjiwjjYqKkb z&9^@sNnUK1%^`Le=c+euu7{an9{oVx``RF;UbNruhcsz!s~=z26rha>1F$T9CI^6*eQ0H)yW(xHSNKK2U+@ZvYK{W!NYp2|H6l{uNt*E4yHgBU(q1 z`94dM4A@5lV^ebMm7=&d9lE9Y}njXg2ftP-o-{Ls}Fy|W=Z)zyv`ktQ+U_9S(Yoq z3t{~u2>|jLSiD3Ab=8@mA|&AG-t- z{%|)27~*p&h*qm_|7J~+@YiEtV=DchG0KUG{I!oStS|Wx@9PDgO`Lflc>L7Y4^R>i zPY8M;JzYPp>_t0ZV&ieEkF{#g!9=w+_Kc{z^M(bc0DpM9brc#D3|<8Kp*jF@)nJ8l zO@!Yh4-=m2G&0eqwFlNb0V37Qw^yUET`EPTpPP( zyEE4$h`@sNe*kKDIo~U_K`?mixh>RSOP1sDru(ju9IXwUB7xN&+ygxp}sN9Tw5 zu-3)Pc;A-BoGzWLf`!fD5PmiPvi>e%Mp(xK^27;nRJ9aKUxcQ5xn}tHHi8DzQ5XYS z!oa$O$8RmO`z6<6dYwwJvGvD;pU%Q=qyPY-t4GMg_j$UCp29z>5>2WBY$y|?x^E2c z*LrmHOu+fbyU+mGTb$?ck-AI){#@W0Zg~y-vqG8|_$AGHd~ycZcyFX^9V6NYX?(Zn zNW}ag2#9Uw6M4LR`@pf}n|5T~Bi#fNw6+oxZbIA_2Q(khBH%2SPhLB^`zG)GXw4@D&tAm@3GdpxNdFEYLeV zn;?^&)x@Z!mg!N+Jk042ar93o;knY}QXES=L;=}!QOGYe0L7sedO<9e4Q6@ut7)wYj)Me7EB}H ztAtWRZw(?V#!ov>F$nUSLcB*IEPY<)bp-6ZVS`-o-YO6SP}7L?!S;UOP_hEIXWZ1t z3w+xmQrBclk69L$KlQdoT<)Xaa1B z0Cv2PE%o#SVNO&U#ix;80nM~K?w3K6I0{7B$9Ch=^`W14WP%6swd9S8{4sk(C?E2f z+u{A6vTQ2kGGK$KSC7`>;l)0$;xJidu49bylpSxS!f9+A*sF;$%l_&it?ejda13bF z7w5QE_1lNGhb zE$yU%(nkQhy-GSWKEWrSm1hdp&IL9WyJCAXc`u_v#=DM}?cmhrOIz@qyH^09`9N^y z!B*!TUXJpAoZluke~r>t;hz%^2^c%`e`8V8PtPPK{(7; zt=P0RvAZ2n9;FicSSky9Hn{_Jf66t8xKq)LWp0N<`JC&R0LtW`~TZn?7;TOtUW(-qH{gNN?jci4D*6aqB9 zt)S>fV1CK@8GF1ZZUBEipg$n5HGpi(t$Qtb*zgY5Vh*!lE7|Q*jt#XXwS42RKCgvGkyPf_h~bN(FdjIR9WXeWNd>x=*K z*#8e6^Pgn|g4Q8y3C<1D^`!|%%HFSFD;Ws@O#rb1?t;d*;e9)1=8l`(W#eAE{&3ra zVr>|2VJ!i?KIbuj$1G)hQ@<2&l|o~iPYP=#a64T!1>VVr63?0`!5N+Z#TX zg^5M8Yxcu1*_$3U6|N_~9IRi3-ylB}I_^^ZzR}E!08`^)4bqk^p&YEyBcA`blY^H> zIhp3xj%>~#2_&oPH}JCrHi}`%3s4901t3FIfkXhkOB@IZ&*pE&_|1*8z$-iuPDCMk z%9Sc@fQyp(_UH=@fUa>1pL%ooHLdjSjFG{3<4l>qN7*xUXik=FC6vPP86i9)opw z)~519Q~>>6!vtpOx8f$;FPT3tg)uI-{}ijhn$s1}WB&~}t#Wm}Bg|kNH|URX_I$LA zp+tLZ1-%S_CI=AnCoX*Z`5r~U&9HZF!TO63H<>cQKTFrp)cwb965CyWQP}2)tIH4Y z`+Nz-C3x6i{rmW)7lG6(is6s(=<|IB0=~~pdJqoSfL*1>YP|Ks#~bCUPedZaxt|8` zWrkyTY4NbnZJ2PW-0uLZfuD`f&j5e;Hr?++WEvC+iE5*UV;#D^e07X#cBF*lav+_G zH&x=#+OLm0)l=}p-e(^M))ExuTp(vEdNj7rM)T~x1=166kGG>bebRMVDJXR|Pu z-5x?oFV6wqV<-$m)AfA5d;qBYDGI!nhFbu}`G!#s5&*8D zcV~H>st641eZwlI^YrKXEZ>l@%p0x-_Z>l7LcRB#OPTJJRUWc&?KL57f{Nd{lWY?wCI`f&K=s* zuk4M+?C>+>dTmU-4U7LF9&lRe{GS(Wqd9YeqxQEmU%Z|1e?0b2kF^tsQ?z8vPxC#G zl&PP+NGkay;ec0qld9)I$5SVS^Poentv&kh=CS25EB*26vDwUt3f)_foDvm<=if?= zd{HJF1&rq6$1ia9tV>i#DVHcRbyv6mX7Ja+OBsR44zRKLsXuY@*TDw^$Y8U1?x*Kk!8|bB(W5{13WOX$H739ZM=poDW5y8d zGh<615yTr63E;UEbnx51H&^ae zko)`k*2qslZcv`TX~Ld21>kLftn~^MJeb1vu7Y=lqF=aabq$tW05jUTJCmt9TiEcP zGyG7>XKwy+Zwz~iDt})mjz`DYFy6PB21!XP{B3xdj7@&J!1u5;1aoJAe3&sEuHM5p zCGw;39;Sx;dPW~;)qVI>7Mg)lXF7fVkZQT!k0EhzIi&u3uFzE&VP z=(#4)&eMPm)$@1&V8cVa2%Eoq*cZ@LP;}v%z`J@%4L)PypX4>iyr~t?%Z1{od(rLe zWIV_90!`-Uyci7OX~@eQUPOF3vb(lkT#O3@-SiFc z*`Ejn#uRtElduFLc(udY0c3m_w#34b7=yyW!c;oo`G6q8ekWqm8jzA;X(LvDqtZt% zTi3y}D3Pyucos!V@+X8KqLdUbHU`3bC$HuTYx>L3p8s7Y|nqL5y-}|-rPHPZY;_1 zP!t(_QHOOW@a#ZoR(oYe#@9+ge7tp(fi1(Tup}|zUu$$Z6)Rso&VXm1qM5$qeT>9{ zJrVG(h0K%IA1K8TAN0NI!E`0I#iQr=Xz9QiH=-cE2<)9zKR+r%4Gd6#Z@M}<)LJ`5 zB6$^_r6@f*KCVg)_7N|hJ$yO9(p8KgNZCt5d7S-DX<@#fG3-?&dHt&BH}&5>TNs&l z=)Gn<$pD8wxLYgnLK&&T??zbF3n?@*vQ+woIoSjSG8{0d5{z@UHld=WIp{<1uGlQ? zx6wBctt52;XR^Km5yiAG$IkC$cJChmUr=6Vqvx2LzPUzscVQw}OGj0g?Vvoc_EXKr zoKV>3d-A4107XE$zYG%O`g1{WonF*NV8Dlr4BNht5v)6{;(McC67Vw{yx%h6^Dnj@ zGZ?RK=L%neW_n?Fe4WQb6VO}x#=98uDmA2#Vr|@EM?kP`RY_{s_5KEx;K;NPrYW|q@RNSbzON;$VPM?>B> z?q#@x{^UZw4G`ffCwyQv`)$90_%5j5jVCtxa?ExAoO{?v-<^HiGf~&!5uPJvf@!zMi{5>pMN`89$;`br>*@(TaiUuH!Zf*nqB8sAB}Ox1v&EuwXL3^ z{>Hwf&$*#A4E4C*Ht)K)_mGOM9Tf<}v@PfVBJRzWU01g3uy=jMiThFykSxQmvEB|${)J2E9{F!1S@fKAYXE$TYOBQ_~Z zuIVQmos9Ur**tNxI?IswhxuwmLB4MYYWrHp6=p?#E$Ny`Q(lzxQn$Os!^j>8=mCE&GZR8 z6=^B}aZIUDdL_F(P;&i*-@pS{jf^_U@ z(N-Vpn@ZLqGfV0sej{rdqJxbsSqwn(FjaAoP+K!*#Zu{BNIxgL=ihkGykUPjV;L9n z>x2uu(so5wmLV=l*A}{}X_yCzo7<5tDq^<_ZaCit%km+xZ{)a&{LTXm5G>kr zDih9Xx0j&ou4;eS|fh! zp6f(=wO*}pBh*Sg_a3F8<~NeCO(kl63Y@*X^)km+*Y3)?4eon-0uJ$6XUC)SH&hmz z#A#bdXQkF@TC;KfUdij}S2!VIMijUjpo-YtZ6A%mD=#Y;Y>zEcjc~sX@@TzP9gwOl zJ_Dl?c6a+C*lPf5@$K;9b*k6rwv9Iy%=S-pk&f~r0W=q6VSWP`DQ_rP&!Mc5Zz!Up z5ynH4OLS8v;rjG0%9mM1?vmzZ3jpVbP$$v#CU3Y$DRnPwMW#z8jYFBQb_V3jH@g7> z#ECn+x2DEFc8cP7kP`TuO+w{P1tMR!G z?|kq;qUkdk%cMYEU~M3t(MyO4XfcZvn~dL=G2%R(Y@RQmzQdKezH&E2jR|^N%hdg} z$n;DN<#04M-|g>Z`oK+0FK7VjpFUuowAXBjqOw#*`y8`^pL3h%Qo_tVSBR>1Q0CujqZ-?*MRs zv3QAhA}PZbr%~pJL;jjBQ@JuszWX9hkIE9^&*1Ze$FCu4jORk^PuTYfkSx!`=Qbr2 zivRAM9jSo)VKs65KYY%8=I8v;uoMK90#;I&p;T#Z357*1fTwiz;0lldSk(U2NjE%; zAUu4$dXe!;i)89`<@cn_5HyFz701UNvF{Z@)rN&aRF?oV1I=F}JYA%qw4NKS+F2)OfR~n zt0U(8Y}hu^;LdfakIl`ae!>_HU7ZIW_am6!e9ExbPdtnbnIpGJr$5t)l1P-q7w^x< z>_WZRkDE8S4tyl8i95T!#xL9m7IG5$N@F;a+ch--4s1M}n_WoYziF;IKX-$`eIU2K)-J=kuMb==MLTNsN0OJTY5A5ic`4dhf zpKo2z)Q<8e-^8md5dwnfCg+FPsdstUL0Vlxl4-i`~tuVs^U|&!H zd3;y8w-{3opJM~=L!X8^rwL(vWNyz;k7y3=)n|EB(NLV(=tJ(3jH+>Y@Pt`kPE`OC z9>%SSu=F`cQF;_${k>ya-43P_uh z|8MrF@8hc=Ivlg@FF|lT!+|^&t&{Z~RhPfoy>wG#PjDyYa( zLjdB;pIR7%ap9~l!?Jfv|A_0v3$kw#Eqgi0V0hA3o>Cf@d?S7sP1ybd>p24)8sO{J zB~OZGG}FXG-<8bxhU9+n>)C=em3I|I3b>c^sp+-WC`+!ses2m+slTxeJ%fSG?eyB7 z#}kBlRqe!uG^#yQQyDN;`X}gz5o!35doaK?|Do9pN(Zr)ZWt#x-nBf5h$}AROr}&) z^alzO$XkT^0-A3rV8Ty9Fo9)@7AmS9?DG?;kA=AE0$qAuNi-`C%|?E`hiKDv1nMkG z`#^0`6j2f5gzuJ2ynfzU)!s@zG?tLdKDLP0m$_r|*-)k>oLb=xQu(gC@0UOmK6z~4 zx1Ooxu@#Uk8~*|05gPb`(9FA2dDiF z0z39Vy5@aBP_IlAhJxA7`)j+@V4dvYRSLY}qU22)kby93-7T3B8J0Or^q6vXc+~=X z-Fy38I3nL)=7cH`BxT(l;)N>@F0Q$e48RkL1ETu@<#nk1s+M$1T(p}e_NU@fTawd^ zAI0gB#@nLlXTJl;_mqM=1~Ft%YpfdeGBk%(XcV$uq`IUQts!lWs4$B31{D)TAPu z3;?@tpLvG0ilL&NgWnR&SdKDVDEx4m`-N{xN*6!VT8&1H^f7n#3n4=V9Jy9MF~awM z?#XE?qg*A9e)Yy>MKg02lg=mHd|~ex=mLB%wDsQ#nf0?Hcet7PK5GT7K zdw99VclQy<-X*&~?1nYVqxEK2xx2s5D)6&60f6bEnK9qw2qV?@Z7In2gS(c8YyzaD zK8$MGFKUY+&v-W35OLpseO`i)0My?E*?TQ3h%ryV$XeeC@z1(g{hO{L)oxP{7jkxx2;#Sj$7~C%oE}J-gZ(8oiowl~{c0F3w4`?Ye{Bzoa&{#&PDl3kUiF!yLNQd-F^2Ptqjwis|}=*Lbm zG4rWeN4Z>Vc)-EvbeQ1YM+vyaxkj$@#iVaX3s>@Wk3%IZvDV0w8=htl@8;T@@!3V& zxwqN6^I&O+t+`j8MP09{vTjrE4) zrJ}a;U_tVsG_$XhCXByAeV^NVi|98wqv0BV2g^6Z{{OQpzrfsGM&()iS!DwbCc4Qy~Ee=N!7 z%-Xjp1-cLUq?AA2UTS|Oxp}S1Y*9tS6zS0uSvl1KjBL!KVfwt-jxHTDwjJ z)PR`ay_laM%UaJC|5WT*dCI*GSee6>_wkBrK$1C*RofbWppUua?kyqjLW~0WOgof#nW*kEv{>PpO`|9SpwrM%_83 z7L>cdy#bpdZ-;F4F8Wz(Z@M)Ms@_V1m*_m_-EuJ8JmTi?C=dfSi(6_a?w-vu6UnCliH zQ&X=s{jO{gZ7xq%d`Ki*%(;LH(Bx zzTAB$e6x`J%B=g{Qg>zD_$=RO2F}}C5BJ%4jsIY;l%kzE6?9)BAIIY!HbUVyRIK__ zmAuDUeQlUQaqn9_fZV263H^b4^p=W)+bYl*YMs&}o)gVrBcXfcUq-Mv%Tu&bi?K+% z1V*~qgsZL>HqkNwTE6ZSzKC1TU8FO`k%-5&X}qlWd}!W_e!q)!a~Vntzw0hnsE7>M zUO&;`QONuAK&}631SD`=+$B@5P#pvV6XV| zsTkXCmiry?_<0&DfONYUx%-P^!md)nIF<(MJn}>=Cx$&fiuRo*+wtvaC&#n(m}=v( zno6ZFjb5Mcn{!Bn9@pUT$K5Omgvqb&OeA#)Z zz4r7%7&Qe$9s4W{dU%N|ao&|QSN#URN>tHyw!O}=^ED4CknX!;`o{{_m;8;dM!ufP zb!vx#Y?nje*_n-=nY{RCnhxv{kFfye9d9!%xodJ^sBR_!F3~;s8241Yumq4d6H4kZ zLA#hNq5{l#c@j3?Nz*&&$pPK*)nj-~K#Pu+HZ+fECN8x}J7ZCpG|J(Nc4_O#MsY3A zOV4)&*T#mns>+(!AVInq)nzUN_D6#CS=(YDH8$}X_z6I75 zeI4%3-0G3O=tI*t=IwZI`P8q&#(ZAg^0F0W`h8yesIly*s@b;9V4+jR>If@mtMQd8 z*6`TVazQ-pN#E)N&MLn*5Qd2tQ5$m!*5ngDXo(wPy7*N7(a?fFu`3T^9=T>>?7#a_ zzLf~S%fx4DwQrWz7226EC8Mc?HilK3uDmkfpP4G#M{1M0QUW$R&Ndci@jOtU;R8^? z!Ny!5)v7tRT4QD+hni_TsI~LwF+6@1eGdyVOi>uy4zE zcbona%yoRnxOx_Y)|+=JaZpOU4)p~0@RX^tNJT*uIhPQ(^EjmozmxI^GXAV_6OY}) z_WTa;3PTL>lNCe&%Hg?u)CMh;TVhOQ#wp9|6kU~=MWWWfssnO)VIEd(OSIB|kaemJ zceEpi_?9L5?GT|CzAIKcH5uOb>Fwus_7JmOHs8+#9Ag9N@{2NJe1ArJ$m4-6h9r_B z)lBk1lgi_}B)@reBeYEWUZ)$tZ{xN8b8R4&=AHAakq_1*V%#deU%H}mRIlc*)YF?% z&6xLo?T51X*HGTX#%Z_LR*#~c-r0wIoB|P>EE!~St2_bRGWNB))K75|o;!VLg(Z#h zdUu=PrM8#d7ONd#A9~G;B9hYPx$Q`p`7`ra+Ci=E7mlJqK(%`>tW{4Oks#F01@4=C zwm`e3GQKv?yL!oZFx`x-iV6oe0GhLrdAl#E`Zzvc7bVglxV4#sfCZ@6F5ZP zKX@5IgGjx)vfIDRGLY;AP+ILk8C$*Tw_R(V!pHr7xhwAZUO;^!baE+L-<>_F97ol^ zpL(bVqm%Y&kSuf=_=-u!?<`XYWihn*VBRtH0oa(mI!LBIF0agCvXEe}2LG)p6RPZXWp*7FsYDHHfv0X8ZPtruXOzK82NyCEReHZx%wsMDfm5K z@gU|)gKE@CB(0spBM?JzsM@76jF>l139D{=aACx>I^FE|`Q;$4zH|9<<3STe=bOA* zjjUv23X1c`;k&^Ddq~LdSqLB`vZHG}Qp9{}l5PeGE$>%s$zVzWn z^ePL~X1@$ayxhy@^K7j-^`x>iufarIEu#)BX}AY!^yh^=)}E_=TCuQb4;uXJdixj= z!1v1=?2UPC(Mq0nsx}9zLT#3N$h@pyfAaL(|Al5juYzH7pGcptHHfz2%EA*%3qH(>|i8+y-ei!cmijg(t?BAi8BBKl=st@W5v?5n)vO z5X2ivb3ucjt9$;mDvgFhw$J$+gJPy)oBY16M*s9bkFhQ3!iswb<4nLGJvt6&13t?l z%=PnaqLbytmCWF4u1{dKq?_EMhWkgRfG2qV>+R(AtFL!*|D;}{Pd`i8>t1Celan*t z1rl?#Sn=ZVf&HqHjAxSjXUT4=;LANC>Fj>ScjwN(Z&M=Ff0gWpYCJ27R>$p?V4jQQ zzMnN9Vr$9s+zd9nUn#INsZP?F-Ls5I^P8C+?3KU4XCn*1@o+P^;`Pyo z&$R^%N!u1j!%Ei$UF70d?|S;_YZj#zvzGhRK3mudid3r+5Zm)pX1=(c0hwa*C8 zZ_QH%z}W?94Et1sD2eg$I_0Yba>dmFMD3#^;$3+GERHp8 za#05Q=xS_kN5ZEP?-yteD2n&*9m8R7Sd^NZ{;~n?PU(~Y*tbEW9af|%b^-5Ga(5lq zPTD}{{EJJ0dFo~Mv|dO~Ye);EGp`l;$7@3|k2+rA$1wP-#DE8a!B=~en#*a&}7Evy^bX^!ryhMBwr>$=>o0wE?T z{Hc||!usj{)W=sPGd!EV7!M5$6LVlJvEhkjOePFZ!U`6rYgn57Wu7?b!z5H@hk@Yu zR>t6Nm|vfKT-L@N*al4PFljE<1G;`D^7Wmy9-f81t5gc?7u zKMkE={ZX8GgP;ET5z0Nj1GG#giv)}CuswchQZH5qKSP}}tLP6z5#3pVwTFQFYLW+% zuTpEU2OXLr3S3R3?usCJ3@vq-g!o)I)3jf?$nyd0EhqZNND3&*kT(0%du9<^3Zg*? zhJfPYyRQ#&{{aL0Dz2L{zNZX-n?6Q9#-aK2=Uo$Ot>lp{meozaSs>2y#ivzYD%8#+ znZL@phkze`>tlt**Tn5xzHh>lM_x`Lz?!}IzJ^B8z`c3^(k?Gc9O!sPayx8_?L&0m z_t{KM`3piwxuf~A-h`& zpB83{89D;iLEzW@W12pA@xgdcYl7l+5z4+A3&1%=rrie4jf$LQI#XI)dy8?GkOMrr zu=Kbu3YSx`#(j91;pnd^7`zw$_%q(K2iza{j%QYaFo1`qI80_x>}(#0csL4Ye)Aub zzLlPzgAZSdkA$lkW+aPVeV>Hh5Sl?^(kM8C1^9WDGZ531hxxX~sknGnlObqS(0Aee*M z&wArY@-I9)QLr(6z`N}gs{V+w^EM&$oyp&~E0`d^_Y4;*a7R}Rq=&(D+hhLT14SMB zDCdh}?mm%?<-Gzzq)V)TcXjnq@{XWv+^4cD(lcOAU>|r%dt%u9b(i zvUSh0;FE}vn~mY-8bK2si|01y`DVW>ttV~{~YR@Y07x$dv6|k2S^Uz-wr>g_7~A` zU_r)tF*eav%80J8bTiTDYF2(V=daK#vE`Q1R`L`k#i#6yhjpt-9I_GID|HS3Jx5gDqzxr~irl_G>=skT$l@emSM@`6}?&nNoSL zeUE#gWipmXfqy`~fuYkex`<;RFEZD%I?JywPf83h1m78+{q}t(1={@UhihNL6gm`w zit+VZvxw8v`Bhw^kWh_#w*c%gVv#$zfg{2DQ~PsuZ{6~19P);kv=|73BYPj!#4T_# z)IzR^HW|+f*5~P^bpto9W#WuVx%YxAh;wCqQ=`wAyzaforBN2P;9TB6;e>ON+X_4G z(Xra{Oxy6wkLR;Bx2mll;d$RG+-seby_wK`(qf17lB?*`D4aBNId6IiqeE(Q&pCIK zYyzcKOu(fiS9Y#`LzYXywlR8(OitzXs<;VnYepB|x{?T9injviz?%2{-?-S0rDkmx z&mMqf0e35LZQKjM1ygZyzO>d(br&~$t@WdFY@z?#)0@p+7x?e#7s4$xSu=TG#FZO% z16N(qzr%Ild!gfL!up|u3g3z&F|n2>yNkbl3yva={4-?C!o11as=L8}^`?U~zRB-w zNne6%Xf2;M|3wn zblVb{+-~g~HOwL~_PqAHuli!}_Yy%O?o__WA>kr)PXy_zEg5XT05IBDn1AP@Clr|t zem+CfTLa%CT)wA$M+!;F%i~`9PW`g=Qh6r#g)NHhGB-AcMzU+T@8c(ik@FXu_C8;3 z5Ab^!y&p?WoNv3;L}rUZ%P{H}ez37KR^}D2d`G%pp_CdPaUF58mXcig@8MDUXkR%p zJw8=MA?YBqD0SE5C;;cj!!`4;yC;=nwq*wQ-K^*PCp|Oqfun$|E2Xn%C90OC=Bx1B zg6Ox-96D|fyMAOWL)e7)Eq1(JdB5ygM6gSmukM|ry>IJ8JCjaVEL`u6sw{mMubD;J z)iFqHp$RblVM$`b=brD!s%AER_#W@gN8?iwIxZ|hTpLC>>mz_UooUk{dN z3x)oS!oV$dlsPhBmBCYr^vk}Q2w`NwEl5w)j#km#nQQ5&JV@&Nnc4ipAp(750=38d z>rAexyM2&(s^c+we-V^aOt=v|jAsD*eXO3_LCe$lSFGi;rJ6|D}*p+Lhz(4Ua`2<{c3B(gm0u zfGDYnaB#oH-yCu9UUG!Z)^zqhZ>6LI1ZMjKRZLF{kQUsH=C)1v4&kohHxnkWrCQ6h#SC3JAEz?hX(<^e%7IT4=;PVFj2Pc)cfYWtxjfW{2AoO{edSE8k+A)&|p1#+{4<1g^^X-p8P62=?OIco8{8##xdl{AAYrh!q$_rGeUdAKI z9PO^)^L1Z8k?8H``!0H1Pzdz$G7a%wJ*DG|mK~Y&P7t;qd^)dCoAxPczkcDDWk|Ja zqzqRdJdLLd^}OE(uuQ);!EnfC2D?@k{hwQIB@}S|bIYyumS{1D_w6i1=v<8~~Fi_h)F0eXb;qe>&wX!kqqA^v73fzHqid-#H9Hn;e=w>M!# zj{%T{QS9>Tjlw<6)td7eQgFRtOg02iqvcxi!~ef^pV;#AZlYKODNDU_bcJqg6H_U(CzJQSrEbXc+Y+ognILZXBi6>h(EQoP2h$; z)oDrTN&>8&Z-|%j<~tw3S{ZpUy`br0r5q3geumCQ9gCPJcDl#9Z?X6N9NCbc&)~K z)ZgQMaq_s)A6o7`mFH(DR66=-57QPY(izpNz&y?rv+3!t>KH_ zy7;SCGEswLRB+SQ~Ibe+yQ=zmgy3Guq?_6Ttj_&c1cQgK>09O)$meq+MShl~=Ew&EFs7Tp+&=xj2C_b$cvm9`VtQHAKIEsdV7;q{XQ_(My5 zP0^L0Zl6?qq$oXouMPwUv#V%65QEIfj3%LUnkP-SkwL_gSC&sgV?+|R<8O1)aF6U) zA`3Ss_1aW0OML+96lgb)M#;w$YRs`Uh<#D=>r%Z8n0~!>;ZQ*UWxZ)1gT5|Q=)Rwl z24$P!9<^-H$Tk}Xttg)W)_jmu90!P4p>6B$C|v)BJbRG z21MG&bMCy`3m)UkLJbXS1tF6)j1!wFORTl;9enJ$KrP`L^pAjAAcC#hHp*&H!~PzD ze0C>ySQq~@koFTx6?&`xvfI)z8FX)X++C6+@#xb|`2r2=d6^&eq@$-(ndZko8F%ml zYK-m&FcbGG6fo$~H`p2a(?N&Y6_4V}FbtB!DNb{)4z>Bh<>nJW*rk2PBaI_O8gz#kKy_fJz#Bh13kf@!j{y0 z9?$+e?ykWw)X?xyE7Oz4Y5#`m;6K`||&{8bNd>BxVR;E5X{GzbkOgz`b2(d9`wSquB8F%b8rhMHL-_ zTM^sp@Qn$%Ud##l2M%v(XhT?o%n%*!`IWY7V8#;HIxcSj9c=d=aKYS1GvZ6l1|V!! z9|};CBK$n(Isr3IA9EWTs~J~^NdVU4;`^8V=sVn}=AfRs?N9j2hI^@c~I;9E(J8cKd|?_IbQPE zUyLQrYiM|je?osQPN1`J(*SOztnbpGD8mKm$iklXPQ)2=JhMKOr=%Ry5MS>~{!1U(F<@=Y{2oHhS`+{y~vQOj)s|DZV?gQ=t*u+*eJU^oy1 z){+1D98#IXhvmZZtR%BO!O0$Qpa1==(R%8$UhN;3h9C0_ZL8X)~?1oiG1?gz){57%!b({;** z7I_Z+THdr&4T%o@FzXepG-}T6=;POl|8Rzy$q;Tbf<^PRlGk{uuC*wqk})N3B|gq9`7rHV7y5tvxT# zgu`6p;bQ^RGT-MybuT+w;0y~jS@)%hSCi*3m{0EB`Fw3Z+w!cg zL77Pc;K7cmF#4&%zL^O2>`Tz|hD#E6v%Bvrgt~Rp;u8gG93hUtYi-;C%Z*|q=_CvD z{&voIl5}aqR+afYZ(iKh0QZ>Cmrl^325*-HR<_{WvnY`qC$lanXysltqi19*`Fgz1 z=NYy<)lir5(7@XH zJ~)V#uRQMpOt)5glb?9v!{=6YmF2kvB8xv%?w!oDH;S@C9eAC*q#l8w_RV;th~Hk_ z>Zv$yXfGL@iX>fMAW%6xK;G|B2SkrV5#4jVI;}W_i5)P1@Px~CsGFDEp0*&!ATCe~ zw6QjIL=`J+x|Vx9$e|6%O9x_M41joq%T_Oz6|XfEfT`vN{S2u&(DB%}unR~)4D?V1 z$qm`!IUmz(vjJW(Hv2P8_=ij84o#aE|)^Jj&-^Soo!iJ7)jaSokr` zAJrSD~EW36{e%RCHwK_K#k1)7bXM;=fq)Ov?C}`$a5GzbH0|oFK=uy{!OF$Tn)1jR= zgMyg9Iz#T2ww@ly}sa3O6wh&slb4X)vf=ADX2*{ z8(>E0ePVhT3)YaUK%Er|DfDXe4!Nl4);SZFWcB2;yUKU=nzs}_h4JJ40Q?NYQ9r*cn%aGzn}{!Q z_+-K0<^bay67*#T3ogF$S4VfBsu4EAJc`d>Z`t>LN^>e7t#>}U%bbD)2rl9AuI>-X z{dv}86rF?-V8m?h{Z$K6r8Jf*=Uh)Xvrc;|L2-ze4f|&9N_tl$0_{HErEj0k16Occ z?mpnlTD-r4AdJI3f?X$$l`P*T|}-i?XIxA8%1avKb^{m}!Fp zMQO@e-6Q#V9{;?+J_Ut8i#gz;ZQl&)3vdz}>LEHfhY~2s+}ngusVl9eH4M254V*tcgh=7RVx`&AbZED%zi) zlC^HzegSJ>F}KUaGWm2uJ^h~sH{cE*Xt~p^@lUU1VF70G_#u(1+navXFr-0mORKuH z_^|F&Y+~r2=cPVI4EC3AP>oa+!AU@;@F3V=oSsGX*k5PopPvFF!KFu>x;Tn`^Ew{p z&{)s=magt-_;`=EG7X7kphx|(7JbG1e7V3{Hh&t*bqL6Tx%0dX@@?t!Aqa~y)6G{9 z#gf{J&I{nUqC}cl{s|_0=khUt@6k@MhELc}7ft#^u4N~+OVs7(5f>_jh}WH~QM);A zY4SO%T=!f7b0s4cbP#kiMsJzUl3xGGwe_5NG9_x+>uz*6F z=jD#wMCTiEEl3N{`+>@t^)uVLP?^GhM{SW=NQO-+hkEJEOIy6&#Y|r0p|>}0i7PW) zpl|nBBvMf53Ylh0h#~2X3smRQcUyx#uuO*juo%>^`FWebT(ww}-o&`? zs5bk{P3C8a$G71u3dk|p&RW+i0Cw?gx(C$5*FZcnv8K4 zly6W35$<$`P&;jKPio6v%= zK|uZH3qJN2{I2xgO_2eBO2Nd|F{9!*)UFRn(tn!A^&70wz16$t}_n6DFRGt zS@^~J%Q(vNnIqZw`B42pE9I|j0wl6EvL0>sF#B)i{CFhPE&C#&$F2Qiq*h2NT>u|? z$PsT1NF$L2z>X|k2r7VujFm^L`0_7#a~o-wOzMpm zP;?F8Qfen%*s~}BBK#?Wi(%Bs1#Mt^gIrgBZ8c*8{n`O%g?RS-+K0T?`Jf~<=Ju6GaU>4&QyI)9T{Q3jrmTQWGcTL>^>l&Hr9-KPtr>tC9W7{{E%QKgu zuIF6@)2jyDu5tkOX8g6vElDyU&k{GAp=s6toDxj@#nm1+7Kw6@8oLR%B0w5d=*qEkkEbvR3?S**)_;U^u5JKTCO=QQ>X|Zo zd{^TQyg#{tAUwi)as&EO*6$x6!=d+c1CDaIxeT@65>tQIL47`@t^(Y5BhL!}(nAY9 zBwa8=n3snX1#~CqYiiYX^s$HJ?FAyosINwM5gyP+PI;WZ2T&O)$vDRju0K!fcG?lc zhQEFI^*0xO`!J%rk?jF^rBGo!mpw8Zx&E}(mkN3*LhK+6cmV=ok)*8ZG~a?@&j~4s zV7lhvPp0qAG?xFTeq>VqYBHpsaN_&VEd){epZ^-Vvv-23ABr(}M+u-Wf4#Z>Mf$vN zWm*1*|M0he`)~gE{rTIH9}7CRN!|bL-~aKy{q@^F{`SAfb-8}>Kj!+I|3OpOL9y5$ zd%aYD;D7K(KL4+O`|GX$$DjZHpMM?jZ(Y75fZQbDX#ep)|8iKDab5m#ImaCQ{%=`5 zgKs?l@%O%e^G`nhAOCuPpLD-B&PxV9`nUhaVF3KelkPvi3a{j!T*Ke=q3wop_zc^> z^Sb)e58EG0J}>|N5Al!x@CWvHoFNIYXrN~vM&K9n@^Akb|J|0fYYtu}{!Vfij>Q<7 zV@MKXST_H682i8X|7F}BRsKhpC;cBo@y7yw?g0f(+CRXhEuT7FmwM>`sOLW@hQ%)jdF z7=UB{exb+<1%ALU#J?CY?q78{9J_GD-x>D8FeHm#n9ILB9`~<09*JSt-zl1*SPFm` ziXs2`@m|B{lefw518KhOdEWo=53gZf=6@_cL-&tcJ^cCV{Eu@t3`_M7e1z$wLFZwtCWV!sa>HeSA%YQnRx=YF&9s3f_@;bO$ z?vCk7Ily&ndVVGOa7o8L_)AFzK8C;lpw!vGAFRlgL)Cqx#1jJ<{CqlBGmhLnRXD(Y z+FDbeEr3g7N>E>;Dc5_wmx&7X3h%UWa?3w`uhDVm^JGf|Wh8iPA-ZZ|bO%3i75K}J z9c8fTYy{z@&)%)%1TNKEWR{V~BXCib^*IH*$e0fmLE2FnJiHZmxn;ii+unDzp+5R> z-qx#|54Bzo>-8-RWhvZ8&pb`49O$pNdsDNh4tYi-GD)r;o{eL^$unoTyzu1>Bss*( z7Qk?j3wNBsP?tKiLcQt&k(II_2&xXywyedDBUjq*HA8Ux>UtGIsT>4Bk>=agkpN|h zJ*Cn#$JwEW97w^H2HsI!1%U_eh}7sg1M1>WEwurTLS`&^ks9~V5~Uqix{`&!lqgm^`-R(zf`4&u{9BG6GT>YMdH>oMDDfjO7zcQj2rY_# zrLW|{74Dlzul2Kgw*D08?ZcX9z4GS3;rM44Losynf+(trJzZJ8TrO_dJuavPTUdy6 z#C42FOhG#CDVQY8=Lsz&^1?RF)A!w*g5mXrZ?Ej{x&ds=z-t|R^uY2uY2e*i*?<;k zv86f_?@s1*;1oZ5w({J?9JD^8(4r`aV#4(|%d$H_u2gE16bq$87k|v0Wk1}wOrcGi zv&hr(nq=X4=_DLsvXEg3`uS=!|LV$qnBF^G;=S&eQ{zIw?W>N6|+a|u>80#t5 zMuW1uw3%-z{wsUWEHZwZ-vUfVlsi4J>fn&w z?4Og*Zquk;RH^%F-#~jbu@S`R+X2u4pFsPyz#El)!BYnV5^}(H>Ok*!EVI za7!V+Pre>iO#X#_1m%xLIbeM9>}{=Mq1LEwT@ur{6cdnhJ-T))aWN>?^6=cjU7jiR|Nm&>}P(viYCO z4M1(iw(V)=&xPi40(~+db^Tw=T}Qg3u$EpB19BSLV8G-IW=1qQ=hZ*?*6UtD4|-N# zT`Yuj&L0%_aG8w}HH@&oMlno&%XEbBHwEgqKq&`Fm{9<%LA^o!F+cHI#UZvEIISZy z*!W6Pk$Z+R@)k+C1_s0|C%tS${y7UM#YW7RlzvC8^Z2j(H_#hlU;M7DYh=ScZcv}M z9>Y1bqf==QFb_m%jaLbC=qL4*xwQYa(ec$xQ{!?o+&60O)7VACFCoFOk;3i9qDG18 zUG0{?qvluqN>SbR9{Qc(0h=@oz{xe3{KDPvXq!L|#%r5T{<|qHA~{f$g@Yo(K}gnb zya#cTE;ziCkT3pj8c5Ize$|fU*$*jvl4=kBQ?f+4dXFHlg1;#*T+%ANA{KrSfkW6k zjKVEy0<0g;9jpA}B{v`lqWLrRgx{9L-%LQqlfVf#qY@UVX;4VN0Fa?{XDZ~CffG!w z8xo3Q^EdAO2zk}RyM4p!^IZZ{WBJqLw@XHQF8XL#jl?IgZFMr|G>4j0DEB{ge*LDU zn%@;(@bBuJgJ&SaFqQXMai87!TZYun82#?l>R5}yd;!>rmyh|oO5=#MuhZ5-tf|g? zBzr-u!R$Xa{}B~Ijj;9|qpfufSSN|RK1i<| zpnS&xUMPQC23HK9Nhp^(a!tO`@Tg0It=eh*EwK`{nr=a zfXeh7jlSI&4h6ibRimZo#y(P>`<*adq4DVdhOy_AU!&QGe%SRlr2C5N0a(|%I-{j> zi%!D;N97+KV{(83@J@O3)Pm0Yf_}^MqV1EB%6qM=!+Gv@A8^3qh#xjlPh|c=Jf%98 z+5j;~>&;~=kIxWrJ;@ZmYbLc0grE}xT+X|{A#=@rF`5^KEcZL!o%zyV{C6F+c!D=S z#osoW?vfwnZ`B4<^nH&SY~Fi--79OxaK8Mazq6Bd33#2?M|^%ij!z26|3Zr2Spd9bU@qf z1nND^3oRZHWIXumCVm%Jynm~h>ifR9zvJ+Y8Y%c+lhq`YNriL+mh_|Fb>Kib=)Y6O z%<6CF5N)sq;6spX)A@N%5WhUkz%+fi6``VlDe_U3zvTcUYsQsY0^hEhW-EYd{k!r} z`3+4oi{-kfI}>yGyHE`PuUMCx3dr>anD48dE7bNR_R()hsI^BdQCb4fd@x-qN{sFe z0~v>rK_H`o27>sSftHyThS45;HRb&XsQ~tD5M8p6=T%uAQ##M%?-;N4r*Eo7_J}wu^YTo(E*o^ zJ`#M9w4WyB#kC6mR%D=i-vgv6KRx>szCiZ^$y*K%_j+kS=Qe-BSNIdY-|wF^E}st| zd`F<2f6~}_yYdgPbNKjz& ze!qXO1kYJ5DDd(B_8Vr}QeKMKbOYIG=<~Q{D6smVX$%%%$rRi3`%KeX(IPR8wLFNu zxb-LAn^1A^ryhp<3_fOLTCE8UNDK%JeCd08V%rTJ9@91K|b$ zp@%P0BVZ-c6*9Z4H>*dR)`ItYz5aHYrUUQ)L~}rh@^R!wE|ltqYsaS$m>qw1Y|rYn z5%czCapY1Wqb+D*Y+I!*s-}GFp4#~b)G%?Y;Z2#EttNJDxJ+mk%;M-9{4}H{j~)U1 zkqV&PuR%1XXG=4QGKLDad9wH6Uf=2)BCT;TERDQagDm)Dzg?INAD?bu`*@5zCA!m9 zh&RH!CGP<o;53dEm5DOy&%rUVI%B;a zA)kYsrfDN+R<=duQ9C-^=i%r6u1y3CsU4`*J+~w^)tNR_!;?^(Ovvoy3Fp2fja440TQ+cq`KPqn%wDl`F}1q$x*pOB+l0A%+pPC&7NL$|^SbSyJxV-YI7R zSRy&kwhemj-}TBGO?W^DJGb0(E3+)JW)wMCcM~t+RDbu|AC{OJxahksQ?%HkodSAi zVP1JzdJOF#UTx@P4aprJvW?C;+P3ejgb7Mvb*p-+C6Gc2pzhgA(9-_(V>8JB>M~<~ z(Uph5(xuh5729^&-#l!5^9-JzdUAla8;8M`6}^gLG`}Fzl*AWN?Bcz8gT?>)rZEzB zT2|tFVz#F~++SaQLr+djW)(UEyE5=ykD~HYQCJtP<{mS?{>uH@Z&2>Df~7Fv;eL7? z4F|2Z6(Lv~^uczHao`v6*fayC@@euklj7cmZy3mms^*LHm3H&aVm_?^N>5t3&2uDM ziy-zqN>}#@nQ4wG;rr16Pv-|s$nwhzp?txQIt8_UuU`NxV~P9Sv=rJ6ZS@wL7$fGf z2Z(fba!6G4@BcF73EDPFH&PX7l*weSHO?>fO8n>S^X~F$rdn-%>q7ipZv^h}l9xo& z8nrP-Lk4*ap-!ombx}#%`dp%K0DX^aYAb!SJg)AAmtT^w>e9lHSMb%89PU!9gGhGt ztkVQSD^kGU%FC*|pp9r1lxx0l_N_LS%T9V;o?Ohvofu4@xvQ>vfgC0?SAfKAfsN_w zip)w2@=@|>CQ+DPv9yd$!Mj7gL;Er$eMCzHgn?%F`tMzppw_?sQU^F3JRq-vbZ|im zU8PXZL#LZ3T~l&&|2@kqB^tLQj6NDPpVe8}+wx;`MkG*5WccBUH(sN}zjhqDYaE2G zh--HU0PnOG09=*tG=zLYHW^Y#n#xK~DtS{%s zc<(}qF!Tc2DjhjDK6ebYDK%{motAJdm@~zfxG@yfjPT}W1ve1lyn5y3fY|p>6J@&$ zf{o#O`V4WrHo40iCfg%+{>nMr(Y~6xUx#a>qtH_UT@Iqwn?|J(%gzMa$V_@GM#7Km zuku6= zMp}La(r2q^9}_+GGpgq|esg_)98i0I1R*pi<1XwjR;4<3x2ry6o!1^oaK9TcQ z1nVdjBNQcl*7OUT8f$cJ)GH&^WLqg45=}&7=T{D5hCZ^KqGMU*wWvOKb_=wBEPx}- z9eE!2@tkx+6=7K*{DNHUXzg=*n;|BtLum%;qHhjYd0+Wh-pmFY96!IUvXqNGl%*%q zQRgRk0-iA_`_o1rI73RcDwL{gxj;8W!xhsS!OpB~Fc(ovsYCPpt-8;PO}67GC7G6E zF=^WaFbX-Yv@b8+;BsbB)Ay}i#mBQquN0jxF^?tu=(x4c@l3V07qGwR*?Rb0qaDq2 z(#wt1YMv#ir2gvNIZmCH`K5(I1gli>wq7A13RK&J9ti$+a=>_xeiaI7NX(R=oyl8On&_9PnBKAnq}YC8~aq zPT897Yyh+H64facU=Ys<0yp<* zeYw`hj&z{R)e8z;C;;Raf5$Q*RWqWWY#d+cbyfJUev2gJy!J`O$a0R7Uc=KgA>VXr z-}-PLpX>E1JLR_aDKslft^yidnPq<)Uozms`_+kSbs=(uHOxE53)K21NxRgM!{@H2 z)-v9PXQ*yYl!IJ-K3hWLk8J9vn7-usOoLoa} zeWDuqZ4G__f0gJ&5$L-~{5t?1>@t#i-Dy{Oy^P>9oWyE~)*$JZr94)GyM&F(?{m7W zbPrU}2UB_wK9>xXyx;jRup{98enX-eL1Gytab+sHcaW$KnGCOY>aM4aF}5+a?(YJT z_>(H7t3`v)#-{>){sF%5GjQb{lfswfeuY;BE=$t70FC)3tYKe(F1~wj84*LUM%rh5 zfH459-ni^6dwVKG6O)+@f=L6gb-HtfIm)HY16U;;j4bZOY<_S`+b5xpr zpg}E<>c^u_#luZ*qzU+5}Kph|@q}Le`*il?WFv#+(B;F&++%oyrR27tYM!ZgGjmx3i^kIXLZz3WIV&4P0T zlXit5!;U;fOt0ixWhQ?zb749w*p5wLWqY|tRQYc#NiTIT;D#KQWVMW^ z;i>{E0n)ubAGqfFPsnwdc?lx-QshOJVUsW>f&Yswbx}{p2tita+mP*=to-*s82eJ@I!lCv!l|bIgXNf@jd5f@>Ow zYm_%rz32lev2m5+C_8}efP@N;Q9&lGV~(;7yAug@*Z0$2pdSLdx{F()vGZTsR zWPW>J;QgVxR)fR&X=i@{kCz^m_Tme4Ss#rVgJTx*dcVEti8h^%H-?Tc;M*O~-FXCnTZF6<2=cd@sid6cPznvdndgE=4jUmmHb`3 z$*yBe{Cye|w%IZyEz}}@iB#ExWqWQJ!s>&^Dnr8?pPQ2OS_`b=H?pP0EN~obaYjD$ zO3%*I*N&!7y+q^y8gXWPTfehgpX366a%`wsHe6g8w|)yF+1kH#gy>fBy()Y661o~j zA*l!?_y7)t>tOleKiGgKce7ftn=_r&nL%%}EuB_|M9hT^(X?a(=}`LPiWR6LznbWMdAgV1@iJhg4kM@36rt%O zt>emE{xpcEoZR)&Nx&HKT08h?$$jU5fHqlw^tA~g(l>iVudmzh7^V>$RxF*1XT*${ zG;W64xk>eD>$&{Uoedf8e&2R6xi@G)=?@%T48y0?Z0##`5RZ=*Zdi0s1yNJJzhA5k z?H>D@J9C|~SzTPnXzUHz&gp>$04n!ye_GP(Dto&W)E+BMW23jj0uhOg)vip&;xO>#T9n_@Ty#IY7R zwzBW>9hq6_-ejzO<3|TnIsToJT7&7yJ>j09NZ_Swh{i}m+{-VV7^aSUTjAuNR|j<9 z!x+wE1RT}vj}ywog=soAsJ(pkV#%IbZX8~*o;*NG|K8SAdL-dW3D^;$YW&#rF ztG`U&W!1OtnYwebhmxAwk?7-Xz`fJ@yPn})l<{MU+v&Jm>E;MBmhl^?u5S34LL<)IcigjVHg#(IE==0^ zUDgsgqDe8MEAzvwt;x{{50`V0lt4ax?@;B-^bA{Sh`QMHN;bgXsMD3`6(At z(I6s(k(+9NK^7Edj)^-lY5m*JO8RarU%i9bCEGDbW@30?OK-bJ(7?Rf_g4LTI5!1|~@P^coS7S=1f4^97 zg=7~m8m${geD|L>OyyrneUvAD)n8{SLZEK?!4FObeVF|8xQ9F>NxjDVk$;!hXYI~i z@ctPuwoLeBWh&Y(Ds)=Gx?tc?>6QmwL2c=D zqS$J%3nhe`^B^JQOK>9&JUc={?Tu{1#UQufJ7L#1W5j3E7AM9zW7u ze0DJMxoCcgm&=GA&Cuy@*k?9p3puPMq1bo7>~Fr>YJk#xDWawvOFa6o#OB%sUkaM5jzIOdO-x(Enxw|oR5ymc#DA@&!`Cgn*ttwPD-n`ezaUpCDk|MX zRU8D}UI;~*>5bNYhpn%R>lV%sJ9^DENz%u)J2v|Bw+vfH23L%r7z{OAaK77Q6NjVmGqQvgxVsuIfX#CxJ4+--Inh?721% zGP~aKQ)!}gi8M%$4Z)y?A@N7#WLCg;DyUz&J5J}$kho4_9JXv|C(yIyW?!zgMviNQ zS6@{cJ;h$uKdREck5I_w4NKg?BVXk+cyF!26ndr8B)L&(@VKKBm*lE3T5oy0LsiGl zqS%}%j>4eO`1hlSq}Cn|ujQ`ZE+o+){3-bpMJ zx>(n%kI)Z7?pWrXOjXwzp&K7$6u8^rA6H5VnC<%AGm-U|LCv4NJHt1bYpX^q=Hp3+ zvTRM#B8IxIs0ng;(x>bpDm@PInV>WOjWV$x*L0;LBadD@-}Ovi^pOeQe(FH9)=>ec zcZAydxyKMR%Pzxydz2J@>G)llRS0*GNXEwyBBcjWkivbxqo-6utjni)A-)~Qbvr{r zmVE~nFzVw(Lw$Uc8~PZJfibm-zLD>9arG{`bZ;x0d5d3ELp^#dOYky-;m&mSaED=- z^ie5eRAhoNl*C{>=C;~4&z8bNMe!2=2wNl_Vk*Y7` z7sh({dauki`3maxM;@8pOvC6gjqPdkUoGvElivlAX3W8kO+2hd3wz#+=1qS z8C;q^cJFeC{8M zCiy<&F^?Fc2PqMciQYm``?hlm31jFZyI8mpP{^P7BXoef%~Z8T;IF#fE7kN4Lwsc= z=z#pEXu9qq(lrVfm#UF^iqe}+!!I@LcBgG)Cu%Rq|E!?J!tZMPW{hS>i``8uG2hvt zn0BS=T1<*WfRTA03#!S3M+)B;N9p$T?2+`_9W?uF_@(qTgwNS`ES!4=-BkvW}& z3GduW!6W!IeFgE8x^-TgpvSdsK8sCy^?EyFK@{B2gI>t=fSHm5HYU&_<)h*ImAESv z)p=)re+C<35b`8JV$y6~rxt~z( z5+H05CTmD6ipIIh_nctc?uOA-fwIsp_^Q|<1K`2k6_N&DVPd>@ywd#Hrv^_Rm3FCs4j0+GtG z(lwKP`x_N~X%^?6)P`>d1q{T8BZ`wbzOMmfu?tIef5Im=p3h15VpRD?xG|8^w+Oa8 zfHc~4b15{e_uYlMHR+bQ!z9x>4~3anuD%?dS@%PDNCQ?x)`PxAEx2O?Z*ce`K@7zA zN4GZ}SYG~)AchV!IZX@^*l`OeO^#`idy8V*6*wWTfq!)ofEveR1G!QflW1k#^zcP3 zF_-DWBk0B$dS41tBWgJ7C)JXnqcC?}3*wtl*wNH)D9fB_*C60sJ}@+AL{&G5Jej`1 ziwpbNhgJQY9yP`vdMgLv?O``2k?eLBq>?4GjftOt>LexD?4?g3HGO2@yVI8+x4gaF z>e&+X4K0so)_Yh(H$znlv-rRz0nvc|G$}~tIzS?{koCB=mtgau`?EzCRd;|-4B1Ek z^23Gp@tuA>zUC}s1B4-Gf%8D(eMGZ(XbH?9O)miOeoh#0BJ;SvI>hV*zKUO-3e?fO ze(4@Q?D;n?v|+|&KL*ZqTYa+_$<)YapEpEn0NR@6L;K-tp~6(+jD*jrQm3K2L6fzw z0W;#`t@-7+LfP!xY+f;(?wjZhli*J@%!3V9Ll05etGOEH7`JYeFOTgf!si6p2VmoJqTo74ejG#l%s zT~5)$$a>ex`M`^}M^gvs3Xh(bm@W5PCIl1# z>cuXy$gxiiK=L*$}Ma+8fcc}SEl5g*rycX=q1orxVff-mYqUrhhr zhuGY1IP30PgWW9MrQ{w2)JVz?oWPSeK~0!uZ;F5z|9*cl6-fVlT$vg}_b(oC*mJ;L zc4skK9lv=}>>vahd~;?26JIa2K_FNtzs?KGKSH7=31bBWrR{qIjKa}NV^WmwBaZmP z@At?+obt(44`InYnjbysc{b?7vPNsa`(YI)eBI|TPPVYwi<096F~=k|d}Sq}==JrE z2J5Kl2--Kt{p4IrL#x_pWvM>Hbh5v;P57Z8%t?`5X(hhkP`X=A6DCxg{jdyD6=khU zd?}pi6oK6$gV&T?X^!UpzN+6EHomPws|cd*^*8pY3-lFD)iZNm%9~kUd+<`q z@Nb=*;yrQl8ifb+xBf0Zb%JumkyO|HEF#;f-yc=8j&)=na4_ge0=L~v-v@tinf>4S z5aTW>qF!X(Vji@G!6)-Z-f!%R0eus%vR|XW=N^XZab^Eq6n6G8?xo4k$@L}x7dl!| zwXK&n_^yA?zRhc%Y>B|EBB2V5ac&|)@8{X2z3@^$a=(wp_f2!(z)~ylyc7~XDOie} zHyqNuKNe8$2_2}D^umN@xa5E1_K@XHOv0a3BJYuqmxK}SWlM{~$Fd^li(oxvycXWD zzOS$MCW7Bw;Pe`iXRw8yqAqG_H{t8`_WUSiU&PBd10*Y(_S#8q_{}l>#_pTwo!GFa z&$NRrIQWgjDDzk{YOj$;=8ngF&NP{diQVL7SBahGmK);Sg(Zua^A1Y8UoA?{nkQx;=CEQeWm^T z=osYnJc^oZV~C&fbmb>a4EjKAw0DO^2XnasX%EyMO^cu!=taLM`gq_<8{9LseG=5J z9t=8VMUyOSG@x!p#>$z2K*iiopfJ6(gf|JCM}=hHuATOaZGa9GXv|%`#j@6~yn2x^ z7yAM+r??%A5Z!67a9J(AK>VLkJRZtM?;qAoV1OcWJf~`eCL+~vk=c80qV8$9;V5I3 zH#9^vQ{?r-bkWVpQK>TD6S7b=h?A%Fm6qw)N0VT*Nav4z2=leWC1F2#Gw!>m zk56y|rbPRjM8;I#Ob!D25+T{4k|zSvGj@g;alpi?3|SYSXcz@2!UF-0&QCXgswnEZ~sJUeC(i>9Gso?CWlT^W|#1NfS~A4qr-bLgE>OW)69 zqtF1d7l>GmJsT{^Af?RE54Qqfe`-JNcI}xoDAh+R0wA08z6>ZAZ8!gP zgv-}@Bz}^DCY8l%kKU+<@_ZgH1(h!;Q)b_@S&y^UbQxX;VRH?i6Y?YD`Ep+UL1dr% z%cWt>_9bZdN17^Bp?e`6wA;+0J=uLHA!9XQ_;a5CU3F;mlUO${HPt0Pmpxi*If>DK zIrMpH4jwXhb>Cr<2eeipR|-F)-*L6v$F$*|iu(1YkR%K8@4Xpm$V-|%;E6DPO{g$> zEPPuqHilICq|j&~cN8}_RqAo}OUY+MgHlgPv(6PkvXz__eciT#p6Y!cWhTZ;m2VM z$JZb|Gvf;|SLAJyR4=G_N%`V#>o+Dd>u;^VgQJ~dh|yFZarH5p7Z8^Szfh6BV1SC= zmLUp3Nz?u?kr$K>NY6vCye6!Am;;qnTz_eR6k!X=wmEH#=h6DO}LZLv1nW`QprZ26}dxDkt42(<~f>x(PVR z7~EsIR*jXxril`7KO1pe1j?0z?vIGte@<57nzzIUYIG?rDc)K(;MBzmXg77~|6k z0zc0*@CB499Et>JVRM#-&@24BmsM4Sycc7%}5Hh;45P7SI#L<6}ZE zkgKN%jO1bw0whbxhLhO_ec%*}*Ld*r3d$g0I3`1M!*am5s*2gPRaGk4mm@}l?TQ)5C`7F=@Mu4<=ONQihp~z^6ws8)9Edx$zR$E~5_x-Dpeuc*U5tTK@Y7Ywb#JmspgI(EtPw7h=~O@K-; zo@&hv<+fP9Mb)b*n6JNelN$YHQ8o()SWQop@a zPH!=CU1va#GW5|53W~%1ZC~{tgaTYC_}*_=CzCj=FAD9lzC0G;p`ifYqW73^qV!q;KK`JBc@gy{sSp5Labrc zJ^J!ysxi#{jyU~Y^;W(EVTjcVS!L7@CgJ6kkbj70dnaZ)0I7aAVy_YDMLoYX)Ec_$ zVL4J%lihjnmct5h)V-gw56E|hAQ^E?Q-R)eMsjat`Z#x>cNcAv__z0v?`##TR5FwC zx0Y(~I&=CG=ji#RWJZ}6gCdFyS^H094dr##um->%g@5{I^@s%6B#8qe-Wn%t+9u#3 z4rRz78+ArsEoqENz8mmT>?8D9E+*Q}D6l-S2J1%G{0Jrhb z@QM<`FS3-(`jTb)GRlWoA#vRbJ`g)&t3C^hfZCqRO6?8f`ib&=N^a+) zatU07_pd0%XJ*f>HI4W$gA{gpt*C8!`%NDIF#(Lh#g+ z>m6^+C%%}lMRa`F;otGyFB>UpYMYdqV)oyjX%h0oa8vs57&hn2k`7`RdH;UtvCL0G zhjRx z%_hvw^KO;~tXDsn!t~tY$gt5{jUw26PAv9I)KwQDX0~{4Xl2AR4J!d50PmUtt;CAo z+;*QtNQ3ZFTpysLCxXn^tedk=IOCsX^AWrW#UjsxjJMHN%@$&R+A2v@*f~EsC>c>il3LQ+m8=(-8}cQ3)hoxerd}8_5C*ci%XWstl}s2y;X(_yc&JR(uDp zC+LC+YBl%CZu5$ARL4H3bmpzKi|@`B>*r*KB3d&Du4s0t?fWeTCrMJ3z`{^pq?BYc zDz}DeZb&rFRC4b9ZU;-RxNt{URW=}58_CH}CS{Io0Z1B#?uyC|!qyN&tf>gO>|;Dp za>nNO)c3e!S8js7E_sJ<>V<*2#{FJk z2??4(COPIkpC;FK7|^@eC0P#&=sKwXon1Kn=F}#W(EL7xB;5(%16k`ZnY+Uk=2%at zh;cj@Cui+=XB?O_j>OE$v-B zs5VkbXQ8>%I6=jH(Qj-KPeT(1G%gO7ZG!Fu97C=GO+cCPiZHkSn5$U&lz~g`dDL2^ z#;@=9;o1Xs1_zeHs>ir11lK^LSD#rSmQKz03xBQV0;xEDKjPX81EKwn*Yo{;@cxWy zs(X-TEfYZG55C9Guh*k_3j6T@4v=_Tg8kBgB#mF;d@m@Onf6ia*+f@jj|f#p%N3tX0YOKy(_33YX$EkLYTfOm zf43bgEbXIkKvV2C!Z#dvtR)N9NA$7e+yn?>P0Rb;dYcy#vg}rbyxkOMDH@+Lk0OgT zAAP0lOoJ+KPz7u6R0C-3gt5XI32+9dx-DA8Qw&pc%=_VEWz>11=<~NX2~VKvEAb!i zZLs=;1gbFKpg;EoPnNLvXbFHQR5et8VAeHc;=~lZp3{agw^)ejT0iCdY@@&9P0s>o zFgPqn7}Dw@5)(G3aGstK8p#(Oa>-VpqABM6pXa%^c2T@or;K-QO4@;I>RPjS;>C67 zPiss}P_a{+tc-A!b$neua4Or$Gt2Slkd4UthDnL08ks|+wogXZ8M@?nt^Q6<@ebZ_ z-%Rb$bTx$6<>mtshaav{rI`Y=pRCPM4u9bVw}8CnfTBVYqvk*}guip(?lhNQzcVWA z@MO&Q5~#Uaogd)uUX~@6v+}hxsd{%Z;g2HM4xvz>6wmO|NfzZm#7B~*0a41vg7>rB z0vveA0OGNe=F?s$vp&Qt5-8zJ!@i#xF{5(!@qtzNvJqVD zG1c#7T!*_a`q){5Po)U^r1$5%8z&_zc-?PVflLWWU~}M?+&#F@jy<*%hvGXDKL>dk zH57!W3zxrkC1cKdbSMB7NUY=u=nE2fzXHOSJTH=!gyvp~bs8x);9q@iIoXZP&a^0C zqJ)6pe;0!GSXMY>=2ju|!HaQXaMTx|;@MBEL>h#cpr3or_|Eh-brzreH0UqJ#%ht5oJ;r`y}$-)hm01xSbsMc5p7L_&eb-%vF>MYPIQ_ zg>6%13n?fX<-s|gbT3L8l{ygrO|G^XqiPLPGGi?A_(0t32DUv1ise%kh;Sn2dFm>Z z)LNT(_E_^ooUdGp@`V|TM-TCAdD+$u-NNz~mNEG|PN9SPbmUU~LSIg> z(B}KCv~ieB>i4&2$~w|qz{SIrJM(!!WfJ;L`t!P(vtW23V{gxurHRndR{6~E0+6G|Rm zm-b@Zhr!!UC+;pEb*q}b_1`zKl6eHR{bLH&Mh?Sg3bL?2TZZoBg8 z_BOw-4bpZ$yQ#;vC8s?a72V20XMCwbFuw|$&Dp=A5m;O!`yeukBIl!Sa|L6j#7iur zt@k8)oZbI6f2B~a&+*#k!#!Ao*FhgJu%Ir-s9_OoU~TFIY{7)0JI8mSG!^$s$7?w{ zFZe3LLyd4cKHZV&hRbi|xRz0-L&bL-TMg3ZxiugR%e0-iyjHzrMe7e zb#bv5z`D>2@P-jaqYzj`h*Cc4<0YW-c-){h9uyTo1g_^zxMuM_Ft-Rik&}|Q$8~V& zBH2;gUABvnKk7q_&6R?X1Yv*FYSyCeU$LwizD$APH9tiYstIdiO-RR!$b>6NrV>O2 ze&2Xhxso~FRi?-(v$_}wl5-YRiz(l64b7GgAtn5p=C=`<88j+w5!s4yakN648+VP@ ztRD$Ls{-etCC8!EOpS_W6d?t16mI*m0=jC58T@vvB9~IL$yNat!Zp_Q>)PXv?Sxin zS$51zg)+o#;Vdc+>7OTGWO?B`vvLA1?Sq*d$0+!iICml4g_3q@#(|WacSI(JM?I*w z^QuY{C8^my*vq1S|9KvD1aSs^-N?_^<$5t~mc|?_YI8%+EMpUO?IjiC)Y_8Hp~N3t2`nV`ihyJ60N%%wvguN`N`J=wxpNpA^|cH?(5jZ|6JZ(_k_D8#@H~^3`gwQWTw(#mUEAJLr!$xh@Ji$h+LVvlVjq+|Sn%h% zh|-tF<+D>IEj^iUxnj2pJQfW!G?GN)1dd=d8DT4tojS>~SfR3=xXe{QC2A6p?LrA0 z?4~VuD_x)~B$>(k@r_la{07$HqIF1)t>f@CsOn*vh?iR*T4=WzH~D0D=;PolE2-tS z3+dL*NfKOr87J@iukN*2b_NLU<-$s-FtXarGqvXB=?z&1eh^}Aa9rot-FSx``p8RW zMHXsT;&Xmmshy)Cc9iQg3r7Pq(l_aTX6*v^wp?3svzUg)hS}EzvY0Z{y*~GCUZxE; zaz^=vnYkCE;m>Bb+K*@HxjJq%g<0&zVkqjBWI6vDs6v&`mGPm_3j*k_JGiY`iQzZ`(m9Z4rzlCr z^oYf&Ty9ksi^BX!1x8w2kx)006!+L^<_DZO-UUlfoVd%~Z01S`d*V<>zSbvj^{$lu zo4xzziOx!!Bw&gk_Yd&j9IEqFO3GFjeDjHJ(=OBon>knZajj-Du}i1kPvwd_ca@xW9)Vw_TLS1pz+LQ!_i?PC_ZQH;y|}qvsAzsJ>Lps0G7(`@&0{^^n^^9 zle+h{y%gH}n-C$4XYRU*((U4NxGxA^uQXEJnSwxi%uKSzF>dTEHLmK_bxs;>BSy>p zWMXGvy=`s6#z2yQyy6=R&B85Zd1on;6!TJIXf1Jv{@G$#s>QCjwkOhzO24sPEIYN) zWUgs**_|y{W@M2KPgtzv0ZG6e6j>mnnZd2wu9txFyoRyTK?rGj*wx z7g@YNxmvSp7F;*1G&fG?#!#g*ScdL*UN%`qSZr-$B9ls6xm{&SXzSv9-7)z2rpAd! z=G^jFFT+N%#Mo!YlLsDPOTX>zkYMCU#5)|+&Wm=@ z#Hl>CEG7jn;F@&C%rrUfT7fxLN-?L;yKB4N2BnXzZtbPsN`l9S{e3pDcH$66OZyNU z1Y6+Q*>Gsh%?U#sZ^G`cmPlTxN0@*{48&b;g~Y0z$}>l%mQ*PMwl=<6XshOoZ4* zJ+-k&1{8l<1kPnuaMuxXe||Ed9jW2lec7ECk(|W5-RVe`Q9_bcK!!Qlhd8-#;%u@z zGgpJ<`5^a}-U>{4bGQFaM@L==0&8*p_?VqsFxbaTmbOJfWQq#}r!;z8Y?R~agj?z< zrOgnGuq;uYk1dwoww)<$mkL#lWZqV)9g1$RSrQO9k1@r0PAx`g?3N3#R$-}brh7C! zzi@L8QI#CKr0MavI2yrWdcQB5pCtQGa?Qb+-P^IAP8i%6%ltuEpSR9P78xI>S*}$j zCj}k_vw}%+d)2dNm=KTecE-02SZ$IW&ecJa?U#DzluLspx?_*=T_35|^cG1asvAIW zgV`LisXp5W`gOlJN~GhGU1nofW66tw;!Jf1QRmq0N{sLKcM#Mm{B;ss%IJ`2r{ZvQ zB`Ijph`N}LRE!a}Xz5B`%;#*!vRQ6!RdMF+*AZ7`S=MbbJZ&aCjqGl-sZ&K{g6Zo; zAJ&)Rg~b9n2LVo{#g^R2)0$&848KL?usamH9a!*GyD((}=ekUI;B>B~#RYpg{57Du`TdYq4gszTdwJt05= zo4zl5ZZn-vH^fHHWp=I-1-j9y@QQosvZlq^qBTjrVDoI+u=^@M4C-1;?Ptm6fWqSxsau1u>_F-Ab2$X%6P8(8*kITzuj%oA#RV$5< zxh1LII&#qV;mL5eD_0vHO|BA^kUY^`QD+wH5^ok%&cziIeKqIRQMp=?KI@p_TJFC` zWpx(NZn~JRuHjZiV=z-z)5>*}3^!&Q#?+RZA3;6`HQp4+J%n&L*u4-S3dFfs6O zb8m0wwzEbYXR#`}eJtdWcrvS)&@p)=g>-pr5UWkylOvvwQA0HoFLdPPc(id2$6Lq; zH*rxUSW!@xD$(JInJJn_xVGoy3w$<>sBBzGow12kK8-Pb6GscbNov}p^lLmBhh5UG z+SN>unR%EnR>zh5)gp>+vm$1#)>mT1dzTH{u6g-FfPPLLw^5a% z@rDhZuCbdM;e1OyGPHgQ14&*L%We-S9jH;hUpM_Tt$DG}v`rLJ2tF55c%=-$T$@+cQ>3J8UGeH0@RdYZVdVWGu^~S`p$# zY8^rh2`FwoTSzB-$sxw(f=ox(vxg(<#supRbX|MJUOg^kj>P=MZa%dUspRIC)Uk-` zBj;@HgeP(h)?4ICG!~J0BE~s=I;K&ls$|zvdox$rgWg;T9jsBbWda6z(<^1Akg5^A z-`=F`G@tGm3=3x4cC>4_WHjv$taREE+ z2}xk*wLaZa^V9X3VlugC=YoC6(5db<>*XYG4K19awL#KE>0-8RA3}m@x@bf3T72=3 z2ElHMfKeFc7PZ2Sv9>dT^}8$`#bKqcjD%1fmZSxb;loId)1w5|B3Eut7CU-Z^xyl~ zImKzku7p@RNwb4Gi!a=9w7jh6$Nr0TC8z(Y`FJO z5Oynnyl;3O=#}lmlSNyQ)@@31>Ux(@ynhexnsBz2DC>RL)WdN+8qNBzz&+-Y*e#2F zJle;ctsGaAk%#cS->k%4m=n&3?dwEU!Xok-OrFf*_1LqN+#++9^{hpmS71F+A00)` zeXw|6?beg@3D z#HX~2=9*Q9kz+^P?7at66wB5&tcW5YqC}M>StUx&qbMLqP|3j|iR6ru83hF;DmkMd zL2{5Bh9Dq0=P)2)fFTcY;y>!S=brcd9lZDc-#P#J*1D_LqPu!`)wAoV-9=6By}MoR zUh|Ia=E$~`fm0=dHRq@jpJrG+)Cwa$uXBCQ;>8){p(8c@{D8pY9VhR1W!?v}j2~5H zzTPiEIn@H(Bp`2x^0xxGiDK-evD0FW4dNDJ)uM#vY z&&KJX&+i4)OoX<}X|&0^-H_zp;Ini2ICPC{aanVSJ`3aTfqnW$XkWO9p zeDts5xhF5|zv&vxahNiFdF!U>Q%+i|F1e>8Aj9Ni!`dQK23oes?lmCsf^q`vpG0QE-`){fQJ=g6;J0G67`aneahGSw5-!fRh%%5U(^Ijgd*Jb5`IgxtDHbTNuaiz^@_lH_4) z8AV}a{(SKDIenr1BB9YJiy6|;d$wYEgQzLJt)dCXAvPfzH6F&Bda|UY_b%UU0)#7y z-Xtg)1#&h&1tlaj6iaueQaMHl>);-bTuLbRdX?us8NYl}l8=Wx_^vkgO7Omd>h-7Y zSXqA7PZ!=~If|v`EHpe=ned7fJgn^RTu(I0Nd3J0$km|t3$y)BKWQLGig5W0v5oY# z9V?=0oxDBT{?ylFnsghFgL;EAI9I6FP-G^Oi{le<{jpRfX>O^*N?H$;Emq6X`v)E~ zY|j@s1N5G#aN1!XR<6@T^dvhnL$7A|?S5c?$akJsQ5&+UwZ9=qy^~e;NVj+1JF7fa z;$+Z$kewT`{(87{)srq3u4SGqT00K}(BQSvp&J=0qK+AEFkzHfI^^q)O_Giu3Yi)N z1l|nbic@&5e2^EfqFgF5c{aIXM2)C@UDnl(n9hMgrJRw;_HpEli(8h^0XLh@gPH!W z%90N27mVhp*;oN+S!xbp%3XSRvlSv5F_WqGV07Hs zYSPkqTm`!z>}W(pv$$} z-K`OIty})`*j}8g@T{(~ic=FCw3E>K0!+x!E%x=s_eUdbU!Y~gDG{EdPGY&&=kMeKE z`a5MYNpFN~zv?^|NgJ?O=rX)kRqZ6mboB?BRslELw zAudSQx$kUvHWBi4ro!~iZC=ht8@@NmiSLJ-KhC_kvS0GCq@-xbapm$IqDx%UT$1%y zgU>zk$#kHuWad-#7*BtD&otD==7p^?qH}bh+u|b}-l16V#%n?j?IB71^yADd!Y7K- zUt(SPluqLvr~WUhI%5hqTsSHzjK1pVz9787$d|U25CW0{mJGM48HbN?65*vDw9!_V#-Ve?m;_`L$Yn3jXJd#waPYKG?( zUaOf56;+w(e2}-WZnQ$$&XeS_k50FlWwE_Teot%m^16}ic@A$nAqukiao6lnQhVE( z%Mqr90)%HBwUT{lq@p&oWaaop@1-AmSV$VZQKp~wAnlq{h`eq)Jz4Z6rE`?;auPYu zn=I;ehdb4tB#q&v@B`tB+&0VNUuv7pJ?_`urR$oaV zu3`o68syzyK9o}0c^G4#b&LPzvW%*$$i*n#1m(S@>u=h}D|8j?R+ZU39+WIRwd*C_ zr=Q!SQpUZ!8(3PHqx8^Ns7^mVG{ST4{@UjXH}yxQEF_f^*~JkVXU0>DxuJwBUn+(2 z66iLE+DiP; zpUg~+dhxXzx)dHn%76b^sja?E6dljrEM?A}6Q#*Lg*$A!5Jk|VPrLVt<`R>E5)=IM zlgA!Yuk%kxaWlxR$*-o=OdyY?3;Lf zy-%D=pU4{PIS^9$@O&~Gr^;yShVdO9>1$A-=IhahSBaY0`K@15mFyJfpMEKqPC)M- zQ9Ap0eS@MqB}VnuO1I)TEbiJg1yI_Y)_|`?aH9JX?&Us?wxnO_XBYJEBn;d>c1b_0 z%e_Rc;p0ilEZr#8_V_W)=u!O-f%&DX<4W?j4wFDH9&r$}LhHN~gC!do4Y6ujyTYf` zw^z<{5^yDFn4~it6A0Do4Y(*nMQFWpiB_QL#nO}09Z?*-% zpzujkgE}W(b$KJYdZ63YT4K-Zy87GpS39{ZCr6m~td*|UKchN$)i9w8N_#dwP2mVV zf50pqlFBZ-o*T!dlKtutbU@vMf;N}_aMvl7i*b}bDbjk)Mvm%%8Ks?*Zi*-sH6*pf zMd>8nOk7c7;G=FbUH9qZ2Bc@8qgihubD0|6HYwLu=_3Jc)E7?lWq|TcwJ7+7mZ~v# z_BCUQ%RNOB*fKvQ${%mjXVW=Y3li74)MjKto}pm;2{z(6uBAKO$@;)$)K>UDz0>f% z^%E;)zV_TP<%TNj59a2MbV<^7+NF}cK+FzJ{Fqgjo!;jr*_XFzB-?K+@#`g5oQ=w- zoYLBnu8`^%tZL$QBM>VpSiu}}Jl&h%RncNtREWL@Ww`sG(QA&oQ6<&&?vzmPqw4Y| z>o3ISOj9T8ch07=qGMnTxR(GQUZ7L0_HSe}?V}C>GX_bjkVn3K^YtSqTjWLI3+63s z)|J|&HE*k1oL39=D9vvz(6#8EYM)PiF|HECmP{ca`_Pm-TbY*kLJ|l2^F2PqWVnX) zVR2vgm`q)gdl^VP^o<+Uxr|ZCijd5<*YBU~e16@frTJyzDo<5ovAhwLfVFyOi?W-t zedTkQtnuwTlrOz9BaC0{p6DEI4Zhw%xf+1vRqFS3r^}^|$w`=ZBRtWPwW##$J;aSX!ut_{{>eUde^(?dOZ!b!Ob@c|219qKc5fH>KkyU-x=cgKn{(WLz(iYSMO! z&NzMHKA{X`ZpleNw7zOTF;TZBvgr0|IA&oSvlD0@xmDqBFqa<1JRAtpr zfOrat`a2T-jKaBRO~P!-^uhz=MJ$j-1=y_k_Vs;!;5230)v)55pbgQ}Z&lh?Y!k)Y zoWFzz&$h0LW$x$Dd{NmQwnR5N{yB#xDUV-Dr_B0mx54ta8 zOY|i9O-1Fq-{50-`}NiBcNJ5NUFTpaQLNpmSV>9KCOMaEi0KrM{T*yZ_LFTLF1ATu zfVP$2l{H5Qgb9#d!3ytk6io=DeE+e_&a>Q+>FhxD`KBe8B8EnVN5iW#`?4Lb}Ilc$zTTI4~qzbWKq z;kIMtRL?5^pu~vueCut{)^ze&cU+mlRXY(Wb$vnPg=e#cv*qd8J4M~u?otVABXceh z#6We0oO35yCWS>m$5kmgSx@=g(@giTdOpU~+I(8sJlISdF5K!^jS+Lzb{~)5(F$_# z7mbSJc#~y$>C{9p5~@DUP95XZ?QfJy;`>f{s}iu@lQbqF5aN5hY*l6|h?7lIhO1+2 z1rvPs<|=6xk&X)!L)!SA~AF5)qo0{J5yEkG-B_S&$;6=XH;$}aH*ER zaT){Ac0;gp5>lk}kndK2fKX%M-;v~7{34>l*d2DrNtTj0~C zGqV|foniAzc2;QGU`;N=^Ig6oZN(wG<60#bb>hv$I?9(V78m`y18&=+KNxbhwuv;1 z7^&{(OD*k)RSHDXl`L!r8GlNqgE?=>+eevRqSrzY8{Ak$KeRG_I!=^oJ0Z&N_~C_5 z7}ph~kWSLhR&yiti_(&LsVp^}U$1$_!|$=&qe`6v-+IYw zu2uP*Bg+*TVQnEr{a?&hFJI-_79r@^_N+X+l>B^jam?U$MB{NTY+82N^uf5o><6XS zT$}@zo?+}`RCD^yucK}SP7Gxg=aD@h6}*==2rix|tMU}xwBV{$m819+<{-t9yeq$uZ*rZXvnhC~IV#Nf!h@`s zPe3ivCM%@lnQ9e^iQwTiuc`vxLoX4{w#Q*g$G?Vh-fr!rnN@ib*Ldpg9e$#HaumIK zeCSd$#?5JDF6Y&yk1R2_t6$Rq1>XvkojXww#{T)vqV3(o#lGX2&R6BbTjbHs=U(uP zZ4%1R)bBE%+Eog16nWHsc~-L8nmR6tkN0HRJ3AYQeB(u-mqw1=sX5~2(>oA3ChlxM zXE87HGezF+j`M|*>$IM?>-q%KSp=0?H2I%ggsy&eFi(;X6oh%zg!v4US=(tA(cCj~ zH_ub6Fin)?e|P=^JMw0cfdOgf3#p(a8`KIl*5ZR^v@wLJ}^GZ8BerG;MZ!2 z+10R$v--4!MBXss2fwnAVdhwjPp`#V#Y~9 z;h;A0J3H4TXyliT&}oB;!sr5d}P8fLwb$%XyN*Xacj)AvRaqzj@V5U1{lxt+SMV!Lw6}HUkg(4+I=v zFIc(4!!9}OyJ`5!XAb66u7O1%kZU<~i;58*Q{XuVP{s($6$!)s{Ljv>iX+WF;|6C_1JA0^h^3@&p+j=49`yHT|Sdzr`@ za*ZDp$CdNwtPStvE?ir_m8fLD%mX{jY8WMm^fEK&t;ks4IGAoP5rQ?{CJ7jPo149> zHN8*QY$d%N5F*1L`Kl3eP1q{)!A)GOee=mNzgfy)vzf<)rRTXkd_Pa$UxiKPRCc-v-;%xWAPRH zP!I1;y|dyu*WAN$Y$VUd&gJC!iLh$U*o@y!timEG1Hay@vaqDYt%-v|1KmUzL|a}y zwq#`X5_W!#R-tt2zoT}{A$a6o%^D-Z62-MYWEF3d?se8lqIjdqLVHBmBWz?WH>vsJ z2(L(czl(DYUx(#8F@X!7k0h?RRY1*}2vazd zuZZwp=6b%yq*7d*@ycjLa58$j*{Dkuf*8%{6KZJo?HMLzmr<>sjM|5Us9Q$GM^M}~ zDZke&Sa^Zb3^6I(SDG2&|ExK53^+{dPWiNqg5!0!UEJ9-6uHD4G+q5%Q8$G7Szr0K zr)F98Wb)<j#5cZ z<&q6XKWAbjjeSw`+d(hD$=4}vJ|jKNWYF#WLM$155HA)BxHbP` zCGV5IP>okySia~OiFjsqw%wfQh=fMChoV}c^aEcRlJ5R70Fb@2d`*Ban#E+W#)D*) z*ZgHo&24>f*h@PJQu~5S?Be9YY}EB)Z~AT%%dcM5VGT%nF;5r^=zXKtli4*Mn{uLy zOfqqbun}Y7`uOAXvPo!r>dl2|`Hv2r!J%h7XnTW-ZzrAU8Pk(oHo7p#E2i7BbJ{vY zMfn-*Rp0!pNAlYm0~SCyI>Y9m{j~o0wdd=K;1falUia1Nb%ZHjzV-}!-Y8Nu*IR39 zYtDR{Y^ivzO|@Qt-{@T*dOYWmu3aC4SPGZgaRvw0^6s|U&X*Lqks5_Q&?XOYf)0TH zQ-pw3A$R zy)4TwJ&Z5M5$S~DtbjZ~F37YjVq z=i$}wcdNvZS3!_2cv{d(ayLoJ5?SP5JoN7U$Ozg(t|S+9!gRz(?UB~d+jm$C)zaF* z@%a8b0wlC(@7-PBh1K&XPrdsnM1Eh2qTYkq*?Q#)OM)t7)n0&~mzlYGwX86-@|wHk z(~0@CyrPiK?W*crXTOKx0d#fmk)`C-VMTuPo9pRfM9H?4f*-<=GW(^I`pn&_!*x^oh*jvf1i#+e_nyu2|*@wzIww{qU#*>B~jg=R|WImGg=>z+&v`HkjGzbF$6} z-YN;x(>MFp9yf+&e&Ae?y2{W^%G_K|ZC8{u0GJaf$Jkl2i8U=7N$dN*bO};qeXh8o zbh_7D`y{i@)j3JykEeKSxypxJA}3Ebaam`$gjMdHeV$hOf<3`v$f!u zT3uYtH{wd-Xf?XV1Uj8@W@R{3oukL3GkRkq5;B;#u<-bcEwtJJq@Rc6w3!KCHNC3M zsmi{#clks-b-;(khTAtZGg%0RlqYwu9p@^wekx@*?o?S1K;xMBAo%gs z62_kT@Y9{4(sEGsuIyvnF5~&1K4de17il7p2BToEs`7K6`|9=l1$<*kx8V(VDj6-H%2vWKW#c+Tw0W~t}=>A^z3ogny%_PdHh!r+&0O|A?Q<_cFjh{pvZb`4eKlZL^l zh#X2+?dR@mkyG?shxy>rB}nu=33^%)ab}g4beD7yW-C|gd%(MQN`0SBQ7NZoEbo~5 zFiJR4QI}mQQIv%nZ*VJ~)-MOh#IbWxjVa{#PY8<0JZbPLN0*3HhDEGxCd0zRCo}{C zFIx}|n=hSlnY&cIpy?yeP`Ltk=4450kP3f3w#dDUo>AJ_pP)T+qGd->G^j0+?;sDn zQGJ<{D^Rxm0&8ba_&Jqxg&%DNg&vxpoHK*j4S@vDH5F$;kZ;RY4=h+(s>%i_eTLJf zrq4d`A{zFN0(E^p!AIx+{LA^SvGNlaP9oQao|PeqxU?IJ*L;F4?Dkm9`y+t&K0SS~ z8{(F*q#SpM^u0sXYp4?SYIGV*-+Dh6&@x&Nx&82%YL6z*-50mozkD1l*5}U1)VKhF zXcPIR`rO@dp3@L>t^?3cl%OHa`&zMTNN_Ey4i}&~Ve3IcO6oxs*``lYkwi4qQ*v`c zOHxZ=3b22DgxH#mEy2+so4b0ke&%7eI_;~IxdIJd?E+B52d+H#W8ELLV;q*BFE^K! z>Df#JMl#L#g@#I5tJfZ<*gv~TYEJj@qvU{hU*+Z78PtpyK4vnAB}wwT zMh27^kHZcfSsC+RL+)u@W^B~N?gP=@LHzSpJ6RpY&c{`sD0L-==2v-A70|lBbw$@| zn|p@v+fRo zKYn!$Zldew9GLTBv0=xQMmN1i-QG!ggH)04md2IKtcQb~dNU)MmGtwp$$>3oyitbf8n1NUE>)vh~(31%SaQ z-1}Jh04PIWR~|tm`@RJ7<@-YauZ1z=T$8jYv%%3rEV`*x1%%ziVi7ts)iq7`%HDN;JH~)JzI`>mdlacIzSbIT|DPBp@3Fn!Ji#!*&Y7C*un=7!vD1gg_u_p8Dwl z{4V4qG6lhV&^`~h9YXLn103B%AT~5Qi!En#dWg0#@6KIH3mlUHr zS!GNFMvG;yp2=EW=urX6*lf>P%Sk7;sIi|$XY#${c=yid*`XeBSmvtwc^I-4qT2#} zg3)NKb+3y96gF@Ss~LLfxycz}cWQb1Cif2p!b->^EHP#t$m{5yuRuO_*EvL{c54ub zF6l9{&udR0A|_zXWlvy?F|w(H#tR{k^1ONbj;@_(Kuq6b1i`S+K$E#Hb|sL_U)r_# zxB`j{bV$zc;B`p`IJUZPPrWTB>-84n-PANtdIND3)ET)$bZ&X=wcc^yG$TUNB!k2?#0l3bW81V-HM$7^O!wx2a)K z@4-BqxK9RJU*f}(VfXmAz0+wN-xHBQOLYKkU|^; zG*dt~d}6A9sN^gAQ<)7%&|YKf7)&}Sry=41QWsP%)Rvjo&|>C7v?}^^n2*T^V|Con zXD8?4&fV#XvJyxkpMr1?X1Qan;dqQ(Vr;WL03(UvT1O$lAaOS@`W#izW3RZKi$pDz znN3?P9dSGp_mqi+A{HMz2M5_i{$4*6VE{qAzeLxQt zuYJIMqqJB!XFBA4~v%K40JL$vLF#IQ+!A30IqG_ml`95%eA3&s?huG6bG+ zIA-*AyCZ1#{rmMy(lgXIzpj9d!Tl+xKOJ-bYMl{aFRs2XLvZ#y{W2dP;4uifrCtQD zJH@xHaUi4Wy-@SKADgxFZdI+Ghi_F)6SXz1WEA7^vu8affF9tcDYZ(y>$^kRT z@&$ceN$1<%;8VYc0_kB-?(LNW_V-elTu~L98wH8voQFPS-hGtYjfke=9CEs6hW?&P zCgvFo)IcLBQyB*47e1VuDV0vl-q=kC{8n^Yt_RdmJ-q!oPv4Zs%7uq=(f!hQctb9`N4R7@`c6&uHF*3b3 zAd&)vTcs&Tu#AXTMHzN-2iv7xPGyK4)|Ms51n_Nzh3tCMiq{BWK|W?aW%38BG%{k( zJ))P_49iym8`LS*&=OAsHr@eaP{l+dc|0feLL&FQhNYaUN_!-vC4mot$A22!?z*GU22R!!V5j5rec8}T(IQllGRc)UVUQW)L0SN8ULqk*g z!l-@nyk%$x%G+C+z@xi)oq=NeG84NgDLd7e5>8Yp7!dY#>jGvDkd@L0GPWk5--CP^ z;T;Q;BqBmV9lel&C)GQcm|73YRYTCszB@gV`-V+ez7u}lU`P&dv2uU5_|T?+d;wjq z+JerSHM2LQ^%h|QNx1PVTq0k^=wxEYEDw!UeQXM`am$-gveYAWDLHlQq?>SsFr5_g zu9*i1;}-^Lx0>H==k?3%M1DRbMwT)WFu^cUQ@x}5t^;AkkXpLEu2N%c)cB_TG#Qp} z%^w7}J1E##G{m+c4yf!%@{b*VHjY^aV3k&}@VSxo5!Au{$c7YZTkc|tzuueKfgX@8 z+^|5aA-JRy<7T~4PTO(VIzqm%Z-_ht*quO^RztwvMc#FEO}thUp>_9{Y5AK$z)Ykr zA~&|xdAq%~kI$~v+fnd|g$-t8BOH^nf7g&f2aN4rcAervBWBL{tbmpbF}2`%WC|8G zO}TpcU?`&d)*JV^OQ6H5nF}ajC!sl^$PC0iZf8@1%2~WDe9h0u zp_~aQ-wjyX2Z|j?5ol`~3RFC5;9#;w?5qn^EZIti z#I+@n!j@Ro2OyC6;?t+YB^Vk6#1Nxva={Me$JPY503RYdXmYs(dR9{4WA_)#9I1Sr z2O!R_^gRYqZ^yJF5&AE^J72#{srur5=OZ985j&3Ug|vkV7LvR9f+K9I5IoL@>RL=+kiRhlFhB2Sbu2V%_KUU^;p?i#Ed=ngT@}T)0FP0HE17C%7JY<1G8VTf{r}2m~YR&$2t* zB1YXK2qnunGt9u0DIS3Qf+9e>}BnL|2HDI;0i?lyTH+}9lSGm+7N}fc;hwSaRlGY z?p40eT)%HHQQ+1JhF+gn>9{HH$nHA z3wvY}hc{~o?$tgl=!s)C{O87`e)+c&q)~el_RcVc#SG2m^U}ENo#FfcwJ|-wB%lz@ z8!41K($CA11#f|6f8TJd{dQc}R?zgmY;*tfxJUOHG|BH9B9`E{V+05ItinVigwhTs zF1W+>`-U5v-;QAkC<%H8Qjpr`JZNWc!9TaKtN%E*--bDK@}Q(|mLH{vaXC(rsUsDc z#wp@)q#|B8MRtx<1jX}JdJDfI%>|zwesM`#fguVbeiK3oI#Lx-keeFn|BEWP!PzQ* zwf`A%6eq0^mg&EXB1bWzk5W2|)AguJemB@7hUYscUw+NW(Kr(Hpdhbb4$k&bcJfy$ zvT~Fn%>{!vgFW(+f7xI?e#yx%aBv2j^~=FY{k6gVMT#6{B{+j6Z-t~Db;-YMuuH$@ zfts)X&#QjNl=}}b|Be;$^s%+T z3`ft!|Ck=Lqv*kd`JJA>SH)6{4XEB{K)&!aO#_P#VUW=p;g7vHp|uRvX}aQuT@|`U zS)J3Q#?0I=N-2NrcUf0sNV#A5cRF|v_!pj|`EkJ-;g3D~w7&G`^wc-(`S#=gkTK^b z-T2-}E|YOid4&d*K+De?w%MhR36$NdsI$EF8|cfzrv%@?Ul#hj$OFzlxc`N1MAY_& zA}{gPL=1@@)M?@e436MGaP;yCFZHMHg|eUNN&+*uW|7ok`!j;5hbuLDC1NIiHY!}X z&L7`B;Vu6eJ6cEdBDb(R-FLWh5+5~0Kkr~rp`H%3|39YY>=D#396^o55!ARGLCx|J z)Zp#KH){OwQ^SR)6ST&r#!2{uOvYdXBW9r^y7Bc0RRjM6b(oA_KFJsRs{z@54hVdq z^5RK89-+?*wA1sKy+`z4=?F{ZBF1l1_z`ipRQO*F*#C1tkMklpZSV*Y=l42J(#Re) z{$xk=|0sjQi-?;xl>TbKc0402H>#`y#V1r4<)hU(eN_Lse{POGFK~<(|9m(Q=W+eV zi-`NL!9{um;eQGs`Og7qjh%7Y;1T|M!r4Bm|3{5KUPRo4BXtDfe+nSW&jEoicyJRA z9^tPioNIW@@!P5)=k(8qyktT0f=n0s^A+d!>G;-gCBCIBwEzAW z34BXgn1KWdzJ`y|3AdCja9+OhDCy@*+0m6JybLdMo$;*OYrE6a-aNs#ZXc<1p(^hF z`MUjxMWUPz`w=V8U;Wdve^8QW@o~7rvsfYLP zw0@?>(}|$MqQY)`!mvU^-okU2j`$hic*Ih_8Q+f*oDt(kY?RQZ|EmEheh$dj$br)a zj}Xx4Q&}%-aYX;wkLZ6C;m?1~eX}Y2h`67mihnhr<<9|YPbc8C!6QVUMlKMLzCWt} zNBmsZKFZ+mBI15_r~lP}W_SQuY&deA@dhFc!a;6a5Rs);V%AsIKJO-coA_o zT+~s7{~rg`ukAKFcIn3G&mQ-Z8WgA%t9ZEtggXdWSOK=cuvf%Y;-Q zgiK}~HN7u6?pl5x$~>=~B5&}I!ds0L`3K&Ya38y8YaCF<{T<5u<0;FEx9D6gAN$!A z^R^X1<(wY=Jeej9w|tu9+#koE5N8Hf1Us|USaW1TauGtsD+6EtS79z;$*^ML(U5tp zc8W33+uwcn;^Oax%FPG=IF)%Hr^|isOLDuGR6~C(_}XNvrDe&46eEP%R|c-+sHNrp zqj121pk+L3wtF#p$Xtg+D$?u9vFj8sE)d@iCS;W-pkTfE`fdH^I}RpnH?C83%e^RZ z;H5}nm%Qs1y~DkwyIn$PakE$H@+XQjd4#vQZuTh|eWIX8KiuQ1;ZgSmM;&uuKW#uv zsO{Sj^~{0&iUIK%ZQrJ-V~*^142Y?;eVe16IkHwniOuVt-{o%%<(z7WK@D zJV^&b1%2W#TE4|m^)~Ec`oz6jzMrCQ*s|;B6OU^7mPOUuvb*UM&uIBp zMBT7skJTsk((+A-I%dIsT95dJmhbzhXBO;N^oWDCeAA+iS+d{JBaYPaO^8o_3f`Tfa(4VVeR_R#<1zSF0AXmdzrTPjAp?tznu7)|Kccm0Id4vXBH%Ih(OKDv4 zPCm5$R}a7OAT?R~%!voBr;n&DI|#C!6-zg?y$ha*_NnASKh;O^lR}D>C0x)MzLajM*qq zLjl0NNb5nyJq!XF1_nZ>rfv3m3bsewuBUJ>Vj*6_he_QYYqOwq_%w3T0L6qwMelAs z0&au_)FO2^@?_kq3e>dV7&j`cMm7c^w1M7GtcLgdtkw=QX*VH^PzR$4=)_zMBPR&! zxI3Qku-qI5UfbSh!s^_9TZIi1V+7h82w^aDE|xyond=Dq#vS>N%rQ5i)sRI1ulZU) zKxX%>$)K)b-#myf4bo+C0M$huZ0*6gz!(GUEa*v#-`47K+CCVwH3E(!S_0Q%<_egw zdre-SFm|}!+5`Lx!WMn^R=0K`*u6O>P@ach!@(L82#w2iIn=~L9!_;_A;XYcCWGkf zak~;~1u(ik)UjIK0_2>}ZrXqk8kGQrZ-oUkR7i*IVv*M9f+^QdbThmT4mp@O2m}XV zAJ#*fp&MUa533JSh(WX|;2_At##ayr>Y*2g*Cz&dMI6c%!;6U_$D;V+U!X?r)pNO> zESRCL@zvfW0XA^Cp;KCO@;_VCZ$oY#|KoYwMY|xz0>d|lL^mRi)kr-2aXIevkm`>% zhKLFj*?|N2WR~o}5_~dUb|4&|%$*$=gij{N4(z}uGh+vmWubOgqG!3hUKVUp*EMQO zKL4?2s*5M*La%U}pLt*`qv&=cks-8bCG{{ zc|Z02u?PC(0Vf_XJL}tB{NXtN)*$7Nt9G}ocW0tWbSIrUeuv9Rn)%*FPK;rWXt~Ch zB|Oe|>FRXZr{$9ZzpHA+`J;`Mg^`s79%_NwpZngm!t6i3+2tt@{>|A>4IU@{(T2xi zsR!sTo=||+f-3yNjsim&OvF28B)(2!qR);&@N{y&+~o*&cKWS7e4Rw|FyQl`Yi-ys zzd`@oo&4@2mv{uiGmg`cKs?&ajZ}DmxbXH7?1kdIiKi-Vw9t8C{{iOTu%eRe&j)L6 zOM2Ljpa&o3cY6M274ewB+{h;!J!JneJwiv(g9r0FJ%6u??Rc#x7e^22 ze@qYe5%l20{7%o`tRkL~oExdZ(L??p(<5~hJ$Nv`)ARSLcnz=hqvVTH&+!2dJmRyP>5+r^V8**e8#1UZ(f2%YPe7OuPxirI- zcoQUkNe!Nr_xIeHBO)AP{!U?P_;M%YMs;+c_=LZXEdR?hEB+KfJh>CJ#+10oay-Ic zN0x`98;8Iy|+TaoXI&guk9}{_@Oqz&q`2=cnx%3@R0t?ui})nab}iz!z_WOgfP3B~*@z1oyW&|Z ze)onr-)D!(YNou*mN|oqCc{OVeR4?2`kzr}2OL#@h-6#z0Bww{@acH_S0T>Ktq2C= zg2reVb%x9@;6lL+y!{<_FSh(1koGMG?ML((i&W1)!_s~aNVD9v{D3QHrO4}fU;4QD zN#Fa@|2U<)YxxK#*G!Su^1k$~(3o@YkEh&sE$4A^T=W~z`_gX(Qoi=zPyN5*&VB@; zF<(eW2nGJHxU>Hegm&d*vEpOOUK(PnAi^>^-wBrEc5g42-+8CcHp)Y>Nka@}G0_sE zeO(f6&fcq(8g+*&PD?iLogRCk7ID5Nflbt9+;>T|h|4qy?4o8g*l%bOKhPx5j55+> zH_{~5(BhbDnx z)QmR!F)du#Ga(XP?m`uF@p1kDAeEKc-FGs7c@$Wn{p9 zS(~^W&07NUi}2Q`bjc-TqoR!|3+}S;iVw3Fip1{qPA!!$aGn>wZ9-V9YSyyeCpqeW&8Kz$^PwdIeZY` zIM6dHrislx`Nu<;eXQ4{PZBnE7f9<@xwQ_OZ=X@aa=RePnLfHR28E}5JTwJHy8|)7 zT}+zs9QQMP7GVI|`x>_wdiqggQpah{-g^0{CeAy?$)|u+&lFSU`DDOk!=oG%GXMiU z1@_azobKC#JpjO$4z>;-niM9sB7xeo+rr13`j(jc!0rz7GgR^1da=BhAea+>E^cvVD z7Z-o_0{8Uvm?ohxkyp5!;m)_gGLUC%svz$%7S*m>dnNZ{Q}_di)XaE!qulkiqrv&s zy0M;m^ypwr8D^m(Fi;A6u>c$%*nUMY8!dG&JvdI#*>q_)MMY!2xn~|enwHC7u6zg+ z;eYOad<4v@el@%k#lNrt8jgu763`y-g3p6_1;eUYLLv+e8*9E~z7-Ezg8TJLb$4xf zPg7hI+2HD$s5IO^9T5l9o=<7bHGw~5h3B;!GR5&RCWH-ZU+eCyu1Q6q+PXf@!zQ-( z&Exb1+eciV0em6{YtSP5SI{u?*K-1D#VKuwk~FFAVZcZMQ#0G!cI;u)SE=^29djBq zLIG`yrt8QR+7IiY({En%pc>nl&|tr&UK#3Rn8ViAMUI5Yr1!HuG^B)m=}Zz=>QXz; z4gzY)o9!VR-glM?Z@lcxL&B_M_1T8TtALh~9~ODxf`=$V_&@?|8>VU^Bi-J0Yt4KQ zmL?@&KNJY^xe6bkg0Ui^(iipP?cwfWcY?S(zM}RO&2%44=iLSjK^5F{J8xy7(D&a) znLV<5-V<>}-5H>o3Nb9=-pp?;-6pNLeQZZC1OtGG! z;(ieAHZNqVkn0!2^iCrqbY}(#fR2;AUFS1HCmJsp1Oh-WUR_+AEv;ihnGXqWgpEwE zel$;hme#|KJvl?mF`+)@5u#!R;yc6I;?FI8v|= z0Ibej)Z-t9H)4HIBLD#IotCKml#lMz$klkHY4k!yg1+URWLJr-f@(Zg^{%tu2UR_k zk~16?Q*Bs@O1e~XWmi~wJ%ka81{Yup3RIPN`K3{4#M)7IQH7LkueXG!5Dlf2APa2{NTFXo5@`EK%!ZVBQF+F2z zhtmh#FjWzevM^w3L$^DOxTIkDZVjuet~C8ftld*g_QMmzNWS_Jij6 z`L*UHrL{H8&tOtg%00^=hc&2~O)m)_pUR~IKFD~bN2P})8?0kwcWdCQ2^4M}rVpjf znOK5v=cU2Y7J=I^PZ4BZ&xUaw28K~C4S?p<4zD*4t!d^JdC=Q4oK@vLl>pTaKqB*k zkWO!gpOTC8xEfa36G72vsr(UZ^pdSusDDfVd~5jwGI6nF;Q)N!*nX0Y<6eqtog>m-O2CAGvDKv~Wu@6e9_9P$ z&S%ieK$zz2Z-0xIabPh}CLv8M9LPS(K_&A$# z!fKZ+Ts`O3i}oDEfn@gD_5~9YwE+Qku1`Ne*VY}LxYfA}($(yjv}l2!2wB*PbiE#y zSv>5@0)mz@O3j?N>uL_a(2%1aIaUbkq)+9A#58SnVc?$FExUqJQHwo$tI~mp5@BFzp0XaWTOy=lT{1MmodcHP>*4KW*fsNjT#&a9f$nWu zjZ1il>32hOGKZ~pifZac*M-YK(si6k?<*OiO^1AH$GmEU0h4nRrfIoD6lh27!GwXS zBX z9?j|+P}|~231SQmcdk^D=xXT_sP(D4W0f=QS1B4AVOJs}z&UMYV=&y|1n*4Lg^7!g zjn_=TLVI@Me9XJG^DXlLNFU5U zt4`E7Y}V`bJRe+MskzS0J2vx8&7J#f@@9V4_FKo?h5L`@x{l2qpxy5-;&v@v-M$>2 zZZGqww!E3eW1V-F*XAyhKAfGM6%UtDdV0El`MNb7Os^ZmxV^J^a`~3`-;;bRp6iFQ zw!GTee7ql;SJ9jF=4Sb-AI{S0V!50iF5^5i-)L{^Fsk(m^X1NJXY;%h?>@cv-d|@& zPc3upj}BW$=P!p_S>tH1^|o#9?4@}X>+L5+?`)ct^`dvUdo^fW77zEjHR-gnTO?h}boXpGQycoy#laFk8aI(K0O*hI$_u)YgJ31R&9%cjcNS#gd`_@cF z*e-(r+;~8o>I(hpdJ>Bo8!Ui zYGUr!%?tAs6HN}AE%Pld4$UjT+qi9xb#*d0+-iiWD)IxowIx6%ls=xq$oFHLXO z%#O`7R_k~xns@5a>*ne1eCJg)9|xN|YO3xplB=WL>*wy``TBACI$Nr&c-Z^sE-sFC z&o>%d!^?~1(O^|So4n|Ib=ooC>*B3;w%K1l8YZjU+|F0WjqBz0dn4|S?w^v=v)0Y< z<8HpOf85$yJ}0%^tyycTIwzNLeLud~I)6MoY+enc^<8`Gy|b8}EFK!ovM3w7`+JLS zt$wz@zc*hr507`}`lFq77y9UM=cV0w7^Stly{o~|#e3ZCy!V!u9tfAVp9xj24I zhR=8Nt-;o#%I2-Fi&`6E)TMi`S2X)xp!nfl+m_oSpUR`@OlYKRgePCX+q$)YPAy>3QB8 zEzDKRmJb^r>-+x2GM>Df=e1;WXEW)2%*{h8eK9VbHVucSYHP63x-$pT+uIzNXXD!Q z)lsoblf!8}*2igC)6ebs*3jHXyVrwq)Yi|3Z^_B#R(${ZxO;rH8g<_~r*ZpaHqiOi zZ1?Hn`h0cW-rJvUr?aED{=O?jNTYv%`Cchlu@Hk;~2uhWa~UX$MN@^1cgx7XZxzKt65SzFi7?(gqc=Yzxc@N99j zjE~n(0JGtu=pBq_tM1uovUNK=d0sB-wcb&dj_SAl-t*Dot(Z;C(_lY&A4k9Lo0l89 zu}0Y7u8hyl;^OkiIOE~8b$vI#zfxQ0uLrY_Os!u&7kAU*<|8+6)>o;y%iTA+t@r*# zF)Vwh#?@uI^YYZ3wz66;i--IDq&qu1UEfV6@Ac(-YP7LBHRp1=k(_Ri`f0DV7vB$* ziqyqH^Ps=?^4Qt!#kJ!2I31NLD|^dUC%sM=_v7oV8||1Uw(IO8Qtj*h#H9BFBh812 z>Md`^SG9-Q@y^rzsHXMoe*UBzd(T(XX@6^XXK_4^+GjiOC(XA`ud&w{=AGf*vTi&w zTRUjuZM&@1yQ8doG{_sZ`pCR!_GX)Dq3=6!S5L0<-l8^o8eENsqsgjsyt~Rziu$#A z#(#_E%d=?`Zx^-wJ!6N(-ui{5USB@cwVW>|5&x;q(}Cxpq)^2R)(&rUuzR_pIwU&gh~(`i06 zPi({MX41OdY@gKQ!^Q2rO7dtr*qIDY-j}(#E~h7N#mC9znKC!g?fj*)9E{BOd0H5= zW{b<=u6f*!_SZl2;wG=3zL@(-EYV0;|X6DJZxj0Ggbayn^ z84q6W_Ad68wcR+MsAlaXxmNT0kCV7@)qK2~?=5fgS?lR&u_}$yUUoJ&&fgB+?oR5@ zS(epX>1B5rk5+SK9u1>I-8Bz_?R{?sk`>x3qCRyISqu zsH5~^v3K!uci*VhM$NRlm~R=Mm7N=HZu6(B!QriWuM&5*r_V`qX`Uz2X{}x#9?$jc z<|RG77&mr?o736-)7$OI{`%)UKB?av6_3TyvAMgM_jxCyR((I&8c&bIj6dV9H}8jaiA*JUyN*c-pzw4?F$ zacgpw>@AA+O}&*sI>c3Hf&T5+~jJdE1A=7sC#IKEKbyO;OdsJ*DAotNhI$#{9V zcd=tGXj~s}8&>aJMBSNcyqlZ3`PMtzeVPu||Lu;`?Z^7G^JwnR#r$+~wR))4m-FSv zNkoPVb{VznYGxr_a;Y)BC}6bJ~BNm^XCm*VXuG zVL%ssJl58)gXTTL^~v?k(e9q!InvJ;xw%I+%@Iev{<(SQVVyMG!N3}_VrC%FU?!k)^xN#SoLpI=iHpn z>6XdUe$+nPJ=>47)^>DNKQXT>$>#9Bv+OKxP7b#>Rxj(Ca zcp>R5&JT>j-fNxvtMSf`xj(<9Z?EQiXM@|<=F!4vG(8?P`nT76o9pL1Bd4>ARd$oq z43*4#(0=lC@_h0Uce~ke`y)>tcg_2#+pEoux7#=K7)kcc)8qAVcd>tE9-uGV{kD0V zI5p3dM|1OeD`oEJ$>8W|dbIZS?Otl0v@Tu_x6MKCZC{!9VeMjdq}H}s+?V`lG?wczC;hoE?1h9u8mBH0?g+hCa2kz30}++2&1Gn+x6= z=a>EG?soI&NIh+)n>uei_2YT=VR*RHJZcXo{mFx&V|F#K)gEq6&67f2_Fe|-+cGN+ zNUoFj_R&UW3>8h+kHO~kVA5_D<=x`>e0+CQTh7vK>$ugN-Zj^+887E&Gp%*~<7}&w zpQ=T>p=-UCc_`j@M(5G)*2=uDUmGxu4^H&$$yMX1`7&wlzjY68i}uOt;iPS@`RmKk zo1uCAoW{}mJb!1W=4~1WA)~weLYY|-Lw7L$h^<#rTyu!e!taE(jx12 zRDAhfboYzp{`=K+?ep@yYhLEkD8D;-JYRK3ch~K+*N;h*Uo9R^N6nUD_1U4C-(A-h ztB;TD!rTe&4$b?OIF9Qh^PpTU;`RHEUzZ=Z7P@s-G&-YhJTT|A*KF*XD?Q#n?#8)F z4qvt>IzQ@PoDGk6wv62kVsBr1%l77ts;6h2o_YBlJS{Gtlhuj2V-612oQkIJcdyyG z@lg9<+}qk}jfZOTkn~@Q(fT=h|IEC(*x1{?oV-Nl&D+!ZS3&1DhjF_8-P!ALFIKDm zOy{@rUGr!$_pSFt>)qpx{!$&ajt5!)Vfm37G7Lw1jak0eH_rjD&7G}_)8wIE)YjjB z{d=>Wlkrv5{7}XO<~?AqZyrhad-dhf*>ls}<95t@-@EDj_~rQgW!PN5=CwaAPsT6T z$5CVbHuh=fXmL4z@0pwSZgdz;TgL`y<>ICPy1BD+yc%7sUealM`n>maaCxwLc`+~I z!(DSzpRNscd%C~z`1sa2yL;W++U%N_kF+r6PR@5WH#f{1Pt`cO9p0Fi>cewmy5;b6 zxp}mc3_r{RZP+=BYrXaRpRsxK^eTS062}wpS0p-EYQyE>d#AM<%^z+C z?{A&`qh{~yW`5Ml(#}<@+#a_!4+hOi>t&WF$>ZxhzP)-hZ!`9{`?`5OGmo*>;Ql&V zX0_#{smy!TNzYuPXuevsKJLeNR|jYHJQ=HI-W|q=s&R3(_Up@RRV-P}T)7AUt z(Q#vE^giF**|;#rIc}YFYK?=9r!Dip!P85BVJ#y10+PX0>;Lqk(Y)`N9 zkKMuX?8SVo&VFm7zW?@mxV0*Ge*dm|<9uZv*{9p)*1fFn7wT;IGMSGqTRYe0zWDq1 z&|5n@)7!_JtI1AZAO6>MSpP2h_wT!>?d|Kd*3)rppgKvi?5W+qd)4=^abDEp_|&|4 z8E!4|*WBC@jpHWStGS7nv(DA5`RjM)hqt@ewetOHI%poQpRccGD(f~c*1x!HH*PM? zlZQ5@E}IMU>f)_i(m9Y;K&N?49gAobDe_qO5bgej_l6+9&(FtF-fG>^YA5&sVF* ztJUev>R_ga2OE#O6LTvXxtX{9zkbKPxi!(NRsB{sp5NB*6V|`?Dt15e^?T|IHEHjS zmO9_vyJ=jsk0yjO1S z95kP9&9hzqUYE{#`+RuUJlec4M|8A1KR%C!&*nYfqVsO9NjI4{ z&d$oKaZx|}xUH?9-)iPv%-#Rb*n4!lsw7>2KZpUs!%z}Vc;N)hgcIKTU*DPM*86(W zgRZsety`~jk03M5-VrQ&O*8lQADK*8d6A0f zl>*xvc>3}PSbO!=<->;{s@lWOtw#h5YEYRUc$3tEZ~kOy7Z_1XRq(2 zt+32)>BSjjfH^9)>{XS@044|gS680Dp@cQr`yD_|r5rpu8!OmTDs~W8j^7@>#6aVa z4#pC~0_3LycY#lnKfj$p`t{F8tv799&%GhLqX%>Irb(eQPC*uZp^uC9cT?QYF>A0M z$i~JM;&_}cCp-%Pu2{zcw)>V|RZq4`8A*9?UDS*LVnuQriXtWf!fFfY5HZ6#@o8qH znlxtTj^Z?uYv(Qm)`fkH9iIPqpKW-6ODFv^{{TF~nXa(w0L6SBF4GMZhlG%edcleG zuxO^7c|`XEn%IXDEFO32Vn76(PK{t+)t2qw!k$(h;1+v$>r{^fAKbg{F~5; z8JG0hDs!x!s~l|Y$G3Puwwb?P1nhJ5j!?)v!?idKIF(tvL5%11tH%IJ1GqS;-D>$L zWw=(!>RP~G00y1LGv6>e5D$QJ?Y_>(@X#R=*~8x#Uo<=Xh?5OqZVb*HK{=U%b;rjj zg0xww5ifFifbjV+YQU#ccm60M0fPKpESa>tf!q@dgs~?XAeKZ&5JB#->Wt%)8d7c2g7T*1_)9I^;_*dl7kq&Y-alC%^X60G7-`PALJ(vSXv+B!N zt2`qBAlfSg6y%;{SGzqp@w@MdsXPHT8#S2wxm=L> zgRCtVSZjqb_pzsbIlvV=^3lZR`30^moRLmy;AfONX^K^qegVNsw10eSMv^A`Giv~70|1-?RAv@E=eGP~2iz|)<;UF<=MNDZ9fP7lW*m3LTLKl{7s$Uz z4tW;|AitficpSD?#iCa}BZ>L8?zLmc@a)1I#ENX-?f|g?3ltKo+Qa;#W1UKMVtYlM zOtGzxGs~g);@f+JFTV0D8pf<+S>3W2z=`}3)@!6xe&2*MH7_emU_CmNhVcZ^WT0w* z>&Po&nVL0a{wMY8V1xT^A7wbhB%V_+2P3(rX{gB&n-v)DeJ6EC5l(=wdsjFF_q#|9 zkLD9*2Kg%A!2y^&q$%KR=HLhGR)gf!vnG~D@_61!a{xhy1%BSofKg^+4gIp-LqnP# z0PoXl5O2~bN)<>XeQovBc7jNV=IdX5fdO0~0jPvZLj9u(_V3kz*fum2UQzm*VrW78 z=T9BHH|6cd*%(87S4DC;6YqXB=Og|g)>!)3PRDeZDepu??I(zJB9Dp)voSZ$@dl0F z+0D9k&h;?ZNUt8fAo0MD(Yzx5dZL7JCX>3)fBiKQ9RBx}<^H#E3jg`8m2mCegH@3T zy=0LYnC;rI-l)X-D6YiR4GaeO3G~GuhcoE~zC$ek)%C#roAQQh_gifI>>oJ$##gy4 z2r$zTey@n7lb0LfeudpoGRgCn@2&xxiFowAuOU(Kki&Y}9oC+xY%-J1!xR^`geOC3 zS=i&FS7h?l?n%xc?VCEi0O;}v;YSuwuIZnqw|+EL9IGUN`3 z`HkcK*#iniZ|HSCcL6XPdm|_YBn%Fah-d-X<%#hipjTK-5-s9dAQ!>jGDFb_w`?2j zBOqR}9ETC~^jEho0gn3XWx~6Ez})5Nz4|JF8O)COaE0Pt11}S+#|WO+c{fS`0ANXA z7Z=+%U#OCNBT15_GVA_puXGzaxjIV3AW3jd;{(1Ac;oxs)3G$o0rJW~>OpfgJ6=Mc z*-Sf3qK(XJ!uO{<)Vj5$1hre6+Kee+;~t^B9mvKzzPXB@z(0NfRnVynYYUi#e>uy& z(FS0n$o84u`cHFkBZwD#^s<(S2dqThTWX9%Od=rAb~H_b>sWbU0P}Z3z4;x~t>J$B z($)dqHd^<4&J%yX^iyXo@kkSxj6x}cCHSCW#)KaHk^F+&_GttK@tk(rp;6a>Zw-t@ zU@ZqjenCNJZ8V>2* zv|S7RXMbf>p>r5#{oPY#@RlFy-=yd}diZv3#{w4|0TH!)t7sr1eESx)U_nO~v!vt8=@X#i0+ANO27IPeTQm0Q0Z*0*8qz0+kpqVY-qpwQEnwD?H!Fk}xC z?J9%43wwus^*rCo2dsC80ROor3UFse1pxm@PjLT=?nYpLWM1fKTCScq-&N2BWWw5Ue@A6BLn4+Vf zaMLjE6M!7(CtJ5iur37iBMbhZI6sh&9pUl1eb)>@QHlo7J1TZKfBp_0_`eFM1;|5K zF4Fq@cnWiwFgDvdq_MHN^=^~?BP3VtHa~84EKjAD-sEjwxtvV__~L22vEhBAOuj=| z_ML!WfBs-jlC}lb+a9n7A@LY(=zw~%xpIb$(O%ypM>z!9+@VLw-_!t8fSQ@wmTi1t z$wweSUy%6WHfq!6f}Y@U*3ZqO?4$#Df&=*CIuKMWId_>Lmdj~u z06s^}I1?J;0scr*VaCN9HG>$8xo9AR+lp!?JxS&HZdh((E9r0W7g?7}Ixv+pF8T5dzh++LScmeJTaGEMYe4|Um*HHQ0 z*>PQRO$qi&A&}j{V?<2DAy){nAXVeqqo8}&!k6)DcELDH3_#F<9DcYh1`%!5p8%^T zH(}|3gDn^A>AM4fd_h5)4~}I8X{9{o)B!*LN=&xtZS6!Rh?2z*;-GLP^S4uq)V*&K z^ef*GD-f^)ZCUYDF5eAS*6#V~f9L|#LfxP_3u@ZpA z2!Ee;7pVq1mm!Y$#KTRsL0%bcK-|c!O63nW{E?6=O6CuYjYkR&7wwATK+o;V5^p~R zE(KsB1;BMZLdD%5rpqSR13qVaP|mcvFa`kL*#nwW-UmAApuXo|C-qqX$0QIe4-VFQ z{ychhA&-+_TLs?bbdW87Jr=xj*iE|t!?bveOM;95aec-Ss2_=7Z<37x$6kjIldKLE z&bO?pwR`cTa~@=i^7B}@hw90}T6mPDOEI%~$&%8jQnKe5^Ht@oBxTw))R6%22vDK+ ze78m(wP6eEx~n$Q;x_VjStcN|ZzST@hXSAuUd79Z<2xsf(0*^4(m3Fn)-kX%zkY9& z`Sbt@%czF{^Ib1KRMdEO+H{M3d7AXhC6F8kAr6o1=MSq2k^=}}R6*eDW{}bO2k=`v z2}FQoxA>E`#YTiAAOd$N5CuVQO4jb9C7C6`p3{?(L&(K6z4o0l+O|-6s;NX(z3~x- z4J3uuV8!@F1uEd;FAe(U4{9Ny76Q)!F8)w<@%0~0J*dHjAXy4~eeTvB6pR-GMOIK> z+s8$eZo3(}vhsO=_h7Io{@}8A-pkdNv$Dq`9ay7Cub^}Fh;+b2%a)eI*pN{t(8c#KrNBe1a zuT?*=C2v-W&W=FwFnxVdLiqlx@hY#2vQLH$or;VFfe4Cct_)+ksY>3GBpwyL8Bb3k zY5&!a6}$MjwVGGMS>u7soR0wIXQt9xTZZ>0DSdL}%OlUe|EEcod@mFBQ@EnkE0M-C zf={D`n9A`IxG(8i;-02reVXe3tN-ftw`i=2!ymVg+*9{yc<#Cz{8oz05P2rQ{CWe> z*=aBBjsf}r{koUxqbcBuppdFSHPw+><#eP5!{^u(N|ApetyeNi$ZrvK4?CJXK78PYGw%bHcBS35fer4=sT-&bY}e%30NhQqF2k(^n#C^`>RVw=^EtU6 zPpChl{SiTan8cII1`(LxI-MQC(50*5USI#7&wpp3;4G@-b|JKaRWo(elEbsjx2>Vt zLSDGr%c) z7KAg*kqHaLJC9D+^98ZA=nNh(_1`;j#pBx%2s+X1TzB2|7aIV|J@l?+X6=I=iT3Nh)YX+FP1r`3$Fl3#xs{Flm`D?VN;zu8Ucuzh_0{%Yi zkOOi-XZZkD1q_SmL!)nzhC{8h@fnwSQcuKmT$g!53;;D{AI<+)E9G+fU%j=cANRKa ziSQeipW7)!}Sk=_x)z4US7t6vP6$?02RXmXCGbNi?9yvviD@gta zsmBY!;4!Qr16u0r@p7Y6Te}#+U|?Si9GUQy&zE(wlo_?Ku2O=`pcL;pXnB7U2M$H2 zPSV3xDdWGoCUD5*VwRnPmEYY2oB-+Kg+`_1130ZN!&-tn+}JBeDK?tvp#EnL2w-vk*+k#ZF#=dQk5qyrujjqEL;WtS z)Ah}Jyt{wc@@|u4@^ZqqfB*?esfp4%7mwqm#R0xGC8(A3RP1Zy7mM z%>~35&wH@v!K7C7mIiiSP z#cjT+Aq$PP}F*9@N*Oc2u@=LlYde|379wzz?LAv zUN!JC6I>I()%)!Z@OD5^X`YwD@AdktjkrRgJzLl4$`@`N>En&P<+V6CBs6og`v5{U z_z*fi#$wc9@ArM&(47b0r|LG#f;(U&3;?rPy91`Qqu87UY7=b3_Wd?PF2^y09NFfN zW*5&-{doesO2Qfd)NF)?T7&dgMJ$q4A!70LLlvumVCn;$0e&DLP>nm#lh%|+w9gX%z&KhaUq8`cDoWTT&01sghB~vB?BtR?g5^}* zOnlRlX%ggA1>G(=q_xewOY)VMYG&aRlrZs>Mgla;N4y&1DbBw=JPlCnIYh@Ru16bc z0ocEqT}!wgmUqPui;dzP5?g;$rCZFgnQdWZf=rQ|SUURHLm6^G%UiDrP*Kq66%}N4<@Fvb31-1lebL1^;n{fDTtNvsCR&uL!EwokeOCVGaxU3#qr<1 zXkuxf8)pgfwZl3TjRy78B>9HNVQ9%TZ7XJ6Q)nbSg1m^zWq88lZlU4YOQ)?6!0~URljvbB+%LC;N?N9&tre0JgeT#tT(F zEWyo9Z9tKGn(?$fb-hLcM5qWz+)Sw~sBOva6v*GqK&ziJ!EmFWu0#ZD$C$@U>>qb% zfXe}{zsW7dRRqJ{(=+vH{{*OrM_}FS4N%H{Yg;vg%{Q+@a-YzG<+B7(Z~2KqEPy<) zb_UxWR3 zlgk)*Ob{PnuiZbHw8vfL#-2&02wYH$`8%eRBzwLdPKv2}a-U!MjPg650hX2I_BoVB zPfhJ-)9#1xv;dDcQ%kwMp2E^HSP*i>!Mem`DMO65ZEsRO*t{vR_-{TgHoEH; z$u>w6F^<5ZAOkTO2J*`zsHH5HZj0_hp9J|Ht(fd)jM8Xzn@(xY+=KjTF$&go8Qj?;_##>k1TFNUn!kv|o{t?nT}Sp}?%icd7}uv;p- zA)J?hxrHA3H=w=2{r7r)4XJ#hz{nItWK0(VX&2y3MTYl~Pj@~!M z`#{ZIn6_v1-a#t9bJ7a1lfTS_R@zkZr??Nz`U!|O)??!yK`mHl0g#$HJycHd2Xdwe zmt_D8wBJo(>}LAno7y;=%rGcg^(KFZqek8t>2**^fZv(qX;f|lwWmHvF3(f{_1N{; zJe?b_dG1nb{M^~M2dNn2na)SzbFKYdFs1=#P3=IdAASY%H%N9Z^-hxSty1XJvP zHAQ4?yPq@)c=%R?3q(-$(VSE~&VB7FoZ5Sifob0H=w?}~9y3u*%|!u{l63*Vz*w}( z_o4(?&kktx*YID!bW}QT(V7Ex?tg+c3W*^$yZg=ju9P#LxpT9_^CSH3yK$u*d}gR? ztp#u^h2V?H2g&p8ee)&GhQ(z4PcLvJomxWeeC9L!KC90S1^n&?$8{vg9RN#Hv~6SP z@BY95#!*FQIm!90i1%KE*0zJZSVPVP`{7?7nf%NK@n;5*!Gh8qSW5;8|7&DbzFiy$ zWOVcHJyp9f3*txELrQ>iX70{f%5M+fTArf_cT^&Pzx^CcXE=Ck@HZxApWZ_`Nt}WGuW^yaZ)&sqCob6~D%p`y8t6Ex9#i9*g#_RR>_reb%bruaPf+ z)HzSeQrSMKe4DSXBRCTrB4>a8=AWbVWRdPuGt*Dr9e_e#lSut; z_W3%T!3JQ7&4VC7t^VfdAeUVIjZQ2$`-gQ=2x~Z>;Zu3Mg4|BunLDMYrOUvLP?WfO z+#Vn<-wW}(Y4z!ynI(2{_%1>Ge^1-)X5@CI4}CYdygY)700`t0U{|Urd-trrSOj72@)h4sD?oYwd9y90`lpNYyfI4ljL~?k z;a>f?=Xj_XRqpLSNqDv#u=j1Ju_jws&@A$y({zCv{OH5vw zQW@;!*C%zKMqkrvXnTvgMDMRr&FV+7nm}Z02JqtlX(XQloIq~%plT%lbtZUKhcodd zz1x24P|T~OC)bz9SNeDpjc)3ZQmG8e`Q;J*j76f8$PXf8r;5IV|xcQaYhL{aYqZ`*^_mPkp3j@rHaJYV)tj-!A2MoYMP* zwW+aW;;p+e(BHCtwwnT<;IzATRrIETTzDxXrj~?E-G5J&ZB)-PNyP`8gR%xjHJ$;& z5=Wth1hy2X=lz6Ybqb+Vz#GTQ@q+V=Yq?)KWkC(By9P>AeT)TQOc*;+{FFXlql@q_ zi>MC-U#4%^YFCOlH}CoL)gf1xt&NVV*5A)^aHp29+s;UG%6?|9!DByas}x-P7|1+4 z{?)sv>?78c?gGTN`Bb_P44s~XH&NK6AgtexX%&970cXPTm|c(sM|y@eZVdG|VmcK# z0|4PQrVf4y1t1K24lRgWi*$sy9BR-(R27~ef*P+9z?Ox({_w+Ky21$($p&I_1Ty%M zx8tvimd_jn4F#Zb9eH106`_bp5KXtI=a_(Kwi=e}r8ox>Ut2%?+iM%H)6 z@CjoK22N;d@0-HIpLqg|1Q3EyLXCH;jNjX6Da zp?2t8xlO^zUA?wI9Ab@dMo-x$?4;^e=8K6qH#o;ZXi#emIF=Qyoj3i>wvnY#`N8I& zqtBU>|J-%{exq{`cpb&tz+qb!%U~_kuEyzj>jQvwhn-1Dl7H;Xj#y%aY=EtC%H<9_ zxx>ERQ4?`-g1nBG00-b}`fbgQVNIxi#N%X@B>@~g@XN-5^^h)GNy#jz)gg{9!eN1ZET}0=Oig%308_lxd)1R>c_f=sy>AWgPguv_Xi>)M!ZKQbOJMH; zilgCt`TKV0{6p{D`M|o{*wib4d550MHx6*OC%|@S3+EI#a|>L=B%$xjgw?sQ?(;XJ zKfm}N)xzEe;xYuB$EHBgIH11b-@K=J_4E<4=}9l|0nR0!3gA0P1&#$p(2Yp|yLVgf zlolo<%%xEr;rD|&hU&HJQ-h#<{eaJ6NR{tk*VSP_(&W7GpOPGGPMLlo( zzy>!{;uQ(_1iKzdEIF{x-YA~0ygktdR1wa=ct~db?=!qzpSFy%6#zJf#3ZKBI?A&y zU-j!b@E~4}6;Y6x?2!D8Bw6sduiAMhmOQw?#+@2!IK~$mZ3Kcogdezs6M)v%d6jyl zcuzXBq!RvbL)t!?+Cbp=O;VakZ|rk;0s6yVfW{+e-e?`TFYNsNF^1^VjN`(scQIeJ1C|BM28l$%B}Ppt~khh*i8`9qPI)lfnV%tpP`(=_CV$ zW^BAK>oRbzZ_Wq$Z_Xf|{Ukc5L(R9k`UJxAaD8+kd-{dShQI#fAM=n#gbBntb{;<+M-+Rcd3(38+ zP6R!W+K;Aq1##FjVSg3mqFlaa+JyRL;Q=6G#!@4A9KXp43!P8B(3W0DLlDTEH88rr zIwsr)fc)uX0p9@MMO`~9|HZQYram9*e5HhsSE}=4t;Y^?;^>cOueIA}5?cggFcxuD zC=n0as2w&AeEu!^m1k%0s=o$+|6UxG9KhBX*HZHtNlTKN!K;Ruf^I|i2WM`_5;ZRG zuP^a#0M#-0uf@LHfw)Lw3`mfe`WwYx(PQ7XN@C)%&(@Irs{_v($+oVtzvTyF zO*FY~)5$&Fo(%WYm|b-$6U}XX#5_tu@F@A#K8^9FSU-F_0Jda76dCB-g!zy25R~b* zPAs}t)+Xn35ERAr@aD^9>!koH%w-6LF4v}0Sj3*;`wP88Te6F> zu`~~4!~4OWQ|I@Zw5%G4>izgEq15;ZYp( z2VwlZ?>_n{QR(%rL_C$J3WVr@^K|6qbt&54z`qhDPdSDqsHyq+xF1st?fL-JyxOep zg8M^q!eR0EDdjO9X;*Q^l!E;!)}{)BES|`sU&9&@J%tJS89!$1NOG6_!TS+ z&Ti1~RVW$j#kGy>;~njA)*jA3KaX&J3m^jJ5(Rb6)kgV(i2~dLOe4!V0S2`AT^uIG(QL++g;4|Tnu0&|V`9nBk4t?w8%u`AG zTEnS(=owhwWbcT22EXaD0KSjw%4%{==A_?)Z>#z`r*&*f-9g8t#J&Le0M6;%G54yW zmVYv@nIuiOv$tR&-XL8Pw}@22?>dzXNEZ+f1PxNNoQ8uV*(%tweLN0sM>c!71{S@@ zpZ&V-I647L3E)`@_u7Pd@g^fEv}^umw1+g$hi*%gOqV=^&H0^w{H`CA8~y2-NPs_p zk>lDk$hQK8dovhgB`VU>L_)i$8vkewp~70I@#*h>wFney!Z~oM#RF=>L24^v5SQu? zh#^I=KVvM5?>yxFyf)+)+t^x^VPN3h^uNC0-<~JW8GS=Sv+l3Qmr}VHeMCUe zPkYFtP!fx`2Vt0KbqqWwZxVktd_|}|IKGA5!(K0DZ5SupVUn5h2Y?RaNhTivg&w8- zC0wUGhtH$~XmF;$-t8N0+#fJb96g&?)CZ)g9z~1nZ~vL%t)CfW>o=_LCF|q9XQ3`h zPrCyaNP+J~zFEIS;nPAGizQMpLOt58jxl;Jaouza!!w zDENhzJRe<0;1Wn~F|0R|Fn%7>C8He)2;p1hT^ifqTCQ+Z+23_O5fe42?zcVhF5dlv z3@>hfhTM%zt2-euy4n1^(+OIWW$kSz!z7>KIxr{*9RC%7FUr#X`rwfc&L*w_x?MT5 zb6Z!e?2$Pk*9jgd0tZcOA(G$#j9T@Bc^SD^C zApZJE%=K9=E#WN9OY7lj{ilNhUdrZ;bc|XZnM63lyLh?xeHNd8J)FS!ZUMzX?`YNl z1>C!?73VW-Q0s>!-s`<=w>yiXIu9^MMy1*Ou~3xNaeyf#`BgvVFJI5^Le$!V)QRXDda=OOhbO$@ zo~71m0_oA5j}S{^vudbswIr`M+=4{>)yRhWEm(B;#wF{Pnl}5vES?cq=UKOP0q1s| z0@Jsbzwt)A_%k;f@#?G9Z>y)Pc&PgDzrBuXkWHjStP1jL7ptwW*k5eiWM5y{hj>3{ zUOYlyERSd~$)mKdJ?P_7tAwq>*hj4K7%q{|ZRKJK6Q5Ivx59Gc*+N*A-ZN#u)xuK; zu!!7PycQqt*j|Zf6pFh8+bJq)P2!nN{s{NBb?4fA0T!+BDJc9HoL9*N;Jm7NwIn`Y zyi7RIlj+(g({?e5ed*tkl;Hck+PJAvt>9@D$`5)+sYUx>TB7vN`dfI6+j&}lgP4|_=@%U@b0uZy}lxi$;yqDvmzoe5!^6Y9U?O6tyb*4|34}CsKT0( zOc1xfxsYt)pk`+HfM5W7Ft{;(>j~~4fx^k$^F{k_f8g&ieetE%hlc*m9~SN-;h#yT z06y-TRW$!^2alhw*OyCH*!a@<*V)$~Qq59fmo|QfA7!)ud=udnnMjd$orBmUq}Y`+K@=zaqQFy!1E97QS7$@ zM7>vApOSn2?OYt+D_S>ZHGW|&D#W9<=s)5kAQ(VL;2hJ40?_NiErY=bxaRzUy(t*u zXtp8&>6T>!Qg{(dvckT#VPU@WqB6-4XlS2>JM^Vb)KqqmN& z{{OET^`%jLeQ+FPC>VNGgnhuos!Gd+38F zj#8S<{^o~dU^8JdXGe{+#YLdWwLKp=O)sRYDq<~{m)_xO>C0|16~e=23z-oB|R z!JL&KOk%>}1D&r95*=_r#g8h0T$KNX!b7>yNvH!J{^eMH{?o47eeHvV>x(&7~H}RNy6$D?w6|Td!!RNG{udCf@?ArgaoL zT+WR%|FFeLfineAO@#eI_G`+?I(T#*uCh?U-3M?TmIW~}VF8Sa)q2svQPDAaCdDh1 z0q4Gh%&d|*`0BeZKXw=3n)}ZcDC3`f6hElDAeiM=>HPCO`76OrD2j$~p1xq2TkQ0Kpsu14)b9^>o8y1!T@_?c^!GafaJg6qgN*42uCy z|MVbSVXlDTvbjq$Ci#4g8O~|fBgm1#$Uwc=t;_TLn;w?|XwK<@7tb-Hd_cBnY!Uu# z%l>hLJG4$8?i7LZ^a|`F-n2@Bdy^F5sRDcat3TzyUIq)e5R$~x={EU~hP#5C(n=z@ ze~M0$Jg45Z6TM8??Q1%lMhEtnK^%l7GogH@W?)Eg&mgjX;T|_QIyIPX6Xglq&12SZ zUTNc<58P4dgpV7XnN8CU06aj$zwUBs0v}81MB*REBrh!y`$JEs-yZxu_4c<52@Z_G zT3V_>O?m@=mSX@OM5S<>5drs4j!?_Lyn=u@>`}`1u-M^<=$AWFr7qrZ=;cAL`Khcf zW}fxyx)vIZMUM$rt_EBuO~(fZ=j z9K09^hhbwMyYe0YmgVvc4j%&FEBo(WuK`2paIYlrSHQn-5C20bIvh89R=4CVU;h=D z5?MLD0rp`%O$E_tpDiByc&k>@VT-a%Lw;dFXZl_};LiJv*$7}fb1&Wp*ykyd7=7%C zJZ5Ncu#NzD^!~RWMSh$*>_u!7jlenHSCngJ(4B)G)A8GGcXq4M5}KM2O(_S!<~X=3Qf~wh`}m631Zz{$Q4GRG7}!X(aKo2kp8Awo~?_*55`? zlL?0W(`B%+boAeM{dh|J!I@AW38zeljnuW||U0#Sp@a~j$DF9Pcy9yqlQH{i}My+yN4yn;-lw=-O;cArl9H8}Z} z@A?d$qv%t)FM{Y= z(Sj%4Wx3V9b!e>1@f4JUleV8`lWpW{3O*D$G2S+0dB47;N@ed)T#ycTJnMk3O-aHdDAwct5vwhHWfZm;IXyL#A)bJa_%^ zas_f>(|m#=_R4#fuPVeF=<}Hkxna{|-`g+W%(aVQm3wX9W<99UcD~Z*hk!9hAX5 zB9_hIF1WsuPO7W(9NiBDjh7ISpGA6td~Y@mm7Q>Y50J_G!~h;yp$PKQBOMRlu*Q_h zKN8NnMaFphzCXxyiPcfu;l8r_{r#`L2X}{3E8N<=$3}VOhx-0qENl=5P;2(Q|5SkO zHf?6wzt9MG)qsHVdq4l?w7_>IfL#A}IBN~HDYMQKo(*NB_7CO}{z0;>2`=H#BPnUY z=l}T(oG15fv;$6|K|RSRLdhE-!f?#fs3-~ETIg7cGteY|7j1Q zGw8S`JYT0a%78-O2f%ck7Ck;v@RSQ0k2C}uN~Rbd^{rU{?GfQekbE}Q^{eA|jo3(N zaF=@PrW>q1$m0<&BuOPm)OYIgRv*7B`t5KJwE!2$!P+pZft>zk?=~QH03?Td^aePh zc=vAw;2p+bQ9oYbc9_!FJ^KU3aJn z3hFzOAi}ln!Ts&ovh~h|`|ZHz`nL&JssU`xX6m_4xtW$~Qe0nyX3ei5S9Fwb1c1;> z5A1?r1LJ?Xoyh}4bD1^`=e}vS%Km5cB^mDC61P5oel&j+LtW;P?{!@aeXCQi|973~Tt^&`p?ful$WT-UMg4kp<2jYa zY}*{`*P_?`uy3-U)F7W!fRihb==*RzUNxWWr7M7+2u&LJm_$s(Ol!m-^)piiTRukttS{KI?ocii7LnZ>||e z5K7?g(hK;-Ocdl@%m>A>a@j-^ylA}`a7b*y^ zkqbyBdu?{1H(`s+*FZTnE!as={~V&4`652=xG(xG`Ti}|?k#A42+3f|`Vz~Z;;Cx6 z#Hja2vPS4j1Kj-%o{)I!%#)L-O)SXt;hFibwfFC^yl+d-IrR;B^S8xaj6x{ZZZl?k z>293-K5{boiFeDQiiS?&sOz)q)jT&cwTwm-Mn&tqgce7lpCVppUfIrxQDTYK`>v#S z{Kr08i-LFz2kX%E3f*tSZ)cB_K{4gy5x><)1oxcy0YZYeMmKV9pGsqC9AjT*ho=!;<4%_QSlj_q&z+l09@p2)TM5#a%b zckiA1E2e0_i2^#sUVV0B@+_IoTKL_miKxo>GyI~kQb)aheyO!^EZ2{4@ow+&&tQ|$ zIE0awxC)$4P(Ix=0PsiW&Fe=~fyyk&-6<|Z3$P=>j5Y{=>{~y^6D^$6CyAroQP<vRl1LX6>q#a~2?o$p;Ir0;}Zi6H$qgV!5_t(NaL)(YWM-wsOO%@2>a zukkCBqdZK_y(gFB^^8Dn)yq3S-Q;&xjA27vUmcaq|JkKcgz1*>{7rve8#Hf4 zj#x7qoI4=$mcLl@SNiRI7ip2@>Hqs(BHG7xIU_=~kGeltQZ&W%7`$7j2vK;i-+8g( zJS%zyJ)YH-c_IE(YtiuJ`hUN3ME#{PD-h*-vTvQMR*}wiR*+U@6R=8(^NA7{lo$Ie z^Ij9tQ}HET_3>5y@Av0?d`7T-=+1+)5j%*+Sh&z5`j=!2NqAJa!8&a~nMx2oHQ&vx+dJ2z2?VEwa;DUEwiZ6+og zV1I8!WY~@IaFSb<&oGyq`}5GEw>TE*mx36_Vf@LDr8mAOVLSh7l6$dS9A1BAdgm_G zm(}XyB<7x4df1wq~KdHpV_!StehFpj|$mSe)VJsKbT?{D45^8@ygC0@kCncQ>7DqAjMR9G@Z}W$-dlKB5AvBKOWI|^5>P9R zy0cS>JVS_wg>UXx42*yk`Q4(-_Sf&ZVeRqed>+4XCHh6^|M`ZX+H0_Hb#@D6`d4j|!6#=KnN_Acfpm?J@qQU&h(Q z;opE?tncYNyoAtQ z{1Q9cu^LJbO;hY+v%~tiNzHI5%D>+y)O!`1S--Ky89jE zO`m(=nC#Q%f4)h&C*02{p!tWOjm~|ZNhF+y5DPqlB?ttUB_cPUvJXyhLiY(J$^QkY7JZ# zclcU|NXp+-=kd1|vn(Z^xy)aR=EA8I-{&swiBX}BsQNsZal3lTDX4~NlWOPn=XCGYOxybxf{8)H*+)8t9;I<#{KcutXpy)A&FxnuN~#< z-T_Ky|K+)80$ZZ`tyo-w(n{tDNZEv2478i#GdJHc7X#t)A>40bH6jF5AJ<(;ACOEv}=18n0`=utdV#Lfca#$ zW6VD@i7@Xzo+a(mIw*g-6BoU`3IB&XGG_Sto9Xt=U&;u%%p&?XB((C((VKa{Db63& z_x1wU9tm#Z9Y=GGOnhK}`Z|c;xq$i8g`tl(HLcdcF8rp~V*$7ep((hGSE&?xxA~roU2<;`md0Y!F}LE-*NRz zIJ_9=qS0kV02In-2r=7o`>EGd~QytNG~zect=I>zL;2oxDw5s z>5`tKw}$hsY%SY3`h4%twB6b;yTS!a@5jK+$(7pt3&86u@1M3%zVme90(o@BZg^Ex z)_{)3lyYf{`<1fNwyP)x_D99f@4f9)Ju0h+h}(*rXz@xNKjWn!vBE$1tR%4c z@$&Q``xL$%t88<~r+|Bb{b36C*!7(ui6!)(*LbGaKl0DN=jm_w>w)m~{(44VUex@u zpS&|xqMOd@WSM#wq@|$r7LpRG+lW~iV83&X<*U%Ud_4`x$UyY3LjUEGo4U7%_Gow1 zVibM7ScA!Z7*NT_2)+J5#D}IIFZWy4;wvVN&zaHux%E-k+v6^LJ}~u{Y4ms&bROA< z++Pg^0T2y@Jj^r9dZ06BC|kH>CZPD=kayDd(_bPdz|u`QY$m^YBY-qyKxX#cdL(s{ ztS4psU5`uU`^tAkG~g(^2@bPf&@FUIO0uThQEi_C5`)40kidP1Ep?)Q$|hx3w==={D8%(1krY@!>+?#l?$Fc5q2 z%k<@Cvj1iccE;kz!|BFAh&>4wl{fKXFsmF{yRE? z^p_PtCNSv^hD(0`2K^W1qhW=V<9Gjd{kmd>J~ksu~KY9*^jtE`6x zFI{_$_JB!vJL$GEIElSuis7qFW4Oez(H9U>rbFM)+PuC7!3)K@M@ad#xIv0LARze1 z)7^eDan9bxNiPbRk8S)upP$J9V$xXVPsx)L%}5RFahfqiSZbSV^+dIE9d~67a{q0F zj-?+L_9GWw8JKgcSHx0A>5y6F@Nu=(GZ*u6cJb0a8Az*$FPw$r5yeO<0^0f*Bqx(tgw;p`DBZC- z!&T%&-R#m>h^tU84M2I(G=m`pKRS5*XfQ^o>IjzSf|4odC)c23&5;`^ifX%i;}n0h zKGz5Eg-bk6#5+Qi6CXMc(GY5y?D=sKr;zxwKceZ*f~H&lRS zbY|z8JyTEI+>39hcDdk2ca5=;oSspz4ObU@)AO*EPNAbS=JD~#Vg74+%S6-Mfm)gN=$D?83o-qPA`di ztX)mUN*>R2Pqq)J0GzTh-i~s4MM|m?%a=}Po^8`LuNf3jA$6)c^_NSm(Yi|1o<|_w zdx>AJ{l(^z0L5qt(=0`O@Bv?K)g-Gms_cb6whEdtHDud7_wDvOWCxdaJfGQ%t-Nk= z`L=u-{Du&s|15iUnJSJ%-I&K@ckV~W!nfq{%}gqN94*(-Id1I^DP}pDAGlO=3dG3I zF;X~0UXZthE?`;u6zFz;dYm_#h)lcppTGQ1KM25z!?%?xlRDkhez0m7(Zl2&MHsW{ z&W*3gJjY55$8mTp$GhrXa{LxwCQJmvjTS7*P4(Zo+fhSxY@~fo^B;D?IsZnZ8?({r z{NAr^+0h~|6p%tw!G{xmUuzgxGZ^n~RPfqBgR@I%zP*U+-`?^WSUS$+-L<;QdvFHv zf){SzyYaq4I+TFC6x9d zij5c0v^PfTkn^>T_;eg`euzXR zRu8r?OiLXQ(lgeqbFAHliXa{)OYYt+@nD!G#fH19qi9+$x#48gV3ub_=E+`pvcaD5 z+dt3YC#PvZZe3FtA0|d_zUMNWGVoR?1uIArv;f8KBt14m-uE2K?*OvR82+fw>ISX; z`CJ&mq=aqqqJ5MWfp78SbDrq_;@!HwX06qEm{`m3@yAH}p#$DsI z;ZrK<`RcHD*$+g`udzYxWy>66;xhx^UfG}y$ZB}ti+o=-TsO<`n^Hll?=|vM;Z!wY z1jvOZ)BE%L^fq40OM9-@K0-%7dU|wG2to~{AK>nXw*&o+;6(8=XSk}&a>Zh}4 zC=8*233~KRqpa3^?6O-pgVV@iQDlm0{^Tlo`)t3Eg$2P)ccfpLmUGQ%Qjmdpv$I_p zLsGwQ3JuSSX%Ji3Ah9`)k=M|d_`{p?FQ;j8t$*|2PxfC`AM~>Zp)K^{_ms!4{SCy6 zSV(g_-X+?j@b$&naaZ~Vd^k2a!;W1?38?isyXb_tvcwyy&u>-%RKXOO4Oky~vHOCXXNy&>_1S!$y2yG?wIhF6-*nt<1pI*_p4$1n4or^8nVBktsMCC{=bti6E z3LD;&b2^OvIh&rryDTkb&6vR_`cC-s7<57&dHD`X z*-Wu(z%m_USn+0yOh9bet)2?NBo4S}5-Xi{` zUGdH)kD*TrRZX)wT77d@t}JY|zskq!_IPH!ZaoE6+O@^cUE^Q&;4P-8bwRl(dCtB> zvM%rj2XwTvZN%TdpJ*2CdL!Yc(~7s8c!@xqFK;+5Oe_gu{6@ML*DbAK;nb`T4@|WY?i2N-Rw-hF8$w_4z3CG8i51L(3LlDl*@h=y| zw+>-41LQyfW@ccTF<8AWcW1Oyh#Cx2M>Gn zI^Tc2^LlrQ{c=lu;on+^slTjfE9io%s-H6x7nX=#1>SKX{JbEEo;mzB_;GJ>kV8CT zMw+I+NO1`5q%j~`K{gMoAGHg`HyK}~*jKRkycRmp&ZuMlp5i?!Q^b9i>t8q&@cs zKub^9#su%Mk%@bCgjaOSVmk~pWI@;ZmK;E+a&x$Ry}p-6TRiWzo%_&&%(*c2Bxp0k zEt2lvFSPh8ZQ;jD;+Y+KJlY>4?8mL0>sI+4h(`D<*Uok;r=Z1sdkv9lFEyENJU z0WvopxZbTHVFfl^cb!uNv+A}ri&zAHfU)t8%qQfUNN_%xq6g+<_2f@yn=9EA@Pm=b zyVtPvt$qv3uXFS|soiYwq_Un|F)ZsRpfaCxrMI+)a^+Q%PJ0Zb&9nC?1 z$`|n_Rb+}SRla$}yLPhoJma2m?)NKvl1dv`)nm%72wE@ z*K$9BFHRP2fp?U3qhyuZDE<iqdL)kIWA7^5Yysa5UOZ+&mPRT;n?PS^ETRe z9p6}FD|meX^SgH;h2K+tTNwIcR&+WXc$FJG_6bxv(hUz<3lEIP_vq>IPlON&SH3cz z?2uKWGBEr9bs3zaF?TYA_5ys%OWyuj2B;#4;IhBQoOa@_*vq{pVhv;S6n?)(xR9q? z#5Il0AI>(Nd7ue}g$p#3^2hFm!cGB`m@kp!4e0fHkLbU?eDr#}tN!|vkyz~~Bx%+g z>I^<+<61_JxlioFDU!dAG=8m%TygQauySQ zmkh+;DNt%89a;EgsnxHEu3sW}coM^eN0Uhg6v3M(7Ia0>@*?r|W?@;~1u8zw0J z@xtyWatLlL77y%ZVXT80PdpgX`y=gA%MOjvT6|;Yrvk4}qW8yyAO$XOgsA`U*TLz# zKFhHtN9;wBqzyRv0>sJ0L;41^9vV)FpZ_>I9%j?nV)BnCd2jdn)t1~10S=u)+4?lSygnw()?0yO$2(n zzlqtVF33`|PM5kl#lv9TI3*bzb4XNRCGj&l{O*k3<-@>&qL2?Av3f#30ZYC&@0Kqj z%5tTP6`uWG$`jlB@Qi!^{eJON57>wMqmZTi`Ka_h8kzGxUoZJ(r6aHH^D>p7oz&z$=c}M9F9o`}`OzyQy|I%XO zXOF8P*O*fD`wJ99tOMk4hK0}WI`&Mj$mL(&P=A;E>hV{IPmzDP@(M+ks2Nff{SwA) zaBKPV9@#L+v7pTlyrVjDu4HxQx2`7ArJLFOlK#W#C-vx0-_d`aVXFuB$&d^oz1F%( z>1l!M%tEItpE5B=`|K<+O9Ly!Ao7I@{oJ0&g8#-_zbLJ0-_Q4)jU6PY*BHk#xRmuz zs9rT9qyIaNIL&!PkMi+np@bpIU>nCd{ z!Lp=qcGm*+(LQMX4gwmeeOQBaKs)Zy65*(TripP1qk3i=skbgy)A>4XG4iGm?Bea; z6I$zdgb7f7Vk+&wv*nTWIg@9~!24wF?g(`Ec0R@9;oN4wmfSKWf)8LUR|A<~Nt<7@ z^IL_8=ai?t2TGH9qsPi(#033)`I{3q-&11Egs-oUS;mpwVNHXCy@OWp1d*T*g<3#n zbi~AaeYIr)3R_F%sgAQ_z!%P4EAdIw^-X$u(dxe$QNA{4Rq>T&A)ahs;4?g4p3UP| zQW_AW-J*9^te*;T#=9yXKM#BJHxB(?^)qu>|khC@9#rV$pj2KBa)NF|F z1^=--B1_E~z(rMudFSkmLu^J?47DMwD4_Y5)#$lpkO>U+H`DTYQyrZo;!vbpP_Iu`=qHI_a*2E z8j+hnQhL5OA>iIr{}`y7H{U{zX51bo21FUqe*L^+rz2Cvdm6DHPfKIb?mf}p-0T+o zn~wqecs_H%&#Vwx|L#B7PYeS}h%zG7ak^ z<9H5GSnAOmoTL^nemCAR=Eui(AJ6UzHSY2L+ileYvtHfM_=c1E0?u?pBV7TW9#!qR zi_1Ps`5r`r{n|;*-+xjpO&Ap(mtFC~NI8J*QD+|_*Ipr`FQ7~L9-&q23&@S2z6PRO zURx=O&CL{OOuc`jt`ohN-#R(OyI_)KA&=Gl4zL@?uQ&eCK*G?osYPb#fuBI(oXST{C@M4S2@)7_+^$~jr~D8`ZTx) zRUIimue9k(x4bx*YPC5T7lU6+i)1vYhYOH;)Cu z*JSSGJkgfg;!i- zQuwFvZI{xJaAOh?(^ziNJU1JL+wSKV6>yY%?dB@6Sc9trzo4a2`&P`0608F)2Ei*$ z`5%q?q?>zd)_@m7BO(Y4eY1nn&@=(1N7sv*1hPDUD!lcmi_QLbf;#odrcAY<)u@o9 zc;oA+!R`<6Gv{yWNRV`50eVf+37Z1p;0cTZg z-hMoqlUswvPyTh_srB!f=l|<_mb7}n_wRyL4gO#K80KKp#`GRomA&{$bUlTPsgR{EH!x zpK7KmWKNbal~aUZsoOy>bucpUr^YLI=^QGVEjL*=4s>TCeQ@?LZ%p5yxj52M=>y;} zJ#sf&kN5!jZ%YO7!jr*$di6`Zx2Z$sv~ z>774v{!IJCY3sYXCjX@T*;%D=JY10wgLM08(#^9nmzb9U5(y|lhERe#6?SDe&-q3j zZUmjyCKOb)aor_qt?P_H6A4{{d=DDflq%X-VTDF#(&gw+M!WL-n|Xhoa*6IFY)&n=Cvk9}<%&pb1U z3Y~Q}`7HM=!V#Z1`TlU(WJ}l9xS6umbm6^;d@DIxd8;=m-s~a zPf+zTq@v#_o(tB23k4rKN=W1csWiQ0p#P2+5rf)t!0QK$j2tnS?f29+yj&dc(R z$-no!YuXf(c(9LYH2(XmB%dmQbHDybiki6^S6k^eKeR$5_k(}LlW*{M^eZl^! zbc*5|I#o9J2fS?iB`k=cg>_dW^1G4HEa@bx6VVxd{mpYeGl*09kJWz`2&LP%sSd#C*;Gp(u3_2v>)6*MrY` zi(cMUbXi#JBLo>$M9_>D-`u4rOhTC{(v^Pj(`)Xk*nT*&ph5CLo2o_Kc3Ul|WoV7LGp+VJp~rYDYwwuP;;r;#ro*jbUw*baVfR1D(Br`q^<}^M=j!Brb<2C zAf&1eid&l-&(=Xex*DA@N&)lJZ;{?&@p!DA$=8!=HF~o%(eho5GKpm`ym<1uo&=KO znVr4V@>f_$yxvrY=H#M%llc9AGM`h&vo(b&Yfezm9m5FuofFP&z}QXiO!m#ug?2ZH zOUnT6O%#ZYKZ9C>lU{QS!TZZ9CtbyZqYNtg!?x`?#Jbnaw-jqnc+DBQY;wmQK&Qi* z3&+}SOz^O4^N1YWO&*k==5Mv>w10VD@!Fxz1gH7{UN=44ZfwF{sN7!VlL-OJng)>1 z47c6NocFdrkyWx;{Z2;D`w={e5)j3awlZ*oJf(cP0MS5v8ne=TzLc2oNZEqBz@yJ% zgUuR({bVX_ulH7L?&WHvzkSEW6A(#w^-Xxm=ClL9c^;SmWA7`|Cht>p&w7QxV-8~E zSK{1P;-nV&DYI2Yl($jR>{ZC@!6_c;JO! zOWR<>Af2&__UB0eji4VrVRVn(ss@fmt{pLZ*KCM~eOI=m=5|Em0T$5fN8czY6FHz?s3hN{{QM?2t*UOMhUiV0VQC_^ zxpCsjqQ6HX-OunLmuqvp7e8)!>?SX-AoQ5tg0Q-xaq#WDZPf;1vU4m+c3f z-ZY^x&~!rmzR%_zh0H55Z{%3NWOU@;clX{Pnz4Ci0mY*HA&<8^3y8$Qq6^>XoY4+ z?vgwqSM7}xr~R#hrH+^OhMI7t=O&?V2otOVOpAk8V>{ns53`n0!j?My0;i5DV0a*@ zX*B~)2{uGZH%EJA6Y7=WWcjyz^mc0Yjoc%$&3{o^eeeI9Z(Rwbs;|s)TxbH;-}kqz zApoZ7zTzFEpB>vUGZBL(vhN=BJ=zrEFLJIJp#eDgt2pI7vC281@w%ZoqN+_>vq zby`2#WFVGkolfR`f?u@Ajd;IJ5MWYbI8ANAz7+*R&F-K}+k{2)_GVCi)$d;G*s*aQwBr0(-3J%_Rjg#>u++wL(O67PDQo-E(rddeD zM`9EY@^8q#YY&0=3hOa&+@2W)ocoy|iwz6JD*XM{jre;6yS?O1BQoT`u{Y;Mc&bC= zLMyx%My{O#{M{8HHi4}j^L#fR)jR`D^u)4>H`j(%Hd1(SNmvF>_xkbezVfK^f-CP` zHu&S8k^E&rJae!`+~L}jsh#MLM=L}B@+T7v6uYm(Xxw^|_Zn4sg8Z4h&V6)~jNW_i zkA6cg-@VFBk-Bj?Am47fx(IZaw&;CJ{J%L3@$0>D0e!hnujZia&^otgc8l!vhGm-?9S^ahsE|}^2An}Qn}6mki(rz*>1L*z!XMJ*+MRxumBk*UasSqmBq$M z>gr;@a%ZRk2%6}$8%Y`tyqHNL{&$Bzm&X?Z{h1REQ+M&9TgNIveuGBbJZM}u1^k$Ub)~)#M*eB7{zrUZOZb(klWID2@`5AX#+@CxC(JA zN~y8FsV|LxgiPjshyHDS8P=QRGO5vvn>{`|WQ^bqwOI%MIM%aYKKIwOfCY$SGE(^a z-Ho$2LT5DH8h)`&W44!DKkZv`;!5-1?AN|Nb?+JIT!nuVX_ynI5#d{K`p|$Xc zysIT=_~kWjq$b+GuJfgi*)D2ie57rjInTCn&bb8N!?&aP&%kDSQ8GVzB@Xazz&vBKDidT#~bCB!yGiqb`0Q2Ff-1G?ZMgqV7;1y z{_lJLc9tO$7eQkf<2D@|@Ce6h%lw8o3D_I&%%_p)-_D!5Mr?t73{vf`-66~HqIX!H zu#eQ$pDI)f-v;@nOJD>z=zc14K6OYO9q5UiFfjVa+qKuq9OpZIs?_j*E7uZ%YfU8Omzm&R>n#|^U5 zCRC7(;}_LpizQy-09U`8rgbdwtAA`TL!Dt&*}czCfUi@=mRoB+W`vcH)OObrHclyY zf0+#>U^!boF-y9>(g55R*v%`E?7`s_v(P5BGEWG|bcpc5LQM||;j8K&8lq%Oq| zDySxb>k}WZKQVNr8(aVrh)iV+;|ccQyd%#< zD~B3tHvQc{OB;*1#voyH$Vl6hPfhwdekpxyI6>N4gl6*xnfG>LlyBu5lkfA;Olh;a z!(;iDqmguugVar`^28KzdW*%U3%W^J000v3uJefZIlsU{v{(jefKn5i96i}^Prugv)dY#I-_g^bnT1l!83NWwSZWyBEh*9Oj7h@E zjv2~q^NL_A_qOgG#{Qr}W#0y@+`oB|889Wsrn|*T!F&3vEg_<{Cy_ejvze=B|G&Oq zbWyDkQZ5zIZ}KHy1Exw0yI|?UdI#)Djxq6ecVo?3KlwB?&z9Hs$%;Ic(XtIbZAF{= zJ5zXap|RnQkG!h{mazQbPvB;6dutq%bVyMVOu#H|(kpfEE${u*IR-dp(Ixm@45s(3 zqA1rJCa{cdyq8sXEIGKrs=$o=hrM)^TKGB_YBYv+m;vj0RRRH}>&D70`=&EihTG^( z#QFZ=AukSHSF>tK`SSS+fkUOSn1al%XP%AnTiKl;OkMZ=L@n&*;*xw7JOpCFzPdtX zb`zt;?&``Emg|Khbl=IlE;FqtuL*WcOyoLetS-889GiU% zsHxbtZDR!PgJe2Q?^9a$3G1H9B5Rcl6tWUT8u2XuhkgeksWpmDhOFg(0gGo&JJ3XC0W`Kvv}r;owY z+jI1ve40$}ukX`O*0DG5Z0`+`dU5sPjFP>+A7@1TTZuLk8>8P=rU^#MT)m?rtd6?< zy#EWjbs$5dGYc#`!k4gU)geCWqmNTtxX}QRaUmPQ zlr#A=uhH3J`)+Px^q4g~{VxN7u2;sFL#Xs@W!4u6036CmblJA;S>iW>XK%T7MnMek z;x)WAy9tZ!<|E(>{1#5`!iYXdOv%5@@DBogA4Nb2~DywPzK0dNZN;}Ji--q7w9J^aqvPf z>A#8MD2F2+EzYl)TC3u5lGIStSU`!_a^OTPw3qPmi~Eo?Y7 z&Ss=8vuO_J25Fgl`g-h1ewrjGR%pXWy-?%oO)qT;=Pea02o?2eT$-}E9Z z4HjU=qt)K8UwWzfFpLM0ecLUZQ{0DQqY~U70jUJ8o3B<+-)5U*+e{n5XBG=he@1plRZqN$FwI@+09PoxTicJZiUcyXM7kSi zYC5=jX;iiH27Kn%`){rapOkd-UFZoB}1@NGXe;E z0Fi9Sw@^w38Rkeb=-$M|Nz$8UkNmWHgopiv1(#jLwNoy?i_X*IP}vkndrb}mJ`K^` zkF0I(rQOdYT=C$UH8;!>~!}lf_*%zg< zK14b>Tq?JhKYx2I%5FKImp~S(LKBGxkcnEEaZ0`mXHpW6?(K(DH|IztEZ9NTNP9Wn zc@RzzfhqnA3f)Fh!CsNOLQ~m*-EeT%g3-hBd_|lj^l{EsYnMp`dZZ4m053vw2Kfx> zOpoD+0l#MsrJVoD3E|^^A~m0lm)&6^%1muA7&-880gj6dqLb^D?Y=V_+(Q(!y3{L z=r5W;7zWL-{`0n5zHhpwAB$z$Q20f7WBdbpv-uMR8##my2k?h7$%>`EH2(Nl$fv=E6 z-V{@j#wY1ffl!=(G9C>pxjSHy7l@qH68_h}1^)`^+7^I>_oWCdrBLE&eBAAdYLgtnLT$ zNpq`+d&=$cvD#YetyCHdYr0_BE!3%2@fEFc%Ep^-gx$SAhT?|lcC3*2+W~pnMj88w zn*<3>B>C1f)({`eSj--j7h?QUTG@9Hm5VTWgkQm!xBx#uz`v*8+JsZLC4pGY0>awv zm(4%Ee}xCY^QS(*8UO$i&%s*)-st^OtD`vXXWh(~-7*FzK7PtwIYbl()<;wC3a8`d$($^Fd~xDZP+!&|m7W&&h%h*HHq zZ!Z6>oBN;p>VBvMc*FYXes*#?R)0&dK%(GAdhRVZ@zd9#ZwV-umUHhPrOy=~u*Drf zW1JWJqFIH{ZXiUCER89*tVDxuUJ4Ke{qTxptq-wm2EhJ=1wH+kd=(Ts}MS?5+J@Qhg!68mcU#}$9`%+tLhqgP5iBK*}uF64n+lUL*&_KT;kU=+) z(e&0@y|rvukT)44zXDx#L&p)Lv!G}sXBVOd**y{?=b$-;ZRQ;b zvHjuKM>zAg0Q_F0x-NCW*@IcVcYk88JT^s@hdL zX8QC%Ym0ef_Q|o$IT^`0QG-tJL)FnQKQwt9c9^R&J?aCq{(6Wo&-;%TpucunwqQ$$ zF?+E-JVgFib=dnb_qCao{A9G$PIK^Mzy;Kp0G;?{?Ld$D}h$c!7Oa;r8zdp zi9tvbvqZamK3l^Py|OO00^$)a_meGK6Fc1`Hfts*+x03zVJm<#lsA~py&&D~QWip= zO7@ZmK)Yn3}wGspcoh^hVr zjyFxsnY?kDTle}xN-O?m1VcTv6+pV=d6#yr8o^7eUyYIqmcJ2Ai5R-^cxhmfC<;0G zo;rZyBHxLayF5|g`?Jaw& zRA|~Z7?JnS_kl_J-~IL98m62s}U+E}Z~oajbR`ft3yO(4_2@lLP( z*HXcz^PfX|oibr0kh<*$g#1^wlO4Mhzk=-1&dWV_|1A;qbn9@!T?b|m6=3W}QXJ{U zyqvs5T3ew<#w|+(=>~H|Z0%AtuW-|GOI9o%7iU10w$fF|(b6HU;>B?wb{}f>f4?`4 zez9ZQBHj&ac)=a95w@AsAg%+nY3wl4?!nqgQ=HBX43m0w;VV!+a9)<4Uq_DcdI0?Q z*B4VqFCYni(-#5~a16c4Tf=w-3OA(z@UM<)j{kZHYJYM4cZhf#&+SkgB&ru*pSU)( zcl_78trw6~KV(g#fHU0?LV~gFn+BaNL6=nf!4!BWLFMJmLDx6#4f5rL5+HZ=1GM=! zu;|H~g0%WUuUSNFdj}qH%o%!2TH3Uee%Ci7cRZpu`Ce>BO+JGPlHr2f0i8e)1bc1R z=H*C!C&%Y=LqnTFctQ8+i)d}(Wh=R281npH^>gu|P-TbrC4{1Nc0Ni(GnAk30#W#T zW?0BZ2LJai)o?~1QNO}0eVL>F>s{KRuGz~R%SL;swL)WS_}_ce0U58WTMxwMOfuiy zwa#0fi2vKeK~~$W>g}BdvhPflYYyZ$*%>Do6;9ey5rPUG#h5akRf8D@v_LFh? z9Q*>fa^xcg^fhdMJ73QQS%a)aBFUKgi~OOk<+p`QxCYk#$rKH`x3rfn1DGJwXIEv< zK~ASFOYRIbe*x^`g79x?Bw7{d|q@l35$R3rlp%8sjP-nXm7AW1hHz-$JTRCIvsKqa4jt9S*O9oQRL{Bg({ zbootZ1?Y1Kk?3wg84F*|Mm(XNh{ikJ_*=9?1MjF*6)|l+$;MG5<7WfKXCd=Db=2!) zE>Ypz%_WdAU)f`wuI9Fy)$X{Tf!C-UN!6Qu;!t+60o=oRxMuBp2?+n}CQEJ!3X7S? z`Y>Vi!8H5sivD06X>o`shLF~EY16DuuhXGt>ueu#nG9~y!M`uvX0Hv|{lo3vAnbb7 zc=CpsKc99y`vjpGi^DUbc4B!!yw2JVL+RVB7I2~^ zW&S;c`$0Py`37)9zjH~@WK5+xIwIrAH3S-+VgKX#)1UM*vlw7j-i zhL>3nSvgpeL5^?4 zTveRxl6Bq0pI;6PslR&%1jbr%Q%q7Us7YR@J(~|4`gH;8h<*CQ$Xw>&GZt0KfMo7+#8Sq?&%9r9u`xgCBZt)}4QvxXT>tZkBuiSb=^SRGSV$(b4 z`JkHIQ}huZyGRC37yyy8mG_td{%3L8Qm{~p3HbQa8 zIr{yQ#ztbwEMhld@Qk;;E@um>Ya7B%u}Gj7&t3ib&SxrIphr^M*@18Br1{1yXU^aS z$0Q(_QuYaR(_gp6n;3^AUdCsW^6hJU%a;kCODHM z(lUZ~RCHo$MK`R;}1%K(rPd0G3bIeSl z`?i3E-2nA66Hv~aW)J2b(0|Rz_i{1dBd;0FY*4-{b8YiXs{E$U7^t%btw{QmL8^aW z`S>}DwkRVlsMBq}br%*6-x!xaT^ws>%m+8Uem$j(LOfs%^n2I`GJ$|wE-sW7t*;Bg zIK?jyn!*B9n!Tm#PBmF=7}@jWv=aG*;fk=uy-RiW*I_9U&@a(JHxRIQ%D8Z<`4Vxq z8Rmexv0>Z5PJN;7ot$YIa>E^y2bJ55T_HgY_|zq(nYP2<nWgPC-O*OpEU)p66u^J zcWgRh+kl?%(WwLKoJA2bI&FWl$?qgGW~zv1q|xM`>m8W;EX%gi4WSzegKwC$;biVC z3!QYl8GYlLeXQu(^{zsW9{ORd)bL|}M(_v=f0~CcT_=z8hjE8Y^nO%py4k@WnG#F- zs=bOU4bX4uTN7!H<&P)8pd3E0BqB|;vn>w|@9)Q3d{~CiFpH=|)kqJ2ElQhbPrg@Y zQs;|9%dC=n-uu&Pb#V!LAkI*&xYYW_JT@J&@YlGxLMWLF8e!!=-}L9Sc`BXNB9s%% zdQm=}%WrkvJC*ca|NLdu4(qOGHPb*TMoaJwHDEV6OZ^4WxqR}ZNmG$a^!v2C!$5WA z@of||JT%gt%J8|QR%4$>YSP2o9@V6x1R}PmK&Q=4QlbiQmX}7tVIaVJ;95oneE|I!PW^-LU?nn z>_A39H1nnQ#ZP#{BIKzeuH6|0_NYGT@*h6Ku#S@``hn@nK(|LZM{X=1e);Q(@dXUi z;0;%f+#Z(y_MBR&x~dMVkHnd`nuiC*LdQ*xdzfTO6a$!yNk7jK`90^Zc8-9|tItlQ zO%jb_+Jj9Unj;XY0(VnRqtx-nC0G>s>=9rz2H8Ok%#))KeW|3@dxlZ!Eu(yUn~hXn zAK=CKs%dvC*U}87k=7UzbdTls!NQ6ty6LE1aO-dNovYe&nhAEr-lGitRDp;~juvQ{(nXTMw5Ekl zAMygs)fqR2^c1uPLn$m}b{f%NDu~yCTVQnv>lGei1NEP; z*ig5>i=M$RmENx9LN43YjMIz=a!KCJruLyeyWVP$iRMAU@EC9os_S{|Fb1^c)Z%bb zHl7k$$1f$nBczgPM~Tl3Car#t!rtzV??2z0SdDD28=<5kw8muP3BS`#UeJ=sPDY_- zfa9MO(>)KN7x%e)!a3^3EUjW3GD!~MGOg5&r&A*OVead-L>R`aQw&j9$FkX1zJbr^ zja(Zn41zG$vh#O%c>|r%kFstJl#8Ou0{|-CE3^(qgIeU2UXbs0DixS_EvRym63*uEO>ZCAe1wZz=fc4Y6%>xogngxnk2A!fsG2 z($}q4(I)!1Dt}YIKirxPa1hU&tYu@@(^k^)#*i>$JIar1)pa!OARO8HOjL*_m&?nL zrl!DHF|&gQVk_*h=MuNp>6bZxdu(saMI(`}Ce(0*^X5 zi?C7+m{04*LF{aDv|Ygw1l-eZ)vBC_X{=#mydTq%#B627FS8qaD?mcVf>BiC4IG4& zVH7Bg2g#JklQqzeFfA*e%jRIt`{P`?LtoWg)4Q?Ch6Wet5r7SuJ@4Ej*aYt`fnzSY zA6!4@gdW|ba3-|kb{O*PR!oj*9Yb!eE^5ay8Em|Y>W1d=Xolf#fwDon9Zh2Lm@21c zk=MMLEiANIfx0XnN>@MqLO;&J?7kyvi*nhryL;01J0lVIcYXVhhS9tBrTz*9$5s@+ zQFZ=t3qb^8%Lc$Z>>aPZ6+4U8HA*jbnDOE4gm)x|SF$$=n+tPE`#%prq<4iy$n?-3 zCz_z!9Ps-XK>P3cxnHhrL9@kxOAkTnRK!sBQAXzb{-L6F*eH{oQ2JRgK~L;3PIVtD zfan-5yhxkQk_{%h4g)8yB}*H!=g0wuK2(wYc22uED7dnLQAr= zO#99FX+8d$2#OBk6jAnO=)e_M+;zq zHAub&v@7LW3P&{{) zB`NXpno%J65}*gqEi4pYqTLrLv0mMyh#2XP8sKN;P5(r;5dOxzpIDia2FV_1(qN2J z%E#^TL54=6qn*0OFQBreld^46xk^&rg*m6XvcPtG z=cLs^`Ry`jEHXKiQ>gu%*^n|P^($N?lg%%>C3g91|BiVM9P_7~y?=#8^SIaHOA=HF zTLl0U{=JQI%p6nAp~{IM6@0WkNq>QE&!KrBSFsm6Do=2>^)%t)n{#x6i_sMKn|h~b zk|K0ZCF1+so`fEu5V*$cR~rEbU&OY={VfEddzP&fBPy)G(AUqKx(WaDy(XDu{!zJ_&&gjCJ}6o`@@yipE%1u~vt*|0+zp0^=w!#|+|FDC0)>Yj%)8M%tF^s< z$U7$Co#YTA(^?MpE~?bV_sgQK4YedW2TE|l#QKWQ<)|`-BN26bQoO^8#LO6U z#PmgJCP7h)U%^W7XMyYI`*m(m@PkA4F&esPaBJ@^(4oe+V+Lcib$4Phg8?_9Hk)- zGuM9bWUaCu=~uv2slFoI6m8>U*AVSf{aAheX$cv{EIbQ<(|?kA^;zfGoj zQW=v)sl8VBt~Sc`M|Sw3bI=;AL`?|`3tlXtki$|%((e#L*)1D@bwJ?x>@Xw?t(3$ z%hxi*c$=VtQ9b7|o>=bw7Ve7;6z!V5F%$`@|F?G`wJZN(!C`OxHC@@8a52nY7(JAL zyS&2W`5M4yPvOB6z&(B4NS27#EPBWxPR1X+A{v9X#Ouv|72u%pB|1{!j?Yn)h=N*( zWCgh*Y;O$KPOg)$pAXEDs;rL91g|7#dwm7cs28b=$~$D4<_?ftS}+VO&Rb{21r2@{ z3hscE@6@(|E*~g^WU;j5AvAzK3Kz4(`s=>}${tU{XpaQyn#n)52d>!EtX3E{U- zjzknOj3tiq#H-~xDIS&VMU1%t_vVW6qWL&~9e;JS;NJn2sr*o7&o}h1uzdmR2f!mC z9Jt8wA?{{Y+E3Yljv7mk5Ex2&75dAlkLbM^bE)268lko*c;T@i48IvTV9Z-!2FQnm z3$lN#h}tP`p16N6BHh~LFNtoiPx+y@dGJYBOI1IoI`BKYEmlp+_?oi(-TBm$&G z{hj*jC9)w%%nj$A8;E|=vd-6hybhn6^YSJ@?F)8|Rv+doF|Q9U&Jh{nfxMUWc+EyM z)|hAf)K@Bxt8^6u_vA+B{JjT%Z8x3)gFT-4gRf_8wnXvwRv=b@M@hpEs-w9PrfY9@ z@DXBKG{4>)h8tiVd4_-D?F`X;$@gl8ozO$$wnn_K=@xV{&t2l+Maz&V5>C1U0#fxz&8$p1RSUQ;F4WL_%Z+WCMWfQ z|9jr>3E+8)N0vDg>o=+Q4itVjrOqaGZYxCXUq5NQKv0y3X#k92t+1sVVZA_=2h*Te zLnAt<85#Dkwa6DD4MN&6BrKcxYaOlOSLvs=m!Cn|p~-PHl6| zcWU5*@1faOCFYcv>a%tyCbNut{=KywnnecBaJnKq%e3#B{cPK_<>NTk{jY_)%;QwA z^9wHEp6~mA#cN&J!C5F@x(v#sGapdO-_BLOKJX4L5*;)GSqm}pW5tY60 z!k>Zk&{~GN?5ww+gX&=+DfJ8`Le209fSq6E4*_;AAaD`L*tA)k4;(C7%l;(Tj zJ|tfAWm%d=0=D|_>~99az;WLKQhYsv!8h;@WE~v06=(ukS&(U5jC8zU7x7z$F_I4`${KBo(58Ty=HgON}@$I=QS|^w8>U79QxVjGjAo&QiMqQ2x!03FB z@^?nh8HxX>$Ov%~Ig;LMI-Nrm|1EM)(^C((?KzzT^jRRIG9)$ij-?1bY4F(!NDGLa#-=SMJPenIyvfrEyO zY(FRfG1`z}*UIE1TJ5fc!maLc(9XX{UjKY?*caxDEn;n3Z%{5QG>4=1=z1HB@G_tz z)`o!DT}|}6S|j}=@Irm!xBv-7E2>Hc8Sagsx*Jkgo9!e1k%;U!zQ3AkVLPkoFBX>R z5iYP$6 z6l{8N2#{A#&OuWdkNC@jmz{Y_*RWpqLbH~t97T~NWiKd?Mm2oym)JY(>;Eel4El~F zw^6NShbAL?HZ$3Ko9hrb?y&->fJ{%e>JExX?F66H2?RrPYp8#jL?w_$g^T;vPC?_Ih} z2aySi2yF8%@mKveU~v=3`B&smnM4JglhR)pYV1{JTe3Uw>=`1 zUF48~BS!GULvv9@Jsc=pfD&Yp!&CkjyXW#%Q|>9+ch(%_UU56ev9ODb%Qw>2q4|B( zJ`cd13=V5sl1wkY{B4E&ZPnE^Xo9~cUe=3x1bMU6*K=2=E>kn}t&E?dc$U>&h(?2@ zztGPIScF_zc#aTxuS_)JBc zEWc9y2-Zbs`DX>|6?!Z`?}LrZ={xw8Rs`{{CMUuh9&#alS z;Ookgxtv9K7pQI2WLv5FIp)gpKFZ-^YCjAS7)G! zLzm0qe<_^7{?(jwAx;>7d$2H?04uu#<(1q#J{2_>2!ahT1Lz|bh%IX8XVGfLd~$*- zj`6!yYGj5Z&#U+Kv9VPzbzT*;Bwg4lbTrj02iwsnmEfe)arV?dIm|w0ob!&%P0%z^ zzOe*fF6E7aumQHlb$W?F-V<%%S&qY7YKpa(M7}_3m=F%f-!I}FaCX1>Pys25eRJv* zM45+c=zCZ|h>ssVARz2;k#MRN_|=<;AJPq>HG6;c@Z3$qR^?CvNrDW1547V)$g(!e z;R1a11u@nwYIj~aVowsTW>4BALSB&-X^dx3E+NPLMqIXVU7Trj+ci>$^4a$iBwPqT zEk@Q5j1Ib+~gEZq+Q@COb(2KGgzIx8*Z&7U`2jw2MH&&RX3`2ow1p#(!l z9XLsv`M>}N_OaYM3tV7gy$cBcUSPjxUB2Vzk|ys(+ktUkZV z#_Ela^O~N4Gc&!H?W(GY>i13iN4T<&LLS(!&hEyU7Q#?<%=^)t`9)`3g<-_)GGGHi zMZ7PjIwe>k_BZ@M`HQ0{hRz!_Fn?_zn(i)~N1^$#=LDPPEORBKVmsp4?^fqE16Y3Z zs4$WjD^nQmEDZoo9q>ucCI{Wag;Q#IH6#hq$}=U`Dg2aA@tZm6+AD<_kEN}dxDz_H zyt075r|&86?GggVm^7c`1N0CG_%KlW-)WY~LA);Wq}C&1--yp=Y^)6QI5QlNl!_~^ zas-KGYDYME;beg^^7$5JJfuO2<~{iaBwb^V(@F94gB4E&S6bhnxDLeZ zJC}L|kegM6L+7t6t4SwP@WBR$5~q|5`L`!go0cG~SeGfw*lGUNdK0e@bda0w<}hW+ zFD{yT^wR=qG#9~9zd)#q?k$lwYv6SGbno<4>WS?PvZafFz;*5x(vABM5i2;Jq!}_c zKHw2$gUmn~XT*1@KZZIQ8;B(hMMMaQMknzEON#<~R(p$&YRr9??{~|ME(a1R`1q*4tidlUP(7;UYCBd#2IV*G6($%^xpk4p zgGt2G3wn!&#y3rjG)_ID+&Tte{aupB<@DvYl(m4(RuIj4b-isu^$B4ty%kz!K;(cS ze>-akX>2~mS-!XsyF=>CPFVEF9ur8>e>=efQ#h03a$D%5Lf8H5L~8Fv}^5V#?S2!f^( zU$ve_4KI}540s~eNg^^+-)J3@I7HeW5A`{F*ZMBtvD*cT&G=nV!{vIaxp)uj_4LGz z4clvrl{4|2C;)9O%{u(mCzG$2T_3nLUbydzxXt^KEGie#4C;Yshn}5B7vFre-+b)P zbi8+CppEPABYHudn_7?^B&-sPIau1xK7nbDYPieN7>1Z{8Nc~cdMo9V>f46I@Kfg2 z?6xB-iMY-a*5gheRp^9)wD|ZVDCx^ z0#IpT#O~Lr^-Ad_RG>dx8a^P%Q6h?-{L4vmf&Nt_k^+v87@?Iy61daiMZ zm6rh_o9tpW$uE;RPojW16E8ZOxkxTMTK=yKJQkDG|7|UboE84hs?G&;VF0n>bzEVw);`y1n zk2HR&2{$c;nlQt~y9boTIGfc)CU9vCWEj@c(Vvj?uMUw%BHA0&eN%STEm6 zP4-Re`Qwp&cc^A9G2FpRn>y>IUV_3>v@re}cxftWx)r~9>&?qm7T+((;s7VBagO+Q z5(}6Ol_89=Q}o$s{6OYGF!wn$Umu(}TBk)>c>Nvc99KX`+iZR}ruzvQVQT8w=&u*c zM{tkEH+I_#xE(C97|kBNu-pH330|$XDq)tRT>28*KAoA`lBj6}WM0XYP;V+GtxqPk zM34Ky^^{Bw(QFWrQ(o4aU{V1qC)AVy)?Y2d{QAK_?wTp{9P-zIj`4T4nDZx=Hz8)K z*Z6vl)-HtYgcy{@*}jy!c<%0+W(x)&hMB;K4PZDQp<&$0(_3Y zZmyrm#Y|hxU|GYjIMmRg%BpW6ZH*jJROA+H(EeVYps1F{1nrbIQLsHSj&s|uChBHcBEW2T{@#Ci)BK@5h92#r{P~4zTwlKs!_vtVBb2#@q5H5vfZT{_ zMyJ;|fF4!%Ir`GgDQ&Q7UB28dP1H_`M&4 z%56fMRgu#YMEY$nw@0XXzInyTf*Y&VRO6}AdrQRR`5mHs{Nnj2$_@e_&s z`8xx~p~LvWMI8TohPh0D)Mz;QsMw4%q30la3TU8em`+W9X!= zh%I4yOW%)RIY8UFJPWy4nfIq5kKuw!ZP9;17G;SsuPblkq%use-PZ)Oceu3w>y-1!lE zWluC!F`$w~vZG9iPYSiDrF4ISwt+|Mr7Div zM0`wo28|l&)kX^9$gImFp6|QrfK%J|63k?jy>sO|T0>skMZ#Bm!Jb5{8?bEt9QO#j zy$X`y&H`d0fJ&g4-~-;xWc^e!=B!5b{+zLcv3A^HOpE0JY+;=t%L*+77(0##kPS0(bf zZ}jOcZ#BX1{Uiym9OgOu*PR{&WVbd9;N%~kkziZ|2d5`*j&z+aI!g3yA%>YG$ySLD zo;Y+ygH=bE5s@3UNun0;&|pFrl-`)DF&&=x`|v$}UdS`%a+#M8^C!KV_V2r1_kk=$ z9;64zco;r^_@XLhjX2g>-loNes(2=jZ+YlCHx&n?u{6SS9D0mHvX*KOeB%83>Vn1w zBYDhGCfq}bC1frT1mOcQwVDN@XcQI>qK}f|-v^$J)7OO<1#6jooQ1dRVHxK}J|dUK`GV!o=^MN!5XJ9K^k zO&CgmN$Z^28>Lbi|Ppe_PNh?hHj<{^rc2Q>7bw&=hu<3H>%K%R1Z zS>hC0@H{qqdp-X3{maRjKi{03SX(Pq|CfLmt}#O=Iw2!5V+zeyW}P4?K@C5ob3y^r zVz>^f4!^P-V~M#>Tf})69%%$dDbfZGxitdEi! z-PiLoTSZ{1(HzZYczW|FB0m^dx5ERVQ8&viBZYS48>{HWF0X1i#q$6(quOu^LneCG z#wIX5K^C{6*ihKA+xzLLx#0SDoz#f?!2RP>MM@tg1e27_RY#Y1{(1VO%3}T;VJ#w3eZncHVp^3%+Qb z<0N60v)`_C-3$>cHY0foi`~qro*us*aXNJAUR5PQl`r`!It-LT9~J#CQn(vA46p%Z zwXgVgFv4L4QVa*U51?OLD8kC`Cy}XEt8n@9+Ub`H@N&Xy$*GsS9n5{_QyM-eeY0j7 zzS!VP@)^`jO53f+oRKK%-Kgj@<rC%|xA7m-Kjc7dvQeC-1*PM$j>LO2a%a%+fVOq?#&o zNvZ+Ol1E&11u(<*zrso2*&x_gk_(qA;?PyZXg;|O;)W18(%)Wo|Hro$rBFjOtk5_y zEks~NTRTL7e|{-q|CEV4w-K=cv!~ZAym?4+{zu!F^}MQV+r8&k9GzDp(U~sFjpPnv z(-_l36kUM<)AwNDD*rvRy!O31H%gQ|B*tnkcEVLkK`9;8e2WumwzzOiJK{vm#H4-8LR_Ewi#zp)pN;mOB~ z8@B!wNcuX>-xq~6&w8f5|E{*7o|okvC5tydiu<=jwGGW!|NyA#9Jp&X!&!UXmus`Gbw8>W%kbMnPi`J~KP-?=6|F$6`GQ ztQJl;92Yp6KdGl)%kHkorJ?ErkWYeFn>HTuR@FwLP(s68OeMh=jh_-)`L_*=e z@#&k*^;eE!Q~KoQ>mJ>IG4mkMLh0}i`Tet^wU^^ZIWR$-4$bs*DzKJxY691ZCmdI6 zOJnfV^Fk^*ri3HjLqj-)&E8%vG5L%>r&3Qp?`$2jPX77y+m%|#uE!RuvmPV0ryGmX zTBC9XYJ?YTqxg8njhLSGtvxo;ArH?Q(R{q=%lX3TH#c{4yiNN7Tl$=Ei~ByZRH)+5 zJ@>z=erB2P>vozLJ7q~XC&OU~$`k#~L{M|?*T*d&jay?3csyN-cajmB@Zrdq^nV=k z8r}P%q%WVt)3`cf8Ju7K?Xcfkw4>nbb)|J7pMHDgx%dlU8J=Ut>A}&N&XvBjaw3;= z{pG*&IT^6qBe>_`{eIKXJld~h#H*~Q1_Rz5<}4}NQ0ZunXWvgxZRKq#TK}DZusi|F zM>`l^VSJloajLe=m}iWjkX`G(uig02eAtV>cq5JW@EWm5`kb}lpiPpMx*t-HUOLS~P>t*Biq1^!%BOUh!0VpFjJ8k3g&+(6nRAEJl+h`?FxXTPSdPHeUC6oY zPSA;*o?yD?&u^?x-M{4seNyAyD&tP@vAb;+0q$*Q;{LmbX4mK2cKdM<@mS61v*(*= zq5f~=8?q~99IpPKYfJb4&$Yd0I4{Q&kM_5+4<7n=L)E^Fw{Y8%9we$FzBc(VU**vU zmg;1GeE;2LT+=QeQg%pEA(Ogzy`OK!1=anCn~Zwtr*$7<07e0vfyM-d?re{F+}G#Y!G$ zU0?cpFj64^S8Lw+M7J}|iae@3^*mY!L{CPVu)k^jLJL%C_u}>aq$9O0g(@&z>p|eR zm48bP8GCDch)p`6rp~!Vm#N-^UNn0zr`#mfqV(Zu$s6rW!woo|HY=Y>r|sQ8z}0gj zJYXlt1b(@$9t%cbe}O~)o?6Oa=ZUdd3G;MIqwp|M%IbcCtZbj0xR zwArh(YwJ;VT^QEPjy-jqWv`iUcjs66*k<5`jGO4**rOUPCtJxclZ18O5}^>NYz%O_ zKal^~XMQF3ccJfO5^6QoTaWL-j%()4#=rA%gEu-icX*Urs-af;x_8iqf-UkD&Jry& zi@_L*=ktr#ovrQV{(Iots2QR(vC$knOfnSSeE|N@4#2gs>|q|-p5t(Q8ESpkofME% zNM;wtoiUuAdH1i2+*j_P+_`kqzE7!G{oLPFM~X&D2Hwdrzx~ZSPg#~UP59{pngZ}3 znrLSW^rl|1$U6Gf^YZa;UJbA6N)K;BN7LG`Q^gA5gm{2ii&^VFmWbs?kK^9-Glcx)f=+T+IHhq1O8P91O3MS1#mATtBnusS^C&teW| zRIg48Sc=Zr<)3xaFg-Z4YR`HQYxkKEmdm%jCh!)Dr)?#rb>S7B;3@6=^Q&_LM*Q{^ zyO-zNahIHF?rw}_SC4NDv!gY?P3)>d=Ud^*h&B6@j%JKVSJS z+|TEu2L`a(eK8i~tz|@5AC&72Buj+i<@7yo82fXOF>lg_A%SrB$c#30ufoS=^4ua584 z+PaDkU_Kja?Mf9f)m|{x_UI1gef39=;20qRw?**@ybYtgCD2rgtCf6gPJH2!YoZyA zuJxTXa<}jK_sCI0#|ZNHvz*h`{H*7qvElGN) zW~CDUw|sBaiZ!^B>AlZ+?e~s7X~H|k-8GT&!YqEKKl%o9p*d2tD&B65y=WsSDObNa0joGD)6#SV*WaAt;*pw;2WM{O|+w}9u9HzQ~Oc$sz@Is$-U zc+sC~k`2_?*7e@U?Mc^-vTy4t<#v*)@%>Kh{VV8!I}Yb{xUl3Y!Q{tdeF3xy)yDXf zSqF8u)Xk#o$oH*a{~p>6y{-3fx;6pmF#4LQR?Ax`=e3ebRFLdkiC3?l`>^bKzmJ=M z+A6U5(0xB8@$5!MTPDiL$t#aE1XHkkFp{QzA%rTyFpM}b@8|ibHcDxVb%LK_@PR-i z0Jl_=7d@`px?$)@Yjy^fG4-_`8^NL)geulEL(ZhjYHT5xQK@})6qBp-xxT*+JKXeoa=9eC^>JuhE?lwIA+ z?;(Cp)@FjjnnvqMebm=$$|-<3iAQ_TlQR)UUU=I(uNB05?E2YyrL5;3pL4OV00$qJ z<3<3sZv5wQ>>LHS)H_uv>wZPV?p7qrr3B*k{Wka#t>j%iy<8PAE75oO2Dk>y<~|G9#j^X|{)fr|z@WCoP3xmc;j!{2G zx&Y-q91{IK0Dr*Lf#sLu&8dR7C9RlVF>ehOW}CFT@Bg+UiId$W^)v=MEU1E$I_hr{Z6|0<#$T26g&o}rEz>&xtPYWW_D7J*bz^uoCuRWNr2()4Y`h`l-*j=-` z;Y-SFouP`+711m2h{c5w&`8AFNZdJMwz)AdkRWxFo28Mwa*!QCKsX(gb^(s`cZ z4J6?Q#n74y$%#3z8sG|tD&K-*c|-;sY5DhKR#rFt&eeQ30TDcDe?+lEt;2Cz()BW^ zgUx?|Up)_`we!6CRHcFn(Nn(}o~G%qh5TDEvhOVcpu|C=$_AJ%3+q5p(!IUiZA)lH z*`}#|yWd~0*fuMB_;I8z-OkruyqWLqiSm4Z_BZfZ~p$D{`O!0 z`t9HT{-4GCGY`q%#(Yoy#vTapy2RhU`IP;Q{M&I&#{cs7zb^f6fByg9|GMGd>tqV! za0=mW|MG9YZ0D()r+=P+!@%$V9_J%`;A((eEeVjy8d5p@UIV<*Wo9@yXJ#F zoja$e@g?Njr$sNOnRa%yKk!pYz<(s^eDvY}=ltoe9J=7?8KNs0Rzw9vOo?0YI%&RV ze50E{M7JuxOF{K+v@W6!yHb2+70f9Fcl4H2R@sES6HQaUj-zg z8Do!ogL>(0<q>tKYL>C*V3~8#3B>( z^eES#4PT|QFduhhXSs1=?!?*r-Lw3dhW3jep5V9F+WN%-ZFIwXnVru1{c_)~9r;`j zck#?{ZQ6}gWDnqDsFc01B0Q?qUF&-woKbxC#d?;FVkqNPHSSfc03iqB`|2`-x?S}; zhmYy!R2suL|K8(d@wWG-b_)1E-nyMKpE_TKkuglv}c5Dd=;inW-_Hre~kXuluzqFN!9wYa_ux8jtYlBQ?3$Y&d$qv>YWANK6pFgiiF zO+#E0p`rfbb=~!LISn~CQ`!&`@co0wfPYQj*caC<*Fik-fJxaWd=~c&6NHoE4HbFm z1a(CmzT_hKJ6#l3nzUlR`c)RWV4u<%E{=kZc=GZ*$$q;?VVUFr`{_!-26%x&D ze%h|D%B*%F$to3w6<4--v8i=6j;~UF`9=BHRKxq7v}($gG<+a)?|d7@?F7(W~1 z?&A3EM)56BL$3fyHZFdM(Jdz0@lMpi;x50CM z7iV-esicUL6N$NzcpV!87dzF(vfx-kdpsuHs!F`WEEq2i{tR7%OzF_K({{qw@;n<8mujYysEQwi3s(W;}U2 z%z=1Jq>SnL1WaX-@-^vUycy2FuV?WRJnGgsWZ9Q_m#2yu#wdl>Fm2{GPP#BBtH`6N+X1)b=C;u9aOry~ZVF5;;U>Zsoz z<^FUZvT2uV5@1Fd-T!myuJk$a120}%PPsbZYt77Hj1^80lPR1kXV|Hk8Uf`IB#2E(iM~RB}{D| zJh%?VZZ$q-VO4;xAS$ZB4LiGc6qy$A^dX7%wq%PeF4P!fSn8kSYuPbQnN15BUOol4T0Uh6w5&%-fR0=#8 zWV_cM(RdbZBAxizQ-p-d50r-NNhifjLGxq@Q&K2qbY4;Z>lnJW z>8z)hEnQbWi3ELG!0S)n`Uvccj5J1@mWUE$XQrY%Q;fs;dw7sEJxPfRr_h;7dDjFw&T9+qx$gd9o{@;*FdhZi z_N!dg%MqSay~tbHf$EGGkJ;kJrrG_8>nz2A$l;rNE}vWbDIHL3 zfLrqOd#;^blAXMQsIA8!f~^N3n75Ysw&+D2;U|sRg=W)H+7!L1-U$~FyWkK@NhA=ehlGXcv&1LI^vKaMMtE`|${jHYiaZ|8af;M-!I)0J;oAXYUDE+Hm$z zQD_800&tUCzw4QlY{Vzmi*TaQlmqgP|XNWQ|hTzPaiE#Q-;9ObSSAlKsbh*dG z#@wuqkWb*)ZWk~cXQYQM*msUABrw}S6fipIk?y|h%VENIY&o$F@|XPkt|-4VxN#JV z#(4$DaUX~KCwn;AJv9Uq##{MSmOp0vXPh<|lSHZ*J_VS-Vjqa0BR4<PTPU%yUa!~Q%bU6m%yGI8|d48=GUzop|170_U zE|71W2B2Cv9zV6uQ-1&4OY-KRvz}_j0!z29M0_>C2kR)l&ab47?rmQ?qpx0F^tRfx z2is;@t*KvJj2+(JqrM1vXLX+)C=HLHyr@2>dIO_Z7@b;!96>zv?16LsUd~@$(~OGW z)Iz@J_u^IL6i1^owm46TZfE|yqoC!z+&s!|$Rto}8weN83Kv>u1!BbolrKzIUO#@H zHjn8A%Ol6Q=*v9+;{4~q$tet*+d*+FmXZYPmA&fTEzLt#zv^!IQJakD=@YcJs%%g*)z;qw+L2q)_c$xVxo-M8lj#ic|Z!z=d!h4!UX`tm6DbYf3Gb(J*j77G5Tg4&oQAH2$bS1s?< z;u{Z(r8h!w4Z&v%?EaY!dcot;+S&+{!FaTDx&LA@O;PMZOR|U9S5_53F^vQnto!XP z1B*SnxZx6ddUG%9&{tk^@!D6LpGSBi_2af+y&l7L$soK3ix6}Yc7o&3xT$ z-R8jMkWscCNDH(W<@D?_9C_TC!B0%$a(ED%-=chRRba(+^zs~0ibQjH==IXbL=Jt*`ioFQr$?+nkS>#U)v%@Yx z!;<0Ut2f3Ps_zph;p%>FMKJbqpw(;$xCyVnl@h;zVUHAw_uC!70KcT*1~sW5fk*O! z`#t!Y%>r)`qAS97mXM^>#ah?)ds|-9|o{3}C8P?l7*;e7rX1Lw)Tn+TUaxZfogRk1b6{ zf=-u)a?mrz{fvFx(+j!(b9GA7rK*N$DubKZqI2YX9;oexROV74(jHF9JiJ5o7V0^0 zSu$Q?x$X=zoLD{yQD8%nsi_N{Fnhh;RH~DLa(l9F^Vr5u`@B>EXl45GeT!$%f{ux* zBZ0G`{d6=l9#PD?Jv>hzCP|xa`%X&W(i1RrP(q?I zU$5fnn+&{u`VKZCm!yq292r~-gR}%}ah<}awBMZkHfc&bRuCQpnh(gyYRq#DO~;KC}5dpCh`LD0i|KSLc(R&f+QTLgVp^;cds~Km4qL=6n6> z%lPut*Ps6WbuS+bChmD@*=<%&zq-To<1E`w?&mVLAo>@RxN@z{?(^92v?J%YCZYSu zvOIeODd)z$;-=|V_Wl{wSDa6oK6g*c-bCqs_|=c|<`u=QaNWYk_p%C)`qSNo+-6Yv z@Xx!+_j~Ryo#mM4Gy8vtF&;_uv+Y|*_iXB}D$nmX%RejPTk8HMl*vg_`BFywKW6v1-N{oUbThr0p;PfyV7efk_x zs7dNe{P(=id)Eo+F!|PljwxYORiiEQb&729eum3)@NPuHI_E?@z`O8)r403L8##4g zu0XpU{g%@2WHLaa9-A6K>j-%UG8|S5oul}r3?7L-s1>F--y(4g;@QwKF zGRC}+bPkGtW$mn-DQ1ouYos0$ibG5(()USY_l5=B=c z1HIZ^!6OuU_FZfCP^>N zl|Kh}LQjY>y057!sIvAwWUHu&;74Qw7D>9*6`#?A!r}a@1za$*bSXCA=?7NFxkD=i zIyx)b1rt&1>R~zq2h4#+_4;i?ABqg@p#+6lhy$7L&*suxP>o%p(WXPnCwBRu8 z_QbHnkG>-Slo*0jf_s7bCJzkQUa;bNwxas-7fQ@W8AywsvJyO5q6)*U7IXwtA{<1z z*kH!H>MV#)J#xRwkYAQShE*bpXn?K07R0+P78mQF_D%6*URzcz5^MfJYr3ol>GXZ`y1x6ECs(gCyxQ9m4b*2= zS=`!4S5~`z{9G@{Wx2zv*Sk9B{J=kth z@K7VbzT76jd=CHk+O7X|HtV;*2jyK|ZB$zxqIkU-E#d~OUGWwod`c{8K_MSpVpD z0w^Av4g?$O;Lcb3-N+jW(r$k=G7!QOn>r|0XZ~uiZ15AJD9e$vG4@a~PqRdO*}doQ zg7}RK?6uE9Pt=8ibXA3*ooTCZsI)^tsC@M>KFKcUC^&c8p1ThA*A)wt8tCWBqXx?8O3GPjP#t$zny-uaxV5_x}$f7ue$f=`_`FpTo%^I zz6apwg)5g^tTWT#5yU-dahVvIGfYlylq^*}L$UaLyK2Ef(+dY4#RYiXBgw$;y68+z zb%f#yhv&WkSjIZ4iTuT6#TnSQNRUu~NGVO>+9Q1QYq?B%jE z9~1{dHTC8Da(vx@YI(nyiK=v0!4x0nwK!lps#Uaf>6yK~ik@bhn~NKJ;^x$$AT9E($JK&fIPS zmW0viXp5uA!7>0Y{>5*Fu}ElNUI6TtIO`D~m z)~Yj#+w+5mgCM1d2zVb$P;y>&TWB3-59YQus=$+YOaky?58CaTP$(uhSs=Y8JAW2! zh|C=IwDE(LB7HT&=ibA{FR1rp=$d&|0X@#B_o0#+9K*uc-5E7PpL2&`G*Pa>C`pjA zvyw9-&9QMSA3x~Ga!QF83Fk1pJ+H$&d=GCeynBt9XAVkdXr?EC&#jtb$SyBJeOy2` z5%l%!$S4gs@R<=ZOhNHG&X|Q~G=?FfiFQ*jP5TzeK$Lp^r z1=X{uO*(r=z=qazD#Q7tdX(GP(@@rz+{H#Q+-mmff+|hhy1iXE(Z1WaG#OSp5AigL zA|9PEGyGDp?gaqhm^pxNF_#iBUt2C$Zr`DS&`cByR;h;DHZzs!n2Z%C|Gq@(@V;JS z-Ro=%-qy3D%vt$38Vu8@NB)4TSspT(nDLA=dF z&-HspaAsatC|{0jXV6}fL~ozHI|x0hlL%#giAZgq=ErqQY*+*nO7=TrqDy4j>>X+S zdTK^s9%;(o(1Fr_J;T1^xOYb!qDLI4k2sDoswG9aaVMD-LgcLf&4N-!&vNo|B<4ysRgOfXK$w#xz4q@}+&;LC=kog)$G)Q{}2@1LZTL2u_q zC+BBfalzIoQLP7#P^JvajFV_oqE=T7&)c4Z`cypa9qUr|Iwr$4H-Kk^&KcH))9Sl` z+l1Ud?Cp^=sP4{yX@kiQ-D(^xf>crMZHHo#a_G&CWr)Vr0F8S_A*)btDAqLPyb=|V z?t+Kv5uF$$(FpX7oz@8?xa@6>xxR}V&OfdBfVDy}5iXz_1n|J`wgpp!?`XOnHPQ4iBvD@t1`lg$ z--IP_pB?!WlEz{0uLWXK*=Q-HaFUI3H2@#%M{aAkfcClwNx+vzz8szojN7 zjtO;ct|L~<2R{JzneNb8-u z>ezwCgjZUyc!M33!z`zD9#lt(s9o{nM8$eO>sOVE7~x4EEe6D>b779NUtJOEJ=u)) z{;Ex@dsbaQi=NSK&ZGR*a@noAtCr4zyJ2>4{u+uL&4ATIllvLG&=2FXdg4LwkugTj+r#Q3*l%iC`jqd@@84A{McZ}WUw!dRoTiZ)W z>5sfivp8*nGPOkWIP8xfwXfw z=^c`0s#9WdGT8S5Z*1apJc{sL=Vny($MrG1$8&8W@@r%Yf|!Wmu_RJ-oe)&(P3s#* zQBE1$PL<$(@lzrig&+k~R36*)s^4=>GknQLakIQQA%5`lmEIiHOi|AJ*S?-~Pzoge zLhBX9N)vsVCce3E&$A#1AOrcgbX&&=*J@Z)7B-J$y^|-vf$JRE@&U8Q*>_?>20}R9DdW&6DZs_AH?1Ah}I{0(?4NL_r} zqp3xlto!v`p_thD$mlMSW@BZ8O0reW--2soyLs(uvtwy5f7%kW zKGCOqoIR^0?l<5}(=VQYZ1%i!EPH-hS_4$aG8P&bq~=Z=>Y31LI$ZBWq+tcnz3DL@ z0OzR35~0x3Hwo0+_5AC*Ks+!WScGG!KjO>7*zO=|&N+&BpctP^E7fjL(|&m0MbxLd zg`LxLW~o>uQzz0TqxEt--B42hRtIH(ViIe8{2rwX`10#m>XV3v;AdN{=Q{rD|WkHbS3j`R+EEdJ$yLF+#Ko0xr#X;~yNY@z}Y_Wrz4&g#nBULSv6XJq(j;13&*1 z2>YAML>CPz1`%I_)1u8o!jpXo#@a<6Vks_|S`rk*yvH*WUqDD2*Y$Y;!M4uf6?hj@ zqF4&955vIif-s>|?v;02y2_+~PN?V1&y*d-u5vkP9_%e^9azbw0~Cat(@`ian|TJP ze$HI|@qPw78PBQ&9A7~#yfP0?-xH#@^%RL!`pla7#>wLOzX+L9QY^brzA;#8Wa_!Hd3`^F7mG(72PYO z%h0+m2f&K2@{5trQOOi7`xI~@&~vivj!Kwp26eFX8{&Ld%T^tIs`0o z+nU0g19m!ES5_y>;IDWEQQ}u%5RZmFVE7z8z(4?K(vAQQjp}iX7N8-Xv?q<$6r+!< zB5s&j2h{`duE_7$JxvT#6zN1Ci=F45qPlX0fLBBts!Ru<}!o71aRa`6i(!rVO5sQg4GiRH6gVTARSv1T7X27esL= zRPX!2^>v98M80dJVp$u{lkfg0vUgvHq^fXF_C8jfS@2Nh1V9|Z%*hJPD9 z)Q5ce+901R@4l(oVLLzBg`4#|H$P>4Jm4ekl1pd^v)3?03^tXaQ+9s6tiSpdd|$w= z6sQOJ`Q~6%GcSYba$w*cr5rP4>yfq{%J}lK&!CH1bLiQ{4MzETwzo)YNmz)%%0Evh z?nzH5PJPemsrj0D0IR6ERC|__(KMW98W;m7u+znVo_VjM(4GQ(n$iqb80}xrfiUwI zlc(k#{*EP!)xJ=FpH*S_C5Lfya~oj0xiA7Y*6%`}jnee9Rl&Phjg;Vuv(76Q=aPA% z?Xl>NV%}Kti-G_4Nr=7H6}zyV(t+cqUx)tF^Li1*v*UhCg*T4sJ@Q?Hu|K}9*VD(B zFAte}mybR9hZxGNND_HSq~iVNP_YbeNg4O-;IYgW%qh(L(UgH*UF#@PmQg|4(+AVU zqlplrs;h$y;U8N8pOBLxztZS?@7oL9MjSf|O6<=echWvI#|V#QCZ@=(P(B0g7gPk` z2WaK<1yb2qeIe2yAn-xdBU%xuB>J0BRftw(KgPxk>Q*8;*F*Z8$ z(%xZSt_ZUof4Y4PP`fFKMHPq?#TVfEbpL+8IO@GX#hM`>lQhkVBW206x2KpG1?6Ud z`4!3m^^5in(1;$WO5UU1;+mhWhnug+qxn@ zP)+xVxfx}$rwb$K`chvJx1c4JL%B6{e%P<~F9RojyMRw8RB7kl-JSBPH36jQ<|N!r=2P8JI+B<(2ujtc4Z^g;=JfQ}MCy}>`7 zg$gvy4;aGn0t86Fwma&d`2e|!fPo4P8gJBkmZRR0>N1*_y&&20`31Y;Kw6z<7cHVoZ%N^yO9Fg|A>es zW4i5uT`e{DTvcEel@o_4)wTi9=lPL1;dP+grGeD8^4J~Q0jCqGd9;#gq5NfJ-(0c!L)(M!Ex+ly*Rk4h-mV3atLU*et5to92KD z>J`3t+cU813~=1YDCLylz!30V>wF4$X=HQ+^AVCgL`fO~adVPme|b&Bjb1&l&JaGk z`Y_|5gHT;;6{9l{Ky0*#r(IBQ1mdy=>Pt?Wx~4E9iF`2S;fnHZHE@P|2k=cr=o~iC z20~h|(Fyg(&K|J1;gr2t;N*M+%Mds#u-UcAi2+ z2}&e}s{6d5ei&48-A6#>rm4h(at~JX^D4DV`dK$ya&8WSu#>JyQP>4EC8 zOHuOM!>9feH<4DpDKPxZ*)RHb{MvH-loG<-(Yio&l(!0&ZzrNfcP^g9X>9>utOtLq z&ack{ToaZs3 zNU3tlR!DF*JQFQ6Ba}EHY1&r`dKvHg@TpcoWR0E24D=F*Y*#G>O% z2bLX$kd5=z??LN5Zl3C!!H`KTCt4~0A z-CqOQ*i@4#Ub5lT5u0CI!SKt(qkNNgO`g*8CI|w<9p-m+`Yy$;+&7E)b52EmVx+I* zU<2AZrgD)63S4p?&Y9i!@Gd~fMs5A5?4avU6P^BmKPCpaI_qPAFt zrqGK>@VC-3)7UxOHRf?bZ<$`0Zxa^ig2LIF0wxKNH!JXh*V5i)`Bca`<}-Esn-^%Hf8FWI5 zg!3X0huF@8&!B^bcz9Jn{V2It`sW5ZZ!GhCX;`)*pKv8lN)$dmkVlz5xF=5e_Ns;CfTZ_+Eb=;QPBf^asTwoo#8Vc|>8KUrJNO zs@J%qNdbtpM7`Fk@^U-e;|5e=gBc?m#~tmxw@1d{};IYrLWUK>rA_x}m!aeUGZVFvInP==PI$ zHtUPrw%_P)rlnoL>ljr&#$!!o?E|#ieC6Ps67@Goz%wqt+&w&@cNALFKBV#VUO%BA z`v-(*YT!+|)^SlY_P*Gxaa`fW#y$(}b5?b+_=;}#u)hQs(q(?|efw~|3Ge6aoO)TU z7*%XUf{+z*Fwyk#8t2DBTqVen99^hK;2AC6;~j7ec>9gq3Bn}tHMXcndw@%{dMO$KH`rmwaA-H%U*Y3|$rxvQ*KYIJRF0paCjX4r0=`L@NN9}Bt@4l$y1xhXv_E+QXO2wJ-zyXd9dE|dB(g;slfT?PcI ze-UWkL45&&oYz zm%kV;!5!qitWt7aSWUf;6~9#S7;1G2jr#5)EK3dOjyaDFA^F1-!{=8Cx*ifC zLw=D(o?upG<$Zv>cV<3BfgZ*9bl!7|M~Q zie#ADdOGs8{q1in2C?h=nb0h|(T)eFzkXB4QUiyGq2C5=u@j)9iTL5r$OmbnTf6-~ z!sw;ZiJt^bfWIMBUdi!i9m($ae4Y{8 z5TWPC@~Q)F=7_f)fk!+z{`-H8QA#kr_Vh+8N-%K8b}m<=Wi6pz=in}+f6>+IYc&3( zzqM0%N3Kq88ER0DRF2E}eerE+iLGLyt4Zqt6(8zX+5YWdS8dm!~VQ>Cci6ZsEn(L^>eA zz$=j~{h&vCvk=Wlzc}t1rtd)&*)+bUS3V4Q3$}I4)P}GR7wFI3`;$w~WnVbr&$klr z1iGKEokG$rB**2tD1PI3}lP8*A#@*E`%aOZ$UwsfOw3Z7i1#Kz8FiBX_Md-ZO>g+qzbh< z$rY`M1NED!qo39y0^im`9x9}H`NBBtUVj#RhzODo;s$kzc7x}EiQI6`v%diC=TSmf zs+1t+<%Brqn*YJ8pSi6aoi1wa7y55o&2*f^n&$V?JwL$Ks7}`H7+Wlfb9M zYTO7TiLkxa5%25WB)LI}UhVORpzpEj-cW1RYCZWT0TQjE?X2;2Le2$&L6#AnFd0vd<~c zm$ZxvjTAx5OQXMH`xhT2AN5^D_!r~_O}py1tLxO2sSX z6CdOJ0ujr<9$b6bHe3?~0pIwfC>~#{&5P4-xG0Dxb0C5Br*~|0$X6J?ejOs4;!-~z zprr5@S&}rh#aw>#HfGpkS{|LqN6e~Rs=Fa-EXh&ssJ)^FbZpZ$ru34*U6K~gdGe@Kb}QtgXurs> zc=vBEt3T69!4{TCfl$22Df+>(jS39k7{j5sX;x4b*ZpP|INbksqc-WB}QL58;Y;8;!;{#$mt zo;)ty9Lgnd+ybOBRt)+CIa8x&+N^^1%4(sqr4iQRk3@&w8|Z^tO<5E$`2c*As(#q# zdGb{6YX;oqsUQ7zxc;j}F?zjkNrA zZCwMr@=q@Z(JUUFwQBUQok8KVS+Ah)DJkKNECxQOI*?jbw|LlNI}N&4PwmM~UnBVD zWg=pjC~hubaG`Mj^l-s;Zpd#=m-M{l$RRU4x_-C4mIK;*cD(prktZQUMC8gB`XOfx zIs37MS`I4FYe?ai*Q)3Utk|(?iQtZM9PYBR7hnTE7~R|6=Lg!@5msJ4;fu)un7e@_ z@P`eZboAN39%iQ_oc79V9D<_Hcnw(&5%a65l`nDu4dpd$bjZ1+xvR=|T7=JiZ)$n? zpt}aK2k8KkeK|A{L)`?Ak9!Z7Mj#h3gwFP4(R$^_C$cJt6Dwz#5L@73pjR2)?eapQ zpBTPD7l!H|e|L11w}GxU=8jI}8MG-tDgtX#W+Fbp}VbsPNGI{ zze~iI_Tg!vLl{*szob9(R%t{Yfx)W!E4@RNa;2Qow)z-&&cZYqgW0=$GzPXsaOA9$cxCN|nN6FAw$+ zTv_XjK)&!=6sw1N%UASJzazw(sMsy*_!U29!SV`hLMO)xbLh1u6Qx|Xl@593{R(wD z6N~)Dk&jzMu#zc$D5BWD)LIvFpoY=-*owlL#jSSUB*;Dkh@(eiev5a-vP` zHB%*vxvvlZ(@4loJti!uy7ToT#}jS5uLznzJnaH^)t=fCeDJZB(IrLP=W*XPn9$k} zc_uP-+OgV{xGg34$reMLmIOa+eD7Bb>X=!dMZlX&r@MieJ3?HRaozifb;Afi)9O{n zRNg{O>?lV?F~48mI-lv1EJ0%8MfK63suRb8=som3-yTEb>*M=Zp21Je-!Jt18t5)@ zRVH-B#yzp(Wx;O)r#{^L_}DIS661NXd|3zK@kOv~Hb zcR{3A6dJ@ulqb}N<8ED(xDT}!ke6LjM0s&XEOB)2oc`Y5rfV=O^$xjZ0#^h}GQWH9s@C=hJ8 z&0Oi~MOetM_-$u zj+8irzZV8wP{HSV`KdcWNKI=)-iBcj)*V`J8*sP$o#aD#c>bgVI-r{lJ`{9!D=i7{ zVLWaL3~xzt>1^4*Le2+zB5DgOu`AR^gb~d8h=hl=Ch8RE)p~EKJLtbidV{d| z53Rxbbz!nE5B;}LNBTo7U=ChT%gy|8hWB!=S^QODh$ME%gADl0nhPr_*!%aZ7c4s& z`saN{9DF;uB|q^FqaL1j!~l>X2an(&kNV=M(L|h0E?df_%FFrip_5@l{s!%Zagret8N-=@nGj#vhTPMJ z&fZ)>v%`&L<68*6lM#IV#G-eNKL~sM)<8&O?XX#}FJMCs>%Y3#h%l9v`#~IT*Qypa zdkfvneB;#{TO}ZAKe%|-ldE0w`S*LI}egO zl7n_s?YyOvcZScd{T>3Bq>J+#di@Q#+svUK%0)n)I&w7Yt!XrOZrC9K4aq&v{y-+q zK7tLEa+O8&g|&fa_*c=5MdXXK-X*NrJVG*RL(f^E_kKgYuy&xbmi^pqk%!z?cEgFg zH6Ko*Yrh`HlGc6WjZ99y$VXSXPPO21qB8}Eavz6s;YJLfp=T3p+BzT_DDIjM2ooJi zLs#C!YD}WY^BZ~Vs$U%6$+3s?bi40bg@2fi*T7iMCMDKauD#MUCz^GuX>PMA8pOKx z9?UPqhNd3!gI9!gbG<=8Ec;e!>3u%Xe-__^igmtsKA)P>Zp%K(uO=>v-K^VC$KODV zq_|%&kenW6IsFLvaQzt}Cq1 zI?#3y?mZn)Q>WiIk>=~IX0g28tj)UV$R_0sf_B#Td(waH?W`0o{wYZBczJ>tBY#?i zUCaF{$iq)nkHZ(&czMMRb2uM(!a55#&+v(%1bsFrh-$JRe4h&%Vg_lE40rSob#tRn z^lTZ8!!LNt;(FY>r@MmqeRO9|*aZV#D7{+*4%^ckB9gz6&sxAs=#nuN#C=~LbvKN2 z^+2r5RDW17c2zll;QL&#=YxMbdiFUutrbLq_gbp}tt%)`EV@!2f!j5T{J6BMve=+` z8}Myb``HQE;;OE0y7(fAJ?C~f99>O?xJ6x$wO%0~e=L+jcZ(67`QLHsXpaSM=qF0B zj;Ui^F6H7;fyJP;_0$S+Ju1NYdExCj5ULJQ$jamK?uU4)Db(ycX)yQAe9xjLY-pX? zDypKF=b*j+qI(Oqpp0X_g+bJp(l!XhQ(x!!;hAq&m=kK5pYv0^D_67HhEVZ}A zPx$ru%Im#{0%im&ABem9f1CrfFo;+DAxe)yzl3yR+}O2d#*!J$h<-9E%+mx;b$mkZ zlQv(=kMmt*$0x#M#T;~zTp7xeWO)-RW1Y>;qTw=k;I6`{Q)p5JBRJ~blM6YiD>D6AWE`IzhfSnc?E)at<>MNk3o=opUn84W(qRTe1=8Lp zYD54`<#=DBzn=A6mO!uZ`WmQY*XoSEPT4&k$knjbw)c>OHU-q!l-_f>)b}a0n~KNy zY~m(lFXXYi*Q3m)_e*YF$#tGJKi*yUB%IBl8xc|ArZmvvRuID zme}=N@Kr;A@i&ROiX2tpGT7YvSnGFr8xLz1_Vw?;IOvTi|EeE`fEa5Kwnyn8EmH_?cf-E}*ZO>OEtvTp03kx0|@y>rP0!ul0Zs zkZ;uvtQ$t_^}Xv-U++)DzBk_+C*+XciJ&hiT~jA^&Tr|J=gvt1t`4(nmZ3L}uQXIx zdxRPquk{F9k`d%RHOHGvVbem26e3OfL3%WwFXTioXw6vh(@YGWg64_bWNnPmpeIUS zo%9)U=vMsCod`PjDo#|Ag!7j~C=muE6f!Mhc^K0CfgB&y>1gwVeu83I0n^a_T=nB` zvfWnwue%kpEZ;yRo!hYwXzno%jVynx9-!~TkH=(0(D_O!JaU(9Zco#+(K6K|TkW4f z2M{zB8huJwv2Tjnncz|}h?>Q&&he_B?=dUGytv9uhp~UBS2~B#*Txix2#@9C-s+IQ z7+nacC?KhzKul615vFKK(1AmAtw`Jelm`4tx-+}OyVSB?0A z$OU3;*e^INa0@m62kwCWJHCGFJqg04xp_JUF}lnvy2GrIYSIrafqGD(?-;#Li|~$b ztEr28)IG5p>xA&$b_C6mQGToPg6}zHh)W{M{)#S$n(3fn5I{UBG@PPtzyG!Ow>}U- zH&R5;_7f+X=WLjf0{5I51RNKFfLI(U;c-zZk@Mv{g1hQmUUm>Cuo_vBDUpYmEqae= zNu7hljrm;il;}c@t*Q-_sXQ8Tm&U>H@T{~qEadaP0hopgIN88Vg>^;)~`@v?t?Mk~;L&&|kw8)wn*LKhw)5y<=e zhl1(B2R~W^efw85s2aiW>CM3TLz}O}XF>@=yl-m5XO342$A+3c{qIGhVjhUOP zO_h;Mxp5o6{K2xlANqdxJy!<$)mnWa-hRmLnqZ}N`CF@%ChFkj)QFb++8rAm9LHcd+UX6368lP?ph&e{j}$(BeVL@PHjN1wmp}3E99=1 z5N2*_&tZQ*LM8BgL=Uoh(XmGyi5Cl)pRFSlNE8U%6n(#mtYK@r^6&zccdhy{BY~1+fmc z;(43CQ}u`$>+m%6E!y`=h3_YoP2}BnM1D~+4-0ykS=Nup>!khKAWsm*3r>sU zMa4X$yNrOqWvC2vbYUNkS%FWNt-YeNWhc|*NBsfjU%e$G~**3+*>5cuX${At|Ida<9Cx0QZ2q76n-#E#2LYan;f+*TvyjrH?Iw)gT)U70dSA+ z?cOUb^7Notk~mLhaq+B2_NlJ~o!QS4gx~#G@JRtwE0+=(DHMNi#IVb2FU~S#lFByq=QF$vTu$ za9~CQ?&r(D7RSt%3``KJuM>ms)7mOJUGg|fi76w?PO20Rd8p9%yYmaeROptSK~v#x zF6LR5UzWW>bf@ZPH9OmhcI5D-}V?O5Xo zzPHUWud5K}7}eK>Nz;`SInED9EcD+et4?=}m3n2BV-?IlD=CEKrdu>4*mRpW0hd0N z;hxU1{W<*k2<@DSCIKmDi!Of=)Z$nr!s=r@YtIE{u0{^z*S6p9lDb;@g#4=vsXNX- zzNp>+4W;9gu@@Ky{7cmX%ri|LSCwOT?CX`iAe+S}DhvfRS7vZse7*F>Z>YA2&oHin z*e2)<q#IjiJo}hX-2I=UI{xbpfgRSpyw-ZSJCDceB!C1NKo+=GtkVdn{~?(}BtI z&lP&E;v7A54Dmd8f@0gsd3F_a%+LXkXNXPseO&jEt8S46J&5Uqc*|c72*4$a-NR5! z?i_yOAlDZV2!qcv%~Qqe_qh&t^C<0_B`Pw^o^pWv)FXQgD%hLLFLSy-oj1_Rdr;Kf zByk7we`;|1Dpu}j4;$gh&j|rH@3gqv)j}u73D)+o`%-i=jC-TMKcENv<)M1k*mEW@ z8|~A1J`_6yb^W-1KM?o%B&mD+z5FLM^?r2Ee+>IfV{1Dt3L-jvatDp!CT;cQ3iD^6 zcLQ+n*FwCRori%&@Am$DjMIw`5sVDIYpNNmNFtuRjzxH55T9OocUew_#tD+*2}O(N zaJYhLiR;Ki#_?Ft^7yxbrak8RiQ%s>{MW7YCA!CKrs*-w7HxoFJC( zFahnBHQg**sdLCcCp&nV~wefXscB0?W>!F6N{8E=wd zn^n;LQ{@b5AxY?;x+5b3`ID$OVA94LX1F~ldJ4MtN>ceekywHb!RcDt!4!tbmwZKc zOerd)s~GXO6tFfm%^eq=-F~=)>VMm!zc@YA=drg}K2Wg@{WOBOxe%;Z1HC$)kZra! zb$LJV$n!bSQ98YZqUXugUoWN?Mz6{KiMy|Q`5&;W5LOCng&t`C%z1zPUR}2>DU$!2h0wgDk1JyZXx?0VQYfA-+Ctj>lE)^P z#FawG<5;m)05yxm*b6zyb>?lN9vT_lZ?!kk@rZ1Z`eYQ)RP2)Ib8Qe9-TmuvTir4s zMy3jM(O}ZME8qpy0d*ZlfPvChQ6zlzj=)}K(;GvgJXT4gZ3;Id6E@Ifbg|ivm$w9VMSwil?ZTA{! zn_(D&rc73Jc*Fspdu%Hd%Ms!WoaoeY3q3B%gj^2jn4DNAh6cF47V*n;FDF9H z4|&4k3-G%rIC}|$U!x&31s*-d-z-0!;L@L+-0PMXF_;7O69&^KOxt|HjOIIHwcFnR z)^?mzflS{TSywwrD0bSQ=YDxI=}`u=s?j5tL!F|*OMsqb3 z6-Ik7&&Tn3MxTLt?kQ|Hy%TY~Y#~*NfTt=$c_r&V9dj{OkNrr*WFF|AMrU5xeX(BD zSY+U5`sQvwg_C+>l)U%Za-Im|Up}<5W8HHn*3W`EPKKm6{U5J?{Sa3W-@Nxu|MO?h z&;DoB|HNzl=QHU4um3RX*<7!uIsi?dy$535YwsYxAi=E65@z+hZlBaSzt-Eb&sy`& znkljud9Ry?+Q4$>;jx<5Ny1YKiXgD_T9#wGUXU%`xrTI|fK&aL)a@v| zn7bcdh%)kCm9O`JV`zcoK; zdL|kOprNx2njFUS87(J8FXlB=qm`5018wS;-|}~k4h6bEpvT50eF}BL3}wradUyuN zbR1q=irbo8F@k=W+k$7^@?vR5CWX6ysqa|D#ps?;_h!pxc(&(v>V-hRKz|8nT7w|Z z>0kvHh^=PL68$o)XcuARjlb9HS_8~U zkN4v21XT5qi6UUUuG(npYZ$-GB_r;}Z|4$-Y)lB7e%~L%L`1Va51W9%VH2H1JS1UA zwUDXh=_k5u^VxmxvG!qY(C+mM;Iq`-kbR|N(Cgo8t}XPqzv2n8F6f8sPSIDp3K7E% zIT^8s^?+^Zu}t&*PJCWGXH}y?`k%`gnYGJjRl1XVkD|duCR4!dO?cjC|0M;-E~75p zJ#5L>uyY&J{d{7hnQ@+bZ?~Ro<$8TxSF6-5vZ8X=tR?B?()O(VJmryv!hJSz%IFz) zp1tY*xTBYRvs13i5~Eo0Q&S7#N940kkLS}5;O%kgi?)StD+*7 zPut0J=(1-SBj|pb&kP6ZDs}X3`3HSWA zfq3o4IjVKXW9Myu=y`b{caUIZAsv_D?4CGUT5zd5>9BSi$0A%tzM}h|CblFY(1K3U z6>4Hs8DSLF(l!kHN!Hc5U)vzmy0pmVhg$c6zDfDHcJ^7Bk}%Rl6M&f2Z73=RbLG9` zJA{I2;yYR8$wfBoaXEho!DPhf{O^wW<*H>i0sdLQ^EsTlyLo!^*_UK+n8QqP#ucIIMHwXHXLdI_#fgWT89d zUur#31S~WWdLa5tV8nA#JsySeJ)Svi>A#JE&$$0E{xtY~ob|h&$V(E(i)#p46cV$B z9K>1XTLfpL{$iE?NH@LuyxgUliEH~W7Pejs`${zNln{JJJ`nwk7BO_4^Fg*csV>i$ zC07`JQRt1j_5z+-yqD3g@%Ze1PR)w>IUf)}>weJRy|5ZWl8?*#Au|L?0Qqs0lid>xt$pHJ_TZq1;Xq6GbBqE`7`|@ zg~#&7>tDef-}`>nG!s5yA0+GuJpY)9sYMW<4s=}ep28U8960yO!|KCXPCF^8qW3l? zjx2J&K$@6hTqD~K2h4kx3gDiYwIQC9p;J9|1b0sPN03X=>)GJXzf!Z3;jCKTW~+0RGl`Bxm;UoeLgjWIV9uycydJF*}v7TV`aZ@ zT5|^u15G7R-_4Tc1b_xootN}d@ozh834ZRw4r>ilgYtfps@9RY7zK&Fk0dQR)_ITb zv~>$M3o_qq>L|x@x*-6{)sW{uc*b=yGwZS@xAQBYxSQP0R$s6|Bdr}}g@Sv3R%-Lw zXWnY7F^b5XK&|>6nPQ8{UFqp`lmkLx0~P3}l4ls0NB7zoz!{m)USjNciRTrg-Ma?p z#154FgPfcZ7k~Fa+s=U9AjV}~9`e(R5J5fqRb`gThgy)AXr4a-i|pg`0e+XTg=5*x=?47x4Y~7<<%E?ke9^XDmsOGJF%bK? z0|I|T{WIVY_smD$k+F_Z)$k%Q(^iM;+&y`o2~)QUi7g1_#6I`4T%xl~l!#x;M`1P> zhEKu}VqMQ|>G@~1E&6~dcGX(*DK~Rb=D&aYD!kD{-0bV^uxsbao#=W{?+d*npaBri zBBmw^b9(;tuBrphAL7{gY`my(Zg0+?=a`+@e|sQr=S+38%uKZoneNIYFx{aGfz*SA;*u#ACX!d$Ic<=WfU#b@V zONDL>>YqpHV|*Yt+FiywXl+<8Aqm8dtya)bSAkGTaA6*7wEgbe#g5Bx~-SWlGWhwd|ugT zZZQlK^rKWtFtY+Q8&ZgOg}-cNowrXs)w7{i4z5wo`)8fB;(fly?~#`|b}_T>O)#y~ zdCzB1g?B-Ffk0`8_F{cOELDNgddGbRJDXC4)5=ZI)2vY+%Y!Y7!VhwtTfMSU2{Q{@ zsCl@TIA~CmV3!f_UY;*pd%z9mbOTSC+L%%6gDQ^aPUf_zV{*GRJBg)Ht z8ZKAKJg13S`H_@KkD**%*fiYBq9iw`Kg-T|{*Rp@zU@BWmWQ5)TgRc-2i;OYOnWOuA#!Ei2u?zs3jP+8^S0j<0>I{uN*+<|MmODHE3r*{&ftZEf&}-oC^cy z9r;a7o}b>C4C=MJZQ9AcTy>e68^m$nAI4E$d@mxB4(&S2hfeA>XBh53GZ;2eIIyB? zoNnap;a0|@wSUSMj<^?nNA0I%g>UmuJZu8i{9Y?`zx%qggYI^@CSQmxDyI|c6q_Sr z!JYjk4epvIFI-?o21t>vNgs5C7{COm7t!aCPXJ9rknwMdJ+4A<%4xMWyQ31LCcp9V zb@`%AY<%b;6I`Gp?(-+9dvDrsH3^J$>C)`vM8*NkVXE|$dn=W3SM zc^9trS-g##_`-+Nprz}}{aJM|goOopYK-XRk$$@)so$!_ddR^ia04mmt@f}ie1mmDMBCbfpR^?ZCjt*EW03^g+S(cZdn3q9uU5=^K-VLaesCEfp#4&hW6kijv zt=BigHz!EFfe#GqPYAghm=z}(M)@HKzniwdu;CiL&XEQ)G~N;AdK`Xd#}Z-m=_fJ^ zLnp5VX4vQ?f4U#e6JLE1L#x=T@o4${=JZXS!{OL@=UK?e7TrzR%P6VStz3TIQpr26 z8us4wnRaBS+yoyJ2IHo7-@YN=HO^W5RuqhTIDXbNZ9U1=MTBQ~7~h$0Fhc-(N#E#P znk$CmvRQz<9d zz*P+Y1cps;T68nMC;T$ZPsLOf=360Krn_|Rd?`$P6;py2r__4-8KPOQ!y7+I*adRx zPT3^Lu7(z}WiG^RLEp!NSk`w8doG5dyRH%)!B-xh(Mf_n?FNGRn@w`H&t;hCcvYlA z1DF1_ZJXlLpFJabf0z?YNHe~!5elO-c0|lUp&22vLEjhjyq0WBH4X7V2$s6Jw@+8+ zkiGZtZ*#zuTXH{#b>g(gal+5S$M%DY38oR3jp#0@_H{Xk-{6A>25^b!AhUm#)#+OaFlFoIs? z(sVWpr+M+OHj$Z<_Qg8E>2p%$uY%z&)Fy2|>95xwLg%3WGU%s0^i12kE@FH54RVCQ z%Ktm=%)Fiu-xjgi-*u<=zTFjJi0+uSCzw@67_9s>D`Y1@Nyu8$wpD4&k9+?&SW9Pk z&$!7$9I<`cIHcZ} z$=JUL{_!P)3!yWGo+O;*pZKC)NwZIq@{>!UTI}(l%ZCfOtCRDwp^$~G8MWQ{W$z2@YjfR$UM-V5*`TcypZH@FW`>Cg*Tg0C_aXFb= zMC@(#=d;S!{q~{@smY*b!@}E3cKo{E;(_chug_Y$-Az_-JcK2`-k!+Erf&~=q^V<` z-cioEc8(V`=w6H}^vOYM%|5>)5-{xZy7#xwVHv8Crg6Loc?`Q3$k$r>@!7qX&o9=y z!iJd-8`N@&4V$ie+sxjx4|5t0&&e!krpqusU6TXCKTlhJ+)isRn{MTbtv)xU#O zhrsXijz66G+BN+^U-zpZ1lw^gR3l1vsIxRnhE|Y)?CB|enw{sh-#So313TkWWhUn- zWDEkQ8`F`sqrP!w_4|Yo$<2JKQ(D6KMfkh%!Sc-G_J_5e{Qd;nm42+ssvqxO<+oeo zoxDbmXZb~K8>P{q>qa-~<(}GPd^vRVyrCn7J(9+IJ@4p%jR;gRLc*;1^IBW%{iipx z!yDz9F^Awmv$af|FRZ3viTjl?wxT2LDIHr$NCL4QTzcWj>OMbL*=mlB1(Vp)FVVC_Y95JhZlqjaW9&5)%lebj zzTq|Dm1mP`GVD)lo%Zrkj@QiJjE?g!-$}VKfm+b>dfoJAv^m>sX_rrPN~rB2pJ>F> z%l(d@G%LQGtA#YU;AC_$#Pdb~%RH zGn4YHSZhrho=J-U>aZP=eK3#Z>wY%x@*uthoOX3}Ll1knMTgob++ie}(2QzHEqeUt zkd6~%sz2BX)_Zlz+=(@Okx?i|0%`6SpS57fL{%`R|=0zg0C-?dxV3wT$C7lv8JuwhfE}-G8Ywhl()q ze6@x8dJh1L9rO7fVf;{98P8KnFxT5ucpnQL$aK+(qWA?8o83Yo=Wnu3_Qz1_DdkHp z?1w-ITyXUTW-l(etAO0X8^r09`+V=UHsU{DLT67ze2ta#p~NX~T4ycnrRu|B$QhF4 zL&znMON93Lxz_=%CU)UIv*hs63xSjT7xkhz<9H<3GtS7>lgZE6{M)RzSXj4uQoN(Y zQG=K|e5w|ffH!KK7Wc1k;RyRa{k}I}KaHYmlaRrZT;U-V8=!;BvM|d_7wp z1N$kvBCt5B)JFNSeH}YvS#HtQYc=GuCA}Vs`96~do^=TQ7*m(jBwFpS+|jnUI^JEZ ze=!>x{P7zeHg8RnE=FXgrj%-qwX3kHPG(cFyND(teMV1GauZQDITpwgO^oa6a$?}x zV?cNLu>2C6cyK>O#EMN&v;-f$Ny{f|1n}h)ZAj zZZeQdPe*=up)Z`~sJQ$de>YA$3Bx1(HYE0^Rl3iVTs^zol+J$+uFmPDj+w(xOM;E{ zHuIW`U2*-0Z1G-&8)urUV3;$1dc^%({v3v7NZx|3g_7Ch7=Qj0<$QfrF&OJ0dyEtn zWYed0=)&?O^++4y=e7vD#S@*L|NXl+m*;I}cbx?I)exVoAyMUo-UKZR; z9@{%LmT&SA*(S;H5&2y`j(Mpuy1OHlIOm41{l4tg_;U|y?%K$0`3$a@6S0aB@QJuI zc!>Wp>6rHVFS%)_AWUmwu{YZZ+D*ha=2eW>Z(@w$LgVp9SNS8TSz+}H9-q=TeF_Kp zNVDeL(4zXuZN9nl-&_6}MG}8_eI>Ljdr<1PYRV@|T`0#HaGWdpv$Zrf?bnfYZ=7fT z4VH-N+M7qG5s{YlTeuqY!AI$h=O>De=s!B{RYKY-?1ZmFMjNkm!UtXG$ zsKit3#g7F;Y|@pVR9z?N_v(B0dP&{%adS0}W1lep3oY80cxNrvJ)3xQ@MEinc(N`G zgN5>mK{wQGZXcWZ@)EZWufe&?TSvS&l=AGg^o3MQ9?@Y(Ng(xy&m-2Yqrqe^{E8= zbV;{ry?g0&E*O;QOAIp;NODI+ShJ1IUH6<}R!vX!SaOG}=g+9`+Fii!AgImq)(n>6dq)=v}fV-+i+f~v*h7mI#gx2M_L3?a^Z zF-fN)i*6s>X~tgDKlG2QZw7mIHH5iDbS_>-h~9%^=8*}=g%Jl$BDM5pIpM>iXk=I0 z{(W#GTb*RT_eLs<@nCCQgLt?p=6p<|GUXt7{4bu)W6Mn?ilQIHfSiU%h@3Ob$T?@e z{?Q)1iM_m9ZHcN|=k6__c+wXsR@a=LlAqqn8D1U{j3W0_N9uh%{TdKpYajpn*Hb0` ziO$9QlxKUeod*Q&UdA`G6@B+{_))!)A{_s2K-ZiZCBGBJlw1=ZMHMj}#Fydrxw24s zh1anZR3Wm1#x$_cKb7hFLWRGJdv91Kf!0JA-!@}q4@3|??nmkJlFIau`q3A)&b#B{ zVz=-Ei*fl{D~b*}pDx1=?9W(BT7vmh59ynk1 z6-%$yz6$3F#`F?%kx#G0mfNBtT6XgTLF+g#SJNXF~auI+2IbRIYnh$OJ2`}+*q zbw}f~pw~?J8Rc-9vk?7|+Nov3coo{Fe_KD@mmH`vgtdvn9T-g6fiTX=bw%yW}o)v9vBmcu*mHAAB7N&$c@{e{ppPsr#Xi_rD#ZbToD^8MKiyD-OLm zDKxB)Y%tl8l)ieB$FA)-N6yqL7Gf^xc%Y_q$Mz~G3 zCB!A`&M$ns4qEHxQAe#Dq|w3eCy~VWytwi_f}UmV)4up=OGr(NqNxbn!pT}6LWW>F zPcAjDFPJ0f%d{pI3xxCbhFV{X!ndi2&-P|GKyauj4!fhyawXmz zRO3O40g^zmg^%Jhu`)>eg|Sc**+ly%Xh+KxC(D^|igQ0WYN=T{(W5DK(iiNlt4Agwnd8G+&m==TTs{S~s#X z%}o93u-hdDr=tqd*I+O`!y^^>DmreX46a>UgQR{|bFy0IoMhKR3Os-)Aioyx?bDd6 zIO3drX`Mru0|E*CCHE}cP$kO^ZH^tqPatlHT;Jc!z>-H&i9HUhTF(JLc)ZQcV3zF|E`T_ej#g=1fa)c(T>vteg3wfIftk+??DtO0a|xkjLC@!U;2I|d`(#%FG;(N|>(!__$g;9Bw} z^pOU!=%g(sR9E51Uy+uxnlxw|SD5<$K8Y3$;sc~QtOk!X3-N5b(qm9^qM;fBySKPE z4KgBNqeY3zt#fREnv-Vc#Qh9A0j#sT0g;)(0PKOQ02FQNNZPU?RUdAEW%(6Z&{Sof zjVdzl*cryXkQ-J)=*n!kaH9cTdY2}w%M^XG1}9`r_~4gx1(B`e))Y~QuwuPM4%GI$ zr|!-Tac8|2lj>LMYa9Mu=(F_q0u-g#&UB_{tJf|f$9m0&Oi{w6rYjCG8^5vK6b=6b z~e4`3J#@fOPlrjB`aoVwTExqlir zIn<||4nSDkAjeCD7=k&`tRh$qg|cgG2-LSduFOnx5B?zaIZ8!%K|uw06@S~}l-OsYCT0Rs?}T!wed zua<4o0$*8}x$ZZYM^y~|r0XA{UmIx;$M0z5TncdgtJ`JuUHhyucU;Y}WK~$Epg3v?<8kHPXdNn) zAlI~JXJ#RxZXkZ`=*JWJ_SU7S05 zLyIf0n!o? zP3yUxJwKF7t|XhX_Ho+?98Zl;{Ge@1{SfJp4`jeaRmofWcB^(!uV3E^4WK6x8UtTC z?R6DX^`<(TXFT>oNw_SIC`M|tvMwM@ zO}Gt%zzUT4i!vqf2k=MF5l80P_ZI{#;w#!}C5?o(znulwX1((u#O#4kY;8}i@{c6B zfMcGqwIi$i<@$(Ha|rYH$amP5sDggLL4FUsJua5ePvI=MD>MX=@&&McC*s}ays95s zX#wek3P=cnndB@f)80 zlV|4EBB+XL*ZJ*d?Qf4LnX9dSs6k5nQO z9}enQ8a@w_fz}K{h}0yI3{x8Ms7qLboJ)Rb`<(`nYA!X7`zlpgZrjdp=+IxErV6xG z;j4(HxnXN?L)mVBqD#WwI03P@mc#tdYrEp0viVXt$IE`C$Y}!gPCCIn>O&lx_B$yV z`;H07Dkuno%ME->DZ#Y`Mf;8Hg9;O?qESwzOx`wEWN*|>nNfJx8feuSn<_XQgk^(# zkoM0J>E3SyWG&XwqOTHQ&%q{i2is%n+|ndy^yot^sWdufC2AiKMY_LdMJP^9m(a#Z znIUZ_va^j=H1~y46wT+K4@~x%HNk#otxYXOeeyP2Klv<}c5Bw49^{!zvA_d)(`WFV zO$6VD9@bZl4UNqRvSReJe%R*k%HKV@4T#(>9#-EAAwG1xEDC~k2bcgT#W?(VxAB)5 z)imzV@^!R)UvnpT5J6z_nU}A}0lR+KXvV_JZ2$#7l@X7->^LL1`2tS0fKhR$tE&*e zJ0720$CnWx#QJX+`@g$mvTO}(Is1+5prHcF?CgqljgY5YOlf$wf2xG{ZHo#lVE&M? zxdYMT##G;;i|}a~jQC=HRGGi${hK>%H(#FDN_~&Sz(_yVFos`2QHH_=fA;2a3 zv*h2y#2KGG5y70-i+&J9`KkU)5eIhcCP{kn;oo{UVr^7Muh-Hs<^^xR^O^^yNyI>d zq=$OIkANp?{6d>(8rKS-7^tGTnjn`iRW%1j`s^_w&=O8(y#Om6W^>(@MXJNKfkLdD z!vGwYf!okvtzYt2jAnU83V`@&ZB$c?0R>#t8U0TTH2fe;uuyCu_5?s$4mE@xIQ+F* z^T$m@o{LoMgvvk`H%AW}ec}j^5ph?pmyl3R>I)@y__=`VLITl+ zbS~Xa6M$pGlrn;^6hK{wRRDoOM9Xca!z3;oF379rYXLtq{{(aQbKsyn^kOenKKQi8 zen+405MV}Ptj3#hB_>V_LqSxLE7eWH43ohl)@2&N+=5FE?}F1k3GS>ct&{$irUXu{-c&o3&35eUX_#3AVea(S#WUVs=T4jVusZcA-eOGcchj_-&P}?w(kVqsnT? zVo;D=%LOat^jD(D$X1AU*x@U$kpwr@aik^={#xzX)MMfM-lU zJix01c#I&@bCOZCPk=WowRI5NHglq7egN&7}VQdg3j54 z`qIgCLN7Co860G*T>)XKnVE~P8gVjCjkwTz`Cr5)`0e+pg2%i7WO9mE*74JMeG7QNE&)N1NOh*g+Tc>a1 zWZnY?Mk`>EZU|GmpxHS)hCFc}Pfw4*k5}>~e#d?;Rj?U1U99Wn%5QV{6IpFx8X%!L zNe!RsXgoG&LR@Xu>XRv(=@5 zL}Ebm7bqLTkkdPLi2k0j)1+7M)W~KMemjI1R~j6iBc4GsaI%3MC0L5=S;n?^w?;9(i(X2VIjslXmd}X9#Rtjr1EFM_)X< zZq_c#ss&ZIPj%0Fp}(1PHC~}Tt1JE6le_9v`D?e^@6(&4j|_@4eOpHTk^Z|t91Q}? za7VreZh%FR!_;xf%x4}ZPFdk{XcYYG5jH9n@c~8NfT=af2Z+U9Oh|M>kJY*V%Qwi!w7=5L84LU?m327nhw!7IO1+@9 zZf*F72@;>%s;KX^BMAwwvRXm8<3_u!%4NRtTzl{D5%vZj@jyCQ_j}prceOOxB?(I< zC($ib2)YCAb9~6>Z0DTW>`*8H4mSBtBRG@>wbbsTM;nmZ20$p%x20=n1f0QMTuj!r zaB1QLZ1Ihh$)ek%CPa4SN%_3c2e%~CP$X?Uw>N%8tK;?TkxHe)6SD)c!Hhaeu?7R= z93wpb;0+cF?6UyaXfj9%C|ZeBw>n9kOZAsY?xAc;?Wiw zBuLiXb7q_Aqp zo`e@=2h}=R3p?T4^1a%n8SR!@SmNDY)gN2`bJCEfrw5qz8RU@%gfHeA2v&LXSA-4eIW5C>-*08UR7C?7ca$lHE7cPInVmAnbCGru1vA}ia+-8rh?=SuNcU0{V$v;3hgR#8KEr|(ra=Kl%>o|yJgvq` zf?4z_wSXi{Gk13GCf`4UldOt2SP9Phh88dR4UN3=0}+$0dc=&Ss}eczBM5Zu*U3Xe z?wwIo_uC9WkAccHdo`#I8NLo3n*V+W8wuzaDQ6+)>{8N8ZOd{{p2j;y*X<_#L?H$A zq-3Q}l#W;>BEv-e*@r%FT6a1945x0i~%S4>J%=u(kYP%V*Bzpst)a!@R5{|~Q>2Kd9 zOZWp-M!j#fAiXvvMI;z#PiAE~zn0`>OVuss%xN>xVakF8`!I%vfYJ5$O+Xs30;g|5 zg^sclz=Q3Q0|E1c#j*`y9oXgDI&eZBIsYkFtSXc`qL0_mEoKuVaD&)rr@NxZYK)6c)bL}+HB zwQ4HWX9!71VUc0E{ga?CEt?Rs4s%wEp_#K)+NOXEXP&K#TQuakeyZL6Gy7PJ+m`tB zi%-qkvUmz46EC;DS=-&Boy#PdXHgC)1bS-;@Y+lKL$-FtparNn2z*QLf#>jzA#j{D z+`k}Z7pywE>DFz;&Z+n7JtloOiTOQR43P#qacyO``shuVrA~1p+><)*+o&1``+l3G z$W=;b&2{mBj=YNi18OBFNBI|8Nnb94}W%P7*}g z7}16DTLUeBE-=^2zx}IO3}8p5<)OR3RA8^&S3|yv1f8V%uuLoAYMx?hF~Z@617=G9 zX5;VsEQT13P``_i9&i{Uf!(--X4#&OlUMz%&GLtLI>LWs{GpJjgz!|+p^-r61?4m> z2f`wjkmPdl+55`5=wKV|Ms<7{H`w4uQG{5DN)7R{!&1`)d7|37T%kg@_S>PLp(P_L zMF30gFi(4}E7UiVX?1~b!cq^JkHDXM`mT#h)3S;xvf`;Q-D}JdgZRc|s2OmkYFJYP zr>f+(PcWTa&jQKX>qNr10HQq#Oy$TsU2;@x#;S2uD|<>e4KpjIaPtUFrGSzYo|7Z? z{E3%AR~h-p>C;qn^znd-HwvtQPa9DkFj%9FibNuz6yIonxftPE?@@EHk9xg*|J?~T zI-*{OLG|0Kr^W-MFb70}fN(yw*J;SmA#L7nBC+1UI=%VUME$5hyyaNms~cR0MLe%s4>>yrkHS(a@vXYL;|8DE1t30%O?;EEkAS!-mSR*d)x4(J z`3|BqNncRzB`%yA?|eoBWY?1<8cr^Tc3rlm^|QX0Wja6N@;-7A^YVAfIsK|HO#Y6e zC^MUAeHA3Aq`zlDlQx1lGEiVx#4^3fh*ArFHyRy!G3p;2Kj{xuucU$^Nd06TWm`D2 zHU)#G?9YTKn-=aXOCyefLwG@Eei61V9=JK%A_1ZBR(abZ)(UH5C&FRU}udD`3 z@T_%{pjoXF87(K}_oBt87^1!iq1S^~lk??1Vd}p}+*H~^*_h}?ze$PE`grwhimlpd z53Z9W`Oe;hQ;GbOvgTO2pXWi87;!q?TB0g;sXaFt0ugsbm_~f3`R-v3ee{Tt&fcbf z&SJwaZ^Kx{O}QJsghqiACyIS|>LFHw9r=%PN(M}3Z>^dT{XHajStRR&o$+q4I>u$% z5vM#6BCk?W4|l@7u@Gxx%otyAXQt*g9FGbNxYoja=-1tzlugP6{)nBN3>>J8XdQv* z@aE=VBtF3 zI9K|zTZQ^$^-dAmZPpz1*z=S0sme}yK^XTMW*FroVsILy*QSL<`lY!Zi0{ty^_sUu zd=^5>ZHTby?5U|uw=<+fe}I5}O@`!K@CO;%E93=aNqI4GiRFxy-x5<+sBFaJ7Y+ZI zNl}zr($`sTxaj_#=Wnm9+>~iJo=)-o9Ef63g5bu%=YjXn1nq1y>yeyZq(s~PNp^ml zM1kYR)?i)}MebO1ny3zb_jHw;pqaERL;WT_$TFf0YsK2bkd7qD35<~VdX}iU$Ik`S zGVvET66Ur)drzaZ_>AZ^IEbRAM)LYRzmX#a#kiy+0@8p#;%K}<#powFX$q}7_};?h z6ZJ_%clP-Kmog|+jlkvBJ8UF zot$Beok9a-*VLOY(?bD5EO7|!c%`lv%-}<%+;LYnMOK%Wy1PiB$@@tJY#fjlmy&YU ziW`+aKC1n?&kr1Gpcv$E`L2`kYUs<|ee|dq#wRd&P|rl0;PmPWoPEta>sDbat0rA0 z#dBcS#Pva)1X9ZlyPx(Gf4$M3;GWg1cJzm3Lpu|Ulfo#8|-e>+Ax)*?-wygA#kMWm?NIc*Jn2kxY7fP`{|#Rh`#3# zMBimep?5j*WANe1MqL}ZrF_Wq78fU~J3{=mXNij{Iobf#xx7XL12gAX*qUZ$n7^q< z?89p{I%!)pj=>)M3^k6U@R50k{pEQnXC|rurU(B3vExq7w^7ST=NpA4jW7`J241+l zyR$I6Z|$3b>vA~Zz3OjX-08PvlaK+w1$T)LNtwRn$1@(bxidfQhejmrdrTH(&_4WK zh!~KWspYEa^;zE@I{I6mEb4cioj6uvvsGYmci(SkR1k!Tb)v#VxUPLWaRSg3r`ax& zbTFzRj(<{1Gggg$d9Qn(fe202X<@M5nMQxw1R0lBS)hpsTJ|U*HnzypT6TV@{NDU6 z#_;8;v(*tO7N3^{nUvbfPE3ap6dv(rj*`BDfQ&c2k>8i_RUQ~nY5uSelT^QtExyd3 za}bI-E3}0LzkgUfEG}8`VoqY<6xe0@sZ7u=sBypT5KwTOyRP_p*{Neb+~=1*yLzo4 zM^ZD2Xe5M`P{tptNfvwu;v@iqycy@6OO(Oj%5-j&d={C#6jQ&E>c4lal)RGbl;#%! zT7USof|!rP*gg63`d2X4gMsMoH$1oyjD6&}ne7K|OBWBW6*Z|}!brGwSzYQ!Xt7`Z zv~Snw)K*Fz^Br@7HDkMd*BuCzDpH3^ zw@pF{sw6GLB=ReA)aSrX(EuT-CsT5;zv!~a17{oI(*2^dHf$aJCJa^xEPl1{?v8qo z5l4kry*IQ#0~H^vqf#j?k9y0IV|@=+-DCxtIue5)yWW(rS8A&X+eD8(SE*oCQS>K> z6>wUc^F}v63R(>&`&Y}B6PyyoGA1TtdgD`cpAI~BwW%R(4`k6Ln;_{!fMpu{({OE{ zm+ohK5}t2zbCVzXqeufNVXi+GkYy*IM3|+2+97o0PE;(91-|Tdh?O~8 zC-rxZ|Fnvhrq*&O@7AbJ4v)ffGBy0R9ZbjGH?=9gck;8)B^v{uOEMKIiS-KTQaeKO0=p?OLaD%$&~|>Sg#XL;a+= zPeU0v`27iknFPD=a**m{Z{S|3COf{z%GXw^+8XyiT#plUJH%(?D0&g;rHt-_!j$jk z?`*ayD2J!H87ucaz{mkP`U~uZuw7e`^z9NXmbSp|;xRtzkuFR5>(6_Kx`+L7$5kM- zcUI{0B@i9dHE7*1^8qiQd`njCp@oMh3R@*kt*>#9wUchuBSGpWJ^DL=bM=;GtK%fq z{eaEY=sCfi#avT=@~wQXk;^N6zCS3aA=y8^tV4L98@CXN0r#~N(nS1`);;vz9wvN2 zVw(c2l&3F;v+=Ulxv`{s^pGI4`I(lVbP*=DgI)FCDgDFR$ThRDr=Wfb@cYm!Z5v9= zIX5PX1NGx+Vo^}+T2Oo{dKKWIAK?a*&X-v)98F-%933Cw`Yjc-pL=i4=t81nw+L-* z0<%ZblVNuL^8Z1nu1+bp_!&N2AB|&NU^&t75N8V0Ilw?G@Hh(ceyU9&J&L3_q4iQI z^U&x7N)*o;^9&YW{l2gE%tEZ&PauZZXwvvm#(gTVZple^R4P8i-`7cS!-qV4U_VRP zYobslcLil}`eqX^?dti}dWbfUn@}9{J*>BSI2BUVr=5g!{5(r)b4lxjFQKa{XoNgM z!6Y*~Uqe!;Gk7gd{!P5Tf;%|jJm4?;1xorbUpq2}*ahKXUA@_%f5cFaqF_S-DImGs zvWZ<%Y07RPJM1y*B?^K%B_9Q3nh6mte||(8(y8)|`=q0pnCyQoj(-fzUa<7c!Nt_{ zbp4^>F=jpf#k_=mvhiy6jJ%a>E5=5uMlEBSE9|dUgd!yA%BucRyUizZRS&;N_1*j- zO)X;kV1@BXSDVcElJ$(pmJW?dXyZ%DVhv1(0(VyXNB^N`y;bw_@dOWyvjgJ#3v)6! zevu^$g--xo+5My48T|^G85Eal*3RWl#4!mj`sbDHBtAA&X)*A_rK@%aNwL-Tt$DPl zInPQpWl06a#CE$pX(K9aLOaA(qzO*g8TvB}{u6W=EnNqKwA}WAph=;iGNM|$*zY9@ z`qy?H9`%3ZtKsF_c3ymG7DHf-M-N!%eoEq3FJQ6mzFBPnI08lkV!8=HMfyb!`0kFS z%ajmD?)w?RqUr#2Y^k$0mSJ}nNn8cR*J-AiPxn0z^heN-30}s<%)&YzCkiee>1Ejj z8KQm#0Q?)VUk75yN^cfV6(l{WKvoePEI%jDQatWbsa!mH-+e5(S-mpWe(806;2ty5 zurjVb46r@kJ|5jdUMK0Q10Ke@8T(}@U>?tq`%OfKs(Y}sF`fIB*;~NH&>?S+o$^^8 z-VYxrik%CK!IF-&4-K4)$fI5u`+?;%wRQjRmQrMK8&X{lNa%=m;tpJ6Cr?raXLgpS zX_EL6sgS1~BkcB~wYUh7R&C}cSStGfz4;KS$HhY@Br)2>yQHTE!0?^;!1*n2{Es+m zEj1GOf;PjfDkxTpT&{fi=~U3xN*o8kU<8-gN!6%#czvBdbX%z_<~`1$KA!>Wrgn+}+I^oQ^=iJ8^ zRkbsXK2yU^+d~)Uby?>ZW|r_&b4FZ!*Th znCx4Rsaa{_Q@`tDLY7rXL7CGVaW{gc)P3Zq%GVeZrd@#3@| zT~@v|bO!Yjo*mg2{h^%Cy0P#{%C9{?Y!JSnzyb@cwtJFCC^2+dufzeZFp8fA z;|~jC-vzd5QjeJsbm__)nFcU^Xd5!l2S5p5F>XRHr>2AC(rCC1@Zux0G^qW*7Wjn9 z#kt-<15m%{Qi5CQD0?;%lzy(B_26=dq7__`3%kj5W0EB1jK`FHt|~wgDJJOmfWhnF z?GU~e^Ze)+_QB69E(p9Rk5@4AI@uc?1$6ME9ufg8Bur_tP5kYQ3w-aK?z;FREYY-7 zxi9PrDVY5aXzC~5M-BF-k2}q#Xa-g{>y-0<^4k%XrTGUqtp}h2l!@ULgvr9Vgb^wSyKALWH@$pMMEuNW3 zSM8&uzH}*jp2D~$#WX3vKqLqp*%j6>Wt=0V4^{g;ZD@Cr!}2a?ec>sKtCjZd>PN6+ z8WcA25L8K@MM-Zto+|SJ9Nvg-@QSwk6*;C{rrm4q)tOv zUsAtQ&d8NWcyj9FRS16BiGLCH1q@8wWB7m=vcQNUHTX?WQpw9%;s?-xk=A z##1`ODN!BsT@6WKaX#4x%Y6~%ze~1D0SL$I6CqB8l47O61cqfm&A_Y6xzH%@2}nCT zX-|=z>#mWLC}7gyS}>m~-clL%Zi=eP=d-^W_T5sI5sl@JmWL) z) zNr(4R1pVEPr@ZexeZj=YW_4tx?z23Jn@ z>0Dr1fE9O5v1igsbTMQiBff8+)Ho&Ey_nS?*lHzyRP)%4?{FC!r__A=O1O~OLw4tQ z?dAwKV=)=F@Jc}hoXqEW;6S}lEWd~~Km8ajucla#*@fJ}0*$jD3o8a4Ni|{sH|a5y#T^x6p}A|JgZ*c5l_3zf5qZZL zb{d{OGzNZhUHq{v;j!vE>1CDP?_H~LIu*mtl-AQd9Zp*;U}JNKuHY; z)dQ&$sDxmjqJm0Io1(S#DZ-z`R3n&;&29bV@eAy)BdEd;)TI0MTky11KqIu6ytl}) z-AWvcak(yLcoCQQ6@}MrrrOJ&V7t5D(pkJFj`zzg;X$M1EIPOXh;Q64&v@GZ{1ZAQHgf zwyFbI?b8l>P;iIy#FQuS+jSV@RB?|6(HyqbV+bEF4^Ff>4{1wWO3)?<9YdWOhZdx$%8*v z^tnqxY`V12&92l^eM_K$k@%~fI9hbJ&lZTDc&>gFjZiV`NIM-!hn1)tygV zabI{+25#K=f%-&PQqLO_Gr1H}hMK-^D=9gndTgEmK#3f&tceksh*di(uM^HEx~j!1cXq4(@Znk$n1wXXZK4==t}l)S&pd&KW;=b5=2U< zvVPIK%D?HQ%9tNqB=!SxqQFuXv{pZc7hi^C6iNgNh-dycsa|(xS@;}6N}SEgL#$vQ z*+gr1+Ol^lkS>X}wXD^&RDa&IT_j%%ODL$hc5o1OXEeBgsq5qsE|H7zHW@)$7(0n(e7Q?`2 zqz;#akNi6t#Yd-kV#9acFPPgivbTQwSa5AsY~@I{r)qP0n=_`~(bHEjLU+`4fXEj4 z*GBid#-!%kM}#UTaRF`;UC{W`(XX^$D))AZK{g)#SJ9bj;R)h$1&LIQ14D$?CRT|4@?{Rq1Is-4-cr@%Y*hs}$cQdW zr)=I$mx_k9{l4s`UKTOVPkVU97F%&x4~7eG^sq=dit~r$jJZmatkhFF`z|_e%GkF- zw9%UNH$EMt4i49S-zh|sNs;WA*LkvemaiY7pTFJ=eSr&!>?|@$7%2!3&^3Ux1ym15 ziM0Xt7Xa+|nK2FiveISdHcjly8jiPu(tBe?F)K#wc(yVGltUkV4TF;D0`SP|&7pB+NHTMZZKtD;4DU;pDC;>X zdEp!4{UpPAkN&Bs$86A&Q&gzb-z^AIvV3${iKlw4lJjB zjiIy6oFAm?xfZX<%follrBcza*^OXtvbYPI7p)F+Bm!e_lM0>z2=Aym&qXLSi|KFA zDBW5AZf3Xm;HDYjvVP}iO5D#oF8VEp5Pn^duVq@@qaKWdI-#_X`}^({(+jCX8h_l` zdB3J7V?AK>vVMz|@;E-7pb8RN(>%qqJ-dS48}Rwr9Iiy1e24}qa{8EOUs~E>^BrP1M4-vvA^>#>_0vge@;8JDupJrK$MbHGaiy(M>u8ugz+HNE^k zDuqixz49F>FQiR@E<^{zGT!(jf641hh?!hR3uO@5;?Z7wPN?PJD;63oFwZ^AOnZ|p z?K8QyUM{JFvF<`9I|J3Hcy~I>9nUoqc_HKMIZ}T%7pJ|y>Krd=znBVvhj{3I(BvtO zTGS&fOWk^DDTl3C0R~Gkd=R}fjT&wI_Lwxc-kcr(bU!Hz3nqc%$my_3gxm{(^g@AS zKzkiRuI!7n7>?;mp5^tZlA+K5seMs2+AfSq_7plVBeoAY;I>MM&wIuq-VZP%u7(uPe<+MkB=5V?djd)pEsn%o28)PBlCM2D^misj+|ncOr3H6I&>d8Yj{p9=G9oo z?1!o*mt{NLn}?ny^TJm`z76(9mB^9&W?~l;)cp$Y=F5^XOG6uUkha}~JqI6TbM#4X znLdm5N?jw$`)B-FJO7v)C2=zQTpCj-`tllKS@4fDpe|vmb|S0Z6<$_tW9LSH&JWB#*+9C&nA@W2Im>V4MM) zn0}eR=udNUEk|>MwC%8%Q7GxAgYotjm?BkiL@)){D~}QiO6lNyD4?E4Uj~(zF*|F+1tPu@)DU z`o0C)j>bns80Tu!W$x)@s}nhnc`BzpZ%AnRGQ{U#ks2_T>%qr(F_tK~VCl|_=O|5I zIZJQjX;G^Ac&B?sAK%Q9C8c5omMFa>vnX=scE)M6a57^7_*+U zC7H5f?x*pPy{mIi?6+)_H)gVS>C*7(OF5qMRdxRv;fT%N*|LkTRnCRYodY{o9^vQ6H2^2aBc7K* zNQ7}YE$gE+cv(jH5(q_PYT1iqS{1ci{r)CM3@-zkGkGhH{8^s^wIhjs6!{JcBP8AI zNboT_zY(_YqM;USIBm29xBFV@p>U9?`R2lNroIq_4(9L?csW0?HuOwO{Py)m6NeGrlHxE7SL5e87JTYM%3if!qY5& zCB@fErH>V%Riuxy;}_ID{`Nh$7@e7WE)#g;xa9Xks}+5hcH95~{OC&+@oP+1>;Cnn zk85MYO*YAG&<0t(Uoit%`tV1zutT-Zh}hnW-^;)iv5(m4^~^c+`_$iDlazRNC$;V` z?|gcESW0>pk|6U>sJ0>l8;!7A0A(@x`LeR_-A40^9$y6PH6f~2+V(K44)TD+`@THu zlg|M`kkpQ5PTC&vf!#N1`IiOHEskZ#J(!HkZj^I%nv8NUdq@5Jj0=~u$Q4yb@rZH% zy#~FpJTR6G9QELkq{C3FHRq5nhu}NVT+0|-U~bR`Pj}^oJppXRZ)1vVE35NG05I)B z`_-lS(fwZYiKF%FfS!F&hR-oH40nn`#xytnaR!MlwWfo@0r7a$?Uz04XO2JV(3#x^ z{*D2T2Ks3qD#kSNcDnA6gv4OSHo$M5XkwgT{rx@Qy&N3SMz$(-MLZ)aumCk50FSGB z>v2m8hT@}-ghvr7+{(SI6|XBz$J{^m_=0_ivkOb)CMJ&NQphvGBe4gksaYnC7W`6&{S{JWvG{Z%7(W z({j%(5Z`*f*onfSA{`94qM%e=ApHqKG0ykWRH%juOOf^>EHDgQF~G6PQ17VNP1+VE z=*4fR)3y5oHdjyzl_UYX<^7nTcipk$7#A@5G!OkQF!2#?HK3isx6VQEb4dq>fou5tr>v`9xDpyrtndd*;vHtqZ1Xe zabr6`WH-plg18wY*vVq#^b(K8I1-eAKJKi4p!Xz4?#snKzqQ|6va2jy_pt?<1*JLd z_I*u;O7meiI3j3@V5_Kb`x)wR?^D%>@>!`Wjd-%m6RG~<e^hB}jZg7KmB|=f`@=A@!JlTE!adx0fGi**;#$$lJlJW}cI8?Nb zB4p5zvfvLib%WX~Xn>HQKEF59`EUcHCsv(Ik5`cZuAfvB2hwFCh!<))-WjVW)jMei zNs)uY1~E30X;ztRpr70bt#4{8=!|HOeD{;3hjYu)5jjuj2<>QVDawBJ=8SEO7%N}9 z<8u`7glteTR;_&`tYsUsQaqaxOjH44Pm(wf+Yc`d+tNbm&t8n|EnWiMSR+5Tbwz6B zq^@CDvDYnH*l8~Q1U&IkiaZT$LFT?koNXlq^-pGa^TzGU8jNmhiNDG4F_mgJ*kjxH z&yg5;HuO^=0`kri*n>MH&oH0u=OrB!%`>uS`1<0pAD;tEhdlT7T{sD2$Y^V&WGtBvyuotW zpvI{D%88fP6ulmlnd!s%GB!m2#&XS?Wc+zvo+3a(wgy}KhTYr1h%cGk!njZk zlgct^=lpW-ha|yHCpo%)q6fOXgqcM|sXu^Idc*LmMC5UV4Z~^?{FrOGi*}$^;W^D; zYf8jK`4kvQs(>e?;|Wo93HQ>w*e=`<%or;D<+X(mA%f(jRri|SJ%K9bM;XxtYZOf3a*a_LI>DOkiEQ9=z-L4@m(;-tY7{!UPd}EF6 z)fh-H)br_Gnk&ck-rkN(pT-r#UYjA)srKzCCX$~K9tKcPj)R9Fk>POJy2HqU^ab>`u#dOmVAVHes_~xk!th2E zXbV?sLW?gmBWZ>Eb^ci~=F@rHr?)DM-Tm9EQ`}DRQi5F?we#C=SW5#cn-UMIT8~+w zQ8rqW*I{CNl|gQyICL>hMsJS62d8Mfoji>MP`NJad40&G%EH5=Ei_lY%Qz^x<5v<6 zD{N9~t?7+UVp_UqJ{qku#l^^;(}l{YAn*R$fs_YC4UJdl=@~?e$C*6=_|9x4UU@%` zft(_G`%LbX1-)Z(A3oOf{81o5~6r6{Gr0 zJ@S-H*!;?OwUN~UADM5tQ?KE%?3j8>4dy6U6U3PkhOb>m)*O$Ft+=(t@AU@O+qP*K zoWmKT^yFQ_=0$kZPtnNpw48wi?U$w}iv##=ifR}*lw#&)NsW!)%-1oBK7t7G+^leY zLt67?`0vq`kLs`)*0DKt+SG~noxl0M3aJg4uVJ95=&+n946@-I#ZAsu5anMJTiAU_QH8-caX4{< zW}k5;KEySd0YALZ3Gs}+kC|zpFT5n`T0*)K+Tn@TtZJS-{XvVyo9Dmh&xdm!LV5GM z6`khBK2HWs4@%<`>>K(Pn!KqL$wj)z9>!qkZnWlR@Fd@npV_W;qE$>^6cGM)KFG`~ z)j=rDnv$%3Ga;RRS6V3qKudnSap3O7ztM}a67}EEo7uqrg zS9lAp+<4Ov#C$Cnd2yV$?@?{}JQychwKA)`wp)tb(l+TY`pq70&dsxfOj%n@STXcC zZ7N17Jt}SRS*G&0GHD%RZ=b<}4%N$L;%#S?dM5KE1qm_VVnY={=o0&djgAFBZlM1a zq$SW!*Ai5f-AgR%dMY_aV-29`bi&#ls*hZ^B`2u}<~~lqxLD3&R>hh>(MjxgA>9Tv zH?ak!ueaBWbXsZX(L6ES3aCTtF#b#6NI(m02ZfxX;n zgY8;oh}(KvcimX~rU>xj!A)jd;G6i`H&Yn>oLu$&?Tds}IKg|fnnyZI!W%A$>jR}( zgMuHHI@%rnh~9>|`;X1zu5I*(JbW_-iZlFYXjWpe^vQ-HWZGeqnTXe1mdB@%o9z|5 zgd?0v^ALzT?|r&kedR?k_cIlR{LYzO-c&H;99MlUp3R5O~g4z439PgL5m9VaO_aHK)TcOjX-IU&? zOiB1}d8@QOxr{xmn}i%@%G1wYpu&?0>MXQ)qO2tbWPXw%<(bS}HnR}O*|&1+1q?HR z*NV6K{dix?rKDSpR>nCo8P6=mn?sDAE-UbW^w+WFw`T&Ffl@ev8Ie{0x9-9O)vUTY z``LEm$;7|e#3Nw@g5As=I=na}N2`Vb2MW?FS{=`*nY^(|!6M|Au=8XvZF=-%mrv%} z0bb|(0prwcRgRBx?%qp}&4Vc2dG5B5JVvl{e)}k9aq4S%VtzbU-4g=2Y1=#Hh8)t1 zN4*xa7rh4F2x|?SUUD(U?;N;_s7+wKnR>D#+O*9|>9o!d=e+FdLw~8q*v!^FgI~0De%eDOr~Vlr1j92_wF44m2av+a%VE?z0@LH3XP{HCQ<)ERc75xif@W zE*DuV#qvfw8SDCvI2hAk+m~0nevj@CxG$Mz2YMsiuy1xfJB>)fmBwGa<=oU~B9Pa3 z$7fUoO!G?vf|9TZ4fo)TRxKPznY$3+GlDnKB;&-^W{Nst%Lm^n>K*o?9?N5SUfuLK z93){;v6N6F8BGdr`!>U)a>d0K)#-qVt)LDJd)(`-*!q-}zS{FcKB_4q<_9@yp?~q- z?1YeCcDOD!+zV*PFz;FZAJaWReD5evAMc*LCwKf8M`W6O%~=s_v$rCUBMdPVuPsilnCkkQ&9 zIJ_4(xX2Wmxp#m10;hm_Emef2@{uxldN|f z2(4p_LIO5Lh$2sycXQOWUO(~`ak4U9u+ZJ{8pU(GWZ)8sU%^vcQu!qF$!k?m^7->y zDkz(U25A3#@RD~s#{T>zANFB9Y8wYdl_@1oLVe#4+CsNX89xm9F5jcKguip~?s!0n z-?@ntZ3F0oprMf)XKzT;5gpO~JNI|C6g-shfMJbW_2`MMp46RtE9wSF5?S(LOM2cU z-T89#Ulg?A;d>Ze0V$O(p{j*g^h!{(tzK;sh)ouT`K@jvWZ31Xz-lmm?NMK%)8g;` z9+6EkYr=qS$RYxabrCeg$`+x8ODpSDoK9%SK?o$WC=8aB2qY-K#myGWv$Mpv)xF9? zLx0O@jgA8S|2|7abZPL>9*df?&>Gf0cqWH!kp05#o&YD0hl_fS=y7i=?+>pMI-1;t z`4RfwEm!Ajhq+QrA!F2y_%0e0n#q(X;ornmNPJ(-?OFfKrwEL`*Ep54;}keL_G_{CDAkw)b6JKqs9 zvEc5Yq9{SqbCq+jMGeFSvp+@zOV45OHS}}ETvzlX`Rbc1HX(|rC^SXiNTu_U{yXke zb-Cj!|IQDV-{0ja2=~QR#7ju&`{XPB%tG;{xRd9cBw3pf43i{EvFVA3FH3Z6@6>pn z+V%GD46dSgN7zzTlCecc;eY2WUKGz)#K!=mTLQ-i8APQ&5kw-1g)+O(o~}7MC$Gls z@WfN@>JLXYF3PudA{2V#YYG~|QAZ7{zZcURj8rzuavY&1!%Lohb10}87~<{6ZX&s_ zm;nU^9y+*f3&VYTA7kX2^BcMmeXv#`sj#TT#iy|R<_CAp@(XPL$_-L*oNIos+EMR^ zy;*wT%4psHt+y@J+uO6i?3-Qttgo3#R}yXruMUNym%Znx(_=7)njc-}b+ zwR}D7RXhA66jn6Z{p@$9C%n^8YD~z-S-0(}WVm!Quyqr^bDB5HF!T4&tDi(hCo~mL zBzH5Yu>6k9J^_x0QF>*)%5(4G%`z}l*e(EpPIjNytLKs2e8({u>N6UAMIaeWN!9|Q z$eO$|MN0DinzYaJQx$Nzl~z;N2OrIT{IC&&q;*9ptTc`4g9hw&ALS>(370IqtsjEq$?yi4*#y&2C{0>Mfowthpotr*My$_WaM!{~jLqVaSKok0s z!dU>%BTJ2CovXObZ_o`*zJ3(&jRQ;@=ZLw2&?H-+JI4$BfINhY*`Gu_HcP{UUBFT)pYISgYD&y;!8upWB`^-Bz}Ac)%1Y_ zo(+nm=%j1{ML9rfo1=%aWLvoR$)4Lb^%Liq1BRu8Vk!vV)F)}njUD7wzQQr3`xN>q zwwUeJI5<^o(M*&_h=fOp-Ljf>j*=#d!eXX@T}hhEu&LKfuZ2RCkh=-;0YN190Vvky zw}|WGv#Yrvm0UeKJL>&VpJcBv#nSO#+ZY}-;Q>i2AkSy?KKgrdA2Njoa=cb>&I(FnI z>cp8&ZaaB7!=4IP%o6aM)4h74swjL(afr*-x((A$4@I4JwBx(7VrI|N@Qvex?%~f% zsF0M{q3^cNYRZn@%!MTIe8)%C1pZJCxe2D?Q(}qzWL0V+f3V_7g&r0nCTJ)VG%|@` zZh@D}mO>+>3brVdDO_QKi-mP`F`(=L+8e`2q-}b(vw(HjP^dk9W$_uu$IeKl`*Gi zzWv_M0rzrL@tL9#&;b&Or>r|U`0jQ2N`ulUWOLfu9WBXZC1=+Z9QsJwjb`^vnr9(M z5-byCTvMsbz_?5eaF+_Kg1+xCqw+(=%CPlI2M_c5%wc`Ce;Df;L~QW~n?Czbo3!tP zd)vDNt#)Atl|l87Oxh0D#nUg*!w8Ke>qnOnbvwkid>!&6irO=wSd$kq7pSogD)lxs zK1T(hOEmVoie$c3li)j=u1Aq<&<;se6tiPz;16-+`XvSzVNxhHL74aAK3}qs3uCSZ z*rPl{zZkCj#&JdO$SsW(qVt2H#SB(Cnn5GvTuBIM6uZU)@-dpX=N=@}V!S<3(B2or z6FlPA!EP~opc>M!9s$JI8L#W7egSJXuu^s$X8+|%+#*qS?KO1M`$+TjU==>e8m!Jnt`}Ny4ZVJ;qK;kb)Q-jD^k47-2i;zUi&+FI`jM zqrbx*KX`cl79n+XZnj%m-mggbeU<9asG9!vCEL?icI~i4O%Jsf94PyoW$XLf@=U>AIm!fDGypF^(7y=lwI)XSl&f>6V7n_Grly(tI8h9ml-f`PBp!s!t4YkT&$c+r z`|o$5qtf|Jw7{;anraG~Y#gI&-l~GrM3u7;bo_L5o`O#G<5hx@C5fURt|)to;}LFxe2DZ~9)SJR!rTI-W9;}0`!ux4i|&Ex~*q7ljf>BX{a%jge#Nr2LZ z*Oa;7(G2{-kOw<98pYm{S&W(>2Uhkelmr3QdD2T=GbW0CQc6nXLbJernRV2Sx1Vp0 z{`RkqvjJq>Ja&}MM3kdep9uJis=!=;B@GK(6$f;J<8DP3>QcjPiqddO)AwO%Cv)(U z8&4Zp<@r0e8M2zD!O%5#hx{V5J%)7g<2kak=@Yfg(GnOWoBLrJ;E#hb;w5-W2o)Oy zjpO*|^;MrXkYUZkpI$0qQZ#)<&69Y=FMx>MIBFw8<$>-`eiVK)WqsanjPEKc!DdgG zo%l{ZQ->P1)EsR^fe;rR=+1>t|?)ecefQ)jp_0!pl_e!H^C9evfLOF#?< znP)^-mNJ|Tfy6@}-?psYT(khPA+pvw*g34}K>V11vT$znwlJI6PAgDG&(o0-2pTJ6<<}v_NI8O=p{dq!c+AT5+a!nXM`kTTe z%KdUJMdmMupq-s2xe*Gz9_VTjg#t9$_ExXk35i_thSGd@GGEDah~N2@4_BWpL$J&=%(IMgwlBLFabfmXeqzGDs+@WDXb+{i!I~p+ zPg_wKG0z3Cgby0l)`_fckT5f6z#bUuvCMv__B@`f+uPmy`?jlkZgX__Qb66*lIJzX z1h*n%4u>JL=Al>?7YoqcEWq}V9~tv6V1Mirhdp$D>NFD{ETY@y@;Gr!xq~xIB{_bu zVqs_HJ&B~$1MkLJA5+G*5ZuC+D)6hsfc8=}OlzW{Ih1%qEDw0iBP+Z?UK~B}B5|G( z3xek`Ld`WWJcIP|68&v2EyL=qaf9xkVQf}YMC$i@;SUqI$>InBE+X=vc6&5H8dtl~ zN7Jszxk=`?k9GXEFzHRp5`EOqP|i3ah5_A{9UEEx9lZ{zw`J{Kl!xa(Y_G)c^!~uf ztlYBPWIPfx0&g}$=^dG+J?^~>F==r?TO1YvIr}Pn-b2l%8{3aMzmjCw`Zx=P97-Z) z37Pr}-yO+r$%mh*4(-KDTI44c^JK_UN6OaZ%d+bNm~vX^DdK?)`h?TP_Cjo!&Xbzz zckwI=7K2*1LZISEA1T$G>tzD@p>JR9F@h}A{pi&>H*jjIe*!8o08ey8le6YV2>kj% zbut77KS*2~O(w$_zOhv=E^g<}MevNy?u2#=m%G<7Q}WN9WVYOsjOqwrjJnX~m}cDQ zBG0y0Ctvz$YwA5R2L9NQrZDO2#X;qw+%}I=>BX}C!9)t2R`LCCYr4IGh2r~+y$t4c zT;`0b%0Q&%hC33IztU$Qr{;`@E3C0>WZO2l;J25g*^Mhp96CCoOfe4_`0Cvh>QGP{ zt+o=*^O=XdHnb4lkyOUe14`eY4)ICvxp{P^(0;~vM^)DbuD+!g8>M!{xxyeQ>2`&| zhYir<#LCJ407+U3bVLiIUR1Z_j968ui}sox4PO!4X(9K@_uY)rI%wZ95P<%N{ypK7 z_WTGScT6ko%v`Va`8|O#l9u9RFOhY}Z5fgN;?V91w9?Lyk9eLwzNX8K>(}wsQN+|} z@H?5gSSn~p>SLBqVv$;l^20G?LGWMAxXFel3GL@gUPqTuRhuFYn{?AqIGis4+6DS? z%tsub&%R;-qOj`S8b5tfE9mb9&Q|+Gkhe#G{`%DZ>(gky zxxdf&f1eRPllr`Dv95XHaetg)s>&By?eu{3D)*K}@%4m$?@{D`?@|2nR|hm9s15p? zJVxd7NxSM5Cyx4fJe0+Ww62&Q3;yJb2`T0;5FdZv4L&Z?Rq<7x!e5&$R7E7)dh1{Q z@9+Cx+mz9`{b>dZjA=KYB`N!bB8?}uF>46S>{QNTP{c()fxQ^-VlP+W_Ig-pKV9MNyBMn*q$u!qmcW#2jP2}tvw z80i0E0EqI3lMh*fMKHb`s$Mn+zC`EX-?(%BYJSo9f5vEj_=+lku>bC(ZvS8Rd4n+n z`DCx3k{5BbQtj*)hi7qG)f7{ikAVh^?H%$MMFLy}f8px`9 zMAWkVoUuKv!5s2eHmNKGr_AperBRa5Ouk=%bk|IHUOXI(7C?&*vSbsn7cNj9Ql^JMnMD$;}tZPy)(Y5mGtN8l1+=^E$x^&QC!2PAoEgLx-w zwg>SRYHc*i11i72U<+vMd;=6bu9Woago8vUEy(L86dR^Sn8l}`Bi zd45_8hFF^Q>7L!viFjx~uj%`c>xcPf)$*9*AsxS?>=x0oBYtuIz_c}$3LRn~l{dX2 zyw-4k<6M~t=>lW_yYJ^onf!;N%zD`M3!?I3?UEZ13im*Q4?^eLb*yP$*Y$ej`f;ql zCm;|06JeHmFBG}saZK=iw}nZun5WLfE zv+ZcnpG^6h(1*=fGyX7_&MNSg`i@gvFu)Lj;281~l`%Lci?H{LaQC0j#?~ggsEHzK@t;iU8OMip`tVpHwgu$C`Ic;l8>r(+BZ_ zB}VEjn4|6~#-<(479({USm;d=uosLa9GX=OLJ&0WgKJ#xhj%vn^Ksgz8X}M{aB-LW z$=Y8U{;dwS?CaXOo*Ed{W}U0zAXm%GR(L>j3h;7K(G1ti1u#%us@T9kUr66L-mkCzrhyP!D-48vLxK#259%V0?Pnts(R&3Uu zL$Y>!i;~p)*CB~}@q=+O41#O4d@e<3W(A#@MOYLN>*L0qNgRk47BY|}X=P7Bn;R)Y zecI6Oj>B2K_{pyiBGoz!f`UZcEf`0>toCv2T|#onN%9SwQ}uI3z*WR$qBWfzjE`xx zHmc9{u|o*7Xn!@wKYSCu1L`$=I$h~ZjQYk!SW`u>(x6^lkMiRp>Y2~Fvy`^5RNVU_ zoj@1!m(N;Lka&^b0K>J<7yri{w(;YZER8RY)GhTNpZwNsqM`IfGp@*6ezE^BU+jR$ z9L-0&Xn))qy+k(*GE%VB|IPo0Peo89dZNt^1lan+KoQ2)H?C_kD#hpfl7IZ`DW7)J zqCbvwj>{Ru?=J`616Os)ACEoc$$mb&-@Wajn!XyZ{?&^1)Ks~8|Lb%0c*K+V$BW6x z{XLKPX{5(w|J^Y9T{?V|-=%-G{QOUTRQ2?IJk-yxE#QXZ<99=FEx7da-`X-`tcgE> zH3arXzd$1Fxbx#^p7;GSA;U5v%g-0+lE8+oli@^FupvAX|N5wUK2Ac`2nKM>7l7ZB z*iPz2tI7Pfdlj-adLZS2>YWS%+~bIe{2f2h=lOf8h87*^^y@t*NV&18*Rb%Me8ZKc zbJyGb$g9hnDW9O`SZUHp3{@WnkVo$sb)Ijr)n*h11gr{hZ4^)3TxSQO(e-WN&^kE< zj)TepRIwgU?`&0`&;OtC8T*-w1aO~a*?+h}8<>gzv0`%mQY{u?9D_TK+^{I{M)6#s)ew*TgN@5FJ$%St4z&xpLT{zovf z(BC7D(XGZB_>QIeh|>f`Z}$#3fcoQH%xI#lL$&vJihj?|_L+hF!x%yXRXR%_mqIQw2` zQ{bzk;$OYHe$EpP+%wSYG0ux;b1;gW_c1wOY4&@*X=${NbYRT1G^dy@0FY6OnesdV zgevjUg>BQ8E#Ad#?1;V;1xkJ7Mc9%+S3US@&>j;+rN27God$skx!81i{`*<|-``$V2$DBc?E!6o%o5 zAu}jF4m&F`V}9-_dDT%M<>noIBL;$|=1rghxuTNeIR}{yz~r*YQCjU~bbzg1FU!Jj zwD%3M&x7VSA%skhx^B!W4a-~MujN(=76bj5M6Ou!9J^JoT**TI@YHAt z0(H)ShJ4w+Jxpdp=b=%25EzKN5uYz|0ilz(P?j;)8^2dPd_Ed(ZVCH3?)$1)Xn&KG zBF|_thWArKDi19Z)Zw#FWcaMQ96 zJAXScPy~&qBO1Z(fRqglu_l>e83e=UjlnXp2{q7dmW~Ij1+)<;V*MMoxjO^366pD+ zWS|d6SshPN010^2Z%L>t{+cnM4hwS46&nF{P~_$w+(-{i1|4iv^aIpt{IM)a_CEaKW14V3M>nG%Di`?Ju z@~v*+G$xop$7oPmbBobheYpjS-tNdicX^Wx`Grb%2KmY`8=kJjt|$sqI!G*^5mP2~ zm$w#beSdp17KVY|C|M{VkgKQl2qvF_(y(k^sGnhYe3do-P~(-NhU>u!&R#4YYH&zE zR+F(I9um)vDdA)kJJ(ch;w5bW_aqzuul|WT6$quPcKsRMgfA((KndUBEJgEgOb7zO z@#;|lJ~4lD@SO z^+iK&oZlHpejo;3JH~P(KdL?)(K7x5t*Ptn#r6jg88x6BOeHLVyS~5N>(wjEHoV2# zvi$j|x5!!}vW5EApvCS`77(aup|jwhGP9X0@?6Ezlgx9Tgd_mRaNSx10CB%LBu+@- z_+4s#xlc(^5G#<>PyqB9RsXQM;Q{V+y*oA%&2}e8_x?I96D!!qYt{kubO-h&13%Ay zvDG?rnSbL3>Sw#s6`?TP2qw}8*%>*q(~c)e`z0}bUP^z(DuB9drr2Up0SoRqea^r4 z=9$9B1F3Z`srXeNE~rze`|(TXfw0dKh$o#6D9DtrBRP2owf&mE@ENCDtj=HQNiaEx zcI9v;V0zn@`7xnU93*`Ke5+?|fZ_oH1T`9`Fl4x6E`3)2D{OgyO`uSW=6QV8t5AM5 zjv3Pu(o8Y2zUGLoh=ySW0O08)XtFCnb~aqpYM?yZfh_j-!0dt`T`(NhKuz#o__N+>vW=QVU-m1 zJJ*FoZ@>NN{ZC8nK1#wcDB1J*b*~+80|gPlOAe1S+g9Btf4W_wcPN!31~vOZC#p(^p#?$a%Nl(^yTi;!5}&NW1W+W6V<<2K3V?nRfCxT+HrT5e zxx$tEzP)itwJR^$6$Zw()qJyaB~ZM}(*vc~j~?(@%b@G@RsH;LaBd`u=G$NVO6sLF z|Gbc&ZuoDkJcwdhl$P!cHJDO3yjNl=6#ONjyv%b3`Rtk{PVwWP{#M`L_4WPJ=#nUj z%|HEbw?M>+|DM5#l2`p+Q}Msif8SI1z$8vUN`-)4(g3OUHP~MAU6|tAkWphMYexuw zb2au*Q!#rHSx_NeHCadWQhcaKf6tRfkh_RR<{LV1lOueJNyGD_5JmI1aDreu&eNsQ zq3@16K-f@oM9q9$LeT!S`FWXQXe&nIB4GHpsKD1d<9@yuDy{y7xla)Yg+A!tBliVd zc-?dw!u?3=_d;~qXV9#hfq&1$5}^BvwBzXCx=GmG@ssUiEdS|D5;A2)dsoB4V|eg5 zfzKs!U@!LH+RLJx#}V#vq4fbSg#altSE(s;prdu|BYn~-3faYUQ| zTo*c50m}#K_WZ3W$bfcQq7Adc-$ukWoPl{=0t z;q!dR;pcpI+VH8hP`Wbi)6ndOx{~NSb%mtk0PupQ-wBRmSWsx70RJ>ff_HRNR5i^BZ`-NiwE^8I#XHMd0^aAUj`D^uz0itEvP5DminAEq+@3 zdRER8YW(TU{ttWavFp6Dg$K_46c=~O2AGp!Hy|;JSbC_WB?$Og=k>$#$~b&gqx&LjYvPIoMtK$`9ZaQOMKg+QR^ zwV0_aUONm^=?$oWYZ^-ABIl-N#)wW8A@c9loEaBTxPZZd=9V~Rh)g9EVBJCMg@L?!K?ue51^k%+G|Bs+OQ9+nH-%xbfwAT-_c}h?<+>rwB)K+v zqk$51-Ah@q+FASxM0bCe#+8S6e~j2}%Ln-Rgrnyv3iuxea#h%ZJ(d3R?Eziefml@D z-8OvemZqxbk?%?=%Q)uUF%v5kq#}ruGqCNwc7|a-;tk|4?aa#4; zW}gC;I=;uugmc+WLm;*cX$hFTFqRpItIEkw&$)Yj@43Y5#?O20!dZX1_wbDs_uunY zPe2ZW^fzAx$ia66$`wqNdZCT{9_skT8UO3;ZW8%_lYfdQoXRQ*;aB@&y~q9Ic)n?? zB=>o}{_{WokN@zGzn_1u`LV)d8&~~5|KmUYuYbS#kAMCrvHEUv{*R@)@_$eS0u&zd zkNd$3|3LroIP&HH{O7+n{lEY9-~aaS9saY+*BBIa4EFY4|DP9!Z5_AuzpUq&gV+C= zRSP)c`CtCF`w#!q$N%fUxBnkI_#Y=(b@4Zc$NqbNy7Hf<{X*1dEUi6VK@VG(MsIOC%Oaq_%p=QJeP0m|(=E)!$98 ze&1NOkyO?x?PS;gd0VYx&*)^n5KIgAzPa%E3PK(H#g^b7>+_Zdi^|5w<4Mm=E4+Q1 z=uJG!@UQ3F#e>Koj1`I3sBnVNe7`KGh~}!&h9Y$zL41Hhq^rkJF`yFOsPk54RPmxm z_tH>JBZOvsJQVJ3pY{iAi*X=6U$2c2SQ@sL$F*J=l<2gSn3sakEH&KR^OrN1@LZ`k zclYoU(Rf?8Zs07=f5C9ZV?rgCmN*1*OX_ekUrDty08{iwC!LmLg z@_lxQCD4j`nt)bMq0AiKHcF<0fAjST$$a@++ZZ~>!b6>8ZQ#zL_^CJK+@^X}*P+~8X< zO?)V_z5PffOf4~O`(7(`q<-=P&l>(4^Ix~U$d&^}5n(L;_+>pX$e`A(9;X(02mWi5 zS*FS>({#NrcSQ^8Jftw)WLd@U_0JItd{GlwIK?n4kW42CtT!n9xBMU9g3m#^KqNXA>1%CPJQ#k}3MbSz($GhI8_8tFc{jG) z<9L_?2<&&-+#!_3bv${NZ^{Q|VpicDr`sSe3O_zpIt%gsy4>7kZoMY0*_EMF>+`sbko~w4;_L1kNVUTZMML&b+jGiYQOQJZxg@3kiF{5Tq&k%~i#XZ>uJZvjz zVpVkj10r?wCLa{`6=6jp!uj7Eb=n@eFFM|^`cs<&u?J-{{Fy5Cky>(V195eZCcn@s z0fbI7fpfvUd#!#K*@}y>;h&!%-=hb#_L}&<#s8RAo>8i5z|M}h$oaDjeLNDJ6TDBe z2V@qAx^jB$0meH}+jyCTGc)Tj>?|#&=lv>n9QC2_?o)-gHxnaY9aC34oZF97AjT|k z>%ly(b%Z&8_e-gFpR+aVAy%xGh4943?`gpI7A8}3_uuXQZ(0!eK+M1U7eOm{w%byy zT05Hz^7tMor;V2BeHJjyJR$n7O{1@WBu1aDL#ET0S3m+>wW0Hk#$zy!_@;l2FkA#7 zvjb^=-hEL?tv{GSYyg>mzr<(`#cc=-%VTDrOu5dBjR`^-Xk3*sW&W=Z!`uyUdBilS z7E0aypbB=^_Spk%7iahS=IO2t2l%VTPyv7g!3lx-NEYvBy8Ek6Njlo-gCuI3AO;>d zAJ4p2ItuUm1Ctn4F)(BR7e?QqrE5VA?%k+fZ7+CN{jL!*`C+q}TDyQi1vplr?E(Cf zvjnPW`#@zBB9^sLoPlHrHG(n4MYju)U`>|WJ!h1`h|Pa<#|x%$&+l88$+qrpnp!11 zX`9JcYjMN*a!CS&%mBsR_ai>J2Vy8TP;JR^&p|QiFwZc*EDXbWH$Ub2c_Tl_!eGv8 zd-$aQe7==c6HuNeDDa)d08&HZQOe+jL-#-)jXp^5@qnynNgRGuZ0~;Um$7Oz;9v1* z={(@uVMcEOR+^G!!Q>aqGj!KuR+Ij*|^+|#+9WO_7UzxBW8&;;%w zu{(X7ApklyuV0rA8!Qc30Jd}nK=%aq0;Xy3+ivk>;ZooIW5%6L)_|q`ZSv&|NqS)m~d4&yy2C-8Bq z3GX>b5(XnqWoFOg0IxaVulPDFvJMy=i-8L?TBOJGZ=PrAnCE*5zy=`wa#OZpa${$0BwsXK`i zz%U^UZr)iPMB&DPfcW;E8DF`xD5M zX_r~pZ<(9BxrqnbOsG*TYF%|iz&p~$|Nj8f%Q^T5sfS?DcA*tye&d28w0gF$4 zq7e#vXUm>ThQi8xv)>?kJk7GA>88uQH0%T_s`Ku-5h+fe)qojw+959_H`UhzjKhp1 z=0>3`fU3iPbZ)lyy%^{HM0rzY$~RY}ZG-)E{y}?i?O(wkNt7U2bBi)gbIJPMZu4NqYxhi%WwC6 zz68UmCcV+k1pWu}ZVdElpa(k#OXoJD-n{z#dRlE*TQOzR>d$}k4QVkH1$^4HTOX>J zQn&8Ka$>hVTTrzaTbBvwUsFejxnlAbDt+>xbGK=wbYADRuT+5Dsq)qN%QHJ55%M08 z1wc2rxLZU`Vp=3gLcRTJK#Vt>29mT#;TfG!eqe=PW|r>;Y5@2j50JmXjIz5vqjzh~ z(^@=+G!i7vL~Pof4yyR1s6MykiZ!nYxk_VX_2b^h&8t`vtpOp{E>gZ8dL0JQn9o?j zmtPd2jmie{NR{Fo0HwrHYK$<~P z*@n+dp<|12#byWS@&24eVA<9Mh<7lgPo>{LC8vO0M{wQAl8kpv z{}|k$KXRck9N;P96^VF+YyN_{_qsd_VCNdnBQu|JZa4Ujzfr{z>TJiSyYDs()uNWb zKje~}m4{`iDg}KkDLH7(vWB&ox6mDhoiFsjHM-hwBI)g^3VF{A9&DBs!tqP5`yQto zVk!qPmw+VMDG2rM(D5kF;n*?ZI6%nD)e{dO;P-3s zSm8~gkXqq@gx5V^>%1D}@eTk1mkFS=6Jpt?3WJc)kK{8&a2946o`qMtx^RDziM6`d zLT}jRw<@*Gdrd=FMguwKqebw<&#ZwxF`UK94sn=8-E#eXUSck{BoZrPh5lfl=PlS#FGz*GK{FFArYpDGpQ8mMlsiChO`85y$_u5=F5 z6mTAp5ovS>< z9UmrDv!nsz<9=bQzhDdSH40R%BK%?6J?TdlZF;u2Se0ST{7oZMEHI%V9*@2X&lp{- zFWto_`s{ETLhYSl!Vo`g9V^U523lnTx{!$j&EE%f!jliS7!#+kuO zXvK)^_OH>cao!B`4mXd6N4u^8J`r#~Cq+IqKn^ZvVi16T_lm!d3lf$q9qrn4I~<&Q zSx4$K*WiAs^Q!N;fB4Q<%(|@@W3+uwizI`^7%(s#Y0<5vUi5^&0LRKeKkA(e<kcva@eKaN3n)NS9y`&^Y0$y8G$p$!8VyOtaHgIQ}Ev4!t(!o%mDZ)dXRI6-~_ z`QlTsbJs3GyMQmZD~Ge~?vvG-70hDi(wPdsGzm>sB;I}6LQxh@T!p>$;oTnvfCN`7 zs2JQ@vD&YxgL+*0q5s`;JDP4oi`2+H?)mbMDL=l_6)hhNXxpssq~A|c8Yp)G*BQ^r zD7QS%iN7^ZkuSFSj#JsK-93$kioihe_&$v@ARxD~fuV@JbFzCaIp7Qi9|}CGb(GJO z@ngL42_!b0GMJw=sA-5jLoK}T)dg4I$C=r=v9ziu1N?i^+e3+JJ#Ff2P5y zWqt&(bSkXx)%ANi;Q-sVyx_>Dny~p#T&)?8HM1! z1y*vDqAmr8V2fy%n@hT~P2nHE#sKUB6{KebfZZ*Eq98an%#Ob&%u`o5EPh>CxS}D} zy8A%i__rTGOVs(jM2rLqGX+uu8}Hstzo5WiY9-;xce{JMYwPcT5b!$Dd-RyN35xvj zcpIOse~+5KVSgHB{J{lKYYY@eIRBkv=>l~Z)Hgo@aR5Om;0Z-mTjyRhrc3t#AIY=) z(XSQ4B$io*`BJ{e-!g+<{{9{k_Z^hKW8D3Ke|ar9i)8C0^)9t=C?7miUiynn*9l8L z8@*i{;5-4;hAR9EU-sv{_S?IFw-st|Jziu;-}k-t4)~$2QR3hzv!am1G-}I_^aA-3 zD?n_FL@rmq0%oDR_kV7{?h|M`fpp{-{ylich{P}Po z9p(0>9_6cIptG4Izp2pHd%m*oULit1%Qp@NK2$_y7rfvfGwT2!XClG~n&0z4O!h{wkV0wrk!DK9WQ^Eyjz$PrxICMO9E zwRb#Pj?B7x#p(KakOsWHjj{21G>G(#E^QY1cck6DwE41;24d#wVxSk4X2wD zpIX{U?GFZU1t1mnufL-F-PnD5twuSE3lEDzuwM?d@Y3V{KBDfgzmI^CEw()BwJLbbw$z2#b;)P_%o{W~folGLx6$rLgJGJ(a&<}M0%?r9o`7@NuGgiL9>Ax=&Wj|94nOqK zOa$I!yn6-C4)ji-rrpD z>DYZH-+PGC4yc?Z11KB2OJbV$sIEp0&N9*CT+wr+`#3K=pZB;EN;+ zg4NzJ;MvpMl~E_z$Ur8bHA%p=@e<3JW>Ap)hM*8LPkev}$f$k)?&tY$zCDP>F{89@ zKj}{{?36U#3zX)nfJ_JAp}XJPAIz>e_gait`o#d2b~Pz>N|dWJ$L!kLYe0U#!BBy5 zgW5>%oNB_>*;kipL5NW7VFc@7_;eX(6UVP9PV>Cm__S#Ed;r;lFu; zdwn)~JkNrjLMkJ~zxyjbXHir4{R{jfaoq8KO}9tYyyv_ZQfgkr`~B;F&Hum6|A*VK zVr+gNrN$ww5?M_t+fH*S0GNnIEBc~9_0#cSQ{fO)K+Nuu_)-n9;ynSCm#G6HmEL3k zeFODUdyR8HV42utctBoh5{l=71Zz|1%qK${>QnV$r=)6~*84B9lU;s%``z2QVvSZ^zZ|huAm@h2IBPc%4jQO_ck!sSB zNhK#fM#bDz<#RoTLr)v9gLD7&9gY^uKU?UiPLP|boxbkkIYm`@tRKK*!yW4ECjjTN*AjD!V!mDarO2t5dGXZ60um(3 z6SWm7XznKsn*%inv?%mL-LkviI7H@=X1^HE_fC4_*Y+Sz#epa9{GEz8;k@VctM8e# zsHs?rpeeGP;4&KW7v)-3O2}-S7hg&!-OceiLS%exOGm%qzRo@OUm>k^<*sAVx$a*; zcCq>*XNr3=q++9wR%1p_B_ep@mpD#_RTCVit~px8vOUGMUQCH>Ys9YW!=j@FA^4^1 z&kq>YH(BW+w|E{#fOdkWcZ9%k#SsU_@$yZ$M}2x>yjK#NK{NV{6C-m4^yj?BW_grP zkieB*NCTh!sEAE^qa@q3gp*GAodUF~mcCKm*&aLP&AG(eRTf&1b~Ng{=2m>Am@b8_ zMIG@ddtFV5XIfc3fRA85w!c7O^~KpVRPi-@6GYY2j+c7Xa3B_t3B^HXSB!R^yW$Pt?n)-6i z=KunPlN=Jr#!sm8a!VlSG@{(BqD5S0;B%%1hOCrH1!-kcSe@cI$Y={!(2Le2UTc-p zJY(6O)Ver;+Rn?~{5V*<{}1Sj`K{WLYV6q+2jm376R{XuHtg^CjEJA7Feps3V< zT+=F3o|rwruIIkUrP{P#k4nvmeECNtz*E{dMg@H1d3vNEr2%{i(Jp(6@Yq+VL>63=;#4j_Fp!W-sM#Z?Mcej3 z7*`lUtVPT8k*u7j%xU>DC|K=@sSEd1%=R-Wu9WrOZl%DS8r~+qrb9p9LvF~#;aelR zpsJIp*L(8O5KDRlD10ys&qwfz*Do3M8yU1XuebL;b>0B`Oq#6uo!Akl%07C%>hr=| zeRyra#6`)O_K_ysFRSW@3(6sQs|=o+mE~CA8i_LBCc?HqC%X`oWTmUUwZq{MX>|#l zbG6pycE4$d;U9kqp>TrzluuG^ceng~*!TPdW7<%nC48JrA*&|HH$9YkgnsSzejM@> zSwW&ykf(5aer8-<>5N40T^UjMi7(Lx4Mu`w-vuA_NXRIf>%I{$)p9<1<>LDH~VWkm_KYC*go#9>DDllWx zIHMe8cw)e6<@r)8$H_+K4Hqcd9guoayxC^aNSsskwYTLP&z{ka)ZG8n`_g`04kTNJ z5SP=Bf%jm@&uhvvi7AdRrnhw}AeeKN)^d-DifS4a4C`psfq-NfM1WoP(Q zvLf_77lsD~K6uqb`JjHvgjxNhT0YQzMB4*53gY$hYA1iI8P*5>6&luH*UfZTffE}( zLfL!nDUh^nd&fW^$`LvB@|(ZsBOadH9!HVvp7f4UCiG#gXY2gt()dc*lrVVX|; zRs|54)U-0xnJ+ZgUaU0H5u$Rh;@;mY{A?yufLf41jhL_3)HdBCU&AjePpt7VOZRvv z3R2azIA~Aw7qlh~V3P6yxf+V<@7(tuC&a9qYD$NOvfa|`(Py0)CSX6Gy?ZUnjSZDrn5C!Vn8mq@LRD{|txBg;f8&Q7i7z8}``@wR5rOlUX;>uu zcv*-ajK4;H?{v;Kq`N`(5^_RrT&Daf%ac9b_Dl2BjYsj5mrk98B>=;z4)olVg4{8F zIICu;8i2f$-wRPxo4k{Gy2zi;+usWCcdl5RV1+&qGEYRSw9fk}(B&@m?aN)+AjIcQ#3(4y~^Pd%i^3nT2hpdBvR=PFn9OyY~rgsl!4<6 zQP!jzj33Ay6-1tN*QCfFygm$oqh+aA{R9;i6mF5FxvVMlfFQ4Gu7aaT+vsl_uKEt> zR&U|2yFfi!n%7Y7PlB2sFX0UI4&oSJ)$-P>&eBeS@+b=$vPytq2gc*)ZaG2|zZA6Q zF5kvxx7;b0%N=k&T-$q<|1}d*%?H9KkP=Un)n95tZ+cJlnz1%}ntjRgph?0Fiar|tm$lLTGSZ_T`P>{_ZJpR1tC`}Re}UF0q4Ts}Q&tGl>V z7QoANT{*r&onay3TCiNt%PMTEVlaqnDB}>Wqm%m zhqeVhlLPxOS>KUsydN*GEz!~Kbl&y)F5foeWMi^I>&j$xU@k4dMf&>G{t2cO-O3qrO5DEedd* znc3vZ#WYmz^4r@Aw@{x^P&*tasgvmv*^68^>>722cs0Fy5*g5q0?;h4_dtBXdjO-q zdCY^T%N0LFCxZ?4{XVEPqSdxYeo{4UH~I0rdC{f8Dg!N; zTBE1{Bn&09@H4^UQEh)`c`9a~lFYH?vGUx1W5Ae86kky=bh4wIhop<)&fQW1)yQ{tOkW8lwsCUjJ za5KI6UNI-bk%i9@9rla3UP33gx{b`^$tHR8E1yqm`>b+;jI{!=UXVqh0B2U|De$j^ zgr2Xz`ak@ZMTRcOTR?(Jn_9r)Ct_S=s_S2)4}7LOV0>b3IRtM@W{Cj}B~^vs{6u7=sB?cHbj^TK67 zr4^kc72uWl@PO2w>8O9Gqha2@aPdDxOSE#nY}?E^Z)&&K3?(IDF3g05vDN zL=YocL=!<5I*s=c?n7+23?jaNW?maZjrp>;2{w;V+#3=BEn5)7dzBk2{~B;U(3Yu( z5cDr;@|p-}7M_!2;9QulA*F4_-=C<FpMXIibV* zZGUvss_d!wi@|+=N+|FN;J;!Pagw|HO@mY_Td=~%X-!VuUgM2D6~7%*^f7<=xoA-T zvTUTZpOT&Ni?K(6F$abZFo889(=%Bm4Zfvqa9>r?<1TOqeWBH9ssX>W|GK}yvrF$I zt-}##I`vI;ru(_4EV5rh+1j!~vz#1H$y@dRo( zQ?Z|`5$_gG>_ZmPYVC|SZdwE7NoBb*H34^T$$#KPXRM zMe8`|k`VA0m1D$YbxtBlv=NJEA3kqF)SdXRGM^?h^SOj9et8MeS{7ltx zlQ9#924R?NNO19)0rO+qW>YVQI5J6u{?NVtDnZTEPA%(jx4uORSjm8WR|8 zp93*v$a6<7134S_g8)fDw!cC(>iuRBMFCj*lX#+vG4{x$(l)+P60n@9{R{2Th;q-7 zlCjz4_m)db#)#U;h{;2_Ot;#aG3Scn=kBA}l6cZ_p_g(l#yFqHQqYQKPNQPlagVUu zcP!OpHRrSlU*tEG2!wm7N27fb7*n)%qv^hO<03j&SOwSjar_yH;Ss`r2>kP=Ry~Zk z`J5O`Mo!8|zpI658}`&MzQ!{m{FIqp=oW!(+bcx|tC_;S#<26l?W2k2>IHH?G;;lX z0a3TL*y*?tZpIi}eQ;FF6#_U8tgX@nL^Ou3!|(b-Z@A-n92|mbKX>S5^AduKaA)dq zT2v@$<8!kjx)oa))Qht(cKdrdNNKnyP#A2`#LYRQn^uO(H3U$0!one351_etcGoN1 z+kcU)L4+_UN8xA50+D^)gzGOW9*yE)Hn#;+y6oLw9qd-#2~E@1hum|q6zB-*&Gw;c zLu2s)maEp&q2MndM03mRW>=KmJFi3BD8*ofL9B^w)a@wS(Js34@G^{MAl_v8aOEHV zV7!&aWTSiidSyLo0@qUlOU9q!_Wc(-7^J}=YYt57M;po|px+p%NmOdz9pdO6We549 zYU`*!7D8ZFoML~^pPEOHQ-2Q19vAMUOT~CaW)zB_Q$3JWrP%<95S9-TW2YVC<-Q|E zk+y}9#obS@Ap%y&$3O3PyCoed;5gKn291Jjbq3#jAWRY|3l>{&K$+Ah!4Q(5Belm;0 z`?UEO>OM#sn>7tWZ-?+xL7Q=j67#EG%ZK*3?R)XuS2}K8Xo;RT^$)b}<|*N;P7ojF zoBi-L<>OABoWBqDmQ-}7%gcBSw0)KV)3d~Qx>awV8pkYFWI&(@51 zwOg0`nX4ZTdCEKj{GJze0tze=KdrHlb7d@D0R~_6Ym&3qi$yl>V;QpTB>B?@cs6(9 zRPIE?MRlK1dP|I)ZELluXL5F>(3#iM26su!Ul4GQ+ek_b`ig^*&Rd}1viA!~Uh`WA zHg&dg_3>>-&(QG+QM>t-$K>M_-)Mn8wP}QMllgw;JRa!-WfeMq4-*=~npe87LU~Qb zsxAP1GYaTbWGEk;s85S*IYaMBGH=-!m4BB8<)ToRGcB`YjC!iyh$h`6hv}qYnu#g% z9F;UjdR{EV$ciuf@Os7k<|54_6EO!(#I{EU?4dY$He^*@jTj$3xR3ZSBy!fBH0r(Y z>o^NOD5vXeKVuu&2A&@QoM}!E>~b?#euw%46CyHyKCp2Y75*UJnUs`pzi?S*JZ40t zQvnD15P{C@VRy{WW>tcfZarWm9o`*Y3I^*lQrzKq?v-4MJz?5Aed0L#aF_F>5^~#he$Zvgt*Aih*K9)Tb|Kz#wJv|31*i;y##qr8FwmR(=2P_f4LADKj_qeg z8CMm~XE}prq^DP_`7>|-bngE4BT^;IQ|@Mr@9hTL2tAqdC%1CrFZF}NHu&U1Sl3?- zWa&Khzz;*K`tAsf;h?g@87K_1k*%+gu3ZP87l-XLvSc5C^S6$@H2E0)rGZ-d^H*g& z-;xa&oQLHwR*q}&yZbatZIzcwz#n>ilgBgZCQo;W1?F|nB ziTS000EPWVd^@A_M6{RwL4QmstAX$^ukB;5N*QPBH_p=1O9~r!5*N_nP(SrTFcop2 zsP$~A!!tu({v|z{>}waY0oMjoPz6r6GWpyWGn1bX*P>3d(2`Fg5NNc-f_#h@lsO;gMD8>Z7OZBlN+w<&nY7P?XWMmdZoYT*E^=!=X-cmImA`-n-aM7L3;riQT&QX ztL@9<@}=Xr!-KEC`PqjLCj4Fq`@)p6W0#S<_%(R+peWVRET0NLo7WzTMga9+6eh_vB2jca+X>mBtfgkl z+&3tv<1D(z75sjdDsJI<@4{RSn?SX?9U71?wSLTGq7^)&tV#++# z_}!s0OS?iEP6_9<(xZg^juR>-VM?5x4C>p@aIs2aRnaiL;bM_dtX-PB6<;cDvRS+T zNqUMVpC3&=AFADs+H<;)(+t-`@w$QQ7g~(%I`4vl$F_jF^jnWx^6(B*SE~R?@iN8+ z);I=~2L$hy0JVXn0Rt7K{eB=bt7hOmTFSF*mnjE8K&K~~Yhbm9_IlTr%w%LBQ4{g< zUV^IccAQkjSPAQCi;_A}C3m`0wy>9x7#KYNK&Y&hvX#D^!?tA|H>a#mgr*B_2vo6! zuk?&NWjFfPV|8oi-`4?!%q|bi60V`{stpq|?i2n2O^i1fanbo96>I^S4w_jDJK^_J zN8s=E0CFmkl*)5nPpk8aqSN!G*2Rfm!l1!zZ<@5M>B8r_v}yI(z6oUDX&54r2}c&$ zL2QRW;N08=JhsEeOfo?G^}9hc3(oj^V^0=iKiHuDllz)2)^7--zwi;w@S44R+`ll9%g z+%qamj{+sV*s`*4Wjjh~baqh}DF*sdT497LUBb>%gy7Qa6zV;em<4q0yp3DjkG({E zDwq;$T1;ocU$g}MFql!y6^M?;ZYKP&p9kz#=$j&W(?O~Y?k@TivNV(gcp#Ad4v$(=Whz&E5b=`&Wi9B!B(>60;Y+m7C& z5Ks(KYSqVmD+NixE`_!%)WF*_!YeOKMp6x$bht==cmQR7k<=A4lzG5UNEZ7CP;?9~ zP~=a`38y27CEDPq&fXJtxyjP75`CtfH4c*K4iC>zY1znj8-FqBm!aM1c+P(0UL5|g zo;y{E@A%ehes8t|t&dw*%+Ep3znJnEY2oN%+QoVeSiQGRi{|q~W|$Xej01Gu&X+*D zJq{)lk;&DE5L(~j%e&eiW2e^^BXh_2YnFtPw2cxX4tPU4L>z#n|xkuGB~)7Jo4IwM7&GLU3kJKHK6gx6_3dwfcyD$Om4@Iq<|LE)X;{$VxrWUUPg!v_C6J! zF=f2*3KX&vlvAO&Bu|!P;cwyS?TaXi=6m89TT+?Ec?vWA_+KJ^>4y2dHUx&hF>dx9 zSLi$$p)c0%6S&Q`7oY+fQEdp;IXFS1@0$Q(43M=DYW}59st|JPmcAC65H=? zr1Xh0`~74X+hjsvdf#=O;4A#uo|la3hpsNT&4YSvT!2w|_d@p51&XV+$KAh}2dv?0 zYND9(Tue9^b#u>D_tp0BMWwjAz|Z9m`1XGdhgZJQnCeC7{|XY3bZBL(V?7 zA%*i5cc-#T^Sj)dD7`;}7B`64P`#dKeG$za zJBDhb(G8ueui?4p5|DTi^*yLou=ZAHG{#0ZLAnOtm^Ko4jY!Pz=Laj3e16p<`*E8h z{<9vTF+E5aaSt>(l!=4ma7R7_xb$ldUnDiB9|AcK9hSAvR-=c*kJ9d=6_hi}UwB~vE-|4!JHAj_Z`9Li2ZYc=?0%3%Q9o~D-)Bi=! zxy$>_^k!OvimI%Ph%y;10da$`-378OzKYK1=|~3d`CI66^L8_x#14 zaEupqYQDE+NL|M*oONX#1E%AwusMZc!}BuLYlM3N_igxsDXziH?0^AK#Z|E=4Xc}XJE z`vv+9LpGs%D?ku($2@!5xot!~ne>qJ=J7J$^L309TEG74=D439>WtX4G5sXO&Q|{X zdbhL&_bI?{F8LPc>phG28KCpy5RXgUBC~xzKE6?FQx&)p5g3Hr z{LaZ(_ozHEl1JZ5-U_VmNo+~D)kkmtaX;hX(d`EM3=Hsd|01jC^Ss-67IPv%(A%=_ zppWK0XLOhPGpz~c-?Co?m{qOcC)&;MliC?7tZca+?Kn~q0SvPY1k!G3j@x>J^11+r zQZ!RlhmUUotO#hpf!`$wT0|fcc&8SiK`-;ov+iloCTx4W(kYsVzTjT^yY=E32$09= z{oSQGmuR28d+)#OJn0E&lVB!~$7eKA$=ItSN$0D?tB03oN@6H zC8g4Sj$E-vL3b$Kr~@Gg!p!oEB-#mBNacIqmgrlz4S|vmL0|TJX;703x0*3Xn%}+) zB;ici?g0FXA@o|sWq2{z$DrItnF|G=6`^gd)@fHOASIczhKrZ5@d;hv{aCMN!T((OWip2(&e|U&U`kZT;#Ae3`Aq#^X z#KjJkr+!oSZD#RnnC$#9h<+@7{ZEz&2_g*Na%D^$-k;t!BN(@Z&BoM8+R^1zDw5In z`$W%f6Qn15?};94U^0-qF3MXU6Ide zJ1$tOxPs66^AJgtYG~o22}nYjhRn9}I|Z0f_co3BV4T1s@BQ1faawWeH4lvb_D41V zKb~>$SFeRt1v1f{vadHdl04T34Xxy!r}fPz;jAL)^QsjOxZSpwJ%9Fr0lLPQ;$6I=^q+D=4TUDi1V6T&82Ca$C6dt ztZY#Sx~WC#H{RD2?phKvV<8aKFM@7R58Y32Mg2~&*U}6k#$s?jXHL6js?Oel@T{f; zC}0Oq5av}FqMfx`AvkuVaiAL{*OW;lZlBdaIp}x=MB0@{tktg?ZzR+2n588epw=hO zH8-zPeC~aw>hK;f;%B}!Nv+Wc#i9!Zf4m{dCLxP3fi-&(A#=@uM|JEp?qcD9+~F|p z-+YNiyT(N5G3lt7t=K?TXpCnop8fh-tsb?^yh0|}%QUOb_oc83v7tJyPCL%FxL{@F zFCJ`ebjs>0X|2U^nsb=E=_su&qwDi0y*9bA$K%>651}s{YkswR_!ZH9@tIB`6>6hw z=WlaGDyi4-7>l*~+SxBaBt2JnR(4>cBc2OeMz!J7GHW?NAuwXOI_BgO;{6#RgCA}> zVd+FenK|2G@M|Y;=m!shW!3aI>tw61K&^BcS@xouva=4-Yk^%P5~C3r1V>f^bN1fH z?48uYdAHp{_IWoZtL;aCx?ge}T#02$oY&WD1i~vH=Hkm4X=(+#0&GSi^y#1F?s?| zO7Tb#1)fJ*y_cmSX;ZVuUYfgPoPGg5;gr=CD(DRC`=g-FhPOJ_;p`sX1NYfXfkM_| z%M%6!M#T#76>oNZJ`a7)PN1QlI3z)^UM*kvLhp(PLN$Vs)`i)f)Hr_#nzV$15pn-}stJm7;nHeq|D3#myROzGZ{=|Mao2%%p@>%v6KZKpmu zE!_3E$18B#-}PU%XUnNj<0+8dL{f)fje|y9^`(p5ee3ntu@$p zJaz%f>G@1?LgW3_owsr|SCSW7!`U-jg+b1X^N-rX$tgqdSJUYUC(AvRNmiS(227i2 zdY*{YL!yox9S)Q7FV^UoZLS}TQL%Y&;x(vIuwnI96*VfE$lAnR?vNYAv@#MUi$Dh# zBW$oOKX)cOMTNu}m}$$=8K>t!J1~;Md}DO*?o56i1LUtSz*qG(V`8AnS$Zyj<_kN+soBB0&* zP4KdONo7JkR`97zm!fIys8k`3{G!1l6A(#XFw!B{fMRRZL7+; zBWg)Q>ZB7?DYaxeCz%cnP_?i8-a9#&jE27PCfZgmhKmM&K|W%Q5#bx5N@jCXg$w56 z84da+>R{_GI<7a@Wkpp=delZNT}CqxpZPpBQD3<)(ZHV_XpCo!Wiqi&#R(!O@YxT5 zNXC??3?9pn+5E9U+)^nseO< zj7V^3bY7#JW;LV8fJhgv3d7lPRim+r-Spf@pjA&emLh2TrG6u`;cE3Xyb+Vf&dpb5 z%LfV%x%trNHamR{0(?Ki_3;Qca(W-!#;olR`uXA$z^uJb&ysaM5WMc3w^nJ3IF~a$ zN!$x4U4_0C$OFi(;ZD0yJ9+}}eEaj%k4&bc0hG0{cR{CX&9t@$aQgmeoiR0TvJnEv zvFj^G=D@R+CMT|4gotGF1Kz5*MpceH-TK*i;+qw%A>3Mm+D01WGsU!{6+|69Hn9t! z*@0?$@PmaPdz^NwnF6VD0G{hCZ{)t9lDE@$M?Op`Nz%@aRWPi0wBEH2g01Z(lk~Si zNO2xl*=o)L%=b#Yyq8K|_?emP6A!GZy-1ORE|z(uAxO$b2B(j`-CU{5dsT|7&%Vo{ zb|K*#r*utO+%t&R9bVO}h8I13(902bK(TNA`h^3Gos1ssr^Bb&{D@_g<{+XWwXT9d zG~u-*8~UpWS8w(92CJ%tRU7MgW$!mCW1cJkwJ9!o^b+|QjTrpw=;rr%o#`HM5AQQm zf$r1h5DzD%70()LMH&adGg7DOZv`R9sWS|^T-OwbatcZ)T4ZepX2e;1;jK_%DNQ;Z z1jst4c@s^g2Qwv2$1QMEGmyyW$D4_LjDx!XQRQdgwdYslBDKRLDBe0vzs`U?AuTWJ z>+^(58ziHiAQ>tnooJqnbJO(+7iX>+$%KrUy}t&6-eI8V$QlWL^0sBCU`CGZ75>`B zFk^TMk!FDZol50?hw}}FiQOu?i7$^lfMCdQp^*Hq&%XBDxLERfh}^;EC%2qPSuuOm_ir>n@bT5CM(vY$?GOqghUtH9~b zMGHm!>@r~HGgyPCxrfg61w^0#Lm$I?KkWzf3T#=M9JuR`ku2e z9#26f>Zy<&8s^2FrVAiOF5brb_YlhKtgO27@ zBVH1q{6KWreYGL3g^f&-oD;H)eVPfDR+SXi^i=bb^r^GXU$|I4l=h#?9jM2-eFIWE z%}%-iD^%_p7jL&h|8n0K!Y08vGSzjf9;y9oH#W+QbxEpg0@9)?G}Vt$k*GYRU;S&W zKD4>sAcqZL_O{>wl3x-kb){G$>uVswn^8<_A_vtO08jeWfj^ga%;+14 z@A6ClFw^zZX%(bIt$p4*v{~%@kn0!P4-zMpA z`SVqs=-Ixjen-6WWhReKLjLa9`f*zTk#DUOrgYcyN4{zsc>=CCh$_}LpWZ}Awr+(F zs&V6+m%MHMBsod7U|3$l+)AtdfV_$SCii>Prq%1e&@qYBf&%XH<^I=aUoU*ti;m>b zaH{+Bg%922bY23Bm)Zl~8DPt^n>*2LB?}`g3aELnjG<3JWru-aMRJ*}P0V-G8ug!y8U*~a)pI10{e|u2J{^)a26q)2 z@v2E}Fn}3QiF7)wkmFKI?GLXClgD5g^OD*pIcg@c*TLaq+h%bKYfh@ z`8hFH?CspzcXUGh(P$7n!XjA!JXtg{KHXwND|ABAD>ho4$tHI{TXFE+{szsXufa3;Y!4wMNPXcF1t67S7hQS=VgzjC+;a%7X$i~%k6<2!Z>HZ=q z|9xhbaecRsd)M^hUZ9(y-{tufd9m9bJS32Wq3S-RijIp{v#b;S@vjzA#aX6B0xw6V z%XqJ%QR>JJ|I9Yb6P$~+zT=NktfY(!?1f>E9TAhKj+`=ZOv zcxNQn5pMe}MHdjdfj)>cGT^NC9*S_1apq^qTZx$&!Yj!Gt!qLq$ot=nUmz)#ic@v)n&r2+ZWu7xoNaRBp>z$b9S8YO>aEm4W{~Ha^!`#8@W6KPSdRBqVr>* z+?sJIwx|HR0u!Heg_2Ox_v1z-fzY43UVnpqW#AMIlkg`(?x$1=`SSu2 zr%iR3q!ZH~8r|~Fs`1G! zs84bOmUv)>hB3s*gDoGy)5*ff2>JYs4#0I9QZ}Uy$YJZtu?AodARhxR99Y_0e!d*H zb}fDNjFVHy!j7=uA@t5`8gJLJ<6z-S=DC7?9kVOmHSK&%EprrFkC-(Ik~GlI0|Ppt z{ghC4Fx#)yPvLzV3QLWV z&aA*@e%&&g*SRB(z=m zGz{6o;e+Tb63+eh_cl?W;1*gFI|!y_loS?YvxJgVZVv;dGP0e#82#}qz!Z`@2h8t8 z9^C2nXny|SHGpQ!6FHKqwd!k!7n@lw?zb#RG!|5o{;y)b>Olx$?J~au;$3XJ0^%8G^~R#wZ;tdj2xv^lsQ0Sk#A;% zoWn2?e0ZOgENkc>HiPUf=D}^g_La`7djLIH<6pMN4iJN#6>eCiFFE{H*X`V_joe=C z8_o<0_TFf}R$L8#V~fL0HF5(-K1}AyAeU8K*y^2Sx?|chdW?(!*~KM4o~gJC@L)@m zE!)8G9p4RC#7rk%2aU4=!bGZg4$$W}yaxZ%Zt+^7b`9vZqYh@(M!pud!yp63`UVK{ z3?^ptZDL4RnS7IL7?2&kQg4H3Tfn_LjSLq<)aRvJ3?iohpEj$a%FqGf@Hf)FnI<$S z+7gb02Gd~_Ki)f+KiL-u%9pf)adNLw8Aw0o4W{|E#NqtcXb}b z_9)HhspWezFTW4Ws)legLJh6M5wK40=AX4QR>$H+)l^KU)oB;Cbh-9{q@-k`ZcQ1& z{uOq*I169;Zzt9@1jNx`S;1HHsqjM6Zy(f~Nk_j|W>dFcax6(1@f(USZjuf2g1cy~ z_pTdGi-{GHKj!%>Sn!K{+lFIeL?&CIOJT)cQ$hJONAHCts;1?$-RIT$T^of!>E^zZ zh6b=1QhuyB1s?~Z#!%A|FGSkpbUzRzMR;dIkhx*)XgHye~9_|cFt7p&0k zF`Ho8mdqzvR}KC|lL&AD$;TvGN}%EX?1Y1@L?j+h8IRV22kn{C z^E$3}*Unh8TiNeVe9kWQzBq4@j!AYt$n^d-_Q;oHoM_WpW;p*YWZ`jk5M8 z;6lrBW}n7JBXyCP=%>7X(-q7Z(C;Uk^a}Lo8nd_k=-%TB7Xf}R1sYjW7&Jp9NQVOT z?U*Q!dylP4=Ff19ACH#8xJGDoUn>U5WpB7m0q{eqexqRz~wG zez|+s29@vn&Q+${Vd1Waas2QrKKq!@dkubc%c)hE-t?0_chhQ`-1=p zJM)>X=_DPi7A+c`oPhU8Y>U8Qy$cA;;Q1b@A+Go5Ll~Wr(1MrSO*GALV*(#gHDFUi zqitFC@Hgd)*mJN0QI{&x?)~jIgh4I@vj(Ad7+Pi;3KS+2= zqi&PQ6OKQ$6Qb#!s-!F|!N$FxzcbxAU-9Bn6TpDpSa`}b6|uz!Q>}!0zBb_#X*;q! zmr$_TOhQ)iGd!m5Cl8L?nrAkflHLFALZEl;)7LNmC@(+GzktuGSB!JTUF#icMhrz# zZj1s3O$%cu0i(&@8__xk+q{-YY{2_5RH|WG>?|@oaANUzTGw}`RnX~axNWNwI;Ioo zxM{airQ>s!_V{FX0V93h1>BtMatz-Fme%Dv{@u%WbA)U$wb2IWnJa~re4HfW`rV=JH#S`cZQqe zr1QSwV#~@KDSg|f?RFpk&8K@oI6$M7jXOCNU!TuMs0!u79kKTI7Pce?!kx0c1;LjG zai@S^<2uc@==>+im^7JVS3na~#`ro`$27hZ`BX)7 zR*}%AxBc@?epkq<&M;R&zF}&RAj%^Kv(*fK!?F?yE0^QroS%u)DgUA!s@lcp3pY9{dpN^oF>R8 zzSm=~HC?QZEu*jH-{s zF-rrs`P$A5z-@%zo!f01sI9sKkr@RKen{?eKowFK3Zy^3F(pH!ku&tPq%q+EuIAn6 zxF(1A9xN$!TSM7(WzNPEkiHb~cO>)lM=Fd+xIiE6P}$GNR8Sfrh{frb-3El|blx9- zMB>T)B6NYLmJzN%uSk+mGlXG)s37=A=?K>!0bRK35r{8cwSFmz4Pj+yb7j_kDSrOy zX&JctF>CA`%{O&)V1SCh_1$dc?J%H}`|eRenSWJc-ea@RUBV66QuIbN`ku-9$j=B` zuH3QZiFz*ysVLh6*nbxQTWX3DIWM$JqP3+q-6(2#ilv$A1C5x`FK~8tcyp^kKiUQv_JP?z1Hzu!s;gA?M z{6Rsw&a428)g{jrO-^yz9w%8tnYv*v5~5bZIB zzc#q7N!lFIeBu}8^cr>F{H@oXy{zK1!e!r`w}MK$+B%%Zz7H{7PlQ!2jC{Orf9XFD zt$t6mD~Qyjy?v=Zp^*1~o_Md@R|tY(E7Ftjh!%qS)z5o<9|iA_AGc4@#V7OYS4(%z z+8N}tbXAbWnfPJFV4EIG@oWd#HUuyJbf@IN)dqVtZ2kXd)ov%SA)Y;^xM*WA$NrP? z?rM{f3j~{91&iMg0S~5h70)N~_yjh`{F*aJ%MVHjGd0<|-qGSJT@mBGm26#{UpYFG zTrjcsxBktSx-LM!)w|?mPTwsS>8l0R==3hG)+89Hi|RCHP(N7Hm5!_)k;)u-$~I@R z5_Zyh>tEZ4B$%5|y7@Vqx(@NCRp^RcQ>kIyb=r}1L;pM-s(g{kP*?yDVHWB8#q3GP zVJPEhih3y!1nt*tl5=K%HVXT`!z39bS==0u!~g8gTJWL4=%m=8Ov;~+J=(8o+ep5` zfWTkB)1S8!|2l%YqwmS`k24+;tdz!UQ(MxnwJoX{>kYIdyS3 zJCoiKAfSm3W^59R)~V*MnO7f3ijsOA$>*zk!%ZDwu>LwBW|xm(hN^He=LZ8tYL8h` zU4bJ{`o=QUU&0}!z&nt9=8Uh272<@K5nHH zriPxSE^}YkA-LP`MuGROwPNeHOmE|_*jB!Vp#bkbx88|o{Ph$uJe!XAlu5O)~oxREJ9&8lcAjCb3~*8Qb-%C)kx;2rxYt1LEhRW zI`zf?S2YqP+z$g^2YCU2P?yJ*4nY>PtOlXtxY&{pV}cqP8?v{LKOfT$!HST3*!hb= z#W=62gnh)Xi~V2j&SG0tcvb#5PRTi}YL?<6W{{dO9$RXZ>8&+mz=xe1k9 zi)M=uN!w;*YEY8?9Lv zG8u2qSykvyyc(>0H=flp3kQ*PU3Vq!Y@+`B;qJ3K6&j4G0~Pb!jT5>IegAl&w%ti)26EKVbP|?A zcNJG11z9o+mGt+lhi3-Oh!_AJJvgsO-Q|RZOmu26s17F3A^doC+^CMU4ijJURUwM4 zPmW-uIJSGg)NzozCewYy`6X1({K0FTwR$HR1Ch>#O8((yXa7NZMHA3AoJ9QL0s99G z9d2&(_IBB~M?S{=zc~c@0wJ?)C)z97$VJt420J>yKl)Z8o`Ju(1Yog2gf#LcJn&hk zl}I{`E_*NSq|H>KBkJtE_N;nRUoft4EqYCSGEqxiE19_hu4K&1Y^-X&6zf5xcni&5 zzp5n!5XIBp%?tH;ry-7bd^^;ov>2rHqT2=Kk!}7jMBUyEpWSQYE`a@AsP5NR=icNy z2&1Ql3}juJYH$Ab@N2E+3VK3)X*30C8W1wi;?b}yYdGCCTm|;!t4r$95{6kCXUL|5 zmnDgb6!zyHNFibaIp;j?@dY6td-B&8IAi#2VsYGkXn4;(mg$PU?dL1=Gs=0d7Zt89 zS^3$T3c32$3c5#lLU5hxtjNLHpFq^Bp*fnk`dm3n;1i_6$QQX}NT2zhmj~=y4GBi4 zYux}TEn{TxlfO0oC0oR2ViIuQD>|~bb>KyXyRn>+fG*-5qe|IjP zXqw3YER1f@d2RaU3y|69!FPXvQ`CjY9C4b_!6`~V$v+Ps{4T>9T?mIUZ2mhKbShnfS(w*kx1Z+dy#B>t;5NUrL0Y4s=-F&_ z@AcDO{8^JiaIK&dOhcf;%9+bUqo@0hS#5v3?`Z#?_pZx~N24=2E3R+P=XHWKFlJ9Y zPM@gqQ{FAq`3cu$9V?X5(MO3*tm^O^?O>p_y(WHz71z*Rv<-aG2xl9114v26W5Ckw z@Ad)_qAgrA2Fxw-7gz8=u(+N1kHQypDvF3S?|P zmE#w0Nd2D_IEgjn|if%2r?oU!0G1k-C_@vxO;RoR=g;<+rHRt%f z1`gV;E`SRn!{;*nL066%`Z>9{n9=lFs`S3z_FZYYJyC(uBLvKxQy&+#;gR|3W>iHIXzMrPy`6J*b~luP7>K- z>9V#v^X;>L{XkJ1ht|Akfb1%J@oBy{?c#!Z>53EQuwoW7H^@A|kA6iR{;JpDX->c0 zKhIuzxy_rG{XU3c>v*c%fI6dMNF;)A#Y-=eo##Ig?+z$Bvuy?bX*~?=Na(MW6l{(R zKS>oD2d0HNg8+}t08Sd8EQ8nMtFzB=`F#T>&>;_ctX~d0Vr4ccPg}>~kfHD~P>gkV zL)~$2A5}r_FtjFTPzoMi7H9(nk8>N7Zq9%oQWL*vJ+!wT3S^h*419)lnko^HPOdcC z7B&YGL3I@f#5AJXM8qon&s%An$`u%d;{*o#=dWr0!+=)uW^`{)*-}4-MMelKT9y5X z$RveWkx)TW4xS}7K2WyTXO#Rh`WKn!f)WiZiA76E12XKHuj}>cUyZ!U{Dr>Jm|GC& zKbwDo4yRDp2pe;71a85;(i8pZ_;DQYfXcOXd$$6diqO_zIvI@?N-}{=MM6VC!^L{s z{PWN2W}A>OP#buQqKGwT*HCEfTAA&8^*7Io9Wr!_zA^~LYL23iM05iVN*y>}H|ONd^r%9>Fm~npk1s?L8>o z5XHMA|70QP^$tM36Cgm2s?m`fb7F2>JZ3T^nKQ_N4wbNG2+b%$05%%a;WeyF7q27yk`&n2}+-kiYvjGRZ3pP9_-YPxw@h%f0pG<2jKjpU?)(=nfcLe^*C4-&0q8X(=bkM``* zRD7=;BQ$_0*q}DMIV-=F{1gwDtWN9o&ns{S{K8h<1tQ=q%YZf*-TM?Uf!sb2m^w_q zpBDjUCP2TfbGc)aG8OXj<6*SE(6cZ|4fvE?99`_r+px2foGTANxR~k3vQD`F_Vwqk zNu-SPXyWR_IsZE+uyNb?qk|2b7dpb>_o1}e>{dgbX96xem5PR|>(Zvy!cM!|HhXqY zg$5$8TsGxd^&0)?7Qm?8Z_|JMt(#WEB;QUdCOeGcb}f~jxyAE$;_34VEiU|S_W0GL zl$tvC9_EjafZTf_X2PD4#smaDKbqf-6UY9A=npK|Fq64B0v9JU#D8PSmuQH5Nj8Pl zjT~tl$Dyn?RWaVG-M{@u*D9FLG-#d1>BWwn`d_aoo&w<4Y@SMIl_2`exieEcH5w?|y41GZ;R2kc~PXGn;fBj;^qi(sva;h`hmzFVFqu)~f!Iw#=k^>GSQ5M!8h|`DHp1f+ zo)lT#K5}IpR(2>?J%L+yEKd|rma7L~WR?4}kneQ`-e5BE^su!5nCayht4G{-_&gB^ ze|Cjmp6bcZQ8lWLU~#d6 zzYIo$?$cK>)y(RN6w{d%m~# z-D?NWT1}E9smf2A9y39g(XAhV2}e3#+sOP3SJs+w z)axt4E&gm66V8JIt_7J-G5x;QHSBCcgh1;r9fVv;W#mu??@|WA=?*Yfy*DaDvzA01 z#aa(gET@km4RAg0%XMi6V{{3`>`MW!oP*ezn%Cx4m=z@HnY<)PaQ^~Nl$GJrV0_wK|v4~11U2YXh;ivk)pT@iqZHWF&$ zqkG)=N2%__Q-Gskk=j;vtaauGs4uvKZ$CW67e$Tup%@whq2TcSxln}`fr%A={FpB1M+IXPwZ_Rs&aqc4L zlyQH)>p3}JaMY7@if`6}ft>PZ77iWbsf%X`R=PD z|2QgjmfQ?*5cOJ^S5=h4dQ1-sIzm%o#j#LPG^3RlSz^g=bbdB-l0qCXrtuC^i7Wd> zsVC+2=sjG(Bb5r#?+;O1n63|ZY8JWqu}ORqq5Vol8I6~CE2O0pqkC-2H}sLXAI={Q z^+lVIbBEQj)$^;--;H?Qcfi{i%zetH)m06yr27^mLq$^D*_7xFES2!FU;8W`3yGp698z!jQ4Zd`c2bRninxK^0+4oZq#ZLLP^K|QV>^HLKq11` z3KkP!@Q`j+|8s*@S~qvHH0-Nw&Uhp&v>64nT9PA!&^Km0nw|r4`lAyLZ!j2(zy-5# zpE|lN%s(cvawYh_<`oK+%ODK+zMpZQNg5EDe~o4{M0|T0Q?-4i<~%-uaJ#$z*1?UG zHUugqh+Ub_S`8v>&uh8*l*GSt=9;Isyii7Tdd-x&;$IZ#6!RuDwIW)heQz4Jre#8; z4fN9|5{|XCNLULgh&?J8h7|?G!BN_4v|`b?vPPSG== z4v+UhAR#-(sQLq1d1rPF_iry8wXpbJc4 z8>q{YaXg-vd(8`Hz*AK_d)c3XW=`4vi2Kr>SGgs-@BJ%Y-LFJqo}ws{ZEUd3XfRV0 zwajF|fSIRO{(EZs+}rAW=vEXZ%8~AS&h_KRhP`W7tyQ(EFq=zw*zV1w%+mdmFHAs0 z$@6_DSYAzAToLc*-Vi_hxUu9tf4mtRPUB0LTYt36_xFCyb=ZSPPcmX8lVj{B^v=I# zbbi+Gy#`wjoZ;bi)gwoIPmx%mHk+NrGTspS-C`4yW1(a{@*-w_Y5zKno9zxUHSw-cvJ0Rvo4Tw=E$S|zy4>dt;hc#l(k2>2YN&t6^RUoMNFOI&iw-!$->w_gsQ6!@siq|LW; zKCEE_9j3!)`{cm=b=SrddCpqeb1@8~=%T~sp1L(5 z7YjJP?=~E!UcG1tBUToJo)pdV*X!lP+VAg4x{zb&b*|Br{f!g7@8yY}McY|ld*Jz0 zD6H$VyPXT~NpJ~-V8GHQY|NfbFFK(U&ON@D=#F?|%q}i^m zKBIBzL%^Gg20UNlD@P)^cnMr1(r@#3RI3~&N}U!aHSZVA5=%QLhgzQ=%yRd|cbc1PzjtHjoJulk&yAA!Gn*1jZl36AFr+n$wO#|eWHbc<$w zq6=L=xYA#31MF%nz$8XE+I&S(_)d#yaMrGz{nSpv6AgSI??`)7OZgh~FJNpCn>FdS z*^S~7%5)FQ2t7hQW+?tAGk>#YaY|qA9lNbq1OaFc>(JlI#iGc+SpEB2PpUkwt`&vGbufbk| zSW7)&HNRhVvS@=b_*<8aL^5P1P-w#(b^hd`x$qSr#nEKh-q+dpCh0soKc^BnCdKQ; zIln%EyB*HWZOdX>-!{kBfpZ?EfQ0lNUDo{fg1iNY-pzd%2nz zF?#O%iMPxh3H6i9C%S9O#!Ef#X+Zo82`$LCqv>mg}2Bxh(7To#pc?ZOxUx z7UbSu?r@F;+-c7r2RG4kC-r@!hp1lOO+B$WV=+<5}hyKzbvw;mG}(v z*S)t0(Z|>R`ekFFY}N&x#@8t4NAxJ=k4kc?O^juDE+3fMuz=3Ln?b(B zq~^rexX$2>fe!64rJFd{C!1 zWfy{ThEG+{;E()@dLbxPZ%SGYk+z$O<1E`Pc48>4ygwi$0Tgf=hvasZRUi&G!a zHJ;5w*b(Y32%2W1XIh8(Wc36f;eIDzJ!=d*OsMZ{Qj}6KGkO&xuPR!%nm^&3$IV5w46lnfLc*DiliqiVx48K)!>nOxgV$52HGZ^19- zj{p|sMKC4i>#G6}zRJ*r)x88LUl3Y=18xvEMj=C^Z)Y1nyRS=nw2{W~y?m`3kg7e$ zFdS!jAZ)^(nfc;3`}(k1z>N;Lmju;Na&EL+AC|Cs!Ryp3B72e-sa-Xdozcz!cZ%nr zUe{!zU$DQifpQ%cHoyI=rbtu-d86~2;;p~4&wC7Ap(exg$r<=eh z3=2Pdt*E(F$_1>{F7NLd_GxhBwZDxN^5f%v&8@+A%?Kt%f$0<6j&;8Lzfnr<4! zxT@8Eln-PdZRwALrYTDCFd`)8gScV_laDmSpBwgKQy%c=9n6#VQ~>I5zyX~N*$#0P z|C0w2M6KRmI?^2rnCJU{Fu?EJVR$q;5!a;aM?22Dh`aR`9zKoJ?r(%4n4YHT+FY5z zJ>We?Z)`jhd`rSX_Ij;;T#1a~`+3HUx@qu|AP7<4W*XL8VBo-Qp&+F5Toxw*a21lb zK8B;G0;Y8#qRWb09_!fjClBRBfY7H+CyKY&mNv;{&CaV@=Yru%iQ%ShB0D({r}o+T zC~hhm>`4QtGCh%F%xh zo@z*NU-LlmnMp~c_Ty_>_knW_{)&lb@Q;TC^&CL3%!%?^s3+s5ei6s*E~or3&lLYc zP_sQhKEd`Ht?$;xvqV|CuSE0MfL|S4W;l(2HSJVOU*v8{PD|ho?PpmFILLV)e_H&t zn3^LEew9n$e8DoQrvu6nANnJ~xp#;Y>ldKgTDCYV48trtB&SPYoQKo)897gNr@&~Q z9T@uVMA2rnl!5w5DD#aOQW5xYSO$niLEnjM)JdVk&dw*oj#zt^F ztgd`8c#Z7H>S)6lu;E|(2O^ci#3aiE3UQ<_kX!Z>%8$!QOdA>DU&Z03z5nSb75zyvWX%>;H{LzpxGw4?g6r z$g z1%{#(5VJBgjh03HI&0flr?8yrO6PtXXF9kxRmt-pfID zBmrj=jLIJ*3VbRUM#3x-UCFH&}Ppw4;l40GX=YVzT5 z0cZ0?y32e4{II-q@4IAKPx!aQrSwslk>~Et-%|}Bs(5zyZx4YtYrNhEhmKridHvWr zi>01VMaTTW#HPZ9aMn4qcQ3Lt&wbRmt#v*eP+pN8$9Wd~lx|tX4<6sMEcwcyUAXHT z^{_t4JdjlgnxBXKsP46MtR0dk8)0!>q^;F!`c>Vr4PoKO3-j++Oya(UV3_Z8J}LRclqcP9E) zfz1UWyFAZFQ+fga{>7ljcL|mlJUniQfCAX5?+XSqrMaQ`Qa6dTBot}oH4$}+7(4iY z&n{q@$)YpJx$)%Rp5QOWmA_7Efiz0p7GsI=!9IL~F$9%Gz{I@KV-r}L@G*L=yR}yP z!HBz2;+aR~-#e;%Ik(#+jWj9lh$2bT>dI>59HQFmPa_n<-MYV=+2j0B20EI2gkJ{*Y^d30ZF_%C&W#^cSX+VX{@O||8mJlBP|E#E1BT%e za7zmj>D>e31H^;7wWQ|pJ=T1(b*@^81siV}fJ&m&0m{B*u!~ty6k4$08jlOWuwR{r zR?=mCLtK74e*{0R+eoZfmsytG3}#Z_Cq0@cATmBW%WRrv8O^`lnEO6o#!pjHtikxx zQStk}mqTneTcC@gpaEV+I-xMbF+AJXb;V-C2V2{5fjB*G*34leAxp;`j5PLuS8Gg? ze!*5=V=w^MiFT5#2G)K1b!bKdl>tsAaPUn6?*CE*0J-Z6@I3r1RnNR=gMwhCckFC~ zm%K47wHy>5f10oyl_Mp64@e|6Sm|b-&CJ>?@PU*HKgTdpTsagEPV~s72nQ0p4v2CG zqVpvSLcsCR0!$Jj5pXrDn~)l>Fr><8tSOA#tA#nA{+B-JP@6o}d35Bd^o0_rX${!+ zdWO%4tf_H(B)>gbgSz3ej9rMX3R)Cb*HC*~u&LiSQ2G{pcQpi^_9f*7<+!oZX%`$^ zF0O3Cq@s8Q7_a7Vm!q4qlu%8KzO3=VT4rI z>srb%Fodr4s%0CVuikqBDrlbQ=F(prMx2mA_IR=g{pMUZZOq05@WZ`1?V)`+^C<7| zI7Bi#_S^nI_)fshT~lrsHuS zStkeMbT-s=xRhkwe&*q7Sz&R*DR@7crJS(S*PQr?XPy>%l16h_KRoUpPU`U>hOO5L zKAih=q*L6Rz?uz@Pp|X+_I*KFE6fae}em=^O^TucUQrD1~jFvR|=6G(3%}e0=tm>UjuXBNBmPaQCJx}wj z9mij4|H*r9Gk#dqWZ(Hncqq>U`g5)>&c~jt`gHo+fBL7tet-Tp zrt65e)vIX!_MiXyZ-0OFufP45OYxri^sk}V)4w=|0J2H^wG?Ci7x~v=O^5&bx4*yj zzyJCBfBgLe|5m5tD|wAC__Tld*B8Ij*iPd=k6W9<>wim%0lsnj=imGO>EHbNzx@6E z|M-A^Tx3zd-YMQTZ}i)Rdo&&kahKm;bXU6l%+Jp^{HxI5Kk{%o1n711c8_!P55s?4 zJmuj;R7fP$WecsN9{7(Xl0_nttKzDJRov)pL@~Qsz9+3%-Y?}r#EQnLn>ha4_o}tH zg`4okaQ=)w`{W<5o{zp)MBilwmi#rjq&*yt-IJ)!miu5DgH951_1Fwoge23CRee8o z@2>>NAg>}LuKgFdXy?uw6>Gmqo1pN}iL_^k+O|Vo7JE|x-v7X&Q={gVay|6cV>`qO zi#O`qxmeGmByK{(;7aOC8O!rVmhjCy$(YHmdy~$Aq-o2qEEdVvkN7;IT6n7*9nHR? zen{_Go>4DdQB~I!{e5ji`~U1`LHE@BYgc(ar`zor2vQZ_>S4ctk6Z*l!O=XDmGYLb z?ZEG2QEPnM^W)G(hsx6$z&T|JJZMIm9XuTx-|H^2RNvkK#y>bQZ6WS3k8`L1f$+P7 z-aO`p}gjCUZ9XuA_IF_S0Y7txFe}E7GYM#Uq#^zf09(8IRWUOwZ8<7oEpG zqr5cw3aD$QLy_W=czfx=gm7*2(;9VbV&Kt4`?It270r|E9q|4ox4tf?1LYRMV4^zk z&d6tMa`4E#oPURUJLwH@s2b&G-Umy0kUILE;)=%1{L;Bplyt-|_XYTfZRgM*vQuu2 zH|M;4;2j;3>}ztQ$yX*d;~tIR0^XzC`X}+*M)J@&)p8`iZaUuq;A_{Z&V z1;!->vp>)X<4hR8OUHpK)htcl!+Y7}{1@}N$L%b1z-BXGfM$zw{A!IIVQt`Btz`w6 z5fmy)IHP*%Eu^wWx{7Q{57P5~IUNy+ZTy5KKC^%4FSLkdTl?$>3RbjTlna_A7@v+s zi-djMqi%cDu&zw0VHg}At^=*B&$Ri8a{3#N(<~=WD@zDMj_gW)c;5icQE?OrDBHI? zy^*zj-EIj`o}>N1hs_Sah?%CF$wc@xPSx`%1kUmV4%7QB;zr|`(I|&XyejARyI&N` z5{8d_KUPAjZGcTe4EF5br-Rq&4^ypPPE|)mO5@@>T>h@_>}N%$#xM*)W2UOAd&P;T z@=~l>GmhgF2!P`f8@6JOIC^H7APGnB>!OAIet_qVE2`0GiZgjd{_NQys?iOd4&ET6 zbNWyehLp+@-Tb*ceyJOJ@vL5+Z=*le?aN(x{k>+&DhzrxxI>H#@1aTu?Tzpoc}D}B z;zm+4oIJl_-T%xI$ZthrE$F@MDfn?Wz)JVftQdHYLVOc=ZRfO}!{*id>AXMp^$L{f zeLF=>esit+Lz3;RE@e_o{53$=_ddp)HeIIvaR2xGL-Dn;XLJ@GwY!2;$fi=R9b0Zk*@^CNSr3?})rlyr~uhhLo7yvMl zeL8$@ON_qa%fPkUQplxEUvN{SeuT^yw+qElbsjv<-dwE{Ae=W) z&a-oV7Ojl-YQ3nPwy7*JSXD~#E%uHQ*Qv~=$?E;~(G;n$G%7P#t=WxLl{BT~xL3w9 zL7ZU*@C~tmST&NI_#%xl)C-2vTQ4)4ZUu@L`mPwCgnh8&yuYI!AKA&5;1<1Lqs6+f zc6PmjpOi~6jXVLfE5!G}(ge(%w~4BEDl1u*#amc1;Cw|o-vjZwU1=r)zSPgPf~WV3 zJuQ^CQ!2|&bkwaOfqSQJ3gNm79>cP9JFE2>uylEWGqKk@E1J%E(R6S2W;x)O{S+Xr zixYl{C7C0c7V&IFdEnw>hx*26kEIr9nO3Vk#;y3gB?ii=H5Wz-OLliJjV`ltb3vo^f`aJt8JL(89l7QUts?b z;e&7C<2nIXx@Z=h3G~P({mu}nnRTzDd()i54#Ydf#X#Qd*t^j{|D`^u z1Og2&LKv~o?26&M;z#mkKWp|@4bs7qemogGhcTzq4usD9&7?7byB{v6;&G@{SovL? zgZg@s-c={i{~i@aMdbC$v^F_tU#M(g`1P z>ziujqjIyzY~mAC&m2aNkIt(QD@b35INE~d36S*~ygk#z83w*%rR0T*q#!~ygX46{ zGMvj@aF>@8`MR8hSrt1v5^!+M!`>U20Z;6|H8C)bIS^oj>lf7^Q`4V=K|V{; zW1v=jj%pm@6zapgHvOCjy~}YQ)VDpU6MtTpc-O#wgsrmK_z)AggE)>uKL}lVMP(vc zggjs?wub}W)HQ|w^7N_Uxh_>cyCZP1*M;E6EkyjEh>$+{5YHIOj}2KEch4oElO_Uu z!{5Gyt@6TVA%Qra(a5j;{L34d?^YfMS{hCmKhM*&iBdJ}=|l}ki&LSQckJ&vAN7DM zLBI${Hs_A0b{|i`=gIrWJ}PpV6Y7E_P+S!Z_6wz>jl>a9_39Ld0wDg&qxoA9C2t^D$arr>ZGt9WYcn^KY- zrRru~B6;=iSv!eAXEG%Y5waVl>AzQ`y(n+V4(C?p%f9c2uiq~W>g&{HJZ&^!npDC> zFP_9Io^5bSrw5au1uDzt?p76K#7utm2;n$@5l1a0?yrlKW3c`Y9pdk+g6LZOv?t3` zy;RFCqP(?O;Mx91+cSzJC1b$9UwSxDFF}Zdh*kg}J{qs~neW{vP0mm+1y}j~3NfJN zh}Y=dfgGE#E4hVmEAyKB4Ff~HUGtu#T%MN?3izbn$9ET32U(H2%f@OIoBC)b(^W=w zomU~&;5*|DX@619>NVp+Emixikx0Koy~EL?tDe0n5h9vr!`$npDXy!QC|GK?DUAb8 zT(wv^uLtWOiP-F`f^cV<+k+1JIu;G{+U-nud{U-TQUFec^a+k*h0s_-tUg`V#P^!l zc&d>W`L|BUFC3;2FX|&^7<)3Y39R+!g7z$bqlfjyhLWJXS~~R|H+YdDhY&gP9+tBm zd7h_yn#S=4)tE_+Eal|XxAF?OzMKqzK}>;W zPA=2P_B~I_?HLfsWunwGoIcBjHNIZWlEKjIJ=sR(7M9~q@XL?>Lf$K<=bPjKo6|lG zMJjiSlEHVRp3ehTt9ceISo6TOu zV%VFZV4C}Zl3){SK@?d&Gu7Kow0oDq5sce;o~DP|RHOl%c2nYbmi_wYe|lS#znc7o ziM7~lmOxWQ&1O+{Z-M3M9 zgxzm0nTg|qwrH{>OF*t5uqKB5n@56|Wl0*fbFBE-@)wlf(J19~JVC_LPD>E$412pM zzuFj@WdX5G#bKYQEMLC9zw35xAyvg}?wFd>M$oxi@i45PFazhl z=9rI(x5Yphzq!Q#BJ^7KeOR&%d}cQ0gN`(5)4mhm$wIj6rxO+=!Fn^dlc!98tpI>@ zuC(*a+k|h!P21jt;}hacn7k%Go*~w1doK+LC(kp!+2_K&Q${Qb+Ch?&77!|zpw|Wk z?(@AXt1EJx?Z3tvV$Ah$JYZ!yJ0pA|S0Sn0+pIDXduCykaopW75w;E>{@oL}-mdlf zxI?-Fz=lhJY;75EQ$k1uA16WEf3*ap#S6`rnZTE;S2~h6QT^KE~2KUzyWxVm6X<$PTRKHKN%} zZRYR}e8vz{JOLWAh2?Q!&+V+QGnhZ#z-NoW0Pa^0lW=M*0C(Yt_)F2Atn=XnAm^sO zM*F)25F-(;&@b-BKRVLxCERPAc2Z_u!0(kkffr_Zts%vpwo7rB5_`kMhn5O1%=`S z{_O$o(XF{euG34Qan4EKK1ic-Mx51_e;Bf#C+!O~3yW?y7W+65?Pt^XJTXgpi(6hI zTr?^f==V?IlPRiMq8Jb)>Ev5LI6sk%G(nO!Cg}tF;WF9!J~S1NCpw#*pB7Le#mJwv zUff)oM|p;E6g7O%uQ;QdjRH1APVp7R3!hGb==J1v52&h! z7#~DvK0o-@N7`;4BCcqLxiriW3-PeMF-6%$MD}h0c5N-x4k}$tuyJpmgdJ=N z{IjPiXa-TJ*IVe?v)@9%!FjftJn`8jM&slnR@&f_zbyp0ImA)_jN~frGMnr6@T+5N z*utpV1hB5eLOo^}`_T;`CK@+MJoYo{`; z(}eBZIS*XeYm(OB>`6NM)N!CI4mq|9j`5q7cs2qQJtfBSuZO~~K zXGNrw@?AH82yd=Pmx3YE7sJ9l-uS$;V8vVR0#3|4Z@!1i3+FI5fHjxz?j5K)aX%h$ z6Xjf2h<&tHOK0gI#q`Jxh5(Mt{gQB+_^?>)1Hz|p*g`w=JJOQBPY4q-P^?Xr;wnq$ zTSl_K8eBSi{j2i&M~@@zl2c}LJzjbL!efVMt@lR@0c822zQ3x)o6j>&gBW0QyhY#w zyLH*6tHfYRC1yU#6Aj)=_lC_>S)}jhUrY4nun|?<6U=lR2XBzxi$%TuhEH?4e8l7n z-hVED+)2`K@34FUP(9KB1V;a60$>@$29yiD7h+72HD&X&vlt7i`|s_-eyAElG5U9w z#S>tzq$BUeapDIa>0@^NP!D2E`)-si+;TlxOz6ug-5#kH)}XgHZU?Qs4livj6!iat=X4Gfd`m1qrYYdNdj07@VBgX#L|=qvGcmmH z`?a|+ni5VOf3X%pe5lDpr$5Xx&-1625$!?uBy1T(dn!u$K-vXOcJ60Xd#(RyF0b1; z2MkhGznrvmL{%pM?E4$daSGs?=@bl?iP;zntjqE7;RP#1`8kwhD*{SqL%@uElc#L# zSWKV=X$0|ZgT3$oFT`6AU(euh+Z#P(Uw!vRZ}d^_E6w`0b^& z6nLk{8{H>v!hPbu95`xo$&%>Z`feV`7c>_JH5QQt^2uoC(Us7Et(Ke^Q69qKo^x54 z^8qjwtD)InXhu#rHyaz)Shzw+{pn;r;17!YVqI;tvl{SwT3rG&B3b4ge#sY*RXg|73>QUk zV&6|XM~Dc%9!Pq5vaU}9d{=C&!PotwXRPEOj`h6D?nNaJy+`g!{4b3ULH9&1*03U|~(p2sYa9UjbWR|o5 zJ1$*}$L0GM4%fOjP2b(q-tUqB zI^=LJn;IgJTfOiZ>2`X$OMkW+%CGu^X1fqfJXu$SPeyt6%kRgzk=!1IpDncBy5s?o zl@#hpmkz%OFa1=ioUq-Q*ig>v(_I}fiK~${D+Zj02Mm;Iv;tpiEie5FG%rK*&>rWN zVUj=iK6Ks>z&Zzof5j|t&^2GbOXy-WiIQgl(ll6brPm|MP2_CqUk(dc+10w8=^Ker zoV>^Rz1&ag&9*)ae03-am%A8U3sUC+h6(oN@pP}RRo$Eh1G2JVaBP2^0UJv2Inu;( z4W@^9#(PbEejN8B((+EzG)zTqwaak7obaY7Hk?7ctG_bUwY^sNPk2_eBVZi*VqloC zfa!)@WqmCb?p?r~oo8F!TLC+74-MkT!3a!$qR7YPjHdPMU4Tj&51ac0ze3b=YX8)! zE^7C9|L^@Qs%}Fhui!S9_ZhW|D>?H713pQ5UAiUGA~%wV(-Uw$FZw;pBa$K>HS=!= z%`~;s5oqraq24KsGE=(c5T4yW@Uz%Pv0yVV8K*ng)Erq6;!8BUpW^eyKix5BIgX-H ztxP&DjG*=-;%1iUb#C^X2M;mnJ9z5hed7jqNH+Mc^p`)Iahj4TaPW2*jrE$N+3?bl z;40LYhMPRki;?0e{^J?S@dU>L`SjQ@6leaT{8av#JRAeRziIXD8$k>yejE?=j{uyl zB{Z`HjC=CIpZW=eaCHF7&a&p+>x6q2+B#A{0WgR1aAQTmn&{W~!lMZN?K?A#gqB&i z!2Np@Fej^paGw4@Sq z0EKz}GmnF(7C2MYsdozs468aJ?c~1Gv~sO)MHVTcz{N!bw}YD~MSyDJOzMMwxN2q$ zEH#aIhIx-Ci_n1dTfhnT#rg&_j`~b=mSGkW)q^hx=d*F=GAQ>rDd*y$6a2V@5hc#Q zxiP@f00O9u)RM)7LYaZM-=CcIYw%DF!4F>)ntc3u|7I-$qlkaHZ;bLi0&f`Xcz9mU zzyfOzfHK1)k+fXsN?ALo&kRA(I=yp;_u+H`_Bwo6Ec4kgOV`V4N{L&ZtokKgN`S6I53m5`%tyLIF#W#Ihj3-a-fc&>6hj zZ;sw0>o1cH@Zwy+k);_iNir$|Za@H^d02o~q7TxiZv)5Y_q4e!61sFha$qG62W_FG zbXJ-$ZJ5X1U8gQb!VZAwX!a}WK`E1ALOnnoY_Ug2RRe*n_BaR4;eUT_r^l!MWNU%? z73FJq{OPNQWk&S^dgRko8F`I@^JdP%+tQTZJQxCe1J$@)?#{ER$ZnF_s8AoIJp{Hl7ACAYgA;JT2u z5tu6Q?%yCBlp|j|Hkp`0Ef(VqB*~sl$F@a#^w{`*vHJzVTY^KqIBP;}0rCS7LlXy` z*N)_*`-&j=X5XIUhI&XP(}3M2yNsiRIYPa@5R1onzJ8g8TC;n-qd~U=9Z`{#zsc~G z3z1BWs1H(31Nr!l9eJ%^+^*&;v%dH6vr=E?Qwrw&v@L8-n>UgbP&^i>bN0^`q z3+x3!i(-o-q}5}X6k9AO)IGm)&%ZNj9bVESTngXO2kMRWZI z*}0v|MKH~4cSb`Ps0?ud3R_?=!EfG+bqwh){ zcx$MA?BBdybbe{1(MV^k&hoG&#FUK!>hd;}o!5}&Ej7BfrOUE(;sE^XAb#)HO#uii zVAqlnOvL?n=`^DnC0bv~$Y^f#RKyQ8ijNIV8k__ZQr}Ed^c4ZwqWJ*E2sVc6-z>AO1dtxXyss%|A#2qw@ zN}4j$Ku>v!@(rB4x82P2dbTxy$FK~@Q^f%0xin3`UpW6?OIfju!Qxc+zRy27ds;7F zr}tInc;0z=YDM0WhCkk9wvgUiONjuh>>Ytr<#w7JTl)_* zk1A`udHxxIDV^Po!Sx7fmF^8i5T+hNoJ@*q`BA|*XfieAmPibTiynp5G&d>{vWy-{ zaYyUQaeZII$J!?6WF6I^L(72GWde0{c>VsKQH&WM0&kTUvtwGeoNvF7H-(lj(TddyUGKuKUxDf9J>EITF?J~ zKijHc)f&HZW#Efn3KdQ<;DM;!aR+*La`f43q>obzlow>`B53*gLY%}u$Ld(M{vFtS zf@-VuX&J7Ibk%;njR^k^6dLskctKJ05ZqN3)LN-J*(g?n@f-$p1ybQg>$7Ok6kYl# z@JX?vMv!XXaeP;;U`VO~!A;scS!lgknu?t=pAH z5o;VEd9y;2BJQAG*biW?$-jPV)A!>P;=8#BxUoCiIlm^gen)Hj_RhO=(60NS-cBea z6N%%mf++!UKNwk-fr};qkE_;ua@V{4?p^SECCb;$CiqUarfj0UUkktzK7(Iv`H1Qa z%=CK<0l7cfczZsb3foxDdn&a1WG^iE^>YsIJHZNI8NE>W%$shwI0w|1j-2!%IlV1Q z`*9#NKM+f8$Duf;zfb)g_dC)vv$>)DD9tAR%kdLE0^X^A+QiDx*9Xefm+G)aMM~il%=G8l(VNtS7{6D5*N>(-S0BE{ZtP*6PF{Sav$ygFZvDje<&Z4S znLZkJ@UpB^p2$=Ge1m~}&mSk&??DRJ_8G{6HPMWIl-q+yc`T7X{33dQW(cqyeE0kM zHp=g-6(%e&d6r#qPMkh-zBfy!~X-Z1hJHeMoB0mqd&aEQnG)Fnegok*(+_~l%tQxz!Wa1bd9uJ18Dsoc#Q zIxRBUGf0^~ZQ0#5QZXwf%XcxZXAXoS4X4_Pj)NRKO3jd?-j{gRl)LjP2B%{Rz__1% z@0Q|yU|RdJW4LuAO|iX|5gb!L`C#;@6m}6WxKW)dT(s_fLr_k0*l>Jdvdpq^{QZUh zl{-5q`!1M$PZ*DDpG+3q!>U}l*9i~V#uQ|okds$-EO%p(trzc&xLV$7e0Zr@ArC&9 z`x`Twp?DIO=>?`ZEtQ`qKQM0R?lSmphZ_P#Sm#g?EWx{m=keg4EK|aAnYi(zZ%S(@ zazRPH3I0MG$IuogI~|^}HrRPEyro+e=mPfAnRxwT3a$v}=ksxnzw-SY*W33(Jd)!~ zoOG@!8)Ng2W#*Vvc);W_#0HXGe(--oacJX4UKT`F~_#;h&CV6l1o@Exl>&dzpuJj=M& zVDF?P!Hh?(omOu4nOC|tH?v{bnBpQSWZ1k42J3#m1c3Q41&)ql0o4f`*jDe%Uy93` z1;vfv-|HGbw)uJU*ofH6PnY;~htV(B*ss)h1*a)5ElGF90etqwx_^RC%mlDcWGzn5 z>#0^JS=6sW&s`XPeRfOOlWbp>#6z)!2~L^To8RebYjPQUtWBro!mwV~dYuLFXg0+= zOR@J_ZEYd77jHOSE)U}Vbyv-av3!A8b8Fn|ejPJf)R!fTMlRJ4&1vCi_1SlX7oN`f zpr&*BVs6*^MozG4;U#Xp+qrL|^$K`32)HNJ)eiMio-g3sh2dU!iSZ9S+U4y=Si&yW*#K+@0kU zV;?&)s-hY1k+c%mNT9kF)*FIU6Onq&(Vcn*e`=q(uY6Mg!OIsT3YCw2UD8N(k>+Nj zDUnLN?o6<4=RbsCJ76m#hRU--IYhh2&+E;Oumq1ZPQTQ-s0h0pl_De0F~1giv=@%3 zw~m6vAJ!P0;t6mnIj%twptCVa#n(MlX{srm4wPhz4F{*k5!|MLADVb#G~R4zh5NjV zYjbr_4cGu2m3Em6u}-sM<_?JKtrg2F%!6D^L(vI5awezPhUhwc=1XxalrZL!_ZI5R z^{n(~&WWrz(y6R`6T%@j6TZi8;&qIQ#|eCM4v6;MAth=q2)IsRroytq`iO4ro*Ghd zC*!k#?IHEX`HwqvkFVuC?a%jeSf677@gr@!1ySfz3@7mUYV%@ag64nd|7}8KgBbA ztMQ&@;{YqLl7Powr12#t??!X<)Q)CIFI9v;#*xlvH978Lss*=RoE{(X^6JTAD4yf- zyAyeyWX>1B?M0DKb8(~fbyS-NZKnPDuD)|*vA4klw*G;LN`njg3`&n@nbFN^mPtTQ zJ(j3llK9GlWFI{3@G@b?uQw`+;Fp5p>R++@C%%5QL>5j*{s0cm3x`jMQ5;^A#K;Th zRt+eMUtme@a8BnsC|c#jU5A;`mN4cWH^sQs_)w{ru=DNOToiA z2Dtd*xcw_zv!lEp4$X0=-A%fYEMDQy_y|#o3YW*3_BZ93pC6ku85lu)NIrZY(mgrR zEGFoA>2}#I#-|q@d+`4QrX{&zf7y?)JOlVp+2Y$hmO8R;rryvHrFoBRtIK=}Yn{c( z^Cx|HIeBNUuK{r5)Sj=!ZOc92@}$u?olEga&j7G3P+dAz-ppx@r`ce(7>l$oz`YLD z#x5v7OCSlJ^%YL`fu8AQyjN^1&3HCOP5Sa|dwHz4Z{xq6w>UNPGkts-%>#7leB$xb zJ%;Xik9FXN^aahf_&v81wFSR!|B6TJiT3-7R7H$Wetp>FL`C-tLaqdt+%ITc3QM%i z_KHli)bX(vOg3)A8hd&eKbOn8^=_Wg0+xLB)Wv^2E1{x>_aQ)Y)BtO8$?1SJT=&Wz z_Fo^nN{Lto5SPQlY*s&8ght0PKOG#ZIsaznqMDI_J@2l*-J8On^)EeiLW^LMK=nhAf}6tcbS_ns}BB=n)$> zb$kcckVP}SRxOhKCQi_^+!xegyK|TP`uOM^@?zp~?_YGfO2GAx>1QfUWxyWta2e9rZj%Uxx+*5`3P8nfSH8GJWZB^aLa`o?NxB1sPcYs`{K&5^)NDTiQ! zSyOg6WaUqNLg~t6y=KQdp(mzIuEP_WW$0mp9imil{`kv@ELM%yOQb{}Wx({NTs}qx z3BKyvf)#q;zQ6Yq9Bef;!V{Wdg?cJTl2TBMFCJ`(c&@gqO3j;2(x3JCT*vQ##i#U= zQ`dgf&J|{T>P8Uu>DR||@xQzxKP%N1W^2aujY*YUy-=L48N%&3d7<7b5A$~L@>akj zmz~yw7PmpA=R=QuiO(L_;m>BoO?KU&Ip^n=RBvv1STeI+(Ol70FpfvWDb`uamtSAR zQ%q;;@I{&QH@t5I|2v!x2YY>u*Yek+1}u)&=$?2<;Y6QH+$tS8Pkpby0#+9k<+!)> zSho7yoICxxjK2Rp5$3%doOg5k#Vq?HXsxIDs(>k|udcnI9QjL!dwYPhm8F##{5XL> zhx0>uphiMUpK1QlK6zdgtLaw)L2Hj0vZAnWKG_)ZB}N+kqOASy&lkU0tV|+CDOAl_ zZ+0wZU2Zu)0wHZp*3{q(UALS4lKpQrwM~u_pGjJ0d;CV*!J~5+-wa{(N&|8aP(sFfM6RnDB<6$R5A$m0`O*-}iR^ znxW8yi+&r@`EWZJTr9p>mXOE@35IeW#y4;P7sdc{-cy>hHC6{d0P{c~Q1BV9H;Z;Wf_tYe zd0R$T?z>k|3lB6j|I6K9+5n5IHiOwOw$0d75|Qin9Q&iq7REc0{fgwT_%3z({OF@y zUP{7nkF-`_8rmP9C z*L?@~Ow))TT$u=hWZL5JevjDG=oYL9=K!!c)x|v#s#yZb8?4^N8{nQz^ZVI0Bdk)i z-=cs%;JHp_PZE{0mW;*Wf^!U?-iqaNOz?p9rh7!_Qt`S`_>LejaYK;kZp;&)U!=@<4qGumeJEob)I4i#B#oI zh3aX&@bI7fIf&0%<3GGmmGFdL$~G1arwRF1c_mjS#M4hcS-yA` z&b;;nIFrnOSeb4=Ea$t&yapXrs1PR&V-u?Bf8LGQ=7ATjeWJ8f*7jEh0za{BaQ~Gb zzu#FDzc(vUIS8 zGlVlVtqQ_QuQf11Y7oO`yPLt{wG0Q9^9-w=5z1O09HGuT5%)ONcp3%xIy@^^o5-Kq zLxGy5u-nPP|FrA|)#=lsC^Fcu&rFks6mauZcSD-}(@pttPoeLm`~C~M#QxNHzpbWI zX5yDB${&j^^=LruOVMg}{5A(YHWX*iNIzSsPnH-RS;3mCOM26=mj5I0@5}K&BSIDl ze*0b7$MYv=H-9*l4^uGq{ixbsM%X6oZnMRRLrnD!31<+e@?A(fcZcV7(=xBRE84Ii zvWa5fhPwDce_d`qe4tG6N`Qe@WTzR=zdB02R4(Zls6plohMa$QOS|mF=qQy0xw37_ zgGMER;&3utkF;~aQvU(^p*$5QG!E>+#UOXSp)a8frPW5EyP^@SBaoOuqFrTm{!TEP~i}DxC7#1MjrEzSo?%-Lx(KDG%v1G}GaAa2}Rd+VGO8V7K zEF8DzoeaVY1AxGSm|Q0|e==7BHm3afV=wSrx`2Db7px`y#LDaGGBm9idi!^dRKrdT z;#L4GbEOG}ip`LUXOu^a;8?T+^7%(g*Ph2TijAWF6TJrgNsYeR6ex;Z3XAqgptZ0c zNonCO$#`o)UzTXT4fL#C%O>xE6cEl^O`aajU_LJ`3-yyJR9t%t8X_1{f};$bWlY|2xS!#CY_pq^nM5c zjnf1Os!3_`=35Z?&)+K&=&@2H{sBQBlb~_ZqvW6B=K2dI8*k<>Rtn|o9h=sD5zo2iEtZXy23Qk6_Xb zYC8O&mTC2F6zxGU-Z*+UB+v21cKE7Q1mi|v$yEa$imFSbBX#(SbIel)fF78?XJ=vg zU^YyVDXGJA@h8393q*TWpEEUiA^GKQ{H-+fSjj4P%3Kmb$_B#}=x8_>{50Wqw1#(s zEQa$psA_VE_GQ+warb1x!O^bH;>DF22b&NtPR$kH@;mmtL}wE1kF(kw_j`Ow`Or|I zx-+(Lucy4mt!~gS+RbNVMfdJk$pHWf4#?W+`eOHFt}cFtdoM5j=J7r+vTWWkBs{4NOQJtd%DayDQ(Kw0aG#OJy&eK)f8Ndg z0Ac=P-+D@~muLaR!8v2r;$2#5=ntvfnRVDCwZsDfp6_Kar+xKRV0RbbW#*1~KSTRa zH4y~%a#o4d@6WFq6|SBPpWh@@aO_He3wQ56dYEmhSuT%4quT0o?$brqo5EPo0hXu=ll3r)3)tASl7M$nNAjoHwN497@(efujngCShD?eQIJ2K zyW$+&kqwCi9wh8g25C(I>%YZ|5kGl>ZZKc};GVNkDGGPF4OPWf=HF%JlXUhW&Yc4J zF1lWkMDgN$F%QN&t0QdT^Gh(c1O~ruq@Xx?l+L!As$O8;>dJXv+jg7Pp}7x~GMwce zqV5z@wLE~Id(;ZePAq|=jichC*rKnyU7Ol?P1gH)2V}zBEx_Ny*{#AdbWd}qdy8N{ z!6(l?PVNdPV=xJG#_Q*cJvoH^v)5aV65|92uRLsWBixdmtu!$OzKQY{P4s1 zbmf_)g-`Yrdkiw=3rlrLyHkza$yt$FZAb)& z@tbWvNV-l`N1U+G!~xd^r2l5;|K3$6NAMlDPYiE8nK8bvn9)yr3h-H$6T@CTuR1Fl z(3d|vtf_Du*)jF+B(oM)le_Je_NN!0rcCxtB>N*HG607@I{l2m{ElA70|c5)9X|C; z+{x#h*O(p=xrSk7?k^d0S9??Y*=ryYC_jj@BZ^g6ImKM2wUG}=|8h^iKZm&kfZ)8} z7Yr%xa#W^{Y&YJbfjNr|AkU;iUr<-;kUaKV0$roCe4F{vtn}{!5=zLP(nehp)ysPe z`8)9(@rnGg8WswIzh=+<+x#}OHk64IQ|Llw2M9i#W}lhGl#2g7mxT9cfp_+?hP<(L zZzMGZW4FqX)G3jwpN6$LIlQUY9i-D1EG@v>TEN?yFX9nn%Nw$1HzSLR%tq)rqKn_p zpk|@?Dn;7%9G!kyqlPO~lE773uy=w-i{Xm^3<~!y$Wk8dLQnAGDYyK(iHjoix`Tn`cW_JW9?wuzpnCpYc4I?W#)hQcP5Hz76MlHk=y zpIz|IURC}9(fGQ(t$WPDdk)57PU;7}kYz*i8$%zQ4Dzo+yCB?HtuY=)1j1WY;;{Qr zElZ)&>Z9#4yLmS*;ZNe@udc)p>)?5?xddx;V2z7#%C$@&6shcVdMja@V-9%OXKa1# z@t%EBvcEPFX|Cf@E%meS5dZ=aPF@D|TR|kg1#l93H$40ep`+6(_{-Vw+#PCjD6!KF z5>Jvrc0mscSIMH43>Uc90?TAAFii6g`oB)^;WGzIhZg;6?RAxWDS=5Znyul!dvklm zZ*_Xd9Gm-7{2HF`(9E>a0K6x6qU1=c$_^Kb(Ig4>u|N7?#6{RU3}P9UU$Cy%Rr0@n zb0!<{5c(&0uJphFnJ-e zVG7Sxz%{FEUCUB|((si2d>G>9*UAr%i6qwQkbt@)`~AA^U!8}2D^~zmL~e;xUi1;o zL7D8;=;Ft(Q(jZo&%X4)O9$`zztn>EPBL zHj&ck@5T;uH6YqvU5g5GzB_xt{`2dOSStUTcld0QOhGJ^|5*q?VxrTC{&rq-wdarXXHr z9?l8?PljvhBS@P3jq~aZu)e1hEq5bn1nQam>3Vz;nnmn+1xz^yx7<3HrTdK1Ji$x` zTu{5M&u2zpf~-b2D_mViqdu?ubK68MmudmPa7U|Nmg3ct1AE&6y1T`H`*902Zj%4zsXF}!}i(z)|SNQYP+bzbSmm&k(pvTO!KjZyfkkl|>O}Ge!RrQyQS;eO*@llst zrTlUw7MWbkYK{U3U>_S+d&flPdcE+J_NMQkWv^+q&M;SdR~CP&xLvR=>)w+W{Wczr zpY*u*&3=P8uhz1{eGBK+Jz5+g;J$?;-v&`{q;H>3>}_bE!frQP!bQV<43Osl-%(!K zrxt3R@%HM7HS*5=E_rF+!_Htg--31sZ)}n08m`|`AMI-QS20bDS^rufP}9`xFV|=a zPY1*`CMTx{Y$bWVXh6kMUp_)@E{O{BFy~aAT9%Z?3$1cDrNaE%*Qqn*2y5ZnCn!O7 zc(sk%`NBf;@p!%w>Gw(W0$?407VPSHjp93baJ1pY!Y{Y)bI(%@=N1`9KPW1l-Po%1 z(tvv;a?ZN5{icgh#uOqqz1vU%yf}=2+PuUY3dm~i784%BT96U|{3BG+Eqz~Mqk2i* zP=0%eRktTvg4&(N-KzPCB7_9_@JIcIbM)jk>U*28!8H8eYYTo4%+zY*d}%E~y*COCgQug@zSOpJ{ymz zuRCcmbgrDP*8WoG^=@L^(fz0UrFc4R&fA+7{1M5?Sfjot8r{j8XC?fY-)>>}7 z(MVs!?NmsYoI-$)YnhLmaQ@V{HW=#c85=az7Rm$89{7%axvh1C?%eqq4so$cYe)F? zt^uS&6)j~W+^bH~A4-gy z@=lLIjP_>E1+C2g(9FpYCiug8`sh{1eL&H-_nUc3@v#Oe#)GDs&piRq`_3F@&j7#~ z*75@HKhJ-(ygV zXLdV?TmivK2|kx_jt9m8v}iVY(f~N#!hPE2BY&6nmo0#&-{q$-F}H{21|HSGW9@Db zwcbM)Y)Wext6ys#l(E#LO2ygwmQJF~QxcQ4kBRcn3GQb_>Z78C#Fg#w*k!-E$N%_2 zwCFu$1(x&24{FwC19|d8z}Me^x*oBAEx2-{9p2rY*Y!R}s&3SC5R)?22NA5T0P<=9L(MwY-l_V>q%3jFgKrJphixp zS>VP7nEp4Tw;o*1?ksPwiHMnM;Sz?3TKhXDDOz=0bt4bzQG&d!*!S^B9%&8FQGZai ztiLEKEF!<6PJK{e{TkzGN+T!BJrLz8F#teV;{26}@s-9WI+_%14&v(I)0k+LcY~!M z-;xG_{gI7if}sqz64w?(FU##uhmOfV=`lt|ql99V4uB7RTf?f zP5xdu^nZX}5ZwdhUgZTa#%aTPNx1*B=ZVpI$R5L9jH0HItm|Qow~$N$!9 zfNL3PmE2G00KxxJ-6h)vKEjx%{;qwq$Kf-kn(JPi0X+p{W5iS_ad689>=x-g|2;6m z7x{xT0sdeMcVH3ekss`Fy_1!^m(M!P?`?mz@p=deLpI@8o4Y!7Tk7j;FRGP8-%`c| zOo=P&V82f1GApy*%<0B$<=A{xQXcsMap(9=o572n}3NCg^ZBRpZ@%25DaRDw^vxAxPHjr zKsdXsphI=DPQT)o$If+`e;lIyuC!!8Vo5lOi<6m-<6j}KGad04{% z9+9syq2t15-#~WSn#GCA=8;o|XywMPFOuGQFma0fbKqT^g+wWL!>GS{w3Oub1 z#>|1&Jj@My2B#&l=1zpw6d2ZMK4m(qo~+?+>-t5KFymJy7ODQTVlY0L_pc|DO@rFJ zF$PLGNG0vl@QgDDS(^5!uc1*##*_l!V(k~?pLYuPDlAJY#@!GzmyMblt(o?3Mm9GzB#ZxFYEB-nagj5EW%tgjFjwH}i61OVw@t(<-|W#M4o zN+<8hZxQIjHRmD^|cZUbY?KdUyx;v!bZSz5lyh6F%Rj7zc&XP z0rqrc;S1&fKrz^P9%49-gQy74Q%@y&-XHwVh3kXQ^R)Q8<{e)1;|Xp_N*wmq;}2~@ zj4cQ^JZ8<0nkUQ4F}_HkOzdARb5<#jyyXr+^J2_(p!K+=)BU@jnm5;mc&Sytue@6(6`-O3DCu0j+9Z(iZOkuiAHTqg0+3 zvU9&19d=$!Vt>VNm430&?o~x|eQ)Q<MftMC;k*!7xc0TG z@wYcZK9Nk}Hn+-*PKEUjrwc_k6_UKXjX!>`5s#t{%U*1Wc`^eJJ}+AUQLz*?<1de3$?%H;z1i=$k5^2YNE&7Psg z>E{Of8xyma8uwF}B&;w0TB8QB`t^>B;vM0)`Ki1}e%qpF7)zfK#HJqp#{bi1f0V)x z2Bl+q-B?RbJ)YzK1L_mO^a5 z-}e^TS4^Y19O}c;Xv>uV3-MBW(nY@&MSj2i$(7B)M8uV?_ZZaLK+e@_xzma`c~?d5 zuG{|P782&IG+{@W@{?1$a^S&F3^Qyn^3M4HVON^a@=a`J}kO7KFkMy)D4;{V{_nV+@ zLw)Vf9~MUo%^_kN&ol3smGGScnl^OuqEf(}DTH*X;L6#H<^ZQPv{QBX*pS`~m@Xw) zsMmha(Jbyi(9=&ef|i$7E7@fHmg3415^+2UDp@)pXq zKvq(Fo_>szt^q7C!hOC0B2NHWi=TATzJwNQ9e(11txEezIJ5}s3ZPes{FJME{jkW8 z0Cr_flZXsLKO*ThFB{f%VeF-Tf zeQp|G`-xl;c(ye>cqJPUTc~gVsjmKwDDm&M0&FKfHE$&%Df1p5b+=o}J8j(PhHrr9 zTPh|Db#KO?oP>J%0AN04_pLsYVSMavbL(h)=fM#mIB=g1Bne-E z`*gJOcQ(N*g}dy|r!O~wJ*Z~&7IIDH26>f@RSJp)0(L4#LfQFUP}tqt6Ec1J`sw;ZbY2lw(WjDenc?akI}AZ-vim8 zd-U4KyvoeS;ND(HDnvPgsoQVsH=Ki?r}sXrEoE5r_H%uG*5S zRD+m6em-i5e5zy#2n7*NfM_L=stGXxIe6Q3t$GP7u|2!pc zBKscef~Z!?3=DbpK%J!$i2hbvZYZ;7DSy|NsFgKvl5W4vfN=@Dm#MM9y@aWG=cRh+ zgR`rh!~9|Lpho0`m()kwIX6JRejVf1#Rw6=2dO7#qq_-ypNar;Oet1@JQ)mMVh+Ao zDcko5o++&JGk5qKWv9C+a4bTgyEz} zeyK^Xj~|(AJZt`)qu_ce(JeLF6F-~>e|3y$H&V~~O{nq5o*~HF7x_BaHULsH6M&B4 z-pNDc>XG8za#I5_u+M%!7kpkI?EbH}MAk|EPBI*xpYk^6apM~T&9m|1tkd5d<6kWM zEkEO1$=2ZuNb5|N6@uti{)t|OSx#(`DVdlYG2L4+_$gn7J{VYheVx8L4U$gP5PHtnwxAOp+xX=7FKk-mm9T@Lyf(wnD0}!dmL~JajOxf4&3O z;_Pd^g9=^!DBm0Sk`KPbB^|Y=j>aOY6$!+(_>)l(B7*u6t`EY))+5;i*f_TE6`$)^ zg46(Tbm@vVF|XIc+)&SQ`awI-!t)cXFP&H zk{^?-WQT)s5v2^bgi}t7qA7pz^9xUvusBm++l{&oOVc~|pMQ?>Wv!Z?%|f3u>w7R9 zJ=N7^Z(u^k02gYozLiajE{v8601Hkjf@$ZRA^N>4xo;|h4P>bGal!ruvR{_02sf#@ zpx(w0_86BtTGW^OpB`wXN(8IisqAc}pMMVAgw=ex@>@bQpI6cf3>ZIjW0Fl{4Dd>K zQZjaiA-l$cb^q%bgoobo&j<7uWVb)PYA)|Te_9LBSFYm$1{@O&`Ut=t07fwWeBIv* z=bUsLS)?!9qrIyh2KXfiQA~APS&K#du;<$DT9oXiC|NrmrQ$Yp2k{-OZl@=9%S}@e0ARFHh*bY+ zUc-AN0ET`Lf!b<)r>~Y+`xewKi>fFj<;p4vUb?nZ$!aWaSat9Kx01=8N`juvimelZ zW({UkjR53f5%RVZiH2v7k@FmIDMOqUbc`1MvY;L!E$M{N$OPa z3W!a*iuR4V4G2|oD&m1S>#D)?EC3jRT0Mm|4Fh={hh2aP=4;WPM2yv6n?z9*%VPj9 zYE)=R1ihgQilQM}6x(Xt_tc!F{)$TQ1>t2%?(_mkaG(9Ex*Q4+OvTPav+ilY6T_RLs&?B}i4;(>?v;D?V`0<~j8!})!n zE>Nq2d4TzvH#7|QjwPEg8qPqClma(*@NoYYZu$>?ZU3OMuKa|4@eb4(B#l+ww8!^b zbXVf7m(39JsRdzTp4@z4lE^WS#N$xMBA`^{%=ySCV36lkr0DUBrT&cRN-I3|6P!l0 zl(faVbg4V9jlbAaN34zUIUD3HHcdIRo~D~1)x=DkpU%o3k|KF=7DQS^k{s-_UaWG^ zh?t!2ZoPTbfo~r7_0e?|btXbLbW`gQkoVDzXOxmxDp2XGevR%rk%8V z{f;LtqapBoZe-DWSP+;WG}yIc$U{<5Z42y|t%HQMM!E20iGtylqw+3^{?>Orx)3l zTr+k@zXw>C^gygZx{5i#DCb_S`H-@or*ENbtl!ct?6>5Bl~+0@Ph<_So0Os)Lpmj= zpy=&`D4zU-c>FzoaA)U3iKEn=>m#^-PyN2mcOG|I|By(oU61@23x^xr8Lsuqc%LXf z80(j5?Qq#I2OCdQ4K!_hRCFNxVj;^TDEE+%ad7y;>oCMjwMq5YNt@nVVmKu^M&p04 zl9)Q-8d=U^x!!Co4shxNa3aDSBe`$&BOUrXz_N#Fx6(YO0ao}Y?8U2IrFpz_X*N#* z3oZnAWx}%FjpNaRD$zSnkkW>og^kZ|bPX7RZ*kHbLvT^?NXu5;aHe|3;EIMrX8Qu` zZ_1RU!^#WM{Qb5z?)He+B3H!g2=!mOhMxsYa5=zl%cy1ki9X6N?jMJ%WX>RJgOS(f zF^!Fm$(@=hfHMMM#0rLqq6%IFG%3&dgs0l=^PapGC33Ui97+$an;Pw0OnM){R^XXr z0c$Nj>mdtDmIiJgu3G;V2?H6jwSloM*+mA z=Jq0lYWh(_)4{qg-TdArR5(rA=kQ-@H*RA#qhBgn{%r{+G(O|aXk3dwz6cqH)}vHg zDrOie+_vBI^-wpr>KvsiU*B9vU&SOyj+N}x{8+U3uV?KHFX;R7k_xM758b_&2l`Bs zJfdiQ_TMTP;`s3``1kt)oUT3DSn{govwO%YeW`@w4Qox7Ty)A?A;3E7dF~5|rkQKF z2!_hGjq<>~NV21RuY(4eRejVp11$#4o6p@yrhj1WFUcg*4ruStoBku{f${Lxngqxy zb4CDKzMnfI(sa%KtKAtzMS%I&lFuu3bY1*9tuwSeIiGQqQ-kVJsoAFlj{uGc@MKuN zJQ)elKo$a2>`mmIzJmlA{%VdTRl44xA)8^P0jM|Hy?mZEK$eDin_@op^J&fPfHDhgftVYoW9 zLJ5$<{FGmQ0ub;Ivn-0b6bi7(qs{_(Q0@b%2ih*HV&Z+xeGBlQVokX+_NOr{QK?-- z*LEL?|GgxVo24lnKJ%M3f*L^L+P~a0#h81^Kbq(B$i{{b6rJ*!JF@gW@w#|`wPS2x z$3*$-!&wV}xqg1%UvWP??QqL}v-JpP-CDE67#erAq%WgihWL9;-tai-q$2=SoS)`w zs?5`;&Ssc3?Dl@lH?B6Q`VRZi@9VWftZ!0m!DF$xhu_ca?}^r8#-}xUlGa(2N-^?E zb#(C;9=j#oE-prgX*9bgxj+lEs$D{*>6a!j`mv<_d?9T6SHI<+JqK?4lUnJnU5(D1 zfR13{*L&HR@hbV2;}j_xI$N2~ie}vS;<1TC%2;SAwP7D&eb>B^{S?Iy?yEdrgQ$-l zxTSlZ?+1%A_K$h2z1uB;Kj&B7B9fVyMx3FREri`_lt%c~MpinU40;t#fD5{$0N_8}n5sn5PpbGRkrWw-dvBcvi&b)8M`ZFD*kKA;n z@ygzrD!pX0r;VdEP^u(xWhcVg8o&C0X}tMp<1RB=mC}mdzWa~H*Zj$QJU-Dwsy+YD z*U)q#7+dSb;cCag6UMUAYYiMr{E_c>O1_HwFTh?PRPtj2fm;;8x7I7nB$ z*la_!)?rtn`;@ZxpwkH@-;{Z*$kv`d+^^gvbsfNGW^4auluw7<vd^J66uojA?i~23QCBJ%$^X=r|(Hye>VW z4y2aXnM6`yB1chwd-k}t*`p|>&{Kwako;pizd<3d4!w^4<5=Qcg}TC`{hc?cjz~=a z!p9eqcVcChr1k?~9^fbr=cR(xb^MQaVkW<`@t8s4{ZOf8ZPgu6j3nX)1nT)i zZYD}wdns>chDq-O|_6>KH2KimyuD> z!t_IIWm@d!X=YZPH@v!z@=~oMYq?@(pw1Zdvmz3E_|qe0 z3M(a*^9S|W#*@L<*59=Oza3vTm!4t{_hPzWjXt^%fRv+#T{Apg1}$nT*ZFib4ubWL zYO0YQuFHD5C{jn%UfRh>nIa-<02K5nHd>J6d-0n#ov;$V($YX%%I(65OQ7?nVyZ?l zg_CvU^|Z6Y`AknGZ=TxRqiA!eM;*}jidG!AY89ow8d_Jl=^5qt0Iuhm#fh(r!PY(- z$Y-SY&IdOX_rB}8#qRp!o~>wJq<9;4Mm=#2V$>mlp}sC&xvBe>81NrddA=L)o&{TV z?l~KbcRnv=F%;prTGBSoL}5{Vo`Fnr(piJS8uE*|lYfQmu%_wI-!GkyW+GLfcgCO2 zWx&Wv&A}GR{bpi(662>&DYg3gb1Td=h0J%i{M-nc?cT zl(&3QJJ;R_W+9SaN)Fhq7_eBEU@rmOtbjh*(jmJ#P9+qK!ANYH2y@?A0ECtMo9w}# zcs(zrxyl>#JJ(9;BSOjIa26krdIHHwD*fz-1^YOTWu>YiP<}gB?`HIdx0nQ82RT{8p*VW`*=qo)6fcnFowbmIu8G27< zdF!NG@dZh5z*X3r!n}2<3<1A|-j=^2+Nzb~vM7Q3u#Yd5-grJSr3s6{-+d|`;9)#J z8Qd*|e&l0aU$hFnv-;td1Dx+w6qtY9uh5XHFm7V`^K!1f;-n534*9+Wl*D!SgSWI^ z+3k}nq=vKHbMPmg$Vtk z=dka|qe>V9dJ6GG)X5GA}!V?If>S^+s`Exljc7-A4_5s_fSOw7k$?<;4@+mUJLc&J%9D1J+8g z_@REn*dXYcGq>O(Db#QSPZj(Td}k&3ulB*opPhLOPy%KNdc-6B|0?HO^y9}>h;Ck&Sx1fDwz9<-6Vg$hR+&m zoO0U?p51xFni>FK>8`eiu)a9XFzNM2O7+2#chPFkAs`rN3giZ`->e36ZNg8A*A1{T zi8L(RZa3YA788J5i;#d`I0ih!$;rf!plh80Eu149kWc^0YpC_iZ7a!c`4@UD9bH}& zdGsd80_4M{)}EK{Yf4jPBAW*Wqvx_rpf&W%`CBV%9@X{q;L1)yu8f1e_K~&T-yw$@3fRZ8yjXyNR!`>`bJ@TU)*}F4Uf#iQZM9kBn~? zG~zaZ(Gtl&&tKv6sadT?l>i_D3=tG?4+Go~{0=mofl>y+p3b$>lCt#bj`^C`9_+VGMmhz0zorolmdK<_+YH z#aX&5;*BL7@p$2dY)_Vz(4bIneucARXo!z`3?__mHNtY<3i_LuaxR&iOThc6p(inZ z^LAV$09eeo=na_F4|)Z>!CbL^qghM-#1vcK3eYe8RR>dx4SpgwTnYT*5I8LwL${2EnzAaE^TA!_2tEg3YE+9{tNqmtKop18=!q=^dhS>k- zYY#o1h)-byh-Vxic3%N??8Lg5XFtG+&8-Im5jH+>&mvSP@(WyR~ukQ_UMi?Frc-; zj?Dfug#M!f%MtAeKQHL;69ns1(-=l>48yiMw7UjU`_t_;&*MO@Ni#t#VWG!1VB;63 zL!E^*@bCc|nuDNeD)-z9hUPTCKZX$eD441I*{86t^MD@Xcn5|l@IHZ(K~1iIh3pj4 zmtcL!5|{zd$lX&%!fGq%$9oyy{r>bdp?j88F5!W6+SY)!UwHPCcd#_4eJ()_Cmv!A zH@_|C<#gL25_c*z^1?EVUHtQ_0ZDA55!1465eS-;Kd=kelPF?Bt62su={(gMcpVgc zheLPhiTfE1k%#)vB}n+sC%onJ{F@(O1`HC!AT?titQVrgs(mSVKC8V(&&Ydv8d44_ zwm}W^h^$%I=i7r^Z8raFetbaIMzbEqWlcbko>$s)sr$$N5t`^sAx4TWCCf18M9}F* zt^LsDIr<>zHTlCY0O#62u-0}8dIjy!EBU}H1k7_{b`NL)9WX3({90G=pIOpcXMwh+ zIDqfF%-{E!H5N(1>^rc&eK23Cb9f;jV`k>;0^W$4 z@N6!Q(Ee$nb@LQ>Mp`6@g>2|CXv3-a;Og6_-ohgKIM4s@JaYJK3(IK^)>gOd2}K)U zzA~)`^fG%L?@#*_n0F&Tq1WU>q!z&XJ$xU45MpOWcSNGJgX*^+dnzvwr{Fy{k?MpU zczV5Y{NYdI?))^uxNpXC!RNbuUXwufE;HuQ9Jb(S2#GRFZ9an0Gh|;XdwZ(ry~&>12NQEZp_Uqw(+*!tpCm zDHXk#gnh0^XOU*xuB&t&&kwtxmgI(PojxG3#?a@Y(YT$e@B5~%eK`5_Di&WFm6gGg z1q(;wM!V-WVEbjEDIk7Iz+Ef6P=Mv?GOj9 zkUYU>8HT=L%jdv(NhhsF^0XBaAITxmGm9h$cy?1G04q{H(isN_Gy1AhG&*b!9*@Z# z{t-F5KhrL1xhP}@?m&pf6AilXk(#ksoc)4Y@jxLu}%r+n~jXGIBSC742HS-y6PgT#VU(^LR2ApD~a6efRDnG?dQxxn~(y za#^afEkMppgDE^_#Jt&;8=zf}ontq7-Y0RX|Zf zvRaN=)MN<18Q=YgVb&2yHR|nu#x1irvm#YA56{ir*t!L5F-SZXT*X-)KIqzjIor?U zT{$#;Kt1kakWwjYB8>_;lAXGk6FSX4_d`r@DL_TQ4PPO@6Tl_-B(njC7!ld-=?oTx z=W+b^zXv*~%RY$->BINQ*z9B^t+~hG2sT5m9v@l9F|3`f2@{^=Hw+E9(5Pp>*+M+Z z#jTA5&Xz#+FqOT`Gminm?Nna;D&k5XNZ^Yx1-uj_Q{>0nm~GYX@Ax#!{1&dgZ6de&UR;W1KZ zh7zfq_GCA;W@9F#m8d5+Eq@;ZCg=i%<3dNsdsI^*?7;LR0rjXPz*{$?T4D3S3pcJi zy3hD4`Hc9qUl3GLUPgW$FJh&G|6kGk2Sx_s`KY8c&%Hp@;?m+zOxy5>uU;&qcLz#j z-F$653BNpm;jI6+HRt|c{caI#t`sB=qZFOPj_99DBYCwu-`ex}Z~yr}{q_6jZ&S8T zcn=t)``f?%^}qi8r+@wJf0D}AJY;{3<(~b;(g%o8hrhn%RQyH!bzHOYfBD&5fjgrGl3+NB{?C8)|7CeLbN2hSa3D#ZA$jT`kR(NO|6BLp*X;khE4$OP zOP@pfmzXW(b1L86zy3u~@**`-*NSGYp(>EDLB9VtG}fCh5#RCh1+TUb9htUr)t3 zB#Oc6n=t+JXZ2RtqK$uHICnwUJ_o0(AE56Q(0AG4O@7;K+8>XnK{+&++ryzdi%L`K zbtlpVA`RB6tmE9bi+-qcg6D^Vi~}9th#?3>+}kxP(Ovg@!mZ) z#z=lidXEkDa3_Db^T9Ji@+r=pEwwsyFDGm82LOLtWbbiw0QwNtS~+Oug3>cMGqp52 z)+xV6Hy))95Fx>#u+m}5UWTpulX0PTG(sntl{3=C{O~Emae|CS-i(7_tJ|%?a~$Hv z#v?f%>hY<_%G|B@MHYQMot40DcUj?tnK*Fi=Yi$`Cz*JB`5Vr>FXHJ7F#u0Mu)kMx zHYOPkoY{JtQC)+}_b06*jmB5hO)$Q%~F02bu`lX(e zCV^uG_~;q!hD|^fUub;cT3p@B<#d0nS>?R1fX9#%cS)9mDrLzt9bD^#3vt_y2DyKtw*r|@qYV~FXT9X1i<@qVtvI&S;gfRD~ zc~09Xizmv+d=!h0M-FTTiR`JU9RtaVQVMt&FP&&0JoU%)63njXe@M>Q(q(D{5!u$E zobKzrK%-^Y#Zi^2_!jWa&X@9`-Sx-k5ey`EX@6r4ZHngKPq{$!tt|k${MuJw`R6TU za$mr0T~DpyUP8FmIuWG}Q;4pO^Yic+4mht_1eWnn^%(yD~Uq!zyn8SXjGPyUvtGdvtoThUMMX4cbG zz?-UC=IsjKi5~RUM}Op@baG`m`M0qm>icA^4m`{&(Az0wR^lB zkHCz%D1RLuS5o^RDI~B5=Jm0vD=nX1F>q(Ocv+~Od^~VBSWkEE{9SoJUnlZ&IM#}4oyv!$H}{Nq%>Ewwao_=cr|`|EMn3TyQb z8z9>c2WHFp!T0!cgH8DUK`WA1XJ1Yyk!@ZrH|g-uunTvZ?G}Cb8#C3e)b41 zYzkRKbjjv-Ul$c1BK;X z*AXolOEuQ>T=>3&>o(7?W1^hCPKSxWFH|hmC8UzI_atXj;K0l!JJZEAl$iv}e`B|f zmrtO#l-BxW*7fXNm+FlYvRJKZrVW7pN59dp7&g}(nqao#d7d9|pf`^RC%)j%hm3Et z{c7u!z~>Jl4Mthdny|jrB0<6jZB%-3h|SuCIJuysyJrgb86h1MaJ)?bv2hMavq14( zm)^64Mi3tZ5yO?{H_o5WbcqML|5o6o6<@Fs5m;#itUKiFd^nv}y#9W9+|%=Hvau(k z>1QPW*4{WA%_J-TjQ_{%fxQoE?Pe^FQ?#XceeGi$QgMXuj(_P-bc9vKGiTB8Q(wpN zh4|SBlD9>vI5%HN&T9m0xlI(kJU&G3A)1X&BF?whl{OzifCV4oIUmzI?(smsJ<_Sw z(^D_ds4sy#m>JlAGDG0vbaZU8lJm0MpAewQx36Ey&YLJ^fpun6V4Tlmns3iZU|5tt z-)80SV}u{!*t#kg_zhX$tY zpfM{Fyn(0~AJzN7CU$5ZDnkx5>BP|VQAU!s@{D2FnLppPheg9bT%t(p)xFh)hT(WM zq%2X7QpMr8V$TaMV7B<>KN2fg^WqPg{rZO10kvv|qpuy%m_EwQsXRd0Jp~;pl-JGh z#1yB7l?il$Z6X@AKW`rM`hN3>kK|j8)?%G(?}E;)xSn7S5dtMLE3`-ol%l$E`5j%4 zxl+I;Y&jFwEIM#JWl|G9;GBh1eijOY&7d*QC=R>(m$$Vdo;GeEKIA_A{BxOFonL0Q zQlhl}Zww%(f^BFqGqJ>TMWBZ&(KVQa<~Y^?-wVGMo<=G#OtHgW5$)8P3`>0m)A>QN z>A5WcN6<63c%SAzc>RH0!In#u*7t^tnrL2_ii*wSXOQ#RQE-O`kCPU~@92!OC;v^_ zhDqAsm)-XF3j~YEaNc|dz;+_ZJy&+K+h5L19cLos6ANM00@?5ev2XVJp5a@&;xeJz z;m-J9`4&1xM`O{qe66!D)V>&o=gLJi`Kj%4{6{4S10_TU9;y{^8^4@%V^GzIq8`GVO!?OYG> z7s;$6yj5q+VTmr4v#W3ep4M!*jndVd57!Bw0l*MXJ#ZIPmJm-QYrtkZYmw0gy2p78 z!p5lj>*X;#w!`}`If4@e4M~nBl=%%2=eTUC#a!Y2w5GLgz@W`X@v=JEG7D!Y}L z?dzX+8na&~qZ3O9Wd?a1eD-QQT%6D4gmaaV&(`T+_xOj5k_wXB= zG_78?_{CV>1u4|a-bx!Mh@aZYZ1->s^zw$<(e586u^!WU%0-eS@tqFghGtn&b`|d9 zm}S^^1%NtxfC^`efa^H#%N72dWBUF@r`9ZX44ElVj zdKO{nbyQ9-2OUXWkfc)GT2qUT=+~!%l&SHv0DC#HA4C+Ok}6MGad2*6SLZCx4IQ?s zvE$JY@VLH(b8NIX)bgfQ>z8BUs&>O9!NBd?8C6!0c?Z~G^keOUY(`8(a;DBxalKv0 zA(~vG2;~#X{Cam(kSwGQV1V~wlb-w$rK=+i9qao;1B(|L3Y&tzMbe*76UT8b+B+U` zq|4ZL#rJ)ua$QNe>CM)W?1-n4tYmNwFS*7Dod7a)3J2xIVZzcfsb3(8W%>y?WIPiU zenpZ);H|zd;2|@vt15%v5Kj&|j)`vP&*MXB;GzaUF1}&h6Y6$v41a-Lg)jvf3%DqX zY&sYe#irig&KXnpsyA-=lY$24W~tTnTca^#Gsd!KdXaALyC(U&-VX;Hl{vAop0I<(p_1+%1^Xlb zI7iB`7~`1aZ<5OvE|6`dqJHso-ziEtp5O+V2I0-z7EdRv)#@0)rl3v!o*%JB*nH^~ zqGN$*fB=;a!4k36d$WWL2bu_&c(J>-C0?V1_Sfjtky#0K6 zYsd14WOeWGKhHyQN+n{|b!5S< ze;)0wzF;4H9AL0LIM{eY+~SVvz;4ISgJlysAG8L52`aDK&wyhL9pR~-)Q<%k(a_7e z-3@6X&S&7Nj4E*BO({4F;nmm}1%7;0NGQvwb7?zG%Ob;)azIEP42Jorw4)O>7UN$7 z#50D?x(eJ_N^N#@5VAcAhGs1Yh$kp!i`SEiw9#PSbm$%1`3?a0_zs3IjmqA>lj77U zI8O2!9#S9io*sCyI9b<7`bj9Ls9f2kC^UwQ3gq~G&@R6{h)RiyPN~+3*rt6rb`=45 z2$ie0o;pMflF&kBn}rB5<8}VV0*}cRP$QXT*p(L_T5-ER4#jZ_ajqg*K|5_s(gO!x z;#Qp zwG&qefU!cS1JTDsm*>Ob>R!uNf4*HLz;)rn5WGNT+bpP&EIb==3wl4z1bg0!c1%R% z?y@G2X0~Yfj?KM-ZT*gr$20q2opKG|lynk<=rwBR;rh6~j>L2Fj)ag?MYr~-L;5F6 zfnr)-(U>Jp+#Otv{WWdEqUyf04TB`O!vsA1eB}ImEu1tPPj?M{S7Zaej?;wW(H?HC zYDVScKt9i#0RZS{+xuk>R)!+8o5$5YV_W+Z@MX9ohq^ERn{#esjR4RJjpW zckIXz7vucD@}WBtH;n!<1U?!x7xD#i8V{-?8az+IS5>)s)B}Gq!SxO6ER^5y1_0ul2hXpd^rt(l zh@g9S$}EKnC^hjUf~dV6xcfk0Ia&wahUcbiKyE%N<&m$z`jT;Kl6l@|q3h~X8)3KY z=mU>I^5?%iSh9sih@Q)7_o$ap3u}TufYgCaZee+SYk&i?EuD$yPe5F1w@77(cHkZv zF5d44Va4oI(RIL|kgoDYLC4uh62*>}h7B4f`{liWBjr(wMh2~@4vgmk50SV2i=*kJ zzy|_|zIs`VI`io070nj>9Sg(4yuXGforM;4fG-Qqu*t)AQBUECN<%%RB zZhNY2JO2q@8%INw*36f``;ItQ$4>q*mDdxEs`Rlvo{;vR|2|$$SNA&t>ky0ym_^kd z8m{y5EW6sU^Cxc!dSdSZi2=u++4dmzd62*4#g7dc@zbrp4-M7@S+;>oD{JYvol|DXmG`Z ze>Z~JHI9wsx`D8GIf2Sr5jOe5%Pt6`hSrp}jYU22>UNx|d|^gdx3z11&TK^&@6gKzgLXZw3T}*9lnI_X5w5IJ(YJ7v#r`ff&hBkx*_}*QhHW1c3SUPOuBuOd!lCBmN#rxj>6#1X)^S6)O3>k$Il zd1MnSu^N|2(AdisjbjCnlXo1BVebi*jINkvK^%%k|1oy^3>bl7ej~vVW=&@Cc=Ql2 z1%yv9Jc$cGiEs}Q=BNhSdn@9~)oAQPgpSgKcHGNVum8N zE{a#t+_DmCk9eA#fN_9HWf7I^kFcid8{#t`11Froi^B(UQ5(|KtFCuyZy>DgHyeGW;X!`MnXQ|XSRB9~>~TsE?SdU)bwFbA*= z4IDVuE)YOWJDvCNhROncJPN&#MksB&R6k6B-Uoobsv_}FBQlh)vlOv8!@^!7o=vXa zW#z*dxBz?{l};_d%c428sBA+eXI%#nG-A*Qr}*t*`{!s4(xoCZ`)Q&)e#B#0k0vvW za3Kk>?zzsg`Bfm^E#5`1uMd$SACIVoh(TeQ267UqXr$>H@}Y5h0YA6F1snwTUg!V@ z4I6^zG|q$;r_WK2!1{`|a%jA*z(2iF$bL6 zdUfIbvLPCY#n>OR*@(EQeL#WINQ%HoKZK4-8dwwVo$IYEYKrkuhITOiN?;>2egnuA zH;lC9)sAvn!5x5yRu4=-RmI6ytB~sYK0>7*@4M*_cUnyWN=SsY1RFo=Xxnt(>qWH0 zH+%RL&J~`^Z*N0oJQ2wS=xl)lJUmg%j37Yban5s!R>FRsJ)nH`jR!mUGdk?^xHi$a zg|(aINBXQ3ur2K6{YLl7Bz4zuYvThR|Bm3pcvKIBLrln%7dV+CMm;A8Oz)5e|FpR% zkK3X#wX1QL3^z1t8;XS1!Ch63ml6F#&jZFE;g3vB?vw)qS8i9!F<)OoCd! zZe01U!uc@?6`7M9my^}C0{VpB#|f|#8K7L<7*uOGa4cJn7@j)-bS$-$T@8(ev~Ur{ zwH=i#C%95NE-E416^LKP{lE(mT;UQ)#@)M=A*#OZi6MR60C(>m>D_*9DcZlati1ox zvhw*y%gTTBySmNYV)BtR;Saz;1(e z6sY&;ncY6z1P-g-E*i_|1CKj4%aY%HRMrdNc0hTbX@~<4ah`Esc1Lq`qi_sQMP=~7 z!6v{%8xzG7d@^zn4@D(LABp75_vq`%pg!w+EddXPIFr6|2otOEkdgrL-X4u0K{*UY zhs7HVxEE^j9?o5|ai0GQV7OevJ|Xr@&(Z&lNX7sQU~R?{s3pUZ9RCrcG1d?J3UGday z<703>OtwS=lF+Ewx&O7Y{9;vu>izCQj^6d3GtxC2@7MFOdcA2Ldn4a7BolW&_-8`k z`P&`#@9PD&oSlw|!lUwG(&;G|Xo4)>-j|CZd)fslA0Ogq^5Mt+J3wdL{o&*P+LXAN zu~135r34=d<(Jq9odY1Qw3D)BCt2sJ=``q%b}Mu=x_3gH7I^5cB2K%rl2~v$pLY$N zS>nPS@CI;QgnXHpPR;nJz5 zVB5+}`9ngYaWvKiKte&h3GL+|$-a>P+_hk5m?+~BuAZvsmqXa2H-XNb^=-`;K-%E9 zyx#~!YameW#EVf_Eit&WKe!O$7aCQnPxf$Hs?UcNzQ7R+be-Ieqlk-{KU_wu-RGLl zOkAxr?xB)5AkpC$;cC1?tt=75@@O6xV7s^4K&=xZDm8XB$(?|Dh_CN0$yM=Am~OgU zGt|{0*sPzBymr^mc2I#-I4QiGWXF)y$TdNeUjIyHk_A#8XKP z#D~%owY#hr5BWLQgRfhk|C|dVOGM>=?Yh9goE=D~Qn+`*Vlw+u1{4J}2<-7FPd=Bm z5Ivu$z33{K_}yL(?A=2BGw(YrcgR1Ds|RdxfMu;Fk3G0YMm6q}RDOobXvE3UL5jIS9IA2JuAh zj}VP(!~K}KQT>-mgc5FKj7^d=9uP!X`FHr|?zxS|3b9r-h!Hv$Lb9$;YnN+ke1(a! z8d*GHOft$j_&xpn9l&|&5rb^yy2H8vR>H!32dC$C>S`dW6(=}{*=4PL2c-KXwxU$U z;AlC4F|k~4j_Ea(OL%}&1$+tYyn~g-CM%;+d*_gt_Dl>I)iEgSeKVlIsby~1#2+to zVjYjDMJzNdS&ZPh_{itcL4Tvc8>fs#B9Bv8vGf~_dhE4dUdI89qDrWgn4HEj!dK=A zfR96^&af!`o!kI_0k@m$eqGp8#g{Ucfd_=-{69@795YIIq<$-o*ct>d)Xfl699fC=q5N6N*Dak!?$VPx=>tv$)qBdQLg6DADU6w!H|&+ zYk{x1$MQ;=i+}kJ%(ZcGJxO?e4U_BThyQoa$(M&dbt&2KG=? z@-YVPb_t771`eNw6G^AY=X{e~DNsk-9H|Vgja|Ts?LwysL zz!Y*)1X?53rSsECBfn4Ou>Ofr7NFBP+1lAtCuJH))zaP^cyMqaA`ACHOr} zsD_fZkmmEXG^_L;RCJB`1C5jXjAb(ZXn7V-=DX(5bn-^EEM4~$+4kjvpMT4U2Y+*$ zVUTZ+lzVznO>-7JG9zO=_dPadx(_@d$uhomL~RA^a${?M`eCdkSg6BLtRea1P^jJZ z$eJ57@|}uC)&w-Gqq?B(^~2k!KBS;$^^D|5ezMYA1Hr_LMkHz4lNvf~1y3RW_V-{` zNXq*xzH{%_2EL&h3#~ZqSW`j67M9OPuO0|F zy?9peX0PQku=qrlsPvp>l%>#rV|Cg~80h_>e1OlIm4T%e?{(*Qm-nyjnVSx;Q-2EE zYk9h5=v-vt_KQnCe>GRW`}L2n>sgbmRx~4HWz+(k5tnI`Nv{oJtB%d!KoPwx4eZTQoh8A#j9_beKxP~M?_2k9qypReV; zJildCN@(iKyb{!h#tNqgUIBe}VefCBb+KMI#XkX}NvyjzNNe9c7=d_vZq=!8DF4Tu zr0V^4HYaFI7B7}XDKeE*kJ-s6{1chZl~yOQIXkCYSLE?yFi7u0O2=>d!U8QcZzCDM zfk$CV9B&t%5>?}#_W*1+K63WY3sT)4M%c~?6kzPj-v&&P@ zCLO{MD68|byPY8{RL>MouY-E_6;`l1z@oOD+qP-4Zre{9YS+gQZ}nNGLUimcE7TwGa?`+A@9g)bVHM`{g3)h)b)J4pT=3`8_&bIuGj)hjE#Kab^-gy&BGDq?S zsCEe965JL=@i&$<8zF|WUs|vI5q#6b1&VgaY%~pcnwCg-&4wpCtKnQo%%W$p{K*4= zBbP{FkEFke}n7P`OD7NDsLZOsGfs3^OnBT;l$4O z1H_kxw1xxewf*NAm$*Ejx^MkdtzY_le5u>q3~J26YdYbSF*B<~ztmyuKF25C?9?aE zzFd|iDVt*jY*#>*LO5JYol)zFb4FeKl3l)3)K9(wwZ`p$aa*7Rk3l-HkLwY&5;|$S zMNZA7<@XqKz}LDw%iQ$wX*^WBpraA%xPvQIK~R5!gBPn-JF(uK+S?q2O@LVK*DjuO9NU9Bm)9_8 zbdHUO`?V(4+agvtE}rO&s17w?&#*Z>(C9<6EELt3wQ@?0_xCa`7iS8o>B}EYfA})s z@0Yo9Mg`b&ETdT7Nyd6D+So9ZXRzH~U-P3*nuz3YSYdc%$GA|$9o+Q1UJ&*;0&9VI zV0Cng?{r^MBH%yqM(y~$^kY>Ar>DRGw>vQ-7kmeF z0$V`L#uB{Qa8Xr-nETS<1J@5FvA!_{;>e46h36wrSu}YqW%mT{i`j)V|6-kY*J<9i zwJYhj+dAEW^5aRi;QLN0li%4ps%7Xub$EF$_s8jl10==44^Fvv(OTeYuYeNtv}gx$9U*>)s4>$Aw0Pq=LWrrl192$50RV{`dE_6k>`aWl0pY4 zk+I$j$D}Y!2)yij5s_COO}+c<6us1#*MbxCzW~=-Bw2sCZEsntKkCR5!4QM*rO7l` zAGfx_?UGwLK?n_9972mlYxH`WWq*Eg>1*|e!OL=7RTIvxYxiIa+})y=*tT(TUCBQf zx+RY5y>|pW4v49J=}V!GQi#l^jw9L&Nn^JNlc*FIhY^aR`S8Q!{LQIP6Bh+HO>_Jq&zy8ds~H{+uI** zSAc#;z{q3w$n?@QAJ5_3HNkys;u%w9VUxkvy1#WUpO%oa7*m#-~D_}uRbf8^aaEh;vDYp={_>A*FNxxC5*zc7QNOn2v6Ux?oob)9~U^f zclGKN7K zN0VGQ)l1dI@>ZjI3u2ih{v?`{7m+#B*AFV^1&n<0%aHFOU0oZkj-GY+lLD}QqxW}a zgvdF#D$1K4;%1L1xONwK0&0=^Jd|qXe&e`?$4vbs?YaKICOVMD&vYnU@P{A4H@|6m z5CyD#c19XBbQwh0MpmNj z?rOPbZYh`YCnb++YJM1wq$rAwe;AwW++O;sI?>j*-Romzs9onO*yl#bPQn8x)-o-= zPVHU?lD7e~y0szWnSHvtm)A=NdfT^DFc1~0H$5GA;X}sBd_LW#8B0HE8P>I0#!qF; zrPaSv{RdEk(LFOKpw-W7?J48?&|AT^Ju*rbF~4;jR@e}@KH-td(*l#<+UBB>s+6vz zFA7N>pqjR!WasLaZ)tZ%uu;)oJiSS)dOz(Q!&BS-GlKEWVvIM)RY4+LZ^t*d+_Vy? zEj(YqH8p%oeE6})uy@=&pU2kLZyd>R)IOovx9V0QA>#|ipA@| zW7xI}7MHdtFAWd;vBhs9t5rx+9?CN%ctDh_o9Ey7#j60Yu7GjtPFl>0-&Esh+T3hx z-(GXW?owR(Y*3TG&&g@r0(Es%M^vx)0h`eaqw|Kf_bCC-Nm$xRH%q!zzbT&MOgtpY zgP0<6t@lV3Av}@CkhKK&FeOcK?wLu%y}IlZTMhYKT`If(NlxMSl2FX)L9 zwZ#+{_fqTVoa9&kjn38?)Svpjzl=(A(Rd0UpHVIEmf_LZqq3Jjqdkja6~a+Io__h~ ziJWR_;UrJMnfHjF1K+qm zKMy+5q-3@3WHIKmqt*?Tp3Ra{qHr?y*+Sa$_gyM?Cgw9-ri7U?_lI%~&JaH4L#l9h zEVHL6L8H4*x06=KW8*9QgWe8LGM$@ReJ4WF&Dk)DGT;bRz<)TxaOaOxt`E52mv24V zx|{@ItN=KXD(OpN&QJym5AJU8L!Wcq^{m-Jq=?WT+F7oZ#$r=&OfiY>x(g@$vlgz`1-(-0g}hcTaNmZuLE z;|+UMFdVQM;@fw?x!fdQr9eMukLPq|lD_S?E_qIw?DgcRgnFLw_-LN5WeR?x6ba2b zxE0(ebHxaa(9QOnD;3j*aBq@`)~JFtGKo__a&a3Cmmu_&+xz_>V!@_)|JeZ+S}oud zoIw^hN*r7J{z2U}CZ5%tLq=Bdu(~~@O1ivukINdif*)$}XH?cXG0Wt+L8x;OtR;C$T?mF1&-@L*rz=mM(AMUDf!4)HH; z`FoaT^_h6~r}(_`Lf=Jb6h3zaJ`>Da?97RMo3m~@zw5N=xhp1qDZHw>oqp&3tqQ@m z)e?QyP2Xkf+@|3~#A+V$!f6il+u>{M>e{Ep8@#4U76GzW% zi|2hbWM30(l1-+}`l4Q5BukJ{%5-%;`kH=6Cz1A@mIDdfpQjkdo?ZC7MAF$y{OfYB zhDGZ|)9yq9k6R_*4+1 za%)Of(jSX~q$lnrTvWd(U~|Tut2f*JyoH>llPp5@YCW2vDk-6Qj!)|w_c5^by+0lt z9QHtBosYjgaC{e%Ye`hx6U0K7x4ZJuNzEsRpum@TiUgO3SMsxka}QY0HLI0o8x#pq z-}U_&8FGwplZNH_G|$QYleO~Cd#_drHLaV}v~%4hk9FS<&wzJ~XD}(t^47fm`!(tL z>Yib%`RZ>R9_)6ywgXh!&trk+){MoV(cP#mX|^64HE|j0jU}HE+y~cax|7^+LOQ>r zcF)9lW~tz_bl=JI#8**qtcj5l)&D}STmuv^>r z8tQjLxypQ3lW2yxf@<5%vVChn;tYVuWG6V$n|e@~^BPYj_QaRXNOGF;n0ja!@Euy| ze}n56m6Lki4q&nM?5>%q^LNzYLB{~S!#CCK%x`oncGny7iE@CpVKxX?AUUGxjt&fQ zJFa)GnWCIIJ(cb3=5~I)A{h_ZX*!oY_8mzjHP?5fjp46P>o^8_>or^N-!&qcU4CS( zee~pF&``aZLGBehf3%59sksKDV{Vrx>{*VTao11^=IAcVQ9Zs^p{3Px5-Hc$C6dqj zVUbGt5~3cZ66JkyRD&;gpNH1y8vj*YUsAk@y!>kGAFvRg*c_E+BhDYmbl+hRKIccg zP{r}~GP4NSZm0YO(K;K3dk{D#LQ1OOrgU#^HT z&E;A4o=^|yua+H6J^P|QLY`eiKBd`D+a)bOlpl2v1^K-ksuMlGwHuwB0kzE%QMmW| zYpG||9c90l)nD%;I92)A%ZmCYN!&vUF)ywU$$xOU&GA`2P}~70*eU1_+C@Y? z8G}S}O);?X&U8aTjo$H-MGRMzpM+9#n)wz4kT8R*P);CvHg{??zZ~2)%9VZ|9BxXA zf-8=*3XK+diBrF|L29byDZtW*8(-?>E8-F>7_}-(`F)f*ZZb0cBs-#2Ja;IW zdh}lh6*E2Z{T0s-M3Vo2OXN#=S1| zOSBmafY9K%dgl}8Q+~-TI%C>BrY(GIdYw(6U9-M#NZOgKOm|btta`^+S2=xO=g36! zD%u#p7R4JSeg8ZutXg&1IN_sraof-1`n?-(nVVp#0BHCcd*ZXCDv3y!a#b$EyAurIW2cM9nW2Q{$6s%%xZYpIWhjOb_bqX{+`S^y+T)~N-kBYWLw~!ub~qINFq32;YnA>D ztsq)H2l{|#eYF_OL*lOD64~&kgHjIIc|S8B#y(D` zFIA-B{lh*Hpel9je51zQY{p;o_A|CM3G+ALqR0x^T#j9oj$zgGALi_ zp_`TdhAme?%UPbZXcl;LbFWUfGsI+RZ*JN(?Zg9iJJikSE&&x$lk&{sz!3=3M3^Hi zXxK0Q5f_;0cyS&B@PLca5dO-|U*ge4{4i6O{k7Lby8@t=8kZ9H$rBI(|NJy&15F-wW0DW&}2fb57mUrp}fJC-{)lb{QKT zzgdR=eIQ@1|A4<9A~V;Nx2JHJ1aiV9+64^=I%@VcT3r{=@4J5NDn6Cn@O{;-P~PVL+nX1mR&i|PE{Y#cRwBu6;mev(c^Vgz5JrI9hGG3VkaOJO002Mj@`Wnuk2a8)IiQ!&O zu|iZIo)69X3P4d+DR>L8?$+yRr7CNYhU{FPPb7b<9`WX52-jHtj>j_q5~KX?fYn*D z^7@)&|4-K5tk-oW%MyF9U-2Pd*uvN zZ36~;6Zr1Qdn67q?7db*?6v-3DFV-(hU#Dh-J=94`=H0D+Ic_fns=s7)3z2vusoe6 z+&`D|$C+eCPHXYi4U+;Ta3ejM#rQp9$@uks`=Uz}lCOP-B2QYU( z1L0=9TOE%|N^+>j=oy~5Rimkok1)ei#8Fj5%yahq^^*;61rLqOcPO2Iz}gl<`|_4l zk_L$Qa~uz_x?zoJqnOX_`ydmyH7HLT^HPwJ1>>%ake}2N42s)w_QiR`oi~ATc8P9R zRXC^y*6Bc5Wo8H$sFMrN3%;CWldC|?2^8h!bcbi5;XXL=hA!nY+069EP2)I9%>w5I z=L7gQ^>hKYH^jgJVDJvtG9eg5(H|J}XvPuBK)f&B^eBiJ?GGw=38m*Qku6r=glu{{ zSU%9iltb*4Blkc2lu_X#cw*eH+tzm?w6Xu-e!J(OWd+#0Cz8vpM?rPkj}4IOnw>kz zicm(nzyDP7pPEwl{*H3*-H2#-XMhH`7fZED>!NdyjCv=l&TxCldG@vVJB-mpTd)?Yr%VUUO05`O5`B!Dx8v>^b3+ zJN?f6CgWI*$91zG*IuELI$Bm{jX*L`+bK?q%S)XEpWwf=Ao4MwyFuODTRa{z zu`C8t_#r=1n2!swZ{OO8RkmM`O15zrh7f)C0Qg3J0st{Q^}RYVi5*OjesYg8b$8=Q z`gfW}_-fuvx`$jjh<_1@bWE1=5D-7!i?!C@=yf z*ji-qiC#ItnawP2Uj5X|$rqkZ=7X`z^>Cf6ad!aulmq=Tc62St!^Tj~UuEHu^h(DC zue@0z_KyXocYx~yXi9N!nkoW&B}4?hVe`hY($!2gDM$rCpj7bgaIH)$8|mtQa8{Yx z4Yf6X&D6p9v$jn!_v|D=EF@$_dDuGD2Xuw2y)*Cy{dgPgu(bRBq!RYU7|1~KfKSoM zS2bEh1{l$*n)>m69-*?Hb=D=Ha!@O}vlo%kX#BP_buM0oYPg^lZEm`KkE)xUC{3tQ zGV1vkr>-tV_LAN>nIIs%NfxOkVJ2Vb$bVMQ#tg1me=!rXanM)0NSgx*^fJQRH=gw?sac#e^l&)sL=Q8CH8XuOP zn&r5dTqdwMR7fMTNM1NYui^TAskfiiHc^Hwpw2@12*hU%>yU0m0f2$Zjo1r>!FkqQ z0%RE}I0b_49Yl*$`{aY=fkiJTAE3m(F@@Oo`s^J}eWUKXv7Yt8zMt?MUYc!Qn5)fX zV*TQwxj8D;r`w&*kpX8JMuZW{>zD6~YINQpoKgJk)!z1nkas?au&<-GO2ziu$Qu(# zKu@P_)1MC7WN*aDsLC-oLvbM6L_zn3b$YWVsxQ?0^~xuk`HH2Bqad12-a5=(K;008 zK9T+TYyk_8o8X8jn#)FLuk(Z@rhjao@JLVAZ|up{^8+pKV+pQhAogQFPPp#I#RuqO zTFR9xC%x(QJxYV5RoBZifcp+iec~5xK$kYC>xk`Wke|9(dJKUnrkCj92!3BCHM9x(6+WLo`|jUfSe_4;t`3)@N%A~&N^J!|3fh4$V+XAp zC<5F@&?>gPuk9{Blg%vR=ew8@{rT|y_ZbrJlldx^W^Mx_oZ6x`_FaCpa(hG_S_6zo zW#+9YXM|a&sbT=6i>IfLZb)`1=hQ7X9t@^N7Zg{D&&m8S@0bjoq*NBO?3Q!<0-iWH z)gV5rYjMAVB9 zK)X}TyjD~_TT7>)I#ypFSf9Ixda~hn(xp_Ers$-o=i}8R^R)M&ZgfuY5=(f1y_l{S zMc|=uYnZ&A55QAx*R{pG&)F<%hP$0*Ytu$cwmP}looR>1_KKo;5X18Y=lIqYeor3j zbI7vq^}}Ckx4+Hw66Nznd@Fv8@EYFLXTsEM)N)HCC* z!v|dll+(2tF`loRSOwJ!6}1rU-}&3WSjy)B^TPa;K3tKSrJ4v9

~D3);v)9A z@_YN9oV{lt?`w5tReHe*FO%AJkdNqBwt`u$b<$KY7t)I_*)$}t$RzgH4PPJOgB`qQ zBb`Mjzt!$?qpl$*!zB{Zm{fLf1n^o>$1cb;=1hFzp#;(fGj!lPltrz#O&ZyzLFAx0 zM`Xr*aIY1XLN5V|*5_4y1#HC#=Rh7tC_eG~4BD(I5+(%NVnsurGI7+(geNX##I*+Zp zd>841=D3c_H*tRgoWl+^U6B=M+&H&$Pi#;P0Pjif>)w3Q-`q?+3x#ttsFGP2B2Iq>lV3BWl8(zV7v~;|OnmO)Jxw{$_&N0@u2vXb`oQ zHyW$v*6C_yM0i0zHwh#L+Tn5e4RP%!eYutb_ph^8X@Pkp)FVdbmD693t;n+vfFaDK zY{Fg$h1osDmkrOWltd=Y8{O??9UuuNGRKtHb-?+!Os71{mv^5Dh$Y1A;o+)J5E_{kE1P=UfxX#f&E< z6HCnm9johJ19EQfY<$tkVFO}_D+8=mWUkNJit~i+ikR2ne#NED?`tH3{LU@P4-We5 z>mV63*2rghx;`%qUq|bUi^12o`qrOV`+I)Navkw?BgfPX-X_PhflzEVmW`3*CQW|Y z8(-;?x3Z>qz1k0W?l+X~7QizSh-1=>hp+Z6D{D9;PZ*5pxg*&Qj&m;NesZ{sne^9%$$zofQ9 z2f2>2+`s;lKv~wxQoztFjlWv(jA2o@YCLTk)F1rWM2_@2wICyYJZKZyJ@Kjh#ef}* zRcSA7wyd=E1B}U%LbgaiH+=cB4nd0#F1~A3QtF`!1?WrV<1xwF9I1J>4g)ovE~-%5 z*oemuxOZ&xWpS49TzddD<=&0q?I3f1w$I~vdFJMKhnDLU%iOMC?&`~DSh=qTyplo} z>Vh*_pXVFic?=5eyxgdL$%fL$m|ov>qlYas?BbI#U%|I&eN^9ru)gG=azG^1CqW0V!<2W*&<8no9_H6 zRo)2aM9q_}o~uei{Y!x6{_|ez_M}4vJ&AbMcIx%B;Za%U~ z()V=ojw(gKi{ID2xj0fu8L@cGu`?j3T8E9Jw1nzO(f#W+obdz*#WbzBB-5C2&Wj+P zQP9C$`}wl!C1(E|RBL7ZfaSB{zuC2I>*tupW6Owz8HQOYCVt$i<%|r3yMq4n6m-X# zVwY-aOkWR6DfsVZ!y47Lg1R+ymaM*dz^c__dsFx($%=E=3+L*UEXa}C5D=C(CdRJW zR|Jhc4aWeKuGOS|yri8oq2|pa=dzozJ)0LSGqEhuZe;I@?3^*Z4=zN$&+AR%DYg?* zFyK5cL-PY_O=x|-`U1tWiJ#!X^ZuF78{rhazlz6laQ(q?Wm+E4c3i#h!BX^U8yH+h z>#C4dm-zAOj=4QFXwg-ryyXH`XMG=m(OVy=p%fFq24~goO_w#DK>0tR$Im`eol2$-`qHwmlwZtACS^XC)<2i7b(ai)}EWf&&5)C zjh9)pi|Ua9|36R`k4d}q?Fp}?U)H+HUUKYCpd%C7P7SJH1a|aG6lF{XQ0}0RxjV~_B#7wiz}rh%OIyGbFnL-boJl!6>6mz znOa;0jtv~*t)G&e4Sv3%EN)QVle%i}=h=Wzo8000+@l#?s@<&+?vdQ=)oGY7;pf&` z`;Q%)B|BZI3$nYPET7C~UY08j3m%l=LfsB37+()(?w+G9Np}P!f#1H2bVaZU-C*A4dl<&SnP7qRVRm5x=9pkZk1 zEt~%6Md)QbFU_+L;hvw->c}CqHxy#gl)!Gu%yP5XIe64QWI6kYTDOAWq)4i|svlB| zz2Ok$NA(eH8OjHXbtcYuwv_npO=rLW#@iIto-gklS-fv}<%M_DD^&LAoG)1g3~jau zs<&oBZ%!uOnuN3SWfrG}@dc=L`gs{ruD8TXdFhPZ%j)W>gatOXfq)uqorQY=gS? z-&dA*H%zrvS>D|ekM)ZZvb04DqBh?fQK+g>zNSu>k;bZd8;Q9m1;5-Bv^T!5O9b@uzigjT~$fY z(T6Ra-s@{g1(WAj(>QAY@RyyLP>jT@82l;TJyt`m$l`uKFJ%2(-Xp%;z|3^TZ%lERBZGYN8@CDp z@$X159IWG*2Ancnr=h^`Dkc;y>uoEqvp=cG^fn%ElVdZNLLAA#G1)-3rLIT)zm&clu*qBLty$Mqjsgaq9hKa#npK zb>#)Qpk}|HdqYe@Jxe%~`sr4FsczUsds3&GZvo-jCSoDNPI6oYpE7vP8pN^iMJaH- z>9|;@qjznyg)g!y1B5XyM|z)h&{>%T;lV|?1#7C#A(;2wXSC19XR>?CtKGWfSj)i{ z9UU?6OuKR+>`~;&tdUj2scZnwz%8CMSEzRz)VSh$T#8{mZJ^5GrJ>f?V*Ef|9-b zQt@wUIv@gcXn#&*nk^nJ+^%SSaliOW9spwp`7t+9j{AO7V|dvB3m#!Id-M`5FK)$S zr^V>@B_2e{WKb{4^WwU^Fzsksh{t56u`C$P04#I2H(hM+$Qgjxu2P%aihONlMGS@Z zVc!f7c;^0R`JP`G4$1%Mdg~(ZFbua;`C?xku>Q2!sY3R#4 z>-*;68)x;XGXUAIHc0zqV}BFU%=1YzREHay-gV!*Bx_hR+Jo|HmWs_B9oh_2%-2mM zv37<6*Ij+1G^TB`1sRa>eJ+iSE8`n>ez`TDaApo_T+Kv``JiW`ofnON?9Yzm6JPK z+Vmc()R|u^ES^!`-Y7h2qlS8g$}5(3%_$fJ&FL`va(&oAY<49xQ!YypO9A0JIKb=m zVXJeL-hIUS8cR>!Dz~;>My!?;4JM(rhT@#T0AE0`L{8zZ^Z`yW*|SRe1>81-%fcU6 z6RXe+H62$>58bzd1(r}o-k7G(m)n**Nv2R2CA{N<2)G~KPYkVSw0+lKEx&!D_}GpwV0fMjh6<`1 zH2#PHM!9g=e1_~Z`R$Q;VeTXj5#C=Z(bKmF<3_VNhHtWwFP@M#(eV`@@l8ei`j1xN zqtxpFN72#56mF9OV!vF_68F*#udLq8WI>j)AoW8TUlEi=+=nH1|C)b0R2pjrGtBMN^*Y@!MaadT7(UiA0x#` z?Ca|(6_Qi}*LS-_lTFR?-*@PTkH(VCJ*o9m9@;G?(0tu6rLiAzWgrWWs^tf(xzWmv zbSl(xccxbDF$5iOG|jwbM@n4oHHK=+rnCoKZL+47HOc5WpV^2p{H0Fw+kDr{le-E4^}S%;Yxrepg6IBDDp$XTJ~L{4KXp zsPt;tjFU(s8Mg?A+_2_hM@LV;6X%{DTpNHkRPyY^kQ3o(3D;kk#0af|I3d!w$^*HC zE5Hpt!}5WcF;iI?c1(E>3c!1zAf<8=xt-&7bSH*hIqa%5j-+?J+~V?l#6GOSsF+G$ zZuIGrW8n1MGGPyvEpIm*CE=NaTc6>~ug1hUggsI1eZElmR8WO86Bn;G-uDUJ!qLKl z`RT>@^YbOvU|spmsUucoCe5@eJ2;fa+gu;Ke{t7!ue^c4Yw|d`K10oyy@H^S5Eom1 zytHI^0$gxp<>RfR9JnWjCJz0*mSQU5Igwk4{nk#0QHlz2gBDS_G;ytg- z(?UF6R39!MBX1XOipSgioi88KQau^)tMB#ev+JI3?#q_h7o+r?rFT2PR-bvVp0SL$ zhdSxBA9Qv(sxoIzFBc_m3)ZV2bpyzlxJc=`@g&IwQ#B-0NQ6~mpN{(B-S=%k%rAb8 zE%YL50AXe}tQYp~kAn@aR|&pCgUu9VF@7EDF|EAD8&r0$eWjCpD+< z!-=c@yStsg#FPb~NcyO8dCrjQ4)-%E{d|4{ss4D6nJ-z>eLudTH643xG#uFXvu z+`$b?GqZhcoqq@4f*}e(l0Ly`ql6xA07##!sJ#Akwh<)J=z~Eg87} zB!K(EXB}x<;iXrrpP$J7 z=|RP-(}negy)XyG`GRs^&nQf>dhY0y%j3ez*3e)U;>#W()&~Z^M2NikU^IQ)_G_EX zX8=?(7~Jjtvs`Y{o(rn z-l_WFl7OBK5nAlFucHS3vOd~RM|XH~VKTAU@xYz8<;FTe`=oe;*7C4dnWHNw8!prf z$rp&ep&k63_b+@rb)m00^Kr4!TG zS765PJtYF^YrZS>?wL}=yBEuFQ7IWy-$I%G;VTkTDcakVo-TH92Dgju=Ciavek+}B z`1iO(b^ri#nT+pmIx{Gf4Uc=Reb{LHwUcwaX~XA__eVG0-9-BA$yjn=zQqtQ z^M`fzFIpoBxi>VytIF!4Q{pV&C3e5_wzv`XPiK0#{V6m!bcK0OlMZns?BF-os$8=WFWu6R9TraOBVJkwGQ}!jRv+p9}3!a~YS|0%O4v~G47wDmLawr=q zm}2(5bOH#z#!lxLRwouq_#u-u>s3>lbkqZ5=U}LIeI4<)jSF8D4m8h?Aeo4oK?(3< z;gl%-z3obOJz!lQtP$l8cB6Msi@4la9~c;OUlw_D`=flfrb|#DXyf>`PgLsUspQl4 zH)A7j`gt9Y;6e2!nc0o`-rdh^qhO#^*hbD~&3HI@bc>OuzHcIVX}zE%qTC?a3HURC zvNo9YI^35)vX|rFb+SsL$Us9`Hj871kpT~u?iOiKAfgeY4nRK>@)f}IgS<>5U{9fp z9W#H)Sx!shfxQQ;8lobZTxs);r#p&f2+q3dD&X4Y(R3kf$t^ry4Z@Far{wT@Xj^ho zkxFSjRg9gT%v|LgZ~@sj#tDM165rT%W!fD}V>5NA1;r9v9}jAUhfT!>?hi!>ic(Bj zkm=E7dG1C~&oCt@{y_4laIZuUr)thGNh5^R~quKE;mAP zTrbn}jvqSvGAhgKCMy@_bsw@vc>~FwN@osD@_LSkqw|spJ4GWE=!n2r z5Rs`jPIClYw`(Fj8Dh>DdJduR)_lREM1b2(N64BtKn=gpuqIl>o?m;o@Hc7jZ;>A^ zkvdSneo4gp&F)tD{EvD>=v(Iya0bHCOY zix1p=-c2f(CZE~K2h?Ky9=S^0`V0AzN;;1_N{PC8q!J|gc=59l4la}L#ixSu$w*E= z!}b2Wlj}Ls1kb#5ZsxP)i(KkFz7kVn-Zv5#Kk;o3X8&$|wN3OGSOf6$Y=pYKVHI?e za|XCvv1U$eNOPbYgLvM@B#mjT)&(0*heEc`JurJ>6R|Bu*gKNMz0hyDxm7-n zEPeU#8OI`uUdNd-bI-0x=1sXn2?{`}t>WfR@1^dB2)t4UeMZ##x1(uzE*;<#{Hm?K z$|Bz_fWGF|;yIM25zfbzCW2`QzAUhTP>UT(vOMpPvVdM*|ImFnhG03kY7f4Tz?6#b z8O}tAo+SJ833`C`i@s5+bx?<)o0j-z#{7A|;2A`?6qcaVC2}JW0)$J-J3pQ10|3ya zp-1T5X7Dw;utp2uLHE;AJAa-wzTdAp2|%d;w-SxZZnt~t+YObhmR=IlT*lZY;YG{@ zA{X!6M?~=3=CL`hv3S^b$>WbdbkZs=@`Fm1Lj8&HV3<56;b-?Qm&zXeehS9+`hHSc z@6aK*;|S79Zx4^g0e!~Aj8lPF8Jl0DHOc9zh^rK55Kvn?UE_{?+$?{G{sRPcYoT}O zTUk8bNeCUE{AtFaKl`ys*#Aqd_4Zog_MoiK@BX3Iy+G=u*^TFUj{V{S23^KegNT1# zk+23w6VE_Xpho@(5Z~|~=`!C=dIdQJO4$pYqxZyc?Nx#_etO{GR;l;pZ=@%bu|Tq( z!r(D8)%HyApH27<*V01vPP}Bg2=(uqhJZfdr$^%|3r5V5lBg+3Q1@Ae?X|pCKU_dw zkHfw~XW~qtZ4^uT#>KnCqT<3*%p(sjgL3acmUu;PrS}|d}LeV{11YNK_Y|ROX;1WR3`vgQphFek7ahz{hoHC1= z9C1PoXpMk+yIK(LWYwg+Q zd;)()z_j%s>CgcAV#IzX)ZRf1nqj^>^iHm!m+r#6q2-ZIX7wjhPZW#;5zZgzp^|C( zH|+x)szo=ExkLaSAeqvpP5X!|r!>QwSaEdDjZaTOuGTr7uEV z|C$E_OxHj7^ctAFo8B}VLA?}h884Bi&qO@EJ=Oj+ZxMRK)Fyu>*cyMy#1KmW#55no z4bdb36%b{LV}rV&jdD=o+U}BcQ%~w(81@1Jb+AbSt_8+WBU{5LhH-ozPx}y8+jGFM znP7Yb-Incxt3uU8P$*l2HQP86_*{2M3}0z}=Qm&4sLUU{cPI|;X#50oj^qw1xkSRf z^fVAYY4JAWF@H5AK;l&ar!l@HS>?Vso`i1oZy$m_37%qCR}7%vWcvKpnu;0#Trtxk z!kAPCt_lz?^g`dES5ZimbXI-<<)$yotDk#Zzy*1z1_{8HIldMkGS`^?azE$w`?&qI zKIJi@FS#$g#9aUz4B=Dd*lZ^QsMB>~s~;CV%}*y6?gbBSZ`0)3rk z=<`7W82li^Qq-cpC#X@la~bg{HAj76q61#z_IN!ynIRM`)@i7}zzl2+V$W=o)X?@x z@5T;wp+j(O=mim{Tg@2z40W|mF6>T<m32U0AN!9ZxjdkvY+RPldD<>IDB_11mijc zv3jRGYMC&~^yDeeKU^Jn*PGJbi`&=>0$z>X+C>U$zfU8o?JN+EmvnlcAH4v{w`&#~Z zUBR0D1$4kh9LK?4V!g7Uj<-94{NVn)j=_MQmdxLnt!b6lsz6>*9CG3r zT*i#BFS_U}?E_#N*>E0!KDLkAr@ClMvJ0(sqthXr4gCTvGSEJ6&}&m@r;1U32oSns zAwA7oN3)cSLhbWAm2bn@o8JVq-+8@ongdPQe?D;L3Bq7&f5=WDn&XUIas=VuE*{ZYPjsqT4#@Y9pcWMSnBq|gkBKlK*;5_;~2Yi|wb384HP z`qIfD(Y)X5aN=x_gQWSmtL;HrGK=KiIjP2xlk~c6^H|p>rNEe)NFgrQ)}_4*e7W4g z{ke;n5v|4^vwCGE_2S3SZwimwQ-FZHEA%1~-XzZ++CU<929D18lOFHgJ zlAmN^1<=a}OtEE82<!xZZD zUtiY?U<&{bo@2kK>qXxav_`B3BIJoB93j$}}R=2zpn8gxRLInK>-N;rg>f2ri(bccE9 zT4tQVoPzvJWdM8R(VmA3s-=NF&mPbQSURDKH=#=klGX_X)j)OUgYu(3!Wh@qLyU^uadgB-&daEw9jvM?T% z*4PmHDlL_r_>FG2f-2Dbqyb(c8F0h*YfD+E1&Lwhh+8LfzaN~ z6;?L6t?pd&jpkLYA(%}I7X!owO@nbG`1wl2V1%za^jE_-5dU4m54g|#%VBD`65jB` zL)YimG||(7HR2=S9yl-YnsW3}-qT3=fu2R@>_e+H9RwnhFx|_f^Nx$Fx3WMb-zV4* zdYRHLs&{oNc8jRO^8Er9gagpbe_0dYnPrQ`i-Mq;9=vPQzqk1|efQj~zam9WjLg*x zM{{*7w+A;EOil~~g=i3`ixG(3pK+-jj4j*t9X&tx_q=P)$x! zsT2pjxK7^kP(_J}*N`Sx`#aG(^o!KL&pVj8;6hmq(9dCS`AVdQzF`tzu2A)yAxs(u z{85C*(cK7(bk79fryu@szf2755~fwGxGjfxiA!`pYjRb7ya0q1kcDzUhz||=*@`rR zqWws_1iz4usVSeO^m%jcNWkbH1)}UW2f2~0*?;z4!2evj34Ut>Z{xR5@vsHw|G(dv zoVhP7V4rD?65AF2d9;GlyjH8b-~aYs{^hUVpT8~HzMw-JRsG-o{@4HhkAMB^Z~u!> z73-A!HCJc$7foJvlD+)(t(NjH?62#d&Hvlq{_)ZO^UvS^;~y9NTbC_S8Z8mH+JFD| zfBdY=urB{?Ife}U``^-P26sIE+wXJ#@=ss?@BjFGA9cSE&Pxiu`nUgw%goD^Mcu#u z7QV?py~E#>PuqQp&-B^;>#M7#{rFi z8vg4RwQKhKHA`O@93wG?qZpdGu>VK@A8qzpqCSnL^e;Z!s%WV`{ont>2>il;Z_yaR zU057rxqto!PJxyivMI^>rD`+qaT)}ef2C-g`sa^nSz1NlVmL|Q)W2f?_g^yq`5FX4 zVC27E{?j#-*I)Aj9%^2yMxI{eS+>|ME|r{GU4M-+KAqu2pqWk)cCh zz`m*8H^cs36tVcgp$C|0`EU#HQ$2(KNYecpfWL2HJai5GfZ^MMC)rMnhnEx=j?lX4 zfp33fk-NlVTRo+)O?rKZaq?8_V%4f|5lQwXku+L8xaFTet9ING9&CZ2%mZHg?%$rC zuYsSq68y&uEb#+C-vI7!mzVZoy=}5Lh%|feE|rrKtFiN`^2SqrYU7<>$>GCbCno!n zEUV7@JdoWNP-_fbk5E+)WLRfhm)A1lZ4^$O+vbZ`c~KPMk#axAxH7(9UOP|ZYxa~@ zR8RhTmMqF}&P3>4&h>R)pU?E{H!Ugc+r5CxfRCzc)nNVotQ54CeJvR`P+@HwNp7>OyWp8Vge41N>B`W#BJ`KbXjz)CbS?8pedPOu>h^XmrqYa$lLYHr`wyVcSxz@%x)1NoZW>P ztp#u27x+1A-k{Gnf4yA+Fb`a~h&a-*BIz0+RNqR!je4XjO%)b^XfY&1pr`_cUn(#b z5OxzmZjE_w>!>yE_;UYr7e$S7z;QDb76~aWx54|^P%J3%K=RxkP~>rce%=b}oaYIz z@gC*t{kiWH;}}j)VL6pnn35r^dM}1Y+|jEhZ#ou-i;bejcqLqY-osE!M7}(pu>W2W z++{I#Xwqkz0Eq4do!)DsuP-W~=Kl40i>4rq{P>a&fIvxVMX!xgU*66H`bYqjN*!N?yMF=ebFTZdao8N1erk-dFW_aF_gj?YSe0}ZOv}^BN(@NGU&yaoIb~-Coue?rPj&(b6lnN6Z~Q`t4&s%4d~t=f3Z1PBU{WzdRX&J9n#T zq!^cXwM9#Jf%~n|J3F{lTFKOCUtPB8N7C4km;|eIFJI5{={`QKJMQUaH!lA|FxvogJd z={Np`X!ZoDqOiyFnS!5R8emyP(aU&hhZj0c;0lSM=5DB5rkzSS5w0yWXPdt=0tL4q zKtx?C7TS6*Iq07W@4-VwvBW7E+9PnEyL{I62T(NdX=~ixRX<_$rw8;sQy2Ahal37X zpm?^Y4du_AK6Ou+R}`n7u5g2n=U0o&rqLS@ge?z0L$Jxq5RP)T#zBKy=E%Zet4N?%=WH; z23%1pXrdTcw)v%d-4#F(XnRjlgzR~W^Sa_9F5=)G`EMiYMQ3M*`lB3J*0X->ZUQ<4 zORkRxn9H$!!Rr~<+u-TuuQqb7LHR_?nkwKzBS|ZUxx+HWN(~pB9E56* z0tIw-pxOD-fVg;iQ^#hbF(3 z{F|@WVD4V>TU|i@m>(&V5u*$jdSA@D1yj#k5or3nTT1Dwbm8}qpbMXe$PY&APwm&f zf8KXhILKMA4rq4p_-GPOg&B<59uC^Jd_-b3Wgr%FKQr=0#N+o`Uc4H*!%9Qm%stgv z&F#9!dhwmTGt~YVOO@pf?%b>^t#?pRkV$EVL=4czF^X_66cDs$1v2LlZ0|ZT&OnPm zR&0E}EC|8FK}<_2mc2j>a4q1b)%!T-&6GsIt4kLSWiW1`v3&&>pNkR}f!Z+VCE4ap zj+*xXIw4T4_si|D;6Pi`7;acy#1Z?z1?WX)a+EKum+%09aPS!fiS&C>gz#CZVCxSG z9y%DW*FvbcOjYoDlezxGJpf-Yc-sXJCe?NQf3f!-&8{**l;96yf$(UF2MH_TcX)&l z2ycG#%M>(Z;PYx4FQuJa~-$G@+0EwM0V{@us!1RMTWp|B>ZBFUP$M!GZUi zghWdvc$!gEOoXsbf}lv@z=q;^Zk6rQ%qX0NEx$NVgh@ROrUqUf5h**D9USiZ{O*A_ zzrH?=6o=7hZzD#pxgT;pPXY`>d_%Pr+?5y>V$o>^?0%+m*4;F(^!E9naQFfgJ5AWL zRWFF7>3YGx?pF+GMxfT_3o|b01Nyp*0ohZI1r7=7oxp3 z$X9>h+HY{TiU1uUmxHJ@{JXiT!*w>$7Il9C*R>7k(Qk*ei@xms{vIPEoG`gzATB*a zNMDAbgX>I(a+fY>v$eM))1l4jx^=OFHVa$2xC`4f%VYWVz&*>!D2H};enSN>OoDH_ zhVW}Ig2Ti=A;6FR7i0ML*puii;(O{@BNpffWpr)Vp+e1d8*2+r*aINxjPTw8e$LAK z7+%?Rtm_@Ayy5*T0vb@&w#zI7dO~JC8@P>Gg!uED=QGYEvOAO_boesNJFi%Un*v*X zE*4OP+wb$S;kBE<)i2(KrE(wH=@qa--#aif^i`=cJM29;h!7ky zjK1dLaP?o%;mK(tEc$6s@Ysb7FbAOmJ&V7;cc+3)j$ifRP_JT8EV1t1TCKR;m zX9Ae8HaI}{4-p1A9Imc?&m;@6d0qhEMAJKoA$dGrGm+Edl^|)bmOF#BT!8^#F%dkG zZy86YOS2k9Yps#bNjJfPfuIZeog?rhR8OQhgQb`2YFSl2$TPcepnTt^=q^DC+}ldI7gxKXaKjI9(#P4 zWWINPLi$1livimQL+oE*xAtHE$EhqDsQ)jk!|+a|$nC5GbAn|SoQX0oUmW~4Ye0T* z+`|~4`w;aAT!GtIwt`LA7g4C4S#v#1hQ()7n9jlHgFN=@RhGcnUDj~ia_7ofNR$~0 zZ0awGLoQbQ1_P;_Axf`@U6Y8}nBjRG&umkAOmi;?fp-*8CBpuJ0R>c9OZ1{c+?A>F zUV(LN&Sx2y-J4PX@?4VFGd4bZoS?-O&UVl0S`uMnR2>ULIEec&U44NKUd<1ay<`{z zfKK|WoU6{o0JPW$M(++~cfYoJw>EnP`G75-sh5{-6A^WF=FO8|!% zu(rBy*~h&d%N31B`h4-@$}jqTaUm`RYr?LqS1Q{S%De-BgCPPyx9-sIAHl*)|Am&# zr6_Hlhe-)NwiA!}q%Z>W?+MXtfC_H)zy!2)VVtT?J}#3phV)s$nWHiiEp+*X!vshP zHtKRFJhw|b^nIcM+$G;3rr(WF4stAxw$I5h=PAwukz(izdF?3SFJ6Az$^zMTRoqEQ z2WHFDnd?H-%tBhz^~Kc1H{d z^~3=kRL03CpVUeyih|SJ+l8F6n8wq536?FN_7|Eq$tWMb91MtZoS};wyJRlidVPza<a0o*{F5@%pB4cLPF za^3Op7HpoTRSq~qZF7B@KyNUp+Oz5fA-B8UjZ;0P4|GQ6&i-CrUS_M61C#A&6I}Z0UEumdUr%(!={#)O*3UTk zhVKZHp(qLPA4}C}Ibv{F)gA+^O+Q-wUM9KQ=Nq{3d%M}3dC<871O`EsPHfNL5Cf%S ziL4B`owmQ>J`&HnqPwe0+SX6L5z`@O@w~VKCqVcwbHfi{{+c)s~f^m=>lqb2Vo77m#mf z*wcA~iEIqudWKqQec2N0oDkuKTp560kOt6b1c8ZPG@*RypxyE{4T$gDT-i9*@Ni0+ z?7EzB1&3F>B+}v=*a&aTQ^v#0$BR66lzuGcqY@mXHNN+7_jW+S2AW7Dpir)`P*i$t z$%sA{Rc=&Za|FQok^D}=25_*g;ee&%@UCYVV0Aunt(DB#q1UMK5w;K|Pk0CW^=0{H z&^Vgld~WX{d^>Q4FUP3_Lu&b%6VW2K<$dc+pMpg$h6420eXXY~D;xwZ*5tCp(IvAU z8jJyzr7giK;i%HGvYcN-NF8v1^b77$VV8MM`?}l}H2mETmILK|=87S6j+W$r-M=s| zrG^LdFc7;XlM+#dI=Id1J24ba@t?g11DakyvkuqfEAcwc?+88IDd{uM76ptIoT)Yo zs3n~|#5@LHc$sQ5Il$g-O_$@Ocj?xHxgh5bBp1l`UFQ>+z-G4G(B^k^0&BRWAHU!} zAFH7nM^>E=Ao12j7k0G5G)28V3BAD1`$DtrT9(9<_MdVx&l?$;iwe}y?2!d}0thg% z)5U0!RfR(-0N+MlaDAU%d&70J6M=;wv=2C7Jc}SW0R!QB7sd;~cH|@gwpM|?65ysd zu+KQOqkRbYQX8Wf1ce_RCg4FV$9tRwX|9Fja_d708H1PjuV{r7N~%``aQ%p zFsb%F5ho^_HVGRau?fdH_J%8#t+^2{e}O_7Tc2c*JJ!VSgBLFhu)HwPwh}7Eeu4`e z76TKgwp%m@ClBOzf6+8O!JxNNU$0kQ`nd5R0sd$AA8=sT2E2^!fE}R`U|_c(!29QD zLe6+escQk7PToa{_AFJL!1w3o#+cQfTaCk|G*GHdIv;;J?=*I!;mIrh`66Hce*4)- z4shc46A1ogE6Up&yU(ErAWF!V;2*$+kq$O$DO>;Jx&R&kpmCj`9{JYBZ3ds)*k1w= z#MiHhV31{5VbQZ%G{x#JfZW6}?G$g=pwyoU9H_iZQ4p4Opx@5iVN-1pY@^DPg*w(3 z`k?{aF-X7l0B95?j<4wfIjb&|<+dbslhw=pCJTrQ4z3N4X#~LD75W;1Vs_S#tv2&d zC;y%q4pQN{Ozm-ice-XA(_+h!YXOW|BKub7zJaRdYM6+#!*zqnU&FO31-qne3yr6( zc)Vkoyqe&;$sQZ9NvI;4p9Yv7;j6=kHcAhRK39*1o(?@_axAwJFHzAf6Ii(qoRi3Syj>rSpU_tr$TVdiZ z#CsCZ{o)=&f6I4XcZOju9ZYb%F^^<~GLWt$la}Rcz|YbA-su-N8K>!jMZV=z9K3&3|8P!iAPB=C>~!!c!7nOfYB;ft9dxgi&~sFVV>oqe|tI%R%;oZTUhxi(6DJvZeeectEB;dd%SqF#lp)LB0k*(S{p6~t%zpH-(I6vvP!+S>} zmp(>}MTd=q7=&6opmMqf;Q3sJe6&ujb3(;~t)Xu%0#+~uw8SiHp&2d~128umc0h)2 zVKf9c#tx)zVMfs*ax6C$BJo9l%Xt85!8!OoDPly-C(!{ZPA}_#+5CG=>vKT_s9M>4 z2dmVgCGnqstX%ca?G@GM$0BXyTMNP%AKFu-6>&!_11~LE0jIlaaG+#V*kmM{EJb`& zSn_?&$DusMG>x_AU2WqfP0)~rJQuMPRlN!@Zk5VlLGr$x^S;p{7fybBxj@SaKzK}; z(JVjra;>+daUy_Q!1cYU(VA75BvTSUD_3zUN%qtF2Dd5PvEX$T9nUyQo4H0`VwZe< z*-wWAtFGRZdHsLI9;fVLy~a_2gH&O{_Ji>+mPf&D^4;ylL!Aus*JO3knEZ0$({aSv zEWxDNS}W8V1p>^Auz{Itqd>+^=W3pLh}u>Or)-BUBb*s#!e$W83*eD%&;B}-zXCvD(PD+K!EULdy;EHT z)}i=2lE97TL9zZN~8gkzck0mIOwv4*uQeYx_zItp4g9< zb{OJ);bG|uWptmYD80cqfgLGSe^_M$kc#R8Y~V|J-?%6E+RjNtce~op%V$RIb3@?J zV6HM;>2Y7Gh-#MrKtR90j}Pt^a4fhV&x`qWiOUB*%Z``lWLj(y8ktRyS-r94J_0q@ zFEAnV)to3YnDP90=c;yIuHj=?!Vycx9GnaAd7M$HvB%4~4HZQ9BJrUI=%b04%YAl38nvUU$m*dOPRKjt4j< z+UMfe0Px&|jRt8mnUzhbU<_N}-hfrZTz`2ZJq-?RezksHINn3z4K#zgKmsVNrieGC z0AUw_)o|Xoi?w2jx!`*G^HcqNU_c(B-vP$*et_&T7!TN7CeY*QlRnw!5DI{#rEGDW z=b>Fp{sE`xMsQFyHPLxO&p%n!nz_kQV!}C{!2#LPI*k?Q(iUr&5G7-gS=I1;L10JP z6a24j8(ffJoHB;-sQ>3{pA#vJ7yR16K40W=cfel0=u;bZa|QSXETWsH_GDNvzoH^v z`s-rp?F)Z=D6^q4@IE8-)lS;=*LbN8b#u}?+|!!^E(~Qxc#NVr!^H-wcVqs~YyV4L z!$o%j#A)YXkrR|n&cBx7wv~2$0$qlmpCN~%8J#=-2(}aG6veASJL=nxH~|(;E&jB9S5pFX>KdkX0v1z*WD1_k3?8 zhUH-ZaJ2NzAhxkkSa5i)pf42`k2x?$$dy;3Ies>zuwgC{z|;3Hrox0rkilNDkm;3d~mfZ6fT-|*<*5GSCtfF6WC@Wiw}VtKOY{jtOKh4H8D_CDjwm^HK9 zLqJi_KKCoY26W6gXTtcdCd@tJ$1lNw;NAzl% z_7g?(@B5M>_R91a*u4H_HeUGi$;=7S(+W^Jo_`xp<1l8Saw7XK|6w2ee3DCa@s-Et zH(Wegh-{_dY$YUN;qW2e;j={r)>Gj&p$#F*=gBQ5(?9Q%fr^#*=HS8@BF+X`P=bGWvYSWiX$9)V;EuvlrM09QxE zntUqQl&j+v=1q9=lnN(U??r`!*82b;)TV}MEI=-g8;nNwaPU_pa4=b3JDfKUi<#L{ zsGa2jV6Wz56ve{}Li?AdnGMgvsEKBSpA&IX+3*C=+XRBkY*(}hYd~SK&EO!-fCJg; zlDQ(Z365k;GGo^ns{VeeCjt2S(`ubTPUD4uLAXA2E;9E^aQ}#WU9|yf|8iy@3|<1- zRSU1vdIP}b{_d4wK1F}$D%2K55%YZ++K%j_h3><-{ft4s{BNL-{?BXwf4l~84bX&v zE{SkhFrX+BA7aP$OA^CEkg{18 z1leF*iZKyXMOE;rYL9y7vZnv^{3S+!v9Q$U&|p!R2ur*&9eP3M$%D`Kc*oZ#PS`IP zipNHGq4CB8+>o zMh1srg=0%>ENY6sLfQZ=Z2bfraVF6+4EtQ}Q(Yc^m{TaHVZyAU>uJsq^<$g3%yxS( zu8mv&lxwd$VIN4YSN`aNBXK)01ib{FVg)pXvOj{u+7HNqCOWfx`<3^*0{N9OYzo9` z?^nSH?hLRpuoKhH!Pw&kaG_-5xd+EY36rS?>@-Z>bz_?D7VZLlh_)t;B-w8q74Z;n*4QgxO#Eci~w?0U>p+;-#q zo&7IR@clqk6HNR)9%po&4omuV8yDjl@11LbMwKNtQPPcodk{5cJtdgm^+ar|a~BnM z`hYWGxQd7KV0L2MGc|q46QU3K^~3NDhvm99QA7WLJZ~%m z{Z@cS5%g7;l9kZg=lcK#`pr#hvq!HN(BF`-+yEDXnPhS_e}KumU+>>c2li*(VPi_I z0V-ahmh@YZ*;5BL2unR?{+XB9=lf7O=rUdu^{GmBk?(nzDwo06rx2L+R__F-zi$O?J8hnsQ@sibnc}wO8Lv`#X&nIIt2#rgg7(;vYz1yv@HD*mSO}o3A!?nIZ z7S?TfHW*;ifkDg^3FYWPva4nW>d7YeibRy@(Z3MRECQSCaDHB}9uI%Kqkj#L_!uQ> zvBP(8nI7_m5FgAc^{=34jJrYU>l$)cH2?8~`>LMx!_AIH<-=rECOhIYHGbl0|K_+q z-RE=u64%}SAkN2hMC#n6+;;75dmUbzKvAJtW9Dg{+^ptA7tcJ4`{LMYrfj-xKmShN z;sB*|-1qjF?tCs>UuiKP$47~xpN?A&M9~LF;LJ*}6$gr2cVJaP5W*Wa~C0v(%18 zgk0x^09ZfDd#DbYjBIbGCsdLx;~8oBpK?R5I@gImjvM0np{2zvx~SXHba4nB3B{Zh!m#RFkK%T26=yia+4ec%ep` z(1u$1Cwo(S2+6{?$Kr{dt}kBDZW9BrT`?{d`O||D6{+PlS;?~|BER+S%XUd}O@alW zJ8X18(l97KjQ!{5JA2;cSQC_)w#(ujkkm}q9xk>x!#BzN!F8Cs4B<5_1w$W@eTCsL z(B|)c48e2-<9>4R=-)84a2`@v48B=IqH%?Bh4NRc06#pH5&+PbyQzxI0}z`54psOn=ai-o#=U$X3MZsn|9B+JF z>E|}_F|UP~1(!M&82sjSd!Cg&*;dM+*SbRymO=qeF4g`W=&XI7=V8FUNlcDV(yy4c zx}<*VV-k=PAlsVu*bwGkjnh?lJ)&jH7XRLfGa>edB4#?Eg%jXpTy?*}ME5so@h=!4 z7GS(`U~a$eHN!_DCe(WPU|)mb!!=~yT$5|#C7G0w|ARg1>)=Ji9BnsuZv?LP-+_O(|m}9Ip+~$dc8y$+Y!CJjtiWKf;K6(o_Zwa2U z{KGm5k>PF^13)&M7pC4L`uYCy4mgoR@>;}-w5R0-&yTf>PWOGF=YZ+UrfFWEbyVQ< zLEcg5H2m7{fcMO!#C?FV;y6J(I~gU8`Jy~0C^B3xbNw&BK!4GIFL)Oi->)~MvrN4v%Q003LS8bOe-cGl7 z{D?0%YwcV8EjHeJLrve8riur}jTASY6ggSL z>cTyxS$AMEW+@g0a;;&$U!YTL5qtmEF4PlYl*Y&AR`QpPml6z5`7yso`6kP zcnY)l%hh%-M{HDn9QYp}e_r4+2=Ch-%WBNpse_NhWAKG>GL`Ph-zEZ>X>c@)y)0+^ zY5^s%2@bICSs#(g1c=8ap>pfk^JR3s0upxES&Msoq3=egk598tg!ln`$50`D++lxx zSozGs-1_j*=KTWg6-Ve*o_A+PJ!BkEjxT8e*o~?rNnRUjI{tDgFTde5URU@OsY&WF z;suPcM2pB(L1^_?e~4TzdVYZbEqk1r3od$vrAkT%&=|+@R|F9xMSKiWe!X7hP2#yC zSn6MMtgjY%S=TVUJGcf+rZJs%&r}thDL0>RCN`BBS;Lc^!to%Fz85Pv$a7LdY%k5v zS*jDuA}4=ay8^_^cuEzkFl&(|se ztdg(9jw6{UyT{3qoWdPS^vN@r^PbJDOavWQaZ+RingsL7rN@-MsJtxex5#gl5str)-|=wF-`@IJN#t2O#a zhy7w6M?XBfA%6qP1^Eym7r*#H_Tuwx)g7xl?c3P=$EgKX7b=XyoK@|e< zv*t1K(8mX3{f-XoYVh-6Wu3BN?nUM~A2wo`XKFLN*|DF=;_qsFhrnSV9xYDYZ_lVG zVS_UXY(Lr{ut!af*+D?(%Oy!?niWD|=>W$|l&ZhX`t!#tVG}rB;wGiE)o)|QuoiCg zmE?fUTJIud9Ix(cx7j>Vp7GgW(+xGOsUjzceYB3L0|H-#-)rLQEg`aqDw9Rn@drEU z8Q|qx_IhZzv?)9Xa~Hy}t9TdAMha`v;)Dw3*(f4karfBjqTat_DHn@cZ2(FL8=R0a z7QwZY0|yuA$6I!NtDUVFPBL#?A8gfWbPbWxYa6wtngY9CS58*E6)J4V0q_$85FIgw zjhw%}Uc10FX%IGAGyC$ei`oi2`1LVt3Z z`!d9>%9XeuUQq`bfBQy4_oA%%px;O94kvr180L@wj);-6qk7n^rMn#yLGiHho1dK6 z6~O|J<CfSzgYxvTsW9s1A-)bs!P%9h|!oI zsL&;vdxjQ)1Kn*`%9PLP9kkHa+Uolm1#%7JPm6j|#t_5#SpYI#0Vl2=u=!D-V{{H^ z>J!>B)9VeI<@QQDTW;iz3kEpLeemz0#X!&;sL!wTCQFoLyBmSJ9RhZr5TLFDG=co# zBd}UCi~eDNA@E$peD}B8&rzzOFMu5vG+Jz6eCCj9tnbg>iPQ}&K}O|%!PE5{H|1e| zddR&z))l3NhD7c1k;A={OmDI5XP!b^Ao3SlR_*TW!&}U0+#Q%4m={ZCUKp~j_I_I9 zEWw(*`%pH;L8ex@4+!1Ddq{fw?W!H#Df^=@t50o+ zPVd?KFE_0OvVgkJKqK)Z z!e(+k#Bo>({VZ_>bNr!Pm@tuB7TU){1B6<|-laKH7fwEr{nv1;u)DS;Cs}oPk~Jii z&2PGnt565|e!W45vp=GnU|azYL+GWQ=SF*N1F-2<*Bm#PgWQVWD`^P41?z&g&%W1yk9E20`F2Agu}2dbYKySIBpufu6uj4;L&vXTEAo?UMgjN2 zmi6(mUjQ~#p3IkdF8@0#PZqsp70JlKQ0bb>eWj1vn?ZnTQPgD7aE}IfhKIevV}gjF z!I|9KMzS$;vJyG{;_4V*nJ)vgc7!V}6i8Vg+ZwX^kz9`{!fU1A|Hq ztEDD5!-(xk5vWq(M+I+Y@p@tv_xGL3tbxP+E*1A&PzmPUJY-iV?o|uVH4GU;EUJy4 zkMnUqvcqw!NW6)&ivRfw2LE}4lkF_-EDXMzEZe?=>easyctZi@xpOR%-PW z{`}{g(AtA~=LGgWZhOl8UHdV9xY1m0e#1@N`OjZ2-qU=w!qOcq;p68E`T36GzxV5y zo)qunpMOFbZg>au^(6da-M%vY-AgP_a@@=5Cv7>px;Erz`|tjF|9-NV)Po%!ZlP<< zNxmEZeO_2IBzc^*cdJSKJIKFSJGiLNr#Lckf4-L|1`tnh35-|C|MNE}v?NK*O;?8O z6b;5KPMGHt86QXEDoubKZUWRK9@ei@HNfs(QdqAEbNU>=jue9~ZJfniaLxM!3pb10 zt#W|NuW&x69QbW)rd0@K<@&l@G16t7#so4Ku%Q^>`cNvk>rSzfcRhwPdHbSs<#U6P zgHWq%dj&875bVtYC>v2fRG)6FD!!@0zp><&nj6)f-;+PT7X~1W5w-%%INr8V?BW&| z0m4!uwJsPcZialJ|Bwj|zxIVi)i1)I1VBAkHnij?k4Xa9iqW@l*^hJ>wbt%?FPzM(ppG zy8a#vHm0p$tawsmq+k^vHe-ML?^tl@YcZ)XR~y#Al0Cq`L$jAKhhe;_?YvrL=o!|T zKE4;W6a@L+C6sd)d4g#HmUb5cQ$NrrxTS#c(65&b{W5(Ke&jS!}}dhn($|boOi&!<42I=sPg^Bq!#>(Cmws2-X4!X z+ZAgs^Npg)Pfr3Jn%+(p*5^QZR06%BboMV#GW$5+P(IR-cp~wCdeqe}O%peGjw@;c z{h9B&$KBEWvww%Pr8jK;%U$|vvH7O)ef+yO13$7?|L=X1Dj)Q} zJpt++aQHWouEozT?Elu&N~<*Rl87%Lx7Mv7AjWLjtbc1JKAAJO%+~&rzOaEk&hfwa zL)LKs-&@vtW#s)EM*lq@eor*7ZZH`YQkS$9h)Me81z%x{VSO|H)dFwfEpE)>GG# ze{ytilK+{)GRJR}u^e9giy1WI02vD$v8o$vD_uu8y7(uXH3*6y-O7jbE9cU#s3OsreC6bj3Bm>C_i5HK|<9 z5Z_$G+k*2X=jn(gP>=gOXJ3&CW`U`~+3(GYm7P@G>7wZow0yXuybxjqcrj4fkrm$C z05+fTEY%$(sQa++`3q}e8Pn!-b-OAzrYr9UavBV*FMM2}S?jT+f9xb1)%uIHI8_9M zQi)t4k?IKG5;*u7(6K54_AGjTGYW(Gsj>2C18#>g^UEn$bx)HX7n!fGpihe9Xz{a+)-%S!UPA^Fq zQ+bek`G{b&>Na2!wtL){UfeNScA%@TK`=r7{an#}{$S~^&IQ&fmd!BGXhpU7?2do(lp?}DOH3e@9$r}Ah zn8}xhwd7A6ZL(*w`p!_gfrG)kES&i~4li%^Y>)79dhV9&Mn^W@ozD)4_IX8fWsfcy zX}N9v%Uf6p6}C^q4gMv(wwBx@Zl6 zv&ItWskvj`kNU1Bh2A(#%y7ZvVlv5JV0|asZd(iqtelAPN2|97Bg43SLRXMuy*_R3 zBu2uol7K)_3YQJj<;Mlm(7hZl>`n0@=g<~sl;j#vWt08mkj(9Dr8yG$ z*QLV67uDDQ{e5d0)wySSyx$`F_D@G!p+}fo*}e7q%M(V2V1=%0Gv%jGVxz>n*F+Cj zW1l!CGs`m`s{=`(4O8O@M?Np=B=7`jNAl8VGW(Ys+@(VtXKwwZ-++E7Ga6u7Q<#5o z+41ID=v8wD$dCd~>}4zru)LZ6@7L`sjnrbW<@wuQ?#b~Nw^if5d}PwbDBlVk-!ZLk z$;oU#8}WG8L{hrZ{p;O@xI3o=VVZQQ5A_cxz}HFA4;>hmOo0(2{^@2WJ0?a&SMmcFvNiG<9>0bYADIUMR7j0Wh@&Y)`*#+@s00#WGp3!v%r&OaZ zTfOE7Aq0$EXRZI=+SHo!2pEl(Iq`hx4=yo7BpLRx^nO@prRi|HpVOn;4F6U~AF$m( z244SaY4XltJ4}v=P|2%SQr*IAlI8Iq-cZo@m8s*GKGRY#bewnMV(Q~o`mg67o;A$E zvf!Q*Yy?cw1whpYkm(aE2JNE$#Ph5@ zM({h3f8Q}&a*LHi{S$L1lFM(DNIow3j$Z!B=Ab26Lw-G^pNouf{7wvq50-fa@y~bR zh^Lb9CC<{KZ`F=m>)}a?!xsHp!+>iDgnr5F`ek+b8HHiZB!V+2Tkd@@{-1BbEGMv* zX+ea*2HGNh;YDvm0C*t3gZx|X`z!fAA2F+6bmFR3ep_YojM~5&R0*aJT^Jd|kie$7 za5C?n#{j;*i22L}ZZqf5i5~CFIKgRuOl1^}L;C4EGMA4nxTG|tg>jRq$Q!|nt7|nhrx-3&?>e}1uT1(nfM$8X6%!3J^g?a>0AG+A%7TuPc z{H^buBg$5-I6~zKuL?S=Icbo&7TENOvhJ|Jbg2DQ%q@UXHMp-cQ(j*YYXL3KJ-F*s zd0k%NSSAAu@87K$$80Au#}2=doF*s>VJzR48_1+}x66I3$zqdoJi~H>8eo3XX3Qps z0VP}!*^fgjNg=p)sItW}u|QADSt(A6r&b$%?fLq<_tZd6!prOHUiF{b<*K;7mU)JqGS$xT-~BUIaRq((s~OA1l5b1-ni4 zTfX*>FLP^@z2VZk@rb1EPO}^xzCT|j?X;5!dBx%tr~4M(i@a|&SnRc?X+<8%{JfOt zIIE)1!n!Tr3!{rvw+n6V`TZ(m;Q3b3O-gk~B&x?)(Zz=Fp_a8CXqU^q@|TV^zB1rQ zT!JM${7EtWNP%3|&m?;;-QBLE`s+;E^>(%v{PvjYfInwne+3R7H(-IxayyhYzA`Mv4}_Wbs&Ab+)M#ks5I)0Ori| zZaxii)0)nGS*~G&KNwGSyw;!B>u6E_;KRll3+5e;tg43FJm=LpEXwQUr(6zoN&wej zt`B>sGc?TZ)K}VoUAbB>EK_Qn*A`$u{Y%llJt7Ly@8zCT6t!R8w^#Clx%>#sBfl!` zbBAZrZ)TRq4lGN$h-``mh3lmO$ z;=op@Rw0b)(3TL7gx&*dec@V>r)jwZ+8y!r11YyF5P(NdTe$L=sU@-8rtV{mD8!YH zHg0t$qxfXxgh>}E`xrP(JPsnf9GC1h8KDQz$ha@ydY9F8%f1s8*pJmBhtiUwKoj*mx8PG1}b>9z1Y;*SRN7ssJ<5&)0J$9N6a@%45X z?2=-8SSBj<;WB$U0p}9yCIQUsw z8xy1IB}nrH+ZjJv^qJANA@EWI=ocx@k|ugyR?X=sTDNaTY%=zlxdx-s#oF$N!Yt|Z z)t>32!oo2bd-U|c;pqHKm7iPQ-uTqNbI+okmIh6)$S`8jmrR~4S|S9@5d^B;2S46o zUZ@@kP}%1O&Bh65l?i|n?{O*b?b=S8ALqBcO$p%`g_P|hTI(bBbF{}ie@eY{Wf4jJ zfZeCj3&%&Mo4N zBW7I6-5h7D!}!)KDHKDh4qx$`tS|lkiTMt;mq+afzyQV;9!5`)L=-wxwR1hQyS4Bj z?jPUbvF_#`d0X0Sd@3iTYDz0m61N!>&l^_lxI z4I`UhlVj~xWO`~DVZ&M*j{_F4a*}n2g#}WJ)49;@pFA&$3trJ0D~K>I0c)ruOB7)r z&)4O=pAGwYU~t+1kjVahd-`$XUGL9c=Ob`_K5(cQ1X|u_J%i|kgE2fHp`lAFeXJcu z18cW|Leeo>c8C0&f!dUNR$LQ)XC;Ico2W*>?{5uQNN|8&LdBRXqPolk{Rp{rMp8Nx z3=0Q^8N93oA&w$0gSj2T&!}TV-+ONSwk$0Z|i$}G`DVz(~YeH@IiulEs zUeV{4mXs@5hpq$g*Q{;FQpji;Z#xR)9I`C%pKM1?-J8hV29|MA*TdpuZ`4fZVEIY;6I z+38jDbLRTxvVA0sVp$LNQ8ASCLoKkk-KIG=Y3E86N3dpgJzsm$|+UMXWFP)hB@|&WPM%G9{Gv3|t=e zNPs@7USaa3J3gDKwufd%tC%s#GU*NUP0`Uj%L`Cx9acr<|yhGA7mHiQ}q~4 z=Ix8hgU_D0Z*X$!3sbkS4;Sn^Nr*6CFDsbLa<<%>AH&*#y@8r}S1@>gZ3 z;v&KUAj{z6PMzMZjk^4Lyg1%4?QGucTJ7o%Be-A@VF%Cf9nFOb<1++iPO*pI*g^8O&n>|LZ3 zn`LF~-D}AN`h!NT>=TT$Wo}H~V_tGC03xTa%KIVn&imJ^RU(=!sp;DTSK%Py*HHl7 zuY3O&d0*D+uCrzLRq~ELi^=iIgjyEB${y z$Hq{#_gc00UhDC?de{Kvp6z7OG}*edAZ*>BmOg-D{Ws{N&v>_;sRO(yF21q(*qTQY6(XXcKlE2gNK2a@fH}M>w0_%6m z(N(hxM20WT9bUm$0kFE?u zvI7QW0cv7Tvx9(8GAVC3&D-p|o9(c{j0w~g#x{73pT}s=fVIK*ga7uaE5G~D2tp78 z9*wv6kKy-w%gLW$z56XJRX+0j_w)IQ+bk&gH4zCmQr(CFf+Baj?#BH#^H42f#j-O*2C(=d)J~p3ju0 z9?|zFt1b>BdS=hq6JiYxA+A#x*jG#DxxjmwboNb5s6}W#$UxY@w>IRKCZZWB2SXao zXTSBrhU?zA0JQA?xC-9qGSS2UL|ypv)vuj2(k$DwSZbAyVDPeE$OJz;o^$XpJfjH_ zsLzFsWCHGt2+%Bj+&x^L&@s8)cV`TH=?9<_&8jngu)!Rz z@$)?nPpR>+&t-!~LeJL$;=-r2#?}2qYuwj#+$~RArx5>wp)~?+k1CMg3=wRvM+IjU z1-;Ck!v!oA9nR}t$#`aQoJUWsLP2(x$U4o|k^&k9U9k))?uJKwgsKYW6VdHB*Z z0c&LFUJ06dU_&=EyMsMzC5{7iOm_2O^{n{*?UW*JuhjQY8`-y~HjE(B&=1sCd*o@; z08M0Um8}V`jmvBvAS_5?{>Gt>SZ4pI4ECjFdO_|81Q8NEe_wRi2jWC(JIB$L;8s(1 zpO$nyg5{>kf&lwjci4AI2uBac!IO3eYd^^K1?!&20&+s4spayJ2l}3=<;@hKmxZw1 zEyvAiT-Q>oAE2H~-h>RFGa&z5VXrWiuIswo&wIGZ10RoP4@?lME@xklvSr>I3u4AD z!0!or&>l~d%YH0zKlT26@Eq%d`d-TtT3*0>vm37e7XZ~u`f67b#_U^(@jC)3?& z+JzH5mcPB5n#gxTheqwU2=EtCaTH0)-Va>mkkr~^D0E-k9`hZOD@I4~;>n{Z2x{Lh z&*q=qAxNRNHz+{|?h7Sp{FC9hQmgiZ= zaIe?DzJ&j^*9eAtS{e!cSF%$=W)71p)Nh79E~tIIUN=Dixm<)s5Cr=$jQPSAauO-8 z{sU`6Mgy<|4U3=c(~2f#G@X{UKzjeZz;ki&XW+MvUiB5#EI~IuwKvN!UJm$Vnx5;C z0xRev!Lt;_#a@69OV!h=0YsI{Hpuc4;D}bGg2~ZN@CE^yf*+|%RdFw%C}M3b$7nPpJSNY zwq-FyRG@Jf{?vn3AW~2ys8zAi<(`l-!K$sW0Y+9B;<}<3AT=beLftef$DjP`S$S=p z)qo*MrAs_Xo9aD7h zg|(DW!>PxR8?2QAJb!bc!4I5S0})7*ZxS>LcZT4m&+3FvAGZVhl`7NjM`Po=`ha`u zt2ooao$r)zPN=Ur%(C3+3$@kn@7r&3FYVc{uV5)pWpG~m)GT4f4yEeh0sCAo>MJ83 z1W0GLowy)xVWGHj`EG*0Mdjy{LBji+kHRYLf0tf9jVieN2=3ph2Yr_pwj;hN#EY95 zUb~+N`}JG;o7Pa*bYq}*#W;1~^D0;?x~MYT@=p99EQqhN_5wSg!_N(x3kfi{3_k6_6ypNA-Tnw0`fxO&#`9%7RFM`d{a6G^sJM3#lb<5(p(()r*Tk~6O4%hUD zy^Z;Uamd56?RgkTsBc%eu>K98BSk?En#Bg2VL5#q;B07B5_^umzOOq9MYcZbJD%{H z4)C(QL!4qj4{6iVn&je5nFnX>He_|j)1!gEG8XYmN(C&gNd=^bPQGOF0Z0cANvJ1~ z38fNU)dl7+&bDnq=#-&@8n1$3i-N-2KK|v@z{S?haa)%(0bca4EQz(>#u*aZVbEB@ zUtwrzXdYjG$5AimjF=?p%7^C>lOC|AcMSkw0XS;%x(`954?ufqI8x=9ZaA6oRW|G*fGAXu8BF&e|Uf7hXp zz_5Q{|E4+LfBV9JXUzBC{f5`(+dE&D|MH*yGxiUfqY0L$a2lr>g2VomPoVjoKf_Xk zUt~#!`Ug$W7?063Pn>pw|HsYvYG{h5Eq`Wwf_*Ww>i?n{Uw8XgW_(+~yNk}t2uh(!S80W$FZ;KZSs9vJ?0xWp51H+z#I7UQLGrhLMbRydNw0S9(Bk*zOVq_V`9b%3W+h6 zBHQtdnRZG;qocxRX&P++Dns0MG$nYjMbRICA9Q%obzK#nFJg?k%XN^)XOqa(hmX+h z=pKTJllT}30EY6Hl;0T3a|Jwb-PsUeiHZ9Gf;KPu<#0Nj;UB64RslryW^-yj)@@l! zhEcv;wKFbonw6zqP&fsU4Q0=w3O2CS2A~f#G_apuDirxC8G_FV`m_ zmSVQPw+9s3yhQZ-7@e_~KjgKPhPQi}h}jvT7_aclNRSAh9|2AdA}b}`O88v**y+J@ zF%AQ8?j-^sA}c#UW8cusVY>;qi37NpePeIj?-=KI%q5AssfKg8PzXF@7o62o6PA_VJ=LxXvb!njezW7$A+A6%<=u z2S#gxsS8e5Zzw%12+)R7MbS6>jJby5SJ0BtYEO>4ZtD3L2{RC7P~NuJ=KlRm&g+>l4IxbYes(u7^uoE^t80{)LgRk%Kq z$ZF1ERN@NOQ8!?dciGnuSd2KC-F-vxcMcIH;xj)^l6MZ>TizTpTrCe!K%*`n|Fj<& z7tUAYFk>1Y0r8cysm2V$w_Ao@8JyR>{H&j+x!%pWSEnoQC@O;2j!Yp#fR?{!75^ca zQT6l-;qq9;U%SJ|=TKfPKn2@VY}lT-@W>i}coiU8+^UZIsLgO5wwuO*edH(eooj@z zk}f_Arl`^^^@3u87C8mG^Jf@0H zP>YxeyXXwF3#2#Bqp@KF2-Q~77{ZenC_us8{fok?vf-Jv*Nb&hUSLE{v}%fS_j7Zn z#`_xzOf{#T!fkn9q~xhp)qyobWw=`{mDVqdtpn^CWM5K!<7{EbBI(yQ!ea;~zXv;{ z3Jeh*oQJYJJM z`kDs5^7pknZU1dm2)1bzG9L^uB-S!(EDe@c>3_vqkvy^nGlmX;TmDWX9!11nab&PB z@>jiKjyCgxVy03WT?|xi?}yM4fDa%=Za|19=!X$<5~UA!4%%Qf?hi`Vn*hU z2UhpcYL(g^x6<(>$Adpm-fFYhT!#g;n?B*sZw0j!h&pL6rU(Eu(-9M`7>jn-{Wa)X z;%NzQC`J~l@pL&$JQvlFR%}Q=UqS|y?f0BKO_>fT<_?9Pl287P16U3Lv0xyssgxN4 z9KbVLB7MyQ3_O5OM<9U!QVPg%=T;pp#C%APPiUri{w=tnviFTA@+sQ83JnKtvaoG1 zwo@Zltj$J6O}@&lU`WFxRXoQjTDR9W@n94R8Ze!Qeu&=E0*F1w9d4Ks8e5zLw}1%Y zuL$3-ymiY6Xk21Iy?^#7_nX^RW_u;s4{-`tLOsDGLvlL@G0O$dwIZ(Y02>w`Vm`ia znIhiCe)-wUe*4|!Pa5J&hQ8_VGs<7z4xXpak0KA!+eBWgU7(nI(>L*40f-_V<&5(f zRitpfT!#Y4=fKLJ;2CTVJ{x4AKwDjqEDyo}ev8;FziH9z4uZqkGT%H1H-21dZk@5Q zUecU&p}A2C2wOf%KO*cO{nZ~^EV17kCEVl)@1DWL^Eu!3L#jg+=f)eIj~C$g9X3bD z*X|JAg-lxyigX83x(8-W?kRg?%r8%dYFORcOCVmln^=Oc@hjOoW6>OvQcwhI(e1bY zg#$u)H9P@JSx`9kLt6U_tcgP$m|FuL%8k0GeKkrT4a+9hWWs8{`<@n!i#h98lwKc8 z?HD;Ui{bkpO<5O^cwlq(G}2ceLv?5}m}ui9^{@8>ib`}oPJ*2FIV9J!T$7XligQ=Vg+nLq&$65p% zve9dq_xQmfx%+(G-;wxDO9q5^D!PVq2?z7tEFIY|b3ct$KChy{Lw5E8PkMm8MwO5-cIb&7oO2bDeT1Se67yDNsg@p`Nt8L`3+_N=jgTp@tuKYi-LCW07>7eZnA zZTa0~-K2Hlrpd}FP`o2lZ`Q2|mL*;$#7%~$L#MIE^JTyB2b(4!bA~8)z5W(E{xKHI ze~91XER2KH&_m-WCxVc2w%}|s*ip8QSC1&u16=fT^1~);{2|*|Euh#bG{@Ks0AmIy zTDVHT3_JgB=fVA>2q8{5C@9Wg0(vGya*3+!Afil=J?F65~!*_)~QX5 zde+`slFh4TRx>#^OU}N%uVt`2<{$I*;jrh%6}kvbv$DIPcsJ87&{G2g3X^ZJxPVLS zt0$G61JAOd^3pc0#KPX?evM{D2k{fDx?sAJ0T0dL+N^oQ#};&N6bhrBapI=iy?_6u zA85`-(JCC2#NeS8U1;ogp6UtgW#sz#@TdmGag0lGO`?5?H9v*sa}+%E2A9uWrCVhJJ(1@^#&F3A)M&m$Pg+}|UWk|ep@Tkml{2l9;hMe_*#I{y~C zYw}Hl0tY5Mm`xBLT1FnX#k^_$eS@Zzr#Z{5mUKSA;9?G4sD9JDP>WWC}~aDCXh_n;~zw?Qb01Isxq=#zbkpq&w(F}C7 zy3|?^lj-zgY24l8?t=*j|YwV5`teEOLhWy$Qd4Zu@H$xzUTQGR?nzH z3O|^n!o#5FptENk7gt1^vzB+s`B~=x2Ts7=Xd2sN_%l0wU$v z%x&$#_Y=oti5$ZR+v+T0!6I&{Z2gALI74f0n<&tP=`?>3pmI`pozj#wNNdaBkW2jNgK*-BEcQ#c=nQC~a&m^pToJACD7SeCRU!9BZ^`S-XN zHLP?|xhDPz9Ge#r3R6z=JJs-y;7Cz7KDJB;xY|YfaY`xXnHN$?0D?*LId*{c5(p=|QH32rH9 z13$?slRf$TJB87;Zccrnvlw`q?v--c69}?BRk>nCSlnCUbi({QAA%g3U0SfpAeP)r zfhf_sCiYM2Y5oz3ORD~c{BOworu$>S$LGM;Ik4{zTrN8jV8@Dd76;<@kDl5Na9e2l zBnZw;Fl{7e@5Q@<#*mW=_97*eSFnGW@xx;wY~J`6E((pi#GzBl3Sh0Td>Z3)c8RDi zl{aIZXg6Qmmn*!F0tZcjxN8b76$YIF7Hft$U|tBb@b9KO#%TLoedP~_d`K3*plZhw zdPCvQm(lV70Z1EIImqLd>mQavPjM1>kKsDm_|In@Fc2$CNjN#hyr8hrr?62N_Aw0O zcKEj+4a7)jBNkjh`j#%trVFrm-4x)fZk7{Hl2lR_9Nx3=ct#N(MKh4AhuG*09N_dM zoX8LP(jTW8p(eZQ2G`y8X%bBXaOxw+_<`g=VsBoBEp(r{9c=l-! zWX3jHLG#bbrJF57ZI+jsg9muh>kHR^!$Gh!zCg#r(7( ziXOX@?mI^?{^FIw?fY6oOtDD=S|uowJAU}QZuH|-sOwVH&-^reg8~fDYz*E`8H$`y zPw?3D{Rl_*_G(gs`P@hUKxfmpU7|aMk08dXsJkq1zuV82RTbw4{xV3%TKGHn(z~P- zj?XI?Lb-(E&{H1+>_$)BQ;Um~nuYP^8YP~6-_GTg&w*K(G<;;nFCo+!nr&VpsHq`~ zHHb;%TI};mq3v3KF3i2W*4z(`y12jdTnzGnc^Osj;=Eh@X^Fxs3g!Iack5%=lPv*( ztQMfSJODQUG(%HRjR5@e%cbmK!w_+52$_GP-TitU!)_2f=}{SInond<`wIU0*$)5s1B(%Zscwo-zP zy~d-sQn}w&aAtIB`2zVRw=`_b+lR%UAjeT#Vec%@a-b0A#Emq@XYXy1KRz9K;PY9@!JF*bXl*2CcB)>)F_JyDWDQ z|IvV$f6Ic)Po5Ppc)rKwnywR9Ks1TdOa=-;h53lTY1#w3;f8Ba;8za= z`&&D97JP`kQKa>Iac~ItJXi5IKzU@A?+N^L%c)VmSasM72Vh?jHh%(?4t)N2{9@280@@A;(2Io4j&Bg8Z#pbSodZT{sQkx z>VGD!?>%AN?I&Do`Wq;98=V6pRDmK$RHDrI1K-sGoC!z%40`}y-=T;Ms^>X^io~m@ z4~1z)KZFHrXgt+G!{Qz6;Eq*)G)8mK+~hQ5XjAgz0u}<$30%wY96ZUG?CzgBXlo1xb=+)_ zQ0tLf9UF;rvk#;Ud5i+!Pz{qUVnb0gbp_cFd+7rR4(kWp;&NExD{^xGQ&eDq0r7q1 zSvT)sV*zp}ND`HyIBWw|cgiiZud^=CfbWRNwWemV+O{I(QJ6fK_G^uB)CJ8@}} z9R@-S?x^x|twDFQv%`@4juGxfo6pHgck6# z)M_ie1$tornfQWb*e1LG*eJz=&Cx#GD^ZPq=O`K+Y(D~^!1a8zYds{#5HO?~uc{QoVP3N6zs_^=L zS?|--O$Q4bH^=bopgEA>b|r4Gb_cl;v{3e@9qeE5mK*|Z1h@tA#zA+qNgsH7eYXnZ z^<#g90yzc<0~!=GJ&hKal6gln6L$okS1=aS=K&cAa$cgrx=E4ec$JRPslra+V{#+d_0v|dl({?eKBmT^DY%SKcd8*gntugBB zr=y8i^J~Zoek5<|3t;{;!Uq?Al+$hYQcN|8slq6!t>S1Z!Ux@X#DGKo%Hwj@U#H$OBlmYw;xx*&P`1&Da_xAP|@22_oY1TWX5uJSl z6W)!AsRt8w-j#ob9I&BIac};>r*|k)3!Ai_A8VdjJi25r(X~Rqf81TD^gTnjFKF@% zh$>m=&HhsdOak{4z>x>XWDdCZ=u2ATAClQ5g3Q;Yzj}e;XwUwBx%D6F1{=pQ-j4AM z~ZxgmA z%kGS~`v-vsEQ!f}*>%dG;hs8?X^H#b@6K8>#13B&wk*KOSpdNquwE7JtbfJ_uhp*e zJq1x6z*H=#J*LRT`o^o2;A{YUe$d=<>g1Zfv)`tHb#^&fpa%QzrL(UIhq3J#bL7&P z4_N@@;Vu3$&eC-fQ(#_1Z4gI@@dKje0$wdk0r38>^m_5E2+u1L?E~-|haheJd8Z6w zS6ilAPVZp0+;I=Vy-u@#>kL5RG$`1&fa@{4>JdHS9I71b3=05RdRixK-^}^94)?#b$`n+Cx84W_(xPf@bt5IfCFS3NE}09`>xAY z2OR+TzdeKS@yX*wH&F=GA?Lr*@!5YyB^0AChyi!UmEl?H`1$;_y@Y+x$p#5wCv|>!TlL2M!v05Q zd%<%AjD2xw9FvIU19aQPBl?WG(Npk0@z@aDs|JUa zhJ=Q=ATgG~*Tdjw`B|fd<_XmR@CpH&(jl`L!l{=*9&3GQ4(_fb!~1G?RDME$bclf` zJ|<(V2u^m_1pzjzI{*v4L-e^qBk}UDrZ_!Pb-#H-+%901-!XU&9okE8e28vFT2q7N z0|33oT?y4Nns|^e>r^5?_$z(%{MvAUqJsMe z}B3Ilx43O(c}WV(;W#gN9y zZ~Tl!)XQ@$)*sa>ZC_#g$?77;@*D@!F?nTs&Df;aHJ}@xYT#`U~LngA5c6Xk6Z(8 zvH>{S?y*|G&^W-HLJL^-Y}i+ezA3i!D<6V(y!%+X(;3o$7p)@R z5rDx_4Hkc75(f*$2~^R2?wk&`cX#ok`3kWhjIF_aM;@~ApmmcbS~pOi$7g;)!GY{F zQve(JaY+*%ehxhUpckZQ{zTx#a(hxBm*!FRNh5rLq6O2%gW}qn`uWB)L#-vH38z^u z+>`q$v>n@4)HuSQq2D>rfp?!Z$%v;F#99V9{)sbO*o5BqjXMgkUd_h7frQnw2T%jx z;{ldTK5)(#7cFm@C0N^8T#$<~EyCAS8hW|b!Nn|Z1D-{mXS0f{yUe$T0+mJ$HiMRmAIi#@N3C~slel`K}toj$yiK`@270){h zl2$L@W4kv%2+{Q7OB(9!(!%L>Y=h_CjY#p@ARQZ0OQN%GP(b|rT@&GX*nH3em?YK% zF+SsNe<83A0ekjD_(ij8nzG2Lw>=ds;s-kbQt#tY{Ep+`DVWD~5ql|MPgE*+l(hoe zr2cJ=G_48RIRNheghkWkvrfVTD5=RN$ZH|7zEg~cu5+ydGv1_r$?H54Y%*Mw;~7L? zvl*o!)P-D25oW!>b5h_s&+fIGe1e82U!XieO1GpXsxj1uq z|7fNo+t)LyhUF?9PCw3O@Js5bs(T`;i~T*Q+%+~!()w5|;1{av!BO&ynJ%mb;_LFP zVeYMp$?w*bjF9{M45p^5!cFWeZ}jHY6b0oXrR{+eCNvRK`W)4>H}<`Jsjhv} zDf6tRk`fcxgoQQX(Lz!(R>=@{j{Ffy_FFq|oj z55h>qthbejYhC5Oyy3ZcR_@E8JyrA4s@$jCR|~8=MtJP5;83gi-?FzEe?<*Y?-@Hy+}LeUWXx8gvZR z7%wUxmUa7Z*3)Ct6=B5yUq$lxC3|~8$%OM%;R^(Q%d71 zZEHy1=V|Ku(|BW)g0`#$qo2V3qrX>JriBBu$?}b^fXo8Exk0i2q`G@F5_dF=#oPBZ zb(q!u2X<=Q(tzqfJeS|GkL-c(G}M+KgxTC>Eh&vcjorTZ+`qCH`3xnobp^NeAJF&_ zym730B!XJ`@B@$JU%0nmihyZ?eL2h$q;(5Cg&gxQ>_iq+zJP_fVkVUrl`r6PN z+iC1IUFzFuitwzH|C!e;Wz-;ks-}n4W5%}wOSXfq-W$*I-)+nuBt#oclHIkNY#a`p z_f16rc<4h6pcAggxKB&`0aPEKV%w)*x1b0<$1Tfx4r+2tc6Y2bMR@6~m7F~zNd=b{ zAf|sf^5oM9NbT4YV#|4!WJ4M(#}PdpX_|Q2Zj*yzZSeqm!{8)G7BHn0^Cu=n*4%Ar zm&@(1lSZw%c;+%IK5{!iLeY!2|_=baJ<_+qK96bWFKnhXafpH-XSZ%Ghv#nYeJ#&ciYFr9%r#V~ohO~KN;3JNGx`3`;u zk<;XoAUN^PuI61F>R}vJitrW85*$b>#ZAJVmLkCc`REFjW}Zu=YQvn|Wq~%m_Txnt zFX%QviH6MO56wB?k00C{tu9HIGGvBWEA0MwnzuT3w5cpr)DG<$<+kczsLKIQtYoB# ze7}2r1^L#OU{MJT#|_JIf3&QMHl=_G^^{0(9$*@>f&UGaG=%a@q6!BdH425^rUL_c za+yB^cWJ%EOZTqEvK-HHk$moc>+#ayJI@{kH!0oX*G!8Ft;9LGNSYjo?ZuL~`$O`C z_m47O5^JHN^lNd&6=!Dxp9~KG^af`J9dNz2vVE?AqW&n6h*sS5{!e;2^lSP&-8ewEM0ecj8f9`y1x?MfSNjw944dli~*1ePQkmUP$ID@%I5 z`TX3Fp)E(1U{hg5xe1}g+uH${p~SGKV=fVsgL?&XuIx2I6tw>=lEK+Nw!@wuiWw%U zSqL_>U&fx8}1AG&#H9_0A(6ZybFSWlp|ID!6n@3vpHS4k7?KXdb<; zIt>$q1HvnSP&S;^dG1aSgvx=Tj)RPWpN`wF44itficLfsYUK(?LA72Suv9=%P9T4n z;tEfxJrV)%Xo2}ewVU}MClFMMkDjz}{wLLW!kPi6r4_EaNsY`+bXn6N1L-4{yO(~TvzQb>D-f{3O;kf_k%_>x3a|vKU{~s z%7G^p^5w#TWw0qA|4T@bNr&mNhJUy8#%P8|QCZ##Dp1=(TlQb?+O2~8jbr`bZX$Dk z#aC4Zd67&9QQSD`6Afba$gL&zJo9()5MOCP;<&6wG!pZONpZu!JfU%2^v1s?R|Q_xXV&gu_GtCUA2VSocT?irJma@7=0Ozzqnzh${Ch?#vgc4S`I+clpM za2aO%1u0C_NzRu^X((O7Nvv@o`gtAZx)b0!x}S@;?l*qf_m`YZ7lFOW55E?P zX#egrxJgZlumUa_#fzaH)Uz1{zK6x8c}8rbR!kNoO>B&mJ8@tU#Zkh*ywp~Z=6)v>Br5=-4`(A_f8_B-@OOOY zW8p|MhxddnzN0;HxAEdxSC|JU`M2IvrZN--!vx7BaNxP;@5hM%-|qu6P<#xwqE>rc znGhBGdQe$Kc%&l3xj1*MZ^1R#3YI=_K5ysiSF9NDI1YumAG?l!uf+HZ^10!k8|tQK}E@0X6a;<5w)55_B&%lL|Wv0hEOf!HC+J2&|N z*hGmjt;PKtL9Hd{coikuhihllm$0HXLnps|4Dc!jAXnQD94fX6=h<|$ zOIqCahduC&(yFe=5ia5uYH#Xr&Qt>s?P82LLrF#Po}(m;;(6T0*`ISFFZtN&t5=pC z5r6QBSDnb4>V?DU;xFBjA;4mKdPA*#No6eib^_a+i=vKOTu;1LN zGU+@JS(dtvW+p>{vBG9(LewtaVTfXj-zzh;^e+nlkQcEC`6vLNe^y&>;9mDwHUV8* zeYrXcK2hqZn!B-^#S|VQPZ_@PQv6IBv*rWhcoy6NfEc_+c?{m*D~_gAQu7WO=*yi& z)?=T>i1DwwW933A5Z5%2-^I5U0fr5g5Zo0h^o)<}l($e|S^IrZ+<3`@Z-#>9liBIr z?L1xO?t00%Lp?QG(x_6{%MRd`oX&g_Rjo1aPS2C~9PCwIXrY)i>H90>V=u1n2ZZ@y z$+bq94)X;gw&xfg#To(xy5C5>o^I&+%)v820m^+i7m|I<2DFI{9R=QxacOLP(Y4BU z!-lkzP+F^T(5)Y_Av;VB{UmE z7~RC$&YM?~uu9?ndy3KTdC}qh^8V?5)k>sTN_tf!;kt#L06=cRUt|0Uup^eL3gAA5wkf~B8>BJ?Mm^G?=yz|R@IG@Nnpdx0rJQ!FY&UiT%3BeA}) zEf>j3d*(YPxNi;;iZC(7QMj?!57m_4fkBM{Kx4sI)^$4m4NqNIqI+GnVA)B!`8GC1 zbm9uN-#=(D?NILzu11km%+-eU7k+*{N(8czjaqU2{~3|P|SA~_AL(0Mew#o6^TcQS*pK0 zjp=$|UinY}***>6AgCqvjZ*Yr0%4!pu}Br2684WZ19R6V{)z`{od&mzi~v%?+f9cI=-P#p?FE%3y<*vU75@R8}_s#TrF@wwOF2$jZ%j6o2(Km@&0jTjCxO8Oq->uSIt5*)E z8tCt>HMQrPS@if)g@BJ$HN&88%gSQOnhtmQM69^ka#y5#9$NU<3Pi{)|)-rN|`&h#O1fJ`_15|uwAcJDn%Yj z6IzTu7CQJ5#DQORE%;XjxO8~IcI~NnZy&EKKw%qAY=crp9rsuI;>Wls z@=9d@+o)cm8$Bs2AbKG#$fl{D& zOn62;O$@|NJNm)bviOj4r1ous#v~vWucmH?2QBNE2zb1tD#Vl0q8s@4N^x%?K)o0B zSO_<<`6vAfl6eCpM7sGur*Mq7{=hj}W)Rj4;wu{UcPse(-tABs>V2PlEMU+n>w<~{ zp#)0<8x#f|6S;dJDd?==}=Cz)#Q!BkzyxVY!YL17Xy5C3lP`?sx(ZQ>=3oa>fwaiMu zU^KqGwZ`PUi$GZDC>}R1t27MHe!IZvW?R3}fv-~aDU8-~yk)w*z7rTXCO}}E<7>5r z`(p#LCvlvXb06Ts=(3Eh^hTF?I<9OntGZCe-enTU3Vu|QegOI6l7N=8)qlEEIA#+wVEH|JYoj0Q5U-^{6&7*BIP6zY;)Ytz+}SgI@i zT5pva%3w)Kg19Bd4OfN0%ddb?i{zTZm=Mg~{LZGUMjoww6@-h5f>18GRv@>#E^V93 zyLTcMHi=o+mqTj+7glW|eM@gJkGpQyy!tkeNW>~*Rd-QMJ|PIbWyMnOFmn6J#Xq={ zk=mhFXm;a@?CznZoxjs!D6U+ZkSjI`^;X{j9jn9Eoxe$9h{IwAA$US8dz*sFGWbOO zf@Lola~e4Bq+hqE;i*4Po#skj|0{WK*7G{EW{2JPulR6o`hxc1Op!($fTKu}lsJo2 z5*%KWnN(Kt*w`$iZwjXS~aY7M}ARXe7arz?dHS*X4Z`OQzo3Cd8GvvN=7Ix;PBl>3IJ6ER}j(ner;) zOCj!qGMyUUJrOm#*EC)8pD>!5$N5#Dwn^OZZ|xiU`)*79)hSwMynFkI6+TD5Y~rI~ zc6;kD7hDl6aG(2=NJJ7;Q~R1G;<=48K&U0Nn|g7{le&X3(YFE4pCUOY)#a(=y*{<| zeXLVu-6--hH&qfO+PMlL58Bs?2qgD!q76-j-8FjCn*?|A!M!;0cbuHbq-pwX9%Zl!V4F08bEw9PfY1SpT1c|n^$W@&`lMmF5{Qds2<^%&C!AP*pL zSFXGnI|ne_OMz?gb3Xs}oxt2>rN!==`*9BBWg**UYW{ffkchERhI`;TQ1 zalQDCR_T(nungR1cf0`oN2KLq|Mr+cbJZH|MT1A`^ghmxXbuN*BQImg=AO=ig6uID z%}{_&mDi^sb|kCYbPym;Fm`){KV@{WFx=j$_2HKf{GspMI>Yf?U#BSDuuzwLT(Jdv zf?JOlD2xl*h~tVRc$3M3{RH1HP>j{ydX8rn`b$hVSg&WClN$R{%e-0|Fbz=zZNcVnf81nN6wgb-o+_H>#b=%3(EJs#{cNPurX8k47}> zZhHgt<#EJR)3R;Zu^Yf=Q^6z4ONp;xjNWvi`hIjBbl6LNKH8fc_>blXWVQWe?g;>j zs0#OW`B0nTm9G?1^c5=YsS$Lr8sJhFb{?+&u`byTs(U;FIUH!qBT9w?!-=PjmmVMR z-kpy$2|dB*pTlH+!b2C;oqY${V72`xa$3WWkNB}7$R+M*xM>#zJ~i%GQ~Sp`;k80<)c)5N3Ygfb%*61 zVpL5)fGeK!If`^zOroF!f*UoDx}^Kgcc!Wa9Y%11oG0s@(h-K8sD*e2jyzfTO9eRY zG@aft)BX^eigT4Vppvo<1i5YBX|Q}a@+rzq7qE$=xCf#Wb%Z;C>~T*gZtC1x|LiYO zajSWn-s+Fr1mJsH$qKOiad|sXIj_FI`F>STpr@3(y`*>@|@BoQ&(x=}uDG^y{ z-MCuJT;F(OTVVVPpK>h@T6@S#*EflLSF8`i-(JcSpj)P5+FB1DE;PPQ;PXV=7UF|e zx_7ZHyeVZa@?>|<4^~*PUtH|g>-H;@91e4H_Fmd${~A=^AE1WNVy*?uk)OkMt8YoX z;K?gsyirHGeH!NvfMy`K1=D+PLx~9Rmj$vyY*qjYD`=JFeh{*!nDC zzo`fBR^Q_o;MI5+`7g7AKqyhyPXNGjdYY6%v#k55h8mM!relQGHrDC`ePfT>cgi4g zC*jv-H4E$P9m2De`u0l2Lz+PQNEjB6kDF2x{0-wG)s^e7C%-WFMkZ`&rvQ`7@rc|6 z5FE;HXMvdJ==KQ9^nw`Lufwl5wL5#ggIF+y{Rii}iFm1??j;Y-!WpE2RW`dHui)9b zEZ@NKsIQ&^+Vh+zfLU+yGxGHMRwUfC?jLO08_+iZT@B|mA2b;N@iQu)D;Rdw_HbUK zG>CZ_?*>%|D#nEe{#r*5?r)e>FrJ>PG%B78K}!{W7YkrF&7BiC-bNSTRN92xx$(vw zN&DMgwtl_Cvq;PXy^`FD5WsBz@Jx~z&&K% zeb!e0(^vybttQOc7m=zU=Ne6xf6gaen6sf2kR!RElaKr}4~C4%0^{irFH7i4^owYd zRK~aS-r@1_G7+z(b>X1c7{>D}@yWK3_$F{~5R@sL1d#vYAaJ5GejXLE1Uk8$ng0an z0eAg-R>SWR2qse`@Ey0hyFDGethCqw^**Qp;Em3C6kB!t&b;Cio`7gE)zE+dF`)d1 z@~O=KZlBY0>8haF!te2En5y$!D|`i%i(%BlIkfINJOTRZbhs^dwIqx);h{PMiUAr6 z@d+hn0pk(wnLdwnHKZtubg87T@q3zLYPA~uyP0w8+vOeqBs|Y2;rbrNm$c2LBfo-s zbI|hT+)y7A0%}10HeA;vp5Mm1V2t_4^TONcRvu}LgQVh;Q0(gL zjv~>{HUCVkV5YWkzoc7@uNG)WD|)*P-2cm34i;R^Fvu?A{QPdUzHK*I6p>kMU9;9l zGz|BY$@bR?X#2KR+T{u!yn!2uZj;l-$`k*Ytpr#mIRKtb3h@a(+V{7YbtC8JwxN{H zD>W7i>VyM4mJm=?GwUfPFN)a^$~`}zK}y}sy6W7%IDe3zV-a4R&VS20Egc_MO71;R z(ufW94CzQSER7ff|1eGlGBIXT*?PuDA%@IA@s z;)0?i9>~IV{Arw;_Jewd&{zRL#H)Y4eb6P8om=lW?X5OX&g~F!qk3FA7i|IwUxHBgd(Ii<@3FNfPHN-`r}w zP3eIU?C9>#G|EO#Y6ks4GtuNCS2PIqCYg19d z;~46xhI1CSlqaoQ>BJ`3(ypSmW`c}XAh9{gJ8Nk|wbS0WCyBSJJfNBF&6R>3m01IX z{JqqJMcoc&90tJJuHCzd?ydQTdk3z$B4@I#m`>}4UHn%03;RJRz{PU5;y?q3Sm-7( zv+j?^Ik(=jhz#WVI%>3L#l-0I(qFeI&*Kt*WfPL#`KXJSSWbUeIqRFFS!J8#Pfh zP57m6FX>j?&f$O`7>z6@&eE5glJp`edo6A4U=L8^*Je*rGEy!aB`EbTUpDgoc4Waw z*K4k(eh9cW?_t;N#a~oWbHs*e=P_}O|WP6%~3@hb>hE===1 zby6+7$<946+B4x>QPq+3?LGXR^?g!&>8`Hc#}ouYiwjgkbMlSzXXkO8b-cbmpPg?s zB)D_dh4m3I0Q7n?h4;wU;CfXVX3~hbFV(1<@XlpVNmW@m2Hg9$$ctaSu6Bk8AWaUy z89`*SF`Y&c&VE-3CMLw(b!)Vb_dT#O&ozP}qK{uLq`o~1Rfnv5CsVOH1I zBd}(;19omeaPjTqU=N?CkAqP28e6QeAmi;l1x6PZ#LYFbh^#%JtGW1hK zJH0!TTj4mXLX9uD9r@N${x3$fhw0Rx8|JYyYl*p*9v?b3zrwd{^zlx!J1_BsWT|Ta zP}Kz6t#eg&J_-iMOOaOxv#_s;>5}i;IU5p>Hm!_=Y{kD>{FZ3}oLjLS^pm`P{5h1) z6H5S0SpWvRzVSFS`6s@0V{$RN-yAvMC!p^t`S*EqEq28?TF<#8qV?)_fblMB3x}TF z*4*DXgKBBINztFY&fk)86EmF2950;I>U;cHT=IIYA zLb{oXDt>dlfflr~J)B1)PyoP-VCHO5sZewGV?K5O$EI4F@c9W_1y!+OXz6MG=-SZ%sWt(!}I&pttX zDW`f)ZtnHOJJ5hVo7O{m`J$c;mTJ8QHn%ddaVXARa0Dp;#lrQsgcKHt zaqOp@&7L$*OIZcV#vt?Of+t@_-z3jthfk7CgyPMBgHD#aqWpWTBT-#0`6cMM zEg@WaF>L(x#Db>`_vQW0#fz^n;0j&(+)b?F3^vcK^8-+Ag|e-c7g`KB{ACv}dv)QN)bI)(Tqz^C&a3xZ*tyZ=iZ%m?x$OF7 z#`L;rZ4loMOR6eLEL2)_2Jgr;RLncsiQW7TZHgM6Qyna-A59sTn`)B}5JaJ14u5m| zC)D1?!v*Tv*?*ofD~fFxjSfK?fkRQ@`ZH#MMpil!X^z0Ve0N@ixdL&^>V1Vyz_UFq zkL+;Iz6!Hg6#Fe%hBA|#Yfz=h?SvYPi27mnr3i7B{~0+V5fsz))cUomi!Yq5z3TiUyv+$Ci(gPh869$sU-LpML1( z%RR{ta;**5a0GEG2>A=wPN9wiJQu7aAzI6NrQQsNkglb(qkO>~fh)Ro<=_2J^xd#l z#ENGIx(^ZS1@bjjt({f9i*?z?!n=;3O9uPX6f;jlk>A?;^@iu&BJm~$G8SDuEJ3f| zKwqUj3e`q)jk_tdG6?@&dvz8I?}bgz}Lse z>Fnhh3?hiH0}Xn@kS8&w>0S>r34 zwgb=ACvpVkl6kv(@_=L*>P* z)CJFn_8xM83H`xkL5$=$>N*>yRP;>N*M=8tYR6x)kx#^Zk6*?w52c_^0qp*KsOz9# zpJT>iaX(TO*1gWP(7uvbZ}ak|eeR%Yy&BM4$L&$;?T_yZVCwGNw2o`q|n<+QKgSE;ibYw?vV)Pri)cC8QA zt+p(vr3Nw3XS(p%6aZKYa-i?Oc`n`r>b#eQutRN`8F0&CEynu5`K(ZrghiPO{W8-< zM@8WpkZ{inW6aCNAw}X6`Re<1+0#Nd(;Sk~_|CBHU^YW{t6sLJpJa(5ZAn`s!LS?k zb#&+A3_DNuhnJ^eI>=X+yye3>HOJ{!Po)t6pb=kU=uLw9J*o`w?m+2UzvhE@^@qzn z-|Ue^bmbhcel2(T-NT}{S9*`$zRyXd;LI+L!Q+H}5a%Qd_RM(yCY)ANM$CN- zemMI&0s`D%7kg{z(%_K&VX4%d$$@$Yy~u+)9Fdi+bW+wKiFJ?z1$V!dYaUbd$`dXX zULy}n*52a*uo8b4Ptb4Ou*Y*sK70-DV!m4mP`?$CgdZYq6KK+N_lB3|-X^m8Qj}eE zsE`%BiX!@xVjanM;8`2MU$0H)KJKoqK|hMry4H}Jc0D>(>fw)i0WNLZFX%sU0M zHfaB;=>b#K0X!3h5|ASlkC)EMhig;Xgz@MRiSB5p2Q}(QgkS}VewwnNat9YZk%wvR zciVrc0-qP{1~51RFgU0NZX`N~p0S-rEcz1N0cT6iYO1ckGdo)@E(-*UGm8t zo08^`*Wq##1;vrwmnF2Z6?AmPa`a^fIRo$yc{JAF=M^8^FQ6+B^R4zqESqV(vsaDQvk0;*Tzv}({$cU}?h z3xO6!SJ-e%<&io5!yP|l0%OS69_?+XG&S-abyNnzYyZ^GZO@w=fsPI8KA5UUXmqS4 zAN_zapa$CqSWgP(;@F6(1gS58j82t;&{d#*}5VaiVZVecnCvR$T zM<0dS83+MrD~DLyA2b$>q&bllN zn?Ha0^=oL9ymI8Bwxl3;_X0dOeYfBDT5sGuYy||3gx(v3J8?HeM#b0a-8e%T ziaUD`&coJ))R1;W?P)xp5~!0P^3Z)3z%3?r{80`xoIUf+NQ*vDDF@pBpv9Ht8 zPyf(ra}+7g{2BD^`WQ^8WEdB;ggSZ1qiz~ne7&Q0WkVHZh}#a3|Mg=+xgZP)6-76K z$_oItKD4HKn%dPe(m?Z2Tg)kvDd2Xnds*W`fb zK^thzg+sUo*}U<$M-J)@CjxoF>0sny0Dr+#c_era8bO5;@DW;ui;z&-1JiGccjSyJ8~*+D`a}>e*0wL_WIw((z-(85G39SE3H8bU zT?*@^c|NfO8?K&<+4O*SRb3T!moJ4x2&&JI31niYjRz}cvb4F4U*-FHw@exxO#5x||BN|aF$GQ$Q{V*!aw^$HagN(Tt2hSL;I76Mwda6a@+xOHqc1qxtUhQpWL|Vs&dpNkKM^s#o&e9+atrwi#{TN~bKC~u>#*1XcvbQl*{{;f z*grJKrdD&sLYT?J?+TzV7aSUr`nSu@F|JUH*(+r&-Prko$|PZ-amm15D6WWqUyX75=!DG#bMB=HVNZ+x#c&cVMsume4)+A=W1Y@*VWCq zRxBl`%-uub#KtnEVA$U@79*?;k@zPja!Gju!W1b^CnXW9LK;agsFTjy1Kn&`jOZbn{{jhuP-O8xE$quv@ z_azRQ#oFmUzZ;vED}$$_Rd@}A5YV(xG|Tu$nH)^sgh zxw4gQ+sT1}V5fu*J0fR!LMs{qYSCh0kTK>wre9q!4$oPsRH$5iI$d5B9X}OwR1W&FWAke-Hi@J+rU`Flzcz-{ zgeP_Nw`26hqUs&r@ontg*(y;OcaJ=M1Kf;l^pvu3mWimx@RGTmim7wXL^ytihMr2A zm!-nj)L2Rk0OHH!($TL}XA1sYFNIvTTE%LyhyCK3ptT^ON=6mk6QjE<9i_vvBLM%d zy{8BlK4?gTonySNJe9m|%vx9+cu^{}%3ovGJbdlIsfn^$%1 zY@E;IqD08lOsy3JIXW|2BYnAMd!%n0o*imJ^S!F-!@1&|OV2D}eA+#WI^~i))(%DQHz(LKVA6zUZ4?j#7$V7R5)`) ztR;PaS~x##;VBMy!)$mZ(r}J zp#hF+D>~AI59Y?o@($i_oBgaLU}S43oJ~{&+l!tJ?_OQHz?#oRd<*N-20bOx`+^wk z=JjyknfnN8p9eLZ&y1JrTXLfLIRwzdyE_`c)hrU$K z6m{)3)~!ZxxI9#G)6Ehp*q zjf(_61JrPi(zkoFp&9j`US);#Jr25LvgcHBC6p|7-mm&aT(YTqE-{e=GZu4^33ZhX z&POoLaHFeIm6_)UX7dyGDB^c~ZzDl|DUrY{6r3OLtofSX&bDETo6y(kM?-vMe`#xs zSe%);ql7*kf+mQxfW95FClf(S_pcd!*Xq2q%=oCN#G=txvDq<*9e1=OTpI^LXfrUe zzF~qTt|f;C>14|QX(mXK=`BF@HqB48KTZ1-MB(%>X)O*B^d3(p{rEeF*~(m;J0;BVV>9e2XpnTnEt33}b%bT0_e`xz*YRGayS_tvLEZP+=P zg7*tY_p}O;$b&8=KLaQne^-534H`|ZzcE?sQhw|&4plKzwM zu){Re2!+SP-ct8{O=|a*HaI?p!C-^X(Awqac)Tne=y}Ov%U;` zVPqCbm5&i^G_3@AiBo+*C+jDHo#hMqs{K3`f}7hepA(vdzbj{l%iE(^k@JAy%mZ?mq6b2vwVQ{3X`oWpf_vGWJ(Oa>rcRNkg3W(W*S`W|< z!aief2*Ewe@-DfU?-3!Zx<|5<2OznxKytnQz=KJgX$fkDl|ax{C7nQ|OWwkw&D9A<0-X4iHg zKz*a7I9Nk0hds##4C;UjEEs4%eLETyXX>(*eI_AmotS6K6?g!FHzk`;@OTNuBL$gO zxHi>3LT@^;g_?}IdRxEw7HnX=SXT8U!Lzk|^#K1eGiQ;YoTe)7tpT#^Lrtw@e&*Nv zouh09#A^-e0>*s|HEO~_I#@?}^io-&7PMC|IAnV&P#b@G(U#|9!R0|NRV4EKLxp;M z;^l&(MpY))eTIH<_TqKKB%SEOdQ-}CcYbR zg=O{MeRHr(5x+FWRT)U8`~12(@9AFsye>*+9VVz9BEtR5l@~8>>;t{hy)9sioOZ9&MyaCH-8Ij%Jxx3Fj>U#n|oS&vNZMi;9Xq2KV}sG05aS zLryI~-ab$HO5#btamK#*R70}BEiXyM}@_2h-;6!Xb zb?@zBh8LQC<>GS5$@V7N!wtZt3$mGP+KZa8Gk3Y!Q}XUP<@wT@{rg2xQOY*5g6qa} zAA;%+v*1cxxDD#q;n_uk8k2k^fY0)k?T8pk5~&rgM0CX$Bmc&WziLf{1@||a8@KxP zw%>+wa?d4G3GTwrix|D)j9?ar_Z8h?LPfA)1wCx2J0aIj7e;VrsXfUylv}GF&`OQL z4bktMmvnoX#?QM5nkR?GLHRcSTM(GC4_sweTY$57sAReaf~?sf<{S=`*^+yZsyk;H zTkxoT#pKhFO9yC&M-xnR7F$;`UIZwtI0i$SNBwP|c>!wdFf%|a62IRosOvp7acFdy z%k#Cp~`%t>BGQ`7V?qXG*!wZXo2DJ>s*!@DK*}CT*3U!z={Vb3!$a1}S z(P42SS8pLciU{?Gg&~lJ!`dRKWf^7l9z8$C&o2$`oRoQ6xZ|vZEs~C1ynn$>aPAjR z|9jkoJiJkBC;4~uCxewPD3DbJ5mN+pa-r@0wEw~$b)-v=W# zNrEQ1>usG4S#K;jL&e|;oD|x)29^@y87J7jEP$XImC)?WYCu@+eHKc{#tL+c0H{C= zqRT;CoGI;ZAXX4`CV-TpiPY*%`Qt*dDB4O!SCh_0awu`{Uwc-B3fv?bXxxX`iaWH` z4!~Hg2@s3tYwf$WjBpJU{FenPZ>A;u>>B^`!eKzpVs0RBQGE7bIYT5CSB`Q^t@(1$ z0KH{b-Vn(J%j#&lY5psGU}&X-x`hXYiL8H3UpQmT+`;!l3#dAREAS8?z~ zsO@e8hl`CdPxAKjb%}vIi}??VwdQ6%X9hFBin}r@%VC&9)6! zRjL|(H=zvwiC&=wH}nfyoC>EsY&IeS1mrw6+;R{fAP}+zC_g-l9Ey23GFme**Xcb@ zS$)3j`RDrwy#eb3e^ya#0p0>~5vbuRxf0$P-{S(EMYP0|*oU_BB8) z@)>%a_wW>t$vi6Pe?mwtfW6*_QODR1I!6WxdC|t~CR%aqwWCo|55Dv}w`0F9(!OtG zstSDC>nE{UG*WZ`d)Ll$TAHA&^HMLHw4aK9<;sI`oY;wMrJUm+;-`%0jh7;=8<9R!T z#e}RO)b0o6)p$q4;-Oo65&43C^>%_Rc5WbaQz%N3Ag{5x5-R>!ys1xZb>wrV#y zl8jtDKW!+0ykX1|7>mD$aoj&6e3+TMANR5Sz9<`dv}n9$E8-Dp&^JPUf~T)Bt=zQJ@E5}+ z)c!kyF%bed1?mhrUiAAFgWiMwi9=)%CMzG#9r?y`*ImDLe%P-7KLBGDD8oW+V_{b1 zL-*`g0Q%kb5jSw#N`}jY-H+o`!TIJgMOF6}uI-49?8-zPkRDC3eiG8GXKqmK=|907aTHCX9vr5W*+Mx0ViL?}mY~(GkoIt|xyy?L@GN zLbwuQ+Z~F&QX%wqsYs*y?!Wno=gkp~8PJU4MlqXia)lS0#__EVsvx{0+$+!Yuj z&xAtI*|;|mVFUa!w^Z7n{9v%#6X(+KH6ytZKz;=Mof$xL0Wq9FIq)QKmK9BQN67O) zhu#&H+j(Q#rY5)z(+m=x4EKsMqU__ zT(+tpy)FZxHZ-gAdwiG69j0rj;|_>1$Y?SL2-Nb0zFJfehJB~BX4jSk9sBa78d&lGzZA@gt1D`q!_ke!!*az3? z?D^#ZHZvxGxjuyjH~=w))Z-)OOgYvr&)$(e`SExs^y-FcemcQ_-#799MPv2<)$c6I znzDwLX)3{$nTJ2FA3PkL79&mj&wu$3|MbW2=PyfsEa=$9b^n+D^vD14_jmvJ%m1d- z<@(A0nCox;2SZ^8#bST#^-}$T|G^*m{J;O@@0b2xfBySl|9->2yz?c_;w1)m`!E0U z=CCf~y8LZ9#~i%>m#m(_H=cj{UH4D_>f`_A@7MS7`*-2IWZ-wm+79UjEY`(jWi%5A3hl1t%DW<`@nmE);>`|KX4E zKW=fi=70X0!zc=;NtVTFoTYK{|Lp%;rFm8U$2*VvKZfFu1-$Nu&pYn^0ME62>SSH& zq5q?v{~#EgCI5;uEcy2z`rqD!ph1m}`6tc$rS9_o#r^OLDCu8mj>RyHX0ZR^ZZ!7q zx!eD`beDfu=|~#m@V{P2mSGqiaLLQRy&d=Ox*c`lF!rxFU<))uaRh-g|HJ*bf6x7B z(9qmp3DDpa!(b%wZ|{~4pHJS!zsr+++4H>ry_QyZR%il84t^QX0GyE3puQe!;p~`x1*8r9++Dwt{Sv&2eDK#b5%~zd|3Rs*<>uZfScfL-xM9_;2?=6Oc z_6$w%imSk1R%BlXi^)cw7lzjFWIJ%E{y}CLS^9WZ@NAAgIa$-;K{L?}cza?+cjdJ1 zyeRv$)={gWJ_qV3ad;E>@^s@1jhbxk z^52XotmLNADS~7N=w+nfVq@Rr2_n3?N0(^LYg<=c-Oz56z%Qr+(d{w?|)07T~5&ksy^P$4z?X)@vn_ zY^wqIRZ0pZRadSoD~TfHn=C9EQ?{~Gvwgd@M10daPmw3LYpF_sHpnt+epzUJ8iNHmSk*hhXZJU5wtU0-i@lOzeXD2hTFogcT&lCLIF zJYJ@&$#9Y>f|ihjqa0bt^lHX{WG!F5m&<2$71^zpd-Skxb_OU1zRx6pHol>JZ}p*7 zEQ06rQ&yO~`AAiXNw#h|!(AENW;5-s{q`H+>yW*A6Q~pPQJ(-S7|I3#fp=pO(j5;m zR6|8nkqGg$HX5H3{#@Z!%!ETF>5I&(@uqb+Jrqr~Zo-KvGqAO{UVpTKL&GpUw~ETI zdN^^R-aV%gia21enCmYONyADHM>zyuqKt z%e?aZgy-9RENE!@4T`r@`c+j|9T2}u4LA8y)4T8M>6r_fCnJ5lxKOYy>mhK32AB2T zdbbZW5BMa<2PZDUUQjUTGL@mB(2qvFEGpk0a--Sp>a zFUiBzmDv=AJKST~jXoe^Rg$1+j-uzAXXmsi=$a(Mk9w`x;)67JJ_TQzWpwOWVja3o z6T|wp)x!zqJF!obo}F`P%{V5A#+hX6(CoZ_M5aHJy8mV$`Jf<_{g?J$D zZ3l}!8VJ&>geKsG80_`zLpKcO_C(S3a|hr)zJD_kupNZoPVyKvI1vvGg)Gan?PG@! zKmV@5N|6el<)PyoEM72Z$PoSWBlP{31{tSVq2V$QH^XLg9yM~*Kyehf(bwwS<+(iB zEdU(9+_T5%qUiQ93Iik-hsuBb9F`3f5+k9Z z2waG}Hv+~eqVIRZv3x{1V+!(v!^vL(PL?isU#(uilh=s6@}HuJ(L+RUX)6TR5*jr8 zjH4pZWGSq(b@XEsrZi&9@lP^PR;``pa%w)TNEaPnTD7xqCXI9G%UkpaGs?K~7uvO&O?4;xO=WVCIz zH#c~7#^N>kLe$&mB^GlB^0d!fQM!`O`7nIfHmZ_WN!Z_ivzrJmX80$)f1j4PA`0UdSZ{ z!6u!k)evbO#q8>6_am2gUkAjud!_61XAn9GnwzR2P8ZM&F@pvv9RqSZr+#1I@%w1!SKS!%BhxXGgfS;P*Du=SEL6DHtdF*KsD6gZ<(xaggtS3t^oC#$*k02+OXl zp_z!FN{6_f5Y^dn617%&iS{vx%}z_;G9db;oL}mPI_31PnSz&^mu~yVJ??k z#GjX7#IP4=m-v1pFHq6Z7=R!Zcy`=)XRdEIM~^<3UA9$amQ7)gqA057O>6;{wfB|* zpjVJY@1~0o2b)-{+htjOMR?&=Hg6Abp$#rP-4fzp>HhnumhSMvssbMc7s}v5vmL;N zW%+ggfBzlF=VFVpex`E%*D8AaKw_;lN7z3}qFr@lTXSbj01sQVMm(*)te%P-* zhB~pzw>Dxy{*d3Jm~|Z;|IN}_EJ(^MQTTya;I^zP+}(D# zyYuvyp8xKqr@N?>$P?cIGV&YgsA-Q*Zo3C3_as&&OY_&704ra_elZeDyXt3hp#|Z+ znbc10+9@FNs-4=l0$asS0yZ`$9iW1CDrU#^MGMi$Aa*V)yXxA?VrTE;R{na;shK)A z(_bHCA6oDmNm6lS@8H88=AmJyYq!!0BLIj#Ho}t9;_B!}qHBqrSK- z$mF7USv*GG@)Q*;cg{7r zbeD)LfkKv5qJI+99REzxI_MA&6~Fx7VNb zdirmvNwWwVtNf)Su0_ucqL-@X#M*I>+M7g|l+Nm+YDtJvY-b35V=a8p>{on!0wSeC zGo{&6zKE>Kvh*85IX1qRA#5F8gLElIvNQNho4caT!G}Ls+cV`Kn|%Mz3rpMN^q1#k z*8z&Cr{|k+=er-RynB5_;3hj~tbdzrQtfi2VAk5nSRz_toMf z8hIhqCWPriyqj8^+;8$qOWa}*N+bSa8REn3r`C8t^W0m?ORH=ovcU9KlGGDHEuV%Q;G1n-;wxUOv_MDKy#43{l6C> zO@2D*;pA{jA;xjT;Y;T08c03+R*_F6s;s819KCRi;rZV#bB9_#8YlF#X3qROol`d$ zuGCALb=woi===xjv-$*1n%g)6%IfNe;uboZuZUtr$?slEasnGjzupy*Uh?DDFIQFY zn#)xPN7ITGf94z82pv9%4{=no`|r`MZgHeI3b1iugM6a_GP7curL#Bi%}G?}@mnP6 z4XeR^mv4((oDpw(%H*EM-g4KwF+6Ae5*zR=^(p!W@h$BpF8%l@_J;M~5jz)+I%J^G z*=CKpYAK=LmFLgInbmduk>25w8;q!epb%^Wu9_>8_{+e((IX706|5`EUR|-W>({no zlnD=My%skums)Y?;<RF`~vDMnTR=ilPYmfa+lu?=%rgs^yRBz@%ymsYf?Wj2rM;?OVchRT-(at z;7ra}UXBM>ohec?8)Y32`zp;t1p=BL-}~*)l(n-}N|_)e3$r7H8m7A|i8r_7`CUyE ztuXOWDz{$R_}YH$6lg5&|6en<%)2GmU3GjQ&f2>i zt9oTP^YyQ3h@Q)@md3;q-MZqcC*fCJM^`%3O{}`F#Q%LzI8Vk}SZaR3L&KeAT zUijsO*b#g&r~-9rWV*aFh2;=#`nwp-u&O(0Hm`7*zKFNI zsL|Rn&k7pm-WUsxGn7@;2ay1e;MNt?@1Hb=fBh@J5g6ro3>HsUlFW}=y>)>VDKZSG zipTW~yzON&%YKlrDkaiIer9{H#?tNraceKX5N>k`@=%i@!sf)q+M>#Z2b|<2?W?}* zHrc`yml60dGk@C_m{NbQ?2FgS7HYT!K}8K(w+>tNtF$yuZN%;wbKEsa^y*g9x=;%O z>+R69aEQz*;gsr0`1CkM%J4~kq_%SQ_SIU%V3^cvFUKqwgbdHNK8E(y3u}3ETh6}Z z-fYS@*Z%~Of3IWwa6|3%g>c^b!C2zV7|LJmS;%ZBvs5t?G>BEe{IHnfP8S9K!3q z{q~PsdGsrL(s6&k`Bua@thLxLX#bHUN7es5A~RCO_i!p%GEDEGg45o5-_z~gAtUW#f#?5RK`b~sNr zGz{@De6OO!D=1P_#!qjKV}1f_Fg+l5A-7q!C5>kDa-Mgl3Pj!Zm(62$T=`+Q{f?JQ z=}pDC*+q$RHSN&4o>JLO(aw0kz8c!S<0KpJ8pxh6KGFKUd~wl#SS*AtXuP2Jr#JT+ z;c(@WK0F!H|J~c4rEz8YJd@RodHe90t^DU_Lb!Fd3snh=)}8qvHuF-~YYsSGN&e1& zPFwWN^{VKaqHa6?|DDdHWl-)`9Yy+iR@SGtqS^V`o{Uwt5Zofxt&3f-N~HG_D@*NA zabt>HYSm_t64ve;b4zPWDJ+7wCt(g~vx^ZPHyQtHt2;|!r-am|PA?CG#+zY4wa}+L zJ#n@f)`~$SowM~iLjy%Zc;4sV`3D1LPA;jZJ>qYDGQZ!DfI{o(-73UkWd6mrI78A7 zLVkaJlV8yLJ+3xXbO7e``K5@!O9aXNH(xyWPpeuf99?<|3-5KljeX`spl&w79*)uW zT1`{a{kLw29irqM&bT(};~0R>zvnTg?i3zb<`E5S-yzPpuY62`2F3XMnd|%OzVB53 z9TlQgu)Ec?TF#Ce;Re?#6!EeI0pE1XE1N1md@^Z2YH`)40@S}4Gv753i&O?*$*R?D zid3K1W4F?q#@zIlzc-RQe@oNLs?J?2^~nWyS&#qbGEAbF27ZAZb0T~+kqo`b+S&r= z&MfVozWgq@`LZhY(Bf5zZ--z1>83fE%=alm^sa^-T4f2%QnB8$t9C^($5y;Ge|MJN ziv{y@A}J?L!SaRqIx+A`U@AiJZxyk=NiDFTa{xHxx;#!x5D+`(+=B~TlHb=q zH*Sp&ZJ?&vu@9s(y~q@Gt)WuZe{&MLx{e%8GdpjbsN`Pbxd(@)dD%;ke4y83>#N|# z(wg@zt8_^}60e{46gOvnZnkY@!bHm%?Qag42=CDo}>1}7+@-9l@rYOemh)lbg{CO0H z#+cl$UUp!A$#l|Uq%pny+_Cw>`uMKlDYn{KcN|g;hd2E=ueZHl3Vk`pJge$U$arb*CXwr?9eH%z;iY4MK2UN63~VrXZ!a;MHA!^~f7#@DLced;6<@_Xe% zN2Fyl{lqS(C{EfoV2OYF{$tkKj^A8^nuuMDPW3K1;*jx3QS8LeIBnbH-zFF0yJ(_t zlGb?mu1)-+IIa|!XrAS~TkiH(LGWJ7E3F~q`TuoY0f%SO`pHcSy;}d~#*R~=nSIBw zqNz-xJR=hSU87s<7S%+DB9N?VxO0f@@-TwIR{fIM|j@&SCb{LR@pRMEM9u6EM@%~M$=ylDDl zXZ$t^`V$xV6-b*!zr^fJ_NTJE-Wqsyp*3ykjC%+XPQuKwxtHVwOV0sXOmY97HAikU zGw#c&wMYM2SzKFd{&lf0{6Niu{(jc}es+6AZ37DzI9GPX+^47c2WKUNicbr$h?NPi zspjLaU55J7)sLJ38wa?UOzpPyB8D3q-YjS*Jae$dOUQEvkt#w+U`YzXZ<*VXeQODZ z#fh2~#lLsF0%F~iSj>PYo!-Se69I|LF{=j}a~(N`GVV*X@-IRC*5tQg&bTGUxNbqx zR1|Q82wyV|snY6;JFW&NJp*QIohhnENVEdD%0!A?eYAoi4$UD`PzO$!1J&SYzUX;^ zT$TzMtl+}s%ER}B2f@sV-! z=}zJSfveveGp)w_;Ye`rRl$Fg`REhxMo6~h$an|Us^PTu@)aLj*nNpY8HYH=oaG;~#qsHEYNK4t|%#jSrk~ejSzopA~oFzw-}2ioL#Um=CAK z_%kdq?(q4015{L?VYGO!FM6a&KdGU?XRxh`MA3AKYRH*9G{GCb6X1d{Z*^599`H-L za*U_wp&mc?P}(hTFvY&FPQ9!%J8)Kf8wk@j8u1<~OS-^&4MjWhL~0r%gTr5Q=9r`p z$YWaz@5WyRO1?OXO6tTJ0^n}3#6Ar*b&sMi$0cDy5Pa1KU}^$7%!oJd1H!18j#CWt zdVL`mEM;8U>{7d)gbvA(P#QI-gwyF*z&84o}AU2nPR1} zh>7w2cyDf1oiFD&@Ue=6bQ#6NSd@Ax=WZp2_eMUlL#cu!?3yCsmFS)$T$tDL4bPX> z7N#_Z!I-Z8O1&9BeVNa`kl43QGlNt^nzeeA(4x}$%RvTF@HkZs4WI_*v0PZ9x!5m* zU&@u>+hBn+Xcdd!`JG>Tz&9+8J3t%?ob>H5xpq>!-9zs6MZZNg^MYhAO2xGco_En8 ze}0;x0RHvMj5Z0bBekj1JSL>_ASw>S8`(H3-6(gsFDL`vWr=qf_n)BEIL4=7-f89P zrV`nfCwv5-sUI=o=*x79tNP75azM~=-)jw8BJim(d z@E1E}TE3#D#sfVP-G!Cya`_BQ zjl~&{&E5~^J(i5(@A*J_<+%R(SF!vUIjJyxfNFYLmltxfi@LpbJ@p-NH>l`%;A?Y;c1!uxSJM^@h^Cz ziw7nQr05sX>j`m;xUxausom*~?T=HUJ5w(*`v)-U)idR_aXE2t}8}pTN+6ltDnlV<&X5g>hd#a?ySZbc54}xX(I_m+(*y?%H830nf zX6y4eK2L3jH{*Q10F-^y62=?1LSuW}V#Y7sSb6`^%t`NJ)58dxnQood)Rir7N<`y?fDfHS5KxfHQoU4Q zrhu$1Aepkuw!3THMelyUWWiImcYU6`m_=dtt8fEXcOV{YusB+Ov=OnQJ0&%9rd^6^ z%R=mS1dr-kTC>2U_Pl_(gB&K?eWVHTUu1=3V)d8Tq&Tlrn$~>$Ez=3uL;RfU!Qtq7 z(e3AUeyuwp5G*+@yT4Ym8s;}e+kMbN;1E>@PLAI8=N*vm=;=x~ra{O>K$gC!d8b;~ zVT28omKLOp!(q`Usqv#?XAJlASeawR8siGegM5g01O!R2V~(KjFg^j1Ma_ts@6o)& z4R~=GryYNs07deGYkyta=Cr-=j95_;{ym)IMbq*eD|W_fPXaM4zfRXSBTu%XOg2N$ zwqnm}Je}Bnc0$ANs0TO)>gAp}&nuI*N|5(Qy}jZR@F~F8ZbS)a^5a&FMt(j# zCQM(nL^Jp_om4ct=48KOn1>HB{9 zs`}<#D5Gy*H+gvBs`%wsL_DnVcj~QhhFA<+yPE1z(M&iqb{fFpZ&9AWWt~}TK zRiOrvo(_Qlg`)uUsw0e3&3VB)sDUz3+@$(3V((s>U?3MgINFH!uJWJ_@=Kk${`{Ip zZ|-$y$?x!u0d^}k6vI{U^LK9`{o6kURwtG9>;U2I*O_Bg~?c0Sa!x!rHN(F9I&9*Mn6{;R0I)5Vns#`_HYd?_ml>HID>-$pOhYn za94rP1RzVE3}o79$LInC#~*`a2Qm7PV*J8q#P^Jcpud-WMaG(iG^?|Q1BpAw^r5wd z&35aj(AQuo1c}^%njL=!ozV9il}%bf$~yn~6f0Z72x^w@HY@HYk|$Il*46_^@CX`z@L?;wWUtDl(~8h1u|ds;w6sj=uI|*UeWGUTs@s6a+1Q|5Y`7T?UYfa z#X5G_NUP1cuzchRVhh4(isB5-Jj&UycBHcAd(EGQNcX@N0o}xaB@JD|LEW#-L2o3o z6XjZD&}SwqQlGIinqIFod-&QjFYCR~W~-YV9X_DH zDOGwg{pF!yM*(&x)!_5#CMs?K6)r33$(4Z2z}M>%T42Rr@TAJCiO9i?J%k?8>`cuk zR$jjQVi#IiGb--@zhL^Jlw4w{0l0XI8UIzmaN6II*)`pge{9jdxV-Dp_&o+1c)dxs z8dFFl#qh+MNmzr2eb9XP?I*e8eX$M1Z$Z}&dEU8AU{{5MASj;n^yH=2Z{~5M@c^#Hs4NUOv)iYg9(mUvYD9B04qh zc&~_)8jZIo;*AtjFntb7p&8LWJury9Jg!V%@R^r8-#)~j@yjna@z_4c?sOKdJTxa1 zR8-vDNozfPTV&pBqX(+?vsG~NjQ%2_{d|KPwin$+ACgrZ=wd&2dc>W#2Yhv*fvJNA zA<}>~M4N=GE~97=%pxFak0z&mm~wn&o10LtfC``t4ZZe$V5*T|WH}9h!i>9(aIrnm zC#Uow0uHU#*6DDrUUNJM?OF0&`7Upvu3fWU1p7{rqs2N?bN{FYbOGubVlg#q!^DHpJd z;^@&9+5owF0lKtiXwC{j8Lt87cE6Nc-Cwlx6x_JEl`p!z=tt5>d~W3Y08E&S{@_?H z)Q%e=a$PiZC!?=JQ>9YlTZ*qqE4Y=Tlry+^{`R1hq)@uaSIGvL+0libb7rmyIaH+u zC=!K_8M#Jz_NV;>_OJ(qp_J54`d!Y;>_$jTdc8zBOD5##VtluOUyx`9R`_?*>!?d1 zzD7Nn_68sWi!oCPOb_OZ9+Zmi-l#NazquM?jj6{1D=e275{Qh?&>3mzkku|M!pMAk z_4pHxRZsNqX#zm2gLLwclKSS5ugctZNU|IMnyj|f#dJ&}u(*-?O0pT4t(++XAz}XX zI0F2fggSZ0h-bW3&8&brq74D^n>!o`3WN6*F(5j9F;uEmlCvq)xbU5BN&}oz(3HrQ zl_e@Y!tp@q1TU6peCS@tjTIcKrMh{R{U(Hi)M|k=xS)RhqS^sjTj9TbJ76GA*W+{4 z1i68_02f#Ne#Yhu$LiigX1CJ(V`W+Wdlzopcprtf1qh)umF&2zes%L349CsPXVQN) zQsCh6<5(pg3?|J?r!#&N=8@>+=kIsP9QV56Pn{vO0Bg&w<5&9QIdYPt(z0jxSJK67 zL*hGLdUVp#uS+^V5h!I?hf4X-&~UX;gadB(@3~RtcKWHdt!$3T^@e2Rs6Nso_F0|u9YPXB@0|VX)*i1 zd(-t-vsqbOR~EG;>DSwzrpH_}3i$*6I`X##Kx#uS5i>{dF$v8hy!)WEd4$@pmA&Ks zV^_h*KQ=LHc>uQE0zBA{5(cVz`_lVJNqnkJ`b{|S>>j|$tT_(1GjR<66(esILeVTs zQfm9E`LrrUTK_$lAOe3jKKd%cee&JpUne;%D}@&3$x2^A+rZ*mh&-Z97pME0lF&50 zjctNnW_$S%12wJN_(blkspO=jxT zU8e#RPE-JaW4j#TrSBZrNUMwhToRS9b*@`7llS)wQYOrOBYkG(Ld3i5R=_yXwWc(1 z-e!W{tNt(htcmk^y~*F-i)5*HWLL~S?9*N9PG8_nm28~w@SFY5Gd4dgrxY+fzQM_@ z12;Ef)BLt$bWS#Q%&vTlx6Oa;Pa}WM0j7+EV>#)qb?3@`Ochn2?B7~gc^*Lolhvm2 z#n{o)VjKB;R%6-%oG|LdaYtsstjm1Dm-6&$|KNB4iE$oCwMYCtWBt{rx=9+s-ftq2 z*EMjSYcT5F1C2a>gMYJcb&#+^>2Bn8;9p0;!bt-YBJV%9{P)ar+%CJ+&Q$~o!1<-y z(LX6Axt!-;>#0p38f3OAb?RTSpfTgd-=`E!8T@-+%4_lrQDN!1*+*Lco>xv*@++r_ zPlwmCV7r5}@%aF=Moc4VedR}hhU)K78uqUs0{1nYlI#d{?pR{E@Shd!(E0Bj7YU+m zAyTazsHOUdA{{;#nfM)~p*6k&EQqqiT)f;cQEYC!SbJXn|N9rYv)v|{K*alzUigE} zyRxF3mV@<7gTV2b_V00<;qsLa3&#)5sQAET)tFzvv@haZ8B<@L7|*Hr6gNthpu|1yqS z8%D#6r)>~i(j4Rik(7B{qi_VnaU&`3OIO1oD5I+agSuh2l_u}&WYcL&v{X{DSUk)pnz3d>)WCN6&f7XwN0@=>&_AZAAfHl{BpxQBxBz_Vm@kAG9Qd?p z*M{2|2((d_*~nJKU?1DgBpN}i8H1|dSv!noULOlzf4CPPi?t4La~#_>4Fd@4odfWJ zWNy*si*QkYGC;J*6;XWp)Z=CdK+uYcoHR?BBD@`Xm>fjP`up-ZLmsG**^G9_^gV12 zI!8TdFE32W%$(^%Z=_4TDlXw^Bh1LZH$@gi<1e?hJHdh+5BwANKkXVTOxBpccPHc> zD&-H0@A`C%sTR$f`K$(B6)!=2#^WUY;mx1BA2w`4(8JiG$PNG~+p8o9)?;@ZD&TL< zHo@dbi%@9b!{hFzQk;l+v-zj^J7e_G1EWbcJkL*7YH7N;{G}_!6F-1_f^*{?q65PD zCWr7RQ|YO1qSvxhSVz!lxWCeiRN->*2@E_Xc2WQaWJQ4}L?gF89^e_(7`jsvYmg98 z2IF9ZbVaeg6cVjpU-6J{1_Hnbb`0=N18>z;JtwQb%8zwt-SdbpWBu$1n)(_cp|;62 zZmf-@wrTvT@~=4jUeqFuhg3ivu@Y(<*&Gd)n#3>KRZ+udB%B1w;2octB{pxSjAS7% z)(WQLD2}^7bz{kxDKCmv<2P%uB)??8Vw5sdbsz8yV5}5?u?a>jQG};L(Se{ zF$)EP-^U;)fRdK4BfbO#Ol|_J%7o@ob15XMKRN)Oy+dTz*Ue=}42I#-;+gk)tpsnu zUFh62&}{txOg%Z}40LfzhuSt1t8rXn1Eja~8bkxUi1>_2aQdFjRQj&s9MeAY|Pb6cB`5VMZF2 zmecB7i7~-)!oh(R(uthO+7MKvtz)h4I`x&Tl`{2qBC?1Udlts#GnbHJdB7lnPhsQ> z)ab1)%J?-4;Jnyj9G)K%jTPJuYujrCR!C?w`^%sHvdVQg($@02{mAmKl8BjlfoH5$ zY{VPhkr!o4mVZzo8&fkY`IG40aTrJCpXUpESAo}XL*dKD2zbjYExxDT4RYJGMUhbv z-8)Ja7+3Q569v&4w1q%IAme6&bjOqtlKF?+tsY+Q)vT@=T{gK}$%>OY??seE&Jjr2 zgn+hY&D{cpPcEo+Rlcak(o(H`>2VC8gW5q!kW9j(FF=&G*VmJMLI8{oeEpk~3o#fA zmSJb0e`eg>Vx5TNNE?ePO}l|?eUpzc5@clq1;vGvbO9JM(MF#smQbW=aed>v@JV|I zoY}wg&{qKb2rbSFHFy)4k9jwi=kGe%HdG?dspiQ-%OK&HSbu>LMwBzq`6xTKYYK@a zI0dNv%ad`Z9<>#QzAqFS7~%r~Sk|v>^@)WEW4HWVellV;M1`_}_i4y5;t6md6yIX@%SiQ^e*={rl_tGmjKOl zdduALN0AauDVY(6*^Q2-CalIU7=R{R8UrXMS5};YSmR837eWP%*^w3~uch(~BTBxo zp3B6r9~!M?UaveEYy)ca(-4|%W^~>>%naG-D)bti8K3vlB+cP4x!Eplgh*!)R6Q$7 zS0%GuoF65+!5+1{LnUcaeok6$zAYwv^L)UmHU%v-Lm1xC2NmR**UZOFX77ATP{I{O zk#+Oy58RKIbdxT?P~6|#dH|Q~C`#g96;N$nra`+(`*BmR9fQ8<179Ia@yP%dWnNW>V>! z$oMAn1?-P_i}ll9)_q;ZcJ9vC$^dzpa**sE>M!QZdX6&M*tQP-o6po-^=FA6^!+@x zVKk)XKzz2PZ3?3Q+|8&7AG2&o)>Dix2%6wmlaW8KFn;+Nj|5Nj&+r~SgJ*`qY*E`& zu!__ZbSL=rTq|m`Mqc$2EQup=i<1>$#+p#gYg*n-eq%vAyQWV9k?PxM$y7O0|4%RLV%kl!zV6f>n@K&JK z$f4=i3>Pdl$?}EetEeo_U~0D4wuw3vp&5FiC1j2Qj!vkUe136;=&S*}#^V(x)0Hj4 zxl`+;VX;{=tLh}Y#`Z8>0bWQuG_N~q z0Ooea?fC2T!67*7KI`8(+!IWkeF9y?mj3RKzy{$pT=@6m0uF8p0X za4Cwu7O4Pt2;HlXb7k+W%OS(gD?|_HdPSMPhp1rO1sYF+qaBPPtOmen;?nGf(XjLyrF9_oFi_+2(z0r@jnQ zgbir>9X!9tVIL6q9nX-uflE@=!eOTlg1NIDzF;yYfm)Ptc*`Im>1=s^f40Yw@}x_X z)fR9E8bno}*pSC38dw1iiXZaJj{K+k+LOI#JsEyNy_YZF8lp#NR)l^bF#Lwe%iG1k7)g8Wluj^7n5XRi62M{6y>9s19R|IH0)P;6Vk7LmXcu z8fI&>Y`(%ivQT|8Yh_u<&#EvOkGGeNOl1|44h2C2NMq06*LwL~dZ>C$6NNF~Vp3)2> zg(?|5iiewJXja@ADS;zN@@M?#A;Q-~B!Bf{5?|R~0uU-Jy+bwB-?>_Ct_Y5AMrsHx zQ1xFy|CXTj5)BVNCa)~tUlo{O^!OBM<{;(je$6+8*P<_i2Lc;26(S^j;S$SBaJ613 zARDM~i5(t(I6HZA)f;kLJ!Z^=$FKm*peK~&Es$?e58X`3)?c422ROiKb#bR-C_k(G zW8Di!K2RY2fO!$-HEw(X1;;sHC$fBWvYbtg zjQ}b!a#ygrGmu7Og5FcYg7u5htbbI5Qhj(?|ZOu31ro{E#h5v5FSciK$(@Dh} zd+kW^s^81RnLoozJy2+NXR@SR2(Au?Dcx86x@X{Pyy0$0#1U`}tp4!q=qn4sS}Mhs zwtJ0RyqPO?yo&XkV=d@E-m4TPge)6j1cWz`7cr4_hln}AIygHwN*C;b_R|p`wh;v; zwfE~BBzZL%WiQn}GrhBy4k4`_eJ|Zgzg!SoTl=eAXZI7Y!X3wf55`c1a!0E(j>Sz! zcVVs8Ujf;b{L3@l-@bW#HhNr7ZfHM@wrG1#TTw;M9U?NYg0)5fz+44s#@jHA0;)-f z#w!>IK=tMb!&r6l7km>?{SaSpz*tOY$~)U0XAn>7^e7SpY^~1K2oj1jDAM8cyW&Iq za0m4?k#0&axUp1r+yVOi8`nD&e!x0IMZy+azw>Gyu#olUa%}xop!0xKr$dQ>mCWlS z^v0h#2bq2X3&Q$LMYRrJJ$B^Kn{+DlrQSo{ zfA$-+$AjZENFBp4;JW>{)lKgFP|w6U}BfCH!B3{MiE*-rfK| zJxBpUTPjWq=~aH_B?P;9+_{8z5b1=1s5+i_J+N#TCy%!{;Fuenf!6|J`NY-vFsw6D z{~h;-CyW$eHQ0DGU$CxPd?+TmqmHBSwd&bjd&pN&l6AI3oq_H1_CJq*Ff+Qig;>Yn6KbZ2= zM$Yl90rywe?ERcSqdzpwrAt1h^qbSFH*ekcG5ZJ)YE66*eEliKBBS(^wQY>##P@-8 zW|j*`F(=C-8$!&6k}{mXd+xv;?ike!5g@jDf7Y1bayrROZ12R{ul$%F-RYL}kRkAKi^6o^ z*BGF|zb$^U#}mYB&K|>Fy(^V|tt;MHVAy1bahzv)tQ>e>p*AFT@e&INTk-hlT}bvA zB6xPJ?TF$V!aG`?2migCyeb7sgMT&NBJEYPOJhiypwqyEuMEwGzXGPH0>47|!Ph%F z$>TWIA0c;ZD<&% zoH#>w^K#b!iY+gAoja2&;4tj!rZ}#SlLjIP#wr+SK;5xn=^*<7DYLo7I)wX+AU8nt zu$3Q_Z}WGbYA#idejx(os@I)2LUsSXoHkc#EVREM=+S@ooNGkuUoZ8%I*7(vSr78( z4DcC^{a^kZo6kpW#C(TnzzQne8a?57Qib>PTFc1YQLdQrh8Ij zyYeMz1p8MA{f&de_$yWCL0o8(Y=`Q@W;3yx$~1DU20>QFTXgwgDb>=({a&RYL;U1FbIn}W;ff{OR#vL z;4~z4f>>4{ibsB7P#*+RNWree+7ya2TeYxU#q~Oet|v?BsZJxxFeZ}qFSVz#f;U1Z zXzCAbM2Jp0%Y|UFmvA#Z1kt89N?oUart5?cAjUhKu92Kp0D-A5}>X`b-YclM0PJqAtdU)1=#M}{d-rlMH zclc>4jJ7+={_*jG0;?GES)IFT3eZw^H-WCQmiGX;Gw`jHU(qxosy677I zSN|A!y1zV1_ZS#1c=&+L>cbNnujKaxx5-zitCdpfI4)=!dKx~?rbR+Rq!wG_4X)#$`lemzYk?xwDDT7mxXBZ+JFB&BbKmZ%3Gi?T=+A& zKV=8=mu;5{%gJw_{OwNNfAX0p>?wULj|P_wz_Jz`X!!LYjSm3~MjIh42_*Sf20eCX zZCC$%L3H9gHE=fC!j-QM&CFQw>vBO=NQ9v%DTZLH(YaACd!*n0+I!PxM~!T4^!xc0 zJAC4V%Lliar;h6zAqfEz2qB5N+u=YA0>mU{=;;4`Gj|QHVehJH_j_)(HRf1J9cF zw%(9Y*P%vSjUXbA(~yw1DpIO9I9CO%@F|&r!z|Wih=rTx5*R(fa0}e?zC_X0jDw(> zj-<&>$(F-u>+O8%Y%OkF0wCw@)Lt*d(hdmO>@gb&P?i z_0dj5lh@5mOIc}9d17(ahkbHj0S?Cr|2*^@mkZ&H$F^WKPJ+49JOiiwCD z)QM*MyK^ynJk~+*9y!ECRg)*Pj&-*8eQKU0apQ@ce^Agh#j@P8_VxWmM-tsC2w!ll zM_a}1H9@m$wZ#}^KsSsf*4Et9PL<9pz?J*zu}L;3Oq)X0KvIpFm*Rxo&zCs8NTt8U zxKo~7n^clh;AM|-%^vb{KZSa1@oVAd3YpC&0aq=5;P z+kL(1M}*uelU=TFH~M3XqtvAf(~hsH+Pmqfhjtw? zI@?L4XB>@9h3WVvUj=>xF!he8&iH+l>bwx(HxU3)0X(ZBRWwb@2tDH@Djj?P759>r z=lz{R2iCIHrRWKgyzpCK)A#zJUoq6KECn+?{AP>k37GQ%e&aL%KYB-B#c)qLB)C#r zS2z>gMV&1%2Y-u0A10*|$jPF%ihhW#k)KN+StsWMXeVE7`|<|AeNWwA*@E;&>?dI@+1=C#@P-r7+a$xP;|XRfC`9f&K}&+j)7%;4apqmVTZ zWo_Ka9e zjk6*6N0N%&&P-Fm0^Cg~QvOC?&1AFm0XZjI$yN$B$aIhC5;9sfd(Q?sQ~}Zj02C8$ zCflgyqb~Nt1o7XbtNFa-@Y{{&IY)pmkx!VoI?CBpIZaGkIS8ij?I^vbOqA0TabWts zIWe7icQ|xgNG*NOPK8!K3W2vW3~4%iS5y{vJFB`G;yb`m)S5f!fKJrrz(j`~GHIug zKJ*m=$jyc}ntf;iw^2=-ei#nwvMl_-8p4UmER&OiM7>CAqo2`&K5SNV{rn6X*qqXU zpy=`;f!uJq(2jC7HVp`ctN}6f*Z~k>#ZH?` zyyXvzRjS6(PIg8xN@r8bu=9oL^SXWMoBzChvqNuCxXG zwK+F9MXBOUnab5S^YsNLHge&1#uR(FgolQDQC6%rwr-(h@SFyvWm8OqYceiLhcs%^ zWdJrtAKHz~P?sy*Y-Q>dDNzqwWwdz_EHhg@N~2zRdPWi(pO1m;shYY(0eQ>n1d!!u zL;NjeWh>26i$whp;@l!%%=$>Hg;UN->tw5>i%0L!>`&|R}V$>-_y6rRyXVq)rgIjpqkI#>JZ zJWm938r{Qc1;5Rz;8Cwhh?-G+$w!T)>nuf~rJ~~Lc$4EOasypQ=hJ--^+Q?Co<<`G zK^9AmUnyP8TtcsLhKpfxK3dx(r)8n({qZ@m)!Os<{Mxg}kwZ~o1=fH)h#1U9gzW%p z1o21E=}YfOdgHFvjqP(rp&0{(>~tN`OQ)o7NQJKJ-Z^N!IZ?<3BN~)Yk0ICX1-DO< zFq*H!xE?)|ZetWZ8LOow+2ge-Lq+5_65wc&Jlbn!^f z_Hyn(w30L8sMUd(`jApOP7B>g>~ZA^D^Eg0ZnWV;=g`BCc*^;Sc0lF;#)h_mxd6W<3KwMYuL0Snb6!4eLi8|H=dF_hj+ObI=>R_z=ZM=TP8&Z9OLsJ_%+c7Dk0dIycC& zJO%Yoz{F4@hsdk>knKl6K8yXueoju<62PjOrPv>Nw7#iSP2vQ9N8>GitM8j~h}~ML z{k_B+8~7}wfVzH&fG6`a3yHZA8sK%H(TlvQWQ=8s4Z{<)6qS{)rgI<3i3yOiOtevA zAPS(}QB1kW!3UAaGk9H!;sX-Ly2vl*2Z#bb?+*ZK9EKHGi{4!LK&TQcR&2AJI9f`J zINt78N?t4e<-YHZb!-9N*$9v6dbpIAgzvz50Rp?$!WB#OWDY|6!qBMCXi^4%7^a+q z!AVc!VZWjS#2x+IK#shD@v?CQ6Dx!8&W%xat~l~4oLoUpDAC|2944uZ_j=aZ;98qb zVT499So(>30H}25B4MrO>9Qt6BFqwuwQFPHWoLt#jtqFfS(8e0*{Fn9VKGeFn!)|g z$)zHsNvbjO;B&uHNa#!f{7yeSu8!nTkHiC$ai+3~*Ak_(?tQktlLvwwACV|^ehH9z zm;&FmXd*eTbR6;vbmudV2X!n(Y8y!hC)w(Fh2lHbZo=k*AQFOzwrLxxx}5`3jq4kM+{r^i)75HOfdbmTo&ZA zAhs%r#}GZP=M}kUVt46*%iD6z!#dg2j39`9x>_k3o-|EbtaS3zdq-qeK~fSwof%nK z7YXlRE-HJ=<#Z!Pg>E*twXP5MJ4_ODgzNt}17WD`Iq%9z;VkOc#{k z11H3GkM+fEfKv&WaDUk@x8_tBo7J)_i0h5hBta}35Xz@=b_vps0=kiF4W9R0k8F2X z;GXWO2#zdg2m1#pVuq<<4DaXHdZ7kg=#Kz)v2AJOZViBUSv%9T$!AXVF5H0Nr+g4? zp)12Mi!DI>klDeYTtKRG^6Po7pr&r#??GTFfHoZM(C60|~4CBOm~ z0#AqSo{Q!nYMJ`Z#AB!Fy4c-fY=1kg*UaEiFFp2V>|^kZJD$(q*f=`$ptodxNeYk8IFM)6Jdtzza+_G|pxSfi^SOtEPX5Tz*-$ zW7V>V-i=dec*~wNi{6~gvT>6&+G=rt-lE(AJ`L1aXd%t@;&A#sX> z5!d3#yAD`m%?2hmJ&w9C??{ab`+bkJsN)bMOdK6(5Dk!zY*p#JU6O-UZF;?c@o+XQo*?0kvpv&t5yDfcVrN>1Ce zyng<^cZ8qPnuOoxiSjL8_=8Mt$Bey8!;Xs7+}^OA#&6n)G&g>J8N+*;uC*L5v3=^M z`)fzgIWUqgoaUXCGc|r`f?+*Pp=O`+dM#cX?=WA;KFc53*ooGej?o8Y8+T?_ynbl= zJCeZfrDuTKq!qv^b?R&8K=yG0fx1`o3dP&QOdm=SEg2?8yc0J7_S{D|CB9}GzR%1@ zK5~w+Y6&%BDG+X&lOE={p&TFBc6Jgl&FFqPEG@U@ot(-wHad~Hq>?!x=(WGjj;A1K zc1k&V0XGXar(nnDfnWr8RQq4AsP$^9=tdXySCe+(~^md&YgY z6^8oYSl9?Z#`-Y$`{Qz}Nr1t#sZnmZ#&^4@REYg5Wb4&l_9hbelzP9uE0`?N^^_WRP>3LGP3*Zr!##Hur5E=!|6xZ|dITy>?!=f7&4K3vc)qLM0(W$)1%~@*! zTJW_y-5$?)0NhtOp1WytuXfd41c)`zosF-rZ~lxqf@%cW*4xl%f2z-XbPqbCfaiB; zKl1jw=Ugext%L2QOdu=?%>>y_^Asn)rz9`*wsd&_8!>IU0{t~6%WSwycrlNuvR151 zWee@d=xIw+Yqm+LzF3{>7{!VB!EXwTMZH@ZTt>)u(Cv=90KTBqazVU-TeS5YdE zyS8c3)Tx*(iJif$nkHwC_hh}jdsbyc;=Kv0JaTSoXCk4hI%Hb{j()BjVGIL( zz3djlzvLdRhIhB$_Zq)=$%TI4*ZyRcG_8az7Yx%*_sE&KMRX)D;UJ+Jy6A%Zc!WJ_ zD%OT&vh`>c>Wx0woq6qf?!wF!Iahd4$%K< zP!UFDQM4Qwz9-a{JuI5tn5}rr+z8MWKQWce>xlcs*xzx0Km7uJ6CY+cvlGBVd4N3$ zz?HHabL;`_z;yVmSwmO_K=(4wtzajtmhBYSB88{%0KJTs~=RgYQmJmcfcIY`MmE>HFKKu?mF_70yvMRQT>m=-PIza6FoPT5{ZNsB5 z|GWce(YQey)iAy~N@%33Gd~37acNvCS!f=~)(}rCcQRUIy?Sf5P}OvbNAH-Q`vN6IUVo#gVPOn&n5(F zJrb5K&P5=d?lDgHGG7L?enGd^5?$AVNvVR;DUbN973gwlFfB$cibqN(LDSwkbXqt# zn`rE_hz)yg#brS}S;WB$F2-rRX#%tM%iwW{OZU!&ak?Q1{YJ&)0TIq*ckZQagZFnC zlfpEF`<8%vlD=g+w?h<3l^3VGm_|^Jkm32?@^X|K#@AdzQti%Yot-EbnU(7ffIq94 z<`i+m06nPP-Z39wJa6?N9!Q4O9{7weMcDAp*sxw`?W(klR^xnA@>ju485 z*lzOoV!9+Vy}AkO+kAt0JW3&7*44T!0ptOC(W1k$80f)Y7^ZJ-8xD=HSAfx{TVpji zsm}B^2b_1_!!rOOtM%7{Jru@K-p=8XJD85MtQ$O?0Z!oiZ9I4_5qfQL+F$6Pr$fMt zM@YFh8D^(v^a%7GFZigm&ozo65$d*}Ugw6+!}1r_ov?COp%Z?K2C7dVCzRAVRJeu& zn}Hy|phArvV+}>u_R85F`Ylc)DTlwM#$&(F+#yM`SZrDJ=unKD z%n_=_h-ECybvX_hNnF`FBRDRq*jQM)6UoxoXdA+u2Hq^F)=c2`3hyWXkpuW)QI%Mz z@zK)g`~W->p8y_n-e1wSuHzW5kKthy%d+WPQ;l~su=|-iBdvY)&zQD(2x|<# zsTa)JY=-F6+)}2D1KH61SSKIbO}B>6TgdUzFZ+^mLhcZ0@N4{-Nm0SkR2(%Y0Z{54 z#EFYK1NyVq9H%TqxHrE&yY^JClGCzsxrDm(8ad?G^)i>FvK%JUs7I=@&xO}tuOYJs zut#P4B(xs+`CQ+T_}WMv)YTd0%$h9GJfBxl)OYZ)EY_(P5d=TDe^jNZ0?!ADH|wMo zY~XL0B^s>9+E}A<3K+Mzk%TkgsCox8wKY*xJ~QF@L1XptVt#p5HF~2mV9E$^DdSN{>;{SS>_TGMqPoHtlvqw z0)3q5>HF{ylLgS>>Y<+f{e|jbz96X2bN%S72OaCh3}rZ!3$;-!`2Iz%XWlt=`Mv4W z(PhJq424*chYm<|aB{zO?zNrk41ZZ2?ICk|ixX=+-$g8M?NbBimV89&?V~xM$DN!K z+CkDI8+)A3EvCj4a&@?vEJX#kAIt)e+0ix)xSeN|WyzjV89%wIFq2(pNWXQs#ak6I&}ytv zcyr8mi*=m?Y*OIfN^ZccD+8L1!s@1&!h&+xalc&o2ljv`an=b~`%|u@wX*VzVL;Jn z&{4pN4obBjxyag|b8SGKj(mhAXl3wrCASOVKKz1CD15zz)W|S(5nW}^huVBtH+5zi zM7X&tDm=C@Kg~ha^()|D!6a40HnEESu=m3faKvs2fC5KH+5>A`$bh56+&I903?y6d z=up^ER5>V$!i|}o0QWSE8slQSxpF+i5)fZg0pK8r5%kXSUi(&Z0B$uTBQjDr;A@&2 zL-cghvtS%rO*E%kocU-ktj0U=k94+v;L0u4`8JZ*Q+v3EMmo9O+jj&#fa%czkjt(D z2)rK>o4=pW;%cPN}>jl$*t_#5V zut*H}yIwT#88oVjBF?R%OOKQWkCrfLGO1&XDEXRHZnE`#xv5LtF~_VT%ewX=pj`Kh z477pT1N*p%H=LnEVcinh5yVOugfD=HWpHl-K`th9r0pH6-W<#oKWytEoLQ_t|%wNRjQetvze|34~dRWJkJM(_b%4f0|e)M zWUtMl@obpvX}UhsHd!;r+a`{OQ(o|4tC4(Xvxuq%Eg_~w5to?M7EQM)_e+2mjGIZg zkC`VY?bCkCJmV{R$o4Kd&a`y|(Y}#TKxn!6N}T(4r6$yo*%Jysv0d08PKWk97`4{a z$4$dScKp|xhvv+kI5)W*_wc`5Z zOuVM)Eoszd3#OloT07>65+-%e_5dej#X_SxJmFHIdz$C&*|WW|d@(X%KkZD=)7|2(+esIot}P(kqtCA&I1hrMy&wXl%w~iB zAj&d{ZwuDz9%6(FZCY;K7j!^?NGLs90P>vCMKY zxK}ewn0!>52PWhXOacf*Z}Kc6-H2F`D#f#RGmv=3ny}Fav8Z0qxW-dNe}uy0Ftt19 zRPM=e4sN^MDm;>&3(se}W1LWtC>~qndL2S`AI*+q1aW7haS7+F+bkZ&0T_+bG>wzX z0Xt8&lTN!FW3o3Y*RjO}!vX}7}_Ce%Tp1a#(ZqW zgRz=bF(pzzJQj_}@CCpMd1BJ+LlhEJRTPFB0e@*B&k6CkWP$n4hxEKdOAN#K{l&}^ z;O{44JyO_4J(HJ;I<8aRX%V=mL?L}q9-V6|9RZGDX|eDIoy>ZciOZSc&@&^N+~F~n z#S(V=bZ+-GfULce2NyJD5MG;rs?CSyK)0JUdN~z)LoFT=h(l-7yi@8v2h^;w=D>+E zyFUf70cbi#1Q`>C>`HoQO3(H7XjdmiGb%E^U`^h5v$|nWfm1MUVneMi)+a+{NkWzY zU{b^|XU8&$FGo8+1K!!lgMch=&ns!V&P(^pSp#D4T%Dx3p|0frn($%_7^}8 zse>c{9VbJ=f@gj_xYrrqnW7Flo48K!jG;g{1gjNFH7r-mo83anq8Jr!r;~N69!k1s zy{(Ex3drN>q{$8MKJ<)2FAdJ29LogdddZ=FnC5MLz2+2qThs1DA`q))0pxZ;Puwg-pbCB=hS!=UwNcKSlI*CT{ zahWJCXvmI@(nsGsmVJe7);R*;Kw3p|LY8QACfshSkXX%NuG-v_tXz!m%ixrUE6dX7 z+kLkYSAl#ctn)<9JB|hhPeQegJ;ZTw`SV7x_=S&oMrlciFC-sesvGq|DnD8*_Fj>?wsHF5ha z2h=NN*E%Xklgr)&stkR3=yc+vGoeF1DZ0Cy{M;;zeOjz4IlY620GEeM)dL?yLh_*m#GK6J!5eh5eo^( zE2{HLKyE9#dhGX?ZMM@UmEXh1N=!c#7GcjJLdBk3>`2jh?bSQ^$<>yr7KfLUecyvFlRAjcUIBe;rs znO2}tscy!D;vEWCX&!SQl{HIu*_}Po#}m->1i-r#taAi> z^b}tYW@i}S7RMo^`m*iR08bsaWqU_RRu3tu&H(?b!fy62w~`*t{?P!TW}px~z2Tm( z?KuG2cYFTsHJht9vLVG1Y}gsS1Kg^B@f&_FaAcTxG#I$nIm}$&EGu?Y+PU+WmD#Za zFKm0^!F0|2tZjmDUuk;AB6Q>>8==i`tXu<*H|{z6(@l=at70~dfGZYn_wPhtsMQwOVtOj*2c`wmu$YcVXb*(rZVzY| zRIg^ILC%-1U_-y7P>-VmCjlXG-(&Z_h%rnQ&zrcQMD5H2v-UNBs(mT~@dLi{;bdJG#s`1K1+qs? z_)Im59m_)Hw@1loet(~RkDq&Ge4O|9@pcs2>w0vh)omq6*#KlUsswHO{>~|;VjW%h z>=s0c7+n$sgsEYOPnT?A8AX%Nm^CBuI*0QEfWow>If5*^)p~{QJ83yrrb=gWR%0pT zq{%9hTc&re1H~{1%D|k=%vPWPLtlHw@!Y z+K=M62ecOD;29@rIH|oJ1;(B82Cu z4d>mcaz96aZD>PJWm-HxpCthJhnWEGh&cYF!}BZMoEmpp7Y4apmSam^`Ee*9 z`x*Gq1wjY;Ij?eQ(sCfL#>J~}h2NKePw8annoCXb7tE8^FnRM;aSx6T=CS}mH}1=d zXIsWQiCciVTdUI6J6kN=^9VR6F;6LcIaABIU}tk@e#k%!?3EUnaneNLhDntC(iF*R zJ#zhdf2)*@y1(Ya$7k$Eb>!N1;E;=9@XF+30fO|mEF}-_k#t*`T*j3TG{9DYjCWvb z>b=Y4h^=RE|B;Rc`!TteR`K$wm*s3%Qk&>%`+KvuVFpei*Z}O=y1JZ>$VFwH8K4ep z20Q_hirTwMsP(8Fnp_PtTzia7CYPLGN^1a7Y+z$lwomABqYKk&8-1oVA zzJ1P@V8Hq{{^*L5@3w#s{3$?#9^?T>z-@_9{()Ya>Ir zDJrs`fKN!Bvdl$~rl?zKH#n-pGi=duo+j+kVk~o8kgf=b^~qds&Ua6I&Ru>Kz-rpp zgCsrx)+4K{8gGU!@66jx7q}S4FSl5yz$fxO=a^*58}ajdE5!o1VZQSKEx+!vv%fhZ zdPc{>7C7EY;c7DMJ3e7g#x^>ZmkLvg`v!-62v1$&xX)Y+zb@TG-OhkB-|Y?O%Zcku z*u5QEhXs%bQmGg;u4@QYHWiF)P3cTnM8Yc{$&5!6{0 z{?ghWV0!1uvbAeSdIX5WYrqeaq;*I+OE8-z*=DHS&R*cYG4M*ow1dy5INn;N=ze+N zpn3!B{JP(z0{HtNXp2h%+h`ryx~zwFzlO06FlxtP_c()lz;ibuaL{s!PK;}yt(F#4 ze=Vmu?FUZVtLh?>Y`#37WxJgYGR)b%hH;`HONs?tp&ww9aOYs_Qg8(E8YaXBShFyW zbUfh9(Cib}-X5gIKWs0W)XD~AY-!b{i#PH{eV-C2!`Wf=`+jxG04qHe=lfm*Jd8zw z1#V6|Qy+N%x#9U3xL0w5hZ})%&F4KBe{C^R`@Ek8-*1Hp=%rO~lqEq2aRHEUX`>aS zk16NB z00<&oL%sF85AW+ zE_QG)c$kU@>QA(8U>tnkXe$v27X+v*tXkDYNrt}y{F$_v&%1~H9F#B*`~shUja93Y z1924~hR}ij4)Wx-4j0Q4afo-6yKoDqIw^&|%H=9X74SeMS;sRN4YcAYv! z4H3gLf4yB8r(Bb zm%%`z{!!OULUEgou)g7&+S*Ot;dYv;5tsmQ@Eo3Rh6kOvm#w`X%Y(2vX*QS$ndPD* zKn@fFxv{S6+wGMlajTYf48}EzLy#&3F`e=%UVPSnBqm9c!Dhgr&G@R$MYG!P>*F52 zJ9QudH_&Jx7TzR;X3!EKhF#X}m?DjL{V^-9aQ-nrJWB9+^)NIKg#otO)6_I0fGH9z zYaK@~_Q#6~)YiTpOW}@P!-x6y`GirKhS!tZtOFQ+pMqK8b@YP9Ei=vnR|wH>GMv77_n8V9~`# z6hWKGJHD{88|EJqk)>J@2RVSXD%-2a2)T(LAt|$LeZ0nNCf4<+-eo(bcG9kW{QSkh zK=3XVy~tA&F|5HqJYp#8TFl9)?520vx#+Nt%MCLR*myBZ7aEyWBA#AMgQQG#YmySt z^E8nWKD-^00o>TT8-1(A>|~nuV0PE3lkt3Aa=8}*6q2&>LguO7L*ijTVU@?ZX3o<( z599-3SO8vxxG4L0JlycDu-pp0bK=cB9uDS9}*u`VH&dlpKp$?Uzm*>N9L zeunmWzW(?WQnQ~za8I9%|G0MTG3yJGw{>(whV?{d>GR3Q`#z`XNAmg&VX=LyWYPJZ zdrAvkSjW!&(uv-b^XV+RRU2Jx!JDIMuVtQDx(WKB>`_Sv;9zpuPJztA`$+G%g=lEJ zPxm>>Mwb^Zz(X|(ea@LCgU{1daiUMB_k+~LY|hh)b+yma zii3Ph>M*)6B-KqVw^bvX89FK-kHAqttH6=Rw`&`nn!AZ1uiWv3!q&0<1Sa5D*47`7 z>Jb^wRG{;>ON9`PRuAjLmy^i%!rU%g?Ys$9-WOT}cO{ScbU~hOP&)3hrPo09EA*+= zjr3MYQeRSkV9U%|5nFR+iw#{9!E;36=~0VL>j9~dTk{Rw*X;$bR_D=#?OUVP6}dV( z&$-Fcm4ed<{9v0Fm92c84o6p^$4jHpm0}3?)M(XGo=>aY;JrQ`%@?z1r}d??bi6Lt z=b%2twFwD_FTc|A++0rhw@i&D@OWdnFN^kkf=Sl|*6m)E`#@>vIDz^I*SNH{{BQ!Q z?~muGVB`(fWu=zMJ7qN!03jG2XZGzEkq3 zd7n|+`W%_mt9GA`<)MM?c1A7({Wh0Ohw0pgMUBnuXLA_vIRFOBm2Vty-7>Sx2Yl|V zJ%;FmLfG>nK9S&SWH|BmlYB5|9(c0&0=(Za55@8wNMBZUbJyJJO-Q~s#N?|Y=$IfU zfHsBVOf0QF-HY`yT|fQ!m}cckeG!6|{&FY&&Ca_5q?WrMRg#U|?ibBW3_Lft+r38% zy7S)NUfwQGUJi7%6J*o3+af+^-{D&juYvx2Pi$ouZ)ctbt5;g4`|=^4eqNnaZqq3x z+#cnQH*4J0o>-iwl=$D~fwyeKY2%$9RQo)PkhLBKxA1HyGG~ECc-s zX){Jo#bD|&-0S%3#P0CURlDy)Nnyb8EItqx2>uCp(8|m9?9$TX?z1rov3whzaj(yA zhABEZDmxnWcz=V>TQ$KOCauhkd14sCl%NVyU>CK@?JXp`ECM^$u~USuboD5SA;R*w zHJqQiXT#T0XU_M0EpL-i{$pnvp|FW=;Ci<82@3V#!)n`P#mE!E{j}-*ZljHN?!}8( zNtmKuo&c!i&%GYIEnU&K6g4~$=QGfL4tUORUg8s(c4hyL@I$ik%g7p+c|MqHc8$Adpnd)q#YeDDCQ(?4~l3ZUX z-a2QcC&1Q)H_1HelNx3oG^KbRHRbaRA8;F2=7ymVy39liEMdZ zDhq1*jK9VwVB_PqvcR|O^hh}x9Dm{bIm2p~N!qmW^oAa5Sp@)e)l%FqfX=FryeRO} zeE8(=0{R>U7!(d*G2B~uHZi;#o65ZyL@b*jC&ZfA$$dVUyzc_|(0CB|^ zyL)97C4`daeWz~qSUrEJ%0;N%C|%E8jf{ts@-pX=O@ww^c3$(D@8|7wgt?fCG2?#NL9-CV+AG zl*R|F)I&sg_YtU85464FYh9nyYh`P-S-33z={a>ke$+g-;lsOkYNwtcEZg5(?pSia zy=O)5J9{n*N))PMbo-IE`2_vT^TRk!Lsr$@Mc00|@3WpLe@wHt{{j%>v=w`{eNjXj zE&*Gnn~LGAQMZj8V|7yGAp^RVj@cZ5nGTO5#r1X4Fz=OUmiY!qjqR>PyH z*y$1C0sQ5{BqN{%8}qeEqhI=SW1}9hiutgW2p*W1ouE0C|(Ow%qj+O^}L@!}!%q=Ti@Dqt=_GV2` zt)P6jt9nUa>dl2~pT4K_8L&k-y6Mq`Wi3}7pcNr>qc<HM10+6Q5-mEK(+)QKu!|u4GQ4( zOQ!t0+HXJn>l-A10AQs@W8!8-r>q1IfJE`|RfV{Z^0RVpc=YIx4_!GoR zPv<|d(El42UB1_UbO`;gShPkaCNyT&#x}oq(8B=I_}}esI|v1U0{ja4?Pa8dFQ7U) z(gH%VifVNK8ZNEQ?|uCP|F`k~8+wbG*Z42+2D+#QCc65%r_xd?r*Bn4<$OV7Fu=$W zu|e#35d&^vF+)M42?+({$i#X<357SK4d~yzObjFxstrP|(n}{_JNl=LISc+%E=YgM#ooxk#@^7(+Qi{^k@y-y8oxHb zm-d%F|Bm-6l5e?x;z>~h0Kd$cl(>+HjH1YI_$$_uhGyJC4XXEO zDL>208>--FxU45wN8;AC$jxz%M$EO?CilRE{gKRLbpiI(`=uxKtD@qK&iOW1u1JzZ z2nZuvTq_)zh#7a4yGHej>&>*vJ5#Eo+;?n7X{>3kJU+^ytpnCyt_ieT-LHn@BAh$N zTHWsKTUSO&t(e%HA1c<07Y9pi*PgDH=q$?Bw8CoHii)MaG=GL#AbDkV&8>lmdypE6EBK@7Lb%REFr50T%2D-d1_(e} zBnyfab%X?ep*>~3J(^Q4`@w8w!AaN8M{(^wSB|r6OS_CI>3*4g1TQRAVeaU9IzA+s zG%lZ7H*_&re#F;p?o?omy$<9@ zG%tWneal}Mv0fe*YB%ywGjc$HZK$Uo)g6Pv*I({Di zn)z=;nJz_X*}nPg8zcm^se$uY&Qyfpgw69CVh6>AU#rVfy^H(TGi)Pfl(3@7D2x09 zoK#tflR#tT9E}m9tXYl`ydCrKt}(QuS>NO<=|07Hkn}~YL59ZwtCyqWy&-Fv-v@k3 z8Ez#&kx@=q@@oZiiG<(!U?mO97ti(6Z-!8S;5}!E>;R(NrSt~hj35l_+c2s;$Vbrfja|hLZnr=)2q!k4)=C(e2iw?fwe^|<=k8h1TXKyPd(n8AVEBy;#kNC zdO5i$V_NK@QUKO8J{2k|KrG~U?0Pf7XOlW5f@FO_{6t7>WF6ud{1D*n2Ui2h_Ri#? z_aPt_u<8`12O=NaK&G-SJ?1dcK&yljF9bd9$8TVqy#o!+hn^YL*Cf=R04knxpz!O) zz3+TIQfvMAniINMp~7u)gm#?1grgq&jlUi5R14Dl+(?gp&82nBNDnZ9mjfeUDs~^+ zl!(-)j~Ye(*4EJzEKAVQ$T3Pr2UnD8k4S0g1)Y)<919x1SA49SpqP=dZrAqRJ@z?> zoi(EtPqgv|H|$2avruI}F{3aco(TL%u-Ejf@-WDG=7VtX0)Lfms$+$6Li-+;Z{JKs z4w1Ye3nSK*FlALVqtp4lW}v;!#;E%vJEZ=A@q-B1eIRT#^S)mP8SgiY5^iMZSh$5? zmG%=^N*zOS27-O;xd!O6qaJwO5Xet$Toesr4hp0BpjN`{-hx#A3EvlaG->R!{0pQO z44Q0EFe)JwxuvyRe#}Rxb59NpuDj#wVWNG9KpMjWY=AbGeTMHIUtVvB?(3a%JAfVX zU)+6<33J3;_l4TfN0nPMoXC00qirPznIOyB)(-3oQjQTRl>(uQZGgEJ!kV1Jhal3N z9IU+z-5Y}~gBqlI73)i)fY0J6hBKm#AiXpnJ}9R7&!-*!Qvvd*T zbYC^#4DS-SX`1m7UjFQg2HjGSvEl0Qh7!4ahk0#DLZmjVqH9BW{xNM-M)5XOr`I3^ z?vZ+{A3pSU4$tbnbLNn!05^z{Qs8E$V~ebWV7@mtp*PlI4YRHd?W~w<<)s0EI}bV^ zQE`XWO3wzQG3g93x+;E=NlFs+u)UIs9I1#gfgd6$CvpU_HeeG`1{$tZ6-!P_Aj2IwZi_xh7(T>N&xGh$j`s@2xQG1q{IswuV*VkG{g_ zB^OIrZNg_OL;C^cH_wIy%Hs{j>G0j@GKw6{Z6$v4F|=DaUe=jgaRgd(PQYgf$C2ut zBk*mL1IVBfpRT{aAj6yRH!z&mp|m2SCc{mTj<|tGn*l`e=qOpd^HVJ6+!UaNAh%mU z&@2qDG6Nz6*Q`Tz!geWk%w%6L6z8vAE4qzloVJ|mnB{{-Dy#|t&38W!ZFNPEVl~;; zjUj&?q*f(YRMz3n_W;849HeaB)ectyCzGZ{#k>NQ4FS7h4R|o z3Trp4t_AYY9eMvSsA)4U8qrL13VnLR`=uhj77C@oaTT>h_{_mkkaKUlDDEk1#C%xa zf4$i*oi#D{K@T`0gd0Kl<|yEpUZ&2&@#i&^9Yc;hSC|N=hY(A`b0;%IW7z$^y`qq)@ z0uYp{m_$`_MyH9UG9&C^b<42At~LU z-N#B%%-tj4e9j2Fr^K;Q`nq3O4qxSj!~mjJ2ZTN5?qO!M#nAV~93exNF3L2~$J6cI zxHahHcCxXX5&m98FfY;RP6qf8P&$N4(Un20iuu7FOZ^~lD_!r@tjcA3Ahg@iMI?_m zTl(`vUY_yUsW4Tn?2i8X&)kcgDchvrGS)?8%Ip=^sieqXd)SjBjr8;Fr=X>9&S_Y< zp2>VYrn^9wK6>tEhI85{y9Y)|L>c09VjuQ^ca0iSy(vdJ*{>9?K}_zSiFY>*67dlPu$d7)$IdGi zd1F_%@^ttgw~?z&^ykXmN=M927Qm}vx_*F$yetWohLkav-L51|@n7*|VOIp5Q}9WT zpgJmW>X2N}@O<+O?aa|d5?W}Du-K-s6Gfxq4{}b#LT3h}LK1Sf=psH1JB1no(Mx0g z%p6TeKGNJ(kH-IOS&lY%CG}XQMmO7Dho5l6x8(V97bv=mOT_fkxjT*Z@wvraii>{v zz4BwzM=g-0aZj+v&`J3OnJj44LQAP+#?b8?#?>j1%S z^sZ}cR@ZT;-dgB@utDPo;;~LU9`LgdXb3Gp!<%@<5{b#fc1*R5O_Ft>X8M@=@K{KG&;F0q>l z1wRZ}+2OrAq3MVRYw=s%b$;~prgxVafJ89zAH-!$So(*T6JpO(^X zmL9L!+7eAa1@W?wNtIR4e? zl7LE0Wuy8KstbVICldYJ)Kn#?FLGI#ovT2%>;s1xzny#h21FJdCRMGLFw=uvWasq0 zXs)WtZZNayy6qa=m+W>H{ZlxL?#u1T)<=JH!PPxi{1Y1VW33HWGI)+08W&fC;r19> zWS!Rgd#>_bQL5T?lQQh5!+@*-_Z+i2lsnkWbx{1!6E$t`Peg)1zp&A*flq^ARKLE1LyY#OD$nJMr?K0sWtLsX46XjslZbKZA!& z=iw943%*E6{-Bg6_)Z}>32Y%I)ePUr{F%HTK=avU6!#=;lz?!lKq?isIp3F`f<6D6 zF%fvVFUd|aNSJA%U*4=#ICHxTIEcI>g4pE;(aB2K7b4V19TKEmIF4YfUwcGpiTUDzh*YoJY7o&LjmU_2vLw>g{1mMz2>54lRWl+MQ*m3is7QSk z(k+gW5lfAI4Ag$Hx~^$Dn6msMR6$zwk0^zz#e-2Yn)^H<;*#2-T5$ zz%TOOG9#;BZqL-K{=)ZN$rLCGBS|FN9eG}Vp15md)N{JiK%`jBKAUtjG$QgDOTJJ~ zbioYG9dlShc9OnCxFiv)A($X;oj8~u`-ZN@(3V#4DpLW~d=_#249+Ye7nTrNUq9?r zRU#o(1})X}ru}`8|MqY?>0ov?1jfDJ-un`%FfQo%Y%vg2|-pgKmcgS&d!bl zrwu0?E%PCJZQPU|3%7YUV_qn!%P#^&n^=3S4{kW0{Ck#gu$?Bee7L#tJI9ndH#iHU zZ5lF!I;c}Ev-cjC!6cZ;-#X3?WIq`hVbw)%aC-#r>oeO~YwfdZ)#Y%GW_HF(iAgIQ zMdJc)PuIZWL*|;+0M&>en^RL$ElDZ$CGG~rSGe5Q z;BU;`3M7jc3K7(#8%~#dhN@DvBQ458s*7OPl83$%j3jkgu98yci1CMg3`XIsBSZQT zxJi`z;(p%PRvN?YxG0+)%;m0DGYe-l<-)!LR2pakJlIk3{ElK(M!d#PS z`1{XdX519|-+sw1jmUU6M9H*mcUS4uiCe`GBNJ(F$bXB^7iOYKc@t6>#wjY|D?z0> z;(?@&9;aSfd2Z&03VW2siY)(@xJl417Pc$K1&kS0_B0`-bWCztr(auD=m$7Z>9@P; zm5noMVnNBK$eVQ#4u#Mg1Dre3R3klVh{kNHr<{3fNZ9TBiXKQv zGPTs5FAInJyMp{Z;$ej*#^glMPDSPh1yrLm_^n&w{97NWd8zbHNYts-ZY9%$m7AGU zTVbcX;6Sr8=yB>a`t?Hcd_kdw3Z>$8)v9*rCZz}|2!#wPu;;pO=dYH94XDwcEmA+_ z`71^;?xcx(g#{_)v5(~{@3r0Qo>z#Nthy}R29aR04r_(>S5qdxkvi(ytysV#podsm zawJ14!ruBxv_n6dtCuoK@QKLt0xuM>0R*l;)#wxvhy{LZT&`>ZYKG$W)`PF;xZ%j| zKIp3cm`pi|&)*l>ab7%f?sf4_TS!xfg&n-elGljsc)530gG%VLxxu$1lY&MImf$rz z^k#UqSuu^h+Z_A(Owoa*IKJG1hrZ)tr;vV%IJv011v3{^6`x0XKip!8V9?6J8is6>^tMgHA0iJ1t*e^67r|@OPp<^$h;h9_OjiAy9 zT<%~ql9PgoxyBwDhir@#bzK99l`CV{jY-wUtE~;IbQQx~^%}3y%(S3EC6v^4t>G#< zrmu6yA6Lr05x~eN8AEC6e`w6ZArGie8$HT2*zwN^oX$n~K2A@$?80;lVw_--5akz< z)g~aV9yc*Hs!WFleWjmfi_V$Av8+O(Z8Btgq(AHU4F0~PCg?5Rfp4XrkxQ~pJf*t zmL#-_dXk8JdrCkQ@{wK7UatcF%d!gXm{k|Yn#5;ti1a}UW|Wy)xEKfx7%d0ou(1O+ z-A=(Dgfy`^{jDZ-XpyGr@z~PyIe?iCNHciiL{8icr<;1Hqy$>lIe0@A%Fi!Mbx_CL ztT)u{!idH+%`8}>An9f6I5C%~lJ)gqk-}W!M$$=myLm}0@0KR&>=MRGwl?v%p=y`b)EK^Ykl_sV2`%%fEi_Nl#rf|uFa(Q?dotRq) z$)If{qZ_P@f%`nPWppYgJ<#j*wPUWuT2V;O^y0EDzoVTvw293mY@92C~ zd*ene;K>4TNscL)*prj)8gJhz?xA63Oja87SxJ!UN2QIAuVMl5_cqrP<0v(^WQ?gu zF;c1qZ9@%ujp6bS3*|A!`|;t(zgCPgISY{6d?V1QNs8*b_AV!`;wqy`sNP2Sp?XHc zYU5Ii+!my@&hPw0#N;e73Tf*l(HlqmN$L#IkkHO3n(KboYOj8VUD0+_T7Au-%@O8P zl%m@$##z?+}2j3n1t4gCU8YMV`0syay`}Zns`QpsI z`%|T9P0Sq4Oss9}js8&Tv~$M4Q)ek9MO0%hLeGpI*j6}bgz}SRmBI*QI9S(pi;;a0B>gb>eo6>#zznm@ z@xHwwcjt}&Rx}mW1m5@Cl8k(62i1G&;cGAd*%|$%q~zeJ=V)f|dwJ@=dZRC(fWJDn zzj&iYu6kCsmPY?q-}dj&el->b_GY&KXGi(JfW1zWsgwSn<9zkF|IzDy1>36s_8;X= zxc{8<-;sF1nT%y$+Uw`v`tPqu6n`lY{*S^bor}Gmt*w#$zXYfCQXPD4+W*%oVHP|9 zfcaz>esBQ;IPMxln8nTa~Lj7qSt}`vvby zV-Rm30g9&cR5d9v<z) zR}a#(&Sh}+Co}ofP;g9JsRS$t`-qIZ#VMA)ug2z*s!iL*J9FoT+;a>o%(RReo|)Cg z(2L*UHG9-oOq@s^)@^i8^2U1J}kG8ZA>S(y*#BFf+o6|W^pm|tBqOqMZW-?kTA;JLx$Xz@SF zc)PfHaOV&^jlg&iL-!7~>xd;vee`pC>-H0M)yi{5)e0|{mCj=wkC;&F=hXK{*MJh1 z6m8`vm+*MfQ9f&7Yt7>g?}{r>cs+v!rO5Ex>hnZ|L{A<96CcgJ))OebxQ;VDck+tV zljscA;qdRJ%&omgchn{Tde$cI4V7-!82EmC`9{>a zZKp)e-M0a;loE14HTc|Ww{w6%)+-nYg)uz4lncph&_E~U@-voHtg0+s(vP(s8nLFKSF&C%5VYxMYpHYaR<6%;(P>!_Px z_noAYUE~gHb#8bi1@%2m9`tSN&-2*2d~|-xK0d88Haht;NvDU-jCHQp&4;Ez)hz%F z0C-)`e;RN0w5b0uL|A{b(rw)g^bAambaZ|ld}+Pigqsj+7RExlFZMn<@5o0dCYFZOF0Y^E3R08A z7m@$~8sn3nJQPN@-1=Qf$`AzZvE?E@huQawt6dT&w#4Yk-P1P*`5em6Mu1*K=n=D= z&oKHRX6-CsNRv`v!zGC*>fxg>f?2ybF?*jR}p|68CbT zp2TvNaw*vRVZ_(1t$OUfArs0MEd}`y3N`PHf1$_YC%)ih<$~yQ>m*91R62jSBb|cX z2}I7F^Ox^R5ql>Yn@1h72H(F9D(w%rk?a~7n)LIzH4V%DAlc+u)wMg0^t8MoazDa# z-|8=3*ksP*X3W<>w%wOrmZb+s5@SsVCGU!6+JU_EN*(a?_qxB`IK_WDZAu zq@xC3ON0zlyIYwj;xOrwhVES%0NpKuUufL&y z3zGy0$|v&rL?7!>uHmHz^7%Y}yLh=L*sd`X{70x1Pj1+h?+1bRoY?S{WRxp*&Uw|1 z4oEb1Hz_cOW{p&Y39z&(g_>bNtFa=G2EwG?DOJ_-h}9DZ+CER*iGJYsY(W8BY@;}< z0dpFsZquqV=4<5+iVUiZFl+5?xYm;Qtja|WX?r7=D2m+$3Q_d;GL?O!X4?13qT;#s zG*hQ6aP=q|^x*S#Ks`x_d?JM#Eah+R!S6nqsPV+p1WJYVlXm#XH*7jKtqM23y~5X5 zKSGY(P-plibxO5fdsrYmhC~(EN@2X&BwuBkbRHR@7HL4JAALjrF10V-=_*=hy!h+Y z44r?J5L%%LH2&nFdeT37++HQA&4?7l|dq`)yw-3E52RAo6R#Y%#cDt$L zX@!g7TO(S8R1;f&SG{Z|MoIiZa8gEz`NHy3x#IT_rTr-(p(@tP>+3NQh#6n=DkMwA zle4kkr>Zw!s%sm6JJ8L&g&yv zh18@kaTl2ojVlceh}LtdFzYB0E;(+{u#^gNSuyW|5RSu8EY?y-mX5}D9YgE-b_1}g zK6Po(C$BK)T4jG|w^0veaQlF+G(p{gEX&?qVF_C*gMns-rjPuZmvIYvzti?JHo3uE z0WA>$2ke5==>2^v(p0@pHZm7>8wPKRyk2WMc1uZrvY8Bgbp)S(Y>@g(d zSzQG8r*UH@I>8YoL7FXmDWFOq_B=Tp#7ulJXzaEi1C3$5ocvlu4MQAAcl9$NITUFF2%e=L!VfUP6P`{i;gUIeV3NxIsSmguY0{Jb8)#OS{u5LgL3?GQBFVgV5x8( zHOP%E4GKeqE`Cl-VS)AJ$)XBK$j>Vni)vI*Fps`di7tq#p5+ zpPWAyk%$E?1}@i@14|LPqA$RWE1kN|0xL1+aIr6qm%^2r_dKDwM`5i!xpbWKXsD@2 zBYI55RlJSZXEZRC4IX&@NXjf}mX(ZG)trf6RS87Jr;z(dD7$=KqGQF`7EOYku50x= z(*D>wC&S*rJKqIi3$T<+pV-$dqgB&#IrLEYsyTjQUV?L)&CIr2a&>w%YB}Rb1J`jh zs^zC?-+5#o^=Ro_#ECHLcFKeA@g>d7;6d^JgsgDkx_pllll!qo&;@g~C(kc$GnlxIRb1-~=XhJBoUqn4%kZwOlOmM|$bkkTn!2|8y{$T^I~tius;TKaYk^}4H+v)*G#BkeKJL$b zl@f3`3Ra8`SW0l=iY%Zp5=d=2RGnEE64f)=qqVbZj@OKz+U%dJ6E_DVqfr?AAinZ< z%gRYqJBatMyU@_&5t@OK9+z8=zbkB(#F@FdgSbOD!tJnm2i>5-u)lIrt|O>6GYxYT zf2IWlnU^U-U|CwkjcZnnQ~?jg8sv)9M5ROnC${=VDf1M-sj@?*v?PGEaw*fG-|DLZ z^-+^5sOFn^eDQ2-XN^i~j!H8zF@t3wwt1`(aP;hpma{)oq`^F5Jnn>WDb75PBm%pd zv6L*k&5sHA{YA0RocG5`wlg>k_PXWuCVmM?fzb_+r)j&&)$=<@^CzVGwX=w^1Z#xX zl#nCTm|FT0dpT_Wv&`w8?7Y!HMZ8a+D7zug34==P%yai3U#IlpEw82l2mtUpaeqIh z6E9&V@jp+gy^)=hnZ1$2f6VE~+25uqydnhTtsw9<}>eZg^)=T{eYrM@Cc}Jcet=lOnfm)Sw8cl zXP#D&*MY<1+S7!-4DEGze?RaSPZ>b+&jWX~vDL9OayGL3kCFd_HV;Zi%9ajoy%(pU zrQ8O6c_1WIm`s(eTwV{!uK680Lj{{K9MgOU)AF zPwd}omVcL0a8hCW`$P)lE94(ve~#d<38Y{6(!W3D&#>eFDaN9#rAJ z5QVueZ8oe5Q5+Q$1};k=7V=z&tGy+$iYMJ{b=lk@3+zF+V7|w@80%?_;SX9aLM*4a zbR5r+mUnE&q?b^TOY=X+6AHX*r!#S~_XFUHnSB+Y{8r*8ZuX@{p~4H~GO7w>UpcD* z{%Jf@`o2j9tSD&~ zqLAnsRu8vP?WojajGw->-@CA{3&-{PYnzXu`B%K8bI}Q|Epm%1DQL`|@4g3z5-V8K z>zDpea+{IoE}>XQa8>w5y6RM}HF*#^cm9KB#^Q!88i@m!UVnWZo+Q)CY_Lnzk#iTM zopb(PZ#AwLcG4gvr74=~1V^+K>qi!&cN^^2X%XnS6VGH*$@eu5Op;Q&cO8#tEnhvA zg}AX=5Su-&^>}!9n|%55o;wo~R9!LFJY3z3P+67=n4OCk1G^fKl$q)h5AJ?5&P@pD zE^{wy^i@^d2P$DyHMqaTiCCag6@np!NG(ttkROmWKDEVRCk z&0j|h;=fal>GV=_u`_j0fa4BC#LZ+~(35+zYJhKu*#ZcwaO9y(v2+C^70d|L(m;6n zX9Oz0P~4R8MHO_;UIw}BYUW$YiLxEWA?zArPhskT z3V9iWJG#c?6EYumzFYXvv6ObA_Sq5W9?d=%>_zBG4+-_*O@;i(?k5zEaT zEQX4k5Y0QuIt|M&2vo{fNd3}JzzL<%Nq!UmOJq#$(Mssy?@$i-CP~TRU@Y?iwG|>z zc^|?#V;ZtQUZyuhf5$m~z`|2gZkF`Bc?(DTVH!_Q_`_%e2i9w62a4C z=d4KimRMBpdOet9Ax$Q83p3Ky#rLw|xLM^CG}3L}T##-!W>LNmp-C%&&qm+i%za=t zVKgnjJCB1V*pz*Gp>qPL1X2tjMbNnIp{WRu3$=HcnA zB`MXbJgUpkr>UZCwuIdH4Ce#(x^Z3Hp2v{9)G=Ol@9(uT{-tpDi_7qbRvOq?>gntJ z>sMMcTQ_U{KPnslpp?Z*nz9S@DBdTfrxhXEGl5m`W?17P;$UWP#lUjRsEUYsIn0Q0 z7AkNhg6>yL$eaBD?NteVJJ)F^AMNop`}E94jKZpsvon22$c@_Tm^D_+CCAu1v9RyT zgJOi(yZ2t$&M}X87{+fR_noaBQ(CSX!1{gPgBYF+*qWu`l+Q?xh|09i7zZJtl*Xl9 zeXtTCHhC+;psC~vBHnKQ45M+hR0xIx`sMv#&n}iuk@%&?6bNa&3#LxFm3KGQ&zM>w zd){pA6Kp75O!^8@woIz=19eYk8=@D^xM}#y^`n#ix9XQ3c_Dg7tv|n>zPyQ#X^!zF zv-FjgT@tD^F%6=K!i#1$?i!n;8zxo)4bAa;wcJ@h^wyh6QdQ3A$3DLA{mfF=hIHs} zEp5wTOmwE_ermYpisZJ=IHX=x!lizB?nH4vhdj-1+ScY4BBTB_ZH_V}I zGD-gk={t?&n=b8TgW}fSyCHcFh^?RBO79r{P0- zKCMK1u4rDU&`}6_nFn*VXla+rm{|flgP>>SduDZnC6Cr69(I#dfsRPv5qKUe2YVVLBU;pO~)5^u?I%gDaer z+wFmDe3}AyO&OM#_7s0Q4EhO51)r5SH^e3nihPPf3mzm>`+#!Fe>aHoz3}ibvr$%K z=iLA*P96f);o;R35w-iXx z#}KHuxvs}+7tn9EZyYzdctv=6=9+qUN6+2@ZG&60;W55=>}D}}5Bs`Uz?#v}nqC&m z>yr5U#nSViCfwm)rK#UNE}a*L#mwHu+Ugf);D0gUQA*M_FDATsM>W#k4z0p^&@LAT zF)I|7;7xvE#iDBkLlj?Gn##ioCi%}p6(wqUg!0~t<&?9OFGlK*k@lasPD7@+V9Vj> z^IcKPoCKh&WI5R_#>kG}mMu*{cabs^Tu&Weab1+gv%ycDM{`>c7j*XO%T*s`-xsvY$sp)d?FEM;p+$mZXSUObQ4$C>?u7D zUtNjdX1ir_5s@+>m_7hWU7nqmUDy&#ZJ01!^`nIiRkM7};I8{>@58C9)rQ5aikd22 zsWt2SM>a)y^}8ccs^%D6Q?$@Ix)YEbvG+EYo~jf^vdY}Ns6s+-PRWSvsQP9$SJy*_ z!9=B|;m#HhUi<^BSiJ({P?4U`ev&8Sh2}EV;ZoMA9l=3QV5mr&epaj&!37swVB_Td z{ca53G604t!`HMf2AE2~~hPrwgU8 zzaxz=jK@}fv!!xFJmuZwSo*< zzMh)xq4A|8S^jCJ5!2Hs5bHFfTN?dJT@sV$NU6{sco1R=RsoM z8t55Yw}ZUcG7Sjaz!C~2B$xD)s>k27W1WH0F#E;zeAR*9wc|D1`fGgkOJ(@qRd5#@ zdkbSr8<$@y_5W)<{5354)uUy7dFU@;)+;QXfsM7XnaO{16{KTEVLIsHdPWZvBXLoS zP1?yZ@hWHz!mXHMl}?YMunT(eS28?v@WalIv3kCeyg#!b8_AS43{#rd^spPoK{iev z6RXd%fLMJy6w^f^@B0N~r$Hu@LPry>d;UX=jy+4BWo=nJ`?J%vB{|AMG<&cb=YkSP zQbR$B{S~jU*1l~`m5-dX+qh#|^Q+DND<9VSvh-7CW2M`|hx6O_WglzQ9jSPMz zG$2#L{e9^CdItQjGDm-T-Y;B>-w?%K=)ZqR{6LiF?`S7f64I3l6cW-@GoxbCQj{~J z($uu#H0!_T7^ut~@-bh4U%~%*Ab+8|(EWMF`!%5cUq|tR3&{IRTH&v7mNx%BAN{I? zuW(5(pMP7o{HsvJf9v}n2jhPeilNc(M$pX0`VZOsgX0Q2JSaINEiF$WLDNS!LfubS zSfwH+-A^q~(ced{BKC3+Tv<+5oJNLrfSRIzXjD;^qF)uT2N=jFGMo5UmWC|?5})s- z7q30}X9@mwv6$M}Sp2g1{;LT8x~BgsLVF|QU*Hm6#P^TfmRAw}%5BlW_#b?xmkBmB za{Wzsuf&$LmlO6v|C8VnH3zLGr3D6gsyIQ$i1ao7yD6cSLd|tWl>*^6hG7N8$)Us~ zm2fnV@lY|1O9`Y{BFSALadB~!pW0QKUodEDY&abT7d5$UMmOr&93GxVc^)uBjPR}j z0K4|qJEh!H!6v_AWM%VCZN&-SAPfL9sXprw1i>~tK;N^SDGns&(HyCJ+Ib-+=8VQG z^(p&$8qB_A`dGK&8($PQ;<^b+k0GS|-VlBCJ1T$dHp3y~cj5xVKH+-8^tf&KiA)D6 zR72l{zY=nIjByy$1V>~v6xYOD8Lic>?{6Tt$Gz z#u4=GJ*I)}t_y!HLQv1JR_@Z7w_8;WlSVg_;ETZ>`gX^ZXbA%#N{_t3-W`{iCD0~n zScIU}jQHU;ZnvU(efsGo6YOQej4Q4Tsz5tcnZE011TDHXffUzK z=XgWkq>71?PFRL@O?0B$K2*$^ZIzR)6EhqcEbh|P)tFxY9kC2jUb8|8@HC!96$Gh@ z1vV-AgCnYt4&UcDh5wJTuYig(*|r7}+=Dv=cXx;2?(S~E-GaNjySqCSCzYkWgZs@h@D>+ri_C6JFqfZwmy@4K+?2wBG{O}{XKUDWVq`m5t$ZFrZ z-5v76b>B`rztl?ijLjx0qsW^ba3ya`$&5~&2caz1tr-kWfE1CfSq`#qHXdDbV0rLb zr*yYvsY7FAs3RZu=Gnmg7L^sI>Qo1DS%`Q&{oZ1dS8aybFt~UxPud_pqo2?|zsi<) z`Fkqo;3)7-T)wugx(pPGt@s(DOpFiZ9^(?^##&c{jv#TqA@ARUbqPUy`m=2z{d0QY z5>?E5^-`h_cqU+9w8?F3L0Cbjv-m6?kLs`BahEvHCV8=TryZ`Wd`aOYg*2fELp7uQnznBCfn7QmQ|9wxF@DV@%rX+zP#ec=|h#3X9lG zmSUg`-ksp(uvxUV40!unwhke?$egDVqOZPQzenYu* z!S!-WTY_@}`)3K~pPs$xw)nPGUTneavnLzBhRIx-m0p%m?a_ZJEFRZqc!u0YfWkom zeu)GP${uHYM!Sg0^F+dF0fx-XyjB=)?!QYinBhzv-hT1*0brT|$qpCX5!buVK)XbD z9fEe!x~qa`K8Mgdei?S2X9jTty!bn>4i1iCK#a2NmQX2!MzxT>X~Brpv?CYdzU<(% zefaDxfRXl<&yKd3jSFn{kr@?5cDfEv7BK|t?q=D`e$`tmb8n*^;T-Yw+1}B6tgCEX zWpV4jc~0M7cp8~C;*H9D?%SYlePP#Qq9NCvxjk!i4tk9#@(lKdI@#(SntRF|+WgER z?e&!kzt&`!O^{lUFq=1&wjCiE2FmA~VV#*FiF-31Q%t~>zT0Q^SbNMX+(69N4{`)4 zWrSzSp4L)>5>5IFGllQCC$VYLL<^=w{ksqXb#e3==&rW1vlbRP9fWpo828heM;nfG z3>!oT+!Zs9an9P;uH=0lpm6D4E#Pi>125r*izu18+K)?2V%mM*>LZIob^~n8u45Qs z%bVQfk+0t#lE2I>vDeey5>nfSg!t5py_8a~O*8eRJyP-piscRzO_28FBC+{KU>h}S z9Ws{ZcnjE?uF!mXLZZaen&}s#l_)4LD_f#IQZCZZZC%AAi@F}N)JJ!?XwKMVwciAt zb9o?L-V1iet)s<$;<(9w2o=%ro3Tq9@1xkEZFtg=;dS#uZk6PP3&0WogyvQ3b$Q3A z`*5}58=bB%`85@#j>1VBaOd0IGDSsaTzfkE<7GdAE@muJKS`}BVWPU;OPbfKvoOy= z=lUWSt>O{SrVif|6ty!Zv17#AuKLU@_Ho&(^j%b_wlie0?u%~W7c@AC7|+3b9bOB) z2WZVz{FkEl3a^t-QDb->n7!UJwqBslh`Y#-3m?yHAu|Y#FX&@iwC5T$-rlhu9pO~j zo7Ap+x9M7^Wyhw8878PE_wAN0td={X(_Z8Hu#0j7H~OSPcY=pP($aCP_sZ^`Xtx|I zQJ%Cak{r491up1YVNEkOVK<9x^~$z9gl{Gv1WB#m;0PU;QGM%xfsHBdnV1r7V^d5XROFSXft1eW;sNs8qyN2c>#%vu1pTPTq z{R{fr^UfXhEY>}qr3IwqZ;aZ}vh||c39$L>$e$rBTyz3dlwf_vDiZa-wfImJeN)Lg zEO`X8i^H<~4)QoYV|a-|x;PZ)^IeyCT2%OzsWgVWRwE6^oQmy-tlS=k(e%lL_i|#$ zy%Cwt?oNHYFn85A%@j_nrUKVth1i9{%~@(bv6-N!56}Ny})caT3Piiabxs6V~`aZs#v_7vc2BYFUnzPH~yfHd}ZLWQKkhkJY9JjdX zYmwjMX;>FevheY(!k9#pyQ$TzCbvO2h7>MCYdmgV7LMtuARpBStYdCE#tmuP4-Z{jw3TQR8IaOt|ZUv1qQ zh9ir5!M)LH8EVJ;F^MnfRgZF0IeX=Vx6ZeG(M(Ha6GTNW38g|tlfsIImOj)X&`HOc zpLHZXqS;xt^8UG2tW?@aB*Sjlxt(oRcT@t;`VLTv*EFnIyXVI&q?)Bv_#`Z}eSIuT zZ+bD=hggM>?}*d#lhd!*ld>O@8uvQ_qb#oJP2&BmgoXL_}mIF0}C zd83(Sjj~JNQYd!wy|$jRijWjnGRY>~>IYd7ff!fW-S^i`JL-h5p7p$mwl+l%QSpo+ z6yrKNO5+c5oF_K!8QikC8}N9G{ySqndkgHG9|s6NQ!gfzt<0;GJ!(@=F&X^jdM_K5E_Y~75cc3&D)rMi9XB1?k zik@_C1WN1`g8izdr{NQ7r?)_fQ0}usPsEy z9mj>WyhG2$7PwgTR8N_UoK!)YmDw~OC>9ZyeY(C!B=Vk%E>3Tx3%nf9(P=eSbE^W2 z_JUhmBibUzsZuM4AL=5V;I%g*C(i-A+PIOCK*1_LEvPQ`YNsGI+)j3zeW^I%YYV`l)#X65rm$emwr^ZA;C~6R3RU@t!{+nw1m@KJm4&%~Jn7_MQlp0G z&l>l-Zw;JbK|O zKvT?{3l8RHuaCZsad0agjZMIS;JK_@*L>U`ih(3<8*glIJpCABF5wo~#^$lvLb!!e z6iwgsMfcoO>BnwrgQIkq5E!G!?r9-cK~q>{5Jy6JewEThUE00s1L5I3Ox9xN#3<#d zXF^OVUk=M6=;hrfmWdit*8BBV&$!!x_09Wf<_LASy^%3RdA4d7Q$NaLMx+I|bjhG1 z{ZI`*+OSiLhw^D@A(gYcel*M-&^}u%o&)z%m-w>lS#s5#lGQu3N@_nT^N$uORA7zu zlCSd&{Tdk^K6eJ2J7HzuXlV1RBE?VN$TYuhM=O*Spl+~zpjkcf*oL+R{`^sb`<7VS zIY>dG2Y0Rkr?08X_R@L*-dhfC%M}y)s}YPM@7cB(N0N67LfS)hIoiTh&p2_@OoPQT z8@WC4`XNbHNFu}4OzBlnr#Yzodw59|9knUTuB<~R5%#4aMh=Z(`AaXe+JH`yxa`cW zg~L|L$oWUJ)`*w4-vAGxRQkTM0I9gZUH{;W9o3Bpp$5rJAn(k!=(vwj|#%ZSVPu~*=SBOnZ zGI_4Ynmjz+w_l!}+S)a}(sfQ=2Q^BlZ9*j>EQaaWqZO+uT0nt=-Gv~Yi!Q2@o*vM+ zdKppQSwsjfXPCd%A7P$0Ha+=j2DS!%?x&=ae=EBCu+!}}C-VXQsWdq5fUg-wWEu$xV)#fFK z0J(u^7Q(_2Jw19v26)bs0Bly{C8FuaP9nkB$VJUywj?{G26f)JyyxF2JI$Ya^P&KO zq@OGM=bGbG|4BdDLHfNg-oNVM{fpo62WiC08o=|gw)=Nncj=13_$RVt1MvQrAPN9Z z{TmDXRL9mvPV~EGg-{J`KNmjqhwR=dWT`Ke5FkE@rfFfH+=z?@$R5|6MZ{~#T4F-PL$hLBxLfJ%vw~ot<*Q>GEqxuQ%tWd zTQsiNt(@pkl9x|)!Dy~Ya%uI|@|dZ+VgfEzCpSo8PT6gl0=3K1?09%#aKC%KJRHTV z{vvACVhG~;Dema1$-0up`Yf`|a}f_;oRXw36bnm=B(B}Czn-ISj&;Zb`tz>KM4dj5(J!9mOi_5ub2 zqUr42(~M!QrrEei+{wlmEr@HFL+9jfX=F_!c1npxs)x;rSPwAD0{hNp{eo_^Q7Nqk z>x})~y1Aaq@>PnEF_z`VAagmUb_ykzMsDay+NOK}>@~?LZO?R*D|)&Fx~7IBrI$TB zNLzJuf*;*pYZ!ObtCtzPtV?7~kN_4vP*~8Tu0cRdd9}(sWqJ;SjT2W^x$V7EFO}-; zAY$ru|ApK+sVwDPPg75JNtYpTY2y+mVUKQ!;t^8$X7VXcesS z+(r@JO+(zey`RjE@_YBq5Sn8f1a`{s<2k8bx{wwTlrcs#vLo0vNE25!D6#Xl$AGOF zvFugFF6D)P=t;mzSN-s?L#jWA3$l}V`(7fm97I5!yuC8ENAM%X$i&&lwe0L#f8wCf zBS%V=ADBdIoNzuf1nry;c&5D})@6Q8;DQ4f;tqf=d zc^zi>?3TzHvQN?cPTN#*ERq*4+l8D|;XQc8YS6#Cs7C9$6~V*Bl~&M5W3(Wt4~2%- z4RnKf5}|w|A^}0GIDj-Zl(8}&sedOBDE1CqiLexkD|{$GR}fqG*vXgwQ&$Du!vwb? z*g2VzVt_m}Ola(&pMF;;P8_>)P0uMaEMa)5@gQ!SqUNwj$t^<&2S}tEC$AJP8x2Zd z{+W_I>Qe`Zn>wv3QvyS8EBWQfS*k~GSnGBNU3d#E zYG<*RpQnNyw+WJ+L$tZ**7-KRXtI=Q=ZZ&y_S72NV{mi1In^hPihU~OR0Gbd4)m?P zqu1u3k@Rz!ti5|)aoex@PU)Jr)i~rV=>)Oe9t@X!%{DOejLq}s&oSnvRYKyCA-+TD zZH>L#z!Kq_cX`K)3j4Ss>CDNa#jBv`80;&~Nzo!2HMNaB)BZ{}VUo2MtK_#4Uc^Zk z_k18{*yZ>_sAz^1B4c3S*R zaZ{1$bdQ&tNUX|@ZQlvP4z$tpf;Ca`ikSnG)bS%u<~Mq<8mXc|_>I-cn6OHcx`LcN za%65l_+|D)b!U5fSALZwW0#M2h*XZmM5OKXyrB@&20loUkPY~$mx6pJr0`ujQOfx<78s-o5hH=b9h{)h>tfx(tTmYozehKgwgG_F6!Dfj4WL3Gev;gZh4SJ zCv{%dekW$hngjHI1aUS&Uu>@ZnS75pa71@;GjkVNcs65807mq$N zHRCXX6xjmj!Afv3ST@Uk3361RV|CZ+IvTP_LO$Se;OZmQ*dqPW=m#+ z0^XsqDw3cO@U&+{a%!TsL(*cvyY* zuF4|&mUskS1ygZlJiVS<5~6t#Lqx;HbJ^-BkWBKZ{Jb?3C6+N882pLjsTMH%XS{+4 zwrQqSMQ~zo6a*HCO$>_GKZ0{Pue&>2GuK(;I4a!1JeaZ_0$VE&33E%V22w%m4AHQs z-XVi2clIBSk8tndK%c8&a>BH@!X$GLC2c1z%m#OlJLA=^$Avu-!IRqwvna%Ej%@ag*(&Fvv6t`Lnx|bURU;2Bw+2y0$kvP4V=;l`y8oFzah=%?}LH zMoVyq>#PrqGCibt8<%K${nb<7%rT9NgpG(Za@51k>_Ax2_>Tq%e4UdeKSxJ-VA-r% z(@D6Cmu;VAVvXRJTi_m9f`q#3sWXy($F^9tk`s8dt;=aV; zlR**_LKu`Z6rGZz)CLo8mDpoykKRkc18<8Cxi_#}4m9Pxxz$}p%BJ?$Jn&luUZ79Y zQ+1`m{umtn)E`TpijXkaMrkh0PAdq&t$kArH4kJa=r^0!uMW*$n4InO(3;u4KHj0@ zbZ0;6u!-nTBxsrZaT6huY<(-rXkAVd$#lJav%a?aO7nEk)HOt1SzH>e`DL>_6qU|4PZ*^wBj%?awDu48n>OaY-}XovP|*WQtJuX1k+!lx1ZiVK9u zKoJROU0ugYM&si7JGwH*?rR8%WLVAFZ+$|C>9{cyrZ*PpFQc``Z73YGF z>wq|}P3J!dXg;$YrD=rF%36tqQwuuQ`i&|cGRozPjnl-^b<}z<3x8RhB zb*q6J8?!RH)22k5C>Mvah@@G`<0BA7f5N2DlSuaTe+y{9eb0*u1W0zQ`Dl1TXu;|w zdduX+CX1QVRaIDfw0GugR3tR$1(4MCQlB1{f;`utnz8shwHZmp_+K+4jwvBP?}#S5 z9bo<#s#bCb?Z}nn4DUL7U`J^*b7{SOCw!Tn(M?c+AX6!C1GHDxZ;QzWXOoS?AIa03 zLt>;RHW&yxHF+*g*LwS*Yy~GZ*mL&$7@a1?c()5=6|o3^UE?D}pjO4%%*`B3SaE5f z#hEZ&Ia^8vBeuYalE8&{ItV*6dfz#-HiVE8gT?M7^Klq}N?;5_OSZZ3K|eQMHuZ zM1>$Jsd&X!PMP(UY~PP0@490sCC!}ljzc1z7$2-+d{2}$9G*5tk47#oHH=LEirPE# zK6mE5sSYhAeS>3?ZDWTr);s%^t*?%|Q6{`kCAi`Hucz&f!*ZJ>P$r$Jz9om%i^pq7 zUq~T-m`w1y=?u3)StAPuOxlfqDlo|vkNn8Uw}3g4F>b}R9iz2>ev~)yDrP-`wuiWc+#E(_{3NIeys}C@6^D@Gh;g|P0 z=DumoF8Lho!?33|egRiP>QEw~9x-dX!#PdgmlXrjrrf2EUf$hp72H+Df9sRy3%u~w zW$c!4d&=og=IAPH%5*bL!dlo64z)M>T-zh{S!cfU{L6#eZQ26E9b<qXlm@_|h>!0Qf%P%bPLje8j|EzWC{?NMY0M_`=0K(s}$E_4ct(O%M zyw56}=bjQ)PCoPG1Wd`Rn%0`}V=F5u%@m6gXA6Wym!odocd)JA8$2DaIA!d{YlnS{ zp$H=R08X>)e)zVbu@UbLOdf#Yj+Fj|cszXVd6$*(Y_v|NECG^>_eE86tqSpyl>H|A z4y?T!gwP*J8Yef$7|t=JpKOtxbdMrcAOxTGGPhc!;iHPnD>-6$RVUd=3`(tFLoHcV zG`LB9sH_XW;^Pmso-j?J;X-_iukzY%iFw-_uJnDw?fNx1P>3&2l4TaK+MrK?v;O;w zVT`#nsh2ESmt>@MhEbQTrssjCm%E>7jDVL@AWwr}D*K9n7QnWzEA7J5Q}F$e1{bCW zgIMpe6(^_E5g6g=E4V2Zb2|ldl0YXBs9$Gk(sZ4Hu5N4YsCY>+Z{>ox&pUe8cJ{Jr zh~C_v1|XvXu75kjIM;vxk{jV3{@olNKhI~H&=){m#_9IL)S5<2+upanZX zcv_Nvu^>s$KIbrPcRwZ|PsXW>Z`{pbzy$h3O`3D#!cs77F7MKjnt2lW#f|xOSnTq6m24P~=;QhAgczumxGdmX)nfGKdaXqMN|4GyL6|D6D zTrZs!XKgv0!WnNNr^OcKAn$?7KAO#~WX!qjI6l^1NK z6>uxe1W$o2kZnj>gYxWet!@~qAMg@BKc1!{(nJD^Z(EWKcg#Q!RUdCl=6=Unt5Xi_ z=}xv^NCswefXiG(%#S;&_Tvg{lvT8SXY(!Y<|B<%o+&PS-YyP#GveciYZr^Bi`hoY zEjL_eQ}$Mek1Ov>{pe4P-fc7wC)T&%TB6gbz+xBVCawF%ABu~_FeamT+Mx=au#Nwa z*49rb0#&}uYHp6KZ#{~7PxR$MoSg?a@=XkXFWP{BhNIJHAroaU8z&?WTLzCc@f9}T zXOGH$pHl6kW&lJpVQJ+S_2fH4Y~>X&x5Ztmgq{jxs1dT+(MuhG6?>eoD3E#akfTTK0-^QW`4ihp&e3GwsW@ z?LB!MzBw)rt+PWUD5@nSs%glP=iN)o-BfESV&qxfQ%v^|4Cg70svhR3>(WcsB!!LE2??kReE4u^hk)QzYRTA_a?+tJ!fY55@*DLGP)Or9^4y<3&(BIc|5%OZr%&})eAfT2(eZoB)>is@4*wMw z?e7_Rzp(xO1olt7JoaBS$j>qTf*bwyS&j~7mjA+<{VXxU z2YB)RU&@U9!i4&@SqD2kV`DP|9dpOe07jJc=ievk|66zDXGj0q@&8qKg!W&P^rzSP zJGSj#UH^~W|Le;BL*tf?R{BPE=72=ZFF3saSlRVw&;Q!^&)aYQXORE5p4-?N8R`Kd z{5pUSu766;Pn6_8!ue+1x) zBpN3(LnCV%IvRF%CPtRu|DyZthsMgtL664V{$D^-55U0wfzn&13_zGtjmd~bj7!SZ z%F|8A$%>DSQ;LmJhR~K!iGo6rCm0px?;h=;CX)dZE7(JeQ~2Z(3l)?!{C1obYr5xh z0($+k=YN`fzm91G@Qv{QVN6zfpUsQ`b%_5M)*t`)Z-x^QIr$UCE#z&qttL1@Frvb1 z5H3HrUs_Jxh-`SgCPf(PiDXvw}1!G@F&)=eG>Dz$3JYe)cXW;MWCJ^wA=>K7EjBL#8tqqNI07tcf=^qpH zlS^#>UoQ6d5kS08awLRSoLZVXgm!2|ike1jgjyD`@(e7DOicANjE!_POm$4G)c{HG z9Y%YH@@W7>NFJIY>`Fm?%U=F!N`isTKTLiYDNLTuUxET!zS~!vSiI(Uaqxv>gT}z9P+OL%^861`z3Jr3$Oa;M;)7g@A!Y`41aP#{y}sLp#dLhIGho5Z&iX4d;r87oVk*wF?#Zu~FKDKkELICa^ ze5k|NIRW)hy(7_P_E0(%MS`5R|9lK~M&5cvx3IogKc}cdb5>t9=&?$aF68tT zDVe)9P6HV9ScKT61k4=JW7nFj6xV*LaVRPSS7ngN}l=7b>n~YKpZhmMl*diFendk! zMh@ealzE8lmai`Q7nun^o?U@C$XvFmq-5LWt~M1FN1O_u!kvPcGLGl*oMApsmiGki zX)e;`NllmP1APeHMf5amDIau_NGuc|qA)%*)x8o_#_k3hf@Y?At<1%q4QVt-nWvS0 z7B`JWFiO8CQ+Dsa8;n}ViT4d?vKi`mdZy<}@tpUCj!GovUgSr})^rb`X<$ep{M4dY=MOD*MH&NMyheMGztO%z0!gbnd0*o+T&b3H5&>=03I zSS#oei1B33ef+54?48_H2za{v^2%-o2k^wDCFcrNm4#YG6#auPKct$bzg8;S_ewQq zVI6B{T7g;5kHgL}yFGp`ph9YCAgHDcLj|orfmM66!GA~pjgufYm8I3=6Gh2YwG_6z ziS*pYQfA0mY+{sTQn5wyd#JLT*2tsSRJU_T_V;I8Ymh4;2}k{9(+2EPp-Voj!&Fb7 zfXH_QJ}5nXBh=A`MwLhe;&s`k_2HUi5LkS9=AM8a=7=8P#cs{=#r)191!n9lI zQLA^RGI#$mC7o&aLI1s^oE;jv8Gd0WhMVU^@irzCA^&0A7M)Z$bq1Mx%^>lPt5NTa zry6rA%Aw?K9lOmI*s9R;vqRHth#V`SS|}6fZq&K`%)Og}YBBAX-DulA&tzSfuGcTq zW#o4gmg$P><=-)jZFY;_00Y~wKDG79#^L?jtG>N(V9~?=wd70hq0J(Q#ZE20B9+b`RlpM?7Q>` zhb8BN-SEy*CwYt=4+4}#uu5x`@BHoqYh9vkiSo2uzkH?lOLsH)_I%Z*0(#}%~IDn!*KabqMSM=jBKtRNQ+BSbZ zjDK-U{)_XtPr3W^GRtpex5*Gfia#EM?Gq1zb<_*_;My&Zj3Yfsr+5c|DE0NLNF-<`KP;TzHL^fsLM!x3?Jc*qk=` z^_Lh6h4)kJM78s3NtdV%&ecoZZ4EUu1x>--%^mfDaXTG7;7)^$?3QT7s5msbe%_?+ zt(wSRSbS4*^pBp?(?7F-A)|$`MR}^F@ViSJA}wjmEMN>nI!$JQo0=g7pwD~fLo2HhLz8I)*MW8YJ36|$+>(*D_gf-Kg&E0-i@ry$C z;O&2jndxRAwgAEuLTY4-A--7)xi$HcIbcnCqJ`LjP^+kn-8f?g_r!hEvulF@#2Z7m znu;rIp~y4ElSCWzps6^RtYN<;o*`q^LwRcXTIFC+XcwSc#5RO#Q z9~f5oAmFuZ_EEg_WgDH{_Z|A{1;pI zL`BYei46vDN5&b!IW2OSiv-A~;?CC+zlx1^ID z+Sihz;MXj)D4sKuo5*YP-fQWMerLu(00pcrY_k4?12FFGRaATW zVBhC3GthEpj#Aw!c>L-g!(!iDjb(i1R;Py4*DvvS;n?vwlUkH8XZ^cdI~3ggg|2OG zy5Mq-bnFNv9pRkvu?uRMB;ywc3X=4Ws90>~vP62`xVmT&@pPO%NAtG4#XaeyK79`c z35$(&jZCUof?I+xUus{~CFFzcmYSC2;?&=&&cBd?U{&ZRj<9F~HvO=y`nA8G{r#g% zZIXw38_y4}29PViLbqP^Ol|7YZPw(B!t=h5@F zkEGY*jz>u@aP+2Bk(qTv-%AGq$9gl{m(E@ir&Ov%mdeeF_O-OMy1%-@p^~|Dflx9F zEnBv9b=DvY>)z=Au9)scHEi}zVFk%a`|%w>L)v+NKMvyZKAg7u8Wae7ExqZ-xC6Y` zcEu971>@ui0hn<0b{3HoLeCyyvpYDNe&q+$_M}QCc^Vs?O18ENME%`s&6Tb(=bPV+D z9Bi!rt5OmL@|-e&@WJX8W&00AVhW1_QBl#6VR&Z7%ZF$`9uL*{02Tm24E(tAh>~t`~cFLJ=lsBlx%`MRnamJG{!gRQ5 zV}nCcS7XN25iV-OzmnQ)^pUl%QGX1#ga4o;6^yH<_}yjOHJ(^~3Ppan%iwqre$}8} zWJtXW%<;UOYM~c>b7Bi!aA-xtYmC5ykCLM9hmR>L+J!C%nUAuYf`Q?xUm3Z5gKmVa zBwb0$yAFPy#`r4QV0pe7VFQMCH?W-x@W`vw_e{Vm>tK$adE&A9NC6*K;dZ3TXd0wz zJ1w4AOQOw?(8>D5UQ8|H-9_RK*S@7Y`U;H4Ex967Dd5Cs66HWmnQl-ks=Am~gdWAf zBOA(V)4P6nlV1_991^a=u3CVXm~_o*$MBOENT=R*$=_v;Jy};=H%+E@vg}Lf>i|Ma zm;}zhm}Wg`MKoyLb5_0fY8Cuh5h>_R{`Xb|HG31JNW!F z_RQ+S`d{dx21Ti#)k8X-R7OTw9Hmg9prHdZ-A?ta34%+wrG<=!MD3jFzIGD!>7+49 zOv!6enHjITyWxsC0qn=y0h*X1Gwaz&rk<8w44vmjqInKf$W{fM2ji;J6! zE4|(Vf;TcT<{a}<8Ve(Aqlc53T%7pUq_+*Q*DrV-vFeZE287@oB9`i+vmkf*9e$<0 zv_~H#B4?Qi9_o{_3?TMO&>Ar6;px)CnptrZvb8y;{nsvi8zEDcVC(yRfl9h_R}~r) zB_bMyt?F)8Z^J#S z$Z_n}_mVJN+smv@&c@&EzU&(?M=?dBiPQ!!X;mv(}BC3f{JlJt6Z}Vc`d6ao-+7+@LY{u8t^8yyc;D&R zisBJfgz7bE`?OwF>(7H7>$$70atZ1ZEgBQX_>v9ShP|G(5t6IL^2V}B&q6`!E3}`r zt>IwA?39w})eO1yc*U1dtpQErUzwAy#cL*YxaKh2v4^JYVwT>O>tfJCU!(-lp*ioA zuFG}rb!eZVwS5I}eR0atW!fc)%i`VnX*cZGFIg}jx|_l8*U^%rd0xU-Y}STV{dM|P zK-0#Bp7+H23uNsaTsis7Kh8sBl6+E*x!Y1SI|Yx5=E3HYezrN6eb5-c$L+ils4Md- zgtdO?hvy&m>-<5Jtfm!l6o5=z!5NVvl)1+9Y zHmVGmgvoa^TL*>^Ej4Hj9LDa&T#!kU@X0-at8`WF7pUv`=4id!0%f(#Un%p$D7}o~@m;M#K&)c|ZF}P)VA8WpJv>R4FO4lg91l>2 z6w)0H-S*klL?ma6eb>2yV(OVh7LqMu=Vel|3vZuM`2o8C;hidN3pKnw=p;Co56p}06J)mja>i`Yhz=wM$gHZFw|^_RPrB9=q;FlGh}Sre3`p_ohN7>{J6 zQ7;825OT0YEp3yN4lxUeS&10CYy{nw2ci}7S>0Dmg0I?P0!iC3FcFOe;?+35jaV2X zpJ=Dp!(4UhsAOuaK$)SZ#IYyLm+joF#uP_HBg7fJPMDEuE_?FB+-HiF7no7AQ`58- ze-H{ZLkFMe1v^|^B0lV%K-5;7J?gyUt}xb9snSB`e$U+i-HLAiEfQ(CpR-)d#hl#j zX=SSVFkA8{PkP$DQe0^cJyfG{bV)d3lQU^!RD8#oJr`L98z@wiR!zF$Xt+HKjp739 z+=cQUGYHQUsSAGMyNe5~_4%C@f_h`&m0ILoV(>8DjP;?Yv1d`=_7^Wb((ufcs2@I> zkIabip^x8_s_@p`lyc=MTeOyhFTMn81^8;SZa8B{=4DZ(L@y4gFm1hbtjLY0pg+dO z6JuN30QGirLh?s_#Q*NXvD)GPV=XfLrOKjenfBE~YR&kb$6i`6hD4W*YCz`xvqOHK z0$A&-V$sHuF3dTUgD_*|F%wzEey|U?ae$v~DwLF48JjlBW4o(bt$z1#Kfp7QM@Ris zY7X*6@BE9rIZ=oXxrpq{2jpV`S!y{`h2R^6E@k@g4-yWoWL)jyb21rGk{U`76&YhSIaD?YT+j2C5nUu;hN;{jI zzX0X|Sv9i=i|&2D!mX;@W(#Pnl3p|z_fzWPG{T-9 zBnJ+z3Z3bNJ1UU5OO2xqJGDv{u@2{&j{E(?R9|tYJQ-V95*ND&QZJA& zbRl5?q~^XmW!Gy18mk|)0;sPAADnS#lcuHG^@e%Km^!p%Q#hj;QaFoE+x};owf$%D_rTQlETp;Gj4@CziW-Rc9OH??&puPvTN zf>ioh9sG(v$CxyVOZebxWYyWg?8@gU$PsSKFGr{B*tOO0o#klIwdxa`Pq8tIi=FESJ@FgPf24P=WeuGtcI`5k zpuq<#Bhzgis-hErdMvNd5eJ-3fchy5pszO|YKHXfmz+txW}$Pq=hg(ql(2TrQhbI$ zrmz{;UB@@6b-h@5RuvYDHMQ1q*z&|y9&%6V9M3cD8sXc9)TW_ga3oGfi!()3dzQc~ z-bm*`kK;}Q+c9@V+5{hg1Dtj%lo>fTM%o%i*v|<%_GOz6(up%6B7ToCIPsuaVE|Ar zKeg`P>*F@SxBsPm$LzxVU+80ieHWDmuTQCU+%MlB-?<+;1}V3X3+Q`hW|AaBFmVxp)4nQ0^P4E@5R-VcJc6RU!vZp*n)>TinWety2@pNi@SC+lXGh99EJ#Ji=;T}&4o+oI{e(`IYv~k3a1a_% zlKFGK=a7k2^0(Ua@1{A7(+`bTwqRw3H__d_smezV86X}&R;QBC+z#S&q!9@WY_WXf zb_83yK?agpJleLn_f-x?#aZZhq9UEVD2K0y0#d0LjDqVD@E9=LkQ^`Ew-cf|mU>ms zsDtdf9l%Ig%t_G+BJPTUTAkCM0aO^-3xpEi8Z0wLVSBB)lcrMP(sXb!`{6@B1IEiJ z1E>e2qJ$w)J!NEVm;l=fDmv#{GmF%XYte4)5@>CyP3s=D66Q8~ifNWqG{=#i%lc;( zmDghA&*|+jdQ7`pd_i0wbp!0{wX9=!Uy)Q@KAIU8vgAaHw8be*8N&)R5_F(x7t{^A z>eVNSxA{wY&ncZbf!4gb!!M(EdHcMray|TBcXMoX@I?r~di^P(e=obA0l{y4!0*36 zF@KGZ8T^K)`fo&5;g2LXs)&Cvb~C<^fWRhc>38*5shV@b+7*EZH%nGCf#JL_=R>z+ zV!C63g;QMhhpnIg*mVyY5n9Dj16#*KY-9!D zc|MZfW{)OR@JX%@W8_pO-n7IOp=NL+f;QECC6slL5H&g8S&~m;c0bTh&@He3VRK5pr zRlq@WuGr9E30ERJ1PQ=*WZRm=1al9}T8`-p83 zZKt*;nLN}EW8M~z&DX-h3=ur6h4MIEjip!UouQMNEiG{mbI!xV=OS#WPPV68iVxWG zS$7bECVuJQSLFYXsILrbtLeIi0&Vf)6mM}WR$Kyip|}@!cPqsuIK{O%q{S)j6e;cw zr9iRZ4#5M2oHx(+=lho{IWw7au9>~pT6@i$%fn{5H#AS&CEso2!n7IhytElwaciel zq4S%w!WmH>U%VWy^97k;{2kpyA2QdCjRwkIPrKd5Z24aAz_UI$$O>(w_BAQKqGo{Z zjPyKS<~>RrSrGfeIcLwNtO$;hp;uS^LAwgR)WX}dwu2t>jbEqKJs>s8VLM;PJ>9=e zPm*a$D8*xco?<3OGOm40!ZZU(Ei*4>th$1gU-n+b-!+a8U2(yAI+Bsbw*j=?k!~R(@YXdNjeJ7@=XdVUNWg+GA&LJHQXk=cP(12faBgSox~ z@nu;(@`$#BRQJJ-l}ZntPp6wj6C?Ia-hVx-&U9o}_IOPj?YcKR zpd4pr8TOCR^alHby0;G%i?VID$;b2k+1*_;(;$|k)`JBH)k3J1VWrKAmhl~hD;QQW zJXa;FGSeH(xo*jx4+(Vbd^UK4)BQe8Jw|bPH0!&3z$0ZfbMH+FQ32)A)l>j8WQu28 zwI6fWy7Z$#nlxP4Of6;W4@hFEuWXRr&kXeWk4x;nzR{c6+1|=+udl>rc{D8YwpbV$ zQ5jpI6WSY=Q8n-UgLE_5PS^4UKa=W-Eb%AGt1e7zScDbO9(5}arX|CV@ZNF$HO%c} z9KUWl_^*&K>CS7Bm19H+r(cJ;UGA6uvwdzQWnLN4jXA4tTVH(>jPCw2goA_?T2o5m zOFik$!%9tC?_;r#zW)-!`oMVih3<`V#w6-Y2~3mLvFQQK8#Cfy#WcJj`o)$Z9d7b& z_ifGbP;Y}6S*t`m!|yg9zgUb#j^fCUmj+C1X@8WdTu-5q6curQtvi!o<5puiLQ~5L zc~|RHMBcAys^-nFJF^C+YSmye6usNjmHK8#-h6r&Q$!afoVh}bCF)c&l(n#x$$DRW zHY4?jYdU-E6o<u${4oym2SurAj}FFKSpaOLRG)+OHRr>!=9-tjjBtkh+9!rx zE%i>Kz{_=^rDp=4G&GFAGJo;u`a&(I_4{&jQ%Gug#J;8v^M?PjwUt?BhR?+b&t$lx zk2QhT&?!|UXLnoQSw2j7F;mX`K-ch^_BUQiV(V^zL^4wCKCn>Q&c#W;sEX$Y#g4-F z<3=Ge!cQEP`m)!Dzpl)Kh)J!>I#U?u?pmk!HLZKzF}o{kTD<8_3FW6V{OECDt6Z3% zO4_tD3hke%=Duf=kZ@RMRFhe(m+2UVd^|f9o3c4b8HLExpID)SaJ>v-sR#omEMJ<3 zyEpQ?IFyuYYScpfA7^rJ=BVB9rQ%HRzd?bQ~PP5S6`A^ zX}Yd97!t5N?NpD9@Je#wD_0kB6KyonFSMv!-NJ<(qr-kPs!Sp5UVPfIzngHokdbI` zEB}tQ9cY_ZNvAS3=T68d>gbrBb{QpqPDi^K<#FrD#@(CD%tEO*Ipn{eGQ}dd_NUlV zR5T{KXld(A&#F|DdXuTV$)p625`@DUm#4OqmXo2gYRo}${Kt3q%ji;)D9p@F$CEDI zGR>tjvy5F9)xX6Q9Y8gna@j7!xxN(|CK=BwgH)4{;+k3{O!6K{$?e)Z@}+xyq9r(l z85uAc5dUMfi4tVITW*y*%Gio+$}(7IQhz7hOG!qt!(=rrFmI=;knzVScWn^UD?1iT zk!jHb?ya9^vGT8cbZ%F&Tl}+73-1I;fa1$9$)ky6_4fLkuq67)PJ<Eg-8P|(W=l0+2-|;gf8BSd0AMEK2{oTt z$zMOL7Hd-)Hgff0@T9j3l+sL-oV5~=kisa;2*~Yrhle|M4lemt+_Jv^h`y%bl_An2 zk2D_({XEn@S>cun1ZP#GJdFpmkzp zDnuha%l>9dOOTDCFV+TS=NphC4@>!QdYrb!=K^{y2&3}z#i>Nk=cp4v+X+dmVM(7j zv+^>_KA!dJEb^#`4i}{i%hM~o&GB$?pp1IkIdQi2GZzIB22-B{RXvzNzRw@kp3&#J zT}U(5E}^kXOzR`GeCfzY%&NW*cZerxFtt`fAoS0iu(lSV2h%RG?R6rx)9RwpKd48E zOwDsW>Nsesl@c$EUL+GE`?a-rwjq+=ScE(VzKny&t zLq%7mZ{B77h)vM`7>59o8@3tqkMt^&{z7D}zVgY+?;npQ6zJFOC4@T4lbb;HH6(JEV((LBiDjpW^ z=nnE6(7bCAC)fI&&Zv7g<3x&H z)=4GHiu3k=t92mFFz$ELL0_jw4<(Pqr0nEA`*O_wjL@@dim?nlZSc4=78H0>C+Fq2 zW78X6?efrC+V&e#aDvXr9i6gkKXcUNvovU$M^ld*|DJ)p>G~N1do%PS|L-f@X);Yw z-n+jWTNWAoTW2n4r=$*54+D%|ai|tUrun&_30hcLcTe$W)aqMQqXvqC9LZUhmQ&w$ zrn9NFvYm69y$70}9nG%(Fv$Rvj_JaiQ$-}_K z>_z!HTex(m^g9}8oTpM(8KQM|qMhV*sxOY4d?d!k*b~PhcUM+z>wZbj?(fS#a1t#W z1o=hgigMWRyYIN<=kB55A}?dTds{@cx@iTQuIDRenNs=nlsGONOI(Np7y`9v*s`Xg0&nL4&7VU!dGN0DR7*{s!{f*_} zC+WFahqFzZdod-Y@Vwf0FU4Z({}>2oDnE!RQ2T~~e=EGq$M33(h*e^;sdHFcnJEr` z`#LcH(Oz6yDR@t(>ic?W(n)rbhw)+srfd^@A)C)U*I-%ELgI4mt6-?yueX2cm#heI z@msRD2kbUa}y-QjL|2f?%_IFmT^R5_pM^A&+K8O&%cV2MfKt)$AGPoT+%Lr_& zk3paD1J69#>Iw9-^Z6J5oayUN?KT(4S@hEndqYqNoR{7S9ZSRm@mI|6@C#VG@oYYb zur}|`SMSBTD=n9C;vb4gvpym6t%qSy4;|+!2FP($P-W|~JLM`huxtcqmJwLd; zHkxVD*WatxGjUIX%wzLTYD&_dEjH z`7yYRU$M!IZ*QzcrAYBPXTHUoOr62i4&ip&48Sm&q7vWlNjEGReqq5rxVFZCf5bj% zbbp8~#rAyB6ts<0XOEeu!*t zr1oS;-25U8;of6UF^jqNhJ4nIW|M*)NSiAM1A|LzK9x1Tacf=Ig`3ThL0hhMw2^Dh z$)fU3-6e7uKeo?YQZeT2t%Lnzb4$o}t-_|%E~w5!a=W^k#n5fpdA9pMtX9bmzr&Tb zO`*dGn^N%&BCxF}E@z+?*48TR_DztPdA~o6uO%JQS(+d9nVYE>l!g~|ULFOutr_ZS za`DuHep8j;vnH`G+fCIg^LuZuDno8s&rVSq&~vSVI(D2{`E|8)P)|na@n={pcYvp7 zBn~{<;BovIZ_~4Od>WI;ppWuWaqYLu1|dx|t#P{$*5mSgkR&VHPm*m&@8_)x^*4ES zk4FE>d*=&#~uQZxz(1W zYAX^+vFasWBw;SuHa^gtn4UJDTyZJZm6VKqshg~j7Z*cgrn<>ZT64PAq%|_14qg8! zY@`J=hRQRs&G{(fI-80e<1|3Ob&2qO!4+}ri5#`-(In7XHmgJ9nYwK3@4Lu4Z6t5< z#ID6XYsRk;l6N~7A+6zN?sci$agX=2&>Na87UZfAQ1qH>`yDlP#W9IKLjf^SE;l|$t5cYOz4+kBA{&x~fz1{%-_I}wbyHO2 z7m>KYWKl463u3Iwtge2TNM#LA>h~GDiutE1m4Hok%>T#x8%!RFEj3kC`26%m&o$o4 zk8<33f!B$gF?(Hb^~n_V^9qCI*(VnIT$Z--*pG_-5{ z;(6!{Jz9j(JUZsS<})0?zi&Z6KR-pnp%lz zf$Wb&P)K*KY;z%sdp+MQTD;|JA7$JmAp0Ov<}?qIh&YY8@!`r~wxPbc^RbwdH*L6& z^-=K{QS&=7U4W@O?Rtg790~3r;FpK_3 zGR8)p8BeIinJWwAXeJCBI_)yC$eW})J;&Y{_k)cQqWzE~F}~@cHhd&8gW8&uuGNQ~ zX1iO$36=^RmKJReFGFpJ3M33SlmfV&rwcRfh8@yEiCQcw*8cInq=A=_VgLE+xuwLs zKp6h%LUcaJ#XQf)i>6*X8joB8RLmlunFL+z*>C!r|1GlBXY1D?9c-Pkb8q#fU@hZE_MS1fJ*qQ`R^hag+zF3f9KuF8J?p#fj2@<_I~L#VzM)1J zg+k5{u?PXVS37veW0z(mbwTg74s^^eAb+2?K5jl<81X`W-DU<8(`m;ln3Cj-dw>t9t@%5wHy#%r@64L-zo@AvHWq+~q*27#bQ z%*6+}z)2~{EQsQ;*yiy|!|jzh1xfVt{9p;Y_FL4_ShYCF6rs4_+%B%mSfc4kkt18j zW<=3GgE0oOUg2xZbD$2M)KE7fgpwjNe@l&ngPv$UPrYv$_`#N>toRjwhT`*|I=}PR zOHF^3n>DdUzXaC(dX3TvhdPX6z=kGDY=XS%BNu+;H@cwmjb|cO zKlWK6mKF6mJKcD%0a%IC5QwGS{rqk1I~>j8HW`O=*KOO1$X~w0LM<8sp?Ao`#!Gp4-1CU}M!Es^xobH$-#zBGo|b zrq5c+q{XS^+5TF9i}xH=A?H=W*#LnJ@9|K>%U7CCS3c_!I9=$|FFvJ0(r_xRz3=Fy zq)Si72(!MUmpH{;*66rEoL-Pic<$V?-U~;z9ISw$)40z&RZ^tRk78HAN|nx ziO!!r3m36DF9^#O>4a#_PBN_c*6Yhx{vJ{`+&^6Kxt6|6P5s(AG|kJ02DpRvx(&m< z)y(yF(^QLod(ylAamuL0C!ZqlJJsP5)J;PyV)eepE-aOO)*D%oy3^$akrl8M3;>`69YTgTW2+HcuKP&_?QZ!V@>_Hzs&{@ zSn61%4zgI-pvfpnz!uX;4_RC@1rHl zJF%aqpTbJ6ng6V5#H0s(M*GSy>%W8!6Mg6)t#{Wd46L$ni@j|84Ly56PaKpO+ZMJ& ze83-v@2l6ugkiE%2U502`sYe4;mtQ{%PbNwTs!QnzYF9S;lG~{NVI*db&H=oKIkEH z<4fmVm!rOO6cfC_Vo?ZQnnmZ8+)g|eMT;mk<;ve9P4|IZ!?0V3l@f|}Wh*oKs!KU3PX9+mpNw;wiKfD%iIQX%Gj3fd6bWxCQyNpzb0Yokw-=CFn$t7a1MF8yFGcw<6|7oWL}}u8z{cX4jIbms z<60DIHTh;vy0}|9JeJbVS&B1FLInLZ_H$bwG0b1{4^GH6iNY={p)NHAuftEDM(Z^n z%_H9HvPN364J<%;jJ=*DSYA3`ojjy2(&W4q$@tmNLC!UKH>nV85HTv>m#rPiXOR;2oa;lTX088S@(E7TP z42D+Y{O*(jDJDkvm>*wAyxY$6M8m60>-SPjrsG5?C*lQm=>_j-63#=d?qzw@5=1di zM%5BK!zvLlPnf-D#T+I@r&Ubt8oosUj;Fx-b^{tX`ae46p(yhxOlZP8fp=D6RIOV1 z)iQl=a&x_&k9HzpuSl8@+B;+_8p9UiyXc}tw|9AOjRcUPM9%=@UMU`Oz}F&-YQ&2g z7xL7bq014=gb+I)b%iaR2NJdf?c?&}aKKIAVJd7i>vG>L!fYk%lTFp+$bsQ(Nc zu&L*|RC#+}#%^lWSeqZ^wx437w56<~QAQNd5#n8NdM=3Hl(y`3VnNr{*0dwYqA)IY zMG|G&_Vi08W1_)dMGD{5AcZs*J15Zf@^y#?yu?7*U8iL%ejJ)1HQsj-%bn2XScUI= zVfyVA-Yc!aKC~7ltH^CX!}6&i9RIrJgq*^x`&#FTnZCt>p{1_%wR*MTC1`wGN8pi# z>CV%)M@1wqVX-YyIJQ!3Tg%HO?ext1*Rr2}?&}W}#b19J0xfm@amn$HxLfy5)_I#( z$Ib1pmcg1M94H?E{b<_w4z?~&N5h-ucg`W;_AYwxbwDxj_LXLwT<@@N`nPWYcc}eH z#@`@w50>pGjFl#1Du22*kN$Vm@1kZ;n=2@DJZ#pjj2BREx;ciH3ogl_37Yjun#Zdk z{_bx56w0vuuk#0J2wckc_bXV-B4K@FWA4A05$}b4yTAsnl*O^FgU(HE&liP%%2&uI zc0NJrC6Wd4_$OvFU-AUiUzz2Y=LVO_pP{oHI6Dx$J8}6kj6==}8wp`<`w57uB%8=+ zZlmv;J0ae$#?}0#K8o8KmZd~ASbvolC(MO@Aw@)q2 zp&l(fTqA#LhyYSvo}1_gh~t&n@nYhly_e`5{pbT3?RF>TESA z7x^ntpC-JUnYv$^vJ>zu!qX@`Elxul$blshm`xz`HmNK`0sAy}a)h>lqvOkHw9boH zYQiZWJQVA_C$~V$vUEe}iM*oYy?Je8Y>N^(%sa9J;QvrdYDg-J!Ho4cjd$(Dw9FJG zN6WZkZ>C?&&K;{@g9w zyHcU9bp?~6M{RNc#UsqYb9&(-lqMU@hA+{OgZa*1jhXTo#n^eJ5<2Z~cFE-5fzRxd z){eKn@Q{QgD@By?waMAIC^5z4tdFthE=(0n5B|PRu zNAT(aZeRASZLnn{%+B{aR9Rp%ul)d+*PtN3Sy=Y?ykq-9Y(K@_j-K(vRG3>zs!zi| zz8j;aVNw4c-CVfWQ%FU5=N?TQhifiX40_>UrmS-&HH6)sAYHz2aPTV<>-&pk zbp1Sg`rO*KBl6N@`2^Box*<##*B|_}v>@{takj8r_)0o1wa+`va+H_7rlVzZEhLE3D|G9#Qgr?Qe!Zrrx0>i~7(29v}~(Ovzn9*kegAKP>!MxBaE9!c{sOlGq0vxXVj zMsnkC+phBh2q8R#!2ae`&Bsl4lp2qHMZ>~viagdk0$-mfN4`CV95*5DUdHYGVFGk2 zANKdTh;D_u<%F{e2nkzSo)z!h`!1-c>!l0i-4l^p9;rSCP9!&$dZ9`ygFNa9Xsu0V zX{3gBH`5lMUB)lXvV9p?nS&_zJB>2{{*b$MT{};0kurYTrm7H0!j`5JY}dm8M>qir zLl*79);r=w#n6_nPNdJXp&3ZKVYsey0%Z83BXn-YYz&P4gWmO=mr}v=W_~^h znsJ&_xh*n z(4stRGMumfttGhjVvj2Z`537o;Doh5eD7MLQ?1R!q3SxXtR}#w3~yvFl;H--P{k(gs#T$r6_{h0?KZ1AW_yFi-^xoj?XU zg;}^-M%E3;@(-x8j%g7U-IfLke&#%kFY*G+?MQ=vIfds{s=%uq*J|?PbA66Cl|>U< zpwEaFSD~O@kkSY5TU*43pI2c?7MqwI2jc9h%OO0qQ&3L{=x=xH&)~9WLe;Ip+?IUr zMI?s^39B6;nY~*e64J_Q09{`6h}_mH+gaC>1D&BX2+spcCNtd~3W=$cLD;Sh_08Zd zGnHW~EQk9*1FhZ>O5E#sfcF)M^lw7zyijQpHHiUe04-~ETEsh06=M*^E|XEQhwk^jTQ^`6>w?JTr|N2Hakm=o;Fy55 zYu%{o2$eGu6%%w%F|{z zbEk;lGkeb$D{O7=D9?E1?^?cb99rEyWW4O>9w67tFcLt+)?9Dhc^Sla1n$(saUuf@ z{@Z8mHL~YW!C`E7Mc%)5eg(JVoOy|8q;w`BJ0lvLErSYJ)j8b%vmF+=vgSaVoA9JI@rU+L)0GLO!CelF*0_1a_;t6<@j!4z1e zYUvstlzrU!+XPf@ZpnHH#jLT6YZ#ZwudIX`5r=-K!QWU3U~!&Cnno3T&#wuMetbrf z5hzX_Ay<%cxhvRG4-ytU|0#EUbbWvUEw3ujp5FGl)~9+NhO9?s<3{6FOq`JcKVy6^ zAe9ggp7Ex1VOXzP;?g31t%8>Q;h@9!{IJk*Mz9tzLHlWY_%H-)>Q2%%QH1-m67X1v zF4Fy!oLe(caKU40WpE#R1A?swsC56PTv%gGPY0nW@H>fHXq4y2Z-^g7sHTnz7yJY90Wgt*gh*^VYlpiS_C5rsMa3Y%V_jo{_tg zM|!aJtQ>b>ud1ZyQ&E36>v-4?zRnwwEljyQa+Tpa9)g!WqI$=zZHv8`CbS56E1`<8 zR`hc@&(`zX5t<4T>-4!=3ZQO^xlsS%i_sYULLFS%2fUm8&1sM(G3-{M*10h25Ne;4_2v1xl|7h+E zd8_H(*|+|l{cp)-t?|d65e?SOZJ>%D^?ZYo%h&8XtC(d!k0C~N3`yk=kfURWA4y?#yDHh|Xyx(X9KBwU0Hw%cI}junU^u3P(El;s za7ZU0QpxhVIwSK2d^3v*YOHTnH7b#Jp-Wx1tA@EhZF(C1{`tB90(W!bTE-m!9M-fE zf=6IL{fQ9%V(8n&^qnnjZ^1i>lqWMoJZ8sU5U|(XR_v~*MLqyuwo~3gZGYN2i;V@Y z?bgPHnE6jP4fE)>B@|jw`lNVoCb(PuNF1^dxcZ)uAm=5Lo!!!*@Dfhgt$RVgF8-Qd zghHjAfY~5nvm%cs%H-+b)5+Nv6}NCwmFdAb2f9XS@zHt71pSnm>+}&$2{EAwzgN!F zvG(8>OLP}UU=i4KefGu{B#~7~=I8s6M;G6$>u24o-&7T0vx8N1eyjCGUa(PnhrBgcaGZw5=x4GzobdoTx z^-J<$A5D2k(SblhjrxM@8m`sy_qxBB=bGgST!$Th%S9rkqG)Jw8ta#i#efBk(ABAK z3BfS4AILHa+{Zr-$m5rRysd9m43Blo0Y!r7lJaeY-n-v{3L$CQB@d8_Do=OM#+ssC zMYzS%ZA)NmT%aEg$@?#MnVR}Cj{#~OUw=$n5ma7{y6hb{W-(MYsjL65_@;%!6bsJa z_*4$`>S}*L-YY(N4kCFVd8i}?-o&pq04Q+df}RjGAF7bA&7_1fE(kk)8u4~Pi&qLk zep_*OvXdR^Qq?lo_5qKf(2UxA$cg1@Lxh21PO#TKHCIXh6FC+}Ksy%wJZ5VkdZA}i zy&uFtmq|qOA(oryIpw<@IMoI{I|=3{?=&$z#>+P>1;>KBqQJl4;RHZ z{bAET49GS7vdB1wxIRI9$2r%ev@*3G8`P_7KA|FoQ`SE?rzm z*LJDoJ5mGauB%NCVI^ORfmCz9aq=qUvJfjL3jwLK#=OVe^~^vS2&YSsGE-O zV#<$2C}`p4MTEC}sU@Y_TNp`b!gMr;5b1buEbse^ojMn^?20A7P-<$lR()VzLIu&J zAhI^UB!4T9H0W<4l|2BQTYEWwe~_V0P7?)|gQc*OdbcYro&SLFITSamwTm9 zY<2~#>7IWGzkjR6UQ(*3nl z!BNVw=C35rTvXP-`CCQZR&wJ+QA6OD6E90EFUD`C7ly!768phjNI&$zI-m7QsqLS} zIwWT1%8s99{RbZ)#XPn;ld1>#x-S1VTkEyflDL%q350 zn>`&oZkm8%dX!cm=d)y+n^(odU=K^;IzD|EeApLxQ)lz4lKWLQ;NPEDXI2fakpjL= zZwdWxph%IV%UG(`h{MgfbyPQ5`hI#8wg4-#Q ztv70b|E3J)#mapAqT}Af?I!0gab9r!5MIpa9HKYMwF)SU-y)!TZG+Rwa@$9BnIvmD zz|E?ib=Ze+bFqJWNi>>#8P;adCdx<%R@Ip|7HOlU{TJJRYUcF`_v#O?F zE`QK&<;cYicf%>*#c4CFOGa)8U#GSvMN{L? z@$#;ohC9&ykV>Qf|e zoLgG_kvy^uRxZQI&B^tHD4;zdd^G#BYZxcxVb5G$95ECk zK0kI}_U%CimLw}kMj*h?!LjO1sqv<_q)fjJD`~2z~)yHoXt8p(^+AJ1UhAuipe)xvNs>XGqJxoTRqvLyKVRG$d zxT#Un7b7^ISUHaq%|WI{aG1Ri6xpxfeBKej;}?0_mAScaUnhwou9H?45V3013QRW4 zgYih+=oe(7hq3rGsg=~p+#8G%QR4JiZUIJ{?ChXx*uWjw)p_8s%58;WC7@D@^t7 zP=nPr@T*LXaifzDtzq=;z1}JQ_!yF+bb{Leo2YT`Lsr7Pw*n&f#}nU*8*_NU383TY zsVhfkBz0ZdX{OrsU4$D84q4>0$Bo-b7`cD1bG%dJZDlCRHW&D@;YM1KgWw*JNy+4c z?3yw9x<9S_s#4FmbFOb#C~bRme(nMlh+N^Mr%XOQktE?f`2Md`p7%fC)IB?r zY{$VLT;wVWD1a^w(CDswEOA=>GFFRB=C%gG*RfcF)$Ymuv*AkkAJf@aGc?Wb6z!;u z*E^v1GIWwlNdc*Bk*yUf68~J_p{7e+tN67OIt;!1A9zS$QMWSA=hs^&?|n{8Te;RZ z`jdavPaBS^Tg-hbh4H8!xnMydEE4r%#)&J~PD`7d9)3i+ba8+}uv?glFS|)N0fj*+ z#Ec}ewZL_MjLcR3{9J~qzE!XCVGOJ&|GcID&I0k_xlMw~zUsrLOn@T^2#mAS`7SsH zZYIjfK!xde42J*#w!jJ{K6u;&&6jxi$BuHp>ZkebL5K9Q9Ssvah{|H^q*STM(5pD zd>csNr)u|oPr-}aT%T`b)i9Q3xU%hHeY2*BM-p_P+Wi1ks4!yY-08MmNo9Z?nB&UD z9z)<;m}kv}49B=|@Ppu$eMc-#>pJ{!U>tm3hcg}I;;6T2fO=jPc)FYHrF||X1RL*f zV`D}^UDjnIW>yD(Z<7DdyHc%R_i&e#^BL>u!%t&;G*j(VKB=F*~9)GL~ zTT>!(MTv|cX!(qp4!CC5ZmWI;ROoIB;mok$;!e#9*ahM0GBSEc^TYvA)4u*mvBm#3 zCliUjc;U2e+{-I5OkOW97XSW6s=TRtWi&{V_>u1;aarHhEyufqjV_Lf5A8!Zz<-I> z;t6XLsT5ng-Niyy-Sj@6`Vb!+lrKnK(=HP4fwkg(L*fu{VuRr&YRtPMlcndK^2D(p zdO(&OtLr^o60nBPJ0odnHYp1aBAex(k+=ew$XRQgRw$&Rm9-JENK?^r^hl0!{R0u} z>#g?Bj$uz`k!f+f|LwEn!AYF^#RBiIx2Dj)y-bXtHmX1isNweaBmTe{@0xw(^k>s) zOzpJ+>DEueR8btepWZmE3tsS?jF*lceJNQut!ZjH5}Wof#HlK!NL__34#ZwPTO$5q z#~@x%Yf9>BqCCziA)le=>z$@oOaH$|S@1#|^QOq%=vSXW2$A6v+p6wQSYY4n{ewnZ z^#F*k3bO^(j8o4yQB$i{IC5o-tRToBPEnF=6}QWxEs2GzvdsheVVaqJcYP2l`_9gq5*H+6r5&P}1a5j()#|KHTj z5BCmihV*>Wv(QSKCQpNqs^6%gdco-|hN~8$|2NIdR5beiHQTE5037r7mH*u)JFsvi zs0YGn{dQ$WqMwS2$*!FH-QL5!J58VRfq@27@j%u46;wp_migrSlL*cVk5Bn=gp#_( z&o3k#aWT;UQ@{N}i%~pk{K}Vdn(@EQ9sYv-eO4NasNh2Pz9Qu~%cuH)pr?tcAYw5n z5SoJWA_cOWCtY^kB!m*jv_%xTrtwUM-2Lu)GWSIJof|qByZ?(m*UqpcKaMC8FwZ?P zo#(k2;TL4*%C}ly=$*jU zhQ1zvk07U4FlkfrdS*EI*9P&m9B1AiUf>&L-^GCV=ELHp8Adk2DIpH4S3bb+MI0N@ zH_Nlz&C>{?w`uQSP)`gB#He0UX%0fXPsU3ZUy!OZz=qik7`Bwd<)Wp;0r^dJ1@TAR{ z9SE2AR>%km-QYfb27u$jdKpoKUzMTKiJ)-Z^Hji}V!trD8!Gie*_@}$bH7y~{10UE z8BdH)f5(M+1;+yumSoo5{(ZCTzsBDYRF&P>-PfvBVQ*({gI3o~ho+w=Lstj1`(zyHs7Oh zGhmlGPRm!hul@0c!0a8XHgOXXE2+DxPNsa=_tuuf?pIPvWvl@mVP7vF2SIjxg^)$i zez9&PtWAgYb zTHo$e;XdBH*x>f3wG(OZai&#eQ#GOw47CXKJLi6)!1A=#(@0M$`C@s{^0%h|w{|Li zGfo%2Py2UG$R2CGsN1kE~w+ehzaBeFk9&Ip<##a?%L+S%o z)Baq5Fd+E6Nq|!vn`EbhNJZs~m0T4_`XKs&-soBU!HA&X^Uek&j~s?R$Hx>U|F`$U zFm`k?5{78BVJdSMaAeB#(j3iq46xZSVh7R>VhNqRdyq$aGVZ#?aACIh%CzoQ&h=;!i+%v}&N>v#-aSI^_}+-$+QuKMS75q{dsu-01tImFXScUN$5JJIWXEa@sB9Kn7)|blmHn|A z8+~qr=$~{^c+QQ7IYmWIT==jt+$dIv$|sdSfz4Lh!~hSSm>)g%29%{{Nq(s;NF{!|offoS!g1mJdA zP31@)2V3#r%?QR|OUF%=ktBdXLKBBIrRw8B=A{{bhMD**ftrNFUsch}tJ?dBnJiQ=W+1vGL$##~FRqCpR3o z)AZ?tVQv(WCxovLF;dnQuRm-EV&R6vn*8_;4dM>2DSr@cbq~nua`IXMOm&b>EKcV& zE6BT_mu4jsY6~7ffX;pGRMIc8GjJ_G@Qr}cto7+!=OXE6c%9V4My1&11raCt&w1pX z{VBb=h!{dgWAW*H;}{Zdk-)zWT8;YgcBKmQZ?Nw0|5{Do$jUq&0H3D7zv>0rjz0ip zC@@0uN&NwYT7zh^=grGl)fv-m{O50{<*M;gKYC8};vpHhcd%+{&7nnq?KrHW5tmg;dGPuo6mp+t*?B zh>G7&3YPpIUvC{1)%S*v(l|)SfFK>x9nu3M0@B^mIZ8=O2t$g1bmx#FAuWxxgoKnJ zJ*1R$4|DE(zIWaG*S){B&YE@BK6CcjXRp2A{l3qBp7&VEAnPRbUrYU3{QDWgjUa(c zA*?E|wZCyfes=(!Mx2lpTN@^iyDq1)V~nIBvmoaFs7**V#iqrfe7CXW*FzJafr*jf z!J^&4j!D(ULr%!0{p|QE6#-KpE_e4YWT<)6QOF_aO6U@sXon9T$Ux>z|NDSi%a6|? zZN1CAH));|5+50PT?eS={2iN(l&{}xYD$>YbpMIrZEO5mI42_$ro?Q9^Qn)B?|J=; zE_bgS6f#xBEL)Iji~^RMJ1J|s`I^~1wL$@#W5~<$oQJN)!$y&NwxyN82PWH>2HD|51n8j)topP%1AIq8*b!Wij)Il{tB;BjPA9Jte1 zWqOcbHYh#Fp2+eX0Yziw#>c!`$%hy{XbOck^7qF)^4mU;a*rQl=fXFNQjzxOJkoA*I{Y&Ij^gJhEjk-0Y>56)L&7v-8uD zN%ScE4Ti1LbbVn;>|g6n#lYyLE|%SJlTLY$qPkY@PBF4SL14IjU;Yk%Gz2<0XeQ2< zl&I$^wmPNd>l`^D6?GD79k)9blx_%X{==Cj5O;7S)Zwt^WkH+!4O}f-zAEtl~Tyby3D)Bk!E$;EsSHd{7zrpd3{kkq(6n{ zULlQ86xC*RD&zlY4so&nr#U2#1+4K8cOgzSkjbz~!hn-ukuLw{=kZigR#x?G;+*Dm zb(9zrh0muYdrF2?$5Jjc;_Uca0VL9*J>IbJV9^0~WF`h)!;=<=HscV>||S?JfV7xt0pScbjrv?+d~)gI8I24Y7)vAWK67Wl6^8!aie|3 zeUb@wgP42m=TcUcVkE$axDURX()yBD`+MBc|8`{xm@X6|{5>izLBebgGRA6r`Znh# zS{b)WJ*IY|>$~+|{KW<;Uh2kTYeWVMw)VX0WFd3It=g}^eH&`0Y0%TGj9HVL_gS2L zssv%iFgBGw`HIIH69!kN64n8|vW}|+Pf6OIZ4f1p?z{c88M`bJ!A_s=RLs3-4`6wB zU^PtY*Snm}wfZQ+BAG3RW5F3i+@;+*s{N{2ve-%+J#S>u3@mb&=(lwpt{4v7^GkH?AYhVoQ}HjBfSRg=7(k;HEDiRzU{ z1HRws3@T*yn;F!xX4b3Pe`~@58_b3h+>@xQV=^Q^PrP+9PrQg!ZeVKv+k38fCzXa1 z;UWTs5)M8QKCopPVe@h&jslWvPN=AyJ@x9C?mG+8goyST4VwVZDX82BSF|N4#$tmh zSX9X|{DVufKF-2H=uuzsP2mc!y0C&8+$9zJ^3%dx5YGHIA$v=HmHa7Zw*W`vxHP#85AiRCnjHh^Zq{V@#E=cnbUL zU+ZfAudE*N2J~AI)amx{(7^7!>y6Lnh~!ZoRZ)y|IPqx%YnQ$ZxgHLN4=hE%}8v zpvjIs-S6dVW5u=k*jq+W6hl0sirhV+Rp<;j3LhU#IcO|CEy!P83RYvHdP`mjK4-lq zWD34{e(EjK*$wBUj>b3|UP7ASLQqre3bc|qkpq7QoTLYa;a`Q;`JifvBG)^l4jkBF zDvy}gPC{ifkY?*|ae4&leQ5i#2k<7yuB;*ootb6bXIf-%O z!{)_L@Dj2F<1cC8T!O`wfej|^28kNEgmO)^GYv(+;~H?0L@eNvRHcd9p@Oo@N%K&n z8sw|nR}S=$BihU3VDmQIx%dOF+(M_twE0588^ z3+D+R`zc!pK1N-+{KGDALgmIYj-kwAOciUI>-2PKP|rSnI*_XA>L+hRO2i(V|KA{U%$9+6lsQ>&I89kF;H3;C~EiY0Q=0hV{+lU0~MmXp(42&>M8hsoKZku*p?2 zBsm^qQh~}UKfjRRe_Ff%9t4aLqr16x|NWNt;9o<~LA^~rYt1N}=r`>q_H!2#2B0eU zAv(!X;_lp_$d=Rn?bg0xS?fz$86pA$$^9!@YiGhyS!Lfy_JMsK-(H9vO`X&DZJOUc zZUi~y7R(a^M)a$q>Aanan3Xg+gZygo=TFz%R};24S$cTlmKJ@8U&BN$O&M>h-_Efz zHnc*vd#L7_sV$;#g!I_YKvAsofmrS+wgc& zZ=Jia@h9a+FK~C)4rkV&ih)kpw6R!-M>?0Y0e@{pHTXOb%^T+P)4&+CFy#sc`SPd` z*$1R^_r&2YRP9M`7equ;ayLJE5KsDwgQXtasE2mmr?vT(R>1QKPS4JJIzjvyuZ)d9 zADrtOUtr5AWT_!aq9OD>lXHb8-u%OX-ZRHMAi)X7^ zYZ4}EQFr`A={1!pi9c$$`vIPwv7nG0VeM$tD;Ot>1G`YuO%2%GP>rk04?IQpo5vzt zD|3On2csr1G6wXi)!f@Nhs!^s%Scn#0y@?{xH;Ff`*g!o-4oQg+oJ-XAm5e{@F!H2 z-69NpE!D+N5N+iEO+W1cdJ(ilmajDDAt0U2nx@k~3`f@Qawi%4dAnqK?HkDZTtF8? zeGz2^PR0~iK_yMvx|Am)IXB8rxnTSRb`VcW|*p>LHy9ZWZy*1LD68qFCp3!W>X z_?&(L-G~?AD3|BdIKYw|3BN(K|PZ0+sp-gWagApYHJl*k?Z3Dz{hg~g?{ z>@Bo{s@JG6>?508Weoc$o?XlN7s!rrZ1ydR@bivYZ!Eeg+B9GUgzql2o_MT7x74@*-QHzp3qek7gMSex zutB$(pQ>1CGByE1primvP=P1sr=oaM**rX*a*7|E8I@frShv{hE@s|Qu1&+k0ev$q z`&(>SUXnIv7!OxqFm;WJpm%L{5K~9m+KVUTvlFUVKz@V;6##P3^Zzyh|N8&2t4o}i zJXzn17w7xitM(}s&`N-dQKXyTG);aG>%&cZmK1g9U;DMsdOm-2B2;;^A`UPJ1Ygqo z{%SDclt*u}xd=;#BI@nU#BKy3zv^g=h2&3f=^U&*^BMlym6#N3ttI>PV(2$~<2<9h zuXoBWp!k^uPFlUy^%N%gEYV+5Y8{;Yz$JgL6A-?psInrkfMoap{Mt3m^)5)jxVLlp zh>$(i{A29U`FHsCa~?}PLTXkY#PNWZ{!08!)-$z-dRC6n55F1r9%mEQ$BTwP`4X* zl^%5%|E{ITJW22f1b4_uoJ1TXsJUyJd_T{)iE1{j1JOLNL?F5;WxW1jUO1JDlN!n769+0VVO@zpvY{b6_F#^_@I7F23AWAd#S9?Vj7Y)%gy(`{qR4NR-%ay!Bkth_pbXBWH2tNXm)*;& zy^izg+(^<7VA5xWn*@72;gO zyP-d=)IM$^y8d9}h)ID|L$|uRb20ztfr1u)yr2XqsUJ86fS)TIPXuZ_+Q<9%a~&F_ z!T8C==pCm`l>Q&pdCUYNA;~QZhZ}XIKN^4nwJn;OS~xmFEv>n3$1Ad)Tu@MscKvBD z@ee3p3qnG^67Vpo1e35?psQ^|m^1jq9B)h#jU9qP#6zpvL;s_p>tB&K{i zR`U)v=W@SXYuk_>^?A*Ku+amN7(-^KfTKnO4Ebgi3h>1OLmgbky|c*}+DqR%_9e$W zI175(-qvKQQQchZ-YRMe@A2KL^8!#hpw(*`BK9aPg?6Ag_4+l!90wTU-qkg|pO6cg zX-CcSbv{)C36M*~#+~$mZz(<9aME>jgMCFljA-G!j(=?M8<=(Xuzfj5Xwb#UlE#ZX zt=I+HM&pTqNK}Rx<|QI-HdL|Z8}(LJ>UD}F4QN3f?%eEP9N*?91H zWb$TLA{{3X^2x0RXOR!#;UF;YXt6K%wiQU|JR^1H*L@{Amb;d5%^3PqkERu?t-#fEQhOy=5rmkLC zt_~nX<+VH|P48Bmq5J+ref>)K*b%vw*V_UeoBLVd(O|1;E!5CzSKaRU#L?N>_Sa?t zCl)rh_sL`-kB%pZnQ3sju3T_X`XmXDPRR^t*AQG^Kd8?4l>8ND=d<{$*to(`MAh^J zKOy-&cHj%_t+lJnZXkSBpF0;|t)M+Uhksd{P+zXU%=XseN+EZx+ThH^GXd$()r2p% z!+SPTNFxw=;M-8;{fHNTa5d)d&W;W~XS2zR?*TIfsp$M&(W7i}hz!C~v zVD}kqNditsdWkINF2q%sRzH4{%oYd+8`mx~OxWhkL8FST%INPA|NPJnjg>%l`Tb!l`cYbvd!@k2q=E%X^-N1=lJTcnI5>P)R~ zz*}`5=DuF^*JB-GQt`s&eM9JABOj!qM6=he1fDJgTrgNqlpr#kiW;*fQ5kvtY$cB` z!FSMm_sth=Y-|FihY!}Y$SY}RqO57uV{c9*{D$GbJ3DI~OI>y?n1w+gEb^6!`5B%0 zoX>|(YKdYdi^2#&uQrD+6Gv?Qptr|CtZYZ)-;v|Q=M%OVPMZ7Af#<-}MHKZCv*Rt( zpY1NNj#$6SeqlbjmS`e_#7%S4v+J>bX)4YHqiAMp@X3G;wy zCZV9}U*H*%$x;;c)B$V{o9_h5=K{q;&hmS50-uP@LZOzMr)d;!-|`a%cOzT*{@lY$ zegh`!N$Ce@=+qv3jbIoFJPiV~QM)=hTmnhrp!z3s{{E={nvsL6)=1ZR^-|&93%}w+ z=3jV~&6S)Wuh?j)8@f{${JdR6PV_#Zg$?T+bJ1SNI%jtlM?S2Q%k!yYcZIvjOltcAOSs6oP>*h*Dwb_ZZOsOU#{bTk5qV{lxp>SNsZ{(gcFdi!fM1cjh8yNi z^bP5qxaTcX3=)xGb)rq4Kk;)PuJRK~`Z)Ckifs1=cJiWUPOHUk<S2}{E9&6CQu==M zLge`ccRqxJ9NpSi0`L7OgC{%o`dQVQHn5C-c&vi8duswRGCm(!>=>B!Il0k0KFQX< zw6d}8vvVYzd%X21TwC8a>(?%D zmwEpY?#&||o7oHJm3xvPX4Yxx_Qd=vXMNvRqtA9${)a;$Wm}N3DW5T-j=qy8*?4CQD4NX=CTByQ8{33sEZ0Udp5STRw*Xy3WE~95TOVX(G}rQ z5US1Jjj%+uSJ|}ZK0A?Q&VOq1homO$Z>0d{DUXvrhtj9AmUq3$ilAszh!%eHo0Oeu z>k)%bgy#dsvZH`S%u+RZBzOOTzub!k>oMbN7(XTUT7+)-+TheR70@yUvbX<`LD)bEy2k+PoJE8mqgwiR}eJcBi zo{;^b?cL)?^cPqf#4X(pKcSog{$M9G8uWB~UcgTqYuCSgX+CnO97SvYVEH0B(x^~= z1T2Ye%dkC$MgOn~ft&m!&NU8ra4gx)5;jM8=)SQu@%5Q4m@owOw9ljaoGx|7^H;S1 zzFQP0&J)@ehACj|186`UaZfHACnO^nXrbdA>wrC;Q9XOnHtn zUK^9z7{464lq3I!^+NRMA5E2#k#_fup{H+jk^<|gc6@#+3is#Tp9j@Ok}gool}|6u zn%82AG?;m77P85Xdjtm8YNOoge+U{qCtrD|AdW35h|H4Q%^U~*t&14-Nn*9X^mTV( z{dgd9o0cgVA|taVb+c6fjOa*R$BFMo1ptd(|{O_HjJ=*8jQ>1J4;h+Q=C268usN(S z-xXNbgp9cxqW=!?7P5BoH~ZKt8H)G{Bs0FNC4?_{35411o4VHb3g>Cvi`gfJZx;~t zQcGgbJJQYka@;K7>(QH+XyFqGR2^oE`V|f{lScm;Oz;RbJMb(zQ=upH{gxp~TD6ZA zxqdw=)2$L+>-;z$j=$a*<-O4z5H>A~T)A(|E#p=YI-XEb)QiV$SSf!BJd#sa(sbxQ zER$4ZwxPckSt=M}`oR=VH#2ZiOK6bYlhdXh_B9k!L-ovcikx6OxM%)p%J|unKRvBB z#K&o33g(!tXa(FLE0Kg$@ z^HS@wEU}`wb+67+>YI&NH}uvB2ZoK0vXg6|JaoSZ9F;BAq$Vs7KO1W&RZ#5dEcmY-ej;ZynLbX)fA{#9OkI-Hi*l==_#%5me&7<$gH?z-67#mm7*`V*#2K{7_fUbY zE@TfFpu;!5r>jNdTH_mb07BV`3{YV9<(xL`_ZL~dDd)+-tJ?7dxizY{f^~pCQ`9-w zhur&~X*{$4m+zobPxtl^v|W-YBfR1+?MZutp2ORzmR^r1MrqRVxVZ9_-{x%W39kqU zD_&ohRqpoO`oEZ#lt^NT;f2$WQ!EHH|NHR~j`ipd$)P&~2)M{w#tqdw&>sJ=#jt?| z4uwa2EoEa$VyKo`OTp+(oS3 z(3_f2YeBE=u(;Z0AZ|d76XlZM&2*F%%F5&d&nf=&! zo$U6raDpmtdwas2_;APf)!*?`^LH}l-oJfu9g9R@q`$7sf@MQVG-RYv*aw@a`1?=Z|@gr@F_;_Qm%6<8uVOd~D;yHI&yjCmm5of@VBlBNP z*&TaH5jsLQdqV2)7X|iQ{l~#>r4a_bcczxGb>dC)%i9$>3UCe%uUtnVvO5OL9v>@f z(~BPKsf=Wcytju6XMtM6lBkP}0sWi;hJ(6Tg4?1-N61{Sa6=%gPH9j0?qPoE+`CI+ zah9-zSRy7I6$e;eIprQaw+}z2A$-4=y&Ec66MGG_=|d+x)cVcygS7goOJlx;$>)dj zdb2@$X&>$pF5WTvi45jqO9krte4T5c00h9?Qz8#=UQm>($h1g2R6^O!c+>mg2v`!; zhf_F5vBuzL-aYCxrnDpEWFf5&roVO-u%ePN|LmdC&|20z0WeRUtOIwN_&$wJBJC(*-xlIQo8($@wpa#%X^+gnz_l|j1z_5b-jSE+S6U(P z210sf@SI!qzakENS?yFz(KFHp2GA?19jp3F_$E!1Y3QB5LoQq_2Di5Fi*#uoJj;`A zz9;X;*awmEuKG20!5;V@EM_uw&OF$@i5C16BQe&DLI%n*c491(*V|R&#RYCPU}^Nm znD~_d&wpo>SE5aJTgiBG%A?yy)aUj3b2~T z?T+IQ8>)h5uY_-Rwts~3)`7nMcM;`9Ju62v4dwldQfEXbiIL{e0E%mj8u5}?7=Khs zEg-cFUcJNAB0mrO5#y6{^DD168(4|(sm1957U8CF=TYdvL%KT` z5@xnH!7E?oz!P2R(*Brhc~Kd?eHFZGuXb#6_qmc7M)m?^Qkz~!RgQhG>pKiNj1%;j zzbbrCI?#fmp5Ml++I5-oU;lQY6Y=tZwh=o49nmF%MEEc-Hkk~$EXY9^^XcI}A`-G6=Het3_6dYlki&>#CK z{iX6?)3Y-xsI9X~Jen-fjS(fK`sUxNOMr}dwYe{sLAT7~f{4XB0fVz0E6$UTroQO9 zr!{WRTi;vwwUs4r)VAJf{vwU6;M$m4?71m;^vV~lD+?w@DPSJ6JgI!MFIfVko-Z-B zc&<5X_Tsk>P2hY|i$+)LECE+(91;1slR_2$S!TU2-vaqEhmE*x|vWFnJ^6FN=%NH^&@{%F5F7)Uz5NEm>D zzuRY9SwKvq5?a{P(zEz$$&H{hv1Q65LS`lUrgZV=zh#K734iKIUEqmhas!_-25Wzs zlky95n(w_4WZh)qb)grJ#!X};ZI>f0r*74UAAIm_p?nkRft9>o`&C#_?Gl|JD<4Px zCUVDWQ>gS5CPC<|x?{!b(Z6OiI-y>8p1K3;cNY5;?y`+~_}~>ro9#Nnx%AdVk?vvw zLE@0H-|(0`ieH>GxveurK|`37nYv~#MTvF~j`aq`(Q~TxGI#m>lIDHFt(&(waiuZF z{cSEFC`8lDJH}$XxNcoH?w&1-6BjmgHi6(2bXD$@{}*0t9_{OH`Z|#}3(mIKv-|w8 zwXOP0W!B^H0o}H|qcOAn`H*~@vyPsutA8spT_54zj%@@|F;2-QS#U6?YlFLP= zfbO|3$eWB`q;vLuJWDS@9$4f)5fd+EmKL?)D28j;Uf*Jh5UFS8hcqG5WD=Cu&P~Yd zhb4^Dvl=Aq9z;FY*ZE2`Cj1}{Ga3z*cC2P1(-@Y%K9&Q%Ke(8914%+>#d*w!+8((7 zJYAzcJkz*&5O|;Roh-3*3p29>KJ^)uA`xb0lY0@W^d+C)Mc!;sqzaJ+<_nRUr3jD) zih&oR7^mN=Fc4Zt>{tP{mq=3GIU?(@uVfO|aE=SAO3&qn>P~q(K<#o;BGeGN=dFZ`b^=%&8^J9UVI+jykYE zx(+l;LMt2Amdgac7<4CDKV8rG2E}S}Vf?m3s|3-5zwNBGp%1~0NgeF?-Wje*eCN80 zA2?mOnAdAfCcbT*=S{~WwVKqCkO+IiBO`wdi}RY*gfjQ0E*Hv!2_!sDnTY7f{T_KbjHU-IPE-p@a+?d zAW!sUWA9B(NHR=f1Bbs;uOt^p<|N&fTeA9rQmBEu^+T!5>lfUXd0kjWdZDB1wQ-56 znTtKbpZel^NQ?M<-8D99@9b7Tv@+25NxCrpP!^H|l7&eZrNR<)l{ef5(O+jI4|f07 zk6~1anBm0BF@BfP(MD5rE_vX}MUSxx1uc>Xe%^^~4A333rbUOYIIVsM&}P*S~k$q=vOWS_~66mw5@#_v^>Aeb`^?j@wX zaRje+u=-$DbB!yucJnGV*8>ak9JDC5YEklW_WbT{2lwe$k((1^Ki9o(ABIqNt(rr3 z3rS#wG|f(A5L<C@Vl3zJ#HN^jF^ z9n-d!qkB$>{nxSNuHYeh@rN|@>?f#)Ziy8K9$4RN17YQKGB0+efSu;N5Z8FQt`4Gk z#|JZyV$L&h7vHy-987f4*a3vsoa1MXBF7U++;hcI`Ko*YCDv$iIQoJq_fU@u_!x^T zNx`fm1bmFc%>?K*=^l?k*T=i!JJ)zK;r&0bkLwnTP0-{LKw9{sAO2awyd;*40SR|1 zygwg)IAHn$^Bw`c$|RR?uMBmJ+Cs`eXVbviK^UU;7GY&UC>YrzPA&I7&%;%WU7J2+ z=5~j%wlMK2X~DO2Z;wQ#)w8XSxQTM4U|e={Pxz}MOzv{uO|2IB?R;u<1n28gU4*ql z8(Le`BQcNe1~3qOfo`Q))}|!PO1q8fnYT^YRGKgw_4@xmHeQWxpqod%+I{YA^B?BG zo=xO+7V-kW@MZ+%!~*^7!9r8?UrH#4CB+3P((Cjep1RSfQfdE{{kn%$(?H*_Z%<`&2#k0^;X6wef?A zkd-L&?(dW4P`_sn&AO87Bakxb{?fI0mP9y^s92h)p!Z}S*6l)mg8 z##X)}9`jNr5Y}XJ3Kk>8kZI3XQQKEHz&OotPf*lP|rThQ^cYUgCn{ElI zHEK@mf1$>RR+T?%IlCnp-Pm*DKK1u&3H9p*ZS8p;CoFv2yl8l-&(;KdsMhgZ|G}6? z>>>(t3FEAd>wD?NHQs#H=0@@;Mp@9+{o`rY1JoJMau5EPK3BD+4KpL-d+~|55_|H5 z`fKwbPBQK>eal|Nv5L2e4}OmMN6+|>8k4QXj%QdbN&Yc5u=$Al-+PPkAMm1JjV&Zm zu$QNTH6r(>aV_+VvX9cQ?ir2eUU)QRd%DVo{2Ke(bQbE1ew=Vy+`#E|Mu~_v&@l~8 z&+@)UbeQRsEs$znMr{Qd%+OeW}5cJvHR7rIf`14WCmu6J| zLm&~=+(R;a7DK)_H#>htvzYw6d2c6CZQ+4%k5kilD?rU>FF z8)j-t_E5zr7pe#jwYM`&@Mpn5+PTj|;2(2tM~gOBGD|ZEp!OE1&1Wbb@`FcQ^_ynu z+I|t1=CSq9$YVI&Hd-Sv4U!ICka1_&-CE6RIltdu)Os$ns?kb0%z-|tT#@5^al)0F zFHbl61E;w$OU|CciMT{qJBvh&YvukN$M4O*mf{qteIH$U(mCn6A&u9J8H96FSCO-M z9fDB_N<(kIUho(E()iwD&Xf;KH9h=n*%KQCBEw4m-r&53$*8H+FiqAwAos0Vpk!#2 zF#7DpgJJ207Gg`H)1UWI#kfl6&;0#^aLstmQt1ucS%Bw<*y!>Z)}mu90c}=b{v;&r z9Z>&4>~-vhRWVzv@?%j*F#*b}-%Jf{3L9-EL`!W3bC^%F)E!TP+$!~XBeq-5zExm; zsnh8sYZS+Li_v7f(6c3;nwL?8d)@p`J%JmABER(vv3U*~Prri7!x$uX)#r6k-L{8L zAVj}UZjFV2;-nw%0MIai%>b$|mZeE2G9P7R7rYb1bRHAck<1!X^tQ)gNXDEmJ)J>|UC{7ebk=SPL1gS}WVTQLc3W@OEFC6Q-6a{%>4=sC1va_+( zg;|850$bRJqixM^x%SkuJ}Psz@r5b3&K}N0lYSc8XmtNw3IaaTFrr-;C1|e+1ul>G zdbLRUi*Z|?t%I->KVZzJrouuGBPJa;cLfG}YCQI9#|C~O;E_6|^!0CAsE^4M&RW=? z7|FQ@WlVPSzq5gAI5WP(n&|qlD_ijc$}k0=99gnfYR&9>C6|>T1V9)n511 zQf#HP>H6_#&N&_o1_1)@L|m;p{RPYdPVg#$S7)ZR0bh@#FNJ<-MbYK7c@3jY9ctY4 z*w!qNPfa>);jcz;Lw4W4yLxV2^4{-^@{bHp@S}iYu?c9PwP<>fc#HX=k9=uqb0#uN z%>8Stj(zAL6&2K!?+4_cJ@)Ktzt0JqNSlWT0A%}D=3q1wm{MX|91a4U`q-7jzTYs$ zG`M!(%^l-?y$N1>rSRf2$-=M4?i&g@@d{t^8$`6cb?g0!2=Ab-36rE`3S!xJf3J!8 z*I|!LPA_A0muW(7K9# zddAbZ58zEUPHpucIiJ{ryu^3Or&l7RH-pU$2~J*vXMD?}V3HrCFAVmt17VZL#Lq{D zb~21=RllgUGh%XK3q4yPCmJh=TV4wN_!9R(;+eMp+a3Ok8)}#tRP-Qg?5Ez-1v?uB ztggT7MszLmHV7vu_Jcxy0r#IR%@YDR&3wX*fpbETVbUDYBiPC^$x?k#@9_I!q-yYt z66vJ)fb;TFwLq-jIWFCvz|Ebpp=LK#7l7(0$vS5HEhGkB{4lZl8aa-G_V)>Am7Cae zbNR49aY(8A&}HUkd_34Z8i~!QA~Kb^6E72nI_u2m>l4lTYo?Q&M@!&B;Uojl2^}-}I&w5O{R$O5!>Fno$3yfu;Ie<6rY9SFw9E_cLu%htmykn!pe= z@8v(S<~}6NPt}ayi(hl(D6eb{PJL{?C}#nVVi*@6*EbPhm$)kCMg@?(Y8Bj}EV$ zhO^}>p<>&TTQ@c*0l&3f{IO#V3Ce_gB$p{gUmamsJ{}Z$ z>1j;;$Si9MUiy4&^lrnFG~+d=sDDzM=Q%X4zgX^dP$ApBw%u83Tnk}JH}L&9XO&&^8UhWB6O&X|xgP4}w}=NO4JC)b zP%%K$NgZYR3!;T7d!kre(}$Z{00`gHjp)G=cq_fqFnMmH_&>uhCPt=!{$g=tEZQ{d z7O`vMnVk#geg#k%YK#jA2nJ`}6!fbfTg4kGCq$K|gupcn(#ql0bHLxH_&oXmpU6Zf z1FjVp`yOI{%xh;2trmT|b9+|f^;e)ufRTx(rhb8^|A8oppP7q6n)UM0t>1xZ zl0eAF3^ewd5KH&Z?G9$A8G|g$0lIjA-{)@Q#!&y<c z)PG?P*A#F$pwOHA1(6_5`0?Z%{#|6kq07%*fa66B5_vOmGdikIi-2FeJX_kRAQ|&2 zd@$DW?$>M~2HWMOaid8=B1?b7-9k5qC{3ztDPkF?iOI|g&ULiVEUQ@V%_5J_V`s{K zL8vRv;kRF(H97~Nzx{h`6^}&j|Hiq}KgLh+NvMC09*#m_qmg|Reh>(bc4GY2Ih}m( z@eua>R2Bwh)wa%2cJP1W0{%mumpOzzho{}0FNOJJ26p3c71xM;O0TggjE62~(4(^f z>CcG)ve(3!7$1dgXr3OEZTS*^^?d76U*yJ=?L(F>eAoR>uz+tl{(hItGW@uN;&o6@ z2{3H}xTxn_gak@792>mDpc1#oLz0TD+kJih$9jgl4Y5n(979lpA0E}o7Za3wh(K;L zU{vW|H2e9~UxWq9a{?%ROX1ey*Q36M=zgWViNLi>;p;U=i%pR5a2OZ4yf3^=rG`k8 zO7PG(*_jh;bEgKmYgRJ0!%f+TA=T6TY8n7cRQuI^0+|(MKBWT-rtr~DE*mkKx=p_G zr(4(>q7I_k*SGJkqfm|3CJTLycDANd%7sVg08@%+Rg$~ZiAhG|+w-^L!UtVrm#%#` zm~^-Y1Kp4<-ZS$iJG=?@CpU!O7+LTbwzDHR^MuK>chY;VR0Pnv_YD^J9++KK&KF#3 zOkxTY=4&Ocn&4J2mO(FKd7B;6x-G-c8pS|b}O4(tr&Sowl-ct|%H13PC7~g|`mCo#AUAW9K z!V_uY5SbCol6zx1cfrpd3Cpp2Yk7IBVmAXUt%yk`atcE{GoqQrO54EKoQ~J6nDp7z zBYn>{nbaPS>32jALFRajF@`;GDGzqf+J%ANYR;h*pDF%zeOHJ$JL;?j8kEyM1Go|@ z-*!=fUT0R2H3SWp_rXWj^Sf>+MAuI|qhD)OgAWX2d`B~)N^!PC&&m3LZ%meKKGh=p zLs^s~*2SaKOZNARzaaP{=dk+(H5l;(hhoj(Qq(cm{h!a%LJFtn?dnN~Arh9jVK{HA z#F#!7>S^}jVo4CKE`{^lPfv+69N4V31pXpe`*LOR4fC%=XK_M%*%8W>UD;;n2;O zPes7`Nc3Et@)wDLs@d~~YA7c@nmkgO_K}q~xHs6!?fFk=XCaQLh_;7!&sUK7=p*&t z*rTlP9z3#G;YzIrQpzGJ%6Rq;1OKSW09u+V1}w5S65>|{zaVY|@W4(C~y|J#5%3L0nMk0@N>y+!E<}Um55eFa6tZrMjZY3b61BT)!EqIIW z!)I35Uw~m`^f~l^*3d428566iQNHL3^!MVmo+w+3c@ct1$u~nAL%D5kiJx=z%Eqrr ze2K~M>(MRFWv*4PsNQ|U8R|#?Y%6^8ppM9qg3dbya z?M0mus!)vVDFA=W3nX)4tSO3tGxD1nn%mwrkM{eUpO%?gpHRe5Lfc!#bb7WuAP~tk zSx9uWzqw?51c~|F8_DNWub9c#A})L%T5-=q-1w+qy|bcGP}EK7>U%O5QTWJKnmK;X zVChCG;Im97%Zv>ng+(@g6_Sja<*Nnvn$M zyH}Y+N`HG17&cr=$9*-N&H4p$U&cNBJ6-K1A(zzL?{_}Lmb@^j(5o{XB7c(9&{s#!KrmZhG*{ZRAn@*gT{eNA!_B(S#cZ zM6dAdD1?6+ksCJPjM>$kjQ8HLN)gi1tGKDz^o+?p9&3O60|HU?L;Kipzrq@Se|J!% z@n3JlNo|d9_t6vK;1|JL!Hbhgi7@K)U7QjFA4EW;<_<9Z3BBX;81WlW-n*&6SoeFK z;WG#HK7o>>V{YcRN0-y2v8yy-5wHGuL`y%9N!d?}7?*m>)FVt{8XFF+N8(v(=%cnsm& zFGj_(@4xwytvHd)ryzn~ccd5f;=cDF<8VV)CIMur`3`N%98bU*JNL%4y%UXe-|mRk zUV5)~6`Z>qhzrwLV)6~U3(cJ*TuK7yIOmR?0siX$NKBt2yokX&EsZUkP5;LOE!etA znL58t_VH2jlwU>_E=tsvP#r129d`qR zcwv1!}McNC>>y1_q&1gE#28LD#4uuJ5#)vwirQ>$C{zo|&pDN+CT<}?OFgF&87 zc7Xt{-|Wq?@1lCRD&#ZF(NxZ>2hs`-t(z5q=WF^A+DZ1ShX7X}dxnoQCQO&lSn}df z^Bz7NL4Z(wEKif?-n9;Bf&6b zBouwC-b_)B3pw zuT3niRnQA$IG^S5DJ|+#y}9RXg}d5^wPL6jrl_A8?DwmJc^cu-8?6VzYnaf@jQ|+p0#HXpeq(pI#Sz z6?gv>nd5MPs*GV^F~&eYM2d|CeQR<5H~;MMaqUqp)n=Qb}=d*9xTF!8op28_ZU9Zci#Knr9qy- zE^UETIOjv#N@1BJE;)i@((7h1m_vCI=mTIj`h~{<2kcEO;CzP_=q5>)Wasl$`;2GJ zJ0iiErkT2z%zKe4^#0mQc-&_OmWF4w2_|H2G`PD%W4(|ml*Umvmy+J9hv@gl8G!S@ zTxmd_p(PbE%=usPOq$h>^Gev7lzDA)FPe1G%7G>0~a|zqFShBYr;+{?%2j!b~ zL;oi4Z-(iMP|q&L4^g#ljt>pL%rK9!0m{y%k7}yW@e0xt0axo4spNeYy4W6Y;EU{Y znJtTDQfxQMt+GB5@te~H(%OrF_qM*$=9NV`gWTP5e%E2lK*zL0Dp-i}vCF9Wx=Eli z=d}hKs&`QJ`*TH7_s1=nZ(T_s@-%}%UTdHV4270JP^t$Aj|CYPxu_NMH+)3uegWr| z{!n4R1#9=as}VeTvl5ZgfP6I@KvnD4IW8$hNL7=XNOrWw4n69EM>bO1gWbQ;};8 zlm;oK8-$@tx`yr?X6C%}d)Bws_x$m$cdhfs<{UVzv(MT4-ut?*>$u-FMuPWiZpc3ihrF`$bg*yA_$i*XeI@!(ozSN@jb9?zL9N4AwHj+XTm>O3xD z=iBW{;+tT!wp%$Vl!|^}D&Dl#mW(~k4vu*jXZpOQ90Nlby8$a1F4n|- zF!#e_1p7{EZS`#AVzrxqh&b~%>FD`24Y5$~0Z@JtZexSnhmUMICJPuP%Bt)C-b-$9 z)_9P6b*bhcxR7C*e72Ha5KZ!FOvW*bK`*%<4^&n?W>V}=8LDS(HEJZO=^~x$ceV~-1cflI=L4>sm!xHlwc;C>RuK4C&M?#< zsj-M|g<`_Fi&LcZ*_FZ~QN2f3TfDaa#pyh=X9Nho5OAWZ3Jl3Z}|N4uF=n7M<-GtQFlAg`UeAXOh1rx-1<#)2)3$c@%S(qWD+C zmTHom(>$CacXEL&yk*wjmg-$N-*9;`^=ecX0eh>yQ!k7=+n+(16X=lR`f5bvedq+| zZA$jFZH9?l3@_$REn@)u_OvS0+9^S^1K)+mqQI1(Iuk zX;Nr%wpWavBA`MImBzxb)HCdfOkNQMiLzEB?mn}|Q5pN0gxjB$#Z z^@0mk2X&$b#xjy2PLnI0mDAA%5|bwq5AtMJ1?4Q$M_Jbpb3fARz*~wB{(N_7*bBNg z;`Hf0{P{gmwRpz=>^&QcnyE;7 zv*ZNmDsF6rRsKGk`eO0#xb=;xuD}GL=4VtCFCZq$iMW%?pW}=uM5zb>u63A~UH~_K zUIBF^31;L)iM3=d#rR;?)9h4^M`v2+lr!YW9Et)l)N@Q@qWYm3_S9rm14uMsM+G4R5nJh=XoaMdX9%67HYPs zpcx|?{3U`qH?ff!TmQM54Cm{#T{)>JkC~!Np}_EtJc^DMqHm5nJ0R(dWv0KxcB`C9 ztb(l=(Xmrtjg4d(DZ-VNZ8;#hhx1lo++0yA5J|%Fhba4st{= zM&MA{xFXRLeHU2A^|`VjZ%PCbM2ub?F^3DbzMsp~CiGC7+_qAuD3#4HQ7;F`_w#?< z#lOl})-*wXD;dIq$>hN6flDDRVszM$c`F9ksH?fE(Kj)VA?eYZe+S8Rka}kFaoDBV zB5gC5p{l!DMK@z#G^BjO;oT>d(6xq3G>Sz_8A7h|bOu;Io-vXWRxHwpCD?U&wv+F@ zs&~j(zsg)qxp}W;{X1{_`vfQL^;i2zO)7hvgEhYrj*Qe*spmW^_wU;^2>*FzvNlb; z&vsQpf<`O}rQ0plIew>o=k@dOk(}-0e4k38o%{$QtSs1SRF@ZYNB#EZ9{(x28}h#w znmT`aTKZ$f(0E%`YBM!$8>J=FcxEjIDD~gW_ff{Q)M=k5OJ>SF`2aIG^U9M!-4V6q zs7Q>b41u?v*=l^uFkvSLTz`E%d`evhuuE(FT4E7!Lyb}*qq60IP4ey}X0XsB&4rs$ z2A=Ux0plHO65b;imLsU?S`EfRjo5=0?{KhGYcXCtr4gQ|ajtUr2SKgMNy|Bp)__k-! zWZ%U7P>-!OC$hAFjt;i=@C3*;`(UeU%ki9(;u`8iy-K>}j15GaxN{Sjx{FtaBS8{p z&~GqoeMWb72p#11(OJUb#)qn%J)jm%qqhA^?Sp!>d?6xQp_rLm1>1Y7qHc|O>iw}) zXuOus2Nfi81EI0wX$2J8+Ze%vk(z)kT?)pBm4zwB7rP43tmegVLv;?Hk8WR?Sr#P& zP-&4Dt3H0WH{Ih!O}kdlOl$e3x~Q@wh}fNA!| zoa39Jj}LaU0NfX3Ma8g}9~!+IFC$GKtQGyCd63F46CM218>SzxEk|So%ISS>Eon4b zBSf85T~)!B+6qV|n#XH!QQq(8z37Q&ty$m7-W6thw|~$>8_HldoBTE1l+TB_RRw;H zC_PjxKzk@G*0|xOg=#Fn=S@0(V>>J4)ddb8;4FQDVOp;Z5dE$@N}V8e75>M*3TaR? zE*om|s;)jQ;vC@YAEN5)I9Y!32Act?MfED42?=^vs3BNL^i7PTZF>$1 z$+GxU#KfyM^VM#nCtWJx%@$g*!lQ+r%k;3!Z-rhHtW$IdJ1z3K4;@~NJbM+}p!;gw zt%AV+mdWO|n4lYz;Jr-tG0RKwHi1mOF~mcsV5?~e z1U-sPr%BuD#L-5Qeo8Dc*Dvo>|;<9O$kU+}EIJj)vs_jV(bP4t1w&W@H2V&i`ed-XVR8#hka zcmd$$vewb@COjHIXS{!eCUKN-#9RV(%Jfm5ylTn!utyo=0PGa#+2TrqU`V!gv6Hma zc?^AA@c-0!rXGp)mgFheUQ}90-zkZN zpjrM{skWk1U`c7f{lx~|R}_6$tJH%D&Yx}kek=oiOj55sT{gyih*$4An=)_Gf6MOL zT^aVCd~yG^XVsp(i?}>Y4@N-lW63={NnbdD;dMSHf8plq-z+u+%k}gqhOVdnL7z8p zj~QEC3_$z3^|T#5GtLB;r@9CE{nrquobgT}Ujj(}@O;za%i%?b-)Cv8F)JA~TWCVc zUf*9CBG?)uI_}axBquk9|D%5Zo1qt^*Hj*;N2f4<&wR2TsO+z2q#xAzhM(h|+R`nX z`+`4ys5s-gwm{8VLpOW`|`ba;}SD_dBJm@p$ftRKcQL0K&&%*{S_%pAXDRIn` zTNwCPsN}%PA_4x&D6F5lT=^$3?Lz;jAB5h*(aWZP%>>2MJTw39y(a$`QE zi~p9d8vXWg3H#8fIr^OcCf93SK*a)~7Z*_bAJwua?JwA0`R5=8K(ung;XeGX#%Fje zy11MMvi$!(09J@ep2UHI4{5B3gg?9=)co@w>qi(i)2F!BcU@qJw%M-LjkNllif>UN zNy>eu(#%##1ZORnBvC{uvrgWVbbQb{*WgcWvFF3q`58v~!FZ5w_?x*~TtvZjl&A{o zB_!2?!Mc0qr(>%a%9YygNg&#jw+-vZ;^v4Y1}k`h5aXCmF!=WK8~;ahbAe=~_zK4a zA>8lV-}l@zM|Oi@`_AQ=a%d zB|}GlXP%YI8QLoFZJ2%ILxcV<{-x-(tNO@g=nEb>anNd~4)~Y4lm)7e0jXmW?w&Bq z?oC!?zfDCbQ{6G@Ip%A9{a@A|!@0ci5c3!P-Qc8VCYvk^cO*K(|9rxpm-~(#_w3zI zNh{v$7lYuWT7Od`3+u4o!d!?H#d^I7@2S#}Q~DUONwZ@M?r&gqOcn>>l*%)S%&g@Lbng@VhhG3woQtu3LsUda8ORTrbe)^e=5Ptqk6L=tYz ziARO$^vKA+wBEt-`S-1GR7;5U>PnevMS@fqyb5w%*s<6Mj~4ZnKg&tCX$sQ@IlTJw z@b*0wW`m*f%$#<~ou3iW(Pji3{~viR-d6B*osh&P-54Hesnc#}8^qtV%p&=&>^#tw zQ0t!s`+>sgk(XS9TV4Q#@y_QVkPqgq5wmX9p3nvT-Y{Yi!$n4w|P^yN9m&j*Pz@JzuKQ>DjOpI(I#u6vBL?Z2by#9=F$ zDi=jqAysvpi@Z*{y~2%8Ik!pQP;#rVWd_sW$joB)YlZxJEeqAx*;f7=@($w_wUVLjdg;DaHHgP8lAWG~A2rm*((Y+!^r-08Bcz5>L8EzrzR@^GQ@aO?#zK+n{$fODy zaR|5IRw4MGQ=@d|)dE3^TY(s6OrLck!!}Bv}9@uN& zh&-d|y{Ef-&wFNr8$Z(et38`6H(?U*=-x;EdmC0nO)oS8@>%k2B3j-UeMR0BJEwxy zmBQ#-Z5O-)Y!2AhasT5FUAF{|_ZCl2-Dr9aUe+ja*64>n0T6|=!- zu!k_N>{?Sv%&J)Dhj!+AB1BoBow$_^&(QvikQV8(rR;>X=O&St!a_B=QwIvaHv`Z1 zu$}p5Qq$NfovG(on#SC=Qm*jp@-&$9OULcxxk%#@JyASWvraOUD6R0>RryBIli5(n z+mflM;KwF=bQg7GF{;u*IJSBqub#n6A@NsNhj)l+l%D_>%!Ct za9Y+QPQ2dzT$o#wJXYInZn578z44h7!8TZTli&5A+zif@p1!OpLuYWt4|JIEEegSe zO}+k5J*(XWU;Bt8X<=$ja8|%Rc}D$ya}#DW-5V)P9665f zK&*KTd}4_#v@bA3e)j8b*sQ5Q%#)VS*V`uH>2Z{WBXKOQ?*qCI#<}vZ3&mh&a;{$J z%`Dmq`BIoBSLm|4+BHlTiX3rEEGK=_ok+32YbAw73lr)UrUGjhP*DA+>mam$%$IOp z`>2ON^mim9_Ko}Y`fU$%U6A!D4t+Tgigl+;2EL5w6j{jdkP(Q1Da2LtgcdCv=_*e0MM z!|8#k9{r`Oc%xJQKOXGvV05i@^gk;LoFZjWu9|coxMLs($PIx5@YeUsAF<26#2e^q#k}dW>HF=WmX>|J9EHvzbtnz_c?;Jw z|55sFrB_ow+%hxdeYWl7fnVX8nhx2!0{`Nv``tAA94oKjqUSFMrDZmn5$6m3nal35 zfyBQbe3b6f+mTc0j)rpFVdCFNgGtc_$%l{wXeU$1r`0#@O)Lu^|J_CKxagFlRO;>% z$MzsnwbN}44FKVtEy`5+)P^#H+BBf}k5VBVguWg%q3|BvBDlxEQaXh6okM+7n|bsB z_Hj@%iRojiN`M@e@HnL09GPGMS*AE*=Y9L%wRx=fKK+M~*R6e?h?SN1ydt}2Gd~}P z-zR1esVoPGd*-YZZPF&3><}j)&q+T-5!{+3Eb}&_vOWNmVg4L`*U&3?KHm5$;RIM{ zbMmg(y3fKzMHAUDXlVplKJ-1}EJGRoL6s$ePk_ZhSL6F@sM!v7^acp39g1}Y-MmO* zID~ZcVwyg`hWbY|qFR(%zGl1!UI(CdLN;+8?d7LsCX5{+9%JZG-@~K?PUFdcWS{+6 zUJ>==FVMMnGFT;88}{^tJcBff1rN!ocE_?L>?##flwWia>qm7>HwF~=JMW50#<4r ztpJ_)9*}CS_xFM|$}wgipesbdn2d{1+|0j~W6x9y{r?8Lm&$CQ`PY5~WRoX952f80 zuOMyf??=5YP(eJe?1cq^2#A*@Fl73FkrK+rpjujgtk^kfM z91G!F*5v{lxLlb`5A1h+Q&@T?Dq&I&Qi#fuX((^!fJjktyC~mdnL<@ZaqRKwahnW=r zZKn{=&VwvIz7WpMEjwnJ-~myT4`X)_I$26RSt8R)FP#SL>}9ri*(&vawzF~U<%_@m zy#+OR-7%@2jN)FLP#+0Yoe!JY7`Dq81Ro94q>vWaJ1E=@hhJaqx(BJj^IiXPC5?mP zkypvzeV%}2#KTu+K<`HzwUAK>#E+BqNMT=adp5723I&^pKW|5ICO#4$1z|{@2@-8IEl+CUZK ztP243Chg$J&Bn`l`o3nAzfej;5?Fow06u)%e+5}8MlI$%)d847VEVc{>`xUTR11S* z*D7u=kkD7v=#&iI(?YOwckh;myK`x71=U4DOiXhkq-AR=1cJ{9Tlj=FJl|U-dKHWav&D({xP20O7-D_Wn?u_Rh7?DcY}`?bh9-%K zZlWbNn{#>H+F=~q*n#6Jn_qpeoEc8;N`?o=U$7Mot7K!>yu(YE-;wwm<@pLJdL)s3 zkU#kH_}=u*T2Cz|aMXwW%J2$0Gr)3r5x^@8s;@&=F}eJc(0&nRd`$nZt!R5=)feQh4zlePs8`6HoEm|mO)x&FoMN>8)uO_Fq9uFdY!C=a}>S?{qvi#;FHXZWUOzlaS_4eo2~!sbYe70 zwQdYs((CY+pRfW_Jz*aIiG3)^?_gzv_DQppq05ULsg9dCzi%@n3)zZMWjuNsXh@cN z*CgI8T^>|0F!vDk^FKx7^X;zBB=eyVl<|-`SzV^xti7a(XpDoy}n~l|7pR*FA2nuy8#dRZS z*^ydXfgF^wvATBhB`x$1(8cd%9-OlOD{`1#M$HQqadDBdQ3hxZiPJLAS zCgb5j1xsa25UQ2227135h{20=(-RI;FacD~?-NU<*^rmR<}1(93jiR7U{&4dQT^S1env97vEt?Z-GX#ohpDMnp&|%xsQM_u>~y z1x;fse3wdxOh9`Q7$G3&0|iAuLLk-L1GZe4z<<;Vy1Oy~uw!e7E`k&-rwb3-=7$2c zbpX_#gU#j+Bryg2g=q2)1KPaziD)twWzGG_IoPQZi0Q9R5<&mTEyO)HCK=t>hb)s6 zGI#!W?ox|Eog{4&vhbhVNZJ-;Arr!~9<#vz-~Bd0ASzY#{c42)CovP#Q^;Mw@5!1- zr+qo02dq5Er%!bq|NB4!3vh2tP}QWKSEWfuD#~IqsK?-MQK(nMWS4;QWm~axPX_)K(Gy`UVimFP!zVoZ+m*-uqQNQpUD*?Nh{-bVPoC zy8c+j?B+r36LO%HY%#kt!|7MnprvIHg)@$m6pOyy0*gy-M%EA)t%MW~jYfRrb9F&T z9&tNMzs|{?24RIj_o8;Nufb&%#I3lKVS&zK3J!Ynb)kD{M*PMc{!$IWJ_nazVj)yU zOh3KZW=idKVmKJ*VHhY(rm#iM9qrYOACq6V7PFrmmHdNC)ORqmU>~xvqbA$GEqu+j zY;XDX(sX#zxAHn&v|f%*sW;rpo8;@&2LFB;$r<$5wSaQ5&~M)d+YD^rwW^VJs+sf6 z8NRWNSMzh~-2|=ITP#`cwJ{M5Pfnc?xp*}7id%2U;!*MM-nU#6{Mp6Zm6rG@wBwiudF95z zgFM1no+6pUSx-k(By6>S!G{{SyYla_|_CHYoT%uno@=D z>kmTQjmpIu)X3sHQe_IRHg4x$@KThJO3+>K_wL2X?^{73z0!NtT4-Z`Dw`hYS?;T! z9PudWd0jI#P)S1MmVgIy|UTN{qFT=#qOqaoKjNO0ne!@ zQQs;^Y5tT3Xy!hB0d7mZtLhl!?)Qj~*t#xw3j;M z_YL;Z@awB#EH=W%%*?-BS}p^Nsb^W|P!{v5yY-9384IDp!xxiQ>7G@92ajq-Kl~Hk zg%*UFx5_1y%!VEr9Fi~omo_?jfwEJam2+ruZX-=Lzg^#1WA@tp!Q0Cm?StzRqR2(zB4klzbOvh@ zR>-+8SC{=ZPQ%aYI-Uxr_rp0zX7!1xT>8(U!V1#+uop8d&uTlWKtUBa4pX-QeCCwQ zhZ8?^(L4NT2Y%eS=9^Pq;bNgK)y6r%#9qZrajqH7P(t&Hv-{Z*c@L4F@LreOS7EO& z`}jSq-h@MZpowFDzE45f0$e`MHDBgqCO!|J@ACKdYF{)k&-o8}d*nP>dkcA!^wRiS z4@STgX)AZPModIPJa%)XYY3yKL`j?e^!Z!+lZYMMaeij4PHdFh*`{z^$JY>>BfU3X zoU}LZ1v}kmFkdKdqx~trT{RF7q!(Q?$_g3#Lj#BwQX~bI$mwFxiQPH)vE*G`6$_r3 zo3DaKNXSO_@opNSq8M_{8}Dh6+y?Sc<}4E4>#wMa==0d9-N}A@%d{}iUN7}nN_}>r zE_NJVMKOlmX#mjnJP3z6;GA;9E`H$em*(?nn^jjIOHoDJy^QK+TvQ}mv=N*6=(uL< zMZ5m1(XNc_!@uuyz;S3UX&a|67I_w0b(+Ng@#@is`j9cX43n700>j_8@>Q2}mC@=# z!iS%dKDQ*ZmU93m8y^%J9R;D@B{LxUkA+6?>;BzRHRgD2Ihuvm-5Ki&zrrM3O$(21 zGwLF4l=tH^Ujhmnf)Cvj*y@myje|}&YI82y0f8PuRUu2F=daBPwDnS5G=B&bce;kX z&AEXrU_5(?Ui^N(O=%meCR&At@H}R!3vl%${6mEWArYMP#%=pPbR1+wS#b3>eLL^} zGoIRNYvN!N0xFW$6R2*8kePw#z z+x192Ef|+AS&Sa}uH{N}gy8;v>j{mP*y|l>I&CXhvXfGfyo%OAIoN4A7FS6|FvHVe zIz{qkD-Te2abv~c7vmR-%r2iAw$xLD<6negDTC@_%yfdK1*}qpYK}1$F7Nt|JX`dY zBop+U++~8t!z^EY=3i?U>qS9pn0Oz+wZ0YrH;GN*>yrFov~y=Uv;e?vXIoazIc3}S>sL=GIB|p!=UFKhI^ORdF zHZj3RSt@F1Px}|B)}Evo2%jsBp~Io1<~C^+0qD+_LjxBHHLzx^s01jZLzKkczIjxd zGA_x46zBTvGwtlOnOvsD!VFEZ3u4p`AJnfw8?h3@A4s24a_e$ulJ+w{sffAmj3{8x zf0H((IX)nw1ZD{u$e4pG!t4#Xc|!lw70Mm|CzQR*3T$EH9EO+FT_20H^V?Tx8O%L? zx53=m!K^X-41OU`%BZGmnD0w2dtv_wLHRgNpTGTsXyFZ_aJ3q5Vtl&;@@pfR{Uut2 z(7!0Lj&$@*oeKoJKKX(F;ZgaWkmr!8JBj(gqlO?2;O>?@f~5v-2vE*3$)aiHqP!lC_ld|?PgH~l&(>i6Bhbj%c~rVqg9 zD2fwCo{uSkCKEbZtQw+Fh(l8G(=bGsvs)!U9xb)?vu#k}^}k8ma&{xMy<}d&*K|F> zm*OM8saNpqxLf8ce>;7i1Gw+$uPNwITZ-{V#htt$-oon`TG^M>oKPVd4Lan`m8p=T zJ@Ui>{a?SV#>r*v7NY^jH4_@)orl-2xk_uMw$^o|X1q3>@CUdpZ_v$OVdxM2tGsrh!TFFF!K6TA0@O zqptsHUuBlW1kX@@>D}fW7nUSp$>DjE2_CoS;3Yx*rVvqnS4;e zNO`Ra#@Cu4$j_KxCy*A|-xOh&K1{){R{kIDr68iz@j3EH>Im@EcHwW7UtIooJDZ|n zBZ^jq{pyD|{jWn`;?VeA=@*nl6Kc(YiE){+$^}|;L}O`sB{g~L_amwKjhZ)~adDbb z`LXeO{%I7cs1rkw>C;L08#cg>$q)^s1#De6T^3&L`Q!)prq8qzg%FZ9Hh3R&{aWC* za1PoW;A~Ej%zWuaBQPo6Fo1$Gzl%ZUYwb}EhCt92WfoJ-0lbc)k%VBUI3!4~sSlMP z-lP_P%1AOAz+e3G&`jeW5bDS|@S8hkw==L7NU7ni11?3S)koyu8pljTnHLb7pvSIj zOH_A*k$h;N#P@^VVe{m}x{=74merx8@j8eHtBxfnK&%#(5RspCcZVbU(u(4H*^`%} zP_W4nQdjq6&(Xn(l*F|ciDWM)KW~^WRM%29I-?kw?nS#o^g!Qc zg5^nW<72L47A}PXX288;Ifkjr#>vJYy~86tXH*1bkx?80yMiVH z%jIpkBeKi}hK^JWS2TBc>8NMxpr%N(Z*IPOQ6kkNN%$$zXIitIMx{o<3xEy)IcoPY zpxmdGK)jA5EA2PM)Hyg=QtBd{@;P?wj>P9NufgZTQXf58np^$-xxK(rNwKT)k&Gp# zPRW7}xeiV@*4lGc6wYvBR$}1%ZAU(E{EMA~o(t9reePp#+1cgr4+MF)17;M4LK;=0 zR^}zN-`zn^8Lh5#nw8#MiDlkFr<2@xpm5?RNyP9cc68E$O%+W*QPewSvzp@1qrb9W z94{`~>e-gd2~Cr}HVRfd(aS1Z&nuJ%?DFRAc zC`T`U8!V3{VZE-5};}nEhVUCO+O-#LFeOq1| z`rgj<9jf6)xbiZ^q2!fW4>Cg^`^6RHrz5u0IZ%2in=K{e{G7}=v16wk#|f5^Efv8? z-WcqNTWhm+Rra1|)63g!i(av@uEEUvk%=Q_X%g4c)VgRx@SBc?j&~~9#lNHeXkPNA zsc$)#iyCzX^-NO8975{ai=!4iDi0rov;EZqWlSW>tu#$ge?IUo%iH?-zh8VBaFPP;>Vne zGTW6ot3wFn`v&uA)7&3ny(=$Llg9ZzaQz+BQ?l?%O9~gB_yaD)M@^Qtjg5sO*GJ`W zpD}w>_koq47H%W-YAifL#Qu$s(Gn&C2!4hP{g~iz&BKcy>DZXfOe?dt%#X~+}3fD}j?h#Hni9~y6iU50~$ zsrqFEZp~i7MAcdb%YOj%uaLj<`ZFI9XI8C|0gdi*V!NwVWkGYz#WlyCopPFrtUzns zo#T~`N}V-I^hSO?zdeXYsS@B7x47Jp5V(g1xHfzbx&9~o(%GA)?RUd;C!iP!T>5HA43Ia#O90us}n;(|mh?UDM4F}>eUt6U5XfaxgYsyW;4Z&>qf!BZG4 zq5sVC@8o>%#NzAFdniu*&xAg3z!bCYESzRyJk>%)nZV(oQ&&6f2L`a%lleDSv1orQ zbilm|T|jCc)6ex316WAY<`CRdcu9A-=jfHQj+O-VR=jp%Hv!YG|#2qMPN`8Ek} zl0ee7qYl+*X+IjI0b6ORDN{n(xUN>ab`0K4HI_(H3>{Nk0Iy&&Cc}R6%yP*)^qwj! zHH!+jmYbh-8`~t7KhQT`f{YEtIxjK#?u=AcDg^QwW1FLEi}#7dXn|Yl5OSK_rKg|i z*39f4Tw(~8b?k#wm#X&U4lox;_*{%>?!D?mVrpV6(I;oLedSVKDXilvS)&=u!uQUs zbfN7|p*+l?b`b~ztk`d!^9!#5Bmn-f`)!f!yp1Bbtyruh62+`pqe7cb*CYH^0=X#O zWyHn4xq-`k*3m-6lKZ)`kE{!vALb^}Hx3^=V5P!O0n;$P=r=5P*?%+<*$oE*khJ5F z#+;ci!KtN?PM>hidxoG?4{x=cCvgI?e7>G?OGaNFg!1D3o~`Hmz0$ibU-vFf**WQ4 zP&S)q28J3@QPZPWw|++2_Wm^v8(4{UI}1M@?cIaLWskxQv;w?|1zQ!RBUq2_^MGXr z2d&0`Njd{gRltstnTs-RR_M2Oo7_G3;+Q@uo!yM#Wx38)) zg8gka>X0JuGItjoyVqH-#qMMg-SL4{I~(S@T=2JY$7NtLg*UwZe7lc8U+xt70_&{YBS=gb6j9UB4=R!lQ#;r%D zcyB0nbELb8`GScfqN0222r-j|A5L95Lavq$finSi)S@53cyxKk18z5Q{h7Q>+2g}K zcIP>OX&3BOi!zHP7%;gti!}snw+zsBh~XsYDDy@0TNSXf7Q=A|!;n>%WP_l`hE7-C zzuy~3j-FnEQ`qKG6RTG<3i6evt+E(&7y`li5WQ^(@M8llK&7DvRoE9|1mdlI7VJK{u*$-a2YieDqbSiv#UUI(6X>zz5Ic(kpAw_Are>4| zI!m>H$x*T9_KHu5M#3$VSZkuXBS#uLaS>v>_9UFcvdw^M3l^re8*6eAX@T@$U&{>rlW$Y>iU2QNq+n4LgVhU zyR4c9At(2Ml%VX=DChTta-PQUf*}7WHBv$5mn2H_5f0-jJ~tnfQrdfkYH;W*#o;3k zQqt8f^;zH9!xSbNpNy-ylFMh_eyuX2lPV#`H4kJ-$V?cPB9c5V`-qh6&2M_xL{CiBYvFn%Mq}Sk zR0~6h~ zCuF^#Fy^{+oyc!aA4O8=K-x}CO=bd3Om9tSOW_Ix5E|f$Ir(q-Ke)fX=MP+oB%2}Y zwa|7DKlt-UlmpcsZAqVh>lM`nBC`tp-SWUK-g+}Z3iP%}FP?X2tS6eTxXsDyGEWZ*hiN#WViaRyPj*GyOZ>v zi+1SW(SE36aTk%Iq7>fxrbjyV51cCGPF)JRDv_6%x{R0@}d93mN`V^$p^I;!oCE7X|{OuIk5>`k;4Hf?QrpuDH z{Db$&(~&=NaP?W|M-2>jwi`rMC&ua{71 zdDF_;Z*SyQd$=b)OE2(sZYLWH-$P>A7_0mMG85g_s=)~P+Y)K5eKE*JiRb$n12_`i z2!|M4O~9VP)LDtY9^ILLL~MQ5Lx|lCD4afAp$Mefl|L9b5ZMhP@B03Qcu$$x=7q2N zWWdz+LOxQ{T6c50oSTWQrxPthDK6beH`jg5CjZ*-i%nbAeA>gP-C?;Z#xTxKh>oIh z%Xj>%ZM|X5L?1vMq8)P@Cglq@%;Mhg%x8C+1XINhrHnc*VU+OZBxMjj8RqBjpu@;o zyQK0>Cj@;k2(PR%M3)Kbg;{gIio4q$V9CIj+8d2E>*nMphEPQMB(SCi$k`v+#koE8O|8 zYQ1|>!keh}9Q|%P+*PBZ^_t1WORn1KI3(o_Vj-4p!GoOgE#*@R{mb`K5*&Ay{V8-? zVO#6CPbH39#i)!@4vj0_vI^)>Zr^9_OXgQGx0hwTnPPpn+xY7h?6#~<;xRhjUumba zxW{eSOXkc5A5q`Q@!~DVE%*=F&o$7+E9RGWitAQK8tjt1q!kzY>1z#3)SE7R{);K= z4_$U`a84x>qq&j+g6*7`o%ac2xgd`}+Xr1*EF}_`CD&@p#W_l&Baa&m;Vo7*3>9%O zSW3=JXlK6@#aX{Hy3v4^-jv`*&1}hS=DV=8IWoUeY*Fy7c8QR|nVFDsRz@he!L)g5 z71ncEpP-+2j-;Dn+AWO%B#@#&9`?hMa}aNIQ@hfiOV>TH_BBbrNT1HgSJL$ULf2jd zh0P$|?b_><9t3Uk2Hk@d3OZJBwY+o>d;1YX#r+Irr-XlCu9>SgV!dPoIY zUd>EYQ@A0FtK7o4gK0LxvPwzHbQCCc3%4@KN-T-b zJ4HF@u{eiNx`mPGx5z*n^Y#a;JaNW|0+^ z?>#zIy2kTgKOhE?JPt|dj;Kr};(P=Hcbb-@GJymEH)$wjuK2NqUdBq0EjbtQ`SqfDCLl}u_HyD)&pk9pS>KXd zxStZ?Bwsjjbj0*k`M7NRDWKH24x>SSRXwR0j$_1L9SQ?uu76G2J#n?Gaz6IP7Zm^9 z)WHnC?&>RAiFH-AO0s?!QxJS)o}lF2BM?eNQX&5l-()ihGwS-=~yDse2(rskz6G#KdH~zztE)HuEzuN)x_x{F!R`9#ze)9%Z-*8hf>!IC7Dg zlEHdb7i~Gp`isMMV)#Xaw$zL5*W_QhZo1L1vID(0u4)4k(MgQiEMO44B$b+x}VXZc=>=gNGA=HS;P8N_Rr^DEwPFSuSLRlaD8TLV$d zy1E@X9Sz;_M|XkN=2Dz#B*tmi`}Zs1(aL@I9%sL~&y12j#DI7&*8c`tgf9~fYvcU) z%!p+|)6MD6+VSL|*}qsuB+De7a)9V(ej&s_nKRQAlYqu+ zmbQ`YMDOlCjuRGl6Bzh+lE%P_Li`q@3_)3%8O1bsPm0l}`EYvl^J?O&CU__=AE{(j zZoqLpA?FM9rT+Ep16PXmSqWYm1#ms}e&46y%-42SJ328(L?g`^!IxS%+_8p;-b!=i%lU0ysIgv z%Zi2A>T;a0vC-RG#XDlDCv_lQ&u=(O7`Xm7L+^^E5a|uJp0CVGZ34M!Fu;AEeZ*&C zD+^mEX8LeD5T~Xf`Z#;q>v*87tMpuw?HdO9r*KOwDF?eEmAdGv$nQzBG(ac|a?M}wS#SooD)sg3G|zVVlKu%dIM zgo*z#J&d$w$M05~MroB+R*pIR64X&UrQjpK7PuVtrMXz>Db3Z2+@I_yL3V(i%(E-U zc3zGg_&Zw}7AbW^Nf^Do1^W62J;3XC!6BcxXymX%=?7fNS4Q$wh!*?b6?h{(G{af%;QdqmdEZR4hC(^!q^&8;BhYX{n$FkUmv2I3d zcVOncZbm}B;i$`_-T2j?=#+mr9JgghBV`Tk+GRcr>u5NZG>}+`#(Mu-A3tnAlXNS+ zkQjgIa9P*N*Y#u&Zj-GTQ4r;f^RyBF>t)%5!u#d`rdc*c#I;5neyQP%wp`|;N z5G99{2C1RDBo&FFB$SenkdP1%fuTfFx;qCLW}p3i-*tZHth3fxi$7-ezzj39_w(HG zxvskxU*3ehL#UCUvEU|OM(p+{8~)QX7@FL|Hr7a}90%#R_@YltB9Z3I0T+Zi_jNDd< zsI-VR5~RbD>A#Yiy!Z@OQ@sGc5brGucW$%~ewOLb2SDQnGClHJbR@re95!7rOBRS6 zlsk+C|5%c9fH)`!*CU;gtSy1!KG*+}*gYROA&F!5APQL$<5VD2WQ^ZXs zUB!L>4_9q5X&AQ#Jwz=5r>ZWQrblBS*QiC}Fe6Cb&DwH=ZEC?cz0yb4D=7 z&no49HkFRf^UZAg@Y`_Xs65jG0HNPhc9;YPzk@lRG&z#TwLdnB8B#s64@5Yt@C1EN zBufKtKgKOa2#rZ;vt4?kIZo$v^Oq8aRc3K`KEE%R1>K zwoWZYEaxCjots?yu?hu5WX9$a2FzOTs#qG1@U)s2fVKZ-W>%+ftNbH|pdYFNd! zOJ-#Y`y%WIOlOq+huR)k+A}WWZ(Ijh0qKQSdfC4@R2)_L?QtH@>rIn zdiP(IhxW(ze?cH2l!+!nS5lBI$-c$rVs6(p0L0V(WUR1&X-yF1Fni&(R;ZWgxNi@H za*EV%^2<}8?ScGzq;S_KLUNi^xo(W*aqMuL86-reN(v~h-s%31aiudRB|>jt%$=29 zO_!0v7)mt-lTPSU$ffRUjp(9AlTT zOn>(~8-FbD3kV71zy^|)lAhykL7bee1A#^YkC!iY|M?);ufVSnU+RPl!czV;^xn^!Rlt1b$H25|W=179J1$?hKyKr&BX6fx#oh{=`3Tz8 z-)n|g)20V2F?fi&CoUk-7uD;C_Ib^=H=5lCRH*@ekDOG0t%7syfwc@a1D^X?#WFm; z85O-8^oYXqFJSfJ#a$h1RsjFfe~0$}`n=1K->~27{-cs@`ZX?J07bdGQR+)3{Q>F3 zDg@8kXXDWZa3y&?e|6eX9{-G`!_&n@ofmH;v8@z|X4U#3w~b~THS{DSvAmqsrj}(M zkEU-t0jjiPHw49+GerTzM6@g8q$P9yKg4RA3mC}QDfDzaH8&$N`M^dO(evVEY`D>d zO;Y-#6*j62k{WPkMp47cl%=_Mq@@eBRE z`BymC7%Z`{;%J9*4)U~%_GEk`p9;Ujg>D{W-h(qHG*Z%6e3CCRULpIDe4KX1OoK1#y7itaaDbZGPhhDq53MS7mHlK=;~7j z3p&H$=3rDt0Ve<|a&FS(WA)v3HgZW!okR9oYQx=4-W4;QVMYnN9(89G6w71v@%+7E z`NO7iha>_kzU9A(-I=J~K0WNt$|)|0Wi!tGtQUIl0AfR0z_(@aewML;D_eqH7 z4N+U7h5d{1j~+0I*^$p5F?@gXFrPa!j`Qvo|rcar_M}>N{W-_k=m~MR1Hj<5fx!MWg>q z-QrOmspMGdE%YHE!(x8 z+dRyZA10@V%|g!^fmrp9Pl!(vCA(dgRNAFWbxC+B+7n7S%fFTFUZ_qL?BM>k7sJ=} zB`Hq9eULJB10J4x9ZJ&A-lE>IusZ)&VP62bO*2bVjQE>7$k|lIF;OxJJtAmBX3vKB zG1)Y$^CxWdF(?@E_V%?>>Pj<=l6=B*Td#S;)C_Tlbigs4OPNT?aCuNu5_|pjWuLwr zZ)>W4@DeRl*$<^qkbTvHCj9Y+E=Izig}3%~XpNsSzQ&Mh*^ls#3oj(*v(@>9-bpoX zN5U?Q>=rVlmNT=&^Q?MrDNgdYTc9g7EsRC-`IHl4D53$0FUOxJYK2yc_&>kRkAEcV z%c+`#&CFBL9rj|8d;LjG%lE^L3~cu5yosX?b!d|%arbsqc^Yk4+CUcZ4&Eb)y(grI zu}p3LkC<+Ju8<@8o}5Tj@Sp*-QLD)Exy^&Gd0mq8HR8~4eB^eDGZhaC8|)<_bMF@! z6VIbB)+sTS-ZXr?9E(pU#BRCAOt)|G_-sk_;@{h?k#H?kDSpZoB1%p#2~S}Q+tZ~< z9s%_z;jXCf+ODM8bQBv@DcMzYEM3|akC6~G;(hx%G|1B?wy4mWI2Qp`>vNniKt1}G zHoG1R`?O-76ajrrU{kZ6DQZ{Dz(1t*p6g3LqZbt`+Wdpi zavAi5>niTc#?uAF%lZK9S@r? ziGQCoVdY9SA9yYuL;W$2@g>ClV!)yM!pgP#_3hu!`#6^vz;9apiEtRB1fEV7{hi;g z>cHiAl4RcYdcYR*>3NU=`E^S*;uFu;rECT>0ep33g-!F44&CPT29@;W(2**wmc zlL4ngaJO!RKos8NQ!)(IR3&V%a_~#sqRJhO{k&wPdTW+%4LBH62+UzGK(Eo6lhb$i zV19UzvgOMd4`y#+8vLdO6N^By8OlfPv3Oo4g>I*3OQ}q$G~a|6uCabtG_?;YzdCCD z>mdI=$4U`p*mz^30xq?}_lIn>>Gy44;<@pEZgVQ--d^#&vW`^#$j%mAJSvRA#oZO` z{b2&?LFRb+hiV?9h0jC&0z(&N)2$OF5jH)LJ&JL)2vOi8d1$D)u<8|FN~+gosW@~% zc3e0=OgZnXsCoe@8MvddisHu9(2Y4$e)pH|&DTEKlr5_#Nx20RM*y9wF$-55^+-OD zUjSPAZ^lxOJfMG`Fzx@pw~#2MvV(ft56B9lpl&`!v)}YUG~C?hZ|<3}e|0|uQ+S2^ z;hPw(au-ye8q|sgRSHE=NTSv^qVMV=^L<7HI52>$6Q4Kzs`=!Gn^~#$=<~FdiY=A3 zoU{WOo20!K`Y>U(l=+knN0rCbSc_l1e{W{{&cMsfc*R<_MCnI|=V6)&pX^pe z8<$~%cjXmbiBWig;jpp5VadGAW+3nCdH*9|N16p6)yEnSi&Gee^?HW= zj^f@|pR>rb4GX$UZczvdLs^*}N)EOlDIw|_YBy~*dW~FpW_^&fF+;Z-NX-zl@S|Rw|9lgl&A9gf^Fp%ePWn^3JZk4C)+?QaDEYFKR(2HXOZj=<4HzsCuy%N53d6k z6^tB#^EKFKR;?>6&WLb7%m*nf)@h_)e``Jf$_&Tlz$8qrt3G!cI-oh|W_AaXgnmU? z*o}`<%)T~n(2ZiMe+@t8Rm(wdeHAw`Lr|=-6!USQrK242vr*fao`Hm6F3U9ZHMQq4 zAm?G0++Lr~ZFkhK`#JpoO-7Y~1p~1h9KnfXb^hA;JgF6y59<6=!!uy{US&$u4wX_9 z?#?LTA3f#TQ&^0@deP1xdPb}l93rZJ`$OOTK~SbHWwScexPKC==hCFT`O=D~4_cc6 zoKwE~*as4VDwJ*QXE2#Wk7K%e?5<{>s}6N+kKy*Oc!jJ*K17%Bu0>giRXpz$(%wN# zw<&xX?!@oGgsk!LWARYTp29u@%hl%EVnfE8=5e#T7hjY8hnIhxI3>uOJ}yXZZ=Yc+ z;|AvNf{pc%&m@33Effn|`}d%#Dvl13QJ9%+>cglBEBEuFJ@t-x?8(vM_$BF$Ur@0M z#(Ns^X9T=e%5sOpYFeA>{3dDHE(Qg4UpzFQo)j6sV}0p#pDR9$t61^JTPvU17||nc z-{AcNViSe;tNOGE_boTH35;7Qp3sd_vyDC{ROi3AyLKSqc{q7#bKlY;egkpVcvB@0 z^%jwnN?>z1UdiYlhr#+>W9XZrmZy8#*`GjYAD%qM0P{^xI5@lJ;ackQz5vFCj#iRC zHs=)QLk~psaf(?bZpxQFOhB$=))GIPVU#q`E4U~VE~IQXuWs9PSC?*!fLSX=272C; zuGbpcPu@qrFo^e|}&RJA*qo{(-XTjmCvRxctN}zC30I)VDB0 z>2Xv}$kK4uZ{!F6GDn0FAN?;ETA5g^=&t&|2zu7h37%YYLXAaRDGFX&1}x9qc#orl zMpB`IaBx28pKowXSwk*>XTrb)`A~s?e6A=DomJZlOZ^dDJ4>dA|AC@0dh%NuE>w&L z593|=gjJ_Mt=Qb3{gAq%_gfH5=x?jI``i{)VDn>h5ikm!z15w$HxD%leoA-vYY=kb zw9~NuzzdCi>Hp}he<-$~2pu_4$d3PAScv2^ebWW)DQ)ge+K(m|rOYORPFq8u*|3I! zEvR6|RwB2dn-q3f-5I|W zXxQwxfN61PM*_50H{r$|wU>abUi#DY$1V@>9c;>+zZq{Hh53GQTU}m7ublk+kjc;W zjNP8<=?OUFcxD0`GhI4dALIyDeN+2C6=Wp)JB-0R0r54LFLn5o08gud2~8=rY(9W0 zb++(`zrCgo_orAHgii8*1#;5B#gid*2>1cVG|nxH{#qgzRCk?H_JrINiF7F&`QS~C zC%CyNMM*AHf(a07kKTvsh02GDg9(%!IZy`Wi-c{lEyP(PBtQYc5oYy>x!As)a{uwq zmzHF8mn+LpTa}ve1-WA5xxi#R73Xw((=^7bsz9H4Ex%~{iDeoj3i<`wC9JWv+yRapnJqIy=-Y4NfS&dY>qhQL?0`N z8oVkoSU<6xk3@iY5tA>2f^=%hxQV7=ssS6y=S}OWMU0HA6`L;~%GT8*KUtyL zb}AHeAa7y{|5+uz1#yAR$(yNbxP zfWYb=s%WKoD)7^jND3%}PDz9m0t3Yfz z3rIdMl1-UA>|P|rv--Tq zUdI8n=)PCy)NRUBdcj(DO`k;GqAEvwL=FEW24!^?+1$OSMA#;Oub=*FAg5N{K+`?O zxuN;1R>3of>-Xa@!phy0Z_jb+#O2nXFdlMp%V49=j-jBxlJ=d4vB2rLP^FUT#c@`UO7h9W0vnByr+Ih3U|p##xet&4 z=wL?I!?6HAyxl>33J21HUem3x^a?>=RW)>jg!%Bbty9=!kql}&nMURk#c=bM%V;kH z&Q$~^@&77)c=2?FToy6sbq1oiH`j{Bcb*!?0PFpjK>c*Lb`-bG?}v>u-Pr7VA&7Mx z-L)rRsF4(Ieyy)~IIKr37{$^U;h4^imzqQZzY>sWB8r|r-#FVd3%74)1gYgh82A0` zQ&8f15V;aqbYF!cYC0tw7k0r($Gf({qWH4@nEA%fHf5zzoX`cPIJIVc600c5 zez0II2<@OpHw5u!hy$-{V0BzjI9Q&d?H6Q>Y}uZpJiGWc-=a(*PE=@)VK3t@zdRy3 zU!zDjUb^A%DcrriCp6~T4wHn?{Jz7g)LJn<+_DwleYO{tTz{@sZPo{{w zQ{vhJb6^hMS0)?oAS@($$5cN$QP62VVSjw`p8<%AN-H5A#>6SJ>o_rZUhdj$1G^_9u4Xl$Uibwg-Z!Ro+`oqyCMd0!G?W8=@3 zy=y>UbtFVR5y^5iDD8K68}J9UZGj@Ae3H=Ue;>>g#m#@JACPR?3wOtq1Xw)KaUcG= zKEc(^`N1?l%~gp}tJ~UE$^*~h_Qc>_gArgDc5b1z-#afwwm6)^5)FPDb3qrL3juLz zh>YZ$aV=tk9d9>A#^WEG7P@e$A@IhrM0B0-yR6yI&?)yAB(*e)#wgk3l_kk}jbysc~~nZt27Bk(TWc0c1R zn~QB@BjXNF2)8>l1heg{fVKDj3dJ=uHq(o?vT0qe@8k_647ph{Rnd0!)>~^774NkI zhq@m4a;>iXcN@v+@5X4gMrA_yMmDM5!D}QU&EzvhS0dhbcTu}g+-eiTeYo&9inaI` zmC=1qg_1P6O2xA47Z|*`BPn(k>aJPddZ6_MW9MZ{wHHAlg&9TeqW}67IE+P|bHG^} zF3swE{)KMQYAw-lu8($*XNt??;a5&P*C7W@gsGKL&Z}yAn(TT>t`c$G-_qc84fP}< zJxeTfbbeOT}S35=}g(J&h5(-NO7!F!edkxJ0J56;JUOskN3i zo^+=K;XiL5t6_od#g!Mo$@NLZ=!ysb}){L0~Wi2?K zZWqn|LaYvx(CfUo)?xGE_YXzvz%#Jy#Ap3_oG%smY^pm%ss%52iT>$VRk{`&+r;%8 zMGlZ*TMBKfvX^_<;;ZoFfS~75oZ2l|N2{_Y#*iML5Ib;7Jueyv(K|0%;4aw{YKsh# z&_2}Ht=LKNCzl~TJ+@g)xRPs}{!y#qGT)zoDH=PAQ4LnE^Mx z($;s}fjd&BkyBzUT4m&}`|Z5{;ud7_IYrksX3b{-E;zlm?L5aj7bC3dM$S)bP0w)o zFlSzr2vv?48)_-~<){*E$X2p)!NE%NU}9-TR=5jTmQ*1V7I0lHu-?rOQQ*1L*4SOB zklZ4Y{v7Q*)83EUU^+~o@lpXpKJntY!&5A+ai$)~1R_5Jm0W5U)4y%JUFG?m=F#`D z==J*pRIDKUI_P@i7rmlLQ3uWjW_{9!JY0-Z^f28xlOY#sCo)|gZJ5RNOC5@A#yL zwc!K~Jh7pX+U4l9G3=H(aATBBbo0-v!;bqi(iByQ5v+jwNDhP1VJ{nl;MT%^oItE{ zBPYW|mrImDOW!B=a0z~4W$pSDOYv9o*1g1}YDq?f>;~u=fBFI&svghg5k69QGdX~wXODoJPk1M^srLuc-*03RtGKn_E zU4&N)wV5q=0y8SJ53X(gt0z*%2~0TbWEinnujOIuNi~0^V?0(E{%omq+&mmK&w-)5 zZq3eg;{u|J*A$aJYUg1F!Fb1B2{@dMjz;tt{ZE}w;)nVmH1z@n32-PS`x&?*F@HDf zz4cGSOuw{pB`x3HYxgg`BwI(tRaoRyN>)ezbMY5x~ZHJEY{4mm(i}_$biJ1 zmOr|>OFt*M0zgK7VZ64F>>kH`gzee$`^jhX#rqce1ZTjb zt=71eVBZ)kvbxOG0@*r(0uw0W@O>?#;b|FwPpy$?vwh=N9uNp8{NfF6Rh1OA zBPgeBF0DDCHj!;e(L)8k87}ibC@1IlslzMkky)c*W$&eI6j*nNSMnY|o3S79 zc(ox<#H9Dk+6WzWMQZJRj^VZmp5>?L|IsMy=z}Y|-AGvAEyzWau$JIL zzU{nG^@i!bT%?7oGJjVAmm%jPQi32q*KaE7;-}*d(}7_S;ASu5E6eG)uRI}|ZaDV@ z_WYBonelKw=t*Oz1ricF$c4Chv$G;`OyKz)e?|t5WMxYhM%ZBZB;Er54MgxM*JS!Q zAD2-c1%lxRuCZ2G?2LV=6Pe3dr%H&X7KN~Ifqs5LCxw&YaS|jnun$}nc1eHwi>|VC zd2m!c(;j3^`gN1G^%|L`*qcNYRLjA`^)*qu){GiGZb=m^QqFm=`5~A+>k-(|&urZ| zB6n|iMl_!)h|deBKpUJ|^#2Cv)C46|KQ5H&3u$gajzEoKu|bYY=t0j81^)l7msCk( zvu&i3@lAAVO5#;iz|Yj;HjGv(Cm~2pdcxH+c|g3#-)y9$z0K>?J1fx;H+eoN=dRKt zh|N-bcXV=&kjI|(BmjtWgKFIV;J+Gy|F;C`yq{`itB!kb^cCOmzRP0Itd)71b;R1H zoMnS6y3Yp-=_EGbfVLxO9k^e5KQhE&n+KdwUL^P&Q0+z6IdTDUG}W5)#A(p zn0eGiP#HKze$L?gO7r&-m64NX7q4OW-ShpJnA;F<=i{b9d0xtj?q-Xp6{>~!oW6FS z?P%cz;W^l=fs*^*U_(TjTK=?OCEpum#M)~4r{aCGPmABCmTPA|{6qlYW32J_K(=#Q zNOliC{`hz+wF$3y4n|>b9=V)fk?Sx0-ZTIQ*ye42|IPlp1-YpkF;3#E$%lc;EHwrL zqo8VEwB4xPJj_qI@OSm8B#3zJ&f#r%$JyK>xd5C+7(z@ok7_)MIP!e+|Bx zRYw^N$#->+JTy;8UiCmeNi9_xYxO`EVS-|N_ddql=yL_RPM0wlbkcRC$Q3R{CM&^5 zBD9opdZYhLU8{_3t!Skrmq+X!z=%m(C5$SsZtwRSP)hH;t)q^7?K$EkSJ85vAoeuY zyTHq>Nzs81b59}VR0Aq0;oukQUQ{kaOwR)^Q-e*f93+OP?1&B^Or>}j%&ch;IekZ{p{=wnKJ>+dcKu9-c^KL^ zu3M#ih(5qatrvGPn)Xr=ZZi1C9)%uU){tTZ%fa*Z;M$8P#_^I8Z3Fh$%e>pJxRl=h zwJpVjlDSs08<4vPv^V1no@rN4SGcA>w3n+^3qL_7Sh3#bFuUm73F|XX@*ZWJ`2lL5 zCXEh_Rv0d|a;1i<*D7!b@ZM`NG;acL%5ffk z?Tl+VPtZ(8@;tPxqrVk?fvk|owPNx@5mKp({)R0pM5dKSXSuO#0MDTyVZ23Gj1KkV zpH;CtS_+zXk{9Q8s0hYyhz3wN*4|+0@=HQm;EG-k#P0at+oypv9a#tb^Vmm^k~QE| zo|+#587`3rNXS(s>>pS9I>q%awr=fV#GlVHzzdqv^o^bFG9f;(m7)N<7kk~+<@QC2 zmm;$F9jW?NGo|NLmgq(mYP7x*_oApj-yCT?Oz1)DI?FKHl@?32MaWB-UZw&IJ+bPu z22l}}01F-!JKbm9_qo*hN0;<|g;*8uiF2jC?z(FJo3Mn(YyT2532s=2Wa>K@OfpC% zp^ZwZy%Jp0O!9!;Zr}6w)@$5Jq(MeM=F1vyo~VNv8K0)Z8#S;ODZ^grh%u`@l45$4 zQPTXK>5Ah-pPrarMWZpy@~{~H&d1$SKU^?sg2I9ncy?T_VWu@n8`z+$LWZRrk}hUW zsi24YJ$v1pYiXImx##n?;gidy%{1n5k|mSLAq9^gXq^YVC{ZcFviNa2R`x$~^0{SU znR>x7*qD&BUCQmXWLj-7-YWWYWQFmE3{I4;{NvW89ebp__UKZvR9!P;0kPW7+h1n* zY)?Nfd94ZNsa`^MF#ml~S>E2FTSxQfZ1uxPe_&|V-U+uaxO}88-KelvIo$m=Jg<+x z4H2@O&`f3Ha;^+7`&s9V?NlFF_>w`YDK(|L&HC=d!e%^^iU;ju>D%L=p(%rp;6+G^ zni*sRZfU3@YOI{w>t9*nH#_j2IWD~=`1M3VE+G#94!0w>)zEAwtfM9}?Z^Yt*j8Ag zp?IP0&yTtvTd&DMAQBz_ z@S4)N(JkTc;mMHCN$}+PF{Yy26A2$dELAh1HXCEC9X=VOyir>uNBvCqRDVdwm194i zaagG&@82dqwbQa^D3kD+?8Z-I?&rW6YW~Tc+K1*REA$7?S#Wti^h8gob$&vJKFBKU zVvtG3SDp3@`LzXB86AgM*UMmoSqGG{XU1PYpTSm0*?X2cCy{!5zN`$_8h z(TfK2*#{d)AZx?;3^|@y!3nyc01;f%oiQ+-477xXqqf83=U)UKnN=m3$f7&2Sv1G$nC%XL71|ecPzj9 zAyb~iizOrXsg^mNQI|gkGeqBMbQeeL6s+Ir!=53uA zV>;g~VDJgOPc@mAwGZPZ0(3yk#OR*PVXX+?Ysk{98YmV(O<|@0*@rM*N}%b7Cx^s2 zOFu+~>L;Y&&D(GAX~eZ5*QZ=pyDn5$+lwcaiHVdZPG=gXD%^de6vAn7U^ER4$b!Ed z8RptzrYd+j0G&Mt-y>P+5{r5V1Z?5{f8PB6-vglMmH)Mva)}zB{{1}jasFBTcsQ(Z zdqXN=0y+?aMPKvq{>oSc^&!O!?;ZbV%MR!C#emV<+6v9Rhy-w4_Z(hlFxJv@muVIs zzxeUHoT!z+Z)c%z@jVc5`6z!S(VVcJbsW~uYZ7!%Pf#Olc|pb?d}Z-|mMb+d_F;XV zxXD`#dt(UpS|<6k-`6nXlT9(pk{E#!K89jleo!NEN?im{tt8%z5%N}rYyHxCsX}sg zbIgK?K^NtmHm1D|dG-)!-8KIKVOqV|Yp!h+&VT7VaWsfQkE_=fK=o39cuXY`2gduB zWPGyE2av$g5Ouws8qw%{^&YT_rV)+k9UVd7Xk=;pN<{i4arj4NjPZ)){3Jhjzd)eL zPRR7ZI8N3#qMmTpg;;-VPol8FAh^)_-=f-cKeUf>)=v(vJx9le8#?b~GZ%9DJGL|t z2q-D?wJI7}?tVz|j*^P4#}11S&9+B@I>#@+pJ7$2tHFDraN;Gh2sdc{d06C0F;6@n z!|bhVb^WhO4Z5y$<5*mt5*R3#CfAA)N6Plv=2y4ki!*6ib7PGhvf#TNEZu``I=={p zrcX5i6ip86pw9xKCN|lQ9rSD4U=H`+{1OChVeTDIh$tpO==;0^4BBD)Kr1%2M<1`O zWUS;0*j+A9nz3Gx-!+6udk){~`m=w}QR7`?eQ_t;WB5A!F~&1O!A=TK zpt;o0{Z9CivR$(fRBYCE<%q4qEcLK4XFKzI`^c-O^rP|}J}xtXxf&JS)0b}87Qg0T zd=XHKi)7n=$XSa23rby^(#p6_gnFD5%hFsh^-9WWuDP;T3w@}AxJl{$N4HNu^qCgc zDF_4mkmEB5y9< z>SQsb$6|H`RWVc-+A`l4<>MC`cj~$O_nhHfc=PQ5{&)Bg99pNUhzKA{Q;2gEAVv<3 zlCzGTPrHZbQLH<4l?;OgRl%axaabAhS^Oui$4~M;r^2iz-h+7J&uDD~r#Xrel~WC> zL;?=BgmN`_rA5z=nJ6=+1r(!BVl4uBLb*58ERrR&@*nP)$Ji^9nlXFa)y1eI%)nt9 zGlq-+v8+z!raMo?HVA;KNjWWMrs-b!}rV z4ubW1%Da5PS)0ytX}}Qd646B|d<9iCFQIJ~nQYN+$6aNFukvZ4{I98p1|TyrKRtI- z%#8@VpVJ{B?qUNPVl^FzpCmN_;9c^ZvCU$Zq?=~?(FUq{tA7~yl#LS){=Kb@9sjSo z@?HDRsSAss;Nn{gTmH*ugG+y1emeu@G5I*$#ixXZ?P9LBkmDnh~@I_RH< zmOY!;vVSVOE~x&+RfBr1>t$k)%OABZ64yuH)fD9NiEJR=8&{Uz7Xs%qP!C}|lIJ^5 z@kub|hXWuAm@oN7?_Y73BK`twmvx1kt@w1|Ff&>gEc9E+9I#(6*J>`td@}xZ3)&Ry zBw21%tP5Hu{yplpKPf6bHaa`e2UP~Y3V0D;ktDX!X zg!0)Jt1BBAM%Me9g$F@XmjzB;#IJ^7g|U{6>S1%9QB+Un9o(Bln8$7&D`7KU*;Ky= z2nbe+Gy{0Q{^RhH@I3`haaJi-iYBmf)P0N*;b|&}o~&T8VGRpgf@U7J4dVoC+0yJu zW#2C2QS|51@OVk$Wo${%*8Q0;w&^f>pC3)0u?hBJ`?d`Sz!P@jmvz5J^&HGpG!Hwv zGVeZj@UB|fPJcDBg$>%Yb}hB}D!!Q|n@WjDFvI}*{XmlDndkTiuFUWs%F`VCE( zIN}F8W|)TYu)KYN{hV)Bi=J?*F_5)w7>c@CH>2sMcQ51z$r`S8T^8Cj3u>1WHO zPZ}m^;x(&?M1NTkt>crtVUMTZt5GZ}@9MTSyW@tt^1@+y_zyJpcKvcZzf(qZN&6my zl%9QJx+92oe+t)>P$W~!e7Z~6n7A?!R#b-_kpGhZO0KLmwD{o`Oz{g~e`cNsYn@Edk%-slrcg&KyFt zm?sjz?ykD+e_kr#S;QBj6ETpI2vPc)3lkp4wT9V0hQ@{RQl=)EfZUDHUR3us0L`QD zQrm6f$z34{C9?_vjAxhjrmrHJ4ely$kgjX#UFh9JzW zv3eMgjvSPgMU{TEoW=?}{a%zJUX3UCPAPcWYq1E7Z3xoDhTJpB`^o^BEVYIK$W|K@ zH+;23t($uWIZov(NW$3>(^C?us>06hwC(z!>$HfaC*S}aauK{zn!E825=)yP4!r2A z=2~MZRB~#5d|PK|Pl%G9rhkQ?P%{f~15u7)=%&PEFvv$-pe{q@^-jQ8pLhZ)RQ6?~ zotpy~#QjVzixw3|stnVNgRPu)JA<}~gK!!?sy?*qRj|zZ$BCn1f;agXP&b44NU%f8P1$WZKw%x9(m7>*4$oWRg7GV87|U7OU3-ZKvy9CZHUtm<+B*&%&S2 z!nvRhr!4P+99P5HiZT0|Pkh&+>zxTyjTwPcmb#$$hiSXmS zZWvCurO*#Top(0Cq8r#XVviy%HOR3Tyw7zN^D`-kpVfIZfJA`kq++s;4BdJtURavu zzss0t0q9@HN_k6^v9lOWJjw}KD{@-YQ8E|7uB82Mw|fVaq%AQHK%1~6p`jfOe6mY# zAp3DBC&6i$n&F@c|AeC4m+G45CW8+No7nLY>2WtUI_bqNAXixa`wsogcx!_D=YaQ5440bc{~ zB%$NW`yAsyctZA?*rHvTJPm2Eu$HhfjBPiTg?AhOOyHW5)!W0MyUGHgT%&mWs z&=;1KKSUFXwb;r6#ZAOBxeM{nxP(hwwt>IByMCViA3Wp0i>>NKynELv?1Z4ZYhcVo z9TCm}=H5HzoX|;Z$s>VF8xtx04!>uhEP|p`dX%+l!oxS6i~g{BR;AQhc^~x$P<(O* z)9DZrFPaJaB3Ei%sZ@77c#Ea}py6sH_BaAz8_36VpGy&ix3SwD`zqBo`-Y6Q1YGN1 zhHD=D$?H2UY+n$kfBX|%@|{*G`cT`Flh%_Y|r`gA-e+rS;9Kq;52SA+F9PBK8Y({rlAj)XI7P#UFDh+1)$slX= zmQMA{Cp;3P^i50KlCT_bNQMY3_PPiPzbz(zlwHB|vz%{|WZjvmvAV`q2a9^0+2N{f z|CUK4<6om2$vhr@hDX?c(c)Oa-btVC2;X117mJ%z8tM+pd1>IK-Z%H3XI1!vyeGe5 zCLrzE4R_ZHLrAQUoF{yPYr7x%`$}2xeR%^>zG2C(*|XbuQZweEI<#3yO#cgqNP#2&H&ce*e@zxGQ6%iSUuYZ^DU?I*|3 zFZ4~mr1(kzQJ#qu^v)c!A@9;hRBAbH0Pp+zm&QP0Koq!unG<7pU2HbHV_pUJ1YkWR z$J70_4KbhU4xo4x=E3-6_2Y!ls4O$GToWFTg`#A*0`bl{V72Y$Bq`YuOlZ4#D^a3U z#S3?fwr~R6t}!`2@Ku=-k@#qu!F%&FTVx`iC{(iry)W%CokiCqru}>hycUa_yrXjL zadpvccx|JjRnQntVRQqaL_Ja< z>j3HOa~A9(n-0p*`RzPR5bGI~BL`hv1Ped&MdahAHJQ3KX}jO&`36A6s2C{ZY=$a16LyWKVV~Uxl?^ttwv0S4Ty);2b ztzX8x!t%cW-b0Y6ovY;PDg0jm&mAx|1ui8?Bc*WM>k8vR@ow|xZa0V@mqzt7g2hcR zliF{gK%c8oQv6?xy>(QSU(^OFN_V$}gpx9VGz=lqohm&bjUZhPIfS&*-7O$pA`A>5 z-5}B+DUHCu%zNkeefO?)*S+_T`^T(x*883{ab}85=;wid*17!`rgTt-?8>Ut8LLW%9 z?4!8@{itw`u7pk8gU=cc^o4#hP#vXUpIq+314N5%&IbL5h?GZzNl!&*^SExW3vzW< zZLSsOT-$8?#Vy^XhJs0YK_`Fct6ZEXbyM?vF9?;=a_b2z#ZJnM84}LV-AMcB)Z2w^ zm;A*8Q5(q^m+;rCRDbd71RGpXI7uC5f&-`dy$(!dOR_6}2>o^)SAppxT`Wb@20$RW ziyHp2kx?YnZp^e=C(MlLa$MIyZ>`ZEjv-lMRq*_^dn5WIzWiFJ>xu0)PRQKLqmMzG zt~ZsJ3z<&~gI9O_DIaXsY=(*EoC$5O43%4(MQU(;C)<4BY-g7jq?9b#M~hM_{3+Mo zRfZ`N`o`d28RFXY?E(9T)QzFm!q1UVG2h-7*i9R^&to3|_aML?B2rR_9$v^gJrAyy z4uRr6&M~~7PW8*lK~p2o;^4)CKQN6!u0sSb>QAb@_Jqs9JPJat%)h(@ipsn>R>GlVDE6 z9VD#?i)<2-txhqs#Jy(PzgvH=B`QM2yz)-pi%RJXTOq*iQKr*E>g)QeW4Yatsm}O$ zQDt8WWO)oMEmkc{i5c9I`fgNOF`AgFzRVxLb9IV!fm8Tt;@I(LyyUQNc5!sNh6Qk0 z3MY1g1SFcq$LGXEE8AiC11ykFN-rhZiU?YLuCQFT^zMb@xd|@TGee@L#h(&bO&C^8 zhPcAgX&=hJWIW(GKaxU@NI;2on~Cko_rn_f(Il@jV79HO3}njQW%wig!n`oF)t-@pk}fn^V;`uRr1@oYX-lDW`lLV zr_$y+WWn*GaoaC^4$BfA++Xnj^-8LkhApUbHg9N|CW;bQEx&o-aax6uVOpfx{U05s zR-s88Xg@j;%`#-gwP#b~ia*=!xmA>xZj8;d9L!awVdPaN%wLO-StnOw z(|~a}lX!Sa+c4oW!h)Dq4}@+1bZtR(3*qu`t7PBNG#I0vf?+2t=~q2-ucrK5EqjR= zKPkHnO&xoQ@HVbQkI4$7BCiq))^3;j0(q+Nw!~$n-d{#VC-`Op(wpWhJFlxL1~0Dm zB4KZ~J`3kiZ+4DWXQFLWIN&u`Qib5-W=H==e0_YX)1LO%;|KUZI@kk1w`p+S-BtR_ zhUU2D8yowvVVGbBMzYFw*dIMZF5uI?IN7Yo_{#S?KO=5IE{5MoLPu^RaruC)-}!Cl z^RQq4i5GvOuK#tBK!P>g))V;ffrsDuAHl&S7)9apa9FR45?zAN-P(w|x&<7MDnqJI z{BZwD_31O&Modd;Xa97l;MUHrS==Re$^8i3n=xXt=l|7*7EDQr>kOJkWoX?s-zkO= zEc_#ln1*a#wldD9kMLz*_g+3Q25DNl?wP4!RM9|;oKEkUT9rg899tO(o9%zc|6M8E z`5Bj+g@aKn0zJv&ZeJ)~StX~m=;a&y*3S!5b5atIk=hnen5;XwkQjq%ZzYxcFs+`P zPKXIxs(Q~fzrn)q(ChSAa8Zr))Y)_2V6A2L^ygZSXLp~Zw2>6{_n<4WHoe$Ft zU6j@8{vM=>^L*=SKl~Un8EDpmvetb0*{nF~;j387nLmQc{TG+Nu#KmME=loPQiT@N zm)k2BVcMP{kzWLQetTe#(1)ord{CAgkb@P(l6-Q=r2(e@ATb$;JH7I4Pq)In#}q4= z+8*xpipB7}rHG$W6oZKGI2#O!l{)#gvuP<)uf-v!k$?L^-hd&AsQ z6W5jR65WH5N0bWwC=sTTskMo(@2Qf}=^JN;@$7yP)6t*NlInD_16wLV2gKv3I;M(J zVBl1Y5_HuQU|?=6LHfS065UAHxYxY<7-|Ag@H?zSuiq>B_??hEXciHJtm!5RTw4)f zEv01pV;>hl->JLE8ba{7G5adem1yW58@X$xYGM4x{>k8}6pl4&_s?qbQuIbJiqpJ= z2$sMB+v%gu6r`6*pikRf0Ks125kEE|){yPeddIlnjshX>*mP6vSnI(BlswYXSxGV`I|nBGfu;r3T_C(RDmprj?kA8TAQ*I&d&70;52ZX2o2I|`m$In(@Tib zyEyhXdnVSuG4xUmnIkC$02{2-lB2&~6`Nbb49z(0@tqPbOnpU`Scu+IW{0(^-J;Xe zkQ1U3fsry2jpVfK07v0-A<|F<03Ws@)q#940y6X#%UOyJvzY{^je8V8{eB5~#a-op z;-lCvTY=4_A2e4UBDpo`JMe7cLD-Zz%Tv0*Z5b4M`H^5bvu!NA zQ&&!vAX535@NXMFPIwnYO9@uYJv97jbOQ_eRsW*>b56m?&}y3Wg`6bC*XLe znT?8NP4YohJuph0y)}vWBgmI!>WrAoygu*MN5;#-<6n#cLWlVhM}weI;1n&&-Cd*Z zcz$zl8AL-&lvkeOJzh% zq}=jv=`UXAf?&_KmEW)&2M8bhsSp^TGEOJjHMqd$y#R-=BiftFwD484X?5yc9yp4i zXO`^?fM8N=`@Sb%8lbjPUGvBt6uiLIB1rqAGH4uNE+q0Vsmdw1{nuWgD8?)%X1mJF zAzX;7)#CNGH~X~m71iQu>m4)mBXLYC@~Eewce|gEGFIY(-QBT30 zxX&Z8OSJLS&X>LkW$Q<6xZ*pN)8@q7+WzcJNyM9kbrF8uoCS&d_NAlsJH4^+zb&h6 zZP4fYb9N2E)PUc;R$)+A3HP{(p3lsiY9$n`pU&4CT{hPIDxd*V-`1k_Tw8^7Cq=Xl zns8kk&t`ug6uADu`~eReRxVS>>hn=Xb<660jw6vY{|a0`I)FA)WMGb$wmCrNS><+A z+H#%Cjk4}cu*Jw@(YavP&LP0YZ?~E8Pzm>Orp0ek^v^q3E}q?7Kr8FfanzG|0S)Wx ztQ6>Vf$T5$C)oV15*s_NL++vUATJ(MfWcvS9;TG6lfMJP?g{<9_aGZrfp~=a%QP{4 z2jjb7&(69H$5{5ZmPno7@8;cbvP~+lk3$TTWaa8{!&}aHh872(9!h;}_EQiU1Niq6 zuUMTSb4VSEa`Zw3p2YHnEX$lTD4BXH|K|Sbm8{!JA!CKsr4W3RXJ#X9KfDtdP|jts8z@`@Ex#N zU;fsgcqQaX76=wtaA`!lMD@8(HcWpuio~CuUoqV#i;V7bK@qO(!kLAa0LjW6nA4CWF~pXS=#Dw&nG0CvY6kwd@b`5@w*52ZDi zWOhnhPIU1t zKa)t3unW4fCItlWb1MF?-^Ah12cjhg+yOfHi0=(-@$Yv5HcCm1p* z2neqOr!s1AD0X`>_vQj`l2a#{Oq20)X^xfjFOgF#0WAF>@9jC2L&}u9dP(#I+QT<1 z9Ic=ikF*zH4$qU5p42r!mGd1RNR5KwI9c{fHW5aoL&SszBDz3Lk?LGdBHpW%_}ugt z59W$vH&Q3Sto|yVpEd+PZCtUj(Rkh(g{#Z1(*~Gu4mXfI=}E4l6;RQ7wgv=55C;5w zi@lTFVQwt-LvhACvitd7niGfMK#1Ed6N|H=8GMcAher4M6m>PiJrRLUpXl8;$abg# ze#2ixn-d!>A<6g^vpd*`?GG}t3!nj6PFXD(G-C~`}t2Oi~~Q?Wz6GpU96|S=4Qik`nQTtH^ja^%iPvX_qtOjDxC%BXv%wV zsgo~Y_a{H{+g$L)H2!D6SxOFagtUCsUj?<+WgxCwbor5OU7*pt06kd$JirTcEZCuh z`W5-B6t8_U+Rfi6+ni~cuA^Yhyqef=XPX9Uf!T-MT}`1THK`r9`QQ7A?kwN=7_?mL z|Ai9l&KjpVj{kKT5Cwwh;GIhY1VRx)_?A?$3*^6Pq5Tz8Uz+OyF%3 zZB6)64eo}{RQa^t@&M)e<^}8rEbr^6YLSPkDhJF&4q%Q7HLAI@jNcM321Y7?Tc;}Y zbElM~p|tptt&$)Zy;~r&w$%dg;RpRs$%k|FKU?;oQ8YT801D^N7w5tHpo^ZnV&JgE zg8?w%wAm9>$2{f0Fq^9=qzaAKo^FMtX zKi0MO)k(}k&ewlugV9k>kx;*PvcZq=dGwyI8oUiSJA@o*EH z|9tz#6F8Lpt=L#4jq~!EP*+M$O3UK6x+5f6!}X3gCAQEl8kFoV4USgV)#q9Ypw)&BdVzcrS4;Lw6dLU8mgkq zrW?VW=SVWUJ|Q5;ji8LCp$cd!QNXUDfHBaBBk0{&bzj!pTdD(ra^DITZ+l(}67A&J z^l;SgbJ6^kWSgc}a;IpDUkpaR#cY@1Jsmp+r5~-rdU?9PoBGKeGNB(obPAEPOB)Ff5;@6bYT4JnKwWP) z?n=Hors^aYH1cJSKcj;83&VcUhQF6Y=pOt5ViD>IKE!n__m-WLHkI8#DH%X##?DR! zTL>u>vD!mFhfmUcY{?N{WM01r%< z_|DRmGY4rI>i5s43iuMwzCPF%ineBecfX0dVr`RGOh7+3*i6Z|ARBN*FVPVxm|Z4H zBXn6PtcmRwo#kTA%v>JzwXd%x>zanE%Y6^V&EN5tIUu^XIls+c+aOIrTrviQZAKSD zUIRSUP?*j8Pa4FfsnqYy^p#t_*1jXIV+2f0*#{P=;U%j8SCOl$9aqT>&QWG)spZDQ z;af0pkLU*+(HVajm4?&DMfI)TM`e6CfM&AMIi{7kJ(c(%E7dskM)9amp)K4^ZTZCm zYOk7IRz+V^y*k&wSRHm!tGQTf%b>qIET*>SqyfoF9}F9w7z=;f}d6puFC}*?ruHC}j40pDnb@7ijF@Mq8 zfRLV~y{29(E5VLuEPFIUkv#o&*L@O9Z;heqM9XJ+ z8bm1Ti@8#EwsH7+W_tq}UB8stBtbs8i^H6!9|4LH}3*{M- zpPUw}yGeDS;VN5fIts4N1eWkAJM2!kJgHtFdw$uR3-AGVSh`v@HW~5?0C&=2TjeQP zRNg?eni%?;r5<+$Ec1fZB7pT_r*tcFLQLjE!$prtKhTWg9he3z*a4G3O){$;Xk8M( zw?g|!PV}Pz^jlXlnTlp3^svu*t zw+>>_92411{4nJB?}lFk@l(0&%&QWR*lE*J8JVmw!HPhiazB3@kua8ph9QHCkE2w)AIl(C_Zr$AN)j51UI9VR zULsWAHPc!3sbDotcUU;*xFC*Mf0or2vU)<8U{Yp zbzzIO0dV`L0n7E`-D2cUf$KbpM3M$Yt3x5rP@ePKfSSnk*umewaDE{Itn@>x2LH&; z*DuXN<+}cS_RigsS$9D`x$qMm%EOnaRb3o6%`uNw6*rD+KY7RBHV2jPbK%tmJb;&t zwxfBiaJ_DFaR+PC&YpuiPTL^#MmG!~c2L!VaL3xP9x*UW;SFom&XFFi3zBssF6 zgS?50NGIJc9V~j=mV=&(2bC)WB^-t&4@GD(zQgy^yFD35ZwJsk{rC`E0R13xQaJdo zte^O5qJL1P5I$6M$NCi9-J#sO$mL5~WNA$M=4PF^3tFcMF5IZcKjkR z(IdV54q`{=&qavXa+e;!Bk{*pfb#pX7M)SYO0 z$n*?v&a>F7{KUeq)<^*%cT<0cNd}keD}J&Nv&(Ma%m>LDEAq)8^p373zP^%xyiqAg z0SPhMP7vN<-6{Nw+xy)9L{Up)Puv|+*R$9GkoFfat!jof2HMuOJ{$E_^Af1=SUq>84hr?LHBaoL6$-DD4TWr(HSRLFD+uJ@bz+i93iWbo4Sne#-2cGV+%-gb^K zcy4?q=w5Y2NZ){$H@-#zB6T}^&eX4G1kb5{s{;Wp_w`(J0dz;loQUA>IKL&fj!QMY zxk@ilaD%)Uf|Ymg<()F)xbQk-l(v18r+vUr5kEB|CLJbBkqHepYIx6TlMA4;0WP8# zx3{1)Qfam!kZPY)F`!1g$T@oi`IrAnI|SIyg*uDl^C;xb*Z(8#VLl{vPPCl92`q>X zvRnldD;`|F!Dz7u?;Uc4^K$q_)(6fHk=|R6gTM2$z_Hu0QM^m1|$*`K8or z>Oq(y;uF6eXCC{0zF@78YiJQwIrAaC=i@t!<@g0{oqls=yt8~GlIGBYdQP-E_Zeon zT`6pAD(v%A?xm}zi}~^s1XAZgfC~$zL1F`_mxJydv1#&+0NdQlEQZdoS7h{f5%v;gpz+RNYEb;!3ChcK$D%*fc-%&De@$C(dW@5QMKu3$9uk5+3HJ8|!?ecZCD< z8r5$Tx&yu3KDo;B*dB?I7S*}z3H68x`_#E6i}<{L$4^pJw!Nm+yif}4M2n7Ocj9s{ z3kA4f3r^WwQ!Bc05^KWd_EtXKMR6U;iUqNc#D&UXq7d=vO0*|EDd*82XjnyNSxYYS ztSvuC*ynS7_|G4vdcr<96i=2ziVuX-7KMDAt;{$&Z}VNfYNVA%d+clKkD$%H59oa` z=-LgIdakoL>O}o3V2V&-9Ryk8Kun_0CIVin^%wL^=4zoFv@|z5U*ruCA+(5`7g_Pm z6TmYD&#lUBa`FmMRVKwx{Es>C=@5cuY?FG`GQ~ZV=V)(>j&dJ_n@Uxg{C?0k=lA_z z(FSP^Oi^DUn2Z~gS5sa%;bwYi5R-P*y3SD4#vl^WediZoCx%Y+9!Sn=R2XtDjc|L^ zk7&4{zkrpxvP_z!-`X!YRzTg389tuOe+rG_mIKAAAOr`Xcjv?1%XLW&e2zlTirwo8 zAzuZb75%h2>Z76jHbg)92y!jYGG+=)J%Whi(hfiw_!GS z9n3sey%shIX-WTq+*?6IN{BD+H~S|&L8ymO^#xQS@m#5Wi@E2-ZdkARYYb0=icn?+AK#~ihpO|sWMf_cLo?tW0p4t300i!9uxvgEn+)#KIUjJDF=gTu0M2o_4 z{#q}1%%cyEI}def9s~Bq7faJbPqiH@hcZM{Cji|n83Y@7q;1S0@1SNZ^`BME2Rya8 zAaQDoEE!?hG@v~4E{i7?o&Mq=zx)0?&RotRyFMmS8%|jv0%;zf^=bvwkq2)TYZl=qwE z&`h5Dce|W$?zAg9Gk~A7o4<71;l?>H2px6Ly-p?adSm@_38tNrhIND$A|##=;`@Om zXg&-V&GWw9PLp;=FT2Md1&}5`6HdD!Vhm&Be4lT=gJ)-7+J_U6^L>nHJ})J4;Fc{y zhUQ{oTv>uXQZgQU^1QO{l)E$<)wxwUVJleDlFL_Eh~ zNh~ou02_NX7S`+en(pgcnajDu;k`OOStAUc1<`Jp ztj>?8ov<~Huon)SR1&%rnHT`#3Hx5Y$rkoPO3~1=R-IWf4-P*gNuMHZf z9*+MoRX{aXr@cZPpZY}nUR<>&swms!mg|%k@ACU-An2Js@0)szYg6EN8bwR zaX`{BTaDIRSlGu^P5Z6HOkWfZ^cK5ScY538wL1}woJ^-F<>eg7-T52(PMKE-!R$CS zC+!!#5oyzyXSQwz=z&Vycj+LXaPsre&k-59z@b=61_lNq{H2~>#2<-g8GA#Ri#o|v z)wS9HisGC&6w~q?`xd&3|n?c`s!WAJ2QB!r|d3`fIH%7zo+4vajavp(8l-vDM#D;)F!H) zK=g%%e(Ua6p-J#3m*wXl@S!4c%lBmX(lJ|Xe&5FVzb-TQ92fm=x}wBI9Piq+)nM{_ z$`*3!iQxuaUFfZ8z9{mkM3Rn9a>u(R%C||JP!dVmx+e@8sJy8IVO&FzG|DoH6z21F z>@MmfsBKm>6-0Y@aV4zzF)73HCu@?dENI~Ky^1T5)gJ7}cKZxQePIpDMCKiHS;SJ< ziViKWSG3t8#Pagq^E_eLl@HO~PWLx|$rnJ!bzmpN$ib{3dElNfYx_99#9XG2VtCpQox%s{<32fa6f{|4iFY0kcnud zWi4kUS94ihU)qR$>Qce@=ooPS(0BilqSJ_tePdZ4M0jBXE%a zM2-mPv_bEVKIAXm!x(($OU7nQsdW!_PdHnI+g}zh(;9!YB-z+MUIq;pk(L2bWY4p{ zp8!pmK5{7@+w$!(!}6CHttk54^ZR@)`)8Z9Mmu&9YlIOs`;s7&^yA4()4isvL8{kS1Z}2Qh-YguHf+nc(A6_?|E2TiwESqhi^BZ zmHonm4(L?7@9(&L@^j0VD*j@^?hw4JrznC!cBX7zA`uDh_?_w0IRw+Zvs2*6fJ`nS zB??9(UfW&3={jwG4A6FC*PsRS9kxY-ATc`i=+7L*UfX~!RRpKUk}Bn{ni1%|_zn*h z#3baZ?-56x8}^dON2?F@(RM&}Ep<4P+FbwK=hx$46&vMeE}O+7PtqIM++)Qov)r zLJ#!QNiqCgD2sXAh15slTX2vPB7_~X4%EhlW-+5q?CvZv2WfS9A5oMnPQ;2ywNHJZ z>$MiCd{cinys%ClKeG4sk!T7D|L@aCMU9?5Uh~VNe-;w)%DoIqiqse2CodS5-`t=4 zj0;VE_LeuR0cDPO(3OUt2#apxchAH(sc3OF#Om zZzInFkqq$3X_!NFWPy4UJKr*SgX-&QGp28paOV}l3VqihLgE#yOCF;#lGl>3H`qiL zJHL|dLcV0UYX>eMIO^U)NoLUSzdkipp%CBvaQCz(|SL~JMkX-x$NKO23c)Uhr!dd=8czOM+{$t zU#zLN%wHQl&>%K{Ca95x5VHD90Hdmk$RG)`Z_=D?;7BN3rLJ#eSJ{%Q!-3n7{x5H! z|MX_1g!cDy4JN;VK8>sAUtM^U1RP}i`PgffSkm&GMj{F|9`JnqM6&ecJ5&ToJW&?> zWw_@m|4Cao1}A$Pe$?JF$46?bQ+@xo6XTWSCL7A{8v`4q!SD+E+$K&17B-rV#(%HU z`yOAq5*K%o3^Loxcvk}SSL*vwth8&xIg0%cHX~ zjzaf;v1Ko9OdW9Za|EzbDIY=E%;qR6&kIMa)FPZIYz%j8GTk5ihdCQta_NfDP+hFWt*!6|<<-s1cx7njR z%Mt&h1Muo%_hcck^NC(6)We_HyoM>6B#HwxlqRj55zt)_3N|ZXlJ2@r80vZ}gCTaE zUe=;T9O9%t+HLR~SotyG4uQIMpZ$-~vS&n5G5%+a`heQ2^qB7mp{7}=aMEtk{}}SB z3^v1oBe4<_?5ms;6ULCS!;kzuPEKw8cplO|BlnpSJ9d{CQzack*alz>J`R@e)KI${ zAa|>Ku0-PZO=`4mg)j}1yzHp!J5SC-VA!e_Qa!&{>1M}6e~&mTxMVhEKcuTTe3N)K zV~1pf>pH`4VH-QeZc`KVuok2aNKoT{zZyZXnqz-`S0s_>o=c6iS;6S+o_^xbmW^l) zwfk+DMFUL*;QnKB$*=(JsfLYw`px{za}vBC_$1K^wM!ObFmK<){{H+IQzU)b(dSQG zE>~o~lE_ez^sB$9=}?X2RD`AjM$t~WJ@wczF4?z6O(>&+pZ68e}y5ApvD-m&XHay=f zMEaS>>Hcen4(q6mojD*HzU zBxG4FcnP3)hV>)*jy_#Y0+YrY4im$q+hTwx7GBA=m~9f%f)7aulf8N8WV*4_A1E@J zX_+D`Ow4vO-R1JKU6KlB@&dN5%U;`nz6d4`RZhrHnWVgIzZwZV{UjR`c6as zZ(Emt5R2%vhD5r7t@Bzp7REVQQPxmP;LEYv(1;3>MJDST08_R~$MQyWN&Qv&j^&Zw zzfPkcGyLPjFgBgs?-Gt1ONo3sCHl#kb0`jmcwqYi#zQ~nps&=2BBTnRpMGHTt;Kp`)}|!K#6r;?xfc4( z{<6rrfED966rBR~W*CA*;Xt$oCPqT)MsK(lAz;LP{)+=IYR%@X_ExvV5RPaS@^Fx5 z&gH~%xjWzqpM0T+;vlc@Kh;iIPWqfQ*{23@uHWl?@#mTVP7!wbufa(IFgm#rF}BS| z1Cp>xy-7G}zaqJp<9K4urH%&ilwNOrNQ&bJ&vJ2xm|TMk1x41om8NmgFc5abd@TLL z&{4DYXMfCEcB$e0D=>}+sPIqswcQ9OA+P$)5cKeuih=rV%q!%}1C+rvzj8^VlPFBu z{xy7+X`^{##aNP1TL8fN*wE~XxM=GaQfANC>7nlkWFyz94SBCFdXOJ{iWheBr>o=E z{lyil+|tn_yhnbAz4mmyjG}0Z5BSeAn+w@rS0u~KxBpo3Qd3%!rHI%reedvIV{iX7 zJ}dG5u=nlDp63vfuUsjn<}bfO8JiSQ?;;G`17Gn9#yF z6?#7?;nnr*WCK%}Dthq4{(2XyzIx(^?6U%8OS+S)(bgLn;%NYjjcNZC%1wsLFr6j= zo{hb6KarLP4AkSJpW0-sd-(#Vvq4N@SFD-Qi#WfZ%VM`qekf0NbSIMS?kLR`=%oKi z2r;GqPnhsRdwXhM27m{kZ};#z9$tWtcJrl+R9@}gvbJ1ITu<(o-QANd-|jl%xDy#V zpj9-8dml^+2(Qg#AXy ztB>nzb(l!ZZ`g0#D>ANWQ@}BH+@}9D)^r~FgjxaG|Bte~pjPdU)x z4B9;S*F2qG{d){g@{1z1bQ%)ub9!zkPaUj=&;BRprJ|!?GEKw;a`nIN*qJ@@H7*1v zjMlXkK=dZ+Fr);73+q%&0nJz(XmY|&1M)P_suxF&l$S_*iJGQ7uO27u z6;F+i4_!r&V_O&0XG`GYaZX9t&cpd~?;n$M!wlCX70Z`$$rC^m`)d`7-R%6M5?{cC zo}?UVpZr>Txz+z1nJDeAqB?D>HjaVMVB;+^umPedD!K0GKQ9bN0Yiu^5H7aO(@ZU9 zYi;9qmF!p&gi@h^UovadbX@jSRHfB&m|<2Mw8wgcDa`>x+)nSF>fGrE{%VshfTldb zD?$^9f8|*K&oYmQ=^0~kO4*t>PA?u`PLQtIGC7ygbNku9y%A?Q$hT-Z61Op279dW` z%jEWclb!UANoTB-{@zkq)xeHU$x*viy{$RB{0&eUj;iwrDnpwzMo}-*hn3O8FI(kb zz=mnACW0u!(HM`~Ru;G?vUB=-W?w9P0j93`+AVE{9V*!Wh&nqf(~t@+hZ+M zO1iMB(YP|x{VekzDWd4dseXczhgR8)%`Gvi&enze8+=j{r@gn&l55quA>IiU6>y8M z-CA<#9P3{!^ci_fQXp#Mw{rphE|K%Y z4ZYOoucfir72tYkTSW=K4C91CCYb>(jCeTlL%kt8bd(Asi4%2CVRsdvk%1Lkb76Ck z?$6En@#5BwM7adGMXAPNp=Nx(RiV#*bZAKuf09Bq5`p-o$40D7F)`KasA1qZLi^tb z5VkwsiZ*k={3{wbrQ$GJsKuLq3n&uv`EkoGbXVfu!Ua)JVms*^C=1VYF#6Vj zLhK)gt975YRh7bS>vn~d$wgUCeC^AR!h%aDF!!T5)z{&lZ&Qz;WE>LyWFwR*a13_Q z6{frgnQA;d)%TiB*LkA<46|=^xoaR9(eMWrrUIR{={6YNB%cjF!fpt40x2KPfZ*30 zk$WgqMPxS3NdS)qwP5r=4@@23WdIcIi@crBtSKkduc32k$9qZvrz`J1R`v40I>-2o zP#2H*2QZ|sd+*%&JFOjRxSWEavl|@X7X=Xvfby z@n7AJnr?PkF5Yb3>*UmztzeX1WtEh2YS1KxC3SHxGD7cWN4%pcDqu1TptXRC7;Am2z4eFt za9})cV zJ)jeL%X+f|Fu}KDcH(32)>7|T1+Mb7qH(9m0F`WXdmPtG)G74)fKafJ32=+G>vIVQ zrouLGLM2vD))%GK48U)($)sd%GusZZN0LGe)o)sE8h0Wv33SZC&_Z-f4O{@WzQvSP3lLdIMNBwoO(^>*?qU=sZbL*AYfzSKZ>wgV z*+p<_s(NbBivblDyoUBs@Y730NW^(!E~_9jfH~m=6=FKDH%j@@7kNorQGu}9fDV&d2vs6hn&D2jhyyIN4)@qbi&5NcoEHS zI}Eu2a_l1AQnZ2NJ^awYaE@H5z!Gik?G=HlhFS&Up|)`Ov=>sf@M4z+$(!tA3;WRv zvvjY^r+|H~UAPG;}2o&z=$OUu|`wn$+&QcSkWBG$hG5Pge?nA5VfFSGC&jQ1acmMW{ZtAThV#3(6?|F$x zt$E#Su=eunDqS2t5Ta(FMVgqh51(0Z?U>PJqO8kFR2dBzkEi?n-fbycJmRkRpEh?; zi(ibuyWC5%lLfmzf{)Xf5*pnaq$S7HZ2<%oWp7(@`F+97Bz)0T=0&U2P=w?dtnc5d zTfi`0Tln|#;TLT$E44p=k2Ko<2_yAXP-*@>h1cSnuFBzQ8i3@wc^px?JAbSYH4xFX zItOJv>0{&!nZg?+;FX-cAxEuq8qK8mgcoyEqZeE!({pC9sb{FO6#zDp$CqTvxR38H z@FYBH9bcd^kaJhw9u0dlh7=as_M>d8i|1T-Yd*Smc0ddAT+5L3rMe|qWsRdgm)3$F zq4W=`-j4Apu~Z4Is#&mB*rXN*ThDP$*yp^49{w7VZ}#OAs}By(9UopyC6$I}|9Oy0 zdw3g#LxAD325ATz5}QlTwsuUF}{PtJ0UMMJrsV%U7-lVXtI zYuu0)5A*#8<-Rq*O+ofeZzjzxtv;qot!_ORBMJt`sj|U(`bhD=(wL_;5=z!6OsXn? z(u|E=Ts*$?qC==;-7|wQ172{)>3aLFxKG3E zOVk?-%Wa`Yl9J^CN|R!D^6Q~3`yc(LsW9E+vzv(qB`|CnKaKbe+YcGzo!=^DbHw&G zgYpofp$sP*@ta+n>$FJSFZss|{c>9JY6cWXRiR+2N62#Zy6>1eyOQ;@+()O;DEqQS zMq8beG4r(07KJRre6O`lUvlGpEWGgdn=&e!>oXTMF*Q~{){5nwS2~-Ss4usg1Xkn~ zSV^9LBz@jZ_+_e~8P|@JwCoR|YEld}CdLSAhu=>-VGXyFvA#+s{ls(`B8`8QNRmvG zIHnqLas+)&a9qGs_rX9<0T3~;IaE-$qhigsa@`Zb|I zkUQl|eOmtEE~Yg~X_oT$pPzLl;`nQ0Xm5W#hN6hHv`wQoY4ndRP z!97@lTjTB$Tmr!z0tD^g?(XgmK@;31f#8tf?rsg;d-nUCnX_irnsa{4nt#Qns&`B6 zr>dU1?z^7IeA9)MBRHNZoAKTg1&25vLg&)eX1`VMg>?Eny&%dw6%)>K%>p)fU<^)1 z5^P$#c?P|=tN$bOWBw!aS4w>WRw{`h7evP0b&GKN=D3x(UM8kn=5*FD_b?(YA6kiD zaIg}T=#~ACd+ra>RcBV)M||odL#v-|wAV9LSwFx$jHmO2GTvwPQuoAtAv9CN z((n$GmG4cw%)&SYloxb~Dd+9?N$w@0$WePx^bfeNqshQjU&Cp>L1`%G&b8qtztjQj*S52es0|YH9ROAeU7pIQAK%U#OOzaIi;#wM%mur< z%CFo?C|^O0_fIw@G+oM~8oneXq#o)I0fTEef`4;R;cgmQqRIz4((tn9@0ph-0Nqbv%9&h(cBN+EoOe1ikUrht_Nh|$3mHgLQ9L-QtYrg zhZ-*Bbz(>J>}k4+*8BpD1W;R4`5VA4-P0OuExF7Qn5jL!YCb3>4I?B z=$Az!3S8?j!YM`tNo8!(q5i&a*z~}c7!;@t{!K0Xp)B`fB`jIT@npiI z4$SUU9p>%I%}~6wt$>YucekCU(=ok%mW4>$@{1Jiba%EVD?adlRaq*pB9`fxr4di` zKd{v46$qAQuNq0Ly$C2kGs^6Qswf*(j#WG zJwPz@3UbCa{u@xcblnYSE=j0G z=6$4EG#7y75Szq?U5jZt7sSICic;=?p^-Fm9)Mh_M5Ygk9?rQDP>D!LZ*QfYtS|}~ zH3K_?tbN3_-9Q=5ca5eR;9NL5+=wUBcsZX;HH2Xy&~PXw{}to)v^jI=!`{b5ACA0I z|581iFEBV9lFFn~N+ni% z`y5bVdr>Xi&tE7nh|O3O9=we!eOLshv6P9DCt}KUHjo6c(Htb?&e%j2`f0X1?RuWy zGxORgRGs6%JP}K8lP8V}_ELNBKTYMpoza@HM>}8r23LdZ$_KnpTsPUJHI9ffEF7yt z9%Tbnm@Ffp6Ij!PnEB3{&M+#dY(Sx0`@Ey`IgFIU%UpWKGocWGsR-gx}`4bq*c$J^U8dI)q z8;9n>&Yt2J=XVnrh+ekF3oADgW%o)!c=vIv>TjFqTd3p_q^{m?^RR!oSeqTtf;Mc7 z`i7#)HVilp)|tScETHOMKv=EpW8WFt89x`tuN*na#yDBU>oedm*1USzro;Oj6jzuP z9vt=quVc*Qg_LzT5Ov*CR12TfJK5`f*qn|#XCg;(zeKsbIamT-M<#CHR@t|SF}y_G zSaPjx*Zh{%1vja3j&5H6-P=GFEE@J^w}%Kbcmb4f@VsIS{C?Mcy$*B)_gDQl**+Ws z#0^KfQorRbQK^zTd5#CDaO{uB7%Bv(ph@-1{T2u-?%VF=jQ0=SV~2SS#IyJp51L$W5QvBWRHN*V zq1p3^rBVS1y~S8&<(}!(MT$^a&0vHlUO4#;?z{9$ZimQEh;}8w?ip`8{{>Wk)c4l( zVsXNXe4P8aG=Rk(ujZEdt%IBXf*@M7h(=*B`^n_bh{O&*Kr2jiRrK?-Qr0ZZ_R}<~ z$bb~RGR7&@ETqho2MJEd%;2F3jyWeQwZp$tC#_2#_L`}J69b##^jNJsp2#&Vfb{hy z#QMNH%_%y2n8`_`4M;XnVv5Is>Z6b%9Y_7_vYNosaoj}H%xC+QqOIHk#MzGEz$mFa z!qaQ=?st;ODuBSU&f;3d5Y2bc__(-M8g8H34(Aq7=6i$oy1Nbt?xT3695GPeg)J&z zfW$&?3Bn7Qne6-1+PZ%W8WsoWmDjmLX`C~FhC$e*kN*-Z&JuVuiQDEjBFx4CUDE3j zOXdc8{F z>>cCTHTf2pWfIRRZqRr^zaTcsqvrZD+3^1?vXcZU4y`A0{eNElf2;6+QTzueGPMWm zudvW9P@_MZ02;1$uEpDeg^Ihe-?l$ny!A4%N&e@kb(OH-!IOH|T^h+6mB{Lh!+u{M z_cxizQ=yh#C8BGmp={tlVRR<|k+E9$6Y+CKlRJ2W^m}BPGiRSo&1mK@Wbt< z&xUmlXlFxOCN+1EXLTR$atOW9C>`Hudkn>BQK|7XbH>@w=> z9mPX~k*&jg`Sh0jrzt*5fF^NvTX8p5wD`AcZ)YUr!oPVuSt4`;KW%#c$V|)TY@2@5 z_X9VW+MTZ>|4+`~#YhYfT*joR?U0f-Lj-kYD@^YuEU8y%Mv%(X7+0W4Y=Dzq@U;kz ztW^L5!}j|9v3dP1a(@cy95?!JCSS8Tf%%uIx#LG7MH~A(C?6%wmuT#kxLH2%XnlV3 z%#+bhA^QWSmC6m}^TWK+zhg&vUG<%cM7$TFtlLeVBQ{?TZgW8s1?3s?)8ilrcm%2W zCn6kJ7-QLX-v;;m6)tr(|70yFxnr>qEO!%4V6zp(V^`A{Df)0e4Wh#z=(H(l4dW9F zi?=VUz9Ay)b)KFonV9@ki;sLz==pkLWE5)I)ZH)j^*&;z)tf%%Up=3bUC{8*?P&-~iI?B-`s$3IYV{4--6t78X=T#nh*EVr5U z*ceU5XwBAWh+ZjyZ}O87JwH85WBg_^Q{2jvs_GpT5BsOZpZ&rYExk@(T$?-M3%J$3DSr|HRWGeO~JxhKZe>R7MY zxr_D$IcdWY)Trk3fpQN0T6DgjE>Vj|Wf}TbcZC*%4Erc0f*k6fW;; zt(-?HoQwCzAjFbO49$_SC24O=6XA7S9#6vV<1qkC^Yku<)O{9r+bs`1-Q}A~g)iblhV2^2hC2OKti% z6%s9>r^8Gvm|dD(VA+a*9d1rw99JoFe7Q#MR7{TIE{WOo_cCPPy~Kv^S7@EemB=L= zjkU+HxIc+Q3cWr%&L%3!ZGqVL^}kk~qX-Eu*C!6~TxOGGOI9honc%8QMCd=>1 zMKnQ{n-QOZ5fH!x&rG~~M}{9)+iAMB3jj&wpr) zR@8k4V?1J#Rfr^+b^sl5ZJzJP1eWQ1Nu<5Jf7Fht zpkO`==|*Q+pl{0S`d<3$Ggjiv=0VEK{N|y4Tg4QDycfM0&KwrUlo^hauKNA192ynC z^?+4$|C8GQn(^sPCv||KQ6Hq2e(OCT+=XE?9e4fuT4>*j{bosoQ*Xi%vB&G`G20eA zr?%O3K-s(YMhYLxg|>pZ>|swSJAY3Qm>F>_B1iJER&i2+*U9KG#WFva`81T8|Gu30 zVF7uE$wmKqQW1HiLTyuuZsOJN&Q~OEXS#Iu<{7e9Kf}^+K!OIt#5=*@o(?mi*~8^6xFI1l}qWH-JsV^uc1eDl9D_(dS*I zmhK$&gcs+3sB@VlUMbyDi^;ZS*cC*jF!KTw&v4Sq|Klz}x$WK*JNZ7=-ApE0#VpMI z3d29Sp|9ZMRLIUb6W6+woav$#EA^K1yx|&q?3h1$8*BIT^+tpAZSGk1pAGBG#TBZ>y;k7GoQl)rg&i@$`|$rTUe zq?DHoxRe&3Pu3i{nj%xPC*`euYtr$~zUFHPB#;e2v`=XiK&!_WpvU;1YJ2H_9?p3t z8{RAS4g=VXS(mUf3iz>L^(dS})v2Tcu@#=R&~966&gj zPXN|m6Ldoo34)LsK!Tu{fnMIXBu8WPu7gk60f^E+Mb1> zAu#~btc7T1&{Mbt3=Bo-dxClbMDIH2%4a2;5MEys&5nS|kSj>7FV7XEG(=9mfLnd@ z2{9xU2xV@skFrsccNJGKG*|ah$SwkfQOI3Gut`WEsoZEv8C{5fvqCQ-t*+$5ZTBe! zPBV|B!oQ-cs&%|K0u>;Ss?Q)y(DH5Q- zSI^<*WF_jr015%O-sMDg+P0kbqe&kJb_18Oa)w;ECR5=6~H;7~=(EerUM;@Qph>w`O4;bxz(!`31NbU)pm=OhI zOSo$2N}xk*d>b(FCg_;N&o3)2n65S>6Iz8$U(5VbB2ufu%*Gsq<|PNmnFpNr=wgEz zZBd^Dq)3_D2g{ezccIC!vJV!FR6IYf3nn==lBc39E0-SQ~KDY`63<|aO4gafLdPq1J2A}*GEtt*$A?b zzv){z0Ia&D|2on_$jye+ui25ecV!N4q2T5-{hD`l7N z|75bPk?6hgE!n7SCx(IG{aWJ^9;-we&dJY3)Grc$OKyH^7B9^E3#}XL?{MEE|Ku$k zUY@5(g_T7Ee*BeDDO+c~D1yy>Jg?Z*_K*IJ*GiFo74CS|rH0L9y3 zK9E3Ti~tqQ`Ncm0qhP|Yog;|S737~r;wvyxzT&jkXzV_MR#X7i8GyW4aEbz4wgS5T zIN)Id4~~ z^MrI?{>&Wx6+!~f}*Dvg4PZW7A1*JK|`KG)$V z^d;)ERUJAVEyB@@A1M!K)*WG->NZ4V?-!_*)E5D(-?T00n!!34@|H(2qc;zUgvdP_uPGF=3SBfO_SV6(1i>)A44*62$+ zqKqs;;Qy}BHWEcef|Y&rQgDRm`iKD&!OeTSDtRkrDKOO|Bz+2co#J*xe_CY9vz2^q zRu$&wj_Bp@R7YRi)B@)g1jXBC3W=l&wJrbYm0vkuLe(#Fwm+{UoA`qhlvWt7HNk>* zY)0kPWC{8do9w7xkpR6bOK(g<`X7FEt=L2#)bGC<+#RBr2-dt(Xh(9N{yv5-j~ziW z_?v$H0(PPicO89~$^S^5rvf6YB>~YQ2-b}NPGjqF=3jGzM_H@G3mv% zpGV{`zx5&RSM}DTs~TphR4u3pk|#s&Kn<)7$hV`~C;k6gL_q$CbK-q@iEQ*jd_ioA z=l`~N2cQL>o-4{t8R=BLr{Ergss7XHaHZf@2s~x5kI~Kzl0idjaKs^{7e7U-p4I~Y zu7DnzXVimrG2vGmueIB8NJ|hI!8HvEU#74HvAt>C5sztKN(8N0#NubsL_n9 zFnOy=5yOAe{nN!eX#6P-^4i!w*sxrT1Q&xP%=BfZ-RZ0KAuZtG<#mo2uim1xyF+|1XpD;4A3G#+*X96h?fYCy=KSB1h=;1MKd(ND;on9xOP*` zzl?Og(RlxDTl1Jw|1tb8SbLenYnt#YGp);r&6N5X`6-QAx+Z*>Mh?ejU=sFK3`_iu zf||KNy8T@L)9;9kd=CY~RpCBdAH7R__NnW{|G?1L!ZnL4`s9kH{N>qU zp3l(njjH=S8T&yD>PrT+U%Yht{xt|LPs)JX;!QYQ9UalU_E-Jgqe-gn z^dFTe_&l(=!pC5lJ3LZm7VuA;P}=XIuP&e+uNw;m^bB;nu&+#@(rE!KxXUnVPe7s& zeH({#wP6j#Lv;JM|J(l|g-BeWrJ9mXAaXkqRg1)bK;dIJ1+0HlSW#o)g?JFS^91%& zqzn*5Q7$iz_mOB13MJBJNj`gL^@gme3i#?OUU)aDYDFx`i`fBEi<13S}k`D-fFrR`vjw}tn3JQ`*pyI?b zI0J6GV|%TN<%s}XvEJx9FUriK4`3~7s@vI5Di#UvbrPr={pFt_||A?E5Zvvq+ zrKUbTJIO)sC*M1Hn*vCF7f>6 zpnDy1@(+#nqS(KU%XYP@I4tS*FU5f9)A3R97f1=9R`jAg5cgLo*E|&Lp18d=BjCC& zdF%Gz#O*oB4JJDldgGDLi=qsSjt)8|Z9=7e#uHFkGUv+T>m>m@Jzsh$Atxk$he7Hr zv7Svu6OzZ4t0u(U@sj@+#Hsf!$-)DrS0I7C`3Gg>V}j&lc%7Xt>81v- zgaQkrvHWsW>i_v)XL3cYluEo01H&>#N8BW}MMwUxvk8Gr*7WqcNB@0;k_J;N#P^Od8m1^YV8N4Wysz(?QRzikZ>CgDc=R@BtgNq7{ zpJ})pjQ3;v=ISLT$|rr<;^7!4-lv#$rBwT=v!AKTt&gA){)5Q$5@vrnS^~)S9%?!R z@*NDB58u?>;QY69b@d69Tuv~N_;V$P|oRq@p7pdcQh3pCV&dDu2u z4xPuRI9_-^rN`qUj6~@suPN7A5zK&go1uqs9RlXB$P9Vmp!V!(O0p32PQDE3XBqP50GBGm)v z*PyIId`S6+3D0Kv(PY7;umcIltiX+bpYjsAD0Eg|UR!%|_F;u9%sQR<)@(-7V^(+VP?qzO*!c%H@h(4jjiwLum@)3hyBe&e#>zDYbg9-hy)`8Zd-tk{1)FVy zI_Jk{PqWX;{kc9Iv~}!Uf!c(^MN|2lfanuA{U@H;---+S7@Lc0N7$DvCr}kQT6a02 z2c;LV@I_*jc@LWUPr&H)#m4F%l1(#MnUU}{Lcd~SV)kPaf^wzlCCSl*6wU`XR-bv1 z*6*Mst%X_jcax>bdRqir?TNgWfjfjmKraQ?zA-i8Ktc{RPy5_|@q=y(1=IGm; zWe55e8!ky_?6Z(M2#0JYnQ(H))tS4fbU@z?h!p@{jH6|j7#-3Nd3gI!`5gt7AGDvl zwRcH7a*QtX>wzlE0p}rG-UWGqzdr|%N8h8n;vV3A6czW3^DjW)Q{*`<`ODM9VS z4nqaFO{(tf=g(s$pBgujS2j`*IaKvTtmO$;?0n@prv<-c#YPR46?JJXN#dyGi60L< z{eG?~D96v1Be*o*10RCOH82^_eJA?RNc|w}xMa(O)<$nAg8?*Yx;u!XyRvTO;Dz3r zSgUYC7 zTiCQB*>Jqg-W=A~-}(%BU<+(Yo7-!jL(tZrQ{XU-3}bp<-S;UN3PyJ4COZY*PIk}q z3}yN)Y&{nTCUPb~=aG%+8RLx;DtcSo#YnX8J=I&jl*d0s`t2KHiX{Q;_{RIW-wBJW z1<>o#@ta9ZO}iZiIhGuNnM?d&&WKAno|{@M)0HrJLOYL&*)o@?^>*;j_=JmP?aPBa zSm(msB5%oR$m?Iiu^iUi&HjCdlmgO-Jy!;6+*Q_(wTL+k*;?5$qPc<=j$D`y`1MdN zd!YFHj@t+ULGex1;%hf;w~BY*QE>9h>{W676`sq@4w_M_@yBwF@&UjD@$BzN48FTc zesKZ?%hz9QE-;V;lR323p&v7;hO3}dcq!>rJtn~B2|yZ4q>zG}ovLOA18Z_E^x_Vx%-|E% zhmm+M_V&&UVp|N@am0Z3@9hyKuV7pIQgXClMV48BzUU!EV7|Yh)4<1~_-BP0Le>RX z@H?arM(APN%;IP$RGPsRZQwH>VW%s@hvu-S75`jyR%m|%oHs(6t<3K|IpY@0iR|mQ zCy0d3-SKtrf z*;`(=6JV5{>}UaQ5;qx<@%Nxvs|=$FcXYUfONZAzY}sY-h}E*8DI(kd=$WT1bQI5o z{s9s;qL*`YciW&PjZjsd45I*DHE0;(;*ata_^ri(>tMMIjk-rnt|*hKa>z%uBY9jb zm6jC3FE2)J4ebL7u7@VE?96-ZDT#L+$Am#NhZM8(lYj_q39@)=rF2%|!xwz+1=$LW zWsl-;ebMel`O0NvhZk6zY=%TW34wC!tN^uvKv^*YpN||9uK5vB77bhQgHN4u{m6$lPM;ZvM|41y z7;3bdhI?0V)9-~8@1oG|Ky#BapRYgk53g`jb^+w1kZPJI$&-h?IhuVK-cJ=)xQWR^ zH@3}nxg`U-^(;{IR@{?CyOFyeV6H>BKNKjLKjpdJhb4`kR8u2V1rT|ExcSB0z-#Co zLFg}wNf4@D$SW9H0(00v8F`@NZ#yZym<#BxGWhs5@&o5HT9?2(?xQyE9=H8G&agE* zp)%Y><590575eAVtlwNvIBTB`z3yp4BG!4fg=uQ||Bs8!C__e{&jyTFBXf&&J`@rY zdp1Uz8v0sR#XpI$3tg!vrttQ*g_yR5#9%}V$rZA3I#Ls#J_~S}B(@=pj-t9Y*g3p; zxit#VP7@Fl;qKr2YW@g79_p;DV&h~#O>rLZBtXkiYa+Rxl&lGGX$aQ$77_i+*X4#U z&DwS3n&1&r5VUP-__0QH%&{h5Ba+lSnuDU*yiS^th>Xk{H`?GcHcxVpi(ZMXGMOD# zv_UeE@RjVf`u9|ByGkBTc6NbvK;9Ta&&f`x&0A@pr)c_*22Lx;4~H{afw#lDf)w3s zo4(@dZR03T>BQ;j zM@+-zF9;4rxkM-Z=)+#nIfR#G7SbWG2F~hspwnf8Rm;ThDjkRnEjW{M3>*bUHLV8e6C0-eNd5AY#ZnD@YR6UndVYNH6gstDP27*VfQFn6YsHT^gE_M-Qp zkOL-L3v zW)X4aGo~jFY`Y%M6&1TvD={oUc>Od;lKc(2(Qx!n_K4FZN1h^@a6k7o2PI+g)(;la zs7y4X5_rL@gsg5t?pG}v*;UP>uRH>CnTnuY+(X2AVW*|&hMuNLeM6?5^!ITl?BUh2 zN5{q})U=jO(dV_CEhT5E#^@qjPEuaP@qpDPTZc|NZ1`4W*dwm10gH?23IXXRG5E1% zFh%r_!v8xbU`~&PQ~2_6nhVzgJfUQvo9?AueYi~YQPYN1#7w}+?}Wn6E!(Nh=*wiL zYUqPRl3{p5$7F>rFzqJK?`tYN9BP8zN-1IbufiL%w+*^=M=tUtnnUXUwWrOC+~;u< z(SY}{9BL1Hg|-!Wc)Pqmt@8=7M;@qVePyDOblCECS|wxz&o&}Qs2pGKGjcs<*J z@*d0|Ct<@s+<#k0G8jz_(i#PXzlqtuKm09jZw#G3qKGxU9v{?Z!(J@OivMBOl8>0` z*A@SpUGxv%d#LP~Bp|ONN92OHOE?bL<4m1G$g2`I1Z$CFtI+^WVK7ndWnxWB#BkQ~ zH=e)5Un?`UdqJr@@0~76JHAkz zAkNtdlBmVhC#9wsE{lK-{dgbkq8KiJrvcDQ>Rp{0!2cc$4VvWo+$+SIsQ?MC92=l2 z&q>}5cyaE8Yuy3J=~eT@_R1leJWmi+Lcl7(#T-4FhL;_=D4fmv3^Th|ceCTxjr^Rn z#AGdEjqvl{BWCw-&zK0&CYMc6QIXZb(E4Ssvkbw09;5)8{-D}aHA@4izTg5s@DhEk z6ar0RM8H)oSyU4}bS4CGL8@O4UFP=^1LUyKUaa3?4;)(AU)<#oHPgbIrpLaC1MANZ z>TPeU3`F;I^VnW^!me;+-miAPfyvUzoJGM=fcg`+-t8 zKf`+$Ctym|DvgJeqX@7opGZr(>*iuy8h)4X6Y-j^Kh)!+J{t}DnWj8kQwAaH&RJU< za`epXk-ZjyE(3Hn89dwzyo`gmIVTAxf9)ueMMU8TdX5^(d6+HZsPq$(oXWh z(QUPm_6~5SP5P29{WKuNWhk?}LW@2PJUN=4_F!%@(zb08tvsaPXyJrnJ<@3!0lLMd z2xNID5$cjmU(Rp8=egE4|I9glsMUBtc044F-9@U;C|-so~9MwHUSgcr_N04)SZg%6^i z{|X6sLHM~;>q=Wkqg&(Jr+B*_PBy{EiHTv6+Y~nb_i>;SmIstR#npWJgpfTZ5Ndru z??Tj3(E0Q&qkPWdjH=PSnc}@Nhz$`lbRfINP_+5`Z6qL`H(|jOC*b=Z2a({a^iBq6Rzl&^rL_;vF#LDMm%B{rq5dMneW{`A~=W_Od~oed%2N0C3%_nIoxeAa^51VKWwmmvbH`B=ib-k} zQj7|4xk3NP!(!tnD>Rf3Jw5z@wPq=aE7>Xrd`?A?!Ov)35RZns`U~u3gcIhyOJMpr zCyb$L*ahf8@a_aV?;$zVK8@1^{n%eGA=JCXyz!I zOoepm%akWbVYmvxE;4*;B1TD6*g-hTbx+sg=(RFx8q(J4;0y2UXd^viV46C5$%zhFco1>Ez(PZM~xfI`pCe~7y>Iy|pK6Z)Gt7vFc+ zadM}?1tW0v^@TCu)^fe%CqS%&4W~uh^4Ba{lIs@ThSruDXr4lM!TFDh&D+UTF<_R$ zcZcg*)hS^LV88Chf}cqMk3bp1bG0`<^BfQ^GyA&$Kf`*e7vbP;{xFjaFy`?-)jh7c zQQ<9k81KcGi35{(@OD>D_Jh2rzooo#n4akP+78Dv^>Mt9-I!Nkw8W$Oz6fY@6JPRf zbTt11Zkj55Ia@yLXpB){t7!n`hk&{lC*5~t!}Jxj*uhDzmcHGTo9jzdL(IOQ!8FeS z{msRrGQL8ISF_y{4a+=OLu^$5Yhv+tncg;w8)&DOT<0w*>Q?F;U=dzaZu!59%7sV; z|CMf`C)6M>-P@Kq3Y5 z_P0M$I{SJ1Hbl`Z(*S!2q5u(liT0vxrc@gw7})NB{Z~f->`lki)U-P!@XtXAgF}IX z|G%#u7#B0MXFy@^qE~aqJB!H?H#Mmm&yKg>Vlk&buz%J53BX^tuRwEv(HqulKeaQ1j1`2xRZuQTA%AJSeEOK)@F85b;0AqNEx>rwI z+wAe_hRqaPLRW^`)B}wlVKotHbTQ>8&4G#LbgmM9;)T_3diSm5Q70mW3o1JHXFr5h zkY}th-f}t$AgqGW6x(#Z!D%O*EcuLo%awLzQD0F4ktUc_ukZ5Lt# zW}h%A*?~{LOM-oV!KTxT;FyLu@;snQ-ZIz}cOmd6c|s5yWY`sJ>}GdCAnF4Ix+s?e ztXhWjJDUZ5sy3W0da4VlmDk`FK9sxw`XxPNyd)gTZ@Fl{t@o~=eQo~8vT&pKS<#4f zxhZGjF6xVFYNygf#2zLNs;*-Be-9EjAF(MTKKE;EnQ4&ow$S3$ASyxMUD@0n}fBI^PvVt5%H#PXR@uVjZeoAttti zS#J0|h4p{`_%`-htsRL)=l_?BKv{un77laF*v-|)8g1Q)O(FxYZK7RhbvYts^5Al= z`#kuIZgY#`k=fm0lUEOKbxF#{{@5(erSe=meaR$p{duYg^Y=NVF)p{uQxB7r`Jabw z*QT*D2X5`xC9|m8 zrh;(ZT!VQrg_^z__IM{NhXo6C30}>aI%;(K8aLSMWlPBIm(`CO6peo7wT^YYG#;z5 z8OSTsB11=xpQ<=h37Rn_r;q)e&3G&7f_!AM0~|5V8V=da3)qHCoQ6NECeir6H&qFv zCZIC-NBGg)XH8_Y57S4cyloUUq@4ecQpN13^6l;)o71(YNrjQzI}VN8^)+&l&8&z? zP19AO^l7r!BY?fBVpk7p8eii1J#3pEqaN;^@suPl4pUSclYV zoV>hV3u?i+)M$*5zD`G~;x~mFh2;oo#LYrKKPdgfXGMpV{UC`7Ltvyd&I<9}eL`{9 zHC)T(y{S`o;c(izOO8_zzzq3%)l>ZVYxCmnE0x|<&!6a4goWJDlwQZsFQxA>VytcPSO_(w7j0Lf4g`?9%R^ej|612qLZ8 zBEo~bEUqr~UKLj@N2PEtIcH;>yVygMi|$A&`dxioq2CS5Edc^*Qdih)Xqim4P?E+s zSsb+luO$aNC9ylMJ;S`zgxFed`fnWxiw$02S&0=Nsf;mK21U9hhzK|5$G$apxQi8L zv}AX|vbERwn`n4pUBI-(B`Z(T`1;E0V`GJPAi6BJwCH$7dx6U#_ES{4DuMB2hBiyJUA( zUhMpoRoYs_xVfea>*^D_J^a*1id@jP9;o_tII>7%V8TW{sn_X2-@=mbX5jYiFKy8u zGqo=ZHZdnK)m^6ae;`53P(oTw>_exL{6)t4v%Qx6zKK;Aent#$<`Ix~Y zR!{Yxe%bb;=&yD1>s+)9@mX}Hd(#C*%^yd|ekG3s!D@&yOt)oRXS35*)DW2LwdCFT+n^Nk*Lcd66^2@; z--*n4cm2(XUwt`@fM7!^#|aV@YLvz!TvmB5 zcjHviQ?lAz>d8}t%ne1#KoPCVLIkB5?a~G(rpY*s=JH+b&Pur$sV{u zJd|ngQO_u8Epl7y&dB?e{S(_p6YI2heQb~6`XD*o*WU6=U^*JA^{sex-^UY&xf(i1 zjV_`ZOS-+U*9%*aS-Knwx+QI#^(K0EM;%P#pIe0ceCB*?i2a@{ z6~fZv-kpAud%p=A2j2(ZgPP9*(Qx5q*CD_qO9w~-@_7G(7Z z$)XxATvQq_0_Sx%!I1vUe@gmx>e~y%+00IP8{>BPM_a8PCFUla>ZK>@VcQnw^xWqd zevW!Y2$|Jp{d(ktm^yqD#2#Kr!|WX-D%XguI1h?IccHF>%HODzgOMFFZiN7^Ch#KB(3?#PY*vsMmnKe$+J__Sux3m+gme^Ilgi@9EmtOW|hCubL2i+*VtI} z8Q)0>yRw8!9O1=&p1xjV4?<-drnzrB>yt=4jRQc{{FGMYT^e>vsOVvfum#|^@qLL=o7!aMPRrnOg z$|$?z-S{J{%t}-#{JOb=GVG;PcLdhkDf=McOW?BC7w-h-+27dj47q5v$X|VF+j0^N z=a{UsKdo4>D1%mAknj1Msm}-|0~zQm1XGsNB3$OEzy7%@6m70ThST;)QnYNf#E>7L zhh!gVfw<=jp9Eq|&4zx-8h!tSqLM?DB_xsbj>5~4`Agq7NzMA;>d>5A>3yscHD>9W zz99ZfKv$pt!u@G*jOKTg;>gKCQW1zZhh(cP?2rx z;ycf$!MjU>(@9>n<`g!*JzdjBRzdhN1-8JjQWF#9{9)Y6VDj%6lSpM zI%FVaGv51$aQRK4>rI9yxU#kzJYM+`r>D9DD|;J}T?dDxF8;((f#!+UV8geDz!fz7 zSz)5kb=>}qmHCmDu*B9p;Zvkqdpx%$9_%~{$*x&0i0bNB*N32|Nh6&WoJtV~L)Yzf z;tyE7enY+)0UE!w9G&4fwe}$g?+kwBUhl9Lt_$cLw#hpN{xd@JR@?_m8q$YO zuP>;PBk%TSgt-2mMGXh)BA#TDQ+%}&i>Hqyv?8Un{1IiRdhfEQJg&O>`)uBC*W*e| zf*G!1iP7(ifhRdT+y3U^Au}6s%Qf~H$77{$tFB4xn*Dls|9C7ET{Sp+bWmKR)z1qV z-mc#*0-umKgg0(CY&@#PJoe%z-JS}0)$h#kb;c-PbH2tM;iZBQDehkmowB+dh6;^8~+dH-ZChT?&}u~AxLmYfIx5!9)bjdOmItZhrkouT|;031h?Qp2X_e$ zK@(gDcXuZ^49xWHJpXg=``)^BtImgWs?L|GU3+);wC`T)C+i9gQ8xd@G8sk27av{v zczm93h#or|3(+e~!q9))eVH-(Jnl`o=Qq^jO0}p9xOBz68k&GL$u~-rCC{-H(Ieao zIJyH`lkrAj-iOtvV71Ha^&xD@JZ&JkPa;|GGiy9o`O1gF!PlWWvya!wCrvm%x*vog zFgQ(Y0+-j!Z(uX3^oO>Z+)xy2dvFk~vNi(}5c=*#a;Udi+xPV}66cxZw3Wa;yoc;p zs9R1Ke)M*xH|nCB?D_ZeE$F0dK7Ufwo)Tkpx=w@JPj-0rg0k&UJP?+faIR|C*Elzz zaJ|u07Dv2{_xYRS^9vgtA@;}f7E=kVM4^n88{grxKVjRuuM=-0yQu{U@M_#CjT=1& zi*gDcB#sM4m=b5`q!OpUVSPsf__!27oNF9jS zN6}8*wa;rTDL&7QQ@nw4kMe7mM=G*sL#vI6R+RsCPPYq=Q@s1zLO_?tMGB(S6qwH; z{bsP~@uEDE%Sw*t#s67*A7*YbJx`H%K?T_g{eu6Il?$w`IW|y~SBY3Dd{U|KcM})w zGq-dT;EHjGUUlGbx*rg(=>q}WL|F~6Y_NX&HG*cLVDg|R)td@7m81h`<^cz20)6gt z{yK);YOy7Ium|gF>}P;gJusMwC|8Uc*rrX)qh!D4e^0Dh0+ccoEq=}+W?KNXqETzt z0B7EOvpUi^D3^hCY3I8h~k04r6<4@(_JEa9{0feMv6&#g0vd6eRB6>C-iJ!%tks=F#Few zlimZbGM6FS&)~cnC#j8IVCbHnf%pDQYA4^F*ky#r@enWSqTVP*mAARp^iWHb&S0e> z#(7V@3*LUf2lvFkvwzAr1qPhd^a^L|QYB5 z8x?G7Zg{O#Ug*m%z2y;Kixj$lZ=1VI~(Tx_cKcLmbTbEl;YAnuj>YUbNUW_@|#LmL+v~{X1)eA z*Gp892REGKTqU{dZIkUU08`^;X_VZMVf09)|IWHc7tmVXf^rYGF%)s6e#T?Y&)$z) zczq_v?5hQk+->PpU9KHKp)$J~Te9ZKQ-H$*q!h_ddEjL?qL9F7$=@lWU+Ld z&0g^U0~<&%$B^tX-t+XDs1s?rrF6Y<%(78=XIG}>`A zDvrMG^F8cYb(J=l2)p<^E0M_;{arKwO~QBDzhhI zpxkDx;DrSBnn-b(ngU3Gaenmt9iAti|Gf3>{Q%3NZ6R<6W;bEFW|SQf0|Gy?Ep1lYIV*&}~A}L0~mY;A%f-%!_w*9net0hhomqRSS1Xw%i^a^ta2N(BKq@Jy>H;A(yWa;1^D zkju8yOZjm@{X>AAlGNLuQ(_)$T|gkBzpU$kQh9)BgKOFXNDh|8co8vx^v97)Jpm9G zr0mzs?zyOwedr0P18stF@zsMf2+#kGJDWcw(CJvi5i&#z;aGLTpsM)NN$S-2KiP>i z)+8YM|6TeTN=NWI>vGUwEVr9<82BFY50|;+@@pMKAk{qZ!D)!|`5MbR?Uy;{CVZgv za$`Tvfw9+H6OS_<5#vLnHpg+2!-URXi4jt4xu`lKT-E5n&K+m$8s8W=AEJkwGk@p~ zxVIQF^Y_}m5$*kb1d)o#(6bK8o!KFDq*+s!mq*)zveS3mpBrpBW*c|_OMV1yzvsE3 z3_>$@yvM2dcjO^14s289BS*|T`+v@7=ThvVh9~($na!P?jMr4(k*+&+PnzrCVHFyQ zyrd=I4qfCLSXZ^fS@=jPbtGy9X-EZwFGC|}xG7gF4J?RA|ID}dw(+9z;4FB290lXI z$x+LuE)b=1U@>i1#2cDVyaLKTacGoNv(TgtBb9&sdR?-n`l@)cP*Eh3L}>BVygxkN zSv%sgQW8)>Og-eZ=9+GAyMgYj_l=n#OGklti7pxY>Gtq@$ON7dKpsM3_sVo29=G(O z<^27eQ0;}FT|zGOkccDXJ!LOZS=Kd0!?jv>TbakI%lt>h zdh}~Xs$y-qe=D z0Bg?YkkC&7C3E6<3`GMeYTTu5tTk+nPk#JP6U7NAFR&!md+E z$b9}<{Cn+<(3iIAk%&3?Pl0CT;z137WG$jN7P!SNtZJ6KU*0V+3=KP5wZ&SN5Y((YE&7O$lB|9Pab01 z!&M~W#WV6OD&4b*e%y`FUi{(DEp~Up>d4-|DK9WaX=9sIUWxjUEOBJmsis$3HN!Hq zd9i7a#&URfOFQ$Ot9 z3>|2zLt0}q%#yeI?afMK_lu&8Ud!a^!1>#ojiLyXZ6x~!OTYzUWN?3he9h@H3#xCP zcSy~?{DzqKtLa3I(NYkRqL_GZN6W0x7pVkE&3x1)&Do$~CP1upevFRH<*?q1cir=O zlpl~Fdsp8;{mU0rN0 zIt_cLo6v`@KYWYAq^!-V!d__3mdFEik4$J|nS;Z;F5o_|tdmNhY-0qX>G3h>P7DCy z4#`ZH$;u0vjR*U5SBJDvA98oJ(pLmo@<8L@y|A_d$~Oy|8Gec%gW-BOVcCSb)f#&D z=Bxe&nwEe&b}gGA!2?2hIWDU3ybt)>li!`s?qj%>y_C2Ie`15UEY|po5s? zy%hXN9CS23u2alYM-|FGT62zZBJ zyVJD-371R7ixsT%>##XxXd54!P88CP4 z&ffODw{o{oG@^psw>td!jZs^Y<{wBouGdCI5f z@7=`SB6ezGlNYj!#|K6s@ryB-X8SEDyTgnDvCkmd$M~gAt9E>O1Go@P_M+7y z?NJ?_X5LMqlBy!ge9WIzMSN+fdeLoOms+@rW>LOhGIcJVV9*djN>YUvVSzDCEGn!t zulzoSNk?+^XhU~E8QYg|PI5x<&DH^}C>=GZ(e4>4rryY8D&yBR(H5GJusH5p07&I) z<}?R@r%_Ur9o&^6MaR{xTe@@GMpUJCy!M8B_wdzL6S6Tr7dj7~%Ddh`t84X6U;_s! zbv{uaZ zG}qxcY#qe8qrgU&LVLdSC}=Gvj7yyp^A=<%R8prK{*hl3%0o%}Tq-x$+q0VkooP8hrbLQfSwm<|9=3Q(ow-9fvRn&&=w zq9{vj>5Q`;w1mLG33p*EWT$f{yz)pNx#IkwwW?J|@Ti_@!6(@~DkIyY!eOr_`UgA6 z`pJ`3{whTrM)}iNv@b+z?jKC#`FhVWG5K#FnlL}_yH(9Vn^QTHS0@qDiNikK+kMWB zG+|XC)tN>(Pi&ij)*`@m?@5aG-AV+aefyu+Qx<@a8mjku0iAUX&^ijZT2oKwk%6Q4 zkd1!lqHfeR1JX4TPkkQrCOre}7`DUyGw9h*mp=>+0x{eI(YBR9YacL?aK&kV)lQCp z+F8wKd;Gf%+OWnxJ_2=g6zz3xf*ryUxO+&oq4F=D`~$H3b{m&H3h|5ymLFbnMJ|gg z=+1a&_+$biM_26t^a|**t^2!AhTm8%=8vE}CD058$`J7XaW7RoFblW-{{1=JOnB1h z?=e^+l>R-edw5}gcR4ut`yMWzdX)`v+l3itL-i4tr%Zs#0sHhOP`qrl>eudAjOxB^>0~uKgnZE9m)PCAO5UO? z$~ckzfEV;`HfAQ3x0zy&qH%KPl><((3;)TT~ic=jXrhSAjcJ*@fyuK9bRs1sP5o>$mP3E0d~Uv{3o z5wo05A@7@{!7Ao!&>s6X8c?&K|01l@J@<6Ncn9j&V9HN$#_X_dyG6{(or_}5;GhlZ z5_p~szTMRyBTePbNhVJrQUH-&QtXqT(#XM_bi_>VMEWO0NW?DF&I>sL5{{704^o2B zXT4|%dO)h|#t-=Jd30ye9hhS7vfGSmr2CqA2MqB#X?6n|%$YYK=?s8tzw9*E*Cmqng@=+ZJAIa%>=SAyX${r*Y)Y^N*_kTXW2uE}SJe@ynSv9AE) zU+l_U+skL7T-vG0C%-1iH?^$c_5S5`yW`|K&n`Ic{bh9*BakOg9O9x8TORy0qnoUu zC!aTdOdUXhM7IR?OP;#B6*zan;>c`64BXjnb@}n@DxV;f(3`QZ#6Tfe?XWASTJ$ig zqPSEA1F~Rm-dl40ob}F5@8+#euBha9JOt}RizMe|^XU8fB=_@g7V_&;<3pxE&W4}!5o0?Gmc&RquS$3kVTAKNZa zzO-lt=9tIj_pjRjzfu62wibU=Cx=snPSRxXMND3aUdX68p%qnLXIUygL3OGw)2bat zdij?cQm6WdBibEvI6?+!mK^`PJFft|5y5VRqmAXBxz+n4P8p#Ez6kz>5p@gXVDc(6 zrM)Fv4XSSZm&W-Kr}=cdtg3@tF4&rC=Z;g*sQI@gZXO#w?j)k5cgdqO9+V2MQHl}> z{weTxE80k0Live*7vrpSZMmf-UC`?>B&$&k392Jb_rw!^MNQATv5gv=%q|`CNP6Y$ zup-{oHv2cvT9)-U&q^fuUwKwhy1#-Uy7e*#fm1SWu_MxDg6jOMFXf9oAa zH$)cB;@fc*fTm8?J}-Rsrq?8p`4hIV4^%-n%9iItZZG4e=JShv0-Dhb4cqxuG>omS zoi*LlN=R|Y7dexqAK9*{H!aHvt%u5l=mhZona(w4pBhF77yJOG9D7;aX7jQ<>l*5 zNhER}`J3!*AFOA)4BxQJPgDSiMm$pewtB?zdC1r8@Spb!UhD5XP?!+%Zsxn_xw7W@ z^JU#ayF75>ZdpxXt#R_g<3f5DUui|hGwU@COBsbU3idD-MQwTo+U%96zVaKE`^U#q zf70wF%_vSOaUt>7OqSq3`s=lXkl?G$+$5Nbi{VkqP-<0V6 zGjc$J=$&QgOLD4j&EMLvI~QNy;_`Qm(4n-nZS>g#Q#_ql+U&-_Te|%}o6vy@r~*4h z9quKKhY5?0AFB6~n#}tzy^`T#F_hBd_q!$!qXp~5D6w3}uAsU1JJvg5Qr{~ikrP2l z-&}+m7}CqF8o6Ed>r+6X6$dY?-uN3e>-2?{8!@bU5xMp2MKMU!C!U1PCW))RtcK`> zCPPsn??(+^YlWxUC7XCA&SM>s$ zd^O|L4i0siQC7mgJ%cFUn1EKj_qVi5r=Y|{VC-_j9s|j@>KF@Bauw16FB7h!JgD^2v=C>l z*vsP*m&;1Bgax&H-~gn{!^eZqJdDDo@8LPr6>-cXn!p<-6EbBAZl9-X_qjj5x|}m> z-8;RyX)WRPCQR(SjnF%^(`dHxh{&2wP;2+(mmj*wSZ}4xTz&T@q@i<+gMNs&%fI_B zk5!H`p(mv_yOsTJYl)Fa^41@=ymnr8E|g>}J|ca6s4c9p@X;bMW`%)!@}%{rdRM}y zC7n>c`v6(w#52ZpK*5S}7e?2|0t*IJT;!LF%V$kxcwE*w^+W<>zpDK1 zg#d-v0p1?57~Wy(L8ET22Q#)vqgltywzX z9im<5UK1l|P5gju&x=~_mb%Mik(rGE-4+hd(@l{KOEKU>JS(=2r>Suq-ZO^;u5jrzNMW z`Ps|L9@SUyJI5msTSyg+UMwxSgSPF$`Xlclpx+g==^4(P%g%ON;izfE!Sz?CwtuM& z-R-n65}A@ckE_9jZ(P=%F`qjGZ<_s^&^bb3kumslZab#a8lS7Uj^!Y0Wzc`deWe+V z7eTBx`HnZDnM?rD7D#dnTiP_@3O)yMT=*RGic(@%N-a}5ph*4umT;Ec!q+2OsfS~m zl5gwrSJi>zwZe-OOOJSiyBAA26@c&oo570O;wXROzOg8xkUUCPZ3PIN?e^nJ??6X( zRwaqw{1d6=$nS&|p!&DO;ag%~k+VY*-oH34X-;DCF%%VW?zpvC9YWH%Ec@hk-}rKY z-YgxFs~z$1MK&8FoU5A$r)-KL%2cpzj)F*X6f)>!J>R%_q`sY@)9jAf+p`#&`Al^d z$OnEBcu+W0IHUUtf{DWOo7cB2^(R90?y8YGM5a8Bw5d3CPj;`j(Z)810>oPbqtCed zdO4CV?`5AxoB#5LLH_AX1_L1TwfhJshd08)D<<%=1z(x`GbALLA@t>|_AOoPuBrN2 z_V0Fosn4n+|5BfsK;?CHmgSy8T_oQ}&#d{gBXi9U`0j>w!8>uzpghKO z?vkgf`m*v>FNGSiRNdYPfYY0_6^{Za(vkO|#K8+x>IHW0`G;NS9%YDoNdQlGzc+1p zK}73DTWTmSq0FJBeM7;10jRk>XBJU-5~BTKmWK-%)+E^!0%9LYX>6U-+haTsD@T2s zWp4qi3Sh~9`|XPzp!hrBO`npUcnWBOtTe@>clq}a(o;ytH=yZ%@O(_BM~@EUu;%xm zDp@_5mA4Se6R0CHjw2ArO+v9@zJCoy%+$@bjIp|5L=-aZ+?`6%u6Y=4edaH@sj&xu7~DU55-Yd zF~7}ZyS_TgGAh~y1tY|8Q`F)gsS-`+x`k4`0|~&~BeBzNXL%b93RsV_VU{}Ox$u%p zUoV%;?@bwa_NXHM5weGBg}T6u=!-a~yDSx!dm#hoENQs)!-cyD-R65*V-KFB+$Rz( zuP@=YT!2O%R0u00hu^32QmAOE`9}_0-uv8QYp=IPgt*)lsrrm6fUDQT8BbK7lhFxd> zCLfcF{RvTxwrIbU}G_Dsn#w4n7SQZh~UxmVwLIVj5MQl1E=} z+f{bD=P=`e+hTtll%*Up^~1G;XalG>Ia*wP|5w_c{o&eP`JZ*RZm0VCVA4Opi?EuE zSZ0kvK*>DaUQnRDcnV}qkDQ8f=U*DyGEO<5aTR>9K8Z2V@sjRBlM3zDH;48VJY&Yf}~pYz)7l zPvI+e(Le}c9#e0g8Pb-Aje#aQY@BLA2LyM7E6ukqs#ej?l;pMRDj7I;;qZ@rsn@xjd{ zCX_)FmwGv_!O2znTkyVdvDaIcjOqCnwj+Qd@e8@4)Mcd$!pl*WL0;qER)g^dzc#>L z7~_eZ*u$>@u&V$U*+V5Vg*Ghm_O(me5zrKi`^>WZd zFBZI1VM;kPHxbMAn4g54OmzG7uBeE9+LW@`f4mD@cCMnWk=9`li(A|%M!Eloc>r@Y z3ZJTke$ZtLKv68h;J{-4=i|P$W=w>P z(XG{LL;A{{)D#Ep_Nya3oN=Ua>a&l??XRo&&D)G7AvbRvvB`SFyr+g!7^3@`OswuC zGM^Xq^IBW+<~-qxz1b4KOJW5!lq|B#lTi*lS}~>!3c@xDQJMZQckg?MHF;HQX(jY? zqpk65lxOsA;;|{)FJAw*lWJqYc_J!yF$~#bjf>j=qhDW}jbL3+K)YW~?^k?7Ak) zTz(RE=^`?Jz&#o5Sd$&n%_%_dLPVqLd_)yB38&DcUZSe{IkaWa}25mtCBgkc;Im- z4g=CO^4mG@!JN4mIR^{wKRF@fWY58gcJBsj+QVjG3*S<6b5=2RGOL9|u$(51lA{nB zOq1>7djwQ{o%iNjKO|WkJlbAnHQ$0ViOBZvh#%g{s3=4m%(AA;`?mZvrCg5~GqyYU z-lpiXX!nO)LlaWK-i)#g}(Dhr9bGdQa1FJDBh8 zF9DTE@?QdKwfS#3QM&s~9g5dA&kR<)^?ZvmFKjUuGVpcx3$>r9gVqrhVGT7DB8O>- z!M2qeF#v%~f-T=xm6c3{Lgiw@5YWOx<>)Wv#O)Rw%umM^txnjeuv`g zXv=;SLIOK#xAsSULhHzFJ-0_CH-~=oksQMEM}-SwXWmsi!1Qi;y1zKo)V5Zm2RT^f z>t~dsk1Bi|?`j8<=v>S>g?*L$`T&hemcB6`2%^gZ46MZcQw z0+#shGoPhPR|!*{4ZFRN0;uwZWnXVn889QfRCwr`vcrL<;Ib{j_3yKo0^6m~CAz6R zbztV7^~O1jO`WPGIGA6Q`GS*w29ZHoLJ1J8T^+pK*SYdx2p#u05?|U3BkVc#;^~+e ziqaXr7V43ImbCe)bBueaa%Ghb?tVm*fHXGz=e#To(T{V8bXHq5lDVAyB-}JHzzLVg z+ox@NcG^GBoA3VOAzOS3-f{%iWY@;U!>-J(mv;|xZ7RBq{2M>4BmAfq`n8f$2vWrT zslP1>FC}#QEQwG)sM|<7NBB4L_nJK?oqT!&ed_WK#L1p96bx7#q%~A@r5B;5RG>{NNnz_@LVT)jCN6%@H&7Tx@=cPGR0O>9R; z*ZX@v5^j51So0_39r^hXugf6YBxG0=V&YgVjiUX`KUP7kFx&?)hTpx{ZvzVHRE9K< z02`(SO}F%%dljWi<%rSz<5+BYCXFGdht z_TkdtvQ{==$tnp)nzoaD*DD3*wgGW7aiN-V=MPRqy+)8$Vp4VPURoTrko@!|zV*mz z)DETcU4taRX)9XmY#j|4>q=y@ZqK|uY-)0R7bkuJA6ejCju{DYkC}774s9H-X*L$B z7sv=T6>0dHf7c!87!1mVmcYO$^hlH{-h;L&90BdZ-6v>!6;~xY(!5CY^=F76E!vU` zX)on&6bK>HV}lT~AHIz~D%G@_n!~DnCjMMkqB$fZ)v@IJhVkWi)Vmn3Xv^p^_J^$r z>$Ar_t-;IDJve9m@02=l6p=fgV!O!!>HRVAb&Km`s^3rMXML1-e9j=W-^k7($=#=f zh<5>Hps`9uxoOQltBI{|n)4Hm7wbF1hyu>KwZy0o`F8IR%{w~6^-U`y4ioE2u67@; zBj%37cq-b>>;@({(ufj|yAkcZt>U`gJyLQ4Y2N~{`2Rx9@sqC^XVslyAH-}M7OY&4U26TfKXeN4XDA*_no{3CM`B9)UxJ}R>DEj5nDl| z5UMkO=uJn_9#?d}w2BXJ2G}hfu6QhmF1~vG%%^Z2jiR1l=odN|ccne#d669}H1{@t zGBwphZRm5gD@2lB65#KQwWZ2;Y(2M;nm&G>mbdYt_j-yN8!XBQ=}{iyt9N-(0R5c# zQBAW87@F!iHb+BhWN@?tVh4}OZW|4%_(8^_2rIt}{V|R%Pi&*S8ov;RqHH>T&bx6p-KCS0Nq&A2zxl@=0EHvd~k4!Uy)Vm0)A3YYY{A^Z|mJhnRx1U{t zSnxaw7UvpC<9rx$ymmzNsci(PWfBYndn?Z&H&sHSdb}dd71uvM_qbFWEq3I8lLTTc zG43G)XV9K_A1Qkv2E<4DAGv9)0J;2tCE@$D35MSCBGlIVJu)tr#qfJL4M85X!fwx4 zsCQR;X!2c+W^?)sVYe0~YOo5wfi>Emr+>K32naJb>u!iQ5v#(MXXmH(Tr?4p>yfET9{1yJQ=)EVNY6vhLy$n?752- zLdih(7RJ|@Jl17E6na^ zMtAI_Yba=q=sCSNsDo@?rJP>zwteSmd&zMO#I}vR&VMA}2#5$277aU>il<({3)w#- zdK15imU}iH`Y-+fIOA@Flw_1KE#&f77(}M2 zXdnrb0647%3!xQp?x#~rvlN&%r_ek<%;Z?b`v^1=7qvaUXE(2CZ>HeboutpWK-iSYWzMm%{HJHpw>}FDutmu z3jp({fw76v^}?tjTrB57!NcQ^w^VY6M$W%5ZVU3yAzGngS0_|*KIl_t5R1$_4!!)Q z;TJ1jVgu;&!Q0Oi^vp|F7pbC|9YFrwvmC?}l5u<5`~s?XpL;>o%if%fVxTQ*Xd!dm zZz>pO3jIrh(aWclcC^R<%2uAl763`hA19CUznB_~P#y4G@;EQ~S7pb8sD4IhWA*@5 zZybsieV10z_s=DC?Q!BJ?Xey-`~selnwbZkhrf<>wL5TD+=Eqb3ndK!hCM1WCsZUF zn81%OEwZ4YwENDp{@ak|fOV}=NFqlL^bt_A3Pr7KLo15yQ87Qk2SvEPW{YZpqhNY| z6L{(jrm_bzjU%W)SRNUBnCuz-=;^LV*LI8d=a13;UkXS^Yaga+zn#Fb(novGk!6j+ zU;e+c^H(aHE+61%!1w>!F6!j%gm^RMUe)ZotJ}^4tQ)!BKRFR|HFpXJI5?Lo$>F;& z5|pJV-O(-#OdP<90BsIL;a7m-p_&q+Fb|ju`eUy@3NgNgb%qt@lOi>K1B)NIiA8I( zpo$2XCvmLwzl+~0N58WzXJa;SP5UOL)mhGlaW4Zx6 z7!$9TjKC4VrUXQ?ct%)&RLgysYy$Rp5m3m9)F66y3roI+mz}@(g6e;gri!vGi1x4e z`1Sgl6`a&x^HqJnH|9SIRUQIq&-wL{86fMK)ok!@+VyQcr`04Rs%ko(hAh3S4y44; z?UMaXyh=4(e04~&Kdw9835tq;o+vx9PW@k**O1fbDp2;)-c@;3cNY4DuVAHXOTe@S zA2snfc6)43rT}xuq$j35JtSIjkyj|9%Bgq7EtMpd3{mK(1=r@OIQd#{9j<}f+d{(} zEFrWKfy?&NG*i%#CR2J-qY#zvDahUxdk^_J#XcwsMZCu%SzV1gQAx;$zUx=2aocRl zH>r1iK*nLJ-KorDH8_*ntW1sw0Zreew;HxV`4j1l&}}G>2?$9}$$h5qFo{N+RNmc` zE9@@o#QFTVsA4YoVF1stKhC0F7T2LlnpL-TfS>6;|{UU&L015Xhr*Ms1iY7k2 zfZh_5Dnv~$Dj15kCAY;d0+9mL&Q>KR{06ZAYq(39&2&2xE?$IHjfm{>AyzR9WlH0H z+D&bytvwyHgVEAM3^Ap4qt~!?tNP^xMYUX1zUA7ss2jd}CVN%zHS)lamuadVc#_c<+hqrS6nTnV-PP zey)!MAjKzv(jRE#7l5^G) zX#N`{PSY6uB;~`Z|HmyTIKRz=zr3PM+r6cvoMkxziK)i6X!N@_Xq!A`x5}o@Y!gb8 zS7YXMUnEVe0_adve##C;o$eSd%P_I>9~nf~$K0%bCU!&~3`yQ+Ob!6@>16w_DE44H zvvZ9p0<}RF6KM;z1SpfScp3^`xL|x^Ba!o#g}<<_LJTpdV&lw5fCz8#Z3p>||XrzityaOd#F zf#TEhj_?-rXOO$d4>93!^OeTtdgV<(n0KYt=YoLZ_4fW%wv5%hP?*M0MYdem;oK4~ z2cP#jf8^_hr$g1ie*{z#q% zTG}a;P>>hv+uEI!?7~cIeTt`n3R|&US{c1$D=Be)gV${5D6`oc%HDbc(N)Dmqr_3W zoPnMRNegfV_=`>pp%rkw8D;rH;rzIaxW4;)J^PxjGJ%a z0=;jyRehmb4X_Pu(GV^$=%pWx=~w0|n!t>+;4vazv_p$tqpHF0gHv-p4od9+}e93w_=4L($pG-_nLj=ijUFYb|Q+1L1k7 z>s88*%C&egA3$qcwkQ3dR=nHC#s~ic_GHH)@Vz zRDPSr+3PLEHJnox4-C+lao0-@p52d*^u!aVuAt3thiIJJzGbn?xW|^m3rW$k`MAbe zz9K@Ey#GOWOmkpc5d7|s)v)9ri+*|tCl+JGGI{L11VJ@t=-<-zW$x$*UVmjc3wc9c zb0hOU#EcIVsZ%DuI_btxl^Mc(_ylH2zD)5s^D1IE;AYgW*Qp<^ko5Cy(YJ>!CdT)0 z@V(_FNqG*T?NllM?&QkjmT(@s=k+$8Cfz7>mhU3;M5$WVwg)i$x=k*%aI;2;XaA(3 zo{QjQ6k_p+YT`REr&~fayn>gO6nqbd&hgGIG5R@PC|yG@ZFY*m=qG9Hz8eYOA^Iwr z2Ds29gCFU$aiubRs2wcUuCo3j5v(e2q+X?`mk1evTmJ|K~I< zVPjMGM|t`IuGy$2(BC@uJ#hY^EI*+&m*=~;p92!CXLT?&2p@&??r1ig4b*wxsdj>( z&ZUanBkbYn_FCRlH3bymqnzmN17q`S>zZ-1)Cy=~F*GSoNiO%2r#$PH^bV?c2K~8l zUi@tRjF&B-Bg< zXKL@l@gAL*E`atm4!vHGj#kAj=sWE95$`2StxykA76H?&N5GFb_n^^xE$$D#^CAHo zdjKM0)lTu0aBvf-Wkv>)QzCY(&6_F&{Zt*oyJ+_jyEsU2jbxwc7 zwfP+`@`67ox43fr%E%98xyaw08>TO^^zm@+vShhOTNC&c0o{oDClvZbx8_#rbEY81 z1fLVONCQ~QTb@WhHNNwgDB{v>y)~@R|b+(SVx;XWSzbkV}s0e_bbpjbyAa zsVQ2@ZcyQ_sKOnCyy;YUJBov?k0!xEXNB>o{7V)oQSHM3?f;`Xk?ql0VvzSN9MYqd z=BlV2+<>c!B2AkDOluB8mj~)m7w`8WEndC@=J8^n*nUyeEq315Fm1Th5YNB!- z&bRsn6kw4O*M8$FM}JIkCkM(Kg{ebZPw&^{>_UX+^U_eBPk7Mw$YPe zDNx+cThqHI#@zGziR7;7^m%FH(*TWUZ7qg>b#I^{>kX$(JYpWmbq3DW)8~}#ppjKB z4uJ|i_qB-F_Q1e344F;JG^UGU%7)4+$J`QZ}KJit^S^=I$8dZA2# zJ^f`mji;VI`Kymw?eD$bIyl#~{raT0WV6ooWc#94$4i6_h6?L?5{{p=`|*1IPU*Le zq0b(r7+PxHNY)3UP>O41j2m?nE03|s`WZ+gDrrgynn+tKBv&lyYXjZVL0Yo_yP5zH z)0OAixImA6^0Jo*i|izDf-Yf7wajprx3~#i5+486JsMjKC3N&w&U;geZvV%PJ5BGI zaMB$xFLKgYRAG-+$tU;y5z)&4U8|XAYQ(Ch(s|3JV=u@!zo_$mCSjP5NM)WrWzpv3Pt5sZk}uE8&#S&XkG_oS{146uvh~{U&KT}B;$#7MJbKBEf6*n7H~a~BwOc{mMA!BlgR3woSJXxP54VH% z*cj%_7@q1^VM_))<`oyvcrebK@gG%{_!JBN$EpdTmlk$Vw1?7OPCFs6;gd@lt=aU555p$cme*415 z!y3%k2du6?y_KWh7;M!TtV#5TvP|=;g#v|%hjj|SAp|Yt-Zz(>7Ix2iIfR;jK7z!D zLPh*6wr?+mypP5`6@ER_jB|GiCy-{IYVVVh9ccVV(c+q`{7ntR9}14vC#EsAj5dH0 z*FhKLTYAV49`FUYIWI>NKRK%*T*pqW_3hJSu%UE#9z}OC^v3RKgA_0$Tz!p#zaP+- z7pWn9k1sLbWZ2b`POZW>*?b6)^gwpP zc4VtjJR=S=Oed~=5SLdvqb`T*VguW5nt#&h6ceVB%wGMtWkknX@w{u#QSi~g!hp!8 z-xV&o{L#n9RJPRX3wMW??k50hRNG@hHdV7Yf}&S?GN{K6+Ub^Z7+jrmi!)aFPhgu+ z@yimMSH)7O6dyx9L7aKzOLzn9)GRma$Pl;}(@_AsMlRu>k}6(lj9zh}yE|y@IfFpH zF~u-}%ihx7j=Q+`$)RGQIubt5q~L+)s7tSH9og>Un{)YsEc*>u{7t@?LJyNQKF4eZ zSC3`w-Mz{y2`dHC+Kh8ZS*5zj?lFzX>IV#au4Oun_F$YRKdNut*I9D*lBFdg=-eFK z=j=r{gW!)1W}bI6KWjsN^VN|MA&<4X9KUGAR)3GaA4>n00fUPqwgCEp%|(-84v%32 zFIoS2KN#4eLil$S;gCVEyWze%w-^im(0di(lx<)dsFMh|k6|f>5?IPiLZ0jJf){lm z@`V+Y%1q&DZ5Q-e-jU*c2ye9go$)Pa2*Rp3Ch7l}AE4pZY6`8znKzV*IoHo{MD4EY zttGFoU8O==LKZ4yp$b(<$lW@5L9NV){b8xVItTCN_9*!cQAmm}K-{w(8?DgR`51^y zd=dj)lz(3F^cIF<4U6fjU1=7NV~LcV3n`PTlX#6rYuuviIN?kh=_XDx66KpDLiVt- zEr*V{iZy@Foy*8J#9TGPiME_afmv~lF{QUb_^p#o<-8NPBgZ#)p=V>&|J3VEJhH45 z!{lbemQ>=qN@EW|`;&}r`DwFp+2ja%JN5Fz6;uG#FL{$7A7SzF&N!V#`@IfnZd}Vi zhfyEo$!kUyJ8Hq_aztnFt1yf!gn# zrUoNl%GFb>8Kic1T>Sz|j?d%X_CBn#lvb|gA|KD0Y58GFR^ak>#(lFjp8AFr17fP1 zP>Tt9Qd5CIL+7G0AHIA@yFjPDFhL4@N1p>yvmdIVU`;z7XH^u9ls+FDF-XndSM5WM z;xUDbKQrOzSuXh2`nPti!@ToFAN$rmMDe;=O!rG@SGOiAm3UpkCA#YB*GFRYHW>J72nZYmKYB=zIli@kqV~b%IQps zt62VH|Grx+QdSMX(WWacUI-~T`UPmc{tOnWEz)gBcX<%fr3GhMCG<2oCEcpARRNqFz3$a`+Wa} zdw;s{JZG3Q=WO=uv)9`1diRP_O0(CXh3mfOmQa{Gy&K0XIcB2aR{F>=>Skdt`~{{* zhl*&VV^sM&KTzfRoO6N|J{lqCAa!###w8fHk`QY|c$Gz1&#Ioj6kn~!086a-J?otdDaB zB1$RV-)4p)TGwK+DGGkbb*mdSP(|4Se-SgAOA~Wb3Vv9=nox zr|_70r*x4-yZq8dsSa1mlnYncd&>IIa^rnKxzj3Z2{d7n&|n)H@^Q9t2JC%tN_sv< z!D>MMbRSj_j7$57H?o99$_ticVPBq^rpp@QD{htq*&Mjx6I_pdinQD!WEEs^ zNp|wfoIec|HJM35IE^@LPX~y9u>a++bJ2 z;S^Ki&bG3taAL>qO_)PIdCDMg-T>S_3ybfJom+j&nVSf326Oe_ATDmKi66+lu$@aP z%zBRnfojCLhp_3&)AG1w)cYWc`ULM|W=tICDyZrnd;C$)v8>|qlb-L1t-?FA0iac9 zX1haWF(;i#a3brinfqq<0kDtDcx%Q3V!CJGtVd0lw;tQUE2Z_aeS6Qd^}XIilVJqX zPX9T?uH+Z&S93E-vG?|AJrxIsehSbpy`{Hg@dg?B0}I**P_8IwSX*KV2hf_1-Vs^H z04;zmA7NhkLfz~775!Ux(z!oLPXG?<+ggrNx1-X;cOJLr@7PuMaev{8A`#ypHdXuf zaBr-D^vPq6lnS6+hl@{K=QkL74tlfLO_ocG?Xi=c13vM+mX5|0I1cXx14+1)K;(xt zxWeWSw6)LB4muW`fdE}BCbjk5_U-Yl+ARe1PVsgm<=bn>vdX3$Cep1<31t`B-;!8( z4H8)?-4zgfxK}{V{WU`76u1z=zR+B)FcDV%^B<0&W7HyB`PjscEEY?B;Rs9sdSRBO zH@89@_-v`o<4%}yK4P*(h@Ja%dq7;*-|`>CDq2TR;gK&w7#-C`PP=H4u0FX97|Mc4 z1kdfU(~^;G<-9Fkhr_V>ffgsbcw>u7cx&P=iQBZF6z8wC`sE8zeyi_d5*;|v4JU^? z)+ruCtK08CZ2$4`BQM_s2Ev;3!0)s6OYBTs6dK03zM8ZQoHGbXq;Y=!klibNYeo${ zAA6VLqPvGu$+XQ<^`mAo@^f4dY{G1?T6X7V*)|5rH3&n_1QdR1EG=W`N}i{U=;``rs|8wg9#R0pb zGfj~zJNWg9^xZ16F(SCcC)c3AUR1Xg(t5h8T&+(wsq9 z^=LK7@mPP0pDH}Qq0ql`ENf=EG!Wv-{3iGg2y*LNNc>)Np};jU?qY0w&z%1((d=7I zL*9|+%9a6{Evfc&)vv(MN?4S}kTc6!;!+TSJsN@by}p*iGMu=T2V7$|yy)=Y-sivF z9^JR{wS6fOWd!i=+>|LfZr~;YF6RoE70#JiGhHYE6nw_Gb6O%YDT{B=pj%_*O{~8E zUzav}9&0As@AltQ)G+!piic3W6z0i^BLZ70|FW3jo6@K`z%zGdx`xS|+t8p~{oS$7 zCe%h``_r9!*8g@o9*@@Zesibq6nOsh!)~-oeZ1%By>dD)w>Lida$db3?=Qt51G)^` z6)(XMIERV}xr8JE17~xY)TF>j#9qbzn@DjNs)KAsJBonB8{#A0pA)BgdepfprVC^W zBVV1R{#Y>x$txZF`1m+Sl7f?hxf{|2`+*CKSKr_N1ejc9Rp1zmlftM>gkX-s@blz2>pMt6E!yc_E&Z_-t2;I|}OWdh$S=AbOr~@pj z%?ZUpU(X0r0{5qX#yv9<%7QwnrVmJ8f*FnbAkwI<`llBd#vcVmx4*WPGxd@T`>Q(UgBY5OJS5&LryRK7Y1=*ddeKZ7I*wPZe|dKkai{R znKvh{7ETOqvYaKsf-5*KvFVE!3VmP`zJ+--!fT^zIuz@DS+S_zTbo_A4`ng2Fy9>S ze0KH*Z|???ZHSVEz>l@VD&(`%-)98j(;5EyqjusKrC3J+525oP-PG4$yaBO45Izw* z@2g-H{&EW~2ej!M@@@LW2yryS(V7+sQ!PJUwXL1AU|Tt=m$H68pEoHE{(fC=S^Ip_Z^t~}m6zeAs9N^7;w|_Ie%y1uYzU9R(HRu-Sxb6f0jH1RlzrVug`R{S%_oz7^h?a57OhH|Nk!Wd z$=xRAQ(1!}a`%az-c?m(>IG0yyR+*btvg?JcQeo5{mnNba}<(f8RJb2x2Bn$?0fa6 z!yq|W9?6r$2IrT@R^M}hD*UfTE6rw@ZAuG!u7qLvTw z`jj;)6v&%KWdO3?DgVh$8IEUH-Lo*`ilC4SsN1TvX8^@BJ8R%Mo@F}5%{!}O_f<<{#UQT#(D&?539I;DG`>;tg7Z6!0DYuq zVAgVd+WKRt{R8aW(7^szCFg;ld?(9Om)(>|O?=>j4Km#qp5FLjcdp9%D6#JRBGyFV z*N4l$*9L}NfAcQzpz;vWxj#J4U4BBAec^$}UGdbm zLtJ#2;qIy2extFmU0aoJZq+^Q;BD9aewIvXL4NuYzZ8aO4%JaWcA3yJ={{Cl zn_umw^Kd*{ChJVOdj3AtCW+!g=9g^e!>v`SyB|@O?yhk%q;CZN>e3njT_yW-IUpe% zKbg|nqUvWYpg$!Na`DHO8Aqop?CD(=XsBui-*sg?F)U3^s~lXo6b8~tK3xFj zUtI(p&q1G50Dm*ZIuZIn-ZQME$)8qNAnyb#WL;BRexw`rhnAGN_mrZMxHC}<@m*YA@0{{Q; ze|HSjkAKvo^j3f5^zvMSUEAu2BzZ7Xm$-mYB@MC%QGY{2|3JHD_BR`XZ z#~1OhJ@3#xZ>{Q3Sf(dyx+m0c5#cj8op88m-6b?Rdck-++EJPpK_9mG$-+r(Zp1Lr z|Flow?hBB}PpqB-SItF%#>%x);c3@wIGy8I_U8^%N#MGAb{jA0aOJf9c}m~E=M>}Z ztjbs-*Q7Ahls70?K^T;}9R7LM=&!|w&ng5hX98lgzIuqS=zDqwmcK!43s*K& zE9%{J3epY}4Ln^he1zZ(6bdRCUyCCOvP%klOpzBur2jmr$NrVOhNv>fOzbxry$?G9 z`9pVVhqAcu5BrlD>pWu===u2LUKs`QaV?cB!^{Tj$vETt9Jwd9va81+D}D=HGT2Qo z>=G~)CGjLcy-=wZd1KaPOax6JGIn*|!~2iBZubf$SMNB?O(-9~wpEi^?&!mgyAc+bdoI#ReGA>7U@@P>ics_TL z5M&^~8~@t%(|&1|FyzD^u%dG)Ml@$oFU7}RdInr>lp~uJn1t7r?RqGP_fztP^9cUx zR`(#dlnfI_f3v~wG*)A3BzFcL8S=f^VG8x7IOESpgc_{^+@fiGt31L&FHq z=^bLc&`OTMr0=c~fz7Xt3ez>KF68_^SzLBmDSO?R-4pt^G16EjY%&R6A+uha&LdzO z@zt3E3w?OE*)8z2FW!@?1MFp^YWE+-8}IbCcJm!QT@2}Q4zKx#@q|24D*q#_3sY$1 z{?a*wwN{ctd+oY3DECNpA0|<{`_z%7wT_L~#LpgJjV|OyH0$63bms9|EBnBYe$;|# zJHT7Nc0X5?K%I36XPV93DXLjbf%G6Z{@-P`Bq~d0J(3%uH#It-%rqeW@jcazlAiR_ zmwbF@ZnO(OB(5d!^L84qg4E~RmrH@Coort~2>yrFw#IsEl7e5}t*d_c;BHnB9wh^D z4P&&Yk*40haXvczYWOp)30CqX1ZudvI;hLL3VQd3p;eF01d|Y6-5xI_eE*W!#hWa2 zbkjLBF4+qm@U$3b|=FR0ZpvlR|e{G!tIz{Gp6G3~Is5xJ6) zl?tcO(xxSufzDQZz|S~TGS}!P&ZAqS6Q_lo&MVs;%z~28wq&(dqqd+gsd~4R#bS_9 z<9yVR;}usw@T`qSa_JaKu4butmN<%o*>(4Zk2cda&Rhlci9WP9m$Jq{V^E4AEstzL z#Mg+;vzy>5;b)w{1k+BBO8V9Zj8}FZ{j4M7WQaoPp-WsF#V)smF72xC7lA>>|1k-} zFrgFv1)rhqav?zyK!GySwk55}NN7qDGC@Vv#j_@M<^s?yv!W7ApX2AP*c`$>$Gw#m zYW8H2A2(f~&t6{H0yUj~Bb?Khacb_V&GY=_M=HQfL4<9`nfe)dgmxVf4L52t5n3i+*ZfFn7V!`_ct-@1F> zn(?2FsA$Y1d+{rZfU`Z3!lL%SBAo^$BuXL0?r8Vg5TU`>Ys ztAih!oj7Z=yh&Z;dc46G75G;bGzaa_SZD`45nn6XtoDCQ$)}t$oyUpw^-pWGDC+)G2AW1?X?FE@JOCC)JO*QBA{VNN`y04%6Gvl#Mwu2 z?h_CN8%cQXH1Iy9(|m3U+&BYlro<)jAHZ1RNRP7>ZOtgK)baip#fL zf6+TvXpm>8KutN+1rE4z3QeWro`lhDqPMVYe$5%+Q3jNfEsta3EjEm9mQ}i=x?lQM9XPsT zec(HOGPOH{=CFF`jp?;vi%lh??3NKCPeLr_2^W|3O>H;LI&cET?jQivF;ENyz3zXY z)h2ga1}4xG(p%Jk7Z&2YUkS@!V<1XBh}fC13Z1@rUbva=CakOSF{uRj@nvtj_7#D{ z$fXgqoLADD z!JjoL?-=7CzvPw4PHerJObfrA_CMdI;#Bz#rWRxA<|b83=6}rJ;KONC>CaDI2qZWg zxLZ3?vJ?+b*Uxy7`3-M4nRr_!md*HUSxpWF7BI)+PjX#e<(=H~+}_FQ*--THZVL+d z=B?5Wa~sAkNXWeG+NgV39eBPG>*)I!U^dD%jXv5_-7J$un6u>e^khyhOR+1Q^ye&? zE+vk=RlJj|Ud`09{G!S6J{XVhhDPLQITN$rF{hME({x~Bf%#{ddQa{4L~w6vgXw@a z>Myy$Qi$a`7BlpFrnh^sLJCdfO-hnA2+~)-{`JrweZ`jVK!K^;Hk=?4H(n5=17e3? z_0ma5`j0?eBpqW~Jg$O1Y(fxHTBA)w2>vSZrLLSo$^#+5TJ876{5n&X(~;Zt8LCRL zvNHN`2$JQr9}rY2kQw{*<@hPrkf5w%HLfsQdGcW8Ai}2Ryo5%s4*L4)0dtk|7hSs= z|JCvBbadT}_FnFFm8sv<>6c$q-~Ym;VMb%4DHb-dM$X<2_OX1#kZ5!?k<}5y_0F+Z z?l8O9SsU#?pIe7dD@4T=Piwa78*gF8hR3SHGt1<02;UE1crsTggKkeS&3}T|P0bc% z>MN$=pXFOEmcuhW(kPUqm72;px*PxW0KcGhp~d09bn_Os&@X*YET~7mGro=S3flAX zC=1SSmI>GkdfE9mE6qD-%k7g(c>DUC_Up}&g2ypi++{piiyW3VK@S&S85sdw`p)M1 zkqH&o=5-T>* z)WQ^R3O<@mIV;`75y6@3xzw}5KT$^FZ34Hu#sdR?(#}jJJYKdm+_VTRd5W!0)fOKQ z^sGEf3XEns4m4esbkfYpZ4&~}9hpk&YucTue~oTry}21VkLo{+Br51_jVVc&|4RJ2 z@^Gn39JGSa47gkj2k00o`W%x24`9%dl%?SHe+#-y8}Fa|jxTYJC-xnn{I#K?jOZtD z7<#f~6@yeDaNSEk99?^%&2wrTclVsvus!H@bb#FfkIf0~q zxqd9m=9};4;Zn#%0D=pR_ z`E0tzW8m|3Xq<;tS+Q(cl@Epc0Oe~RSUD_UyzSHV@uwiYR8Pzap+`$y^?A<^tkt2L zbMI<*n4^1?%Mnv*t?{fL=(JFS5|bfOxS|lw=o4cU;AdTHg6R!7ez$M zV$=m>U(Qt~5a9AsXf`hRGzrScme{6}Y>|3C@QV6t?^hFdUw1lTsL9)D)1~R-Q){D70&GFb19 z#g9g)4fq59+$8zkF*6J5%Hf!Zv74}R>0tvo^L{xhESPsMdtLGqH71Ft6gl^9@oaOP z&@m=g%k7}nQJRSTe$UDxSTIqDC{K2smdLu&=^ZFnFT`YW;8?f0Ih3f|T~eyoTj#5Z z>z2FQ(yc7&a)a;S1z{lO1iK`y!DpF$D$yp+z!BnteL4O-9p|Uubf&j+D)gCDQGNzeIRv9zaZmfeKkUf8pC5J(P@hmms;v z)4JxZ;5l-TgrVj4yv#E(zXz>(0?@f^Wym*|y&z)kO4Jn=0tfXeY^SZ;YMFvW|J|Ad zM4t8tWF?KH9NI?1wCkoinlw$9Qo&uw=7~%$Ea?C`wdrrTxWH8$=C|DL+4hkH#qi&^ zZ}Lp?p3)8v5nVpOW89H|9JwEdoJcO+ zq~&QozjAkqurOya*R=8UnyaFH8maG-8pE?W4`1C27w5KzJo(Tu@{r`(Z0?nc<3jfP zIYggQ(5o`%e4~- za`z=YZPI=&f-936$DnT{we9| zC1T|xCX#NG?aUAXr2Q8HqO4)t^xK+>v6s0o=w|9=#R@)&2E8)!UqSqId}Bd*`u$(| zAoJwS6`Kr!`8qDa*7hLpxjfS0)S_I0P!@5z^F;cTKq|&y+Ghka%0^5$51#Ord%xQv zF-=oz+ws$!Oo`+iB9tJL0u8F*jYv2#LY(}5!6Vayj?ObFC&AjQ=xUKM_hUAuu1ew` zH;)vQt7JfZ2o1K%+oDm$5#j`(rMeK5Z7?2#1pFm}*T<5Hx^c?0C~y#Lt+w<8TV~S% zZgVk8qV7O#6*%^0KsVnaapuJ^`z1Zn_GhDfvK|5O<%q1|E}$s9fQ&VYKe|DpqnYud z2Y*fN(lt^-1oCCV!#!#759hJl$-z>>z@chD&{CF8_Pq{p@#FqhCQK~^|KlT!`+z=~ z+SU5yjeVHs)%w5ezBnQph5+=t%hM*S)*mOJa)^hcNRK942RMDW4qi~;#nScnKXTPZIGaI3^tyi zU76Y78PMHw#O%TfccH`wW_T^v42VS36Y{YG@bu{20Tx#q1)pN4^Gdny?WCg7V?FUQu;Z$V z(Z4sgGncRDn9B;n^haPVzsBKy2ajDIf69P<`uflNMK-k7GEey%;qAwkW?V2R&+TXe zN(EwC-}m0%FPYm2ZPF>Vve#MEk-!WQv1apn2E_Gib4xJXjcQxAIT-(-6ey!K#m#X( z(s0)`A*M}vfkSkFtpBu*xXL%ZfasjktV_FetBS3k*tKfaq;|l-tK{vmk>Z<#rxv?j zDn+yotz1G3(yI{}P^lc#Zi#L=)svCwyF51Mqk~{0`EyRMUUASF*}kVc$n4>(KeFD5 zL%b3vXk+1vvp+^%XXI+EF@%=ner`8?gP#Pg$U4Y{0W!^?9tj#%ma!b~wI>2jTj3wjuapIQ1o<;9ELxr(aP?n(9 z+@7U!Z1IfScEb*4t*9SM^SYOvZ0pyNtZr>U?%e2ux7|zc=Jn$vLF#|e4RjK(G z{IYHp=Gfxg|K6A^@bY)|C!2X@hkne%#n@eM;QjoC z(Jxe61esIubq@OpXD+m~P&|!7pNU(UALC)oU$4^RV>7qIh&HFzj4w97 zG+@Hw?mw$9;-!sca{4x&Ym+wm_C>B%)AIGO^S1%YD1D(Dk_oBXnv(4oxJ7;y26n(C z5Ka07M&g=1$8+eJ(aj!X=LPMjGQuE5GwdKKB;Axfb5UKXp*w(EP z2N8;Zs+st{{yf{C$Sc+kaP$bu+ijOPFdey$?NK9rQKE*WMnkyogib=8o`@h)lmWKT z)KO>=?d02Mxv~02q&yqS?RbGO72kgx={f9Z^1Fm+2q}g9Wu}C~NO_yDb=OynY%j}F z>;Agb1dFk|y=8GNpV3DuhoS-_l^x<+Tj?m(eQ;Xq3^igdPxT$V@6s6s*_A+~l>1$0 z$mjfw2NPgRU+vH7-rABQW|eorfvy{jyWqQDGhk%8KObG-yw`&lR$eA7%}yn-3x_Hf z|N50Nxv3v_&{5nO=V0+Nl*WVNWF*&I9x%Tb9sLlKx)N^&&I9oIuTbg$v6Q>YE?9%w zpIyDd7P~epN>?^Emr(o#7IDz`7qcohm35T-)FumwT))$S3x0SmsWph|jn9Mzx$nR~ zV<&=$?xTm*@_=V@h;2u~5Gem?dOA9ha?HF$CjI%M-o@lUQ!aP27*s*o@}n2O^4FRv zW(2Q!3|%#o1*$kb5Y3gv1$>zg^MFmVRRls^rHB;E@Y+MF65GYfbpcb$6}0+8BeL`h z8=PXg*bwLBSwNzqd)nm^Fb)m*4%lHVd!ghHk;TSzr!RE3VNaE$)7d=@4R0e4H^Q)@ zCqKx0d}9z1fz^UPhW6Lz)8v7@*SeH5_px*Hy=m()PbRlVzgU<&Y!+W*eK2+FCSP4X z*7i};?LX;m@#zO^SLSiZ0h#&5TUrv|0nTeK!zhT_t*zy|7>zC1-6h{$pA*PqDyan2 z&hBeh=@>rvXr}e$%q8+?8-gC)QA_zK6MCUbbxwde7yh!#C96k(PGpt?@|=F;0q;-U z)d~Q5wOuegRsDhH^U=ETQ0DoWTvOG6Y^lkdttHewSK$+5Q^XsuLz;_7dFUAQ9eqiM z%`UjZVsiwYm>hfoR zN-8uYeQ%vn6JV1Xn7MyjyzrGAQ>YVHHVXBM=Tq4iB>VS};5JehKX36-m(zgjuP%vA zRR>i3^_#)U+0@Wl^ygc`o7Eg7?c%9J(lZe?*FPJIi^3+$j+!A29U2 zIf&dzJAE;TbFDhs?T>Bx1D47BzVfoE6O-DBe~Ui-Dw7%z*yVBYZUUMCUF>^u0SnMW zqmm7F!2;Km>-3zLuyg_+e|>LRWF%;Gn7V0yrCfSpA#1R9e}yu`%=agrJsLvcEfNAP zyMSwM6Q)4rk;IkG)qMPKbHmsNBCdG0_0!QK^-G8ydkh(RN`ETM=JvS0$HZqMPz4?WBZl5{o&7YDsifq zY;zw`13HH=3rm{pVbnXoP0@^@YMj8c)Jf{&)>Rlgo<9}&o&GPsIxq#6DZy8Vj6eOT z&J*R8DYs!QDw$9(cdBOSr}pEM3Xq~~s7LZ$NBImWjfmw~Ga0s2b>DMWdljaZMiZUW zgDFdm5d}`nNE{&RwFmwY3r$pRPO?o;KAODW@^&S&%GI@a|CQ3!tp6JXh15>QoHNbg zsfl6PI2ufaL!b_CK8>Zyyz?9}OACRP;B$VVz`AGPyuh|t@o#PCLs^-_{lyk#E6?Y+ zTFcJGN1;-)DTa!eT4Ys?g+cnh~@6M-5la9q8&;-Gnp6Vlz&8Mp*OQ3zh%r@>l=$^ z-}+yXrJ~>;^~1xk51Gy_^#Tm2E?^?GG!7Z4EUi*Rx2iQrmt*?i zSH+fiu6+7;y*3TbTdZ1cWmyIh;+>s%S~ce;BH$@Gup5ruQ1VxKvKekKGT%|Vy#MBL z*dL~nnN%Wqz^%S0gdGsCJQ1Q`_bA294*yh=S8@JR+ImSq44+Sg9pU;!)7E9$lu=h+ zG4;m{ZH8LL9*m`<`mS==VR#x~ak!>gHVK`sA5SK@fIYu?*1z!`n6&*9Ot^GlVERxd zYV4t8=JM4;bYia58nHd5i^(0gznpu)@sD3qVQ-7MZMeKxR2%FD`s*J+g}aEQumBi@ zSFC2`J`I5)vtD(8WArGPgdA=~t*iN$Mxy6W&P=Js?xWx{kK-}j#jS+tA`>2Nwu&}c zTb1C(r*y{&Y>l6xW&8c*&lwJap@O1v@DfD+*mlxT`00;RR~*VCx4=<}v~Ny{dvvj5 z&$Sf>7>thA!yk#No2fz7kmSnMR*$7% zy1oeyelHxf9@I&lLLJ0>CDu5v84#HUjK6)bJ&AQo5Ogp_;mEs+!uB8^ zWn$$+b1XwjN6MbwjCo_2^)tK^OaH zDqr|tJxpFG^X_$xX+aQ#Tjsd7``h`Nde|Pcm^2W>5&WNd6{7C~FBVd&wWLsR((=;= zdKf+_b}cr^e+A{A8@#=}aI*(z$w)nai{*Qfi!)J`51(W}v;Td2#wr55596TbM(y0y zHc=sQCAqaNCt@G~JP_5>Ac%WOAC#--VZ2KB1C$Wcl;ospB95sm#0gj?uHm&WGljO($6NCq2FA4BXX@yj}CW~lA4)RPAtb4~l*`Joza9|b8~k};`? z_t#rNNZZO>it5; zIfhm$MFZ>G;#h4Tg{Ln3U!f5PG>sl-W(7ROtKuwdXJ++9B8?xihWN+Ot956s{xCe3 z%z0t0O++8*H?MeT6gzd-Z}5(LYVzH&mk%(t9j{`UjxxsD=B8a%pIOE-#AUwx4MRNPXDG!dvIpoy>){G_ zw{w?x?-Az3>vs7mDmxZ4f)FaV^dX;`C0LE;bsor~m0A zYwZPqIaJ0KbJ$w4Mu3WNNiNrzCD%#L%-r3R_l1HWhtx>u#U{Sw>odyZ<T4d`2iN3w>)_^hplr(^ZjP-Kr zHbcpimImh!ZHa`+`=&BBx*Er5tRAVyjoQAem#X)lNx~P_TFqSoEpy521nZ!9xQ9y$ zoI;>3@0hK>H^(Le1O5Ee{%EhE6A)Cr9LPAHrbdq-4RH+HEvz5&pT_odB>_h0bN0?8 z*8o%~mBNV2B9T4_e0iM>Z4=?U-B5>&b0bU2- zj;2)1!a@X{k9=1$2k|^Pkvr&{4)`nSnPi}*@H+v+4gxEBt8xNKpZN!6{ zwhvERE0nC72n*X0zD=i9sQiq1Xt%&k~xEAJjNZ~*z}Vp3pLKxK zSGx@9<&Q5w`K1A4qsD>K*%@yMgYzc#l2sxj7)CZ+DS#$gjEXsbqtMD5vX!W?PfQasYOf_1RP$}^2-{GwZ z5^A9of|DoZbbc+oMGFETF*m>=nfJ_WAyPK`Ob<(ng}G52ZIav;%U{pKE?`CMLbCyp zqtF<_*t>za4fUd8InO6b;ef5a{$blPF7i1X75M}g%O$%CawTL#LbEEj>^N~>3j~i@ zY8i!c>pqzvctr?|a1)9^%?D>CH7R5j6+>hblIk2(tcGB1>$hnR8lbi+qpaAxm<3 zyNZc<>Kx^5v4t{5fw1-jsBrEu`-7t+-v-|AWhn(VH)lC8U|d2XyMB)L5qvBIHXfsI zW~D`pf^!Y3GN7RWey5*(9XkPcGQ9SGy_kuCgt*7q@P7|5qBOM992c-+KOMY@ZA2Kh z00re|eo{k+0b4J{ty}?$S8m|v1wQ5fqSSQ&)Fu#&t^LVZOOE9Ekbu{8CB%OLQ$BjR z-cv7djj{PgkcH_F$A&=Z<&asyJ>kHPNauqKSbnShiX*Egz~HJ`Kma_>(e6%W$Bl3M z@vFo-EQsajR(o_$#Fs$z$xt|h&XMH#h_vDlBO;zh_l=FBd2)l}MxiV@wMAWy3VhtM zE$v;&#yAZr+Jn_YLl1w4kGOOJ9Idb7@RiqqQ#-1=(Npz*#9;l-6eohegYyfq=Wpq2 zsm~L4gfz;h*w+AaY&GpRekRbIsF{I-I4*0N^k-%h?g~>;^4J6Ra@JtDldT@2ye>H(kLz%eUi^*qKLoGQWc4PezV}Ft{~^8uRdH`8^?1 zrshuIeDD?S;T^sX^3-2kZ@dK?V*lcK#9{P5sx7zBmn<-!B*uz2J--WJ)?W$R{s#Li z)x16%d0ukC@f%zmU34nG3gZ(0$!125t9g+B1TWs()oV>HRg<|AaRn@X0y=@$>J1N= zGoUUHn|^G8Sg5ct+DB2t40fJGNa*yBZG@k+4%r24#$b|`=mW7DY)FeKqM*CXtX;5Q%2cm74Iu;EM3Qq04_jhFX;Lw95l zP`e3RcJ6tx3frK5NrALA*XymrJk;NgUWK7riT%AiZ+`2KygA|KcR7C2`#{DE{iGW+ zaXxp`5$Kr$Ed+L&2BFMZhMmKiI0ivd97ZRGgkI@ttDaF}zy`kSb-3JD%*Sn<0KWUj zG~65{pm~7Z>7dYXIBj#;C&GDIs*22NpFp5Q%T@PpZC}6370SV>F)`5PlNXXWhryScevOi>$YQ)++`O|Nu2$Jb2%fSioIvc3hH`pdt->@8 zd|6LM60iw0lvoDKoy-VmTp!fgxxeMQSigb%HO41CU3d*(*lu#gob2jeo_k$18f6)k9C2OPPRMaoqpeOr8TMQ&4)1p7)WD`t26j6l zRYQ$&Ki^rIKtl(q5%&RzFVXk~jKRgK)ySXl9MWj`xSyma^}@!O<` zGJ6Jal*XrYZJdqKZiCmnSMu!y;FN9Qt>fMU7&cVbPQW94J8F!86FKYoXWI~{jt_5J ztG~P}PQXY^mPGd@;kK(V0>O5Yz-^F``GQ(6Q1Qm&${PE0t0+n3I}Q$%Ru-iB_tJ^@ zW@nffMq6UtHuZFO!QnRoJ3L~-h*9W%?8Wddjz8Cn)4Ywnk0o~FN$rE?7{1R`NVov3 z){<%ie@_fX>XY-2#biKV_Tk?azk0N6`5O>D4a5(`)gRp?(zSkp0T+`OZXw=@a^Uu_ zT7Q(U!n#>vMxixX0Y0wkRl%dsgmat(lk)jYn3qLi^hsQ$z61Dg)<^NMPuKqD;oyV0 zf2T*M?&?DtwJrggMknnxDAnBU`%v>@94&Vn^u>Y(BP#OklBQh=r+WWF`aC`wqDq-D zwoJPTCjh=+N%lUF`FmEZQY(<_*~65ZLNQZ#?3`?lC8QjT6Y{WYdSQ)lth==+xcekV z)F1bVwjlkuZWJVsLW52!I`Se>JH1(=XPb^0uCfT;Fzkb&;ZHrUfLv_6o zGa^?NlknK;c1CqOt?2~3^RwvvCc-$7xeqs~Zo^$@nhJ+sn|uo;E8LHkwEjw@pBTUg z{5XYh9Ef2-8);LWK<#J_`x;bMDwt!qo;g%Rrmy>ef2DkQypx`JD$ARiAQL*hA}cnQ z#lFuD=*0nN$%Eizyh94L|MvuHy z5jhG0iYs}Pj#L8#FF+dR^!}Hthh^`IJ9+sGvE}L|38o}KViAzUfO*>*48ni~GdV(M zc8X7%20T`w573ay=sAbvcR;=TYc3>o*z06Q8(a$TKTC?AggS^U1mpn~~u zNfOKiN=)M{5Y}E!lcl}Ye z5nxK33TzJbE?tFPkF}v9@PRMTKp0U9=HXvFHSAn39*(#g>(R@l!LdSzuZ4@3i`QVs za3BPbM3K2}v`Lc!=OWOcL@{6l^74)9^A7NY@8exC-@S~eW_h6LuOcS~EIJ^zp;>kT z7(XdWDghW%+)YS@6G{M)TNGcS=%m^S#H*Jkh7JoW4Zp__9fo}yv?@cPp##`=e0z(8 zfsF;v`+-(H`a)BlR{#n}&|nW{J-G8*ar0kw4yH4BCI>cW7` zzNN~>&giO7vy%UDz7m?83wQ-5Hx92Q7>Ed=r_c1DOFSm)j#S`zs9pLTWIE@0xek;~ ze?$m^=Xpy87DYkj&Kg)?0a_F1Q=|fCf)_@K-^kEIc+;VBc`oJhud`?m5w-*Qzxg7i zYY67`w8kmxR>V|{-1VyJZh%7krBeGJh~Cv4Frfn1ci^hG*5|D=|4jB~(d=qMVLuT+ znE&@P4%t^yOM@S=PJ)`}0*qRr-NY_k##o$3%h{ZpP$qV$;eSDrQceIfh!!#wx_F!d zb?h6{|uiOO<~VN9WPG9j>cR?q=Jo%?&>pP zHlO@cc3AP72|Sp}r58aKeG4X^2LUr>cjcF$lIA-(R<>1ypA0E4mR5D%@iccB8x+{M zGRDa%)3IyxvEk&6!9&cs?-*1vY;F3?IX85n#s`-0A(+t+k2j&AF@+28eTcZN&^4D(Gt?{3M zOCx7UTxIFd`eQ}{Ai^eXfe$%P7GAp-DdqgA0h zg%?CK)7w)Z^^V_7f@CxWqiK3BjcVa7k~us!u1Xz^_+dmtS4EgZ2c~?kM}N~o1ml!x zR;9D(5AKgt_k?rSb9_Ssr}-+BS`-s#A<0_W5r&LxZ?Rn?uMyyi$748i=yR!1uB5oY zsNFg7^j|8ta$Ct{YJ%vs94}s4FE;htcpN5>SgN0XK6@@WaK^;%I13chbhW_*vXXVx zItr-@F2pl~YCLU9Z1kFvhau-Eq2cfDE)ojGiNv-)yYqXlBE|%)JXM9hyYl2-{d`|| zA3{A9nxE3xwL)opO%L5qrd26zJXouUs(`+SJ!C;x_Cv#Yrbf%y%af)bp$FYx%fX!@ zi{NYz$;>+{Jq7{U9h>JUE`r~$;qi@i8!$06s*?+O|A*+uf!=|d>)0~^tG;1(N3eEo zFSWHlT4brtt=NJUpS{6L{kGMu>Z=IB(EPGNcnfL3D|QKZ8!~+QzLsw>&%uK*+1}8{ zD50+hRk$pw`8sl|M7Q}tFkWi31RuA^0cb#mrZ-5_Lm?^LkQ-#{$%uRR` zwyFWaJ8U$oz0o%vs&H($*|*S5t7_7OO>WV9(|5Q&1i+|N_zP5e*HqfU#FHST06C4&S z=s>2Cn>in#px&F%HizKS9XKn${)~#Fpl79Fof&X1h$=|Kxe&3SEyMd}^J!KZEl&3yFXw+XAa}{)7gAoF#;UhFF zcgfBI2De&TCp| zRD-ggfsVVL z(es)%bAX(6S(=_6WG`4R8?)sNd;XqE%GMTN>$wVGzhcjj0cZDkY$<6SW~6s)H?Qik zwn9$cucG$!p&Z*+=he#5p%g^rwg}qYt)N84-*_#0W483E|GO9#nml zwV|L{j%C6n(U~%+85#6w)WXIP{N4EYOZGRk_#kj{2R5^Fd94d#5dRwStt7;MX?^Ce zNM1LOF%>Q9;-VB+ORHLL)4g8CwvOH0hL<2b3P$;`Ns+FtO=x%6zfF=i2}B^&Q3>Q=*+Euu{3e^<4C{O`sIWL`uwM-y7hb7Ve4T#Y0GvR7Q zw1?)8&5ClR24vH1^Yn2KNVw$@TGc3Lyx{d0Fal5@%@oQO@)e&ww0X&g6`Wc(5E@BJe|#9rs056Du{V~jlzSvf(+VM2vTR%5$xlDjfz z`oD{{O{sgp5!IuC5s8D9h#=Dd{XRq^?V0KT^Ti^N!6gFY@rQTr6Eyhx^IFsBD4@5_ z;~x_gK56ef$xTNv7n0yZM3h7^gQ4MtT}+Us=2DI;gm!#c6pe?x^&a~T(Gl0qnS2Ch ze!@z-#^zjiZt4o7y|NAJqbuzN>&E^>c#0%&J|M}BU{#g^;FA;4r0+|_YEvNlc3n35 z?wkVac@mC#K^}hPl1n1rgRAh-@!Hi9hk}s-mc}-)r0crk-V5OKCrcs(hZejuhtfyE zN_u_OX;#XXmHe(M90AM^oEKrG2YQyQAXXt;tHm+w2DEn5&Uoqq)Mt(Xpj<#**~!+5 zt#1}>vGrUlo#LXFcAST~*BPmPqF4B#{$5E2+rFG8Cr z#2saRg+U z`sI>pikY_JvtjG)+NpNB2xVnf|! zQZb>&Pi&P^Mv|Rm6RZrW-`}PnzX*G*n8TlO>#o&|eH$|O8L{8WN={3Fuh0Scj}(Hj zGaxqJ_+S%}KXA8|N_t(~hjBCSY%8Y3d{%S7VMdF)S1>A2%bV1f>#w(#O_md6OY$8_ z%V)a85etkdWnjpQFT83VUDGl_ zTz0NDE**s5m*5-XT5h7%r+?xGqEw_ht=C_`hAZ&Y&8hc~S|}bUQF%;xbi$1}+w%-j zVu^a;Q-QKBk50Szp*+twb+`ZR5;O{i7I2lU$y(AajlEx`7zA14N6q5dg*P*_QJZp%!H)$K9jrCXc8gsAWlVkt=_qn z;7H^;PF!1xbg3^)B3t>jh%F5h@_f#Szp>C#O&2OKtR(~@pBk1%QOL-O=EI`1v*^R- z=5!dW3Z+84{MKWJ&c+>K6Jq^Jc_!YrUY03BtmGA8HAA@si+%VnO?SPuRiJLg_mkl3 z3qD#6qc7*5bBP=S$8BzDp1eXSy2<&kZwT4?@vRU#&C#Dl=#!hn+MCzs0;Xq0y#a9; z>ADieOEahcNq^-2B)C<-He+qLg2#l?Fr65cpEN((#YG$_L-=mg{z%~ z&!pp831OcL)7B#&v<$gj|Do^K^6}>7Bcs>gD`OhwMC%GAqVFI-m9NS-Osi)+$#rcx zJ!i1|VBYNtrV;~d|)vRdZ9c4S;gd65(B+_%)a#DTaj^# zXibmD$EQ;Ww#z4Ch=W|BN)aU#VYu_9+ulOhNpxYs@1O zDm&+um)h zxtV4~Gx;n9LhGoAS8cAC@n#wJSuYAm4T=t+H$im3968XR;(jroWfg zn%3f*(k~V$-Ld-i5#|=vKGc0+V{~15I>%IO6R-Tmd2z|amgDV{(hPe;;=KhU=a%rF zR|Wyazcv@HyTW$W80%}CB_dk!=Ad{DVc>mrCliwxE8Vj8Zu4|$`Q2tW6D@;X+3RN{ zs<4(wYoslB7PCSv#3LIp7ll?*5=Yt-yq0d(Z1VjFx_Y$-CfIwdc^D-5C*4rVmk7CV zYQvCdeNBh}Yk=SC@kIUbqTDOq`PtDd6?>~AW%O0wpB6#GwQZl+b+_n?7{TT)Ht)v) z`fG3E!rQHp&2M$t>c#`IoBR0=(%sVK;}5(8t}IK(3t=zwU{=0!Ut`+P{9!_4*c-M# z5N9;j)hh^o-#z;R&|B+QVmVJXZs4hmBOLO}tm`fdesQ+QV4PpXGwX- zh&{vg^XU(Dt{x$9$lDl4>p0h1XHKHO7p+wJTHlm~LyT|PUkV*k*e{wgJ5N0CnBpo$ zU)e~y{iIAOiGQcfA$CMRTe0+WXSv6k0`P#vx2#JkvFC=AiB`<&8k5(ce^%Cw_qwH< zqJC?m=lz{}Gz>)8WPMr=U+2^lDQljU1 za{r*2Kc9M>YMq~n8-aS6&k6wF=bjjkJ5`90*mERch||2SaS z0cWpTS+@D%R_*w+vrj26CSMLMgdV>*my7uDX2H)4am#SMHOq+uIsqB{CN%v;E|iVh zaD|$5Y)*M=nxQDlY&b&)dD^M|*cTF*a{HdTz8MmVfZU9yAGbzMK13;1-sG+uVb(t@ z5_ehaqy+L!)!-*V3k{?DFDXOevNXmVF!qh>&Il(IlYpeRH z{GEUYs=w)Si&{H^`Rw(6WNu!c&kcPT-JGqAnoS(mqbh(86lLqku9;N?+Odih(mk{g z?$Nn!<53gxw-&^AM6UPzGBR z>TeFB@6?ho(nv3Dt-$53{AKAK$FeTR#dQ!q%kMdSwLJgnx?F}i;>_bYR|A!a$12bM z3Na|A^q$~&dQ273bc}cmZyaYuNXi$Xo`g?th6_(_R8+eW!`}gCSWJG7(@EuK z91fkmYgKb~BBq35mI>*top9K?6Cj15OfpTsnOlweXK#SL8bJP;EM%2aw1j8AQopOZ z;@ZNt6abr>x(%Nh3LsO&_%SQ$A|B#;#7yD(nrp;0IeTQdiGzIop0?*Umzn<$-Xgt|=&`X(Dy3c$&{LLr#urTKC{411@7>vShNB8H$ct}4-fcl* z>;MZPiKp)@kPW>A11nKYX?v?oc%+?!)us68lrM$#Sj{W#C9K!yY^%wy5FJ&g&HO)e zUVax|?OZz@=Xby5Nkgd^yue`2(iJ0-t*D{1v3Hou$|x@d_T}=;)qUc!_{1-oqR%sqk=_vn+uzh&xbA{e4I%cy!Xl4`{D`dKieVh6v|H>z| zYai{L){M;(hkV>o;Rb92zg=7Y3#UUE*0bV|^a7^{gim2Js+$zzRFiuua=(as<6{qg>K4jx&CWYh8a?9;NdDqJ@BAmwIBU1GyLR+-D>)W4*_Cc{i|^e_cz1 zb-nVR%Gc_q>z?f79$1t`$#0IQ!LiJFR;uxN6_(Q^tNyEB!=Q&+a1Do;Ip7^C{KIGm zXVPq6P4*JQi}+}y0sl-4S1Ygc!qtNZQfB^x$cu(CutaLF_SbwFEX@u1@wi2tEJ_6~ zce6c}8$E^zNr+_mEUZhMaw;X@t@BC({T?~>zVI=GtI-zwv~dmB0H}xYiS!Qi*Ml5H8ryIp-cKlL-jR~ zFN`~x)5=unFE;7F39_2iWaw)5EE9#7Fl-R1Bk|*5%OPgyEG%9VD|BRqtmdIN7`qtC ztsSmpk(7g~`T9l(W~oagI(pzN`mpm=;gMbXhJ3h#e z*f`l?v+II-p9*U!|ER=bKT`6KGo;Kv-JOOjGhQHm$7MUSx*J z#tlNog=CR$jZjFGRyT0_Ur!h;*CW3*b|6ln=!vj`w^SxmsSk*#FI?ucyg3W`HePBc z1PblIrOuX1rF3{@&p=K(ztZRn`oDgMIg?<;OX>14=y{_1u^;8+WnxJ-T$8gFT{-oR z{2YusZuZKUOEakSNrni#@~dwhLmGXwSRoXw_sJ`0UQ>-3%JZ-;EwEm*EE_4w?mVA< z(Hs@Xd(C2HwWt2IaxPdiRbWAaCMP;{O+>Wr^XkI>O6-0eSMvTfbb)6h&bs_BkMor3}?6E0LhaSxFm0q@Xr z5A#&vd{6NAM@sq7*+`u`XosqdCl{2+=nw5$(x=K)J|0mdA^bq}5qOx)2~3Ew!bz_Y zq}?ro2n*C~{<9!r9{tut!rl?Mha5H(CZ_&GR2D%HMG}9d%`4$w& zW38I2OKg-+E?m55Q-Yh$bpczAkpk;C*JyFp=6y;V=#AxxZKjXS+$1PS!G&9__Qcmi5_FD z^m>7O4qGBQzpP~VFd-UGyJ7WGfZxYs2ENp;G+*O-=SxdQu#*}J`kE3k3)CZfUPvEw z71&k1|J#b-d3{z9x*ZjuPXmm~L|Sh37C70pLCpq7OJHAf^yH8YbZKu1pZ--9dEbtW zD+7Ko+V3ETEjfOM;vjNt851%=XtE4}nU=hh(ChgCSQYhmHyy08IQouEH5So2&`#(Y zhf5s(nAiks`05$H#xx+UEdfU5dBey;OePHC8jVhA~L=nX9gO%-LVJnEMojQ_KKysPq=6_J&>_a07*e#q~!el=q z51pP4V`oU^`C*p~u1PLc72Y6bDryF%KxYQ8+}+74}@a|CtxAVe}%F zMQv`t55`E3fmR->V?tKKiK{(0jNN4@7a9r;Ru$Z4rW${RzEgVR9Xx*F%~v|($pb15 zRWOqe&m`Gk*tJ6WrnS?I{j#q1z<5HSx8d4X#4dcc51xm1;mwILgKx*$(-;pA}dzPz#31Gw?P-&zAX z`%M(aO}%DhPf0e7w6i(7cmKM3glnZheaf4{V)ENcc5tQgKq%%8Ofu*_0R8Hvj~){7 z%y#vqk6O-#JzQ1~M#-qt42|jpFd;%u2_>MP+dWtWH(F^df9(n;Tpc;PeM|)A<76G5 z3)fCxN2Qjgn@}Ww9^{t7roYLg7EGt~+@aJr0pqQC1Q7 zEB^Fa)nmN^Tvh+C|CpiOw8YATsvOREA1%S(FMPx>pr4I8-1rBJ>ow54cedYd9( z;Za$Jm9bS~r}$h)Qwe4AsfYBPP30e!jWPIu`lwLs81Z z77c?S0X&SHKNnsA&eO{QKM@HiEo>g(bYKc^VBKr2oz?(d;8l~rKj2G!yfn4jpgqx_*^(x`GC?(1E%*Z%gL%Y{wHf3QjZiK1mxs6 z=y0${Kn_^e%D{I=v=fdVL=OB_(lMUEr>FjpLjTaeP(*Qx_B`p*P#7qPIgUrlnY)=O z#CkC760+1`T=V)LDIfJ!6K-(xElh@AwM(*t;GhSta)+{f-%P`-l!DPEQm}BlQClB8 zc=t1;zVNK{trAYER9NcPLoF#fLWpD~w*~*IGKjw>0ruE!YdGRAuE(acVpi6ZPnN^$ zw`|8oacL18rA`w9+1)rn2Ti#Jv}&hMoW^}+#e^hf z$y2Nxw&qu(3|vK#Q6;RBPgi*`c9Q&6eedA+)Vk}a!&HAQtQ={F*<+Iuxd?_-NeSyy zOU8G~QW+oWj+p4t3kU^hN_dO3xe;cdKd3YOnXr~gkh z@Q(d|@`aoZ8`QMOvaSQe?hSxLkPu^w@k$q9A@RK*tq2x;1XN`#ACU*dfD3a~CECT5 z$!go(6CSaIz@!gJ0ke=jIq(B?a5S3R0kk%72OA)~cVxtws<&GQ)AS#=+c7DHUgH8rv7gb zO4+$CSUdb)QaBnGeb^`K?yQ!@VOOP(fGdH&f2pb>o5sE3YK{GW)2jA@e<(r0D=DX? zDXV4AN3dhLzy3E*Wg;pX-Z*o(rns}ZAuAkH3Z_ry`I`{1GlC0r6u1l3fax%h!y}!P zt4CpgoVRjdj2Rc2sUpK7r+0eGY5S4t4p-=iEf!6f)Ws(|f0wRUlSp=+=xwV0f-{yE z`qaa9Q1vNP*kl)>5-h=+JPmHNi271o+@JI7Gs4arb1nU*?fZ+ijEoe(N-WlC^gHl3 zAS~~$iYzi2nAd~_1On&yss&(4IFAq!GB{R;0gKX z+!LZQ1DH>93BDyj7)keYT9Z3YgRa8*00IqnKwFK{Z3%{(W;zx2ZXFNBG=0IRiFO`d zD1eEl*{x)eivGWr8><_CdJN#GiX^0BxQm$iSW4VoEL1_78kv{}S zwd=qC!uLn|39HQxY$i&it}(nbzc9i@Aytp#)AroP{@Ed7Frui#!cw?S8`j`ZI*L`{ zQIduicirArc^#YUzZ5Wt+&qmFvm-;}(~KIZJm;#E2Yl};bi`-)Mxwn_V2o&TjF;$v zv)NhB|AB~I!;WeNq<%dU4^QvVVm94H>@w5-!oU`R|DROETj!yia+=OUxcH3xK8 z5q~>WXS9)G4@KaUIFQK)FM>cVr>9QycUGE}A}WxL#>MOg=WiJvPmG9D+DVKczAxPH zfB`sX@bz0qo(9n}(H(e8_!6^$pkMHpt?xN;V_ddm7g=#H$mm;=xAWf=xRupKj;c3= z6GXekcIDfEd2fG^y~U{%GO$XmkiwH$i$$#qIH3)Pvx+6r_<#}7{@bhUAF@2avf6)z zyNwjt#@w&qbid^?)C@V_`}5e7kxD39*~KI1{C9!u#Mo@{Tmyd(&}R-G$1zita30^s$cC z*3tv3c$S~ruRx`ar@Zf)-}4e?kPb>kIE;&F`H~A?;>&Bti2h?wo55nEvly2vS9*>{ zS3e`;8rG}Iz@msQZ|KTeZ3#LQ-xZ?^F4ReIrKdkipWPP9czfUcd0n+Bl(6TCMrmI#+(Vm+ ztJ6pEIQqSI8|_5OSaO1RG^q55OD0tI)nZ-9y?-h>585l`Q6T<|LnOSL&bu=cX|ghh z@Ww}Ty@%VE?b8*rubeS&2EA-t1E5j_x7O1X%GmiVQ_<=R6!=-Ou9f>yUFP+X8xnuz zA_lOe`0_GPXFXK8Iu)~Z#Wa>(Zt>n7dGgG##)10o^6_bh8~|@ydk64Xi=o)TgA|jU ztc7{5?8-ar|CDDmcvNAq*FE}B&*TrSfnOsp4ir1?`Mc-e-rBU5A+){x=gP>&xE%1G zP1ND7z4^7E8(9qRhs_oZru9t9jpxpz zbT|?XR`K+P;&Zo;%QtmnSG9{`PEd>)>b%E_B-v>CXB`#K@6qM=rZu#CANZYh-tSB4 zF)?v_d8*QyS(lrwrTx#^lbc5~Pn9KRh_%cZt0ih3IdVi#^ZKo$w=!g2FR{Jk<1o~T zH{<0NZ($1^`%WnAxsVgXG_Z9zy%Ng6p(n=!y}fbMW6ngGXJ2h9$-~Aw+d}QMzvz=s zh7A6}&+{>3xQYS9ZQD!O6eZenH9c`}OkTZ3ekY50^KWiweB^bF8#l;YYT+dG_+x8H z0YSR=hn?hI*~qU75|8}a{RR>&ZO-P`emZERd!H%L^0H>xtF$%Fx90trDe+f5SXolh zo&LNyNx;MS3?sj2CRd}p2zq*sPjdnYl}UW~P|yD*pwVAk^`*j#5N!0wLH#aW<6pen zC0iPIs}{kCF}LaF(Y{dBwVYNtjoD2ob~`&ggVNsID1G);5QE?x?8xMe=$4S6c^hgK zk3PZZLS^6c1aenb;ybU4W-8)2j8BwdIKHb6>l0Ct!S6Ljd)*dwZq`TNJp^nsuW)4ZZYWiaeDAbqk49&gw6Ew8aT`d?FF%>yz+L2<$B;Dpo>$fDjy z?)nGcOI*fMH;t+&ceEP({;H&>KN>VY0sVRo4<0!*Ej#(*)Psnv(xZJGGyYd(P~bJcs6HFh9{Ja&>gaRlS{?$whU~p;LnOK5d~k zm$$#j7ePZGl@G4!z5?b;Egt^u1W>E@!ea=cr?1TgA)2|U#)#tC)fF+wRrKh3Qy4dz)r%$AUZ z?nT2qNy8kfrXL(a$(OX#acGI6oC%>wIO zuN}W&e8^V8#e(;jiW!2tn4U_+Rg0U9)b#2vDt=I`;0^PiQX`_9-hZNF-KkOnmaSkh z7#19#b)?&ej5>%j=BN_UDf?aNW9;2P~q zgbon#O=x@;t&9w^aiym}6Oyw923OG&6Trd0@pqYE6%iYMzS;=Ei|Ke(PUqEb{*OCb z8v12Q5?x04`%<|Op4Jkd#5S;`<4B$(K5xK;q9E<*QDFIbsB9G&5A78rPSN!-Oclzs zazBL~H+CULs$q>>EHDBrTO8MJUq$6Fwok+NVBkNeR139zVGTo7BJPp}f2QrZoZ;DM zuX=|^JT~JR5tY0C%lIwG5qx2 zdD77JL>t+=Oi<2sl6zAI^kqc!`Btcoc8L)D6ygQZN`!&!gW%s=h+$^YR3`fHh-JdJ zo3)|Lbl&36e~?Ak>7R}`or8E2vdfQWHcG@GJFdrWyZuZGlI38RdGy`r$N)BW&fT20Q#J>mbst4M{Cqt7Ct?+={P-pZaPVv$u}}(+-fVi}HC57v z2q>TJGmC>r0mYn7wYPvzohdDU4s^#v>LL>5fvCUA?9i;~!zY zlFA)N!Uj%Y-(!}nND)mU%#vS1eJbI|rA~zQC$SQ&BuzT`HHgeV6Tj3BPuPU$^vHW* z`yT@CTsV1U7VS?=eZ#JZk9)ksBzA^z&HC^NJ&tv=ZRQd)t)y2WgDDDGqvmPbiVgPX zu!go->o~Pxpgrs~ zTdQ*i+2#1oj6Ru*`MlEt3_ivFztgv89IM`EOVZ*hRCUbuuiY{D*l?EpXk=ZzhV42c zkEA^T1Pn`tk0BpsthOWKU(J@!VzZs@khs3QF@mnv`|4|t0DCy#jUEBWw-%b}J&?QW zZS`F#L?sjCc;rFnx-4g;&ix^!-5qPlF)? zApMtBiWfD!nOh0{dt~BU$$2-)?uuo!$U)+%S;Xu{6g_rm(XjbND}wI&=V>jZ_p;o2``9-XVJF!uY~$} zL_$J^KJeFQ3mrEco!y18ISrFiK@b|6Vg# zttwX(bAM3*h6#>5**9bCQziJI42ia>5i0Jg2!{azm3>m=7>P^&zsth2=xovRGCX$} zte}#C>a^`oM5x+Jft~G>jNs@AJtv1he!JCnB1F|0LnB~nSbu|?p-Ea%HPa>UqMLdJ z1sF~)F@Bd?<`B(MUDlA0(I5*Hlu*t`nfZG!>LGQjaq3Yw2b7as6Z&!%J68IYZA_pI zcuwUJDL+cWO`97dk{}Di4Ztw$Gz2@Giv!fHsdo%{*j0czZgg(~?g^fWA<$QOu7U2I2b%G3MXFxO^G}~N2m?F?(^*nGun6ax)i{$ze`(-`ivD*l~C_gcq-Z} z1r|sQ=q=V(?RX8w%WrgTyXPW05ZAlV*g;ICXHz%)4LRUN=DF!gm3sZ?iJc0~ub4)? zMDi~Y`g;^sXR)=6P`85n;((`AssBeGPd7UohV>b|67>>AG+EN+1>k7Ac&!uIDhQ{b zg&929KsyS0F1z#bQE&qn-bH0cyr-3%-q}8&j-+Jy2AXa&2z!}i0m%x) zuSzb;UlrFohNo`X*K%tJP^D0cPYU@5T~wB$`$qtoCHTq7hs^Z9w_bmur@Ljaqwg(X zy{!cz%?=tk^RdTCR?cauus>3Gs7yky&FB=P@6~cs(2qD2l*z_jLOk!x{K;biW%pZs z7;&yR*U0l2=wH@5X1Zgf5yLw-jUfoRMXjzlQ0|}fqBa=tGd;?!?zG^c#5U_s*m@8+anV@HD|4oi(0yc50ywSijhiCx$@pc683DO0_ zMA`M#%E5N7C_v+l=!bKCi1TB5RNUnd8-XHxHttupxvbXCTTja>jjOq5_-3(oH%?w5 zW2-~n;dU41x6=@w_yi|B?$rLsf7G~&v>vp~n#i`@;@8c+-=_Pj>vM4Dp_L`*H7fHH zc=ZsDF8|y9;Y3e2I7=((EbBHp7GQw?KCZoJ#a~0**@)nL!Mam!rcht$tV4Dcfm|u* z%IJj%_H~5pX=B%yKac|j|Iyw1eNO8qycOe(@WIE6F;Xq-*HP69aH!>n7Yb48M6_XB z0*^{o1o0+BF0+5<|8IFy5EHK;$93Ri6WZ|m@WF}VO6f8ma^TT2E6G#z1({I2&!-cC z31%@`3Hsoj6P@?zX%J&5`=ui5M6l7pnY1cdHR$_YNkU#`kKNJxXP0KOBwe^zrzR|S z1ot9j?KqZZuH)hWW#GYo!k=KjVY#L{?S1WysVlvm_1Vg(<*A5vs)!107f@e1kJ|;_*_oZ%2jsxuA5ilzRoxxITq`*) zlTwzU1Hb z>Uy70jCg9Jeq47@v|q@o7%`8vd$j{>^7R5Ip)BEuS?qqAA#b|0{ZSa7j4N7>=X^>AIKL%dGTKxo;ALDh3!C|Db*(cT_#Ka?O z55*jsm18AtDn=uWQEJ&x;j(=5 z>xHRBdD7$f(K0-aGvB*W?~sctb#_{q5i9~}86dXaCN3}>N&W}V5wH8%z_0=$PsGyk z4TNU(d;U*|-UV#=L7vjrm&Lb${;osJAKzQR$lKK~vb7NDTc;+u@C`g%YrF3;0`%RV z=|tp^(7wxk~Bi+lnbh2)i5a`)J)W`R#sdeCvp_&JqiVnlw9Z znu{RfY`=Ud1-;uyvg!!h+JF?n_H}GIt%P>(p4}(JcYnEdfz>L+>aQx~60M{N919YX zc4Am_L*;kAK!qWHMCY*$x-#*vf%Rnhuqgk^zq1I<<=07^pc?Vrr_%_qGMfu>Dge)b z?L&}^Q9@=kW1nCK5u;fJw;yZiG2-I>p*VI#uT#rqW@?pz^{da;%($2!?rUtV*o}*M z+|&m7{Hnl9;G;r%e=|*@u*gJ_Q8m2H#IVL3A&)JdW@g|oBzav-4e{q#15WtffWaez zQo)@O1#A+BL;G@ie|35_3~cml7JRkp;;X;xo`_vEo`r}+^;^#*x(ZpU9pQU}nwDxRk} z0(4q`vbmz&!U>hPNc;hwpV%q?E{7^Hm_@%)>a~?wN4UK1AA=O>e~Y&>-qHVl=Wo_M z$cuzr3&mhr2%3C?19}UNaYud~c>|XKdeFx}+>cHAODP}B|!@JMLf zEBMoSNbm`k|9G45>N}Kk4)-rSVME{iJU0NiM?(GMH2_oSCdqC zA`|KT!VEn2*NG#vlHhiI>edoLbUUdP*ithxXnuJUvcGev05IaDq!qIv3lXR#8$S9M zxaQSHMR_7bMKci2+@Jy-eVWTN=q2P#QB^$PuwvS)fJ9u{-ZyoV&_rZ63FzFtGP+u& zQ7b08(>3$pCvWLNUKL611OuNH>7#HK(vtG5y&y_+3CrF+ZbMhnbJ$u(Oh(itIp_l9 z4Q251D#{sqMcV|(kOEpsz@5-;7bw)U){r#gr*UNva_CVy3R*Rc!^?K}Xpj)2E?rTd zhMT;!&KNMn2XUJ0tjK3%33SNLLgBcsCU_F<_MxMO&(Q@m zReOW_T!p@}d_YBX&opm7F=oZ+g3EiT13SxMdVdSuz86w!8#Eyz_@9?3&{qn=oBcql zRe0ARy^^=(+{nY+MSl^xF3S+Tzi>O_6%E&20tee3MjGqbhNPix0-jNWhL)m4<55#O9997{Dz`wE}g*LZPc=S zJXnG0arsu(673}o#W5B4ppk?6-k+mDpWr9p^WtUy$57G59D3zBaf2ax>^e%d!b6{o zB^(_CEI_K(wuj1p;KAD{ZYoY=xk2)fjpl;*y$y~Az692Hx3x=PGKWdTn27irD%_8D z=wA>6T)Vn@GAckNj-3_%_a)Hn_cv^)V0$m{csDo60$kou9Ecbc9lc}~3F4%h&rYFe zxX|f0F90uRT!{QIwBGmJU66$H*1VZI1P=yDiBe~RMoy5i+MaM9?OZ$uXUr|&UUj`; z1*EBb*^bdgLSs_`Qm_XOQywtHbTqS`V1}gf_Qf!8d+9?Sl2t%p}c>6VzchK;_Y|;Oq*U3`Dkj(ih z`~nG`uzXgK9=D?M*e8{Y{qV+7;pNQA2`VmsF~Av4g4sp9b(0}Z9xBe<=JBGf1P`OC zsnGM!&cdnxBkH;Xq5R`N-EsEWE6&+_?~HS1WUB}vBO^r`h&-H~S(J>N$Y>fFm2u7< z85OC7lNpt>59glW^Zot)`2FL~^L(D?p6B!VyyxpZ$iR@=Ba>s<_qWi%4e5$XFK7cz z8tMTel)wwh@7q_pF&7Q7yJC5mjt2=4V?UyGn=l>MuAkY?_M>1Xmy-V_faD(}>ZaQS zFdHoDz^V;|@cBvY5cYtP0`(n&P5zQ7;<)x59aB!lGQHK_4 zXjj@Zw}q&IA2j&uOzb3GC}g!9^>_?KY7j`tFLuoH;t%o*w0 z7t2;a(#DKX1lpBs==LikZBTK*a~+r79OiriZe7pa!neNubzKYC4n`pkYYT>%M@apb=yy!5pzTNQHZ{j}ndGS;mZwY1+L0u7^5&A1qi>7HJ zgoJ}>QH2mEu05S`@TLC~=pN1o{bQ4H;Q)is@q%oND^S??O|?UBtJ)!8V|_)C*B|1a z%dNlO`~<>=k2LIOsmY4uxoKG@btT%3d0I@hRGF41ZCjeU&qyZn^MeQ6v=|u0Q}ejC zzv{pHnHYbwrA2{4uJZ(4weMuemCo~r5ZS}>1z3hgYQsH~mM}LUPSqb!7ETR>N<7FO zEL=K3dUpY;0Qiy^IqlHQa$E+Wo@qVq7j z&-1ynXc+`Z++|)EV){V_;*f7?Z{6;^&_7l+Zu>z5RI6-73aOiKGOx3N{k;b4)ct=k z4|CKh84XY%A!O2bpfcwN_%tdrU@jl>J$@2h!_hQPuIe>OhAYK9#Eci z0YTL~9%iOd9Uous8 z11SyTpfI3bY?xnmxc^v>0DAF!kAQRo{et?&Edb@vaQXnVqz$Y+%50e$d*vtk?ZD2E zVI|D|1suwj&nQX&sd0$woi4O(9=0S$FuGP7CO$Iv%Fu2N19(Xu6wg45VG716N(NDo zz)^f-WW-kWGGt<}$0F?@huVXL%|#F6*fQyZ;!mGJ780JF%0|IV^OH3P@P`qxIu5f? zL8rI3JI3(OeoZ~Qb_D^po6t;arM8~ZwtY937ZYji{jJ=%NO|hp)r&=nvm?$vMf>rS zVHr~EY& zy8+S8(FW55aQ>cE)*#*nicq6%2Yl+e77*Ptrz=ot*Q%1q@edH?fI3+etN-~#-MsnU z?`#l~wtYjz21c`X*m1NV;cfQ&oR;g`A?VM=?&GVvOET;RGQZPb0Aa*B1Z9O(?x6(G zE+eeez-kGy!<<+YZeAl4{};V-Cj z+{HKEAmgLY9ALIZ&ZJzldO$cB09_AQZ$GAW6Osn-KL$*gcJW_Y{iLpXu3>mwt}T8U z-XX|*7+B0H*cmT8O&R;yeY%Q~8fa`&ZvGnlxZqLv6^MIaJT|wB2Mg9Qmf#x-B+qaF zx6(=KinTQ=gb|#7f$Y8i8@M`V&+?TZ@CQTvcARZOJBvh}-t{(!LQ{}pGovOVczyBR7PRJ?zm4q5CCP0de=fHSaZ?8)!x;~GbbvV6ewyaMmu;eY1t>{CZrA#cs+IA9Fa>Wj(?A;K+NUarwI;J$3u=YUv5;elt`> z#?&jU8hMa&Ax10A=mjv>mzUdIo9Gk50|t>Awe4HUc!zF0?V>OYF~SFO~g@Kas#cVP;;CqJ6uVo%U8! z$BDWxf+%6_FYwTQ5=8%k;>*U1)G3wt7UV zzvhOlVHeObV+P^X#>$Wn|D{o)$OK^d_u=^$%Q^U8ii~TRTW?=(ZN`Ng>})7{O3^T^ zf5v2apWYdpI(b%)_Wn$R_V(e%?h41W+{nNB+r> z#=a_kg7$vB@SWNwa?Fv5mgBgZ0MM4wkw~zFN9cwz*uG?#y&PuDBZi{A|2DufPl%l5 z>BLA24am`izGR->95e4|lLWRO30u=3eOXzAcwAcny^ple;7=TF>Knf~8N2}gQCyq? zHix*t`Tf*gVB9GPe90>*vO zQjm%?r~i!4>s7Qw+En)}_!|}fUcYluTeybNKP=ha(UN&`;)FcR7Ti0}J1#`YyIcaHs*=6R>gX=Z=1lQJkNgW8oT1zPc4*KU828S(i+v2WS5BMTL_9*dAJ*WWaV4eq4RzVX>}6-lC} z1V)w@n>adg*z)?(1r?C!^@hup>TrCp35lbTctn`Tz+HC4J1H*id^`3Y>$@g|L1DY8#<5`1z|6UOdk34KSor{(xLxVhHN2o$))v7|F}E9Plakt#<5Ah@Gxsc$zS7 z6UX*%J%fw*>&@5k8%?o0|JtdUn8ac{Jt9g0v6(b79kwdX9_We;mC%sO*}Rjp z?-&0_A&6s?Booy6WSSotrXWoA`=t>CKHIiLHRUO$-myyNeb zZkr55Z9o%A*?^13Ur&wy>X~x$rOrQ2>2EhX#f#{BzvOH*M@Ak(6|(R4!p5&N&NZRd zWggxa9+0?l)IhV~d3$O6l3}>gDVWgA3eQjVO!^m~%~-(;A#5#wCw<)W6LIGzO*UPx za8cSba&mc$&<5)Q8QgVpdvWWTrpUCo*g3nsE8<4sub4LJL;{Q1uAzj4!+YsC!paag zO$wadQ1MRkgP_4cIv6~f&Qj_bE8?5G`WwOM?clPu0n+Sl!l488mQx5C zD=ep2{>g=mkE*;kt^3mW6XR+zO2pSAtm)>RlQ9mc##f^ZF(mRCg4>VCkM5wmt$%2N zQ|ziKTO3Ldozg~+ob3zQ4v9%~^dam(!YMA1=tkx9>MA$Z$@$wj?-8#8?S zu}ceu@2inb3hbjKLC&b`us{7WLWAklnyan88kKK4)%9v@X-TW^$FS>*EsuYq;?J8q zzdU^;aOYItUmMjPn5;tYzrmxB9#@yO3`Dj?Ox=ghgkZ;v)hrYCXM8&Lsv1l>1yZUa zp?)X7Zhq>wzw_42taXS(n8KQq$vpJKRjC*C&VxCMYh*r5s%_X@zcPq{|5MqB^tPc(COPz83V)qTLmDr&5z5Upz%!=Sz zaq@M7xl4*SO#6rNhnR<&4W7lHbws>QU3I(ub(Zk^*8YwTUc2tngP=Pm|3bIA-Yp*O zw`z}2LU)2UJJSb)a0$1MjB+Poj+$~|ZDyijjfkFaeH-*qW-n?Fc&q)qRH$!DOU$zK zQy71i%h3tGP9QrMWd!Po{j?pmfDM_zM)g?6=W$ApdQP*{S|apL;Bw0P}$VPftlTmEMwfd&E3&HZX&v3$HdW`0&~gKcv`MWNxgoi zgYDWqE$v*Up`?eeKbdenxhF|LG@xu{64SZVj86!YiELSD&7B3YKIKl!wS=F!f1XgZ zT-X;b^_|Ku(6<-59>~aKXZ^t>@5FfqESQ~=_hSav~!7#6LuG3M=KaL;wpdos6J0% zDS&;$m<*&pvVE^TkQFp;u=O6H()5sBAJ!LO*M_ z;3Le-blU?mi4WT9DB2=|ark(|Xa=`B99c3o%1+cI(QPaThF^Y;&~O-cLJA=lF=`Fr z@sqHFW`NN!x004;qBG{m{v2$(K9=#I_8Cg=>gLB&;qhVUlR7rjq=bcA&ZevlPT=n? zFJq($O@6mi-U8a9M^7Cpm$+*{vS#ajVh@66VQfHE8V_1yJl;4fQz&{^R*d6wi^BlHp2N3~|kjam`V^a5Sc#w2C{{;Kl zi7r?zc|#Iwl3SotP;?qT7a6sv9=GP?O61nT@Hm>=(-Y z1c&YK58njxVIK|bStrsO5sS~H8Ls3=Likv+?p#LgbIr@}S364ZGhG(u8o zqupSP_~6kCBe&G13u8a9BRdVk(BA08-_yU_7=)3F_hn$FhPh`RAbI>>GPWh=LA}xt z=8q!cTX_EVa2dLH4E z&6!1XNPxv6y&F#mhcl-AM=M~PmOmqW4MQD0(2zS=oI-KJ#3z>75#J;X(w*~iu@}xf zW-G#8;Neg$EW+v+gC}*B!+mC7TI@{|g~L(qnsP!+^Vyd)-YIV#6T0ZXvi&oDLGQ{z zUXLU+w7+`}ek3Kss9(r&rD%Ug`{M;HB0vIGKHOlcDR+`(fkBw8{PqND0Qr^08+J|O zze5&Axq;i0tU~#AtvW2D@dHVSt#3DKw4WmyNcbJ|BkQvkjUXS^OT_&ia6T{pR-$1z z8|{6721HrJL86Gyb1)v({f;3fXODoA#3dHb(+QZpmJLOkcoh%Pzj27xK*v%p3n5Lm zUqoxlt&B;Wq*RZ%8P5F|e9TX_e|77atrrC!%b*hH>k)_EQB!r=j$*z%8X`U!pA&9N!4$=} zzj;jn<#Ig5xtpDz9Ph;n*ElF#NID!F+>gd}l)~93`sF^$A@p$0PI`v7{^X<-PCDQH zl_2u48ce?5z1PVMk^YRio|?T?E5pU%akZU+e3;)aaV?l{fFI-rZ*a{X$pA7DqdFyC zmatyA6Oq-IRp(Yqov%nquDU8jhCid{2~KldBT1+DHp4MK4=kQ!ykA) zlM*!Budyv{(wQrqSK-Lqq~k>{ZZ0)!U-#J}iP0noo7(y(Gk>w;H*>zI8)0a~iIf=VYZBw*JNd2zQ&T}cL(D2 zKAt(PSvNY#(haLKzJJ5AJEA^)*q*Z;5%i2wtt6fw^Z}A~cLpF2)~0_qY=ye{a+j_r za?i#nTre#E#d<}6nX=94jR79sk_2hgz-3(5GF>X!M8!!(lt=4juq~uAn{Dk7!YvoS zK$=7(BVxJhLf{)Vo9PK4is2&iTg0}J>M2Qcn8kwx?deD~O^iiadub$+|ko z5gy+w^&uQ7l1G2X`-X;)Tg$)`t!kR4xUg`IZd6Gr32a=tPD-G5LS=sM&>5dIuA%v9 z)G3^qcQg#Y9V$A&^Gt{n2@Ea-F|M95%>D|z=>H-Ek=m|SH|j4iGgfD%Y4Y-;76tlf zqQThDGVen+;oJw=0|jl;nWPdrsMa%K;{yvjMWbHq9Q#C!XxCJH>Ow*Zv{-`y2dtGM zB>MG0cwjsZ78|GTSq>9&yBpP-f*|2!B7&`sSGpon#${kT?=tu~k{At#&2At}VR43f zFw-aQH+Bh)w6|{yFZwXH`1_2ojLMx zXg;S)6?+c4X+6V1PP+5Le~J@;<|&nY7&R-^_IO44&m~BYdbBxqE{2pl98^|vHmBCB zi2g8-_qVDGjHY)!H6ngJ_c%S{Q4r(T&)7CYmVLTL0`_<%lszLiVX};$$$57iELwTS z4FdI08N*d<;qm?WGYVz28>%3e28sS%$k}KcNa7GqB4W_f!{gu! zj}ZyT6{TbzQ(bsD+#uK&!sWPx*8}d5^1Ig5gykBLkaNm_(#!1xe%I41bany{B4W|7 zUzPd{UjfR=7pQpV{J-t}STB-T>C@98scdN>7k|kGXpqNL7){#~fP3VJXVx*ACxkdf zsX}h$2@OG}(GBzTl-qvu85cVkA#f^Kb{GHD5PKhLm^t=~w%N#eTnk)wMW3X|9+c`l z)g~>fq2l@52L+i?dN2Dk>Yx0=UmF4;q*nqWvGQ=)bL>$L<@t2rdq8A?VIe`}C)hv) zuBkUV2hlVoa~$yb~xJ=Lv1A}st)Q`D9+11(yEl<-r=oACn?<7a{5 zBb(?P*PaO~8|}87W@VqauIsH4VMrGiW7;s>1m5y{Don17?RP{j4$6gzq%e9OU?A$e z#Y&UxgprA^3Ky>*TN%*-M@8OBO4Kj!)R6vPz&q9_Sej_t!eg9Z$~lozG$QkAK8Y#`kBlK*rU|Q_o$z;y#_o zV85G!mZPrrwd;JS_`m>>5xT3N*AxuttXKviyk)uS>YGs*=V#WSvd_T#;}Ao-+WQ92 z_Xu%e05p6)n&kNdrsf`mfvl-pdz&wJ=-U2T8(XmUV=hJ0(H7=3?1{qUa3eY)wB>_& zFh7G3f$YQ*hV;mt-?e0EFAV9VEv5d@T*Xm(k;8gFzz?m|b5V50(h#(}UdrOSKu>96 z+aXlY^2T7{$+1%;`l!8%Qt0qI=Z;V^U0Yq~>N)6VJOp?4gBg3+OQF|^g(^qeGWNhq zUqWVDO)CsFpj*S}33Z?#4CU`gzfXe!B0dC0pi!6au6n+j?=hQoKa-*&XrY7!5b@{?`(i0ra0=(KU`&M#5 zm3j=_O4-1b6ClqT<~3Beigg!1bn9S8gSzrYA`1(yXE38QcJa=e`FahO4iRVncFdpm zkN>&?!e|cDg?w^+RJ>~crnZ;nxcB4VD`5Vk505_qe{bC%>^!_*KSHe^t^~DzJkfje z>2!n>6eRt;8&y;zOg7uhzrT>6#3)mb15mUrZ)cmql?2V~Pxd)sGWt?5iP2$7zuE7B z_?l_F*a`3eFK{5V1pb`F*)o3Bme}B@jWxJ(%xs|NTn6Zl2Q0#FM}cwvKKh?e+OxLM zTng3k?8uU8N%xh#&fn*?+HK^+fGeA27TOK z=I~+XA0tR!^;~L)wWmk1m(FMh}cZ zokI8{r#r#=x6vXygkNEo8@Dg*Pi7apTc)Mld}3x89)&bz3|e;**Fa0bCj5Qw%4GLH z)kHV*|9-1o6nYf2tWZq`DB0qaz=ff{ii6~QFg6?Ed|WUA^jOB6*zpeby5 zT9K##R9Bw%AUc4&s?&QK(RbIYoO0O1?u3?KSOx6to0+J1u_J0X*5w6+irJ%KJ8|@A zh8@DI0k^G@4c{y!2~_-K*I&SB(Djzs8m8TNfe%t>G__$j&G_biSF{G3+D7DECQ8{O6Kxy9&S&ec-u$Hjb)cfB zZF4c2t|i3Y-$(xiw1pPR5)B~mY1kt0#4i5x``ry0*fA8VB%I3js^Fp)>v8S#heq@B z_{4}_N9vb(b>sH)Of>zUzs(?obt5Obs~eNLlnvo&og^=+>43?`wrEa;klRKKbi!)! z9+Yqsu%i}--y8>y2N&YO8*91MYSg@8*uk4Xii!GRZOZk0CgftXrr<&+>>xvZ1gaQ& zFdb;>@3Rzzj*N0;R4|0GQdf4z>=ZM>pW7#epzx3H?A<+>r<5@>py2vA_@l{?#&YDH z0#eZ%s^1~BEn}{;C6bIUXjVlYPbGQ$@p`RnGrhmi+O}UqJMOrb{1sps?mTsKq-v@A zr8VPu9gs)EByZ!I#q7UvLP6X^Ch7_wQVqulM)#E)XKKv0iBa)L!{|frMIWjQ-(l^_ z3F=DKpnuK@F!0vNEe-@&Ez*sGnsV7mb*iU=Z3xsr70qmK#le#Q`L}4hA6XCR*G^02 zCG0ALd6&!PG~@{gQpfphoW{6V`_u#4ls0s}+<2hVef%Ll7}8|3sF{0&4krJ2nw8+r zIM5_aR1ETM^Y=NIiXcsR+)uc6JVhS6){C)NC8PnyzSM=X=r`tbvyUE;@XL!=*r{)| zNT8;4ZUSw!oyImkvXM@xUn3ajRNRI<%d?l!JN?N&MiqrhIOsocsMVeBf{{6qaB5ytF`zyP zT;)9$B#-q4Umb*sL;#aebiwMc=NHn=v)8ZiR)0L%OWcF}O#_`p5+Q7pz}=5@Tr`aA zvW7{Cdj(MWac+a}a2f$2&KX@~I+0t4-^7NK<)LuR!er#iCS4mgD(V)aH$EChgpr+v z+?q-F5ZL8G>5r>(`_O3-qo0SjcF+8q2tUAFngFWwPTh4E*~JT^xS@go7_$YI1x!n> zf!i$GD}=!NpihqeKJ34)TXushQaKe+K$s~B1N^WVV2;+TJERT3*v%Nf zrW!)>Bvs>qd@|cBLCYw~`w5f9Eh7H$D%T|8&)bCDD(uO6;{^P1UVj4Zm^v86LIzGH zlR)Zifw2ju)j98uL9^X5fY9Xkb7^f3nVzu+9qg4lbsD@?c(}h{6o6ixBOK}e7e$;+ zy#j4WM2h7L00YG!ZjRA0j^cnmD{qow$^_sQ8XH%W2?OyMVCg#K08oTFDr`R$dmw-P zTsCb6>lCn2GW)M!8eA&3gwWyb?@`DD66jKO%~K+J-vm6D5eYrBz20!YkcNqOmV-(@ zXoyY)-?}kE+xGU=j+g{+yLQ_py$me6P+nx=2Sp+(DnImw_P!O3XPn@$eEl*O^wOPO zCAeRKOwTgR%{4Or@FD&h7QsCFxu_lMa_*$N?5!ayVwoEOo~nUXQ`e#LYr~4I>LOqi@9^Hs88u|wJelQ&lm(E2hT{gqsa=skJy|~z%?Q=3Nj!`*H zr*{Cs?LcVP!7QM`0`Xa%__~Ns{pQ^nO82SX&ra&1_PA;iMM!%6FYJmEf+cYL$SqYo>lb?>D(C} zhR~pxSC8CIe;XgIVcwT@bbR?pd~JPzf6nkKRxeoP>jP)URGzyDEmmx|Mg)gmsfJz7 zM)yH1LQYWn$MOSGt{;!DAlM*x<_$?_lZ*GEDH6D3&E|I^dLCKt=vVV3KUtoK^#kwP z`>zeB?w(p?-0C{+6v0w$Z)|np-rQz=ue;<)WahuQ50E)Au=gJOVO5W;TRVns*Ow_i zNM=5T=3Y!N#0HueFGvntY_lia;qDHa97?&E(x>)y<%%_lrIa|ARwgJ>8;sRjJc(%Y zw3!@kps;wUJhDVhOY+||Y(!xL23fP5zxJ}O^x~vtG+u`;INfAox%?2_PxdN)$-t#d z6r(@semypZQ-^BKo(VsY+264o6Is8s&9x{(5B{@*8-iPZITDoc* zvJ_Adyt>{pQ~rb^+vF2$zN2{h`4g)jU+<`U=cVcbq&*>4{x~s5$Ah`g2Qttu{T`9W z(mX$Qm1xSbvCaH^2OrqWrsC%qY0${P+FNzr&0&Jf{GJ3X$kQPKkb;x?edDy&==7B)PIpgHDhGvS4xv*5uFmtt4;eb~^CmQORG^tEU*B8wf9 zlgVe!RlobW4UZuLiMP-EFtT;%*OM1D$aZv0#$nH@m=J>o<$4$y7S@>`Yd%D8N4|Rg zv2bQZ47^=su!EeuZEt=jk>la&U8ZXjDHvQWY3q6MY{$!|7cKsR&MC;UrEV}hw9EE6 z!x`aEUr-wSN|n#HIn$ml2kx*KC#{Ocu8x~|lsq5Z>Clcj)g2+Q zV_p9~+UhcT?XCoQ%)es?_uNpqGG=FIIWHz1-e1qIS;@UAhg!>Qz{i`}v7W{7 z1ajq@-;H+`1l_om%lU>h9EDrn z3KUNXx+EZ8(uLa}XDsopoA*}71+n*iQq+@_Sd}{Zlfu9{xBWKZwh8guz?s#X?|9EL z@|F4gxnzD5Sz_bU4P@jwtBa;U^D7p02@^^Y`b&z^MQTbGf5~aV#F)%8+MVEKNjSf*+rtyc! zqQT3S!SBl;o^tr9D3&4h*S?-RCs#hp->(zGofz}5?c^gVxtewvm1uQ!{rW*H!5tw-=p2Uw=F+h=soTnIE z$uPD7l}kj4cw4`k=x&YvY#Kc9mR15zUHSdrbIxiV1qf2KF)KpCwcM{_|3mC*_4>?Ns;ivb*XC5+G&{ea%L5s$nldT*30p^{rE=C%VK6} zCb2_;r`u6Y$%gGxH^%bGq8;0P3nB8?ZyJPR1M%xS^2NrY6c4RRQ!x|ua~8Png~f*X zmSc5u^Hx*0I;Hb|woUWMX40+Jynn^T68ty6K5^sCB`G~0-G?%;wVvvc9imC>y%;C| z)HLzKf4AOs{0JdBXNF1=qr+<4#@}D8XHT6evcf6xx7C$-%yRK8o?tLjj=Jhr_jCKK zB7tFe`xKVv4T52^T(^FxS!Ibze!Bg)jk@C_(PUCMW8YVmxkWtS#fvL4Bwb zzI*WAUqNYAg4=zICHKzf~MZ(^UR8>k;3(E%Y*>XEs0e zTzP%Ia{BH*Q}P>dv4>Or<2Zj0`y#6}_F6@D3)CPR0`g==(~*;#pURCNMM)nCf`7hu z)-Xq^@tFBLon_d9%)C~y{-a^YJ4@aiO?8!V=d7LM-w4CDInmFhwQ<3%2KBMj7jFYb ztIlS8KkIfJ$mUt%0{K$&)SYMflSuf9X&<|iF`TY~^tNt!@xSyp6ccuqEFeNkFVnyW z>slcgF#7jf?<%u~5!=)Z|NLzSJ3OBWP*QHh)B9G9HT!N~7+>n)Ozy+zbC=K? z>}YQN!e8_Y8lb0TzPIE4Ju#1er9~$W5mdM_*H0Wt48gDvm1&|G;k4Wc+xoP%Z$!*{>e-F0Yo}pK5IUCL7 z1ky81TFbvDBQGF&Z~=y}9!+r%&bgTyM^oihK^u}{74O-O_Bj^0r|Owbv8UYNtdHFJ z=Ig?I(*$j(SpY3^K`e}e!R(JXx5vms!9F>p-GpG@&~<>YbP+c7e){M1mip*?Uh>stI|HY9*rB4m~f_x1Q_+U{jfh#*wU^J~GWu?z+ zt_v!;6_IDo`lU$)9HjMeUJSpF$2low^W$*W{a}}FKZi#j4&wkH;%DPQ2}d=VzgOqW z&eoIuPTon$*rPJB&i0}1on!qV5QT0kyMms0Bf@&BAs=_h^Qp>*nFCaZpq#&VA1WpA zx{Y64uS$N`{t})VwjmGA=YvZAn*e7e!d`*Q1eDH$wb_tcurl2(P7 zH5Ae;aUXMKT-hYwvoUmA|ChYneOikNYA8L$MUyW@JE}ET8vbFbD%;XjAg0TRIpo%p zlGAhQNlTG^aAnauTBnjXa{}z-%4mKPOv`6vjYO#3-t``zAf=t*Tgs-x&QIexBVUYn z!nz|`2R^z0 zhUPi2Yw@@{nOsiV7zRl@nE|)DXDb=#c=*B3ax1fa0>?u>)(x2J#;-2PsA z5Z2wbJU|j-{6Wq9g`c^~0J`0uVx=A6X@)k8t}Nyq?PNbvDW zalvM;O?Y{_pkN4`Xsz^A%Tdkva%K{U@PfAK$c|SQ73Nyqtj**v!kQ6 z#AB!ydUo4FS7X`2RQmRJRoV9!cT1mRb$%G6;JC1R{se}ZoQQjcOfv5bA(D@iDS)Sn z$lyB5rD?PR&5*$x(gV$XB{`311>>(57a$I6Qe#Qa%NA$pCwNeJia!{Ry_gjB#EIoa z-rD?U+F#qq6WZ50UG`?pmV9{%@DlDRqry@HcEk6LhOpDgxDOE@g-RRH%p5?s4>s`7 zTc4Nh-tGx7Gebi4)bQJXZQ_$tzAqY!uL=IC6l!|1*xjYh%U0fM=r8fOig`{#Hs!vU z2DeCgDV0;M<`@_XBo4cE+%)AMj5HE&|AOwr=;~ed>|$ehxZ31N>3_+adC5%=zazlv zHCuvV&Azmlb7o1PnYv84EO46X6ZHe7crPWJ67#z#49@DN~Uotskosz zs6)hgU2*~9+O-z8*?vkj0GArr3%!}Gm7tm%&ccr`~GAWlBDG`RT-4a^Rb)y5wF}A?Zn*b2^XML+nB*u{jbGC8o#E#8!nVb z$*nJ2;s@^E6HCq>M}=k-q#g{orZK2`LlJQQvL%jWhUdUTb{(rKaOv4?H#K4Gct(v% zoUO9kz5AkNWgXL{1g1XR8(1mix|W9?ghBQF=U!aoLAdPF2C(T6d@NfOT8hp>Q`{mx z*g=TrUqBoek00bS8`*fp-$osM44oq}|G0<8!B>6@Ho*AX@YZQlu4QGVnDoFC{B5i} zVHE>Dto*dgFW|BXp^cH53{Ub-UBcz#hIr1ZqKQI^GjHg?RD1%%u~lQTb-1>gzoaIXA}XxuZXjk%UXZiX+EdeN&` zgCvhVD_$`382_F?6PSxJ(t^;ifW(EtFrwnpVz{&(Zy_`DR-A4nKc|_fYOduPrsXGi zSrc*%{Of5hfG$g31P%3)SCz>?6s{|u6xu1uc3it1i&$t4-aM7|G_KE#b~rW?;W`E_ z4W)WGbDaO?;*cr)OM$&ju$dR6tZ)1}YTF}3;i!~A^G6v~5UbjXbO>TS^;=Cq`FoJ> zPsE1!RV5pdk`vKeeDg?P%L-J_AOJ7arN8;h73zbuh($>NH{h$#?i4_ z=6<9qM1oTHHI-OGbe!ATQaQ0@T=~(;ZnstLyU{^;7NuFO`3$TV6o~pJ6AGoDAL2ed z7PEwZ`FG%~_c7fE$8d4wevW_-R$v`Mx$W7aqu1$azjdx<;=vo6s&VyXzZxN#8V1pB zc-%XjP560obPCP2{>CrCB3InogNp){C%}&zf5iQWtsH4K>zK;44|=NuwVjwQ^cUJx zp06i@IpTdv=IszDBdUQ9;y?IImA-Wbu6!{q6XVi)pL&Yr>E|R+@2$tt0cOFx2x7-V zu%3gNOW6I?j6Z@U~`m7GmKieh~Zy%VoBGjT8G>}lOd8Us`~ z!6*yDmFxwpCNKuAljAZKu znpdZ(-2PDW`bQa{rT}-z1MLMl9`2HuQacGA;!I(r$m#H9ZG4G0?#qf;wh27a09CC@ z!H8|^<&s{-H5(SvTm@*RSll-Dj2D&!inx0^9?PbJCDHtp6@5li1>(Uw#SJZC;>;bH z@3|yuecKxg=tJ8#*G%%=`b&llx~wvh*qvUDXZ$y>26|tHCy?+Kk{t_Y1P{HZI--$; z^?dhS^c)n=xszF;1!b#SJs2<#F9^eWoeodAwlc!RUE)-%cp$B}y%?p-1<%=nz}_Fj z3qkOG2)4;3<=1oyT1o45-?o?S258Syym4=9TG-&sqsx)eP-^v#|IZW5qXQMOgURq} z^9ZP6?hP0L9@rb+FTFV?trVq=Yq)Sq?3Sw%aWK8?PorbsdtYXIdw2lePGkI`O}H?} zf`BPp7uGZIGLJNYB;l(a;8Ug_lF;UPtHZWQvY_P+r2K3(!u$7b>)^3vHB=U0cSP2DpV%P4?5zCGfw89J1W-_=4CFCeVRJAD(CGr<|$Vvp6`G`B|A{#QeBOa8Mq+EYrsu-5Df%aDZG*0R8^dvA7 zW%ZZ~uDImhjN*cCY^kR({o^-Na8II5m3#cVl?%ONtt0c64ZZ0qaQv}9hMs%zyDwxk zN%JfD5)F#`osL~SQj}}Jo8Hm|>aW4r&gyU1pg06FpP9sComVxO$8*)vhu(;2ktNK= zp(+UvI&s0_to(f#-(Q+Ub51IM8fkU_mJUB4JBV$WNI;ceF%FtPoM<=tn->$|2Twvw zpBB)d)Izi8}`4 zekBWsK-YSx+KBsOsu*ZXi#A)+Z?%HW4u^T@et2mlU~3}a1$Hv$*$JY(b82KX*>n~ ze4)hY)~rc?s)g&X05TAganTaG4Z}qun7+jgjrYas%Kc;->h1ch)QPKvTmevLPVTFz z!NeHP7YrMi{R(NB07UYe z#Q2p2lEIR)!^o8JD%y>h2XnY3MfKb<=s`*yYOTK7zn5P4`<35mu{rG5->e&L@0Hy; zfsXdrT-vZ0(dglp$9OP&X)Am0-#J4c)0g)CfWEr19!{eg4CuzSgz`bskl0GYV#gtm z{`sw=#=@o^+SMkDdgDIKXV9jK&5dl_e7^;?1m?fgF?bACaODg1>#@}NSg(9p|FL!T zUQ6rwo@Cy__si~3gkSvx$6+l;m?o$5@lie}zml1Gd)^rCjikgo#WPhC+W`tk%TN^; zcLdqN%nSPhEIO|Y99NGs;aW02t~Yf&BjHmQo{R2yo9%C7g5fUdI$dA_!;DQwyI4HK zjN5Oo#kXbC8&4-@TKlq_X+J&kfBfx;nx_Wr5Q^G+YTckp6*l{fR^fpI&J(Vr{o}>sVH6hs>yCSBA zEH!o^a{6md==te}sKN#>X?`$KbnSY3X z*R3<4aP<>W;VXLEkZ(olP{-Y50pBL*qJ%BwH`^fj;9#kE`gT*DmUcyh8+1AlB2_6W#ywv@o ztv!T-2TW$0)`v&w*Ck6`@M{0fI*uh0Q?+%u?Rv|g?~}eq*GVA zrywL6{P>fZzlJu)hU|Ngdk;0Mb>$W@;n&G#NDA$9!LJ2U1!Y6>=_f5z#e;=gz{TDR zAaa(IuB?9s5k`haR=2V{VM3=34`aJ}$;?dzFfaFcIP7=C44X>u33A!9a1SE;k-{wG zDdei*V-O=ZRCN8wMRYw3Kb{TFXW!Ih`j2!+ji~+i~N2AAcK&~6XPDv|R@Zm237>>Zn{$6PSXzD>-g(Inq>E-+ia~_DJ zerdlbm>egi1O3z$lb-Be7zQn?-#0Slz%Ux~dVgN(v74=rJPmY#8^Xleyvf5w98_AX zs7-_*FnT6)L~zjGVzi^C2oP6!H++FNZr@3sCV;jYn*XI$;U(fgbFrv{1JHtWQl@fZ^156Pn)H6e&;o?y*ZfeA-p7Aw*X=cu3 zA2pf1oK~^A9Owra$9SP?(-aOBqc1lF>?lAM_9gQVMtj$LQ1L5?sf=lV4$g_)%hSQ7l{~#mE7*&=(E9wFY1HDjuwC)Lu-x*X$l45VNRrP!r&Z1 zd0r@5FMPSiq#S>J%!?nU56yV@B*SXtC~Cq&dZ^q`Oq%l^(mbEww(FcLnKRlkgmyDI zNPml8N|&+ABM@!fIV%<1-vVoj2|6W-wbz zj}4t>gr0Tm^7!4qgZ`A|h)AK%CU&F`qaV+dv%id%-gy#iK<&)6{vHf}8(Bxe zdloe^aTi31b1vhHFo_|FJTIfP_T8NA%lDts?x8L3N<2_rDnYy1N3OQFA|dF zpD4Xmfa^#+Ok8KJ^UKN&<@8!k4>cof)psvXqRXky&%m~4cIAdI6uMgIcxzRohsuMK z(LJ&&ou8Vq?`W#Gg{N#nPC_qDs{3_vx2{$dq2ck;VAzgmZ+^>GG$IgB&oT9!E<`(z zGM5{2N2OV@5rRc4-va5{MtlZsRRp3O>P)$hG&pO+^xRn{FG|0*Mb}B>2P>v)JB!P= zO{^IIQM+R`aA^cqha4h##r>Lk^5;u_cl7+590tLgGP0p~-?R%`Z)$Q0<)-ThBSM`F zYv-}ojL_|=<)@A_nBhjbSzi?wS;`Pg=m|(OuU)z{7?u{54@aU84$^)sQny<(M zmCI|PE^V*F1{?;^zPp!aV=oe(o_OyAZ=~&?CpBNppjag4e9K@Sy#}q!asxATpGqmh z=30{s5vLyLR2YlgNJ;SY=I>fOSKV*g{>cs{2yTlKx8@?37HA1`FvG9<*&3~Gwl`av z)Zs!jsbXrUf{26UUk*>ajIR9s$VK*X#%&?qQ))jg-g^4Jj1CCu^mGAc`R@HLgc^Np z7J(u{>5md=kh8`qHTSD|5O90Go)FuNEnKw=L0>Hs+`9GHI))N~eQMqig8nh3)qy@V zWLKgkJOybfjK~t&npHhpJtO|kyEswn$tAu~c2hKki$Z-_7|Tr*^cvJ+i>l74`O2ZC zeA>?EJ`JHYS5Vgky5^k0cQ+~MbaRdG#{?Es{qtOSuEO>ADN2Syt-vL%Si%JaZIY_J z^~=Qu)mN9w4R6AI{*^N>vLd4$B5MeB7ukhBwAI@#(T}=Q-7=WJ*33Gs204u1J3g1{!cSLy$Y#wJ#E>$AdiT}Nlq~y#m z<=yHPfv;Pfqq!?)ly6MoGUSDK>#gykxUwhKM)%~>-q+h0L*E@)l`j>d$5h1qRZgo$ z0TT;tlIRiTyF%?r>@KpX+i95_KGyXGa#K@u;C2%&w^9h8oIiC>Fq?J;qs-#uqA{&U z3tf6Gs3KQaLiwfd^!g6^(YR)sTSf!A5pNZY{+{Xj%+t90Z-ASkDmSY%*ShRCXkSqk zd12Op;?h9rvphv9)~=Ea#n*l1vne8+rp-a~ICOqRe9PBl*J36F-|j4X_UC-et8)F? z*9GejrU%*xpQs007w0vZ2I)Xm8KvwRryGnfqw|h^;#}9-a}}I3PW9ZS|E_VYvaU<* zT-CYRxM}4amL5}Km^s1-w<5M!^V`ww6=B=l*Wq!%f|`d1rvp9=bNV($%raaVO6&xp zvyT6)9w8J#)d%8S%K>-QCM;mGa0zEzx|hm4CO?ia{hNf?c;n-1HhOSyLs3(b)=9FE z=41@V8M;s%);AkEmNc9kxuQ*XUoSr@qd#|bkF8$W$VqQ2FZXQI8G4$p36eTjg_{LSac(pyU6AlYR+ORR)2t+$)DE|OfkSqGp>D1?LQCr zuSer`MOrcx{I)~V?`TarX`75F`t?4m9;v<{dVhwMP$e_>L?10}BYnG*5 zfiOXj58rMbFc>EmPm2^U4y6n&=rPpra^=Abh^CawDUYdm%n1n|;f4=Z7Aj&_YxeP@ zZhDU?np)pYbRgOEKJ98kDIs87uk9uL{aNz9qA2qibc@@wYebmD-5{ww^P9-H%f#lA zkhtT$Hk04&$B_rc?ETk*0Ob#&T+f}yqa=Rx%crEJr}?Vam;8wwFdlo!ocr3gHl^EL zkTPu87w%0BccdG8AA^d`Ua4r5D?}Pzus?8rWfsE%w1PPJh**2 zsFOP`N~XRrwl~rL?1j}OmEYo{&uS#Z^E1w#we9#2m9Kp=4l7a=h3^|<%HLFTWw@Xs z#Va=rj3n&&3^B9@%hs;vjdXmF{L_?O!`p*$ICaB$^nD>{X=Gs1T827W1r}QoLNn1$ zwUa|JDuZIyY9577FZuWdH!h_3205PW2qOIRw~v@PhedabPSgx@)IM0`_oRvrt=6vN-ZpaH=No?uktY&rH(tY4zbpj&(Ek5^Jcdq6|{+Q~B zGDZ>nDsgWNkA7FyLTlp84^t~>2H7R!|CB2x>KfrJ#{rhtXBBl$#;u>(-1E+O5!*bU zFJO=0sy2LcFTLiIi>#dvHN5uF{(8u2^D~c@i{=d_4`lLF$_S zP00CKK}zm-r$~Gla*OG$EKC_2ZsZIT9^U8ot(9@2DC)l|gEav!sW^%(ot>~q9~JSu zRsMkpY%FM4YB!W{q_ic9IMH6UdsREow$+SIMO}>r&iD3n|3SZZooZ9gH~ScQX`{k) z8{pBoDyjjk_x-l8dtSc@(jIHKY_rd^mK!oM8uO(SbCtmaM?LZU|CqzJ8_6gEKFYd!i@hx4>|-81+%aA*mk+WJg6Lsad#x_$w{ zTZyiYQe@OSGlJS}T{6l9;;dQC%~G1%Ijv?vEu&XEzoeS*0 zfo|68t*Gx$7^HJ)xEX=hHend)xCKP)yA$M2L7nsz(_%5bZ>`F=bi6K>6M7EeIx>ii zM6{R;P~~MMs)Q}&M*y~a8?#k-g95M9c5nCKp3^ zlyZ;-T85uf%Z?*lh&xrYie}KzJ%7>av~z~sik&(do?YoSP)LEvm%3v47`}qz4g1df z6-JDOX!3^X<@($j-lFZerx^-eypKgDp&Re4U!JPIDpu_G&V)d0qy_!fv~nBXOj_^= zWP`hoOdcxLMg=gZVPDl?i<}W*WSPE9e5Av*=y|Fox{vlWR3N9hvV4?l;%5E#J?A^L zRn$c0C2=R(OclJ&H~_8+!!2ZSt62YI)<#S*IS8Z$nRprhpQavY_*dPJ9^g6 zO>Fy^dLH8m3(|vwZ|}s>GpY;b3kfyi`xutMcG0n-qWzP5*b|MJ({+?prCzcMVz7Ih^P zL;`fN93KXRa{CCQ-#WgPhtfo1KeLXAC;7-{Di2&n)v5{-=eELsKBKf{+&^1C_7-|1 zC?P<^WQTRKoTjPbp^PUCNwrcM>_ZcX_du9knu1Biwh2Vx z|J*O}!zGe&mw0Xj%DXf){ywjV?yHR;<8ZI^GPHj{N?6W0%vn$!l2yLmbj3LXn_lyI z;*~eGpdqos_rJF@BNR5|f~vI-h*D;*k6pv7J}6k(y?1TYk-`2t@J)d z@uP1poPRC3^E-h^t5R?rWl6BJtZ{bKVF| zkrzDW;)$u=u4xwNK3?d*lpUf4)dFewyE*rGu}thW(=*}p7HD4N?rk%n5f_I)AX zznPen!5|`Vc}G^#N&lSh;aWM{zR>Nb8-x=e@)o0w=UZr3GER(C@S5I36rJ6seUVl| z7JY||V)rT^0KM~qfHC47{%K>E1~FflM$)!YJ0UHiDu~a0feu_HeRu9%js<@In|Hs9 zZhYU_ZH;ccqcrRJV^ewW3+u#>j`^2ignz8skE^%ec~FX!jhyG_4JGl8Y$V+9MMZ-< zP_YG0M=uEPyO;9kt@C_uk2w%x!jC~is_d}&O{sB=bcDbog+~0-KTSm9@ub(@-kNTN z&23m;6&%mP`NDH(76&n+nKfQYm*HR6ScNFZXi=Agg+TUsO!DQ26rqr0yzi48tYA%i zDb#46Cr$(;ljp_qF+2epwa7$pfN>Av%a4`L;m&w8|`tG zk_qI*qss^NMeold*2y^Qo@)kVRDqSXv{pXrUcl0`6lJ5Ya>O_ZbmorUJHkX&Wk3f9 zi0m{EoLF+X7^O6^G$*vI$uqI^rM*v_?DD$$sT2QmwvAFU%H57l13xrRGifmaw*V=q0ueF`tPl*3AYUYttpFz=akCI z2SW*ltmnYLYvX8@RT1V%QvnJ_1wFz?a-cMwKDjVkb8IsB?iL0Ps{BAHB1TMqcL=6g z@##D}1n<*;*Wuv0hpWA0@ivm3asN6k#q&~PVreUS@$huAEi1W9p+5!xpuQvk5z>dA zL6q&!VS;+}mGg_>bd|gcTm`php@mH4D<)~(&LB=kkkLt>J8t`WS|Rkwm@FR)e4vh_ z7a`5sbCm-wp3J@%zCtW^cdKYzJhw|eML_{~h&XTe3{wgQ@eT5}J}M*h0DsVnO%?31 zk?#n46jLEYk3Jg_mwd)=8+aN7OMuP<#nJE3xM6c&i4j8U7$RbNA)vV^rK%_nlW6#cHtzpXJ`Dgr*A4 zLS}EKG=6}Uk(~)u-Q%Oz>ESrYSXRFF+#dF)@x1Rkg5ERZ#Wg&ZpyzozDRB*W^CwUg z`|Jj~BHKmwp%bl~S#9q)I)6Hnir784Z!IZ|etb0TV{J@6%xdtxplfZ#_XnMyvi3@s z>fqAUIAHd3iAM)qussTj`;(OYX6eho^*WDJ1C(Ve96)Md>J>$0y`9OrU!3^mIJzWlZNOwcswc9u#wtO;i>pX|H1O@7AjF z&E>i#*~y1D)l%MzqGi66!cYk{M5_?A{4wMaY&wKuqog9})6I%>r2Rv3nI)RdIk;-G z+$3S)F8)XGgTCZsmc!?1Ew+;bu%uR6S|BcV``b4+WXIF+!FlER`{z~UU)l6AzY~@v@f{z zbOth$E&kSeG9Rt-w0u`|h@s|Mj=?zB3?lX8T{`MGzQE5vwW>QV!}{bG_=>7wc+%B1 zJ(1Zd$tN}a9*jI7ziMn^h8%RI^zYQxcevjsAHv?fm|W-{3N~>{tCmCw(hOsbBs0C(1g61(w_}T8kQv z5>81`UbSMU9$#h9$3x-jY(q5-eWve`tadH&v9x;lpD$ zw%MgYs0R8z@$5q*3;=&)(}6P2=`a8S)U$U2XPZUD*&$pvT*n&8I}2*c<>$^ zzit2J!zi@od4{s=R}>F>ab})Ny7+}_wy(*8kKjDD+ekr%+ABH(5f4hK0BH{tna60V zFkohb2B4nnttxV`phDw8oofiBcFP}M(=KSC_EjmZEI}>@sg)}?FUtI}#>~>(q6Iqg zRF`FL;$#AM884!p0DgQ^ye0Im37@zF$vnmn?K!KtUxSX;H=Ojb0Z#%Ez1b0a>%F?$ z80LhU1p;Ix?WEtECrv%*&*>N}vGDk!!t@C!;aL_%sP|sA5iY((kk3 zd0gEC^|Wgn-@jJTR1D_p!%jG&lao4b2D?21rfNafUkrU5pAH*y%!H`~BUJw}8bUAZ z=km6ng*q49;W+kRp+V0b`~tOP&ZEGv#;DzI1*>Pfx>*eU@LD_|kM6CHyIOGtY~!*` zX@(hYNBhusAj%%qIGPELnQQqvpmor{QJe|- zmF@oclnymVFwH+lIgvVk|KES=QRyq#K6y<^o{1Zqe~X3!)m+MUmuD5dp!RH24qqp# z@#~y@ZB>hR*2yV{GP!<+?)D9YhQhsVWJzs^x~E zz>9)0wSAb-*v*$UAavJn18bodtZX)v70yx8u6+s;|1^XMM=77D7x}wA9#HS5OB(Wm(5&Mg)!x}6}`wf7q)>FNoM1`ldgAs^yIxd z?_*ezH%Br}==Ct0HIq7OPg^4L@tgepaf>yYsGCDsWd~w>L2>b*K2LtgT=3zI1*Ls~x&r zanhwskrf~N;xlLzvVvOc)VVyhD^>Bk(B?+4FSO`>MdmJG6Mcws#Ut4RZk z6kMh<)KTED%QE#)ogBv4$1-+YmYJASscYsG9&(>kiN>#F9EbU9$)akB92JV0Iabe_caQ##ZnN*<)9Zw5ckL z&ABTy)lf`Isk$I>=mMfvsIkoAWXE@{{5ZUJ9MJ!t|kIj9(}o5{~5fJF7jBJsx&RTH$EWrH-@M0or*?@ zNYAD>vmfQKE1uk@oV%3WA4Pb<^#h_Kafz4T%WBhznn?i_; z*BxT>b2O@BcR6k7CbxPwWRns5b<67obn7R(Ip7<>Ca)2Z8_Q8^c{6{dYPpg1RUjzD)lI?33(k^C@n#$p4F0pyVYE&qjzsP#hFlj!8Dfei{XJHVPl)wTt zWl|{3HRo!N6BkY%BJyqJRNjpJ&R(%M#chZ{pd!-%3iSuol41%NaxTfrzj@(p)jr_* zTFDe@mLn3sU))Dj47ybAhAd61<_C}}RL2=P`ZT?$ zBA*OwGjgz5wNE}ecH|()qxYGzK?h1r<@di2R^v8W+Vf0Jg~(D`_s-t|_S_9lruBcZ zMzJuzlG_>!K9jS^N%fY_hqo*e5sUOLxeG}|Oc)ix^+?Q*mHNdV;LC%HHHT=O_CVMFAROosPvd_7Fa^4TlsGj~O}K>$ z`MNiHvlY9T$+|w=U(7oe&s);b3jBeE4Vh>_zy9ot-#>5TNRl6$mCT{VKNh*#-w%X0 z%Vo=M5q*cK+_L%ee_=4llGaLh03}=rbQ)w45+Ubl-c2QPwqj#N4w;*cXD|kK7wFfp z|FA@h$hi2bwksaX?O0YKJGQMuhqV+#Amg&jJe9ifI1i!j;YQkHuy%KO19PY$XIp%Z7pNjZO&9wK+vEYKvY$xjz&1 z>g?baiDeag<=RR0hqFS1s7{9AH{dodUVmIrJblIR<~Ql}q&<{ge6d7!Xx-$l22JB7 zN|Nne+zC@y<~$jvv?g{9yy{3_a}mbWdPgYZSBC8CCNrYWh&qLjCPP<5Fml@pRcZGdLYkUj%_>eIIJ1Di09SSWAy6^=NTA_F?QvT`j2Y;kX8$jyYT#>= zTZ6+}-|l&=WyP1ZnuQVs{=?HeUgC6yRpb*`HchEPJCTW^xbyUq>MVSThUq3&YM*gi z$YMV{|HBKS(7zu_L!I!NA}Z^6ov4Fpt9I~&3id-hVe=>9kR`MV>ff+9a%s5FZ|QFI}(&6H;>l89ElJs_w5w@@5nW?Eiv3E`A$nwj9iIe&blUz+~ z!V>%8aCt#RXW}2@tKG%vr2*8kf&YDjj;1eO8q}b&mh?;1RN1@yjZV1X`EGB1@_NpL zkobcM`%8mpM!z98eRxYScH=L=nkTM!MXePwZC>Uv>q*FtSb)A%b$1(i5z)Yd{doW8 z!i#eMRdPs+_6@z0&@YdIha!90)mn91>xjO?i=d)?gzfi%(47)9NP-{w;HpWMpH2Sd zdY*FF@$$m#_B8(+8<)`#z~mEqg22k1p)Ad=g|>zRq1zo>ANBJEUcK=cg5E*~@FF07 z7vGWG)1LnM-t;0#YWSEdwqI;ByetTOExBd8>t5E1U=kZ90QEzi8gtlj$n7$>G>Bku@}jCgn)`=ByJ6R$@8{dj?P!8d7eN|HD>LQZP&=F3@c9PCw74*s1QqoLsZ1is2*|dp)8*_9LTR-DAK=MjnBGJf7Cuez2|jHhMN(Q zvT!+!3`1p~AH#Cg@~5ECC+yg#GZ@N5amhQ2LNYF`N&oI#rVq7~71Sl&>gR=(W2;0` zA-B@1{|mS2Mu>s)Suz+txC5uyD%REMEA9@z&a(3SV_X>>Cm zCmIWkjVpVBt69}e*RX02jZkF<>F&V?9s570Gad7&SL0{>Eu;Spvb@h zSkng4d15dQA|@HbC&SwHiyTlWjVO$R>pr4YVt6Q4&glJ?qu31hALqc5LQbh)7DaDAx|pc@{`qqKv8N~mGR{Kj8hDgf*$I%w<-foA*o;74=~55YwtNTt*t)i&gaYNJi)Sz*YPlgW z7U!?pji_s&3_ntmB!!GYM_Su_A>wqVrY500@_I9jqwCE`|lJk&$oF}20 zzwsBds+K(k=(L%@Cl}~NYV>YCs@$;5)tRnxutjAf$XVmrnaX@HtI_&11SLr>>DDz zFubD_MV;+MwlRKdwEU}-cNT1VTKHX0{q_q^f0?I4Rc|WXv7Br8SgT0)!JjzPX-kTA(iq!quH>+jX=6e*!s&DCh@?CyMQ-!q zHZm~Q#0eJ#cwCywFI5f}eRgIz6V-J7rQu4YT<}cRmm?{Q8}00{5R!RTs2etugyZmz z7GK16l9qiCu)5@2(CBOTZdki6_&$9R^zEL`{SMU;fF%lUFCr$=QUzDaQm-xNYZGw> zP8=2+);KgQ!t^2V*_A%gp62V!haDX)8<2z`WD+G1NIk{+RN=e&E;dvC;}+Pd%h=~2O1 z%!s6wQ{K5|#0!*qtj^we4rPfvJ564Y@8gIcnb#I`)EhNA`;kz_)+^op0_VXZ4y#s6 z52gaWD1*ImO8dz%K)IV%@AA#DYc6;{lHES1mVVerCV7r`&2^TfT9F02nM<%VJllIb z5au483JfI$#go`kb4of#TTM2*d#0pQMWemf@PC~82_c4och>KPCUVhN@OGgdI1Qw~ zgc-&Eh60J->YgW`q7IV$kL17D2612SFn4vMh&5A|SCg)U*92d`wsCvyooK}f zpqG>u3$&`xQ3oAjB9<8Drr-aN%+Kk6=3?#osG2b;!?xP$sh|M&YzNDbQ@DkPipb$S z%(mCb0{b&u_WhVwpWza>JoKdEns3v><93A8v(3H!Zw-^1Cu)8H9eSlNox<%RpuQ)$ zFi#js>P=|=m5YU5w^XQm36o3dgrMnf?Qb6h1Fc<)B@jzUNFFn&2-PI=#sfn}HHWA6 zu)qJx&=Zd3kWeu^sbmZ4z9%DX$?ISY5Qgk9d_j2@&($eYf)L#`<(K$M#!X3NUQk^} zJOmZ%NwrXN&FkM3xCZ*|>_lrgu#N!MB-^BBFgvkLprX&hw*=T8{6Y-8CL{p$oq~xj zRGoYGYymjTM=c``B_yAq4u)j%&PRYNvDO}0>@!#?((>0s&r#qPP*wt;R6!{xJq93? z&)U}?5aO}KYW43d9&}LSpqc|!>LlpTY;`x`H|kFxcN?%&Yv_@MOF`tXA?&-9K5oPO zWCOePYluYbK)&$-MWzfSLzdKp-_vKol9jl#Ag@L)c538t-Mc_C3+Qioa;>cR)nw{H zt5}T6S?FNp*_Q87t^L{Qa}+|25aS-EJ0EL7T2>U)hKpqR0`JMV-u$l2ejqOR7Foxdsc{rb_tO zg8q>ju7gL+4Aswd#EwiEX*lx1wiDma7=^9GX-MB=dkd%~mrWnWkm%o%Fw~iQhH+DXFFfw}w() zUb!#x=PbmyFBF3BT!W9E>Y1+6rC)!Zf4G~h(hPkaj4k|vp?tj};|f#8<>+R-q*xa) zQN7zt>|PnwUj6o`DQ(#teEy*PVFBP^4n3Lq&O*Ul@woj_8;UODq$;7>*iVty$<@bi z<6);{@gt0}YB960LJ#no_-Kv=K5}st+#1Tl@m}jI#+ELXKo>aKa>s;tz50I(!-unW z_+k-6j6NdKi8&?@p@u)7X0ZmXsM^c&-`XQ|`~zdg1=U0<%>-27@YFPGQRW(-&b>YD z7`w)O3=QRNk?`ixXEou}l{gY#;GJlUC50g3d&deuOK2REGJG}%Wirf*_b(NKr7wj}ELpn#K^vZzNMs65 z9uxX2qf|y_N6}KffuGNn(^dd;^a+m>s)jBk6sR?4WkelW9v_CtipcNpXA*mB#{UurFDWy$2v8tNZ<{I{Uil|YP zB?aSc`ua@uP99cJq>`DO=kz9$ysMOmbz^&b8f-iK70d4ZnZWEC%;`7z`CO%Thn?mm?C3e6Dw2Ydb~5;cT@_{ zX{p#44^mE0C0h{c3+Z|=J<3f6%L7E023s z(&ShKTJq!vH|4A&80736VejU{W#vmbT>9N13Mad9^XrJVX?-{jxt(#-7m9`@)Mu0N zSEL|SZ_f`<`MW4t$}U_c))85e-cj+Z@ZT-?k=vBFjccNRaIpNteF>r>@zAU{E` zXYdeGccc?s<-5yuwSvSRo{%BLK?y%(@5#b;k3XUTXJ1t86oj2q08Dp*d)6yMLbeps zNAKC;Q%(AEVj^5^9%e4)TtnO{PEw`6rLZl;mMebke)uCgSavxCT028TO1OfpX5D)O zp#R-z@>*E}Hk*hPP*8bn2ZNek>79ht_S?Hf(J^E%+tt?Sx$e48I=AvlES zApmjV_oxmu?5uC=oP~+()A6vs_OcH?_W&44#PsSHLKj>$F7IKTZ>vQ9C!UMg=KD-S zwB?A~P}1fHEv{gQt#XrOwt`%shL z)%qQ7wow(By+Wb(DB?SE6zGWg@jP#G2AcT!?=dCs2*G-$32Z|Yy}g`?xIJ~Rtr<#t z@OyS0!2%+;Fq8%Py}5pU5&}_twjKEns_e=@fTD3?)Uu}fUG9@4oS9LYkqM-_B?atC zUVmi;^w&we_X5qTKiJvAfE>yN1!NxNt@F3H#PrCNzto=8jUlsof4*r8?joec6HVd~ z=09Fts&9r?8(Xa*)shWwT3At>wk86+k^Jb;p%-KInS^t|%he1@uaseZ69%>5u0*)S z5v>3G?&2GrhbdUMWNFDLxY6kYu zcS+GSq7;WLA=O`T(A1V*H4JNL+E(3F$?PA5H0X_w*~5)Y9>dwH_TAr#;YKPmFwd7GopZw`ngz2*2vfFL4Ve!F`q-|{rz{Z@Jk_3I*#syv1)1Y>y|81Q| zDBeqXQ?ZDAS8oPB=R94ie2;gJLZpF(ZI?HIG@lB}A!P)(K~%5>?7T+ZJ3U3mT%zubd5m5nVe$lVHsCVa3% zDEWUYYk=J@;u~k|lHIP5WX1vJy9pjOfFeYjz2y2tmf!7<5QQ-nzlnoDg5kW7Z1r-= zJtB^p^k!EjAM)1Y{CrUqa8}o6?03AeZ@Q~!FW=cvRDZ3T$MVO+8Zz<~96JF~FQ z8}f6!;$Rz2#-VquDF-5}?_>F1T|sUIznK1EB{?sIqAKCW#9&O-l6TA&Jkf?AQgx^f zUsd2h|BC`I+-V*1v$O3dRsV^xuuoB;VA@X|Whf6(9Q-d+El)scMMa|G*q7rFHpZiY*=dj#bPAh;6bSrEi z5i$B;wM@Y5JL674uBh%Vkp6ld(Qxj&IT6SGJ*;Q?r6=(E-I8J<0U4e3$>8=8^U(rw z^RaX+WqWG4h-f-+G+&HPuBkg*lUNGl)LI!yD+au_M^TC|X&|YrC(*S~^#_{V1!T!e z_d0e4bK9R+WR21vWDH{@+Ifb=TY)lWH(2%FB~BPGGyr^gOX`$|jsh7aG$Hf(MI96J zdi{PlRQd})1g$l5EsGH`KD)|ac2u+{-h7d}K9q=X$q#+L0IPM+(6WTW4c6Ksbt0zT z=OK5XX7o4%=oh_>U4qOaa_vaCkxllJ_e5arD)>EN2@3cpYyr#XY&elb ziu=?Qb0T(T9QabCa08cy$eIMgUZz7Y^OYKTVD_dP{iaN$PrDHPqcBgk^x5P}n*58SEj| z>%T~_WF!g>+?JOadQf$q(Vg2XWxc;B;Z%pafE-zVd)ky*R?|+I!+fG(zLAl&fa3cz zP#SyV5i{=M^hbz{ql^Pf5GR!O#WTGzkrkpFz#f`EZ?}R*lT+yzp#OqY4@?Wjpd%ht@j=kAV-yqiq2Yee+$@>5 z=6E(EO#N12#qLO4vThiOT?*TL2!o^at*HyHx6h`eLNPGv%z|D4Lykj$o;N5Cfz(9C zGt}c?(gdieznm1k%|JQ)NiM)pw&5K7A<4m`F<}y5pe&E6u~80hT!M~4#X@N0u_GSQ*pF2_oggUBTq9l*4BX1vQwb(K1M2Fr+~7C4GsbX0jWfjAYf_y*&lex z=I({vf_Z?y{@88>Y$sYYe!$z6@g(^7T2v=&T2GCW;X3O+@4*R(Gy=HDeo6ih;Wo#A z2E@s$!PuNxR<-xA>Nl5`6=aL~4>_Wr!FvEx0-M@5Esg+_KY!1kfF|ng@0!3*WMdz2 zgtLNfKYJ?ax6}E|uLYKusdQ4tw~`91qyJW4NM@i6`(A@dq7}ch{MUo;L@ifq?-V@W z#?w+vu{Ae7F~Heh|9>|$A4`Si;qRyzo(f+Vwr|#W#{wKV4}*Zev2+1}NvVRXK>zMi zoQiYyKkdc1I6=R6oixzVlD%LjBUE^beH*IZ;$%)S1rb}oZcXg#3!z|BOf593k&_%} z1o7hj!+Npr9Ju_}raHiYf|{A;YtPQmR(cQz5ANHI?AXdbedUzo)@G<8-#=lx3A zeq{?`E(Xk<0e9VQH;DmHZ2GgZDDS3^!91sg?KoFh`85f(9{)sqZJ|Oo6UPlzwkQ% zhQUtxqd)I&6(8{}xCS1vhbbpwRjbu7x|6iCh=~Y?_e9L{n99ny>r_2ADg2fE=07PQ zAn+MLl6s`2>UWGeuZ@+RxQ;W;<<> z*;^>u1JpID7eW)>pW=900hb%?_wdlNI-^_&SPb47{jmkBrPABFXWgS}n{y!io~sYq zi~W;SxKS#18hXNv&S0P(pGJYZ3M<%qw(md3__shZ=a+YuTm?V1gg;w=LtKadAT)N; zwIcdStLhhlAh>wxvKwhBz?_rq?*iSD0FM5V)ff zCllSPMn?5JJb1l{nsr-RM>Kv!+WW&mt1{)XMoM_u<*O?1FXrx{y6N|){gLLOGgy_S z3qtc+kay*)9O~he>A9Rws{v`tkwWW*;h~wSq;`|jk z_*0nnxiNs9vJ77}Uh?i>`h{|ey>2Ps`Gf!X^5l;L^ zPaV(w(dZupDL&O={|T#^zlOebbf;lw;-Cx4smuI3+Gk9@e;{>gsv=&`d?)xVdLmC4%#tij%%!)aqIUS4d&5Kc9{3}c% zajQI%xrgc-qxSJv^cW*W&`kI{4E76}A%hCU*eRYK<2Luhi(whxqWHlp;npv4jM>z8rCV_F`es^=5$ zJnGRWjsm?{XpH6Ec6b>6ludd(T60lgDHm5}y+eDRe~PKqe8W>%I^BrjYT&ex-hCsY z7k^i_qXCUf$cTjJO|unm#`G{}+V<;~pG2LbytN+RaZ;dG2mMU8Uz}qS(`k9jxwJZI zVHKjyz-yLge7CdZ7JO4T4`ui<{)!Jg-iQw}?Bzv*craUsF_B0>Hd$&c1a%xU&V#}?TNCR(;8NMeH zko+rFCv_^G8`!9966+4ASH^E#j7a??7=K-t_m`gJhYNpZye{ZfaSFYb3;Fgxj;=eN z>hJ&4)wSNTx!1fRB5@neKll8dGS>TU^8z~(KFO?_ zpX!1Hss~7kv1?@=X+7G&q>!4LvsM1BuUwtd>gQH%<#__tux1S3wHSs=Hn?1Cp*2q!S?UQ3>o4mxIaeSK!mg!{S>QD-|;H=(Wi)j>`Jm}I$@vO@jb zQ@SrAWZDS!#Gw1o@7?pLx$CF+cbi5pALjiPZ?q7r3Q;B`lyVN;SBFK^iZjdC`WbOI zhtloS(7Hps4ld?<0s&b3_OV1u$Nh|T3kbmqO0`zqD@ z*grEYb;3r#nq=7&)Vpy1mkWQyB}A69uMGq{W~Y;X@AMdZT0$gR)E&D&h(V_B^UZ&3 zWPf}uTAy>Ezvg#I`JA;8UT@teo&Ku!jOH%kYqnh49xC(-yBp;qwDb1MMNx-Y=!RjX%~WgqkI0UGQ4Wp8yua>JW9P!7?Pn zToJdQsyf@c9&KRHUHFTr*TV!c=D$Z9k40`|bNt=C+F$ZyPq8GWoWYWYTV0AgfFUFR z=^Bry^uIS!^hLaDRNs<%n& zcm6s%=8naMWUKw|Pr#YN$7LN2FEFgnD=ww#^}a&<;Y={R;I5ApikWcvdSU}d5&`^# z^v~xEmiVEmLSi2daQhFbh%3%Cgj*d;kFz@p^_RTtW$IYd$O_7!^dL%W%@tD}1z z{9$h6ks`0|k%PEJli)n;6~2yO5cj4LpT1{%W_bfZVofMz;|km|>l%3_6JOS`2eYN* z+B>GdD2aD%jLNV2-I^xri1RD?`$fA~2u+n7CiC|1CPEnhCUj2S3hhJ%0;qfzX+lXs zi!B$(l+x}UEpO_g7Q|3DR&yqbFTC(117SEnwtlg!Be|*KkK9c4ocj0?Rst6A%9ny+ zJ0@6X5EcIStW$xI;6xa6)EwxcFq@+gwt;#l72#4259k)<+7n6_4$hQ`=O?Ye8P?~< z6$_u3%LMjfWubf?gJV%h&`))wYp~ITOci#0*le_sOrA0pS#8GY> zo#p!ZcgoU}d#2KRCU*7uhP)I=qL@u33I+;6MY$Yc&;>t)@$BbMgip5B)8s>AY-UKo zt^V11%#q`8>#>PCcj2Rzd$pP{_sEYTkqoHs51PmCblZ z{Sr&puqTw;oT0c$W?-_NbgY@m&*AaIujs))>szXQdC>uvj%NuB8-<=yJb!htlA#+b zyk~;{q`mN6e{ON;SPH%*}b?6 zPa|9x3$g(oQ|)c6Ko(uh43!^BH(u5GDm!!`O%u64yGT_9UKgHa{|R5VrlXcikIK1> zxdS}>j~nXEb@2{Uzmk!K+R(i#^w~ZyoOA8xqbjZyFv!@*GW3@CCC8hBN{3oWh4PBO zb>M$KFt`T9KKd#XSXO@flYbXHE^2I{lyal4W69cTR{fTtx1s+G3TMfvo52N2R}B=y zwVpDKMSeh7Uc~4-i-y7<@A}{~yWlo`>eA61q)CtlCzT$>58nC%{WTE7lvU~_eJjWi zI2ko+4j^Pce@`j&X%$O2a7eu^dUL5bD76jmAWlIxh+`P(!ROl5QMOFs8NXeckdNe- zey7Rp8C**Toen>p8DD#~Yl~1m5+Y;x-JPM)$&l-SmwZPIROY2*m z>W01)57AFpNQb-*$%sGLllPaWdJ?Cp$mlBJnKnAG2Jsg)-aL=+w`2+iuJAuwzAjU> zV0+thh8epb5^hZR->;`z{7M3!Dc0(9S8@NpjnD8J(+ii6 z4EfRDugFw|wfvd$RM;C85K2hvFCh}`6&4WwpRK!1FG(1^F}s)(rNZ2$y6np;!`$_= z#N#uXX%Jp1Vs;;D6j4;cV3{%Wmtz)9FPqCQZ!73m(g;J$K?aGR#Mkl8_(J`41yL+5 z!_3Wf;{}8vJB*SmS;ObTEQ&wNrRB-eUu)2Siedeiy|^k%SGUn9g8t{;`=~*p7wp({ zyYfO$;Kn56f_Jm}V&c-JAMj{~n^DdmVIQK-hL8K(d*C*%i`FB&wvJMffOp}TqZnqE znQJKiwqZ2G$UlaV{C2npDcEX#jOalbu+3sad$&nc1O< zE$fPUKmi%s$+pPl!y(O8UqQ4n*>#bmxsC8Id-}imw)IZ!Nd^k9{668K)LV_uf*Yq$ z;L{KllJB&NW?*B0U#pO3I3+>g0(sc-UY8d;hZ#r33J}ueLdX?A2Ylf zQceX*tH&g~rsXoW6U_sd*vZuj$0zci~K---R5$ z3Auv59L6X}&!CuW98s?gkEo90=Nh6z+6&R|H{<**Vy@v(#ZGc>A&n2)-yrhkot*0{UwTHd6eg~x45#yyR?xnb7@$3D{oTWcTKMu@c7C-<{||yjhqp7ty%%%eBG+&GsWF@xW+WDGv0f(6C2tWSX+y~ zdv_XFla-8Y*kOJ2;J`D3tt@$>e6te`cdd)#;Mzl!VJVAyJK;)&oTB!HNl#-K4t(S% zh$0i*S~f-zrwNzw&Lo(=d3k}?RKsNc2Y9~32A80JOd9K{z<7KqslP;9|6wpdm6vpw z7DdlM9AcbcIc%hSPP4J0_u;nDw~Yj!|9<*gZhsW&PN077m&TAE_4=mkA`6}!MQG0F zS&4-YB^_#ncN3gAYgArflobLHz2sx9F4c1>er_J}H7hW0JU{1|DfQOb=y`;1+N*e% z*Zu9M>^pd?g#Fa>r>{M4Sb=C~(_c}*OIL|2c=J@uMR`STylC_^U%JN6!$^%mc->U} zxw4J^9EE>cZ)PnN!E&`g^d8IRVbzVMDDC^aN`8U;W&5cb)aPT7eM3vutg?%984b$> zh690Tnz@RfK%EDd1d{-)r55 zA!hSuvgEfD@=^>RXJlo1;7#wR;dKroNLx?o>#~DtPApmmmhj=LZ#CB4s++l#ZZ8!V ztGzd8)m8C8;pyC3lF8G{XW>VhpwE;vdVf?3ZuAv5$dW>0oWk{&xtYShOW$Y0H@*=b zzW_J4aKcZ1)9~7H#Mh5=!_Y?dqUjpPf?SE||Dbb1;F8AkDBcK;Wq@}1Rzg^5e+d`!-Qi+5X{U?S4_$ltsJ9R`z3HnL7>Ic%F7&!E zUEmfgB(m$%w-O1rTSWY+IU9}Tlu&a1XmwtBkF5!ki92miWf*aOW_TIKpY6$X=D_`i zv2L}Fa_YGF9WHA`2gV2aJ9n$gj@X0aq~d>f>_aFey%jnSPGm{O1 zwg*~v8@sSKrfB;MLLmITp;jB_mxfCc`S5|9x&4kfSR$(VrCm#u_XaUbEWSg=6IlZ3 z5_+d8QR}ap%%Bc$VPcJEP~hii%FAaCIghSJR{2q6O26wb(Y(sbD9Ugy6`_utIgG@e z?Tz~weLgPMP?cyQ`*63rHTkC1Pvb&7yVZ|eX7JR{-Au2`n(&N8f-_TNS};D$Y=#6} zmRZ<;d=S3ZBcaV6W$+wizajaNB&-g%&nu-GdoqP1Zdxs9D8qM-^;3bDEVWXf$di=p z-Ez>J9aSsyL)dI%!;iM9c&^Ag_L`wCBrv?|#WuMvwzBIeeu3-lD9o&jm=nLT1b|%ebbU;JK7qF+=_8ADo-$KHD~S8jSZ4>J|a4!2JOHP~gg~ zJ&OBt-3Q;04D*SyhdEKZw=^!-3&DK^W^XKlHH?=#2Q;T~@qZur2Eu1g{K7q|zWh+! z{8ZOX~)VmBs31AG;M>meV{E{bYqbYfo-Dc9a>2;W(bxig&wywp{ct5Jw& za$eO2eOwilxL2LMzdjcSp9r%>;7GZq@l#als# zdj&7a{JM${wJXW<^Qd}-Z*hp=sYeNKP-7ilsx_bS}K7YoI6Ud%an z_2R#lnMFInvw0Js+uZ8?XhszE_b3M{P@7et>s~>B$vbCte3=Dxno>&ikCh$et%Z8P za5Q-s3Cn*)K1?v&9Lb0(u!Voksp7XaON5j9#!U?`;FCL@EdkNdHQ^*e(@=oOZ7&`^ zq-T2<5_Pb3ccs{U{%q8{@ztYKENQLpwg5~dpCV0Ac2OoA)xmjZV$AOub|`fSD&rf( zR)OK?$|i5vrf`2Aqu!y8iyzD9-M5Xl)8zKw;F+O!4V!n)O`1ZHbq6#T(` z_d=$X9VRGn<$iS#rrZ7`(#tb*P95aLtO01N*n5U~eCVJAJzzNPr$8cB)449>P&3p8 zdFJ3SdzEtZil4@-UW1pQ+XBlo>9%ptPuqJ3xEdC`PRAv0DbmKfWOE_-t9qG=y;&6X z8!%1c(^jIzI`s`A zhkXA+%kcfHMsTu%vwARaR?jgjQ>xjg>VZ3r*1M@pK|ZWS*qk|3^#8p07*a!+e`n4z zFMiY(fmD}e6@;i+_K2oa%Bzr7;5j}N4jm^E!K3!bDO?O#?i={beu^sTz8^c@K7xCE z<#sU&wrdg1X{N1%g10`)zPYvlZojz11|?tRC5p+M-(|t?hN^p!mVkEg4k^-IX>-a3 zO1>prHUYj}X#5-Ewm6t$x{up--TM#UNKGmxxBM+>6YEVvw{n`3!g{N609)YbIMqd!Nzj=cUF zN`4fvAZGfdUz+uS10?(L_V@1onJfe{`*S+CXqkE?1*c}&8Hkft%m=S`>Hd>10oiZE zkJDKQM!>i1!O;E^gir!ds4;Hu57Ch)Y4wKmV*PG{+Oi@UmuC$D&t|&kE_IR$SN!ll?gR%1 zeZwgSG>pH#jD86ok6!4B9MJ%MMk@zDO9dc_2&&#uGnqg?d|~w2Xm2Tm^#iRc*_)VQ zXY2fiUR5sYeYZECQiCyZp)TrrrLMv|RK0CsX|g$ylHnE0>yV{S5G^#tj`d|0vGWJ|H|5TI zovxfi*u=Ti>T_?Bi%4ZL#I|=e0XExN4aH&+W2(jI_eh8-X2S*| z;10HkvaGwrLnwhk#y1e*bDAFYaPcR-ykh+=F@d65Abi!^{9LNJ6U$*}CLh^F{fz2e zG{fRXa!rfO-bmW^AIw5SmX9qDn(4Dlpe{AEg>ZFZBTAi7vk~s6WmTG< zB%m#2wqlF;8t}hA6sU2{}+M zGCZU$sBN6&(e0T1X4Rg^mTI@Q7Tm)Q_HpBnmDDe7W7!d_d$8HoRkv8&tvN^-h7jgb zCv$pznLvJJ+q}LS;WgKNc^JQX`t29*^D_t6^8NUznO2HAoo#G#yf-(bf)Z8w-39fs z9lkNeIM~cY^l&=~fyQXrV?JyW{>ime=Vxm2fIHs!{ex8UE5l5huo=x+_~{MqMtdsm zLeF0AN=N)*+QdDN`i{`j&CX}d%R_eP+rp*f`_&@jlazjevj^^fSn)l?4b!zFZv(5$q--sgMoIY!#0d$o2S&|(#7OvCK#9Y02Jib6R}>0rypp1 z@k67RNE?rczO!>0wqWq(j0`idDdL&bQ$)Osp_Hi1-kpC$S`E1U@@S4kmLeOGTB91O z36Fliicj6&#HFaUTj=|y)$X%Fc+v*paS|{U@l=7l{{`~Hn$dM0Yy0~R_t}aptn|;}CHe3(bR2EmgM3bA zG@c>Gims0dPeA~~v<6zz#_N~8+pI7{D@T~3WZ^w}Ukan^2qti9O^}l~+6GukKBRP3 zM}nT@QSR6V4M?5p@xUoX0S3R#yPi#r&27j$aJtrSRVlL;k3Xrk}&|M$l z8`vv<%deJZX0e51NE%Y9=|4yh{k{k>Wd61s?asoR4AjZ@Z*ueSlgf{F2iHER%PU6g zo7@7FA+ro3lpGuxX(3`0HP#_}+%dKJUZ{ZS(sW^V5pZvSP!D%>tmeKE2$H zFC0UP;4oM{?33DwEm-&nnMM&k{HuM@B`fUYrilHITzX5z3plcnj2#iVo$`m3k%Uqr?ZPkW(O}Q&`cK_JCIXaDR@M@Y7!6b zC){I7TJDlF>P_Jqq@-|P7)L(*dC4#nQS*)gnTdj*SKD}(LT|w^VF0>@^n#{nrpgey z1oC-a>fss)MM*7>AT8tXzQ@Hhd&LfCPy6z|Ij%?Mv=t1ls6$#x0xsN;qCx&^4zbs` zjh{OdM-IH9a9VJ~_8nv>FuoNZMa=sj+h3PY&sp8$Rj@at6Ydszn1QH~OtAu-Zvmcv z$kR!m&!iw_#G4rQ`)cav8~32LV;hHx6JS`n#)!ulF81K;fZiG9{))xtJz{M1nOv&x z`zpj3GPVy}U}dew4T>D{O%t4IMM;y1qmIjjc)dDsj$#X%D(Dv*F5{i2=1^(sSzgU<3oKCswpc;Nit1d}LwZq48ZLyW;g1x+d*5TWdlYfROd zLmZ=w=8!+W<(}$Qa{b-!6t?-%Vcjkbx4(M}H(6bs0he#>^!0`1o@PQc|55+@A)XJ~l7K$4hoa8@4 z&_wh=Xz};IF4996?AQNGk|9q|+J^8ua88v9Wv^cA0FLtKcM`%Ci*CW>``h**k&gxI zO+||QG`+bgge}7RBLf_uHo_5ssUIp6O7%hj5j&|D&fGTF5 z(ldu%q{~HTb%W!!-lxMn!wL_+`rvD7X4*8*H!SJ$;-e4`OAAHdMU>8$xK}fWlk;0o zX0jL#c)0Isoo_fWVzSiFD^DNv%0scn-v_Zjj*FjoZ)u!Y-gN(ZljH0$*gFqt6e-wR z0x@C%G~h-Dr-bD@HD8Jk->X!i_{&2n=%nrG%SK38`^Mu?cofJQmSpX<*~T?B`~o+P zF8TLA`;?VC|N9Cv%H}WGgrEr`hyOigC=ql7o=&uxL^C)FneZe~rdX^o6eGmN`dC*QZSnpcb`PYY)YC6T9hF6mzVJr2I?@}x%JzI$sQ zLMWchg(p*k9{vUafT!R8+9GPYc0zf&9UmXwQa^tr{6F-96+z@lC)TZ5)_g4mf5!|FQ>O@;kVx>; zpkGmeWV_)nq`c!CpFIH!yU^FjR5~dc8HMs@P`NUoJ=61%LKOv7M;?|5H91wT{-QyD z%+4UOX<0%(e#|WPb*ozNOP=seU17`ca-}?NR2R=Pi|I^StAM&I{4&yv5_vRk`fvkw)(&4~)kA~+1(#cO{k&1lYYDEdA(KPb(m1(r(x@@s^Z}@({C;_0Uc%YW1 z4g~8ovevt4UX|H1MK-m&D=H0BJXVdmFN=RaIL3sRlF*W+Qg}@Mt3RpuNsh-^v`lrA z)j;?HHV6H&H;jdvewzQU75DPw%9Csc3Ut{z;Lx7n*540cpu zNzU?=C4d=-M2=uzXGQiwww7FZZU}%! zAvtkWl6GH^FlYq48K!VZJ#f8j$?-Z>zH7;a{!j2!9k2tTWIo9TrF;yZy-oRzuV32b zbodj4jYYI1fl6hU{x;BL5`O%jI$$3?v~COhk+(t%#d!Z+v|)T9(Jy_hIRK?bF{J=k zZxw}mWIf*^jCLqcw>a;qfikOi?e$V0X;}EMVZtoX$H##FXjbilIH&G5oz_n%9YU|C ze7W@{|E=3VjIDiU)a#I@skq)M=l-q$#1smJ_=xaX6@nec$@B2C@uKyOrp#xLo~Pcq zJ<#3!qClfH;Td(CPNG2R>VY8PK%lEdjLeMcYGKy;^M(mMT>WZ@BHO2EXX5HvRcw*@ zfgsEX?b=SuEvJszJh^*m!nx`z(x-r}nPCFF~ zxnd5`UKR-nv8q65PQ{;iq&e9RJGWEF3|W^Gn`n^h|CTcMBZGyrG;ZqCsrh>`Ag#UX z!rYDQfg8L-j8*koLljmR(X5Oyod<%fPQ1m}keVe3=cY+hjh=z3Nq`F%Obk>FC{ z82CcXU#{8M+dw9JCdh8C&+u>&?7oCRJIZ!Rwp9;i>`-Nu`nacHaYq-OXlQO zF&p!@FsJC9bcL>DJtzGr_py9xD-4Y+%T)r^xXgU7DNy~-SRTVulyEBbc0AMFR5UN7 z?-RMjzi#JyMbYcppLapN>HDZpHPpx%!>f+d^w zm*PZ4qQHN)gx^175Q!g7*zBPBc_g#SavFXjJBt-62rX4BeE8}NAS z2UqOX81JXz=N$Q{lOAO;sK7y)>#Di`Cp$5{2arxSFID&<&zMI8FA{qoEX*k3nmewI zL_m*)2pBWc?7~rq!H!(2>-EiIqB28C+E*^PYo>5|y^z-u4rB17c8`?8%tpf7$wU)BF=9&BR z=MQ(Kwba@n1Ka1#p{T`T-po)GTFW1j9P@-TT<_Flg&NB}t1QsagcA^Su8Xk04^k1{ zw-3NrSI2i#G033l-|$At*(l`V>rdL)^~+sp0`8F;352!cgmy*4p+JP*nvB+Zx(?vh zw{#)D^d1U!$Ln_HkGx!L_Qvn>Epaxj7x4(5;S!X6##D-z3@|>oH~`f{*HbRu^q&R3 z8%~qboc_A2IhG*G`laKuA^M>6e=?<0V?9i|1x(5Y!tOxc-lxf7)aNaZrX&Gst1h`~ ztWkO#GSaEQ>r~+`rr+)T83}3BEaFowMjyKuqV)NTAhI==89oh*xG!^b{jJHQtjB z3dJzP4{aP)iXU&kXmwY-{w^&FNh2_Bqkt#A(5Sa|@27opXBt@fcWSaX0jpJE&7;XV z&2N!G8;Ne2b0B+<;yV(SBcWVtj48|W5meXMNGQYWWL)B;g*5+DW8-TnhZaDS-o3k5 zSeM!jk6$T|180Kq=1X2debL6oWzsxxPagk!^HZ*LFfb=AYQO&hCVtc9vlm+H&*m`I z!mIbn=q_5Afo*>~j%Bzj(i~XAAST@LujD>e^h?)wS@ikxXl(E*K^Gq3k3WszZz{Re z)Laq0+)Vg@4GC^dk)Ob2Yx>-on%7G&PE$Z;wtVT>$}!d_g+ypUSd8ugLDw+tYHi_f z5Lpny-l=_K92zD`eH*9*WbSkQ$;qF3J&`K+wDgi;sKnH>2!Gr^M&llNXTekB&1X+| z>nr2vcljEp`Ac>pKC{$+-CXPhYi%X&yvVu2nOn#wlPstB!a3$NE%tH+`H)}@M10te zi${*UqR{k&`+1;bd5<|3i=#AFdoLe(kB`MnSqPEGtBZd@>^!P;dc&_$VNrnAQ8quC zxRVHjp@|Q*1S;tzVI&t*9Tak)yT)k!;kMRKAL@MFZNWj!AJ<#Igv>(Fq3i@D6S?O) zfIVZ-r^lD!AH-ym9Y|W=l035XdTvj8)6@GGJ=I?Ap9TW|30-kMkjG!hr#y&LJ2o?Y zu#=rljDK6UFnZ-DwC+GU*W8@Yp8aWns~pihFCEY09?4cSApM{BYfQ(S+=Jzas8RE9d?bl<^@PiD_3@jq%zA+h)3EATogUoJlEm(dFe0`tjHT*ubh)M!TX znMtC?)d(5%&dYM^Tf%JL|2N!>*HIF1k?qK!8*7ih)GXY5X=<1r<;5s#nvr0W$mqfG zDTtZqQJ(`G3S+CqNnI8O-HX?=)q{^^@xQX3==BgQAFvc*JsYLTONZggl92h`$n=na z1bm@?u2~+&8{SY=`Q6p4a%VMjIFq24hT zIJ0hywNa)MFq?8J?5=oxnpF^hiPzH=c1K3kk2tQPnY8wI=&xlOS$9DSxQg4@(_Bim zN+#JM3ZL%KZER>BrFjg8R%`|a23)8AOHOF^^o&-G_kUkVkqlKgv%6sy^Oapn$%$=y& zp0u4S$BO3aggpNW*%lvXcs~WRIY$Z(!Sz+X0eSE>f&;p4IjpVI`x3M|Xa|9f?TacB zD@biy;Mv(478%fO{_?1;_b{WpU%K_uNN)W0qT*4TFn0r!x~DSKi~d3eSZVQxS(4PQ*kfpxb4i)>dsK0VLEMO6+~3@PVJ*4)jv%|V0bRdY5ir#XX-!- z?B3DeOFL0zKe+=E5!VaNxrvZhuucJn*6H?mmJ0CY^-$m$uLC9>;sI{J^?xfpL}(k& zzPYH9jcC!gKbLw`%&OuozR@w=Z>gQQtbLl%m{#T&X1re?S0d8r}F$X70U(-s=HKj_E6Ac+J z)0vl}Q0AJufxl)@`nzV#Ojk?wrOGdt=$mP$ab45r+>&T(oNx6Ic9Y4A`Sok|gz_-Fh5;vzf5L@hNP6)x$`dKhWRfgGJ z(OQyV%Ei) zIZm81LxC6*=RlbY%!FfhEmONV1UPf2b{VHMDQY$sp|QeGXRHZsgsR+)+}BJk#;+j2 zz1Zc6idnNKPx+CSRIgyJi`*^9$L@lvVA!*_DKc7Q+t*K$iE-CBfBSF=zPTgWohjRe zztC@l;K15^xJ!79a*%wL*DI3J^&HeXQVOK+UZh4ae;v{Qda&i4D8yX%E3(O!_JB&p z4pYh}jY=GC34P>*7%S=0xSRk?Wysp^+oG)t{oxLcBr}(Owu6w^1#yULDy^gA!HHKW zc;U>;hMaMM-@ZOh8aMU2zHU;COln!?C!dhZa^ylhOUn+T7_akkd4Bt4B+Jrfl~9V9 zvmDU?{#*pJ3}r_VQ-^_G_1#Q|o4UR7iZg6Tcf`tLOM6^GP15X_#_d6yS8kqRUNUhP ze6Gq!yS$F!>6fP5cCcFFx0l-O>{{QFoJ%%8CJNmjn$vJAFnN2$Wyuu-OJ@8Sd@a84 zXrJ|4h6y0G>>GP4Q5be-U@I25U(v$#+>NGdxy@thH+KcybmGd6W#Dbcpt&`vMC2uq zRZPZMR4MQgzZWC&kI-2|Xl{Ad%0cz{6lHTdi`Kq>cV!M8Uf-GOF3VwU+h+@yoEyO| zzYu{=LL^F(n=tL@4;HG^Bzkv6$}I6p&UFWagcnYX*La7B8~3D~ZrfwfrydQ6LY(NW z(YBXL!>PqU8ktdRlSB0Ro(avBPu+`hg~>RGb|u{vg$VcbE;9n=*SL?rWFf?|SoWih zp-m?5h2u^X((&Twd%PE(n7b}gBo2N96QU8}Hvj%7+VXPy+pAN~g&V0(#pZdr;R?Ji z>HkG@XnnuRZa6R6ft8Bra&Y?(Q|Ozw_%-V^@WP(Sy`U0}Z$YYiso5F#jI7R!t>cwp zkcamWqW~*Rvxs&?F_Af42G*b;n1Y)xCcM6#8O5-j_%C~eN@SM0rlJ#dLG;p{AeK(l z$l-1eAyztGzG48~u6flY3J4$6g(?kZjJ2Jp!8BM`l4~xdy<1IeZyYsfs<@cA3~Fc4 zVz2nYJuCDTkVK4HD;qgf7&_B5Vw{wS9{D7b4F`e2O+_D&8cOx@e5`*3-CMM!$`We& z+bf7SV0ut~@Ps$Hyd^D4rT4I_xwB4E24z9quw%LbLlS3CX{1~oJFo$BRo}1g;SxV> zI523J6xAcVFXNnNfjH~5PVA`oqOus!z%Z>ug`VQ5&a6iac7XsC+&C}_6 zXc?5`*|1l#A30%O2$A8@@DE< zxRDNG6I*=UtyBa{1$vaFJV0pMF?nGMBg$Au?*cdTy436UaRhi2P`DmGycnA)d{|p% zrE)6WwG?d&SS6;?6(82pnxjvoiKU{Zqiw&ky07W4JLClLQK(P3I^the^bzU0Y#mariDBC9 zsrTcVv>053x|-8vx|Gc6QzYC7E~Yy(;;xGUmEV31(d!Mj$xp9D(}BsK*OEXjc|XFC zX{QmQyDi*+9ZgiRGNI`i(Snb3g8MN}>A$)5OsubcJ?I6SY{$QXO{lO>`mWKqLrJmA zfaW2N32s(t`{z)K5lIZg8uc3Q@eET8yy#iHW&1_FUyu+s%%y#452u+Jn<)?TW&gb= z0wgiA&5SxLlqS6J@X3r;l+#P;pwG7l(zXKu@z<;okz; z27L*I)?~(^K)9rD-XY1TQ^5g`yfpqN&=3h=3Xk0kR*G)`OptYO>y_8q?C?)KLfvoI znBh(pyvUwKJ5#sl!>z9as-QBDRzf@W+XLM(Cao*UPkmap!5aRmAqq&O=&iZq1ux9% z{wcI&SUFee>A4i4fCqQk%c}iJ7QAUMRQ4FPxIVkI1ECWOK&i#E7=$?|uvH%C;Oin`m+Ce2jld3ap-`hVEKZzT19oU;axCr0| z-x2OVOX2%1DGxvq;rQ8Tkab$A@QAdNi-7-$LXehx&-qgkVn&D#>5}Ii21{DaT8jZsgJgQGmwcf?O~Ka(ZYXbg#syTrT2a!teg0^5b_S z{D>SX0Q#>xl{mZ6lMg#w=PTI3jVO^2>z#{e%OK>Dc}rYrro?Z~z2etwJ)9Cfpv`Ci zozv~#!p*}J{OL7cPQFBgKX@-%A#Twqg$BUMN@EHI$P!lPs?*@pp$p(f|AYDG*f8gC z<36pg_Uq9}>|`54v^9!AcSU4(_{}Njf%gBRaYOnbI`Y;dXH1*eP@D4Ty*Q02Zl?iJ z!gAs)P^ssgbxL5wxGsS$;rI9)&<_w)IA@5LmS<)0!g3I3LOo58(R!o1uF2g9y{>zA zrb$WrCNveJkp zfC6lKEKDjNN&}s!UjM`1eK=DkH$?I9#!B0%CVHr?_aA^)95T|+qdeHZIhloAG*u|FTO)vL0uGpX}@wqKhBAl1x>KVsQVExh4fd+_tmNLav^VNE-fxl z+)pSbt+v5} z^nBbgd-~w_@;hg@f_^jw7mbjlHXN8=|z)B{8f94eD>-h z7JMbw)M57!;rS^h3JN#=wj^|XV82(h$}lkjS~45j)Oq+n>sc%|fed!M1SgV`D*s|L zCO*^ydPX^^Y^1JWM?#?;Aw4-p!}!(g6@uNN)aQFqkQwR#FyVAtka*Dm^q7W6XppYE z**=i|XTY6ZW=>#$d?_ek0_^a&S&UK@pdsR}LOl}Iy{Hh>jx|`ez9sj->lFLWFNPO5qz}xs&DrlKCx*cx)UXt zpLf;W#P~Xs4!_PZyM?&0KtdwGjw??gRu-v)^^u=DI4DD$#5 z`R_TT3@+}7l2wM0OZ8)%T5Y7`3DKa1(9?9_47-1hWFFQ6E3fZr7=p)1zaZah)}1*Z zkaF70iRrv`v+a_}y?vY|Li7xqi?jKywl@k1UeVx*i*PN^3h-q|U-Km`taG`~!${m` zSGG;a&bOeb0C^w>%_U|7voC|-*1_}G?I&>1Bf6Ma8I3S^UHM<(*DvSa|9Dk2g4e*l zSIC16jc5qfb{Ir;1=5$__6C&3z^fJp?us_2SV7`nXzwGCyMo!k4cTdE(YH0-^e{nG zY5c9D&X+I_)5u8X*%bsD^qID#<3kqCiysQRqqW>ghwpfHpD;$^Z0`%O6Fx&oZ#X$` zX%c9j zXw4Y1&$7d?cvg(`Sd5bVgnPe9pnx2he>FC&6WA*dUiE}w+F2~(Z$Upb(Gf0nDNlE3 zCW4cpnCT{d(uP9Q6B#YZj(?C#f}fA>lRwZd>^y^)j=%Ancm$<-LSwOLIdJ=txC^Lv zSnOfQn=9~eVLqBi%MpIt{gM~ir1D3nde@Kh;)Gs)!%75)8keKT750(TSeU! z4<8OY=2Bn^n~s|N;BVEe&vUL4H?m!lLU))+ z7N1T7=flQ88;2sPU1vgz)A4sLyA8w{wxH=rJ}%|q&H{J$yc~JpEy#KiQyBNmJ?jE_ z)MySt(*bP!Flma5`g_KDSUX~;x3z#|AJ$e1de-LS{5+FN@dR{F=_kRRknl5D= z#hDa4A)8pmiCHb1n2SP=SnP%^VHMV2Q@bnTYLn$2Ai6s%C1fk;I@LKp{=@n`zj+I@ z`lRJplziV}K8y)|FDxYhQ&??I4SndmOLF<~Zc)C6d>1pOok5PlnPgEt`nvJFC3lG@ zO;ZfmX}xWTM~)1Rkd&uL*PQ&J!V9gI)B^{_yG-Q!-C7bq+!gg!9ibL={HLsVvMNc6 z+#QAjUiFA`Z=Ft;YkVF`OC9WhV+xtEhazzD==lbb)xv4BNIr-+cPac-Y`TNA)B5l7 zyZoi08r9WY9$#;AaC{qjb0NuLDZqHE?QCy9g4SIzye0DztcXX0vmHkk?||Y?NowwO z2bU-NA}@y0c9Z)R!B4C5UDXA87zj2J?BNR|AI2vor*m`yb(*XbXSEi|B~bj~rnbQh zH(N<=xLqj#>?A#d=6NaO^^RCJa?|D{a2m!wP1-Sk_~mY{z{2`{``FS27WiZGaGEO7 zL+{;VmjM-HWX(zB8*9ub(G?`{JpWCs6C& z0QKSR6-}4n56A;9Xz5}39mEpCPvy1#A_>`jE0@jHomQeapnE*;bk#xwrCMisNW5(0`YOp>6vL*=4iJam&+SL=38EP~B zfRtI`{~Yr_F$w`veBM8-EBBpu_pY&=PxD~A)K%U4M-3kdS|BzL;VZ@KvD%AySg-AG#R4}K1ogNImYt@)oFN{i3yc9nx&rT4yYM&t{O`YqO_O|)VEeriyc z$rS+q#+bFdxDBe#^rw+iO)#$GI5&DNlQc?8IA-;I^;1874#)mr&+J(9N5y}k#a&ac z(}{w>Q*!_N>~tP&`Y4>cCC6nGcg*UA>YtkZvW#z}|4RP88*S&_?-N4nMaHz1Kw7n# zkDAh;YbA226C04ECYXtQ*BHpQJN!Z1x<=%@9XmfnVVx!S4x;YSZ9yOzMrU#wtZCVhd7)}H|gh*i0`7AJFn`^HEODt%QTD~Tw` zLl=%DNEgLqFZi@!6P-VCkLKFL68Z!F>dWwIsuxwN4^B1 zQ;1fI+~yj#;c%hd>>fph=;O_70oDe`8)1zT8lStbj;9JyFtINsDQaWgidX4XS?d0g zR(p=#;A~d@u%KEmsg7Km$B|jeUX}^RB*>ZJLI!ZPex4E$!(OIg{{62^3S7La{)rr1e!c~P5l+7OPFEtdX zRXl%2vOcvGl@Huxzj0aZDBS)G^jn|#f3(u{|z1|&1YjccvPYIqt z4=ZcCaJi$!Ru3#6I-<_2ChYsC_x4ly_X%i{yQoeSz(YHkgJpQRwlTIcLQ}PHRLa>t zXPYI{nq>SSqT;LZ2w|6NP32-HjCtv74CBece{XvNTZdNqT7g5rc58CTakfj=A6Zbp zvw)i$z2gG+o?xig9xA$j3w(daZ%j9k)9Qlyjn79O#*q((qGA4MwfHm&(^3Y0&%vX$ zLYWG0;TKXlCh2z|eow0JrF(X3M!3in!|63G#@Q=tk=*LB9vGl!tlY9i!|(`ld5!&S zUAFN>ewHXt;N#5Jq=|c$X3zgPMKk!qd#8sjj7mSxvwx{@9MEx5dvM4d)mhNhmR$XK zWbM|HQrhPM2@{8IkiaQ#&MBWvf@VR$cF$1n3t971x#h%lY{zPI!9zq%3#d=C6F&gLy>Y7Fc)9Ox)p#)rJ51 z0uxy$e1;?X$_z5?>G#+jB+t>KpH&PJf-NsZYTSKmOq>WvO)d!P6~gIY>k$C~)LX}) zy>l6n4foCl{X?XbDN&K;8jRm)^|cMg&W^IF`pYT?#-*TslX6~`_IVBpP!`bnjNg~8 zxb8@d!-c(q8T=n@*ZoiB|NfDgO|r7HvdK=z%64RwJyT{V`xwWTke%&hS3*U|maN3V zG2X~t=Ohj})^VKsd-wStzQ6ErpWA)zalNkVy07bbJ)e&Fx-6OpZM?9v$N0ZmmQxiw z$R(Q0ryC#Dy{D4%Y7|EYa12!$^olV1w}2O&#;aJIfDWUR*3)O06?J>5tHStq`rF~r z(>P62*yI1mmz#NEOpZV6(8>8R-yo4N(|9IZVUqv6WE`GZkxenCWpn6Xk`s<@uKY3U z!-c|Gr)V+{Ee%r3sI2d+7_IRQnyU;pqm$7*F06SZOh!;AE&eEi{2O%xK6%`K`B}L4 zrtk|RzupIQ8ZTS;t;xU!{UFp0EmzzF3frFjXnDZ$(VzZePJ|{?9;bp#%*8B)i*Fi{ zwDHRf>yzDNA(6$|Tr>50EQc%Df|A&XevjxmK+dfd{`;U{40})5#lHajhCGa;xgySm z!0M0mAYM!c@<=91#dzO8 zy2-!9r;W=ro6;AAxF#9*=oAzGu6?1A=g*7Te1I#lqC2aViNEm}&w?ts0eZS+Va?$j zlHRs^g+x1XhA>-E7uLQ0ruXIg9L$K{;EFhrrRC8j+Z^_x`y-2Mi!yyxKu43gE@w?u zBcscx6Z_VnG0R;kZctB{-`30X5qL9Y%x)GCm#?;er?xBXCR{dq`s&5vNEbazNqRYI zJAFs+Q~ncn(OMV=Df~M_k)Cp&m$*YaCNETeMd!*aB}PbV|IlxW3gVMf*oI^M62E#s zCE6%!p322I%c1MTzguFktl$mF`#35YW3G}6>$oBKV$mPle1@K4-6W|%b3h0KqD)Oq_M2& z34QXU2RFy*IdU5(_JK9?+c0TQV!D5FI`gDi%$RDc)|K{0tc+iDZ<5?$GD%5$GXs{R zlCDox@aav|7-_;X^NU4Eiy6u8;mk9N-gaM@ z6yyJC;i!OEBt1nn*~Tvn6RFE|@s{9 zh`0L9hv^!fZ~uvd>lIp$+j>b>b;?F@DkPy_VabO-u}+1TiV=>exW;87wAG&ZLID4a+jcWX-h8{9g#^Mgmub{&+NxpvN zoeW=nt14u=@tyIn#0{YaF0aM@;RJy>J}QTr&};KAOlAPv;nvU@0JAd)=}G;Rt&i*P z)uGqm5YoyD|7-9`HxTJSIgn*asrJ*1%l4gAw;TqzeO@8TjBqMU;tf~qYN9l&vriMa zY<};kT^c?JS$PqRP{;Lh6W>KFsf@s-vQV9{<@QH@)$$^~q>OPmflGv5Vrh3WZ{a;8 zVaR8a(Z5VnT=eAUmj&_nBbqymNpkPphT_JPH_c{r7~^4bOxk3dZ5Ls?<~5<*VY{*D zua``&6wYc?s17O3hd*%AF(>OBsn7!Fx$r||=@dW{;>I{d^B<+uNhP50j)jHa>`k*^ z*K@B(-!xFqMlIR#vmFO8E!v)h0I}*nAqsP*c!QBrTpyNKC!?n>6Kok9Hij5o(FAR? zLOfjVQg~7OMOm^v=H^|wMEDtNyxW7+Xl?N^TFmiucMx~=2`MJH(1lwWlZQ~GugYk5 z|MvMoF_H4V;PR-qK-ZwN(89Jbsh3|Iod!AJE6zHAwJz+wb3)1Vf(!PaIL(xaFlb|n?Plh5ByGPR3!9dUzF(|N;)Tnh z@3m6VWSYIM_;UtE2WZUAIQN-YqS+NnRVRbSIq6=dfpQbbpDctxSuj`>3aK9`0y|K57h>Zq`7f2#=`ui!oRN= zRGPqzS+z37E5IUk+SAUP{CN$j$^Hq%j8H;9;=PgP%bQdz5!nJTX>VaT9+I4!*JBie z^Twx6Tw&JW#!)snyu{hXd(txvTdne`SHqYuCE@155iT%m)fkIsO~bq}{wl`jtG}yR zKh|4uSj)&o&EN6im@K=Y3n3`nB8|AFBY1~^Hdm&<;G^QwRkZ?Y4P36mK7jG%zED1T zo)_al>@XC>m*;wh^}i!#XsQ3A`ENxzB&961uG>cSMHzbl&xcpY#RfW!v!zYz1e|xu zcV_SkyTX+zvu$BP@yF}$E`=>#^Q>rJ|5cNV{(7@;nT6k@?xcm~@*thI%nwzB>~!OG zQs3AYE@FE_^g<&ybQ9Zcpv)fQ$vW`U1~ZkA3$ot zHMZ#mT1;?gpf6_;?vQcv<{z6YP_7Wem>W=!=uZ}057~8*LwYQi^cY9P z?7X$_bX6$1S(}P8vYId^3sF8k7n@XoyD?tL$-ON5uoNvR{vz-p&u*f^K-jwo#mg`H zB2L$>H)k%5Q_zKLlr746F^+L>br>oJAuXP^!b=2*~me!TeuiF<*RGfMEcaq z+pw9J(O)+{DW?x}5qeybKIkCmxL&m25~%9?o<$lJ4hkWYDV?Z2Rvp+3^is;sdv1An zqMR8q8?_Wae4$xRoO>?*Wma9f6&smlxhiTO6rjYP0Q>bR?|9}!u_+3H%X&) zUQM>v%U%i}J)n!sMbATT3s6l-0ARs&@7yCMKZ4iJLESsHjm)XTO4566IroM(qclQ3 zzd4^4n(mYaZ?X4Y8wICp87F)B?do(P*5^Ie6^M@<$ycLTP#r`<${W?D&x_J^tH-)W zRj#)SZG~=jU}XHm^pzQCQ?E_ceVXB#cL_3NEhFl<;{UQ{TaB<3>rJ*-beI7)*kb;1431d z%Rm_gnd;>)i~hTxH6wFT;IrhLDmT-OKTt@6(_Pe)1y`1+Ch&$SMy(;tX)@zex;1(3 z;i-kGC=(WN<8RvAy8b~BmXWTywI=W@V&M$TCh*bRwJuJY|4O;jTHLWf=DpNtI-~F6 zHLUz21f~rU@EhvH9%hPk?F(@236nn03VZNxS%JNI55zRs+zI@-kU%}_3r9*ND zED`3KkT>WGkw$cA(LeeOhRA8Oqb@SuN!4Xy*0)9`t*8oUiXy8N`MH{hS$bSy^+rSr zFEAuayGKFfd^VtGU;7QWpBVVD+mA;yu;GyK+jg=p7`> z<{w`Hdr-$?PU#jSs)J`^scOzhKAQ{^Ot@58&cpB^_D7eF8ly?njc}HB+s>R95Kh*Q zTCgQSwxufThPUSs-MJ=Vmz;aW-5k?X#%Rlpf3^0z*Wz+TaVfF!@=^zQCG zWNO)`7+gD?S<1^<)1Yx{?d6Ai$cIKh-0|8;M!nDth)7&^FChMKq!_bws@?JkETWUq zs~zVQEb|<*;Wzm+ov_o7bN^HWNA!UFi!%7RgCvKZ6EH_@!ejY_c5-ajIwRTjz2!$0 zU8L?n_!qF)XCP7HbPJba)<_eb;6M}%` zP)0-rqB59Hpv$SALzw2j!Dh-IpG}IpnM!}$#II#QlkYMol7qvKfh#&r+s<2w^J>-F z7jaMSCo~nBcZF)bC~F$;kuT^5pT0tw#-+lf9;IOcD1&YB`UcabFYk125E>=%Sv8|Q zYNPwxQyNvr#GJ*4*C!QPSbz4k@s-1oe*NJAnvlGT+z&X3ht)&6L@LA>(58f&+HcharCAH=|l>q+h^zgMw zv!l55yR_`;o#PbZ-kNlDU#GzGMjlx>)pf#==27UHqtcGti?Wp0_}fum#h1{1I~hy) zg583z)nBV8CIDAHR_j7YujQoxIrF%=Wc1f6-1d-0!F{<;EwBx9o?QcUHJgvUsLRps zAH%?*5bj-cvoWsoQ`S3O$m?(Bn`b9F#-nk$Kh8YlClY@H|A{N+U}!h4C&D>ebB~2# z4FZ4Co1f;;et8!3OZLCBM?jKNTY}eQlh&jWPJ!5C){-gg<)%gy^q=F$ZL^>JxJwELzKg>_9 zq0kt2$N1#Qi?TWE2+kV<3A|NyUPau17<2L!8%NzduKYc&o-zms3%R z-tlD8W{m#xqt@DI5CW06+Q8Z#AudRJ#U=xvJ%7S-dEdo2!719cwkwCTjSMhZ^SJ}6 zIGlzg-{o+OAY1(omDxe@f~eh`@neC=h4IL@$;IDX)Ue@^6JPqV0zS%FSm z@~r{6BDLx`7izt?QUb7LKFx@wnx1uT^2v-rNNyNuIhEYh-uzW!piuW<4T$M;*fwMsdu4VxoF$M?(fbW0pAZ2?k=Tc zjB^h{axvH`bcbxz$Twi52V?y5AL!$;T&OEPTO{}z(e`>oXpPKdWaq^Y{NyR#;-Sd? zE!o-r-u%$`)n#1J>Ekp$K+kV?c5)Z)k15fT>Sgh7s$3?ODKWU`+;!1v=V4DcTMQP zQhEMNgi&d$EMUi(D}bANgsPE)ukSm`RUMECudOlg4blUT##k`WZPDcM0(Yb=Z|0c>c$;m2yM0+R2?BN(-1 zsAT+huXgzYiPtPH5va5*D>N$oGw~2%eY{E)hrvo*R5P4G>YP*fd0+uK6)VOWCNxz|)%0-hU0zb)YlP1J!Y&NC#qMzB&vw63slA zYaQ&MJbd6Tl{7u*D~zH}BA(Ry;uLhRBg<%>^W~ z6KKAMZ&ci$sO_^D;zOxtV6d8$FX|tlNfG-iLb%mo?|1PNSmi5jBW(ETq*LBR_-|wq zy6_O`OB}=zd^jLX+A{hM!No+UEbVjROt9^6c2hgxIOm41_$s7veZw)4H*#?3g76RxM2-iK#{nh7;C({fZMcH!U|q?p zzK!mY-p`LWJw(j?J#08bj-Cf)gj3eY&@lxAqKV*wtG$QRz5nmP*tGK72XNx0ErZek zFh4xhQLtRv@#R&<>G@TlWLx_xQ1zHjW8$c>^b<%ve}s)Ce|`#&j+)js0%ig}q~i#6 zHcXD7n&m-Cfb-doz1OEFrSNkp*a>L+mOLp}0o~sd#%y?v!*a?HK7t1Z=*5LjuW`-m zAn<^z5D)C5^~5U(zHuY%jmBB+eU!_ zZhW(*JAg0KdSrr25O%}^c3l@ncuWer4dVZFaQ?e5YCmwHj&PZyMrumcRp9&SS_k_X zq*#4;X_OWA<{Sxm1~&Zd`arbCyh?)1wN;>#y^kMc*E#Es+yckqjYSXf4+uN$k+3Lb zR>F?41Bk`iW&dY54(S2K#D_<2_=^k!$sG!Scp%RK0?G4QEnK-vVt}85fcR0$@#pRN zoPDb~+Qwkxw}}w zb#j8+q=m2i+jfou4i|Cz1EP{v^^7Li!A10?XTkCSwb*bm#u4cu)%+gZ`ueL5W4mE# zXW{ZOJOx38Ki!XLd$kTh=BtI@?&27t$|M$2T1{QHaHxxgx7r90#Z!A?CdjjahK7Its?!A?C`pEoCB952MHbY!fJ z8PytVLU}JE^piAtA+2+%TAJX$5~!p8&f&%-i0^KI9%KS8P7uO3CYOm6bX&pOz9e2B zA?$)j^(cYQ#1ZS}NGa99QZ+7vt{4m%^EJ|_e1`1dd_~;fM(kH66X)o&0@`;IwMuCr zF8IGcx_{~Pu%}=~lO#Oa7)+dr=jq_$8olBGS{3SoL-FZh!d+W24#H03a;BX5_>oKo zaaT=7_%V#V&-pqITOV<-bs06;16*$0d6y%F^OAkU{j+D=>hv98w7wYev0(Y~Xz`(V z!bI%|qnrpSP`k`DsI&_Fte?E|2yNMLB$EKTfnYU3oQr`ZNE|Un{`~y}Hpt0{I;zM% z$sfy$>X2>4IMUBMUjxGbkUu|!SJ1~{909Xpo3~)TJMgDyUme$xgP(y&59!us>`b~FOeZ)7%!_|+ zT;7|rnQAeWHn-AdO#7Cr#zsH{Q5xUEVGNf&Wq~4#Y=5k}95{C2kG+MvU~!hSlV&0! zy}UAo=eR#{UGhcT+TereFb>KxX#*a8mXTCCfQR>g4%rj&E2&RbfrHnT1b8;F_!;rR zJ)ASbmv6@mZ5dA%TYWX#D|+f}3=V!4W(}IpdDtBOv>HTaIE8Ge9RP-E#V}Myf5EFJ?h+kx5cQt`y z!ry=eT2zO>Q_kww9qVJUk7w1|?o$iV6yE_+OLPt+F+&;EaeYd~onR?;&mdy3U2C?v zGTF!CI!gTv__LL`8+iHUqp^6*55QBXnpCJ24RHX%tV~?a#hzie@M|4dS7!)S1=cl8#(L~PJfZ)MC*=rX z+`HxaItQw9v(|=ZJfY|un-)RbU%U6;pKPk*-th=|vf;lY3Xl`=m;dhk{*$z6_Dck$ zIC~#Bz(Cu9rk2lZ3b<9bb}?=QKLp$ z)}RY9SWwayRNVs=KB>$TorotYjfpMX%rM(KFB&Mm0Xy#kK%-=vyr#&9KOIOvze5V5 zBg&hRsZsIq#SH3ht|ShI{>}laE5p(P*V*#2hyWq+v9b#&FFRSEZ36#i6j`1Ti+J;`2aY-u!-@JhjlgFFJ&AIIikl)rLjt8u7A_;72v z@wRHRCam($Qhja#H%h^$78p5g`@Bs@Z1 z5nU_%W7>!7G+Q|TfqAO?+q=}!*@`)l)P(VZRs+Z*yJg6WtRj}7XsrZ`O35Y z^j4lo9JGAst9`%8#t?;T?0S%PGZ&PD0OgFv>OxcQ?ImXV@UPkz@iUlWt8u$y!7apt{%4}|08>=W)%581kaxG zGA84EMCtRl<>_B?u@HCwgbh`Gs}L1+D@x*}*}LDk(}czB7Ujpyq>4YwW$4t^E~9=1 zU#V}t-e{!SC2VLd(SaIK=q!!L$gaz1Rq$3fvR9`no-ORLY*Reijc|>9vgE3}vVk=> zT9JME@^l^p=xaZF?hdnX9rCvyvAb(^p)<|Hr0u{i)5=rxbDnXA-E@T^?P1cgvJjZU+bXk~etNm` zhfseF@FDD(UJ>J@^>BxImgr2Mon)KMWITsix1W7fUJ_Lus?<2yR|>e1*X1QEtZnl8 zEXrL_L-CcE%0c-pN7MYNxuAgy<{6i&#=O*o<1Id-KI{jo1o;X) z|I6oYc2T42R=rV-W;jum<+01X`Fp(@1xfYqcY%%j-(b%Bgz{(cYZppldOEmbS-0Z< zP+hS|A)Ba}rkc^ia|`NENPY+H(Gqd?n8W~8q}VYuE@LOPBn1k5mZ#U%*h)ENofRs9 zirmY-a!1cqjR4&kz5S8B;nMZCX~G$W(F*&YYs!=CDA$$R=NIIt@-dd&SrfD8cNm9q z=`F)3jxQCha=Z0Y`sxUDxzeZkH-~8N7BU>NSZ%a~?wnSrT}E~`#=1SBE?NFIobb?S zC5YaeV^?Ciru+8%9TPL6@lJc`a(3m)K)dkapt;(zNOnnvt$}KnV&sq> zj|=v013Abmk5OW%b!YQ$(jM7WBnrHz@(lXCHI&q~$r@adfN>FK#SXAx^`|F2I>*>e zM`$8W$vkS@q&#bt2c>Li)8OAqnr;96fumkJaCE z54aInT^2hUZcRFPI$L$7ms*H2_hIVCn;5xo8m%1Xm_a-QcI*RSkH|H!wklo679CkH zmA9APDopnV`>0}kB>L>rS~6OdO1`~73zNtv!arCQw1Q!8=gCEyGI7Vc6UGYlHvF*s|6+O7at#Jb{_ zEfRom#$x_hC7$rbrzZG~aD7W)A1@3y7Dr*XP9a`R+w&$otulA2+Wy}fKpC-1+*_mQ z3UscXieQum919pufUhXHP07*rgDX(~%^OLGSkxZaI-%JlC#rrLV60Up!t$nCIS{U% z)o;TdNl1NvY7&a)bf?M+Q7bIyKylkzln2@ROgK3Vf(uQg5Xxt?@~dR|%(f4A$NHrh z&8$S!1ZZ1~o$S>?Yr;Xpv?u`&Z~=SO#OGxPAlL~CIEdp}Z|r8PL;*uzWb&!PvVVpA zLp{JFEcmV~iDoa3Fbg%bbeFo38`g7I{{=M3lH~w-92l+Am9TsCIdA|wXv)Yi888*itmvTCL$|C zPNnOC_^8C&QXG6+xp+_=;cS8|iwpxzZ!oM&?`MC8IE-HVCh%95M@-|sgd+!wYas6_NeK}H-wTp3Vo%aLH^+O|#GBQ!MqAAQ z;ySl~uwnC~7qg>nqvFM=g=9U3)qrF%9Y$tZYKH>@vFAb9p8 ze5B~FRpGZOV9o!=E(npO?A#!XhS5Pqt&x=G@+HuQvh?B2pzQ6B{L777`rC_%fKTDJ zDP|OSp9VKlv+iupS#OW@2qe+=sivLuNek)4m7I}O^2llwM^1ca^S`6hj!T1I)dh>x zf=7)zvR(g?*6dKmL)SSL-MCz{#%7(E2&_J#--aKkcaZ`rYe9d{cIB!{{p@dxe!ti& zvGf)ZSjI+UfEzhsJ@*TiBHxxkIWkD^I^Mku)XdoE&`*~|);|4Iqff?ht7b>6f>x_4 zuX`sj01x1$FeA;5M7eEer9KD;>pRfwg}rO-BW{)u(_?qyMmTAcm9~Q1?uYBC-@~ZC;5RtoyXD_q&|9C{I ze`VTxOo+E39SAy5>7vkWxf5cniC*7-{2h}{>}|9@t4;f$(ET}om$cfSeB-(-k6UqD zDKwfTKX#Ll)G6L2?HTQ~cneW;6n#%oHm_v}5_W%IDEBu&k4>K~tdbYAzIUu;W$BW-dqN$XnZN=hkn~ugJYsY}erH#q`L7*r=TMv6zGVOkTA)`5dipiAeGJ zPj=Bbn^-}%t3L+XG494w>4LI_(?;y^lx{CSbE;lzszCVCQIbCYDI z?DPEJ0O2Suu^9!Q<4Lpc4f;l z)xsWp^owHE#w3mfdlJiMC2GS?pejnyRGLe`U$ar+)rr>fYr(q1r}I zz{dxPKQz;V~rV7)r zhcf_;5LGAHE*Ve)iWEqy_Tp{?exXq9!$o?hEn=h`2RYsY3iuG-(q#)>Oh=&6Ilqk2 zip2BSu@OwdDaBTheH`^hsa`K6>cj2X+No*;$5%C30MFnQoATg6r<|}`K#`!)3JSu# z`Mss8uq+J@P9xQ8feX?@52U?%x_KA-N_kuVfsylaafU0cr4jpb@sI%^dSCq#)IrG% zfkog+Jp-UUyYA1jPqv?ogJDA5T&XS-FCjmmBR`J9Ot}h?F-?1f?Nf(fT$=T z=N^W(QByRx+qHc2XF5=sCrukC#%!oRs-F6|*t(jQ^9Lp!%yAvgAoJ_#L${e$9op_3 zn7QcHh{BV5As&J$ndRkq-4B10#!%Pn^enyjy*YqMkOal-bFcn;rS{JL%IwRhQc6yK zfBm-n48cbV;f4krut2sjRkaKROyXt#XhFwDnv|0ZK<>_DimRv`&#TdTavH|0hqcwE zM5F@0B~(=O<}`oZ)M*%4q9!EvVE~B=gI|qY-K@+Lum>4GvT>emC4K{#`m2-f-1S@_ z#jxrNi!2-mM6+Gv7Ur!aR%;cYQ%f+l8|#VQ4$k=TG{1*K_8tC39|L|p@5*&{fXaxy zf>|qw)F^+BwykuT*D#;juECd||5ksTHWzLvtETTC zUvwn!nS*StMcel<&mF8nmJb-`gE~Lu9cF~UXkW!o&Dr_5-z=nGyH887oHCl@%A%>w zX_x89;XNHu?A{SLfIpc>7g@C5Nt%EC8>H^KM&h<;I>5Sy^Y2J)sCv*yo|$r5Q3`#re7uh9iCIaV zBWAGIE|R?8G7!4uLatsTtctE@k|>PPXb1Z53nicD&sFJv_5@*P!Lj+AfK7akVoLiev7cye1hJHiwou)CS zCD0dY8y>4c6*;_3x1uuq*H0}tj8QSj}6*Ij8(USeC`Wj4fVgO zUFw71hK6wFG@Lt|93sw7_Lnd#@$k>8rMKC&R7#)=rsKImdjl9}FfrzWNOBWnr%qiU zt*AxBN(%QcW-&+J<3Qq89uRn*d(8sc>ik?_3)jN`x&&kkE-b1&&lkmr(Gw7CJX|}r zm_-o-2}iIH6An5<{EzGqazbFR*H9X?HIWVQklc_1XX|C-A`jt8G+kgbr?@V{-t!|U zCgY?so(T3_A5r~lzz&%Sf_t>Q@ciX76&xh=JO)rW;Ae=e4w~1qk|>ex1(RGAC(8ge z+odmZs;fbt-+wx&_>S3xeIFsGO=klmxcyj@xuQxb$fA4yVJ+Mo|Edj;bhelw)b-nO zgDFP&)PlKThHK}BXKjw(nEX1Qyw%jCA!R*tcgb>yVbjbhAr)bwTAIK-GrA0ne8AO< zJk=J90wvLX?&Nmo@ojdz+XhxyN6|QcLChi8?_;RR?cO>&fLZ5uBfhW*)G^k!UX50{ zjM#QwO4v&~pSzf=3ky}ts`3$YjDeeoIcobIsxnp^FD+b3-PXyK-aV(Fn6eXdeD%Rq z6MT~zPoYbH6Er_DM+@j9dw%R7F$s+dQrCzAwf30B-laz_JqKtc-tBN;;&IdG2Lo6@&1)Yl2y<)o>`atuOviA@@sqK>pcvcH2_ zm?13fi5Ljj9oTtn#|FOsD*g3qrhxsJY=%&i0SzT{ke==&3+WnqqEo(ek*8xHUOqjJb2){Kz@Sk$?$0J^IeA%c9ur3;S zh+yR#;|Au7QIJPBZ}ONvNxMCwPMp6**iq>Dz-0J(5yNMw-!Bae92l-4F2Gpvv&S>e z*=VH%=X@!vy+lhy){;(7)uI;1B;*N}hI4@KCK{*LlGqCws1KS-MFCf%wb}tx$)|%v z1vH@2c?AD^2VfxvD#%6Zl^cMp)Y4wSLTAVK72aIP(9e5c3qp9FZ zkZQp6?-anfISXgo?2*+0K((V2NQpV{Md!c_8&cN>YO}XTJLn%GWRg1$%*qR@K0y7A zFdaUhv;3>ESqqeV7bdyg+?2NadYB7M9a;^^vz;8#mHe*{qVu}vcbw<*x-mqtC=ICS zQg^=g`2#$x1WJj>QVXrdMM)Rk0N-EreA{!UluLRfOM|9=YInVIB`PlJwY~xHXU`hq ziDpNbHO^ya29?sZyApGs*#g?T5TIrY=7m2NM2e(jgr>ut82>H+&w4TLKUEJg+EW{U z)@SJ1fm(J*=Fgsn6V)HVR=|!=Lwr@GBkC(d3G{XIGi1;BXrBlHQ87}g#4JbT+_(aO zYg{e*5Z=&Q;i&MQ+iM;VDFXVb%{LIUTFqcdp;RfsBNWDc(|h_j9%+BImhRp3d?pbo z<)k$MmueV@$WU2&&DRoVoq5xdu@2_OenlRf2p{lN^Jq?gDV9+baDk0ISF#xU-; zK`pVRasS~^nK9t70$69iX{euSCxltZcJTP@PYm=YK0Yjfm(nBmgDjk1kDMGnzA8K? zHK^S9E|1pyH;#14@u)E$zNrSj>ZOzAiYuAwb*OmJLe`4l{IZx`%6AOn zS!ofR=kbT?ZRep!Be(;&#rI(yQH>;da?L4o_v(P>alp};rx$sYx?D@|H-^X#+1f!i zrpt~9BHL&_YVInC^Oly=c;38v*B;HIXIs_FVob&d=T79 zTm}3z9e(S)A5eYP6R2Z4sMW}J>nh00VFh?7&v}v7E zpq51kU5o}~wXF>GrAnQ-+>P7~+M-KDzZH1+@} z>%+99QP)17g-5ARn$<3jTZnpG&W*o|R$Cg+Qv*}*BMzf@)LyqLUCK3(D8vL(>X%_$ zUUH7)%zpL!S$mX2l7sC91bM_uUc0z^;dU9+erfmnZPjkRyf9B6=43n8sQVm(`kJ{8 zfk<~g23=gxBT(o8`TLbZ$^=M_)Aq#a47p$BLj`8>LE9*acoYQTq=xx)^`E-Hg49WN z(B-4%wRH#_vL~J00DsNgv<>02jbJyNOMc(={tWUSsKhzP!9$YI4Aj~khrkOMWuAV~ zOf>F%?qE2l=X8!_e=GB19gzn^{ad@{F)Z`_g@SKbJ3@M!YU}PA6GKx8I zzPl3)c-$89ME4lw|IH36?La*>&f-8psWt{^hI-2E`Yq~JbeQM zx(i@6(7^3|zP(!19+ict1Q5=Q)d*;Rs@B#C{-eOu-bhhUGgwPS_Y^TUZvSs5Drl?e z(9sj{>1-HtYDv^CCQJq&tp*j}yz4l*9(1wD8rgXohmzvMf>;Q%ZyHj&*sus9!jKB@_3m@wA!c1W>X^x5)Lt*Eg#ic} zdoF5djwT&gnW&RMRB_$|97HE(3lm~0x`Ui(l z&_e8cVx ziGx>_*$rl|4$zPL%){aetZ&9e~_d46L4i^yKFEWk!Y>hD>x#R=gc1^*zj%xV=tPicY02|C3#Mf@= zFRA%N#{bIB`ytY;%2%hooz?dSo>Was_pq>$vS25I5PtNVc)x?`Y5$X|U(hgxZ($w+XORB-pwI7QzW87Ncl@yvy6WcN{47#? zJlNYA9~fg6xf>Y>OwXh0lFYfVQ}02li2DtG{)AfZ77xKJHtNm9q2*@1Su~rWj8bIs zkF|>G|H`s9JvS})is$#z0n9X^PIOY?$KK}7!0eZJnYq}-VmTU3>Z8VD#TT<*m!L`K zYbUSg_Y(L%?GTY<9Ka2(n|}VYRv&J>bu>EocD_KnsG#4f{Q1K-!fv}3WGy=6qDJvd z+a~(}e)T_}@KQJk%wT;Sw`bd^Qdo~8jMB7tys*7wa}wi-5O2+4dbwVgAEN$l*cd%6 z@YYdj0nN4T?A(2e?ndiEA|S`M*rZh0aquFTvS_fg=Hbg!{Cciy^CFXxc>tHn+i7V^ z$5GnF9M=+y3rZ%}&cNRRHV)UYCaq~M&msaXVwu#1`$a%Pz!sJ~M$yt;^@307ZV?ZC zOrQfNgbN)I2ra;^wRe$_cf&?&%wtiA7IGoG0pDGRR$o3=Y@&vG=r>th2l~xeoxx<2uW)2kz#+%vN@H#Jd^% zc!NY#LP=!rcSPRMPU%co4Uu>E0+FerjYn6FViyey!#?o0l}c{2HFKAitmQ+wrk*zD zs|*vuz4Ukg1JiEJRm4;5z57=WkiX7sjL3z8~I!pc9Uf!KxWkUG!!HzDT-C=tSb>$kn9PlAJZE8W~Y8DyOH1_PR|jN2&46XfPD5|i^`TKb{)?luHc(y_=LUs4^QPE&sE{-4{R zkoS%2phF%DTRWhgI`e@3TR3mQv_#08y@GM_WRk!qFQ=tJO$;o`s{$$yk@ApV%VF^# zcTQ?})~0T}B)xY&U?+YHEchKD0s|u+wUtjx(pF)T3y;Rm+A z{6rtV9|--3?S*e|Xh8h@JB!7>_|wbxe4MnaK5?9b!2SFKI@rJoqa-C!%Xd)`7h5OS zA^)j-F3)W1N-3DB3~_x?gNL|Y8z^q-04(`$=4!nvg46rzLUHgGhEmPj43~VAFRptl z)vk@!a{#%@O5Ns?f+8M-oGu`cukpqK0MBk1J6s7)7w z!mo6AT)?RXNTwDM{N?`nIGCzw-k|w{%|MW6UGZKKM(Ss7L$It_G8+i8)<4z8vx0+= z*|&6ltyBJ1-p&=3eyy$kLSmeRok}0K5mMsOp(0J;8S-H)w_(rSCozD~kqoymXjcw~ zz2G=PocN~?<>$zwV%xVMHOil|`A$Eb7&sSY);3_0pEcW$l>RIk^=z)!vv;ZrXi@f+ zqs9^8vk9<#sl$DDIT3IOwezlMtGUzKR$n<9f>D}{D*L{t6nC4*dBwA8x;yvQaX!@S{WKPxia znt)IX2c`H1#adbWAc7q1mvf7+(bw+{pD(g%V`t8G%r%9!LM+lsOr@%R`4uCZCpp`? zs&7+Qh=G)~W-hu_XUL3{P|aB)qrXdM&2m{e^&dlS-JCF)OlYF zeEWze&FnO`0%%{zH#>FNIyo6Sc|(V9d|d#Kn!HDTJq_N?$+zgjkcES{^C7P2d+mj| zO>pYz`n4)^GT~PtkRK~q%7lKPGhXJ~UOHMT6xdn&oJ&bGtDBGH1YgMhgwj>B5eo(B zKO?MTfTlq`J3#T;5?9-$eu71je`^k%A7D0mb2N3422xlC@7sgWpLu9m{sY%ud3E%E z#C>Tzly4Yr+Ej{4ma>H;OB71hDNBl&5V9*fSqFo`REWqHBHLukPEj#oh-5d}!^>nD zV;y1`Gh^mG@Bf?+=X^Nl^ZDZU{^n=idEV!}pXa`>>%Q*$VHHvAk2m>3Vl{lqVpAS^ z`pe)EWmcIvM}{desI8vrKbv-V-?WjdUAO~$?E>}kQQYLic^&{yTAyz}5@W{(N_O1U zeU8@j64bfx+ZeuRT8$6^ycwJSI9d3#TEqL>7%sTz%c<_z8*LY%gQ-01CI4#=hukN^ zM2J)TfYDvAQcwQa>EQ&L z#n%|C>p(JXykLyuh<6jme%-6p)^#sZj59PklyY$Zx3^EW-FyAZz|0BTdD{!#K3}C< z-16t(OX{ubW&?(8<$EM#FGgp91mrI8KI%+Y7xA)UQU9_&;!dS+JD7nN;WD65jjsu(_t zO^{-aGn=Tb>+9GVsAu9E^xN58^KzGyDt~v!LdqJDQcR0~9z zFQ7+Q&qg+I0y8e-PR|b2*AMhA@T;-cgF`2CoLM*Ohf8Po)vafE>r+>4pmFPOEsJpSl2n3P$&VHVlp1B+U29o3t3dC$}Lkf83>Oz1e?u|a_+xGU;cHIOR@@_6g0fNXpi;(PKSjgw`b)snF9M(^}bU@P2n=|Q9cSNgDUP!2%tu`|72fUw78;`AUEl%`j5IO=!_B}Zw zB}T!Cfku;~_4VUSKUA+fi&#H3<(5*ss1BAd)J#qBFYg8;8jr9h9@fckCQ;B4+snH* zFWb5}vSu;Q8v`3hp%!>A@2wJGTWx4gNH@I#J}7f&09erE?ZI;EeW6M43eR4>!W>77 z|6E@$-YBtv`a>8sOeb>jyiIDD4pLK%tl|INA|kw|4564miZ<~l=JFk`q$V;hEqO(Ke-L{bSP@)lQyW5mEbgO+!-u{iZLdtm2A`JQD%ONHs z*>o@s>WN+7ZtyQre6K`*V}t*WBWrY<*b7sq#fsSu#U zwu4vQ+3M@<_easqt$}H*aLmXuWbE*eRW^{c^#u1O{5T7L=VGIS9T{ zRgb2^K#Kk6GfoG=pIi?AZIE?h?T{{NP|r2<{q#(75M_0YTFEvsl=$xJ#qAS&c#=yh z4dOVFdpWNAikBnO^t>`ZEs>^PE2;!Pi5iy~n=$k9G9l-YQungnYz7$s})|Kw9rx7JewDoo-DY zT2J7T*8rtOg>v^Bn>Bt!kF~FReLU9#*>3h7xq9$yK@c9gRb^EFR4ED_5W;K#eta%$Npf|>g%0e{Z;@y810AP zkI7K;EyfCHlExjm@^IKcfEz zp%$N?`wl$}@1V{Scv4f0-5NJNW0fY68x-(}^BXj7iJzw-BJ#hMjkF-tw)py>eSJN8 zFNYm0=@EfjLO;5$fZ5guBR(cb#7|9ag!rMh`z>w)O0?9TPwv1Bl*c{?$w+={@g zbA9#Tpu6OqlN!>;uN=D*Dxzkq^XRtzIi1_Bztz?Xg*&uTW?l;w>)zJ6!_UK$0zv1L zvq?&9Y|9!{TfEi#lTHf9^AIp9i`@S8H=hZ*EAy}g0yddF z2C-bDE!ic{8Ro6wGJkPAzJwDEK1YE1gJf$qYSsVoF^kKeP+J1L_8tsG4_*a|mr=3d zM@#u$ajR2j7&ZJwn}Awj?Xy;Os^+fuuxt|@y3f!g?y$l&!?~=}uA8r7`cj-KbtK%Grcz3I zW&M+r)Qd5fReEa4u(eXb*Ot^8f}_|QmP&fE85LH#cqz|y8tg}vDkxDe9sEZr`pV3~ zv2$o29APFs=#Mfaw0n}It?z<4OKY1ojNZ7=9)nAB_pA7|dT3l?U0chjhTj&vl~rL} zCE#8PNA=TbaAP5&t_1jM9%(JiZ}Hve$vK~XDWoEHa0%ga!{KuDn`%bP#ZZM-DS5^C z8SfW?pdRYm8D`7FQD&0*w|kVMt1JV|+nBJITj8ofRp%{r{0RI))K6rzS=7U!9YA>= z^;lFZ)|9MO-Pff;tag!S59gBbTBC(IozX$wLW@Oal!kM8>nNr*KzEeyKI0fzDMGze zBh?a}W?)K5D(us!iD7Begk_(;QKIS?*%Be2ISG|m3XSnj{@yJG42hN+CNzV{-45wyEsntS>Jhs0sOK4t6fGmHn$MGYLKVY4gP~jcPb2E$ z{DvVm^@t@E{f&2Rv~oA6aN|e9g3ZVMu{AiMyiD*IcG1pBcb$$ax2sqH0wqGUkU2& zs_PB%SPM(sz>}aY^!)Mo@%39oi?oyYBIv*&{-wXY&CmEg7h!!|t(}Mw3^$+(M%6JN z!{J{Ya2?48;w!_nfhja__wAFVb&AH}N-(zXBK7B)7f!#d=DWjDa3jj`d$>p(9#6C` z!Ak0WY_roRdwo{JewoFoe9qH~#l?vF|GN0KZBJ8N7yRU1xC-NaW~@WeXqtZL#I*{W z7h4D!1XACArRc5#H8x7yx2?dFhram{L$gpX7X-0{g;x}f%KjA8zikbh8uaNb9RjIc z^q%9oiX~0>yVpUJ$Mx57@L=JAa!`FMkArIX1UfVMT9G1kC?ws^iXrTAu*m!qYhK;& zL%^5ttJHp>jGIh(>t)QnKTUUR>p!y`XFGIS=54>6MM~%`!n6>q<PJT zet&j_Czoy)Z@Hv5dWqgCWc7XdK++OMSph(^Nyf)+?O8EQ&hyPlT1&wUm{Gf?0 zShSCuweP>@s<eF0`=$nX@3)4wgvrHfNpJk7hd8Ht z`THREPcgTXBMwv2>{X)rUA~e^T%W}=^+~ob|`mciZZgc;wj40)q9D7FRJ#} z)z~M;^x%4BRzVxKW6$RcQan=HpN3(0rC%38@7w+A6S}?U!kMS!;_XuU*bZQg9pi8X z{FV!Os8xl)=M*Hwew1??EUEM5tV=Z#~II?H3;my?8A+CibbRZ)INVSy9aUe^K z&d<-_@-b@bj7m0?=-2-Du3oEUK9;)#EHL>^>EQW~pqaw?T=Ui5(zk4d=mmv08*KOb z^vD?i_b6EF`%>%5yn$Pk=Qi(b5XUx+T*w(QicnspfC6hST3uLxx1y@>s=_vz+1)Rq0&(|L} zuTf3rUzc42pm~qXL}5ej>-pmi$>Ry?>R}Von;?dEA0;_U;Cy{&P0b zA4vvBf-Dz~55kmxHBaZiCVU6Yc=Mjf-gCBVe#vp`T>7q!IKEVduWFd+tSeA0)}cQ! zSl_iFedc^zIeeP5iY}wWZ_V`*7xMNDX;p=v0hI>Hn02nnu+z0BlqADzpL&|FkgcO^ z5ymLWT+MHwKKD#)&090@zJAaL-}__eF|0#_kIxbze+Mr0#}aE#^SBx&CVg(lT<}P! z|ElUQmc`FwNkcI2KXs$IW7v%0)|k(q-lHr>?MFBf@^$60lTXOblo@Q;l^{2ARrK}? zhJ9aiR68P@zIjZ8IM4>I{_o(8567D(sn-I|LP3pf9^ok$!yfAk7HW*Z5aNGlEj(`^ zSSA!P0HV>%e?pY05H`Uh9!9Sk1EgI zZKyA}A3(NHdR;1%%ebm=>Sj2G!=szqi2|<`R`{?A^ZuE7Y!%eIU7+ zBl8aS2hjWbbj}BD9fl_Q@JIGrR5Fzqk!?uq>0M^*kF=TN{ywI*JmT1UDYFX1TMsl- zc$dF}Y#+YmI4m=sDcIoxDd7=ARY%k-88vR#;|;PAHPHoO?(448Et;EpoCL|r9e?sw z#jOH2UrX%CRhIQa$=y0K`kKBGc9NPbjE^lTW5i#Y&9=v&h&y(PC$t#CZD%lDnXkuS zyaVhSR8|)-+Hjhn<&~n5#rXEv$WfJw3b^}@ zHDPIM|3#D9ql;)93j{@5Rjd2Hqcu`A!cbo&s0(l?OkRjc>gCoUwC^>a!Ya%?m?oqv zQpKiz9Gava@NxX*xPu)(u6CPT?{L|^ku>}s;I8fE@dh1)PWpZSlYRcc@KFs)Rul0Q~iRg}o#;E1>(n9-Zr zNYhIF5)*0aQ_EAjLJ}E{>P56S5z^GTpad*4MF-`Q2-GlnS6QPz^%v=`YeI)$&*qzQ!HF&JrXP{oRt9|QofN;av+K4+<(z_R2tX>s4KQ2IDc&A@YXN+HIPkpG*@LY_Lt zSYSI76c!@IGsHA`+p6USMN5bcVfN${erC?UJAGB z=UB9Rd%1U(a!2~K85O<#;@BTH|CV)~I|5{D`CL}X#;j|~-P~@OGW)6c8oSz_^y>h( z@>_naLFi?LZgZ`pvxBErLPPhRW`%^;x;5M%a_x;qhgUK-nCd5Ap%iZNx>qte7H3EHgmeB3 z{*k9X9j0#D!a|ydsh`#k6wc>=WQmu*?*8{0sdK3~Vl$QQ`oPDkeV2q&X`f%=DYS#y zP&lm=*uOw;EByj`$!|YJGK9$^@E9nbjAnA98Lo%CWY+JKC`Zd=SYIk5^u$tn+``=^ z6uNiqR}>aR@ZmWiFULJ(AyYoCpoK;A>@ z@G=n9^YtWrtGxfr?0L2FP>~#m*+5+LWdivmy!wz&-SR-SK$|u5lIt7QlcNgMv*+1r zo@rGweAfTPc0qUuw!d`He!7Jl!{?($7uIiMMA03KG_SAoo3El@AvdggdCX)1&lOlj zOmJgm2ADS4lt^7p7_PJnvrF3$9rT6s204U;4;i`J@$iuGBmEKz#r3HPrJ5$ZbNEx%ZO zV+!2KB5MBJVA8 zHPr5g_w<;)9A#1~hh!9~{&j+D73Jv_G6e0GYJ#O(yJXA`rZ;fG`0tP^RZRaEiLq~X zxL|;p$M@paE5Q0ptPS+aohr)oo|J@hiLLEv4)2JBU~K>PIG3S;s1qs=A49~!aRp#1 z^BO9jr0@guA5ksK60A6-e% zM-Ucne8b2nJ23UkKgua#?HvwTjL}PE*KESLiZ8VH0=@AukEH(jK*9E&zkCc2|HcM% zc{8TFSui&dY^6S|kZ9^_ZVG$$Z8ooOJ=i^Y43>pr)T-40@iHgxf@yE=sADC~CNs1Z z5G{CnUfrL7!FGPnwbT}%SJu+)BGzY}JJ}`$sF(52D`B}`x5>R;i)isZPblHe2OOYq zj=xhsadGCZ7}n|`fuz7g$lpLvB=Tsds5()~>7(}+Bc|^Wxr-3;d<&<~$3OD-p;q@K zFcy>3hnG+d#ZRyT^SENer(~*!O2|sL2G-ulY6NC_qGA&9jdjMW5JKZ$=FV#~gcoA3 zXNGDc;9^gDr{Zz=WwZwI_I_j~`0OD7A#az8XV=0pT=?F6w9c3Qp&gM+DOZ&8o65nQ zgC>qCO3*tWAU@%8Wp*-8t7}_==W7&RPvF=G13klcFee6YrVd`F&ixF~cbf)R#0Ue| zex{(TXY=`kyPMayoK=e6qENxPaXFmZe;6YO?SdE z@CYfC$5et$MbPKVfcfD(c!p-b!I-)Empl5O71(uI=L#+HswK1L*Pz|ZMN`mE%q*yoF#%f9j$V7gtao=AV z<+Ak>*fVN;EKs+4Qnu2+MU{Uf4x1rWP*{5ID~opi@j$!ae>!5+Bn<(-`4-h^mG|MK zzYqAn%t1yn&xUlxs(5PUR)>b@92C*5zcAUXz2mk5GrF{W&m#=eT2edUy@EmiD&Psn z5DcX@=5YxY{VRnKd(Z~RKO*aDyqKg)`P6LhyRJDiI}-xZ8PV;&a2?)^NXWR}N#3Kt zt0MHqTiC2A_@DY33rAHp{|r^v*RdRn<=0ZgsRcXuFO8uY2V!qze1*>H?*4>ST{i_#a`Fr2khg!7?h&49`;D`;a!cY^jg1^$x?OnDATe}5m-fnjR$ z>5EhoeY4XgkHVUKd!ol4iiKwzjG|*Z#{snDTh<|R)d2IdgtdH>3h^5yyhqEXs_#yS zpU>>$0oJ%Y@3$Ur2&uDMwWewjTqP%>0+_B}pUMoN!|pqeq>}n>`~~tL{Z7XbJq4=R zQ0z!t6-C45^WS-#6jNOJihAirUCyN2$48+Bo2NtarBCiA5}*9r@X6CY`K=7pxtQ0r zJ}JJ;`11Q5>bKZB>N(Tcm%M&`^d>jcG5T(=Cq-&?N2`Zg_U5zidmx^h6w(WMC>gE+ zE#43%>dmwN{CA;)oF}C?o60Hp5=knp0fd##m`9pLCaP)*M97(pqW2EJ2Mo^52`oZ5 zdA_T;S>ph;=xylBk24c~x`Ru4Gm$IkDwlR>t=#kvZowSkU$nb$w-qu@-++r$^yla^ z1zt+js=JENs$h;6mmR>?*A0{{jSwr=(4V8+{N=rMwkg$!7WU_K+2sI9YKv0S4}`pY z=h0r~T~sN0QTIo?&zQ%eot)Jl39Tzr)UwP+|N6_&3(#g-KV{~vYz_XoNUg#MHcOkn zz2re~z4bQa*EM>lR@if(x5D%`8LiXtR2q?yi#lVdMPC1Ry6CKa;OGxP{&aV93^v2B zma4w~d=}^B|HFkz>O{6>GBj$v{M1H|Roo|#E_fSMf^Yc6P*-ZqNOlFAE%{(EWb-=P zDuVzwR@)51A~S3h_#xt_o~u-WSX9Ldte|q?=x5NZ#p?jG4pa-7&1P9dK87(JAr4{P zpmrX@A6^5p&#@V=ggmSr$j-ab5u8QkK<_6-bZMXxbz%GQVuUW($E^$5i+sG;5Ktcy zLW-y%<}H+hm%eS{nMY@Ff%@05eCwbkL^;MsFpEmQ=Hc*Wk!_zMad0)$h&UYn&(zr` z4LurMvlnb~wE= zPJ)V_mQkMmc;#Rg!`y!N;yU;~C6XbZQO<-Fd_aC?e1xFbm+zv45}S%eTzYiIL1P*^ zfLY?FM9w#SwS(c8TqISaL(70(B{Vtv%CTJ3>)S_p0?sn*x34Y~vbMAJ&(*OI{`D~G zx+?i9WV@qa=Civ|<`CEU#w*p?ItzXLL4f!iFT~-{RhJ_*BL6g645}thf$HAMY%HNS@7BKCD!Gj*+Jo_ z{SU|%n)BJHFT)m~W&Ta~K^OtHxV-5OGoavuRGuzj^&}rh-o(Kr6-;X>@b_tRqnr-r zxj#fD_cT4#s}{S?9^yLW<>*&%7$I*RMt4o}$d61m0Ub~FXZqk>S3cX^xFOkCc>lj-u!FgiGCb>xvdPzYqEp62ypLv zk*BUX3wUI|d&aCdtv)NfVM@652sFs!qoZB;NHdNNlDsa6jfrnv=e?Dm*Yo8x@4kDz zo1=F0&0p^VtdH~G^!o9&y9+FER*T&hoG6R1C>hnRWM=DBqpDBVJt9-HuzH{5)Y=2d z^|6K5Hy$}nBsc8g>JJX|z|-S+nc{D=JEd(wDNZHChvAGq=76uL_BeWREY+SD_V=O8 z>WhC{T1GhL8H0X}t#MfS%$7P6*6Px(@mZ_MF}~FkeY0!bs^tT3E#1D4`8`EC>Ff@? zSQTk`k?Q3JdUF<|_3JxDU_Gt^@*Z5MskwpAv991M)}9-&j_zKT#FEeVS^>IIr|X3k zsa;wg&w+U0B1a{tGZifQQ!8LH%K992%+HXlOFobF;ch%gv@8z!KJmzPD?{!Q%Bru{ zgtL;PPU^HFxkC$kIFmKr;wZ zi$cJWBPRI8zj__RE6&uZ7PkAB46^_Zbqzs?BXX^$HAwcFo%bC3JKreuO-*_TcbTdX z12Y2gy9%4XXUVIU2^TGJG1US^F%*WWa{Kc@ayvvlThs; z|2wBQ;Q8HQ+3sYUq_#7-j@?2^@;mo4lZmh5)lZ2FZD?}gMZCS8-&zK%My^FP> zz~fe~^Hn`yZr5`3cQesUm)-a`AiLdeCZ9Bjs9WZ}X(HbLkyUoI=;&|Q_9KZHEI|{d z99;&EWcLK6?t-zZv(-A_cEm*b-fo;LnaVSGuOA|S@_Ryq3BfBE%s#{3ziblZPG-oD zhsM%NUab9Oqu_Zv`cHL*H-2M@BTY#W6>Sl@USiYWi95Ad-1?hsrB7H9?Lu=_K7QOK z{$;`ESOwU6i38^531H{z6C=@Cj4^>FzN78U8a$DeSiK%(G6vX~#VFX`TWtq?te(@8 ztOL`wQyDwir~s`oi0ht{vXnFcPv#rkLBPoZ^I8In5uLnT?Z@e8??%{BsXpd6Ya6-8 zWMs&ZRa;YVkmcIX1OtALM@Ra?t4Cq^%EMsOKy)Q*T;Rz)fji{;H%7mA=c3M*f}=hd zSsn6}2&_wW5^Q=+}j-XA#02Rsxwl%gmW1hml@nU@Ip=HJ}9+SVq9pn zgr>}t|M0~|8Vepb)H$;Z^q!!78fcb%#FM|uglXv9!J8vQxaw&aQ_}^_z~#M=V+F() z`KL|SsDKAjZYvmU@xoFE z;M4i!kl=5G2s`sWHHD1EUcVzO*wGx0eP$Wrdy;DJ{hL^cO)`!)Lo?}EJu{zeQr-fZ z?5OGALV;DVGP?5cI`>i--&}N_W0Ss)co#8YHNDHZ2$79SYuo<@A5navm>ZCt`EL=4Lfj3QQj~D<%uG-C`U(N3v9z)B@K6su% ztLZn*`muyD(VBj}ujhvta z^^#-NJ%K7TPN3$CAsLahIP(YSWZ0P-mqR8oz8JAC%RDr4s+Eaf#2Ff&KWPin9(yM* z&d5C@;ls5{|AA2#9WOk9&ayo~wyxhW6E>u_kP~!kt3h?##X$_iMF6Z@WLaFK2DKL* z0P5FQ9Ps<&2+xJ1DU6_mZ>jaFJNRf!@KHrT znPc=&U-N(^*zHAj6wOOhq6R&b3jWSQ_i0=Ql>3{Ti?EXRO^<(`f7{RGnnRn@LBjLEDo;UBq91ujtYhXOLWCndE_GaO+un<~lW3BV3DS+m zM5*XH|K!i$jyJclR3#X&uQQUE$7pz*4q2(5A9(c>Yxs%V2iADa9V}=tGNCjD3eF|o zk!IqZSL3?!L0Z}99g-%Uh7rjUXBStbF5R$p(Badb4Pf{qR`VF5aa0wH*RYc54A!8kdQ_|1vVUu0{(k#vD|DPz zb7LzMw##@YQ7Djro6|bk2XrbB9U4ISqZm|x7t>W_2j4{RNmYN(!YHuO?G=e@x=hoJ z&jRmJPBKhHZ>0R|wS3mZP1Mob%;68s&!B{ZBA$}A`h?X?HL_=UezhWnlEGdvQ~>B% zEjUw#z7Y7fN)NB`U7-&K;(Qwa!L0LF+@5%{;BieeH$XkRjcYWjpWGW!(ADHA|KwXR zbwuT;@EDp7HVT?pf>Jj3a!1XWCG)S?pus~-PP3yse7A+S7ur;Z4af39NoSuobF2!; zMROHyMGC!S}{yIww|P)D~Hfn0<^y((B1V06gmN zLIMh>>lW6%7d6Mo6}+LZ_Ays$A4>jr^XOaDrPYkcf$*I1u6d|f#Gcyp;djj>h_F8S z))I=%O`W)Q@IOt&P)Z<4Z6cu*9V%<^4aVB4OJq{(@*6CTjwa~-1<9{B`M2*}p6B-B?{1@Le^B$Czt3kSfaC}v5 z@-7U0i6h|>==z~ls2Gh4mCzqa9D*r&1e$wDPJ!akAV!C_LSqJ6xYVw$E@6|<-KrFr zy()Kw4VxACWrv22czCo-kXd8h{xF+fGBkhv8m&e|7ZO_7s81etAv?lTAj^zKs`xOW{t?q+h__<*y*^v zgq|3Ir8K7B*>9K3M8l+4uYtw;_^)N?MDL8iQgq_}BjJA|uVI516(;}9A%v>=h+vW1lx+?elL68XSZJeY6pV<}opt`Vqd zkTLqxByWl1_~7^(*WT%WcH<|meb7COSN|WtOjzR8|JB%P#CeW`&rMlj{WEE~`O}zLtRS5?g`P^fY~f zJL&Y#;SzvEcwe!4DV|lfp`$Zf@OP@$*CL0X`y$e&?(LCXp`%wlHd=*8k)^;t(}W^Z zApL>fC5Q`&^|6$Gk9DMG>TeBAw%Wr9y5h=nk-I{P8}GREK5uODwBQ}c!A*j{Ze0yx z&KM8;&jtZA9|L>Q|n!uQ|MJ; zPDrM}E7-P|wnxHaP8ZoK`s{7^k?bxbtL1S*8$S=@njaDFP?Gj*KkfHV#gE|QTUaS1 zVpouBFnw<5QL+4tjkzBw*{JY5qI0;XO;Lf1(4j%hY<9hw;H* zM^(@0(d2MKodn1nW;ZV|hl^S)`m0XiOAA+E!ZJDRSJxX)iz*SK7Z(PuhN+)CPq9^} zMPHE_Lq|m5NoqnNzK+>G&kn@ zdSmI4DFey!F_^d0f*P_;$G##@+2+E%C$}saM!8Eyowc(AY-+m6LWLW16^}337jHa$ zai!oW#kMbXJaciN>S{&)hhNB9ofG(NXo?Htn|$v{379TGi6o?F3ibS@aVlWm?NoM=W!L;Q+bmyp;(G;X*)_uic61BTqf z+SfPZ<%rId=f_-EFeZF-)koyVL~l;)?^n2C82UJbij2@;T$|`Eq|v44dCIb9KU`l6 zq}T>zjQ@E|mow_|e1IUG&2k|oX2UHCQIH#gJ^ltcbG!Kax7Al+fhgsj3Xq@NW2y4A z6zKQ7v5}TBZnWsli+o(hJA@lXKyD%T*X6baZ@N;9Az?YoPSJsa@9=mxTRpNQ#rG5N zs}b@L>dzI(fin^RC4GdH`7iSog_U^fT)aS#VsIuTY@~&=uQvwoRiRiL^8=9jG@Kn$ zpC&+il-#GQdQ3@sTzqcUkNf1FNatChkHXyZYJqJA#lb{CAzu}my8jBmopG|&1?rDLIH3pm}$1U-$s|AWZ zf^6a;_$Id-_`HM(-TIHth0||c)AzIRpfU~ZJ;?wM_R)Of^YEMC_A3ODEKP<|V+zry zNvC|Sa^kf)|A9r$$KC{ko&f<9aqd7eT~;k0YjU~HI8pt%7RC-l}lHH{4Hdg9t+>Z%51>) zPS;YBuDJPIj0JfYHE82P+QVj^4Z@-?;Hz!TXhv0JS(T)u7G;I%?w{3hjdO0j8SfUW z15&5OPO2@b?Yv2R5rZ4PUw`>xI&S!*W3~&;$kk{T>GWVZG^s44J=(^r1ru7*KN3aa zR)N$!QbM`ki?*$9@hiK46{%f$a-2YU7@q+({w3Ln(D4piP#6HN=06TzZd|IpnNlC& z3=-nY=y>5(nwW7)nQXGzQd4kLIoOAQ`>51bhexH$Y54rfJ)R^u8AuL(bf|X(eIZUu zJ>-%#*lIxbCN5fWzVkdUykWK@;#w2#FN;d{DhNV`Sc1pl;6L{d(g@rnQaY}S)VL(4 z_oh32xiJ{_oMl;I|Ew2vOFQ+w8o3d6xay-jlRIU>4B&AMZlx#94pc$;5CqZ{!rme6 zYV34>Wi)xj#PADdzC(;`E~L;LPE`?1&yzajgVu^ryRkX9#93i6L7C2YwqH+4B4#~2 zcY*iaSVi#Nw!0q$WID24twkkkivlRQ5(?7$#~$>jsM+N&p0WTwUqS=6$bc=H(oW-B zdRosWsrs>~ROd>Av}*oZ>B?z4OX%eSRW)SQ-26q;U9giik%+xSnS24pv1rccBOa#n)0|`Kt-N5%yf3o8vxfZ| zN|N^TN8~HQwJ7j3wdl`QRcNF5(xmA3YMVAR2kiXU3l->>SjP#12ba4u!O>D;h4WZ~ zE45&KMKpY!{b7&BLV(S_S_0wWj>rUG0QuvswM9+}(9Oy?_H(;$x=HH9_nQdi{t}vh z=hYo;Q99}JNZNl2!sPz{{@OJyVc=EmYB2e;`2LB7#H&AvreMzUkA6#<+}|ad=Vv!! z<0oIuL0{ywUr_i&j!zi*_D8|_x-h;1^oiBCiDNK0oAT!(Z9+$>Z*Z(Sl$U7puqOwH za0gotAw&JnZt{kg==#*R0{w-^FMbPsLhX~cES533t#a>jEJ51$OHH}^Sj0!xc9z>E zy4*)VIR5w~a?swVg9RxL&DYfuqGI|Ydqo!pEU%c&B&oSJG!`zl_6c1vCA=JhZL8kk z{}>~&#L8GI1sY_iO?LSZ$n7 zRJC^kU;Zc~|4uxHk-clTauZvsQUsd=n51&K;KPHkV+kULs<9x9 zE|)vC@qkJ*GPq$*KD7>FiFShmJ>d4(cum}l1Lzs{USO6yg#5n0yFZ4;o%^gqSe~s( z|Lk}b(W(9TjuL|1savN=0lB~CV&0S4NPG5t?VIwPT6E!cCQMH@Kpsx6?wQv4cQGHK zEdGM~Nv09Ghc+)(6$YdS?1YX3={0sH8f8t0&SPE>0oHoDsCTB2VV`=aPbgX7+=>N= z^(MFC`WML6PsiT?SS_*p2O$^HZBhAs1)ZzfV9DD@Mt8rQL|R>#nc+NnYXJCXXO}-q z{+=RNbG_g8y?Eo7EpTl@Bka?m)n6GKO{A^p)E@Dlz(4yIKiz&?sfwRjYb>5FTPZ<< zX)Q9j?bdD1k{b*mD7*Wwz*>6eA+zR0fi*Pc_dDVZoIp>D(_oTp7aflgi<(s3?AKYr zILl?y;c-cgTNzf#K>8nrum=HPh5R*R&32%_&*Pld8rn?m6oo>Aume&w9GViDcHbqq zzXj{O0H$MlU9gBHF0;bQ020rzu=5`P6(T5_xp)}SX{b~GAZi)S>^d=LK89X=eCulG zX*xPiLMb|ig?XVS=}I$#nL9?Cb3O%S5;}P3=!oM#UVApk_+|nL(FGH@Vef*dCD3y4 zPFh`_+Gd%*j}ya)c=iTzAmN-!vm!E#Xo33`_qMT?pcOa~w*$POTDL(X#%dsH{gpje ztXa&OiwO4Wb@cRYrsT|?kwL5XdTZbJ1@eGvsy-=E_e*zgJdi#{n0q3~;*3zU<{mm8<}JRMu!DKc>(cA`1$4cgD#(=Y z&qYZb<(^D#v{qyjUI&-I+-v^w1uS_iaYD)#JR?dZfNyycHXaERuLYN)fAJ>iD(&EN zlO)>;!%I3xrVPf>`pH037BHR58$@f6ag}0wz|<2~A2cp!A@)dII(p04Ly@`{+{xZN z>9yU9)vccmc3SS*em!Rlzeyv#S2Oyq%alx2`<5)|KyH=cegZEBC2tOX4 zT67drSC?PqNHoRHtCswEx{Fc!z_2&F@b-8L{rA}p+s_Lha(e+bS8uGNO>eH}juaCn z)>y3p*!CeFO)ge+Oq{q3iwO;7iq~NnQZ-yWnpR6cB_)Dz9h}J zj!ko9VR}Wpe!}8!m@V{Xc3R>NRsHcEV#-T?-~M5|gUNFZJ-B{o4?Sbb&bWBoIkZ0}g)&nze7I=YQ;2Z=I_sPji?;J;9c4%^(&8oXUspmlCv9@74 zlrq@A>rRflvXrXamD^Yi9*OCme*8xts+g~SbZ92vhp$Ce)DK@I!F;D}m1LxuQpgfl z)0G^^l|2sk=OlaE(IwT82wQMfl{tSMn|-u4op2GUd+7kfUP8C{_pYPA@@-oW0e_1X zzL@C|Orz&MK5)d++Rs9;=3T>UW{nFA;Gy~Xld5BtJbLY>Ja3?;%xmn$wcj_gac*m8 zHJY6L2IRL%EQ6M$m$PJCzIu1yyTGZXQnmeO!C{GjVN*Z$kf(k-xm`5l%np94r3%$? zm0n^rfHhE|+grR+T|&!w$}mlJXc#?P7*gssla&2=fL>@5ILCt=&*~@Xk`jp_@0Z); z24U>h168D9n0)Vc!6N4^7v=r&jsm-4v@8fD>^ChAIOB8I>UvG+B}Js*61?AglHzWQ z@RcJzAAo-koU6QBni`8K@F>y)yq6q5vrg8NTm&uQ;Jw4?p`RWoFk#+}Z`5zaSSAh8 zI}_X{LjQnS7I(i1uV4mF>2>8!?yT<(Ot46eLa|Nh2==9miiixIu1#Z)y|7@D+H&xg z=_w>%_4T+$wYrb$;Hs{)wr+3g1G$hnqJxb0!3^BRU}&ia#=)C3QSt z1d-tpvVNmb_Gbr|O8%l;N~286%pT@N&*y6_%&Zuy%QmFZg~NKMMV0qFLh`pjqGk)9 zw*OUK!y7D2ARCD@?fx$n4$mCFe}!^Ww?64c@WoS5xThe;t;;PkjbJh7#^7r&9kA=v ziH@kra6+4>vdq~Hf%iVz%NYNHf1^M#m0ajQwuAA$ef8$2yorM)LMF$$^I`}=sy!N3 z2lMnsS|*l6B!>?9T6jno1pZQXl$i`y?Qx|tZO-U}v6rF56!F*z{@s?3abizj$c6_B z$$#rh&CXQX{*piy2@+4(wS)XN_r2HA?;F$JAIrwhZzVih*!cS{ME9~%nqt*gW}(=< z>_X0sEH!1yCF}6-togfVIbtL^?@CM*>d7#xlSyEpjYOe$h z!p8D#2_1_>bVT_Vq`qz|jaYFek0*UHU(HZ6OO5Zvhyxcp0{;4m`+NoItj}7!nzr(G%Oa8{Wo2dP zZ!q?K#gnfZ8A*?eJedfe0!d$xgR_15>!%dkkX*HDcBchfu&m_6>6(lZkmm6C6P>F- zajp#AmhwY8a~W-U?|8Br>ydH05|Z4^%bAHa`10pG+2)u+R|>l31`}V)hv*z+iA?xS zG7uyv@$pG3-;L~@qT}QSA|zKJr9B68A-xsLHVF#8ezTIJ=C}_NOD1-D41sl>%E;P$ zX%z%{fe0tQ`8OjD32~xB4flX~`NWKGZ=Z&g@9Z>pDSOiI}W8;)1x3Mw=ulE5j2%}$m)EA9ZrJR(I z5`y2<@8qb_?k;X@9dic}sjualh9NzRew;@e)*zf3t26{mYt<5c$wP&TSd|iKvNs+> zB(83QZ}gv6H8jGSkZ8r=b4Y5olx~{69Svh)s5Wzd_nJQv%If%6f?;a!EIwipDW8bP z3qaS!Wil2o)L;&(|2Zq{Z}BmGR((KeU|N$sIC8-DLd{V{CkFb&k<3QQqjw-_)Wned zQbWp(#Vmh&cM{yszvr%jboeX8q9Gd~zxF>ky7F+S{NQy=zq>!zI2FcFYN%nnbjAbx$&U=3EUoPWZXPood?&rDh=YGmx zdmF+98?xr@tz)`LtE`H+>Z@`cqpbo#C3#S=mBa8q_@1(bzW!R!ExSNrzlDp0l$BM_z;|F2v%mo;Xd2jijMNx!c}d;K}y}Dt={YF z_xVyKaX2z-mB9$4LeF47HL>x9ju9SPSMyK%nK$RP9_YK?7O*3>20hS{lEHnq(KZbS zZFWcT5@A}8!NQBHJ9_J*+a*k+V|%F%aM~zgANR-M(fy~_!bitu-Yv||t|thuvt|Hd z(@QM`#qW@njVUuqoc_7 zC2(j+cR;!k9CEIm<>>{9yJGEz8P(f%P!^0GpmDF~>L^^)Xw}{&zYY^Ii(*jtdueLzwmbe~1RpSKyx zzkxh2u*2AHnAz^~A@Ka@BH(cI0mm+hMPX@n0(#+7Ww}HgauxBz<=y76gdUtl$?uYy z9B%NmuQ12}LwcnI3UJK%a%p|N@@OenRkc?%s)gkKxs`9f1DLBHds1BWyE)VA_;Zf~ zKveiVdk;i)`yEMtgcHE|(th|$GFC7TwxQA9vr53|7+AOl77JLQ-Ss1_|YmcSCOu<@n5!s@Ie2)+u47SE~e?B1dRt%ykUz!dDFoi zG6~p!3o|JxQCe1zFm)dQt=U}`SfF4u4S3+q}g6Xs4}VUX($?b&s`U2{DK z`RTp_;SQ|tM5jh{iL^vidld!@L$ddzG$0x7^&%U2qxJPTSpx;H7RjJHY7aU&Cb2=vbh{O;sH3UB2`@sp$Rd&Y1n; z!_d#eX}g#`?-`A_sLYlzhcoO-P(}GmIU~qo1M6~no&=6cm&WcF7rMKs!!+2vs{(%O z`qfVz=ALQbs{cvbnXVAO!#iHvAm)j6tXlVIt~i}&zhjZ4I0HF>*H*;lCrAu20x9Bm zBOn*o`iO3t{|)$I{JG(FF{xDEbQTFj2OTocU|gNJZDe$*2mbe?z1khS*?x&b zv})9>Nb?Lp{JIt7dFYiHYe!3a$`*G~ket66=m!a2o>?c^M+XA`mAy_rsj! z0o~zY_+#}x&^L$AbWZ5-yf3zUXkwwPbW=f6*^7E7p7A@@dPFCl!AFZ9`uwf%#^!ho zV)EID#b9Mm_Yc~G*LbX8cJ)pU*LQEMC~`irm?->&U} z^!oin$s@Z)g7mS1$QlAfkp}IVFQRPTE7J$(4l;Pm;2$&vTu`p z#|l~g-yI!?6MKK8LN|ihawXP&eDXe#yLJ%rQdeBe_8|%#48O= z?9Z8HjW<>C@QNmD_rLp(^C2)ms}>&~xZAza=yC55ye zqomv%4lW$dJiW5e&a0a6{J03jX7HH|-jP?LQAp`s8y@s)Y~g=ax?iXfMy^%2wkqy? z8$Sjt2>vd9tp{RIamN$`6^C|F!UggIoQ02E-W^q#aM!lkPPl`~`#74V_MeT%99KnI z(}g~8%A`g|;g~kb^%1hlpq!L&M_%27`mdbyp7MuivGmerjc{BPBAPzt2fLn3%pSYN z7dXBwYhn4R*>Pcx=k2T)Fp7T87BCDh%11xZ4czq99Bw3?x!6re1P(9FY-CLJ?6|Zi z%>b;uw)gvngX2lp-Zgz|PiHHM%~XKGy;YmQ(>@AKFne6k^rU_go8gfv0Gj{B4t4zM zRyA`~fE<3SDbAPZj_##4rM^RFdS&Nw0f# z5Wu0ACEejTy&b!CWe|LCC5(dN_D`gWhz(BJp|*Ep1z7BOuHEnka;&#*dxL zAtnA~+z=jj3tkk9fSy=!9hQ@xj$Z%YkA&(gvlR)DT=$#g_&DgxYlhNkZblA8LFKbu z7Q=ZVT$HQa&hRjrwJ}~;`-fMQXA=D?{pDS*ZTNn3l-HZ349e3>eTr$cSsJoTOLw!TvVgz0mJw-c3lZ*zv=XYA&D` z)QE6~Y{gd;!3=2&%cD)iQ1A@=OPKE^X61aq931Uv1d5s`QXh{mQZSbE;5vgS)?btv zAnTj<_4Z&o0V+`qe2VOTT^%c`Z7cl#bqd=ZNF(W_IG7xtiOv7iX6AM{Gb8%=$@Ws# zG=@h%RhS~i{rGUEROi|ROyN1B)H{!V%cUsny?ph7ku&zphyWZS2al@-2Dh*|a)b&u zc5z>bfH-MbK8Vd!r>}ONrZ7RNq2Q|i0JGET9kUd%;e!7jzh+#zrxiT2!2#|G>Keml z14sDhJ%Bo=Rk5w?j#EHdJj5rG`;GeFuyL+0Jc4CmR!9PE`0K@;C-xUw~n7j81{3jGFt{s3o zF)R%tXM7%7O6>0v$N!AJS=k}KU5JGkEm8ga>?c9vGuv-oMDT%g+?ub^^s!sA5-Nem z1x+QJ9S*0ye3`~%19HP*L^NH*n zh3+^-?hMz5qSJ`&s$J+AyU@F(O=7AEyDUC{ z>P^YiCls;$=0ci9HSs4Rz}7(XhwFQlBb4i?AfV(+YpOQ??sySc@QLB{!3Hh@xmqdO z1iTltx2wDoPg<#YU3qH+>*FVx$wl?!uC-n^YW3(;TxrVUb$9}!)V^<;Pe zIBn$oy+v>cluX}qHHKznL1b7myj^-IjbPK))!-WHb1AwN1H+j`EBL0-+vSW)E*4aH zfx-hse^&u69ew;^)rlS))gt=oedn+%dma@J6zIO$nz39oqG?a7thW+g{fSBqn7 z#S)(`@p2zfh2ez0zMh+wUlOl!q+lsl7NYMSj-@7~H2PJ8>>{#E z3{Yqu&u8o``$jiG^9iKG=CANYq=(SrS98h;s6_qr0>U$9v<}bj5Ih0fH`mi^AKqxL zP)IG3dDa)@Fwt>?HgRh!O$MA>vAOTGjHzh*?Vm(TlalR9iU;t&hcR+r=1;9ltQGsk z<$*A`xsO;acAl+=0mU1?*CfWRMy?qQIm$&MHIz^Q#oTLM7_PrBJC;o$RbyOJ-4u5S zt6QVVl0G?%OJHU;oU_T_GLm@p01BG&q-M~nU#Wfdf{TqzpdU$XCssy7H3QMM!u{Z* zoEh$u;Gft<=}i~}xHTC$3ayC|T2?Y=xTQtB)A;~!>bfSg+YHvlSU1oz@;JLoPd|pN z(bMJ*$ZK+ngSdeiID|M@F!H-i-`^{~C0YQ+B|Ch5aOBRgi=j zIp+N1X+8V$_Z?{=n+{aCYfK+fwOAEBBlv{;|*R7dYc=j*#@cWF& zUi|r?3nl|VeqcF6QuyAn#+<{aH+F5HV?qiqBYX9HV8P)gTk5zWx8}ECSO1{8+ zAVQ~*qM_@*9#?+f`KvE8t>T$ET(~OcjasIWEhkN7WfL%qVo`A+{2oB<&k&EN%@{u2 z)9Pb42Zq)mn5TXj<|_j6{Jo^-cWj0{ka0QL0p2*DvT1>i|Cp;3WG;+bUvZt2p0MNb z!x}-4RcDAE5YuxjAy`JhW?cb0Up$hj*07v;Nmz(FIyrci>Czd3w;a|$%R z-G3l+LUiBr22{Q>(TP780S_YB%isV5??bOw**ZJ-0!TVq0meR)34&Hq7L}&4MH&9G zHo7aA&gK&R+w-C$7^4F?pSsKa_tD$g6Hu!XSn#Gn)|cKl++am-#yq0eXGLwXK79Ie zM+`n&mx*MSNGAsKGVxj|*CF%AzjIAUM@~tiP(y(fv{z)FM?#jCZ0LU!eIDZccPNFPW9scgVIhtq#mjWe;J+x|3{qsfi(-acV<*)e^ z5^Oj0W}_JJU86;43<6C?f862$&y*H8NK!NmY(V1t_UMSil9;I?{pG4BjP|Z%MIcd6 zUFgB7>@uKE_B19Z07h)2s|bgJiV%k1&3IDWvCbd9%jg>8)6NIjq?XMo?CtGO53ax= zO1?PZI0Z+3_@46}EltmV`3{1jz~1I|mfZaqSv{~<0sO+i zT&sxah;w*N-u#E#IDh)lHe}o9Wx{0SUol0?Gv;TBIW|j+C}5)-J~;Fe%L2l`wHt2> zv@H^!)}8Up0a26IJ4CJbUuXoBF!b2?;s>m~;tWx5E?P5ifBtNic^8fPSMseL{54zX z_qDINFT7i=3KNaa?lmFoPPR$yM;lXeT{m(1Z2MGCx`UVANg)6yu&T~Hj3o}6i-L0@0IxX+MhtYxdr!8Ljs0!iHnF@Q=XVA)O}dr@XCV*tAI^a@~4v?-!2&?m98D)2!eWXC9e<^GMA)rP7{SgSN~fm^yZkUC&>RssnY zo$#*zGl7^6{n0>d#{W*z?@k3YppFKJ{*up!PzfCIq<7~bZA5E7=KDpDx4FOt^QIk| z_I|sXD%a5HH5bb@}%f$2>C*|IJ}Nn+0Q{cyVN ze=`bm@MG{I-WB8epwdmQTA_Z-j%O@LAwVtg%Ck7*tr>X^BEaNiVk5-@2C>B(K)us6 znSb|@%``3V`Hr#UPsb+A4TE8tq;@1&6s*MhbcgyJm5ZR!o3xH4)RwxOTTvafM~Vr z=L%O>vi*w&_Sddq>A_#_rPHc)a~&7dK<#Q!5|az(!5pFQ4Dyy~_YBFE_w8iE%G?O- zGcz6yEPc6pDTH1>D6Kd{S}}i)hE(mioIZnn;KF-?R*C&Sk)ZR-2rjM&-dBM$xa22$ zp93D#_ZMKaL_9E{kjUO_ieN$6S*!}^_wKAj0w6o#TS-6Mdqkh}4vy!6N<-g-HJ5|v zFdZh))yLZt_S)ax2UL~-*{W-RTh&)I;rcSsN28mx_Kmrd#jqwO`HLLgM<5!Y z?}yOY8Ds&>(~Big)m38qC8J78^Dq>0<`_CaKA%SYcK$b?n7d)>f;F^q90Lo5w$d|Z z6m+FQxs5rSVf$>bfcF`%CLJSV3oU&!2#zdW!IU(ew4yAC;g`11O+<`GYdoCn@?;XL zDr$xrdMQNRx_z&00syhUE`o%aJe#MFFfiAf7P>KrCZLc;y;3EV+~5;UINq3p3|)Lr`N9t zyx|sWpb=ObmrfcAL3g!-bdNrR5E&lSGM#h$! z2RHIU)yq|*O<1kpx_wqU$Dk9>JoFT_Kq{vthNb5VRQ@Jk!zvpPRqz%le$6TZC7tS| zprk${efFCG^&DGe8uo@^SZMrNYrn+rYrREXfA`3;^A9b?vYRoQQ?-Hg+(lwy&@JfE zS*Wbd)Il3;&6y#q?d#r$R@nY#UY*T@F)*)zaI^7U`x8L%0q0QHZZK!)Bhe9$p+uR2} zk&+H+*LrH++-hGtrUjNhHzec*{CgZ-%e3enWik{RM{zH5UdC)b-WCJ?m;Rx1tM~rc zX~YPoyR?fccOJ9A>h?#-Q!`@bBXXKCoJU?qIx+8%WyrFL8Khjkk`=T zD8U%gUT*gL+${&caxU*_$j1xXO{G7tJoQf*6b`KgTN5yn{jjThb-jGmL{iH~>syzPvC03l zToQw0|07E@e`G$z0z7CgbK9}JW=x#6!pMMKojFms;02QY^4}ox;byK*-5Y~WY`lnR>bQx?PY^k z%8Ve|6fl?82bDQ=IbQl}r~M1}b&DrJaGC+vGg>P(EhADC z{Uzg9IN&xiguWL*)5g$?I=E7fjix05o)|24ZDATAF!#k*n2u0w z3i`|o9lnUVI7i>aZOz1Cnp>U}v`in{Az###=l$}Kj^TgNCh%gv_^n84sN_G0BN`1o zbSe!qIRxH!6+%)&zYi>iHfEoN?HB8Q{`p1(_;mrwrI1iTdGPA^P`77)r zvR%jzdky84We=~=hwkyWm=oLE06tgn{3(ces78z&@h=9Y^aql(gH24S2LfC~) z3RF-xNqk^##Z7HtX!fW0i_81OjdwiBj(FOH=}$FW){y8lu<*n8uQ5ceE#=y=c!@4e zXqu4s%yituh0t@GpwHlt${^=7)}(_;70iv)(GY@eUgA{fXo5nPp#3lpx$Gqf9ZHRl z1KN6^pNZ`>pT8&LA+(2}BfXC)^+(GVrd<+_oN*j)Cfe|izAHmJMT`7jff37D4+bxO zcC-mtSOikdjFjQhyM$M@(QvVCu}bd0&7*@r`0MPu<)EFESWznlb^hdWTLLCufxAn6 zW`G8hE1HgU4DbU+2W!(&FqAm?>XfY`!+rZL50|jj4$&m1_{Z^n$czEwuCM7ziaTQL zE3*=#DDd#waWavdln#PZ~jQC>9E`6Mj!A z3PXc{8Q4B{!i`?9!IxzN5}*5D;5x2rLLZxSh6_8Z|5m|_Sm|b5Abds{hQWl!i^f;g zP#!U?8J;SQ{N-#blr;0kqje;m&qt#$mPeAFKmq6tbIYg5u6!=@;M%=Fm$nFlV`; zce(xx73(LY%~JwO+Q2R-|HPh=p$y>`NZ2$fXU9TAdG`FL9ddQ2@ZzUUQmp*IPanH>YqSx&3|ai=CsD}+{rTo+(%$Qkvs6dqU-jXOOjNWm4< zA7CR=z=G6LoAmgwgJv4>T`G)wR$)RK3#TDIftt9FVU=Kd;y+ph?}8?{1PXp^)cHy$ zzbq(K1I-ie`EPAw7VrSReVde_(Zc~tjJqzJ+OjZ-s*6t|{_X9)YNi9IMqaq`N{IkQ9JigZ6pe zpp$ul@7N@#XK}zt*$8}(t>9y)b-|mF-42=*g|WT?kGxe1>i^vXSWk~ZTCjTnMIruR z(^soZl!AlJN;N6CIlgBZATi&p%HSuAY*UZ|PiWQ3AWgf+n+0M|+1&@;Hr7o7X|fL* zDQPRu+n)675VQR*&e7U0>dwyPQa2vzwP=DjSBUp=K z7Fcj6;efjtJ1zs$&f|Ws<&5ZKuXcy}ff$k(WyE^8apKmOTbk^^p1n+SzF9m8HipXt zNVw6aAKtJl5cJf2Y&m!efX&7B=hb>H!;+C}1A$%Lkjna7iiI$sxlDD+hL=VJdubQ{ z0dN7;2Ed?Gk-h=YaM*)8_nPh z@l^~dG{-x57#b$f472jo^8;F-D-TbyCj)>sPXRoIG3J57e<00L_ZyZV zx$ZagvUZPy@OgARdgwj+702@qoJk6)`c^H_+GDf;$Toe86oL1r)nWpH4SpK7cmwVSZ(mgcR58QUa!nj^~5sJGoEX)vV!nC8|UO9aajvM!%tgsV))N#q{<~Pw4Q^3}Tv_ z#b0C-xaX_PbOVI_&|4NspC}_75OJoxG2{{*2*KXjG=rlZfvV%1 zYI#E_uP&)MjFP4uDK5&*2qj~y%ncSN)D=euX%QsmDWb@p9o$?wsc$Uop!m|MTuX8K z%myR~`n!*he*j38(chICF{wXv5*0iWw}A^8iCRLy=1Krnp%MEYBdGb@dy#aH7ZoMG zOTxYI`OlJ`>;xV?to}TAN86)vc;d;gc!FSr6R3;LeAWKfi$%TG0OJjiMZC@kDOdb+`kIqC_ukLnTZw0<5i9Y4}pz z16sW6o$Ue1f2C&!Knsf=TOp7e*_cB@KynL)F%Y=^CySD{Ks)}4P8^>Wcb&)9(w^-a zKow~#9`wCEdla2WlzAl5h<*C^VuDT#>^$)6Xz&NXXV}KoWmr1=iC;HRrt z3b`22$bY6=8Zq&nJRGne*>X2XXcF^Qr7xV#ftaCvHZpXYTVF3gkXtx=FI&Z%7qFY@ftDvLvlX|8j=I=VfWgp9%-BZ-yWbwUum5IA<@9JF>nJ?pk{gPH}f9qLcHWJ&#Vj7X8#2JXQ;_mg6Og1c#*ss^@=)q!y}Z8dM3P^zP+8= zo>GR|62<|o`UOC`*0 z;)W=tjxCk-vJuyVBd?5a-H))F?0Q+D?3G2!Mk7yJ#9ye66-3TnY2;P?Hd1$Idv!2< zu;U_jAvd&?hzOppB+a82Rj*u{R!Ynq>DIjJS}&*p9G@h3Y)TKrXqXDKRj!r zLj;kLKTj0j`z#cWOn!3YW}xuo6G`1T1bFPaXhpH8!%+pB5b@iWotl`~cH5_-gd-j` z+OXT}O$*>qEBFq}x3v_eGa<;hpPfY8(+&9G!S@3rS6*(K1=tJS6clz`$!py|%hEAn zEQBZ-&K8MdQDt5dj6igRFYmlvEUOy}7F$p6(brsh20yDj2ZtjYO5>BzkH+&?&oCh) z@3q)OD>2zBeMWFgO-G#))!4BA>dK?zyF7qirPiTCSiW~ShF^YBz^#zk-SGa%zqVur z!H2*GTj)+P@6o$ozY4V`P+1j z<91FU^-E#__t?Cq$;5Wxh0rR&VC^=-o_WD~E?)z$^j0rULHp$?F2ySEAf{g%XB2!G zLI~f0n@z0Vf0zbBob1pM=r}I&}`awK~5eF4pYPGu- zHQ7G-$-W0n*9*n2Dp_9qPScRkGf?F^J&xNN_~Z-|)b&knZu)0L!wn>uSkCc|I5K6= z{OAv{QP3hl=&Nk1%WsWO?|U`!)=yms?%0^HOIi*z^x=OTIHpC}RkgU}@m-4ffpr!`CpOX`V!z9 zQC1%rBBg^o{a-X{Qcq^i=gRxjlXYeb4raPc)tiXhk|j0?k6rTcKgnJsJB|Jd=QW!e z)<2UfDo=&IqB>S>RZJwrGvk2SaY^RTr#R5mKe;EI;C2SdT}KduoA3$ci}hdZnG}1f zSz`I%$W5hE9*LjMnbTsjaXZGL;$)t9HetMrswKC&>72}v0n)giI}+`Ybv5ma&&{FBS3h01tM zKhihb`yRYu>2Y=B^rZN)3BL(INQmDxlR71oKf`fc#YA7a0V$Iz_Mvo=No>3!pa_qw zxsY&E5vf&MoC~m&pKHp)A6z!t((DLNL8?gRvxM#DaIN7zj#N&Ir6c&xFCswnk5C<+ zO>)T^pWF;Sqs`0G_Oy|HkzW!4NiBN!Y6_u|`xt;gWsm#yH^B%a$kLe`Ub&Q z-L0O-`q4UQl1}?n@~lrcu~k{fH=Di3a9LG1F{l(9cq4vlvo6*zd8*EOA=OYfal&~x zhv%|GRDEI*!u<^soqAefZTOEWbC&GiGW&im-9&|bwx8jmmxt7DAVS26E!&C?NOP_k zgvSj|@5^n7qAR;(P_$)-XE@zm9~Ie_IF39W9jiaxC*EbVjIbrFCaa?nb%{rWHc!L} zmz;<|QZK7#hem{m+svu4(iY}^m$DR~x}0^(f|%S2@yCSlMrcX#R#!KD^V0hCgP-5G z^;wkoxtH-}B}apMkZiiAADYb7ZSr7C_%#&W(UKJ$AMsoMj^6Q1KQ@ZK@FZTJaL;-m zQ4(qr%hG7%C?NGL;%hoWbEzUrH_=;D5wXdwE6YewjBdxeHiyGQ5E! zy?8gSt2*JVA0J%sSU&;J64s#G zu(8I?Wp7SH7Oo*!^ zQ{&t@)F;JYe3o+-{^z?JPo1|6k$O4OK6hgDM_T`dzGnR>PxLcTK^`1CdJEYlQ`B7l z{RZbri4yLSyZD1a`Hs1 zzhXAYCIayVSv2=g??UR3?&DTHdfc!@M$+z#SW&)ci+hZov2iAX;@OzDH{kpMueFx# z(h+{);S9_HEC!zcgB{VgaEX%8rFYBVH zdR)FK?V`!;tRJiAVmLX2$enR9@cVn5k|tv$KJX;a{(@xU`y?$P{9_mYZqR~@=W^Lc zOG_@Zb%{hUP(dh{CUcj&E-fN>@}s}wFQ@Vd){hhXoA)DG&kRXD9XgIz-H&@?w8)J>mb zBn8=#_0@QgwWZLn@$y2Q<(t?#u3ROK46+l8^^5N~UcCH2Z8V)1ZFq%#1OcVg5(xK% z@Y^YZ!*4+g??9pa6&3Nb9YIXbEzOv?Gw!VkqjeKM;9s$>hNaZi*=Z`avT-SkhGX;# zzKVu_NJ*h`#IUz7z6@x_| zTgHV<)=Xs=H6ini5@_x-s2ZM+ruy>@NKF(S)^bcs{qUY{mHFC(8#kdOA5hk%#zccA zN5!sE;N6xy1uHcKW#;z$!}BP-N4V=k0{##ss}6T@TEQQ$Y~TEFkU1rxi4=ZA<^TFZ zG6#X}d!?ht71u$s*L2li^_i`spKq&_f#Z!d9V+Ib9vaA{DuCzjh1Ma!E7p(UjcZ-W z@yuMEqRI5mQCeQ#!b~d*_+aGrY+4@IMy;uYxal1ysds#zI?%_jP_ytO-EncK$y(c2 z@hI`02(&+3fA|bI==(%?=SIq>l{IB{BaDbYfTw?_TsW`;_ucKBB0yJbVJwoj-j!;M&0T1*zZbnOr#BR z-A_|t`h@IQc|2|*Y~-!F6Yi9+BUAiINKv)zIr8_!(_;)G>lRB(xJK#qlG}N?6WyW} zVA;93yXSR^{MQdK^hZn3>fys9Jy+DxoV3Zat2Ah-ihkXvN8gpA&vg>fiY*qNylHd^>Fe0_5}~?D z;NusQj}$2Ymmj_@dSea8Q!*_&{?{PO6myWfAZ4(S_Zr1Q{0!*k-;&k#)l8F}11 zW*X5#%TxTTE`T4qDtFIQA35O7e9(?oDPcC&pHIQ>%*rb9_H)E&d|_Y4+h?qba(OBr zrMVZrnyJ+b#@EV)n_ym%PcJq;rq8;Kzf;_ZL?nf zD`$4(qw#y&);hR@WvPja4%66TCk&PD$z`_$iEmt;v1tAP`QLu~VjABU7RYxmv=Ny% zTK6}M^&Rz$jxUYnZ%g=_p@Hl1;lRP_ydgZtz7>`$Ag43RPsFoi;LCi&+qu(+)WT6+ z97PcvUo=Is&$$VzT2{9w)?*fOBWS2i3lWyCUZ_CvertG~sK!Or$-G_v%&PA2jd`s$ z{9h4Mwx`^6kv`Bu8}?I$XI(Lsi7vHDDnD! z&WA^$q*=^sL%g$Yx%^BT&+z&F(N0OY6D`17(O@V=EmoVkmY9cgqyK*S+0Bvj^vIh( z$Y@~`q-zhyb6xl}deWKyn)bpv9f-fm%blTUW}FJ~TI?21_If}vku@3_aNyq~X=_u5 zcz-8ybyd|eIiRL*5xEV>E#oVy6+q(q2<|t7H#dx4Up`n0tH3E30S&+O(lbl^GaQV!ub7@_oJfOv8X19*#&~Jg+1*Ybuk0PSsHqj^68qx+tGnVajEAKM|e?Fe~ zIgefxLfbxBNq%w^GZ4gd`$5`7L`*~@djY9$r3YskDK-9d=(jb;o9y#2m}5=dHFg9> z`jd`AF(~t;JJ|@I)i27vLn7}7yVaNI?NYl(DfZ$8ru$?{4ts!B#k`(QBjBKtKS-WcJyeW;2#YC^+v5+bXIEX`y+nW=hiEw zgeN7G(EFHP$Z2?Hf$96cVXN9EQ%4FcgLb?4qYiXB8<)?xaCY`zjfk&|Dn3jOo10>vXwb=R-?g8u-|4M_p>jleOd z%8F)ps~>=aM|{sti{46`1Iz88ng#A`=G#AT)BdjzLN4HdgL8yR$fV`)n?dsMZ7R3^ zSx?=&Gz?nD_XuRG%135XE_tD}YWED|^>3)s#6G~XMn>&VX~Q++MASuQX=HzP!c(rz zM`KNIV_E2{PT2|Js+)X2$bs)BmgB&o1O<&ODKNaW|fb9K}As!0}>DZ(Y`}9sxD;k-~vJ zxb(v1R*3&I@tOx~hCc4Y4Ksr;c+6i|LDD6Be$NF?x2Qq~6kT`LiaVp^vIJQ@O6GadYY9uxrWj4l(l}r$3e|N; zK2h<9xaX}swDd#@t3=V?r2@TblW}*gk$-+1waT>dA)HUwq5Q)9K`NdRUA;AfX!(BT ztr<&WiZZR}3p(p|Ord41AiT5+_`(xC>w+JSCy?uq3%6Jde}NXd3Yi$5Wuas$y)ZK5 zr6;tYS89s9me2Bnho=M%SI;a*OD&kF=mUerFt!XRO=IRc2ZJsh-@w?PD27-z1V2J+ zJZh5FAq0-Y@fMWJL0fgmU_D8R8ZSLgjUKl!l&WPPn&)Uo-LZTGu|QZGj_i5_!gW96 z?+>f&U?(vXDhTYOlW5y}FB=e{{m8`c-zvh-bnD-8z5L9y0KW}%DL(rz-q=Vt@mit) ztJYDWO&_#ZMV6|9vVJu={d43;au&WcUtm|oMgOtWaKzFdX^*-xs{TCG+(C+IV`|N3 z`K)+D5~V9Als_!({EZvUUFUlh5h}??lWEPS+hWTht0?a?8qzT4cVf+T=?O=zLTA35 zT&i>nH(9udeA0>;2w#vz>V|S!=O|qevN1)J`GTfWws#;Ad@jA;Ox5n>tP~OSl=ZDJTCimFhBi=?qRP!$~mspNq*vadLEc}B$O0>Wh478r9!|(B|&m@Oj(OCv@^A3_W$5{vc zZD|kq$#T_b=}#iwTs0j7w?!n2#peim)J133{p94H4XFtgfv?l#k>yV4B5g&yvx7&t zQ=goT7kYNsdxJ(s4f5{5XS~B9=4hB`%NxK>wMI3#ADOotfX;F$U|CkrKCk{3KL>BV zd8uS|nc{6=_p2!WP+8}?s%2PX(29vl2@GQ!UZ<-~)= zm>BRi9tgLQdV5M7`B>qY1ajMk#n6Fy6BwUicKtX2B!;A$= zss0ZPcmm0=(TVC?`wLhrBR5uhS&1P(ShfU(De;JgU(ZuJb@N;q*uF`;t15GfPJAF0 zxtr|8ox$UgxId5O@8rlr41t6x$Wh-zClTGc+4H}o%AA#~j_*a>LyowQqAhVrLdgQ6 zEr$XsDbG=UhY00R8#;B-y-xluUexeiP^b+^rER?~hLEV_GcH-&kN%f>pRG6CxCL5# z5BR$cNK@CjKQJL z-q07RV3W@%t%QjuH}|A3_te4YSLsWEhGK5DDL_5}hw7vjUK z%pm@*5p)7~n*RF1(SMHeeRCqfd7Cj59$}xaMZ@GVyXXsNzUe{QQON1*T(AZ3MS)(9 zFY9r4;iFVF=acayWIbDfLNDedSag1dek!+opqc6>_U&W%u8qa9*l)rMHJgDx{%CsB zFGpz8i9YKAb!d*W58XWgkyxK)$> zMC~K)45yoyY2@JhV&_d8p!BZpKG)v4JPN3JN1k6&jo`Uk_O;z*c5 zkC`MA1rr}$Vi`omH7{pFeL8RBV1sis{9g^!2%_n~L-}c#^UJJ{buSIY`)9_$b{U6tppK#`SV=&$^N z!m*sdZc;NYR$>NEM@2|&0$cO!lZdnYq!CE7*QB)nTK|7Y`i#f-^>?8v(_$*noztQ% zlNi$Ur@%=l?i}BB(GHR<=2p`JLQtZwM_vUAy<1KEKgzECpUL&B;;%kp%aQoVROj&Fvmn}&Sy4;ZEU+g_xt-Fe1G$B-#c9Q zuGe+FUeCkp%0y1z1`AL$xRg~#2&JxJ&s-Cm# zvhAD<&c9%`_{=0gtIF-8ulseJ>p_L9dNUAPyRIcQEzmXA`tQFotQq4g^}LcRvNPxy z=Q0t`W+?pmlUJ`WVKX^)r!dz0N=eA6W4 z2;gn`lJ{vgs{JD1(g0ys@F=X>XcKt7hON}Pc!z#l-;oKH(l@0{F}zQEmszP_qt=oy zbN*@2$Vueg-KKDip***M_v_eqAGvnyoj~LR5vPYC&gUA6^69gi!k0p%`d zCwc&fvG`%5E|#V?_03ngO8|W^+Nc?NBrvVA<2slZ8owbQZ8G0;^`?oA@b2Op>Icq| zQ13$W!$>ShLwC*(v%nW$Q!r|dnJ0QlGSI&EG*Rmarr&Sn${J%cAkKde$OftKJplfi z@U4#teMJaiu+!7PBN?bhA4QtkFC1l{|D6QuG#^eMPZZ~Go!nA>DzqEiInj#n_93ts z=Z)57cjLgieF8oicM&7{xV6r*0~27fF@oy2W&4XuoBg-p85W>%t=LQFCFdDj>l?I98U>(Lq1?mt3*oN)bX6hk%kv%Xk#;VELcNGP|fhHjljr zQMH#iSKQ^u4eu1#@94;v-T6VMqnD|X&8RKZ8=h6VpP@J#uuKX1`DPzwBKGgr7;0>J z+FIvp+4wjwcXxBxkg}v;S99>lwH0il-Rt7xD_AMa>^Sy^SaH!lYKeT3gK@ZuyKk_P z@c~f^khQ?%o60+nwqx?f)10^@@A=^$_;4_Bjh)dx@ico}Od}@rdsLJQB2+`;HoL-?;L$e zsRZVWLwcQ?=Vv-P!$IfkOVp=O&%f;SnnIa2FyMHwQyCdRAB$G=92((55Z7WZ~P00yReFE~*)&0T1PNeM3T9x?F&SdN4!xb{vMVxitHRDKOY@ zEKP8ReJm5_YP(Xl=eo*32Sz>%yboklg2#`neYr*i&YzlE!8&;?xef(tA5d~`Nm6mE zk}DnTG|3B5pi1SJr5ab$U17<4=*KcESg&&9o)}GF?RypvxF5v?Ls(5e&8~J-wyX5fktOKAW@ycsj5=;A zctlM9iP7a0*Do!sQOg!s`%3Ct!=YZDqgF|b}F@f{8S|Gwa|)kgs_l)*jL^yI)$ zs<5f7Zr8s2%T@iPN@}<2nB=DX%fm50dZofk#N^oh57)R0=}FrX&mbbrm!FDdyrLSY zXmy(@Tu4j_M~<6c)kI68QqK8SipZt*^19}Ya{tT9!O^W5XkKck%A26L9lnF(8A8pB z7hn>tivBXvV})5P^%ks@T+-p9<~)f?rpXRnxD!je#yu~GpEi@M`#}A5`FQ(h+O_3- z)#k`MYoG;2_ma42_YY*4%|&)D)+xs%(xxTBqFeRGqcNo#YM2OHeZ$%1VeT2G=Y3z0 zbGm$64`IrpkH#NeQ#e|f>~}lQsNR$lIPWVG2!<(@^F@a^-}cbdjiKWL+e0CTbk-@~ zN!m?%TO38{j_h-f1|8%`qL8e?H{!pWiJb}7?57~)*XUf9t*7*n3u1id^a$%#cf?*5 z+E8uLbYG{`_co}4w-ObVw3KKSfREJk<)4* zr~Q7eNF5EEsCVNahEZb@Q72D-WQNX_ykt#Z&|;m+h@Yvb_6^bhCrrQn8!_Wu%-Vcz z1xVMha+#7h!+83)m22gr)Q`%W{QAKjcvCGtWgvD?WlG-O<8q_@0d!!;MqD-M9ZMkd zLI*O%=UuMw{+7YNkWPD>%eU2L&=Fj{Z^Rkst}d}>@0bFv>&+#0SE^nJ;d|2$5FgZ3 zGUxoH=Ss}FS(jX2S$tl)e+ZpTx4r+T^}6fvtA%{;o_POF_h=H%NU`ndL1{0DB;B7i z64=>%Xo6EcY_~L)s4E0Uh&C2>%R|70!b)!L8O)n2yRljXMel4&csprwUDZ(J6*b-o z!%RsFoB3Kqu3Aob$DvHPF&MX&%Ul5jD>zU7dfIX9>XvMF@EGRtnNdBRELw7WN^ySr ziF9OcKJ6Ojd6-wtRP}G`yG9k>+ngyK;L7_+_(crXc3Tg)JIKVz5y)68sm1bedGuw? zPdB%R3KR-tFDtmcw%yZw!K6;T`9h7e7vd{Cd$k~3vY&DvavAV0|EG#!%Bpayyv&s= zO{)KGyI19qS)o&UYH0ty9lrRkVDokU5#ZYg881;QOvUB>2Ge%S&d~fzES{n!VWt

a?shyl9_i;RCp^B@SQ2=EUvBq zqkKhlqL3+4_w=Wye^`3>Zg1~{$)d^5#o8rB^u1A`lsEAyJI6m;=n9mI0*`%Nf=7{6 zH`~7~q(SX`efeg*-lUl$`!nt$@8tT}zoXgjvra9)e&qSarp)p!m&hG=jQ5c2heuVk zL#T!tCoV}#P9-JRyBME%5a3bpey0f%SNl${Kg}^HB6L+J`gG z4KHCfYe$MJSlQ|w0nk3dq0xV!buA1k@lE*V1L;Gs3_tm@q(I&5} z5*acJ2o^#1M?7{>&kbK6181)_=R*Q!0aFp1CBVb%)&`0--_pjD%ZfbMxocgegi=@n zmtTuEO6{e(`uw)M@k!_+vdMl@+B!DBN>B%>xrx*H7H(b*$v0l(?*mMM@|n;j1fN8~ zGpD35K|$>v9J89-a=_3uS+L73ciIncbdU zKW>~%Zn}&#F1;R9jQ6QIK`hit3&WlLr$ke{~$xAwu;34rF zuRaQ3r}2jy=e7-Tgd(ZU1F1g>c#}sa*YKD7y6zx1m8AP=u2^1mBMX(i0nZXEE#|DM zk1_Mnhs3IG>GLlQt>pVDRSUiIaJ6Lr!&i}TpUuAItz%t41T`Mmrs76g-yXnzRD_kB zIF6%sR^yMxqpktVk3`T^pR9QQgAK=9@?`->m>b*BcLbHCwqVy}szLhOnIgL#(=43e z-R8GX9!H6~SuV_v{9Up!bAE;SV!nTlo{=L!cD=432Xx9|G~de+uH#mY<@*$VZ$G6H zA@B%Wfz;B~1T5L=C55_0l;~)?)4OimSod!Cf+@_`#L`@xk`6Qb=5~TI(hwAF z#4M^BR+ip{ro^Xs)kP+u2e5g%oUi1waE*kt4?M9}P0t2IM6jhe#IxH30=8EKst^LG z@-QV&8=UKa${bU%`81UutOJ~k1M&6Qf@P-DW^zD7aeX=T;7S%M05+_GrG7XL>Fz+5 zRUX;C^--fhC=;A$Qg>bZGSAvST!b2alnG4UL8Fz-UqIX8YG@;L6-enySEbwgdm%~>W(43PS~yP z2+H+)OtD>#X5utVnp8`w6I1H7->oyGIe-migSLzx>_GH8g-hlC$sA({4E&LU|o8Dw7kc5<1LB%8Z+#eDhsAOdGUGHI?uQ^Ms@(DRsOzZ#0saGMlUOf@=v>JN`W#k^sdM2P&{@1N zoE}LVUlPbUSaP-YO#j3{FQ`hS+>NFByh^$s0n~&S=O)ugQA6p440Ihjv*ooZ z;$q`_IvPvW1hw7#^cf#&YaQZ0iThx}4gO#94D@e~s-!IzU^#Al8sq_CS2N3?eQ^RL z-)mH;$fwXb7%v_Dxj_OH{@>m}2626)C9_<1H=&p03bz0gJKsi4sjY=kjo@XcCLYKo z5Ypi*AFa?NG&~y0J`c`H%a^oqw7>=JSjgx_pd9({11}kZ$8i8oPwW*A&BP_?Bfqal zk?r!ELhaUZ#a&`do6gb~9(RYCHn?pM)U+Svt&aoFGC^SCK9tPeVK)iD`>x_u7Gh}E3VIuz+K^sWrWi;M@NqB|wN@b43`lmk za+0$%r-#5?BlL@IB8s|E+hSe@kiX^f0?XSzuGCgBmu5cZ&vVL;fw=9g#1x5?T6*V_ z&Lfkv)VL2ZSfC~v>SVF(+WXwWmWfjy@>M`SY(9%I890 zA+xvhsU!`CTo$grD;h#?CzLYKO?2l;U?R@BuM@a>uzw^m@uY~!-o=3>L77qXc+ zu9LUz$+M3d&rw9kO);wBA-5!UT~U=CF{Z$n?%TI)3007S;M5_M%)DB!12X?8utHxq z4v#DZCZ}En-LwZU%z#Rj(<{IffIX!>%j0R4X20Z${{CD}8WU@y6&c=o-2&L-DUn%( zRxJtmQCo4^R{aG)nl-tJbr`sS(?6B$_8lP4H?Wd$&mLIJpb4=ejLkBR8FU)fGHn$& zu^=!`p`eUjgAcw@kp-YTCu}5`_IvkS5pn3Ju(0WwV<|Y z(|?M$9SZgqkq>4Eo`}9!C!PZPIx; zEij2EYYRTUgOssKpeLRNB`B{C&$0pKU8{j6OkB|`r02F`GXOuEW1G~9d=C1aY`k3% z%t)r6K#+SYAn7o-bCC=D9gcovCpRf#58A}f?M&)`(}hb>av$_A7(aXw|)gpw5Cd`#~lF)FnEqT@}jXrz3 zR*3=fUVic=vfy3Jho{*|)xC!mL zZgQFQiYpydg)}`2mRSVOrW$~Zfb*%>Ml;ZHxzWf*5JZxsnOcx%pEAh>8CfXj1<-)y z)Ucppq1_AxKBieS$-gMh;L?I$IFk&Fw<4?XZ$3z9ldFC;sPvZYoeTm{{_JuxW)@p;FaOip*$e1Fby_+o0 zfr!9i=C@>~T4iYvbB~bn{`C&3EJUUcy9AYNMxHlDhTM!v{twVL-p(KG#v>sz0P;e% zf0ls0oSk|J?duu=E#h~f4YXi~>LO=P$DXsk3&?g`zurZe5VbRLE-S$I6YYXh2b7FG zo#>!glC@u7oZM?=2+0OrVY){gj|+;;41`dZAAV$K;#386>RRZZaaD~{r-7g6(D;X# z9|$8ktp}JNO#*2AU0}?!G>fQ!Toe6RazKgo?1~NwoNwn|LU^umwnJY~N^w{_>mn9v zClABk>4gabgnp&|?@fXhxG{xK$xU;O`yXhelkkbi6+-VXFsCYT`>seW|5MoZEj^Od z|3SEM_Y?q6X!Ofoc9ralW8X&cXQ~`UyU}mI%`AWx0{2yjIV3c~Id5G32Yp%5^Hiw; z6<7Lv^TH9v_%Zl|F7naQAYv>K63G(*AeRKkSpnK8+KG$0Oiw+(Gml;X#19~R1P?m- z{P@pRQVcv&&`}XdUcS1OE8u!>m>~2L6by{BGAAzRBgnI@p)Mli*~UOAUM5ba)N@k# zplZ@AWk>Sbu4~%5QJZy2tWCZ~z8*5Djojq^PXf#siIWp$|HW8ao9faql6XM6EG4OUq>UD z>wlkFTwh&6T*7P4=FSF&f%W}XLa~w_LApocwAd0NlzK6OwEaV)=LyrIpA!2{Wd!)x zYgsuB<}qiZzDny7r0rsbKQ$w(TtZHg(1f37I;sEtQ_M0JxQ$g7I)IjpL|Z+ZUqYBa zr${9+af;!LaS-a&_s7^L!ys7#kbt%zvA#IfzEu7>rfX(hzO|a=BDi!>RZpZt`46hb zG#;_gYqnnK2e2`-UB=~1^zt8qN{t-*=d0kawO@BaD;hvn8B-TP9n7&&;3v{-sqJ)d z806oRfN;k?$_`kJM*nrk-Jb>0pWdGB;6^`^GBA5$b!Ymk>vrxImw-ia>p{-}8%NN@?9Y{dJwfoPuzhi-r zsp}$+#X!i(oJKVe4k##GLhLEOMgl4BG8N19UGo73L#FHAModT+QF+P1^<|@ImsIdM z$dYSBTPH|_Vthc7++I>B0?_$!Jp@g|H|@-vm3{txrfhmlIa{w<5Pk%BtN;$~Kj$gt z2U$(yJri1`eKL}UpQ=-@0TZN85P=)aoz44NOoe0XPd$EQb~Zm;a|MZbEs8!i-9=m67~Qyj+mx8f$#LrDJChSpiad;)b^Uy zqHHU0bXyQa`9V)4;VTVq-}tI@6!XTa_*i3IpNZluk6JmyjND`+MV@xjoSAF1@v&Xp zdcB}9#3gAeU;&hi1`%N^^v~aNg+rNpM@wYJu#}^Mn6Eefc(p(&Hwgpp`mnD#1kgQ# zCd?v3KfpVGw>$*QsOYyULu9GTFxVevN_5cDOBD5JbrGs+8uHGq5QTQaJc%WpET5*1)>kkvRf6VSCT}1qPEHH)TZ3kJSQno5%(A;a(VFQw( zLBvY0BkfT16Oi?Kx1UlXv5P9Z+lt$iP{f$vUgP&kly^ZdeB+LqDQ-~`*QrCYa z`ccwT{H-XT47GYf{%k{dhcv-q12zwzU+1ftSZus!N}Pf}zt897$F|?#nwK7}#f7E_ zFza9v5Z`*cY2wN5TrwQBDX{@V;!zan2mjrH|5ThW;hi4?FK09_Hp78C=g}1PIkE{6 zrXwksT*c8Oz#eHgXowmbxE@ys3*T|PzWO^H_FjqzlP_I^L!)i0SMo!|>&`pO*L&|{ z=#2sDlE)?A&GFoDxN*G&sKN?xFZpMr`pE5Df41?Ki}z zr2KV$weR-0IBfZ*6^FpRJrnPEuboxda7fbO&Z-MzYwo_+E+S%r`5|NPap!Gw;uu1a7 zOP5DWGPF}=#TgB7%e#C}7@o1WTgzyjhV8`4eKP2FyR|nUbeug`yoRtXFh9)~hUO2U zWu=hAC$|NUg(HSG zxI^aztA_59WkzSosV+F+P@D6K96ZE9{Nvy|7z%&wUvi_@>=g8r*SaUxurA3*;CiLszx-7ERT*JWCoxfTDgpeGltE1+(~seh$$!@$U<0kf3E znL3o!FU(-Es@Q_1E_7E+FipqaI5(`lpVLb4zzPaWjTAm^%_oRta0s}7z(zu|l75(+ zs@@KoGVR_icdp+rg=E~rCup#2*h;BC9I}zuw>m`uA^$v8@=dndeD?~l6DpqL`E^M! z`S^74vr8m7Jha-LN|S~B{Fs{4FP2m$kw;N`Hsg+oRWBIP-4c_7I*n7OGy2dsD<~Rk zQKDSPlq)VzX8sIuFEjXVgsEpfG*i@AYah31Sb85%*SoJgm{jz`@ct~^^zU4Qjj$+N z+~`VvTlee!(g|knFM$Mt^DLDj5Jk*M@qe>@?QgNU&PDRzR|OMC_5bcT)&!AV$p=&( z9jP=LIGnd8!2wRH?CLTjN%e`L|JdQ>x6U>iV?lMq_pS9>iKv1Znhu74$T9yWy~}3bxg19V?C;7x@>wPC5B7 zDX6`LxLAlLezQ3iL44wdu-;jJA-shvCG>9G(|$NI|GVxQV{2(y3_#*S3n>#O|8TGu z*pJ%{hGk*so9)og`1<~A*}5#28pfKjBzKEH%D;Q*rxB5Icw0*FJTZuw#RD4aHzec8 zCi7@pr1*n60?d72neqBqG%q3Q1E=nJsk(d7M%UR68A~}t(?9pM{w-&uKP#+0V2CLS z{LTBA*+O}2YlK$}!4oZY&-whl=^bjvT?|`Fvcssk>I&-8U95Oj4|JO0(0t-INH8z{ zS_mE|3B)&Aw8cuWaPm1$l^!;3H`MrOzALXf4!mLLosq_HWe(i^`~4NXv(`9FN(X75 zBsQyCw#=CHV3!x(H3^E{5z)v{9p@xO`8*pWGfhNIa=$7hX+*mP*kQ$8a>=X^dH-)A z=ix64__qty+%IZy2N8^#+H?cUlutdLQBI6_zWVk}kH<%f%J zM%tIa3Q8I>2Mqaz^rCJoJB%5oiTj&F(3I65i*1)aD#^_;Uj|-bY+X7HaG78;g=^JY z;aD=Mh5tTenejPuRB$5ty3V~w`NU*`vl6)FB%Zok^_&-6u0OY?eQp)5UU?zh3}XbI z$2xQn7eg&B%{07gDEX4nP8?7n8FTLNt@%?cT`%Y(NAu*0rU|>%f!x6NyG|0ni7_Wm zOM}jHAUakO>v~KFTp$yV_E03A**&xoYZ6$Jxv?`V7po2~hV9V)52}2yl~TiI%7ZCk z>vL4u&#sTIbr3y;p_gV5qo0cPPpLKuT+n}!u1bU3#Kqj`{6v9oM&wrv8R`Vik~0pP zuH|3hz!D5n?5nxMv1vNR=P`MpP}8ELR*Zb-pTL! z)zqf+FKHLQuhkCjD3cL3%%8Dt-v&Prf}ZR=gWZf8YKA&G?Y56JLqjBMJ`$W=8JXiu z8<`d;OF$2WyVOpu3kEhwTy8=#qiY{Mb$YL#gjYSqjKF8_ls~`^LGo-#iISK&^X_jd z{s!+Kx301ef+(A{=WUD;QwmJA-DXCZkAg>BJW3*_U|lk_DNU#8jZ| zaNDiCmAJSiXca9{)IlBw9IQvVA0G)rv+_pgn2LRQXm-PjZw99NaAb{gNOQk`0f9BF z{DKs4NfthlP3Dh`=6k_Ne)k7%c}m&qXztQ5sX_X;HNTWc2Qd?3Al)Qz04=q9ebF;f zKMbA6a5>dSayFBw^4vYV+hpTl(uh?Eht$Ys7Y(u22-Rh!xII3n2aZ9-jc;e$x!Waq zSn6DVsjAl=;^XL;>EdOg5wN=FbihUr_ty1LV#gBr$&!jK#6U{kQgmZf@A&KM-76Fx zY{RcwDrdhKwxJZ4l;0i0O>!Q8EW?*ufQ!GfHxL5X>Iwb3d#h!1;;`vERsP4rkY(h$dOAejIFMRbeFT$+o!U> zDpYrJ%{A~R?ZjLczti?kJ{+e159pR*vJx@?)z>S7hkXy_pR$B`&m#Oo)3y(tdopHW zKb}P1Z{eUcxce7}LwFI7qmbZmUEKbbgk$;G7G&<&Eu8pGd~NOb_o+$slPIrdqGyr$ zXEu^^lFkMz)4|Rx?Q$e_`ibWv;_{nU(jKj1>cQ=)eAvwT*v3Kd-b+`*AHPJ|jaF2Q ze?i(~mZqo-?HhwM`}4|2xE=d6fnW9wV9acszI2Y4l}F!Uu&qw&)hq6_M)u+qmJg%E z%2N~ZI%jyBYy@JZ!I{r__j@w_f*$KUg(-*fcPW47Vh(6ad{ zWYitv)bq3ReBFjV2H)5M--w4TGt9e^q>qn;Pbq98PNxNp{PC781&RxIoXM%;pLZRz zQ0iY-l7<&z@ZN2Qtn_fM&tqO}R9T|I>7xPz?>`mAv{1fECBK%31cL6iMh>&g4-b=S z6KpHV2r%atV8FR-R%LoVP-gv_!m{%Y^~3*^V)S8fu4L3}Sp%9Dp)iiX&YYD_PyV`^ zAUcK>CRdf1$ImF;+@Cn(g0OH`!(8sk#wru#qlF8z^pW<(V?IG@gc6BcmRvG_lN-xR zHLS*%DVt^0lfX}BOHtC{=H0Z7y`5{NP8&gvaz~3ZXI19WgR-5Ew3sdO*J+g?S^NS| zGLG*!+%vf@xc~y7%V^ZtxLL#YVXcqIyGpgpyxF)L=~hB#k(L>)yT4(VV_418iMO(h z1WHs(koKzW#Rfx`S zFB-#>@!3MU&tUVEUX*Aa@yUbyCEX%GM8ro%LEEA|Cm6HN@(+{RAZ2r|_)bc^cvA|?8>%6;rd2PwBiwj$BG4h_ zU%-_7JZ0{^luz74=3ScaNqm!-g8#1pc}xX&Zn(~;?-2iAxpi&H_&9nv%qHeWpF1?@ z2QWEV>oF6cm#u!EwPNJsU-CsA zG?7wyScX zwmH|>A+Ow*49iQzzKn-|o;Xn}_uBhLO`l)6OC)pLQhFA?*pR~%*c~8l1|4rERwiOT zTM(}m087PyVmKD9Ex1E=P9dME2!}Y^QTGNQPZs4Lj!dlPiwLj*^p4PD3jSdkqunO{ z2Ghh+9jrqls*-Pc{X#r#T(29U9S)%tyMYqzA5z;5kg#QA9fGoS*m4KMEo9g*x3T<3 zNiv=*%)1|P30O|bT?W$8N=W;ox#3RuvwXQsgwFUax=!3WsL~J8g=e}k7n4ign`fbF z7h*WWmRCioM-AATLR4AfC3gqLaoE5iEIQP(#AX1h*V}>d61~J1ql;fvS z78=^{%;dsX1fuisvPhFn0jqb<+diz+(jYKf^_&zsPf>$sHcHfXcOtH4Dcy@Fupi2lyBGdS~O@})`RHW z@t0E?bI_!D(Mj$lanj!3MI(={ld>Q)DoT$s_yp51ulu0q6iyFmB1}sk$&;}cRe`V+ zZzD|crQ+nOObzYG4!Ym4uL6ev_Ss?}uFr1wX2G05h8bRjC4L`e0D{&Ne&=^h$Y*mB zJy&z_1CJZ-DVQkn8i@&=>(B>!h#;nQhXei`5rx&u7xur>bQhmZosbQcOLEsc>J?2z_B;!a0X8T z-%fFI1@q-D)r{Nx$E@mAn4m+2KhXoO;+2(W;ven~bT#x15KF_0vTih{8!j2uYM(6H zr_I|A?xHT8K5CcS1NO0gdVy`T=@rmxX~ZX_vof>2-!%-4^uP*zS4& zeSgs+9F#%n83p8HU z##rE}T==kz2FDR{JRZ*KG;tlz>aN~+vzYRbZuVUm>Q3s0{&oy%e~j~;q|QbaZA-W) z@c`a`;U-tiZyOyzvDdtPK=S>T@Z576y?Z+5ZTYyI`6r66XcRnc2d(}nI#z*%n>>^N z9$4uGb)?u$SVH+I1BdTVx0kpUhoK)^xpAn6p;dT#v@arUb_%?V&pfdZX(zTn(uY=2 z8Fl$dAm8MFQ~~m97c?OM<9w&w0-cS?<=g+8jJnj#GoJ}u!H+az*It{E z5EI_FBwo+1Q5@qoqNvX{fybUxr{$OmlX)-jr6{Jd`^iNp*|>AJIB^%nbTS)dF{|vD zfr={xjr7sIb%0Op-jKaR8TrT$lIohk!HEnl;2=V69^)^HCAc>v?3Yx7#{N|Ch{z5h*UsQ1qYKlFS1JztGDp6JV}q26)O-YIiRxnZPq-z4Q7#9L)~Ii7VVwk!lif zeG1|PS1ur~0^WpLq0C#8nUv9zyU7gUUuqv$fasfGo-<@9V$2I-S#q()jZM%d$#BSR zr}V_e^7jYDJ-4u#!GAkuC_86^+~ajXF}ydQeFvX5oN$@zrUMO%JY(JELUgA>oh0eY z_}XY7c{m3K_Q0w)Qz04(0vD|DizcV@n+VRav~eQEyARO=VhlG0Z|#S}9I2wWu?<+7 zXzqt_m=60&KA6r;EQ8JXz|bRgt5U<#fsxPv@XJRHSr8S&!BbH|rWIHDARLCC;s$7u zHpCFep}FK|m?*+@k(LJ)stUM+@#O{&2-KE@bIm(_2Q?d4Gw2L-#K}hkeZU2y*V_U+ z1q1JTjetaOSorJzrG3`8qnJjp2e3gw6u~-q$S_m>(1Fuq80@Nf3a!NfoHfG05^|FX z#NOgS3p{KSgXCrh;GS|O(Z2Xay=ynG{7wW65Pq;7p6@h5O3 zG2V?HMYFHCXGI0;lwJt{jIbC|k}&-h%^qu$bN&MvmW8?=d@FG!fBxK!&uPlUi`zJV z6Zb5p072f6U>u1m5dI2wH844RmtMxP3gl7U;T_Br-S=CO`0NiJEKIVV$tw`tb&I~j ztG(=8Ajk=Bfv7L7F`z=>fs*oQ^)-Bptw~hHbD?k;GV1LNrB(kFv^|*(@P?gGmcWY7 z0;DC;e^PK8T3|hpmfRQxI43Df!?`?V=hDP>`0&|oJO%O2RcjI0#V?)&?O`3_@67P54za+s@2dn7;^8S! z;S;DtRgGt^7pWv){SN4Sb$Y-MkCDVS2<~uXp%Ubs7+X}%m?OQ1fvis8EW(MX2aoX2 z$PqvI>@*W$8_vl-qr|N-vAQZiCo)qOj5VF#rYgjl6BpkeOlw`q2{&dQ?LzIJBp&7) zK99IXugSu>bTuH6{_h&uAyk~GtluRU5*qrk-3BD)Iq~jC#!&gc-N2$wyS=+tf8}-J zpkdXMTRPO~3iQnWqVpf|x4~XbA

eP^R9I*FzXurupK0*MiKonKABg2$wx_hI}aF z5Le~D?gV$!)xN@+ky zBy4vO^`nJpEudA&3*Vtjly@dJ`u@7t>4`%zc{=T=)1n*wfU24B7>Xe(Ppp1_^KKM% z`Uto~m=OPs14koK)E=+{!G*Ug-@(>m6a?+d!V77pjUkpmguw2z(zmL&5Bc80XrE7L zRa2*pyF*8bfr!G!sEhDS!i|-wnMlcdyaDn9HQR@J z^yQ%c8V5x?EsjW#a9o*6b*f2txUU8oGxD6TmF{<-=MV+n3Ta@3W6jwQ8I#RWn0?~j zEIRP|i542vTyhx9u$$ux;e;mT3e6C;O~Qm z8zN$**;@u2Y%JKpgYNfK%R)O5WdTlxR}yR78!&r!-UR07YwP9}Ti=tJ>nWW4iQ|n? zvBdf=656`q>t zAcwN+CgrZUAcMpY<4_vPF1ZdMU^p$ggXyQQul{{NEywN_)KI7O77TJ=G==e@#`1DW zsUpzP9#OXN;skxUlt~+x-!AT&gHg_m-(uqOhenTuL+&jI1JRbwv3p*{G+Bo}(O5`X*&VSAu`IL-L| zFKZu#@#5ga5^B&okuD5b64+oBRPl-_|9U~LO0!k1Axwwi+tY!X`nOJomCn#ySVL?AuM54C5^Is?`0#= z_-$4y_=rd)l4J~*JaP52YCZo>K^Rjirm?)80pOdzf|7tKmnX>>I5zVsmO!{H^lT&? zlD+3jI(&sCvvP@69AT3ESa%AfERK*5r%xxPE;-{F@lh(hbie&S*-n;wuA4vkNjBqp zPL_=DWvw->0}6xz&usA210`%PY<7-}fkqc~aW*CmUTp}YFE>?7t{}vVGgW2nBIm}y&3c) zN1A~8js);3>9L}qvoZQhs%^HAm%EQQY763V- zlk@f^%EDPgq>uO``m28RF6wo#%nUljvi^nvQ>dnz4k`_(iC_ymBBh@#aPmEU*|zpb zsRs@ft~+pY1=}I>Xi*;2tQ?#ck0cP6Q)pyNrWL3VGWzWUsw60Igg%@rKy7Ih&&*r-xJ2GYle8JAFQm zsYc1Dea}9_`j3shoG0&l(QD@TarE-s+u+j69=R-IwE3SiS!qjjcqp&Qx;xZ|_^ZyN z>U+L#Y1B3<_D)?gv#CdiUmI@o&*}R!DJs-8DRX)_=7stY){+~DV*A6O`_uaE^2D7jfU^rRxrRn%h+Gc@K@ZWFgsMJl)WVio2EKj)zFfWXo z0;(5UHhUZB=7R%GGU6Bb{Rhls4+65ys5sUge$3o-Q(wsn@MYeBbU2;~3 zQu})soZo@3Qq*6jIM({MHmy-UQJtBY`BZ zD-Wf=4X61nmShEI2Y3a4&uB8SOI7>SmlHI2{!09koV>UCgx97S_MWa#(woUOu%VYX zOHTLa#)p4c``;JiP?d@GOWJFxxQ~pO|nSY`Dl(82wCg?UcVSd>T>Yl zM>-E~=>~A_Jd4@&KI6YnEzUg**q=m;W=~h!TRk6;S>)2S{<7;>DrNO$*HuGXyF@)u z`&%N)4?X9Q7xhH9Iw}^>I{RR*VmSz0%n0&8PcMJ{>%1=l+)c zaU1yUwG=V8y+E7g^G062x6|MY(@VeS+Y>a&JL_>6uqQw=eD|E?9Ad<2;S5W5D7&bv zWX@5zY*b8gon&6R-22HbEb{(IIK4?&t;smPpj1wPr~Y)*t8cR=i~ zTKL`6f2l^EgGYlWi%NHr`X-C-z1bi0s?r!rE&O)LdQZ;?C9JI|oTF}<^ogok;QyXe zDhG6+Zl47^Y8cvrkyO0q@7ZxyOs}=%y|!X_dT{Xsz7-1SlWdN;#ry(}TX%Rp7pSt3 zsG0f~jf<72Iq^TLzB;a{|N9>VCNW9`HadKRNHZE`lp>M}(vmVdBqel|qeDR&1|p%P zGz!uk18G!3ks%!tgR%R)pWlDKzxLSRZuj2Dxv%p)&)4%jkBv{zgn7~KqpPxMQOLLB zaj0uzrW**6_nw#lr<+tVTT>ke=h4!-eQaU!y;*T=W_Uom_h%ew1{X&mE|56wun zI9w($_KRIy$|LmGSik(09mZE#i9PSf@MAssG_udChzda)*+=U&j98u82_=*?akp zQYedS2GV>!JLV9g<0RF_6atv-#cVsiY$4FyE#caX zjEvwr-Ia6EvGF)%Q^`}`A}`)*Yr6%-!ng&-$OmA%8V*t&24}6;$)-IEjQv5 zy-BZ->1)F%b6e$#PUz)k(sW1)5>;$V;~QF=p+Sf9JD5ONLr5erfOI*e+VO!MPM+pdmmd)J8wQK6LNM$x6xQA!@^9!6qJ zqsH=ENt965z3i%bQCgBnJY0gSzClM_-knu4Z;QG9)RrnX;-j3e#yu40)McX(nX!L; zS)J)k2;GNf`eoOrLbgLFo<`G2K~(8o7_YvQgdp7dTNUmd~vGiUZrESB;k0`~wM zOfPuDDuwg?m?fD5WI${g%@s4mRUbx(EaXnMSmLU{{kbrVa;3vJb-hHBE_4=vgsf}< zl5_0(5{{1Wd&<((zSH`0#C`3uCoB7rj1|Ey;cHaK!#6gTMCavPSe;xv#4=zON0X+v09=B*9MvQ!*KbJrt6zEdArCe?D`opqlJ}u+^kb zopQ%$#rE>w+kl?mW8YAt@8pY5h6;58$())*UXEr8-_72^Rq#aXX73y)$Z%QJ?>{Zm z$o8DG$Qd&8_AFpY3$q;Y7epL5k_B<~At@qM-@7-M-Xtf`bJe2km+3Mu^RwW?9#MnR zwx0wIaG4^A*Ph~TRSvI5pBm`mdB0a8j8omuusPzG;jjAHI}MiHE{@ynPtSNoF3l(? zTpXvRbpyJP(bsO(v)-I!=ARX7`=;H*BUu0Q0?%HiY>K)VDN&uzYA@m7Dn1id56p60 zs2q_*L^wJ%bwYOyl~K&|nvtWkw}z8udr%EkRZ;IOi-twu_tEK&8835sJP5pn|0yt6 zJ+5TPR2mH&6-a%b*ZA!U6OB$fprfWNxi{ib)euGBR{mkETeZANFue)kl9F)?3PuWNBtyOq?yrw!FJo+-Pd4_Q)iAC~R!?6ZmcZ@Z8s8;A#FR z$Lzp>TQi-aCmlF?nq=)rP0YA;%H6UIjw~gejyg|93)}b;6m;G@}Jo3R#P4$WDl@?X3jhD$pK~aV08KH zRhzn5d#H0soxD3IqD_!x^=AV-|0cWmtk^%aP0hh9YRyh61@Sh!fpYf4SV;61EZcKu zU$xZ%DvT~B8gg^djIiYRD!a3GNb(@kI^Gg-5B@8fLH%S?>_oogbi_oNtKjyLU)Lb6 zO6d#z(0LJWCf?Ml$BF7MgAZ^EK_%2ps}#eivB=wnyJ1lUd%shbTR+{BS2XLu!gA{k zckL+RWExQFqzcEmTeM8}W7E}^kz%mF&Q3yCnp#f9K1*a}AlMe;)6hQYTigf__fi{! zyQ&Aci_=+6RnJ+;)6M?k(3nGHr)u!G+M}Aug-&krBzR!{blTV%zdz~H;OO;%8itTt zCd&^94s3jK5~7~Byo+SJCUOlhd5ia@#~WzMcX!x7EQx$w^tI;v#nx?iHVw0iF_-%7 zd-3S6ZyXbehR1z#gyDT8j~tzW-dn_Xj?biG?a0;`D+=U($D`cWM6)_hY1H#yv$M&K zaBiNh9)m+<3}X(OFh~@ocXC9nF;>iMmRjC_ZOh+`Syxc-MdpRX*C%pj^g|UeqM1RQ z2){tCoO;y9Bmadsi;6RAGySB;*>SfEU*@TZ0D@}eQYi23o2RsibUgD`h!^(bxuq!e zK<2mE(~;jLKaOo&d$lm|(ZVH{w7Ack=PYTJon;Vo?9vEV%HmhsfXNpZ8fZU8`9jn} z91mY5!jsiy(!j>(JIx^mEx%0U3=>8F$-fOxKud_!hed{QO^(5fWpY@{J zZI~V62<%}atbpSDXj1%HnW^83xI`7vkZCA|>{fl%(!7XZ{pGX#YQ5~CCaq|fXSc6w z=Dv=y%(J;ka-|>+zEuJ($V74CBY++Z--G*}IJU9RZ=u9z>rwSsk797lDm#w1BAFGz zQf2x%Z5Qr__(;Bg!lRoTfJQE|P=6SGdQ<9+?iNv$S;=?6ZggoQv-+ecIc~xI-_NV$ zYhm4lIk6`ep0kbQZ;nh+qP#h|uUF9=$(-V3_3z}}EwVwI;ed#^x;jbuTq3T+p*B_R zWX^ZV`BB=$k+wl2kPd>NKb?Yz=!gFQ&}t>>WR?ob3AE4fq`PCf>uG* z+(&(J4&USL-F~}{ZgVSfqRVq}#Q2f!jrs)jk0#E)CQm?rJgjsW9sRtu%8by<-Z}5` z1$_wp7QT_08l%31y%%$FE1?1?>+Rf2L#{C_=7_8?OO(qCCc(kg@I_CU#vgR)tf*Ig zKk~zZHOf4shLFi8P?dd6t9NH{vC`@2zPjo$GAKV)?E|!iZx&^o{$BS{-a*(hf2N`E|K5T z5%R9_ulmQFLC!;uRJ2Fi0QL%x&KjBGmuWNA&7pwfi|0w~R&z_jE)NAKX z7E`jjA+Bl3z9p9RAv+JR-#EQN%Fc*OofS*mpyWg#2Ip&H(INQtGbYebq7C-aBDlsoyzXf6J@uSb({zXzn*pe=9hNT;i9=+dbw;{m!H&I#(}NL@??g>v znrdlW7qY(2&Yew&D*=#}FUI_~<=pK_fB+4PfG(ND0nU3l$FN+wr)Dl{2B^FtS|vvp ziN2)-&&you1LJ2Sl;UQkMLGA?5x`%kOJlLJCPe>hjFVfb#QGoED0y)5L!PeDXjc1N z&K0`b9I$o}KWj+68Sj|&QFpJ$jPO*`^NNtv zb2@pd7Q$Xatbbw06-IuhpW$bg^%>`WCc;U?55MGc>sBPei=@w_wBT@OGbK!309-^C z)aYnq&5&oJG~z|+y&5>L+pLl@P;YJXU;p;;3oRZd$CIxd>yHt~?<}+(F&*q@4OpUK zWA)9%g-}hKqWo!K^tIXGFYLp^2%%4?qacg>j+mf6i(F<$3>9z&Cb(lH2jX|o@Z)+V z6>!Nu=lSY8>oNX%?7riP>R~PXr;DDfJPL)W)Xbf9#QRKmfm!2PKZBZk{fVP+kYF(Y z54kU4&)fgDYN2gBF{9nBf7PnuocAi#@u=ui$CK(Sx6<Z{Tv@V18B(c#f?L^-66RLsd7B01MQ_iN!}(z|XoS$8Ot z*B${!wPZ35AWwv$NI9x^8?_POtJQWqIbl3l$ZEfm)|01CJer&~xy}UKo=UP}IKbJr z;(!PDV$bjue6-ICEJ)fswtWsL(rXf)m8B04?#64MW0O#cllWtrj(^Y4q+#cOI60iZ zoyROqRW3dUXj8)5R$Rh<&TL}S>ZWX$@rpG`cOG4pAb0TFM8C|y#`ObPggKhGmdiXB zk_bS#Rgv^sxcVG_FLH)PCuG}vN0z%zdTAf&qWO~^ZyRupNm&^%u6L$ToO^k@`PKoh zQ?W%E+}_u?n~$E+$kKNzS{Osi9o5|ELMEI?@Vm;RB$tEv2Ig8h)YkLuOCGk7y50^f`;xIOyF5CL-8L4qQQLsOj z=QD!z^pqw`r@f6d7btU%Pjrw*94>7V-N}5&-y-JF3}_pHx(YB_DJk5=N$QnJ3kh8s ztwyo?zaB#S#ZApSesCrdH2WGL-U`ITLt4y40NT2 zpDNXb2xSr29TEZ8JeVONvMJYi1WPdwMf z;pF3VX3g}@r?(s?5naf**IDg|7Rho_t8zDYc{9OnGN{^a!O$(hn`>9*OY0HhRCuxS zV({;=7MN{j^k?nI)=TL0td|Yj*=!#E{W@wOoRgJpXk42i_j|crJH6^0kBE@EYQ#zV zHIQU=JgrrX?vJ8GC1$tsGq3hDC<2R$#1LKyc_MBJJ3BI)7DOn0t(j-0ZPLG}e-V_K znEe{CAz*Jr(-d7rcQoq5sBEuN{Oxlu?a#@A6X30RY`=JW06uU8X2wFOIpR1hw6)zB zs6Q$xp7wZoLOiac857cfM>K@`-u{Q9&`)>@&ppQfD2qkgl+*aM)f>C_UzX$QfBfQI zDN-C2CS@-uvYG;Znby2*?)|Hxg@EX?x=awq401&7rLdAfA5wjdE<=dax|%fmiikV= zbZ8+?QsT24nN6f4t?oRzwDk?GF%;4!U}d8h(}W&g*7LT4?1*G;&HPUx)Y*B$8ogV0 z)~fR$ADE9R!z=sKOQ(@4=A-P|m$rw$D)(2=kCcG+I3mtCpLx^p@(5+~dj@qWIFXy4 z+z*|*=WfIF>#4hf~)7WI5*c*EAH4pDwk}$iI^Ka9 z()b80c(Z-Wd4=+&F$U&};C*=8NUL&yFxs^(7Ci`5axGBVA0qdXuvF#1)hTZ>mBs! zv4reV&7Sgk>Jbm#maAwnYZdksTuxcSVF^zUk#jG$tI!#DEF4k35nlw3Y8LwrQj*}R z9u8{7xDHBp*HIrPI^BIFn4y&v>uOQN6x%;hU?vlvbU^(-BuqUd`d*GsVOX@6F!v6s zv?K`{8wjiDBPKf@UhW;1-2xCs@)L?46NI_p>*m0m<6#eoBla0u^X^R|(HA`bDNO^i z48!a>XZR)MC;fw*TELvU{#AlcV;tN@=4o+Q?M>;PH0n_B*2C!528wD?Vfw*tY{R|K z%swQ0$9$U)ABuzo03ZZbTtcmQfxxKSET^3UFSt-}>&K@7 zA0`#42y1AG7GVnTh3)vytH?9_DM?K?f@TxlAzorW7Fh+yx?QWVK#+cmY9Dma-LM~B zBEiI8*!43+GjGY z@s6DQ*@2j(n!iNdf>aO{Eo=yHplkAOx#0iXL}O?d7sj))yLw;!>yTWc@h6r^hn5ic z;?axAEVoO~w0R~Xs7W6j{8zP8vRwf?Dxxhmnp^J5jvEBQ2*+c_BF$c;b07<`CudCX@K$!thg19vE9H}3;0Eh&^fu@)y2pX= zIK%qm_Xf7*-nI{%cZ6J!)JJXT$YYyOKgtzIfulAba>mfGdb2ek$F<)EtT4UrzDL=Y zW(QvGWVOpjK0?W$yffgw|9i*!I_HDMu}0#;#rfsm0Q@J^=7pr0jfl4qh+axs9H+&= z)F!xo4RwZPkaP|>SS{uK0AzEji8*3;k+!pwIb*?f$(!hEwua(!Y_x$791%tk^P@o= z)fnh<+U)jtBfWyFGrxG2Fn$V9#-&aroa3FG2md;O=^EO3zKrkZPdch0{G@F1Aq9** zK)jtJ%+b}ac+o8fuJukFPAdGGKgYh2qpMg4!O7g(adwV=0(y=0+ zeoqHaD3mj1*P~rg9q%ZePS9k)+d5mneH2J~Yxgmw5#<)o=hB6ItUtQ%XMX#i%9Q;{ zv%`rwLQY@8ajv@H;n1{)y!<)d#D94(=T%uf6KZTa$F&YlZ$H)@?>p_W?)~II`M_}B zdtf^sUDL(c2hOa&joN#_GH2#N0uhUWnt|Nh6V=jdP~dm?8NO3PeL`5iltg&3M=;|j z1uy=UqbKfw&by z?&n!B?d2W>;0yIv=h-da=4<66Nb?=96~HSODi<8qPENoAT&ilQgtfv#@sC5<0OC+q zy1XpPN$~XE46vP1GavC{4c$!+E)gK6i}Z%RR){?lUY*EhO@`ap!SHC4dZX)zyqSelLpQc~2h1Px%cz`J9EZPZ zQJZzBs24yXswf?}+?x}-KBE^9to7CCDk*TRo9;?cvkYqQ(9Ef*<4L{U8SiliBVoAm zCV(Q+|Bk8E8--iB}mj(wqy9Uj32& zH+?rRT}3`deG5Xyt!PGwczXXz_hg^c-F)3IkCUpO-uGL{SNkWG=Nl36>)SYa zN9Pe*2qLk`^?d)vv7RVwzgQI;nNw8#mQLcQ9tVacC9LZ^B}M4E?M-KI!JMLk6Q-WF z4+|6tt`KiXkcF#^RzakkCb@8_0MJ>Cn*+?E6(>p&m7vYV<%&^XzF>P3Qvxfsl<={lk@nG%lB(z{ z=g#N1TRC8u$_>aK^3EUo5m1}lvngJaa{ty@#d^fg=NBD1UDhElP+8*<(xKr5*1vgL zg@rz_X+9a~gOMVbB5L8Mb7wpDtH6Jn4lAmZnfBq&I!21h99BY}ynr1Ecd6Nsx!%w$ z*e)E3*ve|PqVO_YC+1Ix=}G%Ue)pi3%JPWb;kvK(x${0Ng)?4jC(q;KtT1cO;L;Hl z(xKvYHE4GX)XD$V4N@AL3U&7J(7LY){Z_BZ@7 z1K(m$JwB0jRVD5)J4=HzaYq^o-){WUZ`l?B9{yyDf!EOB!7h(3nj4x+uqE=<@f(_6 z8KOJZZ}=ZmqNbn$2Se-!W;aCPhZ;~JW5dQU4UNDJJpFo5G)j){)Z&+&5~IllY`>-e-MMbAH@EOBv^lQ)1Bl&Mpyh( zWTDM_zrTmqpCa|7M`xie%FvHDn&|@#Yd3D`bk$)9A@R^f9F#&YTvz3% zim@vssP+!aF8y_ifkBclYhP!*b7!o*MAq#2w|fkFu$4n?n_V2(GXUxdA>X*VuAj2U1uQc>?N18RSA~eKH zS>@`jxxYVDa=$ow2TJIlI70g9Z$WqrofcXfT-RLV zi?0LJ$-ksa?ZbT&FEi)0hElbnixVQ)d%G~?OLjNYJ0u}X;&dNi&zmjEQ=uYKjQ3;I z_Xa6G7Q)UR_x#PI1GLyFI#hpR$lYKDa{p;k0uAvZn?I6gm0WwJl%bX(QH-^b%I@He?JbQd7Q)V<{9acPbzwlc>#eA0I zV^Zt!aVx!^Qxvjk9P0pKpLbAc{sW9xc1qp%NeJ{B;LA6wezK@U#wElsCADVRDaV5< zUw9UKC&^*O;6v@1v#aR_niVr;(8LpOFJzF%$2%j8Z+pXk;1k?e+rt+jUiX`e#zH#L zLy7U~KIPcTlwm^8)0{3G*PBO8eqWMoS`QOps1^234V%^w8d7WhM-Qdx&z6XE-Za7d z@AYdslh7H<CZq(qrW`m7d3_Br`-=l7GKFy5dK|K|4xM=-92FeA0oM`ZL7zr+r*p2&0HNWL;`7<@PlmA?SVu z&WS}?l;Q7o-uAwo3DIv;c5KO*G|=VdEnO?EN%g5o0Ckg4J`4<=Xf>M`!TM9UzNvCx z#BwLhUQ6XE7=sTkK1Q;j`l=|TKq53Hr-*G-Gqy{{4Rh#7=9%dR>EI|=e-q=774z%$ z9y8|mLhEIrgBzHYXitMTAH<@UQq&oyA}e8#dH&e@1S~J2IKV*lUBq@ESUkP(y*&&N zm9Cig;y&6B#tkB;KB3B;6z*|D+CSjm9U?ucfYj7)QfqvNTZboJJ!IUnuf?^;j10NH zLw6rTq)Ug7u{3S#$;I~Ja!i|lqd*A-JH)Q9b1u~13n=fB`fs1{O%+U$xiEby-kvgb zXwGoV=#q;11MJLK2`_>nxToFb`-QcvS=va5=0|hHykZ#y+TO%p!H~a{`cMFefK2aa zv>imjVFi#M-rmsU6@?zxh*7WS;9zVyL{j{@UJNHjzbL!M z-jn$HRRdHYi#F3?C8x~ULb?{SYEGfZsKCvs7Jw7>P zhHw9%AH<1aoo+)_mt)aa# zzF$}~5|2XId=p{o81GD%(RE4|%oNe=`96|*Ug{}f^mdyEz9}r;G2VXZtL8&Yd;}FU zH~%Xam1geSU8m6wS!_vEGiD}GFhp)qILs_$}&yvT&!JRfS`^b{5< zcHa6oA-8Av2D?h}2xPn`+7xfpLx261_+UObzEfd}Vh!}_C}JFp=KoTRzU$xs8~o>!vp z9b7g$*pYjNX;3Xgrd%+}mN4He)V^@eNKpVUn3aY(d!;&YGRiIq-Oh8LlSQJ+d6o`7 z{U{NE&5jRoP}|;5Z1|)D9uU22ORd(Hu9qC>CCtb51d339J^FZ0QJVg|S)StaRYX+q z3#Tpn**(kn2pruzmAIY+_8X`eTX=M4WChmEaX%#r+iDML@Eb*X=RlaGW{k!^Nist$S>z-Y9CoS{~ z&ohZ-`Dm)wBg}3`4^=FS<5XY$K*sz|z)Kw94@XKJV6AyycOPN`K6@ywp+m-|B)``^ zjD%RFWH8<-@sjBWTCZ|GKYc_L4632Idn5_ad|e-#stKVxkjtgox4YZdxzDY5UE#-* ziwz}GgZ!zepuwnKr&@U?_HLj#Bbgi z`Dk))A15#t7~ehb`Nc~Z<*`?+j<@&9tjNUhL{= z@-*1YX7@|wnLW*Ch4#R)wC35TYC)zHzjDpwQ8ng|)3bX3(jWr*wwuvJta9_A1e71=lpGCF&4)&bD>suF`RW$2g zkMh|>d&OlHU1NJ-MRWEBbWqb@j`z!M9<`{>S8g7$=jUQjT9Z8M9 zP?xa*E)e$0#~9#snr&aFt55r|nqj=AUwj1Tzc;)ncyFPhe|Hh8oV>I3zuwOOj*&GW z1AD~Dc5m;;QVBFWKEgX))Zjo$&Eg>eYw2_KY~mRk0UJOrma9f`e#n&}4fP?EgDNAP z8b%DD>F8F^xJdslr9z>-r4aNV~e)L)oF-_OL z*)`sA67)(gYPs;0tSRu@&M#X^t=Ezi<5XCH+OJNTg#Jf1nV~LgdMRuUdCgag?*@Y~ zR$rK8!4%b2?yw?Wb8mFT@v+?69pDYvNih+q{vH=SBDKbDDIXw4vXGgNDMFC~*1cHD zKyg8CbAY<{WlETacn_@r4!o(~_%<{U&;P{2+e%)#!!SoHuifp2)e^eN{=_?s8Bqru z+@!c8$?y{nnfFZQgVSs77B6%B!n$5zsuAUZ+@#wiRtRaB4PSi+Cw^?cGFS-TyuA1= zm5nO@TUOGYp`JJ+Dl)c$(57y<#RE3{d;`+gL1Dw$fO@F9imWER^-ms&F{4I<3|)|D z;{B0s;euJ5w6J<_0&qnc#Yw51E#mk|&!jqgpvOOCTY8cC_vHN5i1q%>$yp!){6Rfh zH<~!LMU!A; zU^x(6om2pvyWl)c@hs&aB2eKKiS*P=H~idV}3R zBSJOr&-L++&je6xl9C+&!}zTh*D>_?*!;!Pi~uG(ATm1YOPKr0(+J(#YQ8cHtv;#& zfx7d+`eu?RymxfNeo(YH)4obNGf4oUIKtPIoT<`(A}6cKo&HMTtW5o4$VU$9Q=|=d zxzIUOm~)1Wi`;oWc`FEh9=3@-Zxd5|fVVxR-sUG+_{Zrm+rh>QOQ!i2WyB;&S3j8o zELtQ%pATrB!<^fEDnQ8H0wc6BX31ki==X{PY}~5(AhN{Cyx<&9yMi4mtt%UZ+Ry@> zpX|fcyuILVkjy*f?nUsAp*Cc*&?SC`Cg`tcexK%~^7hWQ8=&s7L}d0I5Ny+sC4_ju zk~`Ug)f8@#{z~4z-;;VV%&j@1X6q@2eca;%sWa@&u&41V8t=o9UkUP4_$I}*khDfV z<_td{#8SYIXd}BPB?N_#=o`QjQ`m?d&PtMxL9St z&3>G^7=7~7xcJsLD!~62+5lmKrF@S-+5fZsh-QG~EL%OhJc>G0`N1Z6YP;jE6lfdv zIYa&V0n+j36DyljUd2445pfoX`8~kToJ`;t=LyJ=JzMB~|GAn*v)W_bwqHDj`hReI zhO^2$rF}A*973@QcP3sx=h@s8Nb{naI#n`n!tV45NfOel4`_oaO)}4UQT{<=9x6ji*5{X5VQWz{(b|f zHuB;>Z3}pXT*@k<-QT@b*DH;*=}A2$!)eP!?a8aT>`CT+`~;BPnm2X`e;D`JAJDn?7M?=xoEAe?o&A~5?A@NLQ z>}(D@~DPaDa2i# z5~ytM8NT1s!UvAwf(D91HD7jGKbpH8V$isKApS>PylwYVs~9LswmdLoq9eT5gO|JC zdnNt7!y}XD9SQClX75k>ulaNh+=OoBUbFQGIgt|{jivZ~7FeBta5As#Z6eiBu2~TM zhN(~A;MaEM_w9{ON2tscgAjM)&hW*kNc0u+KAC@W+WlP^eqNo3K4btD6$fcxUgyZ` z^KAA*8crC-Ls393>S8jNJ^s~Po2~HoeIzM2l2f%Hp!7?U;scNj(H_I-UK-9Fh;wsd zrrgpQGk{it!U8O$Cir)av_V(%-c_(~+3<=Acp-qhud?Ej#g~(<=6tkMV4}j=KzfA0 z!Y}M|!XwcBeN#r)t~(I3+93C=cPr55ti}65&qpaz&92ih9}F`kme@CUl|m_Zc0}TZ zAR0r&Nf5H{!$Wo-gTQM_&pvv*WB1oRP=J}x#=X&LSrFO^1&mEFq=IK7-=*@7s6$~G zgteWa^Jd0h!0nGVIWN;O?lC2MVEhVJ*cgNmv1T#D3JFA9!y+aec?)2lUpdaPIv`iGUTjJSw_Gac^Q7caNc( zC$}_hKlYJxIbpxcKA=ghQQ?-G*sH@LCpSjaPKSA(o?%0C3PPYH{Fij6(_29C zIbPu{5d1f{dDt0pO4~!Y>*Wa6Fbk;@e`KOMtQr4uKw8u!9AgqjS4fJ{s0pae*!zRv z>@Ew;@!;KMn#kSU$fpRvP5J2yh=y2?a+@}2sfolYplCa#Iynx*1?&g!*fN2dQO1jD zF(576NDBz-VO8Z)9wp4>lAB-R+Mm6$S3F)*ak#X-pcDk*2JrxFFK>eSquDg3k#l#9 zVHk_)TO1@%Ii|5=j!F zrSNcWvj8{IH$c_M`7+i`a~|;{y=ZTl{zt*)6U<{FsE>yGfC{uMktA$2!uO_lyhB2s zdQlgM3ZFr`!#`)(2jp>xLu+gPo);iv&+*HDf~28J=B*m%*mqK$uIutm*`%5|-3=^r ztrHX2n6n4N+kQbFP-{xiQsmivme9rUFRK&lGv=Oif`Ih-J$rb_y{qQLqqlA9AlXB* z=3(j~GsXuU1@9vbugQUYxiv22qBfpri0va6L%~)mA`7bkcW+u+7zAc!;l;KDx_+0C zEL5{G87jyFis<#?(jI7*9@9XW5^4Z#=bx+ZbF6mrQP7?-a_Ay@uadJ zP8%1aLf8L;5*RJ+U)s)4cKVQJyxtdbk9RC_3YTS`caS``U0}%>_Ejb%{9FF-rT<08 zqLa6`4jG3hIuy)D>Y<{lAh!8uqOQdOQM||WI-{sLBQ$yw`+=J0PBY)u+aoQOF{^-+c7d(3HZG7(&GNe#CO`kRbmD3h5O+K@TP`s zWWsh&BuFwtU2l1&xfbgGowlA~L~ZW@lU}#f8oGJpqYJ#8Ov^%hWMVHfU;_wgc+wu9 z5y3;)qNcoW0nF*ZrAVi}4|bgAoZ%brKhprT0v}E~?NiaW3ZuEirLe2C+wR(mgk8Tp zlJBw{1qhd%5r(6{hdh`_PmbsNm^^g*;)W$J2mK-kuP+r~B4ZyO3OVzc^b@0a0-PQg z%Kb(IW6|V7LqI>ltU0dk)Q8kq_ftzD{0zE@je6#Uw>5K)ey|SwQ;Z;i!p9w9IQ10p z)jeNLa0c0&_vjXMe@N99+ScpbO{VbjAnkUc`FjM^Vua78WfYWt4nu(F(c~miR(Sv) z`M=WpxDNgaD!V1?SASC#$O*N4zRaXFj9z-ip$8B7M=)asrVMQO#Wk^Z?>rK49kbSP zslWZ+Q2ZtQbyn#ImX^LiBb_!n?o4*Y6!ZgcTSG#~-2VbOFG}0O`$p6xL%J2J9{#6k zGj8WEuk22r=PN7DInO=`mI$zw*-O_IZH>!9P&2favb5sD|eB+zD%|Q8+6bE4tK{jJny4><3iO zZBwwqwrpEyAc(8Z+aUN_DW`ZV-+v=3ytW~Pe@Nt$N9E|H+D=NyBjEN1ow`aSny5~(RF8yXYfXFzOjXtt^7J8NEA$={2fEUdlbgS&(T0=5M;+o6IUb9u1SWe0l7kDr8Ldgv9@@ZSm zbC_=xU^UG5AhRg+d|4Y#=I2ks!wATZ5oM4KaPL@-wP=7|dZ~kc!{z9HojP`DDBjr!(kw_}+7}|L+f%XjrG&bpMmOq(`Lu#-tVH zLe`Q7iu-Ce$b@+j;Wg%*K|}HndKby{m|8~_?;|L+7WKbVrE;r;*OVs}KIJS4GZpM) zB~pZq<34&)>Zon`Qtq?T7_CvvxPZs4Tvv|0ug3KtYD7ZaUG31(zc>8FWseWMlf7A+ zsiEl^;8P^|fRiQ6H2jyHMJWxRpY7u-S9aseDYX;?V-gykjW{<#ZWJlh6_wh>ljraA{M`{zaK8F9K?ARNhvBOildy&n6-7Uv>$`J{V{pZs39lzZw$7C!NGJWrWSs!ci^p1UblOOe&@ zgTHoz9@{Z1O&%uaYd|9EjOvRdq2DnvAqPp`^z$()Z`oD^l*%T9_7!B$t}dg0b^h7; z$&tEJ&B#Y2`<}%X3ahQiE>`*0zvk2#=#WU})b_#ZUkY}<`aawC;Cn{{?mt zqtJmiVWn{9#NO1h{M-tB%g^=I+1`fa9!;-T6lV__ABIrw6Mj4Ue_gm;crWT&4{JYC zC|pvHcyy+8eEYy~3z?UHIHyrfe-U(HHm1^Etx_QJ!kw0?Whbr2nVSWOZZ;10mA2*L zNL2+3O5umYr|#do*0r(UI1NnMM7;RS+hTL)>!!5$i|#I8?LYI6_GU;hjTfqS%yrp1(zj~|1Fxb z%#PMLL|)jmB&H|Q8WHD*YO+8!IsT_+ybk?*+#q{+Ib=JNim$*MBBlS5NW4!*2j1l9 zoAY1+RE|CSH5q=VLGH(n$)Ovk6O+=9=nNRSR|Q39D37P!kpQ`@=oZKF!|gceN^$fT zzHIU^X>d||U27Ne(v*B>gxcpS6X2iE-O0m|%5z7wx;E*rJw>yiXZoQ6Cwcq|Z}_GT z$T~6CPS*-Rw#|~hzH=cn9uB8>CZli(Brgu88F3=}WZms`s7}DHs!+6KLEhVbjn^cn zzrXy|(~J8xnJS@jvHwQykvgHWmT2fsuH=kR#CZOAOf(nz^5z#dpRxN7*?LEZ)Agoj zdr~y79$q=)aO&nrwb{5S45c^&8@9l0)7_tQeBb&Twq$Il$-eem+@B-+%C5rRgUkR3 z^*Z*Pr(e}k+z=-3VEx&LM8C|+|D)`>|Ji)Mx7d4>+N)JmwYJ(~wOWdzD5Z93?bVo# zQM*=AHAC$xTCLi|s7=hO9W#oGEg|xJp7$T{{o(rqUP&acJkNdZbDwjab6uw!GW?}? z0|QT}_Yt7~*+BNu&n&zO5_**n8A$UuhI94nCo)b+$lzn=5N=Mg39JhFaakAT*dtNf z3epU7WUuc}={sYESFaZKTe#N~-mw%C>YT-l6g`UFseVDz?(VsIJQCwF%(&~x8dwf; z$1>U08C-ZwhaRnug@>IaB+Cg_x5^hC$~7jsBGrv0%Rx}!o8Bou13D35tQ;X0A8NER2tAugg3Uyl=|q@&PuMCR`w|N z_inKh`f02l;2G=ck#W?Lp7)J3gVV`~!c`;q-<3Jt+eDxJx3MHM;^DZ`sxV{nMq#oz z50d2&U-9pR6jcuH)TOQYF3wbmk4vvd-MUX$-vdFial2PY*>FqaB-~W=M?tl#`j1u7vHo>-=QEy$oL zS&ud^NitzCfn1yzM>?;AL{W1Cy*qpwt**>qs7^lfl4Jy{vT8@sIB(KP(oWXlC5ybM zIFiKkzam-YB8?ZHxKE=?=BiFraMKg{fucUCE*pIf6wa$XoKGUz_@Jk$DMK%)-d5$Q z#;tM0H+JJ>i?U=|&d%%kt&d5Ry9!AL( zR8mmmOD!Qqfb6D*G^8hgW7rBLG=0t@ao*2r{sbdASiXv;W&7r_P^=FcnSA(tvbc>v z$wau8RgE_b#ZBLn=0-)pZ;B+xvEFzv;T5tb`Y~`5SAH`*hIAo^%5#5Z@0EM;0zV77 zW;&&Ulr7?ZaoAI59OA(7lvgOcmy%gwP(zvGW>`TA8o^*?X3y||@xxU~JH=p<5$U;` zE6>VE7cBVPQw2=OCM5Hd`8C#FksL^A>LiM3y!=BU!SILm=T^zsssI&L0u{4>T6hd) zNeVx^s}9-3j)K3qdvU8TzbX##iDb2yV(cjqlM*}l(`RO;d+4WeuVBzC8iL;sk{oZ_ zL#X4%ac~qNrE*7(a(A{3H8wdk+7bDYmj(UX)%Tbrhjt%&Km1}ICcxGQy?Wd3F`3DN zwzt41NM)M4S{|GsM6ZT38WQTajuIpWJ$hbrngS+8;?Z!Av^ z{#MKMGrZIL3mdvUpd}VkY4PDIItQOc>s_A|(#{$Uh|y$luB%)G67`c3ps|teG?zYB(To^6#(Vs~?jb>jN=s5Y+#aM&D>ua_B$vfGr z<~{V?_e=&NSFdv7UikgaRrI-Zlu9z?*n%J++QbKv#;#-EnUGq8NIh7c0vp4{BJK#V z0@gL(?=Td;jIN_}r(`w-$Zp^AA{oj2$D5UYo08cgsnhEH7?%wLa8!d%$^gGC2M^gP zEWeCysSpx!njNVCx|^L3Y}{WFTT&-CDB$;jbUxw%U7w}uaSJ!KUynOQg2Tua)(gm7I!k&byP&dB& zk;46zmE=FmGh7JBW>T3+2T>(OJu+9%_$|1JX^Qs|4s8!7booqctK?bUiKK|NB0=y+ z0b*H@kmyr!XCrr>4@vXVpX0|G!C=Q$1j38mowcn1DzA(KsesQ|%XL!go=*cLGfE9o zxaOlTAIC-IqA%!6Ujfrj<3egCFJdZV@n^x+Gj|}K7FH{;4gRpCGgIf zUhvfMc;)xsZrEzPp=M*KMA&l|5&@}Hg*)!Ipp&>OY&UrqHXe7CrER?6#ZVmX@{B9y*!()c$>HHk}TWu567zI z`(5$6Mgrw72gQSfD`g%FzBJEF?1|+8TW5R%D%GOBH{K4S1uTPv=s2w)&&U+nw27F1 z6_C)D6PcgEOB4ZqU5=-C0}_OF8)tGS33oA|$w*=HJA7Sm-5z<`1-;BJFrNs&(=klB zUj$~GtE3C57{bsN{7PCkq^Jt4vpr|`1&duFk-G!8?QrXFW`lpWOOGR_zSg^|#Q;G* z3nK~#!batgmzvB$>xI>jXHue6_JhC_MK2xyD46XuOxK2FEs|3$$szt_nc!dM}F*EvJ|&Bm<+)5xe3P|55?2Qv=Gw3b^_;%`wIk zR2E8(R16$*dZjJA=o4j0@d`rR)E*>;y||V_L7$o4+~o}E&h#_TVj^=j|8ZLZII7kt zzJ6)sL_9gohx~$aYkWD9G)1Pq2l;DE1F+%Ij&%m^_C$zWWZ{~urAC0J^$(JEa#n#) z=ZKfS?<@%*Ia{s8Rl`bb$H|8nkoBb)5(=>Y0UhcL&GqYmteEj~R$>z5CDup-`idD4 zrFfFw?DKLO3c4hk2r#f#0Z8i_VXeAkCfvw@ItDcB!WA094{M91L`-a!&^ZReaaiT~ zpTIy-r2BvYeIaCNZIHf@%Y=37*$ZOnIbzf|5X4Bc1zp)6U;0eeH_I^m;UG_QWULiR zmKTIvHYKJ4(o5HS%V)vNWz0sOJM|Gs5-%1!ZE|XJQ6%kH=a-agFmKnQP2n=SOskxg_+ucM2gF*_`W!CsR<4tKJ{zKrV3> zvLL0&KVTak*}rk%N#+_3S@4aS2)E%WvSuQd_lIBBJs)~=c^YNs>zRMGMp}?&cV%}G zWUTL2idGo3+h@AZTraaPKw)N`}$BBxuSL+LI{p}||tJS`^6q(UKLFK`f zTn)DI1)oJlV_d8QM?xwVUpNWki)hKm=${ga=St4=^o4?>f4hj!wEm6=@zuo-FI2oRwv1@}I zAdM04O*ybX(%VJ=)OJS~a)Os!RLRp`Ne6vD5oz!I@_L#@d&=fXDLeT9V*)D;x~d>5 zW!)&T<6_&&%K(Z{>O+|tU*@uS7q-m3#R)=IFf%R{1kMlSlo#zmglo%T9 z*7rEq$3yHH{%Bl>rF$*=IcNo>PWA_sGZaY(LevWBtG>Ycx zxcpOw^MU6)H|~Rb$~Jtxjk8HLAyBYD*^SK;st9j@K=0)$TV}=n>^E@ z>IP<`ElAu%G0+6aa@WEblYZas^;&!wz*iE@e&Q!kvph1Bi%q0tKBLO%v+t%lh8rBC z;d^6e&f`c)Lz+$-Z|BmMDiMqt{O=I%U57rI>#MG|ypEmXzb!U+sw!*IDeHZdjo~cm zsP9OJA$^AgiSfRXl+GNc$dpTiUbUCO4Qs>`N%pgCymzrfWx45sT3GWKUj32|1}CWk z-B+U#@vA(`<&jg9V~?3U4ztXSSY8CRfJC&bq{5Bv4$^}R-M)RMWi#>F;SSzT=%&k5 z_#Lk?!UKr5WCRaSB@krF%z{Tq@=u27_z_|~Eq&)R;{spZMuTuJH z0@e?YRLG9u|E&I9)lhXvt(5~D$sc}(QdI0u-jDe?}BQmWzVntgr0@B*e9 z_Xw;^-hKsI1lj3Mhju+~4KP4mD+0%qkS4;Fx7|$MTpA6Kw2MqZ&o0oKop;tM;GX~0 zviCC2dbSY_88G6888XPL?J1I)du=>oK~?zmd|U5Q-koJSn)Pk z=sBriqhy8#@3o0#(+lpYR&AsvgZ~6?nsPaTZUa^nO?k1V6jy728p7cg@vc7m93k(~+<>7QQu zxhY{PfhKJD%Zhy&JE7m(2*=dtY#NhK_ns%fH(_9i&)kfF$f><^pA!%7tv`~S&C~6* zzj|rQ%1HalgzWXm?ez)>7%-?=X~>VJ9w`G>s>s+HW#ChkRM>4gMT!Bi^|y9u9jx;a zm^HxTw&iYiquMX%_w8%t40uHR)d^@;dAJ@))OiQ_{1glX%Qs>ZLqWE$Fp6?4M=Q|BUinL@x%LFqIeNw|<0=&y~OYIV$puHx=M^{{L>c4S_0-UJAt*Qz^P(7!xNsyn+0q=O)SVnHdO&p-Bg7Gx32 zZ#ysnA@~ziDQ&+#)qdE}+b(7@95)rqheL?PJaJj{%^;=3CxAK$Uv)jPC_ke7+OgtO;3A11 z5^6&);Y4DKs1zn(g)z3~z=Ga$-aCSbJr?9y^~0OPlcX`SM1^w!%6*S~A?6%N=(9fQ zA}O#!WP914$d+20$1x7Y@I3>pzl0Cgp9^S?-C46cKzWGu+mkFGo66Xe#bNrZKmz&9 zql`!k5`%|8OBJ!!nUr}NIGK=FcRH~nK?rVHQ!>})yY!4Ylj1ggr7DQE#>i)BB&)Ii z1wZ+EY>z8}w)||m55ECk^#}JWYCzVqL@DqM5<0z>K*5Ze00RDgC;b4jawlRb0I6{7 zQ84Ae9};yc1DV9X2oN6WtwieIwtKo<=uJHNWPD~t5Tb4lC!jns19R)nF;>(kM@8*4 zgR^o2ot@25aqHHH4V3D{^rqO*+-L?DQ>Qt{qesWrskuhqt}5MW!B=?MhsxRG;S5RWeW)sPEXdo z^s6e$UwH+=+gZ`wCPu8liA=PD>gR;2D#0yn$a%+TJQ&nWj)0ouHG!o0g#M{uOJD@0DA{Cf5|=Avpk>JG>$xoW?SSVjoW;B1aN z1n#m<-Gs;$;6#5Hhkv#WLAa2Nq=6R(buK!mFMKKVY!pLZsHB-Hi?K!By}IsBrxuR1{!74ANcyV5Hq!nkC=aqLmTqL1RF?+XQ}Mozw@a$FOBM> zkIYGRB?aMU1tgU2&z+A^0jsI^0gFpdv9IR{h4qI`U=0MzHT#wv;H|l0rHa4&urE;4qg-XJ9X?A8SZCZ7e33+cGG$CmMZYsEpH2a?(o)Pmaw-C zn6`EU{bVwbo-8kJF))aY^ZFyQGc}2yx!^xXxUENvfg3VtBX5$0pKke7ewtZbwf*`z z#({$Vg#Nk%q&q->_WIzq^1Q;KLX(@lM?t8tq3i*u3-HD|w9b)E5s^s+N?g!6>+7a14T?hqYgNsyTqbSQ}eM7U`s~hldmHP zGaijMeHmrtt8&HgH|A3B*&nNk!M=){ebS&!@0ui^DECV*UmIKlr^It6lCC29 zQ2O^glUSf{_IBF*T}jD-RyK=qQp#<3!Z9k50{DUy=&+ChO;-fxE!YK2RhaPt;K{W| zJh?%&4(E3MaaL5k0x#HAyr*7t$F5O=LIRKY)F^H{?_eUJSe>J#Y5kvDfj^2Kf1K`^liAo7T+leb)qb&SY zv})m>Isyu-2~qkCaM@@9kARuznLMoa4$UwSBo#gjMr{|K$hNbhL?j$`8 z^EKX7PgXCt{)nNL7FxsgJ;PVMR~V6IKEhYL;L>lqL$bfd0|GpE^o-X6rXuJZJkZaX z-CRf?5jsN6WvT9a*EvsC@PyuUS3U0)jPH*SN0_k!JN*&*ly8$}ptv+`2^8fX1kUS> zKCICAdD{U`_S^FP@1OY4UwH~3eqa3HMdx`{oF)e zBA1<}KnYcmD{~Qv1pBKuc+ulHYA8?2f*I4`3}TM-YMxxo#e*l^0oqLmYz)@x#dEzv zXzkHph4$5+F>kr4`bbv%33p=_-2%V-!E0TYhQ<|c_4&Ymc4WFe&puqNF%irr?|3gY zDTw{{2$TJq15e?uNl#t+yc=plebG&wmyS;HNw2=aFf%a zfg8NRZOBrORj5{$OylOs%R|b={Qw8|)8nFN>*rmFZ;`BTYZCMnf5-NjMHw=VAblQ&8!IGv z`H{>tvHU5{NXpTLnW9whz@9(XGhWLc&yG+PZZJ4!c~aB&(d$a~>zDX>*Urf;hN+T~6U)a^(s?&twQ@>OfHf?nzM z&O4#Mz1Pt$jLvcJb2&(XVDbb~tC3&7kQJ86`FWvLVFC8G;)*?ds$j<(s zF8~}=dzm@#>ac{E9bw3yu1Xf4r$fl>c&dPn)d<qc1_%91b_hF?h2i45wB?7R&A&Kn{tDRdklY!`0vu~GN z(9cg|anV0O$-k;V#7yr_uQxvY%Zf6ky?SAv@)dw2(aCug(cm@mHXk1F_pMg;?SOmSo_MT&8?)Z$M5u+a+^3_4z^I^oRW@6m{8+qHsM<-O}i zGA|f(HDpbCRc1CRL=|`s=e$`PXS$)`vfwH(CX+uua&1+w=RX_(jz89r_zZk4aR)0S z>D5GCsrR&=iKJ(dFYy6KFB_9l2JN&Jk9=bYMyNnE^cTj~nbf83s-A!TS`RaT9`_^k zi_C+4>=7DF5{j;B1QGADmlbFFjAhJv;zX94m!BI}Ze~TuP8+Fxnu&=uVl3M*2NrK~ z#bB$VS5zWS7eKjU;Si>yC+x{yS9?Fq@ReMhQbn)+18Cn@K3Li0PI9+y_~B{z;B^t% zLyCt@o}3Yb(%x+tTf^Zks|P6W{C9@&-#mqyg5}X49~>B|PHY5P5QO?dEW5{}xmW9m z6qYVn8mC#PYVb-KI;rE;X5(fJ>oTEz{My;#1%cBb*^FVPJABqh(T-jhq3#nAyRs#g z6vXW(C9QH99+cy2_6UBW&P>*diy!)0e-`q1yWgMpE2*k1eD8d*w@UnWZgu@8S#jN$k&a+%NC;7 zRE|NR|4ysrahXN(QN^uiEZrww!Cq}bMOOI(Z5!K|+UZ-;_=((e`&)fxuN3uk?9od& zY5Q)kzLt7G;Iv&<^>dvY>|eIM2-&sxc)M`sPgk@=j~N}ySs^TyX^f;3Zph2?Tc57$ z=xDssxmR+ZMRWTm?M)M%R7T^M6Q$z9E^24NGTEv2GMcD>QCKq2x+J;m;NpUHS$3;# znRUDsbAO}4Pty2 z^-Cz0`A+Aww)f>TYGHE=sCLoDPUDEJ!XfJRL+;MTD^RlTHi@E)Ef(JD6?VsYDrR1r z12jV4!L)ckh1|3Kkv(h$xY?PO*|lGt5W*el_KDU|F9zKlyzH$Py|M)NwfU)GkJP|06uG=w9cFzJUw5CA8Y*GcHpOJf*#LO@XbgWBI#yBZ@KrN>E5_`a(n ztIvpw6~{)bc(ceLS|+^}pP@IsJ*B}p=EQ4vVgv}?nOBLkGA%c=8#P0&HNHNQ51~K% zFnojc2=SuW&3Pa0NngMI7T)Mj35XuB`~vJrqVPuF?R@=MMPfp1a62bB@?rSW0d>vL z5$l$&_m5X~UZCbI^5D&9#!aY2a^L&Ad18OwE_?+1ZMzXLXGPW8%Mc=&LRN*cgQDN3 zo@JT^Rm@b3te$uECX{^^9={@WYPH%H;werqqTj%8X6uxOar6+gbszQIK^yQ%a0LF-6#m9r3c$N@2)Ro0s*l9} z;M}W!SU(MtL*i2B0TzAg9ab!**W5xR!oAp~fQ&hfh!KFpe-jY%L$}qP80YU2E zL|j~#NtXJ)DPu~LtaUR6!ArwiOBz2c{ieoN0HlcTgJ}Wjk0rHKIrfZGF7mDzuYmE|YqYBiKJM5uR zQtw|X)y5mwV^=P^twn#1e7b?XbcZza-VCIFvES~WKz;1OICO8n2A1D3BSW)t#Z>{d3-b_XEeyyeC0ExJuvNg zQ)o+F_bZ^hF3n4{8k%XIM}l0pblvhJ zvj+n$DRF>b(eQ&@HZZ&iB#%U+b0SS*PKAq?uMIWI#BZo@zJni{UKes@;Vl($m=!q^ z5KlTnBMw4O`w8XjP)!-Ya@6Rtu5+VBHCA1fEq)r6?)*so1hy)kx9{zm&WrN0Vfpg;6?P_0jb44Ye zQ}doVE4G3~*1EFeYsGyj>=KI&`Sv9*yX`xn82i~RzDD$RB=X=U@w1mE8t0eE&_#4R z-zZ?8hzV)=b<=ChvhEldF_f4|)ctMcF-zKqx1}QDr#f-5)C9~q-wx;7%bDG0LTz{y zKQV>HA5Nw^zfzi;@hIrArjdHa(w=)emt|+&j4U8VVUTb)@7nh6jCWwClS~otsz-tP zaG2b zijzMR$Q9ge>K7K$&s!a>Ca0%|$@sguq_*S%r3{l-l&=&+A27D^%t$JAE$qw;5yjuv z9m4Gz%WH!Xlx=k)* zg@eABhvTk91G|Fq z01G%K{eWqf0wB!XKZt~P<;e-)%DwL?y?uh#nth(xpXXjYrQRXl8KBy)ga)kX%YHPH z;8C^@HOJv*LVl%5KZtIMwB{s$(%T6F&uEEaWeIf+hw2@;_X>1I3sy#A%jWej`TgReFm!mT)FGS zKAknyMXh*;U%vltN3suJG~AOAq@4U-9&pLq?QkLFEJbMUb9 z9R+Jw`cqr4DlOPDJU`!11c&lkM8h7fAP51kaJ1GqynR9(i~PUu$(C8kG^NQn803UdO5hao#7g{qPMm;~wPmS1vCm)V{n1?l z@OkgDpAKRm;xN$Z{i6P}-x;zJY{jhG?$}%W@3GJkqC98WUeV71b#8dVFetY$)Xl7M zjQzzq9}bu{$vX1vgKhTmasxdCOiKhk^Dpm49aFl$$jaHfw72FSjKB-ll|KPF-2rSs z`AFU9w~WZk8J-}vdCPwt%I;j-wt)PH9RkUvh4g(L8Q^bMoGTZ-%dUX|JVy380h9FK zBJue(j?S4`7-DUH3n!Z}E1tY)E>_~W(!C0ZB&eu6g-J+&hq z*)s^7pREou$P0G2x!`tR#^a6d2%Oi}p)5()fghX2y8A|a_s2fL;VCS~7e*C$qs-^* zD1wXFOd7Hh!t#R*G*UjFb#|grvy1-5=X4sc_BgG-ByiN`p!LCaXF(E{rWr{n2cb$ zHVpUVLu*sVgc!pa_lMxb4?8+hK$rZ;J@Ef=W%t=aZocz{1SK0W?6b!edy81~~tI zbNY-&h4~nre*~tUkQkrZw!NO)>i+FUm)nCA75IfhoBq;8Z$4vreqQT-e)hQD2PFek z7ShMoa_XLak-Q(VhOk2schv5YANE39^z4a_QpD^>BAd`IlVj;{v*VYC@wPj%QO8M06m}W?Zh{Zuo#tn zbBGq0b|f!>Bg$RlElJA{3l=zvKIGI|>YcV*zxS<%!!7K$Z%5>wa%R!y4YGO}X`@Ai zNdSUW5@;~PVYgQ$V!9*YzefeVh({TNk~MBd3Aq_xe=J+IYf|Sm*9U)$zS|y!M+x(k zb^(}k3ek~1;!dqTOU8#4Kvwly?#fb%-{0!%L2N{^J*fAqI3@tWxGhTwAktd&LoNXK zf=E0HW9>q{2_K3&;`OHTSjX9HdKGeHA_4fmi(|5;j_HPu zZfH?kKRvz;aQ}g~?XyM+gW*ul3npvDCzJhu;5KUA0nCm2eYO9Yv}_1?_(|``GQ{3t zyM113;qe5_2o$>V?TN?Ob{tBrDM&zo0YIcQj#foy0!CTvpHt`H0Z&yjzA=!Nzpyx| zf@4vOA>Ptk%N-k9f8b%zfm!vcm9M)UbLN+egc;bL?Q#=53`0E0zxcl&K(|zM*+*35 zU1j)L_YHwBI>-Fzn0<*||IXO)OPesv9(C}G2iLnf%?hgeF-b+ADvt`?w^IlHIv5Qro>){{wm2*co2%Y<^N;M79IiH{%L$8kYl-h zWt^{%Hupsp=b|hzLixVG0wd9+!tLLk5UQB&cR5a2=1#sJS ziqmQQ7(%94>m3&EY|7WakH2{T4JNpa&Vn`PsrYa%Vw8kr9{yA5YotJtt|1$sXQNQ=6@CKiMqwtdzf%H*+tAEr{>)2tItH+_a z*vB=IZ|W%deDmLQl+{*d_v6ZwGY>Isi2aj^W=HX)HI*CfeqOEmBL02OpEKpDo7ajG z=dHHBPVi-6_e=Nh)sL$Qf9i{S$(~8Q6zynjNpuZMfj{@}V@z0K)x8*{;5_~k$g%KF zleH=6p-sb9vSaTJ7Cx${@7uI*3LlOOwfpV#IsJlG{!vkVS{vBAWN$zw%csn-p)zx# z@`=1Kqt}$-LSZd`)2A28IM2A;yrvRVUjFP^zZ!eH_`AY@F+TyNq#jtbQL$l{O7%(^ zgL&j6mq(6kSSYQ~Cr=WmNcb1FUR|brqa8~^5Eb^5r;2)RXUj*QYDDy-UoD)(K1r>I zzm`_S?(2}m*_`Q!$`FkKomSEm(JvX{3i<~UoB*?pTJB ztB&dtT`Yf}<6}u@7+&a>hQ5ARaiv5G{7AhQl63WtHDia20Qce%B$TS%`~C05R<<-2 z|EG_R-MUox9{%7MdvQ|l_q(?GFI#P^nIUn@^?TSOo2$peV|t+$oK`^;m)V+ZaZ3mL=i;x|AZ)aNMD6`9pCPjvzh*G; zNV=OJjreDp^PUHX!Od}Tfjxe0lzKaOCNi2NR?`*Qv6nLp-OvC`EjwKNl0wIz};X`mu00o*Trv|Dr4^k!1>(SshKxu zi9V0#f>+h$~}#Wz7wjJQ|i{Vjh4UsHp}t1h`^^=7{ElY6A(%u8b0BpsTl z?B-mNN!)r1rqHFfu=`Zp>>5pQdv7$drJuvN%Vn&n7Ya8!{k{0ZtTj>ceAg`CElm2h z2Ei{Eay=BYJOOK{oE ze(8T!=<+7=zsptf{z!iLHZ>6;h^zS_AU&ofn~5E?je?)yYIZGzZsBU&^JxYBPj-t= zYOhwIMkqy|laa7*g{t63mvezTAdKh_zx{I2(eTTA3#v^b&_b{er{Sg)C2XDJA zsnjN+|5~a7%juJDxh;On0XQ>OvtTTxqnk7dFyo^uR_{2sJU=MkqzJw-qOu`!rQ0I>liJgW7*FGcw^k(&ZKHZQdtx0*W{J&+bJQ)JiuWlJkwm z3;D?fme~=(oQ3&buLdk{=SQ9P#?#a_D*o#k4UepGP8@L7P zj@d7LN}}}hqh0!i=@8#EEW`3z{Kmza*b&t_y0?q;U~69u+*IpZo46-MJE4re2fF&9 z7BLob5B_!Jgj&W!9E9rutKW{fw9?18QT!NRO7n})c%uq8gCxigE})rQE!ITu4i=A^ z@k-d+{FJl_i`?7%G8oHFIO17qre9V{*$X7IOjNk|YH;t2b&X-$RFy*$_2Fq&!=7Rl zR^nG9!a#m^5LU|c7hKB!ASUdp;&Rq7p2>o3 z(I6}uE@%{w&-p^3ee=$+M=~Ak&*s^e2b+QJg*~tt!}w6@c!ET$^h5;Zal$gRw0NO;s5+*`5~vYQcuB^>1cfwR~+Px&)6=Y|U3G}wgl zskQqBOi{|S0#ZN5TTOfK&cS}MWLMUWlb48X*KQJNTs{?>bRT% zXAuo!(R(sfySI4*syPSqR4EW#$-lL1lgJXk_`9m}o>mOi;*2E)pOyLQhlgYK7N1t% z1aatIz4Lq%nx+4F*>cB|{uP=3Fs_gP;WV!EU@IBBt&+uxjf|vj7f<4IT5YE$Drryg zl&67aFLzklGW+)+|5;64iCQxc(d-v2D|d2;)A%KU^`U20uyw%L_srqEfWGWzo*14h z{+y2_7C1ndwSw{I^Ep`Zwc{&1FPEDn41FQM@ghe7i+`gjpN$OAM3Y^zFY!h=p=;3k z)Yf=?GzPxVw$c)kh|M_6pI5AU5DqymwSwFni z(~~sDcf_aMOouD4N$+}PPdsAhxU#$$3}!@^Gv$jN&aNeAwxJ^8izBjCRhhDdEi6lw zxEebkKouddR4KW5KnI=Q;;9F=L&8~poZ4`ezIIB0oxff%9qEt8qo!PYp{pu05nu;j zvdc2BXAxflOgVVUtVXE4#iDK$ex*cz`ai2x40#K5+q|L|BtNyK$NKZA=qf zqqk{7Z}H!L@eFNybIGrX(&4xSdsTdQgfD6_w|D?f0P{6n=G69?bL%D30VrBppSA=y zW4(igb^^1DACHMuyK?7!I4Cio4;~Awm9~h9bZ5E~5PN0DZciUPo-1wSqgU|$5{H%O zE--%+pa7Mg(;PdBPjZMBZf}vi(gt=mlPkG zWn?D;hcxHLyxwM-Gs=~0{{2J~b+x)h9WK!oJ;j>@N2`Ea>CEV~Pi0ZEw?}zP3ChyX zlU2+1@a^6o7DP|=OoJjIX$P1MYF^@*Nde5Jy82rrAQoDhr2yz+JIn{4%&Wk{-7mdn ze4=37a~)eZua|7PE-S^{4`Apbh@PL`q|4XYLw?MK?Zf@Yj+UHM=RJ@rVMB#0IM+6t z1#auLlZb@9+uZ<^GH314v zZ{iKiPdE&O{>E$s|NA&47c4B`$fL;4Uc;~`ZTzyMy(PPOjm$v+o8O`{2+N9iG;SpV zyy$!fAFgu2b96@uHIWg})(5pMm0e$TH=&{%ltCAJheJ@Y-`u;OD__hGZdN~bhNb_d zG*x89F(UsYt@{Jzn4YYEpiuwb8%1g_>fSAh{#=bTWWS@^< zyhqk`U5o$iwYLNV)AL&YX0&0gsC1RDdW_sO!JAQ(MyXsj$CAqRxhi?<&Ba>_MUvH1 z)}fe;y+FC~**t#*Ec8{8Yi7&w7;Ej2-nM-<@Yum8V8|q=RUwu82qqe6J;FJs-1a%` zLvp$PjwdHT&UtKY=9#+Nf!hgS)a6(cpo-R+eoO#WmP;76l23@@JV3d$P6Un2-4#BP*sB}=lD>ZG39 z{AJS>3B^43lN06e??a7d8;~#hgi*$QbgMqwhwH^;r>uv=ucx&PYmd+mOu9xn%B z!m2gKK6<9@;?q?8>$(q5Z5b{{;jf3k8FTt(sfX&7$}ffizo-f}hc00DnoT=$63{pWtD#g~k^j33yABCZlV zl&YyObzz{RvF{^r`_$IOU5@hz0iAflT(*^*^(iTK4=3!o=7sNt5`q-I0o-SO+F%jw zroD+!J#bzz?nL~`0CONtl0AB{01}N2BvyhNkUy`92=NRY~fQhJyvg0wu={!9=&{vH@f7&cSd0K zT$P3HtpHP)y-Udyte$=8r6a+SJUh(@^R@>^LCLGx9B33QE*7jwVrd#Mfe2Rc^TW8cUmjkn9 z4CU+k)KQyrO-Rb8di7Xro?6qt%!PHsWM228d--4$Zf6Awe9qA1n@N_NB8WcOV02oB zFUG^$!}3=seMLoeq!*BcavjGM>>fu#H4-OkWOc4E+LOWstQQqdMCrf4?a2_O`!^I| z0~UK4<@K-bKly3!8DCG4Ym)lBzxVZaubwZGERl7=ee2OadFfeH@l zHXrOO+VQw#HE!mb8 zPcHi&2zUMtoC}TYFm7WqB%?9j>4>w!{eqT;Z6zDkuLi3zg#p7_elJ%oKJ@q6W{f0Z zJ@45z>3^SkpKyRO)K8425Oc==RnwO~OSco6<>2TJ4Qv49ox`60fY&N4+8TQG@%%FwVje5mq@5> zv?=K(v;`y0+(mGie@07P4IySH+0iXQ z+7;Ezo9#`^UWVFdH|evb7pZKD>|&1)1HW0@{Z`C#P_eT6sO%B1^wa!{cezq%M52KF z&&RN6Fh9$+FQ!96#ZIm^abDC`mO}D`Vet`1*E^07Sg!wZbmf6e{(rnuDC9~+$Pr1% z9mb+^R-!|WNN#dhjAiayZgOq8k`4+rvdod&Hn&3N+>+bl*qA-P=llC-V|(^IpFN+? zKA-pd{W?BU*=!)42Zf6U?Tom(R9;$jX^4NHTYRAXI~qa&H*e7k&NHaBA84YPlMzaL zD(4!Hk{hHr%bHPo-%Sq7br(H7YveHv=hklRH{Nm})Y4XufMwMpgz&y7)QQtnDK?th zy6eb~UIrB_dt|Ffr_(KS-lsDJsw97~m&$Dl)T6`UH-{R`Oj=Pn2V<$?YRK-YT?ex> zQd}`>Jve*{>N7YFD2p37;gD7rERobrl}R3QL%m-QMaRlcq2W6CJb_mYs zsgyP28SQ4cxcaPiD~fpq*0^T((UPZ!LA42KfH1_Sb@St2e;i_VqoOcSF|Bfnn`T5n z4c0>?|IQ{V^ub2GS6!<*f)(EcnH%QMe2_tbOFEhv6Tt&0flFR@u%CK@s_bmds$Nnv zkbS84JKyG^B%tq-ByII{QkBzRy7?mk`7B#XR4J@Tq@3ePkb*ysrGx|`j zp-8KYdmAO@-G%gA15~yhux=Tnc=CIf$)vSH65X6v zuOeVcXV-*-gAfMYA&g(~r}ZMQiW3XOpw%Iw-qxDxy`oD1<_2PtzsHd|q&Y^D^{8Js zN&cg3(lG)Uos|okgDUMHzavyAuusHH0-d|w_aK{wxwp5K4ni-kldPPc)b6iR?i#(a zS`9CKY~2nbMlMW++>!<{5X4Cg6>PM*(8RXmF!6Nj1dyQ1ogaMscIl01; z$h%V1sH-dZMVnce}D7a~c+ue8Ta%RkR|3UZ0yDJBLR>OkHaeHXDu68l2P{WeVzZ}#h?wt_LMYt93J5d%b>^2x3G%IVb7D@91LK0*P-IX z@~^uQT~9a^*ctrO%n#4+23!Nxs|)vBU4EmogDR@eUUinAkh;F$b2kltFjhrz2&@t#&;_Y-U znob%i(m?^xR6GqQYZKyl zXr#GnK;kGfkYM^ql}7qA64GhDI-fB)ulOi#ZF1n&f4C4!uaHzr&}y^gYLCiGs{J=@ z69B^zbNBUTk^G%1^=ra3Vv5$#)qUb|CN#4}qO^Z|gG0LHt`avUL%;c?EzZ!(pTwcj z?%Br>ou^Ese4KGLIUL+%k-UmXO4mB^ghJ`ZJ@oJmluV^8$d~NXolg^~nUlxbfHJaT zY)OW?U7%$wwY&LYn(0;_+0>)|TPxsfWnL1?Q_u{QS%(Gelg4|uY~V_Eq8rE$`bW)b zMb}{o&->w3^!wJK*YVRf3w_+@HQQJT^t0X)-NJ?>k&$rYnjH;52X@jL+6gkcXB_xx zdfs2qnLY!(B;bEV=jF$ba zrc@mO=OCmY*pQ&*|I z4Ok%@;ax1#!Vu^w8F}D_%zPx(gIXM;He24uWe}Qf!2OobSJ!QptZw2wgmB|$&b6(fi zQKdM4R+Y?Ez1`@_wcnkeH(!=L2RT4qT=v^Q&egKzxZVqW=YgE+{a)kKmlzn^Bm7qA zjwsYEZ-mk4AP-fQ7KAapQ#aX9&v#6@j6wc$6Ww>$6xF%_Tq@aCasQIP?Rx~h;_Zm z49@K-EPFhb?CC>5OvCGgQy$3LxJj+hJ5tj)(6YV)iRTD(WX{CH4Pjl`^e?mY7m4q5 zpDDlG5!s0nq5mvx^AgnEko|6guPRt?-8rLZ)8rV#W zvXZbji4~oIvx#7EQ_21$&^zLK?dtyr;$Pjb<;O5K)YNlG8rE=`zB)ZjE zpFJov?>)JSpvT=K*#=r#m0Vd>#z#|aPG#K!M0l8s^5W8qh!1J>!(AD#>8QuD zf>W1Pz7ok@heD?`HcDF3^YGbk^^MM>u~G4sewFJ;TOk9v2vsOI7;KrESB#wJ+N`y?Y6=r=C)b)LNST+(xTTs8$IzV!C; zxL1)Q$1Bf7WrOsTj;99*I16)oYIO)oi|E@{s%J9 zioi@+x7pd$xW)#iM)u4i_)L{AmwB>OKOex+y2o`S52Z_P_mOi7Kd!zFZc>g^DPk_NFNgc>y*D7jI=@AL(*AAVTv|69loZyap^equA0)g z*MX>PL%Y`x87avP8&pMfFP561E_d+4LLI2s2scbdoqcZ4QTgEtw_zykZY$zlC)Mw^ zDoWwk(?g_89{H5EsJovHI^Sg-=Jr)d4pEvYw9n5~xZ9)ibk6kaC4cUGhZ95jxs0rM zQse$7WZQ5bSoacW{U^}({E%(i-675xR0*Fvb4TF@OSAE)-zf`jymjc0dH!-OjPsFm zZ>3W%b4_D@+E~Ts{Qc1gT;{-JRJ3^foVNE} zcbp*8#En5^Pb|q28e&gmKORt_M*?;Fk6q#ZafaPLwYVl$BQeS zvm<@vkm2+}kC(dN6C9R}Ut*^*(L#WLxOH20n@K1-S>P^i84zFUPcq^tZ(lL)9%g9I z?ypvtz|516!bDAhfSr5yRqlFc*_~7c(YOB{Ct)M-%90s$l0Pqag6SAeV`U!ifXC`O znuxlG#Q+``NnjeS?qWsqv(oLFup0?`&IMpR6;+$(YybrKq{w;&&-0uW^XSJt4Y(s2 zUW=yli+0kb(Z208Jd@?lArjUg*_ozxgn5QJcWo+Pusfg56dm|?f*LW=ccqzY1L;&B zQJvby_`vfrvgcShs$K>Eq#0fsrs&0^rgV(-wYI+-lTBGz_pH2BNa86;d>+Fytwz0FuU*IB5xY6H5IK`mps-Kg32LOd? zXD)#M-nE+8wxT+|eDb^ZgAOL6_UBc|?e#Zc=F84U1pm>w!S>=vUPfD-g*@I`wXto3 z0Sz6!(9x2*;3oPYBfz5tZsELBcSj1UGlqHQrqoeEHzFqqE3tVdGY^jHnErKR8mTzn zb0zvgioq!7!0x+Cv=y%Lk5=H1oh_U`5?X z1qO0|ikA|EfvF$e!z!rw4)~2$IoCrl^7v`lPxK$%ojXD6NRi^m!Jh!io}Xj8jUnr^ ze~G#On5nw!uPBYis@C(U;msyGEG5b>fTGXi#ebo`R zu~yMkg?8Ek5~Zep7|6DR({l7|BC#Lj=D ze`V+EUZq#Br)IZaE@Y!R@`)1#3bX7WNk>zdIvzo8f7B1HyY?;gGF|id2ahwP(6TQ~ zi%gG_Ky7b^SD0JR*F>tfMZk|YRqAKZvhpA6SQKLY^cvcK%}e`<-F}K$umW`aH5BiM zqJ^HFs<7-}$S!=yl{*_3w~UwUp4TbU1!aG^pQXUpvL4@*MEe%Cc0&l)o3>3yeiDi` zo-5qsGU$yzQb8l%@1=B+!1klTaNBhvK2g zDe|wy*)y$l=ejmc<51=rDgP@QVeKb-fD;qYe|Z=>qQm@7e&&Y5!xQVI);p$^AF7_K zQsBtA+bWU}ITN-C4l?4D?(aA}A53_4r28?rogT&KcO{7`BK6Z(^1|W#+&|)->7$I> z+Y5*ST73AC0mir-?~_VTmKM5nh>9)6LW7Xh8%JHqJmL|Ks={3zK9j}@U`U&HZNBxD zss%UQxUSljPA75e!F=hgj$UtTcg|+Tp7GknG$YOl=PLm2rY?P`+rKUxE+najz6_oU zyJVib3{|3O9N)wf@&O)7$f-^1b~;kG)wGfe0{Ml&Ud6`FQ@p3EI~(ti$m5r8FB8?w-vj`oqo-c0qIQObF{dk zIw;5hBrW(l)SKZq3@)&;Tdd(ZJ?FY%+GEi--R=Kv7oi^k0iTqr?uSmPjJHJkg?PJ% zC7hoAn{i;Jme+!KVg!4PNr_ALmbIBD8oHtILvmxvN6B)v*@cV?>Yybbvt4g8iI zJqNAPK+zo1bQgY-g#E2iBM?rV;J0HhBvESXd9GkEFOnC0FT=<_mEO2%%qTMICoq4O z{D$Md;h1*IBMd`czc^;iLh!ZdP*~FkCMu8SMZKZ__=J9=qmH*~FcBb42yzD&u|c|{ zQ6;8{ud57{vZ8&*wwPX9DPEl)8`c@AJ4=Yo6w6DKiwN8ItGV%PID z^~pd8HKy!htDRJ9V(|jvqALi4Q?D`&094mj@D%!2NAM+Kv@9T$Dn|l!D~lwJ0aTd0 z{y0Dw;5?>I&2(2hkw`r*uB)f(OVTm{P((7s;&l9Y(<2(>zv**n=K1Jn{iMW9TNkdN z35`5%i5L`7^dN&>^QQ;CB9vgUJ!!RQE9Y zv-6d! z4A*ir^g1Ou(Xvt(`3ajdbkOJVel^RrM<>8vca2u$%wPLJAspNGZaW<}Oqc*(sMN#; zp<1t+!py(32k6y}3nOP9q|XQnL0;~mbKuz-*$a1p0;TKdm}$u!V|V-HLGLIzR^J?= zupy;%3bpNFEOQ0p74qX6^hE6S|E){5A#@(3SkqLktx_(TmeJs&9?z4~exLS|-@s+} zb%tNT*h2;T-ueAupSGreo6OfAC{3hKp=17)OI)oy>)uiu*ix-_IuEo#e9q5kc3%~ z)(qdBNAOX;BQrkUGLUOn5!gedKPNnl)fM{UNvO__cT5d#N|}*mr^$Z5Ko-mclZF3X z0|HFl;U=g@0cR(F#_{}V^0-fvP47Z00KEC+!{RWH;t-`aV8>ge`~qPuN6w{4PYrLi za2y%%*hgYx)~wbvq_#1?wq@OwH2U>I3{rGUTvxf+8Eal?N7X1EieoRLm9e{OAa zv$FrZw#(1*IZvijs(xoV6@Fb*>H;f`+Rnlm8jkw7n4BJt$~T{8gGlViM@t*TsrAcNdhF^qy%|3u$#tEcGGiBgI)AO5X>YB_Qms!+{}aAY`*-1 zyJ%Ucw%qkrUFksw{I@zDQeKv+DUlVDynKpt0Rup1%y1m|RC#KWi=^q(+k9?_)@ka^ zAVT&n*gL^zf}6az-0i2bBd79=od0`kC_m#kDYGpJ)MC~yw!j?@AU`227=iZXwEYQS zBJKon+!ulJh@H?aq*M|b6dDFmY#7W>(|f9Zy&Q&7;9 zzZQaC&pytSGps5rg!!_pBb|()+`GdV=@5>3pM^6bX);e1jpM(?CvBv*2OK7@P z*@AVq0HchBf-zBbnc>RH9f#jA(8^DfmC&0L1mTzYxg$!O2`q%(J;8=h(lBn`zYB?Y z%9KQ8QwvOv^?dz$ejmz-N4pGGZUCF;l#^jBq|9-~aV1Vt<`~2)7J#xGliD{b)f^W{ z6Ya^Us`=*#nEBqO6cit=(1EaZ+hBC^tifv#_b>U4a~~@4fs(#rA4*#x{fl`~uE7(# zbq0nm-6rOM#(rpk)Fbu~ij1lfGa3g_$F52JMimLV5%bIi0vkq{y^gHYbj(lLlPalU zLc1Dy>o|q`3BEiksjM%&q?;Y@1gX5={#xd%m@PCmdAZ}E&;)Ia@hu4I=?cymO3?jF z=X2sHK@HE=unH2&36;`l=L7*%soe@*57KzFzXp?^T+f0c78aEUQyQJTr&U9;-xt>F zYJb^LFL};T{e|+z@EU0Y8|U?sA=F7vxsz@_F{=nEPU8?Ta%-QDtq{-t0(`#*)-w+- z@81g$+829slcc#cz{%u>kpy;aJkYx!@&(0)W4U0a-q_5Wg>Nx$6p-Y1gc|YSr%Le|_l7pVOWMVTRzG-Gq@`m~0-A%!r zJF~HxoAtCMXf0Q`cHm~ePX(i|0$z1pBg~~QS4<^9P#WsO8j@(E+u2L;`lZ4vLEI3>;g;KtX!yj2E$%v@uu|b}HCLU}N zD6x%VF#nv-V3Yu};RX7y05dNNswZ@wgLG4ryYyvV68O9}Ua=W)ejyqq1Z^Y}5Uj@X zv$PqGWoF3tPJ?BXpCBTNmzPGue5J-7VNh>yHH1Q#pzvy0W<@%}nSI}YMq+49JS{!y9)@}iF*-?^4XMh% z2KK9Cl3z?CHiw@2G)M#fOUcl{&NDqH2+1p~D04ec#7LRDBaxp^z|8IOB$LW@n4G10 zA(V$iAKDA@m7B!G>M8y$u<;6-xuO=rDVTB<3w@ND`OCdri5gtG1O)u4!#Nm@N- zXAH?Pgrp)s)qi|W(#!b7d+R!2gVxOv2w3QmtAfNV3;cFQTMoHrxs2VQ3R1nZ-qh}$ z7fcexIZQkd3T@uhaWhF%_wN5C_LX9sx{onQEAlDbTuf4?$xwVVm~io)oDN8xKY1Tg zfV7it+=1}Mm%EoAX7qJ_xW!Tc+Tq{)d&~@cAwG5(q%VuHn&qKo|2-YeOKrI=Y{kdG zsBl_-2H(Roig|Y&0zbUFf8+69b)hbJ%ULYDl1QlBgjrvC>S4s8%(MWYMt3HSs#Gj@4SKBdJK*@AH_c@0pCwXRa0**<6i`bI@_;VfBZMu0G?PVH9|0R z7we)^(fnOFv7!%9s6dCOM&E6{J&i%255|IzMMHB7Im7pLW5o}~PAW?8?YjIoZhTdi4>PSt6aJHO4enjMtsX}= z7g?QZ0UMZeZO}mH5Bpf0+Se)ELlWQH9lQ4Ahl=>4e!P-zFFT|YZai}?60}=i4%CB^ zocp4`Q8EANoY2@m^e$e;ke+6qkzf|nt8J;+zf4s;bUmGg2_(MaEEE8#<15#`a#_-- z<$rCZXp#3Q1Fy@L?AQQ|?bwL>1t;cEY$_xrcmo$xW2aTjwlNUZ>U{gv&UhNt&c6up zvVv->+QtnUA<$)PvxtB%9Hm(r&`e*niVGrT;#n)9J21Y+C_u|9Wc6mOqnHpK3TEEw z#KlL-g!UjGre3DNYo9(FilvjB?AY>*ZjtA-ZNL$O1+7RhlgbJ=!#FbUrWFh|#SJqm z96~vYw36v08$87Mm@NF>0)8RS8hOaVd^uG8M)#`qkorcKL9L(9mm$z|TDuE5xhFQZ z7*01gVZ zvTFT8>1w<^?r~t1k^j*Z0uD!VZi7!P0L+VqPI%j}yDD@b{Zdky2g6cFj1(67E^l`nV4j#V@yyIuN9y~>K41az5cH$`z)9=^HFq*-f zD+f=cfq_PpbWM=@XHbpZ&7;JDNqV3c6^<(D4N<9vHbGMKC4uUsJYCL^ zVPT5Rw$=b6XA%c3tHX-6GDqt{;ebSu#7k(Z4J{QB%wzQ!V* z)C-0*9W2pzs;LurwC{iakQeD4aVx#_X+*30vB(KGZ@F@F$cYX1plXhpu4!w_Pt1IMnc}^ps;%;EqCi-9J@~*|ouo zdQ@l!Q45;a3NnP5e^e8{E(H~e^OJhMQ@M)k zu4-Y#v3A0@)b3FyQ&dtMG-FsR^WX!dJ3Ypv-dz{YFbY!n6GHAdAxRtqbUJ@1A^^?> z7NI}eACCZoPCNs~qyh}PrwRda+5NKk@@u+z!a|g&Dk#DBmdB9-_l#MuQlqw*Zq$qV zo_=2__EI8X+M?|OeL3UtdrrRCLOScM*b@(3nP1~~5Ft}a^b#n7ni*mLCmw^wCO>su z=-cg~|9Jcd_BEs;q$=GiOtZ$zo;EOm_I*B0e?iT|`{Pp7sDQo%ECR^87k^XwMm6$B z%|uoq`JyhpGS2lz$Q0vLlt@{qeN=2{fPK%#VWwslJ-^;L%tq#Ug z;z*69R~TK8pO7S=lpEYM)5k^~ic!VSQMqb%ae_&6*#3M6`L;XTsiUIashlDoVdC+U znYP;{kl&57#$X{^A0t7#IqiBt7?9bEBJJ1oE8PG@ZuqRf|D)>l-DBdC3#7@%h^s%p zqf(#Jin!9oeSvtgyCF#?2*r`2w3GD@k^1``WWMp9deCj+Gr*|gN&K>f+m1iC$iAf3 zp;lG6Ap0vjY~r$yO_bAYH}XvDDfh6X-*T)0jhjU?fw%9iO{po3PW#7jqz>LUcR(}=)Jk6yjV>f(hZ=bVq7JhM z*i(*WE;<04^?t9P5tSOyhcSU8jTIgq727+X$M5zGs{E?FsZa8%2&VZZQWn@j*`T+t z>13;qNAq6O;p*RxH(Ull1P)34N@{=_Af*!4aX<3`=M`Q9*XP@+Ho%iR{%7SOFv;xM zxZ$JSZ~)O|;<(*HpA^=wEJ0W+R2OgIl+t?;5>Mlgf$Ozf!k?q(alg8Y3JM!1DU6Fr z>JWes5dU`;T0)@WTBZVZ;T*1bFd+x-x8`6j0o`8Ncg6j*pc_2?o%vF_EI;X^YGyzy4=Al$? z47c^+{}LSgPnwau%9)=@yhIj^u4a7=8+bi~{F z6nf{2YaK=Kn}IgN9#3+(cLpMi1Y~se=0SRGB*dq-lF2B}745tk8TZ&K@oM^-n)8x{N5)5flFV z6m@2${*DU$>1jf*LzF_+2@=-Jjh^E&E8g+6gsxy%HPHNYX{j-r(%zzEhRQ53Fzeny zvd;i|pY(5F>8@n~X8T*3nNP%OGdbP{j{_--o-0vryX6Gv z^sEzc&stwkT*U$44Egd#zp)^soHLbk z*M@M(zw)x}&rT7k6TA>OBX)Pc<0et|j16E`$MM;1cpG>nes)$onEvbnj9ibsQ}Kar zIu)OW4sD~6M8w=MWMC++Z^KWYml##V_tM?|yL|ZTMMOXxCHKm!qDcs#1)KrrAMBN8BLUzYSk@$jH)Y{H5cl$}CE~oQP;F}bx_Xc2TBmpTYFSVV? zN)L1^N7pN3wrO(O8yzFHiNIrXmS`~%i(-45lnA42D$3Y~a8_MuNg#{3#USZ{yd?gi z_zMDj!(fx5NP=f`3BF~5+cs*>i1R0-^S0o6l?RExWM<~K7Pn|t!=WqOQbA6{;*C`O z%#GJjCRZvvko#IHfqLA^Q1x`#6$}d`E!$e4qaK2OI+1)lfucD?TTVRU|43oQL7_ho|_vV>o@1FnxTC4qYyhyek@Cyos=8C!sMnsle z2w-7vzk0hGLQ=+)uKi65o>F1zyX1kF$MdqKsG0J;eekR&j~D&_6mOWf6zcjGboGIc z%!$6|^>kEp$pOvqS;WL%pM8n=o$o7xufeo^9{}?-KUXP#J|9?zJZgdnW{Ow1XH08F zjETOt3*#cc$VJ5f>p6aUxi6q=i?sD|6U?k+)gMB2@WV{dfVf6K2&2N)<S+E7fi7(}{vvov*!g%{9UB_kP2!KymOG<`=gw-B zXG&bVg0W553k_{L4B{Tm=G|wozvP^h%y-Z!-{e0^bXyrsegon00pGk?sX%{jNhDN# z;ZdoDC>-B-2}#*6-fUYIlD-eI0r(jy>Z)-9C-K^MccGsE$Y+tLNsUkvC?xG|$16ID zorn6WozyKy#lHJ>vnk=g>V}GJNNAJAYhnY(+f#?Ge?ll<=z=tqTa9H@I4MhcwTiyr za1C18&OJ_rz=NG<-8~4%zvNq{WCV($DS(KYXQ(V|O@Bhnm_0g9lx-73;+_#j=e`^Y z2ygqKT2T34vgA zR?gu(3QTDD=LC=fZrT8?;#%kAXOI+GEQ+vp^o(K9HKOmShX1~{;W#(3B8ty--Vms{ zzV%{mhK~CE=EfA-7x~EsSdyt>$uy=|NDs_G%QiYE&g=?zc6VVO3-d5FbH?yec?6(U zc>$v|jokUusw;E_qjWQSA!Zc!u(AwG6#Sr_TRd!2s*xi0{tKHgaty z1j{!)Xn_}JvHGg&g39i3veaGUz@BN?GBiQ=z2r2qLb=;}s>pOh_>&(*TG$oxnf}R3 zFmox{^?ruBABTF z1fYgDjc&Ih;-#@1VLq*h&l$hNCn28?)yX)D<^yS{D_Z+yAdJ_#1bv8_sW0`ohDiTp zTKCh?WF2?#F-f(N;!**7)B{MrzlAEX`k{qE?i>Q2&evAx#6Gy841US@AlDG;#0*SY zI|UWs1ZLQv+T)J;-ki-^2I6{-^af7dT}G_xLo-~tHElfb=I>Rc@NICV+X@ZhCTR(0&@m{_lEpE853*l590i$M?PA-gx%klsgipuJqNc9ukA9 z(m5k3Hqi)>>fZ`f+{!mN(hl?a{B!rJyvj&+XoK)?+S1L~Kl+V)#v%xlr0n*c`EPycC>xHDg(oToAr3zj1NI9<2M``vL1 zDnTBM2Jo{Ish>vLF4bB1>n$bYqy+?+U)CPlh@-&=+Li0vct=lgn=y7hedgI=G`2i{ z!-A8cy5-Jd35a*VNw*{k;O{eDJTS7PUJlp4n0T~Ka;H5vsai$6Qu%>p2K_ZAK2^-! z{zM^S)Sw^&H6G?8RfJG$HmU5#C4c8b7bCi5$LWQ1c#x~uv44orj_(1#zP3Gol#GI? zy`T-SS&^w23h~0M>=PmAsX}c zRmfhFpOa?RJv~(B=9?(NJUAMK#&Jb9U`jyzHkc$x{di;^3dSsO=oV0Fz?Fe z@VPK+?E>zP{Xl+r@WeX%VP>o5yUB=!XF}yrE#WJ>MMdZ!;|Q4isUSFe-#*ml*~2H> zg4{zwYMv@!g3N67*m9=~Bn@BF7(`>)4nmmK?yq`U!D3L`U?mwPCatyx<&#!`xTJfS z*SE*)3TUuXm_EX2lDmE*1|>O1)^)i^%49gZw!xuZe4l!dt3(2w$lb}tFyCukKmS@o z^YPwWC~pM|e|5#B6On}`kbr?y<%Pl?L z19M_#K9)M+q@>zK=Ry0X_J-dKMe}f7M^T`0RjEdU15Y8nz^9@*t94R{cwTCCqBZdc z0hMnQyn?rE2DN@zpyhm-+=95H1YfnhYnRGYQ;Et|{DiggInL!~5Tfw4R!}r=Tv~Zf z`C7zxd8cx4?eCQM^Z)oW)hN=0rGmv}h-|a)Vm#tZ-7L#w;%pW8Gh<#5Q<1Ix%5 zyID*lfZ8hzr|hx|LIz0(Mv6xnflv9{#5Rxqe>!tKfeH$+LfQ%VM&vJnQ{-!+VZ+X0I8#1pD}U4L6bs0wB|Z&t zYPs;b%sNRn`^kZun68Xv9ENg61ImQswloCmz70|gug@Y1lt>zU+>JiSt=fsbU^pjQZ+I$Ng!_ffDPB8t}VE^j#Rtd^Jh& zwhYW1jf~Mz%mZ&iS~R^3-X{#5Ki%*2&M7(IKG4s$)v0#PR15AnWTJc>i?d{l>5m>m zv*d7uGT*v+>Zcj#PA>_uO8|dqHgsa9lyt{32m>JkzoN+8znTOGTRd7;E0~FrLN4pe za!$qSj?Un2O+OUN1EXa&V?}Lw_#Jp6;F<2 zdWU^5kQ4XLo6Q9I-%+~*F^BWYFZo~%yzyReuNRjgDi<+#mJGZ&F?V4aleNcb8IH0E zn;u}i68NRMzBLmdCvOicSvNN}Z3ZwaoMqvtu!(7CxR;%)Gs&5r^`5;I(w^6uQy;_z zF069+G4BFJkZ$Arg5PMKQF19Gr0J`zW2v~F0cGH_Y`N&KK=ei`ul{k!%Pk}#ZsF-D z8k~H$w++i{{mce=BH66Yo(J4rwaF6Gm?-rHzuQns-4OMHwv7~d*=D%T5M&I6HwG=^ zZ@&$pk%};%=exj-MZ`pN-hJT1mct*HLrSzbk@6`$7-)Ww(pTE7QvDKv#>w=ag{VPf zv0KlS{YWSbt)e_+Q5x82i=96L(l_|n7BV{1|9I%P!lXLzZ!n=_s`VPj-c@KtMK!~c34M$~cGmKpHn)1iGa_GUSG}&{c ziKY~}Khcz+6^^^z(G*Sc^Z&dZDmUT=h5q)Hn%EidIE*wD9f7KaC;xMV8aWVxBD;@E zy`LfFk>-cMvBbVkINL*J71djYSy|S1Y_$pO$)JkLNA!^NV9J6hKe z=)Tt(UL>p~L+tz$CP1D*>K!x%+P5LOPGdj4hmjwsE<=()?O5Fmx)5aMu!gVKJjOip z!}D`Y&IQKZ(AUDg&Jf8)GHDzIL32S&2Nl)ok*9i`3`}Ro_}`80Y4DVvB?RX@Vl{z~ z7u_OE)yMs6oK6#xdR_G^*fV#c57>Ft=T$@xejNc3j7P1rsNQ8ubgpVT0agR(OXa23 z3b8Ntc`eASq@a4RL2}Gmb;mz9#Q}^Xl`)H9@wBGIpqL-lNj*mc+q@VauScgnEE(d^ z`!dKKNEljs3Ztt!O^3&m#JZ8N%G1#hc9raZuQW*THHX5ms`<^HRyVF`ag{INzt3ry zbC4A-_na&{8vMVj0TH(MAMX7$#O(Ly1gf{7p-ykeQ!H%nZIS=)H8|8%k^f}Fs)=y> zX?S%NQCvwEq-QUk1OEQhXo5IhJBkl%YcDY{{0K_WiSt9a7ZRyGuvsz45O{mrLB;zN zMDn#dd)xFUbE%1~4dS$YS6v=pxTy;Qqo|iCRIB;b>_q*C z{#$-!D-{h!-@Mf24#C@A6>h=IR7-Qn{MR1VI>tYyeDTjb$~hp0?Zx#5wjzfANkKlw zkkqTme$*}G8!bpIGU%kM={QS=$NAogrNXT@*K;2%;l~4ew+Z#@YygzqOTo-TKo+zt zT%MG9WEz^exP#FbNZ3=dF6$Uu!=rD(*`PiHz$Ow5%;(StpG&0t_xMk?N@}lK;qW@D zW|53xuV6OANeXgeLQLy7GVwMb^X%AR=HjLT^!OVqtpROal~=O+jgMp-+bM6`KC?soyKEQ^yWVXk535k53%7u;Ae>9V|1vp7SCkSCdrAX~yo=0@uPDmz;@drM~9 zVPCjq2HeU@xpvUF8;^W44~J|sV>oV8o_oHgk4?CW67hyBxtqyA_1=$#zZOg&;ilZ2 zI`GA&VtKEc0|xf?K6_9daCzJEst+}V4+6Evj|;ZkuLq(Tr}40 z7WemNG*(XGa0nBh<`L4LJBJYIK^Fe$dwvFuE$-_?h}z}OAVmIw$R{%Iplhs1JLI_P zDT>d1wYL9)*+lSoBdwvBB*_LV1=m8P7 zuf&M>SKjFt(LnWpNO>Zau{6%I^5|D>IvQ){QslG+ej#W#IE#6Qsy8x|nD85^yE$y* z*FlVTP1-*)i>t&4{?XB6HsD*N5;3$ zO=;5H)}qAcdr&9iPU3^Vzr=)o#Yx>Ol8%bC`-p9;0fg|tD!H$n5tXwM}eS4iX?&<L@wdZbb-tvya)sByKhSZF8mE&pN2%sc zUo&ogNwoQNU8)pEzou2n%F%rl)LKl zGDK5yzg!OId|~RnH>$kRYG-4%VoZy&B4pT{7E%gyjrKSdR|PfXUQ)}Kt~ci%Kl#e! zIm(-wS9lihELxuLb66wpi_LGJj-p~q zk*a^UamitFn8?fRm>Wag~;*^>9h(6>1|dvd&XvFMbO(q8I>amve@v-I3*)J9pOdy#Tz z#!5Drfc+m=R~gV$+lCcIKomrjMp95ofsw+bOF%-928qF>yL2MbCEZTCb94wu=coyz zq#NlTJKuT#e!m8bQ_pi>_jNaKJe{c%#!b*ls618ss$gwWQ&!_=3L7LS?KG>_0>Vl9 zc?!d5JYxjeKtTwSV;`5p3aHwiUyv!+)W4AacUOyN%R|IdJp@kI4IgG4DppY0Rho)N zMT#@V4HkE2Bx`AH5l>gz6z+Djiut~W!=8BEkD5^Ux3r_9Ehna@#-8P;42L9bkoz4x z3hS8)y`|P%L4Omjq&g9;6w#z%lGubDTyp;Pck%|KS88&xbx;2`#djH(K`||!obPTq z!fJ0YbXf+yrY4H_A^wEy-=wA{YcH)SWP<-i6Tc}Zl&yKKV>ENK_C$x>Z}f8me74L7 zTe9YbsrO%_&NTRz(_KZTcpokQffcePsAb8OE)F+(Mj*IN6vSA;pVw?K)R6rthz+*; z_sPeSwUw=`i?GD$QRQ1&pEykFDO7ipKO881C^X)cmiYW@x(yo1sNIjHmR+7PMpIwFZMOELc<90IQ*Gt=pdko--C!2I>?x(3YQdy>D+) zfiT){ed!SZB|I_k|Ed(UEACfzINvikVx*Q7sw`idA(+@$1kbi806 zLtO?mR(O6_$UN|gQEH*V{Ek{JFsF`uCuG7$^;r>)`z;y~hj72QaP|pb|5)!|*)uTb zx`|_e>~gF1Fl(H8&Q4|b;h(ryX6atE7cXADS1#H-rZdSwO{mlIFs01}WCt~QpE^zK zyYT(IZ>-_tn7SVI+Gy4M7m|nTlgCl9T|ZX$myx&E4H~@X8=segaDH7^C4uZ9L;|8h z)GyjnQ?04=;OBOCE{BO@5bfJH7R~b|+{I#o@Ef&S8-kzCr9XDH&QdDBb8S{mV;9hd$>JYTuaI8g*kn2Pnz-Mk=&}*`DTe3QndM+Ye%aC7L8&yRM zv74Ksrr8*xE%*F2!STAdA+cO2fDQ8bFLM{@h&hK5A0=JSC`0bUYFN37C+ZEj2_9KU zzus}SVuU=B4r*o=gm1VC?3z=TwP50> zjrIEaFL-KX&tao&C0;Eo3d&kvv}ShY4~Y|}CvWEps|~&pMCP9Eniu4$^^0pQo2%3~ z<#dyOnjPwwhsUjZ(W*o;7AKHfu^8;~(x!g@>y=`~mati*6RvbG*-EN6PG4)ecu7+D z=7&-79D^GPUL^$>Bi0?W=Ka*ZC)`;R=NQB0_i}~KEtx7Lu1i)q6T`}ogoF$&`u80P z@8H_^Qap&L8E~kHfH($-AHJTtMhxzCO84~c3w*=cbrYxfgbc(-KmNtP!~$ znV*Wz2Acl+tch~Qb;2o&!LJjV3$JiZzpM!9wo7DbA*-}Vco<+0m~iy?6TFbcVs&+4R#YB$w?11!ZU`a zf`M}>;TI>BsBEV;`akfb>^mX`kgw4?<9(qnOW)y*8}o0xWph%0ohSavtfp=P)%ESz zi=aEW%0ppFKBK?84S5@wU%s|%b;;l_IUW)}$7G3&h&k$LQ}S_7`Z8$tvlW+wG~L5X zdlf%!7nY^vW4HzGs{C8`F-zZ?nI&R%#!V&p4+Oxz@GsVNa=V`{E|Z3-+CBiUSJ*yp zB9Ewnxsa;K!n_z0))SiKoNdoGH4PNb` ztk{$k?>~L8prpe$r8xyKm!^f^@=POx{v6@^ZR)1=aeH19l~;b5OA7OE2(8Y(w&N zn_OsySXhpyOzBI4OKv_D5I(LQZ}eO37r|p|UeNHpk|&QjD$%`Ny$f|8oNxYVxRU-W z|Ks$`6h#)M|6J)jG(E%1T&(JluclJ1*~v}d3I4ahHB#>UUlyvgo3zO}L`ie`r154# z3(f1|1jD8-i`N9D9^}2d^2H4wRSPzpi!u$44af&G9ltIiv-WFpBg3@R=4jePa<$U% zZ8ra1r3?v(R)T5Hwh7ZaH|Y-ivL?LVlbm1fXFdjcWkt}z6M6APPar0H5q@??*h~ZV zG%womeSV$rjSp2x9L8ofF3v@^64#yTvG9Y&z|JbMQe8`{1iFEKp)(V~$m~02;Uy*6 zK`qdrN`p=Q1XT$gZBl#hHotr=iz%oRcuqM2yQ|~r{(_rk@i{$jSd8FJI2#^rQElTQ~-uBqnF zX#<(ux4;BD?Qn(r+_Y<5YmQ<-e?+h&#ev52cd_5$2Uupa7t}&pX%dXMjspKK zh)>f<*I=s@w)JTfsD%#DThw1QZ#}>38xMa;*Sp_;#+1|;uj^W=RO%?E+jPJCF0-JH zc0LdQ7DlL*uJ(>z=-G?8{ICLGX0}hnly{YC9z1JjQ_Q~8c&pAkMjGdO)w;Ym*xFB$ zlR1&sHK4ZeewR1VAmFI{w4>;~*t%)ePYVbUZYG@l6oEe~zb&eH>;3>s+l3wwIrM-T zQ*HD~%vlLCC#CZ4GdBBR6zKu01fS7f_P z{Z*GML#HcGhqPf{@j;tS^A8<^fT|MRpq)N8j9VgW*6bBjI9J%6h-8NapQ0;o7OlW@ zeX%58;z1~C(_HbBk!p{wyiSw%fI!9mL03U;Ujo~zUg~4;#=cw(#L$^?!(U%Pp(Jnv zim*vnmPYmiwDyw;!E0m_K5aqX?8OzdX5v!j0L?>i_=>tN6%;Rb^p#Pvc>xuo=CjE7 zEh3;gX*@wE+@aV_zGk}gX?YZ7rErdc9J4x{Pm9NzUBwwBG1A7y1z9@bd#`A~hY;=P zv&e=y4iKhG%{kBKtyungq4%R8&zF!b;R67x@_$&SZV8$)KAA&>&j|2L=72qVLQh!8=-94rQG(?{8v;gC<9R|7(E>+bx zg-#K|w0yseE}^E7z-Zm3Nvb5q&Wp*h+NuB=KHQc$R=SVF#B2ZG-k^Adflt0S!I(|W z)qdF5_ocVHizgHg(C;JZl(721L*1s_+x|8aT1f&*xf5%xqb{d!;p2+W5K>oy5WhP)WHTuf)0Z&v&RZg~v1{<_5ylGiGxNY_>Vx03yIfo4$vVphqLT()%W z4gn?MM@{&vnp9?xGw%6iF=#!9{FsqZXHbcR=_Y^!Frd313rqjw_0zTdq_OZuyLhAQ z`J?5)P{cfUw2n$->n9ew=KbAhO2IoeHDj!F{`5n-JE(9R6DfL46V8y8iWf4Z)+wH2 zC3fn>acAG5GVp@l!w7W(DXDh;*T|t;OwO4RF?3A$Pmv(>q|w}-V)kv7ST}XKy^euf zO^k5k(V@2*v${!pSYOxusEMaTR#xyeV~m806Xf_#N8o4O63dJE7>6|M-MEZ`o6q;L zZO_pKxvag!7LYpUGNLGqO1;-&`MW^7OK2Z0Mu3}ZiB&VM8*a7lUsL{j5#Uaw1|QT$k3`+rXn}*+~rLXvni%k_#w-Z|bTu{&AH7wyZQ0QyyO${tKk%<1J1b ze?ESyzUjVZ>cG%1X(C#J>1WvGI=jGznN6@C&H*H_nkpHJW0nBK0 z$BV72k1grIUK~%3u4~r=o9*AzDe=>O>5}m0uL}>R5K9z_h0ZbC{q5k33e=uKcatG* zM}nYP7fL0+b=wDLV}3``q934`LOsjC3I%xcR{X`)^adZ z4>gm9xSBRT9L^6dj{&&T5hW`5P1$|8|PHU#+T^*rx8|=*aBDunWD_?9C3x@>Ze^6s`_uY&> zm{SKcsZYc%Fp>B3!)-CHCZGm-N4VRpQg|RLEwKVLyGH*6-P9(}1u{s*wdhHoINa=l zo_v{mGW+x#HZjsd->d|iSQ)k=QPKq|&nvqw=<+K667J(TDavUPrz&N!3iXVMlXdzH zSBccV-N*fN;{3j}935o8gh4DKb@ue1YDFJlJJv`5wIBlOUnVF<9U^@PM0NuyA^V?R z5gp*?*Xx`*2 zbP#V6@PR&3x~XEn27P9uc!562zm*P_!;>9Y6etoLd6w`Pi2r__b&Z6?E$$`VBY#E%pCsa-wi;o3HGQET^d?OmhZRdPx4ba!+lAK_w5znp#EHd3wn)P@8IXnS5 ztP%@Jt}qjI;GQ%!)pLMT%`6-gS&*hVWhIjJ(#qk{?yITo`8cHc#~Ky)oNlV1-rO!c zI^MD^k7_3M?+X-* z#qt44zP&)*rZZ3U9NZ!7)N6jgZ(;7Yzb*Okd`Q$l(Bm*)!uKP)O~0iqYCdGul3qMu z2G)y5U5H}w3k#1kG7})8xh6x9Wgj{^Hgwdp7z*6sA1<&?|3@j)#0&% zT~EsXBkkxSow?e-^)o7s9PYhM!VLK6>-hktM<&7`P~POKI|NQK+FMG+?mBW8fB=!>Q1*4c_eNE$$T+F zUG}-vShZN}g5nVI*?3c)?GW)tjan{8^VTjQg(e&)mzS;*Pglo~pMBOwn(HU(+i!K} z>2f!g1Ae1#2k4}7s8Pc6>+akSG%xW0*CGh*yT5|j0pcHnwvKjbwpm)z6o92jM`XN< z!-lSgv129ok`Ud-e0nN}lKhi0dI1RGAy>j5ytk1HOw^^i3v&0XDZbd0cqF|Mdv-(f z0!ogcxC7|?DVsk-&1yQo3-2%YS)bQ`X*(W7aI^~BipJmONPEd^f0^GFhETklGg{CW z#IvxB`>3Lf6-`I*>^(E3S~nF$NAxa1k$<>dt|}|@je=*lhLb{NH1&FN^FHH6mk8r> zT0ZAL_6#6ZE_9r=Uo(BeWOs@P(~Cz1Un3pw^oexi(Tc07!JEMB(6u9QtUe%fxH|^# z!dP4_eYC?Mj9268rk3-6+tWfuM4~UtUuab$RIl}#3`@DMldi|pec1?s34YC)VwkWa z5ZdBC`uc25)u42c+6KgpRmzlA&=R17Yv3mxVF`y=m`0P4uZ-qQ5ODNudn>IhXcOpM zoq;=4eEy(ZZD6|-)Jj5-j%YFF1aw_feB7P`ia(FH{8S@lbzLn4K5kd2t(!XJ)b>7s z=-()A2o-UXR)p_#Kk--S2W-<3#SN2>$C@>8+P6Fm_~A{8ze?9kFiY1}Yo;t<=0U%S zq#TY-KG|qbePS$=K!4~dEye*~qWjxm=we9xqS}DuBP0n<)sMueyf1FJx3^*dhk&{h zv?g5Ncoff_3n;My`RYi8uSQs4lLgbX)k>CB+g7k^*dG*b&^VMX3NlHKU)-@E3Xa#2 zq!;X(c-Kxd+EVow0$ZHHBfO`G%U+fo98$Z^fM1sjjVT;eeWG-MPP1Y>MARaz|1^jK zoQU(DN`t~D5p<$rG3WW?;7?=EvUItOJLJ{ox8%QEBQHBvB@TBpC+dS3Bv*h}?r?#s zti-Yy>JB_Q;FHT9kj_t931}Bt3yNwj3u=S-Tb^_#Q``_7bo`wtRlYdkeid2iTCE@Z zb_Ia4tM`+l-bbQ`dY{zD;l^zvW=%fAobm2&{$yfi3)%kr=I{@vG>581V0OyxYr6C> zPF02P5J~jQTjR7~S7)o$q>z;+RM!>a+XMR@Y9Q`F1&Tf*_f!k7alf$Tun!P(Fgxwa z<~SMz#LW3mx=KBg?!~Qpf4wYkwaQ35QUhSfl8%eXgO+vuq1?;c5{ck>&siH3)#lwd z3vva>D}Zq$sv_{#h^j^gz@#9K*9GKPP|txFSv-26Yo*tGqAgZfs|5uJLWcJGJ%Rjh z>7s^x2nRk)Qwjo-T-p20T@1IrXUPa!1D}||TRyrFZC4Tp;8PFvHD{_8(^$zP~?y7bR)9V)sS_DP8d{Kpw|06IL?ZO!!S zPewet?RJhffRAV1;J(&V9XEJ)ef9A$G6rb=bnN8KtY9Dgij~wBSpUflT1HUBfQ&@I z5nsc6D1uDVu3MsEuyuaGVnRz}IcE)otps-+;3T&MRR4P<(~G$6p2<{#73dz|aW@^0 zW^I_ag?PIBe5vJnj@Gtf>;e<-mEz{XeD3dfNmI2<~d>C71Z$Tum=Zd3}Sx!YX{~uD~F3aKkGPr)d>yopba$Q#UcIVd2(X_)X6f4V+tIZ=lLiG889-& zx(o`CBL(yEoddAS#P6xbIp-sYn8_baJZMaR6M|paTl~`6FYq@c1hwY)XAi+1xcUi5 zPu~Ffxf0aH32>LO2S9`b|MKsGT_2YgE&w*6>Fo*X$cP71vfxGViTQ*e{(X`wJM)1< zI47^})ua|%R`2SStd)ULoXXUu4-n5T_`fjB?Bn7FE?rk@RO;0h!1|wZm&K!&n6~~l z$(cHPG%f9TbYXYHP&1EVbzH*^dg*gzqD{vC_W^)RpOYC>@06B3{;;~Ul=Lo*MPSq+ zVxfDqUsqV&nNM8{tCg1-kJPEHHf46_PT(<|f7qMMBS+2~jJ3@aCQtROsS3QlNt;!) z6Zs{>HzWeejDYoEcg3GP?zj6cThNqk_cG&?2>dAgL&zKB!WJr@%8Q>>v{!JI z*|#vL=`crLRmSv|`9I}`~PwP!gkyllx1`W$*unsVzH(4V?QF=wal~bFrSQyB_R=oZJS=r%0_YdO-ecG7h zTU;p!ijElhvxu5>m^ZIGqtDGKHp8 zd5fJ9ASk>o6NMrf;p?;$eRt=xyKo`jWRrMWx}#+MstF5C=V>fq&#Z;tDqcYL*9?)%s6G=Bia5%~dz5a;9ZI4$natqI&vO}QVV zw3K_|XX|n5pOR$lkvjaj#(AJ8S#e0~^aYACF=*q{^QM%C;`b@kg)%--`)R~sFS3ZD zs&b2{DMJzP6X6QZ4LerA`aVS>UG`J|V2{)AHjdYRu`E7J`d-D{46wNT)ZdC$dqJBn zEfW1y+ATc^oT$Z*M~)sc$E-2kTXTvqk((iOk_O)p&u=_-;kpo0dEin&09qhx%2^Z^3NlW3uss~+c z!5P;Og519(OGu)Yspg<_nEWe0om!j_s)YPh_k}MP<*Df49sTc|Q68tOo!;#ERZpp{ zJDdeCiN(@-vmXLFQSvVw&lEbLzC6bZq1os8R=1`Uf4iNc^=iK}Xa{ColT^x)-5cHJ zw81t_?T@Jiw`YT)+&$XC4*Ht zgN(1s?oO+%jMO`9?2>5HuE)NHc{K8Ga`p~kKSf&sN||-!8ML2DA{={TR?)8G!J_n6 z`}0t-LG_Ec^?ti#xK!nL6nz%16I#O>)2^}eBgsL4#izG3fcy6&&?I8hyn&^7;Y%pf zQ;BN*SVJv21za4sUvz^2s{Yxc$S%flhfk_lfc`^FyW`IH1*TI(84qP3{b9$>vHxID zOjW>}Hs6WN+u}{~W^F6!!w%GGrET0}0rMu91PXXDIU)qVV1woVRq@IVS_kRneF$9P6}IPy#A- zXs{KdfCM^Ssh16fig2EDOTM4b;*>t;)7{Yy?HO2!!RqIhs>*&lD@GZ3nVeDamyJWy zTK0CV4cW#s!nu4hO_2t+Ft>)zR=ZNNNaw4M60n z_umFL=B|{t><-ZU%rPWcE`f(aDLbIpJ~>13^9_!N&^ZlQUDm$j_f@+w>mNj9_)U)F zZ!EvB>v~{;#EzQgDFPBiHC8Vo(y+r$r~ z3Q@|>&lO-|O_7rB%GMn%B1Htc2`yh^rp(Hm0N>HRuL$nqfmB+TQ=xVA1K@lGb_gD^ z?ltI%+%eX^U0D5V26a8Ncsl%ec~Hf1aQG<0o-kJ^{rwBAeRR0-%Z8$n~Q(?oke^#1L5H6$? ziFubkc1Ca6#VnurmivTagLlMkQGS<(#T}{*umpf( zY{(ieFxc7)AweJu)hF~xwi|oeR@>#Uvq*pW!Ot@5@1vg~O8-m+I-F~V7I#ihRNaKd zT%KF07PI5SZt9$J)W>u$H%{48ga+ezBbEOtls}5@uo7)L89VCQ1a7p1+FaTP_@WgW z^||SHrM5(y+T&gY4c-wEAX=FQNEVTw(#~zoMH6_w#+=(48yAn${AIpo`#G$MJ9Zf8 zz4_{!8~!Puqq~&Mx0a$XZ7WsQ9p5R3(6GN&8ZvRo(`=5Flg&n0gU-9A8FLPSJU2$2 zYg;kC9)IkZd!T#Vy6l1pMN@sx%~Ivk7fp2vpzzh-LT~QF<&ihA>gZO)UEJU*c-55 zr;d+A!N8;c7;#zm%|3muwrm4R>M?@dJ&=hhvevfl82XBsn)y%GmWUs`!)6_ozj`SV z{jPdrhv6s6P*J3@D$rxA(#$W*w zcX=(?%+l*R)deIFgPOTG?~$T*5vEX*u~M}Q^#b7?aC<11)1R1+P7decL>)z2^WHlT z7#vRV2KU9sUzO>D!?zB)WP|Y}pCqd|eft(QW=4D3V$Mut^m3c$^}{SGVjA z6U1HC1b~MnHHpIxUu#l3@=%8PpnEVIWq33S;c>zzs9`1c_2lJrYK0!lb^dXU&Ttd1 zs{CVNp;UB3+y z4~ljY1WX*tH>!33q2u4Jf!D}MLJE6uu|1{bqEM~B2;ZHmgjR+>P`!Y5J=k~?SRp>s zxJX?M8i_kTF5b@gnq7<`3f*de7U@rHp+C{YV0rG`^kGv>(t{yL-@WhV0`jFQ&;e2_ z0K@cV;IiFez!x5OVg08g8w{XUtE*mx25?0U4s-IbA6aX`;?m5asg_zs~(klV=w^-kVpNy;EYF1|t*PfWHVqY%YNZ_I3s5-IR*RcH1U1%Xp_3Bdl zp1>4~Yx*2_XM$povv7*b$91C&wQieZ={p&79*V!w)7$4U0nd9M;IyByddtn;_X^8k z$UuN9oc+sDc^p)UwtN`hDneCnz6OGj+DRKYU{1I0f%w(r_poiLr2=h5ptI?$m7Wr^ zklIh4B{>=e`fhH#bG7oo&yW%U?Ch3@WvqBw*R)v@zs~LGf3*=PWs-ZlHBam*C$K}$ z3{~3p5en(dC_&j9v>9rj0gecip(D?zJ03zcHK6FX6LF(T4H>0MfwNe&@xDOxQCou{ z8}eJDCK3GCNWjQgN)I-J#02+i*1i|jSBQ(e?LdPI7;vtgGl~=`6$D%?ZW~&~aVc_4 zTi^2!p;!;&6dSH}k)i-z=@Qg343pup9A9=AGUk3gczyaRV=k|kJPr`7%l$?Dt~MfO zw@+%KO54d%|6!;zICH(Vwwv{N;B`TXW%kDuyLgzDDjC>c@tJ|-7mE>fm{9<>E#P9r{hb|_Y(5WbclW<{5G zm>TTmd$Dr(L!Hy%n6YhJazC^d*2TT-92U~(c6{~j3i@>bTZvzcY+{W|rB-PBF*m#G zTec5j7BfHb?Sy_cOkJD7s_kps!`V5WPY%(9Do*wZM;9T#K@om4DlKOCgvzL42i2Ak zP(e(hiCMvylE&3VFq=9?1G6-wc}lSTJ|RQuGqb?HOyMJ9e5HU_dCEo764NB`6VRU6 zYdTY_D;~TRUVLkBANC07Mfdh7SgFnt6~AZ3ds_jryq11yu(~d3s{tCx!Kx#~f^&S! zVr--VPd3V4v0Zq4)2>OIf6PgIMxpjg^e~704?Uv|(I)+k{_Qsz*Yw0*meEQF8tk!r zSv&XF?PipXl_1yxuD6R1kIG5U(FrC&l1bOdOlcVK$sWR@e4ugLFDVrGx`#=FL9=b*(sd*ITRBPiLMWp+#|hcKbJ1-COJ-D4#Rk+oCgc zFob*UQ~bq3=4|>kr`MwBOrrW1L12q#1oSB+sRX5~CDOiQ%urKDjEPKMOr|BN3k%?o z_>=+V@6&M*pvZ(}*i$jF>a2|9N%b0TyJ5RUwDUr5vuT;nO*q9|ee-)2HE%u`WAOQv zz$1~^%LIwHV~2fT-r)#-JIO!HQBJCqh1#7ZdgxT}EJN*A(z3x7RN_qs+ln0lXrWkV z)}%{zLOVPMF63Y{1{pAdExW4hUxYnaIg^pTpD5+gV$jxU!zA`8p!I2D6k)*KQls%qKviQp$)IW&MYr+1?98s zr+u^qJPLh&>n~qA;y%f;o$pjmxKNt69{P$v&7|-H&!@x7Z!ThSj7Qsctd7s_dQ=LF zv_HhTf;>HYiUPz>n)AC-zza8yEX#vwYy2w54Lz&!coT&80=tAm!r$EU^|WTt@!rt! zp(-2bE}vp{jc=h)ioHGr6D{hLHm%}Y9uuJ*)?}#gF9of4f!S4gHZ@@%h!hE4g}ixP zpf76j^*(vNPSFkz0pi~^lGBQD6-ur8`e8x~W;&S*e%3g77fT_JBi&_!QOjhL-}Al2 z`=16^5EF0WVOk%y`ym30i$M(hEgZ$E59};=kEroV&nzJK)~Vg_C%%-_I51*#EwOx^ zhQ}^=@^j^Wzlidg{df*de`8andEcyz$XE;Q zC-$3UJ#rVOh<~Vn^$o>GA0nczS5h*511+CELY83`t9ryXQ<)y=n9J1<0hBr zM5sfS_U{$}VFw@JY)eS^0kIR_ecgiKTLcP`DfUGBpaZk|F?uf>J0#7XeaR;+07lRd zK|23&)}Dmh(tI0Wu<8{#$6Me;)3e=R0&#!J;daYWEuVk$W?THl>D?q}VOv#al|c1$ zguYF@BaG5pT9X62v2x?%iRvm_B#Pv@hkBI-JAqc01Q?j!vYJF(U-Dc*^IuI-rXo_D zLO8Jity#oCf3;WfX^Vdo(mfMhHpFBCmJPqgUZpT^=)Cad)T=N%M;rb*Yc^d7S_1o5 z&Ju&9hj zzX(Cu$0b>|KS@W-9TIc#x(hcvu}N^W+PkH;`XQ^T?kqq3r}!lwsk8YdbjT!74a6qo z9Nho1T$|LlQQ+XG^J)2LPlRXs#iUuHcJPLs)&am#Eo??}a_19s z^l&XA;biqM7BB`~lk3rqPtCv~l5A)F%KzBu-D9(iFQ7A)M4G_6rQ?yEOOP?_YUqO$ z@^g~=hTRo<{xEq%bmj%nkbH>1ySlI5`QgVwAGnW>cJzMh%YA@uap4mmt_|pC4ZTL% zzDxV9TswcI!U4u)eu%LchFl}3`#w|OO}m&3ZY{ev+{%#>2nG}+-kU{>r_HigfXN|O zt3s@J?|*J;q@Br(Co_B{^84haK3KDJE0T z*QcwTVi|+)K7f~==O?AKtoLDg(osN#HSu=<=gu^4F%*ZI={3r*zoL}|UWqkb%fbxB zn=qttcGr9Xq(^HzR8fFY@^Ebsz3nzS|()koQLKMx;s zR4(q~JG6~ktti5$eyaQV*Kzv?_)tuT zW%fyV|1N}U^>5(v)|>h9{spAWAyf><^t@qB0DEv#dIYqot28{`Hf{pP_CP366bg8d zT?eLnI_u^!XWeY|TNJ>p%f;;^JHqN3yWLrQt4^B9+-lu5?9A*Mz@?nFy&0ljHEK>i zKXx(s`NCKnm_|DNz;kts_@b7P%8pZL_I@?ibD}&z`%a^I$hdB$ zAiwpz*Not{QIa}`#vJUY zUFa=t_osYbR9QkQ6e0FLyLbO>g$P}qEk)i0Lt=hxb<|f-Noed$_%)Jkulim9!k}Mo znG=6eoz(Uu&Q8bX1E&AW0`flioFIqLLi|TxZRztBs7Tt~wBgz-s=i%y*bL&1dmI8K z2&kN7$(Xu(G(m4h;jaz-)gYTYwX$%&E+|jV>9jnoF4v_6yqs1P!2_$zfKDiF)prgF zJo|q&g@!j}-Uy3Vp{)3^&m<#IyOZeTAA?<2P;d9zInl|I?91}qX|&@IxWvug zo-*wj_R)KgZKGG%gAX|5Z;-!VzuA<>X_v3BieJruqUpO{2J{p6%5;J~;`OLl@hL`9 zETq6O^*#)AXqozh#6Vts0Fu|IP;=RLD`O4ECHn^qYp|1c9u+d=sF-qa7@449#G54Pbx> z-4jhGJ3}DRl!4sw@*Z{&%eW@p+3~K8OPRcR=l7pKma*#U+T}s9hTc&Rb=a?T-~YeLfA$(v79mR8%N$N zP?&11iS-}NYv0X7r*kM@^>^3qb6_VusT{9C7^fN+2e_UIJ`wGU%58RvE!~S$yo?5u zg?UCc$WIY}2C%Z%zE~B7w<###sLe=!-4ir{2gFyPXE^CUU|8#{oRF>%|0#e$GD8d~ z`XC+xW~=4}Os$g-w_5j!Pas-+9SOz?5qq^S-T>=)v|kZ?$%y-NX^MDsULx^uYpjE( zF}M^qc+9fNB}iBY$o%+irnEU+KgY+Os}{ zJw>MRxSw(Y4ZoI{IuKFMDb9eXO`~4*N19}ttXlK&DdnObTW!nijs+Pu%2NT)P5xxf zL=I6yGDwZZtEwB^xSwWx{Q=+PZX8hfiD5q)D54)v_^W%<)A{{R`Lh$^-k4F$qd7!x zDZ^u3H@4uK>B7ky8-r;w)p-q{@KUos;>Uz8xR*hrX?`X!wF03F!BbmqWg&#-D z-#U!NFZeUEH-emfzD@=>3m|Cbl9beu2+`Xyx+%fs8r%iA<%8+n`QFO8pnWiP@grt@sUMj17)Qc|_RCf2#S*3cONdC$rm?(qh8&Xe9Z9Z?l-xQh6SUnFJ9?-m%&oGZ`zNji)B)Q>7yBRdMf_$t-bpC1 zchB*m*Z!fhz0w`e#Le!JQ?$5+u{t411y{zr7s^ju1%@j0ek$3gRupx3iC(+ZLdmWZ`k-% z{oZi-6&;ran>a7lyz>V24<3w@bui-2^^-qtPld6bYdMvfM?>%jH14y%wSGwFM>S;M zLaFpwD2*Dw8b%(AJ?NmwBV&B%-(+0h*Vz~EaGs94mHi^tZCa;jU79_Em*XAgd81>X zWSmO^i=^H559)|Xqb?-($2CUzg>t#74Nv*&DIxTCc`ouJ?4p`px@ZgH-n#&2pIrK0 zqTH-Bn<^JapKp0wuWyk-0NiMpQcI85kbU~3l#7h;_+zZ}Ku^Q!OE_S!k4}mTM8-=i zy4kfe2^ou-NBA*}s$Hw5l+C`I`6Vq(_&xbm1?i85vLY_yFoz3ReB*WHQeVJILcQTo zl)CT18ApMFJil^2LayVNTY5ygr26Q6rsn~riT7q44SUbCpxA;Hie#cg$!3Ie4X$N!>31+wT18O{a2B4$U^T4{GsC?)E{JjQXcg-EY+QMdDgINS5U@HqdE zknn#`38cN&U#i#Z+fSK3{9=c9f2%hECT&T?T+j<%SOu8Y-EW8zwpFb4%X*$^jZfs2Q8y&D59cn58@Qa5%|{s1;LE$2x+A~I^M;Tp>d9WVO`+h;yUO_(gW+## zW1H=(C#X2}Xrl0c9|x2|)yBFW6h1#-ish9OYQ(>a;U(++k+s=hW0l+aD>{d?@OiuF zq9&7G)$inC<OTuP}P0L75xlR$)D7t!#88qRoi~iY|4CAQ1i^?rl**9 zlmFo!xP+8@-bnH^uU6n(@n>h3QvRW!f!hFDDYP^zEr#!A-p1*pmPUdhUyR$NOiPt| zdIhbx5`{3|o6#2K+~F6En&WeVfnoGxU6hA{g3lToepqrpHGZV{rOa<6%kV}@+Yb|= zIY}R;JdaQ;#~U zkBfORj0Mt{{_fnID>APb9+Cc>r|dP>b&=6ZJc?{^y2uo*hFsiAX^?oTnYC;KmIde;Bpd&)v|a<2EZ6 zv6+s8R0Bs9em>MJ8AY2a!H!gJajeXK0B7(fLly&$(*JdGW(Pj_oO-~aaT{!!po;jX z6`64{f4OeAe_o4$lW!w@nwcYm2ZaB z3waSRA#%lobMwPCKUR7g*woPb05QH$fY-St9;gFE?~`v30TF!BH?E*ZhVXfp(-D6s zXP@%F-qPt68IQ$p=p0(JWn9{di{QxTjD|;U+B}S){46L%zbS+N@S-th>l4e2uu5ad z7vO^|(FHBUIP@FvAxGQGX9ouKM2a2DjEF8Swo#ktFyr-)v3hQkMYNHV$K2-qMZgEEXHle2;PwK z?Se~pMK?B^UDmv&!#&NhO*65B^OA`1w1>AQ>YX;6%!;MB)Yi+w2^~nI5n!{2P5BQu zR3nBWD5co;1(n1l-#^hm&)<~`q*oDtT(5aYncX;Idh!|l=I^%qT~GoM;z$~CCxxG* zpnQ2#%n^fImTfBJn`f|_d8P-yy5oS= zqs~_s?IZq-K}IR6>HE+5zUH~UD@z_;opj@M%;N&QNov77;~15%4```=X3>jqjO4ym z^Yp!y(d_ibvo%t2i$92ro0k--w*J|eBfmQJpMu8YXBs#*zFzJPFnwM3$S^JC!_v3a z)70JE4ePyb?_J98VC)$2u`Da?p-+nhxXWvvb$jfSJ1wv}|L%TiqUO2hHyzNVB>riQ zj&Cf%&7i1kX?7>`mY7@*Y6`IBd0?RC3B=Q}C1lCmoJ1XT(W(1h~Suzw@yCsI6CXFrrtJ= zD9|^8%!jmB-|z?Fna8~=lA}# zYuB}%>+GEK?78Fn{oM2G8_suvw&uTos3=?0w)%d>Kn?WAtiWBy{;*ZeSiGQBYlu92 zE#TCD=ah{B@AZ#i)#@g2rh28CQn+L^Yx@*GZgmCVX^FZxil4O+)CX@;*pvFSt8ofW zZ4~%b+OBC|Zm!H>Qu?8r856~WINQg?OFXpW6zds5^S-e>rji=<=Sg1~66iYxMXnfE z6`cit7cALvte^dZmh&q@SI$Qp-XGK8F?$MXX@j647$>oymERQ2W%vd3;sztQM{NZo z(3N2|4}Fg;jA1LY&eAF4vkE^G^?F(}`+ZX`;Jo{(7m?NKk7Hq@i-wb*I^SI`hKYl+}C&Tca6THU(o}Lyweu^RsKuRkDz*SKNpTk{~iBta+_K(qF?5EiZ>_H#TjAw zg4jN#tC%dF-O+yshSL`tmjVT!afYb=q=J|M5xLK zL<&$}jXhugEtpt;>j$SL74%Ivc(%7zn$+d9-D>I8nwT} z&3^AmYBMZ;lhiV&Q)fP9z^ZmfIl;C4*Fh!tsByL1C^!ODC1i+ku6E_`p5U^gb-L== z9{%?ysII9Kd#q~IKFrpxKLu8N+E(CqaZ@+LbhtEb$PF{A2EDo9r#++?0e4?r9sOfa zONA1?ZCthIE+dX_?$~;#eJcg`a`D=ZA}C^sZYZ<=SXuxY^W-qPiSh{Y@(o5SV&I}V z8~C7Oda`IeQ}CmRlvnld%ZfNo#0PiYRnXG9j`mhnBo~6E(gANLbwH!jl-&l?_TWSo z2b_p&I4g9Dkg@PKaH-K!zey7kTx*J_KjCcu;SLf%G|IT&Q8I9elsGdeqnv^?ogOh8 zR~_(Q&j95)YeUVB<383+_zkN>mo?e0RmY|cBvpC6Z zyq}9}J#8L)&0KFkZAC0cc6e)e{zLKG&5=KC53p1^dD?z%vS)W$Ph+vQxg^5gyFH-U z@Z>ZP18}|&9_)u~aDTaEphj1foz3hiw`XZM2VOcq5B+>}rE8;wD(2=86X6@W@4qXAu)MB!+CMI>z?;e~Uw#CAT}on|!(hz- zs(wp64=v(J3WjMa&H`aU=Jm*qtYE5}71q(*3p(EIg16r1eNF8GJt%CTB}nA;Yh)me zNF}-q*ShR5r;`M8p5|>%Ji*ebN%~RYiJRHa5tK{M%hs;BY8(%o)CQ=OgqycyU;zJe zs?dyy9ycoE!*1*&bdX(A$OK3cT+XP)wZ^RGZXrAV@d)Ax7HV(7DHEzTjqeT`B40S7 z(Ibg$(3Ln@Iy7Z4$<`0E$b%qnEK5o={WfG~FcdVA+DtY?Gd8{&M}H}2_Sa(oYbJP& zJS943r0*oonlL3R9+2)?<<*%n(cA#$^<=*dU^fX8D`wgKt= zM@oCB#(U1iZn&>mnK2KaYS?Z4d<1xgDAnUKENGu=g!Wt#0Hg8U`@8av^97i ziP=-e#eUNlHDyJOe>whSSP02yYCWYZF{3I?9qRwd!aKwAqa0ycqRj^CZLM*Hv3*|9 z?EPZ84qaKK@qU2EV8~O5QEj^JJ+h;RY=D_@CJkZwg5pRVPk7lxt0$H`a%e%sbl-XUHMqu`XuY8&K8exzvunng~v$*REzIDP=asv`7=_Z$-ZD66Y*lw^rD5~egz3kEz9 zs;=wr4 zGRwLI6|-s-jG`-#3)RNi+TG+3xR)Lj$N0olQ8wPKb>P zFs}uh zVS`AOWowGtQCsuW2>|gReKCfJ{eualb7~g@Mb?eRqyD;L2fppsukyavpu1^q=T?Lc z@(FjkvG(mWZy%Q8Q*z>Yj{7ZJr+k9_J}g95@fv2qch}TXr9{d-H*QXylZAKrdg?WY zgqz4fwky=b2;86RWow2H_wN#vf&xq^6xEjsBKbiKMRFtB3UgDUdHpB=oL30VAYDF% z2vPW;gEss_$%tZJ?N-j`ENM87{6RNim6BwgJHdr%$ip?_m@YI*f9Q39zFOwLz|N*U zp#=I%u#6?6$-Q1HB(T;y4VD%5pugkjv%4PPRHaoGoumxNs17AH*lzqtYBexW^SQ=& zLw|C7h+(9pk_e~Ekr4}e1iQtq$~|cg3DbG5;DrledW9*eeRdtReU;kFR^fe?%(0Km z8HsBLbZ3k1X4uAfS>D9($$wV<97?Ocv{_-=5Rw zy&C>AD&wotII0{eGT*(iE&BeOSI!E5XoC!GgBs!&2P>SNgc|v#yy(Qr&>?Cbip@F9 z1a7+R4nEs*IJjm1v)X#7ezXvl$_bW~GE12zXO^G9`btxh(uCakdqn|kn{{J+HsGkl zLppWrut2U)_WXRvTVFKm|0Z>NOps|v z?%8(TtlRROHS*K5JQJFfOyC*!@sG9@;+U%ls4=^M-x@T+gRR?nnyg1kLDj1=z357x ztBhG!K%P!h2Js?mR)~w;|1s_;PJ?8^{cj~g=a#U522$#4J8`1Nr)gWIf#a7d@S#IN z$pB6N##Lr1#rTMS)4(EN3ilW+q&ygJG}}W_uRQB@-2&>0qVQgdD@R2d6qOE5 zK|=>Uswa<4@s$X2PeUigAg`LEQtt8hkqyN8{vul6`V6-QC?Z=W`unFm$pvoAkmfWt$5o)PnSep5B)!P+wpVorWqLpca{#$T249*h9^Vdy@#dI{7Gl zPz)rdg{jmF{$3e(fc*j6+YK(t(=KNn0KdSnk5gBeH&*g5FsF7EG){v5WqmKp&6dD4si-{$LEb`Rh9;q` zV+yl@Kx;m5`F>X0yc^W{yh5!A@AKfoPPS~#650`S0Y|wBB&GF4n6Q=MPNuy#4n#Ga zwdX659cy4TC65g3IU`xJ@xuArBg4QFyFaKv<;FfB>4^cV*Us+ZE|rPa;FDJQ_u($) z8S~jh>`yHU;3@7X(w8;|dZ0N3XsPzLOqT-0ehpBHS`%-3FZ1jJ~x5>TVTij4JVt z)$JQARrm#=u_8{469$k_lu5)d9M~E%C}oZy(3ivH>71-C*iRv@>omxPo{e(gq& zgSVxFLTDWJJ%8c_Yf8lh+&p0>lk+c-5sS;;xzpD@8S@(=SD1bJORFux=a^NWa9yv( zl9RhV{0$8GN!^x&x^hEihGeM_CI>pvtj7a%#V}^i4XA?Fx8e>g;b12hDT}c#>?tar}M@V4l4wFGi5RsD@eY z@G$V7qGDk(v7p1{o58m@3+=ljqiL}9z@O-CY6{tzpdE0C``NdD@I1K-?7zxW)Z7{5 zlXHbPfc4N(1O=y#A&KQMAVu2T5TL(C!>sY|)jQsO_;S5a9-E$6OQ-a7M=xYs?Y@^- z2Po6O7zw$L$q}R;rY|EsJ9w2yna+vH>EX8*d6|Ng0_V8NUE_U;k*DX>knbQ-o*mbb zEW>(Li(#LOr^uoaV7!pm74-U`p&%e!0;uX15R3ccvhQYpcC7q&`M|3^plBqVwqSxL z&59_bF4;jEQ$_qb2hJWigJJBFFFP1PEKu-J=;f_@fX%=cpfCBM4IoQB-wV3WzBUly zs)3+>n+S({j3Na|H$Q&AjbkK54^6__6bv)U5-jSoy7VvXfQ|i_*RTm0?(J{<9;E4& z1+U8?Iu~}q^G9(YiJ?Lv=v_W{A_X<*sq89wUYz#^7A4JH3T(yED=Tt^C0KO-iorcN zhp12h>)!3*x1;ij*v(OzSupRPsf5hv)^L6GC`IDqF&lAdG*lrODAUDS(_9kB3Bpk`?-+K65io1d zd;z+?*bRlM7g#ZEQpi9%Gc0LWt>j#TL*t(bAJ(N|MCD5s;@6?sn%|RCRnedo28Nx& zsduC&TZ_ueC8aKy#fo1}ud4u5iDl8vdm$HgdaXMrzI#d6c%POUk}Tf?2YL4xOZhdt z2)n*4S0nay!e6f7mJ`E16(9pcJR*FMfl4U^-9wk@@OHoD1*+Oc!1Y7voaW4F9u2Rb zZl3Ds?h~BbprUJg@Dd2>6=2qw`?>-*J%-gYcOniNeyvR=5#dxz^0!|j$bCl+(>Pn` zWQ+HjiVuEnTe2caWAP(unE){jv|FB^zv!IPrIJO%v^;TL^FVw0xhj}uo-_=uH zSM|gw)0u!yXdv0t8`ut?3w;`Cv-i*3UOGq@F0$Mjc`)4rh<@Ff4CW<^JQ8DEsVaG%CJ0tcW1b^KJj7 zI(~SV9KU{lu@N2L+#T*R!l!ZjV&=_h%)1iwwp?HS#NuqvND3l5-0j)$)=*orao}nvbUC_;G3#Y9 zVO{z{;(H;q*QoY1F>UEfy_=zFp(5$+XO?q1)~R@(hM$3&PhSvR7^2VqacB=FuZ|+BY$9=v1$O0dCr@e)nU)4@SK|CPAL^>$9*ft0QS{oV^GPV#HxA@nFJP!eKBpZSA-^mWGk1i%_^59-wc1)ZH}eC z!{y9kjztcRgd^|^T~0frE%*g(B}WlbLCEJ$cF;5i{T$-HJ`?2AGiOTq6ZS$fFCIpp z-+L$e(4}b9a|qG`{9+C)U*(^MNdRHn#l}t%z;6Ch@0u`O`(*BPkLdO_ayI|QUT_5S zgNiBEA*PE5Pi^?BS+tHxq%vgaKXp9@HC5WTF@;N&o)Pqp=9}V4rOPJzD?SH(6P4II zR%=@ZJ#U?*t@3S~@ux!0R@#ZdU+Tr$}0Fy9r-oan8k*CJ6kV6t@N zx~@~-RfRv+M{&Z(loItJW3d(KpYA-UGQQQv5a7iN9iCDOOe)QcyPb=}>SAscA?3o& zSTcp6KGK)OKQI`eMv4{0@hV)Lr%GK-DqjDXdvgkvXB*1vwuHB59qqG1o@8WNVmeH3 zei)tdKD(ZrB8};|a`_zRl3B_}g-WG-I`>tZf5KBZ@qM9~XKf7rKi*f&Aig?kXB2k#U-~R7eMZTu z?{4alOBsy2cfFg*ttRD{|6qpppCLZUFdyYKKOJ6Z|7hjVC;Jz?&{l;qrsrAB2k0rj z##ke0>e=MW!Z@Rb*JN^COH3b)eK>+8H45|?Ld*2dUJXf{Ek5XHchg73KO3d}``6%w zGc;0l#IP!FX({+meL3!Q%Ng7qyr{%Y1pn6i*1i4${>CpoMZX7)yKk^w4AVm4+tf)> zUj}bf|BTm5*SUkaE_oK1-#rW=QO@ZM9XTYM2y7I~7{G*||LHfL+!dB#_n1uqXD&@5X1Ao`AXye#FBneyS<58hgnY< znm@W+B_;_?yS;M0>4wZONz zG7^OGs-0_Ek9s&ny{bGrS+pk5Yto)SGzd7F=9O8*)Rk6juwq%8`GH33w&1`ezKPU7 z0l43T6lsasgn9m#-;>2fw>2Ye{c!ijXxgOA#9#uOpXJ3qZsc*KV`Us3?26#YvAB#4zt@+?-;WK3>rZ>gaIUiXRc)xU0%q| z%#AggSE#TnzY2elRxvYIT5LS_k<8M?dPa38AK%2^I*R=CHmf;HY|7BRd`$rF8bNVP zTk@xBMC{I(V)%pQ5BK})QxCHu?2&T9ksonX-B{FnOhnx)6s-ZIDt)B%D(gjnMXL2= zlRy`Vd#@r*f@B;wz8R@x++ge@tlC|HAG;ab8LP(-#_OA3257Y{vHQ8JbV*$)x;4yn z(Mk#A=LF*fZW;MB^m|C3xmNzYkLh3`N5onjG&Z`P(Jl;O^o<)t9vwiJ&8kjcX)3v7 zC|rySMmkD*8RmwV!8Z8Trqb}f?Lf%OuAAmAD*^ftwo;kcOUb0L58R%t7p&%%%L0)f zt9CQA7yccXh3G@}js4xt=Otop=M-0@o8BDnTVuKKar^fiiQml?_}T%k?p#;K_g?px z3^94vXeX^uo9J?fA~tG2+}?!w;A=sBXinn&s<5!9^p0L<9R~`hJKwHp?I*(i5Wt_h z^3n}W@QL`{_}F|`mIF#qutq4;Yq8N~1pgrH7)5+R7?!@hl1K=mA^96Mc%8Cca9~^v z37S)tdJN`V>*&~u8FrJ6+~z=36W=ui&U1<`=HY9YB!m%$!0)OCesa?E!s=Xptp^_L zxBQ7`36GWubg_Q_-@!i=LjSF{!`uG%W$<6G_^IcWy~M634o?P}xl}A!4ojT*o54b* z`tE1tQWB?|z9I}qnuKV;58b`Z-E!TSe*}p;Zubssuc=6vwen`Pevda*emnX=9w`U3 zZ5$8Q7oUg~;W%YIagPzIOoR>xLLccBz-rvEJPI?IznhDrI@#TuE{9 z{zL!UciryZE-Mp^dH4RE*);vDq@?%XO>8LK7;W4bm3tz8YFD!uR~n&rPnV9v&U>Qf zSJ(gi`Em2vZr7E((l6!@2y!gw!8M0OzJRVkH{X&!Ynjq5?8&6u9y(i{CVH(W(N%p0 zD9^Xf8~AC$#0o#$`pu;~C4V-_el$NI3f)6F_++QxpzXj%ykmh!5;n1h<_;Oajij5A?Zs-nae+vjV>4K1vTT7(P~~O^^NsE zR+E8&E1P6x{Jc=&Qnn?(Fn&#{*Ah={MF>2kC>_5wlV8z``KlvWw%1lM)3coAdA6Ba z^kA+2l`&U}&QI4<1JqB#vAIeG8Zm^{zAe=&!*bzxD9=7s(R=kPUH^6ANCKfuc-eH< z68@oBWfdLb3bhj~csVCZM+OrXEI1MCuDiaIiue66sYj7__9;prk#GD1zCj?;=E_+n zM=dfmtYW5Ea>R9-aQt`bo;fU)`d5B;v!;lgs{Yw*FR$Cbbb2jKQ7?Pal?1|`KJ%Oo z1153fQ}OOkjXMPErLOJ5Av(Nm_v}xC#Xl6M9)k{hr16vYJ`VRq7&J)&758R|Wqb=be&YR}N6g^pJT$VP)GjLOD?0E7it|{D{i%66I zt?wz$S?td!=Z8NFspJnkXNQYiK*ryr27XQ+D9;ii$?n;i=~Uth-8RtAP6_m9({_s* zCLiZANQt<0{Zo*S156V6tinqq&UAFy%y)OBIeP(UpnaKCGN%Nbsm!U~R=6!*J?fyqXsM2$jJQLcT4z}0T!QoBo?^Gf)VU-o>318o4;M!UOY;F=>b_>R6eapIX zkm-8Be#S|KOq%fEmt&oGjZ)Z&)iLw&vZSJU0N!TE;~ z12=RS^|69sC7*arViWXCtpUHJUkd-uJzn`R*=7~P303D9Oa;$qP8}2RGLpj_h9Rg^ zvtO9G;d+A)TGQTYKatO`%q#4EPWkhIx*WgwFNqqR5kzC$;M%D^C9CGu*5xp-Dv#`m zf$;>rb@dWAZU~?F3FdR6+v?(a9*LqdwYVx~^IiHE(^WI**c$4E?dZa#R1vA!!li-2 z%FG@>2;!8;*Fu&e=T^5NX z*j?nSL_WspI8QYad$S3X==>!nqfA1($t#venxGl* zj>MyegyQ=eqQJtFsIxwiwu=< zr+VT1+^E6z&ijxz#$)h7!DK_!3~28$Njl~)&tIa+zB}2(dI1832pZK2MrSXHVL1jR zkD+2O`{MWr7iRzNxJ@N)S(!WR$Qcx^&}B7uK1bWVF|x+wZ9CYDRORyK~jy0Z6()53Zn$9|KfVg(fg;dO_7{Gx;BSLD}A6YzpQIjJWoQ! zFyhM-y(7h;JNKA;)K=L_U^G5nRC!1_TknxU1b!^xUa@m?SYd@7RiyQVnF9#VPfH4) zQS$OjjEyYzV+4YpUq0>_Mur;EkN2#Y_addsIO|5L;Gn$y|Mf`B+VLm^kRci z*nRS?bJwsBxzw4Lq(S~Fe5h8b?o}Cfj-LK%gdqMif$~K`v*w4RZQ9bi>1|kFF22sk zkL`Sc-4++dt#2X8qIA&AIeIog8O~&J_DW1PzRa`jsug1e?q+0ru{w*^uKkIyU6gQT z-=e4&c$-_e*^*!$v?Ll+u56d@!gPEKqCNS=T4oLwFCsl~(Gkd*1;{nLtFNXWoJZI% zbiokQq1?(G`LPTe5VZv^)B!u;8hlAFWFz{BNgwiIn!^{PDF$XCJAQq8=`>n6s)bd< z)F?15($w^_k7{3q5BkMw3uh4amim=+8P=7!OZNve2}zy}X$D9y?{)Q(yhebWV4*8| z4iRFgxKVTguc(n07KdtUan0Ir;QIzhGM_dYdwlSv|8`D0Ye2MLKZGvf9)YL^k;Z=( z_?`4tn_|8B@U{VJ+cQBQV$NdXrn%OiUHzPnocmhCt6M#n{v5!;KMOK4C!zfAg$y0%E}o0Vz%9d)vGbC)@_f_m-H$9%W*ft3;NvSuzpQr(0x z^9Xhn@|`Zy#LnQ)rckZN6$|*+?h_GaQHp=(5|`cf`p2-m%Ij(4_?V6T&u!Dt{-r37$1MBwH*;95s_|^QS$iVVbf% z_H7k*?-DyHw}FTorkljg@xPvQTH>{=pI6|s0oFdm0Rw>x`0FxH^`8X#eBR3d&(EX5 zgti32S<#BZ&qv-+F_l@)lPUsd89(R;4yB2yUzOuHW2pvW4^<^ZR;8}_|)U5USiS@SuX1MgkC?D#=5v4M%5um;&2+O;+!i+VRKV(m8jg49XU(`MtZjr za}Vzmcw(FH;(8Ae=&HycS(|rkvZ${7;OYhl28wVi!oL|)H@VxT9 z{0d(7AsK~BDwNt6V`i}5O-MOsw+f-#M5S`?d%EyC{I1M&#sE4takX9mZ~G5rnodhn zK+4TrGxxuZ$*T;%HR{ss=3Yp07TL^4SFtPoNzjG+RHsQ7c5B+0+)i7C=Lx-zqA%jL zd0Pe#-OURYKP#|Op}CwC^>>-;G9V4TDhs;v2II_N=?nyMysZHw36_Gx4w+!Hn-}yg z@aIpVY8!%72>|A>DLGXTk#9UUE||aM1SRiLxu+|q^u>?#xk@7@@g~rI7fr9n0BuOQ z0j*z7q;;4C5vufZK1v9xA^OXGZ$D-d;jp1Qg|N>_kDxP8nzhSBWrhS}=^D!u&2d>E z(EvMEOqLAG1?iEK2TcIsYl5@I(PI;t!lh~Ic)o?YQ4hMVI)=Pb(r=_(E^shdnKl5C zqrL6;?2pUmFLJ41McbI&4d)Js`V}MvAEC*UVP8%@?y*kv+yfko70IuGL(AUt{)HP;rWeXkab8rn+2R^Q{oZphnw%mv-# zGQxy+ukecx58KLABKMxiTY4o$Nv2WUXL(5IbZ9un9W->J#qvORA;f9x6=-yu7A#nr zp1vxgNx0N43YNQ%pV&NXz1HOzj8&c`p+1fAa7?25Qf1S@tLtEQjiHOhtn5IpR|bKO zjkRQ(A%JN1nyp&xtc!Co?xk!QoHQSMe`?d=9c7X7I^bx}r=x>j1f$*xhFlPm zSO5Jg*dOH>A^2pTk=Ninr9MVW&7;XFI>P{Vs|y?)3tbz0I5sP$(Z$ea$6#6y7l1iIqyr<^9Yot10&t@|jXnLB#z-&t(j`3Z72)811 zO0hssog^Fg^&v16>qTdH&mcVuACyy=bh=LSpChbe`MS6{F%ZM7!ovte_zxYnA@W>g z(Ui%(fS16s^eVhWiIz6f!~@rBI&*b?61^KN@ez?%RC&b}r{i5aH%I`T{B!0dj+ngP zZA9!Eq!8@49~4M~j{TCaNXXN(Mv=uW$qk6-C~YGmk3k;?5Em&N{;+GoInO>IEDst| zxuE~x`oDm=vNEvBK}`yy9f}eKAtZ4ml!)B)H(=y!`F z@=znXNwBj@L2&o1w#6L0iJ}0$zo=AFWIn4X!^VAuP+$H*u()yQar!>omt6MEC@{x+ zRqJVOYTp%>i_Z0T)|V%hOdKXHucc9E|3vS~1Q=NptdET37{E?t@}>TT(;1d+g=)O1 z2*nwR^#q%BGvcS1_}S@n?`HL>p4`3bUiS&siuSLX_F=N!Q{>#xQ=?YgTms+1J9Rs~ z9!ho8Zw>D=Zt}dAe3dR&U?{0H(I%0utJBe14zI+F#`@{Efey};N)=lG(q+r)jAjS#>d2J#%QsIj>TZv z-Fx%-B%e!X%){b$&Zw-(6!k%AAXj;EOmtfxRFM^&c>=G^q?3m zW)Gg{rQN`dEYKoevUnDKf28;EAR+BnKLJ#ae<6!Xy8$KX$XhjNPNv7^zZ5!aJLSK6 z+lG)e1l(Nge_dJ|)-+E`dS=+3x)1LfwwJ5QEbK037a%A$02p^BOD`#?Y`oc2cx;xW z#}q-t<663UVGe+KUG-HCyXR;Iqt6@?A{WlI2JrP}r=xpl9!$20?lJ&50Lcy?4c~i$usIo20luDy3Hq=D3%twMI7txevg#vlN+Cg@#5;3 z@*X4Q5`2h{tDB|us)?0^pRB2iPR}R|nLq-x>F7ytOnDGVMXLtCc;U^U4&H=Z-YYx5 zGd8Pse-8e{*>dTG?p7_S20pV3DM<1$KzT#+hF~FHXGPkG9P+~MIxnr>LxHptY&VBi z!2V{B^98-=b%f&}V;|7Alx2)$0oy{yR|?NE%`wY8fRrw7qxlQ<6k+F_5)DzZ^BKERc$z0x6yF8r*L(z_KrwTeoJq_y?2su#Z}MQBmU8_rVIe$|o?5N* zE1H~&%KAA@=lM<7v(>^=JIxq#6~dbXIjW2QKBd|ATq6eTir5PcshmQ()}d4eC@``^ zuv?uFX+@O4JW~D2oMv1-wD^Mm-GK`m^c4m7na$O?0VXZO8Q4gZ1Ce||4upe4@@VEC z<2DWh!7x_y!!{lF+4?}l7qtLAIbcKPx3SPXILlM!Hh9p@;pL}i=kIw;xu9W1@4!5< z^^R#uxpHLKpbn?cRk}#;OP%`7uKK9OrR&Yi;D7Xx58qRqJ1R$;=OPCgS|7vv>`d?} zR#mma40k-#)G7WRkYqdyQm=f6J7tzpl?H}-KawHk#KS-|2LBU>U9ob(Rzks-a~bKx=$dQKIf*G1KOtgJ;Z@i-vCt%VQ2nf!&S8|DMskr z(96%nL!?~etzp;>lwICfTaR?mTo_~Q;|6%T{XmN)L}3qMcLT1TLIR&;%8O|dicYGq z589SMzje#jyBr2afGC)sidgR9KQ1)REr2;KLeoj3B7p& z>6i1l)EQ)u5bFFUtc2ZG9{PMHp>7Ncd7DeM&p z=MZL4zz!Tdd4|!^j5^`=jp>EZr)@1N&Y39$9J&B)GdgrSib*iO%%((S=q;=vUDwvX zY~p?bu*P)}{^slEK||%d8fYXARG>DCrRicJDiN=d4v>^TdU`l}2^*oVZ7m>DKueVH zu6VL1i)kFReNPS7z>s)TOXjmK`~MiOMU82Hdqk=4fQh+n`QDlPZk0$1-c$+Hfsg^% zS$jSB63i$n44iVrNjxwrNHITnq{~pV!4Rp@f{vwRf4+4sIBa6p>emWCTi`s9nN<58 z87g;G(CT8Q{K}1;h-bLdIhSB1sbzM4Q%r|A5lcjdt_{9?cXsSwwa)&F`)Z{36YrFR z(Vwzek?Vr?8XN73!`-tsyJWK208i}qp-Ocz;8aBi4Yv6<;_3NPP6|K?rS^l1@S?l8 zVtS;1#&i!VO9^CG^K;w?8i-}T-^~&_tX))n(9jeYwY)@s zxv;x)^2Hr(Ox_*Wu`X|`Ns#osS$p!xn}(Wn&+-?*j0}y(^#~S___AgC(PM@dqij!b z?oH($v>OH8#RdSM$63b+!!5u;w5M|)E>?74yoY)m5_hzF1F-*cHwd%biQfG+{}Nq_ z4CU7rsIqd#9p5~JA?*xr>-A7^^gb#j@$%gQp5Sy?Fw0lzcsNK0#YG9AlB)X;X>>7- z=C<^km13Vkf_u>QI#TU(NwR*(2gA|HKJ-}d+6HctWXXCH3W~mbQ2qg3zfNapn!q$F z%ehw-jJp@g%s+us5-wS)pc$z83KN!Y@%nrDBcEGYSM5l zBrUn%NJo$+rb@?=8%U^`QLQs`phbLXl zc9AVxqJ|8okft<-lvgm-{9w@XvMut_5UieJzzh_xc}M`31ZIMD_r6IllaG>|Rdv zLm*m-L+kG>IEqg+&2Tyt#HJVC$`|K<^4!*P8uF)racA_q-R?{ zm8rIXV%g(^4}3qs+O%G{FLk`OJ158rIv$YNm*QXei|;VFTKM&A!@4DG>saXo^#f*m z_Ts(fUG^gw{iwze@Ln_Ihvo7~=HTf}G|GH8^#U#a_G$sQW^ zL1M4{YRd=9N+;MjiNS1uxM0io;Nsksi3mFrxibLiDR|rjLhX09qUj=`wji=2tf>cN zCn&An7Hnhzi1&F5l3)&>QQP2j@F!#92MzJy2C{rEB9KhP(o|7UR|A&MB$}fTq)i6$ zI2+*elQ>OO7kHmXG>T!ox5PI=*3;H)mW!Dj%M8HrPWOr_-=+>x$vK1Z6e8y+VVE}E z!~P96aR9UhaA93ah%J%TXQ80>g=DV~IjMRK>sY&?nd`syNU+K(NxhIJ|F?8<(gEVr z{psurRN7N|e?O8qRnJS~QsRvFFRtlxH{jX0j5OLUU;WGP{A1<88|JF=&`UG^ zb^1k%0+kt&NBNiS3tpFph`b8AH`coMM0-g5%Nm*HSdNOvBQGyE+NUCGe|{8^3(EWa z#83TgwKk|ZN!vtidQS{!wK#p~!N&1$!I z+tk)^n~R^Ze=la_t`E|6HqQkgNB(&v${>)+`IJ8N;g2`gypcLJ>Ur$VS+tF7ic)-S3R$*Fi=S*9 z%}XEggmw}G3ILloDMjbJ&5Sri(!tCI?%v?|$WWR3&M?Dg1=Ft(W53hK%$-f9v3#L7 zm*TZd%|~na^AMF1Lz1sazQ;_McIGz4kWvcCEg!wSNvN``uN^`Xx@O-3`|g_0*r$^{I0q z8D^{RB`3JQIo+Pv{TEe_3CyALDu_`pw*1DSrq4s)SS~p+_N>1scww(h*oDzGcZZL@ z^P7NGx!$i%@(;znhpn@3KR;CV?-@s_ha5BeH>Z4UKK6w-83a09Fb1U5dlxVzc94|h2U#(jul8&e|{F_dhy09NsM|(9(b>{uO zWUH=v+Be!43L_LO4?yQO{-88lP^*;H45?xH+Y z((}z7| z!7t~F8>SCRSKhsA6L9sG?Ap4LCX`gS_Q-ii#JH_b)#FnB5xYA5p8;tVJH3Lq(kjcZ z9glscDVUWcHviR>uSV1tkHbz<-p__Oe&fm8yt0@mn{xO8Wm`-ues?f6LbRe4s^fDU zZBtx@tVNXKTVL{boxoWYpGA>k?m3Q(zfZfx3TmBHJ*JXXBXr0^yR2`9t*75gU9Hjo zIp-?6{6lN{?^1s59si9(x8P!H1mjR|Xa0+@k4U}am+V%fB2uy+|n9$^n4=9@WOgcZvf){@H4gkAS5O;+n+gE+1Z)fB|C6% zg?~3(XX>-$?HGz5ooNQIc#Y1nhBAC4b1Wv?|I(k~g33N^5SsogGe# z-Pe;@yHgX%XPFqHu5ey+qxRD##crw`=L3mbPqk=n)u^W%=0*&>Z*`cCa1-|kAfQZI zoP)0}`9j{BYKfl=b@Jrx_yqnMX%`b>rGfiVKGw->`7!nCNI^2 zRK2%zL;F9u-JlZgJZNJPCBt+7;$uO{`n&QT;JCS1cB4;>MEtFx>Re~8UTDy(CB4Da zSVO^nX{Y+VHL7Eq2^OO3wcko1e^PYU^Ir(hzOmMfX^wT8t@8Hp`%ZI1wst1&k_DXQ zhQUBixq|8LC3ewo8nP`P?hKm859)Q_joz|NxAQKPaS=JNZcS|S zZsodK=94TVLZl7|hV)p_)R#lS;JMi<9nJSstzsoM?liv7VgjAxO@AHt;9S|49e%kU2z z?}PSveLtdpB*8AYcJk8e)ty`J!>y-FI^=x1Tus7>Xo3A?sV0Cqxi^6NVl;odVWV^= ze_Lfulkyz$Px*JhvGZNKgP7;5;6NR44`8n12y;Fd++DC3RPbwwptJSQ5kHAfS@USA zO@a2BZvmHl4{WRk%?(=B?wR?x=5{`q2=x%Pz^ZXRUl6&bF0j@b<|RBM#nJR5w45wo zjiX5hjwoo`2GrGi&$hUO7oyL#V73}NojkO8-yU-}QAFfZ*l`Y-T4$OK?xnE@TN+pv zzpiLiOpA4HT`)vrz8mk|=5*u>x0!1!aogOuQq)%8bW42Ws7;-_StdQS}o!ZscV zYufv*{&?5$O-`~NI_&u%>qspvc8F3Z-JPit+-?~wL^5rQ!xSgSPzn$3 z9YOmL%2Jdim#Rrq1AaRp8sJiar59nnpw|jDj8WS4V|&uC{}dhOd#O z_lm5&yl?^`Hy$1HDtwNf97CuvT0v(t$&#nhiWhu(f6Y9L{}?!j zgw~;AYk$VaI>O#v!1;VPW56`Uh0ETH9{kgAKEBT&>ti@Dop`!u$uct!^j$mY7+w42`t^|%OMzy%KF4fkTUkem^k_e-@yf6zQfoqn8O3QsJ5F3Iegdp z;0jfOg~C1x{5#)?srXcX}C*gkw^(kt(h`SElUvo-F^RMI~ zdaubBH@!!#LRs!KMn;oNXIt}jOPv#?vj3JkL(|55B*jm@uW}g#>R2| z>oqI&+7vauJ*J;<5w=0!FeXUSL-=c!Dnr}cuh?ix=qB$`ru5bPCe}oFct;>kee#({pBh@S8l8d z!|^BH@H;$IgV4jiI*_lnbw%DaY$fCQpfc`x5E)eu>=|(V6scJK^TrGy#_K5wo64u> zi}R5t{XAOQZcf}Y^q?l@Hc_OSTdO2grtq_38=pOW1s=wgawv0@<#umS>A-(_`$1e9 z;C=cGlYdqJHLA>z^5FV`$Q6qzHGaMRJdIaJf#WXRY9BL`mhY)DjGeM1Nd!x-&1&g< zgc?jeWL8m6X0CuAyK&O{CcKh#LZypPS;H6c6OKiHOsdRO#nKpaCvP@26h8cc+%^6; zcm*;e(Fw8NAD7Xb#m_LoHUX`JT}%5zPgUn21=h}!?@^hP_*hv#$${->sG-cJ4e31v z;n95V!thmm;1sh1i-_u_;t#9gJr+-1m#-ZJ+I#-MTLfuoBq8oUtJ;n_EhET$4V&oc z@nT#!Dc9t{+70WI9oYVWW1D_lElOd5oX_pS{yObu*6qGxF$9%YjM5q9D%0;|j6YrN zS^2Tr?=|zI`X)!e#BjG%#j5z5wXu{f6%6kqxU{)36RQPS<28RWF7*00hMvEwXnOxX z$=N)YQ9m9K*EmTMHaqXHA0wolp)zFBp&Gd$uv&0yyy`j12i?_@Z+_%pCYSeIDT zs7;cDyR)lO;d(v%4)wgB(m7Pef6iwkBxYEt*R5L{N2{0HYnob_l%@U-3@tv zOtz&6%@@e8=9VC62QYPHUaP-18kTVdYrbFm4(tK(!+X7lCPz8+S&F^jF~atKRzrMJ zWplu07l?Un3_Jiz#z<#oh_G9z?R*)!xnm=RTK)67=U@2;qA`fko>4wVtcU6Xrq_(7 z;CA@E-jHX}I)9Xdqq#Md+9Lb2YA|iLQ+N*$Eu$ZIflrK7KxoKks3@s&2~tdr1SDnc{6L{FP1%Ya0IR+&MZ zh;K_HGUCTw7>7ChyObFi#%zmdpY*!`#bs__;p!hJs&142fRSnOm51=Lo&ZbGBdn1K zf4Az*j1?iL2H4COmx#h6#3g0i$W&bXGRimG^~ILlDHZOf-AZDSA;#mup(-P|=3Vdo zc^*1JO%&7kL#S6Fqle~9!SMvpINhO3`oICCe%sU@v>#WV7AIPfkt+kosQ9x&DU{iM z)4kjTnMAfCO7{0nKdl+u)F4%Xj{7`8$FxsP`}11wfnpu+^iQhC+e;yyXATU8am+RD92QffDc;7+cp{~+Im2w_!Gat4WaEzEY^Gsbr zhQcoNR;QpYV*yGWLyUF9Oj!ka&cwf1->!ktl?P#4Dx$gZc4e*H8Iy3gNKVXoL~Y`> z3*j*gm7+HKY5-B4mN=f@TU4?j`7~T@UYGWuMnQ}AXBW=ROI7OgYrl^T9(CR$_dx87ZZWElY z;!xSm82UpshIzcjT*?5=4S3Mlv?i` zR!$n0fY8B(a!$iTvUUp834W1<9lWb3*NcS?A)@8niQHpgFl-QfYZjlyUQ~Q_ia5@L zpoG#Ox|1qLHHJ;0`(Doq?pZn{f;enHy$ZQJM4pdH8Wg@$z*v-kh&+2#`%aZ1vfQ*G zqe40U9L*?lJLypWu0MvxG*ZjO zQd2h*!Oy&Wg!-_i+ip-5q|=ix>HtK`IS=k{je8^uZw=tdP=-%%<#OAxuC0f(2|~Jo z4}739vm>A^*BbN3gjCuF1StDF`Bk#2?lWknl52II4&LU^3;~f8N;A0aLk}7A+@N>t z+WmPP2QdTp^5v)*@Z&yk?yP5T*a1)QlhZ*AB{1I~=vL+T*S#Pge{0~F<+! zQgn&5TVtZY7i2uDB`l`$|9AN&v&QJXSsMHm3iEjbEjY$?RO4SY8G zQL22(@RFoT7(h5&eMI2`5}~U-tT84Dh?Z{!wPm&_BlzH)7!QwI&p6dE8Al&H{;&oc zuIIGcL!gUh{?qjBsAd_uA_w=(&M&EeDo6)byzt`(F*n%c;r^8bM}X}p(PZmDjfmml z@_|PeGuC2?;FU@%CLGlN-n4 zKC9De^=p4u65RomjfKd}L1bxTPY-l_R zI-nXMcy`Rx43`(5VwJQe{H-G^UgwVbTeq!@-LHo-p{bB`MyIVN!hBUE`R^um`&V*L zi0Qx>jO4bzwKV+6mkBlSDAJhq69EP)zXi~Oj;o^k<5QXMRh$3kibD(TM*|F;Tn2@pNJu6NGR#3kQ*rdM17UZc&S`Y@EmO-5U`E;>yPG~S05mZv~tmABX|y+^gY2gIG8#- zgBjG5<=GWA#9PD*tfKD6Rw1Wq>#eG-N23u(@=we%82f72vu3&p3#Bw7qZOUo}`e>w#Y{_(HV$;COG9gpC*pFJeH`3%Iq9 z%ckASBXG8Fw_d0pAcig(+OPh;Tb2CJK9Xo`s^aUa#1m?Jtf)fBZHLa)h3`{iK+?h) zy9>`0)a>BEZt14Rq6-N=jXyh)L1p@(FJGMF*Fw-R z;}i*qK%=MLiuXxD%0%W-kqjP|m<0$AUm<_-n`|$#BJOcfS!mcw^dN}07_I?)z3%3Q zVZT-^qB`vUe}w_ok@rF9fMsA=NB(e&^!}#^%ww5nh(#xEstvlyo+_8`EF!|geBUN| zs+bKzvzSHO{we708Q<03fV}90s@uhN;SPybOD~)OrN95%x!0r#(#eMux^MQN0-1NBp>h$-twF9Px zvF1aBG`&#uFKIi~N(4;QE$utP{Amc5r4Ro6Q3D)%A%61h&o3DXQfN-Nn$LqOM|mD0 zY<+-Oy+1AW5t{!hhk?%>BYd)q(!T-2OTH4#B4X>mTji}c;oB#rcL7=P*>qxlQ|5Yn z9zwxLJ7yKN_4D!rg1q4vKZwA=$(qt_LoDS`L+j%Uz72?O>@C%eL>AcML7SvMj76k$ zNc9gib(4lhNe=F)bBS~vR>N^EK211@CtMAjgy2Na zpc6ZE7DN*$12v-0YZey~ySV@VM>I6Qcrk;+wFd@4GkfFz`}@hlBF{Bqh^%G|rnzX- z1Em#2{}|2Q+;`vCMe4)}zafz6B(PO&`#R~?$tnJgH5&c5W`Dg5Jrrx%%(!>ieC%@= zhXsj`h~xED6fP!xK8`}tvZa5YySXm`!O4gFV88<9^3etfCpiUSZ`ivxnl(so@s`|I z3M}x&W(0+dIi3hecy^eWH#y-a-m1m#rP3gCv*8W=Pczl*@CA&Eznq%ww9*q&)P@9S zzfhmX`?_^1yU`%OnPyGq>0^I|s_dvK`Ne;=amDT7O3k< z4*fi~hHpI#cX7N){6W0WGnnt|5}T$t;HL{iG}!^6RegLn2TcircW0|U*H`w!G7Q7>sNH`Tyvfo;oouyt& z$@Io>&sSpjvUCV*c{Aneq`Pl%4m-A-VRh+NdSsN&v^GYy#ep&<%R_91n2>prIV~;v ztHGo3AxaN$7`slBRMz`+{XCE5wPcxpb&(;QI5#;&7SJ$_Q-%YT2bcdeq_Dg`~iGu#QvyZ&?5 zLgMw}qGaBD)+ZCOsf$e;r`K%Oky&V;BzAfHkTNdVpyEVQb+#N+cC(&Sd@UtvK4{Mnzb-Y@_%XV4enpU8xjx6nnsRC;Ss-fvvOimS3Sl)0D7`P6;Y=B#6?s3)`KK-{Xs*OI zhLv;Gqq!=ou9cvN7z$ZmW~%ku%LYvxFi}E4$?E5_3z=ybguRH`MbPMQT8#}vym{l> z^f^V%IVWt*cdZfIiUD3x2=N{)QbZ@nAdQrVT=JfORl}RWBy`#3&~Y`)F_-5B((T1 zE%u@0XRi)L>o)$U9bZ0;u)qCvz;)*|hM`eO$U!!`u#j7NI?E}{qua(JVW#BR? zC)V{Ar>YsV_!r4^px%eW9VH5Ca@9Kb@o?5^qklD52@L1unBNiJ2M!zy{}qSz;Q8VW zTKWzCU1~g+x)tR8h)2;rEHhW5IP9$|Onk9%Mtry9X@`lsv;J--yKKlYF_mWa9YUzX zau@e**kb45{388@t%dO7+M`ZIw<833gL_A@y`q3XrkZioG9NEhA5OX?gZrjQsE}Q6 zR7w27j;*+u{`aVRy|8Rg_n?z2=Te{1=N5F6k6S%E-kANhw0`fODkH%^A;d{cbO!ZS zVo`z3{nOIanU^ho{XzK>nfg;RoFtV|yEXb3$wI`e_wVZ1z^Wga<~UP|=nuTkHX7Zt ztve@o>(56U&}wK*bA2ICBf65-qT(cR2QoCewggQW;#-Js0ap~hY240Dvz==4wv`LM1y0e=xn6h}2jZEBk zZd~H{UC2?wUyUV_beZR z!o#%A^`|}PUsJ&{3hEI)32mWBVg4Lp)NCwy4s)x70t@D&FKzq_ z5WH4fv*^@~!7dF{AFY$ioFx&oe_#Hi;`NGI+~#q9P9?Tz9yrnWjW`|-J2Sx@&X2ad z1SMvWZ3u{Ve_!?97~3D%q~#26(igtZoqoHZuVA0##H~9D|)Rq*Zg8K zY+l#`7aqTQTjB&>E6)wBH-u^3?ih&pMjKXLaanl+&*~H;($9spgWk?>3o`~*moEOb z(%e}mc>Y@qB?m3ri)ATZGY_K{szppGn1SA4qv?6Mi?uaZkq^eY*0^n4WNr-1jg3#> zTyn!*1GQQ_8yq~YgR7!A{6ptx2jX0yCkb!_Ub;4qn{02r&NH3<}bQ3+20zqY|lt+eE5kw(bCBLgyE2H(d?e%kv|>d(KaDL-X&89b4h-fJ<5D|I@PW+N!>GUkCYhch|I- zc5F3|ty(IFA|3(S^VuJ^qECx9!*aX;PJ5=g!$NS{sx~UUN}E9I@#{-K68CLN$p2#J zonBzO*uf_*rH43cLe#GGt2@4^C0CxIh&DSbejNY(U0Ys$uZL*%WHo-`oZ{zHDvSHV zM7;Wy+KVJfWG8Waua(>&7xB-_Vbd-SH$)sSvrfN2X_3At?kiXvo^jF*1FqU~{SmXb z&{f?`j>~ zVX#XfPah9A*MD?06kZT>C$6Mn<(?^S3X*DhPY3SVMX_QB9K+DopobywCti1C#`q@q zV#5SyTSwA7B|kxeQHEGM(t)wG^QH+rFu;AgS3H=0$heBCs?QWX;rN&RpY~7$QSq;b z$}tLnOizp)xj=(K5GF+087FCL|02oL zmPEDR#AzF_bYc+X)FtV-`KMk$=`Nju!H!1e0~y?>kNz{k1@6HeU;b0ug{W({sew^$ zWo=VzQL``hlGLRh4Jo&dz>+?46`)hOoW_zAHZMCxwE($q_R|bSR1?0gRC!=J|2wA5 z#LYAQ=H;!G`rkxYz!&q|7ez~Vc=LMDJJz2#A7psyhCJ}B>$+bDFs^Q%-<^wPq9;t` zHYtHYzYwAY$$>$3>d)A$wVKBrcxUDz4k~WCuVLvF5OuG{1uZ-g`KB{!SBL2o22&SS ziW!RVk&%TiWt``7oV+av40@$%p&Jz#w1&v^Fx)qW`GsxlBbFQB#7s zSYy{Yp8s;iXgQV3fhn@o;8DTny7Q{A?eqGAJcLaj@y3Q`i3C|tu>G!W+q&Q%f&`mo z!*b*((fE}x!%f_B)iqvq@14jTdPs}y^I;9E^@9yf^JuvhI8|>$ShJM}-2xQy<}UIW zp-D+O91lWe;z^LxxN4F;M^XFpIO;$dV~&c}M)a#LD4l{Hatoyun};2&d>MUvKC&+e zvTu4BJ~YL$ibe0?*|{6fFKCbgCU5p(Y-g`+Sawkuv~Y+ zM|B9M1`w7>OM4yK33nIYAyj7U7mD7-05=3^$G-Jek%cnw0WDwxjyfKrFV3ac7+R(S zlCA_j_=dN5xC;J{)Bb@x1Imq15#n6%AuI1w%oL38_ydrDy%ndl0nIgq4@;(+friEc zmFI~8x%Y0Q!L?IJf+$wKIQ-kr`PwRq1xRPf2;gb{*()h05z$f^k==Ms@xFEt zh$s%I_l`M7@D>r#*FjNh`}i;oYQVef9*aVjw-}nU*&yZ97&W~O&H7m4+MR42phEbq zD@`oVf6?i2azoTP$l@*s^}|iD??HnYC3zduX+e4!q#GWf`JZ~8gju7^L7!n#Kyh%0 zAiwevgMDaxP28VD%I)XF21?>!fh0o*p~4xwvoRKX4@+wS!b# zLHK$;>D|a*O?PBLszVB`PT}ISpQ`ihrs%Aqj@#AhtNnR(PEM66N(5}`w`l)HwESVs zeoV;)_1Lbnw^}e)fmO`5b2Z*}=3Zu;#0T5Eq)=ZTYK;#6@;F z9zPDw>py9ZPzhhgne{EU8exB0lpIz%fmD^ur@ z%^_|cXxG;Zo`LPzPvzkg-N!@y*}-Ms3q4p>zr@SXbF^q8WU$A$bnga1Da#jhODk+; zkY8QP+uz2j-`a25UG4i)e~b|4f6U$7K%jqmf_CATeg72u zyry7=-LAys!5-t#UhxC;FMaB`9o$(srvov-&qSJtmlk>u1r;62h?HKus(O>WSi?Ri zdEXnNI@JQw4*XV5alnyj(*{17%oAeC53{CF}~f;Z$6%fR1*su(*g1NzttA z)5z3OoO+$|0Lt9uk5P=V%$1R zEm)&sad`Tshz0a-z6vwQ2KB5`m+gxOEiw9UkW8ThgGj?D+2n8qC~>JP&}%Y=JdWbcLAam6AO5T6h+%3lG4@TklSPz%)Zy zAbym9A~P8`#SsQ*Lhb-4CMZWeEgK=V*q{X_Rrm(CI~OP}Lj$Z4AX?!40)qgycx%^o z7IJsmwFM@!;?rxs!~m_)0l2Z&P~KlPv3RD9F=OEpwzq1}-ix4#m^Frj^|4=VZvPv5 zJu%AHX=R1@AOC2ItuLp}1tQrrlH2ElQ2EGMMPnFT7Of1Y3L&6oN~WnB(6J)JJsJ@= z7qx0X!l`JTXi@735%`-@bo?N3?} z3Pgl(7-E7LfO8piZ`YcJgd|}w29eM0dmi802}4g4wtu{IZ^*&C3`i}p9j#0{t5FFe zvm(suUCdOQ-_t81ZZ|ZS+hb0Ywt4m9H=vPBo-5PP{#BN@LgUsX75J> zId7Glruf2{07W5sJ;(4LLuxYmtzV03%xkr)r!H3H?k85Q&AH0!u)j#@#&Y0e!>ZpknwJ(ls80eWHPlg}P(I|R=_Ta3%p2VSS3nY#F8zJ{Z| zn)f+PWceO<6VZ~FgVM?%4|z-8mOh@G&1Pf0*0i*X!$aA?AoAPu5EK7f)}%0xHJKiq z&`qdtaWRUu^d8LluL!%dPr1~yf3xFudCyk!M}ibHE>QJ#11U)efrX1DmrvH@P%!SK zf@dXu>RTyv=fj)Euzg2V`1UMRo*~@`GN!`Z9sq_T=XE2z0Qe2)OHVKX3KekU^B9&1-}IgQHxbhMTe}uN<3JR$? zK)i)8dmWp2VC++KUY8Gk7^ihRcT7;PnX?Dxgj@xFDfOJ!t^Se}g#Gxb;W5^Z68di) zCNrMakPXjDjgS!1d4G(kjx>Az>Htwe>*Ihi&GVuBn(K!#1tPnHGL64q1&MNj?eZk6 zs7#w}RWR)uL~Tfv60n-60l+i}(Yz8gF?#)R+-e)yYc)|T7sXq|!#}wCSRnjUl&o0% z%g_MQsKaJTu)rC|F$lz4Y5Yu6NF|_-O?=@*1@t#}72Orc|7`&9188|X{k>^8jGWUQ z<3d&5^bvY*`dw7Ia1W8hd15xVSNOkwai5s|eEp900HuK2+LXv`H;0YSUbOM?KL1rRb_DFyoKW>p(tPI92kpoZ$2@v8K z+H&aZkC>R;>qVPX9zn`r*&W%n{J@QU)XaN_oIVYkyi3E-+$0l-MOaNYJ3t{$CnaF! zYYeX1U(a12H%+V^>mQM>J`D{s`njL;nk#Z}1$X){ z;4B%X+qGQFx5ceSgX|{O*&Ft{ZpR4qIEcj@Qn(w*wkD0L@GLO81}xu!2)3ZmVJJoY zR}T+eibv4bn>2GErX;N6NJU`0ogr(q`aMN(-W&I#8z8yEBgpM^{0{_ZMV$iuHoSSM z*aeydbrV;T=NATfudbeap8bfX?S%dF@hhFa`QX{pXZ<}kaeKy$bGQEh^v$8OX;=`w z{?`j@H}&Q@Af#nYQcx$W1Wq)BJ4=#AKZJ$sX|K_MYn=+|W#bpC4B}=UWLzh$KmY5w zy@RXvTe#2SuQr5oUS-H7Z#sfF4jcFj@t@IX>j@z9JTXwIIjoV93ApqXT}h8mwX(2z zXWicu0X^<%8ZL@ZX9BIq14!X{85U>SuP?;zcoG&tPM8`Fey&}2{;$VejEv>v3RXA$ z37^C=^l*;usK{itku-|JlWl<`fbuXUIts%I*+QlRwkS1B09v{sS(d$g2vt zg6@oZgx_=qgz!?!TrmM8PE(w=>4pr_o zL*2n-a$zCIbS8y(F)!JFF`eAphKp^|6j}_W49}$B$+mvtzBG@t^kSe`>t67j32uIr z!porGDxPliLB&vsERlFSiY6i$JskGXsjHN7e)D8yQQMPD4Vk1>noD!Vbr-R=)AxeO zPt?Y9M4mi&C^h1`Ff8mY^9_WlX{HjxZLk zX3n4*fVTSWnf>A-Q3;(c_YB{8vd6lQWC$$Z@e|ecl4aAt`!${WNZOB>7IT*-_ckgsdsEe zTF-Ed%kY|gqDlY-@^rZ`9cen6s>p5l6DR!JL6pjL!H;~r*jM&nrTf8MPQ%WxktBXi zt)g?)_6%B1Vo7?2yv>hoC-~x!xq*j>ky_C%d;R!P0AKAlTU?y?ks^>uwS_`6P<<-Nhjw!6Jqv4WN%k~vp z0YAsY;=`h8#M_fQoZ*ZxW@^K;@R2RkI=?0~$1xmyQ{YqEI>e|7GG&ZQ#V@!8QoXd~ zqf74w%c?|-2R!+_Z^k*uUx>)@%^(se(q(>K3a{_-R|!o(gi60Hia#y8_#H~IKhBdg zqWTj1Ku17Wcf^6y+mQ`DiQF5IwoFhlY!4Ti5S58v{`=66nwXD#g~+dR<&y71KH!qq zH}t^~6jU%PJ?n;}D{XrAw0_i_$4qJ6+%bdf`bNcl>XptCa`6TIpLCj4$djqkU!{Lg z!Av>UrT-`i=|k=*b@tZy@dDiBiTPeCUd_reh23q?L0nNzcOpg6m3&uQL3pR-vBflZ zsmZ|RhRc3xtl;$`ak*w1%t_OY#KH2qf!F!nKM=2^tNLD)q%R`%Ls(0PBSr4$`iWX~ zmEh(TU7fcY7)!;)_*5>*QE7S_{_eG3GEV|-kxGmkvKLlqbT2K9r{CF`zufYBTYIG! zN!2Fglz{lAeEVU}r3M@+FVQn`Pn#^ z*&tU?;d)xsphHr$0WOEtb&yu?^)zy>TDp9tZ{$GaN2i%!C#+aJhnv)GHhQ!K)AuUO zQF8PRE8yVzE`IFEb$i?J4b?Xi6L{vypp-ahn`@uEztrtg$?4?UA5wuIm>e8E;5vNl zw5hrUeyHk4obKNH#;mJ+%KC=y*v(#rw^#C3T~(X5(A-y4x4qyrj^c{%S>FyLqo>eU z4{7a$6|tmGxTEAYn;*!gaI!Eg_YJ~t#=FZ4P*1J+NhmVBL7IB#XK8gP->&Wcua6&D;}cTQBXjXK}BxL3DPitKwWj}A~L4c7{k-LUt10dz@IQJjbCdlc8k zR#8`-_bJT3bCVV*&dG1{1~^)yV18fEgY`evj2iq~zAsW_0X!uc>Px$Xa(+3S zk6x_tohsgl7aKv!e!;wX7(nvu;OB{jZ+tjo@qT#SDOI$Z`oLwv;g7w@oQPV$JcB7t zBHc!>Dm?k#X$Fq%&6@X$SlkL#S2B&075d=Uh=i4Zj2@su>W7j$Z36ZA=VxRVAWi2s zsQ#G?KA@od9{LLmY1JKl&#a%(4JmMKMs6yn+J$z9SiWVoa^j_mc!lgXBE^hheY(Fo0h9ADE4wa~rZ9`Fg-WGy5e4BN-1?`Ia&xcR zz+t*TeSsn>{TcEYk(4lHfg8TGyhXPaqHkDDKEX1=Q2K3FB@D<6nUb1_p=0c{E@N}O7UX%8s)_ev;Go` zNiR^;ed;<$b6%1VEfyJYqt9ogM|K}gHiBbG_U_jb*Nv`Nu`fRuls3hHv<}$#Svzg^ zJ6{`JseaZb1+DouK_x10^+z5Oz*n4l3iX52|6Wnwh*WVxQICGCm}9gG#p{qayY?00 z7_`Q$52b0PF>24Dk~sr$$o0P*d5@$5+GW1Y2LgKJ-pMqr&qMSUIR9-y`ib(8n>ZxX%nal>GPs0)TXwlL#cttVK5p&S6+EL7 zs7!?FPL@_4nlEteg}KIMzb?+^kJGfc9WzNP&`M`PQ7mZgVRpG#ClqUaw^zSd>6!qk zVuA^sk&X=MWag672lhLE_o`Z_$DH+q@5B)EvFG`Ji9x9CTI|T6N?5yR)z?~C9({OU zKNVvsvmJR??nx(%qfgxj2kqo4k}MOA8eha=M@dDM1>~ERG)wJ%Uol3m*@RqaBd=L&MJt`04%iAKx316@a!N7WYr>+xUtsus)RqSx9*ow|N@2YaUGL}1^2NKolD zWOL=v`VIxl-t7zql7}+I;7h{4_>D&Dx<{3M&&UK5DF|Z$`B&xHV>pmVQ=Q{1?|akB zq6$z~3>!mm{pAdIEgW=3;!?)52UfzHUSAzOSdfk*4SP<=d8FHw#cSXwkO~@TjpO~` ziLDCrtR$xXWaj3fFy`@zmL-PFCS8x5`z$BFCqv2b=FYvR{al9o`cAZzdXU=}dQRv^ zX3NK3((a{?)So{be2?LU`bKJn-$?`D=(K6x!vEw<7W-fG(n2ekfx!;x3wRZ$P=ky^gE>h6=z=uB5hDHV3xbOA8B6@l-#JN)EYX zTPk3925UVIO%5?B{$_Bx+djSA^iU9^+~WS8%>6ZJB+^n&t}K1*L6%)z;d_MfW7xPW z@rH(}PS$3PPy466!`U5Mr!M9~%oW0&yXU%eLgdiwR_M>5Ep8wCtIq^t5SO}&S0h@O zT_fuJqUl?psWH!y!;DYg%vdrUq_HIu^284%b3U6NZc7@9%wEp6L=s_yLx_%;CihO$TLUHqZmp_xM7$GCU^)Q&IFc;S-Ir4u0Ez1+6<_%JKZuiB*@5~dOSDsD2 z6YLwA5+TX$wrMqSZ+QgL^|HRJIvOgcO{WvbcN1=PTFaMxx4ZKX;qjaHNiA4HN5?k*wo8;_^u8%G`R}bGp<#?t$`Ey3qCvLxDbtzY zM{eAG4`Sm2n=s|h;SNjjv z6V74Ab9H&o(#TCG=smJ|pFq2MJwyK zojZ^}s>(Ir6SP%%vX7AFF7X>yiqCn(=IETljHR6p z>{y1K;_(Ez!%!dcL^(fi1<3c0y>vJ4Ol1`SQq2lWFT!_YyhHvb&k-gJx zd!-1S7%|2D+MP!&a@Ri5N!t=mu6ml4kc6#a#=jlbet&}!%SYHqFov&#<&+#^`_7dK z#N@$f!dE3mhMY&}hNFfl);OTqwKxmX;BQ5-%{zJnV2!wI|4<5##ZkIDyd$Iee4-~i zquF+cMI}BgD_V#kxohP`e}B`yI`a{oA3#30^}IImY= z?XEr7z)2XkDRzJFYo z`ZeKU-^dX`L{2W`GmkW^>xfqh?{-}da-)yfKz;xE4H*IGVtS6>$iK{g^Vz?~fIm-{suZHF>XEKgw$ zvdvm-i*e^L%KYYbxfNjS1+2I|8e|t|kn#LcBkhIun#1y9G@JDTNqpBfq3{LPgu`w` zn;fyIbHxF4|zYnQH;wIFaw3rMgVk)a?bLF+JPi;rAtxz)v+wRmYu=4N%)w9~T zA-7Mzxk?;Qp)lfDcLNduirLFN~M1RK{cRDtH)|DAH@~ zf;cc))w+A*o`_RUd657|kna0AjXa~xC;?Q!SMxVt#-GCuhexL`_!OY}r%s!EE<~YD zYgVuQ5kxgckh5GJ;8<-gK{4XQb|ds*`j<&#a#NVa5Nz~Bs4N>%3it) z_LAg&rli;@mR6$0bg-9zeVL3mo9W!?u!?-0?KX8s=&FArb}< z)|O~Y3RclB%V^ip>bOx#B!drIDdVZ-0YzZF?6wnGR-rtVpUX)54UR)<2rt>{7?xEh z=a59Nrh@eRciDYGPuAU`$?lT#NO+B$@sm!iA*36Zw@qv|oDun3nc>hphnK`%c;mKx$##fTNtMi zL;ghkxFGd){8;-9ij4Q>?3MPDM7ig1*_qe%$sh2r_|n713j4s+kf)j5fl|eNm+htO zvahnlY3(D(Av3q*iayvQv;A)9A;HaMq-cNqPJJ^oQH*;jeC$at;){^)R!MAgln#^L zI?=e_h43vwj*-XKEN98lVLG=g?|Fl4<^E?nr_KK(>AT~peEY#pPL zC<<>h9DB>G>>J0(CRrs1S;>r$(Xm%JO4;FL94n(^9nLxT_x}8T|D5A-9_P5P`@XK% ze!iYBymRM*-(j67t{S_2jRE{QvbDAx&c!sb5rkOhA33Cgz+2a{6vFEdms&GowZVKv zW!Xe9(?f}3bSH?NVYMlZhrlK({WLPxhP>vFG9jR=dCv0Ha>7#j5L`*}5U|B#F$l7n4_Zyj#3^jvSQ5KOOn_qg`}0p3Y= zCI!ecfszE$){z3!gDgc{#oScvxof9ace>dvtoMQB!uY@5RND(5bcd>P^qp9 zOWI=8ALkhP_26=aIX_(6Ibvo~_qFQ-(}4RVK4XiX?Udl1DNuy2BL_8;>2XJ~$x)9` z*YuI;!LRh~y1U-m?_UiK9nR(sP#>G*jsOdq(Y#I%LFd9y*;j!IH+lZ*IGD5ydXCK7 zXgY;9SASue@L(F{UNx##h7nr?hVLun?4c!logj|K5nO0(^cQ&Odu%0Cc82`(d=Gog zy-Ln~`dpB}54anux?H6*wIi9|TdI}^>Gd9htGddXza?@#VN-S+-GP-K7*`;xdto-E z1q{efeW3t;Ayn|)a-k?c`Ph#*2%HsTlRv>TUN7!f0?V~Wj=gU$5v_`4T**Q)p7tRO z4|Z{g0MHO8By$JWiwM^ujIDZ(80>~6{{Sr2VHSxuq_;4e9Rp8!=})k?YA_^Lo-l(kM(#=Qu{5_9c|vW?h@z8Lk8$9v9?ZfF|p}?(9K4 zNsJp+$w!PXN=Z&-lJf#8z+q7nD?+Vgd)MP9T;6yz4s;_0m?!7;<< zrd5=Zx5X~%;8<{!f)@MVCq6b%w&<$GNO*F@DI&=c$Klw0vr#5NQDj+226s{M za+pQp&I=pRQt)Ivj=c&}hVyxIb`M(hvfDhG`uayMRUg=u(woE4@Ilhn_fHxZ$_9oP zGCdvFhL4wo6fd{S?SBuU+~gTz1;1GT^*Th_L>tvIy{2gO$Ao@u>RPc&^g(9(Z~E|X zF1XXZRK(t<9s_!E_cXJ3jPB0l?lnH1{>yJwrN@8q*<{yaiJsrjeh{&y7|7Pn)ZmfV z80VxYrd|Y_h!e!_OZ=C=86I-KM-kPYDNjF(qT6E689md;{lf&F);e7*1CPQJs`|CN z%!9kuCy}3@VcH`e3qEp4DdO_?8rjC?mfy0Bu;uW4_PFR2@vi3;;;Y7OUgh`z5Px#< zA!H&j)d{1qil)50S!n?_s9Z4ll^g^6s!HxRv?5ljEA{t_=^h;cll`-PHt};TYv;X2 z%&qPu=cvZEzJeXzx>LXz|4oXL7FC$f$2nOD(OmxmZNdqO%>2{2#5Ly+P(5xyaG{D< zd5G<6_*v38Nhs|Na7a{+O|0t$bk6J#n3frZJtzSNOsy{|o^d`3v6GmzOdT)108g_A zZU1?!N`&h-?QSx*ug8P!HFVl|9&o$(uhA%`+ri`HJH~2V`quZe)1H$R8=0sw<66E% zkQN)B96X7a&D9xMwb98h_%urJoH|NwcU~Gnu$dIi%5k$?ukk z2vxt_ckh$%gZ_8jcG(@*^8TgDtp|k02D^X{;+0%fC%neVXwG|qU=kh)uz5I;0uBRZ z+fO7l6ex;boHEyU{L8m6^8zEES7I5*3ALB%VT)IW`%U-$@0|P^qDp5 zzu7z}rO{VrjRo6S){3K3gqi6x|J$8lcmq+l@Ewr0VAQ=|m|cIK7S5XMKmp%&OyRM= z!bZ`mzq7!IZJX3c1i%ei$0$kl$5@2uI7absPLhSEPRip&-IXZrzursWgrK3>Wy|3= z1X;?0X7U>Bj%+u1mNOwS^DzD{RZ(9wLhY;4T3W+*NC~z35mVHvxISqxTig?yllr&Ey&$0xDoIU3Je1 zzPdf*wgbt$N>h)I@iKXV-^A|`@r#q{nnoMeRAD2uag0HA?ZZwy7-n@?sLW3|8Noc| zrSrB1DZcj+T_x8~kI#QUs01-7Ag3;PO^L%#IyaDP*cXN}YS=@2e)#hwdM18iqXuOp zKcV>se7|Q3@q_&V3G%b~Lgo}iKt#w?+zx#(#quX(HRxf|<{4>ML;Z!Z?DpxK6h?#_ z!ze)UKX0-M4Fh@is>%du{Q1anMEiAf-Ev(e^7uk9fE9%9H9zgOw^1E_qn?#t&&V6dqwND0Oh8keeuU)!EhjISeB1*p zs2cN1yDozUC;0V5!`EOMdx4;dB33Q;$uaQXJ!&2>kcY$2Dz{eP7?_-cB*HxZwoXf! z;xVv2oI4Ypon#Gu9*7$ySc8X`!q?cgFcbqj!>z@nfmR%jf+amXSn&n%djkR{rwi{# zfd9Hm7Km2fK-V^cBYLu{#NOlSDnL05Dp;|#V0mLMvtsYq4YA_5+>>3ht9iaSPIxs+ z+FCbYxnJ2O_^UG zu~C@f`xVM!H$pNiNTiKspkO%j=!XmhuR`~B>e><6b%D%Pg0o8`Nw^K%KRv!fdu+}N z1}70&C0Ija6r9dhAsXF1(-4KwHi@rYB@e*}4I1}kAV1h~-WEioVoCuQ5FO41XQK?E z%cgH-pPFwEgYlJn5|GnyoT6+7mNz(@`GT%LMJgYB0nJw??OF@3WcAD+EZ*Dw_Li+| z4e1U>g4j#~-`&p#)0mh;ByU%Vk6~xkU{?tjiAQ^AXHw z3-cFTRtAf821mO}KEt4{oURhjc(V-&68`1-f!z-d=Jo6iyCX^3%|&|ENW&8>g8BVK z5=4ppFEl0ZTgChp9dOf>i4i1f%Jue-h@^46vZt8-)k_IW$twlT>x!eWd@`UA%}Wh< z8P7aj0rfn{doy(Wi@`@jtwgSB`$*J`)-Q|M3ztSB6wA9FukPjn*Y2N`-{>)tmO1d7 z#2hELP>q6L^q7nkLhtHpd6}Ga8CZnYE8@6K@BfuJt*u?ZG*`K2ns;RUOP^yT;*D7* zB!;gUO2EoH<&Y|u-+=aWHX!GrWGIN9je){r-^WsxyIv+Ob(jB+dYPzu{9>UJnld_%gVq8d(>J7;zM)b#!!sm;BTC*i~YF4OCG7-rvYsZ1=q{@gwxLw!yC8LkJLdAMUSL)ASlFD^8grJUY z*g8MkdlPnjgOS#RcHvj%)A3nW4Rk;~GqN7xcSp!(`~{ruZIZX|LcQqq287?DBYpYz z%znp0GA*eq()p7xKA``iB{5(7PloDCE~AKk-G=oi7SPtPq>Wrf7u^w6!e>tW4nwGKfx)Dh$)kwhqx&d== zbQobMKoNFo=%yfwnx)8(xmqi&*lVtu-Rz3IddS(Kvh+%cHe2;Jb<5kjFe5MDs+ZsB zXs`Qp1WsQKiB?CGGL*eZoso3^{~nGcoBlkd$r7H2*-VNo@eJ0&H{Th`aL2X@HQy0O z4=1if65%h_88z$U3A|5qEH*VgG`=zf37f73w*;@aGHCw=nyU z$9d+PMhF3amb)_-D*g>wEeiX7W4d$QQZe^`@GR+yHfVvlx>ZGE7< z+zmRG-q=!nCPwUn_HUTto}{u1=<{d-K{J_JDZ`fSAZQU$a!kI+1&%3W>U^L#~f9`!sLarDVW)3zFkMu zvOS+~&MLLL!f_Eb#nvF=clXhoR^y6CBI$aC0^vs4Eer0F^w3!fYh7K)=uK|DI@ z^Dy4F-+f||aYl?!LBJ$FJcNWZ(U zRIwrAt2lJ+CQc~rOIW1A_sBER7pjrRYXLuAY(M&;EW3)*M`gk1-#n$YuE@1IY_6;bgGr!5 zBt_|!D#FsvcxnGq5?@#y=QS-_t6@W)2T7Q=IqRfYk~*t+f|PzLzrVZK!xMaVrvPH8 zR6{HE!s=S-J>VDk7VH^xqN~YNpp2tp+PV$>YkRpKmW=< zqwd@9l>2pI+xBVK9;Wa8!Nhi)%^H^#%?jV&@0*&q>vFV%U1n}8?h3p%;of;U7e)#O zpC07RCr73Zu`Z!{Z)`E@!vf$B0M(g z>BXO~!YYw8v%*NOjc=`fsTJZODbvGa*ZgyXJ=DUvF%I!l>cx-W9{_jwuKj5DPmSO) zzfj7o#MhG9rpV?snr&r9Tw&o7I3B)e5iy$T6gOU`BR;lGNfITiKDz(?x5L*c*E>$~ zsNq*w-`v8C@yK0Tjrmr@vd`-sj5#d`8wn6MtlS>zSa&GmK&9I zDC&yo7LTlOQ~2y+YrI!Vjd=f0gT7#4brk>j3JRYzP~9~;x%ckQ6QPT0Zz)rj<*dx= zNH*;?em(o+a%zbA?S=u9?d%{if5Kb6OKETBn)AWgxAJF~BnY?Ti21D|JEf+L`j3wG z&-}6aTWJ>+c{V6X9e>O8oJV#~+InGmUuA?T8m-@9zd@q)*R7RQV(Pi)ScL|>;Bx8R zOoLbIk57AXzj|~$9&=o@{N`K$I(FPwQf*+idyyRlKAYxc1FNj)*f_J7)|(g$G##U> zVt{TP(}ruzIH7jrl2Pd90y0rdkz2)1h%|BIp<*ITAiotTFpB^N<>Vql7D@iS!8)IP3-7^g|tK34bgW94u z&sNTB!)Vs_C;Nyiu1qxEHeXBreaJqjD-P8qWs-`|5z!V(6<4?VaF8M)8)Upmfj>@( zs_fC`VFT%2M@@W&?3SY4g@BxA*MljEP`nQ|={lKVlRpw3zqUr#TQ7+*UDM~&N^@A>_kCPVLV zn%5WpI%lcy?SL+d6+7}AUb2Ukc<8Dyx0HXZ1*@#pe5o-knWLEgiJEjxQ--9_;b^N(A4?^+T669#CB@eNbgdG^-d#d3S~|-ZX$Nc_I5_YLr5qw3SlDq(mV> zLtgE6IrZ^{m2DYRF&73O<_~#Dx36=_(ogO440g(0-xNnTz;w&4Qkpf;9)?dHgm1w5 z9EB08>JMSIIy6_gXxS$MLsb4`!7B-AVx*zMe+k!Zj{yebIu+Czf;N=qo!|(JUiCL9 zGT+_p%S+2#_KmXV^`)%R-gzC!121BB45nNY=}&0V1{1dg zsMWBTQ&2EPya3R7ZFo<91Kck3M{%!Eup-d`U(mp>tked`TEq@+d;9vUfFX~@WW7-a zyM8NAJ{OPMC3Xos?b>~t3kwtDCdjwy>E(G-+0V!-vcbk~^t@j~`ovv-1$B0R^>NmN zM7*=-7L9AF|8>cC9Evk*0WQj*X2+w8uX(D>g8U^=|J1-@{$3eq0K{*d8Rt}7MABgA z49sj9?UTqKRJ9P)jR-(D=*|gG$TX2&C&`a_K{6 z;r}Rb`OQ>vC#=3c_35fd=D|#At@-AJLrtRPOi;nc_E$zE(CyL}m~(5)oXl8y&!FBO zKKQaw$Uq>x7befV#%Ys*N8%Zb%l9QdG5-^!f1tb?5aO;de0$^OnB=bh=P6|`)O>d3 zi~26&t^9r=S=b-tXq{x>>Moj2fx`Q>s6Jex+#2_wS zi~JlI;m1Rp`+KHRkCXGCfdidCA1|Y+C7ty0fY>fe13C39JTiy5WF85#jPzx~2V}e7 z)qP3OFnpIWV=s0M)NEy6?;JK7l9Z>y(yjz#-TEVM+^LMk2nwSc@+)sB&n;m5*9P^Q~;F~|C}#1%I(^_d$mO9H2mar_%Lu%-oS-p#;p;p zKJr?`FjCwz`(OjF!^T=0zbh~Y`P*xD zMK}Hx=a@&+FyB>Oo4p97Jrjcb8dw!HjD+`k*qAxgC3bK7bmhlEZt^ozF#f;SJmU)D zzf=O@Zq*~Ik#Au0x1aMRP?H%e*53ggtDYzWY;_qmt>wFM5*oZw$U6}!+4nj!-o5kg zb+*Bl8#p0v)OH#yIn+^D1;pcSd%5zja2ztQ*E9PsR@3mJ??XUIqeADE#T4qp;X(#* z{wpKN2(967II|>xl|@BYw@oj-WELK3L&#XX%5S7~ZB;?@1%GKf7eN228ZckM5NQn$)*T+aLL##Z`G$Sz7sRh(hT; z^CwwVbyjiO97@{sQXnUV6^C9fT4QL^t1+x|Ri6{)7Mm7RjeO_HKiG)~7L-hhXi`oACvB=n{!NCw!jS@EXWIpT%t`BKQfXX zQd}Qk{)xxmbXu}#2m46$kd1n-zJ$}3{S8uIz0uupSq%2`9khCAv?HNQbuc^{hZANW z1yuLAjtiVAhvJ=Aa^`r!c6USNg-pq8ODK`u`52*CDB_QLUdeON5Qw6TIG_UL!MvyG z1pxKLbcNY~Zl{6TZK&OB1NO|Gg0=agCz#rGo(YA&x+vmJ>sH8)UAPIeu8Qfv$nMlEq9{?X9@*H|-a>Aw+=Y*4h!~@mW!V$F zp2n!22!z952@B0n(Cf5FEMiO#lMCc_0aX&Z@ z6tUHip-;WUNTR!2ty0Gqvi;aj)~(~9D3ISNJBmN|_S|!yH88OBQaSB_-L=;fb;ufA*)-ZHQBHk(Xd4P*F4YCOo{mdP`2WXr!2 zr=PA_*X~oV(v5KMkGoQDZ)f|u0N|a@t7iW9zQ`bY7}0c*ct^)anb`GRRh>?af!zq+ z+Z^*;9OhZW!w7F<7Z^Ah;45_$wwoXc$vim8d!SSiFMUYUsxW}xBG*CO!&9n*lXgvA zUGs4)xdM9cotJdeWEI!zEQxh-0~h!^-}FSj8-svJT(pWc|K0V0_rCP}A@(qUev)79 zlH~=Yu<6U+I@eO~S;AO;fFfRy6-a-c2+L$Xt5m!fq(HtBKl&bzJgB_^y9o|^q6mYW zji3zg&x8L;A2LaPc_u%x-rD!4X6JgFPzNjfEH37YawSNZvX4F{T@t00hqG!Ky@~x8 zQujTP2QmY|Y2DZteK)`_obC$0^1! z00X+ zhG6Ad+78C(YBxz;{q`hFy1_eJahN0eno`A!8Je1Cqi_X|2gt2Zd_m^&d#s`dn-ZOp+hpoJ59w!1u)^1-8jyl)1a zx9QxT47_MkQ(OX?vqgCOY7V*_b^oyv2;i5Q+up6p{+xM7=gnPs#V6{pl&=LJ)XAw& z8)xgof0-Tojnok0bW#^(GosLCnET>?Tofo8e^T@?GBAaNdT)ADXu%_$EXGMSK*Pm)ELF}9q5F1^DBYpqvDZ@qo7EOABKkk`uu>mACbl; zZqFC*1E>3YAsKK;>?BAVx5a>7H#sn#voPM1E+LTU9_-Bn3h=Rz`mN+iLi)yG$=)vH zoarU0%sX9({Z0e;%^=EQn8E`gWypOkKa^!7P1r@7`Srr$ulJam%|aJ}bsN(a09RU5 zK6PE`5SqUG^0tGpYIbBlqK!_=gWpLQVeWqewB!krFa`%=meBmPQ9keT9LKYm@1Y-E zskISSETm1H_$9yIqqNtTF@A#Yuy&&a;LnZ#!314C9QuW*kdJO!LaN@P@j&h;>MHev zDK}aH6qjFB(YfEj5D7lpfy53N_8@)|2hD&mTHvXI(a|@IV4jB!$XP+I@(#MP2IcV-XG zsP_`CO?Q>%Wi)W7`Lo?}U=#IY2ZO|QOg84SOd_)|m;KCOr~Az{5Y8I=@ejft1(jyt zn$#j84X@$aRnO%#7fdfL@yP0rPqm-chS@mQal9pvJk3-tfM0Gb+i-zT;UPhu1;E`h z)_1EsRZ;VylLYw$J^2wOUsrL+n|1D19RYUQ%|fu(Ea1Ho`r|3<${e)&@8l^(=QLo( z1oSibWaI4^2o$oVTJH<}=L4j^)pV89QnL>gJ?-?8@a;LOSzIdyo^N(*Bzq%H?r>*q z;K?v-JW;rN2zn4PkHlf$B5i2)tDtt-%$=lZsZfrgqJ$e^-M=H%Zh}%hDXc-B5(or5(pdh%w_6*Vu&LNt#-(uE!0js;_$d+WXvv zV~Xr;^(OO~C54zt zRdooSL#X{P5UR;*Zv=WWY5QEE5Wf2KI<*{AN5>|D%R4Qv?7rrJUOse?)qaRwnJ_*@ zaG(lCPHQ#aGp^K8JR_d$l4Ne|{X@4UwLlJ$?ME232^1OhUXWi3MaN>i42|&WzEbzSS{};Gz)C9ZhJztRBbf9%<)5d6^ZGazj z{VIoGDmD3FLzavVBx~udH_hdOVd5N5Q=?jfZrD>-@VvCkk^Ia;UitCljx;3da^ImM z`n%)lrB1gGUQ}sMi#0RZzO$ruX|BbCgxJA0{_y|C|IQqidR*8>|KuE!e~9x~M}k?7 z6s%cZVXNjnWc0g6C9)HCL+4hGccy`M)~AH1!3(sTo~ZW@b+^XFO7KXl5+D$NbT{eW zn@xDLeSKWcQ~p%oteINeLD^kNyp+8o5!YK-`|8KXnQmLly8(m)?epLP z{hE^_?d`Myy#l~8K5r9}HG9;=-Sj`ZTU=mXVx9TS4AmPuveikFuB{|TjUwd=P7+pJ zr=lF@IQ|U2{>4p*U*PmXDS*kSr62F+M}1X^_V#4}!`AZu$rR6rAT zBc($Up|^4R;&kq&?m(n%p=lB{+*-_yLeLA2X=MHsB@*EvJm&wU@^Cj$(vAMV)8o&( zH~aMmbPq~BlA-B5npc+Tu7T&PEgc!Cm7sa_qPqQxg;VnC8QqYDsJRv5SSS7C7Uc9! zaTIx{V-dG%TMvLuzzMa6$nn?T!q`eA_nx?}aCP5K1`@B|3}stJtNDyyxvZ)_Auq!h z-y7LPj=LC=?M)T>Lw^Ji40?6(6>*9y`W94Nsp9ynS?Xc6<0;Ot*$2VUI84Rs=F=khDCs*y#pqDvh7ozEeV(Qb0Q z1{C6CDBiLopTQr#bR6gYEH>SvXkEKlHf8D{yg=Ao6q55KG4rpmmi5e!lF|BEKhKf7 zP7IkDdU(L9(r1t%+=4%h$0L(w2Sf|-NT2!A7k3@%`bVRO6xmi2uekiysPFpXZ2--A z7N;440u4(UK22D-_j8g^7lOd|T@aL9M$cS($AIUl3K1ID7eE~v$wjN^OvCodVZP1w z{GIv-9!BsRTS~B8P}#Z2-wX;V^~Z77reO*(<6{<3vgfW_xzHR{jhNfd6mP1eaV7~x z>%Rp`g?px>yx&4TJbT{2yyWpPo`~2&6hg}-XU^6uI8#0sBr2u=E_%m+=KX{&#li-R zP|@VfG77Lu1hF*I-3u)86rWh2@B|Uq{uTn@l1@%^3AJUzt5*~8$isn~olq2K3O=Dp z1Sh0Tj}2r5?<-n+iWWNR;^}^3nk}!snCaEkW)$1lIR4kb{Lhu{ku_K_pJ$iU8%beD z9T0PxL1kkNZC-h|M#!RU zsmPT$jMJB2DoWZJA=G*|9y=xIA(i<(C|qykzw(YK*7Y4qGjh>x_R{nt&Y_carRRjF zLe<{vu@2ZhE0nfeLt18awE+B`WI0BV?dqfP_gqkd4UgP~{lu8wzh<4yQ+1=Y!1nB` z1j&(2pzSM^Gm)n7RumNj z4y~9S{p+71B|)4)A)AQEV*G1iN5KZ^YZU{#H{A?R5%Lw1w`YX4WU@(Ts)Kn%>~THy zK{>ctQTg-nEY;}*@a(^TW%PB_+q0d~EQ`=NRN-f-S*4g>n`QC$qz#e>@k^#&wCW~u zueJC)qjc7z)nOe|5E$qHDf3hFO9QolMskfS^Bk3J`Ljvr*UXas-n1?5ub2gK2&=$i zJPe0mP4Qiwgza@cczD}EmtMU-3XgnWW9Y{3R0?wNJEiLFwo(8}C`ed5LxZ;6Uz9C% z){Zn-xC}I?TE2-4f0zfR*%K|~!K1dXbPVEgXKCAG$O25eT*^!H(2Ly09RX)$pK#B5 zvHz{Eb0y%)2x{Hg>E@sgLUA5tFi~SzU`A194{j%A@)JY0OI)Q2!4Ya{ST3LxW@=-w zmG;UxI*08UMGbYv8vDKWTO9o_r0XYKY<=NfYk>jZNA!>7&L6W~{LcvNN{lIs!Y-VExUcWJ5COYT@cHo}diJI&n+Cx7$u^djd{Ump6sphcO!OG#l zPi?Q{7GAy>X~xYbRz2*vMW|z;D{{RS9o9!Piut|6-=lM56nP}Kagot{6!{-BXgrfv91LkW$@Y-*N;FPThw`s%a z%1TxDg#PEpjpsX89`E@)iintsV-YZ@tJ5L7q>8M!+6uNicT{(|EEww0r$o6Lpg1P* z^0-!z?R>KkXYYYS~Hb1#!6#NQSL{?hfOY>P9 zr3S_6RofZTY6G8RK4XQp5~V1Nx11AjlZvk;H$s}~DW`>;deU(l-Su}Ns2F7^Bdh); zH~+%!4OQy4AIN3q_@1&gd0n0fof41@PPi&^;RwwOx{?L@*@7IU@&KAf9M=r6D_6PA z(B%i(|FhBe^`Iz~nhtPLpRxX0EwUqjZ=`g3*_y4%7b$d*<7GM^zi^Bk-=peA{%(3t zWtsS`DNU@%QEKj6_Q>gbn}ftl_;WLWPdLxWL+HS0o$W{&!@$W1Bt`D`7#?FsqHVvB z!AZWK%~LkTDCP&^l3YtQNN2YSFjU;>zJ{~hhcZ@+^-N!7hWy+VkW8(G`49Y z85Aj33N#|op-@FRM?LP7ckmK<`;^s?i(~+O?>^IU)9J-zrg_kcy^Rs3;wwOjd>ZTg-Mdlk)`Hi0|z>GAB zuG?=F1SIK2FNlm+Z8CcDC8 z21K(C4Naey86=V8xvvp<9s(V$tr^z=Ug@eQ1sH)Boe0X1AN=_1{y1r_Pw|!uh%+JN zV zAqR5n+L4g0=S{A!Oo;$L{-NKH86~ule^oP1>peJb{8Eaer!h4`$J}8nO&*s^9MR&^ zX1<)GDEJdttEv!D`$*}TA$RK8w)P}ar=k#}JXIvurw_-axE1G*kI})BBRU^@#+j!7 z`8gY^Pdp9a>Ct&(Tc1o!(n$kAT+UsCL!tA%h9F$UeiLzqZAVhMWkVwk+gBheaaW@{F%aF_u@-SnypfbVb(wa#9Gj| zIoOs$wzXlUQoH@_^Y@A9l}3vg_pnHy1qz_!%p<0mr+x7%}s}tvQ53bIerr z!q#R8wR`$04T(1330p{YD)aBjK<{CF0_g+Pfe)7_SrY_$4&HIi<#jc#AyO+^@*1Dn za5^T%zBQ1iF?!scKzVtmwu_qVI`*U4bp-k9L%&`2%L8EBmoV6@pSILelC#BrN9fVX z{`oa5CtP-^+txwq1H)U~;o`^Dk>`{j&;8QSE2BTiLel?u{|cPcq&cEgx7vJVdkIgz zj49+%$&k@z;`b9oPT!w@B3Ch-1k>5EG#KOm z@3oh?%tmX%U-V=^3z4gB0Jyc4c3on+=frhFeKrNkQ`o3H`t9*u=vVNQTyBgX{Dp$H z*SN_E{+emWjhsXBI=86yW5?cIN0Q@BjxND|1aP(7Fk#=cEkUB;5BdpmaQGoWXX^g> zpVbrAan9HO-t=Vuj4&0d{FKM|jS^~Kv`1Hfaaux``TtA+eV@(#$o1-`g!*WXC-0?k z+_*vtcqdr&CE{Qz?MHSGY3`|~3-vZ;D!8#FzNGAIH>(Cq$}~CkG$;5VVS)tCo(@I6 z5A)y`-}Eu473}qg%icnldYlS*ePlQch!;#QOAR!D3xZk?$i2Suo7=7wrF+t9M~hlk zJnLhh+(!}h#afvA&}yqr2L5sH)FMvN-~vcGrig$Y$wVf>PGDLx?w>^KryHq7+Knb> ztTGgD=cQ1Z2ipk3GCy6(G_Z!Ep(e+N%1mb#Vg94^2=q8fsJs$1(2y>q%tF+UckX(( zlODAy!~(ql)}_UN9g)dyFcb6!8Vy9q!6m1}0FTnoGa<-m7HaZw_5&SsD)7fzP5fja za?TX)%z$%VLx)Zi4Je}YAL$P^l|IfxCreE22fP(?kK<9*fj6x}V zSC0?uBzx32N*R6>Y+SF^`p=46HEth!lQv*O`3D%R6$zR^<3S~76O~US^G*>qq+@-) z&K087fc)id=nWNRT7t#glSx32A?@g@iIS`rAayYu1nFX}@X?=Ye-X3?-vkVTL7tsQ z2J6EYec{Ky0&ETq`M`E|1wf<+D9hM)8Q%sa>Y|Dv^C%X*u9GW5{*g6T@TR=*J}v%3p`9mFv0(hUmR4MaV@oi`@D4$BGMF@$g}a z0p6y27&nSM4al|ChxkpqnYJnUW3iJS zoJRsBhf@SGJ0;M`8Bq(`GCs4Y2W=NVzR*a}!<+K9ti=PNF4E8_87Qh<;@R=wViZ<2 zJAGeD$A$`9L*^Y!eOW;VdU{Ysd4}PXp|uZz_N;;id0hyN+~SKN?;fbUxLB1L@mJ3h zDShC=2$mqZZOx7W0bge*9GvSseZO1H;PL7g&~o{=aB7i&k=o1JbUboS;P$_NKAsaO zF(Q-aYx-7++j#lujC_U)Qfz|d8}t3K)2*}}%6+vT*9@hdkT;;j^&K5#A`{pkw znnUetYUmyvQQYDzyCG}pihN`7M3Q5RB-;J0U{AQ+2U#hKvWNnC37s?LXOOsW8chGH zLvjA^IwRELdGPIkuJ0pps<7eAN$)!{zN^XsvNSHX42M?+=|gOL;fiUjXhQ0XEW|G-%d9C`9%LffHFd@ejYpQh7>efTc~Zqw zWXKdjezlL~;#XAwkJh+s{YKqN5u&8bmAHw7pne1^C_^*K18N_h|DSRff^wI85vO9( z+s{CHw;f4fQ?C1V6~=bj8WsZuv&nl{ktL4sTX^}C%`_zK_7e{G91Cz-42Kh3!J93YZ=7atiQ4gy)ju(9q-?jYrUp5UpN%2)OhkI*#!VfeWfhU_}VU*-lQ4r4Oi8JqAwl zxhcx_{7ni!D9W%_LZk7)6qnt-y&hmk4rMioRI6Yhm>0YsB7I{~n=0W0sSy4f{Umg1 ztX-4qrRF_!_4FLu8_@g>TjL3~F^i7%H&}$-i>m#{%~#7qCA*ZwO=0a4e23^Gxzg%s zr=^odA>?)FKK7|(BRH}-*cNt}kEyG^X!3mh`>7|rcEj!fsF%J2PJd4p6c}5aS1Wk# zzK<4(bYt4aND|GB&jd=nKQITpjRhTVYCl5MQ66G34{QLk5Xv{{1bcy7$(Rb%vHtps zbeHyR@Sm$>p;^>}nSoo&LiA18lGgk^Ea{$*d5XwM`Kg^^+t32QdbJizMw8BEy|_Yr ztNTR%iZuY`AWW84Z=Og(7IACww5>Z44703^#VxC{7Li(*H*Pg=q#Eu&vzCAu=oj~qn zx>j{rT7dK^F4NQzzBH$-FTCCUk&b^-($%Mmdxq2gK3tG4I}$46IwUb&<}6VG2dEWn zpDKu{jD2pT*YUaX-lKIy`3%0j3P?a*RsQQRcMz9y4{lFtASmo=YV*!Y=Jj)k({Thz~Y+t+gagP)UVpEA}3jxrOotKeMA$4IKLwi zi~*XaC_l`_`Oz;mhcEcWRhIHK#KS|3h_VC ztuh~>KXg-+8GbX!PE);`iP1ngq`bL8QYvS`bFQS zRU_aeK_XPV{iVeRT7Q@Ir9`U%=A`3w*V~jJ_7)<{on)}(JNor+W{ADtTHUv}jC_cV z1ACjI9^Rh@`u?4bgX;BB@~84z+N9JC3aeQZE^!2NzSAirX$<{|gK~dyn7EwO_PrYF z=>H(mtC$CTXHIl&5hdZ-#38;SNdMj2GrxSgB^?9s7=ECbI{4AgPuM}gP=ZUIGq?ubs;gf)z;vU!Toj-_WHGng2 z1hdYScKJpxaoO})lAb0d<7paXk6V4iAyx**n@c!Jsoy|a=fy?Venfm6_1VehGYXA1 zYc7;*1;;-^K9ub_s?;glPNMm*Ds|jn2tmd07o&FHCJY-|aEL_C77_+F7=IAoYLQNM zFBLCCkX~p0U;=lUG~K0%=raUFIy5G$z~Qz-b$_UT(pa&7d~Y~u z{!XixeTn2HAwsY%-ZZBE+sn}U`MUiWG6(1SgI^Z{rrtgZ+>zjKLs%`PrCDNe z!M`qA5AC{suPMcfj+QZ9i{3N~>gQ}fOC!T~=e7F~gcr_s*AGSz>JM38;a#PKGmZim ztYsR*J-MpIe^pfPNG(&Csu9Z3JhUg5i6}0mt$!GU(rPd|i^sgH#=i{W0_-gXC0%C+%0Tz=r+;{C1GmGr+}mK9T-Y;GHJoV)$*Zc_@?LI{%Z99OLyp(Gcd6s9V44W@)z1CM9h4a!l6D!y zNJd<^(r2**U%wh%Ra3lWC14GoL4++bie+dcf*y!BUI+B!WIOtVeE8%0Fqm-x+yKX1 zjqpFg=mj>ZRHMlHLPWH8N5~NFwtB2^CfPyZ`XZ+a?c4rff6Nxv_og)#lSJ1*S|N|a zaFl@9HL}C;>#NKGwDt|b;AgM*yeWW-M?|IdC$*9n9a?B=v=st9c z*yn}5!=v^Sspi4!}|IlJvXOjxfj(qdL6GZMk!~5K(YpID?lD zi(!`X-*Ka$l4)r>sE-S`G(b6kMU5VKX9s3GoloAj`ov{Bi3q#TBmX=B+CnY@Fh=zN zest!zHpniJGw!5C#`W<{xkKm?{K|0;2+hNDntxaDD%AS#mT2_}a$fm8iEemy>X^0(g1J7qX~razxf`{I>?2%@c3*7{3NYo0aH$i&o%)?b1_d50@9W(fGB6( z?_03|ZWxNb04{mC_VsTB+Vi4&9O7IsdD05PYu7?scb<+)f~sCqRg9umuYphFf4bR5 zi^Ove|98k;?_2Ry68G6tj2o`d<;e7XshTSjGrTbB(JsHCR_yK{BWEM9Qs`%B88 zLuT{s_d~dmZDPBWX4R-8wkCmggX$h!7VycqxJ=05e{Ok|ovK7k(jTR9Q1vq4IkdEV|*3dxpK_ zDELaLN($)huqI-2H0(hMmr{ZQEwa|vhI3o#kn}dzF9)(EU7L#aq^TEZ$mAkE@0TFp z4Y$EX`Er`~928p*Jv`!%EBko2-Td=I*5Z=(dXn#P|A77t4 z)oK2mb<#I?%7-S4$XJttVRBx)EML!IEz#tTyVnXPEWTEgY)S7>$xyOhdu+?V2!+kj6NK@@(#&)fF{ zw|gO9G~1haJ1)V?ez0UmlL|y7!`rh7fa%b3B5RMJOq(0T<(GVs9Kunmbo}%`Ja>sw z_$;7Tz-Ab~b~Y32<;%;8lrc2Q2B-V zcZ%{D2)jqm<^`5OqlcuzcIe|z32*tis|PIA!X&oy4BIH`ArxhlW_{)evQYRf@HJpH zlTRS^=bIDB9NqhuqrLH`5ES9}-@y`FsK?t`fao;nG{Jah_y;*9<)4xA#OSA997U-J z0!DGcV0*MHb!iSH8;XTecy{F+L<3SR&(Fh6oTXzOGpO;dlk5G6RPZ=bb_(Cdp3>&O#OkH}1ecp9KBBSx4OLeJH= zjWTFKMcv*(QJ2#QHX5IRt(XYe_T#q^4390bHnYJFPOK$2a08m$ep=j+s{-BxiZM1* z@(&|pFGFb*5;p}gi?G6C1fc)z=G>oe(eR-pIHnmQ&jg_TrC*|?3mQ$+hF!$I1fdZs zOHyOH<>718q#J?%LLIJ8^G6Zi5%wYjg3$0s2*Xw%-KLA?6NJ0rPnA~kG`aHj^FIG4 zohIKVD6jFR6Q*7~9Jbug)6bMuJY2rk@ovw+Ud}cbyGQ%g0VCIxtS~!YB}JZgw0XYU z8b0>uKBieAXYcHA%n`qzsmQ!t1fJcp^by+CLnS%~FG09GT)eM6e&pQcJ}=oCZ+^5P z{M4*Dya^YR#C40HHjwC(uc_aCvHTcTmtEhiJa44{&vy5RrvDrJOkxe|fQ!&Q4E34D z?M32dqf|jfJAP*Q^sc)Bq1E?dPyXMtlEN?AB6r?7-y+=RT&`fzCS1Kx=egbd?EFIN zPUIwai&dH45jyI2BDcYIo#F(7?m!R3*^*44|L9@Zc zNxK8^?y#|9OSrX%<(OV6b(7PWIB;X92DWI)zKhmQWI$PS!e56!mdX+D@4EM9ab z>UT4B!h1Zg$`VsIgX^~PpO9;?@zPo4dv6FAzn(0@(L$Fv6~`az{M4@_ zW}U6@Q+qLI=h_(#sI8u&Nd!JY*C_oNPcLtZ-fW$``eBQ!gW3#XoN=uHPQaxYyy^(= z88o9@IjMe5AZ@?@kyfy8<@bkb&II2(Kb<2pKltU7OVk^}Mz#)ilKlI1xv;ulgqWLd z7LjW(6BSdQXsko*%f7nVa&Nhe9K!j-_jT33Q-X|7mg!z;zQWxs??1)2Bns?HMk`(} zX9ewceA;mi(jZju4kA!q{fc2KW#UIa(Vv*IT&;=J?Nv z#ON3khyry#o}WMCza7`*%;sC2Uv(fEzp_?dH%qS5^xsb9-0AQzS{&F>_Q(eVul||G zub!OaO!+eSg*pDi^g(c|;@QDn&iTmy$+Qn7flT`ZGUs@I_=_G^laI6H-E-|W$%xb{ zf%d(w#ou#YJ_nX%H9Rl!u0d<>xAZp`&WT?nKb<>|V`Z;z{1raNIvSO>P84#z9)S;c zdKXFEe2|n*$hKP3x%z;kS^Kfkv^nZt+!u==>i_ArJ5RIUC%LZsX>y$@*%4Z82D+%1 z|4FvDdz~o%KYH!aJ%bPNk#^^-9PQ3_g*y9Z{%yYvHl}txDqRfb{}X6iCA0%Rev)9)UL+VzX)Ptvz~V_U~|jcNlKkb;IF~mB#ZeHX9_4zrnLM4>bt% z$%q_+s6^(QF2d49VDD}^QbU$()0=H_Bt(6Y9)2DDFP8A&G6{JgPrl5tqu6RfKDT!H zX$LWtR&08=+xpJU`DRhmWy|yd-TzZ?Ur3JoeYF1apYsuVSVJl|mjL^dh#94BUZr6P zn~0cDs5~aLZl&ci1J6HF{0%*Vqx+n+)MKy<{H2s|#jSzUk3@_SG>r(h z&&ZPHAzJg`n-nj>s4+)HJ9ISKWzf=H*bl=_K^fz{t5Au;bY}& zCMo+n&Vn8~de8TU{P`?$953%7BxQ2acYxpTdyc04Uq{jksnpF2w3jri{tV`S(WR6a zD+B+F$M6yjhwaQ8q2nWokGkPA8of`ap-Rj0@vHCtSn^SxX-!3f$?I!NUA)mj=W{xOUN+{O z|4;HfB|IO=*wNwHG8p7)9p%8_P5FC`r4V%U3ryMsHy>8LZ@qc@{}V@Zu;bJP3K89e zO@j!j6h!Okhlb=CbcFk6kWd5dHMwRt$q(PMntLN(>wEYZL`(bm?=7j4H3-I$dXmtV z`qK;uE$^0bJ(~TVRr|;3(YnKO^QhwM?g;NRv${AU_iVHydr%=_?>&2vn(O@I{m206 z7@bkPt74&Q1CF42DdE{URUpoh_dn#nX|u!qfBNWI?=*(WdZ4-X3G5&pTv?|qhS6^H zCB@=Z7p3%}h*v5Z2k6W)VghyZ7UFl?IKdJuVCDex=Xh6YwdwMd-L3&!o0#^X)4J&| z-L(S$+Vp?UzzK2mD_SN-nK56JyxqKJexrwVxw%^-&S^EeRV>O=s4d*8WIVwm_*Tw- zbWElV8k~Oa^ z`jkhRk$^yHYH(@X?EGy-je)G$6B+_f5|GkA5iKg$z~uNHxnR8*9-aErIQe27B2(I$5^@dmYkuB4XyWE&#@N7J18i=r(>qn#=__BC0=nT= ze7uKQ+gO`iiKJQY6T_*G6Wc@2GE1SQa}g&o*XovRy5T8oUj&3HM45{p7IcU zVC3Dno~3}r^ko+N3jts%wd~kp;66oBwgnl3aTc5Oq?PR%w1?g}@b+lfKZ<@Vt2E@) zQY-2w132BNPwpvwx)oc%qJ4J=!7IDa(p81Y?7l&9agmn>h12IVVskC!^QaI-U&v$- zv!eJV*8q^IH;CVckF_ca-s4C5Bd`fi( zRMf)}Qf*>PDTw{9_B$9?6F!~$js)W0-NeZl%&*NwLA6dom$9cu7?3~x~$O zp>U7#N|42TwEz-?93P>Bu+>sIrb8W}aWdgn@+-d>w~nH}i3HDn_ibvKAe=n?v1a!( z*mPyp({Rd^1AI8c5^FCo)_klJ2z-|(^@;Sm2=2b*n_wUp6@E}tx7s4lz(omX!GZki zJlf8$>r3(-cQAmTR-ttx0@aO@$mRoP36_U7-bZLok4F0fl4lXi#S{9_gE##Il4$Fi zw#a!O%3^F9Zv(Jr9$)@e2yDHQ`(=GQ!&!)VV=p_xS%S{kY;=XQICU={P-FO{2}O?T zaR{ApZ_Cub`ng-GxEsLop6}6e%QMs|0c%43B8j6pn+&I3D^-6E1m~P@G^4GVW5X#f z02TQvOsWc_eQq~?2Pd4iS&|S3Fv1uD#y2>py}29*a?~X59x)6p@$ueh~z4dwrMP ztafo&-IX)h*-l$&CphkhQ)2wrLxmj)&v;V+2L``=6gQ|=2qqORcjxb7aQIpcm=>XO zQY<*q&l_ze(S!!NjKGR%mxsTPT2_s3Nut$!bqJ$KK_2BJYU`_?;$J&sFnqz`&VZhE zJ>!WuEseeNOx0}pYOpT1uLs&nUs%?seipC(*?NB007l(>Sblx<4yO4~+1Z3}TN=|m z{`T?g(w_yAVmM3v6>j}o2`%!ZXBx)9#qXqU18%0^+&UQ3HcL{6;{HOzdxcg6qXX~H zb_ZZ^xi|$Z=dEN|c%udg<~ev7~m^0(!*ySV zr{I*MS3H+>4y|!El1rpZsQI&{RF?{^eS9=VsEwd$_J)`QfM{ zuv|aB|5H5h-Ki#D(;EjpOtV`5HN;Nex3H5Z9EU?LhyCb%D|o9!ji=v{(4xdyp+G^o z7zsj?_u=8Iw@*fW*LXY)on9VjH-pp2iSO-}ApzYn{Yctl&kqw6?p5@d9A&IEwVW_3 z*e%kEH73{!YyH4D_P#)f#v%+9QeGOR9#S^}D#@vL!j~6KtRRN?deDej9N(xeeQeKK zmn)7dB?;@7{!GV68G!Ud~pMvcDI6N z#iltgHD_uULjn7Vf%#6W4;S4jRrLhC>xqu}np-Lq53M^H$9}az%_2kY^LrJZ{b`6C zNc`oTi~c3@a>}*q6mRgH>#(522+}@^0dE7R*qG_TK2Vuf?$51Q2Hk*$^Z)>NExI(eK?W=7)AT5_WJ2R7wOiJoJ$ z^Oakk3{M`79MP41OE^5d(J@c8+}7#N4w*9AthR`ZznFA@<~1$uHwX3a@J82x_i&4+ zSoepPXOT`V@*>kDD$jEJqf#zf$E}IkhW4v;xAQgTXjZQLiw74(1KatG4&n>f-&Y5h zYCd^b`^dmpj*$1Dw&o%=2K5>wK7MlH6B8QdzVJIa~+{XyF!!;6c<{#o<2k_>*yk7!W zDz}oejE>wl96OT#o|Xpr0sHa?UGQMT@oQd}NP{PIDANB{54YOQ4U&&}haLF#pt8mP&>~|;SAG5bTS#@J3ARi5ATHy*3Hh!d z(qJIj!C=_1>4F~gZygyQZlgm$_H8Q6Lkj4i0LQZvI+u4^udOTj0wb2+;ke*%)!V&b zsYY@xqQZ9vuDOFJ(fW}$u!DI?Yk5XI{b}b6D8~E7mz|gM5HtC>{*pxx)VnDk+4^8P z!k;IeF&9A`E}j=%B>eU4*@JSesPDwq#vgr5`^_a;Oi*TQn#=GBfABo`ATdP<<X z`P+OLSNcm7LEYX1`9VWZan=3+s9gWX?WYJ)8xiI3>{oqbeYlpca4VG1E$CNlQs_{# zp3b$;j#7DTR8;lNhV%pKyRl1E&|#F@aLcp&7Q)T{g zhI-MgDqps3(#3?nSn0?hQ7;m2M|fiqo(gi|xZ+2+f@a8Q;yD?NCjIDF3ZPO6yU@~q zP-aR0+OLHq$U-?m*p7lH`?b%VLxE2Y1Ev0{=S(}zx>md(_Qs`_Y$z`#M`-S6zIk4S z_5`%MIgXNTgw4QfM%$>K=i?vrfBdJC(munHG^S7T`TS-yGWLNfL ztrJ0Ti;2s}k{O5mQ0u3y)(fQTuBn`$f3RoD*&NqZdo}t6@c?Jz0pVWkl122UrMi{*;7d_}bJY+v**|o8^TGC~?Sa+I=AhUY&5ciYFdsu)MJ?abXg*5I2m`Ty;D-^)k=!-5{bBAq7@YP^kD9%s&&JR! zio00)kaSemPL6r*4sXG`>^X%MhIX#BFn}t0xvv(weo(&si9NOmK}jIP``3qKFh;j0 zz9j-q>MCEf_7LZNsxxPGegt+yrcluC*|jpP3Z#AZ^sNKz1KJICAuy{BHKVh^OH>8O zyb~cIW|NY7-9$&p$%8E5F3YZF{}%12r#`eWIE;My z2s-FQyV%N>@>Z7DwtS}!ypbv^XgGRL@?D=IX@o;GI4V!E7CghBld)-erap`Xp<84* z1sd_->z7xG$0w<%FN~UUgw>#2o0Gc&3tfb%=6$|x9U;6CmzqfVhgoBFor(A zd7v_a7_2B9C&ZEAAW3(kEPPqGa#_uM1{XKp zqZk#PZN$~edOEmcgy*RD!X^nz^)B0Fic3h;l>ZCx2Teph$MNtJa+fJF5?W@u3DlT) z=(yMg!sv-6#Sx6{>aWq~;Z|pF9N;}3wz1Lzq)*a8H;>S_n-sg&>Wz+OQi3j@q15gX zPRZdK!*`);<>iuYgg{!ZN#;+p_hNg0KmhkiQTY+iuWUVj-T!+BCQwRE@};6w0P# z@>9TFusWUy*k-N{LcQO}@J7LFv-XxJkL;A#rRGsZF`i$s(>nG7f15COg>1*h{# zGsRB{{7rf(=?x)PYLIpX&KWK8jo1pB3i%ISh=>=0P4y5~s0_MA)wT=1=Gi@)gbsRe z#O(L-zM-@7gq_p}CteS9M5dHaTt5egd+8+068Y1t6-@v+}p!&vC7#P7v-wk`R>**)@ENaA4e~U&Yk$J{1DJh)$#&A@cZ)ORXaKbLBi0k87}q){l@?ERN6^ zQ3V>jR(`5O#$Sn+8z7R6MK2`E?>y$_FxwB@)u^E^I)ZRUaBs|AJ$8Wev4)vpX&vmU{NYz+H|lsTU+ z3FMBf!B6h6-KAnIE5lzzgR-tF!G=~iSgd6mI6${#y5*FDsH-_V%No85eyHQ3y=-+yj`ia4k6=THMYqEGU2KLEopTB zgec5aB1qWLTFDm*XSy_dk(gHNzTn%2e{?POXrAHbtyGfm88J6z-wm~yxU21&Zcg$l zC;n{jUEeE8#KN2E@&$(@4T!pJEvHk=R_T&qL&M&l^nvF@Cnb>YnFw?$x<)#5Kri?e zh6=-`HYq33SHy7`gtYi$WHlLncchYh4ilVuY`W;QKc+qM|2e4JDmio7>XZcXQ5ZWR z3=)}1oQ6$cDzxDfcL$M=r-#_@1w?U?m6g81q>6;)H0PRkpS~GfV*4Sd8}*j%{l81U znQiE}vRms5k!=wux+5`CCAXouu*RbWtN1LNVeYW8#X0clrjgj5n=mTX$JXs5ebXh2 zrwWy23~qPCe;qB*5p@Iur(kI=yt=b#j#rf{o&~nl#;ljqrIKn{wT-O|-6TF5skxDc zy|uQXIUTRti;a@Huw7T8Vt9de`g&a_ z;5HeS)3LAUCBlUkQCFT3nuQ>}YQ|Ooke9BPKaPbRaeSBWhBwBKjdUE8ck~4Bs?hGM zwgkHLM0O!J$R)Q-j@L<)%-`(g4#OO~aF{HyCHv;0yyUro61wZ$ZFvskB3H2;iSZwv zqwG1Btn`$U7o#m%+#XhWAz5iJxsmPg$H~?q$nZvk?_9aamTORw#MbMO6X_k?VT&&` zxR;R*l_v_C-@uMktHD1()KFw+hwd{=!w0*sRPcI_snPUt|jmZyPwU%*kw9 z4|p=|@)!T`*tVX9I72c^bcqTpx z1|lcA!vOnDwuG+4cyu;P+yBj}U^#}zSEoAh(%>7=%h5P>WfW=AEG-DzrE{F=LV*Ts zv`YGcg%Oxyk!yf-7Zy1aQVR>AJ9pUgO2c%)YAsmzlq&7!)EHAWl6FJLlV#3a-CB#; z*N`RWNuV{AvZOealbIo`v=nA_FMf$lXM4Dqrje@H}9!8uAG`!Qqo3*2oVy!)L0z{0LlW~QYYX`RzrXcvjb&p!8+J;KGE z`6R26H87vO!E1(YP6mLwRfgQ$ylr{Vvq0AW7r2o2Dnv`klEQ|$q?^Ex`1ecX*oAF; z(R8@ON{GodueLWQr4nT*9h)JeFqL_*YHuGXH_$62J(5u~ycO(o$H+eyIk2PKrYE;F z%uf*GjzbjQ;B>r|P+&Q+O{<(W953QeuMai|SeW-HxwmM#Kx7Kg*&eJ~>R8V0t{8pk zH{|$Z5P6qkl=CXx!l@gMXEpf9@F-VwaL%LWD7$u4q%)ONZ~p6i8+Vw*KW=_#p1b7i zV$RN8Ti-wRJ^2QRg`iRgae~qO$vLnm2T6S_ej7|TTi21nZCJ&@l>@u}@hnWlhRzRr zY{7KF(Ih!?*ge8AzWIIt0qyi;(xU7x(=e&)i9M5@p{v}z%r^Fwf7fqj#%Sq%k8I*N zrHjs6YYMnm4|}0s9=yO)wZZrU#6{h@@PXY;r>jmU;tP>!xec+1n1iAl zBJJ>9y*2IaEm?hR>St|sX~$70R~QM~S@U>%%7wDosDe7E@GQMbIbZigf5;eS$MjoV z_tMpX1)@i6Ne(PSFTRdT%?-X`Jn@{L$Cb;+BJP-)o3|?Qj|m8zLO?iC!njFIY%#9oe`iaY=W8ZgSTv z$Pv6l42d6qeOKj5(|u zi=@i1i&P;E4z3N;uqxUS4GeXO7VjK0;!Kb+y)E{x0=KizwCh!867S z#A8jff(-kU^-?2yt-f@MMZ%NC+g72s>I@UH{ZcAz+$zFCpobb26P>Wfsn{_}3su&8-Ig5!Htb6mO)c3jB3^D#ZWE}hc=a${c| zR&1Y)xN1EzL`7v;&G;ja;I=|ZZ70(h3Pb4ZIeCdMdrtzk>|0Ai6+SchR0X6L-%~G& zZSC0`gk*$Q{#sr^SlB=BC;IKk6l9mWW41&Yx-h6`{zZa?Xb14LfQ&Gy{~)>hEOW`X z4@|a&fw`^!)TL0(J50MDLRJ`KukXZpG6gr^sS~ZYe&;TTQa5t9qEGlOo+fhry_OPl zf+&=eGsQXuBb1ocIl_L1vBHe)>CQG|s<|)$rmIkjrPle-$bQmx0c5f-xwN+J|Eqd(}LvmtxOMoLR*j-vh^MInr0?G9ajd1cZ^ln3$ z3@!}wj~`Gx++@YV6X+{cRZ^NSJVl6WnX$eR96s@7L!9!t)Fi3d?>w3 zySpIgN*tg|EUqW33xB$P9I{AwafN*mb&I<4$cv&@R)vEtM}EVrGjtldF*O9X9StXD zK2D|2Muj1DHqOEI?jn!W# zUBiZY-W%RYRCbe*GGCzMw}uq}8@X59-YE#3-oAF^W7!lQ?+ql&1rItRi<+A`%okg; zotBvHHLgR4S3q9jewh_c`y6NQb)MR)JPG5`8dCf~&VB|E2APbw{$#|dZeH3)nc_bh zA#{>bZnR(a5#4Hz;DMG?B53%4wq&S-kq)4HC!9HC?n7bVK^G;HZo-UN4;43|D)IF` z%cxd4MZ6{QEjY*?l7E%*R8Luv#Vz7u?{pi8pAB1KN@Q@sM1vJsQM;Pk1@n3yTn1F2 zTHh4`V8U`M*EM|?vBFgFH>>AR0x3mF3#8j;gYdNgRyx(|_y*#y zfQ(#!og9Dy+Io~t)!d$RHK^*gMUiidkLJLRY{t6lv|tbH*%E)~w1BRXv9m9U-c)<; zSE&?M5u8zdVC`r=-HjiN1FEX7*^Cu@ZI<0yUEXo_$=j)Mk3~{n$!8P^7@j)zH9&!ivNUYqSj|UTzr7`3 z8AyMrV|zF57*nuruKEl5dumhx1J_R2J1sq{x>vJZ?+v0R+6^tPj=+@~oG%`~pK<8W z4ozNe)}v$Cck>9FfYnC4s-%?l1E$Zt-t;TS>pml;8t8whE{|NqEIOt&L1y8M1B#^x^5}%#W|V z+Kx?}X`mPVQkgh_RIh8Pk~7>(PO#y0^Z>W+zpP{S?ETf3VV-v!=_o;2bh5g0CyyEa zWjV%N^@`hWucxI#!YwIw#>?t%`KydX<)6EhonMMq;<#Po!ZvNv&i8>Xm_d#0uec}k zdeQS??zWEA6+Win+^~SemyC@3LW`Ha!>#t|r<&zp4?Y4HsCyEPciKkdkE?+C%PrVl zVrzwW6_iv z2QwO3?l31n>HABn4oaD3m)?YSOAuhd9yk$JO4!*_V>UR|5DZ{{+czCORFshqUOJMNweO|@ zkzZyrCDqa2qjMKCD0@tw`6QTm=q;V>oZcAs2Hj#nvAFJ>`WwTFRZb#Gtm`63YnR^^ z1%>g%P{*uQ-ZHxJN?d>#zA^h0pB)1Co;b4aMU<$%* zjLn)^7Lb03wM}*0e_jr7hncWlk}?=#FH$Z67o52GvX@v}ZTt5t2zXX3Dp@~WNhV`= zyM$QWbW%hoq`BKtPECP&Pf|Krj#&FW)-x6g%4=v6wfZ3W8tL6A(%n-2`k3sU7i|Co zT(xkEBPiJ$;wFWUAJ2!T2T^ph#uu2M(VFz8#}?ziEqDGac@Y^F@v@zQvbt8PRw2Z+ zt1@RUyi4)9ejykt*G8+(!=>IDnTl_{;RuN79p?AQWT+b}G|*7wb2w-pXJdWruK}C4 zwuNzn=%(Ybh)v~9+E!2K`GSp#37FU807uHxp?dm?7m07;tPGL8?9ywE1TIdG2I4cV zwqbHN+(p1J)`>lfZt2SqN>w}wQk&9=%v3qXV4Ov_=XDI$*?;bF7x^nxLWW-ycN4uL zv&YsAf3d{*{&x@GtEq)_+0VDX2$NPAA1y<5^r8l3$Ws=w6z_JqlzRT6D?-ySG8WIHpL!Vdg6S=g2SLcwB?{xGOT|xk~gd_jrU@it96xW0Oer)1KufjmbugyrN^Dr zHDtfnKgcMpI9#%(QBS4N>XrICJzEX)yB%J{ zr#EiGB)I+6_xg4zFKSAQ*sNh&*!RS4f2S350y>bdhg2KfB%%fUFe;LjXg4Ndr}jjo z+nGT7fDjlx3H$f`{KvYcI);H0T=VWMEr~U3KSJ7wzA^%QqBs9+gi{K7k{u;qJ)o(n zpuC+j=r#>60eZ}KNUUZI80v_&pB+HZc9`zlg}T%pQ&IeRo&DYkU6@)t#F>Sx+)CqT zyGVwAJEqdu+ZT%%um(xMEQfH?=p-N;`lAxmJZZqmgI_oNx&n8bK9YIb8I(W@ocCb| zSlFIxPb?CZ2mQx@%HL8s2dbbpqmeC8RlNx9ll41IM09JGy-(V6YC+x~L%7i}mToZKSGrbP|{aD_cFtkQM16Mip!wmT}O~UQW7hgIlfKEDD zz7qzK>~nh)V!Mb~NbTiTT439s&(+@2EG^vaRsysjwu7m_w2BSxF6jKrUg11@Lxx&` zbg0GfxiHR*{_B&HyZ(QXJ#5xTUVOB9#){~Y=YQ(1Z6j)u_3)ucvU!9|t1k`vna%o4 zfYY(ICq48l7_ZUpc^Y=aU%u@Q5bg2*+Px(27GwAb6hl&tr@qQ?x4Dk$fP|fG8*3WT zuI&5QZdllv>W;(OZ#?|IFF|Gw`EKzx)>kJG0OOR%sY`%Fp=;|Wvq2YBfjsc$E*J&k z^N(f8f9Qw^SzOF8(6_5X8UKZWR6;v-v=9Hr{4ZiLF{NwHU^5ylN7}Plb1}#v;pW2% zru1F_t#Sa4oe^0}47LKp&sL*Kq)r`_EB-ex?XVga0?&o!g*+znw%h8NR`)lM_|9H2 zA}Vq$v4dOXLxlBe7q^N`Iom}J2=8X}XK))a^|fdld#K=8g`7pJSZ-kkFmf=89uejG z=t{>OnpNVNyT$xPLK?+BfH`DR4xF-I+^7f3id2KrDUb6}_A^XVzOMakuyM1Td{o&*8Pv1pq-leE~AOHRD2t7VIBayWCr-i4=kx zD8C7*?<@-pGgg2(L)!@oo?lQu8;fcWWeRBH$fQ%Z;Jy!%3(`-q<-{JiU>ccp>HM*8 zAvrc{hm{LwOzD4)p%S#~Rxqr5i3}DtU0{GF?slnqnvFG|PCY6h`>8gaB1?`=-TsJu zlEO^tc3P#l8<0|qP_;Lt2O;e%Yjq5>s|*d?x^ogZ2`)`}x55b-U4>?d7{zzOejCvm z)w>+yduI@Fnr3B5?}b*7YXRIMjc8MGq^q>_&&(tE7=BsN@*EE5HQ4hu z^|Bqk*F_1p_O~w$vLxL8{YdO!vmRrW9xBHE6#K~N!Bq2i2|oVtK6Q&*g1%bw zizn3q+=P_gx-i{1^6nY4cj%LlK$WaSwq~oBrED+ zEeg4^iGw zMCRod8uzo}24gPVFNG(QZg!dpY!-6tB4FR!GY{DM$S~0UPFDXHMQV;S+N&p#>f^hg zFk0N2H;PE9I`?yVnr$u#qTv33XvXuRQqN%;k?^ag(qr z*&`TER_Jl7JfQ8_o9Vd5LKHu1mK=HG+3G!l`;St!1PLT|$p6MYmbvoMlVH4U_!z@5 z;Ik|89gYeC!mI;tciX4>Kul9zO30BjJWD{OAea+doYrtB(#^p5A46(xN}^lXx0;0I zw0hiv9v7;NvSAohoNpuhCSJHWom=>n|u?XRj#-1IJ{S2$tkv>0|mU0 zAWH#qvYZc83X^7aLVzGyyYu`vMk;dZ^(#7cq)9qZZOJUID4i*_K+rG+zMBNB#kpZ81CfXd!r)lQD8!^RrPcJm{{R%LWRT`UN>tB;pJ zqTZO9@W{LUt}biGA8YdG*$Zp(&o$C|Kf8NU-`A9e+#~#9YkWlCzF8I*Kxa=>?t-*a zl6h`j9uD!KMqclU;%>VIZGJ7L&AhSR-2MX`tg6JfRuCU))E}W-?jLFBpV5r@H=))5 z-sm1jsCBnh-p`Suq&o+AB0c}nK^eJU@7?!NpaexuwWAfE4PhkDy||Ez@O#%`y6j3C z>*k|biQUCb-5>jVH1>MW#wb0S)FDgZZ-X7UUor;O>zIqv8<{^)`LPXgOT!7(ZXEPY zr|;ddw(*dL?pqqMQ!}v$PT*dTMEpOVt~;Kp|Nkc?d+&8=kWI4XMnp#TmYHy^vgPJV z_N-*fwYSJj#M%U_Ew<<5ttagWdR|Qq- zQ)XF6c!(WntiFCuv+6Hi6?Rx@?2lfuPKp{pJFTQ`yvU|JLiqi;P+$yf-Vs>*YO>h( zo@Y+`#@(NH=8h3r@8o+=8{{Ev)T0QpX80Nlnc@Xg_ZFvJEpBedP2$z>75@#O*>Ohr z%L)7!a6IrOWUkRKC#q3ZnXFRD?h=jgGvLk|BUA=R3Y{ZE@z=P5qjrCD99rpf@5 zi#A1mt8-+efYlFGa0=aTwwpliHnXygV(YcM$k<$IvB_?Jq~HR4>0{=o&%-{QsH_}9u8_VX{}NZx;XJ}XTkEXyN$oN01SSPY#> z=B*LzlEctz#%yp_va!Q8yLh%DKQ#jnX3@Kr62?PkTi50(WZr{c>sUFO>I4uo_?pXB zG#ukt6j6ANfLZR`XS)4X0C&q%7z4ny>1fD14?pSP(8xqGbE2^y`4@tG}iV=H9mf+DX-IAA@m#U zZ`RC`RBea7RbYD^*3Wi)`Tq|*w+bYdDPERtr4ee&V8nA?_mV+DN_fkCyH1Z~)qk)O;kRJ`rT*6J(u%vJhaesTCeZgV3BZ7K#n2h*V_ z+Qpr~XC}t>ocMc}0A-B((DrL+E=)fvg=W_Qb|S$*U$WrTp9KgB4OvNu%G2m2PCxMu zYMU8--htMJ)m2sK_{{11Qj389o4trO-(aum6@#y*l*=n~8j*w~2N;yp#sA1v z!J32cdW+%~>}J_?g@CXpkAj1vwBMZt*wEmJ2(g-zzpW0JuIEd88%D5TloxM(FDF*s zU{fdT5oW$Fw1|*qF1Lzf5qmk zu8OnM`hM@YjHhV)-gi1xq&ckCCDW#!wCe}z^$6~hau|aD^9T^O>>cV$m9?CI$l;W+ zLv<`mVOXZv;)JbPxtdw7*vy|$zd#U2bqpN=T~Ud|foDjYVS=DSLI z#iL;O7)STkl(M_%v>;W|Ma`vg`?6SV2U_%8Q|9XlYdSCI6>2`xy|fRlZVfTIas8Fs z=#S(2cIRpcN~LCv%;}1=tb?Z4-7VJRe{fME$ zd7T=E3x6BQ7O(P2tVYM-8s7qZ!h-HJY0qKEBH3;do!R()t+9}>RcGb+raTsA%4(%F z`{(KWM5hR>>7QN|=e;V=)$+?g%)p~&XEQ#KdQItl3EGmvyIq`rr6_L_@uQTgsETd@ z_Q8t%sx53PsL2VoWdG3n9tM$cRD+LCMmOZ%#Vb}B>eEBn8!A-jA>))XA3Y!n;{8<`CBv2q-b!P z_t?oLUiPX!;6ke;EfoP%~;qS&5G;V&IL_V~#>U^X`;h1%icBX|}eViv_S}=%v ze;V))(>ooEurE@f6#kfZO-P$MuRcT+;wx|$rmnXkW7zsB z98IO(K1sjGcu)!wB{HW)5^?D-Dfvrwvlrk8OHkL=$$?w!WU<0yjFq^O&AzCan`#!B zrrIb>YoYhAUSOKUBjZ>pzMLs^I zUsd6PBbE=p*m({uJ%x&@emDZLo2-r61kn$~v!Fux6%D#h)-+8Y$E*)g$mYEtrRY% z?z;_jUoeLrb`=EnqYm*Tetcuiz*a5Jr zrQ4QuE#AglGXNSVHyffA4N~@!7gs*un0$EXa$x%N3XIRmma0+HVRr5#P|oVq!~jzC zbsyr@a}G}VmYQFo={voc?rYB`WgU~!n9y~(D==x$mo?-LjV~w3ZLP6sp`7;K1K&bh zT^$aHM0ywNo=->F7s@1<3%%>Aqy^o@i45AC3SjgE83y{tok8#QN74Q^6}95CU5+pB zE;xSYQ!V%6{weg3C;E=7LTJ)NQ4TM727PH+pgIy#{0fKtxTQ=ipMI*B*GIcIRTLk= zf@D5!Qy&K23LmJ61kC^YIndO!3aRNHZq!iKGDwxU^qO~y%N}R!ZUZprw<=3>^yv9M zzi7mr_;fyi@1jb9Y4HFRgW;yCZ#AQRTx)a`d%RSGAj@5t`GcSfTCLpFV;kk!jWg0O z3t&o(3#l05Twm(pcd#9DOi>RS;bfj~+-9xK{9zG{8djKdiSe%vwnVm#jXieum$<+FCvTZ26{mx5e}Y z4AQ#lI1enFYb@-*Z1j744D%3dbA)L|D>fgoX;TlMOcd06;42<_KKr3u@fN@xs=9R zC8@tKZKn>bk<&q8JDud_fuxE&Gaci*FxGV>yAwCm@iN>y%b|6-3jg>mrlZ>O^4XZY zZ=+F;6}+JCdLZ_Kipw|jVwn(F64BSGoAD*`=9qklrR-^UHhQq#_3X*OOLqnk;BF!I zps+KwlnT3G`r(fKo?_-@%5`vl4-zV7D42W;G9_Di>_#rEBR6S+lyx(=s$Tm4@s!Ok zr4kv51~=x*wUrFUV;g)y{f8u+Ho|8RrqGW-eFc~Z^6 z0|b>_^(kXO`G8$Eoyb9-+K$<3P;EBaX0L~x4uVE;D!%#VK!C2W{VGDs&K&Ao87K_x z0Pe+`XbxnpyiNiYm?KXI&Z;X8;QyiEbak0H_PcX|47o3tmGB4+xO&vg-zP5ZUA->f zwu~v*cJ6^HuIMV8Q|Qr!&)qFEw3?m`pe#M~Nd7#m4;mkUM@*`>k)18*BQ-H>>NtiE&Pol9k-S3&QG zAMVr};MI>ev;iCQoq2!5{{jv?V)_faHhcXcw-#ZW6Z6cZGi+(r?U_T(I6gy#`iFYh zp8RyXQE(UIh=uVoT8ITrzg>YTfd&Q{BxJx_S>}0ItaJ1s;`8JXgPND$Ewr$)mN43j2ReDr8f3Y`v2p4~=ASv;ja z#7U&(nkl8_8irg@kL6S-txc-O9zSqRtH|R8QT!~V=9?zK7GxFJXwAd7z9Y_2_e1GA zfI!g~t;(>~Y7JkX`R4oMsAI|O|D2eqm>%s*wt6k2*iIhDz+nAzmoy3yDj!TbMolT3 zlctnFCf@VK?-8gxguu!*I$CmKlxX@iRR{!+IL^Bp2zs}tmJhenT*|5{f7tu}g99PH zhWX6`J0GfF0WG!9;U#vy8>piTKML|dr`ke=#$_hAau^AzH2)gw6VDQ?y6Ow01Zh04gbj^1S?8oO``gkm#?!6 z>Q`i>#f02%nkmTmo~A7|p8H;^RX8e)vbJN@N`i6?Y$_5D+eH-np#8v`sv9#v_%Y)+ zq9D)5`Qy6-?bflS?oRbrmgX1h8~0e^|szsOAWl4kjNF^DktLc|c; z0!=SqJ2{mVTM4FKcKZTNEjZ)=4bXTZgGcUMzJaS?CYdvFqKXXL>}-tX ziBuEEviC&pNmn-6Rnu3rd)9J@KBUTA;9rq~y3LOm1};5PU|lYxxT3^s8Ga>khIX&W zjr6_zcOfVe{Vs8EJK3XAUKIgsb%F4St}-ZW=c%}4AtLn*WA`?mhsrEm z0X0vpuv}EGpE9nQ<=cG)<+xkuhY->`jse`BWs#8N12lKd@b>dLQ_6@x$_P-&L6xgz z=I`~~UUJVky*pX{xGv}-X?)U@vRB#>_wVvRF#+mkIbC*or)>G(*Xa!;5X%NtV;=b>QXONtUiis7&cUt@Z5=4WCK}3Z_`0*G(c~Q)p(E^;7$v1TA_kqAe90) zYA78Fev_Hbo5e2J*N=TlPBL$^^LnpkuZKnrCZhvSFO1O&2vmd1#vKeABDCeQDmXiH zaKSpW@``(&?2UgI_%DG}^-PCd^)?`9v*o@N}=-s4TolVqT`MBUdb*G2}PD&5mDa{Txdj$#A% z19fp&wa>TeO9ek+?b!8DKCqXEX#Vx)`cOEuFRNZ)FuQ9_yRb8IiMMbpGTs7u}01 zjx;{4TJ72K#z}?tE1_{$U!2J3He10Jk5{6F^z5cNpMsjI#G=t z+yS{~uowhE-e7A#s^hQa>DU+O3z)MY;(nbKtd~|=;o(~&Vba#i7Qur44SR)YD+EqOmFN?Qu7XyJQg%% zaBuL+Cy1o=g(?a=TZ2@JnqS>@z;8glevE`{MzP$VG&LysN#ojTfR+UNw@*G8vQKP*|Zy|OiB0I(Twd-HMn;HaRwLu?B-XX1ZL%S&r3tH3F5948J)iN!~)y}I1K5%^=zy#NXt9K*R!WGt1~xk z#t?<3Dqy+MTEtSf2M#cpnb)dD=Gv@ZD_1O;X5;~+D^6~%t!H#5>Yme&kvomtK+N6; zk)g)8O#s8bI=PR$l`Bj!mmoB(c!+JDMh^7wz}{8gE<%5pd<zMk4>pZhVMv24zafh%D1feXloO3XCQr z)LC0wB*C}d)BKmfs<5hkeUA&2ED<|5#vwDwwGt+8Gy?Z~Px85E&X%)cTV5WFdo1F0 z7%qLx`!gDH^>a1?HEF#$Z)pt5p? zog`q%`^!Zap_3;hHmcgiTI)$!K{dzBfRdkyN(g ziRry@6rmr?euo}K1w__%q-h>GO_iV25z$o7!4GYf?6d1<4t)nMHiY9q%D(r|5p*rD z)A3;5Md@4T&ty}}_o{j(fKmk6TW_*1nz64@{7t@_`QlSN%8foJ_( z4nQ!b1qxfu8JoWj>!h#ym?cpBk>TwNrk*0{AMZzcKOu(X-o62odiarV*kkn*gvk7X zoPE7v{f1WFHtO_j%Sq9z3d9EBJ~idppR#E98Ci!61J7GsNu|GDYcrV0wW(Ll=)-LZ zAXg0DB~$`t!0Y$!jnRfy@U3*O^jrcg)0)V4rtW$^oHna{f|DZ?h2$;(9UnG&ALB^7 zMQkH61u{;`22OfMjf&UG98{c_>Dtee&-!Tr^@C^hxEuyh&D}{3bO{ldesy5HIbn(m zm_Gi#vH=H20J(w|HqGQ(54!b4raX>=q#0VR+vI1xYCN;7vnUz52}GoAn&=Ni=5>=k z)5VoLsRKNU9If@A`D`zJwE(uZA6|TXIb}gmZl1FYpv-TxlY4qgu;T!oN10AJ^Lph!<#&FkMmvC9%~FDz<%$lwtc;F^V3^VGY-8(>(6!b*+={3s}Dv}5un%; zs7RHxCIUL-XkpSAMiC$O`op>G$HA$u)Lp^f4D6 zVfFJ_ArbAQzLmZpE&hTJq~;!xy9J(;6|${C?QZ3zpokGt?R_V{`wnZM+5H7FI-;d0 z{{a1!Z5)(KKKHx=nx}p(5x8zj`M}NRQ`ksuJwo&~ULKUrR)|VJgE}3V)Nft1-D6DG zdTfxE*OgI}s_n}(k&C!f>!2UKjf&Rw-TH4zF@=}GK)1AwB3C~R!~V)#1c3)N(6|dF zhY0-{#!)~;>+D9+%n}0*onRA|3&^+o-2-nJ4NfLQ^9ojh&@Ll71^3zxk(Spuzqip8|9s_WWiS{=`FVE>^UQvGbE0kqF6++i=kU+v?*w|ClF z_Vs>)C3*jlCV4GXV)ZI-+dItsvo zN6Ix)DSrs49x1~FY&}gQ*NV(I$}|GTzfla!K!<&9hUd9857&~hOl0iC*743nW(tpBICUTyK}BY>VDzxPeBBtt7Z{r=$*Po^Hci69?+ zVYJVdVhn@7IG(>th}V|^3G8DvUXOL9!0P#M{}oUa%P0qgL+{)Mv8F3pTOoy2<#?U6 zU!Wa|?zF(ypawcY1tLuHmMzfwNkBlW1kq4FsSZZ`@N%~8UmAtNoj?dXkUNX1dZ~z{ z0$n01QU^>a5r(FU+i4*4!D$kIxI6_K6faNUpF_v-4gL0ejJy7LkbU%>@@V<2BRGf) zGddzdou!}$tqSs`g4umBmacQ>pQrt$CeN#iD(> zrKG90cXX!iEWNP5ns)K^A$-Oq!Sk?_fhRH9u|SvgUXi+0pw3c^gf+cnQ@!5kqx0kX zL;Lab_*t-k_0}n+0nJ+~!)@=LLQy}2!BnQ=x8a0509(kyp|^&?%ozHLaq-7L(1MV( z(LYiP0v=R(RbkFO|CTa|*1B5Xt)(9+KjV5UwSstRfAD$WA^}+Hg?*4H767{h(t^I! z^}zEGS4=&Yu_ksN*jkfQO`QisKb}m3l})haJofbTa(Ou;IIPs;#LO1|Lw zD9!GQ}2Nf3&lZ);KOO8zUFrNLWnlAo=ffW ztyY@7#u>j#ON*}N8NaGzX&Ed4x5<^3XD_)7}kZI6v;A3O2O-Cw@7(&^q4`u_)Z zbN9Xe?R`5S4$UDJ;WE1L_v!F|;7dOdpZb~Vwt`1+o0e>M$RCJ>C%5v^?@)2@Gf)C| z^2SypZ{Wi9twwGP66MvCJq3?4T8D^-0UwBt4<|6e@iI#OJ(UN>OtYL^4dVk`t38Nb zxQyr)s3J3d#d#g-+LJA@CEGwag~+aa8fcwLl?_cBzqKHLfM{T3%0^(E)~SW)lOLW#6!$f&_^q;E&AvE zT@>l8hZ&-Nx;(0ia8@c(P2LZhaN+I>ADoL-Lhg9*xHS@uC{g}j&Kqn7>yXnV zsUR2rEp?e)9la;l^U)3^O!CByLf-x>1?Nw-qfgd^d_teMU;q5@(eUf7RvgAtJXDN0&PkQ{UbPte!i|D~3MF?9`30}guMpvX zg}B4&N>z-Jqi&oGrtyDNhjT+F#VXL+=&QE)jI)nM^9xRaNgs`j`84L+`RFkR;lD97 z%`l`w2=}AWvUI^jHgJITMsjT(^X<=%UK%KI!^S1BFmTL+$|k?U;+E2w54W#^W%?)c zL0u8ofOSZw#C*aC2ye5?jhAuK&=rc0mw_mhvw;uI`dbWMXRQ>zw@`U{}I(0uva3mM1<|v#dVV|A z{MtTu1Ic_9gwP3>Xl?QLR|H1VmtlE&Z1UifT-=y`7m*bW4^|%+QlK@l>FeVI)8vs! zY?TJ6vT*2QxS5SHSTVcn&(}iy<#%5jl%P(yd4afr?^BdCU)xn3c=)}UGXYBD{DV@3 z8-PDxs*jUFy@2XOeP{=T#A%tZ^4PhG&p)Fj> z(qGUsDyRtENT^!*ASvD7LZp{$^Ozmry0Vl~z$O0AmnhJici|k_Hz_o==DjNZ;(naW z4I2uffo!`=OW~UsOU6D!cn>I=wFuAAC>?2U4B==4?g{CB6=lKQ+Bd>L96G{g$CjF^ z=!eDsJ^paw26J0Wg-1zmksi<@o7JmLhmRt&~G3iLdRov^0~mkclveAEg& zr^odc5YU@Fx{+~Tp6{?y3B+7GZC-x#1uharY~)$6%S8(j1+J)*qQUpVq!wWZ^g%Gm zeUIdZeax~j=sBlJiiS&2+1+1w^~A8dGnvaY93iVU7^l@Ys?M0`pQPhT);l|iGxrq9}LlbP595RL`E#H<6KE?B0?0JEQw zm-JK#`-*f1H~}IudNH@NhG_TWhJOr`_2b!9TVxS2H&C3^jgcD9{rWdr7s;hq5h1SzLr4D6Ul84;5w1bdgpV z;OK{9-*aS+9n?O1^Q37Q+PF*1P7I(3C=`9iaj$Cyyy}p}TD!WOPX1l@;o5qx9-rrO z?X*YZ_vi+00e0UxDL=7X$I4&Iq84f*oWfNp2GwRryDz18FPnTMcseWJ!Y4Jj?DuUf z4vAn7l8Nx`#7UaF`V~Yo?SzwmMDqMsD2T9?y83!Y32Ccxdl4Qsi2Ntd4fvF!zl*(3 zhO~q?GS>{weTsi%&kab@T^@DJ#}~-y#u>aE_i&zDyT(S7 zLKkt~A}LAn;)Ht`XppWn+@Y(BrC{p2-abo(gYV(s#PRQmVBJHmSDRAh6MV6LEN5*%?3}Zpw zva9lRY5?zpPdbCTLo|zegnsADTp~+ydQGNM(J6E@9=@~?Hy&sUZ+XZ0{M7&#RY#+X zMn03%Xey#hOM_=pJ6>i(X(eJFTnd4&@UX6gbarxascvEeSDK^BAQcOQ0b1Tvqyonf z{Osb#Js$o^i~p}Haa17iQK|!A!)n2d{sK#on_C?p#=^Uj;EQ_^s+V`}v4Nmv1uM`E z8e+-=P?XZiA%&;4Pt-H~chxQXE@b$NLcHDXRa{*;UhdQidhiy{&zKC8aqnq|{T6GJ znZ(Pm@hov=fL4X4Jkr}_;|MJy5dDmDN=%=JD;wz1s=_JszWHk0OCQ}>EYn^!zHin1 z6OK-d;hR9|k4PRW3`3cEOLkxI;}Ih7+=Ge+b>Az+tu5eUy$Mhith{m7)-~Ck zZ;7ut$!MC@*t=`HVnN6BSTb7c8P(J|gc9;)>*~s6vpA_4fq2XGR94ycBsvel@+bWI z-HGN!e+|+npASn|c~Nkbg>`w?M?#LqJpOA5y_0)@{z7H8_IhMvv0eYi5%}RW@tuY}FP;9Qz><;)(Cxl;+nMK1Tvy|AGof zk8FY>ZZs24Z=P|vHxWE977KujhoPgo6@7xhjz*c_L7rx==oXKnuZE8^A5s}~`Mx_h zzZkM)3hp`#kMugeZY-XZfTGoC{dLggH}sg|IdS}ikxCJ1N`kQ2ip9i$*_t13uL?Tb z(&Ez?D(iT|0~QLpXQ7I)h$eaPZcUPZ*CkKL4Y2F`o%IKzSB`)0{IAin$cT zJbd20Hg^$T@tUxO9n@`z{#W^uKNL=u2+9oLNRe<%F*{~OHQTNIFLJ%E`+2n#Sl%@_~<%dc<*Bgmr zY|m1)nWR+EjAai&`=Mpiv}f+LU6K|@FXPHxz*>YUQ&GoeiXP+{ma zL{@r&Vt>6`MY1Y1jkGJ2s~|@S9$9Q}VnWIQ^BTtFzH-3LS;SvjZpGlD^a9BE$##e& znX5CAi&3CM;Vflh2N;~G&4g`7)#nvc#%S6HZY23xiqH%A+;qDI5OQf%A~m~KDjSA> z?e8*2_@Yg{g3;>9#8nHemHrNh2iN$!(g`WK5F8L_sXmW%Q&CJ#=Tv#4NcR&%7j&3F z`~}nce%51wZ}`ErVcMVul&nV$=`OmyBP4|18UqrQ-YhW}4ug`V_toMDxU5Oc0q58hDL0P@Iz%EHG)G9AChQT5Bv@K7)O9cOow15<$mD^rYlZ3V+6?01jr@R}Wk zbI82B4ZLMm`#Oi+7~rK4A<}z}96XLZP@vFOb96p9LSEU)_Bc37?2dUe2!Dc5mqZZ~ za7-L7n=t_Y-)Kc-HT&fi(J_FBZTIDoRC2n8WEJtLyj}dzTs;@4y6v_d@%VsLdk+8H z4gU2_&<#R^WP9WDFW*tP^7&S93uAKn>=a|(D0aYHLPz4@%`J$ay;Zz%BC&84G6UH9 zT%vIb%W+P>g(SAjWj*UiE=93ofBe2t^1^Sa85q9TMl$rlYZ{Ws0?aIMOyD%7I?O~53qwL z9!1~{5$JJVyN(h9<4tTPvp?pfAI#NpUlvg4isqJ@mYj85F2ZJ*akLHXntDeyqbx4qdE7M|%nlF_A@f3qrBK8x+fW~` z-Yfl3d2n_WQp=v)U_|&d^!a&@fs`h%z2QRCE6xvW8o@?@$L*OiNsZIU z!Rx;kt!!vy3LWXMyk6}_fMUFMp8904*R2k1kmm=OBK8FYEn8!m8sz229t&NUGp(lS z&}U_jx!Ilm{vk7{tRJN>dVA;T-`1$h^xgM-qG(y)IDVnJDQ-%!qfN2HTs8;yMtOTo ziVR$cMD2%1LV)Ej5jH9+Djxfx4M$nqB7GZjN-4qf_qTEo?s3MstUV^BO`16fhDWml zTHWsmpJDsm7oBU~3-8XG7}@2WKlVJt32QGN@M*!{OTjbdcsS-c)Iay@arAhV8xdx2 zjmJC5IKPuN5*eIBcbTT&Cg-X?JV0g=KA*a|F6K1;lVdXaI2p=ToBt7NKBEz$-&cBz zrH3eC%8E0`hsT{)KNyYO$7eCx{}#XhJ^4G!TOt=Afg)`|QCVpw$cKD&KhO&!d#Z<} zEWqzLXSSU(|0O2UfHH0VEF^g$R^*OZw4qx#MECcCq9x5Q@dIY3^gxp&CQ^pGMj$bT zREM#av6<=##-Ts^aqq;p$K*`{AHDD{60QATIFDE5Z>$sl;6EJ6ouwJA3*KHikq5vW zD(}V33E31x9Yit5!XdGN1Bka~17$7}H&)hkj*9Gk9Ya8$MC5TbLw2?`HiEtE$cZuF zGEBp5eyhDfpq>D=%1e@=BOzxfiIii10Pf>k0>8EMiLw5izO-aCC)I;hD1?xu37XOP zTJTb4NcOElQ7$Gb{9ok&*wM`&5fpfYx-62Wqk+N;-;FL~DI4GllcV~>a-bk{$H0UK z8xjBMmWJF4hHe4=q$wL{zWXi!B*A^Z*rnNzcrxF7CDWznq~aCp6LEz8z5dFHe|eF$ zr!xQI{QoK0F*9GlMxZatlPUb(keL9E7YPl!#zNzWLT1rVmq}YxDfZtP+3*-YZRKtd zTzIX-p2jIhJpP1u$GPWA|`u~-cMdd z8uPxUTieBVih!?@ClgCjYns_l8~Pk{L6d6kuL(rMSo+92gK&SpYXO?@7cS{cc#M0J zR$DMnKbKWN@$LEpV|5nCxh(^uOTW~)|io(tF_Zc?E`7w0+_XkaH z_jAFHvDdizxn|q5Cw>bhG@!X}XEv~q1O5N|#swb)jX!Waw^l)&HcY;i%8emftEt5r zef8_n<-7yjsw$Jg^EhGX(p%aHeNdHwN(b4M-NrG0f>x$rKr*YwP;cmy5d57c&x2M? z-}SGT@!fB@@(FZKRccxsEwLRxfbggv4;jhwKZ6L-nbBcGZ*Z~g;rPjOqzwcsKd!;N zicy0X_^87B^pCo*^09H$Dsqe zj(g!xNTQQk)2A|h(7@m1IaW^SM+14suWm@hm5rFk?mct9e5U%!<)Gr^Q3CfQHo{ox zIu`+&p^r;$<|S0t@IMR&zUb3*tpzO*B41XD%soGlTTtIL-QC+k{ifAa2b0)3;Voh$ zAElN;4p7oo{tk+}fEuWjy-^C_PhJ#? zaf}^t8Wf)uBic7zG@)Hl(F>%r6|+Z&dY?Bu>h;d5j~@K^O+ zg%!2Uvg+?BC)M!;&lg$bZ<#8<>9G{e>>#O@%MBjm-xKKH?+>X(ek^J>&t+O9*W(Ue zp~l2_Rc6v8kjfUEr=dZvQ3&yv*I!WN7lj#1+W`EhW}2bO8PsUuN!YuHXRjkx#(s;# zDdJ^w5eLFPMX@~*5WGy}T{>os$eosq79uW}TL0)b%B=LpZT@qJKu-szd+jo-?^+vS z%qIT~Imtl1;rat1_xofG*$|eQMMjgJaOTzZR|+nuqVGQ^&zSifMj;6Nck88EUOYo< zNk{IV35qMDBw5yx^2*O?$YKwsX<4E^11e1RyR=9lJ zLn}eteF8NR+PXoz0>;s%({+q_L=8{bDO5?{amS;WC=L(Yj$Jo18H*=gvR7jRmZDw+ zC#^yx%Hu`eD6SZ!E9&9UlwA=!L@tON6$OxTmJXjU0mSb=xkeYf7)#_@qM^Yzgzi12 zf@w@~T*j{hCa)P+f^f*I4Il}Li%ubj0-sy2IC#NCvT#3Aeqzx5%8N#6Y~R0)%S*Re~;IXH>JAvm8O zpkH&(32d)G4&wY7H*csI6D;^N@~^@Jhld>DP(~aQSW}f^(KwQydxJZOr3HlGu&Soc zQ2~vamn^q&=z0Zs+zoA8gXio-$uGcSVxSXP#v&Po+CNZe-!{3lK6&?so|DCyGuNZ1)%7vntraj zXJtgF5pg@1^Yr9j*CD?y;GryYCmJz3^Y0X(@9y0V5hSrWe73xZbz&k=BklSsPp-$L zr*G-#5rIFXJ@*Cxjh^ao<~`g`2b2offuGJKJX})&SZkF^C-2sCgX@J>dQ74`k1+bx z?cCEIQKa?#ClLy5)gpXS=%^1!vjm}?YG<2Zv=Ha6#!2lJ0lxF?)=2%XF*1fOIgKFP7e=oHa$P?&0muhQuc!f}n?X2o|I@QSFaC0Ub?A@q zL^tIiRJ=6$E8_|nDsSTu{IRZL45tlAdoQ+EDlIqSs|L6t(T%414c+w*Vx_N)@BgI2 zMzDaHODyRa(S$?GZcfw@p*!$<-6sw54UbHzUX;JxL8TN(Fs$9{#39Xz?|2VhgJ8|D z1+5v!0g}YXeKrYUI9^!v1{lmAGp;sw(Vov~#@(lZsGF|-8okU@{bI5)Y7Mx=+Y^4X zfE5D-5_9Yd1GZP_+^-kj97P~$m~74CIBNCa_b7`2{Y|YB;PiGU`3{;?`%X2&g&gc= z92}Fb_HJ`!Wcl)`+T^wPN2+@a5a?(yC@!0@s|RcaF`joG1H>cfeQ6SYQ5XAZ@{mn26#-`rS?bvz-4vpa8CR&I!VQV?LOJCwVq8N4ne?EtyTlrbFkl@VG zBl`=?4a{v-*2B-Bs_@?%hxW$+*sFkI=y*tp?s@xiWSRpPje;S7s9SEn)I)sAN$x1J z$!qu3D2PRTsPeCSlf&e{uG$Oj>{$bUdSB5s;l`nJu0A7!s!TIm1TQe&w>r{dWC}J_ zOAsg@chc2!q+R4WLbaC4_7(x1@l6=6i9!CwM1W!O;i>Q#O0a4fxT*!FO(E(9Z~K6ePyKiG#!i0Y7Or6<`nry{abi0_6>eO*<{{fE@^a+Se$Oei zHGGEx@EB~=r&)j(kL83Z@?lLM3r;uNS25#~}#w;bIWR{dC zj+B&6Y-LB84 z+Ki$xMu4<7S{X0ewecW-3>L=EL1^)dQBba05;~IC5~P1H8(rxT!_~C?l8N4bDf3O+ z?Gm{+yaaxNJN-Q`1ssi(Q!>TXOU46K$;d}wQB?8qMRBmXK6f3Q7u@*CCT$8Kfv}+#u7oEdj5IJ2Z5;MWoQg z-QH1#3ZdWR>bh^IuR@)2{p+^9 ze4K&Hh%D(${oUK{=t@=ib6_hhYc)|_-1n^KpH!Fv>p=dkbzILjIbCz#rCby6*mBbb z1TnEbBh2ayZDza$hAkM#k41_#Wb9k((?GUC#M zLthpye#{zKI<9o##&zGthkm_+zJxEkN$F&?A-O4%0<#%U^H}5+?$*Z?W{*pNjh9B7 z96fvjuQ+|*?Kh@H0^H<`;%WyHX^+Spe1t``3X@t zi8v+e2`oS1X8*U$W`|0PidD4(@Y(@12cS2?w7}F>0!QVtY99MC(4&`O6=@mE1IPvp z_g2DUFEIFj4WaSMfY?{t3HHlmbCTG!eso2Mk>uN~j=+P2iW|O({=8#O)IC3dXLb&E zzxP7Nxiwd|zz~YBo4s=^*8fdLd%$_7oJzygm`jvN@HL2k->To}W1A$zZ@WXH_xI#B z%qWq&N)JZ^q`V9eESX)2tN9Xim~6;*GZZEASO20z(WFRbm9ErlR|# z<;`(r{M69?-uS9+r~Yr$Z$QCL-GwquzbJkwf(RZ$3L1tESD)|RE7hIldC29ri2F1& zdLZL35^!&3nmc;mbpu9gXjTfaxQDT^zPoi$jGaDgqffV7P<+KdKl8iASK@{OYhP&m zO`R~}(9_f=E=A$FnR5vFt?Nh(W*ozilHeR>emx2OhZ){bdqb4YD9mLn?uB5SQgb|u zKuFSRVx1RX<#hcw!k3~~&orjmC}q__fq_?wed^9smy9_^Vvz5>A8<{X5Gup|Rig0E zC{-(cWxFu$BLHc)OvqEXYa)yH$XAcv{~RED5|8-W zUjJ^WEF?!enQ?RLFlL{M$mRiFLpYN4)qv6twFcVlCjHD=nks~dIcRs*vmZALs)?hrxV9$M_AE>x{M zIY#_qZM@Vs<>2_-(W>Dt@1zK{(d>3sp)oR)-Due|7ooi+W$IE$Cz^QbX1l{cVgm!2 zleNzrHJ}KKX>dD2p+8WgtFe<86<7uTX`!hW#m;R!N%hrjf9oHZtz&ORxL(=3Vxf zVSmGep;~d$RN^*81I7sd=OvSu-XhtfR&vfYjKufjvE=<6gw)0mwcV?0U$O+1%r_de z>a)2Qx!NV(QLVh4juq;X4M`OSL2{#hVQwYBlw02XPp|%l%+XbfJ{)0#gDfFQ)lxI* zoj&tW_n5IfXdq1Ds`T}G-gkfADtehjuMp?{ER=j;CApPf8^y&LbL-;{Czaq;{Z0R- z7g^Fc4L+;aTr$3oxZuH}F88$X@ipHt{LOnbEN~`j(f4hSOg=kDBR(J8O8u2mX2dBH z&Ftg4S!}7^($YClK`ou@`%GUXY;9FHnLJo{e7)oIrU>a@Iu~i6 za>?J6H@dQ>SX<$4n61Xuf4?u#)pJu1BF44dLR3o^ksUP|!G~O)e zktOl%Ff?{wca@Uo1_znAt46kuH)q8e@6Exfd(uZ?^)x7|M+5d$Ad8;$hl6MCB|x&p zoK-d6o1JVr*ns>5Cgs1}=_8^L$5Os}f}l#8Xqjf#Bivy5OT-haElLJ`gN3qJZyV9O zL@v{G>dENJQWBfB-tS=IpxgAg{WmDZDvC z3Fa0XB5Z%zD>PV-|Dj&aUb$m^skoYn2kRUYA``Td9~ebMF!AZ5mkXj9tO@D z`@W3d;F$@ic#z;Ay?#ws2&HP(*dAYUgJ1fs!HEP1ow<;SP+iwD7|CQnbx5%*IAebe^e&NH~xn4>H# z+%TEsL!bf<^7cnM`?KlOe9l|G(uAMtC!vqOVr(bMM|pS@fN1G|_B5d^yz;#|xT~+5 zjincGOEqbXKk_KI833N_*aYr{mb+l?favFf*zq)p^;fH&w z0uLSqEdk8w;YQ>hGiofF`#)3egak9!oMh0AglF%95vll|NSl(ZhCNFaOsNE zEAdiTqBK}{AlXi#YJ8J2!G(oONPd(Ta(SggJ@|+yGjJ=_(4^U`L|Zkpv=w3#yG~>p zyV#(Jw7%5?2=k%W=A07a3pRB)3pszdSV!^QS=oI{%teWVm}|yVHA$v`F@?>!K^Qjp zz}r5d16zMJVEf;AEF+0AdbTPxqF6O-{kVxAlVp=g@qAKNENL+ojtJ+$35X1}P_H|< zAViIZ!Fsq-o12t0etT;Tz)MiOcg`@Xt&5G7V>lf{vFWw$U?c7Yjp_+50{#&TPXyLc z*vj&a#LE+E%-k(#d$eX1=($I*SML9LcgFX>nK`8$pk@v-MYoQHvgc2(c!O z>Q(LI3nyi-9HE8_1`%@xW~iU$&<|qC1W4+7lxoBr(s&Ed6YUe9r3~0l=_Z7Xp6c@3NVT zn=2P-PR{yl1fv)CIj{)ox5b}8fWxN)^tfaz+%c{G42?6i;J@|gHR0egEeO?~O+VHp?f=l>820R8Apor-LHqaN4j21EP zb^{RG-s7XkxwQ_zGQ2@R5|UK8yl{GQ#EhL-f%mq7_fb13(3_#kwxHCbm!2MJ6kP3Y zdWYnYIx|P+aT{D+kXoN@{(UMrwf-5L`Kgn88%}UiwuC`Tw+q6bu|$cYWKX^cNHK>h z%VJ?G%xr74AuU3J5}QDXb;`ihvx%4`#OKs78aIfU;{4nAja$5dBW^^t>rA^Wb??GJ z)kV+7rn5uQ!0$YLptCJIj|jCk-xWfi=c(|h%~KQ#3{?w;wIn~q!`aj0ckRyLFiLm3gWoHP+;e^3WKxr4<^6RXt_8vgFHh_ujz;o!lS6 zHgd48TK@`P_;OIMdvhUV$m!iTA_xmtBG`yJoND=;g~v4ch&7Pm%`N9x9S3d_OfUtj zL3Fw7ZICC1DBck;N8;v;>b8o4N3t#M35tLMjteZ%#SMPYQ8MV6CEc8BLpI#1HvvX8 z_d+s)j^$L96t$9io!CC>>y669&)a=)yJz%R#HGwm8#C`_${thSuC4Xu>ybA*Pzff0MwkNa_-(=`;#F|3erMN6d>x>eI(=n-HBnLd z`15S|zB&FRA28<|w_MCPq`1GR^%GWkNErr%>;zZ{!P^2pE*&&yeI?>NJruTJ0e!$4 z?&KPl0@tM!18(9WP z--um6mdKbcM&WC0fuJdx7vf?Az_;RhC-9<{fFR37N0QnFLIQiS*+gFG*bqa6)!}W|xLR>(oG@e{QRup@5twpnyRKSOz)*uw_7Y#egojhF@Th z;uZL?M{%=9m8~~8yRZr8TL3n}U=ekVK(FW_LIFHEb{(x@J(6^YQZOTv#0&`XB8@Rq~(cLlNdc{0+*=}&54gCD+iI0IS; z#KY~~AA(vbmmhCbjvqAKyt0g@6{#4`B*g9isMrI(z5{*`akl5mae^6n=PSbq!qrqj z;}zmgpGvRgVo(Jk2@7X^*yj;gcwKmpfUDz_ERI$>TPNXSDT(_ZUzr zyDYLWl`;Y@XH+P*>Y8gnP`n0sw_^xzL(Sv%TgzrX!$Ga$=filn`9q@7mdYLHgF*PZ zO2m@G*rQcex_W|IPDF`IGKaXO-j7^uNRWd+>!7nUrWvxfsleNu+S|h z`Jf%XrmfA2_IFE%dNx&iHI?AT%tK9}CaB+t$j0M||sH4Lyf+KfXKAVoZ1~ zdQr(ukQh4%ogPpeJle@EFSdbIQi=s2S#C2j@!Bj6N@L{CrqCAb2mJO^q&8mH<^I zXPOHL{rY#>?^LRlyI-qVH-PvhO}apf=;1^xqQ%X9lr9XxFF>yWG+*2zRDXdZ4kLMD zFr*1&M_7 z^uIA~{1%_)AVH3TRPcftkhUaOh#QGd|3!66Vuv5#&}*Z^q76$9qr7?id;;jaqq*1v zYM_i-5sbG{&py;jLyMmeh-bJlX%fQZzY&{}U8&F@YPk6&l3eB(49 z&$pU%0>ZDBU)>r5eomsi_Kr6M(*YBVj0X^8X!h?;df3uW2LK^X0nLhw1aQvl$-2@|*|zeS$UgK565h-_;1}tIJUPw7F04tAil+(pJ&_g(qXj+$()S zWs96~Y0D&$Z|{5GSw&=XoPx~5rv+CB@X)wk3RT3!r+vb#64?2~C~NJ#jlr$`7(H^l zh)-Wy5W5{B72Y26rbb4Yg8^jk#a&QsWlq#xqBNQAJY>v#E3Kf zX?juH&nVmvxP^xCSAG`T`x8k(TaR&pE!X+Fs3A_tFBHHQ#R}Vnb5Jx@1?_~muko=1 z{vEhqfU5*L?#0Lrpv37V6>8-{D}+ZC#EULrNH6pvV#A+>E8JScw_@e z|K9FNeF5>dbLIjmcTS8)r|?8C3vjjgUIwmi30YG_5J%iZ=;W7&s1{MoE872T*#IXY^Wr4wlDvd1&|akrY(ejOB>VLxI5=4+H<*>OA%N zFc^lH2fV1u)KORpX8hyDhol4UHZ=VRxgL7l>3JV&bc}6zR_CS(ZTijD(UA>pCs~1a zLebg{jsqEn$hC1Ge9nHnK9YGD)T}_Be4g3OxJk&Gk@9Xn1%D56ZT_}%`aL3P{MLDD zYUP#W)GoI}HHwsU+_(8%#>-)fTp-$CDDQ1oUo6c-1X)3Q$0w6$^;`+Pbh=-6HOv(thSY(+B|;lxso_ozxIw0GB7eH3 zOrp9PDvwKcsm?ws5Btg@hrOl;Z+7QMdOxw}H5e29!-|&L;IhwGO13!?JvQ?Pk53L# z1WfK6>X8OZ@wseeB_20({<%Fx_YQu?O1L+A_K!HP`{dw&t#IgjaHY&uAV?`^8nURT z8}B=KQxd5etcjK*)N(iER=)j7X8YV>eea*A!VwNm^gCzM6Pov^?T?4+73qjcGaqD- znynR@^tkyqiGT0{ShJzp5J1SUGlI`Rrr8JB^ZoVj9I&R%Dn_pwo)mYv_ak3s?os5g zvg4v(Dm(NVmV>Qp3H}dI>m672)f=38BVMu1u{~T&hqZ8kbU!uTW-B#lOjobEOWg1$ z@t@s9K0=o^^!OdSF%jrDub{s=jwgQmvd~A82$1tzAGEwc>>QB+JM%*&qw1j>0?$Wq z-G;sP$YECX2w^<^3{<&t5Ij6*fdq{}|9X)~O}5 zdjE-?98?wD1xdq>FNxaymwTNt{d@Y-Lc&osbM`8)ZgfH`nTBaNeYUHArk@I7(a$|# zk2j+U)C+9&fM(q;qroUja~kTcK$9}LE6uY!NA5dg?)jbVCLZ>7>>9cFb_o{!WhA6B z2$ZQ+{O3D!We64bO&bMg51Vqq>>0 z{0qo~#INgnz$*_m=u!YJ1-uPn_MF(gE;B=N(*ZiT0P6g4@UwarI%+Be{WXxtVf zB<4HwLHn22*P~hl1Cq8&JpK*C+-CP#x~wfmn~4NtB+Q?)f10_Cq4bh}5~b$LAZ^}EDD)Y`HGg>M?1>Pz)s zFBa148rvZTTHM+plibwrpzT#6xdr?V)qWm65zxaHSak!4WHh^EKTmDk*?vu)CH35R zkGjE8MTbZUK1{oEr@&AB0z&jv!{4w=Tw4K`t&z-@aSdaHc_8yjZ>^1yI<0nS+stwA zVS{6{Zhv|2tY6s`C$W;I!fk$q#(yBKX4`VeX5_b+hz>8~9G?(}Z(8t9dV9m9P)~V| z%di*Zjc#(p7SJKT66e#C&{MD}Ks<%@2NUy4?%QPo9P*``Bqj_o-9(0F{<3rD*Mpd? zjfm&`{VB54H0th3Pz%|efSEE;0qF<5=2NvI>?0xM$*T2O>^&_-x!s)w;uX%|uTw4C z-s#uU3wEPefTu!cYY(V)2ECIh&r?a7>7TFJKnGhg|9ag`zp=1XlaPOq%w|RXW;g1b zKJ7g7vub>i3t=o80m7cBdWib;psOh8Br63{oEM|=$t3?{zpY0)jue*+K^KT735bNR^=U52(e~z)Lqm3{4)n-9nO=av}q%t@Yb$X9#BEflLH9x;VH22); zM{YV8*vG3<&|E$MNJ8IRL%m8KU+YN9VMshuloJ8$KbE1f8WCa`G8d2f1)< z=-{*1>pU9*nUC!I1}65zHitb7p>`HCgP?A|fn_MrJHEyT;()S~NPfLo-24_;&>^>N zldrMonR)Ry3CHpYE>B%l!%D0Hef##L^q-QY*#tfee$8QDS8xaMIDQ77c~M0b$!v@QeJPSz=HSiN zIs4pC)@@#?7WMl1ZLG*|Bp}Rec3Z(uhSb@(m+<1qCfP@ylJfF_; zDI~=ELqXzV`s=u!@;KSwDfN-p4gY*paq!U)CBHi4UcU3uCto_*<}Bc!Kz@!bc#CF} z+0@RhdBqj_yz0)rp8P(lwQO4=gn!_D2 zo~oQI+f-ZZih*53h?i)jBmbR-tn1ow_@S09!w9kSDPccXN+px@@IDcT;K#JYN)fNo zP70{A_$&LvTD{b4_(4~y7#^V;ati8^v#19*k>f{pG{o#aK*2_|VfnLXT<6o^CQfAk z_ej!Ef+)DbzbZ%i8wc3(9`wCLUeBt^i4{{(l6AO@)HAg@g8!%?+P=3F_4W*YV;!g| z6*&|0M}?rPxI6HkyEFxNzvaflg;+ikAGHBU_rC&NU~~5F!oT!~`}hZD)3vig+V>MD z2c>A<+>!BEL`45s?IHq^^^h-Ht$fx{qrCeocYd_(%~c{t@%_fPX%1uOq=>ZHuJt@v zXd0Z@VOcFOT5)1emlCg=Rd!@r)je|Rg>m4)?7A0JX)OWl6^TG@TE&%=OI=;B_1@T9 zezmJFy-{;mV`2j%(5*P5%6rg7`Hxo$FC)#x;vJABY3wTz{;27w(sv)f9U&;raGARd zs2)|k3n$s;w?eQuRJ@T}K}5$hJ)~bz?5C1)ja7uj&_0>$??g zV*PfK5*xx6Vvd_#TLm0f494~njTmE+?{tW=5D9+!xR zi~PwRklA2ARow(0^IT{%K5VJW@rapw7(g5&T4h|@j*lBQ_9WkZ8g>c9f90gf5zhI-YvvyA zEAkw}i__^Ux)Rs&59FMfKl{D38*&?A)6V1>&GYy$tDHBW&~_SBQ|K|f{e@8j3PUlM zx%{mycn+$1k^5}>JYju|c<7?=%hPD$_Vc0d^2dGh7^EV&zHxd9{K)Uazq1Z`5k|kz z{(D?wf`zqK^X|`?6=lkxE7?-ub+Fm57q>#mEGx95bV|O`GTwYgOQ1&O4c&S3w6`48 z2NGnXyg@iW(Gs*^Bf=jwvJVGH{3NdYHi*{w>?Xlj(TR&jC%cRG>B zD&xdDZ8yT2EQ+D{=S0oUc;}Tpx*AK zRm!zEw<`wL^wB(~y)>LvJ8+wy3I9o0QCWJ*`Kj<9Gx@Uz>o(jeABR8569LDY+6-$E z(1s6fslR7zc-RP}50+T>WCqpcy43&Hf4SryD1Q$)C$o`UN={c&-xj&lq?QM&M^W}g zU=6u)?k2FPm%e}ez(CzSNyyBDBIu5bvp2(|uyt3ZrdkO4Qbo3IKat&tx93-9!%K zv_iUQ@FRhLETNAdOpCPxj$H3m$-L^^iXLC%&VrwCyrqa22XRx#} ztgG%E4<`7UKP;hh#{QpYUz0CFW_r`y2%Q4S-2+E;{4NX?n#&=18Xr{suTP0pl zO|V|>_gTxm2w(t1l|Jr|D&6bu-Q=sA+`SUarR#w1XZ*|Z*?x9%*I;%O$IMF)dN2St zI{@Jy&#I5%0+&fPAkq4^8G}6*erk-dqrfBPD(8p+rJ7PqHWpC?z}(}&PBUli#VYW8 z2|>9(tv|D?vmgi6=r$5}d?(~Di#QZ4zJyTFoZ$Nh!KC@m)e!iT3fQQfnGw*V2)?;o z9JA-fMbwa0AN3QtQY^C@g^^7LNwX!SK12Y8>RxS*Fi(}6k4}d;&ET46MFcQ=mo2BK zd^%KP&0z)vlUp_;WXiffbNSBpZdg#{LbacOk5O0Y|6|Fii<*hlxPBQ|iuw&xQ48 z+sll4#yEWnPea`G08H0>HrN}fjaa+W3GLE|g~e2#z}cU2Ndc@~576e#YvZgy&1zgH zbf4*`4hf-jG|~)Cg*2;K2c&jxPeQw*=kh+w=5`QepU2 zbAIX~M1@zVlo5xva0V;XE(u`$IqB+ak45zv+&0-}-y!OX!e7>lB@n((1k}XydesH| zyuDdpI?vODrLXIi;-|K*K34nmzENbiq=9Ix%#T|nS*MRnuwsoo4NsL`Ff zbsnaf*T8Io>&9*mK91Xfig4t$(Gr4%gS`TKbH;(18CmyGc77Dp_2|;5R~v8x3Y50; zHwkC}e*&*w9cTD@j`lZP{FRAC2x6BQK-#+;mv&@L)nJRwQgp+@z(2{XLYMJY`U}U%~@lVnzQZm-1r3XO-kg6 zha`aUE)~L+Fa$-(a;dykqzBSFhBc26vF%sWgU_cZEfMc5#P42IK96W;?gcwAPa9PI zQF`y-%Ast=&_x799n2){#UIPut?R8U%>-mslIHTjRujk=3xkV{nx&aTd^{b%q7)vH z?v4V1>2F@tB!FABv|v#LaI`qG6m|gr_^p*H*2Y~LvHcGCh1ID5O=9*@L@U6L7zH^1 z`K_m%ANt_{9rbJuh#iy$`r093-3bo>!3tqyE5K5ZNMd%Xt7{n&$HGI+)i#Jn9P1QP zmWbCKz&%hX?A)?Etg-J2SV;<`&=JMU7V4sR@NQlTjdn`Po97dB6|%Ni69eWECqCOO^1%v@=^?(`7_afOWgY$A za#_t&;f%f{*=GtN5~dAm4j^$&#K@<3axmR*K5X`nt!SJ>!_xh=@r*xD8^I+}#oGaD z=b60bzg!&BSH3Np*=Q4=^M}7d4f5KY45)V@$fs@h;LDnQy2<&VAubNh2tnGqpi!{- zM`A&8QR%IBZ!O~1PSgTI%k?8+8!Y$bOh9p7okw7fHa|~gx)(LX!Mr_`MWfVv6)SjRu1q<9+17epw_3IGE$V z_|1-rRYcSeN~(0EC@~yKl1Qrk1A-j5H?FM|ha4b&fdn>@t)i8rn4D}V?;uc$@?-r1 z9H~!efaqzH*$_g}O7WA|cd_vCBrxB_lS}GW&BVQVPb;^4Clp#v!KMNw2HY|i>IVg{ zx{Ycr9b-Wd!&)&Olh}-dytZ#xqUwt^Id-twP9r(0{KVS0+6r}@&?b)QgI1pWhN5L> zH$aLNPX$p!6aW9dYN!R~_sRZI{m{JK3&NPjROdS&_`F<0MdaOe`i zoveW+)+P}D`*`0PII*xbZpg|r1Uj!tUk9gAIer;oK#8Qnq7T-~<@3e?)-7hYxyW zN(4q0_mnOmF*ZWg-#&Ln&kek#Wl5L7m3WAyEFxOiQW9<4yQ}nY=muQ62p+ShX2A_& z9|j{C2ud|Uk8Tz@6HYuv6#reEp^U&?)$iFP5kBo_UX8xDm!uhZ8%*@u9G*2{*4+K&j17Ml+poB?;d9NP76IyB!IffZCVyOho-)|H}R- zbP{Y-=2)KFSw|8~XbClsXI*=r#6p|7I}!r$^Zfuz<<@<`BACmWt;1`mv|Tiz!^jR} z2?Epjas432QoEVE2;MS#Ml%scxSI6-Zme<)C+T0YHoTN}3A#lxiu!q}(}IXqpEad-IZ9OSXX)FlLENsmq10>0Hp7Bhln@eoi%e9bC5o8h*9Ozmf!&cF$iOUNZLwV&Se8Qq)hAGoNKgnKY+C14AFSIRs4!Z(QKj$%RFxqUA4Wzud7}V6E8-M=T z{Pq~5b(BzpBcsCn*UxUPw{?1_CaC;Twxs^SGwYpeH8bXT%n7V|1!3!t;>_L;W7{ow zV;}+@>E(p1Aj0kZqslgLmf|ANG_kK&@by_+K;m@{C5whE;pclD`dT~3^9C;D5}>sy zn8YM~zFVy~yD#y5xof{y9r%aGyiJH<22@q@L!3h+KrD;TpYXp{7m{GYKKt-;-N=(g zl3+FQap>+j=&$o@oCG-Uv*J9el>L4ywHzn}HI({IyGtmFfGwFskS{lz6N>KE2F;Y_ zJ^&^k=c>kA&OVs=2a^FExP%Sf*r*PwkV+07OV=3;B0^tSJw=Y`yVTrbH z(4}oNSqU}ku32gR&(H_)@mx$pv-BLRh4+SLDHkyRJzTmyxc4|Q3svrs_%;vu)*aeY z%u7|yFvnHWS*8p#S*E^m=WOOr-Ld2Ag$HDY9kAQ#u!se+q0X5nKEdMlo<_Xk>w)N+ z55yX<#xl4id97*o$`M$+y?UR8%5!;SIy7r0vdpsVvSj;Icq+Q{YR8)RHpU5ON`4#T zs|fi*KcA(Ad?7Ihb)%VqB8?y6(9Ou)YO|qtk*jQS7HpsgCgimK7ZSjHi!65 z_N_9#j*&eZTWkFK0*xeU%vv6J%a;LX9-0M7YP%F}dkKZla|3pgDh-R~>O(DphLmPu z4CX7E#h4GWvmZa9mN{}qf~l}UZ9}7T-sGubwti zEzkrLR*Hz(EVBY3v z1(Zw0jyI*fbWIEr&Fr7S)cYcK%f?b&s2cO$^w@cdH?&$#!TcxY<%)tk&1o)wrEYSp zn%jm43oHDrqVSv2!`E*t$ozJ>8!=BkZcWZsW3WH~_BVPqE7#n{PscaGBbO8NSk-cG z4O1YtS5mjpM>S~(2<{^yU;a|pYWH{ zUm|oW#g^+_{>86>$+N+n$l0~A{!;WT9^x#^OSKJB2x-x6BRvd*Mt=6WIUhRzKK0cYX6JeAohA=JO*3?Yd00uKii!6yl9;`odgL24&wQ29w78#+4oFxg3_4(1p^?o*7++l%^_Fm$b$}x0JpXc4y_DFZl0qq)*@MUF zSz3-V{VdAM=g&RKi@e@p`o7!VIWvUWDr#EIiVhkX1qm}?dS=p4gtC?C+Soavre*t4pzfr|Re&iTuXDX=1U9`iN6pcRyA#oo5z32i=k z0loL#vWG7&G4S3u_8Mxs13W0_VUM1BWc)8X$2EkEX{a;mUyvIWb(!5<(t0t`L`n?G z0ZHXb6@3qfofTKe!23wxC_i)!ERDF*rhmvn^|Lu24YBv3A7_`n6WfD>;eyf}GsbhB z=Bi#tk%{g=OqL`PuDhTWqbOzx2WL@SFuNVQGmV^r8|Gf~uI$-zUU#ONlRG*?a@uMm z{4lpH`FGwnDaJ|ezfUIN?MaBie6Q@Q`^dXI#e_+3`}lNrg6HVjI~+KfKXGEx#WGrv zfiqw3gt}41Id>3K<-z33)AOjo1KBV~Q+Nx`EhuPZJ$uFG1I|kP&pkERvzKZA9i=b3 z0WjCRZ%fr?AL=hxlCUPfb7oAtOok~C$~xC(m$`Qx<7+9U)6J3%vleL6-0vA!a%QtY z{dd}rXzrX@!&Z(41{qz}O+OW^P!1}+ttpKBFu|U$SOPM4$Yi4tB?dY_>y!9?3kJ0+ zCH&wpcBDIy&7vNU!C6gcJ=x~CY|a_SV?mZ(2+TNR{XPO2I~{a@rVeS3T6f)-X=csN zc}wnsY%X@C+Vs$FYgQ_;HM5yL^4uZ;W@ei4FNdT3QTPow&s2UKW&p|0b{S zglxHlD|h6q_)#&23~p7Rw0HJlmPe7v=x(~|XBjoi@6VHP+&QPVKOd=ulC>*Q9^@lg z?&i}hR|jaL$TNqm+QIlAmV8rt|DWu4633K+S}bpguUktIv! zq^pz!m|-y($o5G$&kwvI1DHoUCcCVZgt46xJ`b*N zA3T`UeeSegJ&XFNz_IoL^pnHg@(hPfvS3o&Nr zFsty#^J?SSXQdNqYZ3$3Wus1Wb13vH4@{(j4zrX@lgDQ#h5hJg6-&GXWW&evU|wRr zrXJ8hgN*PP+-*v36ZKs4W~~ACJb%-6UwC%T$mQ2BfqPF#`N0F^%{p01iuxXjIxyL2 z?wV|+=cgbUKQRXLG#^GzSh7$?#2{?5od=ctlnnIPjY&)777;<{9xEc+XiQMb-i8f~ zXz=H44s9y5KYsskvcTfZ3kudAX z0m)%`I<%e2a5}{a+86vQ%Tbn+5IysfEzQ`X4kIm^QgBfoTsE?w?`eB%0*ZlJ-5`bM z+Cp+t3trO3n$HqahrJ13pb*hZJz7h@_wGADqQx1hX%@typjO`c$p zJF;~}Wssz=!!EJKV7SMg?Og3K6q%u`Wzdf+d3@Lt&vp|%7g#Vc<~)|orqX%zk2i~G zu=)bRIf43Ra-9$qE17+t6_86ZS13P7(kMD;9`IBaI`nj&S4CJWw!}a2I;Jy zFdz1S(N51Bg9q7L@w#JP9VU=~j!zRW=Vem_Ryci#dPyN*v!rAh zpN^?mwv{4e?V5u-GDj}CXJ%8#>5JQ(46%wekIHs%-L{m2M*U>V89d-+?NR|mNJy$LDHRYqO|o-};?%F9z46wlM^f8`ysCq6T zmwUJ}Ghxa%7CsHx%50D}OBX3gWF>zKo-8eu9Nq~2ksdz+Dk@Zdjv=goaz6>ruJ^JB zW^5enWIbRg=go^tBT01ZgC!nNSLPh#(W`*GYo2hOk6jRLOo%r7y_8xjESkH(kQRiQ z3jN1DctNv!EqUrH8*~(JErFvuN1_gNg$g+sU+aJLsdOT^m=-D?-iOUzCJC&Dtt(c9 zU+cJ4ptSz$Wd*^+9&Lxp!@i~o8aVr4S;IMeRGAaF*F60=m5dJ-Te*Z{R<-;iKDdJH zC#PINBuhm06e@3&BGUCmTfF7%T21uUADyRCWQ4pjF3~5qQDjQdo)cp^4nMRl4Oynf z2ev_|42j-{it+6YUq9~8C!`+f!~kn6_p^DkSeS{}ZVFz!dcTVMAT?nG7%H)XUc z`L+MOnLEN19?PNKur^zyUh!syqw4#4CV8~E8c_YnFf^O3Syz!MBmvxZ2~&~s`ZAgR z2MG4=U%y2!ts=~$sC_xKRcX?khYHWtXe=in+;@SId zaHr%b0$Pp)EY6&Ro>lowF;BpdWtU6soc9B4>%?|AcjQTs3<-`r^W5@&o~X@A{5x5a z`a=nj!S{uNz9e7}S@o|7uUHa7c>$3<&)%U$+ghK&aenR9H-|P8^kwK@Ytg;#QAkIu zM^&brGdZ9HB_iSemHP=67azc41sRj}`ZX4QaNvOANo(ycgj<7@L5*1tCF zRRv_L1!Mpo6RIKMTajv#Sw#MZ?eug*w;=flQ)|T%{vH%Hc?MrrKgf5G4=W4luMUE^t3fWHTb>ApviMjz=8{f#xg zL-2l6fj^hW=HA91y(Gg3Weo2tUoa1>|xYI*ohD56a~vVU|viEzvIL!+{`9O)cW>3OX}$L3E~*vZ)pCr3DjS!J98KG zKqHHnR>LrcC4SXSDA=0l{Jfo%Pa46*(Qv3xY2AoCcO)wQOUsQL8tebAs~{V3Zd{zw z9WV*IL}qIkO212i04kr@yZ=EoP~j1bPo}S1wYVpe(0U!dsTerWRSihax^)3DnxN!W z4sT!@d@aO}ny@|sm%>@L)wW|z&qNl1ML7XbOD%zatq{57sSUGD5t>@>hglA`{H7JZ zjV@Ob$QhrN8h@dm4g8ueQ{v$J>AI(;a`CCF{?p$t?;qk(ev3#tHiNv1dR;jMTN1R( z+Pe|%ES%td0-E&oLyK&-Kfdelc#$ZAvM=Re@s*#N3}6^@Z^0DcH(Gb__dzMw8ZOX} zrDJmYY<-H`9`O}{+}wD*raaN$ts_w~bOx*DW$Zq(tmQBlHn|=!b6SPGmI3aqJsDx- zb&|wD^oDn9-dj&%m47B2oEtJ~fao*v2;)XD0QvPZZJh{sSjnv6@po2L#gJ9Q5Jq1l z*9~W`#H%;FVinW*mx4@pHDw-JN>wtAQ&4;Mb-QecO#~(*Qodd^Qgb~<{{m7J6J@pG zouOof>9Sxe3YHZV%}5z~QAE(0+W;obi4WpriFNt@vQN~R^C*|4B9H0Kn0_c4xw81X zLzzJ}iB3&$)0m6NRuKl-r1u`641y~fdBuP2sqE^Eq#%8t2Rjy!RTeZa8G zb&th7BTaew1K~jmM5l0!p|ZJi&DDKQs*(fuPN!GKW&>Ay#?F!*E9I&ZdGhff71~AH zi-%d$9l!ELyPm~heN~Q= z1VKwNBmb}f%#^P0O?}s9vmW5h&kWc3vaZFrmbVojj8Lpe)otZEhKviO7f3UWf(j1B zT`5DhTf3c*yX%W@ksY{hFV>?eZ5igx}T#!T%e;F`d@2bs?1E=-93C?hcF1%+Yk| zT$dF4{UQM6J#mXm6l6xoRNuJ&I4kY8$2vg9{8&!OENCc$v;zjsdTDR*X1`x3Yc>}i z@29PA>I?ARKBlmiUPjt{I!O8>I|wW_-Ig3%whW~b{w~+8D_Q3DFqUU@uNdqYF1t2) zv@QNxm&9Yh!V_KLl)Q9R#~WS4%`ejDEN1<$aM8x+qA`nFX*?J1JUyx_dDkV&z4WSCG5Zmt*^teU?-qPL!>^o4)s8Hi1#2w00(m(lKO-7J%`hAg} zs^yDRBZo-vqN zXKnK^e{jDZrfM-vMam6tBQH&15Ax2LSlZ$91~rVHPuZRu%kZ=d7R!>Pw|sr$|AwF= zH)PQJclKfM=nN3_xh?>bl|~e7qwgkhqV$Y+4w!0>?0}s;@LpftreO42KfTBqYTGG8%p4`)V_T|(z3{x@09+)jjheqm5ZapVhq`6l*48GPIO(ID-J|*n!>MHB+e( z*pQ)UV^&)Cxg0V6Xr5n8a}1;lwS9T=lObWCH@^gIe&v3*swROhV!))WbSstKBpc|K#?1f+i7P6u+q)nbQxcInkX3wye1+yO|M z@O#1XexR4uGy19fLY)ZA2ls!py?H!T@B2TV6v{G6*|P6aSxc5?NLjOsicE@xqEaDp zi0u24-B=?_^J}uMe*f|L@AvQT@i-6jcyP=)_kHejUHA38 zuIu@9Qxw#jmvQ^L@PbJ2U&*>voY)NIZ$_o2G(qqFsOQ==LGaq~=C}&vqvvghqBlW+ z#)N1K5PuVwNsN!k%nCA)2F@P+7;$uk49?yo|D#57#>{^`gU~fM zoM;yP-#)>y3tFIA{zR!38%#T7k@#*1G~RqC8_G-idN_a2efT&OVO%xBlbu+y_ImbuF63!NsmU^q7ztZ$$c z-Y}(>`PRSEgLE^jt2K$NFpo_YPBNOpIY-*hHDfeqFsY?uO6y~1%dJlZOxBJD21hhv#^8(0|I}E;KQE;r!hOkvZR}stkLUPW*N1-x_+Dj(XxPI_9=Eo)<~oM!Z9sd z3N9efM}s}zcl2w*76kbCI>wxJ&cFPgRD0o#u}JNi7GLjMe1QMc*)&%v-?xQrgGmg9 zmvx7Lb-Dgdno9ogS)*1Jd0xPe`K1V9V4Gxqlr{>avF${(r;@~F&$(1{^5l8GLyJ@ z-`bzxgPhmpdwB{sSiz$vPezX?Ib}lTCqXx6eqE)by~-0rJ`eN$zw_%j&4+oNlNig* zXZ^yJ zMHo1VEeQR5{teYAu(zSDE@KPoqJXDGTK;NrjEeDjB)SEqm|jlm6ypPw)+EKf}U40zLUhB z^+Xae^e!X*9;r5qD_0fK-e?JT>=mJ29qe4Gt;<5Kfxh5t&5=OB`o&1}S{Zl(IpE4Q zwy7U={c_%VCxRNS_~^;nz$LN}C~)$sIJi)|R|&FOzP{Nh^T>V0RuUlpgwSFI0Crqn*EMNL>;^4Yk(Amd1e!2+*+VQAHTiB=F85&cy zYMTQ@oed|&GK-={I;Li!b~7sbdK1`vyg2E*`Q?H`mr`~nkr1Q$ke`$uj9=#l2oD~N z+TO~E+ke&Y@}_Sn*}JZ=1`qi)`|joi=6yQEAeU!P9gC+q{8}t!-5Oh==bi*t=6B+t z;CCutxO(sHj;$06yPol1h0iB18_&D;aDca0=zVCwDv$kZ&oc`Xi4W$qe-hS!BA<-! zEkmct{5l{fdxW{?x2@1pO2r&(slXV4E@T5xHpb0%u>!GGdFtj{r1cc>ik9S~;D^g- zyb-jalIi+DedKe+hy77nTiC1o{hNRHPMe^0{VGd;))Zf?KIHW#fAL=7y)DQY?gNDe z>fI*5B)VMkCTU44JY9*KD_OT>Lj+tczU1~8;aeQh|3$o|WfZE_v~Qk?Aa;IR@e(T8 z!Y;dG#Z%rQ-`)wk)~d6=loD%n*oOWVu?rKq)YR4G_ttlMsbuH$+6(#ldsWpRc)GjB z_+iUjKX~w;!I?Z!PQVPyE81_Nq*{z|oTzLPUq%c|)!gA<|3r{FFo(1vmW zDf3m6DICXU)k}EUF~YF>`RW%1hV6IVhKZRWU%z6t5Fil!>j)z}fq}+Hxj%bADVoFg zTrNO1rw~5tpAFJ+9be_&NT_9a_xKK7m7t;rBcwj02i6uWA+uY#j&z{SKHg`2)p2aE zmrGs9Bry77NrLH7yKt*gb-G>B83_p+LveGK=$BiMaZE*+8SmFF90PBymzg zN=o9##bn{^}W~D zub>lpuGR_Yl6i=pKa4m7onFN(zS~O|{^oPJC?5ZcByz8;;E4o6o9$UNUB4Y5`|h9l zN#7~ayUJ8{d+Eo|z_Ff-xy&S!dMuW>nL`P?$2tIJ z6T%vFs+&k1kBTbYQZk4yZY|$^>(wQpD33(0EzAoaUpzO5_&OW}1t41H#zWAvJ@t&& z6>Np;P`2g!zTP>5j2Gn$=#)?>$d`axe1CXtPP`9j#NK#qJil}DDzZ5GC^}xQ2>v*& zC|_Cxt52e??kpw}L&Jhj+A{IeiqhIC(}*3(R4RoSipojdMBZN4o$0!B2}2B~t}wRO zT-4a@!zVGjUEM&cY~pq%JS5~C76MhsMs$t!wqcYjA8q!Ed~Gt~n#A4Mee&UFK4@FV z9^a1~n<4mS5<>~JI2ML&&0R72?v>{MJG;$@E4^0Zol6))E#o9(q8VcZV_Go%$Wi#? z0^t}E@bs!fGVSL79eLFqjCcRP9lh?s84>VMu{i6^C~B59-?{3Q3S`hn*;*TYkU_Ea zc{s z!HMb5G^MeK*{6okdF7xJw*~y*>m*1y3m(Aj>t} zHg|P*hb&d|S(l9}=!s{!-MMZeE*!T}8qk!{8K!c?1%JI5hUeH{|NAmCxHsXeulSKm zHqX5o#ZIkq4oW9-g5B!^wj?FuGYM(ad6`pfj=SY_)IYwb%ytX0e@gbNQ6{-w02EBeg#8+UH)uNBwx;he_x z+jns+L4cxbTrKMuj?pfhwV&~eEMTZ}EiU1o&AjrSkoSQ>=-m)wb=LXRO%0i_Dt!7W zORkHkV6SS`%tR+vut8yTVX9v;i{%+|A}-7RGoXeX37!aVVT()r`_KVEZuyZfEJJorY)6lin`0R zgF0FsOt;^2QRN4DE>`f*5FX)U9) zz1?AYzUPH6g6YV|v;A1a8V~?{lGsRmyN3;xQZSu7hPS| z9f`%rZi};*UXjHz%d~eaPQ&kQmYo1+FTwEkks})@O*q)#Q<~ZMFeu zr|TL2eK>^`;Xr6R?1NEtnk|f8(?k9uycp#xgA7X zo8N~BKFP%#-0*#pmMLqBEE^SR)vjBfL$Z?70Yr1%B@pAcDKUd%Wz_w*RRI4=%@Bi- z|8=w$>!HMnSByvBn>|g-EIk!eM4Ci@~D-PSZTn3I}G^qCyLX$&F zkh7Xu_29G|Vwvgy#N(4WM@CA<-3yNbweUg1>Q{(uN=B~JG_F@S9q=3I63hIeaLA1| zb&hGWFGcMAPn|&tIIDYx1E~$O9fF#mMu}-$aQm54|Lm^8mN*&HNZQ)y9JR-WTGsNx z9c^nEV;0E#T0SwC&ioz&-KFR)KrjBtXZ80X&+CGeu!vWfUDp=_Aoo}h=laY8fjw4Mp7d~eK3+gd=5J|ujFR!eRZTDG#t+io{WQgC zvT|<8GA32EU;ye$*6O(E6#)umhVLDwW64MgODWt^_Q9H;KI#M8oC&9O^9ou;Q{C0A zDd&YJ*ciCwPRff=d2#$m7nXla#t&1a*-Z`3J#6lF?)Cf#4UMJ)MF|)Me}G(`8|(UOUfJFh zHHisiP|2YVt%y05l$_Sv?w^U|XHg#BqhK4aZDFm`#J)31nL$EU>nN;kr$jtNJ+31& zPlzi}4}t3C*p4wBG)f)La?JZ>yx#fG>a<(1L&NSlq5mQ##T(1-22&&P5wx45f~F`J zSm$3=oC4TOnKx50t^lv|!^;H&qZRbXGLvw{p>F7D4&l_abep;glvTJ>wX6wQ&lA2Q zS#OOihaSKmhDlKVV2q&`>?=0!7x%z9j5gj-|HdACI3DlVW+AT8=Ure* zb)b4byvR^YwB?-l13>~n_0BT0tzXxo=KWsy@4lq@Y!FWZelA9czr-Rmexd_;RJ{)l zcylnT-6eR)M|hD=Iv%Xo#c;dN)Fy3kE5Z~&d!)-lgmzza&1i(v#z!T;qv#|lisu(v zUg{$=FRqTUfyZqmp9nnx)fd_(@zNVf{R|h2ecV(Uxc4 z51DzS-NIOPHq-$<5RO&6eI}U`>6{gSD{3E1k`;63Bd(p%=^R2%wn_jqvS*~idvt@s zD=TsdS+3QJD{yQ?QR+<2Z5AQh-rUZJW-&7O&|d_+Zh&G~e2c(#df{GS^;#J6@g@63 z#H&Y7-UFt%iGgp*?C0|MiO_cyA}#>FU?YX$R9ocfKY20tQ*>;1K<=vrh&xeUfJ&M% zRZn>v1`pjc)UiQ?N(Z{N_OKZw%K4X2@5cGqk9*il%Gz-HfcI6ezxBtE+MVzT@pTGL zQ!#M%+Bz=P@kjDJQia!hqlUcL`PmlP843CS0yko=gAR;_m+$Gcoj4l&pH+M@mDK(G zh4@Q^a#bDlZZONZa9=~9-%n>-uM;8C`Vr}abbQN1zt003@>^V17;m)o$Oq?UoO6bWvz{H22j5Cp8 z+_F39w-3>ZqnV-Uc$$?D-*fag; z2czcEEyRq13zSRQ75t7X8gTwB!XW#tss=IfXZTa-mNneE7Dv}fm0%TP<;M6etB(X* zeRMbqyt|M<^_Plz8i^U;mzcI5hSA&r-u>@PC~xq8MHv$eO~b)i7dY_NUDhxF-m7@Z zOjw;mu1DM2A(b~KdS*;^3Fz8!iG>MElK1_>$Yb!^pIZ;|QxEnNe1Ko&KmvlkykL`4Jc+GY7+%G&*bH&m1zh5SF6KYbAWoMBq(+JfebB=a2PY{D!2rfuZQzih&Ni zXTCGvvyjynMvLNLo_x~Z@zWi%kTWK`n|dk;bLpuV$2@^Ab*3c?f$8tsFj_rpRXlv# zIFCW0*${+5oqp~563Kn~m@TW!1%4dA8&HbGcnqkZm_t36=l#HeWDwWvnB>!$6Fv0- zosQzXR|BS(RW&_>Gpdmr10#Y#^xfNL@F7Z_J8CVv*$l)FUiZ7UBuTZlWC*?**RKsX#e#5j zV+U@!Mn&pZG?D6mak8TDla2}G+re#|T zOH~TI=gTGa-<5sd?Wfe4pY!Bl4*8|#r8W9&tNnq+oYs3UT#ke6kmX_0>($YJvpYMZ z?u9dv!40eM&|K_mOpT&IeX_%;E(Ab9eVT!4{yrbm%Acu!$1%W5%uC0kjw|avL6qqH zq}~|aUY)^jVwLt5@3h1!n^!S)pTsb4U3w-THlP>HbyyWdJg5J7PlBKk+|%-IAc`!of+ID5EYMWX$O1WS&4iS=fS2 zcfjKUxceYO{of_qekL9rv<37Cf>oFE$EfWyS6PAlo1%~tB!&1MNd;IujE$Aue~9Aq7-#MPh@ z75IJfv&KCEgG1KvK%8_>kZ5-*CowTSZ|-JlDq>$Gso$&Y-4j%;zh&)`Demfel98I- zs}qjdsWs)-(EA=V_h4&}H!smBER8hH_<(gj_r8HV?iB<6ew?vCoc1SV;mI{}hHYnP zmpjLUa9VTZo2FC@2K&BYtvOON;>4}S-wUnfChH`jMt`wP{O)=m=a*gSU5=N(i4I{A z2cDt*^+fk5Tz$af&XHPqdzDAN?B*T3=rv!dZ@WKBT(>Bf!OmAr8U0J>ic(nXZ4cH8 z_W*yZqzX^s8e$7a*L4WTd~b}IAAYtup?0M1?cGMhlu_}$F&)b}(tjpXnc{o)ol)O| z;-zXp+35B1Wun@x#v7g8WvVLJ6$el=FuxpUQyQ@58S>`kSQ*|+ZoaiLY$fj znw#LKJAbooHLBgcf7w|Ddm=I4%1BkC0_|YEjk>jcmlB{E6F_o}mApGXLR4>YNCL(08b>hRMo;`kE}1 zr}r2AA9Ek1d=9GuI&=Ojv?TnIIm*nn=Oy5YqQn~v70DTY%Ak+QobO4IjlTU}3qk*~ zN%z?E61l_N7L$kFou6B1#Y93$CcjYXzz@ezV9(n$CepPN#!MmT~dsT z;tlklSSWe;K63lV6WeyQ3}E^74g$Uw3$)frEocogje}hW@hM?9T$x^uYR z3b9^lUOOBGed(ItK6)Ecv!skR~xjE

{<8B8F zrdjLFx34$rPWzuH-B0=a z4*o#|9}T%DePP~1*q-%2g)R>8Ir7qVm&8SbcKX~K&^8d8*_hKB#9+}M@9&YackE`TX9`rAz6&|K3iVbG(i z-N`-QenE8F2|Y)8%yI54Ecu*w`@|#IO`;eWsXj1EhB7IRgHR*xIv>Z8US&~13G~`p zkdvK(8MBN(DEKKe^4`%1{y9sHTCeh)-$Yc{Z`tG%+c;JwM`NaZxAQv+uN1nW&iy4# zTkG%KoyOpp{aLF)yrMrk910mO{|#8hS7qTNmfW&k8$;-phfAiiX)Z}+`vYWl$hjXv zSYTSYuw7O*%~J)|FT61Xbp&BA1ja-EW;ff>L-jhSITGb$n*cG9wrd~y(s(^l{|qq$ z-~QT9=4(X0s^T#13^e!g8O#*NWbB)-G4Z~SZh?ckrmO=2#= zY4XB7D$VT-h5QVP56`?(ZP0rXpf>^J#M_iFnF@BMtl0`^%~O3e}#pC)XQP#h8fPZdVIK4W^Xj_*l>S}jbpJ*HC2t8*gm`M%*1<99yflhcvGo>8)DIC*Pcl39`&ko5l=nVihWO!j_l!^3}OeB zsVmLI#H3!oQ9F!u9di6%P`m9ym#mkoHo?{V!=IThnG*1?Z|Ro%`waLiwnc8^+D9(x zOf*!NSdVPuZjp6h7x6Dy)|$*S9^xg)K;#A~<8{onoXfOdi3_z&u!AwVZZ*dr?hT`i z*A2ecJD@d@T)i88fo$$}?`p65*n?F^O8qv6^SSc`r>U#Y%?dWPN+>sZ=DjS)Ix)ua zFQz$*snvs0$0=dcE}`Qq`QYfXYg+^hM(&l>e{jZQH@~yEvfQ;8^{RnR_B;R|9mfB*L1^O27yi&PP-{g& zVk$MIuMDpj1O+Fudv3h{j_DcDBel@p@fC~8h}S<%$4x66qy59mc$YAK6OJPilep&C zfIz}sU{S`_7<&=6>R4G?(^^nXy+AZ&Jf`gQz_ytJ6#U+rZm$6SrfrZ;Y^0srPc3s+ z^B_-f*OLrrBB`Y(qrg4f&wRB59hyT)FKj)qf}K>S6%JC#*tz|A*s*ftU#?o{5M}^c zva1X!im^b4P-0JX;qq*!-nsrsr>vna&CJ|8l!E%}6-B8B({$Qi`QqVq66)EFm*Zx1 zY<-}0mD1WQtD)jaT7f72o(*U4v$@r2V3`UJxV> zZC-nH49?QpR9kz#k*3Dkg<~CV6-2I;!jy{~i}sXZ1cu%HYtHTNph z=YemyvE%^PXTFj(%7sp(1oRpHq3!Np{p)~?1dzj!kEO#Y5eX~;8`4OV!U)>!9A z31XM7=9T2F<-cl(-69CCgqxuy45%0a@5+y|OEjnKw2e=jFZ+541KB`6gwzO{C(cp^ z3mSvxEDsR{6rdel%dJD+`5^Uz2`*;=Z2+HdY|=g7< zOy3I5MFjV4M4cyG15glu`qf^)h)4I@*;{}0g$;GNBjT*vIHG&>-^UTn7?(V#Tl0!T zm%EoKQtigk8ZeG*TAo~Cn0->p2Wbr+C|2t!*!+kY`dFBTQhYWXA2uGHQy2d(wm+?- zYU%PnUsq2N%iKDa_!5|JG4Q?mnL^|sSw}Mkl|R86qOgs-$)!21z{L(B5D4|$lTBNpiGQXI$9oApU70b%?roUBhjITai+)Dm!$zvqh10tD z$c|%M|M|M+4K?eF?H2Ian{mL2!LA3Nz&-ZxI$6$`8A&t=IWZWRZOF}D{rEYtTD|Hu zQ7K5!OLUe|Nid^8QNItI#L>hZ7GQy_O4dBA|MSCew4B3L0%aUG!F&0gu)5tKG|A}3 zIIoi^jWvc!U?aVuj#cls?o5U=!jf6CfSV2c>$car8+0uax_9q@zPGR0zjG{lS~bLx+$o@g&Y``7Q%1vWUF09n*QjZvFaR-jTHgGgXhb;=geV+zA4G0y zy;p4!Q^4)5d0IG6;!4=y*)o^2gUtJ^NC4hk5!!M7hwJ6|Bilvxkvp(McnnNw_M$C1 z(esw^X6hE<677B;L);oIKC*haIOy-@m-yFBG&j@3caEMX42bd?ExI>Djfx<63$P1_ zCszmA_Yw|HV`>C;drH0-qbEgGMBqLq{;Gv`Q7*Qh;D5vcKsO?a42sKWYaD?2_;3Q& z@DD5P!>-!2rh=8uv;>f-VXX#O(TYA3pVUjRf>p+nX5}MJ91lmTav46`Xk+)8De zyTm_?#lb`RVV;xcU(I$~b4OW(TcMF-mO_&V<1@DaC!+OgBeB&OfS}(ro3U5qF-Hbl z-(E?>a6P`bO<2-a{1pufslHyLVoE!IWeK_j{|i^u%4l?~cdYd}$QX^yXUQ+_j?}IrvR2*Ujbxt(Q6ps|AMx20=$_$Qvd7T(_A0sG zv&WgokVwll+^eU-o#P0&Dz3E#5Ow`FR$fnh49lW~!@t3Il16bs|0M``3_wzOKCc#` zfRyB*0B3*BJ3fiPa>oub?! zzjzJn`L;&}ap!ZUC`Y2Ejq>pnvxol!$O%VAKXA%dC*!XwS~|pYmWkvDXi8DC<4?u$ zSRp1cGU{pI#7$Z3=}z{brzwSb5c$i2b&IUwQ4Qh5j|$FOJrCSo+_F)8Jj3Ea1_=oO zPK%x-k0XzzX_I-zF2A2nTy2{{6Oq=y@^F;DucwLeQxyNkQ zZp=p9eE4C3hg@gN){u4u$YJ`PI3%k3f=~H;T#j;n!bs!SBe~eBw2&UBEv)4_AIB79 z^WwpNYMGYBF~Fy|Y+^TRQb!1dy-J2hd%V@oe4BZlLlgW(zp$6~hBV>!UHCe%7AxyN zKonk6-vEjZaP~YlXxng$#wSS`LlO&wY>d76!)`xYlG?DE|8+s2e3E!g=Qdk2&FwU) z;3{pAK)qdx{Gkhx3bw3z{Zwsl^a#5-_v(6$_bg;TVahXoi*kf&j5p^Dj(2d9Gu)q@ zlT^;do*W4EX|QONSrQ}LqLDL1-8n|i^o7l^!dxietS2X*);c8i(dIb^gYO^X;<>z_ zTe1$mAZs65e?4?t+*tw;f)2Pb?i3&V39UH)E74dW*kihHfrbMT_L+jeiuOHj9dz6ZeN zgKlR({CDy@%{w;z`JH)CXZ%#KPpr*Yz8Qkm@^xbn-z_qYEAR)+n}4{nkxqiDTTdx+ zJ)R8GeE4xaXcS!TjXo>O;}9sqUOGF?XVc9X_BCSB`U%YqA%{B=7d^HgkI&a#piO-xH?s;l_WImA?wX_l z$nfUWqaX&(W;-WIvDeS2E%gvg@+$M$l@+&J*Fnx*-TJ$K1D9waGmdDJIBujUE74L3 zI;U%BxsXHbCg|H9oHnS)Y{J}8qa9*F2Hx^_(RN?R4+7mR{m7Y}nT^?!~z}>Y9 zHQe&G;hV%Q^gn8Z9RjcWwbI_{SRcg#QJwa%C5$lQG&ji6nQ_{goE0`3R_D1%1PQhs zsndL90iDw0pOCX{CU_HMdDY)E6&|`JdC?3pOHsz_(BATPP9O&mEgy0nPpqI3SnbzQ z@}DCpW=MmS?4`|Le*&Y=HPJv9Kn6{Ci_n%HUv^9ea#(2_HeA9yL*E1IBA88lrWx|K z!oMDPCPNE(Z!bnQqGB@G56WSC7x$4T_$-7cfUmu?ztne&P+tTQaD6=B4Tjpgk8WYz zR$V!`$#p*YcQ{T{{s^X4rw2+TUB8FE%0;CTNEiXV+4> z;GamBZh9r`>+9AE!JEfBi9i+IhtV zQV6r7-Uo<}$yU9}wijg}XU^T{(C0i<;bgoa`3|T&E@HS`*o4KRjj6Q$%Bg^WQWkS^x89d!mkX&aZv=V|)2KcWxStHSF z=373EY}>fIsZ`+s2<_P!#j%akXU_q2A&KfQU&BLZ41fM4WH^moo<1Z)>(qa1+mXTe zR9B(@6B>X1CGwRg3NSR+#cLomY57IEWum`n`s~`B4=- z(yF~Gfv_{keeA+)8@S@RvKY!*YS6$}7V-o^tl4s~SNWM*z_jt^?yEX(n_gv~fp7B2 zEGu8Ha%p9U<#FcTzM{n-?sZab+y8n2GgSYvz$0~08M2QWbyQDl!nDAURuXjgKptpz zC(DT$En&Jr`sD5xn0c@=b!-x2G~Oj~nT~ysCJ-lsbiTLRT6UXrvATd8)O%ZA{FAvbk_k3q$Qde`@vClL z*d-Ths#XG0WRtH;w%3a9&OU=56xeut*CM#ei(gugHj~GUCy&xnr@ImC&;&p4kLen=e*7>c%+%OntVMNrYKzX zI?f!L*%bzv*(^`%UO!VLII2DCe+1e8%hsbg3lZV!!}>t!?YHp;@`8@P{gW`U<0HM3 zkOTNFn>Ftb1WwGmyXVX}wH`yRKUm8~C<*cBERB##dn=}V212^23o1IDesA-y#h^ty zD09o!J(G+*rnxBL1zD^W0fmYnMDAN-c4nkgf1Y9D6HY$a-MPB}J($`%g5a_&Oz*va z!zL65nd%{pNiJrwpFelpVzeM+TXJG0t-gFT15io!6V& zNT&g^dge_oq;!&*u>`#s_-zGag2{Ed+iTeJIYkY)w(bBO$d+ly`y{Tt0SQR_5~$`p zNAoE3bOw8WWv`Bp{MpwIGRQ3z@UIWCyA0JE&blukfpuIuW&B%V8y0iDbFM*q41DX< zGR7F4DNo;BdZgcJX)b@`&qRp|Uvep#Cx?9B#thgIO7B+frQH)pj^n(IPgKoHCX&@`UULnK8^EcC#P0)e^q!a!(dCxb$k0Zg~&EmHqiz zbH(mbu9?o!=nYGe(l+@TjV12SDltpre+q_72e_t@pzztU`{cTVqSSeL;Idn7xlBr6 zUlpESKQ?6L_MkmX<815sY$yZ#rm3r|{Z++}qW@wL1DEZ!iIWLW#6aNQT1)pbdjCw; z8W{n0nqI-DyYctrK!_khM6Cn~Ofr`3+^TeQ*Tm}I?jX}g!35e&VH{3>2^|0FZ!3?X zZDY^KV1KWWE#cJ?-Hyk8>h^qO|zmW{*zq8;)hx59c?`Y;#jWS>6y%odeXM<5`K2N z>A=H~#PIUaDiSAa+L60P7pp@*FLuV`g^rIZCrT&;KH>gR5Pgr$gns92(c5!>4(*N# zJ@wZ)&a>WKpHs}H_W7~Y5l@K~cjMy^1`GI}rn46es=dF91KD^z#zHHrN_v#Q!=BJu zSx44E(=ng2L7^apuk0n~n+QC%!sTKKveLW1rL1&6u@{8dNWM6_JL-Fha`+nSh{Ul( zuGUaVrGD`qW#@C{=gw#KvsReS=akeUNpm0D4zW&K7UQxEAf<@ z*javEF+Tlmwj-z8qxxB6g$#??;|9yW)3~y*Aq8|vyu@pt?g3vJTX>m~p zci_Pl*ypGAmPj9K_A})^`u^J?sWv~n(MsB_4bvVs7Ow4@vS!q*iG@9W7y=pjnmyO! ziOI9p+Z|Q!FA{s&^PjCy+df7}kx$|kvpM!-+`Ux~Z%K8Av6uY9bDJuq?O`r|%}4yp z`kZqPomsklwm2rEp#7Fyd4Hlk`cCQ4x~B=M9An@l@4<6rLyza--^E1r2y5{hv+IpE zzilfg#GbP$C6r(`V_2cza_dbaT@%OoSzWyf44seK>bLE)k4dN=|5Mzbc&b`|o0o@} zdP!E_U!Qxn(Lt)yH7#81rz0p9>!6&~-8b>M52rK5GAwb4{`rEWTfeofM^kaf0h1fM zAzayE(Zv@YM*WW1-giIk(hb(n&Tc-*I~x>Iz%9! zb>>$HA)YXm{GvQm@K~|}nD4cv52b8bNz{8h|K-)kNxfp#jl!xvyk-BlMgH0f%hxtT zqo&X)_V7|pYMGainy*KJ!g=P6T10#=z@)|>B(V}@2J{4xAL|#T#X{7}X6$qJAtEzt zkxP0*dcrA(q7!%aJxL1WSL7kSXB{=vR13OLu;58Ve*Ze@PQ8cu0l8%1sJt5n zy7G%gqfyR>!PERv#$!ToR*&Wo$!upgiEROG2&W8=*55XU7+Ki&$o`Zj)-qF#09pYX zDi@o->;ih08@`O(yK+?b;k>p<8RprHn@$G%t(L_&B%D>fAweG#RG_;r@`6v{9CM1r z|719OFUwi0!a%)|RWOhX59LlC^a<_9e+0QC$uUI!>@TtNMWIJ`q1I4E@6*@l+q-e$ zCG4Nhg|lz^LJBYZ^!xEOs=QqX;E^N;`W<(ci~X$h?DBJV*56qg)5pwxJ)1cHveK`H zBz;BJB>%)GPlbLg(x}xZbbpN#OG>v{7xUmD`d;$UG4Z|S*=EbuFD@?D9+>Vw`Q-S! z%7?csj&e7DNQ)6R)_RPI%y})wJE6g=p}Vg;L4q!g;`y%xBNlbUH0fRFY$9}Q&i;H8 z*FMf=mJVr13ebie7rgF!%m1E)dtUIRC#eK>y9?1y>rDYcM{sSEF!TWT&-oAQNWsnHg2i8G-)$=&cJ5zI7pEsEmai8v+H1bRGt$%-T+v`##^NR$jie;|1{7@;ro%JY6Bv@}eq?`_4{JpX!;HcKYq?Xxtx0%%MgfwSGvP+zA z+7_?8dk;11cHx!JsS6V9r$c{_6ze9j=Si5h_WwdcTx!xC>|Ei{QwEf_`A ztTTpEp+VI?uOg{2d9`8we1?Hg3y~J_iCtpiXR6~*R;lNJtO0fXc^K=*;esI*Y&a`1 z_Gej>F%Er4*r4|YRf%6*g)8=VNb2P>jPQVrB7RF6H2TRz?lLs2TwiVE1i-^*yg<68 zL!=5lH?z<2t+evRjARyT*<&z`anL1mwva_U?AP889r5U$^K+!l@h!)O$QxH|#X~=nUhwmz2y5b%vRGVFK zD`anDK^THBS;w&OrVa8IzmC$&d3Z}?K=@or93=c)8LL}*UPbk3%{7c&_{l+&#p3xj*q=TDB51d&1dY z0_7@kfSs?0<%nC)f($Cn(A^-2(jhG(B?^L414x6YG|bQqLk@GE^L_3wcdh#m+_hW_nK>|r z6Z`DX-tYZ-m*Q_^*uxrwD<+lpxE|i~zhSJRH5eS9guo32n~zkpiI&qzOJ_f!;)%$` zp#16X6iVOSVMJf%^uUGe=QCMxa8Y;6q;c1nN8K-+%eoDHso6Rjg$vq*0debDIpeLv`#uiYaLu6t~Ba^eMGNwukZG+f+G$D2%!LIk`8gYL$x z;GJVr7>%FFM8~2qSCG)lW9IP}z`{8@of3Iz-8SmlqW7xLvt4{}(gi9V6f<71; zQSY+ATKNu@J+ynk|HS8ifBXvbMo(H_i(Jo!70EHh_PB^hUu@ziZWkYoj>wO_8jFw) z@pUeTg~h7h*-BG=;@q8%De6#2f41Cxx-?7|1#uktDf3~94a4#i5_q&Wh>9ROPO=a= zIa3+~7`1=?g!6pca6uU)qVIePav{v+*mh2W$z^d(Y;p|w_cuiRBDiibD zv6*<9hmgnye4GSx_C1-Kud~0|u~GP{&Ps>Uhj{tj^a8fz;)rkMwmbN$u1ezi>A0ud z6-Q^R-+?~59>_-3eZ{z;aO6rv)YaIKPMU?=z`!P=zInoB8u6PPP8hUo*KNo^30}ZW zig;=4kk#%WYf`f51lTilfzJ(mFCSP0P(+jk~16PnhX9MBdD;uuL)2 z=Vi4kq|dF)Om5Apz4{3c=Ubtjx)Ea#ry-Klyy5F0%oqF3V17le>ZHUk7`@(28A=WQ z1>Cn4A3}7O9PUs!Q4OBwklyL%#+GwK9GYw)EzOZ7*S2gARE4x3EU&U<3_~ut;(OOi za8rXin(x+g@qc_$--ZQs=d&VpgO?d!gfF?eWg%|(lprxwsui{;(%mr4`sw#YG>2oW zX3513x2F**!OPuK+%!aQ8zP8C;e;w8A`fn3ihjG7NH2#-Y{NEm2g6I`#G!|20eFwv zCp4tN(Nf$!@XaLFrqOOrk9Tail_XoS z6kIN=xEi3FCutBx#6D3N>10O`ux!UeQ4t|%O?Y2DIgW5!O3`7IG=3B>{vTvOzp*1Y zfkGFRfK>pomVBkCnc+0S(}L0eJntxT#s?l9-Oyd6?f~ z{#(UFFoe!%L@?7&D3t)1nOg7&|ATzeWp>9lb=}JAERGQTojtGB6?Xb8tR+3{S=;(E z4Rf;T1_ko9qLwIw0|Rd3!!$*;o66KoIoCC|y$CFR8kh?`TNNN+l^bKdGaY{7We84Q zOfj%~t73>5%sWqG**@U>INONy^A8c>l&=VL3OeK}Jg{61zW<0e5h*J(T9Th0txmlS zz#~WblioC-{g3xC-zSwqvDA1D#e~)t(2)I2g|j!NBgGr6c;eO_=k{T zWg8mgga6*%MTor9|1haE;sKQuBQpWJ6M4sOSgss6*e~0!4S177T1UJx)lgfk3&ww~ z1uGtE1{*UBsj8U^$}6TbwYO$WHG;pb>j* z^y$t>kWw{Jx2A6SVEJIZHewaOB;uV3Pi)64cHW}7mWbR`VD%H|yaQ>bP7kYpU^vH1 zX9f{v4Sz*L^vA8sgi0&`em0L!Hr=d5(XHa%Eg*AKI9gj5_gc7agR{>#7Gt}}KzSFK zX2c_#mSIY15N50jY?o9T%=A6JUMW|2!Ax6Z^vHB}7FhKSVJPMiRHGbXk|yL02uJMd z#+tl&B@4OONF`dkTZsQ-O@RXIXej^SAKfGG6cYyXZ`w16Pnt>)(^b^k{fDIw3~#=D zY?^34*+ksPis90%W)a(z9INat(TUT;x^^DQ7*;ViWVP=bUt!rgEb>YJo@i1HW#waPEqf4_z5^bm>~W4fk$r&z;p zA{Ek&fNI+~{L*~{3-Pqlk%%@R6l+apHzf77W56V`2Wh&d@ol8`Xd_`0Ox z%*Y8jMaXP-P_ZH_0+~;{?y;qo9U;RvKi}(fJMH>#y#I*$McK21z5XblL(wt?JgZ+- zrleE7vw+7~fNCH++rdQwSdy$v%hIa`u(bUUL-+1noV$7T$t-38;#OPp^w^$5kI>A9 zd6}7^coiQ+kc92L7~B4bC#HKrI(~kE|0e5n2OwBwOT>WU!!Yi^5b#;qnT%j=ts4Q$ zL4khZbg%Co3ub4h5>`y=6x5OC`db6v4Xy^x*>_j<3b7yEX&La7Rrc9TP-?ZR%L%xLKFwk=!c&mxx(}?3FESRR04=u=W{f;iP6LCUO~()O;^9 z0%VZBT!Bg9E>SX!5UkWntn@NL9}JAScYbvrm*S>8>Y5ixMi3f=YOe|T|4uH0t<9yW z6L2fg1|=eB*oTz?6_Mpnm{S9U`~hb+FJ*KT@o{NHkVv5E!H{h68D8^b#wP}If*-J; z-eTO-MF#U4T$y6}*gvgcJQq8qBfMlIFAWKLFl988cRu8dz8XtsqgT%T2#VJbl9ArW zYEM`}>Uy-U*?*PY26PqZoe{;=OsC>KHt4i=UgK2PW`7mdX})ILEZ zf>&63F{_-ys!{-6&dE}#lad_2v>G7fSso*K!+7|_`S`G#V>n%iDiK)oZzaT4FJ@K6 zW|{qfh_jJRPbo9L_^cN}=mAQg6}tN%m9U-`O2$J&Y|)yrwyP4BtYOKtEInesOMSWM zN;O{QVYLi6;P`l*Xc`Yw)H;X);UpIVkBabt2is%ca-Hm)rSGf)l*G1r+c4YoV?5Hb z2*#oJv+|pWzLRwV>{Kfq9w+#`CE^;F2yQf!*2t1VIP+n zHWiA`x{VfMKMLb&IR}k=DLPJphGBuq!DPtNOeOx5q|z# zlO~sm!5n->9%9&7DQ_AC#;%BN1A$K&;OC#ymy0T2|8dg~o<`h%s2$9Ho>a-R3q|= z?=qeZ)~JmO*v4-Xfevf~%NNzdB4F*1tm@hp8$~~%1=P}tJpdQGrh$Zc1BL*D#Tfev ztf67dz-=HTVA;A`LTk8;3vNhf>^`F4BDinL67oH>XuP;- zJp6<@QJwIr=zRe}E#PIiI}`F?`5Q?h;z{psgKbz_CKDCWJHV@A`U48=$*NaIt`>1f z7K6D}$2fcULm^sl7O=j#JkXBNEaEWjsYZt=&YUBcONRa&@9;eV~f;B|1W3o+=6KCc#eT3K4{6+&%Z5!gId7 zw?sji4 zmdUf-h(p|xRN{ezmVZey)`L|5WpLDxVBMq@^i-)-ZdR!_Q`EM5G&&aQ{c})jHxWJ*Uo*+E$_wHqP z4iPReRC($$&|Ju2iN0j1&-nI6E3^@l^t5zd==vUp5DHe|jk3+4;*jamZJ2J@8W9Lf zb1W#w*qW{mCS8qnX_CANvzhyg?hr)$x&cM$&V&&Q_m+6KnL_NvCf4GP)Ph5sAVdC|`-zA(l&OrJ4 zSTC9g%=yvE4CX>bsyD8UDg9g!hrk2GC?t?rlKr)q@1%m0QeB`K&FB^q58Xm^+jkFt zlbxbL5s9aG$)M_1fjB@sXW`m>Hf#u*ewXbft>C}iD(8BHH7;x0-IgYF5yUkq?~RuR zP+wFEPTWREKyYIP`XDcjbqJz4gNQ9eC9OnQb50yYG^wK=|5wS22y|9b*eT}$hw#P* zGS2XlSJba8s)18b!}Wgn<4468r0?QW-#Qh~@?6D4fQ!q9+JbT^CEL_WlWC9|vBJW{ ziey;_O>Bvd`|1UY4OfaL&>xQ>VCpg(|4ZQ##(QYcyER98?`D>T!lLTqDs0SMvXC$E zxl_2=c6S4ZLi+V?m6GUY7w*xhtN0;;QDH<%OLBbdgfx?M{M($BTgq{_#kg+*BN3+P zJB-M=cq0p+gEf36_?RpLm&pj4dkCy-6@<(RYOJ`K(foR^lC(t zY@R|Gt_j3L(0ktj_!8}JDd6cY_9$s6;2kjW{?R%;P>ioRR9}slPZM*%N=-TZ^Wn%k zqVPGWrv?7HVHO_?LUbprzLw#dzK}R_g9fREFdhbTPZWm^p&Fue0(yFY2yyn6W@U1} zQpC%}JaAb!FID7Ox*BtizzbYs`y5~^VnM)QzH|)aLE&<13y-rd0VRMo7MKFd3$or+ z0$&5zmvmh8DY{z|fvKg>*GyqPpcYbhtBX}q@bzD~R18LGpMz_M^rd8!Bs=HO%t8EB zTFaV$Ef~bDZ?Pg10csUMI@vYdNy&OM$>KgIBvg$hs&&lUdJ>#RZEhaJe-!jg;Z^Y; z&nlpb)DVWdl+2qn1fZ?;w0U4KxW>J=0%(zdfM2EmXwOU6MuLR{Zm~7CH~3$7RiAu5 z{<8-g<+%XT>24Z^(qu+PfIysg1k)-$@&y+W4jOEHwcvp^5)0V4{RaPsx#v=ifm>6| zySZn{2+|4!g%waeWAIr`8dvfgia8k3X~n%tsd;~<$XtIh26nG&YW+8%F#_Z-d4?1~ zk?fTL{|(3-a9uwn_2_E8F4#&KdBYC>QLu>stGa9(@7&K%vX1lLYl6C6RaRtQ;?v#y zVq_)HL2IR}+1^Jm>8i`Nsjtj*FRHr@6oG-U^{8Bm{rlt6mrmh@3p|zN6J13dL41`& zpPhNQ^j-}f*jms6c_3`9+?M5I*q>GU8K5*C4=d%QK(!4rQ+-YX@kYZ8|EVnl3SO1d zZCcmB?(*}xE8}loH3gDo-i>~;5a3G*qRV(@!$Sc{DhhI0vna9)fUfIfRi#?AWSaSKFz7v^-zYF71;$?3BQO2yf!)u zUfyhw`Z!JkIv~jtR0=A){scBUum1}^pi37O1uEw=UJ?;ITjy7KA3Pv`kE2u?lng?V zD0MK4bq*T<+v+}sgjHw2-_>%Ci7Mb1W@qVoScVWNisibT{sBTw{Px9dr|w`kn#c*E znfNs><3dO!fKE&i86JRw;YjO7)(CXsaf+Gqpx%~hwipN+({N^u#?>;YCOjzxuJ}P> z1uH(ZL`rF49Py^1Pl)W)w!wO({r1i5sZ}i#n3q6;PinxBQznld6)gs6AA9$M6pB}@ znI0t);0fPqkCLeY1i{7GL$HiMSb%*c%_Zcd6as_h&4Z$h%?4&KfDmX8;fliEhy|0X z-UDlJvm>$)5{69(J-Z}s>zyo4fFHQb8@+!7RC9tFud!sr_Uvm~i~axeFQ#qY*y`h( z)MA^HzS<)RfvuJ?deEaKA8q5Zo5hqMSW0s_{=R|*GR9I}BkY>Y)G{wmF(mT5A6a0g zuxF0LqLsZZ6_-DuV@1wyIkYWwpdYbi`L?`V7q~h`NTedV;Res!-(WL$Zl_iAx1E^+ z(da@!Xf`Kx$z-1F%<)e+?+{i~_?+_Fh?S^!<0AW%U2#WE#LiS!ti&$cr}>_oon_~d zihh9;rO1Kn7=6^qqrgI4?CG;Q;{vs1KC?1JKlR?8M$lSZGy0Vy}(L`u%~R%6Do zV*gKQv#~h%B{d`&3j#x_}azdCs_#jD+|oreE)xCC(` zWxBP$UpmRX7FXQG*N6=pzphs%X3d^zE*`0Wo*O(<9nk;C;#7jEK^X#*?o1UeDuXLu zY~eS#dXt{Bq|c#_5`uQ2fzs=(n@250p_z;P7BinaiTR&$w(sj_TPqDt>^#|@ZiEoN zI`8wIqkggo^%fj_MUxl*dg0C>#xe>=Dg+g5Mkklf=Wv}j5hSy`=FLGJyR}SA&~?SRT34GPLBJzITVkS5$YflljnI$RlwT_;iQkU9*8y zxAtVv`{~omXt-oJKUv$|p*^gd&Pguwh1%mW76{s@D}{52>vg!R;Xr*n%Ug~t6@ET4 zq(WcYJ6iCNk-k4pTFChW+TnVXt^3)vKb_Rx`jp>ZpAk;|0|r7hjSQ6AtaqDP7K6nE zuzn~a4egkO1s7(Xi#T7JR*3~T#}(VB1Vit#!C@&a?bNdyNp{3AgP0Dv1dr8|96FI) zo1H)=)!~XA(y)OCIEG*0q}g*Nu7T{0`OrRy_U@U0v|+%IcEE7wR@l&<2YNAJAtzW0 zL}ImM35>%V$_@4Df$1jOF7#p^%qx46Ej#OZ2mj(*w9PqhInV54UPE?tw7~&q`}&l8 zdTFM$6po{Y?miHLcQ>M!{FL^)MNOvl`K2?J8ISos*~Z9@lof|#E<5xI57Ef_6P~z? z3bCK(#%~7jbfiA!iVWApHQM-n{r+rj;Yv0bH6_R zgHjr1AasgEG=D@Iw4;M(MCeNOJs$tgEc@)VZUtXXA^<>(CK|1eV{Cn$I>S7%)-SS$ zX@;~LtM0WP+3S)&ULGvt(R?>XmkGCW8zLq!->>d6FKnEPDb2sF_0hKhyq+0ixnf13 z<_Y4q&1lkz1gxc~4qQ#)CFE+tIg}Bm^%iP!tgYPTd2^TG2YpyazRQAH5xNOcXJhX`_6%kW zug9$yAdm4u;^h8N1FnZI31vzBq)C7;xhx#9|0119C^g5LdV$Rh-b`s|>A+{OV>Kf6 z`nsqopEwO+*FNjc365qJmJ6R3dU%#VMeN(-tdI9T&0p&nfV=hF9J>~Q3i|o{Xqqv5 z?p`FOS_#C_mqbTeJUw@m!QK^bF`$-&*sbj_6+wERU>^6cR z=c`q9aVa}!mKPsI_V1QL2s(PRq8GSA<$0 zA;Raqc_0d8#Wx!M^8i`U&ON*0$TwIj(DpfosM#8+MLE0zOAuVO=E1;S(uGd>r-c9kh9CDd`odkzhkdtqO#E1Ue88kW|E*Ci;owrXn z^rn7qsbxieBGk$xlFRU_0KAzSj^*6i^=k7s$wi=oA-m-{*1Eg*4k^`Z!3l(wY zS26a+ynFW;YM;#Am}NHfFL{9^rw0;v9&`tDKf_uxV}R|OVpr!hLJ4Z`6;)y0X6^K9 zB(>m|a(l&{=g{5a2VRb3pa}Jd&PcakSM4MfOd}S(u&OyJ#q;`7n+9H9|KM-$pi|ffF(KyU zFh&`;@=ah1`p|SfZW<9#qFfqYI{&U8RU<)_j0gz!cMQ$=v-cUb@5lkJ5a*(N=!O!a zo)`wP7R{IB4}iJC%yBm=mh9C35*!aE=Wk;f$_oC{51ZXQ?lVr^V{(yxlGCHa4#LOE zo99R~akG2>;Ply6LOgcwOuVjr84e~{mjs9E$q@IQp(i7sQD)n#w++4Dn^$u!HvVp6 z6RhQ=BX!)Gf-Pjn5|s<^5sr}50f&7m?P#vpglmiA-rV0`zJ2@cpNH^bBod>d&gr?P zvmDTar=cu+Jqjt<4i0pi#TP+A+?fg$>hxOk)nXDQo;y(rizSe3T@I#?Win!Kb3 zwtbpZE%sTUicDy-$L%doeDrwC$tjk7o(dnIHg*1LzOFZ+qSoY|H8`@J{V#GekXt&m z4-nPst3a#K(oZGQ#d0!j0=w+3<;tj<_YtAQ3cO$$Dp1_l1DyEMZU|GGxez+51zCxN zB0v67KYacEC~LWhUMCKWe|>0$mn(7z9C9$jTWZZe^yTnAB17HSn?aASmRQrQQjtUY zOa@m?+%nYr#1N<*E%xhRlqSw*Hi}MNXf+mmv64=bApkx*cV%bmcHnK7N?dtfgGTYV zTzlKa2Dw0YOnB58&*g6fa?+be_?#4!bZy1`{f*4dsL-+)?d2b~e`cBaXV3VHh67*V z`NQ~?xgGCuZ_W5NCKe>!WA~jY?RiOt*;D-?7aH@TJaJOm`>ff zXWi7r%4H@IHmrCz$NB^|`W)KhQTa<3@@p^lD5(0KY&#w$$fKaJO#s1_d@7pyuWUw& zY%%?Iez24;A4krJ)bMOGL)zSL^bfR4HTkdI9=kwT1pIn$POdo#^_40wH{?d_GcNS6(OktL#oB%5 z&Ts_lO4+Oksg)hnxRv}jd4{#rzaMAbi~m|H-KTfySR}sh+3#sr3A9Q3t+6G}-~4hUu1?sFS8XL9h2lIV#sPy9*&psfd2YlK@*S{E+~w#)Z#Dfq3tyBC0d9;%I~#-;H#Lqx4%3K-o+s%` zUsQOO3CBHUp6grdhWi}ouo7i@9&84{xn5uI$!9|T8{iRsx01H7Z@F|Q+uBD1*f}Ox z7I$tZ!e*_TT?}yLNVWADd8+Bp>{>*0Q~}AI3`tVkF!`X(AYJ{z2@&()mp5^nRIYe9 z5DoS87LR5~W9j@OT+<^5uQ;@xgyuc>!B@e;gAd_}ZZ%nxrl{$&^L-M~^DBuK*qH5% z0@pm$gGO*k_dyKG8p|dEz2Yz!W>B3)yc7XTEHYCNk+?p%!3taHnQu1vaGs5oVEQ_+ zyXM-)k|Xtag}V0I78Hs!&W+gp`=Z^t0ejdb9}(W*D}H$NyajpgD;psE;X{&FPtY zp8X5}od@AGk40_r``{mB#76(YFASJwVo{r7(XZVGTUuqub9zF0$1``NzGzsH^QFro z!CO}N+$L=d0@%r0S8~seEY*+BRv}~Z$YNK);V&ZH4B0>MFH+o^Nvs<_F{r8m*)Nuw zqR9Ab%5R-v0d^=FTdx0d9kS2Ntl*s)bnteM`#f2H2)-^c0vKg-J$HPDH{5xcS;mZI z!!X7sE||SoBT&H6Fpi1$i9y?A&k1J*U%)o3$G3J7gYG-NqL% z3ZhVCM_UUe>Zskm9nF7-pWi5mjh5=oy{#u@oSo5&gKcNu&h&T(xw)4QF25g~gP_uO z=BN7lqkpd#JUe3|OsL@&qvbom%S&ASPt1pW_e#;UQLWP7-vEbX_q1fQ4{-D~k^2Szr&vf$d+e#w)-)OUin# z5M0Wq`k3HIKd5PbSRY<~0h4C%VTVsikkds0^V$?Eofi8?(g=2V4R_umr|iJLYICXO zBKBs_2Y|(058R3H_Ycr@Taj0Q<`F_CD^C&N3k8S>29nT+6^9?l7PeQ0JNzY8O+U1|Qt}Xycwv!9H%6zZht^-dwdr@aMVo^MN%w`$5y;$$+ zHfR)O6r$=Gj`--u(l~ESZIWS4$4~u%{`Bt&!i5uVuR{<^+zpcsA~IGu9D7@94|@3W z^yL=C;S5#=4u;|8x~YOfnc$&a?Krc;&4g)DeCJT}KD#Q|Q2-?w={wnB+S;i>afS+< zmHpNve4v-8b8WNu0d0G_&|nKBr|01QS69?@C+7gmK@a?ZKr|X>up#}QdClGw!=4iS zricE&7QK|cU>+y5hQ*z7-PSkCrV;fIzJ9~t^b_6y4{bv)B7fA7U%@J0`-1*Z^(zlS zTzL_dvIIDIFPECjdWvW@$bY=P(0rL&P|5=@0S6v=!2sjsrB+RDPEIEnW`Iirvrxan z2$bbWmrIojxHhKd;-rOQ%vLf7`y~f|@q4)&xg$eTqu9wNO9D&U;}3160s3@Onv1 zzzHEofP^Qi0jCIZ(Pi~J7nMiL;!)$kiS^_Jt+o8J8Gq^m z8@2fwsg*sjcfgeRd`3GUwOyED+!v0JF+e@+hWRmS$aS0|`j1ykwz^sye7fU|M-bl! zVBN!K9~#T z;n@CwsaQCEdi+kjG@dx(IWJU!iVs^>!5A6d8tz{sgKQhgIGI9m8Pw zyO5W!zm2nykEAfV2W7pt@)M?LwLHoVKM|bRrcAqh3LcZ+q+Z&*7h_C>qu(phzc?po zJa807Qe#L0BF;*3F?2;G!Ktk*gH9PB)tN3koTr$iMlX`Na|^Y+pvx&sbh!U&UHl5R zvu_KSQeIR7Hs1*^_%MQED&4TLlo1*Dc2MH{+h1LgNUAr^sO&bJ0q=0${^Q;pQ2rAU zR-L^}0EGMtl0j%Y&TwCk2;c8qUL@kL z6wn$J2VPYsMx1n!yV-a*PmlN_pJ6vn8zGOm0=w6741n0mBt)-iMZxSMaJT8wr7)z(*$sS=uyw! z4UKi%i0ogk;uDC_Ia7;KJOJy$m4dj(Iu3UM5;-*C^@y=B+#9ap&Pk8pF?wj&*OZuO zU`li%808`??CPEG=yn@tFd^v?_cadZDS0vWtqTQuiivHpt_}hT5S+kVq(&n86S&Wp zrOQPSTPwdrM1IF9O)bpksOUtZRT&jfa>pOFj`xaw_z_$%3)e|Jiw|tXTzX@rlv-=_ zutoFROP&(mRKs}*$L25Lor+1@px78Gv{OyWxQPrH%so zQ%NXl>R|?4?_!Tz?VD2s^76oZ-&Z|>SxkNOdKXBSFy89&wppJMrv{)|xSLGHB&6?h z(fP-u&WhChRS|+|x93pRbs8q9eY%eI%V}knm%iT!@Fm2Q~jRNV?^v3m-%0>-DIisuq{my{EP4J%b zBIZCyaA2CrSAT0sO5HoP>K?W9`iI2%Umt;<+Oz0V(_3KX?^A(5jNB)V%~neVG-$Z4 z{6F`y&SB`GmZRQC3ay>4A@TC9g`h`&3vpk(nSOOWJOrYUM{URANM)4Cdtzg-@ynyh zKfKZurtN7{r^3`t6^UL{#bxtfonwy#;nrJAH_Qx&1~@lq03?Im822NquN)Tdb$SRHqwKebi7%LS*t z5}Q0}h%rXq#<^cPIAa;F!)nnFWZw>5%7N3PjBKk0`g(?J?sq}U6jKXwiY<3;1$PlM z0eg2J)s>?4a*$@@B3^0<q`eRSGIa^2FAatT3u+-os_297Dd##8GlPABlF;b8Vs~fF#b)ONB6`@3SMj(%N=;C?f3@3k>swOXW zeL9}>^LtyuNk@N&1FlfIB3wGm=f9q2q^i316Lx9eo*M&<=~x7p($OZBJSLoBwidiR z|M}m&gZnuaHro?HrdZT}vA{ zC;#vx!E8bL+wvk?+#*)=M(dy zXIZ(=RhJrs{Mb*?x3xTR{=wZRq?`klW_Eud*r}iz;o0Fh<}{t}(vd};A%7zWF`Cal zC>Ha8Fg0PH({7k@cCkWd6iJ^31W*{&JrWx48bho8X*$S#KttAQ#kFPxdN2<;#Q_F~ zjUVoB6Wm<4GF*Ta&Bj!*C)Kwx-WY<%5PL8K1c72C*b`4C&2K)opayQY`Yi{%X)7sp zbVq|Bw_*O@a8mOmB1QCd1XZ&xfN8>#_94!gf?6JN4%E{S|9q<^dU*Ez_Em%DO^P(v ziLuV09M6G1U3;KkeB;S5b_m$mW?&-gB*dOn7CZm~E8lJ>0S2n2@>`cM5YH`q&Kmen zB@N{1ZzwWtA*(fa^F-p>2209twReG2&Y*`Z0N?4+FR;Ey&+c&(%+{}U*#|$6iz>JL zG%wG`u6;2UMKAV;?Gr%FovyLfb&Kh6wL=K}ndY$64O?0!3EU6tPIQO^BOvb4Luw-& zS1W0DJ_-ojxtYCIr$^@B#xeA}ieCUPan)&F2aLes!5xR%Lxcxa{n;pxpb1pI6w#2} z1YTM!CLg730~ZCf-(*q{q?RP3LKm>2-&_xO3xKlAF)@wwwfl!-(l$}q=*8U)!kW?T zzp701Kr+ko5~&&>>!DevGMEcz%%37AK>`Nz3Y$;H`x|+4H2{Rqj=+7Mixf>2j=7I* zG;jF*JAv+9=AU^zs#N@cUJTI>{{Q=BoUHsFfmyOHN|yg{;{HKPS(mJZ2~c^{wbPvY z0b#JGCM$|_YecQY{6L$KlUU@jz@T}b86D#dr?zVU{C`H zYyl)H6_<+uL2MKB-(_}*8wu*gqZ3wBwih%wSV`cbP`TQhe#hs>C4ggz_mKO6`3Zx} z!?OG{OJ21E_r4Q9vaZB}<39fNgZcXerT*s}9u)c#7ce#&c2{+So7suyMP@Gwcgn-u zjw4EIK%w+_st~pNnmi$`-oJB^YMg-_6y!^Msbm1Z+-=}q7fA}rPW3H(%@B7Ab0bw` z)6&-xhAcPBZdNJNf@|NO> z9E@+3?Y>Zfje*?Ac^ysg2axRzo8lU~G$MV}*g6DM&fhO(O@1v3TA(IU8L7RDMAvsK zVD0zc&LQJ>5gw9xB$lBWKFUQu6Yalp7HtE~P2;@i=BK{EJtiJdL=G>$c*$(Tv{3|? zfk6Z1TkhA**U?9G{L?)>FthB|6uSWO=Z=ZL0r5{KEd^+iVyO%wV{G#eq)l9LIBg1?tYUAeWz+(E?k&KS-TG4h( z4B%*FGMcg`f<7_YRedJ=_|t@8KP*vq%1x)_k6i`e!J-G|;wiJwmy2qNoSB$x%Qc>D zcRGHQs9*mSa@%{IWOy88M)Nx|ogzMDuSr++eEu$f(zp24QhRHu?Fc4;OBnXM9NH?? z0p7%M5MvREPqm%7L6Kfk6KZ_)qS7R}Z1%v6WUc6H2Y$HLZS2%sk;B4C1Ei%JNT~XC zF<^L8MILt{uY#e8xL+86pbwekn$zK0R6trJ{ z8N4U+=5+TZ^M8YQ{&VsQq(vD{;2z?6=@bp=xu&Oo*lUd#=KT0H1aMa)4Yq7>F- zRPFkq>&54pjt`{TI`P0*F6sMXz*5$nvUCzL;x;@Nj4+|W`QgJ4&puv`b##*80?~X- zGl<&kZ6KugtrPF*!7CMzI|#cMUWGYO8#Xk&f<>V0!a*ycfecbg0)SL;uzTNibBW@x z&ja`BsK}OHS{ZodMhsEg*nD3DNTvFV&4PZ{X$Wzi-crCgmuiA*6R5+WVBZcYy5tT4 zhRHGpz%*0d2On4MQ`rH!EyzcZCjNG#wc8ig#!BI(Y zCN>?PVlwue;4qpSj9nRD@T&fVoP!WE2BP`FfT&1=pKA920=M^0^qQ)dt7k1Q3Os7I+SCT^2-3u(QYQ6 z|90+}Jg(;_&N+5TM;*%04XU`zR*eR8m}>iLO_;Lh&J&N9EVyUu1EXjMF%pD)|KJBM zuP|~LWM6TpAtQ^WC)t;<%JMu1uYAhv**aLJ8%Y>j<2*^}c#!Des~b77>^YKTPDSq7 z5J&Jhi(EOE-ulyw`=|9gF=FT}#W(n-5ud=X6D_mCw(@!2lLQ@~$&41T$l8?0_^0~J zEj_qmdjjkE>)Yb1;nVIk!9^FP6O)z0Oy-&mUP2St)$aM$(-9BsDiT=%j&-xPUl=Rc zZt4TNcKuJA+B3zr9ztlD6*I#Cs^#-2&JherPThM#h#K|Gs{5DivN+dcIxY}67ENz=7Wzsvwvq(!<+q$ zydOI*7exeeRg%l8g(Mozg2#sjYSU`?<4&@*%-_u|fpq3jugdb4bmkdj6jd84z9=c# zIw06K;&r6f$0%fUtuwvwc`DQCOSRY8S^P0)S>Qr-oj1%_7TszTQbpWD`51b{MTGok zN{h=ok(xk64*s~WQ<@E3+iVNgOp&L6r2Q%J$un(rg_RIy7 zb|)lT-!;5-VB-d@tC}BFp)&x(&Z0)3{eOi*Y=Xzops~4`67=trI6eAUw_yeZJhHi7 zfUo0iTsk#Z?rc3(c6f$vOpXY$7wP<(;gt=|a2|NENgq-`1iTI$pgnWf7;Bx>ThrBs z!mhl^&0Pc{ae0IV_##iiKuJA9kP3<>`i*QfL~kz0{#96#yHW^0mD3S+6~)seMwin- zZ3)3G82XLIS>1<3CS3nM(QmDeh%_JlrH6icCp#;Oi-BX*G&}2kkN!qF{Bvi;B|8^g z5xQ^;KIYoK(KLTzv41Ii9+DIMXjt+8dik5ra3|>v+h^waJA*CCZH)@V=$tXT6E73s zF++RvXbhT_p=}1q!;{C)o`rhn3!aXP)pgV^=tsj#mlGUv{waU!o?2=UGBX{QrYtqV z_%-nb*W%3OX&B{-hQANGAAZ1k!Pl@4Tp0ayk~9T2ZU#J2KjCQ+>B-hQM;1hX2y>*^ z%-t2v0+)+2LgWh2`@n+{@Ix({QOIKYaAvUHl(5=H(p9s}hPYesSQMLOHQafcP5u2V z^a|EDI3cN-xw9*Beo@YqK@x$j_Z)iuobN_qf-NUl&3IxC9~bz(L7T6Av;u8II7Do* z{{4%6-P(-jMjcb7jPMS9mNxwP@s^_bx~zwmLy#kZ9<30M#eUO1VS z#8)T3`Z-ojv!9N2VUZEO2jv5E)20^#9`2{^3qM_$#1EJo(Kw2qYIb+amGLf+{jyN8d-FDo<)i@-V7kf``*@`& zI~UQMQHXea+fkjLF9KEfp>R5QrNS})EuxRPW+zDCCBT;gf5OW<6S(5@j+w?UpgF2T zQ$KJAID8Ct?Mt=+JeO#k(r-@F%NHt416HPAxVG92-St0 zV?HwVe@JWgyYn}w`f?VNWM+xS{_^Ah^=mg}x*y!anel4r zc%O&LH1L+Eg{#&_T}j-a3S49*H5oxK=ujio?y^O%S+o#x-srCZ-Av=zcb44$HzUtO zyWpY_KE8|yJ8IY$n|e!7ALXFC#Hc}JXvkT4fiL}~4V^3m-&U);Je9|)K21gZoIedC zo;~NksFXKSZ`NLyzE*>xQ5e*Zl;)Ha1`!Uz z2A%$Z{aU|(Z_)5oDDXK9Gnptk1lRZD$bVZ2sW+>3bEQRgT1}tyY9F$49Yz5TZ(Wcl*3P&=j zj=eum9pfqwt=qrkzAPW3{^qq@%@=j{W!K_WchUBJdA23ilNwhA%3fS?=T=^@$Y{>ZFJ>p(;~#g%DF{z|wC=QByC_my;zyMil}`QcIy-u-3mR_ml9ENv?G>`E%;Q!@9DP_6VNRwh;dO4#aMCRNv> zz~@TszMqO|7K+|-rE= zbKwT4#`Ci3*Lul@x#M`0v-=2c_vRX%rf%2`?lTA^Aa1wy#B=Q*VfYL*CL$`AuiQ!q zI1C4HYjuoW_p!3DAuS^Zbfss=rbOrrRblV;B76DDC>KfmyXdW-+~Mn0vc}ECjc<<< zdfJJes}K270P-NCE2x2WddrI|Ty?@vXDE1WHV)fuOAV!N&`$y^nQl16(v_nJ{AT52 zhc@x^X;W0M299N;y9dlKS1B5%SXR^~7EEv34X49~vOSaSDXN5dYE|v19*vy9Y5)AH z2U(UOnRK(kdr(V7`_cDM%h`vow>%5HYEmeT}`u_=kxuv0lCQ3u0{X|dn)8} zE1YXCh_a7buy}Qs<9j_|SyW(Oa5ctue{sNXIZC+=A0V!%lgyP^xv8V>!?bjZ2Q9wz2!zRQg9$n41W%zp1|FmD4yioKcS?;XD+vvKj()5zl|)Wngm^#I7d zk*Mfw8aFK*w+gY}LCGa{!uzt50Shqb>8@512@@~)S9rq3uhrK(zU1>&x4c{%T0+kvW`d$tv{SN<<4DHy|;ZDl?dEDQ%{t#~y)+on4jp>?1I?Zr>uxvT^gb4dXH z<@wtw?{xr&dQvAK+V`|Ep>ky>l`dSRm!|K-CrF~6#Og1WhmG)W`syp z9sTqZ6jYJ<_<;s$kJY;2k@|6%_J$VNimGw zyc8_2yO!~Sfn+YY>NJ-mWhiDW z`sU_A7q4e@B1G-VI_IH#UmT=&7*_m|D?g>dHZDkcdrKC2Q#Gt8lk(mtg&ZZ7zI^f; zq~NJS@$agZjdVK|`#8!?BbS4V(hubnFF>_VRws*C%#Adu8#}&*LS69HB8rsv(D#4;(8!qf@T;CI7Byn&f86z_Z}$DqS@D}na`3MN>HAs$SJ;A_H)Q{B5`O7S^1gQ1_OXJVKCykF_BmUyd~nY!U?+I^mc|9SS5 zSh49NGy$IH(K2IS)&fN_ti|_{qENfNNr@0)QzUhxO;9!1r{h0=|1r9Rnoq(L;`YU? zf^25Kxuf#NzGn;*ua?(3k1184Vrz{o?+cK9;D3NQTkY%rCUT`eHc2+^<;u+twxEX* zD>vtOK=r3D>%Ci`6%)yv&)_#T6yFmN4Xd0-jr5L*5BSz76X0iuzPYxo&}7R`4^Uck z3sN#Bhl=@x++1r_l;_EDctFo6e3HB9LOM`o15}KmxY<*<=&BPP1u$O zKQpOS>(KUnz93T)c|4KTD8NhTa#j5Ap#xZa)sX8har+&00CS2oJ&333zRXsfi zP2cB_B4LU}f1U^PdcBvVg4j7 zCtDqG^F)+iF1r&Zc&*GM9<mE2&}oIlg;F;%P@1Qrr3OEzr!;c|WOh&+W8b^|Q=6e657h}U z@$SD!>K2!(mG1kG{0xPi-Fu*FZ+%0i0Q(WDh~01|`8#t=*u;{6jLS%22omR70G9Cc za9nQ3dmu8T`*`QW_2JNcM2RoY4$9kML5*wp1pY=x^rSQv^`K#h>NbLBlTM<~)k^W5oo5=hxdfsKM%T@VUOM>eg;x^=zu3AH@nPhc(*o5L=s*!#Yf= zBqDdo2+v7r#-xF?qV?m_bfo<|K)>Hd9MbncAQcG}=TYVh-y77xE9(lap;@bMJ;6TN zcZwvf>Szu@lY`ML&~ZQCoUgz?XW!y#4o;BdZg)u8`3jVA=kd4#HccI*D;I&_Wx#XR zI$wbHNr~oPn&E8qDk}$Y57@p_)z)*J1ymH&kmm`MHAs8IeJ3A#!v#L9=Q!gVz)OjK zHcUj>aEMIsS#6`U&lh--#wLK4?y8%xuQ7mbOkgMQi`BczF(k~n*;%SHl)9(EJ@*mA zHORi)LU7+{!g90n+)3&COr0x}peb$~@Vo@HBaKz3VJNGwSiWrfbV*1s8u2*Z)|U6j z{=FT9t(|zv1pf}AWD~1INkmy6?OeP76?dx|{1!WW2~~^(+5!&!;!PCEQ^&|iLiA?@ zun_X98oD))nu9L^gGAO}R02}9vx5Nm!~T)eN}z&@720^qa%}9nx}!;w>E{;@Jhw^a zUkm`Z$a|lPTxTANht6Ou4fI6jM@4@c7Kzkn6E|XpQP+LLUc==&;niDXefUGtQ>uRS zYVaP)x{0ReK>nK7yE$qow%=9RR-A~kiaZwFM;JIgGy>!%Qu5B`)z~BJw;WmXe3Bim zlC@2-EB)pI`5Z#U4fp7+pXizGBW7y`DLKzj9t@g~YlJ5fu8!V3K#-38%WrwK`-;Bg zR_%n(5-?{S&IVVH=$U8e59i@-A0der6f5ZP60>_)B2;|XldJi1Iar@e^#i02qcg;M?8e zKn2g_jZF2+N($Q%i-oAi3pcS1zs6`?NCPi#*~c*#(@eVi_7(X$!qvHn2y zw>!F&z({@;^7g(X0kef}iv?Qo^TzWO!YS{q>H_yreSlPkW!O}t12x^Rh1EF#I<)aI z?8_&8+bjMz#8dBr$zQMRoSX=;+u;L3;<>izqe%<2VZeTb6X) zb~rTkRhZI)ts`X|4 zS@d1WL+C3o*5ODPzePc?Kl@V^{9?5&Az&^aO8v3`$zR4{xuwj#L0Xdf)bWIES@mc} z8+nJ;Pf{F?_6&u0MK1a#B1U0CrjR2q5oOxT&!0(fW8;$p62dhWJ~`GaeT_{atY6hs z?yC|le-@KyQY=H7=B;iJ;1r*jUO^hce0*|1dcP@%lIxW`n@E0${sQPoetA^D&cgN| z0`7*>_(%J)W*Fe-sSQf5mK#FMEGu+gxVP<~={0i3mxTZvk|oq29T5Ba7_w^@0~QKJ z?258){-k2l*e2FVbqO9|FW%HT42fUQ=XYE07LR-!GKe1Dcj7$>D4YH^iO~rgC3<$e z1Y#EgMw{i$_q#l*9pCb?#J&c45~kS(?57eJ^~$C_uZ`=dIApW|;c!d5+p?%G72w{= zci&Kf)@yiEEK&afq+(;YqW}sPd$%}7{B~Sy)#226x%T>K59wt_+mAL27@@fBR)Nf| z9qJQ!{EOiMG}v&51-q7pAR84uZc5DqC;MAF=8USoL(mPNM3asUldJFc@jfYiN4*A3 zFN^@#IVVdtx_@OJ&UU%2a)8E=RxzRM%c~&WB|%7-y8W5)zEcsh&hE4P66)}I==FpbP=a*|QS9Z2J8V=n_N$1+vs6#WaCch8Emo~cg$PHc~TB`rKLxQt* z!v6|(eLI}A0;vbP)7P@h?tzB`icoE_2f(0SM_ocQn?`V2okGcraKwb$(Gt}g=wZCf zk56o^5C{F)>8Dou73d=YZrY;}K{D^~a3LW{=JjGbg`bqZEh!@2ldCj|bln(0bI(6< z@4MgmP+<9y4xo*@~;qep>458OE*v}sPO_t$nZ2{Sq!V@9|+x# zxJU0)IdnnqsiV5&i_n0l9N}V#WFm#B9#tzeNV7Kr*2^_}kO%Dgpdqb<<45!SLhRUb zIwTTo_-+_clw|*(TAREgpGfq(v>g{->aCxBaU{8qlG)+tkj+$UcY~1|KITJq{0oK6*YhU2gJAO zq}+IwhzP;lFB3|aCxe!lYWJO7u38bXb`wZt`DC?*7ll5))#*bMxz=>i2-{Rk!D40P z$}{KucPug68G=@zhryCQL@LZF6@@>gTeC9Mm%bPBga%nV8u?lJT|aCdV)@__e%L4LRv+rDMg%qr^UgvS#>exfc?10T@9)55ASSfOL3SOILs_=qpj%?VUW1o?!cr zS=HHuTU`C*-Vci1PvKdvkUh=s^X3W)jBdl|gl>(~fyIpPm{1*I24A3E2wg?_h}{ z%_QA+Azi{Xn&NC8V=YrG--|tZhlrNEJy24VWD)fyM{NZhzBXKjGeX+9z0ljvJi#lE z?i<#}li(7*3%JnNrH3_tN9ZwApxz2vZU9absidURC;$H@fF?bB&sxj zmA6n*EBrhQ=L@G38FP}kAqxvAo&)NvJIUO2cGi500&@G2P$J}uKb>eGK$U4*GXqdo z+?xlrY)1J2L4}IP0r>jpA^;=G8|#nZZt{QUMg-ZY$ySpgOEacR)Dgh6BWbY(e^ALi zPAGn4b_^%JZ1(y$lO!ltJ`}G!`_vu1`R~Igc_&AK-GLfXu}UP46dIEGD`#BF`z%2= z`F?cFqZ~xzM{3tf?(j^3%5k)H{{pd(x8M}++5s8v#K4MMeqv`{b<28nTN z$0{u1Id+fZ3IFe798>_~vx$j7_D!8G%O-7l1~Uq2Il3pGU5jV?n3$)RMxn4M6vrya z@qF?v(uCnrJ684;qSG+-_*25;HY6^0-dvvl3CNGcxE}sT6w(>m@pdhxuWRnU z!S{zRBN~GGT-!5oTJeNouA#Nf4Niu~?Okz<<)x?<{k6+9>5nM|w`H=T{;U0WZ=#Ew zqjC}4?t1a{wjhaV{(=RUcJC>c^RaP`4)`(ZGCXMSI{I3{*E2Nr7I}Kz zAXK)%LPt1KIhZ&3&Ux zDPV^J4-vfJyXTRVDa0!xJg^n;Km>E}y)tb*X2MxgG;F|d1xx^_FM%+;lQbCKpqjNN zDNw*N=%Uw#<)ln8uKSVu{P5v<-w)CEsC5bfJ&r+)f?UoaLq$l8TrNC7YquTlVspw` z&UajG@bN@8%E*HhqCzey@B&0l$#zj155cF5$QscWq<3zA*r_IRH^7Kyp zaD~+lM%~86Ng??Htolc~@W_*p~ z(AoSJiurF=v%K`6i$Q3S;rRSzeMx9bj#9$nMri>A~jBjjjZrlzn2iV8fpBJI~OhiA}-NZWRVNpb&G#-WGWA~s}_hOfW zn~QJRvM{^FU-N!KC{;3oS**EDzoSfv=>8reBgENsZ->F@s-C@FnZfIx)r}y&Jxe{7 zGc(apQa_pJm==X+Ea-RQ8BgKNDrjsa9~3gD0nY8&Fa8|StU4`4^pvsSOATLUMkD*` z*?t9J%bP;zMB1g_{rr5!G*MC}cxW_V`AJfOub{>RvErjsUh=v|uSJ0O3`NF1v_I2{ zG%7EZbQLVPbp=>rk;Le|!~COy+EoX?${iF^Yp2LARw!pX#Pr|)eeWV9Y?sd12S^MR(EiJH?8D= zJM1LEMO-D@w-M)@DReefI5zw57$|lwOK7JIa_J(8*#k+p@iIq}XP}o#M?u$DWp#_P z!nTKNCzm8e=0}EggvB81BPKyt*@6X`&_2aBnJ+9NvF{CLV6TGS%G5UIxp&#O$&@1; z^t(Ki=Dc5Z!qZeZ&QPecE30`rPkO&a8& zY!G_aB!7K-LbzyN@)L|H|LQ@V)@hzaNZ>ixRcR$y$A_XL;l)SIaQc%I5zp(r1FgjF zIug@6m0w1PBag33U$sf~d)Df22vnFPi z@&=EchyUs>J4lD2JpE^mT{p^8mTQ7L>*TL^8~;7=*;gU?wZOIYEL35FUq0)LwC7d4 z{w&>T^&TF2r*_5atc&>H+Rxz64v~}TxI5}k!)+0Td`qakRk!uy$HX=)<-cSf=eVH2 zzGFV{X6tIiC*!eJJ>6g_qbRo;BIMNb`JA&aEYoU9Av|HixT^BtT=)_f8z zaOrItF9O)oO(sEV_y_S9!v>)TI#*Z-#VheU;BfO5!hfzey~b;x7{WgFL57^Dkiqv` zt_j<*264~#Hp8(DddciV-NX$gI|CstdBKp)+j`Te6*@)j_S_>y8P&(6Di7b?Cs?N+ z*h!-$l=6#VokFFS{MiI|upY=M86|G06>-mfi~29Xvd63=3SxdK%SFFW8% z--$U;lIcy?c61-d>$^Hh3T3}m4wj)cXotOcb9_;6Iy3Ud5h5`<1%HhIkKR3ABPDSS zBVndgzD@4=34l=c>5Pf%F9~zKzKbYHWuO>DZ0n&C%9iljeJw83l!&NNN2;CHDXNCp ztzC*nlML z@@&|kJ@@#0K7==|07;q~;|nGsohJcPcesXjPI#C1F&xqNqrJ;ko@YjDj;dDBUH#0Bv)j#hZH50x*_3 zR-HnLQ1W^zqF)}y`5nUaW4UC5YBN`b{XboNkqk=0?9wnEztnv@t7F)uxhm}V~BoVJP~C;!>=dN^YaqERDy>cCV*V1n!2v1)4sK><^C50 zCqb>8#42Y}hxwIC)g5yMbuf)uosC~IhO}KF*I|5sn4QnlrhS#vicI25)@CB4IwMuv6Y(Y!jo6dUO$Lcv)f1zJm!p zGD+l*NjhcnFSFDpliQ!>y!l#IkmMJpDt})ubhG&C3EX9f(B;6Wv-wx|_3D&3r_N?a zu8t7~of!2NEVF8Uim3l&2DAT)>j|1vH#Wn!BRC<(TueI52d2Ii2{U!qn;x3U znb3Ix_g$SuK2oRFL7sWsux>$cK+EIE|LhHn6)@aVM8vcJVYBl=88YzRMbyh!yoz;FW1Q`ed{sy{Rwi zfy;S*_vmDT%a4UNy#62!09g%w5n?I*dW=Es943-5r2L~-2+?1%I3?)M1)|@n?bBn- zBnEwx9d=|xEszR&vGsWWZnz5^3h(}n_iKlay(BzH_+F#PCou&9isDDx4DBk_oAV2dN>bz%Ax&dR`mQ@4gfai}Bc$@h*9Z-DjcOV0o z$B&*fXs-y`stp0X4S!{>Ub_6{rTHTE0Ml>j1OdR(mI82LoyoM-DD}tyth^yZmv49 z1*VK^-skJKq%f>aTO9<3FX)kvV~8Hop)vF8 zMG`(Q+MWVm>e{G#g(Sj0WdG1z_zYz$)eX@5sem9ZW|E(WV=yQHz|F>BL_ak#B1#@q zbi2)Qft)a`6%ln~AtAjJ)Y$n9=0ZqH8*tfz<5U{2Xqc&?w6OLa^k*F#uA$M@DnlTh z?g%?IeoDZ|3eFyC$!NYqSe0N*z!`fk-cjkA9T6Y*^=%5I;iH`8VW6n0EEw*_6oiG` z1r8PLf$nvs*dEjsz17`V9)M2m15x7k6b6cx82`#-cG&-{Cd_sU&OeeZ-%p()Q$5MAM%n9G1ZA+8*$3_5i1QVjQmUt z>PR7B^e!v#pP}kMN-==)Mo}fPl=$X&VxM9Ss{GOJPr7}E@}~c>s>YYj)COY9%aql0v`e0V@c+R+O!S?F3=EHGnAc+ z5~zc-5j>Z`=<`H^K8CnZ_fgp<0m?GN9IOuIgPPDB8nAb^!mNx4m=L?v_Yn8n5E6d5 zQw_Ng&%WCzMMTw_iry<9LZq)=rXZq7=-qFXO!(5v>2$XV*!`*whpWpJpzsctHySc% zPm(vP^(Nux(IS70Pk>k{)X5+ej4BE)VhUSBuM;eKWA*q7pbDY>TPLOYDiKLaRmp2; z+y2TSS2DTK!@^ua!&eZ|N4BULiR%*;y!HZ*s^3#8fc^|jG$RSi$_9E&?K|d{82-VV zM6AdRlt4Hrqr+uU+th4F6D2;7}6FGPEWf$xHQ)D7wPCdO1!_7$VO^$yc z5Z1s-m!>QzM2xCny~1SsDwc>%mt9n^i7y*HDf)Cg*lBrP@^8aMTd~f#akPMo@O4=<&(L#SBFpkcm^bbF;Yk% z8|gm>YgKN6lV`zLE}l)~#56Engl^LIsv#{g=z{-E-g{HsWKc_f#OJy*9hL#M*CbC( zCo-X?hg=+-Ey&VmRI`oF?bRpIT^f`T?k> z_jE(GKH5*geyZ(azyW?@^2itrjIAo{m4$6 zL|!W>dKM*&Y)MKbSb>p@O*%gBaTa2RbsB%Rd36}#Xn*ha%b?whV+)M0<3+88NGQ~u zTxR$;g6HNB&F)ka5?=EJF4K3&>vta9VA^H}l0Y*%8?xC<^o&UW&MDV6?>O#Y47d3Q z-CRSCXa!{$gqpO}k2$jtXX_ovPaHU#0?4+fUOnRTQ`)=r_4EbC^J-^B&`7pLujxJ_ z{NWQ}gZ9|>ZnB`att$e`N$rnz*7JKTHlfXk-JKJ>qDp;+5|ZkP zMx>J55(v*3Bm2?qNblpHc3zSIIxXE^(FZtSfw)6CB}mI4H-m)a{xj5|u`FQryu0|X zH;i~t?)gPR0q9=j*NxZ~oT?|Y*gXu_6y`dc=-s6(_nKE}8_9q?z=b(^fzNFdbA+bqD}1_>My-N=#MF{%9IlCUdR{~(h--`+lj zTeq8MG$Xd3icoZ_gU{wJbQ&?~`3uDww5xP{nS4dUOx%^1n;kL;)!P7_8yLsbWRLq|bery31(kqzlE)eW-r2 z99T+ZM!mLfQWI03wc6VSgg^hHL-LuJ&Y1c>doL zxWckw&h&0Hx)o0Ehf&G>zND~=NQ23NzQ1p7-^?}X{c;PBV3jw}VH}IGz62#p<*zIb z7DimNuE9D?HG`ty8C34(NlwP+Hiu1;wuGs*iFhKgkFyhs~=a9+Qer;C~}N=Y=}BV^C*C9cp8NH zD$upeW*-*)eePgCcE({ZiS(tzc2Dp*)l&5V<%1nD3f(ok37O@&>P8yyPFwdR&aCem zwEN{L@UG$wKI7dV>i$K`bHUh>e4b*V%O9o{^H82QOYC3fIR93OBzdN2pY^8T_h&mY z68M${Vk^#W$F;&O%c&n+u^h?*NS=kn;b$WWB`G*lqmfZ}R2M#=eu*)@I=% z=wv)k4oN&!!LubezPGir@c!1o{uwas9+RJ)W;MopeLz6?(aUDkDqCc#bTC1-vrwR&?k6LN5 zxc>y`^%P3?{s^6v7Bn0pNsY#CCLpZT$}iT_2`!?-mqAD7*%PXXSFqg)nc(%g+}{XI zl|(1*HFR%=KOV)EST$YykZ$DP6EKk!R-JHvez2nEa{F+JDk^JXfoNv4%d0#aTn*Hj zb_?~k9(*z`g_zTxIp)jMgH&_r)tA{ZPe+!07eXp?a{o4%$0+BC*k<3rz3mA^kVz~x z@D+B{dvu<1>}$Db{2=gVnlm+?qfl4HWr~W)+|^B=o2hzb_+(jrHrC}PUQ@AF!)vys zXyK^*y$R+v^f{94wY2Wv#pjpI-_uf7?uKitOEouq)KmH#(jMr<9L;px;PkYw(g>O2 zIt#nrv-4uS^ORgaqt6zX^I{g;J0BAIgF`nPN8Mg{T=FW7sAsSgCevB5-hgt)wBpRt z$HQfAT#;&t(@xu@&P(*o&NM%s3eH8V$6&;23=Zr0t&kXt=&xPkOE&en4@x6(w)7#B z760KXS-L}Z6!ZF4`cz`B6~IrEs%|A%M05Y}Wzo=M<9kt2kGDG?+D0U zhFbTAgm^0b4WvA&jYgR?28=xJv$n+1do`50eser7Nxb%Cd5M1RZg@akz4(Gx6-IXF@>&h8Dg+FI`hU#rH2vcoKoX#7UYbA zKcsth{YKaP)K!bN<5uf0+wRt*z3M`MnWgpXK`tI zvQ2HWy~jawBRKkU%cF)!MBQ3$Un+B|8g?_rtS)3~eoiZ<2{G`P%Byp3$nr1IQ8`3z zwv<{sjh=TA%KFNl@-Bd-J67p9(y%!|B)cfGm=68!N8qDe6~f9GHmvu(EO#gaoI2N7 zl)y{lX(dM)npD&N?s<#uI%aXbo6Zw zD2fQ4&ClGboY!w^^C_ACyLKzsVPA^@awp?3&wRtZPi=a|wj*^EdHWa>G2d}xCa`Rm zSEC)lx=?>N6K=ve7m4Pga=fZdALzyUU1fz<>Tm6gOz=ROaiz}LsL@)k48fvcDiG~Z zsC^A*UcW;hf!}@4xnloNr;{28VO}+MAl$=h?Yuo>2UEotWAbDvThv}%t4Axl9}V746|MG z>tPKXI7#9N;FXO4E1hOJ3oe|ePMGbE;H2W*n6Vrq$-RlfBIHiYTptCICy**{yOKP; zYSvI&8?5km0oSzKZNqXki5~}^Q8U6_bj$21!qFe4eRx!H zSWNz`o4dCOD?uy-&Bh#z^4nmCuH*ay_Ko7!!lG~z;WBSM zX@+0^LuN@GQ5@pER74RjW6dC;(D5D1-we}lLkvimlT&2&-tdVOX?zt7w49YSA!@Xz87l_7(+4aYgx5%g1FUh z4r>U^N-3bu3$H$Tn6<-T&D2<-ys@K(>wYyW7CFbWq9lh4yn#@fjXyf(TmVN%)T6_Y zFSu6tQ?$Cg|4yIKo@-D@H6}9<^%OYINF2+d-zc5r%Xg&&TrEW~wVt zhbGA!pG^d#GQmF6UM-BLN4$wme@U3(N&nHY5rm@UX)qeasL!S2tacsxo^y8KxSr}B zHxd#Kz1hDGo(4Ph2hLgty8%Wz!9qtP%XbzLG4W_f9drIVm|YC$x!d`=iCzL70hf8w z*K*nP>_8qaGg|ufV9;KTW+{x^t~U^I@QmCpFtYR0tkjQ)SX!Z-Avb+FTmr7Q=29?9 z^%YB-QwE{&D$Y${D}X}x`c?qeoXSY_-N9(hmCG8HeVi3%k;C_l2l#g!l7iQBgK94~ zdD;~Cu=;;yX)+8qZ=!IebaY9R|MK4BN|UEgMg1>7eD4vz?&%qmNFcWUb}(99Yo7L+ z-rbI)&ns)?+Mr8yz{Cqtv^M`;@$bG48U0m^F$&v3zR@=>_pJOnj8aMo8QwmL57 zCM0?vRd*PA%zr-^+|apldw;N^jQ#a&t-z2J7thQFOvCikK7tisc`YM)RFN>5~;1QSI^(cB9uJjdf6=)$`EO$cPnDGUZOE6n%b(fVaZETKMJ>C zc;ojV{7n0LA^r-&Av~^bjLKg*R}`*$+W6fV8(;{Q_Jdm$S3$gWM?1|n`s}Y)zKDXd zbbIZlj-5QO2cA6$gc>iIW3S`-e_jfg(bitRc`2&oKfft=t#Beq|CSC17Q}&Fk~&O7 zv&`LOI2cQpbppTfw8_v|t&w3;kyJKJMXSXQrM;;Q+-K`JO1|7>wFmJ>aj z_2eHo?MN6W=bN+Om#^Ks-&Pa6`~4|d(eX^tu0xPR^1O!@`mp?W1vPXBUZuc8%8$s# z!HjyUhX3XYPbx3hV?v=H1on>9_^{wJ?LbOAaIdCc4oc zJz;I^t|l8#!)KK^MYQ^k=9|1&ack=R!BRW%+YzxWB-QS?WHNK;mA(;$IhP^_KXAfx zsGU$m*QcfKoYJ+{3AT*TMw!(eXd(U+*N zs6OPT@$w>p_kf*p_caYZF+e(oLpX?HVW4s@j@l8#)+bLQ5cXyln&C8da^2_W{6}rB znim``RMnz#7YiYU{Ti{kbUr>}_}n*B(d7nC$UXUO3NH9@X2SE?mVg(wRE?NFFrvCH zPC-C8Lgy7?oo;!AeP4@9O)|?|If#O-mVRHWy3`dD(D)&>c7zy?bHe$p_~@R%TLW@M z&H!ZNwdaY>bTtJi)>FJ3Z#Y^ZPV;^Z(Xr0v^Mg!Ng(;`?vM~c&oJN;fHg5L;rLNF2 z54X*1P7Y<>VSV2z2ce|GcudZo;PyG1)&Kn5!S~FD$VU0W%V_o4&h0;=$2WJ(Q-0=9 zFa(`6x?9Tr2oOEtf5$X`BoEvX-T(orD-T+AzMi!k9({Aifo*)*++=_74Q^o}pqS=y zqk=^iBX#4uBi?hwVHU`OYGZcAAjV!4b>h`m3`zS03s& zv&G*}>GK1c=FMMMyuLj`C>iJ8b3jM!mX4n$_x-rBKRDKt^YG)|d`Q3d0uQv-5V>l4 zV^$fB2U8vi2}U8tZyxlA7;iv@mt^*MsOQGE(tlDC+)&zftvD0hk{mRAS!r(Ug92>%`4Y;- zaPUBjIPs30XlZ&>t3WREX+j-mvld25?+jRp_^pn5QUCVSdj!?oLH#!CS5>mmq%Gw) zrB_3fF43mGL^N%1>iog2C?=j11iYwECD8=)mEVb-J?5!qK2P9f@D~bJoFXv@c~V5M zc|M;vr3sf&33w$}crdsP7Nqg$L$sdda!DrDwVceno~8~@|5VPv7pploco$cpI%@!K!6&}+H#CVmlmTgr ztpN4l$JY;Eh9KKal=2jr|k32wmX$2K(HjNs(x?n!#b3+QPXQc0VubS1oU7&q``wB zKX6CWcw%q=R3vHM!{KMrj%Ih3G>*SGlwq%1{r9E(qvP7EsujESeLSR|AG60QoksGI z?Ew1?;&#Rt_B#9Z-1qXS+Oyi3O0)@&b#M(0wOO8x6)Nb=%TgUU>3d74*j6XT#tK5C z#zPP#kESV4w5VHui<^B#o8or|2HJ!vX$e3c8y--!AF>A+4( zeqnt5<*aHuDrbZpYR0*`SOLq@JjB+|p$mkw?X#G&jCF|n-XxzJndL5`_t$50(NF5O zxJ?X2YR-CCiTr=Q8@W@v7b7T2*TTN1y&MDBB$D@+e*wAGl<(=>@c8^FjDY+GuLMu9@&m_Eo{*4p@7&QaH2Gs3-_e`2B8~$ z_3?K@C*TXo*IISF(OlehJ$j8rU^ZWK8HsuP0|cHz&T0KejfqaeO5Qe3~4Qr9*Lj?ajDB zs9m*lSlr$A^;{fW1S^W`p_!}t`ZjIO)R-9|N ziEsSDPDuU(U>cZ(G4lcRUJ0>{*3+Wz=O^$Il7^md|DJy|wyk%0WC3+c%LZ_78bd3i zXQj?J;_bP{cq5OkCDngK%@!|FYyHr{;io9}>`!=hBn_%=OZxq_JT;Ld>ON_e z&=9fZ&z}{`bkyHUr2*GY-(N7thfa@JA^{sF;g0vH}e35vtuRi{! zS1$YJ$LE@G30`i4x6M!eS2Qm#f1G&UkJcfa_dw5utmRtIr;G&5AZV@{B$gg=ot=3- z>ak7tP%;i7H=i?vVkL1-hF8-bve zFXIFly1#?`(SO~?=l2IqwknT3*K(ae?6@6_@~S{_Vh7dPPsOR%k2AHY0&9OmUxt(W z-zyu>Hr?ArJmsRjP6nmj=NBSi=^P_J98D^1=_)}NpjcReZ4}7hZfmH0EjaqygDc7f z!bbe`X-gU?)Vo>GdH_G|U1nXO+yK>R2IF^#vF|P}{p#}(i${GE^iQsReRjoq;KPx{Ki!HER+G+WBg{Wk-N#j2i(_li5bZbVK&8uAD zLf8K=PJekkVPaW659L^d2RP&`)g%&TG+A_RECw9eUsG*C@Ya}CQsxn{H-j=&w;_A= z*n{!cB+Meeud=K?5hVv~BWJgczR?%Mn{fdOd`l@uqZ?C(v_For@Hx@KxDP#s7tGpm zA}>?h=$O!%$5K*-_`KEsz;Vl6UF!O*Yph6owlHqq!9XwdXw=||dg9jgS(TwNdptsZ zHO%cDA|LXzyU|&WQSV+3P_g1)vWiB`A8|d2Tano?-9e=CDg()VENNBup4UEtvm;~= z_0anRG?JSVzpR=toN{+-7OSe;FrORn5`pR*L5l-HMKSFa)vUU{uCJjY1ZJZpH5= z{lVbHN2#wHBAeBTuc=72O0)r=&K!w7F7sD|D87J?pyCC3xrG7vg1J^Y^<$#t%LlxbP}^L?VTb@VZ9dr!s&Aa8 zn7toeSArb_ueJHY2mrtub`a#esYa&Ov)Z(9{$++PtxIg^Jmzx|xfk-oIwD**oAXB=ChK+@{aX%2??I?Hn^W8Acxinm=T z-5z}tpPh7d5CsAf4j?CvC?wwmjSX-|zA~7Gz68tPkj*2p1VpFo<7|*C*noD1M~Wqv zo?f*$Zfpon=>FRS4TZ~i<2lJl_}R&$JcNJ7n9=c!5~+lXPv@9(4J}wt7P11lEFJmQ z{iq>}o6Ra?Ej4I{n;Pyc(h@5QjLps*J9{1o6H)hxVqoQvKYD+(HHKSxk@Fv2@r1ft zGu*r>FMXB+{2#Sgrvc8TBz}UM`<*<-G^=EJ9 zyu%!Q$`4O98x6;R#&(R~Ku--P9+9(nWVh16x(xJ>;dek#t1kJf^A0gnll*@qU3)y! z|Ns4T)5Va;HHq9xa>*qvmiw)cE>teL-&4q9Y%ZC5?n$LZN$ylCxh&>>k6cnVbH9|$ zM%&o-{=NGC{_>D%d%wNj&)4&H&Uv1r1t}(Qa8In9IcqPpgCAnft;ZN{xLgu9=w1zHf1=f`8X{t+OvBoZZSpr2)rB{_v)hp5|n4IOEinYM$9 zEbRpsuJCL;)ClyDN682mdSAo&dqh1bcot`TZ`J%EDab;(nCe?0$X=VY0Rxa&;>6{F zp>R^?vs+4QCz*)-+ICyj#L8x$-yMh$kMdZqRs8iR-lq2}(8d6V?VtPMAo#=NyBC}5 zKWw3RG7FdM|DtZlqDGrogAv@(HB32aD*DkvJLXgmr79HI`k0k;gDg=W^=z%SN%^hv2EXwnDl5hQhmd#|#-v*1GRmzSLEocx>P0vhTG! z4(zndl_<7MdugtPufx@y$(PHvko|?i;81v|2{)eI-`%Z?LiksfFxOGKGI{~kM3(2X z%NlTQ(uv>s&YL^pF<2TVtFc4^hWd&@Z%(#7=iq|(7exjIwLsGcc`XXz35)f^I~zZ`(sn+*-&{lV7r zPABp35>ZH4AQ97A+>nVP&M&nZ%nv>pmeZ5}eOzkVuARy}|nrregk9J-b)G zcUFB)ip|ZwaGZ+KKBGBP2{b4-FpT#Z%lj$=#>Btkt*s;*E3q#z9u;i|QNiiw2#0-`^?`x#Y!uKWuRm z($bC%C64TZ{hQ6#e-U$#jZPoPU=5qNR3+c4nZfQeATi&AlvJJv7wyWkF**Os)4Fgi<)Us>ZYG}wR9 z;%)5>8<-a)&*S~jjTu-D!>Lwr!G(Q5oE=HM#Pb4GO*h( z%!nf+WjwneR=b{~TOx^4A0{#>i$?Z_OMSx&OyAF7=+RW2_p-gceKbAioWd^i5inCw z_Z$=&Xrg)mGNTUW0gVeZI}zc33iu;1s5UJW1_OZYMHs~49{8gryl^ZJw=@EJB{Gaw zrA+t$(J+P7sd{(EHe&wPt)-Di)P`GqtKcrsz1uo9Qb@c%t{|E zhN^{*Z=+P53mVvzqo5RLMY^}0m6W%(nldgeN;#BTe{BrHvlkH_EdhXq**pO0X04~k z8$+AS2AIxQ%nRm!1iob`bYBVZiQiuH1V-(2#NwO~eG^6AbBh;{o-Z6}r+3XsU43l* z-X6Lpm>sqZQF#de)wVm%N&T+P9N2bVkQ1N|ZV3ts1AF8(J(#Z?r2!x54>nc95B$8z zBr9=RqfeocB*`{>@3@#*;EQq^{~cg3Mt~Sma;}^{d_;NUu9*LdFw$etTh|2_ zXl)glmyfJ@mAs#QMQN$(t}hsPGaRnly~QZX`CWi%*PnHP=6f+m*^RZ+CzR{6>^IaC z*VF%7jPW*d%eLQCOh5ki;<86O=EXi4@TAF6wo~+{5iT{CdAPdEum5&6T!i;n5uim{ zEsSIDYOhN>Z~VC7B+2B33qITkYlx3B8ny-)#v0pYvR3`SI zZWe%lY1Kprt8vBuuvRN-8!<}ZU!6z8_wuNC9Sh}UdOsbf+?jO@A{qpLY>fMBGxP@e zfILzHvMuoG-O6C5|FtOP&6~@Abhl#{^Gu$-|K^y<>N5)Gtp-{q-oVy%|+fI|ICOKKIZ7~KBMGR>lwemXAku7kd_(%epRaM7}bW*Vp&?)zb{XL`Gt*Lu^-+AD>n^J zz}Zv`7G8*zD{{gb8P3B4BmRBRV^pr-oW;U9hiOkpq!Te2*ZeX<2 zS<(Bz3D1%)2_7CUb@0C8`~p&xhRG&5DUfnw&v)tkUO2`$rAwI(vf0phq`QtHyWcrj z@%tggwV0F7()V2O07zONgIs)=bQrBFY@Oird($!tec5ZcN&b96uW(^ppqc-*FqcFb6^!7`NtgQx~jfUA= zPhyMm?2bR2ooLm_Mrscg#oB&7Ily84-21m-wz!={G)Q`2XrwyZcs!!Ors^3Lm2`PK zf*9fL{B$+S%)@O9p}=BJUADCoTZw`f$JiBf{)8j>2D@}3Q*bGgENmRcwNE8t(dpYH z#d>s`XDVJaVU?5ddKIHzy%J?*2?jv)Bd5E z#!1rRn|rydQOv3JZ#$a()ug5F)l>C^-FnDDP~+bQLStO|4(c@wccMnQ**|C`on(a^ z@&w9_H%fSj{g4NDEiyf2Wj|{k^?4)QKI`K7m*p*+=il@o`PJ)4P9MYf-l%Go=0 zwx|5cmbOK6oR4IB<##NotZqs^I(IsK+cVB`C(XRcB^Pt~;_2xi(vY>Q&>6OqYH2fi zk?$Sq~Yf9lP3a&oJXBtU$o(O-x;^pO)TOz?UCI>++PSg<60#9QS9v5qUQ~b<#b0fA(ianHukpR{ zcsagF>+GEfu_*V4Is5kq^F4U{R?PBlHB&P0#sG35mnAI379yFr24W(lozLxpeqRjR28E*xxxg@?71UF4 zcL3_pFi_P_NkKI4-LuhfJY|5P?j!m47|4BMdp-ydKE)cSF3){WcVaBeBh4O%3~j%b z5bhIDnyO?~u@87|{iIy8+RjJX9z`QR#be$ve=0Y3Dr#Fs6T|PJyayr|SqJ{i4KU2f zHWd!{se0PcJQ5He+E!{P7oDRU=wVoNq zYAX>?+>@2vDVPIG$i=--hRGs)wRqPcJQ|Zj?6ka3jTS-LMR=nx^wxf$I*IQi@9@Q- zn%*&+HfEd5Xi?l!$}`Gy`zA^`vRQI^C(}Fideq)Ac!=O0kee*ZeQ@E;Qq-hFcbuXe z=t&>GAPfWIdY%NxB!=5bnN73>ZF+|w^QiQhk1e}miSJjRW3`TvoS8<94Vh7F0eGF;$NdIle?Zy5cza$S-j}r%eZZI%Fv*EUZiw9 zX6vCcyPAVe5z^aX*H!Wa>EL_hh1JjE!N-lfgM`w#P1#O`g;-A zR{7Y+F^lH`tH1wbe57|=&yL}Jq}+0biBECGlDk9_fAXm;AMU%p9%jkz{DPru&|7_s zAK2KAC-ttBO`QtGg!IADU(dB7?`KF_XW5*NCFU>P`0l$XAn|v^Ow90!nI8Ww59_8g zCwdQ2oyNx(CpT65wrcqRzB$AaBt9N@9dZ0!TbAjC& zY}HPttmu#uPPM3~nHHnCXFKF?&t&qvFepKasoYK7Pr4@UK1pGe?3+xAToiid{%!#- z8Af%&FFfU)ANn-3ZLG0y8ABW?!JP97cG9Z#aMRjyrNx3Ru{`Fk{W^nosOs;QWCVkyd*`R=K~ z(6-~Pu`?+or>|TANI&lpbenJOzH83UzGK>V!DH#+}3qv%rfyIGv#B5MHtz;hZ`7}xox3Gl|l zK_d-y@&zKw_`LTh(eij>3SCcC4vkdqwcEj2v%Yg>bsyJP!a(RH5*=|Z)PS>A_53~f zYg5ygq5zQe3wx7t1#6=k)cjsg^Ldsh@v>^NCih036%o{#xQMs^y^f{nu~}1_tEn=@ zWALBdxqBdB0nje!LfuvN3v^}Dfv@TO3p>jHO~&q%kKn9fH-!``(@8-*q*pHtEGdm1 z89&@A;QVL)WPOIwsygsfd*w7*b=FGv1PraJuEGIN zYC~FMTz6CA(+8}o_Z}w9sYXaa7&r@0O^(BHGvi~*f z9Oq_#)_H(|uS88=>xkX{)Wc4Zy_Es@?7ViMzwju)_%?-D_EX zMxE_CV^4QF%niTcb2NNj>eL!HWd?Qio}C8OW@<;YDY&e8lU$^I)#pYu5_Yim<+L=9u2w3HJ<@47h5Slkyi5y)ykBKI&M;`4ZHLPn z`o-NLiyH$T|CXnh@)`6ie9kwoQRh2zJk8CIHFu^)&f;jrb#l2`1} z7|Km~G{61dF8IWMoj#Qeb*26Uin_s(xPQKDC+-2>*#e$f73hN$TiJTJc%-@`~(vo0PwjWGsNegYJ6M(UszRH_;(KZU> zVZ~=>8h;nN2fh<~(X!D{ zF!O90NETKAbUsUV8G^ZjkS_X^IaV)=A)KTpQNU=6?t?=&WnmLMU^sHow@73Ui1q%) zk&ZhDR_5sgyTT;F0jl=8vnjag6bYzoP%b66Lh#vL+tY?8jrETN6K^dv)AN)PW& z=&JB1 zwP4-@sZmIr#Rzu>12YtQqqPFp;+~#I;axJ+c<0sGY1}7@2vL`fwjx@2U8&3WZF-;k z=FIOH;^GjXxBTzw8ZAnlXXeBMkhMg61&k_h_}(~(R@L1`Y(L8m7v|uSVW>6=@+__3 z8dNeFf?jZu?v35khdS^b(p6P1fB}$e+H8yJgo=;Rrml1pn2e8~)Zn zQ()@oI;!@H;f1F6F6`{n;EoPEC6Xisp!Ul4O-YbX$W4u z((ePH>m(%aRfwHb=K?6YP{}_$ zT-M8~>|x-0I#+ks(vaAaN^5nD@+;IkL|1jMgSY~;7k|;ee(6w#gNWH180$N^T?;KG z-DS+GO)-dSxb$})x6~$d!t^Ob&&_sIhbGU-$7*tu7R?kQ$FDT1$dKy8*he1koys1;d|zwwivam@jO1k4=wKQf(XQ5yy>yAlLQ@FB<7@ z)OD>FL%b3lxyWrjtlvB;=v($uVl}4Dv(*uKo&=odl$_lwwnJMnX~;+x;t58EqT#|C z;50ZaEg%5lup&rmLcHJ%+#Eic9%;EJ!TWF1Xf827wsi>dU4FSAO)E%P-zAaDk0-eV zUIoLg*|z{nj_dLYB~vGd%!8{N{GQ*P_!RgtBDc_}CHd4nA06zBB5 zo0}L}C&CDbI}_VC{{%A+r(D<=*o8jqA&*8Dreda_DMm=g^~KIhEj5-zxVaT^oWBU> zwp`s8fn&9Jh#aFFLpHsqVz!_!rAqmGy5Qz9B#O~-ld0KNMlWwH6ms%m-_E&zMdSvp zlC{O5#TFq&*hU7dLHH0FZ)0Lb;t9{Y(;>;~K^*|>d{y#BpfM$&`0rBAey+H=rXCc& zcaricM5^7a>_0plec@i$wthq2IM7+GbR>ob17v;N?8oLnD(0~l1@gg@wp%S0^)?e_ z1=!BSh7^C*d`wh#9e4nw6il`koSMrv^h+>Xww2k^P4t_@xU0mL&Vo1Vl3qbdL#fj2p zaLM;zsA!CJ$uDn^NGvU_?=pC>>y1h1`!pI}QD zDgU7JA9#Xf#RXgiW~L?^i9y6GApZNG@-&k*o)pcIrsAR&Tr5VC+9>l2Xksxi<7a4$ zi`~b%a_dG+vkMtN-VfOj88?35)S#2yZ=-Eg@!@&YhG!x|Hs>hIN=CjQp805?hpiXT zR3);@nn5NjqqINKd|t}oX@4W|7CylGzJ25%J^t?y9OE@~x0+-nPmJKNc;IS7CY!8T zR0p7SnZF0wB%ox7yzo!96|F+WJO!GC5r5FG>iF%1CY*p7V_`WYtk>4?VFRddCbTMrK813>VDf@lfY1-A@Q`C;>q8JxkXD)I?GU;Fauwld4gvh_g^#Xtkd+nTd4 zxDXOTqzD3%1#8)PW6>$AXZ!Ewwp8e_hFPZmQuP^;Vk9>jq+*n5(rCDG{L(W~jo%{u?G1N_iHiZLC4>9fAqk#sG58y#OjaxqoJc@fW*-2nUf7F(C21bM?hHU*p9!*^w=Pz8krF!)*64$0Yc0uaY`Uv#nT$+yjH#oGO8V@2j#3!P7RY4(3Kl4B)AEVb{&J(^@>al6| z7^Z)53bWC=xqBU!S?d^*?S%Gwt*l)}Wckg(tU&pD65dH~q~4#6^*d zGHl%K;6-gn0=~X5x+5pncU%1cMUFcE3UW`+X|yVaik@$QGD_+~-5l96h|6GT&R_J; zWo)>I&qYaA6P*k$_AJz$$lQl_*?tHzQN~Wzyk1X_a}z(?%P!Zm7Xea{DZ^U5Oh(g` zX|)@-Zq;2;z5;V^4~_qK?R&q?+}(XQ2p8RFtqP4TA1#&g9&mK^B0rVszsef6V=GO>|(BL+i2Y9{0C!cYH0=HQ|p}GBio;JmA8DyCePdW2C@8+yi-(9F3}8>r-Jh%8{@?|>OZP}r{D%iNkF<^8>CUYWHdgEF4y$*MW@lj10q zUhi2%_&6Q!zbnZrR-X?~vXIGniolC@AhI7qAw3#UDe5cqL$%{7h!HT_ea}*L&dK@R zaBM|5sR-K2-Qnh;0t9F7pJmvKr~>hDQo4B|I1!E}dPNt5V0mj_$G9X?@13fauTcGJ z-xZOYLn2IXbwV0G8uP;SGZ4z*oZj6&&0fTeK+7Pmdm49ZGYQ?U0_qmQ!1q08mBd-c z)sZNk&T1F$J7h9Au*+&*maWi!1ySjLA)d?!LE1uF)!o`RL2XU-gXsX_f|zu`Gy*Rr zTkv4LVfE5?30{y7tyIqp6=h#ZvSGmRq*3kRt!mMN(8KgGd zK!p+;cQAN;S)N%UB28%$kfqQv(mrrG{yc$CFoDlH{>2BGtgEkK@{ugOCZ*vk8d)dy z2goWmpRrE^!B?Fhmf6(Bwqmb=V|U;;crHU~AFwmCPr3^ZbII41#Y{jWm0h^Oc!Fut z&HO-edI(chgC^BWGIL{}dd}%J-G|dv=NVQDq@WlPsY}ta0W+v`Yr#!)ws#p)%=#yBHA?9fZ6Gn0=s&7ItrG-!Kl>%d zUi+J?z^H)^vvGoIHt=HzMP(A*Xyy`<}u%W-}A&ElHcgq3O zGCV5N{`v&=>ePZqD`kyZkBEWZ&+e-K=UFD3El}*NNj>494sBl~pJkm=&E+SD!P06wx=8X)mD=Kf>Oa1(#1%y#=15 z_)d_Wa=OogcL+M+toPjR8{|uSoCl(7eXTj$uf-zi^FXoWRqPCi9I$b}`)u@;n=Z3! z&pJfz_p!Ro*)H+{J~6tDk}*&u*^HP57G0RsuN$g;c(hl*Yzg!}M*1Lg5UN<*3uNLt zVrEVoe=#d;8)6IeDtzw&=*XZVFa!@>*5b;1IZFNmegAwMjf|n4#>0RO$r-+%ugCCb z)S8^|@aFe&)41U|ai78;X)PG;lxaZ>=Jb(M=1^x3uZS(Ag^yo>V0~jyxUtGrHO_d} zfXEl>&-4`RHQh2lkaN+DZZBO=U&4cEk*M`V1I0dQJdmj$z%S7w&dzB&eQwx9D`IZ| z^Uy)~l$ZWfenX>gCue`wk$Y1M2GU^-$Fw8$J8M+=^43!Ma4dc|Q3U2cT48d{l&E_w zcon^NDgky&iV!jXzi$f3k~{h4KRa(d`a$)aT+22YvHjpmD5wFB28nuV8DGV`O!nX7 z`D~;encu1QLzB;bHwD>+9W<77b5f0T3x)*=HfB%c(|${k+!t`E_UjDXP(zdacY^LLMh`w%|u(foDXztC+b zpH6dXBOW=79mK6NRR~oPG(rMaCJL zSRo`%C_3(vC!jYf_bG-)V=PDcO4KHXK0H5=DyQjqy2)!H9`gZa5tVUt=q~Tn6T{sS zYZY#}k?x}13o(DVK3w|8V*(v&>FH0c>pS_$&fdLZSiujE3efto3dh2HHb_d}Jzzi} zV&NjG-)x`hnvf+~GX7g3L^6D-Cp_UDS@P94`wJ3l$e^8$u>ZKV*~(Uuv8P@jqcJec zOPwtpOhvi3k*m%N3F+yd{uA7DlrU#Wm`C22JjhdT>l+wReRb(`^6}I8shBU5_ZkUU zO8b0d8_vj-`(>N z&wE#>{t@&2tZ<0~zr@-lf<3mNf$x&G_@Nuk1(89rKA-;BX8utgpp^|Ou32Iv7p)M@;)ckA6-PTc9-U$00CHeVobL|N1fIiW3%-g6-imdKE zdT-({rdrxoE`T}Hk&x}J+o2|~-Eo?B4ot~txpWJ>%>&OYLsXoQbL@S2SH{YNzMzg1 zx4w>Il*F?-;<>=v?>6C6(Ed9C@*d`}hfHcL%-5-4N5-R|PO%Qlk@cD1E(?hp71e`E zx4zvaWQ{vzODVsEEtltofNYf**ml$r%DJ#_FWbRLEuZz;L4!A!k>=(@15SGLl;?3+ zc(Hi+ZKDvke&HJDQL%+zs>Tfj-$m!4yE^kVGH+f!{S62}%Y zWISuhSDMcrfaJ>5FMd`z!ON$El+@pRnf(yEeZ*a|u+VYZ|F+SUic`(U3k%imqI-10 z5~W{tWjuJbyw6q4+7dju9dfalBRSe}S0V44mEZRIUC>vlHRAd6Xu#0u zGDE@v+s`=3^w#`#Ls8xp7Q)&+Jt9_Hm~>6kagTf(*7TNumoy z#I=2tJh)Fxun$ox$N1(?n8Gt$(;K^nmQMO8IgKiPp4QB9^!>c2xX`1`I?rd@5XBpK zMbl9qSWAVOyNNx-(rgN|M#^{BkXP5;KTi0X7?IP|V9XLqv(K=0o9K7j&(WGWF2S{h z;Y;$|0$y{Y(t+GPZY84pTdxoMNV^eAB=7HeUHF_&F_3_i+HCwt&JKDFeq*aqR~eX9 zefJAJID6Zu(TU{H;;WQjJF=f)@mtozd}k!@MWb$6!ouM;?3>U35qA#TzU#2gE{6Mj zhKJ_80sp}%4LRovou_+1EUAR0y5b)*sB=_I$;4w~((@Deu}w)vy)e_;XN)WGyBmC} z{8>r(A9tfTNo6Iz;+`WKs_tAZ9tFQI&?`#Cv}VrOC5uk$agz{dw-V&Owv)&Bj6Ee z^sbENt}EF73}X-TFC|Z`eiywX!yki?dC^4l0giN(yCFfP{}#ZWdbc=dvjKRrRn-q( zj#SW&{PJN!b3eVhSKBfMw(R#bTLg^P1*1JCgkQVA-RPzaVr>TTM|-TW{k%%-Gw&Q+ zbg3gi0!u^Um&8UfB1 zx@*Ltu7YLmV7r(qUoj}TmQ~EXm^g5eA7Ig@3XXuWc@7~Mq?Y% zZHbLq>fIsu;YvTgIif$)lUp2^D5?HCQw;m9NN`jgcHMIB(!SaeoaREmcUV&c`CBJhXY|~Yg-Jej^qjyXjuby~t;vFmO1cf*y~XH>!!}rkY2{-^ z-PrE0?*DUl90!Hu<^!%#DcnYDBaO2w&YhFJ_ z<*7ATY#BIeJU{csJX>xQnO;NsQrqE?!^@UBGgji!+T}4#r8gAozp(6+oML z3~dqj1-k%MhzAtPPu(UUUkIwZiJg{0+8qI3j-(!wGODv-4S-M4mLS|9zeD5okR8l} zm;fu~jRTgl3C^Fi=JSAayEuE{q7k%sg8ShhL3cU(`r~NrNGCDD6XJjEt3_vamJ>AM zF<$G-#R1GW--8zG$hXQ;Q$O&n`!_zCIGC5*<L;dEY#u4n;u%Ob4I^^VRwJMtg9e^N=dde!9g@J16jUe~?KUTZMdw{^ z#|3jbMrs?NssRf4g<3{@)zrsDfi zxjhp@74roDx@_>t9R3~nFGfUvdo(eW>TL`r^uEdea68*Qc2wiFrt&4TdeAdeJjJMI z2};iRm};~_`F#z70X~na(YSSi03QXK4pwXc*Q?$6tYSUAAgoC6bs#fQAtC#t47%~u zvKUqmmi84oaSP{i1jrJvh#%Vt{VqSH;0Jd#a_lnLzvKs2i`eBHNbF?#3UumE?F-Ve zQ79kUIE4mn%p$dAk~7b*1^qC9x9ELWpNrfjh-M*TsDp&+#8*;GgbmLXzdPzMu7^37 ziKs+m?&+mC1x#cQjE7_*xt@oZgZ50m8G`=~=Af#x*O5KqS+C45NY04Ucyd=!9GmQQ z8%eWj`ndno?U-A4j$vc&rPyOXJVN?T93-)WGvXHzd$@x)B;{qWCsjr*=D%X}>Dh_6 zD(7>8UtZ#=IMs8}n6BVFDlltlnJ`_)91Vc0;?!Y3z4zZ)&dmsrvaj&!lvqo!?^wpbYW0HT%i-j-ebshcjRt`+o&+K3#Mu1@~y)+vO z(l;Bucjwm>iihi-i-UxYhgqLlP3ZVC#_k?G|6miv*RQZ*}kZ*J4s%?|l< zXjVO`a0^}o1DeaSa38lDak7f%!&_fl+Wb5LgAeeU2}9oSZ%SB|N_FVuV;9M%mQwAZ z)lR@4Ir#byF_+)(h@Zo~1AWkS&99tNG0*4Ktn$jt4wiuYfw)#?d!|^~Fzh+V7<)us zabga--}%v%f=LcvAG{qioQvapw3eO4+3~;kaB#ai^7CWi^1JnUe4AFlc7n>P;`Nme z!)Ixm4JLQw@Mvvc>3wdrtIZVgp!)Oc z2Mk>3gTuYYI*7Q6oK`=K5<_sDK z0a8S3Vjr-8v92`g@zlmv_S#;_4Ykb8*$fR0;?BR;mC&q4XJIpuvxpYy-rq0-)Y%CV zr9(zqX3_6K*C)u=8S6$6HSFH=1mnvYH!YoM#|DW(DmBNbHu1KW3ws}mVK+Ci4okeA zgin7dys4{gSPIR8CoC58NFyLitn8IK@-|mXQoAiKr^kUhC?tKj3u{WoZLw!S9_^EN zh5NxMe$T%wfKKA=V-H#5!yo@mWfL3UuTZs|BBuSbKsEMlDmytUzkA>b20wd;RC+6^u4)o(I}v`h|$ zif{*|S`kYikJ-NW=WnP_aCbpFmS;<~A3EWY9HHz10p=nwjNh;UF^0*gKp*Cgn2t7P zr-KfSa?r2lTvBs8UPe2zaIVaIXw9TRzx4M__{RB4@J1tb<*dOCZ3Ts3$>^=Dm;c| zkkQ8hQh&fM_L&SwwX0_F>Y>7|LO!Dv5L$U`wUA;;oL$~ovq$`||MlP$zDtEXJy?ey zLk>=J%`kRIv5B_ijO?3Mu-{Fr4b&9YA2WR<|18J)Ava~+#)y`J`6#jjG9%-_X1X8J zI?S!D=lKOai9d;G2yBVo`?OJcT6P8Xu1u`VFBJo$O&2@`L!2vM%@(!aB49g;sObjN z5RCPqF8=ZP6t2Ly7@)iY1&utDZfR-nd1G5LnwF`@y$~6HQ5zG5RVT z=_DBnYa{_#a*=b9Q%zP+cMJPyB+GSmXRB ze(zc(@%k7M{u5lkJPpawrD4Ae$cXXVY$01{vs4F5-(_lSrUuMmV?#B!pGoHi(Y;*eU65NF%> zf7MvsFN-yP11B~htO_j>%xYZxIC%xb7Y>TD>ANnYf?$L&VsSf`dDh-@8ZrKRDUJNA z|CuJx-l_T>7v!yx;S#MarsH|Jy2hdB@YFcse zm&mvvB5a;?P&={?r7-ck={3I|^Q8$uiw1Qx^q z>s(=H&?6rnEWh7|JxfxDjdC_IBImRqaBzMIz1pB~=%R!g-PnsuB!SZ@GUGf{%;S#i zM}<4NwbUjNHEfv~FE5y$9^I=Az#akKNMOtxq?fNg7sz@by+^WH^^0yd9|uKpX+3cf zge8be?1xy@3*vxidmTS*TDE#{F`e@AKu;fDH4U9AE*1(y5%!1Pp96HxEt~u}=qaHh^9cfaaxdTj? zpJc58!n5Jw|q)wpbfpx zhCYaC1kS^!O`c~9F*0%}YLfzX%H-^@Sm?|r*WK;05WVe}j|OPm&YLQ%S@?0aKx}ZJ z;7P1OKfXta4TPhTjeF<4x%XF}{+bBvu;okE-lyHxYhBlhE@cgd(H<U^#Ac8g%FfTHKEt!Lx-RE-}`0l+>p9_(g z8PEi~bjfTJ(|ZwWvz@Yrjc#A=S2Ue3#@I6#D<$vT%CtZ~kX$EQGI*g#oI9%1__9XzgPZ2zHD#Bkoc|-~ zyW^?+|MwFa$=)N!$j)BLIFXe-%I26kl8{Qqab#y?XT>qI;v}axvJ(3?nf@Ru2W)Qo7coy1;Z2P;fdZm9et z!&>tX_LH8SD_biMjUgeVP7Zg1|0bio+V^%GrG6E)7O{*Aa$mL!@;V1~)A&GANk1Uh zi@K!B3?g-4_Hbu3_v?kuh0Vf2!ZZa6E{0Od(QLJd#&R1&s)O1}w^jxj`M-FdR83G* zDZg2Zw4JJOY&e=1*h$p5aqt7N=cr~Z7L6$87@R|_{qggapb(@RjiWlhm!}oI**yfS z=84|Cf7nqNmt^?sa;rP=YGE%~77*q}hc&N3y`+MGaGN2TSM z8{mg%)3;7RQXGV%n)k`(9cfEXf;sPHs4I#xBN0WbT~wnV81Tm-w=?_;Lism$t|pu& zBWE$&&$rl2O^_m>J~_6OKqCIRy(d}sB3#xGL436<;dHxfywCh>dsN?#jjVd#jJyf! zS?zbmu8a7@G57*?p;a$40Ru*}5nZ&;#sqMyu8tgz+@o=mVRJTO#&P^l=q%}3rnXE+ z1mcBo`x%Tn(%(=R+>V|LO4_*5!t3FJ(Op+v@I;34tszM7()%TveEpL@kGp8lY^n&u zbWsj`{c}|JW@(SX_x8ViJ!>Wp7djB#S(?@R5HFq83c~s|PeRdGKb5wvd`4i$s=a!6 zpo@V%iBQOTAsFmp*VAs>%@KtlI1ujcXt<#0Cqm6!SZrNkVtBdP4M%ebOuduiKH}*) z)u_e8g>L-(3Ny=;Dd~PuaF_H#CnBa2CcVW2IaQr?y?1g==JgWitW73gRuKI3zB$Bf zL&IR!rtpe_bT>k`1`+6mD!khLpapF!E|9&If7>3sKNJ^khHt%fTwtGkv&#u0`HxZd z-rBU!p>d2q{2fv2PMZ-STAyW3!etYwcM|mqxF%^Xj=6WdF@z~YwLpu7 zRNWTz-)o2|4)mE9Y_*#L3f&9L-Jx}66u_Bf^EYI~gj?+-3dc=O{Z$-O;puY$@g;au0aK3~0sIag(hiYnqhW4#M$ilfg>OjUv6usDhFhkjGjQ#pp6M7Q6ciS;rf z0p1aeMRRS1pFv7W7g*T$?JyA4VGYi`p7dZkMb4}WWInJGk8+yOI}@XE^VjOyLwj8a zo#Uq(hpS{4m<`J{rZm5ko(*Dh7h*3Zg;UFisL-z`E)#*Uh(UUF?9gV&BF-|%mNQI; zG{I0O)AI1GH@;p=ob!@R{K}er+0*U_=4DN?&zABrCeqQsl1T_)G^#WMa-tXpQfdOW$`L@5w|ahBKn znT@2TXkFtM>G(Oe62w`EaF}qWc1Q}`^Jr%%DsA~(6!;S*e;)lCQ~p4>j+9y zAw`35#1d2qKATG?K}t}@S2*KQQ6ak$j@W4zJ# z>bsz-_D069SAH_#huT)XU55p5iWN%+YHWT^^gSt|B2yd+-x1?|#C7V6%!Lm!ok;k_ zAF|T@CA#wOvmEq@2OFO;uV4ZUrZOs175n!6KLT?ttE@I@Zc3vRlPKK*8uoi4m%e!E)yh=yW&Qa8J` zjmVrSikcR`W?)g1 z(!bsAGv!;}pQMzmLSo@&#Ub7yACdlg&s}yvTCJD}>wBil&$`U!X;3oG-wjw05p1NKrZY&i`L0 zVsC0vbv7t?ZT!JUIce=4^}If1o>LWuVvng&IPrh*Yfox2LOoyZ#etPiTwd>%mUV)O z^UGmbXX%olWpI=kc{;aB$$S;_BB`GtS>^ZCs$3(;LDEfrs+wpG3EX@e1UiHDuV_Px z{iFiu%bBOMr|~gv8hsrVR5OFw44^Fno!Fk_VW3VbAWzdgrx&W*PacTaZFj?|7HGQAt>ww?+N_xP^zh9H9`+q?w0~ENSwvZ`c8Pu* z>wEEcK15%xCoI-q@PH<%6D&UhD-Os{1*O){i-_rhN(G{rYJHN6VEll>wpt%h&HoX15CyEB(%zA5u9po?JtTt`#k( zG4eaLow=jZb>psZVOpQ^3B-HAPn%a-)p74Wn;HG0U-Bb#P8v`v3YC=I-Xwo86`HDP zo&YV4Y6=DSe^C6iQG*O9CVHxB3=w1AYeBR9W{Na;zvKdI^~1g3FA*OB1X#q=M5Zdj zT3Iy3EFi#bYMD{Qqz?6Rt-?JWd&K=0{62&Tn&;((nzmK+rY}FsY|*c-1$YgZl761) zXM2)nZAd26m=2N|X-p+|FVG_3{S1v37zor%%?obv(2)HCUm{a_p(&nQ)Q7XHNmz ztZTZ6?lZ5}43P+gT&*QVc?9ARXM01cNYfc+#{L1vxJiyc%mtZPKvcRWgtmC0=rSdm;iiEL{#V4+b7u*dk(m?wayLrE^u`5JymRu9(+t*M5eI<2+Ko65wW z+axlL|I294j6$RhAvkE)yV@5Wz-mI2`k&b#g>jW6OOUp+!UwOv5OwkE7LU~4yK@^h zQQ;s^C+x}-(ZP;i?4l{)BW?j#m@N)<;kcy*@;AA#fJ{gJauY+^7Kfcs2VYQLIj|li zc07JI=Y$6(Y&GCQ1)86SA4u1>2SdOcW$69yMm_F1SPGEaDM_Ax+eO@Fwf;f$iR;Ze z{|l_vYOZHtEUgak-KNMc!?3 zP!U!IUI97rmh9v!&@8eUEI`zgZ4gZ#KxD;+WG)`o%Y z;j}+@MO7b@eephZYAa$O1w?1OTBv2$Y-9_9**Cy11SGrB4iX*LuW$2WHi?y5J{Z2>=YQtdd-yv!9h z8UOyOwA4i?WF}GqXK@_r0z0&JvY#(_J@XC+7XL zV|^af7j7NSa-r1A1|&=wz>*8a%W;hWs-+w^c(0nUgBH_#gL7n)9N8$yFuReLq$~6{ z)e-^(TFK<}Yd-pAU_-lza*35%i$F?Tc}sn9>|FAZ-Exbk7iz`%p8??>ohTA;eJdA< z^sout4_`#3a5Z)!)eGc5n5&MHjOU2^;AJi0QXv%Z(uY`0=zrEB%;?yWinm^95qX4RTI!0iyf2!K#Fs zLMPZa;hf#Lk+OnSZLwT^7RRD~>D2ex6 z)I7Z}D;)Z#Lhg_qlwW4>Et}i0SM|QT4%^L+%>ac+{hXzZ`a7X0O_8fA8i3%pOtaH0 zrg1vMaGfO-H`c4tkFb-)GXRQY{sI1U+}zCI5gO^QU58Z!Ay_lgG!J5L)jM16NhSdI zyDUI6XSg=eN~48PK%2flCObfe@BLn|-4^Zy8$gW_h`TNCalekfL#GjYHU5TIE?!`X zU%=S=#RcnpRHvp!y}=LRE^`Rd$Givrj~^i(F5#zP0K41#u8}VnIkdXsgzX@lbwPRj z-8cCxAPPHN9s(dVSasmu^)u1kN+c>q=-Gq z;HLe;c>yrs9+HT~eSOxAXiN$-Z(M0lcS5X;>{tcB&L{U;s470LB!cgv>e%m4OjiI$ z&9k|nhfK3OVEk)*I#(AM;BKQZ8Fp7su(JL)-RRq|Uhm zP(BT*GLaXvgkooWfXV#xG9b`hBCU*J)~W}$nNk1ttfY0X#Z^Ims^awD`lnIR&~p?r znDJ^`xDG{(0OH>q^?4Eah0rtZCM-DYf_`%`n=%SYI#YzKe182e0%!=nJ^3;B64KJW z-R@u%{-_7AlP%)0@D{x$PQw29yWV_=qySh8jU^Co56$S%gr*=L3w9$=|GJJCd#_`w z><-NvX{NL;z*|)3aR7|VSP_M0I-N5Oqni!7fOzQK+BDisiU%uWNI&J=zre$1w%DH_ zu$GNQ1MJ=#u5;fYq9l@exWLZzRI+>KMZzAydI*}@hY&SU%e~Evz1M>s34y!iV=BHA zn02bMGbsM!2X9}20GM~Y(_gm22a6a+GpC<-x|q}fQ=@5{c;9J;!F1%x&AQKud13* zB4iE-R!vCX&_W!+t&)+^fY6yf9vN$QwL)fX0MWQjsU(y^GJ{hDTMHrnF@ejJ^yha= zy5&Nv)3*_0aI%}jhkF`P2M&af0`dHUu+yxWmkVEM)0v1Y5BNUYSfy!JQLLI|5p1jc z+E=HIMjA!sKyGU<8|-W7o@A|u)&rOuJo@mR^@6SV;lcs(bI}3PFpX+;T&HP2{8{-b z=<=4^<<$xVFiq&BgE-{7p9%kiR)GJU(V~rTFMkLjW}KPd!w(51Ie=s6DzKA^4>yCRdks$(XSgq?<4uJ^<)A>qHK-^Bh2XENpAI))^4Eg%T=iIv+t}Gets5B0-T7g8J=x3NhnsDjHO<1M9zKZN) z<_s|M?80~^`ei>3m*~HU~s7P5IR6D45qhB z?X{S2{p3e1Zf04m>Jd8AdIUvo;?4X&T^yUPxh)mibsw#UMRB`41{B{emRx zE`MPqv;bb0e&KPBt|Gc)* zVFam^X&=aNEBvhg(OBKJ`ncRhzH}UA(}L#okPH`15sTeT04!iCmu&1KZUpd>Plik`f1% z|I7t}EdvYbBN-#6GKR3B*T%Cp^I6y4Am3%)YVdk>I(KEuz2g$4@ukQVCUJi~erK}SaVcF@3)0BY-650o^Lt!c?}y0aJ9@a3bKDf?-& zRV#?S{vW9F0E>nlYJ&mhxtNFw$*g^ZN`B79Ud2R8@<`?Zgj3`Qy&9))|4RT-^{4DW z{TL=PB8si^pa2!7g_DZVVni$9fYRj?FifhAr#5U$IJY^18-kb|TpGI$4*BR=e<&yw z{TsbxVlca5Eg#KSJ(B%I^d{R8Y#141DZ!61U3k3ut_uwJl-!6Jz;Su&oDCr zTN{Rl0^dMQQwI^=pDnmh={*(jHyWE^Y)joN{4%T z5YOb+kJ8(nBGu=Q``<}EH-Q-7Z6gO1uUl`D`#pFAW`Dx{qvTEH58I$w7ti3h=y#GK zlEiMH8q`|&ImAG{NnQ|lL>^LcW7fv>?4w`ky1El8se8L`KI&MQBJSG@0`sIx>Ajso z2a%hXsPsD9<+eZmuI$%|R&^395FTNP)8?W5tid+G-D0oBPp^{15Kh;Gom=>FXV_}- zRyKN*j3l0?UdCX-agZzr5I&0uua^R&q(H0Y{b?s5>6eWjeb9u|OACYB>o997wSUbQ|#@0o~B8=N8~RK2gR!@gcfoMz>`E&aLH2r^dWXxaEAi;})Vw zb6D#$+KWyhIoZ87ErA;Gv_Yx*rk_UCzjEMT4^*u>$GtCo3()BmPCYloAw7;#&pA4@ z{b+%TeD5at%K5dzFBcood;{1M<(!%zu4%FODgp>&dTsP*Z}&ZiUr^RRt+iVK z@fHBi;Y@qf^>lo#?(oi`Pk1W#0kG729RrfCBVBoqVGeX#Y$pBwuwb=lSB6=eZ(@M< zBI>=64;Ll`SLIgt_3L?@v-EQYVKkq=F}D`jp7emrhTL?L&pTw8yuF#0%#%9_)*RUv zV?Yn(F7G$>_t*k`k#JoW0rq@Y_7p(CGO7?UsI5e}DYJ22NdqMWm3%M+g!TMIivYJN zatg%W*=dxz8mBx(%T}|x*0|g-u?$t!dGkR2?A*n$SuC1B0!^+6U^ZicpM|(_ z`fPl_2;hqc!a^c9ES8|Oq_lI671X3JNM1C+#Uy|jc?Gh19q%jbhv4{|uwOTKClPyg zo-ai~$$8IyDdGm(P*o9egJJyca{HI-7c8JTUUc7yM_z{Je20c=`fS7jqp+L%xtrwI zflkD4L*+>dc~Dti^8M-KMHCrCGjFNA<=lxU7uR8yn zRJt752*O>q9S=buLQ>IJ(zbp$c}6inSaN!?3F@+=fC`o)nWOvwCWa}E8t8uSS~Qr( z7hoU+^lC2?IHR>umBPvN2580O-+%N2;%?gm&&*Y`+74<(AAsk9ZVM=^ zGqO38#8_2Q#<1g(3T7rUsL;`DXA56QF4C8R#$u$=kLj|1G6HZE`{V{b)9VZFn=e7m z0)Yb*aCT51u}O|SkX&2Qk59D&yD}&_9FYi+vH#I~OUWxh=JM&@9bIGP5u zCqU&ifURFxo0k1|-TdI>!;R?I40(1f$oer@^^w) zAqL0$@7t!i2_rN$TYxzjg&;S{kMhKt*+!^7s>in{gP{h?VPf}L-^b`i0MUeV3d$Ji z=sVv8j@?tqPv@#UgBiA{Nk`V#qUF!A03>w(CgjoA7&^3qm%*FVRS_|=tApc=-(u%o zcV0wZOXD`D*c?&YsJ5x+Pm`wpan+X0B7{emj7=cHfJtEFS~N~AcrtvVu>VG`vB*+5 z(7JGef~x-6SoJ`)q1q;_e%4*UM15WR=L#wk2IzNl3Vc_mBV1o7%{X2Xl z+xoa6c!a9Gt0Cdy87tHOz~i+0rNiAfK(_0GL;xF*r70UeOUKujp=oX=mWses~^`#&p!{y+QyGBIYz;u86`WDnPcu5*g@K6TgzAKW_j>{v< z%pM;h&!Si_ZIC6!*Njku>wtOA1Q_Hz0e=;>0eDpSq1Q^Y4^Z8~?Z&Ye&x}>O`w*x7 zF0(-hA`&bb>{SCyL?IG1609VgpIQPo+`HEV6n|9=JDY2l?~I;M2N(b1KQ*Tocn zJ;NOIFdsP@Qr<4Y{S?qs@C~R}v^N}xxW>lqIn&B~osI*`D_NuLH$KH!{PI&!@f))^ zZA`76^%G|~XjlNZvVASb=F6+e?-|ayS-Gu`W3TEwb@s{V8CYfJ+^yyP3tf+sA<^qN zNO7EV#>Cnwy=Gu5%~c6VeufznWed!qFWGI+BDsg3J^2Iw>b!bKc{+;coSV@%2x0B) zM$a}}Mq}5S;t`@eo%eFT6P|)58$J=3=g-ez(Nnm8I{#*v52ipFLz(e~lJY5r!Uxkg zNj*GKCS2S9HfLktV*dv`<@<}lKx!tL3`_W(Y;*rojrNbq@&5^U9D=$VzG(w?k7%EP zKZ+lcgmwF*OwGd3PT+UEF^+Bi@|m|aWggGWa$HY2^a;cYJt$sMmD@S6fj89-FHgdI zX-amR5>wsY;=ZAh{-(gcPZr^DB;As?OCDY8;r2bYh6qaK)9zpf_Wa4|JGZxU?q9}- z$BUc`PXFt9Qd%(ivXwiP$YR167_UukvBLShXn5o?J1j|Wpwa!T&7jzId)r*?U4y{7 zxuUW76db}kfF|eJ#rj3mu(l3qz&^$3h1`S6Vd0n}+`TbjnGz|O`l$eQ@^OS)AgX7d zJ|w<;* zP)QpjrTboVeIsS<)%4Zk;XK!{v$)~Hte8Y_;2K`l-l0tn6H0zR;(@QXI7KlYHB6fJ zxnhD`5O;ctai2YzrS(cZ3*T>@Mh?&CUM^*_QI)o>%((YoE9oyzUK6J~+gLqO4({Qd z$4B1NSmB_c*x4=E1o!A?D_n<@@W#|Max@}T&f+iq{l@~ zvBUG3Uu-k9Fl|cjsg>*P;jHDcB8ti`idKTr@`&qn9;3>qt=-=Nt%H>`l_W{vHeM}tuCI=@)1?krouucRI_bEob0uelrE2q$^z zk&7kJ4*#d=cSSW`{&lYpl<|L(*~97Un`MlC%V~nh>Fd=?--TTj z?>|h3*vJ2uB1++h+2A=E1pX&pHJri!CLZ@qOtt#;%q*2JrVlF+6C2_oK$Xu4r2Ywe z#U(dgn$@J&M@+?66x7L*>~ha18_$)eR|4u1aA4fIkjSqnFZ4?>DNvqN{F*nlY?xWU zPM@5WoAJmI8~k*J*L_eb)HRIhPYh){NMvR$Jc^8$VNP)8~FZJ)0iYU?<5E>!w4 zhsCY;Y;S|teLuyit#5EXPGz@`v&>WGerIechaDJ7@+D;EnPAG{z8DT&@fiJBm1y>B`?Wp#D&a*2{ynjt@dIh|5gjx#KnfPIU6Ar{rA1Y2h z+L;^%vz2jb^^q5gs5+2dP0{|k3#Je0FEIUYq=6ihZg*eGJ)p5t-Hm=u4*F2GRD&c} z1-w{+XsN74?e~8owa(#7z6Pgz=NBc%_RsU3iS~OCyKBgz^-!!oK`X0O>Paso`0z~- z@UYL*_cLpiux~$1Q9wDTcPXwZx->7;cFTBo`N=|Ct*RzI-6FNn8!Ids#< z>9@Me?;noL#lx>30dI>}Ur^Vwuz;the{$moQQal|K9$r%i2~=ruifwqR|ESv&n)UV zD_=!k?SA!uTX5G9@Q=l&h0jFO9Y0r-i04rKX2OoM+cPIzb0)2$^i51Bb~JBRdmDeQ zG%a*L>G8usSJ%+%?e#4+r^!ae`yx62tzttT3DL?Ju=w5=(~TD~x*)Y-c!{>ra3AxJ zEhHA^?B`2&l49|P0uJyWEBD}Sxqbaim`>Vc>RO{i8#f_e#Rev*u#Q6^DMG^@UUq0? zLSz3oU1&X*>;|_~y+x+Q=}FFgnNqIAdziMKnJF3~YxIQL@sA^4(2;6mc1$g$ z;)Jv@yQGRHvq!bBPYNWzPkGL0(wZDQE~WX5s+j4qc|^{A(u@`yk@m*iu&C&oOsQ)V zQzDKK+{*8bxgiSG7Wy)-gS(1N7Y!bb#D$#6werj5xjMccnS|SFwD}DNx|B%$!Qk?i z^scJ9C^no1{_l4Ov>}|}nX_C*KCb}|px;2Z>K8X|5UtnBeaIv6Dd};$`q|C!8+r;^ z_xr!oj^H7|TIWwOXUVEW&51NCxCVbE!R%d5cOVT>EyzBlC?;qZ!yp$I$U_ky9KMw( zJ&>0nGVScek7325j2_*9{m)U(wtj6eh{!<|#Y}WR+P@W4Y_Um6dD(1wO)81YRY;DT z?`2wFFr8f7Mli2Ho})72ASEB3m=Pl~&Fvv<4#qwZ#ntmj0@h*4>L<(54M{%j%MJ2f zkt724$9GNe9@td^x0S@CiiVAy?i}7R2mX{%>0ElvIP9m#5X{C2wjY#+fAHG~k^wjF zU5%7}e3FNZd81fYv zwWr0Mo(F)WM6BZo5W>jAyajc5&6B@sqBwk3;d8^gWLL}awvNMG8$5+z_D{W?5|2WE zBbO;9jg~0e!Pdrs_1^`dX;?Mr9Dqr^{_ZNattM` zTfmFt8C-hBqjB8i1UBwNU1p4nB{y2H65zFZuv7`WD44#DN_;ypM}%%Xvk=n{<$a4$ldml;e-@D(S9a+WmuWqdK^5& zHrYmr)b?rDONvF&mnYekxzkUMDO2lpPukr0JX}+MDp{I)zp0IL=r2;97Kd7hXJ_n~LM;Yt6~A7YBRSh*qoD73~5#p$pqgOqkE@GW*WZD0nEHhxAQF@JaE> z#d(Zs3u!D0bJC#-S8fIxD4SrE;6cI9sZ7L)RtH7m?UZ^7jeHRr!$pO4vT#T--Q6+BfjJe=cCE{=6xzSEpnkW9ePsV_>YzQ8q?9>m0`U)8&}1K z%)U3n^9kYV@2HbWO92tcq)*SwE?_DDf%9Y06Qe7^r=WTB)#x#Q-X2SuM86u*vVui# zxGI7&eXU*s>%j;i&!t^7kd;tWH2@sk`GZFZ)6ciLMFU6G1$p1$XG1MqMo-@i8w>R$ zf~mm!R_eFI#=5aMH${aUCdC8nB-=FQjXe8`UZ&DEYbk?Cq{PvPkQQxs@MgWSFvfJN z7i-0(y=@XsfF+-uF+Lm&E0&M$n8vXQ#p%yieI<%YC5TkopvEAAWy5QOiSTKd} z6QY9MP17j(X6(y9o|V! z2859^B^D`IjvSHHP;>n;r0L-u&@%I`28a{`U#*f?X!5>%Im8+z6%{C00OCydhZ*f{ ze6#Oy1CDS*Y3f=CU%ycbg-L-;^ho=K1(4hAe?KV}@+$*axk?K5P(*830?m$DRCk5( zkokE(R&Ct+fO;y0NmhJ0`~X?4Szus#fb4}WyXXE;Pg(oA0hUkeCVwfw6(Vp3Ty%-J zyGHm$&{lkS!uahFJ4cFAp3|ZYI|dgij=27;#J47hDGBcLg8ItVzWLuomvMB z=IgI|V!T4mpyh0a=U0fR+H;k>V8vM{AXn(g7Pb>vr^}w+^zVlKPgljGS4}M>t4FO| zE||$`O*=IZb$yb8e67Q|>+5Gcu@-iO3Z(G;kM;Z@u$mNj>{TLgz6#xtcc!Yz;w zS6@Rf6eX|6T$+1&_7}_=p$B`y#0@lfR7PEZvg!zr_o=^}Th5Nfc?vEyzU6bR#!HB84fMtF6rih+U z3hhm$&%2v_kNyBJ$CCy_wbDNX!;7tMIh|d{`qfqbC! z9E+Gc+c?(bGAF<+{&!I8z|2lF#G7<IlSTBDW0x`+=~377e$fI`Kf+CqVF4b+ZKc~#k)Fc`V7O+A zl4WErUFv!h#ctmpA>j~@++Om>rybXw0SmoQ=|xnsj8|4i1_2Aa z13n#MPyMnpNRy^CdG>m7jUWMEuW7}*dw^15ee!V!sb}i3O7ujE+<&TZMKKG(bj+kc zq51h3K)u>}9`u;tp5Uar;2cVgG6&Ap-^rQXBL)6P+OE9qbZrSL1Yb&ISC0bgu+5QP zjFIehH+CM={s4kmk9C%SV;PgN#zie!;wNTBK3%RZ&r0^aA&E^t2)M*b!dL( zTP1^}SQ!74G%!^96Ktbenm=wUfNu+bwXRX8gEmP?8LDWc$EAq?~4{v zB8u&-;+V7Ket6OeaRNDVGsjWrd-co4B)2S~VnDE(;o0x^LCk~k-_Q)&xX08o<69|J zc$iKpk#p$Soxd!4BF3iukkt7z-|AVAU&q7ZuYkP z_g{V6&rDcc@@FC#c`Xkp|S@>WcIPnbM^*gXfX+|MPz&)a^M@+Z|pjb2yqVJ`$x}x zsyrP^6r!%(^^W@W^`I8uhbIgMH150;I3A$-I251;*fCjJN$d-PNvGMzJ_Q)gHDJ6g#mdjOI@s5qC?1&!*aR4Q8~2tg=w_09TC<0Zn?6K3C&~+1 zZc9wp=N^asNDul_C@eu+cn}Cp%_Y9M$?yW%m^u~`c5+o!>$F2#Onj1n#b$AW(9+Pg zx?i2pLIM`@d{U`o^NE*Ltc_#X1qzEbFXmEpC)aJ<} zXl;pqLcwT=1L`a6=?vYv->=?mL$z$1B_!dcmc``1>YLvDO%n+~AW6r<5KLXv-SKB{ zk>hg?MwLa^gr0TK* zeT7Mn>?3;{S=yNI--WB!^EPt=n1Y26Zpe35lK6-EAUjVt;*kW9v52z%VyHk8&syGU zoxsR}^hs)%=KdCtE~R1W(zC>=T`!dU-Cm7;<}*>Sl#q(DJA=--iQ6qfW>OV`G7=9{ zBzXYi^dt2xs1r9+!BqhV0<1VQsk+w>yz-0+oGXBb^jQ96!k9Y})q3D)3L&1eVH%R$ zrhBWr9QWRXWB%@Bm!w#ppLKQkeai9`7n!pPdOK+d$*kA-%YQ|rhUD&%5t2a}`GNmZ z#yP|=hC>YgB*FOJaL1$|L6CvezyVNct@N<;!1eGQC&61#Vk|#%3mIsj9~5as0anjB zhizV4Zz6U7W$)J_QFc_sjg@e4)@pq9DoAi(rWp!0&q}siT_TmG4p4r)$i7-6@aUyE zIVW95?+QOW@D=`fcvSDp%Z;FMt(QkAn}7G16_paG(u=@paGA?&a^l9jMB*ZfV%{mVAly`BDUMBMEI`JiAUw!j!7a7};ecuVN2v(>jzxxIz@(>L;t$ zodB>%(xTS^Mw9ON5rCOk8pR$mQ#nd`fI5|Zrw#aYxZz3x{N;9iW&&yQ1NRqiZxX{W zij$FW8f>xpbQLVwc>q4qi`u>zFLUlE35!shPg)b7I`I%B5w$%k>jt)jpKE%$?=qEg zIfqVAeF9z>m$8DE7nUfnvakpw6$n#y`E|pwMK{J+h2aeXc2wjONFvv+=i3#3$}?!= zjb5m201hzR)ni>7LOKg-(J~|%@f%%BZge{ES9o4sbZBB_1m>2Plhxd7<@1&ht$Yz- zcLfd*Rq@yR0A=={!nbg%ob->Ub1b)kiU6Y>|L;i6@3xGKWunKh^8S8(^lHqm$SePo zKVN^IJqsy)_mvK;ikz&Gl)sb#Ju(JVQpCYi&!{Y9U0)X z^KHSnAvr9W!uOev?nEwmHOXpSV=(4Am~N~D_~XD8p%=!q^6k?XlW+X+Ur#%ndMLoP zReWg6(;TJz=u*@R_xRd5hQJ51WfV1bK4K#UzAVX=fH(YMsY7{LH zL3v)Ld@9ttg-5wQlAW_8MGSv_;Eh`c*|vv~4==sem{wGArmfGi(;&|~Z1|XLG zT?jfn0=Jl7Gw%{5KfF3=1bMy$-IpFaokBc$M#8sk{{6NP2YiSlY2X)Lhup-tj0A** zGh@zqErcHd=P%op8otR7t6{104$ZD^7FCT9J>L!=q^M^xD$a-62}dz1hSy_9N1mk@ z4Ke=e)I=WJ93bPR#Q~IpQ*!q8+|{23cbhCW7lMn)051!p`35jJ`I-P@JcCZQhj(=Y zn=MAfx59AsborwCBue1?v(|&}K)n?1Sx9Jh?sXtl1yJ?JX@9v={Klpesq9TNpUAm; z7~0HniGH3idj4`D;m3Y?FKtIhUm~ay&vA0#22$ErE=X`jlWgi)%$o494x~tZjwK*& z_)sZsSat=)6eI1ZZZ~UJR3yF&G*HAFME2@4`mV-VpY@$>&Ix96LAN5I#1$<3b)k-* za{!eT>)UzXIN=D?m8h`+hbVk zv9V2KJSAz9L%T{nz(vvVi!CG3YYZ9k^gUAy(5)~1v&BjmVACE}h*IPx$$t1y34|SD zo*K;Sbra*6F{9O6qTiG+!c}1YCVr_qapv8V;$^fzPo0+U5hGPJ{U= zl4aoi>;ox(JHX*w5;OrAl_mk`iBJ!} zlvRy$@<%=1bw(Puj+|AX?I9y)U&JA|3V*BSu2R4W)ZUMnYVlTDoj& z@R@(|66}98z;EscRL(lK0_>#iTCDyno@cidkuRFJ-v7aqfN{>F)Ymw42zjkf z_ctzt=LY`A^G4_R`9)t6tGx`cZZ;0WpndI-mDDHWj`-w-3*q{# z0pFtelij(}k@+CkVIF8#X2N>LB9tN0Uz z49fomD@x|~?B|~-r-(W1XOSuAne$F4#6^`991 zkKoG2ZqvgE`;azO<)H1{S?~lNzUc!*N@C?$rTBmhw#yFH%0?T}(9l`R`x+J#3f1^8 z`2t}{Gb_j{=Bx3?A;&iaZLvT4uIXzC+%KCHjlbKIe$q5o%t|(2%dr-~1L4I-Cn2hj z{kJKHt85K-x8_60o(S!CO$#D|L0PFleBf3{)Za*I>}w) z4^wd{viiSy(=H1$VlXk=T(|A%HWw`)&=JncijV$|ITp+04g7f(|H{I<{F5EV|KTlSiZOW!2E?!p$! zYDJokL{gIS&7wP@rxf7T&jj>0DT@t8A^9xM1v{Zban}EOq5pfx?37!(dQlRxjM}{~ z6K%4v%JO`T?^dIPDjREjwC!d|UW<@kU*}(LYSp+mlvcfyJ&HZ@1FRO()B&LfsNJBsA)L zuRhCV$!#Ingv6%yz#`PV^2f7vc5CBzhCflC%{Vg{k%HIKPHjXRCaS756W7PTDLz$n)Xy1&tpNEw>7DqCjuLphq;9gYcUImYz%OiOk zvO&2IUdHjU!eFr|kjQJH9A;jooo-?~()fm_W4oYW0%aOa?3egv>$b@oar3hUgWXdd zJNM;(uBH{2mTmgn@7~tCK;_hReGn=JHx8ysS6RJXRMXhyrc;)&xQ_>Bj@KKA0zm1r zR#k5CIk&z9jaKRS+^D;r!oN<3R^rgM3cotB9GJ_7*Ml{Q5dw9EP6`|wTQWKpvO+d8 z2LTH0>uOFrwaK+2;SR6%cVLE5g7oQ~)5Dt&X-~Wh@<0oT^P4gAf?1lj8hVp8gs8m= ze8ZjFHw6UullOilNDk*2csnM>OU$9}e2PIrBe~|i-zu^YdV-GD9GkqY*OqOd%hSb9 z3|+~6mEjlS4SHUw8ick5NwI#izy? z@?+Q5(JRbduDz&;ev@QleW{IdP2-nw&X@L9>Ml$e5VRx;9vw6_>p`URw_7?bc@v1g zrZ$At+W9bhxN7wu(n(%FznzxZnVv`?_A}-R?!`nCZoj|j4gbGSHt6$J*4pYnEUa96 zpk}~tdKywCF};j=fV=TR^WB~N3EQZ&uB8}M-{+I0!c5;YJ)Z<*pT8sSW}WNR@!a@U z0E|(*Ju0GBlSr~=7v1qyW2@{80KTNFn)kb;l~x;_#Vyxt;rZKqCcya9g*UUHw);khn@!H za+eiv1UwyM054NU#TN}8#mu8;zlDFYms9k%@Zhz^i z-h|k}6LDdvT-NseN&$PHb-_@*M}L?4zxnFR>`GTW^jk!?A6lja{pA9cNttEMC-Nt% z-YDwQR-?1ZJ>*v|Y#?@~uL_KJF#FS6CwUP@wYWS|nh8y$Ao^(H?Az;#On*WGGmMyc&Te6 z!xENNXZ9s>w&G?rg*{92Y7z82knQE^@eV;f^Y`7Trw<$iZqSX(9Zy79@RYeIwtk_5 zu9@>ruwI8cqIigCpU9wHBok)fm;WV1aEA0W>g%&}L_xuIqQ`g#e74vY%)jW+3W#(U z`ACgrT|t>wNDm%)uLTO+AW?ADFO4FJvk_Gfp6Q^Zc_%`E>tK2Nhl6|2UXQ=!eXN$+ z!W}GU+R3nK25wCx_2gWq@Xo%&w${;;<`?p@g96L3e3JR9A_T|)EPA3UU6=> z+GsLs^#4DvZxGw89tE};wEq8jXcat(Waw4#iL>9GhENq>;N!*>o_^d0_o@>B>?Al4}eFAp-din1yxLpT+ z-sV@+t!G{X_O|U~=h|bmTDNjZp7398GrYIgfHv0yjBX{NLGDGt_u7PZ;9a%?@slGP zU~LX5mucMy-3DkBv$vE4xVatB^$875II26pfK(jl}{E|aQZl)c+hl|Zn`yUh7x2X)>FRd^p; zDh#GuUzM@H-cnj;)mp_Mi6Lb)P3xfI}#bKjQtBd&KNHFZNL6Nwa$DXsQ z{XLyFcpnPn5%ckmj}gHXkqzlJ;I-`4vT&6}RIQsd!yjmW;G0+=4~(3+C(Hmz^tcT z{Iz-XD%R7~)Bv+1*=c+>f5Lt0o&qBg)oF8ACg?5Eq8g zq^%A$#HQ(>2tLda((h5WUzw1XssvWHGS5SW%OyUz;$KBH_&x3ng=1~fus^H$@aBDY zw}V6VHIdxH!)ZeoJ5MI@0q9HjkX|l#ljO5xZczE>GuKcCNcBaDqN4%B(ejh>z2koi=M_Y}3}xbw%joQJ=H zmm%bS`~D@5R(-+%zdlk{D0waXvt(-$|F{&i2y&AU4)Uvcsj$QPN7fMHYyl3v=hUiX z*TxFq(=T}1_r#oQRhG+m1#Mo|)0sk%T zRf3zXMAp`0n0Mv#M;ygDC`oQdOybjp&$WKYSg8_tQDXNm9~$G+$H<&hyBD)&j2M^G zKWUs$KPck4sphbr2qRk@fqhv47-%6_M>_!ESMD}q^2czLJl} zX5bUMQ0QPpq>c|}*&{+b*M16qu;+XjLo~p+uXLltY*9LAt=w~FlcW-2H}A3J=R*pa zj%>R)O0W?mfoJaYx-~W)$a077sa+~TSnYOQI{J>jOe|p?Xguj+2mN?f$egQWl_Kt; z*7ATjA0$`Nlar^k7$s1Z7(biwOk7&ch_Uc1`UmQp6N@r^9#p3!_0Ie)@SHo{dKQLu z!b9?W=;O(aEn}th<~>~cNDseSr#^GfH~X1Ry#|fetwUC*WO7(I_-55?jQHPV-NjTq zMCmlWjau;OQv4twz0>iVemrmyBW})AcC=ld_l>Uznwc=rM;23XuA$cuByl$Zoaj7S ztZ^6EkdjlY638N(UZ9v%3V`nyS=;7FG9U6zDsLxy|fl|dArxKW}G{I6s zPOvu!A?{AnE%)@+3z%9g$1$}7@hpX3ejvP#Rv$uV9;q#IwywjQba}b4hY(+`;`TjT zQ~6tOsGAvM5Sv%1Xctk`K|(i!}3Yvo6?Se}Lm5e9}kJESbIAP@chb5o+$2 z)zaGBVcO%iNTTku%`bP2x}uE_fGB}33jEE(OB>FTLO>`4PmVdu9!q|YA1LHum~Ggve~?2TIOiCiTUn~;_X~Kae6_#QgHCV$z&7YrYmt=oFCCIEo^GDcpac7#8CLRo5 zwHcR5G1v4=1g*sIwgT&~`4fo!OzJ3j>5aiEtOI}Co<$K^^7X|Uhm7t~(fKJX;?LEG z>3wVLV~yE9DpH(<_?=g0-2vRR!N4mz~oMyH{ay8?{ltHo!6pj3+Fl zM-z$+o)4~tFXt+ZlOl^0vhh80ap$afxzwIq+u?3nj-hwfOfM*pbW~dm2Yvtoeba%a z;DMo6!hcXepf%hB3llAP`Ym<{n3U+D9O=EM_&Eh~d=h{I&llh$k5L-SI??X`P?&Uk z*lv!7@eA;x2a6qQPJ)b0(8mF2-5%95)TQTuWn8l`wETkz%+prAZj=6U*b*LEd}O@l0PoA!f1=zJbtB6iM`i@&|ip zDzu32ZRjoL5-`2GvzLa z<$Nd!_JoZU1SjpS%W2THLx{m8_WK&}XGmL3d;mjqaL(mC;SkPUgk_2Ua_N^HRqT1$!zuIj{sWw)YXbXg zNGdG1Y6-ro!Qh74xLjvw88o^fod9ERLKe`xofgey!cWxXvL;{{tRmaFQzq=NHoyB6 z!FO{(x^h-(L@44CY|r>q_8&ZG_GQTv=t9VxADr6n;aryPLQIh6V8~P9r&~p?G;aKr zlutiVU(d5+)TU6LUtql$CmnO8YTeoCCZ^)GJr2ltb79N~8k3dxcpkd2v;b@B{1zq( zMyNUAYew+Qms8H)A!v1WXEH&*ldMb*w}3xSTlFJ!Jhz36nA=e#4aZ&JJh-=peGy%l zwgT-Yhr%>)8phxIuwMmLGWefU2XKd;LD=4s+E=KCw@9pRb%#~!q>41rh{L}2LM3^z zrVKyGxXk72eS_n-dkO6IfyTDgV#)&h^XSHEoH3#OK<737{V>(VsPxe@(5F-UIYHA$ z2N~p6Xg?2K+vYI!lKu+7TJrO(eq`?a-GoAj*xJJV5W4(6x|JW2*yQa(O#l18Mlru? zXf0Ukeoc9TWixWkOMmM93Hnyc{A(QoFss3*GZeN2yEw{Rq`!&fIm~Dn>0SdKst;(C zz}5o1^yU|%Jbf^8FrtA`C3PwA+ za@L8;IX0z5vw%gC*A>AU_iu6E(@%kH*lb+j71RPtD56S0{i5+YtZibKW7X8LADxvp zHQMwL8bnJ_9D{n!WQjTS!6RH{92?K8WnRF-7qME z>|m_iY|FXUA-^X{n(&E|_hH`;^>^R|(id)?ED+?sTOrngx{n7fI-dKJ%!FB%dGv(V z)o?!<9hnUKzT=h$JLLjy!Ve*3r$WmIIJ;W{rmc_Ee70qH@9KMg-ECy_*RPFHDC}8OfdChwfaj+svL|Y)IpY^?hOWpe9`m6xij9PH>G0T$+ z`cP?9*-tpIDMdYF(o3NrMa%}?j(3#u2>*fMY}nXDShA?lPl`Dojvsg(;E~2C@breB zoAXeQnj~sr{Fx_h;xv-r-|^KsCWTlrlF7inSfV220^TWvbR<~e;< z(BS7&)YV3@@CCwkyu}-o!}&^REX=VQUQ1w!ZD%$*P}$AgBBetF<#P#NuzOM$DH>Jy zt$tQmJS1-f)-8H${>sPu3hJQPnkDR9t2IuUoQ_*;C=^X zfXsjsL@&k;B2P7*Qv;o3=LPw^DKBBi0M)@_HD^f>>0#CVk={c$2d@4ft6fU`U9Cn1 z`cs0ML)#D~;K)6rx5xLtu=65Py)$KbAPKN$j)Kd267;?hPn+P@ks;U$3;bMnNHNNxx!*gZy14%zjJ>({!|dTME=DK0 zV}0>^;3pef7p}j>AI$PnOCS;Rh+W|OMtdndd5^IfcuEaj+f#TwLq)U87)5l%C4So5 z)Jjgex#4g#b~!Xpf{sHg4)8x)0#1CEs?Sr=BJ0jIIdXyt%@A>dPVv136}`=dtQ<*( z+aJFeZ8qlwIZmmSSM$p5?(hUZQLz7gp$*O&Ph^MfqItJ0Eyn}Lc6wWW$P65MHIu`O z6Q)`sG$7)WhL=VsVWXPC&l&i1pqE{^%>S`=3BAv5*|$5v+K4+>T~2E_(=@Z-4O!pyPXI={_eN9N$Qn2p~2^VLaJT|SzKOTEr82Z{R!DJ6@e7X;^ z7Q3PBM@#+4-)b$P8ez`fYTe{}QklJVa@ie8Xdm-4u-(Zt{-BsRFPLoiHWsc5Cu7zU zPwIJ6@Yqyc!buYX0V{b^EP$SZS%_Y0`HyC`b$>~VKq^&oNvxl3w!cW{w&|FhQ54dK z!W>BCW-jCt9ZGp%w`X^~0B_jY%<)?xC>NZ?1{HqDi1`5(-;FC%5;EuBY(3|FM@&bi zmYRPXGf*iY(DD}dh{pDwSebOL9kml0$vd45jZq%*vQscf_r%ko|8?IWL1A}oG(emK zEOB{xJkT>7PJ(|Wj7aqg+Ua{Hh0{AuNsIzoJ#EG~`%!>zmjy13KYvsWo&C;DL<{@E zA!OmFXRHV8b&h?{aP2>+I})1FKR|bUJo>ygmmuF@_PzvCopZbYos84{x=BgUk$DTz zL1@942-@5t!ob&q=58LSVcjO!5p_RTAPlrflrZYc!8Q$0QmflDHra!+iN7YtvCea> zRL2|OGrldY)J6qY}AkFrZh}we1;R1D$SlBrBR^zV5z3$ zx#%^FmIU2B?}xx;zKCI#GNAO6NQiP&m*A9(P&4#%x^)c`GKQ9-ahYA=`@=!wM~*Vv zxjY2$M;|DCX;T{lw<$f@&+hOPW93CQZw0fiVeW^S(?{K!qNJ|*&CpTPJD)^%wf)l{ za|i=2u|D)RnNOI2P8;Nqx``3G=QQor>Q;VjT<;RhIeZ6F{UkNEMW{}pZl6Z*441v= zc5a^mS+zNT_rb@UQxEKboVlHdPe0lC_z0{l&}52rk}Hls-J(FcRh=dhVmc3CH_zi=(igd+X0h@ z(TLmlGQ@`7LoV7Yj4Y>*5lKnG`6O!0p_=A*tv{PRTn6+A+Fx=>;I4DK_d@?h!U!Iz z+jGLx*^kzl{pYjsW|v)ji+TPFb9&g0QE^ET;x9Pz)d18f^7W4v=62V~J_87`nuho4 zm8OvE5Z8_QPD-ly*=o&VcB&tb?fmI-0sfCmcc~YBo+ej2U!F8LLK9L5oq|+eW148Z z-4qas(_jL0bP(VrGnfL94{~UlK65&{ObM*eRH~#8+KuAP+3m)Rx|uTSN*Nk_x=c)+ ztIOr5E;iOx9zKKkb8~no{lPaq?WE4DwSU|%cZ~t$v+OEw43R?H&dxGZko-JtgWbr7 zs+vy=S>KGmB>r)qI`sIfTZCl;?_x_*i87P&(C=*72VR5x3z08!&)krtxb`cA9xkBg zv2;{NrAueP-R_!CW@0Jr?G|g1Jzu5EpPb3PAW^OX*iSDJk*%(y4w7SKA1*>G7UxpN zv6iDbn=2h=j$+;O6I+pPiP3a@R>$qt%BZ&Xyxi zK|Czi88&2QK^4QRR`kPP@9L*_83BX1&`yI5nU=uJE@Tc^oe41gB+@RD3oT{jM4SRS z8OSl}Zw=4pvX#6xybD^KqMBiD_dpimhX)Bghne$its2hy!0|f5ggNNC&dux$8GW&VrlFV%As%w) z@u`V#-0so|=n+Ovg#P<{GExy4Yh-^OCxbdZ&MA&lnJ7&5^@RAJZ>%8(TBDKIxM+oa zVg*jDA&_bKyK%bn7i?e@P3@5&DsDR(PlOdUsN!!}{Izr0cs%8k!I7^dkxr~EL-;?? z(I~)OAWCY;!32MJZ+`At=_#*&F9>M!JMZ1$n~pSY{&aM#O++&(SKV?4Ss~c_Rd#z(EtcH&uf(5~lTC>kr?0c;LB|upRD zMTeK1&XqC#z|5suj(yzn?)KMh1bf}-nOi4dfquK+vvWyn#z8U9L9ZWcp4Ej{X=CGl zCcQuS^7GiD)Z3(WfX<>bas~$p%NMM#7a*|MRaDBZx?3KVftMSI3ycCwt~F*0jN|86 zz}?l;a&f=YkVrVs_EIf$RxG{(%OlU#$4OY7 zaOW&P`GKAG@w;v?St*7reKr()qbcQmg|a@3Gs?bTNl3=7g@$XAW!#}m3zHr#f@TNz z$^C@nk$*No{F*1;;VL*VWYh$`Dz3H#?CZ3_JoE(#rc?sDV(f-Ij1C}jj{#Hs&V(ba z&iN#8L|Xl^0j(fW2DF}xDEqe-n%e);dr*f~aB7@;)Zx36&{A~pR8p@D{wl4|<*|Xj zAT@qnj3T9u3Yy0OtDLdFO>?n5%kVbGbsXN8DXwdCvhWSL?&xa>6<^1TsVKbRBDomd zYMe*hDijz$yEo@po16|B=?k9kRum%@vg6M{6e)q1DLO!MA1CeACOzBQ93T#dYoKNg z;Xw(7^Hd|IDg|g@@yLK9w7`2ggoo1AItArOEkHdrfro=Odiv}q zK}{fdGJB^tIUF3RVl`@8<)RJDJqRp@QU=}u)G>fZ49*X3g$(lu-FCQ7D_rbRglOUd zFS&EkKY`r|q#0!WqEOoBF2s8wIFJC}d?}o7LqIN;SJNHz;(R26clan)nh@6%1G-@{ ztBF>U6b@>-gi&et@m8etN}au*N7Qf?Gvp6^Ye&H2>TdYr^b}Ujiw zcI_42S%i0+HIoi~$9?p{Oi zhk2!JLJ1!>BYub`f6cK9Ks>I3?cL51^Z zPKzE6stYDiSksm7x zgAfs|Xyg$LlN$+_hG~Vi4+<2ZuIxXJ`iMPpt2709Ii*)$hJZH8Xr6eviAzV;jMyy9 zjXp96p&;`TW25G=a(mk|`{Jh{J9CzT>|7bTg9OagkR;L$VYX*ch~m9twGh0QurkNu zX>xrxK`CMJk1-q~g-}$6)6gp}8acKIG9g%uLpLcsjK6`B{@?V}bg?{M#%Sn~Qvzga zhJ=b!dShiOAyZ$d_&Mj|Al2_Y$mhH2Xt^jK0S)R1FpyShVzD0QXn%a<=yDc*AfBuytqA(LI-et0 z9lY4%O9{9`W_$Xl|4Kx%Dct-|)^LR)|r9PQn?fDzN%l$ePA(B_VVKm1Aj^QX_k)^_BYr;7d{ z4?I&^Zlgh!fge~iSf7ALQ;Irrtc9s}%LrZrL=SGYwqeQdyVZprX? z;GSC08g4ph6vm!y01tl;On+X(1e^Rq5*kK<6_^oDa*^42EJ>*zhr2VJ+O*Z}5m<;1 z@=mQ&jVG~z-Fb+FxB>4|@4wIV&3)DZ95xoWux}PPNZoTn6!X1=LWvjeNNm@a>h!6p zM+b^xpo*lb!UJ&c1+hFD8f;}bx0wK^C?x=-32T4hS9yd_Kq+NYGjYISiKhv{>f`j} zGor*iT67ld6NZPpOAj=l%%BaYP{_2E4|hOkNO;oJ01YxmgWEkthPhar+?6IZN+6a) zQ;5=R7V1dV$Xh^s?pb^bON8lEQc7#y(@B7#38u9pJURgKyV8s&t@{jZtem-p-VppL zq28hm+}nSwD-0ju4omu7sPT0och31ypsQbG>J+$l?@0*XdFXdHaQAp8_CUvB*79T> zt~nPxU`r4N`y%~e9D3mYe>C^5@&;KaV%amXuR{~ce;}kC(08`ic635dK`i#aZwrBB z9o7^Clxas&N&{we2&-5lTH$yP|L7CpTNz9$5Nw<8daNN zqLj6rUO?WK6%pN~gxSBvK?~}L$`P4I10NlK!$v-9QYJxEzkG=3ACeBpVU-yp4K3!e zoLa`Jlml@oo>Nd}F@SiHC$9q_h{X+%=_|S=B(tn@5KUizM%H~EKOz$+w#kl^9MxUAio?Yb22>s zbq2(dLu1qOLC!0Rr)57bV6|T%uPHBJEf%fybl_J^E{V;?93qFX?fk{crLpb&5`1V8 z$+lwz)cG#QFf6y7f^^Jmg47?v66HXmCRk z?-?+UPGQ*J#B!#9GvdIg7F3FR)uVSrz^XqLIt48Q!IZf*j6Brc$ql3KnDY4d=3LvJ z!EE(527h)faE<2S&z=NM%;-I4TuR1@QVw1TLd#5BKM=uvrrJ6HxU-urTM*cFyy5k|g`)uT8FlkW^n7=Yu{2cD27yli=DH-SEHkxsjFyg+u9 z<|Npsdu7?Gm5xR}tvB=@I?x-yPXDzMT?fKHEaM%x@}bvQY6%Qy=WOks&=CR%Rgf;B z!HT9&X46w}?T@tBLih{&K~L}wnOyJE4(DWhCi?-fXhm>!0r5?XXLnu@S7B01HQhuw zPmsDrV(_L8y9wQcD+yebQvYC?DL}arntz&TYCtPIQ%o&a=|^YZ*h26m>AG69kP$-; ztmGvxh@9v6u?+!Ujd%r)dr!{7)FsC<&zG)c zH29HrVjrsg1V9}N(oO~_{OQtSBBIIIV`uySXfO3pQ4B2Y@G6!Y>iz>&D+#aRxK+v5 zahKLG8)|KgX5^i+wD$k-rfJYY6hy_t>CtzZ41LrI8bdQ|;0ykAzg+%{d~z3eiC(Mu zEpU7PFdm%v$3=s|0ZG*I!UCw1W<5d#ze=COE(K{YkXjUNzZ6(^7xm<6m{rk~6LVotm{6$zf= z?SzF2`u>x>{3!{h)sM*~%7cyRR+sh!?YiX*_Vry3SU8|3^ihYG$eiD)=vPY#!O#>m z2#CX=;BF5O`G-WTq zs|79wP$kCZ`{ve60t*1J2~ZDm{@sJ9j(X0p`1u-x0K-!MpAt}Rn;!4g zWiaIMP|ad?5j)eD0D1e;3SA{>y*KcTJS=2xWy!&U6(?qU@|u@fND7a6MpS;s@$%eOEjQ1$3?l z!8wVi^OBE#?1!cf;kGSmty9X_GkLsv<<|wQs>=IIaodcLr61n80I_tWHBy@9?r>G} z`Uj;~@OYs-V32})NVWe7tQM+QuX7pOnd<+;Lphgto-@ z>-Yw@wEQbT?kfphIe73IfYD*?7y!1pDj@~CuRU!$4hi0pKJ1IDlR5Xl3=_|+Cl)7R zJB4wtkJ&ygaKltBF98tuE!`KMY&NTrT%sO0%uLXlo32I*h%#0E&9mT0Gx<6$#BKzY zo-!-_K`EV-UDAN1q!<#QeO={>7suK6LQ_ikQ!gH4GarbEL&~SgoJWc;&On*fY_0rz z6@FmRDDZuIc?EGa8Qqv^evEF_Gw7<WXX_y-@mMKS+@wylM z3?1!6Cp~`FqY?si$#R51^C6MEjIK0gPn%sqJK`ys!6Yo`Rc-kwAL42!>BoILMb3MJ z^6T_pPH=5GQ~&?l90v=)de%7!iBK?lZwN_LBK~zO3mo=!Nx0(zaOA1K z^IK|kV+*4u^!m}vgNqM;4bJ`Tv_3#L>?(?pZFh<~t&ex^R9m=x`pNVNW{NS|9PcoK>@+fR#BdXPf0r)|oLKIB$eREg z;AkI*S^O?u>%Zi|5rl3_H}eU*4k(4@GG$ ze43ELHjWZ8=zx`-(X+J#$A6o_D+}kBedjbN13pjKM_Bj+!lvvE3#{oQ&$EPv@(%R< zMELw-0k{i%-RPq^VfA$nWfl$^nGddv0ycLHMHn^^?v#crnTV-`cXAsD-J>yf?Vk;8 zFsU?55aM?L4`vnSA-k%P7xU=!i4>^13vs5pG;m+P^ygj4{fUdL#zNv$hShPFYVcrF z|F>6WE@H<#B|`ve`@VGm!;55=QR@J|s2h(loWZiunPomI^GST%#Ny^1m`tYrdXcM61Bp>)hhWOzac$S3juHn;p-WP^>hJk1vdx z0ovgw#gyqktTaG_h*;18+Ju@xEA>qTR(i?n2LQITCP0tw!qx|WNCB|M@+1-*r&EPO z7oj3*`MLtk?Cqk<0e#$**kF;lHqXZj;G`(n_XAM^z5aMT0H$7@Vey6=Q_f0)?K%*# zPmMkr+F$5?K59u#ExdW>9MwV^p541w+UURo6FVPo^|REPRBqv7=wI3Sd}uzvI<|$3 zOX2!@Ag<*qgl>fopiu7v7%I~z`F8;LdgBkoRL~y+D;Se;EQZ+-s`a)tY@8GeKW`{_ zEB>$eEvk#a-90oHc-j5l2MR_P7M)O^Y=!r!dSw^)lRg%J?~(%e|8)%k>))&aaq1mW zFi2bM;p*`J{}ra=E@7tQ{OOf<=Ui3uue#DE7}+3Yu{NPS!GHJ}y(u8ztHzlHNasUM z`xcq=1+}J38Q$i#I?fo{+%Ac*g=TyHzfpN?(SC#0UqO$xm$IO2Psq|qBHD$XbHH)v z>jt8WdS6$fuA`Il1Z=-OIV)*@PR>&;gxNG^lU+`hwC-2ABA88t4tGFCz{B)d zxfiqYG0M5HiBKYsQ7H6ilEUJ>eFC(WBIgmABJaIXZQI{-5V+E1LSuUxWQAPz{Gupn zQ_WKso|rQV+({E6Z|eR2gp3T<3!@+_sA#RT%ZOhHL2J=?ZkJ^Qi5$gU(6YeONT|%& zMZPhLb3#}+F`9%!7HLu3E}f7WY1LiWX_~u0!V1b9%-=wK-JiSc2mbt3W3KM@D)UY! z1OXuld%0FV60D@dyRpRZKE$M?Wf4sd|P zSNp#qedbMMftF`B5cjfVEnzH23G~}=b1riKRuTf0jQR3qbl@%#%?N8#WC~=(Q=^N$ zuWvN|VGIzhE_(gW_Zmx`_$Ku=(+Jr%r#6h8@Q!scMrV$D=MDZh)s%e~?bCNgNP2W47_pGce8CKovY(r6S0A5`gi%7xLb#yH0PmHdOCs%>5l_jDp6;s zGJk+t6|BvC8!W3hWw?q|?ft#dU5tLf9AF}|WN=Kls=Xj+A6E++-&s`Owv+6nIE712 z-L@h$jK|(^RpPrS3Dhq4zSH6K&nTC})+!cCj(n>vP)mW36{JGj#4k+A7o`}u&B&a= zyKG+sYZ;dvcK;fVjBjf2R!5ZfY>xw9M!&OKPspIJ);_U6Z#_;|LVKogh>auW2K0Bm{-i8J+4A`>TPt+h@0?eBGri49~k%EM#HN-=!5B+yE> zsK`WLT_2it{df=5sUW%N{b2itGZnb*W@p@6w_+N7&&`n5I0laK;&bSv^&G#VB|4nHlt;!%Fcl~SdtBPE{ zs=bUn->d3_-31=|vvEVqrrxpKAWkf|>D)_0THz_>fFn2TC6bF+2WyW5ru!u<@(~GM z)TV2L0*=0WzW@&|X%3k~lP0yN>5M=WyaF=)0uIx$bq?`@CR{K#F{DWClwA_mJ}rYr zcZj9#0lkk4_TWC7Bh4kjvey@;;CE+B{V>>JhV{D0v*b;wISp`)DOlzgn&esy>o0Zl z^6W9eVgWo|Cx+ibm&_8%1mU0+`WK=qU!?1TQ-`0~O}{^Y7eF~{$bS_xaT+3zUJsV} zi}==o1vjUyQbg3wQZVoRY#?^pv78Z<1%tbr9G6?DfZeY-$_{aTmYjDRPzx6{AhTtu zkuuf3N3YGDhB8vA#Nj-86==sRJ#g3L8?8g;uHQwio-;vc(`GT$8H%IV@oMXn8+Nwy zhuCyK#oFpHkE-BU|J`px9=B?%R10#ZUwW%JsfJ8vY$%@N3Y9tW`+{I+;wJm8;;5<5 z6H8A|49Cw)!_oCa^K|d(@3bM0wFi;UJcR1Q_oI97Uu=FkW%~S7WTp6+>+xU1+|@_c z|J4oPhnGS=jR#FBUOm0Dm`Z-{A^V z_opEq>{?e}=A4;7+$m6u_;_Zts})-@U#_^~uH)e}?6p=cx>Y12Q)l2^_I||ItXHEu z+3#X!{UgTtWM5W?n}FNJ@AQRRAT(N-@tm0pQEWQ>8pm$Jp#JHFHgB%uGvJr2EuuA} zenRP{f$21CNwclE`UW=hdS{i}Y^(RY^H=4&#(#>Z8kCoP`zC%K&gjXXoG@s}5s7x^ zAC6ad?Gty+=YMS%;U4DEo6hp@8NEIBSiYR^&9yM(^wCJurRNqmQ>Kly`Cj?Qm@TX? zv!7z!I#+1|;K94#a7EA8F-3Z=5w}HNvzBQ)ix&_w?CiCu=}Y#xi288B2B=`~XW`Zt zy`5Hcq{@8Q>(j0YI|pJX7;j>iV}G%SyWgY4Wl{g_XbaV2Yis?%$~|FKw$#w-zjJihX=_k+1jXf#1}Md;av$=dzl4w4#7J?|Tok@v5wb zMb2->(m2;+#QB^*E3qFUFhA?IOg(3wBr*RoEqvI=u_CjT_~y?GaxL`;p~sM7cKy0zl11}C_)lF#kYHkf8{qnFptx2z_^jFV&pu0*c8~CX_t)vK zPi=~Y{c~`cWxT3=-T%q0)hk<3$gKyVzQOE#skc9^$%n_D@%Z~P?v~3?;BwIKuN!qu zLmuA(#+K;hn^=$J`xAYC>JJ4fB~Ow>D}BVo)4-XP`gv9b0YQ#c9`gmL^61Zk+5PF} z&2-+Iyo`o-&TFCbG(Pa5*78DLF)~Gk+&FoXNh{y)yHh8<8A|Xpm(F|hg2AeHqY{h; zaqr*0J$H3@;NkH@m1M)4yd1ZVW>>)qh@4R~`Q;7y>5gAH@d9{He8%C*Fl}BSxA9ih ze0{YAu-PB-DE*M&XS<-FpRKrJ!|?PlhikEQB^ARrn3S~uq{-)`WM<5tv)rySF826k z<#F+{0N0WGwyj%39L6%CKlID*uP^mD>N~mVq~(v;zm;d_6ii-dk^6P3c}4%HS6w+% z*IrwRM{pEI|3OF6-<<{b@|R<~cRTBaBu`r?Iohr@OMaM^#;nsdd<=1JH*4BU$tHU+sn@_R5};dtY-uLv>P6$PdP+K zrP~6hGJi%|xuv5U);8gHcoBR`)rO5ougphZA&E!avD;y&4?Zka#*^ zgm_Ga8Y{k(Xq>88B!AmdT`ZuA?gtLxhhIo}htOMIEOom`>dr2|RW;6r<>6R-=p!&z zUn%V`XxOY+#2f!{Yov7Asl~xg-tgY^Qt`VWku4jj3~{#DwIj*eS+e1exrg3rr^iH$ zpJ8cS^!533e1&@n$CoeN=y^FUeIPM0e$(T3jAv$+;yE?OyZ_bgRovKb-{1e|b?4&A zd((EJp&nI6$8Rrhc{yOkd0!O4N&%KwF&Wo{*zRn_VDui(EVr7+hY_25AwxL3%>|3%@&ux;Y}N0pZU}kVo>@A?VR1vu1}axjtcB5)(xjiR=reGU zeZFqMEjTpHqba{e6zMwQjCZ>jnsjqf@Pr{8U!mq1-_(vR?1KS`;TJxfBtyd~C)~<; zku5K?=1aRs$zcI(`VUXuVF4?|^vwrTi!Mg-{6ikPH74TwoFUUQY4+uYR?|pa@Ae_o z8FL}iSNcVBk}s{ZQp(5p+g*=|?XJ0NMIOWPH#NehzFP9p{VY|lGr|9b13B0-m;IM` zewyHKVVpJA_3rlt7j)&!eUA*`=@5MUx2=fe8oj3JndQ}M4aO>eSw*)J6Efy#BNcSj zCXm3BGr#JKK1X{PMHI#HYEC3v$vnKprIeA+e{a=oT5rmup+b1S{)Ln?<8%u9-?J51 z(Je0p(`&``jf~UIdFckt;J)arS&fN%s%Wa^bZO&ZlH2ix>u(&>v&1(3ViZ?c!#&#m z+vC2-<;g^?F$vPSbMccZyK$O_R{?84WT?mOvo1B2(mX<0M5GEwo@cqD;;-d&#j@v5 zpIS}7`6a4nwS4H{yULOQWocGPZb^h?e6P*p`!@UVk?otuA5=J}M}O8jdX*c&h_)Ng zUAdVvk54~m9_2Bul)mzBX|I5PtZj0U9-Msg0hWs?9!paSn>H#yDd(%*oMNw&&Et2N zGNa2FNb_*~ApU8F^+!F=_;`Q#Gy-DHRpus3u5ja|Ui|-!Ag-byC!Y<|t5_b31>k=v38v8K;o4LbX5M|n%(*XRfH zpr2*}vqZ(HX8;c8&lwc+*bZuwC7uT!l9M+YIXLCzMrH!rvAzGWwiX7EE#!fCH|q!Z zE;aux9SE^duIK_N+kzCa71OhYvM3W!N4=1T@TTZjj1$EbRAm3tr!j{8L7h*vAwdyF ztVIvtVNY#2bX9RnG5k|IMm%2)=0bUGacu>no8P-q8F~+U07G2Q-zaQP$Q{< z&#uPZ^f!h`VojX|NoukB5YxDBciqqfJ;(*}$r&|j#9_J=t53)O30OLj%d_YCSQJ!%sTA(L;DsIi{P zRRiwD74U20QhNk!;}?U#x`qPXUAA<%jkE_CecaX&0eHa*oybCN<*Z}OMg$qr_*&sA zo|ULI9SHog3=7FnG725=vsEmaHe0~T)CF<#D{p%2a5q%mKk((_qQ=W5>?VDl30#AD zeU6oU1wXAyyq|#l>4UNvR{ai)Du{x;4_7|~=8Dxd;KD-N;V;h z&S8R&{=Ij>%cXu3*A6|G5hioo9$}hCW!p7N_J0hdC#6|r>YG3m4vXxwmJ6u7mJE#c zhbd}zr|CO2kl+D~+GJHo+8xS(8Ewr?Cth@p#UoUZb@psKMu%HPHA^TCne}O!x&~{? zT$*y|O;eS>q>U>nOMIKB+R`$O*4icZ`xKyE_0@!)d8|^iu3Yc}_N=#`E-@MUwVwVz zp-RJT*Os!GRrVxlF!C7Ks`v|RR?DF7jsP5;44_`=H0&<$mEC8w)dzF^90kc*Njo`2 za$Oto)C1n;UoM4v9f=w^l)?q4`#q?9;`1DL^AHr*du$6wdqT!qm`$&}H z>bND=hzB@$7SHbq7Ae5_^0~+<%$w>i_!+PQ?V;8$dx0LH&yHr^A81W|(1ce`&{f$n zFg>`vBNmO>CgK>pR~kyjSWP zcvflgi-dOh=B`TZ1=K~PY^2JjUL!eULL1kXZMD}>d52*OyU1F89dzT1Pp&XGu4|Y_ zh>6GiE}SZ++ytZYR;qVhM$BH{RwO)S0sGC?DkHQV+Wh2q(R6tt#UjdG1Hz9GlkW|i z)09qy{*L_beS1t?Bf-+>F$Z`vGwKgOo6LdivbXT^RY%`V^qp_qPB}Irvun^7nRB=6 zH(SCJFXj)ldL4jkT3KD)p6sbNHF9%8DiZOAJR}gS`4;Z)aekqyo)y3Nd-Y=W=*`WJ zLzcRR@&u&M*cI^1e=DuYT$izGa&gUZYNSdsBTZd97y9_Nuf$Zby~61!Oj1@bJCSVU z#P^y8pMPz*3OT(>rusgZ;v`z@j{*_>|28ouo)@e$=rwt#Ml$!du#rZ>2r~CF*uQ^}S;8w^ zaZ<*;mEQmp)^l3+#)wyrZ@4FgnX8&kqlBtpZ+fm2V1p=HS3Ce+z}zQd??v4?cx4Jf zH-_P|pngA?vWo*B?W4P9bW;>*1o%wwR_hObouhcYTzK%$RGI4P4t**KV_BUm}RKMGJCSD+tg6J+^b&Mv!@`XJHDU_!*LWc!(5*F5zj))7-GlA;hxwR`xZ9ui4q=@F2 z{A&T<@&G4O_&51Z`uf4!L6^Xwm6heJSi%iq~e%6EuZqk83^xfqcyL##IKuW7;*$#XTo8Nqc z6HKrpz5ES$PIP8r#|ZY2K}}rS7|lDvIf`}>l!%n5o1STi!oNASeDO?!vbi8)Oh_uQN*O=QNO3Hc(j#2w1v(Yuj z7=bHcVOI*ptv$s;hDHD#fsjGf!Iwxu*7$25u>!=k-0byzwDYavKj$dqw8d5jyvq>| znx-4hYMsJz!o70D6W)-fEgY`RV^iZ#EE!2@DOim9W(~VnP9$BZ6c7?3-Kq=rz`T}@ zq$FFL-3|guOfiPw2nXU&r~v%&pX}y~qnXaQHaNmrE?5CPuRz|tzv&htz?}Ud1why; z=Y7_Oxp;l@kY~~@Nef)QadO%x#z?p zxHs7dEmOGWc)S}29!~BPvgtR-`%+~FX4gIgE-N;#pW&7%V@KwJXW8u&N+Vb#=Z}d{ zmkjK)^#@1-*B^?g)MX42 z(=~BNsQnz7*&qglSQ1Q`Zk{x42L&w&49C6VOj^1GsW~$ryV4*aYF`Y}@7BsBxe@&!guB{&N)&Yc8A2QHpn4P0kg|PiXEOi-v`5zkeA@7+D$~pkg z{udcL$agx84$~G@EZp<*?~&&%ITLO0t)G2w1d6aC#uzg3QH1FtHWcYXn}MVuuE8Pn zTEN;2^6EX15s4nKxB(s&HSofdAqFL3hY`UN>7SV=6E07B1>A2u{O%@ry9!P@_~Vo?8GqY_Z(mwtG-qGp?sjR~z-tCoJBNX(Gk-bH zx*65#!V{Qz{qKi>`?R!Y^7=`ZswG7c{dAfwY1H=DCw#HNsxCZ0;~L*XV9i{tfyfBn zYHx*-xd!~l56C}4_nZn&{qV}oufc0N?qy3zna4H;aJmTS*m8EeGmd0p6P9Fq!C)aG zfI7HzK8l@Dx>1fPT6{GsIP;_n{)B@6^NgQu-}VwPefH@wWF3&gUVVvtx(s{*_E*GI zNM`CGuW;~p1NKmvBB16a_#sa`-Mh6@vEgr|nTbQ=m~nMb$o=FJ|6dkATM-jWsS#6f z*rRsX?+f4{?A|)|9it3QdiqObj=Eq1d?XVpC!2y@>mY3?bwiq{_tEaq-CGAp!?0Yv zx{Px@zrep39Uc6y57;rdnZ%3*`9_x`=*;Y1e|J$nAF9&r9Syh)72$u%r?{LEhXi4~ z2guc;#9G&RR5Eha(R$7C;*7=3;2F)zWb8qplOt|fEx|PzJGM|T0zSR8{SH7!6*nRa zT>XdrW~Fm$0sHhN`UUfO)<{JiIB~iMV|Rqi8~`8o)by$so}$Pgaa9>gk z&{{(673nl?8d^+u*1bZigtqYxv`XBW-)LT1YlDqG&4x;8RJ~XLb`Htn=ysAfAk?IJ z;1~e@{)M-YDO6)AY)~A2gP9_ z@K)v!Y0-c8>-ft8&#vphlvKCP-pY%mtDfW})v7Q1lKgU*j^9Rc?7-N-Pb5}7B-VW0 zO_)4nKz3}2#@p$euk)LK2~a&m-g_OY`Q@)hz~=#w&A;0a?x|OC26DvcM$t0z5M|UY zQMRFCq$LT}92uKL20|5*c?(b_wM>i}`Vg zYdpF+&qJm05V(%FIYj=%$c>Tu!9byM9vWU2i5paSsQbc7;Y9JYzA_{AQvhh$;Fyod z9u*XO4t)fayc|L9s{93fZWfFPzLqROJG2Tx)r$;ldZ!*W44xn5KJ%o^6{y%)Nv=;x zKz_-*!rid=kM_PG$-UdfdexDIR^X2K7N;vAqV+=8pvfXPfaw#p2i?d;SNm403J)AF zjtVX*iwMX>44r}I8WeoXk)iR1wzt5!k8gEtMURP8bcrfHii0D{Gl7VC&z!7Xqt%m% z??iVkgI`hqlq81wst`deeEiY9k5B#tHTw-4?Aoka;&%s52?*s4pkKeY`dUFO@Sr2fxZmA5QP4&EKlVemQzPoaTJQQ0I7o^k#(R{DD7bXToZb(47 z*TOldRaM{TK_CC_u-a<_+|Se{IuJAUsoP`K)GHD0)iL--IMY?KQBB~ipl`*=&sTxk zwtI(2m1MNVA9popW{aH$5AHC8FVlA4j5@kaset=nTca=MHaygpYMcB-S9vPfOCtlD zt_VNt4e2imIO$#4T&*>>vNmsaAwLhQ59_*b?`HRjyQ5&9$4xHJs`L{Z<;4*4li%;Y zc;Zhw2+OADM*Yh)fiXysxGIFBW`;g({!KYd|KM!w+A(}83UB)-PzihxCM4>uCCNV( zBc(|;*)@$I6%+1Go67*@x~q67xN8lmGD5c=dag8mZCZ;G;bE30XAeB^00;QINSg;- z;+~abF7`jtLycxV;yTev5+RNHS0BLH0jt(xQ|5 zmDLH~S@%sAUii~)8rhR5!SScrtB&VK{X?a7#ANje0CFRvUH%kcB3tRBVb0*R{EC#pwFZfG}SI$JK5A0jU zMwDk%#<$R%y{lE_o5kBPAUYjkK8TePZjdYD#jkeEu4YfK_A_+2%b8w21}%2BjqVzj zb>!64uWXlI_WEcJAAqI%LLBx9P_?o|FkE^U;#TEbd~$B2oy3EnZ*z%LPsxl$5dA0L zr0wEN?K_Xwd3WJylJYsY_n=e;>42THt#E8Ig+Sy-+UQtot6k4L>l7J~{LG)B5l=bB z-{9?(^Z9wd-H11mZwV>zE7RumU$|?~bLE!TQtEOfx8uLLIz*|L(k;1a*Qk_dn#u-) z(8PHcd0VW1ES?7GF^J_Yr@go{{$jh%2n}?{jbQOs=;LzGE{ftiZ9ky@7FjI z>0{TxyJyhVM~I12x3ZTU>t1l1;ZsK~_VmM93=K8EFG63KueiH`n0jJqB7g7ILZ$na zg^)kkTwjQcxJP?&W-Tr!aqT{lcv*c`H)o4WpoLt~l>u-=>M0Dpq%kHjG5@%gIUHat2HTx*p zPU`9XJ6%tfD`9VzYOsK8vF_m#e-sk zZ93uayg8fPHezi$vA;z1dlITW7d{&fLRV5bM`N|>3&#;+T{F*~vLV0y*5$WfW#F+c z<1pmPh3jK*CD5Bpn*s`B-3)S<7Bf6Mwe!O#8)mP6LxeP>e;aP}>-gu{4_qnxK@{zU zVn;0Ggdf#%F9j+e$N;~iYvcTK3Ismh^|RG^ku&-yuwPinvo6DOhnuDDI{BSENCWU# zr4cuck$P9(V@Z^rbq)8}=qsMuN}qh3 zY908Hpe5+FG+a2n$}9!Hi+sc5O^KIl#!-?@Mk!73lR1O$&|VSy774>tzSoNB2Tulf z8wNT2m0Re|X17=wnm$k60AI?AcgFvmdc8w7_DIeBVWp6S)a z=G9CAhtYrWuhd7YHNj#XD&#ZTKeRuDqrt9!{uA5@YoA#q71Q7BDjyhz$K3e<_xfG3 zv&U5H|2%&xdCwWQ<3)X2-vah<@k4Zs4<}-V+)8D_3 zrXn0$or(xvIWI?JKZKujZnt3;b~wK<)q!K=JcAyB?`tf!gVTTyxLbZ3%r%oRs{`;UHtM<$9rZ^dZ-~}qv~a_kYkZb` zd`2CV!ia0(v)a=Z_tee2hCi3%W(Si3tdJ|v7y0s@wFWs;_wbSxcsqqq(J|PJu(OTF>&jPEn%CMdlo{oul$CF$C@_?M25X?7n6JOhHl*P1 z5&ZV>iIq!M-_9M@!rP#^=81Y zot+Cb!_sa>5(%U%l%c;oV z1`;%}yKBl{3rgXafj9V{FRN~R2QOocRdTy@>V`(Xh-vwzke{qht$o&CD##!u{@I+f z4>h){+V83KgUjS~nfodI$c)Vwn;rh<*SjV2q1X8LBlx)i z5{;I3BJxW*;{&*v{q51WxcV6?Xa{CtH_6sT%LKwEP!p-@my;86<1KU$`}phi2XGc1 z_rfLA&AHQW@Z!g6PPWne$CJMiVp%ALdvIf~oYS|(EW-F$1=d>s%h`509dN|#TtPeT zHzEK8?KM6ySPSc(!t7k+a_DOZ!^=f(O?69_`9QuJfD||rRznhGkuFTH+AUo`MZ;~< z^;JH?JT8*<;}TC_!mA0Mv&QZaq;c$hJ&;hPI?gVIQ|=(|2~#2hmiUcB*2xT=rnKf8 z@jgrTP)6lmOpQ-NhW8IA}%bwt*K6+F{-Ma_~40XmDQ39--;g2=}iUI`D#GdJrO>4&292E0!h7 z8bPoI5Y&yKY3K7Enoz1KcN<)fGM51YzYUl*O=!G_PM2yN#3UX$i1Oo3}z+4yXMF*C#ZYDPD zE=uKc?V!PI&uipcUMJp!j&Omgh?*dNa5St@MqdT~rz~J{wrl|B z>V+S7q5i1offQ5x@he=AwY#ReF{Jh5xZs2rN8Euhc(}YIPBr14ky?yJ2ITb6Ol8)H;K-x1)jMlI&k=5Q|;? z4=JKFdeQ;g5(hDo-ZKIP{u%%FC?i75w=6sv`upIyN=PgS)@^ZRKtG+r>`#2$Nc%rl z8|pOo)=a=5vhO#nX95kfkp`CLb9cW$ru8{&lo2sps zjqwAI%>QxM7<8V5wUa6~e9VhPN#LpI^=fBGt0}3z@>#s#M94hKk%JoC94N_Dtej1pFQN2!gLyISH&rSMZQ0)(S`x3l(;KJkE;;fs~k&E?1WoR-Wb|mNo<3juglJYAA6|A-Um|DnwR-G>Nm%?W&~&8b%}dFU*ZvQ zISHSx00~V$&gi1>Tc{?6F{pxsR+C~$0E%>MQeY%e+&}3jQin5Qip-&a>Oa7C4%mhMLO-Y6HA5ZwQyq4mBe)YRM);))8 zgMj?#It$5rtdr!;;8G|L<_om{uuomytQbz-U}*K?&%hq&Eh!XmFKYlL>hPANyE`?CTT1HFsItwtSE+b%S{%(`$FC6C40Jh@M%zrGJQQ$u+z4 z0C1o8y*SqGJ*Hv|8LHUKK^6_JNJ3T}i#R#bKSSU3D&p&h^_(VV|A0KN#djTC+kg!P zwrJ{G!zT1`ash~;M+gC|GAY;t9RtcY?~(N9=@%EP7(goM3BTkRbc1rYVcz5%RqZ*^ zPdr9S0UsE)_aD{jw|?lH>EjZEJi4b{kvKv19MApwf(2NT2a4U^8Kxk~ zD#m&T>OefQx*2Vz1CwN<4ULIwQ<-M)1NohyFCV6A5;Jhcl*2txulq$+x{$wZ-s4W> zTIno=*k6EHBz}8FoV>sw=#@jM?!Gk+FZv|>@-tSEOKsJDAK9CK2BS=r3~0o>YHAo% z7DR}hk=7ikK0%M(x$uHmxekL=GE?oKx~BRFkS!LgsaRuh9dMtz|3?Q*Z6vQe&k5Dh zE+%`X-7JOwYwGnrAjMbHk3ln({Fo@oL$9B6+(m*kugLY0(M(Yl5z6`wDE;T?PsBB- z*V2jeV8tlqo0fU`4&X!qJ?yn)MX6|v$K;O3Hq8Scr++ZsyG}wisMIx37W|MSvJ+u} zx_N76_@KxN_@yFx{Tq%d0H(;tYe_MTeUk!k$EUY>>Vd+Q?+%ms=54^$l5~*19)z0j zOD0AA6ajNwSdx|*4H4E{CkuUuYwHTH-0dTc3f7b*Ci!pP_N0KXQ5xoeLy=h7Qa!Jo z^VAO-h|he);D|yk&B<6fX9f!OLU9e0^UJ#qMdHdm7nm}IL7Gha(E?(}$Z{dmtL}O} z5hR}ML#1cnTO&Fru8UY^)kkwM_=Cyf0yGYMu4qN!KPA|>p{szgFdzKe6Y6ZbkL}Rs zc6VGn$VPYAw)xSW@Gh1|@M*^{VTHi{C6iXJk<5?%uaKft<`CHj>J!?N#ewVJ>kJyPWlv(@o5qdXUnTEHZ-zSbnlMJWWKlP|rh; zBi~S9aFm{DDIm~nRez@eM0Pzq#P(?L&*bJDWB}2hb!;Hy$=Ok7nKcL=AcTu2t|P!z zAE91)AQu4=qCny=@KI@*cMv#{_>lS=Q2ZvCY4;c4vK6FT&9N8+L>_x~ zdw?wgb!v0BS-h5F%ug=1!x8hBA|8ObslHP^AF(7hY6L3*V%mH^pXXxLSG5up1mI=L zOZ26Qja5mv){McU$2HmPs#lJ*U7`+d*)}W@9*MV0_&$3ZK5t3+rs^t4h=63Ee8rsx zZ)dW6W`%TV;)~`^G{|DDEdIQD@a{i|(2{yCx4{eo>vj~E2g+sF)x~IUM*^dPRg&vC+BtTE6H4bhaf`5hijG+DhDvDX@p|eNn?jD2Pes%` z{c|AeOTZYxdl(Tw$EX&ykBn>>OkBao?+!I#b0B-0&lmK;LNcNq>X)Mz5&QSQ7dMzf z<|MVGrvs5;d4K!BU-Y~PFP#pY!2c!q`sF~@z$exq!~$e1j3ENlx2ZsaBkZQvk_(h( z6d`8IwS1#4@xZzm;!&)$5Bw$TUpR~hhx!XrFY2sHyybclcX6%4~ElXyt zNXl(^2oO4v`}rKV+AiJNoM~SaALfyy`gy{A>e33U37Qq%VA{p7PcRqzk{M;O#D@ZBN7ty zJ@s564qx*oNII`|a)5YUUH=-wgI$BslI`R+?c9T8V>)|_5d!MscLwk*1v+I1=APujzB_rJ*6yqZZ%S^ zZu*)=eVm@3Rx@x}=zst}fo%bkrFTzZb1f^pW@AtW8C@{I^D2TEyK;rA>CA_wfUGya zdw*4&F@3wVO$mHSHUp(OWrc_wKyU}S*c)~=(smt)PTsHDG15+O-`N(d%|7;1eX8~` zf5xI73ZM(cQ#2X}Il`hSJ7I@QH6?I{6MJa-iRI;nji{uVz--Xea`_4G7$ebL7NzUS zLG0n=g_ns@@cHAzH;cp03xS&4)|pRYop2B_e_BgF0O&B2Y6RZO0l_~1)zkEs@Ai{6 zs8@QCzra7uzQhz!CcxdUSs=vr+4;easeUUF9Tsz3+t+>4jhA=C5n{LdcS{zDL`}r`WC$1Sw(#ynZ3?E$dOulVi}j8CSryQ zl>NGN1sD?um;#BVfqltJXeK!cq@&W;?Mm~2I@-a5(su;nEX;JV5^kD{w6u<*}Mz&&f4)P3WTQ?y67 zxHkSB+~k9)6u=quzF%gbRTR4Rk9!|IH^-vJ5B5?Nna!tzJCrB?hUzHgrNP$XICHCF z%tdemItYXeEtCMD2(IoY<#XF?bBK~xoub&GXFwgIaT|8|a!UCKwt&Y~|I(sW(+A|? zwC>oQx6JLxh^Hb*`d7#!aSua@9hBtx4>ZV1@*ijFt-+9c^$(c;wzg1zeq@Ez1J-z> zx;~NCULpd-1d?=*k%V;W)%W$4A8-Hb{CagC&9ufL??g7go2mM42bJ_okeR%n@*V6I zpzmDf7bPbhwtX=DjpV>6fYg;rSmG;yD`ep!IHeqH{rVd7d1px85!aUY&6#n?A3pJBp|@F?GS5i&M8vYFgjkl^eanF z_1D-rPsOP+@vkYOag-%5c=BNv#F>1k31T%@3P-T;pYV4+n2SVrH~EA9X80;6pMp8? zTE)ZpHk$zDTfhuMk>%jY9eIQVdl&pvih%O`5z4T0l*dKI?ewobF9DxNgzjX@iN;lc zU9Ry?gMVjHZBMyLz}k3oX1FZIoqKp{6%Rj`XW?q$XRmJ<*L z=T=uv7t+i1@|^O{wP;RZPcrbj{Rf!Maakfdr`zgQpn`Qz?0(S@l71ulj5Y)uH48`% zBc&N>Q~*xGr^#eN%f$zAu?11aSaMN9sB#JDJ z3Vm@md$pn91T-n7_SAFfN%-+%P!aLPlk|L6*MCdXGxqbUV~Rwwr9Al+dEQa8*qK)(Vu0+O{Dxvzg$C z={<3eijQhl9~D~i#!Fejh>oy8G&1T3v-8Com(<3sTE_Y#uFRWL=d>OkHLId-lnV_)|KTL4b% z6hmAqBips-1a+o2 z^>RoKP*zf>?{CU=;4%4$(2JLtz9AGnnAm9EF$F^#zw8wrlMzZnx=hZ1SsM&;5@LCZ z4yYSvr~$EpW{-lEMQGMPZ6iF{4F0|!(>H&L?({N!e;t?xSHhd}pOJ<1%QnDsOl>!8 z5u5rM_~2qMaf&u4p|^B_+6O1{?j5if#Q?~NwgG&&8!IKq7c{`yd}^lMGk*B`fE@U& z>MI{@cz)gkm6ee^jEuv&JE+b+>wO90kP0aZbDQ$d3s586nS2@1SUH#Wt9%^E^k-TA zFDRRC_x=q8moTb94v?JJJXIY)Hs4Kv-@ShDy4WKS;TU}G8cnUyw1ZL;7I}ahiJ##} zj>G?aLD#DesEAeIxt37ZpMw1TXfXS*rh^%?r=IW52MMrJIto2Pq0AybVxRtXCv-P}B!FPZRY*vmMun&OHi!F2Y#ms8K*A#rXt$-Nv(BIlP zFkGs(>fp6T1o?aICBbZ@)34iQAFo-T4))J-IzE}69Bz1yZw|-3yuJHlmSdO(U{7JC zBJCskWZH>CwW6G=hmM{n?MMqxy^E`kuZXaUyexG=9mvYl3( zq#hl)W@b8mZTaBNx;_YjCg&kH|F@ec_4>&XnELhXl!Hxxw8~uyhTw8q{-9G_D4w`1 z>k*tJ9%Hv*S&?hREna#GgVKm!A?;VTlKX3!_R%-1m#roDB^xmc&u|xpC8y3{C%L)o zZ5;2;_ROB{4TXhfAi!xx;9<=N7b#1R`uMvQHuvjGV3Nqe03s8C!Q*H;LXOV6uo_&zVnE9{MaF8>J7%KouJ z1B`J9Oy67wCPfs*fSzAFnN`KBcoTNw8k%baJGVyl=tz1VRTp-OHhoB7AqOz!TUd`^ zg~CDe2Xnm!{P)d>6+H-1o0Awq){sk`_)9jDYXI9nOiM%fbjw1*HX0i?LBV>wR4%kF zM(2(vfI`@X+Ft@KDHu6Vj{_hlU>D~meZr^XIQ-Ui6nW$AX>LO)D~UzxN1Y`-s6Hlj zD_HAk7#6j+7F!U zN2rmIAW;D%EyC(VnNMIngiWWo@w1CPt#86li0<_-Jr;5cc-baiH>q z4oeJoNldh7jQvu25IFhx_ms@~aL1FzAN{{xeK>E@lt1fNvCr~ABQZC(?~Lb%Y0O_N zQaGb8f3sEbCLHpYA)M7Y3)mTV4&0L-*T(4c=jT`##GKq4>yj|BruG{qz#wX4bm0L^ zXEuN4Zhf7Mnc%glkI6X2%rh`l`_f{*o)q!a6J)5Lbwo)eE&^$5K^!H^`NtG1W8?6Tn!-hLZpH+{3IPpBu(`P&D zuD4H|`)Y$j58M<_P+%JFFcGpj)#~5jWNM5yUUa?nCLFyTZhTHJKh^%JJxEDqW7j@t zy0rCFq#AvjN#<0wwUdzeXocB|&=Fra1$*O>2{;)Uvi0VvfiqP58v1*tiFzr@{OMD+ zx5BRgP?13XvF)Beb?K4bgmUZ7a<6sUQ8IxHMH0GNnHp(xmz2k6Q15K;VL&3;!35laU9Zn>BgkQg7@_=lEv* zINQkUQC}^D%)2QCLPd`iSsR7k{M!q7RqN5O^Y;`G?vlucy&_xtzV{UV=6-bz_edpu zLGop1gq_Z8>~1<#NwN3(y&n=ib&nI@=ceD1O`FUT*5o;ITJE4&be`CRlLSrPzc`JT z_+T74XE5+nZc3H>3rB!EdX9|$%lfe4u^=dmhiQ%7GsE0eGpx3d)O4^iuXp>iH{vA* z?`2!N_hM4&*G+M438~>{n1s)>iEYCnit=YQHyTr_D#~2L8`E>AXI}|@$vj1832e^~ z49k#;`Yj0Wy5!VtD{T8#^>6LpW^MEjdLY5t=*zj>O$d3YouRd!@#7x;%@D4^uxH=t;qiC(edSde!;&$mO9=e|6O=U*7k7Z5n8nS?^ zDXX<96ntzo4X2b-3k8{0{9O{LyOM1-_VQ;7C8as%fqQL!o?HA&oZrz5f;9fy66;kj zRv2*6ND1dT(>?ws7N?`YK2=-u>WaQ=|4>?)@K)%0Hk#;2Oyd?FEXq0m^Je}DmeeeR zr>|gceBbP0v3W7mwMmg#d|GQ(f?Iq#xx#@jG$QJ1YO?Yf)8UauKi9<`)weUw%aP`6 zY>j0He>RwoqUlE~MsDOFF;XoHuX^(oZ#24!oL~NMPAT6_-_=#6>~``IA{83X%aUuu zEzXawq;-9LzE4a(VrjGjXA+ZGH1qUbjlOw)i^{n7L{@26Hng+iz#%x9TZP8?s5n2S z4qY{7z4>J~DdXC_Idiq%t>GY&7++{CxUYM!HBl1XL+=L5X!N@xj`n{&3}n}Nb^*|3 zYdoF)b))i_?MBfO$~7q-6oj}50UarPZbH(_T@)i-RTS0 zf#LIKHWR|lr9|F9#g+_($jRf^C&TSOT5{G*H8BGQ21jy*lZmX~*5*}GJ`kqX?Ym)& zXA!`Iyn>kifFk5t7ea++s^+1MwdF52p^G~Vz1J*s53`$}cP)JBl2E4&AO9EbVo!D* znjim1$5l8r;x`O#JbEg%~hQl4LwQ>ParG?5Dy48BC$ErUhNUy**zN^yW2b4$KD8Un}5;^l1EOKsd-JaPNP}ytS?U<<%7a ziNk1SlZ$MA4E%xE!$4|&h1R>gp)He3CX`q1*`Vdd-%Msr%y;RSxCvoG5fozA?;s^> zDX>g6EZxvBU9fXKj74KoFa{|~R>z3Fz264wv8Z^LgPt1l{z3S`05rK+I!QfyaYuZbD6zgX9^yB9o993(ug@N6uN)Lsl(QzZMi&zqwy- zfN-Qc9l2344?8$i4}PPi z$Q-2Un^8B2y?N-1s_0-QSEKwv#rL^8oN?ojE?%91*{T%z;Fj1rF>A7_z z|2G{rmB#b>R_GQ^DpGh#brk<;sx_J?n*o@&$RZE1=Ra!8GpOqgSty)Q`o-mL17G-j zwGUxRe0{5#21xg0(R3ic6Cnl6gPuh@NrIs&rOYw1U@X!6FHcxPCBKj6F{;oS->%Dd z;I7QN0=T}h7nTc#4a^eKlND_=^AGdD5c1z8={Fzs--GUWeb?_e{rrp=&sHfClrb`B z_@-rFD0DB0ty(|h6vEVH8>XjL!dDz>$ z*FBZ$v}}dx`fft5VqkGloN>BissBUb(430N&3D3rSK^jEp+A1-K06TIQ#5wZon4LM z0C`GOI5St%Jw*PqzkNEcbwsj~-u1-;yK&(spMqJ5uZ;@a$Y7A@lJGc+pFFO{PMTA6EX$(Eq#g#d)Q zD0j{gqW}Bu=yr;eJewpp*zo?aYyK4MpYs$~eoU*Sk{Y2g zg7NVX>Q-Ri6agehPT)GEmp=$#8}vgkEP7xhZHn_br#lIS+H#KDQcA&1FCesu7R0qj z9cKv+0ygL;#bhI_N4D-;Rtl2Y7r)sqiO!$JIk}H45(2D>=cHy@ijUEH>_6?J#$SOgKc4&@{3(E zI5U=cG$)XgsE8%_(+k5LNU@ji%?VFA)-!It30J`PDQ1C~d#1e4W5$8mZa9s0pm1`* zl&#b^wM{moFm5n8?T-3sXYr#2HCZZ@ACsk=>;6q>!0@feg($Mowy&+p_bEW1FSMLI z#L+@1&^bne-foeI7f7S# zPm(Xcz2Vj*4cV9MhFlz&Rms*DTy+$WCH7vUW2A_;@AZr%_!Y8ehlN_xJRqDsS@lp> zr?rd|GzetxdE(vX;q>j}J(?PpsE00n6g6oy@TfFp-|}bO55O$A4=iA|UV|;%%)EhR zQ(_mU0#bgSkDT*p{*D3-!>4XIRE2;doIa9hAIX~r#>>=r5FWV@E2E@%woC&FFCRcC zp*M#dOMTqxVs>IihCA$}l%BdDS?Gj+etZ<@!JfsAR_`Uz1~c8WdGEwB=iqzVy;w%qp?r&b*B=n}g0n&8R&+33^#RE{wRKWK zZ_l9f}`L zUzBY$I7m^31w}j79f9YH*$>q+-cXg_m+LhK3Zb)b&?0rouk4%d_BPXcC@NPV17dg78(w*K80~OJAV7`w-Q~a1GA@>!8JZ_Li@+u zt&aA??cF)vp_V(>S#ZZrM-;A8+^lpG>m~Kg$G6xWybj$Qk_qk!^jNt54p#ZL z@3u9zz=@^p-X%beyT> z6taC!o_N)>26^2)pvO&U*PoKlCWsh6BmTlEFqQ9!Tl)9YEH|M*)MTC2Bh#jV)ZJ(Uy zHaoVy>(#J;6Gd#4J?^JDmq;GM>o_R)s@`Mr%a`{=Ao@0+iuJu;dBoTiyU zms@|zY$>HF>HP1iQcBjA{K->z(UZr2fppP6hRKK9<}auEu0rHNh%aCUvzu@Yr)~D_C7L}7G<-K{Icx1!gC}{LS?snE#M0m7jOe4 zxxNiszX>;kDhY>^^WN{TD)%^(HDS% zXd2-D3#W^)Pw!CtgcKfQFn6k1;x6ob=Wy-u5mcj)hD_oyI3h()fSk)Ul&bSlYq-s0 za?<OTun-k5io%HT}yQ z6#IOsG$N#O*%21g|L)xfr(Gt%d5^eHAv~J7FOfEeMM|A8HjqXpRTzCs6zn7%_57!1 zzGZV4E8frsa9~3vW;n=GA6yI8rOr^MdDZl%F!{Xe&1QMKMz>W$7{%|e4vfixO7ru4 zwjo*aRqZ3LTi?sTkGU9jFnyWK(N;H6Nk?+Ys^K3Zy5F{ZG7n5v8QnU)KwL9bT>7x{vf!rOjJspF|^84pav*v z@Zprf;Q#Az--l3>%upBeqm`>%Ue-T~;r`RO_d0QaI-2`jl!uP!?ktio-Eh6}vRr6| zpdp=38?3f6o$C`YN>+{fLD+m2N-Mq%LXD=7=BGTXjzgzJ*y~3VWiTRaZfB}s(2z)Q zBC+88)r3?+*a7`Y*wRBITCi`hWWbO}59%rzg2B$IT=OKbDJ*S{l3M9s2Lt${;sh1d z7I2v^l+Gs*@~i%ela{o3{-s!M&uZTk9Zo=dau{sdT~)|oE&9>&792-Wz#W6MxZhfH zC3_x7FO|v|xDF~SSJ3RMj^3g$_-ERZS4ilTc20A1d1Qi2TsQ|nSo|NNt~(y;KmKzZ zm$UcYLiQ%*jBJtYk-b8kl$~>k?7fQ4NJtq`b`IGyC6sZpBC>Nh_xXLkzsKYE`2FJ# zkGp$(KA-pJ{d&K~b66OT1xR=@bci*z*vr`Z;5d#^>F1xNmtlh0L^`2zf$jQQuo9eh z!w1lc(13r{8qRH5z#9T`E#Frl zfFeYBJdYgF5c#|wKSmVd0W~pZD#V~}%dZ3TUxREljlEv)jCh`zBH*KGMd6IMY|>V*!3*jDmQCO%B+jCty}i$Jk_MzePp_DDnCboeJgE%cyYhlzBsT zbN`Sg2N3o^+SH=E%l)da&gQH)eTkXj&N4cKV2{B95?{(noT3Z8DgNTlskE@MdQn!vagJ520~4Y z1;h(=yWX;7By4(i+V<8VHk!Qs+vjFa0vV=00U~yCfP8m!U#aQu-Dh(xoJD&e=B%rN zu;FaFk4drq6}LgT4_bd8-K|Q~U8L`bcN}+48JpCs%Q`FA$0!lo?#}az{5w%mKU?=3 zkm?1(GlEinA@+*zSu~8F!r!_c?1|MCpVXYJSO$ur{IYzcynvOpwZMEF=DK6?#9K;2 z@HJUrK7Jl3^Zex*vXy#s!2jD&r)S<-L4g=`a1cYs=ggMIr`JND#PGzGglG47U0Ma) z3Pda3A7LhwU!Uw|LsuPBc>7duyze{8O-JFYwi|@*dIi?m@#M`9n{{SAdvcm5P!P{^ zZ^M_`bv7gYC-2_i^5-9SHXA!E)inH>&&cNwG$mlVz1QV;ezP6;bDR}0yf0&J zBYwK?sNsG(0b(r!UzF#xqEUb8+zIZP+dyvQ0a>(WUWd-kMLGhgDk{7TO&6>xyaZkP z4Bh&IW>eHD3oN`uMd08rT|m!Y86yVXs+{bmO2LL}*XW6qm|3ruNa)8$=sE6GZa8-; zoFk6i_Fug0>s;Fe{X6YQyB^ANDJUI1} zJbuzhTAHw_@6=?zI{X`0*evx}1n$%82H}9_7vdQeBMQq2{Y70NoO*8U^IQlYIrYZ7 z5F2~95Qp4@_l1igL(KmQ-vJ7?$zrIcmJWrBWqz*n7hQ5*elV~k_J1%0ZOZAb#J4%d z&$alyZW#QQ^BW+%8#8sg1z6n~5l1ztP|e^);~hb|ST9QY=ktREQ1l~J1=NMz245vW zcr1(Na(qe5@(86pG^%NUNA%%;XEh97>)CY9Tl>w>^s_w`6$|#RcjPp}0B`q4oFU9j zKNC_>`wY4}exiYY$B7DKOZLNtUoFPFO?QMjQ!8qvdU;c_XTAUZ!wRE(L%1Cm~%(}4JMnGCNM z(zD6>qtIK4H$Qs#kP(kcD~RXy#|?iTE5#jRS`5%2^)36@S-C?O7%UwwM3w@TX!(owBcxGOm2`V1>W5L+}|;P07C98`0Qi$QcNqpu@^3W z^fXX}d9{e*6}Q1N;=4j+mawZn#cqGK&KU6z!zICa3#bmXMVFrp4&LmWz2VN)4}`BS z*`kPM^o@ZD^DlZ4*2#nSz-8V{`k~;DCf1H}5;?zvJn1g^HFBceskX6^f@{e8#yv%O z*O0q7O_m*YKpBs)=1g~O+DB0^jC(D_HdH`wbF{#Ka9z8Hf;+#broVeY0Kz3DMsFHW zll1+)vO=Txd|;`cp)48DX{J4Td?iv+*NeaD=O5a`q$d(yEEgf2Dm9BycfEju&o#|6 z3j>cGyC1x8do+KX+y{-QSME^y?Zw;2dl7Q1@u3~lxH98rbM4B}RyK0}O2|-)TN1kOEV4&GZkoWva)CNk$vu zI|sG`8S;@8jn*B<1)?|T88$yDXNJt*W8%6-Y|;uyU2+X^c)Ir%t~~A4;>|O%bL#*z zN#2iG|IzlmvvOqi!>-G}?*=1#hK=7F;;$1v9@R4CTUNE#o=C=}ui9z@RRJ{RXKjid z4A>Gi?+P2||$#(tzSjcC*JhCcHDiaNsXOJ$|RBXK9e_sx1c?%XUk7%LfqF4<9}e zA#9#lB!N#joRbVp+%C!3`F{wygj}$!HaaDOJv4{qC)n++Lkp##YqyD8zYO0izqDoa z0%RjsPuU2Vs= zncv;ms8n6kZ=ko;eeLnRt*@wit11R# zrsb{5GhTu3B){#6fyNnJ|5;hbC06DM;uBLE|K%&5TW)nvJaTla_;OYXY-f!U%@B>q z8P^Po$KuBJz=@5-VJtxSU0Bq%h6v0tyXw^W%7o2xS}O3QKYsVtE0Cw)v9pK{p5(m#ap!A!JP3NAzfM6k zWA9dVjs=(|Wzwg+L6?hgGq{yCVRO4HbVRfxg=nTXcf!(R(6K=V?28Zi#h@nmekrAa zak?!`hK6typ_KsUC^erObdOL2`B#}PF&W#&C(^F7j$NXMg8L}qrAj@h=Izh41&j@2 zEL+Gxr49~I$e=Sn!!cP7L%&jSoWemZmJ{^Uh!lifzo-vAbxMVHI1{uT8M zAx~D{$7NsQ5!)<`00zVNajg{K9I6LnB}Srq?e9-7MD-ZakE4nf#dce^L+Xp0XIlvA z@g9d)i*uJOLDyB|v54n$dHHI9iJ+Xb}3eRDUSQ6G9H~3EkX*;k`rD_9$;x8+?i$9L;LE zKiv(|jUFcy{}kBERqhaZ{&s8$+<`tEhu~qDAsZRI8F`^`B94E2c03OO0jEN{fWnxT zI86P#h2i!Un0osj>HS9ECY5pywG-6DPRsGaJ%e{Q2~T_QqI+H~jQ>=q@zHmzGkqZ~gN6g0P*J-tcMkFXH7)j35y(*}#GKPqq=F?f^803x{U*Y!qsuTP)p zD{X~FE0}Z^X8mrM-B-4!1u$fz9L-%Z4Ha6c&jR>&hNNZsKi;3=08;tS(X+>(5eu|l z3-BIwXv_uuC2r&>Kev$@*?F*3a)Np&8>#*z(DzXrIZKFTLI!XAJvy}^P1yNq79!t_ zE|PG31uZ4paha6Gg3+1=<|knh-;`KFaz|j-o00FNOF)8twcd}Z7dF5OKmbMowbNxB z&o<)yxehBL z!dn-3Y&_nr2wEy<01B-|v>mKX4p-FgdbQ9!og;4A;|BXVh>TYzhTxn)Nytq!5m(S{TpzL)Pcm~s}Q|JrD7k;oQL+- z^*ybJ)KohmR@~W8RsD=% z$a+^lPYW$l(fB^5?O5GB5|*zEv5f7MDP80bAGCh>u}rakM7?f0lBxK{F^fSnBS3BA z@;1a~I(l#OKrJ3J+ET%Z&b+<3Xj@~ZaPs*Z%iQMV-&?wuN5uE|lJoUNth03AUWr5i zzC?O0S^Wk1Gwl-=c5U%boiCK~z8GOyek+MF=5-syG~2l(EQ_P4IncTM5igDT-{0>J zm5}<}AqZIpnB88!;WSaP_S5%d+OUA8h!VU2I}xzcDv~( z2(U&tDx8(;uGy5amQnV{{d+*j~+`q z)Md)PGZ68vIbntB6-`BRk&PalM2LRY9wua%OH+Q@#AqtZLC-CFhw3*eQX8C zmY%En9K&8+{c;yz)y(fiyyqsqF8ulPF_Sqv3*TQsPf{|}^|BadEi#9TE-ppRy-_+o z`#wjSM>>m==9547O0^OXP=-slpOiZjM5YKj>r|=yHeIQmTyY_`?%C~2@@G`vGL+&V zqYi&0WaDqpe<_Gkl}xYK{C`&duaXElsTf}m)A8KX#7?d|F2iX}gY)gxFaAY+2BM?R z1f-90($BeG3ZSpT_4pI5xz?msxzJ<(N6VjBNxD^!-V|W;sXIC&O(Kubm9cGkMt)sr zZ-94|yMM6j15hs_wD@6?X^efb0UKG$u*6n|O&9*LnF-$Pf3*DMf=-Z@Z_XMvNy*V2D~7TLNHL|h>1VPoqoEw)Z1bW z>pBhm#kO}N`h4MhmRdq8tUOCCp{^L=)h`2scaAS|au7~c{GUU5Z6VH5pi0j@%ws@r zh3^I1rQLsEv7^b0)pzj_hY}O9sCyD@(SeFDEI*cxboA70#FG2yhIU@{A#HxEyFryI z%?@Oq6g&{g4E3K9e~Vs@1TAbo9ahD-ke7=!tgQ8XQTXKXr`;)dQoG~b*NMkEFY6E?#Ph-LVk~nzj;DEy)+l^n4yVxHeBKZY7^~D#+RQTZhVZiH$>oU=MaX}RK z;p}QWWb2EWTTNJ@yA-@`=+k@5BCg~QJ(TJPU~9jx%>@eXz0Uc#JeH~H-Aode=1ZCz zmK<157br^ltN#bY40X#ED;X}B@7NH-5eiLYuqU}UNPn62f&r#OVoSm{-5`<)rLHF> z)_7JnzavzW18W$QYkPR&cg#0jQ`Ptl5z_NlUH0!prXutWVpwP2cw~a7pRB)l)Q}SN zyHOc|IU2l--7Sw^NUK{(@jjjMaaBw;!_af0Z5+c;OKKKgoPRIS4bLs_if(DvdfmzO zUWp94R=h;d_}eq>p*MQ=QEZwMnmpnz2c<|9DZH>sVV!67QM15aSTvIp8I{3LHCrLZ z-{or&z1mK;B#5)NG*%MeZFrBCVJVjOLI^w|lqJ(TSEcs$IoEj1YU2fs9I%;x)KG9- zy%zdr^N7}PDK>yNM|_W+OF4o3d@Q4rg7LRISECoXSVQHI(zTm=USlW4V7!HPI)hS6 zcs)&@t5nY>I#!exTY4Y~J}y&4v$HYQusK;0X#A|t6?#DMOEhcP4@j7{ggi)xk52Y8 z(dYk7(+~c`PaYtFdR!s-Wenvf6^xL1u1E|Q&c89#vDMK*Q{;xWc|+xU*LjI1gdF;B z?C(d9^TTw2tYX>&hmjY^MLJwjtjN8|j{aSK({HJ<9DB$)tWI0DCQihI8Tu2MH4k(m z%?GGS20v?9&r$kW6&&i3ykzb`Lf1GoLYhfJs_jQ587CEeSCI5BPxKv^4^Zm1FMzls z?;M3triK;tvF9-=z#BaXawb`2=k9D??ye{`#Hh8={8xQ^yZSX!9q!_kN}PmKvAq_L zWZMFcak}hiYV|nX(wJj!*pp&(2@bP{6{2^Nyu>85&*#QLsqG#urD^S+iN`?XkHbJ= z%R4Tpn9=OlSl{Kpn#`G!46xD|6RAEQeaU@VPfj%6Fs8v(lL}_n3{BO_ITHuhajU;U zUytAH$i`T#Au+SOOrJkuP?MSdy-5Oe$urKHBA-N#SXOwxY6XZQ8e-Nix9apt8DRCK z&@~r!HTSrU#TyCz?C8{QwCjNP{-l*H3F`=igPIiKD0kwB5H!Sk5P`qKV=YAAgJ4G+ zEG-{jBc~eLDSV^X4n=}h+)e^=3>s2ZaN!BlULto>xSEV`UW0d*9@SUH^W1fVcW$zCg0Uos@;_yTqgo?dhheGxM@vCV zu=t~g-jxklcz$T{H}l?OBY(8Ykl}U6iizTVf_q*|#b?$qsSWZh5+z>Av3W)Z+&xL* zy*&okun0|b*P^MN<3Jx|ZrP*paCxg zl^d;IwExh8Fi*_sfs@9wtou|ueD64Zxf4$gndS4KW3Qf{ga@)-naT{kPM#6-986g# zm#yGXqZmtIGmisooGwJD^X*{~LQ9*g_w@M*8MA0|pCe%>O4cw1-vJ-^4IEz?76L@+ zbZL+g&nEQjO0m3-o5G7gxJZJ$Ci-^Z0KfiWy2>Md2vk{qI*^omITllruj#z;*o8EB zmo5%Xy+4WgRmt7P<`dmqRvtyjhzj5~M@F^t zgYuFUI~oClkV3z^kVDceEwg(O`eA`d0z9NMF9g7~HV&z4OElq= z)&`~Tn7Q1Rf?}Xfk#_W~vXs|y&gjF`8%X|oSJpi>ieV-i((0j0q~XcnD=d$(goiBB zUdhb!g8oB+*b9HqRWkNtDVFG45nYHX(ZHpSm`i}4gcPxt`NCZ!c1M>1@iOxE?7^SW zbu$kStM;~;JRN0-YEllRzkt?=x29GMFl)1pDY1s9;C>lrc=CD^*FHJ|k&%D<6|L0w zTix4-c@7IRIRY(!O*Fa`0ex=M_Y;x}ZnDIhyW+J!*etfRQfqHPER(8rl7vbFn0x;$ z(T$46LHrg!D5k@R?>=?9zv-D=*T-QVM1>`h|6YwiGr&$w-c<-|F0b8nS!OFas<}+K zXNOCZZPHkOdOh9pYF>f488$P&+%E|*;2_5ds*NESXQ1(&<^k_{_Y&saKJXPDy zNXvk)ZFFRxH?l(s?P(iEO9$^JVlOdGZo8navhNSFz3AJ!n;#^ZDkF~65)5MgF}_QA zEjHHuak3h;_*Hl-6xqb*f?2qDK zoVeaaGBEzmC|6Os=E~ z3q}rd{JF%8eUv>JV3H0uhw#k8;M>HEyxxWrAGpgvUTTF4X=kXMkYJJ~J81TIiFqpC zdGZYEc=gp_EYPQz3Y}l9@fx?wI(<#xd)qs6*g6`BJ!p2Ne zLl;20iT1aoENaPp-uX=S=g_6+W%*zh{?pyy6s0VjhlSLKJla3A@xhDELc(h7_4Qrk zcR+EM&0o{DSf49X-5x{{M*aar$JQ4a0n+m&y>{RzDF?hFzy68tW~hk_v@BQ%o56Qz z9FvRY0DVTM@nJTs*n8c-R!cZ<&|Eq8QwCiL3{4NCD6iZJR4c-wzT-s;fE_f{X%o=P zd5OB|RMKe#>AHWIrX@Tj^h$Y#kzZo=`q~q6hC6=__d$*Ja3E9%AI3&J%2_7Yi}52) z@~h2~1B%{+QzkNXTP*kTFk25~ql3;ZR?o$kXlB$t3GvcSc4)sT54^WvB@-w@J;8P; z3Tt|^est@UgAy<`)GAwnn)G+o4-0JlV{j(BGL;X= z(nCNC=$Dh$Mt?sG&UXgKF_jQUi}t8*RlMhIio3#bY2MkK|wYd2@qXThpVqQ=QV zYXaTZH82xT(h6-wFrV4o=#YG5tb)TBf(uos^1q7ENnpzGV5Y2-q`cU98k%m4q}|GWskO8I-n3M*h&Qs27ANjOsOt=gbiPz(qmZWpz1H zhC9-re(|?;ZDTQ2{DW~KbnGQR2=4dSmIr)IAAQpsch)BVrV^C~CZ=QTHEe>V zUh;a`uxf@*N-zwPFU38Sa@p~G#r6yq&)Jf7#>W*7M*XK{X$CPr*giCD*G7HO#)P1DN^_MFakEm!P8}y z-IfHvvTudhl594Mf*!f69(q$a;9JcPOWEvO{IYt#()0we&Z`#G_^JRS<2X>7-Tw~x z!2+y07W`@Z7nLb7R9>Wn-Md*L1*(uYzL~_y4R;-Td8w)0f z(((i$+rZ|LOuaLt%Q#6kpgN8?X{_u^2`@BcDanPx3t#+o5f);_2HFV82v4cM{vhxU zhzRu+AYd5v_$94rMJqpIM|T0ql8Up*9}us0TO1Ft{q83Z2+gv-(bE7b!n1^rg3iHi zIw0um?F)YN%kMuV#n41QLHeOeA?)fi{ENqycDNa}*6(`ufai6?BaFKnl|61ZFEIZv zK)=c-TV^H*d2>xDfl)ukCf93VDTAK0KT?gAnUGVu;jG-QC5SCCHG%@7udi1yH^zjw zb26MG6y6-@ftpK`>k=EAK(R%x_&93AednE^RE+3<5!UXE{lCUH%Y%MSFq1`UT-sbY zoBX*x3Vv8oGYOo9v)<~S_YpgG_NGv;SrZjSkY-FyOw=xSS+I#|+BTI%)SOKNkJ>_t(XL z6QWeL@t=&VnA_`|-K@*KWA3C!>75pCq$bxO*CgM4MeV&h^dV1zny9wL?W1yuev3y~ z#0iH{XAhsZEU}H0z(?^6GzI>r0Ob@XFs+mWasPW^wu1Bt_S71IXn$I_l)Q0-UAQypr1It#-XrI~({(F4+FI zNNmTZ-dHz*q44Ib$;*BHgv)X1^&#V03N`=x;{l@f4lYD1M{@1mfWM;M=1e&!i?YDs zDuyqf01lj#!zBaneQ$nceXvozoFqS$#|zFBZ?x&-_b02Hlp~G=Zt)W~UMbpQ9SKqi z@?R~kva4BM-ARy@AK&a95hDcWXxRhNY%hyW<4^%|t@3@#-oDw7%`-wP#y_$%B@q{m zF2eS*VDS!Nd!)oOH?jXl)+{ZL8mAI3=@zo(a01Yp+I$OwH`S4@i!W3E2cgMZpYINo z>LhG6^nZMj_x9MCu*HbR@rQ~r+zCqVyBJHNT2f^pRWh~1k^sU62NQVFp04QQa5uv0 zzBcTi`2J-3=q=|#^54twRmALehs;U0;sGDf>_Pqy)ja;{If~<-$debXmgKE@@>A|J z46s|Fs)qV8-A>ieF(E+W&o+@?{PH3)Tq` z0RbMRm)%e2<6-e%)ql8a@u0&Kd^oMSV`0@{hZ`#hGb0dfQK-NV>HI42dpScy z{IVlGW_r2ttv`-WorQ(;B^ayfHxJ1e;kd~vitXjcz-IK8)b96(fS0i_lkzSU z6wp5b9s`NL4>XBQ-%%p_OBLsc6UtqjAocla)ogGY>E$fOU2jOr2pH6|E1muWBQJ;d zfak;^D!{E!mg>CzV%;MaI#zvt>qvGOEN48`D=8YW>OIbKTfn37=B$q-(bKbu^%mdylJJm@Lz_M$$G{g+%G zvM_;N7rH=wr{CSmbj))VX9$Y%bw6gHYBOQQF;bj3SvJ19k@>yb=Dt7v*m=^{CxGf4 z(LGm9xj$L6w>}^YxrXi}#QZZR$FFyL5v`_S)x4%u9V`rmn9;JzGlWN-co4be*}Ml8 zQ1wt}O9>5hP8&IZc}>6782g1B_%0cDa-e~xVEJyXACT5IEgX>QjQBAKaD>_!%wiEC z1Bg0xF$56)bu@o%u;poyFVSko&>V{}+)dd5a`WymJRtR-x_5Sl(6hB{;xK;p62HC6 z+CXA`iD@lLMPX6V&${gjQR5^3Y(dm?Jb&HE9)ri>fiy_ez63E?4`{jeckjBdrm_aD ztpbi~f>y_Saj0hV4~ZJ)=cl`=4}>AEjf!jx=*;Qt9(N~#!V1$qio0lw>1zA}(0HYG z{nIH9&^Z=!1|{9r^k;yr!T|H5dM-;~kPn^7%TtPy0kz?TZu>kyQ(i_=5I~91W}%~f zZ<)NB1DCJ?%8mSHd2p;RGX?1X>HhTb3ur+fSfF;FRdU}NA!21$O6;PLjLF!=`CT7# zaDUl=ogP#)^X6dU0Pvg6t8D{d5t6?u`RL!Rzl4-dJCHOl>b2@dI;Nx-*Ga%kR@w8~ zhQw*#Yzk2L*i}P4=0Q|asu?0q#vNJ>5^BMbES-bTsr)|5S{%T9qzjL%vk*O{qZ5F9 z4MZ1w&eYR&vEEY^G9cgZ5tqAB2qwj0k|314?O`3Q)kFO3SyV)}&H*MRHD)9YOEAH` z`U`|5?VMd8@^8W0zQw&+m?}SZK4u&5Zoii2PDAl*k^0BQiiO`19+azid~mtM`hHLo z=!acDyZ^h}iOHAl98vX#{$z9#DC+86I9+l7u8R@@&fcu|?y)EVYK2fIP$bNk`1tR% z4RAc(y~LQI`G9onzka54aR-5>8`Lvf*f-v!KiS(G{u)oL{l#A{7X?DYCl9!ClVR~j zkAoSL=+XOAKX-@HVDa#aGZt;~-qt-XkkBqc`K-eU^ofdle|dg-8I_tuyrk;ZO1T9@ zzXu;y!QoCRU}R13uBr_3lP^Q@jShDF1Jr_>%0P6mJzEWm@9HK<-thRuQVpEL1?k5y zSQKyJAei}mOa>MjP+8~oY#u6%L^HTNO1uQxFk|=ws1gL;RJM>1ZC(;HUL2TZNE=7K zYtdl9A#f;$-3K36N7WD01xE|r|^p5dphCc8(iczZz#;*w)7eUlBBArQT?>FOOBnB5C0iC&e=6J>zVB$iKF+5M?8B$%J6J zC`p5!Co-1(^2pY-F-5NXeR-UnGOO5{*~AQSZdIY^{{Fj275d=-*$}*>#D=DR}{2=^neKrZEsD5qj^GbJ1!bJ>(g`FLjIJ+u(g%tjhlF@gOA8#`d+BB zP?5U02hrSr1IK`D|FDw$Z{dob>bhS%(zS9;@TQGf?AMai*vJ?s*}~0h|9#kc#865( zIMcr!_g!LxDm@5i*+xCeE7jU~*NgID95UztccWZkzZfy(iDq0_`|M0s6#MFp z{_sNEmML5y7}+zBwCz`|5W3S@lwAX>8h6%k80Xh*MSeFM(`m)|T2lo`{VfebUvg=D>3YK}RvWQi$lvq%>eX#4?iFns&D<5ouORB_pm=hNUb9s!e1o5@U0d~udF_R|c+a@| z1MfMXRYebJmCw3@s?ugxDQJ@%k-$UDQ+ei6YuRJ`@hP2Tl%H&2Rc-;6 z20gc>Em#&VPY1K@B#A32p13=9v~!U7ucs9{N5z!H;hugFbl$9Vn{@JwdF}C)mF8hRjd9qLJQU~EO*?@1JbDY^Y)%dY>L0FZp$6v9&#aTk^YP;NX zWKaveX+QEQ{O`?gHXL=1VcUs}UVr@~q{KZn#0i8Tdj&gxg8f&YCmCK zaFAu8?_;6#>|(gS5{oj~rvuoP%UW1gnzx|y@ro;gq>=AEq%(xBr(DF~@TUeJ6wB_W zY4<#+@~6-iYt@f_9Bf_`N9E=ji~QoxKKA@GvJ&*v#nLb=$Sz=uMDX7H=6rR5P++IW z?8=H@`obOItyl-Lq8K!po#IOD7m^`%a96LBk?@e4?MWpL09?J=yE%tu|gzto-_ZL6S_SP{0x3f}6S*#|;bE7JV#sHCPsl?6(@_ z*HwK|X}%l49|~J}RGnwNdM|s-bGPdCf!PzB2LrdtN-USy(*ruYy%HwA^%3{hlrbK8 zy1Czsu&P8v|6m`o(oVSTxwheFXv-|&L3iH(wB_IVgj8!d4=ml;o&25Tj5sSaG%V`K zR;0D;nph?%!#41+hsQv*QZ@~>!~9gI1pfExUw8El4)^-`kzZU{>PTG141Un`zUtC! z8qicDSZps9ycb;=%Vsvj+{Ypj9`3-{mxsV*u+;5O7V#y$Ddpb2=vUOw7!THOyFb|J z<>veRkD3|lZPjt51s)c|gPVF@UtiluT%X<_&qnvDm5ZcS#Gt-sVAVxl9!VwLd7*d0 z3nZ4wQ!*{6t)7*NRtkdxFe;5CE3w^C4p+HC^4b<=i?e+yzt8#btMVSInhPxfErP*S z>+Fs+y}FSa1Ap0i@9WejjjgA;lC9CL_hMZ)cec+cX_@oo+%oTrfyGx&RG}E;Clz(5 z=#kvNLYd%A_9gxTSUOKGDgU=WP0u{&C}{Z|tA(zq1Sgg^uNVo3?^A?_3`FMc@8%3O zw689XAPel>=^!hQdc&eEO@Ew6OB2oc1n?jnS9}K^+U+)X@t3xOTc#@?*PyHP#liGW zpPKW=LoykuV_n~Fkjy*_Hk~rrTj2Prmb~&qkH4AVcS+1YDiJ%$x>s_SMzJQ z@!hdk9Ow}VE^6Vuwe2~<>&3Z;h>hv3ik7{bA?Mrsz2YE4YbG1C*DN5Z1Z9R@RjHa4 z{>j^gcCF(0_=3Oq&Z}Bn3~nkFukwL;uMj~mT<}!4o*$>SFLMqd2zv_ z(5UQpsi-iOn3caB`1kjFXl(pQWS2YL=Nu*_jGcmmY!V+Vv;B}Zfv1Q5;$+}f5Zt~J zbX3q1IZktRR)3hIDDl2qYFuZ`v`Pz``hrZkHC-l*}$-;T7Fd<8O-%XST!$d z#ychGYd3pW1nRulHzfRL!nhsR-|~prgbwS6uE7Bcm-%lx0`{ZbA36C<{{HxjRXeC1 zv}?Q_vaJ~R2!~DmvAy>e)a)&JD;?|ZCma|xpO{wg?gQ^KYYM_kdL_5K6Eqzk`bjd- zmBGip%hMq9wS9N@WsnFRH`2std+siSDNQ`*8dboMY853dtG?X}%H>9?6N-TdpVeEn zl{xao3Zh9k_M*6uLnKEA7VWGNk9mHzm@{|x4>0F`b2Hr?S2~*5vKP1V*X^xPESFQ2 zYTx1lTF0s9cIk@rzA>5H!tT>+R@64FM)xdIR{~fmXmw~G4tNSY!c;iY$?R;3?2+riJ`!fGE|abHl}`R6iTjBK}Om zC7?4Yi`3r$OaYquo?NAbb9pK)n5v!0^IdXfpX%IKVz?0-3r zW1Nw})N}^&lpJg?v1Um)TSTdGbl>90d&(;-v3yfBB($K^E3GISZ>F-NYpZ!$^U4C_ z4SoT~eX@2qLL{N%oRsA|2%J2-{%%cqJdo7?H+vzwE)* z5ciGAKQYLl^-w3u_!NYUy{@5DLPu1XE}q6F0F^sfTr{Na;t4^n})O$S&MmNvh zKwD7gq?(F*Zupev_VCwCHTH~P1L{q@r8D;&kOZ399Q^YcDJJbX5N)-Ev9-X-#hQUW zRjQ)Vt<;KSMaMRn{x2~cMvCY_6VTd=q3yd@(&p&6fxA7#G#%>>ObbA1<4FO@PEpkr zL4Kqr;%RCn7O^falz}VsZ`^fh^{cESp4q(9hE8pcPN!Kk&kWX}jdpKHS?qe^qj;8qe2b>59=OVg~z!RP>PcbL6=>j*ue5+FPS>q)pAeZ?JKwlfonpFB)AQ+YII{8m;=+*L2p; zPA^S3-F|^85`B;x!Fg0K8P&;NCu{5LO26T(hDjaGKrQe)QZG`z5{v&j{weQDLPtd_ z#p-yt8u?)00!1f((P(OxFSC9t4qWP4fLU?nn^93X`5+;UlDP8grt;N9MdH>FOFuom zboxgj74pHfh;r54H!!w7v-3TxbHw;FT1w7|gF#~)i944@mrJX|Nd^NeW6#0wU^JLX zHqx64i5mstt|ODWIqNAIrH9I|R8fq(&4F9sIodJ85I3D;VaPR6Vtkj)uTKrI6Dk^Y z>^ya8HD|?p8z(_kq;I!Y|5Eo?T*+nl;(s^x#fQ~VXI+dAlQeaEuEBL=X#l1C7oZ5~ zD_zu$MXsCW9$ZIaQT0Yh?xNnA3pLSq6f4IWDRKnYfLL6ZwfvV9WnW^HzxMz_u7`{g z??!JbF>|{6K7hT>5iY9IttDF=i5tE!@>SJWzO8x^*o3~shzU5}h~K^kV{`06J*a5I zA|lz;he2l4frkl}AaMuzc27wbDb5CD-sYOH8@1QM!8!gy5qPwnGZ{>Eq1n;!i`y$mJ@9iV4Q z0^~s#=z)45%%%}d_5=AhlA=)HN;;qYz#o1ju3|8Zhn$xBI`)qNZu)fQNh}uwcgo5L zPQ$@T`R{YmzQwInLvo<`S(%3t5K$EzA-Humo09s>rS*L%@7}nq{g7JROficf;+SzQ&7J&lFUT1rGRR5-T88lOtC^)Du3+}~x`=(v&u1S@w z7)@bC1`r_+6Ma=}JT~5M1N_Kgt))9On~?p<5WB`_*vd^E#p1|Pns5Qt=%x)zlEfS= zf`j5I8C(dtbwP0}#Yj~3s$Nr5r z#*RAFf!;Gwy474^gK7kM`nuz@*{v16zLyd28;!x|Kco3hI-w(jVFs?PD&2O%^`^cgC>WF&mGBOYF8L`gpUVqqgwjUE2_?`VT}c^H6ug~J@I-de1H6iB9Xai%V;m$8j3`WS+5HXiZh*)Jou)>j+OG@T$l+N>)p zs}ZMv<4XigS^Hx zzJIJ^4&7T;1#aA)>2^~W0O5kl9nJ|Ly5kG}CjC*d%nD9++<8$Pg zu|`Xm#Ae#wtX_8Q`eUxT1;OK!kS3Z7;y)SdX2J7rVryAH|VNk7uzkv^KvM_;T^_4>fnIl@l8B zpUXIR?wEu8{^a@J7A|sv!s*>NPp7l_Th%zpfwy^3vI|61B0V=I=#G~wzUp&=L?;iZ zARS01&qlEc?r}}zk^f07+BoRp8$Ny9G(}u*Lw5|&l*VwJj|1IJ!E?k!lD$>YtVoM| z712ULRBJ4lB_$63$P`7&lU-tDoc|bH#Um2g1voW`lRT4o?116{JRm`sL-!D+fqU#h zJQ4hFPUd-!{Xd4%OlnHVm12*W-z&j`O{9 zdek=4XHUgr0R>D(D0>|@IH{jl=d7#V>aW59{wG`$UC|1(zC5kR1;6+SSB&mt2*b14h|g!-`JIrK{Gb4R$;>C*iv~hLr+@V z0o_Nx2daaEC~7d1@ISI|L>u>_h_=VgKadkzHSAJ;1f6E3XDa~=XyYHq%Jy87WPscF zBwD`4{^$J4S})EL9zgxW=GDqbxb)phAXtLW$&Z<6H?3sR8CDS8-Ftrd+%*WcApMZY zl8wU$W$m8=uAXqVRleHw=?{q2F-utJXJo&T&?V?VatAtCa2AzE%uwjPSxfXePp@RW8N@Y^11dO=CSyAAOBzOi9ycdC_uG^nK ze|R1?F5@NUM!CxO9P^9+ziI0aWSy=>oJ*^Gz!3zgt1JIn-BU4$G z+N7`K3G155Uq;&Oo3ZIlT|<`6r{>%vBWqCsA7lu-JL5h^)RE=e@x1Z6kowa+fi?wT(WB3(#$2_92Ki zM7|JcwmEhVl?TVkTix%J$mpM<-dHluHB}F8P8K>(y!G=WFu6mt_b#~4WhDBYGFFO} zqLvGU=v#OBRX&0|dV%($FSOK~+o!J0LJ1uVsUaYrFPA})dh4; z{r~&NdC)Ps@Av&4uh(@wFQ(I}nfTZF2joFrCt-K!&Bs@xo7&)Lt?bVK+LnGs|7jxE zpYz@rBzxU>0|Uw1c7E2n3$C=3!Gh~>)@Y0e&F#zVAT@^oYOhvSDq3cxKe*K1m;D@V z7Js!kgk;&eMVG`24w*pbk)u8C9}mt_k~ct}YB^)Dc*WsyJ(QP(yL6sPQa&&3Vz;kq zT9cfPaax;1#2uP#@M%cDVQ$vGRk&&8Ff_`129qM%`sZP*Rn#2f3+Ods0xBe0W_u3S z2ab~6%u?Eleu{p(*uE=V=vFDbV|EJ3!n8XxSzuNFfrG0p>;>0BS${6Z51OCfpmtCj zK?}{a(K5x8yHxC*cop(18|HTMX5fQ`pzyXLmz&>{~5VulmC47U=S3oxX)r$WDyzCCB|C{!#n;ZUJXoDxYL$Z{Yj2_~?#M^9# zn{5(>!AE?Ha1>7;sL<`Zl3nuYeDA6!=P%SQ)w#c^ZAC=F*JrHz&ULH5==*(vJ&(FM zG+k#y!_OXp%m63aD{Rk~z3zf;N3VQHd=nsVz`@Y~ z3ZI!aJVzT{d+8@is9%SUj5ZmaV1%>?EgNZB9Opan#eXxnsKGDLjmo;}fK=@@)VjOS z7BSdJOOKL_QJy}?_b<)5L%b9#j$A0gym^nWoSFxyy==8?IPm4BVR6*s{^mjyw?cV3 z1hB&?^+hfYWOpJtZY;EYRo9aHL{$f`wBTVxz=Koa4A65$4{?w7-YS%~WIFJ|v>uM+ zR@_Z_W6D3wrZq6t2W3*Ye6lW{Aj_o5T;T!!IraFY?4WuSUPmti$4|s~lAF`4D!qFe zExtb4cC=kYj}tjRf$8}W2;zbfxQaB2J12RP@$<#oLxYd~-+xqZzDH1gwB1~uw+Xv2 z_t%J%l8RVVKbIXshQ$;L3xHp5Y4898L*$oK+0_)U=Wa7OLFuVBoBh&D$Day`|wZSWF zF8EOn8WJp@FQn-*k;AEc@TT6>ck;lzcm?V^Nf6Yw?<93#;R0c+8Mz%!ky#ZMbB{t@5+@`KVo{UPGACVT zThJ9acdkxjuY%8hfzYiU4Yl!UH^HWIE{KEho0aA|d%a_fF7q*b8TH=r1I3Z{PFT^T zw5IREeh9>H7np;xwXrK=uNTl^6<(Ao>N~aEop|1w;Ff3OBsY05V+S?7`!BL>3uCl5 z8PT>iph}#CghW@UNH3g=|2Ruxh`ZX{dli|dx$OtrNG61p(>KeKJl#k2iUdCfm6%17 z)(U6@AO2D=puS_X@{^amn&-kJT>kzHIUsuKEO^vW;hCF4CFvtK=6m?#@~R%g2L}h& zg@4v7{9Kj`gffK;k??dE9)j_Lt0~QO$*bUTz%}UEEX});1Uht)WsD4LB6-xlSE{Vk z7ob`TEK`oAX16l#Tp&p{N2M1+G^}R(#dnk8t65@w2c1AWm|gT6XWWW$UYEh zort*b4pnncPqHO+zOmHzG&p~;91VVWSM=ACf%?is;-zg`hWn{2!{FVLxFg#W9ZCSr z4z^NhhHx`Gz&RIiugXbxjALM zpws!IHyPBsNwUwhhT@&{7zK>)K6j%WO})9#^fGXez^@hTwJYCx*qOI-cR{=&qAmQn zSo{anPjeshhvB9*ReS@>`{a>6Tfzl*fgWUIxx7&+)My&(MLGIXRGI)b*b6g+Rg z&Xb*w@13rU3%k9LBZ1=G=9iD*w`q34^tK}XIe#c`Fzl5x6d}-4W?f-kLRH7I3yGAm zkxX(7Ypvy2-(jd;S<^rhLGi*>H1Re=%lOg#8<6JEmLVMEuw8vgRTsYqr`-eXp!51l zRTyROh^H~Ht*9wHU3VATF1mv~<7B+=+`4lEB2kpq7O^j(ACOn)8v z1HC6xkCD9wdls&LBh%~LlP6f^wqNKPhlz^y@w*Z%{0B9h%-#O_I>Mk&E(~h;uZziN zt>uZzIp3&mT0Vcc)K_}zmr6aM*fk5$wBTp1P4e4kQE~Zh41FJ48d{gE)^lt^@v^NL zCth*xVU~hj{OTOD;@FZPs6~81R+rr-+O@uGQo-?oE$bE&u8{@yF`qc@>+$Zhvc^6(BymMQ;;Omn;Z3$iU{FDW>q-NRUuC(`cE z?Y>E?9Gd@P_U!J8(2H*^mM;e}mQNTh*|HFyKZj_V+-<*kKUINe{_jiKZ4;v-sSaZK z)&77Ix$RqWN4!RX12;J`SMZq(lI2sxq>gGr@11u?ia{oBNZXEt+$Fhf5_1%BF@H!@ ziAJxk>Ejp1cKBjO`*+2=WUKlWL3z6V0Fao`en2MU z?zCOwK~~pTUjFx_f10h|>}>MUW3?)35O=3qsOhfs(W&}BnjGT85h}G5NKY+te4*)H zKTn`}Z7Tr`Pev@D>rIVsjPnkL=uN*dAS(5s4i$;U3u!Pl6UCr!8NU_oCLcAWxrHr_ zp9e|R-*Sk2cgp#Y;@HFj&CIk`k$0xgl!<1>d9de}F4n{f9#Ywidx+4UpKAx_4!aZ# zdty+Q`Tik6H;^GH6;fK-cgja4M-yn%+*%BShm<}fiEv1OZ|A8bu;kJ=y=3=7JgZn) z^UnK+uy^BQMT#`1=pyL~{nn+bdIG}1UpdGP8D|=HpRbx|ZGX9c&#?RnoLBe;935leNA*(NZjK<26vs z+9KTO5jGS$Xizig86GhpCDruwiM=9vCU^Jg*-?)FWlrXYX#Sjfz`+-mQ(=+8vYaFF zTC;y+`Jk^;7_pO(86tsJ?7Ip5nl$u-G&jnxy!?=^RQ{w(7C-re*Og?ft0dn?B=ba7UgRCTxZJO$R!@l0O)}RO?SEPxma7UgG0#GT?0HCY z-~tZ{w~zj$STDmOJRd|Lr3d)5OLyTzw+)}3#$Mf~4JA>X)f7gCB^R)97pG{VGN$Kl zV5R!^G&EK}Vtxpm!ZeV6FXig%Li~Gz9ZA56p!-)@)G0udAM%aBP8BW4JII`J@VCO+ zI$@D!oR7;nYRG6Ljcp(%^GS`^o?C+EU6YkN37Uoxqa{i~j5BX?z?g~2ozPjNl2Q}a zAi7~Haeu2TjO@ZpbsN!JU;4)*OyIhSx7r3`d9zzkOsQ?{7w;gApqp@p2C-hum!SI) zdZ3COg%{>kj-oF)t(ZS?9;-_`T&-0K)krjVLSlwl2YL%D60RAReZ^$KZ^o<3SGz9A ze3pjkCttmEcIBYIqR?Z?$=peGSuQ=57UuHhw#L?==2HlXqgY52+YRw35sC9@q=-T&=2a>apZFnc~H^<5?mYO5i? zbyIx}PwlOMrj8Fa1iSfWG%#a!2d;&JAMD zrMdl3HYBtU(mXHAf{ar4=kQCPQhAWdFV1DAnLEj(#&wn|?3ihQy9ILFv6Bmr*1;W$ z`3TY0H0;ihIFzl*(-i4)%SxLRy%S{IbH^&&InmjG(5^RSikvt_`yd75G)b4@9Qgf} zf@C>`P}u|1$O7ZRY-!JOi?M|E6Qqc_Q*hW2LpZT9)a3bX|Sq$Ue$WwmtB0F4reok3SJS8n4F3wWu1jgL{()`k`4sETZ%N2zDz-9 zc?ap`nDF3DULR?xBa!&~hC&X{Y{gihIg>J)Lzj^pByT3dk&&Epz_ACdv{(pW%>eAS zURQTXap%2OKt~D z5;j!K%{W-SlAEh|^n!a3bES?4DvX|XMQ{;Y=*W8oiF1(z$Yc6S^n1wljrmIDl}?jf z^K{5M!CUxF;(LgQo+1sFrn**E_5^Dwr->uuQwvmNa~vBtB{*~H;KJ!wG)F3@O-2eIwx0Igp`?+W_~!_h4Oghlc)st z#*z6c`tBM1WaGS3=g8Nj;=z+-TZ!#6rfa8Yok8wn83{P~C;41fYi0;aLF+#w(#sKf z!%kw7yP+~vKs^|4o{(lpHNy%lEQi8G(KExsRtKm&^-A`f%a7eH5a4>kzH8vTG9TG5-fooWcbb%mr})(~Rc10E34Yst6Wgd_VR z#x*z`Xy{iCb_CtS9nMj9!R~XrEc0OVTa>jZLWAL87TS9{@aJm8^c8D-fWWBZAEo(I zX}5#UsB;Y%>CM9uMnF5rB;VtN(M3A?j9hQICk~I>>N7&(;qffAuCl0JGwA$jvU=5J zGQ$>Z{ui;bEnS!@)e%V?Fql@-?#GTSUwe6mH=;Fkl`&S6E=3cgvLa?MFXKTT*UTDf z^`>-xH;}4eygx!`e>sCi5&l5n{6Oy;O2S8F%5|>Wm!x2}SR0o7kdz-G7N@Y*_n+wd zu`1=*p3`2=ifP6dldf`#rsQ4pR9B) zx{R7!)sLzOPo)j1fXYkMJ#Yhiic>AuHE^uPDu2btLn%x8mX*_~k8S>n{5;5K?&Kr6 z$I-%*F=d8ESbFWFF7;k~bo7W_G0TW0Jy@=QfRqzkGs}x0tR^HXzCvC3;7e48+^-ao zI8Dk?_G*&bv%5?Xk$ONw8hUwC!^`F|n7zvwcelrXwQk!}&z`t^S2lP7UDC@VoD*am z#4euJks%z*K$4^=uY3TQs>^I$Hef~VV zX_|M`9Zqlyt&65ejuM5Xvp+;fIFuo@;ye0;nv`k*DUZFBIY;DGzh`9^#=Pn!Hj`Bp?n)&wPQ7 zqgnlTGAqA{V`~hwE}f!3yLNCCQ>SL1mvx@(B|I=_hs*JD{5Ii3+O-jjaz9xdT4%Ky z)p0P%JtRCnJBYEJebgQR$%2nrQSqic|GmjOM(TcEvOXBYwp7h-=tNW1H0aB4%ls7P zwG?XIMNG1^=;ed;Huh?N~@qRX=sN*3=b0I1@Cn#Jv>q@j*Iv)O6OSy!A-*?HtatUB;T&!XC6K} zXZ8G>hrp45hulXj|B20R>w-~b@I?}wu`kWo%IW=~sTjvz#djr#9*7!03*W9x$I%B zeG?U}RclcngMMA(Uqxq0a^w4BBZ^ECwvAv_%5r`q!jUB(u<^ziarnsS+x4Wb`;qoZ zDih>Vzg|&UlnTYeT%1xeX}{lfsP%-eM$pxUZ)t^AdFO#z!>HW^85Z&NuB58$wmmZ@ zyNMn7-9x$c`gd|-X`4^dACM&HOgek^-v^*(?zp!`k+4gox?twc874?_dz!QXy1Mlb zWMe5$b7vDw8Md!?-WvBh|kjGS5{2{IbN?+)HdC}Qh1bwjaJ>g>!sAa_B}s$ zoa#v~^p1?oPFtBl41E@SHu4GbaEErR--6T&gx0*fH#ce%`0*lGJZW6s>?F$t66%`b zy&{|S?o;fg8MmyUvv-o{3|k>Va*B9G#6ZrS-qa9Wg#`_CQBM@fqG0&0Orud!G5ge~ zF|YycEV6_8EVmIkC7Zm$!29UvL@re|r+Nh+LUSWlbDDgPN~9egP@lj}9?w*Dt0bQ; zSjc{nz%mbXO)iJs+(m_nY;fw$5WT5fJ}s_+Pydubw1gL%*Ut^k3pE`;OmUx|=BALs zg%>ktK}MJA+bMjK+h$g{H;_qom3nC^k_LkUyW-6t=S58NZk67=q4<9XNfx>+#Gu#C zuJ#Jo4r=a6)7gIP5-qP!KDhjw^)??k#I1BTRTKVGLISEUH#kr~-l2NEg5Q`W?L*_< zdHTz|F66jWdIH7Z_4167-1G`(-H>(FoLQ|6Q0cbGBJ&mK@L|Qrn2auz+xE^ak#iO& zYD^K!3q!N-&DA`5Y6TsYe>QE!bb<8kMDH3p8_1|5-wmhqnL3#t!yT|Omyh~un(iOc zaP8YcbKX0ttkyT+c*4->wE-GXa*$t?a`RpCMI~BYqRYsYjNLU*io#;Y>yw~QSVJyaXx{_;kCUygitjJTlC%b)@Q0IS-t)In0wRL5C3i`>pth47Y9=YzH`Wqsv zM#sFBS!1D7BFB8Tq%9s=u-j#~olLugYCoNPeE~r}Ggcgt+%Fc@iDUWG!db1*E;O-O zyz4X56y>{9!v#w6!^mD$x&W7^kE_m%s$Wz2$>?2(cM(TGa>AwKG-GcnLx$fI^uki; zw;YtP3vl5Is?QjExmnr8E%`n7yWUD(=Ng**gI`{3>Sf^rDofNd1n2ueWrzb z4`6WwNnB>m;w6nvrc`p5LAt3p9RKD6)YuryYK5@mA%X<=EA_n_7h;Z!uAKr27N#dYh-G=+#ec>OX#e$4IK$xOH}fu8SbwuKEjlS zWuHMztK2rFw&`S+2>ky(DbL74Ku_8F<71%gJmf@&3QGQB=t*W(W`!>|Tfzo4CO~AC zF7q-hnqEu0l_fPBoiRrQFv-VTgiPBjTi#bG$eIrup!q#%x$Re|e>3izWG=h{_yd@2 zsojTubSkaez}D<~6$E865^c3DW(x;#2M6z&zWy@bMT5TwECs;d!B5>9SB`!>R@YkC zB7Zjc+4`mnJ|EiWRt)tecOE*?l2*y18=p>5mwgKk`g-CXds>_C#E2MR>e!C%ii4|4 zG^SBYFcbmtQgHiXxq_7J<^yXLa4D({JX(=)g4$f@I+9o`B1pL(3pRZGPQltT{8SFR zXOUOsllCU$tDzm#f7w02L00V?lfVxJJ7z!e+dJxQaMMO2^f5V%>vkT4B$O9WnIaRv z%OZIC(j%3OZV#0eS%eKP+hk((dvYGytblr4(5zteG(&SOBL}tRKPFm)f*+mII!ruP z=_Jd99}4Y)zx#`)Bp{F5&ieC}rEO;J$E}-lEL7pH6wu+z^8l%RB^4DhgItyC)3ld+LmfinUfvsPBLcA&s15w8QC? z%Rqj)YpU6aV#F>QWdoIULsnFK3>V;a0&4NM?#)Fxx5$lv#U*1XIe1Gv_YKeu8{EVy zyPXC9`~fhT1RFAbZbEEaTPNBq#A=w$W(Ji1EK6L!dHzHE2r3kI3si|1OR+wQs zojP-Ih!#YbmoTzQ5Kl&LH$lCu&qVcrMQ~g^y0Ebwbi4FXaI}@hty+l{g7*Aczc@xN zWd$J(4Q~c;o>KBzOn&eF9B&V(5rjZ|UG_ebcivMMML6|rRvR_IrJ|d>EP(q8&y5;d zKVHv1kepXqfbjoRlAMCsCk6&gk@3sKsE4o%%q6}{UO)|L(=H?A7YeJo8uMQj3vrVOe4#}>x7iS{-LS`WmL5xf_4L!=>A;;vYD~6g*&BYu&b&_ciGJ0Q z$Tj_r0NOB(q2_!HMLqXcU#@e5dFiwttwF;B#FXr7hl7w_LAWq~Vwd0H#5uPd=26^+0RPHNgZZucLm4ruTTmyC5+WRT@A%um-;D> z7`8|trkemCl%Q@sVzSrqeEIrHJ^vsJUEtI+4rfUYe{{~l*CY14Qq=VbFOe2C3xf9G z?Az}CP|ewStCg^19gGIW$+Q7x^lRMcaIHD7mF{7FaMQh{c>fVcj-2y#!3w$v#~DtM zjk*4siTcibS+IWjQLs`y4*4$aL>be^Vw|iI=5OiM7kcN^ydL+v-zO%{AIw|jeee-C zueEs~AewP<8(5Z%^hXU~5AFM^c!T7tbOo(fT9ScE)*uAaT^0Tz;K~K4hVnG|^R4M- z0m>(JkA}TVJK;6Bm^BHoUMy)o$+9o>^$fnoeLyv<Zp!WFV>O;ig36^Yk<510< z6Q9`;*in#+BmZeS2T2MAwVWY>p zV4De1c*lb$O|qkaQ%N^83y@eym@D)wo*S zab12n2_84RIQR#0&`4Z^vh!{wP~1wkoH!{8QoJ8*MjTJP*nGSjuri$hAU*ezw4;8k zvR3ZJTT7;CEVX+D=yEv%{23R#OT{&1#*+k|kB2qV%7;Q%&6-z=TK*8}(GE{_03uXn z0$Cyy!|RH0Q=W6nVVL>U{F%Xe7cmQ_D+6G-Kj=|@a#Dz^$eIGmwf|Z~0&K(ZR5x5f z2L3Gbq3buBpjN6wwr4+G5RyUiJ`Xq9Fx+!57p(tOog@msw);)RgLpb++PTwwu(YB_2|T*v-s_9W!sUHy>jWRSe;=g^x`TPwHtkV7YQ0(A*kD&Jb z^siEfvpNRT&ID}-;J$l)+KEHRu}(fdKJ7#bgUK-VUp{*<&AuIu`>jha7q{>yi>9&n z?CGm&e8o$hPwPUdK?4h_#sjCbq_g3G!V3{0a~}3Qj2!}UvDx$@9YCyrAtyR#>#*7 zg{a!?sEKFmDp(;EscSA|^d%2ye%QA%O4<1q#&%FNc>4vjF3?#gwah#h=^f)d$f42i zIWw5ji$@krSeBVS_K#rswlZx0BgxOGiq@;$>CRs##u=Vi6iyUT{+gDvh@c&jydxj= zW1~`@DR{p5@*dKyv6E}!wW`|}WDU*;TnSh&ca49IL&j&SLnJ)*JhvXKnEF}FeBShX z6^g>zZ8f>8l(<`^?{z-=?IyTPMbAmU0WRbw?xU*LcYi`Ymj~mh8)+J=w_p5u|5Qt* z(i0L*-{YQk!)u&!1^5Xy(uOfr2AAL+PwN5jo(IgwH07jgD#zf`->|yGuJxi_7SN$E zOT(zA7OnLQ(&uemh`Y?5H|1%xwR8**-Wzd5vQSahe`3!&{jE~>=JDan4oha#>tL}h zVl91ytCY8ES=3!-Zh9uW|7rpK?%K>H60J`PRA`PR61o*WuV~ZMP>qMt_Xr+;z}8@v zGmAhDBd040`T#CCBC*2&*kiY$p{(=hI(>Nb68CUK{IAJtRP}XRnqxSOrKemxaO%at zuHuakpL+zeV|F+^A-?6n_w;%7y>YF+9|ZcoT&9_51XM=WlP1d_^l8Ove0jmp1-+~x z_}`^J&A5P>SI$oNYy%FnPAW&CoBoNN?T$Pi(vvbZ>E1XT2}0!UeGgt|(TCQ*|6IvJ zn!qdDXuMDEW&LEX$A|Hm0lk7EeW5QrMzrQuo88pB?U$HIpG9mgHukDDSMQ_lh?mrv z&Nge_xI27v1}J(dL|(p>pkX*6mcyL;eZ+Aq<4t)q0H&yE#Ni#k(8INCW+C8};q`v# zsbg$PS!SjN8F9S~(wOXGmXINal1OwXTC13gFUqf%qlsPwcwj9*7s-n7-8$oQ;}PO; zRjQ@Hng`*q^jseZc{jB?s+BWkJRZnW84T&szWj9(s)6z%XQe0lpy;lg@MnGcpc_8j zs$U=MMJ8d9BaZyP2uqRYf$tne%r=rjiWDgj#ZmyO-PRS-+qiyVuHrEYCcTrh?XJy?)tu(fZ19ntpQf|YFb2;;oXZ%pD9`KC9b^+Z+pCI8Qhf`|vj$~x`^ z!B->axZDqyxGF&Rt2g(a${mCR+PcALM*PM?vki`Je+(!5E!7UwJ_zC-( zP#u--CS=*kd+m?jS}z;prE_c>xN*3U={F26e#B(SWB?{yN&=+bd`1+m-Q2=Q6WTYi zk~wm4F|sqjaQ?Rq!dXhC11sVJ3a6|2k9i*zv{OqhR&;g5asG+UpW@(*)eXmG7+sGt zMWHWKUnmMS2C5WN*-W0o3eCO~SE`K47R%;g+t=i&yWttMU>i{9=jN_*3^*~7JGU_~ zP&u0xB3W)ZoG;|ZD`<5;#O!9#B%thhCw$%G@PaV=_=rO&BB~3JNWz7@eB24FjJ)(+ zfc6361}SmNr{%nitgsXR&z-ei_%FxR9L<6S!D>=ftPN)7WDW(EyOk|#;DZOyKbjjgg{I%5*aU%7wZsmWZ0ExpW&*c2?fq40GA^LEzh$gsYfN2K@i|L&VZzS5%il@X`fB6C z9eS+UxikhPg1DI}Qr0uBbXlirWEacc-OkOj`8P5rv61>X7uH6KS)>Vi)R!x-X|R8* z($DOV4IJ0_>RxSyFP zMdk7z9+QNpBU+TlMXo5^IQRID?I^U*iGE~zoR3{)>vy@rcy~|U^l~02VfbRh3)pfm z{P80xYl1e!r052q6E|sp1}Mfnp_8;>XD&?(7T%wNC`T?^U;4iPtxx@}&tcI>j~C0^ zO^)Ky1vJUx6#;DitDS3faX6#DPLV+4!w<$Op+N{#9KHg=qy<}vE5AiOmod9Sw;qTK z*RA@lRR+bnD8m2EHYWt7C29EP*NM@`Sd*T9ACdbC?u{cFWd3ZGVn1K9f^3z=KzM70bzD216L{LC(Zcj*}tftS^ess8Beo`qZ&@CSEUV3ft>=6syw z9m=pLr3N9`s#~y#KIgx>b;pDXe=ytqL-;RXN$Nfl3>6=0W!5lWg!SdB?%z?6vIQJ2 z={Ho(1KF)T`50ieIVKqqh02mY_SAiV8<(AK4h)6o1cIx-H(HLS@aNJ|akH+F+RC7|cocx#bY6F!gY{v_sRd?dt z3|r&oGyRYzuZDov=??jr$-e@(F|L%F2EO(BLHsl0Sv+1e#^&KE`W8QQ_}~~`S}!*N zn)E+#%{1db#%0$wFMfZ#1$xtsqDEU&>x=1#TrP)cl0ZHOBvRA%k5;1GTnqHYKbAQ3 zINHBu`qwALdJ;3t$sFH5kan(q(rb6n}4@ux)hPQ++98OZ{)ybDXZ3 z!YqbF5m~k~x)qZ7jR1pxydahG;P6L+X+^JIrUu%0UqaJ1BhAb3w9j<`=2g6KioUT{ z-cmnE7v`6NH!U&@y`O%6)Dh+|cl?~Z9-S9|@CxuWW1a)UajQhtrknBjLHS%cWz+uL zh;c`F`bY2XGhQz?tS49@2lRbyNatZF+5yH1FAdxn%*Fnrw*dBv`UWSH{JWm?s(kL8 zjJqgiRhl4r%x-FmlZg26#9+)3t9G;3F)vvJj<2ct62h09c<>g{d@HEXb*kg%{?tq} zqjF*kvy)vW4HfEp^8O<1!gx?C`C-+$W#%rRE)NC@;JG5faRchIdAYji3UA#%B~+86 z++xsvfcUfZMj=KhUW6kEVQZ$Lzb_~>pI>>hlk4E64CCLVK*_0c%CXSVu&GS$*Byd3 znWijQh#q@zN^-laY)}FXl?t1kS-jTsijr05&G(Jm@IKRIISZT>n;&iQCt0uTFf)e3 zBj-cN{y)ck2wBExnU2L+!50eP%k>%Df}h!=!(f9mh3|KhAI`T*bQi6a+g!OE&L9>E zm~g@h@uoVsb`HUvm-VcOot#|72VU$v2=ho>F>dHJNpd3MJd{DCd*yBY#Cnz9BCPx@ zq?+B`iep85U!J(`DhlM^2`Il^gU2iR$E2JzZRay)IL!dr3T}++^~E)ek<{SR1xM*A z!1moc&J+zq>7PG)^WsnmXj_2T6a`AWb{|=|wJqvlxuA;?dc5z#CX`lg${t;1AQDgW3qM z(hIMG>a|+Y$`m`9RlXgwV`>S+HM)7HMYv@85&JUJP^(i(+Y1+&KhDw=3ih0#z6pR^ zjlbEn0#Q2jECx*75p}P^M=bYIPR~p)Eip3>LDdNJW#$VD=>9Ee#@yrls3}e{QwWk@ zUpE219l;erc8bz6n#FsJygJp;qXVvPwnPoJ{%C`cjq?ev`oM=iW5Xq8C+qi6it)Nb z(PfI+VKWu1b_95wqET@n;bfQ{qYS(64nqCr|Gd{xe2w}`cz5R#3ENt} zKpS~96(s>tw!&hMrY^VQ@pS83k=MYb&Qd5Q>kvpTfPf^DEw0m4Dui^cr9d5jxdl z%10NmPC;FtZ5kLUES*B%RsUOg^x{|Gt250R@hP=GhFXOr8qc;pT>N8OcH0OYZ|(2) z_^{(3p)a>#X@CRayUwzM0$hzngIbmLcF@SjaT+|vzqn=Ow`%|MU-|uHMO^Qij>$?H zYUNX+R9eUqhTl^<-%|UpdfZevzcL(W>3_b*ktmm^{=LU<-5z3ry^T?Aaed|>4EP^F zhAG?KZaEP|UXK0@bf@3|2BCg&45T+R#MRaG_S_?KEr5}nhx;|keYi1j-8v7qmXavP zvATPil8o0XW*arxzpoEGEv|m|X}QIPm-}+V9IAa>eA*T2lPtC&ScpSLe%PSYM?B8wQ8`ZhW(HdnOuhrFwwFp{C~NO?XN3U6 zmx!ehNw$BF=LrngMk;AB{kD zYH#SFjH>+!kYo+HUDJ5L>UTM=MRyElPK8!&p@pegB3%|5OSSF%`{YEwPCUyXterrs zytX6xIL+oPQ%z(oWpcjp3eioY``4IkC4fpQYFVpotatLh^twI45Zl#KID?Zr2I&YSAEnr-x z(h&aP7KBhMx($gWqx(=T{6$d(E5rJ*M~{2plRMM3o%ql4fUq-IOm}KqrW)eE8nYRn znLB-IFAW3TRC;V?fGEWqwI}#1Dgq>|pjBVfg|RW3d9ut;11xtTc$o%NPTLZ`3cPwM zw2xB1FI`=4Dt`R@tCueZ$xlrtKGL~#ToK=t^dq?8U#~$;667X0uERU_Zvl{GM7D?LC8THC z82gQP>>!tJb3HZeKqY+?s0s{1K*rMwCyx{q<$11OHu4BwRE>c+vc@>oWwft<_84$y zFt>Ms=S1&k>yN6~4pTQo@d@0^&ut8h6GyE71)3Uy>KL;P!NX2FC8QFHBpciqaV; zzTwo3eML9cL-R%Rl}QDX^SnoN5*@+u#3}gBrG9bo#iR1=mlfHlET!N(P&(0J^HifL z!^c0bUVCIHOWD*Q=%7NC)`0vo_NA89iEe+Q^k2X;rC~_-(qF*lYUHHfJgr#&c?_nN zg${JPd-j41+il)`=Nir#HVXCy=WD)}3m6R=SBo(6>ffa+TWmMup40FuACAVU9zj@O zSs6R*PIB;oBMxDR;l7WH$2`F$QnJ}}{o3Hqo)E8hv4D_pu0pU$+bAswlOU~k391yJ znt2<-Q9;-c=UuewMQYG!0`;vvV)QFkEC_+UYa<^Ipm^`98g3WlBr-d$Q-kaQ~$ER~_s@6!6&+Q8 zS8&MOzh!oojz ztpV<_IFxT<|Jypw+#1~6N?SP_+y!K_?fZlVe!MQNUrZwji8p=dSmPcoN)8RF@62tmZr`-p)8NbJ4Hg4969>6EpCFF7zcJ_2}>!XdJ;)EkH_V`rIF-Um1UQ zIRA#ponmg6T^ea^-MMJ;4GNAu6{qU5zTP9~ej41Kfqt_S<8M!@(*~U5s1L$zy(=`4rkK@<#rY;o`k67#_!ptj z>SV|42Sd5=v|m7h zCv;3?rDI_|_XYHlEv@TE)T%}VWz&SQA!fA^oR}L!orb_b8WwsGsd;SEfnpX-&ldWH zx^V^4nBX02rfshAA(;-KdC5Qzq|l44Eq=#%oy7V3hvV^mNo-7(F*`cc)TD6^Iwasl)ySZ)cMl z4IWn6jat1?M=VjNEEOQDvJhs0neDp)fKWm3j_2GXZNhMrwd{j`>(SSXPe?@?bymwb zVm&$n@5t~;!{`0yo^Z^mi%*usZN#%)B_`k>1K-Y-m@QyOwufpOU@l@}5l75@q)}&V z*n8j9K#yPM&GY<@xj*eL1Rdg<49O0!br7e2tmiI%$0x=wIG&w#th~G-oA(abr9#gL zAFkK@_@?WLIoa&!n7l4@2)A8!^+u4piff_&M3`Ow$h!6*X7}{Vav08Mo~Y(qH?JPo z+a9D^^=78XUk#{O9{=`919g1?^&P*Wf9MW>XZ#GoT@h9;kePFR?pQz0Kn*iAWDde`PEO2PMs|M|BL*shc8ha&#X#LSCJ3p=u2@cAjtx;9Kr%78E&uo+{1bw)6`m1BJh*!i+dA+BDdPSM~lcPGs@ ze-g&~x;!#=+5+B8n5*u~{RRH*sm-PWO?kcHIOLv&Ci9xm-p1~tP~g<6YK>^)xke+k z3bj?;!&nPto`PT;DW;dL0sCAmEwC}**1E$CKD1E%$vkwZv0>6yU#_2b54BGm_+x(lNuA%A-%@rADH)+*Ss3{u&@%pNMKjw#ld&X$wwZf&pQAjj^aO3!wy02 zEoT2Bmn|(Wt58XBY6X-SpJRt8?Tk?HghW&uLI}xI!Gx9+QM`Sl!gIPRd zVDhc?;jndWXfYoCHkgU~EsvLUX$-jV*XhdNXSc~u;_!ngn2na$v%^Qlu1=A@xYqjQ zaH!NuwwwX$(F*wAf-H?jN)kM##lbr&Y5-!%>qq6u#fH6kN5x`|ld;fR=?(Knyu34Y z1GYD0RBl#~cRUwTJr)h~toK4;NI{5VQjWu@R^KyE47cL^^ATTxL&^E#M}D#x=N7}U z(66c2C5;+4)7@uKmjZ1S;dMzT&74%yf*>Id82ImMU3rHQgzhA|^!_(ma=_WCGvKW+ zSZFJFcqDNZpDCXeWF)4DJe;ZXC`>yccTyeC2u_H+jlnyC8p#h91aDm0!2(g9I`P23 zln#J$$Nc^If76RE{Z)UF~M}+xi3fFOD@c!0vQQ-N5_c3&YU2V zS-w|*eQs(%DbfTozu$jelcg~${T6+0Mo`YEa=6VhudWgW z8F(W~-J0n}<6>;4{1d1aRf2Dyf=CixB zOUdBIzovk#*ps_~AktT~X&;4kBq$shzFY*O;@=rmZ!mz|(JE>mdz zbt|m@C&sp^1Fm=cmp%bSdK86vgpLzA6h`A#$DB16?(~HwdAF&gdEa}3`^br>BCX~` zByAmZ!nGTFMa_#j4UlOgNx zy+2M?yaF3Vw#6*$W(7IyS^qj63)o%gNGS(dwbqkSAFSWJ4X%A7j1tI_h^i&(d;!Ts1(wKTY)jN@Udup)yYP$|f3|N(dnoj&+KXnN14EObCT+PWH+y$;#dw z90%td=bZcZdVjuu{{G_exH;#(U-xTV*YzA1_Ap*ml!pc)+Z&Cji%v(~q+E=&Byww0 zTnIx)s}5Lgj@1uhGs7mIq1Z2)cs`xcC;nu19meaOLUo2~<=>+SyQZ$K#W{@{Yxt12 zX{!4yue4dnzWJ9cj_br+{jW)<&dL`Eyt{4CJ+B^6)t0D zK7%n-SEsr$7t^J8!}i*}onC2wn^QdV&LKf~{Oj4%oiJ2DP({4iB`r!3d+WDk!4KGn zMM?uFdqZyeU+N%WlHfDpf|J-z)9>FhVy2OzBi6yU)`TCn>w1_=j(K_PQs}|;| z(4<_gPYvm!(*!nw0fjO zPiPdqQoND3sGF?edL0%!z1x1L0?bHZ_S%OEAqOJaVH%By<1IZ$$LUzq6KT`?Kj5Ys z-gkUf6VofJg7)6)nFri@V%ZiFZ^;`_VF@uFsCUTuVFZlpcp=_CEVhxIjX6uOD`k$Wz+qQ>Gt+YO7r-gFY z%9pe5rT-~*Ze!E$&&o=x9et3)Fn%CH)qlA#tBm3N_xR4jDSe}A z9irg{rARDex7|BV$dNx%b=S_%pKSZ2ktgq2#D(u-{^g%n!gP;B{jJRP$8R5SEDd^x z`e7nfzdBq3CbCx<+14_K!t=!Zm7Bb{&MPuBBKmm!e~4r7suT8Rw@MNE6BJATpQ4wJ z_S8*Gj!uhnvaS8iP7*$v)$d`BfAW4!63h~DaqNRZgS=+3iA52gTW-I5j3EbksT)WK z{za6;5u@oz9TgTs#I z$G2*jX7YQFr3%{r+j>(Z3${ZyPw+5NiBaPd2HzSL!%RgMtN0g1iYgn$!#>z{3)usF@Q?S?pj<*TB}(r{t1g(-OL2;z4g9R}&|+u-wf!jE>A1Z+ zb7YT!(7)x8-mMb=%UU>YtvnA{O{d^qdWaj?UPDzejm+7=KyBEN+ChQ8C$DZXM47~f z_30O>?nj){GE7i~d@nmrzhYFkHrjYQ%b(~(c%JA3Im_HGtre&Hpbd~|^(X77z4Z`f zjTiF-*2S~pc?B-EX1<|SM;_|(2y>|Z_s7qU28Z4ZOLo6qcJBPw(?p^0Twwr0`Mkp4 znf2?bC&`LZv3v35P^>kjqVsTx>Joi+dKv(2`>9!`cA#KY)rqB}!1Cw_ANinboK%g` zo1N->$zRQLp#y*I&He%^XL-3~9I8bu==lK0*~cUHGYycdapF;Q_x>&)-&Mu{SLm@F}fh6>LvoZ#C$S1Wcd{J z*Mx%uAzl&A`uOkjo7D6y%RN%m^y)c37~dwGEd8xfz!(oh^;I^;91a+AG|q}D$a{mM z#3w~{zX%`ztSBZOACDtrytIadw^?T};$a^2B4M^=eWF8X+NS9tqk`gJ3a%Txtt+gO zo&(OWu&q^`#)Gcl&MMB0F;Km6%tg1n&1kl<<@Oy#1yCot?J#s3fu62^;7k_ZNlZJw z;tN1hKQDl1_+m}xa#!)v2-{($Y{}aOK>4u@~0%{ z0osUVgC87SYZlrbuNJfKJ3c(}xss-Xl{-M&3R>UY!ZHyTMG(`i2HpY*)bTSqsWhez zI+TkkKTr6q7M;8HS$-iqO)W0w;ZfdRJOYXdTQrBR^XT+`_GRB4LJ`F?*30iKWcMVH z*mfMUrJ}74fY;j%DNFC%Muo7!(&-GC-t;f;mTTcNS=!SzZy-(+XNP5trO4ZU{vs;h ztd3*JB9-gZbj`(=ezyV4BbkA?@kDlMa5Ey}R6!#SXtb(7Yp5-O zDQ~N)j!kw@dqd~F&D6q6?7Y)yJ5aYe?e)5MZalR7btKDu9ya|HfKWLtPCaLU&qdI_ z%hA2rAZk1s0EM4K%aX?c+Td!z(x3roMno?p8ondVkN zfcs623adV9dN~Yk^DmPf)ax-VK^Jscju3C*|Cgw(YxT%~Fkj`R2bi=$HunU;o8@2t zBXIF`UvXW(O*5*q^X)M7Nc8uZEwuNP;{?Z%z{D9^NN~!{TiPOvD=1*o4OfPG8b6 z=e;Uc0=|z2_2%ykBh|1jqORZUyfh7^;yZ;npha=Hn;}Qg-$z~UxkV<~1U<=AFj4#` zz44gbetM7L=GAXO-j*8-I*~#AAvq12pCPWspF4I#&0mAAzkZawi^@#-F{Qdzt};G; zHumbLSLuU-S5lCbL4@m$o6q%hb{R(j@Z-1lk*;Agb#Xqq=MzY%#uB;`|7;V8X%V$~O^aw|K+c4Y-Q-W)ji&M1y<$=X-~k(6qu$ zTn6pnyle0~`5}rYdgRH?*y|j+vbl|ut}GQs>twX;me|s5GFpX++v!3+0e*D*sqGC# zYQ`lW^xKZ=oX(#cgW8*8Bktuhelr%k!d^`zrrHHa?uqNgDn2B5mgB@7PI)SyRZhUe z&?}LRPsc07pWcnu#sZBY43OryuBi*bwW7LR*p4)6kxk810g33q#`Z8>0$j}-S$Gyu z$eZE}r>(p-n!}-AuDp|~uj`6-wwG!^S{eP_a`logl0Hj)>$!Pxd6ZhCxj09I`?Y!j zQn$K&Edln;xn)VIC~QZZ|M1MXKHd<7tqY4-w=Rp%&{ILNLxCIEq`i3YC<;s@Ki95~hn&s|`UAfw5Dn;(CHKpAnWatISa9XASJp>-C=<_OcSzNk$) zeq4DRQud5i;wS@l6iphcPFlhiwb&nFwzx*3pF{5D-2M|1wA6Ww@fMb_<7VRNN(6Zt>)nIE7~$7M48GTk%z!dvf(E! zBenh~?HwQVVN;M(Wg{o~7s}?DFcI#d;g?Wz>@mR0NGNLeNyTL;QfVOM&Pz{apV#8p z^?SnlTMm*D1Ohy@vqOKOg}CW}?)hgGNzh1J9x806H^;xWudrym?~ zL_$B=PWNoc5a6aptztIaA)I466iMxAO~6=D5_=BF9kWe+kp@eH-xfF+>-XalgK6t* z*HLe*jQq&x{4(jHIHOHyW^c6B{mkYKL6{O-v-LCLh(XxETO2!qF@heu)sa99SczZPV(C z&d8TglOHwNH~GBZK<$Z065c{z|4#R2{YqeXc!1IBWpTErjOVE*r~+1*_53z>C-*$4 zx?gYz{DHd6-)S7b0##Qg;xP+I`qUE;Qd6bZ%TEm-?oh9yl+8GeZ~ty6z0bTViBKPzFBp=2f&*p4SHcgDyR2{drAiG$*0zP465|W7q`i+!v zX`DTG(v(bu$Es|6S2dAc)_=B7YcxS7*-M)4@I$;yQpPv1>)!v3aQ5x6bm3OGP{6t- z{(9eG-$H={sJ1p}UNxC&%eyE75aRtqQ+{a6dHN(}{FTjZ ziVIi1B=FokXCd&t6E(=?pCN&D{nClbz83!zUq@TS?Qed+@M_^yvw2rt_;af7G#4K8 zE=6(xk{*t=o?c$6G#grCC|JBKvM_UvsM}$Gnf9I5^zu+_en3z(=KE7QxTGbBqkcEPPY^@(~{gvPQwT)BX1-;Xo znm6mhW%+i^n)-KlP!?a`sh-YD>3v*6wSD6t`-M7waliw=02@XP_?U0Fc7!x)f6g4m z40$QvDV{m&zq=%OLT4VXcr$nq1Erp9Y+Qt=?bS9W*M*-`1l`b_rNZjUA@ZsJDAgBv z&kp=l1&6#u5Ceo9w)l!Y1zq6CT5<=GC=_HJQRfA3Y2l7g#0|*_K8=%h>Hxxw%f){(BsG_lOd_DXy2(Pl^IBrG=dDLmtC`s;Wwdj^0NZS zCd(f-58?kU4|_8lzk%cktIoL{cN?G=%{6d{X!Qp5$6jMQXC6dkNB4Rz;dc!CqDxQV zcf6*#wIuO7M;g4>unG*6p!$H1HjwM!VkDKa-l+zNP!VZjD)=2=_p3vOkZ@7W5eGQ= zL*FuJQC2Ao6e8SnZFY6+X+|=&*)NP=e*+G`L#pUhW5w?XM06$zkdBbw@6H-Q4w321 z7%C58XsT;(l$J6cZex7BiSZf5U<&DtR^b3?L)Xq;ZGax4TC3 z0HS_>4pczEJ#ud~Ucvb)y>SGIxp$T^xn;5(?vKj*GtRgwH#p(;!e^4TlL z#&j-khEYsBY&hWVn>^1Elms};=PjIw4yiLUZGOI$@r}&BrBqv*=-F#Q9xHW4NuNeB zv%22CoT(y}D}XdDUR(|d#z`p7U@sf?epHy8$zX!6HK9@DQ`esel2M;hoMQ3q(PmuG zs=(Wz<*+m|##B%XDy3oDriWbmC{9wY@+8C#&aGO2QelRqzjPFAZ@xaOYCeR34{#k_ z{qiHK9UgM$&id>nimkN7dV>vAYOmX&&UbIMCR6v*+N^#ncJNL|8Gfe^dpVv8-&g(s z3OlA;d!CazuePwlQty9LJ@poJ-ML;v@k4+Qdz4b6o--q~0Y6f<^23-gS=WubSREN^ z?(`NB@#UcNqBX_k!18Wqch1SnI~k@Cqo}<{AYl|Sw%$|+s?2p}L)J2TLsht3&c-#K zptFnUTV%8nooVTkyb-bCL_IGAN?R}JC*&Y>vsftUDBw5B4;Cx26Njq!`)>PL(v&Kx z!DM|_af;#xOZP!Ij-7Y7`z~@1z&8M|eMsl(SqL#Tj#tINT|VCaZmjtR1x9q+;gKY~ z=;{7tm<0FH05Bq6O+QLy7X~0l%aZQht16xkpd|IYbc3F$f24eJa zIfUxnEz%eIu%32$jg7DR29y~n>4)H~6PH@4f;&r;eTap`t58qE7xr<+-c; z0v38xHtq1$Z6We%mGt#ez8b7yAHMZu^}@RsqIA8?5BfhJ>Zb~Zi_Y)YMFe1`_m<~5Q2mngRv zA3t(KqS)iSH>izO4v2hhOw*m?6{_Ser9!N){IVBe^@ofwNWkcEHzSWtn+oXr1gO{R z`$sVqbKXGbVG;Tw4=VZT&bDy5J3Gr{1@Rf0{_LKx0Dr6W1q~WlI?EB@t4P7M*$PDT zd51uA$W<*fJzE8w1|QKefJE_RpS-AC0`tQpU!s~JBfu*Ez1BHQ@134MBpi;S=;53+MK;&}FZTvsews_zm*1>5I z;3+Ep@7=Qf<1cx^!KX&vYv8811;l65{6}$iCvT^EBG+j4U6HH$MA5ft05O&bmNmR*SfoB91wH zQP0w^YC^PQI^QBlY)pP+vI0Cc)aTzayk07DhVxy)r1{^Nujk2 zUr%9k=v>jnS=rY_1mFE9K@0J^n$t@6xi=mJ9`zM1+uxwh)9B*zRVt>wHYpZT_c_RQ zSE*+?--ACU8iw{LvmT&R1Amb3{L4@s#_=-NWRyey8#os&T$AIiaONbJ*!L*#{Kz-1 z37*I&LGQpny7undF-s-6Y*EMtB%*(25t;Z!)lCz z5=kOm7Js`y)v^{gvOi(&Fw?MX5wo^=f^gBQE%%;r_9n+ReT@aUjpodh1eWDg!4J4r zmDeW5yJmSWj&hn<7j>o1=9p~2nJ3#GKDIQ!^Dl(JVj=Hf)rng+q}MXS-S=8@-69*| z$hXIMB4vK?nMzfp6;IgAp}YJ_;=09B#{cPfcz>HiYb`CPfDi3_r+i~KwOZ}Zk3n52 zk*040?7fP+@yq5HF2&ynUI>>rwx1KZm~rLPa0*-AKWfq{K8;~E!CXU+W4nzdYN;Fd z{nvqrD$7tLs{HL|7L!`Fyx#t)M!S})E8ETc{y*TIt-{=W`cn1YmBT5v{f)-ss7T@V z&RA4z4yj_CyEi_oDZn{Ft#(lwe{%6$wHj$V9&dbY$&iL@6{b?=CE5TCUL5i)ZoLNY zmefz)V?*&RiUh5iI2twZLFS0ePcKnnk9eU|<CC*mVPqBGuJIJ5aVy*vo6B#PXRr%P*a-KH@k1QU zDdnAEw)bE`)X{ZmRyL{f?(72*<^_f&*34%;Dwk5?AL@L3Hq<_p7?L!qSjQk#JH>An zzsGkOES}&2@%CZA)wbPh`}B}i0`5J(Z+Lm8uVtl=jZr$8*Sh?wgdyavCVpNQxgGN4 zY;e(YC}z#@iZ$wk=P8d%nubm$`G+}QAM6;pr~XM9fArX`zmeae3(l;!n3T3Hf?l-VWI(u6}52wQzeg|kh$+YFu zHvjB`vhmvIqxWcMPNMP*HRqlHt0q7t??;Y-h6EbvvAq-&0-9%ag}7sBQ4siFFL zextCgN3WA?DJ7Ry(smW7xf}H*(3RkZlVna)2lR4(m(C7dHZs@HULfFYrn8zfEb3De z2jxxt4B;r6@?O0-(SMPJv^pxiYN|^Vy?2U)sNm!37F})G>C)}exZcPpCGwp<``?EV zT)7Z_WXjpv8`t+@-psc~-WsXgtCcTZ;a3UFXag^2O_y!J=`2N6r+*SwecsUO1XboX zIo}HS6v@>EKddynB$W%Ec#7}B5M@J#OnYwYJAY9xb)iE3tTfQkDo81*_NU`PE$NHd z4|polFybNbB=UNv9Jk-y9?rdNvy*k%Y8;g<(Qqd0!3DvZ#j@j9a%mXYOWb6$bp1o% z!b5Ai0|xks;B7%XCc=*DaToWqhl$OAgaCB)`A50IyQ9CxFjU`ogV28q#37Ya{j)B- z#6?(%V%&~UA_^B746bdx>NL0j*X?~R^rjp#xD=1EzW@)4Y-#IE+P$B7i}wRTs>1K? zugqI`jA!PD!MB=Y5I;(Z6?}W6TpDb0a=4I$$olgp@yzq%5G|zAnj3Zle)$q|h5KUu za!kvypxlm6C3)aBPFvXonTj~Nod$cW+)^$j;x6=j?f6bKPCi;+4%m;e8{@1@l=Dcf;_YuA}oJzwtLpk{^H9RY45tFnmp* zpC|Ml9b~r|X6`_Y-Me+Jzi~?r6#`7|Iub}`LDwdgvm;3HN|grLe3~$3#>d1MQoQE7 zznK@{_8YBXdKZTN^z`8hN1EhyY_C+>wMdcM&7^p*Xz!XyN$hdsv+unOg&Hd`LE%uQ~1R7u_^`g^Mc{FLYLh$*?1GI^w7WoqDpyw{mu_)xF zo)&VMq53orEkx$hGbFzRDVzJtxaX8r6^6ni4B=5RWTy`U;;)Zg6ES;c0ry9{1R}ht zzeg-fsI=8pin@t>8dfE6sBGYCxr+daDyX8rhKHf?F3bZ`8n z3g9qow~q>r{B=zJ)9Z8>CZ*|c6=um=)|?}UXbWE)^OvK4dbZCUFrvhe|6Y}L;syh? z?pzqUXVh6rP50j0X`XlJ1e?uCvQ>ijng9Aw6q-mTGm?E5eE+SA5wF8V$|u#Ku;rNV zJ#6Z1z_c>WUoT++!YlBoyB6spQ^|yR5#&vPDTbYPmAz;>zIPIOjV#||Hag}KH9U`Q zlUhzAi~Q%d_=i?TWzL!6&SQoXiD!6Dty7%ztFOFO_%r=E)orIQi|FY?lY4W^FDaVi{mj&JmO%KF6)v)%Hb8RzNMaKbiD#9-C zQy8@WMzb`aA(vXU%IOf;fm@yf)_DxaUmI=5lU&^2^gJa7C2ehEmK;;l4=bv7Mw3K* z3c_%tvaAKqjNPqqXcSIhw4*BAY$_qZJ@=j7s#hP@omPk&UU=AQy?i|{$Xqg!WF}Q_ zB+x*$`jioOBb2UhlbK-~SVd+w)JLnGj{elf9k~S`{9-qKz(?Cb`!!ev@Qr`aN9e?| ze_xJKc?QMIN87ce?B=R)lF_EYE$tJfN3BjQ_rG_@e>7rtA@6@`_Dds~2_GZWc+8=U zR?bzHgTk{O8l=ImR;vf!XD}bFD_afpPZ{CDP=v^l>q)=?FnhFMZ>aMVHN{pz2iFv{ zS8;g2O}0vYLRTWY4UiX-5Va(s1R7>PAq{*T&z)Bp64Vi{My+2Lh+6hK5)u53R0L(& zXWtA$ztG*ca@Yn!`_T5a>aW+1#;n_U5gSpwIMxoO^Tc^~pXy-X*O|vdSdGXM3`~8& zxI8AR^0KOiECKuzFVCU>Xr|zWiE=xA{HOZUJHyY_RF3cpoM+R}nJ^&0|D42vl3<3W zB^kf+nu1TNJj7qrqb~ z2%zR>22Z)WO86hyE6x4Ks#-r#Gx#NTk-OaIVBHAVKm^F#{@kOBkq zA|2tIxLdzjc(id%FJGDKhKnMzLu;zyETPxS{dyq$KvEB1M5&#Cx<5#&j%$Nu8z~3Y z5^ox&oA30PLiO^_*eL;%=n&_{t+V00qRJ6VtG_+($C8FV{yAR;yaKcLqBU`9USB}m z^;-PfE_$&i_?M>14+vTG?6hnDtX|?0J+xXJf2FYlXA2tyrg~48$|T@mM|^UE{A*cO z$rL=zJv`}i=NqZ)aPSA zJjVGt5b5jvR{(Rpk?}1~M_4T{GQtuf%CJ#j>6J~#*s{j_CqK@S$${FF>{>X!xKS~}JPeDqg7M}c*e&C-@L03-GwU7N(hVtU)r{W(32u?G0dK$5 z-}tWyZ?LqyjGMq?4#9{3ymirpNaDV@h%RMZIa-YQ`*yKMf&vTu?)-Lm%p8x&jyxZS zx}7GGZjhb+{esnnAs6RFP?svX^iuB2E1o3uv#M69Jna)6Kj2kuy`58yIu6Q9+mt0u z9R~;hVVizjp$;dm4Y&NgG?+eoxj)wT$XL#Lq{KlBavx2GvD=(FK|{UHVT29Fp^08t z|7%ho_Tpu(h>%+9`X+KWKi`O2dK-U#Vji6|2Um&9Gn|8-6D#HF!N4|rkqB+jnM0o< z{kyXe#d`!_CD!LXzjPp?r-Mu&7wn_5Lv%7X90CWgGTBz^Dn3=Y*ztE@I8WzV;5sQj zt84iwvX)>d0E}ZhSw{&)$$xKeBOacTNWCE46f}DYTm1a}n%q0j&(Lf68=-&n93o%R5B*VBO(1>$JuR0(^CZ_@Gz)3%`2H{kR%;MuK^G3Y#xn<_t)fK1w9 zNQ1es(P2?YU|o=vS6WYajRIb=zwrh_ z_Bjt*^Qh-X2%iLS3Ae`sfM_@PbOS8xaw@}W;}i%i)oU2Ol3#?Xls!^cSrl31{UDLgrUh@H_8wxE zAaQ6mMbaLc_>&^WNC4aXqctD<(EFXhk3#ZCKiElI3Ei3>p#4XY=w;&^YI+pOY<%L; z_8)+UqIp$lB+pQbjaMp&zfg%LM;K=Q^oR+7(ndUX=NPFAhVG{{oqqZ!k-^Vsz6zLVjxAS>lRxd)2G0Mm)&SS!Gxl>ar|aFOEmDD!Zp

+>o$I(A~Io9t15qC%NU-N5dpKixt4!Z9!*{T=B3*9ddSTh#Zm)nWhzcvTQ zJd*Ob4QiX$?WiqZ==(sg3LD|>i^%eSzDZs1HdLs&+|IvC4n?Kn+z04SoB$V(THdK> zqBmS>MR+{Ds=bE}s>{1U-F`umb%|kSMxH zfJ^kwOt}!J{L$flVz{~8N%qkL^!R^08_1f=Cny!$xJr{Z)XgN#Qo;xMHPof0u_v)2 zP)6P>lmm=ctNj%U8x?nflTh^U=6{{m6phbpjU0&N{i}5+ww=%+3^6bz#)!m%Ge(WM z=QY%R+TsZF&+(?RFnRTElf@axg{fV59n~z=XzTSYc>ZGRxWST@wOsulkRrb1$&E=+ zJ6h#?-Ba}Sqgwa{9Ec$3AA_5})gRdZG*CvBWrym=xybb64~fT#ni744J@o!G z#4lai<1vRGy9Ank4i4r8-=P8u)q4IKYUWwO7!=J#>kWKc^*i1km5o9rJcbQa#eHQt z)V>=rsT?Yicx$zek`K%72RGyI+X8UNP2SUDxb>sBU^(q2tAM{xaR-=<|E$&nmmW;p zagY!nGnXpy2o@quE~Mlq4$rB=~;Gd4FS$;H4d{vYsNK zlvxIqMlybfvefYcRCLM+9dD@P(@8k3nRm-sz`I6juB)2*-eg=$c6zy1XVTw@Ck-w7a5S^@H!1~pfaXR!a9YHVxS2Fi=7Z|r za`kxj;|7Ot;}Ug?;A^R0@rZ#`?n)hvtYyW&QR)e8@NU# zkxT5vt6M|o3iizUGq|qdq~p_IWj62mw&Ox+KWh-I>V!A$f?cPll3y3In;UeW29c_>LUp3c$ zFwLJuHqZ?(pF#yt6kvpjBt$oTSu3U_X>tlbQ6C2cwlD5~4en91dzEoMcz8_}#*6Xj z6o7foJyH8UJ%jy{nC`iYnhZ?{S<=+3tt4N{Aq@luQGs5IPJrvt5xyhcKdVfax@_;uO-w1K{`!3LS+8?xUq&qId~YKYr6R&;Dzub3 ze!QGa^3R>8GZnmSwlZSkd|J2mCM26VGWY?M+$Gf?T(ljH=3y?GTs1hO5T8%AvMd|O zI7YazUIQUnoSJFrg!|7~kCI8#Ptjygg$f%<^1k!kv;Ma9`hXHzrhwl`gNFFH`BJga z{<9Mh`aCnuVF*>;q=D_6Z}*21Dv$X2T~v+P^4RC)tw}myB5Y{L#$!%_Ib>Qu69;ov z%Ke~GfiQ62910!iVIrev&h#Z5{cMGKbl32R*4Fk1egrfMdErEcEEiE+WAJ>|amke> z*_nnbK&)U#s;DR5+vo{@Gok0u5}00U!`KCNvF&u<;HLI|h9>`Ys><02g*Cj8XdsMx z8bb$V=Vi2@7r|D}{i>Lrer0@J&HG~tqYdO;r$~gj?*A4ua=O7*RZLsEsI_VS1N%>rMdFxF2plNg;+9JRGEIXg7xDoRmVggh;w ziD){MV6X2$NJNY_ouk(87Acf~=Z28h!nU!JF(`h)g5?c7LWwzi4 zgj9cYYCU%AkOL(r(vd{b6S(~5JpPs2Y(>o754xK=j>5XCk(b|`c$ic!|W zCeNdX=6=9=LT&f5+FUGeN;k$focJ&>H#$jxy)Sgt?k> zpX_Xinh!9(hovl6x74wzu?CYZBU*144Bq+9zf2Eh{$0=EyN1|_&@$dPZtv-wQa%4- z10nRo|J=+IEz0**h#>zuLL1TK{*WtUc>Ie%hp0ZGFcrD}s>OM0_G|bs+4A3dz8tEa z$>b+Fvw~>zwx!D=c4%CKyx$)j`fL9eLUiijtbb)H=Pa_|h|lGZTK(B-=WyTnp_ zwM7i_Ys&0)&cQH`h+XerdeQ=m_}9}>n9MczYRoiaOBFD34wS3+Isq{+DoKGBigFd2 z>xp5SU(Lk7rNm1H{?tp&frT#E3tzFrcMkAPzC=uJ%XS$!y+M4_5I|k8tZ#);ev7iv zis>-5_!MWXx5JfWNyq(J6c&O@xcGwth1m=|j}iMpPY%rUWXM3Jsl^gMU1+B!z47-S z6$Xp@6W4cTB_9uuCz$LTvnYU|1wr;M`xESV7x7acMktM)x!R?74zo~xf*idiPN!Nt z-P*J#TN%gorO+)1xn%oTK>DwKC+kWaBGIGZLB5`z27!Bm)IVb#DHTtSrVf+C5P_BK z5__`rq`UHfCFFnOH^pcw?b6GgL9}&{7|GvOBbqJPC z2mG5BQ{J?!mrQhpm5=G!+q>zw=ebN*tig&j$!R0nfZi zM~w=8^;zw*UI9+Hh4&Br2gF0*=LLi(DJ4*BQ!Sgt#b6`SnKzL#&XHWd5n`kdr}AvQ zzm-F;&)N(uz+4V^e6}m{Ftnr(N%@hi3y;=euQ^-mEyhvIliZu}D23zx`dfs}@_5O8 z?|^O^D5Ykpmr5SpqSz>`xm%laY`~8LiU#?MZ?VoGzP7aE(q-xI?8$>EwVU=*fFmR> z%WZ#>Hgx^HqGHvqu{}t6wO4MIRRH*l#8eF3UlwTV3_}?uG7`UFVSiI;S=wK$#Vn#j zPAS}UED`yh&n1{Eq?f z&EjTc8_elt0Kdp)gLn0Q;H%-uXoOqJ%Rx=jnHRw_0V)J;$MznUeKVor_LuBD#*`xA zk+Tf;Y+;zb+ToL}h(e2N_1~KP|0z7IUP!+oWL@$wqz$XYnArFB67kESt^8@yqn5EN zQNP8aY1lbJ>1cB=lEqV{?LdQhpVs=%@I~~EK15+l787^V+XVmFC9B`4FK>$|FQSz7 zqW=@fEszIWKi8_^JN+j6isVPeUq9L&A@R3}#i3$F;}_CDOs!5z^LnM}FtL7%W&Z)2 zaY%Y{^Bl_lS2t|=Mb|9i4V;ertAPuB;~6|v067oi$-e3*C$2+#*|>yiRa54~dQOFp zaM55-R$yXoKe!(ZkO;E8!S7^Y7?;ao=m)9K$JhoVgyXRLvtc`s z7@Eu$3{z#^^qdO&1JcLQ_gwdKD(3m{$>9y!h$=_cQ*FEH_jvW3Sq_SKKBeO8hImsR z#G7Wn=(Hv?MTg5*?aE#z+;&G4D2&Y3t=12ZUz6s$cxG?BSu=%Sw|D&j{g~(HkP`LV zFXpDZBV%EC&1m?KGX!z3^hg8TQMwgrA+M#oc+ZTze((OjpwlG&y!dC|c4eazGiU(( z`Jb?2l-lrSAt}v&v4~I5YSxJgQqt>NHfyNwUmd=IeA&NEN;)5dy?q}j76KpR4u!$b zffEXOFxH}ihP;ZXz4x}f7v6l*trgD4D}_>JVo_a$mB*5y(z*e<2tgx6{VIrFYsNvjQ;kCaRY9@DT|tsBU>umrh)E@ zkp|(G3GfJ#S*@PV_+D{-T+`#Kw`rU8zw&W(Pu1nUFCgZwZ+7ZIr6-5do$LSb>=)@E ztYS~&CO9v`4&>ujYhKd@1yg;3<9?%}kDVW}65hhnNNnga%T_KO1>@bQ8`@ql0)_R0@d@ zH=S|mi}6t{5oZ_N34d4TZ0%Oq?jqV%Ue|ht`gT^#&A7WNAD|~0Pdo;MGsl}iazjGg z!FLjvM#k@BHgZA$zoqUI`9&GY>~_fB3>)t$RQ<)qdsH8I$DtHr zjMPMf*6N)16W>8GuwG(=_q9RmWpRe7C&kfRWdAij1_>=hF!hMtM;j6BjN;P034VCV z)&Dx^WW#(hlID=MQ%kzfA^rE;Rb&SOdt)d^WEcvRbwQ;ni$p7a+%-OvxSC5Ql_m_( zk*g*|Ysa9wmrBlN!EXL$GN#9Oa`GPH!_#3TgaDuT!egTH@CRit@@P5M<);Jp8|TpD<7=D%{vF4FOoWDHpFqVN?9qa~bguk;OnFcEauQ3sE%FwX5s@0y7j^}UCEJ#EEEJUKwbw_FxL)fdRn z(?DO;XYIfKnM~~l#e7aqpwCcaeGrcukMpZ&d@6&wPUULz=7oMAeZY9VxH0L47*c%V zLJ$5m_CSOlH3tHIjGBxNoiQ@`ON|DH|-9kLRp0NS|rAZy>ow_iz>qqk;0=|Yg%enN;0$C z9aKlZ$VXa6aY^|dZZZ8oiPH1= z{BbdcEIrCaC-G1BJ{%P~-pk|;6wFW$GbNLA)fv5JsIb5kps&Eh?MR+uJZf7Ca*a1} zw1xmpw7_$ppR_1q14X@Uxt0ZKpOqdSXJ#(sGDb`{7w4qx!`??MKwKu$jajv~MKiJt z^~`lT|M3KUpi&z&N9NM`OBv4PasA9O+#R=fGOy$=R%7^>zKfz~Z*IHBL_Fy=EA)mm z@aJeNcMcl{T^zomxA10Ka-ypdeGEk%VbPUpZ(^w>k`>=Ez^=9 z1W3+Zw%R9ph9S~`=`v6mNg})bF2j~{KW^*_XIaX;aZnCBv#-Z;{C6WfUmd&CWl{fi zX&zQ@cwCx8<>bnG)^fGk%L~(>bE>&6&VrxSu`CC+(r(uj*Tqebd9LxnIwlxd@~ntMVU#;scw;<8)z8TIhqK1ArpxoT(-g=u zhE+1F0_5e%z^PX?9Ai|*>%&$(5_`AZGmRIyR~IlJ*Kq4lDLm7g?7es-lIiUzz#&GK zP;!oSy-$$6pRo{E|M30+vGmFgqaP(uBmSHZ7nJ|z+=;xHQ5JFhUEbGcw8 zQH3c#Lo~`CbZ&*Wx5DfX!hY7aZ*J$>3P@4oDck4gokK#)r(XKCG#DJ3sis=ih`6v$ zye;ZXxIQgOJn67MDjl9zY|*~?d@m*WmQ!eUCqkte(pAoaJuuHJtSA;exiu)jFzyXa zXiemOfo*8~b7PfwBp(f}S-3E2gcqZceluEU&5k^cUCr`O`j*VN93GInOA%Xd&|%Y6 zOtu%E?%Y6#MoR8U+u-Q(zBRjt#AzP%lW?hNy@w8{tpO6&MxD6n(yddMs_y6Vypbeq zx2(o>)s?HlS?|&0fEIQ#50;&?9<}a?$@> z2n(E~(q{A~PIi3d`ZCOADq01A-HhWu9Pg#JbS1tXDBV<9+zjZe$t>Q0Pid)r{ssbU z4TTdjtsX3SG+FBhUuzQ*qKG4e#d1R-_M3!0L)3_E(Pp2QZXK>46fx zgwyW0Qbq1>r9Z0`0y}xOLe_VQ{puv=+%59*ipp7+&=DR63qhI3me53NZOqtb780Vf zhvk@TxND1ykhfIyX-6FWTX?Ci3fgzgYUZUFpj%d6N=c+*j3|7;9RZ5rn9L~t>T-w| z&#zpD7=cQRtH;q$I~q`U+J&=0*VFIbhm)@?|AN?b)zqkK!g@0Xx!D~iL6s@&VL1ZM zMQ``a>QC=K6wqaY_=MjzK-@b2dLL^ce&G$h8Qj^ZzJl$iGXHTys|=~IIr;{6s$#xS zrsc<%P*q1xwo}h)MNHv;$Fh;hSHh6XV^c<`NI7Hn^w!F&0_whwoGWUkxvWQ)Qqkx8qGicqWZa--=>TgW4qTtR~p04Uj1kM@4A}< zVJNs_p+dm)vYdJY)p6M7;^DIy=)6pdamXIx(Mj*MMmB3ocg`l00ta46v6S=DiE4mC zyTp2Uwjg(=l?x&ms%=Np^gh~FDbXhekBPiHpF@XA-P4V_ zJ6X9TyLi1=NgHnJzI`wq}G@K`)5-_CTI?L?{G6r@ zutI4Gvp+u|4R9x;6g1U305ZmvEwcpxWX@fUa9oJ5x0&DWk|--8%Ecykr6B)K-1Si34PC8$Ivje0UV+Y<~U0 zb8kSi=gpN3EC*|oms73@2WQUbE`)4q%r2fnf=hRp z_n^vmDK_c2`lIfrCSm%0EwO(;%t4kZqGo>cL^YN)=1kaQH+#s#UsyriK+$}Oq;x_&nZuMzz7xH*1y{)Ep>diYqIRY#?uA=-27PTiMV;*;PHg|5bhs8)M(5N&bA2LwWDF{7mS{l{Hhx?h-M1q_G=tsSPl)X z$@w~5m?y^RD`R!H5PQe}U13vw(w%|bsE7*U8=HmbaP8amz48Q2@eE}(zYtKWTN>J=+VP#+mk`?8hW-shZ>8BVEZ6MB zF`w&62+wI{i7g06D0Sk&f8It+pF%ip6R}9I9^9k%!a9i4+_);vhY4N9;AmRM^d$t# z%wO7qPKu~6PUB5oFi=twF7ni5d_tWrBdv4@2$Z49A@<-9)vv(jp%D|IvS&x- zd;YTfXW%@}ups8v#xQt`vA;AG(z#EB|HAoI-W7l}LJ~0v&}eSfGfnU-exBC+dS1O%I>5rQSx@Y=C+k0H8C`Qd5NH9XDdcIve? zJjVC7MN9{iU{%M!{`dei{;TP-8+|JejP$p4d;zP^Ys51~KjA?3V@HStb2f3Tn<&IQ zn+62tkB)3C6KazGlQqQ@EcOu*!8c<$*?R;&=$mkfHRpYy`d~EA(^t0;leJryK=LDT ze8T}SrgW)~2F1YyHX zzUfsL5eOr^z+Eg1-yq=Nq#>-yQ;W?K9724d4_EiGO#-^vF8bt(Iz3Orp_K?28X_jH zdupK^nWz=QtuWIWr7y5F^orE@ zP$j^wkpa**<27%Nl_O-uYFhRJ>G|3%&-??Bq~Yv zO9O#kg3r5cciW}@{eHmBq(`1{Z{=d|HO!W@UK=73KaEW)oyJvNR`AjI3{mahINP1q zN(^nq1*;S?;nPAV+mAyL)(js~_9B)kZOrzHgN}Hs@(;6qtPPd4zA}eE3gSe$oqj(K zn5}|5Sc`v)2;5>4rbt^f-<&=Rh$;Gu^EI#}a^dZ6QU5n4qR8>B7>oKr!XO1}T<=kK z;^1jErq$YHZ$khQhkS*CpLwi2Be4SD2CAVrlYOl|Q~_Il3K57sCQ)$<2=v5;P>fyX zDPDJkTY|i(?F8{y!*_VmAS_saImj$nxQ4DK)IJ; zY1YpSUw0G*5Bg9mfvRQZ@k-X3}fMO>gzh)L_|%?v}ziV_8Vqe02aTt`8C1? zklg+CThi3+6U&~@@O>A?jeB1t&0#)(UwGJi>vzm^()d)^dn6fMNR#OB^6e__?9F@U zThmm#v&Pnj!a`DW!H}0?R5vOxc7rS#@I*c#Mg2x?z($M;$>t)HUOm8YX+l(zvX{4M z9+|>>*A!-~x~4!4ubIQ&6zIc_5KH1I3-KH@i2s9_#| zHdumY4c){{6-4W#Ef6RO1$ox9C-uD#u)Ovxob;6VDaL-2;DhM;uxXdd->|Lrd}m1M zc*AumYYuS-3LpCbSiGiWM+5})g1+;I*lA(r?N3|M3|x-0)n}~kH^H2fO(x$MwTS6t zo(04;M}LNN)f2w`hF#=6+ZkOBojkXer!VyVthr`9PkD5^4z^1=CZ#3KNVI8JnU-u# z8*|F2#YKSSXKd8-tW=_cHrT`zk6TyNY#v(HWr zmsJq5&w%$bH`LZhyhw=|Ex*Mon{_;2`jk4#j?~z+J9M|FvA%#pym0y%aaB}-7)4zD zJy-U5Qz3q1#A>6$QFGai8m}80! zRIvz_alZ(M=pPN{Uv}JR!N3h4&KO zt}+q0wT+o@pic8YDX6e@tc&p<$h{THC5)wn)NwAtY>uB22q#Ly4!Pb7o^;gXvQOUkD>fld7ceh<+cpiiC1u!MLWBDSeoU=&%snFxfyRCqUdBFs!6SsOrJV(*H5?1lUCP2yP+#$yuFBM^gw|m8W20 zTf}O`jh*^O(z~$QI)X+0P6eMq+>Dh=S=!*P1$6I~`d@4cQ3?P*QKYIaP;!iDD8Yvs zLZR{FZMqi;Sal&HZ$$nK&%v7pCcZN~tC*`*DY?|s>ML=QwP$3-=>V%5-;{+(TlG zhV1N8T8^>{Q;OESk$14(Vre0R_GQZ`=M^AvSr3km(%bb~Ma|6Qf7z*-_hm|{7X)Hc z4c^LzUd ziF1khCR_QlO1P|#)$z14O)nT@@3OLB2z~>eV{!(q2T7sZsQJ}M>Ba;nMl0%GfdFg( zN58aNyeD?I@PHGqbd&A$-yy)A)mGi2x#Z5u_KD05KRDeBMq zIqEiN6KAZ~R!~$CImLMF7Ym8|1Uloo!qM+hl=hmY-Mgxx@k%U4sROM%IiB$e1@J8! zOk!AuQlz5yly`o3G=`m4sZ0jX7VAAFC0pnGdQJ92F6Kl7H-6oXlFci$%~h%@lF@eCcB zwq#7|ryY#kRO5+{l$&!r6=I%Uec}xyGnUKhX|Kb((vL8P-tw(Bel^EKv{&dyO^@`2 zQp2#9Id-LUcH0BUIKxva6KL2u%CvYz4fNJ3-Dtm$^pCD9^r_ERp{+5>PUSkInoEZ! zrWSSvhW@57Sqtlw^{=M1t2A9`x6O9-il#A-y^^26)H-GrM3ZQ_{|$;Owa_1KIdFkMaMMM%?;7w_}BN?36xa+UG9@hV{;2LS82sJH{3{}rA634M^4?-99nkkNpOV|+@`0{Z4QS^x4ZUCl(!YT_9&Vs-!!0c zRk&3C6UfnGcY~H9X_R zfteb?3-mCcHq1N{#haPF)pb0CbF0gaaZ?VSACyA%yuX`a{GIN)_& z>-TEK{f^iKsH9`QduL6PPkPjVd>=Jz7Q1T3FBOsBz@Kk|ZEpGnm^r}rPTU`lz-dbVz(@sb%d zH>G38sJ-G|TcLQgaadmSCjbB5sPz4Zu}LJ%{&jkbs?cw2DAQ+t&tVFo))Kta>r+IZ{jkO?uOn|ERs$Qv6!dV-hoeK^2chd3=_p^t5o+pP80HWihsktmWOaHC)b? zp!|fd`t7ofwr_e{Mo?(+)0s>Ld*c;BSarmA;1eEQ$>9YKy;$#3L@o~Az^a?WJ5IOu z^%&_?&0^MCMrmmUDx!xVHC^|7s{A4X=Hczi6m;Sl;eEC#U|u<|jba?<}mi zG5#5uTIFAMkcSe9IR7r??B(o@EBfz$G0$R&JR8EdCg>9XYE`{i$j@dM(RvwaYc}eb!&_>d>JhpDW${58Zt8Bv9lq z*Qud1b+InO2M!$IGPS&&@#^q`ArS?`cxkj=JK9Ds+1?v$!>Q?xAm=)aFtEKa{HJXr zSo@hB^|{kUrg@@Cw%Hp!&64k|h&581{W@ZWbcXrgj}#gSZ}-!WNHLLc&I587r3z1p z?h`4371y)7)Svo&qV(8qJCrZMv9E(lv(>=9z6L>`9fa8N6#kv4frnvgC&z)WLaz1J?{^Y=g7dak8IhTKZar z$1c||Q6n5{u|&Gcq$C?_C6b@j)Q?kj+S$e=x+yqg^iF9vKs-1^?*rQ< zZsnFsJ!6kuzs0?DQxAUQMjt<-o?9}H;^qd&V%3M{QEA%1`4B2c+qiZnjU!5`iy8mo zc9r>3n0Y{JPv}tnep=XnAE`yj(ON6!91sQy8DsP><};l(U9Y0 zYuK3;j18aYH2w0z!>gSD(_>YuwArIheneyE-y4@jB zZ=tg+8EW_Q)+^jW-7;HIxj7!Pbl#zvElm2Bwut-b{=HD;p_u}3CM93BvJ8)o5O_v$ zRwfxVG-loo`Z%bMnpMkVl*aC}&x|Eg)yXPLuqb&q>Kvx!C1DTYcBM0#rZV&ZYR^NL z`a`Iej|%#{t?S}1@gVA>9_G?L$l=Qg5kc=zU&CIiSYei6TQxb9Xx5YS0XDQ_NIOBI zJ|@SX%<<_OTWNpcH#Bs}FB0*>=!b3_1;%z^-k5%oeK3W$-F`6Wwg|I^3NUv<5`Y)J z?yMgPSmFkGwgq#qz&nV|ee)pryg@#?{5K5lw=|beKaW*!KXmvyIsC&Ze>PP{@XKxveK*L3%82tYFS#uG+kdZaE_+T*%%X*5R$fWUF{g=!E2e`{s)%uf@-VcCBRhK&T zgKz5YD+i7_zf|HYL1CqX`_pJDXCUsuq2&ba_W7{X!xr9k@WZ`zY{J#`FVeZ})nQpJ zEO85S9!qTZdCb<{mPwc0pZFcxuU1@gplKc-p6Ijz(%zukR&ldXH_^W=!DerR?OZwZ zkM5r)G0!69_`s!(B=bfgP!kP5*6y`U3^_BMCPBhv zD9cjS=cH;sJnI&q8iw6HvZ~pES?p_Dp+cVVrIgy`Cz^1^Uh9TkhmA|O^;dy)QUUUt z%6WKItFk{>l;un7*GEZR=nZdosYCXFH|vW87T3xP+7j z0;lkMZUjX?NWsu>Hp%rEc)R3Gba^58B;Ws8kAC}wG3s@4QkETyjO{@`(H4Hn zz-whA!4n82K}_ zeTBXsGk~-g+f^IMwq%>DC7gt7L<(G0Yj(LibZGRe_n`-nV#Of21o+`rJp~>bn-GuN z#(&Fo<6gvF?+&`72p7M|HMv4W82$RuoBf*qaXHWzPdx+u1L~PB*;$mAqa!TU=gwIe zXJ~^DWEeEw5jgA0-KMW|EF+&nC=XroF5?A1+SU*Q&iq=T@h*6u@(inU(eKh zF_>LPIlcsKW_jw5DAI0~P9hoWn_b%}6hdE_cTn_{^bb_;2pBT*T1UE$T)&e2%@M_B z*LbG*Ysz;tp7`Es^Wy-xyA-ziz75~JMf9jlNYUww<)qZfq<{A(y|vkO3rcdu5fO}+ zCe|{7+29x;E*}8U5bq0%lW)>;)`TKHC79MB*Ht2dR$a+`!7CLv@@1A_zZwsmllbds ztU_C`*Z)dU|I~ZNKN5LzGaH-bBbT803Nj_si0pMbhea74@MB#CNCXGTlFe)nflsmq z3!&d06#{k(TfeZx09)+GY>Oe>$zH;B8Y}S!Qgc`7=z`XJ%PNA?(U;kO<65MZ%&S*6 zm1d*YIXq{jYL``~UTI^A#%1E-sEJ2jvp;fk);cOL?H|g|>(@_vR)gMG?-Q|O{DI=X zk{AH{vTDS*cUH*AcJ)cy&?93gZRj1DrNJk(P@{D=C6s1a{4>wG#Z_a*4=cCM^*Z}z<#kIAPMfja|nN8}1dqzcT2 zp-JWTs;C%H`Qg@j3iHob);FMs!*8#znJe+I&62?^`BG)~K6F~a#2B_PRktL?qbh!3>;y}i}3(u5rMKW}V1s{Pqm^ z#9r?BYZm_Ov2B}vG(7BCIP2@)6Ea64i-q;vFHtA7+*!3Ak0+yV$XKj0;|)HDL(u^q zL?E$QST1>QAl~)!MeJ^~IxX<0OXhTq4XxaIH7*)LeOpiVpRA-1VhW)9z|8cJ25wEJ^06Q$2f-4L!wWYBxY4L+z*63g;VNqKNY%b_H<&YU? z#J#u~B&5o(8L4XzIN*>dV=n9g69E-M%AA7{aKztiS>zDLgSbUwctI#@}ZX5Wg z8%wl7LT-oAP@5BLTp>M|=XGo7ImtG{f@&z@&pC~%G#7LmrTOJa{^keVz@^c3h_I2o z&wc&Qa_u=qV>3MX|Dm+?dIX z@i#3~u*w1d+w_%XTM*F+%?}mS8?uOzM05M)ph^W;|BC2n zay*!`nXOT+clcHo7UknyV8-fOsS%DX`%ba){`vaw^i$C4S+X6iZzLW}il?;CE!5gg zqr4`q{M?})`8O;gW{p{gM?wUuwYQ7kFd}>}3c)e;V^J);s7!KHYt3?A*dxrUi~Y$~ zvcdSm4J`31*yJa+nVol#x`O7t9U;e}|+cM)Ku?#ob=JqzzV`b5R; z$#2Szc8=8nA;BWYMj*~allB37PBwE;s2FrA&Sk)g!Iri}e=^ZRfZTt3Vy?_PRO`b06B1FxN#=bzy>pnsl(G1XuZ@1+| zZ!y^6ZQSn;fo}m+0(~LB^HX*4kBDTlt~i3Y`g|{B4^f8`9s9HdJ2Nx=luf-psxaA2 z?>d*cOU>tMxH2j3n>z(M843Bm316>q{4M^Oz3Oq$`Rp^$U^mSAl76vi;)Ki#VVgZe zA9%6ix$w;DbxcCJojj%1AI+#BSDljeiEqU?Ohhc;c%J*UG2wXpwslqL2Tfmg?bNV; zYzIGYVTSCD?9PjwI6-qK!MgIX!<&LX?5}Jy$76+^3O2LJnY@{hXL{~2Rfy~{Q6?1T zfBHs@rbP*flgT~-C6Ii6!A4lrN-}BYSV+m;*Qe;H)Pv+zQ5lD;d4T*+v%R`Qeueuo z(QpwN4VKIXo@1o$$2XwwX-eq3F8z0MVm>U*%b2(KC3)j3;rnumTMEALtS)t6c>myn zMDi)^4t0UsFjm-Fq>On>;WA08%TM1c`Ee5Ys2Q;7rgZe5Y&&ge4VO%@Qe20lj}vQB z0}u5v32nLzQb5ziy@_WfzE<}P*<(8pD@d2c#$E?~d`17O0q5GzufZmeICtxo2Q(iv zP1qc%xriOMW3cty8TS`z{YUCI@L2XNBwEm-A^FQ5>~`lcgX^Mp)4^vG0kS4n5E zd?O-O5)EI**$1DA0-y@lUPGe(;2DJ|ia%oz65MPGrb#Bzr49=5d64BX=~50EI0P7=J7aXW?k)lxooeeo#2rP zgV7u}0gm^Aj7s0iGhj)6<{`!@)Rg<}JS(xo||H5SQ4heIMC8MNLxdx3IrKb-h z)`bEto=~JMh&EhY->O+UX>)ateeqp~_V_$e_48p%2`JDnIM<^ofrJ@Yv*}60hW!ju zTG}?swab00W>3?CTsc;>gGuN+-p4BdHoumC6gW@3P3BmSpAbdg(xWXrD>jF!-+j(f zRqUp%M)Rt;Km(&B-&r3VfrAyhF5=V{!A0VoHk}NpwPQ?(6qJ8-tEssvdv&yBvl|VM za^MBz!2He5xv-i-u$ox3&ufRMVW<6FO?U=HH_5sVkQ{OCMsNTgyK!CWfzb=B`n2=i zHehJx#K(FTek_FQC4!ch0Oy%-+mUhYYac1?Y0%k2k+$LVok*^#DQ&e|xB_V2fQ!lv z;v8aR_GfM%Pym|*>AP(Rcsd*y-(v|h2ROr2l>YJA=HSC>=}+D|E+?T zjuI%e7eb{(u;RH@Sr2X?yiI?tPapBhTzDHX*;I$7a60}rVGmU{dLFsQ&hKc-sMX8? z1AnLXy@Q19hYytOA->f`G6~iufonLuo8VI?g}h$c>V#fw!Uh3MR=ok{vQ~{&=)Op2 zrv*6M?uO#6=J8j+nUq~G@SgJ_-VU(pK3i-F)_8Dd?tN||XkdAx@4G|~t||5&xzBq1 zkUKg2H_WVfgJX>Z5mER7?+I|EZRf{Iat5nzvh6%e za6&^VBmS>&_uMGo0q3=Y>!iyK5)39=^aIG-4YqNEF&lGX6oum!ROh%y0#JwUdCix5 zr&6L{R0Sq<=^tO}#5m~Kbjmq{jK!4wTZk-wMcTdsm$$$;S&ByRMpI z=I7N{-T?gYf33L6;ZN>su8wxws1Nt#TBI22(J&Q~zoGfZZ=d^=Y~1KpTWaUPRqFyMv1dDRs z+VR^ZePd(x5bjX8DPRh1)9jKx-?m{+*fW1eW6|(Gu!KvI(AMybiR5s_b1dpx9aykV zdU3-ecnQYC!CJ)*g&)oXjvd&G^clhlK%1p?_^0w+SeWp1fEWW-r}%qDDyzMAC<2mb zVZbnF$)CXqOjJDopIVwfIU!sN%)iBdjh#=Zy?JpP5$p2yQ4E@UGR-4p5R0PI6C`e7 zQC6D0o{}3lLBB)WlQ5{0$4!Ve6_FI~d6eDMjC1Nv4HQxbMrmeiq}}dN7Pb|aM`=G+ zHxpA9c#4e$SAJ^^;za4ZoozJ+#hBlh6lJRsCC?oNRjc7d`~xo|LuXQ0(EkMqRp@2fzUF`5B$@nwJL&)4z~-|91uWeipT-14&STC5;r|$cZh@hCC^Sl zNXbLxsBff3tt9TQS=Z3E*a69xJJ<#`MxAs z_eYWAB>tqFT4h_!))f9w!^P|A^e2SGA}GlElXE+3Y4_O|z!8yr7FO!e=|0wD(wS&KCtrOCjI#DuD)A+=;rKPW6v#@giLGzzzH? zN3Pl@fGseacH=EjchhT1hFaQL^bW+!3b~7( z68VgyxM+D#uVlv!Msr7>d`w;M63jBHr%RJwmj-^}ExId`oj+KA7N5I^Kpjei#)Oy0 zqb>{3XpO51e++VUSG{@C!1nG@RwPYvEPstu*cfMDDO(m3>kbpU31M^3V*uwMf(+D7 z3&k6M`p|G(vn;JP-#(>}7J65gDWw}&)V6mA4gYt4WV%S1TAfr5=_JUt`G((A6Tnb^ z!vXSXWtw+$=w5sF!>n??4_h=9T4B3o^GK4}f!JJ(#5fyW#1T z!mn`5<;MkEF4`KshIjhIzF;1^Ha>R@Q6|5wSV?kx`y@BJ)**VY96Ll_ zCV>yU8U3UI4Yv$DK`AP!iG1N4&0Vm6$uF{OQ_?tWBjou+9wt|KK>|Emq zakj$4Iok+~ry3715{wBf+}l%TLw;cSGYD{R@$9<1x{@pr7`gzBvoZmdB+%xhL!6*Y zNm%>MT$A&E!dNSq-2J;?sG65RCICBwPkR&tz&_@mX*EuS+=Q?SVF`WKV*CT>}%<2zsBq@)CeaiK&=EE;ac5EG+cw^ENGvy@eQBpgWf3UgJul ze?B_BK;P^B*E6(2cPvSo9%bV;*7r(s+9a@SLZy4@5UsK&pt-XjcV8RzA$khD?R&v~ z@LMUCMNQP*P{Vx!mZWqsQ0Kv+hx0VdM1-K=GDKN@_&%-bI~YF)E#>iTn2h9QwXLx; z|F;V%7P2bxzg=wn+E{yTm0&-zii2cmZ~Z3qB=vyuX)8fpCAvXSQZ^=#u$8rN{I$*q zxq^Bv50y8pc4y~W?D*x=RKey9AmNX+c=^^n!px`$m?4~(G${QSF-aM@>& zuvG49s&q(K?e2P`_Cl2syx~FID_qQ6NE>wb5wL&>Kag0P^{=VUzY7WXdrHBjmmQLf zH+~rO_fz26H(z#ldz144#z8)%t*k=n&j$3+C0Ia5NazSR>BnyO@6nBpyTYvUTbi7) zRQ|U|-yu)J!^QiKVK_-IPCn*Zf?;02x(hN2g~yzw&Sh&Nyq$mig&q0IW$eR3eyQPf zK(bjEr$kGtM8{5MruI6~?hYipInr2_lJfDW@BM4q;sY6#5IRgf5YN?eH#F&=fu^We zTtu&AD=E1|_p&ufJQ8=}q32;;B}(h+w%fCqrkl?V_1vpZ`Pyf}ATkvR%Zy1;7{mb* z2P#!Be*yZ??*A~YR@!WJrJ#yu>sf_*JQvqVc@5Z%rGD@&Qo@)PQ5Ao>tEAZ2TUTo+ zuxKmG`)uYAP3`u5Hg)ZO)yr+XGvDTBmhr3BccKqw8>_-5M9%(`1H7dU#@D(pYvL4ylaQS?^rnT9zun+nO}R)htsEYR&5v@Jxv&Huiuuj ze=uLAZE-lE_FjcBkl_Et-xkwl`YS0K`I||?UY6(sdH>!C@7@|)(s)n1OKa?S_v74X zX4$Yg(Vbfus8XQ~FVyNPMO=OI_mk_Mchj;gwA<8MG@V~W+Fa=oBHGR+y-=*;(yzR@eH(%^NU}f z%xbA~>DtH&*A!mg8mpPTsK%}5t|#hepLHym>QcGwe!(8vI6giuYoQ+K=Pp>;wun>6 z`}PomT4rQDS}vNC>OR75-eMiq6a^o-)f$|h!Z*F&R?E9!v3H=CnsELs1!4;y(1i<_ zz5ELkD*KVM4j2E4@$!y~13%GH#QU1$xuoyj*v1#-s_%$h8ie3o_u>=YtYa}Agt-SY zmFF9E%L1ws{r_s6#YRQl1|O{a6CF)cqMELe$}+dclI)-=Hy*ZaR>&uu*_eQLG2VPu zTqhP{7D#-Cbl(4e0VqTIMj^=Z4_ygwwBK8Y;|E|=@uB|#7 z5|nHT@c7jB?k}p=tP_0wyjcir3CjYeqgp<*_spGsQ@Dq?ltd?VJD4O2+^6XZlR_F! z{S8y-ZTEjW`5qN=(j+2{LsA;la=kSI|0agBsXJII0Ho&bZ_;Q_u( zGBfDq=D)|YpL%UUA8rc|(8ozgR~s5ge^Zq`^vZ%Hvp$?}d}U`-GF2gR*Xvo<&%3f#5~6=EIJp-+TjwxJfW~`h$nHBW720$H+yx%t^uaU_s0>-O@hND(5eUggG*q4{VY6 z!hk8bD)p+&y?2nt1X)?tKU~SVD2-RbHF19rRW|f`Cg9SwA|`S;-;f?e8oaSS)wH(z z^ZOIe`h;_O4Zw;7A5SqCHlG3eO7u!G_bH8ZL#`}*J$cf|??VlHYSEw)o1Cod)ag|#8)JfAZL za1$HXs<%-wJw|fb+cd|sUj4jJ%eQ+v_S>lU3b?52{o2&gz#m%u{^j$= zSfeHO;1WzIB_c}da*l4`??#j1S7*f9|CorHc-{55Vu=|!Mb1IT&Vi=Cq-5Qh^YPWo z^1IC=7;sVJ3lKSpP3PON_k4|;DgRl<1DqBc@SH^-0(h(^|BU45RNAR?Hs5>Od!Yt! zH}<*)+ZGZ~bE(~w&?z!>cEGGQDCz;vvK)15;ah!9?cMTHlL|>|jTY7xPDP{xmr0Xm z{gP+-FF)MZwZbR06;%kN-CM{H#As*8%l;++oI^}TKQtA7zj(>Qe+&!(4TSv@!1AAI zvf?FGM)?(=mT>}UD`A>$?|Pw6nb(!4bbF}%s&BcKRtg?4Q#kc^(pIdk9N4@ctjrHo zS%_TF7>D4a8ZkJ{Tjnr$KrWlq^dN5L&L^Z#Dc6E zG7u+;|Fk%4QIdWYh}Zo5WL9~Pqd@`!A6`zxdiP6?=gI9`hkM(*J}CT`y{km#!I8u{ ztTMo3as4&+m2PaW>=)7^Z`+t8{mHYgNnjs((X>8l={4@oU7gEcO<8@ng47+p1Rns4 zt*Xu(_gi(j5ON;LvCA!XXaiTB96sL85s3@JXo}rw?WE>Z0ik-nvAEF~Us;)Y=Slp_ zM#*#Od>Sv}mgyl~x9>#tjb&>J)h%77aYDNsu{S@JT0!)w6M4X934t0DwVQugq+)00 zFAUz1^jI(v9JBqq(844M)u;1iPj6*03c0R5V>|Bn($*+Rv3g13qNFB6`QT9G->^`(~G2C_h;ujOHeu>7>Q2uRgLPC z9mLhAqExe4_{vY0p)~ry)5*8nc&g`$+@))9H z@Dt{$?_B0x)7wLMhhX3m*MRtA;s#HYN#YH#Uxv_l4>%6Qzqq#%_Z6_Aj!35SjUxSf z4DVbbxeel;xk=h@;^p_@5vDdXo|@Bojknw z^{3rh+mMwAEpnC<2=JJc8N_+09NT$75)U_I_>{u>Qo|ru-ha8mgOX;yJH@rKGJHnJ`L+?6*%+ULttYgyvN1Ux2hkUX2nW)||1ja6>Bh9f`)idg z)FI+R!filb*XMCq4@*Wjq~at1u?-qd1@1t*KYhnKO_w#HC0NDma#KekN8RYd?>T#j zh@jvQ3jJwt=&g!6UUdQ<1=}H{f+sfyGUU@ODI0`>s(SU~g&HX>xcW+$v+WajGefEF zZM+zEu;hFvPV}vb-({!FZ&!7m9@0EX`7(F|y(;(mK~#1l{X)d`kmy42fP7a`mEYKK z#47GIfc$EQbcUz>69Mhff}FF5%&NnKdlnvnk%Dn9PuY=mZ;xah#8u;&KfOw$|wBFzDB`8~wb@Ux~G zHuF_VSIh^Yxp5+>*gc$0D5QEE6emq0o?eBvN+;bm*T|2&Mo!Un`Lea!o8lz}r>m4G z_LKO(nil)1yD8;s@V@UqxerfmOR9+pXkQq*fLFukRs}5_o6o_Q0h|V%X7hW6Pb1=& zd0sSUj|_|7%CkI=Be^kdmw<~3&B-_Z^dbq(sSkpoRf_EsZaxD_vd!@gl5fB)#Y1`- zvQ+w0_sC3Z{-pTf`6pY$mEWGTB@OpE-bAjLVoLw$As@KZ!P zW{QSHFo6PwUV8lp=V}7I)Mn;j^pA56>v%e0u<7KAt|W(0<@sq{_mA57iGw(v07dlE zCBF@5qx%j9=svw7RKg$yo|-+WRef|DAv#xn__JZuY?aeo_e;>J!s1>Q+hHHeM)SnN z3``qsLVQ}EZu3ebW9i&yBwjlkBRW+j15H)0N5h*?J;N2^|1 z#jCH_8Ok6vRFCEu(mhhhq9>G*;cRt2?<}2Sa4W@Rc-CNzG^npy z01Qny!Unm%@o(TVxa89JYP+03V%6`@MLneP*Xxjaqb5i?-DYT^3SwD8k*f2Mg&N5%sVd+U*U(ax_9d5U zosiX(+DcvDkkwcnJ%DLQ?=r;`%`mGFy4;IP09j{asN_5d(q)7Qiv7aX4+3`IUEc`r z5Kjnt2tYG`Rv6a9Wdjz9%OWR(!VUkxVh?IXSxn>oLsU}yeo(*eI}M#cWiMIAIlUi- z0BoBF7uMmvwOmlKw?UvOUrKrW9h*RNI>(}(4)hvft5=mv^*ar52O%own9pymKx>@m z0h7}))EFDViCnO(q+LF zixKdlvX#&}sU;;+YMAv{A{W$~j(`$?(fKof1J`jAmL6=S-q*c zp*z~=V0@>b1b;v<^p&K1fYWLdXT6Uw{sHPE_`I&kp_SI(FmJ9&JgTiRSP+%iSh$KC zu(cP~LoOaOx4feWc(0lY*~fgR25j=2OSg%h;Ya|HK|oS{dZr@M?X5=GdWc~AGPJ&Y zJ6?inn808CsB@XUX930-C+=iR$NzlLA+mp@*aWH}*uw>)pvP&tHPr9vk5~zCeZU2-OP&jxLf*qO_8ACYhA}7a2 zkfExDVn=0b3LLFvQIohKTfx+pw0vsE-!jWXq>dLimG1w_1AKh%%kx1g#k>uRf?J4}nZ-;u2dHpN4XLG=6y9$)wF!HCzSw0A)xlxmEg6W!>1B=z3PG@6q zyF0|A(+BRDk6k4ha=cpYf%Z_MZ?=(z;1~5*1v1+Rn*&FnxwRc&ANd^!>-*psz%i42=drXJy`)Y9jx6)k-@8b=XUe)weB-8+|7?KIN|>9g6Qr$bQ>w1Z&ZCV+Eb!~ zR;}gYn1yQSjoi&Z3Q%QSb^-0$~b+;Yy4Jqkb3d? zKT(#A={jD${uI5>hseSsm3pHKI@Sl?$U({S#^l_r)-1gQd8Cu^yn#l((SJpNoh(5? zCTOteeZG)sBY5BMK6>?0x|}>B3JAmWoO*$~&9>Mk?vw^VxYy$WaAN~6pP~>C`I;wR zc*UffoT&p~HW0B12%|r95q$DbW7_4u+TCWsm0VzH!0|g^*p7_f^(y}I)|)-FwJ#gf z*u|K_Z?Sz&)SR_AVlmsc@EH7>3^-dm_)w}Vu1v8G5AVz;nib$Ge*r_Sew`lIuginz zTeaHATB7Tz5}djN$HEp^*pt3XO*r|wvTQ%M-3G#}k2ccO(HZxsr{TAtg&@gr1CM*? zhNuDIvtmKG!Yzc^v!7d|btxb`l}bLnQboO#O3nTPUbD$Z5j$RN{bEtK%;nYfyRvx? z-m;y`n}Sh3uEAsPiQh!2ym9(>ufu7L3H~`f5wOA}oT&87s(0*JAMH~6vPIMAO3#7+ z{0BOw#6?x)f^e;CY_CXO86pJq)30|(;(EtT95p@#!BeWlXY!~42{JGtyFW1^sb4_Z zc4Y8KbevB=E*(e{`Shw@zQoEbhMKJ@%;Vog`54M`^rc-~jm-eg0Rw5yGTI21z~(&Vrkm@9;|oVw<?8 zSGn2d7U3n9LJGUhuM3v`!2Fn#*Ob4E>aPIf?EKkdfN#6c6ShTZf&x?vr~ds7%T!4w zN9hA4B|c}c^~yA?jEW z$%vBd5JJZ+*?W&{QCZpJh>Qr)J0s&z*~%`mM>4WEk=Zehd5(3Qb6vmJ=lAdL_r2Y| zx0`=BIM=zZ*LpsmkH`J#0^aBe#dkdtX5t+Cf5+;W$tVv+YE8tU z@@2x(Z}0Z?_BpVAdjwWZMx{s2kC1K0js-PzSYDlh`&Okn;GR)FkD7MmRqJ(S#S>QL z;RlF`nE&at!Ov-a{6%Xb76Q=pSLQy1E><*DE`!oY{*gXQsDhOvR#RR_b>OsK9Vg8_!_r=voSFA+J&KgtK3PolckbLJJNTY1@LuxC1}_P_5vy{t8saa^2Y5mGBKS52Vsd zMITw|aM`nQ5qBc)HJWQPgAF!*#?gkre2O)H2Yy%JcK=L5=;A{Ef45#9qK$klPfwTJ z6i;II#e<90O@!YfLS4An*B&6&)RMEqWi1hPl|mI8VYVCjs~i^IEKm>@mo~LV9Vi+4 zn9PX-rs3%y%+oe1|EU1la8V#;Oc%&c+eTt7WrtPwU!pw*$Sq<@VIppuhd-Ixg&7Vl zaUbS@YYRS#Mx9{f8*V;?wYJ*zUid3*<)x*&!%i}OOBZDT$zfG!^Trv6*`$J!`3^u80q&SRWgr6R+QCvl?x;ic&jH%kBJkwr6 zOI+N-B%>~sdOz<*h&RHAMA;~m6!$;HZED$~34FYI^`4$p`&}$5QDKa;8{wLvo&R&S z6L6$nhDU4@{Jj^eN26T%!`4)u5Xi#Mk_?fGy^xvwOQ|m8r@?RdpGZrDiRR6ue-sF& z2w7`zVYgSL^0^n!e_@1qVTRCAtd zvX}2uQ2y9sa6W0fan7AlB@V^q|M3a#Pk6-X-b-pX#){rN%CA0dIm#9sfe#X&;`MdQ zF@iwOl`lr-yIRObZxXtgBoX(CTlcS>&muSb%-+kF?AT{d?rU2_6a}nM-#Y$Htwxm8 z+dQiwZ=H%6ns&*>hM82O*i#f~O8deqA`SNg*{ZD(GfURym zcb-C?agz4^y}{~8RNxR6^+Yi+jXOzLF?~JukI5OC%M$l?IOrx_RoM`-Z3*wXJ{8(Q zGxCD2QVN=0p!OrBtoEn*4R1pvanEqkCo8!fwa&g_q8WguK6E$iB{zMEUfCKB5-UHWl&fw>g%tSBuIDe(l=T2D zekrBgETTx1$8~HWwm9|!dwe}$zbG(Ke4>Uc!j;CO;OmHWp#}GUr>#rI!F%U($+wVJ zYuJMQA?+X~e5b?-^&@s~;BhlZXaPHj{^%ZRF*NXvPzidGyNK`0_j1{B2afxdZLeIW zmSf9L+Vm>U7VkDMUr}W7iW3@ad2g!!fZffmuGqa`Yz0N}9Fv;ApDg@f{&ig~mu6*J z#I*r<1J|70?GLiXDyC0XsHfA%q=6zC5tTBS{%p`OWq;#v@U~1W-><97p)t`cy0-~S z*>03)3a};)6=^~e@PfCz4^RsUYuR^#Z&|o~7z^K3xrn$qRT(av!|6NyfW#Y_5*R$i z{>LX6-L>{;t6i%@U!; zZiYPt`_Q-0K~Wo+k%m-5hZHU39<0`P0+0Lel;c}%#go;LS|8g}vLsG86o_5GhFEK`ymNnyK6t4L|wYOQW5!~3#nXrY8_*%U?cOj znk0si{~kM6my2Q%p*U^@jHwQOWi@adfK(FBvHxk@SqaI=J6!Zp{Zk7XZP#rbAwQqA zDL*z*JzL_Wy0>Zo_Qe3+wBX{XtCVwY$n2Tfr@d1McHVddJ!a?f+D8`d=?>9BLEv7x zoNn?V_SQD_ErG8A*6X!tCq>4^D0qvG#PM&h^RBbOB&|RtQU1mQwvw7H*lcf@d24MDvAM!0h?s>sr02ma^-H8`RSd+FEj|&y0zfQ=d0x zK@84T7r!`XD#XoFosspqgARk_ArSxABOH2-xaFcgal%W{x@G~>0?E!P@_U+*uro+L z|KlgU{oX5B$3x^3+ZHB}O_KlnF`OD6fty=vN8N<-xi5*o;sZu9b~1!kxk(*&UK^{Z z)VS(4BM-mp*vRGAsQW1)3tGf6ll}CeYPc?nRmCD)w*D^Csesy(|3sjOt1pXM*}&y4 z{E+9|;o>nhZ;lBe_C43;2jW}bDd4qHpI;8gcjkZcX96cPXQ%7ys&mEE>8gl_x&pAp z@TDBqU`|K4|8+z4ZEzmdxf2DBW&c^G@i9MxRiPo>OF+cQ(=HMo#Bir}6Q%e~f6!JT zn=7AdQ>NjGTnP9Jk;uF{?xb^S)PuPPo9+F+AJ(Dt#f~cr?;Ad!**Q!v-==H>mjui*5sgGIbPjTL8s>J_a_h>~$KI>-pgml?6Q#}76+{Jv;=$Ctu_b{Z zFYbt9J=9NM69=ufUwzl((kG(DI`nHCw8k%p+(K;aHFiTj4{2%%x(geEjtvT zy_Fx+Gs16-v2O=8uCOC9x+tM1$koK#2)ZM7wzMP%62|FPTGBJHeY!dfvv3`txjBDR z8~;ctg?lJU`~2%CRo^S6ap(Qcdu{UNeZeO#`Om<(WmunZ(Us&XBz{j1+4fXXmh(Jp zZr*!Y3!_ZqV;{e6@rdSXPF21*k--$wf)P}Vx zFV4wd%!u_f-n(L!y%LgRe@n)2XQAPVi<4$LVpc(~s_78_>X;mtkDhOxrY+`%# z1OvHaWu7P++lEJwo@$*+bqh==5Su5~O~_R0rXP9?5Rz}mQ>6lhOjw;Z=Qeeq$pjc< z2Y0>&EI7a(vCLbvufQPnvY4WWF>id)avs?vp+Hhccj9)LsGPOhj3wgo`Lj!ZA+x3< z3hU$CCVm4yjG12^$cnWTV2Oc6)Ui`OuxQ02{gYb*UZ5B??c{&4_Qf}xCsz4f?(ezGoCwxIZ34-mO#gM)JV-5t z>4|h2NiR!aC(~CtkZ(b^W@pv_Y!DvNg=Do|F&+)}nrRwLkyi9wy(}wvDZT%ILQR;> z^$Q%V*^=N1_k*d1eAoxDz!)RbzTC+Z>UGQ%{Np~dZ9)^RjBIPh-KSrt)*!Jx4feYN z)%@|b4Kkl0RQ!&A&J88MAsWyfpeg<-ui`sX^8G-lO-vW!czpqWwP0GMpv^VuW@0cw zsyLo;yh{?ZxY$pPr}zxfy$-GE7DBEfzfshD3?tZa5y+Bg&DIj1LXw z*B))QoAurPvZsd@Ql*x?Fd6j@ZbCM~mo?zS?MBg@SilLegbN6E$)@K{hg84OMS@%y zoy5ag**pLc81%^pQjcU5L8BM}NsuZeyt@Tyz2`;p3i|fOI2d`MI>+a zBTmEEHO|N;lav87?all5);QFUb;G9KPV)22bK;0CsM^cd`z}F@%*r}@2ENUgfZNTF z;|8_}y*t%AsAnk~-GreAQp&yaLf^=?kgQCy^0nNL{ib)TW>q2IHibB>f-n-k; z)#Ri@WvClnBM8^cuxHgoucM*Gp2kTA*j$$TcO9r{J$hYzH

)O)>DwpU|3^Msv7I zTm&@vv4Z`6FqP3hp9iw{f3w<1&Y!3Uu9v>xlscFG%{mgjgt5KpaA^L$*tdM-BKoBt|cAW#za^;s3r;g1A4Cx-Mn|m6&}=A7Q)W9wx*C2cevrZVAH?wPT3C3$aLi{-D;FvHXwV1jOX7QQ zsC`-}cbn+wXCZk<;rZhUfsq{U!ZW)ije}0)qy5%!l6y-1hPsv>EZC6_~7EhMR@cjyBto*zTvOeL>(1u>@^r1 z>QmcDCGmO`rPFLDPOIgdXy?MkI?T`Vy8jodlz8syWrOgZ$0Q8xT5L@pqY28vs{oy+4 z5EJE`uFCslwZ!de{OQR8A^Dt|S)M{%Phb|Q!eV=s{;seL{BQuZ9h*&X~9dsSmDwHu$ma`+GK&xv@Y z*KBKH5*|;bNrUyPf_(L5$HVNbqoqHx%Mh_h^(j?Y{V4 z9^y{6aIX%iAK*Iy$tTtU)mbCfk4yo-sVU(=oYLR1i`*-*!ZHKGnMIS9$|n6EV)Xju z3KRtpsYlJ-2 zO@-EHX^OV`I>vUYjd!*;ej-0@8#=GAO1cnuxvcQpQ5<8xmOE0c||TuOM4 ziBkxQ(Ek1XIiAzW*79?^RR% z$_u$C1t|jb7T!*_0+akSuw;tw*%5f{0!L|}r3K>+;KfU+@E<2Asw#iUKuF{&QNMwZf)aBxq(JMcez=3|~QyY~2 zrm1KBuf3SBJ=@`KSxgLJ4}B48(EFY!4Hva?x{+_T_P|w@2}&AgOFdnu(n|2wJ5E9X zJBU+&Z7qt=1DUPSwNI8aMXI>I^u5uLxQUzz@41O+PHW>@Mo&Q}5wh+G+_0;8{coep zfABbY#?gO2fq3=8lcAO%20^^)LZkU(#Ocr(D=n5667o}6M7^0~dNp7Sg?bMO9kPx{EcXm{t1vqHKKEVR!0a7^K3=pLfh z%4nwM4b<#@Yr%obEys2w6QpVniWA6F!R^Q|mjZ!{*^9_UZUXOAnxGnlYg>q40RFUQ zXP(X0^ALL?01n&{Gv$GT_zu24I8fP8D6j0T+=!KIeXb*v0FS9Af_sOfShcrd;#5aw zH*#RMViiIwL8R(NN(Y(@Qb?9h&km|&Cuv_*l=)&5kVLgUAWaE;xn_%Oi;YA5$GwOT z#qkvCQud&}{?$Eq{d9<16uZ1fAZyu4kG*^g5oT(2fTmI|5=VIS(k+1CJswbXD__7{ z^lxuY6%&S_iU%g-`$Gt{RKraud88GN@<;~5g!gGVou>hO|Et3-Z7$O(Ak4*C1%0lD zO{D(xgYJ@pwwAE%8tvpWDR}0!xhL18!ZP}=jJBtQ1{cHOuGS78WDlslY`2ZzmL0{3 z-s=FXqWu&2lz|rm+>o6oV;mMI4`+BVz@GM;bQV@>qfR=@aQIVJ3z01Hf|ViU%TweG zLF`qALOBmV>pHxGiTz{l*Vq(f?&Ek9an0N`4^BpNAM~70tKjBQ0R)pPR&+=3{#fgf zLCE_Zlr*ltv+pnbYc*f&fA*&18nG8I0!eSFnQ2B4ayQ90rWUC)8lf89XMR+x`Isfx z@dR0R)$%bc_qMG?SA+Q|CcVUF@*|GlfD%|a(#X^b_&tJ zZsj=M`u!rc)3ZYEgJ-{H8R1*dP`W4$WI(E5Q%!jQqwQUrg4Vwq|1LqrI^*q>R3Zp^ z$EJ9S&~VK1;DPlz?7NnJhTEg~PktH+2}Y+VD*3n0prAe{6m*#+p-g0ZXSbcm*^1m3(tuMKR!)}_9Q;XStzQjHK_0C@Lq zzL#SQ&{V?}r&D;`{lYuyVB5mf$=1KO=j6}qlGDuQ&T6SMV}Ghie*)>Tv-1bTNd0*6 zfBCK9uN7TZsA;TQ7?DnFrtZj9cyF+l+lTUx5B?*pSHw%=%$Nl72c)(5$wP@ADfm=g zS*IMvl9HJsoiX+{)DUe(tgWO}9~;~=*`YuK-Zv+_r;|nWGvwjLv8JN~9DyrcRx(@E zw`3j=jlSfALU}5fi7f(3T!eT{jnNa=#NUG4#Fd_W2M`tk0lS`d&(8FiDzoRm|@ZQv;Ki}o-?ipBFQNaSAYaGT4cqIWQMuY#SAyqw_bk= zoW9RJ6r7;>Ol-(p)U&sS9{`s1JMN%P~bBDS(e zhLYuffS=Coi^}QLt11*Rd7f9;E~o~7y}H62xnm02@Oqr@oBgD!^pJt($r!it@F(B6 z1(L>)^VO4>yj=^`kuxf`qe{s4(6E@NOLiDv=`76SX=)LT+6i5Mc6#xv45cwp0LMAf ztaH-sT+k?08S7LZ*!G}r=iu5jf>Y$jKBGt5YAzWYofQ4^{aeP7a#QhpobqiQ)Q_(; zJcRfR)4YwPhDjesDMjn);(TXqlqn)$@nly@v1$2W64mjE;%@Vuuq@j(6Ibno1+(p4s^Ax(#c%PH;GD+%XuL`Zsd^d`2r3 zBIVBPXe2&X@YFXSqW$|kJCpH}-CDsdV%-9iieQ$?6mBBFn9@_G1*;A}EaotHKB^~e z;a$5i6i1L#8BSc!>AEMQs-#C-dqAs@8`&yXFJk3yRanwQH<{O zn!TDGvXtf*OH4%g(l%<~`I6G{Gb~ESk2mTw+Bo2rHVr{GlJi=mEt6Z;jbP7K{g%y( zaV?ADyz8u~R5TO(P=n1xde|NiBZ#D_PPYLk$96yogFfg%#C^dr_(Z}?`dP3!>y_+i zRLpH%PsB3%T`TZ*qD&(zHtgFVHkCWI>q0cW3k-O%+vY8Wu`=fHzBI|6hAif@U}O;NW5$|Co_e3EXb`l-2NPMnFT;Pcl>oM!XYSMMr@ z>JDF0-|f^jmOA9tA&eBp#^Y;deiSkcM?ZV)n#I+!Odp+fBGB}ThkY3x!fmoMG{r4R(>C7oY_&`KD60&Vz-i^~oK}w=m#?Z;HKS)8ZBjizIor&bwTfy*#k+K^2 zw*e-o=r}jRm=lZDZw}ZxjjB(E(ndt_0{mMCnnG6Z4as;mK320Uph_LaH#kTJF^2Bx ziH>pe5rSnnO2kI+4Y_P^d8%1LA#&-v_NxKHP}AIQUFbeSg?>Yh#`})rw97Ec8#?x6VoL$ z2lY#L1$p3>W{-%sjzVPz3Po(#N4*jrL-SgWJ=;ru^Lz4a(`*JNzw7HHLby%+<}vp` z#(iqu`w<(y&&Rl5UZ0L##il6559ox)WF1 z)~4VKuTb&8z3g1O`%(;2j@v*{wHv+N(7T_{R)+8CLSEgXftGpz=Yx;UiD2}JO$RMQ z#|0EQXPgbL%jnuc%i{7=mdAT@mF*m$khHFu)6esocqnbnlOSCDXbPxOi%*e* zx10&Wz7GA_ApqC%X2ZTCA+(Mjeh9s(m`o1~xO-sD@n6+nxgUlr)m%TW-w-7FLY@MK z4c1r3ToDXvV;GdVdcxE?OKSM9+3){_XP*=~4Y++~hNXg3w^A>{?8VCUhTf9d7vS-( zNn<*Lr|;ZHLPVy0XZXenhm$9@=zRP)aMGn1B^?>aHO@}NF#kcMBzGZ z9gPxSgC2h#GwawxIYHmomc8E;4pj>u8XTfkj|+vM(Tzru6tkQu|B z;%T{$Dq@74cY+{o7MgnK&gh1S2+)#!zup{!B4K)1zZgsbI$ZH59YjBW!x`3(yo$Zx z{K^!3k5TVK24I6hDK~@w1l>%b0egD_4;eyJ8WYVluTZx=j=7N|jgna6snzzO=_xBC zum8nNS(8U?BSegNS4xw2?g_?@)xiYb8ktEFJdE$LgHK9Sqr=zMQ9!Lb^=?*Sb_g$^ zV7r%mXGh~o4$sLi74rXKF#YUNLr6!sOkEr+Sc&=ouZl!f{AVntQ?ae=PrC6^8{ImV zL}h$L3b~SFB%UpDN}V>szi5TwhZKyIwo0*|i;) z!{Z^ngeS_s7Fh;RTwK~CQ@i;6cli?(piqZlToNO94}U zSyIWO+EEG?pnr4_0a{<|e2)JjjAtZD24y4R)6<923jpCqQ&OJr#q{HGDUxnsZv~0D z6E>O8_K8Zi1~#et+T;)ekJ%*uhRm)i-*NK1GLl|?-}()RC587sqBm+THD8gucH|wu zaJ5&cP<@v#W;3O3-R8l~Au8oU*!jJ{+n{$RMfdi;_KV>oiD>8KKIZGN>!19zNSWFV z+>1n+V14`zjhL^~2!OTFTM(Qb5#fhQDE9EVeE947>t_Gm2@Bsp_QdymfH}#szD_^D zJKM7c)aP^|4`13@H>ir4!xr<8CS61bKTZr;;nP2M!A_u>`$o+STgN?$f9Q2h!mw}( zkGbqjcqfGa7&D3rYkIB?0vp#vr`io)ooUI};6iG6*jt`*OAA0`EsH(!y|5tF*(2(2 zVIXdGV4EDn2mNdtE+4KCD$P%`f=&GHJpFLrljAxf?hw6JN2!dwmwbr6_P=6ct8mK( zG)2gK23pxYExtptaNN_YK9CYMO+3NquvrcQHc0(qn$0Q_$y2Y6Lu3Anl?dbdOV1UJ zO5!%ZdGZj4vMqi3WDHPwfaZflbtWQQ!S8q@$-N6)chRq z+UQ>mZ*QIybre22_I?<#3}-8tl4}@O{MbOrjYPTnIR?|b;4vS~XIs+1T+itqPAFjg zc77ctozXQf)q>DLi}3rfjpf@_C+zR#0$j&7C6hPfW=jsR2=cJ;J`28;ppT@!c>!v$ zKE~U#F;T(e@*z))T0(5#6h1Z_Ui*JlMasB%IP{}6ldV6icNf@R%5@NkyYDVqMTh(S z#pj>rX(=4?$^fiXOKa~d!4k%ghse!#hQ-VqS<47RT#Fh~sTOd9mqHOFl)fRk{HG5e zG(0ovCFvy&W^J~H3vCLi!Z6B7dSxC_xRD>hX%Qj?HBA3~*+Abcdj_s#9!~YBKML(7 z3{lY`eYP8x*O2n%<{z(tD`cLU6aIROWVS(2uKwN`jwYht9ef6|#-L& zX8$j^u87ZGZDh6*xh5?YX3IOZ*YGrBw9+`W8Gv=F#ytppY#Ttr#Gf^3Rx!`y+nBTG z`S?W%fx?;b0P|GvWX=O;nx9rD&KK^B)U z!?#H3<Yz<+ukB^8j6kqHBtBf;n{{qrB!r2|D))h`EgU=OrQ z!`Zs^Zx*PwM!xN+c~Ugdxpf<=8LM*Lo|E88HOp6R&GyCCl2d_Ai8VJpF2zL({}yIupKzA-dP}1NL;y=SmHyOxwN4cf!VA%o zX(K#aE3_mI&?D=Y)fa$ShCZ3>Hz{D zQl9{$9FJV~B~Wwsr(+N1V39E|3Ak>&u72o>KiXjnHn>TPdTNA;8_%b}mJ^aMn!py~ z+O|~_#QU^9!Q4iovVO9<*L?2%3CiT+5>B0ca1?Nc-M=2Vb^rbJXS^NV!$gjayOo)y1CSGMHw$k-X zV6d{w#HTb;@`;Taar|W)lmCDelv1HR*4Y?eB=}wMq5He#%GgUMV^@Q#9-{8;Cns;@ zKeb37|6~>u6p4yd?fCQLPPB@~$pqn8{BvsatfSu3)aK{qoCXeAhrV+wx8+B2!?>;m zk6hDW$k>~&*TZH$musPeQU;z|U-n%8xi!%6J3RjciSsA7%NjNka(mAHS@h(mCM4}< zm;e~3JbW5kf1{TE39*LhsfCuR>?!NG8h`QN{^OSaFanw?y>x`W^P=ao>9rEL&z>Os zU6XXCDd+HXjm!YX4DUH|HXnWac z5%Bil6&Q*|6^uO2C7NBPJL1-|2$TK-E`XSeFLi^Y8B&|~W8)R{`is~N~ z^6EVQ{(HkzkNX|`FsNt|&EXR$Q#h1Why1jIQg2iB-z}8Oi4`1E#Z;+=Xay0fT|lx- zmw#XCs1tj4jpAZS$2X<7dwaSR@xgXs7a!*a(NI;9&1Z^N0B~3uhpZu zKt+slReY=zR01wG#27u=OSU$8Qs&&u7*@ggiF-}@9$xD!@E?E_iw~amsI{AB)_(mb zo1U9wj|ZZ;Qkm$RkK8k`B^*D6^bW%B|9lY>KJiBFo<}(-+H3ri8p$t8e9X+lG3J_S zO&RA3Txf2!LUPgIW}&|IU1#zAz6^&F-*zU$6#a|bdDIgufNV$poLsFFX4FuL&>;F+ ziRC#KgF}L(jq-SYQP@lSS0Ws++PPJA=gj+7f!SJ@_J^D;bi#Y0;CBOH<1!l2UQf5~ zyxG)_q+tjqGrO}tBq%USzR~*h03FNy=%qv$w50n#qS{NG`UVF611Uk}}RK++hGK_yx6>DL?*DizaSZ}?z zT82U-dFzc4j=Mj=Ac86S;i(p70H`Tm%%ln-&F&^2mMtWzUlwFCirxSVB+8l^F8c3a zX#W81Y?uMDVFg9oky-am3XvCyDy=WpH(ZnD_fE&gTl-n6E#T8-c6JFr9vxJg?}rh* zYwmBJ_ZOIyA)E)$h_q53fp81n*irgq)(fAudj!ks&3}}^8=sdViA}dQF#H-GQ+=cJ z2*jG18vH>Z`EVg}0ql%J^^a9rbxY)bUs5!wYVcwoQx!1u+5G!{1GF#-6^I>PCxSke zm&>Awn0=JjdtG}{Q4X_JRwt0?JSzz722VvU(KdMlVQuvq<#La4;=&GQ2L52{wXXhd zQoIn!7{6e64sOUEL2X-0`WoLnvs+6kC7~?)u7kb(J#Ms>P7@^4GkNNNwU46+H8j^$ zNK?O9Z)2_!CQRkDoTgAdN3B@t@m7RO;rk`n+|81)WEIBG3ddG;b|H^-Uyh;%whqE# zXVW~_)BogmBO^Z*wx7WCn18@Y5c%mguV3_~=D7cE#1f9b(lzH7*<6mse!kEwe#l)X z-NYJfRrkD+lD>7#3TDj6KSkQ_^;~y4L5gq=7Y=l_}E|T z2ZOAb^ZL)A$6=R9jdSa*9292OvSEXBa39_ON_C75L{#KCi{P8@VBb>EyiQ1uG<8Y6 zYEB%CBry_b$O&`{hKW~#`=E5Y^`BL}mF$GWh39Kk6ZTX(=zx7zY1krZp!&!S!Z~XS zUj#r_+u=(jR7{TB_AXpwH#5gciY8U`gC6QJxe)TGMP8xQofMpt$=j=>;Xr;xQ{!ul z4lvC-$be&KGMA=Rrqx8Q$@|wjk&0Gq~W0LzPXnPgGvg<`Kn|-Tea*Kl>8x27aV|Y3WU>k zA#dV6O2+NM0-3=<%j~!Qt%uS1kR!wa4E*($9s}l^$-k+jBE+=AE zLgUknT|eVQLEfD+ElqI zReR3>2eQn3AHa{gH>p+~nlrDrjU}>I!D;Y(y_hos}6f`pZ-QxK-1tzU|tRkTVAhwoHe7 z;yt@v72rIrcinalftgM{yT`DA z?dxV68dn(enjwG(V$8_H*hR98SqCQJD$+gbQ$C&{kJUTlnh`8*$hx}el26NNZdFD| z!^_j~hTu7GXD{Lk-`PRvVXx#GXn0XVyBGZz>k{)rrF8}!`rsI~dZlKo8YgyI`vE1g zBT^~g+0oFBOgPP~=|Te2n(lhA;hG@RTv* z@RWw{`M9v3;-v+T!cFh<34t7UK+8LYs}HcXg)LOksPk~0Pu@zZ17!`2`^v0!@8tJ$ zN?uFw{;eP4wH%KG(7+m%MS7NlP(7L-?jRp{F8WmGKUc4QzY`otwP9kJ zR$YJAJl5>8pg&>j_y~@k`5V>Odb5YvKk0&EbGmADf{9KWNa#kMndBsb`E==(S;5Ty zJJH&}NS9v{&#zQ#8IcKEVx{=17Kt8fjbnmI^%cx4vv^D6P{`#LSheQ#X^Ct)thQTr{&E?W?|mF8g}J=1&GmZX|7puD6Ac zK$iHY{#7x{&Cq*&*9+=#6*`t19zCL!Q$G$eNwjKX4n@sM6Ib}(Vhz6+J-(W1!_}H$ z4Zdked0zxGcb{*PXe9fV_6i#^dJ#R7$k#W0i{Jmo%v6WSYpUN#jA>HuU1WHBgel=^ zUTs;sb(LisZCSlZ!;g(_{zkO9&rkQ0t())4bCJJku!aQ~03NGrfecp>MlifM!^ZW4 z%=wu%+Q~{2Ykb2rB}`;>8=f|z^wh_a98B7;vV=@wSuUy`k!~@vOP$FRAIUd5k+?(D zB~FpgYWeGwU;)J!J|^>UL+4Dhc+JWq2|!iTZv1xsznR2EG@E?Kef6s2$kV2g`$R{3 z+K1O7K|#@Pc6yV15NePy&blTK=<53NjJC_@B(d+&5{(ujjcj1Roova2R zJyu~^52QKmg8T$GuWCYbPLHS$^j_;|zDAiYbDUTfI!_Z=V(}=vSu;MJh<>Vi^xkud zgw}VuY=vz$5mTAP0(!514-=vGV0o2y>&r>`Jlj*2+eeq;Q5+Xt;fJ@m$xY4IX(-xzo5UudFL)kS3wnN`3uyx&J}n;`AHx`(FR5Vv(MILTZPE8c`uar6 z7U)}#FU&mtuUxiCIdeth$nSa)1Ed#DeD}qZA!g&$>{XNL=JAWL)+s)&8D^FLe$=${ zHqGw*d~2M$&{)9{mOtty*Nli$F1>yhzSrYzkybn17?WL#CwhD@f53mHYyJ0)P(vn4 zF2(70l35qV_%QoBq;Tbf+=ii$`jPq-dbmja9E|y31=f-7RP5QPbkJV%A8(zbFNl!) zTTCzpv zR4jXNqg=M~XNS2zwe-w)cC2RsLvITJ$<`8rQp0}rs1#uAB&6Tw4J4fftU9>-vum>y zbphEaY7s8oi;%a&;o3mbE^BRPW$f#RbJaf8O!C`*>G2Ll<}2uU8quSc(CMf@yo9Mr zxTaB9fipiBlude6F7IduLpQXNICTez9$){2e7Gj&&K4zgfZDM!ng0a`6;V`3i(`4? zrn6jr9aQ^YkdmJyqEAd&-TcQ~N7dRLb-Cf9+HemduHr%$-7Hp&Q!x4fHA%}40)bl2 z2JMapJ1kz`?CWX_R%klLGWQGJ3{2)J&7n~-ilHad&L_y+wn}PqOKdIGmEMVB<})OD$d1>d^E3aN>1d~C{ze%ycIRM= zMSu*fzVFq`18ta5El_i6V7vY&F;WPPaQU_IoL?V{@4hCX?}Ca%-8Z$__;Z@RD3J)r z8e%m-1(x8<$g$Semyd-x1C|;%R*~E8c<~8_X%hpm&=CQ?)t<8ghRzmnvByC8BAnFwBOCOTa@v!%I>#D^^;~UW@ zuG+V%*DOYH6@EIxG|+~rK;uS_Y9MLQ`I_^#9HFxmGIO;%384=sg<3W}-9?RH13$Tc zbFW#ssNl$X$6@!vMRz8+?R{{XgMh%DxenEutapvzh{0H}mt64s`Q0Zqj;AOGLO?AG z9P|za)m0Zp(24kOYRahb`Cm}J6w5Y?J)tH`80oQ}?)0=Olfbs~o?w2JnM_Arrka4O zSfCBIq@^&KUQx7M%W_<})xSxzyAOjQlN_29T!p4jt+_Ji1ETK(eIsV{%)VUEFBfSp zOX^bC(a%OYg@iV728z5HZ+l!tBoWi(f})jbWDhW&%oDTy@$J70A;pC!$ODy$-Snq| z#1&HqPqsJH8xKx#vsn0dbAH6MRrqy`on1^@T0tzX;*P5*tqLb-Re5kU;kSL5)C%3! zt8rv=uu9y04iidVoXarzOLrV7UqO0>Iu(A7V* zDqV4O^k&%6#$OXS7VM`5E;fm;(khc6^Yd?lV)OSLC2Sn^@E8@``bAjI=jVd*@8`yF zkwj%XRTPSgP3dk$uTMMDP+~JTAiskPoTb`WA0|BF-3z3C8gQ%Xdn=6lp8DAx8e(Aj z{4v-Yx*RW+)cj8n!r2MD{7$q-IJ&p_LRi{gX1>ne$4T;GU6#kSshg1BrUtuba1-$p z9b5L=G|X6^qFrUgUXXQ^b*(?=F41+`c+-U*L)41UwuO=x(Fe3^|H*N6%nl01dfM*Z z5_{B9Y#Cq?(7LUp(&sln9J1nA0HUFXCTZ{PHU<@$k<}ot(uZ7;rw`=ihBhk9CG5Qw z#r4G}%dQEmL&?>klP6L(Mz%HZY(yfwYOK<3m9ghy2tzsONnjG;IwlUV;Ibp8lRP5+ z@!GMZsf`u+iop#2Sn)&@6By)A48~e^SBl;a|64P+@D3$$Pw%aTSsrzM%nQ%KXJ~6_$~C>_?RiW$uGlYg%Pq8_L22^9B0rZhS}Q!<#wbUxkp+ zDc3~SkhFZ7&?>Qv2X+liu}6UO<#Fi(@uATRIF?>e3}Y9vX5$#rDR7i#^yGTej0}i& zroQb}!#GRkR%olhLj>$Nid8v$`f&6uu`}eaj-|CYxh9U&Ay#*@D6=e*PWsptpxmA6 z+W{M#q|-$Q7}L0AlGbjPAPaNwfu=hc*qmIc2PJCv>0Z++GT=iB{ShjL%Nzr+;ZGjF z#8q%}T=Mx(2{z%u+P}<}2tPB8R;&#jH&C0te1?>(_=8nC6kSE?dAIMOm?9U^LrfRp zhp~!IZjj!ciN4h><3OzBucwoc8?GAm2qd)1nkN@8GNv~{V( znbx~|T}a=}UzCg7XsGNv7R=uIs5Y*_)^Iql$`g_{ZB#xxMGAgayjmr*$DV#85#{uV z%Wj$|lmR6Wx!YOpVR<=uynkn0vzl&yAbpC&Y|=rzvmll6E$^D#ntbk}GXQb#p~FZu01n8=e1rf3nhoOpx_>R0%90mT5cC=Re|HgY#7 za2087w5F3p*OI2iY9hMjN2WaisU+$Dt~GDQPJOyo>yEvdZn4brq`r&u}eMjEw|V5=?~5rFi&{ zq|-s`mYJE0Og&4Qgu}k6*4|Kbyr6VzIF(ps0WgB$-bXY>X3XRLK8c(nRa>S=9SSZo z_Oy;h&$O1%w*_6G8&t=h-&71);F*OVO5%pGL=w7p6&Rm1#F;R5@1U$C~W=V)#NEGJpmm z1NbF7k1d*bRL8rgYpt<+#V!{W0-D@;Q}P1;AEvG}9;z?yml8shP|21eOH?YP+@ew< zl_ZKuwrnX&2wl4pA=yGxStHpJvP_}urtJHWWvmUxHZynbdCvcN-aK!N&xaZ2-aF_1 z&UbwzXXj;>xAfK}??JOn&1SISi^8?_EYOpm5aQ<-V@WieO4_+dj!;-QmEHQ_&6C1I zLr*?9w5P;0%?YG>XT=YR?nz$Gz7ypzV%H`fbXqL+nSwtGJ$&<0mL_?ci?X)p=LIrn zUtjj%)%E(_5(E32a{t|EwJ6-3@2eE%p{zl?QmDP<;M3>M2L|`<7`P&}e9UCjR&6U7 ztS-tV0ypZcFNf-IdY^%*_0`}b8lPRm*;O`tZRi~v_BT`9NPLC3jjaUi$ro&t5>Ito zp{M$u|GN|P4DD{%bF4WZIEZfwLM`%M!q0oPuU`IHXieDb72){%dDYE}tzPh47FqD5 zb`@rfNaEouQT;bp!`d!iY!bw*oC$lL^O+UxDYoJb+l)9bhn*hWRX0}__*T>&cRTMj z5EF~BJK2fGWNjy^F#B-+^kLl-Ceva>zUQ-duKdfyYF@YUS9&T>TOXibYqvDv=LBU7 zdl_U!Rz-|9PA(vP_>+VYjYu;SjZ>5WJMO|ObL_#7f&SMkS%-3e{2m&$VA^Y$RMo2I z_e;wdnaO%_%{P6i{(CZ`R^rX3KIQcqEy!k#VY9O)_k3;D1UeTs_n;bU&X=R$6{z00 zcD0qXFDHH68qHpaQcT_hkAeRMydca@D8_)@Q&H8=sQLWn%GkliV81QFr-SQkiCCm< zy-kdPT3shaKyA4{X)R0vkNO9snu1R^*v)KSLbdPhT3slVgdw6~58ybM%Moo#Rjqh* zZv*8>8p_pObPVn+Vj8;H#P{`W`MXE2vL$U+<)WgK(D^?ZFKQ6))(7KxR=u@Nx4in* ze_J10q1nUxzQpg{0_UM#Fck28vbBLr0-vD(*O$IW`CZDp*esa`Tb?@!seNwZCsIms+<0AW7Z?YY(fv$5Fd2(B)D5H>-+R;YZ7uR4O<|mQN1-)yc z%EyXfYs>rAm%_M)yoY-)L-I$>X9vYQ4F$Qqxt|MtnV#N(Frjh&@=v~sZx-**w2>6z z$c~STue6e0J|ZX%jsCm^;<#+7zR*84`<;VH8%IQ~u$DEv+KA>Jc zYtTP88MNmjoLPsBi$(`*@$>A88d){u3JWxe6NdteT({aZ^oQ9{uJ=xPC=zx7Kl30Kc_>w zQ9psXXQ4bS?9-mgI;6hZSs)EKz;g`|OM9o&iVyKPmAS3)t3ZQeS@?b$6LGvC{%KqO z3^=vh+e7!xQVwo7go}-vLOVN8cvXxhZ0AAYoY|kXOc3SN=EWPA)=<~*f1r+6|6nFQ z6l|V{mDS53qDrn^Z>eTchlE0qk*r-!6LgKu%Bi0c+m^{ z;xS=%&@?6C<7J2q%bcGkU59u=-J;cO@@*-|Pkz(gdl*OT?lhSFx05ixC&3oo%ijRJ ze`)vB?MP&c#|p%PJJMI?E#832pa50#M)c`GO~b21_W34(V;FHb+$td#4YNw`hG(*T zzb)!v1AF<(Z+|=a8733Cxfw6mz!odhTc+B@B%vZbN1R?gus6Vxn!r<`&^f0<=r!H| zbQK(MH`D$5`*`-p}HWf_$eSN97sZi zJLYaJQuy$V|7?DyjUx#m&@=@HF|mDFYiSiQVN8pLy}%I@MHgZXhT@lc8TX8@CbLsV zc_@kOMxCF5knFDM=ns?x!RNMKl*XUC$QFb_$GL~IpDjwYcoihFNM1NUCv^qJJ&xDK zsCcO#z;|nGxNp=_K=!4p%!6XxJ{wl+%Yj#bt~~)axI&W`#i0$~4_I=ALbUx?eV6~$ zF`{Ak?#Ej88#!8V_ufqmiO++Unzb>=YMJe0GIp+eL$o+qvY&Mssk)X(Q5cv@vs zP|#dhJ03Xm3iw>zddRl7!!swk9bqp`cN`y?z)7Ey_i9X`#bq8R&j8Zt=>w{)E@qa2 zk>x57?m9HZApgCx?NkO8$=6H8U`|EO%WXd4j_^{(UAH*3pscp~^!E*a!$@n z?|cIW3>FDxz{N0l8r|#6l?I+(6in|2a<8=xY`C--li>b=ctAKlp|` z&R&r@#!4lac1x&&OL4s6RGftTp6gjadv<1mZp(zl1HE9#$d4U0<*;;d6@SK#>hrGI zwrllELF-QWIyO?32)*2|Xv2Gm2NndJKuHRqNT_%~zgt!B%%8GOxjoYcx z5Ny7+u5A)0`J{G^;-GKm$6EOOawnD<_;T&5-#-u&Q}p+KF(s!-^ikoiMpAblI#W%a?;$Y>(GbdDMJ2F#$oCEqM~n38pf%ra&NdupJuts&@dQ zEGfLk;^L1PdhY%>>!@YI!(9O39fhO@{t@mOX9{S2u3pD+H4Z@+?-)xiO4))4J8Hr` zif@U~%_q~Bz>eBuYz%6{Q2dmXiEWK?LjCIv;JdNWC{a#N^T?U_6vHXB@d} zd2(ut{4(;^TRgd-6XCb#VcmbrFbv{%Drjd zitb*D<(1O_j0qja(KfU?MfusqS(0XBYP%(c%lnD3rC}iIFY_R=AgjvZLiz%^T$M|P zb9IUlHS?~=3)i@=3qaU$F&Ot5f(d2mm6Yj9fRMBmJiR_w?F#2%b+)4 zZW+e8=~{TCfzX~P^eai$1$sW4rySt=6dDpNJ`R@;23fF#dzxYR^OPl5<20WSvpdAY z@W`Bf$Xzk^8Cme)c~^J8X@e0HxcUm+*@7s!pFZkO8Z#!2o^&I*1>0>_L*Z6^^Cr?cly&b=#q zvpUYCOyTvvo^QnKZ9c>_7G$9%JEYu^MY5dpqe~XoTG46V&UKd+L+;BTaMrA#byJWI zNS=(Gy>)h7#gNI48r$61k_jG3o;~*j7pjS={Dq`#HWRJ^>?|DZM4X+5`d!Coh}wfw zpIOM_pTii1B^2|6K`tJh-%13Kd_N!ZtQVB88F;^j4}LfDnpaT4xU)EIw5VZ$B}vUb zwLJ}ea`quFJIbz0cu*T}ZjCY#m_bvbz`bS4qC+%Ypr)G60Odnkm{9)pJ>jdM1FvVF zR6o45iyhTr4*%f5pcWSUd1hxlP`}^X(Q#du1%Bo_h5lWajFM35aZADLeF7H=?N)p^ zZ0AMcTC&E_y>hqyXz1xwh6{t_nSEs&gET;GY)@m5k~g1q#A!`$0+?An6@ox1?qmwE z&b%l-kJ3wW(ikfA0j_bRLFO)mFG3PG0Mg$uzmCRRbi!X7fh^o->Shis80PdM=t0`s zF8mLw2$F9Wkc+;wq1?uF$xwGiyxG8TPe&Wvgy&CR|L;8Al-;rp8Wqmqms>XYUpt)% zj6|~I?C{#ApKWM%+s&#UNNs?E#hHS`(GhE{4DG-wrUl zBbEXzYCm&y@QK#2Yf;vGF^wAv(=26udeq^=q4RFo5`3`y^AUkm-!dXBPcnT6zNccT`)75YOfOrSuJ8#+1g^Q+-n&M56q<1tM8jUV?b-Hn zq3tt;u!xFF>7QT-^zTv41_oNVi!wMrvXu?=i~|z>IluKbxN?x~X??5nG;Ni$kM|e& z@#YFS)wR&C_m+j|C?f{wB}X*TLkN8m2TZ zP%v9(ZR+y{_t{F6N&_Oh&h-|rsVbgZB!mtm+o>;dQciY(@~49r*}4so`k)7Vm&gOd z-&2W}31jsf#PS#2NXb$^y0<_u`N$#%gLLto*F@r5aE2lH#1}`4+czws$%EH+g7ejH zakUP}F<+9NCh=d<9bm3wEp(k1V7qO=1+gUjeX zOEtw)bJss#TQGFG{zUP6lG|aerqwX>CW0B6^y-Y;Uk)sxetmZysNT4$9{|NeZerVI z1;J~P4LUn4T748VdMQdeyH=QUqP)y(RfYLLgWVTgey3?M2Mc+)i(MK*g@^uqp(&D5 zG%8RLTsw|Toll2TfEIbzFJQQpa}~YI!lDA)tT)eMVf~wq3eCw~6nc+WgqAb!vLzW2 z-AIVkdDzv5&+6C_7`qXXM;U8nN_LNAhBtgq``h{$fe(_HukqT2?*{9->Ffl zI=3|CFoxDL5<+fpNW**6gG?-nL7oCV6fZHZWK_5b4X6z^H*4502>GV&0y3SzSf&i6 z-4A=Pu3}Oia|t9j@iz`+;2+PP_j({O2c%jS2-?xL6Hi=2LdTl;2e-BSCBbLv-JOWZ znnHbvkh(a8)Z%2HxY23z*MKlwaG7D(57e(DY&!v7k*|WzGKEvMad z8tQm?3NYZPtyb*06#4_vwYvs}+-lvZLw=WTEZgyU)sd+WPti1-p!U9#1Twh2}Ksu#NIUPg%hkOe5LN6e7*Nm1lXZ zrck8}JBx0x&bI#8!suCP^{P8;NfD{M6VC@kPqe zRQZmDj25`97_$1?U2d&GhlwK>T!b~>UQ^(21*hv&FW4=invHg4xsG70d5So4l&&K8 zAzXAdkA-Z%w3;mwzm@eeC~V{I|9}&tGl1)6=wI+slhc~HVi=XfZLx$_WT+g}2RlXH zzK(e`Hg?lJ&%$!|vG@0z89h6BH^EEX^=utN)8gLGU?4%p>2W`hN^!`T+eNemUvL>N z>16!VLDVi6cq-3Y8B7YD17j)^JZ1;f9Uvc^#;{BS7eZ0oLwkF=8$Ewqbq?sqkQaE& z+FA3DazKZi`^_+2vA$y$m9F)AW{YXcxJxI}f8Xb|@SgxV~=-0bm9D8;G*3h}SQVFI_K_9NG&Jfk5zVrcm zW&Z?c$sdS;kjRi$Dx0|ZpG8IYc9{H?yvrXQKLQK7M|IAa1nh4?r&ituo>(R&xt$g; zlLcXooHy`c+XtFk?ZvD`t7tQS5|i;?Bp zGIz7rK54E#-STCmP%q5SWG7a86LKjUqih$tWco2S(8}Xe;arjh9Xy6dE6srWTNv-K zVi@@D_Re2`r-8JGa2Jm&1!D(*+VYO=%O>H#=d9qe;q@d-3JEnzgP)jdvOsmOypv0M zHlv5C!mTg0?vr!l z*2xF!fV%xD7BXR_EWbbfKMgtaPw6cxK_jMfqeLpYqDq%!CRhU931sskGJMjoIs&hM zghp+hK|d{!&yA2dFgx9ky&|!(IDE0CqE0JZ2W}`RS{cJIi8KjOrygNT@>Y};w)Ws^ z1q7ZJu=n;iEO}MT(`=}DfImTd19<)w*3Z$ELy0q>p(2D#u1Zm;Ibc4_)m z8dE}j$_39K#h?kSm-|#Y{N&ZSld(k*GfD(URw%}~ho6wAh)Q2Z9vM;5ubW?%AJ%%! zq*XoeYWK;nx1Ud)I*neLAnNwRH8T_pUhLbvPA5n6lx z@e1cU8NI#LKN{R&i)hKhAo{*#hsTckxLuil-3WuqdK?V~+xcVBBJXMz#9l+ z)xe#tGgvMMT_p~DTqN)I{t0X7o_a{ftW!rsNTBt)aWtElrBl8P#rT^?)On8r?C$=# zZV}f-!;kwIh_Y~E{0nsCFkOW=tc=?pFILjd-}>M^?(;}j%@*F;{Y;NV0E$(YFN0dD zz!h)@OT%0eQ5)$9sW2uj{|z^`^%ZK72n;sh0W$h@;-Sv*U+HuNLU|&f(%A-r}i)j_P2ZIgLM6NsT%m8{WHs-_N zR`m;vn9Jno+0lL&jItlT`)0EwxvfdzbUa3E+pV048{VEzok&K#cU9RfA)F5uw!{D< zOYXygabd53(p~vR4hFeU?w`=9Oa{^<{Y{aD+_s%=uVy4rr7a-gr8xej4lSz^*$*itonIuz?zkyl{OOSvuqxxjW618oKMc7S{^XQ6_^6b)P8tADjHet{V+{;gMt|Q)br2h4V zj4j6t7rLp6H|WzOu8wVsuU6vQQAx&?_+9ss|It?YQ$1$~|439K>$+jBcJ`e#q%MO8 zFZ@IoRe8Pmu+nNNe{<;#f#0YP%Q@AB2cIt7ajR3h>IxSDP}ejtOXLsmDh;8hJYzGb zRb$?AR!J~6-JVq27J09fnx~a(CeL1rYFm7t@VAn4iA-OFVIvq_^4y)+y|S|igM2*y z+XdZAg}|um`C$pk9M7%K4}kg^_BHkEzV!TX`EevHcsCFW-pw}Q-jPY20QySAcNi$} zu;eONLE6D2Z8!G}I>sl5|KMGGe{m!xw{6q*OP#3ZJh#Su{{2xG>BQ8;LFo;kw?a+- zVA?XF_68n@0LzFI9LLtVde7W<|0kfhKh4Oo>zvkCId@lcjVJ7%gd_H^oXx-NyJu-P zw)p$13AO9e_m1@O$~uM#_fh&RNhH2G#mr!(O-r#JkCFsme!I}b=Rm`&+FE42U#>rR1UaBkuDqm-XyJd0%aqBxfNMPz zquTN_>W{G{Kh}CWkD=ODvA@*bC*Ps+A37zo&8#eho7r&1&Z|3=XJ(+vP*?WQE`h!K z=l-xI15W2DMT3;5_8%q@wP`1Gy=*YyAHhAeiL^9E9DZVcjZJ(^Yf>8Tat41B67;Q@f$00(|eaLyS%yHrD~VXZrvg0%P@IgC~^GD<84UM zab=m`z!2^z>kfpih0MbnFT6vB`pm1=$Kl9VBbPQ3wr-d9xH=7Y9D_wiWy*u==kDL1 zueMyc=lWWeQ6g*Nz1ttv!OK$^QAwg|)j5?i$ZD%P)9|P4AZ_uq7#r#^-j6nz`Ed}`#T)bGOqScv z9!+KwO;RpKXTVZW%9B(yL+Mvl5b7G?ZVNE0u|X_Xh%|HVP$cwDuPman+<4<-Et51u zW17sd6NAMpj<&bGVsHM~87dZ3#gx?7gK2%-P=d{3mzu@x5qM-HDhFoB_*ajPj_sE6 za*hwUY!}{rwYnr~xBe>Zg7;kMlTTaEp75a)4Ke`#g1Ngth^#@ZE>B=j#iQr_KHY!- zya9ie_a88slJb`-_~$;`XLFI*L5y}7R}(2JbGHX$lSK{%0AA7r?)x1KV!odJ48>%n zC+hhsDaLc|@MTAKg8zYSqg~kLF=E|e3|@BpD}=PQl8{lTa}iE#!^obwi>Yc%PdQiR zp8uL?P^C(xSy4Kff)<>Hs^%=I`!gim!Z;z$0NOV$e>~;-2V{P5qPdxxQIQ74wuiR0 zFv(HpL3A6c{(B=#wD6Jmj+1|S7_6M1#oWrLIG8FQTQG7eN4(kXvrCe2AY(%nI8-jI zhZn?NQtO5TbKBPnK3qN=sOl$V;%sdj4P~Ly-(XVX$Dzg_i)6=n>IJFymQtW^T@sAf z^t~7dzS$h|$3WIo{{5d0;HI)?p5bZ6o5UU z!ajgu7z#A?AG^*F-}1$zVuCA%o{K{jR@ds!pRV@fE~_V_ec25)2i8EW!5ca;wCECx zPW*K|nzw@m-lZOBKwl|#Flo90gu#naV4~Dn&+2IauBJLM_ntu#>tJIE5n|%m4>z$S z4J&>pv4C4)$O0;PW8<_btOg#ai_|#@rWfyL<6(24BuAP>(d@+Rss}&PhGz;L)@jMS zv#4)o%PgDeGH@9rcNK*j0r$t{QVlbn*TMb`^2V!x8CG=E%TgA|bM(1J9A|I^3A-d< z2~PQ9$NNCXqMPsygOj?dZj~auk zV#YKREK-BE9or2_t;-vpccUa}|8rziV!v}D1ePByq@O%&IziA{s<$XV8jBWO6b$G_ z_~UvE>8PW7ARK^>_=xOGML)76U4!mSwc+*jxIy)1T=%c@pMq)7LmxVq$Od_j4>M_$ zi!Od&cQ91G?t7d*tU)q3~lUzII^3&GW_tN_dmKW89C0a`vqr z2cEtI8}Jt@py)S7iwp0)%anYI)5KiELd4A z;wfxQTUEDM4XHfw%bX%mAT0$eks?Y0=uj+tH~*?-)YpTzD1(PM0(WfH@U zVNM5t{q5Y$wpG@w_Lo(&J z);nO?0cg!fo`U54eZy(_GGMJ(3QLH0ZHD0I&M3f2?UV)weEeE6f_ID%_c2KDI(+#H zF$0Izv#LE_<7&I}IQ4#m%%RIm`H-s5HYoYapmdg-beoR>wM#Q_HA>387kwTiJCqu$ zyafg_B^6@qwF@7u)>Tw4Nw=cq2L91R5O|jX$+Uf=6aPGJEE3egCB8I`vxhM^$Ok(< zc&YpV=qR+71^h=0WkQVuc}f2dSZA)P-t7SE4J_9nh$~K}`e^8$)Ml5$(Rqg>rplM7 z_r+}!L*9;O9^CEJi{)Z?BB)iFodA3Jh5U5r|k8O7?^cE^c;z`ya?7=91-yG8@C2ynE2xCGls{Sa{~qN<0XCRH|0uhufNnrPO2;KbM0aCMW+O9}|A+uswapUM_4y-Yt;N_x(Bs z)IOJUW}xA;)72Zm`OgcTwD*PferB86=xqu-A0=3Yf^z*!Jo|u6;^Jh36v$XnKdae^ zsyO}G2APnyWWdkW)DC*JIR-xFd%A}jKG-tIN!47dmuCwV1Qh75AquluLzlr=iOkSp z1_jlX30RjaDuRvi=$nSFhAHCam)E~{j;N_?Rx2nHEgMePGJoKnzFR~1R;@lqIp4~~r8)L1}b+2sWSk2gfAkoB8 zlpQ5xrdz86#xzA`_XXkg$wWud5%kmKp^T5{tD#V#_bwuy+nhB?raG*FjG?@ZzM1iG z9Etfn2w}C((+SNWzR4EfH29rUdm&2o)Yn?|C~+lln{BN61w69Wh58;%`rUPs&>PN4 zb9@fCzjvGo8LXy z2IV@Z74LY>o}DCY6aI9by(S}9b`O%%!d~_?M+~Gij0jm{Q7g@qBxY8H2=B2ExcxxH zwg9GFD@gTxzAw1{4RDlBI`m{7^z3M8R0X3YBFUg4MfIjCD2)CU_JScm8rqP&svHU& z{{^-I3)iqsTY%CkOOiM61V)J}-!hGHdHXn2t`dM<*`QP`V3&x6f+Y(+xWJsKI!J!K z3TD%|R}vb*)3QTzhU#Fql-=Q0z#cQ>?L>yI8nk8h0~PQ1SGOVUy370Eo~uB-HRJ*n zITj2}@c(8%f`6mGOqWpB(>B;kMN}Aj1yc@|E#!oG7tmR{>2F-kz~-l7I*5e}tCvp7 zUQgZ*oqB8zE{hQ@#PCfYe%#Iq<1u!AN?YE4~|<-{ep)6$$SzGC7yeW`HX$s%J$Y@i3;K z)6!z(7b@zT#4L}G@RnQ(JVJP$qTS~?P$1<9ra zFP$}1J9__RC}xv%kBtra2galC*)owU=?uE$TO&6pnL4TQJ)mtFkysYbhhZr?G4d4& z-RRn@*2`3sm=@&!2l>wl4`2K9%NxN@^3dxXN9Ix0fX&nTsbGXJxBbMSA5TxIgY&~W zdEhg5Md~I|I8N2RE$#9l=E9vl%34IB*~7X>9_N!zEiCuK{bq zj&on$(DNP3m!XDJ=`uOt2SerhRmfpXZ{M89jtcu1A6VFl_$}V!tDJ>O7}W%cd312c z>Z10#%5|teI=vVU{%WhU5X0cIpfSllL1Dby&WA8Ud5}H@cdmb?Am5a|iE|<9590AU z@W+y5x)ckcH12k~C51QkLn`p<+x%l7gJg*SC!b1x7{v|$^zPLh1OHJU1<_E;WQZByA2*c~sPAFznwj+^> z=(sW$Tvpy|itp#gdCs=+F?@_bRq;#CB{aWn+Q=U`E>jvE=HcByt_R#&MCER3s736N zKMh>>8Ffoc!Zx&=no0OEu94X`_W}1g#s9(RBhV7fd2*xZdE89~CEs`X++DnMuk{yP zb1%yQrfTe^62ir8Q%_^?8G z19B$`p8go6LSkm7Y;zv^jpZ!*D;BW+ZF=C<0K_prcZhX_#rY&pf^D=}|5eg-=}#J1 z%YnpOlvL8&6m))PZyGzwu-RUpMd7>lYhyWGrA!|W5}H~jED~-i^t)KnRrcotYNe*4@fs-VsT{QnyF<}n0hxm# zpSs&1Ei#gDCLAlaUncdXa^EJC^cwr&p&{p#z;m?avr-$0W7%*S*%De5n?@zrSW9fLhkgBBn!XHyqZ^4?i)#cyhm$-%IVqq@tm`G})!!;bG zNc8deLLs5Ix!`sN>2_9gxDHOkGgP-X0G@Wh6L|nHe895+tnj#y{Qv66eK1YaCy=Gc z^-kE32PE=474tpr;neM4v)yr>gC}E$i>yqoUwN|A)VDLc{&qJ3LlLKP&jlvt;@=*4 z36iZPfzJ{l%JBXi0k_le7;k42J^Q3op`A+OPUO)CuAx0nnbr|!*8OYw3#^`pwAFQv zgT3)|U+8Snl!Bb-2ys(5_l;e6jK|D(lbaW33PolVMTSMpRG)!IO?Tk5&9%f`2;82Q zoYGdE$;HSqToXyh7(~Te*XZcqjn!tGUY0KP+Zo<6U6fBy?&ppNzt23KVYi@O1}|+o zk<>xsTez_x7%RH8tIT}pT_NyU%^loDJ%k>)@pNU!v!{OXj97;ms0@o>fSE$IKs_|R2jWlDTat`dY zQ-l8#1H^GqKB@T#f(?!tM_CBrN?vGiX&=~~k_>(uRhdDj!zBgy%Q3GD{`%V-_gVg+ zq`hf<_YioGLtEFDW}#(OdVX7TYCB%fA;{TlkJBFig;PP~Ic?q*_e5CV(RC-H%^nw+ z<7KRqI2Xt~8E$P|h;&N}Nn7wIbz*dAD{$icRqXj`(eQv5eF=TC9aMH0eU4bC{vZp- zOnEXX5l5LWP?5D&Uy9=oGmG2n{-Fm;$f4yOPHA+>D{`W90L!zke1abj-c+*>BK&40 zfN<(Hewzg%H7C{(yo2B6Bs+?%9;aPRjM)Fjk=@X23KUzr>KtGS0%MJzxU~xxFGi`WG^r@v@EdChg`PI0~33rvsx6 zKtJJRKPa=|d-WPn{%TI%B&R^Q*LHjb+_svSMBFn$>sQ87elIW`&br_a4E#Z1J4$vM zs?P_**~^p%bq;v(nJIJ*^wi zJDiLN#&7j;x|&FQfo>gc7&dvZ9VF88S6VS<){*-3r33PL3?!!EU<*iOOU9x9YYY{y z9j*EWIU54<@U|fCChnjYiGSQznc6Aiybec|og>I)n;{_2n_qb^6LfS0_}!jHu>c}a zy=JlvZZVqIDe5t;U2ia|)<$N36P;JT55YDu^@Q#Et|uHuTW_|ID$1_&jv|zgPT-UjpvEx2`*Lulq7s2O zcBI=L-?6ZIwMQ-G<Y$#KsM!9EP?E2! zJJ5ZbV#4Ed;q;fu-}IL)h_=fV$FTlzDjd<874jP*O^%k;Xr6V@Fi$gYycWR zrikN*e~Gz&b(c|LkZtBKxSsG!ym*dE1Xo3PJfQo^Tmp_s?Qyp`8_{)q%W!cqjK-Nf zC-ha&1K1racqMEi9^^KdBp6ZZ_Vm%odqGUefh-ghcJ``n=b4=oV(H1_hRuOY+H-1G zlW6?N0lfUc%>@DHz!E?dvUuC{*DSQ|0>7GtjJ>%;UBql^0S{LacxZ>d%V}1TK_$Se$5+4I!E3X~Cv~Drta}fTsqquu3WLtYI;&!*mdN zFUR`jtq6CnI}^$B{ugn2Gsy$~36ftB>G#cbYHe0ijqDlUNQOyn`Zls11i*^4cB$Tg z^2KX+?n7zku7wX`_J?dq13hm{X#vqK+U%(EgAOq5iu@LeB(kxv9Qn7`bW>hdYpOMj zX)k_WMB-r>R`KFY`ZE93b-UyHvfnJwC4C)nkD*eNvA@c50 z0G7LNiA?K?+eC!{aO?!3iml^>2|7t$8?1c&{Aq>qsNng#R_yZUq&pUo=&y|xxBA8r zILhnnVj*H?S~K7C2|EBbQS<)vz=;QNP<>lW6y_mKNI6|pNOTP{R}v^h^^B%k{B&J*8 zhI6OE%u-I})lJ*n@D+z+*Qe0RnfDG2ws=>ky0`u@KyBV#kc4X8Hx`HR-o(3fy5v7_ zs|y~S9+9Jves_1RB5hR~I&jik|RYU(G$uJaW@FnZPlOp|} z7pnT-;AA~Wf3ZP-wY&3W-OqUQIO1f$w}7l(1w#e(%>hgrM$b2Ietn^Zh{i#{z~%Nt z5`5FF0j^av{5LAz|I3*F4G`KGU1hmU5xKHt3JIBWeOuNNg2ATQ-&dfnWV?DU7Ij({ zNb{T+J7fUlHb0KRi21hVcb6pJ%p@RC)b=p#*b0F|&8r*k;srMOL7a=IX`|AEQDp{l ze@u~37g{qyJN`nzTWE=>W;)eYF>y#DM8VR6Ye2K<+D!_@{)c$)AIQ?pjm4WC+gAc~Ll-@gpeNpXwC zElzLDhMGolte5idDtdHe6hv%qLAlRrzs20zNEfR5x&l1^QI7sop7W(zG(+-h9z9jY zg?CC6bXjsxYWL15bB;?&`0s_kG zm`l!-j9!KPP3}bjeLN1EI1XEh-5qN(H>RLG4UknE{Z`S4K6c0NhEl+Kss*+?Nj0or zgf2OEqp#{i?;}~@eL>i3YFH!@o!H&L=<)RqbO%{-?5MSy4L8`56X+b7^u$tv&~+Qi z7J|-9V!DrfqdFP&c>Nm1;b*s~QtIN2v&aeoDy{mA-m{|=Gkb4O-t(&aY_Kx#JS|eQ zbpaip$N-Dsf?iqh3|wdys>5>E^5tN$oyTrhJ#oj7d1jH=T;x~PU-jU5ep{)^eE;uT z&H2ll0eS&^q|%Y$7qfe-tC{14U|VpibP=gXGOG0wbh)Tx(?E3mjEYwf zq##eX9a3cxO7$P#q)ojl?y!sRy1!ms;Pgr`5)eaL>M&-0N?Iq$XE3`FC(m~@fa#p& zO_wFU@p`7!=ZsRAD}>gPvHO!FGHHt)GnQ?iO1aDquEY#Ww{pm(;1 zBzcW38G`C1T#kIg84lOt|IZQqvzkE9pIx#P-l1xBB_ucKZ%&EQ4yyK?QrELX-Ra z3`7Qk01=BG3yHRUm`i30_c9p699xch2kPUDu1^vWOg z2sSY<@Y(qC@XVuCbe!iXgXmIzBnhLgas8AEV&-2$O?0=?k$mTQB5K~r@{PvGgScV( zcZigAn0;Nlg9+`4Y&kYVxqYB!9R2L*nETAVIc*7HJ>WRm9F~ZBP(tfoO%mpPZ>t6B zrGPWQYKFqP5eBLA#1{VRbYj#7`K8^bLB9;~4^(=HzsLe7J+%q9M)CSOMEd+B5-eJW zlrsd%8ARBTk_nUR!Jt;o%@`aBT6y#gnp&Dy_bP!I&i@o|TDcGSDmzIO%xfi(xmH$Q zS)c<834EhQ3imYJ#Tg!{*3*A*)B}EnKJB6txwxTk7XKmZN5h|)$SlyB4@^Q++f61r z&k%@DUE&5wf?STuSNcN6p5Z~#y{rABZmy2==u zLIiuG(Fa(RG%j#%p6!il=p%g_Ya>az7L3-BX9o?V_zN)_5!L2=oY09)y|L6yNRz|Vt&Ej z<8kufVd3$px=$>VE+L`#SrHay6s1DTa`^9cYRRZa0guls>M?43IbS+JC3Q?}|;{M7( zo0rar<99twQHnZ-@0DoQT6mLD1l%9eq(lTM-+14%uSUrgGa(nE{F-vwOMG=b{}Fsc z_FtVT)>9MG9MQ*#?Rviu#s(NCrR4pq(&Z6r$?#2AR>pfbmKj&;B$HL}-jBPy4w5}H2Ky>DH~Mhn+5w1wXMac!Ac=mJejj35;1(VOkMGNCRLJ{^KQ zvB{y}cZgmx5j!hMhEFsx# zkrdyrPuRX;ciKuWLFfBmYS3@Q7*Glr!Uvdo0w{+TQx{-aZ z$9}|kRa z#S-6-d-IxQ(ATw!`1y{I-d(y>tOxPBO_b*jYAZBMl>a^9b32oU_!3;dEt)%Ce#1A# z*GqTZU7w^vEA=Vsf`iEplk#k!IMrlm|21;W*LaM*!xt}jW5rr|&i&yOYZ&Tkimy`j zvtm8ZQf!_s>V=WEY^cHw5u-W0g>p(to{VWqH=CK`$jW+<&;#Y6lsthZ?58vYq*hq{ z+z9XMRetBHSgyZBjA$cvCo-Y;E-^ucho*9vEQz)$S$~KR?Y=pzkv0#Bpt8dMXHIWb zKjOos;;Q#vZMT&iDgGwx-=u8H#_>d(nw9g;IUdY>^Pe+YMOed@-ac&-_Ynw3bh||) z38B&urluorI6-RQol5a=^6JsbeHrWBMOedgj{K3C4-D1I5Aig!bAEk8x&(6;jry!U z9gG~e49t1NDf!MxvD}YwQ4udGHwP>~A(RtSHqN541_~Fh+(BhykY6G;ry-4fwlkCu zWK@f)qB7ntO?+e;FPvg@tHbxD{RcSSJNNB~eiPMg0b=t$?Za+&4TxWp`Zf_ov(}H+4TMGxgPN-2oCuO`zt$7B&igo zE!(~cdKc<^B1ck*3IR@RS|APHFl>z za@T3oH$=($Oj6#yZ?Z}!qB}F3B{|RN13;fG=jb*GLMmMn{iQz0c*s3P2?Q47@C`9{uKFu?(Iina{>oMy_7T&ZZhj+gel9$l zjQaq@myVSvl2;$5pQ@R}lMf7RF7IFxxE?&)bZE-q5i`H2SKLn@@VHb=ut#HePB17z zr}*V+=Dvivb8?;|D&eGE(&TS6I0rUC>%YRG0&R+OzZmP4&tn66zq)w_+hlRpU>~(WEOD`1O2vSm zvW-Z0^@~m4)WHtlHyt-}#S6(IYABETP%>(m*lDPILK%OV{M{&aqp&|u`oTh>%UVJh zDqGK69jlMyX4be`6i{Uit&`;F)KfLfi$kNa^S3G=$vGpE? zD=|-?RAAr4_xncf-9%ANjx@x;)pB0@lOk%@mxrL)Al>}5tHOBi#OnK!Q-vZ?SR5{A zzt-c$f61Y(Py`D$mYjJ6eOU8BQxO*TAM^WWuD8)Op}hSk7}pyAreU27UYSr-=S@9g zG9n1U?M3O&-Tzgt8u2-oUW(au$Ncn4T$~h|2to}~Y7MRNNd_^b7e#Kry%|zrv&jNDsr(C)))=8iF zPQo!+ymS$>CE*J6>V*oOKv7*M)sQL6K@z52EKS5 z{g&TnH{`g7GuSPq}%zUcuY% z*L+o{I6}*?aXL+@F11(zDQo?aaGB5+X>AAX!#J#g;FI>eI4ykbW`pF<&+G80Xw*ol z8M92xOG-Fy6tHJXSy#2ol;k}l{J1zqUb+qe1h7G#>X(V#5-d6KFgAmx^9>`WS&}Cn zvJsH83$I@)$Hzny_B}Xnp@w8azhu8sqzYi3p+qjy8TGlz?jtGO43+s3daUO1kYIhy zb1Rx6+_c0D|3$C5r%=z*bQ;gaCum;foKTI`6m9N=b3Nd{(1bB;lrF_b;5Y8$*?xXu z$)S@2I1DJa5ehX8?<%ly+zxkD)MP$8HJv!kzj+wgcKTtld-#j-zT-x?t~gSz%1CSuEvbUFhX!=sB6qRqhNJ$yu(P* z_vcPD7bt67N?ff{PUUobNXcaH?6&uGpPhLOd+0Eb-zp_?|+4`H?^S*$HY@^&1r#FhTSo%`j zh~jARe*in{-OSeVLzP;(@rA7idn&gp@`Sf1V=|YIzwuSQn7g2ZdcFl4n{FN7aoZ5) ztptTI7F4F5e`=8kg$cEGgLnG4qvhLLH}fk7H~gQPp&GfAu1;8=q|~ycv^EyEFTw&N zbU3-S?FW7hQ8-mw>SggWZN^WGp-X#i!krq90Lu0wx`ox&gpCyO-nup;^=)vOhtFuM zP+fg*We6|a*&5nFNl^7kglsYKjKNWW|Dam#$m1~G4Yb z{B2dbXwLqe;qlK{TonBGov8}&36Yd3IgTgjm@2!1Q1|@+E2NfM3z`Zm^JZ-V@;s@h z;YYT{8+EFC^obYF`F1qzgVRn;qvxUwzZM|?#=2L{Fa=r_jq!}b-IU69S!)|<0oIh( zmCOGC`Skwis#mCN{c#|jqxj&1KDUoE4~O#HH93LD#Z_(azq0y9#Z^r31hEa@(;4cm zIu&4T%pEg^Vky-?n$#iT`2!RdS2XdF->?-6Q%%S72l#mC)zO&(TE%kLe8n}n?V5T{ ziDJ!dZcKt7aaLN+V7ZxqjkGv+LgV#JvL7Y9|D-Nt$~5)BG4~~Hbzj$W9I#*b+c&2N z8NFH!g<5RDO_UEULCYgF_nCEH0n9D@k6hIbW9!^_7<1Eo?t%UHm;-cklqWkH4;0NF zrgELTQe>Q>!y24Z<~u?$bT3o!s$VLsx@3DfOzY_+Jww$vpRNg|ZP5<;vFdY)YVQ#W z9mdp82nDZEJem0`GCN6`r@PSE$(rPe`h6kbzLWLc$SVivJl`SNTDpK#Ap#&Vlxom1 zn!?RuJ^N`cwV+TytlZs#`mIcSX#GUGmd+MMX)Z_liH1oKaQ)`9Hwz`A%i!rh9k=le zmTmM2hI8-pkiJ5guk^h_)s72t1sj`Bi|JSiZ~N>;VzggV5|E)x>qYCGNiq%6{+L#P z01BeLn@UribK-MjDx)=CpL0Bfa|A^mpqF7)I_5GO=Bz2|M<4k))8@Z%`@B6(pK34@ zUNhs5vffybM*#OlbIpeF&b@Z(`o*3(d|CM;`1N2l==hn{HR*;?SlgBTpIgK=t_(kt zSHb=Pe`L%}$mcBNVg=Zo<5h+o5UBvKsXY7WUrhJe{|Wi6q^OsF%wSR&*UR2;1DKh`)A_C_MbZUdgOeqO&xY;9iz)|-!WF!qVx@6=A(`JOGb zxNw`n@jemlO_nl2Eh#SduvH_vg;9Z^rv_^2_-LOixN$E3cw4W5p$Zohhy1PfrhY4} zLXMW>m%5LqinRzNUjG`Y!`a;`bfGo*d}%lbQnQyH?p3BKy-Lu-)9C15 z&bB4DuS7b&g*fBUFtkBa88eU|`w!@q?E{ls8c6yTxcbXio}vP@jv8Y7;#i-a>sRmF zk7IJQIgM4Q@9SNBpX1?9b1_J!Mzh7O`Tjossh?u#z|bsy1c^mw?o%X|EI^$@?aewGkgEimD+<3xAmwban{7sAnYDxX)>*hjFkw0r6x^z5vvF5yxYt>V@ zUB&4_B9hW9Dvd9*bZTf(U2|WD?*)6_8a@{d9N)X#317_daC9^Xy2RjqQy4D>$Je}F zJb-f0;;0<03Yo!V@ahX>PV%jJ0hVr53$EN>_?^PLP$5G^wp@KPaLp zmkD)99Ng7v!UJDkY1uJyCG7&syXy0i#}46A<8ikUE)`(xLFz@S4U@I(VfS-z(4um;W=D7dhiKJTs);Hn5r9MS;|l!hO| zeFy{H@Rt?OORxhE+TO3e1)n(5y2WzK%r#%@n7`RXm zG{h<0l6_(NVLR(%FzX7BI$qc2w=#_ZmUkO3n@_8n*2=5)$;AV4saUs}=Lcx|XZ7c9 zHZ+lcz6iij5E7^c5B(hiuf>C>k*3Ol!^Yz=FPj>F-e`%#zC_yRdI)D;+U~in}5|2%cUJ+AseYVg?}D)`e80qFl13Mjw!yp z|MILy(s06Rqpax^ueF3Zo)X{P;`j&6O(*D>V+KuqGI^aJedVaq^cu>I=Q1E0;# z(T$TjRvGwja~=LZCq6j9Hz-xcYbvKRTU8ie^Qwph#{Q1cCET1b*B*HythOm22d%;3 zJ9cJr5vi4mI2v6|)_3HsdITuI=|)5OKW?xBr$p+d#afV4_Pd#BV2Cg9Y$WaHt#ROX z=k?OY@FjrfL&bHnc5AGz!Z;uOLxdmYN3;GS#n50Wpoe03$ZV-Gj!%%jKUB#2MxuIY z_I8V3sjz@lBJgeze8ng=&*e+mm>Xv5gy*gv+ND%dvV!zjp<<}=_=c}>LXBMtuN}qq zO>gesKR4$isc(2}v+3rp^-4O-Hzb6gvd8gj z`%iZVuw|Kjn}0i3{`^P+sV~^5heX&p;};un>ETP7sO;*D{?8{+`|0qO6*yC9-!gwW zp!)-A^5+?GXHqb6!(VAt8GYx?g1CFZHy9{*a0%_%lQGVAV$A-6&EFd)`g|%$onW#11Jn|ywrThmkXmE<6tzV89+w!WvpNUgZG-ak-)z^BK=E1S z#gt069c1KH{xgRdQHA>VNLtlN)2Ix2JT)pOZIu!1;fHGRA8O0fVh!&pRihqnA6tNU z(D+gjDdz8$DxFpH*~PXWEj@s+M%g@nCfmShgP3S>jbj59?T;Pr%a8--QL+vcBLg_gZ}$H?P+EppWxOm-0h{xq3d@h}82&iP2q|jS=#MAw z0$TE)j$y}v<{q&eUm@*i;ZR{lM^6jtQYVL{2IPD0ei-s>glPvuAk{x2`f=&3VwGf4 zDi6$#O$OCN67>%{cagG9!H6$lPY;nonDMC?_lwxrc)hD6X zy6nAuTO>N4BN$5Re7~v_OoLP!q7t=o>FbQ1-peti`{_w5y9hdFo3Na$OE;VfwN{t%xK|-FsCDB zIOh1~f_|6cCFKWi<8>3yU)Ctw zPk%3)%U+@q^u-T?#G2XYa{+^0XU&3p0J_V4X2*!zQUdfqT;zaBkknPlnunzhx0%7z zm|HX{;-)O6>g9L-q+ZX84~uYEvS)>SjA~1(5C3V<<;?BDJ^nk)FNn53QNvGn62PjC zU^P9+=i}t*q@`Sso-(j3`dCZ#i;-~dR>AZRd#N}bCtCMLJ=NunU1}Yoo zE%g}-V+Ko(CS@~5@W(UzTC`Kt_j2PmSY`tVKGk9_?9&@;^bdE=9=%NmPcQ9r-^Noj zFx_dopl)L1(b%X_RM}XgWJdN1JXktnubn>O)tDn$LuQ?K`QvU7y3YPmM zKkxRH2r#_xAw5B-shU<*J!bx%mCZnwS;N`t1Re9YQ8MHaD=;a; z8QcB`W>?TGcVr6>&3R}6%4_h1a$bd+7I3GK%Zo?0$3ek%%t>g|VdGf)4oO~ZL9kG4 zp`H2u-!D(H$MM39mp3gTG@$LHQ*;33<{;j=GKk^A8qkJs__+-@z748=F&;@>?c4Tw zTS3>eQI4gn;qadLhZx)lI#O}&iAQ`#%^2FVC;ep)m_?G39kkrQY3|#4YCWcN=Eh$w zr_U+!kiP4fvmXW89b!~M)*jF?HkKONz2HdpX0#ao-g7^aId5>d%AEh z=>8J2Q~@C-pm1s&n8x9znV@{H(iyq*aX+x-?uVITS^PPMlzmZ)yF`LbzA4m?_aee( z!0Cd#q5p33iMy`KJch1)#^5@gl zJ-~cZ$(@lvMS6`6rGj~=XF@oXjV}f3Cb9uH?Eg2nkIpL{v+%eW4P4ycu}fWqE4(6W zhnlbBJkQh`M}Y!W&r{SFc_jnEjkK#{Aq|;@hKHkRpukW`E^9R45!xB)C~Vy_EPN$P zziIuJiUoMNxXUpDOj%d2?S1K;o{(a&wq<+>zmd9cw+{b)_HMGeHT3zH#*lcUnZS=1 zI0>FD!VkZ!3G+%zA-HD5i;*<$tl7deQ1#m67509BZlq;Kn_w0Fis-({CaCrVrT*Pg z3b$=bD5}@@uuJCcQ&O8!8Y?tx3{Gl!P9$(`qmN}k^CXmhw$F50%a!_|?dC0HQw8!l0{|5gxPGggBSD?)e<(Ovl0+^!(&TC10A9o3MSc69Ie8mC1 z$6!+K8*r_-<@&G#bQtb=^K1|JL5F?$3CW_i-Z|kOvj?B>aG{};w{EI$ z9tQkIU9IohO_-t(+3;^voG*K>%Wv@B^AC_pW)sRi8_TYt4=gj`>Uci0%qK(Pr921vV| zeCPEUB4JKTLEy^3)burmZsR_IOCk&E{%;iILulPBxpa;`X6x&j?bdNXd#u}AOfO{}BMx5pe?_|AGiZ9w81Pt{>f%RvEHldV80;kgJktHV$HCNS1D>-| zH}UP>T$i;cp1cJQ5o73%Z$wzW!^srh`5!qz_~uh!vwd@r<1C4Dec|POiAIzI#`qF^p zD46_d4AibHtya29l`Y$~-Hnrg#u$3SU|l2NYC|`NN)bQISNU z4WQS3y3nTqbJUJU-%bKg$BHp!``T#mvD1J`nCOqBaUky!D-=|2_K&AplGiWn0)J1= z=YyEf3~v=`L85sQ5!L|)J~!e88{d|KP&*34~@GE)b303jtpeq`l z(Tix|9`R`(2d)+7oInS1gU!jW?NO_yr@lKt^*nkp>wg= zj39zV!|FHzvuj|M~jBQT!?l8cziVLj$%UnADSh%E32|9??qds`u@3dd=pGB)9 z1dI?aU}MH4Vugr)?KxP4vwZ-bivNQg`{}USN05R+2fUtZ&!w0V2JL<#d`=2orusZX zapPCJ2vTTzIQiPHHAffUDs-9tq2^TjsVE+6HIv%?LmbTzRZ@z?R9og>c zpQO=a=Y#L=c{GQ=c${%_9DS97{wFJ1qbSRHR;DkSk3aQI+L*5)B}@DvO)QbhbNwg}5Q-1ydEmb**A9fV>_N98Zl1tTw}{9(y~tjzYjNzh-@xy3 z*A+zFK|L+iXZ!QTbOCaANMyqOUAsql^0}+OIH5X4=+VrN?D#6(@zI@TS`bIm&F!IFm_7%^R7N@-v2WbA&);5Q&$bNhg z*@buGS{+@W{7uY8i_Lsy-VfHxJj(~3x+qwZ+OtdFPSWDfF&{w4=W-%p22`nG6{TAW z1}9SHs7i#kyFs4ZFq(plbf^zgKz2Y#ZrP3!`^tfOZU&@kh+95;``rMvKCNx`&%cIZ zcg|nzLpe}et9bqhMq<2R$pfje&0k!)$-vMCpC`FjD}=_W zWp0qNG3rQ=Y;N}C!}nuwuX7Xv>(WiCW2D#|aOXTyND zD(EHm4#e`G6RHB`^gqrAC1c%|lGg+oU1q@KnKR~*yMW_HAT~07|3}tb@3!B@FoAIR z`~g2MU|92u1!#`WZvA<}znU3T_DAwXlcV zv&Y9&B-mDt`qWWm6x%FAg@9{yJ-4=!(@g8D(!-hoEymz>@*FdbE_q{(L}G^UX}gC} zD@SZ{OX@{o3VzIBqO#lmyn(hBU!EN9ESa_<65QcH^SI{zH@=97JpNPrniVZxfXgt8 z1VR0~hw$xd3DA{0=FvB`9oj*rUC1KweZ7BV%lMeg^N2790klNjHBx7PtP-5=`QeVo>UafGVRiNeAe^{ugR%pq)#xRIV#+RW1mXlaE6-7L5?3 z=%}lAY4kHr8*r|k#3kzuIL@R1hYQv`7nXMNN;^G6|KWKfe8BAUCzmZ+0%o2=$Y-4n zWiB=J0bRT8!7`i#7M+&q0o3cHG6PIIy>7GNTYwDDiZ;7HF#i*nUxcm$w2k5GFW`P! zxB)HR*h0|+z4YiO0ksGRUYzFY*a7a6F*=ota8J3F-8$aWM=4~He%7Ucy}t@oxU+HB zM^)zNO{HrPcG|GF77!wsu}Y07ssI%gt-dr+Fr+SD{%fF1)OrhyZ~9S7fK$U|UR9{N zdZ8|7M{`+K82(CjvAJtxK~ES{S=mhwk{emi?W4%Z#w>vqa?CHw1J(s6BdEy8R=o&l zFSz_E9oh`O_#CqWj>`OT+6BbYfAldN!g(nE#5;SuBXIF|K=fa5)5@(!YZDkaGG~;T z0z-2W0-`k|+bH**kvvQVt6`)+k zNj3`Ye!9$qWjU(Ls|p-4+%P$&1MU|$ZbYMRz3PwTg@o?6;W_ju|nEFLrWw3r3^lcq)&Rl`3Oq#VQ(hZH53QNO)aMH!qiR!{F>3}gmZA&~&boIuDRoc+tKD8PNbgKsLxgbtI-lM zfGFi*|;{;B{zN9fi@mVS<6WKSAkHekX zy<1I61btrB=wCRp72qztX?gs})@LQ__SZ0l$0t5X&>4iiZKbKm%*{svv>Ttn@94e~ zE^RfH@eBd*5c6!4^8o$au?+E;M!)8(uXT2hH{##ZUVZ2}WO`n$FB3W@=vfsFl zofhCXqM5Ta)CX?c7bIQv447!0n#W(72B)!Y#yRtkw~rsfVagAdEs1#Jm8FPZa;neP z;XM3LB(KK+c2xST)?QE>HgKoJs#>@eAgCYOjuQUXuT!6W2Ze0M%O28ec7cnWm@kYO z7k(YbBumxOs_rpE#Z3peL`^a6qAgcXi25i& zF&2kZ=)yn+nJ8BjOdB0;X$@(f#LwZ{$h1zJQh9;HI-HwA^3SvVNg$u68@AGKLN&2_ z88o%koZ?v)du&GKA~-~l8d{AMD$=K+7?{Bm{GV~SRAteeYQLHvUowM8UUo6R@0$LB zTyWbDX}yC!yA4dRk)6c?8$+XwdkYbFX;l|6CXy>%C#HRnRW{M&?%r~+_O?&P)%$3) zbAdpj$!XAX{OtlVv!tuIjQv+_rX}b76*ZIfs4WirmK{sYd4{~wAk8;Wa8KMt1~9%j zcb5!*x?361Fu3c2Cy|c)U7mB%sy?x**|uB}1?JEk!TL0jD1e#o;uZjKd5Q1AGZlWdzFxEjY4<`E!);bNiu>`3oVC#>}4wgXXc%x>m+G7cI z4ftgeMQ(tsrOhc_s9(iR=D3^P-Em7d<@a%<_u;(mMP>Bzqy-qv=7*2J3mCq^KJ796 zJ`7^PjZ-kTlndhy%Y>(YBu(`F(||r8EHPvYHhd>MTsCAQt_Lh-~F7M z#~y7bNIil;F<*BE^KHY@iw-Q`ei#ZXt0`2;!Llhlo^GFnUoFx+96w(G6-%BT*-CW_ z^OsL3qZ|r7J64w}eOJltWS-zy%zS!hUU(k6tvLCMTORHuDPa^?g(yT0Nqz<;bciWF z2_^^pRgxm3qJbU0rGOfh&tGu{cD<0=Xt0f|{Pl4%E!6xgC`2eDqb|=1bnYR!VC0<$ zX~eyai7SFE(FZ>otI~Xkq1>0y*;2N8EVhCs@6iiqmz?e zT9;@ZVWm-b6gWd(LW-N4ng)=rsCt1wIqsXgA>V+XZ@d6Nh0873ts3iAQNgJRGKVu4>U~# zH&M8OPXs8#k_qC+!DsUwj0J5yn4{*B`4b)Bp{?BRw&8bX$5aupvFi63BDAMJ1;PAo zL+Mb$eyMi(*$JA5jJG4-fC?y1kt4z&5#sQAHX}jMWuLW68o!pr2ckxBOW-vp@IP)X zU;8d2C9`l}K|8d;Is499+3)fm_=a>x z_Wlme#OcB5HJBeCAM%9l0_v9pA$v{a3TNqy$I>*Ze(6EmB3e+^wr$8PxN z1zJ{sA$*rU1Vgum37-091@5^V@3g1xBzk0b2sga%pl>h^RGzQZM5F(P6W7e>2~QBX zzdyu4rHgnXPd{k+K+hmMV5>9pR+z#7+X0$cKF-6RaXu7o9%m>eMv$#(^U;EUqFa!7 z4&5q_vjBiq3hP$QQ=hm`e`WyNul}<+I6ru#Izxe;#qUew;#vDYVq7O^D*S1M>;u7* z5}85_rBFQ9jfKYXgnwex1)=+}k@1Nv@(vX6ByUV$%dXHSo{G410!y=y^R+n*aQ+(i zlt}m~zj%nLY=a{WB|%2>-S~kOtaS$fO$8K5&YFKng2u2exn7&i8U@%LFD+hLNB>pM ze8g_K3?gULOD|fs|s6emIY!@Yw3@Mi9Xr`*3%)G5}q6XCAc+ z(68m+@W6Xb^d2foPAE#lj^+-Xov&51t%N$SkkFM5C+`t$Z5(^&HE9!d@4bT{yIOqU zVV4}C#_A?LzSffvkQ>3Iro_&KuI%zlI92yFP?H9S^XpEa%e$xrZP!e_Mdy`i*(CDr8Zu!O} z1!rjL0^0Eyy5sjX)mH4S^}&tbd%W|9Q-SLDA8HJ?+47M^_#-n+vPq(3aFE%8y2tX3 zixyAn6Vi_-y*fbGeM!x51~;IzL#Qfzhk!=hqOy$fxX zq$msXcDA5#bIEOeqU%ZWXIvNzKCuJpjLoDjAbT_;{l}%l0K{T5YdN_7D=9byG`><| zsz~^>aW*C3J}skM0Jn!vnZE>Sqf^P7@B0Ok+CB`^Cphx&--QZ2+#guuF;7u5a9(${ z=n{>4$8hY4?&*W>A~AaD@R%;&U|aOcBA5GGJ>L}h9gny~yE zD!FAYAxOXqfiBC%JB^A^R@Zw?UKLg9QfH^o)Q}mZTa7#07J|L=1%c)h$M@7p?k6BB z-b8?eG6CwjAAGxWaHzL`#B@6^9;6OCG@rMko?`;8(-T1MT_o$*8K=F<*TuWL9|MT+ zwnz2nz|r#K=v|;`(9G~{P&)C{jGYv=Id;Y zM22hgb!q#BUtXa#HQ#1`ELhX{0#_I3+{_x8Jr3xi~bs z2x&|04VZX4{^K2-^~PY*^=KZ;%kF$DE&u?&> zza>b~pJ%=2(r?)3GpEiZTMVXVdiN`+d|w{gbAt(Ez5o8m8g{wUDDUYkq#bT-_gIoR zlgmXv`DxbMSmkPe%h%M$ABS=p|IdGe&ZOPQUhf%^mWu&izXl4yk)CxjLnffbVMX8RwED6Zl|2RXuOoaS$eSKV@Zi4NGs)((1brh z(Wk!!0#G-3o+}yakm6w8G$>2XNEG_0lZS?Th#q#T=R@5@prrnSWVcI;-%Q?E&+zZ~^`<$Ghr zpIw4fHfJYkh0=0I-#9UClj9q*-rmEXwosK?YLyQh8@{fqTFuZ{Z`um2PYF>iw- z?u8EzBtNNp^2+Sn%GqF}+t-V7DCL_WA>|HH)97{nt+lKUx`P8()L zskH3)vh%=<+s#K`228A-I`8RU)KhxHdBGCgY4?0~b;oY9x++ApN_TK2JI&6T08K8vlB^q0)u z*K<=-0Hixlx;LXI5-E@vj|WJLkHXQMbTJFWs9A2f^Cs7iP<%x6SAJ=TJKk0lr?p+o4T2Qz0ht#RGYrQ|pR`d-Q zoZBQ~Gu;*5H=l(DdS4@H+vF}HW9HH8@L!ud!Pfh#XAZVy$$hgibhz`pLM= zR0f~L?_^e|2GbC#t;pxjOYKm*qe0v zK`-&*RIsAPhcCzfy;Rb815wyd$4Y{B8PWoQh(>4x&s zoSEk9~>a=@}m~<$!*u-^vNox%2T4@};Qgi8Ekr_D{`zt1SjN+XJKima4`UR~kx; zQSkJ#`#`~L-9zn0Xg8pQ z)AAnmB&;q$?nvfeSXU=bLvcdLK09@$+{#cj&HLs|p_ZhDSFTQ9@AT~Dw=WKoy}dmd zM_iR;Uj=-M>P@!T+OZzmzJ`h@2ZjWhndIsL)qGziQ{~?B4w`crYkv?wKAh5RwWW0F=FEX6#HZXj53SMGpI+@3)1EvoG=K7I z8Gh;^;`p%lt00En;e^_2HY0bAe|)x$`O2W;m;-4=;`$YZ?RS4-kbR2Jm1MPB<^Ldw zp0(jfa<`F3(xjvQdOu8!_eH~%*v#hLx`))IEB#sA*^o4lO<~SG&_+Va`?^rha+6bS zT;?bpl01!Fv0;Tq7E_4N{Vp~<72v)|fridIKEF&&F6ibzw{V+}UMev8*ZL2#p5=t= zuVrU!oK%E&^_o7vu-;o>p%g++fNOHsbIl}N96iH?p}z2>mZ-=@?sR7Gr_Tgx6HvOO zaW}6Z`fZ16_3{@NkKwh0>W>^?ptuUHLjTS61$$L|Ds?n_${{Hef#AP z@`T{G&y)kT*vaGRy$QqKp2V1UT4vr~G$Ug&Q(Z27v#z;5i&3gJ@(@V+RTdGK$=nn% z4lJXdDTSm-^gizUr?(41Z8?&vHR3B5CC{-`%h}7suH1s+mjKQ?;irZ`9k}Fn61{+Y zjKZmZ2sTufh(t5XG&p?E9-qB>4)^SsXgybt@E1LyiH+rH-q=ar&I|W?hyMgo4j@7b z`}|Ez$?}a1Ovy^@lNO5{!`x8BYam^UmD$iz*hfagl ztb!Zx{;gz{OB--wdSQ9W_naO6)?6bGTlv%(^fhQhQ1f55mo<6Y!@{6cj*^dwEeq*? zowrpkkw^4h-WjA|RF&pcv>cd@Z1J$|MS^H!Wy$JLpJ)YmI@x8^Tk5}q(HCkUB_n2< zp8k|i@A&qHbi`!&OV|qhaNmrfS3U%i|8$bFLK%5nWnY#RQ`+TMNRq8xn(6uVdJ$BF zhjK-V4-P@MVl}4ahSGZh^Dd`SLdt%??#%*I-Z$ z(steoEPeg@*e}D_mEU;?pARogykBtnp4{flX$o~PWvxfFXu{i1t3;gzCBU`R8P!m9{Ws_V$l134MN|RUj&`q%3s+oIFK^z&r& zAk0a?K(xM2|LmZ*@RQ}c-KRk!#igwX$+x&$*LwPSHI%SY-8BCfUsoN~YW4 z56{_l?!NclZ+wCs6cExnGU~;l!~KowfO9iFhj9)VT}s+${IPV7XzA#(7}-*Pce8mf zJyPkgT~#bW9juf-ROWSuixL-VfHYwD-%uFk6XtGCYz^W&z~ufp1;1Qb2Eha`p9oMb zK#^lgHh)j4y6}{SU&>LBuG(R~|LSH8(w9lF+6=nz1FF3}Q0)=f|J=xyIyli+yp8wC ze>wJ-6>7n1Qc%Id-&kI!Ph8KMms*ut;g(#B#orH!*T3FV18t(VfyepRILaEa7lw!#o4qoQEM;HBWzFP7Vj#(#&+J5Z(~Zez=kml$DHV5RPpCgxq#omo z#vnkzNbHNMOiT7wlNx@CoOI12C8%m^6`ccghPtby$_ev>fwS)teYNDW<#;K{hkXj= zdlNDZ;Q-ek|1xrMYUPC%v$bewj$vq&#bJy^{SeaqgGIdo@|7`FyL$2TCQpFtX?0?QGNU)5{5BiTbapEUjD;jX-*)lejkW*oBdW-Sgy89V2%k z?S+K#nOWbo93h{$1V9gNe_V|JPC;FyFj?m$`qIl~^S;>rTA!}c!WX3ky`y1Hh2Os_ zb_9QEekqEe>c)LA>y*CeaW(lXW4Db=7i3B=HGAtr9PPbCwHX$M4KwLEB=ko1)!S4! zj*A}&7T~zVrHKWitD`>siR-PLW}mkiy|GAlJ&11VYDOkO$h7asKG0mc5q5+77M*rdQjzV=`OjREqK*#^De+)K0C*xINl+6`i+QGYDZq4=OyZf z$Pn|z4b^j}Cb4ru=;v$W<(24INPqlx6lm)kR&j$;K8ky>d7Bp!NGu47ONAAOoSY5l zX-Tl4jRif^!-(=@GLjbTo{*H@jj1ztV36pEkNyv{aTbOov$D5uUI9!p6SLdE!qU6?1VaiL7=Pb|*gVPId(?;5#3wK0;}iX_GD~MWCHz0itrZax`dX6+U|- z$L`@ey0v+cdnhQ;w^7zhgj#GH&v~yp2o!0|bCK8N`R?T`Mhks_f^H@ywxlx84q(Rs zmiRAQf`t`#W!ekOjqd$^S|?e)^EGXgfE#g@ONBJ}lAgw(yZ(d6eO+xWicSbCbHqmr z;z6Ssn1#1`HS7L@oAxqW^RZeMS#nWu-9t;5#*c@NwULrhx_{WBpic z0*UBI4S>upUnF}AuM6WIc#<3=&{9anBm7(cQLpV!j(7-%^RZl5GH!EOTswjvCAJh0 z%odq<(9x%z7un4=$mL4uvY_ZWe&gW#N(5rRZT=W3dM5y2W2pJBRC0y1?>BO->4dDK z4I6IVL@N5%EuX?2! za+vj5a0f&N9*ZC9g_cw1Ra7-hy6d&YPgSPL?u*jeu|Cu2wxN05GK}tk;^IK~?K;1- zD*$B6MvJx~-!uodO*1v9%;R>GOlNK;5(Gy&rC0HLTw?e?dmxBOGo!trZIWJLh$`#k8q z89i&{M9&&XhHSh;KCb5Pb=YJq)ab((RNUM`fj%TBwU7V2wM5e8J*!WgRGe>1Kd!1E zmY>QX{hIo0MI701$$cH$9ZgcEiUMy#kRP>4s26kt_JM_qk=noW4s+fu_{a&|_Phne z_v*st=qCTUDBBLNC)Ezt>z9;pVf0vGXwW9$Q2t#4qWNS zL|?c_qGJczl6+ZwAFwG7bSU#`F>k)aYrFVSJ!fV+iV3R>E~g$L+OLjazxck~mp(%BHb*F2?I+-_V@x>zCIH0kS zMjD9_fP}!KLJ^Z0s+@XMz zY`hgD_#NFCc=ea3uP#LGXLn-7pxnlL$G@buY1LFV>m_cEFK4CI5{qo;YJA144WNmu@J79D9@m0alf>U~+ea>@=Ru_J z4o?ATy)u%%0FrA(@_&auwbkaSy19J{SAWQ{jc-jdBz8z6FMhsq$ywI#nbPA0rQ1;G z%MNd{_;)R7QCqdW%|@z{&KNKxvm~!1K=Nn)2#SMm1MiYtfJ55_J;H^c4fhHsRUw;| z*mAdHc71XObItBzN5!vP8+%q?WXo5kLD}wkhZ8qr9Ywk#dO%-1=sG@( zJ{xljp!>}}0Y}J3&m+kTaH32YYwRPk@fZ4wG&$7*6sCv#j|vvI;&0?tBYf`+_g3<8On>91+@tP#)0GuhKGt``6Q2wqOYQ*4CAfKavJ3XZs0O)KURijl zw=&IiX~yd|MG!P;d0@L6^@JJB+wwaXRicwlL1_q_=gyBhgbW3Bx_kB~xX{k#qS5_| z+HMjq3quzIen(EYNV1*F22&QELGw48N)#1$FDI?`;Zzy0eBI6;hsw znL$DZP=7~W&f>n?5q!oH~$Gv~C^L!B0A`}<3j5jn}F@7m9UMo?PgMt(df zm8qmK>ySs#fTmu56>k*P?IN-KRrruN73OkAxp5ZG@aMw!TfC_x;s%dFXhEEQcX*h~ zLINq^5nsFh3*3P=t3q9szf4X3GWy!9;kuWoGz~!A+N7__YPz&CT~VRwE6@l9dvVHM z4Jz#35ebbq_wg7vZ-a6OUVBvOlv{QD$c$+7pA=(7O3b**OdD|l61n`Nxq~|StJfkx z;Hg4;rL{K(Bf{k?haj~hc<{eX9&<`XKf2T2#rbnbuwdFt&x3RBJs~BLH(F2n<8d!3 z_Q<-;nMTu9co$IiCkHKMKI5@9fxJZ|Vjxls!j-zAbIIkO6=y?(@u|6}Hl)|MBOYZ5 zmb6ZxVJB;kp)z@m?vDV&?moAE)Ui13XGnOOnez8s5>M1b<)Nc@DYTcxz1i^N&(Om+ zo*h|YFMC5m%XVbN%ViXeAo@D4sGB>O={nfyVWbyzK|n+tHy_}YX>s!!A<8W~XfE=7 z^p3KU1zsV1r^KtlY$0-L(`vrhR!x&H=SCdmTN#(b!qPYAYri34$q~aupWl)@( z9OV?9r#Z*0wsiP5Gaj{HCcO};c3MbaWw#-rhx-Xg5i&r>N9P-XiFGhb$5}|~{LG*L z9}KBj^}rT%$ePO@PHr8GE8Bx{ozu;Inj?HF=fZL$%N}vB_;bg0>9M%?Qs6~EjJn!& zWeM&+h?<(?Io)#WLz6`!OUUy$Q^JGM80(I;JwSLS7*2#$Tx=a_kITD@dh{)6+7$63cOcl6O=RRD%6C)93jF z`-M>mJ5?56l!&GojLJ_{)~GqbH>k`!$QrI)@S@2Y4t0FJ)0E9AdR`x~#m@_@y3%ha z*@`KAPpO;qxon}QwcoMq-)7`1nC+!jP%rD6)09Q%X~=5LG3U8th3Jr;?VT-j5Fsqk zD++!eqpS)mYB7`&%8PE163nc|P2D0_HlXw;pAGZZzf{<9eKPY*dTE02^7sjb-lnj7kuf(!Q_!b#c}pg|g3d={X9fJtgt z$5ffUX?SXas4S>wSIZjyV$W4&=2IB^QfhMhnW93~&>eHo9m0oC0tRy})IU^b)ZOgc z`_KjJcl3?!mr=L#y--0|ooK0A(f)qJ*`z~a^8BTp@2cn&GtCEth1(jJ#*I2;e!)|h z{!`H!l~R+=^ds%&-sA9Bk5W%!ja4d0Hmhc@1ix8(I;{@u!U*E-a&)zpyJVi`Z^Wi_ z&>JVL@rt-?n*rix`+X>FU1Y?myx){deoPoJ3_TT7bA=r+Nea-w$ zb~TMl>zBsZf9X9c=6X(JipWLQyvi*9Z>~_x17uDP9SMwszvQt)TdguYE(KX`Qaic) zE#17~hs4o~|M!$2c#7Cs|#Up(~q^jwMAmG`d^q??So+cW9L`i;6Z zwLD3F>IKcnI(MW@Xk5PoG5EK_HZbpS^x-d+nH%`+)T?#Jzv zndWzL{jV>QDk}jYcnrp=1b{gU z!ER7&YWlTx=<}*#WEAtCeE51z&viV?{h35TeLbg+JB;~hLA^a_6cVNk0)8i$L3boR zZ+Ok=NIoqk3`l0aBXIjw0r)%pXp{+*H>$@HZPr^S%nX|Pn}f?Y9k*kyVm=~jHKIK= zVBV|r4;~fN`_|>Y96M(4gFVK+{Yn3rGps8qDX>wt|7@hu8e;SWBM$4T=16n|d`OM} zviPZA@Rs4O?P4QeWL@YChB051hAzqNCfagVtHTk%d z4X-F`PFHM?!-=-JXng>8PF70CV2azn(&(^|%9I!4ylR6{`(-70dYJ)Dg-;7?Qr$qF zSg4mdmM{cTIqjO+!jMJwPi8TJPpYf*R{1V;#B&ft)e&~2JETHLvuIqRl0v1njW=#O zuXsO!L*ZGK8IxXS>?^+LT!Z)kB%K4qWnNMt|UAFK~oQ>A+6|tB@ z-oHX`suw}bk3I$i(W+SPYy$EkB`)U^!WOo0E0&{-BI)#oxV6XQZ88{D_(lH{=VRDW zPw2lA2lj&GPQUC8ZGA9#s5$kkx)wD4{7#1@B7{#dy)1-&7xvHb6sVw2EVIow_nM-_Y$CwOodd}a5t*G0E52FiPG{t{66f^^lcTsWeQ{^LRdTmvBj@&X_i3zCHdW}ISh{A?c>Tz7nJE&3Q zP&~72pT}=`vr6t7B|tn*Z}sk*ho8(CMENJ>3JIk^32JMr+<^JMwMl`E_v5T#OHDd( zfN%BR_&#hAuMT^06v$8k4i>_%F~1N4a}Aha*-#1ZtJYeoaIOl7)s&e?ka%{K0aFI1D6I8Rb|xju>NU*f(Ev}z&?zi=~ROq zB`D=qIX;8#8pLxAqgP(Jl08@8F`okx5aUJb(*@AKo_|&)dpw^w04Hio$9S58jz#Q5 zuE`)^6Pf4pn6~jM_wA{*z7w|AB?FhEuVv2LY~v-ta{KFPExnFwi7<$793FYolpFku zYO=+2T}z6pACbLSF=71}t8hK__d#$jzeihwq=C6?JwHpB+15D4SbydeTD@prWKEEN zvg|9N)UQZKn831udZ2leU^j7g6-*#gu85n+%A%piZaOy`Kp*$Qgq2^Z z!MOXOKG^E>vC2FK8d$75{uod5cKP!@(aKwwmi18L$lg%GARhDJG~%O40cby;@7SL5 z5gIINkHco@(lT`{2uT0@*ZAg-jtCFLR&YxLOTEE=T{%lNYu52>FgE37DUK~aoRl!f zQ#-hzCdw;RA(>Yd7lzLJWng|J7-bfVxv^$ypUW>msdPsVJdUaoLG#xxki<>guaoI7cb|mRA+lT&qtI+y{u~Iv&u;(J116Qo-eSh?6gS_wmLnH)DM7EkDWt& z$d2c%yD^FRLcCkiVUA>!k_*(`ES>7M9u@8`?cH&@(5uf)xYZ0GB|w94p)jn4JZS|z*aLWadY1zYAB5Hn@It?yJouzD{fluMM<0LQ4xWA4$ynNze$w?#ahyCE6Ggi z0}$D2HnY$#Hh$pEcX2lnSFY|id>_7Aj{ip33bNW>Q7M>Jp?b&8l_9f9uKa4AJ!^vnw>^@4E$e%RLT zMG3w8&zMV%BzF$_DM_b@ogkC~d4~``q|PLz;+)Ao!T*}BRV=ilr0J#f_*N$Y{d9XG zNQCmTH1xdOR+!&T5i=EB4yu(A>w4^7k^ak^?se)t9Ya`(p-V(LHNV^$O}a zEckU2YRlF8LJAh!3#wzZ;QJeIIZU$*#szTwCi$*>KkPncL$AF`Dmq53ZaEHG-y>$W z*TZD4Ri?;3xN^u3Mrz$!VL)L=;%-YNWw+(RNb|mXLH51&&iQt(UGbh(bWsM;@)-cU zxDh3oNX)g+-vHn1hwbrQwjmi~+`=gMQvuKmy^zHlgv*)Egi?AVn;sPq)`8oHCRnL-&E5Lj)o_mXsi`Ptl@ zP||a*zP<`4XlgU)XKi63Ou&^iUUkOv@I=l5rL8 z*_RN&2HR^c1wxpFDPgU}!ZGE7dRl=d!+clr9lcdwIBfnYLW-iUsEmNg*U7n%bCTCO zYa1o^Fat>6oFS(-&vC#THPYm^@ZaR_$e~k%R%N`Np6SvgaK}Te`jBHnW;th3JpT|Q9@|i}y<@ZXWRry!ptBxpi#!!ug;UP6| zv3HOzZD{ytb{S+3Cy*${dA5ZPJ4g7iauVkJVUUp|+518SknJ?22{(zg+T$hi& z03e5F;%%~q`IcLsXASRQa8FeJHL5$o(gx3*-f#TK$&z5ktnWm21Y$E|_Q}UOd>aa8 zkZ!^3jjZ8eM2Yz6F!d;x7q_PoKlqRE!WhxJlnbA8dudTsuTCQ$_>q2ARB%}Coy*LQ z-l{eu4NBhcG`W#F&_e%AED6>CZc%4-^-@9l3sia|!%(0LmIHGvO^2a5k_IPyK`VTc|A4fU; z(#X@D)GiOa4U^l~f(gSU4%?eE7GiU$-z#fjStF&^VyZjPVxV`Gz$XZ`E4>&-hmgp;_8Q-Y1Ga+Xq>GQE*i2hNirpk+12^E=?oi6GR|yKZ!!Z z)~OpRJD73!sRi?Oa*TE89;RKG*P|BXb0az>TlhNtY#-%3S}igS_4nZkiBNO#4q14eD^ts!|Zt9rTsk16=d>CT|)+K0Oj z$cB9Q-w1~^i=P|xQDY(lh;!-O)BT7)ZXb?xQ&(SLM}lb`K_yuwx0Tg&3zA)Q*pM-U z*!gvU46WTwv-Q`x`6Xsj5Ic7_^z!!&b|Xan2cA^*BHvT6vwXVHr?DuuCr5bc&5=1w z>WA=@t`Rg~fKn=4uBn(_VpbERvW_pDGG}D+9mMh3^R^(uNdTOeO(%oz} zbwogv+|?Wk#HN_$Ju09S0DWp-tH&fsyGBwAD18OghI8{|MxRDRo5|clNTv+|Z7SMg zpsq;k^hE&s=7yY@>#y8i4_#Vsa*5U7oLMnKF7*$d<5(bDkfYP_=^i7(Vg3LQaKVaPmib_qmKk=sY4NtsOlz@ou$MHOjZ_4jT;XuI@AxST>ePhzzua`owy&u}KVBjq$ ze#W(B7xMz^_8tm924keYTvwB0TJW1SIBn*p@^Swz9B~7apH*8)bb)HX(sohFHla` zlywqLe7#0A#cwrccAqIBGDo0iDvrg)=N8*Q3V$D?CxEjp3egGY!1I0w>Ex`W6eFo_ zZn;CM8)CA{0{KB92xy2Eap^^7SBLRMIY?E*uc~ThY30+g{UV z+E@l%opoFuHM>NWbn#<Jgq0gFMYQsf&f) zL;9-@{_n#%ViZSA-?hOj>^PqTArr<1TmN44n;SvY5<%#9G_n#$_&xh`7WV7~^#z** zK4;Rrs&O)C5>Mlko<$xyj<9j$u~90xR_}LH+=KQolk)*3GL=;v+7BNxeybW|e%FzK z1@+6b&lCj-gt`i6g|67~AohyuwZ;+t{*bWms zz<)%#QIkvVnFLkoOzBNj3x2RRoP&lN_@)#&*fDY3x(Bk= zyEJ>hf!EkOmt;jbgiVVwO(fyYyk_|&%8NgI9|-C4K64m!XXo5bEM?Wm)$Qts6tjh& zcjqv~Vnbcp%9LcoNJ4sK;Sv6zdw_kQk;jm4HMf_e{9H0+#Lauu1WNw*MgIEAv@myV z$z$@#L7>_65b`%@yRb4%b0rTwoyOe`ltuRj?C1Vcj9YD2 z;7rE{gHhfk)?^|dE+og(5Xu`f*WR2?b06}03@rlVczYOKf`EVjq1sp7eCm~HwYonK zETHHKljksIsuC`o6ZSy8MOfr^LGoB8&1AMh#A^ru>c*b`vuff#tv7~naD07;CG^NP z_yZwqPBD0Fjy2?5IMsGhIjVqSEKkxswo(I0@Hgqnje~#Dn$Kwx zvles<-I|-{X@9Ps7#N{D&m+I4XuN}QG#H>bV0!Z~u~`2pT~3W+vW_N)jgB$UMeWK@ z_wy&0Xbz*NFu4qyCyi#`zDWykSC_-njWma{i^%8@PSPWCZSZnlhgZOHNi=m)i14v% zFxpmc?B4tO1TPTKS$DTU4sS4|6iRrl&W7b3@teB&kb-dz(DM zy2^3sO_Rx`-7qlMoY)c-VZ(YOuH-ziZc!fZqVQffPA?)K=sK@nJgp@E{PS^)^|zFc zj$H^RJps{_sX|}`ogar%^@4^_w%Gq05tNNw)lm(`*}x z0Q*i(elN1p*d?>Ky8HLmMnWOhxSv5h(|E9{9%Ih-leuOX4Pz$F|12PnLO##hY=(yJ zw~RVD4jJ_33nB6uAs(EQOZ3%UWf)4rd#?czP>~aWIXZ|m`(vf`mcv%p`PANK=aV}? z8#~}QCVZ@LvH6s$*0Z-s4_=A8EJfb`eWjzzLU*# z$K+GKZZR7`AQHG4=qP#Gr!tE|^kSCBN{R(hLU5~w>rODQ5WWkem}>vX$ZT7r zyT>Vggpc7V;VbD9Q;p(T$-CPL8$x%gplG)+;t!$MCXL@xAd!e&Vx93dSo*1wG?7Yz zN5o5dHqxU2kny?w(Xb{iPj`DM8MPR%<}j|H4Q(b>RvYJLgq0z&Qt-+)R?Ine1O4oy zBh1hK-UI4+h>!Ip29D1S&ZWSSQT>fn?z@(zP-j%Ajz>m#3II5!SwT~Xk9-^yoYV)( z+vCCWs;;Wd@Gnc9g4yoj|2AWKm3|{+K2Qn{?iW9rHSz^V^r7QMsUE)m|M>@9$JZ;# zR#PMB-I}Fqhy$j^!_MhU)TDl{1Wz!@hVgguw%|;<52KFhOf?aZq3u#h?t?3%+#K_qRqkp`i>Ap$R zvF~lxa1+}fJ~W9b)@3fG)!lY}dT$Q%rTkp|OFWQMt{S=E(^qI9fM}6r7ss@ww4l4L zWXTHsL~Tlr;{E|U!<0uKFN%xwq1wc2lfX*)MNmbQWDJ96(F3efM{vvlCz${h$KD>bNf$a-hTdhs#4KT|bINnOU>!({E! zwQ;_svg5xVzpc8d!{qeKD>KRM?>k8~527~bUj)C@=8~8`Q$s;L3zw;W zFpO5E+-9wy=vP_z{wak1rI(y3-;sCJteaQA4CHyLT6v#>{Uos1Mn_yo-||w!-JL|S zi^7}X#SpLbLwqu3VoRk3K0*;lL)l8H?b*Lc6QJO{^CC0ni)uS$Qcx!~97K|QXeQEa zwfnn_6@E51ZG0rY;=qX{`N+4=sc(HW9Lwt zgmxJgTLe>c-w&U|RX@zSM%0Wc@wXfm!Rx5N{GaI#Ah9 zY|>l8VCj-~*;J*I_9srb*kbK`!~)SmO0AqAR?Y2v-5=wVliZC`_z0-wqwE#yew4$T zy)$XdH#zXL$cHK6_w-WunsvKyh_kD3@y_d!+p6uJ-49CUi$-9TAHwc^_?ISlnmFuK zj0VH2pYu_5`=LWzSxsVJaaG}HezFkJyGT|}sO0nh#Qoo`y2^D-T>Ml0KP`5qrlb{E z|7fh(cN4VPgS;E{vuR=f%kaoTcS5PXJx6k9*t*8*ewtdkIbbpR*?^nC` zXGN8g<+ox^{rXdS%uv^cD&aWe_wWsN`1COP@<7m`fK}ad!YT-M1M6;L_ot5htCrdyly@^h|CsACWUFhH z(&E>U-|D8|_6hiE;#*LoB)W+nEf)^=5zne<4ScK+Jyy4PTjny~fu ze>jsa>4CMc%e+{p-`tZEeZK;<^0W-oeao^b%t4yoBe}8&Bua|3+d0@JtkZ6OurvnZ z)_#5Y+XKEb)0R_-Qwxc(U)(<~Jxh;JENY!XX(@Y8L|+s zk@zmVQhdsmo5J+?+m_Dp;+ZEu{#ht^`9?&Z+h-xZ9-Mpo;gI=Xnr_Ofm)(_hSyJ#% zT|f=?W{@R^VwZA1z02gKlp+0(WttDB+B#)AGq6i$TRI^O>ATxpA@jSe8&9`~k;Qx& zg7agc&-H5j-!`ef2U=n=!PiwB#Lho_^$;xU+ZY${qpzv5OvBc32eiLqHO_78L_o<;d=tXs}s@S)c+wC?oUcujnNH{G|%^X4>y6#woNHB-rDD6MkKTk9;F z``g2Dv6a+huT|%ubV=9jPAR|^I(PFI0zdnq-2N%e_w5+wZru50FnUxR`JzJ>keD*_ z*X)(Wos*`n;72t_t>O9=LSaO^-&Gd4k<&w5MKz(T7i9%P^H05^%WFNx=V;T-P<>na zTPV*fThH0sPt&=Dd0w~oA*I78WtBX>l>6a#aw#QAiMIyG1F3#r^10$j0~?tyIp=V7 z6cQ3HD2ySO^xwxty}-||OX&WvW`d;h&+>>ah5#DfKzPs7x_L3#NL`fFi_lZo*Jabn z)pgZ(rPO2#A(8s)a>SbmvVh3cwzPv7>AK|Y&}3Tp+ne)F`1+B>qQR6*d*pe^bjd=9 zH~5e^Th1Msx%78#rQDdkw|~=hQNT|Q+d^WY`j4RDW#=0@Y%wBobE@B~ukLf@{@q^L zB1?BF~0T{&JCb>S*Iit1QzyC;5ZnPMM1k zX=}Ff#Y86tN@lzS{VLqi@S#uPy(S>2b>(a^>Y?ujHg3IsBoyd0esmTB znt5VBQ6Un)Cx5{sQdu)`M&y!_Zys)q^Mo8Gt-4TH$Y8on7+wV8-zvUls>h15YQ+19$yC3=AXJ{Q3#Co-V^y=$?@lSE=uh~kn zKE&6rrWZ~=j`@+ihk3Bs1;nvklJU-eWl>oova26f(ZSvuFPluVWXF#6XmG0Y7*Fr0 zW_anR=Z`y3KSbw?8ZBb!T5#)~lieO}d+z`ZwBUU2Ys5BLTn6l;+;kWv`9K7GhC?;s zn=!Y-L87S>b1N$Qql?*p!hliHFR~%~bCHR-r)PR~4SHcTcI2Fsx!}N;xws82jZgm5 zdMMY~j1-{kZ~7*_P^J*-$*VdGYTWvUqGj26^?njhre{nD`t0~$N)oezQrfui4SAqe z=Tk7;yU>N`#jta!#99( z=uSvv0U4#G@Eh?rXB8av#`(_`NBxAte_M2G6s*D>)&XFF0~eC+JMy5z8Kf#Dg$mVj zTYlV6aVE6vEj&;7pV8kKO%0lonZtJ6n!hziHtpK1m`a-0LpzWRFHhuVYf6iA;pj=iOL^+tau+>2*Ta=DkEr)71k4_ zxuRW(oZ^LYpmk;HLl0B~NPEbr?KIQo2%y(GoId$q5}t64@nQBhzM|76U9!vSeR_-N zUKo~}momyf@*Cvgc#G!Ooh<<-KY%W)@O{VEg_lq0KpLFl&p`8 z+UMdv|7yz`Frcd&KEJmC=N}5ar?di>3b_suHy;sNq|}fgaO9@s;tw4aGZ#GYHggyl z+0+HR$@5-O)mf+mHpFF^e11)c9_nZvL+Q-&!~~VtNKtgcSK(MEwgF}5=pOZLi6c0g zKp#_`=qv@bhrhfQ>#UEev?(oA0QYctgS`Z6aZM_Wy$C1GcNzNi1!2wJ=gKD+!q!*B z>>M$~1dxp3=y(R9lA1m_ld;gW^( z&)3#Z{xO?O>wOrS;=G6xaNhYH))Qs?=ln_}ckyGXFM8CG!kpj)rN4rR{o>jZBma*iNVs0I?pJuA`)@r}~ z1B7BhepPnhh6DzEMel|$!D;ztQwq0ZwqFO1*l8QSpKLyziA&!M+~#ti24+az*E7R< zsM$?@=@~9X6>jU?`V(xY-PtjQlHq1&N|!|9%}}BTJt}4+=(dpmY!%%Nn?KRZgK<|) zqd%(IYh&F88recuF6G`PQ6-_(hM!gKm8@*5qyFi72yyhrrA;}8IEvXZLQ>bXK}a5) z-B;eiFgYK>?{?S zwuQ-)?E=#7^LLWY*qazW3NP(!nCML#Xwe#n|lN?AoO;c}!#@Ncbe5$f!1)c5UE zxfvP?T7gBV@oQ*RZ-#}&!AgrF$rFn?bkw3NKYAyvpf47c-qP2TagJ4WFU?f=JeaV> zyE7_sQYx5f?1(`A%~IgjYByGM>CA5+jabMQj7}hymK6oTYEK{Gn>2I};e;Q_&b@nf zw|o1M4q|C?LR5gF+}1h!&*n|-<;|+11!1noEefe31)XdaFPiP8MzraKZ=y2pP-6V# zj_^rzOq#7`gTKFcY=o(4ADxGNiqHp)b@@^&mBgF;(ANX{%^A4nlns@|*v*i&=c|3KBjYUWGqB2+h%* zS%n+7HDx0jN8kUD+N3S#ZtZdH9|E#BpAE->*CG#Z>c|tH4Qz7al|LU8hyG7 z3H&$?%5<`zE4vNkhzoJ53$ngV}qY{be#a+prv2-l-CnoryG1f^wbbiLsb!rG5 zNzC_qw2r%X>PJ7-Jg%Def~60AqrU+0Be+tHfn~sw&mJRvFJ&ZVai)Z4R`$7r;5Ob% z*0yKgT5=mdx74|H8G2>flnLa+p2_;Giv&Vcp*p_wp{m2lsqydLrtcO5ot;!llFKIH z5BCj&Gyg2HUw^oO*gl+`k%lFTaJK;{BbD7_kKJ-q&t7iN~3W$sliR@ z{)2%hKQsLgQ+|B^kK)^0dC7Z=vDeuE`g&gNu$T!FiCzB^t_XZc7;7AhaEZPfkza|W z|Dpd6j-Tqf9O464&+=(q>Yn(%jhlsjHMJYBt|%_RSKl_gTdFBN#!F(8aYr9%T|3d1 z{k>E{XzBAm#J)jU@&^F;m-;4&stx53ANh!~>nIWQ2C&8Y^_pI{nfFYrS-30>Ng9_! zSe-T?sr2Jn8~9-_b>aj+K$)xX6lj+nHX`tHVP-0p$bDvXer`{2pSV;U z&3Ac6|EcYNi!7iR-*+ay+vnQ2BVYi%DEw{!z$9Gy*Z5JhZ%p4O5H*RcsaNM@^?6Z< z$V5eFl0JJtZCLxOr60`f@_0^)2-E$es5L0b#e(#QzXJo+Ny zjiJ5QD~(DRrVUvUOvZ`Tg^mVme7$fOy$LkRZ5D|H`CC2n3u^h$McZL?TgUOwh8fM3 z#;K|;7KZ0zD3dLGNbFKdv5F$u*l$Y`WNP7DeYbYdVpkXH0*zNGNfv&W>UTclj)HHDL{Tc4JVzv!6O!d)jsvJNcX)bwjtc7OIVI&3p&6j>W z^6qPL`2G~1f>aB*C z2Wr3ImJ;oe#K_>69FV-RgIhso(wHi=i0wpAncNEdAKKnL8p=2R8xKjDLJ1MsitHh~ z3=&0_vSm;9Ee4TwEFrSXmSspn*|J5Jk$oRql%2uY_idP&`?)^fKc46Oe!uhl`8#L+ zaE=-7x$oni%CLfKGuYsmW>d@sd1v+Q4qNsa62o^)0O zPFC~$*3fN90*CirU#9ODQwFuqJ*+=<3NR5AT)S~fT$SrHS6vMosv(^Heb0IUi{(*z z$}6_&tf}M=tbg>xI>%#uixFx>| z$v9JiZ5~KQ@2%0FS>|!CusJj7V_^8~y+Lfg(ZY~d7mi|;6T!nXtnKa-MeL@Q!x(%bmWc_+#xPc7dK>qpLvPN~3g3%HAZWUEYew{S0=|nUv9GIVd{wIz;qFj`- zCQyJ7(58H0XaHd=TooSBqyG+tfk^l?U%G}dj1(qU<4#{jBq0Sy7v2pKMS=(81c>7k8Jlr-T?P47^ zI$kg4Mt560@}(L7791_Lh5w5%qFQnw_1e+oLkme_r^kaI36hnrEmK;_i-u-$Q;iRI z;`+&r-Y$m_l8c~o-)nDVNjG#pcrae_5I}`BE;>8QT$K@&9P-$HNiKF7sx#og8vDF| zbJ~6(ur^RuvzuF7ijrrFvp~g$FEdd>*+y#`+x2n;7=i{n<=JXgvkSqe*<3s9r1mfT zvO6u|t#w%KE9z)AUO-VhHnGWhJ~5ySL%P?612s>y`;OGg>2!icu}{o}P649dj}tWo z<#~2j)L|V~Z#iVdA_z)q#GNf$m9Q)LPEG3{qg&*U`v_0~G&6Sb1&z0VF}(B4B1c|8 zh8mjMbBxN&X9XE)^@+5jw*^CLiW$+PcMW~0pRSaZ=R*TG@oIMcIpk66ah@b<_xpVx zpY&6xSD8j+4D`y%;kV;+ac-LUWElh~zNSnpiDh#@kmWWkF}I<6WtK7?OxZSA$;J91 z1b2pnK`Zh?EHAfQ7xGqtOXI_n&gKF)%!A)BSr6T1y%_N96_R9T=-wtgVgWdy*x%*x zBw~gD_f8!CBDf9iZ&54Vz%}IzLdPThHKsLnA@ zhlrE9^Zb(SYcCH)J0M=lE$`9u6}Piuk>@vXz7X4_RN^)B5@) zBtF~4I$`E}+$-*7`{l8|trh`X_S%uF;{XaNk&1R(IHFb@VbD_P3R$! zX~#xP44g@4ultn`ENC{xL4`I~uXJQ)2K<}&n4)oRZy5EfaBi&5M(;eATLOJLH*VO* z!nZ@_As9``B6@5#N>GTg&-U0UzC zAhnOA^~1)!Uf$NuzM{om22w7B(KSU>L72xJ+)*SVMcs?_S*bS*LE(^K0kSSU9jj4R z7I}FOMddml=qkpvco5i3ktEY(s9^3#16|*$OGT8_OkcrFMH@iDd#ncOFfHrrsdYLz zQ2ya|rA@MAw%-GdQRq}HyA?Q;ZBO34xC?ZgxQsI2EZgq?IAyFR^~MMSxp-*A`-uST z00NJY(hnpq&DoMl#4{eH!PY`Z{n%IM!y#%y%ISOK!ol{=r#^&C@*4pIC^-(wd<>yz z3wkAFJ?7nOYCM-i6g$Yk+UN1$90^_OW3`Yn{kc`SS~pI9gn33lQro^}OqGcxKSa?% zyn-J5&moCu0&X|rImoUrL->(&*1@-1MmHD+Akc!We|uvj+s|+Dr@#bMfR&H6KiFop z4c|xlgSEM(=`$kjyR~#~Bnd=10J=#1pmPqo-h!mTO!>3$c&uT11y}lZM{8*82_u0# z^A0M@Z$SwP_^w@11RR5Cx!vzE0uF1aI)pwp{;oDkf|Dl>x8UJlq3ZOXg`ndL;Q)WQ zfW;V?PQQIm5$eT9+9%c@208tM)PsIHNJF^f>`^o{U}8E;CxcES4^+91AxZ2Qs^$pgQBjR3lBFm#04d!V~wH1NK+Nk{_*$k zxMSAX4y0B?C;?^S{b@2Pjf%Tm+b%77irs% z4D2laKNr`#PBABT!R^44@@cZA4KNb(c70aGq=o;LQUr~zFTLLi;}(cnCO1^gK~cQ9 z&vf!(#n>$rSyL4I;=XYCIk{=*+Dk)v1$WD@*(;Yrkza+wyJt%oheLj5e=hX(!B5T~ zvfunhs`Nj35CT^|DR_8IOqC8QPWr&Wq`HGSrDIzD*rOY307drA0@!-TUQkyUbA5&3 zb>Tjwb!DV6^dol-*oQEF+jMG!KS-L->LP5Kx$jCu9A6}LC+jk=!_B@5svMlgJ*Cly z{{3u^iKb)v3OYaiVLq-=E%0!d$FnY2LHailCLHC7CI1Gg4bIN^nR0tC^BdIgnY`zW zk1#?!oSDZ@R2{hDfI4lbr+N_E<#o8X7zw+9;;8S<{;AC zxM#Eou@Vy|)3%Tj!9=B6rr2HbVpR0G>bY^KKvf48XvJG6>iAnuGrAcb()kyf*iebn zAr@p8OH&y^=~GkG?!hJ(-_58CowQ-y?ZWnuD`9y%arDgYi=P?SaMjQDer|8hDWXLf zNcdMxm(y^RcTd2TIhId85<1C>2{O#{BeN|cjqg}UUb_FBScawv$bB)gNPhfDpq{C& zSuT8J4cw5-uC0rqNMptp%3Jw>2}!e^aC0A^xV5vjR2du3Su{jW3p@Tkt~$SXKzdgS zU3_;$4-xPW#eNOLcFKt4??Q-WnU_)A)KCEjX|*QGT-w95%Qx9vTJyZRz0fE!agLXi zuKO_S>wTtLqTmgo(MkD^4A4?|8m!tk3#>pxS@|N73Lh52#7#H1%nTDD@p1ZP%DYmw zC-4xjaOpwhqF z@U}PauPC=d2rjiCx@w&mze{do#4>TY(u$lDcoP4u#&NPFbcyUc%2NxIuEpCVnMu=~ zI2e;fD`8Vx{zeCByoQN#Qx6Dqk)^8}!^}13UIONY@XdEJmsDZ;yeP&VU}@@RImJAN z6y86NhyN6?|5DdnRuig8!lYXi`A8B)UN-f~^;0(G%L0*hTBGdyBxb?-v{J}FxDNc;3Ww z*z@hzz{l9&$$w3jbQFs@G<1U7S3`wq0!DzP3Ir=%T_pF0@hwS`G7j12G{}ID--ekX zTL;YUVMO_9T*>~=jINCQMmK|JnsyCt$`oGa6GrpwIob(8r#)Ti>^;)B=cu$wM}QZt zAr<@r3~0hG+$XTl>4-%$n$ELJgkF`EWzC?Bv>Lcji0)U`tlDa1$NhaeE6*IN=6cV`4fd{-D%U+7-=F&pT zY@q*9J$D?ySNq6szTs}I4MpQQKbY!sU@5fIXc~H*R5bYQk-cI+^dyL_1zxa3MhUpp z$hI)1UtoL8E6L*FyrjFsutE8Oc&33jAZ!4C9@~Ki1Y64{wTON}sn*HZYfd(Ddu<0u{>) zRJo^|&3TXWFt!B#Gs8&tWyqNF^c{P<$tEp>k)epDqx0t>drBW2*sTNn2vOv>koyHI zVuA1ezk9|z|J>Wa{AlvbFJc_I;OM!bBvdb*@#V~d^I<6ndnVWDy2`?y8}CH)7!+njIs48}Rp`)IW~_u& z;vN33rcUeBj5XhGUtck2{HONK`?jl^A6fXV zdWLRjn`Jxl{r42AO(d|_G`@fw=-hoHwYF~O79opD088Uh`whFsKzL0q<|@w zAV^h07UBDwk?`>E7horlLjyO31vU|IPCbCh?bpy?X0*UQ=JMZP@)gF$3J{VgpE5?HL4Qc(FC(1fyuP7knLoukA_g)T|NeL$J|d> z9AJN-idXdWX1;m)|E>`#lV7WQ6j6%QG35a3E}s9gnn3!yLMON)jg5|n+wbO3xoj(x z`3>>x=;h#ekJx3nVK`fEn3%cr{!;685`HZ#V{ad6ko@n*51_<|#O9HP_QxS!W4SL4iA*GXBBgVXh?8)$S`d#J5)K*edoTOGAQKMZ-7pnf8u_GdW54T?$(N zHn>HfMr;(a<-fOVhM8Uo27=?e=3SP?g{J$ln$OIGs+-8O`GE_01yENGC*b&UQzaXK z)31=8hwuDdp;4Akl*~Xq_3Ha@WHciIvtr)$Y!^dC+pCZ-p|+A8(QbI{3}&9ndJapdqN^Fo%qBcQgPRn5Tz7WZbrCO;t0 zEU4t@`8oBEqnOkx_@1&e;`%?x`1XxR^B~uQD?Xe!ubS4zbNY-b>vwQGGVzrt;~kH> zon~vL+qWMrK>)j?S6XOxKAP0qVmthzn&~CQZ@6%qfde$QSC%r<)iKmNicEf}`d_oN z73uYv%7}UqDz1Z99c4oJAjYlGjL!w<#gCVV!CvZkt1GOLm z?hHNeXx(rwIM`Vc^yF7e%XJPniH46ju0gC9 zanRj$%gwOM4Y#}0{Wxh~z0o;FpO;2IQm4Vg&mPX`$sk!9cyiX@rN&`9$h}k0GQ?xF zE`yJCb>1XU$SYTYlnp^aqq+BQ5lW-I>2)$_o!N)~oo)^Vf8&s`Jhu*4TkQ2FdjZKb z0nJH+B#9c@pN5=Ql6ySDZq1`-J!8yP%_jqylbxk%Aw_Rn&J+A*b$u z{3BJBJQu9RZb#c{Zqtouc%6iaM?^szUQ0A`c?}nwiJDO-AzxVx^G#3{iH{{D${7nOe`^X3E1&NMq9|gJWUJNuN8z zHGnXiC~t31n@{NH*K266ZV8X;5D;oPzj<4e*r}`@bB|Pc8RVqg!C`_|8sD57IO>6Q zk~TBwIQ*Q-<=S5dDhKAICE>oky#lE<{k=Md{MI6z&S=_{)&bvZpUUL1>1qG+#hVCk zHmcKKQkjio51;v0@J$-4Bg$9)@7oMthm>|o_P;NagD^Y$%flLDP$EW$Sf)O0y(wxQ zBsV|B7S~2iR(x~Qi<0r3)^<~-vow~!eel^!^blQMrG^@DZ6 zi<^{>MHc5VBbpB4c({Jndl#mI7N3vTxSZVNlth4Ji6q( zkJSp@^x2sNKGoT3+Pgg?|@%9_ix7RpvB0SqVB#iq)$`h zybW62ug8?cy8esyrBgHPC-Z3i;p=Eiw0vGTtz%XrCsq5Uen21psN)7tuYt0W>A!J?VA+i`i*3oTe|l+3 zAa|~O@u!S!JuBuxZSY)g!Az*Q&nKFxBCOU;#JY}F7b4?lOblaKX`YB32)USb5m3s~ zx<6Di?d4yb+G;jA2Ch%C&!~Z)??uO`Ie4pOMfqZ2t*C=YCQt4}bPGJ@)_b)eR+XeM zoB2O17YUDusb2rZp1A<+w~)qbdtg(de?E=T_5_n#4}BuwMw&4VX#^DU&uC1Sthp12 z65W&y=k`x9?i(#5JIjOI~DW=5yo`B1A)*cls{(NeIvDm)* z9Lhhnn1JCaWuhHO``#5Xpwug~xjACqRW`MzbL!oX95bQ|hjbPCU}}DxNCEN3yGSj= zuGMu#XUaVk+>;XsGsvnlj$yyu)E3DUomG2FRsrd+E>h3{-pYHF4Mz!dst!29I2UHX zYSq3qp27p*O<7!>KOPhPz^VBSqO8&m=_eUt^i||QgyA3dOVx$UX=Vfjv6L^)B#loT z!0?i{uHT7-(Fsx|nU^4E`H(yjy0(@QPq<|&yXhC9cgx7l>|*F#|Ke9*8M-4VRpVO4 zM6iaQda%h}n#yY?u@zy>fMVklVNDIAR z&n#LWN~EtC<%F12G39V@Wea)|adh`0KXly#zQXO1gUD_Rur3CTtT-zb<4j(#Z&BWb ze2$96wLnitdcxgh>_lyBdC&U3lZaxlDw6@!-HBwJ0*_}v1XY$;cwoUu-VlCb*Dg^m z^{b`9Z4n!ZEXA}+{w5hZ7l4X`+}aqr6BHf4yptXC`nH}+?x{se>OLriUxJP&ihz1p zkG~2e6t6Itfe;+o-@b~TTRA@5o3Criu+{UZzm^gExV^IN<|2HSA++lKX`FKc3-r#* zsvlxL?|IAtjJ}#~K0dP~;*QjJejI1dG@NmQ_J*#o$H2To+8SR2L}{2(3cWLzErDb& zp1{_T&oUD0+!|b-LHh5R+{eU@(6MTB>{_Mo1qW}rSUN!{;{+qR@m>F;?n;B`Z_tzz z=84BE;b0&{;5x&5yr@LI4tKHJd4+}O$(dNd^jPZV1t@x6z!qc;6iOs*Q}qwm�Vy z-IbkRmNdLtz0=zw7U%qK4Bac4c3Bsy%M=IN6a!+RG<$)PPpe9tfBGY{@6tGX9RD4r zmc06Tm)w)PTqU4Ap7fxBWHr;rxq-|#dCN5gsgaUS%Up;{o_*>ABq2A2>lTj?7z=w` zu~mg4Y!mr=*odxP87IF9aod##tszBWo-E)aMmgkucU@U3$|~S)7Afa^T(>VN+#uFZ z#Q`8w;0?geE&TeV9Afx3?43_qcaf@K{C$9CWSWEbUdh5wG8d>^P>{At zWnU#=Xu)KP?itbP*YfB>Z=5+458b^sHw7}kvz>bXI?GEl4mBL_HiULD!SnF4+>DxC z^2?3ij3EVOuiVcka5o|LU+cAV)1ndIZMF5xgJ|WPb~hdoJXE!cOkn%uqnLp?b{VuV z9~tCRyWPPn@+MZ+YtN#oCbp%ryyeuwRyZE=SEnfwfyEA=;AjHhb-{anz?g%5fuk44v?b z!MNfeZsa2fWtMvd{+VQTg26_$!rkyIVW#7k>dipUA6!EpM;7JKV;jr5!4F0&0)C94 z$VCj;yRoK_kP5_bd1lzWODma&ZW+?OO$zRPOD22ueHVFHe&wL{YCU9U-W8`nVX}+N zuH~u=ub4UeAw@)fRE|Bt;U$ME%cBm+{dKUYA3QMSGw8Odd6WCABCPrU{WSx{V$HJ` zjAPk&0u;T9U>;LW;-PG*F2#xQ2*0wMW+H&%Y;#6{&bh_F3o!GCL?W3Ll0 zkxryI4B=uJ|DazUQ7XNW$fwf&W(jR`q2K0oxR9b)$-X_|rcA}fA%yGNEii@YT$)m$ zqPg2XH0n{1JjK6QJN|`^;E@|Eu8L?Byi{Xg8u#+i$lX#)^kxgNqYkutN99yho9(I0 zag|}5=}y;}%W!I7YMn}FIxi{3`zc>3_LBr_b7^!3Tq~NsDsO~0LOA|U^s0v4BH0k4 zEKHt*>C+%$d%R@N$a8A4QFSoZHdDSgs?avGDxBzT89kX;Ai_D+8QYsJeIg9*=4Gh) z)GXhXeTKUr8{BU3=p-p0yHAK~w6kF{Tg)AsdNu8s^K!yWKd{OjoAf`lTY0cvc>Pop z4y>07x@PaD^&rC4i;ebar;XclVKLQV#P;>cw>$yojx>bR4uU7;|32+(RM-X(g3d;v z%4DwR1tMm*Gg4M-FDi~QL_9s(czPN{KeEKYO~UU*`jkf^*CdlRvrg<R2CE}G>LmOP2pJ50g8kL7pUe$2Y^G=HOGqLRwP8mla{08+4X!R^jRIsS)i>I%D z-Ro?$*{Ge;ddQF`U9mF=zDd4Qhl`n)dX$C*{{SNH#LjAa*?QQ%y*i9ZcnX2J_qvQj zb1ud31w^*h1U|e9;FkYTn%A?M^k|rwmTT>j5W_f$vAr#W=%hc$bKE2ZvB0gvr3gC% zHDS%W5HENh;V*zwQ-b=72t!Y8%vNoxA*A$5au?He{x|$f^C&O2O__&@-waA{*?8K` zdaT#^Xa@sjS)5vn#`=YQ73n_x%rKrjUvKgtY|7F)Q+{2=b{G6Y?Df73%^9%w&7VUp zupM8<-j0}B}$Cl3`x#d_KJgwVW{eh5hq}|MuMDCjJp#?RrXR zr%vy+xOdtdO4{y2#B&k@NA4(S1}oOT`tJf3eaY$irTM5VT>c;oWBu}yvC)%Z>M`pt z8D!XH1NbgPdPKdMssv5yH=DHHCZ)1N+soAA9q@y_lso?`T3z8j%PjI+llMr zE2({b<({c*%S3RguOT=n5M zMl7mcIOHlJ(W92(lIKL~#ASaTEX;Yq$ObsDwN!p_-(Q{%%P;Y;=99Q70_#3TalfP; zud6I5PIIPz5_8hVbFy2fwBSastmN((iMyE8|>`JAZ9ZDldz?g|FT8Rom1 zAoyHHs5NH?458KAw0A!(80SAo$zQnmw&R|bZ2e@)(9qcT(^ObPfQ|RbCE+S;c60Wx zlxM7e);EK+P#g8nmXXvWKMQy-CJx2=jeB_MAP#(V|C&688S#GKba(Rk(@YpT`kC%6 zgWKyXU58sOb*B1_#fpn7uO^jHvQ0ikP1UM6xfVQ>D< z2b139wNO{M4EJp1y)oV+{eeIO4 z1D@;pN8;)g1U=T67$r$NorI|ML>}SmtJ0>K*lg5lk`?V(7fQkC3)bxt>+x`*^skj` zrEI^TWODZ2o8gFc6_#Gh5TqUivj=>MTW1#ZP4mpoA8mp2(i3+uJ6`Q*mJ{p-m~ijQ z?&@sEBnywJ52dxjN~d2wA7w@?{5rpybQ_lZ0$~<%@I>NeH^cZH4b<)4;>D=-YhSe8 z+pY-nTcL+uxAmbp{l0V?AN%%u6gPK`-)3nW1HTk#$mwvTVAg5M- z;6jsDI{dRUZVIP-^Q$X6p(9lAtLyf3GZ{uIzy#4lExpx#X~OcN~o?=pj%O4U@$L!HB zr&2fqGWyy--4DGlvn@cFJD&9CSSNXSk&SxGI)KBV}8Cu^ym`PeO48t

V)N=rE0%2|h}N>(YhA7Jd8(iX1KAp}^clS)~S?~|TL5Kz7MNup6sC7)X^o1sy8Lju2KUa}s~#0SehLp)Hq0Jb8G zDk(mf`Dekm)-vXwCDtRK7gmbIJ#1L{OFPm3EM7mBp^+jS3NS~+`b}CSBmP3h52Try zH;|eS`X%1S`U%)vndp>F)S8vt2c*0;Z6o*qnJ$Fy4LgU27sUFM;}7LKnH39My=`nG z5K|yyT-GSX>>x|W=y^mCoWRRv-tRf447wv8AC>2QbDeCY8)80FyC&8R{c!)@dN=m< z$48(B`Il_7VBeW8_rgdb)^9KIs)#ru*T&<)9OVvj4#ZtXg+)X*vFzCR;UA=)_mh2; z2uO>7T1Zy{sJNzkDU~asKd;{@+RM_EebZ9P?ybSym^&ZvK7a28m=W(!qqdOA&4?El`iv(yIW>3UY)V)f z#_u|RT_=qX1>8CSty;UXW|0!<(u(KdJwBTiOtTPHyK3o-4|0~^Lz0=}0i!;hkIJg)M37wy3vk^`q- zih(Jnv4)&ySme|oP5q4GvLA642DP?>NVeo~y!e;DQ;)?^%Qa0xj# z3DD1Pfo{ITi{G_M27dd< zE6uY{y8A9zMcnz6^hk?~%}y4OmTiBKDyx4{0%p$f$f#UIZ0mS$mis5qG^V6P4}?o| zU4u!oN?{u%SAzXXfqlD#yOqaZ=vPaOrQm=R^G#Ivv%OoLea=(E>o=`0>$mJ^4-rrx zG+v9rma~qo*<<;n47BRP_J5Ny4L>ruBlQsyI+@Xb3t}J7UYpt*clZrJD@@{%jYT<+v?k9 zPZYsQiyP!z#bY#BQlBE)$Ue-$Q_~>&ueAo-#$Y@R^HKC<4Q$~Si1lI(tPB2c)5ICi zBsZ{?NV~t1wUCTKM&HSuhx+1%AzyzQ=Mrg0u8W^yTd^}Q<9vMCj?kHZS<`EqO-3Vi zx5t7|v8`~Ijq`&DY0f}SAh^YapclqUV^suGZ|$K9ZfGcrFpNid|9OXglf7^{EBs{` z@zz*I+q3h55+x=P5>&LOkuzd}(heoj*qIWMy20?+R)N>FhNh|V2z9xw>B=V^AcX;1 z@*s`m`>Q&SvXK%gBFl&sXNPmymFIe9`nU&2KWsm2NotMy4qZ>Em9dJ{{)40*LOwuF zLzyHPO)IkI-h+>|gGNE1XYyfPI0!M7@mM4KML&{plt(fA5Je+?@!fsMjtlp}-L#?` zxn*+YNQEkp;2b%O%yOR(^2&U6qNZfOBt7D)Cd*khgfzNwZ2*~)s|g9*?Yr#xideu! z4odi$x&O)T>1tcS@vv1F{sN%J(8`X_^J;5!3NA!1_||0pUXiYJZQOX}Iko-(%^5Zf zN;gh5ow~GGzlfie5{!%Rh%JdJ1Wy@>UI>x<4wR1J#b*|xJ8+5-$oS5r9GHYyByI-u zfi`~%UD8Th*bQsUyZ%#5&QG7Vdxxm__r_RMZV>AX@VT+S1`uVRS?i<#-daIGVO3`p!QSJy z&5|Z@>IhtF?LxlxMGHPygk!W^sMzuFye2D7`=A#1aloZ$D@ed8(#jYA%O;uho$Ys+ z7b^{3blJX>`S;`Y1@_wEv|4;EeI1ZK?TzZs)Z5j3sWkG#@1ze*#63OTylxP5DVxq# zrTuG$Yc2l`(K9W+;)z8SQqM}Sq1PVnA_W0>_&muM=-{k%s{C%#EU9g}BjE^X2se35 z-_{OIRiL_X$=Sip!tTQ%`;v4kR}A>~9#d9OZ%ggk&Kk*Znx@=a3^6~q!NqpGMd|{G zBnAd3;ttywk|+FQ`<+F&RDRA|Ry@W+?ner-%*>`^siT#VpsTuyzDPG-0UdhVCklX8 zbmp$KKPPGbiH~*MRfw?e4%0K89Txet@XIoNnyJ;0-k`VJ#t531V1dqu9oCI~2NMMy zj)PM!dM2I>s(MKC~WFU&8xcddW8Kpi*yP6z<8I^5G3pZH$BL{d&Rj*ghXV z*lAw#un6J`cq~vF$|!K}L~S4G3|g@2=U!Ev(!SDM(7OyuuLEU_p1FTb3G>z~ z$pDC69|yyk{x-x*tLLI81v%K62Ps_kK19RW?*PiwM1DdAI3K4L6&J3&nIoYJp;HBF z$d1eqsCL#|v}V=loIn5O!&!~#sO0b2y@TD*gPfsA>~AW)y*ABuq%c^t`wM+`PF}%) zCo{d?78acQCbyPkXiiiUHaKMW#`o+@gJ@SCv7H^q2EbHyt92;w6#9`RP@=@j8)k6a zNIXnR0R61cI*FDOjJqg(0i))*CZR8o&DaSS>+$D1THUt?4^iJO;WuNE@o7q=eOJWq zrDo}Yu4jDjkU$BkG33PVAem8*SkSsAd*FOOQdjFY-11aJB5@hcz0w9iea~n%Q4B>8 z|B5uwYv-By)&b|V;!3LJSVyWe-5~7$a<&>pzDb|xAwRU(uFiI(54Ijch+Hrd3;Q~P z*k0}BVzuiN0Uj-emuf(?X-%SAg!CewnAT?LMn>w+UMAn#;mC;8^={PW0D4r(A?Ozu zA?yCSIT8~q4?=g=F(?uyVIm}01L>)b=^w>q>RqNzubGtcS(1DDkV$k2f7 zjR)JHO7x>;1~a}bT5HxLbW^|`^@Y)T4={)rNAiOaz_OjAWZbSU6xct<0;xbv?MXKp zfhE%EagDYIQW8vK#eFV}9J0%V_Iv!^L9H5{2OwV9gym%K4oYaMBdQXJ8j;VqS@Ili zyVCBs`K&!6=w514XnR~x1k|*^WIIG3NFoL((w8#JB9sOPI1cSvfzQ%NWFH-b`#{;? z7m5_TZ*7)NEaMXjhMV4KZ`*|<=cbdON>y{T!POJo3r#$Haum6h*bj(Qy3a|K&5V1< zZ2G?~r!iw_sgV9!l{_dB{pm^`G+`d`U9p_Ro%XIzbhB3KY1SK%mhzXs25DQe<+0tg zDzC8g&^xM2W)L$)!_sf#BlK9D*77uzvrD^S@j@*syY^hflgx_pl+jjnHTKp<<(?sD zXQOms8)6#zRUA_@Zv%a|xqK}WR*ju0P!u5(n?ShA;XZn?zP>1rL35FeZX+eW-y`H5 zZ11(7hdyfTvrV}05mQ-vrNh9$@Dhr(&U0azd4MCUmQTEpG9lVMGjd<()7(SOSii4O zaxw7+`o-uk<5$~nIyp`{v$Ue7Pm8u30|qkBD=-!R0kav2OdDh9*1YsaG`d(bO5IdZ zKl?XmsSdo9q6D7q1=SI5hcuO00i-h(U6H2vH==T>a+yyuLxi-uCL>P@^3>#L0N3Ca zR5yHMNgC@_Z8lU=+W6@KUzW7+j({HY3cT{;8x^k z_k!=Fabd3eG03UMCKvA^Jgh_%IDjs-NQZ;_+cN*MQ~dy>$r)PxeZQ$Zvvl7iK*ONU z)k}ch6A7`ayGi@iNYsU?!$C(0T49FqRkF%5TSAlMgp4X&S&Fe} z^*8%1pQn?LQf+T^TH-|jIsLrJ4ra^{#0s2a*7fQ=Kk2v)frr`>ldD>ykr@e**lwS}cW zy8W7;g!zIlv9NJ{hFC54PiTMy);MYwJOcFH&+REw{0!Fu=E2R$Mtiaw!Oc0 z)Ah6^AG(OA*UA_M#*MlWQ(bm2=W-s%JvN-GnPLuI@{aBB>NS3_kXrFCOoj%t)~r8w zk^ui?H-1c2oVMtT_;*k&NrE%}51Em<==)86gyRaIOz2Wm3$4?^8I2v}BBKB7)H{0q ziH00a?-LEJ?Mi)O!D=D#^us06k5}PruOET2V>R^wdhY}tG`oYe{xqZ`rS=DzS^Z02 zO9dP#4!#V7v8@dA*wyLVuVqCATOg`2V_BN*bCUZx0OCOI*VP}LSo;qlysKfo@DXzY zbvYg+pW{1){P)GkqL<;F6oMtUtcW-&7?Vd6L7C(M=yf`fQ$&;B5xTVK0y{TfG)cov z?-_x;3EX5d`;O{|AfmRs-VQ2-JTt$S_*O&OtPReY0NxVDGasqDw{K6v(B*&u(o@qH zhBPy#DFGk|YdzSr1A=AWh7yu`c9206=am76R?$9);;HG|v!7*=E>%$6%bQKc^jejW z{@aC?wAsV|f+Gjn3;<)t5=da~@)5#tFJ+eh!-*0_a;QIWfng$JwB9tR>wb={6pCKM z!;>gCb$r2~-o(aC3`ng0g%W3V_K_@r10nE@^q?+WgcXX1D$8Gm0>%O52^^GiKtqvYI$!K`h^f5^r^#sd080$w*N<@53fc*73Kn;Y>D!La5yE1wDkifnP?bm6U*7A4g zulXDP-(SaQtUvFHi2(9^K=0Nee>%Gpd+j}(Ty2AOlVce%aUP|dwY|rO(i~fegn2g! z1n=QOk};gPRG52HlHACENWaZZLC*bZ9qU=YW!^v1?wq%2En=tMQWYAMF!&z}+8?Vy_)NFpkX;uc*AdK0) z9u!XgF}54ozR6ie#8|^WeDfc`5tMQDKhA6F;Lbo1^xb2M_eZX9tu*6wS(<#a6k&4;xYGAm;;7z3tVFJkLafK0<=<%k^{iQnl zC68%2wnzM^-irt&bv2pQ{$G zorIZrT&%!-tJay@PGnxy9a=YrSYNfla_k&Y|j3^9;wfrQ;{H!&ZVk*d@w%R zV|QY6{CUr3wD+fU`H3t)%EIfn_Ht*BdB~F$!Z1WH6O$Z`-=*D&o4S#;Evy{`55fQ1 zyCpkIoDX@X0BtCl*lMz$YSp#cFL(O2+gy2M6ZmloPh)VQGQ5dQOat7z!u_nN^np*Y z5!21&yBLBMg>WDE{c^zUYt#F?=jrYy;*FJYa_O$tfyHZ7qkB12>u`qv5%8WL09foI zdhz1NuX}Z^e_v3#vY3-NL^%QhwX^Fahc{cE-I?2TPMJundxFpL@*I3f95)94^9D}NhZ+E1 zmnFm;$L^<>h2*(Y?mq+g)cl34Z8|x?+t_`*Sl&`r$r#t=&-;1!5cO>*2Jv$9g&5pD z%Dbr-<9|32h4;msUjGX<;|-g`w)`({SoS(+dIkQtm%2eGWEdpj8QeS&AKv#GHNda;sB&7RJ# z8(S?jhSxgx-aw&~BP(OP#ULp|)M=t9b6fb*J-)xvXzBTN05ai4?VG(Ai;r_zeIjbt zDX^^ee`Kt*6>xHIDX1aAx}yK5jFsAqLJ4RcEUwsOZr@_~`&o`XnX=~Y^18Uh1m9of z<#icyqX2FwIr zYb;yXGtX)4UQoVNQM0SxoTR6Z@CApBU$)Us&A1;g%-i3rUr$CbhH2TtL$<0R7bThE55?_a=t$&ZOm3Ea(=#44EaA5F7|55 zxSf_n=d|Y_T?zlnJ-13ffI^pCnAMBunTVC>vHN|yB4!nkAl?@{0J=-ijho@f#ij8> zR8)Dr{qlMt#nE#Pnw`o2EoAlF3P8464ps(4B-i1nH-_g#X(1IcxZV+VZc|8%<*!NR zxaNoBlA98Ay-&6W2Sj$e-s>pb#HF%Rs)61(6SW!uTP{ENlE#dyHz1DchRvlxF@!C1gw-d4P7 zbXf1va${LQar)ac1zrvv4kuB?<@I^jdKIgtWw2qr0YXZa!KPPdq-~C!Hw6k^q>vVi z092|j{y9R(3$$(1uY?CoN47v0mM<8@34tV98hx_xp<8JHmy~s*nX2kd`g2kjMdHXA znXM*paEr{`{%j*e`fk6hRz)%dOZr7`xglulEjtStlP#woqIU)sZr>ABxwSi zOvlit{jBl3C|$bR0kW*>zp6FI=Ij6s^`vyD2j^wo4?t*F<+piAMpLGxxqJ93|Eb}c=i)&#isCv zr2y{y-4YX92ss0>l|OB_^9j!eL{i;la`!bhRejsYyQ%hlVg`&q-mx=75Jl?Jx~a+} z|4?-B<5EjR-aMw!Rbavq>N%J@@cF#nA`7Tfo(3>u^|}+mhgxE3_y2Rq9+7P955Z`3 zA<@@VJ_BftLjR_ijciU|W zY4NLM8^OT*-5BqjYjNc-_o_b=H!SA4g2;N!W3BB8@J|Ee0(nUPoG$!tc__lF=HwWV zhr(O^f8?RSMY^v08;w-m2PkW5tELIy{uN#9z>H%mVO!tj%d+ST!sQVnC!i3g*J@+P zz95q{i3hp4z@8%J!aj=`gm|UK`xkKKTsfWfq1CgcFLe9W|5hF^(W-sfv-?jZs*vDQ z#!$~BSLknWQsDAxZsINhMICadkbp-!)jeN1jWhDQOsZ2Y<=j7;tJNU2@#4{eVayKWeu21CuVk0dqG13xH&}xCfR4 zBJM->Viw=I?u`Z0f)2~M&%faj54v7L>)Y^1Ch`l=Tl_5=4v2Sk@Ae4*hcW*jWvF|7 zW`W)I;p~#6kSh ztXA-C7Ra>Q26zQj9xU>qh`*pe0hA)c1a>1oulo3hxYUiHb7C>bY|Z~nXy$%>PNbz_ zcPvsXZeI+Yt?8DKbvhL{_uwh?SADnddLPH6WY7kGqAG^U+tCgUw)h$jQVfZw3bk)J z27iYFov5kPSJ%QptQjEw0p`VpWHI~OxoiLB6+kVDJIIdGd9`>|wY!Uxk`g{dB8RV&c6C@tm_(Fg0?QO&_T(AynezZk+ll!B zZ+s%Hv#qA<4WJ@(fghXiAaDNt-AIxW>d9_Pyi3TG)MhCpY`s zCK1&a!~$hHU7djvo!%n{q#23~vyt<{8NW1^^IKteMVD%9m^S!kI zU?du*Z;{EOKgtNp*%5~WXy-sqD$;q$g|SOu1~f^yGVJCEdN160TfJj7Kr3Ud6I;p zp#4JrYe1xDdhrO%uPo#xVUmE7Dw$Hrgoj(1*H=C!#i><)2F9AG=hk~k&q<4iXpoj{ zPwj}6n2sZ39XXu=j4MCOZ~%Q+@cJXO=L#J1E{M>)49V7iWp@%(;!#zOiDx~Ls zp##6fC}TW|*5E@Py@}o;2yi=+w8@reMK7IM{1kbg=pPy z8Q%c@(FMI8+;PF3u3hdHd@P0PCX(T^|9}T6nL$Htwwpi8KnL%fNSp1QYv2m}FK+ee z6Hp8f%(bJ|2m%z2I60&IPK4qfNM@glUV%L3zMamWH+;x;=m+Y`R^PTtSANc;Z*6z&g% zXtZ+47L!#Iu^{rJJJ&yfBOBBC@7q9jzanw%y1xY-IQFzJ;tYN7UTxnqe?nd{LP&B{ z{yN-C$LFE9z<5qrXFetvVmYRF=du`Y{ixq|A5fgZ!UNPzn)mefB(3DfV9>{41NDAl z#KGT4V1PO#Zrej`CRZQ-7=pfyElW-FaNce7XNlmwa_L=9?TwZ3}g{)a2AIUB=fn|5)Qy4x)KtU!_+2 z9u*oMM!TxE0ixEDu#+ZiB#&S4=NboZ+SRmNQ%$UlXQU(SS zu+KS&*QP%rm=Gzk$n`T{a>EBP%UNnxCh_e(fIiwly>@ty#ykZF`G?_0eaQ#4dB=M z$5QCuwf36#fZGpIAtQ9}AN>&@&i(E(LJrws->pGxxZ(x-BW(Oj%oUB!Cl=b|BOK3X zALPTH>Yp>TU}RLm-||!+oVVu|^#S5g3w8QCK}7SV9PkRinqwJVMKKV^$51c@`d1$p z0Vm+d$v3f+B^pGojV_JQXfVVoY%~-K>kNIA)=!)~+4JpHHYuWO)sRY#G=_=)rnI%H zZr(=$q~539vXb?WJr+yj_qdtK4#jQ6m~j~HKOlDc^iY|^-@Ar}>Eexi-%r>-ay?}r_Ax!=HVAMcW3(!m#$-Bs#jh>E^X%cFItmAdHfu&dUp=2$IhiWE&Q4_zqTw3x>K z)j7(*X!BPFPcvz>hXzC49Q{7ch~P+cepDC_hI5c~+HquInv38!+A zN{p+_K?c$_H=W8u&Y=Y!t9gd7Id$_{!(uM6;55bseb%X)qq8L#l1f+w#^mB6BqT8+ z&Z{BN{qWrv7U*yJ*L$cu{Y0kwG+vjRuS{#PD5C;yknJZ{z*nbymXOysWg^aPq@mzE z0r=Ed-S&_mt0GVIFVnU_rg;eX8=6jmnTn~QwB~((bZ430qe!0$K#aVfZ6Wr|kLMAb zj-E`r^xntyZ5XE=MI*r>_=At03~IDeI8epSnpT%zpp24%lj!j7aw@hC6pf zD4C)eycbP&urjhgw%^n_z{(^V>dFgJz5^wW53HET6CO!;`1EX#Z(QA>CQx)5KXRn1 z1(09HkZ_|9bShxi(RKyo+Y>gizgUW~;U>rkCE+lCYG>tq?8xAJ2IT0DSA6;wHW(N? z&7@++$4xT0$7}YDm>{Ya%ij+p1&+8sdCEQDh&7WcJq|i<;M1HGoPGlA*J?~ZPGCUV znJf~SAZmO+xdgHkdq)3GPlB06MH?iEi&YYN!-Q`3W#4@<8<@s2-me<7U4l(V@1@go)xIgS*$FfM+?1-9E zFO!V&7P@;iA`X(@ZepVxo%`+UTwvY>dQy~!^RYhh&YG2;r*&fu#_NLXK$6Q^IyXGp zE&LL)dsjF!htb!m0ZGh@4cn4{sutG9u(WJzjwVONTCa>(b3pII9gA=mH}e1(8D%qn zJpYhZZ}7=y9E{uGMy+8GMj{D?rL_C(?QI{ihVtNPCD1n7Eo~T}ecOF?S2jlMu_Wra zyNP%Yl`1e_3~#OZ%f2^bC0K3KUyZUFvfm_% z%brh!R@BJKLo-YolXJ_x2siywatX+meWo=>WNvL^l-=mSap=99@+2&8B#NHOyzl_- z?fkY1REOJeF)+d%4<(Rp)4bpfi5Uzx#AYSS^!4vJ56$=+%c-1Y@z3EMMeGqu{mgQzCD z|6lwsqVUIbOoIT!SM`DFI*CU3WrPsrc!j^WDH9z40)n50M}vv(hd0+MMcxf2Ldg+j zdmwHJantAzJ})NrZW0B&;$FZWX9n5(BT}RU{td^{uZ8_(mK3h3UrmS`2IM!SRPMsB z53G}h4Vs?E1rhkDxo;r;R#61bE}Wgs%$j0c>E?LYog2;cRndAqU!GjPVAE(qNFmfX zCC-~`5dR40gup8)v)te&);HU1JCOQ~ShNO(6Lu%S)&K90)h~1ry!$M(rQGu(B~629 z4~!XQLWdSeyc|pYog}k4kzO#)F-r++oM8+7Zew1gSUGpzQ#ZT8PFcB1c%MwYA1wO5 zJzRr)KkMRDYY?FzhUKb&e=OSQKzY&61gUB?5)S?GfDGBJ&>dN~&W{Qr8R;M1vC%($ zkAIqnDR7cQn)bPkT5Iqdr8yA@cf3QSQ#j$VLC6B`ZF6k7Q+Q|IMfc1jTK&flD-URzlNNt>+dXX9i15jWk-|YvgD66YS1lf%Ih@>T&l&E4!SN{ zxm_9p7vt?ss+qwBCm?~;aEwSGkGMwyYvf#pg@?9=u!b7eN$3Ui(YTLj}B5)#Sk*>aY6@ieet78*W{a?R6yl zByrK_cU|jyAN>FiO3>etaGXoH&ImD0P(DL*N>5q^5V&W@=7r7;6gI!@h29=(y#V}W z(j4_OaY&V$T@O@U&nyp4KDm%qBo3m#3HL{MMjrB8;P#YRVA7VRrF8UWnL`_Z*^}CX zJ$P{h(@b%X`~0QMkD`!8a#zMwZEQ{H?4xu_oIhflheFS}+x%USSP{zL=uL|{ta|qn)wBJF zj^6d?%L|8Qg}?n~;v;M>pOwX5V2k#r^3!wzOKX$B@$yz74$USU>t7hZ-$KZH4YT}1 z6n_^ZJ5gDH^W>ZeM^V~OlGuAgZK6P^>}3jv)zes`Sa5^t4xLlN^#9SnX7>z*Ua0NB z@M@}HMQ(;&W~YBb1@Pb>u@Ih57kWXY!9A| zd94WphNFer}na(rJfkvsKB$V8a~RuPR}`pEnY%?!XurTQ6tz z8}kzUVnxJ@xV6JPnQYYDb_<`3ejJp4!S6mD=+0Rx9-gvBm(Mz-M>2emm(9h$E?T?W zSSo#=tA}dbrL#~>2#VG$SnYYs94k($zfjB^a~9X8x9TtAxGGK5&hay30tn)zJ$-=u zKmK}bZUB^$p_2bC#~-GWXTA+Kjr2#bTbMTFfx(&d?|nT~1y;aEo*sO% z4Cqvwg<-xy4o5l5(Q+~lxIDMKa05y{4k^#RKe-=Xf3bBXHII~>z4U4USAK3h24IvN zx$%*Ou_*j$>{?m!tj0)q^U_CJUN1~31epo@Ompag?+V~s)ThEAcH`Lu`qmBZF;Jiv z#ZLFUb&5Rm0{3>SM62T|r~_%ay%*~NFU*}<^S|l}6pah&N`AdvY{pC;6t`~qTr)E*&! zr0TxPBAE-;T?A`O@GWPrN4pXeRrZpZ<~rzdt-*WmJn?^hg{;w0`?OFNRE%3n!(@m( z)F4Qe=FY-|aj+enK>|&slKXet`~|(>;LjnmCTj}(f7J<~Dzul+@HP~F=Zm0*|AH!; zZ#_uFl8qJ+yX8@MOvt6zaY{~-Ue7~%C?pdy*O}c&(ZT!n?50gjS{&y4WN_hrxH^`(`FSq40Yl|(E{)nl$4DA9xt>pe{z|m z7uxX_Nx)bIshRz1+H{1Jrg*5>4&l-Fl#x%=w93jmLVx-A4!^AGZKW)VC(`QoYv|)#HDV+*vD>wim%v)Dm%ai;!EXz!TTVr?IE&ifzI=uXpT-+w z%SK0w`y^!j><-j2F>%`%KtzMtojBx+tJ@E|p}B7@{0 zwxe{HQoD*{y5yz$&Y)CYLbYj&`@+w=M<16`Oe_+DN9CjpD5@&l+d98nDQnD@H2Gd` zcI~|N0HmoBoA0%Po0q0PJ%!|!<#81UAAOem)Pt31UM>tuB)$KkzT+tjS5A3L`SI^Q_1&|maoeJFO$$PO0EQ$AuB&LacPQh{{!yQGHF{2>g~DN}WSPcDfuZuY0MA!U!8rMC$joc%NvmpFqQP0{R~(DOrQa zbm%xZ!)$5NSHTMkWplg1-FFD8FPM4GW@34m7dS%KxAbW|@AaZXkf!0Vss{bM3pI9c zNe}KyZ<-kOlp047_hYrsDtO&`6ZDriSFbc@k+KBy_EB?fc;}xVYk)E>K>9u+UCtUh zb=4*#qnWAkMX2tm5+;7r`{gpHGd$RvCfsmVB!RAI^;A!uUjO@@w=SCsUSiS&{cQSW zmL`&2`@6*=|F!1K!szHJNB1jshap>W@H5iLI*2%x3iu45Lf4D6hVw!b$?s2ELu$`` zq)_%YUksn2E(n|z15a-#+vk@-KZ+_oB@bD{OYgKU^hYR{7316#85~{B0{y|(7UpQvH`|l+t-%rdLy{NrNSd2O zQM+^A&C8e+7fJ9zusIL)TXTcvVfy*XQY)~R@3<*a#=UeZ5;5`HLS~OKdZ1lqZ_&~5 zgA1f7?{ni)-5jN}UA3IOX&mP1{kZ5_C)ov6%}`dSrRv?a`lFPUjDlyV8ErUYP;d+1 zXTTQYi;i-M0i8qS{T4_UvnOhA>mV+cA%jwKeMke-_RaTcf_`6<)##tzdX%x7N7L+} z6ra=d*Baz(dBd-eYEgtlaI{nFw~rcds{LnI?KW^b17H~|;7>rNVFjhJl9&T4fNr(f z87AaM}uBw!YSXR`V0__>Juzl9;}(diZ>G zrJyU=+hm!+Eik&F9nRZ>KZ|_F$jw1r^$P5txWFt32h^$B!@w&kr9XQi!FL6YJnN(Z z{N0!b!0btRPnSvf<|1M_Mg&aJvX`H$yGvsxt2z6kdq|%7HQ3KiteOC(_x0#M*MuTe>7M*M*jCk3NUW4A=#P|MepW*x>fTS@l6$AinBZ(znhsAPP}jKD6F~bx5UMd_aj zmh6wXa92(Q$v|!#X&r%uu9q5$6ywl+cexl;=Gp_@A$Vx0!fuvXQqhcN+9~n(E6xkB zQNLB8sfvBLBNxEwgsT{4E8A&s`p3-@{vL0dU0(uggIiNw9K8M{e5uP8dX$ZOnn@SY zu_|Wf9Yd)@hfJcsI)QfB8T=LeXHGu2mtrnZxOTH{?^!Et;*ASA1Kn178&mNU}38`q;^X5|3RZ5_$7b7lb7#*`bLIE;D+ZRrTiob|Jos9ot~>XKfUqgc z_qr$I@N{31bLh!bMM>;Yk5M*eEg{$A*MbVDq5*dbFXBcFvPfu(|LsDoB)IhU^cH?K zX(~e%+IS6L?dG6d2#T(B8!XJL&Gw*Cjme!o1^0ybIn&O7M;MKp|ZEn7JXaPI~yW?sBv^7G42r1nq@>iq2phHy@Ml_eW%9Oq1c|mkbOl z;Q2zar!~}DGgop~7B77HBA*)_-+k*q>QfEk)#uzL%$Y-4&kjS}W1+M%J95i^d_yHOh}DOFF?~yb{hb)SOky`; z!mABR@Anh?KU^c!l&q({FlL9qdJY?7fJ#G`#_wsW zg}P}>tzmt?wV$;sk8|c@GjSa%<70;*`e2)jJU(_E_-rVn3e}rYj|K+aWN8#)tzpx| z%Mrd-1XoPX27BR}^jE>aCUD_J&W)DCa6J!i*P+f-$A2wg*~JgGP$K2p2*7x|aXGcy zim3YG5{o)?s4%Dm_tui4b-E-a*Ft}bnDoj~%R^KA8G9r7K__)Tm2t6I1chF&z zxa#w%0_Qa4bpJpjo}+U6;p5^~d;?zb zSN;vTYF`VIw5d~T6|ft~XkFP!E3yr_Oz0g1<5Czm^K49Crc8=Ice^t2cj@bx^Iwx_ zd7Y<=cK;>Hxf(-Eyef028Qdn=_I0sQ$<4d&_EjLp;5#ymk@(E^Y<|P$h-rCgppOXo zTU92$B+_NB&q)UJ#OImPsUGUp_(Rk1@oEYl*NrKf#`k#jKXR*}CBL#Mz)nbRUwU3# zci4pwQ5`GFp?hD>;I_QHC5Km}?O&QNr~SPZ>cPz*Jrlb`Yo2WAI1iUc1YD4==*urz zRZL|rx?DmWy!oOp$5~_8f19dF19ljNYrj$fFM+|D44meCp>9l-m4Khy3%e*Ht@Vh$X%#g^3*0e>=oa0%UYoy(Z{w+pIrSau!l~aj8}3fyMush=$;~ez`lLd`9Rtwr zsSNuT-uIp^6${V2h=W`8J>3N0;}-wIodfHXp-T7q_a?AHk`69vq+d}JnV)I-CB>?l zosT>Jp#$^FU91PMpyQj8*ak+(NQehV>#?W~9Q6m_`3O)7wJt5rQk=RWbSdQwQWKq@ z;%EnRZ@dI(&1Vad^ce}fym38VHSq5?n3qY9^cBtq+`o%=U#XHnADZHipc(OA7cEDO==|_inl0Kq7#RJ1NBn;D(li&Kp4f+c)FijaZUk z6fss@3f$E$`*)0*d#}ud4t!RZpl3*89G>y(fYZ$GGZ1%VHvN0!LV;S7?Qzyfgn`2z zzX8}x*1*k8&q{H%A!>J~;|Hq4WiZ|w_{jMX6Lg|@pe$8e_rn5HWsmmJb*su*PMDEF+(m6g5WdNQatS-!V;=Uj z<_o1q%vyNQv-cBEbmL^sFf-d>_vzDHs1;oV;ljSvZX|(V_`IUK-x+4$cfK-rusRbOFYNb1y`yyh!E$>;Ca?Z2X|9#9J>&{3pal`uv0UWdspiK+~#g$ z0>5U@D>yam)2B_6UkEuLWd!9r$hxqMop{%tm|cRuKYy`>eb`TQV&b3nE3TezRh+P= zMI35}4_%HPn%1IpTfTa+Zq*>u1lBDz>@;wk=NyE4?hxo%$`wn$^S3X7To54?g3h#mw{O{!MbgK-X_e+-%~mxf+#x9LdzXHEOxI11q*iP4$jXsI%RMh z9p(N5cx%Gav9K#ubUV|FV39hHGLMC0Ig_ejHGq6Cts`qjc;rpO6J}md!Sxj#&zF{v zbU}Eu9<3}q+UYL|xMHa=i!=U+)r6tJeu2;qqt$s%5w0@$=CNt!YsC-~!kM=sdc74M z-43aAN4^RhxO7PbT7YFh~9g;U$HD-I1IZ$9QgC^WfIrz6J7A2QCM}Z&f2N_Ym-d= z=)s?m`|Gg-x&c2gTkVG<-xUpAnnv~HFumvvC-p>0=9tR)wS9MMU&3MIMr|pcp@O_eC-C_LK5JQy1|(@0Nlsyfn3MI zk>cB2Q39|IKJfB;=K~fU7#ACfAx7u6hXPA`EFgkE4T1N^nt5_(>~OE?VKhue5FsxU zI=(L5_ZQ)oqCRCr=)>OSS?@O2LZkYLIx3q8FwIZ~a7-yxLlVa*z5vBV;=iAg<0oR^ zH-*6ZKnXxi=s4@3P3C2FaG&iaPu#t+j^_?+vMJ9{OcSkmy@TcAdjPB^vF&r>V0s3O zBOWs2L#3hsHq_goC4tHDeyNTcHS;%F!NhCwxofVOG@~MS%cdK9< zBr!IjvJiq2QyGBNnso@S#|-Z7Whsg}XK*Ju>6^1UO}i^@-e74D8V5HQt+m{VwyA3^ zX2?XaYW|`sg6|;Psb4WFvW^{&9POvEKZJ4Qv7<~M6Iyefj-?M)gLwH|w+VZ<)%l(E z9zEFqo`)UnARLKJ-PxuSwL{^RFP6~=9d~o?Z5?cK6D;@%$*Km4zI-;&wEj+_c@ zp2dvxBcT4pLsAzw;K*hDa|pXs`&E@@Kr|dJCJSr2(0)|@VuV?&1OI5N>er+I1&BUs zuA@JT%iVwIML5Cp_pwE$Hb43HP#d^_{v&OR7c3?DH@`NfO03$6am z820)>5pU*PN<`Wh4|zzHO>B>EA3TG3bG#W`ZK{hcU{hS2m-$q*@#o%C=H_O~gZaP@ zVN+t=9PCdzdfb_OKWA5>uidy8m-_kf5&cbKM@yzkF~k=S6`k@^;K183@rzwA1x-Z` z)kWAXk=Fq6q|2}2Eknnp^ve1216C>-hfD|tv>$)W8-MZ9mr>+%JqD>3^3A{B!OpdPv}2ICGrRd-W)dgpzK`HjiNNK{18-4Pk?A;|1)F}N zXVX7DR(bI8va37W81I-+T4rWLZP$LtPwUKSu;A1N0&4weCw$p@oTL9P>32TGbeF1Y z*7ovEL)%*LKWIcR4)~SwMO@^QQU)>dY%xSXL!K(HkTIW_YFCr5dT=)^qDomw{QiYC zBe`V^lE@x3o0~L2io z{?ELusN<1rTUh4bA^1+l9fCfqoOP-UMgrCv{oT^f65UqeW}Stse?_W zJEBfhsqf|IXSKVNPS$e9n+KFhTU9~vf>$HG7E z5HN5aDws9TX$D(v9u6A*loDb|cr5zG-K}<_|G6eY41c=)H2A@Kf6;=4$>th-e-E+Y zKY)ween@QM00VCD9njT?hko4ku_r5V@nCL)@6Voiil`Lg?lw-BNNT|U7^eT^kTz5# zwl8Q-Q3EFQ;N1~nif`|?cQWwD;d0r?peV;3#UG4f_m4DKQ*W5P=*$d!T!MQ|w*=OT zx~S!pf$P4SuUH-fygmAHkk%Xo#YL1@5<-WZ3VmaM+dST~Fe(kKLK4v`co7=Gw8isv zlSgK|xn96MgJ5dzgK5-MEV`CK67lw0BpMxmaapw(&JGsoCc_;ACeYjV(;SW4>eHA& z^&2K0iO>k(z32kM>s>8?&r`eIll??Nq-etwmD4q+HEWf8c>l5p;E;M6PIw)2)~Rsq z>oMIF&ZI>mEt3OTpm?sTqILj!!yE5bwoavW74xwBCd# z!lb|!BsH)L#Es=CMnUimLLJZiL{aj!`qTJFYUJ-9CaEXp6wvGLzUQ&ekG`8U{xk44 z7d5YI?~x#689pGchV`)z0@DM*cbDj8UhXusU#Z-J7{P727-=}T4UZe{tY3RQeA(k0 zImyc1k1(9SYDJJLxJ?kSS5FV6dpq-NY$?cZICRPAhVSsBK6KU1|4X+@%FN{^a~G6? zcZT%xad(>J(YGA#NFPUiwhXjHj;D=zfQ!mx+|%_)P^URu^eD$zDJ{PERmB)6Q!ah@ z(J-;-?P`f3I!#YbP9YRK?L>Kfm@6?c45$-&TofCe_ICQ=(=P(^}NY zyJ8A%O=x`ZoIpf}=};`8Wm*+-z;35Oz6J$l(s8RQxiS`=eai`>YPX_Tlm z*`Lu@{#d*9ZI3GyC@ACQROOI&W$-gUm62X<^4qr)7s&b>VpsoIpX=LxD<5$Q5B8=9 zdGLoQkzWh1L|^IMeuVSLTkC5u{1bYW5*7PhxVc-VUX3h#CA*V%BMi?L@e5ZIm*jX< z8r_o#PDL??mPgy~KUvKJr!G6}`T0H>GpR{$3}6azT)=lHgHtiD8qC5rW^6`^RLF;2 z{`VtJumW=aLbpeX)X7no#RQ+NRWFXM_vf#PT+!#?#gg~Z8*g}@IrDiJ=&hflL~*10 zRne#BCrr7&uD$`Mo-3ZX{ScgLntP92NMIbz_;bEu#1yV3v&WlL8vQkd5_RDF#<>r$ zuo{*CVU);+9|Nm};8Y0is|#`r9|NUeWx6Rsva3_2n$Pj7s|N@L=YEAhXc zp^jh54KnWA7M)Xwe@rp=L{F`~0o^8UBC5NyMQY@g*XPQ|TKNBa+6LM8V`lIT&(`AJ z!1H~$wFmP1n{jKs{gjA)sBK^gICWMPFM=&09NSJN32m0dO>0 zIG3N^c=tr~4p$S~g?s41s z2`vQ0{FAk$Q?S-p{gkVByXTW-Zz(9sX3&}duumd)T=D13iC!Efym#9fcqOE!nOxJV zPX1Q$xy&W8ru95o_@3nf-Tlh=`T%+(>XrQOeN4M*$6LtmFsiT>ZyQl?aqKDur-!B3u+z z`gc+z^2==GZ}y1e|GSf7KTjjO^%t_X3tpAHIztOUPIZ#BAQKDrLNogG+l1|Sc{Mh8 z^S9c6t6LZU=>R`!G6$U67xv>AE#SvkB(h+YFQ5NsfXkZc`38zwSwCH`oKPe0+6~%# zh_jk}1x}fTb#|RZhwW0LOsE+$dg2ZD#vOU-#jSLB9LPgN~j<_l)7vF6cZinlc#fRlz&GQkPK1gB#V| z$ga1!e8*{%$Z372Vm&ShIb?VK{Ra@gO;pF(3V3s@QeDCUes65tcZiHjLtXBw8_aU; zJiSRYdLq=#jm*%Uc-KGt;1u+L?;}9sV@6sV}?C})biXptR z9cX6DiC|FG>AxgqS3bz<`qOnwI0U-)!1k+6NNId_W$UhzGCBEGmi7=_V)6t*jXfbN zi;@4?b>N%ZMwSlZ?`{6PIy66FzXCK}`O& zIC|O0p0o^DTrpv1;(eizU5aWxTYHg27 z7m}+nf{g|ifyONvi5nf55myJZx_5&Uan0~UZLxXibexS8TU)aL7~~w@$)-lHxT{g zFMRok2(m##P`@7TBNmKWb9y}L$DTNjqa+=}`qKi&cqma`hz3h6MM*6N?9G3^TyY8Z z_i?R6Oy`zuUBt7sK0?iZ7Vw_J$#cEil#`V?9P~zXUl_|bc9R&Bx90G;_W*dM*Ox*g z+3uO2*g-0u8)3D14eNJI3A*pJ4#MTur;Z;IM_1`la;7P&LhzsVpGO$wxPT4}kMfZO z#<=P2<44a!8;8>S=!l=wrxlMIqz zyBDQW+USpL6xEaqW;;ItowH(S`CgaQWq{RC@)INWkC*8=kZ|J;Qk;J+#9L?+27ez) zaG|tr@r8844;ex%#S*eI?G&Q#o&E`Yr&lT{(Todh=IiTb+ZfzDMt|G{J59WJ`=ZS_ z_3jL2qVkBk*bIg)jlfvF3FRl60<2-)jr97+&&N>rdy#v+V#Fq(r;lTK3x-;M0#k~@ zX9}&LD;xNzV_~gbO&GGv`0TZDT>NaHAVF{xo=rahbBQ~rrk{s9@A)7Lo{`7uOR!MJ z`=K)e*ANR$PbA^2W>C>L&{KzIOBr5krKNvAe?V`1LP3Le&`}1-j>~x-gv1?z^~6WISKy!y1>ZIgQ8KYDfEql%`g-cAk|v*o-mk`hIYn_XtF)J1U>NL+UsL z&H6)149w&9uH>UPX@nDi9Ap*hmE8FK-rz{xaU|qH7anDL8nm6Pe!dF%guc|Rq;?2Y zaTDGT181 zMvrDO+t(4y-Ov-ZHskT%&3KrvnIR*wblS@LSg)_TB?7C*II@rOLI^1{5;DHxr>NR= zkL_ab9a<2_J%sP_VkefahCu2n;)f|#C8@A@QiV$K*9klLZ0vI*yHMA3q1ba?SB-MN z!tHC$9;q{PM`mOs7ztR^EnPN9zVOeO_kfWLjvP~tD2T`Om*~kde2Lz_*D**PK@|$N zKCuJ@+#4~hTm<@}HTuao4J6xFyK?lC-LWH@{sNw>bBC&45Zp$9GqG%mHOb0BP*L!( zY+_n8CEXnzf>W5gq)gTp(zVeoQ722tbL}I2txHK%e=XkapQ^BIPiggPaj{GYbXJ^v z>u?aSB0HBe>;D1irv^$aR-XIoN5agn?y7(XP(dp4LvMa19Q<>n4@jl4X!-+izDM~& z!`T;b*xc~tq1jRe2i)vR)w})?CQ94;x{LD#^|y~I8TINO?9=vRWXE^7PT=&8<`8ua%7S|8>TOhyD)33uM1~@mtXHq~K{>P7{x0x|0 z>?K;G2G`L)9@V%#2|(gB#t(^rCXCLDx?`wGFLbMZDWVFwE&A>Pg2@;MlaP?wEa7oa zXSzB{G{>adfd}o-iN#g_-ctq}svQ%~Mg{@+^aQkw^45wN?LtloR=fc->FqclZ#gp~ z#H)R3xZt}*w2w&eP=hrwc`w>VP`a!zDe`Ru=+5+uA3&6(Lwc0s2l)70s-Mwpzo#)< z`5CQBWIIuP??83(TzVPoUC)=VggSS^(B}c@mT?Tb4W#d#z4442-bAImVPHfqtS8j| zM3TmJ{K)WUqj<14dGKZ1h!Q?8A|Ua}(8*$`bx-hQ=r91abd|o@N!zta+JA?K;~X-5&qZ@GGTEbh7jl%t~*Hnv-#{v)h0Xa8xyMr$t{Qn zpnoXmIGZpS&HK6iSdJFxDa<2qDhq)J<5(fvSJZcGe*%`D-DU&PrLFwYbHU*y2%VT$ z=U3Qp^U6_5)Rv4S5`g2adB^Fw%YbI3VHY$KJG-!lK6Afj-6=N~oO)5c2wZC);J_gS zJ&S;imlgw4CN2wJQ+WYv{acM4o&Urbx14;ZiBv`uZ2FJ_o6d*Cnn*7<4eUPnvO+?7 z7@R!AUd)R?=5+J|zf}mYlU{PnnG~#Tne_Ugd-cs3F_{?%w&c%E+JG%QlhE&jK9zzr z_}UYx^w3e-PR$NTZkfw?tM?W%>Iwh~^j?FL7~x|~CN)lQloc4q|8Q6hhPC#-`F zKDil1}Ewf$Y%ov2HRRZ?sLD3M?{5e_@aq+0O3voU3rA070bmX#uW_ICYh#UO=O z-QEDgE!zCXeSSX6?c1758clFbPH~v^k3Td7dpw&aA=sP}DCC8G(8kua?_@$-zo>-SI%>uMf}T z2k)7NJ4X6jp3yiGmN*0e+*HIW>c2S z-Jy(EMGbpr?55$~vndn1^6U6*flURaE$HVz6K^ywhagOS`8Wiz5w11vCh!WcBI~`? zm?Kh*o#|r=R{R4T7}SpQC3eAT@AY#KM{Ns7u3;Y>@<87zfAXcDzebpMZnbkQt9opA zyG=5e`LKdzv7?x|oRX9n!7Up=2t>gbA7<@Us%0!25XdM~2oz?saC zUt%+VkFBRsU}Pppf)Z7_{`L^;D-f%wBQ}Sc$l)&3dF`k;erxKGRHD(SNw2TbwFsW@ zBx9_q__3s$E5seU**;71hGS>EH@a6tZwH$a5ENx~!(nvJul}OgbbhtYrk}rdnbXYS z2&*5IR>KV4s3T@r>>0nHz#&Q$bF&4W=>shs2Oxe}1bHpYZ&@?!#MXVa#MQJ{uaT>? z<|y_e%eS9KP_Kkt(Sf(_r$tEAQm`NCM3KVEiY!w&d9R1_`nudFsCxatPk2@l{MT23 zJMzjP!!PwoUBqtK4Cb;qJQb9o@qt;s5gJv#b<|A#g0K;hPuo>)G)nv-fL5vqH8leT zcD92X7L*{cWS3y|^b(?h%)EnS)M!vV8~X0*SBTj_CZ``VQAvmw>}cb_C(>}NNdA0Y z9&~n#qv9s<3W8iUDvI6%`*K5y>ww$b`MuPSrtlwY^uh{c{A630l=ig}Hq1-sgrDBu zEtm4JXB@{C2d6xQYd^is7oXa6d8r3Ke|B+-Tnyl291rho2PpDPuFWB=euhQgzcixh zy4PUvuFA={Ux1+f$>OWMzM|5zbB!|VRyR4|^|*rG1h&5I29Xg&S*t`)z=;lUEk8aX z$@F#8(?0d_Ad`B{+%;<-L?2P~y?{FE4DW>#Zso(hBd(`B5oMZUHqv3#+CeN5kjFmu zN;P6!lE3-pE9`BJRq8T;`$rgZM~8N5ID?i?4+EAPpWYZ=WuiVTtsjWB4Q@HR77@DZ zt6P;`-=pX2PN0fNNyNgx3$Q!=GfU*|oX-L?huMn`n3!?E74>2be|=;f9x*cLP_Fq-WoU&t-+wQA(M>Btm8yR?@LsDLBep5oB zNhUTe{KpDLE532!od<5E>aP%+;JwgZ*!OzgElQO4#5@81=@9G?8)tPs{|IY0RjHNV z?N&T~n{Ni@sb1&C_sqiUpT#MvrFWGN*g z18hDO+XRHW?iHt9AkSU9n4oxET2b=RSIASw$`fJb$-MI%ktNa_5#VjceKByDdUPF4 z?>K+@20(8t`0w`p*Z2F3h|nbx?QAxQsRH)jjz=PW_nA5DhUk8}ps#eQGCbAuxqVCTY}C4bJOL zdk|0zJGbb&%p5-O-D8pcAa=2DG~Oz#5>6?yMAoI3`PX(_$7!qCKBNsONQvqz%4#X# zS;j0R*=HyGiwiVTKkNc$BbyUxnDt7~MY9~>$4&hUhsl!M*s08~W-Q-^SR=BXr!6%; z(|q4wM0*yE$$N~@VT_s6SF}VBYrJ>}$3A11rYb(`c$igFiAPJkifjQ=i--LXxL!4Q zEPjWysruJFpM$kq5PtJvKkd;jUiP_K?U1`Sfj^xSWqb&))t|6$?j#ahyuG~N@&vuH z*}$LGdLaycea^BXa~jsUwBM8QGV+t6uyNg4Szgc7FZa@i^jFCA$6;9vxGl3=^=04o z-I$Ih*!PY&PZz*Lc#5}G((8<7eIcr#eD+T!@@{fmJdK0^22T_CkFJxpt0BKf?SRiI zshPu0l-7UD3V0x;$CBm|^AYAAv4&k?d*fi{!8~6? z$M&|664{ffAIh_>I)lkQ9g?+F%$$Z(a-?1&Rs9Q`Hn96?r1Zm_Z{J^}WiLDSEe!o6 z^$zAjlRirQqn*mC+gC_`^VS2H8;&0U{#KiP{H=$PXN}Nbv(tU|!vdoRkVzB~u7B!s ze{bGhGOnKWVtN9dHE`Sxx4!VyJ*2*xCxP$&HTQ!^&pRfL%S;JF2(_PxZ63v~yyKFX z?L5GFQYaU`^J=#@LC$E4C;-owhjt(ZQ0f9#zLxQ*M& zv9kJJ@M5Kh^3phtM8+}dAFKyM;Ye$yIMfZf52k+w#BpLG3X2=vHl|DbZXQfj`gl}` zh1RF<#cFRZAfnvjs*`%oFhTFS)%tqe85k9PbPzDK)B5ew{~Q_ODRNt|$d?_X_uY^n z+@yB>p2ib`4`Laym><>cQCm~;vSa(6_+Ekyhi%tMiphU?L;tm1B%Ny=)u{eGx!+E> z4^Y3s5%~8Q;T@t;I*xnhXznUq6bA<7QJ9^rxV!7Tx6Gj#JElkq@Mkjo+__iF7`6T1 z4z`q6oshA@prHC7`=pGmTjjI_v zj*Am!Qgy-ZYx1X{v<#QV7>Ivx`@I-a_^Ziyai66a(LT-e6&2b-#Mo8zJbxm+c|__6 z*b&M2U&DFTncb9e_wR!p?lf2a9Ys;i3P`sCbi>HG-+)wYH2L%5is5^YF%pzk{=w;s zu>X}4;)Sbd`pm)nL_tm@{;tm|`~)T6Fu01G!!tLo)!Behxx|pO;CLqQ?a$i#dpjkL2=mUwB$|%|Mc~> z*r*RJ<%}a_c4lTQ#HtccjD=MfYpF3SiK5;#oO=!pxuy|5##jvvhMp!Chynu-{rvPb)CA-IJxa%kG1xJf zG0;*j4uP%$-aghojy3`gj`jjtzOL`>yqv6UoW1|^6XZvKWN1K$ftzZDYu}nem zycGEF-A7~#onNnLcu$Wa^a&ska(6lk0d5~}U9E?2z`pw{NYxtH!JMCwrUZS*_)8)2 zen^EnpwkN)d#YgLHvfVZc?urWG(7cYdB3{v_%hoJ&AkI09`P;T_trFphlCaDU*`LP zld%G=8zz?T&FOD3y8_`~4Lgr2y$mBg&;LBXbv>z7am8u;aqHe(*D)uo#6Sp;8;Qmc*b zh!f!lEO|NEc7v2KW6U+ov6@+X?d72$0d|bc2V>UT_7=v}PQM$RyT3UT%Mb4ewFZB` zerGasve(_GO4cVSkNC80vdYBqLslW|;i#vF_R$}EOSiOhsmc|C#Cu%oq3TJq=RDM% zeDTYTKtj83b!iQ_Pap}WAmlsrxKIvrGx{mEr&Ihav%9atD~-E7)$KGCe4u{0Z<(~* zdvrml3ho4UMOq$@Mr7UP0K0UM^pX7S=kcE{Q8oWLVYc{He$>(mVw5a)11}ytSp42{ zi(fvp@L!wuw0Y>S_PH&_Rp|@}Z~%4H(0Uw%-fB-i_PUuU%&zvW#SL~yI1rb_*kn#; zTF5vhLy`%PXLrL!Y8>oCTnXW&bO9Tar_wH;TQ#Pv#zyyW%azciSbo_Z#51w0YgJEb z|9nkr>U22s<;7rVL_}}r3Z)4T_Pt38e)ZfwB{a;l4(D@i%%+>mQEl`30ev_w?m$}* zJ$aV;>Cr6UxmNmXeni@#J=8PQ8)(43nsB%1A9y(zp>e5g=iSa&qTC!5TCv~h%A%01 zBOR2C9q1P`u(j5Ayn*x4(tDwGB)2xA>n{y_Q?EMiD%$FZ}YUlVKlc<}8WO3zjcNSJsw6N}Dno~qa{1zU8tCH7yU0>-v% zDyp98PdT*%Mhg2X-A^za9J1Zb_j3J#|Jp-xqq@_t;T7APu=8&+;%A>#SSUG(I_tA( zsgWvjP>HkP&+c-$=CF0G<4c{f1JUEIXSKmML%ffn>|kO9ZHJbkvVCe|t=tv!Qdlu;Au8I@ivLdDMsd^YmE!jg(*o``Bqw zKE;<<8Id?S(+*u-{Y=OD!S=S6mhy%?nBNt+^P79MC97<#c4u3-zd*Hr$?PV#{UlvpDLtx#9W!)AjMHY){*N3od>Qb&#S@e$BOZE?+Ob8D_24K=94T z7YryTX)NP9j)n)=$48~d+*!-37~{s&nxN1W;r~_iwURjh0sjA{yJ8qJ4?0^K2T~jo=f%Z+P1e^ewCr1&8zN~rft zp*Jn`FveV+6?W59xml@YYIMF% zBT7l}#aM)B9SyIG(&|p7hWeoGM5MUEwJ36n|{f?wc<~=I#qHXS^813Kg?&u zx-;Kct}d)RZhwT9jBQ^Y+-c9p#%|klnkn@y)e1(lCLjj)-SIm}ks^kKKa3idqLBJV zShoU4rfJA6>zJsj9;Co+eFvV3`@p`?v$_pE71@RjFf$n{nUC3R!u0NOHb`~jAz74k zS~(3VD-W9&!eT4C>6B|BIZCZ;3!1;r(umntlBfe{f*ocMStdU3vJ%hr;H_O|mu+0EB`$XCtF>IPUE-@BAf9h3jI|bzBmaSd#(Ao9v zvG`S!5Jrj2On*IW7xFM;{&eaOd&!c}Ubpr(6Mi*wMw>0Fc{Lh!Fv!yfmU~~z{96!k zslsTH_VIu0mBwFX3p0!&Pph)__ahAR@u%oAA&kE5$eq1iDE&R_?)&K5nQvxXNB#w8 z?8`|f2daBQHepTgj5sWXDp{r1&HcC~I9_&-jGfviRDWHS<_qJwB@3_z6f5?Y>8TRG z^WNLqJ399y`}Or7!h3{^U)YCjjNz{GVc-h4{z_<-l+9Kn{SSe#Nrb~Z) zxq;enI_u6!l#@A`8!vK2?$-#fad6uH{gaeWEj^!~|L-6zEXteY8pHeaC`h2D>O<8N z+201EF%7E!MB@j~*T-qEBnj77)}^n0Vm+E(kpi=&%HmHb5mWp_V9mp=w4eJ$?S5$! z{^J2`5du+qBQuvz58)%6Wh_6E)%nkcMm08!ula^XhqeBN4KNYhl`Upv{3F#5*}s`< zIVMH~!6h-C7pIs#sw4Fof9108d3vCUlZ8>0Z*L%pmcqY*D`~f*PDqGhbt1YOf1&%RSWjUwP~i(+z^%hI zdU>wGpUl?e0F`-Px4t=V`j|{zgw=bMY4wKX4}!3 zk%30u!~9L~>e`ZB<$<7E`^m|Xx2PaEjJ19LDJ34Js@wNPEW0fd7K8Hn)E7g#S`T7V{iN6lW?O zG_oi$E(C1v!nMbXe3sU=#SpPeT;t!&G)BagQH$=Wgrf6nY*T&q^sAM}AE`1sx*-*F zx5{V6W)^E-Ciw&mDFL3Qu2N;kcUho%hA^`~|9s0*o;`oJ<(?^JtEm6#_q$Lw!x1G* z(@XtVQ`b`5uV8r@dPdK+)4so541LB7}QH; z_Zi~jib8eoO<#~lyZN4u{YgvPdTjPDEZRa`ew347t+~MV=ImQvTwJI9)r=uD*eja) z7}*^0%lY9Su01;~SG;&1IA$Ri&56@L1YU1gjog#Uwit6f@^^5^`C)^z(S%!YICbTp z#B^}JLcivB*mNo&AVd_cVpB{8JnN6MKKnj^w83Ai*`GJtfw__eu-ym($l}Sfc;t** z>pFL@7LkxDG6QDO?`p~?TNyZxvCCEVf%2bG-r)lM9WHn(1G32!@?jGeW#iF8{8{{& z;oEs4zwV-8bCl=7sIzmyWUpJSbkaMlno~uTwY(33?_}8RMazJo!rqP$pjNWGkAF9) zb(W@4q4gj3OjW0A^UzoREc>l0$&zSX z^Q}_|z(9_H?NN@7_+BBe(S#dxo3K!&6ebObo6v>svubvkNmAg$2x}7f6H_q5+n0Xu zyFjPBQ=+Br+yscx`8FS7x51h-F1QsU+ZhebG_HyN-(YjTValChE6hX;-=A~iFT734 z<9$!N>|-FMrakJMg1k8w4x>gAUQRBXnBT5n>LS)V%Rhabf>eptCf^M{(rX$26fM!7y8f~oJg)GGd-j%>(IPE7XiqLHbCv_4gUDA4<)=eCpcv8)|; zXbkLp6y@{kZud4L70qpY-hOW>O_hItKx+|);MzdBwM}^bh4c_Gc+T3EWuWTm{B81i zCi!cQNe1eD%#VDhA?)Op(>qeKHa*OLAtdBb*J70N{Uqgcb#>3{j9Ca3_xbxzvl zJy6_H1;DGKmB#v7G++-rx;oZM!Kxxah%=!D@fu;r&$bP{j|nVJtGbsJSN(NC1&+Hc zhu|SKiJza@xHvnL@kjdP&=QwU;Au#4xxxnQaLikhJONUh;AOf{F>+6QGD@$I%cP7H zsp0AHw(twOs9b2!_9e>;)t}+&-=Agj5s|&rH4OPxcSV~L4|(wpAtGS&R?79{mv?{?)cqXDFJUOZ-7_)}hF&Bb@rczn}j0-NBDV?Ed9% zb0xzveebqK@yK%|B@vPb87RaXSo8JvcG!loN#HZk!wR!bbF#-i^xsd)Dp%?S?dJaW zOXIof>S!WuzuXG4}nwe1+Nz}P&U#;(r#i-*vHVyAuvrTrP`De&p!D4cWuGl zj`11Yoz=RP^Vus@f}|8re(Pf9j7XIj!g$NV+@j*<-}TYozb6Hrvr3y^hR#1`k%a$7 z@3vQGdt-DRyluHhLRukcPQ+504K4j?8pX0U4rYCi*Q!c@6-w(Y0H$6nNqIrAa7W_7|; znO<~_^!i4^nS#Y>jBCObc_(d5-CBsCDX?7q zLE_*%tT;$UkpSwy^%-6!Nz|b6A<)+!_U1`<;9F~BUsyfwbY(Ap8(PCI7@R*ATjT}D;{(84!Ui0Mm zz+%7m^yFx`wJx2QI5|-^u2qQfeb@b@sn`cug~9vDU+lE+hrqk@SA-)_IQ6-%Mc3QE zU-BUDMXkth@PUAO)a5&@!i0b2Rlsa0fnMW8IPcqmYKMq1WSnQV0RQ($wVv_r`$+@~ zMvACjK*TlEq`HI}nqays!pB+vp<#R{%gSm^41UmCUBQ+@ z**ufY7bfm?t0^Uf025^3y zxA(xehB0sxomoxn$U8LdZze}HzrVND?X;r|COgE|a(V*X8t4~VK_PA~2Q6f|bxJnH z%z%$B<{9MSV&c!&RwkQML7u~ovD^8Xg(x)fqbjGVTq>fh6KiNI&cgCxOH_n-im0^# zjp26l#;ns91gf@OcPm>5(;*G$`s_xve0{9DC8pUUN?fy zmJ=-5`O6V`E)r9zJk*Bm;}TDeW0*T6(U z{lfj7W;N9q7Yz6}L}|@FRu8iD?-NlX(Y^&mY4H#~7MTQ3rL#2B2}zVMweg?3{!3y% zep2A`6s<4+fBl_AY2rU?B`+sM@GB%0B`$Pl?_XYel%*o;-Ib53Ne(E=>?x)EBs`zB z&TFq#>^%-=lA-!mHlk6=qcVs7KGe?!SwYt-ToWL$GfOo(#+W zUDVr}YAmfr%$j9S265!iro_kHElWIR3ZSKabXtBh5fx?AOGxoJ$Y5L}A~G^dL)M-g z_1snbngx|^wH=3K^BW&P{^V#YoN1*3y#2J3)B00C-(5xdXPYKL$nXFG4vn~_0h1}%Z*Dav0I zfRpGsOVZWb76y4#bJEJ>1Mf5>N3`n0UMG*8swz{YM16dTm+!p-6z7eXh=1=oM9pf; zxfS;|{2$%(_i;4hdz(E>E$*A&6j6aLPsrf|_m)&2?{FfgAn!fsdpoRe{b(A`dG08& zp4I(O--2q;!b|YAlJs49@H=jVFVc8QDHe902m%t@^?e6PNE-z2CKJy%DFCL?(lC-= zx--H_N~|V-_GWWq-xs%0bPs$1b28bHE6)sZ6zVYUGA>fEShI?zM2AQUr%mfWoP{d+ z{!EOgxaL-12H5&8t~op4Bg=|HQ}6R%FgWB7RwaK#SP-X0Z|D4%l@B4B^;lXZcK?`d z8MS3eE00>dygEAak`a*UZ2Dedwg=|@icuU4g-~UnYYsCrB{$C?KA+uRwH?-#tYB)E z@zH;&AIZ*K_55S_!-E%BN0mD{8M);t%9b7sB|j%lr#vT?*$!p_+CB#t@eNJ(`+02e zvH=U2%YkUdlP^zJf2JN4!}~*~ zA$wAkEpCUHF^RORuEv(g!Hg6i*&Cvc@6#>mC=m_e-)OVv)cE6IY|j}H0lBIgt~!Xw zyev$lU98?kCtG<5+`dC%2?*6B{4sP*B;8H0FVv5^CBD()_D*qRUzOBehDCKh|G}$ABj8 z!W;~-%glfJ1ttXa)i#okp_d23LwXJ5C^uS4%4EJ9n~4Kcr*o5Red-^z`&K~~TMW~sj|Dkq=DTtN_?)^HV?CU@YP}OMm7j1iykOhF;V2Av1kV!RX3Ubh=Q9^c4?Aj57 z62Xno7)9zsO0c3ufU%z!82Q(*Flo}Srhwyf+hm9x=f*A?rbS6I8HV z?7B|_xu1#GNJRq@{T0c)d+`t_;vG0ce`k<(^5ByO7=r85gBy`}9k829qr8#q8E};3 zOcyNBkMkawY!`yfKs*T89v#LsTk=8w?Vx{^L+ke}f#uOboQN$E@=7(et`VDIu(N73 z@D1`gKYz8zmFE?EI340km{2KjR2Ta_&m73&awwCGOx?0=l#^_HS#H-iqytItF?mwkbOr|H_K;L^uWflEJhs8VCT;pnJ=%~?AQl8XG!8l zWiEA({UVqU5H)}%`!F|N7WXdhim7O43t{YeKZ%q&0s>>)zdN7;ee)a8Ccpw%6>JQ7 z5y270G`wncG*jL zdef2=Q8A%l=Y0L*GtgFxx$ad3u(KH5|3q!UPaewwT&~^9l8?~z+;>cF6xTq=XMkM- z&bRrneUCtU$11?K%u4ev1C?1& zxU*oZ&-db`a6}#RB=@^7^L6{~o^cSU6jB#%S6@JU9%qDJV7U<)MhTo zq$^k>?>p6v#=RS(ZP->XSLE1v4VgcA?J)n3GiG2ij4GJ+`AMQ2-=?L_y+SR{PmLC3 z62K&C45?q&fdxg?e^)uxUa-HIGjvnHrjjgw_NLhTjMLxmJ8_(uoD=uf$s;=3aEr@$ zp9=To8wR}XX)Oke9>Kn9A9R94gM&{qzXT?Evd3RDg=_xpKLC3v$nAr7wt?5_>|3wS z1s+t!Mnx6@{pX2$;B3*7cj$0Zm$!C*?0!;vrSo8^<{XblaM-%0C?2v3GKuNjr(L&u zYdv4IdZbVgj`WrhW!UA~hE+O;(JBxue$OM<5z>Qxq1e1E#CXV0_YY=p*>_ZkMd%leXm!>_q02(-WDrZ7f!GpTA{^PYY#u zcn(4*4UMsPk584JLLpj(CZ-Md)&HGOW43vwQ%Wv73A!R9>GxIi6@dTVg$xBF z!1&FshXlW!`(H#`Ek`R@v!RNDC9zYH0)52<*6)%L4y^!nxE>On`DXVu#3CH3zZ75v z%m*tVy`e!<4Px3rZNO;@Cd}_HPhwN7+A!k<;Jb&yhhL^C+i@vXbII{bRA(tqmEyyVP97eVPnr2>CR&(qqP8$&{m z)097COA@cj`n{JK-c=Cv1k(n)o>wK0w2jxA1pf829wF%$Q@n(!E~=N9Of-|N@f45c zj$aQX;$@>Q^Vo$kRfTAq?4hL8e=wfQyzHT%?+rz*hPp%#Z%^s5&{=a&Db!~LF5-YN zGXtBcn5}R++EQFLb8~#R79EniRyIMHj*Gt~-Wau5!(A_$`JF|DI)2^wy(Yn*3cq)K6=q(~y9?=q`q2#+W_EU>t;?Ct^)wzPY3zs!A-Gq;L$H2vQJ1 zKyPPNImw5)xZbL#vVFb_c$vA5gxh|1tFzOe`i1E+^ylDmi;Y_|R3e&nBo|fx%pW-U z1{%%ILR-cD8%bIVa`6hpjJR(j>wHgAB2S{njHh6=pp9=7s8b%!T-yL`NUwf|IYPog zVzTHutCT8bc%&1-EVj2IAIA)}wEDP9+{Iiwl7l~8l-w1?pTJUK!#WV#8uP-H}^0HcPF@{?q9fK(Y9bf%hG z0*V|OWl0t^(t;+Es#LM9xkQ_;QyAB+JR<;^#zhonrYAE`{en)DQn1yk*kF@B!(4A+ zWq6?ix3f7Hr^j=e#($j<1H~3Ndm9AXk|%L1EMB)~=_~4yOA3*y2JUomgY}XH_01JX z6^-EhDVPd&D=qEatD`B}=w9C9o!+MW)Ek!Q70D45xbX`*D#dT6W=bI3$PU*73}M{J z*kW(o|8OHmmT*~9UNi4C@xB^q<007=KVXp~#*KWR=G+yFrLemT_mv>|Rb9?DM2|-s zG@XepzUNA(7ABIy7uo7LMoTIj{ANXXgEsL8ZmGzp$+mMuzhT2LYKNp3!b@j`^j4ke@# zO~IB7`VkN}5>evJ$h^hx*FvKB#?nb%2y^gamLs_f;SGJo_X^nBt{Ita#9$ajxQcr) zt3{@(u#}3TDO~uXmAy0xTT~Q>3~jhM3R46LF$R^-D-*l772= zHzKLeM3V<0f2hNqs|583DjcyS~5ZpaKyYKPm)+|DQ|lS6125(-3#K(d8rA8UFQ zhl+0_cy^dv_J@6n6dBHSi#`#RXl7d}#f>*S9B_6t3#5cx2Z-rR=uS*jE&-l`qlmS- zv|+qUR73_LQopCKAh+&kA9%gETNYP%c=^;Uvf@vZc#{Pifc#+CM)JOP*~ku62MVQc zs1c!y_^k$yQHw%;k&@s)e?zm$RQ{b5r>0j0PMVT3j~=_j4x+LtN{1ZC4Zi!8i?w%1 z*!P-FT4mQ^7^wfH!lH*CI%{@!bYX& zMx&(U$sUdTW<-vOTyoLCFMr~0qmC!k{bbQgr+=V=;`vnL{iXEGE5eSRiH@dN9%i5o z^c{S3f>Qr#QKv7((K?$$G-aNGR0@=1H4$waHNtPnTk48;lMYrVr_PKaZK}#i?h4k{ zhf=T&cD^8ejrHqMK0Et$Yz&DQhvH%?gNI)%4#BzN7S20S+a>dNi?+%6!Y*<`m{ID= zYW-uc5xR~vcg-nJe%F1aOFQ}Ul`y^8U4bcl{&%BUl)!~9xtJ=O|*?{3^TCsX7XGG$N?{vj%QJ{ge0BFur< zB|O)y!lj8w?|?!uyoT_U6yt4GPloOFJ@-KcsgQ21?~A9&#AH|I!D)J#7wff=jTQYD zK}{2X(NaoT6`y@Fi4nL4elDOj_J~&559)aPwoAK5W5O<=Moc>4o@X5`(GSxKREo(; zngwI%PLdg;ncsiL*CZ-d%Cy_Cm~WvIbG9f?r6m>?TbtIed|T`P{#L65@4F!R0&Zlvzn^lj~x= znYpQMKNm)qoDs{tdCitA{*+~0DL)q*NEWY3I!)=qFYM||(8Hbcc+#xLu4sxKh0G%V zv;r3G;eMA91Jx^pSqP^EqMO@DWh$9XSZ!t&k3Afg0WjiW4ksZT1o= zH_UtDA~U*J3YN)&{dCreGmzS2HNer^gFSqbONE*gPvaZT8-&A)|9RFOY2ge#<`1CI z6yvei$EoP3Wp@}Nj>$&Kuq6-0`6W|=79ri^CcEM}RGcCM`c};}6@Li!%qpPUO3jP| z4eVNl5~K&bVq|xN33~Sr;_;A=8DDQDFLj@TSpGs@idrw0p>Ox*SD?rj;U`p^Di0~` zC+JRM>}6yI-t})v?uZg?8QcVUk_iQuH_0uzS}N4YO%*n+uDAJjamj+YbA4dqH*qRI z>=wCF$Ujxd4`MD->O?2zz*C}FNNkVyQ;?I$%K*w$p}&y#n35^RKpvKY1XHd2fJ)xr z_EBG`Hn0#0Z9B3}!UjAcIgr)iQLC?qUbW2+|`sdTRkCcuAZ`?&k0k%T7Va-{5> z+MB2xDwoWvA9g26xbcKnF$<3)S&+ZO!whTUlV9dQds7?9113l(KwHI1w?z578d@4i zW`Up2RqJv|H9^m|;tg4Mounr7vV=GKj``?0;lYfT>#qSK9ut8Kk{t@)(wx57vb>fe zD|}2DQ@BSrQ;c18JC@sOG>kLnS=<7~Txzd(1}uk?=RN55!9_>ij?!F;92fMWdl!jj z9Cf-3)OUL&JP7@KY97?aIqAQ%^+<2;X-%n(3!Mab{kt`}?yyZVCx!b-Oxw~$ZM5Z& z)xTaLTp4T{^F)Va#=aOyG(+H~dj<_wwlU3qJbltTu805LagB{&-R*K(#H9%Si9<-C zJ1LeH>U73BTZFZ48t7|Bk1CUg-Z(xb03GiR$KS<~cbguQs!*3QuE+ED#oPNA0WTrgnr-X#G$?7AyojOpU zZM6FNj#k~B>;hggou^T-(;H@PO7}xIwCf_8DA|&qGVn|DWd6?jG){xH3FMbN>;m%Z zGDwT45d--gT?xkRf6*%IoY-@>->*0k5TB7y4I5ww3SkGSJ^ z0aZsh=kOsMD%8oRyjTsMHbD09)XxPguq=ivl6i<)#5fK)FZLvLkLRufrCW$>7DT+w25eK`ixY=DFlAVx~C zq1@NW$6UBuyzkd}wCdr>QeQ^S6f^7nCa0IX=p)O&nZ`zO^{3q>*F~)`q zNT+F_Nb?r=oY<=;hB#Spg`f|7Q;3IB{ltbYsJWa@{lWv_0b#i%?1~ZC?n~+bjn8`i zZy`<>5Kbbb&hQSBg%<m{qdRdS{2~>I9T+iL-?LljD{oYGb{7d;vUH~p(wWQ z&#_`wTtK=z{~@^Mlo78b?rn0(QS#M_7f<1j1A3RY&!1S#g(nO8Zm_ObArUOb@vavm z;%6I1XNCxeUmS>(tQ$|>52}OTS^i_|d>?+A*@-@@1v7Qf`u73xXBV0w+}eVB;7f|f zUB-Q|*^rLmQ#QeE`&_NMo6q~$E~}oU-mC#0h3>D}p3z0P8?iRq#XKatZ);}6Lku00 z03TrC*@*YRdET|5qjjclH_h&j%3LCu%lqXeY)>0|rA?U9f+y`U+Hqr#%?^R>OChv( zs^3bY)Q!}iXJi;O7T3ETUVO#k8ySt_*O(Jy7GK0AYr>&4ffM|%DV<&H?KK^K1w%uN zRrsI`#apoqy9?~vIbWqQH)(Q0S=V2i3ynow!RW{)ONPLzbEZZ_Kan^2zpBa`#P`8z zsj;TdcWl6#C5jT-PpQ>S0HFpS!Td?G$|A1*&WU)1vag6-3YU!4*YgA&?!Wu0<4H}> z#^GS%QKIaz4dV`9OqLx2SIU+dlqeW~H@cQ={p?JLn8G8EmQ~#Y|C7*UyXT{`CJ7IC z%eiUE|DHdK2i32VX9Z$Og)AC(`MoCC=<}u@FW$0rnmRX;W>i$k!w~wrlZnErghGzI zQDq5w7=9Cx5Q87+^eqR04l<`dvLd4%jVOA$6JmFF_R)35H%cs%G^IeE)N5pA431eH zY9@^1J^3NHHP|!G`ms!f4SE|^0-SYmvT3QG>XLm*oQM4VkVqR5c$Go=ja82o!4~)a z4jmN*JfD;(#IyUcT}9ylA|e*KAU@W9#cFU_?)0H9{;o_E&65t`H(+s)wqZw-Jnu0z zU}W;;xYXS>E@6*!Psm74&^v~{1p!CDA|dEF5ViFATwsT3oole`+l$Ld$&eC`F>q3n zIOZsMK15O0!$gi1CAmh2f4}Yi(W|)fr$^=svhKmozQbBli3p7x^!A2(y0ixjzWj*B zfGp{SYaz#nZ(}?{t`X{sz~z6eSN=cBNhWkf(s{GdmyX6!nKd8~TKANY`LQejm9MnF zg|1~6_s%p3Vy1OfM|P5)3yh3N4&6`k1mkg`)KkEka6YnbER*+!$z5c>6dSEkG@j!O zwy6-wSprc6BxJ7M*d)A$Ufd zXR;VU1{`c+w(QKP1LZHNo`>B4E7U{w!CW^0GQ2g0_=@QUkTHpuRXfvdpj(J?BHB?t zBT5!!C0!x&Ve0EdANr}1Q`66M@=wUxR@y73oa`8slFug`4pOIboU1YLVc0JmfgM=& zp(8OTVvucb3z6Fm=+BY*6ynWUmvik&kRs>!rzPbE!7U_cr%i){^T>sCk3!A7r)@l> zV)_d?xLjdF;O;V-w^N`@D`LMh908sRStKzYVZEtec9p*eg*I<_uo~QZgj=9dQ<=(V z{j;8_o~#I#)RLuH#=wRAP0wpp)n!b#_6j{#gZA9qBu&t1e3bHJjECEgjp4g^NCo8{ z-Kd9?8P7J;PbdDb2xHp{P=S;Pzz&aDz$+xvByvc{>nN&=OYUldD=Ap$v`AI+P?L3k z@rbWO*<4=zQUD~ylL!##g?gXMDA-J(`i~(8*{1__z>+S4X}0&}mSXrRd3W5pQEwRd zCFbcLd1XAhh{4Py@-7v4MccoYwnBsSE>U63nnz2Th}AM_Akn!62XpKYN z{|@4oTC^-7?XjsC&ig`NxM%ZYdm0FmJz^2O&2#~%L1Q}qhuiG0B4{vAfGkz}5GKIn zk@fBK=lT9a&JcFNPSAxe2sgKhg!w!T!ZMUjPR7Uxdle|x-7I-G1P;R2K$q(sx@f!@ z>d|HU7rG=z08iH(Cf#?ji?qWC#PX}BPhOy42@CQ%wxD9eIwJ9@>})q{BZH%z8{B(h zj!1B4+Z1UbI@b63pdZrfvcqJtYG_Iu0~rwpv>JY?qVDb;|0sr^lmarHf=o=;?WEK! zs%N@^Q?H%KF4N|v#JJCEUaEO-1B*Fpz9OkDw<=a4M*RSmAOVCO39k8JB}W%=DT1!t z_o%>^W@FaqA{6G(O&OhNFAOvY#F;KIqypHsAL^%$@EjW=Isz7*2rt!v1j&#HFk&bJ)1> zB(3LZ*;u%m$|p#G?F~fh_-sFir?|eSIX(#RU@sNT{!(cE**tstK8tI7qoAUQTWd@Q zD*jx~vu6|joWs+zHAoxNlRabJOzPbS3I5YJ=$iTn21CN^hPNu~`PyupW>1QKFIC~& zMf`$*S<69H*6iyEudpE6k-tpZ3J7^=5DRVBbj0utL zgN+Qx{xMxK>G2@h6r$O-93{P`1ogc3yaTjEGXz>-Y;29p2bI_hpjXiMG z6Q>XIe;_z-c$PoS8EC0mWl#vsacDn&!9wOU zdU9_OCxuelEVd_@XYFrBB!e{}d6D%oM$+&qf+2h|y;=NGj@KWpjIB_^ju`$XzbfEy z%Ti2YlGd4@S-NQ7yzpW#Wj8dTz1_@k#Def%Xs!###Opl}v$7!Mt%=z2zTZ4NrwYIj zIHkZJW~;4#nbjvIrEv0afNszjnN`2V2ZqFIMO78Qv#&hcgdHT48Ia{Ff4nbPu8Hk> zuem8zN8UKEV(dN*Vb!$F=GJ`38083OWg(N<&S+}=Of2HAr*6IhI=_w`4k6x#<@;dB zXSPT2EBZQnMM6;`u`f0TRu`;Vs?s@zE{YhYH}pmmrFLNC-Yc};{l*)hCzKay@5t}4 z=Pl1AGN@D+#OB$Ch(C)#(MAA+eXYgMj#g@n_{O#EB1KjXo=`X~s{LkpzVgSV;w!D~ za!U)9$f$wg{xcmD+k)Na=)oUnr?9X6&CN#Z`RVo`VIu4?oNK1}e$>`&|`7Qqfvuh~nZ-R%4^wqf)(?4_hN2lV*+2}xyj>?=-b zg68049cb)1yPCEg*N_}PzA92A!GV0etQ3Iz^od1ftHsm$sG!upeK<<0>;tn%k}2z} zH$_^VD8y!P$IAvCH0qd>*;Rndq|bITxUp_#stnB4vp(xUpSsN0pz`5ZQ`37OQypL3 z1$<$+M72dEMC0d8NJ=W``!${eLvc5<80xMU>?h&0g#ncC0=Nlq>Al|0rYWjrkKIp}qQP zS~@1xu{(HZ0fL#?OPa?rJmABmSA)1GdpKK9#~*xYo5Jc3>Jsa@Xw2$`TaPO1#s!f@ z4}R;?NEf($FN{Sn1R*TZihf`I6MP;eQrr;M!3X9yt!UVu}F1D_s(H94BXJFOqN zP#WY=Q=qP9!cM9`UF|mp>gS6JfP1Li?r}GV#_xPnvxj9_5YSR>^aSLA*C37!kivG~ zxB=HRgou6Rl^fN0O_Y|A2J&f7z1$Rp!h?!qBNF54<7XWbhC}eAPQ4mCBF@I9ON`AF zps{6l_{v^Lqx|SzSbO#-UF}q}XB>qN&6H@YBv8KNWIjhK!RYD)r05(yvvIO@49v<) zNGs&bH^g%LMIuF0ep`gFzj>Cr_JABSCv*G!`guJG+Hf`X>Z!Sgz!zy7%A6c1!p)PZhW_hdFVfLvfst8RqyxI)V{J;7i;$TnvKK9^5ys0FmMFJYf<&{Ca~$f z=oFEhXlFW%k#w~<)6(1bTzDzF#G95FF^VtAbVw?Mal2Gm|1l~?7n@ZO*uIU7W_u?Kh#=>&~Q`Aq+;qO+mnl*qTxszkH}hi&Qy>DmR^bj8$W)o{gHmh z`m>H-BtdyBfPv%FYcUd`gf!)kAF$C!c=wDq_Wd^O;2RI&uXl}pYgLJGoCAja+t1Nc z5GrboPF6tlbHQ|WIwW-?C<)oCLvnbVFRv7k%mPqHN)CaEo7f9kP;*M-NbmL9$5b>x zmh!gUfXpp%X)zPAdvP&vNL(CnW{w5a`7fg;%%) zS-1}J6I;552EQNt>bV2vc|=elKE)ttH7MpmQ0lzQ{*;Zp!)ucqa>7)QZiWr;2yzLu zW=+R?Mg1sxsh3=(-0S)d&Qw&}s+G;gLT(TzDM(#dkf~;>2+`#?ohXTX%?X`O>;oGo z(&N`oMWjWq{CU{hL!ZZI(mOSbJA5!Sn7t@ZO%;=*Em4w>O~{PU6q_*%-^b#3nqr>M zuZJ^^Y8P}z$oIlTk8_Hm)`mn8rJ!YEWIXRx9cmMWnDf(4`d4HDr}~nH_9XA5^{!x| z#Zg&^=kn6uD7(Rr$ynuiTFk=wOdDDyUVIa~o}5!W_(uVeq&g%vpXT2q;tFd+b%X&A z{>6|T7zyD*?N6zKO-pp4dPzq{+PYk{fi^3nY0_i)MXGX8m$#naAy~qO(x09EI|Wx2 zXbn#>djUUV_szE2P@3lECeo+qnrVxSaspXT^?*=@CP1p)obxccbWA!K+6ECN>~859 z9*K>Y(IOE?7{%)Fslv4Aw)oav=R*8n2UuKH;_f==|1iZ{dLP`-yu)}weL2~XJ@lGo z4BXBD=1)kPL|bs(K=V5RjUM$LDHK zbJTAkN9IMyqd36EQYXN3PwduuE`9`%`)f14x5 zbQbcCQLl<>H`T&_>(*&Dn)Rw;^4+1qp-qFvJSg8nr{#-u=u=)#NMOH?$(y8Ec<;UL z#nvWbmmOcBp9$Y|hR6r)%Ug%AebC+^U8}@%vhJzRV!x^)zulS@DnC9s`ah0M%Onph zwfo854w?ai796TajHtFEGw+fQ<;E@DWp&VnPyQIeE(KE$u(&EWr$_MK5pg-@T- zl%OaOItbDddIu>&0BJ(#9fBYwAc$0@O6U*}5UHVsD!n7Uh=54%y$DF}ReG}dzwhqZ z{jeYQoINM!=E+UY-1}tanLE$S{AS2+Yckjh|1*s)KnV`kLtjFHO&y--!k~{mL=@du z(dl;;`3hTgv3HOQ*)fC!#^9^((5bq*@<8v4_M~VIR{=im4b?nkg8||&gzC8B3k7L-=jLS(GNV%EezP}XIRqz? zMeOB-b-%@76w^+%(!XI87hS(!g?(*!qPL!-X)O4_?dJh=cE>015ZsSl?)5hoU=!eA z#OMp~-|rBKlYf5v!0VM!K^!Bmk5pATfQ7;an*N74vT?t$Z1RE%L+<~uqW1VwLxiaV zn-(r0j~(^KLZBf|traagh*Y9}tFZ3xd45!)gDu@Kf$!|t`45eLW8*JfAhNr@U|vo2 zpV*gg@hFTX+-`SE=_#?p3Zz>&8Ez8NRL2=zXeL0`7DU{SeJtHACW{d%69#1nLhM0a zP*@s_LhVBlKV9BabGV`q(6V=s4?i`r7hfC7mXf>Hp;<|%+%v(?Pqk{u_S8$Uungog zcHzLjwz!K(HkMGVijBfep@GG~_y6k)3wb_G;{FP6TD>+!#gVu`rHd~j{e_NwXDs2oiHVp3w?ha*Nf8@NzJk8$}!=Rr4SMQfT*7{)X zXY{yy#8udD$?cG1CRi2b<>5KO_IQkTh+(aH*I&dS`MfEX*+Cg4wYU&?*;8fXh5qgH zY{+cN)y4L;$&-<)MjPVxYSAqzlsW~~bsuBy17K9~-8HfMD-LoPUVL5J&H1`88^JMR z<@Do5;>*m<(~n`9YkxyR8@^P)A_h6Q8Kj(#o~YG`^&G#rODpE6yKK4J1s9HZr6b+r zgxQ3n9#&kqea-s{FR9efzl{ABCd{V3-I6h#!9hWz_N8&kYpJ^aKBeml^{}1-Oq&{M^MW#=J;U*Zgz z6sbrze~eW~yFJk4GH;EV@Gc>77R*T_48I$572c_`3)s#H<2bH=pq!$c%6E{`IH20N zMhmDGHinG|&CSH*TTy{R+q~-z@1Ib8>Yrw2Eqj;qi%0WlMbr!jH)yXJ^k`=mfAC_) zPVH@-#Gx=&e?TqC(npe-CR1c)mPqs2p`v1;%u_neSo=e!XEM#IG>-slr%7o+;u1F* z+fEC{{a?_IBwW1w0dx(}O<~7Ja7Ng6m-+;wbp7aBR=%j&V^-}{CuG2$bEOIWXKfvn z7^6v`q&Gz3%1Z|}l}m_HKMJCR$zx{U5#hL|(&GxX9nR{9!$iWaz}cm7h`!!u@|v1- zmA3AD=z5WR`XDmPQ9UL?zoo~VZ|j^vRD2>oZ##Q_mmH!~UgGh=G-`MkYksNP!4t<_ zTtfRxyA}H?oShArGj~^#SOVRaG$ZLF5HXv1JtCiwrynVOKgI|q2lV~S7EN>HBg5Nj zkuQRlQ3$sGt#Nmcgg6(drt-C0y4p@f$~YI;r}BvaPPJCeZ=(gLzco`KmNer;&F-^M z)0l=WFR%@Kr-HaIJ&q7&Gcw|F4P5Q}d^>!#iS7=K?fV?i0gGBp?l0At@j12VFFn2> z-{k92ES4X>b!Tn3JpCvu>#V)PEQQrbQIC-dc#d-Hc_FRyxp9g=t*;F!%+^4D4#&KT z{N-EKxVCQF5<^_hsN}FJ>^S^5u0Jn}7#ol*;oP(2wd4kAw(2V#q%BCGBk!Q%uMt1H z7S10q?P=$*n@a}k2#at zD(hNk%SH5mUYJKJvJZTJ^LgwiEb7q@Up+gLa+|hExvua{QeCdj)d3C(R#ohk?cuWk z2gaqD-yKvBiB&MJbXwQLJlwXIz23o&5W!?W@y5a&tli}UI^^JY(Q5CPI?i3zV~BgH zCc#OA-!rDaj%%biIufl=&B7SEe8+P(`DmD7$Wde3!SB1|Avg`@&BUWH00DOv5(gRR zxFZj02@`u^H)p()Z*&H}Lreo{lIgaakbgbimG2{^qYM~>8@e}1Qv;_7?}<;#rpqa@ zD!SzFeiZv`+sB*bIlT91eEn{laCd>@v7=N0p|| z#A>>aQ@g=xZleIh-{RV+*>sP9yB8f1%Oq8Cg!M%O@BH82Z>!OT|0@$y>~0efQ6DZQ zG-5wmos1g%?)pLg*W!(z_%wg9THHe&_`k0b&a;zPb5WJH#({UKPvjg8O+p&GD9(<< zmir=nzg7*D#@?rN-xs-`jjKPSXTaLH^}ILCE?q>Ve^Rg$*9HLx4cOsR#z>04xz1I5cwyQ`MA2Wtf~lTQ1mKuU zH0z{A-#@U!Ug=_A^?88fUF806TexeJ-FUPvoG`15YQPS&-JBdl=Vt+r9%d-jzK&mG zy`@!i$!Gb&PspVwp8O~)weBa(cO-Mj4CLO6XZRTR(8c>!S=(g`=Q_ePp5Fgn;ydac zEaNRs*Nc-r3hN6Y7u0Nx!-!8%Ofb***C0qI1T9dLR8U92(O8IQ^0vUuO~-s8jpWYm zP7?3}SB(=gNj7AwoHQ{J@DtKmN5#c7&&4AQkt;7K9>2B8Yc@1LCVd?ha5Thry}oPh+K z+UDhE@M)5()Ef*^ne{KXRUHgompV#bgd<05N7VEw9adHy?joM7@7!}kJcW4t=)qRU z1pAPgIDI2#5_a+iC6-<#(YO^k`n%f}F6~?&S=Gki6;h?qBj-68?!4Og5%{{c?qQE&9pV5+W?GP2e6W z2h9iKo>!j>aGCw2k1CEUY^Q!KM4h8OL?TudW!GD5bD5+6T70TBFTZu}8yKUa8TIa; zE&2_gTJGNDW2QCL$6gf<$jW-)C7X0Q2Gg1@acK@>0;=J%uIVuS&y6-9;Aqay;&as2 zu8!NNzeRTso!1rqu^4Q_iBDis@9aGzBr*VE^`d4A>D>+t^44D{P4k&Sf%TD^Z? zFV^b+B;!H?!gqf2I%~U;4+Rnnn|yd+F6f5~0}3fB;%Z!p85!l~Ea4QA!?WM%Fqs|B zs1dGLNJ@u<|E_N(aoPdqip@#ku2{_NE#>8H%)>rQ_o8Wd@U6pqmvxhGR}zy*uWmc# z5|ZdN(y4qjtm|U%<#<0kB4WC%_m97-9VO-HVPu9Bk z)|XF{RA@;PO}s6{W+=XE1`Vgz^L{NU8leBdk-3Po%XG;}NHZ6-uEN>7W~q>f9oIOI z`~g|FC>G4RWY343H*df6!cMjb-v_vQKN;Z5I!!u9gN*(xEzQ%XOYBkMTwG><$9~wt zXM6Lx=&`y39Hc2(v#>GOdvbL5$s!RtZgIkNT-UMmYapyOMrWZa9oG7bq86z%Z03KO z(m&wl)${EEm}aFsqpwTtV**U^;Ek;~g)O{FZ%MziIB5VfFP4>nditw|yr+GDQLYDU zVlUZvUF>Q*D}Ngw_4g>{9NRD;`>B7mZ=g+}=5X-1LeHEsciqn|Kc3K~95{<$Ryx_ehW#c(;oAriyHIXyRW zJf<{90lo<@L>DAYsub~d1qU|R)@J0v?P$`1=e?#74$W*5H5ecxt`{&C>Zh^!pb~R> z$@!0c3Q=Am3Ze;pLr{fr|Mc2e=dS1TLS7K!g?BjrWdVP8`-C9fmEDNI!*&%A`R(|0 z3heGnj%_;3{2Tt}P=xSJ{k@-M7_LSi>Wtwd*+yb$j^eF4srW&bt-Djb0UkQl@q%1l z3+}#O7X3&+XMlI;!3-Gp+yXroGD&en7iSA6qdt8IW%c;A#_IQy9LwsZ-+5nr8V<6m zz!8AwoIltwmEp5S^VJBPq5Eh!v?Kd@bvlVQ(21n4#Dq_tXzDX!nwSk4lvvw#ws`(it)J%p^1;{#Ki{gsv9T)U zgjxa>9GG66m+3h5yY7&1)NA`M*MEJ&n7;M}yFH>?G6%pgym~-@+zPJtc#ASA{ ztYmSx?|;cY`y%sg3_G#p0p!T$!%oQprsjYd8%i2l{u7|<{6cX1ljXR(c@ zImhB?BsQeDxsi*1QHon*-BZIASa=`3`(j~nbT@!|=z9cKWkdJ|@WA!!#;PzX*IELT z_{Bt}BHl#S&0kF_;U~nPH0ly@cK(+WvyQ6|5Tl1jpaaJ~16skX4zh*xQ?iG)>}jro z*LL_|aO{d~-c$Y>gnueqoBLxhjrr!im3}jJ56ZPsp_gFEMq;%ZtQ`qOtuyrvqOBg^ zLbLq6aQa5FA>u2x09-oCv)H3TXVW+prGnV?TOD z&9#T6ix`e{*RCMI1t__{4frdT>O)=WIv_<#v2@pY1MdhE!`Fpt;=g+5D%-*t^wvMq zIM^>+$ZgCq(5sbyTNQp#%7U|gqG~wtZdx0YqWVK{s4%1n;z zBMiSNQn8XO9XvjOE@UfL?8Z9 zqM90Pq@w8+8eg)EW=~DR1=QKRy^YS8IlsbWmhdR=3dA)f(_xhKFJ-oJ*LFH_&W~+o z0affPlKZ&mzk883$qnMgsc=3np+O0UT)VKs^lD5l-~PNEXZ>Le|Ht=K-=Ky$%@G2` zL6RG{G=TppD)$q6YRXmf=ffrGRTTIP_relsZC$y&t^#{${NQ7jO|?-L!Yj~gc#yAP z@n`&+Q)U$JcDgc0_V@Ux ze64JmLmY#OfQd{-@)JYk>hTFOLS+DondPon$Dadk=QlW3sSNKs^y_eNOC*^?H2nx< zs2KG=bjr9AaZ7+)%Z*2KHLogJW?oyQ!Opxy4D%TsI_J!$> z!MyboeS|OdK6qh1r?H7$JmudVX%^C)L{8b(QqU&X-y4ClaJPMC95>-BU;=M*Z zqFTw7S|1u6mr^LHH@KXy1-v98N5J0#`rM*I+65@6;QT{19rEmB)pF`lcu`pa(eIPl z*>Czj=`S%dngc`M=Jv|viM-b#Mm~%c?wnkpuYDdPsXq>(Diwu>qEp(YCyp~yH+fWt zcQJ%iR?q+9ZpyL1Py1h2rZ*0y1Hom_;gq|FXy{T;OFy%@u8T;R`G6? z=a<@@f<#-BQ5o@2nQRJ0#4q6SXl4q9rcvtsv`H&1MTShY;CXjb_oU)96pQ*L{@_hgT3Qf0R>^Xo5mFrG_O}e z9x9~>{vzBe1g;?O{ii};3s{!U{z)YGXYjd1oQsg@fWr!g5%~MfeVv$0f~#~_trp^o zDJ!i#_?>)*5PhrK&u#xY^N3s{GV7>WU_2OqpwnB+PnW#U;u(C44Kqm1XrkhblC6ib%mP}=sOJfL75CuHI^n{GV ziieJ`WL&EN1=9BI{eS{%>!)(mfCBn-AH|>BxT6GUd6uSt$@Oeu^wD-_kaaCsQ?`QN zU+Y!?P{8DoN&-%Q?>L{Q`m?Czs|@*HilSZca!f`pOL)xSPFS~F1!T1$jULo zP5Mrng8UF!|K0s&ATwnpmxkj?w4IKzHb3&qxkFEOy z6?50j2y6<_dNmJ=E4e`Rv{gYKDCKkTVP|7Ef?*9=_|aWvHH@=ir@KL(sONFCUh3XL z5H`H^G$yuq?iwQ{0w5ry$V>jRHn+v1`x0GJ7{N(S9!I~OF+w3_-JHY8`bCdJlc5(PaN^knC9r};yO|yTy8r2#! zD~){qMpvhx1oYZ#AZz}47DC*#NY1{(rU)sbbhwJ1rgbT2b+EFA*&k-F2{Qt9(sh6; zQ7g)Wikw}liiet0(u0|j8c|AW8c?45u##a1GJ*mPm9OFpGe2Uh7&eu#?^I6(VYYXA zb@#+m9SDG|d!&#!8Ji;pOV(R1@=>0O&x1(7qUrCpv<6Ul9+QEd!x_KUX#+^yt>^Kl zD0}(c$y?b%WPQx~`51$ks~vyLd=}_^v>ET!zcO9^IuEbo-NU{*K7s-RDLP;upiwgG zA>7KCArM(Onv_hI5>eY(|KfqB`DpWdP|&o^#GP-63AXa`Gj}U>n01F@ArhS23tvv) zeHF3vX;}Mu^C~vs^=0-7IO(L!?^5XGW|O)Ic@r-S{IbcLV^l2L-1%%!@t$;!gF73X z*sMd$Rb=v8g=|IRC@gFC3)+lqx~q-XDFm5fSOK24j6^oX4-I=iN}np-i_DDYda53_ zlERrIVg{yfe5n0*Y+v65M%67dpRBZwOyJI zI`82P{Hl!Jtm*M1qhKz=Q638;|A>|QIYyQD8bM6PQ~t#)4y&W6M@F_FxI@j*?JFm7 zOAwr!GHwiVeQ+uTWE*u)HsyUch^Q4_m?0sC(!IA|FjW~Q{X6}`BiGMAD*?W?iW2!y=a!Z;^?tZyKb;kE-mfYFCu^$DW8W5+F%Hs}nwk zia5rflKR`9@n;&0Kpv8Tz#>AOhLv)A?(`x;ykvkd>jLpc(&yjQxGRJi3H}gwM&Ela zfD_A|Q=+e^l)hZ?Bcdkmuo6<`W1^D+LBVV-LZYt(cK8CHXfyvp2TGlO1w-w}59vL= zQwFTe^rLIO>RcDd_IsuSJZ53iB`e0HZ=JhJP$^=DQn_%euD2XFs1#EC8!k`|6%8~6 ziMk(a!tB)WCVw&Du6E`AF1UVWXF4dM7?lDsrswLMv`onBb74#5Wa3B{B;R~;KVD|B z(`9nnMnMBK@&9u;1BiW6&_HXCgD3v?@a)eDFx6ZmT{46e`(k@%7X9aA-6pnI>^MQu zEwnPj-=3Hc<6U50b2^{V8e9G^|I%8;!KV!Ac`P&kSnPNe3MDK*yDbOLy&|4}wfp%W zct=_|bfk~uYifg91)t6l!1k_&w)L>EACQBn4VV?LN_XbnMcXmz6?@n&;^W#z34J+( zKJr11K;}mfV*d#3O&gM&0uV4qkOPssjT8ep6j zMED#PkVVitjmQ%~pCmd>%K*oTpj!i`Hi)O!4|;`Rz#5FaWBpuKLY|vnpfR?V`|CdR z(Ju2F1<#~^Ban{+T@&`fkrfRCU6e+4_zx#Jp^nhrOsa_x|C%pa9yl6Ld`oeuX_Sh zy=rqDeui)QYckvXM*Xmhq*zZiW;K0EWGC5HtxDXxR3kanmF2)5GL@CouO4>mXDQD^ zWc!QnpjOiPG9@Q51Zk|Bs-O+DkZQ?#ESU$=eAP_(*Bg|jFk_6}+K)ch(r91vD_aIi z&5DF*paEw?lnOy)rjo!BT>ln~4pgxf%EErTZlEzR{Qj1k4jS_OPA`j2K^2re8%6e& z;yU{vodo1T-2P3{VKpbex7=Y>{(Enc!i@NnK}LnBZE8VshWnH9#QTYV)55!-h5T!C zwsdXdAA#`QWCSS(VXo8;V|JifHsUPrRD^);etZcy+CgFokXF1D?edyAOVGv+xj(6b z>d<>qocb#b5mg}W^(38mKEc9{$3ein0aT+NRNNZGX$e2!W%5r7W;KK(cKG9J@lz1C zh8%5{T5GSB06-b-5DhVOYuTS7d))~HQp^DOJykRrB2<1fE}LzsFo}{Y!#l=GqO}Vy z$V}T}6EVF?RWs>Sz$!H7)ILy3f5UtN^<#Q8Q#RXX6j~@RyH&6PwS0=zMHm;YuwRB~ zCRj%m9RZ$pN7_I0Q?Lkv8nV=$#7J;xiwNPZEt{&rt;C@ge0+yt@hiY4bjcOz5x^Xc zL{=6A@fq0R8_q2T>>Y>%(&6%gr*-cG5wTK1$VfDHMZmM^j*yrA3E#L=xeP8%-{Z7^e=WoZq`+nt+&y)$j|a-RTgXFG63>?Or$Bd@fC`5p{xpI$nKW7go867acD`JQJ3&XPy1P%t2-W+3ghMD>{!}m+;E6YMq}{A(fw&trWahXd)^S6ONL} z11DXtj`}hw`jgVV9BRJ5BbY}0_!T!4zKtG$a;8K`NQ!#zB)+0LC4)?wU>~)Zx%~Tx z@%_g~0MzMMMy3G#SJ)1rh30UpYhsR3H`g8rFP$C!{5nk77wzHyXaahn#k+(ZVAiKL zQ%Z)cM{K<2Lm|C8=r`aB*o36|dNOf;Ho%*C*+i^z>9fHQin6`~=#PG!LO}m)e5u)V zHvA+__RS)NDvKp+SSI<*@I3U<)fSmJ4+lX7Ip}QMVG)@+1C!mZA><0xx*p;AE(5pn zz^V)p8s8^~*S;>Glw2k4>%Rv3xrS@B%sqICh4SJS><@Lht337J|Fu#$)bGZ1M<9LR z)}_?EhrU?@Apmpf&&M&E2YttK%G~x z@<}LE$lr%0;j-x~MG+(&yz4qS7mk%`Ly(|>&cr;Ipvy=8_frgb3iZ;?()cjENGpfOp>`?SeAL3EnMtk(?wr`TWd3b~<#OTOwXVlZB z0PX`LF;5g`q|BH;fRw{4sS?MdRL)<;+T4j0x%E(ekpDDc1SK5ZoxRMndz_t*ANPF( z3vj6tNJUoazF+bg!M?pFnzr1+>c0<==FubOYtIfDsm8dHzn~N8_*lNQkHrTHEvbp5 z1tSXyQQ2^-lQ@D_JLUmG?3S_@dq~px6LO;$%^e|i7Q6u1dh@p0Yh#oP03X^H{& zzDa-AYLX|d*~We$XAuz!HkMSh^8JchK7F9pysb&4z$4*1g<77NcgrU!L|O7EE$jI2 zvtL6FN}gT!6rTW^TB?5xfz$ip$Q8BrB;OE(ak?T?S3A@aGd3PnTh4h?d7bPqd1 zs+Ifl5G00j*iFxo4BTASeILbEZH6L$0FXgbp*a2;6G* zgXX>V+)MySic0QdI%4s~Q>cjWP`iJx2S$uwTIm7-OV$&~=J2Rtjd zaMJr>llz)@s9m$VS{J${S;{pE8LnH2{X=M*4-Z;&ES_%6eOWyAK6iy?xXp0PyW^P{ zaN#s7@#8=kCNPNluBPKR&9hC5X{~X^!VEkBqUt^OGnQJ{Fify9mT)Vk8)pKvm8rne z59EmVTcyP#AuCf59sWDuw~`B0No)@N7RZ_Ig1uz*1q+43O%jhtOMm9oA*Gv8_Ob9* z2r8u!K_^;70uS0WrUK4cwQUtf)%*LWNiE+a5$Q;f#(c~dFM+AcoyFdxEJgU0*?K#N zfz{AYgM0C`5?+_Ayd(c`e&Udn{|jds%qQhVDK{Nu8tEooBrKeNAS`?LbM<(GbKtx$E(wb}B{I?7XsTqF1jlE2XCoyqK)dck{iP*I-C z{)UyeYQ+Wt*tBrH>jEkbwg1afGZ)uxc~05idu=C08n#D%$5vh7TVoP!4jE2^eE&WC zbvppAAmAG|@~amLMT?h|Mgn9s6a3iP1+z8#q8IZzH-uU+xd);1(oPvixE5KwJGW6h zE6RB6I5mn^y8V%i(Ygq(sgf*7r(H^4v6ml#h%Khoz*&t|Cc)D1Aji7RUHrRvf7nSV z?@b7jwwFYROF{%#e5}@s)M2U3+_camQ@Ank<|G!(4sU$cv$O|SIRD^nOzfnIGq9Ao z`L8!8;K4tGISDr{%4mExx{eaZGC38?@L!;`TO~Vzw&S#hik-c9pr4R^e)1IJwXQE( z!YSIL%A~?V+1K2KC|VT2Kla3XhglkKHOJS<6oOiYw>w{Y&jAr{8fIH8yk$xIV=*LDQX>X85qVC@r;j+e>SYX-wz#&4sN(^HyST-I)RJi12MJ8Q6=W2zYl zBJI3n)bg1qPvbSj-!D)Y9^{QPA?}1T(62khx$l8$2OKe!4?$;(q=9+o!%e0w0yqO> zk+Ib-1K8QL?g|f*Inzpc_jC(eySDQbx;4RI_lN`t^7wHD@di-d14?a*z0jfbK!{TRKd4=nLU<1XBvq`}NuI zKp8LDW)L8Q(IJ=1QlZl~%A6!+q~%JdH5g$2OUXEd53H3ylfO>@I-8^w{Y~*Bq)!CQ zi96@Xjy2sy%WSCx0`~3MZYtAX=-f{S#!?GryGB*M;lhAZDmoX#eR325zjTaKM&bqh z9=Y9z+MmN}n=LdmIT4!iOyp<7kUyRBGP`JjD)|c=r&Hnw?JB>(_aAS%)-)J z>`mnDaP3wJh+1!!zXSa$jNMhbI5KaeNjA5Ty008QE@B%AW4`alZ7l`ur%Tt>fvW!X z7Ywxz-F8?#+_XS^I?P0F|5j=~z*?${|KrNZr~J z)N=Q|y#JUVYj6G0)apJ@xmBfG1hYRfB<|yZ&eUJ)`G77NN{bA6ZkYS~pOno!MtnYx zKKmI0_sUUxzsB>LI+GJkt$^Japf;&*b?Aj2xla+|I`D;rwQURnIj$prTQo8_`L6Z`#hx$^bhNB?Dcsc$8&zAD;lU(1IK> zc00Ta`xOj`mgU~wd9ua4f_OC7Q-KHcxqJz;?bn5|Xx#V2+5-pTZe|DUM&~>JZ1Q=v;7GvGLa|e4A zMKu)1YxhN+q&4O>$Ls!&;VB}+X>&(@`td}t)20fZ?amV;orA zUA&(MvKRs7(xEvsKBz0ZzIygfy1UMyh(RYJh#fD-<)y_GEYx!TilIMAHkqUEQ+l+z z{^!E1iLRMdT1L@bitlwz<#cS^$8wai{~m(ObLuLztjxMOc6OJ$J|K(R-bA-*o|RRd>E39}-;NGqY+WNBPeMiFc>B^&*)z*!d-PK(40M zjRh4KyEWTZVyzJLZ&XWDcD`eBZf40;%#fmcBLD0m) z2Ui@lW-YZe8jn4rFrS$H3^q?`&7d1*P!Jy|Htce8>i2*IUT8Pr61MC9>rL@g=X%U$2UDg z_)A4F+K(D>`k)}t`H%c|i};Ff2IVk>`G7!^sHLZUVrcGaB(=Yb0N6Z!P z7%uJB zs6SyvZu_5G%y7jnUhO1o`n|pXj=_G}U<-YCRPj$9=N42zW}iJtrtLT0kHC0)p5Bzk zWZ>+sO;1_jB8n{ymA zoMRRJ0+cy!Wrk~*>;a!I{&Jmz1zbOPB6AVZ$GRe+cL=MAaPt1*t~y4b>+NFyM`vX#v-QI_^S#Y+&r+>q zYJ7j4GMn@@kqy>)CpvZRw#Bua4}@c`cD~k{{etMR_{-`%Ve_n4=lfk%jP)KSr@|Nw zT{zv=Q-?d*t;3EvbM<>0tO}RTCB!{%B2XB6NJ^>IM0&|_iePK)oomY$BKq; z*G2WpBc~YZRtN|60m~hFx|KEcR@Tbwpa=W*Jm8_PUv5K}@ss(m#AzFw2IJ_+GuVUI zB$H#gr%24oAMN>%hCPo*_bZq!YO77@C#TDY(#kZAVGrRK()I{?`ylhSP`Jw>d;yMpO^f#67$8-;^jB4A#Y)yO*CH%z-ZJgvq z8Df&SCP#-e{6~pPf%`9GV>oSQ3_p|)KEY`y8drTRs;llc32Be{qa@m^eaGzIvofP7 zm=iR&pB;gZu$#n^GfCB8kZW`L?@Bh5TW?CgpamthtmQA$Zlsy)YVzqQWZ_%HL_A`CnYP%KS zZ_PqHhW5C>K1cAADcLAcirxZmzVBeQXIxm%5aP8ft55IH&brH=OH}l|9&~QU9AF!k znGdhQ$#3W1&V5X+%dq*c<2R>*xARvJEi7)y{nz-vzx-cAcoFjd?ub1@JODZT+v zz!RG3*5Cin;SH7+H1XLDi|DEw|LpcWcVUpzQ>1V4e+}aSVkqtZ$HNE~4ie>0s$08z zE|Pt;mb3Nxr{c2Ck6}Uvj-;O=*WP|uKND9Po{yne2OGTJ=l4a-N z($Qjlb+sRS$J~+CuX-84UtwLwarLtZ9F8=gk5(1jc);o zfxGD^zw~U3?Jrxa|Dh~dCx{dhLhlwHucmtWGu?O}%9_9y+CI24I|f~qov0_=;1hl5 zLj@e3(ezMKf$-H;RZvN+o36nJy(l@WH335s#f0udYwa{1kb<+|2bFa--=0~gSX>72 z+)Vm3OW#?gyoyZj!gsAM)ptX4V1-50M@8@Grr?*r{rGONT)Twke(+DQV*Xah5Ad!< zQ|EF_^_Ml?4e$<<4l)?BAM^R*m@N!98=UXZh z3qsMvkGM(i%BIR@h;qF=w`Jtxn9l2_bw07LtYZoUq+NbSry=ytFlTF@}Tmb6UoBDN-V$_s6@z9KpyVdfVdMg-;Gj z8+nM2pLRo$m&&SRd*@oBMxqhb#&kZ9=HSg^5Vd))aY!n1=579HMD>g`!(>^c@T8~3 zFH``W)qk^HC1vEax&5yLgATR2^nE)i;E#`E6n6MXaCPj-3B7ILhld~!t%9xOJ^CEqf+x zsjSJkyPnE=d}u`^XH<}#HxW+adK1!fpk>^zUOTLsL{gYjeNuN=HRjPMl`XYDH0*D7 zzsrc^yr_Yh(Y#E!F;IOw1qpR4;;djbq_Pd?j9LkQzh(>+xS!V{R`uDBJ&B%|_VZbw zL4wirwGDHG2;Ph0n9`U=cP#DU*eXzpn{&T>VMS+Ko*`_TGwMEf&(b>3^!b9i#;yK0 zPJzDtLSe@=Y8w>XD$G?Qy=&N;%GPvfP;oEKF+U0CD<2`@5Zh%A{-!TL#VRIL3xHek zKYjg2B#iN>_e6ceoZb$9WjN!+hV}BVfkx?JL^?{_9EfdWBC$9@X2ydQ3C_!JB_{sGPhb8jsTG;X`N2D)4{7UBfHSh3Q7dYiv}1jx|G`${Nw z?D`2Ncq{8y-DWqN{P^xm&B zIG{_v6w-adS&fAUrd>r9su{;HN>U8^N1OlE)@D~NuV(E~}y0}<~Q-WCrpe0o!p(qvE z2>Hx(1V|JK6sfIb%V{Y;`jdM$OgCk1$y$yBaFGd;Psb4I<)tAbZk2+HnfUt5&xUOT zvkitNU*s-)NZ5|r8V?41;)3)G5p#7b(w=3TVJCnE-#>GDZlQap1I!&k6KdoVW_miW zPl{ln6~UMyHjmBV>jF{yK7)j0MgJgATu6ks_-aM$B~Jc_Rx47;*UQg`J?)RV?>g%Z zx$Pf*06o+B_E&+CxnY>_p^~{n!WF$q8-7Q1$N=Lj=Z`+S4W}Si z%;mPxx@!^%q`>3vKeA(KZ0C3bJQAjr17Egg0KVk&!mVu7!}rM1BS7ihxWDwfoW}rj z&x;Kc$vGF5#^C@=Bm1j^JXFM`g_mAf03J}NMndvAgT6)%R0}I)=p(#^P4MkRe>{}O z`C=1}agL@{VSi<6^hl!wkk(^#1qWZ_xfWN$u7SaG49O-~nBx<<&*>wOL<<@TeIata zMcHhEpL((fOIMK6MV@rrM zP^-z3g8wc)iB<012Iu%^6zvIWe*gN3KqRyw{b*Y6@rzbg=NHjz*!MK=(?2B^&)C2B zRBXJ6UoOY-VH9HElbs6$H z`Jy+%%~J)%jUEAgU6w7|2ynkzb?%#Fq?t3@^B<3fHY^u&Z> zs!x*7Svnl8zO(pd8rtF0kj#0~PH6ZkG#}iW+60+l6|=D-VwADm#{epcobI409zJAY zpVa$|v(GcZ{DCJENI-|xWv2;m3Xfz>`A3cKksxM0`n|-rUIfn0Q9Gu<4v>S=hB6zRB~W;ULqF{#Kw;IF#Ia^M6_2G#Ktw#+=-d@vb?rgRf?F~+6`VveV<@(DMZ zi$~fz2XDwED>TbSB`d}tgD}FaA7)&)lx7PZQdQTL1XOPu6O&lL@DGqBgD*glJ?c4? zxOSH!;1mz+0+I6~Wdu(GE0w7wkS(Z)!o17p5h^QMPv-Jj6i&J^py5fGZ2 zqJ6PZY8T|4aeNoTIi7V>nneW^I$>hEeA9%DmnKYezjP7Iey^>&-gUHeTO0jzCH}Pz zMz=fUpd(fH4(+gZ$y8iKk0Y{FdIRB|GbVlmf}3(IF;fD4xags0N+`fr{ZJ?B~WK^jXuv;OrvSmEC(7`Q|(f6D#L$Ea7^j z7+2*+(z93qxANy8IRoST&{|IhLqORDg?2$mZrgqA5sB_~kJ`VvHGQZeNiwY|ma+pk z9jiyFkDc<)Vtw7GGC8k$htg`ia-hz$hY7fKvR_LT&vW#u!Y&toCOyp$4Bb*%B=7SE z8N;oD<{oRfdFTAcTTFT%*d!Kk1|8_Ltp1~taMO2xjS)q@eqf!eEP*`{U_l*7ULcl{ z!j|F>6`!szoiXhGeso6wN;r>gcmDlsfUurhBkoAi8xrjd@yclt2cKv4rkaK~~WdC^d0c}Hmh~lESYq{=Lm&XQZ53Jpui(4T^ ztCeyZI#4+uW`_+A{bHqHLfqSXBQ^?Q=oSUB^(}UXOi`2cTZ2*_NuND|w6fyF&bfHq zO5v0wEZ=WT;D@K`O%8D(^sQULI$lZ5|CjKF>RO2NKFbiKK(uvJC1g{}1Kn+4joi~3Fl zW3Qe)&AA3$GV}jDcrgK7XKwFpoP-KycRzcf84OsjAD9K^x7|xF#-8&2qnUyTxSk*Q zIbx*M`Ho-bEP#(mc03Jp7G4s0Pu!t(E=RgNpVR3J(YRUtqSL58R-6q-EoawzJ>08( zc)dv+Qx}3IkDfTNzb^K#om5Zo{RM>vQ10X68#c2-1$Ul)&h)RQhYm>$C53F80|9(% zq1ah)u8#zEr;YGa%*<<=q!*po|Bl<{WZ;5~L^$5p#6>sZLMO8qV`Dm*oIOWtheth^ zWB#|ne7mkL+0K5VZdU(z{|9OB9o1Cyb&HBLrHb^fBGN%h1Zkn8ARPjP-UNbx(tCu6 zpi-2o2!u``G=b2IARUpGfYOU}>AfU*=l9)r$9Ut7?~eQ4xc{7-l2v2_Zo~*VDUq>{+N+ZF!O@3*7ho>~CIBDh z!cB>5Pi?1GR<5>s=fS?xxx|IR&DR9?Rd$=ln>O(j zfeWxZt0Y&AC(u_AXwN~Q;(G$#^d5WR_t2M)7t~D88l0FoGrCq53cVabMnO=zhBtRBSQ~%L=Ng{?zMM^ z)|5hA$Y9EDLzOvf{`Iy$$k!)^OBvNOCNjg~txIeL8bi`x$05C7LnV zI7;FqGR!@$P>U$tXjDFhdU48`Wh*(*-a4Yq#r43dyJNE`??1NzOeCQ)#FbCVQJSMQ zxsoH!`ExR*JMZ~H6oE9qBHNy^XLm_uteC$=*9933hHo$D)LdJSRq zrKBacsVW}7Kn+l_jjLulh(LI6zn9qT1(yI@a6Q57R7(B@ho{iAw!Kaaoa8pm{$#yieyU^&Q)=pRG?j4-W>pUIEuEW(oU8D1D z>vE{u{CkEkrDnITA_FTC8sELje?OZmrx$wUyLW8T#^^!OhyBAJ!CL`&=VklbN>s;< zT=yW$@%_SDytm)))AopzN;V0SXAdE6Z^h1i3qqph>cY;C2H)$Ql`bg+Q?3bh&F4Xf zTmGWtDaqi*jUO8s?j3EW3Q+D&JmQG+3O{@-`fhQdst$bNw6w5*e!xa=`LJWlLTOKJ z&dllkz5<=W(BMk$PvnZ2S-xYRW4{tLuO$N=2h|_EyXTO^;XBYZtX<-@Xb(P0OGwTv1wq>tcMyameu1^d$-Q^>%jWU@G z6+jYM&;PE%jwvrpZxFl&ItGmIyvdC3vC6RcXYy4kxxZsm(73g&9C49~HWL@$N=-4X zG&g%zIzr`pAYW`&sq!Aur{=+Ot^DQ6>tcieF{24;Rv$uA0uf`LW{pOy`4Otd$|(Y* zpSP|G&_P&Ft)7l&=z9KTq8wh{igDIC?i4HKE;|d}6~hf)lJVJT)Faeq_hS6F`imL3 zxYR(m)_khuiB&hXD#*<+rQYq@dLU-{MSo-9cN!|Z?p&uXu?Ez*r-obLP`DIXr~kZ!)IrQmTtbyJx{?FJo$y7t!l z>*eg6{gou;f1E1Y*OB?w%1m+ifK)j zhV!Agap7+WM)#BeIg|0bB8#lGYI;uGtG8 zg3J!B-&Q_pe}d__HyU?OPh_T>e5^Jq&A^M&cX#%pZcFcrS%3b$y-)P)7_7=n;>)ITd;zp82x%3*`9 zrR%kwU1Th=109P)ego42d-y7w&D53o`a<-X!h~e7F~L1TT^e zS!R-05LQZfk2tpwiKlymz|~lCDfTD5c2C_s^PHiBNKmnoLv#1<*sA6*r!J5}x^T_ekf(>IJNAUv2my2J5%5y>tNt9iC3{TXse|p;S9x%KIc@Tf0aYQw9}og7 z6SE;!@_%OW$N=|*7hGHmuor^RMcKUKyP>vSwO+4$1`JSc(fe2XpGLsLty=~RBVfd@ z$Uq1DGVvK2rXf8lF(2bTYt6K{L!?O>InZ+&C0Y*1Q_(@t7kRRk#vNu^00$u_lz5I^>Fz6}u;-eb>hpY(8FA#U$g|H{}gq!Xa`Bw8U zFcjNCnLb(eJV+m9JY+`KwoWPc2m?`2^Uh zYe1*z*xRysu~T%T$viWZOM9}*dXO0$$29^DFA_K$mMRf0^0&OBB^rfL&m8)SU~XS- zb!OiYyMsqIYvsid*ja~nO{|X$V&3MLW4SKY4sYLjn8gu(POtvhLa!jro+{&6`P830 zE=3Pc_SuUu$MJVw;R9{dmfT#i@?JyCkIL4g0#O2UjRs$F6x&v|Iz|4Jq0hD``a_Jy zn@*P+2D)l>-hKTXI|<-IhX(PN4<&?RKK}9BQsOG>8==eAKv#{whM@Y2qoU)?x}d|h3~N!pQF`z zHJWj*`(8-EZHQiT3T=8SV9%7p%|(-u5U(|wVfL=5@e+(Ve{6iMwsTEWH(B4G_YKFh z4P<>tHQ`%uU=@|p1smO}{yNgyyvx;f!Xpn(u}==GVYpbk)lPrgI-zpq)QBz{xp!#~ zUHMwtaYak%2z9v+By~SiQqJ;C){0l=HOKjeX8Zv4%0BmbynqLh3^HCFV*}Q9;#9G^mpr@enB;>uCS5$1RVz^xd54qZX+ zy1v@GuU_0lT)a@YF)oa5ypi#uP;a24r71o@gz}JqKrbH@0z?_RdwKA#FnTGmj!_F?(h(%0NFH)m*}tw#fOhU)KLFM-@dTF1i5})kWHz;&y#yZOdwVjg&aNvj~Iti7z{7mHw6a{|7 z-W;vD;r26~?JqF|`ca{OzA$JvXgAAB!QvI>I~B)8asMsx)|UiKldy#XfcxaB6cT*) z;sOqh0NE?=Oyb9ddvp-zQM&9M^vC`VMaRGvxzAwECsQ51_vpeFAP21G!^GD*Mr(Fn zDmp-tP0!bIg6X2TZn`7v-scddat}j^J?b20=;TZMXQ}eB-rt|VOA9zB521c!054g& zyRQMd_8Ls!LI7C<4B**~0n02NXWcbRE24C(+Xu}Ao@eNVH)>ozWjW5Hd@~?7$6h>t za=H8?W`sQKN3DaDqALR7TAjz2YIT!F_>cuX_8{=;>AhD~n*n&rh!E^L{PE(2iB;Rk zxNy8cse7vBXNbcWKq$Q=?Ozl3*BiC1!c`V#Ms!VcsVFAA`rc7mOhfUJzwWevp5<+)c_Rc7H?Nl80e(C6B$fHA!csnZR zd-E{-NjDx;ejew>IRWOkiQO=%e6|#Q`18>#k&Tp?hH4dJ@yI!tmLrVQi~n=rd*PXI z`uUq?2LdcRqhaDh-JEq6BU3u>gmBwptbtcUd9wL(uO9CDrU>30udSFae*-q%KrZ3m z8MY4A4oqSkXPQ4f^~k6#`Y_?AMvQA1RVl7-vA*g63ERK+< zU$tcgJ{T?b&l&-{2bRqp4cieP_*c^zCp$T-L8JftG%C3s?hwFq5~Z@$`im7 z=b{p8m+v;@(881hZxdwxCr^C8Nd0%V-1+Usg1dYWA%?E?Al#8|6oJ(==^;&&2l z&(JUWf}xsO@KzZu3OmM^gyZXvVm%oL+l7Ac5C8jokoir)B#OGoUj)SSgl-SaDxlLf zgz)mzXipsG4Q_T6d4>*BB6ik#x!F9nWFf5i94w806l)zJF4}>b4<)*Z8g>jKhI`Ai}F7H_Hm z7nfqpKayXijMo2AszxH#wxuzB2?PWGcNBm))Ws)qRfkyY`r%wo+vH|Iqo6Q{muDHA z?jdS8FOVQj` z$)4>2%snrSUR$gthl1f@&kR_A%^`PkcqfBq?&(*44?0nv=ly0=Z$dxwZ&j6bUsXYR zK1;vK=q@aJ;2=g-_v)q!)Je~@Q_BO^ z+sN*{z(h;z3{~TL8b3xu219quL9QqEVKI7;(znJn;}a4#g{v|SEY4`NXs{t7)uvhq z#H%7^a$hh1eBhJ2u-*?33l>8J!k6P?xO2udTSw^+HSqm6g~SoVEW%nWhucFQeT^F* z?wF11mOM6zPH;qfVsh#z*^mfHHMwJ&j! z{k;`Gll()mBCx9EX;yNO#jnCFDqUXtLoS-G?KN7^%TBh%uTX2Px+uv`=ghkzPH*I! zoQK-5aV5KdU;^sVF8o{G>$IRgHkzLP|?gw);&HEHKfOxrN)) zFHqPiXXhAwus3Gdu*c9p%ElpAWJ`O($RRh?%;HV$)VN~{C!ZTus=ZyiWaAp_kBd!E zL3)+p(>k-B&z0X~l=3^(T&7H-qZ1sqE8}yOOZ5WYMDHMb5+!LUCfqyfStv*aY#zr2 zv{30zZi$hCJ{tSZrfe0{P(@0bu6aT>$*?|?vZ#ltk|{f(ngq$GaSU{a^ZC&2t@B+Y z)XJK*jN`+-OA6FO2@=rAcExpJQM^G!yi`w$v}o5V~4G{9oIoS0lwOkZwh(Mdz6@y_PA~;Rg&r1&)wNp_fJ(LcLcV0S>W5U zUUzN|R+m&6@*zb;^TKijU<-HFkz3rj87<1BEj{jxa4$AKP@c76* zSKJjrsp|JEh_R0DHV&M$gr9Rq8_l(iO|9?h`r`ljg=J}W-gE3*44a><6cxFma~6VP z1fQb0^B)+)^uGnK^Tu;DC{i)O?o5hatTyWGgYvt6l=}$ANqS)`;rlKM>J1q-o-I4_}=SX+MBMZ39(Lb%Z0}LB)%s=qk6bkk{)M_IrA|r z*1AEx5^~#NgOvXGR70#kDaDWV7eoFB$gYt^bOn9W@0qzPfydyZ8@iRh$JQIn=Bffk z5K9o60Se50-9?Tc=F$6{EZS)9-QQu|dx2eg87srBw{NHYYA?-6d&|Y&6xZP!9h{dY zEV?B4&!e=vuc}_&>=xFzUtj_(#l^r}o9%TIR;3$VXhr=dmhkPqhto?B{rm?l?9Bh} z_ZTjxegghz{S~$6vR{N9HfAZmvmL@5mPAB*KBvN*uz$Cv_{8b%Q+8G>cfV+JdDvOa zvL&{>cVJ6-NOuy-ng1Z24%P8Jx2@lu`61h9f$wO-q8)Sj$&@_x4A6^C=$?a0vDOwq z!RcMyGeI`>jl{rwiYc~Dg3|6qj9vM0qE`l`j>SaDTckw^ZRfq7w53oUREKnB!wnCcA2GTUH!#rN$Tx2jl-Ja%HXaOOCLp>FzBK^u4X_UNubv9gq zqr=+k`EoJIYFMlfQ}}m|>u&Hygot6re%3#Pjpea1J~sqmZ=CR+B~Tdb!=e1UHN9lw1c@KeWRWX&-DfzA2lG*GTO9G$8`EZ$#h;Vh{-3T~)5LfxF_ zGSr)5i{$F1F)lwj;Vh_6_9gayqn3-J(p@V>6v}An*?Nc-h0n!xlzgXI=aqDfM%U2k z%673OKffc2UrXoQ3v5FW=%d13bI7o*s03=}QZ%~B!X&uU^i79miNSUk8;Pte@{^u4 zbDgW<*f*7xQd;(ra%59lF@?#=%(YT)Lw0HxOW=MM7vNEgT{mq*WH^RAgUHkL2bfLdrOZzXt>ZDeP?o<6KB>Lh6vU#1{FsGkCFZ7iV8B5}Fp<;*6X;LkwT;eQE zQF3upwzl10WiG&h>DTT%5Ox6Xc@_-)4>* z9L0`DYF4V>sbG8Q>FN%g<-r}EDxDYe9>xJ~cwca`&1`pId)SpAUE1q}V zI@sNl_7rC)2Gic)?^zgw4Kpj0{m!eCiJKQ-cF2Wnd~tmfy$K2BR1q%4{SvNa2;g8& z6PNx>;jhr(s8RA6y~LFv*z;Kceku4m@45qx?PF=!OLAaJ4Zh!kumUu$P1S9c276~i z|E*+0S%6F2NaTIdEUGs-L}?k`=$D?Qood5gE4VhQp3f%e-)_utl3P8_H+2Pgm0+Ef z0jF>}`lW6lbdA>ui=?Xr2Fp^)A8O~qf5phLNiK+H+??q%dY&=jyB}Kq&vGfs;4==n zR)wdPT6%LdMYq_T=$g~`MYGy#q_h@_dCgbZdp@5ZW}ph+LI2MdD{wN5GoI0q7sa2R zK^>DE;Lq`i9i&20zPvN^28@U9T-0SRchKl-Yf+qsW?#ghK{;vReiG=#-^GW&$w7PM zunjT9-DVnL>uOt9`gVSHnsy5{C@-qd8zqH;9xjLXmQ5t7GXF>-W2f0=7vva)?dDxZ z>OGeYreyyMQKjW6f}56{4{D1(6tW>)yi1*q>BRNhkg&8VcClt6F4~PdSV*d~P{j}L z_Dc-f-a}ScLoL|{+g?;QSo9Vd(6Z?De~9Ez%tNny;%Fh)sv!s!cK#Hp_fVG&a_yVl zhH*N7)cwsZg$h}H1AT-;t(coayvdStzfrX zhanD(6IX$%9oVazJe^y5cC4MVCWB8K;1K#9Q=@W`` z^i>(q`S|)jH5@d5FzKd{as#D0!P?QB>z(+`jUOR7=q#VDf#lf zlddAoobFG{7H{(}dpDJA_P=f2TWS2gfRt1$tEAdAv@j=Z3FmCAi`#=zTTPxco<(pZ}!Z)BVar8)p#wJht@x+ z#xAtm?tc3hNpP2w+jL<6n_`up7D9YW2(H=s74!V#VL+atZusxV+!wBz2Gi0^QBQKfC)clYLLBx=2ughtSJSKjpeB@qBLNm>_WiIEtOJQF7amD zv4C}J%yji8GDFqDQd-VKUCzT?7m;Zp@$%T#FosKw;XCdU>zx!x$RYf>p{2ROE51MM zlC0XzD#(dHv`1(8#5_f}{^V>yerEqGkJvQPP34LH)WQBDf9WMgbeai9S&=rb`*GQz z_UQu5eN8F9P6PqVDDc<3`hAvvAidfOkgz1M4WQm0=f=h4VeXBn%TcAr+K1}-yr?za zxpM?pPI^En?dI-3xWC0)J0%j_6mr-eh31FqUSpcbBmVG@1^N{&(VM;IZoI;%75)6t zz{>PhFipkW)|TSOjz70;XaV7wjo@Pw3bT}tZ=%(wQ^;ZVGGBxe2rn^lm$zGzy)c|I z;D=vn^ix$qr?l#Zh?ZZ}-p_N=yw#JhVYt2`GaFQ20>! z-qm?x-gTAcJ+h13Z`XekM9OYTxX%u-AFGXqgPzJvY6``ZJkC#F-%Fx7XrTh#IF4WI zWwUM#4dZ9f6}do)@!I7UVi{9~_w<{}V&a=@)X;dDjI-WJ+EyXGtATydTh#XcPl(!O zBEv7(iE;ko^ByB$^T~dPA$ke8k&Kk4RG*kw8jB{eS7!zZvh}1Ry$LoK+`R^A}z;6Hjmj7_ONPljNn2jXJLy^P)e`e3y7na1O!pU0?{n{dbefD~)l%Yt zeTn3#^`B{Cs_)~d_YLRnUCvJ%0bdP*zWiCIC5yFMS|)^uVsGP-uDZ#GopX=VzElpw z%(~7et^#?eF)4&$rUocPut{O(@w+^4XiBE6{O;LTRN)c4Yf5pkk{AE{O_J{|%d}%L ztzX#BUG0g`CTlUqGaK&ZY$k<&oYv_zg!53;O6rF)IE$sbb>W2A`kvIM`{~e+PG-bI zI6Wrw)|7(FA9B7dsOYw_P+N$^V;U=qn|NgVJE-_}sHZx|X#8R52hHpS?g5|6A~7CR zp+tzn7>uv?BEXaR~9>onVFNgO&Ol zZE(1l8%-Ea`qhHSu@bv*{jN!fZ0Na#AOBn*{Q{7^QJNBCctvAI?oqYZ>)szH=!@5ej7U5su5EZSj>xp{|4ir~d$194IOAc;*_tKQ?E0Jh#uMS`7Z3CtG0CP0 z$HSTFIx%<~?Duc|Iu7CD0@SeY?LcVc4hpTlbshHb?u}W;911nIF0qX!5%c>ylI8IA z$wcbZ{lqo?T2CKboB1wnqY%wTo^l?1=|#$&@{=x;c9LjEyi=jqQ8R{s!+W5mW^^%k$9IsB%szb8St zh_Q$_^m343{^HABuJPyL=E4=`odRm7ym^;J{DFr)zoECZI|uhGic?`Il@IN-GqWZV zHmk$vk#HNI{C)nh+O&wwb z0}U(8zHmm==QT&miIA>a4#f(3d7tNetwoiYPF$uh@`p1Z`Q4X!06U+Lfwrx(ersbk z9Gs8pW0<|{8;Dqg_qNvUE9FpVz!w0DBKYxlys{2)?^5QD!WMEnf`LhsAa3vNjzm;N&ZmqRyQl+LCC6eDTZ)5%E;-Vb!ogtMEn z3_86JF8_0s?`!;pdW*#^#Yx8v3IG$B!aC@y9{60 z6cf&3KAMqTV~Qr^(oMG?YtWwc1|b256tl_3i(Y~cBOZC@^b|^u{hp=|I27SmCVeK~ z9{>1zwZ(9p7?D4l2S+)GPk^QEa($vWM89a7WJ{w|y5|09l)-a8aX$sSXYZA(DEg6t(SjMR;6`}8!YS(-tN-+kjSga6gX03 z5GHis#syc+p@<<1;zOS0`cgVfDeFQ0n12Z_&JBIhmdALyf= z1TAckmNz|5&X>HnIiwGUF?+$W!lLsv&QMs`Q%%M17#K3IbUBy1V3!A=ko7^|CFWw zoar*CG{)qGtNn}OIMY9RC;Z!zEKiFz+joN(ktCj;Db@+@r* zu|BcckE}NasslzeaKw$(<`*M{c>xn(SLxiefD7^q?f1^M?Hi4ZD_0(BkH3D<l4=BR49IA5d1X;Ej#~voj4<+O2@VPh7{zs zG?%o7xej|;`_?XwFnqFKDCA6HI5_7YMD$%JBxb@@Z+BnE`0e&14oN^&W2G@jxC;+x zR3QyWFuUp8y^WWyv(1_o?Moqa$fV~h8|o6yi~t7lidBc`s?*i1ac-H!LHM3(^oz{cuSrG4{44Z4_zl-*&F7pq2ARrj)B#`-4wzz_R0#JG(xB_rk9zI zVUBre|HWGU?-u9nl$d`F_-UOVuc6uEo7T-ql~a@v-8`f~o}nf-dL*lBob$sU*#p|M z->AZr48}5<_VL^(x>x=KB898SaUOQeUt{SyrS_uZ$n)B=OTwSF=0DH1hkKJcf8rO2FhDXboxTfM(ypE2rN?kV z8$xTosL*CipAMxS{>atr!hP>MZ9o;Gu_+J*%P_%!GldiJBmeTx%ZcxxbQ|!1OLLMG zHaVG%7MGprup00qm{5g&DYne`bM_@k;i?pn-g^D~x~_Xp8gfQOtimSiXs8mp%}Vi+ zZseI`uQ;MPT81i( z3Ri7ebOGM2GLOX^wQ>v78tm)ua}D!iB)y&|hk|4Sm^<+$=o;*C;kEPy6|wpCs{T8L zaNCO+0N88@#WLljaUNA*x8$s|+P)=uVcPBa%Hc&9CM3~NakofPwrJ=ksh@%&VlL)P zQ-!vIOHdEs{!7*Mb0{QeV~A#V3=z`MLI5-RH;X@xgto~K-`;UIDr-` zwS2Xe=y(!e24C+fWP~$feyY$uZ-0*Ea9)J_wF!N2BDxZ`P=yUp3Z|8d_s5Ce-Fbh* z65=F~`}wizvIH>{|F3p&%aY0SEYB>-cRq)8-r6aC4n#OJ#%}}p+cqX+G(l*LZt4No5LJAdM?Y+?wJtRTFP&bgcHV4!0v|A@)1FV%>#-La8 zZ=etO!LQzuJLW&d+0WW6X0;dv1D0a{+TTJ;2>=;YeAq+R%%}rr#ry^jRAx-4+0gvF?SyciK-CQgb+7U=jP_Yn#Yo@PL&*ehz|Ngi(u{r8R@#nncJtdDbyMS48Aq3VG=E^E z1voFvt8X#obsBxC9(OJ5)kF$geSr#3grmwWH z{?yBnd*F7>@^Wj~TxpIzF@~8LYY+C`(o~;jBMlRp820k6&z@g{2utvr!Fu>_#|uS5 zE@>6x9Ana7f`npqhjy7aZepb_ejDBs2jLyHUv4;*aW84VE2I$Mkm0cMBuYcZ%1_!2 zc@H6>79aJ*L5yaj>z1(pSb9jkZV!c)R%+1-vPSEo@%drhiZ)rMC4c5)w)(~Fl<`ze zv~)RX!Cu*^-RVYw2svg@H8k5?p?%LAe||Dek2 zdH|-3>(56!wTBb5hq}LbtASu$qQxb#92koJUZcmpCZ4C}3)a6?s6hTv?nq29XZGWcw~ojRw7FRh~oEM8`u(WVXF_($u0F$Km-tBS?%{vBbp=rsWE1RPQVnNj*h9jXggf3u1L z8CVsio^K)isQfMv!+3buNII0*mVi*k&rI^^90!hNWoV2V47VoyIy#*UUw-pj_|>I2oAW0U=L>P8dR9^Rl8x z?NvTSa!~IE4c4B!GG}%KJU&8^NsnhN4&)LCQQ9cfN3f+kw}&wq7R?^{iT7v{xxn)G zO#CY27G=QD?flL4D8*};DVePx;37p~Q9Wkah^N**VA>@(vlt)1~d zFlP-qO5Qa90!jLNTW4ro(6JHQ;`LMzXv(Liht^tpfbl;SKYY$Bnv+Oj$!z-^w^+onJF^+Fn)i)(|Y$v}hrRO$3`3vGY~LZ8tUi4 zt$V45%xS=;nv02U7~*S#ZZJ#}!FMdM0CnZC&mvn!h{uRa# zK&`z_ShDgHMV)Cy3z+hX7HwVJ;R{U7Z4%f^3nZJX5J-B1SKB`&13!AxB>%xhW>A%# zUJk|Dxh9B}+YYRFOu<@IisKnL29g4w(j%GDyjuZdMcv_&RvBk@ogu@DTGNtX7eQ|i zL6C{}2f`;vXwbb7BEZVT2pS^3xnE_1n;_6_|76;!6MuxnW~IT?>M`L7Fz3!}YovF> ze#BHO*7WP#>(&iq93->{b<+;Q4syWf2fWu-r+uGu0x&t=%%!+&4!y36wX@aA53jbn zFM4IGTufWmP&Zd5F0^-_xkPEWFdlsdpyA05ADy z1VF37-R(xtv9~YQ3_IeC7%2@A!BHgFKx{JJNL6ZuYak1Ier*EPC^P*xm7m(uc_Q5B z;D4=wDyC}_C#@$m4+W`sLSyc8NIr7x9>|o5f1vZv=5}cBoWFj8J1p;A&u!4%9t+1^ zTL0S~apVdCakR|jd zdk-dsQK~Ditdh_0eoFUH9o#B3Fzg!-W5c;7!ri99QuioEQ0y(f#%F^=j+iyfNbQ2R z3$`GMD|2JwEgMKk3-Xg>6QY2u3Huw1i@*cI=d;NVI9pe(7b#i;Fz;_SUKfplZEHg; zSWq1k#?~4nAi2kZZ?;f3zYNqSWDZ>b69tYmAP5}`iQ_OkL3|{yJ!bbOJIsDrfja-e zVLkr`7dx{22a}>qRA?o_^=o>CK#g%m#W%uTZ}b7Y#o|*kElM>%k+AApSQ?jcRlUo( zj_jcJ94A%MB>E9`+l0cgUC%}vwwPdlKA;#p?&qZC9)^EVUM9X6wyEv^`F@Eg`rW3b?=I$y@DZGp|rX__p+f zNA$v#aUPAl_>_7)oQfaEpRV&7iKU-2^9Q8y9( zP+3nPN7Exev+2(oa?`3x_JY2$#CCb4-Fi4>l$_bIOo z?zR5cHQ%$M6x40>RQCMj7`UDry(lpejR|muIpMwMAp=q=x*G;x$&xtx>Z5nuHf9)# z$$fX-aX@AQbvzZ3V3Zu(m_qcO)|K5+~@pb{PFDr9Ne{ac8{#(d<9wfcnb-%@Vz{gAGjB4VQqYkPUDX1Y3wLXUU;QTU1?4cI^> zx9>gu!Jc0?nOIqR`!J=FKap&f*iZiKX;}c{3L{aVed0D=SyfEw3=FzB!MrWHvxe$N zBF@Rt$I6%p(@0LyTiOo<9O_r6?N;ap^jg0Xr*4SrBGYs&y}bqX!>uaa(6rsm1!X6l z_-+_+NZO&n2^%nO^YsUezUFI*{m{3s=HuD@nm;!dD+mUO>b^rMuO$q$^m>@roIEXA zAkCnwgNos>sGh+4O(&C?u=Sl;#kSg^qWv(H#h3c83FpZ~d{7LoO}!NRd%S9j_p8u` zLl5fRmGL@K7VLm0y_sX&dT99F>|R{POZK73C}o&2PVH-Z>4TCv3qTtm!YaV|KG24^ zXP}_%IXsa0dJLS}@JQ){QF7{st#u?= zY)4z~OnMO2Zz|KU_r;QUe^$A7QcwQVIxjcorx;#75XC>p0)pj@Dq5ScVZ}P9*5Vt}C zP$XY~nv8R?Dng^5^BUPYFID{e^R+JPXj_S=Cxa?tt_uw;c*gKrkx`M)^M-f{fo?7m zV(}xpag2=-={mAT+OTztkf?>1+TuO#Wg@!14tQAUi@i^kUzeRW<{U#bDFpGDNx_nO zl`Q~6mP)}H$K7Y)HrZsgwYbt>!Q8vWz zB@ia1H6TfCILPEuBX^vrXNX9cleBIfL_Mtxr9N>pH34K_(~0gCRhKRt!-htD?T>Ox zAL`qJ6(ua$z#;`b+@fw4RNj0?t&Kz(|ftZNeYp2Kpo zu`L1mE9DF0JC6PLuQcLi#QKRF-^*Z1;m6*u`mb4zy!JNI@^q?oC)_8{Zz09YsL4Tc zZKdq7jeom>;2jSOUGCv4-$r>J#OjKs8*;P4dfHYc*i9_HuJS`#(ugbuGPs$2r*b_#b0b^l}DFU(1Pn z`42S4xsCXIJrZALYpJTxrhPm6c@OsH8Tlx&smSQ~=Y@9osXXkvfLwaP2)Gbg)VRre zP;|$J7)1G9js{3Ur5-*qMY+gMfYZ{sK}In~Ryk=RXI7$`>u~%rDG%ruAw4G^hD{S> zE@rqwQt)#Zf2x3sIy?O##U~4ZKJ7YZJ$R46;=}8(KP?JI{5RNh z(pa-%7;=fn6<^98@4;6WWsCu;>A1I6ofsF|)y{Jv;LVVx1j9PZ4f23sooi&O#1x%W z4nA#yfflw;<%`eYCt`A6B`N;!Z|867kq1uu`gqq;QDvX(eHUU1o1*d1_dE|5R)PuB-EE{Bn(?Oeq<^TP38RT z_|HdJRqO7JYERy;X$t&$p9kM5KUap$|2-@TvdKw{4X+Zys>CuqhH(nt)8xju0|gM; zo?;Un`Hk3=)xMp6<&HLzswo>5xBrW|_l#<)i^2v41Ox;W5CjCIgEXm1ji5A%^j=k( zbfroSRisPrgih$vi%J9p1f+%{C>?~*OK2g@H?wBe%v!V7{F*;!CAs9>dk;7J z?EUO#Kl>P>)8$n#c3N@DJM9Dy$d5OyDE%Wb&2(&7VrbvuQs5sK2&MXln{jlgdUPA1 zOf;2RH8r)PnGP zQevi7XEN!|x{C*@R8pEd{FU0Y2UR7%WA#x;;|QiAvh`PZ3PjmHnrr%)A7)I!T$vFf zrY-%+VwbcsLlm@aH8vzIQ&vfNrRLW2G`4J&+96oI8r_H%6X~kIcM#p2l5;OxI4s^2 zscHC_&k?Dk*$OktyyKph$?g4vn^8=8iZa31Ld|`bd-Ti9%TESN^hNkJz8pZ(oS8dM zWv7mfm4%*l?xa9)b$%V?Y*s6qzO=!-05a=)UIIqt9RGOX) zuB1p$39c&pqfZ}w4TJM%CjP`pNaw+w z@!Nu}=`TwmH!G_34U+C9Dpzgz@m>}x zSJ@*WmuP8LM45t4A&icRBiW{EhXbF~aUIe9!Q$ELmZgBtyUqP&e7eIv4(17l2O}#8 z@{Xx{MZD|wcy7l;+$0oRVHuas@mP-}xkummY}5IL7Sn+ZnxlOa?A;-B#fq(Xj%u*` zLOCTN5vQN}wYOk~HJ_)R9x%LbJfW9;3R>Juz_@3-!X)U7bS&+{WwGG0~Z8X2sI zFjt+Rg&ht;NHwK%DCu2+Vt00%$=BJC8G#pgZJt>zw;%_b8xm1 z{uwvFO6^s^n~~;M?KFFQMJJ2&oT?$&Q*MpZuA!>eaPH|`(&E_@<13$L-P};FSAq9>2o%8VYmv9-u?!rp1^fG%?vaR$8uAlND%QmIcxctD&LKaf9PyGbuD(~o zXtciC_ag=;jf-I)7Dk^?x4vN)XBu^5(Lk=H_c_lZih1Fs=4wtH7paOy`rpXellyIm z^+jMoyK>IN)mASB4E`o)$rr(%J-&_;pRgzli!@ohOtJRELKL4nh&!v43(P?3wLY%n z+a*Z_p3g$GlvV+TDPSh%xRrObme1|Sz^LbxZ v$U#8=ljnhV5IAMMDD$(Q^xPfR z63)kO4yBc44qjJlMn(}|RoC)~yuHzt#Yvj)~aZyEW>6jA8`A13%wO&e*o(k#KL?(CC6c}{Ya z?tE6o=-hJ`UjSCdwlz){AC~`oQY_z>xjgGqBK?gZ|wSp==AT86|X~AGrme&DE#^;|54^Zn*!1uk<{d|&p{fNERSB#T- z5&WSgU{%cLo1b3HIUmmo^EW!-RWP)1DZA-BfcsN1`vRxj<;Lzft`ZsX)v# z{Rbovm_Ii7Mv~N%`d?JsNaSdYt*B2`&@XA_{pm}Cucy2%B<9Lm*JpiE{c%BhI@f0N zKGsGbJGdh-<@xc%M`8z_ojwjsz7}I2GOK!|WFiOqtl=vCsxO`}+r?#k1luC)X^pu{ zzakCSOKq&Nc(Suyxj2f%#UIx)zJF%a0zz)iv4|IEw40=X{Z{Gwsy`r#4bzb~5F0Uz zTh`B-A(v`G&W`(jd%GVfn_2ROCm;`|Jypr^CG+yCi4wzJ6JadB&6HtA2$dB?YmMm7 zwFuQa;?6=AFGhKW5EN4S2lQHOfb)xUZpo8*qMd##buA_CY5jL#;Vf>)ekx<;N^dI4 zjZV7fQ6c_Fz*U-pQ@*uRgoa2nivhOw5t(Q%(}SS!$ADTtmu}L|ffyt5T>(Ar9Fi7S z5q!PNi{>gt`-}C{^1CfQs?-dbiCl1;Rcy1Nh;Xt`r{*5P_JcMXR_1j6aPY@8U|p% z2)GD9=%sSe&`5kf#TMHw$gA4o>G8BekZ&v_HHrQylzX1hn{;@3yQ1k5%8!8LO?9=c zF^^n9gpQBU4*cN7N_&$ju+F<3_#879^_NCvQWems+DQLfJ)lqn1%js1u5KrG}@f$RDsMkwvabv^D4OK>Z+CpJIz+fT-vSLaf?R zoY8HN?Cj5C#D+=wjnN6yV4ODi#^&D;0BDsY)*;H+2xIak4>xA?aPS@S7;KR?J^o~> zqZ8q)Ur-_DhtM%h7o2ReVQ~+LugtoN6Th!W03Rw zN3-)mW{zlBeoyF-%poMmqg`3VrRj;~_%p0cuX~JoWUafnd{_?JV&UTFAY}^!h2m{| z{oydqVpz&b0AD{CQ03QKDUqCftKM8fQO<2z1n=g2-)&%UN-6*K>gu$!G?QHs}{{i_-Dzd>3U1gyqx zOVbJdDm#YFaGhB$IlGiUG5!ulecy{Fcr0Ed*6$({?i9PW*yHKpPSvbPla}e3q}Tcj zSCAT5BoNbPw-1}c)V5mYVSgGwD$KAChA%nm>yI+U&nwB1MeP0v0E;YBE*t|uyrh-J zYz4R}FF&;cxPB75fv*p7D=O?Qf{nh{>qpdnscnTF(BU}}YgY?ZwoVAF21uw8y;}3a@d!_j7tC?_XrL&Va(96%8%)l=(KU0*4;`C&(Z)Uv5$K6 z8^G5g>?SoDhHvzXrEdv}LK?HGy^yO>^)J7GJKN=YlrvzEAt7;Pkdb^J1{hFo!IG$bUGwe`cz^9EA}Tpdtt5Js8wV%S~+}k`+Ky`f{$X9+#%(V zxR2me_ez6ZoZ?yaSPGHO_Dz#7QeIFi2VHU_D&n_H+xG{i?E6KJ#5~CpM1Puir^%W7Nl0bJrmyCmK{p}nzv@yloT)-(sMdl+3)ga^ zzP!z(pJ3i9xq#()g=Whjd*|{(y{852jph#{N4D?KfNAva_axheSMV+=YNO4lgqp+K zZ~fmrwQA&Eo;(qkvxt?uuwbhiBfY1va!gN@c@hV?ukzhKdgYyGI@+jWpZEUq@jyz_ z$6IgboV-)ihIyAI+#Y9c*n*P)lsEKRv@HE3JiD||H`=9$MF;=Km=LN3&Rr-3sEMG=+FP#H~&wQ{9j+{YLT=^{l5&u0K6o_{WNK>SaLZq z6IbVYniM#-ECs7g3uF0-vZ8$~(5FI%*vQR?Q@qb{FOM3BG(CuQ!@GVDb7TWuJdWio zwI@IxkNKWfH8Ofy$^83}BMHs?;hGg#G19a;aclZCDO<(t4D3bui<|7YiHKZq3I*M~Jxe zId5L^SBFO6>E8OQ(|zVy*01qx>SO8fzopC_Lx{ol^-LyVLMyai^8nr0in+v}>gSR> zaJcX9>n+VsfG2wbQ=xs1kCXqS0MO7e8j&hU=jF^f+gu5agfv~fN*Z6ArR5Z|Kyyfv zPQ5G4KvHkeiV<|KQiBT`=5w1{+bqQNmn~f~ydvr@w~2Jlp=o36OHpH23ffbJl|Rfn z+vc@QKuUn{^w(Z6xu7fphD)<#{v_91aBZ@MM6x(|dbp(H^`LqXdd0!thdkZL@o7zZ zp7O}MQc`c|o6CDnv$tFYio`U>n?-jkrU^%)U3@-xgQ?ZmEzt`|Y}c6@>nqV=&#=X{ z-=6noU$;p7?&BV?}=WvZL18|^O@VDmIKX=iq{-%+$bi>lv& z2dx6Qx{Z_n3RSt?l&)LS{3d|U3H>zHYi zh=V&47Z`pgf+<_45m$eMKDEgVC9wh!LK;P6-##wBBfA%LFwCV&s!)qqQ(6v%=H7Ac zP%U|rp*RkyoaJ#RMRcd7UFW)UXn6Vn3HnPk#7YPfe6P7(vu*ca2RZKBnVzw{Z{~O4 zGGYnj@HOdr@`Cs%pGBjZg(~nY5xT6iMR&+RJaya-k;Npn89f(^VrdtCJSax@J%P~h z2n2-RJdDGG?Bo3cgJrVY#0BPLYEE@j+OB;Tuet~I8TD)3+pOoDu3^S-r_=@nEruMfzxn=^o4SwcF`HYIM5dZQTeDRQ=n8SE}Dp zS4&|7Ao-?&iF759A~STjKt1nqIpzB*?dIRmfbRxERH|-KggKf+O*EX&PLcc`XzRK2 z9`k|q6LkD=_#aY!oZbrXvEMw8Li+7;))25F*-3;Z%541c3jg&8!!fFu<#kYNUFajd11JfXh0Tu%*gHO zsIgTo)g89y{@ME9iN)jSe=`$hvPcJX0exca>_nMe3`e2fQ;q-F5-@tHak?+SB=`ZN zRFC0z6v@ESBqK+Yi%MtT{1f7W#GF6JQK@ITSQ{h0TbQvjZhE_^NCo0~HBwxvXSR@U zoHd>@7<~xii))Qh!fHKLzqxTGIVAu09mv!7oa?ZxQQY01mK+Fde3G|i(Tb1xofI-b z(v`H681ZfT*w+Oreleu9c=E!!(WF`LAXP~I^6{-fnhF!&l$S#;n+`^tU$T!u-Gg(c znwC-U;S&19vp_pq$(O>@;5el-&j8KJGo5#CfM#J5tWnc7z{ZDun$>htp^MA_4e2r^QvvF`eYZfeElE;5KdbI=n7r~Y-fPmgfQtA>-AvU=3`9Dk1|xGo z05w(}Yf~F!h360gvnB|Z9WY6m7RtXlSGU8FU^fM#roS3}2IoQ1fpK#YMxrGqvoEXf z!vO;n5nXv8|NZ64uTeefN0Q3K1?+H5p5Uom#ftaIz_;OgxFvs0vVnS3I>6|i1* zO$ikC3SYuqU6Oi4|C`=eN?K4(AZ~3{Ya7`okZ^<&t{rL|nSy z$q@ENPZVy0xld=iv5;;H<;rg%K~z+W8UHi@1gVErtJYE#7vtBy~Bf+_$H+0 zrl_j{P8XD9w8EhRR2#!=%o7N+$bt+WR;*3^Y-XOyG~gU%QT<1AizDPmI+q0qf%#qD z+P(%FuGDuSJ9pikoo%$Ix|QuQ*I4c6kUHpx59nc2 zHwn?OO8im~soM2F*IQiISUCkZ8<*_@^T}Ci_X0JEMR{qBE+H|qk$3Na43HtwH(=W- z6JcET9l#)M7y0E3sr2OgMj97mYz!Gd&-G+o5;qoaiNoZW1pXFD63Ccz?jDSX{`bKA z8C<%E8{L96SSFKt@PYM8b!@^L;pUy!En@-H8criyRTSl_Hz0Zb1t6XC$+m9LZtc1Kdpf1NrE>1v|AKj=qJp!5Lr9v9^5mD~`|k zE~ZD6=`Z*JdH({Usg?BQ2-c|)U?;(thx;(_+u;!G70&2X9x2}z1vHome71=_gpiuH zl-^(dQpAI{9zg6V|61L`i_HMFyp8%8oiXHdgPRxdA91H_REsPh#WE&<9z=|A`^L_q ztn;6Edr*M^N23&32%Ee_k9vGOOg*;=Q13Re;&%)e^MThOAowe90Oy6=5ErD+( z8S{i8`ISIlCkbIpP*iRePTL*{G#Ok%P#1w= zr7eiZQCUWU&{+6sQfc8n+(?VKEA_%htu&Jza;6uK69s!-C9O88i%4Q^63VkrLk|w_ z03`UP`$=uxQY_GthUPz}^KQa>G8w+CT3-^%7WyaMZJpZwkB_s|BAlfD8*s>cCrdi# zKc`WeS^XZszb{WRjO+YGHlemh8LcHqsVQQ4w(tEfFc_O~Crrb<=w7pbERg!W&6Gg7 zh2G3_+P!~7L^3iF-8$XaG3vek$?5MLvtsbcuES)q(UB&ntRu^kho~anm@kI*el(D3 zTYl8TdiKZd!m%Q)_Q|T#7g4tWit+}J6@>3IJ)B&<3746nn+h(%UdrMzAS?&Y5AL)Yev8NTHJjA6kRscyk=RNWDg zSg(|EyymW8&5Mn%A9c9uHv_P;wk#q<#%AVd*%2-*zpax&kCtSDO9nFK(bN1SX)vfveEC zFA%3Xboa(1R;`>sTjOO+F>m;(fW6uXq*$Y#n-Kw^W)ZnNZ#{O?FTC zfnsoiPLI0L26V?lt^#WNJ|2BSY;^In19p0CHrS72Ry$VlOiMevzhbR@EvgTSi7T65Wu;SvKX{qJ2}&X*Brq<{>LAa%Kqx3YA<`Bv@7MwS;m!Jt>%1taV@XN1z>ym*Ijl z#RKHve$`P{0ejz-%^t{*nX~UkWwzo3drS^kiYEkm9~d8?{Wds=QNMwknc**44)PwT z8XKV$zgv-)H`Kb%Z7bqsY8twwv^D-cpoOjlK77>>*!ir{(sJ{>xEz)A=wSl&z=>c& zK?qtO_wbnx5_MNJfSdp2IYv=-u{_M_YaxfnOO`Lt<>K>qWT-d01v0)mtY0bs^4)^P z$-j$st(v|%^~0kmMtlia!v{|d)^TaR%5A)3cm7(;&Hvm>^59ks0sS`MJ843S-MM#A zX8y&V^LaBNdjbWK%!*swCaK?SphRtHTB3}`RDicFh*|F-qMssiPZjpmpSnK*yc=ob zH{z?(Bv@Ih(RGkv>#`{kD$~y8LPZwO%{xjrFU*j6ElWryST+-!kBZ zBC?0>OrSQ;>7#S@r6ElCUUF(s`Sokh_INEl8rh<+p*ZV3U~;z9UxAsqeH3LY_etb~ z4dS_dKr*4>BjtMT-c)Cp`tvpY*wvi6Zq$GvW31{6$7509=l^QQpoYP3D4F~~S^gvi z>?1UG^8pl=?<7^rf!#7SyBAiiUQ_{zPltwm*WKHzhvKRO9;xfV-=mr6tAz7CWr}2vr+xJJ^>lXlvz87I;EZU zel$+oeq5n|4SV7Fg1@gZ=;8Z$Ct94ww0FUx9Tj%&jQLKzq)blSmwH{)!kQSu&Z-hU zv`fy8P}uXe=JqlQ#FONSwK~Vqj@3W<7oxcZhXpX!LOKiI&o^D%h#;8Fd}SaFb-EW! zK9KsFeWZAQpNrvxON3|##p^3M7CVszA z3O*mIU>?;?%31wBI7!yEA1Ld=me~cW9;rp<-ss=^gS;wT{&`#+O7Dps0b}PqASZZW+BuU0eS^zYsmpR9N5i) zTNq3rM`<~-umYB{H2x^Pf9k83@3*M4`nv`~gw;fGOIc+AD$?CZj@{h5z4|haAlpcH zje7_FV2#^AYBIZ@949w~(q5f4lU6=&Ch5J*YR0XdX5>~_5uTQWo;!T3P#lCVlvXUY ztW4>@$@dF{1#~1a9!Zassir5{m)lM#kBmSEX1hvv;ADk+>4Zwrvy}}&?E$KeJ?eeVUcBEZ}=w1`bmfb@839zLtK-E1h z85Qv!Ki)spfrU7ys8t6RSHLotlD6P_F|&t6>4X*5E5T~@8HhY={?bs3pX!bgAqg%v zmkyjTYOz!Rz2{#qAP4mu52VK{VRA9=sPSGqFWnkJWGKK^04>S(lql~}PEt0Fl59S9 zz#!gb1GtG?3ptnaw^#ufK7U%Mgv~Fs+3z^{g_VW@Udc$$3Ia3c(Gt$B2v!_&%q@f& z>#QTJ=b=OAXaPs^xc=)Egbr>|{||)09+ZQ~HDEJW!Om~gJMGT!*?kAOf76YhMx_YD znJPNp0e?0o8XWvAXhj(O1Mh!S{0x$8{7mUw@(yw|urYO7g*}%)5Amy<1fB4^u+i+q zw#%Ld+w!#l9>~{+s0lQ=@wk=+D52uZ>iVR@tM`hE^)0@q)tte$Twy~FuA{^3?IQ-v zfCQ8Vb(q_h&Y2-tGeK71{^aa7MdmqFgo)Vw;zv~hZmIv1@9ED~2g2DeNjB25zz^Ap zzo&8i<0xd?%cN?;tA!q{W9P-jl2~%0YtY0$)L;fW$B-g@;o<j#39R8uDg0d#f~K$!V+_xq&c+4A+)qfYdSx%6*G~OI_ypi-zW-vDUqDhDA*79r*Lnb zTq-H0JTu@uvST@4?6SV|a`+Kw-Qv=3X7%$eHT={o@P(MxWI5H&q#_EUFa-q1a0yXW z9VCpcG0f@jVl@2qOBOHwr}sx4@J*|`WC=1&+n_qUO@W3bzh(mPP}|esO=r~*!Ey$=l@*3jUG^f>kRGY7;V~9s6vDxr4QktdW_pSNMm#jK`oGyQoeYlxXvjQ)3vYX; z{0Lhli<9e-DzKdB;b@OjLMeNTsAKuBg>YhYmUC%zxp!NFx6lT6;Uo2r!0h}EQG#4` zcB1S{b8q~bDp8i<1xQLe9>_2KPzU?jIw^~q7*RDaWeJq{IvJ9Y z<@{R_>>hX5qVg7+c9lM9>^=s=>b*Xsf*DKnPuh=w8%nYTr9dF4Auw$#ISjAC9X`K> zmweTO`h|@?Fv1}J9XU0Vi`<_^`4Pl(fe+cn#X8ozKqRrfAxKppLY!kpre=J~r^(8X z#9nRvT(<&o?*wAKY9V}Tm_*$dPD!?!w#%cg(x_kFQ_BNa{5?|cU%eO;$6H@a0;_-i zT@sj(_d&ilV_@(REaXuI(?;sQ+gO|G{#0tHZ4<@)`qyG?Xu8#xEZhXmjzZs3psX91 zlbj>!1d9kKt*#xxj{h{}J595BPe2a*KNPQ;Lzd0oHMViOojSgy`qL zN#CCuU^ip4AM^k*is5Yncrj||TJQULYD-VUlv_Vvvd~N66W?R*6F$O6rHnD7S?G&w zz!MAIvC{|*te4KI%IcR$K!cGxeQB7JtK8U;BiMK4g3hs#AgICr7lY|N91|+AI(YN+ zCp6meuJ@~BygclV9&~NTA2VEQVo6y*oxzJ|uievzd~S{jrP*8-BB-1qiosIVt61NW=!fo>c@lV;cmVwe7L78VtpZ;03 zV}Ay504wdpK0ve%x?UW0l;ob!72KEZv>tdTn|&uP`SfjhL01TCRdynM5Xpu9W6W3v zQ5}>UMl}wJBELl)me>PzX*`j+nNmPtCtQ_AJ5EBFq1@RHQ}_76GOW|~UxwxCRs+Er zTWF>gq2=4j7z{iE`oi0u#R%A0Ooxat(2&+*#%wp+4u8m^oF`)&TQ8X<+3?|mt*itc zF>f&p_>@5j7$A+ zmKZPDagAML9MW|<=u7X>Nocj322$kd2|5wjN|yWu@HuDvIB^IoC^av=Ye>*hHI{$L z!r%W~Q4G4ao7A~KjVO4SV8R-F1qF7n8J$z&fewVt%y1x!yp`U6ce%+~Y%1I*j-e}N_uZMFzwZC;%> zj$kfS{~6glL~VOGQB}etQx8+6!YGdLF{6B&Nx}iG}HCg?`utN_w9_U)ZU=-)e5F(%#CYUk444cXD z8IYRUTyiD_B=JV$4sqhB?w)M8^*7O%E(IYRV*8{1|(dIQk91h;|I`8o%2cg_C&7J;Bq^hqhu(npOl zCA2Tg{KS|pl65|nP<&>VA1mLKsNL=mpfDh1oTmBWVZubjHIHZB3X<*e@DY3+d`Ul{ z_?OhiskUQk0Nixz`Ji%6spdkZ&~Fj<$) z&h<0Wsb(ZRFxdqVDi_UCgMp5MHU%RHWzpQSYe>va+zz{}fqnM8j=C7lba0Nzo~e&B z6ClEFwzPG8=0Ak6e(7G+1!0upPMqX(AgJG2eZ<@FpG6HV*5-utGsvRj#nA%PmyQgm z3n30Q(7_m$y@ZcY;64hPyO3n5{brRz{X{0{aQ8_SY$OTe>-qaqJ>;=Aeq);Z1N09Q z;Fo&uHGef5ucVUU>t*8d31fZlo}DiPyYF4S4gdjyIu8Ew5p*qGF=rdr4ZM4pz-i4+ zHu&aKJlWs{GTKhz-|fX!;C71M^nfjUHr-?_&#e_6p@4(NJUeu@wuJgcA#@X2-OsEn zp(J}1`o`*DEvi?`HHAZSHRqF#gyfy;g90lNovC;ssDRR-%n?bHo!pDt$@iJnEnlx( z3LJeHb51rjpXcV79(-g$80N7hQkCf6z;VG3{K?^9{iF zAlBda=Tyb1p(Wj9Op2Z7W4q0Zf$}LStwQP0_7tIA2RZ8ArJSnuvz=f+?_e8%?J~?{dZf9R*Cf)L%H`kD(DsN_;SI8W6 z)8nWlXmSwg`-btltcj;Q@{m^VJZIQD)aoHD9sg z$m)ia$X)te8eSXx?dtAC`INWKMM0a}(9>28v9{cyO^6iEbNv**Hc;lgFMla1$V!n>bx)8I1KVR_VJ8pQt$T zn3UuxG4vdf{+7D5pi=Z_+0N#$>^^jJDV1>i#QXSLduT-JLO|>daCNpvI>yTCJR1W& zizIHNpIIVw1s-%5<>NH`8pkv(Ucj~22IS=~<>iz-mj?L{3OU5{%OOnD-Ynq;f92+w z^Mkskl_mCPzi?r_LnH~z-e994y-7EI8_b8tb6Tg?8ZuaZFMjw287AokxudcK`@QU7 z-nJVpd|>+3~~Xg<~8)gKOHG3I4(dg*)j~VPWG(4UAsag#k*f_ z5JwutoiENwR%ZXr@YU2;TT?*6aOIPRBlIi7D%3GPR(z5T@stw;L$-wAX)&40w`XM zIDR1m96KcQ-+7m7EBXJ*&?a<@IK=v5A!SjAk*REExg6F3auXsUkgU-p2cBY2h~nQzX6c~)wmv-aqNcYg`|n$LK}$q2)DUM3veU8}EQ1YD2+ru?9g zubbUG13e!OYdu;`C!s+*scqV#C8u2sr1+*y15Mi7_@>4ynXd3cur#Nj0l3XIpKg*qX!_YmLJHEY|iCf5GU7@A+%q z82-K95|x5J3ry5Z{IP@6_!tU!IV5m^_#YLl5yH-ko^m@J@ATXIDV*^Ub1@O|Cj&uI zhq@GI#dZUOjzzrVkiP0m{a=dUJ%yJ$;7?d*PrTpyPVblV23c}^0Gd>D-mKc4Ynpo+ zmSjuh+Yo-d?ou5ev`GxAzWc6G8tU0>qf?QGjt{&P0z;fbLOxRM(+Hum?8!z*F8#-6 zz>l(;uxVr|2l0-x+3PiSYP2SmFasiQcS&jWK$r}5IRIPdx{124YBL-rCGPf znP{lZK-{eIb{A}ZJ|xN8rIQX4B+@>+N_llm%zw4C>f1Y5H>dbTGtn`B_tK~L9y-2a zt?2|-;M3SfiKv)lzEaIaU+0q*b!}*WJm-#zF})w~EssD!!>5=pZq70-I|Ctlq@m*& zuUPyolF;z(XdnEi{TR?c!w&G}`~`MBE=@WK@4Bc}C4Bj- zT#Llg3Fw_2xxwzt=>NjXEq@>y+DVuY2nTb8V;teF0x9n#3aoWrbg4+tT{>vk%4t_D%(G?tH8jxBy$t-myUkO=G#CAenxJEQYxzF&aBtpho7ZPPB2^xOPHF=F9t_IKESBYv`N_NZ zwz;qn*9AnW>(r9Z%t%ge@O7(;#9+V!o$KxmVrGCLQ=_NZ=3XD$LG)>obzdAI`=uyl zpc!Ut2rN(w(A>vrt#ut40|@x5JBcpsz!Ku0C|~OXirn7@D&?63epHIyGNo?AN@mH7$pX)NLOwpYj=ITHv^G|Vo%Q&{q9`$I4L}J`PRAwzeE?R zDJF8}zp2ZnTLzw)a(WUciLA>P1s-J`^Jhdl7rlQY;(;8Dd(DpN^OexSKb&mydc$B8 zYbQ^>1E&hQ+xG+r3Z<&}3~B|CuJ*qwos7wgPL8l^*8R2hE43!Z@1WEzR=spRbRb20 zB$l}%d45!U-k+t?n@g$NKw7Sa20cY<#cw7Enmb1x46T18cYZ_YiPB9qzy-1-)8*cl zdL*VQXB%ey%wvr-o$!ub4a-9giwn9NW>*uYPOZ^z;Q2HQqe+vH^7d|}c2}2qEch%9 z4$TTJncruQJ6g+U86h5gPt15{lDSVmd{2@N9ZxZex`9Nk`?= zJk?O=4{J$g4Cr{c6N6<6@i^d&%%kNll~hXXdsSLNf4gPi)pa~QS|R>M>b=x&Kl_ao zCYik;%MaMm@zM|SRPJ(3GJnYyy!)Z5U=-n}O^si(2JCqJ50}i!B!~FeP4FjweH%H$ zDeZauKb2QV(wz=x*nJhU(N#+!hbw2_F4B}S~q{j)SCOu@`(&IBO(BOdLIv1Or*V; za*&tKqtS*-QZxXnIW+nswJfVb?Y{aytKsEju4l`>Q>^2u0mLNQv;j~b{!n+--MS2 zYY+46#Cow+b8Q0q;?E2XRQe;s-N*3+c_=jjNUx+)DR zuygvfw}!>&A5~nBpVN6J(hccNM$=}_p6Q&N=NUIxg#-wu9sWr97su+A{X?B1^S+;Z z3s;)HtUARhUe4qkw#nStxRg0*Dk zr>)L13k|PWdm{h>ZZ!^6q+k6rDB+@uw!MsLItwOP{cw-?oI6Ro7^)J(mC(jvGf$7aL zzROX6$ufmnk5s(ewh#MD8>cS%l`&x0Mx1AdD=uDm8} z**rP0^Q4;`Qk;fegeKGLJvVCl6s#Rr@S+8~=4sHC(a>X_5yW(rp1Ta+x$?m@;yU!J z#4*bw+;?H<@DF-xb45`-j)m_ywUkrKtRm~4j>%#fo z9sgawR>9FXw>~VkPWOh?9L@GNZBbchp+T2#C+L#Kl5bj7NHJfhK>c^Ug7W&W1+7LK z{{h<&h{zlIkk2e{=n6*si+VDO&Pu?}@@_kbB&~%W4 zuY1I&fzzN!=+cbZ$n(k)*D@mJw#h29crc5Wz;mmc{ zgl1)x@`^NYzczCrq&wkV<_qWNrR35(#e&Naqsi4 zuc06>w)S*}XV>JPxfjLR&M(j#oO@G7PHWc^VT?t- z_>7+l>E7O+8LwS80(0A+Z4M2LpS>vq&Wd(I*t(=K`0&o>-67qSFLsjmD9?5b@2PeY zn86~DT1;)DQkDXET&^ktA0FNRwiaxW=45xMnBa}~T7V{uKSu^ra*Q|F=mz)L4u7`E zjvK*4ULjvLMiuARVp>tAH%ijkPme^`JHVcW|7jY~g;bnTwF-M6XRcrQEN*U6Jg<~Q zdE6>WlbhHPa9wTsT@TttsIliFrSJMMvM^m#polPbZi_s=XN>c5zKNl{B{au#d0euP zrjH-%PWeE%gzoFki4{=r*3_KsycIM8DPg-zyPofaP1g90U!|>b{$Rw6kzAht;4Lo@ z13a9{)T2|<^vij_*~!$xb}Jz{&=YR!T2Gmuq`BOer%4|9m2`=!`7_~HXSwf!B`l)Zy}tou%Vo- zT-E5X@VO}?WTDc=C+AK%>69*eDV#o~K0 zUnrLJBBZIy$MU>8*6u;7xu{_lNFW=DkU$DoCiXg3QF`&JEw+&pGr3Cq>t3GI<#yr8 zo55Tf4c~`whrv|)%N4cEBNx{DY@apv)7;b>Nqzg^8>J*kajrRz`9Zp|30;dWrJ6s( zh_(Vf(##l{+|WP|TrI^nb?KR$E5(1Ww|d*6%fR~P&Sq2{*jX^q|H`EVvSYtNHqCF# z%`{S;l`Z`%-wcHpR-)nKWkwzS|H{cf(mGppDcjO?me2UjUq*?KQiEp(ZY3y#1K~9o z=EX9SXnV$V*RZ1n-2v>UR9th+tsyO=Czo@tij03g+)ygk^VuBpciq&!vpFVYx!FVx zR@m+cv2XP1{f5p#=_b$7IjW=nzLVvX($v;8;O(XLj;>)b9Qj zuBrCX^1b-8Fn|oP{SuqJIaO}g7Scj6o!(?2yTyMuK9tZD367JXy4n%4v``>e3Xb!H z#qx`0gdCl=?#pK;Ic}L-nt6~qijYxy>x_m-uwUc1&jVZL2I~ed3KC}AVrK`%H_cV- z3U689Sa`+o^k-GRmDr2?rTkt`NAn<)Cen2^ba{}3WA9I>>_VsKR8h~G=M;k$qAC8a z4~$gYpYpp>1UaimH7sy{N#}n}(Bwe^yGEA-Z5{|2ymD$mNc^z9E2Ep|^bP&wjr8CV zUBI80jv|ePL8?ZvgSxXX3a3j^m%S!reybUMLqnB9?eiZ9>7tYM%Fj6!Fg6`R0qNZQ zYxna0a>*LF6z?9tYbovu$zzuBMWDbgw6%Vr^!56N`9}HY{ zFR^{f&wiTx{;a1n#NMvCvch)!UpZ*qWe!FDFBd1q0aDtX`591_Q%5RIUMgxsuaWp5 zm-8_;GcT@BS@79-K-uPT%NFF0YnH@@2sjlPYYW zAR9~kq;CJTVS@I}msstFc z;F_f4hz2T`KMWgMEbR*(35>C6-#bh)k3oD3awcs2!ATgE-CIkZ>Ip#BgZ|Gc!6-~t z%$%4q#Ay1vj28hr!>SH`HT!3(k=hS#q~gz*X7i{ESvmK8-rh0LxbW7CumYJ={6eWIy4~^QFy;cJSbLX0H|&A8B8s7J(vOQt{|cfo;-G#7qq1c>kp&)1fEDW z=OQ>0V-V6!Bfb`yi)M!#sN{UcgOKlNU6+;UYn66_A+wjn*Idap_}asj*>NP3!5VGaha);z-psc>ujc|~Igmyj8} zrT-Smo3b4^ znSE3k$f49TkBp`K85tX=50<_{)#L1VW>;7PX4NT*K?dbv!BX!i6&hJ6VwhpOy=HDc z8e^!T^41VH{jWqVG>GoGU>vRa*2-tp@x4Eaq`EM{zkZP=2ag#kveb?lf-n7dct9>A z*PL&B=LwExVj3_A&_G$sec1a&Ep#_5S;0lddf?$R%~wCW|77Q)EV%DL1ns5i+FM5g z9>rOz6SI~Z^KBK;y8Q{yKcUzRBJUZi@AbfU+=G^)_}$J!ry2)CMdT^W7pQlWJr91B z{#Jk1wy4ieN_$%H;nMYoU+!?Q3D+KTBJi%ae!TpoALoR7x2HwEIPBHY{g&qBw^m-z z71GUXHw0(X5o=$$S<5eDe*CE}59luzUu2s#-_76EB9ThI3fHnnc|jF_vNn6Pu;ne$ zEwEfgJbbLH_k@Pm@&zlCxyM%nnGLK;_C}P>&|V5p_V<7|z6twPF1c}Mm1UN<6eG*% zFI_jEXyu{^{E#o=$(ba}vqk^s_BKuxE`QNz?1yh~Fys<^O(Sw4Il%5L&wmA@!?5-9 zXNX2!kSYu(*#ghfi-W9krHIwWvV32{=)|iHM5*mR`_P>6P=Q3dYGz~0eR#t8a})F7V{5iO-A1RRR0dQgFb0fy058WjV z5xmN`_~pQI>kX2uAWP@ah(S#<(e_>VqtC?6DtlNpn99u>MW<32xnQD~w2kDgx}0-A zL2Jm9&+cG`x}+pAPTB>Fmjy{^LA)l>@RMW+SmO!Td}y-B$=}LA1O$lB%-W>hR?oRu zuper;>XZs}TS9FkE{G`<*ZzUqRwWu^IOqp|ndb~jTL2$*7$dd?Y%16BJA#@rZ`EvP z+q#(5a_CYup6jUmfrlS6LgutoJ@4ZFs3zu=Lgmp(>0l^@H^lm4C4-tEQ5Ib|Ck_dC z4(g3NAHNxdhxIA(KR@#fSV1UaUs^_}If7J31K~)k$@hYTY9JaiOre6-i-oof;c>M? z)Q~qZuJ_)02M98CL>F!}T=)`m0!hz{9Z$x5}@_h-D)77<%5u;G~x?k@i0p7jI zJ+@$BzQtu#l-^8LhT)49KI-3BkKu|no7zwRTt-VS#oPX2kstqGV$lB?i2lFzaSRVf z-h;8e*MGo^`P2PhH-p46d;4KI7ZO3{Aq#ds5DGgXKY%3X8xG^_U-Mxq45vBDWPfMk zUiAM5SO$c@OAN@&sC?1yyaNVxdHC`qnsdDfu2z}YltukcSm;OmVAx4VB{wzhpYy|b zDb}5vaY+X%;JQcG(u+{li)=^<)Uowmfx;>dTaOuwxELo zDb_O$&U|IU9}#(kZ6y^rb4Z^Kz&=MhlN(2~4-g?ryt%y13QgTI%`<`~x?k5rjo2Ov z#~(z}Ssx^9BVEHj#dPw4UZCi4rT&#?C$y6(IoW&|4okUl-hFON1&3E`I(|@AW67(L z5#_gp+z8C^f(Ayzsh*QaWII)u z2fwViDaBu+1b@48AC5>xVR5(_%l0~q>#q;(lUi!sDMq-nUQ@trp8uq%^Bm{hfED({ zxW9eh8z$4pGU;ayM^lgha@~hIRs)~=f#CE@(;*%7QJ5F{66$vcyySjl=3{n*k~u0| zNI%+~zTzL$D2$T_I&fcy)DcmqPmQ$xz^jfO`&})g44TYW$MZ#rjLrT?Smm#LFaIMe z+&fILIuY^Lz-JAgp$XvSFigoDwyLUUx$2g{Zi49A0TYpN%;(4kqwB&v=3~$BdrtQ& z@1^G>*RCQfnmr^#+geeOu}I z{;h|mkA6xE*XmG_YtZ#$rl(&s(=I&!d^lLVoLC=pnFmuxhzh!#WtcYngzL?Ht8pOH z@2t%d5X_h4A`Xn4S#3=#H+S3&)XxighWN=`L2)<1afX(AogII62ud|tZ);}{33A=t zfIy(9Wdk0Tf=c}JjwherT#E*dsj`A7*JD5e1BbK;q$a;C@%K~zESsXyX4^NmTEqlh zC$g~usWC%B1KM10(9XQ}@x3_mbz^RLO4kpJMB1>Uj87L!`$^Xx6~D5!Vp$XSG)^3! zYH+0#)s0vE()F|I@Kfy%Evr(kQ5z5*@LoVo-~dx~9A@wfxBUD+q%y@0)E13!!8=-4 z__d(Zcg39AWa|^-ZXTU<>2~L5U(4+;S`n`OOtscWRg534cYsYi;`arF@tyyyMs9Lu z=FO6$RADn53o|%M*_n@fCC{}ao#`!!GcS;Tpt|1$YvnxBp*_m(R*3?J)mq#tiy)zG zVA7y^^Ef#6&I4CLyHP!lRTZ$CT@+18#C`N(5qrFf4aoDFc10DMCdB&T*PO}K>p|1VKEDi&odGxqFY`OX6sev~k2lWUFL=hhApQs{XqjrsZ zmtm|0S_+V~IB#9$a@0`dUNVRf52Aw4NKiucJH@^l9D0JYiDR`GWM*rEq$X^X=}C7Q zPW!fP7;@@G;aSjpu>s|;B)R_3Ah>M2r{c-BGjVa!L?q$M{#;>NYkVn3*U5^K6{w!k z{?r3$oPa8omty5|Zh~Ao+fNu01+%ngGH8?-)8b|eefbHFV&FAB9%@V$>vz5jVhfJe z@07P1CBHQY>9wep>*dC{OTvk|Ne}IfU0D=yS(JCmEde_+)gKx8{`{tAIfh7BCl+%~%e_*$oS?G`lC}gL z3Ifo~9mzGk?Bo5H%LXAvUmGcJ0V%DkxKCfH1yT|$i1Su^s1yRoTKxCcC5QrPsp9Va zG5CK1H25R;;f`mEtk)_QrUb5UVQ2Fn7V1C1Fe;Ex_Y;JGs2y;P@09vNUBIq@hKKbB ze-`776&A?Z{`}8$G8KJO@E!;U-eXbKc;IsLtS$MS3E+jk{b!=7NF*FN+@ycxn!*co zZF^Pv2KUbfYfMVh-Nnr_c@uk@9Ek%J&JP{TLx0yWKr@oFq%aDGa^GX78NY+N8#{M% zgGG+vO%VQ*l|^N2>K#Sx&g8OPA~V_u7RjrnT(iU~9X23+4Cn2L616Ds)I4R&b(P+U zj^C>3&NOLD$nEDlM4s(yI#ogsR4PC+?NPaRV_vDtst?&;v))8WmS@HKx>qV<5sXbx z*W#7=_(PrwLZpJPnt*h3Vmkh~6}!2MGW!v_#Y;)xYUUY*SAzoRXwWMD+nMNH$NJX} z8^qV&mJ9;lmfKRiqlFt-f*1*Lt!{x{<$h{)wV`O{Cc=kgd<#~B0L*&EaYg6qZpoj7 z3Z*o|cv$HkeTA9<`FBQWyEHuMUj#HpnE%WBtZeZxayA0E{w(MI5)v4_P?IY+9wsu1 z%*x}J^Ywr*S?#TdS}p@^GyYLprmxRXS!sONukaNU(fEn`1y^X!UtsOMaRZXC!xRl! zySA3)d4uZ4jSqCiG)zJ{;Y1agOHMga>Du(yw7JhSue+A*s7R%C^M#%UW*^n=>~Euz ze`i*$WHvT$)T=m`wa#m>;(m@F*348#*+{>zZywgG-ar;k6rdLk@bBZ07LUAL?b#sB za-taq{)KIhf3c|r9{$(!UMRT$nqs$3{1vJL}fj&Cr~vV z67y|373p~}@LCS7#~*Q$f=LOI1t{k8y~tvHl@08j(D{agU$~I%_Hwe|4=YA=#fNZM zGR7MI5hTHQ9isNOoD4FO5%$C>2rDw9gd8nMl3K+UV*aBP0Youw!kejZ-E1jsiRm%O zDWSD{k36QD?neK{s=^#RJ}7+so)~9pRQxS~L;C?mhM^Zd4}8C2h_EL(=ieGWqtSj4 zH+c?sOn3i{G&idyt>@+s*QR5*YU|DYjV`ti94>c7X;cQN5NylztvC#;mEsM$~oP7>25T0k7IqI_Cj)lp7N`1m13 z4Q5ul2sy)GkXBQX`RF|T-jDgGL{w8srsbi2d>crK?!PmB7fdLayBFRWF~V}@nGSc< zFJ8qeGn^N-8iX^<1Re)ZsBkKmlutOF-6G~2L=D$(Z0m4tj9Sr(YNjp^W*(ut8eQ8nNx9_bwhBUmvMcTrXCF|&(5)f ziC%YBTz0aDCBCF_^gSEn#sKa-FcYM;V=nYSm~kr?MhcF8Q1R0!jjYCFBPD-)%P1|# zeesvr=j>qySsD^M8OwV)iMC7h=#aJ@}MMe-E0qg-{$;gA#T%6BhScAAcWEY>fm?XL5U0<^Sg6E7{F9n(WlC&364myR6f%M)Og#ue*4 z_%B6^C@P%XtHiyr9aS}$L&2oec^E0JM}jUGHO6X&xF`CA~;Xs3BKaFJGD_m{*0ez7Vk4vJ(h36=g7tw`ej~4h=;DlRN43rKGipWB}2W( zrR5=l)ct`Z(EB2Cd7ZVaZWRYG->^{DzwE`(>~tNSR9$ePfC+czP((9~cRi!!nXZ&l zoVwY%okrhewv;WO`RZ!24O~>PDpB8@Jg&-Qs}X}ymTwoq44}T=sU4gj!mK)XHS*b= z9xi81%F@Rozni3`8ezEbyB?L6u#)s3k#BuWxW=(Tzb5nZ@9;j$Xxs|iT$J*o7P;ORiX`9P#eU1eMJNmge`4SKw2IwU}CaLKZjc(feTA!i_uV)Aiw-;cNpB z_+;Pl-ZsSyQ0ppbcNQ~|Rk>NAAB5}i*=s}2>gVq`!49c1)j8DV70%e$r0|xBWURMm93MmvcO1|>#5aX@w_q( z-(t&Jo(J}8ci?5>z#r!Ikz1g_-zZX2emAjVi;*agTtfQp68VH5b%E7TX?fDMfVmI* z`DAWQ=N>+0XDtTuVvTEKzNA$ZCdSJc|jn1p`-^WJ#!hL`3r z9yH08MBsRZMKjystS!LKsHm(~Zaj}4gC&n96jyh7tyzv@t0?1paT~2I9mGi%MC*0T z+K`rj`=_wAA?m;tw@T8a>BA=Pf$Q0?AlJl|Szos&W!CijZ*bZ2+3}|i4(W*B1PE` z5O?z2I<{5r)cT*&KOSdEntE=56UMlfSJ7BN|}Ejc`)$t&FTag9q_!0khvbR&H*t`vg4|x*O!RY4>3SX%APH{toc2)#9HV0 zeOO?rc0~-?OC<8vT0HXDea3KRl7}gRCS~6KD4tczZmkS8NvM?-mQ4RwN}DMl;gN*C(b4($xy$+?a5rKsoK2AZEx2C+}d z0+{<2{xPJHrGw}0wBXF>+LNquA9KIM=ijhfK3Ew3cyK|>1L5|G3L04R8w#Sa4+&T9 z+`uVUX!hqtBR=~q&y|_}nv43ZJYtyqt8q%} z3NHrrCki8$MjrXO9~GMmUA#+eABN=EqS>>=0;?C^gr z+G|cn2IY=q8$3~k=^}D*O;V(WC6F58r%&dlw)mAavM-9^Fb%h2A3>4-rzPh66SJiz z(EKM~dp0eoQtS;y=bI15JG}A0tL`!Zc=_7U5XQazp?aZSETz_|2!ar3Zv7%T10E_J ziBazNIDN!-#3r)(E*c-Q@b}BhRAsulf`J(xKD(Y;qQ~CfXV=vM{hui#EI7HQ=Zrk8 z^_6@{*~4E>vo&P4j!$w@U*ifQkmOXtL=tYNF$>E*f(f99t48B1Q}(VCd&Z*ihcgl5 z{Ye}pF8xE8I_z_HCftF|03C4!`3^t2d!f&zu10o)%jVN9BA>3F+6z{H^;xBpoAoloT@!!)wV zkM8kXCq>?ABu}Pb9sl^fWFizuVQfOtO-t?wAr4NSbmov+bBi+Wu!BT0rp3 z^PE8ct&jbCwY01L^0(JS8M+h|LN0=0nm@{}gm&WxzUuYQbmO_gM+Q!!7#l^3;)JBB zInB*bQgY+9he#qg%`QKZ8xLU=dcA+7y}j3W+0x+9Pw}Bq5{C!Rd{wm`;tt&odjNPB z&x1F(t{WWIHSMN8LcIKPFgFKQm-bXe`bY8L0?`OM+^t7-#(ZgNuu;~u3%l!SGVBG= zqzS#Y)I6vWfBg5iNGn;wg5k_)%8^{<-hRfI1~AoNA_%)fe_SvGALR)EUlT6-@)CS@ zZjIgg;S>louc}O7 z!}k>V0Aoxa_)lgz0nF1J+WDEUSPi$ofOY@)q-#QmvW@hebuBvJYQxzq*Hl9f@}8M) zdsp_(@c9bNnj>;W7@8^$KSP-|xI(HfsNHtH#j@WfAu4ObVNN;g3IX0Ob^3rNdeQAD zKSbZXg__)f1gJyh{mvrm)3;b0{PtOe&n;e^Y%Rc>hawghEvhN&O7gksJfp0I{z zg6Rc2KcTkJKyNmQjq6}G)V5npg#Gtc~lo}h6mNO+bvAMWXFNsXr);VvHodJTjH#Z z?(C!7xSW(Fx;R@SPtB4yYm$A1Ys{r^*RYC_!kcjm;8hqUhp>Q55a0cF_Z{Hp zU%FKy5JH}MecLhbBVe=6|A`@^#Rm~>a3<<3qrbv;9!d6TaM6mlOseWnc^xqXm zenV1XpHRm}t@`mX^cHflz~7D5p4iIQT@srWpV2*dE+K;kS!yL=1D`0{4zv!=>ZMB> zcUgYb=NL?1!n;(!)8-(5M)FZM=IBs7AnzXV+?l!nEA*j#=N6HX`qR*>%7Fryap2v0 zr@hrm7{<EV5A(^t9?m0XtNL+2d|tkxN;kSJ_<1XycYB0csHHL;Bey03Mrr+*OVPF zrRqy!81d2!xv5MU;AAz-&aCMzL*02E{Q&;x!@+=G)3+#Wkf%fMH0K)yKK@@og^8b- zM+>la&z$2)FEGE@2=F5bBtMDh>|v=5MWw$tEVZ`a%_tSIBmGp$8!R_Vp`Ncz7vdTy z6o4UJsU$K!P22TlJ`PinB03q7+PqoidP90IYdA9m69yA=GCkpVN=L zoI(UGfZvOL**_5_6XQ3#CJzjb;m&;TBG-mk-Rte^yvp}k5jz=e&op-ZqOY?8+NAQtJBM`~ zyw3$`J!WWq?~IeT4sqZUQBN@voE}!1hGO~xPf){{qyvPW1oX68vSCnf4w)LF@^NSm z*?g#(7L>Kg*s&FfvCflnec%Q%o!(9?$D9Rcy|^PcF3U&|x(pPxFExV=lu?9fG`ruJ zvVdgVOnX`s4B9l6pPlndNRt|q%+2(03mW?xDU#^5i?{Uw4Zi?msCMp2(*fexGPnC# zE(-0-%J>QO`-Z6O1HrTXrmF@0CcsBDZdOh_mjC;@PL_L0^DEuPfdc4(5(RY3F^cUl zs|mnJZA0op9BJgC@fF$TONH0bJIr?nJK-|$K(q7Fh`h$7wE&)2^bGz}3_CLTR!A5JT^~wpRv;E!`Rk`otS#JP!FX}(H z$XMAC?WY+gWG<5cV^@5!`a!;Udw`{Or!328lkm~2gM1mmD{3@+yztQ6AywF6Ue;gR z6PVvy&|iW-64vh2^I?mXrQ=%^WM$TG;it`50B#}fWeoizF-MT|Zi}$3Vvj;%pPrd; zofIK$Yw4GOSBnuB4fsx0eyc2rPmp6qvn{eiN?k9>WSh5-dofTA`Yi#X#D4sK=()41 zc)K`L?rYFI=#K+uZNz$*!|(dek5S<-aFCnPxu`qM^S5FkxPSJJ@oPhR_tTzG5&{bE zT2JhX*x)$GO-RB%$Y-nGpz~m=za%`PP)i6XiuTkgMG|(tpu-Qsj^UiZql#1>5*TXx z7Hs%Qu%l<~Ao&>XdDQQ~SrA}c{ma&p!r1Yg>|-_g8jUk&l;7YAN7K~`n9InfQ9?WL zicXt(GovOFay#L9R)UV%Jauq5(TQ^Y5R_3icU_G1mh<6H+AQW2 z^2Gt9XrUha$VI`BqlA06buNyTSb^>&MMw4y1?g|()+#jBniP7Q$K9pQ73g04sMcSRnm=TX%h}I{_YVQ=7O$}{F%KT++QkDqTOlu zgSPJ`?=OVK%U4+IKR=kowT3E`iuA^Q7V2^JzOM6nx~OsUZ>8}b*r};{Qd3=vo}2a> zGQ&s@3kl)j`XN0xKMs}$ndb1h&?uKU=0K(O(u;fy-d|hfYqVOlP@2?P${5@L7*-$0 zB*e&NaCKBrB!fYAYEX3SY7Ev?p=4<9ApHHkOt(@L2PBKtj&kU<=W*U|tsM~h_~<~n za%A;-;s?{d-}8W|Dq{UD|FH(Xd2sr1wlU~k4$u|i^&j8}wA!I}?t?rYM~t3JBlo<2 zfPSYRvfepKFy8~2xUcHcUB@QP0gw=u^ZrjC&ofA%E&z^4S7wzJC9oQ<6i>|IiNTA_v^Qb zbZKNoQD{=*H)4YN>(Ek^S;Sy*QnjlMebP4rd%~0DZak)JDd3UEt$VkinaMbFXe+pw z>vM_LjdwZgCWAtc;StxgJJ%7s{?FF9%;en)jBrMIMJM$7CHPNm&RUnOn|lL4C&pn% zN;AWe`S+a#sBMXfz!X6G?lsJkpbnwkY^C_WRA(EFC3bMk334* zlY0`-k33$T!1fs61+t~O45M)S?h+ybpz+8-FIKmZ9ZJY!yc`*zHxq|JvGo8XAJtPF~$b6fddVsB^mV$jpP>2kBc?*Wuc|R|78g7t(U<D9UmBU+K8>VzoZHy-2RuGydZ>cW@F%qMQRvT{l?Et zk9O;UJ!EtwmA+EDvx=;82T1lX!8tY@d-P$eSbb{#qR85PTXJ~tTk;sX z`FG&H?e*KJ_ah8&xdnhYFyWal3ao50n4QxF^jUs73Tk(vc$G{j2q6!`m2k58K#E=v zzS$tgIbT@bV5~*IGAd@w&+%^3qe>qRKao&hrxN@v8`#-*RWtq#aS{x)DX#q!mrsf& z#Tg8!c>wRoYW_za!y08g!t$ezNr`d700`Wo420A_^o&YZ47#>$7_@>_r*ThxJ-e{* zmbUVAg`q51tETNPuy^CJvKnJM=XR_!Y{h%VK3o3Vn*yL?zYm8GURnH)<1Ftt>>SJ5 z412y$pgn1*QNK3)>>xkv)fae>BWCgI=@LVdZ!@=h>hBA}hMK~w{5v7dQ#``m4x@Kg zGG@mP2fF>V!`4#C_OtRw+*3S3D_I-~hw@D`DVA@BVls&PQ4|>69^s;Al(DMM+k~&) z>O-*w$;%tn@5O8~5$#PyT+xajYQ89Atw3YRs7nXZFBN>dSE&@_NKMah$=8BmGvU!7 zorIic@7t~$62-#Z{d*2Xdg0Z-1JOG{Wi+NMgR_TDCAr0)s!iHy+N~cxH6DO}`*y>Y zal|6=+oO($bS<2RpVhZ%D{qGzs@ihz7oLmG3*Tf~ypt`pQ}(HP_bT{^C$t0ir%YD` zdL7%Xy^>ky^ausTK^KR>z@N2%B zn$vsj){&?j2iygR6cV(i?eejw$advA?ap3Spj`K(8^{$oE0Y_@u{qJtmeEY>6|%j*d_&TC zC4Z}x#zHRDw6Ss6dsF-Rv2TAQWJ#2i$tBQMX8QXQs)Sd|XnPt&x?iT|yg2UX99_I& zQgyA&T(tZ@_gBipseNBmsMne&iYdHITvBv>v-O_#^-dsh6GqmOJ%iLDADf^-o3%XUK-#%>~C6Gpa^eU%#g4{nu_$1@cqIO`dc%-1upz z?E;mq;>B&Zyex7I&yLFDJ6NGPhCDSKSbgFdy!+`9-JwFZ1LxhMm_LkN^gZ^c5lf?; zl^2dHt=UKj6j@Exzvx@@4(QW%i$d)>-0R6u%qQpqmv}!IpL7NMupvXHctBd;;!jvB zoT>@fPQ+(F!W&MbWd4P`y!qVw;H7@(&Xk8Fka1%-3;tB|ZNq7UY=g)uY2=#&&tqY# zs%zL_DfDm59D~R=P4oN#puAZeXBDUOkB^{;I+8-Bt@Eo^R0^iKPkEj6)n(17|fu7 zrt?24T7TXJ68?ofHe(U)Hf9{|Gg7^L<{9+zW>UK~7+`Cp{d6CU9xzD#;xt(7BN3eD z8MKx&BP=3)(`|ooy1Ywr{UN+MQZ2x874tZ7tC48e@bF(N<8Bl(m}fMNx44+mHL*Q*X`d z0{=>DoPR9bRK9^+7IKT?FdNL4Z{^foe{K(Viz<1;zoQXzWv=T)i*+A0250sa`!kiZ z%pmPsT&v#4Nnvr1steEE{#2|exXuvsziy)c+k*eH@4wspzv{X6rq-NRHS#qa=7jv8 zZVvqq_VsbT+&26V>pT2k9^n61MZ54t2C4on?DfmrcSlgCnLT(mcM*}Z^EGXo4X012 z>ar*I^GQ#qo^6;0CsZ8H2Y1nk&Mr<|xHIFMJ`4B#*tm0vzX&O|ap%E84}P}R$pYI0)#PrR z``AVIs0BsdZI4SB%cW_S_hALEA-7c)?ZdO3%8Dys-jRiuKG!&-xw2o+GjM%A9QFyX zTu!fIV+*=5Fu?m#rdc{{FJh;3V{1Utt#!y?@=?5P zO`FofSLTDxANLf={jQ^PVf=k2PoQj6{E;eE;mz;i8Rhy5;(qMI0n^%hS=To|eB1Nf z%NkPV|%V>6G`IJn7?pU)akQ zUrPR){Yi)2Ea$d6-wE3TUz(x#X+_tiIsKp&n=UgCH-V}7jztj*sw=*JUBJyavn56B zHyX7YzGZQIV}a52E}q50Q}ZITsSDrKYSfn$G1z3^YPZ`re3Nkt+^d1_>LVR)egC8! zv=R$uYSK9SI`1Z9b_*jEm44!K^B+w~6MkkXxtmwOAoSb29rNV5H;Pj#H&YPG^9 zg-PG^7pKbar1~4a@8h|lK`X&_v(3G(5W==?*2MEfLKeIJ!ttf4dEcNV}IB7dO2Hu1LP4BF4itWrL-{Y=ZGvU8D zZ)f%gCd&GD@4Xv>KYN;OI%WS%!Va|gXApqhEy{F)8Uy}&^L2Z{n^Vt1)!h7kafM>WS;&1(>hqyM*PN;j3Q9d(q;MOYrDt6`KaCMmC)L$W%S@|nd^K?MiH0ar`{lG5^ z-l1OxKwr}a#_2$%k38R zx>@sS-MJHJ3-hBn1~9nS=bJqaQgvU*6$Lxo=#43!sOOinn74QPqcePT_uQgx*u3Wf zpOe5`6ZyV{GrsDh{VJBWydvzfuirM*Y;ca@uVYo%ZNaH$@(d!E=EEszuf#-Z+P>@` zo*rh^hbe9bGN*lA+=#OI$)9QfF5e|O=+yw?Oh4ipC<7vG&bQO=Za!LBB_HHStG`ea z3o(@NKMJ9JJJ4SruGsJ+lI4YuNxOB!eM#^-=xb$~Gu?8~L-`Ht25sP5iVh-n+V$}6eo78u;-S5(hfGr>pgzj(<7>47i9x*#1t=8 zZg)|v{)F2yQYXJSx9t{1CbjZ@>2~csqMs4)oSq{tlFds|`hkXS`Cx24A|&o$fGsmk zRA*FFv0mh`z)0;*HW{Ojcm3f@cF_9-T0;Nr1rz|oOJ_A9A(1JCCf5;#_=>{NdCwamQk zkWdLmn%Xb?`#xu~dPJ@`p8MYdJD3ATkS#k8)Od4?>SAC#_mOq-kOq!%zS^8`r)74K zgCzw_a7OHq3+7@h*BoC|gIQ7pykfv^+*A7c>3K__p#_Uc@o;gK z#~{jEZ_ES3EGDvxRwj8k4)zGB^4|+5S|m6s$h6`-og+BQx3=(Sa@vkiHZxtAB~2v9 zYQd!^>b6JhvE;>oBB8yf~Kr}10`lYjPCQP{c z=3^u7S&MYHsQ--i7R*PEo^sy37aOjGEN)%$_#iuTAiieR#57i4V|`JaIw zjp+=WTogar*Y=--=G_9Fna0IYiEID#W`ch`R9TUi`9atYUz&_n2N&r+ryy7Jw;QJ(dPc@+QRqubW&>u3{aPMVlrQqt~rj(YgH-G$PW@!De z)8Kh&=Owh-;H$ha(I-Xq%9qR2^qlfa7>O4n#ovs0T^;wnJ}@n9P^5kl6p#!K?y0!c zE&`&O&Ho4{Z3&7Ir0r^im-iSRi~g=(eJ4w|FaCKi$8 zUEqrK=EuiT<>@0}AU6QZ)!#h^%xmkpN970rP)*OKL!whp7JR-SrX)+$_u1$M%f~;tj%f!mhZ}8g}dQfyPYm z%A*RO+j2VwgYcI#5vyFd`mS5mHOB&q7Yt@}gvtKKPv)4v22H)P3qGxTv_N<0y9E@JT!phm_P9SLXk zVg>z=IQO%mo=IAOm&f3j{Tx(0_hA@i+$jAZp@Am!#@(G(kDBgpWUgG@QXi1?J(lgJ zuGviE?a1xo21VC?k!M^Z`c%Hf-<3(0`{JATQWD%&vDUV6+}BTvJQ|-6w7WpnHK}eY z^OMUp>hg`7Z?A$*e1!BOG)ldE*IWRkI3NuhWby6HR8UgwbHo)I54~%v9a#AGzX#jo4Sv%5g*ch_mt^`mVobVDS2eB8IMbYSFB(q=IoEd1Q^!$g4CM+pmAuT8VW&RcuSzS{7B6@^8l^XG0G()m$&@1%>liCn%pM2_lM!?{ zo78NOAmWq(&qKaThsM4}(ZE@Y|`0M7y zsK4M~aL{rx|E!v)-lGa+em9{isb5WPpi31bNnE5196UOCcw2KUG0wBzGW+RF>a#{B zOLG%noFXj%z56cw?lIg6zANF_SATqUxo4@%_RQF0e)_A@o&rw*0^Lym$S@y%u>9y~ z;EunqGZWNw>%fs|VX(KSexTaqDjCcF=XjqsDz%<8I1)TSS>vMD%RU(Rxe)Y>OYU~? z=T8SU6_l3ox#*3w0~5!SQB|CfWiHBwHs^(@@7`gMlEW3ax4R)RDYOvZd;$H`HHiFT z$qQ z_W^R7s%hkX;9ggJ1bK{r#!m@xy%@3fvuP88?R+^^2{~^etqPQ`^|4bC2lAowENTE` z6U)@Y*2FmO{c1!zuokl! zu>@RejNj7fI~87=f%Luf^ajN3J1E*kZGL#Sr9*-@ zPG#f<&alwUHm@|lRU|{qej|$Vx9@yxtq>3YpBQ`3sHWPl-PeX7Dn;oaC|!D$PNYZ; zq4z392vvF~4^4XS2!tYprYOCmNDD{}f}nJycR~+Ydp-a6oHNcoo@l)syn3po7Nq^w%14o1@QWgIc^|N!u^a35wd|4gpaVA0{wPhw|Mr(j?`f; zj{($P!qpHWS8#y(3GY_1ktX5)iKC$&pzCaCIaSzrOgaY*xTt0pn6m;}43Kf1mqp?|3pg0VAH1U47< z!tlnrC-rjrY(T?KwaV+&uG5ia{^2#mxq4Fd+U~smEIbzP{uwW*S-eNl%K=fFgS%+X z^%kn*tmk<{TjN&Ei6&hwG}TusFAdSyCQLJGmT_dG(TxPTIe_2| z&eF$_|G1JwFSVbw^9!*P11nou&Oe+*P&&rcHDgDz_K|?Bn6PIn53{J;D_UZMrrN8*1X|3u==y2QN!8{yOIcD`jwkc`3ip=Z z54K~#e|9Ap0cD1sp##K=c_Far3V*iO8W)`nb|g3uVuO>~m;IG(WB>=(SCHVBYl{Qe zX-mUo2V+-R)An_?b14$hk^2Xok;JtGHJzI5Tmz|D1*~Cd!FFo~3i`?D)i>NtVE5S0 ze!nKH^E;EJ@%TU}QlRB+M*gxtM;ftE;soD6w)Mw8x5HY`>-^s8u`dL?4N~&xL8tHD;%RRuz5DA~E#2 z>X-GXJ@wOqKY8YIeWAMJ1^1|u7AaE6XWUcsoqIorXGMc$duiSjP4$XGAdrBT(7rt_ znEQ@38n08sen^$Kqrd;N1C*U8L7v?*NN6`VCBNza<|v<~PE{#FOh3fWyvpGZc19QtPBgYk;ym?q!$8Z2#utlPBo;Q4iNsiSwu4HtFSV)sK zD`u&Xk~LKd{*Z8ZWVOu#$qIWh`>HT(M?0i$E3tXSu+n(iuSJ)GU99VM&Mgbjtx~Jr zw2@;qb_$PPHTZ>&UA4L^uFq=U5i*m58%Pptxo|Q_))j1NF(9IJO+XK)y}oJX;b}Z{ ztC=IsQaZkx_)?Bx;zXHM>=8q5?5fa5DLvMeLQao|%B&nGOTM)%5>em+g{+gKdyO+s z!7^?O*tu*2r^XQ&&!@fr?Uu|B%A&Q1dPzd|R@eBuw<2`Beqvhm$Ff=nt7i5L^~RW% zqsb_3`=_Ojm}w1v*0GCK5ep^9v$&C7B6I)zaDT^43#rtAy%Gu;>d)p+O~Lg|%Mo;( zw7Y<7H;@Tev|;3%w!Tgu;&?!uMSR(r57g^&=#A}PGP=`qZLAxKmow7_J{fWI5Qeyd zjD;!vF3{Nt{~q-LIY;gRNqhf%Y;e`|6m-~{X3FA~L(<+$+zCrFpISa1EC~x4d7l`> zE_VEkIMqF)cd5pn39Y1SFVH(OudjlEUv^pMqPx zkQDDmkRo%7Y<0sdgBd82iAUhRO`hIqD9LknTFE=GEtz3JyO*1GjOLU?O|?PPkTR|RV^ypp-WJxg(>AaoJ!cP zz};p$T<7Gt>K(5Z19JrbUnwQwV~=a-lDgikv~A#8<#-KGeU&|QnUu1g$g3K6k+ zHlZNi`i*;aNg;x&(fWyZ$;K^{*-&c9GNV`M)!&QMb$EAvEp}xO$Iw_X`@+YB@AZKY z+-ue<3QF|Jl0p??dTy0kMX=OhEM)WP2Q9+H)rcL4ge*Q zHl=N_pP2P^+A8GB8>AM;Hu8F4v_g&KKGR5Z!4q_g6cL-J3K$)#DTr0rlrIi)YUr?+ zO(_avYabfGF73RlVC36cx2MukX>%QroH~$OC=;cY_z)=z?A-L<5xzeOwjt}&O52WW zR-32MeG1xvGBr|=*sZ1?;Jr=GZf;C z84NBTU0&&r`Od({-i|}i35d_SIS6?`hWdn7a@p>=hOA zqibO=f?U1$SA<0wKC|5%pf-7dwr`=~VyK^BS(g|8odmQrh))|AfQX5ODFO6T~q*J46^Y6{gy#*=1ww7j5fmSDrqA zopv6z^(t=#8DTY9re1tB{82*i@q8?H0fO9^Tj)CYRJgvN#U&ATPGfWJIS&aTbNhCH zxEn{BvkQpi&3(?O65eJCTsl`6S_J&KHcW41zP<*#4)0W*?tWn8UtN|V%3x65S0XCU z%10-;G`;3kGTE!N?iWQCF)>=@W&ug8hITSHtkGP5tVVD`I-|O`RD+qznE#`FXtO;` zSGhUuNBVcXVB&}0-goO>GFY~Z<`2-*qCb~!6hZqly93hh^K>`vQI;DNvb?3dh=lje1=yT^p;sD}!IkxSh8*20$rw9Y{ zWymPV31tnt1s7--<(5KvI&!)_XADvDW9x?|Wz=R#l zXt1MjJcAdRWLN*QeP~OAK2Q`~oBXhf3=bRY-3*6Q-7^B}H(Q zSVswRxI1mRu5};w6`2ADkbWO)eGfks70|V;jU`~#t0gf@wi<(tLnxD~U$!ZEk*WP< zx#|B)s?lm3!szTQEOcRQcKK;_@mfykqKg<2zvy)zhfH8}Sie6GX%OE)uAx_(?y(F3 zEpbhn7a*Pzu%b*xLpJX|a|zAF-z#LRE_?+*6y674F4C((MXP-{AQumcUK@2o;qxe;I}A z0sLd!N|vBt`&DXL(iqWB4^w1^%qtTX@wJUzpiVkJeaK}iCy{>``jyY2f%%{1I7yI1 zpo?%SrSg)kt!1|CtH`h0qVRnp%&EX-toz&TNbmn+MSdG)e$M*eS&>KU!k4wG!k6Fu zk>niojc?A#;94LjNc|H1|F0N0?8>s8l%e(3RGLfEKKtLiJVcmk_Lp{8M#q#f=q-5dLAP0okrleaF@Yt?t&Ro zZUUH(h?^GU^>O~wgv!*C3BmABc~STo2h`WGcWltK6U}5$>?{^eWE5`v5{LJDBiEGa z<5V+jPu$C^>#im9iKCebOTEaQ&%b68ET35-Hs1s_t1BF%|2$vQHW36 zsw*tYo&Ch*VSJaPta1Ii^Xb1+H3!`Wv0r%0SI^%i`Hn-BNl#W*uK)5YaJQc8uF-{# z_nX*Ik)`6Q5#ld6DfnQgQKtt1tEUJmbm6Qtm&9t=qu=Cq7P3#Q?T+~E?T_y3_^<=! zOJnWa5(4)xORZu%J+?$wa+*qqm(v&uyPE88Hg3;`e_{w$u{#RCE|Gzq_IQ@OS-^ReQp zvn|NRoImqXDmgir3KPcx$93A`@u|#P70$EIv_}3cC6}IMl<@TVA z_qiMWlP0RdZ0imZ{-|0VJSpPp=^4+4e#INUjOvyVdA+(qU6R2YPCvpd#GXb7+&`Sw zl=lk{O*5~%Z}!lMifo)b34bBF+KFQ#QPE=~RCet82GCypNos$2Q&v?86hCsP42SKh zVQbGJPicWCd2VdnGk@v|AF5e_iJ#|6y@);bm5|A$y%d+0lLGLW>NR*-cO#mD^^%#E z_ga|tnIULE<;*EC9t+2ulk8{N1|70Uy-IwC_g7^7dYv3UlTrP5CE<#V7Fn}01);v% z>hb2Nl*4AC_3D#tFV9Bf`+%XBwO7ovXGiN+RM;>2&u++0@+ej?0h({*hW4Gifs4Vy zUqYSN$gW}eRr4Elg%JL~ca^{fD34>)^MATMD{R@icn=UtM~9JPV5jY!K9^Nb3o z_^M5uRqU{w(3%1I+qATT(1lY?D?(jZbaPtU_sK|r4*JXQ`HAt;4x97) z*ktqJFvjgugzRINw9i-lm#*#m4KbQ+bMBj}*U8+3XmLrJIx<8(m+_fNVtCa>n1Q;B zs!^>k0w^YH;mQUbvo6UEIyz{_lMe+ub7=t&oOsa)mz)DqmDG{nT@~!9tR3n&FB^u9 zx?s=uqbWWKGLJHS>P@@MfJxzQ7gaW8qtD!K0EfP7=k?)Hh?l6nVY%tG`$*6aM80joK5v7#!DS&@;vuE<|d?Y=U*oKf1# zw;ug5Js0=O7M8YupEaMuu|o}qya4Cb1`z;+E}Chd^rg2NT@q`Uaud}^lH-dAE&_}I zJs+_T&OG_jKXw;>+3=NCS7bV|TW&I;d#6&M_^8t(@t?*(gbD{~#cDQ#^ED6p0MTQC zwLS7pU=@Cx6O{gJc0y+^2gh;iUT3yGdTZvV^hHg7hpG}k&=1%5(6hO|{py3zMM1~h z1my0?89Qx$Hi66{(b^YwrsRUje&aUlK%F|WIfsL}ub{&~dbco}ft5w*qV!XNaJ#)6 zznp&ZAmX9V{w{2TDV;wcE&lo^Pfu4ex$b2YHO)2{6YVI8T%HrB$1<4yjj}usT~ZHn zeU}`sVYFJ9);~tU+O5alC9EA8c+GTF7ohX|dDu^q0G%Z9Q5`y&3P?u)@7qw<kUh`RN$ zV3J{beH_g~cZvFb6z$;4=JZqdh^VJh%^=*Y#p~!Z4$1r|P~$PMo^$QFuMV8=tbS`l ztAbpBA)i_BCdSCs$b&(oS7&qrLVp;m%H~wMgo|}K1+hMc9%fnP(7#<4I1&V`7<)h8{E$fPUQXGsUx6JxxR~%1gyTujXe{ttjt}HOxD&Fm z{R70L3eqE15bSD9?e{bB7ZN&jePc%YxZST6X!()4Pdw!ss|*592~xyE;% zvp+M8$9!_opxY!+rLXHe{1NzTJ-9OEAU&!(XOikYEFwr1w2Mp1VaRVfoT2Z!!D@2{ z_Xtc|K9I5fokicqYX;^BT{o#s$=nZ|WcBRGpLi+M3MDkID`a};o)&)+sUdwk)m`ec z%k+YO3gA6wk^{VEsMMx-HdegnSZ!nuduxv^SM9xz7S{_($yNFW6O2yFjmLEEe^gPD zox2Frk*O}P_{Je(bvoY%G=zLk z!25T-%i@U}VopP-Q9EB6^oMfNl;$EzllWW0*8qt)z5KmK3Zs2hRIU5#X4D9n>a=|yP z6a;1VU@~D6$m|l@ z(1Ed!cP{3G)XHse2K0N~?niju!832uv~LG3VsPUYpE~!RGCf@D|%D+tBl0; zl?~l^nc!{&;HQkuA1K=NF_|F5YtG&lPEbdDxR*5D^aU@`K{;1Z#~2O&14*K4JftrB za%<=hZ(1QjAFa3>P@8PvocJRndD$1Z0Nd+j zY`8!IgMnE+J4x^l;FWitOCSI(f%JPhRT5(yKrooQgVJ@4!uGrvruiq+W9nYa>7IDt z0<;7X$|dBk?Qt12%7Br%^oq*?mvic0D@HW1OUi$tn^E*?#9uS76H#_}9$@;=8&T`zqF7v(4j3E~gJ`4KHP}{YU9=`Jmq4lOZA|u8k=g-p-g#HSx7E`^Rn>yVA^ve2ij=GUt~6IdpsyWKzi)IXWjCXVXSLuqh^Y zLmc?pZ4e~GDwscr=9lIXG_R3d-D_U3A{FmSp?(U+-kYw7*1eUYgVx z$v#GPcwE|gGMBx#zu-R;c9(R&&EwtU^=!%@Px;@k{PeI&Y7 zYW|)ffZ7<-THY=DBHCpvX;#}2zHXy)v~1w-@rYQX+Fk3%V<+K_uc1|!Qs zvh;HxaJfDvMqvSRU**@9u5CqnI|qyf)zv78d%;~+7hhz?4g+gZg=hMHv(zMucc-6U z+h}(Ou7$(f-=KIg-$X1tlP4fnbp4)VS6bRJoA{C#OfjN(k(m41W>=i6%xUDAx+eDd za&lyM5u$bXbg#LFov-%4tO*6D9HdcN($c#e+n(8XuQBT{S|x5UM7CrTS~dUI(ei)w@c(baEclIhU%>=*V*uPb&dKql z0C7}DorU95g+RinWwKrVtupt9{72Gw z&2pMl(@v^uUGBN3+$cmr4Pk-LmOJ6xUnw&r8=e#r#|W&C@OFM*RZPN@D|?7wK`erR zKth`o&b2%bBp3P+*Sy-mG?!z{+*#|cL{7WzTno>^2vRUeVKWjN;#T{GI>Op?e}x!Z zsb58=018*qZKQul^}YXzop2{9*!>3!Wt)A^?hA)hzuEtyCrIxS6w0SYz*pJgIt_|W zJR}!4G0izyP>;)h=?Q(>?c2BeOELmJ!B`%Z39@-U|C7*M@U`9X|K=x7N7lG=X2pg#XtaxdP!AzK%*S;3$azhMZy=+IYP0(yXr^y(AP?$n z1ryM1|1xdXh%@P_Fz2#v>L3}ybGygh7!=Jt%Qdn8l^@n1JaUw!AqPVMAbueS%V-xSwz9-DJjcs-oltujp(cr-fG?f$D4a^`(^k25cwCr00KNb3lTyQO|fBk zRc}QjiD0nk0xF7HdrXBbo}xE3{i|5`wLYw3H9I7~!E~1ZXh5bYxT5QA1Z^jy8AAF3 zsFmX}1j%la#FG}sM3Fma1Gy|-T!`pNk>S@|_7g~;XK;OM$Au-WnYk43U`8RE zxnj*&>O3z)3UB?)BO!sxBX%ZtI)6A+oOeI4g6k79(Y6#~SB zjuii&vN$x*>f zLP(J3>)k5FxLM%xRqtzV!2#Q$?$@HaN%v<-No|g^lEh3{t0P`X%KgSX@JUg9_Yrp1J$G_m>PFfP-O0c(p_x$_CrijCq;3YY z>MYRW5x58erNukks<1n#Okm~WDbuC>hrvzXBjij_h#-2C>HB*9D_EC^#ADA5rZa+{ zNQ#SxscO}xotqr^(Q+-JE8pMhA=L!w61K2_#?6s{t z5bv&-;;C|J4CJ`EIT~htLgDQQb7CZ(ye}3&-DXz)+a(=6MzMi>RLzFZ1Olr5`LXHn z^nzcSf1#+Xgf1t$)@qY!^!RA&>;Um#>#|AWi8*d%*45Nld(G?Xy1^w%B**9}H}KcE zu%8aftveyj>nTY>5kI^Hh~z93TURgM3NbCyXWi6I_-&k#+x(5_K}v378kUk_9a|@K zulVz77YW@I{p$-zG-73TpU!x?y#NE`bjCp6BYCCAHUnPQYaCgA7Ax@cEQ_%FfYt23 z#FDDy>egt8ID}fh25CmRmCaoxQ~#*z)IOECz5(>LpB0v^9Q1hVU=9Pu`>rGcnt1ok z)6J?8$!N)B4+mO=2R7w)$L!eE`Kr3hH0g301aIZI%r0im?I{^yxYKOcy>j}iYk1S| zIcK}e^UjHu;c!m;fj&{suSVs&r#hGR<{E0 z;d~wi93Qe!B30Ew)T@&%2vQ5?sg-ZNSH%yJB^NL?u|2_`PeSNE%!eWaT)4}QEbIPY z?T;e2gQH~+eAD-HT2{?3C*{m_lQD=q-|*_neROmAYP!LsJ`QPJfTQNFsG?TEY8J6_ z_SCCWZ%5<5P=ESkY8LX+4P=s@{(y7c`MTgJSv;HaW7DV@Ov&P+@F}?@hX`mGpnGmk zdK)yjA3a)K#ceUfGUYg{S@lK5JV3pMJLdE?86Bg@x&k3%bsB(A@{ zl@(b8q|dTZWj4YOgyc1p!*2*(4H7mQNCcCrpLa|^z(5W*t~3iBCDl@LZm6%-(WL_2 zcxL+ucZ;=jURjt|IN*H;J6ghHl?hr}-syu=%;GFlDv~V+fy1|JnfTYLyeXUYlB>}E! z3J}1PYN@w@bk8X+w1XM{xq`T6BEfEw_~-RAl?w>z35)Ip#Ix#<_%+NJ_*5&ugT-Yuor2NN$;d!G1iy z@+=N-1Z^P(91$l;=a2;U`;MSXzs8axtZ4RX73|S5a14-1<>;Q;Z3@JK(_SYyL(J0g zv=U+a;H0cUmT4Bbb1rPsOUHJdF-Th{!>c!qz{EaLy`k3(t^L6@|Zvo-|j`8ltd+(hfK)ZLxv1PSsA%by@*Is}eha504 z1h0?mt{IjH;*h;AUwpp;5??jpm2zd^M3Du^(sv+-pR+_6IC$6%q>S5D;M^7gvo4xc zunREJpW38zi4uB2T8dEfAnOURVLY--3j!6ackF(AEc;#wpQ8w>0gh7&!gC|A1`bfe7BBZC06I1msQ{|_r%qvfw+QiI_JpIqp-t#n zo)DO6W46V6ZeRvXqx|`SIt}bEZG$ZA?FEJ3-)ZMghl9Blv6@K{rTG962<-3~e6uQw zig%`LWmq?t^>6A}%JX4y)Q!y#C@vG_O@RJKD&>jNz-TA3;I!p^Pf~du~@K}{)Z6r%UZTZPbj;O zq^o%K0Zy}KLQ(mlBnwWnbY(k8Ab~Wl;da@VO_4lnM;I-@V}aMyWrD8Uv?#*ikc!|d z5!%db@G}s)-}7bw5o|Fb2QP;QS}Kn`&<)#7LT z;<|WVQ)c%;rHnJ$DgGCM>SeIjpNX%vrrqiuV+qzZZpuV4?T&SK1GXAjh`Ahh#*8M=T zGIFm|xPi$y1hHuOfhG!O8)P7~!RHS3-IB{uirHjsW%`fjcOnaDy)0B)-9fnSgO#57 zg2w;J@as2g^b~v}fo#zGqZ;DZ__KihZ{Qqh4e(($FJD(8LLLT0fprr;q3OJzV~bUO zA&lhxUgDQjopqm@AnNJhjobJ<0t}J{iLzAg6Z-p>neiRM63!8wcg_^K5K@uiY*-I{e4>YTh>6@u4JC@p#Ep+CAMCp$7@v2UM4d86JQ4C zT@<5&+T7<}yNfw@PM`>pU@N0=_-`eIEt{M6C;U|dRTVoJ{0*d_})*rDbrI_ z5)%t}Vp$uNKSFPs3YwQwykh_;1-K|0`baqLbjv@yI1qR8g5O?p>aBxCATw!~ zyf~Y0BL6V~3nb`&pd`rd3*i{l;*s9H6-o9DEN-~|B$N6k!nb|Uv=sz;=8TC#;m!go zPZxmx`9MM?eb?)5r%_fY^r_q?ucF*$2LTalx~C$gwL=fa%%GTZ_4wxI9(Umvp=4AclKxhdOmO;PyOa}r{j3X9+)*5Dn3jKIR2di zcW!gI@tq$n5+ZUyLVhsXs`lk>hcI#89vLnm5`lfsl%w=!^ltF;oW6#pS6+AUF>`Ju zsskbgM#Q12D@^8`JA0ez;OD|SpKg}Z=O8$hDv>`?ue@GW-0;elj-hhZj%4)u;x5;I$Pj}?*WhXPZlWjZ>=Tj{^ycpv! zXk8i<$^0}F&5+aAY&YiGfU0RQT|`bW34|$?9c~dTjSYy*=MNFJcG=R!y(;Na5I-1I z3^q-Hx1U{UhfiVeqD&o21;^?Qy=%2OZ0Q0^y22f1n|R#`tZ4w8y4|A;R9BKA*@0%0 zx37vJ>fW(#!X;O z&7wsMb12KC^3w~)oZH+D`R$afEc#^q6?XcZ!*sSFq%t>=E;Up89Ch`a)Jd{&a&4$3 zGB>eNQ`K-?VJoUk1}n(4exB)LdI#{kT4mjY*?KM^<(bv&D}i`ULC?MZu4sJSKjig` zNO`6gDPMmYeS_R58Er`2gH2*1w{TRp64=1^I^~<0w{&Q|!b~xqVe{f@>AS9(*#M2yU6z6vtxf72# z7cL)3nxRPU$*OD>exWjoIXmJv`tvMDNx_?APy{n@Zm$zLN`q!7xcZ{q!K7=V?0nnr zvcQ(E2=H26$+M`-o-%Pj`X;k8FtX#@#cgGCyMsAt%bV)n_)C2+ z+nbMGr=S+ zj3igGwO0OuGn;MD=1BAE$~Rj`ovhFc-ydm&GS4INb;dor^^TH#>Dw21hnjf_TA0^= zVon}k+yBB)F`g`q7EW6Hy)21PLRWgm+?||0 z4fD&CCpk(BSGZimbJPP*jPAC6nKoT(vDA23=tQDG!k%I=knY&f^7~~i5V;qhw4b^G zY3N!%BVTB17(>85eI7U2fY7t#Sjhpj86^U$8rz+gj;GZcvluriJx3F<^1b*b{#^n* zt=ACCHA-ujoYWr-&5+H;9XUpCduSiT&hhDtaKot~=za5vNOjPj%~6^^zXMP>Oof*- zQPBL|Y?C*By)kkwQxfJS9b5|~z^C*j9jNM+eJK-uRpeyi{~o3cyte&GH5cP7a<}2W z{RgS>W7MaA@)wcSl~N_-=`DCL3sqf027Q9Y?TutdhE05ftTt@>`du0veGAsU)bus9 z(Dt_H`Fh^}bS~e$DkE%*pH6FOIdNPof{Ph55cqV~&{1G<=4i(Q1zu_N-_KF1tf zASlV5EnaE0QliK-E(6E>T=8`80Nb_Pq3mssV>{}FvJhw*zdYy(nbGS@1+^(LC*^kdaz%A45kfzzS-@RfomVrin$2l^ZRk8%X{2^{`i-x+ zlL-WY1QzG`@{KGMA*0I($~4B8}>?F8A%*{o|gBdlDX*(;;#P%MEH2C^{I%@IHNnP$Api1gND-?y+Q z5ms(_X1ncize9Ma=r;T!9!nxzH>@Oz{C5@q(VvdbC9$kI*mygRPlK997c0HUP1>8}9hcKS7#-RF3?0kg2+*&siFCvk zR(qj*WZumTbz{WJg9-zEi@jumJM|(Fc_&y|LCr;kAh8Cj>?9IeV(yD#+?u5*0{(D4 zxQ_Br^X>EP4CqxN9}rpddweK=h`r)iEB$Q#ZUS)!XlDEl+kn+09d58N`&}+vQhRK} zh9Sh75JSUagM)m%vhfod$LoHzA@Cuvb)WL0nY_cTY)5C_^7t}5xkc(q zCvqdFk4WTvaxR8u@2m6FT!ncJ=g)kV*})d+>)$U--a%K=0&gDC)qe*|mVSS^DWK)h zlIp5=iew3`aR*749T|u=}f3i!9PzYz49yu1Cy)DSa#Gy z0wXK-CNFT$rLatgU0#|=^z!qs0Dre5rym;NGqu|ByTv6Hp7*8{WpTlVIp<>{6n(;^ zY0J2@v-yS2N!6Y%M)DwP3{!HYiWkIUn`1KlzCy%GF)dr0&s<&)6gx>ia2)1;g2mE= zAj!!C0&urq5gl|D5G5GO$OPmOG-_XxcESCZ%J(KC86V35aA6}8?Ku1~wuqu*qkxg) z%Tkfl$w>FApX(TAwSW=6h`f0#PhT@>#;>OyZ&R8=du5qVdN?>PBa?gMYr2k%45zdW z;3UA}Fo!9*VF9jX=GsXshr%_7nvilOs{c&OAj()9BZpOZhc`^EherUS&KMg%p8*jy zr+0ei92s~_>V>`z4HlVqF?OzZVHtQau3jQ#}=4JW)vXM|5mw62T8zYRQ5e>Ca!tk3La{W7b zoIbRIcnPv(ISpl+R$&9jAzw~oU};}WzU&L5qh#g!=e8pVVVSVFlE!v5;=uj%DIl!j z#cGI;Lg?f8Ar=(eQcnuDcU6`UMN&3;5=vu^lKu9GgIc%5+(g01*M{cvQWbgiSr)W` zR__FblD#yPgZp|mM#MXQSlMeP8IbPm;j=2?9n9uUp@`HIB=60=Ia*Qt1-po}w+3M@ zf19JEjvd{?n|e>83vd!2X^2b(Jm37ud(VY)^^1XVE3j4o45vyw`Lq|GuwJ@^a7l4y z#nrHJ&4+C=^yh>>Tphgu;D(4)b`jT4dp{m@jX>}A(p*!4cVhnXXra_f=)DG=D@%n` z#@T#~#AQ$SukT(YeUNb+pJwyt>wfFqGKj`5LMD($4w?o0R8@4C%MbXi5>7|;fo2sE zfV>CWMI>}MD{EBQ(jjjCX^w;1P^o=_n|q5k06&8&Y$FyC-VzUlP*|oii}eSs2*Mnt z*LR75crR~Mi21CD!5zE=;kxn(c4KH8@IA{-)Vn}RKgxymL2ThfKwON9A~^dt25Pe; z;wbq>dY!u+F04;L2FQiz0y zz0n$bK%6TlC(P+^m*77aar?x?KF(I{*BShx(vN({jB$MD%7`cD>kiW^cU`V-p4*pZ zP}kNj`*Y71`~>K#@MdPBoi`Vetu{OWjcKcJ zE<}Mz>3Cg1axH^_;)liEa}*!bI^Dr2kL*Z?6Ov8`x5sVs;Mx7EW&YrO3cRt$3z=ds z>z3eYqmU=>ASR)OPhaG2ENlg(v9;fg(Z*NA63BDzb%t8*3-ukWrEyxwTds3XufPxT z<8Uzr>n6IE;cF&!Q4mwjz4QL(6-+k86w#$0`p>27weyBJnHC-KkAp*FbYhewSS)y% zG3J-+iigx9)6PD-G0x|1`pat-cTc;LIbTsRb6!5F^j*s^6{+>?P z2N5$Ja4rwg+{^lA(CRD@7IsEM{#Q;*;Gcc@n$e~>Z9Kug%&u?JiEAx8H>4eLz09j5 z;XHDv9~H~;rtpH4kLRP{yXU`unXZX7Y8x2=@9SS2g+j1<=8vl=r$}BY>MVNxd@>Qf z%+G&MD#&sy@E_`;bM?wm?vWXV$<$H3cn`{%_f3|4=YPs!ryG_F^ch_S)Mc||vBCT1 z1o#|XA&1wP-g?Gi#CgzZw07rtIt*9(D7Hqqa}0-4-Ylft9mf%3R^#b}`n1Y;7q7@- z#jxcWIa*MQ){DB3_gp2oi^t4t6w?8DjQD#V<$~`5IAE^$o$QE0D{m#K1Tu=euMX=% zxd4`cP~jNa_v9z6I&Q&JxVR;@&I;LsJCUDQdJBw#R?S#Xn%ZGrtDh( zdv&X**}taiAr}pCk)K!}Y315zfN8*&^N!!D-1(teL9VwEtYq?sO=BVYT3Vi;r@lfC zQ5V8rAz8kJV>Cp)_9cJ%X^tpc>Y@-Y$}Vb`R_(B`l2>;M;tQ>$ z<{_`s5ox7Vk>2mLWnIYnXoT?a-az3#8vG+l{xIRbbF)9LCGW^$Nuf51cZ&S{+o=0# z81trgls6MU#x9I!@t!tQbV0HN-_3H89&kiWAm*QMJRo*W!`~Z2$ZqV?t>D&%D7t=w zXQJy4rkNsOa(grh+`B~WU{WVcccYCr?i?h^#H3JSkqobH!P&9M07%3K#+AI+$KNOA zpe;0Wa{mDq(|PEuB7=00(;kFQ9GdOe*E+X&?kXTBAGNFw6ATastFUmA*7 zCix{qxANzy2*WTb=?^Ed6^WEP1Pxk0IA0wkxfP_s;}4CwT`6N1_J7Rp%Jy-zoXhNS zMJywZ-SvukOoehTh;ku_f&2F4xj=O%40{2CR@$PX8XkjQ2vZG&Sn^ ziURk;?zL(x44jyW_gdXeP(E(Oj#!L4exUFX)&3nNyz$AG(%D_yS-VFC&Zeh$&oV+x zny#|6D+|#5OSo4jsm$=!(jq~LThur5!FWG!JuWQPVD;jFp+%RLP0}1MVl!XKLK~4;hr}%^w1R_4A<9&)B`3gb{-j{e^M7KWsU%8+E(kK1%56N^w!E{}{diA?n zM#y4OUTRt}>NVZfZsHfwLx#8L%~FZqei0VC_vegscQVo9N7^a)8QQu!PzR46);!y) z-yu7~eSKDyae!1XJ+&q*1v_A&RSstk=N6GJ`&vX{O-?u?s{$!asomdopdlw;*DzL}eIZ^3IG zm{GVs8z=x{r;_gbh%rO-x5~c{WQbQo5}RUx>wymvT70PMgmlmA7P}e6g%IiZ^2-le_4AU%l#H0>@*ULu zhI*ffQfha~ZNs$XXB$mOaei@S_P)W%c(Hz=Rd(oc|qQ zpqgyVIkjVU9C~IqB(y`JzOTmVSMSmcpWjVO$c+LH&H?GtN#)5Oz(ClRPt)_H6HKC& z?#X&JnTEg(3D}hJRz4{YFE%;WHl%8|F?!c76^3LqEsSRA)X@cRmGsJ>gRA0RrAKco zrP~<(r+_Z9b#u$-U{abX9q-`4AYeWw2wmi@N|g09Q~ooo(smdcDQ2LekDgraa8xU#*-3r2gy zcTGJ>xQ)66{vQf`y)VG>1bM$ z=t~kYj42AXT=IGG3$5^36!vMb>=i?=&%E`B>u4n#)5gd)U?9Tf_6@!T^V}fP%{vJ_ zvhY@%{ENjGD096tAr=#dRcMizy_{C#^}Vx`KI-UwR>Mb{I5{5jhp}3#x31dDHiZ0$ zX(gDNLZ~CzS6uZwDOUGoA^R?S;>K&2UWDqq1p<9LmwPy9IyRL$97IllTjS<|n)Qwx<9VK3uX@aBJbWzE3n!gfA{hdmLyP@ZpEI-n!^9 z(DSWA^h%lRP>PAI*goXL^|XA>Sd@cY3ij#+z>x>R1ABFjvX>a?uFq0PQDsCc!hYc` z&c_d5idE#lK*-aZ52Pp~2)9lNZG|du6$wvADP(q1gWy6o zkgIZY%HMy)_HI$46CW3GM7A20bBA!7sNQd!^kS*RO7~$^tvQUS&RuYoTe^(1_)Om> z?D=aGmA;d=(@4tr4VU;}FXFQb6a;e8;399)Onbsxw31=(PYbn^7YMf`FziJ27VJLy zE@5nL03e0wzLCF6@SYEE^&o6k(Ei$t%l69|9+<=ao{Zxl4%l0txDAQ79DQo2+s}r! zYf}Jd-jc{a9n3p&9K`u$I=$97OtKeDVR-D za4PFq1WPBlO*sTYx)nEOZFo^|K>|4-B}Z#yglKL;aV_x$(`pLm5;yM9-m)7mZ;zXY zgG2dMK&0Q{sNstclG(GE*$?I+?ST@2RwpvUnJ`e|LHe!I_Y2R1aezQtTvwi9u~rm~ z^sA_p#x7HokHIrXZ~Iu`n(vaIc;lOctFL`dOuI+Fi$dkHN~VPdI?H^^OR&sOVtyWJ zux(@_np?9q0xT{t6<(WV+Q=ACLahJz+CowL%nkeArhbvNQzw{HXjKdbLllyePk5aI z>TU-=&Sh%a6vblQ%GAM@%2dOp_kWj?aNR)%=ZM9Ap!obUc{7CdYe) zvKXJ6%TI1)xIiHKfBX@dqC!?Q4fBOJ2I)#zMYT$T{-k-@)K{3moE2Q{v6l00Fzv>8G>dMaM+9YQj4BtD0GiQ zJHDW+tuo_O-gXVBa720R#J2e_dQiNJ!OCj!!MR8C5U2 zcM6DCuO*$tV(qp@;H^dYZx1nFwxI<}pV7J;l1SQdZd?(mpY{cOEgMFg-1DiuHZ7jm zK8P-!OLvjeS#cQP{bbnt`YW~R87b*A;7N5;cNk%nxpIXY_vrU9%Q53U+?P=+p@qY+ zx$>XW1I?R>%GxU|v}M$p=Z3th-}y66Y1fn#@zUR_b-A;(IK5kt4|sn@0y*210B#cT zbJDa*r+I5fp`V5BS2bTw5JprAER@*~@-tnjzL6aKmPa(;NFpO$OJiO6xbj}zkyTp+ z^%6Hu^sKxRWpOyq_b6<03oSm!Wl&X+DRk&U*yIVx^l1$>uRl`NfWKcKIAj)@w@6pd z1Vo*Q(X$Yi&WQTV6L3HYogcW2Srxd;y1HiM;u3(V8l?1D`jY)#+-T-bcAWFJS^2FX z3fcW2lltE9#ZmEdo=@Njoxbl`xQe>`sB*^)y{>c8LIZUwKUUqwYKp&JxW!hsvdd^F z_a*S#^!c#&t;u0UBxi7nNCszMe!_59^i-OXw9|ho6#a6pO&eLJWZ#*2uOGpxnUy1i ziMXw(|H?U_p-Vr?bj&s$T|PK*gr)OO(2QxJq5JT;W;h;m2kzoZ2$dbTHi0|VVlgO- z>CJ>gT~=uSgA+*iqJo`<4nX-3S88TD0VLmcZq$^o)g&iB!N%>2U3YdGbdU7-6Lf$em^;xy8x%S`1BE+a}$RwLzD(p=WROMHNhc@Q1k%Vm8N% z-93q00g}78`W3t0D!W$7E`TL@HY zzGKqbP|pixG;@@;5Q~V}ZvL$V`+H~15v)f!{wKhshOHlHqq%x*szhVN25c1VRAx;M z=PC-Jypus@+j@({CTV{tdzQFSezxi)@O7w2wW&DcohcKB(Agp zPQy_97gET}l}v`qYj>$x(h|rPM${mRtLW;-Ol$D(2a55*M3nT%7%LPiaKv{F2vRII z5e*o>`+&QM`(ya|m6Ej6u!?$@ps#38S#@`CsV;Z(5o3frQlQRp{;`=6KfLK70`q?r zjk*&Vt;r?|QFv)`a;~6&rPG{%gkNa!2+hO5d1GGnAX6JT0Eva*-S6PK1RC z!{M#^Hs5tt$dG#|iplQM#|}We{rm;(m7I9mM>!Z9ZHRMTfDzKY0d=K*rp*@~59PQu zVECePZtlg2cfQN>eK6y5!7=%5OD>1o?o?6~R+dA%>REf+b38^0AMR&&X>`z8<61%TCG0!EGO6Xw!Vp5-cAGe_Hb=Yr>F0 zo!^zJJ``Z^ol7&m1x8T%O*t;gV)@4#w0#Zm+MV6Y5v8p7`yZ%Eu6}JmOD$JgfChRP zl6J=au-$&dp{QxNCy^)9c!?XEC@7Cb?lAx*X8er|!%C=GlKn@^2J&beJsr!#StX^!LSxk_%An^n!SH@<7rt}5i`DxgB9+Kw< zd8Z=Y^5m=I%fM16;4)@mq>s7^V6he77j#SRkGY-u^8~jp#~G57hW%-FT&Zj*l)CNq z$i5=|eIb`LpOzuYXK9;;XUqil%8dpJB`r=5qf=_7og|ST=#w{!oVkjAxINZ#C!V^_ zw(#+If4Qa-n;3{d+`hVyVUqFr>0{3CT3^>(&T`{w$4!$bpZ0LF;WuIuD}CLdvxMRE!dR@k3c4?GDD?I` zYk&mMb0jVGUvTpn9wy$R8%Hex_+T72Y+k1$S!=4{Hw!e0Tv=oVpabznSS&zzsK-wA zGc87?nQ(f7_|Ajtdi@$OpS7O< zhJa%zAQ}Sx(@;oE0(w>V{UiDQ@`EaPtB^tXn(FAis@DrSsc^QZO3$~I`C;zVO_f$K z!uY?*tN#gGGW{3g@%oho1V2J6X#>1P4~Ef$n+4OJYeE%#FW01v_W%VW=dH;PjF8^x zL@EAGg4UT9__(25dx3M^dXb`u^Hpw~+#^2{0E#71*QFr{v;U z7t-sCCawIMxYvQsg#37k)Pm{Kqs)b0wv0@2^GC$b|8CDW3#nL9IAoBu*?8{3;Nf$p zOBnJ-a1zKC4u6w?ZZ_v8tQKFjy-cx0+|%)fO}Y}MHhi{T_qRS~>6B-#ecjpjoptZ! z#)}8qc~2#i-+wpclJ0GNBF+SQ-XqfSsIVY71M1Jy;<0TPDJ9ZwjY4U}+P-f~{3*v9 zT|Zt%xDqxwLY8=NoBZs?CauRbVmnkvV^QZfcBZm{A98I@2R_K2_hYY3$BqY%6N zFOPc*r-j?~OshxFt~Snrkj`_ZL8&jCArFvmSp;=oy^Xi}6_*Z0t_<1bZC#3hn+@EZ zg=`KgUyF9NzmQFe+dHtj-B%IQn{}!5@Nlk=@`1DG>Kl2%6WW#apl()J;1We!J)nyP$UgMk{BMB@xEer%)dPpqV-jS1n{aP~sBE(Cjqi4@kaoi+o#tq#) z59(HZVCzrT8GDyU27uDDS`ltsam`Y@il%!|Izo|GNAzv#2^{};kCO4;i_zRx|B##s zb#=K%lfKX|VCN_FnnLKFTI5}^-KGcr@FR)fS${w8X@fBp1{rJIKHf9AhVJ1+nze@O&3_OP&-K1I$uk}3#GV6fXSax+%9rkIsASnKX@Z5eFGKB zh9EydpX_Mh#^0yEVsLNTzw`YZ z=e8L{rvyLb$8XrKZAMy~v^r6{PccKDplJc+<@K~;!G%VjXy9&ExzVc)H3?Y$;sAT# zFMy&9BAsa~Sl~(2Dv_N0;KZ3#K+B3eP?stmbwVR&d*&^tO(41@+t&nsz{FM5W503! zfbkf8sl&NT6q1$*?mReggi6Zq#_K@eaZ5M1nY`E6R$YRdv=X=B1Y^mKVsi4AYD<3@ zp+I8up>W1!D2;Vt7yjWS788LmwVe@H^bBZVCypfPGf+S86=41vWz8)r>G3ElSCRI| z*oT-9F*Zrvkh&AfBC35YS~?4P->)WOw$8>CI9HAvQ1&A+{g)Oze@}}n16oUY@RuR3 zTV%xxFa1>dw!>r45fK6m9flcbnvi$@DdXVIdkvN>o$f;vlTcLw-S2lb4ZQGTtEWY8 zeO>{2MlJ?ef2q}|Cruw$?K8`_^D6QHV@6{YV=P*4NvXsNKWEl%*H|!}@BL&Wry1S6 zQu^5Nl1&?+f4nm*f#*eDeTU)ll)OP^I;znazmd(k$CMa}&J$psY54Lvu5mVuk$xVH z>hK~c&+$ErAD(K&;xuE>dgDkTdgJjUZH6zRtpyr)V)|RTK^jx}O0fzfT7E1x zIawvm8nM1js`8aN-(OvlHGfdb-e0y6X$Lg_w%}%g>fsheML~4jI|h{rh3@D_{dpJk zM&q)YF%~O&B}Z95dbWsKy242cQ}Q;`1=l@~^6;4f?@epr2R|iYi>ARbPP*=Qc{G0C z;h=`K`3WpcFaT0q#(S0#ZLx)UKuOFelhZuOr8cV^}axIw#xQ#h&%!Yfu9VbzHa}imB3wV1j z*!hxOSw=|QkNI7}=-IXW?-_^$*?~7PjIsh?qtMP@s}9&>QoNXUFvuwkZoy$S9(*<%Otp5WKxKt3rrd z*f4uTAYzX6m|KhqV)jjh1ahY!IMD~xehijw)WTw`gZ9LXQG>PO=hiu=Bw)xbrw+}V zEi9cLc4a?Xx_T^GppKq~u!MqXH)F$6)G=+D=-F3|yCntAU?(uATQ3VF7z~rK*ipk* zL&ke^ZGS`raH! zT+)}`PdUt()?2)v7?fSEr{Cg@#-B zBhU%;l6*-x9J5^Vaq%R)H8FYBG2QG3dm&{2)aI&z>W4yh?%WujV_asnQ=^0Ht35VT zZ{ukOf59~~+2bd^^$M>o5fu#qr5hw4IYOrO%o~F7UZh|;H*FI3lj2?zgA7c`)FE4Z zh!~?hv)MjEv+V;nH?%O(I}Vi&xjsE7;xybPuyjc^!X@d!aW;?eml`c^YQmqs{Lb&m z^!;ZA{}V)lj_A&rq`$Yz-i7AE@rUtT7%ILRsF5D{`pnOFQ$GrtYs*LO&*oPdoLMRX zDeBFzf(K7EEv4K({T}+50-?v7#2^wF3THq#H9E_lqwhL5LbakG_L26*LPJO4&Q8%B z5An@dg8cWaOC6MAiDg8>I`x&;p^4q)md(WJ{5`5`mV|L|!pTjF)}w|ntwYHRp%)XL?O)5d#23S}z?OV!GH*$%BT`^nS!dQ!)VuLF{l z#U+j1q`n3j720zDk#!GpG7oO-#|#@=QX^g6Y3lOORUxNI|dF5kx^fT^tDgZZ?@Eb;rf^y)| znkQk%unYkc@M!t5<4=u)jg8OgHwe+KCGAol6awBtr6}T&bMt=|jEHpV`V)ozLa4M7 z_mj(^M8=bsKunVoUMD25d~{13V>8XQpZw9bqA%=ib+Rk%t(fuau=Dc)r?l7-({4E2m7U*F&-zU~%w8!tX96>bk)lmgvAfyvjcO9ecD^QTD=z-&ghNuc%IV49>$hZ@@o6 zHudIaQh-9%U}L)}6U}DMHd39JHa00vepU9z?}wa)8cmSrPRo*dU)Gj=c-z^swZQ9B zatCFZ`Yji$_rf8=Uuy-ae9OycaypmZR< z6`_cYmB}ugWD5}1cO`)9TVnHl!MP6)uH~n3ex!J2SIlMS+^+pl9HAs#-zYx$D#L!N zW64d^qT(^i_2u09-C?!9J%#`A&NncCuqB8S3&~W*HmKMBKUhEm z6R7=tj1qmOFF=oMM0|^#asAMS=8T`JAwNt!g|ESSus2z^XdfU?66e`d$`18^ifKs@0OPe9i!nC;Q9_J!76S z)Q$uqHVvr3yp9GYxrY!nHfU;5?+W~-pp?+l zzG{86rFf}iGvdMxIQ)zNoou$&FYRVoZHgvP7M~gxlKL;&%fR8ZU`EIXv%(Gq=)|)n zAQO&Cc^nTu3-WcgPBrjC8wz0bR&9LcN=4?;QqtwAYjkrgHtdD8Jsh|i>XpLMxg7zI zOJk41&KIpGMeJn_0jYCIpp8(m= z#VFBL<2`>lLW%F1yf||V&$=f?3n0{0Wh6%-5;oCcnyfKA%n>4&ciaoV?$C`28Ka;v zn3}Nyl`8=H;TT)=EXO5p9ypLEfaxk=^OO?6o0cIkgRc2f`p$l{n%!LT<;M}Vaj z7`WauiY|`TV(&HHlV*erc)NFba<2BgP5a+hTiy>h zVjp65IHvhJD6)ZH-|=PaqB_eLt73I@pvPy&4yaw0KK>@qj#egYQAyXTOvLTYrC<&GNUia6hIBdI zhb>73(+Q6J;5l}yef;t1(-)tn1B8?UiRjtf$fXi+3B#0;{nrQZRmZsv8sBxMOEb@j zvqFieSqX8|t^A1$IzbeSw!bR2JF!4(l*KmpL69=`5spMEV1yK*W)lMs@98Uk%cl+}tI*8A|b?2P@b^lt`wMT)W z0)<#Cm`%*jonGDn2l5pRT={a*k;+UE{mrLShyGVWp@nrSYKs1_-}ScdvYuN3mBA$4 z_3YXuee-Y2eJ=jiHy&nA^WYTtlYQK;0)>ei7KgbMPtK_qD%1HLX`pCi(}kdUy`#H6 z4s8A%7Q+pe`{4deF^EDIpSclkA9{8gwJIWk6WtTM)Xj`L{Qv==EY*NM*N~p@fs1|` zHUXu`mWujS%p+Kpg&DX&>#dZ4fb#eH_}V5Cs5z~h1HJgC}R~?goDZh(eL2`LgFrwPjg$qCN*#=Cd?~QJ~2_vcTMs-jWlPWs3W4Z<%(Mki2 z`zMwUZTjmrv~LGe#20W!smE{&_WO?;NM-Q}Xv zmtlV~dGgn-lc7hsieA#xp9{thhY!iA|2sqYR-sSyI_N7oQp>J!7X?-G-7c)N`|}Zh ze`DD2;%$~vVxvmCg9kx@X!tLp78SP;21BotVyb`1!LwF^3%~Z%#(}Axi2ys}bG9Ir zc&1be8A08#J4&?PmO&?PA_(a@s-TdCtLRb*hmGMy@6JBJ$+uSmR5Mj?B7wZSq<2dt zZ=UhhSQ-&>Tsc6jWP+k*c7YJVS|DLfP`=xVXGN^Hqy;SogE~#0tkbf5V36w_k$u@{ z?+jPb*lo4#qv(BCHy^DjL$0EgM5V_kaLQ`S98-}6(=dhWAeuu;PpoPOD%AR%gcm?= z^Y%I2Gkp{PARN#1fmWiS$Q({hO}$>02YlpO)=3=0bvc1P`NirSiBkrMLanqUtUEyUyZ+K3-tx@zhs`^MzXZ{W6&na$1ucI~ z7d-lJm_T!L-Ezzr9Xs<2F#YD{zLobl<&w6%&PZ~|nGjA<2q_UBWl@-){ zFRXwrQC4_io|kss&{EhDk2(9I_nq=|K3@BOu5tM1Dgsn)@;|AI_~*y`ziKA`BbRPH?ND_eVAhoU72Q{DE^reMwblceyoR6zVMQ@MGre@lWed2ZW6<+I*vn(pYRS zh%@G}A_#`S4yYMD4a3aYN1!e{4sFXESR(qBOdMqZ7NURFZVI`~DP-~ECF@8$dXJh4 zNP4fCDGS43#4U0RPmR3?P) zuG>FogQVD&CVH7NuQah<+~Q>!avw$44DNa-%Svf3R=tT;Fz(P#Cy)Y_*!nF)^Gcq82AMe8dv-Xs={$1|i zgO)om6Bj+t*XXe3L`{)PgotB9vx%-aM(W1)A6Qvtk`&Tugs8&Od3&539*Ls<-eL$W z%(VTsiA_F;pApmeZOh?N=XsM5$6>;Z0RP-s}njd*nn4vcUtb=i5A=&|71oMNqik`CDD?J2s!+X}qPSL=qsru6H zRSBf#dseVii&Ie>8v7;U)?oE#E?mCda^WXA7{9qBEDincPtC=AhmlJ!QxorZ$MHr3 zx~|tWi;#gb$a9q5QjM*1yhWFX1_Qx!_<#Qdj2SLR!w;U5NYl3z?oy$mkb`F@a3Ixk ziO>c5c0N(aOJ>wg`Vye^-xs0g$p*sS)(67oJ;e;YA|Tr&5FcO6w>5#&M#ZCnCt9x} zG#s#4GyN&+cFGZ%vrL&H!!I6?wAwy{m-j#h?IJJa^nKVhpk9C8N$)73Bo;2dJQa{ zfo^mt`58%q7fc!7nu3Yl^$(AJe|$xfcmtLw#k@UhC1SkSxL;$g4c#z<8OkBMzl&lq zzpvhQz5%+rGd%Ft_K@!2$3W2vQ1`6*@ZKvU{DxQZ!mr%^C}4doZT>)|Z$i;S0*uvK z$%}>(NP9%wL>Zy_=4O1B`s@>tbTZx5q*dgBD|OC{Be_GA${}}N28qXPB{I>@jc?b% zCn7+I*{jG%^fD;r%)?jPi8@J97IA^vGAf zKYYb#vEw8W!)4(vMR_0-RE9;p`Rz8T0)$dTBa~IqZ8Y^4uHB-JGf!^PM}!Wok#}Us z1M(3~3WY&O=>dypkI`&z7R5g;)+_qS^yjrZ=U5C63ZDgd3}v6?EousWeZ-2pNDBWz zED{Ltcy0B<>g|H~Tow6+8kjg&iD?dntv_lv8;4D6^imdckne9({f3ZBB{>58cS(B^ z$l|L^sE==CA9qdR^`cx6OsW$0-+1d%5{B_d^xlA7c)ri-qu9l}Wi_K`x$&A0G0ou` z(r9_RF<>Jyrh?cW$E4x(nV%&)VR*!bA_Ae8@Eo1a&#@7dxLgV)s^|N{$OE${@FcNC zJHrq9i^8Ih8`xFRRTo}(Q0E%qoNH4$W%p2*r}O_zjRSHB?jlXWB7iuO=42PHBYH|% z9M{v5on*OLLc`231!yNtQw81LM;&;wl`yE%1i#t~F_{#QHfGL+f!w1-#joB5Q1~mb zwmurd6bsh;LOsm|7qoMqmL%S9p-Q$cl{gf1rPQ%hVICC}%#%S(Flg}iJEm|gX?KYZ zp}T~an%0Ajho2El`pvC_!2Z~-@j}ID%!{+pu2Z2lXrOYrLB1%dw+d8>b^XZVN3ziN zj%VqDJ#nr&xRL$1ailW!7dyLBFtt&!Bh)vgZb!bk5ICFJrpw3{@#DBO9r+0(K<`bV zg$vXY;aiWYG62LP*zTuy>EDm}7mT&(D0}Eb%tmwHF2s0Gmi~R_9wBW&0ET=!Gyx@B z%q1Og2yKY`73=ryQ1cOh<3nndzycmKk}o}%)aSxF18T`*X361&s4o4cQKr0$CLSIH z{^?J_p9i0vjx3AfJ$2O9Gvs!cKKrphAOo*Pv*o(rHhz9t9P_~Hl)*`X1+`Mu-$!YC zKhyn>8=EXVXjkLInVexhIEq-mo$=dk>K1vzL@o*|jn8}*&ej2Nx-_Dzyw{^z4&7i; zr@Hk0ehrkM0S#L4@W5l+LPL2l5;k<6&OapWX)sZ(o-;9cPUpj4R)7U+XXB@TpXqCl z#7xFaffMdMr+E{M_Tq3v4#pY&G4OLFBB49s?05`L}7sC4?ye13I6FsXiQEDX+J?pRs)@zZd^7<@l zR|-4U@b*av;9h$sqj7L|3q_09nc%HtxFe?0r#TwNN=>^SJ=?yya#7W&rMx>MqGt2< zPiHAux7gDz5p&kgGC>gq5_L#aFJGJYwuP|&S}DxxPAsZX?qMGTT_gvs^%=p+Hg!J{ ztQO3?hf;~5`FV&>ZD_G{s@?rhFCa>9 z>Ji8qcGUTSH0JCF)`_~EozClXIG7A8)OetD*X?eV0sUM{v`xw|Mu>e*8a&OWnk_=i z-&)B|9I*~41Q?)v%q8fVCvb}=PIY^*bk1cZCZNSIUZT8is(n*;&gD6+@xaapJnNXr z00=%3prEKQGstUMUtAKX^Oh1fAzN%!cr&96z=l0~xt+WHM+<}QDLQI#;Ra}Jp6<-WmjQ3Z5vTxpgn7<7jELE?i1|qkIzPMl}4EoE0UC zjv@Y9<4!*}2()&4!KR|(gJrjG5 zBy1{MY1Gb(3F5)K4@LN@dn3*vO;P082T@ow@a~t1963xC$n-7jkpC8C34!Rp@vsI$ zuI?0A!{J!~+JUw6fO27%8%BM?sSbm;c7<&YpC;5DIb3A-Muf_OxE)9yOL`ETFc%MH zy?abqq0Lekbv+20@}^*;2&Ny?x-9z^rk;*kgBZv9520LJ`N&F)pd6bKH!P<|IRJ({ zeaAW;ivm=aWY*xa1@yNo-ul&czx4Y6P~kjLqP{+GM71w4S-GnY^ldLR4j8rBq$f+C zv>GO)RI{pv<)C25-@ca!nI{9)K9)ye-CI6sAq5x4C~5$~6d$AZoK3pTwT)(Ed(@IZ zOB>7jIT*tOwkON2L0@)E>Scj$6wiz_(Zl~j;Qi`x27(I{n`v{`#(AZI1|`@hW0%@! z>iKP@uQ}_<3idzS^z}W<={ao|4d{X^eRW>ccCDnLezzrkbak&J#j{S46_FxONSRHA z8P~sKbNu#q0{0+isP5oGe7$kZn;N2ga1JWn+?~R_q|Z?`XfCl-1EsSxo3}X`2rsyM z-WLK*X4T~Ea14>nIfawJX&F2^$2~Os#cW|*1?j77}{w_5JE(spuTZyR6{OG-q{?f7p`Y3UIoD%!Il;XX+L~$pKW?*oae&~y^ZnyU% z04%pn|12jWrA?pKPe~3%shC5J9%-b#;I3r8L1_hw=RW8lSS(|B8qKCUD-;g9t5=mo zr+9?qDcyY(PxBPrPQ!$Y0hs+ydh*JC7rQrX!{qQ7_bjj~FM~u3U+dV&x0Q)3V%n{* zNwzJ8A>G%%#3_}_AmfD;V~8!EafE$~M^DlXn&*4&)|waveSQ{#?neyW>i1Tq?L|4Z zE4`=01;)Qp=~K|M@?OJ3Up>jkTn9CGYQev_4;SY+uIw$09_&}1Q6q(HByPw#q}OIV zYT)v4r}{li2>SYbd$)WD>RGsS!V{(PMbHuynY`YWs7_T0$zFqF7h4)#X^n-wg`#0`O<1W9T z=h$76O}{wPwOoqDyc&eK)*!7DDfe;_C-J>UcX z=FjGf3pt(@GD!bZv$ZxjW5G*5ZGWE4o)b>%qvIZzUVfAZ%#0^6>vB=sF%;!Jt@rv< z3(jAqVtCSPg%&~k!_So@*lh^&z$#DwI!gK;AhF(ZPcLC!p{ghW6BO{Cj;k=*Vbnz< z@b||^d>quxk%DM{S!7Da7mUEERs-dY*62PwCdd1^ihAF*C&_J@O`Ke$MyvL$t9`7b zuzqHrm+#bV2ZP|B?-;6F)}Z-bj{%i#_je^$+(;f#Xdon+8!ZOIF->>Ay*?vOJNyw( zl{@SX5S)7nPNJ2MB`E)E@DU$_iRI93GWG^KvY$iwD8=M$p^LdgMLr&9&I3Vj<V zIjhQ+AkZz;qHrwAMdUmqZ?!{)WDJiU-vlq8#~pBaAN~|nX@e+SlKhP%g8N*0&__!T zE6#}@5~Wkz4?PbHJc;zvbDy4l0a&3ADvQv7kn%Pk2#)`_d*mA$$lm5&8@2>g--Q$@ zcrEF(F}{V zo5_u$wXeNvQ5(VRn;MP?aO9 z3s8vJX)x+BF1*R2ECNHaUmp-&PsM=8&R<#|cZeZ|#aIISi^+`(=RbU^>R^9?hiyL zBwj#~$#=5eX+S8RY$uLyi4*4SdKEaS6@Sj&0t*f!y^dOo!CM!~PKgV!Qmq8oWJ4U^ z2u4zC8v|SxlaKB!f@sv7{f$$)nqXI()5|d2#+U%zy(r7*k9Fhv+9k)VEMr=?B0y#Uzf_Elg+eD&O+LzU?GKE-SNx#7N`?D;c*n=irNETp_M`7B zgWA+#ASeMtST>}82cjA`zDx;xHzQwbIJ9~>@mU@gzy|G~4o!CeEd5Zgf?9lUIY-tKiMvvpK-Yk3a z424~{a0jm>Mf?L+Piiu*V-_&$$xVyHn4d-5Fo)BDeV6Ex6=Xkv@Os$%ry~#Sw_wA$ z!~*9sS?cZ(ZH(dROVn)jaBe!}n~93{6?p6Kv)P<<1&p+@AZX~D)gy85w!|x+TBA~& z%BX^Tuq{PH)tdx#C5N}Ji3<8tN2 zd8V|yaLQHzXf#$lJ3cGQjbIPUO!eMP}Cl()3RiZ zRxG$-cOLSAgmB$-{#c_FEpAEw9XfWS1soynL)or-#tffs76@{vlOk`Z58|1!hc2iI z-85#PuO%@t5pwHm_=%BEvpgPY=#glH8SzZ3r!~{olRH01X;|CS*>oBGzIuU;(VDTC zXq_7#w?Jy^xE6#MxxIaFcG3XUB9Ic>EIXPYafegPm~QCbT`G~8gF2Q zbS=%aJvejcZ#r*U-0Z$dE32!s=;7B%L#7kSe1hT^E~Za$mr9zYE~au9DNmv`6K3-d8ByZv27K z5UQ&wJRI<*w5P3{@$#h3O&XB(6PX^|y%8O1<=CQf(EH>>_S6mU6+s^QMRgy~)z@IK zpwd*a_&f^~GR;ScY4ZnC7l+>Gbc(8aZG5w=EQa0qX7P~0xwB-9nL6qgGie|$R`A5h(B;d=xKs_K zX2$5Jg}yHH&b23a6vf?tN6Ro7dhd}hIJK~zcGLObdbwC%+Q$A|P? z6Na<~CJ>^L;unZrZ&TP#BE4GwB$>>v{g!JkW(|Chq^5V%;vAPx34NU#9(T+N;5a2N zOGCGmzFO}O+j#RUN=WgorKTZ6M)eys7ujMliVMp}8~TZ`oAV%A6f}j1{?vc3uX|DcZr>HX)r8`=r!Tv{c+dG^tx1!h2JO!J=2}RV*9Idkhw-B`XScPA= zP&^+XME}lFAW*EKCm|&hwcgf{Q)g82xJ@yy#%9s}OZBpA)c)j0}k{a&1mX%cd)+ z=_T`?nm|QhNWc*tEMEDDmkmy7YCWz5{ z4I%1?5`riZEqd=Hh!Ua=(W5hlnSJ-~|LWXx{^xw|=iV3hjWKKQ*~U{=`F_`0Pvzc* zrGT4!`*9D&kZqF9k8ApYJ<1f$ci_DmYN^pRd$4G?w#0x(KsbXxKDtbl@Z;q}Yfn5+ z1*r9a;EG$;hUPh30ecj3>qb0PZ09wIWKrZbNQMmEm9(xI4u2rnuuSFBTzb1%SIpW| z=aF5Tmvu<@`tAfJob z8nWj;<%*2Hkwe=us^i&J@Iz}+oWx3G4|7eETmhvG!ybA~QO=ECtcCq9ozP zb*LprOBoz>{577qgD0IG`Yw7pp18WYdy(`=|8oT-CF@BtaR*&0(=3`gYK5I~-CUyf z`pM7wY37GJ{5F#~b8uiMPn}mi7tx|yY&Fk@{+df15xLA4JFj9?HvbIniQ%4% zn1oD0S2;u4N8#-%A$BSjGzasC$BAsjG9`Rl#R&M=0*IYhV6Cegu$ktY2Mga!QTkK}EfN(>~@V zz1Qo)dC{sd6YP*xA7p9+8`u*2N;9zc_8uyVS14LyUld_B5_#%poN>f@e4x`5Yj>mSzYRN&Iih@b+;c6Z_}#7~ zBf8AJb|qenk3VSsZ&EDr2t2!+S_mP|PF@cSdik;Wl;o-3I6m$;>u#Yy61wQFjL5&6 zpC<*PlELG8C~deAdRR3pPT=)?di@K#qR$(CsHdb5(bGpL#te<%qjcC#_3+`*u2(P_ z)=7bD+X0E)VVx0>s!yHJe9MoSADaY{D3z&P9xC30DpW ztFD=5qML&3y30Z~gpxj3M<1uKZD~PPk>3XP+66r_S|x7EYz*W=`XSF!FvIW{iCNi( z6Zd>K+C2I)jh_3$ef3I#TbF^fKt^oKSs1+Mk2S4Z-q+i>Y&WFE^9qW8jOkOiwU>2D z*Vyew(NPPfZeKllJDCSy3J^`dw+5LM9`a|BF%BUB<- zcdUx3bdBrTla#sEkeR{;>^(|R6v{YEVkwTvMZg8^qf2Hq<+X>u@ohl=Nc_{lXjR** z5?acej5M7PGZYD&Tdjwsv*#LQ)6X6DPZEB1;iS;g)P+C^U((~-k+}M2Rh8e0!a5S4 z*d}c;tn8~}%bd@TW`1_*98I!0R(>uDdw%U?#zOmV_0r&BQCRWm!x>eHT4TFtg)Ugc z8l4ibqc{YiloyuhNKhjzUY|63fg%+baSh6}L{`ysHB2BN6|sHf#WA*`vw-DlA7u35q>1{Mi+T%V;MZAm4Iy23`1X zzyF9iNe`RNn1Y>wijqL{!MTk8UUE78g;3;O6PC3R*M%?y+KK(%Mi;zYXcBD71#fq` zIwXZIqW$jG4N%IaPqEDTM!(0=E4^?@oMZK34O59ipoz$*H={E&9y~d?eMc=^#TgERlu1o`aaS0aM zdKPQQkO}ETj+#O0J>iHmcssc%nQN_mY&`K(|1ZE0n~;oC>lDIDhOQKN_`Dsdh`To$ zl!#D7AQ(F7&>at=T(D(Dr!_A~a0}gD$QC9Pl520E|3utz@}2e>=h=~_ z*M0JC*x<0tQzLVfNnBCYw*#kb81@wdrCTsse0!;OKSw4 z8E(#LqB9!oCaGgR0CU-cWq|v@3+ZO=;#kFIXo;%s1iL}i1UvcDy4{d_YQnzu0u^&W zbdbxfH&5oeW+l2<0{jpAd@h7dpGl;)A3gLQ?B>q-e)v*JJw!a9Z*qJIv`TBarOItxNC+l zFgdizpudi-wD%Fxudhkq_e*s{r;L9-3Qsz&B;#t%ye)RimQvo86|@m2^hws6pBs-!kbgs`@JBRg)3Cm z;!6BE@7%|dCDZ#{UkE4@H(j`#y|b0~q8rv0_wO$uzuV0m-W2NG7&RP8zL@VDog#G5 z%>83ci1JRAv%jY~;96#F4g4ta`qj>aVAsD(! zwtcI{&)8%!h%O`h>0^Im>{9NgbeQ%lmY5}Dh|_cyZK>0)_zb-@hSy?JArT(XX6r1M zPc?zDeQ7fH%^?ZFoa3;8c;dp>t0e!bY)+jF=dI5E5h$FUER^1c^%is(u#BvLD{{@Z zD35O#T$ASc2nu$}Z&P@e$33zvA7T|Aqc{iN(-1M0I~QW?;E$})fe#jFDcRoM@BX(2 zUwcoHL_;aB@GPENo<%e`KBc%QY>_bh&ru4CXq$s};T9^29zxp<1lH828mA19>;FZS zPyGP;mxE%|y)_O1C;u?+LLtNbZx2}NE=9ic)IDJ7U%t~?_i|8COx-Rq$Bce)fJ421 z*)N^BHI`l7%Z`whncV*>-?w1l#ZMElH#L`mot=AE&$1T&EMz3J+v zkw^C*?AmPMO|&wL26wW|n*lRB^l_e_8Z?ej&rcCY;iPZhw0VU|;CWEGTk=QIsICDw zE+g@{Y|E@cavR>53a)FI6a3A6fR`0K&r0hk*FyDAotwj-V&PY^cs#sumYMyCy1i?WM46Bc+lO*XZG0oxaz{b=XQpGF=8n8 z;|~~_nN=dwqZE9TnNEVwf`Z5T7YL3z4tmUBZ+?kz1f9Uwew%dnoYz3CzsZzh<{BYV%~*?5 zgcmYaPWX%DOJ=-l*rMgSQTCoqeenUDl;M@~7uxF(TYN4HuA|vDatpV@WxCXv!6jl_>Ky(Axit0aActq9-KFpLZzVGVL6$4H@~OImamM*e)J)0Eb2=)5Lw z;f<>DxoD1fo6P!@bpUy_^df|kEt>bS

~u3L%(UY!=gNIqScWz_hwgY&rK6{=&oM zv8?NtC%(~AR&sSZ<{Y<8|01Y#sfgZRA^S|e1JlXTTODS=cRgqQv5BJ(Yc1=o~({r6rVj->#UHXYrnMnJlEU$z}B zoy8xT}E`EPDS18wi?nzmSt%)Wk zcpV^SEnSRjNC>8?Mkp-$Bl|Oz2#KQZBtSoS7F~`wP;Xg#+2%DT zP}ramEIX9hf84`*EZqWcJL(HzzH#G+PyPaUL&olScEfi9=&lyh1RIGJYK3Rn*0UgzWs+1)vx z75LDXo{-ec%=6%MtKw zEfjkId=JB644rEo-ZO{`JD=R*n$Bw-9(?1QL-i(*J)ptp^Wbk^|K5x8%=Ova?twoW zh-6*rK);-msNz<)WRs-SFShLuErp40C_3uuItkh=%Rj>AnIZcD%|7Vdv&r#MlAPZx z=AZSZKS9g~#I2JkaFQ(si`no5Z>g)Djv7tkad<+zJ%rpo^=wiUo`9VVcKA2%d&kRr zKTtwykMlylMDK|S+C8Ym)f)a&MuNh)J(|EQcDWYk1v)+50P}qBI@^dHr*?^B_ z3hO>3gvdwoifX%r>kSpMcj{+O-x5#lR*A$oaO*x)1tQJk&PKIzY>A<89&G0BV|oT| zUoHBZw;$e>N_10T)0J!kGN*2dPi0(_h|%6lRc>4GkS!yXE!|mlikZUAi;o@-qvs^EtQ} z9Gvw__)(cTwB&1m)r?qOMBXPOB2QP`&4|&zuJo-;oK7;rnH1Ws$|}| zyQ~4Ja>Ij=VQycAQ8H%8gEy-$f7JR(T*J5^Q=lH8G?+$9mmsE|vJc`@j{{;VW)8K% z^2=3lJ#{Q$7mm8ba`!8co4plE#a(dJA$2UF7~X+=t-mJ2fO1kF_vy;uSFQ(V)Gvr{{*liEKlKb``8?k2 zsV@qzx=^?@A$Egy;WNU%f>NdchQo%fsZ*@H9AVH@WvRn`8nrJnA@Ii`E4MT+&|Ys7 zKU!J@U5Fe07X>twFMp$EuM00n2sk%NT)`^WsAF*6F{aCmk|FhBXZoMs$uP?kvKsr} z{K{;|y6CxfWQQb<4sf4ZP~pA|k5I|+o2F|0UHDRkc;~a&_^hWl@pOW4Rv56lk5$F# zKc%osUi`Nr9U*5s1>--8k;$jh<@OcmISf{z=jlrd#cY*B^f~4_1>9U@4^&GH2XC+1T7h->+kDNAq^%6||*(NE2;jiY`)b{x6^(A9@ylN#$GASZWMi$1t;N$po3W^k`xxfg;Vk>}SXDV*4NC87J0<(q9& z%%06?=c)@GA=}pGB3s@kiigwvg=?THkOhVjA<5Gj$IdHR8udFwC?bug-U#g_Q*&4Q(eucWf_T*v@THyz&WP<;Oe9Y<3VUoZjg56fdZ?p0I}R zF2Q^d+a|sQLs!Sxtu@lcM+8K*L6QAse_l$NZFAy|vQxY5w4L#eZ!gil+D58jc<)df zHye_$?r;iZ!-~_(h-%$3KiT1Vqsa+l-;8E3Nj^ELQ{SDrymLg^88OMd1a1F?w@zVy zWaaBz+GQQ#3AbP<-LDE-#P5SV|Epj}>G$c)s-`B^E_6GjY1tJ|taLxm(v$eQo;TyG zDpyivijs0;2e3$a_b%)_Zh@lhdNvq3BA5_DZUU-?`zjcMVfhDeZv`U7>r#xDOsE;v-Ns@3J|Da~HYFhvn$iE)ztT}=I2WoB zaCoJ#;(Zf9&Ok7IT5Y9zebI*czPt^&A@<1x~bEz zeTqF60Mvayt=jOc#qa$}rN=-4&k%012d}o;wRORZ7T`&#@JM+b)2x~c@PbzAOuyir zsgKi=ifrbs8CV^lp!={`!n&sJ%f*9{kp#Cf%ogkuspGe3Vut!ZuaY=rD=fvT5m@RH z(sw|_v($#SH#~PUC!`*A(0vRWXaiJT!e>}sG^M4H|q8 z8*CL-2vswjHQmZdB3BOd&TA??exDbCmkea__SKsu6l(+u!TuPtf>)T&FI``6y`_Ic zN%w`enb`fsqm=VW!nc%Dpl>hD7g;V@9xbPI-gWQ4CYT*Y;^Ro>++1UtcwsrEd42-|Q~oU+n+;dHc^~{y%UJ|L08)I+5m@ z#1S&OgI(OGo{za)xP9|ue@wkLm@d>yabv}cZz&dg9mHib$ciLw@Qmn%us*2@>qe z>usOyJqq8w;d91O7eIw|vyJ&{7sZpOUwd%_v$q+wrmx(Jko;V_nkA$@!)-!F{EyN; zfr<=8+hVWS%&WvsKZBC!j`j_{`ys7c(w!i<_p^O<`9cC^QatF>ta7g8kW1f-=&nK} zUKdwXJ+mtucs6NnMUEBk3ww{3HMZl}UpB|g#T9{IRTj+VeAHr?vq~e~KCtcPU~jP2 zNry|J+Swi|^b;yq0crue@P{JVxrD*;+D5}{S^<{{II*|16*?SXEzp=aUbjmXLjEob zH)!Q6#^2?irKn(dvr4bDR9|0O6<#Q&gI_G4Nv?F~OuLJhRN#IZ&j$`XiyXcD-R?}0 z>rZV?c^2Tow9qnA*JjEx(yu(v?%t2XBGt+y*roKQ!-_it^}qZqNL6l3f#V&+-!iz) z|GSdYRI2cvtjgizLNc$NWd9)_aK%xJJS@UB$oCW_U-k<>wRYS3o!XT5%k=Zy$1~7-&mnJr_P7|?)r;;rTW;eJVrJii>xIo_n!yMS7f^8zJKKg%eLGOR9!}_ zJ8XsVje+$)*B3*r9JZr&hK9fgdA84y>{OE?5uV42yU1XHf6Jp6dY~2n8~l$0=B_?& zej8b2#ywJD4%a-A1E z&uj3r7)ouqsKajFODXR!oYM#<2TvM8%Y=k_*GdogJq+0(=hScJ!yZk+t~1s}&Qpla zQ=t5OLyzLM{w{Y|0R*_90v>O1{>LRdXz7XY%umHiA7Jw`x^GtBl)Y<-(*EmFj!@!X z9Iw4e$h&!FF>K|u5sAAZC?s`dtwuDMr~Hm4p*hTDUUlOIG~@mSl@oo7VAB5kTk4xH zHla7z3Iij)Kt$qpx`L~Jz*NtF#bS|9o2^$2Z5?F_c+xg9KJ{{+KK^k=U2r{Z43NDg zV^PzaexQ~^#muiX(=@HQ0^d^6RhfR^!|1YoNS?t{PN9K~=)}VrSmY#k_@BwQzkpA0 zh=uCwHRsp!?r>_o;I=>7ivLqVY!B?c?m!)U8os`#Rivx)h|O-Zqk!2R?+^ za!bGRJFhwI{bg?m&<(BnS0K)1@Y^2)uk^YVqW=OnFTzKv5LA;r^nDm7hv~bw9kJrj z&%_>o)N<}>Za9EX(}8OZNW=+!(}~4Adt9|HuS?(;qeLucPcpi=>6 zlQ5TiNfdndGTCxWv=$DjO)C(T-$R-@a<*Nk?u^%ytbEW z?WX2uI+Es!2x1qe=g*qVB&=n*XZi6FZ#t`JwOja+?>=y|OZXy^qH%dXJFzfxA22S~ zto3_J$VzfH~H#@g$z+HAKzy%vx}^9!YN ze;3K{HHlC6=qL8U((BN1Fp0#J(OAyE{RiQG{36s@F9nsUbd~dC)^bXms zM8NsIb8KrMb0$vqH1fSIIKKhH{bpuRb~xT#5-+hS$;5C}DA7HY50vaMf_otz{IJoH z(-YIe35jY%zO?HtVO<%Pe#SCV4_4a=W8i7n^eyDNLZJMqgs=+nqp?frcMA?Jg@hS5(3^r0c|j?)v7_BRhuy#K4{vU3dmi@%y`y9ZN- z&Q(~}NwjedSC-6vJ5!~bINXQM4u1kLmk;RD6M+uoEHkXelw|f!g+4BYb$NdOxB%>- z&&b>f5i~hLEb%jvY!d5ISl{X77ff7XxC}gHeI=}o8SL--F!wM-bZ~v_Z=9I7+t)rc z_hy~nMx$5Fs_6?no6hu6l5*kTp4O;c`>5kw%Rc{e-nUqHu30QI~ZAUO`s>-ml~YrQ=${kXwA$uw!$(o26^3qmNf%2thE$ zr1;jp#E8=vNICUBd3K_8@gPO&wjcQpaz{PaG49ySdd9kB!DkAFT>GRD&pTkBcD(3* zG?bkS={#JXXB7F*Xkl%fZtR5$6NxhFLZOTX0|X;xO5oVq;XC|qg#Rt%2uMPmc}71p z>9lI#H>m%z`QdeDhdC`^dI?&T{`+kLj7~QEg1ej%x%Zk)Ucq@rna_#$kYvBJ*WXRLG*9g;|~o~$$tQ_!K|Km^_? z1hdBV+znT)po6e2VI%&7UB1Weh5_H%p(@DACB}xXTipju?s%>VC-6G^WLDTLY~{!I zTo2z@+0iQ|w%%N`FsUP@Ct?Q>hy-zf&e%7H9_&j^j|94`Z@th)+(f2WgSz3MjdA5b zZI=A`#>PP5dgd@mPmBGsPgE~-d5_ULS?jMVS{WN}mdCArAc*&yuV8pCm(Y9w&i0eB zAv{2<+2Q2J&A)GQO;np}L&+`=<65EhtEbi~%1pU#Mz5A1R#Qt)VoV)AY3ctXk%csG?v z&0FAvru>Zl-IK+^s9(P7AZ8C)QFCFq-s4@_qCSvhp5dnaYs`M&$0_xv2P!h)pXms3 z`)-*r##w>_d-slMa|;}qWsG^x+ISZBv6rB`2jisNZ?M(qC6YZU7@AP8L@c=SNzqS* zn8aG0www+M$If_GYyW*T3qu(Vt};!(^ezxQn)!$rQ}q0rNCf3OOHm?5@E6%rLdJf; z^oO{{>7XAzA_UU3)9={-cmlkVau{WSni0-9ene=%Our^m_$YjlioecV(= znT8z0;48@F+t(y8{pVj34{dsDPTx7`Ts+)GTK)@id{02Wi!?S?gSfB#>_(rBbmJxm zSKg^#7dgmJa8@OFzkhAB?BPZg-@mg>$T$m&5LrhE+hI|9_Me@}ITPLV=kL3ciw0_i z`L)xv32gw)s>_&MkAZe}sPwP=#U4;qpN79r>ppy|qU>N_+bdt>#|&ArKF%Z{{A}ML zBl+`ofwJW9&w$l$*!aAF;~*E_EKII)d2n(X)c z>E2#|&-?xFL^Y+PON@%EMY~19Z4Y-d>-RFd_MvE7a=AK&!jV)4L~R{_O0N`{2O<5& z@NJ_L#P7UaKY@DO=arsWi?DR!6|))G!&NOEJg$h|!?L`Jfx6Itf2k9U8njaShf5Gt zO^xTVr8x%IkCZTjH$J0kI?m1Hqh{MkB5wB-!_n6BCj!I5^n4|I@L*Kq|v! z%Z247T`uRD&h+PKu7i*UWXi_3K}6Owd}9MqE_`G691!Aunwr{;GDK1Zb_2tALg;{r zhxJvhSnD*$)qUWB@2J|=iuYxAowUh+s$-WUiWVJCI&v9srv*4*v(2?w!d5R4< zFgSc35L57Z68TZ7O$*G<+tP>Li@3M(xz#nBd2DDk zsqBx29_Q8r;;nQ|IuB6Z%#b#}*|V=e{kvUL6~N$=Wi)__ZVh;!+Wm@X1f%-!UPS+I zH5327^*+k3%XPE7uTfr}6+4^+&eDud)+Y4w8)kX>1J}|oy|su}JEF13{PyTj(6M~> zn-cE=A&}V>GbeL_Fh#Bu4X#rRrjxKGpz_6O!WV$gv(!xYW?`+E)#(H*@>PmAJT0oWmm90o!U*3Sy->ZuHk53!J%)_&;q!`ux_D&DV994iKa4)yw5YnAdn@Q`*$Yd z-d(BQ$X6i>6doDA*G9wo#jjvX&@g@KU*>?bFnzy))`nYPL$~Oi4yA=DiO*{{KYqI4 zY@bkBnq9|q(U}l^92BPKy~kd3q|Qd(?91oM>X8km4@;G?ks`to z+0rk=KUa}Y(nHZ!`UN9e!#7c8Y)iLqgFyG{l){%1glpn?sIBn?d5gU@2hy67$zV9R z{|Tl7M1?r{gRQ{`cCSl0G_^ZH(?HJJV6`PIt$W2A`Rif)+2jM_J5fM{+ug(E1~WGH zc=JDLXVlzcIaZB2g&90p@hDmok(d-mKvOfQ3Uu^7g;Ms@aYbbLqa1~}7`Ftuf&{>| zIZP~2S3s;e8i|vM1g0qqGZM`5wh}{KFeJ7RR6sexw9%@vSH@=RB{qVZE<>RTaS>uP z-r_*O0D?`o$i!^Gl+QY8sz)(t-9!zslmybfRmL z46Nf(^=(T__B|_B#F1haYBppQZ}%}kGaqm>^3nXqFOnWTD{Nk>vJ*C9uo#oYWUV#I zDta}hka&-M2Da$4hVa9<`Ghw4)AgeCvz)cPL$rSv@RL?s@GLv{*N);iBVyUX5t)i% z%9*!uOBqamTqmu|&Re(lvlRdLlW)9z)O*yOC~q~Pi7-yX#k{S|xbLs-Ez7t5^k%%& z;kOe1aS;pq5!LsTYll#H@9|hXdV-C7dy|oawL0YBeOY-fd%&R;u-kG`Z2BR)^G`_`AnH0z`#-&ge$;F#=drwTKAU)S(i2vJ!B!p!F=%$#d} zT0QIqHp11PK@ls>81vvvQ2TF%DZuhYG=5Tskp>AmBV8$2r024j@~vx_47qnhCDq{i z2GJ-EK}-JFv!Z_;@7pHkPS%*yZqtifg9KDil{3LJYG$@`QKFpV_go+s$CkqP1sBEn zh%$87rkGs5OF?5J;+qrEeXjj6-gkg+tUWzbuKK824lWk6H5F$59*%O2r0ks{983&6 zOCc>Hg#=%0bYycx2wcPH4Qjl0MZHGV#IS>j8<{^j4edTktY`jof18)49BlN(fx!V^ zRB&(s_&K(1D#vST_9oDT)rc{z8?Gg0XJ6}+onWp;Jhw_^<6!UgY$XkRGnsQ=e24P_ zaq=O82--|MS1804A@$ikA*~zfYc+fyDouV#n{3vV6fEp=E*k2AzJGzh6!?WvEEl=A z#S)E;W;p+ut-L&s1neS7;s1%e*AA`jM%b_o3J#tSHt9&%;T!3o zVKg{Q=FhZ^#qGMGW`l!GGbMo~mV0@2_;!E4D-lbjYd(0HhFG0yWlW%WxS);|B!uXc zq2<Fe0099x()rgD+ab<&!cOOCZlfLQQ%v?MC&KLa9&~xhLv?b_I z)LAkTlDuKIA)34=Tx^)(FeF?|C~_preZy%rEeiQ)tL*{deh=EfY!k;=M_}tR?Hh^( zByh>ilov0@ix`xZu)Z@@9YqcjLnK1`K*pX@N#O)wkkAwFf2H(ga>(|5UIqC6&0u<3 zx4Pe%*~S|+vfUf%4#9g7YGl=iZ(e?Gb?2jj!o94qFW}*z4`fAl=UZ56_qHBl)nR4# z;VDcxIWz7lc(1GCW6kpQg*m#*b}>g`2?%8f`(xd@kB!4 zzgC1h=ODt0L8bZ9+=EnzuN65G=K!)`+ChTqy>$Y}nCfr?$i`Aw2z@1>RkqzS>vB;R zN$n<@JZAm0Gi3WOe2Mn%4RQic<}*(HMn8-UyyoAR<0uJLRVAK7MHTNLRepVqrm()5 z{?POe`2#i`|LyhN2+6nd+%A{mqgYWQgIAf|KB3qAy-5<2%-~a5iuEf}~MY>}g3k=q~4Ck(zD7^xVF0 zqpL)BkPz-8$a@zyvG-$J4G^E_hQAGm^UfzB{*jTm&s51ReV%;TtmXC`WH~@YS>g?n zh}~S7uvGyiKXxc(mMnPHq*1Fw1_{LsQgh?oi%2J|yn+*_10b^gRi_lpM7#;t5_@jT zZw2LCqAyG)-Th*V6(1xChd`{Rd6W`SMl<{U(Hi~7R-Nxvwfj7G%Lr3?fR@mAV}mYi z$ZABmIlz3WX<}Y8pp;ZoP~Xgmaf&4r{5f#EHXg4p>dv;ZqVYrPZ;Y-7y;gFy>@B?- zF^mNSVvMed1d3HcEGN+a`{}7X|9W@`N-DDD?l9 zX`(yY1k2yB`Oh`Gq4@}LMYU%S> znDx$Gc5-GDbMYa%|85>E4GXvAxIo~bEp@p68}r1r0O$1(#CspI4*v%h%7$f7s?@O-ljvv}^qXOAr_Q{Wdc&)@$Aa7WWredIoz| zb*6 z=jO+|MuyPA+`ibS;e)h6#o-;};B-iwT@tYjmVu0u@0*oQW=*zN_w0Yr99RKG?^$;Yn=J>9g9FM@KLDEY>VrN?{3qPhP+ zvP}q~8E2ROq+!)2%OWm3Qs!;ELEfoRR1Y^BtQ#=X4(`URMr1m%(vYo6aH(-y^Vd87 zx^R<<#J8ewViA;5FG>(}^cso=wv!n>?DR;gPPfQ%4kkK7pd{7(HU9Zz@%5AD9lYFG zXjsatrS?mEYdJ|syYOgV>|P?tytml8cVD{{>K#Zecju6k!8ab**PbHLmX2sxxijc6kl}=e z$!4-auAg2riDrDr)=DD&OGhFF)k9F#eC|TU{)dbDFMo6?P<$SQVJ_d&%vXv-=imJ` zYgHqDd!#ET28|{At%51*k2Y(YX2M$dzP9h6GI%~n6UAK2>s=W~jpd;Ha zqdaMT)_cm2aLmG-J|#A^INpQ@clEfhZOUFp zykG3$b*x->CI_;Mn5@r0;zTN(w<$`r0#!8d2Y<3h8#ifS{6huE&@ot@@Vh|xuXz(J z09VoFus=ZEQ%iWuvRHX+?D4f??X7C*KD3`SbX^foJfyj@r#>|gZh!rEhP+iLwP`IG z%+~y^pEM3yg_lfk**1CrQkIC77%cMn$L63;cxPycZ;oK*2lm2s( z9(<2E4)Z-?j-;P0pz`%K_FIBBjzv@#z*u5?tjU!e(GJ*grASG-jGNW+uPSqR9&wfz zHn22_y0ZRcoKSbXfb(Jq`_~!VTdE=1?e#6f!3v7Y3qVUN1?S<)-&M0bbvWAvA~R-3 z@?S&q&W{_;-%koqW_hZfsAJben?t4AEV}_wnuZa-ZZvzh8-KFflJAr;Q{qYp{h7{- z0Z&?vScT>B_K2B*A_6Hq9=mV9o51)eCTF6cPJ;qC^qW~;CZEfQaH&}y3P;l> z6#dd$WLon4zetvRI5qP0zepDF+>IwhrUpV2Bh!ZPJSVnEu{l=y=di%%oZjV}cRht7R*c$$LJhTlWqq{tBSjt-mTTKc)NgrB_P~ zQnbT~-J&dEk@^_LJ#NcxoqU5#3!|~d{r2{%e`NKGcxJj%&V%Y~(~Pa3!{~n7Q^d;~ z=@V+uxF?zJNtFBr3_^%Pj}ZRy_E;kU`7=^QoB?oa!T7Ou`|BAtYu~C9tK$<&d~IQW zDG5L`#+D7E8CGL>@Y-SEkltAC-0A)674i0wZ`AnbDJi^j&C2IPCV?1`CVpiM0;xAt zdFqwksGP9N?!XI##VV>DSV3$E=;ElCIsz*!YNME8#h&M;nR*D8LR<3XmtI1vlZ1Xn zHG@BiekFF`Zva9jjGcvuM=dsr`Ru@Uk4L}>f}^4w;WcXV$LCVOZ0|4sp91r5)lSV4 z0c)4XJoT|eeQt)qTMG5Avom??+eCJQicVNUQ$xVj05Ss`%>A=>6?XpC;3V^O-Hc8B z%DE^vbSMWN>i&oB=5&JCro!7dd%Fw6fwtuDL!ZB@dP@mbv(y;VlyQ)hI)xeuB?tPy zrM}!eXje9o(roVC$K)Cr$F>FC2{_y(x`|zG5&nX9i+sIGFL54$Nw1H?JfkMTN!>d- zrQKw&3*Du>kuO0!`Xj4}adNVJziuag`sfwOZYg$pte}zIWGJ^rYIN|sdo10SUNqq@ z=Q2QtirLGttSoNTNO>KMank2lF4|@{1Wqo$t^LbC*MG&k)XwxuTB^^#?h(!qIeO2w zQtn&<%x2sB+Leg3+&|`iXlda8?CRFSWWn~&$9?OsZhIT}&*g{&90eJEk@62ndHZqj zg$IlL?R@OF`%!D>mYdNpn~W7Y)cliZCdblJz#=~4|Crjh2yKXz>_k2OEB zT#jq*r87jiE?sqlIqj9&+elYgF1{O1@p;3$@hq3Ktz<&(BZ(`N#rN(@ZkkJ0BZ%rM z?}5mwpW2a(cnF$J6-yYhnSC5YA~~$;ctdst*D$~)oKMzizkVc%Er!v~1H*Bh4wCds zA$^knJPl~Q!8OZG7~C@OMPmAly7|vvju;WlLgJ!Y;>w$u3-q%$W@^I&PYDk<1MVB2F!CD9fweGusQZjZ_H-RP zeK%mL(Ib^jFe^zy`H2d~$)t504(dDkGb6lPzCx>w9>02~m8W4k1>PyDF0BOCM9qUQ zs5xTo@TqSiN!HD}TJDTl2o;K>m~w13&AOhy)pQhNJQDIS{Dzs6Xwy;?vITMq*S~l->}Ax9A+6B~i>8)Ae5octdXRuo6FZIZ zxa9hdGk28cgeT))8*UY3!g56WQCKl+ySQ%w$gF64MT5>bS^Ay@DHgf6uW>Oxn6xIG^MFw)S z54$!lU^$`Xe%B%l&CO>!F zv7EB~s&h4>oh{)e5GXrafv4O`2c}yt)G~!FgzoM#GTa5nuYpvn;c0$3&hHNQ>pqhT z1be<_!~ezGdj>TbwEd%WqzOm|=}MK}q=t@y3Wx|wQ&2$ZA`qkmq!^H*f`uZXgCZaz zz4szTdhfj_Kq!IaKI=2*KXX33Gw*Ze%$zSeSx7?m-rRfbuU%J6Asl}Kp=2J%F!L3U z8qq&dVLOA(++rvPkKkgTuny3gl;jE0QuADQ38Mf$;?+jl9K@<+V_d#3MF!y;h-3A! zZn@)m?M2Gj4=cyW@Tq3+;@Ir+73)cHcYUX{+_uO+HF4vN39 z?T$+OS3NN-{EkdPKXE5CQOAd``mg}OYco5TptN(8pKg51U8O0_bj#&1Z`ca+h}JU1 zJnBHqCDkhTaREwC1W*hW|1gaEBUK+(N+B4~m2HQXji0`qoUzw}aU0=U6^jMeMJageYJVt5y+cfBQ9_HxKP_n(W8$z7uI8 z*XgARH05M}%`zert*{NX_T(RdD|z#Mq2wvbau<`08q1J6`ZbbJMXu&=Cj_1Z9W#%^~8&hK7S34t~+)E z%nE%DtB500D9?!ClzZxH)iaMM`T&_NohB$9!6gMKy`;}zuzDww!j6op{;gMP zFfP|;0_VIi;dYO2nKvqpn}aUxCU~O4&#p`hFKAF5-~2CzD5oiDk8ewjCf4a5A%25CJ@cSOx3GDK+_tL275CQ4X|I60L6YA$2hEkmqr}5CsZHMLDPUYv3fo|k z+xrztj-~IdJ*&E@G+kRESh@F{G_V%sROu0{aOCl!<-+ctUHI`2){3(dt#;43q5&0^ zjFuomzOQbGozXqjY=m}4uDM^DKjr25u}^ABv1G%f=aQUF4U%jX^hug@%Zo=IeR2p) zoBTnAWL43mA1khdmUX)N={oPhTUwC>KJx zXlmC!vk*H>>DEhvF3nG3_UKn(*#0lVQWPhlDNkQuMy;payDWf-;-zgosVm_y)U)>W z^+w0#)kHRu^GRazady$hctkb6psKPEVt+vU^8ru09D>eF9D>WpP3HyiE}a@L7Wo%(~05?rel)ex<=ICq+#;P zcjHbsyj12Ez4U5b`wv<5qs1Ir1GZsYH*txCyVfihS4~bA6ndd_s7Li9sVnu^N>YQ& zE^2w;!(nwACSy^tgQ1T%g6qsO%jxiZ;I{%J8{a0%=p_Fs)Um#SVjG@g=>$99T=kbz zBlaX8Odgm+DE$g7?X&=YVX2lRdhp8BtD$Jc<6n!5m&=hPuF;E&wM7=qTGT&fn!`h$ zOAME8VZzE(hDn@Px%c`nl^f*VB@XN*XGTuO%kVpusI(rwhn609Eko}Jr-4?dF4Xc5 zCNjf6?r%&V^-GzCIL=}o*$u)fd&{=RW#>N#n$Th69e(dr&$8X8%&+u^!tf~W_+9qL z%4ab8ySFYs%FJJS(JiX}c@E!AoT#<(+0%G+pRh~8t$U>lZ_Ns7 z+APS%?ZZJB`rVLT>BKABP~bOoPfY_Pee!YhT`WmVo+nW(8gY17Hh%<+APoO`gF%yj znt)5IYDAjAzJG=ML{z%&UOc-bOZ4JVP*8?FlqXoEetby872UqwNvsHc;lEEW&5Ch( z6@ohAqK5Kl8qXw2V(2pkaPQtutyMh{*|bylQHygT6gA&Id&VZW_bT(s0?`tJ@_2(- zaYHT1g1N%y$ByC_Zw*Pk$P2Mjs6{vwkBY23Uc^~+_X!^^ep5R*P9uCRBXS7o5V~=v z7X-&qvWKPIhX&b7@27|~u1bsXu;KDd%>RN_xGS1T@OgOFbiL8TG2=NnMM-|pWcHj3 zeHPJ^-V}uQ`p7D`r=k141zVg!b;1t#?)9B341&Db>f#I>q49nVh7R(T#wn*YWbrM8 z{!zkzHo^NQw|gOuPm+i9r!WPBdZJ@;8vIb6Y<@t!Ae6U8WTM>Ie{gZO_42&MQ8)3q ze?`ozH;2X#m|Qc68vI8@z2A|XYn*BrfD$FdZlvB^YXvvfrzrm=@tn6fx=lIg{t!$g z>bq8Tw;aOnhzIPi3-xclRgPsOHJluprY>kOOK(Fltp=BOXUV56i2e55t~c&ErZO&G zZM(l@5{8Q=qukIE15eHwym(o5>Ujg#dv+0vYT%I04*G~kB z5Oh_mOU(0mO+i$s@-j}WZ)XniiCnLOv1s$c5^MA)PHELV0~ zsh_Mq%gdmdxSiNImGoS`a*roiM*+D|@1k}T7~`<&QRp>)@DN&KBv#ypXsutoD?KHf zb_zelgB|qaLOsh0W@E&YxGAXaj##+CLC>6T%sf@4lxlcQKEg(moK6EQX1xn(wotQ? zX4si#>;UOGYf3s#OHhF+6J`99$~{_QMX++l-h6&Sw@dn^nDV4d&rNXCZlh8x3O8ms z+2%CdfcCF++;%rSKG4K7S<|GF#(X6?zeju=be8zyL~dR~dOOKB=W^p^X={xtACt$H zkH{*l0d5^jIjQ({)9wJRDQ5}&=1<7GI@Un#dH!&FaMk<8QqC~6R|kU^h$fMov+r&l zzR@`FyHusFGA(@AZ%NNAY?o>|+Q}a2@{aXUjIk5|oXXqsK$YVSTC0GIJ?yXynRL$8 zzNf$2c=U-tFGiVsYSjIBr+IQt=Rwfw-gU%JNk=G;m=OMkefN(+r7OjB2XXt!;diCp ztdS66!V7YNYk;Hcjhj1F()-XHz*yAyj>cKY9VfizdZ%2wMw4nTt|rHYt3$q zXRwq2^6Td)>z50lAaP(S(%Od9EWf5}PI5<{`PO_nJxlZV2OAPt|MZ?G8HPWvEM8$0 zj=Fxc^`^j`OI{z43A(BO)zCB)raP?^*%*Hg^sI~T%inH_geuhW!5=jlBLsrJox)x% zrszQP7P@6j&0bO&CtI0>(+L;tfmJ z9%oN60K0}e?>?fDRLv-MLd#3a`_`yTh9LjY4J!;?mB1}RJY%ybodA^AcUUWu+R$&- zE&d5hPd0-4w77fm{K*BYNmrMPG@HD>kHQIExkK*FP!p*+xxqAzt(3S`5hnU7o|GE_ zebTS~YU6~onG6hY=iF%T0PoK+Y>k~5SdgBz{Mi=QDf<&Fhb@$joJqwtdp%E_q)B7N zp%fR?B4)cNoU=C}1oT% z@0GPPp(yHs3lm}9Z_Hr z+?z#P3Pi>=kLRs|9^1WMerjJ5H|-`coouf1ZI)cIQ!=?WENZfCb!7R+ES6&_OSX4o zr1{nJ9VxK`Hn_hAs_8=2K~(g00X@zu3pOh9o4lk?#K)s0oEkpZRGdV=FZWh2Gjoqu z9o17^Ddt<8_DDleLY)tdy6)%40!>7fy%IB6G8w9|FPF9yS+V_`0uz&IanE-;3&1{4 z=7KlOMi6Aiz2w>jXEc7q`L(A3j@0&R;h$g~t*o+{iC-sZu{r3u;#)(k)ZYBI;xNpJ z;}a9hLJk0%TwgG}`GM%mTY7(SKFUzdSa2p-uPgz5LhxycB(Md#zM3||E_9cfiQg`X z>ettpw@+c;&_8D1;Ps^0=<<@%Uq7|S7kozHQF_*pn1~oif_Y`@(u@<1ES5Ee#MUfl zxB1luB~-^Tm_EUo-Qw#j$;cMAQpnr7p%loyE@YH_#G_ikXGY7>Ku!_$*?xKN%UMis z(kX=Dsm@@tMN!uSF+%%@@9HnFZ2FU5o3PY9rZzpz`jn#^>>Hf>)0Nz|;gHNiRQ! z)~?Oczty-3tq~N+AG$e0!f}F_afqX+u!sYg9zOy?Ur}3b6jsrfY+HmVq;0NC=Ua5< z3DC#?q4V4O{ulPtbnw~}0m<*b9J1XmV{W}hiigJ^D^7hurnrS)=Xs`D{GEzt|Hjlm zv8?{&C)yBd;4n37ond5DD$U=*jbXjj27TYq;1g&nFdq4csLHu$B^I}Aw}u1wRoD~A zsUPlW40Mu4{<%1dCn{U0wtI6eb){63CvX^fql(!TXy{{-w~$MQm3#~*$#rS19Opo_e){OX~0&k*zT+ZGDEIV zpTPo37uqg@$c*LDXE$Z2+wdPit09nAm4tNBl#JzDR6$4R2)fO9g+8mR?=aQ*qA^%b z^Iz-<37aHJMCaFm%er<>aV0Qf)K67WM4`dczHym3)ie)lI%G?KUQC+F3*p zzNwJ))s8|OM(zA(uo~6eQfTQ@h3!jEi9xEJBjX@rZd8?Eqs-vwuehknQ$h~6uJwDZ zmlIipQwd>$$*1Dl7yb_tqLAW~g!xDT;Z%dkqvf~|g&j=d14ST{;AfKn5km`-DkKD% zR&&yi29p%5@h^Fv8DWbltoE$i%|_+kFYT_%)plTuSGY+;cHKYznp1#uyc*g}pue;7 zg1Z;h+?f5GprOQzpDfTYKlkhByi$haKiUZWtgPDPT~O5to2LKRd3-Fke^DTu7A$cl zb#@Ng)%WEnPX2S2bNUW^nq)9*!*k}?-)=P#EWcw7CUc)qPKlqOC;*w9c9Yb{Kih2w zWBqFFeL2-hPmFgj6ppzuMJlI}9)6qWSt(QUf)fIZm!}i*5}j#e8U`;C#H)v+|uSRgy1|fBn*XlDsI0iW}RQ z-!`MpBIvf0Ou($XFAtqiyBZ-Zs&;eXCbjWh_=n~M=Co||mH%ppfXYB?Vx$A!6HZU8 z1TBRkOjsuO=1k$dHIxX#TV7MRH`;)^8Mg+8^zB@pBuP?}1*@GtTc*(&oBh;0_5B0n zAV5tY!w{V4z0sFKpo>^rB2aCjju+0x{LZ(@+;@oC)V6qMN%vxnMNy*SEhQ zZZ5ERxMjT0NGnl&NV@Xt3&8E58w={y4oGa|x+V)xc6G~eGL0=~_0gP2ZGzs9Kb!-< z>q0Wqd93b9l@d;+m~G!Kj+WV=JF%;5lq^`|)eILXqFL2!(9piwP5YIJsI}Lw(9)Y# z>*^^sf|H1v&RCfHqmm#EHi|s!VBu8njX>(Mq_p!nQMI-q3)-*d(S9)Fr#N1k_XLg; zUeQ{yutW!=&6H+qTAKhD>OIop6?HD4rI>qLb?yJv?EV+mU>3SfZS1(vbxkcPLbw$D zg!R9^bpOSR|L3)iVILzb95Ym8DpXD3pG~yJ*gwafqj({6R@;L)+5%N*MUUB5by%^j zk{T~nibmT-e%@*_Uw3NZCK!((K1Pi3@MK?cS9#GQENU7gQu=e-GmPD^bg=G4JMxy6c<;Y-tbs}u5Y2Bv@{WwqDRwR1cbOY= zzX7-p+1U--ueRBzcPlKXB{K<_U%|;tbHBS8!tmQs%_nDpvAB-%S{a^PxRTUVS?-@E z(jVD~%!kEKEApPc%$uWwYYxNIxqnQDh@bel7BCTE`$atpsdpG%`KkNsDh*txVd0{R zmXnewYTbpExL&6%#xs}Or3r*4`bb;{*<*|j;9aiL@6?s|otB!k`jZ5TsxG?=)+JIcLBsajsqo3H`xPm`Waw90zOBGXUY zk$Aic^}Wk*Id`J`Y{Yr<-ZctM{16M`Z3|#4pT<2F*+4aysQlGuww>Er(X!diq{ipb zhlAqAk$g%+8@e~hukjs`gf=YC(tHA~Ej(mK@&e@YNQ}Dm!V$v0a2tojjlJUS_*#`( z@b})^HW!dtahJ|(sE=UW9cMrUjxiwi?ma9M?QS=Wx&U6Bd0y=K71M}z@e7v;#xI2+ z%IVB1g1IJi@aonAcWX@>r91e>zE0;Jg+2T1McK<%j8;AJKy^aAxSr0;kb`o!=On`!s1!<)O zu;0xvdWj4K&r%XVct)rz13XHB-p&htITL;NQu*Q$>6hj+4dEGqO%{NPW0M&ldf~T5 zl>?qB;GVIs$w$(UN

W3N*VAgqcds~-~ zpLR1D58TrRoH)8fZ7bz7^xFS6teS z{oOn|HGld^L6aENql4fkKf`~oQlGc6g&s%f$wo!JY4Sc#^w?WnKTJiQjY?+?{wf<} z<{wJQMD+vyz13BWnkLsB!eUi%!ci1j$9bq5OcLM(!H54?O|)F1s5zHOvX)yu_A=FF z+ji;+fdHlDtm!OFQI4ra>nUmfS&qKRG@6EYS)&+#s-rOHxiTBTE$yeTt1T)TW88ZS zF)MS(r^50^s3*pX5tSWx5;KDSbMccpjxYI1Rs5ML7Gx7T`7$Te@XI@_8@e`z75f&( zqhQa(s$CVm4{k*MFI7N7G5($)t=-i)YsP4PNQK(Xfk3Ng@c4((44s_1q2U1HyG zOkJ>+aAOTou*{G;H95o&b?-4*P0^W!FF03};TDR%*KatF7{;}qiL0k3wRx$bKSHqC zy^Ki{yXk&MHV;*p|4PG|)Y&IKm2%rRpQyjwXEmf={EgyjftFfx@z=NE^m&3bKvHj* zFSD8s)1}&RHGp#!9dk5AVUfIR%Jl8LIU9pz*WlL5ApG;%^q1vQHn@56%)!Q&SzI81 zqGhUMlSjAZC!%~MN}n-?Z6iT{t4DwJUH%%yz02K&Dq=K8$Oo4H~+$@J){z3-JAPq9NKw1cIx z2*mvm<=-H(?b*d-3c{0Nqv@Sh*@iQ4RrDdRS*%^s*>?>$uIV4mPJ4-2$@C zG{csSLluLezK=tF51jj7o_(`DU;J>sLJ1Lz`{x60E-_kmEK5^xTc#~qb67KP2>XRC z&#KRoUZ4HCIQ{7V?;T{1;m_u*iTj+1hn&%rj8Wfh*g%LW)sC5MGvW9%A!g6zi|?1P zo=aSN>d;JG4^+)RCAGp~&313U3c?&jE9i>;6yw6>X zE;y()CXNRdI=Qfv8w-P1dbnJ0&o4N)95B=GI1a9?$zs-@zuzSF-(b!B(MMQS9$bG+ z)@6O!-G_gTyhB83s-rM$La_wrGrq*xQEqj*$Vs_~jJ=4l!-?PJFcTn1KafhMf)h;| zYZ=iG2k7#Z-+$hr{(${gxo`QA`=ltBz%%d-^bD1$MK%W$?uQTh{su~r+Ja5g^y25zE3N;Pis9-8;I{u zTZkvIhQHfAun>{w3hU>i+~-{7=aU(jWA(G%)^o~B1l0~zp(5fg7opmRW&?}>)MDGB z0Z=STi8Q4h@enKu;UHq=U5N++HiN-5GEb3?v_>Ut|iMKNj=F3b|6*I@eZ9Rr)<46uLLoYfT87@59-;x3*hd zUl7QYdUy7_L(x>yIePbw2UFQ1p_KZ6olh34K91z*Ke%3OcKM=E89ciG+Mmjme$~k! z1{`np+_rZcdfj~mZJ+=}KBo6zBv_|(0m5gsMilAO$xNOvUG1SL3e^Ua*k3vs(L5;C zUu^My`i0te*dh{kqA+Q?Ite_*LFRY(KhZ@cL;i4~O8H-C;jlxH9qXi~Oz6U-(ca*q z%1}(v4CRqQ1sz9whh=1j$O*=03Bk~I$@h(G6|?Wro_{#Xz3PBCu5EFpno(U{)yCziQP)SZjqh;3$wMLuetsoFe;J8rVSk1==CZ$DSeQX}dnHyArJ7CvP{10`nOms1zx$wh!kJ@fk0!o$hZhPpwK?%N|5V zE`FST_f6(?ip>;giFZawB+oG}UX^SIXe1-gznVrv$+Cy$q{}!07?wc%`P1A}&hZPAau3>4z_H^)B>v+QFi+Zw<3R}Zj@zaQ= z1)WTrj~UG#Fcb^CmH_M5G%sXA%i^Rt&3#LIO_Lws8Qf!Kfn9Lekd>QtYSzP(}?UT3K#>K2yn)Rz{vBKZnct>_cCL`I#;m<-gINzsEkeS30&EapSn5A-uw95 zC#1;lzXWNISCJ^k*EXgp3e!wzIf4^Im_D*(26MdE_?22VxW}GrHuxc+UP%ylpTomd z)=*G2wqK8a#=v7#MB+3)tbas_65Z%(sn-qjY{ikFIOl8%S4m&Gu?#ea{oCLJ;CtSPl8!E-O5@P^y|Tsu$T-a-44=xi zXsy0g$f5(1tj1uVcB!d6xyt^_XWcp3<;1X)V6*E4wr9ggqO=_-yL%s*f#^zKj#zDw zhyjtF`{=8Xh967b{JOtw3vHed*?=r5lU=Gn+0l*CRyi*tpzQL85q2@V0IC%Z0V# zHkhc9x@AIL)rt*c4L|3l^;F7PhkW)}7G0JGgE64Xt&B27KS4yz4XN07>F+6o*5idm+tLhMSkmE4=)D z_!@XVh`*5^x%pck_T?Pf33dUpv|4k1|E=e%?TdicA$>4=)JOYBKCP)f2HwCRhYdQ6 z?*Yw~Ujor=uOd}Y523cDb#`|0d-NX5m@h+SMPj^-mo9F5{*?U+PjlG=HQdyiyF0pz zMCrdCC`G7jHJyD7PFaP^t(4nRHaK4R?=9af)d!w)CjZ_j#H-OBH#n#f)i@ItS3K_O z3ahvqRafEA(Us*O!1*~qHmKsVrskH-%o&gnX4(+Ea*sw zwsne?DXY@xx4EuJT5Gn{Xu@>I=`8nhSLi-l8IH}mEK%k=(VOeAsY}5#5;}dhi0cL; zr$`}t?KK{=ss_!1Q=35K#Umdv5(o!utQt~tfE-)2- zKiMdm9Uin7qR;V;M0Pfq)7gLL!PFg%aLYbIk$jiJknbT)D~tXwo-gZkZpgO2y#ZR# zCszcM+ZgqaA-$);eW-`WBvt=Wa-h^c?@zpR^09=-*0nXav9@Y^_CKdEA?kY?kGa$j zGj_?Pa7BhNb+&*6O|GXCt(o?vmg(ziucuQ(?2eU@4>$FhPiGd29c#SbX|v_3RA8xR%)@V77fzY38pNymkV#7LMA|AdLOeq0zw{Q&jjz63$x%e(mqFB+1}tczVKyDEh{Y?{A}i`$d+9U#9h}*pMWTh1N}h) zDS?ch$kz2g)jctPl7GiluKB{R4k^K))?hY!BXm6uw3j%3x*Sj%8b3NCKeSd!k0AI# zsN|Lkjkg4#D+<8Xm~zZLN$(UQjWBMuQM$J=bc)rfP5fV?if;u5(6@Rp$@n%In-EHQ zpS*+(YQk=nz61`ik_U;skpV820!5Aj#Vs^Rg4CpBf@Ca$z*o`?4!-Tw0hsZ=_Y62F zV$M*b;6VxCg&_pR9)e;Bpuj>58ACp5g_xj*7`j7{?IBo~Qu22p_Qau%sG-tYL2{_1 z)NxQ^KM>px>a`N;dlc%;80I$`>SYmz!x0+M8fJIIFM?3^2O9y{ej$lk;R){HDQV$p zjN$T^P`zzD_c~Q#Np#;v2f|&4ohye3YD!`lxauyVnmDqSKe8SbT!RW$?nO@N`)D#7 zNxT!;HA>#gA2t4mznu}LPaHDd8a25RHSG?;+=j}kt1>_x3+tewhLP~v=#bLr&AaF= z;uzvQPytW0x0;3p7kYy<@aZV(;x1-<1@h65BoL$?)~Io>LQYj@L89Um0g3^bV!*4> z%C*R;#&4BPV~6hez4N0CR-&-SqHx-xu(d%shB5R*PHiBtVP5n`Yc%*ie)BFK9tYZ& zQN>e_4&Xw!_KTVf3o%2Doo0+h6_DygQCLXBn0Aih(VtX zNgO2PSX2}gJQR<#_&5cUr?o)yVF`Uj31r$ydIE9u;qiy1$UQzlDO zChhEf5E7)@vx$NX5wG3Ue~c*k*5 z7V(B2lD;M`wKy>}_o1N{Nm|?SAH$QGJe)~XMW;{`QMOY)T4bocPYkM8UUE)6l~C9* zdL7N{@r61`FFf5hJ##xO>Ce#@;h_vArj)?Kw5ZkKQh{U*l5{y2a-{pv=9T2{E9v9M z(*5Z%Wgy@@EhWJnsIf=4SdFs2Px*})RdGsJxI6MY;DhCn5P6QJcFtmSZb!Etr$Q{_ z0B4y!?X=J_+*78P1V?G(N5E{oDP~+$Kuk5yqAaD~(T&nF&FXJGpJ{)BsX zhm=OP035-k_8a{Lyg2(Vi4(oBR;{!)zHmDXO1+IuP!5_cixy|b?Y1ncUz7Ya4ojA0 zHa&@vG6W_(^M1+{C&*?~p%YJ8hKZmj%#wViV8$If;m&wn@X8}4HLY}{EXtZRggqk{ z-9_Op%&P`eH`;a`)`ffeRbJ|DjTyxpqUa zykMseeIofqM-@F{f{VGMe){ffAZf{MP%JSm*(GxW2(2_^LGo?nhejPffr^Gn;#v*a zvPpr2-0@nB$XXQLg#43wzKYkyvOvj6Y=!Yw6v0>g#y|z z>`6gdMu1pNGZrdFCvE!nmxzn3RRbfAM<`l4B&W;zL5~+dD*U~bkJ@W27uE9NzUZF@J zNN*$n$OB;VUL?Itr*jX?A8yG`pfBI^(WIE`d+4;4egfIR8*_+UIhXFxMzz@KD`#2y>J}CNTP+V_N z6hP&i0~C4zg;_wp7f>W0B2a=7%tBQ4hBU2*)U1c(IY8W<1Ej2YhMj#Pmnf#*!)95- z7M;UZ8^fe=pnu6*9Fzt=;`(O9-Fw7SZ^VHd#QSDAqLr+ssCsxIsSp5qVVL2F&e4dC z(WuAKn99-c?SXL77iI2a*zxS(ZK#1gvPSRN$ICGT*mxf6c+s2juhvi`7+_|YSO%&r z@@gb0#wpjOM^-MoUlDx6RB{_3BbWX!k`ul}>U%^Qw4NNXo*c@WoFt#@>IECuBG0{< zn%A3J@Sa-iME?XwhS{?KYUr%mQpd~4$0LIJMN%OiZT0bLIReN(AejF>>`6ZJmUZU6 z;g}l+&R|39pT*W^J@t8^*%LHp-;N-5T}i!K9MNs0fvCy<3VRMeqLg&t;=fDz={oD6 z+{N!vCp?$L?=TvvH$$&K6OW0yrzASq%nj~CW__Q+sqm2tq5rLy%-@To~q z0i9UlwR)1snB>AetrJ%$|=QGsw#PP#af#0zIW40C1_Hj`?zn%1>?Svi`WJ<5EJOGcv&l34o5 z`-W^}MO_jGkIzYaP!?j6`Z+NnEQ#J5HK~%l-fFYb=Z(RZr@D->0mgSU)1R$S_(r8E zT_fc(mObEHMe?I)d0v0!o6Woo#|uO!?npU--zF>o(Uk{m&zf7W?6FOTowHIt70`WF z;3*@~3>eMnT6WK#XNlfmG-6yiuu?@)jW6K&NahqxOF%}m5FN4rwb6l>ESPs}DSW|E zVw;3fF5(F9sIbiqSY!0O-5}t4(}1>wI_*GNV)ORNX>FM7JKG0W{ncMxyWZ!^!j8LI zXh4X~@IKa@uPAtXmo$)kRabm9_Hk9zU|yW^dvxV|q&~iT%AQ&iOjMiBtqs=BD)nAB zt;lN~*LJc!dcX4t$?bCezF-Fi^7+}ac?J4Wp0vA2F%uT|?-Pu%xa9>b%;oIOSH=g^ zjk6}Rvjg%!Mj!v>cfR&0I8%q-TZcqn7%347bY|-r)aQ7#Sl_S6zmFMWCkc=Jk>88V z!jv+YJvxT%f3telU!o#nTK=|q5h#EhBl0!a=b~`iqQhrtZvmTq$7_1u{~3kg!^(g) z3L9)gl;~h{Vol7A+?8FCJ{t9A8{2OidwhDgcXnAyHCv&dH@$}v(wOcA%OlPX_OeC;uO z{kU`tR{9=Mde~rwF}Mz!xEI2t^WD-5r_!EmGjA~=h}ko-sO7hB4yCtXCeNhUBFek^ zN2ZU5OgcXZHZNThw&bRcOnx1Sv(F?Vp0Rf+k9lh@dh?F2tLAX#Ry=pF3N9qXqD3SX zWVa7b99b&)jn*|O=Ugv;_9*@6*aBPb&7c2(UHC`P8rQK_p@|c_zJ)Fzidq<97PM)*S`3D zWi#&;dmK^un>AX3S#d)h=Oo&xA7yLDEq%d0=}%1zyngEsjP1x;f=Lh=!>j$`0!h zGdEiLmnhimlQlSLA~%P>>>Irc&gG>ueTOB)9WjuPD>~gUos?-+Vbe?;2yiyTfnf}rge%D) zYv(!o4?&)hB@{cmXnrb>G~FpV{uYKQbT9|6>|I|feDD= znrp2r(0P$2F=|9h-CS$ZP{2m8=g7nO^06?QQ&ziSY6X}ro#}WyLmLpjR|}gR_XNWc zpiW$UHRVon(?7}pIE;Jl^{bb7stR0K$Z?T(5CWy((r-=+K?~wxYG5T7Jf(MH*^P6O zOv%o3+BP;x91J5rT#6ayRAU93g4mMlb#Vm=^G%3!nzc=G)d_WiE|hOlK2>!M5Pr6s zyGgNbIls~U+_8U?;x*nR9WJq;Km&UTMgxbzx06%1JbrZXp?*O^r9pFxc?#AEIty<$ z1ZT1F(!di;iH$yE3E#v7<`T5%*^OZ>r+gfK8cMO|qwkFo@wtDuC-U6-$s|e9Abl@M z@x$uA2x{EQMv!%zfO(L8h}N1Zo|)>Wd?9*Ii^51LoljoR+U>_)XKf?dr6noth?BDX z;WbLPXOJLmUiE|LyIOq{$Ht-=VL_XMbJRAI7RitYo3F0OacB3?@osJXhd-sl0#h;FPr$kUT-$uJFd0ftO9SW?^ zquoiaF5!4Sr7UZY2>pqo43OUq|2$)Pv&8g0TYEipB>eJQEW@Vze0VD6Z5z3C$Wn-=NV8m)3WbH?yH45Mx*{zXT28T$r?k zF>Z3J1U>IuxMGkoelfo!v&LM6=AbcQQ>!Gq>s;hV$S2|aAZ8rF0`kU$-_k}T%s7+Vlh1&{>GzZ{gtclLrFaj0hAr^$#W}DU+HQ}p>;>vRq zlT{_t6wj=oN^@*4W=%9%+C&1I{$HAji=083WmV)%cR_T;&U=jF+T)A~_({2G!RDpG zY?<5KdU3TEO?_2aA-0M5oE*#VW_aee8-w*`aZ+0&3#$+8m&srRV;vida<=lRn={%i za$Fw79y~3&KWjDlq|kdE7BRSg_oH?EUewiUT~FPdPw6mDeB~pj{*j#mvlUDk=at-O zcyWXbfTU{eXT0M@eIDr$+IF7b^7C1h zLbGkt#Fq6cf{oufMB4IwW2T)VRFd9&v(=V@%Qff@NC=?a$@AZd-`WlJH)Nl}@N?WvChumCg8XaeZl(~Yc z>?i)1^*Mq&4-Hj}2P738XDLTUGt|u6P)Iu+q9#PGe#zUIJjG`-a&QRdMy_>Y^HkG`z?KW^TYeR>sV6v_{J|DW8vnyrriKNj^G`uh?Lz7Ih8e=q9Y zeI2$N!Rn}?_dngdpG-10{&n+urhbZL=tA9#lZd$a^sk$@$X1#?Ju40(dY={X!Pr?k z2^=pTBBN$D7N5kw${Cjij(_i$K36O`ObD^AH_Noc6&;Rsm}Y{J6ajD;gDZMcp^m~D zF-}g9m8C^~7^P=vj)Cj(+$&OXW+gE6xQwAV3fI2uE0XrOWirSDy|=U|#phSO`6GQzn8&jEw-EDC0YJ4zd5 zDVHj%Y==Ehvx3Yfs+#L!+D&VbUNcsT_j;z~HOw`$u;q7#I9~+o@II7h-gS>38n>Nf zvefv6IGgkeATv{U{G{Ko?RVc?mx-3oPJ-DyV;#bg_xn)+;1 zv9T)`G5)X?QBm1)f$oq=Z)aj3jM`E0yOC!-%?5z^vNX0xD0KE*@A10yB*E=K>8v)vYGE)w)htPB15j1wJ}Wos+2wiG0a90 zRki+h5;Sn!^Eb^|I2ix@x8r-B(?WxWA*)2E^4Nof>nT$Ome(ogz!UGycG-*On;a)9 zZ>y2q9Q@rGzMqIWy?`vS@msRHt3kJGRjpgzUtN`ZKW*iy^FF9)HYeBFeW~tPK10{& zxITN~g=LKfZn>?7njQELaiOADrFI1dF;gEfr($v;C+xSC`||6N17f?3XgPbqOh);V z2=A3R=+rDtrYImb{LkhPMdLq*PDt{{IgLCz0K<28GY8j1;?t8UMlHHHz}uIiTNaLI zE>D{~+dAB6Z=~jiGkQ@Z&d?6Pb@vL4FfI9Eyf6T))hh~bARHA#SNR-@KMF^?Y?6}g zRf42_l1wt?3DL4rD`+pG=Ne;@cm0(}2m)Z-&_)OV2Z3aYnKj2`HNb#HEqNHO09Yc| z{UjwDykFuP?#F79ACE%vT?`^E4-SH(<0XxoVte~qxQ1SL@|^jCxE~UzjJ6t{e*v8B4)f#zUBN~ z5JiRv2IQwrBr!eYzQD_nFob}K`v57Ry_V!Q%Gsol@c3|fXL5=U!DHY^13a9(otAs4Hc_4U}nbX-OMqwQfn+qKhbX}je z$*4_Mw>0g!RJm(%jPiT5F~XhoHIf`!An>vpY6Ppd%4B8A;YS}t(DfpFombf2kFRec z-ynN6y~Ef@!W(dqw{jl(M%;+65D|0Tw$^CIJ7T3kgTLFbqW>9*ojUF{-4gTY=1IYg z?AEUg&6?n3BUA-WcC#@{yy*RuK~KRtHVMT-;zd@U+swIVAN)ph0&`dFOQ12L0qtX4 zAr%g;I5)GaCCrTS3xj-!G9A#`3$6UfSGn^7r3K1IEzm`^|EQfY!%;`hw2Kh$A1iLqIH|(^$&N6hLdLZmWsg73G1F-lE5OD@M&f#?gZ^$0B zZkSi;)5q?yg5EwG#Y(6x%w$p=Yg5hoX?FI5)Su7%d-76nFklt<0#&JUQv+*b)9R*D zWY#V&2V98-iyfxiEj?WpIGku3~k$ zcUD#@xT8IyOFP;cUDEK_>yg(dqPR5u8 z&>wdxu~3h@g@i6c_+_o7P}a}!GTrzSP^hvyR#T9^t4b&296 zu3dTuTcma}#`+8qMgNpS#1hTME+G~Nhim#J^e)^x6(d)M(A)|~BYeVoUH0wY727wTW+qFyhUY zCHs@ZtyZ1vjAl8*!B@A2wEb-&pND<*XQ;)48BhcF;JUQ~Yp6hYT~T@gv|&RO2Up!2 zoLfzpG%k1IS0%?~ZxKK*Fxjmq&+Yoq&EDU+eb}ie&K!u-{?=qXz#F*NqJeD)LG}xj zGW1wH@JZ)0;6wJ5xGpdRUaCo=;p^^#vM|ExO@q5Ejq~%LL#&(2hZXnOWy_Uf zr{J5wj9Vw$v`-{QpJE9=QpkK%baS~w5MpYlYG|hECt7H9YQnsdq%$mFuZm0)kod^^ z{3%P7oKhXA(==_4LKw9|v6onbMx0)SX@B64e6<|vbEKVmh@-!P=hpgmO(Tq--m{1S z_SMbA!d*=6D2$x}B5xn5bqCcU4pA#0JT-=uT?cm-vJSL_6B&6Mxkg!jGHI(1>21}t z*ZMNdiQm=ArPr%B9%jdL6t%YKqgxk~`@?EK?UMvQzEeS1;flx|ZluxmXHBB;WX4z= zet&}2IE=msEQc@y`iKV{J-|E35r2R)IM7*$9h=aKnUSlqXzK(g18M&I5O z2R*d%i8y+H?pNY0v$fqNEb0yViQxB{Vob114mP0Y6G?x$)Izvi5`7#Nj(ql8?%o!qI#^b23H)>$@f$Rrin?vWszpUB?-z*>R>sBeU&*Qa49%OQaz;_lOk zcG3p#(u%;`g#qa`&gog{=~-w2bq7ZItGXhFC`sZ>NtSq;0WZN07;c6bEp8aK#~BSw znJu>&bpn|+;cm@vda=t)F&JtuQ&taCmMAQ1AU(@34iXLr(L@#0ZA%AaF8$@!Od z?*4mQ-^*OL%Um*)JTPfq&vq`t!wp3i@D&9rR&xv;Vdoz7aR1y(k9@+6eB$l*f?JeNWbn| zV)(*xU+Cm&Itn1+f~A4;8F@k)EQ>Mq^(;dmP(0~RLWp;UsdXGaX^`Ap7a@{FUo% zX|Ob+G5GJ_dO=<9HNkgmWUP1?jTDUy5})ySbQ@(dahS;(r865ptjFDxHYo}3;(6kxCzCXrW#Y&RR+#I){GLo~E7bUzti|!OtJ4I`S*OK) z0%n-m;=SJD^U&f))*2wx8l>AA;?)|K*&5N&8nxaU^UxYc)|MdDmZaO3;?h+1`k}p+tfOA2qfxh`*{h>9v!lJEqjSBZ z`=O(ktg~OJbI^+nYZfxU3i;O7xi;3heB8Mv0NRqJ)Aa)sl%N}95QVs|_4Vw{itH^! zX4kjlE{wRG%5b`w_0E%t&a;kgc~Q{qWkxnSoda|C&3g6^jPAdgJwG}i^{W}b*J;lx zIxjN2S6937;FX6$bk`4E>q6PzCweevvk5r*7A^Y-wtLD?YIx?mDbKp8-eg~PR7=zg z-FEaaO!nMX^Z^_FqK>@~?@naQ{*%mJ5GEbHeBavVp2^6b-OqiG6@7JWy|46owU60E(ak)86QPDYc1=R*DJG(!YRkRKHg!$)jQ?;#!e5g7z{U}PQO zl7@1(begb^O?nNxOwtNW4#+D()h3~S-hEnn*-r9bC4cqJu4Z#h4t}tvCugP2&FWbp z8)wcM%2*rcoF6jX74+kBl(>b*aXUb3aSx+Cy_u={t7gu&(Y>fUSpSgT917QXE z$kH!P(tr&66lNjyXEULbJ>wM+uZMnj@7{HfZ1m`<)z34d5p+P_RNk98+{xkk&goQ4 zIvG)5W_1p;7DBE7qQsh)gU?ev%~9*mt8JIbsc3@?kvZ0&9OqC@1jRSASv2l1-vpw+ zac-jN&!T~WN@9$;fFpX~GRHK4&I6bG)u-3*68YUZa}tbmyJhKYSJ}NC7b3D2-qkOd z`tg1^SB?0@2ohcFU~b@bT>NRgpm;VZQNO_M6!N}R*h75rvW?EFi$gb?&VX%6ks|rY zm{EnT+(Mt$LtorRe}xOYT(h|H$xJMZW5rE>@txxe?T=zT{nhWSVF}1Arf(M$_32Wc zSb?~e_W@ShjLWr6QVCTnA#7n?yo2sttBBHfxe6;KSnH}LYm)sdvF9tehxG9_%O)zb zHB$kX0jsuHbY^K*N^C4GU+D8~7TfjL%y!o}1J<0L)}Q*({u^g+{TF51sPEq33=BPV zBcODnAkrlbf+*dkQUcN?AYBgKNcYg)Azedvr*tR{^L>2Z=Xt+t?X`Z`d;bad%zS39 zInVPtjygG!r7Zm=53l)@``B+dvz^pS=E)0vOIByM&{>=(W~R&dq%%JH7Rxr%D!lU| z`HQDZ^m!%<=`UGPy%loatmeH{xUpFr#QzU@2oqoz1-Jq8zr!Ws|FPCLq>hFBKf)yk zN2Y(oPyheIrGc8_nKDd0Y9g%K67B3-NHkTVhLJA#(@J`p;tQsHUzg)}%XbR2D{WrH zcW}k}KR$1)1o?l9pD&BXAzx*eTln z+UNg{Wo7^qk~n_UZ-jhYcA}|sxH(a_=*Zn5ewjdMvW;utb^NpLVE5{ys@fRyfG)cBD{mB?gz1&CT10K`1uy&c{O-nGoU|}TJv~+uK zt0OU3qQFqBc9}y zSm>-4Dm)rwEYP@H zWry=GY&ygC_LEz-O_IwxDya8uD-Ti0nYyZ#J&wCg^Au05132fo+N;+a_nil9tmh*k zrk+Qgj8)rqKi)A?nm1$A4(WVTXIx|&t}xyyzgT{V;N>=t4QYB|R<}rhwc#y0azSb9SaD_Ph|YPv?bBuyNS@g5as2>9 zK)b*Gx&r6L&L=6(I?B;VZ;CxwZ~DbfY=orcc7pp9}o&ReY5x^$tBEy$zGovs*- z#wDf_oXh#b1)Ijz5+*~P)7duFKypR4_Vo2>()jP`RO)G_+Z`XlMirX9#iLdKvEUc zDP&tGGKWEJR{?CGUDRh}csA%mwD0K!n#6`iA0w~2u(YlOHVf^Z4YC>T4P1tyxZ8bZ zX*RY%qLNC`x31D4Gcq?BXIqXj#<(}seQ-@Lr7kdzLtwt&N#<%_&EDgE(elJb=Ip$v ze>di%iEV01*4J~DX6HxzBLh68MWSn`8alVH5B-@)#*nV+z`to~;s9x3*~Sm}NR(rY ziy_3w8YJTIHbUYVQ5*|9DN2RDLA_+(1Y2f0PG5{6{-HDOKPYLX_5cjlDv&v|1ZjpgDyQn_ zGQBv$0{iM|)vHm-4}hFXIjNzz%XtS3j;i3V3j_J3ERvVtk`nx)YuS!p&`Qw7mS zYN9LHxcc}2;BpfpQ`IQcdfJ6U?D(nPrfm-{?g)Y0DKu;vReH=7(Z`i+W8h8M}r&B(rnDRtFJ zfD7XG0U%>$v{;yZok4uSIy>y@2ZM}#uzZORvl)Q;wA>fZ^W>4(-74OT5uQkwk`0=T zSv=>fPiD}WOby|KzC-$MW~kkonGzY9O1fRFFMc?ardy|Go&z_44{><9m!8^QB2Jqu zHtg-s%Qmm#BG!p^0T>~1OtiYlR@vVRj=mgmDSejylN}9oXWS!}QIYXQ`~xN7Jeq&0 zeU^=qsWj3T3@i;zILd_bg#YYM9I4KGk@oSGUABcz6uZ=qXCDht3bBDJlq|yGJ&mTD zDgTEqlRO2|9E|W+!0l|WU#%JI=bM>!Tp9X4-&7AjcRFLXU_x+X^lFLrBz8se=YiES zZc64M^-Aa3%B37$EsU`ju1@v(BDPw{RT2(}iqIey2|(RHX;S%|Tp;x)y82Z=No$>d z@xf1?u$L?c-Vi`$&BpaTD`H!vzQMeO{D>XH0(u)8ofK})rrW$ucIng|aJa=YbFhMu z71PJ2`}#=SgN_<$@`y>H1H+OMbKR2lYt19>-DIZZ9{-+~o&8=%j@dEOD?C;26wCO>IDc`0@`g@p`q47=R7SQGB4HOk1|NIAokk)AlVaWne0z*Z|8e0Cb} z8?lnrt8I9NETJ>qf3h}jYH=wv?=^gZf!;}jESIIvSiGoF{Ir0e8&APnhQ0)plP>twX@Fg z8sRqbl4~bFXo~-t-=&Afv{NeZ;7hcJPWztreoc%w@6ZbV(G!5{C-#}wgV^5xx!Uu3GJ6Ds z^%=USm=U@dRFcl|Ls?)zNU%z<1XHOi%M{cyFk6eUu?;*snx8rvh@GtMGhu9*?@PB*yR}SM#oE{8&2wCz}4Q z$N(ts4B-e_#4|Zem06F;S--^(L9n$*lTOb~tc|%}CJ5e?-VutOZ_x@rbWGeG0kHh<11=VQlYZ>qQf_(n^o%k4{dF{i2 z2ysZzHj|2b)d^0Ji8p`g2WO6^YDi$z4)f21u_k%cMKgvIzydLXyJzw79Aca*!$rqp z*czfIm_0QbB68eJwG%bNQ6fXHm4`JWm9Kq2kYNP7Ss2hIqH89HlUqegBl|tp z$e+Sj(^6mnMt#C_3v`B)IPEN;N7~)TFGaEwLl+U%*YHtJ+`5PHBRqs&PdvJb+}#Et zI3z$yjLMIl%%?HIhkcWQ|veQk$) zQ-DN>7sk(n;&lq+LjS;r!c;Nkx>P>a+Ggt%J-HzcOD>9NEDEwj$Mgv)-^;HMEUwfk zu9zo}=*pgqDsIHg`zi>nvjaK>OMd8-^thLF&I5)%S|~0Pm9si^3r+o+V@CKew9+@0KbyU znxx=;H)VhafQA6RBk&pIh>qOzL|LJu$>nWDYFibh$3?(9X6P6NxjQ{UvrY+~5imVp z@tCr_TL(0VsbET6wDuNCJMhMTuN==Fs}-8lYYEWbmdPSBO2*^x(S?Z4Zb~_B@u5Oc z2{tGNUv+ELmP-SEJrmcZXKZbyMkRTccN08b+0RJge$*2N|-}^kCMa zi`P(g&=^>f$yin6cGT$UQuScg^zBsugMgvkyQ`q*_fZy6GElvuw>V=Y&#VBMRX}+a zP*JVwng9TdsW$a_04YtbppHl~jAmOp=w~Z{$PTX!Sq{QA0--3@h8MSR`<8&q{n}Imjx*qn zw`I;A?>vg@cCiTft$-WFRx`HB9gjBFiYFs%P0Ch4d|30;egi;7&=uBPXH`jQqxr>IUlSrwPS*c{IIIaiMt`^H`EL6*-R4%DUAckFzhX8>x z0OP7|oyG19p>E8m$2bU(sa}tgQO^gZp7+f?7UA_oCv`^Q_-fAq!&R`M5!lJI*Dbu) zd8pTYvDeKL4Ei&;hh9+j`rbm*kSSH*kUo|QeU$n@m~j7_pMCME#5)6+37&+VP5rM> z`qwOD?SA&>=ndp~4iuyg6#au{7Ja3yrkPg0NHDE7hcp@MA8cJDJV=abxf^7# zAN-LDb*y}S$TQSCgui@RTYB&|j{(~Gl{V_f&~OR<5D&F=$}rsxbXIS0%^F`2b5J+q z1y1}3?fS59;Rp^I-Gw;ym(@WiIP-C3){!UQ0XE(BkkQG|$o;|SR`?(a+yHqne5Hrj z-zse_K|FCl{78+y9Ll;6r&~Ti(4&sEg_C2X@gaSrQ8!0O(uSX$;k7{#6kbDg>tm?D z@Gxu;kMsxWP(c6^I!*^4VNo998qNTGr0H=0Y-I%f86LI6#Lv`;8ul^T@$n1T=#!-( zGHQCc!$F~Yx(jy1E!EUMHgM`Wbe}o3A2(&>WuS+HFr}uG;Fx$XGF5grWko$v?m3M= zKK1l|T4aAvo_l)N6K^Se&=4EHA8qE=6W?oi;B%)(bmA<}IDM?K?7;@7f`dJsQ&e_&OlqQ>a|u`;9r=}`^NK` zv?OgZ>xyAAYN1Rxslj!=xMj4IgT`%q=v=~Qpk=OJd9g|4Y0L14DEXo#D(MJy#?9VB zG5Z2qqPO|%%;LBEMM;9CW$eWXj>Y-If$p@WKKNX1%iJ-}@^bU?6x~whb<8yN@~qe5 z`BL=a#Y*-4(^+SspGyP3^j8b?+MYdOTVjl`=5=mKs5s|X=ImUqfLko1u1?9Y)}GDH zD6cgQj2KVoK&HFV^NeV9)@d8(4+UpAAJ(~PHh5JiT3&AOdv6G&ZwR(-2rqAlK5U57 zY)Xo5N*ip-cyG$3Zz{BIDlKoSJZ!4cY`qcPdTX$y;k~7mzNOQ;rMJ9g@UUe>vuz@} zZDz1-;l2GKeS3y}HjMB;VHe3C%Q+OtM)CVs*v0X`S^nZt4u^S`-q8vu-{|>t(Xov-BjHU|y6LuN>zqXwJ3A_GS&dHbxd;Z&!swA55 zzlB{dJWKv-*cF-nX&=qq>R(GL(nA|q{mF*K67^f$D892D9YcxlDlLK*8`D)*E2R`d z|17BtJ{+CxP89dDE*ahaK2f9iE{5>9yR9E|-OgzvLf+r6F|GQ58n^T$&`Anb{jfRB z#@tXvK1jU5`Lm?bmI(eIODg8*0^?Bje_PJ4-ub}#(ym5+H-DtxMtZAI)N;VZ;2xRww%YKj#d9{IsX(V|F`8_cx#*SyU*TFZNIP4 zPmMEzBjwk=gLnjC* mW=R1A)!)VCNF38Y?@+UT)T;G#*99J-a6=1GGcA3KKovKz z;w*P0xV@x$+HxOWKTT0=2vkc9l$qTrV3Mc8Dly1=8N&`L2cmh|tmGV$qr}rX8;dn+ z0q&B@VYL0j@>TEs!>XAxuuok#Zpu;Rg1d)x)80tqfdq{4msNwoeY8#eUFgDbn;9(Z zlfjNHYDq^bt)fCF+NY-SPF$g48~-Fh(004ZQe{~i)uY9-cCrO0+aCwaF95csDy@i2nr#XP;azAMe1) z8+~4;Gsd@{CXId$Iu08JVe!q_9v+kaNpsSlv@8K#3#C_UHa$s9=A!IZ3iF|sah3L*C zqWPrk!S{IOVe8iL+Mf>-L^OnY_i?3#JD@EOxm+V04?s&LE>vRZg!}K2aB_EHDV_K|YhDG3VXspn4Y+^M z6PyOWA_}Ioyo%Gk%fxnBrNi?T=%z-i2-U(i!O&AhdjWxab-yu!Bn8JTgYbjJJtd%1 zi86GhV-Z|-U8{Y4#Lo@JLf=(q(~N%T=EW|3(~*_^XsWN5pCbC3we(X`uy(J|n+h0g z@d@62y`apQ^vAlsO-j6^J{8+&7$Yp3p@zO+E_pmM=FaqX-oFqYamL_6+sC1l)F(3U z5E16NrSAT9K>g+{%AETvr^`p**ALtgwg_Vmu@D(G+;f&(&rL3G`$65CF_qCx410@) zUefw@9NYSCv4_*aXM)dCl=`!X^zFi4%x*mXR$>7iG(vSmGZ$SHriTW(BEt;pO;WiK znZhYAWj%0nV=b<_M7()MgklMUy_>&Fe6Cc+8gfV~0DYi&?KTpE#t?Se+#&Kn*X@uj zkw~+p%yZeV^l7guh>GKbf*ZkWIlsw-jvw1frj^4%H@`sCONA=@6v}M4&oh1#6)B(0 zy?LGv1iyD>r7I?Xo4dSOptY6#*yH`XYo*`$JcpL#9O7@wFV1pI;34)oMY6RIzY5*o2Q#l!b?cAr2y=~-jcp0lZdxnZEqI0h8jcYu-++KWWL zu+<@Bi>6()A?Nt2^m_|Rn%iiMQg#59TY~gX%WrCYq6TixVf|h=5Esc^i!Ex;pxx>S z=LI?{=+{_3iV?8CCAW@y!2*{_xB=P#h})rC*wT%Obl9w z_L}{;;Kjfp{t6uV`>@Ra=9a`I@=@`WZV0rWC?VTs#NxHnf-5=Pen2%X!A@kREeFVW zX#;t*e6*I|R5r#pn>>|8Gz-%7hZ6#zgqFRxp%U-RD6LR>8yk?rcuDq_S=PST)e07_ z{Bh~$Lp!;rZaof>a;0d>A9OGzAkt*gS)}k?89Za?H;d=lkMDOGK*l@)Bfm1#pZKwB z(PH5TXXeRL??!xJ(;j>@zv9}%4M2=BLTOM5DMYO3``n>Q^)+y-2rF$oUx?(%4`YJb z=bW%CqHNj#aLi%d+sEuyO05q550hp|2!C(X(>T>1s!*VW&NEe7*Sh~hj$sS1o}j;lfglJc5A+zpR6#RDSklpQA51` z3NCxxK;#We!@!v$#Y^tw{m=^R3Qk`zO1FH$Ig<$X9Fs(H2jn};o>sEIGG8;Otz;AU zykIQ+!RATWZBZHlT@~H-^SX>8mL2X)OFjq59WGOF5eC4{dZZbQ4SC%c4G6?@{ ze8a0M-l}__^<#Zq{IFvMUFwtgDxIGSJ@)smse9_*np%=*{ol|sYB4@1`4P^*{w{mO z3}Jy&^LQ(^qw4JBh5YGcU+YCrlcK?~q+Md2!G({+(%A2w%T|UKx58VY{d6~tEfjwH zS|1UH*o3Pa;mWEM%Aw!xFUaOz;sXtA$&hQtm)WmNcN=v#JENKGw3D&ijNn$gPlX=F zL?%p1PlTV6vT2P@@;ff_vG2q=)JeS=bp)g-##fOK$Sk4n$6j10`x2bQGxA?9s#9qg z7Qg6UfT%s*zi84Y`7z(?Hf&YDqYFWjnB~A+meyUKR#`3(xT2kUN}YN$NeF(~arJ)W zW+a1u$|MYAeEL=E>jzlH1iD@1fl%bg<{8}gFP|RE0$GF|Z$&)}2fbC;ugm?em| zIa^i9xLk6-IX=OCQum=z^HXcSiF}aJRDgJQf~RzlsFgZ6MBMox%wsX((Py6bBP!I;L2JoOMuS>K_6pzK~A z<14Qy8HghHn;2Se+FiE*W-^`2K#Lu{BDEkDb$5N{4={D1iG2G$bbx{Bqk>N$O=6hC z3;zIL{NYE>J3IXc@;ODH@EoX9^g7Ut@3@OdnIQ_o$I*0&x!e%8&JEb^PNUrK%GyetKFK1|;w3ZEycR_n`$XD?-T*uM&Z zXvqAek)YHjKs+&w|9my)m^dNbc=m3jC2dG6Lv}inwIh2^vU9y_qGK=4m=Pj#i zZ5Bmlw&Pl+C`Goo0-MA@CdEw_kv)3t309p?j>15Wq9s;vT+U?%Pn!|mTg=?rj6V*m zKeGCAK)(oN)CibR&T3F^C8=bxS{_#$!x-ERPz}+SQ!5H#rJAg|W#w@u9glyii9|lwJEz zPKmE&NjC+yF92CQVe4d;0i${T`k4?sRy5sMG&^5J8zrA)rqnR-keLN;d@y z-6<61LUS@G6u&;gbgR=7TFiYBfuzicl zy#~tfi^{qwAd9pf(-2tH2Qn9meDtUaTR|vQ6aK+X4heDPfhCIEJSBr-WhQTKOGl;k zJe-~KS<=(0`%QQx(w(e!73%_47KB+Ks#xH&SL}&duE0Yn?oiFBt5xBjPd`>jFHxQt zRfKh0-gQ&_W3LQ<5T~EtjVv@4fr|d?s z08g<7Nm2g$V@ix~^?O|{w$nm7l$vCG);paVr=~LPCI#E*5VR;36D(>~R%jGfk%dQ< zl1BqoN&`((Lui!zC8Ek-q3-iwnXg^lE3B_c9d+yu^(84~+F|&53+CuNX>%y`*_O~+ zwgxJ?`Y&{)w*&d@Sby%NXDyVqDpc3w8~<1<6TVfsfI%R-b!ElY*0fD-y3lW7R7_Tt zQMa}7VU=JcO6vzK{AIeTG@;V(x&@iHD(xqD4<}7`g4HN&YMe6o8iQ3`VUULtJUC^= z0?=eQ7&R8~xFH1!)@#pQYua1LM__@Dbeo;RnyE6=S>DwhgCN0(wie9tA!q1VXBO5p zJOc}Kv5F_Yigzs3c%oSSiw&RXKn=Q3?6ca6RPBT-mIR-)DV){}7MO@&bA0pVyQxvG zQ?2X3>NrUOonsL|n)L@gYE(VjN_Fd7lDnwy^w3ewv=$8$Cn@Eq+OUKh*@ZheE$RXn z+BBNG$l|-t8x$m)4Qx{6WjwoXihj_K{aAa|t)|o)%#C|E(J3bj)u9r(A#Nza>IlZh zRzSAnNXy}Kr7EAZLLEK326S+Ze=(s)6Uj?F2_Mj3nx~BM>}&t)bXlI(=wUc$RgyVW z)(1dxtxdLw2mOQxd*fQ%e#0+m2gIzJv%~9t58@Yl)^`99@ssyG2hiF(tu0oNA2(=( zJ&=;jd$ieCs(onh8MHgx*;QCC`3Bvu9AerB@kc=XX-76aM>bMNc0ET{umMckoN5zE z+GMsf_U!v9I6CYf4_WFcuN0fO58ERVOI}mFf&)1E32-xnz->W5?-AE`2yFN$3Otin z!LmLY5%7DI=6;MsgbfTIqr?Fq5d%qP0O1k>jdtwr4q&|>2RJN$Bb6qe!VR4>5Z9id zJE{rVmQi6F07ZE`#bO+{c9b=3lKXH}^nO%oDf79)WSWw`nH7o>&bac@D2mAV8)Y{1 zLnBf6Bu>O6I%3>-c$9?Jl1FFag?=Au$0V8&2zM3q*$WW9ACtJBZ0MW{xR}D49<{g! zl>e3MzP}%P4HrdwuEKYRXLCREa0eU%g>?v!`TYdIF<#57B7HcspZnZwX&ih4U}xkI z!zL5crcZ>~!VmGJhnq}tpF5t9;c(~T_{?3iPd1$4wE;k?4MGNI&c_P?!Q_d37Lt@d z+}_UK{~onhM!~B26YJijPJMvS)vV9A+Ti7E&+Q_?^HcCq@CoMq^V~8!lm`7V75H@Z z_TqAvYg0at`NcwUJ>PEDSN ziK2)x&d32p53F}1tGXNrqtY>Sku42|eEp+E^m5imYNjmn+Y7SW`cr@Ql%JRX`cJgu zsJ%Hb} zYq2%36FIVHiiYKQv#UdanA<^gd$AG~=u{Z6cFtw3DrOs6WGAYgy75g&7}L7kx@mcz z*bK!~)s*!49+BfgF78WR1ZKMN#{TcZ>irH9OSOd|{SMrwd|fC<>R~^_XHdR`wt(X> zq=?FffZ}7-rBi+hKoF7`AC-_}tA9wsf5^~eG$RU~Z9Tk+WSMjkI*}LMEZy^ecDS@> zgwl-PP?J(desT#uq_;o%-D>npv59ZzcZ~Tq^&3g*Gc5TDZcYb1SS5{wi7m_7Be2fPAL!SYa`N&+#ETLGs`HEA^n!?kXjY z33_~ul_Lz`n8M=%FA6x*z)X+fzOTf{ewB+GKZK2Tqri~8{|>e)?D~{P(V=}MNu5Aq>7@X3Wg(ZN{??! zxNgfvZ(xQf7_i^_PGX%d*RW4JZC`dezwLCD?UWo};Gc5<0T^~l;q0SB#R($%pAL8T=oB8w!>3VuQspM zZFFjtoE!DmV-Nm~*$yMxllY4E*Qu;Y({jq6$lz5f8PD;*FxzM^Eg4u6aS#LryifPr zJl-yr-VgCJoBa#3eTZ)<)pqHveEs{|$<$YEB!fnEoX#K2HXo~VKTC%anlQSm_Sycj z|CcuLwg8%$OUJz_m(LC?u117k>yihTRa{vKyMp20WxvyiC?&l#sXb1?^Jxa2vEjjArsC%IC=k6>UYCMLs9<5pX?WD)3NaBfsPFVs; z_QEg@vl@#+O1vw!+MtI1V}b{A@*$OpM}t3~CGS~T<>Q)yc;)EWFu7k55mpcIV`))U z$}_oT^W-4%N_>KgTyxHPnd&xqTL`atd6i1^n&67hMwM)^L3xuOrD|EJn8lv&sW_L{%-zf%57(QmqUV6Hloq&+ly!F*Qg@K+#=3iFJXVK#jfmB^hW2~ ztN1Jwv_$b%hU7GW`#noy?fi#An|M^#r&`hU=rP~~4K_4LM!3q5KpmFju(_<+!Ko_1KnIL74-VxUTGMZW`m$ zkC9F5EdHo>W7 zQzyS&l@;5TCnQ%~)g$-`Ao~HgZ8IfGCQ^x4(Y5PLpz(~!z-Rgzlb;Z-yVu6Fb9~f% zI^EbE$g(3W3E+N7u^F3T6!xiCt8Mr^HBm)%K1jGczcBUwHJ(rc)@x7u?$5pVcD+x& zr8=(_TirX&e54V*J{QGH`?W3gh019we+KS4bGqDQu~)~%VYf`K*?f6Ij-=jOZIEr_ z*^ijvIMZTtE9)ZkVM1L+-y7Mjv+}z%PyodnZy4L- z%?xL8Om7CCgF!to9~sH^(&(z&Z#F$hHhKe2JU0s|5p74MY!o~Z5kTLdHOfpNRe{oO zzG6lD=&_bS#67M#7l>X>HCSR+8RRV?Z6eplEk)@!@^!Syus>R$I}V#Cz)Z{dRin%+ zP6Lo8|KDf|zo<9*(i^e^^)mc{r(PaeeIyfFoeZ{SgvvvYAH;XX--O#mevwv)t)-V0 z*MJ3*WUh6y`%%9-QVNW0h9E4o3=PpfA(^y{Evn6iztE|$e+?jsh{L+ z*q_peEXi1vYn!}G)|YGSoxkV z)ivG8B>7!wB7M9Y+Q1Z-y!*%91V6_4xXAXnxhJ7!0=ua&xh>+PtVGNy^-L${mnL}` zD#Q$N>eDaM3N_DO*(2Qx@{Kwzt=^c4-q3!Jj&I15Akucky1s zO@q|{D=yhOkZNkfoy^{+w$#|Dl9WplYytpX<_RZ-4-BPrDPieiU;t=sIJo3If9+E z1wl>PmVL@b&joj!P{;XOsjn zBH_c3_{+u>lQ!?WlOKjSH;3da=c6KC9yB+Htsb%N=ZVqO4J+dd=YwMEgE$Qbi>9`> zLlX;UqhdbioL|;tbHCMW_clFy;JRSO6FR>rCj0ep(W={g{Q0~~`)5&4)FBx9+O8^P zGRNhcp=z1Kj@A4l%DUD>wG(#?|MUT$I()$PMDAI|tDu1yLF`5ZF{7|?z4J^BbRPBz z>mc}%c|or3Z0DzKEE6nb5}Fi^AfEreqRFm~alPLUL~zm@V-&4jNu)$+u&UYIVLDSI z$e|(-%Q{-kIk8*GM!`=~+H|R0OD2>hDo{>u{g_0ylc`osiS*fP;>#;R{^e5xJM2vy zV%J)1^0RMsf=`IySDpCu;|50I*|g|8kpseKIwBQPDAx9Pv2&d?ydK)jvMd7$4eKt_;QDSxTJD z?VJRxQq6wy!I{ zN8K5BQ!j?{MKAl2quw{?-vJzEooyC`AH}Z>h={x!!2lvq@K+|ZMj=?QDlVajl(S#?ZZRx> z7S7RE)ddpQ7RDz(rTw{y3$@jhWz#XIB;FOjbu?E05I;}u>+@6aE1l`PB>Ks(s_*Db zsM$0%)N2}l6)M=k<4p3p+cgT`u{Qria9N;IX)vf5c;g%x9%Ckee&xjn{bn_>G1H=a z7g!MzQ4+DKpjD{k7+PBpVew*cZp7WWDh?dR(PN%l7hO~>i8)=JqcBqy=*%>~HysrH zTc;EW+vnT0#Bp{^Nxp#WGk=7`Uk_|Z2@V$tR1@23TsZ28C6TD2VU#+2=A9oJAd&7* zD}Cn}J4I{W+thS;n8pp#JLK$c?9GjH9A=A6;wG&*xhMQGs~b0)=Tf_o%k_@zx4viq zsq*yNi2LtiLgK8;#=c=L-|RDk9ezy(gg&>LS)=BL!HXsW1I#b^vAU11H6>ez%ip}I zm&aO3z$PQ%QA?zRqc;h3veGC7;In5C^$IXk9&!7KP#I0mp7wCbXNEMHo6_FqbmT>p zRfl5}VLrb4fkDF^V;H?by+0s&fO#5g>@X1^n$%+$eHQ<2eOa!BSEq`un=K$_n=c;| znHoH}ohbW8!>>z(V!V{)6%3Ml2f<#UK{XB<2zLR#X*IZV4t{%ee0 zX_?wL)K>5$z6o`8j^tPnnr4V$$Ihi3vq=Vd`eD7V`8kotoquBI*kF;-ox~}#& z<=cBSgLDc;w+0-*gxZR*{bts})wd!6eqixzVfMX%FZobA*lp`<#$JM__?ykXmdowz z545`a5F5W0W6IeaJI(qTkt64-TM(1oS1hvKJiByChQ*P(y5AA|UbhQ4+ZX38%Jd~J zO}Dc>7A~EJs)4-q11Yv^)#6W8#=UhHSJxgc>Q;ag-wA*EH3Qc^d=-bUb(vriWXKGi zh&O?^Rq9GrD<6Dq$v(d7sT3J$T^}CJK5z9Je??T4cB$84jLzM{4^dx?RWHZHnZ<3^ zD3wn#BS%7uC_jb2boBRQ&6Y}e6@l%pv4&OyPw%A`c~%5UzCSp1kS)&Expv(;r=JIg zH5`&QMAX|K`svK!kEQF?{sa~?TFQ#N6Lx$5S+JN;Iy-ZIy1;fg`@ zouK7-0lbvwv?_6YoYyOKn)+CNnxX&W8Ilv1E-PKK38*G58PdW zUfZN9mosAC=dDa3JIY+2!}9NEt1LXuBr0!8jzupDzKhOY+Io4vZ@!#ILXaC+cGvrD z`sZtS{bzVzN2M?+FUz192WPy?+` zljl_(0j4v)9ycYJ7KklZCOv(xWlW@m{^@0}cOK~%V?PvHJL2RT{KzRv6jkz868z?B zzYiV0l)*m7&Zy?Jz~OUC0cJ}nZio~uM4B5SM&>Ur;4f7FRr2cVD*=ci6!b~LnPC;v z)(DMJ-1+S;s)bR2nwG6baKIZaG__f@y;!>8Of$|N`cwdrBJ-DO2TAn>N|y(IZ3o$1 z1=>P_q_{!pMkG(9F)^S3gDRQ@0&Jm*#)tspkOng#z#&>_p?%;GWPNbxESMjng~H?{ zQ;BJ+3P4qfJ8R#A3H<$y{68iJ`U9bP5dYi)|0N$15)=@Iy~~*+ncEnR*tj8;M%Z`- z*sLA=p&+nTAk-fgC}sE8GlWa7Vw5KmkyM1oZ&HOys7???fbA3?MUj#9Ng1P=cXfd2 zc_G3Ap+CqXSFa)z5ExbYI)$$S&1Q9%1tQn4B0m;@B0(gpJebx1*QFeq#E#BsCz@|A zgX0zf$`w?~nc0+iTia(!)8iU79@b43DGiP6 zf`p4RgZwQ>rW2v7_7Sd~(L!o41qT>&y#Mp-2waSBV#s>X_O3@KZy;`MWX@`o6* zYH(qq>&(+|?ZC(+M%c?p|J8Vx;Fynzv8TqqHpr*?UgY$=*9qDoafpyO+=jSE=Bn8l zB$e`#CT?M-*TxOk3G_)Z!G-ZF+L(?QWZfYgkCLHVA?(Esu%C0^73yOOnWzk=5(y}b zIxxN^xy7zxJVw5k-fT#HmX!WxE`9Y1_VQ=)FIQC=e2OZ0{2I6K7eA@4lj5=W(&XHt zdE!&a>@(7%GoDkV;U>GAFf(vWn-oN+i6~^tlV^BUWHxTH3MSLMzRo;&8H$)ogkUQ5*t_uB1m(&?_OI~W zE_h)h3ucO}oBwn5SXsQV=hVCvJW-N_Rz@k4bYF~&D2pq|*q~E7#3IJ*^JM1N`S{5j zCG&KYgrPcGlH7$8`~vQL=qV-7kP=1ePYGY@(g^BSppI3Ez6B;H2&Jqf+}>8>4N^j} zDi-#Ly_wiN_~{+BGi21?Q>#@ihLQuSXd}Lg)D|iAWbxEJc^9=3vyD)?lp3BW=upwi zgjW@ILeL>g5*EeE9BnfQFCdU!zH!|LaYYC#glkJjo@ zQ5wKTH3l6;uS4}ocx#vzDD@QSI2)r#=cp=OZA{;?pwT5yIjXzsz7N47b=Ix)oEASL zee;Da$9I8VbdEaUR_?P{7W=>mejebg$(XFx7_`3$_ zw6AK=f_)XZE_A_)q>Qo#4_V*vvaxddW#%FsX;+M3P;86q3mqENCUA7ax|Ls=;$v;| zwuf$t%%*qqW311F+Ia07oGW9`_o3B_t$VvhJ60SlI_;rN@)u5ZXZ!7>Yt1)W3_iCM z?7S`C?dhCeby!liJ5sdLV$(uAo6h(+eiN`TP&ItJHSZVd;AnA z5Q1^AxAUO^O_bhzdelf03!uv4{)}xX#oM2N)>55H&6-M`s!iFm5|P|}Ak+n3!>Wa1 zw*!!N?j94p9^-%W$g!~$C$sM#4Rxsd}CDdL68L9jUN=Re^8rh4BO3qCY52 z_4)!m`+`&Z0+2)9xKWaA8GY{9{h?IYk9c@&l{$WyV?{Uj!-fjN!=bCmeXc|O0jb!g z*oEK2p~;`|Oq2#nv9Sy9pzS``S$6}0O8uzD{T0piV57m(gZ?IKXpK^DfYM;jL2oOeih7W=>1}BE#p{bAyBfLLl8)9IbN_(w& zC>Xn^AAp3c;yu{)^?3Fzh7S-{4W{tED#7lTU&Y&|8ctmtf$$7wc@85E29}#g4;Kl8 z*?WTRhJ(f3YEy^S*av)!@M1ObkalC=_7DS$gnK0;f!5>u)(Ftr;8OTNXw4v-BNqPf zco4E=yfeJ-(sMMhWFpsdz~^p!MG5xPg@;*rg3o4x&1Nztd@{ReGHqz`h^k*$q(?mh z@%C^6eGQRNq7}Q~?6x@>Dy)5=$ELef-Z4pcN%IkN^ zi}cidF2>qj@W1ItYuAE9OZHteg${! zYFS62S-giYAzD^fBId+7H+YU#p{+9@-}OJe8sbm2M(hprQm@2|EWK@Cqf$ZXWfk)B zuvI)-!4c=jbL^SdN4;yzmdwB`($b&d;Nfz z6S>)W>(p%n)Tv6460N?D+!>+XtkMUss-Sfq?c}E)bbEnt+YiRmcMnB2FL38POSfxB zwmF;Sb<+5yj5p=Od+~La(u8*x;dA&&fvYC zIbakI+A| zjh+!2p27@HM+|86Z9Y4<{tEIwf82JCUV4scfPz!zO2h`0czl{H@=Nb{@lLeiY3)Ao z$SEMUs`}{wOKka_mLg4Se*b2=BJ^@h$xLUzP*}M#J~p@cLW7g4?#o!L6@t!kK}>fvX1PB|_|;ROz0+vwA-N+eAJ2By#qC;rQNigA-`SG_~ufH+vw(rra%iIP$xneLN%{ zu6IXRJ=X@!|14OehPL(P;a3L|kgy<3QbX;B@ycQPz;(e?Bppw{PBwx`95cs=~>R#lEn4t`6o7^H+H;00MV?@-QxYHU446it`B zD~sN1JV{Xd$E`Y#%4IX{f2+YbqWgb1d&{pl_diR!3JSL>Ah<(tcTIrcuEE_UxVyW% zCb+x1yF+jb5F7#ohX6Hw&hK>3^h~dwXP&k0f8l=dy*_(i8?yHIw(541$@?VscIk@c zN_GSsiJXKw&7QxvRq+dFN`AgC zE&h93Rp6J|bob8%CgHUjyS?F90qS5CR={w0Hepy?#&5^drRu`s^JI7P3ughIeNVj? zk87dSEVV6HuGN)ulfke0`fonZmaTtAdfV;v=I&mhxV7Bj3Y1puD+S)nY_)r^bs`AS zg{btm9fflOVRlHDB4|85tj*~?Di}leqA9T==!#Lp2|lj@4z-JHd8%(B0*NxvslJ${dAq;ol5ZC>dp%~0EABZXtXXaAA07vx;F zkGxt`4cn{Yz%^Kf9MHo0BgC1ODY9x@3fCxxLs5vUp@K;Tz5;pfV>71O&}!Kx2UJ;H zD9c&)mBPtUS(@ex1F#%Wr&b}y*4^}70rq%NUPe-HQB_q@d};Ag%1nLs06un_K7@Bo z=JQ&WHRUqMrJ_OIAe|^#7)CnkR>LHf@gQGEn$DyT80veUZ-hm=lYSmIjH5%wNl?!RfqgWur2(L*j&taR%9tC{4LA+$zrE!uX_ry*-SfbtRCxIKE z`51Lt`{A5i3*S+mT9}4eu0K5gG5+fSH$ce0yAI33cnN;1;&k&4tI}^_{MO~AB^|`p zm9-=MHr1_%9X7SS@PD|!GFt?*V~8o4+px@Bgcs2yGzN ze<|Gke{`a}S~~u(7lZ%4CL%6t{;!(IOk263QD7(&3CR?%{g1zEBB+nsOy1Q&h+@L+ zjr#1_!ok$y)=fu$IZ@0{FH$Q5LqB0qHFx~wMDhPT{qMqEQiZPnPPn_TvOQ=6by2sj zf4qV87aGj=2ua;t{3N8BEsuHm1H1e4KMQwv0uiap4CN1*q_*tP{}%4zAPE2O!d)VL zsl7YHG?$$FuKDGQw0R`ziT2{wk;?B)I?mCt^TQr_}rfyeG7o6Kh6umu|Cd^ zpsqMBh~YjhQ;NjDKX!CSG?f*xP~bEzK@J?J^DicM!LzJG5<3-%mw%pT5(s;Se1Y$W z$}F;AZW^OhV{U3QhzKH z(M9!3$k@UucWM@)@gJOg`i34-2|BAe+eyg3ztgn3O;Wz>zMtVJ-HQhG zyiC^gfhYy*NI^Wh4)r}aIyOVtv&WUfY4U<|LeT2$K4IK2MiBshqAhOq35m-^02S=c z^(5;k(;HYZrC=qMwgq5X%fG-T#n?!7MW$CO#R#7 zJp=~8BhH7`-8;MK)VUawK3>GwkId~q6EwaC(uP~awH-P$kdU>7d=zrG+>5j8I97aM zQtE;|VCY{FpLcWHk6Tgr^Q%Pw{oh^|`OI0D-Ox907o~OgFq@X`4t)CR0 z_7i1)c)%CZ*1jGJgY68zD}__{EbV`Ky=_&n|K++OskQG{-u^Srdq(v8CGY0Zw{P7~ zg}+F-7XO@W&=yeV?K9bXXWHn0-q&cK7`_Y?9VCIl$l$$1i{Q@3l0OM>=imKOU>_Q& zUJ-}exu#jj;6d>a!tz9} z0`WD)G3cbiNbkd~3LvDhsn*g=mY2*Ijw%)DDsSZn$e9*kRa1>5yT|%w z2qFjm^k4ah?vtIW(GUQ%qQiYSb4ZIZ#So11NPEVDSYCF8*(8)A^B1>iA2{HU0xn~e zUw+`bdNy-^t)M8OPsd$bsiv$^K~YX9V)&vV!FjO|Wi42EYWQC6$%j2{sxj|#1#Z9J z%OGiQ!~lD3x{$4tUD`F$G#8h&l$&3jhvse?Xaq6PrfI(8dE0R*D_}fT$LvrVeQHYel%vp$v{rswyiU_drr^w7 zlLa?i+Al4lv5haQD}4|-`UBMo^9rKNajkt`}L7SRQ_Go z>!2u~BC!RT-~}s%YPU|Qx9R6QYWbKA8~OfvaOG!9T#)00U>Zc2(voCG$+zs`lxmVu z{mGCwM*S%azhF5u$p&zl`bqpJ!z1to&q5RRT@pC}gs_V23WQ)Hu?Zd*Bnqhr780N7L4Sa31uMCbH@v(nlLWmk?<( zq~2t7r_k3NDV&v9xSM?Pa!1D`%bo8}B*W;_lJE^kyM%NO9qj(QzATzbxKCH$C`5*{EtahS4N5T8@h?mU&>DjgI&?I-NH zvXH!+6Jo?tS@c=1?|GX3N%Eit?QkWNkQotSWSB`X-V5sKxv%o%^>HqdJ^UqNtk zFvf`$1t{1R`z)KI&=RVsf6@P~5qdeP1XECJcBd<5sxPiyVHbHD+oSF^Z?)ihM9uu; zJFfHEmv8N*{3w#SMI2@fV##@^;7k0Rz~U z4A>2K-j~GXRtImvL3C67su0*~(lr4A{-%=Vx-FpSI{zDL@0$`(hdUM?rw^iqgZU0J zjRXA0M;QCO-~kOEL_`XrBS$StC%FjST`u2JZo~)H(1$CGMoD|+45t+YXV!k-(E30D zca&ZS2O=q(bPWh4?>n4Uuu-FDLxXF92DpyWzh?itM|EmVL2+{uj=yeImuMPB?9?BfFxWP|I*}xu0@F7I5 zkw?C-4@h;3lw0!ueIHo@pMY%{*l7{P+v=>2=jn75wVxi3ui{^u&TCQ{jYaD_5D`2? z>*sXjT~FBSkoGpZPXdi65VtwU9Ve{p@^h0XqKzpimZdQ9`MxVl zWj{)EzW24DOLe^uWEe=bNsT6;3w}n3X+N^JqE3$DbKq&u_~Ze@7fHO`Zc$c)&FlXxPx znGVFljU{#uM4-vvafdO>1JH>?)0>0+;0xKU3x+(w>|vPGNer`2tQ5+0{nrdF1N82O zD8t4@RY66ctnpi~S>&Htkn=#u2_Td_5X#?-UPHTBr=nQXx>)nHSoazIf*q^n8lm;d z!Ojqw+7RB(5dO<^3G+^gIcuqtOsShS{1;(lPA>?+5Q77zs3n0>0F#wNr6fFx?!2Jv z7u2{&Xrge20?gnH5ojokLoUxaeJ4-An$uDKb-g^Vqx_x>c;SHI?NL#ZRZ-Sa(dtl9 zZCyckg9%iE*S)f#o`W-=gKO1>?^uWH(uV0-uk1aos_dtvSyp( zXT#=Toa3(^VP_mAmVT zdSDtGC4O{$RY3%?02?i37y(B^Cp#8Yg%z$o3~#p}ck;?}L>V&<(H`)y#wpyi)I#CKt#;ns2j`(l8SLXK&g)CG)@8wx>=<(}SP55ks| ziWXM^4(5M*^)Uqb@pC{8G5j7eKHoPZH8hD6ltryKbuKhfpg>fD*!4v?&X3up=h`%M z+9DSDDL@c)kydCz$Ahro7n^(}(YDOzGVe-Ub}SGep+h*KBM+sv00mRG(gc{-S*_E? zi2}(vV;>+iW8UaQQGRC-Y=|@3UT%Ytc-S%331J)TaP91hTVO!RZ$^azITY{N*DI?u zb-FNsREk{hUw342%? zdp6E`qqW<##kh^WlkHHbFb(#qP)m^=s&;33UCfDEqCFDQGN(np?(eY7_ zi_!0K9u3N&O}$2G%^L~xArB*>cjy|schW7g)g|SOC3vqo;wqj49>d$zT|F7Ms~XeP z8BUT*@(aGslET=jf(;Rv}(vUM| z(KR+&K3y6!xoyjJS4H1ckmr*!J4iIq?4#xMP8_vbxr4~*CQ^39jHiBswnwE1UQY>8 zZyI?){Sx>pqg%FKmk?&edlTyC4)9i~MumOvhh7zyyErtm34b{B> z0Q)qY?#oxxSe`|vWizzdpOieOYTqw2C1foju!Z!)PP6ws@WNogH$mu3pD8_V6C z=-wRH1Iv*CXMG`9eQY22m;A5iCx5M{RBug9Z_Y!jHxTMK;K{c!g|`*{EHy1`oo;O< z>8*pv*D>cn2*0;LB0sPa=Fz%n4Y#&7ySK-QQDOQ4olHMq$k(rkw+>@>R=kVewsyvU zZG*^vfZlr19qHD#*3siOvEN#gyfBh}?U2rF_Zfn(wm?_iyJ$evF^66Jn%#F~JHO?@ z7ruM6BrO~Rn@O>tfB;1D?H#Tf2rda~lnq8i-8wqS4jeO(Q-Ay8SMXHs-&Z9E#caKU z@n385evqwP#DD}uwTm6VhfSE@n~S*z@Qyz<;?|^gFjTgGQhWf*)$KP?%*fB|MG34i zT>Lcu(1^Knm=cS4cZCsgb>R2)5RPs46xwhEpT9SOhIpT+61cthTj*z$9rz~da7^!* z>0kfiamOSgb8u(I>B4=cRs^(@cj;cHmZugEQvyIlY~& z4~N^#4O8p1yY#>_l3yvgsB%hhXof)fxwChyzvZ?;J#pYRg`=?Rc}y}l6ZmV2kBF#q zAXQd4Y+*PU5g;uKm}wUL8-1M(DnQr9ju<117^CDAl3>O2?MBG{h8N@Z1L>{U96-Mh zNiX46cMeYW$8Fa$xYltyMzpBa0r1)XPV?KHcJJM11K`($#u*MsfdXxnF+kV<-qQX~ z{tB+J4>@zCmhMlt&*X2@Z})D!clwtpglzXgDvaj-01SsajH`!0`@3NK#~{T=>)HFK zlhIAwoUq&I-@`jR7Js|9F>pM;5$tUW z%hPD{=)Cjyzi*Cz(7%HCDAH=$KZjyG7RNs$ z10Z+yF8c@IYuzOpp}sw4k&5!ph5o!1jE~oAD!JF zUoeya7=3^$6<~?~1E7CD8WaSFMj{r$Dl8g?M*QJl3#R|`V!%NHz-Ql2gnzK(0%xl1 zLlV#gJwxzS59R)~V2V{BlK_tnZgZ`KQ7k|J*{w^-z|iE%=Y9w!P^VYZfsZ6$72Vyg zeUBa!a;w!8h?}Wi`uD{^yPRT>g!an1a$Ez$Yn^GrkSsF}W$g`)Nb_nZOi+u&!*D2I zc2N9N2NjXlwdq7EQpIiNQ()k5^s-Yj)&8^lD}g4+NA=Ez8Q>u&Hd@JZ_@0KK>D_;_qEG8 zSe!+r7Sw~d5gkrSde0Kn@+S4T$;)T@^!kdM9w+caG+CJQETw|B4T@u$P$GcZA>Q{w z)MnZQ&oTtAGqCGd6fMxCFb)U;(;^07a)_}Fmk?M64Z~+RKt^!PeApxiv&#Y^k%C}F z#Yz2}B0mMeKMq7w#5e6lQ^lI9j8kUIEkprfXg|@FfJGDX+0j+NQY^hWWBhDPphh_k zo`PC36-{zP5fB<c;72Em{^!>r2xXrFu%5*0sZR-HfT#5jwWLuV$(Sy6w;D8U-tp{we>06T2*mJ{(G9_k zWBd`6U_7f6M$?027{&7U#X#R}#gmRJ!z9rp$-^|6Bdd8oStBgLEZtt^$t=^Nt*~a7l z*Vf*K>(A}~1N~?Be)|vfKZ~CCA6t910{l-m$M~iHx2?Uu(f>{m>i>Of?|-6y!Ayjo zF(czr|89ulwls+s9=I>mjYTU@WcuT%G9rmfyl9%BBtcjFnVW#mOtF_JJSx>$9sK@JCyF7)&Kw$6-9xm;ex=!BpfCu&7|Q4}$I|66fOn50gBm&B-IOO zLkh&Q^Vp5RLid<&pCL@aL&~HJT!alp`lU1d;q3e{%YEkQ=sTuJ#1vgLbK?nH{MYA` ziozZ-9-d(4&Z8#hGmagaede}{Hk1qR%f3cP`A;>7LE3LLy+uYESB_ry8&`ozBs%Bh zy{;8@EcTm|4UB@377zYnc&CQ!(~Z~?mucYfS1chjISO$-O2qU z`^fXl${@VH+e9lWP?yZ1`Dov|Jc@bo+1E#^RiXDTL6n}_d9Z|t{DA$-;0Ki5%XJ67 zYuZ|0olc$Q&t;{{D}-xw-dl?4$g8=~^Uatz@3 zgw$guTn4uBf)^>Jv(b`OpOK5MCknyD*_F9SjRPBnY%#7yLMJ$`(PM=;f{E?=JAb@$ zakNYL>3WP zr1O#=ODd?#D_b|N;fkr__9ETR8(FPlJjv1|c5uGQ*H3Qh0)ak`;$-4aqFkt!y$Y6H z${B{~yzG%~P8IL?d49t($I4zAT1T;YY$j*iC4R^-;!sQ<3*!daoHbIT^gb_9G~D9Y zw~exRjf-(GN=KbGsmtI!C+1wgFMHvn6x3=DiBTx{iaBJ#csC9i5?zYvddVg9`;DSX zm0FP*YQ?Mvm^#Jdrtsjj5~HbDS2mqb*i{|)cu@~Zg$6Db+A@hJ&(S0}(}7fDd<>+E zAEfq3oQoeh23e}CYBvfJGzG)1-PLmhnxi>gsfRTl2{9uxrfPFSF*;t>b0V*bWsx9+XPFM%oR|*|dy$ZNNys;`kcr0L^51 zkP`hil|x`1OVMHyLHY}oMyDiu8}|tI%U6(n_Sjb3cJSky0~t@o!b12z8oC3USQkII!A^&h{__h!{1SQDvg zx)j?5X=6T`4R@}SKl!kn2hiw17yD-h7|?vY0Wo-|I9R|w8L8G(=iQ}#2cHqmsAdnU z)9=7@A`@ZdCz%RPFvHqDj(b8MQ(BEn5zi}5i{>+NL61wtC*Kn#XsD2_QMAG80J?H( zI80I==sm44B8t)g%v_CIORXRP-@6@b{eYXVIg4hBSAYvW8CLwTOz4boKp1u{)gZo5 z`oR@R$^gbc;loKOOJKQ0`6Yg$7qr*B`Teoaq;+y55#SpapG7k?GI?WYcGl!J^5yUi zB*LF`TYFnBD|EVr;JJhw5KjYqn8QP`#6$A%{Ji9%)!>4m>5n$=GpP>Vm!LS1gkY{> z`rpwX2dn%PzB3G{9eRi6xReIW;lAjxxwN+Pn|)hZk{4 z=K+)w{ZR7AR&qw?|x%aOGZ#CWQ0aK z1{v~(Db}0mNrhQvgju(R*&b7V!NW3m40GZQ_lAVKc!Yao5V~oGw2g#$XM_jfMVS8% z4|)WLXhuY2L`0R5M6X7~L1}TqA0yOgBU9RNWqBDaG9r47tW)tIU0jl&+sItrC<0zd zcyg8^O&7t`DC~L@7zoQgc2GSpSJaqG+GrH!Z^FgfsCvslo+atbj8HT@%>J?P6K9Aw zR#0bX424Gw`6_cyL^K+CSSYU#Ry}8gRA4K>W0RI0dnaO03>iFZtjYnWLk6dl5O-`D zcU%^ylLrT21^mc}kX~^>GC(+sV>NZz=;(0ckMxAj~TaSjSXT zk0cYVRESlY3td{5MDjhCGQcn`DC4~+l8uN}x@Sa6v_zb=r`(s!WTU6BjPNA2NO1Yia3uIuBC7r~> z7v8dyodKmw>_16$K6cGI6vSHM!CLw%Gaz{}{&gie=b0BA5*a35AOb#MEUfdMUHFy< zo2Lr^yaOa78|7KHz26#({W10|1DTi%7*v z@Md#J($?^~eMNE;VQ=mj{QSjg+Qpx}idB(88)RUU0IQ6VFum~NkMt#)QK$jJ=uK_N z++5MN2In zP|HHij?$>pQcEvLTo|*XQ+Wy6lt>2sgaa9B66NfH@-UbGc|Bh` z0P)jlY2MASLs0Q)HBPIy)*Dn$t8cBV0ubv}q5SpVD{3`PD~e9b zX&CB>7=TmB5Q<8g)~m)J$f(k+=#g#jRgue0PaD+eD|YCcXe1ikkY14> z2}#ZFPACX<|Gu@?*@1Qi2McWQJ63@Q|FyN(F3>q;QZZqL*D26dX9Iv&0_rO%aBaBM z`~kpU0T6b&=1{tayaBV5AcH=*xR&TOl_uLg6g8+u*HUFCT0&PAEwH__qy10U*;!Xr zcF*#ko@;^5?qMary>54ko=R^})nxA}O6Oa2SL|cQ6=COWXXok}!)COes7}c6e;D3b zdm3zdqNMs3HvT3`bv>15Js=+ds&J>TR9FkR6a5Og!V3iz+`F^^I85%Z+2}#7b9txK zHyGXRXe>!{PS@cLLURa^H|$4H>VakKozLzB1q@2WbW6+i!3zVu{amX>eL|`?Lg^0 zXBaShfg}4fm{u^#7Srzuts1&y7;~B$t3(-&SjCHY8R6&}LX4^hy=%@|7Fx znj8xt8c3AuOwMt6Md|bM0V!M|>AsBD%7S2(qMMO&+ZglF0J-QBlbuzQ=S73?(>E?6zI-wQ&zr6Seo+}n&ywXWKYDL=L? zcWL0Pqlj(G2%K&M^RMZA3uiBT1MR;K~+|y^rUlm|YF#Pt`z-<>@E@ZBEPxRPzyomj2a+KCmm2MWDUoRG{{%@? z$jg)d1Cr43psl(%9R?(`!%YlnC1Aq@avA-TkHiGx369^Ed@oVTrT7;lp-i?%z z@lz6oF5cm8PNQVe?i|$8!|~it20dwuwAuU!x;!)80(Lqu_FK@s*G|E!$))Z}gO4vO zn#VQcRcWOD|Mq97teyQJ&3t)+_J)r92mq}&uQ=z7@#qE3&kyYcofv8Cgka;iPI?2( zkF`SHxgF;#mCiq8MNqAmX4#UKlu^jwK3i-@_gB;Yj3Fa$%?p|m%Akyc9E$yn7iQ-q zi4lXxRZEbfwo;0qj$1X2q$MfePx**mzMFcCBzchbkz^Gw4*3+1Is^TF>>!ggYE(T@ z+$rKPn}l8KM~*@P;!!TP_PBYN-QLNOpFY2*Nw5L62Jsizr^lm$w(@eCG#~Ae@``bza3w0(-ALK(iZ8uwqW zC$*+u%jjx{N?fD80Yz!+wApOlnT@8x3{2)`w_4}TX7!-}z=qr8uh#7_6lMTY$URoe z@n_d?kBhGBxX!B{2K%nt83ucNU{PhQ`o-p6RBvq;OkH0M$VavlL>P0$blI_otYk|T zb=-qpN~Pb8w~_7G)#{@1d-T&5#gzaBhc5d#<+m!QiPA1(8M;|+w}cw96}kBi)Xr>< zae{}OdIqcw##<@e`)u}EdP?7>ak@?w32`o0{@X=tJhcl?v)T`|%;=)%^egMZsOnTGsCuF|0%E-N)Y5o7s7bq`+}1Tua& zJ#A;Fzg(^nOQU($<1SCuF*Oz1#h=7)c<)|Lg{OBiG}kKJ-AKxNI9k&u*P}CGB`;2L^KmN3TC@_8g}ief#sHCS_0g_~8coB;lxgxgEt4 zn6mlEIVyb0SM83p`OYj!vJd%w$`3-c`9@Io(Z8QD*r0O`fFd>EK@JXarOf^@)FS#W zi^<_hWOt4k{^LP6lOI9~`3ApKH=5H$pc&vBI`n&U*I??L_bcJv%2ik^iBp{wh%X&e z#2iIqbr45GZL{CbvCGdjF zMmhGz8sLLQMGxu6Rl4PEDs3j`f(7eRSYe3;ayli(?FRwAio%OyUMFRAExWz(0!F3A zli?u7)brvU$`h=#J>F6ZDpnJpJFZjfVu~3pq$jo4uhZJ5ikaQVC-tAN)B8|MSi_~K zj0tWs#^g%a)5oXG`EN4kVoEqmrKhd6Z?aaWO1N9cr|rCMvUgBRc}JyZoU(5IlTjh2 zRPac8)@%Lds~8=n*exFv?6-IG4NF^Jq^hK30gcQVt{W^GJvIkSX&`f_HL0dpS9}n3 z0kk+^NFW){2f3saBFxi@ydce8A-a&!h(`!Z%cz7jH5HN6YoENZf(=mnO3a$*6zur{ zS*MFQJ$0d^wnl9QX*0@FspTqY9m9+yd6 z+~>EBWwJDr-B^PL;UL>6;rL|0GX7bOOqR{W7ZGdMT8p}GEQb5#Xq63QD_Ry@ULT`A zpsNq$m7s2(zOWtb&}+tYPE4(=WCt}|u&-C*IDWerGsZ@v-@eo& zJjqtX@AE=^RLN!{1gx-riT2^X%?V@s#F(#Py46SOEk$d+BivtAC+kVBMQxuVU5LV& zs{A94{k*Bgb?H7fVCs?>>1-e<+fUld>zI|JlD_exf>cX8JjO}w47!H3mB5)B~$&QippdU1M zR$YGMXgwr3;%aZrOL0C5+dq|X?o^$*gM;GClYT!Jd0CO3P3(@%rRpR^2m*Ze*gqgj zOj$yZc2C=&I`YIGb6KMWw^Bc}1waJ?IQylO-xB1Do}pBuc!0_1^q0}NUIAt zxK8^KC5*`mnDOqp$vb*cl=$H}?oRSU;gT7?0T5_EeRtB{nJAvK)AmXEzK#UCkvFZ12^&D57XuY;thb8 z3YaMM`J#kiTnCS~8ep;$U;_>W(*|PD28Qy$!?3}x0GUwI{Lzy1Y4R=8@90-T@zq;p z>?z#P0=;2&gJ5@pS|q_Fjnd!YAP%jNR!lo27y!#vFby~aKpP^G5g>*b@Tts4_!vHI z5#sRxc4~zvX5gqwVZL{Bfd>Y&N`XU!{X$qlhX}#HG$56np)_o8GWB%qY+s~BkSpgM z9LD^slJJod0;DX%Jv0NncmpOh;E^RvmRv&zxg%El;i3=x))FJQ`?0e7JaeFsPbd&B zfn6!nh)~smAOsQbEKTp6wnz>su1f3+mc%(o*2&GQUUGAL_`2bLG4BUKNtA)2oL`32-HV8r28y`Z`Pvy+(!nEvw zS01(U=-+yUfL|WIt_gF-1-|AYY_@caYzcKQ8 z0j~JL@R^vtR{s9&HlOeVs2v!g3Gtp%DH4&+iAFi+{Zc4UBR|Gf z7(a5Hroc1=NB_%Z*Ur=|&#^4~JpS78a27-O_g5JR^VZkW(Zs;7BouDbnSO90aA&3I zh>7Vg%2B&+QDT_^%vUJA0HHTS81sM-;kJ0dSO{r>r3_tO+b$)@DiYKWnzZ^V7Lcjc z_SH=D%ObZe+%riqH8uK?ZGB>t`4!4T0zcFNL6k4nSPKSP=wiu`W&$rOD7i*Nxkd+|$R1>*0xDlfmY@`w{lh}U|8 zW2vFWp)Jl@WjSrxWLD8)g=92AFy~Xw%vu>dd47UbxnO3wRECd5NBPE6`TL53fpQ84 z0*DxE<|sWlx6iUU(%nWA{9yqb;92=@jc5ms`rV(x6YDb3lkl@B=^<_K*hx;zGoF5j zfgy6$OGnj$Ho46Nab;@78a>3rD-P~=H7Qi&oA$VWg$OahMq@=pS^uInikJ5#eeG}h za+rBE3+0A#*M`^ia;46Ak+uf*q{8>}4ak97RJ69YN00=5ki2191WqoZav>86z}7Si zxj^TSMg3t`ec!YHN7|~P+j_*nWXZ`;(EuE|P8x-?a>dRlwMvS(RuV5{uWTN0PC83C zgSak1BT{|znhZ|E3fR=KjUNwSo8NAu%xu%&-hI?q@aG${131)%LDR8=%?zv|{tlN5 zQvIi6Pgw@(JrF4mfRxZ#EC4vfkLsT6gv|%_2mn;_5M>fT2=7QhfyLnYj&d9LuoIZM zGx!CL7>+^ulI4Ix2s7FB!Um?qXw_8270dw+b`*+q;wWotRSwo!`81BDt}tp zg(pZZ)33hh(hn=r3AsbTO;T1aR^9>j!-y!Oa{-8S$tacvSgO>=76;zc31Do&RO({1 zuIjjVDx39!xNt)#^~`kVgJX*w`vL?N*-E^q~<~VszEN@VcH>;&oQF%x+)FX z!}@}p5--EkjYGz|Bj!FMmgg$WA|tT79(J1}R-45)_@keH^V{2kWv515x<*r+Mh(cp zo>QY>kwJIJNYEza7;ZEiDo7-(1db(wC=iW?$w6`)I>Nojqq4^%JIAA6FbOo<{p7&* zRoHP4C@qEvxl@=bwyTdl4u;OAA~9lLaPIzvVu_i{(>ZPQ-Jv?07@TVC1-l0E2nW2vuzX5IyH45IK7`U zBcCvHAUJa%2UH`Pp&-Zd8V1QL&8YWHK5tH#<~5M*=L{m16?49X`W$sW>LWw z2>dlkAoLv{x;4+>JJ)KvfYSXP^>yY!c&G(DA6~V9?nsRNWgb6w5wCkuR(O$z7((`C zrbTdCO&IQUYK|Xm7XM{2A_mY(H22gsiOpK{APff-`fj23UE#~fm>}d_aQeI8GL#Ii z_XpQxhjL7s7smT0a9$>N})jEDcy;aWXd7J4KRLByVFsRd3pfY$#Gj|~h4U>uZ zJCz=sY9CyB;bK_zaw;*VhVRTLemGID6#=1TVW!0(MkUT{u*CFCC(&}c>hiblMZ&Gc zd^AjDrUeS2b(-lJZKlCaCP@10!m}=%)$8n}&|DX{U#ZG6W~m|t68g}(1rZ)4}r^nBU6uU@B|TEEWKI6(Nm zwYA;qyV248<0kjJgWmQb(E=j#!j&yAhy=2Uw)66`z}7wcn|Nma6|5LMBj30CvulRz zZI^;%uU$?FV;lT+6aaYLYKz%moS6i@Ze9xk)r5hLpMU!6{e%?;^3MDW`1OWSC`0JyFkDV@#F~^gS^Qp~Tjud@KwaSlkE>2L{ zj#*&&6KXXy$q zEf-`an1t>6=M6Jw=v94+Uz)iS&7!_K4JrIu@;k5jdY-yHp1YTDhnpp4E2M{cMe4qL?;Uv@Y47G19vYWLfD z+qXyZ_q7~9ckhiL6bDKD>dlPqndcU*p#9MGRuRVXO#vAxL76xHl7i@leFAohpMYeQy#(-`s} zD)Cl?iFc{pxTv`LkdH2*_lDyl_YwYBCf_b>da3u&9f*OqpM40h${R0m?3qao;MePo ziSGQoY7=@P+Pz~v0~aM9E!j%2*zf5THQV1^J|d7l6OuovpEHO-FH6w5qHuCf(E#6| z=_)*z^Lyv|ik2sVl#!Rmw$MlW*$zhx{d-`LhB-~9V)h@Bb(m1bUQ*Wv=r>5A|5(}H z#qhUhj{0X*`*t2CB}ntb9z6G~{Zp#}&EP-*lH*-!EaR6SFCG^$QkO@685QVMV9?O_ zK>$FxJS-SJU={pVK9W!Uzt(7v8MM5uYD<;u$YpX_Z8vGDl9#7qf4Td|s1Qpc&TmLg z6K2YbM0MwqtXig0PJvyPFclD_gP=>a(_Ek#gOv2Id}NEbKm9O1Pw8fmz{cN31^ph9 zFRw64hbM;Yx@DN$a*tNV>^g4TIsA#X1{0}F1|zz^m#YOFj{HipvBpI6?K!?lVV(%{GE?n#exD;y1M_;XlLlOhT$QuItceg;nLXYdS3q=Oho_|HT|X0u7s666vV6h zwbkzbK=LL)CcN*FB?w)5_PYP2vE6W>`{wg}cSN(s%_b!C{o)mh;84(ex?$G_)61bG z02fP%kk9@hi%}Sf<^`h%cp)Nh-(YZ}4Ja!tYK5STn9OFdeHf zEsE+prr{&(KN0^zq3F$fu4I!U|Nd+x4vM#R?7wcKI1q`wPqvk!EA6t8WXNS;PQ)nR zZnSBbJ==CO2^!-44t3ksI>0pk;{|$dxRcTtcWu2L) zFkYQ4*Lg|Xex8$E?#Naid&PdX`geMy-#gU}v1D+kIDsXRaK_=6EQR8OXx z6nqiV>OMH(iaCjhl)rD=J>53u8vh@py=7D!jk>MdG)?2sxCOVyHMm1?2o~I(KyY`5 z#@*fBgL`lf5ZobXAP^u}lD^-!)|R#J+2ibU?)_6^boGy}Q9r7BzVmtK{LmV{Y8G4H zqa}GWqLMUfche_uw4{8GwHN$GRj*P=scMVCw+!>-GN2?)om98Ey*p zQT$dsp~I5I6)XsqzqLL`Ao8^pfF6iI^EHX8Zj4l@9$~%Pz5xLh1Oa6@c7N;n6N(b; zTCs0U=GsX;{-Nuz!q2hi5DY{q$a9S@L74^nSE;cmHahIeBOP$LGw?wZ&z*f~9{oQo5bgZ`gopM#EVuRlkI4}KULe}v`*eG{RQo^Rp~M7wh-Ox%MjfoD zdto%MQT82CQn>yN55=LLpfeJa&_9wPz}YlR zgQ$~B939uTmR4x&*{Lb2vP}P8AlmZ9GQ)7L-72$367@K1`}IU=Hj*S0L$-q=;Xy9q z5YvLF3zm#c?%QhENrAiB*Gf5JJ1y-uM3|(u>d1P!bcJyg*2R*Eg!p#F@zzM7^7P?0 z9Hvq_D5SCgZMizCH0g&`MU{EGUCrD4sTu`4GdKZDaUquCLUH49&P7?W$5dMFR6dkM zd$_-?`VEDvIkS7O)G~-@~6N7```t)ECxk-{i*F?#P0@ z;C3Np_ZSYvR_B@x&Ty}?ssZ2V*Qt@@rQv{p z%pW9!7x0W?aC&GGxzI>-&64Vu-U_P#mNk`BUVouNDVrJ;yNM+RAcn9()-3VVey&?r z8nv%mAxi{rIuDR1Zvqj>HGqiw@sJ&VJotjwF6eXp&pq4HcmTlCCjX;EEZ)N7A)+YO z@DZem)zdxNrs3&0&#OQXAq)fO`LrZc_4%v<4@Zbv`d6<8F1`=@%eO+MrVub_ja~AJ z>|2BI_x^{55D@8JLeI|0m_HA|(wRH0TDGqj^-fA*-u<>rc^(xI#`)KO!<{H`6r8<8 z6S1eAQgJbAW!BHIRmg{7uM-5z0s!FQBoY4S!9FnH{xdoipwg!1EIt%77m!Q7p%%us zR)p!{Ood2KJcP`59YT<2in+EB$%cW%C^ucOnZ?IFCNpKV>X(hC`FDY+Gdh3@NUbIJ z>HbSnB+sitEUSr6lAVckZRiKb@R{RbLel1F7SJI2j&mqpZxKv2`~WXkU=*sM9%JJy zO)|BBmxO7p)!w~(TP9b=B{uH^5;?rZ?fk@-k{Br+T>5Go8m>xpPHTx*LV;r=dzr1S zXX#HuHL1p|N=Fs*R+)~5sCq({W)Z`h3P7Y#!V{ilrDxp?VIX&tqNBJ;Rj4aI)#6v= z&(TN|Yde(x36-G<{gM7gs|4LxLWyzlAI=AxNA;-o>qu5EvKZoeB6);A-+iaf^?-*} zpYA-qr!CNpJf5QrL{U`**5#4oD5yswP^4y!@e(isO+$&&mS88mHZ)6ioxNfY|#DKQvUM>HP=CiA`3 zIq`cKZhvLTCTKLqk<%k#%%02$_T%L6*U>`InT1=;u;VEss;P=~Y9tt_@}pT9RlhKO zmK|=z2Uc4uMdNdZm?oOXpV6^qKDQ8;xJL0py#tjcv)Ogk`s;PT_@^ea2N0l&b!N68p$3B$5 zGbF|)*p`M?QHcod&>WkJP^vdVyx8iS7RO0`hMDb?c#v~Q>7vr0^-fW3?)PEW{C24l z^3!H%PIHA#ubr9ObEv_ugQhFqFjN5qEk)U*l+HyuOT@4*O9xBU2;@>OQP+lAe_G?U zFkvTuTw+6W_-el!vFYqy$C~^)L^4BiVjE{pzhAWSz3Da4j};t5wZLX_^q_s@eKEYP zc7#-={fd$zq~TT1AvkEKSA z_a-7|`G8dAl&?;-q-%HQ&WHh@zZ?c7e^dD4`vnt;Z`4i2G*VWGcjzOM zt5);|C*KPaoGoJGRNZ`IoLfVujKug#{3=8*b?mSE{thK?-kc@Wn>|SH_(JL8n?aJ) zyjB@tP&MBcOxOHK>9xm2JF5mPRt#sAnv1T>HwBN~*A+s9yI>3$Zw?|(v?48+Mi zB`VCBM*G@~YPbhPq}3g?4x|ZN{YD&j_c3ud?MD3vh7Q9PVi*1I)^kODxd5YB9rn4c zfI-*#%{iSln{Qh z^4&RGe)KWX+6r?E+Lv5(D~H#lNuPt<+no2`YsC0vHN2TzmjsI@;eG9KTPk;xXX@3~ zj5k=~gUfo+2wvJ9ZRPB$TvAJjzpQkOktR}79!$GzYo#6H&Wai}<>Yg}9V33(oCJGm ztm)obi@h!{wsMee`Rzi?VpLC;xik5$;fH%e&sSU~ZtJbtTYTpA$cnU&Bx}WjZjU0Z{@&=S;@Kb zsUE*V{PzQIUAqxuq5w_HmqrEn&lL-7EoXkII} zqiH_V<37yeuFG7$oeWND5x&qjD2fZ7uO!hjLj~9n0bXxBb{GP>DuQP+y|i5fBO{3% zTlt(mg}i8rGp>YWBZl3FsSG0fOD}+=l0qqP_<_^{E{9>ROJRunVXOjSZ@j}fCj{5% zLT5AOt*8Jr)B<4Y22YRjJ!sqPG@9krv$qxn0*gi{}|JZ65Ajj z+oBuWdJ6sk#Wj`(^gYJv>yqjT#Py%X8hOWAWW`N%##z{6%S!>=7pcBGV}Ex>U($_V zwT)l)#{LP31es!k)a24lLy*VI;Shje5O7-syN4!f38rgzfpDMbc0W^5WYZaSB^%qN@t?)2 zrKDk&nA_N;*-l3AJf%{*q^LN&{$z;iHT)EDB!46t56GX6VQ0<}V- z;aI=HY60~yr{SWeA%6iWp9+xDUEj~_Rqa_`tI&u#EJt@@&=chY;Ea5(fnY_y2S}u* z2{oCOI!qwzW->RS`9Y9kEdkEuI0iV;wBCarKK2An4L;2l7LCozk(}98REmVID@_<= zq7>*YNLH|5M~+h}Ul zP%c_z{5+VIB;;S3sf626-?g*%h_aWq2VXdo=eD`Nv`f@}l$tIy5)Lx<^Ji7+^s(<; z)okajZ6w_1mfyF{M%T>kk^O+wg=SCfTl>zXw#f%SI65^>JGMJIH5oUh({izEiMEIA zreh_-hw`;18JS&?eK(XV8;6CDfEz*y@{#=#2u=oYy#GiM(nys{I3LRT=8dhf1cpd-7oJk_=xU_i7|-n(4QLGkJ~$R z?>Dr~;_>QNKYs5J1#5ruq6k&Z(DgKnV+K|1fi)^Ww_WjY{9{4 zG~DJ<%-%|aly_{1t*oFsCfQylgW*q)_Mte#%%OTicp)O%h7m;^K;3QNIxm?%&Vk;(Ura!+rJv2LIZ4%Folwh zQGD&-vVsK8Kq=r|4s|U2u?`-&?*8G>JK*pyRKVRsYh%ZlG3&>ox@Hn|SpaRV z?aYMH@ud072YH>2^sQE3)-ILuDMgmf5CbFd(Zo{hL|^wr%hn|9*Lc^($AdAEk10i} zIcZgS@7q{AH0`I`MW)`on;I30nSP}-`>G2|Y375~OjF(%ICqw)N8Jv6wo{p|dxm4E zmvQCSba(I6AEC)n44xk4MqJlfH2S$AvI)zyu8zER5|%!+>GbyCd4;n1K4$9i=q}e^ zjcMOzMK?{Z;ua96L+dWuVrNFxUUexn&$@5U^71TDiztXIvt%6vQHwM+D%m1zja~N) z!fh5W@)mv2eA(zPAv~fa{|1I4{tbpA0Js2-2qymOHz~_dmxY&PusR-?2uZD^33x zk60^t{$Gtp{v|jwIVSNx#v_bx6Nt_4PlrEX_D%=%JPhn1L*e}ieZ3L@#D&_o^MQRY zKflfAMqw5Gb38I2E-nPnFb#R5sB4J6`%b99e~0vK*i{j9!7K)iDl%rrH|EW&^?GECLyqL+Q4*H@>R`PKT3rM_`FI&Yqzz` zm8L;0$r)s`JI*JPQebqUO$jN11y*F0$Li`Upe5p6b!}#O7Qh%2W5DAK$&%V=l}Yll z%uEG&$FPEm!Tap5g=}cPUo-5mlAu+s5tFu+`BLyHW~WrN$-UgP$gg`ftvf5wI)U@` z3&(&Ip_*!R!uKV0pEiEhlrCIly>IRM^4_jt&04Yk!@$-L`?jILZZ=0cQhWQBa^~r; zopXWcbx}(#O1U31jbm9-jwUa1diI8Eoj#0`Di*cqV?CSa;QZO#AU{KrAA@3>pxlh$ z%3*Nzx-s`U1)u_ggU6|YSY0RR5?LMmNgF+<+5QGir#S~$-DY^_S#Q}8%3f^>cVxLv zvxpDSxKB$B7`82NQi{4Wk#JPF*<$S8w9dZPQF;9(e~ou)sXBi0c2#Qk$Hz7CR8hVK zi$Yp=5=4<1@M=}PL;JM(Qh>)c-{ECUrt>&F-g8|9&L*%Ph8!5n;(*=cv0GX1|8S^1 z?vK73m&#Wm9Gsi)v!6>D=zWl$%V$07or&eMQ|vYSbROQ}*eMLtEO@rIP>k<560r#M zI1-r_^RuzQ3Km=#Y4x8Lg>Py8ytcx=nmYLaXV2L(4Z1}8@$=V@#Iv$p)&RSFiz~$+ z9lvg|ZzC(e3w{GW{pqvn;Pii)y|RVJiuR^n{dvjZfNq?d;UXO#Yg6SH7N|Ek2=*H(nm6)RFj^8on+SI7DO zRV3GRF+R_s311fh7O>8@A$U0gt-jRD@`VG#ycK!o%^|p%PL| zG0DVkdW4Z&cziK9G#r?VZMvi*3G*`*$d!j}*TnQ0>4GtQ;|g=3tgU&cyk^6Te}kc# zJahhx_H#`C07D=DOVw7D<0*_gWU8DCbg6I+!mUX8We&K(pOB^%3XUQjDTyoHqNH{u ziuKkq1Y=GS@-;`Kq}-L!liO+FmS|)=)BA?nRVgELYep{KRSYN%N=Q6IT+Xz3i0S3k zq!qOaqOHW`RhZxAFJ@K5+)~F$#%Pw3zBBf@^HQ8;4veWGsQ10MbgMp=lt^2s40u;t zBALjm>9yR-{EjnLN!O5!Rwj1YBfr}1!k{aLpzrD(FWC^o>fI98KuV-eoPDjyJzjbX zAE@~4>ln=w{4eYKjDT1q7L&@B!l1QJ{&0Z{>-xoYoUSiS=DIN!>*ZO$ObOeHf1hh= zE2Rv=BY7ltFC1)V88<7+JlnuS4uL`$fSsS7Octh!rsb>a&%{LaN(e)MKqdTu4#bVl#+E`NXSME_uT>B} zOIb}1lwmawS$>*8>}oBz%>&}d?O6bQT62;Ax}5p*PcP&n5d7KGH^SYYlLGHbV4Nsm z&Y{JO1^^zk=}d`7aw+fFy+WAnJ%chJlLA2s+|qQ8U@3{H=lr_Piu^)nm2{pSKk`1wE%j`9fr;Q4LLM+kd=U&tUC?b&VCdhEMA_A<_17awE zcwkA8OX}vE0`T8r#s+T3@Dyd%K`ZM1pvYn!_hstz-P`D*)gF)-`3Zx0lh5u{M zLZWjRv%oo$$PmL^A{bqNFO)26Z*6Su5uKggAz@$-lLIP^y0Nkw$yp(B3P*s1`%B{m z_SvHYvUsVXDsHZuy~i?HPqUMvZc@C#sN(Rh@-Qs-w|GXUZy29JLR`3+htY3TfrF^* zSgvL|zYjn4cdxW2IPkk8&@uF@Skz$_g{E(tf~0X`8F1Wku!k<4#l5zs<8R;c6zzC* z2=6>JJoM|Bq7H27Po4=SjnX5x+PqEPKpZU@6*g7_Tz8H!Cf-gu^ zaLDy&tXy!x=X})P;T{Q4nHYFpiT%CMe&hMgK;V5xU5&qf% zUlA|m)0F0)-=9_j$YdKjld4_`unh!cD>~gO@Z>E1iSiZ7bQ47Os~h)i$_#E9cL%Eb zLAZSPQvCY)Lv;E52|N85_rMN^!B`>wH8(1j8c+mtW{@$>L2i@~uYC}?(eBm3XEq!^;XpT{O23qoz%*k2S#6+@ z08W})&=^12(<+C`aF8IQU04&S911F33#yS1u08QJiMCUcK+uG^n?6Rg$;Y;y;6$56 zCbsz5J%W=T!H#=qgEk(9Cw?SrPUDQPEaVtyp`lXVtcWy7AG~~}oUzwzvEWUierxdu z6;2yENSJqMT59epdr>FeVc`=Y+aS|>FDP6+bmD*kHbJE!2P5)AA21Ei$O@2O@V;Qlcow`II zSfm5A79At>U<;UVLt68#^AZ8N#5U^0R>($Dg(O+iguI8w4P2uiT43wi#ARe!00rI4v@9; zJ4!C5Tiehu^Qd=r4vwbcBH8J(kLe)Ew>+K@%WQ3%N$QfSnAi*vF{WU{#~30(z(Y|o z!DME$ULHR%Wxxx_=nQ6X25-1!-JAq6!>e(NZHRNPH9cf-Cb^<`*P~HFqbm3vF`uw~ zC*l`7a&8zP4{8Z@u(V(s@8-#9O~F(U5T+yOFf-;*=;1=KTC|@Xjw#P*jT$4Al7Zgk zhWivG>g8^X5-DAo={;HeYCUtMi#piH!;r~gCc}YrJbkbuiG;v8cej6H4D(J-#9Y6qx{NWk)xE@gdltZ2YnbyjdwB^ibQIsBd&%1p5k>d{HQNjEOTkD{+3VVXBT}{G0_TsZJ|qt|G*NNqKI1W?M&PBerC- zLaTnBd6usiDr9BE(gI6Lju>&D3zvT_%UxwDM3Gt8Wc_u1egX z%6OlCdG4}7nsNKNLY=<>zb6Ln)y&PGN$p$X1FNZ-sxawBs+RyuKvH0WmCb0ilTgTh zCGOS{sm>eEV5!_}J9<;NWc(O&j!bteLT!LdR8=i%J&jA?at{5rb}E8v;^jsI%?W4c z#w%Ph)-a)u{BjNCo5dZ`A8j^3l$$u}e)->G>fhJ`+oo`PHmqGk3Qh&Bp|tFdJ(}0a z6won?pOPKoo*jTBz#qkrGE_~0xoJR70ADu*=O>etHKki@r-ZETs&9STNs%?cmQ}xEn9JYn9$Ky`Z-j z0Ti1+ioreIwE(Gwo`Ig8p^Kiu+8$sLNV5rqSJbQ7)I0ar%jBYWcDi?(CH6}Xs2E-S zIlHUG4^qF`R}WX}smtwlmF{UV=xI;tZ@&P3*aJH50j>A?uUQ6eUi$Ba2YxCII6=f& zEU>qy`v7s|8(f2NBl7KjgEhH>C~<>GaTG_ZsO1uYxQ1vD!y$bCAwt8U_p(FQ3#jB< zLlq0C)DCDy3qw{5NLCA|j1I%lEmR;xMEWZZn#d<~K)g}Uf0Niy`My9J_%ur zh`kyS^+#@mbpNSkat+f&e>b=u+r8W48t}x1@5W^V zS;s*B*!*t>`O3$T>c|xw)E%~1UVj~N|21$x?R}5j3^v30X2*s7NfD$m8uE&(ED2C^ zF)A9Oo&)#mGSrmay_hJ?#Wk@n(OAN{IZ`gNACHZLSeW;CCJlP3shj;kZih^@v#6J| zPF1n0Be0I8Ygz^k(Ok9?^9}J6-%!p^c!!ku@YT{(yH9Y%5n8fxn8azaAOO@(>KUfG zvX#^^aYP!nM#sMOVSJQFa+x#>odx2kkJfPwvCK_<#Fg>CChl*@WOcvGfhH z%Zl58fqNsu;wCZ<<(g#tLD%pyVsJBeD%rKu) z*C0&84>u&3dnH}WsJ;7T+%$R4W6c|>WsdlM4!aNUF^}=LM)={dO>piI0 ziUFUU+wNTyby>tS`YdpXLt0;|z!gQY{dv{~iYdA%s)8djTZLOcuLV-)HzHKZm;Wjf zimH+u*s>7lx~w3IFZ7!ZfBTEr(D-R@1;^X%B$aasI!RH?hj&qWD z9BaZoknx@)!X9`Bf>^MJk%0Gha}Oe$}7bd$)6FW_)ZNcuaM3 zWZQpiCwlDg=h&Iz#8vFX-T1^a@Wi{|#JB&%f9K>K^{f=uaiG{~1>0$O;Au2=!CS|I z)3}|}KbEHpI1s>pE>AZA=m1B6pnuHBoG^+1n}oUMlK(81i2;iLPYBW5Wa^;*wOj^k zk2g3?vPQf7pEI&L@R0A{33GZe=9SugzW;#`WwJ18*fp>GCqg7|jkm87fsRKp)7fkh zZAXL`_p9^&>x}IG79pzaiG3@6QiTwCaj`qUd%+SGslKusOxh*27s9xmNg0arn+QJ) zjRa+18e26hN!SXXjw*~mUpFZ{#m81HiUv1)KR~No9`X*cxk@_#zJ72Z+ulY*MS`>) zu@HILMN1pV0w%$MApS0)h(-h@(Mc+DKTZjgFf5nJAS`DU02F{lRzQ8=V0i|DJJWHN zF%VMWYYE^4WIMD^9_Jtu`|L2FTAm%}JJn256!<4}RfwK{d9o_>Z%{Dz3CDf5@pW&= zwk`H;cz~7#mt^i1r5^i0eX@7Ds>%_QVE_O;P5*s~^VQ^Wb!uRZac;#Q_-$2jP0G_e zN7+;j6p%hga-LP{Ir+Y@TCuw(Kq0E$|XeL%{rHcU?F@R@^xXkBeOJ6SSj zhIihfbw(KrAOrXvv(D3uGQz|=(vP9$9xwZBcDLBSb`Zs;QfSCIjPUlu{Uo{t`OQi# ztoJC*R1PHZnP^fEccz-2OKk!1rjBkNgd>==CDbSi027A?BkfS{!Sg$I18y-N_q=V? zzIQjrbAd@plZIf2WzHI7Mu|>RXyzFffr2Ckbk|UEC;EHIxN|{`M;jWJLdm9tJM|zn%7pU@|gcTnWvv?Q29P^$4fSN z9W&kw#Jvti#4rYTA`imP8hq>b!&7L+J71DZTyFi0F0`Wr@4LQ-edjSgm2mFRDqKlwK~QDacx!YgF< z^HrnOdJFMy<;01w0CM#%RM?2o{E%SkkYPk=BB!igc2W+++O-nGfKF1v;wD$<@~c;` zS0!ID+(eHHlfTmG7~iN?3(Ixh57jLll>$2lfJUUi+>lU=t>ge9-Xl^3)4oNJOH5{+ z$wl1S$o)SmXCz)S;$i8RO%d)jxQ*=S2z;bL@4t>BqVyT#9sUud zzY+SDSrGr*q)P;I+_sT5p|ENoXQqC1DAY%=_wnW?5r4*%&xmM7a+r=qkHqrGxGl|O zEGg1-MnXJrKQSmSG{?rWNRSc*;MBm7`Z!aBRP73vK{?Vu{BwZ&JwxSJ_q=Y zSW_ZAjSah}GO)MAT)JKr_#QnmCMt2raY*S zj>LA$WR$;eXrjN|qWV@zL1p$`an{tv=f%vAOPfGhEWRO@Fw9mK2+3lLVCq+i&-TD zxr;B&yQHS?$iV>&V$rfn4F-TqHWQtvX(iFG0U*{rq1PjBp98M6xRQ`idXZ#%T zy!&W!kHbjnDMjF>eJr=fPaeLxS}5uv2)u(MP_gD1^7kt*ivo-Iq7yLT$eRKR5MyAfRld$7#k1{}Ts){oZ zuXk`ETCA-EFaK0>U=Q;hJA#0Y6qwDeY$nSQP=TCe7ulAzgbRput>#U}{CeSXqq3Bm z?;om7w*IBw`gNVrZy@dN9wz(gup;5|ZVV(81rvp@GIs$$!yB^3U>zEhy=P?|mc685 z`zDq_Cx)>4&6~A?m_sS653@gf*>>BfsO%W~H+KjkLaaX~uh#!vN58Gw@R3AWEnVGh zB=_!XbcLV_#PZgU=F7^0%5m;{KwZzum8y6B?q#S);SeT?)TJ!xeX=>n(KIp>PjQbP%+Xdpw-^I=SH9om$5|+E@rl_a z>QI;xPZ;g5ZMnD4Lzxs#M8SFVeFYO#U=8aUVL!l}qr?h&%RNCZ^(OL@r%8m&&*z+% z9OnQQV-WdmyYR+#B%d@Y6uB&U&GCeYI$XBK7=rd_d7AgSbH=Ryx}4|_2oj{QjLIDj zyY8bRPDE)hr1H<-R>5VOttc{sPhKAl)r5({BX+IZfJi!pqAr%MoX@X91Z~Wdm2N@1i zK4YGFVfR$J-&kA1PHnVhKowDy5`Lnx3*gj;O(R(*(11suErV?5EFoUzY4 z9LxAauWW!-$U*eQzF9*-WmfLWC*ZRaH25AigDLovSC~SkeM}VIVAgI%6Y?6a-~$oi znhgoc8!~99gEuyuD&kVk_j`vCQ*kJwBq%XC!ph1yNjvz;6FhVhB~TTHatgU)B-%!% zLk9+WNos7um-Q1SPyv9bbL+6M@er*}LS|Q7Pg@dSTU7&Z_j86A$8krC30$jFBAcwp zf{sXMofx`POSiusxu0R_H!$+Ap;xjoq4L2WkKKiBNhG~LNb73I*@io73BG~CN_ApN zDs8m{;!<5A3PWO>+9Kw&;=%Gh%*inU%`vdzxSzBMPe(DAmGM`6u{$Va0-B;SUOfI# z|9*i)WVz_RWKHv8BEX(#@Jb@zVIudfEUE--Zq3t54fx@UHia2Mdltz`=bJU%q_8Mw zN7DedVxzE-CGl@v^q(o56kbFx-T=D4Pu!AWJsOIH(2*U@x4SF1cRjcNDR+=0 zZ&)yIL@#g5CvPGlSAW<782DQ|%!e^D@hSucOZCx0zFf1@jZYdwGGDSwZo;2=n_ z;7G6F*r(t$yWqU5;M;n^)lGd|2!1}NQ;0%MIikm zux}AcP7zvn5ynOl)^ibrv=~>Y7+=4b(6^X4rJH5Wa%(JlrF5|6>}T(b1A3aKQxJE0w5gA>a2-= zs+tzrITY4{&sh8R@6_uI^W{>F&)T9w!ODv*e`yi|GVZTab)IiCmZwuL)KlqH5^|I} zo6KM~@yy!xcbEI)Nwny3Xxelpbvo({v%T6aC-aJ^3hUQMrxu;INP zpVL=Sc5si4QkO0)VbwlT;3%yJjrQTWK2w6h@+*SCwQaGMD6o&-W|I*R@h5klNuN zH3FL|6u+_+O_kMZH(tG8Y~cRA!BXGOFvUz#=v&3o(zG$+U@-33Or7|hz@DxF3AlOD zzTv4<_tA%$OtbY6?dZJgh~1W^lZy~q*SPhg=CX0O#s0Wunt|-1^_MK!MYqPrCRO|P zm_btqaG=ZH;55hJsuqjq6;3uPHY-;za^j1l%DuKodF%7}^SpRbirM@`sUpVZ0QS3! z<_OTZF!vN1FFv z(S&h~PZEA;`%LwTxxO|IocGa#%bsOn*&hB0bq?(2-=2&9x3 z|IN&3Xe*qxzjcMt8DSRN!#p1UU6^k%)*$c0774APV4N7^_VqwOWzQ4r?L67{{Y{F7 z$J4tHzdPq=h8w;g@pirEyQoJk5Ip{L5J>$E^Y*v%srPfj&(W~LhF;|wvR%_P=5cnx zqgi8iwadAO3gPcPsl)Go*7Gp-w;X1^{quCl2oMXHaT&^w`W_VZhrwb-{OGB0$CbW1kslY@xJmGyDW!Or7#DlKNesp)qe51gkRtt&6s=rFLo_)dEA%5dDXxr; zQDIV1|3^yJOc?_|OwQn&i>#4Mur%LCrq;DX!j7geMkg0!gBjcOdVhM!K&d1H@05sK zXt@lqNeT9AIJM)Kl}vcGG}U!f8aYC+Xdn&5@Ut?8A%^FUOV?!a;M3p$=)aQ8Lwo>xhK`4Wa##L5Z8T2*E zk}MQ2IPk8VDuPLBl9TPv7@3qEB7vHho*C~oJ>FUU9XDSkup?sAgaitZ@ zA9m9B*k%H-o=Ya7c~nscSIT`kQ!lL`sJAgH_sjZlX=tCidYU8ng+v3)hor@*f3|nlI_0;(v;_JvTf%y8BqXzn&QIWUq=NysMPu z;@pC2tyUJYoy%t{nREzvDJkS(e-I`G2K^eQT#s)ITBS18h{NyAzCX9-h@){=k99pv zW;R1*F<;Wo8vgm(%$~n)8;-Ts&~<Tx}p;gS+AqMo!zK9t|Ht@THhWq+>#f za{(zIM=mhrn{{T<0$)`db_~E@-;a6)71oYEeEGqzXLBnMg~O-1-xYgs>s0IkF!mu! z5+fK5pBR96r1ITo-j5bC!tyTA5u~d;!Y(kagUJ9Il<@Xoh(~cDZ@wyv9_@QxT@DV1B1r` z16^}=3D-ou`$#zO>@+v-a7oT!R8qT8>6`^X2}!NQu&<%$rrkd&nS3U}(74 z_Jxt0_u9aYTjRhU5r=EBQ?{1 zc)q}y=QsfnO*NY68hknuGca@{$Z}_N|NNAJ&a&gou0%bEd7`;IDH2r(MI0vW!? z$exEbfnU9V>{NbHApzLUey^HYwm$P9h4{lVbwH0n*LVQ*=nz!mutOqug-?M3k|=eD z0fh1z=&#*Kp`PLoV1_-gMP`V6bl^D=k@SK~XC>ZfMv%SEd)<42^w+_Q)FI?JPLz@%*27`DIwt0fK2F!*0cZ4|&f#Ry;S?yLgrPyt zAvmKC#N&*n-4R;Mx;mU|Zd{M(PHeW#ktD^i zT`%7hFHd$(ZlrNv?#B?wa*TLNbfm40z{ltmdFO{~?3A?#Ml~4E5RS?t_u>S7ZkBF- zRvh+&_uN_nN>q%1JkEP4r*tb`nFQ0jXxnGvSa7Es_i1c>gB*{kmB9bmbxJONRX&V3I{EXe$DX(G9&IvZa#A-q+-)ME;xvgQI);r$f@%_{ ztcZy(-Ilvr5AYcD+dCK&F6cA#$iMEE^v*RQBO}%NBu-k-6EZ2mmaR-^7k`82#!;ON zfTfzQCciu+D^jP}v?G=85uVwmK+h7es#7pVkVIS((>{Y~i@ka5H1yVtP_JX1vy(-2 z!1Mrh$ZLSQt%w3Dfdp!xxNgdY2O1{OT;vw<~E*zn-y|f&vG11 za~te{*4JRX5EYsu#o;HV5y8AMy}XGtMMP)hI(Oumv%LAUyjha`&rg4S%r^~JEP&y) z4=Njj6os1ex#4sLc_T6%ZTTlpkh68=iznsjC&=7U{?c6ma47Fy@Xb&f&DFZpkziqd zHO}4{1i_c#tKNTV5-Wr}FBumMeVH2+IV{qD(Z$rK?b-yMCpoZ>0%uk(X9+ z;7EY?2JN0;vCT+jkbV`XOqKspWuz}n2-=21Y)o0GzU+UMN7R4GqjSK2!J080Ne7w! zJIm&8ta(e``oCBLQlNj``nD)91F!ficeNO{X7JcE2>Z02%xgrLgH^iQK7X$G56ea+HsSxDWnOeue9;SGh)mF= zkOukxT`ADNZGGGRdnwRAx4xNHzWoHeWfAaQf>euJMol%9>BGgb}5&mUCaoXVely-Z)gfAf@^Z#dj_ z?CJ>YHq7&5k2+Nk$E`e3k2LpLC?ZBh0x(+2$SPKurXUXjDl1~nYiepp0kzi%aH@;C zww3o6^&d-+0Zh3`pD(Hgo=Iz(2C-i3NpdI9mNUwTza5)xq^!OUf8=*D<$aPEaj_N}@1{(Oq1neJlKX%ddr)Zk4{^zBCz zN~{e*N1v>6!wA8q2v--;_bEH+%lE54#$_mbYx_iQ4O$kTD)O$a=gxWWrjeq0o2_44 zO)W~LBJ6#bGYKMV=uk*axm4EYA@oqyB)V+=Qe7XQ2{bNr__63I9PoNs{v%xFzL%?V z%sKxggK)zJCl19DzB~U?or3)>hK5suUu~&`Im}bE?y7%*^l`sXTnA6 zk1H87_!Kc+L}c96SIJXZMC5mx)-D4S#QAdta11Y4#ENKz$%D&-suhAqg2X5@nZ42MzuDngN# zA0}eEenT9G6~TL7jEpo90SciGwX7;ZF6R^_{WUM?Ix$S~d>ti$Q3Am?8~H>{dK0b3 zM1z|-AyW=*=Kt4V>;?c1U>K0{ulo-OdEgY%tn2@7j`Dxn9!9R{wj@!CIKlfWSnEtZ zL=~o3-lb;x1wp5kIdAF9^4&*1#6tf-3hM2 zt$_rG;O+qe1cC++G=$Lee*4=qv!`lL)%+*_$#ZbmbG|NJs{~TEp!hLjnlHf0@Q*b7 zV@SbzSB?cdM2X0+g07~}k*!WiFJK_akij=^q9Zd=r@MMPr6+LwhZbGy*hgReQBg*% zOJICKy6V$iNfN${XkONtdB?K8_Qmo+!lv{_We!#r0XdtUr%qmedWW*P5s*NXS(7uu z>Bs;r`ob_26aB&}K#FDvnr zDi?WiA_pZpUoeXo{|E}3X#j;@opOsxOntKqg4m`0#jRzOsMDQJz7bX@z7(j2;ivX&+%6z3Ck?$%D=J?dHwzItasY&L2V^3@G z8ijOhb}xbBCWVU!Y@43rvULuXL#sv!<>^v)zNgbeeZbrF9AN62?p$ZKYs%SmK+UOJ z$264h+=JB3&ei^R%%w&3H-W@eGmePsWixTgGeH|jr&b@@}2f%h=j+}!)c0bZ*ZrB+1cNFfrK!?b&9XWuu69k;J2a_aI!j#skC@j?x!fI z803#Tjj=Uyl!~fw74aEDE4|19JJxu6DMP0uO3W1=*U z7qIfX&HpOMME`|NR{uq($Ti*$%OOuSAUUmQ4x5GiI|f42dT2g(yuI4&s#{s1ZZz#Q zvl_GGt87$fWwE<2GuDT0O^6?RDob#*>M0} z04;;+r}x$N>MnY%#RpQTfABHW4vWoiV>oHT&6Rbr>ycYH_4_f(H`@ zLbE&gqt2`ZcJbb6tF;CU{M=8Ti>kd}AHj-1s0i90}pJTIe1 zj%Bzr%iFN=JNB($xW!dnR>Q>hCo5@i#J>#s{_w^Se~L|@-da>P4kY+U!m^nDVSs`0 z(h5Jp^#0VpYbndhM0Zz@lZ^E9e)L{YmWx$8Np26ftEVqt!uM<8&g#~`&E8$77C7ZX z9UQpVSdxGCv6{$`DIBk<(l>|Bp7h4n8aOR!uw}8sfnx%*zcd2Rb?2=2M*PEh(McC~ zcdZ|$=qwwurA1NuVt~jUqYcir&$RuK$y+iXNqtwEHxMpIX);fERmx&I9t{8m0f}ID<#;nE>K}TRBC}=k$!>;Fc?Xwk z-gMOICY>@jKP@-f4}rv`Wl{AkSL*sYYB{7pdJj*_{q9oDIa;WK7*T|Fo>t|$UtG<) zo;I4@e}6*L+Po{`rBj<1BFwjRaS#Bke1G#EM#+1@mRZ_<9BS28@b?sTB!A`7;5X=v zp$TNUAiDFVqp|mki&u5f_fO3Ctm~GUAAXpwBw6@(8vebAQ7u^IQ$w#F{Bq_lBE3Q? z@_wu<$qQmQw#%Uht0kO0PS1EgU`FH!HAwFa1wVXPdw`Ehx^p?YeS5uz#3D*=R4kMJ zP=NFR8zqWbb$6Je+~>`NPt|*#nQld5iU>SSNff&}n}*Qdx6q=7%pS9&_c)7k1T9oc z4!aAKd?PlJA$Uhm;SXOcn-HkolZRJ1r(MEn78atF;%S=@D0f50+raPya?-1+5>bjEl2 z6zUWDy=%7bn((8m4sV}Bw?8ccOp9>+E?`5G;cF+CfH95Tp-wnThpM`07Ta2vX-oNsj?=3xc8sKx(`IU0$HB6p(Tq z@(K>2CX1kffdmErc?Vg{W0()1*bShV&!cb|q4FL_2(U(S!6W$AF&OkQXKp0uGZ4ab&wy)9omci4Fi-$8$x+=ZnUT#Kjb|%F0+`c8fv- zMCpCWXmm%&c}Ae4MoB6GiM( zng;5oxF?f=vqoX&rD9LWY_4M@8bxofCpnZ674XUagrya(r*SfdU9hIik);GFrBsq8 zs?0-wBlx199HTh%WFmP1GJtdh8I(ae?Yfl2O^aCBiI+7qT|zlgvQ|chS0`VFHIH9C z%vRp^K$E=!fLsEQ=G8zFvW+N}>>rR#MKLbc&PI~9^&gdWKLKMhIgnoyb1F%Fccgaa zPY$Zh0a{`?8_Pb#kP&DTcUfY23Cp^MyMsiyQA?$j=$yn4@V!oCQpMijUdIU9x=})M zk)?CcGumbHQ1R3A@{98F>mR`fN(HsuOhWq<>Y@cb z|F5Lwr@Rgx6|NMk7K>)VzfK+pI{ZuBa7ovROLx-DLN1Xx6cwudx4I!t zxW^ogOtSw<>hx&6)oiH_M?Cv{pwn$_vZw%m0MHw7iQCSR+O*T>i;NY6Uh;2sL-;k! z%d%HT>4Z_#E+LZ(nbV^-SnT#tJ=&COThD;5cEO9P9u*T%?b6j`GhCe1c6IvA)bwLC z$JO1p?_d5PIYN733#R^d94x8|fj(WGNva}~Ya+(Qs|)Fp2}OEV{N<=I2vX6|p1_Q8 z(lP6?69YEB~I_Q#A9rm_hv*hMsLp%CcD$lkRsfG)@)# z%;dPI0GK5)`=H8YSn!TCW*tA|0iV^tXl~o#ov(iUahcpnpb~8?iXS>v9M4ZB?e|;1 zR~ein-5tJZijx>?C7$!GQHLs1J04_!(vGu@ERz&Upf+L{Oqt~3YkP@8JFvstaK1Qm zy$UaD3g_+LV;Ir46dO#ghB1%S$zG?dcA+R-(UqPUhj|R?E!_?^Io6o$)hX`$K}F>R zB&c+XqWHDPCQ-~MdqE}=XP{fa4DVsl+CpmpA>ITG^^ zt@6P`=lY6rn967qN70WkBa`?Wf)(Rau!pCtBalY^ z%LM~M{}jLqV=DVfB7lX5Z>l7~ptBHZV7uSv9U#k3p^m91mxwo}Q5fJ-jbZ9g&tq=d}OGPdF;0aOh&ETJENUKAw@u;^(y`$H<^8z{R@AeMK?Q zNpXVznQt_sTWU+woh^a8xG?2(!S$9F*%07R`vc-jOIf0Q<@rDgy|{RAsJ(0_IF<+1SK@V(aX0>plK9d|LUY7Gx3ydI>lJn|xMSEe{~EZ;v(;;WEc z@%C@(W-jW|XIp5NrMgR%Cq1CZUMNOs=1hn1c%f$Xy*Z-|h33%adU>Ldqc@jpV%+B1 z7h$0y!)@0{n7DfD?7g9VgPXY?6IQ$ph z8Gq~uDujDtkt^IyPrh%y!Kpr!ykt%Lv-@+^e=xAJ;Qa&TH@vV1N7Oz~6^2)P@KxJU znjsIuz*oE9EHXZ^y^JMMZtbul-eESY?^GXHZc2k=yGN)ryt6+(?;p;JN)uVQg$_Ue zc)^AF=epf~*?rzQ;tq1Neeo8KFO;sy6J1CsF3c=9J7OB+s}G_6T@w4eH&ZJ%_@EBu zOB_1=V(|!Q=XM2TS`)d7eu#bLn_nqal!M#zryy?jzM(#TjZ}BMpIady(XAAlY}kLP zM$C_M2^RqN%{!pee4H?)z8U29TBsd}0`isuku3)R6ayes-cqDWypUsE9`Hhe#a(!O z$l$Mwz}fPbd$pRO6QBMM+vPhCpPL2%bXdDCVoM;`S^Q;yK2qJCn5 zey{nzxksaY)_L_&lo{5C4TOTUQR-sDNL#=k*Me}Az@#{t)5b} z_=LPDM9rWxv>;|tFpCk5uptEB5sU*1CEjB>ZuR12g>Vf9qYDCb=Yy_~LMK_mgdBr~ zMWf`5q7=wD=KZ6{wdt5@qA2%*G>%=7q{DH^q9v?DQ5FJ~9iu^|p?YLN(&N$Kw4k@F zF;R5k>a1WEvKVWn(533=N77i88bY~w0MIcOM=Jz7HA3GpCR7xkb|}W#C?*6R8x!!a zlgEVGbkKjDJch^dSh1yw#%CJEzZ#6o&W+E9#TTx}7sKOA$e?AS&BZ6+jnkO}(^>P<*}Ky@H`2Ku(|O4=_{A~=jWdJ; zGeimVGQ_(xUT1rccsO&i>7HLznk%8Ep8LczQI)W&hM7bFQvl-Y{1Y2_9H49}* zDeUU(+K#Gu8dfszVex7!^#*O$OsV03@TPycfU2wY1EySrspeC{pA1d=MXMNdTN}(< zfwpB?L0y-IP4|UQzg-Njc6-B3Q6&?}1}yq4SNQ#~XnCi40)PjSW=6*co$$V68=88css6GAvQ8*0@+=gt2GH5KN;*|mZ9~A%`Sx? zz;KbQJPvG~V)~~0Fvl!kX@Ck;`E`MeOM{I*-dRDpGK<}01cMwDmnNU;o{X-L@TYE?&pRLs1Kb9DJ5<{uO<_+V`Z-t%ii;|B0+ePpvYS4lp)^8HZ{7| z{pFDn9fHfQQ^n(lSWn>jD>!iymnhDDnm~T{LzSc-Uhb6HkWH?Wui<>0@V07GoU^eQ zD(Fz;kGXNy3W}>^{}@cdG0&4@U|Z8h&s_j{3)XZu@NYk=XU`>>PBN6a%_ zHJ!4hj65gvu#ttKZg2#7)G2M0tNc}il8x5XuQ+qm*LJnMZN8o}x>cNutfd4ECsQ&D z!7T;kQ)c+2J_`7d6{B|z?}=uLPv5VZYg#5i_Y@8=fxnyipqs9V3-=W6DM$|#*7I)u z+W}22m`I+=h?xa0pLCg+-6%ZwfW0^xjToE$)I55i-P^lKsV10?f_{WevHnln=?M%r zAA`d{wh9ohpK24#Ct_U;aouXmzOr(?|O<3y%rCsY<0c4gVljgLD?X57{sZU}EnLl^64yNnf{ShSL_hw9*n-z*B!LDOrAzv`!S@~Q&uJpbmh$^w};W&B}0=9ilNJ-0~zy+ZU35&F&o>~Ai3YiuV<#{Nj zQ4XD}>rvHYGQD7@uu_zA9^|eu5avf~qy(LqVV1is8mzE;{q<2JaMl~>I8h}-A%y^W zE-rlpRI;F4NHd!fwa_0D9la*4n~&MlEmlosRDFJ&aeuij`{p_X(vl1J?U|=LCJO@R+tL$K8rD5^bietPhWwc|}Af?bos4tfhXB{Hdq-uxnWMWVf z$0^DaEL!#~Qy1lwCLfMv(|JFU|OCrUkU&);lngY9b#5cxJgq^vt{NHn`qKi zy^`X`OoNkUMBeeKE$_ckz-^qeWH!r1dmt8vXh=r*?iz-grIXQIBEl_J%?fP{?dK|t zV@xy>Qbwnbcpf&UytoCl!wJ%bY`TM8?h`R{#*c?rjM@E3tc&}H4>?FQ5^|_-nJ0rY zqQYb84U_LES+#0dC~TE~q<5Cf5N9l?x%+ctG$wDaF%J@*sE)Nx)1nFTYr>7s z%u*WD^Kf5reCfC)m`OMKle;SR+J^f3=M%D{{F!e1@kYA+GXv7(4NkuIR7vXR19ZJJ zI~IP;lKZYs4h!4wM>-nkrd_oBO*f8ByA)&8FG4-$H$Sp_A`{mt%?LHIPgi9qMLiS8 zAWLz{_`!ywe-xy0T7fZ3QcN+Q)9Ap`=&D*sCj)Dk3Y$?lsu|2;463+&WvO`p{-1K> zmi#&FSR+68DU`Hc%b-Jik%9|Zqh(N&h}nJ`%frr*kdAil`V%B0Fmq>CpSx?gCO z=Yky4XJcvtzA#ia2QGVYgUil8pJRR%R&~Goot}&I!JI{D8I9i`GOsL7PcHNA_QGd9 z?aMI=U+=8X3}I<=$R!OM%Kf>jQMXX?7Dc>cGK*ue(Nf5rKq9UHb^gV(HK*0IvFHV1~ zfD{3C@KwR0uLn%gKH(mVqx~;8>%n3o%in>#2 zmT=bfX6D>>DOnDuq!0iJxs)w6-=XT{r<8p91_7#(ZMZbLk!3g#-6*`-p*ZYmqPjR3 z6B-TDpA1}NMc)c!A)ymJb*RX^IB~4Z*kOZ6)w4kt$&hVTs!OudotS^4U! zE7R-|>>PE~m~|rnrG`moC2w$9VV z$&0>MKy8=ekO&}eR|WltuTpq~Y8QZ`p-EMqFBMZ)#mGk-dx4ZB3ZQJ(xfoFc^zx`z^iRTcV59{oxO-EAf2%FB` zX>XpQ@8)MwI_~0+4~n6;$Bi_4pHB5iw^c`wUzz^_6wht?4xB8$^h%Z3&pUKY(d-1* z&MYNFqki$-e7Z~${&ikiVwecI{))w@Ktyhv27LTk{qE)Iv}>*pb-8f_c2<$t5<}*) z{;wMq4|52_OQaOn7Reh(Y8j^h-7qTwYyeqY3xb`v3nWDcNU;74YNVWER%*fEcWsWO zxuZpshYf(;o1*}^->>u22he(&qj`h(fI!b!)AcLT&KfJAFpC1DUE@U24aNu=%uBU zpqwvPN$!c|=*mC^Arjlf0AFh&41W1ZE3f46ElV(hbzEw7Az78PoIWo=!RxoS=bKC$ zG^WR4;hWUtPV^tRq!Y>lPu%ZPK2wmJm5*||r=?c6Gt%M3kE8y)&JvJgphp=}P}#gm z_pxQ7x0;xeG-(KDA=;zW&r{LG6-el5XW)_87>|g$aZ4LHWUMrv5M@@#=#bb2WIw9B zn{CY;!nWm;M<20eZ^<~0vZlZ&lXQM_6zA!rI zP8$#S@(U!*xo(jR=;f*5f(i=HqKI?n0Z@JjkaUJsfzL!c*}lD;$PEd_9WodddQOE5 zV?}1)1fc4LmvP(4pjbaG=1<;L5wk`Rk~E9TksMcYuGDBns^}%O>Qy`#Rew!BN;VB% zuH=YzR)s*93Jlt-#Lb5|QEqTkgiA1}aq8tqF&CROc8S4I%rpVip)&k6#h;w(gnP43LpJs zY)EWe)bhU0x51InGsvhuo215nyUvI$RU4CYutPqnL7vSrbrzvAO~q!%sleu$61ElD z5r@;LZ}((Wv9W4L9de@ZK_VJHV$i1GHUtt?T?c}JpP+UT9xRqAVc0n>aSc+=4&(;x z!Nldj^afWaiFKs?)plaXA=TcJD5jCGWxIdfy}nWQHKAs8eEu67g78{CqJH1g|C(AQ zetxm5FB9|2m#;(`Sh`oqq-iK)Jjl%T76bf{FTvQ~BdFBE^ZJMe;OZ@&qrDoH_oO`* zIJFRg1j`EjhaQf~Gc`&xE8?&m*w?Icqql0tc>cLxq~AEXb3AY2$qMyr?Rcbr)h{vp zj9l^Rp2N7qb%DZzJGiOY-4jp3aCDKVx>jDr zE^+-(DFx6{>vi!$DPF-1%xw5vI^7Bz{EEJlw8jnHo)3s?E^w5aG{=Gjh(+(Qd0VMpxj z@9yO1C7Ai%)rbGFj%bP$|Z=}SO>Q_YHcZRssmZFRC{{BPU&#%<0 z1%C|UNT$tdv25U9(gCH|lon8R+K28dazlP`Z786b6hA~eO^>00)Au6}XotRHUU%7k z<#G49i3$r06st&$Z4|Z zMFwDvspfL)fXG{I2O=OXKgP)DRg2Zg)edlC+>)gMR$i#sbpH&nB%s{MQk8CbOsdJy z;xr#awgyg2=dYzS{j;&M($&GWX4RjJl0BlQKYG$~X6IR8(;Kv|icJl#yH9seAQB?V zn|H?t-|J}*c=)BHgKL{zA2QXcB^>9{Y9#-gC3 zQNzD`tz*ClkhbN?ECOC?6^a8BGSAeO1s4_B^ArNDLbRWXI}cjtBo4=^(!f=Y(0&feu3kl^$+gjT#Vf z&zSo`gV8j8IeNZ0qBn5E6e`CouKm&LhSVz6RoKBv(eu>EvNZ4s^{^W6x7c37r-Q2NMU zp=*sS*HU4PPV>og2;Bt$L51s3@Jj1k3B{1u#HC$RP$Udqk5PqfU5}a2?y6?CYP<%J z2BgWvXmfv6i~08wA;9_~kte)vix|L@vwy=}flWnBE=SpLHzS6(3zZ`H^BN2gG7Pz0 z5kmGZ4R44hKbq*K9&P5T{3ERkCzkE)@n?id|LwyLs ze@3B&!LQbTe189HvK~9|^0RK=hl@f0P4Lu4^`-am*OX$B+c^=ukV_uTkNZE_-)Vl{ z|5p8)@4$-k0Vya~RWo6NbyD%Q%0DU2;uU))v(Q&j5-ckkPR z04pgJvUz21=u%;!EhEL!#4O6zkg=~XCV!r?rd=lr!2RHmVoek==_Es#L>R|nZi10z z)Yn?#`WVk#o<0wVN-vai1c=d?X!yNZF2nJ3_~j|6`C-*3Pb%Pt;G?p77(WCF88)2R z-mT@~ep{K5Su9$~ra{kZS#@^6q=0jx0a;YIS zbZ)KO^}R9$fM)=TB%jl%JdCpzTU9Mq2Xt>4l@X8A_vvJP$f-@nXA)D!-nT5pUNP0aSi2EMdGhMbsrAaK7xFqohS685cq)4cYj}&W-%T>kr`J2s-Ku?K%d7Iz$l-;hCM}o>xalCp{h=Hs5ed8AD=uzVK=p4UU!RO z9CwprkB@a#uMrv@k2z6ABH~iujJS?B_hn^W+6W)kpqSh%I)`R^%l7}(kIAV&eK*$` z&Ch8C>?|4r>3liGsCi1MD@H=GmOL}~))|5#eWxq)1x4qN_;To(Q1it6nMQ6Ip*^o~ zYfh1?fbZaj-LVY0IQ}nZ)n3B;Z_PTXv7G?F59_-Z_b&=o0^V!CNWb;wL=Medc=YzJ z6@v4P?Z;wrTOf|*T5X}fi!QaBBZ$!Fh0GV1PWq#JVoRAv@xC@opL1z^6MjT#3fp$G zGJJNGWRxd|W)qh_+Oji#Hd_d`euL&ku}6mR_wVKDijHAO4*6idFsZL`Yf1VXxT~B0 zw@}``q9QngWJ~7GAjgYX{Pi~9T*`=x24=#X+4^S&k!CE>fx3curNTGxohS>R{9ZG! z%ju4wi(z*myj>ax7tT$GWHCylkZ=i+yO_c*6tS394@7E!W7x&7CetLpQ@{DeYHn9C z`i5Rxn8cFeu>>&x*QtLD`lQHUw7TowxdH|k%l(dfeP66 zaXYL!Kt9+*e8<%EAR-!_bY+a4puxj>onCj2bx8{KKb_|o}oX3e$0eKEFw(t*4{mC(}bAy`D0$mYlKa0 z^q)`9ng39Q2eW$!J_M6G>#V(BGn6ZFVBF0K-(PjQYZGj!7JJ=A*9HC%EJ6}$1F;Q< zARC2m5{AXJc<&g7jEn{89R-Lz@G~je5yHG}7b0w~Be+FTNsmK$x`@CIgqSc4{y0I2 zn#kSNaO^mriqwd~H0oa>f!{kL$&MqHhTqpagq$D;U{_I-7)BF0L<^I;%5*uJp##j< zqpgUXf~6%TNeT6<4B~C1tjc3tk7Yv%V?5Dgn+8QijtPF12z1V42Et;4YGNBbu^Pc~ z;f`2gWO0qDv0d|+4dGG4_R@?9M;Yp4B4Idg*A;G)RD5<^e2!6kL2i6uZhQ`0kiJys zJu9KPe0+5PRxJ#gRRe7phc;-TI$f(c*{F8JX~_)}f;#BnH~;*cP(qeCAexwajET8| zUnxpha}52ou0L}Ool=S~Lzl&p2Xh;M8%UY`mf}Z_6N*_Aa|5t0XrUOOq@Ux^g@ELJ z^tgD|1mrXlyeh41-$e5kCuC{pu2SMtZpx))qPQU7ViyYB&~LoL1f4)pCZJ&f&?c$m z-8)Nelr+_fB;A1OfR`M}x>IKa9N>ug62)r2ip3P3=J3L|h z5Vy1=vtk1Mw}g3mU2+e=qP>1D8BSJ*1c2)k8(Vs zCb?$aybEkX{|GfX!IghZCGyH(j!oH)Gu!%&r?rt}8ILdaNc6fp^&CB&OgRf(Ig0bpUe~b^pNZx_P>x!j^*Jc?{WL?GKJLS=I(PmS0=Q@fJxSSA89VeQw<)zgoM;m1> zA7|lM<=i`_QBLHPoL~YUb0XMsAZ$tdM(K`T__0pjiI4dMj#v%G_$hfr=_lD38yS`H z$v7KXTz>kY(O?cI#@YS6l8vNb3{pynVn;j?bzpu~ch*I2w!I;E&5`XZazPVY@~loS z5k5ZWu#GliG5b_0Z`v_0{+Kv?53JHIt82pu+$9Fm=671+b%Imjfmrv=$$!XG|JHtd zKEb@7$TOEk<&?#|-oQlWC8k*;#1aSKJb8gBJaA1gk+o6Hv`~vXNs^a}sk8~D2TMB{ zZ19}*2{DPyyvo?~%j5$|Z3a-;@)QRjQ*iOq1fDo$_J~QIELR=%2gwOS4~fly0?n0q zf|Fp8O`&@f-B;R~p;;A=NFYTM0@GD|)%^4}u}V~6|!)teIf4Lbbgl2-o(=ZR3u5zdq?}Qc76xWZNaL<@uJvI}l*6mE- zKZ&>aVWv!&w9IW59^F>CeyY5MwsL-J<^I#kOWCF?#`j-Sy#Hc^ z;eeq3_Vg04;(sziRFCbU|48wI7$W|&dZ_6AAF6)@<-32Uc=?$i>sxuLM1VLXku(Af zU)=?uLJ}%*Hyd!STEIpoL}iS%w$8TEWK_I_UG9?!ig=MYF(AC&5rR9i!AqYifn1SZ z^mto6(1UV)O4aPoy#ccScSqSWDb(!=6rM%6{#B*}w?W+4QP;Ge;PQ1rhk(gk$=d{G z%Z!;Nar8Ry>`_?j9g0^2X=2Y|H9H9`I=W1Ss-8n}VoJ|BC~ zNakF(Y&~}S+cJccf~2lbX-z*OvR)UwzSzv{K#IfdJAD7^N&@S0r?0~0C;X#`A#PES zG~##Q9km}Q**#rdtayW>yyoP^ICHZ?y@5swM z&X^>>Bf1x9IOfO1qM|X8mLb&#Pe|ksnn=~AOOHFqZ$cR;_Gimy!_U>UAhwgXn+yc! z3tKBgv%=nt9~K1Z$T{R{?c7j>MAJvv$2&uhU$MB?t=GsFCPg>~0}?l^oOA9+4B5(? z`H0xuQgtxv>RNl8>zZ&8hh#ZjMikC6HdiQ=q;2Bk7)^bs-#FOJe1kR>5_j?q0T8}3 zPQaW3*01*IeBy!XNp_+E05))a!lpzBb8RmCH-ze+yQSU`TD4y8u=An(^yS!r z={3G%@9x6Z@|@!tcDT5k;`4ujAC;!;0FbwJyyo6McxvCkp%Q()Qq8KIy**vBMSmVa zEf&Bl8A^}-9}^<^p5N^szK1S9_GZp*{cLK$S9e6?%zxY!=|uc(viE%4ynxyF?-P4K zj=p_F*c%8xpfqbEZ&?6G<=+sj&P9FgrAC$*A41#N$K{K++-+TJ#^k>U9eymyf$@g> zs%!dCMj4?Xh+>2b20Z}4-Ax$vDql4)#Y%3R6qW9Q5R=mi$?)?&QS-PwmtR_fY2g}4 zmt%{(v^&u}zbwo0_?S>nOOh)m$}6!26mIkZVsY2y(qi9g;Ch`UI+-T9Mpo7Y*qcb! zDzb`18CBi*0IU1>%E@5Z)Ds({7>jHvjecKI!^XNJ~IFB)=NnamhPJOyV{Y&s!Ofap%< zV}8wpPqP64HY8QtE4?6uyx7fS2a9BvNnD$b-7`qW0Aw{^p*zK2e_#P_Fc!)6CIj`tr_3UZt=cry-_ zT!MTPGEgTv9*ItTYag`Q$m2x|iJ}u-w1U%U z!dL}laB(B{NNqT;bHt<)(n#oToNMQs=nb# zuoq+&kX4@Ut>;%FR8hZyx3;Ml<42r7=e;Ejly~cFEs$?NNZAdX?xbxGrp6JIG`w9j zzOPZ_-zt`NxSRFvYe-{(+A@7Kn*HU5URjTPs7V^M;@{RD_GkLYuTf;4)9-hMYzV$k zZxmngK?Upg>0?5cE)ENlOPSysIte)6;`HNP!r(c+Llf)R8+UmlLm{PR5(2JwR6kpA z#?IIthe+yx1I@2|v&x)nu$Q>~JNXH1T(r-27AGIttNv!Vv8`}6kqvO{@Lfd^7VaV` z1`HBXUjbjuQlY&0XhMe#crCNF-!%Xm^YHRYcG&uX7pqif#20P=(qKbHXg&_=&)#J9 z4-*$zs#+sgvvYUv6E}^E&b}|cEqDk%dUN#1rN(!~zi$5%1z)+2XXU1k468$QQ^}M&bb5qkr z70ThHKChS}dUaTIk=hU2|C4MG^Y-tDxj%|WSnn{ERHS!beHR-v_e$BZ!A%i+Yj$-; z{eHhU@+OzY&8ND3t4j8j&pr6{5uU|!n6nk&Vml=iS)|X&++OycAyfz2R{Q z?K=;<$X|#mf~4c^{x9;m_X!0S&v3l{>mMFdbEW+!=hc1}54fFIX8y9MCR$#}IUf=q z0!AM~@I*uF(cU5+d^grY3#z@Z(3ID@yr>@DKcYeQ{ewM@!Wvn^$&5l?#f4zyhLFVu z;iHGbj{*)AAuIuq{A$QDY1oxO81)h4i6ugM+-uz5FGmN$S`&zZt~~V{GK%I^AQGi5 z8hOA1R;b|=7Y#Sbg_s!y2P|V?3}Q5Re!$WG01P)yB&Lnk6&n_b9_l_;24WTl#xoImkX$$KqNZJwkgBbsV zvbTzggWJ|N3x`5dxF@)~Yj7ys-CcsaySr1kySo$I6Ervkx8N>Gs`t0g>D}k_`1_)- z*VVdNV?E=Y^L-qg@e@W97R!z4`3)Tw07*c$zms3Vi7d8>m?)Sr5$=B`Y=xk=iDj4x zPXuY`{0Z{G3D~BA$1+%q+)BHUbolJpgXp+#bct+d$=~SU^Ff%=8VtCEpRVwz8oJ^k zeDLz3akHlxuV7?Yvh>Sllv*VFt+h7qjEk^IVTMvnqWJ&o;|gcomu|0$6W z21?VA1yQW)^;prqR6$C7KE#P;)=$Nw24_Att97|p`e#z+*(M$$Cx`<+o1P`RpJi2; z;BcQM)^?{Y>%{C*X3X;=r*xCqr`U`hTlK%DD%=R~pQZ)c=1%LX&qBF!4RsTQbqhYA z6pR<^GL=)3w_7hQ6S1zR-|H4ilEF`)K=!J?06_VtXZf>T@Y})UKV{`lMEZ3oIh3PO^R_Pol7A!zKO#r}fLt;_ zR;3bH1TS9%4+Q1-PFMQnR0eHS$`w^Uu>qdRBn)uzS2uwt4&oXjvIRSmigS?TkykKv?ay+1 z!;qXH0)tRZIg?0p#47))82Vjby)@hUGheJF2HLV8b(g0Gk&6u&&9&++4fLuFPaBQJ zp^blunqUQ+;J;uE=rth=HUV;*&^DVe;+l}&o3I3%arB$<{F(`Jn~8dxNj955zBiK* zw@?VSQ0ceO__fgGwlMVmn+SikpuV?o2)1(RxAK_iW3&6U3iP%LZMJ@XZxtnO6Blfg z)NhmaYXj%D$@R7=Y_=)Ax2X`fs|mJi=(lV6wd>@z>-Dx9Y_=P{x0?`mmxc`5R#4`v@+WvnPWA%Z@$*NN+ z)oTpS=J95-NL2CC&ysQe?fXUZS)}@wFB(SSPTzJq8^RF;Z0mL>qpy%I#UN?N4Av;l zg9VXs#wtpLly+ei$t5$vRB7<*_A|84YR7+#ZW+tgB%|fXn^xJDPpi{uf}Il0Cr+0i z8G-xBpk%-{$UH272?B#eYb{~3IF-j|b!!LVlid)(EhZ>%G?nV5M)f`1zZfcMk^M=0 zEmx>DquIpzExh2i>#{X;gnzq7kZU4pB#ZVw%`BlqB%=lddLC(3%H<~-0$?wX;Rgkw zKfw$zX$%{=Hgn&Ny4(>*N+*oT0)K-~w>Ezn&x!Qdhqy#oe*(jF(f;(iC9E*#l5#J29gG2(jE_G7I~muOcZ(Edz_NXyjWWA1-}3Bk`HF|wAhKc z!(8~7h$^6C4#sxH-^5a9a>)-QaHQK6W%#SJlt9fC9i$U8{KROr5wLy$<$C-{`b^fMWdhwTr_@jQt;|-y3>=Ao&04{7uw` z0&|^G^|KU_BhPH;Lx%WZ+?wWA@ECrioPH~a7{1=a?Jwfpo@t`y8BuFliA}wdM^ba0LFbt3SYxrHDXbL_L6H+MdVuqp$1<^i-ah^ z8H?T#<@j;MI11fU2cSW`!>au$900Je57j1gEX48JG*}>A*|rboeqcPsgVwdL1&1+v zuK~82spf6rirTl77ru3E896P!Y~`jWcy1ume||X*fivQP@p`zlKeCs{cs)z3RO&VX zthw^%Lhv7WPujeUx|rL@2kz){I0LiJ9MbOI56imH{f`=I8FmW#hS? zr>29QivjJgn@38)ai}}5IFQeXt|cyZFebl`Wd?!btbuU!q|l~Sk%@jMvYWVVz%T9I zO;cy4<1a;rcwNJYpXURx^2kt_30OmuKbXDU7h-5WrvpY+f;k@wP@wE!f&FF`ZwASI zIAtucwg?p;$~qI211X%}rgpw0vu1}+S;9oZE|%0WW}sT@L2L#Vm{|G$>J1LO3BPkI zB{`|;vdebxxBW6X~79G;4ZS_qsI%IPdT>a7&%E4N6!h77QWYMbBU`K-fA%juQmf0hz+2NGV zQ&-Bi-_&{GvI<2nEf72us|$NuOaqGhY{b1I=|mL87`oP+HL7F%qB}fWwNaYpSsJN^ zZn|>&@-z%tpxVk5flEzlt`Z?K{oLYr@CG$Gx+%@qqG9OVzph>gBBzCjQi2Sy(gCUL zlT^yxpvcgeM34t12)-N7ujh{G-d#>QYgn+BE*0R5b3}3Ww9sk)Ak$GrC+sl1n9@-fLT3e-= z;LPf@fNm_tN@1wwk+I-8=6J(o13PY>&Apt~9Mre0LR+;~O)7i6V5_~&Io!d-jCfX{Umaw#piA9Sc?0io|lqvftc+7A4MH1>%z3Q>q)@ zX;{7X4>rH$WZqs%9Gb7^h z#avw*PSMuhJIvH^-$?V>^f)coq*S@aF)4n1fVJgE9B`dy24&}K_1tzL>Q7GR;R z0i7SbSHWKJ9`cjyk=J5>%r4A$eLuV6 zBFR1f*wE|u#<&*u;7RORF#cA`GL)vR1Q+92h4;At1BHW$+KAZIobaIiqdPbU+bOT> zJ?RbOt)yIUx0C3QmdjfH6n@y*j58Jc*z;aze7B?DBAnjSBDHbCJWe(2E(uAWqd-dT ziMdG})!Z5UnKD~BDH}bFuX=^mz|?|LUht&-qlC^|UREC$QIX`M1NvTIn!bC#uL*7S zJ~mVaI6*6Zm-W^62htETmLG#Dc+Z+$ zZoy3fEKIKO!S9C`^nK@f01CUSK+ssa@XRcJNoeVElLErd{;8@ zgTtp^Nw-v2QNEHlzHn>)8`i-HRRO?x6nc63$Sy5> zu(bBa{h(DO9vFhaIrP|2vEtk(6FtE2omI(q-a+MncyhXwFv`R@??{HJq{ru^gT454x`<4O z%2>XM3x_L;}ds2S<7bBSnYW@CO5%C{O@I)Y4qe-6pPtS?>HGE4G}@o9r^5oSN*MD`(WJLEuFh)+7BJwC&K*RQ1pzeZaYey^!q+VSk^Ih^tm-M= zdMIH*1Ak8AE=Dj-OJb{W`!@iP+UmAu*P!WR@s11O)yRwTz3Cz?FR z{~n0wn}Z}*giMr!Vwba9jsNQfXADjZi-=c}Q{!VxVW@4<>{M~^YyQb=p75&>3=z(2 zI2PhY4gxTzcDM14OH$wA1@F(}>&0Xik)RC=)F1+_j!H6=NW zZzVRdIh-2>7ZS;+RK?!KrDF3aLb1wf%@!}Zv|pV+e7*T#*pt_>TPP(^0q(I-fHD;H zc%y3XmoYL_lbr*dTN5DyT)O88gdxtv#l_lps84{B8QGc&Jz$I+ANYuzN30;InnCOC z1H)TB(=`3|RLSYoXX5T!eg@uHPCXesx5pQ(pbh&c=_(Pu8nbh5>qp8$-69lVwYFL< zQ%^~UUUgP%dTST=*Eh_98q7~bkP3!|!A7rC-? znXq50f%Z!{xd@mLPwWVyrpTT;Kvl!Xk|M;|iq4*<%Z;WyvuZc&d|6##yXHomJ(~~s zSoXn?2)P=goWc&#CJyS@`*wODfcqQwviDD0vg#I|g}N z!>nBs=3QNpPti_x6Z)6>Wni;0eRK9(Mue0G>1{m$3_L$^!!Z$rBDP9=t&V%RxSgSm z{=MCQ3FFnPl0`px$sBJFUVf3YZjl$Oaj<+Ur&W-#!cvsWxY!Fb6>!7Q=0DTKAYa>H zlI4Mt6?>l&+sH9F?`{C}K#1W1 z0C*Ua9!YF?*Xkb0u%61^o<>yoNH1Af_L_qOtP&ts3B$*d9=C)Jlx}=)2T|ZcWl%s? z(EN!PmN~BIsh4{M!Tft4IstjoEs}gnGaGIHLT(3_ls9X!pi%OP$@+{bi8NqnMtS55 zcD*RD&Sc=?*8mTDc@i-=*_2O=(ovzsCFW2Lb5iyF3!at?h#DsRwYuf&7T|bx6RQ^2e+zaPPw{;p5g~IoLTsGQW6|` zCLCf9+jm9}EcT&~OT!qq!(v0j^kOacnE!KD_#a3&|8GeDAIY##*Z&dHW6RK7fv?SF zQ%RjlHB8`COGyq!^XWVv*Rsh>F1Klg>U;Bo=W9r9LiOF1nSmo*h=TUS*l6+-_x@geHwb6G}p4L8~WS z!Cz7`Sw4Q~<~SgmT;!^iW#WbE8*}2Kdpvb!V<^V}Di1r>s*2h+q&EwSW0^R0MUjLu z&?}T_32c_bEmzxG^*zaD7YZ4j{c6g(%sh)iFdFa0%?L<_>2?(7E1}aE0W=qP?12MKcO1h>$YlaG2#sTeLSCU|n5MAK zc^b3v;%?TaooiZ*m(g8Ag00rx@w@$@!8~2y>m&0w#*TDNO-GdFR&AAc#tpfi4QC)-!07X3zLp#8iSWlU|V}t zXFyqr|AK0*i?`s)k*mgYqGrcl+4c|Ze$v9<8&oB(7~BPV^V0V~KY~6`{T}?frapfU z+B@_yl=_t3Cy)4Ovl3}6_^|`{x;3&bOdDd^a)oN@ z(6Xv>n*$h<8HAz0@SjjaX~shi1^`Tv#sBqlr2>bGX(XeBlZhkpRt;fT&4*%>gu~y^ z4n&Ve`n%B=0)M*lk+Yd1vs#2AnDW6>EIUWywwYpxRY@R#%EB>XzP&=9#3g5O*#$}4 zc9Fa$N2%|Uqrz1Ng$>d~lj^7=JkR6}4_n3t;$Z>t!_5^r>U&|ThnJyhx{e^kp#w8%yrG)es7)*f-6Xd>e01BZXz9(c)l*cKD0Ye;z@4A{DJB`vVA9L`XtSrjkgHOZpbjS|$X`zv>bl7*( z?RmON0vuxb{LF`vNxG^&M9p-o*yo%_1jFPYv$Dmq1E$t&#&n9ZT)1v|`9p#4s-w&C z95|E=-)ttlYQi~UA`O?VUBm3EVT)tDu+X?*SSri8YZr8=v9cD_5`+5UHHT?1Nr|Y<*Y zVXAEr#X zsh%Qo7kCib@!q#UCY%dY*J=dHWrw8{DRA4WKd`-K0%V2pCGkfl zB3CGo(Zr_G0UBeCHy83&}A~ z`&CJi&ptK@MZUR`z)x3?qHG^s&i?dnN`xpzyTKjk&BkxtY1S6_4 zA-9@wK09~Ko(fmrPCHG0?=MnzH+-^m7T%)~($Sq7K{~m}H8nWF5$PFkhkYRP$)gNt z&ApD~h&2ZL|E#~6P_5jtb^4Td#&(U<{l&{Me%at4xfU3uN%}?Jt#8Hnc~Yg{;Ndrs z1?tW>zP66h7h^jV!mzIC`lv@D#}7%$BX^XfB#PQT7%qu4 zc*D=zCjrkbeNd&sZ$aUwgsi?-e(2B2+yM+@^q$M4jy@ske-^SUJ5RVmFPl66yt?8E zFA zn)8$v;?)iCY8Ob(7fL}FN~RM^=^aY`g3&M^igSuNO=%##04&H1bcl35+XHs)VssXT zeeMbqCI}ap3=`oC7hMBpSrLWuGPKbWTLJ)$HyEnksA@V9s@@S=kO&=uhz?O(0x= zGR;hfqjVLOft0YgVv}JR;HVaYxOz5zgAOiTce|yb7-hbQl9#wJn`j1heA>3axGt9l zZ`Ar^sS{JIb`Clfx`;xZ2#T(-VuILVZ$A~cSm|b}_)r?TVfu!u zu*_48vqNPuo0vPgxOx>t7R{hLH_W!-nDxCl9f-kSca5I}(R-6t5@WGStKJC&;f^oB za|;pm<-}Ixn5C*XTuhO);iPrsSoJccr`;r7f*4Ic6wP`6qGG_2#l>|6^}yy5w$vAtt#xC%VpjsMv*?MUImDU zhehC%hN_yO07?7Pm8MCT3Z2ZLlEs`ZAvu8ro>oyZ(Wgig#5JD+%_l8^CxiwavP3B9 zgWzOw-Si1C@h@;xDss&IWac7WhD$^S5_zCgjBQL+2C+|;(Rzk1K_+}TU2SS=O;>6l zJtTxM&GFUB?bXnr0>`;kjc^L$>64Q}pYF3r;%iGOLm!p;5|90wZcmVLrIrEX4btJa zt?W`Qh zFZt?A9n_h>lB$j+DGEoF$Y3 z*r9t!aU+gVXW&q705};mwuFnP1@cnWUAF|44k9v2eol}A`JOV=@s|~=k}x3@(uFV$ zx#XIX8(9RK3Lt8_Qwn%cV(|zhI=^Oe%VB}v*!bQ+jU9Z+t2TH;xK6S_y_#~vH$fgB zJN0s$>t`;!sS4s0j&57b9d$2UH+jwwA?GRa^$Bs5+dy|YY_APWUk2>|%*w+4Lqz(3 zFaY@9v$6>PO+@gr zy5iJ~3HU22a@3iuK&w$1i1azP#g zEtX+HPhOa!S)Z1ZA@M78g*xw|jkZ*E5ub*WQBUCMJ+o`{s zzGh20b`bJjBbT`9@11~lS|7uvTjRTjGP}%)$o+Op@>@_m)(ied-$80>kJ}NBB}W3u zweCTg%ymj5d`Y6XCZt&<{W$KqWf8GU8y+h6B10HVZzNTuYk3)Eq@QP)mG9ZATBWTc z`9NFd0R6#7=DIj_YT^%1PpBCXipdY35@w=HfywokZZKhm+lCDV6kL3hy{ZkpG9E34qOxg^8D=kwqiXeBQYb$pN9K4%GE-cV;a|+mwD+rV zfaRvPX^27fA_A3;eGF?@X;4#3m%maMkng@POIzWt2JHA@H#5~y?yTm}wW~IaKn>HE zy2fe*D+(mM9!I_IyIYZGB`i&b2`63X$jKMK8kj+$zr~7g94bvtbsD~=$z!F3SsQL> zfbH^Y>?bf5Vg&@P-JNRylAB|Oa%TlO=q;ULou(OU<6CA>zzQy7gsXMUeWEWjc5Qtw z;%-B*#6nHel-P=LQz{iZ_ecsGtrvrJwem zQMiQ>KbgjRktZ`Qfr@}v6anoZ)3G+Dl7GA#e z1|Us(&h@4K?wHJ+G2}XiJ=lKONpkX!n@HKW^w`rQE)Y1Wru{s>U(SsouwjSF=3`r? z(eQfYaqsAJqVS8s@6;w~pysB|xWDIa*6XrIw26`D_l4>&|6a51-Y;B>nQotWj~RZF z`j;(Xe`>#lzwN^z#L8FVI`994@%P>gbu2@iG(%bV@OOhC&Wt_?nGlu^!101+3&&qH ziD=eSO3ms9IWT%5p&K}T3}<}C2K>bn7mZ9Vc0E4WWGy-pDG0x0L`E4D(})Hcb^oL3 zAa8`-6({=c>@TKTjOH~t@{iAll`3`)>Pa1(h7xW9OR#}?``kgz((GDh12?o&F2+D) zAB(6Fz0!aP2-%gQdvVJ`oTy6XNWG1>Vmd(T<&%yF)X^FLUKi!z8|B$>2{XN*FsibV z2_*xu23vf>O41qq{3aP6u5d_tGC673>=K3-a;W+-CRyP7LaI!kKK?CaREz^;EJZa! zAdWH_I?7IGra(!gT`d<(ekXI2S}J-zDJ!*-s;#7PXb2BZCNFVL&EqnMx#ev%Em%nD zxtOMBv`Q{LbB+GKOvcrEB1^*npVq2S#FLgP$2FdsG$dqxL^1Z6h^55>p-+$Q%eu4y zO)J^O0kxRk#5d;^`LGAQGDdnHD1wz$D_NL*31<1LKQ~CT3ATpQwpKiAxlg)Vt5-Fk!EcPxAx_H)IfO{NLop% zkeLGU*>~@bxO_V*8kLFD@8HM#lr=OPITSr9gBC8&;qWR2O&GJRx26FqZ&*A`$^r8P~h&zMrRpEuiDL>yQj{9fGEbT1?_YISu(1wi2&o<7))d; zHj}#Hc29>;DD@y|rwYEOp$qDjt5@&#w=3qRT9)g?&fYQMDOlcS1W{L*Fm=-LX4*f!4RCFoRcu zqrIjc9xr&GjvJ#|QFre?@=>iBIFV!b?c=o)e0{ak3A}7KhW+p|&co6q2ZAUezOBWH z7~SHXA&Lx9e=u4_xbWRy-ddmdh$ei-BMs~0k#xWpg*gK!_|#AL;e z|1yfqVi89EPvfmF6o6s1JvhL7lD3iH1z2PA&L`QYdG{IJk!qFI_-%$yERa0CS-5T& ze;|Fb_O(H!5kI5$SZ;znkyP^GZdbckD62 zqoh~CocdjRwxdbLo0QoEu8?X7&FV(vFMG1-CgqyTKn zC_OnB4D=S&@xXE|m*A;*RrX5)d(Z6iON4ObpHmqb0cxR-)%Z};B#K3SzeQ&909+r3@@75!0e zr=oqdgwQYunz);C)Y-jLt-yow({bp-=!p$&A89BHN--+M6?3%yRzyMj*j&_sd#{Ap ze}pIL*L2Gc>5k+-gTg$&OrM1#W&&J4L=O0Z`glL^Emyt#AN(8zM(&t34Xq+LvUR( z7WN4^=sf>n**%X_;h~Y{yS(!EE%IkV>f=u5KqaeRzy#UDn0eo2l4IZI(MPNlPJ-{A za{*!OonFX0x`DHi{JT=hzjIIWs05Y!?xutv^zPqoc`EwB`-ZRQ?=N?GnGGqeAMF}x zx0XL420^`&M6hBvwI5IC?)EPu!aHU@KglV}!oupbAoZ?yKU#AN_yn=-LRsQ|mJ-fC z`K;pk@^;#l4*EuS1`n?IA0Gp|9{sV_j2Drx08I#RW!wbO0VI4DS z#ZRFj+8PT-qy;Dt#2lz=@f<-B=>bkRtwNd%%QZX2B- ziZF5Y`$%VkC|5piYYwb0rRv$5n1-V|08MSx3@~LlVVs+AN2_}{0hTMi6k4;&JQXM& ziXgXPs;xWr*#QBE&Q&O764QS&s-H_2g&ZJSgmY^_QnqI1puv>@FI!f{hSnw)5MhP% z6g$t&i7*rcXBwC24eHPd5rF}FM}YjOLH0#r7?gncu~^g*#Hp_Mab&%=mv~S}Y(`dG zw@z%iPQti%-1?+W266(?P+U|=!sw~qjyF~iCC&U4<_;VOVTHoL;!y>&R!BcJ?Py#8Azyyx}JiGBJ`Se3rUkiNs#bKmm{PX<4=)AN&4KKt~QmVtedFHp8;J<$I{K%@y^hB z&6u0am_N-hAxyG3OE%NZEH_OzuTHbA&UEg^v0hJhBFxer%Jf{v^~lch&z5ksO%;gA zYU9g7?am7DfkbCxMP@@3$swE)5MVbXoBhdwin&MAkQMR5oK>7x%|2AR z0#^2d^YsF_sRAra5a540{Zs+r0JZBXdvyBHRPpkb%dicibcN^4kyZBzSf3}1DI#*hQ&V+h?`T+*+=iGZjGIs*Shbl6I4eqC@)8`e!&?L?@-(Atp?bq3qLRaiUBKmzhOFxz5jkSE+TZ_)^SNQ)Aw@ z=i*EuJz2>NvBV$wPLiqgi)PE{FxEMiDD1hSDm~o`*&*l?8+O)Lc%}uuX70;*^C5T@ zB^+3Krv*00$CX9blSe6~NlF^!K8G&T)i$0rS=H&`4=AT)DK)mnVvt0=c{*e==o?X~ zNhXbCZG+lFwOL5V%CE*CjF!=woe;<`sc`gps{DA+uw<3@$Jzx=4JOn2c2v5M`VM)E z-pfvTg`rFMPA+=ib_^#;oSl#;ZCGx?LT6SMwnN7gdz1hsnT_5OwAC`qNQ@Rt zN^KmAJ&oal6+F#kY!KPc1oy{{UKsli=a>kRifbF6Hay#`%0wO;pp$RQT>@A=J2gg; zw)nVOW{=K83t(bSp3>$mb6vKV@4q3{s;tKugwZ$fnj}x-0Pf@-Li;<%g0d1iwlL~1 zpZ0+2`A4EI#`)YkUNRW3hXJdK&AV{uN_=9GotHjCfWdEFq9+xKJaay7shvkPoB40& zW@~3-TU9*n!g6jZoeCC z49k4J*S!j^r7SphKl%C;qU8RYIQm|tlRMO*F0`?cDeAQ_%TFc8}*ju+S9s@%QcrXcSoT6AO~|U%A*(gDA1y z_J08h=p!&6f0Vd_ivE@}6PLCmnFV7o&Tw{{N9Y3$nV4Qijgxu_oE3qy`cIqq@(85RDXh z_3zz5tEuj+27;fQE)g)CCQR0^>g+~O=9P(nFPhWgw?6j702GXg)EIh$L zX?FVZmkI%MweRK)Zf<)_v3$Jyio&jDIoIi0%bm)qDXS{p&x%Gu5wk0 zV%N+&akLeK5^MjYPRH+l0`FJH0Pou`Y4O zhJrV?kfYT`-#<;$H9yAep3-jYeu;0|i?R-Fz=Ob|k^`JvlC;wwI|tBNoTC^G_I4=p zkyOn-k+*2;7hq@kBv7owi?0oAYhe8`5&GokIaV9q z9HZz3^h`qSZel@*>+HW&j9lV0xgg5j!J6Rhzhu7kC)JPj4Q*?M1jD+7k)FJ99I(525vwBDMEu!VVY18mZ&x>5^iN%vPQ{4tMlK+ybS$afqU=4&(clEQ5Tz(f?Y%7ns>SK z0;b~H5Hg}CMo9p%bYn%6e@wsrFM#MixA{tcRsxehXTO|Lg`_4lWWBfp;m=h|%nLIm+Lj3J`|pb# z^mWShzz&Z>mw>*0wJyi#iwGXN^v{qrk7(u*0!9Z1n7v94y3hBq zGdb(Zl#-U5JJ+^jiU)ivPuV8^Ka=m=t^a{bJ6R*d?0%;G&@eX0Yg^s78S{b$2b5|U<% zdeU|DJ=*@9aV`tN%Uj+gfpO}Iz69NXJ3+6%_be04v)D*qKIaZhx_)BW78x)o5?p^& z#`Dd{t$!%~kn6M;@%D!d!>U@lcQ)?Bz{=U?Wn8HLg`3hq_ujYP0*wK4)vM3-XFZ>L z<9v6cj|E3KX9FsTclH18sWM#k>ewc!n2`I{&Z;}HW-~Xi3A*LX^K`N-gI;dz>uM|n3QN@iTDiqxl_4&a?+f-?;Jz}FQ!tj(k+{I6q zCvvvKG);ni9x())GepqFK{qN8Uq@>j(eCg#gw)3MTUHdjbSU5uC@villo=W*1rk(` z7A%th1V@`80DeS>b3pMp1m5^Q zZr#?D_$mw5c>8Db*aP(%?MtH>+Nt;hn zNt=?IGHTqg)n0U_Hr*9o`4;BcSn9!}2pk?*R0{%L9$+*vy*=rqANRp92VgWp6^>W7 z2v?ksJfLSNH~^R0(=J?UWYiBiTrHkTbwXUC$Rqu}%x)pwu5&cK1zrJ`J^e40oGo0+ z5Kaobhpd#3w#<~8;+eMGlmb91-SJO?kQFTpjyePTc0(*j107Mav&$gqwo(b};3S{y zPgL3G>DhTP9GQeUzu|Ji+p^1a#etSN)jmw6>p49mIZgbzYJ1r~sdDwhbK1Id-$Qb> z%5wYGxjJ+e2Ye(33G*hi3CHLqr>61{+Vj52k~gBrO~}g46XvgVWB%ZmTg}d2WzXNq zX54NdEI8Elnmbdr(pB6&lifs-JJ2n-p;x^oEOcuw@Ud07mo0qbS9$F&6gMsW+fDh0 zu!v`&0G^=;$(IQYrU+#t7yYd$i@(U#ya)s=#>(Nr(JQWE&zp8H)=n#ij202u5r5p^ zA`>WaA1WTzD4}D(qSz>j7%E|lg)j)na|o0Y-Inle;PUpA`umh*ESG-P!xT7&yzrNS z1u|rbD5QGIc*x6?h{_#d%d!8T2{2dy6B~dXpaL@o_@8;l|J@DpPu@W#F-=CV1jmRq zO839>4ynH_8UM*UuqgUjW5|;z4YICN5<~7AY8N7_H_HNWi{B$ zNN&E^qVZkH>MUzQn6;@kEKVqA16))2E{V0tQiIOWlz9|9KU+*3L-7o z93(fIQf<0{)ntlOB}c)BY0$@)O?!DXwWB!7Ou9i*Cb6?4XX!V(N&DwM^wpF|&Q2 z6c_CqZPzfWhG6T2xoAfp^W3C;(JEaYlSdW@|HUrti+HVrN<*LxS<(^v`%K5-UyYR% zg*#v4CW6dg4a%9Jy-cCm17BDbSC8M#dRMo9HOljU&uMH4JAIuI?yP#d9{$LG!a%-A z^g{9#P1_(fy46z1Il$Lyt$d73h70U^EJktyDrb=&4TPg7xtl>STp11C3AN$$=ELR~ zVm@F%zMn%z5CwOJ;~3^H@M2>cVus^12hWJg2(RJDGc`|Q$(hu*Q|M{uKd$OAih2CN zMt)#lR@FAp0+hS&d(!9ik$TZO&p7aIr#p+w0vd{-`A;*lFrw3nI4SIMdr=<>y7nSI zT=5YC)=XzYh5vO*hm8DPmM}Nj)%&E-PidSc{*z30&wD$;WYxG1~}>>QYGXAL3? z`}Bwh?){aI?1v1ms|&yS_2Kq||3RMj(1Qv9a* zCJdH;7X?}_`nvAyjG+K&wK`?J)@=$CsUt&zGOmdb+_8mFDwtiKdwr$c_!2~_Q4Fo% z6ICNnjZ|{iM$h0Ax!8sc)%9P658;zU>zR)y;xZ$!og5<;ZwR#3Kj>9k6K7aqk1az6 zyd#E6@t-#)P`aBu{XS*+6xk3OC3)~cb^?VPOD!(Ajr@DV`NqU*hS=p;bZatDtNLx zOE|-<0-v?ZlxZZYHtc=OkVGsCJODpKyBAQ3Seq~6j1kS1UHOlktm)+nfyZiC_$)|#N#>P*P3wB$V2+5GE6ygI$o(eqgEoKS0UBDdPJ@z~(~t=8yydUfFKv5_pV zQVB+WZ8%f22m$jH5AmF_k-Q7cq8*R0yYY=n zKK%Gh{XL*T$x+|n?24*K@rb#||I2oP$#dh|R-8sO<4(H&Gx@UNnn?#hp$W1?73z0w zawPxLV+G$81|q5FDL@!v|NSexEWnHWZhe))gAE>6G?bcyz*G-u2$;wgPRq0hM6KW| zn_`RPtuqbLIT@)6Vc?ZfFtg0LPv?oZjM1)BgxNa4y$WuQH&-|yoOvPT35;UgQbmMl zS@!dwbGSI>B;QMi7j4u#DWw@;{>8*8f|5jw{+Y_!r^xJYJj}zX?GnHH^MUrsZ-V)4 zh1VQ}7>`-&UyAL(%G`jb|(Im|WSoWqb3FZOpo*u+92zE>)h^dcynDXg_p55rMkQC z8Q^oRx70v5MPGlYl24JQ($)|c(E@7}hvj;|j(2ZLpjiP|8*~4C!IzvRN?BIPJT+al znbiP;KzzR+GIJfK;h(X!zi+TMN4_n6o5BPDa8^y%|F720GAs^dThk4}f;MgmE{#KQ z0t5)|?!l!AE{!zqZjHOsI0T1~gami@;7)M2r0?Em&phXz=g!Qp`CUJ&R@HiIecxIw z^|hn+>??-JKX8l5qXk6(nM-4B5}%uR#&4TTI*5`vAU!3bS8=mqEEI=WGqlJ~H`z~T z2WZzFA5Kp_DZicz;FXr1jNSXLtPqcX(QRka`4)bhL+fFj)3UFsN!Dvt%m_;o_PO3d-&G z#U8-NwguMr>B}?Qt1>wKxeYf32Z!!Ep-%HD zc6ZS25Y2rEo{%FE!7~uSK=#h%-p^G(@k?f;mImY{2Jg;*sunz;qCG%yD&R^k$>h$y zn9w1mA}RBIP;QiGE+}Z^Hrm~mfFJS^ZQy+shWD#zCw!)ma8j6_S1gWI$QN&z2#Jrr zw_DZ%EJsj10_=w&Z_Q`m+t_R^jR*_u-WFaW^Q|6;ZV*hSB2CbhN!asFJ;-#ktnlN3 zN2n}CTtvMehr1`}rKLqAYJQ3&Uh*YkgoI@VXvjq!dqo*|Mb%rS-!O~h5Jea6SU*Mr zo)Wn}MzB6>;PqvsRU9@Ng|VNIbZ zZr(Xj5AL`XY|r}2U2~3^=(*pr0w|IJzde({m5}SmB)EGZskG;Wog{$922UYlULh!7 zAMNy-44{kudC3^hR&otdo*eX$u@@bDe4AZ6ln^Z--uw_z&+QEINhHH3Q!g@Sv*VdB zG8;kxjB6_+4hz_AECe6y1b%~)NCLqwcw7?nq)M3AjEq7 zx+M2Ig)P&oq)2WAGeFL;RPo3{xr&vVxESk#M_UA_IV`fvr^@Rkp5#ihrSH-gs)jlh ziY8fZna%3oBmS|Iz@f9;X}R2mRq-V++`|y=35C03>cw6tp%y6`Sx7Ox;cVj;zSF9x zgwbvz%Nisr#<@Us9B6>SpWrybMEJKVnC6wM*1#z~#oQQjzw1hO*D`5-HJB@5l6ln@ zu|Y{vLDq5=#9RoCmZY)DrRJ!rRfO0XKjv+k+S&urqj$9gZB@6TEgo%X541)?LkoM* zlvb~L5R>r7V+}QJ_4*;_e{awz1i%M~Af5gnTlW72VYoZmE;E1_WFFjxOjjDY=>HI% zDg50l;~&u(ld`73e~QkEWP?c(os5_94||LMhA@g*#=pm`<1)G@RZG+x{yjcYmo}&o zJX2*Il{}h>R3iGre>fUjCLY3zra#hlW=O+c`}kT}pDn8$89$VdX@-xG#z@n2lS)o`=cD$ZqQ&gFzW|l;|dOXY)Rgm`4CWf7XT6b6-D;% z7X_$jNeF~!pf6hmyE&MgD(K9w=HS${glhu@ar+N#Tk$FJ_eBY`M;SDdUK^kt!)ttf z^I%j9p&gi?|6P%H69;Wn>cPcejyGfHz)re(3%CGy_w!0ElQ&z>CZ61_l1W!|nPiU! z@jPP*W#nw%%Tw>W-*EfPL(2q1V~yS~)PM+WICOrwN~F^@%-W4+Az(dJHOUt&0a~h> zmk0AM3mwAW8L*h-7Sla!OGt)Pv8u>WfSBCf^AXGkRn5~o45_=Xv#X_oRdsecobqf+jTXo0P7V=S((8 zj~>*0;$wQM?!0C~p^Dluv7`EWx0$mMcr|b29M?b)>yY>pv&Kbo=VoQk=UKKZtb-ZX zW#R|2GY{Zfujfxrc(6&_#h`IAe7gzNEv~HNOyg~1%b)?2p_w%1`At7Es5po~2~-pY zR5}=Knvft|y_!&BcWVY>bzR1Cw4wN2H5vkd(kyvqwGifVtCO4g50PV#`BV(x4dA5c zEu_WyFVr%(>f-SF=`ryagyke5b#sAq5nZ00B{UniWZo+>vTG%;Ozu~Fc&P*d{`L=E zolXmm5389*xZSV)Z4WMY3KNinRv?a2nA@eP0gs!NO?A-2q-vbbmNZ+Zr(KWt0iJyY z6LqYBnzx)?TRnoRqJY*x&d+BB^9fHUSpgIuQmwd3y* zU3(h#ZBO^jhA3SZ=Ul4K&+(BH;wO9i1y8Tl@sg?}W(r%^2@o#g?%SjJAl6CGsA9) zlu)fnAiQ=jW#eQKz}DM`&h;a&}T3wb8VJkJjMIiwy{oe7AA&5r5>!&yoD1lBq!2& z;c3E-LH4v*c!#eMRpI{Aby{AY$+axrIM-+`^Wa-LA1odq-+dr3G^7f>O!4btNRdz& z)^xfgad{<=CDZZxj`2L{;>zS_d6}$k9R@S<_g#R;kW%=wnp>;(l)=}n)N*b{2qkT} z40{M6G8&;}sGp6W0MBR_v6kx#p`Uuh=GG zu-6HXD2Htoty-37->cdH0>HX8=S*FkYHg27{14Dy9v!`mhK``WAgra>@j>2rB3r5V zxUtUpF9@S%)$1T@GPL$r4!dre1MIW~eigtr!}kkz3D*mogx!|IcC zvV*PCXS&U;=*9|u4(Ip91{E&pG`$>79_aiZ*3w%W`%q#WsEt;M%_9_6{EyujEmo>` zbmc!Y?pfyKz)o@JX--jx9kCi$?iGmu)I7G%cTcT%n5A!qakx#)VzW?o^onfF>*Wc! z0cgzL0K9)en0Z6~E+0ZAK0-$eh{j8nUygmj%tfF;G)#a&r^*8`c1_f4@9E7`Z7A6F*!~5-Uv-$`vv|SH5hh8WH=?b=i~@)syLWu8SC%m3BvZ{k9`RJ zTX}hvYTgaDEcQnQ$P|;-9?0^F46O6ZLuG`O2MsAYKKFVY#l`8o@pzmuMS`ql*9eKO zX~gh5tb)eLY+kKPjh=b8RGV}B((idBwis`;86TbJS8DR~Dr;&8RdEn0K%E#dR&30u zJ+`4dOKcA{uAx3T_^3rWRgE%Ix*sKW@hCXBHqQ=I8^L#EGH~S!Gfg#p`BlH|akVi* zu(t0w=FNIBaak0f#}f9w@ZIM{fF5nZP26#?U=Cq0E?-dx7%a2BxS1Gm-&|V@p69b# ziLc?4)Av0zwg}tqxbdiv%O7*|f@aYnxH>-AWXhfDxbBw92W8lr{L0>6-qT-ew816Z zSo#B9l)w|}PCPj-=6yP#G-!V)Jpo5+JufG;3cWw}?RBzUUU}{l;f&8+b)`4R0ez!q zm-Oy@qdv2G(dt#*Cwj)Up1WpKAo6Yf+3^DN?fAeSUQ)D1uWzhxCj)gm@fiP{lYZ+V z(Gz%{sXf8;=iM8iN_?KOQMPA&x&&b)de! zrB_ytZ*a#U1r6)MQqM?0(|fn%G6XulovaKME&%+kqxjS{*7v!1uFlK<%i#yifyNm} z#Mh_XU+IBg$VHcO8&F-P@O#^Q0%Tn}y!;LbU5__?fOmexa%#uccGP!9uiKPp(LXLR z1mz0)<;wZ(kwDA(-AV_x|4lhspOXAO;33)3o+7E{y(>hUY`sT}_ee9o#{UZ$pmZ){X<7`Vsabwt>H_Lfz&AB_h4h?*jwVBjws`qf3J127~Z~TvZpn zHKKw$nq9NgqcYoq3*o^m2Ll_6}NG9wEFC&3+Iug&ujt z5c%iUW>hE=86Av)5$^>8&$QbH48{cA$4oK@2cWvyS^K1h`;-W{DfLI~VnlQ3g`j(c zaKl61$j6;$x_`RUyVHxg1; zyow|gDAmpBHlsrvM3pjqOaws>+eP440>BQpZ)AyC_t5|OQO!aDAT=mh3X51*S^yeQ%UaX72GX4l#Y##PTouxD7&0sr zGHirJ4AWDuZxd_|BFY*_snMiK%KZtxzw--KRO@+n_(CpO|6$aEKN1W_upWIoKmRL6!!$MMhx0y%3 ztE0nMh#Ujldr-k(1)^6K;_t?54g#NrO=e)k-x!KW49Nh{B8twUS5<0a{YB(Z=Kn<< zD1VoJy8oh%b-=;DOFw3knyA7rB|u3;o zorsj6zkPMD1^?UXqRC$Qx79_OMT8x*rE0F4IoNDXUbb-bBwlOC~vtN2g>#8qDIFg+}m@YY?t z){n1xh)nal0(VD+;rRh5&}(wfg(c5bNS)3=o)@Ytm>>|L16uKcrr>EU<)~-TeyYW2tV_rJ9EGOq(?i8zNG^WtkKM#arl5cRo`Jgai#d?1uMQ<2Xmb6DKoD^Yxl);cf)5t=XX_qaum%5|# zV3A8nm1&l7x!b-SdcVoSoMmMJ0%V6Y?wdO)wl3{_1Ttka7kMBb6az98t3c+4bc@yr z&EG34vr`Sh2Nn37A1QO<>O^uXD(Vn|ww383Uf`Ok=-st^n~NCwjLw#Rqr!Gp#a~H> zw1(`qW&FP!z{#9gYfZB-k`0bg1+3gUA6z+uJle3vU9|o@KrMc1 zhPw9eUP{T{)|~R#_`_J{xIbwBH}k)q!$lvNZOuhLe$&Xs08J$Oo4=kNhsz-loZV%J zwe9$Fgc~vFHp;i)a5W~hUvo7sdUt#^A%V_uJtG~qb|pBGpnuV zcr&MOTYEEa>~(UpU>3=7yJ(f+c)J9K*WNBWw4L0pxD0UI{q$IHyj%6&uf1FIyF;AZ ztp}lV-fx7FI^AzZG1uL1`D73sbgGvT*Bc_Mh{dTRPgNiYEq8t z_w}rv^Y5F-mlOErzK{aZP8n40KK=J;)UywM9PCSqjye%L*5{VGlV{gh>WJUB`=;NF zk&z;7TL36yQYN9o#QtnS0FOQ~lTBPSk%nNPUm*(Z0j8JsB=NUcLPQrl!{>yN%Ys(M zh@H7(BF%*cO#A-eD{9V<=5zkq&Z)?57{Kf27I9lc5u4Mep4Uf7k%FVS;V!CXukkkJPaFHl$ttmB4gtoL@LS zIeD*VsuVOtdZTAYFsnxGdeiT+xjNXP8233a4&`%k_ZYC zrrq7J{xN(5%Q*8XDzOnNrjVMB5RxtjrjScpi@PP)f_o5|6}?POjoFLMal|srGVD$> zsRln$h_&okf$q=qMq}yCfQ$t8di2kE<9mTkdVsftb8(R@`J3N}~<~Yz;z9dW}%NxoqS2lwxBU82_w6WCiCI)a@sA#ZpF6^i-!4TsYsqnm1;ozhp%Z~A7{!AUz8KFV(E_>}16^y4#4-^aY>&Oz?s z8`2nQN1yK%hUDLyT3v`LM1R?li~oQ$eNL;U)@lzAKAjVYl{Ns$eBg(i+f(Dm}h3uX;?wD~Frm zUt?}D>r*Dw_)Hu&dGN4R4E(lR!aA?CTQ~tA+YwCD>G+55SIZk8Cj1V6qcxBD z&)Ek#zu!=Q?ftztp_TFG)db#S$MOtvb1N#2j+8HDT9m-JJh6(?ZNFXmj=LQPoM2mV z#G{JE1Bcta%I*?1ul6o=`^%G;+~ktiqCd8eN%C*gU;^*%4n*=5eNo<-pB!KJpXNNS;N1xatn4AB z^w`l)_FxTNNO5!PCGv>MDd#>s0rCn1AbOPr2HS;BRy_!Ad2MEQm|!tFdw?n7s-j0tShXJlE_oH`lpt`ZdBykJVN7p-wjh< z?{2I==T?)-&JMv|%CD;~Vk4hxG2ZCYti1=MQM2zQ+CPk$imUFpbxEpYHgp>8^rK3O zNjhQo_91@UGz5M*wmgFd*RUD1PkcFbEhxWM4q!p?8Mg~l1D0P`(+m{&U2gt?POSt; zIKG>)^Go3hw!(V&)b+KnDll!VGLCcOT?)m$MD4tUY zfd5v?+yP7gnSiW+rAy8)$^M5@=KrfRxLUvIfZ@DM8bG;_OkHv-PuM@EaHqah>%>@|ut_)(1-WU4MMx!e)bBF#;%tYy!}1u6exHq8w|QaaMDND~`XnJ# zICTm#T4s4rsFUq}NR&7kW2n~018_u=1<1~CH943mhS}zkseE|u-Xhi`g2FNcx@X4< z^_%XkC{QHuDh;epM`SQCKd~_HrfC@ZREKJxQqb0#8e%v=f5a%*XG&945dv~r4;`BF zs92Ah-SR)4tYfvSuIrpv z{(QvH~~>}p7|OfM(ZOb0-l*$C$P2iWqa=nmQ;xKZ;BAnQ#EZBVi9yM6?!3- zRJ3o%F49k8>33SJUQY>`J=@E;nP+UxI{Z>NuJ60lJ_3F?&B%Cl7xb-dwd%XO{c}To z|5-L?R)Fuz`^=FP7@#SR)6ctUv1ydG<)@Nw>#iL&Zr%PRF8brtf#9Imb!8hD))s45xpLpHP=XA? zEwxpE8DDVASqxXPXvd%RSOL>Jw!2E3_n0Iif8U%-t=FhS33x0D)>*lyjBNJDAVz*ZCKcUmw=AsMTL?<`YosK(3u#b9P-+s*iCWAau>0qgGWh3KA2Ir6iOrd|fJGPX>O^PK* zYdjb)5YxDw#a0d}2W9wGrnZ?@sWQ4#B}-GBN-Ud3)odqs^+K%y8`Em9^>O+q>*Rcy zY|okHPyBeN()yp+D)Yo4n7n2N(2OTs-gj~$OhY?bmK#3*v@)POb=RfN^7*$yE z*pH72*Nw%BdWr7{6IIv+=O&1Mcs*xlUlCjY^}udz-B~f_un1VW<(SROog43}T5(q2 zexH0ti~%^^{LWiMpMRLky*<(1q?OO^*>tzDkqkxT`)4;lOdSL(hK=R7{JLwX7WiYB zZT47?^*IHy{&nA}xo0eL;e*7Gw3)`h%{3tlsZ7Qq6XN-WPS~w-z7_) z{gp#_x&-1`(?S#9jbPJj3m@GO8Z6d4r_5bEyI~Jp4xNo*yCuzuHA&us>GFh-%?D=Z zfx4sdzuwG~Az3UC1y_(~YHH-pHZK?#@_3pjivF3I4c}r+Gn0iaEzs&LM`v!v7=;a4 zr0&eYiQ|<%_^hd8|5(rt3sRJ)O;EHJ!gu_zlujAYakxMk^uzI%A&){+VK~*+X^7eV z#fO7K_Za`qSx`B-;$*~f+3;#^O@!QG(e4}q#HLRCVPL(s_Jbg0Nf#@$)F7!7)uHqU zrl3ig)?@f9#SzV=4R{fifQ>n_avD+Ps8raIXO-ffO?535R&}2HiJh&w|6>Jb-MjJq zU!SAsK-=unED=u*M%V5q&X@gWZ2f+(Gs@GDg{?k35%QjE_`niJnhViyoy(f|D{8B56cOST)yV^@HjsLr|@JbZ$gCpNOE{YB7z>qm|Gt``kC(IT8Pl zjof-YCLEm%i_OL3^%2!;%4`>f@~zU${bfelv{@>KyIa)If4h&n26pstLx2P=m;lxk{_;~M?AxBNCE(&@oWPCyW=@6f@X)$f$>s;0GnrHC{ z_vvw1;O!T7>qIOzJlhrF(rMUnC{xm)c*pP1vEN7VKW@c$h{)CX0`EOy`FZCJ#?&>5 z3s;h5jR90ZhQtVMDzzCITDqCb04g{~LKZ^2<5>0&)F&h0Kho{08mQ<1QY-W00;n~rF~cc000000001msR{rn03ZNuVJ&BQ zZ7*|lVRUI@FK20HFHmx2WMOn~a!_(*c{O-7Ep{+2XK7|tR1E+Jj$j97x=%c2v3M9} zb$AN^0R;5{000CO0002&y=PcdS<^LKr@IGeMPR^yio%QpvzSp(dd4xLV9ue9U<4Hf z1>sCEj1hBI)G;T_S$k&85i@4Y7(h&b1bM4!AA%zYdjER9kNd}S&GooB-RJZ^wO6gR zs;X)8CNA}QZ-uQe5gF~n`jmdZTPnGH4=`{h9bkLgEG>ZuT`*#zoirr5}toV1LMY$H0O~O|mn_^va(Aj1n zpO|6SXtGU(Ss>3Y-)=+qQzINCaPE70N%I|!tyD1L`R!9X)_N_|n1FSQo$D^HYBs|X z3_I6fUfX)r&khj2;p)cDI|iL=0rK^&`hO>vsWcDd9X^*g^f=S8ill-ErY#U z24e1<*|YWDkq|9_MR>&heUI9$7=TDEEbu(E|NV=o z%~p00ejxEx^uJTm&AmYW@HXc1#$H>O;Zuh6jzOmaIMON;5Dt4f~<-bo+X<=+E14>38GO<1+yLEb-3`t(Mdq z75Atl7`#;QhnYjyD=`;W;*-+fyz;Ewx;85;q4m}09X55@HNXKt2NAuGO?T*42VkcL zf11c^vEB9#e{2M;=K~xZ^FHqJt$=Fy$Ta4T-J{}r|Eqz@18zsX?t5|uJu<_a{%04~ z_B8`TO?qTg9edn-H}HxCn;gOLesI*b{uid>BZJ%fM!kA+XK`H#{vL1c)bn!?@HB?_WJV zvK+rPs4Ob`#M}Mz2W|^{ybXDhrBmBYKGi{3F|W&=471f*fc}mB={XfUZLo)E zhqDW6cwbu5oIb05z?+xP-g*Q4Ot%qT{?cD9=GQ0!(G?_UySDA-LGK53P~mlApTDmZ z6`|Kc`N0X5Z|)p)nVv7)-aF#yrOl-QdeU9QyhyATZTL+MTX z2dJb4zYmrJS}v^F7k3$WdoFbylkliH=vuF6^k?zfUFXnOT~KFKa?wFo0M^k{&~@D0 zZQtnkr9jtN1t;g#4k|wM<__FteksL=b>BDgu_gGG@a=gh=tCUuvcG@VmZ%f#0&onU zLb^%3^@x6_W>+u!DwJ#v?;~iHC9-f+CH?(Zp7`kV_ z^s*trAKt1V!i(-X=-y5-{fUlv+oY_8Jd@E;}H4QNT9Wmwqmspql=YJe7W7mcTtZM>-c zK!8?SkV|!V-hM-RhqbuPbT2w?>2WCJeG4#j-roP*vc^I4=mFn8Jic_NxZ40PIhgf4 zJTC3ycQ#`$3RV!-O+VOwU)BGCJyAET}boz9=5%#BMIeHmEIG$qjs`mP4Y6UAt{PbZD($B6R>2zT!E-F&d zzx>6OztG3)D_maVx%}rMpy!>VKeu62rR!THkR3o@`NsAE7gpBxz~g84uzUYZt5mnD z1fyR)IJYriD+eKy%7M=J|CykMX1sBx+cuhBdCUlad0IHRX-v(h=a<#PP^z!>?9v*q zRSjDKytuZd*T2(?EC%`JZr_LJH@jkRJ{+nxuNu79PX*JyrDG71+{+(gI9OUGLR}U|2wROik#Ng}_rvK@8 zAI#fY=4?iv?QKhvz^+kX}_4lWqp z_r#1cR^YsFME^7Mt9NtMfzJEF@}DLaFT)|Z_1cby7L9rVpr`jpH}uw?;ZLuX0i7>B zXp4qR0*605p@!CcW?fj(bbiC7V}UjsL@WоgZP>Rj)_UXm7osKjFNTj{sygX!F zO5$r{h*%Dk8mJZgBSCrIIfQ?fG^-BuQ$p0qZL1LfxYPcZPci-}39<)${DS5?)|^}H zqJ}6w#&o|ID$`=84?r^pFpPg%6rr0?!S+FSPVd@yeG7fOZuh1eh<}X0z^|BsfAEyy zkGOY4Ru?X719ETrh|XvCZhLqlOif=4X42*vs?VU-V@ym4kE|(si0%*^X!ec@8{^$Qf z*wcz`Bl^bWr(WSdslkqeQ1iJ}``^4WQ2bC|#y__}{pfrd3qxFwgBTxqjp5htjICQo z#mn$#v1*8R7$CM>)BOhGALWaAN3Li*!?6MHp!Qn_+`0K4cThjtL588X_Zh!@a?aJ@ z8~wLrin}9S2ait8HP_!ogcx&iwfaB!hrc4@pQ~$mFA@G(+shpwmA~RWcLS)55&!t< zc`$Jl7XKTeSV6ByE5 z`m33-^U&uz2g>W(u2QYzN7i-Epfb^jf3`};XO$@&U70QzG4w@3mD_uj3*J3AHt|jD znfY`&U%BIHF0{=)IcbGwRZMtG4u|aB*^yIQRMDq@c zfAz{%&-lmQgkJ<}!asd}nMBXdkMU1EBZ%i6)NG!qGdQ$>O2#|=>)GjAE|lO3y^pZe zRc#vo!3t!l(Vu1}enoNPcrX$f)P8-t^&OR0Z13pz$;F0rj^Ha7AKG=-;2U^le#K(0 z6>)}F_+N+{oWU;|e%B({SwR0}T{qp#8Qx&%lT0Im$LJ9MARyx>w5mDdAF&1E3IFUE zc!3W#!aoUVt_Y6#!)(UL(wT+k7^8Il&!lyo@V{K?=KT}P>UDvy0Zg4kLv--|g&O?$ zNxgk`eNU)usJJp%5dJA#%B^!4eW6nIgH5WhJPJ^e!@==sl2Kr)2^i#=_Qo2QKWl(I zmp&9~4v?iWzT2jGFepi%}Ea*2@X8(*B@VW_EnM zFL0mH@KEx5<;tla-o3bW#Fv7zl$>I$;GaU8Fn&Nb4TD|q%9D#4cek<7Jv=@Q+&JPi zn`Lio3Sn&@!#t_`;b5)Kk*<9EDB~X`F0@`r_-8C5=p_xuBmTL9@qZKlJ>nnZ;O`38 zA^amP;YhQ4=p8e;blqXFe^ji>;6XnfH(Z* zwA1xn45(W!FKeNDKj<3b7G6X6$2zQAq6~Fk)e&l8NxrNpyvk^E$|OLAx6b3o6PvJ)rMN)@S9X|RK=?sI&S2!bbU+D zBfk~FOsflhe6x+CtK8Z>ObwIxxd{B@3&7dCp74)(IWzHQmL>dCODG zH;(5j&&^|NV{q=tVI`LFPZn#{50h~qrq%HEmjeNW&uIE!k|_@^Z3 z_;?Q!_(u(qe7w8=YDD-)K^KgFRy5>og7{~^)%~>rwlcbyH-zwy5kzq8i6;Cr43GDC z{)k>jLM7wkHux`?EZBEs{4-4pVQ!3ns#rn9!jXi3LLD(z7c#cVRvGvw2=iCn7{))#mm&Un zb^pYC_g)yA6JI~Qeq?!ih$RBn1-}vgS$A=z2Fmw1P5e(Q%%*g^nE&bH0O1>%|0&`D@(t#HHqHZi zkKaSYKLLoK1K&Qsb7s%0p7>l$%C_7aasS^5_f*zuVL!MNBlM{GoC+Pbv~Lay{;9{uyMNpx#6K4R{@_5<<}boOrNGcm1%G+2?t1n4 z)eZQrMc=&oqEkfgzb&ElwU=FX4!pXCH`m_bkI%2}Ru|wP!;R~&dJ+DC)_8gyMf@}1 zc3^dU#q7Np|D4glkSDiqy!DP;&WzBzfmhad*oqj5PxrHHdfZAFil_TaM=&HJ{uys> z?kT4O;h$>WcbC+Y;NohpwLR~Q9mSyS?wyo?`1=Uj_|STGb*uNiyW;>}YazVXy^rHk zKfEGF$X$o{$Fj0de6&sl#cK?SiT~l~L}phI|4c1a*9fBdWot5v@ee>pONf5M{LgWF zL>cFo5dU*(Hpt;WPqeHuXeS?Ih=1NyxUx=y)*O{Kb=@-}_IYiP@puob!u*dnkt>9M zMm#<30MS($|J>nV^qVo^9}LoynNX_!=;SP?c8GEC6kAl6_@9y9xXZ$-4rTnaP76M< z$+cfbZSTuLDDL+DfVXiNg#73!lpmJr^5)TLHTWPde8>Dx!=<>*ZtfY0_~$@?3XadN z=KIkt`7PZuQN;==vO+QnGIP=-Z5h%6Hlw zTyo?S02j`*d@d6H!AxtA3Oes0{^#X&+@%rYN)!Jx$_nC32m0Or_fK;M#gES{tQ+vr z4d59gqB23FV{h#(6Lii^n?E~I3I4k(a8swkpHUbI% z{8>~Ddz%5`9~;-PsZlfWiQ9WDa{IG}Lkksj-5;#2x#sKwW`t_4J-^f~2$3D1>XS<` z{sCBo?^^8*gnvAV5xTay`+>1(&*y@?t>2kN4W=Y8g5FE~&z9T!B#7c|b|^R{Ijyh@ z$bW0Vy5UrpZ4Zw9NsLgv?T=2)snpI9>=6GfaSe)NfJpdfT73kFn;6^detBc(;9Cfm zFnd23l=88NTaqtsGi&$hZu{OmKCOj7vj+R$J-f1&gIPlC?E^vSWkC+4dv;;`^HL3g zUK#kO*K)k))RYus6M=us>_K<`=Ltp1H1y{fRIEzRqYK*f1t`OO$)je(|CoXrW2H^%HnsbZl_&!KK zTM__IuMidSKW$dvCU3UEpZK4}mSD&DXKPp99gKe})d6`sBN@g&_l&@9KFIqS|CAtd z{-er__#fOscAlF8t3NzGl@U>m$EW92>uUyj&faCSDr0LsyCgxlBj{^0|1+BOjNJL3 zw~sZTo5B3g6)l`({Bw67@kqJzKkn`le3UZyALC5^r#{AA9rHg|7Re8=>H|5>DgM}q%x(<=N=)0vJP1pl+5NtHq}^FITQwE!LSKgCKT>|y@rK*uVd z^FQ|v;q$d%{%2wlL@CVw{NRpghxs4x<1MrEKZJie9;#m(AFKYvyatO$550-s2k}n; zUPucH|1%A*L-^;y3U?=v3IA;KK0+Kc;-85n+{2k}p6 zXu7~Z_{y{XwSc&1=6|;R*jNQM1pb+>fk`&axBqo&HbUsmr?#&?GvB$FCFq#{8F?2U zr+i^K@jn&kfV`^N?20{3`(to!y{g617P~5T$j<*vD&s-?Pk-FxhTWTQA^!OXq0yE* z2N3_9Lue%UpUMkBCjG1DSv=k$tenTPmiylH!}r~Q*(G>8LjS5aQUxnLmXrSV9IwOp=ha;;q+5G2{<(k`I?Vjf z5duPKjDNN+r`L}4ujs#j(}2SNm?H#0{L@$}S-TKKUVM0Bh6bh*|I>VqBi>l#rN@MS zj@yIb3F}{15D|HW|1{prp*}`w#6Le4i@dVF9kf~PIkyEEi2oV(c*9)$uFU^j8-gju zl`Z{mU5o9uM+0Y9ce@_f<6r<{ty|Ztf%NNl=EcDf-}@*6L_Ot6?%k2CQ z;hzC_79sxG*8f8B)`)*@f&Z%~MR0dSv;GyF?t;5(4&tBNA4WdaKug3w$)jFg#}50kJT)-wJ{pj*9{iGS=QC{%**&x%GWaBRED2l3Ao(!ZkGz{#*5 zo2tqq{*jUJ-px%&IbYUS<5w*S0hNgAXWj^Wa3pKY@51MgOt_eNudHY4UgAW2%Ek zCsojT%aDa2pJn{Bkkp+fvl;(fPS-%hFvdR-T3E;Y&;F65eybkj}+j`@5f{Jg3U4!t`c)0f8T`)~;-^{v zO1QHa0P#tBb);19l&@y`cD&=HJ(Qr=sA z!vDzQO-t7(3e{Hv>0dkZ#y?T4e=V)Ag(NNOUlT1sCjO`8;;IhW`5&3_PoacL5-|R$ zQ>Cj4ZtfTOXK$YRmrVGl+hN*5gnvGoCgU+C@Xz~!cw>8+|GBYU3wv4rIv9XA#`@P3 zD<`nq$NCrUAQ|z`#%@P?(heg0W7g0O@z3eS%>TT@>x7!SA5Q)72CqZ<+H{r!(p2L5TH0=E!N$+nN0&D6k>Ulz7~++toOEyyL?J!!e9wl_&zq<@Y3 zh&w17|0D2Ev5FF;ci7zX*rE=DJF4Jd$akeX;(LHclx}gy*si;W-mnCVPI2A$kBYtK z0J>MKf4xVHMEJ+HNigD{FmmvN*r-5LjGF{*KTw0~YbhGRGLDx2S zu>wW^y6y-%#6O3|rN@yPFYu2Ire5ckx+DG>>#u@My-qAc{1c1UA^s=u%|pBn>tC_g ztz_F?LjO7;_@6Yjz0kkRf2?exXZ`EvV1zx4f6h*t2{PlK3NHOS&ciW=RZ>` zjC5?Z5breoB=bKFgAnbI{?%kgDF;yWuW>B^2>*oN@W`Azk3<1BOk9r4dj9~CJ4 z&sDt6^eRDhVsGxm>ue6H`|9rDF@$Z1|Cv$F9YfDs=6}qVg8Y=dXH-*9_%2E>B2A=s z>7eu?HPRJSdJnxRAOevhAffjvARr|QBE1O+k=_wPDAJ8|5}Fk05R$Wh_y6VIb=Ns- zowfKtCOfmIy)*MZ&-2b;7iT+}zsq(j?CxhQk5$ER^mD!m%buG}CsRL2XsjH>j#G(I z(E1zTg%4>rzcQu%J->&``xIf1Zbj_<3euB$pao1^Z(?i2_8H*Z$Dl5UFGmhHRs3!T z8h8&3zX;agsV8GJFc-G7f$mqU?b_$lTaUTki1ib?prNy}X9Tt_s?kuraq@?Bhipd% zwnXs`h^sl!tDskbNr|I4p6X6E7ZUhHMR#$l9u`Fruljj9`?_~U55WR`z; zu;r3rJ_v8{a^(IzzOhZOrQilMi@-mX)_vTIHBbypZo7qVq;#gcgpPb?v(YCy!|&Cp zT9UpS#y2W65jr7k&+r`g&bC*U=4cI&;uw-u1U?AS&C-80*YTh=+m7-^jD-H)B~+Y4 zC%QZ$a}WRI$%Mw;UHfEc?_tm!Q&r+cnF80rSwCLoMe@*fx-J5rH9+h6F5wa;M-c+J z1F$Gw-3#d1s<{*Pm;dv(PbRxWoBGB*o?s2)o)tDJ{sWPn0S_h7myurs&eV`Qu(fXp z$+CZ~Vpi3F$B@|RZrL9nDIxIH7nF{p&(Kidx{*N~OpclfvxdLCjR*620gKMOmXKsU zuDLmrnFVNrSx#TxYv}%iz#)s#_@nXhu66cRP3$=Ts9(Gb30`KHt9X1Jt#9fxlPmYR-?^(v(%}qAZUI*fzavyH$7Gxz_y5_WZVnmZ?bMp@G!RjxA@zUqx@^oux^EWKL#0kL4i_l(-Oc7Zvtk_ zlMbVwE>FB1s-M4!r)vqx#JisxVl`f`JDKcU{pq_lx_~Xj#~bg`e z`mDS7mHs^syU|BS`1_vW=f@k;*HB~GM^`y(c;BD>pbXGs2eXL7am}YvU>ydWq9QR! zOc}?EAMZXsQ@n-(JDSGGEHNDK&910j+zn_Psl1U2HOIt5vG2rdyQwnpafh^PLW7{W zxuy|v9Hj_D)*;b>5Y9PorF)Jx%__ov!Pj}YtF>W<5@@UO+PZ=I=!-ZkC>h$l`0D*} zB6QHRS-d9_?}#549R!)wJiygGWNy62Jpc19T;;<;IEq{5XHORVEv#(($_X8N$>i!? zL{3cj-0Az5jDWLjdv8B`^#WbMhrjI6lK9Z(*-B6F3E zFE=_KIvrSyy@13jrV*4Ct zEPQkR;U#$)o&p*M#JaUJwRy;a%b9Pw;uFGo`E0#Xj_ZiOl(7Mn!jN6xU)&8{hWq69 zLkE1hyQ)t;#1L{Uk`Q9I+SktB`8MbeNq+c$+a1Y>u^3}qfs8_c_@Yz-0y*&aD_x3m zJm1`J!kRTTErQN&a1vJOsYGi@hp~Gx_GUNmfdCjW&X(v(!uzz#fNv54AIP7m?`3BTXs4j%x?T_0q^E z1D{k zVKoZie^1P_LY!(x!}rP}f0ib|xsE6fB$MDD7T`e+{DS`E3vkPn1gll>ZNi|OJh-FB z@7uQv;5Qm@=<_`G7va(^Rx}T{W_VXMckf=3;&nt#^aDY! zUif^b!FdS+yuOW#Pis7RidlQ@@bd%LAV|M@;ME1iX6JpUJ3!e(jy}p%qjE^{Qr;RB zz_+L4p<80okPZ_m^yU}TI$H*un)J|l=#xWj(xV``tHeaVI+nL2xQg~en}Q@BOfPu- zqb3;cSU{sBKHu8ZkW*v#{D~<8&X-&mKkuBkCSe3|2yk8j`E>)^&a}lOh`PMn12OheReTq z$*}awUV`mc~4bH@ssQF5@wy)|hiDCPbf!_2uM}w!h z>=5VO;kK9RPo_$!DDvllOM>GHl0GC#N$A60k6pM|%0&@nzO$ zGHeLke=7q~oqUY4OuCedRfD#47T%gQ+|(KHWPdvr0vPV2@uFPM-$@U3Ew*VK?!p&l zX!g0Xc#3-NtiyQ4dqcmtcyCG{jXNw!Btv6og|K}&&$$Hjf2TOw(yR9WF|S$!S?n*U zWtqV<2EglVgP`%5l|F-=F%xIccfpzj{=GIfR6Lt@q9ebxT)dHM`6Aiu&Geac5Zl_z zH+GsM9e8q-9+eddaK4&=ML!L4g+b&=X5u={nu%=>XrxM9RKfdR*3#ewhV6g{z`=cD z#;Y~eJ$_we-Qm%Xhd#wt#}PlB)KFRzszq&mVW|NVJ%seFzoW762HTFa=(~8geigU9 zvh6M8u?>}l-_b<9;p2k6_waq_h-H|1gmO&H%coJRhtSi%VPhwK0=HnzkK7Bj=aR^- zJb3MN&bQ8#0T(e^*?8IuB9h+NnUmb@Y z_gT|QR8YVZ7OeE)xHw|4)o;D+E^-hl&WD9CtjO))4swGV#ONOe%N8y68zs5zP;Ln99@%e{o1r;cT!snr*bL+5KYs-nq zqj9(X-^=095oL{sGILxE)?uETvDow@gY)1id?FM#ya7eTm*M2G&ci8ioDw#(3Uai2 zm{utCg*{dVx1ez{h!2Mr%Vb|KrI;?KLQhzta86j6xcG(kLjx}b7RLfATt=Elcitr| zQ>NTop8JB_k*qiBhHy}-p*5#?A^hisd#99^yLqj6jf5)(>I3xDZfk7IROToeIY|v# z>-acGfhlGyP!voWnNcG!U%ZVq$wWmbL(vP3+8>Y7HVMK%CM&m(M=41XI+xX+-fFIV z@z?m1=cwteby$E${ELhT==02T%|lRwmxP6=g)ep28+GpF>UD-HUP&2|)fdEl+ys_3 zr6!h6*hcogsqoua21rEtn;vkWgG32H%Y=Kg{O14VEztX}T~l;E==ia<;oR@r^X5xM zxx14McP^a6d9e_W&Ie|!_izgg>0voh>1*Jul|qK4;@cm42VNDZq^3V!EI9CIBvMltswNdAp1H-E+}OM`pwMyVl>A*q}+CpQB<%6qhW-G{;1-uARR{F zWw&&8@{Hlu&B_u^n`$h~MJiK{76Nk$)0iJgBmc6c#ndg~ZeGtI-dLk~DPJd28v zc-E;rX39-2=A(5UY-1|}QNQo)=kxFvo@7cYORV@e73xDFf#&<;5@@+de2?7TN45A1 zA;;lo?Vf^M+MAW^-Nex*#eVCuoyAPi&|2^PejyW%V^FS6yCVw#5OF%!I~W}5o9}~& zzEskm0c)GNj0ztGK&;po(l>D(8NeBS38b@fjyr93k_{iXb~^%oyv{bA8oSFNh%xEI z5Oavq8aMt{j)mtVg%mM`MbA}!i`p|42g(2dr?ZT%k2@L%mBj1cMGpK30#Jbp4wlar z5mt%;;#_^Zl6Xik5^o9rQJvu<{>@K9)S@H#uLiK4VwrqS93UR+)bt$i5|TH0+>bzm zd02y&@3*@Q;fu4Y5FDo0*pK5DW=(b!C8mHAM(bs>-yHSj;*UDCxdGz!G}Dy)o*glW>KC{rF7}HPW0G~(wAqhYhKOg--pt3nK zLv}Zxuf2}jj_od8+;Um;^|uwN*W5CeY|L6Cs*$C2Ci}klOx#Rp2q^rY{W_S$tg$p1 z%76Itv!AmRDQ_)m-lRG1){+VUj7YwE&be!EKA5`}j=&m>O6599;ug*W2s%9h@-b=| z!?A1MksIuGv+K>8uDz)e#SEYeHqlyC0e_iG0E55dzG(v5YKD5dl?ECty zgrceWON49)u9{@AnASU zWbdqRCIQO$kg?)A5)V)278F%L`kWPso=B*!Q9pP{J{Z?`(YqnoWJ&o2<@^*qGBero z^Sykgav-=DVz&Yl0;@9hWV6?L#50fF1Zr2^?S;K;j?==@x|n)LuXxqA~Y5_-Z842V@YUu4uDDpfa74k^5nfl2Nx1Wa*wV>r=$}pLicA0@)E~4TzqMB3r^yKE zAOTv&HB>AzE^{d$=Kr4TB4qbyNjp$i&z%(|7@X z(Z~dD^=sY?MgEG0G~6BLH2Y*)tHXnOAz6w<=eLo)f0iMpfVBtv5h(Y9M1D2^+!jEg ztrD0t)4dqUW`Ay;>*}`|TjzW>Vsg-Y?f@>R0!)~V{FY9% zKZdHImkQ*XquO7f)cmB`M+PC>{h&wyoB60Osf9@$)1ixPh`WSM4M?3AkE9H6R=pX( z1pBD$+VeF#SORDoxxS&Og5JZ(R56P`y3TYvL)lfDk3;Mid8W&sT0prOo4`xa1xXf1 zN<{d*?p>eF!Ov3mCD1}!Bz8Qc9jO%{-Z{C1L&4{bO}*!5C+4Z$Z>q#JO8mg^jKngQ zW6%RUVh?5Xi@qd6B5?oktlwsA2bE#68QXcsqV}s}?0BKFHR@9l93sY z^YW$`pstt1b$!|Fzw~V){;&Z1N3BI=vs#?DIf(+zD);!A~~&49c7g9>GJF|BoSQ_p6f(a?-M3MiKF62LfBx==RG}-KOds2ceUq@1KT=8^iiAc`oX1br*?ECA78j*OBU4n1rC>C9R z7b5(da(==iQ0=i@0CENTOl&i@?p<(>j9qWD~qp=G)^RmK2B7jmfL$J zvBf!e_=24N#J(`~KWMosMiLFZ<9wuKmCa5c0IKMIe*;PpT!|COu9~^u!y{?|)dW}z zi`wsAG`H|3gb~M1wcuA6{!K&pPTo(v*TC+c6ZV*$h<6{=@L7rwgEy}=-z_CFb3{XN zJ-a6xBQx7UocrMKpKh@RXO$CPML>;t(}ei`-hhW zn!e24WJqcfUfm^dA%$aQ$gzU&Oc1?!1t2vLn(p91P@8Dj6x{jwt3eQPvtpn)Lw1#` zNY_35_}9q_qufCTJOlXOD}2GJ1PJKt5{U?oN%-Cqv5CtDNQ00!hY>uO9ddiNYvfCE zuf*N-i-xB@cL=pGhO=F@x3I^ne;cWu12mEvznUB^ao*85zs3X+fNOsl*#~bRU1k1O zvA)FBv9REO1si1*%?$tUjxWz+Xob}BEdL6p%vl1d6tzSc(akf{aO)8Ug&U>bPJZ&5 z8$H@h5ump_@ZRA?EzxA~q;psinND=V-R>WDoFo$h_US^jloED^rbGSXLG(<~xK4ijzD_)LA=q zvIDt`gI62;EXtw&Ea(%(Es zzjWxw=#iYYP7xB2OCIU1PhB4Yg;kd;6DOCX3_yEzAZNPc4;t@uKPpN{$oUPM4sLJ) z>Kj_fJi~2Keikc6zW+)JD%D#!j{&iBYU#N+kmC_e zO!zY0#<^lv(WbWpiP?F{{S|$>D7?UZWjRy9MnoY@G7lplRdJLBWLFvC2k#sc zYWvPh6D1@H&pT!!qT4pNF1B6o4s-?L-e48(koGfaa)GzRA{nAXcr z*M9AO|3oD~#QyXR5uPQ2a5d{i$2?kBBxGB)){(G_v(_=A9AKp7Fa$c1X8aU;#yl-X zJ@@Y%&b8QRCOEhe#d&5;$)sD^?uEwcAI#K;(>}=PzFdR4G5Or(ZM{c=-QM$|BRPBA zZZh~-SusL0xj%01iMVQ;fO;}tbYr;v{JaU0?l5X$bt0J*%Y5g;XUmar**fr{+1Du< z2*o&*kla$ZrtDcy_x&l;YjM8YW0))pXCn8P4p&A%Gtm`V1>0xbD$AZt`Wr+|bz$p6 z{knE(NH9ko3($Pn(Zx~}kmM|6K-SJ|D1ORqcTIdd$Ll=X6DGqXm89{l!0X-hs zu5z_pq6^!%Wzu`~&V<++X$mhn@60_)j>k~?h(auvgHT^lc+k)IJFN~u=TE7F&Hur} z*X<~}$@?GYT?~~q7iQ`1@|w6mUzw)Y)FV2~Z%)-e*sbCvb2j8};i?2-fHtqCr{t(z zC7%Y4#c5U}JZigIJd8;9C@okjt#BS= zP{&m8q6TpLAz-`=ASuZ;s-Hw*HSa%p_VixIv5{F?nh<${s>-C-xB(5axH%CXOl0Xm zc?_9! z!R_SMcv7tRAllxHemxS%13YfpHrWo`1hMVOH}yPh35-WXY*4)ddm0 z{VSR}J470nH)z@-*L6lgoFvki1p z)XUNmf-qAdh9YD-ls_;6D^$eJO1}IR!0oq@Mx=`W z|G)76KYjVX{nh{J3p}^4c;xd0@M%o@d$ZuL%H4UtHSPNH^2^wVX}kKtQQAK^qh%l+ zlcQqnkd!3p>bX0)-y*Nbb;aEYW^J>lWLy_v^>*jiUrz7UI_{nd^ToT|vhfusLO5PM zwye{qWON>OP6c-~In1R1KDZ$cn<<^)kd>UhrwlvHzU)eEM`nK~#9T#tZNxvh1Nk}H3 z!?f3eSvCtLyB7Um8&5UY0?HyA8(_RI<@aDkt~dA<*<{dJX_Tnbd||_(v`1Tm+n)&= z|Bc_UtfHfFedIvv^t%hE+Rw8|r$+-{6gJLF<1AgVgz%{#Wma2cz*gedfEQfW4` zC=97>;@;LA6wV@0I@_4ZXYu&(Y^&z@ddVuEP$)LMQ4lsVFzh zgd%<|9q(;Ta-;-Mmo7{awle1hKmQlP!@3~*>VfK-I+6Hd5828s5 z@t#^}rd#P>NvY=}gjE~K0IcfEKXCBZ$4iQ?-mS6jLbFEB=)qpuoTck)0@=%c_%LYN zy(aGJh|MV2p5z1@UJO6E%}sa6WN1w*LPH?l0kfcN6#0xIKVmCAx$W{jV3E{!m)DsB zbF{|`^x&EnT}_%17C$7+j7KjuWdnP&eonU>7#=x5~FaMtgE7TzG_YKU< zVJ=!mVSnr{uXGD6xjRoGSy)w>_!Fj{zaU9qfId4DLO&-7dDnb*OQ|70UH_nCX5!j10S&lB z#fYD{XbrQ4lX1&O{qe8zhPizdfplfg&-*rumEdc@x-3_z1&M%Pw(Hv7nd?E4=R z5HD0$O=bR$kx?f7#&ZPGPeb)>6N%$odt@~jzfPU3RcCkQz+Z>&yf zLpAFB_mEA+@>>tpcQoBF=QM{}{fN)h%zE)7)N;|7L4c+aS7lH23=;F*9Q3yjw`cts zKNYeH-k54+m%FO^SA%k|KI|6rN%I1`Hl)=Gg z^DNdfk3!EyOHK`WQ`By6fO6q9d`|K$v%fAZI9rltH9gEWzT#_6Ck*GdfIHsw48&iF zbBsv|7s)6@kuasw{Y7N6b0)|t0Uq%6JQO{Y&7N^5`Uu^NsEL9bc>N0j)CE*J!KteE zAW7u84y1a!yh5U@bckl5A+FL>x^CZ8)t?7W$$X*qoLoXe9dFC+hK!U-PH+~dfub11 zRAVMTjjOBorv!KOj{>%W>{(5n*#h_zwYjH+1zY-EA?H1}j%a&{RNj&KZCla4Kh)k{ zF#Db(mX}WgxrknrvwQ^rm@R}yv=_|oBNIkjawQ%%_nT|8^fFiMHR@qym>Uj7lAx2i4c++k*=-2AY0?d4q- zd{&bUGo@snZ34i6+%??&Of9P%%=h)XV0u4*v_2O`Y6B4jytc>}sxIlYqZ+%ds8iI(|@u^UJ=lKr{T zHx_eF0apfS&v)0f^jg9}E+kAB9#S}~f=9P&%28WL0;y#EgI_+>AZC-_g5E#oXj3!% zURSPL=)Y8rwij+ZePg!9*;S2Vw~6z|^DsbO(J0BI{oMiXy)Lr?Wl^F;y$Wt}HK;cZ zmN0_oT{2*ttp1~+Q{u}jZn3?GOm<#Ay|bykyX_4VJC8R3c-aKDj}2DVO?Dh46H<+* z6Kc|=@c=l&k*U3{%5_k}N^izc!>oSr)ZWNwUtmHpvKV#n^+kW_x!I144~;we>LUqR zs!pOG?LS)`t>RNzU`l624oFX12~`1WH$7mv)Ll-26(^mJY>zl}1qMOYC_EVxT=bl0 zTkvj;Le;WJ`0VYMchPKbYH+`Ut%b6#fAYiL)&P#rIK0DRBh0PV7ibCdeGI@y-&CK< zDl-3J(%PabNWv#UjxZu%xbG9<-C${aW1gB=Cf>-}ARl86|1onbYc}cg_LpDymJ>DE zp^+54^xTkO=DOiSX6QErVAe@v$N_OV!owiwCBR0ax=K$cfM}64WdbvAKys+tUJ{YF zOL~zAXyWX>9mf!-8gg2gPpCA3aA9i$+K!lI3}uVm2D*^7CJxbam;mt=GV}hzKe);% znq9?Q^QZ?T#KK02UgX5xx$?iA zO@Lm0hrAtLXtTSGIvjULoYr&6_5xHTYZtA|TxLzy2z4(=o`Cq9MXlRI=DGKz02Z?3 zPA5O(h*`6JZc*XqZ0W(%{qirS^n58g9ybc+Mk6EjHY-Vvo}XM6`^nvF^z#=u{pvM7 za?|wQ+VIf8nMQRFd@?lg%a~McZXO=5{J00CgslwIX2qLor}K#_m_wS-lk5Yr1Sew= zaJ6TDEArrBP|vUcG@EI2|5l30ou)u+`>7#tbZ1iol_E0(**f#uBzvDXCKCEh?|?SB z*zR!x0=zuuCe|^e?1BzXd$6Tpnx54IW)^onZqBayr{S%7!sMK8W9V1;-8`9@l@wqQ zs2-3oIbqhUb(8@GMSQtTZ;Cc5_(d6+bU{+z5qty{EVY?JJm27F|3kEqt@!kWB3#kh z&{|@)dE|wvYE<|H#9Q@l(gKSjGp&LCDnxCY&!YBT(G}Y!&IHR>=Bmit{DE%vV_}k< z@Z?pD4J|2>l9*JriaB|Hqj;!nK4Rx2z&LJ)>t!mWr@3GfR7PSFG3cI;X zZ-Bw+{n-rgwgRS;h$O~phfk>s@b7}uS(@5HgU^G{H=|C_QkjE-HX9f zHsIg;*=JrA1N8bNSy~{l5PFr~y!TrUM~a&7+bXXfAxCbo7OGw4D2saowoP`j8uNV@ zW$RPg_Qw16%VkB2$x!k8V;sP?e_MZ~L%CTy0Xl*lNZbBI>iHR(XA|}WF7br#i6wq< zf8{fg>&alTgy=c-Pc=aBSV#>~ce@#NezOTAtyBDrpH152XiQn+!J5?gOfQI6qYV0? zcS;mgj%!f>2Nin$?%%IrMvs2b>CqdDle&l0Pn~d>zTZPGHgw1VXdzF5Zdym{iZ+&( z2Pa}s6KA>sKDkyODMxkhb<9c{?;9At_EruMH-F)L$&D!<0qsw!iLF8Tsa^o$k=hbX zGF+a{OO3&k2Weg>92(HCL8WbI zUk%71@O#Gs`Oe}sapsIP2UcW`^C`A5R&a}+Nmf7tJwWvqviterTD;#Y)#lUoG( zKV1q9*wg9p8xz6hs%{mDeqF={5&)@E!`#f}SI%PF`Jh)3^O-S1Cv9^g_yy6v$jYpy z<~ZWI>C_P1@M5RBTKbzPKn+-Y=i0Fk`ZFGdeytCGp?iiY=QLuSK=Sma#P3-4<=gCRkHmgPuU~iRt{gmoO*u z?MH>m&nrg)MUfZo^hDC;`xbKF|Lbg>wRm7CawVx(fIG}Ah-SOI)NWh!5E<_eT@CTn zGt`uoyokskr>`kvc+j;0-g{n13W({tKBV2UzK@=Cw&>gf z%a&pF-t5Qs$pQ}9Nz;}68}Uv-l|ke{R?8qfnd`a^LQhysxy6Dv-FBj6+5>Y0kKur< z&l&j8pu2{7^g|D(c+0*7n1L*akAf_e0#(Y+rhDN*stQ== zM7l>?_SyHLKZsYw9 zsD01VC~D!5ILt#msm_VuBVjw;O{!6OEo#{^3e%Kx6lEL6;Skncdj&Ve$KBX*SKXZ?AZkm7LF&$~ zeN{dW-5Qd!ul73uaoUnkxz;Q);=e^yn5bVD1NQ^MH!)JzP>mxcmEhCy1Ny(fY^uqZ zDCl&u(-T90z@J!}>W>`aZzCh{iccS*zpp?G=zfMe;KxG^B$J)+hDUEYY>B+;f}K}$I;P1d)iEouk3y4P(FK_UM$G6tvs>G`I= zQ%W@6{T2}~S4Tq$Gtp;L?~bjQg%A&UEHN?yP)}5s2Q-aKL_=}CJistn)*eH^PXUWT zN(eIW9D`B@ZP3Gb7k1U}DY~!szOvjM$gcWx?y)i=UJK@ongCdwgJstp`+BCw)ufWe$!y&AY$!gxba)O#blWc{logm&}W-U9#0dZsL5%-}6jZ#F*Z6y>na9@W-c89vh1FiXUyGJZ(b#JCxhVs=m9tG)Q10 zNWEm4FW3wudv=@lK311{^O$R%WAN1Opw-z_cCJnAChVU~y`q?Buc^^R;mk@7@KWnN zHt=%nSY(u<$K9TDrSXWdM9+x=IenvpqMgWeY2fq1H^FowDNj*O7FyYQ$-0O!_^C2t zsN2oeVO$@L)ta5=e*NELiJph5v}1;3OnP*=X%WDcbq8GZuN)E=V;TjW!()I7(`;0~ zENdzLXwnhFnhq4qD=__<)7&{HW-E||=ujM}X5%-(2?B-cK*K2UgNQMCRDBqAVDeN# z?5S!}_8@wa=9$zbXYm+UG00;Ne5M*WGkeri48e1rEmc&~RuNs5b6rY|4d5V$A>ubNwj}Mk` zKN|a1DY#}O_gbxRuHFMj*fOQzUFf{|K)*Jgv!F-F_`RUf3<+%K7S71-6z}$IPBf{K zau9R}yehM6n7(+oQ7~PgA0%7P>X67ZMGs1-eo`U(v%*oemj2YPHFo#ak^qWwYCzXS zqK{7DPyA5X7i*5^Ecj?yy1-oLjvSg?IT~$}!a-@N(`%#ukNk1yYb&yNNk4%o59mMY zJ;Ll>HvRN|e}(CT1y{exh-j#31V?h~GURX1Jr-B9XY;Sz=2i@8+o4uHpw1vs$R8qV zFZplYTW&&pI3vxKvYVny4Irn#0Sz8%$x$*@=vtNn+JFB2-CU6Rb>PIFkqcco3<22H zc}USw20OQCbpweJYB<#lvL9HU;T04}fh{>HmaF%yf2yi)NG3nNF4rq($>2awwT3!) zTR?Alm{)q+%iQC^U2G#&qm)#=VoYO0qinsfBYY1qixh59pXKwX@Q+T*?-SwZiKF&_=T z(ESxGPjG$>m-_vLX#X5i8p=L3nTxl=l*afBf^5J0e%iS^x!YDj^9M~GQAjLNdBO$T zwYM)gKV>nMt%n!&C>uvsq<%6@3Gm9EJSe-hAPQd|uOYB!#5#)qaPzUx82;`2=pNQ& zHXR^;%4RS(VeMNi3_utTN&_tSdCT6E<8Yk&{ovbp%Heek`aq{(E=B|JBT+WAWMsoWIgRSR6_&3LK?lnh{`H>c{04 zwWUFt`qxrUH@w_5MO;YhOF5dIllNnlpt!`kCe=}z_8UE0R2M_Fo~2e>J7I2)W*2NX z*LoN%w;ICIXj=R6B1kyiZqmPV-|s9euClnfTKyby_Lf<1CR*<2Gb?P`);TII!f-4C zlXTVL|K*7p<<}!~DZ7csR7htlIhZHqQ8IM4#Rniij*Izlz_I6yiNe&KfwUS^LRWJ> zlLv4iZ9v4_qj1Q{P<|=j<}gH@&ur#R z^Y0}Z>P`1Vf=pidZKZ>kG_4nn`-0wyvMf<<{<|gSHnQFJy)Ppi^D2+2B6{Vv@L7#Y zT=NCyY%VGtx3bY;;}oB>YguJ}{uHXz(N&6+KBy8sU-xuXwdzV87kNEeKGZKFjB z+Enf{qOJT%Dkoq>LA;$4D)n&F2b4ie7IQ{XrV}?5>p+0G-pp2*1`v8Ix=Qkcwi-$r zNkicK}?JhKjXta6es8v5U3n|Sr`-0{kU`}pw& zCO@?*!EacQ_Hga-lNqg|8=APKFRZRkOE1gKfES;Aal_`6FRy;iimuxV%=MhvV!q%}*I{V!*lPfnEkftCY&ag~!C*PPG!)97~XpL3-GReu;P%=*67Q>RHi zzbh#K zE@147q36oeH`OQ(i!=!4mQz{Om6t8e_Ig+42xlWL zecjHjlpUEIh{@N8`MRJsu184|q)W^D z#g4W^w4t*;zxS@4m2sR`Xm$rrm*Bvh(;oi`iX5d6#ps zwKB?%&d;rwN3*-js?8&In=x%!*hDJpb28y#(mad|b5-|U0|ui_artfcn{D#2K?|ym z&c7auUznWKj@>rL5NlO2cFX?E5_in(06CwsF2emg8b={tJfgpyl55gQCR#d@gkeB~ zH<1t;_{h7Lwog&Hb^Fb6ZJEMh+TlZaH`}j$UUm%6c@3z#@0<>FeVWTRUAaVXag+tA zt(tCTxFUq!0G!9+c%u)8`5@%jxlz=Q`+aH<5+KoRJ6tgGJ?zPZg8P^y>_<5LMB0l{1Wn3!mCU~@P5{#Uq9kImMO2Jkek4(T+qK7DA)n$Sz)gfR3)QnN#J#3W46~~v z6buv49hq`+j#rr34Hd66gY0C{$Mrw|Xz*}@OyO1>0g%1^@tZ%z{=iXZd|c~u^{Fka zRO8}&JnI%oyK*Bd`KRf453qE3$r($O9Wz$F=EpZ(#f~M{fJ@)?lwyTMYO$JuSrdOB2mWj^LEqs^su=CDxi{^?8w~O`$&6pFW^6B2y9=2Qgm!T=gjwls zT;RFUAAu%>(&|HPdF-NrVgsHzuuqRGO+UQ0ZGXF3U(P6WQq#C&aSR1RRzLZtn;C(= zhjHA+FZ{Vch{rj7mAr5|E$2-pBodkvzSF#YX6ElMS7)Cnw}}=bn}0x~@<}B|lsTCX zGn-Z;pcysPeswCQ_K))Cw0j+N-e0jW)gLCR!us;4%Cc(e3ZwCTQw*L7ahOnW6$X; z2t@;b_%tvL>tErHN+u~6*tl1*eTTEKGzVB5^ik)pu3Dps={~jFkTd1W9PU3;Kh;L+w$b)Cx7L%6KLfz6v{)}$2*5z_vBBtuR-+o%|`sMj>vTnbXl=_>qvi4?}0o1+g zU*5Ym4kA||`+~?Y|2ob27OA?fC9{s$unn1z6SnrQPxkNByWZh|Pv9_N%k!#J=}|S9 zV#i1!Z@AwjrBT;^*%w=IviRPiu&zSME&f-I$=l)}@_O>ZJ2^0}IDg$?pfD00l~DnG zn})h07a8Mw+ZM;?Mxok&=>G}96uWjT_#O|MUaI_3-IVJ;Y{NWo7@6yu zSn${`Yi+-1N33;e=S5;@ir?i=P0Dt^{!Mqsu7RtzBCxtNur!aKs7~VNy)0OWx zCORCSd_dheQH9g*EDlgSLAuk>20g;NFZKhYSAR1I`UpSNXK~WX5hj}DKx;2WADqG! zcc}*Gm7(?S0g7H+`7w5k7w)q+ylS>I$MKaKV_iCP_e@q{Z$pC#@zeW^g$tYYIMZ!< zk&eMbA3q12+`fp8awY!wWg+l_XVI_D;t;#c+?KCOU=#R-iSg$lQh4cKy)tbaZWEpk zc3qudcdCx>bmHS7z$rxVhxLp5ALnSZeL?6`0XBIahT|)%(qF6GfAAGb6LNLjp|;ft zl675grG8;|C#U|-$va`4KL=R$9mYQM(3!Mj(#u%5wIQS-58|MS21QLD~K zmFLTgLft~ER;(9uo8}BY&r&*S)WAD#gMo|(25+YATyqat7O{Um2qH}4hvl7x{q7lg zpt+tAvzJSAv(6nxIc-%V*>Mh=VlAkI9WMFyF(! zSKEfUG3(P~o%xHDvuW@y`QxEGW+g~|D(zC0U-NVwgMEjL28B1C728dQmJX(RU>i+v zSkRD>`1mPy#9gg^QKi)`C^m35y%pPUeCvd5y!lKBcKKNB$NwVkEyJSxp15%dL8MVy zX$k4hWdUg=1qn%MP(Zqu1|^p6kdlTkr8EN4v2;mycP%V?Klktd{{Osvu4}J-adywS z&pl_(%sDfknek2ZJoSbcVgd4+-yKY#GIW|EJcw^=3I|lLmJ{^1e;-hZZDr}vj63-9f-te}= z+Naq5ZXtUjhcE8Z6Aa_RhSp*!Vx@}#J}ThF`fJ@Q*zq~moeD}K{2c2k>(T7jtP_Iv z^~$K4@6}XEoB@{1r{tXS?~?}EjFJ)&S^qyD6fCtVEoPEQd>CujNRRO%GA4MuNxGTa zw^*U5p7N(~$hRjhKU`W|rg)1LzE7YlG~xt}rCmkTqb$iJW?NcF+O{lq1}E&|xfn|p zf6?e990>7I(y9+H+r zchud-`&vE9s4a4G-687Y89#^+PrN7dO|?F$Pt)Cz_*kq^%fgG{I{ssdN>x_cRX+YO zn_`mlNbR=ndTzaVlF7hrXW_@$vTEI776pes=Ljap*WjB?X@Gos%Y=6k`Piq6a3W38W8wgbTOLCvzSD^NW5(dHsHp}|ixN<5d^GP0) zyT;pM%(`eUIPGplOn$l>X!|z!Wv*#pUV>X&Kgq3XMVv%ci{NRYYj&}{ANUp~!wMG} z@Ih{eVuf5}=vJ+#Vc{yj<^tjgSoKd#)ed@915VzX(zYlq%BabJT^>0J#8pl*6?xLQ zscJ$JOZHxRF#QF1P5`FV)RvGeLAuTiert|UuHDO7s0ONePr!|kv{Zd`&) z^#uA7XSyr4Kw@@*h$nFZy<1)PJ?>kiQ`KU4qFqB61wo=8T`#R!y1C2Br9u8U!wE-G z8=Xg|-5s20HGl;J(5cma+DLEIL~dUnMNhJk8)$hrQTG_gnpdmW<~%BLt!$H^beEA9 zB;#?HHm7s3N4<9ntQ(nuhkVAvot1tI#rRh;Dr`Y3Rho(o7(pug4pf`?Ni|eCkEs8& zoa4a_n_T`GP0!y?sTa_()|?hYhc(#Ae}5QOfT&@+p7rxd{dk@_sX1V1ei@8{W~=vz zkHoLlfO(U=_$dX&2$5NfZ|vZMF}C@=P{kkYqRnYh83ubgcLV;Eiw4Na0}zPlVxb)D z*VRXzV8-+E!pGKVaEnMtnEgy{nq4&&!F5|KknVei__4TK-BH$;o70gt$W}igML=^? zjraW>_26%cQIoC*#-G0@!7%skNJk5=>>rY|=C`_aI*xqr*l*;63r4Zy9(AvbToTQA(ZHxG5gV(L1l&lxNCcv>4gB z+)9ysQ1+qu8{5?c8}eySJaXa1!PJ=9>y;vY^Fx})C_#2&1#`3N4p{BhHJAH6E_6~(c--d!mo>d zgIu1Ky{ca*G3&z}2nHlPYLz&_p`bunXv$VRPDXS~5`A91+?0;&(#%A&s^0tl7H-jU zubN$r=SSmd-Ln1B#@BeN{s!WGFc1;cGRQ`dhu!)~D8)#iK3CERrKGk{$~nh~Z1X|h3KFTVk-+t@{fZEYXH z?^|aYiN9v}>QSlJQH#?XLfELfn!P$4f3)SuH{D_*N2fe7F>a5=&YJ<;;+}aACzQo? zh>HcSNCEdrzN;Gce4aQ7Qi{M%=IAqVab6TnIC9@8af^!eX1!iHUSGTn=swY>VhG&_ zZ09b5#Rx~x-W4(17+~(c7>7XEXsD((*9)>BSi49G)ePpT#!I!~$y=MZVs=}-dX0+w zl$;4o7`C{HzMKUwTMgD;#UTpDM_t^iJe_qt$g&T#R|9qjAX*>6?R`=gfgg`|?7(sd zh1!jU4fu!Tp#lS8BDz65lVLC_)LM+3G=gF|)N!4*Ig27O@rC|T`?>e6somD=aluFA zWv(yf)c*@{mxfzk{s1P;Ash5PMp>FhReL@!9_T|9HzEzc4M6CH9WZ3j&CZb(dc4n{ zx|)1Gn!3SoJv)j)6&3;)YEc|!#CF?ES>dlj`yB}^ zS5jYB{vIQvKaG!_rp{w!F1))wujWjfjQ=3@ds4l+|K883YJ zLPZ=``f0P61_>YN?#6+$BcYe_r5^Cf{4TBI!WgTxMN!kRW%c)BgyalLbro3@^ze|$ zPiuT)pyD-5;>_L2^x4!zO5xTR&nEcgwA(~x+_@Q6th0Q56Z4MCYpBq)*!Ibf^YoL; zs}EJ@vzplsXgV>3`opW?o_kcRLcfYbyF*?0U-)x|=h~0dqi8ToM%{UxiW$+&Pk~l7 zozddSPpQ(Zh1^&!$)cVku#wYL5C!`(E8OK`?~2|m^EL;didGBPjoG(COQESbJ>sJi z@6U}HUTF<0fHz%j?Fahhtl#U*RV{vt5)G97KHrFz7mmV@bD1Ls`i=ol(7AgzncM=GT*@+U&t@Eqzka6 zzy0MltPp@>Ct{DWy`Hd*GKklhTifAEqOBAm-fDaPs(FxYYHE~k4|vCZWlbPH22cmQ zoV6`x%Mo{pqh0vKP%s44SP zQeV4;W6Ey{9x%UlDDL6xLq>5F_ZUZiNt#@jNI&kjJ3FB9^XjxayJ)TJ51m|}pb;iG z7vsihN-j?>qCV1-Hh(A;k@%TZu6y+B-Efbd>ej?VF8zMokX=SDep(c5M$%w~5u{e- z1#PDhid_Sn8_V@_l2;DsO-qCC=ZS&9zy>&oEYln09Y=srVy&_#3(=K+_nfarl?qY7zA zwo3>X*w|qX+k7^&IzoL>%`8~%eFLlV$&!4Ed1PB`ccu4`u~_{`X6l01IOPFjC)Tt# z@f%VvRhNJy*~NV=lVU-H;-Z>7t)di}8c>mM1o5@7bHKOT+P?wUHMg6taN9KO5`8ja zEe0Eo`cq-oHX}d$zrefn_`1wy}y(*35sBc^sbi^G{ z+BU^)FngnR44-`k5~|a;?nBk_786K&4Wl5**ioAqg%)kcLtdv{>tHxA;1 z#kIsH<#i!$^*26ix2iNlJa$+ufz^fl$@N6GSZ6p198HP3W^^ zQNXCIiZPrGU(n;f!`$li=SH6;3t0y@iQTFa>=nicRG|^4WA^{wc;J$Yq1N9CtkvIo z8(tBpPwH=`!P*wCU*RWz4nxisi+~JQd}M+gZuJD0oqMe@3OTFCdoDa^=c6jtlqH$* z{FPSKuSKCl8XDOJJ1q-ZSXxw0Va(27oLS!O^{cxozwKiIgX^fKM&7Pu1j^J8X5XcL|{+I#2h7y+_C`#Uoz)b5^9dWy3%Ye-S-%)f$Xhd^R92 zebMFDwLErflO3bIx>(~KnH?MtS}r}Y_s|;2RH`!4t}WAY=GbTs-KH$r3BW+gmYjL_ z!%6{A2R{k%+wP`CO53BaY67#OA94}aao_y51*Xesu($UAQwzDhP?mPR%KLqLaJvr( z`AjfqRrSKoh^rYYjVh7pG16YoA98K?ep?;_pDWdfrhx&3LQ5Ak205+$ko!0WiR%PM zBPLR!RmOn5#_UdtJ1ESFCu81D!cqWO-38vYPtC|_O#pk-Yn1K8cTh>2#zc{QrN%ob z?I)^wj-6QQ$;enx?=dYi4A}5|Av*k`sXd}nzZmi-^8beIazN2VZQXCVo&9W62vN1kbvugpmQ8me6v(_>v^1l}vS zw%>W@XoM#mMaeepGuvH3_&vj18XlT!N+oy$Nm!B=+<1~PRzN;YMfx0Hm2g+IQ)&=N2+2j zVw|Xp60T=rso93TCuuInxBFK3y1VXG`h(*c0vp_Rsc7QI@7abs27lofV{19;9K+|e}!6!#A)PhUPR_$} zifCja@C}6d91ed6o((E1uH{n5{kZ{O_-)pZ#1yqtL9+j<(RD?33Sl(K_9rKmja}^R zPg*$QX|Ss$MCQ03yPl_@3^uS+<4Qo%;5z@5RJ8<+ICEA7n`XlD4cPUPCoh|84On7c zbv0Qi;P!HBRuQ#Qy#cv-LxE*bi9?V0!}$_0046u^=(I4vKZuvTuYOra8 zV>v5|bDIh4FNSA`75b`BAA%1QTV8AtfXJkaQM5t>QYs#eSeo2uO)mcf-9kJOpW86< z5Ts1#4Ip5Pfu&FumDr@1p&Qa}=(dFX%@We{vxamg?8(K{U^Y~M-dv)%`1Jq%L!3Ez z%j;4`pEMb%k<#Y#w{QY*o#R|{8*R%bqw46WyM}8_+Dv) zM!3+_H159(^RYyAjN1JToav^!y%CRTB=xI?wizFu>Av2)h;w^^xv|2kKQw()zq2e-Ewqp{6NWbCQZZ z=4*#vHKbi>C9=Y}4!3N+&4mUzjpGZ}Mpyii%39#Nyr62X&ud2fZt8+^BuC09X_)R0 zG)P9W%y&8Q35Rb6ghZ7!)^j9hDl_ujpX+Z?sy18*7m&ZFqkbN9mG4S}{5psz_ou)5 zMTio-SZ2U4LOnrYp$7y;a==;_tb^|mbk2XkG-x^wAG-_ZPyH+Ul2f{%h`_W0P~?1D z#uB=FDIS9y@H5VdvL@c;?$%Q$HO0E>f5dszQy1c6`GsjVB~WqjM!Cbm5K`MMSsY(2 z3+&8lss=Zt%{J|c1 zb0V&7aLQ8Ts^jl_2BC@ z#T*Q~%sD*r#YVaQW%I374*rj=`f<^V(oXsr%s8xTO9TH8$Z>%5ngxq-F9$bAVnl`= zPZRSYSB2(D5XAXQC++h7Jl9-eWhCZ%gdmCPSDNQIadqTP^fXVr=dd__gGrgbIbA73 zd!e#!uQf1>e(H0ZC)|&7X8T}sB47ByMZP9~&+5IW9bevcBW;I&$iVIQx>wY+RE>Y` zlWH!v7$}eHpY~7n+ao#En-%phBnI0;-%l)*@?Sra%T5@O5HNl8?WAE^yNcHdBDO5- ztDS3;lcjj(-+UF(nYn-@(xkMi=?sF_2lY>1I~(r41~TJ){gT)e=EJfeYSAZ9Q-J+lKx{9Zq-X< z|0MATxZ6DO^&=9Y&YRst83E({`9Upw8iXi?(7z zTe~-diC4{GS>`C(s9eD=w=1UVroCKe4hJ$D6-$+Zy+kp;mae`{q7w=CZVAz` zNTjMjhNx7KCI?}-D+hZTr&z&0g$B8o`=rmDN%rHnjoO?r6}8*pJ_qSMhe7}RZ!sDk z`L1?h7!~-@UnF3CL?|Xt_KQ8WS@!?!^Y`cCNhclEOVs10sk*pX+p7v$l}B>@51_3( zt5ifw<5lO4ZKH|tE^wKS5bhnO41{gF9tDBXCftYV_t5gkPx64eheVZje|o_y*iV&U z+&d^?F+YhMuyb#Y3QRh=Dkcck0gCNgWuF2BhgN5+^q%ThFAjkGlY)~hdt%>&+?-2T z*}Fs4DgOH`d$><17jF>cv|RS64pDl2D{ZZbuugox^yDxoD&oJeAP7>A)P(@XzQU~K zz<$N0_3zNNERW7WSE|iNMBsn@PZ}uEDudF+3pE%Q)dh1+v1Vx9CZJTuMfx2gxA;I@7&M1 z*Va{j(s+IK#E)JBrK}qf0zpc`7?OQ|jj6NVLCYnS#;6)22-?=J+rT7z&x&%mb>P7GCL6+z*Ls;S z?-D1Hw4lWBE541(qKSdU>xAkiVvzAe9#}{ z+-<*vtuLS=V5f#re{#4;WVKG)cqX&PiSD2$E7hsuHx)9EO^lY{!4U6A${!i9FAeXx zK~D=sy!g!_4pdU_7#B-^Gly_R>aQFn%T0yi0vJGR1876Y6n)UTY36WkxSxK+i}m|7 zPVWa=`L`c|owkVK(pVrR0M#1lKK!Q4<|QQdGk5fhbZjPxcvpv=$c=t5hv%gJJMTUU zT;FqA+UblSw8k^ZqX(-lxl0)4i*m}gt?rxVYRKEG{vC`4>+m=`w1jxzHTu_R#L1PU zZ{csOm4gNd)vpp{YO?I;LXHx*g==En54)UqL~&>hiK|0TP59?cw&(BE6d*3lSIX8m z^+sB1-Cnc5p(*(L^AQ*R(Zph&(JLQSeWU~PGmc)T7(i-$Wht8IM=x+^^ka{nFr2mh ziu4H$iW`JvzSzltLH_YXq2agEGy;s@TugkGT%efLnz9g=Pl^8^UqE~neLMUqpcCKx z+3hp6jjk5l@8#Q3-&6mrskkwq$;)LAtj!`FaMK>M&Iiy?TNit0^ECL+_nX$Vs4Y%P z+oWmMURd=)dk%T7Eosdm;PN5OjzB`L!;6fvH70=c3+MJ(QI?zOL0hG!h@{B8=bJhm z(pr)K25$^MfK&W9FYa&+CY$jLV@e;tY3?Lgd7w|em)Hy!%JgBlx%%$3asV^A%%l63 zpl`dH`#Xg%Dcr{r@tMATwl-Kt{R)#2_vb|$#Z4T27Y8Wp?^B*%b>H|n;@s48_%RpJ zdCwpDwrQsQqeCTzom@Qxhumek>;xQ=zj(aT2cD=`qtS?rSOVWMz+lpcB0lxRLQ^dL zb#b$skkzGsEA3)=va%=4zsR2y!9X0WbE}hZl`WjKkiR-E^yYIu?zwIum*(nwIk#j54kyrJ|W9_(knc_~-{`cT4KmEku zZw6JF@eUh>0~+m?BE!T~5w0#XO?w&PmYue8eZ`#utJ6bXO}a$B;o~8QERj4J>+NvcT=-{p}VWf8SGJ6#oFc&Qk8%GKBvKlD~a8Ab`QF98jHJA;12oEBQ|xbKi}xsv~3&a!WPBov_bekO}_BCIJL=KS6;vEScP*O=LcvTB3HtoV-&;3H)}MyEswRR4(}8xOAW+2=MO@s zwWlL4NRM6pTEyrHUs~pcd>m_Ow|ZNQ(vWdAoprHps?4{&a9PzLx~B zHs3G&kHt?*+)fO+8BH59f1d-GsD$U9H4LOQ4$U5u8$2L;n+yW+b5*+5?y&rkQ*O;g zg{8&ZQlHu4FNK1%#OpLgD8ySl+hd0v<`C`wl8#qI-K+~WZju{n(3_#TQ$Rx-({_iIeL};Yk*GM0X{bu@#P_mG*0m=C1-roo^SQZts4Io)j*luCo3$2=eXb{wXcL$DYFNzL7J)}f+H)IYW*%i~|ouv(R zqN32_&eC@&ru?h9Lv;=oe=F@pIY-KmTmp>PAcm{Cy07+ha2%-e*a!<5&aY~9zID@Rbyl~&sqHv-LA*+g|!n+%<>%^igmv;30k6u)XkA(+Jf7?15#;QxzoJ-1-1R$$By6K;c&Msk) zfFjVMaF1u}U$1KTP+iB0IFD+cU(2N^jDdY+^tgBN-L#`Zf9IpKZ9-3L;rhfs(9YBx zAyE8kbbQJi4EfH4aDiM#0QF)8>2jVD2v!FDahF?1C*wj$L`izN_&E!JGr{{WSFUXSkaYYgI>ehBCe3QNsu<%IbcyD5MF~=^?OeA z;lE>J+Is$s7rVyvCOVn0w>`#wNl^iN#u{H1(}{V7Tb!im7nk^*%sdbb-!yoKkBf~{ z(UBQTzx8bjLcoEf(pI{IodfDaUKIMyZhk%)p2aWdT!VDgi`A?dtBMob~k)A*-> z5ZJLLEudsLpGdywNQf*whFl`T^`Tsl_T%y2&Ze9n$QAc~Hc zTzyE#di2fZ4ks5 z_-{uyJqG6R)T(zA92?sV{cZ{M*7KxyBh=2^GK(l3rMYx99F}hjc{=?IO`UESUpnlP5t<0`zyvm zcEv;0bE3DUX{th)(kk{IruuD&m@z>7cWUR%<4ah3pRg*@G}Jb-3o4woZM^)G+;iOC zyWWUJ+>9_M60H$x?dSh}{^5jxqzDk?LwS9f27)f)`;ZcR0>b%Jc%C(yRPN{yTLBT$ z)m+`?kX_EY$Z8CVawZJP-#p^vaH#Q*6=xIdVDav z&)^~u9PZJu{L!jz%+YJxl=>9myS`NOYX>U8{C@dm9h09cC6cymxHk{59JN81UYRFkRs9yL?dh%H{B$m6v@igBgPfG$3QEsdW=UN@GK;(AgkJYjmrhm zVTgyz0PX<7y)#6C<4zx?_nOxlJ$SPk^sNKEBS$z1h>E0jJ!^N01;Q;q9siP7FFk8! zs8&XmEF~WwufPBBdfNk3CFjmP#;Bj|C&P-p%Bwu>y>Y!qsN2y@3i?qcC2HoSN1kWz zB#z3Oroq-GH{ZntT?p`c_#=Y4$9OSz-P;}JlR;0W)6O8!jzHt+A52lEmr zANW83>PS>x7QI7cO`yh3!F5Cb8CheL8Y6rnJx0Ev%ee0TaTmAtV$bj1c#i;7BB_w| z4Lm16INQVB;H67oi`P*q-e#q-?kJplX2Ev$wKFB|C&iyx@Y_?L83oCBUJpYYI^V6Naj%+zG&BY!8KUJoTV-WADE)%^{4GIKBQ7-O@QK<#{Oit6+KUtlYA#m@6M=#2E8E3 zBrnN5Yjlh|>S-nF?!K7&vmdWl)}n_y-Z@iVyr{5kY>Y~U&$9=JHAcZsaqkEgdxYm| zk}iZP4IPhq<1$f%f;JygE#gAZ2wYL>Cmct;%o;6O)dda)ewH6RiP#`OGo2cdt?1L# z>Ng9>yJ4^Mc!Q}-p5H_9D*I|KcZkJohBIe|wcM_5!V~`~uqBKl<7*D#&)#+-&t9Tofv{z@D8{2D zt;rE%f@JyU&XhKbd$`A+sNbK$`R_-#!#MsSDd}(R(ZbCs&Xn}TBqy=OVJ%eZL|&e) zX!17&;$u~@9Tfl$G36uDL?Nl3)s#k z5V_MAE=4*p8cm(56fXQdtG~izp)g>3c)zn&6c(p<28%t9 zk)Y#Yw;DSPhl{@6JoI&3nTP}zAT@r&7|{_3vqWEY-KY3j{lhH=weJMr&6DMJeo$gN zu@KWM_v2d36yNhKJ!wu6C*}$7G*`ieg~XBlWKjYUCTemEar(KQ*04pw02T7d1Q*!R zyFqads8?;mXQ_`Cdr=2^y^6}xEw#ncu)oh)a=%31i~uU39?ZgFPKBq46FE1Xg&2cY z+NFq#t5K#_@$${Oj@%o5;Cuc3OqKK_U>GJGwMm@qcTHU&<4mb{@f37}HrI*{p$MAH9d_9g#Ap0JKKp$$J-B20$N}h-p3(-jJcc>9irpm-N+31 zjms)ejyi?><{A*#BF6EHQT&hbASaM^SHJ(DO}m3{$DjS`AC#maw=Qn{`aKFs!eb}HU4Tq>S3pKmhx8yBk>cw zaoX2z8wghlyl;`3E@qZcp2#e(SvNVMaTiay^;026-sWe)?^bieyJhXg-2W;V_zBH@ z^L)pKO|8SbkQ{gWk(*~kSME~ZbEx@^aed@?ZIxxw^LHlh1ByET)bSEXAScrXa-WhC zzUba8x%Yny-ObniFSyZC^1mIa>K_(!J~~4gK_pm?OSNcqvT|Lyu|!gn+lbj}djmn4 zzsp&zKt|GM*{NOGSDDd=o|r79@%|<*)?Lty+FS28=uzi>s6e@*M1&vCV|z<_WxU<~ z54&NZ4EcIg14fDi*D(b10dEFA_`XR0!Ifizp&sIBaB+^gkRFjxDU}}8C?_Wey3zP1 z@V?`(URtKF&17A=ScXU9HV7fpn;A(7ixSAM5Z z_o1Pph4gH#=vhAesb>u?dsJUj}0rA zl#>!F_b5?+*Wye?np1B2bJlPSXHNJQWUcZzqT~y@2A&UoYO^bfI`?gJ2-pZ45phqK zVP%hMstO?&G+>Xq*Q2ga(KE<+^SgpGVc?0@?~0WztDnz;bE5Jj;fzc)JmiOv3$0sns?%OXAMP;(g6A#Id6G8p|s(y zYD2qZx5YTjyZ?9wKjn0ZrRm6#+u<~@@2e_HnfAG9QCysDFCN8dQnZ9ju3}q#WvAcvdRk^R(lA>b9j_y&S>cu! z1PeE?1b17P?jEt{Zbaazj416s4U;+z$604hH3+@~gRhV3(*N z5mUTxM4eD(I?g}El)*S>( z@mzJ`q#a~3oH3n6#+kBLrwAJ;Qj~EfDTHiWy)h6f0w(u=|D164E91F?vh!#Cfr2oU zc`XY6*h3@@^?ZkG=*^D`_NY1QKMuG230u_JxCdnZ+OhRBZ^*&dTtfZD;M$E3eG&8 z?Ri&hNDT;a9m=;Gcg1%H!zz-JtmNW|QLcQ_3S?xJ7OFEu#S&;0kz1N+V9mj~O!VV3 zs&uB?zPX>TXe|xGLw_{jylVM4(-KCUuXyc{r#!~&8$y%Ear;llp`W7@dZ|0Ypc+S> zWW5WDtl!M+Yea@KY*q;=-&5SN72RqS!gVz9C;M{dQx6Q|98+ehwOWB^@N%mM#7%3`1lf2k zou1G$G~>s`u|ajRUNuX5^Z{$-?(keMB2->rhKccxbzGzI;)jFc>L8k+Y;Pu+;a}~V zaO5$@z+a95#p%@&HbQM~&^Z5Y7BEi=3f(PbeEvAF(W8IQ#JY_`p3wmsFuKTJQ2X`S zl@uA7QZln^Y|Q#NkPehRqK-Zky#$f->`|a8I;_Klzw%zTTh&)o_uIa?BaO?@`cEz} zf(p@hZ!1aoL>4G6VHXaq)weo*Eb~Z{g7fB;VJN)uZc*AH6{{Ryp|Z(2634t)<>4cH zRO{qr?U;^Aq{+%oQuIb0&3xh)nPr;pcN_cjt;&(C%`ZE1$7Pk3vtB3(bt&qG)!t5& zIHNz&0GHhv+PB4Aq*UG#jQnxEj~IxWoQ76zpJV+UUlYPf>Y=uJ71U6Z&qaiJVG?tA zNr&1`uHz2~6e>FAXb7(?Ae%>qEvu`L# z$_l)}_3B1N6_N7Ju%V=$u{p5tT!s8ERqu?+)Y@MxL)oqubt$+*UmX+#<dnA=Wi$8+-Z&ute!@tT0t@@-Hk+g^P7lQO zISoENmt?$aily+8RlMeu1lCqiZ`tHe{(CEzuSr3PPkXjTmT!e;#cc2@7jK-rF16zB zT_l>-cc2Urx*Ie9_Jcp0Jn^tet*Qa5yf1r{MUD_16XU`ggU>>5d(Q&c{?`V3JCP4Q zoNn-wG)EgsBk~$tf90q5-LL-lBX^2YW`;dYR>GdOWEAZ`^8+dMSCidP=J^ zPYlcdJzb09?O@kgGKrcSrN-ZbW!S$>B!dJB9UPD?*JuRqOj=^U&xKR?@KWAEP5GB) zr77DxmcA!~m0U&1BKg_4Jz_{l?V z1L9XA8u2$Fu7W0;2Uk6Ul)CGWr_I8vcg+1p*_4Lwpl$3`gS$>2d(}LKKV#cJN zF_vDaNEwJCoricVv|wFSbCcR_$^bhE1A!;57qDSTQQocPmbxb=Zz(?R7Hq!rMq|-} z>|o++jk*BStEuaI)SLGCA~tRQZK?W%XI%|U>|g?};FuE$(;3Ie+0CIVuvZH@}>p>0-yMYm7|FmRGf^7SQ%DkNno5AmQBt+(oH5=FNCML7UZsXvw? zXa5H)L^rtyjTa_-{yW92^bssvc)oXa(P*=51hLG{GC6yw#z(9C8uvsB#Axl(_lxYI+=)P@fFJuNGu3Hf30_>TKiRzsa_au zXI-4x1Qx_a#%m=GcQn)Kk4ElN*Bpwq!1zxfpXR^bF9)fxa!HoKKin#DNv6=>kMg;I z0K~(A7uS|Vz5)@D^zNKtbB*o)*j(olf$YNpf?M2Rz}@P>QsZ2>%xBY~^|4bx4IBL; zgzY0{S)Woxl1@~XnFd|vje-QKd-e%B9mQq(KC@iu$?}X<7j&S6Zl{$&fq5>mZ?+(& z{YsEuO8YQq6SAA*6e1Z2g0lFx-Spthg}d*AfebPD7zC>Sjgy}?hN8P*v}GIfIuP<- zqPYfW9H0}^tcmD(4+#!jm%p0h;}z$m1%Yg~e7EFUf1t&cNvDrGX**CWc|Mzy2d_?! z#k;XzdEpmgo5+%&a_K%Mt=RBOt?!do>klUDm(n3K+=4}GZM&6B*cDW0rKMS)2PO0Q z8O6Ao&`Jx$t~)$Oa}WeR>HMdL6KT=3i-k_Vt9b5D*V&;`<{OYny6)0mLOX}m#=rgk zFsMtujjM1D%U}K6q^!J;yb)*GBArRTRs4jsyU*U!DNqBiyN_13E5{9Qx!q6tCEC1^ z>9$zAgCu{qf)M_~OvB{CV3&OU3Hrw3@!o#Ct&><$zK_%cezgE^w~~O6Zy@n(q>$=k z-_`VReK|AY#jRJDEiFD-Y#CBI@NCcZj5j2$x{P=K6yb8G9w-|>`f13{QAUtp^G=w^ z^=~gdxE~=l06P;Fp1@n~`pRI}Nt;gJa)x*|p@EAom86(`1k53a*Ejr6df`RBCe$0g zuWk|DkIgBjnN16tcl^{!ajEF`_G`lG|I@kB7in7?AMvs5I<4^$ee~D*lUM!S3oS}T z0X``jKmPFF6$QW@311-OKRPv&ak=dNGR4)S| zV2>KWcAb=g<9q9yhO;-L_OvMaK$PnR${VQ3fUA395} zAtKThYv9W~2ITvYG;RpQ36`KvIX|hCyi}1 zv_FSs=#VwD(ARP7!)M5M$e)i0($hyj-<;CR58aZ#xoPzox;#{ASDZ^^gY7stHy1xa zQ}-78z7|t6)5r%dP%Y1M6mGJBbP+!mAHrlZYyd2)RZiT>cx#gqa)uLRH{2;z;36+yzWcU!~P=pV^z7A;@af6)9Sq^BnBtkR&R zqZT4w$^q?^WpUntU!Q}rkxJH~yHBIYU}r2Cu(!YL zWGu*&*?A=)Sz^oB&x6gf9po`D8Dd;5RxXCrGStGM@ z0$BS*kdmjueQgXXOe4)fSyEKNeLa0Jz@WOVMiUIAOipajx8vOI?BMy4dhE>;+UQbV zMNUs#ZN|!tR-9_Y81P*6pVAO5D$|tfI z27(I&cqoSC>%b{#Vwh&qQhh_iYnt`IYSn!9AXkRB?oEL(|J)V6Zxhtgy>{IW4o^;r z)hheruw&DzqeIr1i?kBCumNgC#K;kY7Y^FMQo<90MN$j7ryd<3`=75=Tj2K)$55q|AB|Ji)$kor2+ z0G2bpJqLqwM?O&3ll`F%hAjb2K1Z>au;qsLSx?ZgTNWlJwPnI3h|OG`@eS3Gv)7xFpc;(LaYtCt#uZRD?@0dKm=_ISyYddX1Fs6gERJ600x@;>=c+)& zP|fguyb-dK?)H!3nd&LQ74r?gxT(#EEdc-M zIdS)aPvR{kcg8~B2AxW zUYY97FX4K~O)iey59D;VzRCa5(0!~cf;jqUci7|eK3qw6z%q~u@A}V~>7mf0!yseV zDMvRGY7^HWoS$-6YQG1XVQ)yOq-p9Y$DV}G^7++MKQ0iYHSxT|)bNW06CZ~BEbzkh zUB%;BF1FqW{2q57C{Z(fU(-U1ei!{^DbmXNrVclf=#wZJ%Jmj^ImVEG*$g!)B)kS5 zy>4xi7C?LVFVG7wvP&>{x}Zw*BfEziw9de7lmDICnqh~52wMgwiVosAEN)!VBb;iC zhuk&q0WESd)f!wXrJe+={di^vFF~4ZEx7&}zYuzjhQP!D+Id>S*B!H9KxNYLEJsnL z&D_i1xGjI-G_nLEF8UKO-$LiI#rL(V_AQ)OR#sl}6qR)R|A(@#42$af!j%wd5a}L5 zKpK94QUimuN{h5eHzEx(fP^A23^5AQ4AKnUDN475bW4YH40G=J-w*fueV+L+ab};` zYpuQ3UhjGrKdFl{lch1&0}2VZOU9>q27QD$6Z6IvGT?OA`5!n>JmHDSD>f;wwC>=X z>hr_M05plV7wi*eNKfv_~>&&y~KYvvok53yvQWVh!>&H9v&HA7*;k=Onv{wOAf@Lsc%L zV+e2oXyPvR<*YPr#`8PCUh42S$s?Uw`wjCQq>;tX$3a{Z+$GrrVJmvNmLaMX)>JKx z0oTGJ`bj^J-s&z%N{>fP-`nsIz)F&RLrS!iD|<&n5!wn|dzuoHSeHdc2_s;0vf$_I z$^np1;`X{{WWC^blzQ0ux&-@Wfv*S0UYK!vVk+$)f;f9Tp&Z~HKQz_DbgNlU7a>$X zCNC~8Tk*Yl!Oq{?lK-17Vx3a;41{T$w<1$i+rrk`*lg0OIdO9wG& zr#Lt{sXZ3|%=J?gMQ;l})*uk<2hYe}ON-n0%i{0i{#p#C*5@@8n14=2Rop92SEzdD zPtoP(f7hDUbrC>Ysgl zKRg-o&H5iP0XbrrH0bqHJVu2a;n=#d0ej{;XE>Mlw&`Rvvj`iPo9^dnJLFv0xe z&i%c<+YQ;1A0~w7>x=oRf2~Bib~EXjt%!DJlBA38b@jXpLf5WK%ddJAg68&<`|MVI zV;N$eBk&?BLw|m+@5O2p(YcZ{%L+~g1~_>0{C&O!#8TwYiim^ARQZB(KY))e3ZW_W zl!**0S>*q6XNEAyzAMko>xTgLMDRsr2k|mTZZNreNv%{Tt1zIzTS`D2 z-xU#KtGPt;J?|4?lj;HrBg>or>9jeD zG(}tLgqQ_^}`n_JKGC0*tA5W~Ljk5X0_< ztV13lQx!&C^W$FrIC^W5>u4;$hZjwW{kXY0r&iuU1tmv}vsmtSAX-{vJ2FfEREB9Q zj4PenP6CZ&gk=QK(&Rr0-plGPYNM8Wec7JlQR?}T_lhQ}Gj!iFBr-avnS^FZN7JfxZL zU{7wojOQ1TpY4&d`aC-N@VP?8q!28)qL)sNKk-=xYgxN@JE(GRgX2iCmy)R5Q9y6U zge+16*z_R!*g10s-w_L6{CG7zvB~JJq+nLoz8={?!4dGz=(2!JZ7$XFp2o+jK!MX6d^o4wWv_POBXSI!spmp(2N^Torx!I% zx07Um53P%B=n(`%nK^q}&k}VaTTG_UJR9fEOiZjVk+MQ!_vE8%X!Uq8OWRCRo=#-I zMD@czNh8S{W$G(gPke+XKp#wet^jW)yEO|ryC#g%CvSKtIlf&NFeW7=P5qv|st*vmpO3c#wANvlnhjF&l zXb0&`p4bgjkT4pwmn>Gd+?>@ugTIif8fA=xx2>=UoN3ssnA`ss9NZ}KB%3`|H}t}G zuwA;vq&uJ81jR|hql>^^7G;3KP(*A>bk=*)#l)4e=I1yX(Unj@bT zIt1amE&|@Y`|RWGeXZ^u&M4+#>P^vB>9gCFVj)TYWf;F#-kwI%& znL2M%&(fE(@-QF8xvrsWZ|0+>%fMw(2W9C%sYuYZn}Y>H0m*|RuwrDJoI{(31ScMn zrg!lc_lBVSV<4KGqSf4< z5kWh27rVxGDQ-GZUE9+`?Ad`W>8f+|3J2a<8@qW)Kl37tkXIXiV>^+VKz1TZ2HJpb z7@Qn)k1$;>^P8W@W;0%HW}KusHp`bj#{Kg7=lABeFz+zev0)WJ<*om};HTgvNEeigcr zi=I?7PTa=JJ?0_nsn^hGI$QwMukt`R6y6UqKH-m8W=(d``PZ2Y2M}g1EIl3#!ljU3 zBs9zHd++4VDMT1kAB_yzabX0e2oE56ZWEi2jz*Gv$9Ir2KQu4kzE2%?*(ER}m(b6_ zgdvyr{ZXHQcdN#l#wT|EaaYq;*dKCDHaK2wjel_IZZb#_7deN{3N=dU=+~;2KFT5~%RGR)1IWfn_wb9A`_< zb55O<#Fioa$%+)jBDW{hwEd+N#ND&#=E#Y_?uidFq9a3I>&WD0n#X97y1eo2_Id-zXXKNUHW_@w(= zX#61046Ly*7Rlm8a+Uu3Ss+BibSvZZeHFv#e@HF#F(?W83 zKl=_w^HQx|U9FXSg!tF{DH*~}tG=G>W`!w#S4heZ;guD-{1i4%pStpAewOBZ{3+@Q zKcP!FKd*&7j|s|}thLWUiaFcQ!>VtMsc;b)IXo$vD#8guvQ4SWmD z%!z!_9l8s7o2_fgA&e2I4hk4l9{yytqqQ`fVv4V`01sU1KCf*) zr&h*l29qW#Q^fIBMhD(a0N!hN6%UqBJS2PK`nfbx1&VOdmE1yzskH7 zDKqTUHn95zE?RJ9{ZUF$GE(9LURW;m&j$Wvu8!*Q=l^M1UGs>I3;CsaP=h^N;D+|3 zyjFR2x*^p~v#pZc$~m<7fp`*~)a$|%PB<~s>$4P*Z|fgmebY*z^}SjIJ4rR8jS+j_ z1(Nw8pVX|)nu6zvV-H2Ezx;0gOH?-WEN9&%8PDLpMD(kzSY7`5`S#sh-IzO~5v)2t z9&dm6_G$RbB@4~3CvA&rk;P}#cGCisF=k`p`z znn1H`qMDdT@%2FGH?5WcDZ>6On~l}#^V+J`?%n26`%5G}`mza4JK{Fo zx^eR+r|swdwns}ux`D_#ey@~+ts?poovrIT&sD{!%kNUN+R8-RpL*ecSAev0bk3w^ zln)^;sml+)KHp8_xEYL1H)kOtRAAf>g&nm;j!|m$u8NX58_Kuit`YoDJ>BHhuND zIMOn1chEZ(4?8jn1B`6K22*FY1(I8yg~CHT)PLR$)xmzS8_|W;juI5ze~lO>=S++H zOcyBueR72-K)PWjc!Z&i)Ng-pJ4TED0p>{_2J;llB#!=q7~Cs*+~Q4UafolS#^J!* zX!<3A$S01M?`UKbLBN0yu(B5~s^CV8BZRm(BG!=wp)g$(gu>Tp0VcyQp5{%~%DE^s z2|HYFLP)<}<(t$nAeK#4%h1f}42OSg!+Y|M>WC@}+j8d0LWZsW!3(j*u7xPY!g(a; zvc-gA%JSGp=C?0D=}pn;;E zPK;F{Yb{TqdHc0p5h^C3$HN1I+`&Nx{6X=I?68300~+>9xw~V{g#LH?@It1mASvnD zx>4U1$D+4fumN^T+5CjP_<=9vOs^03#R`(E#|Z=o^=3mQ+IOP} z@fI<`vMwZ=nGJl1T0y-MX;rL7ey5hFsPSSYB0Mvgk;54ZT;&6m8*n5A;erL2GjJ z`%X{|UeYdhRQOarS(jBwOOKyIqEa>)Px-EdaZvr}l~vK8>&1?zwuai9!>|Zj&O-&s z+b-z!r`t#@crFe%^;nF|dRzR?d#?K^{R~8y}DW3#_pndjW1^1hV!EL98dx{DjJ} z@b&6P1o$DD863C>C$SI3+c9qJ|OKx0)%?z zvLqxXb@;&i*xT8zR`Rc|hR@F6umz%IJI1+SGJ^h#Tmcf=@Vla)tc|B=@kI&3iGn$7 z)ub(iPVb~JrS?SQ835yoW?Qn7W_D`D^nt%Rp3#OAnFnUv#R*8oJv)WhQI$JV^VG0- z#PpavB>0~1PSSVdRGG?KqxuwrN7{*EW|F8*>Y~{ zQ4yDzAm6zW6PX;bK8bNNtRf&MW}dZ3)zt2~-2r%1U@X+MD#M1l{4B&JWVGS*2Fw3u z&j?Moh~nY{z7C?viIUk#%7zEo9dj(g;Rg$iA_oRrz zu)-Csn2KR1Yr=XkpAvC*M zV%eTB1JsR_JWUp{GFmfQs>qIdybU$QO+S2Komr;|IN6fqGBKi=y~Rr^(prg{v(ZJE21;EKD`I#=U8sP-*J$^!uXWza;B)u-8 zG_}oc{j?n=ZPl~@heZY^@~@`YmKH|~vf7lZX_>Qvm=$DV0{Mk9#KLhTCm4#`y&47@Q%R6VPNF)?J|XXDg!t`2|5qlHc)VQcW&6;}Bu23e zC`b}4@tXOONRo*GfAsM-C8=`PkkOUDF->BpSIh3;-e(g=-R~oVFJ{OHzBA*$j{o1Z zN`quZ`OyQhgNd;1J`i}=?Rht$=1Nh;9(@YWjLWvwCAP%%#=)6;r;_wwk8>bS*3t$ls5URkLH;cbFk-V zz~HoAlUxP_E*sM}Af<4~2|X#m)aVB%c*Nl~eY^T>s2sAAo~PzRxmbJ?9*Nrl<2W8q znSW9g-tWZXAC@5V5gnT_nwJOu{U2yY&qYp9qt{uft!}L3{FtNn6WHBcG3AtqbJxR`GZ&BuT_wJGjGjJx7~MYA?nD&9vBZDZMZK~Y!)x_ z^-^MPssLd-=?p$?c(1nG_)t50LTamSLoz#o7H^J1M-L-E;a!h2aKJx;Kg@KALvZO? z_+Ypqt0)vXP8i8~DS_d%Q!=TEry^D69isw$J57JR82rd=uX@=w0uAu`T1vBP`(d^g zqTM`Nu%Hi>clZe#x#v@77Iu+E{Di{aZLJ}>|3G-g80?>&Qqj?fj*Y*}TTaR539}V6MZ?9@Bm$bm z8{+_D{K(@C<+V=g8em?SI)VKU}mf;8D*neuo~DLmA@WxHl8)eSm>!w$n_+X+sIX_6IAV6@?1S>W?!Pu+5H~r{H za1ayRG5|tq0CumIAbbsPpg@UGl_CPq?u1b*G-Cbs<_w3?aFn#Sm&6H})27ax5m=__ zray&MO<)FHyP*tbCOndI=%T!#3+`CQcFAhMbk?3sB`;)wGGYHu=JO$IrGOsr$Q)hE zC_FpCo@EvWDwCdxuSOZ{U`v@S!K!nw#W*J5GAT60bFp{X<3y63O|ROP^_x|o!ZIg}^kdY@0>pgD+&SpTTu z7)>GNZi(D;?}pNoesU)z&3{!Gx3a?qGL~conv{+x<**w31UD|QJ?Tx|)oeVs*H7Q8 zTtc&T!IqGNK~y|EH#Zueke^^AWVMr@9qk-6+ZU_9e*8zpol8didT!_=xILj2)2ggS zY8P#FrN~FwwjtkWU_UKv^39LY-Rr?@qKYCL*8RI%#8^;L9zxmH}V%uynKMXGN-XiH=)N)QxQojgc>~cw|Tl5DOyu z`^7KL9JaHMj8V33N{wxx!_qhRro$rV1`qdfHuUC}SgV0W^s*p=igbfg25_)9>VlKz z7XT-~ zdX?AX)vsB6LyZPH*hA@PuuDV6r*w=(n`ffAH+knJTF!+uBz-_jR4~P5%zoRZ1!eMK zwcJjSUfZYGO5Rb-%ah;eGva5n+nr!B_#~p(ofJI3y@W^3pN0Lqq5F z_|K~B;E`HnoxF2QRCSg7culIpfF7o7WnbBZonm<(?`AUS<)VFs)hBBa9)O0ih=CG6 zDX~2eiP1Z`21}ZK?)6_}N$x%fJAQ91*K6qB2W_*Sx=A^rv$KT%!+q~e7`AS25?)G} zN2$Z1hjL>2+KZNI-%K?Z{~NQJ)?)noa@7TGb-VGI!%H|S;5K>O{Ifvc9BWNc%6eIe zsXJBS^!TB%?OV;oQ>@nAhiuP!jFUsoTCvjsD#nTR2S-0A5@lK!CqjI_>9hoRylI#1 zxnU=NCwe=?QkdAX51wbzMRh`^Wv5EM-zr6hxz_3ydfq0dSnO8%y-Nrg4{GyMa2XnY zmZ-Vfy5x6w8`6&T3+eflnVL~M9kO`wzCVooPw8~#^=@%ThX(Ie!TOLZD{p|=$?4m$ zg+NOy(O;1c$?N+I9E;xgN$xi7CMR@5-B*G11l(UxuP}WDjQ#bCPr8o_^ReW-0YTL# zg8IL=J9<}gE@RNz<+iQTuy|HB#wX^|@hGg7vyHMDd+o`+hXQ@yt|H^t_mN)6x4jxU z>+O-ruiN#H+YPmz2nWb%zs;_cALrPyYpn}%qPD8cUWaRd?c3tkhhjf|CXJPExp_B< zvE1#vtA&FGBpZZSU{djX2#hdn97kmn?*lD?sEPNb)AUPph*wiCRPNl98u!$gF}gMZ zr^Tf?>uzs#ihk9s1Ku*1`~l5&q&Jtd$N$2wBJ<0qA9_CYz$ZA3rbl73N+J!l7R#Oa zbZ9rvMywvDv{61QMwez!DZu&w8Y+qAM9kx7im@lxmC@^6`~d@Ue7`b4k<3YpRXbk# zDRFX*KxexF{5O;OqxSm8bv}A%sF;&yOX{@FSQv88_^=#ARR(;pyPiD$dT+|*SO&e` z?(^<-4&Uf8vPZdXogbw;~&|p*u1%j9X z#x`vM9>9|=dt)t_?7r>0tqvZQTd|XCs$TZ-%YsLH&GG4ZEP_`?bA6=Jq z7HttkcL%t2La#*uCSeX^Pi8kEj4*mqp!--)WMP$f-o_@>@JjK3Z`DR$6us+#dKY$C zAjx<66WV>uZ~L0YyiwgHJdk4JU+^Ixzts#Yb$$Z2i@C;VCF?NM8I8M z!-w($ogs<%N)l_{Ed}1HCEem8t^?gB_}Fmz zUnx#wsNAj!|DHJjRc)304IP~zat?J-GAt>?0dP*8mTa{3zt_8T&>E6QG1j!ayez;Y z;9k+n^nUof&hS@Bo-vQ$Kwk0ue*V{K?#E&@_xq1~cPfvDNqtWu{hZrU{tC7Y4%H57 z;H)u1YR95h8b^o`oU?O7RWPgFM>9$s5=^&9Ho{o`K>K{ADy>>$ra>#}OimMU{xzVr zZ4A6shB|qe((AhGfgAhr<0iVzeKsaW6Q;IAAxxa*2jU-u4~4lSq{B3ofQ}F!ya8^1 zCG1H9o_3R!MXGsD+fV3{$yY-@rDMeF0dD2zd&Zkg6X-K$Zl*@}Bb~<*=>hNKU?_Dj zP(9i|w2_4j^gNo!PsUEWkl!#-#aiYuZ->k#j>91VZ^m@4w<{#wo0RjgL}89om@}ZoTmodZbLV2#KeyOv}t&wE!TSE{dpwB zj<@2QG<=S9itqd+23p#h?l2LlNgo)K)ltR<|KN6@I%8I_OmP@|WjPJmA3rTw__!f0 zeiE212Z^W)F1=$`>^EdA|4jMwilA)GfcYfmr6?QqK7ZSr1-X(=?Jh;`D~4EF^mJ(c zRZKMbS@Wv_^;fgUTRaJL9Qc5vbTJa!({Yf^MJ0sTHf|pK>>*kME}2F)nz-8zbzr|bbtIBj&jCbgyy?@^)RG^>@_oxnQsBDkp#7$ zr(Fn7O9+HG;z*UvEC;yjxT^eH)j|MuTncF9j(_Xj2CM$T;YMixxa)Y=zM}hTn}bIM z88S{**=pxC}j2esX2VE>90u2f!OOj-b?j5n!GKY1k7bxWK5OZAHZpB zX-n!4#&>HX>{5N($n3JhEBSR&jbOY!W@kBzMi_I}O*ZbrKhU#Nw7Aj?CJ;YAFj|ie zm0qM=0PZihJfSfGSDaVf??f_?98$M%6lW_@sbs-L4|--1NFaL%O|gX^U)sHtu(D|a zxG|NR8K7PKDBu`S(;x7YblXT2lZn2JIWV!_{$f<@k>qNRU%l}&mEAFb@Jl4vQ`c^qfYjyw3#7PhHwupI87s>J0P!d$m-9~|sxyN$yc>BXMf2t}F zoLo`c=uJJ0v$eeikHoRf88N+$?6FM@u7sX^5CtGm>eEq>4{jWim@jEulu?*V@ezR` zW7i$PjCmlB^r-68`KIBopy}2CxB3a)M(>qagyrPC*FC_W>>r(-I5+hB)2=xo@H_dP zq~wjURP!e)CRTA1CxMr02DTF5=v56t;QBI^I=g>xg#H_SFY zwxXz$zcYL6xa`NP&p)ZB(g&BVP>wgL8?D~#{_-p*!i4}adO}Z0tM&X*CQP7!#2_go z(NrJT#*&)$UxBLaCt9sn+R-K4h6N8(xlhk#!9hS(l*~u%jnH_Fm4F}fQHHerJvMKa z1t;(C6zFn2mb$6n4sb~*CFQn&wfU;NcD9|h)+eShahl|cRC3mwQQjeD3l3H?{$cW!l%>GDz=^^R7P%6W~C+Aql8qBvlTZ;4y%?f2lro zLM1GbX<38R$2acQ`S}Rlq8c=6A`vh%vPNt75-eEw?|7jSJW!5T4X*^6o{?N77)C!Z zfn_FlnH1RfR9Lz50Q%2DHZ&IP9~GsEK&+Bgn<5pnU9#6HP$=vszE0o2LSA5+Kc!3r za-(UD7Uu9H8MSlvjr+Vd^LsiGE>lIS+s2(xDzId%VT~48s_!$=>H{LWSofUz%-~l@ zq=vEjR>{!yfHoJBHvC<5J__{G77~go1g_fmc+7r1l+ss9V=b1Is;cW}o1_i@9WZU0 zz20t@7b5a?f`>!y@S2$7i!-}4VstaanTrpl`Mm%9^+U5QXd9o^$He-4f`B({*3SAq zfAk>=c|3ridNFAq`eEyXugs|$eZZdDUbXanImc1aMGjxW>o`X=2I`ajI<2p zKH!(Z9?Y@tn`^pgI?XZo*N0L;<{7!#8n3!Q71`$QKxa?z{D!svcMo*AEq(s?yF|u$ zqO8<7h1ux07jdjP!6wJNfVip#ZrXFLB7*fQlocPjrO^lPYPMaY<+~D%fsMYaO#G7s z6I=e)Fs?^T!#ogDl*3Q>JmjW0nKm5J{GA@H+Xmg<`v^+}X!r@|3|CCRhe}A=`P+Gu zHW};ZP}-2A8wL4Hp_`F*%blQlov0Z%k#xuA>M}byIt_NkH_b@O+Y%ha)Ys0!w8vCJ zf1PGbq0RRxu^vA3#Z=#bKWo{{;!DC>=-FPrOP?9JYn8XVE!k~CjtU*`6;*}&Y|nIR z67$uex}h^#zIFC^?6REV4cYB4=5Wndu&G^3m@_(9RiFP0bbRsb{q#S-yhU-+1Q% zBFEYqfp$Dz)DA)%m{_s}n0iEV!qTg_8#r4E*X6ykk1`)Y{xjAW+Bgrm`f}8F-Sc(k z=Q*W!I5xk1zJ^P_T*>m(vCrC|sEjg}MyX#%)A0r*Nu|M<;%PGP|I$Hnj z`x?CuZe}4N*$fD{!^_jW*-I}$+b?d8{)eEA?>-in4@;DBS$Y`u+?JSq@vwk;Vx0B& zz_*Hs;Ht0mvgRgTj|!?H$j!tWijosDJ0@IlV&Cnv{EZ)>>pj5So6tsjQgT8BBqi#- z8MHJZ`NV-EmVCr=UQLhqYZ`@Wjv-3DFVIW&BT8Kq_}vLrX|$?YFRT3lZ8>ceuRDW$ zimxa$TKhqT6$ttiuc&rx?;Zd)iFquQh(d^KKG{!UvzWkh3s_g9LA>FUN`xWcVp2g3 zdz_Q7cAr4u_)a6TTi5wJ*+q51SjvM_125k|1Iu^K$s4s_*udkzNt%F{dh;(WQ5d5m z6E6SHc%q018mW|7b?+;))7G5Y@p;-r?z16F|9gFRm;u8tH2P+W2an^d))`=JGnMbB z22ligW3v2GiO~t)spq*bRQ33KHunEQYV59((+nP3)p+KXW1kj@^;=mAlboL+noFLl zXa9RIlgssO1`MQrDDg$Y7q3m(E9rzFR&GC}n1QE|%FHG&ShIuL9JFO}CIfq9m$Ju> zW}*g8rCMfT!lyrqW{x949EiZ%J`Bnzf5^6|HQb;&0P@`)3NKH+`#i^0`G3ybm))a= zDU|wJSRA(8HRnRSCfVqLCZv~+^%wnhAG9G#_>@gt1I>4t>S+`3zLD<(PqK439Cua% z$??CaiVT5?C4Se9=j$;v;FrPgBF(oaOXkFFJ4;jk8`-RW^yJckk;F3!CgA^}{)6yo zCUXIkKRo;RlSrTVx-4FW>)*)tZD(h&oPBd(B=q^fZlv(We*TfphtpC+hTOXVT4wR4 z(LDNcbJy6=%&b7|1#L~9Fvr}0^cZN{s;xn$`Gr+!VPQvXQs&=B8aeV;JyT8x6r=#_ zOsnZAr7%KzsQj~`Df@d_)fX~4W?759R%5570`{)8IWVT`l}^6Hy=-&W`p*w=F} zj)G%Wn3#U0Z;HG}93vH(efJ3Y0UlKZQT#Vz)!Z0$Sy0A0Y)i~MuKczY!;su8Xt} zaCQ)`wf7B68*XN)dy8sE!1ZCT6j98;1eslcp(%FF$8_txnF(NoJ$bdwPxxz%BP)G8 zIl;BjlxYg<@K2}z8RNjNx6zyby?-Nf$y;++a^4Nc1J;uhVY>J@%@pRn=c`#o{SR_@ zu(XPw-kq;>ThJb4g~IaDUZqrJ_CASh*>Zmv&dL~uSUA@vus}ZJ78l?hRCda zu%OqtiJ=a?aI}5))bxi+M>X;o+Y?UPpLwj9r2d(Q~kz`*ROx+riWZ%eK&q89Frn1doxzfLy?H_}5&`~=A%eFLRPOcJU9lG@hWKm=$t zokiGf)j+AyJ6bMd}mLC!FpOToL(FnaHVs7HvNh?{NFN3fM370-+Mx%!O1T?Q8`^0&w{} zJZN20N8NjE?HW1hmd99aO-Icfa`p42Tx!~c(Sy0Cq#IXY@{G{jaloZhShVYV#QAbG z??^j8M31)rQAdVIuzVtIKb2M{d?sYTB|!5j{Q(e>3GcMy;lW4mbZBTeqvQe71AvjF z?^=JjeOZDQpUhxmUv;1d!NDJZi~MMQ4qv2DnXUP8o0V`PDv;_Ju#n5#azSTz#psOV zy)aS_>D_(r!}DytXAL~x#ul_=E> z+UClC&ts_lSQxXJ0DIEf278)c8C+_(6n+5Aza7db*Edzn+#}Db3ucx6y;&bDf8H_8 z46p*X1LG!&ndJncn^saRMgGP5CW`EdyIZ~0xBuY3VLvq-zX=dp$PV;~ctOtL2f@bA z2ed#hF>H^MEx`A%eS^yK(Lm4U2K5_YbGEquS(_mW@t$zKCI^Js!AA3|gG;G54}qGO z(RxKf9L>XC*HAYjKEB@L${QCM=;9|7;zKDHF4y##+8;Ftwejs5QeukKgwZ?gyx zdXM$=_<`7`YSCA_p14!1pdZzLbyCCc4D@_`48=KNIG>N9rag>L1+XduF6|p@&6tjd zxb2C&Ogh&q4eUDyo;tXK8$=j05CFCGHolLVwxPWytq3-}+)Q(~0Mn+wyx2B1-#fx3 zt6A$j%uOOT&PL1n%XvDEk+&T6AmF5C`phffvUuhkx99cfxT`YQRdN`a(90G>T=l;} zE-j(JX4Oy2m?umo+Y+{9xVJekL3M%dYB;(S##0M(9QrrYgC1{)j8S}g`Mbmcwne@X zVF`?2^v^}Jblk}6XxP%Hn?{!0!5R%qJ#YVge4PRKpr(m{)2UmI)hy&i003VR z;hIXf7y}t%m3!!<1X7#%H8gtYYQtK4J`hj22etF3HWvm>@9g1+2nYMp4Oj@3Ec83q zSIw4{`rdeyFDy&6x{xKIm23_JEvpk7KY#39-V1SOd{aOt-b~eJk!%oJYu4zj-%JOA zll`&~8f`uwvfGz#vpwK4PkPnNRxMT_(vrS_;V^!5kURgfe~JFTsN3*+z7vz&`MxDa z-4@uJ`CsiG@R~wnYCpZHt;O2XmAQeRbS8zW>^)yP0022q>j(RE%JFN}2ry)xexW^aFVRQr%>PU(AjL_m7f)I*h^j zNx%S|kNH@AhsKK&pTj0n*Pe3Vy#8}FJfeVM18{O5T-VkfSW@c+ylYtKRzFw5X=BkF zKsfU9SmLcJJ*CEW!9L@TWOm*?zm0<>T2~U#L8(B2kE<@oqV? z&8%SRooL{4!DFk<@N@O<^7b|fvk}aKZCt!yqSUzbpLvLzxs^T0mzmtd&1=3EupVhJ7koCu@jP*MD9+ILs%YVlB-3LS9zaKvd{u;$ zKO0+S$LRG~v-xVHkjGhO-URO)(T&>*HE6EQWd3Q9ELGD^Spg;0)==QRTxdP(+Npdp zeLxsORR`5N8(2EH`R`*hh>W#>OTDI6KcK2RC&9ht78DWyq0|G+6;GI-cqjBK{Fohr zxmv*YrwsPlqU!$zfe44R3t#e)H+nPvzTCX`+Mbmhw~a2ne9{cP zlDvm8#Q~BP8O8_AI>NW-d^|o<=Mje^~I$k~6j#IY3g|!rcbuN}#6%y;>{~lJ{fua`UXCfy*RsA3@ZA?|zal&Z-?> z;)BD5p*8Iu5ZhXJdXC`vlG}jQgH+^4f8;7|6mX^+#$ED_0@9yP4oz1sx&C{_|7wdbk7KtY&^G9)oiH)MM zB{(NrI73?8YxA>*hy>=&G?RJvTO#s<& zrvBLI_2vv&3vB=%%yMmDubUN#TMjalc~L>Nrab!7M*QY7l7c*LuNMQ5TVj$V+43>! z!QJ^6H!cF=s;e;gbnd*ZiQJ4KJ*IH}?j0`m7cxG&Xj3()s#;QuX^7G6>{ zJTuRWbmht=+nm>GMshE4mcFSbw1dYHzDZ=60mP3BbJu~ptHmpQ3Ud?D)9$(-3s;p{ zGOXyH{#1-c0UlCO{#}m1t7gfTV!zveU)gK}*#9)SuF`S)5;%YX7HP>JT~9ehc%8n# zW!hLd>1@ueo?vwmBcQjj{+c1M=n_%cwUwRu(kDF;Amdu?I-j8PnNx9=x$5Fuw`4?o z__pNbFh;FE$H8PXnY$VTTG5i1t-Bh=5<2OH#lp!S*SZT9&hohTx-XW<(->F|a=|I& z1-?khmN((k1f23fgJ&3n2Ab=e?}9C@KdcouYHE)GNKu|kK3Rc(UGGsw?ITUj72xE| ztGoB(W5>n zm-LSVU(>%}aL;mSDfl4w%Q<3K?c~1XEeD<@^4AP>{5dqM`6~wiX;AGkB6DBS*1mw- zwgBQ-<$;ckJdHO<~Lzk_+{0w%Tv3US<{JrRLBRWt;bg6t?2px#4vS|id$_G~F#O^L34=GSs-W%$ zZznxSBHA#v2f0=42Q{TQwh2G}n_VjX1r5-%aI9g`q1FZQiT_Dx{}o}r$!jg`cZ5%r z3GcOn!&2a`Ldzs<;5K`D> zH93>-T>mSEPM36n>6v%lgnQ96L}L6<&d`8ePO^Ck_F8k{mWt|*@c^^wg-qKQZgJD; zgA(#A&cmKumy1QO<)9R|gyP_6dhD)LdE-JVolfFkGAT^o21DFUhGSk zBj+V_$QwVq9ynC{maxSqbMA&TWG2n-|6CTPrx)0r&6(TS3sinS;P1Pi_WaD-&|;yv zZd7SFzdFzGFp%glb=!iV{ppX>a+y8_vilZn^wEar<(R?|ytD%gnW)=zgXT%Ye!uhb zAdvLDrd+D&20Qz_m;=1xqEJ@l#hL@?lXXgC@v*r6Cf}{?i$nW9yJ3-u2d+OC&yJ+D z9t>_dn@;yLvsnncG~qY-DPYZfFYdYiow*yEA{p&4owO2zhjytyeEj*n*`#}!4j#t< zvs5#1HsCO{XQ^sjAN{_&vRt98e;~V>tlU1fJY*I|)kt>A#@vjM(-=C)_a!L*^mNw5 zbb1lf;!x|euXsN1?`yrv` zOS)eZ_ z_p{>y#<~9COWdBKsyf3^oXYrr=jAt=eQLWnzJ3y|H~YRl)x}aJ&h^rg@uxn@>N1Pf z!vtLx!RF<8?-@$unr;+}EA2mwsaC5I9MZ;=N4anWO$Q12|ItPtgjkrphBnWnHE;2J zOkJ#vlp#}PfN%|fO`QidYB+fD$3Xwjl4h9PZ|oUSD-rNtm_pY{eJ%?@;Pk_$feH6x z)asjS%Kc)}Z+-Gn&0A`s2b^UK32gr*Ex5+P2X*P$e>eMT?|-K^Xs!%QpLlFun)mrl z0#kVN|Do-w-%ziAWAf3mgaIiQRGC*L~jSb)E$_Vw@9#@#FF<{WkM^PxpNl^lq;%{@d>u&KBxk zS*=|3$`Dv>Hn}+qa;qGE8WrK=aIbV1@*BRjTZh-c1lBB6TYJ3R&#K7rNYj2x z!RS@EO<|ii*twbZ!7O)(#@dd;N3ph#`Cdg1M^Ea1K!&5sMCZv)$pVj}edIgA0>AF2 zRn^BM59o2SjHO|*i}LhmiegQd_9?~_?z?NG;<9ft5g#u(Q&NAj)u#S{yd{6O^);j( z{GIHXUXb@5873=9#r!w@Gz7ORW-wCdVzy0JA^^5nHf!sxhi8vZINpR6wQ7I*kKhO& z(Fji_CqZ}}jrCE?^C%p+@-lb6Gq?l<=)tbW%Pv>F7N~_7)4U_s-?n0$KPz_T?nbk} z(KtWAgeebx{kC_6IH<`H@jGK)VL3H~t$aY^rp3{{J;;`N!3~;ztG6AOlW$wkkF>XK z2-ls}UT2zUPFm^O$@X}(JF;UF1UvaglX`u0@Xsd2csC|enrCy6r7=8wXv13RAbak? zZ@21KDdm-iD>3{_#AI|F*mYrU~kxvW1|< zTv>cogJJ>7+$Qhqhlww{YUjylbkg=I?j z^cnlC%FV(AZA0uR-K1T?AMhv{(BBeH=y7@J+a*S5AZiO52u%EJk{A1!9%!Jit{4|$ zncLm^cK+mO&XzN%cn*Yu5nh&^lL=n+y^*~9d+E@W69-RLtA27G7ABKg*`ev{4W&Vl z3bXFAWywWqdm|lACjRE|N7Yk6oXqjzF99QO2pmp3wA0HiVoHaOcOTHjvtjpk_Tl>D zUg;mkS41m4kP|rFJp5yj5*=tu{Cu`~`QP`?Z@!8dwGWV$hw<4Mf{3L%!3XlDZTchS z_nlo;Tr{2vdK9~rc{fCU_L|x`N}n=sRdg@!!S^jXFw{GRs}W$ovjtkIEpVIuYvCmK zad7s8}~(TM0o*af|C%Um^FivJGM!PxP!n%b1*3U3heyY>e5t+-9)t7+k~;aASb-& z9!MEyY@W>O{UPT2xO~S$K0a3$HZwAelaaZaK+}%7nCOv+r4P+!%ZAyAVTA09uM>q{ zZu}k*<-jd3;5NPPN>6yURlh$&ZzzzEmj2{%&;{;7jYkvQ{1iX+x1)n#lT+{`OPeE# zv9-f%K9Z(Y6ozKo)jFFFzkZS%eJ-XW1_cae6 z;3`w%uCMFsVy?UH%13Poh%Tymj5(?HJuio0P@DeaQ^C#RYPGeshwD@29hQeT#Tu@> z74EGy%k4jQ3|F#l{&s`m(wcO?fQ7K!)@oKNsd(;AM9Ab3)<>u;7Og_dao$9 zH`jWA9WFsFx5bV}GOVxVxt(8#?J>tM_Ba^rkZefRe9WxLSi00npGz(W(yL3#Mft-f z5gRwGg5gv2sfD`~SJT6cw}9pG$6UHXP3Tg`&Yp$zC&?zId+U)-nC0_*XrK?uuzl`F z`Q^Cc#a)djmSa^wzdES;R&tR)ZRrw1{*vUOA-k=3^C265sR-j#vu<0sR_NP)L%YXh z#^ykYzA|i%p{w6;I0GSq%L)Db}>u?bQD)xj9)`1 zR=n(5&60YW(6OPWK(mC9kKhw*3s>tF(`*a(CZ2z4c&93^nZA7fS%m3koqF^j+Ii&p z=)srB$EGwmmX|hveEF>|Pu?^QGA&*8s>m+N>*v%KSN)Fqa)iK?r8F007dtqz6~?bJ zQ5%A`Z1bSwfB@%5v!4K-#2x(Ku6R>N>_!^YyrP^E7{(WOY45X&S|(KxGkQ#P7f)!A z;O2T&MqR|8&a||HUrU#qOCXp?q;75oH-%wz-y^7|*S@pe{r0@yI7$QwprL#gBZoo+-#H z>NzCK54zu=8^L#N6aV1veTh8ysCFfr3hJQx=+N;g#VhHYLzjJllPk>9Tx-erd?anf z1TFx(y&8{PtN21FF&-J6K&m$v{-gP0>)npYfthP#n56rLuGynf!_o)FG&I572kV%_ zt%A$!Ua*?;h&?c)uKE%~ z?J}D~d3=>61_a!HdIVAx1^E!b%g=x4oVL`*;~yl{%)csSpIw}a?@Xh$fBlVyaAk|+ zKOhg(oGha1vSHTS1@77Dtqd@r|JREz!ZdgZz)(m&)nt(>C-LU!vW}?gFbLjF4_oY& zRk!FA4$vUhZOef4meJuU4lp3RpBgR|fKcX)7J^O3lPsG%JJw3d9{V8DodX(&FX8A-3m1wR~oEM-NE%XMMUA_ksu_&*MEDmg#sTAdYh}7 zs|uGItnBIIemlE?%SD03o}V#!B6P8b$?E+%#3w)ALhb8H&u#lrha*kCJZtLw)b+o0 z(rw6x0~^3i5zBV9{o}$pE`)~0u=TxfaxZwL7YLtvJW;V*e4}s*Ip7X128lXT)I;dFO=N8wziZZrY4$)!iYPim`y$*n z2<}TnKPp1oryBUCe_lFRKEno-KKtx37~Zib1qHy#cL}+R(Xg6>e;m z6vqe-5i`;V2vG4eB|m)do_eYOdgiPQ=Imx1+HAdL%wMz$8a=oiKBV+=gEK7^v`A&B zvkE|ILcTpe6aryyl$K8;Ag?j1m#o=z&BcY!YKw{(s-mkltDp{pDtI1WZRgq{Snv5&Yo8LfN5^4)Xm_~kRwe!>G$7m zwI_;qa!AO}>lqH&x08@GOe;^z*9Zvex1Z3lTCH;qdVZ=hgkfdZ7%3^)bfefv-WX2U zt`RbZ^qT`w&Brx+{E`VH8a>#ToS%x#^2v9_=0n72IMSLkv+0C3CI4k;j7O&b?^50# zhsmpLHHknVG>f@-RW9~u!&$ZVc6xd6%64{o7xwX1nNm;7FDOP&Y{lFP4R7=33cBU(YZ}O&4F!;5SBff~jq^zQh&etMw6r)y(B!2z*gP4BO0Eh9Ie^+@^k z8ph6Gx?Uz*#__BV$Nk2qHUT~}Mh!5Gh60eppQk-$4`d=+^)aR=uPcwFnYfnV79023Pta0igJkC3-q2WGb+h(W~Mk*u$a z$dUCGc<=0EEHv-S4Y+-I=S^HMxV~kz)0{dCX2V^!I@Ar`$K&0-p0UFD-t-7?qfI|0 zI98*O@NR1V4LD>gDC(17gxA%vBzjvyCmH~Hq;t%zErV#Qh<* z;Zajr9K&05b+!LV0Ts1!_Hz_= z*g+UvfZYfy(?ujZr{B9KpqMnr<`V(lP3Tx;z*J*+A=alJId)}}@Elje`C+c|=otJ{u zQZ3A&Ip<5(<=aMzw4SsM5d`gF#plUxtF6ikygf&MwIArA;hvgayBI{To+(t!Ql$Eg zGfi@Ig|uhYYfXOk_vhp}h9gf*#L}*;!m>r;u&3`-xK$Akv34YzhQA~5f$GZg#X{Cv zI)(x4uJBY~UQQT{R*}{A^)Z4=9*_1dT5Dk!i+&k0-rNla586g7e@P_2e9(zi&T-p7 zJ-q)0h5f9N|6U!NuXGrNUW}{>?*`v(nAQ|4{>PwjmRISy7{7z3m%e=Eo^x|8fYV=K zsDen`x5gl38>6Nns6_7&?r!j{xu8ik_EZ~^jo*U}v-U}(A3YawKbD~Q%O8Vv&a=}r z@JtrjIn82b&0BhUbDC{ns@3eYw#wsw$MkRMIE2-m_Md_Sr~AfxSzX?r&Bvzp>_&H( z(Zm4Qlc70acJ?p`+4W$7Vp+ePJ++s`5g`5EjF;wWNtK*Kv+-Y01A-S8&%CStfB$?9 z9?AG_#qrH~VrljuErTpzHePadHiT= zw2HPI+~#nXj_+EsDdhOlfte|pjJ9$SM(4X8KHDG4{m>LAY&%v5xu2?0Xo4-wty zaIfJ0ktFl*rBnMZ|KqCQrBiH0ZstUzva8&tCCxy_%ObCrKT{(^6AWL2x?*`9)<0Jf zZV|uHSULqYGx3Rv(eY`4J}DIj(YqlO4;~ZM`!Agu09UfHDpqoq!vr4`;;SKg~FOxh>Qf z>tb{-9nvyhh{;In+-@xsb}JGf6}E~FuE7r9vUGtPdOBs8TrR$mm%tF5o}u{rtSK@w zTK{|eAMI#UrPTW|AWvU|9FvvBY-n}tj?U8QGv1;uDfU-l11<=?|Gv0!ff^#LBdXtT zu|FFcPl6<17L9FdEX|4W&|&tjI>*b%c*X*)-t^qseKz+&6~S86=l%w1Nt$`(nRt+a z3+vBLLS-<;%Uh`F+CP0ECdDkq-N=Iie*|H;zaZ6yRU}tf-$LrpZ*-%-nZw$0zma8I>b3n&?HR34G)1eB6h z$P$4bc(6r)B9SXWrRfYuw!Cp8pNnt(s|k#OAqg>=YR)w}!V^*=| zorCLl`V-N>cg})Vvlw@4N0&uG^M$61TFf!}ttc2$JS@pQ;PO^QG1xHnxbXM$dcK|iv8xJ1WHrWJXm zrMZ{Y!9jCxrhXX5facx}+Kw<@S^JUx%kxSbL#3Cj@o#J?meff$BI`6WNrvGuT`z7jUMN|sU7Baui@h+N}eh6-d3^KYn_ol zK#mjDiL9|Bd&fN~|JGx=53buScLN83c1&U?dtho(qJB3mU8UOIIwWSpu_DT2oL zVz^h;ilmo7h+-{Z1hF-5ht%z`4Fl6{sGll66{(NBt?5(-RR7jR*4WvtzQErsp zCtgtcc2;ZC;%#wsidc)y;c9-G_U{BQ^^1;QS-oT)ThWfDu&rU6GWQ5@7}2Mj zb8v~0<+$byyr}R>*hdogo1QmRNO5Wxd7;{T1sxUmkS+-CyyqvET<1g?H+|r)T(1@q z#KVWXdm5Cnk0T|;b~oqo=i;TeIOO$qx)&sC*Op4ZX@Z&z4VU+{2m0p9JHZfkw_sDX zvMrQ59FW6zyU&@c`5eQ&J$>ld2^LB4LcTB7z1-E1US{rMtT;;RU>sp=h!^XxAiU@4 z`Rtq+3crew!H!N3o4i$g$Ht*f}j$0l+(h|<*{Wnq;8iI#?u!#26 z>&5wXAaV5e^#kv!GU+^+2XGRq6Zvyi+w0(441h;76ck3 z>jdfDTYuIG*2~Si37Ct=hgPo*v>wIV1)7Hlfl2M{@OUS#ySlTZXMZDhL!~#Z79b+YTL~$46o1)*A6b+Sc zC}S-t&;zAFgpi5>HAbfm)mo^L2ve<=K&y>WV_8>=b{;-L0Oj@GK&}D^rzd0hFe7Ai zM4?2lSn)>3FFCe^Lu}wr+WZj}wuDaWGo+55u*rLwQWmX6x&Mf8>>S!cNXo$0WAI3qO3?6r}sVUr+Td3nO`mJxR2GG z8)X4q|B+c!66PclEZ2Fz(K?*8i0;!1w2(04FIsF`S^Q2P__}efIqrQ3O!X~+3u3>p z`q$EFgfa5LJ9h7r-&U_ZWZV(PB?lFRO}*E^jJKf%q}sO?289rdxnFfwHy4H~2p2;) zeN7=3XM_yv^_CR(x`dS!N%y~W(SlfLyz!Jmz>bC}yV-u!Vwv>sDDXt&}Wu5-0eVFoXgid)=41{lvmoiKY27w0<{ zxd`ywp;{ue&~q-a1xESTzo)7L`jqCzQU0#ik0DE-a^hQ4nka@rD~f{KJ)~v)eB#Mn zIlODh&ZG7te3hw_t~LP5CK5YSL72T-Z0nEm=MBWj)8{pBLoBFhmTT=i&QVM>kgisE zyEa1yUHcr;QZGg?6Tc_pHBj{=y(gBs zs3ZpUd<~*~mJyRQ_OV0r%}@(%<FFTMNL2P=LY@9)!w{oT?A8 z9}BE!;}5v-U)tNAHZPg~{eb0+0Tpte$Zsp$c%1!QWS=8_?X`Dll#q6%mfj!PK!{4V zSz3R$<(+}@|BahwrR$E!dl5f#E2J}v&5|KSVP8H7RM8nJBIY#6|6>G&T@CM5hJ}0c z59vj7UHGw030`r2uH$-ukyvO@I#=5MiQF!Z&_LaHm%NM{9bw$|5h}8h>Yo$(8>IgZ z)V$YyAETX4$4$Y$wmMTSH3<~yq`Nc&tdtA=GE?lfkN8w|8BXI|8y#E&OtjC}q+DS6^lcCE2VXz5pg*=4qJ-oYT)Md-y)16=_*p)gg3AN-YAEe#D||FhjRQ2XER<04WM8Vi!Btx5xZ z0(m#LwDxDDH8J<(-GU{>G>a9zMjh`iD>yf`S-h*%-jk+)JUK@2OBtV?H*>0kQlIO* zz6CSP*C!$deM3nDK?>iBehW;-}8R$#MhFo#B=e79x+uOp=N3qKf3QX zg)CkcY9L-*I)Y zcQjZ9E#yOH`v6Er~G&YRts|B=RleB}&MA%Xq0;=uPV1vbFXj*(!>xCd`(%?2pU$!d%Meu_rFhwvOD{^3xj zD*jo8JH7MK{DUqAiYNkYwi5vA`c2ycz4G~=5kuuI@9|GEY=8jAOJWEDOhI}2n( z$WtdJc5lJoMfaXoaXS2bX%*cIj(;rSj~XC;PuK`#f8asa{tUEd;eX&AM{m~|tW5px z4|eogPvC0XZ}RCgpxA(y22kAy!O`pnV>3R#g}CYXgS;?PT5bRgI}DUTIadR&0@q3w zl-S`0QAYi4@QljC|F}n^BZj7v)zp?ltrL5(6J3TC)EH&hj$UIFP*~10^A&u#Xl=U? zcM<@&AV{zKJ5kK#w`J)@mz0xw3vL@5zax7xhV`7p?F4g|&#Cnq?WW8*4d{q9X5Iq0 zT;{8RQf-7=G_ZmL9O}2PsBXb9yB6xYRVJP{$FO^yU@_N#6_7=W&({lBlHtTeklG~L zc3e%<>}95R|H)dcI+_WHeYI(7M-;C^3uFNyhrf$Sxs2C(o`9P}5kRT}W3JykEB<

Yp!3Sm zB1+=ir+t|{f1%>r`RXI|MuB1z&y=DGqzf)?8?OR%wiCf-yvwuQT_VG&jS_@PpB zJ_9UFBTb1yaPm5Y|amr7_12zge z7EUpXX4D=Oa4qeNF)M2&e51nAF%$fZ$8!&ZBoZ8E_*DIq%bFKMEG_6FY`{~CP8PTs$`A%22La`0~AAt1FnxfMiL$Xx|Kq6dQXZwiu1_@mgKNy zx_-D{1V~zf2D6=osXg8wM82i6?au=DQs;Ng0Z`5xq7Vo6%&-$^KvcYg1J5@`6(xE&4ka35(3COSrmSR_A# zKHX!yv9Ar+TD2e23W-^xEID3{&|KX_-^TW@KlW%ZmJ6j1*hYw)N~za9)h)qsND+KB zMn6b_r2I>cr?~Ybza!O0wl?LCq=g^~=P<(VIna<8xnG(xe(6Qps|*RG=eanC2?@zA%ms7m2X9#LmkaLm)ug|{U%vDw zPY{*B8zLw=smjE0IH6y4zZfg7uget2J3YwQaU>c=4jyvnb}Rro)VizV{+e#;*<~y_ ztjCWTi1^w+xv9U9z>EI8L*&!Lcry`&`$bE_>r+G~&LP9xn|Q;b5KQ^Z>t26lDfOr< zoO08~VF4HBpR3~Pb@6i6gbZ9jL20WEYTUTpa8`3QY&WZzio24|dj~N<1AOh7Znk*x z1O|?2Zj_9t(-VVh1%FO#H6#|~a=-m>MU9ZdBz^K%{H~0I+^J>)?w1_ycz>-}7HSX2 zU$lS!;MT6Xn>1xA2XZ^`5dL=X(=L2bi@OQY&y`5n6xVzOa>;C-v}+1!i}}38Ovurd z<)c1b^DFDI2Q1DrliFiywRL7iZsk^Dn_HCOCxJp2#O4{{;3G|o{1)f!d42O>2w5kK zb?Wf3CW&po3dF)C(8(la`)usC_IbV*&(TYtt;v}E#HiPJAtw?N&`_8g-)h?4ja5-0 zEmgxvA|wW841d8(UV*HO89^S5>m>w-&PGd8_axIz7f2 zcCC$$$K>#q-nv-1SCv6u=(!A{GUM}l!H~1KM*yeQH?ol{SngMONBj!vFv}pc2pKn?%A-kx z%HF#?&k{f@cn{`p=zk5ZgdEviD={zlSEcqypM{q)m7i0SKJKxa&1q$zM0u;(8h&te`%B{1|@4#d}}W`mU^{WBu)1Ecd(1OKfMH! zlHTZ;N;t%%0=ilk^E@<^JOMl$D?DGsq*jvl5#R9*rMN3 zUk-m6QywUeSBK~XtJ$Ozd+!|T#s3)&dQud!eAdB1AxOJN3%##T+!P$45M{@rBZu+y zT_l23%MiXZI70kmdSJ|xFKpW%UfhQ-p-#TXR}n)SN1n1;b`*V(HS?n*5kq^TCp$Lp ze_|Q;L4^YJ;443FQHb;nDx~(BMu_CK7*uYRso?|JSJRqB{1KLdvgcF_)FCEUsCa z9JR>v<*!e?ci8p)XydX>HRC$1mFgxpd4;a}B3<%>=>^Fy%%)@VXJ2Ft&_aKuQR76O z)ZiTQenZsY%91vdW3f7E;_efRp}H()Eiy&MTZ}Z+1W42IpBKh z0e$~=LazM2fze?x(1R*XK*2~cg8$^2&;3}e$DFoyBwJlcEgLf~R-!@_|k zhed+NKv_cF_rk`0d!gssW=>n?M4OVu|qcXjpqMIEuom&s}jEW9cO)ma)!a ze-z!)I`3S&rLHUUf{Ayd!-6G|Fk(0B_8Dd23>CJ2{SEnI)zu@(SI@gJ0pIc0;R{dJ%J z?!#^>zcK%>HaIO}u{YLUJCN=Z-!cE&O$>RTveYY`;CiVgdXbR7k&kg$x>)XcX_93# zH1KplD05i^7D<4Z9&vAn$vS_jn4Sy&Urf5WI37;FlE~a!G@ByA^5kVLu78)`LcOa! z9efo~ov^!=*zHJe$@~K@U!uZIr6LBt)|$X*Fm;F^i|w+!r1~K3&HKXzOL4r>k?`Ub zF$v(rG^3_$PneT6UN}J(sousVo~1iZzNmXru#)4$ROl7uD8B~R+tWKbtpAn%`z{M8 zmHF;4HkI$4ltdymOWCU%WVChsEK`bP%Ik(I_ZUZe{Rd157?<@V5^3ivx;3R7 z4NSvt4rgzeAB4!1bu-@7J;b!`HeU+ylyu&^W}H3JY~ZYR;4v|0LE{#OtW&1F#(OOj zbrC34TG7?abEV3*y|R|k?Q2hsSp!dyU|zEMS`e`1EHQB#NO9rNi)md~-{SCQy4w3vOo(7imtpnZy?sOf?qgtG>xwF0tF6ve z7;U8zVkFi|I7V*O%19UPF*sid)sEQ9GUo>yU5;()Z_Js~bMdAX@L)%R^L-L+-2S%0 zxd?Hh;d07^INQHqg}U1nc+CS2NOcn7qMWkoblSMaC!ar(!RN^{^!o3MBd|)9XW<}|uu>DU~|Lx?`b0y!c`Si%3*_c+6h}qXF zu4#qn->tR+9gH(CjB!#m+vs37;4rF8VPY*l**h|G&JE7Ur)XT`7s}}$m_3)tQvc&@_%&Kc+K|f% zlZZQ}&SZ6hJGMXjxaIC)yE}&J^^7H7F;u=Pga};qseOC!cPzLYTqxW=f_j5C@9eu1 ze7^zz?InaRcQx7jS@@grt0 zHn0wVFoAVqTW~Td;&5>N*RfPgiK!f5@i@_>S&BBpDlQup9>2PSgQ^sy)Pv(M%Ne-( za8vtP)aeIu{GKIgXExCU|3m@ekDBk zU9l}io*=fntZITY5F*!~<+HMjeZbuCaTpVQk41LOW!1>H27d6*5kt*H-U+UATV*^v zgQdg;rOJ>E;D~v-2B{DpcurZJhP?tTqaM1NQYke_-NA!V%)a6YcTB7CCWIW{C&vhY z0(k-Q4-ljad&uf<8XLpXFSp@4EP@k%6#D$3;R#>n<#I(RSpKn8>1V3P;Kutjt*PS# zvYCn}wD;4d%bZ9xE_-1TU#brT5~RWGJ!FS2YePcm)u4e8pIAoLfJ_=dJ<;Ws5Ku#B zmY^jw#-MqwS@b9|8np9WA52mgdQUiG)PvKTtm@a*Ntm6*6PShx7o^Yovsykd`_aEa zs`Kq#X)=dY0T{`oLE}Ilqd87$+JTc2s1vX5rpH?819scno;gbbkJq&Pm^t~BAI@tF zpl=+~Z2%IOsUpuF1E4xD0L-yogb9VK__Ko}llOawQUQiS0uSd=f0ppo_H&MkJ7z1& zTd-5RJNY=kj!0}Sc?PV!pLSSM&A1-T`tIZL--1VOKTwQgujLK}M_4?FJVdEHzta5d z(3N5gmXpX<4aKM>O>Yr$#JqOJe>8i64<-76!j>?UtEklAei8qh>qYSs7Sso19w|p3 zJwG{fnSz;+9qfHSR85Ls@*C1b1iM8MTJ6YO&!4k1*^v7)r`$ljpA_NM$+=s5hN_y7 z1~Lr#QvqS#iD*&DO*}!$Dmyc{o4rEh!5O7Swh;)6!_nV6amA^?#3Si%k5loSio@T& zvE*yJ!t;vOH^<}3AZ&K9h14;kr}`w76rk5*!=k^BkJYip0iaOt;L*TParwGBrp@WM zt_>yj^N#>MBcnlosBU!U$w|rt#%T)cDHhEP_-Ns=4~zmMHPsSs!<_1q4;(Sx4DNr+ zKfr!)5w3|n?q|jnE?1EQeb>1C?fmawMphljPVROb6z`N}A_pbe%PXKzG zq5Vi864vV)I6kJDB&5U{Ga18RApIIU{fFrKf*L#2 z=-OB-gA>N1dk%ZKDWku6XszVK#D8E;jPY!1fo3Lb`$%1bFXT5DRM5^UViLfOl>_xQ zwZdS%Lu7+3YXd_P*gy89xA+@yfGx?hyOIGp2X(;0layRZ*r^fVu)v>*&f9S#ChpD zi*x(qKR!)6T8h}DAD?qTlYEb6a~@eACItpxSvlo@1*S{m`>EN5>d(dvcv!C3$iHJ5 zO?8LR>l4!83|&UnzKs|;GG&iz$yc=VzM82dn8qq}fGJp{GLXK3Ur&1>?iy^XJA2#4 zmrvh-h=R&;jz#HE5eLGgBNbn48!tSWmM4FtB#NMU>k>xHo@ztlIDH47Ng#zQXWsR# z9II>54^|%&Hjyr7hVp!&;nAG93#W`v~S+Vc=7s*IV8S0X`2ui$bU%SDnD^&FI^$$n^ zhG*p`CH9c2!P;~*aSfO!x8N8d;@(v8&WVf09svlST+dK8s$M6#YGbY1H?rJv>( zB-NwTcyd&eG&qL};=fD|2<}-tbFGSyyvEYF{tbqttKF+pFvIQQaWCfsxl$!1t$ZV4 ze@Avj)zy5P7$-?wPV8Vq(6T1WYY!f)b(vJ|PB{WJuo$ilAvU`O@khBO&s{qrBg%rh zNPpz3{y8Z;3Q)H4M_v6nX{IWOe_lHpo~N2LDaoHCssCob#beKcxb$?rFh1{Vx^W>L zoZBv4j6&|>{XK(i_twwF-P4}mmagn%ZvWvwX(H-3-S@eQPJd^YjxC~t_xj%Rm?!3M z5Nzw7zP@d0E&Eqr+?8xBDP2%Q_A=)u z;@6Gxm(gmO?fdw8FBx6dkETwMRxmXj7g8B)JGf8(b7fd&w?zVnnjdC0%Yj$0viEOx zdC*1r(vG+r3_LAldQH@Pzh&^LiK&T)ta9n8-I9^jKH7!MW5RIuRp1&v=erq6hk|J8 zCCTKR{7>wo6|P46Ne#f=XLilylcS(#XO3&F%Qf)N+R%OU&Qe~z@ry~xmAn%`=l^&cRrtzG3=0Pc(`6Q1_~jEHy;QCGtODGN*HrYicQQ`}C05h&k139P0qKZm#cC$z;Cl}^L9%mdjCZv- zn#$4wrOu>b?C__z_Or&`F*i%BK`={O1+inb`aPfwqxA?fPx1QYz1S7~4Y;Gl=)Sl8*)sxB;QCt<;AGU|>uugEc9m10%Z$6g3B(Riwjj>PDkf_obMaSE zGS+y7xH!-_Yr4XJgfACA5)xobCL-=1#wY8kos%!4+*&=i=chaB-u!*|(d~*_M6;FR zpDWL;vVA+lyl>dW0*%wnPR3CL`RPL7^ec`QgEe@>+gpw3N_dSK*3&t`ePCaV#UIy~ z!RT4ghd$rkdevTEs9&}!MuS)MH2vW8mEM|IN~(gF zB>g%c;tXXQP^up-?)>e!dJ(x1U5r8Gj4$hq5Cpcs9xXFOeo&NzdNp8zZU5SbU&R07 ziN-BI4QRq@;ffB_dlD}#+*-Fk-Y~d)6*I(BG8UFY^i_hjh#CI&~F9hG&5T)^>05{S@cq3SR9eV(j@1v*}a%DPcA)$tQw zS}ZRS%nmpx-RAY-_YY^?9Zf0-C`79#T zBYtqG@X;<1`S$7UbI+|X7!Iobu)eey5|)&WW2{ztfPO_HffYm5Zw}@&)|AHHZYqy{ zZCI#MeRgZ#@BypX7H!@xYzWrpRgmZ06lP%88?~P0(+JZ0)W0q(fBTPcp<|0-Py#$! zC+)e&o2)8bH&4Oh+H=#yC4=f0n!rOp*A<(_bsKi4LhB_vwsb1#i#22k%D=wE!HI{5=j#@7x zI=FBMBg{C7d-&Y6$Hv)zPz%?6taO*JG+E&k_&poiDwnNc-m~ob7o|Z(=cbBc@TOQX zyoey`^4o4~+xA{LOu>%2rVof$&z&yp8`&<|o?~o@dmJ`QIgZ~4bu7LcY-U>d`*;>y zYJK)f$ZpU3UE~&BNZ~0C?tO({%;(RO-RZgw&ThK{GD2Y)WZ75ta;#FTvy%FbM2s$W zI8W!dzh{jk-Iu?AOmQ}dgMHn=PRAAsdz}2lz(xQgAoBuz-6QGOUSfa_{_n3F1~#`3 zv4Q2s$Te+UMc?~e-;E=#;eS=|iHKd7$vWWn_dhQuJlfAST3-4<`S286eynZNr7Hh> z^!6EZ-NR8>N)+sfBytNTR{$6l%FVJ(4-gvaC{7R5`IIeM?ptV#5mey&U&B`(PzF@b z1oC``@hSPMq=F(~l!}$+GFR|VnIKV1uEXCtNy%kiXtx9RA1eg>LdW1c{)Rw_P|UcV z0^gx-rqaEi4zDn~i1u1x`N3v2927F}hn}Kzr(9@rp)02E-jlZEv#aMVi7J@}oTUA& z-GJbsYd6@=F=6VVUg!o--UgGOuaB7kc0=>QOYFqeb7~`EY~U5)eY1pa{z!${cvrNh zyO&2E&*3uS_K)S_!RsRn5E5kDMWi_Q_Z6I!0mjVIMiWC>_|M)~Dv0x1Z7q zVDux#`lDQ-i?(b>${HbhRu3;?9WHgMqJ3TZdx}Y-TMN=oO&>W~ zu1Zhdsh?jx$8W;-YmVdt5-p!wbk)j(@bD{H@t;5>OWbdTcVU$?EcCj6_8U?K4Xb%Ncw%mVrH_ze+ zxk@n~#(w@1!G4om5(RwTZ7_Y`hb&pIu9(Cx+WixQELtdrH+mR5CM|b8d>o&AK{6kE z@{XcHThnilM=I^v-&Ci^r{H}`#V`8Hk^EHq_u7a;Uh$Q;)JgH6tNDpuwn%y(a!Fs0 zTx@XCybA0`BfimZfM5BKVtw|=xIY`sIhVg)bGakM7F^Lzx6`U_M<`d4KgO{>FFUtg z{oSY5^oB(B_z%{uxZvXkr}v<4NjHCOFU<%p0~U~G6i7TTe~jNki6p{K5HoE>15?j7 zt4;C`Iy#^otYo_15v=-4SFvp2to&mtCXq{xe@W7hl9_&ZZ5Zn#8d>H+w+Br}W3$MP2Z>VDsdj$u}aTr(ol|>U)0IaCv##7aeG| zzVXTz^c?%h{a7MO_c&MRSvX;h*y3&^CeQC9XM1^ZfD!nF>UkRo$mWAGUk|K4c6A_{Z*sX$!1wI@jaKi!;4*Gu!x@Q6+8*7@$iCV3+BeG0310} z2*j0)MtyzTtep!}>F%3_Ot>FRv38|Uh*#;9|2w|%Wep6gIq20QDzANZ_ z#L_A#{1!?jG?J`pvJ{cuhSjmkOzK&cuip0W^*g0vY9qZ`@v)P=Ioq3${(v)%Gh02N z6vt}+=1D>^^3%$C4oVT``xazBeh!cCDaCs#H29>I#%Fm92bn11j~?qTkiFtQfs{qi-%J>(23YV~aO&kHX*{ zDnn-MC$_P!o*00P7F%q1XGv-#H4RGQS7zXb!@A213f);d#%zoh9~tJzLScMOR9IY~ zgVstTx9GlU=;+bI>JuoiVP{sZt`Xw?evEA zxUJWJab=JHKi0lGs>vs4mtF-z?}Xk|ibyY^NmD>60@6VQrT0$gM2bl7gd!px1S!(H zNSEGAkRm9A8XzS1{oQl!|M#5l`zvqWbCTVi-JPACdFC1H@F>A(@Y?a|dxI~_I6QN8 zT#@r8nmXClzt;*zJVuZ!R1=avp08W}=}gyz;Lc5AU=mDq6YH`vv_2R|hNa~}xj?rB z>eQEwzK`a21jF)B;H1%Z_ZfEHi$>O9N6U5Hi$?6{*FO_c&ksmHyz zL5(;tZwu^gvo>HIkpfCZbrZF`EBf=uv}h==-0L5{A>d2wsy08XA@Hi;>wLElM)Rd3 zUzit=OzS^=VCfUYSg`)&|2<+wPedZk(*FE-K2^J%v&I(0*n7XS*|`0x3cr>gw)(Tt zd>VKsF~ITGh<&PRiAwfNB+`8QL9L!wgpK?h?uX8VqCbs!cu!sCv}wI|EeEQ48dv&W zptxxRJbinX8~ zWm+W;FN>{#vV0#z5qLHBoLzf;(7W8;Yu}Qf)LOGb?~$^RVf}!IuR#qk5|zLKXLq{b z^c5z0VwRfDi}pfTvj6ZW^8x}>WAk1JeUnMU zT4W@^TtrX>i-x)>9wbBE(@MX!yyRXYHt>BtScPM}G{b``7gUBHIZAiPJNdB&Fj8@e zU=o7HQN#2G404rvEVl-&>(3fAn=W_l!v{R3s(8}8_br>u3)UXTtv_kBVt-Dt>BzH# z_|~$~JyPsX-A#mm8*<>xRXe*+677z)lMHQoCPMFl3euJyrukEMpfklVp2Uz2&rfoX z#DISgbNl`0=#nVl4KY^cQp6yu#fTUu{S?=-J$WR@pW2{;t{=G9@DBpOozX@y`);)L?~RgC=mI()%0w8ti*_ z?1og5`dhNuJbT%o89-L=Fa8OyXJBkORH|X2?nyz$pHKEO(-Io*M%D*aZQKf`+p{a= zDsr`Etctx$B!5>y0}XsH`9i2N5!6_w)il2kn~RzzB!yY7M=x?=bDk? zOkUOxDt!>A>h|vy>r(z}w@dxqKjnnVj#nXy37a1eHE%cheS*5UufK3YxD}<y6Lt4V7gb|N!z6gUc_I0-M4 zbw08;Vvi3^)X^(h9KLd04rP?B-M|hEN~bn2Q#_ofLw?;~R@_eFC#TC=*?X4h{PFql zT=tVA`=LKS9>GahWtOZBiMGZI1%CdrEm!FbEfX`96L)(Hwu{(kHO{sNQzZEwb2I&C zHit3KW7@%%%eF?wUATr53r#qVD=&OgiHP5PSEY%YdHDW>_n155ak&La<2`+28yd$t zYV6FJ-NSH44Vh;R$@W%?lG?th14(j^MlU@G?iL+JGBX zof>F5W9)1&<9n`bL0 z!S2MS-uxQ^c_+YgSz+zXL5w-=Qpz5sW6oIfpGbfHP`m+a@#>ky^4K9`&r z@m7Hk9K8+62JECdG6;Acj;jlL@`XanllE>-$K-#O>dHfoiyv8s!`^ig+D?jLhCj*C zKt8(u#cp%s3_%c=KfIn+8foqd84Zmm%af{VNMK8qS-a$;+iOXd{r!Mh99`pTSjbR8 zSUgL*hF^=|uzjgKcEfQQ-Y7kCd)D~&YK`^n(c0WKIGizrWn>vjJ@tJ`Gcn^2FTTrQ z190Jkx=nsNg*XbIEx^Xjz0dy6up%Wxh_!uRKj?8!TyJjhJ0=YakTE>ENUzUzAGSg` z`-x)!d(kSBBRE>-X9ktc2gUfKdmk6l+vL3J^Hq-jSh6jHsvD9+f2?idU3>LMZ7?JG_W3{Mc5D4yc?4WsA72GH&*o=S{J<<#I!DvaNRh98SX6c| zcZ2361qQ^|pjjMH_DSy7U-_xU_STNRI6$9xis7mA?6{%c56g^4e zy_CDBFSG@<`eFaz57qoX>X!P5R~-q)Dl13za$m#8s5{!l?ZkQ>Xy|6}UXqls1vC1S zI3d>NV0Jj-_m+7tr*Uo@lNIr#JLP^2-sk;A6(FIq@Q?avbGm=|ZvM^NfEd$uxS#AQ zOR$gAXOBmT(Aw`&LoFXuuHl1ZD971<)Bv41=}Sy`qQ|3-b~zAIqcpK$AsQ-&y)m`| zYkgWvsyJR~ZfvKPa{oaQuBG(yliXLq?At1k?p#Vgv6Wb4N)?ajof}>OHb+$Cm1R3| z>#*LFT`0yUMqMGO(Cg@h0Qx;OW{DOq^)fUH>SczcLPENVJjDVYok7B2(k$Ua4Vr(U zxkN0(9FnkxK*l;rdcD|-x&slLt-oN!OA?z-X;Z;V!9NR=_{tk=PIhM^422IplS-gm zZni#S=*cb6)Fm@lzI`3l7JK@$^}$loZ?5Zu0d1oJ{=UOv;Wf|H#Ib3^v@~C)ad7`qB982MqW=H{-O1F2uINw}d(g zO7%UW-iA-L+5PvLa2q8ub0Nysm-_bvVvP#txMkucXgsR@Kp@koq`pd8g5Uv9&JK_lU0_tzXrF6M*pW zy$o*gBwB$aR5eO3cO*=!j==n^nZ{nw1nZ(F)$^gHtoU2-)Eem)%|)Juh4TAmoe4l; zaf{MTunBCpv%z#5YG)W>S63=@xj(gfN&?b+gfV+_Cc=?oYfp4s%l&2j$MaVw`mSv2 zNhd!Tg6ZA~T<%X&q4*a7DTDe-zn?%9afIcoKOBjOOtzdP!zZwut*%vlIe!fbI_F46 z5}|%cO@WNwDHg!q!v|ez`M!ok1CErFyCv|%ZkQs;6i=35(hn$pavZ_?<2`0>MZgiYC&^b8kD(9VAR#BfuBx60p!Q;uEsyQ? zz6Z4A=3uX-z^XL6(D*rZt-kd^SRlp^SP4!L9Tfa`itzr;LNZFD1o}?OYKaox^gVRjCC9uQ!m}dF1db=d<*2XeKx-3pHw7{J=^?B z+E*cWJF?cOw5d6_r2ns=1b;MljoLU)ik7=&DAd{n8PZ)VZt5rfojq$gxdf(N>Fw)B z>#mhQUr^cDlGgj;ysiz!lu&o)K9|su;N#JZW&j_LwqBOGef>ud_xvQtcOUVy+db{Y&&g(8w0a=TOuIe8T z+F1pjeQXLIqg=Q2w|4R#0~zQ_V3>ScDb4GmH>f!{7Qc8!w5f<#$gCeuw~&j9PL0u zBR>0X$V~JTkOHq z*i{{?gyY5pXUI?)_0;w{Y zGD4K=?y_M_+LQmBp0rHmtx;4ct~`1kL#lLy?KKw+*t9&w@@}yr|NY7FeeFz#HdIJh zbdC?5MJ5xM6ioPzU}p-WFWnchQOW)`=(D=f3*(Q=$+55^X-ZqPVJ`t7=zGO3OKT)O z@2g~Qp%a_T1MO8j`sqItp%X4W&Wz(If!+AKWeJIGc;p;9uAsl15doiG5f+B-w^wyy zW-scL1Q=M1$~~vvviYzO7YwQV??0lk12RnCbxVZ+wxk3u;p&5*#gqPCPHY1}S&_ZT z-uv}~`7(fu>kU}22esGU^z_0=j23e+Z<9-%>0y0-^oY(=rc;wvL9~uTlynz{S}r|0 zf8zKanwr`CpytIv*6GSvF5y>7MP9VT62pxJ3)JV;*f$>GEx79L`fw&!bIPweNTaRUzSZ1Ndb+y7poyB27Dpd6g==_ z3s9l)K39en!D9p2bC3RC{AWM-y=`tDJ+vZhu?Qm_%@d*vS_GZ{sxy#Sl3CnC_ZFjQ zd9eDW;|?yxaet)@O>Ik;oSWDi$W$X9IwQj+c1Xv%L~~WCz!w02Q{b?r>o@cAlf=TA zPwd>4HR6;PbzTAI_1HZvV!5EOoa~oHA$=3n0Y62KjX0XJ^bK_Z(MmLIZ9FVdQ?)w`(@~~--*8J3T>I-tNM)$gUnxaH;FjY zeHggpL+VR0%yk2QAIA{H(Q!o;9wlG3%E;bMc?rEL<|@8B=xzR$P^d}WT#xZ(JLXh8 zZ)>DdFLXeAzvpj>AlW}f`MM;dj!z?U%A$Ju%B-myH^@Tu9PXT{mDY2xZ6GO~@kng3 z8tkHO_89q!K*|xZsm`M=fq8#l?*k`gs9R?r*g)p|x{fe_#B+b^FaC@QVya(>^{JpT zK#Wwno_@gCmYWpi3QC1xE}U&MmBuNF@{ok)B}M6;faHdUdf5X=?oR2MiCZWQY+v$8QCjs&r~XPX z+#%WiRToMN^n15!{~I;n>(g;$i{3O9eMET!vf{t1dR@f_Ohy%o_R!Z0K>{z-+(o#i z-mJzYMp<)n|6IsRjQZ_CLytb=DOCB-r4!}&FF+^dU<|9x#Xs+1oStn8MLo>_?)2{FXSvL}09%HfhqPjjH}F5b zZuxncjEimxygBB?yu!Ys%0Jqh#t=;ssrnT7>IhpdV_8>Z{iY#ekTgW>tRe*}c|hr2 zvh6uQv%vYmwAWp;?6U(*r$FuV6Isd6zbr^sA7Q@}^}JOGlS&p+2p*k0AJtUCXN|X@6aOm+a=m=lt0dn@-^TCG9lXpp1u9sKt0kJEl2wRLz7n{<< zqLGrH6g|H@UM^9#4{v10O6z4{7{l^z-|F-8hnpev@7Xl(v4luSX1p!^>`xJym17a* zFS0`@!OhSqOdN-IwSDmg_>6r1J|3=-c;ZME%x1c8w)Z5?U*Zq(b%~D2twbhGJwj#V4MR01h>a=TN-xSl zwLY|?GHP`c5ZB;*0%GXj51Mx|GLc-#HHzc-Ce@8q1K-Pq`%J`pDhGLn+! z^Ob_xZBJj+{S;TGA(uE}u>g_RPY$t&4vgqj;;oPl4;n`#oE(HDC6FyiwocMMW_Tk4 z|FuwaZ!_UPX0B72Da)ww=mzac7KHrI*;#h|7yOP{r{6Q=c?mi@-JZNa!pVdx5%7+W zJ)ge(V==Okg*0VLxU)1c%jG_x7|cnYP&a!Rg^cmB zeUFy(bNx9m-Sn%&Eu&U=YO^rpb)uR59NU=1qiaT2-J>AVlNNk5zMFP+E^(_euXAAr zu?1y=#jptwXWfD94`=`;*`35kXg9L%brgR} z%L{vRnG+v~G&{DYdVVrMXfqS?y;QxxtS;E`a$U&qLIkRmBx(SbVdyip~`nX zLcGqdQYv5TpH$WE8;?l37|Zs%jb7OnMac0@8FK^K$nJuqs4(iFR z2ON{@ha7Px8qVU7lgBZ=zX0UOHVb-*TBiL&UK8mkVLw zW@Yr*VOGG#U<_be98v(v(^Os6AOkIq$Ms5jL%wv=NByEnvZ2P{W9yx%WSx8Xk60hs zz_pM!V4VE*wC2fAl|7auHH(s{y%L-$KyK_aTEQmh#wh};4#|e<6v6cB+_*Ucnb9Bb zy6T5Oi2*rgh_$~}w9Xb#znU2JDuL5^0zCD3jG3f$Y<(Qm?&iDf_IJ)Hm&08EvPqT& z!T{t^-hB(ySi_etvVAN7#kw31XerkN_snN1U=6K2291Hzyli}@)J6@g$Bagml~U$Y zI_Fgl^)0C*mPCmjHij4nXKo_S-t-Sx+XWlD1?&Ip)wXXY>RXd%bfOGpYu&G(KlPEL zA!G5^&I1M(bst07xWO{0-yLTWCDN6mO$wQrj~0I)C$_|L$6oYRx^c+!82Sizn$qNV z_1o#5pDbDFB`YggR0ALk*TUuyjjRX7Wxr<82dGrS*RYm;%Kp-J6F;LypRb9AtASlN zNmX8|9zA}olHW~T2oJxv-#l`zzSp*goB8g%cQxv@vuea6N;5=*!8wC(Lnfh<^Q^=v zw-ORAVY*!dpbg)LEu-!k%MM_TPKkyf;8{55Pxp#>7>r|+XL$-y$880J;>05N`BCK{ zW$(l9*GLCHQw02X=IY|`<~lhP^tadJqOQbB==-}5$7+>r`fc1W82i^DUjhrTzIlYZ zPdR8i`a)50t;A|usIve!aTBbsMfxJVh2P3G;f3u%c_P$p$s&sJ_hu+k%Tt?epOoS6 z621xIG~yeOpa|Vd-@chwXI;-dsrgxXy`bULYb={ap39k0<)wty0!@$of|(ZvBlaq0 zSE8Evao#<0OYuGHP_9*}_}O+{K{jgcX*?mkGVhc=rspbs;kUL3p)$2GPuxO%_9Gl) zuc=k3)QPn@{2ut~yYJadfDXBH$BUnIrR2tK6-+jWdNC%Va_RHOyr*Z+n)y90Z{`Sk z?Q^g>2O#Whtav~~YRkPVg2K)Z?#w13tBVU{3S@Xy@7BpRMPH7u{p)rggk&kDJc*5e zdKp!`)xn(zrCa5!n0Gy|d5x?Ajn3tYe;$}0>RN*0| z8eMgtQQ@sg%820|10vg}n{*tOa5nY|tJMjU#t_eXcQ?GN?YSST5efB(w|*D>d;lVC zUafUYCqAMu1X=0mt(tSK)G6d&sK#EeSu?mP;@Ak<(=6rwAm6ni>>7KDp5#O)(Wg}Q z;A=Q_`{N=ZX9Uz2c9?QO)0-*28DnpJ)7)^Y3^WzD(>jZ#5dqvP5(?tnz6D_=j{cdU zH4`MT-aW+I1Cp(S)EI5pZsiUO26TaI!goBtoY ztU-;9&h4q8GvU)ek0n@QsfqXq=*jO#eZ*ZxeBJq^&6TfTj6bCkU9ja@y~}Ii_uB%m zMpdN_^E7YOObf+Oubj*~UHidY2P9F^mme-_NZG`-Bt_-lc|+)lEbJCG14F=D(6Iii zt=0k_@81J7Xj-CNL~B8#3do@!@&bZ2ug7U8I&6!|OnPTEPvAL*H9EXMY!HMIn3>;y zM_`Z$4dZaGBD+VsnD#ZRGIbJ>Mqn+GSCu+*A~~H1{jFz4`f z3Q1*G!*AnDv3=KPUfRCAHS-h&cbuND+=(;UVSR+1p`F8IxgS%pG#K_ovYnb%< z6G_eNB6nS7RZ?L0Zf_2y2FM)N_R?$Y51N-c%X7P>RPI1_#CJbpM}#HuMV&1T#--PL z;>y_3-URx5fX6b+9A}x&kAhqIwOY}@Unb5rp_6%nZTM+|@&*fg0!d0$>eVbuX}5D>27_jq+OA6P$Rjf^?bp8-3|Ea%!RaN6<8S6=5; z?)X$NH}32MG#n$d$_hb2Ra15k-IEF*^=*}nVO>-beE_c>lt&}AW;zG_6Qohw`(e+M zMBer3IEvZYC&Ex7)4`0MWQfKhBYy>_^W?`?<&N=I0OHwp<&G~{Soy}}nD{Bnm#KG( zNJx?+xhxMlS^m@F`xPKc?d04f>}0-CEDUnbuM|IGA^38oA!+?yv=T7SxkEZO9Olr; z*F|OZ@4A$MdIp}W6V^ng>vSm$CWIV;S2Zt)QG2v2^8;jUYg$;AVg#J6VwX6EB)7M= zSC`DC$&g;lAMVML;`>m zx}&pAb4j`(cx4%-xNpXHv*1kt8^g|iStW;n+K8mr6h62=&TQeI&#SKOEq*OWhbdPm zd~A>r!l=*VN1+#JsfO!EF;DjGz?uh~qA`S?EOk~YYQS8a79X15Ft!tgVq+GbSVV#6 z4Sg3l5nw-x3E|t6KhMTdb%qkFkKB`*pKuOTFvGhwIC)Wy+YhD)uyjQ+pRjyr+9wsv zZ6zNvfo*KrJezn8^Xm(6_OT%7ajw-VWT9GSy&H{r755%shyA1LHS;~>5mGT4hep~| zVT?6W)~$Yt8#*Q9=5dT7ns4oO-xrTX31q?NtCs9CPN+zu`>?An#j4zc6ql}@LnMd| zM9uU+e&^LqGc=DAzTDAs!@9z3>z-$*Ve-g8=WXjjd=;P@H?$xUgiB+_+VLq-q)((tXTL_%Y;8iy#>4`J9g^M)y@WoXzSZQJ;kRDSLj{+kc1 z)wA>}l!w-|il7~jRh!|tvue+%zfMA4CpBpfw9WShG&9YSUPWu&O&hz|_rq7zm~v+j z?-2-%V75(SO|&M6hF;Q(9aqFmrVzJ~?cZemAt@^(&3QdyYI1Ml5 z=ez&$%*zelWI1!1E@BP4jP2{|v5s$09_@G56q4j|;lKg5KRga1zRsd3 zU)@cn_Ro?J?LAHM65}XpmM$py%#!8>&I8fdb-geFblMZ+(-kz;x=PSHjRc}syR-EU z)WXkBc57a+iTm8$|MfbF)k5=a46|WBoTarDfawHeZObaOaj+pN^37J zbPSN{M7g=3;X9^vRXG4G>&{!j92y0PkC3}*tiraPS_RT@4L!Y$E5E{Ot!~PtW=oXC zF0JIY9=a#7KG1o2l~`fh@~X1lm+1`h9;Fls4ax(4oV7m!7niKQW5d~o{-g@duu#F> zg_0>9eE`EOqZX1``Y&2ldLbQ>)W=)9=O8)7Xg#Xp8MXcYk4H=khpIYbb+P^r@>k4+ zL729t-yh8Umn(~$_gB<6-%za0;WydbRQ<`@BMTV|Fk6_SPGV8fx57*}2d?XlMfY(n zsDB!%kW24qJeD`ndcB#i%+Z&Uj7t;B>A5AitN>;t_#Rj97wvS@6sUA52@nTpBto;X z6=&`zmlN3?h7D8#qWa=}3r}PMWtAUD_Hi*1)R5YL!e`Y|(@Y##c=FhXt&dw*Ebj3+ z07qF)^&g6tkGY1A-{F=pYu4COaoK%3^|4%)2|O_&)Zt9ErRbcdQ;0HE>#kSEGwOOn zamTbRNco(avW?%#xla=ni{}$W- zw4coU=IJS**YHu2F^|dL?*9(nMzG5SmLsxRX!Hfy9(BR`_1yD4E`IMg{62l5UoRHj z<*I9SdFlVOp)t_Py@V6=wlE+HN~Ym?`eF2RJww?X$UxOwzy3*Ba58e+ew?QRoS+aKL|RS2^GG#uSj4Gd*aCtG6a`>Rzg-uHekhy@q2XzXV2Z z+zdX~H9IC7J=5r2^CptKaC^}qxy07PcB+DZuF821OC!z`AxM%uZztztc&X1ZuyB$Q z^vTkfZiRO=%>mX-{RAJrFsjWI{ zZ$&s1cR6p<%uSDDb&rAt0dI818+(avbheXJhm|t7@P0;Dx1#3BE@rAx#MVHB`+^C$ zIQmlAhJt}0&A06t-*Bk%Tv`^r6D>iQJb+tWpnE(7`6mA`@(4*~B|kkyEedsxzD(^f zWWqk2o;n<5-@(C#KO#lyC$K`d7EXvnFG0-~Yr;5ViO`|-lqPOs6mdfmuP9G*Lt4jK zM@cFLM#Y7P1<<_p%s*ktf2Z}rl|FwzErhB&qeTQu0M(fjufAf3n z)A`kX_~h2{WO#_>yEBG6|81?bzQ?JTn82~oJ{`%cXZpXUEOWyru+Kg}^Ik8`#(o0d z9i;jpDPwRuTt?&*PPTlL`~~G^pP%o(XYW6$5bwf~*0&fLz%FWY_=rdR<9ZMe`6PYA z`26&=1qS_&o8Gg)w#go^w{ElEgPz=I-uZ2TI-8aWyx#nnn%Xgj(u}a8lsuo`=?m~k zyN0iI9Q2@*lkb}zJ=QnsYcx}c?!@+ipXNap)fV5kv0s9$1j&|_>6Iq?Zx%NEgl}J_ zfi5-FhTR3_>0~+`;T>o9bMCDNkRZc9{_Kjh0bu3E5-McDB!`Km1|oT}AFzUN_g(cl zBQawiFc+T#ogXc|*a#p|nY({WCupyUU6y4GSBz%YNF_%k#er7Oci2s|zLrCiU@FQQ z9DR(*?gsSPx|c zvHs(|hisjJeS{@{_pSYv8u^d&S0b|{Q_}o36nDP_Oiv|*wOi$zv|I_Rt+HM_dG^K*xm>96`s=6J_PeC%b**>1<}xh$%pJz<>~3A?i>0y%*I2 z>8dvYENUu7m z24i%J2q1ZyR}Dy8NZ=!o6rRiy9-|lyPaGg0xG^cd(h=gkF&XWuspjb0Q9nIuz|3JE zNAg~R)@9!_Y5NIsNaK+MQ=e35e*sdeD#(oC zJmAMTmQfiq&|;WloB%iA_PZU#*6#b$Z^(Tvs;y_A)&kTJub`tYNlpjW6uBb@+M?YBlXq;}>w zZSD2iO13;!P6fd4H&W;wH>pv}T>- zaS{V(?|)V#?cT|Mp0RjsGS}WjUPzU7S&`o1A`n>bEjEodL>s#mE^6Q7K7)u#UHYwdWN-z_@?aauq4phW$jJ$6;rvhR@%1M9$WTDl z&KZ6cG)v2}7s^e!2G?~%`pi@BAQ$}1D#MWDJkD&z=rTx|xK*ye38Ulg?zCn(HHqih z{tpD-qvMZ-7TO2VzZ8(|((2x4`v;Pg{({ZFa+}@1`I>FR^+7AJI1g7&0HebFY7+SA zf+JFPa=i3OA(erxo}}CQ_YcSY@^t=sh+D0*fx76@MQmf4)BsAz)^EEhgY(sAxlcDeK8QUsWpyXpW z2C)j`(sox>roV3*9BMPy+O3oTNGA%6-a{O$9Nvw<9V4I*vi=N5c&1_<$EmUDF4y>A zqx@(TGlwcFU?C0U3_u9~dU}b6?NJ2L_*l}ky{f%=U;@!m{e)4@m=ya%brjYt-1Cx$ z4U>`<9-GJc`vC$x_!$}s+p8@*#I#zZN1IYpPS$LnQEI~wCku|+!G73OxzdnH@+V&) zW+<>;q{c@>LtMScE-@B(&#*Q&Nbyfa5;jn^kK+u|*7377;&aku0XYn7n;UJJm_`S? zBxc9b^+v^}h2OAeztP3Z<B8O><@c+oG0=d`3&t;X|BYHfl4TDI%#_7|Zay^T*e)~i}Qe;LW zCMrF;^=52HS{oMi!|aDPOcX2@4Zt+!d84PgS-DR|B3s@xdnG#47J-Jgdhu4b+5Ow1) z*uY>w+-IZt?q$lqrK-1z)zE^WtEEx&BxMNP4#+NyI-bGE9$05+k+AmNjo35dfiRvA zt?nTwGa%al*)elPUz?(JN&-^0YqBfAOT0@gqWc4MVDJZ8|GjS*3!e(u%4$oEP>}Td z&L-cM@MAuVz~q<7=f#2Q!mA^hJ@~1I*}rZ zF7)fFnlhx{*Z92zGv8qL7`RJ-l*bNbbvnTGeMm5om5ud;CrFE&K?vW=R9_IDL7*&L z#X@J0s%Iy@X1mav8|6O0WVRwwP2abk%iXp&M#04tf}*j#!|gnnJtD3uY3(WBldazy zkw|^3G&PJsIwjXGv=SN*dAJMxZ0%ebxeI_2iRHDY?vE@8qV;PvY~C>1k6z1^eNAZK zS6Y|25Wwy^cKD@?VV^zx1HwKSu+ChMRuOWYUeKe#w1o#Gp4X{lg|(DIeaY|u7=7I@ zSk%OKh%7my{liC3_IOEesPe^Y;lRKXkDvgC3WHn1eu;gYzY1C$1aMqr9T|(*87EU_ zz*YRO$@p(8(+tRr6muwJQ{C@AG^8pQI=8McXAuona05a_UndY#_Y z-j;K7Ab9uN=tBULq>GGzT-vtUb&lY{j-A5yrlXJ!za6YB)DNXhKNPG(v&|P$TF5w| znA{ddw*k9@lrLHNgM@2Y(47a7uFN8Yo$bEdeqSsDxo7Aa7<_%UdDz1#4Z2Oz)~ zkewzN1?*%F1isLb9a>l)1OO&0W^*Zvs{t5b97^fPE+jMl82B5;D^hY(%f$D3j0h~$ zGSMc#zudGMH`TMJ{YR*h!&iUtNRvK|3Ga*WXkjw3gwE1;mm(SiPG0db5q^^-)(MQa zTXSq}OcGyE%x4yjKH}T2`Gmu`>KEk)zj}#plW3Qlwj1K(Xe{PzfAtZMtZeDIUr%P; zZ*^4K=(TA_(4wU@DAlH#TEGx6ETZxJLcxTP8!5xXX#cHu4`}*Lzn$ZO-2iQ*F?ucr z7X((|qVEp*S$JfbnEP*#1Mg!qp7lWa;!Q08Ki+tWyii!!mh~3j?iMaLWyArqr8~N;qy*|=xWtM>e826w{2Ejs1mb9 zB8sBrjBbxKOj2TR$?>nmYaBBqRJ+rDc`fs^aq_=m%d9_{bROs<3x%`MCbvpx9I4xTN66{9v}qV2cNQ^4b*`8!iNu=CYFP;ymeja)+G7ozFo(2Qxl{j zDC#g&DS((Xp_BHwh$kU1J~$C^U{tuxVbGbPNRKTg)j$Oh>3jN}9qo;%@ICOO3V%0- zy-eg}w6~f9smx(TK;^Bh|a9hvcxJab+F1>IN z&=cP+n=aM{Ze_B{WlTukyyEpTtFk(VxEvjK80>I7EqQ^CR#~_=1AOS8dK$s7KxA z^Oue-FV%?G+StZrbR|%UjT1irY)&X$x?4j3@}8juUDNM^4p?;?&vWF|2;a7Y+$IrTQ! zx^O#veTNHR~(s4_2W9?c%riW>dq&1~TrGO=;?uQt_gQAs-8r z;04V}{nU7%Qmf?IbP4l&phT4Bi%Y$$Xu2-t8@h638dV-oajV37T~F<1p?`77pOp5j zEhWo5Bn|f4V}^yc8dq`gOhdtaftO@#n(70Am--?++}jGb!mf^{XGhYRJ8_=dKWL?G zzqLLZA9?>1V}=`8G!vLsXUtl>aa%x?6ItLJ_@%5{*q>grl>DwYZu%yv=~Rh{gZo%^3CCT-Irt_2s= zUTwZcpB+^nb=<+p#gW}PPPY!_ZOHscd3SK(iT;bJMPiyqF}L?9K+g6J#GCesEiww2 z-r?+{?QRaPxun%S?@{`ufSU)8yuIPJ^X;hH!QHoS7=l~mv+*|wu^w>a zj<*x0M<$aeRj;xK!|qnY>z|k&J&!|`o6A^j5WHz{%xKuNYl;JCNeVz{38R}35^PAb zj$^~5n`(!64ssGidt^O=+AEVr4L-sMv5~x>&mQE^P}VX%`V&C>uu>Q0$g{4Z%P4_S z33tlr1HN@7Z^Ee`RgE%e3Obn_{5zp#EWOWrQj$Jlhv52y52&+;b}RUdp^Vl@$H0u6 zdydgYIGg!&_Y?j`q68Y_yPFzpQV$}boY_g7p3%@CFusP3SfG9ML0_E61l*$ z?ldAC22oLG4`2jSzYy?Ki}ivh<8m*7c~FjGJuU>VI|ix2u}*Qf_wvTGJdKS)6r!5N zIDje$q_0TYx!NZ8a`C{#5$levRK?Vdi-Zcl4{8x$!Z3TM^HgK>vJ8eMAt^u&1@_kp zu*O~AHob}%A2fx=pFI4v7ge`q7zAltJfm-&pyXZ}9OiNnzqn@O#gtb7*NRs}lC}*< z&~lF{Ig*MAqhy+HnTS#fd9Drj2HfP^w>~>cdqOCsq_)ZO>Qy?;oWWY$W$`fh-#j;= z_m)?IALFUauMdAP#0l^MDMR#7#|tLevw))Lcmj8=PRRID^1k0qs#Q(+8>E zMmP&gxE#EA9B<~bK){aO6i9zaX#xvrOz1B9UUJ1e&m?S=RlkVhZVu$IK4f+YOiJq< z-E#PL_rTst$bmg~?{el!p6-7S5o%h6Wn=1Wn;r$puHhcNJi z3q9RG-b3=)H8lfvRIN!Zzrx}|C(@rpUl}~=z8JO5{TeD`oA<6F{*lTt5LpP|1hypT z{bKBGh@FTOp$(=y#PvFgCjhjA3q zD6MjLf`KbgB(&KlAWZpvWop`e;jmBwA^N}s`=nLFUwf;x;MRKvnG%>0@-RM$Z7>A5 zJFdTlK^^`{xClt;SH;&UKY}a5pkF4Lo8WXT^L7xf@|D6XsLszR57u1#7i(^+H=`|J%v6`uT3 zIX_xoGJH)VV9=bcJH!@G!rk+k2q)CoQ~c$yqqrrbsGqo$XN(L7f)Kk(wC^a8@I?W4 zuln}h-#65jJhSP#aND=;TAL3)Eb~w=doWB>We7$7P=Rm3HxR6z4AZ}cP#XD9{tt6s z`4`0(zO5kL(%sSxN(hT|C`c#*l9B?_9ZPpeBej&El9JNWDN88bwRFb<%g!_3&+iX- zUOn+ zuBS$?V9Jk7b!z+8@N|wxfnXhP5jqw@hS6MQkAN|6uiEG!$7tY3N1xG_$qj7PFju)g z+L=M;^t6A@B7CI2Z{8kGL%xILswUy6jRk^UJ(AkJC43YdatWVc|+iH-_-S_Jj% zKaTTZ4`AFi+iC}wyMj6J)4mug>r>%emkD&pcDaTkX(TIW;KqQ{iD9q~d1jed9q291@Bg|bAeSITEjr7g3k5lBNgi(iF@T&wTJf#H6 zS!!HhYxC1&=4*02)>|?DL1^ULhh8FVMM8QQtC^?Gv!+E>Q@KMCs3fXzXp!)*xCgaCQx#f4~zDoo*^Iw%=dn zYtFnD2!;i2PA!(KpW|ENO?P%k$P3Bvq8l})`Ptg!i`JVWf`)t?g^ta#kjxQZC8j%v zse5EJA`H`MfcJ_loVNy|7(D8yNZO?Jv~lXB;4?m8?yGFE9Af@?yE}~aTFRfk__UDI zm_c3A(uHcHQ!vo74{}$w4rP4aeK&xuVb1{5ZTAv%};Ri~MhAn|ab(LsJ zN&u5Q9IVkP<&5xFgPrrr`6&V2#fT39364KgNHQ*~)BiXj5%^y!zQS+Rzku~3NjETR zZ@Q5FTZ76Vcli(ZBlqHDR$uMnKJ@(SA)N2++rgiIKK>3EzN?Kba(30HFX&dA>}*si zd;L7-1X4#!2J(AF<1!yFr~~si{Qq#>-+_^vDXT;G4h7HgMFhW~3gx0YaI3u}lt3z- ze<{RF{>~?^*v`}WBt;HkZ^iQXeqSo_{9LlH2PcWiO@!*>1fMR}+`1i_y#t=PZt1AC zOJCb3&HrGu7eM{|hMzB60zE{ZZB>lk8n~os7{-S}x#N;As=vr#$61NIy?)3DjH-M- z{{Ki3p(d@3foJDBKQd?gO!HO-J^Ur>qkd8i)#3Pn9sPr1+g-slL zrc+$8Mj?sGsdb8O@xlotV!R2sa+L@J~ZrJ4@}?P4W>Q0HRzpv ztqdZ8BY zF_W?GnsvkiQe%zP)b-7~asFp%Y9ZwgX94%LD@Betzb8y8xqUv?s#muL!%|)%)2AVW zVh5J*x5oi7>4Zfyc^M2ECZlC`_;i7fFT7oH|8 zIQLSDc=F4^#U#!@3tTcOkV>LRuexYw$F1 zterdYYq9AF9OO5{9@gSDY3)4arOiBH3zapXO?ypl(Fe9Ik zDke@;9f8nzgc%+X?>lxKdeE8bK3+}yOZ4*^70mCbV2x2}SZW=n=|k{^Ym{Q=W;!NSK$SVadbon#US<$2iX}`!MH;(KnnAE>Jxy*8FzG_{z>|&qb*v0q!`{0pvvZzN2v2Z zZRh<&o>2ePDGuasX9J}P&Ay=W>h?f1x->6)`=Wl|WT3Ap1g zZu^7q`PXj5Hmv}$AA0Rwu7^1A%;sEg-e0#B368&3$x5nfVztq+pH zooYIcx|T3*1%y5?yv=nxRYqZ&|GG8MsZMjFevt3-@ijhznd;#VsAV#*r96n~7X6oa z!lBz>5&Do+^oAh)gK&ThQLqYg65-{)PrHE`t3+>8_56+NR&4k2Sd#e3rJxL!pes(! zN3RQK$cTDuJ%?^RHr;OFMT!FFr!$FXB*y+oj4ggg$*hXg477*SW(m&k106RX1+wy_`hsbZSQh6(&T}GZ(Mj(=e6gS3QSe^|v*+jDr z|B`+_gEVN&0mNn>88-u#7aycBrVs$jlka0l9FZ6S;08;sKSP`irH93Q?Bq>~-p%Ca zfP*p$H)GaqZ$u!{fYbmRBr2>>X5qM}_C6JPA`nTfkPdW9A%z0ksEOE78_a_B37f#L zVCp2(`)|PdJX!DruJJFx(hdDUc*PtgMiKIwpbHXccl8)u#%#(u1aXW@y3UJ+9yP8o zeZT>budm+qqk2MpOiKD(qGi#xvfnQZn`m?akiA@4nT%Q#y-Pz^j*lioBluj2nZ!`e zYs|OwdpA+|b`PZ-MB)?IJ5?0GRGGm^PL3j~?u8cl8q`IXzN9;iWKPPb57%Z0Dwjy3 zen#Sr42EqogRSwhjR^mNzW1PS^Z!p@XuFVJ)!qsjdcQb+S%pzLSpuZkHWD8?hgD$eNe(vwfQGpCY^wp)D zEgVGnK`;VDd=u3}`4n9i7ECY%;U_ebIRbdW7(kZWJ*p7Hr}cEZciQ}UE*UVs-D}Fe zgS0J0A&PQ00{U(smBUY7$JNzBb;xt)n)_~&bpBD)1iV0L<1W9Lgp#?xk!w4F;D$>A zi1K<2o5@aYfD6OUeSNYJd3K8I%$IG8K$bP82m+tyC_y#@Hnwg8{TdBefI2it?e9Wb zfO+$xc_i5dI_?aY^BwR*Zg629_&Qu{Y%umPd(C0~lhMO+{hiDaxLbJxM&~O$yGK-B z#9k{R1U~NT1739_Jio`f^tNPT)HrW5`8d`l>VQ^~bB=nU=%&AmyuOlvD%Lu5?D<|G zA%sK*Cv|52RtP)xpA;rqYI6CpdtOi-qs)Uvxpp3ZL_s?W;dY~|yD|W9)%eOycalDF ziL=Os;c0+}@7iQT!oZY!wM%7DF1vh%LlG5XYd!@Ve?3W+)81V{S-xlZX7${M_otg_ z;{^ZsJE=|yO$vQkVBo>bslEcL_?Pbs4HxPyV@{vnPWXdQ5aykT{nql=65E1nuws?y z4HRs{eq?j7Iq&7LuTJp8irnLmINy07e5SiM%mZLWO!YdlL}y5fW&$MYk5B|F^Xr*^ z-&6Z8pGWGCLHOq=M3@oG#4=p24Qq93QV`s6Y_pzW-(z;E4d`+)$`-R^{tULTmQeTo zc=wN@Sw4UHP7e7sNF8x7w2_Hbq<5uObY37h3iv$GujfUOoZ?4ByX9Mj(-!3dR-ty2 zom$R2yxxhMz^eEav&>Bt%@46&hNj`M}U7LqKBF=S_&gUUCPZ>FkP*LI2ENTo2#}8Cu<`5n^+>G&K+E`4*w4 zEiD+mz)L)T^%3=-2MldM6CS9v^yENtoq^B;X&fl6a~fJ`2tt7{f2UY{#E`p1A`OU@ zJbWh!oHUXBw{dG;%PnjxsejCBkj8;1JBlUcg)NXSfDAta{EyTtpodJcWyzb6>pUV5 z_P&dhqs5$aY0@1SFPWHa5lUcNTLj^uJm z4n(jJ9s8lr?Lgog!*i4b2CElwM#T;cyvs^(YXuZJ#cf^SsCopfgBA$$B@rSq8JzP> zoxgwpS``kdSfX$%fhPF9?9Sgn6WrYzyb`^&UV@(5f5tdq$aR&~#R_!4Q01NcjOzJ! zJiGofhrE*p)TddFp}$`DL#zYkXFB`AxB+6d)=0dc0BvvZ=HXFi7C=E_pK#ZSeV=g5 zo>L43sXn)RHGTwSpcD=MONZAdf&gVOX_>J*K6S5xu0+b0m zIe{8~&TIEhTkI0r?n*peTJUBF@?L`a;?G2KI;A;c^@=y}P!h*mhpBjzIf&*FxbBpy zLj4fx8UFeWh5%XJhAn2caIHouoQ2-Cjl8@t^*l)J1_OU2kd+Vp^_Q6+TrE*d5^%Ll zm!Hbkxz}(E$q=I;nwX+8I0lonCFc-a(yu`IzFYw8Yc5oBA#G0TSJoKI5QH#?M&<)cqpBl%Oci24I5 z0ssFCUmfQZAMvqkia#(tdAbOHY_Nzkx(L_(UY=0vNv5LfOpCMAebedfW7TDEQ-I%M zm0)WWqvbd!V;L&-%6nq4)(9zm&t*(dA2q5~XMLQ)mvxxH@7YmvS?7Zb8%{g4HmhlK zmy`O%$!I$?kHuy?IYT;pMe*Bt4zPORIRN8N3IVz`AuW|J&BDFP)*_8U{LAt!;r4uWY zVFt~M%!-_$12?yZ^~ZOT&)(I0E?qT;zpTKYf5H@LWz`%pHSRcmXPBRE;4##qKuw9i zHmDmI*mSQj@adP|?CTyDYCCnzna@K)ybb{&7s-8;^*04St@5rZO*K$seWez=-H#~g(ndY z(9BmIV$Xln?)$;MuqtwJh!loX6-9FBDjo~Jl~&r74ZWc9i~8V8q4RK8R)_TSvOlfz zE++b8w=yvVf#_4y4sV7>-(x}zs_T6Ab_ktb+Rr`t6xv zP;K&yxyI9y174VQwNY+?W87&*9qpsa+5_L$q%$p+a}K}iAJ4SZ91xk@|BTRxwXSx| z4Aanld&9p>5mPz9Z73eQp`dDb^_ZTNXJlD&+3$628h=S6Q&??-u9xtgMlI72d{nG1 z#To%EnUo(&X?kQpry{rXZ8csaqqY$By#ozM?v8`n^Ba{wb3>BU2TD~o??rN3WtvP5g9@eyf#_z8BMQgoIdQK`>iHMZ{`3aNp-}#1Q?Nq>#C1~@>^1~Kt)Q&-hLW`f zOvM!ThVPbx2zr?84R_ovWO|r54Iu^(CI-u<2FWzQheo`d&h)z$(IYd)xC*0-FCZH7 z3x7PS%4K-P78)IQZT`B^>X5n`z4KRARdXLUm?0@}3sV*{44p8f!5uC9mmoNzefB7;hrE$y&&YTwCf74{)7QU$u=+Lay%7GitCs|I^o-CK`g@;)m;~ISbx}-MEq&!%2AhGodSAYOQ z^(GJBaI`GvEpJRM5ummv^6n~jmM2tGXcZd~EqRql7y`NjCcEzN9$Dr(sm23q z`i)>Kft$8e6gP3&VfQs9zWAAMjP!}bZ^%;8nZyrk3p27<@}b+VygH5A#=FdlrzZbR zWt80xu~6il4Yew((0nJR!QcuA$NmE0pvI-I>EUTR z-yU@+V5pPTTZ?!K+uD4@b^=ihLb!LxTNi>YN_8K_7dgsuTd)mi|3Iyt&*NSh45QXCUKq)jZieEm{cZp4V5R)8Ro)`1Alm(dQTb=XQs$%tUs z`y+~E4r+MfTRK>#nF?QMekc_VNd4>x@oP9*!a-Z)uJ_l@B!-+3tN5q89PbrqT(ehG#zm=Tv&vlM8mhNB zwQFbSM=6Uu@s8Sd{@$0a*z==7o6 ztTSj=TiS@qCf)PHoDA2t%D>vzfmsn+P7l$(gV%EpQRlOIUf?tyQF$9)RcX0&uo#wS zL{xuAFM8Vs_ZLV_Hw$VxhD-i%jgn zYzg%2+&wuXLaGwG#fa(?uXZse_#!#G>1+G}B$Eff{qh!sW@*o3~y)EvkfuKHo^UN-+W&5ChQ$che)05^pXw!(D? z^rPFZCvX+f!>5wO)YthXB$DZ7+BRF&B~{}^Vo880TFzI~`wD-O7O>0Sh$E%)vfna-cJ=<)cF_CEe!=hOR>zP z9Q%gXHxHF#tYov8Nsq)zj#sAf3jZM9x?gmFr6 zEax_r>O3($)N3(Dq<~}B)P0N({j)Htj*JC zN|0$-eJsUg(1dr@rO{hsUn_@-`1X>IPkCBZ&N*3y;74;8WUJ-7v(LwZs>(g9E&Aoo zulAUq{8Q`LJnF~gH_tVg*PEXbx7@a-TlyS>0AqOi;U^?)ujik^bI9J8C(Pq$M z-UY_ZTzzdSbTToyX_wo6lVElNmEKv|6mn?JmOvwhE&gb9aQ^7^{D+vyv)ql1IA80I zY`&?K>G)f_ys>u)iQVNhL1`Ok@A+XO5uwb=-L==NwHTW5Pz-kFCn)6)=NXsLIf8fZ z)i$>JHikXq3f}B6A5$WR5he?NQq=x{TuFk^RI;K%sME>IVMK#p#3j=jDe6O+x9+br!FokK|tfLHTER#;c?1QMUT2r6>N{BWpNrH%##61E`~qw%mY+FIz$Hd3H% zgC?VW>~iSz)b!^;!y2at>5WIu@OllN!Hvh;ScT)#Rai0>wz~u(65#3chYPR5?EbYj zyVlh6Ki`DNVL+n_tqX;ve2$y_Fa!v+oji_1%B@2t5{D78-@o-z4I=K}7PH#f-DoxN ze;ymS>nB@<4*N{@SC7{UXE8fS)${w5a~7Ovdlx++9nqI_*1}(}#6gxR#JnYZMu@g$ z-yi}b+=^R5W_YM1NDtF0>_Gg&CGWoKI6|nR75bek$8KB5Ov=0Ki34l z7sxq$bY&F-_^!e*07{`!$lNIyeqFOSy?dL9aIG@pywD~CbPx@b4F1+!bsRatRvCKns;Zu%|*5>`V>f8`eT(?V5p1KPvXp%4}|xi z#D00xk7(|g84%B_4GhMzpZ`^1N5RKjBclDljos;|>vRH%vbjl}sGWQNrf9hKS?G?a z89O`Q(Ex1Qp#Zuh$EcAkc+YoMscnKv%H;jBd}n0z6q9>GXXMq04W4%P1C-v+%s9Ze zBEOu@a3bNGPUNjYt+h~;sMC82)KpNEN+apyL@n*DoU^RmT&~gQCZ>AFPfok*&}pwX zl`yenxI@`V=0qownofXVljdXLYy^#=E6jCCyq+I5^6Qi8*&zP+<#Yg1-9(91bml8N zIP{JV4U@jFjfR$3vqE-{jHqfQYuPk!+9GUzsFR-nx#M;uW_3&`cp-Mliky%D0cW#M zXN|4y&#NdEms2L0YEzFMrz9M1Du)FsT0;wM4jgtw>nc=5L?JCBLPh3deEoPH{SY!L z*VP#h{k{5NYq<8B-nm5(jfgX^=Wm6L??8Hs;OoxH<;2nc!H*HfsAKpZyJy1(ss&(mRH;2+Feo~bY z5!Zl+hJD{M(&Es~XlExR?E;WO6v1{oX3oC4Nbs%Z2})x@X>Y*j+&JgK8FLzpD9Gw$ zOw@kn%GZo9pR64(TYB4()D3~11gbp3MGyTM;b1L*t-}0Rbyh*%c!blx6^m@_S5pP{ zbmFhVO6|wRiLy&yOo`k5e|Zvs3KOw0WPPjQx{0-uKAReg7C@N3M|FQPjzkw6Vmq3KhTog}b{7?HKH@!9i(H8S14XVX>TVz8z4F}k+77CrX*9Rr3P@uVoTK%GQ?Zt zYk0rPA4x2Hse}o*kUfQ~Sp0?%(ECk#tiq`HP`IUCgL%b&eWx$KsRp~ZD=h%+)t{Y5 z&+6JnlP!tSTs17aN8KT1EhpNZoaT%sR+k@j@?f%1*nLQc$i6e#Wf*Cx(z*Pk$=Yvk zq4PjcJ24{e7((qj{y@l{wfLtydt?dCJQhR>qt`a>3sHfVUh&fsVF0VPcI+dx@1+DJ z_5qq_MJg-zH)Nyjuc&?PBHyBK4BR3K^PCSyihgu@=ZT#?Wuif1w08CSDwIkh#l?IL zYE)$z{|Jp}(*B3JcNL{7f8gpXPfbF3=GlLa*}_FKIJeA$EMt+Xg0gq+(?vDTsYV_2cBkrJ7#lpvP0j*Ux57jw0J&T~EOA*i za7=Dbn>XLa~mMA&X2W zJfh=7QIe^;z$JP+j_tWUnbP^&a{cd}wLPDbf(P%OdF};SdTPb;J)HYU+Z`gIVa}qD zkL3FN0^Cd1{^vNm{-ngu6)BGmI2k0baQGs+@XAi$8zD;NH$4SVr=w$SZ1fPBhHLt7 zK-|CvgPkGEuAN_uogrz07-l-4EMASi2uZMH&4z4Idi>+e_V#4Ymf100g-t50G>Z^< zofUIK!!gYaR*iPemZJBD<~gFb|F}J-)O4VG83HZ(yIgotjjjU2+Xp&S zDyz8CIch)c_yX%buAY_T5x_y3;QE2?{h+)Z=?R>5>ppuFX}ehil`tZRP(;FriXBAJ zuFiCA!PO4pp553bg%v80_cgc&B{Bl0Fu=343&YnkvVsk4v9`Lj^SC>t(_%FMckTBZ z8WvHs{VzuvOr;ydi=HlEmdA!_7txp$^qI@9w}*RJoWZA;^m132D8EVXwfJt*@k--u ze|(jcQBaceSG*T#zLLU#ZvO=|SG4{3<5qiJZtHJQ0UhA|iT2SC%IIm~ zPxdV(v`FTW@E=p6UOU@J}GIFDANbb2Aqrt-&tFi?v&tN}-d#uAT{gd@>ptV@1?m)FPlL*t#&yp7$^Ue)m;D29)qczv5qorPDkK27K?xvE0(oeXsp=B1^ma z;=O(Y3}1>a2o(X#oiBy*De~yfczY`x4pXCl$ z`+^PsSH>}A9@GaA`egv>pSeGz-PW~$i81=mZ{9elPOlZB5F~l*wW^+U`BC>gkP`* z_Q|~~XO)s?);@n5`fq^CoOx3kJIhG<>K}{WiCJvLT9U!pjcF0d{a_6V`teJV-;f!S zzpVGdo9C$1(^Ctl>vZ=$0``N(UC$X$u8$8n=YK;C6A$P=kT^6oqy$-9KdNjzVYm%i zg$<)D;s-G3anXh#?jTu$Z{aTW0|A@ESULIF&6oC?u+yZ{SNzk9GNrWvB8qYdILTKU zrRjjY9gs~+HM+vEt84~SX}Z2TB=Bf+Z6c<`m=*|+kXgB1tfCxvOcypNXR5F=`wh{T zY6<31)vSty2X_Tm&E8sWkwe$Qo9+pF)U0KLhp>Q=MtQ?*rTfBLToU9om~2x0Ju-TKws+;-tsdk{k06tDT$=n zuWgKbb>*{82G!}DvML>~IS#_6o^vOLjI*u6Je4JSKbk?TXxHY{i%IwpS+3&d^c_O0_t6 zuvh1HYQVeFfc|UDfhOtABJd}&%bwj~@0+5-VX(t}u99`DZ}q6E3Gts&R@~b-q$v`H z+C&P+fLbEjVx+~Q*jx07L7A?b{#%~A zSTn9Jh=17)X~^8xy=J@1Z`TdEcyqLQ&L@Z> zul_;Y2V86!Fj%G5lKAP7Rx>DFz2O-cXk~>lyDMK+>KYq7N5T|JRELVbi4$MF;3INX zzPj9wz`QHCKL4S9)nz+urv4f3JgXk1xOGbIA6!e)VVSd<0ynb`{&oA0US2ujp+99BS9v6CpyvWuqy04lO zxbxl?dW0_9KMr7Cg>7geT(umab(l`YY320B$a`N zy^#I%8j)fAJC+H_4(#N>i8o@7Fx-m7|5mM{KB~NSbFc$mkxvnFg$Gjzq98_>d9<2O zQUd3Fdg*qnZ*@8>$XVL0ue!~Yc; zXi+?0=RklS+EC|;0BD)@P2L!RAt=RLar(m-Rx1_w0Yq$f48`|@5APi&_m#P^Zw;1! zzkOv@M~Q=SMP^7kr7On!kZSPK4(I*kAgZ&Q6#J4Mxp$q)l1%0QX*X%=577rQO=(FY zN>>szZ?%e*6O;kol^z8~j#{6{TV({W+dY#jCwMz!7DVBSksO#w%fA#EIBuNx8hoH3 z+_Nw?7Fg!-V+Acx_MM-gCy9~+Eq{}uZC@;`QJ%a@7?k6~l1RJ4o5g9eB?4G#;ZeB~ z;wgh29Ox)E@?dCe3)uaaH*erKgsyVa0gNA*mX0LPwc+vHt(q=}ktN|^S0I553*MD-Sl?f`ipsROEr0u45 z*ZcRe!2O`XJlF8G^jMARJeP`GCr>C?QGRW+y^df*azUq#Uh&7&>mP(Y%RYTf_YhUr z!oH%mYsQ6amRv7{R+g)(Se;JUTh(hnI&pu9wx3H{`MDknJu;PJfu1Nt^H-%7qAzpF z#82RhW$DQ4Q5)3M)oHN9(~mt!#83fpicmH^&;?B7uEB^3Hy@y-Iu*vmnv(n8gz}tS zu9O5?@*Tfu9ZT@Ags*fb_Nq$LHeS4>mTw@C|bhp`~iW$vCr%R zE0T-eIx6rpaSZy+{ZQvqw?rIt^N4YQ@%=~*Nf1kNA1A*hb3oXl5*0N93L{4QLUO=3 z6F%A^{?Wa|8ySPUUNNbY5NM+~Um=*rsYgDg()blnlO>2sviX2ge1x`jATR|txlILC z=)uOuc;~iHO7&zXov_9i!;YxNb#}Rmix=(Uu`O=7P5Ge64QaciIV?`UsixB-xTR)3 z#s=Ioj`iICzHM4~VfaE45z8+ql6q168&YZI{1%Cr=IMYphZ)lVj)*()#TK2;8spJ8W4?U@ON3z< zw~g3Ebx#`~54z07LaJT=&hB!J>eiSot_k8#h|uw^7w`mpD!6-nn2r7a*u}i?J1WIHbF4;*ouj+h~ zunMS4a+!%}CijUrx0^G+=Qu0|6C(Snu9I@New9B7{X-|@<|F^}#V4f+{ez)ZuO7!p zf#XU9Fwf$6CkZIH){sxR`0$};$of=}J z34C!^6Th*QW|9Qbu42O4GPH^UJv6cY{D#oqo=~4rgFIDOXjzku5a{+zDw*LTyW%~c z5lt_DQ<7Y0;*Feh^4r8!zJ1-qAq2HkGcs=~d@A~C&{_c`kyML)S{Nm`#vg}l= z?f-$3*G1lN8sjldy(q6hb6n?7yp3O9a>cc^;ykU~byK6{9eE## zq74putxx~wV;|U`mjD!8nw+9sD@SJvyrf!Vs$U$sxzJ_u$F?$#);w$&Ig*<|Mp*e= z2Z+WveKA{5AZqs7H7Hr8;NdFq`eIlZWyaR0cBuQ|;)9?s>B5=cWs6d@)6G>nPYHXY zwd*H!>xHWN{Wniz*XF*Eh``^;hufIo4FU3HP1G)>56GB!>f~v3D5vB&@q@=}MMaLAu_d zwVhR}z~mWo*_BOCYuUuW=L%;Sa5K^-;Xryn)1PrZK)){$~O`%6;pzW-D6fyYe%wih93S&ccg@u;tkmEiEfe8gRUr_m~fXQGYJ+ z_W20@xOrx#uMl6Cg%++B9gDz{$trJ}p8^emd4zx5)0K|9Gvkf7*&79gK0LOYss5*L zHNW|c-XH8%Iq}`2EU*z*iJj-@N_YYh@07^eC>TCMb|EO9+ucf5zpokhMs)I-@(*H# z!*w!WzmItHxN1eej7eg*{@^xf=IpN7J7j9!QFQrT;~TCpGEX+ff&~evJ_>K+bdJWY zZEKe%nO_|`A4aZ&2C#_xGXP}bJSaOK^=cdz-AgY=<^J2u5jYI}xkbmg%7*=6#2IF% zR5j0eR$2dHWXPznNuWw-gSp&V;WR+WGhrEP?Nenvx5J{7w~0gg=C|XT1@;P6tjQS+ ziSM)0DzByncT8q~=hXL3Uk3b2NxFV+{nhKnr|DMCTJZdOmw7Phb<%y8bfFb%(UFi? zW573z=wx{0?)-NmHV=cqZ}zS6Ker4KpP;WN9siaydREFS696?y4%*`%=@%>QX1YaNjHV?DC1jpeXxDy;^bjNeFCQnstEZFF^kQ<$6apkA{vUBZCNV*b)n#BvLBv`)9URc}jek~P z;>@(F?q6V% zEQ%QDon4pchZ;@#dCv^X{fzGlL?~_9&2c8pC&o)L~S53DkmrYJWR8WhO~dd1c!wkk-Rc?mN;;+n7S zg=%FdkSiKYz>3P)z`CHQg>>PL-b~yeLU>V)gW`oig2TJiG zmdg_h*tm{lRt{~goHeAR;BDOcP*fy*ODQEkJ+>UbXExkAQ{0Tf!rK^8{e{=;;4C?J z&6RWElv?xG8;R)k+_uoQ%W$v2)XYgHdS&@mIpL+$_}P}Gzq+E9{=Ncvy_GP+$%a2+ zSPS|u$Lobqv}LY@%aSThqbu1RK&c-%baHg+Rp9pwbFiz-fOG8= zE%3{7zx!`SArcm_X~*rrx;8PvcX7W!ho!W&*N?P~<@{0g{H@(gvp;Erm+y7Zc`>)$ z3{>!~tuNpZ?Zir#SkwAVvMjx+mQ0ZWIyG}M{KSvrWo8-8^m`<;Tsv=b%%M{mO3CEE ztgE)Tl9uE|)nPO5Rn@jRsyMx;BXo9I=U>dHIlJGOClIKwuH*?kp?KIURkSgY1m)6u zg{@ys!7(q<^N5rw7T8tOA5Ue%2hL4g1h9EP?g{7i_m$t>WVM{`e>hUhxjfWO^U6yN zU4EGW4gC!ew!X;(hNBOiD(6d|@}i^2*3}PWe|V=-kHTw+f6SVyEQ_TeVXD?CC-8Q` zVbgE*4jtH>2G>*3hn#egT5fhT5}4}Vm!FnfKNeHT^uJkg1Eyj4{IV$36K#v-Ut7nH zb~A71qf4LfF6ko&rq*5huH>v(N-X^WUz08>yBTGg#=ptXh~+>1@7ppn5teo{@zK9( zgT0nY=&6tElv_K39CeC9Up>nfaBg+d*K(uC#QETJEk4f+?^}iPK6YxFm~zCz0s^+x ziF>C5a}4g^Tz9n^e1ThfxGS`NGB3m9-Vlqg*>KL!XzFCzv{}1OB=fyM^pQai5$FbXUAZg`2YD#VXD{Y1u582xMA(v)@ z$Ftm;S`$^B2>S1{Ew=&eyG!pE54bvIzxz`M*||tR%TqbFA+BmK@m^a^iTzZ+PtZ9x zC_+4c$+2c-I$lLBo^24ONqBm6$dH-JCwO|qX!N$B2&~zlecV?*x?>_4Tw}JibY3`^ zxeXn(@wp~l;G6f{9UCZx4%~J6T)VwN3%_Gug~dRo8(12(>i5rtMu2c_<>T%5{0!`k znRamrg6NRth~#{*S~M43?m|Auyx=BH=f#C>_2|4cNSQY#a8dyb5P8V=xYovGP+ zMK!MqK9fD1Ds|uM3X1)1NW{&FZacx@4+oC+9Eswlq*Z30Q>EBtEMKQI^d%W!+E0IS zgq3#2&ei6B96NHCA{Rp#Pe6Sny))1PQz9xvw>>8VlQ?;muO zcC(;C%AWg4hyfSqhU!p1A6$#Aq#xx|`%{JM-3PL-_BnqGfofP7>c7#vydPy0`* z+s%w6qY`@^Q3#9;mipBWzXccmCN+Y=q??r3A*-!$S@_iI+ zUw6lBO4GL`rjl>+f6?`pVNrc=*r-ZNiF8Sq#o{j{=y0%la__zNw*HgOgm_7b6 zrU5aJ041+?hJ9qNjW5=)q?FHqfa^Qm|CNWrUK8$-Js`i0H`Zl3ye5@xDzvy-j1TBK zqx$uSihSJ}w3r4}93Dx?5ky~FP_D~ph7Lk}RPs@z1OJ!L=>~Nmf`HA$R&&AJ0EuR~2z+-XW zr{SD%W9nPyB6s^7i~3@!YF~%B)LsBRUu<6QU2kp;PRSR&f$C`ci- z{&#sN@(aa}i7$@5z|$^7_e74xM-tS~Dy+B6m*emtPFGdF+P6qXl%?QAr!#l9?cG`v z^6B$)(c&Z5jzAv2f3o+EmjyV&(mDg*2l6j>8nKyG=_@EY{keGB>A#s+6H-umub;G?t#C5u*|L9hDW6dRH5&LBK{^;Yqsc!ULNBr9w zr~YL+Cf@700sN3>ccprOgAH)YwAJSW#|o&;m@6FY*U5mc7&o5+I{gC>4shk=vg4&E zoi;(Ec_l*UZ$@h$99xRX;_l)0OcpqH@d3hHT^nQe$9JFn{~a>T%wr^2$8j<+pzK;&2}vcn}DAQ8N6v-@>9w(+ne z3I**6?Dp&fyv|;c6|143S@vX)W)I@(%>zRL0n;zeqWyr7V^%5byLrE0tMqx@a>Xq_ z@%7xHfeQb_>+CIAw^tM0Od%2q>cB8xZ&jYf$F0IduU)H`+kyHnyH=uyOU!dwU;_S9 z-$vHsQ74~%Y1N4xpZhpQOicIM7g`WDwJUK`!=WXuiu$v0hP>hGJ*VhSEJJ-=ga6e6 zK~~p#=*?QT6x-)&)5V((9sf;!%|1K2y?eZ@c5+03g418mox|kb@Po1ZvQBK-GzvON z7D%}K#5Hw!^d7|k6dU}b$k&tGN`krbGXYvb9GsppBM)&qzb=1}Js1bDCd;T+uBhd* zzmcO$Bej(#2G+0AI|3hlnR$+Z-M>k-F^axw`qY5kvfiGZQVWo2B@P32Dq_{q6(3`= zz+{|H=TxI`7hNN@IoR~BiRa%N3HpUxt0(u5?5EFw(!He!U$<+G;2nmOMHh3GVpyau znVs*&lg-m3+B8-^@8hu3BNyjyD~jW0p3nNEOC@k0OJIlF#axa)bCph_0axXjar27$ zQ4$5BDsot;^z!57=f}OOmanRP8|!8)!Cvskk2#n+VR=+nCk}h|JDK}B!|m7PgWM%s z63!i36twkUn!$M;nAZ}qjqOP$X9T?5oH(S~YcTiEMK&if-K=I` zX|dzyGo7DbnaQ>P8gN*NXet;AOrcspQ0{K@ks`lz!CvLbCoD(mFmN$;el9B7e2?{f zVs;zJ6AJr-PVWJNh5(Y~PTIr>J=){t-pfhRh4z%Pzsuuj#W;Y3YXA8fye*+H`yTLB zu$(G@{z%ltv1nteU3Yziu!}N)9z-F0NCh8$DEpkBqQ291!4_;d9@O|yCK>V*O#12y z-0@w1+yiQ%b4Uir7|*1Pngqyfz+Kihg6B&0gBgHD(8NcI=t z8D#O#>lPXb{`%OgbpqYBXaNjQpog7+u^eH#FIjDo^jlN`xhdBkX% zBkV5(Ej$Vq!4WntO=&od(}kD_2stP9(70MnCAjhxYk;+>w?SLEaI-mi$(sm(btx>$ zxgyZ}N^_R4yat{%5F4N zoL%Anv%_&C|F#Xno2wdLWuyDDKVw6QbA*7eIaoNZRh3f0m#o^gqNcuzabtufe2A7B z=)KV$!BZa(vk_BirgkK7{@v7VrCtI7-_RLf4F|VL(!zzR3(Iu}{$$|i% zthg_kwu8ShOL%y_dtuJ_TBIk|wtOk&hatlT-&O1nyCd*gpVOWu#j+ft2c3fi8bn@! zFa*K<$K|m-=7ojUkLUG~T`vrquU=cqR_1tg^+rB?T;9LscdXPe>Hd+rR<|%m#%cie z!Unk#0v(=e`h4f47{3c|z|SqKy_3s8o-jMvZ)T@IPQCngX*dX}^{2hlrDN+Y(ki_! zV{WE0N3xB$$bw;G(Fi?}M?`K>ek*mo94^Sq$-n*i^XdKdI04_;s#lpiUS<5-!MJs2 z_W&NXW?#^~C-#cX^D#(aVNHKru6<*^q}`b5P+opZ%lVS>zn>SAH|&R+XUn9ue(H_5bm7rg%EUHsTpc2md>TH?OG>%qxs1WZse;_s|_}FBiZ;N z1G|^tTCwL2@b&0dJAZ_u)Zsw9HN#kyBkUd74lf;Tf5*u^9!)`Pdb)MoP({*Rn~o@F zRu)s{iNjyff9PbEv}H3ZtIBVUc!Qy0Q@on;b_w{xyjx8+Hpo!|k-7|PS7YAt*nY9Y z4Z?h;h$L9kGoyA|{diB5K$6tLlQwC@acy=O}Hop4~5(LVH7wHNPyeYx! zc`r5q#%uNC{Pk?@HDnJ7mzJG72!P%5NjIBDm1jhX|36MvVtL%OS%v_D%e6c9$BZQ4 zXl`!J{;&&CmsoR}ZmC1eAz*GIM1P9@$hQv7hZxS1h1gx+pI-)nG^27>V_ypGiCt#3 zqN`6ZC-?r0G)R)?8oZXD__rPZ_G?poT27cs@*S}OQNGks^p$R{^p;$>wnSP zhTj!~_H$n)e^R1krM~zk86hY#0DWcZ@^4J4*@NtnXq1X6Nm9?3AT5Qjz+D;bEiy{e zkJ{6Zpdqh6d68*W;}4*jyJA8gu!JuJdQ1jCh{IGh?5kEUe6H(3lrwDf1bJ*+Rl2&d z-LD?`+#{qnQiIz2-twJ)JKVK;rI9uPVwG-(goxJKi}Wv+(|SP$@EE8B23l(JO5S<} zI%oJGA-3=dhPYNDjJEc|p+!?{-lb#_Xs|Bg!125965I+nWZPG7O|jjyq*#}d;NN#;(u;B= z{`2w2&6SnWKs+PRn2RzCm$YG8HhYho4yUAo;{j&z^FZaRVI^+F; zoO-RynQ5H9cKcl$^TtkQY`O--ez6$MHX!Zhz>U=@n5FOcYO5qKXU*0Ohc!$=CbQ;V zoi=?oMDM8JqSxF7PP{+Qr6=MUeyH`Dd;#fSP& zQ_^hef#`UCdam5`RoF7sp7{yXjHBSy;IArhVVj4pafAsT5YaUiNX0M!Z#Mo5OQZ2D z?)(Dwi^-0W{9T4a&wFHyG-sYlX~XRemawo{bh*Xd&lMmF9+#9 zazAaOL1o#go%Xg)=92(O`I6UQfZnugnmlj7o@uvW0{yzi>6$k{fmfu@FjawOOYCL* z4Q@#qnInc0ASS#ShG3;kNnmS3v#;w^W@7`=y(gBQSq(syt$z1OfV4>}xFu->P$jMw zHP0is9hA5ZyHA~X2QynyZUDP}Q}U^$_v3l&Y+8-^9oc2o)zDt(Yg3cYEIF6HZpkb* z;)iaPD`j#ms@{QTQh%FkC@zTAP?RTOLcrmm>g zu%x@Y9SlZlsN~FtOlf%2xX02DUu`8qWyrM2f)rWi5=4ZiacCyC|Cs7FMU4g0&H=x9 zL4UnHier5dc4GeRK88avce;LTjd=F2j0UPWOoDie>IpQ6eA|K0kWKy>^wdVdxh%DA z^Uvqs_AV@>xIqm{D=)GzpepB01G&gOKq5@E+TKgkFPiAvwb}HkF{@F^IHCC&IoOM% zsHkYpm|RqT=uf*oKK~`PVvQuX3is74Q*gJv+nxP6X%cJ$G?SNR{W+LXo%0shULtr7O4r!uH4)($ z-N~fwi4`8MDr?=G@kTo%wHWuVoVi%33n)i>rQFF>&!nDB5CVPOOfwJ4m+)V2Sf_u? z&F*G|Wya?n-y|liNSIWmC{+-0U008?agS#s5Exw5RFz+ag)s7) zTW7I^r(+aXVcV~_Eu>#1^(5+HgTWVT5lC}6ibuZ$9h3$l5Fd&`ebtRq9{e>n85CD! zvyMVLOr>Mk+T#I%pMi}ae&?+2yTIgK9G|hG3@df^KQsRO(@B)$Yr*}-sPPWd74Z-H z=nBFl{(c~X&vY%%XE7E^#!4NmjflQb9sK1c>XneTEg0u-rMMi96##jwUN;Y>jF5ev zRj3}|gLA*2nz!cSpy2;z;!|($gs}oG{QeGA@7TUK^8{$VFWn$0^~JHJMj!ADlB~^= z`5ZEyHP*yz3ET%zd$d{F$Z*JE3qPLyN3tJFGIxG#Aw4G`MS64RQ;%kvKBu_|d+o@f z;fD=RSXW`#!%a=m!=9z4jghHZ?!%tYrWz?$s~n$^Ze4+G^G?3^8J8*LHpP9sLS z`4aX*0!2-8CK5T6t8y!-T+Z6vMQQ;y%|tk-#9Vi8+Y1OBU))S!8xxynTMdJNVp;!$ ze4P6rYLe0hw}MqR)}C}2`1@PuPX28pwq;d#qz4&NKqUU!__T>xy*ra{bwZ@-3LL7K1khX*qW+&L3b`X3z=k$mSqViAf$E~G6&iS23O~+ zP4*bHzQKARZw~t2X8ITzq1mRo5r%-coN9Z?Hy1lvFU<$SV_`Ug!MFYT&tfP&4Ikbj zG&05Ky34__Gd|Qz?C`irhQMe9yoDwFR_PR5BB!bj$bdD!W-a-SxjPy{KaGpKpP{C3 zcx=f!09}Peo#2JxGNHbDi5a-rrZ*DtpB-Va>r~Nl))hI+2$mclEOsSe6C`VN?|9i0 zNgci^v#3g%Tw>IcKe`0$2J!o7XCWsES1#b^g`tK<`{ODQrLjue5%z@?M*^rtVgAs8 zLWbsdVfTBRy7j-k;D5+U4Ogfer!8BBN3ERl;$sbpKlyzOl)y1so?M!Q^kw2MYWQZs z*RseY2&5C6N+9pzFbAOw-7`%aOtVtfSc$Hg-9z*6eyb-!Nj>(`8of|i#vgZpj@K&E zC1i5EBD3Fhxq0m*a~w^4ELooJW8_GC_{i$7ToE!s-rPFEHx%a`^D$ndJx{dNXwOU2VLsS-WLa$r zqSyG&bnxv(HCD1jJ{FFpW6?CJ68!f0MLagz>$*F!O4tsq)H%TBZj?$6Pk!y7ue(058=lPYsdyYyy z@QqJEXJSHAO=wsb4PZmgVPmcOVmVyo;1T@WrMc+T#qD(Cyaya`MKM zQD{ZQAbxSK4wo$;{G8E1Dh&Dm?|`%c~oHHIL> z-rM;@5e9_a z=9818rsy5B=u>)AuB09+%J(xx%!6+;K06K`!)&yfRJV+%np^9VS`c_kj8| zs9C99Bs6bW+%u0G?_$T*d-QS`e)v9xpE)5v(nl$Ps1~FrBi!0LszC2PN+&d}hwy)* zeJ4}`X_<66z86)k-&g?gc|_T^iGtf{Cj<0QAw{dIpSo-t(_j-*6cK^(3; zA5p4J9Wk=o@mV=s8B3)Zc{xvhxE!7@ zauo0^M|I+8&hlm=Ev;QU(@--@Wwlnup{as+sHu;`w~o!L%z`F^oel+Me$}_=sCiRA zrJf^C>#Oj|K~GM-Uk;_Sme#nm0hdx3^W4iEv4FPOlz@0cO4rz*$GuSOmEJUk&BijP zP9a{V;f9L#av9`s=PK#D(0MHiN`Ixj`SKA7*5I1!fb}Po{tSXkg;U`lTeAwPcu)}@ zuO}OobNka14R2%rJ@KQ5P>9p12WK-l+l~PV)XE^cY$TzfYFb3R;PR=ZP<+niqJ!xR zsY6EfU@Yf`OpU8Ao5!h0qU+UJi66<*t0SLrp01+t9^h*bx!7Igtqd9Uf9^j4uZ>8F zc~wH(JN!=Kr6Z&3`>~0R(2&SgN3n7R>R@92yT{g0LRr_grG*an9{%hyITeh&& z{a#7+*H4=`;N8;!QljaQbxM=~A+BO>!~;pX_+1kW{erA-mgC1&QOQ?Tu4&iWF|Orl zam^LEmkEy1lrv3i>qtQsG~Y{CWBrd7?;B~C(Isw0$1;n(9E%M)lR;J{^ry zN2areItE%rnEG> z4f5=pNvRETtwn;9l1E3r^dQQ|jqmxMEefh!fgW>0o=xiBf36SLya{!wTy3A`VgNSp z$mvdpbo@S1b;8po9%G56LQE)#@P|C)mDBYa%v#fbn{5r1@i~I-x4Z5CGwh- z=g)AKDNLtV$KK#S%6Tx(X{X^M)-6g>4{c|s|6zO-wf#7&u3 zU|C;J5vvEgJigkndPA-rJeO=IdkAjBz6}RMsg745`|0Lk+BHQ2CdrHYqSGYmm1ra^ z8d(o!B@kMA4Xixac}aM|>$fyi@-K>J!%kvGb#k&64<`x7^%GD&X!zO zT7LVKD=xak6qTHJi50+DoB0z>e*t6XcOV^S5cT%Ix8GJHhSk~~KE;z^`*vxcC~4i) zr-{-f8?9j3V^Q2Q76qBl$b1bYgaY9N&R%Fi%&Q+FLcM*>B_!K!@>BOwxi7?b>!z-HH z3!ul;-8~G^ah#gWV|S!~f9w8H$x6Nb?P8d7;nJ}%0zD{oxTDaT{0dlio@wekEv@?* z{73ZG;9r23lJ?HE?Kgq@6KZR0QL43$(Ilx^8Pi`dOj3t~pRG}!2Jm0Am0QWT{dRT! z=x*oF8l4>qOYtef10+9inI(t9#6%ZSBBxU9dyD{jCLy&+^k4C%;5}~!1o<&%_&3#o zCbm_!cY9n#u+D3y{}MI!N!jg%-+s=r&5tNu|I1}=lc8;oeD{}#&RE}y%RIcaYgu;4 z@!XXrCZj1H`jj)#G1_%nBuXu#aUCXiG$8f3>aTXa>T#5BivfGY;MCY`nQ!%|{`ud6hHG#O&dI8#c+;OjIX-^1a(ee%gIyzPEgB6_cyL|_pF8->233`y=r6baUNU#sTzJu8TUdYCUpix)g-S+02ueR8}ZA* zvZXVrcHZZ}IzSC-=jxrsSY(CgU;laMvtR^r$=FMp0tv;G~_Hqq?^OLPWP5!5POzgU#}cq4Im z(*62=#gpq_)}uzXk;kwS-wo{^clGxFl1_-(d_)sMGzY<<;KBJLP5BRz+gbEeT@0>^ zKy$TbU{x@+cY3XrRnm6m+$@@zT%1v}n?yL9fbO^*G`_8tXREl<=WrAQ> z)*diRQeQAX@p%94(u4z7))bf2VamC^bii(&4?mCU{bNoOWTj&?Q{tV&@a05CTl7}y zjDb8Q*e;uKErB=w&b$ju^506kf=`vDI=bf5? zT}imgH=2R2;|x`g7&~hkTQ0@R6|!f<4s`>Jn-_lngaUUZ_0fZ$y%Z@xGUnoz@kq`k zRv!Ag#P_?)aFeLzVa}y_3DX7}GovFt5o2|q1F1D-DmH5)4U!K>U&ewjK-xTd+{vS~V7%zP`@1d^~_$Wse zh%@mCz?;7ushVS%{xW_7-~l=mlqb2ZXF$r~=G|SZ*Y{VS$Dio@P&uTau)#Jqlt{%` ze}cIlq(O4;0||Z{hharpFIL&2q}6WZ;N*@Kv*+!`$=B)BI5?^Nd*Xi*nt_&7!Z)9L zp-@&U(Z1xZ*#zuA!K%{ruuZHQN~e8&%t(2~aoz^mgb!@!)T2bY zu5!H^hnse23z|>%YSC~L+D*5&C@5$@4_-y=PMEESfV^z}ZJp^apzo*d8a&k+c54)1 z45#QsSgDl-FA+KzI9e%gekD1!c7J0&vCvC+PR4cZod9=6dOEK8OteGl-xeg4D_A*} zcHWjnhOPH1&Nw*bTJC8DN~6Y^K7VeiGp_wICJ1r!aegx%y>|W_nVo8A{s+FE;-mNP zUy7t7eLW(OHbp%>{AfE?@EJ-(f78H^{zci8$BqRz0$)?UqhfMcH?9VEzIOPbUJyxN zE~F#RZ50L#s@*&U#y`)rcBxt<*KO@D<@mIyYS(QpCwk=gpr?t@J5ESX?c#NH9NbJ; z*vq364I3j?%35qsO`*Terj(|#L3;8Oce;XNU@_SIJWz{%)I=*#o1JOn6v$XFVZJs# zbCB&!3k0)f-Gh@8{AV!zh4HU##wGYqVgbe=LcplA#@vL)7Xxc+DOVUoM1dzJAcvSt z*d+9iH8Z#|8AYSs129Z~(a~9YCJ-p|2+_2Y`SH_4ZiNN+l`8{QhaNUH*)E55C1$)f z7Fc#XG6%&q^7$*%E~RP%W|ioWLY*~cY`moh7E)6pwWweaVeheEx$qMj2Y}1!!oKlq zKvi@nerOgeWAOb0O&xW{f98l&%Q2p>HI#K9=t(<;&2dQAml+n;tR4A5z6Lpwym(Zgl3e3++*_v%c_v4gB5lFJ+S5D#%l zr|(Iuh#ZoFIL#|;WPGXY$nO~S?P)F2-@z>2Sskzp$-8^St0R-{yrLUctRcdrdT!4o zYF|E-s5E>)MI`!y|G4gLrQmrx;{!)-876Z@D?GLU`mTXYxp6bvcN9IHwQQ9)m8MyB zcgHMOj$dcCSPQ?_5dO?X;WVLvo{({(xp~g^W&N~9%lHeiRI619@xdaIUDj1=z}cjQ zdhMfQs$2q5d&{yx9pU&*&SHYq-$yG?9h28PgId0zuC$#Q?0B5R^R&Z#hvC?toqWCl|j3&0~`*v9^DPdedW=Y$aPA< zn0_Om7M*?OsQ6GcTn{nk`w_!d8N_ac*yihuT(foG{}1*&kJA2E*ghxh1cYFlp@o<))3U*65YQx1=s$Fp9aWmjnlO|$eIA6td=^!6g_7(POs#@4xenrFGLgd5zL2*s06V_TaLz(LCa-4%Je= zSfk|(Gn!PZ-Db*YcHY`L|56~?+WpqHgUTi&JG3qoT+GF*KCp;NE8Kp7)_qT&_@1IC zZc<*=y2N~@v$Es@-ck8ZMtdpO0j#3*BpqJ@kUU>`lPf&-TXkYj;-PGFF!Oaz7reVl4XK{>!*mI3-%NA5O zU=JkgAhYRQ)I7@Giyw&i=up>WS5v^p3Oj%^08l4qRW1jPJmxz8#g2%D7^+BXR zBx~nEX)n|S{=)rtdV;A>uH^TQMV5Q@eU+R01y*3pESN$`Wq=uW4@lUi&AE`p-|c%L zifCSs8aidUm)l1ZZ<+P(d5LW>MUQUnuj3p~ezj9M%`bbt|1*df zi&3&P94z;?h(GL_# z!=}=ymev$XZQnoedJ-R*dws-opNWoKvxE`hGJb8*kH;W%rV*zx$r65yc8i&QnN;hx zElelD+vu0uVqo9>-r;e6FXx3j`Hi{!mMwclyJeR8m(7UmciJk94-2=&jD9a7a`Vk+ zUeZyd#c$K~_A-Xb;`^d=+fT$3wvm$ah`m%Te!u>$}IKd@dTk(uOC5!g?j2iy}V36`Lj&n<3GAe9S6fdrN;&?-UB$>YHIHPwm($%-&j^ zn+E%Kksl9X%&skCL-h-N(N}yV+YqoI7ZvR8m_)nCmaS9oS%m(7Q+%qTueN9`$2v_x z?KiKeZpr`?kz?O<^Ue(wkrhtiTku6sTooF7_uS(J+kwbqchLfHkPI6|BPuixS6jUVPeg3P)URehQfHzOj z*L0A*j{A55W$xzUY7MadPt{GD7Hd%Yg@*^nVa3%NQ==^7Nt`EAucyOJ|Lp_c8JRhn z$dg}`^}1e4H2<4)?hE#z#1aJoV)Rapaies!I9+E)fse$Sd$`u%j=QoqcYut1;TxDZ zj>C*Bz#nMRXQ6+$3VZH+I;__z6FDbwUZNJ{8Am(5!3J9zDv>LacBfT7y8n z52_xinY|43#o5hsUFT^vAQhRnQwVnK#UqoH?Vlxr?Tb!2^>|H%O@HT+YT*Z=$E{ga zZtE`dnI6M?5f6y4jpe~5rJDnJQw}-z0Wiim;64Cl%LLGP#b{0mq=s&)-)v*{fyV-4v_GK@o3-HM7TsK?L?>AlJ5L; zyL;TII`75&O+w-kN9&1b1(^)f<>SiJ!)FaaoLVprgnA zBVl3AW|g(>80en@A(wg)Y&s*5ZX{X(i9Wr%LWSmwh7l8l0#!(pjQ;o77g-!&6m2IU zp-xHkIBS_)jR-ft9IqqlC+qp8s^i=}DI;kK&(^#C+y+zMZ(AvSk8$;!PW_PNp@K^8^K*b zGOwC*91>jxvoF4n5@#MtlofbbqDhW>$@P*dPUya9pY(E;p~nUF>=V2528YcEv+O8mb(e4mN-1d#>nS=u4voifu6r z4R4r+-TA4hBiXQjKagM-QHjxef`%L8{V#E12eeo^CZAiB54cC8TAeM!Cc9tc3%g>oK{h=n$9NaY{Y=pjp>^%mJ z{xhsC$KukA;T>Jub~Y3xwc3aRFgv2&a7oXWmvV$FiuSip-A11gJ<`SB7F+r;g|DhS zj9?rkmV@L1_59UOcyOtf%f})y**)^@Hd5JT?3 z2s6F(oY<84Mnm$1`1T)vw@s!6cAPJS=B{X8#$5ap{WECL4m02H>lzuA&5u(^S0A$| z%P~XHEP<9|OQIud+Jgv8bY^eC_!kFcf1l!}viuViWk-I`{6^T+SWyt*JgZfM((w>= z>r#^wD`1;1cD-f2K6w1PE$*e=u9dR4LX}{EKQwRf*C{(=iX(`)eP4d_F>Evt3i78|XK zo24Zb?ep#;45ZwE_ol{*Pz z!Iu1q9}ryM>Nr-3=$zlRUArlbhM^Tv9<>`cca&-@=~O2celiJ&h8Gl}l<7 zSxl=7m^fvTSVfGsg6b4XGO=yXx&D#g);y3!->tG-4hiaDwj6sS6$QG>M4_7uVIlO9 zu&*?cMBJT1C8THD*oYpuP8!Zg01U{^Bc64E{O5rD)8aziM0^0tcDXFCSnU+ZY-Mk! zCjKBTNB4QM@x96nHS*xvHz{DqiO7o6QJwvKB0Gr>@lHW|g9>8OK zwxaH6V*NWP^Ih1w2i*`k@FAMAj5@Rb(+X{NxK*{+1;N3INE3G*y@aIbu-0?jI6EmX=#lN zoDLkGtuYX@?%_Jj6J3hPrjR2P*~$8-ecxG(OL#fQ%ZfuI53(D1tZe>av>Ofohofvo z2N>VNyKY#UyqP*Q|FE|P9Nw`(ATED0B*MfyG|_O%a7rFF1ul<^q7TZ18}Rj@*7&SB z;`n0ae!wv_U3IjZ;rUBB^}t+73cksPAlp}*Rym$l`5=CKV~=sZ$M5jV%9~Pg zI@Gx?0wVSc*b^OsA&&X>zds5y{WKNZ7gcYQHZJwe$3iNXH+r^M-dFtEF2z8Z8&=`b z@jMkzdtU}(TC6v_yv!5$#e?_-G(2SyDGQ)K0^@JY^0m2^le-L8dN&k3M}27d*E~Nn z<%Nfbgab<)!av->3ZI89l5*>69u4JCSx?7w{YP7Kq^8vlKyZx?!fvG!g18>+zp3XU z5Xo#$x{gK8h!+?qzezOb6p$*MdmlGG?Wt?Bi=(t5;xl>bJu)rzntQ4vVCs~LbMblwmo40C_Upe@H zhKS6dci40EH4q*t{E>6~!U%1g^<&ZIjZyXSZc}2VVFV(AMg4IrQrl7ZH`oOaMS{i! zMVrIFlPLhE3oCXPu&(Eg#J6y{B4zVH0pa6s@1*rDUAY*Q{$b_>sMu?8v_hGboZ7BXmn`W%?8#p0+3bP(*dK} zo>g1@c5y39Q$#}^ybsM6u(fs({tY)i>z)WJ+%md_pJdvU;FxpML+CZTNy}FVrjNt+ ze0$E@-(FjY-HPP=zlZH!Z^ir{XMk~Yv7sVrp~D%brKkP*2QOwG?LKdPM$6mSP-Ia! z<~x$HI_gZw*&`mI>5$yy@vCs~Q+6h)p@nM6>#*rwO_Jq~?FTt{7mSX@%}{2MoX`yX z)vxA-wGVkzQhY-5Qe=DN8%gi6sUis)9rf3tR^=#cZu(@L#hvY+2M^zeU1go($?YVh zqAW8=dFkJME0Dx1IHRcP9ZDR-$ZX8sVYr2hUK>z|=)Na?Op3M9v5qMeXxm;QPzX-h)wVTYMZo2=o-hlTm05wPL@&&{}o zg&(8w&YIi8LTO{IMRGfmkZ62_(>{~I$k4v>gD74wBhr%JW#zAo1e%F*P&FNyl#Dff+|;0thExixJtOq%JMJ7K_Ka4UKKA z4#R#r-+8c5>ohkJ{M~K%@lBEn`K#8Oz7BN(`$L-9xT&S${BM8K>wM)1whlmPn?qCC z-gq&+C%2w&x#e~MgOSjE1-z{DS|#4n{;Idy+V>0nGFJ^df!x{GKt*yIU4>=v6P`WH zu|GM*-YuLqN&VYYuG`io?L|q1>@&wZuXM*Q<9R%5`t$P-&Rf`@7K?fFZgWGgx5cK> zLMd4^Nfo5ghd-J~ZFiw>DezSXD&p^)BDXx=aTj=&sNqJ2c%1~*0sCF@WeqV|jhJ0W zri+&(ocoy{80u&RHP7ElZ+-PpdNLMM*;g?II4Y8_P%Qxv2et>3<&ewOCbNDk(|c;q!yp{UzuN|o#Z+%ThjW*4vb=mhv)G+$|>&5f9J9>cdSa_v>A|b zCs)LU>q)@2H!Z!RdI&t9+Z4+eQWtxU#r#gdjAEPh8kQ#t`Q1I5NHdhUgy~0yby7*N z5|{KwrFOnOWU;{|YHlBu{!9GzE@I68TbFV{ffwsY=%bbBbP}wmgKZNc946gq0y1dd ztY($Y)FpiStx8_3Nr_vN8YXZerZPT866_BVPu;?eXw5OJYz>ZEF-okD(|khP{>Evp z!UvYN^>OIvl>+@FG0TuXB~w}bI2avD=RDnsXe$)qepxf;rx_COO{Qr=c&HVzYb&b(zFrNerj4GX9wkFWyo z4aQ7GbRTW~)>R>!xn(O>bs$%gT`0z<@(X1G{r_Myid#76bFz^avS|i@eWE0-r*YEe zUtnmIKtu2)6N*&-7Jh*Z%OKtUZ#pUSby#yRG+X!CisVc^MVd#p5?P>Yaio zvdRYKt$qNf?np*aex{%w2^U{K1N&0n;w~=t9XeyRpopcFUHCQpVTaC{XPf`Sj>!ND zJ&{#lEL~PM9r7@v(Tb9`^FT9m)uN>`=_1_TY>MXMry8Cguz9TeG=qyx zX5EbrH7*~?c*#$X(&v<&63OY7ZNWu&_(Z2SdHxnh5CoBxfPyv2|A-y;m>za8?;Nn` z^FM9(@UbzNp2I(m(gt4`nSK=sgD{XFm1~f?dh0ZKeb{gb)A;V|J^z+^`~6LzH-(Ar zTcvhd9*Z_0UAE_D>*uq`;iPN2kHuRZz8YX)p1Ruo9eC(5)8lK<7hJAujkd3V8 z|0oH+iFC2vEHM{zR?8>3h91*7=Qfh^W^hHTiapC`Fs-ka4_;zdyb!!1OQoT?>jRGV zj`%!yK#+VD=N?VQ<;MgVSk#X;A)?K0=HH`>R-0(~-w5za?f;9Vp;1N3elzjci1hG5 ztu11p29GqqO*s8ngJ6vzxO`X-6F95ur*P^lPUe3p_hfH|@f_AgQsMWpr|A&* z%0Gk^R5GkrQ^Neylq`E4?rCaLJKvMf@?W3FU90hI`iC8EWT^;535!WDJIV0;xMd4H z4ouiQDUwrxmQ_<~Jtu+^jXA;!Hot9t)P9vToG$C%oo-#o;*Vw&*a%ifDV`-HW;Lbb zAKxO-qB{m*zEK`N1?b-||6pdK@sEvY)vjRMagTbzZesP7Cc_Q?CD^kpI-jak;U5!d zl{_U`9wso)NuOK+L{YFA?C~(BkwHL+>Kd@+$oTz$#o~bc`h(gx5fJB6v-1 zWB%dZ1$W#)6AecLY3K6;Vbul9ps7qF73{4QFN6WyPqLoaVk$JIba z2w!Y0rlER)au)Be%WC6+(UqW~AwMhr>K32t9Bt7cp7!ey05{eo1-fuM49YXd{$d&X6ocJD${oRF%>1r5+n)_VO{ZExkeY!hK!8q|<~Y7NhuDpYixmNEun;@& zG%jKS@V#HgU+Y=~xXt_|qT}j*y|uMaP$@<)p6c=oIOzdv{$9R=_F5*~b*B&*FX+$d z2dFF#O!`5ap2}9#Jn`8128?^g9ouA4+s*+s0wEKD@2iEvU!p#KRXt~U2l%pp{ zD>U9^y2QcK1)uJ&;3ofLq&jzVqtDmiJ$HlC(BfYH5D0uuz{m~*^}fHPcUQ#g8IrT646rG(9V=q}ZolQ# zyz6C7DUcXQIogstB0n$GCge$Z0$bxIHfFu&$;`B`W6S- zF{_p(w``>8Bvh{cC$J#9dZlEQ^m!tq#pWQUcX3utDS3s$esmTe!hhJEL5m9+PM~>S zR_r{Z>;<-%{7hR^tGHT>QkE^VW%#=f9bWa(a6XNn=N85vH?8kTwQ^$FbWlVuvZm*Q z_ld&+n-h-=3t*g0*Snq%g&FUzZN`mC^c&;nTIQ^GJ5M-o0oqwSb$UiDb)ipo{>l8b z=0@TT{NynFSoZs1eX0MC0Tw1WaHkyY`6@kE81V@TuG?y%!LkYbu{Fb4Zy^*3>TA1AB!Nd|3sMwu-j+DX(ac%F1vpi1eCg2e83;R3k2CL(k}_MNEv^`=L@(3`FTXs zo&PbwSRR{Xg;Lb6KPuiz>bR}*w1@m~W=HQ>1jeh7YqXq%HQ`3JuYOsRB`9a5*RzXc zEdH0W>-*5+AP`idZ_%n<|3Z-jq#t9_=Fi=_8W>&;7%cEK@xNDxzk3Urs4kcU3Jo|VVwS8Fs9W0184|mL3$)kBrqO&+BiP&IB!yTv_z-%q za&a&IZFPMm!L0Jto2n-aQfnXnpa%=z5Q=L6-Nm_lh`qu90|}TkdkgjrMp`gQjl(>RTx@I zO1f+4ngM3cx##m;>-W3suKWM}Z`L_`@7eq8diT4Z_jwH86?{9?Aa!YsJIJ2|Pdpq_ z+9LhWWcpmz6-0lTuPICm2?FdoI6xkM^2Qn0{)T)$8Gr>Dlgu!dHcu1%A(C={eZ!uG zQ*7jcn_^q$sSdp?%v)qwYn_L|%WXCwTV7{W*a%7*@rg>S8D&1cz~>Xe^6Ds!5(S|w zV`Ml41YaGc90p%n84(8D^I};= z(x(e4N_uP-e)H&Uvey(#-o z#=3lZEg$Dp4-Hr*c?iwxI|Iu1R4K8r1r}vtq4IDloah>RWn^#3$_RFo4i%yc#O7N5 zmNB`4i$3UlLx>08FQ}#obw3>N6<_5V0IQwZjNkmLY3<7`!g^$UE+H^TLAL+wAo~$A z0wV6d3fBWS<;hSax+c5VT+F_F(`E9#NVwiB18yp-h?w1ZKne-zK6kfKe}EHZE(+Ik zrl$$|D}RPC?~~WfCo@6-j2bD0{%4=q9tWe8jhsWi3wUt5c-D*x zZ-d6qjtk8tKj7pXnZ_i+&K%ljHXeR&*>n$MRam5e_n$&hTspW~d8tRr8b_d}(=9x* zbDJ$Z;O9YZO{?hP@P_jWh<)=;x3P(Q>(;}JF_2i>ZUWWrbv#8HeNP}tBl6P6{MX9$ zqzJl#BCLG9drBddP?O7nn$#dS<-hNA&nB8dIo)BE8Su#J&2T9|Id!6TXt|E)(wH1Z^tTM33cG zEbP&M$UOsfv6<3=xmhqCMt=#L=%->;s+dA4I-WFVB@kH9gY?~EWT4Q;mnoj%*Hkli z(^n8BLE0Ip-VVYAvbRyoIMWZ;Do*EA&f}+;Xao&X8XnbSyS(o&#Jm%hwkRs_J%in) z&3~CG)4`Yk{!%76rv2~S#?`@7xWZNX!teMHmcVyFr<*$-YjfBBxylM8cV$`jcC=B zMqz@?9qVK(6%K+fnr@h#Ad^8Q1yvP2qidfSJR0y=ueZBh(pQl48g*3G8ll5C!NY%lwE{xU5p{_z61E`Pb?3+V5V+*xr-7m z@VNOV1kBIM_yR-#!^OYk|JGcv#J^r;37j+K(GwM#w~EcJBHUIrE(}UQ*B}J#H!K2(kh1#ztldXJmx;k(!3U^Mw!2-)r&eel zO+=-qakQ6oOq2djIK|s0A`~^P=T&u*`5A=`5AaT2YGeo$GL>2>EEcv?lD`N~Bzk`OANIy7c}3v4w69^hiTK&?v{nGt_k zPiUN_Hw*098*6V*=qMsHz)!oXI{@0GGf5r7R8_XJl~Nkbg@@KE({{k|9o~SsLzrai zO5TSC+r~>jG`|;t1{*li@@^ zx)R7?5KmdjO*fIk4#Zk<=yvY6eofKmHlA6Qzf^$A%+&mfQ~GEK^@gAA z`WBcb0lr#APH4n4@+tN(Fciw7W*WrVw*KXrOa1k*^|;4-I9z7#_l#|z#eZT5XG)16 z1C?x9FB8_IaiT~Zi)c-&U07DAAP{S6ug#9mAl{2qGori{E`mf)V zk+}WD)vg?1z_85r{umMEz1-7CfBqA-dN@{gQh7sb1>ryq_T+VSoLc*TLg(5Zwsb?~1R(p@fNJ9G<)@3mE8y3C=p3eXK^2+(htJeOWwDY++Aa);z zJc#^-3M4|#BaKopxQ#9FtJw=8+ErA(dOG6IJPD3^YvG0CZNx<5pH3XwFdK;d33C#8 z^{kg}fTeP~lT=1sMNj~uu zp&Z09rI&`j3DVrf^u9QTiRhN`475xM%u=)Haz$NCnODmZfJK|;!9ZX5Bx3HVdiRvbG$8hgh zu6uR~PjXM7J$vBCW2TeJtaYi2%FuP){w&nAqXbN;$;++raM*M^W^Ks)v>)sh0s8|{ zKf+=kPk<7?eYeG{=*1&?{2uUif5pO1&K{&kpnfR5@_8OW3idEi zjPXqRs5kEifvfBRF*~4c;|82e!RVXui;5%uNsE>xBoDL0O+aG;nvv^U2xhwrm?}t& zL#~5aig5wA3X!tgMVgn~N&gRig0KS3%i21;!3@CXfJkY8sm&q__lC#Lt!=OV=W)9Z z*Z=hr==fTgH7Yy&1<|W)juWjvxc*4I`%_7h?Om;D6rjN^*igZ3>#V<-LGwK zAJ{a}icnABbI3!b*VKvDBtaKsd^9&|&Yo2n)YOq@q)Dh5X>qlze?nlxOTfQbNFw|E zOURS^K*%j+t@_*31HM8&2)QF;H3AA3-xVlFmBSpL78&r1q#DqC5g3ag4t6=f*<(3f z!BDszw$sG3>?)}egAiiZqUm=OsjD^If|Qq2_*7iv^T66a>dc1m-WLUy+{wueoa}X( zJF#(M+wpND*((b4QAPM_sV5xHC(NBck{E*_SWYP+GZE=aOI%{jRV?S?AH`hu{GK5~ z_j%5M46EC19|yO)$!;m8jNJ(25lxUFzXb`hm}W8tf1<2)4+Kb5cHNe~sN`2^NF%FaI1ASrsioOB5*r?lO_Mr%R4-3;iD7;)(W|6bZa^8Ho1tmQr4t>ny$mYH_oS>2_rKc)~(BZ z3Ggh>O|0dE0F6%dyc0RFNZsd6)yJt^P&9?^GXbnx^Lt!cu?9jsNY)_ZfT&ChYGY@Z zGS|)8`|eaR!FigKiX#V-nq1B}xu!t6#jNQ`TcEL2pjgfwrWLVh z83j#j(aUA?n7*bn1QFTyYj|Xj;_}w^Aa*mI`gz#p^$xCEGAb?v51ZeAk+tQlWoVi$ z0W?@Xk3dtfl7RL*pNeVF)OAt^-;j~jYwBZhj~QeILPmNM{3?7X0s5>rDQkRX*MnUi zR9^owiDv=EZOBzpJYN1IGPW0q@o}lpBneon^m$9C$u*n5qD^z7;)lTXTz3VKGw%0B zALk29sV2$(-T8*pA}Uj1# z_sk7qm1@{mZ{+KyoR>vF%b8l=Pq>weqr(ncuuJO4SMDimv&)#cLVx7|`kFcozggUa znz{qk;8Y49Rn0erN!EZn zN!rpDPJ1&e`hQ+(d}*)I;_@FH3kiB!#Mg&|7QNvPzu~bV5%5?-?fC$10K>P(cT(v) z@P9rUJP};1{Vy%_zt5-mRNysrwSyggL;pJ!8mis@|EM{YZ?Eh`5cxk@{wrSc|0{m; zzvBNZt9GcvuYW8+Ip=0g-PM602E`$)gHMOxlmew>Tnu_9LBzGYRFXeyWDQ61XJO@&8x z=QFTDpCRCN_((|5+OPZfZfhIN$k8^|jC%ErRn%Rv1wev_d9?rCDS4<_8FFB)2Y5ZP zbBZY*1L8&T;e30cXk9T9taTU9f{Z62*yTnS@=oU$Y_Uy;=SyB(p-H(8;n>}T7XBZ3 zkV@2!gLDw2rJi&*Wa(Nr%worwhN>Y0;oz>VV`+Qv6*a7dEu#MKS_y%FysO$b7GpF~ zoGa5imR*%@S4EakIR6YE)7d{JU~gqRmDmaW>0%Vd@6VXy&rBX7>}FfdDAbKGpz2>o z;%7dGWxREv=0<#>|1oE#owl->rR@pjYY0nmm2wM8My!~gADUIK6-$sgK;o{YqO52F z^=*sxqgCXc{O#Ru+KD`*djuLA$ijKip*uJo&6>Ik!Np~zL&Qe#CbAH3BvY-4hCWOIw5_a)F1HVY&sz(F|Fw1tft%1wEITli@L-y8J5rabDLBw(JxkO}D|$C%?cw z^8)1Ct4W@>oqv4es<3}8z--oX)SHV`W%$45(5caiGtPF6{G1H^TSW_yJic6Jpl@AtM!;V*8|^l$2=VEEOp60PMU zZrobGr8pvOW$TpM*_#R0PKWaCS&fn02tJM!sP!8hQ>S7gytPhwpw|k$C|MT+;xAG; z)qD?6DnQ45^X}d3-*>fIq?lyGFI?&BaUi4HL#-CPE|T&Wgnjx}GT;B^#NXqfavoQo z+2L_A

hiGyaow3dSwwR6{v>B5eljZIci68|Lu4xS4mDSIY<1ltKS%=2iWDUzYh&V@bS!_K})$k#C<&epb{Odb#?dZ1h+>PwpMXp!T^E z4M9a|4Wji`Qd)S+RyCq)FT3^a2l4iK9btE!WAW~J>H^Q#G6KPf8@*ZSEpOW>cUcx? zUs#L+xYn=YM_{iXGRS zRuCV9gIy-Cf^g**&zGJQFPz;03Wmzgrz~fm(w)Z=W%WZrP$ly&HLzQ+V%;3tqZiz{ z9)EdRu063(4Q%o#wS+IMrhXL1QQd`vlV-TZL_LM&fNm(9*(+x1ovS-| z9xW1u2oR3)=WU^J3Dj?w+l_6uf3ju#et%_hDpJpP`!Gt$@^z#_R-uPzG0VByu@ZPY z8WNVwNRe)N_K+cP>OqSs zj2huR)bqVVCVYC#=TnK-Zt3pPzWn~qTly&A6D|HNK0%G zUaTYMq2QGrARFOfVfg3n+J|rNF%$ErX5{HrN?*Hnv&W>by}5%``m-+v*m8^oSRvf* z$mcnHL0qz|xtI*^M}P9SyA_}LG@3N&pVX-Sl8PdQ4`5{H0d^8oHh|{4K1~Z2O2&yZ zvgZXqD7zEgAvA8%3sK}K7_)n(P9z7wQ!G6?Zk|s~1KobAO z%MMR*l|T0EE8{C3-df*4$xZPVVjw+LKp>!zA4h#-I^P;&*?v9t(L@VH4p%E3KghXr zU){%}17>nxxqS&qz)SD+sRjFw%v@lP3Z$owZV9>3&va^f z6q^DyMMWRC?b`zj1eUi=z@wnOgC%5JBvXMkW|&934$9APt`wcDKi#&19MQA<4OGt> zE3|@;0AGm};`C2xfB4bC5=;9+y@&f&VE1AV3}qPC^89ev#mW_nIn;Qk;8Fx92Eui& z)(%#YL0lSOkuUWul24cZ;w;L3=1p}6MbxV{-%yy{%i|K4@6r2rO;%HFmXY*%i04r3#z%-z ze+Bk|i;DPgxOhiiXLObn(f8!TDJ_-Ka8`)$v7r!EvA4nqXkerH+UY9aZxQSYz(ey@|XB?CD)DUa`0eW_gZps>RI?RxOHBVM$nhg4S#tz)C$>lER>}w6jKR%n$9BSXN6GBM-Dad$av%Yr4z2Je9efG&FYQ{*qkE#mZ$S-?~^m{r}M2o`y)iA z1m@oTw5JGqR-z3SQ24FF(={^)0X@3~4sK!?UJINe?voGvHet`qo@ac*_G&~;8-zz_8foQ{soE>(Bw1h$5v2j>)kTq_3f_GM3_{2ONh zNb2!%ng&zV7k98nz49m=@)|rh#kkTPB)6oY07F~xnaaV`mR^M1G<~D~H#xMWM{0WOHTy!6?OPfb%vcmVxX|K>P-eU}Y zM6N0StEgj)LiBdwO3}x-rpVhtLg360L?>ozg)ren?3)XmPtp0U3S725$1N}i98h$4 z_JBRmul+n^3wCkPgbDuS{fx`qee;2hwVj(+J+jiybTG8pE5usxrvl2*)49H9CD7$m zyEEOMnM$n^xiHsJ-!`k?=@vHj5=t8F#`6}7^;05^7MP2=%-bOazZE_8%8OKA^21@?+d(NXOq0Vfr%3}{ zB9_uLe+52#BA|8?;HF97g(%$z+uDhc`myZ>Q3YUU72(VT0q@OO%d`Dn$Us|L!CM|H z1;5vQ>`MPnde!G`9V5e0We*|lSIcg!6OZ;IEX>*y749+7Y2pVFQ#DNPb5yEW*UK?> zb%gltR>F^hTi$N4V!l89E#Y^>It_oWf+imEM~ z8%h1z?5^y(VTAJEZ(X5xvKaP$0zTe5kF)LA&Ri3d_gCg0weDV__tlpLRB=6y7rFf{78LrlC19<;g6dP*3G_O_6mCtOW&ti|V*(9{lj}?0? z1^TEe9`FFG@+VBYm0%xqzOnw`cv1<`h(k+c@qr**2xuDVQ!KiJm~>}@ZSIt>kRHL) ze<$jmRDQrM+kT`>NAQ1;5txRuDUk%;~z-eVx(w_03e zl>m@yZrU$)8n62Mobz`Y{kH_SFQd0)ObHHhm6|R1zU|tn#m%mm9%Kvl5VnTS_yvg# z%lg6-=Jp)w&X{cuV4=gK0n->_O}ogBXR>!ilFr+o->9lPmyR<$$(F-=VC9qtT*E2@ zlsMiqoYz~Dbzg{WrO6~;)M+0n{5={9Fc3Xl8)dTy3vkue6`#f&+%;&YJjWZWF5`Ak z!JrX^>N8JOY~*@SoSzD$5CHpi-#uXBp{qyxnGw?3VaYVo-{Hk%?}hu4sA=bt(MstR zP=A;CTC6|pjk({={j*QeV@sj_VDxaD)AwOP)6UT32q&gP>KO;YJwimhjhDb2n(s*A z`?HRAgwF5*s=b41$|=Qv53X%)cwRrY#F@DdXB?lr!mTuZufy=PYuO#Z<0czwDy2e> zZ_c3~8D~G)6ys^PVa-t4!SSe6P&^$wxf=FPeVh}(=)z=2gK4H|T5+IZKby_jydWw# z&8buV3jdt+(}-kqYVXXv{4XTBo6vc=D<{{+zH8_Gnqyz=Uql#YWNDRXqHd_t`%BNs zrr`0W)={)efP_PJ(?3k&s^n1Ru}6e|#49h#XsKuhLu6o8fp@yQ<$EySm|Xbsqi5HVfR7rAUEnV+RhsVk9vG5nTXL$ zD_NY$U(i3AFg`dqs#xS2bSKB7@3YLYT<>piZ?Mp?x zVuNnQVhO%E9keENx&j7|6hJLdGR#y5kG7&pD!!bu_z@=uUaofIf@Q(W`4MkRNDi~a z*do2$Yo}3G4?kJ#myJj7jCQ**!{@cVDOK=#^CL$JvCqHPM+jz6s|)kA7u&gcTmj;R zQe;=aHl5_GvbUhe?CwLU4$zYVk5>PnzPBeOZ^p6Hcp%r?b$gF~_X;7_dp#p`EL-fM zWVmB}#v*aQj%D@a0%P5^NOwQB4xYnnW!Ua-iA@0nH?**vpN#37KLC&V8YK>}A53wo zGG3HaZLtr=O}!|S*{Ls5xDOvhG!E!IOojY&s0cYfHM8nz3=+a?4>}&+);{rl?*5ZP z`;vS?Ms%cH56dZ^edEEWv4vS%B1LNiZ@ZRkqh&=KnJyxkz()Fx5^B}cTrW3rDo zXt*Trz{f>DtyDXQz$Nv>laFHOY4r*%sD7WmnR1#8fW~Cr!N(zH<9S}8@8Cbd%S<#8 zXlC@KFYefsN%{1akoMrg;&kK$eHB|uj689}o?~~79V%xjEccR~(uM2f>&i2m6A8kC9j{RZ*XW11g(rfsu-1@MKZlcOGZqn&6 zd^LTshyf}cb_-OYBM*VUZ7Wr&0rr+@&l!Q`%_Q4rt!{7KrOkKwH;6Re z;}Wd`Dq~fg4r~E^{>^b-YPo%XwZD48g1%?I_g|6NY&E-lXdFBs+uV?%QMF=>q5<;eCwLMu zc|Oeh%Ml&Qj+6j{_5ID*AfjORmVxJ6Oe7GY5wV)z886MsFWC9g6c72A05&7P{C1QZ zm@%0C6P)oC{*-qu?F^S~Ban?1U5Qk^%j)+QEndd&pL8@FR-zOp@UD7A@Yr{oqUO^M ze0E#R*kZ;vo+XM&2@3kOd(}ht&z7-5I2fUL>>Y}V=%80U%27vS6}v5dbi08wt|h3z z4lJ=CEr#dT2)0MPyc#ZhpyVf}<@|~45BxR4%t5H2z7x6SY%8@A`5uP7o*pK6X{JoE@Yetvxq!DDJTNjmnxNF+q{dobg?+$@ z*VFOLnwuE5lH*xWaC-|=73WLLp5_{LVcgHb(mR;d8y5&$V;FpFw%@rk4SZr`&H|(S^J{p@x=Nk^?scu7ZZOa5Yb=Uw;{N;_RY zs>)?il9?;H&Z~iMBS+C55g)9!#r68Vd@!=hf-&7*hG>27_#HUdTCz!N^LZSXJ^a2K z)<=sFmJq~;gkSg{47;3UFK}Aa-`?|of3U6VB}gHicM*A@zg^l^Ram>ehqo4d2Nimf8f_=Yyoe=!nHk}9y8tU;nH_i@5HlDiQio~XfYjMr z-1^rFJL5T|e$zPFzU`%xy5kyX>HvOz8?e0VMw2)x3d6b9hA9c4L)Vi$d{S zs&wg#u>wA{k3%LGPVQIylJ^BsRdPQL-+-4~qh}vgc_|LIWsPl`Ab1dVd9IbtHEXdl zk)Z>s0^vu!qiExxtn(PbmEnABWUf59)IwzDV^+R=M3z{)AzIv&;F=cRPFiLWA{Gss z8vhB^;^Q9*+)4Mn-JG{u`l_!wU&`n)rwY5N$j3RMv~vLJ085B zEN8BRH%BZNBqBPnsczfD%nE*NcSkyNAp687R6HhQG4)e<%emCPexx!* z_N-!+k-XXA&jH%98wPvcC)F6~rBRv8Dpe!CG0_JR8$E8HAp}+!sUFo-ZD0&q>DpEp z8H99oFUcb|vk36rBMTa{trwB4ihrO8w*#j1MM1{)8y5?b?eWf$g^NG<3GBAS6$5S& z?bhD)#6k zVO`2!ueFLFPgL3DI|VH&?~?>IAdHsAt1gO!3w$$X%$qFPRr~Ec2Ibq_G(rv!!kl&$ z^%h>jSdPWM)rij=lWAW@|24K?f7E`tS-8T>k3+qge;TY4bx7yCpUwE1<8as$-1tj^ zLm{o5&nOO1MAgErMM;EBg!TJMdR&8U zSFgI?Wey%N{$Y^XjS8&8oKgYCkXMgJMkTBga0on%@{4dvx|ljyS6N}q&uCqaKy%^)AnH7^FreA z0Q*BlAr{etk;RQb@uTEW9P5W?-HIJfp#iHLuq}PwcHxxB^YAdIS9kW(JpJ;$G&DAt zY0j7qH8MY{cH_{fgR}>G=nL@0vEvld5&!aBe!(PsDVhCJ*xN@r zPJA`|+(C6e^~aT;FeA#=F-H^MV^62wCtmoI{(`9w;{fDc!}2SbNz>YIT!59D;?Pgn zruew<7Fs12wd$nSzFv19#SIxwVjZxDwfuHKa>NCGa~_)n9Sq(u9S%>(7L|l!?np$8 zm)!)+H+G}TPOdf=IA#6LqNPT^AkJx2!iu?=IJE3;&mG35VouWo(-6+l&Up1)bm?Fv2W`&VZkV5M zvb>)~tOHJ`7Nw8S{G9g_J}30aP7v#9zd!t(1Y2~tEEz%!8cS~EOq2eE`F<_fgxg(6 zM8~xq$ta9nMrUjL$aGvf`mhm^(gNH}33BI(ihgH>#8~V4G%{dCm)PNJ*o8iM2r0`W zUklx5!Pmi7lff~kQ@YtJdw_XGs(b^^Jarfdkb^O-xQAW(u~%vz(~h$%=D$2a96cpq zMZx*BIv);pNK&ldmqUx6hrfO52|(ALP-f4d5G1r_K7kb(5g|q&-hy(9ij!46-t19IWJKl5-LYFM=|v>j?Z z0GKuU9^Ss3aO#F^2K>%Yqc<@(7(-M|#EswaXgPQc} zgF-VX>PeBWY#HmmK8s&-9IbS`|A)mI?RH4VAj9d@w*~gnYU7YsnbTV=e?~y>PMJ0# zuz3@^BLJ(A<|i@4AZ8?>K=_6rUn?i}ikIN#-Uig0wKtzZEEKOi8Mk-IiNZ2K(P(61 zJt!Kxm^EZt{+m(hA~kw>0D{9tR|mRmfjB-qgpO#x(3g~Zbc#qpq2j}V;noi)B4-G` z1JpxGa1;3WfgG$c6cp~gjyTGCq+zuwNDSEh{G={1g5{ zL==je-9k|-L2&P$aP5;ymIVvto4DG;z6kGSve>-7H3A^3~xETrpo zEN472zeZfzmQ9(<6IfA{4!Q&P_r=E<*NE2~(QO{F2JKcyw)7u~t6g~du~P{BJ1wo> zSLldehU>Y~V_^xTra$x?c`g`Vd`t#m!i<`07|^uN0kT%6cI`ET#;-vIdrWjYLnW9kq+0?SQd zn3W)x?N$0Xywx&t%-~3)C%O`UK^Ys6e565f8T|qR!}_dJs0O~O=~*Z~`}myEU*b~z z*UW%U0#+goQNGZF30e6_p5Oy(-$2QG3s_9Zh~XtD(#O#r6tQ*kaB20Dwc0t%%1AYow{0_Q~9A4 z-xIT0kUHoW`EGN69vA1S2ZTOo&QqT}8Vq%@$)en#({+7lxY@!~@ICP%(q^YX>|6hL zcY#Tzq7g}?&hVior?CI%xC&Zyb52qtvVZDQYdG#W%3<)Cq`rmZ#C@er7rUfnIW&$^ zK*!AJYKlQx&di8Wi}ufJcQ=aTC^lD{tj!Nb1Es9~rFxU1RsFf{?cdvFIuL^8C(-%# zb)f6?BoN&6xu%?a4N+4J zvjbL^5qJuVJd0nj+Syu*uIdVJ_s*`UhBby+8@QU&J<%@+$NMDCglhDa ztj(aOI+w^ox6=szvC7sQNy%{4W7C&t+_E)~SN)K?y%xgl*d9(+Hz zWt;sb-k@9AwK8$4N~UE6PPCP3Jx_@=SKRuG+zN8!H8~mWBqNj;^`cBWleMOt7N~Z9 z3R0*BpYClb3J4~jt!6?lt+Lo1l2{`ZaB*2wxpAmTHI@{>MrF^k`J^+7%D&T(t4G$s_$l9ayqK&G*coeYgQE-r8);_rQrFIMY$oglOp@8fnzeFHf;mA(PI5DdxLW)Tpq5}3y5n;dhjZ}|28Rs(0!f7w`k@<^V-d; zM-M$<+&Dz;LxzKfM29ZKVVVF;UBB`96jmVABs^7TUV6r|({<3o>(WD;_pEX3cUZ z#}<8#1H?131T?-vL@%5K=_m?sC*AGnzbxc2-XCB@{QKgyQSv9FJ$ecDLOImaja?Pzc; zE+;6@zns$(J5G|OtyuFU{DvMSuL1#uXjvQaHVH#Do2+?QWo~TrQkrvqFx0~iKFmwt zla()V+<*7SJs%%#aIHLd|72gPloaV1-S!K7Vf*_gor*o%sHOc>FH!uFvE;2zA)B~J zHKlGS2&%+w_ZQ}CqPdo&o+D+0SXu1Tl!vwV)*cUKIG*J91aTRh)T!{lH&`5@$W-ml z+|xTutfXt5Og)OtGOrGN0x8<{e462HTa< zdVg?}<85iPtYE`S4Y!+j)RvK&E=Lq&(*<-%#bd-OzWw98$A%SUc+S6R@rMdh(!dgj6fVD5StcI)4Ta4~`Yk=8x;hjDkQ zKH6$=N*O&A+_4NfMKUGcg8<4RLTtqZIz9%R(T!(R5>MQJs+*;_S<>35Nj79~k%z{` zIOt9br%permd^M&LNfjtrBTNYP5T#-u;@Gr9ExN{-u^6S@j^m2i^=7fGl1f7`13v7 zSBa_sCN}x&U3!hkROFe~Ck19MYU%v%c@$jRBoIR3YLErOBJ>-PjQ16V9MVGk>aFHq zqwCx(+aKedxXCd%^T;4cp}dmZauz|4H%}A2QBNnAX3?YKX8v`=4;2&;Z;KMXQ3#cl zDjQ1PwMOabTP^v~R!^)xM4u>w<(0^~q1A>c&gT6Bnn=l?*q(=)#0f3pM^KJO+G7ZG zs|f`TtoQE_>*97=roD?KyV~ZRA8?F3b>)h6Y8qe4SL(beICKg5} zl?wajMPWPF1of~*e1&Kc#l|6m?`KSRw|jpkDs3Jzj*U?PF?-bL{O1-U3pRN4yfZ{I zf#Owym)$jwyd{#*0gX!y3TC`nn=_;@!Wl7`CHZg+6tHn}yR!{`ENzQdkxbA) z&nbkJXTPf#uX9U*o3EDPkz8H06X>a=x-%`oI`O)*_%q^Aw@HJ*LCUr(SQMfF6y^Rp zn4wnaZpu9MPkTbLF@v}NvRFvrO&?TzxodH3o-TT>x!WbId>@urcXr0VAOp>MIPqRI z`U@gp_Cwoo`k4&#!^^G~keYDkXW0kmfKOB$Khn>D&l(&+@`3ODn|pWNlhA z6udh*r6K618HnITqpj(cN!?wnQqQhwLr81K|0@0T6Nl?wVL@%YFww8e>cEe@s=R2n zI&nUp8bhM7{3VCSciy1>Fc_jk7Bx(7GW`4^M1`jb;pj>deY<&q;val*x!nOXYH+U-(WuY*DoWm z0m7{FbK0WO!pj+wIc1gifNdT2SU@9Y{ByO)@o*cMNbUaaHOn>|HyUE~yW-ke`w`Y! zOmNWatA$6-vN;r#RrVMUASDkot2B<|V%FrDv6@6*y_J<*@+b!HMI-jd;+eh^=vb{m zWhNrtJaq7%FR?Nm|BTL4pPGz)&X10JL*=ba+#k-vJpP$Knlp{!R>;{|*UPt|ZcgTN zbS9d`1dAK;Sy9ye7V0O0-LQtdatyz%9M9#evRK*%!?#W)T~8|*KH`?DLdl)baqEIN zz}uE{W7{v^h$x!0Y0_Xr)uOmGiA_5Mztp)aC*Bf>JzG^=kjbHNOYeJbvt3&n@mTGY z;cU9TxN6&h`3AbDbGy2!=cyyT83{m|Jgo=>Lz)@zHL>|m2!UH>zcLVgpM)UUE+6mE zr|jb@v)^?P?|;Mgq!TanBe(rnq3_7LoWwK98mrFJTU9jnm(2ce3Efhu>*8HNwqi>( z8RBaawfjNz_oQ;pf>F>z)Rg;W@)wiSE4|FOaM3S{_7)vydZ4wk2C$h}S^^B;ElPD` z2PW%*?o_s2nq))HUw5{Z9Ve$4My(TBnVVc9kgi8t9NjP?{ZHGODu)^|)zK#?##VmF z4iQ)`c%rQI!0W}=yT9`@87N`atB4bAGP$x?moNuoexOzKL!e|8yPEx~%jKyzuuY08 z8Odnxn9lPHa~QP5b72&B!^~|T-vpmjn=|SQKl$R``?={6VejJxXKU-Jk-xrC4R$>& z->vVI$@R}YFr(gC3#wUlG^G`!u9j|gzIX;RC}3-JG%i!-X{dg;YgcaOubcTfE@4Q( ziDz788!QMDRa`i!w8s{ed=zw1Ku3oKY;uqAM}@NFP2hOC2Iue(08H{L_#wX z&><#(lVw1k{ljJ|}rQ;}zp`^&T;y@wjYKQd>h=D>} zI)B56A3bqDez5`n30tps?}hOYqX@tkk}0gKjZheHw-|j{B}>`RW61suq&ueA;W??i z#SdH4$Y>|WBNZ(mE-rLj0u-<{$2-Zo`r}-C?BD<5?W_Ny{DOZ~Qjtawq`OnVU1Fs} zq(d4BN$HYW>FyALWszS&g;XnUmER&vVd z%w{40;HnCoQ+=fpvW2q3__&Kh0Ik(2_gjO??Qi`-_!Zm9WVOqg=ROLxuSKwQT^_|{ zSh20jKk@nsSQLLz=|jV_=)94)Pv+y!&#OoR0Pm{gF5?WAkE$y9LCt2tv}xxr&jxIQ zHeP~7OK_P$PMGt@IEsLGUmsZjU};{CME~Se0560nz2NwQJ)ORC5M^O7K5MbNdlMHb zR?PNnH|E%&vwr&XEhN!47YNx^HZOYt%{p$I9A^skkiCS8np)oWVnHm66M890!_FIL ziC8`LmXg2NAe|Du(eXNvGBi6SiQatFo~)QnK;`tjcFVHqW{a6&j(v$sRYlhSqX!rO_x-4(zEiSOUIT&(XSLN`wx@ZHT*Jv8Tep>~0%QK6S%f{Edm|`2)bA^QSQgh}#tuSaCg@lAd*b z9&)@l%JKc?s={c2rLS zyiTRFgBjL=Fli}y1=vdX{eJ(hN*VR_(|ueBxxoyNh5F3zz{ZPf>XHeCXvXVo^jhMe z-xOhy2~a@~Kw|tRbnuse<;LYw1zq6r12U$HN$G*ao1{74`!`0u{Hp6dmOe?G@U(71 zGJ1m|IIHupjZLQ0DA>uzGg6zkiy4HtSutD6bvyKhha>#&B`65fsNm{W+8IenQP;K4;^mnE ze)~d%T(f6(+Sj+{g2-{EhKY?ROd-N?8r7*_9xtmS*#=_}_?5oz*=yx9azj{iMnXIWJc_lUk_dQLVv*HD+(xLFyG>Bt_2E!&CVzNO15 z<$5q;-y1eo@IHobcf>wr*UU1h|D&maE53R2zEe?dg|dfC?-TOio+dMaKj)-LoJ)w> zw7m^X<@~$(+N@zW1}Z5=52Yiij)R*bZ&d5!&%rGjxL);^al1)@1LCg1diytfB;QjH z&ERRXl~S{zib0z^TWjO^ajvmL2XU~OzSIwrY_)@JDyfV`nGnAkUoMjpAOa5(PINWzH4cYNd&>~pX4xH7qu{`=0w#XoY z=n_wi;cG5h8RfNjZGm_vNT;9UEZk>C|NB9mnSm=;i|r=3hHWjb^psBHu=z0oNLh>~ zHo~toJ3f7?{y`h2fjD;bMY$MnGf}iZF>mua#A*h3wmB#>qnv~A&xL#o?>a4CG*LW} z>{Yg7iZsXWkxan$CZ3j7jrs`po%W9pD}^o@uhGe*-x3fnAuz=??KDe>?|NiM^uQp* zh;z-1T=7dhJB)}ul>9y6A%Jt>VnuwNNr$mIXoY{HA!}V** zfVE1Z+x5yjk7=y)dS!CT-B6_H+X;jO_fzSDn(Revw8ylK5~6;@0)ZNzIAkf1tQ>8E zVM5H#ZR+Ouq&uV~;yMGXB2w?Ecg8tyB%WRZZA_&P2`tMZ)K9G=il5gkTB)7@r94Z5 zS7nT_S8o@?lz3~mA#qh(;$hHzN19)1C_3#2Ly1`a+Id25-4bTl){ubqT4ik38uIHN zmMNx6kl~2pfw8bod)2IK(T+DJiN(dgN#hX9wN|NMX7N#_-MGQs|g-m8hHP0HJL!1ZwAruq{$^+u6NO%(ol5ftQpG zx6L4osx@A*d_!!tIWDbU@3p()Hk45}OKY7224b8#Fz0@DwB0SqaNqmRK7-SAOJWsN z&QXQCIv&P&aTsVq7Ej^SJHkjM|Q|5e9&U4bueiJ-a|`}8wE?Hy^1Hz%Xm)hq0|%vFx_;{ z`!Hu&M|>Mkq2WF~U<6`LKuVsM#)=ZWZzy1UvHy|3yZz|U^t#q6oN_Jh20sO9|2q%g z#&y}(0>FHDOp7xGfoV2(yDQY)zpTYQHlZ20Qdx@|Wi;z91Bmx1FA~g4u<4W+V4uA> zyT0Zfnr^*P%jJ=H@PT-(^7gPMCEoL#pj}i3A@m9e`6Gmfb!pGvDxl3?w3JyNRzR3& zGw^}srOL(DDvf-HqM#X`O-}Z?hOUDf zvLAcikpY06q>XO`>9;8S%Rwc?Je1R}4-()BO`K(}Jw_h4^m}U8l2pWu1)!#vjwI%9_P&=6sS<@$)C z6ccEUXX0td;E~BTx#kGZB2#9i;}X2h{y1Hq1dXA@QiR4K@?Df+dW&*6MNLaT`R+;U zKZ5M|2ekgLS_;hGONGA@;x@8bKr3KABnLZry8Z4w$wvm`fZ`0WsLzk0`XNAEkwKu{ zksyLhqv5_R&9RM;&yO+RE})8tP-O7kg#mHTM#xQHZ9ekO?If(>9NKEYOqqVGc3%J) zi41N7i2?lWrYW)7W15ZN?zkj1+S5i}Np!&46KhM4C@_OD^0v22XFkKU>|*;{b|FNi zj84fdyYf#|*WtdP;jV@OgJcT@9(bpf)>)D>{Z&yCbn%B zc}EHOYW|cH#TH7h(|zkRFcNr#X7fl~9wL^(-mqS_?BY}vm=VWxG1`oDC3(b-(|IIH zz1tu+x^oQO$TXYFj_%@rSvI;NZ`Sk|?i23v{Ph~$o(1T$v)TDJ-}k#n?xt!-i!S|- zS&ucwY8L??^XSdRas4ZQ_rn&giR@g>Dm1glxn|O|0+XT{n*s&5x2*c@L z_)Q(MdNFz&z7e}YML*tc3~?WCEX=&tUXZBa)`24FEM)exG!6-ly)r!4ojB3!X&6#q zExm>1big8NvPr>qL`sXri3OhquD&W@KO zp)5g(dCPpmeWDpK#EY9p5PRJ0ct;9iz){h`9YlRTUZnh$JpP)d0(pm{^+%OYlGD^9 zk#xwe_TO0k300FnDz6ob(#KM@G-b%t(8q_xc& zZ3++i+jI`JV*oM@xQ1X-9#KWM>>3*&HssCDuTF#LSZZ$$OcM&8U=V^IvOY-p+3LBX zc|T;uRJkttw_TlvYu2Rsykh+hozq;{_@otKT?rqvh%PCE&9+J5o37kF_q=n)GtAc4!> z^qR?+^#*256O_^Qe!q(*18j2;mb(r4!1zJY?HC)~aYT|_xj17P_)bHVA}ue1qC!XE zK@3^k)IA#&6SDg|xD~nz>Ax@Q3k#G3xOjQv&MF&0>XeSG9kdyxjlZR?YtBO~ImM8+ zPY;9hZx?UTC-5=Q&(Yz00E9bh;`YJt1Dofb>P&WNU(bYSMfzMhPJlOKG+q}ktUH6+ znapIMYRS@Ic6Dal;pX)^I(ByGd%=u$k?77k23#J_3*RaM3+D^GRFf9a+t*u2? zC2glS%OmtVqJzXxw0U*Nfx(ibSW#<2(%4tv<4%#>j!1?facOeobns_=gaQNk-BE*t z@t>9#9hq(0OjGw$I%8{9v`sj7q!6+50_;l(UTKX<-D=ofMSUvDcNwE<-HPa6J=@Y~ zeVmwcgFS~5Z%B?UN3PBM8rFYJCs?-R@l zxcja3{$3ohc;X69?AlcZ9gr4CqsuaPn*x6`?u8j)NnH&W^`W z9&+ETxhA-78nG`OF%EtmuX76(Jq%UM1tiG#7~#4l0J zP53K8jbvT*qg_$BpA7;j&wGkZl1ESVGF0|dzwcIhoAAtW1UOgG@1z~kyn5b-* zb|Za$88TMa-DwnyA)u_OPp7_vew@G?Unr-rC29QaXC2UZQ{xq9kJx(G>QiN~)Zj&c zldd4GGl`&I9+S zH^)t#{8?mToZK1K2WJ?mbT(FF6XMxoGTmARd0xzBQg zl4LK$)=*v(i?~pZW6M9D>WL{~#bJ$ZAq9!11I<_3YhWjDMtKrhIkJu>h6$$G&~2!Y zWwcIi!l(+Kg+9O864+CV1V4|WW7^K&I=!b#(JfNjLGeuDySI`k#8ZbW-rDVDfu=-I z0%2G6xE=oB_M$qi>D31J4#6@+av1dbQEvBF-UWRMDM-9gZmQv1)8IEYFn~? zWPJRc@ejh1ahPnz4&EU`88DpT5WX|Mx0(5cP0_71lNGT>AML#nnBoT4p1NPCUd{=0 zoHr~fQ=XnPGJ)m)M>@Vh?pi2`fxHcvYLSr4VUDew5fKij_jfqRC)kOJw=6OlX!Sfq zY)j7J?Y?|G^spR9;F>V836aYvPRljx`2D3ua>z(K37|*$Dhj8*oDs?Oj^GdQcK!Y4 z>m&WWH{$8UMaQIFEn#iji7s^5H#|@F>HosRabjc#;P)`b+tw)Qgq`)1&azkTdk!DC zC!PDBk=-Rrls~LY6_r_Y!gY0IH5r&Sx6LtMQ8*iO{~7Ha@isZzf`A=XN2>&Er>8J` z-5SE0_3J`(b))u6+;SWU&H?u%o3a%1rapr^i)@YOrFs9W^W%0Iqc6ElEg#lA+Sx z3%Y`PH8zR&)MqDBx9KLpS@nX?{Dl|Cf8$`cP3uesmK61ZR8_8qA_H=xfV}6W$vMyr zC78y*cKh%iUlda{E;5>;gZ87G-TD*U+Ui5g95C-Pa86 zcQ=jOiNne7GZMzH?X~XJ#_ZVSGSL$DaqkS;#BxHkqD)vThG@GX_aE0I&B292<%(DX zq}l#5_}u`+7V)9Zq44tsQ@4aoIzuUdbmPXwCe(=h=<^=tVE_>XG&~%~3P**$iGk`r ze=ex*uyTJ|Wtx2Vqnv7zsadT_T#62$T^ z(f1{7nYd%pA_;wtY+22kj1O`&J`>$AJRl9*;p>eslD}|eLtw+1QfYnI;$hv(NxlAQ za^|FF{!w>S0lX$UH&Nsz8%MmPghfYL1$c+$I!z_jO1;CykN1x7eMW3Qi-aeFE1=fa z0s4jSMaPw;W>>y9-@xZ(pJ0QSl~HB^|0pSU2+d`$oeOMyZ|A)kp1CtwcWqvi$k<01=SzXK%kl%G2*t;m!EMA(=ZxdwI4wvN@J*$)mLCgWGw> zq#T8y*OA-MNI6Y^=ZbLzU+NFLo9s8^%-xX9DE|A+)_`W0pAYZ85Z~mUoFdUbu1ag~ zL*Cit3LG?4a2t1*eH^UM!NI_z_Wbb=%~*^%2ivbRk$Jm-Gr<=b!ZOxq1$~M}RV4;O z&Pht_w%f4X%~eC2H~TXnvtIUJAaF_b6t01Zn#zqW#>uQS8KAVtz6V&Bn=k|KB z-^P2{;VF(H2gBQsgGXiT8Jc%B+4{Ql<8kbEGu73SMSX)#+E^nt^ytv10^Ec*a0DaD$il_Iyj`aOZznpeW~!uUJ2~EvoVT%xIuq^_s323wbt%8gf^Fvnpe)!nZttZGe@)@ z!wpJqQmw4$XcUnM)KtrL`p^@xm}lw;Fs--UuQbS-xAWgKGX;+Vs8&zi@XHo zoiIO@9Q;eSp&Jaeb#01Rr1}}wVPfPf1WvPCJhJ-9d*;~9}PZQ zFC}nNQ*a!6H1Q8$h{VB)@{l#fGbTAYtp4D=K_F5N{V}BuC$|<$NhE-eMaVQgubK5I>Q<Ng_UZrch3WUb<>& zPlcA^>E(<$vz)@6&((5eNpN;9Vzjx0gd=1Jpc*<{Lch8$g=H!YxvQrsIv2G3-k@@@ zi$85N^ax_+TY)1d*~4Elcvw`%x*Bh19|(Qx^(A!Y{8w;^g#Bj=zMcKSK+GsKNQXcc zmx9hQbcb!6Yc=E1?#6zl=mpX>Kpy}~tJ1KC11L5g@^%KT@=9ffi_Feehv4PP%vH3=D5Wwd=K6#K zQ*}gLE|PSt#nUg5lBDlIqv@dz4jdN8ekhg&W$di0h(=IP4&+c`9p8OXGQW)5Ie*Ye zvITiXD<^53cs-`mnG_N7xYJuzB7hG2uUls#HR~}MBpAJ^cC05P_A+7byW7ar{UzUj zKKk1O^tXA`&SgA1)uO>>b4T8AbSdCwpNI_oojhwVLPoLsV*`9}^@(FD)JN<{ILw#z zY<-5~i(@ST21F<4AckF(LVLlAdR);b4dntBHjJ5#>xIPVgm!ih!3;|h$lY4y(v7fN6Z6YlB?+XGw4>N+{v1Q$Y%15Z$d z8gxH5e6I4;q%3)qNJ=ly{G87#d~vcfMAm z?iHe#Et=C`mkSRYB*dj^8r|+##;(mpkoWk%A2)ONgDz{j*7Ed)4Z{0aY2V0X?ep#* zkhUxN0?os_ETM?I*$KhNxjEr#+bP^4`jf4i zN+-p{!(z5;SJ4OnxWWqQh?{zVnd_nX&m!v>^yP1kJS-^nm)ex4HrK7Yq%X-;nIsMi z2Lr-Xpv-6RhZy>UVK9IlRPT590j4$fRBZ%$<0*!}GkDYa4LRIE@_}8OJ079O^|Oe$ z2VLg)>4w^guXMKT`|H)N&2TU%CO&l@`ko$~>Tv(Ua<=co41v!HwPiU+d3vNwje=*H z=Ea|1vpT2nk|5aU0!@;wKvX!`xIHV?F#tAvq*Baw>1x@;v*V&UsX{r!;t^`BZ z1K}h92qX>WVZJ8uI30UULPwvUwl0}RIEHn*;*_?yZV7;x$fA%9fr zwrk6*kY-SzNo}usLR+3*gW$a`)XaBs_`+_-Ao@#r9=w6KYPF>a;~)A#JCQmqkU!c; zV$8h8#x#X&6|uF(Ounc5#G$J4Xp?3nl@#c|6EJ?}LbmpX@A))jDF$t5!rCRLknh?)wu!YQ3z*R?Ag@-y%=Ifd zaFwUqUE>Bn+}(I7D@Q0}YVjQrnRWJ})#q9oKSqvf?%osLigL z6J-3Rr_3%78Fis?DK7p7_atwPzMJiFK{LH)pl~Xi$|}$_OsUs8oI)NyCKJJ4cjT7& z?Akg?0Ygj*I$|Wd$$1j+y|7E0%T$o+(L12SB0-Fn zf~?%Lm#{omu0a}~6Y(fT;f})#=f@?$gob26@&Iwx!8eJz&%iQ?Lcr*|9^y>`RCx96 z@Ky)_M)DLe`|aBD9RkL7#JLoXVzvd%mEpZY;A3%>Ff;6Ya98?-UtBXuNBHtg9hO&8 zo(}(zI+{eDk}l<=TRo{vgYR>KYpFT{^?m5@3kLq|yzQ*oNLwZkS=3;<^+9*37<>P6 z)aV22eBuvR4nO;1@;&Es6@_`jRp0bRU#p#1`583OPD`iNE}ossn$QQ?pi39^$E=~n zZcjrWzs++~Z_kr%JVS|3>1})Yw5wv>S|7J@@JTwQjulXOJ8<9H067#_W+-4v2wMN! z8IgKU9ZuIT{jotciqW`jrrKd{e&E%wcfa!W6r-HxT*==5*f*`XdzD7%%yKYUHzU7z z^`>Wtr7R{HZ%cWQXV)TC!=26yG#&f22+<~^OZCMT7Rc|#zy?yXh3QxLXKzp)+mNy; zZcbt!e~$l0mr;FpKO+h;U$Jx=&pA;qZulyobRKRQo>$mYJ7STJ*Aex*;=NvtSpy?w zQ&)4I@^M9Q{I<&D`l^yL(X`DkQNv7B-QGT0Ia=~Tlb2WwEW|(MaN-&`06A&AR$GK9&$dRFxLjjX1iwS$aAoY6KJiIOK!->=xwOok7cxOeod|6n*(e6De!Xb(TrnTZ~Cwa0-e|M^~=y==UI z7Bq#S$_UGPuXrzV6+clR6_R8sfiXHhnodXqTKoP&zS7QXiDlqJrvfWl!`t)MJDG-h zZcz6iTZ>Gm{bpGu+Np6K?!ZcYXQY`b9zunkfi3x&3W!AnErpC?-pw+W?*imp1?9 z$Y?_DupkQkV^!$$)R_;C(N^zOo)67BU<#-3a2KREWqL|0ep1T=QbN-$GIM#xxwb9u z1}oIZs$44etR>(*Q{dXF$0>uzu16`*R!)-vJo0*ZGY43vhi~0E2FU~ z51y7!$%P*t2n4pa9f`!3RMn%dX{`E!bxppZw6=rT5MD0%i!3s>MWz|=tu8W(u|)Jh zg7E)F7fib*NEY7t1x|4Zi|S=8zGwF@UE>=56sBsyF}rc@*dKYy4IHdx+i97) z=6r#Njyicv*^{c&OAWM*+&L2kz6&yP9v^`s2OvCCQ@)~G%am&uk02)8lMeBer7Yk+ zO;PQlyt<~&=AqQxu;ZC`p!XQh30qkz|Jr2TKP_)EnZs1QjKqd3T2DGZm0O`X_n#2= zz@Rv5P@BI6u{zT`~*3zVOS|qQhPq(_@}~CKi}lCtIww z|Fg|$cmZeIk$%K?n{&-m=~6m>;Npd+CftA#|EO`;o%QqJM!A0ypIDj?l=$SIe^+c? zzh8}h2g_&m17F)X-@x9%9FKUvCdkSOBkSCo51|g_d}ZAFpo%&Yg=f;RNq602$D%JT}XJToNN%WnD{ zN0!gLrL0pT^7@+|r(JDW*+K3CK=huJY%q{x zejS<&=D}Y>Azf_?-Em#wixPlMGM6=cIUX}O?9K@a&%Bc@5Nc{5EmlH5)pJ!j=8Y|Q$o@=< z)_Jq=g6kLk#`9%i2L=AmZ|2|@LRy#AnH?!~LE6;c3@`KsQsEhW)tSM8YOhbD8@9hU zK*ub=Z3{M=`EGc(O2*YQv+{27^H75>NTmE<8aodbNJqhi*0%P*U}nSi)FXpH!v+3A+-uW-oR@mCkve>f)%6EHQA2iUku>L8ec0c!5j=?HCbMww@ z&L{&RYIw2!>iLP69gq&N-b2@s$x`{MxdAS~wcYA$XUW}~4^*yM=NKj9yB#v`xig&; zqETMv)uuKT1nQ6?D^2EBWVFY}GK0B3h*n&WRVZU>_T4Np?r{!7M-LPK{r8U3t4uxp z-ouq4Ou-eaEkfp|sloDxf!bmcF7_;MA2k$GDa~Bp;3h2F-<5eM;3!hbcCdc5w0x|U zlo?ELT0UR$Y5oyY$0`Fs56?1rE4*%!SI4yHv;V;*^Ae)T2$!8O3GSZG+P_kU57HAGvwsu+W&!;1+Wk@!sXEesq&>W$hl z+SGg}>oEE^b8fI7vpQ*sbPaEfGWx)R==7D%i#PG4NuF&MeFFVl4Gnc7e}Sv>{u4WL zSCu`=2Y0?wLnd=HPu!IM_jsi2PY%w09dH9{XD}lbf?%+#*`HN2KKY*_h%#;^rscn) zqouXvQ!-bT$6wg$8-$irCC-8J%nu{-mV-l>wyibQ)lUoK(zSAL<|JKs;{GFdmRG}; zWzwv!DmZ3Y4A7(R<7UOM152wip2QU28B^b*+D5U{2}@Ji(OJ)d1iI4bg(9ON>p+py zWA{LQWb{JM6_@Hit&dB<5`BFv_O$`>B1ucTPf+R+p`Ax-3&Y%NOV{MD`|e|o=+4a0 z>*N-hd533(MRNV&4njlbbj0<|@GlBtgSb0wjLyLTcUhn6QX`u%v#?J~l0H zm*G7OXLRsmOR&^1lr?DcxnauM<7SnPhO*wXM*Bn+rZ0zbB=f@atQktl*JC(dm0Da+ z$AYcCls8ViMi^91buTj(W7@qg|aH9lL#F2D(b>jA?<=(&e6|mE;U&WP~>Dvxl@*2So-uT-*cV4 zG$j50yY%=8Z}AcX>ts&nmxcyZlKCRz(Mm^f-%^7@h&5tUU*(s#r+lWUzWQQqmrWxM zvcD4)vZ*hGP1fOx2URBpe_7)>{MTLdB zOI6Oh(m6KoG4<&*-G;Q7Gu6l17CCrM9Ap2emBC)yy5KYyxlww4a*0_uzk5p!W^+|3 zQeNKNYgtTMGY6=FEVxxBz)Px?xq+mGVfs)cRHKDKCZHxktGP%LZ%|(^qi?B%`%zlr z0t!LL(_Cb+*a2YJOdpeF+MdU~7~)zXusbF?a-9VqtUq95{fwomS+EG&m`0bC{%Tmi z0d>(f)0Ui}X?$`MBlR{g$B^;Qq3lu}pPaH`{vh z^Oc$e-B{zec`+>bSlUuW&9LmmCx`D>QfXn2L+ITsEcWv$TwbtwPxifl_u(JqZiw@> zJnzB&`@s57?;7&?@qg@Nzi#rTQ@I0{XEO4Fypl52htM4iQ8 zurKRXNidB%savWA)K^w`R5^@m)*JQ-^LP2mT2U6rPlk?6GHl#WK+{Es zy;nmOf`ZDo^9}21iMkwRH>2lX7X+L7Sag2xD;td7l_S1r09IUd-bmBO&QUYC#XRs= zMXf+-Frv*ouxa04Jw@Ztb#|FwN|1pD6qSP=PGANm!ZP#8PQ`rZ^`usKgFhr3fQMtW z99rt~b_X`fBJXC?bcy-TqK2WJ%d#W(nm!D)w;!=yXHp>vZ!wd9YdnEzsKk6LPOFxs zRmj1fswr54rXQ2NZ^Uz*jW+tTbC3U5UJ8|!Ft_=f~J#yO;)ii z)(%knsu{wGFEjP-YMz$?5fWOvhO}Bw{y>A*#ma7oD%7!hw{@O_6|&zZHNgii#|rCA zbI>P(&t5ww+hE0izM3C+!6U!SUS)5K&AeHZvbeCMEWx)MfUGn({O;uV0pXKbsVRfL!)bo5=`*l|u#kc(su$sXM7ti?8p?!KnB3ey-xZ$=!tmlMGE!Nc+Qg8inTT9nEYe|cA zXZcQDU3!=HaubMnn6)(auJZ%XP3znpngu@w_y;uVLiTan=zkvS-v2*Ejg*)qs9LPC zFC-Ih<0?HP+wRiK_+Fvq{}k=vG2g&4o|=!zpO-!K`-;KEm;w!9#8`~l?1AY#9xLws zRib*-W!*xB`b%QCb=a{b>`p+-tXY9`_}4G!3;HjKFF8){{$jVR*r4K~(Vz@|%-6tL z^F4tO1q#1q&y=Cmar_sigRgXPFbB5mMVAEOL~g zy7!*WT4(10)*zUu3HxxdyFx-8*ayZNX8rT z#qLyMgwj1f_m{N{n*;=P^BAp{L?skYtP;DI2Wgi>h=#ugm9_BmWq<+xl6s@=T!kxe# z$d-$@*u5&6XSCQoYfT*6=FdK4i?$&@EVz=p`5Ji0pB2M<=*(bKEPG&RI-zv{vv_(6 z7mmZhTpOuH$^3u_b*%`C@dQ*aGGAV6t(VnfMI|)S4*K4@6D<6D*n_t3e=RG=^9O1U z;we^raw!f^ogOv`C+>!P+4{#~w!(SP{MCrH86Zx0_w$m2U}0M?$91TyXxo<;7bE)<(+U z_F?D)o56<%ijnM6#nR;S)ec$(UPP^wgsWxQ{E88&aqD18-`-eX>fXXMX@_(LVe1V zR+rNC2{I@&ipeA1@V0J-5=d)?=!kX(D&dVuTmr8ND#vAG(H-IBK6Im(VvXx-((IyV&F%C?i=dyTwz-{Vl0lrx{kIIS)Z+Ua zc9ry#R6RRgJV@1N+{(Y z{?fVX{|13nLfy)FRR}@#9TT};;Q?S>-KZEg@l#(H5vix|OJ z9>LxfgUfO^_|P=0OASHfbxYqNP`$4GmV~4fa%?dU67X5#83Aue!d`x^UN_hI<6o8V zAv`B^;3rH^^=1SL`=hZ4noIk!?8uXQTDxGT^3q2$Ggh`)=f1b5Wk=Xu=5cPDWXNzkG?j}12W}x6fkj4#9(N!(;Uipy7$9f7unSYOs_mEq8 zQe5F#tJ)Q#6^2`13T-u#uEo3{ACVsQ3pig7RkZ)UnZg-gN`1^I9 zlpU32pO4%__54{dMCq7VcevYApj%fFHT6p!Nz5Hy_u`++%VOi22VH#&&GAHDDcd?3 z^i3e4EM(q<94J(Aa!^tG??7`EGlEtwlzy%eyZWe& z!aQ)PMCK?`9kA3Z_PalPB>gSEn}pYBvhEpff3kl@9z}-D`!){;IeVh758?~GJg0?) z&;+uEIZ2BWWP3qB=2ELKn>hv6V}ZA+Q%&a1YqbuIt5Ek zuivSEYg-GMWk4haFQgmwQH}7y*slB223|(&6(mCs3zQz0)q|4V9xk^Tn$w;=*d+p& ziHpbhu35fT&~At~LCE24KJ^Z}6)@6HS{arDy-0d%MUs{G+M7Twlwnt0+r?1Zb9CfG zppMr~ICN10B_wKlN`$q2KPuGNG`R7GfWsaiZ@J-$aOn6u$W3qhA8kguGiswY?}{>0 zi?;6*yH|!|k(EZ$3NbGCi=28I&#+0aBW`fMt~s6_T}zo-HhtOCjJ0}@ ze9<&H$8aU6W!^>=hjuRF zrrUJNsWIkyrT+#kG0Az)m9)09EBxV3ufF(at@f4yDw;4@*%3x~sZ>?_m~K$qn1`#- z!;;D?<>-~2l;hUldws*#=<7@QwJ7SVTS+J3CN$fQmC#t`Xa!2P(10=Bh5)(Y|C<3{ z9`eUw%(Hwn{mE(avDVA`!UU2@^}5QsB7a2ZXg?$k!h3s8Vv<){$-%oP``S`?Jz zcLs6PevF@;3a0bMhSE*j?>G?Sm`CN?O6;;p<;pO@4O9J7*?

D8P~*A&t#kmo@?5zb`)F6)dvj zR#Q>siDiSWnNGpzeP*u|>q8gh_$2T%IcH57ugI2ZnZ9pE42H0;;011EWuOc=Uq6Cd zv?b2#8y;CuMH71xkCf7zBeE^nnDA^%lF#%Jm6kr21UBzGlAT*4?VN4sIbiZjErb<)wZ>OIMHQ@{+@q7Rv z0w!endDeTDujr}4Rty!91Okvb`5W2bZ66~#bAbtfrec~2_bw!R^OrMXT`0g?6!&g3wFJQ+EBnf!z=yi(?HKC78Z*YYT)H2%ClsQ z#COdn@7k2>;!lccPGKMv+QU69%P_cJE@YSJHE#=D3k^~@GSp9+8Ba0P747QM88LVfDE4m^VOd(6Wp zR8mp@?sdpFgzBWJ(1YT@3*U@&i_TsKRVr7EB}NQAbVLKX#RL7~ih<&bAq@se z0YJDUcDVYXc)p>bSxCfop^9J-ni)=J8zmhKgI+(3!8eRaIPBUL=g}31eLalpJnYSR z*yDVd04SW#E*!2dK(`9c)&mm{2`8n36Om};)`^fn#9vb*23Lq&f1%RBB9H))&`@W1@SorsSfoKGLU&Ifi=5P8$e zGI|^Nkwjbx2KMn4RO5*;J(g{^w4#p!zqoTQAF`styCqyfu`l;jHsjDw7C0{E!h!lC zUXlJt3$duTgtk7>b}^i>>B13dai|NC3PdiHJaH%*|Gac$Z7_atYOG>dd{|p-0b(5X zL3~$C)Nl;6T0fd(D6Uao(1HY)^^B**7u?YuRo({H4~<{y7V0ArErMAUlf>1gC-9yI zW6>r_XC-1HCG6)|qR1v)fN)Pi;+kWL+v7=Bu(#%=xWCpdf5IfsS&|?0EnOay@mrG3 z#=%?d4DPYo6g)BU4mQ?Yv6r$WzGE&Ru?o{N9R2sh+S~QETbAEbK zD>T2f7W`zC7Cwa9v=&RL<*y0c&(`e-S~m`Q^mq=EU{nBy8j) zJ?Es5=BA0{W*FpV`Q_&1=H~U}7Hs4eJ^w3PL5@oxuPs%^DzMjKGtG$qpP_+&H!|M> z767sc|Hnoq7~h@lU;WFt3MSn?eV(%af7Uwuf83L4IaT!jbLW5T$^5UUJlra~lB&1) zQ$$#ir;$aYcRvhygcgwT&Z%VHV2-ev@DO9%r7fH^Y#fCSks$5VEj#D4d zGE%*(-W#%3YF~W#{IyJRd4++TWaeJ z&RZLkx{tEUc5T>OZBO&In)->J>#NDb$T+-3cm3--x?9lM+s@ct-ncBqu`~88I_Ed$ zZ~j%_EMhuTJS%(^y(sydF|52*(nOTd)PRv$?OJ1g9a{>7Ym&QWjtCMI*&K|X(ZoxZ{y_M)v1eAYm`n_H~$IO;?vVyjG zJz{$;>x}Of`g^_n%%tz4MsM5SVU+mOvy)#|>EPYU_u$@z-pp$#J~n~+pNH=L>Hghm z@|#hvYGL(f#GH4MZ=G|*ZYH_l|BhOd7;UiW32vBwqRE;vLZIM|G~F%(qPS9B3}g$3 zLb(Jh#KI8&0Z1n8Prz^WiNk`lBBwe{XkEVYi0<-|ivkEQd*A)_hC+QY!?Vlx5EJJ65uxfr73(si&1Xq`}C(dEaDh( zZUSL_A1ETzO1?xSB+c=ao~wf1DJ)odJz?mW;@S701Ok~~{Zc|?NST@rBGWT1hXKLB zgit^kH>RC5a*jHJ!Ded~mYp=G&qRdrI$aliBnO@QRQ83A7a25&p$n7d z@4rEI(B#Wo&!`e?6PZCk8pI^hfIfz`CJsS=hN-GlBd-{MJ$I3Ek!a%-wIpVnN6yA|423d4oi#k9iQB}zIa>R;FO8bl=pgS}0#7>Sh%m>gsxV5uE`F7IoGtU)U;s~4uL(|~fJV9t)-7u3>} zGV6afVycXm{h@xCM771PBsHLvRc39^8U^NaOCV!QI_yg1ftf;I6?FxJk~O znKR$a+`GP6>;8U!zfaY!y{pQ`UXfPlnp3T|iYZ#~<;YTyZe!hNT^tKKuSFv|m-==T zax0Ca*v{H{9w?o(iNT+C*Qi~{(vU4K1((iBW02tT)>P;2J$PZdut~`j%W=;NyzO~y zTuMd2fRoTSVK&#)YRTf-G_rQ;dE2&uQRO}$yMF06-?}??EEQor*HK$oZ={DTo0+~o zwjJAnLtNqY9?l@qu%RpD`$kJy!bVc zBccv{>-eh+v%ekCW4XcysY^=U-m;%r8#j0=4&;F+!l?Uwgkd#NA^3DQQ3_qa76Xcsw_Gjh94;zdQ!FJ_2 zq*^AMjB1_v#jkQ${J+7u@lI@dMCCJlN*VKmA9!&2@ovA)Qh}j=rtg=d(aVY#^6AT5 z#W%>s*fwgfy9C%vf2k)^kzrHeCeC%Go>icz7>Wt<#;@PkSzPZNWq{IO>4aC;-@ZBl zD6LQxkl*Oe_K=N)9ZvG0GBo%7%51Q@t(d69@cx{C((R5UeWp+*4exCk_dq02sb5~9 z=5$IG+`0yM9??1?vu7uK!+Rjy{B>yBCXW8+V4)tL>4VS)y%#OzzvMCK04f3az^T6{ zgSJdW|BZt8#~p?Bk*vJ{`#*&`F|9KVh8m9erZXc-{R92!mLKvq79a}wQcTFq^#zkD z9Kp;GnV8v8WwtzKsWk4QKj<%jS4;_2^fQqPBx7l0U8*xEVxAjrtk7u^#syLGxoxbr zIpc4T@x^m|?fm!~{UaRcq5nXCJ_r4;ztP`^hz9`Wg7C*K;c>Gvjb-#XBChc}g8=!) zm~w(G4c0@VSm~LOjl8x?RZd4i5wn-(ToCc-HSYFi`>Eg|Mn&P6`_sM2irhx|%to!% zsXFVTP;QU&v95Ri&kSq=-ik|VO=SiYj;+^+vo&>J-WwRpDc44S#z-A+|&I8 zPf#4=b>87dYK6$Vzo9(rp^@gX9f_pxahKD-?EhjqINJ{;iq0UI*6P;4@#*5$;4`aT z+^Qp8fAd-}l^4Qi{A`^qWA@BWqiqmc=P9m^yzbH}5 zu4FHpagKP`1s$iZWEVCVz$7aPO5wz?sT7S>l%)hkML_=SbOjM^?gqE@>`<7Bza z+Sa*Xslha)PrA|EQX;*n5_N~Qej76Zs+UxN25mjAJhE#GrOJl3@64CMfRaRpFyPOd zl{4(;O%~2Bz@D`j9B46pMni@YwS+|Ge(P+VB> zHLE`$00wwGx*XxWU4b%K=Xg(-tnkaCl3tyrJB&gGrnsfs$jsAe z&z+{FqkoEz$x+B(&;L}1IL((9^6)H(#2+808@dT_E{VL}5MSobMBlReQa}J(v}v>9 zS*<~QwuRzi;8pWZ>tETHD8c_KzXKwyUWUOvZJyLF1r?Fh0sj&DKy=}wC ze@l-nwq6CqWW5*>!q;tg!POu%dDz!KJgI_9f3@K_(&hGx=QK&;3ECxZvku!d{rufV z87EryP3lWQL%`qkhwJQc;a}OsW#d)2FMc_Wa75pFy_u|MaelmgiTDeO?Lmn7IW+|9 z$K`}X`TLtmah8zY)8zbiIyPtcQI*+2`j(o830IWSOrXD67xJ*BpGaQj55&##IxVmy z5j+X#?S8>Iv7+Not8bO-l3ziAoM!koMhT7ckQ5-;oQrm9L*E@*uwx(e##syQ8Bq}z z_EVAb&+;o_5kSv56)GEGjQd5Zj}FHWB*7{ziC5i6(9H*7hA4A1?iDUVFE)$A>&@%7k534aBGTOc@V5x9KtE?{! zubY)&aS|QYeYi}9MNqQ)TW0IwT%`=k(O(3W$StVuuyoiIa>s{$X;>|inh79yRX3Er zt9zxDux;jhNIm}Idtk^sJPkqVRptOX!<)U(;pwS=6nS;MelGa*a1}fCvk3Cm zV#1dM&j%<}EPA##9w_i52aG?&OT&!)LHS26(U^reeW_A_W_=#zwd#|uUqg&UXdXQq zvf=nnazd-KQm?Oer5Q<4@-A`#lS`Vs+ptQ=W_=L^qg3(vekOzdHOeubNwou?Y zLH1LrYLQHCHH0=tHPJ#H|8lCrbB^sG6<-ee0bJvAO|L{&`i6lRR@Zu6VIsq#+y%l9 zX39nvOJAuA+Qe&qaG^C)Lo0xnjnTqK$oz~Ds++kyxm!(J!=Qy?!myImr+ibJGIs3E zTDCOqc~cknoz>z`xBI48ePJXcg-Uu%ytTfJ?ib5R-oy7F!7;;%B7y5H!p?9#UW8~r0u*SUTV zl>fdO5&=vAD1dYS^DVm66je!(DYKQF`;_58&~T=RE>AR*nWq#24+McHnH9?xd<}Z+=u+^H z7YqK5m9A9Ke~gu{0RT8e%okH5w*Uj0Pxq&5N4FtB8Z`KNy$o0M2|r$9S)XR<9UWzs z*RewpsS&%Q=|aIzazV0C5Y|s}_ZT-9eYkQgSQeRA<1v!ls6F;@t>I37EW_`yQrA>E zK1YN3{kV6F6Bj^L_KQNt!wOUDUt${2Jab)zSI# zK68*5WYS-af;GMG7Vz|o!BBiq$lTCrED=cY30-{v$^dWEv<3ws5*fM?a7l#e;^3%Zqcc9x%hYy( zw433qU_uw^E-)bO93UO60mRs7FiCbHu(A;C23ne=iV|pL(+uE8PjDZ<#QHr$Ql zclqh}a?qEYsfyRvLQNyfEtvFxRe#<~RrF_6s*-v?nc)=iEa~OSoX9xZ^iuu}7^tgV z+1LA0YoM3#-Q1x~X5lVVMd>Bk6s$~6b)m?4-t*$P$(i|ZJtmT?m60i(a-*maiF?B; zo1Y8DN{s90rD%WSN_bL&TuM?7#9O@@!NH`JHGr$zE#r)6N;BE*v5;s-@g5}KfR=Tw zDwo6gsL8F>q6MRGbxxD;1$Frd*$2QuZI_@;)B|$5rlCoK?D72mya7`RaDaB{iyUHvg1j<3h%B*jp%JffT2k<=`f_E@}p zVw+PrH$%KiIVrbg?V2d@nXGZ~g#G!`?YCdwTg!r#qa|L>GmWt6xM;sRq27?&`Lr|S zJuv_njWMr`ZQjmGZ}ieRmcQP2crV$sTP(ZnvhJ;}K(=JI!WMH}5-$$cYu*#q2(6SI z=mn^&%_Ts3f=AUv2HxL~h>TRZgsZelMhn8paVx@KV{pfQf}6AcCUm>y=W=RpK@sN)Fl%}C1ojs zvOnd}HGJ>AbH(_i+Jf-T!vB;*qc!xgRbE8OP*K0wBZ_-9{Ws{xWVLue_(3U*f@zE5 z--CX(#r)#M#04&6t!-#&yBG(gsg+}rwvFjt{aervdI6)P^b54O-@u|uM6YrW+p~>J ztd1yPHJ2XIUcO9f9xY&ZW2)1*{0%JVg`C0CqsAnEfTfTtd1%y3;3{=8qL8~tddy1u zDs6GJkhf`Q>}_Rj`Udy*c?KW&Wzbbdlnsr*0>2bh?N#PMK+zklgK>|4G(?&lmgG

zxU?VnjPfNi8pG2`tv?DlBTMAWWoFWre-!eKl_;kioB>oP-(GR*;*`Pwx6;%*OPlQ&NDAp~dcAJt4C|7_3$ z0azh`NdO5@AMjs$bN)Qs*l|DrGyxC}l8znX18<23;&a%o64g-pDbfAA9;l%VPYErA zr12r`u4=_oLg7KsT~%TTz%NBO_OBkOlF4KKR}Y+pL#!rW{Hq7*MROCK*B^^Slk>Z9 zk2Km2$>R;#Q0F<==_$MH9FBP<;{o8XYEkC}nmNDsAfn^lZBFK`48+j9t(0o2?H@^c z^`RdfP`)>zkHLLT+fsip6FsNiJ|fMozfkeIBZ=&hr)XTeiDdm#v$ZeHQCc565bm7nSr@2*QOQq{J6VsD@Y_~q;=bNL~pw zrw*kZmsp73!`}elHm%0fNA1<@>8A&{N>Qa4gdM0MnDXZA`UFZWe~Xpj4K*o{rbgDr zyL*TTZ;E=_8K4zF2ABu~$~B|QXHrCM zWl(SBqjuE#qcQQQ(s9>%!AX5LxIFw|o~2slM3~g9y*`fj{Hv@34g_jqZ>-ac?LU&= zd`>+&SB&n4uF`#KmQ>Skt-g0L#J;w|QswwmTIlJQ%jzUE9+bg6lx-wyNZl#*0;X^>=sG_Ua;0Z+b~s)i7*#xJvzr7m3_k@+dxU!Wy2cpY0_H4x_% z1Q(-ttZ~J*Y7Mq(Gb41nj6V7+EW+6T8G~4SbD6;12G3wMt;efNbZ0Z{_4wUZl%UTr zn~z2%SlgO-n@Ge5*)Vvo(fd?#zcJ$Jo})A3G@Pvy`8=Ghs%pK7EoosLTbke{ld?P2 zyo(Z`9&5H0(u1bx{jOqBzs26U`v~`w$REywA?Uea@nET5YLbf(DY;yn8bzng}hbew^W<9df#NS<&! zDm*L!ppy7Lv(zR8fD8huRhji!;am4)$n=Dfm4Oczd<`8i;lU6{Z-zbEZo7=@*36$x zLmw$J5oH(j85L9Rj)N062n7OFn`%CM@uE4xI8UhcrAFd17j}f!?T%*)ea2ifgnbA~ zhfku7ApxYyo4m23rW3reUdgbTsg!31R2#zL$S{R7FUiVKCs;{zTdP3H?`|aAKV)5t!J}fydUXt#4l_LN1>Hh$*@_ zTa)G+zDP?&-~4`FyhbGkKzY*w@KaHcXvn)S)K!U-P1|YKZ{myLuM+p15fPFhZ<(O?7{CiO?_MFV*i2 za8~$AcAQW}4+`u)yxCU@kZ5)@ik0*+gTRs6g-%A&FC&tEoSzH*0O9}@nP+*EBW6Tw zH#0G6%`6cYNv^mo&=9is?5hqw!*(x-S2ShKiIDq5FnvyG*{5fN=5>ju4)q zN>E1jq<&sPn|!pX{Z&mH5-b15QbucZy2Shk(}}mRyRC`l;MA2LD~N%amkET9u75TB zK;Q|ZZ)(%r!yOSq7xbebop`%T>QQj3Q^1?b~Bp`{{-}t?D~g)8lWNysGRA z#8ef?=@ag&=az5`&!iaWNkm1*dFLj_4z<&<^Eh&PC-vQ2lG}cQP2N6}cudXR`G>8a_ z?*nm}VfQ{%FbyKGJmW0-;qpRP_k7wkaR3H=6Md-f!Ld;|zVHloA&VN@o1)X$fDqkc>p6hd2!_qS9-NG2*Yy6hu z@QMvLH-0xU4a;a;Lv_@oJ|X0V!2DEbn3%>b%v|gu^CDKln1YbFgzNoRWE5lcofg!R z8kJl}QDo<7c{9hn*W?OL$o?Dlw0koBN(xCtR|Q=6j2&=4r=lfL^GDU2hY32*Vk*n2 zxRk_4)R%A)2}H@MTvRgGdJ8l5x1;ze)J6d>8F&n?h7KE>1ipoY>DR-jq6pKp{aPMc%?kTOJHMtvN`)G)3S>b ziv%u4p%Muus}X9UE2(J{KTA&0>1N0_0bIAGh>b(I&Z=*b5W?YmFHf|bWol<}apw1D z*Z~k)AXo%is+uRG24STy)yTA>OiKfJEm9W=PaS9=7-QAoh!wpNIjW5#naz4qEfL$X z)Y`EM&r7-{z{>`~5luqrxi_kO5RMHV1r}jAjX-=u5Q4r4^t%vzQ5Pe!2^~vJ4Gy#@ z5nsxL)NywF>1$o~uX2lbM_O*HJ!Ci41D0nw+B1zw4Mp;7HU?$Obpc8iX>1yQLuT71vpF*V^)NNYVB<|@5 z3g-~iQ7zxu1!}CDBn5|u8Q+M8ucCKv!O#}kEPIhtN<9LD zq&~X-K{he(%Btl%s-%Wvt%^rY+{e2Ccw&~&m$GX_3}QvB3XU-xXeMkmw=HkXszh?z z437^}hDd}56Vpd`*u7F~RfMG?KFDp|ljM(@K3#hD9mjK*-FJ=I-_&Xc%3;%auP}_k_z92C+oyzPyLq7a@I^42eR{WMhqeW4_EZm zVI~{1q4@8$(8Fx8LPM#>*r9sc!(4S#L%F&9k@m{NeDioiWnTP>-s8gp470I1SpL|U z^l@=ep|Li3?AT23acMHDvA#(D|D(BLVeHhk?Qv~2ps9VYBR+MtbRi!T`9E4Vseexk z!ob;oofdpe;NxT%4Rb_;FcG-$IF&XCfB=cesX-;kKs2dP)DmqR9up-<3UyoTF#BJw z2RLv+7NK<;Nl7B080MZ(GzOq}G`+Z0LWy#|qL)l0N;b1ZnHo@cu9Sy9rcwoknkWSV z0Z{_K{<$7Fk*&5z(QX*R-|DWlL?~QifmVI%Gy`5C93`2}|JfgmhXEkehTTwvI605n zo#8}(gz0i(B5*?ByKwRf-qX#Q;yMyeduYR-3*Zrb`SfdSL^vrl)zsSDrOfm}c!TDl ze0#3mNGw7HD?DQP4?ti#J6Vj2;arWuM3xY7wFyLu>=~o7t=(gaO^w`5(?!T>b1>RT zfV}*%QhxNiV`zQ2WGBXM8!c_78A&WVEi3BL#6s7nnKLBx31f=Fcs zs9Y%gh_t|~l)W3FjIDh$3IvT(-@{D+_OJ-3p{taemW~B6WVUl5du&lj5u|pP9=%4r5XB^8nt<2bDwJ8rwdJh&p?4 z=-v*q$pa6>7Wvp9a6t~a@xu4|V0wY0h6U!4Q;Zn=ao)x$BDGBREpRIXb|zQa&z8}@`N1Vug8w@+r)UHYncGF73mtm##?yq? ztsSiz=iL@7i!hcVzG`P1H^1>}QESVR6&vIv|uAP<%#asUa z5JB;Lge41&=wYv%iI1W-dR_l|jzWU8LugP_mvM0o&CpVApkj5Z7jl5mB_}YgnsR_dN>&?G&AV1%f!4Ch$aUKS zx~l~B7#n69zG@9h822d}q2PQ58%2Mn{OVG@6?3dGDAof8E4Jv!0|{Vl2<7zd7%%iK z>XGmXNX}Fw;l5m^AbrJr+WIJ!p-^g}UfQFSA1ddpDiT!KE)a&rFH;x$S>GXiy`t)q zV7t=Kh`zolxsy=!p5-9;`2o?};!M zlzLF(4feF-4dIgag~z9_l3@gK3bBKo6$D}X!<3SE@;D(T*+hvy7X}YlYZ3>0k*P)K zYnv?{*xno!U{R=j;q>`v`jzTKNh#7RyZ%wL4ptMcS1w@wX>1SI`V8q|w-E^Qanj`n zDb%s}a#NYo6lw2mKAQVy**Y~xiPRrSa6?N)XnvR38DP_I$ic(lm)wa$%S8Z<;?eio zvyOvCJ?QB}+zMCswNth1kAF;hU= zlW@p&R+?sIoGRCvQT(nH%oZ5=6On7;%DXSC+1f%HKQ#bH{Jf}M<)F5|8l2?loc$v~ zN`*;tEPRQovJJ6rOXxH)Jw?|{lrN~ZzbApYR?>E zL*!Z$MDBHKJYOTL4c9wyye0THW(Vr?Li1vz0*fGv;Y^n4-&DTtII}(TjE@PqN2CVc zfMG)Lm*KxqVQvc@fiA!|;3EZfdc>$94tT@0b)^N^%+9InCnWtdWxn<%qK_W7RQfwI;cvN600>vt4#`Q-yuia?(;^!V(XRLeVN`e>ix4S zw{T>axat0R?0Kv*ZRL%jj?F^rXT$2ulTkyVZ!Ht^7~KA!YxIDNo3Q+D2TjINGYPoG zspz130D-rK!Qxj!P)mK`OWOklC->>}&UyeV+qU!vo(V(|7ZFu9fX2q%Tu9cMUk}%j zx_UCDBX2zfwPMeAn`EI={o2nm114Om)di1eUsK|x=X>{h-bVU77r%o*5V|}$YbRS_ z?y_&Uuss>>di7pqFuhRpag`JJIwO_M&>W3q0($C{_3cx*M;rX=zPfBuO<<|Ziq>G3QgXLJ{L!b5@zk`!cML@z1YkY4u2iy57UjAvLdh`8>y}AGGcu{Qp z&uzx%*HVfqK5SuiFLb7T{C=cJ4vn{+v3iyz3WS-}CH%-qEM^4X=og6|PcLP=T*|E<*_kVv{ zn0UHrd;WDE-Su(nBvFW?8mz3$A0I>N#xJP z=g+O_&+G2b_cvYb&7!~Xoj-&qK$I^)Ofx{jJwPfYK&ClBZZSaNE$MZ=xcSxOc%e=J!=)YHzj*I1{C z>=$`y7Z?4Y2l^wvvbPhTpZQOa8i@fLws5FKhQ7$C?!7=cZ){b`cfc4GA6#{*?}4~h z!RtZ0nf*on`tLR-HKCEHKL5=gq0k!Nz>01$>FpA2ct~az8x8_%`3OG~9$4jpiM31k#twnoM zmz|d3G^i3+G|q=vH?|;l%QxLAz<;b^Dl9v#!I&M4Z5tJ_KFllN4TCmpPLM{!mr600o$E8(P2lhh7{_5x4$`HliOnQ@Jd!3SY@*m^Q1CpSaZTp04IX6;>UG z;V8Vk0c|!N>2>W=%Uf8K&v#p#V;XxxODm0G&tti8O@Bmod1SV8@lzaC&ODM2Hr3wf zDyYbm?*%hT)_G`xKWq`kdzO~6l1ufm-KyFHmP<<49Q>)jM>u9t_$3Ftb?VVg8_oS5 z*)mpnb>bF|vFarI{Sfm+X)2%oSNW7|@?N{pGk}_>Bl(Q{IQsYGO{ZZ%ieWq5RksL# zP}oM~2(|v4lP7uA5p~(VOGW(}r~ccgs%xIx(ku z1KSgyxsiOHj?kgm>q8*g=k&Bi6V}2S1kcI0*>HM5soQ8kXD*ancCC~g5)kV?G-gHY{WA27)9+7EJ47fk!n#xv6dJp-jW@USoc3x=Id(e7+{qDV(htWOw>3Y{7I_1~fSG{}$+rL< z2+4Fb+t0udd=K94ea}Ew>!kWMB~g?F55EIf?jm6kvKmOBjY=P#eJCRbc zqZ*cJ^RViS$;{XD5>zx*OKN_mTGCO3qM{o@;L;|c8)We6UcQ6BV$~8I~EzZ2=4)wPh zCEGXpKDu8&>Rq625Q5Idp-9JQI(T<5AaAuFg0~^UKvM@Nuv&yx7mV6O2>X0@24S`N zhM6t~qZ282!}n6ZroZVvCM#1%_`3Y7md_p_QGRYGVuTXD*yA&F==yq7-G7yc=B=H) zw~E_*F%DZxpp^@#&q}wt$tur>Ru<64Kp>-Ie|Yqy!l^}F$=&IOyZ zPmgVL!9V|Z{i1gNKlF?Kvp1bAiokKgM8Ntzzv&HEYT1pzzaORkrXa3>5dbNW{lB3g z-(m7dR09741^LgGlqCmT;xx36o2x@td13(oDHKX^iQ*~W3`A7?lmyBWHUKyR2jDQ9 z@%b*!{YaO`WSL9M+=qj;ZR+d ze{8HUh##$!Ekk|bIwS88g>~;fe(Q?Fz5SKn4Gjc@5zvXx7P@dP)O}#+37w?}z;G&_ z6?n6*4-I#f@-(nQNY#C99YzL-GUJ`o=gy8M3^D}(a999FLnsWJ(a?8*m?--<1-ZaA z7ZYd948T*cpi%;qCue!oRP2&iLu`8HylGwWW&;`3i>ZkLFE)NtkbcIk2pG=>iu0mU za^9!({Y9E+;W#c-xA#;4bk9LRyh$7>Zz?(EHu{oUs47Xlx`@cqg5Kv}z@?q2#fCdT zq)Czo#Q!bb^+qU5V=8sHLO&A&cq>A4CljyczAzoX@i!;>{B)3;4egX-%z4SXf#58> zzQCBg;-J;<28#(M_c+XK$_aWQF{I}0u6~~+GWC66%_p{Sndb9zi&S>%tEXo~8aM;a z*1Zb6C}38hV`tKG5Y;mAVKvIZ5JM&F*HW9@*VGBBCHd|syQMRz9?S*%rd}ph4x&su zMLO8BHc9+G(kIb7cIJBFBmDv&{H=^74{LtNG+8&1`dE;IS3~GdyWok#%hQh=%UI#Y zK=@?&g}f1^@+JXV2aU^)!)#vLFW{MBs-hzXtr%~$?Etl)1-qfzC=Wmdf-kdu2g~}= z_r^l1&HT<{1XA{at*;R_-7i7=P+UZ!oZ4Dm`X8)=Q(SDe(eIX#@cUGe{*aMHZwG6o zxY$I6cltxG*smTtUM}tp=uN2;GE7(Qq5XWuy(1lMexsy7IC1S zBM?*d&z|Q6F-qs2fCgUp+|aV2vM@@cGKhZ60|+? zUsjymv}J5+3T}SFg^cmNKB!oL!|(iQD}tTqc_SixqL;#zwAy>zFkxF8j}E z7!UOg>BQi$C)`Tzns;O9Mc~wW1*hd?%0$oQ$Q$z)}iswS*yyk$>ecD&ip5QZz z9RF#N;eJ?+yPO(D1QnC^PVU0kDdFwi0Lu%g;7Wmjp(ncjgcZ|5*!wOD>9 z!Dhd&Snp&6a={C+#86Kpt+bnM2DVc@*#HIVZ}ZJJYz#@opB7Yfeik+~XCp|N4tOZN zIM}6}Uh1_`@}61m-_P$x8Hu>^%{^PT;3QG9^QY*|K$6aBvAvj*&Xpa}ngtDJwR5_V z1yGe4s@4(?MaDaht_@rD2?BPN#9><>?jgOsGI3HdVzeeyrUByq5O{4}xpYF0`TYE3jaEqXPL zN{OhLG)ZMD-kq%)uEhbzb5=r(Ta7`As9Mv)gu=j^=I7yxz+!xL-w?>c5UwqKRQeh> z1OW#z;XrKP#yWqVoi0e>$`n2AZ414K7BwEX@?H+5%f53~(4=YKlsNpkA7lq1wh*w9 zSyUfm$=sSl8Df#=@4*yixGa>F6^3gXWBhu^StjnBcxHmRDm2WDl#R znS}`%4WTI*)e^G`xs*y_l&KD&s!u>2lbm#yoVam`eOEUz8}Itq?64PF`sWCoe+9B_ zlrFT7&yj!5E@SfphfoqVR)^5p;NXNW_Tk*rh)9Wo;lWAOHgj}tV5lzA<$UWyVwHOb zi^ba`t@dSqpqvE&mk0ts#{~sqxU<=qlCCT~m-04ZSct-vPC_+SMJaFj2KBCGR=<-p zH4l>)Nb@DdNY}|R%H__*!o10G=;R- zD3$b5ygbzypj@(YRXd`>LI0QpO@i$qe(|{5JgKf8&oJ*)_NLt0iNViQq`_xzUVf#0 zpt8q+l<2eC%MQ>6n8+XZ4Mtlgj6(nIrgGa&ailB>D)t=%Cv)Vg!d=}00)Y+@n@_f` z17`+g1+xziMrzs{z)U2e4ZDu@;w|bMBdB2rtP*kU6pxC$ItH{;+5x2cTDBy{pSc9W zFrzY*25BmAkQTK8-p}u3B)r1+E7K|T1~H0kd=f8aXjnY1p6o?>mOOjT848<5?4nBf zhziF=t0)NLh!OWB>AnXWO2B`|Tu7B`e|QC}k>c_A2b-1wurN{3BZ7AMQ9#lRR;^TrFw=gKcrXP3(lAKHUi-iD zpWYit(#$(apGWZabwDIsT_{O_5_o~u{@q2~SHub}`Qm1$9@fXEGdK6wjDj63`2*qj zf_BozW!bVWI)*gTDbjm<^l;WM8h<<(?IoAD&rEY7ELnN8w=BC-D`V&1&nC(z^xsr@yzt8yIr~!qo6k6o&;PP z!8Iggs^N;)2@1d%{yFfT&0%w`#Aj?B#YfXG@d~TolyiSqWW|%$kb!J-+_FLCE+`)Z zC0x0rN|nLKltE$Qpr41M(F@z#yG!TF2}%cf3}z`aMqm{`76nC+YGsxfTeKHLdSrCDR1 z)oI{lL0Sa!ilpN~pT#1s8VO%*dy(s0t7#NI9ef>?QERPX2jtPtoA_0lPSJp2OT2^| z(!S4qniz-7MLT*;h0C5kH9`uH62ZvG%Dmafz{QJ5anw%gYC)lAYO5BZijW9@(rW9< zZN$6OgYk4GPXsBqC^(X+Hx+vmZXco%KRbS-KmU>jFBvuW&LUqi#xH~hGo)b@aLo&w|CH_`XM%>ggAgC8aU<1S}!Q|Le zJavhHD=uP8u**nmL8&t^1!x($ngQgG(gpL#KL@B& z#Hw~thl(4iUaEsQK{9iX!Z7MV@Z5zafcCTf9K!|DDYzPpzMm0`8)8@i;QuOitzX!i zJ)uI8o7|ZKo9A-p{8l}e<8xuhhZN(s{DkBv5H4aDLIw^59qkS<{wfrhDO{+9sVrLYcsm}tu;D($Dqlig( zAaB~}8~LE!4mi20s`cn{3wX)hU7^f=uOelzXeY^EDZ^@^ASO_KsKrI{$6KKlgw}{b_p(f9Oen+}q#3JVonIc}!Z;iE&Y}HPu=5 zrXh!wMeWcX<9G4djR-Ak-3>)K1b4#RFd2co`JD^s!AtRIy35=f(111-eIj$z#YxtP`YQR#(n`S%=@%SH?-9_2+gdT$^!$_V5*9P@){`Q4 zc@->2&gn1~Or&l#5?nB5FS%91q7y@R9v_&ynBUY zT`{$i5&6I?8vR?tPbS(iL16D;T=djMk>Bvl;?^kPWfGdnsxs0pJ9X|O~ za5}sR2?(kK%*KdVdGRJX2C7h-8L_vB}$`d#yy5 z$WfssF39rx(L>+*^qylP`4bt_hO@Y zZ>--1)@nFcydg-Uv1SsiEkOK5`yG8Sw)eaZQO--A@brmMV2jG=k2zbqdfvLk_*E+E zpzZ+EE1W_j;jPDUensj`BcgMlwVto_p6_}}VjVjR=3*G4(JIRs{A-Gi zDBF0}v0*gX<*R4lcgBYyXw*onKY86B#D^`T#>{0`e?lkbDI&t(g^*UhO-wpjP@fTgc%oWx<)@_G1^z%@OCI+jmKm#iYScsR=U_!>jxKA6X?yY3g-V9^IueYXn-`4K*e7_^Vb-ID|+1t>~Pmi;ve%ouXT3FEi(cHi-WV=KyD zk$R8UYROr4eE1$%`j=yiVF;1yCy{|SKH2avgcGsv*R@WwXqI2$Mv7tmWlhY9kH1yu zCStqB(0`YJh&FI>3{ctDb~br2P=5+Y^b$wvYA68#MUAo9Jv?E|ri_3r#!lxCB|i$Z z3Gg!uJK8#Bttz5vVAHKj;((OD_)j5BR}XoS%al$=P3DfjVWE^?A^`dZ?bL1j6up$h z_~}Jyg40c!?-+QT@};=p+Q=BUGd^#6yXNf!1}52Hy*s^l%J?2zk^|>MV$E)fwn1K^ zOTqi+ka53ULqYeIXJH@MorLasG1f6u#)kBd`S3AzCW8}RZ%U%Jtgp_rsfi%^d&yOI z=`IpjcZU|0h$&2)hEWt(I)37Z=9Jja_=V6{?Hi|HVcb2X-I26T$`nJlx_(!;5?5F0 z*n)i!aYrgD$5==d4%Mc;EGXwn30}&!&oxsC1n~RQ`BLS91lT&Dcd@iPz__H>#6k}> z(Eqwcxf9QUiztJ*I&63RbYXLVPIjDdWzO%Ni`w3g@ZF5Fr@VlalWN8HR#yliP#~d4 z;dQ2;Oz&!^*TULCkKN_AHj}yBq2M)YLt?6b9t1~p;#{*T;rujA{!ngMKKxfOeCk6lck~=ZK{`796I(%<{+2NY@(zh1OmE| zEYzM-Lymxo`=#QIO5&$X8RG7LzZ23BbM{4@H=y>?uXxxcUqH%Std%f<>e9eC>?!Qb zQB$^!j#`Oe!;;KneZfR@XgY0g2}tG3I3vTE@yp1PtNW2_y)RiqZ3?@+Up03Okk-qD zr(5$Hb$ySa3g;~pZ0*9_#%-1sY>x`GHm#1Hi85H(zO!PxSu{#If(NBe^YBl_#9Hr7 z^OoWxMPq}^xcyrJSG3;#^#yzuy731FSY7V$bG&TLT_n#Uud*{Tw&9YrZDq~UbU0-B z9_-}2&bo5i?8=%;0TSWDbLq=@YobNWNlVo!oMn5Ze?Yy@pmcA_>6*<5RP^pd-{Yrc zsf)T9-%5C#VfS{lcRO-@K)Iq@H}>bGX&|t76jn#hu-_0z$IwJ_&8(m66#*jSek7GV zuOf3RknA8ed0GVP>q}K_)YG^m)rR`UaC-eMMpLGV^Mxl?VZBtGCmQv|Ge?fkLKbU& zt-s4LiHA>k^!xto!OtmLip~8-MU-uC>T)DrgN7vbPd5|(cjb$1q>!(#24r819xC6* zZiU3uD19Ab5cdxAscTek-E?+K3MR){d_wAq+1`k#zF0n#pcle6!C16GM6vlwfU%$+ z{B-^ga}#ai(vB{&k8#T1@ z1tfGkp+r4FyAe4M1Oef}@UW?HHQhEWE|ZF6A7RpZuhpD^JQEl(r&9AaJlgR#>9JuO zF9!aNDR{AW!r*`?Up%ARuI+eQA^0uEomw3^YR(psMT^h4{ujU5d8jty&9%!2 z&at#z(B(0-6X`uw(yxhB9ToJ}bKv!FsFxn~akNLA$OoKRV=2dLXCR_Mi^)Ue_(0{& zgB+(P#iFk^4o56B;pOe%syGnCI>f539XwSkWYE!qV^heCB4GHV;~As^+4u#b{Q|>I zL*}^567H!=!h8066eNN9Z_;M|RCP+H2uTG?Z?7scNY0P*q5u#V(=xL8f6frniN3@l zQmY6=P{~rj!|O8E@-bZ{Kqc00Mv_&Px}%Elv}L9yaRWNsiy4&qqU0 z5_$pxAY;!=yeH%6oHLiN~ZhSw%r5@?8Fs;Zm>Ekb`Ts@+A^JCT}A#&9D;v_Mzk$ zZ!T^3g8xn>2$q+ZPbUaY7Bw_31vFEX!%B=z6b%Ce{bSBAQb-((P}nX|DAZQ;^D!|R zczVm^`crdxm2PVM5qu-3@#MT5<~;H z9iMGHH*;6tGpwn^%{Nx}+1SN>#A>Y%#POrbrVf0QBuf=sr-XH>_v(>%ira3PQ-JBn zB!D{{QKPD!OtRcyiSC>&@Rrs1jKVY=b}gk6-6B+;>k9T6-@-*K z%J{CME>>^yPJSP^67xzZT8-y0U;xZS{?d9x&xI#l?8K;!_qGq=W03KFh^-9KFxI6rV6>TQY6P?&nAK5nQf<|6x(u3^MS|FXw zvuM_jL!SaoJ0%OxN=FO8=l~FCKB2rPPqw2E`lp!<$(a*e8(OwkmP}Ch=JvY^H>!t; z9~Ytjz1bbW9biD>!iQ~NrRm9R%&!n6l~NnV3Ax8}ltl*y6Au?ysLrN{u_s`+4vONj zhGF>wBwzp6+riM&emyf+s6>`%37-}N5aqPEvG^^J073y=EW^Vw7MSG2fE310J65Dt zB*VFtnBz6#-v@6U!B0o3pJ&8m1K(?=fiEp%I8WXVZK@fr>1x3s(=C(@rn!+4bt7jL zPe+yDmM-IG9vHJ7IbM}kcAZv>SxjqnRfDwk#z@8F&F6=A^co5e%OIRsd>+Qwu^HP?>1 zG`1$A6Wj}^)ae(k937!qg~^Vkx6V))a1-I++4S6LUN<2)=7x>BWGn|P26fC~TNf`q z##^6xNsGD^6b@rT>CQ$|HxUJfzbJ9^dA)PyPN!98>_C2dl9S~XMq`^e-fEsO2|(2Q zz1-fsc(Jgv=?){`d4E;K53F2xS&eRIs^8Q$>2RilT$tLNihJ^_K)3^AEsqq*9+!^e z6bhh@HSxsxQj&=NX^kqSA#+0->HzAX4B+KGlI*id$sRbVE({*nn3IG~@@ymIfSk~R zM(q#iR{AN%ygM%K5_UW;-Lu02m0Zu@IU*h|Tz-5ogi?0s9o8!D_4a5$4Dki}aq{9~ zOmnqD%jMPgXw4pi&&$b1BHqbFq(rU9bANkyX7>0s-{B83qvbvDECxM`xyi7-b>KW$ zS*6mL;p}{nhOmWBy}dTQN>2S48$tVqle42~Zx4Q?N~HCyUvTR{OtDdVW_4wJRqZZX zawMV_wRF~tV4b4w&a}EHxY_!5lJzIqOs@;tbk7-_=y*9KDv=&=)_4xfcovJivELdk z!C?{^7EKl7Z=K{kU4M6>VaNo`Fz-8v`0ssTEDVA zh3CIIf8AgAWp%OgqwpH0zEH+w`rfFrb7eQzj+3Pa-hP|VrY;*Sikrfrw)SwIA?K}i zWouP1C{`e&MZW;V#0lAr>sKmy-4NSi?QcQs07p@o{gIhgsjIt-DY27U*>CJ?(l)30 z42i7J+O>Gfo_`f6`h?)If9osZe7dF(gTeqVE$8Ied}r(`tLxfEl|1Y_fU9TKRV0IH zU?`_H?|Z~5i-b*lmy!nb7Te*zy+@wv(9H6xDX8xl({>Q(=?tc(E|Tnrtl)%@_ZX@P z3N|hDJ)vz)-h?V_IY1*PU&{AI>4$sIN5zt-*rC_a6-| zp$fj0a&@60+KHmbrFUA(-2K%LR9IZZ$|)P29tsRmdq~IJd?h;{xrvTIv3zYqf|2_h z=Uc4%#sldM)X!XaP9p>;!WaZ!Or|52Zz1y3r6{PS{!v*qp^4)t*@jxFt2Hc<8nWxbEx#V`(IY~nE4JmmOldr2_SSl>wF# znUSI3j%k|iWS`#T5ka2yJ<$R;&>$88ldP1ez=}oxK}7k~R}j6vu`0b0A)xl4;_9mT zd4Rj+(AG~sky*fAK$vjK*c7BY37QQPiY?kB1=9{90!GYj|Bgg|Z~(_@37VEfYO;4k zXW9~Kb~%%hE)ca?bD^6r5@7fR9|ONyLJg~a>@gdbIcO0Hj2ann@8E}tW{%aEe&E+y zgb--&a7VeC&FhYs`{52>j>A3#lvrh*$>_6==4`5L0;$Q^cv1f^NiM{0Wpyh_pylN| zR~ZGty^<}p1w!UlZI1bJxN(KfoCvC_teOEJ)sJ>!?#(aJKBuaA(W#^~{5kG{X3tcJ z{?b?$blbf#VnqqU!=C977xTh-_6PipE@K7L-ljWSpyKVJiF#Vi-(H_hQVFM%DjJi& z*0Q*?1jC2Vl{Q4-NH8A?>rZd2)Z+Kvhdk!qG|vS}fXfZ`!1uE5m~^fLQi- zmZa!aNY;i0z3LqueX?a8RjcqP7Hf9b@NSLcknZS&c1#UoZsWXGdB?N-XdC~4&P|*6 zq$?FJbdEvD;CfhaSG}~dZNJgjn4I~g1*lx`5WW`-vJ*Ps$yItC%%;`_&D=sQk;~f( zChjo021?o%iTEYFOp)j7-M(UZ3jB3UOf7toz&6b~WUeTLE~%iH)=ds45Bv<>I!SbQ zQuok!QM#lJl2};%N?bl#?Y*$rz3&f-nK~AuQUg66K1YRp+|{6iev>iC^>E_MHaz6!rVx;6TONRGj;$x%5DHt(?D#|hTm{Y;{;Pb9zUEY>c9LX6sr zKg&5B;ZntMLMko4BL(=&rxMzK65yun2xGfH-z|x1_C(}v0_yk_%!j_^na)+9S8<}3 zKU1lbP8F<~^s-|Gaa=5rzBq8gu~PxYAg-F$G7*9-cGFR(8+)d;Sdh+52UrAY$I#?Rql$ zi2LkHd|Q4Q=fb4m#*9-Ev8gkXniEH7ydW;Rc9Z<|lQckfs{u(k@9w>43O=aQn_?e; zQp~x*102D@?CF8ZuB%}T0k+)XHmP++54kL_Bq*J2MY~g?kWER=_$ovC89TsB2 zU@`xvBx*kQ)b61?_3OCD==Hv<{MVmD<_&81?X|^j1uKbBjd9=S(Xr2_X@8D&UNpT7 zlX0o1pLGO-2_^_`M7zEpgO`YV>58ggKeJg@l=7}&`wcDs;Nl^7saW#*ZU7hZC|Yz3W_ z;l;fB7l~J}tWxGAc$`6^@$4X3YqijAt#0g>Lur4b=!9M!BtX-d`>n=`xZDVHv@}jj zx9??F9b|?#hGRiqeCh`CoY*Ants;Q5z}BS7yMo@+@bWB1X}~n6N?Sjc>yQaO&xL!V zUMhszKUfwy%{$Ps*i|^-eK1;5&l=&owD0}YEZg9dEKN_!c`vYc)cmPvCrlcTEFG|^ zod7s__ZjfUl#a`{{V5rr`q0dPib+*MzMh4u=M=APc!)fg1^K<$nbI#7S$x8*xLB@U zqY-hVmISnKI3%y$xPWO{(9qFbqj|(Nf5WlyWv+=bk!^Nsn)x4orF)>qF%KhHzmzLh_F`Ir)=hTHsHG z(SP-byCr!pJeZxiinsI)rmhq5aibQc^@uiA2(F&~ZQ9mN^Nydxz=h}|FvK(u%O)L` zZIa%Eb1SOxX_NlGQ^(rO9F{QWFo0MC)V~t#?M%E1#aio}va@@!--3T&N@rn*xz=d& z#~2?TAh#pt1s^gtBObKF{9|Swnc!n^>W-||hH|L%QRy*Y=yVFuurNkVWEzFCBE2oM`cFvJvuWtt(19j0d5O^Jw zlWPmM+~bqupp=Ac&^*Ma48K!G85{(N{7U}ekhY{Ma3yl-c6@bz@jk|#1b zzQZ|^_ws|pP3xWZpZ81heIToo#~V!$`k@YJo;`XBjM#%!gkxZ1oHvha42W>RjRL#p z#-~8S{P*9R{D+^(Q-8077%Pd|ku!DOQY`7MrneHC6gRhWtDL95n3oTNDpX58Y66Z^ zmHbD&yH}<6c<;8ZrsE|J7^elghjX`k^JSOS4{NqX=8$<3Tu8U>^C};&q{N4}?nbwN z`=^${TYV|<)h~CA(PC#?Di$~!G{Bk6_e>oE8@2L?NnyxqLtpCRPL|KcpO2p8xhfym z|tJ#Ik>_GZORV_wX? zkkUJWSGsRh4BziW0+Z1qS@8Lib411d_(sHDQmM@w@cEf_c7sT{2;bq8o!1;v0GsF_vwk$di4r24HDh=qobj0o z%(@ia-ChZ~n4K+|xHlG?1Iks|IeFbx5x-%X>63YlOaHmU8;b8}>14E3IdaJC%SSx}%7-oKtnq`PdM-inF6 z>i7)9Y>yVJzzvG0Llmf-B0&QM@m{yrNd-gf6IRqkSk-SHIy;FP7{BGyHtBghHEh`l zRuIzfsN}uz`cR=r$+DyN8ASx#a0RyII+*ls3G2|OHX?3Y`F1-ob`QS+&yH1bQ$lSw zRVE{q(G;S>6pt}Hevv5rbQ$m?PH~GTX862OVJ|Sk)(~8AIMoAj2^mmjMMLSc=y)MG zzQd=Z%8GRKL2xwKm?B4fi_LST7~aFB(peH~sVDrv2;Q=?SY+zX-Knxk=^XQi&bYu;76o z{lcWUnbg4_1ELZuNF{rW%!=h*@A=aZ7YTOSrkEaqTySR>(Eb}(P6`|XcgJ%m6CDr$ zix3b%P6`+V1pwmbnOEec3h+N4zQ6zgQ~>w@R(dohZdP=(CT5PNPWp5%Hue_AmNqU9 zbhd7erZ(0zwodw%W)7ybZdR5`im(9S(YoNqRp`9NwXPV(POi`ZfFO^6002LK003a6 zpk=X6kK*-Q-4ZLHi|=8vJpp#E3g)JYo}S1k2^sEWU=c@>7vq|D?z5eA$t-mefynv% zE6d^f5kxuaVFt5#opN4)EXm(AX%~o{8ZM6k3$0`A7K^~W%tF;kp2!i_yxg`l)&pd& z^w*s)VMdS3elAP2_2W`YGPndGZfzhb$oY7_soZKua$ZHN_OVZ~ z$@N4de5SF?A35j+rL$T7Vt$g3&KaO3|DC?CQ6p&mYosa|`~{=-p`dDiGS*wBivNhj+Wjo6T(EVkg4x?KECZ8@N5tYs@Q!*KM+| z>KuhQK7Rwzuj_sO0aiP{;_scG8~zV`{xh(i!2U~M=^0q)nHc>yzzR}Yx5j2e@Y=4% z5C%mfA4oDpZ6Z?T*WM%b#e-tD@k5kruWzqxB_1cBp)&jG?C%t}-y=pK4)14TGNJu7 zwIfsk@wCLZuy&CABt@%WZR%dt-kQrbVYFkwdfM9P8vhq)GVDG z8|6ZFLEAZXC7iSz$_!B=O|)ezM_hP7S1|HS%~;idiNU_kQ&&UHc#;&@Hg2r}=%Rz! zD5AkHW<~<)WVZ|4^mBB-JC5j!B34ydBnsHnIkLQtBugyl2xEocOOl7?2JtsOqB^GV zO}W`jj*Lk{z*d*Y(mJ?sg{$a_B#C5eDrB`bpG0MntmD2}QfUFWXvU4kS#`;_OZx>7cZ<#njcmB==d z$@=U}T4A4-CI0wVZM(J|deUDo=#6M%q}%N-&s}%ht!yZ%n@0;D$=-3_NxeH(0GTc; zN&_+}DZ%lDMTgLKQ^8-&&teBFdIn!fcJ_YI4(%(tv$z#*jje`%bCSLOud)d@zS8GNjhm z!>zWw2%;65!LT3CQP2oD?1$0!!l5TvbX0+vgKX;&PPNp^uXGLOG(NHpWX01MMBnzs+ zm+Q;->*J3fPv+hpUU36eo6vQ$DXGqW;0&NjbC?XC@&cZsHm}76(gE2_^!BmLTC$Wc zr~EGpK`wUSs%1SNFBEE8nzu9vUzq&~cm$lU19wxW#N3uD<9~*(p`oIRAun0g zzM}6EAK^7K2+pUm4&S?^P5wpwV^?tX6hGGaj~)KYLjPg>!v72P|MMO5|4#jMHu~m9 z297_h|6`SO|AXk0euzHxf4=^W=>M14{Qpv{nYFRazr@D|Bqt-eWJSg zxmt{WyW0OD&dAlk(#g>1|D2WDWV!#-tV9zf0M$p2BK(}KpBquW;!M|qr0E81hq#`@ z8h?;-5V8_w!j^OuJlv56Re-p%VroECBLWsxguGtVCd_ftI3%xTEtinXJ{maPke zkwO3k6=ZcJu&q1~D9_}0M9Na?=bRms!w%ac(D0uNpnW?Y)U4c9^{Jtlk<1HU^rWW| zZv1#(C$$xA1}a1-Uj3Ypry3b{jHn>wI(0xIk+=K)){)_--m})=>nreoDAt8%lA!8G z>3ks6F+wxIm1C4t*Q+`Qg6 zfH?29?PuqA55`oqE&MdsQW>eeEjGneHq%jPLm>n~5?~Y-b^!MSZYlZ#EoA_CmYpaR zmOtS(sg$bqC&!!wSgfHvE?>a@(Yx6+jwIh3kNy_8!Ia_M;a>!6Ius7D_hI|Ez-Q=b!_J8Q3i7|WbMSM z=9GUI+=Vts1`w`5C#XIPpB|91g*6ULS%%z}c)I+>DMux}MG{$l{+i+$#_b+~lQ(LL z9-75SfqMAepe0n=*cCDvb@3}xR9nFtUIp2HvuQ4G7FE4YS+U_@v^n3bv}c!VFI+aH z7$atGaV8vJda=21gOEPa6tWHGyT~#W@Glx%W6d!Z8E;LMEcRlX4b)bA46vqI(ieUK zre2n9XAn6KrnrSi86^bhtdppVD3ll(u_2@^!kC)PH0P$9Q8CQgae92(?*hD4yk_~{ zfaw$%-eiJ78a4{M9R#h(`s`ZYJY7yj&kC_YU_vV^)SuL>4yCR>22elX+R|%e5kNm| zH3~|3tz?Sf4RKVAuG=L{QrhCTYDoKdn0{42c!QI zXIZKiHtV8jpVVLB8X{FimAFm3DC;fJyDI`2m)1arQD_kZwPFWy`oHa4nEJ`hFTS@N zOo9n%s801kM~?Qpu3KYf6KsqlDl^lR9*ErqkIUCiEL-cCBgL5cxIerne7W zIXz0wYn${A$Qq+64PEk)3%O@r96)ID&FB!Kr z{$(IXbE8ETyE-VQoL)>PQ51|2R4S^4%wH_EOnLth1p7$u3{Ke4ivbCsQUy^ONkP7P&MPB|(UB%pLXCS@O zd_TXO8jhQO^lu=EAAZEIB~t8A@MbhXQ~2n&Kgsb)J;FXWg|cMK zF$n&QXB;jvZ|cRyP5y>V*c0z@6H%61oAmAz#j2&;5q^z{-rjD`c0w*Le`5UjLmXiO zNbo_4YZ}E%{!*=DWrAz<0rYrcY*fjwu9;JH2VEAa5= z*3Q#f@}IFCB6HQzs*Un^?Q4oPYy`m!z%9X@JHS3Hmd#_S#A*H*pGY~M(a+uD6qRtT zVeFKM%SwvQiZ1@LLK|OE9YJ%M!7F3j)@15`oRzU{s(qO`FuCDKoZ;CNXlP=?d3r@( zS)ayzYC7X|w*7R%o5&Ckxp6tOuBc*}tLO}Ke`j3A9~ocIXe$m_Kn-C;RDQcNjo*6j zMq0Y3AxgAZ5CZy4(8b_-OWC;qsR1>$QO3SBHY9P9L6!iYQLq$j#ZL~XZ|3`aySh1J z=cb#8C;Di}Q`NJ)A>UV9cHQN4yk`WwEQ)})n;E!-xX?=*h+2Mg4#;t!0yF~<=Wfn zqJ5`(^|4On;mv~O#?doygS{LK1Izj3;|xVMjac-~e`d;|Jn2I3JOsAtG{Ktj7 zJ*zKNk=7y)x!wi08YF#MFdt-%;k4Lh{+>a5+;Gev(3={CqCf*f0i9cprzJ-Ok*A2w zh}wk=yc&N9`T~4@c%4cGZKImdG0>M0MDmsxQOZw^HxUyZfg~-tRshw={DN~{o*FEEwCBAAs>I7ZT6w_5( zPkc)f7&c+*atw9&AK9563xeUnOL7_CjOaAvsgx)?`oW@kC-@183MTYG`Jp+CJhs1s zAn;I26X-~=I72-H8WdDd!fX-nX_0=r{ZH6$+1O}WI$9~hvA`JIE*(gR@qh&#QkAs= zvnwGoM z`L_(nt$1b4+yZe4jFZWYXwbQMbOpV#)wI=g+MQXiY&A(?yg!#86~O2$Tc!c;eJL&Jk;}t|+XRiYXc(ccR_US zB!Zl3k-L1=7-zfLg#ox&p-(2+_OmvJ?XD>-f5o=(0;y$O{(`}Mv22Mq_5W*h1(gA9 z*nZL%AgPM3zdRwQs`)#*^D<~oTBTY-HFC;?6H%aIb4h}H*JXX2tq~jn(cExvEQT#-5+2xT+wPP59!}A zf78P8AZ+hk1WY`_jJc+tW*oEc*n%4_Q?$8@Go89Fk3Wg}XgB-PT)%#jN+(aw2~B9v zzNG-x`B@!!!>5L9{ir@RvgUDF5x2*tH9K|L3Yp&nN%17(eSjg7_Ssfu5`IlX4<{HG zRj6jWKS3*@yJqA~76)2Sc;C`|$L>iEJf9iIbMhR^P`dbTZjH$G*G*KZUO8%Y3*~^j zt>XdMiXYs6r(L_&4fItF%8SzBQgj^vYALe30jjeFxL*4m1jH1-qw4aBy!@0~i{@Ww zrOT}wUZ9=C0k3VQcRSZ!PJJ~n=)-hBGqI4{8}54NDK-Pru@3rE|MmHkcFc%k}d z8yo=OAFts*#{d>T;oX1sTmSJV{^h&=m;Qv6qMpqv6R zfB<+U;8PwAm^!ywCW$-=1s84j=Tm2ap$p*KZrM|B6Pv%wOIpKi>{L^xr=dEI z>nzfZ&l=%IHCp)xX2ZS8;A_jAaZ!cKUDMyo(nU5J8Aur%jFpb_HF=MI#n>4`?G%?* z<)dzMdh^W}@|nmy%fpn7*VqBWAO2(pr3%~ZuHh0TxR2!nfI$W}o52l4n@+thy=Map z_L1)WhExZ!Fj$}C-_i9hKkpL(xUn>yHATJ2qCC!3#xR(^GVQ%|CrXp4gaQ8t8WNlhM3EbH1Z z|7^z9Dv1~r+U_?=flKc+kgsH8B_&R$(x08Cv{U&YWN--h=6aKFa*NB z?LDh49}X|>8yHh$5{bwF-)048pZj@2@k~d2K-f4^SK&mYY$-VdKb_9p>R);#uteUg z^r`^u%CfRTh^H1|e;_67A{>Es+e$P)qhAwSqk2Xzuy88zHUv&Vu{6YHjjp(CrRV5< zgVuHgYCsRifa zXo7-+>0(|{IJX7nnB&~QBrotoB-+(j)kXPP){K#*XLlnRO4^_?qpFoW1kutCK;U`q zDVUw@Qs~TLr8QHd23*H!6`&KLJV5NrJM3U~Hl8v{BJIL^>v)J;8H0e9-2=i{8VrhN z^S*&)MLBvnV4k+c5X5*wsW62vC9Ni;tuJeX^8)q2khK&H73KlVDeed@9m@63TK@{3 zhM74U1nCU}i!c^_8EaI@EILY2KhlcTAGI`h1^;_xVuf4&3R-p{Wg7(=*8Xp1Eo?6OYwR z&KbNS5v^Lt=Y$FIZbrmDSe&@th~}0INl+;u9@ns)`NiO@8f$E_ZfIIff?d(PM_dbGA1#=rrkkB=4TSk!MK$^PNV7X+n$gAwoU(K2yhRAPB=+iy||8N zA%J1~x2c+DEtzJLP&X6yX4FoD|+Pzir2 z5eiYO#l>yGX=ANBGfH0-E{bSE+xHtRHMEEKZHBZrs@Qm|M$3ZKdpv-+S5$%g$KAje zRCF-dg|2J10l5NshfgpCaR~_Ha1d9n=zEXIK{O=U(Z)Li@{S;sLrK-~jPK7-9(IpF zQc7x)6a}t$xWqZq=qfQoX=Fj1bt`wmNOMGcvzW|j7d3{YK>7HIRt6T%kdE!=ok5p{ zWnmEoxAvRV?5HBxLM8`#Ls~0S<2OeXR@^xMWaoyXd?y-Pk$ZJ+31dAR$2U<2$?h z{auRw2>lNc2g$xb{_LXw{4?VJ`Cd39{Xf>b>Hf8VVC-OJV{PVWV^8Dq&keECv;Hp$ zxk|;#0htZq%kn$?@Fvn~=n3?#$RMEs=L9+n&d`-2i(n`2+EC+wT8mgKl8mLhvz;nS zXAeP8gBK*Hq22xIQtkAsv%#J{i~oU%JoDi-6s9-62b&zMU08W z9F+t=##6@dA}!TD!X3C(yP`x6MUsQL@d+ZNhJr(+z$raFdlsb@KBXYXfFrLB51ts+ z!~l8rkai&2BBwFa6T_rN`FDNLm^RCNgsFnp%vg5^TF9e?m7tq!W*kZd=7-3l--qL4 zTwq5IAAMFp^%lM$v!UeS0MH+ru0O|YA(6w@v1epXVVn@I|Gojp1m$q4Doz!og@I5( z?=F&D);F*W1*F|w7u6nvOleSH<|P#6)*RETAv$ji9B}*Gi$z8QmJJr>Ixz5~l;Yy% z4if%FFytg}EiKe~Jg+34Wfw44FlQ_%F%SM$<4A1_njOQ2PMl7RY;)A>hA*b$xt0ou<5Rf=)!Bw5p>tG3n;L0)0B9zgs&hJt_)?lJt=?dy}8a zA{QV-XmWl&gGE*rk_A^MRMO<$F3T|Zq#ROHgKRu-Avet3Y9fv}3M@#wwdz_x_9!d) z)E*5h?{Y}Ft^B5CgRgZADj}=d<7OY#$G)U$+0gnoAPI_xI-*d?U^u1N@1@;*Y@(|n zbSdiq4t3%`0VohS_U4?|0`~Dcc(h;YO+VElUUURYYZ8)%mQwy0sQxtNfwT#9A2Omo z3cZpBchtlq_aBdDfi$&(;My6?SZ;61FETZ(kIlrHrTwe#)!?2uQiIgg|EGY{G6SYI zjnf5VaL!aL$nDl_p9Cb$u-i4k5k=z|-LKez(!KJCyfRj_OyxHK>Kweq z1x5@7+A?_T%nJEIHa7a53?=qUf|e;#!6cy5@6sS_CYqx$hSLN-`{QYNo&lWw z%loMV{4(nJGiK)6VxA7sU#GEs5^NEM0|&-YZ)^%a&x!8sNr^O9k#AmF0WkF=@CYYi zTo}z5QfhVR2_;%0r#vxreU^D;3X3+ib*6qYlH))mb?0J^U95PC`h~w-yFv9}j&R}K z5F{jMq=sZnu$3bwL>oko1DCTj`34g5NJT1>D-_~0KLf}*9+rx!tR9uonwGD2x~_)< z?!@h{>zUa^-2QF+jw-c%N~~%gnj;{Ds36@eV@%0OJ*_e@*7F;}0;{#DwEDwLY?hgk z-TK^3UmmR@gcCkQ2d^#SMKDx?2xty*Tv3gqT(twn6z;Fb9X<2|P+~PC@$=wZg?$n> z7z*B8yZKpk^y?Z%!N?G{gWBh8%$T^4c#h!Vn1rF}>72Sr7rkg?9KIK_R zvp>RmK!AjCQp^xOb+gvGfw}DC7cc)tD9ucmU6%cWWrJJZg$v(|cI9eX6xY6C{lF*m zwtGSO36J74AEoaA3BiB9I-QoXq>mdzaJ8G|ud-D8xZijV5}jI}|1u>T*dbow+RU2h zsh~MD{zOwxp{n=}I>% z<0>QLoQxAed+2U+xxQ7=^2H6wYIkp^l>#E(66*F&5T<6so=qE$I@PLq@+}o+?M}Au zqA^vR#_cs&mr9wQeR#rd%CDG330*70yEa|tjdG;*mR5$egF>0Z35f+AHX%xjjFGM{ z%u11ip!nSuWnh=yOZ%SWE73_S1~g6``ucBdE98mI#lDAJOIJ6FO4Ur+30Uea{jiG* zAAF9-*92fQh`^>6MHez324B^}&&r`4yAB-*#~+*vjXa245-Ikr5I6PsL)e>UB}-;N1^0RYm!4UPVbXVo(>ar{q-R)_DSvb~Im zJcC+HO-Szi>LbzA;z1y{Gs?v6L`s$|(K0C9A!`tVuiTC!xsWzX^S_J+SXQ`jD&p{e z4F4$e=Z517s-cFQNrE$1E*ht^EDS(o5XBo)*5m>}^PMg4g}fB)*{!2Vq%_4(?ob^aUH{vRS+f5mC}6-Goi z%Z_bpX1}N*`ZM)p`v7@K76r+`JS!|8-Ap|+bz3ql=-$BxQcjbsWcFAHvHSD0Gt1RA zsRTjtfV52PbjacoYMpFFnxZMpj-I5d!Zuna z-#rjvnid@VkWJ0So?OyOM8)$;J2@_D%k0g8bi@t=3oNmy(i+7F?`a)oSW~|kvlIZ4 zIm|H*W63nt0HBuml;6l9*USLX&n8t{RKYR3omx;9xcLj=K@O`91iZY&eb5i@>Q#0_ z;VC-qpmc`#hNS-WQLOM7w~^?!(xfax({=;zASFYze!Rh|0q-c2r6?fK zxuBCU>C?tBe|GdZVcp6(yb8GRmdkQs+0fj&qQ?3Pz0e&{CWf)W!=M>S*MD#>i(Q8#*Ap*}&(mdV?3%ME=2e{s(@wssxYbnhbsvoN=1 zZPcdm3&-B$FW-2E!p?a3xw!r;qrYEZ?Vl&B|He03+1b-MSnKQ9|EIo@Zd?x8hY=QZ z>LWwfU(1`t8G-?kFdlx!K4x?Cto-oce4@>Qp*QJTTgq4{R2`h4E}lXUNlhoVVB10+3Z%-2vhdooLI(`Ttg4PyO!{*@c5;I!%B&L# zYy6!h0}JODcSyHOKQY_ak_pulnMB*TOQ1i_B?b1OE97%_e`fgibIP{B`ht+j4EdX1tqr4A_d~@K6B@lIQ@zT15{@d?3U0 z^cx2#LlEr7{!v<^AAKRJh;gacJ*d`*HMM4oSI0hRrF5bjZdB0Hg|UNjdtfTbuE}D_ z7if4n-~pbNZzWi?n&b|5f=c;3{WgUhc*`B<#a z%9F8KDr4^yis3VCM>{`9tMX%BTKg)()O0ZVZLFo4(Q@H1u0t)(>vNLYX*!s@o0}CN zdY8IRupCW1szigp&LLM`iD?t0+yM?Bz56Fm)9jV^E9PSef2=>*B@GPqY2QCK{(I|d zetH5Tz`x{p{)3wMV|{Cf&$Q0}EVIF1al>qt5!U1D2ilMG#rmnAo^9oI{&JW!Vax*~ z_yg=|!olraj)vSa-1mduYV6uqFi*EQ)DphBKa(0gZ8y!_b5If`TrLBD9r0j<9BFoi ztX~lFJKO|nx6IU*XZC?7r1?Vb>CYG@pi>LflrW$TGLM3q9sxS|1#E~t2LexE7*OBy zoZ*urc)8dbgqRbi$Kex#NMoKZ7IjZWZP6MhPqd6h8{cd~mC7-;dnq>vh>d_IRW%yT zJY0j7JP2bk(YtR7Ap`QzkD4QYTS#{G(XlLN8IVci5@&GDL+^2sR8JRF>!Ewh)ATzIzjw-l6<@t==2epZ+skU#r#&JddYMdyqtGDZl|#aEV18OEtZ-$ z)V>`lm$G^sPNwFHRbRU@Ew{FO=RQNv6rJa47$u@Bs)GxjJ6ynoX~BeOMZAFcJ!zdX*A4Z}SY~Wg7pKdvn;c^iA)b*vc*ssl3pgCBKwOZ465aO7jRAl60 zF?UdLwF&+gtbk*1Mx^)UjBx0;FBdQzNok!%1-?${gHh%TWU$`;R$Wyrj68b3Z-9HyZR%vNp1(VqxC^$5@^NU2pQL zdf!Ez5ASL)qB!-mtm~QI$CT}EDNIvRiy@-|_AAw2xxenFOCj~FO%1iZ-XaR^zCGbg z$wzwv&KE7%J#!j=2-61v36ql(AO|U1~%ldDq>2)yfT|qfs}R1_oSe0{^;Mz zphi_LWA5odRi%MYVVKNv-Vt3nMcegP<3+d%;-gw|Qp=Nme(y3d9PhmK5V_6>`YS&J z0Rj4{{c~^pvm*X}tq8&c00e)#R{l@h;wN?Y=Pc#@&smDOaJ|m3#J!0NP75mDvrgA% z|MC#xEQ674;@AxFL8XPa2VPP#^@O!HLMC88au=t`HC`vNhYjfhL0NnJTi>tkv3;F2 z{WY6$G!9~1bp10?a;$WUzr<*U^ArQHzLNCrr_{*P1xfWv5gkCW{;bXE+N#U3riYE^ z+1t(-luB6luR|xNP@YhM2|8p+GmN3*I7&U9;q7bwePX?xY~T^^V(1`N58BIK}dizt0>w*+RE zsH8{aoe5@^^y_nB71C>{!74ywVz!lE|2BEs3Kibt#o6xVeJ7L1owG4We`?1hqQ>Q7 zTdpE+P#&X|#sr4PqFY?Bg^SAq^oTA_#Z<(Me0;H7zP{$J^#?st#&Q;m;oSxF>}C@E z=5I6&XV0q3iJx$(sagW*v4kiBqv2f0MExUH&PQiXmHLz8@rHID_g?evxia`aDUR49 zM!VHlmgSv7LJNLeON}NHu%_r#i9Jyi8^gf*(kl@tOt*LX0@uI7LN;kgS+(@X=6z`l zWY39+k1LMJ6GU=u$TkBVm46#o&}+<6&dt)3ogU_!PH2h!{br4sCvAt@04k)Yj_47S zKv=6h>`(G6OC)ScA0|rj+#6cY^e7j@a)txyKT%JJD+&f(2GCZ*)g8cNLI)%&HrfeJ*Zx^^?|%dMAPa9biYYU;s}|C&CV?I6I#%A|Tr>5!#q1*Z$L~@P&(a;fpQRcn2@2v? zoB32KZF~Ba@f$d|6&3SLaE@gW7Ae*p+(K0gIT7DH)o&<-iE6xSY7sSpi72YccY9`* z!w(8*mW6#YrPv|-^h-<06%gdk)FjXP#tR1r%MF7gRk=}9Q`}e)FH(n(0v}j-JP`9~ zjA5!%7uZK+&Mva@a!IZydt7;-r2J~PP&K9#jtW)8em{H9^@JB81#d!Qua@#$(hiXJDLfWR6`7BsVY0FpI55QI+c}%tnX!BHEJ@@;dKn=|e1pPHeLkHJGD~jz zM`-(ssqE)cO6@c{^mLLh&rki9$9&xxWv(C?*B}yz;-uG@?rM)?o!ny$9TY{!H;u|*N&yXSsukJwo`q_{5Mu6~axZzAJo$Z6p#!4eh zMq681ee>;_t_`teSe6)P7;o^zGQ?K=tR7-_2nRcJRFt4Y9)3}=d7qsaX<7GdIJ|CH zhdP}sz;44N6X;SUwOquf=mA%WQ$DXzF3c-c|x9-vtivw&Tg}oLKyB4&g=-hG{PDyu)A` zz`l7r9yp$+(UU1o%VZh@IF{nf15};Y&&%5Os0iq%fW2>5RLfKF@+^8Dxq}qc%?zan z`vR75q?tOW32C!3=X!ET05%8PQVaXa=tdIIy&hU(v*R=PXJhs|FRnM9Cn_{X+O$i7 zmaa0fQh}2Ch?h39+~gf?qEhAJ9f;CIIjXN$d&a*7Pg(O;9a^j*)}vb!eK5DKH-pZUP2ZeXz}0n9sN( z5|$@}Wxj~1Vcw-&y#|P(<}qJ}xi`z{sC`m&cF;zOn9pgxG6wORuDP&lY{w0f3lv*E-$qBS(b6f+NR^=l7RjSW-*EHq)dk6IQrKbLhPTSB`$Vi*K zU8bY5jE-1-bD)q)D6_Ro5!>X6HlEUNCEg*MRMZX0na8hI21DU8vkwq`9MY-qI-{=n()+G&$ z-n4~~3~J_ObCt*Vcw<6N7fqZdmEAru&M%P>Dr5RD6&Rx`!|S5evE*ypB&eS~E%+n| z8BV=_pib4Y>s0gQtyz(Ot+v-Kmo4I{zQXaLNX-Y$n+|(~WDsy{ zM*vLYvUg;2Y7&RdT-Zx_Zsx`0lz!mXi-oME02syw;5q@YyFGBQ|X!$!cT1OKKztt-!^QHu1?w3Ux)}U(d8Pu&}Z>_)q`p9~B3A(Ld#e zLihjx*FWcPp`-Ojb?9XKNdXwx+I{9FTD$1!eC{t=S`7aiHZL6?qAbU z)rT;Im;GCv$2_OH`Kq#G;*Ey{OU#~ljd^fL`QXs?P_RnG&Y}daI&|8ituB=fPlS+y zkYr9D(O@F(u)JvNvVSPGx8n|2TTqp%y?>6l1&2tuNGSH33E%THdMo9ib`%g05HK{z zcxD)fadc&dE9dUmArk1J3g_I`UVGw!`(*pUVdcp&F=U0-8l)`GlJAsfPRXd2hsO&9 zAd>_LDlBZ~K30O|@^&br?918s^UT+PTUP^5_9*zNhHvnOLPv}iFMHzbDG`F)Q4zsR zL`?w9SqJ=Ea%92AXhig0EoY=(n-1cILV`boRt;zF1aH|T2mLX-{Juj}2^2vr9t!Ac z|B7V!fO&M?WX`ZePLhk0b`#PRjlmyf&@I!YAuNrJC6;AYCL}4vwg6)&3CSzXQ>#2` ztFTcWD^f+QMpL%S>5+=PS@O509;KP_D>_Oxg>%7$%Llx29*PSKGnhnn<+YC*`ylZ& z3YlnO`U}{0BV0cwPScU&0nU9m00eZb)fhK5^wmb>A|ghkWBN&ShCwLA1i+7QH=N(+ zO9#wEG9AUB!~CE#I(0qz8H>CMc+X)~hQa!pEr7yvUdF^{jGpS<;QbJ6(m>m+)ArEf z)oAjpdA7d$Vtti~qo5^-q+p|AWD(2Mi5MMfE3ir12eGFerFZA!y2L+J#Yx&76#2|; zSE$Jg0yz-25LK}QNl5~82zh+$+Jgb%ERzkzawi-m0CMQ{C5hx5JFo>6`mHQQ{U!1w zXa?DUHy@WACrbqfaLFfCglhVf{}H<>IR#r^hR-ZGTyF`u*DziBXpLHm`FUSHN7s9y zeoWMx{5`&7*oW}f)4gA26YrrV#2G!?$@>TeS~_$FgZ2c76Mj>h6H^}RX_&$CK^?l# zuRLh2E|YSwX?PXy;`|(qqOE~=(aUb>s#lQD?vXP%rG_o6HnsTW$4)C?qL-dniy=ZL z%NF;_M=%-rQS0)_t%q{-7(9a30lK|J?pouEoNxT`aq9NPaL0H`iYOIv+4bOcz4=aO z5isj!`XzX8U<#LA$JV`4-JvMIYM@uxIpc?cX<^cUkGLeYl{V zWG|}_J1i_IwyVhSYtgXxEzpR720EnzjT^x3G)PyrfGwYg{eEoEiCOaetnF`&ZR+R*5;bjHNbzt1WAp@jw05V2I+SPA zw^=@50jjl&(9LU!GkU&Yu8o!>?V4SWmWGt4719%9!w-KU-fEt7(#FpM2>9rJy!vI@6i_AW{#FV)2w)O*~l zZwxz? zC3K&>VuRRda58=Zb<**d$~_&jF<%K65xs+X9#svmI)iTOE2z3IK)YBH55ks68zRmAkD(Vwyc10rVi zI2i6kC$s7&oi;$wTLPJN>fj)w&50{Z;elB1mLSpv_KO zMzW8r_)QNeg+{D0;58)HPoJS8@Pg0)$f)P*vY($wHBXGTi{kk8p0hi&AqreiYMyja zf=aBV+PYH(eyw0%I6q9vyrGyf{B~K~mN|@RGL_uWXpYin`8CH*Lx0>A-53njcOvc= zP&)6fgm0OX-Cyc8wa>D#H%ENz6s=2@5FM%j2gd1kJyC`(26EMIvn6?|cN1cc@@D>3 z98}SBBOxu1{HuhZ8d;!l9zSz+12sU@r?+k}0-BuY5#TaEx-}@P6xod(9F}u(^lPhcMjoHnd;h_ z5SZDAqrCt1<><3$W_-w%Cd+zj@aq;5992`_y1dlIeCN3$~o}69`$_NZQD_<7MTW$P;)hhrrKP zCFHy(aOI`Nx3|4UpYq;V*PzYEH=h6F2@F35qv^` z3}VAcsY%F1Vd*s~oU^yluPk%mIsMq|a+GfIYLO0Q{-M=hS$-Q$wa!ho1q18n-R$Nj zhNcS{9z-FTLAZqRYV$xLl6RW;H%hQ>O4wed|@)Z-rdmcyL$=ja<=m zdyLgxfvPakUcRTHgO#&+vrlu;7R%E#E!8^V=J{ zL*=k^$Ll0SxXQZcXZgNrzV^q##E-Lhz7F|Sr~XwZu&+UUCKq2;1PJvWmlt|>A_E7* z$MTY=B|&w+f7a-Sm!2LRm~ycT`chz*>hTT2FoPI4Q%=18K?9Nu+pN6{GW|XEp!3xI z(YE@cZpmr+kXu7HGG-WpJ;#P*SviF4E6$U>{vAa>rrCgMZ}XRS$_9b~CYvdJZGC5F zXMIvEQZy|8$q-y45sn&D!fSr+0h=#6TC|a%oU-3paK%PRA8|Db_C};d;|wCCn3MGD zu{moxWfLLjgyi^%^T`Q+2-|p&X|-;dZQQ!Nx@*s!5&Es`(lh1nP<&~xWO<*Exz!_z zkBAv>3P!AfVW@kw1zJL@2diA^xs zczfkrw1!fupY+MIS*WzP+6vXYz%R5^U}CVbAJ?8M*)P;A)M(lDCF>)1!A52V`wJQI z3lebWuH^Bz>PXl*tTpZGSD{9t69@?55%D`LmJ}r?cb3KaZwS!7+3ZungL^y;jHG62Jl^&kzt^YstImXxtyvZIU${~}C`WBV(8uEQ zgc!F`;ADu_+4RD%JO+V$CB$vhw)+k}kbo|68&;MB6l-m3h3x8DqM<6dolFUev?%R0zSV zup>K|UB<5&O=duTl%LS6BAyv3U*lfd*6@PIC|kzvKSX5CJ5X`PFd%P8ThCgDA4ZGt z?zVN`lkVhsRhyp~FL*cVP0=xQ)))@@fX0BiZ6BS<^pu?QjorOwD^@DF-8${6@q5*& zI)aFDj}yE*vQ{uduhr}c?3>q;f&Rivx0^k{Nu`xPyt&H_CffpS{s+Zn3>KJvYY%0A zDA5`6k3Vvzl18g#^l;GwyD2575(k z4?eF)wP<~?x}zP6WbO^JYz40Jum*#w4iVxSk>VPGzQd4og@NCp9!B6f_rT#X0hzuM z)L4;0Gq6<861k%xT4dH5fGH*fR6_J)<0cGtLo6w!G9=Ry!g!a#_y$FW7GXIw1(h(? z^=wKgBqagE)**#-saY*4@7zTrhD|zcB_!JEpp*Z$O525Av*v#3gTPiyJ|37vM&=aY z1jU&uzUv%ZmOyFg zzc>r?4y!9#1OVU2AA1SBvg|bP(ST>a!Dq867fipZtSx| z*_oxWyBi5bhJlt34Y{)=vK1NQucyKK>Ra~;J&iyIK5|q@L7tC(V3?r4tG}Or+?t1O zl(>s=h*7v-fS#C#o_qxHbclhvXOvH#cz8g53>-cmm{ftlqQUp|D~hq3Tbri99s_*p z>TBFgnqn(4d=^<4|4&*N(yxpx%*^ZBQypG#Xw^VUe@ytP0%xo19QhYnT77=s-lv`S;`y#Gi5g`}0o*it>LjFn?~- z|A#~WlZ^RmY>iLW<f$b;H^WT&G zH#Cn$L^qr-Bdo|pa?hX_G58@SBY&KaZ%z#6Qnpf~RgS!3$+Pas%ZHI`0ptyfBZA{e z2(3mQDzlBa9mSGCbr{K&A1qNtA67ugJ;7)0W4pZR&$JLI5@5HgUkem<-u%53i_p10;} ze-3|x&}TSf*d~MN_?c78yT6!00kj_8(of6(vB*DH+5arWwaD-gW`a zdGpK{SPox!E-20uPdW2QW0Chr&u9(9*&BxJc8DgG!e-GhIH8_Q7lQ9D{20(}8V=(o zLp6T9kNrL}6)PfLDq+Fk>F1Om8%BdmE(>@$;gQv68&=Kxlr(f?oX#>vx}Z~zAzzW%l~;NL8?BsA!T`mRQA2!8m!*&{ird|b@y|?(!lyP!vv>`K z_nw@dj(+aYVvdRLjfy_TRI4hY7h z{1R(}Z%zg?bPx~>}U(R1gcb$zGBoFGNd1mglZBC2L~DbD0}s$8nmk{zh#0{zo2(6 z>fVHxao9fWZrg~QxVtW;_MTJ5uCX%4D$q)pqo+=MKM_r-ZZcHUgT!tm8zND?J zE~&4iv;crv-FJ_?i8QY+pPPAwh=wBu6yA~kclNVJT5d$nd&amxGgjKIlRa3fy};M+ zi$k$0Ri#ykaUksP;J0J_6cU=f9c{a~gvCE1YeG>BFn6(4jfT2NoTSxgT-Nz+%5TzZ z(nMJcwdpz?ysi3#`Sy@v*I<~B=ohHQTX<*CLe=>Fvs%w@LlanZ?T|rj&<_MQB@)KE zdls~N{re}7*?-BPK!MZSH9!l}-%Jl@@nPQcb7q3z#^r8lPl2FaUB-gbhavEZ*Hr!# zBTUSKLY(5hi0*u57e+5Si(+(F?k!*J`gzw1t}xU9ZV^5NH<{$Bxov;95>Q-@4KO2N zi(w4`uGjZc+`z->x^5w9?7bsjPJhtqXGWzpT8Gg}Hs;?Bqs-h?JH9GpYtMG;X4a(j z7HK%{9m#hRjSv{zb@lo1Cgg#W;8~DI27eI(?+BIcX-pZbl)j{x^~Opheh|x+F%ClP zXD$uSaHoDn$Byq}LuOo;6oC7(GkwTUS<-MM_P#WnAG3mfJDeHVu&(LH{DTe4OGsdp zz^fYJ*JqjlY~ik7nvAPHB*5qlpk*qiI*o2FC88MhQC`}H>l`0TWPMC62J@}AJ8^xh z9>GE%Fr{!4x6*b6iaLQ zsmyy8&mYsmVzyhmmy9sj#8m0?ejKZtoNT>mlB)CF!L$A`l}N=SYKx)NM|arw%t>6~ zkUoU_oy_m?;hGVZmA+vR;pM!Ii3sCY?qn#Sm)D&jMYy2y`DiWIUa1?H%rs z!MdVGe+MEHgut~ zZ2KNzI7xDmuvnE!s$Nt>^rOF+SWc!n>-e9am=|~)<3sL`U?VIOgDY8o8`Nh2NU3n@ zZ}7?!Brj8x9-(E{m0Y}gDl1*lthNyUru-_&V)M&@Ww;w(lHkKRf^^7f`|KvDkT8|X z55H<`&~aDE&*8wbf|NtGk~ptGj%nu-n%Y6Y)TT58mXb8^7<$Pp;GR^X{`V5vfLa#c{j}|E`M}-=mC4l}gjYBDo2BPpJzz~!m1&iqtal;!eOS-Ciu7o&3W_IGY@ z=WYtL7X6^q5^MW?UhE)3UK0P3SKsK!di+wHb++3-WLnac{eqp8qA^G1400-u%KqwW z=J0GVif?)7ha)= ztCc%mIH|a)cExbO@Uy>Z6n(}Norqhinh~8|=4hO&{Polka2{YDAa39w%HI;4~Ma5Pq@w>)tl$~IViEq&|^|V`~A+X zfF?XC`;z8rz!&(t1onim>pL0jo#Y);8+d*k_qra&-7!5Qv3=k4l4pq%i&60m4LvYI8IOY2O z-ls2Mq~p&XUC#LhG!yfJ-OQ5H!hBH4lXBMVSgrIDkFQQMKkNEKvO-z+&b&xK(H%3@T2n-bWqcWB&y3kDK4qPN+)Nq1iS2qqcEwQu0K@Y$B{*$C;S_b{L zTXTkhV04CU5vH+Q%VK_Et>2`q0<pa4u8zEJ z?`_q*4`+gDws0|t?g+UbN~;`EopRMSLNY%%^b4H z{NlGf?BiP>TKz2M=x{nTX&rip+VDV-Fcg8vk#m;G{zJJS6(Lt8}|<(*T0=#G<4dV1`|X&_quF4EbT1L?;JH=vC>n$BZaoPv2-*qR4$$4 zI-q%&ej&p7X=AN8396LM`74GWLLG!|+2uSH$VU`O^IZW0ayU${6NI@mt&;J(wiuVJ z04PYeGgCzru>af~&BSNT)`Shh;B?YrR|!V#ft%;L@lxe!(Ge+}h1E0+7(k_FEn3jB zwHdp-W_`c-LZwK{HGV0(M4PMGXN#-6CRusErUbX_7CZUr;|-uYKaDMNx9_zXy$+^r z9%(*y8^01{!`RfbkawEpTsP5B&N^$J&ydb|m4O~(f(m)TN=GCdW}W$V3_RA3WD$p& zJdwzM*8EzjC?vM-QkMCe-Ck_6@*4Ep6YUYSU(RHMyrIih~V9!~k)6N3{{YnyQzF~j+9SZCI+9iVrBITQ-#WA zRm=R^+3R0ggtA=g=37_*;18Vod!Y5nbNo*&!rua|zeC;s28c@0j7%y`Nll2$Ovutv zD-Vs)4KYy5SqEJIB}j1r(0E~g4)|x#e~-s0{~wI}uek4jrbYZW64&~_AaQKHe<>65 zFO#;i(ABa3Q|9pV>pwu2XS^D*E26d+FEsng2%vOPnx^d)qZvn|lA7ElG4=H75iyMF zl(bBwHo!pm0Ncbuc_sXtv_wUZZ|}ozG4>B`+TfU8o07`sTi+=dfX3_}ZWrty#x^sm zrZGs^Cv!M7G;1uHG2ier)6)7*WD*iD4Qo|8W6hiW44>C*JC*rn&wkY!>2%MG4--%R zevBA)>L*XT5#+cHnNsYUjk15gajldhUb_k>bk)Ls3qv+;v zewtN?f$~+1;i}EsHt>ks%Bx-sLzW4E%4+PVgt*HTSz^-m+Cta$CXM#XukxVnL?=<-d-n|X@g+x^m z>5idYo1=ExV!p$;e5?*K6`Ve4ee8HbH*D+3SMc*NwY5Mv7&@ix@hnexm-1it89A*W zeIOV+N0WTIyn?h>b#)TT~wU?l6*2RygSllP&adV3B;T;qR5Q+-<9)k)SH;|L> z+kT*ypq)tB3L!zqzw8IuFWP&lR!8P02b?q<<5eUjYJ^glTA2R?9S<|9>Uffz`WD2P8 z*iSzvWQRYP&|A8gm54c;P^AYoT@v+T8!}T!3zhzs%8RoL(+>_&K z@dYz%*$7~`SNjCf)2qPrfLz$k_wF*~HK{^qSxpZ9z>y zP0K>XoMsMV=*+_mBW!Ln8|%?!4>v=hXbM}48}Wh9iXg_5l9dOrHDHxcCTvJM1su`^ zk8A6UcrxhNg1QT;30GqhQ==k7Pr+?^zD<5`;d~ov@k@wT!y+%hiQGW{eH ztn4uDyeX+!5b5Nm^Af|q~8)x(GHoh^OD~Er*wd|eB=DY{RV%*dv$@cZ^hsv zCu!Y)0?I~g6@Gd=?UDN}m0}iN6<|&tY~PPoPJV1XqULhr^gE-_ zl?~~Pk1|4!B9cEmfDrO+KVkMO!qP99S1nic+HckhbhN?r@E|<+t{~_(A!$_+b}jKi zw)s%NWESiK!VG&0*LQo2VneehXZR0CZ-G7a9-~Xtk7@fDc>`6=V6x+3^STlWH+*@T zFVTPdYceMpiE#4XEOEI(4q3=j7+KS77!2DJ3y~fJ?S}h?@L-xd=rzjk3ZmTAf)8MV zvduWvJEkxj=X)+J8{1~f+`UkxHm!I&r78D1H7*q_9_I)czJX0+((PHx{z`)%x8 zZ0ly+#JFPc|L}zpN|Lrk&*CGb*i>`FU57uh-@F0zh$uF1#)zMRUv)zf9<)Y!1K&Oz zS%%Mn=p_xEF>TTE!w;3H&OuVKxOot!xf!d-YejfW+54=fq(V^IjtZ zd})t_#3_VOqcrpE1#_sVH&dGce7Hl+(kXTGwiq|ABLu0HrX8rua~X=*kDxU*42PL2 z=U;GFZS#P^GkZWYc^}RB$|4^sAQ}BYF`BT1SbK>${k2#noNK4`Plfd9-m$(kBCe4n zP_b%|s;Or1`EvR|S`(@Ek1D}{`HVa5{Ce;fpXrb_X(Ys5h6xwb*us2s)Pq=YzRz>RQhRz zr|cQT;(7@UIJqjtZlVVOdOA7Gw@S@67nd8Z$E)Y#yU_){v2Ar96qQdsPNV+4J9IjP z3h`8Hw6ER;AMAjtIDl=;5A(AM$70l3cR105x9VtYWBk;P|28~1RMHJ`>?u$Hm@zRq zQq*i55olV*sSo>icnu4*F{=W{I6Gao08Wq8LNH0s@fz2C zr-9q7mXdi1r|U(N?5iE_7SPS@4iR~76%gGfCM=f{!)YNho>ddlWiHDXoZ%wH%$eIn zin?ZTA)%x+Z@ZT!npeKAQSDgyc^|4QVnx@A_vGLE582lmJJ0YFZCi1-8b8xHWE<`y zHXE1o2)!AP$kwMDNmMq-qXmUEK{{qpMt>2K{+3LU&#H^*K1MZslV^UpbQT4>Y6ezT zk?GL^uj%nHYTmVwidbJlF>#uSpJB2j6AW6cS>I|jpO*-(3lt&ASMLT#pzFZvf(O! z;tsoS=;*S$6mT`6Q z_@0{G;xA>uG^I2R(E?>lm8N7MEN;eu`k08@pOzkcfWiHEJkM${JvTf>EV8buLAk&{ubW ziukdgN7nQ{7r`sDRJ=b^P^_{?gu9Q2dZmTUPVY+4a7y&z<8D#GJVt~9l*Q4)_5D{+ z@?aSf7H23LdSF1~xb{@HAxXT7Cfr6U>P$j(Yt^Pen4jn#td(LVl5w&}!lC_=iv6fO zH1gwiGahY4sIa7{=YZ13Eo1yE>+QGdtwA-ns^k(9^-&8|Hbv*jgBof>c^y`l^mtpM z#0J^mUyK${+ej5)7+98SG%O9!PE7~L+gm=AiaVjDdz{(vLF9fq>lh--Cz{Oo2x4VN}ZmiRJYL;Vn4M| zRBXY7r|Jz>(g6hNW0 zK%E13PN12rLTz&G0-ovzci{ZV;jF=<1+=zbVCINv`cq2&cIJ0Y4$ejVq5f0w(PQ!g z7nQ~u5|D?%6*UXOr%<&FfrU4?U>Yadf$NXFpB#gUiDyS}akMK(n z3~5OxR`WAuPP;Z*AMTFHnV@@2j)kTb^bsQ?u`^+7^AbwzROYXyfg*g?ybbD@si;$q zU;8bCLE5;glyc%javLR^jyMcqpe-=CNuOGdm$$CO=;{^}5;lLQ``|a9TOI&g@UJvr z6o=uhES_X1ByTtHC8KTq+E%oDWx+V;7pz&R($q`FHf;55Tpc)qSq+oY{Qo$6=jhs= z>|Zq2iEZ1q?VQ-QZRf~4)Y&p^@X%su%#IIv4Yu3rY#opGk@ z%ksvTp%oEBr;4P1~(ZC&TVYuwtYhMnLkvcVl1{5kiCBQadPT7{|D%KS6T zE{*yoteC6xqI~m&r0G_2XTPLvEveu|d+a-DRo1%n!VM=?76>=48=I8C`WvZ?d2yhD znwI>IHn>x`crAU(&4W-lD30UKjYSxV{c$9A&*LL2x*PhAITTw(+_-L77|_+14=8hG zAuRgQjPXiQg|(Ed{a+(W*67@sKpP3QgY5MjX<+K!zi=_RA4M`_tV80zoSKSm%r|Re&^qkWX4G1KT(a}R(doScB)R?EG-HLE|UMqHhZec zt6rW%Uy|IzBF%G!2A-_`Qj{_`Z7ond9b&5{Ox#o{+>ptmhbIcQEG|XMN;rXGb#5M7K{Ky z%8p`vLU+71fK}Z2U^&um*$sa9Q1czY5s0S=8)zA@S&48?Et`6xwqW}mEPH4yluMS) zO1l7lwz_9HLrE>zd({-;#@i2w@gZ2B5A&ZKCg}@SYCG6~XB`Sy-6h0lb8m31>gM6T z@M)K2FuimEE<{)#>3 zoN3tP>hiZUP0+)L<(tiaXR14)I~=-_r2-(NuBb7YR8{9blKi9Mf0#Ana!h(rz_8!? z=d9mS#T34ueT@7=bZMyfFqSP&QIl(RVx z?r_%pE1C}T7`Nxc6eiR<=WYQk@Vdcg8-VU;v)Gn+Xi36C3p5?$1Z&;Da(kjp7^Neo z!saZzAk~i?LcFSUr)pI+O>IiiME7HGOEW!~a%ddpxk=zmsi>)bGqd!ehzl#O20K>3 zcH+=GBi0^W!Rpyfo1R+vDUu@%O= zG>>Xvh&uodXS9;Z`9u2xqKFxwpIw2uexTiL@js2+$B1f@-fd9KOG7jrhJVLV>NKlN zp1&07c-v43HsjsSQkO2H&}K%7{B#g5>-P8Osb>ds-kC0 z!5~45jIidUAFf;eaYi)^CLnj=SlarQyWVEtRbOxwe~4*3Da_v>&BT*7#sbS=t^gr_w`(aQlQ|3_M!Ww_{QiapL*%x_$P7e8;<2Jk^y$6Wt@ z9N_AAv-E!P8J#<~MTT9Qw_hlLEz#k@&ib`R&sGJdPm}k7-grkBVznCl(b5Gqy7Q1qhy-YJ z0(~_~yed0X%rmXc59SBd5aBMp_m>j5Z>F@oDG;v$QiEniXLbwgLO9*# zw+-sirMel1W7xp_rYl0u$A`QhV6BEmPO$@L*LnJwO`?tMj?0-lI3qo$yZ#AA(Ckvp z=qW&bN!bd>>ru?mKve{bX+8 zT{_0*l(OzA664EuYZ@3NoaVg{!Nc7akPHAdNCf+X(oQanQP-w6Kgr#rO#5ApoIB_O ztW~vI*|su8S+^`uB^lDJLv*CZSA;vw#rDKI^(=(D!t7}h*>=RL;y+xE;X{(IP3JP8 z+9@j1k|##UYD|^(_iODOHq%&~w0Fw&Jj%PP>`Pl(*48-bUA??FZi8@(YxX@oxYuo< z$*RZ<+5066Oql@<9(i>vw7ljLwlfs$pVbWf`-FD7!6^!9jdIWklQ7kc?u;omZK5t; zG7=;I96%+MQZRKh~HvA5qTMs&B~FIDzv6( zZ?wgH5RE}$4e%^lC;-p&W%ZlHX1z;9$0d2W&UjnZ<{wmE$lHGE?kj;(np_U>?XUngf7SY9xMadmNE+^yhKU^g(^u)>ZQ z%zX(=4}pc@a@C7SEJ?cqz8Ltsf7hv!i#W}ZLInW$D?|OC6ZTl&xpR{Lk|F1lvL-3;>(PMn~_E(Sp9LuzQ+&|vbU*ZgNp<-f3+{cDNi@2>ezQNm)v`u`0|ITHBK zVEB%b{}!14)8oG(DA}0sO}M|WzuJEbf`7)_zoG&^zNh?OszZP8Lf`Oz54FFvInh5k z|C0)ubF4u4z&C92i8pAJ(tj5i*<=pbX~~#kSSCqDkE^eqpv&8utdJ?PBC79I*W=ab zXbjw~vl8BbJ{?1L4uq3;HfSH-%;88KZWVKH76+uTeIG(Ht{_)w!fY7VDY;>OD3_g! z)xO&Pwici*{k&j2>Cv#CfOJHlqtW$*DU(`52TwhevyO{n`gnuV6%L~|fy*}Po{8MN zs9b!vWSI7mU2S7c4|0EHOJ>0eCwNIUpPz?&#h(6kOwcK@n`TpSFx!UVM}jlnK7;Vp zt^&0$z@O$8D#x!U$3tD)_Heep7O`AoC)3WiXZ|wXe>UnU{5SpN_ve3Up8oHVV&tIj z`akRT|6t4uRjF?QCiw0%Rp>vzMg#MVJ3zy8xeJepzyb&%))Vx}ncMQ4i(S^Xl#LHo zTuoeF;>|`RJdg7@#sBb{biW*CtSzk{)3gc^DT6v}nCPJzG`!Mbxl#l zJ?G%bbmvT&6sZZ3Y?i}2ch%n})XG0SuWz))jg$P8irBGPS=?jw;Q@Z1Ji2RXqqGFv z(7qs^hKB}hw{M*a&>0w+ULA27&ZtH)Q-lJR9q59GX@$8gm2$iLOn0et);{AIs^2aP zfFX(1TrD7YotlWhSk)e2sA>W%TB3S8qGbc3B$CpAmJE&eLdKHc6apV05?-_t=4Q{6 zjD@`|W@u+%bj{}lDYcb0_Ab1~m<qg*v3 z;xj+EP0#Js?Ppitm7<8+O$82s>%WM^^mBU~9iCQRHwpGVz${D}Z`VS}m$MQ?7(~je z;{2t~$6jaU_vMyXisy{UpspinKMA-f_&{Ya%Gi3YpVc?hWTmzeBswP^v?jo^8AiCH z4rpl}``Cxfi=~fMB@UG69~HYaERdMRh|P!KeY^Ry>Ub_)i?Dd-rw$)RfebUA8Qu?& zxa}|%j=b1f#8hZF)V9)V@PmYD!yMm%mmJ{)>0Q_dEd%qKQh36Gr%}~BOD?r+?+)qM zj`nF(S7)6~I&vb(O1m`H3$>qOSNr7L5vgS_jf3PurdJ@0;Q9Fd{g#C=MW_67jR>b-VgiwC`@RlD#sF^D`>#_4h= zR0bhB$RQ{eEAXwxZY88~?$TwhM>s$4+;an1+scIlJ6~A<*YJ-8nh+2hQANbf?Sghc zusFj#jS=S)lf~{);+KyogjOwCUpgK7Fuf&>S#eSdnkqUoKCx`hHY#rDHu=i z%i7Q^Kkh!qqsnoCHzr{nBX-a9lX|qLdV8JDR^};H1>&uvh88 zg&qzdxxfi4xS3;OW3*B9)QnOQC{h zraPE6nYcgY>(@+TW0v3x`OO2jF%seD5?O)6qijll5N7o2_h6=#(Y`3lI@hH_SJ4Uf zzQNjR3DzQe%))xlzOSh@>Bq)+$-9BVZGonm>!nMIQD=MA9~Y;DjC~Rt-u6NL#uV|B zSxwU$r_XG3vp?0g{=%}do_UCm3=AYZcR^INSW~NksSWKm21KJ>&WtWkb?;`)`lFE}jh+tdY2@ z_i$FXM}SxX2z(r_K3d0UVyb8t>#-4q_yptBIRG>!G64%q%HNvexk;1gp zPFiqEKTvNlv|W+VP+p_)od~4`fejwfiEvP?b62q!%*CzWNzwF~r~u@#9w(iW$1SFZ zxX&e62re3m-U8?i7$%(5vv6+)qMgh`^L%qi>h|U0&O4g1=RO`EU$bhMX7|%$iDf>5 z<~RgE?&W7j$c3N7Hm;ktfdXu&ZsqUZ%dC=eV+7?_ZZqenwQt?#nqC3G#o8w>KQ~#+ zq(LyrnrcgTS1!Km#;2QfHCv1%uJ zH`C~ygnpJzt*bX!p!5zQ4ifM?qBr;A+)zTfhA*@dod<`Q+trMQJ?}(U{)Qg2D^D_1yFIdUbPr;U~kEn@W&5OmqrxKg2&s8Z>?CRWT;`_ zpcy;n*Bzd%`6P4yEDiK%-Hs+Q)u7~fqm5z%3j~H3f`FtU61Ms=jM;Cjz*=%D8wxNO zvN^$VBI;cWno{kDbY8uu^BNceDde~B>$&P;fF z565otJ~i2H^oNdk9R;y50Fwx4&%99TdFmY{V$H%b6ivV-tHP8Mni4?JN3Y8a$VD&O zIM7^PX!ht zI~uA7rb~wRty(sQws3yuB?+KF6 z1O4d)e}sdlWCNv5IS0D<0HtC_FE>}GnQ+)1IZ4dOAkphI=74a5HLOvgFG@4OlyY=~ zC>@~B+OP*7Rfub+oC*<7G6BUcMQtbzS^rf^3IgLD%m50f6Q@swS7?3-MazS^;0VdkXdcvu7U~_sJKq! z-%Dc>)0$dcaC(GQX>j&1WPb{w1c#U?R$oaNiCHS-% zw+b?xg*S!8n6V3ODB~Kmo8sM8;YcxOJ8;Cz$6KE3-C;ZO;PJ^LHkJq6M0rInOHrJMPk0fy&HXY zK6X(Gf5q7LeF#7c7M@GVhF-FDt$s)^?y0v*nhUkpX~#NKi4fn_@;DNY|fzE|)(Y`Pyd= zm~m2hGs1kwtcP3td_Q2cLB;4{$kSaGEwvw-QPy=VtD5|iYWeK=qE8+ zNEuD40n>eWdj|gmdIxgxCh7L&@B_=wdZ*19aE*wdkskcr7a>u2?*}MQ9`ojRpp9;U z@C0uhv#0O||3S3-G#xthlrf-v-P3^3IzbFOt^bAI13+gMYFG?J=-jMZ3_E*@3XVY4CVIq_DdeeQskCFMOr9L zzYavV0#G!6L!g<99KpCn#%*_M`4qXw%e@egqR0xiao5olSz7-v@0bc(G0Oq&;*5Z& zfFFl+#^BxcdkxR~Y+Fg9Q0v4^^W5z4+fZ{TV>Xj156dC|k=aB&;0|zmD2(tG+cnG@ zXeO>W&xv-bgnmKc^k?)hlrw_!(!;J}mBA9==vt5-iF+Z3dk?TxqC*9L?@QYp-jSk~ z7$gp~SB*A(E?45nm^psmBjp#SNRR-O{ocX4y1Fnnby|#SJm}3Sw;s^o?09e>AKM^) zzL2o$2_6pGs$025kE7D7S2biE{fRY9-|SfMq6 z)T(O^Lip5yGko^q=X1D9AiHOG0Wchr(QscqkRL>xEE|>fa!(0O44+}BA$g;FKZA$Y z%l6~_%cObH+JeGs<|7FU0)&G{QrY4oAR(SJ{gD+#rD~Wrm@SPCBe?X@vEKHr^31z8 zsZsqhxA%_Jx9$mT7O!!3mb$tTP~*C>BmsEPV$DrflbRsIDy$_*K4K?r=_rWN0!cbA z)4@E3;^z~!4V6TwpY)LIFm~)sDO`8$)UpLf4FH{v%724b$%)uZuwygw+Aaj%Vj}k- z)cK$jAoT^-N6S%P8=gaovb%!D3U02OFLKt-#LdvHwStOR*0l)ql(Bk!6$V~3zln1U z_}1s2n8T^8xqp|8-R$hAJk%K`*qmJ^42Rj)DvwHmO#IKr^k2KF5T1dCqm7!#gd%upvaEoS*pmsx1J=+_n*(*gz+SST4u2NR=s$X zNz8=v|ATd%No1LV;g(vx(g`1Vqi(9PhKzPV+6WAjla|+viia7kl8E4+4aqu;tzxs)%dCuf~yrL_)BrMXYg<3 z-xn;n?^|8V9@v|w2SzvPlxY2Pevu$W!^R%umqUcqI7knvl%D~+BTv;u$`poF+3~;C ze(cla#t!EflKX8+X{ifDadADqu`HNdvxmX(QPrTHM`4b@j3DrG3YwL-YNSVO%$ewTvm1|1_dxU?@GS^=vd)ZbsY z077IWHio-5$YSxE^4EGRdDlOKq%MJ>(zoK_g;bXz8F|9N`oS7MpbrDLnQJ|w*PS5g zQQgy6(T-?e?4gv~SdMGvpIT9p1c?iclTdhGfZ3a$zszd8@fpAGRo-SFP+ zk$HXe#-KOtyL?VT5l!5^K})qp{A0sm!M20zqj! zVg2I-Z`ouLA;&#Nm3nVD2`7@F&+`2C&4GV_42xc(swU>c3JMKr@$iW5^fH&D41uu( zee%r1OCAY5wRDNO&YZR3>lKmv5kt+{PigCSF)i(}N@;x*@8Wbbf#i<1(;mfOjFsNQ zl(we}cBaf^`@UJR7Ie^5rQ-Es%aqGE@H3>>{j+yQd&jnpU6^*CH}w4f7;K>KlOI#Ru_l4_K3 z9N)Cl_#R9#m4js4b8SLUpF3-H&c=Y2iRp_CT8O@zJW3}UCw+8L zUuy-C#W4>qbuUtDKi@LJ z0VIPONLdqBJ#y;EPrU2E!R-N>%bd}7_0PaQs#|YEdFOG*K)$wj-M;ge_UheyaC1N^ z=xA)GTBan{>YMQXJ^#|w093>{0W-ZBn8iB;ObM)#$cSbOo?)1wf+w~m`lU}c|0Gt- zGe}R0n2ee_+}IuZVlYLgmi1#MG)Ran9qv^FX=e>nF%lT_bb16 zIAeKe*yhD}0S6*15xZ;T_Kdvnz-_4bQLGMQeKi-yZKye%uVnuiGyARub|XfI_3(-S z`Z$glf>H=_)`+)c+2D{Qm16G?oc<`g{$!D`*gg^WVjJhJq)XomSZcwMvvMsH#Q^xg zxG3HVa|uhA@oarl9N~cxRKg}sba0*q16eVoW3RDx5IDF7?>h1Nd}H**N6{w;KLb~dl z5jP>?z*kASX!?HF!(J`XG~(e%y~PQ63LD@Z?EEZLj00ilS4@yjAl)J8R*Y8ExG{mt zG6sWbwo^`Yj+~grV{58k2Hc<)Ixd;cL;@Y2)Xj>j zso_MOoNnT@G}N+oM20Ia)^q#CHT}e7$x*@Hy55wv(eHkm>(F9?xC_8#bgO1YJx=4m z!IM6wJc140=TIGO$0-7>GFN9jovMo2q|u(7>(Rk^!I(W`D2Yryb7s{vJv27b**Ru{ z^k{l6_4@OC(Eq4ElYbaePHGX48{X(ogvKf(>K$N={nzD@RaSjX8~Y2znK=7;l_ph_ zMBQT*XSh=PW-{n4I&x|S(o9|ikYY=1hf%o~j_8*xdivD*0e(OdQj0bc#;b$VfIXL4 z0#>B+AuKCh%h23F)|NfHI)aTT(e4QB5uAuXsrNt@=>BhxAWt|`#6yOyt7X!G@0@`w zPyxN4P0~mhs|6cQqD*8!*pYT>h5Lqh-+X3`wcd<>p}O z)q`^52jj6sjzw|(apf;@ml(wB46Vf4whAo#aA#h0u+U_U%z85CE_Did1O{XPaPJ1e zLeL-Fg;HxYyJS8H1Sn87(ANBpT8{vct%ioQRw>O=6AlwDZYx3F2J+B~8Fj!TyZjES z#g`xS3-}PkT5nXal;-)SDlgO?d)bi|q7((_AB z-yK5D*w#>itgJ@eMd994M;{31O6;dkjZ>`E7!h46?53b0GsIywTKTD*06=s6VCLcm zvb*dQ2aq=!l&Nm1#r#^NLy!_kIwyDr*Q5Fxtj^6G#q8B-sT4K8^B)pwk~})b%Hv^- z9APaS!fIdsCggz{B|yNl%-_b?F~>?mbLJb^W^OKFq4||%vvF+IY7)+tIrXTC(XlK1 zY(XTn*WxzXK+Q;Jz!ItHnd)l*TWr*~XvnEQgc1&w*!2Za+8pJJQ4)~OpxdZaT9r7) z(g5Eg3Uag1Ktbh*op}xEFmK0RJS&aiQ~vlB6q`N&HXH%hJB2?!L_P29-Gf+WtRxLvQ`kHCIA8&meS z%tJwD29H2~Y^h?`AX5a4m8fwBC=ac))d!}6)W!I$eG?Md0PSYsr%(VGe&V_E7bvHP)Mx~ssT zDhOiOR7q-i%QJie;W`>9eK{;nl#IE&<@5BU%#PQhWO8wR{Hp>U+NpQ8S7Pb*)B|B3 z#Wwk1GoWKAfP&hWrj?$b5QO%zV^JnEg0N(h^N{7B!Aa4vVZCx5$DLJC->8b9mP!Xl zIL2Hi(FY8$U(nmsm0;!8zX>A4%X&jl7JKD}@?h(?2}?ZTP={YV5RdV&7lhAefQc=o znKV4L2ACXGglLEJyP=}+Z`PrhPsJX3nS(icX^M(Nm!?Eu!ee`{*nH5l{aenr(-i43 zOD>nAzDVhx`^3jCr$)|&=L7i`^JW+7Sf)Z>X?$v4PluN7c4h{)HlvlwCh-~Wo z$)bVV(F+*3&!l@0YjrBvppz`Br$ngeb>2H0^rMXIQ$pp!sF5Eyo!T2dR)P$5Kn>vz zF7K)dlPl-3Zj@bCzTIEHm#%cVz5Oqt)MHiUdc1F@kX|Y0<-`Oyv~TZSz^7TB8*5E- z`KJsBS`K$I0!5@_o0cN1nT*5btoDROe{w_i4O5LvdCH@NN>cvmuYwlf@!pI z3B>u!GONqKp?`57`0%TES3`@~At^QXTun4Pn8IO5J4_5#khAAogIt`>>gQb$iy#}Z zdzJvbaE`XAZf+z%XcXZ||Aii`}K+a>hM*1LU8IE z@Oi#d(X7z3WoSNE7K=UX709L6i5qYQTR4AZ0!4HH*%O-2!`{FrvE)PGA|`Zi!Nr&K z+npG^$h>N_l^b2u)gpYuo2okCMXW7qM%U+qAJ*)=g$;V$JC&y>@F6?^Sxm#%!FxhSQYEg;^qjp4{qlKY zrgg{lTw9xt$W!u3CS`2X9bE#wZU}%bukYuB<>MiW=WAw{&%>7E;~QGdPUq|I83szt zk%9ifoXFFbYzxHO9zNAFRMHff8;y#O2(`q=oQ0)PnhlA}9A)R<$tWP{r7kd@rX4kd zy%5v(9vE5h;$!L^+n`V$gEj(jf+193iCNhWI;%Zno@a50gpz(Oz6&~XN_e5^#ceu8 zA4dgU+S~}CjnV1o`~BiHQD%NFzo4n~?>CAgWxcdZZc?2B`vAp6r&u5tB9lK8;=!fTPHzTiMKtwM+k>DAPH1X1hUzc;m(mP@|;e(uMREbh96OKQW>KqasjMR3x~{c_#CGMURDdVr>7l|HY=Y3 zP(DL}BE9T_gHJNEl0jIJqI6$Gg&?PSr|Rxh3aeie*|bE{suc$-{O5r|mJUz*O9Mxd+*pjF zR)G-}g$ISLBfJ6O5XD=YZIrY3?x&_a2`F~Tm5sFR{A4Glt$z5NrEMsgyLPys63Tdi zUzoB(mhz*UQ-+m)k^1YDP5w`npF_rZH=EI$edkM`X0+s_1iK=0B%;GnycV{k22|U| z1^Su0iD58#o8{`8ScB$>WMeuFZg}IF=pcP|i%MrQl%^{-ik)la+p-2VteiZc0FwCP zEmcpi`cYOYOR`&Kmyit!PsIUc{qw5!q*6P#=23@R{sudqR)^Gw|U}hi&cE>&P=7;_ zTE^HNSL;5Z0(Y2p?v{5Q&s=&gTK7HQiC|%A@PM#H9*~CyjsAIE)Eutd`ct6u^Kx&A zRHU5dk?acY0E-Aq_lnXz>B2tM5-;4Lrr=ytj0EV7O%Jbq8@U6iXZXlxyY1~{-RO*W zaCidrsv(aHlKl>T;9TF@wLazsj^SBn>_n7J6)u8b$xv_i_zIXC#Q^j)+uZ$Bno#g_ zZrmd)10>xWMp=rX&P!-G@5~{qL)Bg#&c=l;%-Fsf2t~4*`ITjNFQ2$`^H8fZMEYB0 zoaIvRQ@Dm`0eHeuY0;99E8^FNo5qfZck?>M{qT7tP4h0~Zx~6Jn~k8?%P4-+vOfTq zC|z*wQ?`j_hgm!zOKlTNK=#LzNuAcZ+VE#^1si?9ZEm>oMK2qtGg9S^KPNy-eWm1!)#U+5g zN@82fIhg|mFI-s9#`4S^ip%RIO#`NAo^9KDTySq&$fY50s$g!*$e}sJEshqH8%UXI zBh~->4FZj~vPbi0nne70TZ=%|@%Nc#SC?xL-#2WZ-tH1fz~7nGlT0>Sb0y2OR`~AG za(sxYj9+hDJ(=~J+~%ovEO$khNT{Ug*%6`0^bF-gG;uPgaE#hU-V zjOmC)QA`W>NM}xTaXyY~R?@eoIBJu9Gg4hGw(U~8Y)#8}HB*=ISbM@5NqTevUMDj=~z3!j>6_nO#m8S7Zf<4hF1m~9C~p^||iEEsbx>>q`g zkGiL6T4A52ZW(zQsnJ%Y^~siezetKdQj?C5F2c&272*gRrYRi(LixhHj~HOX(ztH5 zi^{d@qcPq9MKY^{!W+v;u0b@ zEk8w?N%T#K-lq`hklBYIi@t|*obCPS^sZWa$`8Y<6~NiF7Wfn^2dZpqw;yoeeC%Gs zce4!YD&6`0L46VC$JK?74*?iL=kv6H^yFtNA{FiZPz3%Pmu@C<&#onnYO4b@RK%W{TWoy?9SN==s*EN&3$O8>gzi zim@d*2m2(PpwefIIENt{mkI*4J`eVZZ}Cj2Ya@dow4wdEAW_L2UHD=L>D#3mJ!1PN z^D&B(rvVE!q4r;~J!*i~U*d9I=MhV`sa2=OZO-NfQ)}iLDYaq^m;8IL9^(P5yr15Q zL!1cq4}eWL2`i%vNyfMjP?`lV@0%8l(r3nmVttq7!soiH9G>@~+8}_(R2Ia+oGmrW z;}e{2acbfQO{HQgRvf;7rAdhvS56N32WUR+0EVVI{C+M$I&QN(92}s(y=TmsB5Sb3 z3hXJ7^zo}xr3P2dqr}LfN%LVyp*d@>E+9Zpr0-xhZbAnBV)`%n#67=VAy$s zUhB7&@EKn?LeF&keg&ss_BPi5XYj6JXHLVmWnSZ)*Wb%K8QRmW z2ReT(qFfZ+0rf;K(h0-AA4s6_zw`ZtbxhKKRc9pv0Qd`_^H1o`zXC{Uzmbr1|Ci96 z{{)czKeC=sQAkQlO-M)2P)j|& z?;8Nwy0U+ui~jmf|9?Hie?!zF2k@An{;%ubU`78l2>8G7pZ+Bc{9mw4|4Os+4_o{v z`5C-#>XrXun15li#s3?TZQy*)+TV$6zsYMwdy=3}`nLhCGJ_7RJD;NJDeryVF-H&VGqrl}W+>M7n z?C&$(F-q<2KR{9X+OPgOEJ-{pTt}%B%=$SA_jh7iA3Z;WZz5cVzd52AzH~VGd-DHp zZDiK0sDC)3>i@TC=)X~3e^=(=K*#!JfixEwH=vV1tXT8D4^YdiqWc;&RSHQR?Hf!4pMG;4t3;tb1 znjgpy;J+C|@)mV7^n1F0P5Ga3@~;bLLGZsK{jHp>4U8Qu^bIW?>Hg*ZA4s=v`1OB@ zjwx>Km<{&$t51|p++Awx%}WovOSRU*@+t!X5Jk&nP1LW=rUIBABno1 zQjDEFW*2Ta)DIsJV}x~=owK7>De1zE#v%*`VOeHFvlD`SZM4*u1g|}EvIq9P*+dAL z_z~~Yr?NdW^Y^0#Q+J{;S1{5GOa5rX`bEq^>o=} zvgkC};|9_B+*ysS)4m!&=Z!<53C#%xPn($&T$57wa_uq=?|+? zOWZQr#uFoUszsUc5fb&)hhn?3|DI&fV={!w#RwFkhw_C1zJwtW33Tw(yfhqudtFKY zjZIDWq+AF74r%XFdP*pCSwBFkYqN?<0+BimW$C@n?A2izOE-Z!*ut4OxL16r6^YaB ze04&DK)!PqFHr|J4!F@=Wwsj7pgje;*I3Ol3q)LG@}6U@k;_VV{U#Z&uC=xb5E-w= zj|K6pS)BoMDSQhQ4^7xjJSX;up$;7go@KdBETRbCb%5( zJM6ds7$F&we;zsbdY8}PtaL7iJKD$d!Jb%JH5Uw(0DlqF8kzOy$W6lM0zt*qWii zNCpbL1vZZl>`iRVjITn|Dp_O{0Tio?a)LP)Oe%&1S-Omh@zZ0hkDP-uOd&$6s`4cZ z14Gl#f3~c-`2AplJxT7qw4?0uX z1U+}j*<*027F2DkxI75aS}IY#Xf)tXc27J2`B=yqw!TgeN=$V=zSv@(P@b95E3QI9 zb^f3plujIzPB&s29WmJOYY?NBKS6QLj1#9O3P6M4mmY?D66OHfk7Ys>?y;6LiUBf+ z6TNPbqhs7M$eI%APpMV{Cv)CezfhCp#G=#q2&R`V8f4(vMp zE*)aw(BJzj32~9uc$<0*uP%e#QDVvV01>eJYz5N*qj6(Xqu3)}R#>}3SrA@8+W>w~ zi(Y`{Ae~>hg$dxf8tEOB<;bo*AU_2T7LiDI-Km2V*>!a_>oAAY;1hKk_|o@ZukP8O zg`K?~M=%y-8;A_g=`wdaqcP!DV%TSCdI7(WXmp10+1|3-?T;3Ej-97nQvZ6c^ksz< z%`q*OJM6lOHDC?Cq>f3TU%aiEJssm=POgf3PekQOm1MMW# zOt@I67x)2WVh7iah*XDDGqx612l|SXuFFhV+bKL+J48DbxsU?uPB1`s;YAI*Z;7=H z4%>{Ry+2cSWBoc6_>svTZqKLS4@q-(xJ;g$xvg-<;mSI95N%7ff~E8kQf8 zoIo|1M`|RVelxJmVEtk1K2xeN>+3<(kAhNI*wC{bFhX5Wn)$^WQrc=6*v4fcq7|Fr zE@v>l253>)>U{l@apoUrsrlgQ-IzU7L-mI+G2Ky9eR91QDbXgT<)S7be9Adooeyn7 z8Sb~ahM22Ei*1BEX#PT257@|VH3&tV#7qrrEl*ElBON_E4VkpNac~YZjq4;^7=8#P|LXsmWR%7xzx?Q@GYScNOlEd4&U|-$#CIKymw_ZQ}QE9 zmCN`qM90}zJWnd&_*|(hu?>Z&&HaOP4=XZ(enPGU8;YEn)WLd?J-_4aXbTa!>_@Z~8N?#nIWzj@o~4 zTf!ypqUoqAfXSJc9|7Q!_g|QAItIWqi6JBx$4kz_=1y}`=DF=~(8`Y?JaD;Xm4I^M zaXs_HCc(p&1|^63!o5KJqm zL~*sal^p0811ULo;bsw4;gn=SLO@#*i4_8q%C!xRy3Slv;%4b`b9A{OYHe~USiK0* zdo{D2i!glEL?iQeQUUry0A+|38$mka<5O?Yj0yZ1mpI47!MHSUl4*q^MWQl^)Qdjb z>|31IX~1~98K6ne56Po`=53z^?!N_?9liVAb9~+?ZP3oe(V`KlNFPCx%Kwh)Tn`;< z5^A`qX?_0w1CNzXFogdxhxZpMEO&oFu|}Je#q4hju3=w4w$%C#ND`1=-iwVuh$dMm zoI}7T&6Y_Ex&X==&m5w2!h@7rwQru6*&zpTG3A;{5o8Fez*vw@NL63c@Qj*&CN~)|*82q%~ z1zP_-(x-AEwNE1H$m#md3Z)xZOKYW9fQljorFOe{)RLAv_ZVYSnrH*==flFI=d}o= zABCqbkg>GOndvcXo?>nI^KFH;*@jcR4R1w1x)d!ZaD4|p!x;k1PVz+&gDIg71*Jwh z#QTx+kCakxAAYLYdDR=FmK@DjRrXHQvW}c zyDrSdZBZOz@RMVGVXWEur%C?Hx!TCbafsiv=7~h*@4_rgj)w zjT+%nVH_^Ix~fTHDlK~pBS-z%a#iK?OWNxM!GVHc9 ztb=iKsCR<6caeY8Aks0@qvhO!k8M$eJC9RP7(I}?jS|m)0lmb3!WxM@o$P@pv4lY8EY$5Y2nQPw z`L>61xCOfxmULbLI``CemK?#omDN`=*^iXfG0BLr;WGL_Zo~AtP|dCF+lSJMr+25g z!w&uk`W-0yaRMH05SNHBB7snZ&9#m)wE|8wA}+zf*z`;M8JCcoeTwmx)}LjItsgnu%dN9(S*t=H?*4=XXwi zJ>B<2QzAgZGIT8Qimwm!ev;N)4soC^I_kmwcfdGDIs9}!bbb9`^8Ni8|Ay<2?(yM; z2M&aOrz&k5C)X~~3M)-l>%v2JQp@?a5j2Nq&kv`^ev(Fl8g}sWXl?`VaPl6T` zn<6FaajBk+B_Y*b>H1OfYQeUL%$o$i9Kpz20QU`h`HU|#l zx*St2D`tBMgIvwrYRYj*nuwidP|1^HDM7P$I1dEj2Rx((p#gJ zJAgpFP%oEm-O^0r?zxCI({<&>JE{IsQvNFK!upe?g^tqM1)UkgFkL3Yc^VFfuy81Mk;8}2SYa+G zS7$B07*)$TBVmv8KBh;ayGoJPE_LAP&p_}D@JE{uD;FGbBstYgL(M6FFi}j=why?! zoEoiN$KTopiTICfgJ6%XwSFQ)?Wm)FCrdCPw_)d*&L7~sSf|XZ3~~MW>7J4UHJ}=G zXuBD)G`4%jOT2Khs?602#7B5Ps$kwod@W5ReU8;we(&8{@mt4X1EXl3U%A>5gv*6L zotJYm2ECN^S%c^&PEG4wWHy4qm znJGHlUEvUUq?HuG+}M+0(N@O^k` zXv*6;U#kmjpFYRL{sE=QpBmAfR{rpOk-AHEYlJBx8H_PhXSjCBmcq;g-gsSkbTms4 zHmRVaeUOQgE6Zn=l@Oy~R8}1WWM(W2fHvI=Je969hREJ`f?8&TkO}p6rCAufQ-0%& zedxA3i#h*0pd$wkPw2`i?^w}e3p_7!_5iLB%6G9LI74`c_2?1B%EdwPL~%FGZS7uq zw-7>~t4?a5?+J_Se$@E-C(++E8i#8D<0+?6c3TZJPt1(+elsRb9qI?jw7P6Apd${&(47#EwMgPqe701o> z%}DsN%CeD-f=fW6aiHCdztp>3J)(tapAXxD;20|ZOd4UHTdbO z*6Pd{Mnk=>w`}T^YA&s_6#WR^S&0y0x%vDPp)EA4QEaFZ^-t(#K2)6Itq#+XK3vOhdiXc+c@6isN z5bxn?xlOt8k0BdJfTB6W)$(yZL$z3M`Y_)xEQ2?2?8rlX|fgedUEQ_NqD*hNqB`hq$Fo zApLVsIZCm~*r|Df5Q~7U!%Ln}wo7MbOg9d2TLyoFPX-`+Bu84Z7o^}<5V`1Fg&%>p zu|5;wx~bYTt_g%Ck{Ey!0+OyW+Zyo4xjn^z^vt*_m6xkl2Ikhs0E8S1V(gLNdHrV? z^t@;;xSfPhM-e|h<>ugtxR>-FtUIBOP9c=Gax|9Rc7ej)4)K#!tyNuGs`xY8hAe+Q zWvqm05S*$OF89|~mS=CL=}QEH9jWaD#mNs&^GC2ypa!@cI8q&#?xK4wS^;wcZu`r; zJ~HpL9G~py`@Y)g>+N4-eA7PYwVpg@yw7RdKcQP6w^v!&8}{dpl_vN^QSxr3));w3 zm!;$M*zb-n?_%H_ve^&`y9yr?m4eE%g^B#WXk;%WrM<)51Uo)w(mNb)w60YH=#!^$ zf`qHbhoGmlQT06)RnwNy-TZhS2&k$oz4)t&Es`Pp6zzC(KT*MS40g`$2c=gVPw|1( zdaPF+|L2Q}Cz3G2*p{VZQp&zBmw31fRBJ>QcT@`A6F_bQBLv_H&ZN}m*nD`s_;BE5V#a9?k4-)&gB2&Y#T z9GLd?aZJxwTI-cl<}c-F`6M*ov1(GOqSYKr#uQbTu?6_C84B4|sYj8uTFM_e5PwGj zmM&LM{cUlemtW&h01dVs*UW;NP)|49*T0;?`Jo;F!;k?0{sDde9s!d6z(J~i52pV& z?DuaYy8avf_aA2AYTOW32&zG3QW2DptPaLJoHtx0m*DUvo3fRe}v9#jgYzEtyc9WXv2#DQg%(F><`^ zLyWICufa)RxAzM4_0ftIm%2)Z8`ome{08ELfc4|>+EG+QaXOax!N`T6oOEaoDNZF8 z%qOT=d`mMPuPfWiZy7kk-&JZXX~rn8K}YfWalTzx8-)gNeqZy=-zPN0@YBaDpu`pq zjSjYlL{^_qA#B(eB)y;I5Y&pc5f}oB9)QAhd;ntsso7>>at`<@EWrxeG)SNw z2ib~k#7bkqxMMZ)Iu3};)W69iTErU6DFd*R!Q@3i3Lut)PsPp;k+(jC zmx77VVnPG-iQ=wZ7Q;#)5}eWq@EPE4ED~i1=5E&mHN$YkI7Tbg3nAc!#%{z$CLI|` zGo)c4*-$PHARy~oy_co+0Hh5PFpxCmIOYY3(gjpLquvUH-GjiVFCx)jOagusn+D}S zNdvwWl%R|>BOkvGgff9+%wJ8-)+ka8<)g?2uOiC#{gg@#KOkb~hozpBqI_oIpH$uN z6^WzW%g#VnO=M35;08))cY+H)GJ?IwH!S!v0;9fyRE3!;s=*aq%*QBIJOw@R4!Cud>LZQCEAr}Ccmrh(826&>?!jUN!D)Q?$uPtPxx19TcO zo@A7W?4ggiR3d(qlrmsn;WR^I66rIl${2&x2fQ?ZuBI0ZU4w0rh}}&UqCpU0LJWMv zP}0#ZJc*Nhz!hzgNje+w_uZ>A!jXHcGg)ry2!UHX0e?bZ&mmllD83(Z^~nk-CjK)x z32bX(Z0rqpp(oEC42U=@)yQ05bka}^d8onm z>bAq7($g6BffUq|AG}FvgJ{V1fmr<=CT_?(+IhAdjRZ#8Ag#k7QnetxxhtTnK|m0_ zwVCbF7;nZrWIb(xC0P0C0kO&_3Hqv_C+Ro>#v`b`zrp>?0Ri{+$%p%CNoiDmcH>~q z?*)en2{u`+);n>0Eu7aM94?A7n$jNFsvo;SUP2)>7*vJCY)}gg7-K^_+m!+G9x}J_ zMgcep$0Y#>Pxo~(F#K3JLQo1jaG>^sIcIR-_A(HNf- z&>5w|BE{1QIp9hMYK}ng#73uSU@2>)`l6qkX~Y{9Ku1FdKmj?2?ugp+v;*_{5dyPZ zGA}edDBBn2KhfmA>;{dv*t<<)(as@22GNX$;(AiA!%x=RkM6X)ol%h#p&Ns|SIn`{ z43Hszew9|26m=PD<3JYzHoHZQ_d+7C*+S$x0Yu$)E-`Orj<<0b&}%MJF_USBf|{g+ zaTyQ;)eYRYr-^gQMR( zv|Q)C_9%vD%7J{B_VRN44CTfa1o&Ii)}SGTwSfeBoLr2pN1}Ol2ab0m0dvCu`|1X* zH4bsj*sZ{Zhpjqq1r_T@O86kYeC?;etncS{D+^)Ws-qd}s=QboqATiW*MLnumJz-V z9GIWTMBPt<0BMYTx@HEeZ55ulG5qL=Dgfoi1-RvafQ)9I zu{_*fKnv!w0SXX8Ui0=*{*Wd)Ws~P#Ne1c?>N3G5d~8GfkzJ?yKw2m$?wy;@D~B*fYY+Axddz0^MJ z&(TkvSn8I3j;8d0D0QO2aDt(lFxox)PhzKqJewG08dEPxy~fhrATekxZ+??;)oSzo zyCc>Sol*sfJwor-;3JLMcmyJQFr}vm1P2=$tOeDuA0PB+74e-o{fzb~B{dixGO?TO zkuwbM$Wg)%3aQE)F|v@0c72p^&`Ak?29-ReN*vi^ zf$*lQJq3!+jgx}_*DaP0H4}3_1$n(+6>#u7Zo5#DcV!iQsvq7p*1E0gIsyzBXel<4 zUQe#mTymzs>f?)q;&)%3*KX^o^h%zD_{sxW%v>=L2=7eDX8w6hB=tK#)RF|90M8DF8`H z*vtzow~RGJhmk9UNK8cd9IZh7?X+`u;`5Gn-{BS3=4ad&%)AVFX81pWIOO(wjzH>o z`leeMtxeWe$(X1+ez@cSjI68*$;W*Gw-1~Bz8f*H&2>9TTjHrRl}K{MHyuC?$;eT> z^VW%$J~fpOq_$nqy>JTf0HuGAbQMn({Eo1o)+Z%k zPFp%!hzW9tOsPvtW}&@w2^DClwQS)^=2e#EIyK~vOnFaV{FJuRyKm18Z_3k}v_?PN zELjTJG_mze{Bk2d)iDjlr!BurHZ>lc-YlT@=~50FlvDe(i>`YNphqynR^O}&Tn`Qn zrhQVvNdkom=+EH#+&nWE@`TXSp+qqf9na_Dj`kUL8NPAlWgaouhs&^5ICAI~1a2mE zk8WICae?+O$!A$;%E4$17tBf|Yj*}Ssd`GL1yo%l`M8OCmRQ?_k#bg+qYB8Z+LBJ# zZJjFfW)+88yeIifdgbDnuc39kCAv5s9?0co4B{CqsG+MR(UyJJ4=(TXVnWH+fqdx9 z4sqxzHb%N=ec4aBEU}vVuMbxWyy&6eS)pV?&cTVOq}MwaG67dDtAkde?s@9e=MW;P zR1A(x_w2}Z>m>K)SlzFUI>$BhtEyp$n_?2ie#1U7Eg)!y?bkhBQnf5%Dm)E+^|W?5 z-2h#m)_uXZl`wZ?s-do&8iHR-Y0}k~x$3)^pzLJe-ag4Yy5wm~wiV|aTA9HgJnE~9 z%v**pQ4qa!8T@MDNTog8pxImgx( ztu1s>GKK1WjL7BFn8($b)M4#o&o%FG}+ z>;*PC`F+7~FJ%odJ(bi0R{mPLi9Ot|$Zq)<&$}bxs@|3jY|*3NHfp&RdL1+w_tz5K z+{7qiKAz=s(5koYuH$g>6b|+JB=KE98~h$(B!oD1F6pW$D#k^^rA{&RR0}*_P{Q9J zeeN<0OUFx-ZMdk-=E-fVJASmXb&Dx*K9l6u)?AV_cn%uiQ|^x@WO3;7*MRLez7sIP zdFL!M$}XRt>iZB#w-G1mIou9MSVIwqcfngcfevnh2&pN9ddL<~;uwyP2PpCJxv&IU zfDsGfTNYDrf??syxPGCqQZ1&6JEZN?oo5dPFqN@w!Hg>)HyQL@t|^Dt`@R# z%C1=mutg&9Yh5e^_}N1W*@-FFmUa{b`p!~S(??ZJW{H^FrEV8vr`^WsE%Y&=D8`<9 zZt|y2YB(OZO3%)ROpiV2dWMK=gK8Q>d#owGTD-lCnp99egwy62%jiaLYqIvGJaFg> z9qjANv_rPs^$i=_ad!Fb{QG$V4d*F4rxgv6uDye+uNBwbonH5!hlk@kgNNti>@S8g zzhdWaXMQB#eLZsmTm8Sc^oGW$Tj7i1YI_Aa^cUtOJmyO^Ux{$B3!?Qdpln@u^ zzvW@gQ!TlZFH=qzGaGWPi8^&ECFK<3&wN^4$DHKa?(F@Z^`U%_MWKN%MhwR4u@9ZO z&p%DoOB>r3gL&pKdj-2@;3%bb>s)`#69a@akKAM;yJ@G4FuHbam8#q@4PDFCx)~Mv z@3w~27wTNEzJHvyQ>5O!veD@Y0oP8L6x^0d&@e9$CBY%>+VA~xGg4O`&un^*>I-5z zr?+{2&o5o=?UNKg;*)C3w!dtm#eWubNb-R%P)58(XXYUB>DRvcqU*P#?4mEX$x@lx zTNirXpJ|nDpGQQkm|}Q;h5~YW+_8S#KWKC>b07;e6e^W2S4885qHL@E&tP z7Q!<~bsNN@XJK&nBRkyO-F*%MKBpmmczHipxfB;_~0`PVE$L9kxe1rbeGH9L!nGp12M)nWr z`1c5e;s=h<{7(?bf3@-b|0jz72b-a$kd*mJg{F@HXQD!7a)iE%ltcl_3*c{0fc_Vr zVIEUyCi`h$uBCVz>yuV}8XKM;nY zPm*=n_+=qm&(UeBAs9zC2PaahBwRYfzi)F3dB-D>a*yijX#$3m`b?mkg)j@fWl}^b z)k72p9+_PWv}(Ru7o%=h^q07qtjjnN*b1r_(eB#^dXrpB5xEK^)~~LW_{1e;CwbtI zT{qXuj_VZpjNtrgz<3>J6aPF%H92Ly_U@B)&p`kCZphKaU9s})VSEbi-;gITGp>1T za51Z%gEfVWb-h8?2)`Hy(-H!d&xnbhP^Ze$d2AE1&)_PAJG{VRgV-t|h$qzCzB$*! zk14$})v*AB{gXHIWR~CK+}_}m^ruHvYS8IC$xlmagwN4nQA z+pexrXe4`jZ@OSAW@%H+45lV0*7GvRG1fhJI zFkD$vIzw2{qFk|=nf@MC?0hh_ECV*y9BM z<(KbR)X{eYE~RbPwCAyvZ0kN)fK=Rzt_5gV-A@g7bbV);KP8jw)u67@xM^$1&rU=L z>}`a{UICitjO@@Wvd4B8Hz1=}fO5A14OoP5bVZrRZI4pi@xx)O89rZMe-FQRdXANl zLu!4%33R0E^RLOLtkE5MaezL9dtw5{y3`<_$hQgHIX7IE-ET~km}9)FvUalXS$7tW zFRD|(KrJBnwoZMQnC6_tZ@*RM3F*GHlwH#FViy4fbq3 znaZBYhSg8QCht&)DREd(5y7b75xz+ljJloxWAG0~-vkZJAln$4L5i^?gMRg)C?bhM z2!z~Bli_jDQb;~pfErdgf|%JVBpX|Q@G@p!jGa;kw1y>fP9(#mnbKYK3X&196h+4a*?zvz{8(=n4yc-u3u~j@4;8;ZEK@Uu)7L zhxsz0g!&!-g8LwpR?XiyRnV}e7% zxhH1_{Pm}J7Q!^Vt7ndhjdHl0^_R$D?{E8I+S1oOgESGfO*dR+XitA)BDt~<TOT^151RnyVs9FQri*bF0rcBx|~Z@W5B2 zoSQ3Q$LEj0F-AoDU%;d%2(tJ~`4BdU3f}PcFP`*Y-0bkr@On985`uqJ+uo_z%hwvw z63P)bKAQ@Ud(ZU9mRm3(k01d;hXH@-ZExnn**`G(vBDzB{&E=k*j#RfrI*XBILB{NuFZag3`C3~;9 z#t#9-JCk7Ln=KGUa7xNntknIuyb7XqjtmJ3C?ZYTn%6qo*OqWFmWktrCr9}!Jv@t-sin9lV zkrtxtB`Hj(Dp&ZEg@Qm(AZn}v* zOQG9T4I)j|L-&iB@4k3Jn$-z~#7A9>IO5B2Ti~4QAhXef@Ts=QCvgW)J^{O+*muo+Hv#&@*(>vu$_`@Uc1+t)s`IN9)8<`WBXs9-Br~ zzNGxHIME%^@T}S=af-$oXLPC($)BI;O!~WYrg7OL?ME0;+4sxmkywgf{Zlox@}2ir z3UYtVl?pFd1x#7_iM$*|rG9$Rh+};w@`WpP*^ZZp0I662Ts{e4F#V$B+WF222MAqV zh7RsM#B--Ct!=lEZ*Oe5-)Lo;EYb2k&>FFN)-xd*IbAi@ZYk}h0SC>83NI8o1kF@T zJ(&oX0^YGW4OPLlfz(5`Mb4_Ezh9Ram>8wGy=FE$DrwGHO5WK>n+x0>j$N#F>iHOj zp0*xxtEKUqNsOL(UU^Y+gY6&HRT6&LhGQebG@}Y4gk_hMA~c~yYJ+m+D6iQ|fJH*Y ztzItrs|d0iQmo0z3_71fw>t@vjafI$AWsDwKbFFLXjUM}OUoQH=jE6raQ#5!gJu8U zl`y?GqgPm31kk391L8Z--*~p+(0KNMS%vTmZT+l->6;3oJ*c*7Rxf}@Q~y&C#!8tacK3Rq~HJu z^PrEejnHwrVpWMgxvdxcDeD|M^J$)+LE``$Kss;YcHljnI)AzOqG%s6?hlGnbySDr zO@1$Pow9B!!`nGK`j+U8{kr~JVqr3h?$$lS@;XG6L*w`mDCNNB?dHZaE_h1l_0{x( z(K~$Ezf9=`#9voJCmULP`{2<03!Z<1;fq=|?bW<8cLLb;wxW6wYJUaUp>Xp_u2F%lIwl+AR(2DBJiwMZP=U zn1WZ-O8#g=f{PHQoWslS0}Y|WfRJD8WS&pzN?$0sikJC%M30$y^AzM2Q=SdDHdg~_ z_#!6SGZau{i4qCJ&N9F78^~7QPl_||DAi>lyr9+wEH(yu_T^&dnTowsYm=6?5-K}C z^TVX*N$|FLMH!E8BU{|FA^MVRB~VOEmW1Q==Dnb+Ir2^wM~AcMMv09$8NL{&tQfj& z%~44!ZBvXh8(RI--NnZ!MAy$qM!AZqI(;iYm*E~W1NjV@iu_hzo!Rp^YR zh64Jmi2LAWP@|COl#dK%f=&5jW+ii88p!aOlf1yIx8LZ|yu?x&ejlc)0qcqR+at>T z>;ylK`txvo(nLF{LX<~jAUL)InP5#@Pzd0&VylsR1Cj%Bp zu#4Dzq@PT>KcTK1ojO4FOoegbY?>2_h>@R!mUC1Vx5LY~k*@r~ss8 zw7ZcNAizBL@<~0q9FwfpS7HmnSD^lq;FU7ToC#i=$H66u<|{AD8#&fCdJ`k$-HH+# zw~iLVl88mhnxRjK(yGqyRsqh?ZN*oOT(gGC{M&e1mQwP1M>)p#^FEo#V$IJDoJi); zj2~D*D+d}Y6#(I^S*1myzVBvU5sgjSgsq2mDeQ&ixx%JuvmKs#zG_I5;t^ao;qh^p zz6JhycA9D^FJ_-4*pV4uU}4%y(vSIP>q{llx0wTaqko_hH^78+ryw zy3i4}PZP;HD*6NxfaoC!G&TurK$}mlp*%eA9}n;+?-+mB3n1|DxTSpkZ6_IVk2xi@ zZx`5ZzF!->A4t;L(PSvBaMq&&_}hy}raN^&XV$g4ShuZpd&P@jPcWIFws1c56FJQM zcfPZg=W^`NdPO?9D#G9%X{WRUQ1#UsIy58C>i1o#9Y2|?nWrcGb#q{?IFTI%vDG@OKz~#zD^l=!8d0W61#A{f=N zUq`epP=zcEMxr;z2_-!JkZ!A=X~9Z!?zs8@dl@B8enii_Y2ZL+wT=Zzs8HLrem2++ z!ikg>AjN`1alaiw+p*ndYS+>ph5R@BvQ+P*e0K5apnHZcWDbrjs;|HzdtrGOvbB^; zue1%d-&<*(YnFh|HV=8DAk3xbT_3nhjlWZH{Wtq`Wu&Ccr8Oat;!t5S&8DRqxQi}` z6G%)*m7we?r68-}^tvCSg$Flv+1F@j)?9d2t@D~4Qe#m~6^%1De z6HT_vB{o4cr+#p-DcuuhJns|idAiN8U*eoFNx@-$9!~QXenZzdObJgO>ui@eTzo7w zPaTt%ra7d!V*9{D-YQI@z<^aotfM17ik99v(>6AA=qigK8@?f(cj$Lref;)~m$e>$ z)Q)+}auCwXS}#v+&Zr~KHGvVt^wgd;#F6$1*#$&je)-;$DzNWMLg}QW*J6XE-u=gR zf}*5TQE>-%9YX*YEvKB-vAde*gkBQ3H1?r)+~zXU25(gjcQZl|$L# zF+$2dKS0$YyM@zTSg{`F?l^PTxa7>Wj{YmA9Cj0r9mI781U0HWzvt|6hZ(RSwoqdb zQ%EJw`PSInU0F_eOPmH!yfPb(UCF__3?}YNhAR1`VtXe=Xe;)X@ZNBTj9Q0juD_?O z9rVq4_h999D`SdNL*<|6ixX_Z(p77kDomWOMzB!!YNlk#cP14ub5w!Wl1hj{O$Eb2 zsZ*aJdT@L1=*Y%jP{Z39Yv9fKksIjE?OzGAf2bFr!^81DU!cV00luC%WOy-S{W3d; zKls;k0o*!qOB=0yfz=<=#8vEo4#hy3?f(X>ZEnBx0^EcrZu25oZS0)|l=AU>xbUeY zl)AF>`J_{g&g*@%su-4bZ}_~GWrEmqclx^ZoGy=k{|j!C2}Z!2LjVB$vz+?(3oY^= z?D^m1gZ)1Xt^a!pOgBvpFEfp;{MNc&*JxZ_mpW1dXssCuy8x$zA`bXIQ&-DLM z*v|j}>VK1(|8q|E@2Ts5750C~*-3ts<>WsH>Cf?h$fy7P#Xo}mw-;|`Z(?g;@qdGw zqm^as))`s8*nEyvD+Tu)u)#x^W02q~++*S`tF; z7j04F5d879h_*esfI5-@G6-f(DqfgNH93=C zpdY+t)Gki8P9_y1;<|i{ooNe{bk^)=7fa{0cKgVe!As2ho8+jDRJ$S{SIC~%3bG7T zLg`dOrx@7Ai$Wu%e|nZs{J23LiiYdqCQ6wuCvl2t&LGO!5)Dn`0Ypv~oC(VYlN>_0hucT2If=Ak<@6*XqCG;s zvj;Ep_TrA-2-8Avo4`qg50MUE>wpol&mh+g$fEG*H*5Ef1lx-W+gv|G1d51SM3Gqc z2@CYF(yn0=E6*yNMAU>Fk6;5UiKK2`(Y?N<8=ENMsz*MT(_qxOz}3iY-61u1YlxlO z@7g(HuUTI2i}0Qe#_gYbANcO`5U7$zA@dq<})1;;p20S80x zq1e&hg6qlmZ)v^J^zpoi4lIHgw-DNhvI1T9>L6&tFZV z>{-)N9g;512%~oYY{Ph07~!jd0092E4gUQNQeD9ZAoc&=gmh*WPCp@imL^6{j`TnI zd{+N@p*7_&;9zEAnn9O>n<+V>p#P08*=XE#QW^>Zp)IAsf~$^ z%Ju3qACD^C#n^znT9%)G3ow8lRxVq=O@GyemW!AbL~q^2^XWkG3JX>jG22$qk6AIz z*YnG!0GN^1aC2fRQVN{Fu8s-PBB;}+fSK`l;W%=xD;(KURAoc!6`rLrrOWERs zMa%Mj@3Gb-XQT@Ut#Wwo*^eyt#NRUsiW}(lim8g{BZ*`f3lzAKP-HTpwt+=64M>D& z5ziK2^y3)Z4G7y{=3E1Oq++V>`elP~F}7<1j^{k(l7||5;DU|dC4Tc8Q6!H=q9Ie+ z^+!wM( z%uAxy2MxGsBDY+_u0Nu+dJm}M4?Vj)25mZ)ZTe4PGo5^S1=(|<8bGA4!vp_hVyov~ z(Z$v$%R^~u!&FHJs?}{|FV{6h^cq?+9Z-d6C|F8*Arvl<(Ck@QUdpY8RkOvH2pcvm zd|a8Wy|zXy+&)EAroflMoyfoZRw{f#h5iek3rWsc$jS=hbYm#M#2%LC7spIQh8VWH z5OcCUO9RB?6#^4et&kF=8W~^JnC9@(n?EPChO&n!o?B)aRozD0YqlFOK=sqL2Ss?u zk2))q+&@fPCl+FJ`lmZBlKSX?@@Xi{b+{}8 z=INmAklbi*poK4k&8^6S6V6MlNmOF2>y!N;hPZpX^|&4L$a-p(f_H@O*$7Z4)cxYG_qwXrCG>JZ(n(p{-#|m} zu_`IOnjeK(SVEAf=fA)9C52MLUTsM?i}GlfR|4c@L^W3yq9^~Vh{_}Q{t7)@%frL6 zI%l8iV6qqJ#-HE`;=bt&3WbC_b#-eY?zKyB6qI|24f3HAj^D(hs8jgcPha4L$u|pI zHoU~;Z~JyNySgqQOsn2nLFPgA_N0UALvXOfEl^He&tyliaX zSjdrA=cw1m2ui3G6OHuRF9gfB)&N1=Xv8}aO~TmNh-#1x>j{Ce_Z0l|I1^h{pXWp+ zh>JkDc1wX>HEo zTLa>OMa@Tz8gDi*fzjgW_kQ^VlQVm(0W@O^A?x?j=g8Wy)k-fT5e_?C}X)q?~x zAW1#!m~~x~3zyR_q05(g?8Nb~9O;;X97I56V=)Y3$EXjy12mnL5~|p@ioGb~LyDDj zo_8~Q*Fd$eX~AD_YGQLD&4CtLE4MX91ZrN10l?kQ^eFbkXIjKj$4kAmnJX_*%mFSA zA-_zJ1>J?zgDvh-Gk3OtgZ(BFEY*7G)zqnwZcIn4CIt0RbbVe#ym9cX)7^j}sQQ8c zHq^L5!)|0AzMQ?Bl$5lwBNl=>;QK=Nfq_PIaEAcRi<&%})kIROwRW1+F)`$jKv_~` z&H%3|io%07cdH0FlBCc4&z@4@OyK6sMO)?=Tc-?YNP!^%O5wa4txzX$Pc*HDeXMXs z?PR)pLG_5hX^(>z!4+i}N_!J&yC}ovm%1Go7@U(e80Ul?nC7fLTwU$fD61kLjJ2 za1Ldijnmz{Cr`Je7V)w9v_`I`oa8{T@pbTdGpgt7^7Zuse#j*_k2~(x8J*^4VNbkaIlGJjgn!N_EzS3F1M3>A1x~ z67r^ThT8!C2j3fKEoP<80i}2fYQO)Ok0A;0C12=W;*PXs<%eE*TM1Kvb42e{v#eO` zCD%!^B)AIbCG(7M!D{?~qE&oB)Wzrx1{4MKolm->W|r1ukF-TbGQQbddX+yLD{AzD z;*WrIWF@@p`k}JSVxDFM`w8A&On3Di#))braK<;j@KmUQ;PgKJOEGO}g`F1m4as2X zDR%4gP#6sMHu$4aw4dM$D`D$8yAFV&I(KTQ1jNY$_6s-pfu12N*4+4L(7> z_v_ZDZFKk4)9B%?gYVhCB{{nYaZXz)=Tt?rZZmM*{Qds4-OjPkVEi_BD$sQ9 zo&PLSF+8Boh@PS+7gLiv4&{XFKN8uw>D;?JWpuO;LXOQ*B_nmOw_ezZ=AIK|!k@28 zchW3j9edSR79eTUhD2AKr(s*gtz$N_ou>CDI=h`C&fO#qa(H(TjbRkS%Z7!5aec5> zTV|bMc_dO|S&p`RPfFzAbYZCTu&)wLW9$bgBv=%v*?`89viW9&l&(Hl{J6MSy4c;n zX{O9K=Ggr9OmAIL-!X|BfAm981ZzQH+F&(_tVZ}8NBN1fQ`nG)!v(8Py$bF~@VzS# zH+^K;?yodZpx%dyU^J7)ic)wo@=$j_d7zVb#n74(HR4~@K&*GAo#l}@G*vKV6`FBd# z7fI1LynV)UE#K7NJ;K9<^_;~}s}ltf?VP(J6K^#XnJ{*lAcruRM-Y~A3t?&G&fA3+ znRZj}Ue5xZ_F^3RhC0S5JEp?ja-xCGfEyEY1{dSGb@b5PQb_8T-#$8P*e&a~t(@)X zh;?x;BvhD(n`dSAY`$Q+aKDgK&`LKq3uS4&WcvM3V34ZyoJSwI$k4u`vPxe8`PtGq zkl{JGE+sdxms;S5qY6oRYdEXRpWQO3c|Pc)dan+4DU?i@*oTCO$g2M`&GW8wk#JS2 zw#l;>1d*H+NXz~~^0+Oqut;_@_*M&8`(tQ}$lh4(X)Pm)1T`rukF*74TvHVfTcS}Q z12a${YtR*9ftEO3OR9Mdi;~%*RD+?H`ik=hy4)+2vu4sF9AJU!O6%EPQ2+!CXGs~o zO1BueE5^HjsA04YfTT!Z1G4Zv}dUL z7%`{ZP6r{MzZa@FsjnC$d+;6@ifO&L(*Z;sDIC$i4v&RIDnVuhn1}TFnYuly(2HL~ zQlMSN_cR2U03Y{@Us3233)>Y|7M3B0L9SLfq+}$vgjU^LbC9!{;kNZc4mcA`hd|#e z7jZ$MfEbGaL^V3BO0Jnez>p`VbR&DtNziU5-xHbkb znq{2S#2pw7XO?8VrawMdL=UFR7RbCCSOOBOsMwtYG_{7c6mNf600R-%i(xU75FT>} zl0tFGpM0t9QtJ49%! z#*17FhF`W&YxN95)si8hRl;5^ndZFx16D0(sxS#_vs>6&MF1!l%_z~|#1Ylx3M1PU zRvj%6zddX5TT%U9Y9hW;Y^_PyHY!R)Iwc9T{B~Zk&qYl*_05*h{V=|#iHmAaMkK{# zC_W4!#3j1FLrR0f_6*r>@s$w=tIXsKl8aMO^)M^=FL7_q<4*Zf0lG+wR_gj{bHI!L zw>+kP&;@a2K(9WA%VGqN^Wk6h%IXBYG35-_pNzxd6RmIf!x@FDq5yDVLq0Q;N%z)Q z2p2~lTY_M>V`oKx;5p(gWEQATkL-z{elH<%s*Aqcf!zb47tQ4NY;OHx$MaKh6*XX9 zDkb%fonU4+csJtKwLk9lOSwex=df~6Vl(CN=X+yh4pz-Ng$-JEVE2A^GI@LU4>*JG z>3d&$WWhFHKYnXu(OiB(;u>k|ewEEM7b@3(6AGIcUbxSqFR5}JZ8UghI@WpnHdAhw zm|{M2b9KE8KA7MV%Fp+*CjC z7;y^PMAuS@zcql-;YBom#|9FuFZow1)d0fz6ue`uWzyNtgmi(3(3EJ-Jd?HBOJ^76 zPLwFLs|bwe?^XLuB*Ij*t$?3eBraIUb4nTUoL$xWbPhk+cJ0s)?@?U%UivDnNM8e) zi66v0`J9UInUg_0nzhKz^_Oa-qyPJ_2#xj#lYp^>jMx(UKbxla%NM2){sGO5LghUs zu+MaVT%Ry*fCA7L?hb3TMdB$c=Q~px(0*{_a{HtN!k5jQL|yc<+8=oPt$dv$59>jD z)3*Mzwqhdc{IoiqsA0f4(I!Ju{gc_~P8tpN;*tdkNUlb4QGc6%Q0VXOWKoDR+o?eF zYB@wa)4yh$6_xHZa$s+j=A<|yF>J}B-q?!gQ+?L0p3kC4H?kFYTj7k^`moq3Oozxa z^{HQZ&9(s}@x-1);Jl|?ZhB{MkG3LUBE0Ixhy8-k!2FDXL;j8_nyRXL74Q5fFm-rd zcM9PNCw``$N6!S|Z^yB)6T4PX)qewTmskT`|AWGNf3!=2Zp^Cn!-CnY708-&&obj# zX?e4-3{o53sLP1ijYzGv3&q3eiB$)>DZXVAvQOM<`NhBq)PgG|vLX_3qN@V?Z+R0L z>%8(Y26EO)CNZ~zd=jV>$zE(yY)1@xT@L!v$$Q`+s(0J(nM8GmDtm7G;;0AUv+ zZg27nQqn9y*47W~p53BxX}`se_+Td$sGd0N0gc^P`hMXYqC?%;QF$((CsYUK|9L0s@B~FZ3)(SHIFe}Ma@i;Qg^p*VR#S7MJx;u$Smv`9+;kDGWym7q)yBgPD8b+gyu zjkzkexInV$+DH?SEBNh-Nvr6nnp|DGp*xdSmm;EHmrVcukhlV_&&pAIu?N1ea}=c9 zt|s}n^`MpiSY+kiN(F5_mR6*%o5IC_pxt|&M;HYoAzM&aB4%X+cox9>G;I|MKbUS! zAw|8#_!sV)lObv2-snSN!@V zPw_jbVYHxT)C_QHb-_decavJlCgLwfBWo(Arw)f;wc$#SadKI)o(weF4}n^0nm)jO zH%7!FhE;-{EdY|ZnTxVYgAEGt$KL_*i&@%!OHcqv$eWn#?P&lg99 z^W?!LB-5TJIg7o-V8yEQB5z}6YUTbbb)ij#jaiJ(K%t2=zYc| zt_ZusKlaOqW9Y4eAHavjGp8I;K8x2&f`6;~t7R;-s`0Yl_G+)^Wku^d8Fd#jg`A@Z zw}(9cM$$PL4v}k~AeOE{(rv`O{o(DP^mT#=iN+~x$Ud*+>$1*j7ou0n`JtYeGgQ}w zuk6fiQZI03ZKthR0?}KB8u~7BsK!L>tP?ODo~vRO5!9R9Q!_hxJa_oCt?@<{t$2OaKNIA=xi} zc?`+YIEEFL(JDO7XlYP}TUa?bl&%BGJ`6N^BgiNL*$zie&>1!fz%PlmQm_gUR7Cbr zfYo^JH<|FMuJ_6rMiNtnF^SWqc6rdNteH*>@Hbwik4+Mfof2ik@k&V^d25}vbINNa z$73e>^G}yZomzHwLcq!VcFISJpKYQ%X_#FBcf|PTV#uJAuA2B^nSV;LwV(U1d1D7KsTaMaB&2J=gSY!-#}S$7CO` z$ep8zDEKC@l`cpUl(MiJo90@v)Y~0usF2uY%5yr4aS#k*khdn(cX~ME*ZlQc}1?N!S`@0Rxon2W| zES;Tyhfq4=3@UJlxDH}M6h+dJV6q)#s=*de@xsn$or7V$0?MQig^L~>&Pn)RlE9w? zMZI{Ga6Kt|3lx7N(H~x)K5D(92E(QP3X3?7*>q%-vohN{8NT!9f{2^`N>Z8%wM#w2 z^)TW(Ms^rR8anSjf1AFI{|Gj{#ysL<#0JbvWy|zTO%UQvJU$jM>fJOCfG6)~1fDB> z1pS-+?>w>MrY(qljDI_lCRO#(%4&V8ze1@Jo0^zrsxzoKUA2Vb=t(*(L=-XWovt}2 zUWgFN=;#y0scgD9K!A8s7EIh3EghkwTr|12S9wp=M7f6d8(z0|t>(a2tadwYgS#LR z@|m?aXN8IJ)UIa|AkU?jMEge|U)4Sy_UCvWa2}Yqd!N|B*!A}sto834)RJ6^p~s#z zVzQRYS193)i;y_CCXB}|!ju2kW;J_Eo_XXopxJa>4w#BItJ11;OSwjE# zj--1ySJ<9m!q^LVjx^Wvk1Gkee2>8GKYZr}3oh06qXTgP9^_Y+OmAW1Kd%Iq?4LN_ z(|kl1pfTv~+16|Bm3G+~+VT1-h4K2+BVeZJ=4uWlXeVdX#pdwJ_Bd(OF#(6U*gJXf?%6p}Q*gjUsDN7)K@e<>pHWO+?UAE_ zS`j|*ji4QA+JPw(=80&0l9l%Heq+dTgOo{{d00gp_84ch5m)tMPbn9P1Y{F){0oj*2DyDu_xbCPc%muiG3B&|*bI@0v*Et3l% zP0RxyYzM_UFU&4f>S**u>Jx>Y<1P?2YM=9u5Bv+7@i-#tj+m$KAN8^ zzG=e$dMFwR3R)>L4Uq}XatstMebo-59S?8pdR&s0)B*uE<0wmu!2XYT?Uc^)PXDxX z3FJ||TtMe(@$QnrdDDlV3K`ZF!`0Sj%$C5X5{mwAx5)CpNEX^ZO z0P0^PdZ~#25gp59H^c}l>Xs8IkY;8$8^2DE%H89egI_R(gMY`@ z%cdHyRF8$wT}=?TmMhzB_GtwQEvlTl!tGQ)mR4 zf9w7%{{TVRsS|K3cV;7OQe&v1>h3LWckUZBQz-ub#J*toL91xYnfETJuh@hmrB2XY z*gs-(`^_$FzWZ3(fCDsBNv@KR|M!$L*$opSih9X0=~n6jHLY_-78cJ^fn#>7MK`Py zCE62MN8u9AtSxzcET#tx#RNKTXh&rmud)(@UFr^7EVzpY8#{?zFT10Ra~0xKil&tw zU!w+%WDB$UV$0PB{GH|Y6xF?H88}3a^^8zaEUhSBsJNS|@JllKkXn4qf2XD|&nt5dg^&D-yf{PN_yGJ z3EEje)v6MTd&h6sEJ=ok^-1pTo6q&stFSf@a2NGt0@8m}#!1V8{&!)#W*_S2Q^d(X zGNBY%%isHtttb2!55Mtv1GBVLl%g_&uxLG>0!(R@s22-Box0pwQG>E2h^(fwquMPT zvic{F^SOqjXnJYdyLsif|Is}^ zq7|!lH8W!(;BbrFH=i+4{MMld!IW%_!4yv}x4J+z_~`4<6!ZOOfG6-}X4qMpU%ist zcXD4qCzI;r|H~+^7pZp}j`jNU-;``}0j*G965u0iSQ!C8p({L0M_`j0Z^Sa~J#w_# z?VFw!Ob59qWm#Yd7f(bZz`!a*M_|)Hx%NP#&{GrkAo_Tuv&5@K%eid8w zrrwiRvAhIRNy^Vb)evlmlqGO5mD)`AN0qOdR+lD6t^_0oiFUZBh7RY5mjG7D?KCfA!Rq^46{0)(!G?tZtm}mt5Nl93*+#> zz_0G7x8%iTT<`!E&%ivBhpC*@2ESj;?y*JHEjttySw$-;BS~LHZ$U*XGT~491-Yf{ zn$VJJoRlMDvn;kR*pt9=8u~d}$yozVw_cOF6Y`3t%)|+umA+rv6yKF{2t@^@KLmh@R&v=5P=L03w%5p0q@_^Jc!&ay-UUiR$j0T3#UC2%s+LqS`#jEj za=;)W?0dogWrY8`3j3N2P7AHKk}p4CK;kR>>WuC9nPwJ{imA$(S;<<-De2j|uu2`t z2SQ0&ap%u1LSHNVeK=$4bPS6?fOa+l-p&UIcc2;VOFJC#)StSt@)mzGt^}$HIdm8! zOaWsW5Fq?N^*LD6$iN%K(kf#!c5-gYjYe<3JZC%;zI$9Q)_Tb%4KUS!DHXSs3g^BR z!(J_Q7wglNk9*3sIS=d0*2>6He#*)dW`;EJHkTIW>m%&E80}AT`%RQb$0qu#i1&)= zs)zAgExy{XF{dUi{>*Bo6Rm(R_vcaJb}dus&r0e_Eq%i*zm@8y4%@+_gVLEO^6q9= z!Q!W91%3D?4;b9i#-+NisZxu_s1mv@7>~~YgRU)fmzVvt$H%|9YtD(Jezcd~jz)Hm zh3T`lirML5*gT1IORK(iUG}%rp|`vxs5NBek|?XGZ)p`J2T4sTB)F+0Ap<|A?E8Gwiz?Jf5H;{ ze64P(?@n(7r$B_(){~WNS)r56M~;EPwq|CkbnGE`_)zOntu>EXaFj~X5iY%$8L}C_ z#>V#v88EKBOMoJr=Z~uJ~{DG-45~z zZ1*Vt2b>APjeqm-o};OlWN1Y`S;?fiwYkz+qH~(HN?RGWs4#46Ys zw6S|nx`1K9tl~|-Dj3$UD*TvW?cByi++kUTrZy`$Q>2bjX=U8+{>NUF!GF{izIo*V z=dsT3XUW?`o7b<6kEw1T3gvNhcTmK4aPRig-kiD!Flm1tjyb*lL5DULHeyB~q8OouUJI$SaZ2Ea8B z4F^#k#IE0?OCcuEN2PRjvIGrYepYV!y#-~mtI<8$Y)wFuX5+S5QwCGXRIzEsG2-cmv5m`X#HHGU#cNq1 z7&1kr2~UCTL5wg696+AA$jiFGr3!mqaAyGIl`MfHux*F?YMBwFvGH;f(jhoo_!Eka z_Y|%*_zk2npPn!<=G~$>VR%Q_RRBZOUV;ANa0fF;6R#DK1hx|ij!PChTM(WH)!7pS zHctJOhbBvg4F!<`KHy(Gm{Z&MEMaR8RGJJISlaoQ+`=V?*NG$6f6cZkvsGq=d?FT& zRr7wIeN=FP5aOeNSzZ7hfBIm3rP)%uMU%;e$7z+{$vwhr{PYp7kPsLdEhM%GDUIBd z0<*F%ACEHKpN0gEeLt}$I>937V@6+EMM5;tvPZE}k9a}ZcjXz91o*pb{ zK+m;(_|qmzzwyURulcvE@F8j2dR;Qj+`?#A*f6jWB+QKe_e-{UVwH_BtNFKN-zvn_5iKJ?FychIo{#1OJG(jqDt5--g zYPTMpEesMkRYPYt*!UYV_E*NL^rc4q#?GxRpwC7GY$|DILi~@f6D>?*p(j_zfqcK* zv^+P(^E22E(_&M3tT=*d^_F*1-V=v>G}`^8EIs11!5HPMt*Omz5qc--@I9@6k=ON* znH#HJN)^h1u}_e~Ay*;NC|Nyn2EFVLaIkpKBrC#E_9Foq0*e@s#AHr5_~S#Qz5N{v zBhHo~-SmdSosWF>PM}V}U@On_I=*%pM0DJDdEA`1*m4ppD2C8N9x&?N&sxhTcU!#o z#{9S^1;gNdgxSrLnexh#b%oNL>EdM?*6euuxYkrRR^aWK1d7y!%>88u3L2%~Z*QxG z)EyQ7x+;x!{EvCF33}pk17WSS@_P%%MrJYI%6@NmhF0ACV@@1wkpFxnJsz;ho%;)? zfymzSj-AzmSNJ3HDL=I4!BS^y{kaDReCyhP)ZXt=^_m3ihp{KMMUj11_OB_P z#@(y)6%SoS8tgm?Ih=yEvbc;%CJ>4^(OB%`4i`a_Co(rqU$@tT<=f#|KW+0A`*q&l z!j|DdYtC6RhDAm&^cwZ4vNX1cCaa;)R>xelX8Kv(pA62^AlI+1W-y{%yEwRClrOZe zuCBqK zSb)rTv&;$<%H@T^!@rpNa7vV36A~tY7!smq6O_xT!N`z2FZ8^0qmv)elP?FeVhN|& z9I1m|&!^k-^=0Vk-*Xs|NR!pacpKlwm}5eX9XeJfnrjFwd>_TxS`cOrEpr{c-B0`P z%A*dmu&u#XIof+vago7DM{%NRNupY|^gB}i1k%sBbegmcsrpXjxmOj~vr6B3rxG5n zXR5uh+Qu7zF#*+Oth;OdYL0mjdrU)6M<&n%v`R0}RpPQin0}2sg68Q8I%a~0Mq{gN z157y>xyI~2vvq=)EF+vtbC{KWDb3_~1gR4v(VHhGcmBuJY_mVr0_$R}{4^;y_M9Ho zN7km1X=AZ@08c3^Wz=?{h^OA8E;B=nGU-j@f;1^3y(r#1g`!sy2Q7p0!OPG~IUSRw z8TKsT9pdP#fLH_~R6WR!?r(q2^u#Y6ybg~vU>(}@#50m_t56H>ouE0=w5``Mh467ecNr9u zR)GXq{X@R8#}-<8tir3jVoB zwWa4cRbF}y`K5$uMKz#{LNY5&-hSIBvOn@4sJz)vvN>4lE_u~bq(Wjuo8e;!UY1e1 z=L>@!0sMW6BAuK+sX>TElyqMd3#pPA{7a9+DIbyxob84jzSpS?pifonH@VpcofuGY zju0n!YCkxekzLa;Nkl#x!Q@as?akxx8ePPnN1szu7gcWdyFKROU8AOVZ#L}|zV@LBW?Z*l@z`h>j4y=NfQMTgR?t zReG~t#`*$E4-@r@w_%-GCUQ|h6y=0j1=Bmn&c*1a+DF`@vUU|g;&tQzSLPUB@LU+W zq@5dZ3|F<71mMog`vZu6hH4p?Z`GUr#7YORBS!uE%4{v9jX}}atEYIP8`*@UF-WpD z#R!^XBP;7i%VV)3Qj`9)r5Szc#3FU|I-`++18fd{G>`D+v)c|jhZacV#13MeM3*o< zp+JD0x%OCocO@hQv&aX1oM z*b11SRfhM?GY+lQm8y&1N?_qA! z{FnG-8~we{VSipN)MMf!=9fnECnVuzzR) zEW?wO;M6GbSpc+TgdbmojIoNrhv2)$OQFP@sx`r!#2E=Ui{(HFm;~iL@ra4SRf_gO zK)HqDr+<$`0MG-Ufl|*Gx&ebQ?-c{y2JYYE0?Ff0w{SEDO~w@Q#xI1CCzfH>=N&*E znJt+PF$cB1d0Uu}S3H#FaSvzDUbE&-Y+s1$zDT-KoA9w)$>5&UQ6OngU`3?EN~;rM zX42i9;$SZ993;aT-qW{Qsn+(cB!R1I*fpwSgbKRY4t4~CL%6RAX*IBe|M0D$_QH+# zQ4%M~CKQP8H8v!Gye*w`dmlM*QArMwnKt70zChn!{J`f5n;RVonLXHu?v8Z-au60> z)cmMatk4DdlGI(IyFLGc6t5hU2^9}hXS`F`AYT37O47mR@~em^mjmr3!bhChlw~=@ z^H|r;9~0XPJjt#MPebp=v%dl4wCwNT)|BRuB& zA21Z%251o*Q9Av8(H>$VN%hED;lLOa-J{xNF;6z|mM%FBgxeZndIV;cwL5s$Xth!d z2yV`3R=Xh3nu$x;win??4%TaCoqOZRClQ zNEluyq2>fj*aafPCmG!~QD)WLC-#Ko%iSX|rH)F^*G3}a`{jQ3Y@X{-GZHF$62A-1 zX~-EFhx!K5cx7BmpSV*v*I+8^@^oHd(vrYMa?l3a?TMAK$>Yt29rDS3fMJ_vAEro^pj2PKnEauc(xOjCCqXus@Xt($loM zZTRS|44E(Jg;$=>)3>>K{p4o6R#9rr1|9qF+-5epX@sQJ^?4REHLavh@fkI z)U!AE**Qz=?Pc7{DxOw{{r*FzlmB4O)FB+0h~}ID8!|g-V9`Gng~04m4pr(4E7tEy zF^gC0BiBJ8*(+?k7msA{X*|^7KrlPPIb%?Y9ga}(&p*cAHH<7HJ5_aJap?MqH7ZIr z47X^t(qu@M{kuOFNJNa+k~=n-Vn&oc&NVrt+A))BaT2(NlU2V|3I$JeRv@T9q$NMb z>*8JK<~J4D)ohzZxTR}w{j$(meem*DPV<pS1$1ZYVW;8ik3VLQ@9jIc3 zNlH?+-3f}I6fn{+^ji2EJ>t??EVv%6PSMJq_q`b&W2ZQn!TOCPXLy)Cp+zu%YXhhJ z15cmC6T}Rgca6W!megIL1!oM(RNHYEc>J=L+q-Z&E|LWIdScp5*1Ps3Mr84ILL~C) z0Xd2z4df8T$~nj2ZT~2o!r~h*tluLl0rxtH!C5e-If8KA`*Zm((npTh*-oxc&Iu;X zH%*`BU(L1l*7I)_U+R-6EA0DTZ@dS#DRQ6wzXwxmyNChE`aqY;baE!aeJqeA$~50V zzJYgf7e{BJy=J$K6KP~(J*VlTCo5!&3hV{gFm25gJ;{kLDt>3OL3M&TxqhFA{72>e zOwH*i#$siHJAyuSQ4Q&wF9j{1eUBIa%rxM`HOqR%@m=jwd?6^4$)T;wBp<8NJ7S>B zls2yXlR7DD+~D)mAcq^Ov*3=jwcEQsa5dWUkg}J#L>AR9T0!pm>2~@2|A>^BVvUVX zfIpFn+u0=GIa=bf0N=%t_Ck#f_Eeu^_pw5!9JQr{9mw`YQ(tnj9u%6y+v%yq?7EZR z`A8o2S~$6bDvFBx47#@lMpBF(3u16`-@&aOPl6#Fr8T{W9CP)R?A+dW*}REG>OuGY zKISo!W7jp-S5WBNUE+OT0}Gcp^^Jsafc0u_54oj(ms8>Qe!)YG4v6u*6T4R9`gz7>AE1h#mV65PtdDzpIR!uM>OP70dlC>3@#j|$*db^>5%_t{ zPxqp}trh)(cd3N>9e`ALxMu?~0cL+IeV=^lYcPP^lIk$Ln`kj9ftysZ*@2Uyadna| zaI=$;@lca8@g$m#@H~KnLJ7*O4E;4zqO5}Fs0_}PKMOP3{o||7#S^>ScIG`S1ZHdJ+>ma z-;U)&6MWwLr+sET!R-DMyS^X(e1YC9dL+usKd1~;nv!f|FZ-I)<1H zd4~2-IK$C-S+KgC!|}<3NIQX0j?mA^LHOUTqO{yFBdowHR|o^Z$yW}XXxPSZ3T%6r zenAE&S@KqX1Zll{Bq{gl6cOmw=fQO$>Osj{lMzxAN+XCys?rEV^epoIfao*$SgCc~ zd?>*QIm?bG>U3rE8#LCF;( zkU}Q>v{^e#`A0o#bLt4JU9F;?9gLTw7rd$!m|#rL?6T_}%xGIyu;&cO9jHa!@PBV} zeb4GUN-`}^8vUVDLuFB@7TmJk2%WJ|f=$OLZUC92^KnVG* z7T?m?T>#hSdCt?8+kMhctv1kbo`}Qb%9v&611~O>Ths8N336{chGV=YZOU$^ zQ|0&gRt$tW_ijWcc4L)-IKP)Xc?ZTOR{Faj5d%!}#~My)4m%5P`X||1l0o*gb1mMJ z&8Y*`>OgF9`QgYA5mW~eZ_>!J`jp6A=X?K4Ey7%S#Ah+Un|NK8)<1=D6rU%*3AjLQ z9t_>tZ9g$LABBYEF|%IZfmptV16KlV zf#Okd^7uY|+O$<2iS!tbXA~EHwWnP@Zjg+Lcwgmim!!0rXGXb)sNNUv)6CkPF<` zs6xdXdTpAa`ACa|(8kKXi&<9K9yKq<3m&$y_0{0RiIQ2LiYP2u6wv!l|0@($Gz>kH zC1Sm6=CCCb#&DBNJENLF)XzTDPRyw`HIj^nY}o$V!TE|+Ab3J=zd<@SSwFKYlGGk_ z++sf|%|E`7AH-8pdHQ~WmwS%Xw37O;ew!rEX+G3b{ZU*-S7FOdGu#9T6*A#t1<-aYVWtxSId9&PEIIGOrkr%f&T{g7`iCdd_ikBy2@1(U8=KXwnY)L-vuqTF!TdTwXIm%!kaiH|s7ezN5ebqoQE>FNix83VGpLFa@wOC`vM}fp z(XlWvK}e_-qgw-`QQ63upBUxf4j*y8!>;Yj03A3gp3&jX|drK%9X=#}?s$1$gNt*fQM%ZNtdVd{@*?mql zwco5+uTqK`D;A{XW}sIo>Lh9;W>h1XTplSDX%2;qzzc)alR*ne!Y&OsIUJ=pq$8-8 zXf|~uZY3fCwC1FWYk~e}P2y>qHB{5`JGz2QGD%uJq-hHhz3EM)`wbcp4Ge8v@;~i7 z3zMS5|Mc#wbS(4Ib27%9bhQ;jBu*m-0;6*+=4XCrXby++Ou^JMbFxyj0aV$px+KMJ zTxa)Z;+*8*GFdt%!zRC|mGfn+98@ioG)!DLc2|`IRV*)|VZoHA90c5lOPyyU&F5hX zD?sla7xX0+T#P{=w7DZtAEg<)%nS#dmBaNvTiu1RCrA{;mC#8_SB+H1NWs@~ZODw+K%^y95w3Gcq@&T1N&XQ%6W2=FYv z=zs?Bpp{$jfd91en*1B;Axv%DSr8DA5fLHu|4^$Sz%C;|&(jP$C_Z{EfJ<*Gu60k_ zeZ==rN3ZPjWp@2Jy}#Ll(yz9=vB#rzhQ`CyZwpeCt&7wZ&~Z<3TAwZMo?X3VsDTUh`A01Hq6_yYle0r3F;k2VZY;C?~?G_V07 z00=mq5b%7ffQShQ0;0t94*&wi6bJ-b?t2te0KkIfD*(X35ETGWU<$wV(1Ab(n1B!< zMofQyAPG$W4-RBVAb%tPDo{Tn02qi63Gi?4j*t`}2CP5|5C(=M^%nsy_j?5u0y01a zhypR9`U?O_p!$Ezi4hS3fWZkB0Kg!Gkbpo?LP`LzKjHrZ3qbV0AP@u$f2owQp9G z5`siRP)G=Xf`CvEFbV=iL0~8d0tG>$ASe_B0N@}14hG;*01gA-2mp=*;3xnNfZ!ky z91Ma(L2wudjsU@tAUFyH2f%O;7!C%*pn z;b0^jiiE?Ea0C*LM8Z)>IDmqKP;f8`4n@IXC^!NIN21^;6dVB1AOH;p&`0;2BV>1Gz^SJfYC@W8U;oJP&5dN z21C(MC>jPuBcNy`6pezS0T>ztLxW*xC=3mQp%E}N5{5>>&;SArLZHD2G!%h`AgN)+jNme)z>9>HT!Itey8ulUM0s+mDzASt7Pa9 zi>l3PjpuX9Qk_P7f#{dmV#Ui9-j5#9bJV?A?Nr5QV=AyeE$In+$sRw1bTA!^Wzd-k z|L(Lknl9-^e@Ml<+n0r(rI!|8EiYhtH4yo}k^gYP1jv!`bU(9QZstz-%b9w8vXUCb zE-*rS?alr< zQ|@^GV`wY-FADYGHfY6`roN(u+o1jeTEiE-;8%~&V&p@HX_4to{`K2SnG2UM89^_+ z!12}#(Kj91*>$;VOb2PQq52$oeXqh@ZPwC*Uv34jN774d8#{btnx%zF5m4t?j*^FL zz+~9ubYtLgER5Dn&Z{eNq&2xIF@2ohH)8Z1lR&XXzwoz`5_b!jqM)U$Gj8VX{b_FI zA^8d6dz)J6>Dy^;*(o+S1T4Rw*(Mrg(lCYYWTBqy?8Gv^No0&C#%wFdOk{K4$xG6g zAOGzeNMIGbk|da$#&sn3n%T2-?^O|hB${onxY8_WE75ezojs<(wSGULh0P59+_ufx zHx{BQ%xbH%c2|_x^ugM!d_?tmNpH8ozEX#i{ed9{&$UPNRf7q~CY0Z$I z>`G9Ue=+hrwt_2ZwSyat@|&qBmV+O(#m}=Jx1TcUa(B=+EF5(n)yyn$+g`Behp^P# z!NW>N65$!7K?mhEZ5q~R)yxUuX9H}tM2BDTk`c$;B^Ud?)OpFV2bGVkch3;4tsJ~# z9FLc*$4o+tFZzT#@bjr#r4C0wdNWJ`LrGO%VhqovyV{Fk*2OF zKhk+Op{(Wa?y+6GQyu5e54AgcF(j>j;kCyf?=!?RfC|mJBO@X=J%yDSeUp|0p%eG+ zu$7A}S#gX)8?VmL;fHTHSoj*Fe&nf!oj72L*73)F*7z9&)Z2AW8&0h62Q!G4dRuSi zU_w;>Qx!voxx{N_-az5?nt%M=A( zElUTNMwAVKJDIQYgb2f0o}WI^D(+;TbdShM#E$XTk-8TEFtiaV)Nia@+1IP`66-1FtDwoenm|!2i}s2N}O6}!0zuCbN>aC-^*s1BlBTP z@*%V?RK@R_41mP@5SBu7zZtTolT1|*ajg0FPz2RQT{_AT)8@X(os{9DsY(pHo8bfu zWyar5e$SF*dY)e|<4=5^e2U@2{k*rs_EjX)qRBX!CW~_SC3fVFvmUd!R@bMQleEHT zeLN|*P2VNXzwi5RlU~oEm0dMRQUFFp=fuLc%P()acuyIR3E=roI2W15(Pn&6a!LWL zlj#@5#Dd4)8F%Djvz|Td;7!U+R z^^0i3bfxQM@k2XYJHv5;CSbAzVnS<>qpch@T@A4dh=W)UXlfnCX2{aP4I_GDdxNL* z6R?U@90-uD6fWgaoFm$^^HWQ25xP#*#bz$o!~Iv&!n)Xn7Lx3wd9?i5V$uhnt~T(Gwd*8s4?~PxixvS7d3*MvT8-hMXcG14n8ht z(e;H~jjIIpX{eb{ieBmERqYD#6YFtd{q>}p`i~G>o__g1vkjCb)5OPL%&Zq`;fPTD zLn?ivY6GVp%B)F7TW4Gmy$v{J@(`rN37gh1?cGiL;BA>lwd2x}nOh@;E1Lm`cI?dk zrkUzaT4JQ=pXAo$KX9@6p;^54N$IAWKI@qKP-l5)tEP0V^X=`(xgN>jbv>z!73^m* z^`ne03bl-TM60H~e-gI)9#ae2)tYJlw4{^#= z`P7LF<;g^&WtujST&tMMf}=?_JTvSmjf;jaC6oHVe2yWXQ#ZyUNXCL)D>EpYw?EH- zFrp%Pm0pNi%02R#NT}h>dz&Wrp8u}f>ThKM254=9sdlnJxf*Wwbni`+uu;a*Yrwuf zfE6lEpws74x<0i3LGCA0Lr-;yLz=bGvM{pxgKMeWlOe8q{_zPLM+BOKc7x`sFN6*; zN=9tMr1oPzI;7fTS9_Cfn`98xe84X-ZfxGD!%?DqUUaQyRfg#^5U(8IoL!UOs3wxJ6&3Yb5kC^ z&k$HnR(2!sE?xnTzE4XMx@dWyx@5}zvRgF$vVD7Rtv|P+2jA?b$FFJhXFh8`G0jzp z(e1tlmdBC+H+kJY>Rjyek6|8Vm(i3D5z!1U#-1Eqi#%^`DM_&0lXJhEr~G@Gs3o%G zjF#-U*Wa%Vx;q%(z8rA6I_ojmK?X&QO!kT%@tQaKF16@y`{^G&Gx~dpMDn`Q>CaMRko#Q)JzdX?@yIuUq{I(;~&2v)QK-@RH8!x)A%ptGW#O`B>;jKK$jSB)$7j`$J~4sK+}oOg-vPN=DHdGeYhv!=&3&x z{E{lc4s;`8^UV*f;h!R_DVGq$!>?(d8{~;p_*xLCVC<_54i+Hub)eEvHxK^wC71;* zNbSoaL>~BEn@6)Q*7RP$FBis-|+OxZ#VT^3#4D_-4nq{t+d{N!X!8Xm+A&9bQltIA~%g zL@qP@wVGDR7h`fkFE6ltxkZpxSMb_RFt2}bXl?KYRX{{v*uHVZ+|%%$Z>CigQFr+v zD}ByGf)PI+h1Lf<4HrbAcfKCgaTV-@Qhw7}EO09mgn3PeHP%JSxP*oDaXc&h{8%V* zFYzOh94s+WuZty|O_relTNG(r)YOi)9nzJ&UK_*mmvL3}Z{LWef(R@Z7@S3V8I2cn z`ZHR42X~dMvVpgsNuuh{9>yn zG}K}vi1O`uEPpn2$5}f@ONPXoyo%mtiIwh;v8o4|QM1W^S7E%4Mh2J0qZ4t(3yRM5fzhC?;U@g7U{tO1r?H-xz=K z?WXkLr{ZZ+{wYWi{got>6#t?(zA@*me9bQwiUj0t>QMb_B$e})hRUr=U_ZJR-_tUw z(?6wB=q0O==S(5v4s}XWW!eRXvq8Pi)unrjmVHYh?6GGPXI?i zxWA4(%~u-XZ1bx;TkQzw-B+~O6!!n^6>!pevO_ZZW{ zVmXtd88*Xr^SL;CsYZ8jq(n&uc3~ni`zhRLvmV~zSDdkDCEZ{rJ7Bk$T(~1lRcM)f z>F$3$kX%#mzj7zBq+PVAO&eTShLs|Vb}#9eE%RV3Wu5i5gsWC$l`}dQ)89Sm8c0u< zuh>u4c*_1gbkd-SOt}-kJn$iW&2&b7^>NJz`WlW`(juB`gL^s9VxdsYzKx^}a|$r@C3N zTG2F*8}HW-%P9SY?-I`u*3K0&$ra{}6}dN6UFP;E=`8&IP?dRi>wz4d{gQX=0YL;+ z3-h%c>~)fm8l9h2Ts!5;ay9n*VlwQA!%%JMg&G>`I*nDc)kjHPEF~^smGSS%ao<<% z2US}aRcHNZ@P95J(dcP7SY%j~mtkFFv0sxxl|L`+8TGzVimDOHSnIBxb<^$URuuHT zv7xH6me)FaPP_Q+`?`_m4NuIP9wQsJv+7>-*JY$MA_kvCH|oX|HO8Xfi#P3)CGIPw zHrC_dH?^oYy(DNJ9jwL@Z#W8#e3+uTD*AA(V>1%n+TY%a3VZjru@#ded17BIctLOG zK3hkl&Bi^K)FF%rqy4ojb*^+f32i$CM?3it^>b7kHL9J~qx~6e2MtFDU3fc)sDpl} zJyaYa68!vqqPJsWD_(TR;cUB=c_;e;L{OwdTwM3u{-E-rR{ZY>UGoYwOZ`;DDL$U>2qkJlFPGBDaQ^$r}j1G^$BQx)d2wK z#h$qF-hhMNZ>V11rhX^IzOXdis4)u5p^+C(}n}&u`t#0Or#=@;INC8}XEL{8H znegFR-Qk7B;km`(>G1A9{cPCG&QB~x9tk`%VWEUpj2tNv91@N82tCvB8Cf(L-S8N_ zQ%pTR@H{;jy$zw{S#7*g9J@;$MWv>qMeGR=M>dM9{;=j^=u!R>PBX*{j4z;U$e~2# zDmW?N{F4FEBHyozix~AXZe+r3+VD$!#&Cw+w-*J!T8_R!>y6xtWH!~~mxvKG*H1QU z5r`-ehz(PSEluXM;pDUtq>=(SCjjmVfMXAQ7M)h*oL14BR`Z-zKkRsIK|gdqt;L>y zPFQE;*_g{dbAiAX14f`rGqy{^%<#7>lgv(@%p!q#VS`Yh9Q#`(^X^q5h3D+oA1ZjH zvy^nm&7j#?_QI{J*^scHAUbmhmSqFm^Bamnd{j9v#*`1|+~n;zd0IkR8Z#=+mbYRu z?J)mcA|(KfXdcdg_8hKwm|q#1rru7pIBX8KnlVmuc*{{A8nnQ754Ee$qG&1ttPpM zo4)SmDDLN^v{8(w8DG&YUk&Fo{#agV7Id1%c^S6khC6}pvxxI&riM63UvI|HbEdOo z!Q^m-t$yX{$jYSK3RTN6E>U{uQW2~Mn?$=gX0J#kjxCmM&0H865BV8)7@L&lXfHko zvYJ~9q35>s_<_5|X1o46-6M|}NL?z5cU!EmSy3)oZ7liKv9!|su(BVw@{E4Xwr7ho@)!QmGbpyY#7Xd7N!aHDLKIWc(do5<9`15KbjZ^fgy? zevNb~w#^8^?Vs&NTfxIV6O!D!<_zzn<$N^pvcIjpj>?L**NVzWI@zBMMdii%hYdOX z%_HI6_@fOx*F};sfJ9{jrrgZRwSjZF$}+P4zLfb<^WNk6RfWP4y>%oW$NpDl<=8_c z{J~B5zQGAHT0+0wY0H&DywC2whPQ)jC(lRdT{s9KkDLI@=HJ35pDWTnSSCm zdE}XO5NT|JbMlQoUODV|O72;6(lWms&h6}u?a~zC31Qh7u%5m`qizGRl z=Q>OHyKC!-MNHbgro7kNyqe;&ev1CK)=fYE1m4CQwO65k@R)A4=1+ND)9R&$~hX^9$hJ4lcro3>TMa%{>JB4bJWimJlqN> zbH3O;c2zhA2N~Uu)CVYEJ6tXYk|Y_o?L__=R$@1uF#5zD&id;BLcE|Mx{!-TUzXWj{W-o{dwhRzyRiQFuTIoo)2o@~)Xn-)KiKFs zR_pa%i`d2S5fQ`jyl5)FP%VG~0IaGeg3)VIX3xX}@SaoXCfdK0`bjLn+9g<#DHZsH zQ)2^fJX0nLB&)$O=a4f1dyZY{QTNttz&A&w6e_2-)s=w#ICQA~X_ojqehkD*Ds{`w zw^FqrK~l(`Ivi3WL-$49dbmWd-iz^kV*g9IX{Q-qO8kLVwM}oR%kzXo?RwqSiEe93 z8~qly{nE=pO`DZgLwrK7#@dsGq6j(#?BADbDg#vPpEQD1&$b3rIhlhOwGgHgDYlV4 zsoL+v$I>C$@x|0vJN~I?_ph-#+5+8uWeUe$Na5|hK|14WoTGcbo#_6UjK8Ayt2Ub> ztTKEjUw*h2hvBlzK(ETL>oWfv0pyU$oXY|;{ACc413C!FWp1oI`7O}|kusw{Up&WQevM*M3u}bFRr{A1c z?~nU^eSjC4%>5>xAgYv&!4NGY^2a`E2A5|abUq&v0Lo>Ur0hQbr_`v!V4||6>E|HQ{3btFq8MsPYh z@$!Kc%q}G5<#^!auxn6&P@J@|v&u8?SH$K+&iq_kKOKv;q++7klX#;2@~Fp$-1n)? zq;yi|ET;uo8PR4KEhm`!Je5~ptxM{!8m!BkPdY*|7Ny_YRL?5E4{`9Bw$`jOZ~pY+ z)~CC=DC|tx<;eerq-K`YC9UyA+ohw6Z3Rb(I(tw5Hxb9)&aaJ*gL8+8sbof|u@O#n zWmefpsSMi}B;FjVuL|Ta9>gZZInrFnJ}30IC)zWySf+FGhW1eqUkK$b64Q{o1$rer z$SxZ#cbz{Cnbl+bWB1;|DV0q7!Ff}<()rcL^n6gzd2PDVpPQ`4^hY?$RI;5NlP%wpYaA9b_3W%-69|w4pJk#QOPC93eht%w06X zBt^kTWGjt+VY*{z&KsPHBggqkiYL>;gg+*3j?16pBIzJqKysIpGvEi!Dn+PkQq0nc zWXI{MnM^y5vd3cyMh`6|Ofl|jJuW@ln2f@#V6iJu=?8VJ!9V4RiOkbhVV>)8i=}vC z?_?@M%%uW*LlHf5o3QWHmLvNs$oIb=BE<g4b__ej z(;WKs=~^#4rq^!RPxjYkj)iqP!XXLO#X8*8j&()^llDghJ=8HFLb&?&rttok{k}Uf z^se^)5$NPRakyL|`Mjd#pVUc`kLNs8@iCy@ekR_iE2T{QL+R13be3)*W=p118H;Sq z22u<>CgCSkm8n@&r)ER9N)=(w&lh9<%BXy$iL`G>%_7I0ZyT6fXSjm>6K?hG{!+#F z#EooeU4bkqS7+uTYrP$rfoG0+T{&F-8z=9yBsU;^kqE+cnLu+9IhuFT8R4I|-^i(D z(f@LfLya>o6!e`$uf+ zek&QXcQF|uC%qp91Ba@$PK-}zL!VFXiUU5RWz;`uY$xi6Ke|*|ayzZ^{4hi2jF(sV z^BJK@wnIiJ;pgH%l%~w~`tzS(KGR2|dPQ?86usT!5->{`0LJ9`ddrI9w!7&sWx?~9 zE%DW5aNReR$#@OgToD@poN@iOe%~f4hZB+`dyX6^#Jr4h4_{Z_`j#yt;~0~?xb(^H zPDGVMAaD85^4K?JYL$jk=H=-IlR1rO2!}w8d!M zL-KL=j|(#zC1!DlQlsvM*vx|R&6|lD=n6>vubGbFrhSv_5G|Xm^2*b$Ey8E-=cpV{ z%9%$vLZ}BvdQ`8Po=MFJnK zxPdt_6Aw2%`^ph#P{YVWO0&7c5=W@2@hVQXK!p2yS-%K8G9g!;Zb`ljck;w>qJW>c z%E3unMA+NP-B8)~r@y>zj2HErVDxf*(YqB6OI`41<}`0f zkpwxF%`QUwokeNr-_JXWlN99Q)@mz18mu1B{{)?-k+EQ?pVxYKwdrMT_CoL1=mO(} zLIYGZvxjmq7^}*2^GmPz9@#JCAt=3G>GzRFLVn7kWdY0Fb~N-3-Vo;rGTiM1MMGTUUdd)b<+LSdLicVTe;;1 zkxP=n2@>irp7_?geGUzOZI(&)_|tTPz7QV{&6v3WMEBUaf=rC|ox);l`@>p(~xDQeBejT~a1p(#~Bn{#~*tKb=G= zdekGG#4ZJqPK6;Uo*XGfRF?v(jR@4O%G<3b)vd1H4KeA~aPHRhmxjcO1%541BNJC` z>ZauA)`hm~A-WBCrS)mW4Txk6K|RJ@(#DECUrl68rlg6(yDU7qEMt4Da(b*|Wh|hq zSp~&z1Pyr(aae6V4u~E{P_GlOEEv@N9@$Q`A?|w7M&#eC>fifgrR@iAyW5Z~RjimN zk+?TV+~+{n>q5$dtj~8*)(_O@XVMqo3=S}nrO%NDb;?o{gVDZ-KI*2vh^fBFp}wfD zKG+5rMk^P60gl1ycij-@TNM9f(w`tDX(82{q$rz=kOLnS1>j|D-O&+z=H*kUQheH* z3>rw`9mtXz5Zn+a(3Ok3kWxZmo2ODDtONJC0rrNSE26M26TAihA!W+NO!?GlY zHc5xtOoqCh6@|k4d}4=wa`X>~$PI$}2d9RHHWXto6o<(AT@?pQUO#%VI5aAvG^Rd0 zWimXXt27H$qVgD;gAUE322)c>In|>(JP4IRZwp(hx9tHa^s>xO{&7y zl)|R}$X2n!ZkGZACE9{a+3`@`(;eAe968?e$efdnT!56FK#I?v@ze1L zcM-{6=8RqzkMf(yyG@OviaTbdl#p0sXfhQvXl#&dc;7j!nRN`%8^eeg#t9h1745%A zj1uIIeZa~*b13|LCbxW`v~z~Pf*51Fv z#y&m4u{qI~Gk#GJhqt8q_E43BagwLHjRm5HW|fv*ydeF3phO$eNPDXCi*%f?Ta6#m zOL;l|fp~&^^Yba0x)|N4mXu0UpYn$XHECjX2|9Hd>z)Y=K+5t>p}+GHTBm&pnVt#X)QGs8G1o=Hh=&{=dXj&} z-asdAFJ|K&G(DSV6I|vJ`BXwoYUF>ACoO3v4$r0$t3+RFwWYRPr+mE8p}2geZFf7D z!#EIOIv?dSAAE=`=pN9z5Vx_^F7QMa9cmPBerZ;pFB_KOX?n)QDD1>JTPdyc3j-)O zU9jlVN*kW5jnEn|o+~JNT)(MRH@#46v(RGN)%POfyJWN>AHqPCF6CsQYg#gDQ@g}; zu_Rz|rx;n<_^Y>CxA1U&$Yrr!eQ`ujvI7*X2zh4LB_yS#6ED4F+`UjYyfB>xX{i3v zX1_FNvowD=H`T58nK)wn5nBfDBKjw1)$`NERnz|b2;HHj#bUlcHK~h>_rEr#7dNH# zcOm*spg$wiGD_VqaB?*PLcKEu5T{nBZJ8#>jBVY=}+nmEbi-7kjSHUJ6{DS02YJ;V`W}UDES9he@5i=i84V;99j1S?Q5pp}Jb$Avaujy+S=> zM7_02lV?O}ws1f!EWD~eYbf%-1*p0_jui6zc+5dK5g&lqoAI>-a)AI;^E% z+0yk$=nb1r*ClMfLN?chB0F*0`TjtnGdLEdtc|4eHd=Fx1phAYx)|>^87&U3DvfNY zD6jAbE=pZ#z2wpp|6Z?SY)VkNq1hrQiL)U&WB6Tq8H<=u>+ibWmd=-}O@2n+KeQU8 zHX{4=B9m`S4SF^{>u)JU&MVJ|Hd$`*%$TU+teB8!*(|ogBPgI=rg}K$+NFx)8)i`a zZ7rE?qARoLX>wMe%dMtLmhFqL^cQ3_X6*SNmvm!u3R5q$Eh~O$iyE!&({10pEiZn3 zPYq2)n+=T-&CqNCllXh^vY7n2h|qcKvL3OKOx^yaU}WBCDE&JV57=xHid zPn-F(y<;hl5AE02YTSY{ABk(y#f*6Ek5i}&*Qn=nm2S)nEL{Y{=?`X^pe3#cJ@JQh zkb}_IJM$xZOZ59_hQPz+Qb`?2>kKvPNg2KImi-O-ZH-H-PO+oyNXYugWOLx8hC5fm zRlyL>(V-W#YwPeR&~l8P=w#;b50{^4UhZuHd_byvZW!P7lQ^W(PYnaR08==@(>ndU>rALQ8Aw&%7drqU;`8Ru5Lsy&pC zNrG%2a89$#j&zQ~9&aC$ZnqI0&pP%{kjtu`m7d@WK;fpRRzDpGdQVMV9prNlu5j|N zafCJhI1uoizWOwLA8D5J$L=-r;Y$Wco+aGL3BORrDE7#HE$)Qlqusohy>-|Lr@_ga z1cxJM$GfRr^o!YbNR2T)tCOhd*)vUB5W`NR-kGqh1BcD|9jztVm9_M{F81S-=akMk zAOCP=AXsRe-_AP5k2nz?p1#M4<t8^2H!KfnoqMq(FqV%aD?qzPv z<>4PE_12?XPYweHW6cCtY20&Df0vR$8>@^UKpTWd9wlrJZs0Hvh`{T2yNL>gV#6FQuxd`M&P#N%C`; z$F*y+zZ=>mYC);jMmXWdx%b<(^f!)Dx8xls{t4=!59{o1T&(yZyKl&N(H~RMRle;XxJZx+rgHt#t%o<_r}!s*^2JY zQJNuB32O>K)9n4m!u=-uhxbYMVa$_txY`n-+hT}Pxk}by!mg1pP;dTS^z!wu(Yx}X z`I0E|YAj&T?lLCfOGie(gR~Z6m_x#(G29}c5BI^52$dG385(5YZTTQz;F-KJy(hy< zdio5_x~wvt;35C_Zp7{OQ4dPDzu^tO*GTK>y!j{Gt>5PIB0&_%um5`Hrg=69s3ET% z=X$-`kN-X~^K$GOHrT8g>#Y6gB(nS%EqduOfcwLe+l%Yz%~d{nb$0L4Zp8rE^D^>6 z9S84SP9O9YpM8v&FU}+6-SkOjXD}(Vd@=Ymt2>-Z!0MpyG`kn}LN=7>d+$t%4zpUR zddLLhU;?N4Kr!_>(@-j}!_k!GB=cy7Xb=G>%>|1}TCLX-=9?2si2~JZRXV&j%jsgZ z+FuxAWV^FDM!huTI`JIHDyzv#CHXTOxjM)7i8Sk4F8xqofK2KOEJmUY`|+=*P?gm@ zbkXO>PQp;w8>3DiIgC1MA?~fgWFDu$l(U(^QC7(?riZx;i}8F&zF7nfuQjsNxV2R2 z{_0>6ZnMudd(k$+?_s~ag)UP_SKF`8))~}AzPvO+VJ$Pz>k5gwV_*vY_5o#@YUtpy)lv8ob0xm; zBjB1=ckDBDfdo)9Z9x3#ByKYO=mYWA{9f~adGV7qt8UHb$3sA(7xy3L*Q`1${3^zu z-p&{pz5Ru`?9D+?_$pe`7i-CFDIiGimm*jHb2nu+q1;%A#N0;Q5JrAxAeDBXsU{ly zft4vvD_$-y<N!%f(vX(j?lU!*av@@GybF!)-6g{43Da$KdHL(F02j7_qd-WMpGo z%L2W-E6B!IrdmvhbI0GZrau#)E>I;A8}`SvDp=z&+Bl1iur1_nj6Viv?!-og3H!^LE&-S ze$#+`+(`U7na#-SFwr2=gttZ-@EQ;vW5IVBdP_|gM*FV6_MJy%r=)z!VKsWHVe#G2 z%)roDRsYf8G=zLvksKgddcrd-UUhpmB>Bk3dE73m$$0?FcVEBZb8oOwWG>Owy0NBp z+NFfjk6~wboUn)!Btb$6nOf`-DmKDPJC)SXq4VL&!SPXEp@L#sv%s8Z%ERvLPrnGO z9!+X5!L<;EbO%g;_ix>LwB(2SjksrQf?IwGu@?hiLM4ux^e_@(w>DNU;k_W-Q$L0e zH>ZA&%LnOy9F$FRH6DIM8Feyvcd<{7=in~apQHxqH+@SsZ)tKTo@3=oe~3_+W2i*i zi(buHw>~`PH?-?z7z!)%kn%u#dUfokl>KT!pB25oJ{>G;rqjMF`TKz0-#3`Kd5C%Z zHvCBKErv1?a2mFBNovuI&GyE1<(vU4L8b|xKFU|>J{7ynl;_$emkg&^|8k4DcuRQkq){NUy9{hKK;SLtu z3QEuQv<{P}{vg=F`t|tnN9%~ERf(;fY(#-FC7Boh>tI_ zDAyB<$Vkcp*rZSS4FI`!z)~Ftk%QJ(l}q*thPD}d9+rjV|@1QE0)?g zwoY)dMoDOjxFc1o29Be@eHY`qR#&hj+qP*k#DRP%2UyE}x2XPHBfD@_S>HxvALC}&aYl8{;CmCQ&qHOoX?EwkI&t&%#!XG1K4Ksy!HmYKwI>ufdX&3 zh`$8EG_cQ&X=VCLc^k?9@85MCnh+G&GCg*a&X_4>2TDd~SIAF#22M*s&A006Q7z48GUj}~11yWJ1(8Jlijd1hx2>1+Aa;4$Lv zFe)Ca#lDK{-l!Lnp|q?MMEx-=YH;2D%G|+(r+RIva{pENV1_8D${)=T^CNKi_O)O< z8yv%?H-IRd%vbr`!1}BjK3(($lH4~?T{K%}*cnQvpz&h5(qa-mEaXzMP;0Md8=+8J zy43jXd~a#6wrn}b22DVZnhuxi2!6({KUDX&rz-NbLJRe`%FRKJ;i2xK`l{{G-;!bU zp4IPmCkt?Cl!qIt_h-wE2W6n{H8wp#cKd&Z8?PaKEq?npubb*l*ZX4F4Mv*k&$q|2 z6*82Y8!q>eRo2TR&5hT`E1h9zhEFX`x92;PMFyiS&G*+Q>w_7eT3a6OZ_f9ZM_XIb zXaI+0#RrSR;J#hs9Xm?hFT=*)vku??AS_f}D7( zy~2V#_#V8tW^b>kya&y;UyPWs+Apcwg723$UG43cwc)TIly^~BA5`=)6&+L#@$Vm0 zjmfYdBBnH~537+My5xYlul`vG9Wn50>eU*-QmOYxiR-{3mqJg#Neer7(~|O(t@yQj ztX;Di8#p;a2vadSfH@01&>-S9$Sr#;aede+#kQE!H5ei|V1fJCOgNRSBnBKRe z-r>)wIsN_h;N*UmFzKv;!D`yK&cuH4ry(FoqDoD+zpV;$|`)Z=!heYRnQ z3a(fhUY5Q55 zS{0jC&uhSV8C`e1Oj~FJ`y-EEx-sR$T&np<@y(IoGll7coRGE%(*Q~h4~Oqt36Fo* zQ>MH9t=|@)_IjrmynbQ3NB^tBN$WXV_gzrc0@L>xnHXt#Y9V)&uu`68{?{Ox2Yubn zp7*u*r;?lWl!L<4Tztb5_j#|vh6p!N4()21q6*86abTq)yIR!e()~-D$W^zg7N6r< z*L!C+bB3ua&xc2NZyo`dYG>mW3@X45H!t`F;XAM|*}Eto z#T+`y{yrJuvuId&4v1NW3@#k_Y3{X=HL=SgGXdA#{@ z!&cho$;hER{@VUwTle#n6{38>UO8of_MwnX5QJoMmnAy{FYN#|CA=QO`fXxT?YxOO z=T9N}lWJkuFAGHcn+GP+NbzH1?(#5u}#MPU8VFao~ePJs4}41^2Hts6@!G; z-=i{hKCZBSP8gU<`nyVkBmflb#wOwXE`>7Ed#aWL(@E~+c`9^x%K79#?!Z*Orm2-i z1-3t;p*rsq9r#Nw<1~XxT&@NAo{mw22K4Z<%&U!+A+5TkwWYdj)@o15+dioE@5@pz z(>-o_1-05&Gum(I*i5}oA9v3Gd|@1+VHDyX^pWb4A#G{jv?v#8fa;)+3^)L`2sp{n ze3yvUn|&*_hUt$;->N`jsB)onXamXgw5xzE%`2OL2@ zvd>tOYc?))+kR~yRNozN(XL_S*60qK8-QQux&=Q`$8F?z)?C}rm^S0dcaMAbbf#`5 zz)~LG1G)qF-Y8J8Sa^A|z^;2o%?YX}xnwY>Q>yr=d^k`NF$v$&?S21^@TD}BR;M)vwY(T6b?xsSle&rs>mg&h^TZx}yd4BUTJ~ymzfUHT5;%h8 z?#i$O0896gpi$<#FL?Tpmdbs&;wBO2aV^q{G&)*w3lu{QaQ4K90B!FHrFYO0+4kp5 zi?rs1!&)XzkATE7Ml+uK`k^yz#~55vZ42^*zAwrbc$6iZ!tCJD0A-e-zlys&`L|6I z+D<8_9Php_)k~ztljbz)l)Tlvn=&Cqczk3t7f=(LD#N*qm*%(n@QG%uK*KT2ps5%a zLu>|r4^h6Xcwjh7qm$QDT^ToI0@WiLr^38y&p9*|VrbO%)UT=G_TN&_9GvaGLevG( z9S9gKX!nos{f3C@XhiwWpZrBMjf$2)i|$94wyG=JIFB8IQWlT11ga;R4_yTAwOedm zs!y9vEJlBfVm4nk&o*28JAvmjxoUfQ>44y>hLt$8+L(;$Ee!*f8K*Z_mHuS(_EBoq zgaFXSRP*LMc@nj?$#phdD|W7~iOPQaxM3_v`68>gX+63|S?u5PN$h`~C4d7HkOHm% z!+#Nh{r^A&nZ0$^|Ah#-%AEfV5pRqK{s$uV7W@A@A~v-g{trZC%csfzi-;<#rT>M9 zn&P>te-NQJSo0r5tPiC9ABbr6c|=IBSXaK<`Co{rSnq|sR!CQ@{|6C?Jl22yCnA&@ z5dR=Ttw{fWAR=9-1|=#ORIr}*JX4mR)lHIneJAP;NBdIZ_0C( z?*{uYqkfds|4nZ=rTYjQI;nfmnB8xAHoL%Cj}kuM9_dI=)hplYD6Se3x%R*_h+f0}3~-pco`0e{9N{iQ-R;F-h(=Pp zmwkVnWp2=V)v;XGda^-TMt62VXm~G$>+`-%8L0e0!V0hbc*J+|QRu?Ll6<}LwCGx58ipQu3tBGE+A<=qi~v|1(spCBnE zTaXT27XvX`mZjs>#N!|e+TTyH`#M5F_O6-dOAk4Mk}vf~UBO)>Ktf;l_R?C|?-(OI zaj+C=e7Udolo92UxtQ#r#@HJQcegLE(p(us=%rQhOkM_No}ze>XE`MKC0-96WZI;d-XOYzHVb8wZ$4cYVa zi;@?D5I8iWR?r?%jjzE1)>ZKvbpJs_8v=Je?~$A#&9lU4#1obnry;ZXcfYQXkv#eT zjkS01jstqwb!U=^H8C68HYRC|Hg*~`R?{@LZQHhOn++N@cGK9&oFBgL_^f;Gy7zzB z>s{~rtY`0y|w&hhzne@u%i!Prg&0$N}lQickPMF05it$RxYf%;FIVkM;jc`SKv!at}wEv8}F z9YUa`%4=q~01!a$ziEg|g%W?RRN`t93m=NiWXl(4lPYeDx}6yQXzG4QpIOMKPi9am z{Ho%#=2XbB#31$Kc`>xasdSETSjrc9@w98aRAR}(+tQ-m?>_?>1Yn*8)Bu(M3xIzH zGSdGN$P`-x(ep{%JsM)rNsYo$%;qDn036GHfor6|zw&p^hwMwM?W zn-BWe1p#LgGgCA zxK9>K6v+7JwowT=v%aK;zh&?es@57GjZ7D#F#s|HAMAJ>n*SZh-bMb&>vP`~Vn2Bc zUMQLRq5LLI!eM~f;-i6TfBMqCJy%KWwX>g%dVlukcW3B(4qCZpX48+~hD}|*2L(La zQqCY+l<0zELI56`cL~@Cd0rM$9fL1F4JbNMIX9+jh8!T-h!V zC~C1C2`6nq8NqFSyBP=~uQM^ORH0w9#!!C@_Nq6K`5k}xU3LTy@#itM%%n;iRU|z3 zwp5yrVBP_3&&Eg9uT<3R5;D*eo~=84G^hdJ}D%PaEG zRg!1`ia`-gqJH**0+TGadnsi$S+dz=kQya8*A8EpPq{9U5PB zps3PqES@G*q-I>}ceN2A2uS|mI1^3Jx}*k+-m>jo>~*s#=r|^Eg#1`Y;Zi^?AO@1#E6%Qt5yL;BlzU%a0%!CUgF)pBxdK$Kc*{PF-D4dgkpGi4lkAoN~6F*Bhmk>S~Fsf%hhl;KGlXvS|yx4-5$P|||7P^Vj0ADc;ROcY^e+bm~m<;82tNj}o6e2y$~ zY-!;0w@~kWjx>a-CpVTyA6Y}UBYj_Co3&|JfEmwA6*)^}y=W}DRAIkSS+^g;+xCaqPD@8m?_8d@wIR4tRUAbxwP>ZuZkqMS63W_S54NjM zww)IWipH&KYR`FcjVJU>b9%(JKGVC(&slHv?$dpPj+upz_?afOBe4!W$JX;t=chKx z$38QSIv0j1?UFVBPC5-b0IH%%>Ah?JUr-;J{musAdfEtoO}(9+!rq5YtO!p@=9~kJi@9}zWq}>0MPv(SiJBKc9#5kDXz|4= zWX{Q)dpZO<4R4dnLW%kr;PhkyK)e^vW%AYQNC=jE z3fASw^*pO2+iq;(PdoP@j?!nL6*Osg`$=KoHgBH}iyUD4wv4#CLgs9JdD(X4)jbp}V^TPy=r*VJs&YaKibQ8RCLVkz&}@`LOg|#$A%jg z^=wwF&sMnQ6)7;iHG;Ie?5Cz|sWj3=liWX6V*tf>YZfHc{mL3jI}KsM<$_-g3d0#n zFOj=Ttq#Vd{EmHn8yyHy+eS*iZ|yJpkl>PrAOVI9SO~7$puOgiRr9_pKKkN`L*$wlNHEhx;_Wg&D(U?C9pA*2 zDy2=%26T|R6Tu`dgCBJNrnwV!VKBBE&Dig@ zGVXqMAlZq-A}3NKLfZ~GzmxKlY5f40`j~A<&8QD}83jX9sJNLJr{NTXC7Tl1gE3@u ziG<$ELpYrI?h!D>_E-1}kQLJzfVwRF+MEpk8N9r9^S(@PX8WcZ(2q&(qtJAJ{`M8g zv4MvehUJeZVZSSxSimI{fyhCN7Gw7%g7KWqG9B>>rn2+Oi@`$;wh#zW(ytUq>VV4G zzwu|rN#Q#FB4D)c5Ad@w0%|_u16~f6YMl>L_|~n<5f)y|WiY$q7>vJG9H=KoywD*J zYV@@?&>=3Q(UV*OYL7oiNyNb_A}Mz(dN@EOcokl+LJGtpA9~6D-Bj{jr z5+%fiDVl;)zxGzMl%j#?;eg}}U|IQugv2P*3!L$B;fgsg*r$Dknbr~itiMSmie_}_ zi+KJ`34q$D1BmI#gvS4fp8HTs*#A{h)OZbr|5;KLgTflp8^2>S$Ra5nDq1rxMMVS3 z)BjOYKFe*~nt!cf_!Q<>7kbQWCO|PD70RrXM503{kwTS&s5_qnNiyd@u!6?9hz>cg z?w4MaJajDgU*y4O7ea6;1i0j;ROo?*apYqE%&fuPPYi6kNsql@(b}9 zO-1xPCuH2Z;D_j+@Z@hmMPG8YjOC;fu5Jc7D|x@?Q)mQV_ zi5spov729Qc)zXs@#mQA{VH-Gz_}tr!`AA_T044#r-AoFxwY21#uW7ZXSL?=*@xHX z#Xe4+A}0+S8=cdi-Mn~}w!wDFfqLsZ);RU`(HF|mObeq8I8AnT6?Wm72K|DmxwU3B zj%+Um{w}R;wLg#Xls&%`*lXp**?v>;v^GeASt;3_uJxRpoE3ZB#pt90@$UmZxGaDm+<{!YXVE{rvy}u8kP=pbcU*Z8?Zj8r#Z07{{(NP(n z(1HPgg%8X;aM%e^PihfX5eH2&L>v@%@FE7zVne;;9vrVZN12J%;+R==CZWP&=R; zbfnXQ~yR0^$l*03!n zb80(Vyf82C5+TZ|;wmurLO;!x2Qh~o-zMCyOzPu;;SHGF%Vv8e*kGkzRBCquaxo*9 z5&imrum~|(pSst^te47+8)LlJ1^|G6gJC}VSc(RtJ^;+* zza)t2Pd+&IOTW&Ci|y`QDEe=2Ve1Q}nK+6;D-q9!KP#iYY-FPLoCI>o6vNio{s({% z@U|Zeo-DZ^xoj!9ID};TKfo|LuMr$!kx`j(2#Ni%aVX)gdA4++Ef9oAUm%+mK@M&; ziDd6NHi<+rk;;a=V?st~k(7`3Vlj9xhiQvU+rmR+h{E?1{t1Q%sLC$NRCEO(12B0HDRQQ{23Cnzah2oe@NbzmiWWJ+10yf1`>-;OLV$-cKa zY_N$zhHolJ(5-ETp@7&{An;#JFro9e)x}ngSaVbnBs$heMPT`z`F4#Ez_4}SQaX0g z54t&M4)$D6inVgr)uz2KpIc?*&8$+w6W;DA;PpOou|zJ^v~g++ifJ{g50h!|kAsUFkjkt-%4d3p0!H{`Hzo#b~@B z#rz`_4loGT{XQv#7IsAQQ)!~&akx=Jpr9JlJ1Q^P&}L^=>~F{hpZXvXq+dmye>-D+ zbj?wngx68s$dAE$&$$ZWF>k-Bd(j|{JQ6$f%%YeSVS8T@;ZT<&E1 zU}t9cG2QFm1<+(cGFiVOB6S`=bB7nwnY3aqf?-yvA$RgfYWhB)^_h$^an9wC22Z&i zj7{VM5s(C~{3Jcj1NVFk&i9u0@0*Q3P-v&=F-0Qg)y|tK2)ujB*BiC@R}lY~;RKYH zZV<5SKqXM53@e`sG4;ow(q<$AiIa;X%v^?%$n6Gs9f_mE*9MM6CO{Ha;trDIY%SFZ z@puZPkechlvpsgDOxs!sRc)g2RdQdN+CDx%&HIdYhP_ufHBibU#}Zn74dT8Vb!@K? zksnZsk&yL^;W3r`O5TU_o0{&EJytv>5&$Tf&cf1A3v9iLX4HBjjv_ZAx8U9{gSf}T zbLYdfPMN```AvQ~Tu5zeDrV2D^8?gfr*&V+!Z+jr%>{p?Pe+w-6*DMdni{7!1tHQ& zQ~R-L_h){2EWz#ChI2^yk#$5}%D18u3%;gW>l)rM(!YzFkDkshemSw@@1#vmCQ2#c3o#*k-NvB)#1XF2wTsjE zIK=us_yO{gPJ}Tu0@KfdR|)`NuWuvJ^43R$D(+LR=hI<+GA4w%ge69N8kLH1OoICD zQ=`;Xihin(E1ojLBKtMLLvkHyxM4;)NiemfW(^2H-RJClf=*iE<)v(%n=_847XcqR zGFzuC1fhT#*NgQq6-7(@m-I0%^xOG!901PW_VKvo##~^vWg~9-ocjA98@k<*JX;2U zCg&!C%iB`k^VvwXY2WMcz1R-25atSz2e7O8Ay z;+LV;t=pnt5nF}g=aq^3n-3C;=S1TkW3)YkWcKxcEM0#MmDKR2cIaD*sXZ<2lDcNb zdRyX#3T+}=^JUK~*}1LL1Nh=xQ-9lin?KIn=#+NtEml2uhI)>@lDcUcZR`e?x~(CT zDQVsOalj4C{Do=Q7BT+%_lJcS_j!%mfY>`uh?f2^7qkU=JYV7}vD9s!2`X!4+GUsV zLvMX!#(mnk=7(OT@ZslqZ5Ulh2v%Gk95uV(LbA_o3D?^Rbszg&V*)L|;1zmG`m>tk zZ6}&bHum=qE4=+P;i|`_m5})k^@8nfsS@6(vh9)1FMV3 zmG4?fJg?YuyYIW9!?*4|7Q=@rrcZypQs1P7G2A~J8$1wEbggu*itiTZ-1W^O{xu|{ zETM){zB71Z*nQ(br-VM?>G=Hpi1VCkWJviSApAT2Q>T8 z091k5!2d(%lh2Z^-xXKI``^j{=~#t*N-)5>|3L=$oS6lmknq&_-^l=h;)Yn;lDFmm zDFZMmq<;ComI2te^g{msk^xjGKh3;Stsee|4B-5GvexD|I=|4vUr*YAdUFp1kJb%jhVpjGLne=o_l!yiJl;AZ zaG285w!U|@fw^H?RPTW0G@ozIad|hx%4M~{LZNYvlgYkqhu7w!6`^>I8cZr`)CasA z@$ADy=(3UQzUpCO>%QkD2mIr#F<ug6 znvineF#?Xy0U$T?J2on@~ls&cjE^N_E2if z>+^)zxXT&0jH37x8knv1!u@kK-@7x$@XK!{-8D=iqaNC#S%*@Eyj`~k>O)KrRa~JC z_MBSpym74O+8HT)Hv*s*=pe`Q!w-|*A=Xe?+%0G@9Y!mS7e0W^=a}I)TNz*5%P%Eg zz9?Pvm!sD5pc?aA!2lriLW*sH!oeHa)-aw#7JMikXLaeQ)m_BjhGKmgK{&PjP(=;d z9ar#X7S6`Xy&dXTu{Y=aY5e|kwYQ_XK2-m9cRdP!WH0(kRuN|m+kj5SU56v`*V)FA zXjzeH1X`4=WH!p4-k_&A`Dl`L%(JvMAv`!>{nZD#YkDx<2&y6G z)IJ$(yZD7&JUIz*7_5amSQp)t*aK%m3N+A>+E!!+A&0Pi=0YZ?&%*)LnQ>xpP9b1H zXtRu9&m_y^;LR4yHR&_qf3=9%TeDjlSsh9P06`V}prkxS0C`Cr=#!#A2WI zSh&f#pP~~%!GZ$7n0>im;4zQIy&pVkpAn*xWgn_UkT~~kK6*60WH=-&T-pgSu92CYp|c6q z;M4`AX>!uwaDm(isg|{0A&x@$XBU<;y$4;#)-9Gem_($;>j-7V2w-ek^ok;VAE5LU*XP=)czxJjw-fI;17 zPZ%VRx$J$OOQy3Au~#%B%5<&z8_mYP=AqAe(^lw26Uq-4hzV||?!_?9HVn1h=M`Il zVnk)jj}nzm>3FzhkeYl{zHGE8BM3tCNE$cQLH$}zP&B|vOC$N6NkxL4sI7OCMH-hP z>WRp;ZDg8THn}qGP=087cADXnYsaEkUOUSlky@z_b_e0efra1Ru44|65jhk_hB1C| zDXc4+D!<&JvmA33urVj2gtiU--Qc<*!r=_$_DrKztyg%jYzQceTHCGu{*my_^5Nd4 zFd3bp#7xAVthQvj0G>xFQ8AVAt3Z`T3_Zm*@rp+vVa>x#Yqx)Vly0m$yN7d&BtIs= znfMjK1(7$Kis|ytvj7#RaX~5mb(Z}%Pp|@b2rGykB<+wnkwoe$7w?YEufQFZfbEhzSjerG!jsP z_^v;^L!6yo?(q6Y>M%mzobYV;``}NwKK7XuY&Ex02Gc^KQypSGQ3C+b$F>%}3bQzv z0a7R+NbJ>kfh|*yN2=tRZG!KI3UN+Jsrh7da9vRCRv_RH98>g3^NuM0 zRfg%->wpr7?|qw!r*ZE_cxV5E@0%{3jo&54u$ET_pHmi}PbsGJyfcCrz#r4n!9^J+ z>(T=*-mvsuTM6g~eem2eR|#NuVpwq<9rY{qx938zUf{#OT6X(!q#>^5Iz*$jgKti& zNp$S$kay^Fx#HYLi%X>W8CP71rr%eNz)#S{%je;%pG>g5W{@N1H)DQxVtPYif&eS2 zfJrUF!q5=u5a*g!#U?F_20k$N7*5+t2;rR`ezVGV*^q0vpc4PkV<@k`rSaDw@^HQu zZ^g2(I}Mvd86vggu*GEOt5#>NW0NmmJ<*MQxy8dO+El)Vh2IQ^74L-mBZOP@hqNH7 z5S6(b(gyrEcHc9zOE(T^Kr~=^4DccdM5P0xVTCCe2lASK|Glg%bQ~elqF>GtHD4U^ z{d?p^OVCex!#DGg-ZD&QethS9Cc|%qXOE~e*V?V*PFtJD!Uw#iqVbn(i z(fj@8*Z-Cb+IO%&LHyBViMLS}k+kA1JQC@GuLwE#j3odrNdsdOs^g^N+qkH8hs2zk zBfKQV69^uSNmAiSGFD0Qgh@z77`;Xq6`bHx)nshrWVLjB&hHF9MhN;R$vtz)h5{+Z zNGYkiUao$uI&uK*@syJUOgq9C_JC8CL08?ddsI>85djT1tGY^pxrahT?Xn zGAoqQfsCPm4A^|q(|U#mIY6TvptXRsEL(4O5TZZh=FUI@|+C=%%#^)kiuI!45;;_B)FCLYEIDlqzb-5rNX+!n4 zZk97lrJH>D3{mAV5hrOvc`_C~ryDs!I{oBY&8=?j{VBskM(uM)ZSQ#P`*SUTxDH0B z4y0EH?^%bKS%=(Nhq_*e{!#}fuE!Fp$I+|D^Qb_DJA?b8-xJ$C3NiU~@)<3dx0Ck8@Un`UG+h^wQpvpR>EE-$mi7lqx; zwY}J;3aIu8aD&kr*;z5!>Mqm{CGY6y>9!jr)8PqY9k!TAR>Ji7#yq+8#5V{I_E)R>;@^rQ%4PKd>vq5u6Ds1 z{@jvdfSSkOH1z&7#O^gnCOj;t2oUxf(Af~+Z$cvR8d32YimDtEQyftu;W6JBwmcuO zKJSAu9@b@+v+q(O9vpU-+a3M_I}*t}5=SEEx;Ps7Ix0;v#@sb1f;yUI%b66_R2`&y z_H}f$Tt zwY}7-_m|n^XWEs?Xf!EHgqN0d!GC|%0KaFz&E`-llgDazqx>eNQWZvEnB<>kYoTmn zP7@E-{#faU{1_)1_5Yv?@(#4KnIioW)9&*IgrADq{!JJBaResN^!K5_k_Rn2^5{4%v#%Th{KD4oKXZoM~|##MPor0HPMS2L3!U)kPtvdFZzoGjdW zvhV7y@=v<3vpw6^k5@C@$!AwmC9;6}=~sc-@3BOSL)7KXYES6Jfs@aRiDlD7qm6In z^TmztJ^dCm8)ZSG^w06)w~O=B-@SS;<9>`Zgwb z^5%JA&v1r8a-U>!Ev6_=)a2lv(vOSvHybQtKZnTCmE^VXY?KL)%TRwZn@0oyD{F{O zD|PyzT8kfQ=d5k2+qNofYT8th&uXpHuR{n2@f6PMM`$C@>+O~0rE_8X5zm`geo@Y$ z(01khL0aqS`$M?pGWn-vH)5T+`B*pkPsfhqUj{@S7uU`kECWC|qUr1H^0) zn*E368)&t@>k3{!WXKb?x(?FHZ#RPSt?P1FVjn%RvwAuF8`|FvSIg6!RrFM) z+FD%Kn7ZcH?emtg_*`+`XwjB6xTHSdEH}gaRt5p=9eC-S6c_dYy)9fsBe7p3Qp4<2Oo`W?!t?v;~@C>iqi(Cup~n zadpzDphc@6*^$YQG6)ypj}|--HAou9-x!&rm;t}7AI+S870tjXk-D8wlc9U6Qb z!aEgb2xJ$>pHdFNa?1ZAUzLho{5V9b`90!;0|mSV-T+mKmCDDbczk~;bnREt7sx1u%*N%Q7Sm&LkEs+)#+Pj9Qlm7;2Q+ae8#Ylh`{E7o zuQ{eTGt#iSjw?+0)u-a?nMzPI$Qk9xri@6gRA4~><|7W|Gg+A2hG~lA8Vwl;AE)r@ zkg7}^8#8}5EwT*?uz(Y=)wYEHAI(;@G6;hFTmaE+9->m21pdTah~RBLIJ!)dPJTW@ z_qKpwx=fmHVm`*>wvZgHTvkSYAtB?oh+e5&UTb0@rQ^1kJ-S@cLVhu0?Y4w}y8M&t z#A43#Z7Jkm(#X#OqPsF#r3%&5iJuNJjb)z?4At1Vt{>=7SDrg_P z>LQlX=$R|tL4~@64g9s}8fqHJOA!-{m2EYMHIWrXAvIPkdgslxpyM+hv7-8ZQn>Wj zpGj4#C99o>clA+vRfb4&8vYK=waL}CMy+eh@ub_0GLPp#2^8J=1kI*0scIk*I{?sj z-&_}6ZEJD1mH;K)1{Mrg+y8jcfhFl{?R%?sK)VFcdii6U@}oMYW^TO1`L~bJ4E`v7 z(O-E#(cz$wqP-E&UG!*eDo10pCf*LVNZ0B-7OC}E%>1%)s#@6f#%zDUwDla30yJC= za=)uIc&Byk1zh}X!~*#JvA^xyg!>lohiMJo%cL7@{w-*H`^);wL(7$!gLib~#^AA5 z`}EtTWrk41p|vZ{GL5|8UHJ}PWGgTytuDx-Y704$dl0{+_M2y>?(6fzDDb`3FVyzy zEF(d)%vgP-G4md+LfaVSFuS+Pi{FRHBsg=f4zM4UKuTq0$VmcrWj9PPafaC04fv(UBH!LL$4 z(VD~a*Lor85C6%iljo3SM0m&pb{3k2rJC*<>7Tn?GNYj1&%dT0->YH+S@wPKiI!Ac zn)`BDYy<{73wQ3C`lL{-SzYyhq9gopMzLKFW+Pm_u)ppXVEpT4__X&^w67<4`mgVz z;5;5QmU{$6^eBDTU_Cse^ClpsBR`#2{=vmP8m zvL=3UYefir`v?wxKx7A`+{%^SF=rO~#!fIHe4hr`3b)c)PfrsQ0E~?l-lt=8ABQ!* zPVM#D5B@?o?74N7i_t9}My$Wxl74$n`*w14=7)fj-`o`4eNOy0g#u)J-xkY4?pOb^ zb@TqvgCTkuiV@UI$M`T7`Um6vGEj&c`ZVS0bDIhcJpG*e#HZau29A5(Z!|-Mp%{RI zg;kDA67~B^;ibv_Fj4P2d*3d7dtdi{`#`7zNZ5$H=HOuD-?*qb_INGVg7HG6{fc;v zex>?}<@q6Nd;dD}TWa;g8uRk$gE?^ZpH+n~6i2pWC(?^EDgWrCDUJwJ58B*z<#KVx zfk$Ot4Pbo?U?&Ld3OB_vK#60bjUzvYBq;k;-7PqxEDTbImI~ed5{hUH zs0&Hb4*yt&wGiN@$AK8cjwneEmRCk5aPxWRB~mBwjcg0MEDe6TqMMkIAcl60h4j#eF0a7Iw1vV0<=^fU;X`l&A^-gOmX;v=xHuqO+m};2nz_w$ z1re27+vWIsAY*(Wv&>gv?QqO7FEPs)kw=8_GQ!ERC>RP*u~En;L=#JjV0z79jxf~n z)c~EgV1fqe{SjZb-7iXr_7-hsrOT1XPm!odp(1W5G!VeO1W^bzm)&h>y(P{c?EpH0 zm>-7$tE&MN?XfJYVQDZibCzEcWn-ArVLJl&Se z5bj)49^DuoO{EjPzM3{RpF)3?!qA=o8Ov;M$#f4Vy|hgCaVOC0bLK}%8Q;c$0uaLh zsAlb9ZoB>$rX50Lbj?ffet-zrqk35?GeXPuw%r#wM346zXj1chY& zvdpAQ&-~ny>C4XN93GQWmg*UwgM^t_5SA?~K!n(rNG_1sc#=iYo;d$2vC1-o#xZFm zJPGzbeU^{!2W*DOYLcsR&RBiUhhI6iabfKAUR)hvckW)iM5#$*h0G_g{U^l4Pcg;H zd3-B&2#jfwk9pMPY47x49C;d3x40Y2xVFc9%ICNN=p*i ztdBRTC^wT&pcV+U9xus|hlli)?_0v=WPE_NE=xVJ)0=}YbSM&IsM10#3p}MEoXB?* zv{o#yN|w#ZQZ9h+ue9YTXR`sYhUsBsMC2r_%0xfqKnGL0b#)`vGD9Ay0z6-4BhLyld{70S<1Vset#0>MD=u z5k7F(N{cNqwa$aCiGq>wi>{MUi;GR`E>eakak5Wkh)bU$r#i7b1R$RWfVo0D$-YVuR3IN$R7~Mda8ulG!!Z*>&N7G5yjtbJlea z=$;kop4aQXVDGMuq|dsHfZiGZh7$K|qx8Tc0$d<~fy^G?yq@#Uo{ROK(?}xUI`GX| zhof`vD^>ebCuMAXFN`r4KoO9S-3QnZgz@Uj=jy#X>yU?~vMlCR!R-faQ`(r(Jvb6P z6&MndG_E)|VRIR%W{}T{gCjdNB2DDzyEGV?`%={V>&^z)Upp9X%FJN2dAoFYH#CNc zwKJNvMbEV%8`=>(Lv?4g2{>V}{zLS-O-DME+3H}+rQY5kZA^-Oc;n$(%HoIXzO$PS zj;TJWSH|ann&gEc=YZiZZr9Jk8puE@^QjS}`eBDIaI75iH{nr9&px*eu!|y%<(ke#?6c z!Ci2fb#UCMB8$9Ai=5p`yjbA>y!wj;V5S6s0Rh0Ie*qGj{{@g>ZV^3JY!8L~Pj4|$ z54qC*x3?G)ynY@xkI}-)QQ8so9J^bAo41_|_Awu^z$A&QEjp0A< z@3U4rQ^|VYJXP;DQ%A+>8bPWjaP&r!idK9%Ziu9^IqtB$$9d9g-4C0z*qZb^!5hgk zBfbd0$L&y9K{KNOYR0xrAM%p20`aqEz6h^}1AM9|cI#sb@eQP=oe1HQmS17C@ag0c zB6ug;9t5gsMZxF(rTOv7C9MQ;G@9+((JJOv29Y=YGKH!4^ICfzpr_E{6wB+o6o@+V z$zeEZ2?L#_EN}aMwu`y!VV0M)xk9F9imYjF5b4N)H&+OfdFu0@apU~B3b-^-iz4Va zQyiV>q~!ZjxoJ*r$OJXRXGd#B9eD$yvVe@7Q)_Mg9w_};>f<~S8Qq17jI>2p&nc5e z>udXI#lh3VS-qB&>_`>cF2kuymZhbwxCNz6wIa6Ci9ur;I^j|CHc|5yeGz--d3(e% z5p7M;F(bHUBnIoEi8-->EMX8`=!~w-Sm3mc z>9&2<>+jYRidoXmxh9k43k-b+%Ff2WiQbbRnr35L3H=($PFo&emg{G1=pKjJ zF6P_zn^|q|u*nY#WcWHZcTu5_QV-Xo;@j!Q*jSPz#wHIi_*oE>(#KIT#cg=#2}=Nl zFbSNYQHYrAIVtopaQ{+oH(qvPJc@$=CnHy@5X z_jLx=m484_1I z#wbyc#JD!llZf~BXnj|tWirr{YXiPMF_#P18)Ky`>(bm!kOPqFqV3vVag8@1id^8E zZ}#jZg!gjRXu6=_}F7KsQn_xi(N=CYl}NuXfOqHUJtz_Z{; zQhBud(NfF6_0vmz6u`W1*T8PcKE$XZPEMpY1pwwSo48wdPPK3Ourc+W< z)=2-UwN1@Eq3NH1#8`a-GX)t}iA{uYcWMYTgR%Luv^V1@@&kQpcDMu0PN{mm%VKpw zNHHgFm{fgeVPM=+Wq&-iZoRbnL)TPwZdV1JnFrNZE9dH5Ac|^7#LAd@sC7@)8e|n- zdjS2Oyw6M4lSK=!{XJW3T2`#~7J zjYE(7RssK-AEgSLr@9XvtJ5_ut&^J<9{+{4w|t7L``_Uekk(WV6wYj z)DN`*M3dXdLU((B{ zqKY!L_M!>jnqy?Hi067NW~K6kVyN2JMMCSpvb5MSCC57z(^TQ|#}d!&uNW?n;GJh{lZIz9!ozS|pp(v0N(&(6axN-F1 zwhbo`gy{!hocp+S?ws3tPONX954&~H`T;Hv58t1K)gyTNg8}?r5OQ8~Ki>3Z@TbhR zjcMKi>{Pu@iH0X1%Fx-O)>-shAdwDE2ZIapFLJ!Qu{?f zva~;zNvrnskA~k_z9omEGqb-jYyVl=(~*o~`+V{=pcus7Z8} zhmu#XHQP-js79!Q`$^h5&VI*SciAyqa?>nF>NQq6UO&IhlHNSR5%DZ~^?aJ1(QaTN z@^J9U;CH&`)~a;u%?TC7xx^jB>F@LtyF%E7SDg@Is!7_WqLsX1Du$ z>BpAH8Rrt(@47_8t68Y?do9oPTQ|Bi_TEEAms@|eQH-u$47FB& z9T}(N!yum)x(kLNX-#KrIGNfYgww(qYa1$R5qM8T_=eH$X&Xa(8{^r~L%k$W=Qa?$ zLP101iozd+H3DD^0_L~C&(mCMs=7-nfo?C+72-lrW`UdK5X}H$k5tgpCC2k5hBDwWiQ<>O9O`?-TuWr+&*-@^ z4A_4ImfV0a3+J)0sNxZy_w_!^jKMpp!K|9$&zIPVr~$gm!HP%01@n=Y%K^xuATU3C1ksmiNwMWb<{@6-U{OnMvY(0pOUZ& za~5<I?-_eqRr=h&V4MFl|4Q2&wMNOxxMI&neMDuREoIy;AzG7(UUm0T8;uql z`+_Hqu6yazRv2+S8uEqF#cq_~r&jDAi&%9OWbNAop4Qm30W{EMtXi8hsz*TXO3Lf` zIGgzBwmaXbS(W$r5p6ilXr9|l8KQ{Q^ck(3xiH6?VkK*2f;6Tq?U7*X?WE~nNy=k5E2%jW}Cs0ROxLX6>zVX;m)4R zi|2uT;(&pM@+tr3_ZX>RC+n`z`)XI?ufz_yxhfd|fcZb;7ybaIXl-TqYh+Gd7!%ni=Joni$?RDa;HTh!#V_O< zd_}<}&iGtQIHvmV+RA+nPF=r%;@jyt+3D%zyPGC8O_i^pOzNZZp%+xd0dkK)_0^VKtZ+fQ&gL9CoetW*jZ zm1&v16p@wfy7-q9Fzc_XTFJ_b^|V0$%dYVFsDz5y^}g|kzFD&NKb|OT_AJUct>EDP zZ!G+^E(H;%UC|f=AeVNMzBXm)w&cuibWBVcju)9iJ$qMud|o|wnLQ<7#mdw6st&Xd z9lakW>l7n+KdpQz1 zjUQ1$A5mQGdvd)RXg9#V72~W|T-uHO-8?!qfBH#5lNi*agWx7V!+)<*pSQ7v9!KcO zMn%18@;*=W*jrscOhulJN@orJDjQ_xA4nud2- zqZogxqcG>WG8@1x?WeD+Eh-vg+=ymvAV&8B~U9gsLV+~QT%#b=YIm8cj?=e|T(V2xWI9HV6 z7K*;tZ#a(_#3&Bx4^^1|O2e7d*(W}ga@#kbsK4B2 zHKbnrjpAy7mRfW1*SF#Hd4e=RQg4_qvziXmrP^^dfO8ccb2LR^k-Tfl^zFPk8-r!n z;MU3@{PWuMm*rA6xOwu-!N!+aTp)m=WvXM7N4MoCqI@MvpD~7QBSC+wM}PHQ{u;%= zdb-NS(9~*{3Keh`_NS_`=zQJPW*O5ukDo$c_;sJg5Phg|d>+=qjo1S8%L;eL!U!9~ z*;~51=(gS`YV2v|3;m^{4ko^O>OI!F#iuVeY#l$X*W5S1v$*cY?X0Yc_2Qid`@bFL z`n=AYxneoV^ysO*@0A=+u|p-agQc*~MfE!joa6>77;ChBI(hunY-p z%)9Cj9kGG1#+QjE*2-A<&2Sa9IJcGV_Dn7vl(r~oO157cY#dvm(8cMX^ZSI z4vhbB!@>*~eF0Whz-Ql(G#K=qMc&zj_c&{*BNAc0P~bzL?LjlQ^bfm*cwk7j@;-Wc zSZ7@HvM7A(xXr8%Hm6;Df)}!X9K(KmXl5I4K=~nX(`0fJ{&*6wd60Jj7t$~f|6xfq zHx-4<*Q;T=NzU5$!n#z0sU2k90Tj+t#U;rKHlX3yn|@0YNiOI&;G3(>1vLu|s2SERXD|KczQoy4ZgslI zNx6uOy@;{HiQnZg6Cg-GkV>Y46HOXx8PN|1=<=1{f8L?~fc;mwZ~2k?q$robGy5-# z|4Yh#D$|17N(6qe$b*`LV5!|hd(T7X_lIu8Lod~1zsTd@$H!sc$I;x!@t()Y?~l`n z$62bUd6B1uk55a!Pb;}kYdudJi0@BZh$pdzzuP^q5RvB!>*u4~=UnCIlNp#}07F2$ zzxA`;)^o~B1l0~zp(5fg7opmRW&?}>)MDGB0Z=STi8Q4h@enKu;UHq=U5N++HiN-5 zGEb3?v_>Ut|iMKNj=F z3b|6*I@eZ9Rr)<46uLLoYfT87@59-;x3*hdUl7QYdUy7_L(x>yIePbw2UFQ1p_KZ6 zolh34K91z*Ke%3OcKM=E89ciG+Mmjme$~k!1{`np+_rZcdfj~mZJ+=}KBo6zBv_|( z0m5gsMilAO$xNOvUG1SL3e^Ua*k3vs(L5;CUu^My`i0te*dh{kqA+Q?Ite_*LFRY( zKhZ@cL;i4~O8H-C;jlxH9qXi~Oz6U-(ca*q%1}(v4CRqQ1sz9whh=1j$O*=03Bk~I z$@h(G6|?Wro_{#Xz3PBCu5EFpno(U{)yCziQP)SZjqh;3$wMLuetsoFe;J8rVSk1==CZ$DSeQX}dnHyArJ7CvP{1 z0`nOms1zx$wh!kJ@fk0!o$hZhPpwK?%N|5VE`FST_f6(?ip>;giFZawB+oG}UX^SI zXe1-gznVrv$+Cy$q{}!07?wc%`P1A}& zhZPAau3>4z_H^)B>v+QFi+Zw<3R}Zj@zaQ=1)WTrj~UG#Fc zb^CmH z_M5G%sXA%i^Rt&3#LIO_Lws8Qf!Kfn9Lekd> zQtYSzP(}?UT3K#>K2yn)Rz{vBKZnct> z_cCL`I#;m<-gINzsEkeS30&EapSn5A-uw95C#1;lzXWNISCJ^k*EXgp3e!wzIf4^I zm_D*(26MdE_?22VxW}GrHuxc+UP%ylpTomd)=*G2wqK8a#=v7#MB+3)tb zas_65Z%(sn-qjY{ zikFIOl8%S4m&Gu?#ea{oCLJ;CtSPl8!E-O5@P^y|Tsu$T-a-44=xiXsy0g$f5(1tj1uVcB!d6xyt^_XWcp3 z<;1X)V6*E4wr9ggqO=_-yL%s*f#^zKj#zDwhyjtF`{=8Xh967b{JOtw3vHed*?= zr5lU=Gn+0l*CRyi*tpzQL85q2@V0IC%Z0V#Hkhc9x@AIL)rt*c4L|3l^;F7PhkW)}7G0JGgE64Xt&B z27KS4yz4XN07>F+6o*5idm+tLhMSkmE4=)D_!@XVh`*5^x%pck_T?Pf33dUpv|4k1 z|E=e%?TdicA$>4=)JOYBKCP)f2HwCRhYdQ6?*Yw~Ujor=uOd}Y523cDb#`|0d-NX5 zm@h+SMPj^-mo9F5{*?U+PjlG=HQdyiyF0pzMCrdCC`G7jHJyD7PFaP^t(4nRHaK4R z?=9af)d!w)CjZ_j#H-OBH#n#f)i@ItS3K_O3ahvqRafE zA(Us*O!1*~qHmKsVrskH-%o&gnX4(+Ea*swwsne?DXY@xx4EuJT5Gn{Xu@>I=`8nh zSLi-l8IH}mEK%k=(VOeAsY}5#5;}dhi0cL;r$`}t?KK{=ss_!1Q=35K#Umdv5(o!u ztQt~tfE-)2-KiMdm9Uin7qR;V;M0Pfq)7gLL!PFg% zaLYbIk$jiJknbT)D~tXwo-gZkZpgO2y#ZR#CszcM+ZgqaA-$);eW-`WBvt=Wa-h^c z?@zpR^09=-*0nXav9@Y^_CKdEA?kY?kGa$jGj_?Pa7BhNb+&*6O|GXCt(o?vmg(zi zucuQ(?2eU@4>$FhPiGd29c#SbX|v_3RA z8xR%)@V77fzY38pNymkV#7LMA|AdLOeq0zw{Q&jjz63$x%e(mqFB+1}tc zzVKyDEh{Y?{A}i`$d+9U#9h}*pMWTh1N}h)DS?ch$kz2g)jctPl7GiluKB{R4k^K) z)?hY!BXm6uw3j%3x*Sj%8b3NCKeSd!k0AI#sN|Lkjkg4#D+<8Xm~zZLN$(UQjWBMu zQM$J=bc)rfP5fV?if;u5(6@Rp$@n%In-EHQpS*+(YQk=nz61`ik_U;skpV820!5Aj z#Vs^Rg4CpBf@Ca$z*o`?4!-Tw0hsZ=_Y62FV$M*b;6VxCg&_pR9)e;Bpuj>58ACp5 zg_xj*7`j7{?IBo~Qu22p_Qau%sG-tYL2{_1)NxQ^KM>px>a`N;dlc%;80I$`>SYmz z!x0+M8fJIIFM?3^2O9y{ej$lk;R){HDQV$pjN$T^P`zzD_c~Q#Np#;v2f|&4ohye3 zYD!`lxauyVnmDqSKe8SbT!RW$?nO@N`)D#7NxT!;HA>#gA2t4mznu}LPaHDd8a25R zHSG?;+=j}kt1>_x3+tewhLP~v=#bLr&AaF=;uzvQPytW0x0;3p7kYy<@aZV(;x1-< z1@h65BoL$?)~Io>LQYj@L89Um0g3^bV!*4>%C*R;#&4BPV~6hez4N0CR-&-SqHx-x zu(d%shB5R*PHiBtVP5n`Yc%*ie)BFK9tYZ&QN>e_4&Xw!_KTVf3o%2Doo0+h6_DygQCLXBn0Aih(VtXNgO2PSX2}gJQR<#_&5cUr?o)yVF`Uj z31r$ydIE9u;qiy1$UQzlDOChhEf5E7)@vx$NX5wG3Ue~c*k*57V(B2lD;M`wKy>}_o1N{Nm|?SAH$QG zJe)~XMW;{`QMOY)T4bocPYkM8UUE)6l~C9*dL7N{@r61`FFf5hJ##xO>Ce#@;h_vA zrj)?Kw5ZkKQh{U*l5{y2a-{pv=9T2{E9v9M(*5Z%Wgy@@EhWJnsIf=4SdFs2Px*}) zRdGsJxI6MY;DhCn5P6QJcFtmSZb!Etr$Q{_0B4y!?X=J_+*78P1V?G(N5E{oDP~+$ zKuk5yqAaD~(T&nF&FXJGpJ{)BsXhm=OP035-k_8a{Lyg2(Vi4(oBR;{!) zzHmDXO1+IuP!5_cixy|b?Y1ncUz7Ya4ojA0Ha&@vG6W_(^M1+{C&*?~p%YJ8hKZmj z%#wViV8$If;m&wn@X8}4HLY}{EXtZRggqk{-9_Op%&P`eH`;a`)`ffeRbJ|DjTyxpqUaykMseeIofqM-@F{f{VGMe){ffAZf{M zP%JSm*(GxW2(2_^LGo?nhejPffr^Gn;#v*avPpr2-0@nB$XXQLg#43wzKYkyvOvj6 zY=!Yw6v0>g#y|z>`6gdMu1pNGZrdFCvE!nmxzn3RRbfA zM<`l4B&W;zL5~+dD*U~bkJ@W27uE9NzUZF@JNN*$n$OB;VUL?Itr*jX?A8yG`pfBI^(WIE`d+4;4e zgfIR8*_+UIhXFxMzz@KD`#2y>J}CNTP+V_N6hP&i0~C4zg;_wp7f>W0B2a=7%tBQ4 zhBU2*)U1c(IY8W<1Ej2YhMj#Pmnf#*!)95-7M;UZ8^fe=pnu6*9Fzt=;`(O9-Fw7S zZ^VHd#QSDAqLr+ssCsxIsSp5qVVL2F&e4dC(WuAKn99-c?SXL77iI2a*zxS(ZK#1g zvPSRN$ICGT*mxf6c+s2juhvi`7+_|YSO%&r@@gb0#wpjOM^-MoUlDx6RB{_3BbWX! zk`ul}>U%^Qw4NNXo*c@WoFt#@>IECuBG0{>`6ZJmUZU6;g}l+&R|39pT*W^J@t8^*%LHp-;N-5 zT}i!K9MNs0fvCy<3VRMeqLg&t;=fDz={oD6+{N!vCp?$L?=TvvH$$&K6OW0yrzASq z%nj~CW__Q+sqm2tq5rLy%-@To~q0i9UlwR)1snB>AetrJ%$|=QGsw#PP#af#0zIW z40C1_Hj`?zn%1>?Svi`WJ<5EJOGcv&l34o5`-W^}MO_jGkIzYaP!?j6`Z+NnEQ#J5 zHK~%l-fFYb=Z(RZr@D->0mgSU)1R$S_(r8ET_fc(mObEHMe?I)d0v0!o6Woo#|uO! z?npU--zF>o(Uk{m&zf7W?6FOTowHIt70`WF;3*@~3>eMnT6WK#XNlfmG-6yiuu?@) zjW6K&NahqxOF%}m5FN4rwb6l>ESPs}DSW|EVw;3fF5(F9sIbiqSY!0O-5}t4(}1>w zI_*GNV)ORNX>FM7JKG0W{ncMxyWZ!^!j8LIXh4X~@IKa@uPAtXmo$)kRabm9_Hk9z zU|yW^dvxV|q&~iT%AQ&iOjMiBtqs=BD)nABt;lN~*LJc!dcX4t$?bCezF-Fi^7+}a zc?J4Wp0vA2F%uT|?-Pu%xa9>b%;oIOSH=g^jk6}Rvjg%!Mj!v>cfR&0I8%q-TZcqn z7%347bY|-r)aQ7#Sl_S6zmFMWCkc=Jk>88V!jv+YJvxT%f3telU!o#nTK=|q5h#Eh zBl0!a=b~`iqQhrtZvmTq$7_1u{~3kg!^(g)3L9)gl;~h{Vol7A+?8FCJ{t9A8{2Oi zdwhDgcXnAyHCv&dH@$}v(wOcA%OlPX_OeC;uO{kU`tR{9=Mde~rwF}Mz!xEI2t^WD-5 zr_!EmGjA~=h}ko-sO7hB4yCtXCeNhUBFek^N2ZU5OgcXZHZNThw&bRcOnx1Sv(F?V zp0Rf+k9lh@dh?F2tLAX#Ry=pF3N9qXqD3SXWVa7b99b&)jn*|O=Ugv;_9*@6*aBPb z&7c2(UHC`P8rQK_p@|c_zJ)Fzidq<97PM)*S`3DWi#&;dmK^un>AX3S#d)h=Oo&xA7yLD zEq%d0=}%1zyngEsjP1x;f=Lh=!>j$`0!hGdEiLmnhimlQlSLA~%P>>>Irc&gG>ueTOB) z9WjuPD>~gUos?-+Vbe?;2yiyTfnf}rge%D)Yv(!o4?&)hB@{cmXnrb>G~FpV{uYKQbT9|6>|I|feDD=nrp2r(0P$2F=|9h-CS$ZP{2m8=g7nO z^06?QQ&ziSY6X}ro#}WyLmLpjR|}gR_XNWcpiW$UHRVon(?7}pIE;Jl^{bb7stR0K z$Z?T(5CWy((r-=+K?~wxYG5T7Jf(MH*^P6OOv%o3+BP;x91J5rT#6ayRAU93g4mMl zb#Vm=^G%3!nzc=G)d_WiE|hOlK2>!M5Pr6syGgNbIls~U+_8U?;x*nR9WJq;Km&UT zMgxbzx06%1JbrZXp?*O^r9pFxc?#AEIty<$1ZT1F(!di;iH$yE3E#v7<`T5%*^OZ> zr+gfK8cMO|qwkFo@wtDuC-U6-$s|e9Abl@M@x$uA2x{EQMv!%zfO(L8h}N1Zo|)>W zd?9*Ii^51LoljoR+U>_)XKf?dr6noth?BDX;WbLPXOJLmUiE|LyIOq{$Ht-=VL_XM zbJRAI7RitYo3F0OacB3?@ zosJXhd-sl0#h;FPr$kUT-$uJFd0ftO9SW?^quoiaF5!4Sr7UZY2>pqo43OUq|2$)P zv&8g0TYEipB>eJQEW@Vze0VD6Z5z3C$Wn-=NV8m)3WbH?yH45Mx*{zXT28T$r?kF>Z3J1U>IuxMGkoelfo!v&LM6=AbcQ zQ>!Gq>s;hV$S2|aAZ8rF0`kU$-_k}T%s7+V zlh1&{>GzZ{gtclLrFaj0hAr^$#W}DU+HQ}p>;>vRqlT{_t6wj=oN^@*4W=%9%+C&1I{$HAj zi=083WmV)%cR_T;&U=jF+T)A~_({2G!RDpGY?<5KdU3TEO?_2aA-0M5oE*#VW_aee z8-w*`aZ+0&3#$+8m&srRV;vida<=lRn={%ia$Fw79y~3&KWjDlq|kdE7BRSg_oH?E zUewiUT~FPdPw6mDeB~pj{*j#mvlUDk=at-OcyWXbfTU{eXT0M@eIDr$+IF7b^7C1hLbGkt#Fq6cf{oufMB4IwW2T)VRFd9& zv(=V@%Qff@NC=?a$ z@AZd-`WlJH)Nl}@N?WvChumCg8XaeZl(~Yc>?i)1^*Mq&4-Hj}2P738XDLTUGt|u6 zP)Iu+q9#PGe#zUIJjG`-a&QRdMy_>Y^HkG`z?KW^TYeR>sV z6v_{J|DW8vnyrriKNj^G`uh?Lz7Ih8e=q9YeI2$N!Rn}?_dngdpG-10{&n+urhbZL z=tA9#lZd$a^sk$@$X1#?Ju40(dY={X!Pr?k2^=pTBBN$D7N5kw${Cjij(_i$K36O` zObD^AH_Noc6&;Rsm}Y{J6ajD;gDZMcp^m~DF-}g9m8C^~7^P=vj)Cj(+$&OXW+gE6 zxQwAV3fI2uE0XrOWirSDy|=U|#phSO`6GQzn8&jEw-EDC0YJ4zd5DVHj%Y==Ehvx3Yfs+#L!+D&VbUNcsT z_j;z~HOw`$u;q7#I9~+o@II7h-gS>38n>Nfvefv6IGgkeATv{U{G{Ko?RVc?mx-3oPJ-DyV;#bg_xn)+;1v9T)`G5)X?QBm1)f$oq=Z)aj3jM`E0 zyOC!-%?5z^vNX0xD0KE*@A10yB*E= zK>8v)vYGE)w)htPB15j1wJ}Wos+2wiG0a90Rki+h5;Sn!^Eb^|I2ix@x8r-B(?WxW zA*)2E^4Nof>nT$Ome(ogz!UGycG-*On;a)9Z>y2q9Q@rGzMqIWy?`vS@msRHt3kJG zRjpgzUtN`ZKW*iy^FF9)HYeBFeW~tPK10{&xITN~g=LKfZn>?7njQELaiOADrFI1d zF;gEfr($v;C+xSC`||6N17f?3XgPbqOh);V2=A3R=+rDtrYImb{LkhPMdLq*PDt{{ zIgLCz0K<28GY8j1;?t8UMlHHHz}uIiTNaLIE>D{~+dAB6Z=~jiGkQ@Z&d?6Pb@vL4 zFfI9Eyf6T))hh~bARHA#SNR-@KMF^?Y?6}gRf42_l1wt?3DL4rD`+pG=Ne;@cm0(} z2m)Z-&_)OV2Z3aYnKj2`HNb#HEqNHO09Yc|{UjwDykFuP?#F79ACE%vT?`^E4-SH( z<0XxoVte~ zqxQ1SL@|^jCxE~UzjJ6t{e*v8B4)f#zUBN~5JiRv2IQwrBr!eYzQD_nFob}K`v57< zN+#t9sf7%p;8Mw#=wYIm`8de#eu866^Y$s+MJI2V1k<;n0?`2kkTN1F67lz`*5Pbm z%K@UHRy_V!Q%Gsol@c3|fXL5=U!DHY^1 z3a9(otAs4Hc_4U}nbX-OMqwQfn+qKhbX}je$*4_Mw>0g!RJm(%jPiT5F~XhoHIf`! zAn>vpY6Ppd%4B8A;YS}t(DfpFombf2kFRec-ynN6y~Ef@!W(dqw{jl(M%;+65D|0T zw$^CIJ7T3kgTLFbqW>9*ojUF{-4gTY=1IYg?AEUg&6?n3BUA-WcC#@{yy*RuK~KRt zHVMT-;zd@U+swIVAN)ph0&`dFOQ12L0qtX4Ar%g;I5)GaCCrTS3xj-!G9A#`3$6Uf zSGn^7r3K1IEzm`^| zEQfY!%;`hw2Kh z$A1iLqIH|(^$ z&N6hLdLZmWsg73G1F-lE5OD@M&f#?gZ^$0BZkSi;)5q?yg5EwG#Y(6x%w$p=Yg5ho zX?FI5)Su7%d-76nFklt<0#&JUQv+*b)9R*DWY#V&2V98-iyfxiEj?WpIGku3~k$cUD#@xT8IyOFP;cUDEK_> zyg(dqPR5u8&>wdxu~3h@g@i6c_+_o7P}a}!GTrzSP^hvyR#T9^t4b&296u3dTuTcma}#`+8qMgNpS#1hTME+G~N zhim#J^e)^x6(d)M(A)|~BYeVoUH0wY727wTW+qFyhUYCHs@ZtyZ1vjAl8*!B@A2wEb-&pND<* zXQ;)48BhcF;JUQ~Yp6hYT~T@gv|&RO2Up!2oLfzpG%k1IS0%?~ZxKK*Fxjmq&+Yoq z&EDU+eb}ie&K!u-{?=qXz#F*NqJeD)LG}xjGW1wH@JZ)0;6wJ5xGpdRUaCo= z;p^^#vM|ExO@q5Ejq~%LL#&(2hZXnOWy_Ufr{J5wj9Vw$v`-{QpJE9=QpkK%baS~w z5MpYlYG|hECt7H9YQnsdq%$mFuZm0)kod^^{3%P7oKhXA(==_4LKw9|v6onbMx0)S zX@B64e6<|vbEKVmh@-!P=hpgmO(Tq--m{1S_SMbA!d*=6D2$x}B5xn5bqCcU4pA#0 zJT-=uT?cm-vJSL_6B&6Mxkg!jGHI(1>21}t*ZMNdiQm=ArPr%B9%jdL6t%YKqgxk~ z`@?EK?UMvQzEeS1;flx|ZluxmXHBB;WX4z=et&}2IE=msEQc@y`iKV{J-|E35r z2R)IM7*$9h=aKnUSlqXzK(g18M&I5O2R*d%i8y+H?pNY0v$fqNEb0yViQxB{ zVob114mP0Y6G?x$)Izvi5`7#Nj(ql8?%o!qI#^b23H)>$@f$Rrin z?vWszpUB?-z*>R>sBeU&*Qa49%OQaz;_lOkcG3p#(u%;`g#qa`&gog{=~-w2bq7ZI ztGXhFC`sZ>NtSq;0WZN07;c6bEp8aK#~BSwnJu>&bpn|+;cm@vda=t)F&JtuQ&taC zmMAQ1AU(@34iXLr(L@#0ZA% zAaF8$@!Od?*4mQ-^*OL%Um*)JTPfq&vq`t!wp3i z@D&9rR&xv;Vdoz7aR1y(k9@+6eB$l*f?JeNWbn|V)(*xU+Cm&Itn1+f~A4;8F@k)EQ>Mq^(;dmP(0~RLWp; zUsdXGaX^`Ap7a@{FUo%X|Ob+G5GJ_dO=<9HNkgmWUP1?jTDUy z5})ySbQ@(dahS;(r865ptjFDxHYo}3;(6kxCzCXr zW#Y&RR+#I){GLo~E7bUzti|!OtJ4I`S*OK)0%n-m;=SJD^U&f))*2wx8l>AA;?)|K z*&5N&8nxaU^UxYc)|MdDmZaO3;?h+1`k}p+tfOA2qfxh`*{h>9v!lJEqjSBZ`=O(ktg~OJbI^+nYZfxU3i;O7xi;3h zeB8Mv0NRqJ)Aa)sl%N}95QVs|_4Vw{itH^!X4kjlE{wRG%5b`w_0E%t&a;kgc~Q{q zWkxnSoda|C&3g6^jPAdgJwG}i^{W}b*J;lxIxjN2S6937;FX6$bk`4E>q6PzCweev zvk5r*7A^Y-wtLD?YIx?mDbKp8-eg~PR7=zg-FEaaO!nMX^Z^_FqK>@~?@naQ{*%mJ z5GEbHeBavVp2^6b-OqiG6@7JWy|46owU60E(ak)86QPDYc1 z=R*DJG(!YRkRKHg!$)jQ?;#!e5g7z{U}PQOl7@1(begb^O?nNxOwtNW4#+D()h3~S z-hEnn*-r9bC4cqJu4Z#h4t}tvCugP2&FWbp8)wcM%2*rcoF6jX74+kBl(>b*a zXUb3aSx+Cy_u={t7gu&(Y>fUSpSgT917QXE$kH!P(tr&66lNjyXEULbJ>wM+uZMnj z@7{HfZ1m`<)z34d5p+P_RNk98+{xkk&goQ4IvG)5W_1p;7DBE7qQsh)gU?ev%~9*m zt8JIbsc3@?kvZ0&9OqC@1jRSASv2l1-vpw+ac-jN&!T~WN@9$;fFpX~GRHK4&I6bG z)u-3*68YUZa{vWE`o9v4bGv2fY**R69Ty_97T(n_nELU4I9HAM#0U~y>|k!-bzJ;u zyr6hCDN(<`?iBLARoFv(@v@E1s*6K6o6dl3Ns%J?$(T`vt=vMN)+mnmG=Qw+l69k zIJ2G9OXkT7eM?qnx6oOfCuXM0_@pyF`WDMJ(<;34BKeD_OZ0gr3h6IdQN0y%-mK=m zRk*QP9K`<*c?c6=7X`Qh^S{F-;{UPMH>8e*{6E4a2S=uV#83bK!li+l;+Zl`J!&GX z+7j*TT1YfiqK1(!_|r;yn&Jzld|#L2c*}PRv@301#CLGT`aeEztOWUgi=Qx77~a`c z`ZY!Ue+ZYt62JTrKUJIi^`5e3EW+NNtRrw9-q22=iAO`?LHS&5O_c!vvCII=M;>keP z{0TJV*8FjzUl@4+sLF7f~06ChiWMC+J>8?o_cVnM?5c8YbSzm@DermmK;^6gf zXlNLlFiFh0HNTg*F7)9!IlGIY%98yUu`1CHIpQg_NjbK)S6WY_jJ!Xt#_5v0r|}mn zN;FBfF8#?EO1<1iuPiv=yr@7*de=n$g|NZZY!( zv$g%{^-iu)^vl9eLg`m|$(-KvjO5SK>kB`*Fe9Gim00Mk7Aia->kG2pDrgsHgL!ur z#r--PbrRrcx7m5EXWJhO`uvIx3N%p5wJI8N8}rLM39vXab9e=d;UNwSo7tEO?uQ@a z-t6y`mqmK$=4W%V?WN?-7nWL&qk5D!OphooR4mZATV;pyFKjx)_V$xowoQ`DIx495 zZ7UB^$(g#Ul|7ETP4g5_tphmcx!SAO8~2?DY^>)aA*P;3os3o6c0b-RQkpko)DG!< zQ)gUc8syKvs0`KZyE^VoLC9%;X~y+;98G`szHs2n*--UI0Hj(Igkc1C3ex`E!Uter z|H=?IeXTr?#nfzvHqTZ4$LIYWcCl!ke=oCO0Wxy7a=b+9U#>9TDZf~Lh~VWmj}2*h zVOFsWDR>4?sGz3tOx6iA-f?{WS9x&r6L&L=6(I?B;VZ;CxwZ~Dbf zY=orcc7pp9}o&ReY5x^$tBEy$zGovs*-#wDf_oXh#b1)Ijz5+*~P)7duFKypR4 z_Vo2>()jP`RO)G_+Z`XlMirX9#iLdKvEUcDP&tGGKWEJR{?CGUDRh}csA%mwD0K! zn#6`iA0w~2u(YlOHVf^Z4YC>T4P1tyxZ8bZX*RY%qLNC`x31D4Gcq?BXIqXj#<(}s zeQ-@Lr7kdzLtwt&N#<%_&EDgE(elJb=Ip$ve>di%iEV01*4J~DX6HxzBLh68MWSn` z8alVH5B-@)#*nV+z`to~;s9x3*~Sm}NR(rYiy_3w8YJTIHbUYVQ5*|9DN2RDLA_+(1Y2f0 zPG5{6{-HDOKPYLX_5cjlDv&v|1ZjpgDyQn_GQBv$0{iM|)vHm-4}hFXIjNzz%X ztS3j;i3V3j_J3ERvVtk`nx)YuS!p&`Qw7mSYN9LHxcc}2;BpfpQ`IQcdfJ6U?D(nPrfm-{?g z)Y0DKu;vReH=7(Z`i+W8h8M}r&B(rnDRtFJfD7XG0U%>$v{;yZok4uSIy>y@2ZM}# zuzZORvl)Q;wA>fZ^W>4(-74OT5uQkwk`0=TSv=>fPiD}WOby|KzC-$MW~kkonGzY9 zO1fRFFMc?ardy|Go&z_44{><9m!8^QB2JquHtg-s%Qmm#BG!p^0T>~1OtiYlR@vVR zj=mgmDSejylN}9oXWS!}QIYXQ`~xN7Jeq&0eU^=qsWj3T3@i;zILd_bg#YYM9I4KG zk@oSGUABcz6uZ=qXCDht3bBDJlq|yGJ&mTDDgTEqlRO2|9E|W+!0l|WU#%JI=bM>! zTp9X4-&7AjcRFLXU_x+X^lFLrBz8se=YiESZc64M^-Aa3%B37$EsU`ju1@v(BDPw{ zRT2(}iqIey2|(RHX;S%|Tp;x)y82Z=No$>d@xf1?u$L?c-Vi`$&BpaTD`H!vzQMeO z{D>XH0(u)8ofK})rrW$ucIng|aJa=YbFhMu71PJ2`}#=SgN_<$@`y>H1H+OMbKR2l zYt19>-DIZZ9{-+~o&8=%j@dEOD?C;26wCO>IDc`0@`g@p`q47=R7 zSQGB4HOk1|NIAokk)AlVaWne0z*Z|8e0Cb}8?lnrt8I9NETJ>qf3h}jYH=wv?=^gZ zf!;}jESIIvSiGoF{Ir0e8&APnhQ0)plP>twX@Fg8sRqbl4~bFXo~-t-=&Afv{NeZ;7hcJPWztr zeoc%w@6ZbV(G!5{C-#}wgV^5xx!Uu3GJ6Ds^%=USm=U@dRFcl|Ls?)zNU%z<1XHOi z%M{cyFk6eUu?;*snx8 zrvh@GtMGhu9*?@PB*yR}SM#oE{8&2wCz}4Q$N(ts4B-e_#4|Zem06F;S--^(L9n$* zlTOb~tc|%}CJ5e?-VutOZ_x@rb zWGeG0kHh<11=VQlYZ>qQf_(n^o%k4{dF{i22ysZzHj|2b)d^0Ji8p`g2WO6^YDi$z z4)f21u_k%cMKgvIzydLXyJzw79Aca*!$rqp*czfIm_0QbB68eJwG%bNQ6fXHm4`JW zm9Kq2kYNP7Ss2hIqH89HlUqegBl|tp$e+Sj(^6mnMt#C_3v`B)IPEN;N7~)T zFGaEwLl+U%*YHtJ+`5PHBRqs&PdvJb+}#EtI3z$yjLMIl%%?HIhkcWQ|veQk$)Q-DN>7sk(n;&lq+LjS;r!c;Nkx>P>a z+Ggt%J-HzcOD>9NEDEwj$Mgv)-^;HMEUwfku9zo}=*pgqDsIHg`zi>nvjaK>OMd8- z^thLF&I5)%S|~0Pm9si^3r+o+V@CKew9+@0KbyUnxx=;H)VhafQA6RBk&pIh>qOzL|LJu z$>nWDYFibh$3?(9X6P6NxjQ{UvrY+~5imVp@tCr_TL(0VsbET6wDuNCJMhMTuN==F zs}-8lYYEWbmdPSBO2*^x(S?Z4Zb~_B@u5Oc2{tGNUv+ELmP-SEJrm zcZXKZbyMkRTccN08b+0RJge$*2N|-}^kCMai`P(g&=^>f$yin6cGT$UQuScg^zBsu zgMgvkyQ`q*_fZy6GElvuw>V=Y&#VBMRX}+aP*JVwng6KQIfFA zmF((R*y22eWLs_<@uEol!b)D~1}JCL_s`c;RaAXbERo-@2B|jnc>pO*uAq)cGK^+h zI_PIBfXEK74OtGtH3Fe1)`l0iaQl{k%l+C^1CBG`j<;pb9`8Jg>vpjS_^p5&#a1)6 z${mk3)`}-1Y)#5mKzvy9)P4g%MbH)2TxV5M($vn5Qm#5!trFJMxKQW8)E3go~=2wz-Rj zr;|vjlUb=`syMC((ykWEYAjUCrBp7dM<9k>j)wq&G63VMZk@&M454n!sK+=6kf~me zl2OkGrJncAJr?2hL??Ad;rMFL0mD_Wp%K{0v)3)W*LkSdeX-Zg6Abz@xQAX)_WIsJ z(~v1u;E+C+3VoFNK$vj6$Ix&I{tyqfb;>Z^4Rls-aLpQD5OYvB;{{Iq2<`f?Z{Y|I8r_9B^_SH_ zC^++RW!8}=-vKt=^^noY(8&G4=vMe33fusBFnpzl*xxE`EkQhSK>SFJz8uQB52sr` zK+vO(wS|*or12qrq)|6VNYaL%oZ+=W5folSbn9cNzwj_@5Rdc+=}nwI4TSw+qrARbqk`(hXDDrSyD%6Ym{XgrhuWwz zD)~e+a0Ypw*cVWRuo&-=#nnY1KrGV6+AGHRhrIH|#PzPM$y zl!L}?eCS-lXP{-SUU{)eAk!{nJ@zp`S|wzw}oN z^xB?1VOwI1u;z7cNvJsISmx|pu7F!CqpnWLu-2Z<%_y%m4U8C1=s>2s(esRGb=GMc z=MM#EIUm-!X*PIOC|X`_@Oy6vq;CkeZU`@Lh(2tH(`-tLZb}<$%6M_NM6+!ox@~5#ZQ;HBA$@y>em0EoKVcWi zAImuu$VTz|SJ=hzzgf-&?9@Beh*|zO%lY47m+jLh|AbxIuip8+gfsuga^9EeC5pzfX(5iA^dv&&KbCVCF~)x^=h|~_dbnoU5jue^ zjBkk||F)dRqmEVoZ8`rGC;zwQTzG4n@w?C7Pi?=i(NB#tf+OYEzJquKVc8)v17=A9 z1l8ZgRVBQ8O)li$E1Ou;MItB)Gk#dfIXyUO!DyYzS0K z43wGODPWSP!YVPydl|zHDhHx@*{tLolB2}aIva~MY60$&%3-wq!}3+{{llu6Gq6ux zH*U&N<$}A1b<^HR@t0MD!F{w%{axt7ahn+|?32NcEow~6n`RO+qch-HjVskzl zm~YNsW%Y#WdQ{2^%S9^T1x?JjQ_$gcdAODB^rX-2;nC02^x~RbLpON+PG*(p^hWf| z(dT(WA5eVs2NCuFW5w7309NClqP%N8ztaz#E z#ggOt`k)Qm}Tf(3=VvZ1D-+eZ8Q}ne@lHzD-KJq&^kf zXc!|bo1uojUoLq(GUm?ocHX}b9&yItLfglol+-6O?+_8@xux#@bwK^*EXthwE2qmx z-`5Y^5w-|p4zUm!HQaNST+dA|Z~H;rn=zHqO$>XBhF;S8b{yOKZn1~c!DoWcQk43$ zi1h8kUd(Ph{#IfE9W+99L^Bs%6sCs;xgx_1>rGO*5ShX$FJ(P&b7L*8xlJDu{~XgMu5u zYdOEkgpMEEN~V>=K{vlZ)Jugb{1nP;xX&|w6BQ|+%)NP@4g|kQX2jyP(}F8G+H70VBUM)SvjVYtdri2WRHVQtw85VACFaG{557 z!wo=;F+yoj2`NOZ==Po+*An^-)8-{t7O8+(6_FOT)mKBE?JYyXUHq_P1zqZs z_$r;B3O)AsuBm(K-jV%;@`&u6nhS-Fw8{x{T6w0CB?k~vZ zUg84{Y{`&o#+TWzOLrS}H#?)5?6i}y+>GE>yHAB4#zZDeN>7BJlCo)yPVzf0^0Dv4 zIn+tL8Fd7tDaKck56CQ`@5f$TDEkte#53|=E~-;$7#6?iUx27R-oI$lC;2ho>o#mv zzoQF5l9=VdT$a{do>o~d5V)e9dP<#oGf4=3*>UxL+1(t#RR%t z$+%o{zd1g^eNy+KQS(!4 zzKML0(NutVcY>#Mkf@b9dy4pTLKAi-eiH0pcM2ELsTVIl#(ZYu$0v1tu}tb+F9VU{ zeiM3vce(l{@zU*Q-E)p$Kek;WdpQC_h!D#MUok>o_uwx-*1f`~zfjkIt-bv6M9%M# ztEFXKkWao~pSkNe@`CuC0Ku5Xi#+uZOIhEcfS~MN9^)&oC>e+%_nR16Z`xhA0A@0s z%Rq}Ay&|7#;AAWdSJ!wdfaU;N=m&pSK)2l6>ZpYR;0 zQ}jB}jPJOMNSPrD!pG5!lM6yoq(Z|31Lv7twu8I_PMJy}{O3#*8}$M%y8$rMkjP8U zdM$5h?l&#d5&o_&L1Z4DpT97kTNiEUf4671LGwK*2!8Ds#fdEhi5NkcG$lPfN044#jn2PiW~Tx(#rSwcSD}{ z28+I&@r2PCL6FkyyysER`eL9`K49ONUz$;`rM|s+AA3O(6>J~b6%(1$5qm%y^pVNe z@;&>;Mda7SxZiSQ=<+^MQZW2kTmK-}Bpxoe7qAZ@?1>-zaid_;Ed1C%BP?e><&%*W zUd7Y?hD|BiYxp=lY53dEfCUF~oNVL@Y=tNFt(34PtK|%g8S$`+kpg?r*iPY1{OsbnqY8}Zc z$@Cz3m*@bCLI^1U#zzpxhg3Kj%nVoLxJMRiL* zq{d@Rdn}I{(%RLWw>LQoI)$;xIq{*nH@r|sQncaWTw{A0bijgT~?rS0ZKOo3*9Lc$Z;s{2iNB+{L%b`|Ra zRu+U=AgWm4vsdhiS+2lCDeh3ssH;`spHDwlNH0;I7*&LITi$h3{9~^Se-Nji-;S{- z-M89=tuuQ{;)R@*RyUm@W@f288&t8FtYVU~)~D=7umDf721!x=`(sLsZ}od!EwArDrXa zwJKEC;~W22D-*s|xqv|+x^-p6*4DI5Zo1HKVN^_3l~K2~@?n)=BueWCEc|7E{cY@U@Y-*e`_!@&%U15-i6FfL&#RAY|I2bh+@VFrb3f61SU2ED~ z$VXs-j&z%y!kVcv(^=lt9fKgjh_)8Y@*!vFSZ5a2G&}KJ(U~_!)<-4g-u2Ze+!0I?j0i9zJK$`UjJ!(`v z+e&roTavq|@AS}7&9oK`6DKLvUP3*63k4XjPIqRMrPTa;;6ahzI?I2YcgM+fUnjW>d2bBMn>%PAqdkq&wd#=KFhi7v? z^Kb_o1BG=6koo-tz%gFSt0H|kv!DCiY-t>P0$^w45W^-D)22^^*}@O;q=%bKa-Tb% zkKu6V;`q#6vrjgh;k5xkstrO0XU@k90Kw#meHN0GKiuBV-v1u8S4P3A`4j8jq)vT+ z&(*BYx7y(4Y|rf?!Shq_QSb@o{qx*1JCp|fF%|f9_4s_+;k3s6s4VvqTE@7`;h5Xf zRKxnxW!%CVd_3D`dirp(jvzZtWNJNN@!@`~5x!E_xv~yk?rK3OP){hftR}$ct2&p@ z7iR!q!67#vGja~+Xtj}Garkho0KV#OvtC7D{N!PR;b`66dHnnRv^jiv%4?0(dycGh zoWhn(s^}x{qxFx({j^>y^)>)nPUe&Q%_W<08BR@}hl!$yG0w;VMGvfZBdfX`2&2+5 zbdfC$hJ5{_Mf7skM{1@l^V&)>Q5H_@Nb20vqPf5JF_COv@PwQI38uoF45XNrd9c(bcRftcGt zbbGN973fqLuy)R6ttw_4T4X1xp1Sc(NEp+)+`4IbpV$n=RMnL9`5uwuK`!n~T?A&j z@y7n|!s`7F5=*s(A^i^ArhHu}N9tie!)H*wgtma=Fr-%yiMMt*V$KBTul`rT^u zOR=XmYtS%2rrj??~6`4g%(br*`DL(#M2hl+)*6FQXBGZlT`SQBUk5y|88d!(%> z1=mUB!70A;NdzVS)3y^fq!u$C!A>a;>T4s?i{twv?v01@kH1f6UY#h3>B-6Hq0OG3 zpcXN{o>IRuRsDWGO^H7i_v`V?b3yXo{44dK(e5fGjtP2vjg=z|;F!YW0xt?U)4)uR z;l8iL$bOZJ8$X1NqGx%el_Pm&>75n)JzM2iEvzgHr5#XO&4r0wD~Vsn8(t^kT@$H- zVO%IS?i=J$*4AA3dQd=797RzatfY#fqzZ;3Z%U7EO1N&zMsHw-C>XHc`%Yq=F4wS6 zJ8fTfI=}67mF<)qU*qRJw)HfceTAk60~DdRHKQP9?mO)E8)Ccr;m(VC;AYR*Piwa??{h( zZ0rAGw&NuuD3s*N*CdM|aLgRLV_fzBV79|kQLi?y)opZYm7E*(*JBU@xJf_UZV*f6l*Xv7Z#rE53KJWMs!` z!o_bYU%z{wOXw47oeR;#a@-=$UN2#PuEnnGRP;vY+pG926tqO~R)*v>fcrg5V(t8g zLYsJ0)~8z0^yo3*1PwMcNJhBIkU$-l-E@{S3#Krmbk9$Y;GFk(~p>4iF~`y7qT1~$Q|Wm6}=U6mEvmM0`vT-77^2_X9c zw{0^eN+wc?SJAcWOrY_M$-rm&8k3(8uDjR9v~zsad^+9O9mujHED7L#NwFE5VHEbM zSF3IKJT*~8bv{VAJijpY{xzOZ0@iC!`|i)Z_jbKczNI>^6Mc*6Qt+VpGG*Ec%e$>aLBJBwKMeCBUt#26HWzGqZ0LAgS@R|%#~ayK17}Za zpHi~_K~vP`#BpLn{=PmB2H2qh6o~%2^uhX{rH|m^b6Cue|LywxKcx@&Xx8tag>$p3 z)!z%}y|FCuZ0Ub3oPYLmsQizG^NT2g9LS&2N2#CWZP=gEhvh|76q_dPLO{;yKc&xx zhHycxElN!`Cd`kbzG9=w1&5cBYPDi(P#x>>vQiZ1PSXdZdhMAp~k8q zTjO&w766b>)PG*Mdk655Pqlx+!c+0AKbdcgVf*yaLw#>$^reW4=t6_#-e-B6A}^0; z5*348VaQJc>Z=?p-4m3WhC;_zmnYA%S~(YQEN`xkCu*tN+K_>}aHb6R4!X+~UkHI& zienKEWXWmbuQ$?vA=5%r;)tQZ*w*Ej%&HG<(e!fQ)6Vyl{@49+CSJeoHP^$|z*lSG zy(qv&L_go9o+XZq0G&Uf#w@ii!TN_y!=14Lvj}pUpJt}#Ki_{3eN*Y07WUdABqyBC znI$ViCva}YZ|sna4mKyDmFeNa$FiNGwKPtbpji2yF4Z;N$t3w*X(D~R8`{7Wm%RJO z-2^|z`MAjTxVa~xW&*pZFu5(_q^v~DDfLVz=a(jV87jmKaq81A(h4=tUgVgdI+5-b z3)v&x3-XOREv?>|iQdqDkB)E1lOWP|-!JDLo@0slPEcH;lAVspQaEztvQYG>q4A)G z@pPLqu9xCHZS897udv9fILbo*o%pa)N5aT)1M3isSWSb~04pxpI*@8=!=23Dr?%AC zsFIXR5^Mqhg|@`56PxS6t_x=-(5Hjwo8w^j(c;+%8>-DBPcPW|PNk1FjH;63DE7_i z0NqlweNSni6= zr9q~+``u3JNA6oHR;4!wARLVr592aBe*w?h*PXQN_1=bT^GWOKjOZ1*-j zd*HfY#uGZfC?@;$aM7yUeEj*mOZ#V0Pt+k8`r584WirR*o1to%!;aPbBg(qgM70xl z4FB{2pE`WN_C)Sk#jBu!8A0qu1TmwqalP|Q4Rjv%3F{#Ek$FL`?ri6$Z7dTkWD=Sb zjUb-?zM{#lj&Z%;4@7X%8)Fo$T}h-wX|Sr<++jLXBgmm55X(AR%{j4K$wt9XQrdK> zTuUaDB`Q!(Z~d4=wv(w=O^Ni`YvRi*LH^}a13T zBY(0QLkas_6z|(Wsz&UDwKtE1YPz^)9%;aN(Y%^23&$ACp(`gL(H8 zlyYEO{I5*n_

%+pF!ca`!5XZlB+8_&~Dly-90=idPw zW}R&ogdfGP42X!l8^Hh~Q1DkKv_>H)wGt7m^DB#NxIl(jVJyOZGQIh_Ku&f(G4!8> zvn4@H<*!^DY!;>W3KMzt)p-KLg(@zgh?KKm`ED^Re-_TsSJed)*A~VnKc)S-i3_#W zlx5Q~rzGALzjZWL{}4Yu`$!4d>2>|5>XPdsi0M;Q73j<~zc(Ee{adFL3ESt}wZw6DOi8|g>@$Ca z!(R_uQerGhs)o*sh7uENx&u};ZaMZgrheJbh6SY z1mLr05cLW$Qyy{qh)@|#&Yt#g$!CT%nVZty=5*vmlvRgg6Jb8S`hh{i9b*{1LA^g9 zdVqNvYwR!)Aez);7=0H1Zhcv74){G z;H40Bz&V*AMWln=pnUIL+Zve3wWHYdevhsMz;nWz=Yb0u>EG% z!qvAT0e)cdY+?4jfG_z_JJ@aOY{p)Kr}&%AzLv}F><_fM`Vbqx6=TZT96QbW8IdFB zs#_3~-B&EK-8{Q=N`}Rey1L&H`(C#TINKNJEz0yIE={+yJr*vVhN^+Q^#dukYt`aU zRmQz_7gyIFF6vf*6WorpJqw^iy&RVyESZOJ~q>ZueN zXzue4N)Sbej5D ze&d!vl4ISOUIV9**FINsTo2q`fnM9BDwi{2-si1MAv?-kpTqL+ zXR9nc&Lk>tN{&S@3cicZUfOzjzi+;rM?#PrSa#R@ZTjbHc>QOFDpnNnj0>A{6vV!kJ+e)7A)$QQZ0ME~5m9{N-OkRtP!Y6nU61xl9(er*TYT?N`gf~2@X=|&_^q%kp|0D~%; z1p;iLipGcl0* z2Kob`c@Y2H0{;hP}&~BbnP6jo7#$l}6Zj1=y?|{GlMQRUp(K7AR%+*E57m zu40rY5|LDd$8S=FN~lf{MS$%TA4QRo^+_3{nRj)7>3Jc-0--<1B3G{>6c89y`8tKK z0?lT1mjxo%uOdGdfFeO8t2~(20N14)n#7LIX(yU*E`#G10m>Cr%9+`emOd*HmLmr$ zboIwc3RNr*eJ^1TdFL~~rd!)*O4H*SH6GSY7AXyl?1F@gGlTpsNTw5^tM(DDozX&S zFa-w~bG-la>j+$oZ(_)L(DtrJCvPBbZDh`Bl;m!p{cfa;JczM@G$LM=`zM304@{aY zbfx~QVmTx$*k3w0^es!&hqX}LDRSCv)5jWNNGHxl*T5&&ftId#)+{0l%;9Zh{)|cC z3id4FPxRSl-mgn5D>pm^RAXeWaF5#|<4-`MuCY=DG5s?!@)t48eu0|WvEhLKu^rwg z`EY5LgmVa~8@b+4ER&hBCH8hmZzfEu0c8CWL;flf+8<{w5p@r;U#un{mW&sSk7^`~ z{n(HqodKD62IjwxI{FaJ*{=$EMbOLgI0qvx>iX-%L~3whqU+4laP7dzBu3cFNdMJ% zm*AL>iLs}~zBb6G`d;Mpyw?fZA#sS1INXM~N9L;886=hRk|u6prq{*|*9r7VF~NoL zE83Wj7-Zcc9FLNrTOsVl4X~ec-xca(3Yn-3rVkheU_B| zW-fj83ik46@-J6a8GMQ=dHfo;?-xI*u9M=i_tNCtqIu#|$?P-IqcfgUq~RvJn=mtQ zOq&!$r->+J%adn#Rb)19vI-{CyuQvnco~YAONCx%@8mhyZ_^xgm}qJ+*19q9y64zk zr=_*$`p|uUl$i%M&K~Z{G3|ir?`7^bq(4iJ+k!Z83P9s?xL?j{W2jSNtmYHD=N}h< zIGD9_YYVcF4$w!zTBsHIY_5J$g3xb8_#u{B9`1x8jd~={pbjQ9+jLA~jy$=ngrpl% zg@QVHjhY4Sc$hGdLIwzv7KBv}!ZIpWfECvb6gOEG*B2GnGy+eaKrd;FG8!ep_Sy=W zFTdQB^Fw8Gp_QNE8OiWI!B^Rs_)|5St2gjaKz?DJ&i*ZKI#8zu8}l!T!=T9Vv_6Z``1eCR19 z&yW&D>Q4z@>e2}6R-lemiM|CUCkUmiB;4Ls;|)?mu__k!h`pKEJoxDywKHVY-&3nq zEryZ%I@cB6ZfS^qdwyBYpFQEys6(UUZH+;8yOlP-SRw zwRdlQ3|q6pK8XujO*vZ)WoAxWZH<0Pl+mrR)%d#x>9ntE(1LvxxGr?TilmIP1`k=^ z@UpRT`eo)K9cfpLU{Gv}>kAzk)FyCr!@8AUn&M+^^R|a>ip-{W^JA>fgxYxR8=Na+ z&-bC#imiLQMmtsP4X8`b!YqSq-)JLS`0q76zse$-|gv~UUgVfwmVX^(qhv> zJe$t=IDQkbFifmUGRxdQ{IduITH#~@@o)Cg@u($J}0Zo+Ne0tPK6APfq z;{J?nD8<{KfYwr-O3j)|ovKaQvl5ZqeIV2YUc;(|Vz&d3cJ3Y%y&mI#^T@HW9csFt z3G1{2Jx0PkpQ(DC^?DtwdmX8;eN};P?uGFK-J(AzP4)T$J^O-F`vQZ*4PJpjPPPL z@sM_7-}Vp#i-db6BZ1cA`_>51+Tc?7KxoY%n)3oj7Ahoab(U^&%4dO;YHI?;VMqe(prmMD+Y zndO9gIjidXyo?KPndR#Tlz_Qt>bY2vxtJv&PI)d-1Vv$DG_+3DUJepYJD)QQ$`yfR z+stQ+Bo!*p2egCUwS$O(g-YdxDzAl_h=oRu1x4;j^pk~l>cvix#UJ+zuq8m*XK`Q{ zl(@7wbiX)4y`&_LlIArnVhnVrEzR33Rz=J$A@vuQ?-y4@me*TA?l?|v&lmB^mkv0V zdU2NehZj$&m(E3&%2bhUwcTv`7qrGLlOL^rOXO7rs?W|=@SsdkZ7=8tJ?P^&^p;^3#FCki1S0d)bIX8HY zR-vskAm8;ry&B?AwMOg>^ir?Hi!8luU!zh%>17r2@~~ArTEP+L$K_rZN?+o^-6WDm z(NI}_mcFH`xrkfNH`xWSwl2LydILvGYi-=%hD83el*{YLY%HgeATVXbgy(?ta!od8U| zx5`Dpnaev~DjRFEi?T+4y>Q!Y|GG*+)Ve!DvnQ>x856nLdF#||1JtQXkP@xFj@%ic z-mKCGud1MR9_{3(A9QSgM^yP&*PXM0Q`xSV?`JI>t zb=F3g@-cDgVkPbhdHN!W;bmV7Kw|(T$NmyC1hmVRUa1~om#+mp{u%Um2@B@=;~n7e z`WMmb%UV$sb#4@}_;q3%D5>o_uIxI^5O^z&^0xeHGTA-(+27?{VcQLS^ri^!w)F8W z{P~S2NkFrO9Y{21ZFgUCq5-Y?4DHVp1!ks zKL6W9J^3VZ_I}~`-g1KzXvZ|Q>!>$-AjPKKEqgffyPthLBp$AJM_4`A2F?F0Sfhrv z_2uDL2NIC5AWTw2?T7KoVfw&z!Biw4u2lbvTlK${KFuHY>6l`1QE2_$R4tT}pV^?+ z=T^3-#fx}7{Owj%j!X`IORpHDgeBkwSvBua&Vg!dJw+5vm%J;B-fKKbQ2WQNI*-a_ zGwpw?!8oG(e>i)~uQ>NVOS=jRw<;jGLvVLZfZ(pd-6goYySpa1ySuwXa0?I|0tANu zHGR(SbkFomubyX~weEl6e(}9NdtV!}_V>2xc9Y5bB=&acisedn1RRN+ggVWhzqeKK z3uj7&wnl~-amfP&P~AmSi2mAEl>lb%Q@JvH9?;ckyqBSFX6h~eds|iDm)Ugp&jlvo zwHmv<;aCCcU=>!taCkOhSX{<$$J3?i!s7E}ck>Hp0iJzNy%&#bq0}t3EmyA9m2;E9 zulo9LKF^k|e@1%S?eymEUZJ?P+~Eq8R_!YV-pp*Zd$4sP2+@V8^tT;_a{^&@NS7jL zJU*<=={+hKL-(R7u_EY-V)Z?L=wD(#Muf?-N#KuT&NYDSU6@CM<#OsJNu!hMir(Fv z#d6HDzpzQaCsaLtk7=ZHIbLmE=_t)m+hrq#W4~wrk+B!#T(*zAT2u|&tKz^lScM$W z!ulh`nU*QCYFrA}D278(h^wK3Nd>+FdG2E~rrOYI*(L{6SzIW~S@xB}$x&IF<_iO` z98jlLA;;F;^j!h=cu`(PQg2aJRZ)Ct@lwi6ef9u8cA7qfcTMK=T9q~BGRUQ(LERvo zC|MXrI_g%#B$e?XUq_nGqz@SCd!KKFMZ1%J9yg4mL)Cb|J(ype7QlqWZI>1k|OuSPCQtm-RviU8=v_Ybz1x3oLmdvQJz|u zhFPvZJpVEN>$?ui!gvXOtKxL?4y)2{Vf@zRr6nE2)|Is*{5I9ChaEPxz3_jyzcO6& z**4DNWE~g4!u(Sc!2~b@0q+5UK%2iW2Jio?nh0$m)_*D7{eN_#yjnW`uNQ;=z9u3r zYyPjA$V^+gp;2Hc6A8%_ulqXiPcKp{ z14BPyP&Ie_3Em(7j2FIdGD#4_vD#0RGUhb>S4~8S`Cl~=vv3vt@`E&8w>5(_{V=-23{wRJ z^S^5%S+*@F|I|cWmaYD&iTK=~9DNIbr$5dM!LdHhkD#tNE{NeiEmMlbzdv?#M>Lfc zu~6VNEkO<(r}Hl+cfqr)LlQd`iI;z#XA%f|hJ1nVhsrFnU~U?tRAX*xGl&Q@@GI43 z+tK$ES8a>hR;Qx7l-H+gS1^hWz-z+OnA(zy7%fk@G0{c!OUT&5DR*iXq46J_eENnS zQVBY%IonCdzrWM8x=m8P?7pAjUTWxlHkvH%UEPZY^}I~h^noY^>qtR7x(@X{I65{% z*t5r#!fEn?b3)MS>^@=KFh&soexfaI^$Cf~MF17-&h;egDC5l(t8*w2o_sT=afa{5 z=FJRGa9`sLKAG^%yi}*)jvXN8^E!1=Vn0%=LOiwIqN?*AyA|v&Y6hjBGJV6%rtv|2 zGkm47_v_ZJv2NHlwnmqmi)j+Swp@D0LxlRjR= z*pJNZKNB>*2GWLG#I+qdGmwzAg?topx7>@f>o`_?U{dOWJz(fx5ubN++mBmO`17kp z0R7)y7WvFsm)+1eZy0&Xv{U~O=bop-ucV+qDy^Rsp7s-Ee|W$b($>Bn3WMzozbl1P z_blyydcAE`u>a+{BB{0SSKj_J&U;4m`z7z@(YJ5iPldloxfcJNZO|4_=j}7uduQ6{ zf8N(7Hg50V^2UrxaG^>U5h1n#OBJ&rwX&*S?kOD4alwW?}yLvWrf32V>p-;zMTdAh3 zQ9)5oC}Q}cA;EdE5M?b`cxw1w?#YKeZK^Tva|Ld{-^(CrZ^Qt5ZMu-HlU>?1(li&B zw3M4)ormUb8fXMD(ByfL)U)c_PklV-pxg6JSU6Fte@(j>sWk_K$e`#?)+;IXh6S(X zPZW$OS<}q-lE^JYvDn`D(+pvXzQveQJ0aJ@L24?NA8tl@QZp5|s{8F`8A(|x4=0)6 zBk~T1aQVMcd<;Fc&GBNS5jXQxf*LX;WVl#Ri#U&!xF~<6(c7gO^-yWccPtM+C#Gq> z<9XY0DJx(+Rmbd58hvU?^^~K~jI>sMTf9!wNT%S-U6TbjT-q-!qOpxHtK_lQWO5YS zNM^1#l_MQj^8L!uJXu{Cxp1uJ*E=VzG@C4{fK-Ns9I59atKq0Dg{G@jSi)LY1b)V8 zI7(rpD;?Etf>E!ItTxnn2%gu0+NqT|FC7RRL}qm26c>T4stnu_RO3fnVl`oRZL2-k zSLxl_1>f1plAdkmc1Y$*z5Df%Lsqo3Z-Sx_AZ#L3IeV?`>XJ5ndeV|5RkUB}-Q791 zf~`cVaC}65ua)cPEQ(Ap3;060cBE9zH22J4_GpCYjZncxL0g=)7>skiCp zJ8Jou4IBCXdT`}uOI(oSgkTy(n9`DDM#;DA;go8UQT@q~H%9#_48LGGG|2{Vnfgim zC&MG~1OOWW`~%B81r+0mq_P&vRi4E;wQd zp{StsiG+J*+{kb(_qR(?IXtlv9-$C<8-l6aH9#`&mjDB)o2m{DaL&Kts^2!_JafaB-VeG<6sEHbHY@w zhXbEJM4RReVP-rI4K8m~B+D1|@2~$!N}^XPXyF7rUZ)Mln5AatPRbmIDO4Qwn z1?_MplaLt^VPu#|Fy0I69e}(cu7UIGk-~e55 zSa%F0nRm~=a^HaS02q2)FI$P$fz&-h7fL;dsj*aVL=2^D+^=BNX&nS;JXx-s9qRo^ zb}dwZ0Ni^o29MA`N0CrgBu-x6K&gmM8nD0-u0w z8Q5tN#oOwvj_2ug6t$lokFVlio6c)e8jVHkI}i~(MC<2t*ZVAmS^<{R9yOOdoT@CCVNwy}Bm9&-mm4!xu@s)@psA8NVWG^%|nP9-))TYrk}qxohdB;}O4s zU~Hfn=H2daO^cV@&Yv3jE&yLvpD*2sFAX*l3nn5BVyudQxs@MW3wkTd!H< zpIMOeK*$Lolspj1-;7>EyI7~9Skt;#^R!s^8U2DCtK}M@^~%A{5SiK#-p&yI%X10y zPKh~dsgq2pn>G9wVPsA(2*4171E#1Yfl&aHl|!W@Jc{nTpzIgaxJYQCaE1cR;0zIH zD2zic&p3T2Pr#bfQT}zkJg=kto(y>5fZ^>?QIb_r)=|;wP*H7NL3e`*RD##NvZ0=X zGoOQN)rRj_hwIXY=~=JrJ+0(5m1VohCnd0YG-B&rsG727o8xE0=3t!TuO4A%9Aqy` zXD?gRE&{63`R7&dBG;^1W5GGo`SL@A4XYBuI9oU?x+uWo6y+kAl??UFBVoQcM`5tH z{IAcRc2dZsupDF0?BiLSfu^-v5GGiWGVv%zspq;WirVSD>NR>`8XF~kbbVDp1hD`c zEoB%1M?)t&7F2~5u09NJzDD|{2BJ~|18juk*L^bbiiai&nSfnQ$y^008wDpD>8Uy3 zCL0VI7?P382L|edy}&ox4Is8!rHwjK3QWec8vhD<%-?EAB6aXO%>>WP(=xTjXHBRZ z%n0O-=6~KPEeomN*RV%7XW&APX`0jpjFB4(L5}5~-e3>HmXwMXR{;*@e|z;Y1o`oE zKn*ec9x*=OHzPGPi4&AXtv7WpG*O^HRD#&`ML5on*`?>&G<4b`7WpYa5OtAOXhO$> zu;3S)d?eAf%;z%iN?dj<5FnvLIH4mCrM3VCQ@7FtnAcgY)5eJc$vI;mAT(p%=tNO| zXAo?NGumEmgOPaHG1UoS8|-lH?221pK*?`Lg#tMg@7dQYt2A}GFo0IJgdAs+-QAN$ zGd7~6BRbz&^8*<;;=H?w9eW=MJDw*YoE$wrH(HV#sE!GHSQ>ja&U&M@+qA{FjlPrZ zP^d5s_Nq{KV4QVI7WPg2DaAnP3{86fqkzTIGz)4nU_jZUcgFr{GM_=vtYV=_3YTH) zfF91V4k>3)`o{DErIk|x%t|l>n^*|$$Cudrx_m~PYTU1jf(dlMaATz;tD;4fk|Hb; z`7HPe_xkJId#)eL!|Y98jOE}>ej?-E(B4=_uOhkCHjelv_J?k8!Zu}QNq^A#WJks$ zzRAv{mdw?UGXi8Cj%=Wmwe`>aJ`yI^-~c7$KgLQ4JQC6IQICt!?{OXt%ArlYMrq9( z3G*QjBcgZc8oPJWEwa@m<%}hGuR7u?o&z4k+tgh>8Mmt%)6^MGl1t>o9@k9O@7kZ} zo*b`_8C&(NFJKs5qfHqr=JHqy8O_P;(e1lNo?LoAH6uP*F)%jrlqT0a)i*eqI>i+D z(h`+4p6wG&FEw2&IprugQ7Nc7<1*37=~OJIIvdj*dOp&SGiK2>Hd;Pi8Z)_V%XL>p z-&Bz2lQKIQYz zFUP;D+XnYKE*~-zSv(Uh_({NKcmNvA(@)9#Rq$K6;C%MlM7k|4sV*xalO%JjcJuXj zwJuptxzU9Xjbf4ST1%5szO>WqiyyE7i^1!4K;yj4I~1Dod@Lmf>DLnjcpKOUTOOhP&0^8`*zmr zUe_ClH=178`@x8^!r*rfV0Bh-Hxsy*3E0mBL?P>F30fP=-JIy&9M=QOkpX9YAy|EE zANZI2ujeO!t*2CPO-*mkL#sCs>NnuYw=spc75*$WEo_}`Z6)cggUHu0=RgR*w?HC4 zuoC9cx@Zlzwl=%B$B9v4`T?CxKVZn$uZXt}V|P}(i{7?&#(!;t$bW#|deI%}*0$Es z<2JG1T9dpml78)w&TRJ?g0HqfSKYg4K-4jZUHqEecVs)i<-r%ed$c4i90Qw4v7mqe zMDpz&t{Mm~32KxLMnv5@I>`$JP{z%!Cx zDY>X}N^od~K>4||cdWnVwn05{;5LP$u};|MOzNK?1;M-%o^pu;T(}s_a7&&;&h0@Kq1x{9TLq?2kbDwyXEG%Q;w_MWq(hgSZhLPD*;u z64df0^|;B)XZrN|ikluM@Iy3NnDZ>9g0>BcW13JRfZ8G6_d?WW+62!s1gsJ&l z(4;U92m;e024HfCu??3HSOyKlXE;DcaLjzzBnY$10wR%uU`54A{hJ~`1;IZKL{r2! z?L||?nyQRbX3H%^0byuA(UpKj6Y|;7Rlrg#y*Xq2Y)qg=IS!tJS~3+)azqgj8e>9? z@@jxHHs1+V1X>X<-9g^N4Cq#_(YAwK=@Z7?_rnhEaY3p1VnppSY=sHhMT+7EF-}!^ zo`a?-PTW+$_n4ORCN)xo>P0nmjXSJGNr+QQo~Bmu-A8hpVf7`&Pr1^Iy8ga)N`~Rn zP2Hltv9#*O>1Hij7E9|((-x(AN}ATS!*$(^snroWw!N=spB+YHS+pY)%nZ{5k5E_JsWG;QAT?0kPSjWP(t@RQLE!Hr}55tLv&s}n}kgJc-R z^7q9+-)+T{jw{0?(Im;kG?^o-c|TbrEWs?@UggOw)1u^wI>Kgz&mx!U`mrS85)JWx zMgKg2fd3BtOU;v)NBzH|f3zpN|6Ygw|J4wMdW-Y7A!^yil z5A;8ap7$SHd$j`mPdCT-rT@3By}!}_P7vzSNxfqfX_^^mnh{RRo8fUT+RR)+g!vX&2uh8(XIS2 z>jKzUqFRXB6bW+dv7)gs-7G&cF({!s`WgVQt^75JfQ!b?IWF_qIRuAZEjjiYr`#!- z|MYu)>zyaKBt@G58^5Sc%gWvlho_>H-%)0+yo`xHp>nz&8g(k4Ktg3>(~6#cR>e^+ zlU)M;)Bp#?~Qp|6MIl7R;xIjBTsV8xy!HUirC<{B{M8TxYuDAZ_j zVI1>|vUN=1L`4;uQS9{XmKg%Ox3?=&aj+q;ZlSVRwF<_lWTt z#s2Etc~%(8=rYIB#2k=Nk3=_PcU`rlZcy#Iqz221h-DCt#*OtU zhRJP>BUm1^ZsV<@oSemN!?moU%5=Bn_FagJ&ddRd*@q<63ur?M#Ip0)jle?pm~WpU zOu<9Sqzhbx4MqB;GyUQ0{4mRX=IQ7=rbxsTT{Ls!30wTv=aY)U9x)!CVCK%FCg(Gb z9h!aSwu?5D3-8OmMo9TjHHbmlZ#2C{MjBU+UiTYUfk`Af=j6Sv6?QE4o0AQUg4dHz z4j3OGqHVE>cbEy70Dzg`$JfJUNsR74YBYeC%SVi_9y{I1{UiIx^UKO0yuRB+D=JWz z%%S;c-?}`CdGXoTN2*n!_bx${p4xe^goylr{mbA7l-(n)j&VMI!3b3iKl;UX~x zB$+%yN6L!UT112!1CsqnPM6-IWBC#(2x3Bq(}}t_CS_u@1A|&Bk>;$xc6u7{1lxN# zB!M0ADb9&rwZs?=5a3j5rJJdxCLtJXBB`Jyfk40(+s%;0>?$p--?6KC_oT8I#5-i{ z#H3~=YmRDUHI{+Yk8Pja*DUN>+lAj4Dq58V5h)v@^1^Zq;P`~pV@aM!Nl2>xk!xz8-;8!u0=v8IIhuSg*bwV?fN@^ymN81OZe$}j96>^?1&wk zLWN(V(R~JS3VIkj7J8d+_&t1HFr#KFzQ^4wDdS>I)4a{i@D!x;k{?SdsLLx`H?HA| zspIw{-Od|XtztaM(j<0pzRA~5Zt4PoK91sK;!mPnsFuA7mR-sjhUvWQk#0^E@A!Fs z!!pOpUKv_Pv3YDJXWS)z$S~qiOdkv52HBi7Qls=fFHtny;@G#1vUrV)aWG0poi?e< z;5{eiT)!`S;iMGQY7dD~DENvwWWjhh4jB?%is^dECG`7^qDqxokr`^mtOuAn#p9;% z;ItB>saRJwoln?R9r$=r4@!jwE*9D{i6_s|BskN7RAYP$q>CS<_DGzIA2|kDs;uMv zH_0oY2+(<(PmhG`r_>3!Hs5X+MTVS($8s~Xo5cp<#&RnjqfbGZ7DT= zQOecLa8>1b<&*PRZ~Ts`ttQAfZ@duDXorEcKMQP1JXH*_OFewp=(cr_KGpjibs3#Z zOzCBHB$5%e>~61Acxi}iz?;_cVUn;WN}0Y_X)CRZ<=&oR9vdRJ)EXZvNNwZ#@&Po8 z-SL1$Bhk~OdLowbO{?nabbDOwB2{A$lZ4*IU26}c^Ec?GTFy0oPyGdRHB|nyc`~qT z%-S!r=F_d?WzyIV6+a%7O1wte4bIuLjCyUrNWbFv8tDMdWO!@*ViH06 z3zbHvBzqh82=>cYkbU;pR@`{&c;`x-I-~A0ae9AVT#ZE8pJYCSl9^xF_8G4miytK0 zolntReOj%LCt@l>gzyi+Xffud4bLkZlpGSEDH|3O)oI~3ww>)5Qnd%wS(yEXbyPPv&wSo zwwuRv@;O-4FQTpYr!50D{`w_#ttXGll^XRQzt8t()go9EscO0u+XZQ3KAH`8u9H9c zu$%|b=s*|yX9gJ1e7pfMc&9j6z&;tN)>P-+rG5vW5zVM(531Aez;hxKVdW>83QjP? z+CGkZLLXCFjY|>FD^82%GjTzWOT{PO6D4S^|-6ad+Q@t*-O$J~vVPvo^VX z$755!RR-U?9c=x8o3A;GW{OvU3q2WD{IE>ujBr30b}iK)zEJwX6-mkf#y{c1NhwQU zxkdRUexet&*Sz`tvCpJ+aw8Go8yBBNGc+=JV`z5Pn;aY^R^<0FP>vxo3{;NY&3 zGfP3FE&UM^gTyrjSz!WvG(5L({oo)WiW%>IV^T{-P%31EMmh!=@`fqao9Rh~S!RS; zw}sgrQ+~n2GI$Jg;tltPgu8fzdt?y0X@<0ogn4I#2jE4R{|*m&1czuwL}Wxnm61fR zM#MpBal#)X)Mz79+Hhrg87wj)dX218@gQAXlAznjT;3=GUP*XzmLg3T!PF@1dJ`B3 z%RY8cJug?(m`mDd6y|Tj#oMTQ%Rrtb>CB8!G(61yvG5aTh&NVHXJ`zCM-2HYb5BGx z8hKbKuMbu|XM|K>E5KuumK}R1Vo(ejJZ!AW0jEO-r;`wOY#Dc47N?U32Ve#K$cT_$ zaX{rn78VAhk^%p;#S`?q1T5ge!i#yR8Zs6kA2Dv|u@5$c=|b zVg}`TvVBa6wzo6uJ38PyRxl-9_>_ixEw>t23;Yg036?zZ3vYs#s;$8x&zoiv1ymX= z7m#e*ri;#}#zF^v8w0)ZCOWkxk|0SlNN~N=05YE>iMNN{d3+&W1T#N{(MiZzq=|CV z0m=E21S7*gjvKMyrv~E1=?2@$(n)VB15p8KF!dnJD-2l2R8x;66RlK;RhkQ3T9-uf zJ(n`TFfAzKy(f~5h*i30L`t+ooV2Ihm&|0Nr?8ChB(+F!2Bc&~&qT>5$!?qs7cBaa zlN3+OjC3mrnEE7stHg?9+L(4bPeWgLtWK(FeaVP2G_yq4s;mU-Ti zYH+B{hab_q>s}fujyVTvr;Cspm;4=UzjPwC$Omty0!JmC#KRZfvXh+wrAzETNp(JU z%{mmsTH?W4`YJOZc`^QVB{}Dr7aS59CSD){K42`Y^PXM!mIs@s3jn+WBqJN;S+>33 z8jJlh4nh+KA+G_@wX?+xQ>-Ffd&q$O0MvlKe3JEivgdpXf&$_!#6DKIXjUL5D^QgT zj8z9>p$D?AgL5M3`|9!ukx_X%3i&1q1u6=8y$S;Wh>?p(#Y*sIb4b$G@VR|OauZ>1 z?il?1#cJBcpS_AzkwF_|V3Po=jFB+C@Zyj3C7MyF0mA4_ZOGhQ(ddDpKr0kef>Ps( zV))zQBm9!j>xDw}rQ>;p7W`Z$eeljITu4TxjWB*rS!L=nrNdCmLd}lSsMAtQFGyS% zv!hda3LP+!pxmV+LZhNIalMpC2K|Hs8EO*c?1Azym;ZS^UpfHs(`jksdO1;IqA(2V zm_zaSobF%wXR-<>uMz_o<&1S5uu-hHF`j1Dzo(%4DoA8XnzACkJy!-K{564~o>gh= zRZ=_wAI2Z9MqkaJ1$x`5YM?J&ii$8MK%IV9nTTAor&&InRgtA#OdN$c9r~B0D0{so z^szj^5Cf`>0hlaJBB+^f3p=pDJ0vKDL>I$ELt54;+d9hoWzYv@Xph)y5CW@glRRZt zDhOs6ur_1`OMHBKvvPRnT+>WLVDQ^^pDN}ATI z#vjP2(yZu_ZSPf)%S}%k)aWaA=$mLHBy&98iyJpe*)&Mk6=!)gDQz^H6Ew*L(cOIx z2}|_VU1>IcZtl(qmx5{JKW%U$Z1l)R6;6mp{L~cJ@O~aagp;!ElTsTe2OxwDBuWN+ zl8QCt2M6~dCdwkp2_wb`x2MUrCo{CC>$Im&wuj7tqUVrakst|4&F)Sp2zCFywb$8! zb_E9uZ16i)fd~J!wbw4tIb~8YVTIQz&{byxfL8+QD=BboxYYasz+V9ncDm+Jx`(_0 zvy&i$KDfA+=rxrl+dULDs7BXPWhYufR~9X>y|bhJPuJO5S5CBMCH zcZ!}$Z&1}_?>TXa|KW5*R?=WJ)^>KVgkw4SI=$nbv{-dTGZY|643%C>g3c11y1r^-8v;jCw?yuSCL9KInr_?tX-R)>BNpnuu;SEA_ z2#`1IM^Ng4W$c~L?gRx4O2l+a%k{wv1HJyX{0yG;vJ(y0=m09B+*_~E7rpyBvisoc z`<(HHMo{_@c6x`RhoV5Z5g3`O)x73}hr8|<1K3m_UuluJlbEL|SOL8@hzYP)f6r>>O--Fo zO?OSrzf3KlPX86`H54vbEwkSXLb|0Q)#=>Zj7+tz+Knkcwk>yQ;H;yFZOaIpZUghL z_AT7LU&dtGIhh%BtM#McyM6iWUoczLrW52kQrOl2GYt;Gv_mi|MQ4h@)rq{ynL`eo z!_EbVush&(&!e)W>~4!PesO#ZvG3BB*JL^ z3zBTa$^Q*W2>gL$l%b1hOsa3lZ-gGKMuXYQEfvyHSlgoslpEbFcV3U)LHfQgqm$k;r?Y!|*5``|_;ciZ&WYO*% z)Y8N8+)oBQX^XVk{0O={Gu;AqIxzNI(7o4A!K=xo?n;A?FDshIHRDxjr2hZ*XQ-^5 z{UFVJd4l$aj{FD!tv9bY=Zx{_1{gU3}%kfOFyilB~L zHI1YtDc?`|h+e*%dW!+vKm-?JyK(08+?3R?G2c*Km)EuIsqYs~!gX zuG<*~dwgI~Wv%+f=3Z29Z5K>kUk%7dwi84cbH#Mov4^Z=OBQw9gI!9c-;KAC?by}o zqVjw6(-y^*00oCG`#9ydDyNCkE@K(GS#Gz48nP9+`3}_1Y>sh)hn#u_tPRFnDct*P z_E~yL-==Z8P8A7pE?55BMQlCGhAsh%t(zGl)nBes3>-|{6DHxm6!2GupYPT&cDuNi zKaBdy&e^V@CCE1j;_250r@M|QtZ6-u=TWOUF>OK4M zA?4Nm00Bk8UCdh6?-wzwL+sth-qoAgkA(KG?ZYVp0;oizsSSyKBofU{L9aF>{MPqdkM?-D1-_Ehi z&owbx=cPcXO0$wqrTSeCqA^%DaRgZB7;CjZM0kKF$$iM6z0O8C_Qo3EgGNOU>Bm*N zrz-@i3M^xCC2Rs0ltdDi(_6VWppjOz3~D@rN)!tAjZ`5;vLEpth7De zQVJ?o6Q4V-Q|n@i87-tIwb!rH+NO$`-Nz^OpRd#VP)k_DrKgMuZZgK?O4!rKr_A|p zGUsASI7_9ct+j8mR;Nn1TgRvEyl%30P)m77rDvS7ZvK-|A*NLDNP5<5{pPC}9i`YU z9~A7jck>NPTVJHAq+|h&%o(m5EE+vF2TW-obEh?_rdU^e5Oe{wIABO18P5l~q!c2| z(~7(x&0Qh7kkN=o2usVTgful3k<@FSys&}|Q2R>En&=el`2ks{i#R=Xp`^A(Z3SsF z%2KK2Drg9h*L73 z!G`3pj**qR2MbG)WKk#x_;4^2{L4MpceQ?Kj0Ti-VN7MnA`8-U&MAefEaYXnGl>cD z)PE|@Llmgl`_DeEOlUO+>8m~k{=^4Wo?c^U1M#||}N?b>O=85Awj>>?M0LHdsc zxG)#Cfo9Pg+r|^EV|-3bt*m4RHC(W-SK>H+yBIUZMx)=p)FeE~R>be~LVZ-pW+DWv zuzrd5;l9lYWBbIIuVK2?N9rv_YrP}fUsWgTNv=h0pCVm|!kMc4BaZ#Nsl|2aJ~m+L zk{Ic1ASl~U+RN*hlFtgykgfdz9ezAJd-V9V>TaN5SU96F;Ar0(7nG*oX#PPi5iF^M z>{P-Qc|x>78~twaZRaLlJ;o3xetftx>bI*1!jE0_`lV+!=P(l#8=36o_H^@`633Ht zpI99pIp+I@C7DxI4m>W}1$L(TX%d#oyQ#wJUrwL0&m1m0rb)D1Z}u2<$EQ5k&+fS! zCFhBTME~r@N?+zPI6(|ry8Ug>5O%S zu$Lbj9Jd_({fVe>Fv#k{ogIwra~2p5e>Ea-5nIWQk@KJ*G?qNKNopfk)BQLj?Ja&Bt!@TeD>HsAWBSGLXdV(+n_q~#2#~5 zqXoB8KePov1p?(eqqyn@eriojNggUte!Q=z^aBO`EY3))3plt=`w=CK$qJb9?zzc3 zdQp`4;W_S3@dCTV?Di+|UBOVRnOHcY<0Z z!6c2+-{Bw*t&mnsJ0utY%T+K9I0Qf&B9ajxh8Xav%t!bbK5Y@=@c?#eg(zm=s7hhJ zcXNRU2D3_mLxlZ8SV4yf!M`*hm7Jk8Y;ZF5bnI+jq(zV`=N%lz{Hv1ikrD!=EW-0>KB zSn3DE1`?g~%PtB+`%QEmjt@_6FB|0Vw5rqj4&e6a4~z&|Yztp4^D#GsQGYbVPS<>D zfJ3^EGIYn9vc#<8rOapp?~lgy(MD+*IcLoWi+O}hxWnID##Xe2)X;`%Do2|E6DW4^ z3O&5xen-Lr!{FwT_#CNV(t}Sd8SH2+v88G2fPiw%*d>nugtAy_rI-LsFqyP5GESU5 z{D)XxDVTA6E=Q^J_X)%eaj?Lcw$e~|t7J~-qucKYkG#Z?4_4vKbx|Lmf&r^h=jYh>(p{h#f6hmpi)21M{^F z?4p{69Zt=j83~gI5>ra%L5c~!igRg!z%r9=@&# zbH)X}<|1si3I>tKB}9OQT9MpAQuPs*$r)1bfXVDIsj$B>@_7NS_`&d*n7&s2{_QrO z@B^qF7@-OAo>D0ik0}r1L5q&D zQUodKH(8#?@G{79$kypB&@#VM3vX1fOjz>5{kG)q((pJ01z7}XG7%+|9cdtAlS>Vv zFB567eP+ehnI~GMnU6tPT7~lK;ClAR2$j6F-(pU-X1P>`k3~oM##8zGih_Z13I+m*7;EMzJvg_|vN_V- zMicyD0UO|1`EHG92afvPpTZOCGSQRpvnc5yZSdGhPRui&euse}a@9*m)q*y;%>;2} zYQ-8o#KS8N?sqjQROFlXxPOHRF~LS-MMPQuqBV+__a%MpZ~AhWc{B^)8F2G)L8K68?pm9)P_ORv4hPFtRemmmkUz;r(;i92I)NzDGz{@ z&{-@1IK+?Yp6rCp2lWU5RPqpI587M~7FetxuC~M0uNEqQTG@psNG{W_zUa~qE7A$M zL%~f_RxVcF0rta)D5G-$h;+#)mIhd=)W;SF-qZ^@6x?Ln!sk zbmxO(iyiv{1Qpp}nZ%)W7j>lOLHw#gF5O|;A(hWDqVl>b4cWu`f}9dB!_$pJ#=0Zs zJ|mXrD$F7yu)7|1ns~ryx*{Yuu7(9IGFMDhxua z1EI2lQ2YLZBy&@M`6&QOA7CYCdZR0+aTBv`6VN&}bs#vspEDz$FmoU{b07y)BbuQg z$MPBm$t%sM_f0-;PNMEic5VV*yJlOv027SkwEZ)PTQkVHb4c<)h8X2qhFP-wxi`kS z{g>HipZNn_Ac$$6VRvRfX69Gd42t{$;uZ)?X^un)a)&ziP7etDHAx`!9UrKwSewOjQwREKX(zYdr?++k%t&U_GP9;a9T|m?sRI7A8i)@ zWicWK&`LD-)HR9CTJ#_c2Ne2lq4!4^b)Da z{9hhmvLe+wenP!f&gprZ=@nGS5}Gim(^jA|cu6yNAqowXiTFE}9-L|)TzcVRSoLx$ zF{Xy^%qM<0QLhyNp=DvF#UMr{&TO#6^h_twa=PmBx9&y4t;KvaOl7783ZZqH=^1UN z!A>Sf`s>28E}Yft?4;0KB-6+jCWyW7jG*4uz^`sq^-cWWO9Zi-;ZWZV?&>YV`He;2 z75?g(6sB!_$4!`DTVI8~?^bVP=g;(f*}AV@r<_{9&eb?T_`bEZ-Ris1(f#8l_q&7M z_94*%BJ;wPEiZ@!vWd3y^0L6zJ^Pz@X8skd7(652xBIhehU{&Zf@H5(*Bdg}cds8Gpw|Z{-LtPY z2dBQXuLnMmXFgEghEp?ySKdZ~A^Umy1>%cB2iWODA3pG5!a@*bs_cjVE{bd&hyUe` zz(`?CA`Wb@O>N4&y+rPk1tIoEC!N4Naz=-^`n@+|J`B`vL5G$&^}+zCe2&wyt1<|b zDNL1~efN&Vvjq6-qlS;2E{-wBlaKSM&0LNYeMz;-k8>_gP}z=IVE{>Qa~1mFB@z8n ztuvcw>sXRA7xa@*h11k8hovOvm5HYvJ!g3&Ct}S5$s}j#3N9@dWG0w|?fT~pGiT^k zeTrY2xf9KzzB&yl{95umulahOx;>t|mvD!hC99G(hweC)NU*SdQZaMtvqw9$HBR@g zn8F?0w@t>ULoo>feD7K4c{@qxzfhi9;xN}jj}AqPKEd+N%KmVLsdwQ=cTr$>N%AZ0 zkT+7GKPejV+JyXkf5I4s<hn#Tj3AH})1?uNG?e+j-l!NAmZz96xvOjUW^U zN&V{0jP9A|7OkND(Dhak#_~-887V=TH~x}>=!WEw;t7KN86&Gny3%-{Jd%tdLi1qV?6^G zB_A!>O0d}P00cn$zv&e<+uvP2B9J~4l0T`RGl)SiOVGKZaB@!30Ny{Z;6MSA<6UYj1L3?#@|K-{T`ApuP{o7Cx-00 zWtiM@k5zy8qH>XXvzs;UTU%2=_(d(%9*GUjH0SMF18x{iV^agq1xM#H;(Y)$ad5@+Lqg zyzh}E2wi&iy8orI-Eg4$=JR}aM6<@tCM5Iy;uVVEP|$n2Vb=!J%b_Fy7fXqd&;B8c zQ5cEl1)~ReAu{s;?H2lWjfF5I^M@`(zNNcD*U>$kGJz@?zhAGWvEib0ujb^RGTs!3u#bMmMkuteqxs!AeA^+OjJ-Y2z4!Z_oRmY zU1~h=xPZgtfNOf0eF~iD^(IpE{Xl!GLBkB`V2;H94S!ixXcY37XJJk)bWO1Htg4X8i@9gfWJU{OEgAq4WPo|p`d=b*>J~-iu zIf;mrzi-<;-8SYL{~x5iWmFuEx~FsK7F*w=C3!QVk~C^} z(j3fV_5RE5`YxLpZVL8M{8l`n!;-`m zEC`jqwLV86^0gI!9*98mHHoTjj8v!|VZGbF0Ra^R0cALLf9v@ZiW2Qwv2RW0+DSeB zq3f{1&#~ta3`8l&bB!)RnFae-sj(CAC(o{?khyStyVg^J=34WPI}gmG~!muxEgJg1?Qif)@W$c|pS-(6krl z+N3n`e_Uz|EV(!t7l$u?C8~wTxv!c4kS2J)=VQ}YN)~OR;ETBtfzZnG_O(i9Z^!a{tXYs zp`M^N#0~r#53M;i``2WMG3a3&?+S`RG~clRMS{>jk|Ds^G)#l2lS>>O*S3~cXzbak zDXOwe|6U;4^2IX4aIW1dvquv3IBWa$L}@mXBojlngCgNUF5(c=f~X6Yj7{#_YS>AE zyV=)DIbu64?Kec2q_*nFdb)IlaTM0Yl8J=)cE$15NTBlc;WiwmQaUK4vH)$lI;u43 zhgC(DdAnWB+x)2-1v@i10ZVZqmf}Kj<8jVKS+d7eTJ2Omltg>DzpeTWg{wKUd#}_n zOXMJXd!%Nq(5iz4F@#fA^D}9bdUFaJ+EU${gQ=~L3H2pr9YGPCNj)HyTzhIpD7r=H-6L?)@$G7#@O!2g1z8&A!YX%4#igInhoRq zUgxY(8&V3M1hReOM34YTOqFs3q211Kud=EE-{{w=k>sV}fPl;&B!d_5jAC$lXcD>5 zNOaAT>X+UMs{ocYl~i7Tp+YH}8Wg*UB?cgdutC-=@zj21q>0tjJ=&(>={V1; zKoB7e1Lyg)BvbYItO5^5h+6truLdr@5BtlvLZzk0Uz5 z&dHcR55UryJFQx_uNU=BN@3ppwoG{*6%fYx*MGyEC~*{=y+jkSr=3!9F=}Pj&#+a< zhheW11j_;d;Nc_@{^!9yFyQ_(Iu)SOrsga@6f+l)OTM8N#ETR;NKZV3%y%6^ zkY|dywh+mNfy5{`U9g$O$2}%9Wwq*;ji&i`fv7V&fC@;hCHLw6OHw4yt3oWRiBFQ9 ziF9q~2gmT4<6%P5=4cksAo`ATC|++7Of~!fFIQj`s-YfZ<19@wwSbp|X|2`Xy?a|G zSH>ka?*kG!yv6PO#Fvs7DIHw;Y8x7^N_9?aiC02_VZ5pp$Yww z{zj_=-B?12aq%C{2b)LrsP^kfRxYv_;(8)^gg@VXr_S|&hgF~MJiezb(2YEvqYOk* zRRz}Nk>e<+M#TZJH z@VDd{xKaGe``9|!Dj!N9g(z0VxVuRc`sXPz7*igjuv7^w22Ss7KoFnyLCZp8;yTPj84bB35E zn#Z5fv1UHE5SO?|31ZkvAT#4;wBHpe2bAM_!KRHAlapC4_4H8)-0;2dw=qCN>N9ZFm1uWR$9kCl)p11#wOU7hE`FD2=34v zn~G4XH$uGF>YEnFNq&Zz?UQ(rb4cl;(xCNDQEl$`Vb}b2sS@(jW@%1yg-x%WncH)y z!LNg+E8Z|v0R$~Y*`t)sMLJ8wurEsoOVtSEQZ7-~hFX7G|V#3{5nK3LvmspXHLIgwDP^_HPMe197MIiW^(kPedK*HysdVGRHQNBRH5R6?%%p0 zJdTP6nkDvtU2^y1Tp~sVbqh8g37sCty)b*Wy<}TT=&FyUMvnI;B4_)W>R7Sl6r$OJTPPC+JcjwNC0iVAd1|@$}_~QEo6NzusO~o`)R)}}#Ba*9D^ady23lf|yV&hcZ zd}EwjL#K?y_)7dLL@#yhulxQEC2!uGCDfZeNbmSU>EfF~lGMCb8DLN~-xf^Q{7C7w z$3;7<1}s(#XO)_ZuENAUA8sQWl#ScRU}Nci7Q63%LYfT3$vh=0%$Y{}+Kg(r2SlXR z9kdRl30wU}9C!CIaX0No{Rf5)!xmx}{qNRuMSZydqgWmGxvhdiKXU_v$41+G768ih zwyPLf?@^#6i6Vr2vV;_7p>}VOy|w;X#iv~R>sH2YzI~JsezNl2Ia_}8G11xza|_y+ zTy!gk*Q80GgWTJk_up&8_+>S`nOv6yizeZH?QvTwcavx8)z*wRSmJ}rdeI18+8u4> z?5bQ+ONhU$bc~TEQc@mFyKHNv9pcW48a3tQbH5!Ue%YJ^dugoc-dc;jE-$unkZ$?y zLd;@RPnWqf`L5xIdqdAxTqbVot=U_A=Jm*mw3fabB`duXL$?=pzxm&ns@>?ld!LMN z>)3&wU4vcwN<@PB-$h)i6<-csJOx$>zVze4=?18-fs5o4kN zP0N=?1^CYu3vA=YpaN4l&j3~{ig(&e{+CNX&G@zd`XhQ@0cvPoE4HI)KGWkq%;T=hT)v$Q zPHGXp&^IWG3!bke(K15?*bo6;Z#;Gw0=g=KXEMFCT?8W|i5y$`oIZuTXo@qggk&Rz z-G`|RBKu1(fTWT_DRB6K)B-MtVXjMIi2Grz0%32w!#O7e*XTlLGv%$Q05sGBVCo3z z37mE+>UK_ujyzx>DZ;my?!g(oeIdep!rUs0PKP+sB05sRHPRf4pil%rrHVpL0{oqG zTLkjTit?+9Qu@1g5fnp(;08rff`EV!5M>BI{vZegJ_y3xi&j=cNRfvn62r1kU^!VZ zJV<1RDYA|?x>r-AjX;dNMx;(>WND`ayD6A`4_yBk(~J_^ARpVJ8{2vc{s6@_mIw4b z#_H>m>Iua4pT-(_$5~{>O?1Xt*ka2|0o)g3YuqGIFaxle;WbGd2re1d2ZadhK}4&`2PaYm)*-IksiX>!0jiXM zN(kW)fM5`CTLrs^CTa<$Yj=TgpXhc!Q&D8o8FeKa+okcJ#j2&GVV0QN*rnM{M({kP zQoE$6IK!9)Gg&4<`O%O!lgY9$+MVb$#N;#}LKcQuDsL#xyR!sUK^TlA{dyuxmnnnk zAOqhgEeV#1Tuc|PfRkyP;=Z2ku?{2YiV(0%eZQ9OdXSuE2dO$sIik)AT*t{)fGA}{ zBJ~ns3W-r9>D_`br6dMioZNeyv{7i9$Ygq$PnwSpuI4%*Z!*tSFS8I9*+auji35nS zOTT$c>s$wPFeN5tGdVKh&cjh*E|Z{m0R&Azz8h*zw-h&#JRnvf{T(E-aUr?~mVdjJ zTfJU@-<+2skKoRY1g1@vhOV z^@^$VA^AWUOreB9k@0~O>BcViu?ryfT#A|u%YBMITrXzz!6rE?!^0`dj?O=iF6^4j zRf#DF59O)Vq|A1~Y-&nHgfh&}A*I0pshC{4ni3QJMCGZ956=+L`xLwLY)>CXA2b}v zn0(yx^2n(YGvb1-An7vjVWuwFni{ySI#wL+rdSiiw)0rfDa z;i9G?e*r0<3XsxW-_Pt-?O9%{(15y|Y>;Y+px1~&T0&Fo|?R9P}8_ATN$EiAR}6}SMA%~no7eeTUxtK?>~O@mh~ z@@yAvJL68sEcvLyO;Ce2g$I>)7i~&CS}HxsME0!-gda&a<;-f^Xlm6^E?Q*#JeZXv z4)~`_84d$p=3;IyFu^wmUjC88@ZVa4k^K?~P6lwi|40$iNR>-CAIkdXjjgc-&h&IIxl7-gF#qOe z-;OZR2K=QDLo(vPpr3~;_c#}_pWBZt+5hdO|C(&TWWWE1!N5Dd>){#mbIFrPKPm9ctE7O@mA)9;zZ#)J18k2lg_4a?eC^<}f&|V$ zDd1h>g=x(Wbu9d`4j#Gg{^8I&;P5Y0z}-V@W5<{=>&K$HW)gH+0Bx=9%!JYLr1{GS zd7X~*tyW*wE|u~rMV8JG10(R!#8T`;U-v}I)+FrLc-O_pgE5hhDMhL|X;pde+gLj^ z?WfyCrrx}p8WoC}ex)?~stZeL=7ZHtQ{EUjcb2F}-41=WQ<<)NhGVFgapl)^ckk36 zp~+DUo*v~!T-RAN`ne&p3Cpyuj=XjfmOixU^!DI+g|hiRX6o_iF4td;Y2Ri=H%+eM z77(XH>n_@2XGYatbtyE@x^K?%@+?q`D2OYwWE})ii!?PV*&=L>UH1&aZ5A)`7Jbls z+2}7JJfbB328JU34Td5BxB!j_CjT2S)OiV@`oA5IFw?muRm}X?;LK+ii_usT*nhu0 zdoinG$%?yt`tkq0Jv*I&&lFN7@F7|1J-TuCKgT1^O1Vhiu|}XPP5&5=SSxw{UyVop zB{(xVChro9VM_MKo!T$T)7J%}PEA+U1Vz;ssfRn1sGN`(jbyh=iAx3$cbra>*q8Dz6N z&L@&mV058P2`PaER%DgO>gp??CE{ImZDx5Ez!(!_z~c-mYQITCx7az}64@wxPgoHb**Ad;6Ah=IO7UbAjk}QA;gK zxgRr)V_8y;CNFY&_J(VnK8%to7PaVOJ)7s?{MpGGLVQ}@jG50zJ zpaOz}$Eku?T_@-gSsnaI8$GAl{sv8_IR{wXW_agWZ`lyaUTq3@WVueWh!4=XPfHCL zwk>c{in=qAa8$V2V(i|u&c4=BdHp4SjdyCPI)3tYRciOg$2IX(QN9I>LRxncM3EWr zYE`{M`?UE|fX6o9;blyw^Ef=-b6o_^Ca@lc92m>ufZgP=TUqb_aHu`*kG>n1%2y#A zoSW~npGz6&eUP5ZXFcnkiRH6X>^1vz9^T>DDGbvrc(%4sjPE!Su?X}y5}6kBv$4Pm z7F-x<^`91nZ)yI#w!*%eI{5%+&)G5!x*0K0sPE5#ojzizQ_BP+iP zegi)J>9gwK^naPXvW3Tr_NHF_dCB2|KRizHnB3?-i2S_y&Qm(QIi2K{XafE2o+4~& zD>UAE!nNRmZ)i+%)gq2cTH%>6y?aH`!H@0A;QhzIc)ftX^Pn2rg+XM0`Pio8YWUShvtUL4s>IsYR)o?OD^f-C0Qmz~$NBzMB-eB?KF^^H z1}-F8;cSm^T|=0Q42EcLV(csCjP@O!^}Z}B>(yww&hxb=>>0DorRl(B!ryD zy{w_3E?S#lcZ_(jX{_JcFnN1Z)D5THq=J4+%2j6pDIRLJwsSE?9-tg5tuUpY(~_1a zaRim;is{%!N>e{N(su}vPf9D!Xz`x`kGkPhglLE~9Z_+@!|~an5>iYt$;57Ygppi$ zd@(pQ9GHu3x}+ls^D`C5m4|KD#Pk{If-!vK3Ui^Xt$C-sX2XkrgQ1!{bN-C>b4>pL zLm&T3)mD|`DU3U0s+3V@?tB zHAkeR+?COj+iBpIXkE+d=6}1YYt;FS3 znBV3vW>v)8QpZWgXqJ+`GxoXjQk-QDjHw~0_r14tt3H;LNL#23cvo8@naHc@wcN`5 zjx$zC*N}`>CU)5)zuN7>peu%;@9G^d*$~6(-4fS8N~BJleXYqoUU~~3sQB&c7|j#> zFYEh^fLJ6JlggIDptVo_aDfZ!`o(pet}jgHx-k~(3EPT)pKEF>r3}I&c_el( z9BgM9H!I3K+rUE(fkGL8ou8ge7N&}(<*Vz@#6TO^1=(*LP z)n?CPzYifZEJs_d6-#cASQbt?O39?!Cvft`liDs`@HAM9Zwx4CS;<`^WWpV`KfRl# z-0jb#I-z&g+^S9c8k#w%gc;+0!GeD=%r<~_e5DS3F`AYV7^WPOP=G2^qxY0IWg6AV z*R4Z-w&+k{SYPGo8sgb)cTh>lzYUn1wxIf^D~<}509+YR5n}@ob(7dpGDL#uS+XZp z9l6tnii}xpg9+*&3P3(C@PudhSnK|o--(DweFa&RS-T)Sxpa=VKom~ zewsk+YAv|U1LDZ-Spa=nbCLhLocZ%lFXST-{Mpkt!rh;f0`E#-oG4(NwDeu_5LYVD6gEAnK0znGg(sYhsDT%1({JPGH{6c4ybftDq0$;l?1h;b#i9@|?5jS7H5EYarYUep69B9QJT$Z)nI0<6;mVkm%kU`dcm>gLv0 z3a{?@`YXw*mnG#Ysb1mwGv6h$ouaXa%5;EGxFu_YLF@DiqzTN0|7*`eqH`Ftz&Vo0 z5W`#|7+rralq_p+ZEWrlot@nwVPFrF11gQWv9cSM}UO;OXCIh*`oupc&VW( zZmyfX$1+(@vy-B3QoO;a;_$BWFf8}Cct)ph7@t8xT)3Hs(Qj0NgQ)CSu4X#F4?p#H zue2sO@Vg_>G4!ig)L|Bdrf-{qq;X;yaNKgRhc2DPy|$*~Z{PA1?Ra$v?>sa-^y`?S z4s7X9o(Uz5(j&LpyiMLf94#3YHdX^%caAY8-dJ$-fF>h!*9LB>Mw>1rb2H3MWE(n@s$L1O4FqH>I^8Pp3~m{B2deu)xP13g{QCJr zbou=WJN+2aU|x>rV2 zjR$brI;{~!Z)Ju-Zj=zOeGs|P?$yC(HXJ|UKsTaFzm$`}G-Cf*ZJ>|E&kRYR7SQDrm3MyR-s*w+_J@GY(wo{To(1f^~K1Q_3$F`o}M4Lq>w)ojSf|DP? zj(cc>HXep2ek5y7P z7NY6^hiqi4UA%)x*5&^ozHtNJy$VO6yBw5phyobgOT%#XaVC&k%Wn^WMPT->6nk4M2cc!V~ zX;|LKQ_-S?SEkTI1-xV=K&#%^713_K0y)agX+BY5T2zq0=roBAkhSwWN-m~b+t4uc zsCRY_j;7)w+3B*6=^)9sJf0BCY;BuK>XNFM*bEUdreMRz7$QNyLs2rpWM;Eo9zQT; zzzfLe3}$c!Z@6XMoCGq%t8t5Mh;y$sJ!EhuxuSU2qftVmD)=2SpRj!=;ukw|ZWtgB zY6*3)v|teXt{om0IsompCq@VCg7X#~x$mqS=S!wGy!wkIwat;jce3Ot12`g|riLDe*2NWEIG3o+Tu~>R15U!^WTRYF?VW!QSsxhre@+WTK)v2g@A8Rd?ukujt^P}RI zF3ckamt7ApDY=q7gX5gxz`QKvd-`E1%rf_>UtX)G2P6e%@wCs*On%Bt`&@qdjk{#ln z9e^ajAH|O{R84`oX+TZ@UpEBjCzF&lrCV&LgskqWZ++THku|{N8&@Y#vI`qW9N4Uu z)x#Y^Ee;ClGUtZmzH}8RLCRSeZYHpBU5$A$e#5u`C1@`|f02~cw}DjK4m1NZAP z)Rf)5m?+J~HL)+zSi-qEQZBL|kBx&^nD=-l4SK4noBcp;hfKAzsF$-&Rk5lgu#Tl` zS_Tc#T(%PP4e=A-P|ijh6 z$G-Jpe3VCWnKTQX1>&fW)^QExK|J3D;h^F8fBRtR%(NNVgy1}}^bNAhiravJdn3Z) zCNd7?nq>Sz*YGl8a5Hx+B0N08@BOM6v7TA|j7D|3wA_^jFgSxpJr5GrQL38jJ*e1<0iT`Q?p+jh zS;RB?EO3cKT3@Qb6-BZAdDaGsDY_`Cf+I3pgzk|0)uSs*)SnvJmLH ztRRXn^qUTU`-|An_-SwB$?z9*OneheKm(bGdj1OgXR6LPD(w&1pR=f<%$J;J@$6@3 z{Oa*6RhDJAmo;H41ox|g%_}c;ebH=qd4u$5A3_s#=wGWW#&6>W{9ev`S;;iQbv0U< zvtbINS&ZIV=O|uVci3P>(D3N#Y)IG2tKWF2vxwI24`kb9x>?gqp&A?B1l}?>ji5&Q zPjwn?<+)U1hHQ=XZOv?N&Hdh5Am3gT-Cj1@UJ2M<%irGU+uqvV-huz#-Xq^R5ZyU4 z+Bpu`InCcW@7wuyNuWwVd;Po29Krh1XqWz`rB_Dz(YMskVV9Nb%O5rzYr;K{@tz~X z9(V_WSg?nYfcJKD4=rR5k^n&$+b0a%Cr%(FDA<2xyxY9JPepM+BX&S%e8A8^LK%3# z+<(BjbHJ`zO6Q3G(z?&UzE8MA&oOsUEO$uncnHCwxywJ4`g0&=yeBDkMBaZWPcg_K zMlQFr{-*zk1-!##{MVVU8UY)F%Tn1~8H#kkQM!Wo< zGqO7Hkni6Kb9yo6mD+v2|A7!?vM_4cHLv_9LL_gEx33a`jz=-m*=!MQM}!ymtMmWs zjO_mwA*$?&eJg%ag%Ejhu{*ze!4ej!zOoxk+9kCY!nmDD8H)0o2tN#s1Z7_uTQw_5 z*b1MHDvUs1Hz_>D$5t(h1~+^^K&xCH@(!`NN;?3)esCb$-bO@4g0vm65P8`}OB=`n zCc%Lq{w|@2Mg%3%Nh)$bP6?ARESJe3EN2w}6o5rmKz-m~c?N@c8Oo*m~q)l5l5O`M$7PvAZVYlg5;Blf@s4x7Aq(df#gE+s|rzp5EDiC|iDi!PxlGcay$;EJm2U zvq z*?amEqPj>ko2-TiWDG;vVI$N4idvCics+pF%)UL**WDMSQi1l>vG<`ZS2$dBQ$q!4JY+^*l9{tp4t*aV(A*~79nNi?$U z!#9d72dRlVEZMvHzn;fC=Knma)fM}i*H)_XnEyPPr=M$3`95&ROE!2NGu{ity$(ji zFa~!b55mtHeCzkaQ)tFJUy@5)ZvBidw4(&?yS|5g=P^E&aPIzMw%|duWophFCCSEA z5ri36d<8-o{)z($zceVq@Dkub>5mMQ7%_psjcg=8`8PRHV^HA2D`fWbRio8<3-NE| z#EGx~a`i4$*oe{mkYMVNVMJ*nr>tIfQVzu0wGzUBPEx|+CRgb4t5>gAC0{YzM2`!T zztZU#->6m#%XQul)h!*B0y_tQMx?;pkWh@R||QGKo1BPoC$&4GTwrU zSS5oaxC3!ByXk1LK39zqgIkMLC+GPRtB0(nuFB>NX?8cx61mns2l$RyQzAT#4ZEi@ zu(!lqx^JXKt77M0)mH89RZIAIrWO3uQjyZDo^pwSBr1J7f~wf5Rj_}IL4Sirn0&Bw zs-jgjL`JLb=c5G(Y-7a!4deP9-DRnd)I$Mt_O<+$CEYKw*e7lMRvqyO5imqM^ z=j_!bn(zt}Pi`Xlw-@smoox69UlO_stCFjt>g+~BZ5ct`56fKXQcr^B;HM9Ag!X#l zk=ixMmq$!Eibm6suwtR?6zfbN0OVkXQ3~MjCgC?eRMc`pdc|&wStSFxi!aT)q^9r4 z!2t|n(XvVn27pU86P>4NCDE?|Al5yh*CTG91Fp2Vl8{h(m6Sp1sp0#)`)G2H!$|5W zMc}4=EVsu`9=^F+DC!{yyn`c9vE~@$K>%ToHx*j~mn3%*66UflI^Ud2yjR^Jg$*^L z`ov9G6l(=h-X7Rm;O{2*BRtNhNtH=^qucL*XvtUo5N*8g2czpdKvkwjT7UEOUY_wH+Sg`f(= z^45>$%gTbvaqfFSUC+vus(1bFWvEBt5GIM#r7Y=vvN^}mG%^!UcDmZ-qzb(v$~rS4 z8$TdPI)R23p1Ad=PTz>Rf^3IVr!$qz(N;dU7z1}#zTf1>StxJuiP6n)kYM#;pIkoahe-5~Q$<${h~7?xP}3L}@Ri z`0;DO>jncS_8oNc$1&Q?*GP)qZt2F;4msaN=Ajo!P!T9vd7HY>TI!pPcv9^x=vszI zq|W3wn7(Hav(5T>IJdz*omf^93t>&#c;1VPF?_m^y*aZ~9KfTyxnptSv`&8@d)b2m zfQs_NdA^AKkw`V(A9#!?B{6#TnD=927~+>96$kWbf=+R7{N6eT84goEW1e|o_f)#y zSX;tQaAO~^KO*CqoT&Y@_JMdqD3D2nTWt1LeTYv59J~T!Jkgz;vClgk%lJdDY=BkB zLG;DGSwlf(R_@9t;Ik7n_#QTcDfp9Dm_nv~OcdT@Cgi6z?&0zW`(bUCJMDEu`8SWQ zSK8pSaaFi8^7pZj1U@HsMpY&hQdVO1P-HhHa4@{-y*#oL@*1w-0}+D4{B2L^gcYHY%n z^%Evg0f4A;>#(r#5Uox^W>;KKTM}PeRReGLbA}klaYu^@T&q(eo2uSi^hC6EszJbC@bz({?ZM6jAQe7en zLt>lSBIdK=!SX)L$uR-VF|gvepR@^2M=_U`@mGAYJ1ArVnxZmZJpNGset|?}x#+%R zP4i+Rz@BLEN+RE3BKNHF2}uqbCo(*U+&qp*|1nD7#ERrDkofllSeNV z#R7fc1gL!smUv2)6ik!SOVj3avQ|sxxTc)JNoB=LR})Ov(EIDi<+Z@6SDmg0OE-K< z$HU9edqSQHdHo~hHOMT(4wixbDT9 za4Q!GCljJU8#zfHMZ}Zwb134z#<3b)$qG zjez;NgrFOB&5HV8efuT=0B`^(`loM?@%S1L{7+5dtyc8kXp(3ED20YTB~?BJQ9fH4 zrtwuF42xgTP*CD{{GXa+=`cT(F0A4ea~t$?DW~5*G>K*cARNo;tciZAniknP6xM>z zSo`+x)awlMw2xfUaJXR1ToKW-QBFX&VI;Mxvcpm1HO~m*j)b_bbWQwNfRJ+TkBH0-Gunzp@og zmDOrDUcFyz;QqbAQs2%n#Y|G@TgB4Sv@zmfFz(n)o%o%=o~{82xOvgO;i**j(TAB# zv-J?|=)CKQ-Ik`4ix67Zxb>svvT?V?{+|{Zym(QH+5AMQBF5za_PdMb2++7N_Y`7W z?<5Bu<<`x#SPU7L8+@&GltFS;nR^5>9>ZKCGJn!A1E08?=a4z+W$RPXgmH{d5`Jj= zO!bMmzBUe=_tAsPo@HU#9{ve+4(#UNo{T;&bUTBX=g+%1mja(X-MTIR&CF`#^K{4v5DS=b8Oo3P9u)P5!D2@I?)-vGj*sc=wXdo5T95G` zu)IVCXg=s%LluE&VsK)wF$5A<1Y(pQLeswvC7vk)^G^(6`Cf-nVHBgt$q(b^T!%xI zi_vr@h6%f`BRJ!VF>K^Nk#1Z^3d|H^d1ZYvy&Q28!8mY%Q$R;(_~aEO+9odJCPo;9 za6-lWOYqfYMws-IU&|j&;5JT-vish|nqrg^kIRp7=iJ2EDwmS3PK@z&-^4q|mA*QW z9~auVN${R2rFfVa7kj=*48|yzj;1h1Cl_Rc8Qb)Fe|pJ4sU!pMl!#ntxeTyL3HEC^ zwd0qSOn9|4)pb-FIYO{#APvOuwHu%_hJm+|NySd+CiOfgkm}J!9GQ6`6pI(leh-UL zY-AxK#mV98W~gM%8cvL3Lm|~fl{n(m=Xl??Qu5hBD1(f~RaRRW^fk(oEEF#|@UEOH zf=OzUlkLys_llcoCO3wS;VjR){8wLWWy46dk^^mu8C94S(`yz9j&Up0?$6{(2f^5a zTJ&TZgi%e;c@NcP)vR>ywZ=&o-?84;DT*@b>d0$nDiV~Te3c<_r4`E`cGCCQW&*ID zOD3UtR8a?4%6&OgFRdV`w=rTtGkDex!sRU@_F$5&?p~-hephU`eQrgND^-H_F`sGz zKuAiqgpqaMRHgDwPZ#A2p1*DzjMd4@4b zSAkU=@4M&pett$K$&q<9N~{;c3l>KFK?$sn#!pj4tjRev z*^X1h;G_^%QvmwVq zb!O25UsW4+48ULCk9q|a){Z`W`N6Meb1M*q!>7966?<^&RO|sT_906WBNz>z7=U=B z^4(|Nj}~um`D5dDZomr}fJ5LDrZVI@v~VAZk<)t!FYr3DaBt6=QxcHDMg*S81wJ$c zOY^cPp;`JLQc@1$P9-gIkC4nc!RP3mEL&D7!$=3Qwy^-Xs@FLKgU12`U2}E`*F?Sh zNI3B9G&k;WNzPzYQoB&;oCQD#=QcI|=}5OoiPEIZn^nbo$RJ5zXt>w*g^`^1+Q5!m zygJJho#cNi3J--{rh_#%_T4(aA8KEu=vB>|OKMG4vfs$PUgwOdhDRBUUr2wc6~ z&0Um|ngL4tjw)OTF&e%C8NSEJo`*JpU%i0r zRDMw*0ocucubNr5KJy@j_`@=FK#xJ!cmVY15LDu@Ln3#DPk{oGD0PPcgz_5buiZ$Y zp5hN+hCQ%FW{7-r;5iYI^ny!gCEjR8kiE`(-Ft%c*TK4-q2x4fsGYw2&dvkZ$Re3; zBCElt+!2J-A>=qtl#(IV!(qERCgzMjPS@Z8XY`-W;bhU_6eyvDp+V0fIHM25;qV;skwemTrDl9QK3v+*$%k zRE&T;&U+}QbSqw&1k<}{+h^ifaHkvhX>5CzbH)P%9Z_r~8fVNq!JskIfShK)ypY7(cch>0)Vmb+RH z@EG*lI~Wr#=ri=lzwVaw&NU$;Bh~sOPFl|sGAY59txRYae}m`7QJoBcrJAlLzdR%> zQm5FoBbDzFp4p~A&l0ezQ!qx5L|hWnK7(nCy?N|3^wx|}uVbCFlSOsF^Z<3pYk<0~ zhyp5s1ZtqTZpy4&8fLa9p}Gg5B_u19DJL`wxjF0SNLE@3fY{EHY&~hW3h)%Bh*O*m z*;gi&RI0K>ufkU<_K7O9Q}U`(^oq`D-ve4p#5ADhHlBf-6>?k8avV)_8|;AA*I>L5 z6`CW(;U}dL!MriOyoobKL}%nWcjTF~y!o@dS(5zEPk(*PHw{-TfZ??dDjS0og_`rZ z;dBLgBQhOr`6o}1vvuW*C*|oU$lOu>(p>>?DDPhI%}^T6)w%hX8AHU2&IS~sU$zq{OUqultp z+?2G!T&TiQzrxzL!ZxSkeRqY!Mup>Zg)?cTt5Bu8ex;{xr3r75msWA$NPzbS?Vez< z%}8aCeif%omH$y?q%Tbf+J-`GOj)SD?0=O<)PKpNbHIPWnlT+o2bumm%jR#ac}w2< zzgRXnd~}_h|7qEbr2s>oU;Ue9gJvk8`EOWrjvDlzSo3-FnMT9kr9ib`nE%I8pnu)^ zwkR(HulOr>wHUT$@Ypj5`?Q|SYeblXRk{EPK=!}eK7X$G56ea+HsSxDWnOeue9;SGh)mF=kOukx zT`ADNZGGGRdnwRAx4xNHzWoHeWfAaQf>euJMol%9>BGgb}5&mUCaoXVely-Z)gfAf@^Z#dj_?CJ>Y zHq7&5k2+Nk$E`e3k2LpLC?ZBh0x(+2$SPKurXUXjDl1~nYiepp0kzi%aH@;Cww3o6 z^&d-+0Zh3`pD(Hgo=Iz(2C-i3NpdI9mNUwTza5)xq^!OUf8=*D<$aPEaj_N}@1{(Oq1neJlKX%ddr)Zk4{^zBCzN~{e* zN1v>6!wA8q2v--;_bEH+%lE54#$_mbYx_iQ4O$kTD)O$a=gxWWrjeq0o2_44O)W~L zBJ6#bGYKMV=uk*axm4EYA@oqyB)V+=Qe7XQ2{bNr__63I9PoNs{v%xFzL%?V%sKxggK)zJCl19DzB~U?or3)>hK5suUu~&`Im}bE?y7%*^l`sXTnA6k1H87 z_!Kc+L}c96SIJXZMC5mx)-D4S#QAdta11Y4#ENKz$%D&-suhAqg2X5@nZ42MzuDngN#A0}eE zenT9G6~TL7jEpo90SciGwX7;ZF6R^_{WUM?Ix$S~d>ti$Q3Am?8~H>{dK0b3M1z|- zAyW=*=Kt4V>;?c1U>K0{ulo-OdEgY%tn2@7j`Dxn9!9R{wj@!CIKlfWSnEtZL=~o3 z-lb;x1wp5kIdAF9^4&*1#6tf-3hM2t$_rG z;O+qe1cC++G=$Lee*4=qv!`lL)%+*_$#ZbmbG|NJs{~TEp!hLjnlHf0@Q*b7V@Sbz zSB?cdM2X0+g07~}k*!WiFJK_akij=^q9Zd=r@MMPr6+LwhZbGy*hgReQBg*%OJICK zy6V$iNfN${XkONtdB?K8_Qmo+!lv{_We!#r0XdtUr%qmedWW*P5s*NXS(7uu>Bw4U zc7ia;f!^0n`jniA9O<2MS9!7Eqoae237?5+aoo>#1G7vdty&W=EAf*m7kP0a2PHaR zFpC%e2nw5N0EJ$ia*Ik#eX|UL*roo(t!0#`)16Mf5mqO@6sU&cvfMmGqI%v^`P8Eu zUrG&f)lrw{aN%fVs^?UT0IIsme5D_e?-CoFM;DG zg^LGlo1Wvcbqdjiadp!~p*PSHt*6rNaCZumZOSdG2P4kA4T=#2C-jamV$Qe0z%mNwoLIdad zRJud=JJS;+3T=N=VBh!=E3I3sR@2wP{n(N|Y_bWayyK-%>;PpZtUp|&*c@t|-&6Xx zc(zd8v?m8jZVLldq%^0yV5%HR;y=sVM)>Q6WWT`M+CBE(wKWtFJsy5$xbwfDEB$qO zRe1T#cg4DZ!~MMu&0P2f@&`xa+{iFY;*M@xX8wZ1zlrCbIZrKktAKFpQra}~o%Upi zgvZmvX^L%caHoUW+24DCgfPH$im%16N_P?9x1tnqvO0{Zw0KtTrzofx=pxl@VrpxUE+_04Y%=0uCfLLXrjIE=PF) zfOnLw?-1%iH-+u7hij;HonVB`V2TO`di>@A=+>zQr%jK*=fpax(~&ndiPqBM>du=2aj|0>8t z|AkFf|3#)hc5zM)n?a|kq(LknZ>>Phm=TJ%2=kZP6RpR`b%L3I?e!q_F%X7crB@tb%RxU{v_RC}xdEraT(_tp05 zE_$^pvY(I)>R*pWnEk-i0|)EJ@i4kreiSzzG$8@GaB*t>fTA?a>i>fWTrAW@p}@P} z&6!`$?F1-gF|c>*!u2^(`NL5*5iwGoF~3eV``Kl67%gjRain>I2NMQDvpe{s&a4D> z@!n~z(0pWKH^SiG#4qk_q8pFkc`l9SjFaw=mUqXFoYen`J3xaxFQZ3}Wwej!_-d(2_IORee9Jtq5l7IHG zn#hnT9IvU;H;2xi^v2d2I4x0c=dAZe{KI+CNf&o_tskc7EE}?= zMN#`=fXE%A4bHXCx(%qt2-+?(0d)NO73}cGSqofukp1{uOf3jdwFE#2`sIMhTQVO> zeOH<{5H3e)GEaC_%3?Ym4FClJiC}l-crz#JA9|J|vt{zhZiY~K2bXK!bkyl4oiaB+ zEjQW^fyAX{QS~fW>iRlrIix^(4^PYe?o!P;TBw2;QG|A$R^_^1T+Oz0`xeweN#S@?Gv{=JA%Em-AKL$4nEa^@}~y+SJTeyl6W3t~97 z%b^FWC7eA@&v-syM&t-JNbd{g zcbKBw=govq)q9?qZbf2>2s}+m6uUZ`hS1)((4vOS9X>)n)7y_}iy95`=ZTlWd%?FFaX<4d&hGbqYK; zLNoJ$3CYqsy2 z@T02^Z=XZAKP>`Gi*WreLGxR_>+@B7)8B%wet541(n(g+knLFy#>3nGmzg<)xrj6n zEQ0#lK}Jy!qz`tPiS<(W>PDmpQs+fUj{$HCf}#dMYP&p@@AhM#b{koO#QfM!`$eaRv%|=RE0D!a)6AA6uU}5j8bll zvot1>DA-0xMiv~+L{H;$%l8s^mS)qPpq#6rR^q@G|@%Tnq z)C9h4Bp|vt+_lsQ+O3zv)bVtZ} zMxdicNh$$}ymBDOaoki5(N9Q14~)E`3`z%z8-_!pt!ab5)xgM8toE#vU2I{7`CzFA* zMq%cqVo%6yu45z`MQ^VsIg}9<@X7v!r4_EHaWaNou%^tBr35LZRFWpD%tL=8_@bX2 zqd4bkY5vXDoK5Jq;}>{4yw%oT4FgH z%Ra=A5oi;4Sz>t!%escUgG9JdOQn_QoWu|Cy-s9O#opjv#|YZGQ9^Q&rE}0T-6;=P z%GqdxO7k+5@e4O(_De~_tVvf;@zeA2i}Ld8AHfGo1-0EwLi-fzq6IzVh0*bENPvYy z|L@*z$^T#8?*C^U1+%n*pychby4E|EDD6{`KWx*<-u#~h7Jvj0lz z^k}`+Y^e@MJo|j0(`{|Cr~rQe&>L`x+s={NwA1H{j1_}k@^5uR_%+PSvR6mxgi+Kk zA(IQ4)1x+6?DkMS+LUWs&w#FW!HcRM6%$bH($!=$T%6Q)b^6WJ^kX!~)!n!6U;ZFD zLVIBgrv7#uEUF5DK3$zjsv?tXBF4q53+a*xMS52J<)|?TQqjIOzC5y|a$D8A$ zA5+UXTW(i1pD(`^bN2Tu|DM`YH1oNbLH!tpo^4CYvRM?9?sqRVP8Iyjv9M4ZB?e|;1R~ein z-5tJZijx>?C7$!GQHLs1J04_!(vGu@ERz&Upf+L{Oqt~3YkP@8JFvstaK1Qmy$UaD z3g_+LV;Ir46dO#ghB1%S$zG?dcA+R-(UqPUhj|R?E!_?^Io6o$)hX`$K}F>RB&c+X zqWHDPCQ-~MdqE}=XP{fa4DVsl+CpmpA>ITG^^t@6P` z=lY6rn967qN70WkBa`?Wf)(Rau!pCtBalY^%LM~M z{}jLqV=DVfB7lX5Z>l7~ptBHZV7uSv9U#k3p^m91mxwo}Q5fJ-jbZ9g&tq=d}OGPdF;0aOh&ETJENUKAw@u;^(y`$H<^8z{R@AeMK?QNpXVz znQt_sTWU+woh^a8xG?2(!S$9F*%07R`vc-jOIf0Q<@rDgy|{RAsJ(0_IF<+1SK@V( zaX0>plK9d|LUY7Gx3ydI>lJn|xMSEe{~EZ;v(;;WEc@%C@( zW-jW|XIp5NrMgR%Cq1CZUMNOs=1hn1c%f$Xy*Z-|h33%adU>Ldqc@jpV%+B17h$0y z!)@0{n7DfD?7g9VgPXY?6IQ$ph8Gq~u zDujDtkt^IyPrh%y!Kpr!ykt%Lv-@+^e=xAJ;Qa&TH@vV1N7Oz~6^2)P@KxJUnjsIu zz*oE9EHXZ^y^JMMZtbul-eESY?^GXHZc2k=yGN)ryt6+(?;p;JN)uVQg$_Uec)^AF z=epf~*?rzQ;tq1Neeo8KFO;sy6J1CsF3c=9J7OB+s}G_6T@w4eH&ZJ%_@EBuOB_1= zV(|!Q=XM2TS`)d7eu#bLn_nqal!M#zryy?jzM(#TjZ}BMpIady(XAAlY}kLPM$C_M z2^RqN%{!pee4H?)z8U29TBsd}0`isuku3)R6ayes-cqDWypUsE9`Hhe#a(!O$l$Mw zz}fPbd$pRO6QBMM+vPhCpPL2%bXdDCVoM;`S^Q;yK2qJCn5ey{nz zxksaY)_L_&lo{5C4TOTUQR-sDNL#=k*Me}Az@#{t)5b}_=LPD zM9rWxv>;|tFpCk5uptEB5sU*1CEjB>ZuR12g>Vf9qYDCb=Yy_~LMK_mgdBr~MWf`5 zq7=wD=KZ6{wdt5@qA2%*G>%=7q{DH^q9v?DQ5FJ~9iu^|p?YLN(&N$Kw4k@FF;R5k z>a1WEvKVWn(533=N77i88bY~w0MIcOM=Jz7HA3GpCR7xkb|}W#C?*6R8x!!algEVG zbkKjDJch^dSh1yw#%CJEzZ#6o&W+E9#TTx}7sKOA$e?AS&BZ6+jnkO}(^>P<*}Ky@H`2Ku(|O4=_{A~=jWdJ;GeimV zGQ_(xUT1rccsO&i>7HLznk%8Ep8LczQI)W&hM7bFQvl-Y{1Y2_9H49}*DeUU( z+K#Gu8dfszVex7!^#*O$OsV03@TPycfU2wY1EySrspeC{pA1d=MXMNdTN}($V z68=88css6GAvQ8*0@+=gt2GH5KN;*|mZ9~A%`Sx?z;KbQ zJPvG~V)~~0Fvl!kX@Ck;`E`MeOM{I*-dRDpGK<}01cMwDmnNU;o{X-L@TYE?&pRLs1Kb9DJ5<{uO<_+V`Z-t%ii;|B0+ePpvYS4lp)^8HZ{7|{pFDn z9fHfQQ^n(lSWn>jD>!iymnhDDnm~T{LzSc-Uhb6HkWH?Wui<>0@V07GoU^eQD(Fz; zkGXNy3W}>^{}@cdG0&4@U|Z8h&s_j{3)XZu@NYk=XU`>>PBN6a%_HJ!4h zj65gvu#ttKZg2#7)G2M0tNc}il8x5XuQ+qm*LJnMZN8o}x>cNutfd4ECsQ&D!7T;k zQ)c+2J_`7d6{B|z?}=uLPv5VZYg#5i_Y@8=fxnyipqs9V3-=W6DM$|#*7I)u+W}22 zm`I+=h?xa0pLCg+-6%ZwfW0^xjToE$)I55i-P^lKsV10?f_{WevHnln=?M%rAA`d{ zwh9ohpK24#Ct_U z;aouXmzOr(?|O<3y%rCsY z<0c4gVljgLD?X57{sZU}EnLl^64yNnf{ShSL_hw9*n-z*B!LDOrAzv`!S@~ zQ&uJpbmh$^w};W&B}0=9ilNJ-0~zy+ZU35&F&o>~Ai3YiuV<#{NjQ4XD} z>rvHYGQD7@uu_zA9^|eu5avf~qy(LqVV1is8mzE;{q<2JaMl~>I8h}-A%y^WE-rlp zRI;F4NHd!fwa_0D9la*4n~&MlEmlosRDFJ&aeuij`{p_X(vl1J?U|=LCJO@R+tL$K8rD5^bietPhWwc|}Af?bos4tfhXB{Hdq-uxnWMWVf$0^Da zEL!#~Qy1lwCLfMv(|JFU|OCrUkU&);lngY9b#5cxJgq^vt{NHn`qKiy^`X` zOoNkUMBeeKE$_ckz-^qeWH!r1dmt8vXh=r*?iz-grIXQIBEl_J%?fP{?dK|tV@xy> zQbwnbcpf&UytoCl!wJ%bY`TM8?h`R{#*c?rjM@E3tc&}H4>?FQ5^|_-nJ0rYqQYb8 z4U_LES+#0dC~TE~q<5Cf5N9l?x%+ctG$wDaF%J@*sE)Nx)1nFTYr>7s%u*WD z^Kf5reCfC)m`OMKle;SR+J^f3=M%D{{F!e1@kYA+GXv7(4NkuIR7vXR19ZJJI~IP; zlKZYs4h!4wM>-nkrd_oBO*f8ByA)&8FG4-$H$Sp_A`{mt%?LHIPgi9qMLiS8AWLz{ z_`!ywe-xy0T7fZ3QcN+Q)9Ap`=&D*sCj)Dk3Y$?lsu|2;463+&WvO`p{-1K>mi#&F zSR+68DU`Hc%b-Jik%9|Zqh(N&h}nJ`%frr*kdAil`V%B0Fmq>CpSx?gCO=Yky4 zXJcvtzA#ia2QGVYgUil8pJRR%R&~Goot}&I!JI{D8I9i`GOsL7PcHNA_QGd9?aMI= zU+=8X3}I<=$R!OM%Kf>j zQMXX?7Dc>cGK*ue(Nf5rKq9UHb^gV(HK*0IvFHV1~fD{3C@KwR0uLn%gKH(mVqx~;8>%n3o%in>#2mT=bf zX6D>>DOnDuq!0iJxs)w6-=XT{r<8p91_7#(ZMZbLk!3g#-6*`-p*ZYmqPjR36B-TD zpA1}NMc)c!A)ymJb*RX^IB~4Z*kOZ6)w4kt$&hVTs!OudotS^4U!E7R-| z>>PE~m~|rnrG`moC2w$9VV$&0>M zKy8=ekO&}eR|WltuTpq~Y8QZ`p-EMqFBMZ)#mGk-dx4ZB3ZQJ(xfoFc^zx`z^iRTcV59{oxO-EAf2%FB`X>XpQ z@8)MwI_~0+4~n6;$Bi_4pHB5iw^c`wUzz^_6wht?4xB8$^h%Z3&pUKY(d-1*&MYNF zqki$-e7Z~${&ikiVwecI{))w@Ktyhv27LTk{qE)Iv}>*pb-8f_c2<$t5<}*){;wMq z4|52_OQaOn7Reh(Y8j^h-7qTwYyeqY3xb`v3nWDcNU;74YNVWER%*fEcWsWOxuZps zhYf(;o1*}^->>u22he(&qj`h(fI!b!)AcLT&KfJAFpC1DUE@U24aNu=%uBUpqwvP zN$!c|=*mC^Arjlf0AFh&41W1ZE3f46ElV(hbzEw7Az78PoIWo=!RxoS=bKC$G^WR4 z;hWUtPV^tRq!Y>lPu%ZPK2wmJm5*||r=?c6Gt%M3kE8y)&JvJgphp=}P}#gm_pxQ7 zx0;xeG-(KDA=;zW&r{LG6-el5XW)_87>|g$aZ4LHWUMrv5M@@#=#bb2WIw9Bn{CY; z!nWm;M<20eZ^<~0vZlZ&lXQM_6zA!rIP8$#S z@(U!*xo(jR=;f*5f(i=HqKI?n0Z@JjkaUJsfzL!c*}lD;$PEd_9WodddQOE5V?}1) z1fc4LmvP(4pjbaG=1<;L5wk`Rk~E9TksMcYuGDBns^}%O>Qy`#Rew!BN;VB%uH=Yz zR)s*93Jlt-#Lb5|QEqTkgiA1}aq8tqF&CROc8S4I%rpVip)&k6#h;w(gnP43LpJsY)EWe z)bhU0x51InGsvhuo215nyUvI$RU4CYutPqnL7vSrbrzvAO~q!%sleu$61ElD5r@;L zZ}((Wv9W4L9de@ZK_VJHV$i1GHUtt?T?c}JpP+UT9xRqAVc0n>aSc+=4&(;x!Nldj z^afWaiFKs?)plaXA=TcJD5jCGWxIdfy}nWQHKAs8eEu67g78{CqJH1g|C(AQetxm5 zFB9|2m#;(`Sh`oqq-iK)Jjl%T76bf{FTvQ~BdFBE^ZJMe;OZ@&qrDoH_oO`*IJFRg z1j`EjhaQf~Gc`&xE8?&m*w?Icqql0tc>cLxq~AEXb3AY2$qMyr?Rcbr)h{vpj9l^R zp2N7qb%DZzJGiOY-4jp3aCDKVx>jDrE^+-( zDFx6{>vi!$DPF-1%xw5vI^7Bz{EEJlw8jnHo)3s?E^w5aG{=Gjh(+(Qd0VMpxj@9yO1C7Ai%)rbGFj%bP$|Z=}SO>Q_YHcZRssmZFRC{{BPU&#%<01%C|U zNT$tdv25U9(gCH|lon8R+K28dazlP`Z786b6hA~eO^>00)Au6}XotRHUU%7k<#G49 zi3$r06st&$Z4|ZMFwDv zspfL)fXG{I2O=OXKgP)DRg2Zg)edlC+>)gMR$i#sbpH&nB%s{MQk8CbOsdJy;xr#a zwgyg2=dYzS{j;&M($&GWX4RjJl0BlQKYG$~X6IR8(;Kv|icJl#yH9seAQB?Vn|H?t z-|J}*c=)BHgKL{zA2QXcB^>9{Y9#-gC3QNzD` ztz*ClkhbN?ECOC?6^a8BGSAeO1s4_B^ArNDLbRWXI}cjtBo4=^(!f=Y(0&feu3kl^$+gjT#Vf&zSo` zgV8j8IeNZ0qBn5E6e`CouKm&LhSVz6RoKBv(eu>EvNZ4s^{^W6x7c37r-Q2NMUp=*sS z*HU4PPV>og2;Bt$L51s3@Jj1k3B{1u#HC$RP$Udqk5PqfU5}a2?y6?CYP<%J2BgWv zXmfv6i~08wA;9_~kte)vix|L@vwy=}flWnBE=SpLHzS6(3zZ`H^BN2gG7Pz05kmG< zoYNs}M|sD?VBWs2z?5{iA(sZ4R44hKbq*K9&P5T{3ERkCzkE)@n?id|LwyLse@3B& z!LQbTe189HvK~9|^0RK=hl@f0P4Lu4^`-am*OX$B+c^=ukV_uTkNZE_-)Vl{|5p8) z@4$-k0Vya~RWo6NbyD%Q%0DU2;uU))v(Q&j5-ckkPR04pgJ zvUz21=u%;!EhEL!#4O6zkg=~XCV!r?rd=lr!2RHmVoek==_Es#L>R|nZi10z)Yn?# z`WVk#o<0wVN-vai1c=d?X!yNZF2nJ3_~j|6`C-*3Pb%Pt;G?p77(WCF88)2R-mT@~ zep{K5Su9$~ra{kZS#@^6q=0jx0a;YISbZ)KO z^}R9$fM)=TB%jl%JdCpzTU9Mq2Xt>4l@X8A_vvJP$f-@nXA)D!-nT5pUNP0aSi2EMdGhMbsrAaK7xFqohS685cq)4cYj}&W-%T>kr`J2s-Ku?K%d7Iz$l-;hCM}o>xalCp{h=Hs5ed8AD=uzVK=p4UU!RO9Cwpr zkB@a#uMrv@k2z6ABH~iujJS?B_hn^W+6W)kpqSh%I)`R^%l7}(kIAV&eK*$`&Ch8C z>?|4r>3liGsCi1MD@H=GmOL}~))|5#eWxq)1x4qN_;To(Q1it6nMQ6Ip*^o~Yfh1? zfbZaj-LVY0IQ}nZ)n3B;Z_PTXv7G?F59_-Z_b&=o0^V!CNWb;wL=Medc=YzJ6@v4P z?Z;wrTOf|*T5X}fi!QaBBZ$!Fh0GV1PWq#JVoRAv@xC@opL1z^6MjT#3fp$GGJJNG zWRxd|W)qh_+Oji#Hd_d`euL&ku}6mR_wVKDijHAO4*6idFsZL`Yf1VXxT~B0w@}`` zq9QngWJ~7GAjgYX{Pi~9T*`=x24=#X+4^S&k!CE>fx3curNTGxohS>R{9ZG!%ju4w zi(z*myj>ax7tT$GWHCylkZ=i+yO_c*6tS394@7E!W7x&7CetLpQ@{DeYHn9C`i5Rx zn8cFeu>>&x*QtLD`lQHUw7TowxdH|k%l(dfeP66aXYL! zKt9+*e8<%EAR-! z_bY+a4puxj>onCj2bx8{KKb_|o}oX3e$0eKEFw(t*4{mC(}bAy`D0$mYlKa0^q)`9 zng39Q2eW$!J_M6G>#V(BGn6ZFVBF0K-(PjQYZGj!7JJ=A*9HC%EJ6}$1F;Qf!{kL$&MqHhTqpagq$D;U{_I-7)BF0L<^I;%5*uJp##jA1OfR`M}x>IKa9N>ug62)r2ip3P3=J3L|h5Vy1= zvtk1Mw}g3mU2+e=qP>1D8BSJ*1c2)k8(VsCb?$a zybEkX{|GfX!IghZCGyH(j!oH)Gu!%&r?rt}8ILdaNc6fp^&CB&OgRf(Ig0bpU ze~b^pNZx_P>x!j^*Jc?{WL?GKJLS=I(PmS0=Q@fJxSSA89VeQw<)zgoM;m1>A7|lM z<=i`_QBLHPoL~YUb0XMsAZ$tdM(K`T__0pjiI4dMj#v%G_$hfr=_lD38yS`H$v7KX zTz>kY(O?cI#@YS6l8vNb3{pynVn;j?bzpu~ch*I2w!I;E&5`XZazPVY@~loS5k5ZW zu#GliG5b_0Z`v_0{+Kv?53JHIt82pu+$9Fm=671+b%Imjfmrv=$$!XG|JHtdKEb@7 z$TOEk<&?#|-oQlWC8k*;#1aSKJb8gBJaA1gk+o6Hv`~vXNs^a}sk8~D2TMB{Z19}* z2{DPyyvo?~%j5$|Z3a-;@)QRjQ*iOq1fDo$_J~QIELR=%2gwOS4~fly0?n0qf|Fp8 zO`&@f-B;R~p;;A=NFYTM0@GD|)%^4}u}V~6|!)teIf4Lbbgl2-o(=ZR3u5zdq?}Qc76xWZNaL<@uJvI}l*6mE-KZ&>a zVWv!&w9IW59^F>CeyY5MwsL-J<^I#kOWCF?#`j-Sy#Hc^;eeq3 z_Vg04;(sziRFCbU|48wI7$W|&dZ_6AAF6)@<-32Uc=?$i>sxuLM1VLXku(AfU)=?u zLJ}%*Hyd!STEIpoL}iS%w$8TEWK_I_UG9?!ig=MYF(AC&5rR9i!AqYifn1SZ^mto6 z(1UV)O4aPoy#ccScSqSWDb(!=6rM%6{#B*}w?W+4QP;Ge;PQ1rhk(gk$=d{G%Z!;Nar8Ry>`_?j9g0^2X=2Y|H9H9`I=W1Ss-8n}VoJ|BC~NakF( zY&~}S+cJccf~2lbX-z*OvR)UwzSzv{K#IfdJAD7^N&@S0r?0~0C;X#`A#PESG~##Q z9km}Q**#rdtayW>yyoP^ICHZ?y@5swM&X^>> zBf1x9IOfO1qM|X8mLb&#Pe|ksnn=~AOOHFqZ$cR;_Gimy!_U>UAhwgXn+yc!3tKBg zv%=nt9~K1Z$T{R{?c7j>MAJvv$2&uhU$MB?t=GsFCPg>~0}?l^oOA9+4B5(?`H0xu zQgtxv>RNl8>zZ&8hh#ZjMikC6HdiQ=q;2Bk7)^bs-#FOJe1kR>5_j?q0T8}3PQaW3 z*01*IeBy!XNp_+E05))a!lpzBb8RmCH-ze+yQSU`TD4y8u=An(^yS!r={3G% z@9x6Z@|@!tcDT5k;`4ujAC;!;0FbwJyyo6McxvCkp%Q()Qq8KIy**vBMSmVaEf&Bl z8A^}-9}^<^p5N^szK1S9_GZp*{cLK$S9e6?%zxY!=|uc(viE%4ynxyF?-P4Kj=p_F z*c%8xpfqbEZ&?6G<=+sj&P9FgrAC$*A41#N$K{K++-+TJ#^k>U9eymyf$@g>s%!dC zMj4?Xh+>2b20Z}4-Ax$vDql4)#Y%3R6qW9Q5R=mi$?)?&QS-PwmtR_fY2g}4mt%{( zv^&u}zbwo0_?S>nOOh)m$}6!26mIkZVsY2y(qi9g;Ch`UI+-T9Mpo7Y*qcb!Dzb`1 z8CBi*0IU1>%E@5Z)Ds({7>jHvjecKI!^XNJ~IFB)=NnamhPJOyV{Y&s!Ofap%n09%V!)IH?NQ)6;FKAQ-jK_DF3wuXDzXR!^jODJ{9*kqn&`d4J$9uVJN(8>a>Gsn!gW5nK`?-m1Rg zNw61W7LZk*?ycunB2-bofw#7)7vo2qKj*zA4U~85Y%P#)KSCG@AY8hF)2Ze5gqpwBp~^9`d)R} z^$!ylS*ltiSF>|>?-MtTi_X3;zAbnNK6-QX$fd@2#lLR<69r$nj%Vejs&{L*ixwtf ziTgGUKzjt96Psib3p)7Rhfu7ZAJ-R_ zbAPs_(Zq$L7z;E~f+oHC{024u2Iq1`N;Tu&8}4pj1f>LXjJ?4$P!jzriuJs=x^q+0 zMHR~7q&}~hB6@XLbdlN*+y9en5cBr$hq*tBM_BJLl~km6Uws!FH1|r`vB6Cdduw)e zM*V)jH}WQz#?7a?eXC0LmCrr+^%0)MbC|Og;9@=SzQsLg8Ry;E2x%W;*Sr*7`Mu$B z3GF)%yU1UNDuSfr?fx(Fx%UYL7SC|J{_7teQ*)*LC+F3E7Z13dS7!dQs3uxo$vGbq z9|A@nLhwXG>(SmK9(*^}LJO+Buh5j&y1b|!-an#2_WgrBj=~yQ!pV$6Ud4rA<%W>O z2H~TJ!jA$D6(KAEko;=MGHKYAKp6EANu6c-IQ$%U911qUo+U<_h3cz(dq{s0U&P9&y{)fF2Si5~3=eAtb-^u()5 z_4>Bx5gZ*$(8Bd13_x$DeRYJXD-svM8rQjt8=D&!2aAJ}#a<$K(PaVg>16SlqVd^A z|9nAgKBT*RkZ^6xL1C3PdT0e1v{VTL$O}@VMNd-#MM{C{;LtizQ2RKj)iI$tE}_yj!%KZKr#c#U@zj|*9>Fhinqt9ce!aRHE9d$DM;E8`GXk$ zgR-}Zii6wMHVcPBQn)9$yK8VL+}&M*ySuwnxVyU(+!Hi71h?QWNvij^&*|Og^!WRt zuh-SOSz|rpo%4MhobeMz6Bf&j>G=&E7L#AWi7d8>m?)Sr5$=B`Y=xk=iDj4xPXuY` z{0Z{G3D~BA$1+%q+)BHUbolJpgXp+#bct+d$=~SU^Ff%=8VtCEpRVwz8oJ^keDLz3 zakHlxuV7?Yvh>Sllv*VFt+h7qjEk^IVTMvnqWJ&o;|gcomu|0$6W21?VA z1yQW)^;prqR6$C7KE#P;)=$Nw24_Att97|p`e#z+*(M$$Cx`<+o1P`RpJi2;;BcQM z)^?{Y>%{C*X3X;=r*xCqr`U`hTlK%DD%=R~pQZ)c=1%LX&qBF!4RsTQbqhYA6pR<^ zGL=)3w_7hQ6S1zR-|H4ilEF`)K=!J?06_VtXZf>T@Y})UKV{`lMEZ3oIh3PO^R_Pol7A!zKO#r}fLt;_R;3bH z1TS9%4+Q1-PFMQnR0eHS$`w^Uu>qdRBn)uzS2uwt4&oXjvIRSmigS?TkykKv?ay+1!;qXH z0)tRZIg?0p#47))82Vjby)@hUGheJF2HLV8b(g0Gk&6u&&9&++4fLuFPaBQJp^blu znqUQ+;J;uE=rth=HUV;*&^DVe;+l}&o3I3%arB$<{F(`Jn~8dxNj955zBiK*w@?VS zQ0ceO__fgGwlMVmn+SikpuV?o2)1(RxAK_iW3&6U3iP%LZMJ@XZxtnO6Blfg)Nhma zYXj%D$@R7=Y_=)Ax2X`fs|mJi=(lV6wd>@z>-Dx9Y_=P{x0?`mmxc`5R#4`v@+WvnPWA%Z@$*NN+)oTpS z=J95-NL2CC&ysQe?fXUZS)}@wFB(SSPTzJq8^RF;Z0mL>qpy%I#UN?N4Av;lg9VXs z#wtpLly+ei$t5$vRB7<*_A|84YR7+#ZW+tgB%|fXn^xJDPpi{uf}Il0Cr+0i8G-xB zpk%-{$UH272?B#eYb{~3IF-j|b!!LVlid)(EhZ>%G?nV5M)f`1zZfcMk^M=0Emx>D zquIpzExh2i>#{X;gnzq7kZU4pB#ZVw%`BlqB%=lddLC(3%H<~-0$?wX;RgkwKfw$z zX$%{=Hgn&Ny4(>*N+*oT0)K-~w>Ezn&x!Qdhqy#oe*(jF(f;(iC9E*#l5#J29gG2(jE_G7I~muOcZ(Edz_NXyjWWA1-}3Bk`HF|wAhKc!(8~7 zh$^6C4#sxH-^5a9a>)-QaHQK6W%#SJlt9fC9i$U8{KROr5wLy$<$C-{`b^fMWdhwTr_@jQt;|-y3>=Ao&04{7uw`0&|^G z^|KU_BhPH;Lx%WZ+?wWA@ECrioPH~a7{1=a?Jwfpo@t`y8BuFliA}wdM^ba0LFbt3SYxrHDXbL_L6H+MdVuqp$1<^i-ah^8H?T# z<@j;MI11fU2cSW`!>au$900Je57j1gEX48JG*}>A*|rboeqcPsgVwdL1&1+vuK~82 zspf6rirTl77ru3E896P!Y~`jWcy1ume||X*fivQP@p`zlKeCs{cs)z3RO&VXthw^% zLhv7WPujeUx|rL@2kz){I0LiJ9MbOI56imH{f`=I8FmW#hS?r>29Q zivjJgn@38)ai}}5IFQeXt|cyZFebl`Wd?!btbuU!q|l~Sk%@jMvYWVVz%T9IO;cy4 z<1a;rcwNJYpXURx^2kt_30OmuKbXDU7h-5WrvpY+f;k@wP@wE!f&FF`ZwASIIAtuc zwg?p;$~qI211X%}rgpw0vu1}+S;9oZE|%0WW}sT@L2L#Vm{|G$>J1LO3BPkIB{`|; zvdebxxBW6X~79G;4ZS_qsI%IPdT>a7&%E4N6!h77QWYMbBU`K-fA%juQmf0hz+2NGVQ&-Bi z-_&{GvI<2nEf72us|$NuOaqGhY{b1I=|mL87`oP+HL7F%qB}fWwNaYpSsJN^Zn|>& z@-z%tpxVk5flEzlt`Z?K{oLYr@CG$Gx+%@qqG9OVzph>gBBzCjQi2Sy(gCULlT^yx zpvcgeM34t12)-N7ujh{G-d#>QYgn+BE*0R5b3}3Ww9sk)Ak$GrC+sl1n9@-fLT3e-=;LPf@ zfNm_tN@1wwk+I-8=6J(o13PY>&Apt~9Mre0LR+;~O)7i6V5_~&Io!d-jCfX{Umaw#piA9Sc?0io|lqvftc+7A4MH1>%z3Q>q)@X;{7X z4>rH$WZqs%9Gb7^h#avw* zPSMuhJIvH^-$?V>^f)coq*S@aF)4n1fVJgE9B`dy24&}K_1tzL>Q7GR;R0i7S< zed|KYt(n~{%{h*EU5uu@8=dA-nD%a4yZ1>bSHWKJ9`cjyk=J5>%r4A$eLuV6BFO+r zK(@a<|Jcy$_r|yu_uxtFSup-q$}*Iutppe2ScUhw00V`CiQ0(R)tvC4{i8cL2iqyH z>pkfW@EpOpQAuZ z?uofc9M#+z{FyRaIVl@Gjjwuz)xgw(QeN<+{iB4=TV7Tl7g3SqqXYV0V4A*rzpn{x z^*%OK2IPV`mh=gkrTiB}P#VlAy+gJyD|b2GnvE4oFie)ohP;4X{1C?1e~#kfOVjlE zu?qi><%I;Ih!adgJ;3ia)LMv=rrA|oXBZSs7EX>P@;E^&ewX#t_XpAtGnOBNDR|GC zU2efm0W3_e@WJng7xaDSc>oH#t3c3Ly70^_w1f9)l?Ol&1lpV%xI6mr_bo~UxqMeL z@q@#sUrD!AS5dx_H@Ck$_5Gcm_x?7#u9O7W$N8!8GEh0tv)D6~wRdC$@?BDTCZp8BsYAadQ&FhNNXr z9%`-;+C=MHBI7IB6~;sm04FYoDlNn{DHWg*K?3%DzQOSE{$M{D_D&n=CKgF)BLX)W zU;~K?oBy)Y87VaxiU#eXP$v^X!wHdfmmC=pB$$MkAPdqe3FG&U5Jrv;Cy(ZY2Y{;r zqP#!IIY%&70UQa)$}Wv5kCW`b{RSRZ>+W zf-=Z{536xc&v9hP#$h_*mC=!hT_IDG0kZ_;DODn0CDhdUV#-7lRtS8-SA2D*=Z=*89v!^^I6iW zSoOh#4bG_k2f)Ug>?^jM&YSEqpPZWPoGWM4tK_JPd6bsypw=mrR#}vMliaTA-0pRh zbWkoV0ANsL9}e0CYwfblaD=9RqJ?7|lmNCPOJ##SW97AKlK z$NwIP=$nHiSA`|)E$&Y`oPgqmbz0jzXm(orIR)wR&YzY)%~!B)w)4NEH2lv4{+1}J$G-#Z3* zT*Itg6XsoAkx$W1b`$!S`ek6VF@1CPTSkPG2I*}*0Sr7pald(B?nW#2{bW zV3Os5k`;TO8JR0hO(T*@E11ySRS1JExd413>+YDx_$T|`#mLI-7Td@%Iqz-&^gxK= z0RVUylO9QIc-QJ4$*`Wv-kwHO_((5VSoWHO1FRAtR|&(%k{-8&4wP-XhwPD z3wFIIu+C)Q;@1EVdwCKuIN6j>jM7n|#U3R`?%CH~()){~yV)P}lzv(qqfeT!F96 zWm8F=N;OR2R7*(?M)T=BAJ?+UOfI)+h3b3rnL;E}AdWyVT4e#PixGUNJ0(2;4xB=< zq_R7itcheLZ09jp0k0^Zs;YO3H`j#AEB45cVW-gn-z`T9PQ}q~Bc2#QU`W2!4;8)I z?MKE9-h+w2WK^2A7I%6Q*~%fGJ<&`5|Z=&@z`(UGfC``7N5wskC^ zVwy*|m*;X*9cF7!Nc2W@rxCRV;oBLPZSS~pmXcu}zb2qQOg=uLx99Sg1t5=G5YY7y z8YvS8WmVkZdc8F-hTgHvKYHD)9q9Ac=k|-f4@K2F4nVn}M^)9&I}*!N*1HZH+Y{Tt zh!I{pVaHNBJK^Vh6T9X}R^XqGu&bO~MQS6Q( z7rhgHAgfs{c4fG>A9w%5Yd${ZOvcRhJGCq@xU9NsUreNjjyfJ#@Mw|1Ck`nO7yP_V zmW%{kCl-^$s4$NCOHP=RN-uqKl%+l{TQ0hlv7Q}PwO(bJsN8O4oP;KgLK8|tWI?MZ zUBO>cGFd);=;k;en_T3om1W|E>Kk+7qI*1bW@9MF04fhV)~bryHl#NTies5Lbw!bc zGSDlOX$fqW!!1|aTJ=52Wfuw=o&9Rcy39O_LgRtYshw>&V`_D9!K$rF!z#;?l3O=B z%79mV#FrOsXHcmNR?}ZBM%$6f>IvAUUNt<;@muJ)TlsR`|8iV=J@EVK*Y#Hb64T8f zJih(S5HfAu%`h78#mxvvhUs<`=PRMp7y&dFckF=!O?Mo_NyudaH3*GkghF1SWtgV0 z&UqTM@#1dQrk!hAjF-_}LxQc=-toKrp}{;|;OisvH^z>1O-)CX?GX}&~Qa0`=;XBvZ-PheYn zRcAn1iT{FXt&6wd%8{$abE0O)UfK2!?S9h2-y2jVt{B_}dh^ovKtF;$Q2ieKx~4vV z586BQF_ik0-Y1XvXtWG4QZH0}3RtlG3rN3E9(=zp7^l?__{T+Ele9?*>Z(y z>d>;Pa+?Dfk{N`d!0?|?LutlC4h8^Bk;VV@bEN`@i)kdIgp-LQ@>UIDSj~rGlZ3CP%67lB2>^289jMM3d^MBRtRK4G&w!2I64>N>roRZr&0!nJH1?FB1ykHlsXv zw+R3+1@;7Zf?vxS=9w-XR{e!};Kn5&+NFf_7Z^-|3=`zOaR3UTAigJLPn5?g!*U~# zl~YrTg~0BMH%yLs+AgH2R?py5%SsCtG^gVU9TIK72={m0rhMEYM?jSY3zeKxsn8QC zsCn0mB*vzPCGU|%Wy?t(Pg6CR9vFw;O@AIhNNs^9htDEImQP%WQ7bEiKOJB@EcQsgKCb{;SQ_l$+6`V7Z&pbAw@~Qhpf)yDzy+Zx-4m zi?YX6>3PT?DzvcfIX|Y5sEcJ&cJ1L@A3KfGE+2F7ovbX%--A!X@pQ-z^l71yu5{RU z)9rb>N&*~W`TWd>l1aL%K19uQtJvq9M+C#g<}BVEe_R4Qwx6soOQ>2OEA+3l0>iqXTTP1#6+DMChPRX-@MR%fP+ zMq#RLC13;S5__VD%`wu3CO%u)((@-LvsQLRrYcydrHxbV%>J@}EX z=c%3|au;|I+VS4EKqj0ERM%<*%4LV86e)1q(<7mbigEXgk;82@@usL2Y;Jcu{-s6P z5wbi7@ntQ*mjCX%FQn_qs!MboeKVwjcoc#X#lye zS+G7)!x(HCN|?FAZBzwvwrYoq&s1H-SHyTY)p>H4TgBF7I&$|HA_q$G;kJ{T^E zG!f!$0r-ZD&SAOWv%G?1AWAvWOq>erz>VFopD?3lPLNA*;|Gc{5 z2`>?k)tt`-(C9a89(_ggV{a9B4L1G^l>zU+?f(iq1K%wt%(vcjtFTZ4*2F2TmHOgW z9o8>sBy2ge2#-h4iB7*s%i@?zarxEz7 zK{Cxuhof{Am4TG7xMGuG8Q`cEg1CA%euEA!U3a^sp%`Vph?1AMFq>!wc6{2lz_>1# z2IRPu$;j!+$h6b=bZ^x9WT_KVtac7M6}pH*od}AquwsJPVQ)Vbw^-?Bs`yYExncT- zs<6ycjI%>!F`Jk>y1053L>A4UJ2%X>;h6QkI30+=Uw4h41krnwRuW^eN~_)p1mTV^ zz;g=`_2tA?4V^8aozL@F!3*NR4Q`J{bc4MU4~0U1`>IoQ;cm)RR*z7meG2KExFju5L?Y7aTeAe8h#%&cjg&F*%;}@#>G5 z7%%zhd%=-^$P3WI0rrs0pzb_&dPu#ADBNvmcqu_?C`d?*P!s@o{s+=S7|$^No1j!a zz}TVwr7uFVD?;!sY5)}dn!JE}SKl z0@$H@NO2>MQD@*#ZU8tLG`56`rv>s-)m^s)l@1~@N`6j|0{NaY)bW=UtCBDw6w-w- z4Y}l+k{ekBn+hOmx>E{xP-5{2Bs#xla?4?X-`M!xK#d)I$*VSaL%2?|K)srB!#6=5 zA3ODOoa<*Uyr~M}6pn6N%pG+vTsL{n5FzI&@%0IDl-odeIc%>DOkW1<|IEt5{zF9i zfG_~~-?Op^|4l^jvgH4VhK&DznvVbS`TSdV`d>aDx%QeLsT~a!edX<@urLaU=!BAz zOuFLcF(3wg9_IB#kypdNSlX`fDzi-`l@LvGjMjoP1B! zpNlZ`O%L8JO!L}N%;FtdaxP7hVqSCD>i&*0T_K!3zd4ZZ)C$EF7wWw}Gb2hGO^c3L09rd#8?hcdg&iOBtSOY&P#Jk|^TM&ChdYLD9yjwMF| z$+hl5nap)cBYa7sxF)1oCH*+=xn&WtOB)_4_aZ|WOm8Grq-%K@Wu%{Hn3eC@s#>M3 zBl$pE<^cV{N9MXXb!y@dPfw^B5Q@nUpAu%GOM%Jtmu@g&h1-T8?B*L%ak!WxpM^I` zB5k=AnE?GsjFXa-rMH$NN{$U?PSlEST}V}qGwjsKR(OPu_FjoSqBNe5(dUUW>olOX zji=Z7`f-8B4-{N{lf9}9y)qsxhN7}*j~Qk!i=%4wTv8}MBuD0WMKV)dlHp&>%(VBb zae(EfwrPk#^&$e5j(rSkSZPpGOP9Y=7m)A1FH2kDt_JM*VK+0?QSPkf(6y^Jj6e<3 zm%7Gk1S<+8ydFor?z>x&XC*96h6yKK=*YIEU zp5uVF>5Q-6VH9E)j7Ac!w{)$)p{#Lcp>r>*`9T%9HMHCk+NZ)lTyfK5MhWvc`4(Qj z^adbJdd~Hw{_dE}oH67&hCSGR*hzBokDEx@xAfT4BQ6j)siyrrzhBOcA+TYG%I0HR zrqS?v2X`tq&&bYtlZr1CvN3@BN=l6x`FaKV%?%pq4iL;4#4q(W(&t( zG>K@|Q%cS1201W#A)y;MeGF%O#s>Vw6Bmt4E_OXW*kmm_5-AA3WJE?86w`gAJ;2h`CS|6Uj6;Tz@Ia0xTLpfIYk zkqIROu?Aaw!b;K^{rn~wAFgmndNMg_*z6L97jmfjF(z5y`$DQrpFaLAWK@g;WGqEB zLLiPZ89K^NXQn_&q+Kl+OnxVGlUgczJ}E1;lB%twacBq+P9`sLPR-*ohq>i#G%Z+2 z>A9GuXS7N#J#&rzzD&l|dLm210H4;XP{fm#D#taRnlvP2enc_$nTVyu0-;Zj?#sHg z0Zl8}#R0XL-o!WO75T6Sy)s669w>s9RV!JTeF(ARIVJL(xk8CqCK;!_Ht*C9(xHez@i8Ifjf3Ags;eAGaB7)V-4 zs*sri@!5Cpj<|d~DjJoE)9>KN`;;{_8#xp`DT5X+&*AVY1x;OzJUWBoH6YSoNwASW3_5OWFeAW96FQM0O>7i~q+k=( zMZ_R0t{NxiUA}~a@ehbaVIJ`pJTm945V;Yduh(@UtzM_FO=^#>6Ib%IgHS>3TadV$urrZ9t7 zf}_2r9v&}vpN<=&T2XiJKJrnm890$+_wD1g5`2BN(+Rw6H-`Q2GS0)&BnN^hA-=7} zi5T7Dogs=0QGYO6MY#28XEeHFd7V-^c4U-JHG3XjoqibO@T(UxpSZ*y{ey5H>BMBk zj{h=>%wiEn{!inrE);-awLLh%dy=-1;00J?^Uf#Pr+N1o-H~dQ)%b0OPb`o;y;-r4c`)_E>I$J&{!M;ci#P@CaRjM`^{`JG-tyjj_9L|FJ=Ch1i&^*LUnO z!K0*C!JPVCd$yxV#+#Jc1g?;32+it7x% zj{v8#&5L+&)vM5j{*Co0a$`qLYyFY9FY!+(^0SBqvxU^=#Z2eZbyHCB9^1WM02TdF zZl|Jsw1m(w2%5N?a@5(qQ?0;*^3!qX!{~_(Z69eU3raC6#T9e3{#Ham``BF6fqSon z*yRBG{{V-b-P5eUZkdL0_=ZW5$u-6LYW5$`Q*BMELtt zJL=l(Uqj!?&3OABUfy+}pw2-0_Qf~h-VHwtr*GJQP4hVYx5Q-pH@=}imkEBpszY#H zG8Xm;IOsh8Vc9*8Q{kbJ=DWP|_bu{gLh9pA=RhT^U%&*}!j#H?<#6=kE3|Bf>joK0nDR%fiCyv>^4acRyNl3it%E?Lt}NewGr> zKl!ZU`SNz!l@9twcLoox_#YnwyB__q){GaCumDX6aAn*C(E%iU7Fo~Y1yT}i+@ahW zp*$AAoo9=E8Se#YUn)#yeNmvfi3eAgEej`t7@LGB2p~iNFGK+N{1QfS8rC`%)`|-c zQN>T8A=(-XN2CQPB1d#mJE%?4mdgO^n!NPWC8w-yK$L(zEakj;gmlqJt4Rc+DsCH{ zAc`=)eJCYIANTda7U|qI02R`z7$%s$~+Y) z9*Q8hVXCb=_Spdeht5?fWfIeWGOC|T7lj-kT7+|JK~lD6=Aglq054ls#fH`<77$^D z^b|YK&51A+17{kS=nd-72@!z-dq;r$s6qBcVi=Tw__0{j5yYvk_;F;twwHKNNNh${ zT(?eaxlY2kcij4Oq^z5$%AWyUOUKg9*zwNL zdCi!c%$PsTFdFu%`7)fH?K~!t80_n!d>dqF2$u6kI%_qze zSr$sp`(jtoSGsY?#}!Es?58dH@A-Ul|P@CJby(uZ&Ws8T^IMqE9GW((i*{f{^42r5k2lt zp8{6)g7ftPxTyjxOc3CII{j1u;Q+P&wOEJ9i$WFFWGWT`%g_HGSP*)dSsDKqEQDGR zq{d68{0A2B$%p@0tP4)`vOnMAPq7#U2!(}`V7m;`#h0PcAl!W!EQIL{m#dF90R)GN za$6+`wEvPUl1IeEl=r+?BM+OeQJnF-)IoW6_09KlM#gTIh(2VVS-mLgwV8`m6zyaa zD~X@R;^>9gn{USH|al80lvwyaO{W@1#gU*C{e)<6h?&sWlLNa#(#fK^~gXI9; zAC0#$A4o$FUS(FW#F<>?fzhJ>t+=)KVSxgqtSVWxKQwEWg9)$m3?>vP3 zNdXZM)J9+P6&uC`iyp^G{B{f{Nt~UKC~a77!a`?O7Pdpj6MK{ZCYg=i613Ga%t(wD zOiFDWi#?6uf)zZ?WNZ-G&jk0!jb0f059gQ&l8S2^pEf+(tja_l8=#YK%UuFkJv%i< zk+%4_T4s;VLknPHPM*@{EpuJAm+!wJ)vBz=8HCX{@R}q~;{fjD9zy#&$AYpFI<_$C zFQ4{+>iI{aF2?!XJ6ie+@^W_f9UMD)-$Y zl0&;sVp%7IzV|^{1`Y?If!*t4=B&w2Qxg5ZPvO!7AI}yf=6oNYa^3si;F`Y%%x=FM zZ4AqNzSq49uB9wEc0c+06r$w*n>hMjrIS0unM_3Z2QI#_GUxKvU59Ms|*AqPQky5p^e}5v42=4^fCHWo03gjj51y z3mQ(>2}R$#G@KkNIvLsNxI;|xe5Ud@tn5^E77Mbvi&BQp=&>f*rlbZE0HeCkgbeupDuqV`&K&Y$oljqUWbCw-at6|=!7Mz% zKxuaR@|Ow$bG7g04Q_6GOtE~t`-;M@W;xgCTFmPQ(C_{XY1wkOm7+*;^PwMqC9ZN+ ziDK8xJ8`rXgA!~A-(!+>#?`3t>nFL@btCgF9%AWXU;!4P^ZMU8>FEM20aU3InfsVY zq|7HDHla{W`6N)R!;7yCY-?n#QcHMtzDr$tW`+6AnEWy6BgYg6_bjKK=90zhI%pqawIab{EDizBam)$OYfP01^7+=Q&mz z-W;Ro1@ufp?QUX0i0kaXRE%8WG`S$k-NBmR?Z0Hc^(WPj^$l%mtBPojmz6(=7<+23 zAG=Mk&ONv~&velE6&O3#a2>xspaHdExB zTh*6fi>Qu^aeaEcv!DKRbB)e=cq}>@$6luwDtF=iX*?D1LxXp0YU`vGK?pHPAxJh+Y>8J>V+Z|@tmo9`}^;U z9rSg|^}r5~LYIKPezh*g^0E4G=J-9BzRhhY+lvSuy7bSGHIHcK5duaB2bjG|4!Y0x zu`@aA%9N6poIBUHV~PiSD^J-xP>0T8+$+7!X|0`J>Xz9b%!mH0FI~tf$Z7YJUrsmB zdrok8ybwZflBsjiu2))!NYT4MlqzBP%fzq>dAj~ile3%BoEDqEZ~q?WkRvWx=AAK_ zg!cIzAw_i-U|$&RYmI+oYU!?bY3K-IrE>kub5@M{cg*5FV123ssr0KnKmBLJQ4*47 zi+a*^^gY`CoN+D-!OL6TB!O}2iM|BgfIC62zxON?%(K`?Uq0s!OuBwz*%lcvC=y(M zRL1ko$gO`U{*ddm7xDIo48y8gyLUG3!@$bf=4D){|Am{vjLaqZ z)RoT#QQYfi3VLa)BHF9b_-S|scXrzF{?JaWU=P9;Mt-p5r1H?9^%=1aSs%CRS@xKU z@^|U*_rweRo*pE+;`zrbG~qGm&w8{hw=(;=(f zA+J5qelUdlDNu?as6i^Iz&faCRj(9Dpi#w*BPtZ#6ZQGQMcY(qu03L-EW+@VJKV)j zmnU+z!!%8TeI79cn=?ev#z8kK5MM`Y8`19YIE2*3^;=dHymTnw5GXDk5tJDkC}XvG|c_>IB~S zK5pICl=vzO)_DS+j%7}-7nw~-F18N#LrEo4a~z{3C0JEXQuf4soy0@$M0vNkuGW9+ zG1d~>*a3lLpi(x(To^T`khqm*KKP-e^JWso1@YZIHjpGSM(L+gws7=j)kb*57fG8> zQc0VVnlftKu+?64rZ(LbUilX0*;wkqqX--xSX2uFULIgHF}*$Mq#yUeFb7~XLKTiz zwg^|8k367fC^!I@+S4vvYGl+8Ib1ECN_9e9qR1otzRYeR-L7*qy#-zYmOcG1m7FbH z$`DQpyoaomkG9N|n&O$Z+>`=9D&6-^g$OeZ^UTSjT0y8?NuxW{y)47|Ox8(P)`34O z*FDR-JBybp%U@R3Lsk~&H7ndlB19L`bC4A+3ywMi`*uStM*|&Eva`z|>9$e{>)<4x z>`zqL=jqvbF&vqMIltj@!rQXTbj5*|In_Q)rRzC8BRNg{xoUgaKdEx{!*klYbKgU9 zwaRk)*10-#6$gAI1_|>fvkAxOC8wtH5Zd#;$&xps$W6$~%@gLYbz}bEms`!wUuDnV z$!6ScAS^i4^_n|Vw$fGHJ(Jx;kvq^WxS?0QCMoep#Ywt2+5ZT4WghKv>w*%5!- z;35+!aUUuk)hMB3z@peFi5Mzji-j-<$a4sk65W>YZQ%0ul=}OWWGt6{*25GyhrIBY zfdw*Ti72Fc%6Q1jl!(e5Vau`qp9wHn023R49iRd;2l$_P$N$|8@=xADB{5A#uLQ@4 zHA?rt^A4%MEgApGJFqDTm0;}tSKe`|l3;;I^NlarT)p()c}K?Xv>;rw*7kJ!uVppZ z%t&s&*rM@f<^CGOnM|AYl%S32eECkxYE}F1CAeL(MLV%h2(TbfXT#n|G=3r`Xz)xL zh9F-qQb8Ah52Nj1+n>?ZdeXy>a7^;8tL-$TSTXDPcX#`jj(;a9;ULV7E~mMb1iw!E zE#OuEMD`no@1Lfz@P+Qzb{ihiTBqmrZ+lG_|8R%1pXJQYNvpBWLM1x=H)zKJ{Z^%3->L zEc*exUHfKs(_;|yH>G9Z0P_uomgnr+aiPZ^4!PSDu3lxF{31KYUf6r-b2|InA5$>#dyB_|?f5Je% zNc2MT6;0b9HM-SO$T`5*YOQ>XOoj{WdMrkA0xD;b9}R?~C%KzJFkBf8-U+qg^yb6n z7-BwPK)#~eci z9}2qmB0gO45dzjsXF`SlbxMbf{9TqXH`&$uq|i@koF@L0W6xaha|gI6ybJ6c zm~LkcA`AQUhzIWdm5=O)46myTzxwsz_JjXHugM7q!GIzTTd10O5i+rkh)772**oek z6n;|truilemVXxoS}yv!?(B@A0BN;4WxdvI3KOX#LxM7{i4feeg-|M(U7mY=rQG-u zM5<8?t>6_tKW%(4@5E>4f|qB?QhC`Pr@nL4yLS8 z@n91yK(Lur17wt`f+K5c7C@(-gb07!iUWZE*YgJWVNt=;Iij_Iwab)gB&s&-eaw(VEDJmUKSR40P>NWaFXD_5&6QpGkDTRVkhOMo z=0g9*;0%{3h{|Hl1SakU;ZtuDtnG@0>MRzE)KY=@Bf3L^r3#Um4Rltj)iQE8xhWH> z6*>p3b*o>R?|xbZx5BC=rt-)a?2$|k5ypB8nvwHqERcz0%S1Snr%rYj7gB+OzT4;Qg)E=y`f| z;O()IEU!`tMt*HLQ?m#G^Ar#CWo=yGsXBg}-i#eFMh#X7U@iqb*GeOYHC zBfqhj^VBA>!DLrlvQZQHB%fjEprOa8Ukdf_=#VPD)CgLwTHDX0>-od1LPwFj3(TS& zkFdM(jY~fK_)Pshph3w|-{9pNxyb*^c7e%rl~FLW%(+kJiMNc=u2Y2BI>5aOZjLusI3S#PA>|2- zV%$~B2G!>R2Ozx(rn z_Q`L8`E7;QEB&}AXaD8R0zr6t(^JTHSHM5Y3vS|swZ)9(D*38JQ#i{wisy2JQ zQRRto(g4VU)|gkbZOfmEAg0I+0Quc{4?Jr zzoj~$jZpC?Ly-|D5wN3n?r(T=T`h%| zu{Nc;yYCs`bFH`3KsZHTf2fjAk*3nt5Esz`YZQm&dcTf$Z%Uw90aqJy|9!!ioFz(G zR>?dyUA39jA2M?trs1ElwZCt$Hb=fKeVf7r0B}}K*Z;59&N3_xWn0q?!Gbn!2`-I8 zZ~_Df?(V^*2`-H^?rx2{(>Mf&kc0$x_ux)&x1{ghXU{z6p6AZYulZd+t5(%|Ykl8Z zE%mjd_UtQ$$v<$5$)g2D0GUf;Z4#fGc*bv=OFD>>IUqeHqE~UVVk{JgSTnT9PB+<4 zX9sB49Uo3lJt@DQ3*eQOo{Zi5uB;H7EVT_4S?v_b-Or(M-FMr@yM#}?{)Cs&nAY;@ zV!FW9kI`BnM{m6?zP{teDNrmIUrHo&RvgU0#ihW6$D6w1l^=-iF@l~4o(MH5jzI$# z5ppRgJatXCH4J1@u6q~8TAktGLnPx8J_05d_=~={Bb4<|GP*G?3YH-)p7vktk!D$j zA&uzszQ1qGv69m*>KXUk0&IBh#`G|nJgfGhmb#ewOi&$NFZW3N*)k7;DK7_UCnegq zeoWqbvfdF)K*Tu~@5Fw;`6h2sFT8JNcn~T7cNxwCs8;csJ~+u{*>Wbm zEuvWqJ_UDpx94gQDYUZ%7sR&92OBw=L4`ghWr_g@h#OV|og>PkYvVp94IBt;H{%TU zpW3#)oD*Vh)ii0hC2I$t-(gXKBOs07SOGM847&ziGGTQ86kXF5_vyHNu$7y_8y=r9kSC>9}aJK;t$SpkA1wf6Rmc}_m@2EsY2Z?A{h$BJ-^th;9%}rXo$yl}Xt1PCdwUv#jvr zfk&t;MO;L^ABVdq=%uAaBx-(&Bwq3*VuXZc2587d9eYI?ctzD)rQa}%s zC~n?4Q4j986>QJ?%3X7gndrISvH~cQ0lz(yz?G2e$RxOXAgQ$Hgq;=iy(V>MS(#qqrg~*tN6GTM_J4_|A)CrPoDHdN(gD7!N z3UPA@>N1m1e1K$-%A%DDNTd*75zSa#h(ko2YZ#SBq(qRmg#SoYY*3u;Mn)u?CF=q! z?YboQJB2ONtE5P71T#R+uvGELLb-~Snz$J2f=62fs5vaM%csigC7$F;vZe3R7pjIj z6^bTVZkf&M-y{C9lfa>~+-bSog;nt-FWkcr?g@pvW9r3TD4`Z98d*p&z2R)*7QWM} zsD#mOBg+~jE5^A%bsT7b!Jptb!9@7CDwyV#tJc6NKE>P^a=+_Jch@p$el?gYVUl^( z7O_D|QbE>o6~tT!jh3Xb%BAM0s#S#889(N2n%dd}(W7^@1Z`EfqAeb6Xb-eTLPHCC z(3DoMdJvQF$72mOZT0#g=YMa|C5!)c4kP!UiRmH(C-S5=li1^*>2EaH8AT3VdDx8dTdGFoB0q>c^3c?{1rv^ z?-vEAXh{f!XrM1!1-m(zoGR$dujb&?w1jH|1#$ZiZCmjv@%Kduv_}~Vs%}{a`A$49+y_W_=L z1rv3wfSR|QU0XeZs-l3_LC()-1@j3{Ct1#?ZfB$0nD3kt4&k$a02Mw7AD)_G5?7I5QRDA^1D&XMLE0`kkb#)W++SUZ5e<}<@? ziIh;SNg%v+->hPgmn1S&k9=E7je&>+b-hBy*JeK6TrFj8p|6Vec$vC+iwo}Q+)xiw z&DAE*@)jTLD2+BSCP05DFZ)==3D9RSCv$CEg=dXL*^dZXE_Q^7mbU$B3!$HS;%>k}?9%jU6kWs*mP!-YRqlZ@Tx4Nr+_Up4 z?>r7Jy00(+UB1e{ABBtG4Iw519VZ-tcsK=XDwPUl3-iIFr46T`n|Msp|Fr zgs|5Mk0^(26|Gv9Xy2>a00O|eHRnuSoN8^4O8gJdUmhL3jE0V&zaXro*zrN$cp_V= z_qegn`7a2gX4UH;YcjO+4%SaKkXtruGx99cgUmrlHCTMZaJ2FxB1;mcEPDC+EEGJ09Qb$zXp3vM_V~If;xn^7ORlep#|7? zA#}s)lXS9!tn}md6jD34Yn-yM+L|flhz){@`?AdlHoH0d$ ztYy~-iLPnH@H?!6#>s46txJubdAC%XbNtfpc_g+NZ?qX7o#$6-^7JZeY6w+v5Gg>N z7%^6C%&0xKp*%}$4>hi#J~{ZPMLAWCGE%x9C3f*BIJh>?4pSS!cVsefOeodNFZX6raZu_P+4l=S6@XZNW|4aj;+xVK6RVQ3n_-v%R>P7;xWQ zTMM4&vs#I-;gi$%Jv6om+wQpWsF2GabMt~`(IL1xKG2fqi2`&?tG&@vwG3$Roy3g#X;YZT;Et0`u+oz#m>xv_`LQtZyd+bvyAG z|D2P4>mkt-c%G>}!Sv_d8=y*jp5rAD9V$qeVX(dA`cpI}v=q8Sm$5sdmgl6cMUHgY z@0Qsg=592xJXg2);Ozf=W#-%G^xN}lGB18k==U}$N%RcUmeqCX`?t_K+2-m#t3)A= zKT~y}zP_bbR*r9Q#~}p`>%vmcNI=tjx8yPeI=-E(3>7W_{H>$-)HT-kxp=P5%m2&a z2h4%S8Ars|r`%uZfnUf)mvS3WU8V4Q+k66KT{^t{4hUV3H+_J2e#CNW$JTb#cSf(< zlxWdEE-?h<3i{>B`R$QF%lh3)2gIs%t>Gzv3ugN|AoIx`-iLYUMY;P`SfB^m2V+UW zB_a%c12D z5Se9oSm?=ad&wQ#L5*)ij^Wmh1D^U3_9C`{zpO&t<^v@nz0mIi1Jfhr+H9jsg5(B+ z@Pu4d7riy2f;*aBv(uw8+ky+>!7K>okf_j)G6x5sy4hL#q=x&H2)HTrN9|%n zbLfSjdxUVqL*B^8ooBj#y3@PUi@62IR7S>|) zAwR_HprpKtBorvs&FMCyLmWhvGJQ-0K@ZzS;8y~`4!3V)iCOp1{QHPrX-F7f%%h$I zwSlcP88}eSWpXfa>J~Gy999Q*rSwtFLIEH(C|C-MSXWvA8c@qx)3*lFoesrHN)%ib z(sLLxEEFd3X3iE{dFg$Pz1&oIgVn z&Df08*d+^6l$8~omHUvDd5C#JD@W!*m#Clw%MvB)6)Y77)Felo>}Q|cS(BDnHyGqJ z+2vGuyz^%UevAlR+5xWfi?Q=bU0wR11?Qp`0~^Sc6fM}^_}0VvRGa?gb&&s9jB&On|Qsw|iw5TOHF@qy!jHX}$n zd0~2F68h!boFk!`deqOG%Y!mT|e;z8!kM$-WiuCfARiP1G8C&o z=7w~O)(OquD=V{84Z#N$_?#apbK>emaw;n75Q4Uq=^|d>nyKjBwS1e482XIPmVTqc zc2>n-Nr$wC?6zh6zZ}5HoLFm3voMkkj!^}y+&Ur8mBO~QwOhUBU2EXu*O}q z{yab}erkrg_U~Rw$==qS^4R#pSmwAtX#Y3!zn;TIADL~S7co*jqF zArPG1Wr(%y_;Q3BG3YkRx8QI!CbVC3H7<_%r*B((GjHs5a?& zTl2d^oZPJkp>y7EgpoSkZ$>fK-EaA15FK=?ml4++BCCjdmi2as@^0r^=B5LTFwP&O zOd$!=(kMmw7adqb5dGzPXJX$zToyhNmXb{InPXCVB0#6()%cCHIZOG|l-42S`Epe7 z?D=X^j_ddJte*4lo5zp{(xnm;Dg$7Lf{@^QW&X4AE{@Tu|$Zi*p46TSF0>)2E)-M@f-_qvk+_ zFGyMsLs9gFwESK=IFms{gA=@jY}gL$W6{;{&?d?|z|^lW>e?nY#il140QJ9tCC33m z38XMyB>Nmz(3(bDzNdjndWVlEqIylU?w{_{~T7f9-9JXj`?u*Hwmu=zHmUH+B8 zbZneoI6XOeuV<-LU>Kd;`lk^C>E^5h|vTnvM{XE(fNNOIwS(CD(#`5SbOdOiqp2i_LMw zGR!jUPBW{#U^ree*cZOP{7UE&qcf}L!%K=1d%6z7=3PXdX<37;=(QvkhJ?-hpR-Uey4j--S(30N~r0968qVw@Hn7qeU#yD3e#&had*zVgAOw;N3hwoR*8y_b8 z4u7LHkNMBp2RXmrP=D?Hy*Qzj@#fV8-ebq|403ZTDvpknFJ)Skz_>iIiqmbsUHXo@ z9S59XTXDprip2wm+r7%}5k)gh9PP+F_LX$NXNxk0x;jFQ)g4e1#e9AKWlawbiXMz? z;dBRRY4?#@bxN=HE_M6Mlb77&lGma?wvS2jZ_{7`@9qvn@)dni-kF~qU-zHpJg(r~ z2?wm~A*J-#(NFeZ4P8iabL%DYh{`GFK0E>P3IiZ|l?4duKU9(1==y$BTcwpJr-qI* zBl7}M7uiqr+sS58PV6H+Zf_upglgiEx!CuO*t1eN{UH3VfXeSe%v$!emS;0g9g{I8MIG)Idv^4zg7-lLGc;43sVD@Usux%6!=|k z{((-d1V}i(o3Zmt;R?3Gdid1!wXiBMZLBhmbK_kK#l7RZ03?*7=NFRSB-$tZk8ju? zPR}TwQwV_nR?6G~OaPgHtbe6T&M(RShf?PMt24M-zv<)Za7*<HuP2icT z4Xam89z52{}Zs_Oja;{RVdg~d z#{>E#AyhbZ3Nl(|c~Gd6?S4pIM3M!_&Tln2m??(Y=8>sm82VI)YM)Zj)|wh(I6!~IDA;F8Q&bTG za#{}^n)0YvkDA@`Kc3_d@Aa|={=^kLXhRXGIB0i;jetWg#Y8GQDMbtq%NPij9YdM2 z7g^$UF|#Ui1*W}Ez8#l|%v81j*tHNsgWT1R}Zqo#mGC%KUaxjv3tv)Pv+TP z?4PJeDIFTcd8*6H7;O~Knp|3iCf^YvE_3>F^x02R%l6{n-TV(7wQWQRQS`%CkF8vx z0$66eW8@{2cAwRne$fIzXI5@8H|24M*?G(QrzFZ;zY`boKjJjBh%SV^a+jydWyq{! zwX3e{oLBy|YPI;2Ez~#xQFEU88X-pOBP9Z!nOi5Ywew|r?+cbxi1lxZ5YbaLZ53h> zbS@QoA(m9MZ^$mvPh#nJTC84A37I|H%ea|mY|T3SQaG;fyVO1cemKp@cy$-_t!=gH zySx2!Lw)~QHfL6V@5}qlkrWu9DUQ?6yJ@j$l(prjl5gv-9W`#<{v|H@*=D48kq7Re%{^aLZW?SFvcvpY>P)(>u1iN}KnXHg(r&)rN<3?-8Y~BG~>f14dv4 zxQBvoNjHCKJv1c)5k|dl^RgNQhmq#bvF{6hb4TmuxJ)AxSA1f(R`5Te)7t3n<>^+b zf9D8NC9C&(gLeuHb1T-McwZ*_ns`cEE^g?W)BF3Wa{XHfO80A|B)JP|P()B_63vU3 z@^7KciN4Z1w4_7u<3VYf7Z2e8C-YdNpIf+T%~ni24_GRHevstOmNK5X#Jd zVfvRecMP_|{%%xCdB-Vi+t?t{2?h1=>#IT(Nic)YOiJK5Ll)ypjDKI55=fH(H30K} ziVLw|ii1g;KO%rHBt$VJB>JLp*yw;s082^0KUXG$JW}aksc~f^<46YQkONGif4e)j zl*CPnB}i*L7%vdhxShpT4k-s^_*JI1nO3PXx>F@fQ=CdHn?}`aCwKKitpOX;YOnQi z`X}q;e3@*|ndMLXc&F0(pV%t%#37i!58dk~OOOX0G(?14%)f^bQkp=CCG_$W>EGu+ z>pK`#Sn}A9j|$g~#fo}~?+6oB*ahb%h<|uJXJ=m#TmbdJZf)IJG3KxcSh?kx&C8t| z@2XmHR^NV~d`FA{INkitTST9Kn9IFA(cYw$&+gfDx3Q57MdbTuH$O}r1S^J(<+uF0 zYp53ZW0-CBSdaBN1+xBi->JE0EOOz4#E`U^#=y-rAq%NY#v&8q`Gro{t#ag|MWgu+ znU(98USj=~LwLFb;#t!|6W@(s(`ySK-4Gfq);y=oT|B#C4_pqNjbXbb&51Qh-h=7# zgpkb#X6J#rqw&Aq%#$HmED!}(kY{RYnVAjWVoWoWg)J@6>MTcR zZpIje4Oyh_%)yD{l|J~asbl|G&<+byl&4Klv=+j5{IHZx8PIXKKpFJI@s=TvLQ`Ql z)z)c<+5N?bgG2Wi|IS%ZIl1Cw#B$m2YHm%0++oq~90J6qPW)kDy|(s)AZ1AxE40)g zsT0+q^arM(NtxDT_$$Q`&7}=^5tV?AIkIvZQRS#q*pX+I;+{=)EfrREp8JWNt-AkX z1!vv6@%>+)qx-9k8oF?dj=!kKRCAd69){P2Al#%mDk5aq*4)t&w#q66S8aBRU+D_$ zTYqm&(@p1O0%`&%y&NsNgysfRqD2uWZVgrk0O9mtl;LRBMTzPg08>rAdO5G8H0eL`dWzDMCRcu!y$0 zr8o!NG=|KB@xlaFobWv4R*-+XCatf` zL@Y^GgUKMJ^k|HrxPd4-h2&^qUMrO>KKFc-B^O$j{8D+CA(P7Ldzq zDVxXI!m5<0l$@^Y<5s;;Cn_f1Qb|}f*JSmB>L)``t$cKDL^_{{pxtUQjm)E!&@KDi zH{CfA|BsE_dOao_oeYc3#pCr6)oaRZ7lrbz(#-v3M%uJlDu%mT)`r|o+ZC~{ssc}X z4O+Dy96NtI5MB-pTZ|Ns3%j?dEcoHXPzCsS@0B4(O{*>ndMadmLH+3vpR;K(ZQ|=( z)$y8V@d)?naaiE(7k2AJEH*sb72(op*l{RR(x7<9@6fT|NAN#x#dnCv)%gPNJ!1KJ z=MBcxHHiyXl4XqnR6vHr2yH6085vr-nacnwI7dPnX&^UvC)s;VGRqt-)ufk^lpSJ+oViHmJ+eB392xP;WCjuZ=H zTAO}{bsEucWJbx>SK8&0wge1&TwDr1Ba+S(L~se)ewH!`cY$7;KmwA)2g9Bb4r;DP|blMviV8n+mb1PwtNCy)k$yEg7lf_nqO z-Jx-4+}$N;#oW1crdY5GWV|149sC2oelIfgu1C z0)j%oPzV$Xfk7b%CF1ngTP=A1Pp?NK~OLVfPjDy5HJD)ML=K( z2m%2?A|NOP1VBPSNC+4Sfg&L=Bm{wkAdwIh5(1zgAQS|QfgWymQ90r0TKyV}ojsn2}FdPJigTZhp7!Cu& z5nwnH3`c?C02B^_!og5D6bgqy;Rq-k35BDeZ~z7e!QfyR914TOU~mKsj)cKcFgSpK zgAi~q0uDvMVF)+^0Y@U>Ci|frKNGa1;^_px_`B9E^fPQE(Uv zjzGbYC^!lQ2LLn(K!X7^6hOlOGy*^)0W=Cg10XaAga(7qP!JjhLL)$EBnXWHp#d-& z1V)3wXebyB1EUdOG!l$PfzbdI4T7SZhC$H?C>jYxqo8O2h6cgVU>F(-L&IQb z1PqOYp;0h2fIx!~XfOf|MWA5_Gy;J}BG4!V8bG2!NHiFUh9c20BpQK4Bavtn5)Gix zAQT#mLPJq#7z&L*p^+#w3Wa7u6DUf4RsH-)S@I3ft4A3ByiiPkv9KNi3;@@^o8#YY z0^p*76J(RK9OEZ{?8huZMNhN4gPu~!YJ<<#B*MvFnzj@@%k7V1{1CR3bT(J^h)or) zJy^{=go6nTq{-A2jHWV3Z-W|Z%)l9<-ur~IPDNj0>wK7X;0X4qLd9J1bdFk%nG&@c z!)16a=Ujzua}%u`KXV*!-hkbU_yp^PTI+du{IJV&Dixq(pQyQ>ceUN;(Ia}o_$-4T z_)k`4y^^X7`a{U16A0a_jE8^ns;`tbI&P1LeekzA{5v*SUGz=u)S=#4YpzmnnB`H2 zqxoWZa=7(mN@dEjr`S z&y?Eu+*Q&@B=kgEh3{^TYw)o>oZ9YlE-zLqqJ7bO(*V8lch)Mh_lw^Ve+%_}3D0-3 ze2B#VFiM?5C^G{ob=lPX0RA&<){Sh67a>;jwV@%bDQt5r`xZ%S;XHljYxYCG%#9ob z@YP>N3GFB4nh*R%zJv+lv*(G?Vm73Fmw%C@X8rk`+{8}efh|7w@ToZFPZ$jk7g>>QMmsYZbA66q);3hRg;^CAf({M{`4NFZ|F^u|_ z^$v^_%Vvv0`!G$L)R#G$M@@5ei$C*D#rMtg&ie#o=g+X)nbqeQV9Vc>88ptQqPsA@Za(X#gC}Wm?Da+HP2o2o}%XxUCjxw3E1mA!pBe|JTV6y5g{%tTT) zN<~*7l^r7beg$gr3!sK?((|PZ(~Q|wuF}C=?~iWP$wt8 zsKkrRTbuy_Wb%=V>K$=fX`NVUdbHnD?f5H}q?SF19qU!i>SBg87o&2e2L&M$t~5y} ziOR62(I+O^kCJWOREl0~9{RVY?}dPBUO#>>2I4R3)LO=NTv^==KhRONpZ`)-g@9YK zN#cyUSTu(W8$M@|TK8kUMFu`Mc|YFhBh5ke@htr4Zx*)n`@T10kKc+q$vhJ2{InZ- z7Wv}c#!?4Znkc`DS)LLAFzw<*ZojZN6B>IaAlqrlV}nD445Ms*`LVn^kBN%de3=Xk z%F4##5&Yg@5U_YI{=gd}Go8)K$vU9&Q5w*z6z6FF9SEsYiLWru=QP_3#K7txv|;zx z(ySWRtqx3t4!+Pj7wC%YBhL zM9AtRFEiE-_@(5zZL>{=Dmqs?rXNSJVgjdQ9!*oO<3cef9iP!74!3VPgHU-dBxNQ( z4=FKTLSI%@7f+zDFSD~~O(Y10l6*B096KYdCu1{W7Xty`X|I;a&m4ZmwsA(S%d^OC z#Va#ySL72Ze3k8NREg&w%3~2O6xlgXh#WNG2HmoIRL@UkkzW~cV?C3J9Gsq&amvCW zGWnWGEtb^u?u%L^JAd&IButzXjtPZp+_;RXIKRtNyW5@Nx2z>2dR}NWGa1>U_>MmxaTHe+9BHF+qO!Pw?|?|^EjN{G&R^lbtk)zecTq^yIW{GC zo|Qb-9r~z+h-Ixiw$9RGE?K!<4{h+}F84X1)-`gU%kP@8^?BT~XuojN;iST&B#J|U z(;WYN-l^P)uTE*dVD9Ilq;=46$(pe}dBX#CmD^zR8cE!ro~fo+KVEaJ&RW_Fom!t* ztu`5wQ7Cm_v)I|77o|>ryzltf_u6OJe1lqAvQyC7E_R1TWVi0LN-&+rJN&Tc?8i)x zx|F?<+mZs0sFJUSh;2eY*LI)s{D6!>S;QQ524zhny6i<6Ps+|x8q?ypfgerSz|VzM zR(E$Zv9YSG6Ry^`{`Qrx0?z$EV`*JWC3ijNk4YoaS$(C|IFh(hnlZffjp0Teu|4Wk zB)x2~uDlTXRz?i2&ix!xpE3j3IhTHrIY>_P=nbHMmyxf2pnKOe*GPO_&MLWMD6o^J ziWINTuM9MO>p4HV(p8gBXKlI>m|;J7t+U&81hvJetLM)yl8dxZaddm@%ac$jD17Xw zwXl@;uCC>N-F&{S@t57$P52d+pB;k+@)NNeGG%awp-|8Un5*rG6g&y?#~y_FoEyAc zS_%_Q$$W_V-8{H_92vgd?cKNCnZ0usYt^_Bjka$Xs><2+TwENzwyB~}FFy}h^lDEf zZp@U77)|B%Li*)@(0Xk z(smV9;c39*G7y`ys^r=Vmn-W1#9-YrLLrS9Gn# zI@IP&@@{$XS+w^$W`#!~^~zb~>c>yynd%MxROiv6UZ=cTcMU<_qeP&)l%4SFb-}UA z2_@tr8Sf2NHvQ!B zP~yF%rMY zK&FP}xe%L`sQL5v4FWH4e{`4`rsey~HUzx+642lppj_nrIYi{r_yyGY$EzS?^@c!u z{=jrsZ+=SeHv>NO8vdq99w0Vb{8{~x8Sz(G1~RTrBt>7zb_2cqzkm4?AjL+76#}6T z3HB@s{_%&>CFva`?igwhxYmRSC;th@{}s&S z5*mjaVLTm9_{d&*%ddhWIP?8?r;v#9xUgy5@c5Szoq>^BCXwzMk*BWj3N)k>g9HE6 zhlIF=XJ~{?YOwcF25~|{GYcXegF`PhBJdlcWNU)tS%SUP!=F%kDg;Nv&%;(J!?BA$ zoKw2CNyBjFVF}EzzacN0-E4;_OrHM@dFrlrwHtN)hY`=h515TgPKqJQjc{c89zymV zo%H$_Wfa9gFL9;0G&?r;&U>+H<7mYDzn9=FiF_QB! z68t}<3ZjR5jb-`e9|j_Fz+vLaafS|Y%eZlZJz-~mf_2%boXR5Ls=_v0xHP^L~yO-!OxRrxSf*yMoClTaIplkQ{HDMIw#xZChIe7 zkS`WkBivKJ&Rf?urZyB?|E^1--bw1RNH)exkYP{gu}B!CGOcRl2zHYsG!E~GQ_~zs zxSR7GrJ~E-)rh0vuijZLC#$fK-oXjDMoa_0lf#kf4zj+}Bd9SN;Zsy`RlcGMc=by&p8V-Cn z_?xb+nG0sh-I5U@vrHll$|k5Yrh$`(73ce&Wh*yjwld|kN#+c@=Ya9W?1t#ShIwJ$ z=5Q2A+R7HZcwRuTA0v-nV2BPY>X|J(TFad>iMLO1Qccd5*nfR$QS|m*p<+|vSBG5D z?P3S~a2Pe6uWS@OUzUbP(rXX+5M@Wj%CVZgt#JO7q}_A9G43W8hq2wy!+O?$nDWJ&fu(N<^99ncHQ{VL^)@DU008 z5o+Z=XQcxial5zWK0W1Q`0(uP@@@Pq2zAk&S{5vx?;4+cJ*>nXr#zj!((0ySyQhSc zD(~}9K{C9o*CTl1w%BRNc^F@Rl%s;$N*DLe?hk$ucUW0dgA$Ues$r&xO->(+MiVEs zI^3Z8gLww^qr5EhVimM|O$5^G)i588R+Y$sDCdEeuw0cy2^Ad;`E*LfH@AG#xeDg+ zN^Ze&!cQfxmL-^+wa)Gc?N8aQaHmhQHC1z_nu}!lw>3=r`TdQs=UR1=sSZYRWHJQR zvkQi{GS#-63_LWIpFI8C7wh{QYI;G%Z<}K|WQr%TtFAvHpft*!1o~mkbtdlhc6U|q z1=CW^dPUAg$g^5?8jr|Dqcnn4zhOS@=1RW12y>3Q+{I$pV)-RLH^rh6)m>9GO-Kp> znY~;T*@vQ*jfxV_BA>-Z!^_5H(k4TXW@&z(ILqwDy^<#(4eiYwwSpBm@szWkv~zbf zbz&`geIoOmw5w>EHl$$N(2KSOxwhZoZFi1Thl_2z;Vl*{F|s8gNu1TW=ZIf&te6HQ zNmcD}Ywh<0&KM)@Pd|02-nCS4iX_l21f%w{a~Wr(LJaH#YSl_Y4O$S1VQWeunvRgvnV+6)N$ z^tk#=uG{wem`3iaDW^tsUXyUO>wY4`ic_hB%)YB4@fJ+%(f z9vD>{2t(6aher$q5dvPceKGd^MyP%-uR+hWL64TfpG$-Bhl7r3hTPCT*$m!2jX zP{(Eug&9zvE|PkZ4xeyXAzG5_4*l)IdX(o#o1i2usNrV9iY}p%9_^7nuHpW)k)f86 zkrDs5&#7A=t9fr(#R;Fj04?Ol_ndL>XjC6(qD@xmc9Oxri`DL zC3j3Xej<;5CO>{bD{o^kcGEK66fpkKGQQC%oiRC%N|VDm8c)+7<8qJ9o(!V-jH8`J z?u9i5!|82i?{%a}q83hqJPQ6MfuqI9@ywdyCDCLK$)wBoN!L#4<0YKy!%6P@$v1RU z$LJOW)Ibe{$uW^UCD=L%D36m^sHw3{Z%^W#38s;V<1 zsz0OVor^8dgGZJkgnO502~=bR~<^19;?8C-&*k8B0uDaC0d8x zu4sxbA9+hXpC&v%O%^=U1b1i_v(C0GtHhS#P6}9_{!I(0dXIkfUaU)BR8Jo@^=7ka zEiw{L?xLH$-e&x`~I;TtH%zK4wVYxDKWoKDqV{Tp*uP$_B zMFZNqHC0QsJR!p8X)}Pt(fzGn`hw}`mqXVt8oH$4JE>+LliD4vt${fTo3$yK7YOgg zTAgL*XxwCe)WVvi2K<_$!Z}cmDbS`o|7;UqAxYhEL zIxON?ZS&rmZ0NWAjd(Wh=OmF-n{OgN758+BkvlEMv1)41llE-bkcaobpAc^mj7<~Wo^J@xsm9x?f5ZSn>EIwi3V!lIpbdyO&U;066J(Aai!*#HMX_`jkx@s5{`9p6}~ zz2Z_g-3G77QNr%b7`?-L;$@fn$F9n|)+-wZx-l3Wn_gv`>lqcq4_nF)oA(39m#OpH z>{?$lHiHh}vqKfvB6^B8o7VQnezM!9iYKoVw=pWW=e-{L!cGUv*C)D93ARpK6L z%+A{EbmTALv!9J$EZ*}eIHm8-rk%Q0@~%X!U#Xo`t2c}YdR|7Jv+eEba+(C8t((pX zbUKHkH){51qAtJ!d+KRTa-8-fTN9xL=g}f-Z`Q^*PrO^&E(M)a|K0?Z3E*_7t~*Db z-kd&hwLV1NA5PLAy;VG#5V_V_C0;lo;BJx zb~RLfCiv)u?}N!xQHgv#zjvY6)trVtKDQN}XBk^(%(lzEz>OyQNS*7qp4ZyMEqLn5b8tL$*48q!xa8yf5c}z7!?o0*xF` z)(z&R=X3*1@;GjYSL-@Z>%KXUErd{M;norU@9}<@P7jb+@zuVc8h$$lLdhfAI3iBP zaWSTzb~my`5cc3pel6`L;?=Use*WxYqWlMV1p4l=S(<%8sksW95Up(wI5)>74ct#7 zSv@4p=sR7`cL5`@5M{|EjpYbUxn*yeE0Bl3ybY2Vk^v*hv=CdOCilL^r8ce_R`R`}xC%ug?^_uH~Twq-a`F^sh18Q7*(VSdd$VTAiR$ z7;`4K?7e(N4=?K;>QU{~QkoO*R}AlSZjmoFo{?YoWpyB8t6Y3D9}?sZm4to|z2t?c zTA1*$D0~R4)p*Mg#!;kjXXX6yA%QDi-NJ0SSQsLMSJsHdx^?~Olc@2#9y@8;dI`xV z_x$A!M%onPE>}W{k^_?_XX6yBk(FBB-|nhcY2!C1bLbz@v&;HQ;sfeGnBTa`b7?tH z%KE8S6W4Pvt=kOx+v>Qz#vhQ^V*|2RLTucA$csaD?zHaq`cxFs_;)>_98JRwB zR_`Kc&A61HbPE1oLNjs7RR(|1sq}k7+*6 zRaY}sOdNNRjiT`o@jm6NcJRKL@&&uG@387I8uveLzY~E1|BYAm*v|8D@5qGV(34 zi?yVl6%&AISI!M!_zSdT9WTVSsp`nGXyMM*rd?!QXoqPWSaknQFzZ1PhJWi_mmpzM z6OG-=3#)jsPBw0o!C|xR_O!;Z7vUz!iXS5cVCGQH3`lPO%`>dRBzBZNW8kljdH#}y z5s>W+6;@N?pYNJg1gNOl0U$USp+i#l>@yH@s>QxJfzLcRAoDO1uY7rn@ZPoRWVVwbZRT9)KBnSEjUZmJ=}WUv35G?#g8+6PG%-S7}Q zt|C%1lD>}__Uz>gIg;Z8ZJ}zHvPMF9#njaBKi*D})Yh}KZ}EQ_*jS`EX|qd?gBjAk zOR;8SfXoMAs&C9c{k-d8A|baJls^wmm-gQgYz$I#VvaA8_0A?3cUMw=_?p?#z@g0S zI`owz+>q8_>vXqYQZMoOTLK_XqWh5T!C#g)ow}F!dmXhzcX3 z1FgRcBdrNa0mA#MdV7}g(jqmC*$92~Itd2ukfCx~$AtceY>zQ18<&g>9W&;{A@M>F z1-M);M-HP_bLqkc!!@i=L}?Iy3=n)2B{VDxY0TV@a1tZQr;a*jfGhbY8W8-`KC;F%jhC9%X zs@z>?F26G(C!g2ngVr=2U2TS(0{5g^_{@9^$umE2fapqi1kD=_`}+taO)_hH3fr|| ze+i3LG~3+8Lv3j|cj+r;?GesrE!3B_hNAeXG{Mg{@a&L z4YHkd_H{T_zRA`F^?`y8noC?@6@%4{%6W0}&lT00BEq_ddzQpIcV3wCh&v1o4!X3w z3W^PSQ_yQ-!5N;W4LY$cl(unjgpj$a2Yz^=L}j4=QE>iQl~p9JEUbN0Hfc)%pEL)SNbM~&e zSve}($+P7^TlamM=+<&Wd-6ofr{k=HFGg~Q z(%|FRxaFxFh}j>dfm(EVM$Hm$bB>eg?Kpz|PV1=fcC%eXEv4S{x}EA7k*(Ro#>4S= zXU&~|F^Yhv^g#4`cT49+dz^dQdU)~gnwGG?+4~_v0HWx-V@(nXz8+Ht@jG6^rFQ=2 zkehD0qr19d_7qZiQZg@sv(FUqMii8^5z6~dDWIL;7p)|Q?Icy7a;RF1I#U>-xcGaY zel=t#5wNDahs1`YRvU;@Ids*>q_U=URTp=$Xm_!Vba7+(v0io|8oPL=#b0lVbNn-V z6MnAw+x3RF8_fs$JcjpK;Ai&)UfKuaZovR?t~8ogo1cXc-2z_X04}B&Wsf*WLYxrm zGi?tD2pIa(BS9E|RT(T;8K~C~s8ty#mnR|DBcV{*BRA5cg_Kaflu#t>{c z?PtGRUcWn{-=nABbGqMav)}u&-v?{Jmvq1{uV2T_@BXmSL5DWr`{2U0$ z8wf=Vpu>8=At-+>N;Utj zpXudr8`g+5O0JV|#N^#br_e~3#7OVgkv@>T{?|eWLDoLd$dJ7JaNfx9^vKAjeE*XC z_@Vp=)~H!tzaC0LPGa=?yU~V#;aTI+xfZZL?bC428r(yoxgG`N^yt#&s9urWwDBOi zsYfnuT24NqKcPo9-$}M)WN52&Xh&YL%vfR($n^QWrJ)*durnncNcyy?^6QOv9 zHGVQZxFRvW{B`^?4IJ?uZ}B<9rV%cnKYoQ6zwJ@-qV2sWl-WS^wsQ^sA|1uJm&YU; z#gv?Q1X1n?7@61KVlf13{awr8uRG{b6sUAY*Z>1AxzWfRP zUS)o7RZA?tfy1`^A5(;_Q$o(FLd&Yc*pjcXzw(kz>jA%h54^WXCXz705Auq_Zy6N} zoch5Lg&iU$Y&T~J7=l6~0p1k<>Q|Fr}_q<#Dw6eaF9fD*lMNR&-2vOIqQ zlJ+g0XfcUzzKTzamTxdYn6;{QzRr8TzU*5=?_kB%LiyI*-T@ina4tdyXu8t6gDw+SGszrEbZVTPAt;y%tYZ_N8jAQz^hvF* zYSBEbHAv_+sH}$Wsb8`iu<7bhsjX41=uz;C_FS#MK3;zlq{$t%jz%suYv8q^Ulu)* zGTN|Sf1F{&T)vSiVJO&WL{zXru)Q&@upyQ)`Q+hSU(>qLLI4-}rewJiPl187n#L=7 zK|$sX>2hN*aqoN6O>dnR&)fz@#j2f|&EZ-jQQdV>)Ad6l$Z;f0#by1gkI7e&#V;$H z-f8RFi?1`hORn>`hI@_QK5Y5N8_%SU$nl%%i)>SuZ<#zy@bM(@yKD6;dq&Mj@I z_HA3)7~=JN#w=U-C|YS@0iMK0l-m}yQ3vrN2MT=#$~G^52#WbX?6qZp%V#a&#Ly%m z!w?(l4xF7HUGrX_!6x#BYI!~6%wlOXZolqkZQtQ*`t9eatdWF+aTc}Vv2Rwb3k!er zH=h1FfC=4KPwNa*Qk8xY8;_gr zHR?`yWaxK^9q%OQZII75^XRWLnpbTb>_x33%gvXMEnT?}x5{3#a%`g(K}2D1zGeHo;CNE4;aNgF$=uvyB|F>` z8y0W#vo=Bs+fz#4GxO0?n%S-21#MS6$rxt#c64XNJE!BNCm)ZX_af&n)VG;3)zDF< zz+(-J;PZ`Edm8Ce>#>c}m2+NKyNiMq&S-~f{xI|!3nkO4eWuGfKZAn)&(4}~CINx- zTjq-<26~t;Rgk%djr0ed8T#Uv8u=G2 z<0l{`QQ2}w7U^SIR*m`TSyG)R9MOVpIL^F$m!cGoqOM~L8ODj&7vh=D;!4hT>4wIJ zO}Iv(S~rt&k6myAoekS;i(T>Y>WLR#tBEe3;+#N6RPIxJ?au=SnMm%h%KxjadARC2v>;+1y{{_@4wi0`fQT9}JFdJ-~j?~}M_ zuya^j=4`ckT|VndJnC${;OubXY{6>hByf!m5uL?GFS{nXg^N!3OY53e2%53J{o>;G zf#pxk$u&*COKAq|$IZ?6V3o-6aV<5scZD-SuI?Oxe^SvsJeg0^f?e0j{@&)@u!~x^ zjNQOa4&q&%^X9bu>Q6wXe|>yCf*$@}V6P<=GRq2hY;64vm)>?36;|B2t#myvR#KWA z)7&ZMX>rj_8NbbCbuUu7ExM6RAAc90c-KH-R&Dzw$!wK@qKzHnN$bgI*%Oxv(MFdO z_p1IsZnk$bM7R0|Uk@Hw%Ot^wRk6+!<@|cjmT3K^C(lNGJttPbjEs-+zw;Iq@)nc7 z7Xx{B=%9KEPnf)M)97iF_-Pi~Eozxi=sD4iz7wnYQq=GIpU_CJvPg11+xvFn`{qLD zF+KIL!rxo2K6|zgeb@KHkdtJHyCA7afO0&sA(+ObGz%_|wZ{PCZ^(cZ0O>Io)!sgh=1^`bAF z90SHAY5JT!I`yLu;ole#IED?wj+T8gZ#a&_EEa^#_);qh7T&+VtA-bS z$&(J}Q5gElK3%Aiu3~M;X9+1)FSHpQf^Z@$3|nK5ny%)!fn3Y#tKB+o?F8GMu6#KP zF=Vy-=<$(Y!iiqH$D`-;P$bs`x<{OX(`Mw4{aC^)8GDYu)eb|kNuP(7y}mgP#xM%U ztvNk^W7!W^tRa=~Fmo@X-{%-R}^C+VJoNgi_IolOzkJ1q? z2P(dt^&hXB2NG>vSA<+$Pggx18F*WSJWlp{D_AQYS`TK=H>NUI<6e%AyuQiRp0D=x zVe6cpwYZ`(U;U1k!$PSa{Q8f+0g)_jw*NEn5OQG3B4pW_#$$k7mpn{^DfD@ixnU@2 z4qNsQ?iYV@!l`7W7R(dA7ciJOcn9U0{|q$ERe0U?J`X0DvqkTD!rTaoRmdpJiyNuE z0mXflqFRaHmPw>bq<2O#2csdfc=_=rQS2-UrUmY%398i=1u^y{AI!Y}%n#`K{jCB$ z&vd(%$x8b^`Ju>=gW6*~oBc&oQ4X15!)$O&>{)&uO%QT7EYx&pw;;)~DKRUXXWT8WX0X)^fMI|yhnT4Y)Zy{wOr7`7(rJ|{6k2AUT7=Jk--XV!8i2)PfN@;$x zXj(ZA-{Y!n-*(}ws|~Z<@p-g8zmwq;Mxd{NZR%NGm+LdM*OXCxSKh2^U|w21Fe$g3 zpT56%+;)0R!yCSfpH?3JHf3SI^W)0@#*8a(@${)6%zpY>FLxr*4C@Z2lD%?i5og0emHaF4?sH4etlyhzE+2Nk!6&c zAL%RyIvQqZju2R-E)=qMVpg)It5!;tpRAT-r@vTr7g!M(i8@(MTaKS9ds;7i6Y3~m zVDf&**jiO;$UHAv*8`k~rT^g|(Q29R5Iyv8t8zh1uO4Iq%kh+Mx;vgdX*sdcahe+& zyE?0yDz`Woog);(azAnB{+dD1dtS7-yoJqL$wX{h)J=PC z7c2r|B=k>eVQRCJL%(2c5`Xyyrq_KP zp5=AC@=-66opE==-s&zI`Nc&L%eircWkx0wERZ~#UoBKc8*?~Q*0=< z0mws4%2bO1XJYZFq)T;b8p@Q#j7{dC^ouX6oj<3Yx#E5CMQo)7=>1xwj5oVOs=vy5 zS3-k?s@gOE_CJBl*~p;=HluFqteOmBD|tEySBQY_Dh;iFOK#f^mDc!>*#Zr{3=%;d-@`)sD zS--GVqo=TK8+HlK`xRv|FWBg6O}?FZ%tbd5mijD*Z1-K2nXA>Lurz$9(zc8C?Ld@h zfcEshS5N1?sjX%q+2LvZI_5$WUqdys`X!!2b68Q0Jtpwb=#+}HxVjD=04)`(0#8gm zT$LIxmspQ}F>bCDvjkdwg5GV?_C3tYvjD(K-*8 zmT%ic+@7jS24q(6yNG0HthSmL;JO--f2_~Ij*I$qf(`uxcuSi6RoG0n1_M$o=P{Am z#!Li%>jmGPeg80|$MDFlA42~;N+;sOt$m|MAdzuonVSxG{d>m$(iYfQ02aXY_@7fc zM%I663p`v}aAj7fKkjo5?Sabd?%<~{Wz&$A7`@??Z!MPws&f0I89#^7%2a1zV9Lw> zo3_XwPJCrF6oKrR8cBWcxW6=5W7hXm#2=qd7E$yolsuY4X9!WO8ZVY9n=V^xIg_VW zW4S!!%;I0J-yKF5{)}_J#(XL)5?)ufSTAqYl`dCbu2|=CvA;|($+g<<_lQ%7{b|MT z?hx!@g^`9T-R`KDav}$=)my__OhzjsA8dBUv*Im@6`BwuLj|fO^fHuN`*W2tK9LH| zby7HH(1VrH=K7P>4u1j$#g>M%je$5$y|I?Yi=ByFxlF~@rmF*Fjn(Q{YxB*?@9uCk zgHl_|-;3R;620-Z*1Mb2jp0nC_O|=GKNkn9mi^7^Yu`+f}-^>uDadz@K=3o8xikj%{L<7 zZx?Mu30?1QM2q3DZ^9(VEjDAoEXA9#^89<7amrHcTR*?5S!~5?m=U{h~ZsSy>>zpoy9lDB8EQNXtUs?w2BF zEf2~Xw&4fmE!X=86&*O7r5^RTPSRDFOrDQF4ZSfQ{xnWa${otUA%XKLjWC?9x}MaN ztA52L725E7W$K9g=BYkG^;F$mxY!Xfg2!*3M656P4++e+4RfaCr2WyLC3Q8{(~(kf zoO-u1u^WE*u2%Lc=B&C1a^cbMjugl9yf?l5SEMi(t_T$oU=q;PI9;3c26lwb6-uXqR+Y+e1WX1 zS#8&&+BuBVk>~6)it7CHm@z80wW$FK+%w8|Pj8l#hOKXAL{XEN>?Sa@3O>8FybwNB zz!Tl~YsQ4SRiuUoU$0Z`qWYD!E=&)OX0R zBd0P}u?AX~gd_Zj9Q2fAZ|HWtx4pj9Xl{nNHeP*GHpJh=yZP-TfUBtBJU6^@jR(3H?p1&A- z*!?*Dw;I-gGI&AP*lfsu9@8*QN-9j*sc94II5@)0)`=^>R?K5*{xyeF3oo0`hc_k3 zuz-NiLXO~vql!1`<_t6>zz=vpM-$C>1*2ob@M8ih&4U(p&jg_ydV$`AabZ zTDVAvAv=a7d$Byb>{Rjt(?1{8#W|ND)kN8$Cv`>NdHpPda~5)nzzT4c6MX_2(psYS z-QusV3(s(lUokcP{8g*@iIg1XhyC&tt|2plS)Ec_YR=3_B`Y@j=Jz|#u+1WE_j+tB ztSvYAvl6stZo^TR5(oK>q>xph&M<&6KzNW+9Ed>WmpR7L-TLLw+{ix8G z42KzyV0E&_+A7i~Ec~Iz@=>EJ4I--Uo+AP9u>J&pAI#;5v8gT20IIl|KcMs1gFa-y^0Jk}8fLkWLg?X=pPl}j39zH1ybR5>&Fk;cri zC^@9x<)Cl8#KJjeGVtir>*N-BiZ`k^Ql>%S3E!%>Ndk}216~$diw(4-?@QPQh)S@q@c!$##Db-j4s^x$GnvHUG@%!5AEcsAEyslhwMNOd3NJ2#oG2dfrZD z5!V(;E9|SJ-cA=O*22|C_tlzjXDXv=OU)DxG#77Y8^&tOT}Kae?{4Qhi0dkY6%Gw) z?vMkDb=8Sw2QbrvhB3}-g!v3d;=VlNT;Kc{e%Fq|l)Tkq-+XOn)Nfp+hAxV+(;z|A+SBNU z9#+M(FfG(N<#^z|vwe>2xal}mVB4rV^ z{eJv|<7DYY#vN)$>`7Cu8_#79O}c=DXwzik*kys>{ho4k({!QY6O-!s^k03zb%y`dhuy}|zxr5icKb&k!}S&a>Vtq@zM*ohJLEa1 z?mzn2=!dVeUuE7{ zD5pOj{IDaZN^Ber+$=eaGpKO<-AW9qD>lyw<_If_HCS*cNRCm&-{Xyqb>B|5UDuGaMiFWFrHTemS*jV+Ex`UW8gt}nXA?cwyG|e2 znqBiVR&M)l(nN$~-y;hypc&Ik@Nxz#k+5Qj&LZzuo4%@7xfn|JT^JyIR|@O=WHeg$ zi|A$R<IwVTwm(8{%WeUL5xmVLU!Jg73;9V zh5`C`$;q|cY3iqcfIt!LiFH$7Ygn3lrzXQ#?VL_7jc2Q&s+IjN!F{UB)?=dke!4_6 zUBph$91(DwbKx`5?)~_oo7Yz&v#HZ>L-Ty_semMP4l)q3y?SZQ`sP+71kpTXsO9pfTh!z4la$KFr$2sasJeo!EG14Dh2 z9rbWr*{X_GI!%NJ;$vuG_s1!GOs|3l0T^Vi=!do`5`^k5fkb6fiz5y5pibc>sS<6L*inyP8MRiC^}=J@h{L4-XC*wgd*2e0_1WX~#QhWDxsW1`%rq;np8_)W5(sr!+AqiPy0~TV}ng9Z?YiKQ$ zv&?35im%%i@H`7FnO!x-cmo=sDR1)g3?#u+0d)bOf0Hi+8KEHjNbF1idtxpeS%9F} zj6WhNp_zA}$sVhwz;9kFXLRhrdhbvGHQYBl$|#(}u^|S+*n=n%u+nK`b|{F03u}-j z)^G`XW2R)Kb;gc=C)6BV9bJOo2XiQTHspNB4n{Qjd+1;|u~OrGq-K7KrCgL`lgoXg z&K&VNdAMY35!wMUSGy$w1@ky{vNAwnQBuXsQi=I%Qn*FG)dJcL!vZSsocGxQ7y5CIUS}^K24?Y zh#1$Ro@y$!m$Mzk>YWHEOt(&yRjuV+`&p)}h?XmMJs@m&VYWrpF zvFmBF+WT>Q`{U)Y8-P^f3oX3^#C__4ah)(nf78uke(JqDw-$7U-i6b8>iaxZ6P$av zgSft2fuD70CiunhCjK)A7}&qxtU5pyUJz9^K8$zvf z2Y6uQuw7rw$4w@F(%Za;Q!yS5W&vr=0-w=~6tYPF?Pd|Iq3a`)P4G7>S1H3P6{;m@ z)WbbVs+RbsnNM#CwX4#tH`_#s{&BO2!-vck(>`-eM2~`N(FSx*x%smuFAv zHho`q)r`#oWj5Hpw*L+|wrzE%@oP(>&=)sGFy8co`RE<;-eMcsaoa<5-txr|TF3y5 z!#d{z8i-DEjjj>J!o6>!U^dJLb63_)@?mSjB7oBByOEz2mRCX+p5rzHkRHTE10hDm z_hVsAMlt{~m5*8RagCUU4kkMkCO?$VLNXlhAjJ=o5k667hTwUY5XKn4z>xkl$6VO3 zpTOqAOW>r(G{hd*G|NkLy9wU=LE9CYn{5FVKI7-J0frszdU3L!hXN!l$kFl)+ZE6y zAKfBIVaGf&WUs*@wsTfZ-7ZmbjMw&~vxK$CiC+s{A?SVs-N+eptgVQqwgexal=-t` zWmYrdn~xJbPJ|QMT*nC=XHwbrS=*5>acO4ngN@M?EE(XO^Ewb0nOC)BdDx`INAvb zGYUMzb@T@&e5vaF>V&-+!LJrlF)+TCT01y3qScQx=+^7qE2sT z`$@;p*;)F=Hg*(jQ$7>v+4a5Llnl^Wf*6Q!f-tbeRbhU|K zz-Jn~Ns}R?pgnps-;>Ma*!78|sKj>-PM2Uk`c!1Leo9{){c_r*zO;H?^Bog#%Ccq< zLD2OU_gd5Z*2|;XgKQP?YXDgf6H!r^ssc6hMaSZ|(rqoD+dLV*JZ2(O2_l2?@ z&=5e`5$&oM0xOgQz@UV}!#Wa8-p2+`p^tz-(#j~8B{C*pd1!^oE(18a<8tkqIl=`e z14V!cLLOtnNI+0GmIJD2*PiTKI7<%=+)U)YMckqy%#eRhO}H=+<5MxrFgM?T1~6rF z>Y|0Z+`%~73j&r9uVv7WbvOc2Veyt#Srj`PHIyzCr<8st5`-@9BE37?3e|dl+9R|F6Kt> zleS)~%REpvVJM{{u?9M2K202J-<5#m=7?&ti6zKza>Yj93mwx+zU)+!m!cY-B}6S8 z7Os;_EF{9$CAr!Yxh;|ljXXPW6NDHjloFo<@(T4 zD}6W*HK8&UhFIe(L);I4+@T6%3aQmG&4;>_$qG}B@zp8Mhx#1EO7mfpf!X9wc}1VB zER@FAmQ0-*tL~{SjmA}0Ky#{PO>~DgwUXW6FPDnZ_jZWYp4Ur0Zd+)0&*785`(v>_ zwoXOaI!)XApQ=7WE@W20+g`)h+6n?D*?Qpz;5V-%Iob|*tKBEBLh7=bJAOwoc**Vf z=LE8GA5H#tpB>kGFIq-m!9I7)AK$peeCmN9zj7VL0o)dGbj}3X`{a-7KC(CW+$L3f zo%1Xm(LVN~L)H3FebKK)dg_A>W(Zd5*!bnjQBgEj6UdUW3zu*|aN`ny%lcIhsXej0 z=*8Y^5SIpL7^~-!_sV5^2TJ2TU=XJ3%71umhp2l8)2?QiKN3L?7yk3$OjHt$@x4Eu z(`xG%Osm*OcpVOuanPvS&b3FJ*Z!w=Ksy6tKmR3|31z$bq>3w}PlSun!3+jZuG2aqh|)}?V25l z>~)2xKXjzCbEnfESk?O^gXPO#CJLd>OT}dN-sYtQI#X5_YZvViyoG_bZi&gIKq6y)kjM(GIPOzf$@cHlMiN|Xoq;X1a|$- z{Qcv+eOpJqM~ePtubP|M*%%1agVVjLoEldk%lq7rqmp+gV#5mdS+XB-ip?I(mjMX& z98>?oG1?fTJP48y&>Ql+icYtn+iV1H?0mWElF8sGDB%p3bk2@+bKBRo;p*!LceG50 z90B$ycAWKBBp68_3LHPp5I_X-L}rw;#k6GEUeadjd7<-Vx2gwS-a!B8ApY9=q9TRQ z^|kf4KMle6)k$NQ))w4ywqwum(ER%d}e@B2EozE?RUi*;EM%C+}mZMzor^Iqu$y)A$-dgM5>{gz*EpLCdV+5&9d z2J&z0>P#plu-Vpyuc1>P!SJhPx6{+==XK}=Hsen29?^r@CaFdbj4Tvz-^|GcPNB^_VglW=jRObdR2&X>b(qs%HB za|V^|YOx!-ix1zkNnjUf%L2O4YP~;Q$u|3MRX+cF^oZ8o`W}}1c~%cUM}iN~fs@D=xMH+J=e8rZ7e>Yxx))BiW|kAs z#5$rEi81vg70Jo&PU4LYx3I4Yk=C#s!=b8d5M;xaa79hA;R-SYxNh#+zk-2aZut`JGvu zkb?mPC>^8)+@7P(Y;sK=d(mcH%tYZS>m<1F7tq*FOa=O?TirfHQ z+>-1j+jyJ24jeVr`dCq(Ur9bNS{H3YN6)1i-E8nZi5xSjXPpkJ8ov{(`p$k;0M>D? zy1mX`6tb*xzgJcFHk~uJ|MoXQ$DTZ+>Z!|^u(Hd1mOZB~UAIjCm9vGmaZH}AyH;M{ zvte-9cgjc0QjF@>sO-|tHhzHoy7f9vAmDak)m^*AT{8>BLg*RUOCG!-XnwmzPM%xK zy_`^AzEH`pcj&y$mtVUc=2em!I~(4B(=1=JFoHj6AI-hLW9Z(fvrkm3FMq7f3kH39 z6f0|b8RqYXiDGE{AhxO7Yrxrf)L;-px(hPNwqjfQ-X1}#u7iBAv8=Xs4ZaCl;jK(~ z$6s?D38?dC-tgw*!>n<$O6530vdX+lauBj`-+AlOyqg`Mh`O2kVR>BAs=2gv_AB+c z>sd`#P6YR43Oc%J#Hzj1^W45U>Ajjsth@Qr8yDiKbAMRkZf~%^`YvbBZnWd}XF3DR zi~Fyvr8rgW|-=?tI`#sRIj zUQBtZJk+?BZ#*l+p{`d!=#V>P_?iyVau%ABMTULRgualZ03&#D#v!I;-$-N5eC%2E zA=CzaaH%|%b*Q0UtM^gqqFr>sy`}UjLVXQxk`Zeti#|U95jQjyDRK3DIu;Tn+sy0~ zOsYsQ6JZWV3D~OgOarA1W(cSD&NrnMLSNOAgT%IHC1*v8+P#uMvhPAlmZTaebdqGT zm_#Bsl!L?^O@5*eE8cSN3V+xMhvQ?9O;dH0ECygIAXyIxiK8gED97-{JGZAHYr zU#XbF`f)|t7T7Ch$w?kZu;TGWvY~rV#R%Y((j6g+Fg8zU*;o|=3KB~kRxOsj`{fI{ zm6PeO3FOKtlm&?!b65(C?g@#dQ@>u8pOF;FZRXDxr!D4RRZz%$g>_0MLIP{#;t=Lh(K~tu8Ii6dOfL42*4>Bjs;=QDVsi$~)rWBCYa_WL zcZqOY`x)#7gArKZAU4#8xw~o;O}?72_?k0O2;9U3ycyAbZyi?_WfF{5|EzG7GGw0R zm^k0Dk52;yn)pV}{PP&Wn2;t_TpTWva{22apVIR*UX!Cy9fB$S_R5SeQoX>9bV~Ho z^K4jET{=4WF@N?SI8$=_^VsVcrZR;QY=sEk|H$|Re4Z-CVa`h5$t_1QN*W9XU zXx9FP5|KI^q`{J?C(xwD|9vjBv>{3*%9@*`b)FB{tfy>v$#VNLTbKsQB~k+M?S-s1^yfpPF#% zZvy(hQtf!V4iIkrYF&@nhZJ_>Wpc8?W6dfA+UE{=@m*U|fbS#5q)du!-9b(8j%F>s z^!InE+q=!2fIEF%9Ig(i`2A*}i|m=U=*2{3g*V?}*3(F_ec#QYdWEIr_3T0L+!^8x zdSGfSUPb0-Y%1=$Yd_}JxzT+NHt>FE_W9^`=D**DUb+9V{kFlY$22?4-Zso`1r6AR z_T68mvLF6j_|s>|%2NZnBr8Qcm-)NU`We! zCD=an^AnhZgL;VPb+Btnh!0bU=LtSL8}8%5XDd$srG?KqkOn{^jl7^xpl?tJI4TO@ z?@k&xfOZ;C1Mmmj{pF-V|APzd5TAnVdolx%p(#5GuXZcHW8z^No68#rYJI`|m-L5B zFx-uNDC0EFR62~^ZWF`&JekgqkK!Y7rM^Tyw>!8^du+jYL=T^g{gA7%Y@EQ``|Y{6 zv0SxUr`hq->nAH9;BQ<=r&(?<_VUE^(@LjJ=l9uU#4;wXtTB5W@uw6F{oW8nEb>TB zdjKG4Uy>fC=?^Z{1=SDOY^9&sCyqAbe)%)A`FZfLtYfo;Ot&n5@%T8y7&SkapL9MO!lcel5^lSOYPuo2(bDum8nN9Q!dvQM zS>{D%Qm(CNfnKZEf4~buGgcJWmbqEx2NPCK8r6G=ilT6IsY4>;${ctZEVQ&W+Q{_FB#ws+noZ>-E+&OB}~Ftx%e# z9oYD^13Nor5iBFiQ}|H_<~)S~O0924JC^;~?R(jNoX zK=wH9-k1EXd(xC~xHR~=n@k>nfNLCr=lygOG9tfq6EaHv^9?$&X=T=NysPD_fF1Vo zgz7MHoPZEKX-y3aEZK#$=*(Ai8K`Nx*j*=hTv9+zX6t4tk!?E|V^5#S6(w3lsI4PP zJGd{RYujH|)I=$be26u4OYtK7{@3{B3mnX*%~Q)USSsClP$wjnYzQxk5fC!DYB$h` z`(p9?xz5ZAxO?R}rHqXt9Evf+eX2nGxD5(+qM0trmi6yUu6DCyD8lcy^;1SR5bu+X z;<4YiVh?WH77R9;s=s+`gIlbvuHpIXyzTgx({TNcgO^3w?95+iU88YX@wki6>|D5? zh+*Yt3%?4az(@x_ z59!0Sly?kf*?rd*7k+bz4Z1GRMfx;Si}!LBB7l&G0tyiyAjY{46_?IK`#CZ|&V3yw zADM?yEIvrBc^$4kk%!qlGDz=v9ifkq|9Mz^h&k;#(p)+pXL)3Zz3n>6J~IEyaf!&; ztU~xg`wm`iXfH1uQw;e+8euXA|bQ^{P;YC3u^H7yJDd`4VkcLsR9Yx zyu^~HSfiXT;>gg9V~XK7iDF*8xVrX0mAbj4N^KKbJt`^nvN_0R_c3HHQ~*oI90UV^ zoKZ_+LJjjarJA*$Akuuol=wDv%9xBkOq|OYi6y0b!jz`kLx$yBU?_t=8LO5@uv##B z>T%)`M+oPXc~gDLnDG(Mro=RQOHSybH95z*1W3+~nI)3~#++qdLf$FODNQ8fnC*CV z%pEf>+boolKgL7O5~(L`b%IPdp;eYB3q94&zMt;n8abe;E{DQJP%HpOMt`dh@!@G$ z6t`_6B<~hc$P!XQPhK`y)j9O}TM>2+^PdtY@5{Rx>-ckUCJ?=F_iG77P^f}Imw+Vv zVX<)Uvn=|PV-f!+L-|IDrKp$N+|!CeO&OfU1bpX0zAuo#2d;VVrNmGm})bs}zPR43&tt~a%?$Fs6_N}cp&9H(| zf?Js>(_A#%63;;&1PC;2DfHFcNae%y1 zmqTqz{Wm-s7PJvLOhka~#UbKZ%F(dBX;SL1^h3zUz!BO9#Q0%^ZA6{aK^Il0M7)gs zOJ%Wv5TvUNwl61;4y}zmku`P)@F|_BMGoq9&V`sZd!p-{L^e=F09y95hOba`#Z4N( zj|<`SR&KgOvc2Fe)s)iF<)y}{hNM)^(PVc;D!#}Gn=Dho!OBZ}@S3&M))k^c0B-0f zp)BZ_N4~O*>s5j+?faRdVX8IM$lS}Omf^`7`-Inx6Vb%h`N>Py_Sem;=;n?=oe>W+ zt%QfE=B~%d-`>z-j`SNM zp?`#E_#I?_KmF3?R)ym~;b8kRN)^ffsrZ2`0mPGRI2leBO<-;)ZJ?Sj`oC8E;CJM1 zvFJG%i5}I5@@myToC4C=u~%s5pU!4bX^D0|v7UG2KzpgaYhGC_H~#11$2K2ZW=HF7 zYD>0M_J_l>@5w^@&=j9SiM-!-u`Z~qvy+g*-(+m~S4Gz@COu?7)*G7upL;;l$@h>p zvR;5<5+o%qHc)nG%QC^H7Lp3k(+G(IW?kzSh#^~&;}5P~k{fIS1JOg!wBer^N*V^E z8+soVN*qQhdzdZ2fP^9-_%W%z58#+OArciwoXJ=Ic-Id>=6ltTec$~zF`sB0A0Txkjte4Z>BT5_|jE{6(SX)W7Ual)_u z@^-rB*eh-tTZKgAs})~}4hSAII_OJ0$j8mPxo#15`XxGFh8emy0+>4%$)n1vggwsh zCO~g?FLlM* zAyIc@*J&V~g?XE@_q%EpC+*8cTr1w1ktb?u_U*+NM2xj4rOmz#z$@+JwCl1KSMT=( zUc2K6x%bw@Ad)K2B|ql$$5ZGKB(B1E&^NA&(7sHr&9vo7R{0hl`u3BC1SE$9U!Sa( z-3hCTr)!z6F8VZhbb{7vocAn+w2kfc_eWdc+2PZH4{Uri;i^FhL6@)i&1-WD0qBiz z%^>2%^~sPw`tWg^OgYa03>IPqh?eLJK}XmR2Hzq8NgRs+@kJ1xeTP5=4@nNwSvP>* zAMcj|1Xw=jh#y2BemWy03`imr7QQcW-F=3bp&-+hH?;L-Ug+0_{yP1~e{ zf>Fuo@71O2)Q8ioR&&tciDexm3o;I*L%Cdp2?84u(yk(lzaC3Wd#>GPJxvtzRU3jR zz*CPyU%yCVtOAAuQXGg*;7?@o)Ny#?P;!p2jaffw#``&mK4a_<(hg82A0WAa^|`S` z{L>liA6$clu!AJj+OMg71w?0TEK_SdQ>+|Qsc>y9H}yT!iu6>;r(dj4saA+&Q&aY9CgvU(3DH^}tr)DMF)pUZ&m@-&kqK3r_)* zSMSDosPe!o*JK!9?F2s11uT}Ch()Z_(kEBi+6Nio!ml>VC)a2Up3!;FuJ(gWm*~7* zdT}KmVrFmd>uq5xt+(~-*2WqeT1+b8y&a&{QZ67(-7B0wovSuWtG1Xv4O+in$MEzt z)w?fR>5al|<~_4l_Si(&>(~dJEJ-&EiP&mQaBrbUx;2_q+G@cVZhYWBHr?J=`_7jq zqda`-s=crFS|#2b&jRaCXDjnwHQc#{``iXFy$X6gSFYt_qx)1d!E01x)C9NHGT5`q!_=z1T7Sh&H zDc69+RN6yoj@B{drU1c6ggpp|9t0CWeM&pP6g{vNf$T&Gd#2)uz5p9ELAK)$@N_n= zK_LYc>aS1|HXd``r@AWIgek8AeVNT|!*wE+@+4|cM8OI9?MD|ch~1zGLlS1W`!}Xe zhQh?`5@(i-J_F9aCXYdk*{e&rlzzoKl~-Duvm9N#H0?Z%5qzPGcfGF^eaBR?x==13 zcg|OjveKC@pR3$*`kg$9tk*$EImS zbIDX!!jlADGS?ZXF$ zCp9lR3dO`e+30vB&j;n(^xvwh5edAd6v2FeR(tVmot*7*{i-q|#8lhtw zS}sYCn(W{9Fn@qa6+jSR4EX&oo&*eJwB&ycCQxa;#}!CU9DH@u_yC%;1%fp1QSnZo}7w}|KmogS|eo2rGPAOya$t0%T;)# z9@iF)lN~Uo2O)uoK_B}B*dGbOf_CH~F0F8b66N5;Be7oscXc98`kr&4m@(!KI)>2o zDwP31+4sr>N7!t@#wF|?ycXc;tu$M&BG$fIF#|}iHhhgyICmT%$<6iE1OXw6G{C|T z`5)l&=DPwI)bhK*STD(2A>h80yTMdsC_TvRZ#(_r{7=1m1k@jd`N+&eocqX(AG=~e zz}*Q6K;Vt?Aog4Gz@`Uyfw)n)WZsE!f;6|pej`k#4LWky;< zY*IQy4#XL;pK)(0UD-&04NVc%^yVNpIOTs%oqCezRo7Py+({GvCHXea ziwW|p?Znx)itp}R`_q_F|7u?sjw#d7hg5%V8r{8e13y~77J0icefp+gn_cxhzSPb7 zY7GxtjN>|Xn04w{k?^2(JBhPyc`x?zf@9Q7hAVlu3B#~+5AN zU9L}?s2hv;xmj2$@%y5=#8-*Vz0rL9Gh_NXoQ>O-Xuy{F0hx>}*Lw0c7pItdI-f0#b}?5Ib6jI|Z%Y_mph_ zdl|qAL$F4=!|6GtbORU_dSY8F4S00f&^!WpxsX$Z%-iUY<%b5Eq6( z=x5bcNI3HBT=3OX66e$)rwxWUVf}tGo2gllA+7lChIqR08A_aJppkS1=b+@xP`q2k zxL_4$eC<;$V}LV^H6InKuXr#4Zs~mPB7kY{>s& zLm_-FF^BGaTeLnEXK*gGT`O82d!^j!;CSiTtW^airUZ_Dt{8FCnJc4(WnZAdmN)t{H--z0XF@?qV&-w~7eE>GT zv;jo6hA)yfKpdhDM zQxu8imvYfTB)-Tvzv^Z}2-+bvLCpP_>e08<)C^N?7cMGu;=29tFIA*Q8dT<6%zl6x z)7*yDu~2WP{*OpAL8sJ0;b+JFJLyxVX|D0;Bc0p;#}jCBY3J-N)*P0oVjcoWcKY&$ zixDU;Q4+El7yP@TA9d-nScp@j@Qvlu(r1F{Ym|*ytmPp;m=!pXGP9qCv|0b`) z+*PwLZ%+EpJzx7jNFWwq^j~aK41{En{+R?q<1mK*eVc-;@GsjGQ;E>ONMN4iFA~rR zbNc_bO=-l8;&o^L+csr$h5h$fGZGXB=T`%V;9+R;#|y0BC43_A8=O^ zMW=XR{dSO`ZW7&{tKVag-dH~8x~~A3LU>B*>@EQ`Kp)h$Uai*^b4F}0B=#EOw;h*F zK{q_AO{P#^K&5pvUI09ja*mV4rqEXKYa9i>$BsvdK?sSzx+FFhbz)v15HORWvOAcP z<-y`_wngUy10WE3hFvg<0Dl)p@p}FQTM#Gzp7Yl-mNdcV4W$szl{8iW1z`M_MKNju z5Q5T3LY9j!B~3N|h%HK$l3fK-6hgGX`oqWf!e+v3Nf1$G_+lS>p$Q$Je#<9)KFAHF z#y%u)ixu3-#}J~nC`gQtI}BI$I3P{WpkFPGXQ#CoPWK{i8O<+dPO)?@$xAcw*CGSG z7$=4?pP3i7$d;WKYQ>=e(xrjt7O67PWoFJ^#O0QCPV~4tnHk#LC2~%xWao{tAJ@Z` zZAWL7g#dl-q#{ehbjtMnc*UWzi>9)&_Sw$0ilEQ3CbsEMrt9X-Z{FkQs@+K68EQ}S z#%QUEr>H3UM(8`j8o+reww;6V%PJGW?bd$}C4E@UNAF2~yDV0;e1#q*#oD-Rr>$SR z=?9R8;zh91urkSoT7Dnr2N_TNY8K!>tChmq>QLx0YZ^Tvc)atft}Vj9sHj8a`0A|6 z1-c$*czl|;UPjR6Rf-`JW4);{{tF;r(o4p!sxEFe82igqmYUx78d`CnLBBX!OtzI{ zXK^4iit;qie~kC!Q%g7QPpaZQXD~wJf66nSJGK%cn*J1r!oWLWFsZEcP{z z?m{0Pj1Ukkf1n5i=mS|=dJxsT{4krJ{L!7SBXkjru;i|S>6;A1z6cU_Lgb0EFGWC- z`WnHD%LCCSYNOp6Nbv4M#n58)!8sz+OV!DmM{I}yQH7s6oR@blov zxQ=Trf@{0@2bxgk3&4^GMZ%Oi$wKm}#vsYMA{tP}!V<269+!+^MA|ztsT*F4Kn64O z`3QaDXW|V>zp}WSw)_y2>usjyg~?Yf`Kgo*YYhC|1YA5v5Z;saFkxkUKojrLT>R;Q9$dtCyh&4ch9sql+io^Ex$mE-miQd@3SivljI zmu}l*Tj&DI^z$~S>f^661D0yd-E{kQt*=0i*xgh$hch8ka=?qGTBMDA06^VOyA2?i zKhA4!=b>5MKB3lUK?fdZ$ywl+qd;e)4-UUF1Es1wi%7{HfU9lb&)m|BKs{enRd z%Xp^PR%(LWtmY>o+C*LvYw70K6As)oI{e(*Ol;?hk^3}**}dB|vrHNQg4xWlAX}AI zLzU|Muf-S=6u{Hh(>CR_6`}qHKm#w9qPFF{Bk@6R&o4e6nfA5*rn{=-{0qN^by^e+ zwgwGE3s;@vjeOO|LWTFsW_PV+*Qwjg(O)(O+lOl}O$)Wib9M~}ZL8q41?^XKD9+Q6 zyDLz>O|5i4wOOQh2nm{75pA&|%hy++`cK{ZhtYnUFR!n9x?-o%={j2fUKgctgU9Gy zAj@ixJ#Y@on`v*jN*p#>WR!auClLL$&H3LgWB(TxMf=v{)MMA{`)^<5jvZ*Zr#{?| zt5CU)J*?@cA)b%x*qDw33c2Sot&f|O>5e0g>F25MAGbNkohM>)FLUW1cO`P2XJmE+ z=F!gg(9ftiCVC0~9KF~Base>_aIk-IIQ|j6KyV2Di^Czp3_ThPw^REcHn-nk=kzDR zh`}cQ=5WArf&JxhL~_7}lY0D(YlIqHV#^lF$;e(RsNxc*BAaayTS!#(HR`~plkrd< z@YnsCykR4GmT3jc4SiPI(3dd*xj--YMRN8(xYe43bztE(`u2KlyS$C@H8IglE_Mw(johGfRFq2-h;jAcU}M7Rnm~ z_9PpbKc~bnoHlPI8}Vav7D!=iEd&+Cxg-u3&2yGQ5{;T{kS3~)px*~ChF`iH&%k(U zoFLD<8XOCbL7pP8if6o+tc@w|FRrghVVYtb%~j;DVW6e&&kmPUu1Sa<9H6?|v&X#4E(dD@S&2jJX^7BfEw!`p#C`K&|uV{0=>!Y|nO*?$hPifGO>hf=6tKUW;Y?L!Gf;r4!c zm=N3A3+4qCOp-Q0z>2sFX|=tN^M3vQ;FklWlDgPeo^FHYQ9PA0ltx!C%LwY) zdXJVne?h($En=&3sj>QbD}I~d44hS{)`#T66B1~mT4uWnLekC>Aunn`PL%>;Z~8T1RZd;3V)lcRx$m?PQPZC(Q&@eZ5VEhvZyQ_TC{kPx(Y14>g3pO57N+IO% zB37&gRFn7fgn27r3-LP?{*^wYt}9{AZi^7dwA+l6g9yY6d*}<9+sygIX@EAPFW2fk z!lN@d-?at^>ia`Nwypi(5R@+^RLdEuVI=X6nt$W<(KW0A-o^-R=&B-V)NfIL`4Rr7SQHx?6> z6?U%Ff-fK`#oemf_03AeJ*b}gzKkevQuZC8F5mx&)d~D6`8t%7YYB(BDuu<;M9*=R z8}v|D&5t^2fFFXgYVkn4o|d%C`h~P%bQ&#we81`$c zmtRs9zU*i)_QW;@{|X_h>S17nmBE4G@*uv(wGvOHQRJIi4>7*|N1N5o!$S)xHy=Yx zj91)ewdf)FuH~SF@q;h*?po%Gc1g?Ml|3P!#>pPmv}Kx1qKu!%QV}duoe*T)+fqSk zq=IK@bT>!bP9fCWsVpL#enlB;)6@2!n8xomP!&@Pm;I1yI^8t3zKFO)$q?iZxX=*Ubqtg4yJR zD$f%EP^7JwjWtQ-d>TrtE5#yLH9vnP8gZqs&bZdrvElu))OuZ8nQCfkp8RF|{dIi@ zsk!CP{Zjht#))im8~V+u%b%?a@9^Yxyx*SduUoXa9DlVgNKA`0)qVP}i%?FVekAT`A%C)ofDy-B9@}THys)0eS#I zfd3LaP>ABg|6}Ll|NUbhHc%t*3G?5gyP-DF++(Uo6kwoYca?(^4E>BD9Es7NuYE_8 zu=KX9R!%dgh`1a?Pge9#{feO@p?=j(2rL5)kYnl6_4WOlH3Fn_+IF_KJN@7ZdcNQ3 zZxZWucsLvd;U3QXl!}qvgU_ku~pAI<*Vp8e*DJV9;`g3(%jWftn!O?F5m5QjN=N^kduqg0#t5-VBbuRZRa z4jzwk2a%l*;I<K%cfg?!%Ai8{`^`FK_dP5p!rWhIWlU^YDh{z{ zo3+$5&g&cFj4W>$n8i&`{reS2z+s*j4+YNi%>?3)1H$NOU2tEe{^;tmCTF7S*UgCzlFASo_ z_6LgB4!DvS&$`W?6O`9B7=@Dne0)`aF$^|A3={& z7$VZ%9b6qvVuqs`=H&|7J4gB*bRsrDw07zJl}Z=&r}@Am(KQLe+lC4IZ9jG1G!pH~ zj-mSJKKikl2%5KTFtO_acq9lPIt|@xBzHk}|2TO4uY_MXo_H`Ui@b;x^pLMr2l-** z!h9R_fNpsMfJsAa_hL=+`!Mf2@^(SVvtnmjsszo@;1YVz>Jw_JyQggFa__3yh<=*;(5bpzO=?< z5cfhJUZ9DjdkIHEt3=ENN$*#`(eaEOrvl*(GcnnO2{6`rTkj`7ybt^T!P{GY#rbyI znnhuS7w+y5+$B&D9D+N+o#5^i?(Xg`!7V_50KwheErB2b3PMun_rAN&?%jKw-Cuf) z{sW$m&poa=*PLr5hmf=*FaC;0X0Q89Xpx?&uuZ2L*R|yS!kQexRU z%6F1GS4THrq|#QWqtpBDSBG;I92UvP=sSvMb?1%#=U8@yO2a$@UG zoOf;Z91kp-jcEk+ydt%xT85hGef-WpDV8Li-IfaO+G|{$%GrO~iWR7|S63(euuQI- ziB3wXWxPG0I(q&#gqK<`fcr+NP=9@1S2IYk`9#kxPW=|%^3Ykr&qzxjztvsW+Eg-A zz9aWrrT@XJBLdk`Mf>BXf7u96G{?2_6?7@$GNUc$gjGFOX*l1tt1TD#r_EQ4g;0K- z?nNhC`_w7JldRUB7)K|)5MkXlsilUF_w^qFwpL^Ov%7v{-MC18T>B;Q(0xN#5nwvJ zy;4Th|0v+PQ}bQlQChcedE3TXDy!#*A<1Xru{W-$`=%)hu-fnMZ*SqiI1Nu9fF2PK=hF>!i__P19bX|3pbniTCf)l*EC3^&e-=z2L?XT9sm zHYiLKe6-&G?4RD+(l@~S`zsObPFxXK{YGJsRC?N)d8_p@t8)-P$>n{I5s!>_Y@4L} z&nPwIkfb5u4$39xs4k>^DlTEWWc%CZ6y9wT?h9WzwxZM+{D=r{NzjzC>U~*B-~SOj zf|%V}hX^bLh5v9b2x$K|Lfk4y2SJkN^ohjcd z&C(cpWo)qyww0+YK3lH~9h&O!V&-@%j;@xA4kyKW6D&6?m44ozV^PM7Ki7uPL2bJ; zWak#&23M+1-}m{R%;crgnKt!#49$<0Q**N4JY6h(t^7#*HhbA*!`uE0ex;1x3>*TY z?d{G|)g68x+K_fugWtHFz?F@;Kk=#1w=*}}$P7() z9oOq8VagubXA!Wp5OxrUfeR{jMQvp%S72Eop$Zq0P}nu?DYPq! zF)y%a^ zI@HO&(12N|kUDJWd%Fbt?Xl3E_E2b5>|gBpc{S;;zC|eTUeSo2)WgpY^+$sM@TJ%; z$}?0Q>?=uW=g$*-Fqbb7P$+0Fam%P6mo=3|4i0r?WM(pmKu5R(qgN@_4WtHT8~w7e z?{KresV8c3gjkYDm$0N2MS-%WzU$- z-KZbeY9T%q?9Id9i6~l1(8s^GwBsxXxD1f%30DB8O#&f^@A=gs?QCivT_>oRC^&Gt z`VIl@citG+5LP7xKDV>7hE~L%+wOB5Iuth%!`Ilxd|fEQkWsED?fLt@g`|d0%cBqR zO^^n;?}LjfgCD%x$ifRtX6A`_tPrU`P2Gezw+XfY2?OB2Wssjw5`badNmUp4{v`6x>i%bYt&gNY_ zlyQFcZf|3MwgtoeKHCMIT{m8Rx!29=*Jj4K`ab$ou8^ZF>P=ufVc(S9ecR8;-aBLY zxd>w8yCm|*R7%O9&#<()t>_3!i1931{PAM$$Km1hLn z$0&k@5fY3KB8xVMf&)v4QNRE`mJ&>KmQiHXFexAz3PL$aIHcFJKQzNN0#&3GsC_Qg zm0(WL06kz?=Z=$Dn2!=DFvoTgeixpSf|+1}p8_{r&nUya^?u_y4okw-|bS;@_WlwiYfs0T)~IQt4MIE;zQ$RzeQj%Q8ml_&PS?T8qpp-|f_VLAhHHtaE2p;HV}IzQq~%dxS02 zaiPUdy6j#QS{XQcshpiV({#O1gIUTdBJZCZ!oQMdGl-{Tez}-gG>GgBdh=FfQhr>5 z;S>FP2QAe6)zJa3yyz20T~Z_dZ}7?6dU7;&2p7A4yDDD2G~v~I5&x#%PTVdtEEYu_ z_7C3f_o*ly7Z%muqteliiRa!Pvv^Gz71-c`hOszBkK*QHH|Rb`BdPd&&7bAfo=&3+ z95VmP`o3sbr!)RfgBORr?xSiuQemqTL#P>$k7OO$238xkQ?idF);)x-JE%q;M|&iQ z2LN(Q5l^e55NCvbb%wc;N@H5!ru&WQXYjL&Yx z&>Uib`bHW*t9u{-K!9x315bq_Kv}AD?=ijpQOl z1#2y@jP|g3b5hulb5Vks`Vyw5Sfo5#<85$|*WcNnzkKCj6RdbreLW=llrzQ??54F< zHm2Zw@8=}$E`-Q7fB9DX&sI6a$-ZM2NjoRP-z##nlljGGu71 zamy?#=tc*w<$&M_>erzQ$(Wa=a#gPyJ>yI5;$F(6*@q_bh(xK&-t`{UvB5xOCiHb_ zK-qOTb$kDnW3}k^4(C&E*A9^rNu1)O8J^}%(9a;S>LX7U!B5k&(VkVtTpdDNZ8!twiK)UxEVFx_wM*m%bL!xG>YoG7%iyw=* z{#@(rV85Zkhv&5Lp1R@s^Ud0?XJ20`FJ!l5KOau3a!Nil1eji~>~*B3M#$>NeY*VQ z6C$MU(=Nnho#Ft>p^mZ*E+u9t)ADb@4_!gP6Osyf(hD)r3*xd3^5m7>jK-_?40%Eg zwsLnxS)=iFV=+M>;VfmqOT#ZU#}mX27eWXT2>p1U?HA&Ty08|sY#SuM5NeJ@2~hA7 zyg~eOgQ~<5h8%;3O5%(LvJ)B)`_e_X9t}1l#v`hUSbxA-LJg4Y4lQSjP-9SLxF>D4 z#RyS1P+12SfR!V4KOTh!*F+I2-RTH`DE;jGXV>sWVW`Sp7<91k7a_}6VtH8-AMYE~ zppzgc_*z6Tab#VT?VnI89R(S}7(ZyudlUT-5<9;LP(+k(q|C1nFJdpmZtZnKXtN&F zl_fgxz!7Q_TV}~PPDnie6f2PvQZ`N-kP#!v4Ru2eXhn@}L5lo3W}OuiL)RVOFC6Ei zjc>P}uy__{Lr0Y?6x|Tz@K!UP&pp;vDshoGE>a=N9w}k*Tb!qDLK;bAln<3;np`eR zlIVE+iEqq1lCWxpWNq8TnN^w=7V)YR4|@3ylj z)%Bza5a~BylKe3uL2XjblQ-3cHm^)b91^JS_z<0-iMksonIuK&mIE= zhk9C`dHElsdYhqonW1*xq)RJi$cbb~#%6e{W5`aRAhxEfvnB_gGrfTT-#}1wHh@<1 zK!%dc4;PunJ(;>9C~pQ)-q1*cZWvCxvW$kZhO;BfvsKWCqo={W3qWzBuEL8 z@fX85K*`TxR8`-C8F@0(tWv_Qhvsh^d zD}r&fz~%N}3tovl1C?|%{@RVa;!uuH&lDUYV2>!FZVg5pP?*zVUg)oYR91hqmb^4Q zaHl8OCo{KCM7B~SU;H4E(k-t#qX0URicMRvSgY8*5nL5r5Hx{r*Ok9EA-`-~i#se* zq@f+&K~Ut=Ua*}Do}Iw0^31=il@~ED!Z0tqS5)L(_qki7G`#aPj|~m*R$P6a3q@B-fn&m*U8`LP@hC`BlA8u7J9$DZ>Ac;^44H zuMqK|(E)0<;O!Hb)eV-Ze~s(!_H5W_Ql!n-FX--Rab~4&(D-@&JA2 zRTA7#x>JR+K*2}-)NtIU?j6USenF+!pq__3$nELx;-gzZ}4(Yg1ILZ z$}`_%`)Eg9ZCDogC#U@EAwXRI~S)u~&l$lbhLvu+EXEN>CGGhp-Pl-^>jBMPnDte=kZ ziiY?y2GsbXOCZT~dc1!sOg&GsFzNjsOiabfzwnD9@r%vF_fN-VgwQTJ3vZJF%q-+V z+hTLMo55Vl58T73mQjO{PjO6SgK0Z0%N(dAu7kL!6gPlxRN=2~-{_-N?r>2^7N7vc zL}fSV7tie$1e|$Cj~U^;29Ft1$-xNX@;)N+_q(JcMpAD54IE=MQ$Kz!smljJ5R1lZ z`Tfb}-gvBP`ZV(3n|A9EDmoa?)IvyID7==8-x*7-*l^YGtU0Z5@U0^$)@@2DkHL+J z5gfKSe7|A&9{4~%FiXIeOrwK@{jDdaJCEb1R346g%^3Ubm_2d4IrTK45MIV2IDjBz znqrH3Dlh!;MCd9w99+Ut*b#27PbTUab0oi+<#^iG97R-9#}*iOIio84&L=iFBH5*F z2DsNeGQQZIEbq8FR};Dkz^^4Q3UY}P!}=|THE$fS+?7miaBQsSI>&_}4Eb}c_4R7N zYMC7H%FzWIWw*)h?DV>(fo<+-$#jW)u2l^Uwl_v3ed6zn#0X1A*hbpQ636^&9q#mT znD__bn;3@qaCQXq*O6#Cg4Pk|KjCC^GV7_aJy<`z%-!YSpsEeUoM3VJ}; zXj;sIJtheNf$W}5bQagbr=i@E;$5zw%`x#HEdViR9?C>;mlDNLK});=kfp;*#R`t+Sag8Yj%g>@ z((^vlkS~LXd>sFY6gSn8hweV1Mx2>!F3w^J6Ri>R9evqcBDg($tNj!rtE5Tn=~+%p zYbeXhIPV|aUdi=`MMA?cHQi>U7(mIQ;`5>zL1dM|?=)^mH3^O#ORrYwx04IcOHWp6 zEpv877E}~nsm;rvsWPua!$>2Hb_61Dcbsztr)uW_6+r60%xc%a$FfsJyo_r3dsI%h z!zw*xL0nQMz(5=rqtW7rAp|?3w$8MdbMn_4HW{e}IDO~(aEmh-)n2#KcL{0r*ZbsL zNoS-0l|gWbGDT``@rhwon=>?sBw3{sMm}e(evS`-YFEEyga|Qzo?nDBDaI3RG)S7d z;ta+KR*}4IbW*-s>^NPVK$lXFTdHU_bF;O8veW~@b(R6;X({TaRl6l# zX|if=9h|Pv_P%xS3Ki4;L3SKL5_)BMCZ;*jn@!8cs1O_;$pv6oDf&*xV*j%~4M5&q z`N-Hz6kS(`f+N!fu3`~9iq$|?dt~_7!4!lT5QZd_J-B8bZ^*F1-j&krh_qiTE7G8i zfpOn=j>Q$9>aR1ILHyZ7W!en*^@tLcX86+DEz&q4O|W|y$=dZDxuJTL;2V(%Cs(Po z8K{+z`Ua-%Y%6PLN5%%;28{t}-lyhPp}Z7Y1O%3Mq$-do-jq@}vQw}}{*-zcXHF>w zQ}cb^!v`QI5;W2DQ9&J@uw6QRXug)ma~|)#DlcHLDO7b){ZTW8J`r8uc%;2oA=O^h zrFW@e9f@5b6<`zy8>rrmX%LsC`z&KPlDz{C#J|h+ZmAOO6!}%UR%z$k61rR!>=h1^ zWuFpO%t0^mlHq13SWhvR{qUmYtyJh>Ze+DGqMKyg{&6>6p>9=bDl^5qk_>>fwOKbS z7~7qA%`Wv>d$H*~V0Onj!7P7e1OZx~YU~nYHldM#o|ZLjd=)HqygPyd&FQwjiOyl) z1Tzy3kq0|@QOoazR>0mt@M_%xK5X-pb&XLvyB)Ix9^=7z^n&*(Z~cfVPVCwI2A*bI zbC{3!dS8ScnVj#VXejoQOuFR)oKvFPUe7td@GPL+dAOK=p!r}TDy3jfd%eGT8B~Vd zH`(Y}i&J&cAmL}Q?&J!$Lz4_P^z#PodX?cAo@$jn&-ed*m|g#Korf#3gBboau=mG8 zjX|u}-olgC{_IA2GGq6tDNU58NGHP~F6yQ1$2Saxo2+kK*SvFnS;t;?>3M;;Vq7$f z`1^MBErwENjxVd9|2{M)_y4d}`@OmK_py7f|Jr9RpdH1-DC4RX5+>vm2BNNSrHmOB zS0?2mW{4gadW`%yyZrspgbuKL{`Sq-pLj3H(k@sq9QJK4s4_+GB@02kO9iXo(Fi3a ziJAMfmp~x{%6a^@TFEqpqHJCUwU&GMJ z+n+Fpyb=Pa97IMlLzLV{r1pi<*g@$@!WdY>SnM!3Qa&bz`L=1G)JUOo$B=b+`%h$% zqvwQ+c87yts7?USYYldOMpkQX8ChS-N+>OJcZ7Hhlu9_VUn(h0X5JlJE%7V_!-v)7cRuXWnuoICMr57+zb{?jLJFdMYTResHz_; zZimWoLnB&E`R<6R&w}+SGF%*n+BF~faXmT*6w~lcJ3J>E6^0Sg9G}@8lL(7RTBkle zB$ClFZ0?3Ot?Ofx#0TrUGit_TMp6QAV^LV3|A?IamH2WN*KLyM8V$Mt#6--MxDamL zPrfm&EK#8~W+*u^5x$9*k?|s+$Z~y@A^pe{7Q#_|gJU}fcyodY43&X9{<$;hS|CQn z7gcg!_pCdLh9$00Kh71z+I(PvfEI_ip3o*7b)6Z0ZHLMtoGfaG%6JnmAD!$P_IBOF zlw=|SNrVw)B8_$<&8a(a*Eii24ocFKGUMNfxPYZm#2R1u#@xY@9(OG%}vz}W*npb9z+WC(Zg5-G*q6~27W{}!}ASeujz&8L}4hXQr zFWwJTf=%{zLXKm~+X0CY8+3!vKb7 z@t}!rU|wh_MQ$oZ$hO*(&PbHWPYAL!FOpgjvhpyxLi1ycGBOY@R?cNx){y?C=#we$ zGegTYeP)hIM^t%=e6!j-5En>g4V9W!fOyaU#e%R(N2OWsuwUi>| z;ojE9i90w6pyP_tXV;(eD45(D=91;rU1EeEFZGj#2?j>M}$XkpqvWEf&BGAo$OO z2J@!UzoiZ#j>8*ad`8mT3gF=XozO5$BluQKf;_8oxj?T_re1#6$M|qCS*PEY)mOW` zUzG-daoqbevIzH9s{&*!2- z!8IG)|NHUAw0lUUf8bxkB9a}os4wupfWNi`0InM14{pQA)nG>G=~wq`$DMB;g$5~)M>Bp2(V z`XG;{vu7fT+}WTc-F$JQh|~((iYYIU?3gYKYSVL)D2C@F9biIrcH$8vn8%bRYeB+V zpzdTnW?uDZ<=esO!GE`mFq88S% zoL%6 zj2+9^VwW}Hoa2=kUtSlL+PE;IOA0NsVx5Z}CY8Q7IPYE-k>37Y{6Tq26kR>=Tl4v> zcQi@LvUPez*(E#9hQg^H>)xKDbIZi&BJF*zYnlJk>(}~WP47S^ol7ER$Is%spDD^< z(!y&IoNXZX!HySp&#vxXJ^l=gr)jPcuv$UM%=Jn`OSlAEVwG1x1m@jpCW@=)u;;Ap z!_+V@$996PuN11qmjn5eQ>@3Tfnm5YZLa62<6{VtMS{ns7Cl?z zo|Z+U)oP4Xu<>!So~jbi^b=+AKMyGAexHxpu=@1-+w5wNARmdsZAjNF)9&uruGuL| zXTyu96Y>q61^gDf{3jCGOVN3BtGxJaB|e1kTbGy(698eK=+CctJ+(g%tMAmrE=Ds$ zN$+Qlyry7QRAz$+@+BdIFBe^3241e)k_Qk}`=p@cRo$gu$T@%f3?qH{z4sT;{RP-7 zWqa_o5HjlnMJ<>EU?=P&GM+>3HReIsc7y-kcAWtYx8{~2Rss*>W;KV?F_@!&rh)1) zG)qhcn7$m#e|_4xi4tspGLLhk10KK@lE<{9Z~&MVxL)qWRWXC;Ha^0%c@wB|Gt9y( z6UVB$z@jUoiPHwlq~e{BUs5`>q&ptr&1*p@PBB=@@*%#7)nLc$GpCsO=d(a5Bk!*% zREnp9p%J%H$c&)% z;AX4)S+H~HLyh`|5!CoC8TxAklHfx1X9i^^;LWKh$YuS+ zE1~V&##SF#j6l&G5wcwl1Sc?In=8V*LY~{+6*QmNoZihgfEHu1z|4rw!tbJ*a0XJ~X#-#pb8rQ$7%u%y>XW?juXZx8$Z>lEwCZdtsSRZ*nJT7PR2pHGlKV>p5B zplWihGHWF^H4tBcIKM(Dw4+(~?Hg(NSY3`5!d9_`+UH|dy;8h~ChtO{D7=#Q(tzKE;naS4=ux^;RrS?J8AE(dSfQXwa+_$ntVTtg`Pj0_MJ0AG#kI;Ap{Pgo8Alu z94VW^Hhl)3`acn)KmdFxWtiISRpafMTMw;k3^s-nETCqyX#Q>*9HtmqCh=T!pU7dQ zi)4ti|9;_33Kj-7Pq>u01$&w=0ZkJf`N*&t?SMVK#RWKl`2H|opMjV1R58Lsyc4<2 z_nY#>I3#D8Fw`t!R0droGW|6>e?RA&(&slBvcrUFBi-W!Zg&}7(;ll@y?}P40ytL z+RHYKPGrSi7E+S<>B%`Rl=!Y$$rb^o#!+$PW*oycU4|9jnbQJ__g?!@}@$hvds)ZJxS zAJ%U+FY-?N4%Fc`8y63zeX3vTkLdP8G(D_unkH>@E#kcnZNFaWQi&ZIYrU>p7tYFo zdMGW-cqN~iQ_+O|QB`k3j>ve@d6XiZ3h@6x9g}?<2%s)nC>X?#7;deEcK-@dyfig8==pMqoK6hI|esQow5*^Tud_60e6cNI`2(L+Mxu zehP&##SkQB`7n2faSw&Qfrar7xzmEe`MU`Pe8U+_!*D_2!KYy$5P2Wbe>I~?0XTp? zK<58^uGcq(;J;^R@5*H(tx&48NlGgIm7yJLXZ)8LO{IqZ4^T1A{$Cl|bM->2PMPE7 z#hRsRb^OHyr4BaPcu~KkO%ZN(DOKICago9{nBR<$f^i|r`@8F17K(}HUnc-TK)%1e z4>qFQ;tg)K(-?I_8^czF1)2}4KXj)ZYRl1>X!_r##r9=SPkxE!!x2y}(I%ZMLgObA zHvBcXSc&$FvPQS)6YAAu!cN9#A6>-No=z?)^kV`@~s#N%q>2^dV_ zUvnt^0U&%B3Cv5~jpG;Or%vGFd@7+7c+3GO%_I_6xRX_#AEnX^>+QL@beDaIWw^{K zj*!%q($GG6@=|=%L&URKgR2CwK9L}(NwXBl~y$Xo- zaI9IH&_Ht~JN%XYkX#Mh_s}%xrDr$O*IRFg>+#j&Tb@kVgWN-Wk73}(+wtsrb-nexq^rqunG`lk4gcHL$8}sm6Ib^;we&_S7 zU`}KIYv#Hmlx~2y8N;W$wz%^(t^*kW8y4HoM!R1liNQLnr;m6^CBOYhD7L-bKyG;! z#m?D&Fg9UR@MCzwxncTliuLO&O5;l8>m~~tKwXZMBg0&dlnM4B3PT#Tzik0Ryb}rh zwEVH{V*?m4$dsNc6}O^q+UT|VA$@Ud1>oo#&mkaLxWNJf434qHXSSlT>S<95t!gJr zk8K&Ix1s?*(ZC%$^#P6?zH1RURGniX)HWLX+FxcSc2d4L`5gIh={_Ij1DdT5`;kpx zXW=f5B8QP~e~AE@bB&_M^;0vVfX;R%^^2YWb6=Amhw;EDU;2IoJ zzxRoagz(Zl*^B0Q?|a&}p}G$@&~9>5G%*rkG%~KR@9E^odaw+93w6(VF%m4xL;SD? zD|}65>K*u_(8%f#2Li+Pf$aDbn7l0IBb#=a+@TzR#tY#L2iFV&Q_`O7P$&BHf^=F#u(7Q08KMRdz5B&Qoefs_3BnAhl(fO!JO8o9Yq0 zw9=D&tq8ARN?-*8J(?mnkOq)7O-jOpCimWiH$AJ}>aM$bJhFr_qei8YO>bgSx0g4% z(ox~#u>M=RN3IAnfNt88la%vazKRga| zEv5l-j_3qaFGfS5qjU(Er&42b3YcVC1eUKRKX z72#vSRCA$YNEk$1yd?g-BSjx8O!visD&bTSS=3gxGCQ#;ery#b*v1MX#W`~c!!jWY zo+Tt)wkq+4p0xG1HKrBbbK73pm5ib1qV!5aK9jgrFIM)_UR~-1W6QpQ=b;EZ`So!$MIKjD*Z2T_XzxH|t1Dpw0XIGp{2Q{_;M?F9 z0BQ>H6y(tm{k2P%KPS($D7m>Ka4+n z6ZYe2KVSpHicTV~<9#4Ip)L8?lf_n|wXQ);DdP3Yn zHf^h#S#A+A$v@Ylp;~ueM7(ka5yG4czxK$-|8ou{6QeI^Sl$q^Ey{^24a7ipwry%-QRzouusJHLbE5 zJPGmCpATWpMbC#i!4st}1V`D;`KF~3Ezxj~tN6Z?xu?HyyUOA}MKN%u10hNxBX_)c z=D)8!*UEhdpVre&21+Vmf`7Z7pZfRE!FxlefNJ^JbTa#zo4qws`G4LGP=f zf&A(O-2sx-_p36v#xVjtzo2~CI=%_VusE=hMUI7d3Zu`aHDK_2WBzd%NTN*5e0Y z@q;7@!@>z8`Uzvc3FA2llidl^>j|^4gn5$0Md8F{{lrz@#PyuS&F;kQ^~7CR;yy{z zp>Wc%e$uIL(s@qOWp~n#^`vW9(hW)SopACm{p3gAWLQq}OLy|?dNPu}dUU7x+wa;tW z)KRd}GH#K;`HRG^?l5HPfr!Cb(@D2F6myhyrBr9TUc#X$#nc?TtTq_Smw&aodwtdzUAN_cN$t~$0u{^8 zg^XQkDh|C?Cus=ltwDXC)d7j}B5-DUiC5pb^SN{c*J|O=$K9hjiby~oOrzrZjG3$F z`tRhN$Vq|s_JpmO%6@J3ZWwwJ>Yyu9kI#MtErRQs$B~pzX$T350(vMSp~6871>^XB zESBI{Q3PJvIh|a=`6@MYR;rYxCBOS{X^1GN(7rbp%0_A|E78P$xGP$%l`N6NxmDC3 z5mIY!$ZF?tCJt`wVHWoAlS#aOl)*d;Zyah#h>hyTQO0~I%V|Ekz5Q_ko3t=E$JL^C zD@x6KoH>J9R}mWH7sbk)&C=v+Tk7geX9@i%-CLe4fuh7xZb7AFo2pP0WncB)hNZ4p z)BpUmq5MMRSHO^Qi$t-EMnPjUklw1FvR+=VySyASm1}(|P z7CWrVuR3Ex@D)`qE#Fu4KW*$p@4jlCF;u2&{~~HT5%3xb!x1K~8VR%c!7&Dv}8S7{$1bmg}{ z2{#tr#xprCsLOLaENkmIJ*?>4Ha@I=@VRqv>1-d!b6Z5+#9=GtjfD$5u9FEDGO8d3= zQoKqpS(Mp0sQPwFgXSV0hRIJHZHjV#l^9v_6&88#Wn8K!U-_wa*Kmu5!_-M%2y+^0 zBGvJ3@Ul=r{x&_}!=i>igNQubZ!V#K;k}FCM82x~U3$?kECet&c=`{gx8uLCYH*Ui z`3}(Jc*1UpHf7Md2#~Xb0#su%b*T0c>iGfOI%v}tC`&nC2~OE^Zz4yvQG)d~tU0U| zXY>s3^Hv+6%vB0dGmHED{ShWUv~O*%I~I8dDrZ9E`^e5Ont5H?%);5vk?wg~G@T7h z94F84NfbwqxZPuB$y~qojiH~|XWP|~rHeVT-Qz;==qci{qFMmeP2NxKN_;YP=ts6+ zd7H;o0+FP1blZo7*d(>0XnDj1qy)bAf!{c;-Qw9O_R)aW)ncY-8mMm$sGn*eatE;s zhza<#Vn>wk>?e_uPd)Om3N1u)q1`Alo^__(R?4Z_lP;oz0L<=hTv=w&(kQKJz#kg~ zo0nvxS70U38uB(gZTWkoN2Z*$KM znriK(XAM8Oj!&|08w8S-Qik8A`+!y&M7gF0ql1?kWOOcU7~>5*;~q0(8KJr>(WKwt zMUS0dtf0hwU1>vi*z9M49Kz&-n~Qjm?tQC40Lo$Z=`OYx&>n(cp}cd!8``_p!Rg)g zaR&|Z1pQ!#NDQKZa_Jvnm~eCXuLbNr)~)qlKMVtaW+=dtA0ds+T$eTh;J+_E;T0^h z<5p8q<%wm3t(i{2=zV6d6zfA5xnMhu3quiyo4WM!ZXIA1@4 zTeKz4>l+?fP(>4a5|5P9nUKA^>3%23N_&jBk_Cy zAOa?2`FYlRmapik!Bz|vkpu#eIr$se;B6lxI&*;ufTm)a3HL4}d-InwVqGZ8Z!Qc0 zkVKlq1xd0Mk``auD&*f^YUUd*D$I|SWUloQ(8c2;0HSKt-hY&*FJFePkAx1Lo}h>N zCv{*8Hl{ZxO8J0Okv{w?&gS@T!}oh5%e%9|W5#soc|K^^yS=c(>Cy?2ex)8eCj_NR z`fSK>b9$`3eZ|;8;f!q3^7}(I@C-^ZC>9or6>8w+Cd#v9 zjKp`%C*!5Ds;AT<6OTIbfs>@Hfa8esOVBCqdlTw_W&z`?PlVpP>^|Mc;f=oENYk!` z0+Z&1Og|3hJ6(|3`}Z>@`)bl3T*kYIBlXM{L!SzN;cx|P%oe@0e?oogxDGsm^?S_2 zCsa~V|L%Ex@=U78;!fQUAV+R zHFf8mt(6eo62q%c`h@DFsnCPszzg4ub&Jkk230Coj3q`4J#<6^y2S(i;);Rdiy;jL zNdZ8(BzCy^p?JQbp;<`8cA<)35SkfIW*a3P41-=jjKMdINjU7<73a|vhkZSa>pbku zdD!E8m;fl8&n_IUEm&ej7H4+$rwf)kNw<<^OiK*V2DBL-K9Tz{d`!6J|Vk<0>8L(E+4X@!n-A0L9s9QR5s(#PZl^X=E8yc zB3_aHNDHy3w}iGn(RMMMvFXAQX>q6vkqSgElss`L8vnd>WNk2haB8e#SbSJpYyn~% z_Cb7COw@1;v|2x!WGJptU(kXCm-URN#TVSs9aY{2)(?$e>K5uF5iNpQ6_do(rYG>8 z24m4CNoOTuA|>qSSfa=#U4U>;LE@TYiQD5zSFpF{rMSP=Eq}r!&smZm^etT;lkr=U z&Bno84}>!;DWaVT=($Pt3n^gIRD~=&YnD{;u@s`*R5DV_=k8SGi&PrF7+N%ot#f{Q zQY$pSv=;nilomdO+O!r+spYQ;+|SnS2;^g*w3I~s(zy)M>)!>(u%z)trAlmwP#RiT8s=VluAWPaGlGQoCCzph$@VnJ_V&y6&CT}j$qwAe4t~ytlIDboXDU z0u}(W2>-`MCK%tH?qB`OxC$oSK7F3D|9{px{D0h&X*pH&{&VMl?8*GEr###$yOOH6 z`BOw#k*ASGqjx_Hd4v{_@y@Ao_dd)|Wah*Bg8$HyB3n&HW6nP(aUgdAxV0ry2Bdy> zo?v9mnM%Rof05UzrCu0EiXdp1p8IxP>%hTqupG-l1c6-57T0fs2weFKjpYL->ZtLB z@Uz0Rl-MW2?p%%u1lNUl%3&7E1b&lW`yQ|b6aZ;~oqnpEgiSlG93;)1-Uq`;5A3lN zw0fAJzRFUW$oY6m7p?D~!x&?f7Gt8w=I~e)qLC$H9W7+2c$^~H%6cp(vzv1SQBbM| zq;1Uj8H95d+1q5;Uq0?88O7C{lzz}WW{d-so}WV!f>+OqgBMB8oyCSXs;XR2431MD z&@xiJtKJ*3R%&00Y-Prf<(=14`@>^X>S}W194fOfv+S!WZhM)k3&#Dxja6^`E?a8r z49;5{le&+x%XV$pTWwGCwwn5hp6jd0!^k+iMR)z{JGxuY+1t+8Uf#GY#j!K?EIQ{m z=5PL0;4ES~R6Hwu6}>3=oH4AtRnkP1(A0pDS?yY5ejQs1glm$!W{wCF6xke%p3%n8 zB!#7ZD^EFelckA&mS4%<=0Q0}X)x@yOm$cP5yp}-}+GdNDY$^K)X4D9e2o}^gkPX}L!p;plYI7(q z7Kj@R?r?H50isZt4I&0S!DpWKf>E)%cX7B6M#ujU$(T+dbJUQMQqe;m*E95-pHD1> z_7$CJT|$Kp{K6?kPbjx^ma1ERQ6T?}Lk zheEjoE5yPO{{cuQ?N7jO^@+oRv?8ZEO=w-d@`l8XWeUWs|{|6)QyLtZTquv7Mr zhfxF(7oGjl#G#0AXEz+wzlVf(mV4j*^@c)yF~hUV_Yf22`w^k)LKW*Wz)N}I9g&}r z$gvrYbv0tt*HH=00kjJ#NW7aRw_bV#=0OTEa+G0Lk=*r3n0qlIe2Y<$Z|NR~8>%QgRKDmmh~>=fRSDh63}UW+_qz4fG(LaLdUm6m~D; z%A_kQpza(04jM!y1%tg!uo#Jz3YZ*ZB4DWcc^)X8w28r=cGsv~$kLE4E(Mp)N@I}V^43)6?mc*6y0A&f6w7hX3cT%k zZCpx4z<`s`H(@r{)N0A%+BC9u>UrC?fl=i?AiI9)H{ZHDb}SWPJ=alNSZ}0G+LJdlQr{T-G-wh|dY;%BS*ehp*ro8wy zkRzfFee3wE3$wo+&||s62dPU+-rlmGSsOQaDh}j1OMFhbHF0!3+snfIuYcK|CaApAoH%u@(I}Kp!g3^M z`gs2`e8l8Lm$!qNh2dbdnl3c7^26?{mDHE^QS#0LRt!I&#S`S4QTAu$#19*c4#9Ti zIiy-9n~Z9m`NgktSp2`ix$#bHdPL;?rJhxws2GX~^2V><*I8We9A$vgU+IKb*WbQ6 z0Vu6d6_DTP&i0UvgdI-up)xf0{mN{xx~-U~#PI%{f70!aBz>k(CJpax8TUXWP^n*D zq2_c-72LW8cplL@BC}^Fe8YPn-28QD+9rZ zCxf<3ME{L~_s1QD^^vT-0Q*0MIx(#?4Tc(y_@*->O8o=<>6RbzHWnZX`BF^C%=HD6 zDICGf5Sf_SQDwF~W~nspqCe;_fLBZjRrE8F3M6A`WL>H=C}N%)ZmiI062=8l^0{rS zwmIW(knzQHeC_=B8~r03=%N2We?ABOufNgXhlmFN<%00XF5z*rGL2>QIU=s{JA(lE z#+Y(~Ee+N~qFCvfk&V2zOI1!sK@qc;=3Ef*=r!*4X8Wn&Ax1^vnETVc$%@=Y`OHSG z)u}q`qEK#+^RcdX{?80-0^W*CYE5MZ6ppRehqEQcvB%#P@uubox+ zpja3zkXl??&HL#QF2J6(7#wIZeMUosr(wB1Hd3^o0Oi~o2KPt~ec457aO^NZ6x_~%YfxNR z@HMMHAOHqZZ7*+sMq* zY0sUerK5j}kI7NUU(f$khd9lb7V_{ch{PWsryIHna4w0w-Vk5r&P3m``%*vvTeNAj z;aROge71$+V&GNtPU~OUmMFphD!&6FtX_t}J#C)UE(H~l)B*ny`apEvJCEP4=ON`d zB7aMdEVf<+#ALk~5yIDPcfr*lG&NAUMEU!hNpY5t-qYm#b~-j^_)(SFLHd@Oh6z`c(M+JfSr_uKrJqP%<`2Zp@;WWB zBoRCb=ylqVft+UeHbx1J^N0TCigu^u}2W?io=L z7xq(;^Uv}tVG%&jITb1!V2t}ks*euG5G27WEs0m%N7E?|FsK$K8rTnf8Fmq2CYMLl z!3hPV_~I>HN)dGIr(xoVd6Kr%h+aw!&_?=UxNc{I1ds>KfQ~VNKl1>ODKu}mYBhad z(27sq49aNNvVliXF*)v&qsm9TB*dq_S0;(K7mj+{A{9y|>}=~dVd2(BbK+e-wFjy?!qE^l%kB^|J`_ z)?&h!1kVR3R4jV7Hy$YPBL|E>#7o1B{XzLhF435UIDM&7fM$Ij<+bXQu3tlpL}(s8 z8?xc}PI5x4vr@0GcBL6fQSvTw0h3FbyxXu!$Yy;J1fx{(`FWwx9shEw!gG%8AQfK@`T<*B96m%88Ruk6yc{u})xQP;VC z50wAD8WI6a04RWS|MM-n)D%@okSVj3ocomFK;&;=p?wE{^~KYo#K=gWe~|UfgTgb4 zLuVig#MbMpshki7Q2Ou&-q0l7-?360)zEOJh%QewlbNR!0uKa%Cz%z?7JLnQ?C4VP zj~5I6j+L%d(0`1TuK@r!M9ddcBewtpnoswqYe%;sKpHgodc6!+^a(#+Vp*SN=^Y(q zme;XE5vdWoqv=AyPjW%BP!QHna`zZF7k#*LELawqSK~2~+^9YFaIN7^ek{Z9u~OGm zIzC5(`Te+eixU?>RrZTQ$HNcAFNjgh;~i~oxp`8Z&)0HcxE|xXuU*u?`1~5)FV)fc z@;-Br7-Z64je<43?-ua%i@{KQP{`cSX)F;)@d;gh0LlPw)3gQ!Arcw75pYR_>E%_t zUJlLkl?!Q$YkVWTrX(96_z zfwY_9tYAVH=`JuJ?i?T;tO3N>XfR23A+WL#?FL$!q>2(~W#whiDz3pH@F~LBsy5t> z<9GS#_j1seoT-Y}*FsGr%PpAnfK`9qOI7q|RH~ACKbhea@hs`(%ACkJ+VoQX4H&4a zUD?@ZH>@O=jUPQ$^_|*%YiyPIaNkdEWElxXGFMa6KlHtCf){opPh75Q%%k zDx04R#!8Iq=cQu_?p^j-ZiAC9lbz8QmiAb@ zd}5nZIX6SRN;xUFW$l_M@tLe~@r3>P)9trk-&@Oqm7^tI&NGd$>9}aWI-%Z>+xfIJ zzNnjBVb|N^kViIhMcPcX%(^v|B8@?XvEzu0XbAw!#*3ToNx1)@$Aq)(EYX z9q0w9tIZ`qdV)vQLrF*q}qb;&cgqcL!&kHu~l9~%1}|i*dvO2HT^f}$7Hp5K=?r^jDl&4 z;@^XQw#EG7#l!_JW36pyX}cH)q^Xr-lD3WMUj19p4|)Nkr1T54xZl8{N<^=658Jbi zORSD4U^SN>(O$kxY91|McVnv4xcm()=!KlY(xb*Ce}JWsD|u+tOyDYYGNO>XNP5gl z`zmd5w2-%HXzXoeZTbfH_IU;$_+`*lMwAVWzyiM%Rqa*gK|s+Ptb=ine>6mz9G2uy z_)IIsJ)tvFb}>!*k^IO;mM<119G(pN11yonVhl1A18X`Hb=P2%Co!1|WM3eKo zaE~*y$;|>>Q4HCF23$uxe4~1)4d(_aLI<-EB_htqjD_yseaKs_h?1 zdi9|n9ZfG5QcENGo zH!-v>Jq%)1+TwhC=PaWRQ5pg|bk)LSuP1n?U+ z`N*0Zp-5c!gvOdeioI-AKamUkkltu+Cvi<{eoG!UT1&(P`7Xg1nWvLzDQmP zHKz`x9hX>$-^1Sk;5Mzs(?{*q?CGZmxJpr_7=#_DA(-;!?D_;sEPsoY;SDt@kfuh~ z#=TvgfduW96cyy>e|0O4Bw%h$ghSx8pWu+idp>9Ku}QKza#?+;wUR-?3LyMdNk>qS+y>1r$h-S?q6DW@l1F zZDmkz<)e1g`lB)NsM2xQdcjG3H@H0fV4kH~ z-+WFzI#-PDhOW|mYL-;fZ>_#}F~q*M!cyh7JuA;xwGC6Zt%xt*UChi7jbi9b1~4u@QNP9Bx%&sd$Y=B)t&mSxVl9gc$p@{Mi7VTJB8qQmd2Y82 z2!)pR(PEzNyK$W*AK^L-I=^*vEluuT?O}bmhQFxlI-!yox_gp@%qmtwk>yaH*l{}H z1*(}$yt3hAxnL;TV0xtXY($QxZcNW}e@gTmWJ#&7IBDkrjB+`s8iQX6TB{u~ymQda zrFuw~&x80zb1>Zod&nQogCXd-VDVt7UTTty5GlD_oaDh?dVz~j#fU!-cOSF%MVRJj zEbt*o^+gnjpKTUSxBC6 zJSsda0icrjKC{#&1b_?zsa2WvS>apvWXSY{k(GfD7kmvJFyX-vNN(a#22>lu;>a+Cb?=O}dg57-#4v9T_9kgsljI*= zPgbj)_UP4uR8o~}cUc5#n20(e!0`fw6jm#uCHxT?vPKD^mi3Xo`aGm4e;F@wO7+J#O=(k~;Dew?2R{Q%+s6`5ywlOtwC zY&SD8YRxPW7fG(TEYJ|M3N@nuPI6MUJf$m(mg34Te&i_V$s8Oc1ZfdlIp#cxZ)6uU z&mAO}FB9YhrgcB3?OFg2K=8lVP&YGKBQ|gwNAVpNZ-xq~FHnE+dRj7190evG))IE# zvqaY+NQc!!6wpcv;>HCIEM&RbRLeG6t>_H5j1Nv)#)3%g0+y@LoJm%nQVO)ni~ z2skEZeuPn(vpNpaAu_hYzgOMwaO^+WuNM#?p=vZ&(9B9reM$ADsm$k)36O^q1W!rX zvv^U0kbJDzf5?7kRsMxX5eA1qSz=p(H11^`(kj}vRmc3mm>%9ous8AchLS6z1`gN0 zY)Fm}o}o%mM)stBUP7CEw5k17O&by`|Ho2BYjnEA{0GyCx3IgdiRR$cl^-jJfti;H zgpRI%HT*!}3CS}GM`Y@z%;d~9LDX3Wq|1HP2ujOUzd4K|XO->SV_W;_hCHq6J66-< zZ<@TS>k4L3J_H!%&%Xk0^e)TBNk>~3bR>GKqkhp~F{a0iZ zWAvRC)RG#NTt`u4=V^H}$Gq3%3QoxW8}_t&GW|*lNkmr#T=$F}a6hM_B~bH6)tiS2 zI?rM%%c;1O#7ESZa1seb$*EjaGS_+wGxoQm_$kyz0WTSN46cR_8<}Dmsg=SM>vDVT zQlE*dYGM*qtQeq%(i2voS+$>Ulq*h~WB!mbL|;aI9L3Zyu0Srr21XTlZAr>vI#x?* zG!Zdv#TJVME=Hje2_~x%YN0EsX%jz7PSNRR$Tk67x21@UL%7bWZ;=qf;d?Jnw47yX zXK``n_h;Av5LzHu1X`+^C!+>or7zXUw4zK)19&Y`7YR=tXdoD4)!>K~y%9O8jU$=O zdQvSB+p*Nzu?o*ix+cKO2Eh?cLg~3Ts(cWR4ITv+VK|LId_xd|z6kWY5PVS=BeDq{ zOH2(8v?vi@%7oN$cKqpUUG}eXi+4v_ZmT_HH`N1{XF1w4jY$nf@@zH+Wy^KXiOy8$ z6=;@{BP(k`Ni|i?;`YHTtD88t&HeK1z*&Lf+|7hyA-z+_yd&KMVZyp;Iv7QL*;<~X zTiZ(HF_|53{S0TLZl!~*Gz)F*2FIoKB=F1wWdrSDdV!!5cJ}FN3)sV^IRXD-W7TPR( zkyA=N0)wPJy8b~nG4IN%!Ruir8?&MK@3qjwY_UQ^sm9o$dfUTXbyP#Sx%`p#%ENr~ctd4g{EFV=!vYMm zu{v1(*qHQjaZsVLHhJvWOz?4OGODq@NdEt$xng1L)V1w#Z8e~&eXk=vb+vRM9~1dM zS~jVFPYc4p*?*lDd`;lvWEc%|M1wF9xbQfYHVA+KiO8uzCCETDsZi7sZ5$pGB}fW& zTkJ6VU#A#b9ypP$wnx!!7{cG`uC+ucTw{S&ed{y>ULhPMna%&%AB=|qAk>E4 zP=q)+kJ_E#M1O?oa$+KILg2e_@(SM5&6(ml5>9()!=DS_5q4V)+k1U^+WljEmu1jlo2g5OK8$M2hSgqq42tV~S0U+)dL( z$Z2yh+DU-C{q8552_FLIIkfBOsRFof0DZo{g<&IcU-Wq(#j4JUg8K=-2jHkn4HH4c zc-MkRWd^8RDEx@Dz^jzK8=;J?eKQIKjZ)vkO#t?=2<~}{tw_G{MJNM#j)k%e(aYBc z4nr6v8SZDXFJ+vNA-g6fBp`pe5M?`}0=I5}D5f|He{!x!`=?N9=>rk*ZGQ=Z)cjAP zyXMNo>4Pa-8Lx&BS0d!xYw}Y-HOO0E*!xq+5(5YX^uRioCHuMj7qbl6F^!B2nF*VV zP(tHcq}8+w?1QYLk;v(Mw;6GM(~u%{ezKBpmZ_8CtS_IL%RCNaQS$Qu-pdD-L){wN zK8c7rdvNI94ztMv55yMv*dTC04!QBd_xfOZfun{6=8@ytm#9Y-K*7Snbbo7Vl+qSS z5`Gzw0*QcW>A4-BX;Nqu6?WrYDe|^E1`5)gf>lu#1Rr)gL-b%FUyXk zOyB?sYsMUD&GfT=CW&^ohlzL6RPI9HF!AgGe%*22#wjATO!h5sD+6{WSK7~((ZBh@C0By~J2R(f1P+g0w?uP*az2aShGTQf{DPb*dP6dq);I znw*E|Ey^68j`I5zW^dP0{1|wpz@>|X+$}UB^PcMkztklsFs+($fJ91G8&u7^R$+nG zvdYMH+XA|)1oapjW*WX~4N4gIDH@^Rd#eCZOD3zg5YN1}*qm&;i=d3CcRM;*MhQ%*a7yDV? zA$+}}>XTr*($9##zA3qrQ1zbWAo%27EFd^4#yRu^*#bgB4CJ*hOIY=$t#?Ey_XDly zBqHyLFc*}1P~#2uwBrrolJ|wjr>~M>1aS(ngPj!wVf(|Bl6dksAtu>Gi9Z(x4_Ipw z2YZpJMd)jrEgjh292HO)B>(kr|EQL_$K6RuY-VE$=r57+t(>0!4K z2=j5$=tPThef{vLfMT$A< z?j5`wn@kyQ`20JH<#|y$BU0UFX8nPnMN`}EvN0PdK8b3~z;!b!$9dBdZP9J8`@v_%>z->hnVLVx$6#AdBHlmg(PAzV0})J@br@ z3Asn42Ht>SLhzU2zffUr3mt(jz&GF{1$BDFs38t`!?ks#1=!5asp}^s{WE30_9mi_ z9=25aKvpP7&H@^ePPvH~dcc-L!oai6Z06{{-0~~fQy^3{B8$L z#!)i~xW%dHpn3p-w}rvtS3*!rec?;n0|h7d>GaNe04v+J^aq{^L=hJeRW^Xe#@$>< z)|y`r*O9tXh~EQ@BSP{OZ2CY-7RY z{)bm9l#i!0&2Ht^{u`u=hzx_}SmZ;W_8z~3lTSrJ!VYVEZ#DZ?Zorb7yXZm5Wftm%+X@U zKl(3)3?4*+l8NiJ6YAs;?w-Pz*Zi?Ss9^$B=>u=<%T~*5CExea0{D}7Qln*>4D)yZ z0AMVOx%&ydt&X)P6*O5&7BYVBfhe?6Q7^?Pmmgk0UNe(s6>Xo$f)kUKsj%0Rmpe27!@B} zb*b-xxK_dILA#m#Miz$+4AgQs#umAOjF>L4FxJNX&2Ua~dw94kcf4=jQz&x=E&|$; zGSU2&NaS2Fu=IAYl&h42Dh+Q-tjw@20)UtaB~e-N?co4TqSktoZiHdbkg>mn^sKrB zU19NV=sUUt`Z$Qq)^-L>!O(6R0f=%di-moWA>8MZ_ye52Nj<*Eh`c3*&(z7@zVp&2 z!gY^?Tn+MIXu0HGp*EG63`m6+wWRpwV&F>_T%|>Nijgo3MhHrH2F7K&ucv+q`WUpE zQ<(?E4Jgb&3^%LjT^~NIVaGv@^Tq+S>_q8X5uY%M%Al?{v>3PMSFZ{rC58x^rW%q!px zgEnkVmDzTJNoR2!fl?vo-Fap<2CZd?gk}Sj(I{n*V7vN{?KYt5G7AW$AF^e~rBzl3 z_Cm(w&`1v5(4d1!(^Y!SwLmX@-TOZBncs@emVOG?)q{Hec3IA&4! zB?r89>d{Rb&HWzPGFEwY;uemv>LmO95c5Q7DxdyW`IKz(Uc1mUfSRWx`HcKH`uF5b zr(r;fVLRPbw+Mew*hb_Cwf>xwCwbKob=kg4Mg1D5{@bUjYo6QFC;9J!iU7FVyFMJ$ z4S)$kISsrD-WE9n+Y_I;k$j$x(4pDuLm=Ac^t442*1{SD&&jvhaC$(g+h{;%E|gq$ zt&|)R5bHZaMfWb`$q}(jzrgHWWBSrB&Xg+RRt|5j=}7G7r#`4Cwyl%lRTfU7dc_r# zfoEm}f3C6b9k!`F<+tT5P;7r+9xxUHLN&6V4DQtEP-LRNGODrKuf{s43|m#+dVDg4 z^=Q#L_$ej8Vc&Essf`FGTIE%;BM&TW#4__3MePwRLBOSvYE)?7;WQ;{eS52^9Rt!S z*M9ViIx@$9%G;_5Yc9WczVOnsA@5OXG`gOnAHGCI%RE?Ej zZM;SAvOmOk$(E99G6ytQ8^OH1NJZ;Wr__-iG$z3bCq3TE&EhAE3S>`ZvB{xZavM#+ zcqN@#Z{~%GFq%N8A~#Y$wVM&&uA*HO8oRQMH@Ebh_G(Nyb~?x0$q{GL562m|H9yZ5 zc>}?KnSl+-w*Vao$#gW^&%h9T58m#5&p=n}r1~`_QIrG^zXMn9B4H7-8c3jxN*|ql zC?f~toJrG1RK+hVu>cI!muR7Z)ekJJ8Q;y0smi^Q?2zk=^HhD=Sr^eybqK+UPcFfv z1-O9+*$^GW1sAF9o9?VM`bg$s9{Aq8mcs(k7wfj`MYyYc>{d zS(Xu1p(q%9eH^DR!QXB3E2>?}Y(B`YEIltbR-7812pu~_uRb(3!G2ZP`=jZ86)0&O z?5J@3z!IZ@yn){Yt#rVYT^&nJxyS6DfDY_fo&6zv(_ED^o}Ky8Nq_&mJIAer_jXgc82k<1=*V z`g&8{f0c;lt)0BLkxtC%=G|b*yK;}Ak?qT*+cG$$8#M_eKAx zGz#^(gk$>>e?z4H*D8(M*pdGKrAjjy6fPF?4&OeVbTQfcd&>fwby>~hT8F~x_B}Fd zx0%QFyYZ&Z1)H-^k8N|oKmT|AqIUm3^o#zpH=Qhsz;VJv!1_JE=?zzE*^R)zAEo}L zAg+KB04b3Dzo8)CVe&{+0{;UA`OlV=B?nyMG_;SKt3y|LVgUdt6iRW4;wj$@L{$8g z1j-UN05}2%;4qu=MFqhOznbSrO&wBv^Fr43K(W9AAq_t}-cSYjaf=#%(3Bb)iHx}0 z;g$RW1pGXNs_*d{2I&NWSrkUyYrezaFk}fhTex;cf)H+mkhA|x#s^{gvMh4FrS{(236$x^OMj zePHMbouvoBa4Mb^c(bk#4R@9DG_XQQ)qQOpMh1v7V^?hK?C$?~z=JRul zRCemCr)NYOI0Mesy$ZZ2U{;}HXVP*I)iUv6HOj#dLnZ6iQk&e@)CsC3`R*vYr8B4= z%mw?VUM5uzqD(tQI@q!{N&G(2C(%20=6c~H{Q@8St&AlPYktTySvQgTSdfEPL+DPs z;EBV_(~ld=SmDM%_+a7;m-h0JWe6yP?`B z4?qQiFSC6I%lgsx#zLyi{LW$oQucwZuMsxgFG2fITtuRr+FD-vAFP8@Tx_<{@0OA9 z`&5zskdZ}i2WzFc*hGbQ`a`eejVTJZk`#&}Pu;jd`@L*MFHY=+EWR>KrU_H9)Op<8 zJ;RN@_4!C9)+HDnxi*31QK}Q|-y^f+v`MBVKw@-lS?^Oo}x}FoEM|x%m zQL^iH)8q6Kv_0})R-E0mWo&8+ZhpdrjPbobs91o*@BC>if}QAjBO-jFm%^2_+I!qE zVOtxHvChVk)Jd}#5A_Y{#New@;4_LG|7nroeprpWoEk+06_fT(?!wq9;qBc3%L}OBN`ahP1GUuIbDCS3 z-hnJ{=P+TlSbitLX1}jk?_>mW!3(j(P){VSw3}@Pwo^UX00ruA^UXJG3`xbG7F2Y8 z7B)0zBS@JJcqqL%*rl9a>a|hwo>}kT&+kSViMaC3JzKWmBvG>Sr|8WE=b|Z6g};23%!UIH6FL}UJj+pzH?U4q-o!j zIQ+REWCtO(5U`P1R3Br>+?qofVv*ZMhHDyQ{Cdb)DgYk#O7cWb85jeY z;Jz-FS7Qcb538n`g$WrAp(z;E60-`qluBWgsScp3Pe2`$oOG9*xN(VnS2r;m@A}y6 zuoqhT=LnpC1+s0FF0_x&k$=uEWAg)tP!cs(htS#J;Dj&s;oQ`SNQr{s!AaFNb98QC zs4mjweCtDEm3s$^#oHsT_GN#doCN@v2m(LH1qEWbv)P!Et}Hy4@-|{vh{BalLN!)J zDR21(^{!=BzmqgI50X_vRvoZ*1)5qWKmDd4URYkMaLBCf<#JIYYulb0Eg<(m_kp98 z%k{7A`$AQoV`boW)QzU|PiJ1=Nx!v>-*!YjR(K;+Yv`Q6t^@fVeMX4U0-m{0g68es z?+zP2@}_iOeXFQ-1Fk8`tvWrSOXKWE{35>zhG126y)YQ zpeu{+XS9v+T%t~gwU0jX<=@DT+$-PcU-&8gj3wC9q!e*2GrMK^M)1v@+O2PpLb>)O zmCh2?wM&)wrq|B3gI1^5?y<;MIWYB=3nd$SNbL3Dx6N8wk}=bDsrpvAy%<0EHS3Iw{1PB)h zkfJBsUk+&kZ4BcN1}131gES*aUJctC_YsJvnTjW(I}qMb_yFU zN9?Uu<|`d{U8RK><}x1WmGi(G3PFP4YC~8%-1DO_InnLK#y~idDEleBf0S<-eNV}# z!_nwj{M5wTx7H6FB?xQKS$ruSqG&n3$5SN=)WMEq^U;hKjW= z9=PblVAG^qRWjv8I$&~#702Y_?G~nCKBMFpL3}I-7+gzu4=P82)%#TiNsxIJ!nzOM7*u{rkK1wxUvp>1E{3MQW6n`>bE;eMY{0fp;mRt(B+0qpfb^*Cfvk z1L4BI2EEZPuk_C?W0j%aElnz~MwMn^%86n&;=1|y0gJkO9@z(K2DJ)|= z%>mZP9%Ldjusf1pg@YK!`Fo_EIdAV(KW1)?KP|&a`3+*WpY!siZ9Sp1n-g#6xZEpx zLQ;JXMC2jeY4qEUD+3 zzj{A`kym=Cg+sjNB4WH#!qji|@p&|~cIrDCZ?Z}udca#|BvP?NH~%;8wTa_$iggxO zw_DRTb;Rh@Ohp;)CA{=vHEWTt5+GUM)jR85w#g5g2ip}gUD~wZ1od%#B>7A)RAVvT zNT)mra>9DgBpoMrUXI#VJ903Xi%TCW5M^a;YPWx5cYPyN$#F~W)KkFD-mws^A|oKp zZQkyA$bP7MGU?ESMA(O%`L+`Rq}6-Xe5kQrfAoGN;TF;5&H0pt^(`sNS?-NsT+A5j z!;I!P`auB|VntEjT8*em8n72ruaDbEdNea(bv=r-mO1V!Al{{$k|$^?fdchr>8NJVBL;W933DqqCT0c?!fn*xFqz6|w1 z(g9zkHd*Yl6&{RQs9zDxCD8K}G0A%&_r4ZE7NZ7Xk2?gtLfM8Fy(O+oWUoVoGX-J_)zv8V@k7`U zxzv-uE;chXQl+DD%rJBlba$$E$_L4Ft;=d>^-!Ko9zh}ZPyeaIo37VHJ<*VW2G z$m-Z$2^l2yYGrY_Wz-{!!`gp59OY)ObzTm8s!Q7Ll>hZ`q}|jPv!R$`7%$Ea|I^U- zSH5)wbO0Uz{{P+V1o^)OgI!vun4og&Tr(3U(L%V zr!4OA=TsPvxXl$|7r)w#L#N}-2!MzWwNl?V5Q#*=TNB0zgy4V!_8Tv|yHgNM)>$cG z5O@$o7ExZH{$RR*&HsmYLgimOzbgiE%5O06-C3__d;O6Tq8i>Io1CyF?>1z^Rr!Yh za%(&j4nQ!hImDXR6cUN#| zU)g>wC=E{)ykN>ecYooFh^e{Zh!o`qbs5}e$O-UGpaFY6DfMM}zxcXG^$9m9#Q338>Wa?}2Ws~^#1lsswDy1Pii8@xPJHC1k zmgz#zi*QUC0!dm~1u6w4`(>vupRLF!Z|>ki-achprDv2Yo9bq291?%X@&e85eF}MV zIPq5hDb=bfQao@!&3f310%rttxm$j(hRP;m5}UwO^hMjlya>}Xm8=9xn_FgQ`t^FT zNxntqq--7p*ZcO#Fukl&+U8xAYCb^rxccknP*G%P!Z1f&Yt$`~DI5n6Tlt+rO8Tdj zbu%(BNJCe-na*LNQ4#dsWwo5wc!DBzG|nR;O{58JfVMM*>65Ni zyNB~CO9C-!5%@SU${~)kH&;+v3b7xfEC!!&ptOb)r(KKGif7XI!ZCN>qCa-3tJ7eT zIWoU)$}j`P^hEyrG4b*3uaymh&%YoxSo})cTgZKUn?ing=br^o;*620>4FpdD0Mhh zbaqW+2B;LRZoz{b?0gB0hQoGQXgLr@-hv+1qBn1VG4053yuJ1@{s`?oKm*PAhQU5L zbmtHQBCe`R#uztl)wH`MUZUO{Lun3WYDIB!I7(EmSoEt>J?tKW9wznvH*aDi@C~GT z*_H!DvQ((GNcdoZoY*$gwlQtqgkx<<4G8$jh9WZMk^OH9mjC)LhLzJU z|07}c^1t{lhDP)iVtMRyQpi6^j~4ul;pmeyQwk~>HRIYz8ILGb%x3!QyMW81=WQy% zMYDVPuS(H4AP>L|F!<+d#|TC5_C>_{KTnE@5a>t$`P#i{jx`*LWmJj7skStf^-WhV z%#gmy7fq)GMzlLQ#4B*Iy|#TZEd4=A!6b}eTU7!AjmkH1^!6Chm&gRwkk6vVmi0lp zK|nM_bZ~W@*;E!=8!y{R9R#ZQg3b`nuf}~SPN>a|&7jNr8WgJS5CMRN;IiD129Qb^ z4qdWm9ls``bcE-Z{#4v}wK9?5%hPD_+J1iq21>MU;m_6SGW(FHpALT|Za8Qb+$g7D zDLzrk}tB(JSs za0dJ%Jd9LUk84S!#<7|B)K;KNLhVnx(zh!?gFu2WfGeQ{gK%>fzBRjZ40oEakPF7N zFdXp|@<^W!i>V0i<-2tKcQ+#Rm~}Kilttn2;&0ySP|Y1q4AQj+Q_@ z$d3~i7|u_YL51cK+CDH6C&$Gm6lcf1lqf0EPb)3SwzE=H#^-;QIVew@wkj(k0s`<# zq+ZTh6N?Iw99Mo7C4p8dhkAnZ;D@x2>i#UD#98>mROP->qb4+_U6afeceFCcd_;CR zu&Lh{6|4|%HI_YY>@y?fXwPzN0V{wq&`vs=9A#`fzx6xY!j@ZE&KtYKSfI)QE`4V` zzg+JX+u|{;9r}m{+U#m?dC@Kg;wuHQnX)i5Duz~ay{q51MvY91JruHB<^t6X8afivY^NJl25Smj6KypR(2B`WY$b@tgzP4iV3RX z4b9VE+g6#;*xGla5+c<1sQBx-lxvUZxUK09Y`d(2gR_6dOu?7w+Hc4Tj^WJ0_5Ez9 zcUoyb*ou*4zu%62yA!j6s#JNu8~x2_gadw!8oZy!fyOz!#>|$m(SdmGArUg;j4M(; zR5ZB)PsJ=KnTt00_^q*x_RC5u+*;hnnzQj3wwz~n?0g!evrCbD7A)-dIiG*joi?Je zf_LX7dw+_K`{U8%ZUmnx-Q%@X6yJUxm4Hl@z^elPI7X`Wq&lyt!DFRcF<0B($&XvN-&F0eZSr)3) z5*Kkplwm?~$WR0jK}5xhiOz~yV7eMRXRDTfL?*L+*`LIrXADPmsA{?&F3Y-1%Exbg zF+kkT4@dhVhJ%e}i0$_u&ASFk1f%5AF7$mRBo8BWJZmgEE2J4(+$62EV$0rIo9R7V ztHSiLfigelsx-Ph$3|UA&)Jc595K)%PBP>`gd5E8UW6Ua2AL z8eVr7QW##8Fnm{}BJ_>d7S zS3!#_Pbm)%U^1~kv@FW@VYVvH2E{<%^E$)<00j}#Or&|$&mxHBwO=1f%kme~jtD9u zd;tIx8_>-Al%RU1l05owo04=A5?EnYTwGcGlmd%w(R3|UMOiukQ0_Ru%yL?r;Te9Y z1XEr?ZCI8W8D?BkemreRkD0b@jS7@K?Ho=!TC7`84!3Q9BShyYx+Q3vZASlpDmlxy zsJnd+|Ck|$Qc_ZyA%>PlWM-%VX;e@el#rB0i2(&sYCuJ#B!&>AyFsKwx}~HfhLny2 z&)#S6{an}CFV4&L2i%{v?(bTkCA{Wb-(>p2<$I!*v2)8F;ac7|zTcBB&DVDc1X~`v zAUDgPoZ(`8iMFSCfRJGKW_NnXZQ8;Fbo{MvvIpHIomdB7r#9BX?)5&-JaN}LK{MuO zgFK26TdlJsYlu=i?hW@KYj;PD zX7aqUkYUa8QsW-18SX=lSv}>0#0FgfC9_ zMx8It4wnlq&QA|EFD?Kwa{z>D%vR!3C*WcQQMWe|b~*=M3s1pMCqiKG@&BzJjleTN z9uUL(w|W420K`e`e>dY-w)SbtfusbODqakA6Tp1QHD3%FA07ikvTvm_z=Pqk3Muul zG1vSAqGBBSLe$`|JrDT__)d)_I`7^Setm7S%w79xGJ=0po~ef$m*;1 z=Bx*7cgyN~7}30+eXMP9y=~QC!wT{8)tO!T<9}F2(y9s z@$mS4H8?k5#}o2?{;iV4*zA36$={q>@UgaEI+_IRXm9CAd_Y$SdIK8Q$fZx`sLLIs zK^4ljJ&}MaB}P%Xji@arc@oCF0=z1NK0fA3!f%(J#OSZ2hQ!>`N(+^Ix*|CV%v}o= z!-Kz7i!;sP%S?BP=gUmM;57=CvoUcJ3hms)x=+0{bbxvtGNN`*;bz6)sLb_XNp*N2UiFagnK(i&#b8Ys2HZWfbs z%@Yk4SQxI2L+b9WGAX)<-%Ym3{JYJfirsrqMI(pOei^UC!FE!j4d)j1bYHQ7sp3=j zu3xTt`QuMu8lsIKw9ft%%1*-iE1sZF-P-4tOvL(4XD*o6W7SA#4i&k;I_eXl+P19e0e(l|Jn{1r&b(ZmeXygmyvkm4LBPF+EG(r;!wIW82vU2-u!V+b zq=4uyxt84uq`2go2h~LKXe)@mJB3iH{l7K`w}M5&|66m={{P(^+z!90e2=!bN>*=2 zxP+%OFJG$1#9x(fmj6RNj8iXGrJCwRMgA@Wx&R+w^|#RnsNIFCkm4$Th$r5LV!9Dw zgo1=Af5`w?O>_!Uj8%%FC(Y!F(@Sp_14%_Qat~n9x**4&C)~5KXa%qu2_S;WhNmlr z@k@AjEhFHKDWEWQ+MMAW!vdtHr8YWGj+%%A2(jfaG`H?g(~R905GmL1ZzUL@O&$M7 zzwmn!`UQb%MABP_0*ht@AV*HTkU!Ln>35}txU=;%jn?c0vYM57%>`EY-6>oIx(Z=dW;Ynv^P_gCy8vbz+nru@n*FRVQU+ zLf~kbV9ml|&E_GYZu2A5JOafUAu~@)MnTwX!{m#q&i;ikVrEz{Wg-Dy#Z=kZYU_h~ zY{uS@-;mJ(!2t}4Dv*~F@<}b+XDQw;=2K|ABOx*w--5i9f(HO}uSqSqlq@Iv5q2)8 zEI3eSr@jPPkZMEl{v`ts)PG}vu_9Y-!yC$PHPQ=3Nf@wBNPdR&%(pq&kHZ7`u@Hv~ zeu{KZu-)&xkMqa8fH~#D5EPJBU(Jr{ytke!o*_%kkdj7%{z^L$zgAjpN&?r{@_0^D zLgJ_1S=1slxn4Qbst(Ijy`PIKt(~yM0$BXJBs|^zM#6Qi(?-sfnIO_U#J7By2*aq5 zH5YK=$|O`tBIUr(hj(hE(>U}X+`eX0Xu-DfZ~~^^#>}%WTupubwo8-g%~xB^9NL{b zT|$b->s8!F4ka1%q_gSpZCR4y`XhgVo!ai+b&3|KLAYzWpzyj&_pFMB7y>op;MPkp zS0&oVLVdGzbbp>tbXct*%+;{r+8^;|krMJRL*Qs*kMWUx**?pifiAUC1=7!Ff0|XK z)V<0GP9A$}JgxY-H{bd#Z)fU(nS)pCJtF&~#W*5!A5g4)3|*}?2<;q$|VkIv^uD_I5S$Lkdh_2FAD3;IrDpnG*XpgCt>_}`cn z1q3d+{TH*da6PchO&XN-J3tE{yApUygwc&#T3npbG+zM5D;Lm9g}(x6-tnc6#@>BP zujYz0RLSUw2iXtYCf>>T5hxfyTsf@E2GeuYfjS1izxO3m<0Depc1OEK6eJbFAkjL< zi+8O#=q5~3;7b7eMJ<{00@qIs1o(db3;%nIk}27fTP_b+%s!Bot_K%7;* zFIWb*Chfe~5z9w}Fs6jebiGpyW09{eY?K4(FyZfi`m->S7p)n)5hp3~I8a*qyR596 z?M$2+PtgG^6Y)BYy~ati`e-S|&0&4sqnhgllW7VP} z4pX8$&f(=!EcwfO$?xNMQvOlpG~6tD-`dr`Q*BAfTg!m=$;$n<&w@GIJ{Dn{XUj|W zZa8RT%rh}NHnU98&d&&7FJmgZ>x25L4SZ@a1AzbrR&(=PFGyhic>^R6e>EURiH+Eo zpXMn|+WlV11qyqIqIQ;ChvF#=_>~32RT89QL>wa*$@P{4X~I}cm?w*~r#!_0f)ER7}w7j>lT;@B|@HyQ6YE%@#iAIcS!A?s#Yee-@j+>Bf!3q?EJNW&a z_55HStmA()ym?ordL@$yINZ8)frALG?=2@8*fW}+%?5WQ{sBT~LUB9mdSMz$j(va!L0 z|Gc=y{|Ee(m++hX6Mh=h)5{GT+`&eZqGOH@zM9@$3Ijqfk>rts?O)S70$VZ}%%EZn zFWrMu!@?)(fr+R{DiOC#c!D_5{1hM1D2Ug<1)`8~)6hzjC5V}mdCG@k16kH99jZ1a z1sagY(-6C*4rS}o#=hyW6KKn!j38lC#U$Ph0;_f|s*|8vQb5~`N&r4Wn@v$4iss+q zQ3Buqo_@FY)l+j;VXaK*K>0vh1Cl9xz);M*{Wh2sD(lYCm!ThUPR85*q#F_eBFUtW zyZU-~CO=6|O%4whV0WWYrp$4m`W=jD=h3{wVqZFV*U?5AIXqATwz|_-h!`(@w|l`N zd|vK63vwcss#HTi+Zd?Q_Pu5%=kayqXTY_yhf=PGI~jBdK*_dsf+d`5AkK*}{=i6i zm3-9nY3g`LxWdJl+ao_b6{VTLMz*U$+3Bvf-(Z=3lF}F+2*M#`8ET21CdW^3KTe^Z6%g4kh zaU?zm#%5_G#c>%?X=;pXl?PdpY4HD4);efgYgip!2;WUJ-xZ%;Xk@v~ z^hxl!@?KH-XVbvW?n?(UR-Q=%IL4=d&LVnMyr0 zQkm{l|89DRy#42knX~*Ff^~>xy>IfT@5xKL08CqjLUJ7%(nAUdj#+ODU+nJ8oF8W? zoJDKf5L^jEWn+p=p+~}}(Gw`|?PbIUn)(x|G18Uxokt%(FDKJ8L&{I zT%+HgZafaLNJR?R-pbcl3}GKPQ5OlKV42?v@jYf&;;84i6o!5_9+@=NpzKRw=nqx% zwN#4Zy%GjaT_x0i{!nxk8~pzf8`f8`p~ij{8$nmG5$0mp8X}ND0`?$_#Mv+)S6Zu@yHSyp$2=niAih73=57jPA3Q%LhV$Hi2>`drfJU3HO0c6KmPtxzrXCF@ zz2*zcm`ii-5?I8>bsMxrszpEL_HljE5`#BJ#WZ4$0O(Z_H*PYDwGL+LR}9-rFY@B2 zC`WN{CrEx8H**y3sp}}R?xjFvsc8CEJ_CYDiHJz54t_1exo`c@Ts?r#+6U9kIMp7m zCV?FIF0}apaTP6%jueH@{$B1{n|{u**>yF3xc0Wqg@^vt0n7b}7fiE!rC ztp{VK&xcbhb&LgsLLd3x+a^MXa+BCvMK)US^u4_!v@-uvncXlwd|PmAx&w1AS`#%G z+qP}nwr$&d!V}xJZQHhO+s;XHGWVOQnwqM4_YY|ITHUpqi+BSm>xovypKu5rhJrFL z_LxOn*o#1TU#W-H_>rUub8aX4u~YD{7who!7x?O2iy%G1X;*VLMgv1NHd5^z%Lk8K zwZ+jskSyIRiH!N@L6fW499LIl!}Zb_%rksMivt5~$hPN$`22~8kv2ER?GKWh>CPM# z^529)0QLA00j=ShWJh%+Sa*2qhUvw6NHbtB2)DnJ>NVsO15!r#kA#)XSdA;^R5{-| z2aScOQO8uMmXF@5j0HoTcdswJX7eh{cFiuWr)Ldre9MXtQQQZs+HPu(P&}Asf(WhR zIVuN^f3xqd4FwAPyki*?5|zloA+XCYs9)*QX+8oVCwE%fVVDU{Xt7se5VoX%LI);@ zMEe{qMY>#qUpUqK9~Qp|=by)b{VEzQ%uth+n82RswWk#2vc?THl5*tAj|1|o;A#i> z6v`B0${eYbAT*aueI>?BcCS)8(T5DU_qSF}fL-qsY@Uzd!JA*&@4-JEF9U%wzaKDJ z2d7CLvW|lQD+T@YwU|G#~ z90N`cxAhCc}&xC1`S*4cyM!(>lopJN8H0s{azfeh2Y#63x2LDm$v62i3RVN z{1`fK**r9yv+{$UFP8(()^zy|>UKju*Lsynq?V$TM!9`J?zfnhP~eVuHuNbDPbI`f zLKaDN&j~)ie+z{#IrtOC9<-g)yQ{hi#~@wlfWM|XhLx@Z)u;!gE7ygX=zfv~YChV{ z*)fD0;}Wcy8be}|Dme!e7HtQrCnVr#+_<=lDHoRP7xRxc{(*E;>gZG9*X9(o1xk5V zGa%8nim_%nx-${+zu?5lB7hJhfxqVzF%|?uWDQ~!+bS;lo}uO%2hIGD6Aa4J(^U4X z+sp~!ELIIg2D)%iWeUf~f!@HIj$=pN;Xe3Sh)YA@_XZ{R0jk+D6P{7n|Ws@NA^c*c5W} z6n|feDCZ5Xl++DrMuVa3UdANr?IL-uQflS~#XKYLqBPJ+>Vu%AOooMNMdw(ZA?%rX zkV1`mmOq0Q#g|mfu~~F9lLmtKzXH-7UaDBrRmjh zN-~gP-qpyw{L_M^#^g+hm|%^Yl0Xf*Jg2n1202`?cApZ;HRu}}d#3h7Qo$=$bqz8u z5ZQGnmr;nhA@Jv+z2E#1zDUI~?ieya6_A4;JqaJogakzXQ3@lYb#Ja)85-`XEvOt~ z6@n%PPfe|0Kssx6cbT{L9HB);i9pHvKBraVjy5^jQbxuT6a@Ny`)DJ1bB;ofR607p zs;5i@RnUQfO^LnyQS#h+R0FgA^*L_&3(`^ypU6Xp|KTGD=3k9Q@F$j%NFO@YG$g0W;Vl_%Bk@tj}@_|EMgB8qKodmbZ z5R$2elY#a_6p<-zNma3S@h##j~^%X(* z$64;UYwl|{Srh_t-)aKGg>cCYK z43IEh5JlpIidlKk1gjf2BSS@`!)4+6Nk^b6bRS!+IRJQftWK5?>$q7TAjX%-cVHAs zANpo9;q*tf5jnT}#u1Xo?w=m+ouX9~e`-Nam9(`}A88|uLH9=WY%z9q)SNvwcYOfM z;F&w0f&_vBaRlT|5L+ll%s#E2WIdWjk4>LYBpLS@98kM2f_jCgd*$E)goxSeYcdWsiR$})n@dTa z+0R?~|G9hYzr9rHB3(WejRXa)6q$s{2IV*giI;!sN77D2HupZSNz3Si0=48*4%dPG zKl)gimH)r`611#6ef(4kfSW=Z2! zl@=+|p5_3@!UFl`th*DoOz)f7;%lQ)v}QxQy5fK~nsbSf*Yp1zCjoubeOd0W2F z+R_yPvyT(;mx2H9h$>oE(&rKyxIeArTk_MMZ=sh?Rbh#wq@t0Rm6}tGTc)0or<n^3Oepxy(Z|rYC02%I)05#;m0)Ttir_}6_5`-g= z*q`7$xsY3GA8pptDClDG8`*YkK;Vn$3vicriNs~QW{u~4HX*klGdwpaH}9uw`W=AC z$hU_Z#F6wfU z%kH5=$-I>CZ_eWUoD|n!qkMV+A8ad^uX>q+KR-VynxX_{ojrdx%b7g~byYu=r^rvH z@L6stPd|6p&@baG^d|_T`iI{{RXx+#*n?wWHoN|$&nO+`mPjuB$I{Z0Pw{%Ir)KGr zjqvxYDuMiEp!6{LsOe9O=|gjOH*LS++!Kxo)m4v!jjDpY4Z*Ep>G~NZ?|ce)A1*)3 z@0Z|rdNp2F@|f8rz#_{xF|%tj4YXO;Otae}N23*ZTgaSs+N`B>Dm(K)N8{AXJ8%B^ zY^z6y!dm?_eUp}y1$9@Xv(K*wO6<Q7dsM2Fsl4$CF=b26eQsn6vq3O?H|6(1g zd5+AT!M-f1zZvzeH~m*8atmj(ua^oOZsX`@#_(`rlV+-ZdBQ?(1umAudG?LN(lQ;y zB0@fuzHM0S+!4^bONxKPPP7M#b3pfT9&qQ1mb$ zJcFK-T;uR0Qiejdo-NPtO(Gtu2E*8hmbl-MGd~%w$3|4o=Ws#B#jKbv0w%fIpplG8 zl*CfqRNa!5p@~jB!1C|D|DTWBgN@OgW^85(Q597o&N8)|v=u3{d!N>9Y7gENd@uE* zyLrKz=yOxZ)v9TQ5loWmas_r~{?e#WYMu`0FeoNEC)n>5;UCU-`P7ItXk^IY;6=fI z|74tkAl3HMxatF#j(arKXjIJEx3bmj)0CJSux+3PT%nk&R|g;rQHEM-nkMMLa>=7} z%DqI&T3x!Y*!>zVq6u+{HMrg?KMM`$liEKYFZ`a~Hsfb~p5MYL^vOF77kqMS(Og}! zjj{@oVN1YWgY-q0LH%vGiChiF?b6jXM3>E&=^HCZ24m?eq>My(N)+KH1LcswUWG&uN)(yU0CsGQm4AQ2AvUS&eg5^5we}f;{rI zO@KA%ZT>*4@<93}+>pP(o~c>Jdg;Q!_569wb1A%l9T99rXr_7UuVs0DnI6natSmDE z->a`k)QJmJmi;?CZHGXPeh>Z=k~rkaiT~^>Bms6r$1wqrD=f>ng{xB5=dT|UI19; zCI}K1YWCsY0@tNs)_jbGg8UE5TNe!e=1@G>$odBz*RPDYSr=K-Mj=ImCC&Ur69OgC z5xXs02L&ZgdVzK$SgcuLTHyVDG|rE{?4{kyRGjVZ{MPNg*5&D&En_0Z*+~54k*yA5 zBKy9lMK)|%o6xclP#mvEo=&?A3{b1m}N1FWh#F}*E$RTfBF8sm~$Tvaoz?;&}nCqO91I)PL1?U1n zd;=YgkQip~)c^6U8|}Nk=mU*k`9h5#8l*pjg7@wQi+9D79RiOF3Y8^jngbhfy!I;4 zEMki2K{Ns}C$~r+YVX|zrY6_mFjPGHNS?Vlw&|3N?hV|iW1h!+T)ysQx>Ge6C6jq2 za!M~%H-EYjeTt~e5A+`IT`U4YL$1WV<;6^QaqpFD;%ATq;&P3d-g~0FdX0Y-2M4|; zKsU!82Or);h|p`F|0$y-v*rL5Ef8QiH=jIsJa;n~&h@5r1rv98__;V3D3b>hu8-Mq zH~R;uDR}@zzK1a`3(K52BqIWCq{kY*v|{4nQY986Lf}%x?@FKz0FvxyaqdG40LdBR z=f{d!GoEWM8i%U33+Wf+KpgaUhC+O&95Ts3!!zO`u$-m<2Yusugz5?z9J;$_mwIVYR zVmcu`uhBKxAZ3EewO2DY)bS54{8uxHy)1+1>cU)fLx*cDBdBu|8^v^=Zevv4H+ul} zFG7cc?dPlu6fajy2-5*?0VPsw;NUUCbs-_IljeE)fs3!UaF_T9GsJI*DI|X;rXd>P z4G<1xhXsA>DC{;E^edxR3+JaGJlwXv?2HhMntPE0{KgRDjYq@3fk3@>1W^eQw>ccG zc8YKa3ajFXXDl0l++S|LpSx@5VXWMn@tDycA4YAp^75xUYhzVkwHxJWyR7ljZC( z2?If*-+ZHlNArEY8f-@{<8_Ad4G(|h`KgudlVX+Iu@Eyb5l5IC?q+`GW>UmHe4T-H?L^wld z$%bOc&QMNK?&!&X)%QB^Pz(^Er`_@lobu;ICb2Rv!OMIMZ_M!yD@4QGx&*sGA$C(_ zTa%<3#B)ua^$nc2GdQAWx%i; zeFmXjjU{$!3=ugcxl(R&$t0_NxI){MimPDD1SY^+q72xlM9%#KvGbRdkY=t-gPPfN z)6b>*T5^u5Zr(|BG?UI8{LZ@V@hTL>hG7DD`Ht02&p#?zG?$E>Vu%Lo_#>)wY%tl? z@v>)mtt)=gkPKfF@3}3A1DUWVOczAEVt3HRFVKHQ9YoeC{aRmRX(UX6-7zI|e<+$v zxJS-BOm?#wRvg8L(9CI=XA(Fs`I?H?VF&KA+en2HCCfi&=gZHCo2a_N!)@vdyLOm~ zOibKmp}x7Rr}lwRdy*H{58LymIy!o+54k3%=!WqmYi!+BpktmTMrb6pR3O39y3m7^ z0w!}st2Vas!^>R+PUZhSKgo+(4Umi|CE1JY7Tte;ad48Tn7dPZs7dECDB++E6`3ID zp%%q(tK-K|!rg+o>t4@eM}){elGz3+TTf4RMsK9fcU+YpEp$%DRbkQzG%l!RuO_+o zcrK28S?xN) z-;J^Q$AE?Ll-%0>Wfy*IP{i!p0@>BC4!_H}JKbGQs!4I(42sR5a;GNzjn%jc1Xh7i z|7d_v!mN)<1~lXWG2W(~C;(sDu~MPK<&6`yk=)6gsG{j`4$BuA5eEBf*NS+}Iajt) z750e7t+M_T`AO$)P;f#PH-VPv3b!l^Z|?l88f%nW5HwkYp0tUa0+g96RJOj>m`Ip& z@U~>bPOyx|pHMvr|b02dlrpn!TSg$;iMK;*^{=wd4@54Z! zbN58`mb!0_u1MjmF{c2VLw{T$V~0qPW`Ife(iSlWo#S$NfJiCshQkNZ3<%qUic7Oj zW#GD}l7nE%fjnKd^~sA12oPJFM@Y`TE4n~!beS#K*jK=|9iS_#Bwc1TjHE0^fug#c1QPc< zitXZF!y_9vA%O6h*^~+<=$u8m+B6A4z#Je=8TBTAaFPmUwY%ebJkb;p{FBbCoO5h? zWbKqs272^*DS;whWCY~Kvp|htmh;#d2`er75rWQg9zmKjlpg@dU`Xd0K%GCdZYhE- zHi@&n9w#%_-k|yIavrOW(N+v?FK#u71*9A~vAV@WO|Pfy&It#Fq=@T=ozN;iC0@8= zX|UmzjIs73F58*%g#n~bI_165$on3MC*~W5oP&_#HYqaO(0J=&26X0lI;gIz`C(v| zWvbENZah`Qpoc_Zqv7SGo*kPoasn7PC&&x>yAE8$_q==vkwkMrM*olM4xqYaJ4d6-0AIJNk9OGoOiqO&%q&ArJJN^6;lPFlHJqPw;k z9>p-bEfX?(iFT4(!7IXgns^bKuRl#xtjD~M$8V`JPEEaObZr1OVU%AboS44M^pu_^OkSE@thFZ8P2N+FE;>-$)vafpc`8t_ zzbT4Z=|@vWgU)5t_hvW#16AR8Q8+C2d@CzMQu51K(kV14wH?w;#%a3JSTsHuF6#no znK>7l?$=W{%OkFjnp@~CaKxF^w_4LGqw(Q!)Z=5~mUG}72MMQpGFQ>aH}(~+nvWNp z+R`3Z^X#f08YRe7&ixdupYXvo{1xSP{JN9Ou0+wI)r|(?wp5abY3VUJj67EA?!Wbv z)aI=VO4xMbHe%4WXjv=XCgL~-Lf^3_5HK`Ar;Odsp*v{LI@(2>QQe}aEg#hO4mx8b z^~l4|PV`T0V1cX+^Oj$qWAfoS)ns;h+ebHv&>F-y32n9%Q_SQ#h<|d4d3z=!HNSZB zuvM}I-EH56F;jNO35&4l(oM)veVUZ(m(SsEN=s+Hdf1LaSN4TvawGEgaNQZcbf)=rqc zm+lnaG~u7~0?54rUZ9g7u~M@|qhpuFKf!M~i=-Wai^RO}7}lZPm=|fhANmpNFgqD` zeSrnEIfA$L%%h0q=ZB;%&wRcS4B?BWZ>3<~1QH;PG-2XnQ|N@9cB-{%sj@8i_M>YQ ze_Gg$@eNwZtW+Kqdj*%=9r!%0!?m_O7F$yQNIjk@vm;ldaVA#$sktGA2s11(&_{ii z7UrqHF$RV^eiv3H1r5$^r}L%qQ3kPjRCJRLxOXjO9JXyFpWsk@VLmn5Y*rKoxOF?p(M4@m}>jqoqxrS)=gl4U%1hmFzZE&wys+FvHYU?LgF)G|JjXk4}XbBw`nc zM}%iaZEylQCy8j91idI@Uo#zoT+E1TGNaA(M-8!#a#G4}>c5@!Y&%x$hMsONeBK0e zNnSHno~s>q!(xfKMM>;a!a`eDvb(M8fRaD%>FI19i}wQg_a?j)oU9GZ$-3NpqC}r` zQVYYUg)l>p9p8heg+Kv*VMT^OA$BCQb!`9k&S{cNG>P=nB3_xm44jX4W47L0RX^G& zC@SzQ}=sT5(U zadhu|P4YOLVjn~u0oOq8*z|#u5J)jTU;gc|L~MK0ySNHt{0vZPmRBXjt6Z&@+6Csh z2NT!Z;=&resy-_e=RA~#J-&axwV!5_lPwkV{mc=gM=D2)b!42v7F>?I=qJ}Yc5GgW zkgBZNEr+kd&v@5%evn|$hIhPm)!=3%mWs)vD~==(@_d}$J($PgyS%2atkDcoG+Hn2 z9Jeb`^tirUWhNb7*gilSxtJCADh~YhylYDLxoD_lol!b*U_eX8l7m;`UY#pFhUZ$~ z@KYOq(bvMIKNM5;n|f8>&nXBz;Ve$cf_z4UV5X13@7z>I!I1xcIP;5UQk`NLP>P$0nBQH-wDPzzW0F)F*VgrefKi&K}oucksF>`$Qn z4aAhNWI_l|M_NQy%e}^M8dNA%ieX8lv8z>}|6p+bSWR2;pE7iCip$Lkl zdJ`xLd&z$rys+)a1EdhTdJ5n;qBqI<3ul&<1u|&2h&$-SxGwa{RG4g7J(n|cts=_N z6W4MU>DutJzm4>c|}R-he5!H5A}Qb?#Ij;%R*w81+hxA1>@|f{Lk# zx#1I&y}v!z(|vyVWeX8hzy#lJPX^gQW-NTiW6_h4M)u7^@*{>Bs&ak*P_zB-cKXrE z$QRX(mL=63)Pk11)H9Azezda^5(vku$&ewe!SjriEWYkCeg?F6#+b>PWd0dE&-dp^quBbX9EF;@ZPbzmFEY>r?2 z5lsuTS(avlwflD)qIeTtFL#`f9N#vM8kC%R6SWr%b#JIG7`(GqcJJk+wQ4qL&b(Dz zR2_^FAMLaA8th1Yd^}sR!L^{XnG_0a zqHQK~*e!%j_3GAVk8xSv;poRl=O~$RAa6iSqfXiL`db@@N4J=KHbSz=#zMm4S~yjB zE3RSYhJcM2io;KT^)fZmqbaB!N$cG38_xrQ^?CL?4r5D))B*TjJ#x$NT}>#ml=KM{ zUwn>LkGT!L!&2o30i`FiwqxFMYs7lGhp;X2KYVOCAaHygxznQfz5`+V=LLWAcfTHs zP5G)Bo~C4SH`b}oWpOeR3|8+)<*7`A!pivc*$_CN?2Cn;-9e)A0}Nm?f*`vdVTC6{ zKwMvk7($#wB@uM&^IMkP_NzxOl9K!Ddm>Y*Z>Z6u4-T& zS#VJ`Hu!n7t**}2#!-u=wB}`&UiX5NsCPTN#>9RZ%xsP5klL4{tunKAJ7sJet3F^d zJA}s3p|Y!imp&e^Ma$`hi5G$2w-dp<18_=3_C*o}h@Qkh>X6qzhh5O1d6->xVxiTqI=aJcgxMnRI5^h` zVQ1V-YMR`PQOE!IcuC2yFA-K$=3F8jh#HMLV~5Ok5nr?xrwXuo1qQf^+gu1}Q4!sj+7K#UdHNTN1h#s?x`P(> z9#N!%PrO(`qE&m7@X7`_cHApAF#|s|trzh8(c6oE9Q3{tU5qm04NY?^3X-YUw!hSE zS{1_E{R--K_kPlU=+VmX&Xq_isgQKuAS7p>JfQ9#7Vqc$Z6};i1@(PEzoPiO!*G7X zPY|z4F@T^&`hlFPc&#+FVc!MuB+a=5(nrhqqGQ*=rUyXoFQ)9UId!1`OJ@ z{<^oLD=y!bLSw=$4<|$`DkmjxcVv*ad&5_x(gNIgwi3JeImq7BP)Z_5DfyHOal9 z5r<<@+K z*e6A=*kuI&9nx|_L*DX2>YjrD!KLr3F~{EA(hkLid{rtWtu~iKp-)&AQH_Lpgg3{L zTS(M5rR>pd0|J(qhoU4yU&7cUE`tD?ys{!{4#Mnn-qC|62kNauZJAK(74r=R|!!fE)`4-BkS zSpRA~5b5>AU~L~o0ICQ>!kZN+sQsQ-c7c-p(Y3>VlGYaM(hP%*&}-HUCh~Y zUbgB@T{#iMwp@`?{b}8bblQkXBdhpZWa1&nN>YI**6@9%boqJ7qBwq^ll?Tu_w{6R z8gnc3WoTXtaWQv2%4QzXkjZzl9ey+?$O z>oae2D=>^{jBUhUI!8w3O{t~Jr;)gEas03yVNH3ZJGzy;Q*9Lf-~zbD*$>Q$FFbnk z=6f&GSQRkyAYeicpg2tGWs1ERpIC>OcYY*7V?Ncl3}E*SY@eTNE{DB1@N_xn>6+fn zaMG z{Es*_QrF!bC*lsg^s*z(s>*4*G}$*c_9pLmp4L0(wrqB(ns@amARQ0$V68rGfp%q1 zf9N#P)y95PG3$5FUESzEjWb+(yFt&|bRmRRbctbnI`wZ(ZuHEOx|1Z299N|5Et3g* zC}-^?E$+Yg^}~B_!yOTJIERqb^xhTTfR28Fy z8}KIRZs>6AoR&7>*T6ub8c@-tFaY5H@xcq=cwfAc$E@e}@wBa@jqh^4kXC#U6OCvZ zIZ2xN?)f+h8aYu*SlaBp3**5HOCP%Wok^?jczhn|cHyX4N+}v~Nl1#o&o`i&-MVX; z?Nq(`7MmOA6)a~3O3ld3pgN&MFULqFFT(k>I)iB*sD|Hcns#)wO47PLbbDA<1Oh6T zV7mhjzyajwPIaU>ixf?)+D`@H<*}-oGU#0#J#`bAQ=!@lHm(2xHSWSnw#R?~fNwQr z9gjKw?<1O=;8Tg}9-RWHs$q<64^X>d>6hk4%Qdy8gmY7Hgm zbo8P+R$4(=T=)3(y_s4yhA;p@I{Qh}$ZP*PvTT16JC8&j1(R+*+ zm^)t6_sJN}U8;pLG>DuqNn#u)+a2Zfhqq(+Tdm8kHC`C=R&{XPDB|`8(SVoibn@-S z?J!b(YfzbL&bN7c2QW2>v=*!sB=S$HwIVv*8sK46f$jPU&)}1F4wuRnWvQnxYH`s!K`L>%*+L$=e5;QY$v4_HFpKCB$ z(!ni=We=NB+g9S_LfBfc+O_y%)|i^(4=yA%9_l-g`Mp?G==?g#vLi3JFb9;TEs1679BJ~l|Rhk#AC5Rq4rSLO9Gq-2(I|Pky#~I%ZN=> z8=@w_Z{SwMPDt1)honGgLhbbz7MmMphOiZ$BKw7wymrDU-|<@IJ##t@HW;3trwYVT zZN^g>_w$p6Lrd`47UN`?j-rbeOY(fHdrIG?bWF;Vt9uhnOxBJz7iuE5x$@oOEoe9toIi-d_@nac1DJw` zaFyd)_&hg!qLgcyrBN_Rv=~TzO|1`+0-5sOLd<<;{1C_M?=HDtn)j~Zt-p7bHKi`pBtq9o zy3BL`9^sMb?ty!Sy$xel=O!Y2(XbnYYsR+0U3~tU6z~w1{>Y9#jGnl5Do`d|I;u)C zRw{xJ1`nqs7Wj?5tT|tsiwjDxKVtt?oRjSf`r#eQj`P>*tIeql$+s#^T^1rz+X#da z&@t_)Ihf2dIJ_Jdzr#*06LRl~qeZhen5vbZPp=(KiJ z>L|lM%_`PdRkvh1<%P$6u`R>Zc1*@uwQ>{`*RN`4%QYRL!2NOw#yNNHdL#VU*Upyy zP5VqSjH!<$cIR8)8%hAL3k-3v!s9g9W$7Gr4x8od z+`O&xCL_Ejw|2SiqKO$h3v)aaL9Sc(e{RU3|4E!6aC4YSgC-dIC16YY+?MlRC0KSj zrY`z2%=q@ED2vI7!}ADaR2T{U(D3blD1S410bt_?jV5pt`l1kX^@uYj3AU&$*1ALJ z4r-*X5HUXHYkSt77TotF7dJf%q(|P@U`p;C+`!c#F|_M=c&YKmn9uy~gNr&u*+k0! z#~XD35$B*2iO5$FAS_QX&}xYPhbCB=m8BGCW$9(;q+=ClX`|=m>82DW8Rw%JrRAoi zSN~%UDoa~lC$CqCbC5}kcYN@7Qj61W7hDlHj8cko%+i1tbBxpT)3USF0Jfej;>T%U zlKC$1T%4n@r<*QixYs-&Km+CPC9ne^XL>FViQ!2~@)Pg(=5?mAG*)n`J_o_8@a>He zDA5XTt3wU|9IGbpxWa_gd#)MdB8}0sM?WsZO(ce#9)y%rm3lgs9Bcg@c2WuSHbsFc=0xf)=8f|#o&}QV$vOKL_v?6hp)T*ZjB80ff zTL^!BZli)ORNB%e6#nQ%6*#cAANI2Svr=Ax@oiJe)<~<17CJ)k9|b-x1GIH_Le-vsY%+V4`{stWX~l${KBaQf+W1k6#RK+4u51+b3}2$L z)T9cw<*Q+OHerp3)7p=6EPxk~95n|=NtZI@z)WSx>UYp@e=s;J;PsaKpkPk28}n1@2gA z!bY}QHA&6CbC$b_(hX6uc#G#u>@M;;&{>G#B2y5W^t^&OH|_PXtSoc0)3@wMZ0USy z^oYA6^=JMDI9TAnG9UDcaTS>yE5nxKyXuU5yiD!XxPsY&$u7-UX9fYvQiirE``?Fp z|9e0}1tn6Fnx&<+P_q2{R5xzA4_79+zMd_|boEu1{R#~9LrsWyhZlP)0m$*G{L=Lb zB6GBG5cXG&6mc-|^)|l~i=2vNjiP{M@z3T9)DIq3k{r~M93=8+0`2CLY(Y+d7E8Wo zGH1q-&zq==qi_UD+^3r*85l{_#xqRINX}&0%(2gJBap&VzzT#i!m+hpw19SVZOh zEDV@217=Q}To*T^9EZtzC^<{LL@zxnEy1w1Qn*9f;r@(c3J(B5B9f)^Kr#UVa;0GQl0~O^n?V3n-HLff<3>lkI`=l2@RRlD8mnl7m1Knk%4g zfRl&7a1kx60dDRt=7uij7F6aA9?feoCBQO5*n-Ev%77f(L;rF1L62fpi06~nnWAh) zCP!m1Aen1OsY`Xi*^yk^UrtoZzsdZAcJq_y8!AlQU&M zCVmj^l&XoCE>gr+?OAm|RMRl)f+u2OJ02WKyf$v1dlt4X)%FVpD27JpH1%%Bg`!xn z z_M@-1;?g@N8tO-4n9zml1Yxd+B?Kra6Jle;LRo&qPV1;R(#w}67MoP=xl1j}GjE1s z$V*h&&$cs>QM?N@xmzNc0N2{+Q#0R>D!HzN5celFIrf{Me*2TpV zjtkXI*?2hTzV?@Xwl_sBb-=4~rAnwqwLz$N_bpJ(#i9TLDG@$sRi6QErLwa$H|+IR zd4p|_>3oLd^x+0&kFA`W9O;)GseF@O+T+S||C`!Gi+AMWiU#L17IpNS#`B4*vlie- zF^Sn_c_iknly9Gx;MlR%t`gK%-LKO{G!=9D>h|a-S@qiVkh1voE$JROSpPpnpz|_t z5xE)3*1`2D>fmBqQGK3Taqpn`Jy_bVne7cm2AZ6OwLXInW%v)cs(pk1{QuDd;$J|~ zO3FymHwlrxwT>~5voLYjf|<00I0gLdq(%(I5{yuSmR=4F8wq>893x{NP(_*6;F`vu zlM6%f%n2AzVQ8Uo?pO2>cP*`_RvB&;^ob9@=7|S4f zTVG)X>qoI|rvI;V2F3!;9)(C*2@aWuBVJR<^jNRh@)VX#Q*vu37#TSO)BN1g=6hz4 z#%Yt3T7VFk0Gne}SC(lNbi4m(Ry0b6k`^kuIUt1d_rQA!s?n(_Fku z`cc6W3$7I34Y|t+Q#QtVK^NNQBr@gn`$m)Jo*SAeR>7(e;HyfPXc{(*o1Z zclQ^I)8rX64ch$B1=Y?|_F!5I=6Fj^#JQ3#Ubb#&r`NH7>&J{N{hG!Oh%B(`usrWk zRbp`<-Y*?`jT_&}-5|_#5&av*zKGx&-%`azQ5g1F#q%iJA2Yd8Tb{sd}c7?~tpA07YxXSFNVR?r=(|+Uw}>nV8oHMYiX; zFL$^#P>n1e5rP7StPL9b?>i#PFS8ES(bwK{cP*8tn-h!YV7qv2F|vQO=pOA%;>zv7 zN9Xv+XFvHjHtNmc_H8YE{a)tde4$|ux^H`{@(M_t?ht7787AME!?W9-){A7bj%SL# zE7wC7y?EvYE$~jNdwg(h0dY?^%0}7|0a!4CoJcS)0>`oKhiHE&Nh6D?h>chIm~2$~ z;UOui1C%&M2FM?YNqO_jHyTxI-&8D$08=Rb$NfLU1ddGgcAXR<#r9ww&W=tea+1Lh z_9T&(uZwZk23q-9SwYuvpmy2aQ3)E@$faNEq1yzBZK<<(70}jS&G=Trl45F`Vl8^_ z8U2dtaV10Nf8m(%h8!QgZgBO+?tf4-kyLzIl|nmyItwLJT0tJjiP8NMN3w&Dy9cXb z^4+@Rn~t@^b;F=NyQ;ou-L>HCpcyW8Q1T>YEDJ@QJJyoEmCoy?x}a}$b(1JsU?oy- z!Ia3|nWRKw&I@*7k_(^a<^&vz@ruRua%|6}1&e|^9IO1=DfCz(r(Qr4MD z;q6RiOK-3_1AI_(XNO8_5)4Q*3EuqH#eNL2PZcyH0R&m`~m*Cbu`FgQO*RVsuh;GA8@-VO3@kJu!A5K4&D$$|Rzm1L7_A)-Ncm7eU zJqL>ye8=XqPtsZG3h_s%3iXD6QXOBiwr#WNv^A}&ZZF6Le(uD*c%XU0v7x_<>8y#79Z2niZYZ%R`vt~n0 zHJ9!C#gj!fFyvoh_&uw`cTLXL$=;!`jOrZH)3LRZ`BkxTpA~w#7}iNvx|;EmD#9vO zC_nn@aWb-6l09)NMQiAdFcPO{nx^R~G`7TIV!zxD<^03=xDf>>Q*@eKB?i>nnzscp z{P$kD+tn@0Ki8(!pN)RO-8%KCjcmI+Nc(5~%p(3HwNvmQAb%Bt)y=b+@w*ls%70W> zP}0-fSU+k$Qx1?)4$3N2JStB;>`7epePNXao>-2|A;FGDY33*A1?N|vQYHP=Ngvbu z!q?1epTz28X;~%8YO9(U)y`2FOB+ri?B5Yz+NNb-2h5s(N?gkOXWZ(XeRqf)g|^W_ zl0m^4jH{L~8tymgFooH<@_5TY3U+K9tJ5zURNk{xz;;lw&B~U7_AH~U|2&4zEYqx_ zEYLzYlW47ZXSZo1rg{$RIDAawiS_6VEd~&mKOu810G^v0sbSmx+gVKX#IfQ&UW731 zW*R1RG5s$C-=vSz#2@KQvvcBXz6Q6mw|SxG&a0VH^jA9pi|?DE;zA z6|3iVc2Gthy2vpI8O4f_R$|dXnkOZN1u5F(ogq^gF^l402Y@CJ4;6>J4qDvYg9urS z6}^o%1}Y!S<0*U1Oy@gt^}a5=;gWPvg;Igv0AYL;VD z2NM&dc{sL|yYC?UTlmxE4&?-bC z2veMUoqH4Wk}F2((%lhPNvx6k{=ac|9PA`s!7mNi*Ga>qk~gp+Em6lkW{4oC1s<*DVGY%qMIG~$?!3Oo0YJa!p(IlIno+W_M zGcZ^hFGHeeH8_tagC&dcZPhF`D6&(P<_H2e;R%a2`26Wh@4F{~EjEH$C^I-HREhG^K#qsA^G6ssip9;i#+2am=gD}iY6it|oKi$8?|fFm zFTADQ!nkICNu*JV~OM9o% zik3IPRhzTVjbhW)4cFiy8)8_s)Mz$vN2NW0nF-S6lyz7#8Z15eWxH)Cvi5$his*l1 z0wo3%lo?yv^ia9s0HhF3`WoLU;~F-@nP>ki(yO#bq6juaJ65KC8;zZ`Mrn*SGP?-XTOxU}h}ZQHggZQFLGZQHhO+qP}nnU$`z zv#Vzyiig@FB=ZJ4c>%DEwu^0PBr;0|Tx5hy~Soku0vhQ^gXB>)N$XU zTmhAv$B5C&<;3^*&#^Z!TR=rxXYC3qVFbf@f{&38QgFaiA6`mHsjl?%NK>yya>j>2 z*sJaJy3wXmI6b#qe0M>+F9GtXQ9aPx(TBntWk=0-g^gcIH_ofv3n|-yQtk)FNYFkl zrQ@4=7hPz*x)*(9TclH+6fcVo)TJ3y)gQf~yC5J#w(L1$Jft4%-$&9)EMT$XNE>mN z0rxA_UidBdbmD~?UTIE*{JvZqf)s0uRJjjQZr)N9daukL!(rZmzG?A{20Z3cI+=tg zpOLXH>9_boAQEr{V)`*aI+Zl^i?a=}%TuM>5yy@~YT_(g zM}wa8I;SJe)=u`KU&A*_MwWf#4#(ngrE!UB79}dlJ;%&lgKdK^opoavmEN&NBVUun z-}C_2W0m&+W0n5+g42y(9s5ZiMu2cL(>PF@X#ng+pbb_yZdn&jpL-4aen%&n3E1oa zl=z^(PGQjgP|e=muzLWunt_6WjfI4QiOTs12B0X(wGp_<{P;>hK%%b$rESj)MwF5b zl_M$}TKn|=hvodERSTNeSip3aolnoSh`~iVE42g@qnv6AWfr$ai5XL*6$@)%>ZGhm z&eoJk0|&o?KO#Uxv{23okmH}9Jm?y6A2ewhx0*uCOkNL6^N-q_hHrywklP{9XL$9e z*gd%s^nC@uvxhtXe>%_l)Bn0?D3d1OKtfW?k^E@gYwFF47(AYiUXq4FoR-R@jb!YV zOx-DCH@Ph!i+z*JHoIu)2LNX*$?^Z&6D_kb78NG>7(OETM3MuR`zh3So77-{O(z_M zeZc51n?EiI_mR_LASL`phK#Otr)ugf9~eO06hfd9)eVrD5da9Z`(Y{}YJmi7w4JT` zBSBUjN19gUZ9T`c%yz(+S_l}}QAw#TVd|mlqd*y~M7VsL4w@H_ccr1wd*b^F?j~{A z2S_D7J&oeUjkEI&w3CU&swG!_^@IFjah2l>Q}0mOb}EAs@CU`%^8E0s3O!q}z`L4H z#(S$u$_LI3ajcy78cC)#jwO(boP~&ijex8LV_=fvs8jw6`Z$>?H8I*Uw|hnR{4X$J ziGOaIZjiX0Jp6GHPQ5Hk$Hi{FJ~}~@8;lnLba-4O9h7t!MoyY+R&nc9EmEuBO~+_9 zH{mFBWHDcsQkIsIcJA-}G_OCbx;9LWHup*O=7xsvWpdW{0sc`;QcuF1-tWRhEcB@q z?<$v0d=!C1_Jz9EPK#5JN79U8CIR|RKD|W`^S!9+;wrnD{|n*SGW3H#|6%9pTsl~*9+mEHg3is3TnDp8i|>(*;%P* zxfyz@3F&cqpkM0kj0SfRP^aq65phtV<<#`XO(xg?01Mj0n5Nlcf33$(Ws|=e0syQO zP;S=WS4T_=SPU{C{C&p@Be?kB8gVQMMKS4QjsZ&o&i)!*gV%Qu9$_UsypJshF!%~& zq8V28vstwmI>dT$S{RWA%!EYYYhp?(9GryP(LsSLDSS^HNK6q+v^~xc8FpACC$VYo z4Xn@ey}f_#l@-%q>Bn@XKxN-Yt;n59_ZnT6Ri4;Q|i=<0l|s zaaXe%Pp=HZ#D4;?-|4buGyl1;gmIe#euR)4pNN!`Fr=;~HcHH4 z6ULCbH4H)a;%#VFx591Mv{HPs!uIK%+P4H0jak5TNjf+IG-{&l9Y2hnVRzw5CZW%^(fI;R zEB;Umjj;?MUshE)7MRqAm=K7I443YgQgPfs+1?~XPnZ5C=3)tDv%E0ez5nn(74cy)R(4$@crtZy37U-TAFMo$m%#AIVXWFF{Mrh8 z_GW9xmik!h07sQSXSBa@N#nROL(#ZNU#+sfkBUwc*ud%j1d_9_cO_~)D+YGQFCMO^ zaIb4213RZLfnxU^VnOi!bfI1}Y~#zBV`cp`x_CFu?2uxPU~0}16I-xm^{&5QXg}X4 z*-2AV&l%ZL@kV*#*rdRZJLmHH#Eg3obln17^W7E2coBCt-Xi)_peOByK0}h5!hOywWwWZTq}93m zsU-^QsuD<7%|?^jmm8ZlQ;yi;LsV;NHx!Jll(z2vO^&$ zjVWmO@V-Zr%IJQ}Ne@5;q{-PnpP##lmY``Lg*bi{wT(#kczGYaupj6)9qw;oals8R zAb@)24uJcaeTMTX1_@46uFP->$6;WW?mq+(NM@)-u@HA_hP?m)hd}HlCWJ&u;5jeI ziYy6oA`yw;N8exi2d7LFs|gl5FTmT!nJMM1Y^(^z?M10bg2n;@4>s#M?w=advfoMR z-gI?ZM{=EwVwQR$-+LVl_}waM{dz&90oXj_UY0aoh3gf4R>_nr&P|d0_EJ65a3suW z38go%MdcDh&qn?eZH$0T5W^oC;0T6i$n#5N5HGaoxTb`$Pb(u4Ae4W6k}0)t)$yY5 z`N;7ZpVF^mLuz)Ipx}DQ;IdpQyX2NrA}<#@uspRx!ZwZp-CDM2I)g|a5y+^#xH3YS zP%s1){XsEJj42@IUIbJ7d#tHXi&M3&dsS-opa156L< zl-%XjdCO<+Y184k{?@lEO|4`NCldj-&nXC2vSjgQ@4R5Z(;o#;%e#MN3C7oU($p`( zZ{~=0in3!F{pI!;kauo?w}0IInKFW2k5<1w9?>l;Irg6}i#&4mXl%m|g$RaSGq}g@NY4+)pAn33;EMllp4Re5HhsxTy{U!f&gJOUJtfuY zP|@}CAi2qY0_NtaC>ycXQL?XU} zc$qWte+ZQb%1p{=(dK#tz5!H5&ia#F!7+?~R2xCWQWKp&L{8cz zS}kZ>>lme7eXA%&_jBjF5V)xi;!@G8Ubbm3&n;ySV6@S{B_UEvc-ue>J80c=x{d-` zF`UC>MrlEeEDUjlvW>EifB0SfS@v}8%)9%`OYiKJxn=d;qdIjl1buy&M`anN-TYOghu?=fb zYu=#`7%C-NNm&Ge&pC=9x96C#-{ysg#;LATheLQ+1c@vECgt@~^yBbn`iesphUbpW zEG2Q!nNm(|HmLoqXhHGi#T$y5woEf_O)c&-96g&xsG=&3Gr5XZy&eFD%Ud$a-+zuv zc8CEZ$TRogCZ34P-bzdmf~tu{T>RnpVXFbF&z`etXXpWnb`*iNfEX9^?o?9*LP)Qrn+M%K-6=}{B9B0WGxbA~2 zr0!-?HMe=loz%zo3y5cYP9B>{tQegkaQy(L;y3>%J26km@M*@7m0RC1Db8(zyAa*Y z7W!#&E3ID>&euN6X|oWH!?U!?QSXmRt8~^Oar~W)vS@^!eNq~Mu&@l)3ToSevYtU7 zlz~E=$##ui(}=pzC8faJdWK>?02IgOM5%xHA~|yM@5_Nio?jz?*gDL{xfThSYCgkC z5~xP?IeCf_0l#;iegoTF2RpKNGX};Nag6vyFk09yT6ss5U0%vdRwt@qjDy+_k2D7o z)(SF|g=rgPRJY7@nTMvxs&4G@xH~aI;(U5r-YSX>LU-8|_AQRqzW~pqqB^=uZ@$4y zNI5%fV8SNCC`JUTy= zA*=l>9qxc>7#L~WDF)#iX}EHA!#LWvu?X=E96+bc8zpBUByQ7J?HL|Hx1F&I^1&iyAmDGegR56k*ZAmi$C!av6=iRK{p=}DGpI& zCYBxXf10F5)NdfXL&lyTGHj9c_1JaxKHgd34pKXR_@_cru0odq%m|8cI1O)fM65}jzR9b8kCV+wXT4sn8!31T?R&;Us(Fe8gPWa5|7m>%DF zUJPNPOa*p&2um1&f7CIR7+K(ua$QN~)({frTLWqzCwu@3fmJA27hbL;Z-4B8aK#C> zwcM`1!B_(jtdLS)WR#JH$AZ%Gvot{Oxv3DC?rgPw>dEiIvNwmZFJx*u5jE7kq&8R9X4=UdO;xK=$DEmhPbb6BvZ}{N8RY)$hC2-e^%hyj zS~@qZUst~coE{S0+CfR{nq8gSxuT1E2EA<6@P+A&oe}DbolPB2K;|VkvBJ47p^Cc= z*K@&tZ_LAU04+e$zZ)MeRZ~M=7BbW^JN9L0%zSz?g$Ek*V@2X+tIVRCdzoGA#n}k& zQ@VfjIWFa9SoW^5i$m@XiUUa9 zw;n+XCyFphBp$21?WZC!)QZG~UVC7Jn>cNj4s{Z-Ig&T!;wamAO5v7_UbfT>3+s70 zr;v+@iJYG(d%!*2W`gXT$2O#XS)J5m`O=forc+?F!N>rEbi!3#^rNeQX*ZEGB*B(< zN~}M1Vd1y&KATx_5!ddz4zu|xn)Ji z_5Klggiu$$;b;QkBhKhUaR^36hPm+e_BB(3jJ4#erX<$`N;k%YKOfg^*}e7bevZdO zwACc3M%2Y)NfncnRwz{Ir-l36)EMF4)l*%iwJcOnuH5o$12fq*DV8bag^uo6t*-dVirvV zxN>-0sHZELIy+Nj0J)#s4>ZVAuCqPUi5-e#`Rh%jY&ZB(f}g)oiJcdrNTYaInTbhN ztQhnmTKeS;TnuMl0DXRP^S?ZjmRDG9bhdSHK4&ezcBn$+Rm71B%NymTIZScjAV~g% zL8cwak_V%VK@Lj_X)ZViC5~i`h-G&!z&Y65%D{{@5&==!!hj%nVKZ-nMl43G2_dcTTo53GZzrPcwbZuG6-uXDS&|30Yfn za8EQF)cQeYI}*Mgm^Yi;eAfdagNT|YDb5(%OB}MD)p#K}PNmE>t3=8d#|3qkH!hcK zO1pSe0M)F#VlqooMq4-QtLLdLtO3h*T1=*`nRQk-)=J}cb9>6ljkip?;KiPJ;1|uO z0-qXyv*4#_)V`dJ+w;BHN@xW#pk_u2eyTXd?9O|clKcxRq%Q^)-_E=&RgsP0D(#I+ zEkxHK;I1K(sz+=q$IhVVAaxJGXC=Hm0XMtKABSw~jX){MwEr^CvEXm!eMHn`OiOEo zPiNLhmMl=f59j5HPS+gI2k_sU34cBsGHs@Jwng47X=KN?N&pEXCM=1^RZSp9JhF#m z_#;H=Q;B$35dg*G70aVpERdmsgP4cc@G&6?0Ev`^Nzw>_B#NnoGASq%2_Bk36rGZ< z0BJwc=Jy$T1M1Gi%JuYP(@Lmvi&cHV3Zr6<#lfn8%E^byyUj5ja+~`n+PCrrq((GL z=xptnjmUle=KGq2@m{Z;?Cs?X!(#jO#x+LjIg5GETl9U~Jt3Ya;q6uU%+ zEdST9Qf&?0BNtYEoX$qfoSZ?e7X4*2_AMUtj*_g0PUH}JIC&OAY$X^A1+fuHBE)W zdT$#1nRA-Lsm}ySS@)*|REYia=3VkdJ9uoC%2JMiD3U*;^LU3k13AhbKP-~39wW|Z z8V0v+;4MI&Tkusy>yzmn84D#ly0SBzdX`oW2YpEVOvbhkiyY}EV+9TIQ^U7tW?B_I z@$~4O4zDr~-}79ou0e2XZ>@HJEZufxB9~kQ-{awcAjq|5cCwF0SZ+nD>c*(l)Q8Q4 z4&<2hul3TVRHDN2eiWwc5@d2-ovz=uu{BqQr`4+gb}7!o2lr-bd_C+&@&uV}k{n3D z;1MMija>(sRsqp~2Ki`cG+twVd=Q@ZegUw@?wuub{mmSS#0_zT;F~v8Oh&O1Y}%w0 z2+skevf+h9K>E)stiZ^<=$Aa|c>4qeOm7 zPTce96U;W#8l8i6oyF8xc6*>XPooGI*ng*lbJ{L^_cu^=M!@Lipa#v; z_!~C`P18tzT*W%Slw0NrxWbC>z4xy@tMyyVux*Ml=z>F&^TqVErerwx$dxvx^D#bM z>26cqCq+ayE;A!V7vL7N||FH83zzV|SeC$qO2r zGF(8v<`^rR!xCoq&{6#NWQLv$=(ChPnw?_eO&XtkI* ztL$X%(1jL9iWQ@lq6D*<*$+3V%NpVzpaca~CN6_ntK5PSDg_xH*`_ScZc!b!pwaP|*WMkS`$$l{o|=&td4cJG z8rOaULHX0xCei&#VTDbbKTPa(O8vBLJ!Rc4E`izyVrZalWoXf>tnKm>SN%HX_v)&Q z-ezH;W6}JScp;`|6rl8vqGBurE7%!jmS{xK&F zMA9-SjS9$unxC*x42#YI;S=OU{TL_06YWM=lO^C6;i$WQD!eAA{wC3x`*!9{wq#6_V75IHKi!rjU}VjKYG!eXi6XW8VYE^H2rRT zN1AX~!N&yjZ#aAG!(5)wX8_^6;7m-gLd-Em%|fsno^7T!(sML*kM^h@uSP5$V`eK&pYqAZzA9 zHn~>iu+`Z%Wznp;2idKK!#+Vx<-JAG==E5}(eKS~&nWR6i{zr4()+&Y9E0UM&9{_vm4x_~+{q-ne2b_TL)6(I6o$kY z@AQjkOQM<2$u3fc*PER}^e!d!uIOUtZ84_xcY&O~M{6dfY&G0Sj-lr#8ZG=;5Ydvz zn-<3$*8A(kmRySAJHo6_ge-n5Tl z4$0PzRl0EY(||vEaEs}+t!Yc|o2#Mr(#d$3z-HZn8t?vjc{azj-@DjB5s=291Ls1B ze+aOL=o z$X1KsR4U|Zi_|>4q5V+A57KVqs2Cx#x+IM`bRdNhjbW@JkP#q48UOfq^Zkuk56IGj zxEf&z1(I+n4ut_h8sH+xlNS&-Knhfp5EyUWX7O$~+VMCYSW*w(*=(h@Mb1>j)pNp|e-bX8F|T)vB|5!mn8kRa2e#{8GMROzcIG zuMsgW$HR(u3Km|DKnxaV&4TiZ7S&~b=&B_y`Vr@{bw!_noi>YU36)}?HcTcc%&v2bZ2{*g>$%Zw<_bekQq zxdR1WL#Gto9&-$1t3Q1V+LqtGtXPclcP-J%+{&;%2TMDnK!96k-Y4b=Ht)>?Kka{0 z^NAjTS=z8o;jLI;Z+@+NBxWQ;%cj`FGoCgqIKic`mvMW8=tfio>&7z>3;QTw+QXSjIESrPkYhd**eR{Yd1c06KoJ? z2%@|BO&VyV(?Ky=^kJ4!%xo_*cXZ{B(8X=?`LROtqSVwZwcjzjS_9i@v!*>!@WWyh zSq4d%9>(LhtiwP5B~1A>=Jbv-90vsJ)Y( zeqd&vL6&_OT|80zcH)+84JcjaO~-}d!=zNsH%`jCr-wbSRm{gd;CnA?gRL5rU0o`M zTUPSyq~tII0@$8+zbl5_VX)FdR*A$ufml$Z{ROmirUbThVnXEECbkY_hdnb4-#0ij z2a-R=n+6uggCxPKU~D3uC^0J5kH-o%MQ9R|I~~;}i<#DA zkkmdw+bjc%#vK-sZe`sY9x`XeOsVDU!bh@cJu0!#Q36HjLa1olyD4(oGJS)?3Vsf9 z9rs;;5$w6ju&Uf#C7p=-4awgBR30Zw_hRT~ugl78o8DW$;rzs4mr`KB+I59q-(XFj znr;)APA@mOJm)TC9Sq)lG=1+xbDMP<_d`op$36pz}3v~H|Ejq6MJ=d1MX>vKuguL*ZBnF_E3>Tibr${1rfX!sVulj zL`xH~g++X6Td`iy#?#Z7KNWh&RbMgEiT8i*U4U8K0hoR0n_*=;|JuXm%L2y1;G$ho zR#{YWuC&Ho>2q1|wwAbSJlYvv!Ba`rXsOm@R;& zJQ24h`w#C7qR-2Qt2xIpc;2nKFPyQ!`)C$!mz`x5xszrh^gn$yyIHw$S)vZYByZY7 zT=Jw->;n&cXmbaemx7|LQjD@FVNq8zr;j6QMcP`a2DMSRY&pFh_7)-0Y7260qozN0 zW9@|o&2ky$$dXjX@pr#g$m3(}m(xQHfq(V0&IUz!(oU(_2b+Kn!Nbi$Z zzJJNg!Im>f9#DDm?1tWb(VuuHn5UXb`Ke*D<$NsIYhHL4HdVO4#Qc<52-K#IJdi@^Y?h6IgyhVE0aZbX)fgk7h;FMm=*(WIhxGQY(^A;a9CQFMQ)bd;l zz2n!yl@&Q7!>y&9i6^iT-WT`{?L`&X2ROMS$X!35zedzvU4;{*Yak{0k)=jzeVxmS z({G?CSO#IDiRo5bhx~S4Ud}oSG(oB! z(?b%ZJoEKNN(0uR@0;zNI)kM=yURWE&djiw*(v4?dh%mj>&J6yANbu*fcma@&~zO9rQtF1Be zN-Kkv=g12{wQ?1MRy_#DQbHMjw~E9G*bEY&2tQ{X=n)|51B+HvxMVANz3j%I!lb}f zc)TCrSkjX# zDs0Tp6O87OE2rR8hRqkY2|?m1QZ!6JAk~7c4d6})HD;S z+)7nmp()JDqN50xLib?L5hQn;rZ!R__e`o+hxY>20+xsGsi!xj(7#PnR?HKN_pan< zMjwIe9Mtr$IstdAh)+YM_<_6)6`&0*qwLLs7ma?&cGw$DM@_2e?flq(#vsqRF%;{( zVr?XM*F2w`fmi#q36@!u^rMO3xjY!#K;~nAu-U~+AC}7XL z!=3y@fxqDuQJXKUJpIvoWA#xl$I?`{ddkGkOwfp@4KzW`GP+XCa0X-TAZ~RJ)|llO zbT=6|O!u=D=xfHJMV;UVe2m1Kz~+R`1{vH}CGeZnOFMgRx(>Rhix*@UHU3!LcVe>C#*VpEA_tJ{e;X#`ExyTe?GBTNDC3a0RyjdRo;9OX4O4Xiq?D=soLT)QoqtcM`c%(^l||%DR(=;R~g# za^m6sVCJL*6cFQI6_t&Lrbw2g0uDRdP*P4fSG^1;YwwNT1Kj!;4%~G?#07)}DLU&9 za}`&&+JpHaM3xGcq$w3r#JXc_o)HBd4?VOOs$9@Kf~jAV64Z51CO{=prLwi@xc%t; zi!1i6{z*b_N9+~4)&6AFiSEK{Iizb?JWL|D53o$=RL$~kHQT3^D)F}9y@dCRUvMi0 z_21Gc*&zl5;b#t!8O=GY?mY4xU z1|m{190@L*&IZqcg>%91VNb4JgzxGbs)+Ohz->dB{M4&#Q22 zmUR;paVlfnW&tiNObW1kntazP(o(osmrbg!YO!I+omNBt(^=-?zvu$)K+>ZE1Gi$PF?~TMKktxACC>*TCP) z*X8y=O<8X4X+QMpKf^*#S zs|s3lW}z_fuG&?6v|=NOeXT~PW?z7RpWAKQqnm5dc`>LyjoImN zN=LV0l(HIf7fX1hyh6W#iSDH%{0&gRX5jDq2k$V=PrwW}t5IS?;bmhv#(Hu;@q;n; ztX_rKx#`8flMWRDhA204GyqRC{Fxkl2x*E-;BDS?T#b!>$m#LXwlEAcQL(P@L%ND zhti9uSr=Z5tOV>_Ag2%GZiY)gPs?3jYgLD6s#=rp>RYWB-jvZFL3wV(C;ugP6r)WL z!@pUSZB-i9mU>)~`H~oiFicMv3Jlc|6wmj%a0h>uUbnZKL)!~N!fB*C>JkI8v+8MJQ3fHX>WzBue88Iq`9-(UU#_9Nty!V`pw zw^+!HN=Zfx-f`?fVIBR2vReMnj}+_!qb@YJZ_adIew%6bx-~y1Z3(9HJoc+E^HDX! zPKw-XlD|#aaGk)H(;= z&woWsvxUD}SttKn+b8pn(DuFM=ArpHvwuQ`Gj=?oFn;)O+@%DC9iS7af6uXK|y*D5{r4) zWOmb!k`;Pajat#)MLs!?Sf_SJU%Xr78wUADC`;+Aqovf94^ zrOb-JCc^z^AzdH?%%8F&LZ@IFA*@$cG2T{UUo%&DRYthhTOxMikh?FU%uI(~FOY+f zu+tZxBNM&jMP`a{iBn>jPpOQxYC#c;kO}Qa&R><4V4wt##?zk!P;$xMIJe$;|Ex7pE0C-T5Ug{1#(_wUu)PzznmAw zOaI$(re$n^31L9@KC=cot4e-=fyEV6=&11oy{|9bg}J%%3A}ff{qkS+vTNrFcIon) z+wA()M4+xfNT(FbB>JZ3`6|vse)h|ZxVq4>EhVGhwPqivRC{E@W4xFb|Kmq)R^HP5 z-hOYL7w~GT(ae9)vE8ANA3)f#$G{;OJ;czD%5!qi8#*YsjtN5)0uw|q@8Ux3siYIv z>9&IF?A*KvB?$*fUqeEeEIG_fz+HG*5`m;VfplP8c#E=l0?~v*)FaCeAYVQi$6GWm zsOScVbe2fjy*aKt>?OO1sMsjNedC$9Z`7T}V(-}zTDLbx_rveGgF530{a5EMfTZ{y zhpe*j?LaurD7X2lG%~LLw)Bj4z^WnNLVkDe>U2}?k>tY;j9r-QJ>v?O@Q-}w z;!s&1b_y7I{CpC%Ax&W=fcBFE0U|8?p@2WCmx>h&INPC3jUt%SF!^wg;OW^=(~(I; zPxoc6-xnMmfRKY9-r2nYEf~4jC1XWDTNoW3eC{&U_ylKr1tNiwihF>+f@8IYP3{Ee zUo!$QN>BriyEDOd$Fwp!B!{Bd;Ni5-oC>(=kBtJpRH5Qnfl$M z#&4mmi|)|TbSvFTnY((}}NxN=PKxlL2eNVVKq?QFrwlEK3{i~K!O zrnK7r9dW~nZwbAULa69J^Gi~H;z9xVW8rl82cH4|hJ_F^P>4i}A<5GlEoR4@ zD9u!e?7uRgl_NqzsDS;j;~u*qk%V+*V;VRYv&Q9A0ZgPumS37xWw&6Fh^%R&jDiU( zOJSX}fIDC0d+VRLRSZz(u&f^-Q*ES_ZQEO;5i5Vn7nj~hp*Z1Sps#Xd!@T*fp zJRGUXyj$%>p1ysEL9$Yw3Py8uqNiY88aeKtA?}AA_`4<5v*Z?tfZRMoRS;rEdVff^ zJdai~`?DOwf0B>;mnEt{hU0;-S-biJjs4iRIB!faNju8W`anDTeLhtskV}pZ|2YMt zy)0%ZzhVKpbPfF%0G>lcNCs~!)N`{+#$J6ho3^glFYs3L{8#UpO$ShHkJ$X1>R@D- zu#-7s)#Ps*p{aqq2I{Gcnr7qdR%bT7bPH+Pe?2+szdhN2FAoq31l@ z^UU`$PsZ4r zxW1A;0z;@e{eL#64*YFY9KrMe4jCO1BtiPqlv``GsGpQFH*!OWYkxSJ*@LuF@CG;@ z8Y7D|mdd*TW65Kl3@*?vOjmoce4XP$6M!rE2vGPGGSfcmJ#=Ft{Q#(*m;g()X|=HMKDp(uC7$Bvzz9~w2A3n z{9dtwbEH`_T<>2#FpO>PW9N3ci=w?%%R|)gG3r>3dTTfhG3zjOV zA$}DzuT?o=OdK%fvsEG-n!n!Ulj11TQ=Px)ngo{R?h-680~a&v#d*f$_H-)&>!&D-91Z!=b$r9Ln>dk=D6KQsZ{3rVAh| zJcVF>^7qG0X^I>(U}Gw9I`CSU3n9^0b^WO0*f&aQp$g8UfYpUPTbJP&)$f=~FIbh> zo5%AE-gHsU+V8=K*3R+%jb3vfVa!qLjVqv0%JoamEwSw?jEg%ggyrNo#nZBb7z=-?YEW_M4|`=W&kD2suEzCtE> zf-#O>l133D5OFkBqmE8f5m;dbJ5>fe8b^wem_T4`FNy(7To}vbhLYPvaemdU=JrR( zL2A3{?cTGB@wrQqPmU+vJwp3m5xb)cLn2!=EB{BxyCAIR;2%MW>O^|}g%zpwH!Of} z$B1sfhHaK{G~tI$B`Z9wNNLJVs>C!O)L~D}*Txm>FD=ZrWiA08I^7-=^jh~YQ z*oy}2jUF&45y+SpHZ3ckdY-5fy{)`kQfC!j`>i#PcyjMsw6FY*(s3Bq6M;{s@+Uig zeBQ9St7=&9PNSDE2<0XI^?fl4f9L*%znvqx`6nbBTu31ZUuP&)=Qn;tUlB~k?er9M z3;3Gs!7@u%R_G@;l?`21p+t{e1|c+(9_2B5h{o@YcNLv)5TvBt@#7H}#GbJmO@O+K zkC%dT)UPBn*2Nck+o0N9C>RI4yKwJ+EUWnRJ8K&G_;9xz1F^YV)X zeb&sij)A?f81KPq%S6_d+r6*Tb51wymmQh-$G&o{f6pmo#|T&qF~9)a_yqIL6|g&4 z32AIN>Jdua*{sPZ;+SrpvOPd{PyN)__h1I*lb^wW;c$e&sC183l}1#O5I%}mq#5_N zZ)_GyFOWx!{(%^Ig2R5KKAhwXQ%+%6Ob8zn>$8dFRV8*v?PW6xymL+dD%Yilo`YZ8 zK#PtsaO9ZA;N~VE=v1n6vUBCzoR~Sjy#d$4E6}nc-9fo@Z}Ne;&#}pNTQO&SL(sJe zS7mYT1DJKJQkC&9EyrR1rDcr)HULM@N(w>%n@ZK{1%d$55&I{y0#L4dl!*s2;DysI z&bqjaknl}FxLKCEEF}~tO!3Ki|pbNJhzf=E)1?tsj zJ8K)gXCWfq#aGDW1;OW#b(XnDu6uSZ!;5}`s4YuY)1oG>OO8kCw~a`#xWDw~(46r_ z<{MDYzGz~`|5Y>+_aKO$Pbi(+*0dX{3lCptv3s*HFn=-tA=|~jQ6}FAyEte|WPrp+ zKJqKhwZ;`?Dzt(Q7UPHc1Ch9bHdV^F%NlG93Cs{9b84R%?A2Af>Yw0>|95Z$|AKpW z?g@aMM~xySXU`X75z8)JdFb7Y3xi^?R)UvlZAqtxnG)m7`2P}P8njs+!w1R8*@hAf zj1soBB}MmNG`mOByYr|-br8iCoKpwKD~P7p!GJpwVJFKjq$UJKUkcUK;Z%MZJ1GGI14Re1Pe-Z z(X13YEq{L7tqyBvGTJBs>bCJ#(!(7U$xknFYliU-;}Tyclzvkm~>*y>7vI@ zBdweEZ`t_&fku*8rT;-AoYlW*l>4q^55n4*Hm}u{Ap+1HJzj__$b?sp-6U!|l}1xI zS_8io8!x;hpwR+FLa0%wgP%2rNx=wu1BN|{5{-(pZl6$Aw)~v4-`*^Z-1XGbcjcJi zbHG^5La(;7%7wA&+V2j7M~Bs{TB>uZdEDDPp=e>Ft(5`@b^2Nc8WXI(axb8uH%BZh zpZ)I%n$#a9KS7KGb(+WbJy+Zs6Wm9t?VILryNl7X=&gI&FB9Lq%hU8`z%2d#*WJb^(jbUOcp zKO;|71YS!145)nXEPeq#t;qHWosFy#lm4&n{|{K#ph(&t!r*z208QC2(-J!Sw)P50 z<(M+^PzpwJlpvjpQ+JFq@1xKF@{DIna7AIKP-4Y5?{Kj*oWd)tR1Ig@*Iz!7Z$K&( z(i=xcDzz}WnXT^k-p-W1U3>atCcux$Z5D~{-~aM3YZM)mlAWDktQlPARQH8%YrJAp zA7Y|Y+KEMvSf&_C^8bYwWafCVf1g=PGbL(XM*czd1H`S5Ncb;R5&kMRSx1`S##CVp zrr5MZYJ9G1+X!KlpZ48-Oo_7|BJ47 z4$o`rx`*%Bwr$(CZCj1a#*J;;cG9@9ZM$*9CQX_z={e{5J$S$S-+N!zo_p=J#vEhJ zx%R5o$k=Iq1hdO+ybBs`n(NfP?cNgNu@{a4l4#2lQ!TA=wz=7^e(qZc-d$;x@&$4m z5qJ9782m5KnqQ0|9jxqv^D>52xb6rwyA6f*Cedu#(k?atr>>luue?a3ftrygn|t{O zYp5jWN*@BIRI+&kx0||_V*DhP)*~=#;O$oZx15owZW!9CzDE>p_<>*&}qYCZDDH zWXUL&#bO~F<71wn?D+)SNs1}F##w6nvU)$)nO0kVJ_>l8G$dfMR$cadPDL~Q$(%XS=X1_JsV6H0zb%LZO=tNQTGF7c>}=5Esm$i-De zHuZnp3rd%gzjD#H8TPeY!hy4Y>s6LM~CdC*=Qk*u4BFcj(N^94(i0<0znKA+$@9*Qcl&?GN^sGHoopc(u z-Y31_*Fw2b9O8H#6U}3vTdlndsHJ)YoDbjUCA_hOUEJTr@{j0d|H_n#V}GKP{gTh# zGDCR(!j9iJY0!DlN)4QrP`{CH1RMV=ew5jo9G(a(Ve2QOCMeW$IPLky%G<`&TYE0k5VgCw|7v zPM#BO86CGIN%+z3P=MI%EV-9Ar1OQ;nNTHgTqt=*By0?kdYE74P&S(`^w$``mFWGP;J?j6&u}RF)^fW7<>2PchMr2$ZB=tWS zT0fk>TUR$R5vYBI=I|L}yiCfb020}C`fyZ<8j>w8F*+(iv1kjrA}4IfOfjYml%jvw zkPx1(nGN@fDVzzZjT7+r>aHcyreR=sQnU;yY&`1YU&hqU-^NtJ=l3+j%dqTdaD|=C zfzK&PekTakqwj?moB zlb|~&r3W9&J%Ha)+EQCBq$;c7C$f{O>9UFW!z{cm_dyO>++8|KM)W&7tAY>PDkkJc zbvZSY>dXf$l{;4$cByYv z+-B0~xcoys{32ga-5HUE%RBv2k;bs(CLNOe-$?|>Ek6yj(88cz=RJwY%=dmI5$CjT zr1NfSc{2E16=4nI^l`)mq^Zzo53)&AaI*Pn(e8RUd2*72Fy+(i;7CvqAli;GB(YTR z>7t7c_O%_~OfGm7uB%=_mLH9Nhx}@~e&LFBZ4ZE%g}B2<$)eS=iRqa55YXRsj!Ljv zI^>UFG*iW?)BXw!`rAif-Nphl4dT(z#ZAZE#``r^ma>5-Z#d7v`rzh<+Gqg>(alU| z5{VK_D1~xSe$eqM4^c5qLyQsJWn)V_tZt@qxR6R;a1)7kq0H)5ElhObnIrH&yha+{ zR(-3`diU->x(-Xr0Uxz--!w!hN^Kh+OLSHNwfrJeX@w}?j_SpNkbv%nZRx)QH1z3&VxL04+mptI8wLSit1bD!?=Gc=??5bF-ufM!9ia z(?v}IfZ#+CM*T(>7L$pHK(v+jl<>ssH7M3Cy?t=O64rCA>89IiL;Xw7nuou1vFgA~ zFkI=eneF)s{dHo6a=Z~Y@+a`ZrTE&PWf1j8q< zK{CiMzZ`dRwnXzli~6{yNzh%$6Gw69G54WxzH#TLp?0m6OI_@QTu9asisMpxNvuXJ zd%{{OAzSn#!g+~Fe71+nxwmBPv*AH+^PMWr^^m2R9A~pX!;3;#0>d*7jC_Z;`5Nb| zn{6FD9ug6{Dbsfe_HRFohr;O!B$-9N#c!1g^rKCv}Zn|g+ni&r}Q zzT5z;H0=pOnX#PFM{dwQL#Wl^JnM|vj5LKaV)p|#suuNYTo*~|D3C+?7O3#Fy5dmU zL4EI|?q+sQu2XNCYu#94lFk;BfAb!`wQS^vhX-1ahQff0x-XekEdKtwSzoEDLP-uC z+0GCJD~_G`9|$7G4{(|zl_v|{nV`{Kr$W6QT49*#RG%jmNGV{RH~m1Y`u`s+I&kgF7w8{sn3|Nu8v^5Yow!6{~Lgv zkW1xA{|p}w9h!?$(o1gt$MA`($V>?mDPK_*iXRVfuS?yGozFh!$!>w?Z&j|gA2bJ2 zA{Um<`bo`6xI!gHiZtx|OMR1^%g404Z%1pZt-gL|BVD^U+#>>xndmv&7ON>VD>GaQ zP_undP_wvh={A#Zn1j9+HpLh_Rd9m#p+W3tL4EmL*Ne+VRf12bo3)~_$UgiL{>a%o24L*pvlAb zDTI>*vFt>*q!&6fnZv|ak&ww(b#hoWOo@gkitIPUOoYP3v?Jg*%Ue^20LuY;dlRVf zKk7TC)E3s>JYWsvdQ9NZpEcvI>ro#!1DUBNy=PP76l1;ztc2w^Lor1;aKAlp6pFwI$1m%{&!G5fDDA z?dS_zre4#)Dwo4E)JItV*5Kcbun>f0Nebu@!SA#f)cKlPQ9G{gxzyUgcmKUU_47Wn zh5WP$=6uKsx(5@e{}|`#|BIIS6r4uMewSL zUZA@NAPNeh0+$&G$WOZ+(O>SVZZ@`ePBH@yqbg*PP$;b$GOXjIC5^bHMAZr!GDew4 zqhN`4rPt2;g)*cO8x|BX6I4dy^7>ptm_-!gLBb?Vq7L za}DIW`)WJM&F)ETPjdkkl3wYoB0%;9#1Z|GOJv7dFy{$8zQ9`ya+!LS!?m3UV!1CC zDF(MS6MgHsX(<;qf|g$l&(>%<*Cm!XF#G{SQ?hbVlUPGI&ihKGuCuhvzSW#mXF_(H zNG7Lk^lrc#W09|pip{hRGqA3Yd#I_vb$Btcjq$b!G%~ks*sjgtL%6vKVpz;r_$~7u zT-m3EZbipP6`2c-XF{TDUJC^L%V2<9p2GORifB#Ag4c=Vj@9LEaRrXk}k+*xJKBaSUO?g1K0Uk=Lo z7bg8JDMgsTSV#~tEoKAVgbOO& z)`}-iuM}2#8Kr?k13!^+T2aA4P^^&6G0E{sL$5Ed&X->;?peFD4;Z>j9|geny&LXh z^;$-)(j*um8BQsZV$M%`LpyJF0is#?i@_k$!(+Ki8b83kQ%TRH{?Gac;zmU_ak-!> znNY-8pZi;aU@$yYM_8VTk1%Q{0yUNzaw>|)P!7SZg7QS=ut>NnCU;qd0Z(a$p#oIm zoWBI4gl3hnhW8y%!r+K?b;&GFcWGqYSKhsYrp0%NSkJgCWLllQ9Ttu7*_9mt4bACA7v6LGLy-9E z+H;ltAJ3z0hSlJ#WuqAm5|KG?8Dl?Msgz#j!}AE!(!5&dc1qRq1>zeJ=SlcLrl~)E zA?VCj=?aWLuJ>yqVqk_!RkmE9B<{fq-WQ5pNLCeID{&PV77x{WjU|+PlF9T=sWy%0iK(4FOEC z*yc$G&kHGmwrCi>V`uGD8keCakBUnmpk4l~IeuQmAZ9^|}(?Y4{!4RhF#2%Q=o z_HZd;r*>(nt z`QvjOtk>Xzh+7`8`mQ)az-wa4*#h236&k->2LZ6A7D@6)TdkaoJSY+B#v|3rq&2Gk zt=0R~wB;QD*)kb{HH_=-85n@kjI1b9&gBXVE(JwOQ*z=ag*dWIlrGbolVg$24UGNTA1O2`!z$tEUS{;7d+#A%89-^2iXd^%*xz~1+f|HN>kb?dj?Y>LYnNLk>PF-OyJp4#l$(kpv z*qaWHl$6md$H`5g6#tq21V&K5ID68c0J;7c;NCsP0sw7-A4Ei;Y_*!}XNW?3BMmtE z3IiyXC74zv9Oi@yvqJ@D*D=L66Vx`frr}AS1l5^%%s!#>ux)Y65(Ot@9~Vhc=w+#s zk9ItT|*!hfaj zy8YaY>0L5~#orvkpxh(Qm;FE9=Zz)#AMbO5G*SCGA#LyeT}e##{M@lu1Fbt&j*x^^ zb{U#NM8ZH>EzWx0DQFIzDezKCZVosYpm(r{xN69=ut*P5m{M%$yTixwB~^cy{OY^i zm&xEBpMJrne*dtUXj!L@o9Hwc@vG}Gz2PV}t*YVmhCV<_X(}Oq^s7*L(}+c7;xjNg zkksl4m_Lz6miPG2Fs0TaN;b8=4c1jkw#Au&4MwI(*1-irCb~u5#g&oX72e7Il?l#S z(%B8>$^L%B0z2N!fcETNO88(t!7SYoU6b=uiWud_E^ee3U{$)5GXmuPJcC zVjqHaYGDmrL6}KQ8XNqe4D7kA18D*$6bS`&yDGf2m$y9x@3P+rWNr1d!>_Dl0^T99 zlBbN;L-;k%Akise-5KHvmDKX)=k4knU9*txN`@aw!Yi5oquMEKu_gXi?d6s@sB`}` z-?UahV^V^jGI?6W`zXtAKqV66Z7)(=#7V=|nSY-W^~{&8!7* zx^a7b^Y%`GezdLS?JGQo(bboDf@rlm^yG>)>k7ON5?S|-aOlUM#qF5-usX0Rom2wU z1)*tgOVb%jSS6}5kp82V`}&EBP8m&0AF}xMl4@y9$ryQDJRYRWHyJ;_krCEjGff=` z+X)um$TUXF7jgTLm~nuYbbO0qolC(yF=8$(n{sYoq)=FuQH)>qd?=5|>6F3~Rg}b= zr#88JO#e`H`^>L+zxeA*!1O%v?LA%fI7B;Rr_M32j#o2;ja=#Y2KMhqTTwLW?@=y7 zT}WfTH8xuKX9kwrs@k=!5%|x&_4>bD{u(46U~PKmb?3uXms!LN=joBD^@+&p|Hiv+G)@S83e z#$7uDfAopQ&Jyz4ENIJled-#2+dAzyEP{)ve}uDqZT1WN3Z;@O-01VI?HEyA09{jY zKjp99(vKx7_{cS=D>zCz?;|MsuDs@Ct!GP=mAkt8NZLFjArSP}MF9}RwpOfx;ll^6 z*9Bw}<3Ne-ZjggEm35ScP9tAZL%=a)BA1o$K5+K?eH)Va@v)4sqJh}wdz@ore!xk<@WbrMbly+aTtGM zv;Mn2K|qLQszU%H{85aOv0gR56G>qL;qnnRkN0Jl`%zX=qaH4(L!?ODh+Q3b%ZWm_ zMa4#^?^N$mOrh+xAkrsY-uJ2aSiEeMyE5vbFmCF*7&b>DleDw&&>|@~?|KT2Pf6n* zw80bXwc8cO=e}WVM+A<{eV)4rB}V!u$l53tb{Fi&rIXL!cWL2qxG0l5PH;Lr{Jlw| zzVhXnPS%W_(`~$KqDC2Ho3vh?EFB29DLU$}@u7|qX!#bwf|LJSko8LD)#J#4`F+8E(O z@wjZwh;8@Ej70?u%?n2?5Qqz2tppP;$ zm*A1hipujQGqjGh0!s~d@uG9DSB79}3?QBqe0VkV^^dJ| z>+_kcD!2ZJDL)sRLekYx$88F@@4|tSU4F5(6wN#hWYCvx;BrXOXqAl`#moaR<<>UYB}{wIr&?76(9)D z!afNKujC1-R%XBF%D1(Pu$NkOCW}$U12|=UlkD-P7>2s0wf1}JQ;@LA~ zdH*ZIM{u=jbHG0=$5$-FyxCGmtL3L}sSJY+N$j_=iZ)(5nB66DkpsMK9a zVG}W!TiU*MaTCu$Cw<|vqNAJ^lWCfIj14Tt*ioz}YxpuDXPxR@-)yxE9K7rna@#g2 z@C>|lN>~g_MuzF3zR|leg-*(w*!(uYbHViSKOJl}akhl<89K)7rVEj(|MPw3*pH2j ze_8o6(3v5F%4o&MoyGx!LQsfA8;`@MC{F3$B%RZ^Lv;v@x6b78MMog#`j&I_ zZLdRve)?+V3=I8@bys++H&wiP(aI)Wg})N5loZO^-t0kX4W92?CyeATn!;8uyE2aN zZSs?efAokBbPyvd=*(Nyo?RYUEDlP3(!C$!X=Tq90K#_AGLzTt_H zlB>YPHLsP8pjhMRdw6nJ#!tTJIM&+OWHh_qX6onhg_sudY~j4G%3(q46x%(~pPGctuPhH>a?bM8P|J@aq)k8Rrn-SC5EM=3aW{mZ!Zq~8UdHXe@(3iw#{CmJ8OF<`Tc z¬(Fj15;G4UVe>}VmokiVmx86zSEEq}INRzol{hA2P)1)IOccqMVUE%Gd!wL}3z zl~xhZB1kCY4Ivsv{d6p48o@VQkK>3PHz_3N1|~yt&n|&9fn_UtRc$g%O#hS7gGS54 z#qwmXnEmHktvHL%vQ0?rv7A@wZc6YgzrjeL$LH zz<=!ns#_Rp!G!MKK1J!AFiMbm$5ks5LP2?KI7$~I!wGXjV3cU$V>ZtlD$gGUV>XHG zII>*VfH5t(C{(?L9TP)y_dS({KCziVFYhF3dz)I;G4Jn``io!rMN$T zK>vvp_x{ccV5y_F3rD3tv@_%dR=t+QM3gUr|K_9uqipXKZ-ucuB5v z)~aQ@%-xbqCoN4wZ`ESj>24pD_Rkt3qDddLYw8bI0G;WQpT!uN*?e9JL}Wf&J%F}t zzern|%ZmpNz~z^ZiIYR)r7Tld3$T@j+C+pHuCiQwF3a(_ff1J{;=-2%PXDSp2?wl@ zxFUMW3Z>I8X!L+P+U{RL!q9NSDM8ZJl?uJQQCa0ckL~S9w*5)Lw(gu?%yu$GKt0lUAvnnL};rT;C19C}lU!!~1a7=z@eWj!C?*qLi`3sEo1aY!vUa|6bEBnUd zng|}b?Uot_rgY9OfSDR&1^pK2?o=i$@LwM;(`z?Nje9{W`$U;q{`G#7IGcQnqj+A;lQNa93hRx#Jr^HrW z&+mR{w?G3lVs@#22lh{0AqvRK@6YxTIG5OgC(Qm0g8~kRTlD0LWt@1>#0Hu{oJsB> z7E>7h)gnM67lv)KJlAvYUUYN|JqVUf)?fl4eV;L%Qw87J0iKXahkOz^_V-3+r@h2C zE&d0S_|#?1m%#N-V?B9KgIZ;sQkUzz&MZ_)GH^Yct2+a_Q+4k_v**n)1Vk>0gw+3N zr3r*t0E9deGDX+$4ekRauf!7F{bu(9;b=@VEosd%BcwGM5DuGSq#&08(!J}2ROgW< zBhDb*TDVy%ZXN}T5=_S{prH3(HG$50UCoRahZ~B^D-?JnzgnNQOo@HfJh*pF=-y+m zRZWGtIwNfcDeej~Jj|<7-fObC;Cs`E>{!&d!t%rU0mKp^5%fQi@%kZ1%KmBcS(ZK@ zM)dlO(-n}ZCCfaI)3{&=nNR744h5hc4lcKhCD%0LO!E&W3*%th>R2E{V?{AjUlL6Q zu|690%5EIzCe7XSyj7Z+&Gk-vys%^)ANHfYdFGZfaLm-G7$>_rqf8c21S(*60>BW; ze_h&o>d{`Pevlr&UJpYc{w7Y93iD^P$VaJJ>@mlfK8~5(M2)go?-;d^tz4BPY^x_> zD=!n`t?zSnD4=Y!Bs?|bQ`+Q%amyM?p*T&lh1gQN=cm4|E}4+yS(>Z5l$@OhtOFI^ zhY2PYo?h5_e$Qo7&aNVYks&+-u=~i-z#7Pbs)f!XLjQWM1ZeH7LW(v z%Ucvk6i8eEYL~MlCMNdH-p|=?E5r;0GvNLhF z988Ri?5wPWPG*j-R!(NNW_B*l^e!GQ0H7M9J?^s@{~tJZ;2j-Oz8zqcy@_Bz&Zy9u z=ZtWwG+U4`6NA(M=|#pOI+}4B7O6`he+ZAPGowqRoEb{uo-t___R1CF(c|YMno3qS zUhSSo*{o+gKphC2`Xvy}-&Z=Qih3{VR&G?kVq`t%W@i>tY2N!3+ z&`w~ST_5~m^`d91carfPerl6?ErtA zY~6e3ZxsYzID*vTae<@ty=A_tfCBC7!W;4dZ!!C7fD#a30t05f3(_M?ZM+`ZpAj~A zmrud*q$+Y}XFZACGG8tzJjkA*08$=8;4zu-3`t!-uhd@dBt66*Xp?mcPbqfA8ZKc6 zPHRGhO0tSJoXnivYrl~7>=7GtfBWi7@9M_LUe2O{^VS7$?=(I0KCtEc`!Lh(=OmKS z)nnjhqM%UEmJt%t$R}(RB&w!lYo>NJZa^o&0DR?#R^0=p_sG)eZ%c3t3@j^0P7MsK z2~4QZ!OAk@SDLjkPDTg#8bcVK0Fl36HQE;BD9o4peu=PWGaz-S$k zxpM%_BWqJzCvm+6&2PF^kZu0^B`h6}-5mQ=;+`IbHSWcfwI0X(cM#D)Wa|V&fK&5n z?N$4ZxYSp-Qu})kWEo<3_^CslZLU1T+^5D9vv!reE_$`&1&wLmCl0f0_G}Nn&uU>k zV|Gaog!(;R_mhc{Zr^5lJfEC;wx`K#SvcH&HSAVzjMqBszTR&#*wkz4g|~84Rvy0I zm^Y(e*9n|gV#zu?v?}n7M6_H zUEu|ssBD9^i^uisgiQLZQdZkcYUjdvGF`7PdS|foNz9vGSYMcO;QS9RhLF$1Fq*St z@^)z42#>BN?Y9wV#yu)7KG*ygG*WJM>0LvXz*HF6tVEA(*_TSn)i!nJwV%QZ7`lD$ zf61&(`FpNh3(au00aeBD130t}14R+rmGCh07k0+N?_{SBZVV}@E3*szOF8#R30t#**Rb7ah`@-t*7haOsw$3c+ zNB43aB9~OLW5}s01A|;O*h$iX6BJ#{5u@$cF5&%LO7I^r^<&sUOH~bD!BV26ybFA_O+O^QIf446fj{iYeB6aSSF@;0jCH;$WrY!b08gM%osh5lHc1u6skbj|CMW+oB_5n%L$Rn)E6oLPpkD31x19=XHX=XhQ z=qKmT$E@K+fl7c~4^{8GQlK(0m0}`1#aZcqs=FA1GyDP(+si}QA?d62;4ZK#)7_JZA+px{j9;O6G7~D=-s%|q{Nz&>W21I zH@ex=@+v>W42Dg3igT*0@ccFUU(7vcpnQw=`IbLm!)7B2O}A@alCXkj(9 zr(c%zXuLBa^y0%1Ag=DW-5%P9Feh6g!O$S6w8%`Sope%FS0ocNCk#LY=0=Gm(FYtL z%0t1{YJg~x2NMVys9dl)MNhdfjNHP@M-dnR&gykOWladNf7=zaMe~^xJja&}7xYt} z^nLq+6KV&eE*NuT!tGq{dKXQB?zx>Z{}MJRnxf#w zt1<j~9(-~-7RlqP%Xq*W&0u-9tE{tAm+G4U3!UZq*;vk4P1p~E! ze{F&7hL;k!0_f-A8*l+8Pn`)dyhp*g7EzxpCsu~PFRELz@3!GrD{cpFEv)+SN_BSS zpsuIYa|c5)Z3DJgJT*mTRs{%KqU(yAqI4&{D4aMwi zqn#N>3_YUeqDFFmvUuo6h+l4&LGW4^w&;TP8VfxQK*cEj;PNlb=#Ajnwx-}RbDAQ; zrTJM%RswCD79mH%q*yW&^(+e&Ek4F%a06Vqcj`|Su?164tVj3>%?B%xK^vpUC|8pS z)rSScojMnG(&|hdAGraq5OF6De#1OaV!Dpk32IJ!SIN2Ver@|H^-=1^m< z`Rwse7M0+Vb;Q?4RUxI909I~(-4@3UMljI;tdt6q-NLVenpLPbCB!5ss0AX@9G}_o zg|cRkF_nKP7~4hV2!T^+U=cSdnq8+^%1?3f6!i~MaM3NQ6sn~{YR|rP{}9obR_Aye zZ(U}Q>>ds{kG48~6dcRGNVL!Z2eM#39B4SGl~uLmBm1eSzF<<4r(!V~kw@w{l^Omc zPkf{oBA*1+z<619!V@O9nFb7J#4^ccMgCC5tB2H#29adtPEJt;951mtF@#*{FlTZ1 z7w!awP|ESW3xLce+W}nWh+VB$j>4 zzrdb5Yg*`A)PS&*rSO&eMhKB#b_UliM?8dG@qQeFSJ*-)u}FHwzH=nv8Dsk7a05{# z^qXz{_9zK+BC_fu`g4LT<;pWWZhUU|GZn7AYVv7CJzSG3G82$y+Z)U!?+`1q1E#S* z))YC4TxFhxF}~sO{!f!{ysq|+OQPkKglp2@yvP9(LA4WTlmodW8dwre*ji`7hdD;q zR!#0lV_}1)?qVhUg>vz8B_-xRu0?4Q6poFl<9Ro@Gmpr?iV4IMQ}T1cm+pGM_({VP0&m3`RqGP5H}jBvk!SW zxg@9YUwZJ47*({6mQKIP*&JD+m#2flMjD=KhpbEJ_b`LdN3?RN8nB0EGRv;hHM?ty zOR{kYb2(Hf`yd5d#kCLE51a==6d4Y#Alv4b7WNAKgw2nU<*Q&0@&MklM67I?yXl&9 zyxP42{tc>#2^l7$IpN`A2FV3cvauSCh`nV{#^ejNjM*^Qu4T%~d9VBcWbyo)<{T;%e68Js4ThXfSFT}_xAguJ!FQ6o`i z;IjLnu{{mewlS`AJE@WZd?PTVHI}p8@nCcp6M;HA4sUCg&UiPI?Ahj=xgFqU zLt+(Q$Oa~lig|5a;5m`-IeMGjg3qUtxLZbHPk-kmm()LromsJr*FUCm`M{G8id{8t zoP^R(r)qkF&E(n<@@azmEY8%SxCG1d;4}wk(n97D5n#DkuUi@#6nD_>*}5}f9S)67 zTkO9Eh67LC;dBuRON`k%;v-$r^{0Sz2ke}a@+ z@9v*oOAC5&_1;5YEj6xNNKz8E_$2#-hFWxZ5FLkr2M?v=oJ8+ikGmgf4w_UC>Fn=| zr&e)^M`1nTYRWD)b_dVPQz&`le zDF|a4>-s%&NLN8whWK~3-4dgn(o*qX#&syn-~Vj5UOj8+6&}xK;sq;~IzDK~Hm~Vj zw-)mNR+PXVt5Nj{%NfqYe}zYNuei8gOkhPcX=E32H)4;@6-Ki646h5CNUhc0;Z0-c z?wJD;G70*TxN1T%P;JVJxypi)H6>x+Y&yc5yvU|5^6EfB z0(JRI0wM_RuQ!F%TN=M=O`y30ct1yBX@l?E6L;WO_)pE~a7dEM|kIncR)@JbTXujtq&{}xsy9dD)9mb{Jz6UuR zR}r1Nhsk7L96gFb&+{`+{fRhD>z+j&&g*$l!I)Q(RU7IPOx_+aLD-e1zF#b{@LNFm z^x;_SIEd|XwmlCNvLMmJv%F2Vj_|rq%K$R_3sUulB^`wMw) zZVaj4rTjG88$#aFA<8a(Yygr+mTTO!G>#JL7pZykgiPo)#c(UD7-8EYxuhm~Y{)_E zK7k}@I%Vws19K+}C#J`(-Kq4Iq}8@7;r7a!QxeMVu-F}zHe`)#RU>u1Y=Z44&KJ&Q z=T?>B`_`Jd>s1RUTbR!JGNid2`_HgXbqL;)Ot9R^Km`P>DXB{HNC*}191>5f&DxAN z{CV7l4Xe+p?(PS_A6lcY=&)Y~9<5bqA6S#8@~ABRWf@+NBhsbNs_a_V3;TMrN_sox zMbcpmtU!e>+7O9kWAdJAGxH#-Zch8942q#=0%|il=-mx@lHi@lCPxbsZ}LmJr=JCz zq0I;&nT{0WT@u+k-OUiRd}%;S(o}qXgFJ{_XDYZ70qm6+78Z{r=vzAzgY1; zm;P&g2JPoV^f{#P39Xj2a31+R&l*!^(_JG}u>oF>PSuCu5#ZpTpNpUkgjdrBcsPwd z14;Gf{fCAEn=F4heVdlFKLO&4wH2wNDsd~?hG%i z%aeedfZ`1(qqt^HVRuuff7qVB>RWbj@9aSirsMqOcEd7aF=$sr8#Spe0cIS{E)3o8 z4Z9zEwdY)$#y4@}?U|ZO!*8hPMoiyvc{i%20!<0JMpH&GAfOJ+RvWYJ&?gD~A3 z;3id4Gy83^vZmJY(2*ng_^9rxBE=krmyD?g$Q;1M>Vz9>=NlI>%ass(Q;#t;_Uoa; z$?P2n5lCZ+uvT#(G^~Fq^coH31ncohho*jw5wq_Gx4{+A-!TU-VZm>mk*`HZhJaHz!O0VA6?#=sS?EM^fAzBcx%rB2CCUnvSr!VKvl{l(;{mBq{Cx08dMBY@>1%ggVIu9 zu}DyqdTa1XfDCGW|ZpPkH&k48kLlD!w;R;soONqY<*%i3rSA@_v)1C8%&t`kmyO zHl4!e_=sppz<1N30Hll{E^PuHnf^TWWW5o_?Sc110074u)c61A5c!!V6djkH)gQh{ zUz(eS!G#-fz1Fakx zfkhAH)D4B6edJD7Lr5|8Fr`C5614B&{y5?GWI8&x7BgvNj=eg>ajUo3Hcb+aiu;_cSJjcmdNx}3*8?*{V7 zlA#%!R-2KTl9Zd0r=wOMnV=hCpjL7Sy#cC{hLNxOzi;|;%xE*Iz{j)vt#XV&07t`` zw`a=S5K<{6D>pMq6Il7cipKcn_~ge_w@dZ-c?(k`-DztDFhHIWui@*z-E&$_npWwb zJA&V~0V1lPno5EH&jG&n!E$RC{IV&@PWWOI!Fs#Enw}&9c%X+fZV~m!vQ_D`-e5%R z`b86ZkqT z^bc6qjq4XXOg?0aGcdSGJ4W(^4vC%MP0}o(O?)`Igq9Q+*%wc;$=9awb%XGy+B$fr z+!U;OQ0n(>Nr_*?7 ztufXKn!pbJq#}Qs0rdYhIRU+Y%W^Ct~=eaJCG`* zO}_Gtp)<1!9&-C);PaI14g(@?WjbO;Jn*lV&(+ztR@;^PmMVvXsf^CtGK9bJC&0iR zyE@Ro?e01w!ZX*|lh)cAzT0hTM2o`(+tP`U)eCnId91L(XXr-Cyue>_=tubJd}+H} zH^J-*A;CFP?Pt;wmk;D8Tds1um&7z)h2to-^G#+PEJbVh{H;Pa@)SJieP!|t{RlS! z5g<{}_c8Xe)|aXetPb2Y6(3^F{mCmwLk~tNNlX7e&6P%^Cg&yJfplUTWQ=-cjT*J| zsOWoNlM_)=&>|o@kb0{#&XEcQP!2(8bpWL0k@aifM|uzsODkbdnJ7^gbE*xA1~?2r z)eeGt2Z`YQ?}^!=E}_VJ{_Zx?W`TyMYkDp zafzNdPgaOv`7~akR}}jq)12UAfbqKLPt6zbfrqK(mNsaPUu~F56O+cbIyD2 zeg5xrpZno{nc3N^&b{UubBwh&#bbHNFhRxZApFM)B#g~rI24i(vXDg0Z$`%-F>abO z1%;WUFiJMVZvCWTgSV0}u>JB6i22Kj^Dt%i#R~TquUhOY>x}IV*jZiD_O$9`4P$=V zmJ)vS{WfbKtN*rUK6#w&Q{;^L;p9#;B(*?(KI`L*VMtNgLL5b>?ibPv}Re6Z|9lp`KrrK(F@RJBrA-SeCj95${KAWkGuj zvOb5wG*;2H5$}pj^^FtxN6d#T#T{J@ZC!L9Nx>4Q_fcyQ;8y2hHic<85s0N4(v_^hf|X84MxcwpT^+^BAD zc+Jn&!C5rE0p+epuI$%*nV%1anIpBs*SxZRtT_brdifussKBdge>kVe$FE|upVV_` z%k@ID0EbN{0idiWt;&HC6b}~en-aElT7Yn4EyxauQk3K$Kk}>xi>4YZ2rRuS?t98O zXz`0cl2^X-o7y|`&ywR^9$evIxaLAkPo=In7$5nyO>^l33Tw%zmFl1zE6C0(W8>Yh zGo3DJ${IM?fgN{(u`5*C0|qe*Vl2p z{Pdcg^w-#87YJ;nFtJ?Y06IwM1#wuw1d;Lgx?0P9@$jzR2dYiCOU>SGqRyM5LjH|D z$UixzHAM&U%r%edL&>4FcFDF$YiX+|GUwY)eiXnwE?3@~qL&Ns>GGJvnQ9ez*XhZJ zN_q1!)`!JP%Nj~fBHzl?hwqO|LsWAe?NWs4zWJR{R=t|Cog6b>oV?s6@yotNl%U6H zKceV|XC6~_7c^P-KQ)LiEa-79owS$dvoBEP-94gts$8qv`JVjK3!a@9E9C-+GD3IU zGIFX?9pXlCMFCq$Z7JiCmdcy>M14zvi@pz8ulG+0p9?3LnanQeOo=Y_mo6^VCkYU= zkn;F6jA5lJ=3T5kEfA1dkPgmsyyVKNd+cLB?Q%}(aq6^nqQ%0z%6%7bSpT7FP!{XY ztqmg28He$qJOaNhCrLz4Gjd5rS53tI*S;5N|F;B6slg0^1+k{8)^Wq9jdE{xec&I< ze`W9H1pY}~5+qjdef5OMS(#@kB&l>k#h(RTkT>`tbPgx?O;9zD{ln9AJny>rG|nnp z!joC9v=!TDq8F=lV;iw{agOJCc5&g^5^~>`?_lrQ7f6Kr+5v%&uVT`_YV#A2uRPVb z&~_Y+Sge0L-5dNTwTN;)pn_uI9*R9PpNAns5kZ@S|bFp>SkssffZ?CmJi^S{E zj(_!X-}^XOyyPjdfy1|Ihds9L;-chlW2|d!iGLc>lC8?HpNu6~ z)|`UV(4sIceQ|DHz+;h(?pNkdi$r9{g{aq;2((flMd@s%{gua7OB zH4etMezO*MEY0f%m*&E48zB|TUQB&*rT!;o%E$;{asZm z>n!{sOFE~eyYfVKls{F#CH5ND&%9I(*)^9{@NZY0qm7kbTehD`13y1(7YufGzNIw< zROGHSP#)plIqQ2ye zdkotefKCPX1Oh-S&U2+1hsjw}o>l>Dm6FFrYE;z9iq^!|uScAc^mqe<3oCz&&-l=0 z8u7j>iZKav{QY4Vf`_DUc?hp=Nyc9@T-MdjPQ}{vrbkTbojEQir~0t+IJUB_tf!I(;TnBkdn`!no#rB?P|TLE;7tf; zQm9O?D>gC5y|=@lZwtl!Uz2`4=Vy&iyGB`n@-+5S(>JCl*yMEi7%Aju4yOoPC2Jce zV_y+Yac1Qy{_^zYc_d8Tt^Nt@>7H3N;#%dNAq9&={r&&R`H$hlZn-^Ddx)rn7W@#WZOvbY|T;T50pHSO-!O-E6BHE@3l$VWJ~$5 zxfPgr$W#4YtW;3CtI(k`qIF^&3;^;1aqby1GKaPPcP3maZ&U~0#K5?RRsZr`?f*<8 zFw9pbq~TtU`s37HtiVwQjMg2{7k{exp>{WhIxxdZrFf@+%$ddCIr`_zW8!#}U%W?ayoDx++dE;TW9A3P&k=GLetSZU!I{XI6yi zsod1y^D<=BcPl)pOOU<`NWkv~om}zVstl^&hJB^-#ay z-x(3-o(0L|u}>?4loam)A#C(JjDhg?m|b61ujK{FyZL}cb=YsD&&r|kJ}%QQ%PKYE zJ}E^?9#t2$&AWW=8E)3X&Gmp;V`HY`Y9w<=L_hiLSKI+1Z52T`Qae!r^NsiG@-B71 zx}T{_50pZSp|Kg+(43GDWPLdcU68XONnAcwI(D=5QJ0{isc!>&oD>@=*xoK8gpDpV zVd8I6`@`=uC!mk3G}KoYlpxrDT$PV<#qT+N7xZNBdnQBNiQuO2p1^%>Eb=m$gCP$_ z(!?@GP88eugwrhXd%S@maq`bm%_5uuS=oCndaQ!S}^)8m>TT}EX@XO^dY|C$i&z{17g01ni@l(cG~*ENl6lHXgGo z4nuKtPJk*RQouR$s=kXUs7;*3@IiUl+pBwg`l23pU4ORz94~4@9it=L1-fp`BzaRR z8boNyxSz99)Rs^H-x>ICen$kZiikad&HzM_7B@A|yfVJT$!L6iMjCQ>M@g<8>u`m| z&e-fF#CT@?MtjCQGp{;tJZ;j<*{FgkE|Vy*RrGEtxTjy7B&A%F0B+w+tRds;Q=fAT zV}Rc3Wc~C`)dAPAUkqmkh}qD;QW0Fw|4DQ1az9Cu_h2_Z!y%Go_&tMh;83ndQbD-b zlbx{s3O#l*V~y~dVZ$tdqqV)f>!1QF@sZ9 zV)xQ~=KD+0z!LO#(N4rZLHk>zWXH>?{URb|8tOzV%wEA;ig`a5lG{K3Yd8B@<=wpN z^7Pvl5!3FI6d-+mlS5mOqu7ifM|`tL!}#K|@MPWL@^ZpAyk}Fe^f_6prc8(NG3Z#- z2I<I~M5PLiKH|Qy>Oo?5*sr<|`syPn{G66czdJu~XtGLGb@qxU?mumj)5u3Q)ul^=n`pV6E`-obvLL3br{-43Y?BUG+bZ*(gtOq5)w>}sDE{K-6MD( zA$U4Q6H~T4mT{yhq|d!wG;vAP;lX1yl8=3Ftd763=|33SrxvAAJ~YWApu)@fnvZ*G za2Pho`(Hpm3UR@5?IPXH``_{0>%}E}UyZu_rLU)b84tM8bz52ENM0_iYBjzqe)S46 zv5tf!PB6!=dg@+R?zRGi%~E);JS0DK&MElg<6yOxvd&Zsd1cgSwnub4BW=0lu?}H- zP^)|NZo3^A+v357lRjaPR?T|ObagBdUekzh|NPNg=oF($Y6~Wcad+bDtjBV`yP+v9p⪻Bkj;No|rH`^rzwRdwFbcXaMMg=rzm zYF=}0wl8=!f)AGrE-i;2!AY`}c&_}A5lq&aT7=c8pPAB{QZ8}%H;9`b?WVm2-_j*B zLwaM}e$q#g_D5j3k#4&*Zk!rqJIc}@x>b@}7V4Fa`=!g3GnmQCKa$Y2isYdaA2V5e zM--p;I;n9(itlovvYpWUVUy+g7JrsVk9dSR%4Wy=+L_vES}2zU zIa_FYsQ7zl$tqiDsh7z9%?)quEFzqDd-4;$-wSViOlOBlBobY<*^V78(Il9@=%#Uw zH*qL1eZekLhSimo!+EO3o=Z)tcuDr9=d3I4fC2q(#bkr9~?_ zCD_^|1zK63b8-AlxL?a3UXaFnjLO_;~%wut=xU=O<0=% z{hpxTzTfYgdWthX@gq=?dM|%7X>vE;rl43O#=0N(Ugic*ei;^K2Ry%bKjdsk_*?$9 z=cd5PB)Upq7rywtxpWA(cI4z`K#HD3q8tVKsvd&4#O6_qmG}aDPT94=qjLNo$ji9-?5nb6;fuWD@ z<*+{1YTU@ejK1H3udkywebF|ON`r&vZh@sUq09^CWUI`DsHaKU?ME66!uBG0_b)Fdo5+NK7P0$wryr!x%HHcl%BTgjZgskzvr5EGhI8BzLPmFpE}%* zIUJbP-N--2QhVpW^z{v6WYfO!b$40nx?{%VhshM*&2QY@NQNJ+H&(d^3^nm==EvpP zB3Tz1#gwxkbjhTR+LkXz>=)%;-Mg5(<{qkYd6ZbI+pikk288pn+hUE!M=|P!=WVl% zR=qD=+ku{?M*htGBe;%xdKova#+umF$Uj3D#T#kfF72P2tabN53!y})_@k4rg5{3u zL6P&k;WNJ~@u>G(?c43sr&_yl1?ZQ5eOx>zHy&;4-aHqD?f<59T5`E9|B{`H=xItp z8&9^~qizP>t%J$O2BI9IjdGi9{#>YLWCPYT;lUosW{MlD*Tpyz{o4@>IxXkoB zHWA4GzSa7r(3J5~2e(EmZ0+m^x$49dTL&w5;k28x9|?oLY@}-C3r+?wIRytoDfH$N zOyX{d$5e#4#p7c15LCmOu&bSW3ZOPcXAmheqSh?ZyURbfzq=Kvxvd?irD$F-@1`^2 zNUwiiIQivLz;}EIce?G1(;@YWzp=~uqWxA3J$)B$bO?{f^=S{CafMwu6tuKo8BE0P z_R?){c=4BONT}20Qr)&LHg)s(CAfT^?-CJoherMxHf!bW!v#A%@wh`)Y8fBhUJ?T7 zT!s14c+^S$d_it^J~ghq)_dH`_^_#iQiBnA;nfwbUEj6V&vIzI0FK~-GA>A7s$!Li ziC?cQ?tzY5e#IH)yq=U6{ppe9XZFNprDq&i%np9He9>?{R2z;|s&`qjZeY;PUIEIk zm!1GMKxNnh2?>-QFrVB@=bm?Q)t;uAkL$|3*k<-02z+_VT4DCQwW>hK%xiBfbHMm$ za7y)d@ynlk)KQH-J5EVJJ(qTsyF*>6=0*Uuq_@D)_Dwq|_xL8$RFmv1?h!XLru-4j$VrWCyN-u0PDaju7>_e&_74(rv}B`FR_A zP^h_{yT**+P6zz0**J8``$z}Y12`6-60GV=R|6J9D4=u-+%NP>&ZXW)69%{MuTkwT8*ccqGhzBPN$tmixuZk4ou5@M z%OAFsCv&0Y4nD(hw$Sr=m>yM<)~BlgxgS@_N=dZOk7f$ukW89x6U6}F*&2TTi>n27 z?eoXQO7`R3%WX%{qA}<79R-9eK?nBzV5U$t3kAl2Y)EvjlASrZPVVq+BHL8D12^}(OF9ewmLQ8B&E>n z&sk{`FEia{FP|@n6fy!0E<~^&`6#8Aew4{0Zuwc3?E=b?tpo$|zso`Et38Wv$mQDh zM)Kv>{Ormln-WyZv-N!^ucc1kHjeBPwfQeaWfSbBsou6dT<(H>zZ=M%yaThI2w4Mb zy}w?|sBpDzXR|+%<={}_srGeRd>*pf?XuzAk-$DZFPy=nR7hFUcHORKe3p1)#JJcJ zH1Lot0kG5%HK1I~)eiua|QmJd}OsqpJJT0b8=g^?EPvP5Z{9agI4SBP1L zPuX}L2?-u1&6dJMGTTAVe8XRCrNfHbK^4W0D7(s>C0l<*2WrDFH@kU)$S5&G(DfYh z=Ft1|hV$4QU-XQ4p;*eFMz(_be5tl46N{hUXwM5s{TzdLU8=a(dnBp?7RBp2&%@;T z)vy?xd{9W`_o5{Hi}r^|A4^HXPdm%4UXJFS*K7?ZV`b@n zd@Dtgh_K#?Bll^Y2(?qr@sMoB1f^F{TocAe(%?w`@pJg7>aAwO-rCA^jHArP`nJ*5 z1uyUKB+EZ0QpsmB=4Ht3Ka=G1)X5tITwHk2Nehk>&V9$+zCw`SvV3iL(vrXRo|nJA z2d{tA!bQK;Rjna@%JQ{qcXBfoGtYk7$tl^J4MfD(CrA!Y%i~Dzkg1LdgD2&WuR9vDZE!9T(@*YdJY|AC3C&ZH z7gmEXL*8zC2ZK5Q2Y$?IsS=Hs_2HOi4W6qEH$4TNrALo;vf#lfj>Lyo&!W>d-| zUZwWN%W?({pU`Y+AFn;6Kz2=+(f(2mIZ*w9`fAW+S@yNn^>;kuSV`VoZ5A`X-eB|= zPeJSP34TcSX3hIU)GkijmhUlMhPAzL{(j^k^%}$X0;$`Z6h~3S0VT4hcBiILcfEZ@vWfGpNi#Hxk~(A3ZCLa+#ce$(Mnv@yKNtzGBV+UP zwbjjDZo^_>Fraqd+`68Lexb13>gGWGCvovT|G3{50GBs{H$`j$k>R1Nxs0HnZ!zxpa4H zsgc;$_LaCuZ05PD2T04;maKC~Qe>xS59&(#3g3R`ICy>th>V);*DsPhgXVxYHzUw?$V0F~%-F(VDH!?xvl% zP{!Lw`&H%y)KMpN6qQyJaugp>-o1lhPYWZ@XwX$;kOv`Z_BZ9$6dqb70i&alZJV{>1Z;pnWPPQ&cHHiazF6_(>G663lMBnqA!g$D4PRJ^R&_ks{X=aBWo7p2q zJ=9x4LqM)2k?ZkD|1521+2Q=m2E+|~)9~Y9j1wYak7L5n-%piSFK?8RqYX4YL|1U$ zG;YL^Trwz8akoFRquW7VS#&Yl1dy(d=2X>-O`X=Ijx?k4tupn_INH4~L zR=XYx)Hn1LBc#>Zk{+5&S6E*QjtCl(J8on^-F^UdGG|AQJV8|2r`% ze091Btab!rG=?Cu^Eo-MY~DPhra+DpAb76|d4C%mX^8r_(8WK*n&RBMw3KT0dJq62 zO#1R*eqidjwSCbuIEN$U%hWr}b908_R7Dm86u}^Jul(jm64xdJ1H)=cGzIu+`WV@!$yds?6&^_^Obu zK0g)#a2c$_APOG~E;_)d#LxV{c3C%It=eDI_i-_(s&eX_N`S(GusLB+ z)53!A3ZEJNgafVLrq>dI{}Fkr6GD}3kJPqA%&Iu6-u1GdXL~7(gS>FLw9v<~H(j~e5>cujj14xR|Px4W6TIi>RDDo~`lN+$wF7P-)8@y4VkQE@8MAv)X#*nWAXwQ^*>AIazbK-095JiEy_NmJ1>$n zpsq9gz>M~hMXwCNmoRp)Vw&`Z;ub%zL7e@uG*qy& zz@e5%dN3%8f;Oc&t4{g|CXXktaHPS2C!f!CW%Cdl((7{sqr{V6^JxdBw`&H0*zft` zpeMSWK5eZ!n=&M@7fw`~Ni~ju32tDsCjNd_qo5dv)#2GkcU!N?oJTxz(O-i=#7O%$ zj_!oW&bDs{XFLdipRGs+F)Hv3=^DbW;GotA*Lk>KGN(#$kpDKiE(R^!q;}{vNt$lQ z+mGx76$a>SRa>D}+7enmG^rv0iB6skQl}&;W;^xq*ZA4SnbWQAyvI^uqjaGrysOX1 zx#Y9@xvb}-`wrD8MqL=MLJzB@a>%dn-NhI2&_{zk9)%=V_-2N8j{-nP7vjyEe4Buu zIegGuG-B_I#jGKp0}1};jf&?X4&ZO5!kQ@uL^)^|D{eZhJNzUTtsHdDCp+FVfbFPj zF@~B0jx;3E($UmPfuNj=SLox1n_r$*EnFz~+_)-MjgNqfy9>=T!@i<~9!ta&p#7p@ zUb$%E4X}3Shg@{^ijh_N*}}6;0DsF`y1DLm&|4p`Nx2~)C&QE0DKBF-qYrQOX#MZJ zm=y4`Uq`8Gd$;9l1Dnn2jfsv$jI#PJPC1obPflJIC8tk zzSE|V1UTMTnr!y*;uF~Us(FcDqR7nixI6nre9cDC@CRU<=gOdV&{eSP%Zlb+fkDN! zah~JMj&@gb7R|zqlX@TUVXpeo_Zu$opTM{t;n%8ax`YO@&@i->YIp5kDk$K=xEzo< z4Y4;lUcU;hW}LtF^lH;T0Q>$~A6`4HrIYsmMYEq-TvK@NwusqU(}dH%EOu z(^LqP90Eqt`UfqBT^CNHqTiCHg0#}+et0l%X8WvsAfiT+kASY#f!kng?PC-`-;E=<6dA z>V5CG7+Woj;f}a^yrjtBAXiMc?JvGvE5WZqy;_t}$BD=U`@4U!tNI)acJu?E#)E!p zp!*S!qCuuO&h=*iCZE*29j94ZL?Ds~aJ304b%&N7x$uh=ei?oskk3g@G{!zYFWHfw4RUrAlB6PO5Eu;9Cd?g8OQID#RRxW>h^$bZvnWwvC2i7D zmP}hNdxdgmd*5*luknZrZo6fv^yX8@HdiX8R4ZVQW5LXNSr)8A^c?PwF4KzQ8qeQ0 z$cJpb);oL6?SogmU97GECQ?xV@0-`)^EBDe_1P_oC8G2#iV_nbg|8K0Yk`W@1wjf? zG!#lhw=KKO(?e+*gP(~^WFiuJ-Q1P}wMg-$ewc_pBJV;?4kB^(M zs~hFKpugS|u+#JTG9`!SR&UQF^C8sC48!z)dt+qtrPy5!yU#MujxUCnz1(eY#O*^> z2*GE;OmcRNQc8i_N2N-!S{BYyy#1I$iu^zy*_c`MCRohBpNxK3Bi$1ESl)<6z#!T- z^c6UqT%Jg(hY)gir#hO;GQ3 zdX}v+8Q;tBgreU+hc70`uh<}iG|*6)pBs(R?#yFP9MOP1+EO-z73a$Ai}hzVe~5$ zC0=3mCrrk8bXyN(dNyy9^Gp_gyGL#Ol%gegIhTo|H(`cU1~qoWPh_QB*zg-wNX=^{A$$m%a`if^p*9}mJVo9lA5E@k46hGjNF6? zGO{X9JvE4TAWw~USCVS(MPipOjF(&gM9(i?YnEockemg%z%r4xrD1%UYO_Lh$yi9} zD|j%JrktB$`hgPpLe)F5G(Ff zqsrV3CO7f3%1zis$W#9fwjZ~0ZNU)H-RFL9IeqeJk4!4E=cRtxx)D{8=%#|EG}Oa2 zUmt1Q>6amePBrfMj=H9W(5&phuY!HZgtYh}p1yvdRg{5vS`cW!DB>B4)W88i-qYo$@50cD8Lah=15{m}uL2Ew03!a+DM ztOseqPh(>Wo-e}BQX;bc7F;~{2-f)v_88IinJR)2=pPT_OeZ(N3_v7)-78rsTi*j5 zT$23sYAsu!g9MMEzZAM7a1Mx^0C6Uue(jivrJ=OYm9OtY1^0k)#|<9Ba(SqP5n!S+ zdWQ}(6CwFs;0{tQ^$g%2xiJSP#_V|UzwUJR1EH(vXlUwxq>bfB4Q8LbR??2+XnZZ6 zw(cf8+~6s^M(GVv{d*^Cn|Ujo-x#tdpQ4bp9g)b<=zxih{9jXld4?ObqPrOFHe`r? z{`yWSDW(AICM@-zj>l7pRPI9aP?i{-l~#tn>LWMFI+3SVz6;H`(MVa?14?5wSmM8i zC`4hW82_Wif+s)|__)Y?6gk&*?*$H8Ov5eC1F~mRw+nO$ zUE~jmjDM(Dh+0bz1pPdbPm#mS(KBniPw~KMSYY3KY@!Qbfhu-m$^0m!UbBgq`>JfJHA<#$eCkP2j?19c{(&yAh_XaxbPr(xxYq_5Yo-=W=hI8dZ8y%{c7I1 zQ>HqW?uG#izDj@!){&{+Zp+{eqe{`&Amt-;B9o6)+YV%!b22_EV={bCFtV-oF@WU? z%-BRfESsnezTeM?`!|JQJTGLOBUP?`dH!#Q8pq)j^7aFxPy|=_3wbifM}NJtzsSdN zRV?RyrIP{QdsMy1A4WyNOwm#vSSt&s3nIV57EvHc7ox;?Mn)N7N)(I9JjqbV+q0hg zjwBBtxZi$K$`cX8DWvGzF&mVd@-Gj1NhuXuqzeCqVNzHM;BYG2bTLIzz7gVNY`BGU zt=eOGqPkD1Ooqnh8BB&0KRLoM6__#R2bT|%A@;@u^i`^l@bN{w{0|k#d7OOu7w~>aa-_=MgRw5S=P_8avH3TUCU9t~MbBDcx;dV&NlLCWiTzE$~Qk|49 zqYs7)hZvZrFd6ne`;1wyimDcDNXYy18eXgY=+<7(7O1csT!N~NKD;CXMP(l+TSTIlK31}2~%M`YlG z?eM&z|GD`=hqs6tCThEUB15!B;p2wBXn(Jg}c1e0>m2m;v!`DD;g>Z^`(*A+j}wRkO%B&&2p1e@BL)=(R_kJ%K+yHlOJ z-f)5Nm||csv72EkR|!jL@;k@^>U@8{C+{1VrvQLkMRZ_7&GPt9d*@q=w;57w={Z313-lx@EcBCMC42sWBu#pRmnGK!*i-VE01$js3<6mgU4Z_ z^aPQOLu)2jAwugtduQ z?P+5+*PqenWCce>M9X=n{0WPS@YQQKWQMNts(+a*r-qtpR)5|hapnQBKU7&&`eiGPlu)R-?$>05WKzHN(P#C>g7o-G5uFW>kp8d!ETci%xu%?8Iq&5s%|{#hP; zU%Fr~S`mEzVl&*zPS^@1qx##{c|O0^tT44`aO%5947TcZ8&s<`CWfa!eYW6E_lz?j zxy*f5Y=75vR7-w!J5h)H-3ZAcCGWfoBV(y@K3y|l^a1$Yq|^MSd*fmx;Hy43aW;mi zt)n!3*6YC zcAyoD%lnU)9g0LOTJNu&P&p2D6Z-V}pLiWcUny_Q>t4$VuC^vrRi2_+>lCy6 zL;Qjz^oB9a;KO>hs-soYZ$Xw>q7!Fxvsrtv=NdQCj0+oFyjC*tNeOJNS@mr?3it*a zwtt}FcIO2Nmke&439nV-iqxhoBR7dTBR9b$1e*15eoR6+an0Ob6%|!?qiGosy{r~O zUw!{*x2=@j3c14laXSS{eD?^v+_bG}EBMfX#fKtD<`G0`#jzrKTl47m=2IVvxxf?! zuq;D0dl+Mg#wh28SDV)&4fZ3~cG$dQ+VQ(9@rsA^!)h$QU$_YOa=v;jmZl*^R2vAL zwZ)sFpdZFghjZX6{aT_&;X&V-)hr9z%vkYXjYs)>4sCPCallqOl_x7BjoG{azr|RH z;(I!=v=)+8I^viDvtU7E(irQp`P0%%K?fG%n0nZ!-UG<+iH`3maaNk<({d7wcV9F1=pU_}*HVsNWIH;C%LQ|{jc>x-tb-~$c?!1!A5y&?Vi2#ed#lCVHtKYDetRM_%CN9vOGd9& zF5$^fg^T{KT<~XW87{xuY6h<^`hceX3Z;=3&YNc31o+8g3i5I<%ZKX}mL=D4m4v9I z(hz{JC5@eM*4zEvmdIyl$@%*+){`8REq)Sd>t>+eyuO}EDpb)IT;{;0Q%j-#i7%or z)6R0<+AUtC%TqBoNp12oED0CU^-qY@y-XvVr2A6dobSlSyw(+8Q0#b%Di~|NIeh9K zy|8JkHR)uZKL74~Rj`r~*U|F^X$SFsxaHY8Px6@fkE`&33@c@D+N^D=e~N+lm#ZE|#{n z!cL&{OYE!Jimg+Z9)|P-y<^ZPJP6y1!{4B6&lOt_!){!o2TGqw|F&6@mszo;`pwQv zKYa0w9P@1PShU5`e*JKjp`#zpB{;sKkFiDv8@pionO-3bfzput9Rt?!1{2e7elo4t zvahn}ilq@<;lZWF(x8u5kICtW;ot^@bRqQP5%Cd+sdE}CSn=5AE9#WnhX(xT9{QE) z$0{-;_WeK<`53l5&Yevk8|mXSz~NY?w^<-+{M=@SZxobPwnCR@i|iEbv&`rLZsRZ3 zck>8bm zqO7{uOKC?#$%LsoUJ_uNgBCUl^Dx^6a$KxFCcn`bl$W$@lR?i45(Z6lBrAt=fw7cj z(OBrO%tAV-1)q+f9-z!WRD0Y!L}EuD)_q0*UVqv$!$t>$8L|g}Mh9-a z6EwbqDuPVW5ZqIeccZ(3Tfc!}Uq|WH#sL~)Y5WS#T{UM>5oZ%$g4eV6wJ*8I(efYO zEP^#O7DZp9`la2o=^tj?z#?9UG~fLOIzLO$b9jgfY<}#n@K^G;27%t6E@GaTQ!(ci zK2x>E5y@Ex3>S^St3eSW$H7Ep48)k)&qdxIkwo^H=Y~`r5#D-7@hV?NZ$1RA+ir!s>Lc?)NT;~Td%GKkK0H%nI9n&Xef-wyxNBF1j z;)&PtdWFQ=p{p42LW7ZXcN0$s^3Y!;1jW;6(F^;PYtSIvsK&Zt-}HZs`24QJk6)!D zwU~>gfozz;4v=65y znneDO%l{V?MK5~_i=duvX4k_9piSS1rk`tSZwzQAi<n4fvoozHLI zOsoSuwz9q#g#)NK*zJPC?sv(n`Nurd%(%v19)XK$q!&!nlFPVVp7@JF=j7G;Ks)3j z%IXusec@>(8AZyUAnaBdMGd~mbB9NptqKReTb)4(NOI=u-raS;;m_;PvFlJ)^eFae z@W#dAvSyzJ$(N;8cZiF;36gwg9WYw+?^mLK4;0n`5Fht%+#pT{BK^|zZ{u*a4*bch z!lSk|zzx1axsMecHZzTA+n$xgGW|uN1cg*INYUNp=pnNP;gP32ipf|*^oV4cX+Egp zp&R33T(Xb*_86-CZb3-XnA7J&wsApLosHEMn9@v}4s_R%j+}U{0An$M3t^v~nVZ_f}D>|m&JE!_c_$tG-Q-KDB-nzV+ zg8IZ+EV)Q@DtiQUb(zD?V>99tq@J0>j{6g&AT*=sG^`CQ z(a;@j6v;)~E~brwTK(srUKMhDl`J*FlFdS7FF&uoSd_(ptI`9M0dWi&a{R$S4kEjh z^h8)mC>G#94DWhxs|LsBvWsBElcOGQr2MTJ1o8-m;-Gnh1TmV0!H^pljU?;B#YJa? zuH?AeA0+r#{MX-i4U)caoF`>dy zcSBGn5e{n6w zy6hYXGrTM= z_<5S2Joi;7=$cXt?tV^iM8wfJvj27OrTFHb4Y|Dvy60%GyGp0G21nr8(irO>3O&bc zdmbTs&8ITY;bO zhme~rUVD7&VS@!hcNF$GlAl0g5!vE$*PXw(=!Zo}`=&bK&7TN@5?zEvi}RFC|UB@By=xAno|eJn>O+%jq1%5ANV&A`KrH~FkOo6`q*+9F8uqIgF7J- zJD@$h>^Ff|h8UdErYk0x+@yS{r`PxWbj=3ec-ZC>e0ouJ$8q0sl}e1e=Gk^91YYjy zMw$5Q)JV(=Pwe(5!(!i692k3C}qvbPVn zTOET^_pvg61N{mMJsL5Ikj?w=J6}edHf#jI;brJ_b2l;=JkAFl|C@8!lO1vHLY>`a z@QY9Bi2F?BzO+f&YN*^+l(ixf-%SlA;@an=0s$h(hS?b2D< zBR=miEQf`!a!NIGp8*mdN#4K1HyWvo?rc_1>r>|-LpB4C+nTkE&OYV7wKI)*M79;#f+3Z55IN3OY#1FCv*tC3R0zsAXSVsfgn<%lz^1bdvDUDgY;e^ zy*B{~JtS}Ny080rp5u7#@B5DT&-Wv{JF_!8vwLoKXLsg2m;T=KjzMo+{$p|gaiPJ! z=$Ee!2w7%)!GT){QU4S6;if)A_8dFsQp-W z((B7LX6AK4WK=^E>ifR))vNu?b^HdbNyUJ;5!Lio=nQ*bBrbh_;4IO$3h>IqA*Sw| zn3~~B-%Ia$46JPQNZ8W>NysjJ=xCQJ_>ui<1`FO+PR8%H<88=whkZ` zENDcJy=E=uUNg`5O65&6cs@2H02XDgrL#wQ%l4288#n?eY_bCtDvf4ZR#ll-pZ3O~ z;K$$+IaBbYyX>BZnN4^nq64hdeCxhzu>?g!3FWMfe%ayDK-^esacl*Jcm=gnN&rlu zrD&9}7H|?usSxO%#y#h#4saL4v9sU!N_jhi z9P%IH0i0Vg*Sy*ucufRVld@Uh3`OQ|UG;7ei z%UF|>vej|jwc}nb6tvip>AX1Qk#LAGUVMay#o8#CU#d_h+8zj3&EJc%2X-`N>w0Oq zt2kKl&D3Ij9~vUG5o3zS>)-KR1nx!Rw`!zy6JTcGz`KARtXtK7r`=m3 zeD94mh`^pkMD@HLVKA~6^9#nZ1$~LX2eJzH=f5bc*g6hDD$FSqc=ziC$3_c{TUD7- zsdRz#GjyLJHkgKanC_ZR4D)y$gaYh=1~OrF5w#OPJ+djI*Xqv* z=jQVrTP9y{X8F<&W?b>T0~Lwu4>(;K+XPq&uX?P=M(bUfrso;vDfvu<|IFXoBywVF zL3HEHE=DQ;uSh4x^1*0iOtKy`CPiM5J6Jj2D&p;`Za{T6Ww8prG;<3KABBB({}XT4 z(UhrMq2GmkSOE7kY@;!szxFy7W8tbJZ8p9%yqjBjft8v1ghAHt&B@D(z_J+uj%q&Ok1 z0q}TJ|6CLF2l5H`^3rcfT|~$TT8<>M2aSoF&64*?j2=c;KwO~y)d|C+W|cHT_)&6B z3Gn;e&t`&9Drs`ry>6UW--p!nffNgbrINlwJ54PPVJ~2-blL>hDUTIM=pmDEB zW?q{V?{LFWZ~^8m`3@h7UpkzaJDjyD8Z*V5r0v*Y19<_s2~hw5r!S%;*@A!UuOWD1 z4vuvqMXMY%?S1_E0slpuK>@i6qyHFuH5A{!3uk-xw!^`;?p0tOA$?D}B2NvTo9L2d z2}?1jEqq4@!D_z7QA4mN%RZ)T`!5#wXd&44+UMC~N{gKoXZB@E`SSsDFYSR!9fL6^ zefqfZ!VL^aWBQFb^jPdaCWpyt4YttTsZn3h|CizndbHs`Bv^c0b?x);i!jo}iQsDc z_tFY2=z=gqtroON#JJ>w)6-{LUss*<=2xbvW^IaB@3gVi_nm$Z3X!K9LB)YJWi8={ zx=Kqa?@o_F7rpjc-8f#-S)1EKQD(mJumG=+b=G>d`i>)qSq^-^44OCiL)rRy3woJ9 zJAUw}|4LO*)v@IBeD|4!b;CVYxZkT*xe@t+7E?sE5E=Gh4}=u@4vu8}ExWcob@QOt z2+f+iGgMI0SSVys9>}os=3FhfEvWBh!>K^q6Xm14#v0%pF=0W=7s2nDzIy4a96S9W z)4|tj?jc5S_b~nB<(J9MO~Uz2nk+0iSo4Vy`%rmF8K(~gHRrh5n|G8x6mK^R8uzx= zxJ+&PGB4H*OAIF9$;8_V6BWtCv-dSFF#Zx1QI#O>x-|wAdFDr*8`HDPKIt{TJGAVB zH(frh{wQZq&23s4GBt5uxE|I`xtsbrFk|%da_99;yEpG)KkBclK=&fSe0_`lJK<`6 zjrXD%Xl*Mhe$}$J265{iIj$mM>!dDj3j4WY*J?3Ca)`{<@98|I61Q*?{h0{009Z|h zQ}R6HXWD5qrka(i4zP)loiD}yF!i*VeSO7v{hKz=m*SuECOaR*3T(OL5Vp(en%9#g z*5AQ3>%^wb;2ge+2C2)0?RNCwQPt88v6IFUV|oIEpmSvx?x~LA*4Dse=lOMEss{cF zsX67?uHTZ);FHB^v1`~sL4oWleNhLe$c!pQ$wvCBATE|}v`n-vWud56i-XDsk_U!8 zDn2G(sH8hVw?bGSvz!q5(4K?@-*6fTgiX???JPKK+otVUyk8EuE`wgGoLMZVO_zk? zuL1U#mOn=ir~D)%s&P_e<>p!NbJmAFCJfH$13*Yu$)-W3y`yc)iH-bv5LcoPiD z#kcVyae<9Yg!7itr5F5fpZGPEkxcw#2974>B1kRfjPj2&oRUv14)+3arlC#)g8UPZ z+-#z?3I61M!Ff)anBaS7Y6 zUl7YZbUd0brkg6y{gJH3I90yE?F*7k-k!3;HA`+esNoj^7VZY&T8TY-Z=X?GC0fW3 zFuV$+7bHcwDTFm%GPPRlf+Sgb;Ooc_ps7%5P{e6Al5vJh2!IWQ8nvGJ!@NkR+Jg2g z4P-h+T@Z0mAd7pAEd@p`=(crtA&Yz0DqJ#RZ$X`tr<`y6394ZxY)Mu%uWp}daQ(C3buBB0Q>rp+NGF35KYS{C6Pa26icGHcO^7a#uHX z5E#wl9hWy1yvwlHqEl=SUI5%J`% z3ARslOMYNgE9r<@DzWZlYi7f#1@G>ic&iV6^UWVK_Ga`fhXjMVzzWbf*>UAdBrn&2 z8GSbwFUQBY?{snT4`)AI8A|m|d>g2<;dGYa`)`)V#XGx|Q&S7xD<$?XL!>)(mDPdR zR?o+&Y4ePYBkC?RZ%k%wLNTDiVYIuF3p9;CeOJH2opj{~jS4V++IpG)%c|IIZ0nag zM>_1aLdXF~Yih;m17Y6fc_%(qcNrlV*;an%Bl~# zurH#p>B_kM=-I6GDIMRma5t}1zn$+d8m)j>xBH4cqJ(2d`-an1-#@b%N+1itfb314 zd#MrcDZK3@$pzSQ-IjL$6J2N2LQZE}7x1b;<^becAmfwxU>Gf`>@qU9;Y3>&q(B9B zkA*wkMeKAQN9etIxvn?^q_(YV3vw-6aMT}oa_C51-`5Y+yO|$QZYtz&#mIC*a#y!z zKIIjzhp?DGQ}Ri?zTuQNjK&bg4dQ|j_D(ddO<7P18b{oOYz32F}8T|G6GcugSdmPUy z<3E0zp~9)^L%P1dnA*=Ob*JCX06Rd$zf1#k>&dR6c<%T8%qk&OEa$q{NB5cYp4wZ< zn-{Ge!F*>6Qtm8X|2U9SjYU!gK8w3MfD!|X zU%ERue{3HYsq@)o=>-o|rIu(_=f$X|$ydujq3d?LbzgQgF+7F^GDG3ghyYE@Ci6oT zCrq<3KjHASU9)pIx(=s>|0Yv#Zb`J)io@m}uYdN_0&+EFans`P-Z3nqi?s z61SsXLd&X>h6QUvi%k@_W1gx-I!UnhR>$0^&OP5=)5rDpnvY`O4_surkNb4G6+~k{ zUMcpPR&NNpu3CrRBJ|a2?~LA^H?@1-@dY7+rR8d3_JpJqxqK#lHX|DKXNRk(i7Nh^ zXG(e~U6kg{6vIfr>vQhMLMo?^9~S|>#~qTZaZNCY45*UydnYzj|KUJ#Zil4#huODLqsR3|%sr!fM(%IPaeScY7Cl-& z5<+v=f{MccRm5HUG~A9i}hT}YHM4SxM>UyjE6`kPW3r$lX#c<~JK#p0+=W~i>a z14EyZ#Wnmhf%Ba6amul%UmZZk%tN~%dF$K@*8z@^z*Q?`SX+&`T-!ev2WUw*st7<>yocBOXPn5K#8gr&|&DLQdUW{NH7JY zl0Xz_`yRaSGlEIU_HK}3#iZz;Moy7%JEF6nsZen{etE{k5?n|Us+3V;dXZbDhqZhdzSox>eU>MIRPr19CPq(RNKjB%^l5a<5eJ(@bc+Xt9Gmte&$BDE z+>Rlo0R$=U3R1_EEb2Z2MYS?Y3PBg8swx&VXOTJ>1(P10;PF4N=OB`0RAOLUGJh5w2F{;mN`H^CX;OXmACH(q zD`5c|BnQ)B1MiNW@$PAW1!_DVo{;?R7=6|Ze#jz7j#~(bJ|@PeJ};SPbY#gU#y=DM z$$UZ*IhkZMpTcZXvLH%en#GR+uMy+FhrLzVqZHULoo9TnnKRF5kmyN-|88ExazZjN zxv?_{v#ez~7&sjDiF$OzPj^PS{c}s;g7b6PIIL^tjI!1g&+NGcUVe8G^0NpZRd>jE zMmY7g@0<;QGj*2@@mfC%@m%;|ob0c^=`ZO1i^usQQZh9&}G_?0ZpOgWQl=k+Yh6%nD~%tld&cxcogEjvE~*%YgD7 zh2>*H<7-vFqvcd+dd6Q`_=?B#hD2~%`HEst+v9Db2X8rsw5@7)5FZbhK3ZjeWrT)n zv_6hj&F?a&=9j_yLs`Z{$8jmFh)qN9VmRfrh z78@m}BHBEWpDm8AoEa;e3i|y;E^#HO4~l-;O!hAb$RGiaBrS+m5k2Sa!0BlS_Q$B2vdT{lq(QigSpCNLvLplZJd^MO~=HF?80$ zW~5RSV<$APu7Z;4CAR)Iwx8Hr6mDqR1nI~Tae~usiXQ3)d_CYPRR{@ChSA2edt~%YA3V$zMlkO3f-$q?KDdO*oIcS(Y@9L zfI#*)AQ1-An6>fvyBSqfUwCzb@df9^EJrOGoq_K__X_+&@L$f<|DK#Sb)(rleG@T^ zUWo})pp;n~o2tI->I4Y{f77guUV`I|LPUb&a<7$$P<4&)6xNy}QQoFZ-UiYAVeoN6 zw?e-vNPtV^f8ZxuF{+@~Hf65z3EiuLB2#?A2>##Qs#UWC$xrkq-YsE{C<8VwKVK4e zt0>@hJfEn*m{WJUK zoftOyXL`0mgR1Gnfy)@sl@DLOgJt_!f5x8L+NGFNIg{h2nusC*?D8D^AD|Jp6Dm%I zKRrGS)Gsx1JqYdJL2evt%yJOcl_66#^b&8KAJz1Oq+F8YUAplsN7_a9`DMmC+KjZm zRb6MU191*rSdki@ZdD^676dI6v>+H15l|IU+_BG|wo|#<@X)A} zfY~rr>+NY!ba@286lAMtu+ZzVS@S8!OuY~BZ%x~T8ZW=@A)l$;cN2iEQVf zjqXOZRmR&|+pXv<@HhpJL%52!mB6Y-KX>AolpC@PLf_DbX;fPWr!$@NIaexKwd8~JPafPEbdajid)dbI(T z|5z)I{!)a0;0=8j+z29rAi{WsYGV8W$RS()WMI0@59r;lleF%t`OWJu!)J0`mck4MN;h?w4 z+-s1ugMca;h*iW(`-m#`Kw|HM(q2s^Gl$y9(BJrm2HZrZB- z$~$2KcP7^9k4%T8As*;P&h6eqbibe6?J6s>YRXcZQtgUMhm2>8TCpBPaz+b}=MdFX zHv}l{MS_Fs--;ub?JlfEic5xOP3ULHOshHy(tfHZ{FohLTTSscD}7+)i8+lx3KqKzI4UP~msjs{tOCFbcM7if;3y@tG*&1<}Tapmf6v{(0U?nb};F8BdsA|BEFPazy!t@}M-+erH>ni@|lfLS7p@V^) z1K+fp0kB(qAuw)0gJ;Ro4f$erk9%qB#Ijv9@Yy?mu}A==C=Z&}Qr_*6$qroewd&SiYq#D47bY}vXoVoK0_%6$VZQ;ggIYtiW_+bjED`X7yv#eUZ-YCout9opn(3^4?by}{F z09Mgh9gIpMkf54XPV=f{z4u zs{Ew&+wz?sK>s=|H}N)05AI57myg@|aM?sG$8X!xFhb|wWl3OF#E~?-S8n?SO46x# zHQc-Q?4nd*Ykt5pXEfl{_EoTH{jYoTDN=4Hl(AK^;>bgxM^JGj^tlz8;sLuOXUovf zzyaX}JH0DRM`za3`$0#4l`e&G@WO6?9}2>`1*p(qQlMa_+Y>8xXDW&#EceyUNE8o( zTx7&~Iy1cuUNHXDgp9zP-K_t7LSji!gojVUN++jf$V5)(#NLJ6AEr7=7$m~V?JpUM zow6U0@jo>XM?OQEwG0Iv{nQpmPQHFh^NHiBdNkO0jpl7q(RwB9t>oQnsHKa=)k?Vl z7^rY1M1F6em+75>^Now_yRQ$3@z~IxZ=1o;=cQyg?*|Jd0kEgvi`Y)GrAid{G^9Qr z0Q9+V)YN4pM$ zQ#u3tjSCX;>~79M*z?`Dvoh=8&^j3hv ze@q_L#b@`1*L`7MO-X#W!DF@rtX)VU`@yOEI7f3IkT*^T72Wr{)%?-_nKgMCpPHw0Ow? z*n>oRG8rp#nOJB>j(YExj#ok{S~kIlrVx+5_O@&6CcQ6fq=h1jbgG2$^%*AzdexX=iPKl7%rwmqCzZC&qbAKFSmf_M8rx<0rr@>%y zoiK-9*`57})_tcl{y=3jla+Xzf*~W1hz`2MSg9>=lI=rt_khyd`BOWfZ|WmIm=q^1 zDGg}`WAziJt6UsbDrZo7m5~Ak1)^XdH22Nq)d>Tyym0t*d((B5(Mz*C z*E4ZSPww}?4ZHOhbyI%%VJ$)MV-ft#5Y7G7^w!ja!KXjy`UWjtRWlw8d}E4()3t$H zG8ZM*G2;;HyZkri8B1-SW+^pWo>m!alN+1%UcC?fPF;HHuUtn*2Bds`?60}MS56C3 zVri7Ms!&L-h*n$>);0Ab)uq2;;zxRGD(<#M0PMQ{JR{Y9Yn_^$CV$gkg`^{0jj1?0 z^BA57QfZ5oBJ#AIbIykP?PevTM*8M0Mnzob)Hjjv35t6w+WR+nl4TGobobY2y0I@% zT@99Sj-P|igyd=D0MPU%87lCZQ9D3QWaoXj_mgNw3SJXc)uEeV1^iid0ZzZ*+(-lQ ziBFsnUgHc@Hnz9~`Xtj~M_%;t3nBTJKgn2xMC7e;;xqa|d{@&uu-1%iuNdDfVf!xKfy$|I@$bpoa^V<7Ld0uR~ zPkt!0ql-O^L^PrMvy+c&1F{x(%#UooESv$<{NStjK|o)@nQ#JZkYce(%K#x-$(HIe90zILBRb zT<*1L9E|h-ai74C)3O9$61xqr;!*u_wL~ z+InG3ML5LR4FUvg!oFa^#d{1G#&-?8rqsc>Ru%2vkwUC^!txcn!cH($Nn+<2X`Yr4 z;d_Hg2$t@(Cpo;G1Tr9kFeBpKK|=Cw)OL!5=@6vrZ9D?2NsUKM-Dccb1nAp)X%bTL zCpnX%Y4}&o?-gY5%?28-a0jq}R?^~hK)@nkLWXENJmZ40gdI$MM)`zm0?=3bfH)^c z{>WNE8CNPb6qnEs?&n_oJR1k;Z&Y=Q^-G8@6C63B^h;33rE#9F^;UJS@|pH3ZCFmf z-tGZp*QBzNUaY-Gp9}Xip1d2=i>~63!TT?XLf64|S+eGtLoRjUZegd>Wwk?$!!GEzu30Wj+oSg{Ti@IA+-`vIRT9v-%|DMVPj@k!;y^+90Ut zsSp4vDn@tUEAJ`vh7S6_CkN}6HgN(moOB{}?$Pz%;%gI$@q#A3Kl=jkbOIc;Gwp20 zb@9v2bcmn7EU8U;cg&a@$iEi&t+fc+7(IFT=XQbc6h83q{t$86NiMMZEO;>dbdqu|JZ5GlsydJ1*V5C#+dUxT@0x%vr)3l2|E!43Vd8)`59{@)y)1dexqIAZEdRbSVI zL!AF^j;pHY@zQBSR@b>SRif%M z=W}c8A|x^Td$ED6D78E)L3-5EQlUk{fG~^B@5kn9?t{M119)at?~oACHy(DiqSxR> zc91IZ!+qInCaT8uki&nyHhQ8#w{wck2KyP@1TBI&^O*e=RCIWeLRt|WaW@@b7V_On zFwQ-3Q!O<=#cjz&bG>7!SFuo#@sulXKOHsmrWIa03!^j-ofFYm7-r8$!!GKhQ?d#F z_8V)8tvoht_ZC9<6Xh!J6cw=NQDol?dkJ{VF2;xZWvyca_w!m4Ec(gBEyjLY>i<~b z@v~BEGYhsJ4lRl~>3%t}A5l39Q1sg|nZnL5{7$?>*~RL6ho1p=W7JE>RRQEmBU~O# z<`o$Ry4rkGtHDo;bBqwd>DkJ zN{I~~N>94~zN5XRJ?P$X)hi*3r?E5w9NWELTiS^NdT(r2Rn=Fj$)6`718-&MR?5k= z-%Y=&o3NEQ0`WfJL`nX)64=tYC{|aNxdQs9fcW>tW6OK4UgfWNu2p0;Q6UiVFMh!<$#y zX}!pTxJ`OLh|@5e^x}&f4n9TZO;6(=bBZuqTy47@Gq}dXa)WtanUwt@&s8!m78yHc zvVdWE<@dKaXyL;|#2E^XIha}cFE&@GsG>MnuCl0Y+>V)hH{QLS5Y_z^XO_ z`{q|NqBWr*<>2jv&-iRAhL1$3y&$|HgnQL$9m|##KX}R+Z`m)uc>~&;^*lmV2>czc zDpYsPOt_to_ zWX;ojs>^59s8jlfa|$g$^@zcwUM$ok_bT1)Ft-V1ofpX~sqIC- z`%&xyp{DhMOktDliRAKa0<)Bh@?9&7qzwxh@yx_o@spVp@Kc0=rgm@CRj8_5KRw`p zvNZLwE<4^XRG4d3tQ19ng45!aWJI{K!IYL=42q}gg}fd}rRUyNVv@0q5?8_#_R z_oD*1DeIPW=-2IzE`3O%O!Lq7Obka_p&UQ^q8!cH@|w?=c#0mEY-+ZPvC(MU&a2Vx zk)kCLJYf)ax3r&w-ZTe1!bebkJV~iX=YKj2YQE zi=dSHLs=wP;lG};=6MEOoQ%6hU+HGfbXr7Nqa0HSqK-at#fYHRi$qY#5fyR88X3iR z&4L)poe(V`K_OL39cLQcvE-ZnD>O&ZMEtl@p}?X%8T z`vvBksqwG&pCcRY9uGe8k(Al0^6Zl`^r~WSzsdKoy^Y37vv<}~8X|O!Tvfk84}bv6Z?#@FO`58Y z=mzTBBP_p})HAP{01#~LH=_X1<1kB^qdS1|@=3k_fdg1=rOZI=AJACKX5eX-Y1-XS z!GhV|s-G#MaT-w?fDzsHr%ZirD96x@cMEro+kq2kMVT zblPzCb~eHYvbB#E(}l%|2zO`GC(Oa)@B{^TUdHHdC)-%X)QGi-PMTgqic5zo<>rS$(^QkWn@FWB2F&g6vrz00|?3|^%Z9^@qn z+`clhx5%;@SKZ&GBL}5YlNSXJ{ofU^8m!1=7Ze1?fn4@PLEdp9ld~G>;C4R9Gc$42 zhW)1RLTFZY?rty(GX89~u3`4T_9C;X=tv@Cq+5#I*EIjibPOZvG&ECrO_&X^vu5P= zW%7k)h19zi7!4eBvUtjB4dF5(*WF0GqDV<@JsO-&5=jYh$Tx3i4#=qF7V-*+su#g4S=Dj1M%KJkdsn)M13ZYg@47Yn3WB0N)ltX8?q~{>bE|~Tv_yAEO z=oEz78S`LnTj=8*tH?BMB2C9p4Uq)DH))5k-te$iq?E{`n;P=Hf3Q3?#7m)=I634H zQl5U^PrvG~FEUMYak8-VET&`B6PrET#78k(_p&W*+TH3RRf#7GwL>MF%l!y9epDn7 zEr1&}epF;F=_Ahs0iSGAKocIKJhzQIh6C@nQIrMRk;(HW-%>Za8-BM9;y}4{{ZMXr zI=e=CLL?wslpL{B*mJPM_-KOb^GkobMe|1?>BlmA?jrjose$>uI-L0(7GL|n4;g<{ zTt9PCN6O1=i(jUWrs~LGHq(Ac`DBDTUVPw*)Sy|1Gv3?3ilg!lE*aPmRdHQPczgUb zc3e!!C_E4P^Bb*}yDMurt354ZfwPT}Wu*%pn*NlNIQ+Ui>bY<9)m80@f0et5V{M0Y z)!YMi*XvX1$<3pl1{MSEPN=xzVnm+@ihgG9^~*P97yHYj>-~u(yp{3(xx3AaD{s`X zp{h`NHX88L#+D_K>G?RZ{Ds8$vs;RFBjXkW5=Z=Ud?r&j6kCLOct&5M-Qv$aiN_p- zOd$IgB?Z`qL(z}YyRXX|7qA=wJunK)`SJG~47+%26pOFTh8cYystfQ$QKZA)xMRlK z^u;^k-e9EQo+wQZOh=&k?8zD>jqt|pz$3SZGtm`U-5Hh75-Mz7_papbBE@@RCWq?i zc00`@d2FgmVVusI!^baT;V__H#WkkW6y(GrDQXp}7Iqe&SL`v7;{k3RqM3`%YaQ~v zu5O~@fqBNuXr)L=_WHJ@PsvdpzDzY66FcU(y)JziU}q-&sd~fwQReIoavnPQCOms2hfLrpO@G8NFZcK{d#!VFk3gTh2U`;zERM8EN(AXP~&@_Gvb-F`bq{B-((0v z;V(%^Lgn){8+C>-YVuEwBoNklt?{fR(SV>Rza-JCJ`4j8)|xhYLJ{&f{*wNjAf@c= z-C72f>b>)YYitpJDe`YLLs%`FAF)W-6Jgv%I3woC)?N?apuHzCoNHHTzG-x4E2T3^ z2^5a7NrQO28s@97wN4Txclq@~9SQu=*N7nCT?rg~>Vj#oAe6lNcs*yd=y{UC9y?)9 zBHNUZ!H=q6G6VQ&0cVO_MnP<(7pwq&xle;&nW2_Qs!QCm)kWQeR+^n}C~@bF%i3YB zo2W2e(NK;8iOV*D*^#E}jA?cjf7e!7T_TA+PL(f_Au^MHS}a)`NldeI4J&}9+Htbm z15C(1%^n~J!5Z|Rk_W+@Hx+mHN#_(HsDx#jq55!CL=H7im`#j^9F0a)T*j1AX)3~w z<5WR(|AsdN*$$7m_W_r{hKcMMVl6&f6M#4fTr9oDaE0KyZHJM(_H!%i(LF=ySkA&) zo^MxQ>$>}Yq-C;Ki-R~Y3)@Fn4Ku8O?fb+8MdUaNqg9R4ud)?x2;1}j3nFlB#1((( zQb$sxNygatu_J3^#tVe)`&QQN1r#HHr&QQ{AqZ5|aFg))yqh7GuWIDBtc2|Iz~nGN z9M}-!SAWHdZoL?0=*|$+m#3@McVBsQDb(n_su62Kg~F+Z5v#5IgoFQ;<<<3QcTCP- z5C`c$%@Y7F{C@&+oT(oHSPdf;2z~nf^^aD!5-PGT`|y^t z&``p@FK7KMrmJs=bz@x%hZr{jnlcl{1M^nj{o)q&M!`B+x9$*eC4$fgQ;x!5D~Mub zM!F&~#OQ#0sNP}NsFz9uQtlsPhkVO|ocwYVRm&WiG4%{0s@6!=7hZ&a^=quAL^<4uBtJ7BOmG& zvgeXhXCR9~e6U=}V0LKaq#&KKRAsmtCdxr!A$2Bk<18!U5ZCQUtSnt8cTW9W+ zH?V9m&YK*;1Qw~M@U=m}N0AjIcYvmH&cK~UHHMdhx2{4&O>r9UF)++qb%xFLSDeTX z_yf<0`?~;MOKnGJX&)Z#=!UGtcSpUTOqrFMkkP&Avg@YeH7ay;<-g*30_)j~Qx=4@3sle~sS@ zfU=$L&h_q&(|cE3F+JT{Q~JhLu@j*Jlc9P6AB}L3wkC?j^?(xz%>Z&Z-lG13>{>|heO76SEh=XAP zIc3qS<0I8!ba>3H+4@&##?_rnDYfX~jH=A!5ZcNLGA9-3iDKIFqGlSvSZF1`0CH;8 zY#-=hleo1H>;>6cKKkap^3l65irT7ruBPDre9Q3UPk#y_}}Z}XhqBV-JD##o2kWkt?Cq4&g99Rz~SXvGW>vXyrXpOg=^T6 zaAvAgN@diG4!3v~;@Qa~zAU8$$^QE$_K$C7a98Z4x{;)Jb5!hvq7x=l;+;UvDpcit zuf4b}tEL`AJfl#s(W@@}E9&c5N&L=_gp_8`P~ofFmWg#hPz%*|MOIV4Q{`gHN=6vN zM}fv@B|EeLdTl;asjTm>ZqOUXY$f&;%u!3%^f&EktCGGWxjD7qs`gJJt;gqww9sy;K;7-Y-7dAl$|~A`ya|$BZo;Qkv^2 zvJ+?0k9S_*1CDdQQPkwpV$jIU=0=K)h&|=HO-cMwnCkpuuaT8Z@VMoHcvkz7__3m)#eV#Q-rE{TF8rj&1#vF0!JJBIUFX_h6t!acmjA&f7;xIbh z(Pet*N=Dqk-7Hgu@VuVUrU!D3;J;F($E>lfd=MUo7=nh)lWgyA6I&0xl*9u1$O4i$^vu4lzzll+}X!IN%VbO(# zCj%0k8e|5#YXGXf6Ofp>Xy9b`wa%MBU#KUvG&x&-Dg&-Ik^vvP2_%o^_;Xzw zo(Mk_<^RE<>xksgEwITk2o=q>-F%mmDGbV?!G?bG!&UUOT{JmZLPx=rnWFp6&;0k^u_$-nS`AbAFf zbacC@Uj(tWUjOP=U3mD~Q~qW|_!@|x`TNS^##42bz2(JPg1EhBLJb=K08s+Tzm_V3 z>P?;?bV{|f+dufY?Z?*T8)yh5L-HoMaxG;WYP|9ZBgRQLz^3c?S*JA3%%q;;*QGkP zF+3pUyvf13S}z+$(?`wk&|>ckY4oO3k0^TLLAs%f-_S+JCjVMCfi`??uPo1lG}J`-dY1r<7PD4)W#n-g5vTu`dpg>Q4dGGRS);kHOYZ{o3zE zM+)sVre!88%eEdTs(&lOCd?Qy;D)_umTLEibPJTN=>MW=UUcJLC3ZaH06 z@X~dPI`1kTlX}I>dH?93r_Jsl%jXlhXiLS}4$rHo&N}ix6 zx~@a_bxq!c)_`(Nk}j(G(^=i>ye7?)1Rku!$JRcH;hBaif!%UK8?@3@JulMrYhM`Y zj6Lkc_=b8Jd2N(9pQgnNR*xI?k9NOot=L|i_jS0pJyA6t;Ie3Ka_lT zKvPe%sEA0DULw8sW~A33RWvk_Dm8SZ3R0CwhlDC9p*N)nQl$xqKq3kOg7hLyrB|s2 zlDzQy?tS;Z`}#j;XJ%(+H+yzxXR@=WL+HV?{^G!o&TcNC6cd!WG)bDtU6e2ew4%zM zEKZJ#8F**CQvxkcRJrlY;JhvD#aXI^x~J3Hg#6dw=F>YR&?`#k+&`4~CFr^L4&M7^ ze_+vPnKF`cXUU%X;Q!Lx_LnnDw!A)z<+GarSW-6b=w4P3^svh9i6&t?KT7;H*Y=8h z>}yAWA3s8CN~+2;8%e*!e5eB{SR`Pp3b-ojK{naIi2!& zm~>tYMMB^0PhTs6)+>wb*j&Mp_TLJ>c|^xNT5r=Ocke<)%U0lBn@7J^?MiR*5OeLy z)smWsjb9->Eu*N+Sp45_FCAbr11G;9GuyNeA%t{nQ7o+wZPhFTx-!Wk`r^fFed`r9 zR>W&P{ny)I)wCBfwJY~xw23qORf$gD6r- zpQCM26Qr~8)ouLC*e5JgtA>y}wJW*QAwT)wwdtk4k!&={S|(`P-Ne(4({_ZaH=NeP zEomd@Izks61Kj|A%9>|a+CqzZ7bqZrJ?$%Zf)8v+C>r6doYd@C@;cXed{p!&Gs24) zfE7XW&SDkxa}RYb)c>&2uG87!|9$V!g<1btWR`7(TUx<3dQgXzkmpYiEY99RoN|d4 zotj;o)o<<#I^?GG7xbw-q$gyT9ECivZ{jfnf9@Xu1527yZ=sMZ+z=QgV3*wte30#? zZ0`i4?EfuihE9Q?94w+CE1`?GGsMcR0`Ab2Bj_L5As{P^(5Da379nMbUw$xy$O?p9 z_sIL6=SNRtt{N3}$p4BqD!a!IznZ?g%Y2v>W{}3%AB%5yJa_}ml>&wXCG2i=wE}I` z4rsr>Y7(tKo___gc$*zuF@{LA;d;^$DlvoN@F$=ibTJ_y!GezxUVSb7JQ-*S3sWG$ z_TPXhf1ZI`kL~IKJyA=6vff=#Yi=UZs_a<1&aTUj74*Kt6n*_$E(VkV2BK8v;bEjj z0T00Z>pWC2uUmSe>x$a^0*TzYt-9+xqB%rf^=1D9w78fdoVBoZ%;iQ`7`kF^<4lIoy*|5vzAuV z6dHK2`vx&56(Ce|@TNeWIo+23ajZMh8ZdzH`uf(<9Gwh0Z@O4&pU%GQ%=$A5XVWNJ zm7f$1dXV?-$YAhPK=epioA7xg@B#RVpyU?2n#2PCCuegWpT4|S3x@X`K}PF5YFfl` zf13>lupe-haCv<%Oe;!!W588=M09c^^D`kY%l1ZTdf1{|LXJR2KZ#qPVE>CuRd+@9 z6!mpq<|ZA0KGlmi&JbOFFSq!(#$#j3&-x-L!C9&=eeeuv_uwu;8d?wxPhGM)40B3B@6rwHSRDW=g%1YZ-s0BJdDUyfW-G& zGy5!IRm93yLts^Z^B`XMa>Oj*3Q7=ahK}AUQWJIKJ4*Zibgu|b%wu$2h>A#?##|7|o!5Y zyz)FzM3hZ>fmUbtO9R}J%YvgLviE7($4!&^v8cv;OvASXm7>%X{RUo< z5p?kE4I(4t1ae*E8oNo&Lb~R=1h)y1<(2o_NpVXn1-ea}?7;^Lj$FY93-Pl~fI5v9 z79uVJtLXDT@K-f@CNi8Ie*pO%Qw=vMi)T~$J%^=M}AR7`o6DXdjL6Eb4xL zgPg;FtVCqVu;icYz1&*xVZY>MtskKi@KN)Xf2fY=etnTRy+cUs7p(fOhyhJ|e9E&C zKzzw6)RLl1!cK@$G6%k)^&y3Y`Q7){T1RmbcnkAe;11J&|MGq~WKY8Gpm>T3D*aY< zaP0b%G+(l5k~SJgvvNFJh|%-gw&G~vf!bD2X`J>0!Y?>v^*iXHT>k5N{9izg!X)Si zYUgWX1D>*<0flv!DioRonZ;_o!DB$``zO8-9O>OD*zw5NYk!z4u8CQq5ziJ^5Pv)p z7#eLw81H97V%^zOL!$8;X+OIdw|F@BGOY+;CTeLX+#y9LQG1crNj*M?YNikCE|3~B z32O3>5xzsXfuiw3W6Ar^eJ4TPD^%8mi$pSOf_Hx{Q61=pQx){nv!Q;Sna#gbpkMN) z3O2$6T4t(DuJYpkrQS3-qSSwWfwLwmO_CZdItpb?3Hp5mWz*j*8-(dMgv@d_Jl*xa z&|~ObEGTP-phKK=6$Ix@ym7AE8EA#_dnXnoh~pVJfnJd*z^55vL7p}(1OaZop=yHm zcS4CbK}4@c6vvY=NT9qt$=|Tvp8{9=GJ+VeYkea7d|+m;ZuV;>r!s>M3*A&{7=|*~eKjqIG+h<|MemV{ElQ}c zSfQHeESE|(U0lbWpYD(}2uLLF%c3OzCfThQmAXLN`lVe=U9L^#r`4xcp0#>KwyZe^ zj;8-AZPr*kx8}I1-27>}wSFwT>t4;%o~XA!KIJ+2x;FFJ&YWXDGQ2*LY^>uto9apN zbIvwuf7+ZybZGy$J`Ukm-70!`*P|Tqa}p-u4708)Z@XOuz4{gp{9OVyT9@5SQ*$s{ zFH&cT_r-)FY4s`eq+j-ktr)h)`^h|XqFvD(o{c%Yse zFkrO5)$4a|xi2r#Q={^?XopoF`ZxT!{xHB*5dO+NfMv?uR?gkjwr9Yu@di#HvZQk5 zV=LK95j_@}M?v6mbsZa>ibbfpO)JpFRFj~| zAa!rvO0JUqfz-#n@x14wEyljEztvVY55Q&$sByIi;Frvq+vluLTauH6rSqw$8|r_i z&tunK?!K<{dx;gfw^@YgJT^{GSwH4q?_OAeYoMQ4p?VUNAf_HkYl7OfSG`zwxnQC_ z@ZpQFA|PX8n(#T4w7xREOmWEQS5}1lXda@)03n37iewsiw~CA}*}yT|PPisE&UphM zPnex4SQp5GddPeTf=aI@tJJMrQSOSc9}|;Tp)PsRZZddYNSbUj`Lklcptbg-=uu4cOE28Xm+EkvqW0x5^UHCGpP)w|8!i?N67_P`)64XAHIP9_g%iL} za3N+Im3O5%I22b+XA*ZE^4rgywLvktyK3yt_=?Xx67x8Cb{I$WtzrLz0mSazhj63$ z+gv9y9ii_#k{kl4+d`p^{*&>F(G^NIi9c}ruN~*A?3#wMdc=dd@2%%c+h+};sE=1x zNYvS+xz!NK=((W+*10zuI&)vLQ;q~sH6jnU0YY3TpVsR|ie^;R(@_7>8pz0UsWqyw zFN9580XYjw`WgdIcq!k98lwU1eOfyD(qnK6Rcy|U`%({k|31#0>}|JKTc|lC(+ag9 z5$p(IQRS`xR8Nl_lqcGBdc_=+w>{>zRm-3U?(DXY%0;{^4A6s4j|1nwp*@FGP8gO{ZO81FzQs zYY3_~)~0jn{?fjd)WR6tiK2{E#?9rO)&9jENRoTTE*yfxPxGv9{Z|@!PSGH0Y~HrhWMf6h05EbLAoJOzp|0oa$3W@w zY0u7g({kA`n)E>6(<+GAlr5S<88BwaLXDO%m;9XzXUnM{^FOd6+!VK1FN!4JkT*fg zSFYAvg@tV}*CvRqlGxIjq3cK+kz_#B#Z?X#Gjw}sB?5vs87*%YEq7pabqZupY6NXW z8f3){c?u5TOU^}1Sv4v$L?aAOElVX0V4eqbm-$Unk);ShvGO=E;UhX1GtMoZPA&&G z9O>gTN#d67+~3-SXOP%^M2izJl9}y79Md5e$tvK0EZ?yw!UODqz^c!IKoTSkfG$>d zlSll_7Tw%3--UP)L{#$c3m>cq=r1%7 z+!%vmrQ*?uy#M}j2os!-fneTwTHfHOt(`a3?W0rErEN%&T{k769>+$kGM!*UW{N z$#(-%B@ob<#SM+dS`0H&7lqQj2fi0vou?$Fdb2x^*u#MaHD=7rhcWu73}$jzyQ}0i zJ~Q+a`Uo&Eudl=z)?OQ4EvRaSrcru;fu}PhlUE$fIi1e+rFb{}%3D6^Dr#;)Ox{9m z3kH)wV1Eljue=8Xr;?tF4-PX23ZycYBUp7Xw@}G_96@X)kS%E@{uD-q`Kvx+n|-w9Ee*YNFxto@s$S^)E-I)Djv${qDYa7D@W{hQN()7^5^X= zvUdS~G{l1qXy|~A=$)2`HCfKZX%Rjz-M5-O&Z}zPNBEa+OKFM;n6f?^c`7(?(!b1NuPStB zy{J_Vjk48lKN|V`;pzjl`l>>;nQ9EAHua4yR)pvfzXhGveW?x#X374MQT|BO=<=}W zM%3&ie5d&mR;8^UuhZ>hOw~N-e`r*3NyV?Y!@KudmlTW_a;#xFDz5n2(mV0prg0^Y_ zs+>}8pHY3>aw3sB^`e_8Z+#)L7n7(a0Qh4DH=|G(dF~(uU^@}a!;7d$H5lsor5v2Tes<@3(w5l z<#UMU1Yq}RTU@2>szgYQw!R{5${Q0C^!uMq`#+X+tZxlQ<%?3&Miw%;2E2=b3>#h} z2X*GGL89OJ7EuF?YrGpzKyG7KQIEUiH(K?pLQ(M(foA9vUu)*Sr3m*~51+7&?pZ2a zCcB}*L=L2PNfpW5RAv>^oREcRqlACct&DC@X z{U=v<8T_y*^zOq!S$MB1D#+FJ!zmp~?mld06vQlR@&K(vFhvI?+{-dgY7{rM5E^)= zWBp3Bn$gv~wC2BCq5oy=y)2yZy{xBYh`AW&2xgGQ>7NZA$vERIT~;!=Bn)Wwtnuj) zT#LZ0V#-PkTo=CA&hbGOO=_%x z7C#!8z_p%8nN~VmcB@l?pM&VB;wt6Oo`G{`wTxmGaT&8>z^{@-^4q$c+d1nuv zfLHxHk=g9u{o4N-IE2C=tTlkK?FG0s>ak3;K91BatOB2O(g5fw!;^TS-$K34rr}Gb zgeAtkaS%}Bg#XuDsG86~@cqCs5dLn~PUdPY^ddu4RcXkOa3(;UT{cFzpB$*5jUrVs z;Vm9`T^B{+k#VUNFdN@+smXwFL{8IXMzDIet^4&pA@p9Bl)sEyVNxU4kE1lA51`ty z!P9MmmJV6xA&P!Ff#568iDHjsR`))p{X?)HeNGg$lyTE>I!Y6Pu`QPXnoO8wpL~p9 zjk^QP&<&8&+~oO@iKqNi3cVNl9pnJdX03vPxsQxJK1BeBq8<*I2q-dfL_t4>X z%{Tz0%5+RuTrrk|avjL2tfT8bNV8rqa{kB~D_(19uW8QquI*pPId6}ue}oLj-W6Bh z7ux9)DZh3iBfE~a3kSyGt5h`!yEqj(3S8jd57r}>zdCB<;gTek$D8rwt&hZ|j1UbP zTe&x0cWqAppWtnirehwB9Zd}Z%XRf&ToqIs+a5$W{N#l{-tKF=v&$T4#c>456>ImX zDjb%C*?We+K$l_y-Ar{V+8aeb*jh7Z$--BbcZ(5g+BFhJdkSVJI(&hJOmV3>3*rVR z)+bPWG28LEkWEu{8*I%E#o}ye$akA-`k6b6rPZ;8!EN~9YXSIU|F@W10#LMq0{YL?ZcXQ%~$8FUL1Nq?Dd1vkLetl-?fPcX_27XU^P&rrQFQ| z<-vfz#t&nL#P)7J!>6Nk)J)guF4zj!q8AxKl+{WH1~QR#VqFAshmD8JqXCsGjK+_@ zpB|c`gbF8%-;tW2$VhH4rHLG~vs=H{BA_dO^sB?+F06pxUsr6bg8E}Llan}(i@kgz z9wWODF>$f9N(0PIO`(;kKos(8RPiQd13O;cKRAd5a5P~C`4uR;T`%V~T$6dl^e!0K3V(F0r(_0BM>T_iC7R_yi+?{=5dy|uim68; zd?vfx0s>epMGiSok{HmB*l8P3a~cNpdcSho^SLTD(R;3R$R5%nXns>89Ir=kdr zog<<01+!QDY{$4a8~pmtr}KTv1Xkkk40I@cAH`vMEV+$Dx{6d8Trjdnxih%P_> zmt`aXOMY*4!VL5b6U5HB)_o$g(*R{MYH;Z0>xw5EKL5VVg#kGl7)!?BqZY0ZU^`U$6c<|)vL-r94!VOs z!$=5>Sba$f{PC}Q0pXtUgfUukvv>k21p7mLokx^^zwXz4EBXtG&(Q#>{W`CtQ2Wo_ zv3LgRXz259i+SE1)BQS@wmXMR%5ZA6VGQUpX+?AN1=O(Wx-Kbx;+b>{_kNv22%qWF z^psy+_1TedK*9(0B?pdxgkbb@c5~Y+Pte_RCD4`q@5GQ{Ike8@VL(r-au0FmF^rf5 ztt`5qa!-OVQXD zS6k@AWpJD<%kV=bv;PY~s17LC$C)YRZx|4msocZAgW`H~h?aKQ-;EnvVWKpai5FuY zK6(SCN|06GDAM`~UN7R>n?}-jDKxJ#4c<*<9bj^=MHusdswOyt%+v(hug9)^&bdi% zul?wJQ;Z9%FA)1+9;Wlo*PA`PtmVIq3^Y-MyIY-}?xiYx5{L<>$Qcz}*{imwREj&gO1k3-hq%QVP`F^}ul| zWtR7!=EBO+I2C_4o|HDRCR%KP`cE{+{n4zy-o;(JeIh8b60tzCmp|2)z>QONW0s5b z;HgJ8Q%4n2e%Z+ZaSwjQ$yVshN+6jw$fp1$dNi^@f9R$1NZlbHOeSvWr)Q*Oyku}u z9o^=V9atteoHpj8>ytWJgsF*-Q3<_y+2ilahB3Kw!y$}|0ik86V7`E| zJp4Cm_uX-h8+3ws7|=MxcW3z5`Zp|j!fWz;gP-D)VFN9}FV0}s%iwdqv=?LG0;Q}K zwLkus{m0H|>mn`RxczzI-x*H)0g+H|b#8xQGIJk0F`wI}9!(>?>m602M32lcr@4W! z7A5;A*XJ#A*NA|XRx|6S*(nu=z5+cuTUJ9Z4xaCpO&alCC z#hFQw+$O<6UE&&>{@{r~no?$sz$VRiPt$Dj?}N@C(&M5bRs_cj6MLm5QIyd=fF-L< zyW!Ot`g^0l?9}d1yq)fF6DvhA+%pZitu|><^np!v=H$o2d$T9L`%g-9YM{5Y$PIIj z7f=TpRkycz&J^}0s*jbQk6LV<9DcOvlG8GC!&QY+&f&fVZdIHp3*s4CIxTsWm}ajV z=H%&X%@vCU9;w$uFj;@=$}tKa1KSOBt>9%I>%fa$K^@gXsWgFITSd_(=#I^zt0v$*Y{iv-Dk=V{)B{=* z&|Z3{@+FN=$$5-;A2o&0zfbFbgJI5GcrX9#-TS_Q!dwifI&qSnwAUxH7@NGa*7o#UCBm&X{n9Z6mKO-C4GI^7329`)~Qlz*Trv;P`4_F&N2lD2F6Q**s=Z@94O$p4=5U&pb$mW3(niZsFag(R|n33H#ntKSzaP#m@`u&kKF?CQ@oAk#scT% z<8uryx*xVH88L$RV#GCA)oAUEP(QOV(ZF9gf$H%jg~z0*@w@9q$!;h2jMFh7n%T_v z%3D0!dc+|6^0*vd@vovBv4FcA=Rcs!+hC&qe`#ZX7s0^pu*nwHf-gGb;ZMjIwuV@A|nr~!I z1b<{v*U;e4^{#=kLlh4HC{B!FM4inQ>o|PL%z?I7`VMh1Qt*LH(F}bb0u%OGa>Rmu z&*N`4x+;^$R;UJv4>L3El`E7Z{F0!rxDJcjWqhuhdJt_P*6k0D)$0p$)Ad+VGjwFQ z#Vd?p3`A@0LTN9c>DBUY5h?k(v)d93QgD*x(ANkE?HAQAwP9+!wio=IKL0+!u99y?)ZlWeDK!x{zjaz90}Q^qm5#rVFaSC zx;sro0eRb!m&(?_pG z?2q7yi-66EYr~S7&R`^M#806EAd7|3EpG*V1THo=s&7z~L1c04{Roj4HE(*a60Ct~@#L^l71B}+fQ&oBJv zOa9u0YAPXl^at(xg;dI1nXsgLL@TEgdJOe{w>Cydx_f$4>ik03&`i^e#E?vm1@rL# z3=9zhHOGik|0d?hcb(;m2QgG4yP9i%Y;X%XmOl%L%=8rjb~120Dj|L6>>%{5;yn)*-~-!+wPw_b#; z;%no&6zINK;|}cITybeFQmqkkFIc=AJPBgrqsNiUeY5}gOGGXve5)k0@cD~XBwb3e z$t36}U&O^#xj1~1&AquVp3LmC_{F202y+3?2N9Q!8&RhYi-dPE)F`nIk#*D9as;@7*Do z5Jzy;ueb^~a>ekEB1RBOYA8;Rv0WqcU?er11ZBQ_R`|-`iG^?&;+27yMR{)ymAiO; zmH&SQ{&q*pg6h%6lElt)+p-(<`%ZO79-IX~EAJT^bU%^lcfVar2;8hdO-J1JweJX8 z*3#`1y{rQ8b7|@DT&c7#h@Ey@Dym}psyH2#@Rb7$R7MrCQ5PQLbh`pS* z7Gb_EoGdQ`16ntxFN(sm-2)@_dQ7}~z(iGB%x)m(&g&1v%}fe!P`ym+fV z7q)JEw^P8?$E_@hW;ujAwxE?VaxH2I_i@_HuTS_Lgh}L6yH#wO(jiK&PAT(vWbl3S zYuTrbR5)!d!Uq<@YbYBr3M%-FB_H`FJ_(x6CohP>Gwj)CWe~m9n~juPJnOHY)IjTl z&VT=Co>`%p^I}q5|M^<>arzF`T-g3`cvC^b*p}!B>}v9?$|vIM(H{HLcxa2GW$yN?eSyw^3gPWQ|LyNG{jk zZ}8bn*hKueQX{Q{@X<(J-h&c@CvR%Hd|Fbd3>dNQu()7JM_kOv4QJwG9D`-;_nwvpdo4E67fxNBXeW;SU>c$Om^tO#8KE8?6_sE32bKyKatLU@LQ zHi|a!JvkqdIith^BXfPYhL2UwU)WZy>q0q{oZM*YW+mYI>XSgGP@)Pe| z7$6|Yd(EyFW~A!ZI#J6J%zX}uFCZtZ>AZ^QJOhcc?Qas&kD#J+1cuf=IDs^3ed6+x zyh#f|z&AmWKu$IKpOP=uQ3OxP$%B8F=mEZNC5*AE4CMVA3weU&0-lamW)DwiKuBoB z@o?INorC(2u*;N-?7*jngCWY*#7SUg+H?GR;4BSL@clUnv|0-C~T_(uERdR%{6uHH*x zY<5Tg+4ZK|5nR1<2n?vt&voUw{d!Sjim>Az1J>Q;Aaq;3C}duZFB0c?@b;TEio>@J z-#Y#Fr?8vC-3~~pJe^5d{0T?4andUepW)7e0ErijUr7iRTF#YRg|FDi9ByFUv9kNC z>OaA|BOh+!a>6_HEM9yFOKEwfw(sY-&o$vhHf+{Ga$HY+H7$JDt(t&)nt$ zxh$MY8h3_6Y!oS|-u#F|8VdN5CfI|qn?n$mhPuxBot>rYWm?GZi=V2W&c@R^oInTq z05JsF4pQ`Aqdfgrd!a0}E6YAjs`gvdHu|D#DxX*+`Uo@19I=e}kzVO(hwDWk=KPoD zwl$JRBQ@m$iVO_|GlmwrbC6wev1~p*4fR+#dyruL<+p8~xb=O;jn6r%8WAlCdNlf* z`%n{&r+$MYGG2zd{2i)XXt!qa^cz|vpvcq{8q_7LtwO%dH)r)L#$ zPns=Qu@zei4a3(?p!KOT*a`$Wjx0tZ1PiNQBg%lU$=C3t8S#2ea^C$s>`xo=gD#c4sT!m(W`W*Zvy_3 zGgt2n9KK*%WBZA|KAp+tZq$F^>ob5Q{}>k_oO){VC8tQM_15q+ZsyE@RWS;_FF6_0 z4maI>dkY}>${e5uA$2Rl75MF8Z@w(IlTAXLXY5)aS7k5cnQ=$v|ypNo}Cg`HuhTr`d(TN zxgFtx!#ti&!4ghhDR&xqiIjpq<*{ziB@#^&&UJA5R$y;D5W4u7$D!6p$BA$C(Pb_X zJ>GV-+F?37ctUhs`e2YX!Cc?MFN>NSp@Ve0SH+m+=1{H1aGwa(#U1sEoo<_5*F9nZA zz|>^AO-Ttv%gRJoR5c`QPZOXHm`3*L1?C&7%~#n4vO!?o{-&0i;LwLBj5mNKGpw@6 zSblb*LTZ%=Mq#eN;#Z9{TRd66`y234+9#U%-sp(*rCq~t|UVK z;OZZr6}z#n+)xrS*#rJ(+XnkHNcet(NRK;{%=iAnYkkPUKSM~S?gm9QEp))0)WsP> z)NEKZjz>*tY=+PiL$qgu#wHj)W_JSlry=+@LGxaaBZ2(Eiu(yPeKhzrj;E5V57`&v zj0l}hPAIHu}NN46rS#yHQ)s5pQKN07nZY4So$1-1uYz>RTP4=@m0G;{}jvrU`Yp$jLV}D zR1opRD&Bts{uW&Yep{r^yr2PoBmAql#At!v_Wr@z8I7ABZZMNEL>RkZ8n);P==oW_ zjuu6_0sO2LsEr?oxkgo^%0HYI^Nzzhem@^sM%zvPtkBg$!u+_NZCtcJdWy}orc2q# zfHL5o;CJ*k)TdlkA`$`?|Fyg$fL6Wl`e^zlimw9fY6pto|2?h|yx zqSUTd=zFJI2fRUk@qNe`nB)YtaA6zuxGIS-cWMf)L`1ITjgJ$o{3j1}OrY@yMc#Ne zh~A3eE#B27qYYX}V$RhZKTUM*%A-6I2<@iWZZB0lqFZx4Rp|pPwT360Zyz$V1qkxF z5v?0f`+|MdF#~kKR+!~afy`R#f|5R0h1>+Itbv+cV8MPfCUcx6C`CDkl zJ>GlyG#lA8&Uo3S^XhE>S~rnUVW54_6e2!I%e1$orzwUAZIybA{C?};3u-x5$_ym` zX1CGTpHwlWy5r)PQNz)3dborU;f&#Rtz~sZOY!p;ZR_S;p1O5KUa>ZoO^K+~ry07E zwCNx#q%X5XA2N?Vjokf*f8w+YA}U2#VLtSbO8TQtnIW5^wDWJhln2ep93;#xO#H94 z#K*z)DmNjYNGN-S7lJCS6%K4dn7UV31FeuSGR68NMePe`rD-G_&<#0t%uwgqYNCI$ zJbiA>Ds^{$x@S|!>V2Kn?EQN7*TK{ls_kZepK0edc&bqI@dfYOqjl>~>yxG^AC0uZ zZU9jOG}B1ZvcUSJkZTf6Nv%v0i`>EH>)WG11zHREjofje4yHR7g>i{e0e;kie<&Hw z*W1MJqf%J;S&8p{l*WO&N~M^QRz`X6$FJN|qm4R^{UdL6`7swRtKT_v=#R}p+ThRn ztPMU+%?*h^ZsdOB59SVLXKna-uvDLjyDBFu59cdCn_M|1fdrF1Lq9@Z)d**5@GMke z2$CtFu)=7I^a|9q=Y$Fy3)@!$nCL(Xb^MdRdXUt|{E!2tLve^9H@MS8hM!e#tU6IX z?AC!64Fks&Tp(JHTZIA1y~+ZaRl5Tox-Sxq(iYJwiJp)=WTIi1dv%;9YqewJ*R2U| zb!BJ)}>r0s+ z(gzdBrZaNLN3a^ah_`%@h=^DzBqRZ{q&2+Piz0e%%pm({*uM%ds5-`bub9WczX&f3 zrb~FQ*r95J-0RuxZ|>c&=%WeGhPPR02?Q`qRb%ejAJM&ND}hdiPlA5cLQRm(+}T70 zRvsJ*2b-F6^+~PlbBuYRIib!+K&$s3ryQ*<8Kzo4Iz+qv0C)CXmUg>k=5(~Or-0HdJum`9ETjGuB~=J z40|90AocVIu{RR6>F~BlWV9`^nd_Vx5U|rPD+F)*^vv)nRv20{;mi(mI$j z@gge3t%69~GXi6I){%nUA&{5ds>EHS0P#izXf9H|iF{aqc*kV+AD#G5#Q(j&^gjZG zYTokPq$)SayBvg<@V|kV>NdU+ObZZ%!2*Od!QK2rOYn=9kl*_g&v93P|4P=*IM%*z z-}$_n7I$(yeHWw&RwVQ@a%M9g(Y_RYrcT_kEaO+QNB)u zexgr`H-6sf*g$Baqi~-N6^$Ok6c08 zCj3+6bt)f+L?M+EoU^cgbB(&%4@Qmag{#k>l1GK>|_8i3_Pa_|1_bKE(}s zziDUDTyM82{o3f;jRw2O4Sm@muiG6l7!c;}3k10(#8nfWBJe_z1lVsG+e(@^;~`fF z&_wS9?$ZXl{QSz#!ql_=sp!#~=iC=3$Zwjm`_s40d?}n-pGy&0$RTwy#$(n79kK=u&*ODKtYF(ZT#z%MqR z*_QjEo+-n?Lt*QDrKf+cc3h{Bq$j4TPlJpJiq_>)Q<00-}d zjY`khFC~PvcKjJp+Vmns73^LKnWD3h{$)W}s9+G03|2}E7Bm&?zN-GkfVsk?fcQKQ zh$FTj#QIARX51h_)Gc?-gvhx%^x|w-5H|IE9JGKd%KrZ6J+xCKpzGypKOED#IpN0r z!4l|eY$bF*bs0ryI9n&go^KMYSVN-`=Q5lv694;SfrVEPU&a0t(a4(J-(w&vkI2Qj zSn>)>KC#u4(uO_(Rz^`bIfg$jg!zOK=9iGk z-S<*ZwjsFoFc0g;-{M<5&&vIo4<9C3{449o!);k@CV{@~ych$byS##Ay`ex07)OEm1I1V9e9;V#( zRX#TMzh>x#!jgZP4VVN?JLJ~md+@h&j_JgA%2c%pOQCzuvyRi+3U7=JxpQ@ZG~L&d z{q?tub02-RUsaw2VK_#BDCBqd90;Uh<1;f89a^OvajR-+<<*qbyk9)bUeBz7CjTn$sJE<8sG}tl+qQ!pf3QOWzQ!+!-KwR5GSyV_HB-eYH z&1&)=ywNEYV3U{=wxGwZ)wHd{?1&4+zGuMx&k0NG0JA#hdjxAqM*P&W1^D1?0JMySX>QNU{97ne65IIkYauNVELPc$!vMq-&i#vL194T6_I z>x)aES7c^Tc{E}}(hDJ>S<|RTF-2B4Z9Cu6Cl0OFC@$_9bmQrD;6`&p?m9XubGM2wRK1%YFuG#yUZx z2(k^0=`q&`73U0UclIy8ufim^-}_A

~;)DvA(@V@Vz~h$aML^=aH(1Zs zLzrozHzqa^2zAPgCybq|6c#5YoELWC>EnAo5Ie@$^{!%>uw9rS;X+tD45q)e4-LP# zHgk=zcKMx^eJBJA35+KMYLKR3L`I}pvE+Gy&@<4~XtfO)e&q5yD`y}=esA>L^j!+uu1A-IZndO|j?S_P?U^W9P4+w$m8%{mT)L4lYjJ zVajCRVPbA}6gSHeV^?HyAhch6k<5YD?4>5@x-9-eX4>#a$J1YL`h{Vq-|nHuMlfB! zHUROEtGqXvnT*p3qHZawM`=Z(<<<|tLwn4!S`WaP1Ti-e17O(&7|0}qlHBuywSymk zJ0WM8kYI4g5Yrz;6xe7`)wnyIIk0-S6!A+RHe>n#O+he055I)GKBhx$^NYFZLQT=X zO7ggAtjnkx2l^bRBE|wVU?#*LT*3$BrC5D_ttQx zP<4)cV2MeqY5F~0Yw4Gr(gg5lU0q!5g5OUs@Ll}=S=yk+>;pIk*ue97J`2b%EW<@H z@kr=5g8-w&0JpE7AvQVO?cMQ`z1?{yDP-KFFX!21Ha7Wd*sHFmygKl=DYUWasw(t_ zDfQWrh`V))YU!I3YVRK#59=he$Gg5akg|YI%sBXuhK2AV6JHUzS3-h#!AT!169E!D zwhpoE@@P!6tYdZ_Nl2aQjBBiZ@W!^Lm2hya>df2Gs9Pu2{buR!9?H#+df7X}zbVAb z2&I=T|B$+dLU)joeEzns`9l8pLewqko`b8eTN1K@gU(Ov;NO67=ka>Eh!80fp2JJ#Y9B9-_^2ozC`3Xy^qN9Md_TlAZ}}MJyBgHk?V)^o zXPxh*hUMt=70@1BS)7+c_ z{H)IeqA%+epCnV%`V03bYNpZ>6>SUG`6u5kwc2OyPh5KIPI!P!bo2{OUL`iMq`#wd z*#rZzhH7oV8i5N4{|#ecP;H;O$xxkevsum$$iMomKXG12Z)hG%ZQOKAb@BurLY^V0 zNV4RBYvMmZbMo@d&kwl}gk>&7?d&nC;COlJ2;Rz{eX*H93qAv_JnH7= z_^U~H@{~Ls0|MgLKL>3Ty>oi}MJ|3O>})T z2`z29f&?~Z@(C&FDX>3?c6K?x0x7&b{q}4~i|{a%lWtf0vfXaHos=TXVZrP#!#n4q@tvfsmYQH=M zO-wM$J3`*p|BZ38ii)ENyEOy}?iM`3-Ga+NaCdhnI3c(U?oNWkBzS_mySs$|!QEl- z0fw2M?_cYzb9v5PukNnux2mi5yPvA=UoDZ&9`zIbry$$^AysG3rJR(}j(XY4vi~fta^3$iu36Rtv3CiHwq1_HKxz&AqNX$BR%*u*xh`vLNpCO>O z6K%zrMKB`fr4e&Q%Wbngl#Jg)Ffzu4n~`MI`rSv`Nx~W_|9;e{tXcoH5DoNhazWa! z3FQP%`=}A&LJHnf)%M-4h3*|M;loBcfiuJol39jsGILs1wpqSzi(wOOIM`^I?%|~) ze1k-6OcWoyX*2Bb12B~vVmRVC4k3SACOyKXM_u6urW}^qmrb1bxf#-fZ>;X+=$LYS zk2Jcb88=8OCmA;;73z3%eW~gmOvHU=^8%|{4IVEE|J};H2(+nmHU@sQI6&di{{2rP z)K#SO{j#^?lJiGeccju~B1}vkH~j{VzWIIfsAZT$H!(IEWf5wvfkKQS(g!`kRBN@v zZC4cE$mp!h3i^D9s5hNC=dnU^E|lJs{|~~3UWR1-?`!%^W=Kp|VP&J6P#Kr1`~Ukz z*#G&XG~VjBp2}j}FC!jZwo=C-ovJK2M^)AxoaU}Jy{hA?RpGi4*eOb&bJgHK5HLRjOEY}4kfA%v~VqYc1fe+n-@;^g12%>lz72Q zo~AW09^j-EsB1>VY4g-`$tRJF-$dGTI>28&H%93y8C-J)=z+h$9^ef?D&>+9+*}dc zQ5wSNJ8j)V4_^wA13?G+*qLHzd&(D%k{`CR zVcK*LUfQ%Rm^Cx1TMOHBf*DaBU%fux=JV5oc{_UWr8757j0el!&A8pi?)cvBL9(Pj z%Lr_y^*1WwQP6JfjrKlY=RJ#^SQ7X`*yb*#t#M9Lwr*|)g7y`BDFk=t?1nt#8)Bx_ zJwVmUVS6zXp6=ggrieAgloHUtOfwL`=+@|0OUrC-BCX{w24O z5%`5K4xDhn%yxd`=@eMgL;o0eLZ5eIp^Pc)kNzU@}1iuFtUmE$L0_cJ1$C!(% zOh*P~k9Snjt_O32%JJq_VgGQ=?$D*xy?w|TmF==k-rf*4O)jutQsV<|}1XX8VHKHmq3kL4mHFZ-(wLdOoJB$11LjW&My3cqXf2=({Vy zFCaU)nGRq8O><4C4j}K_l*$>VOF{(A)lzr<0K}I2%Z6C|%mH8ixWpYA7{8yJ>#NxH ziXkx1qhyq~LqW=j%Ge2=)ZVm;s^;_$(#vE%U(e_NLZmOW%$p>yx;VLM8CF1b(xZTz zo&q_+;$-`4l-o}?aoc$GsSr2$!E1?$byNvsK$oFiw$0$pA%~JOx0LYay!H2;7~e$W zhrhJp06~S8)RKf!Pa2D`QnQxFIP{YrZ2~CLbPr#t-z#TKk%|o=th7H77<5Wa4plPP&&8s)N4kT~Upf?hJ*w&NyZbZ^_{t#P49VM8#N`NBl zR6U%vxRc5BSbQ-n@r-FUcj^?6$t|^%A>(_J7V6iS+(c>nwRWr*xry@3AOhV#*rqm; z`kg_LUyxc7JMrIri?GTs%5gkWG$3lN1tug>sDG;xK5)8d)>y5CK_>FC?f6RuJ)|P^ ztEZnA!P+YAf?7uT<=9VS-;V=6o(;(|*Z;f?$?1LT%O;o&32sICo)D5d&na4Bn8<6- z&b?F#kZG3chIEmp)y4XXn%j+ygWr>%RyN7SBmX2VHS5(yS}zL#kCqy+GQDGaTG3qh zGLRC?kYCrK-Pcm@#1FjQ5LkZ0r=y`^62tJ-r|T<)tXB8+_O^h;%BVwiKk^;#bxR9_ z)GUvS6PD>nM?X^{m621LP|p6YfwO#=;8Lcn#gU%TEmb#GYEsL7fLIDl?J=-W(%!|% zps14TC+VKTkJAPLVq6{83ImzjVq%{+=+vZ^>ZCfxKynx7BGa}GUtw$>gsy0AQhW6i=> zJ$;xSdh<28g|h2rlQt2>(_Zz&7^@@~vU+nBKiNtd{Z@Z9PGuTu|5j(u z;bGG4N=mHWt^5bdZlGOW1+~ibygM$Pu%qL*^y?`3OKPg6D35zjW{$oT21YXdsbT-a z)M-Z9^*4pwxC;nmMNOaQBwELdM-SM{n341ed)|nt_c97Ibvm#7(p0k1I5E%~EJGd7 z7zaY9==SRlnXIJKb)P)N*{27ps-8#FR>!C5yss5#7d)?-3S9}+a3Y8Nf;AQmGQR2i zQ22b4_b3S^Ev3gh7FHy3%1PbY1DnQS*!;b@Tk#0|x8|g(#VlXo8J=YIaQzY8eNf|4 zP@fe1x%l~?j*7-)7Zp{l*+r#la$gF$H?ai&Nuppl-OBWC@=D6qoEk%rEbr;V;|ijL zc$-ZC2Q}9>$>te$mDv^avZ#R_`se`iiPY)bKA7IZVlp<2_r2f;ovV$wZ&>lwZU zdp(7WKR&tZL&#p)aVU!POCAt!gFMUCf8}HI`{E@x)YegdTL#IQP9%=c46^cgqW`FP ztaL|8W&KqD*`jY9b_;i0mWjt=%qOqI%se;0eqv9KqMpskh#2zovs{*~q~%)Z0J5c6 zJWe`cK7O1#q;2+z3-~ID!Ztecbov&{Fbh@a`q!^%%ex0k{&R_!H0-%uQ>tj@G>gTs z!?%yO{SOLoXTe#JB?=OHI(gFu-a+Cam2xZT_wWUJ_lvPstY9^lBuSU&PCUfc3Z0gV=jP8oG+ zI>{5&tuvqB{(R3IS|DTZ?NKB>scZ+$8GIg9=DVNaW{Y zh@7D5>c3{%ro{_>XnGkgx+#6nne{Uvc?P$SoS5_=Xvh}P7D_jl5#a%>bQ@%97$s%8 z@cd&7Jx*ndZEm4xC33#;7<6TTN^F73o_i^7Hcf^u7Ja0^JgNjqqNQ!*e{MzI4-$9I zdHwY|Gb_R=CwEG9AAeLVlT@`-w7lp|oN;rbC!F$Y)t1VQ;H%;CORVVEE^@tM1}DsM zAQZZI8)Y>czJn{UrEyzNhfnzy_@wIBcKKwxd$SAF^wJ?^S{|(ylDb-O?6xfRxPM%9 zvc${K+WU?&!}G0SCtuKjP2@yRME-QYJ-H?N`^)WZ2*xra&}#>?Hc{~-T3rGVS{u*> zY*khSMKnvYSZJ$wSaMPy<$Xr*u7RH28gx3pw%wc!CS3Y%2^(sipMAYg0Xdveog)6W z>)$CNZlq=Bp`|yY0iLl*E?H4raQIuL3u*#$aLxq9%#0n2pNdG>%YN}?o%hQlx8TJzKeEIY5 zOh(*Kl3Wso2ZEO*qjNjpQ)S87ktP>D<1ip&K}vZqfXyPIbcX?nzsZOs7`-(W+m#{b z;nqUGS;cVxz=ur|1}tVN%Gce&q&}zF(?DQ5m$=CguC*8LB&k(>d)nwDHa^ahG#Tu}3=aQd$fPe|RI+nl^Ubbc0j{nYjvf{bfEwiRY zha}I==nPRs>F*&2OR=0V!Pa`|u%G3oYTb=~PqyPw+&V-RJejyz%PkKg7J51TX48aYlCpCMm5#6Vgg6RL0ye zImg*+g*j;WyB$rly05EPg?IB*sD1lSVD|t~u|h@@075qW-g|Rv)}dyiPh)3dp-7U~ z2z*Izr`WI*qG0t|=ad(A)i;%Mv~05#Dsm+HZ@>rhD|MyJxDS|o+f3ApN2=)nl%dL$ zKR87`89W;vF<{AM^Xu5A(u10c!oQeigB_eKt}^qQGc(8LFIdoblLywco_`(kNUah_ zyP%-uQ~8)+%A|k1vpUw1oS*xAu}%3XqNEg_SHp=W5?A-fP%u;ZNkoCdHw@UVfR>Nl zRT~kfL~mRBd3|-ZIQ+xA!2D+iQAwrX1Kr9W8>Pu-*~uOzOTUn18X=3>JQlfzD~gt4 z*XuF-p|Wis{?aU4<6vSpXXj@kN`9l+F7(XZ3E#?OqaZTy_&RobrN*Au=+NU8lJ_BZ zQXNY3$n>V!^a!o@U@SfRdcB7sl>U%ob8}bWYUbQUKh46)iJuV5E4Wh|QDNS5v5bi_ z5g>d&YC}HIcWw_bnAMDpgVLcR-CXa1sd^bJ3bbpxD9(=fu{)N&*0cq?3;Z|695(2* zzHO`-H-NTQr1-T%G~f|iyO3N7-{dRFvSFGL0g{W)@o89wnXt~GEgVv>m2tGP5 zIJXiT>4;m3k2b&aY959 z@__yo@jLzs)NZ&~2*Rz&d+^nNyWvX3Zj$teG}63ZfMn-s1kg*(wuVHx@AK7ySKcle z_uf#R$jfP89?Q^p@M$;uF6ndS9`)t1Yq=6?McqJ{chbN#*daKg2Z@pS{(@0>HtD=2 zf_CdiucfIgImr98>^;}w1dhT2{J6fWsxhnxGkQkU&dt;xtcy+z!%{iz`v&Q@6pPcDi7eEGB z#bt8d`8D@8MBMU%uR0{gwUxp@w1{Y!Qsuo7^l9g#3xUjfePD9H`t`?qM0r^NV2=R} z4_s`VzPkVLAb9mz?#F|Y^JV^U>ygERe{d`O!|ZfZ1Rg0nVYp}rvPHb&tR|#1^H3a6 zt{nD<`SN=&3CNhFPRz0RBaBRMHa&PNmCI+xmLYUpM@{}_RO<`61{J_Y=76!ME5 zwormkGWj%Iqt(IuEsmps1u_qScuinF-6vJV*i^pc~L9RD~h2rAO3k$unD z@8x%^_;2bDQL`{W^c}Fsd>DZJ2e=1fqj7JLs?hpinfOYZb=n0x?P9B&gsoQ$H++l^ z+ZD=ubz!}L&)2$xz|#3YH-8YExT=@FCb?EC1!2;hU7IZP#ht#~KM@3~eNkgY6M=mn zA#8^UqGl4>4%?+yB*7tJgIK9aOh4u%gculrO~`^K+# z-4<>)MF#D-)>1{TJEsWCJN1;vBK_RGa7ja&cd!ZekIO9~-nR~$R=Xm<49V^4Y7#-T zW8vBzkY1~l8R5i~v`eK%3Y%8(4Z^YeRa{O>A*iiY+T)ujHT&^!23t!qq_Z?X>I(;b zA0Qno>asivXjeVl)#&1>we?+9jK_x1p=>`*zs&EWg{l;ZSse>$X+ZC-$^&;hgYvs7 z$)Mhh(9iW%O);m6O~E30Lt^qLXbEU z^KZgkP~VrGEA{tzwa>=?%KH|KP?lu=)!PV2hulQD#Qe4N+|rhd4O63*SI5C_cBda# z3--jO-E#oJiQQ^S(zF%vC7AS+u9A_L?HZma&&H-yU5Gtc`dV>+9OoMO~(eeV+D`HCsz*c&-!->ZqEwPIe2 zz%_l{FwlRMb>2YI=!sr~dC`PjEgAYF-rvFU zFm#FOqC(X3x4pMmt3S&z7x>;Kv6X)|(mIONidAOBO;%IJ@u>*|_6ks8iEny*Be-#= za8B-#_DFiuJ8-sv`nh&0p@xR~TEV`k(~fa2RH78tTM*f+Sv$qvIf78XJ<{Z=(Hia1 zD~AFKy}cifW7g2F_DkTRHu7i|MDXaCkIAP!g8bP1_apyJMe&jWRYxMGC%4pnAcMHd z>>;gw5X8dUclBzlltDJ@hflKU9b(MTQ0FVMxuTj#x2ak80e~^2aGp2Xxn7teB6v{v zUjIpiMG_@LPGbuJ6+BxGe+$%;E7Mf?%3i43MN9WQ?PGM?IK{Q8m9avqiAF)4o?Y_$dpBbs=qoPq#hlByC5YWl@_^78J}V z&nfNAB9cPOvNl2FT$GE9t&N^cpW}F@*FB-R;Ho_n{zGug z3))cfL=VR0&!RaIkF}L&z!GS7X3qjRnhS!5&$~=5^QNfJFVQzA{J`V52tQ>BP40Rr zjGhV2x9m(y*XzR0v)!#AILif}mzQjhuS0F|3&e~zl>#`NX9_dzM?R;A;x}9VTK~t5 zMhPh+M*kDzxue9eh#RhRCA<*iVv*gJ&BV<#K81ryqzvU5b?-1bHENOQyDL;?}DNAS)ER}5~GFiL2&Hq7(B}6S=aSo z{4`C`vBbImo&r&L3v>aEgYwDZ?O~mcUz-!w27S~z(lx&V{Y7ng-hRF{CRw1^Ov%1& z-nG*ug>NIdboK-XMv(rtyQ1M21UkEH7Sl<2wVEQa?7!>yKC}zEXkCfdU&igFD6=<< zkG?WVzsLFO5VlGK5Aps)XkM_Rd5Q9BBT!UD_+Y+V8Sosxfi-odEN9;ptX4~+;A3p} z0na{9GP=v}AP{22d_s^5gou>b5%jL+(DU*pU@!B51hE~utOn5t8WKbe!2 z6=SexNx%H5^}Bqx-1t|yNfTx4YhZ2LJJ2-OaEEa$&?v_?3`ie539Lfo4?2~YOO#iA zX2*{FP90Rf`9{e4=OGiw>Q`OPUJuqgIJm@V7{J)>e)*w>6GOAORqFGl>#p6e$Tr^* zfo2W9&zeS35J4+9)J=amC*(Ai1^IA7=PsI!Pj+nnUs3BZq%)z~-9& zmX`0O{V?V6+cZOwyM7x9(`KiVH;3y1F5dIxg={wk7lSyq+^56!Xn2}VH$EF;7+r`n zZ*|f@=@=C@-VZbqlBMV4xLH3COPu1btGQu3&Z>?6mb04f%9S-6DGc74-Y3ge4< z-nw`~Od1j!-xaiiN@Gtz4%Mq;!;o1h0?FDV{d2{Zu@)M%rIv7LZ$IyCa0c=U@jg!S zCD}dKxFt-T9`zEt@qFXnkfnHV6yd)@VN?iSoSu?y&guyu>Gx6B>;SQpZzK22 z78}vijJgCy43TXUu_FMbl2(2E8g{ts&OwiVPX2o>{)~TXo*3v!IQ20s-fKpEEm^L=JoGux|BFXrfY5|7zh-t4x)IrzFJul(C#8NL9Y%~6 zAvno9VQbh6p${bl*p{C+HdmKr$e{~91+7r+2}p8Iwpy@VCw<~=zcEuVRG!;sxmh5H zG|Kp*+8EJwnG*LKEgeRusaNm9 z80u18@GkuPWvou~*&^bj9#fS&uJPBmt0s>SU(#i} zvvb`nT5vK~hE~jK3UI3mqq|cApqLclV{v*V#<`p4iGWp+KH#O8Ld^zMPQnW8(huI# z#9i33ew5)-OB6;v8&gZ_46A^GJ;4s1zvhuiJFR1D*0C)EFg*D-cIy#<(VwUphNCQ^ zkhc;!`8ch^$Xm4XtEBqh=jM8$j&(x8c!Z5m?LA@@jS)-HeMI4+`-i*_#(c0){5Nou zJ_#-oxUXdx`KT8KCg`OvLyt9%9x8G<<_cau6V&ICLSo-RstYsABWuQ%HdOvJgXB`U zxzUS}L-R#1^W3mysdxk-k$vX^hUDn4pYuS>_a{Izfy?e_|6Pa#53G$*o~y7ALG0hg z`X3rXWEfB6>oWrfZR^;tRX#kHv6xvm)Z|CG9j4kU?I^2gl;H<-gm@R6U-Dx&rmuLN zSyFekHtz8=Dolvn5Js7`zO+eYOxF9WNMO4frV_=W=LEW5zYEcTlo$%S>o$)kOl)OH zO!QyHaU`}nR$@C}nSIB@!qXb+M`)(Ej@eu$1|43{cB5abt6H{!{Fz_u`>_U_E>4()i9% z=QfAu+rmHPt0a?qI$JbiDg0QxlXID9TtRg=<~bI*!DaFnh>SgflUoYI9n#%N zpxTCxj%d*comY5jf~nFTign&oJAf4#>S4qrZsCc(yw-8%C9xcaJ()q^r`IVpERjWP z&eTomT{Af&HBH9aJfYZ^sUOC_!_07h#rev6wpTP6oes=PpP6e9VRD8jv}5CH1I=#; z-^@I`#DL^4-Lkza6lz;mkx6^imJZ)OgFk!DEMA3DW&@eA#p-jAIsMfb$WC90om(QY z)8THPSe_Gj;gGz3y7QHbFeF7OqKv0i*49OdUcQfMMA-VF{WJAx^83XjQN0A?%0-oz zK92e>djveeaW85dyeEi5*_+m(=FKpB-yd7beA{{LNALv=Qj)vH6^}1_c5g)vQ{C-p z=+4XpIV2?dHT)BLkgDsK3?30J1p7P%RMcxv@tDLLXm$Z$X%@&IWPN#pz2YHM3x%3i zT8ha76q@j&dWd>ntrk?ifDGP<$`}F^0UV54SWgsG#N2yNCLNV9mx|=oHLp{J(zn!b`r=T>qc_@e zMj6}zHg^HZk00rlmq-Y|eo3gHW>@>N*Cb1J-DfsdHs4GLJ-CgpnQkh)75g@EPORw=r@#%9A>an{S)BrnR4WZQ_yK6 z%pNU%?+-m(x8mvWkRAVCut!!fs{j|bx%o};-d#Q%d1ki_s%n?N2sor42w^I<6aY&; z|52eDs)GNre7X6o+Hn0u(>rg~3J=t3aatj4egyyESJd;;gXcZslUSXoJ_k;wG?aRg zLmP&;)Dsa}8qL#*jO_1bEWfx+T$^Y6(lRjwksWrLWWaet9yavsJ++0(c&QpIL&R~L z8_&>Pj{_VbIIkYE=m5OYksvIBuzYhSd6~VH0ka>W-i{pyc8Mf{M&ul~=4Z{vfrvk8 zTrauF6g=-17J{}i&T}euh3Ad*wU8r%{ro2-4k8#LO>@vc%tskVzrEg^d!IS#V`_<$ zSDWyF+ZNja@w;PJ_EOM)uJ#4DuxtX>Xe{HYN8Ub@F^aJQ2-d@YnXX(dzO8_Ip4 zZ)*|KG9yqokd|6u4x*NkbqBEe1E{E_UwT!yo}K3Ny3h-Sb~rYjiXi%7$MJ}sZ+t0J1> z;0E6o;agNqd7ePJ)}aS?^E0z?I-L9Ai#w>srr(eJ#rRJq?p*6dhOr68rOzvUx4u(; zorI>*Z&?@N?wO>r5WcvCuIJ=5p+~OAYp6aS-snD| zv_PnFQsNrY?piZ^Uc;bhq~jYLp~Xa8+>xf$_I{y@d-~+2;M{HGspq5b&K=OyrXX_V zrK(C&)UBE$I5uGYmb+co1-_)TRzShxMAygDIcJ#^0FPy&?&_K#=-OLDsw9**#)=!L-TjmW9x^f4734`K@$B8V?G3Q&5f2l zv>=`nV5dHY6EWQI(;-uzu>-3L23@-=?D3s59>kt)77f}!=1fR@fj=}?w#8>%`}y&o z-3Z@}4J$PMtwQXiZ!-Ex7dT(gdA6XM(8MB~1ey96SHbUw<9&quYDZs|MRXqOrL3>@ zJJW5Hg2i)MGx!pDbJxg_%=6ygMtH^cj*OQ;>^kFwhDn+H>T0MlLFf-k?9J5xM&}ur zSyaJ~{OZu?=Qo5Ifua-a zW;Eun$qQom?^xd}PzA_?YohU+Ah=I0X?Y2|MnTKrc2LNvbsPuG`tu80wOfgZ(wg?9s;ncguuev!2w*H}P zMIergxt{OaFh$Mys<+8eF~pI-}ZfpT%ED*FIX)OSN$G&^-q z_EZ^g5#Hs03k6p*CulEGUeoy}iWdzyG*WilaRYt_f~ZSpz^?;+R!_PHabL_i$W~+Y;gfW3 zBaBqF?$uoJ?dWvhNY0(u^f%x`F6X@`K)s;KT&Vtc=s*M4C0y$uxEyHKUns{P3)eA) z4?qroHvG)I!o32u;nux@x4`^Fo9niA+fRwcXq~Pq7Pc~W8(b{{KvcDL=y~3ncL0%o zUG8t#y#U)Qx$arn2YHwWbMNYD2l|@Iw|sJnALboThrzdbqcVl5*C(!0?5D$!vgg;@ zF>6}m?q+Z;13rkULTwcNTrRWq{q_W=gG4%gZk7Wm8e^~2e>)Rj(fBoV-xzpJO>bbW zDfYAb28b%5dC#GYEn^#8V}STJq}hzYgs^P~`$Ke21EH(F>L(?e((n01!(~hbIFZ7X zk8rxOh$%KcYGVN8ZF9liv^WIRd2BTHG;g6zV&9eKLWnTAgQscG z_U$N^MTIk5oLxSWL+PYcQIyNt)DO$?A=g9tTTEz~{#X!TZwW}ZbX6Yn7 zDcIP$a!U#2x)GsqG;zE!B5 zTAjtL5f%}LO*%LmW!h~Rj4lu3{fSn`QTOvql+PrQSs!F;zV-xr7*JZ9uew3@(A-UR z?*M@6|gpE|iZx#d=qfw9eFa_J=7g6V}ZdfGoYx?ta!v%}Ie(%>#X6<}AE z#5;f*pnwOh)cC3~3EBMggzrolH3@_4qwfzV=jpAq!5c&9Th2-uRD|*bV2*wtfR9XQ zsU0398(=i91U2|MQGZO$Cse`st|}w*4tO{BDr!t0R5dDK524Flc56ntzioT#|Ne}76aEl;^O3>!B|+V zJ{Mj?aC`KwXf{OO@d}ZuwBs-sCT{=Aql_|rInp`1_^RR-PNXt3H2;~pK~i*VK|IkQ zb@uk#D4Upwz@#6Z^Guur@a;17)d^4tJX4pwxdV#-4L8i~Uo3@-CA!-glC&woQV z9(IeU0{^Y83J}&9xa+#7t`Zl6_dlQlCi=!NX<^XE0W(;2%He%_+S0H`#TU;Z7#Aq-HHiVYv8xQ>q?qwRFHlP9O4vJd z34x3&+)kectbM@JjY5#$PW*$+REN4mmDH_6z;ozUMom8G%xbMZ!cZ|M*z1vky=35p z1O+Le9ff8Axg`*>(6h154`irEFC_ky!kSWQfApxhl~LlmTX72KeIxYzZ=VtIgVQ1$ zg+M8`vmGIm7>5@)Qk2m6huPpbAlK;Y65TlT_6&iOZN5=yZ%&5Gv*%4`p`d=*hy{mj zVDGcChX{(HIau;|`RYotrb{K?kphnBy4LskGDSA~jSUb?w5>OVu&{jspjrmoG& zZrCPO-!7~6ns|K6rli;!42#9K>2NUk>j$gQdb!~@A*{>_am z`VG`*!1+RjExk10oyg0clC}rBSR9lA#Wbx$&9XkrcDZ?#T(l10WcJhZr=h1qq4%}6 zcoiIY*>L}Xyjt@rV6_DN`^=8Oe}Te;lCR^)TOy9%wO74JW)odsia_`H@aoZ?}jSnjD};VVf{|bakybJKl=>KVKe8=PG?mIuS zEeh9%yDJQ!_7J-Q+16lstuIyX&&J#LE&bI;-% zK)UfSK3KMwlddDdtK#KcB^S5*DQ$eeMgv~92498BGqCWWG)~NcgMJSiNc}7!#qI(x zA!&c8*$USq+ECg59S??=^=r3B%f=>|`bCDL-?$SN4}slB8UnY$R8*+pE;&=c`dsjMef10YOJT z2Vs?&0hjZq%rI}gDPTKkip^#<{L{4If=HQ4f~&!Q_tEvGarMVbf0S6T3N8lK z+q{D-tkpRUeN?ZW%&e>)kj)>mUp;Yg!`yTVczfOi?vfB+-y~cs^mDP)&?Uy17YOPm zjSjfvX4Mz^tSUkt_CR1P+W3BdH5&v9FS7AEuNP`}AlgFPWV_2xZX#fLJ_vg0h3@3| zwk9(64cN*ub};3CcmpT8;}#OHuRODP*yCGd>BP3#GAY8CSy{`mLH~&X8L3g(_Kuo8 z3&@Xhzo3hC6bW2ycr)zCD#qpmQJM~}9Jk~9r`|5s&xOyd+Tn4Tr&FDSp{|0*Z5fcP z$Sq#AcpwpS?ANKSN!r-(d!oFnH=>S|(HiAf&W15A9)^gBL!{1edch{JU9PJ%<3^oL zumZDEZr(z^=7g29(10|NH7}qxvBOIQtzG7m9(C^DuEZ}&ak@G}lf|-q#Z2qv$$t_0 zDjEn)j5}7o$0}p^SfP}3ceQ;|EAEfwlP1?BE_k)u##!m(HOt#pN+k>oc{F*d! zYV2p)Bg8nmy1o~drng>3+ZrYPvHS~36$=>Ati)u@vE7UdG!Ry3xvgg0w-V`a+%kvxQe9 zO%P(D4ZFI7b`aZZ8rXi+jubq*dG;swKhe=>9m3+TG}vaz0XqH$fBj(X;d_t zFx;H>ol3yCg>5nWGxiv6tSSW+W<+ZL?04_@uMIKN)2|PnJ_uh@{cD*ogv)PdENyl4 zpIdiFrB`0$S91NlkV6D}u!3jg1M5uPO%QD))Jkfl9u3Fv$uN4YcHqX_EG&Rq@ZbZ` z)p_u^(ruM=HK0O*v77t@F|5kK$v2Tmt<-ZkI4O;))SQ<{eY=>a@)Q>X4mfq^vBF?t zv!Y2n1sC-LRMclZf|$B6UW3&(^s7jVb)%LKt)}ztyWK06dk#reI>W37PgcA4!z#eu zI{}f06G^;rt>>Pz}q+!{wntr z|G97_Y`JeNYZ;mr4~q5_CL0}Fk5bg)%gF(0%#kg>RK)(dKtj!yyVkI4CUt51c%``r z!BO{8&X>14XCHme%v#ttHV0DL>Sm0_)Ggr3RS zz)1q|x?zE^7w}0fY_s!WC!rOf@LRR#vA5uDZm!RF;wms>6GYi=sjf*=$Rl}cu*&^t zY6oD(#GY&Pf7x>p6js3E73K@B3G4&ae`JZ7^va5EZVJAc<^6(+X5J~jF_0ybA0we zn0j<{-qBp~@GZ0cfhm!tPuo*T_+Py+S~ef$m1w7KSC)#qzY{5M>)jX+5hls;$PtwF z-`ulu9&L89PD-~AW57Qp*@z~tPo|OX?DrH4SohHQ==4LSKPz7mxu#zwJ;K+E2aE_p zfJya6*RL_}k4}|dcFGgPN%sP*Sl2dsyTritU-m}R(`{21pMeXFhc~C-7xT#Uc{QU@|*%Dl{Z=C*YJB_QoGbGyijhiNnVgK73gK5zVl9Pev=+U2=h0&6(rYpAX z@K%((LW0Ou(DFzG@69s7S9@B~f*La-S5xH)HZl1OeP8c%{TiD8y0YM{Hu7DOyK!4T zUkJX@3-g-ZZ*XA${o|8HYtH62!8mH~k97TCi$&Hh!?xQRJ_S252+RSKNoJz863&ys?Vub>K;k4?a9_ zONpGW3Rv4mFUI$~>_eA$|K1qmVzhjo+bWNW6It$}zR)0_{4jl2>@JGlY^VvpM6Kmg zfAFRlsMozKbT?*!GYov1zWeFkflm7^pXegAg0j)mFr>2kHB>JMwdF|VV)TE*%v?od zz+bbqDi4lf(Z2d$Gg;t^H~jhljF#^==7a`m75_Evu*Z9(Z?91&A3j)bCK{;vxcaKd z-cmZgzwseVkc8BqC%9LjiOVZ7M@%HdPwMwysgR1tOz?ckX6Qa`@9`EK9x_o{Mgp@mb`anJsW#jq>uduxC45>~@)kio4ZtVGrmE2iXOL7l~!a7QguR z??ykfGw8-c)bE%mZS~K<>%%cez!T8<4Or5Qq>cdsY}+Kbm1WEO!wvsV)_*l9y8X0t zZH|F_a@O|hm@c>Wc>W09H%69V&L|GsRAhBlvmDe8K0o;Uf zt=y}lNldhw0-w`j6kYw|rBwgdJl;4PZXdW}HZ^RYbqFDzLu^ z^%OjGd994m0&RYG?Y&9tTG#IM%YCA0sow2RODD`wZni~bTQ#B|xMdmWcggWWisEUb zukkIpvB;ik#BcE!z-D4po?w3Ns~tF{$bq{ z&8thH`n!EPNxHKAoF$v+NC1Cqft7J+`)Y4}XY1sNO#gWa6vdfY9k+;C_3R@2fm8RS8LfyfaFp16OHkkgriY@ ze$>u-7?&)P0jpf9lK+Rtk=QP)@7Lzhb@49Wbmx&qzxuZ}5vS}y$CHzPi)_a*eR~W> z{bItxR&IgehE36AKZG;j!&8yQ+I;XV>@NDl1gZ7e5nE z5cbWCupxA8n;M~B;`ECaf(=c0^1$sNM3k8I_tI~;@>u0y!53eT_qEC;ue_1?4f`T^ z2AKkD1$ooAXIQE7hd>Z1s$=aF$kg|f)&KDA)zw^+0wcJ=&4oPjg5mBB%sO09~Bko(MXWJYDNW0ys0MEKx_yGGcklhT|^X55f5*o zs0mj@gR=K_Q+SqsTg==9|Es_CkWRd}Vy}F)2jMUWG?2wMW}{nSrBimlhwvOwJ)^EF zczi@SW6!C`3vX8j>ZJ-F;z>n0fNY}Q|CnhwUY>F7RC7&g-tjXY7$mq4%80T`1L`1z z9REDQx)QrKL$@ntWn-|Qlix zJE-Z>bUw0aYKMRJTU_N1`orV2i0XR+Wz?{SbB@IOee=H4)_;tZ2JZ>qAEqGU2KXL` zJn8jv|BGie#ml!VqPUO4j~o1j-Y7r%50uxTO||l^ z&>z3Kn%;|`3jb1UKM~daQ;$zZ{C7c3u`#UM>`Xh%N(%WlUyx4Qb+&l-JcY*CWIy?g zi3uIEBYnN!F#n`$jqR0(v8v(*Lqd2S^?o7rNK1b8k=T z1`%cMU+@sVcO88IdD`OShCq;V9v6!7!HPnujXvN55~?qC8~095H;ygp2v^~=kbzco zOVMmx-j6}(eI-yq7`aHRYVLY!HS-QcK3JAX8C~rjn9=U+vj|ukqMg~DkE#~YSLLTx zB_mqX9zdYsP4yV8Md}~4S{O(rd12Lbe`I(9D-W-cys3JoM4>nilA7rR`)KSPG!Ij`8 zCgCB4J03G<|pi_nWoebLRaqFyAl4pz*%LQqVsigDaQ@4Q-U&5JUh0_X2MktCZvs#t0|7s4gG5=R{NEQoN;~wsUoT@?NVdMA# zCqu#={!OpqsiLf`>VArGnA6r$KDa1cUM-nZQjA*WGU?yWjx*wby>N8&xU!Fn@w6mh zZ@j@Mopv}8B4SdFS}%bZxuQz7u1mqd-|!duJ zf1QhP#G@7$zII4nL)?7S+}U~QjAON)*?q;i@`Gn3+4|%P)@fWdQlTc3eG;YpM4Xpj9sZ7Q-kREWxa#DSWm@6-B4r=PiMx|o>Lt!V5+}W z`xCftL+vz4@FFvP+T_kFU%Ois?sA@{!{~DaAhiP9ndY~xPteB_~*-YLKw-u z+px{(WuY);+FZLr&P8hg^M?bgArimtzge6sPa-UmShLv|ob@H%Jl{O3`KQcAsp*@t zNSYb4m`^Wb<#JRwcI(5N#>u&HWqp<8QaD%FR;gfF?SPAxy!iZregH2Hi4hKs$BOF& zb618oi9!CVCVD#~iQME8)GLnqe1FjzluPe7(W_%jtyQ+pXhH((%?4raiPTjw=@MTN z@14vOE+Uod8C!R{&lT<@Q?Vjkgu!6^7nzic4)tRB3qDv^FQxC>sgexCF8jAC+HPHb zK<=dk>0{p*ydKGyyizd=*kGfbV{o#rssG>u`AZ{YUZ%*NPkhY*7Lm25UK!Q>U_kM6P@Smsd|l2?Peq+nitS(yEZ zHMfP&)|?k6%$D-0EDkr8BlkH;V@rL<_C9yY>X(~+vv(qFONs&rtux8y?j5S?u5j#y zsFApjMknCN8#*lDzgj)z$o;tPcK;U`BzU$ybeY+XPeoKVeyfa8hW!pPG9| zegye$-x!f4o_12vaRpCdO<0>T8xD4$fnAX6jnCJJq!Dgak+pxW2`}nd;tc1Or}-w9 zZNmO59oDMmPt{ovawFSk4YAUFE!op)YR~aS(}Jsjvc~e%DP?s}fxtnsUP4r-+k(o{ zjIH^QzJcE1Wf%C)@CzycOF*>0u)h9Zyz1|bp|=*=K?LaW#R8ufTnFS=b{rapko)pF ztyP_&H$GV6X}AvmOPcKXskJ(8&S(c*$SCIF9M`LTcA+vrAz3jrbks(AT;132>0=%X zrr7wtZ!bn2+f}}8*gAoe@WS{k*%WB0fT5Ice@#2ME8i=&seL}7tH3$L#rE^OCNU+S zCRoQ;db1d~WysJH=LL^99tK_JVO#PEt%H*syE=p9YGTE-dD)tOqbLTsMHINYLMzef zP!ujMIpu({*rWhoRSCHoBh`Dd3i5N7YkbDwn^&ja!tJfK(Y|rq%#Jluzu-K`fSP5h zYRN0Jiuyf!g5d~dO{Z~!gWothGd3SEUbJ+-LIx_Xnqs_iK-tk4M?;HAU@myoMXo># zu@f0E-{&ONHw67IxW)@sOAx-^CUIcL3{!c+w0aUMla4f7dymy6K4t3aK`gVfTgltoX5`(UZ78Xe%|^=e@~<7GQ#4Z+RNegC)LE)bz|;~7U^YB8#c zG0Ay)x-_6?pEemt)p#|`lhh;g3}|y=V*tz@a9lIfPVsO)diMTub8qhiQoj(o>bH*_K}NX=@k~9l>QzP4dOH;|DQa?j^sB+0 zJ6&~Ofo*aycX7upE&33>g$Q4oGTcaF!A%X@?u=?k>2 z?PM(DmA!j?Oqx6;Bc)Sfi&j6K6MX7h|8!h$jjNzxnDVn1d1uEqN9KTvflk<@v1o`# z8mF@XUrl)x`FS9kC(LKqz?fiR!j+uBms^FDYKq5ky~cN?G~x$a8$-^KCf$+%k%- zDd@R5Jqw+x2%hSZYh_J6$EGcS8|e!_HY51gM7{)2uu0HlJtI;7U8MfF(9Jf?DEJdp z{k2^X=KN170xt8}O2UV`tj6i%O|AzuT0C3PS`{}@i@M_@OslR)PUx@M>IJyFMn8sZ z3u#BA-at5*9oPgLZ>q`74b?b1{KzM0XSglGwK5jCx@^Lp8yN%o)N1an8AD}X(WN9Q zs{w7RpWK|QTYWmADef?JuGXl)XUO-(czm$R(p$KJucf-^3H)anK-2sD0KG`CM4G2K z>%lLT#geMiI|N16?Q$g=`+2)$c$!LLR27;dWSV@onBr^N1$aGPn7vGnL*sA{dY)Ni_@Bzt z04j-`uz7QDSNE=yuO9yIUZYs{Xg*le1RD~U+Pt^WLQu6z6*+!qlcW4^Uo@Ug%lQw1 z9mDAKdldfHZL{uJidD`IY3HOa-P*1JWO(1-6{7{bMh?JiP2GG8mbp1jF-uS-IOe~S z^g?jMfB_J?yU=>(u?F5$;{X*r_iQqYnLRzV}$q<&7Xhl{d=c0E0mCOFG{_^(Gv0=nYmEA*oP!oxPdpjR5FR zEse3D-03Z?gSBTK{f=F+Ns-oSl27;PH7V~H_XyA1?a~V{ZbrV7R(Dk$g-ITB^ta?1 z2WLO>;+<<4+}9LUT1d_>5#9&Ac1?A?`^azH-TwCopDo4ZNyE^6E}@{z`}c2pZQr^F zd|`w?FI%JE{9h8Bpv?c0;5IpI&PHB?=w}{a{BC`Ca4xadrla=9=4%`Np zw$KJ3%dv-9KwW&_8)myD1Oohh=``ZW8f}&io`Bfh)iYqCU+O6a(j)sGh;0uSOyXOS zX%vT?kqpcA?s>!sLfaZ}U9NtMWO>ltl-U+^8Wvyhc3XSl5;ece^zcD|$Nz^E;GMzmc zunzp>`hM0b&9V(8VM{%7K&r-x2$4BikAD|pw%jI*A7&}A=<||lY*!aQ05kBHh9sD5 zNtow=<J#0#+T#=^{~ z>Krk}aKxBY+}uKb@Lk6szj0Vre_0MwOEb^cxHcAQJ&LXvzUyM2J^1%JxFUVw>3 z#VX*<$__UIJ~ITB!o$Wzrhlh_8kDie-6_ggKF&3~6Fh9C_IU%|F^`EQDoLOky4lg0 z^YA`5m|(I0H6=hv{m3bR{FVGMB2eSWKF+`4HSk9bhA%EgA2@8H^ygLQ9xf0uaZYJC z)Tk}(Ngo)fY1Y)#!qO3JZpm>wUY7CX1QTRy*PV6~^@Dkv;i_^2H!L^XTs*Z^yQ&yD zPCZAG^6ecq@SmW+i>cd})fiRofe5M^wMukLC@0#(@j{+OE6iIrpHaJ;L8BJU+gIvh zXBteN)JCJOqZdBf?s`X%C&adNc&LyYaX1(t*t^aMf^iJX4_u!-=gQpOfbvNbyn>lH z{p@lS)G!(H$v59+7DNl$lK3pwHJJ3s93 zUtv*;Z@ene{(08a%GqX(b}0E|w&w)ZzVt7+(RO{_ZBE4z-GQETDE(z%x$C<*0h{t0 zN?^n;o=TbkLtr#_U)|D}7N{i)3aZmqtZD;t#V35)R&ozEW^=w>Yuk_<^?1z^V50jX z9~_yTJeC>_FzB0Ez|R{C47PC^cTXq5uTOhEuq`>}LYdK%_O>Py4eI8i_f}C8I8X0Z zoELzSKCN!c5YZ>8$h!#o_cKF542U!!>z z85-qJ89-mnpV;WCMOZn}OCc{-QwL{DMttetrffKP`g{ClS3C_X5cI{Z8f%dk?%}{c z=V-An`@RK$wV#nVbMG%-$`UZE5qy!?Ejt=9k?T~V&Y&5d<1u;Fxs3S~v&_GA=Zmd? z31*;NFJVkrX05S{wcINgPK(4srXXiFi7?ijoRpOd%auO3h@6(Egz4R~GkD*Bu%}lM z7c(Nq@_LKEZDT)^d?eVaN((i((ows6jyO77-TK~y=fuqF7L-I9^5hsn#6*M5dF6tI z(kF&JIVClCz6$63{z-MNt9VC3dW3EEEI6vjs+8^?{?8BXP^KmV1bc}rmKIRB8GCn^7zK-;TMJ-XZA7#_ zx*j;hP?ti7vZOq>G=3~#x`{r+Z!3_OdykZMM4hSi^m(h!LEP60cRbcWCgrbPg6cyD z8hAnF#hTr2#n3cC;DX*7Q4CLaDs0FcN2TZXvKBwRB)!i% zCd!&dJ@)2A+;0dv)81a=SmLs4!6ZaLfI+s5n48j>%l>-!tcEaVyf6%(;LXP1Wx{V; zKk)7GM;6wjv0unBqH~1pR>w>C;l5YEi$xUm5|iUC(h^rGjLW6JGH>a9BcAO*d&fF@sN}0l5y?#5 z9d(V9?slgQ_`C#5c&aM$we`OhfOXR0yU#fAO4uA=nt?B%x&ypKGFpnDUO153L+09n zve`hfkh8q5?7(ND(_paW#%U^r+mAfhz;0v<@BBTqcm^<8OH4aJgD3W&t9V05;KfIB zR%%x#hf5$)jG*q>tiM0%KQ*$k)f#BquaxBPy>QDvW$ZvJZLVYmctl4+-O%m2HI@Ny^Fb>z;{dX3T0O`Q5U6k?r zh9=!_fc893jKqDvNZ9t~zh$J*vEC6@v)psSOMq*FJQ-&`3FjD`i*sBPO;NQSr?1R0 zdP=fcYc9ffub}3CITkaQs&ar+17Kmr&s~!7kkbRSetY zn(B3SjsNwS;JKxgIXFy4Ku-K0wuhNAfM21zhTFqT^bN_KnCC5H3=$5rI?*P}Lk#aj zRffSNPg7o_NOx}_C$GC^wVLggzbr_5=a#F*(!2d%+dm+mo?qF{5q&3K#MRJqxEHrB zsQNhOba<4Zf4Cd0|LsSv5~CRP5c8B3b?}a)z8{?+Szi8~5B>mqr}hS^Ae&Hr73Mj`*`rH=l${zLxBL=5&tP zlxz_lYt|M>UR_52?`4qX^Fij2#PW;pc@v8TUk0o*LiY--pHaC)fi-omeI-(Zu+Zbw zLvm&n1$5^e(Y~Bu^8$OCC>pO2pFeRuA2uPq+WGF%L7$=C+)}X_-B2FcXE0H-c1yQs z=HEH$dNvz;wlecR9SSPj3Lc9rZL5otoEZv!8QV*2)#-gwKA~tIRk(F(Oq&?%D&p(j zPL|U*D97a~mQ%#QWJ)4rl#u({Y3j3$*#kdf%)$$(7?|EC`7ywl%o+6-mZ<(ViX6)- zEBQ<~jsOoEjvFt0DS@l4Q#=%sT5vqw*76rKVXFv={^$oDr$FH7m~#1v?l)Pb{xVz(>0UHyf(*^Wr>dJ1+b<}!YA&;8N z68V(M0ZV;tRH)g@(WTW2Q5jbV82_~}#YbUXVQzWBn!MczOH^y6O>54}6A7lg7bf$> z)u}rb{13Z4P5c^4m%>uk@h&rhqCq}d=-mtn8`b6$dY=f-M+~J$0gDfzYWzs{{u5uB z7Y)W!hPMztO3c*=p;G1ch>nb_k_!v#t6$TOXc777bwM!Z#E5U47q*+~vb9*+>1|+i zRpuaxmMhKzQf-3~qqk(m+f%yr9h!0#j!8=|-toe~;EIN3Yx%XbxzC}j8499%Kz6sJ z*N6eIYN+YBBH>i%8z2tawi|Xr`3DH4RiOP+`WcOo`KImNA5E_I~km#nUf>eA|Hy z`_a=t*1Mj*#(2#TS_A#zaSTr-G-i^&^v#6l2*b58iH)(+;H50t4~*9$NB?Ll6^*nz zZwx(sqZ8#>PPOCnQc&2#ck_>`j3ivZmdjsWpEa$<6lyT>R4-(a9(VB%tky)i)Ab7& zy&_xwATNd~A%M)3*v%LN{;dfc^+;f}D*3v*uzWrczD>=L2$7avmAu)^2eMl!qtWja z?-Y~PF_sdbo>QOS<@fd4i}?qLxVaE8BYk)cIR#q)-tj4xTDiZa=2%Q-?_?RA5K6;b z>8mAeX5%C(!g_Joce?4;C!!aPJb_#d;n?nD2$f^31&@#Y*#%IV`7x{O9vyi!+x6#8^L0`z5 zFOmLB4HmxO#UEz7Z|YjtEtIQuFKV9Q6YtObR_e0P_enJMb(#Q=!B6{gEy~Qn`;2xpqAw-Ki2?mYN>hu*w#GGV0|c`Y=)u6W z(yTUQ=9>)fg!A~oRm~Voc9rVAKrNup7Hpe z{`G%`Vmz5AK6Ixi04{R>Vu$J-Xpeo`q+iD%4~0g2FJWa&q_2`*fw2tFEd(jI;%ba_ zU_3D#Dl;4-`*9q6jM^_7hR*la9I^9o%48{N6*RI)ONe%{pYw|EQRK7G;X4xFEOj5N zOf!EO>j)b^yoja$?GwV^h;kgyftohw+<*w*P_QP22M~sF6|#6w65kIxr9p|J#gAe9Or62ri%eP_P7VFr zky<1R(}7^-d*g5Sp6*0l9?v)|^o>^;3)en+WmXkSgn;QD4XKfEVl@ypoU1udbc;0c zZBG7{=pK$+b9LJdKJ#nRuBPZ?-d%vr!+hQ8eZEKd zy660&JrJkc(6@tWX=^VFLR_#ep>*dNy_h)dZ1&Soyh?9-d;IPAaL1skU$`lG+v&4G z?_XcXBH@_IlCDhxrGtqyq$E+82OFsS>}aC(PU6E=x48h326-C^65a-R%vU(^zkf2| z;*3Hn_T@ef$pG8pueeI%wOWXdI06P8nRYZ~w(TW^Y4P3c@u|aK=i77k9tXRXL>P46 znOZ{Dh&IeGZ}EAue5N}28$20xN_+edkMl}qKU@-tF^9oo2^q0e93Z)6lzY&e9^9Dv@cnMK zPOw0A>@~!u2Mv3yHN)LcQuV^6A*G1S>5qGV`Re<;?W+U< zaDb_+SPtO0peRw1ZWe#6h_ajVrt`z%wH35ox%uvePq&lcY@Ev`vdY`fhKs}L5+j;)L`GlG zE%P23GjoT!54}u}3=%($a;AGPk$ZSKxthoAwqw5yRsPF2Lbuyn{h>Uy1mFJ~M0rt9 z%Mwn4d3I3h4Cq8rl5844VUjncNV9yo!lK(DX-z=fob(n3NL@EA87;($ z+Ldp5oM-jT zwr$QnXCnRYy#VQy#=a+ zz>LueSZXeMrNFV?&yJx?L%b|OeFw%D-92zmXZkp_3={-M+B*G)-l519?V@Ncd(wwV z90*M{6IGtNcEyJ7zddk2yvIE~hK1($#y&|?QXXi0d1eK+bykT-lLopmpd?k_{abMf zkT$O}_vJL`lzy5Yu~^G*aJFs5aq`*J7hU_J+U-?KkcHpR(xmm8mOITqB$4Hu>l2Gz zH~CNA_@Z@X$ca$$58M15SH9VoD27na6&qW;(wsJXJ>x?YIG5P0(a|!E$61mK>@UV0 ztbP$B!SLm4@R(zfUP=fl9GMJF{h?G~I0zOa>;EXL^FX_?_P_MVgHs%8`}*XZdcuor ztdHKS{~>95?5=)gac_*lKGooaO*%#z-gF-J+RHWk#&0@*M&@dHZpoQ^f67V*vQ(&Y zJdj2Dm;kE0vKgTCTDn5TJtZqtzxZ%`Xq&AM^E^OPw65o=9saC2-K$4? zaQQvIagwBi)u0zRrq%bVh@HQUI%Npf{xU1+7v?nAeIvlK!N}u6Cl-yJz(Ue0OHxMN zq7Oazb+#e z=X+3$cPRF*Q!S;Ozvq`UL9kmlZ*!sw;|KWLl#f?{rkQh$eL!)Yx^7%un-3r^Z14;L z=MZpJZkPKPUSuBa>u&lsfhQBny4bb*>agW!)tSn)$KfNWJEiE&LgP#={Y@trVUX3! zJiQOK6|{PFDZ%OsX^{61_LpP}eIUtY!V^H}>^J01`X7>6dq3_aB?1o&GM@;9lJwG| zHWWpF4cY5iOco^Z%=nZlNRmX1^4h)$nf|o&0D4wF3b}uzp6zLWBN7vS5cgn>hKk!( zQ<16kOW&W$0>2(zAl`uz(V1}`bD_2e?!%|6)Q4vpH;)4ElYfyWlx#lSEdDP&MkPqN znc4VWgeqO}*AJ05>l7)1B!PK?BxcF{B!Qyj3sDS{?^WpWts}OrfSQL5=71AXqc*g^ zu<3Z-=`82{Rv*UceP`U(;b4QaNyxqId47F(eJ3k9yGMd7>!w%9^(;v`@B4wOGZEIg zTen|8awM^^(k*0<67|HKXxgFsOkgg(HZmR1>8GNN`kMijk~@6+Jc_tt7SHQSLEe@S zHJPmb{8(EX{h8iAv;45`D)+$zvWmsEraB>8J%sWD;FGxsN6IJ#UsAkhf}87;(YJ66 z^otmfhN!USX_rI@AyeVR(qkwlnPi^jagwIX3qPgf(VvXlc_B8zR$Co;JwJ@PvJ?{s zAS`>p-PWXHTh_VGbWDlPa!Mfg#{-=| z6B6Eu7u`K;?$u^Gn}Yz~W+*P92DoOf$F`zBvK|7f*@H`O6Y5b<>^zIMIh93Y8K-xc%jEVgB6>9$Bl2OHj>N>=OFY6W>Kr$m{E_v0ig$xALilo~}p2 zg`rmFV4wnZgY#1+-(}F+4#DT_S$$ggh02ne|^4%hh#NA79@G+ zV*9o=ax&r;!l^EDbnraCB;}@BC*|t*EP(79otL&3$p(S$W}9@f>E&gAMWwqP?c7nL zDlGAp4l$P=>GME)s44ITaoz1&F4_i?wpZ@0uc{nFvenUjD%iB0d3H657;`ZVW{+Aw z+$AhLH|hU(z(9v9kUOFc{yfT_1C0_-`Pq%Zkh68eF%+#L3G}fWzP>-JR`|h2atZ~L z=*3HfcpWGCjISq~d$KY7s-z@mwdT&3iAV1Ql`EQ+pJ z6n~yRzq{MUe(_!S=ET^~b+6NhK9o(X`q14%0$3(VwG$q|RH4O}+$2xI&RoZES<$oi zDXoo?C%A+L;~fxK8YE!SrT+X|ki-&Ja+_A;n7a8ly6c3m@5U^n+HAYGkyK=9Rc0S#FtR7boKZMT1f+) z4FhWjA&}a8xRnLIKxCH~wQNwXhpQ-?HeJZn?KVSALBb1?{2yuF9tn&qXPcj~6J$xq zvDwT$p>GN?RVI8lw3_9%@~F`f9B)f>;nwm$(b^&&3Awa4fPuhkbPLU2ZA$#i)Z3`8 zIa}C;J#4zZqFaH;s6;`rO;*J!p#Af!4v;i6@PPL40hN)73TjxD zf<+`z_BLWo_hKUJfgo_5s;69l{5&Vt`x)_B2s7?>o;8K~y?kuekyICfluq-Ps==`& z!~#Xd(nNg>B4y-#LMY7a@ScIyWbN2pRQcf8eVx~a)?M*+Etw#NPM$=+$4GJ^uR|2}(4F;!OIWjNMuoIYMN1y&|3!;xhbnVj zb$foCdj=1_I+bb)^{*mT+L4i@&?8X%ws#m?@rG#BOPP=##a2h#W*%`@>3B)0#QfqXEGV79141KomAAIf<7KNU>g2Y2Wp5N6e;b7(Sdq)awevvGIrEHj?Ezw04qg@scZX1SenPyy(Ej_a9eMQcu+_xctP5oes^;v zv-$jfe^Kj|^om9cV2Xq9zhCg>|Izr>Y|fYmOf)|JYS|V0 zk$@BYIry2~UUbql&N<&tLlcf5bN9 zK1-o9aAyWy9b%%(rdSG(G5EDvfVq>9)DJ-2C(*aD>sCdqvC2EX zpwR(K4!AB*-^zr7Mt&lQZJ08hC$ALgVK){`_t5v%{zgfTu zP6hDh%(N!p`;pY8;2*6h+U%cRLugZnYBxRBRSV<`lXhF^o8Q4{h z9v=Xptv?w9(O_Ugk#TY8BjD7-rX2R`h9RcjwexQF80Y&<@ah}+*I$Vj{ycSGm(Pxu z|CU!TtmUm+=TC@#2X=*xlaR`bX5H;v6Y;GZ&?&X;*->c!Txn1Wc--u)$hUWbfX&8Z)&X!HU39}FBixNNAu(UE`@$p z!afjxsqO!MoA2U=8e#?(ImjFx)_bvFXCsf%v9o4G+bm}TcLHNRD)8rbA8u|!@MAUc z3N-}I3WA16vPF&{%YTWN>OOW41q~rpgQpZp#>M)a|1MSW$NHUP)9&%#+!-5cc2ac! zsJ7zFW7Zi#QS!x4h?Td|KFky|LsZE3I}a4yj8PuY7dfKOd8>4tsT$mJT%AOs4eA*I z&C&#lyhLK=7_s+^l1PAMyFqvdHxa7H6T^lpRIEfNPn@AT|Bk;b=-1zRsOZByBFwMVXGct^&Xzzaq~nqJ z-@v1JpXt{gSDe!jk;X~_0b%+tU*0q3mOA9FzENLzRwjys6CC5H`kR$!$LG0 z__kzm!Yse8c>R2F{5SOneS<_L(|1vjHkIaY4obE|(0RJ~ zQLi&My}0fd1?OMFju@XLjvd_H&D)O*t)7On<|(3LemLMYq4(d)rq)*laYS|+?X3hxAaQ0S$lGWH@ftu1)y@-Wf6^-EF=*Bm1Li9bEh z!EwDsvTr{Yu-HjSR8}4PcC)lYN3PX)Cetv=A%$xjP^<%4e zBV|}rNpc8O!yvT`S~Uyoyujtw2Y7`M?ey4IoNRmW{ZX&&RkT|4?e^_iwbu@RrN+J2 zM*u=9mXh)fnl=wo6gx8)1vTmAp<8|dlf;3b-&5e&YkUmd`P=OP^o^j^t64x72k`sa zY1|O%pObVA;bq5pSqZ5@?n$ertN8Ym{haJB}crAx7TD8SkgARaaUL`LP zhmojyrPO|q1L$|)<|-oym=Gb)gA(Jb&DVWn3fJU!IiS!R{{s>yg7rT;hyD^qICS{A z^RvH>K_YJuHzOnZ&*9JurL(2=a^g|1f=8omAO1`i$N?NYG;TD>NMuQW+%0%xkkX{m zmLis6l8DqS|6E54&Afu)-X#3=Ja($|4~V+_9D4ixWrK47`p3VgR`E#W{tVWY{xL4h z2UhnAJro7UL?e3;ejpH*c0&B-Ijvmx@gU~hM5eEum$r4bvV;FS7xHAlYSsd-o zJV{cR7G}dxCFgJZI&T$ zvwTR?gzh>&2;}qrjlbU|wG2Nlrg;0Ys~DIx0bJDcEJ6as>yHgSMBl=1j|U|bShjk4 z{Ezhvck5%9#@GkJ2K^pYNf!vpJ$N9O88D)FFOv0YVh3)4@Xi)?)H@k76OHjdO3N=WIM4X$h(as!ii!0@$yJiJLE7X*22vjx6r=|fw zM6}=B!$_?t^C%sdAB0b~b6AN;)ot>ezuZDr;k6*uo}PVo9r-G>HfiWvw6is>Vh%Jq z8<GAJ`?x3({O!hK0$AlI4$cR!oP4ab zZROI%&9@eY3h)2aJ&9%LYUY~Gh~z=MP&U$*>WKKu1Ue1--eiB4;&=|mzbBX%4;s+U zGK&4#EqaVnr* zEG>&lByb3UJkz6@L`!~>UvoHKw-~H4R{idIxxuLRbX31BdJr_rZT#Ta1DA3k_beSH zNO<)*xcn={zm8w>5obs3H9);`>Q?|;T;<0uD$wi93bYER;q*TE%yNF$2?pu}9q(BdBu?(cKRJxmQkgkV>w z9$1Px=DeT(Iw>fBdfuv@co-sXi5-UZzEYI&bAg^_4>pE4;mT4t_xXf%g1XNC+*H?i#aM5Iu%Fmxz zX_I#cTe-a&2DcYri3n?Zcz1m#FduoMK7%>J^5M}Fi)GG~Dj>NulA@G*?=Wy)O&ZYB zR54(dxe*t;%Krm$!-EF4OTasd1hVJ&!y_YE%no{FyM57}84}`5SXm8Y%7O_{>mzOVgir1ldxQqeFA)y-CpCb)FR3HXPoRfbxXda| z`@3h=EV#+XpSu%SuNzWBZd=t4b^R1diQ00Sm2`W3?-iKd9n~G{8mY|bVQD1%M7LJC z4s7nSKNoTE>CEc3dGl5rblPVqcG8Tq=st92h4~E_LPnp151t#^!I&Pnni}QnjzE7e z9&1GDYRu~pF^cM`=P{I9=9akGS8r_m8pW0v4Zk1VVqNA~b&KfT*Po$|_{nXBZu-Ce z`v!Ozm59L97sv3>yL+UlN$B(k`B3OX*lR6p7gq)EP*n%uj(QPDUl?nOyhP;YHP$!% z{LnPg>u-KqYHEE#5km=XZ4uS!+VTK_BvNHS(b4|q67dnl=Cki4UQN7VB3q5P@O^B> zH3xFzrGj)%i$sA@Hzg}Uq%I=R-ka7<#~6`W1AM)gc$$DR=9UK!^6aG$ zVj6qF-|Ib~lAmiox8Hpi&G7R_Sc3lJ$HJ>6NpxM%(8Mt;a4mtRpWe%*g4wIJPwPWv zyhv5OGnftl@}v&Hd0?yrAuthOQ`E1_Z*885-#_by3DYeh#%O?9&ifW#Vv<)Z45O$9 zLZNB6M4jx3x%q1MXN2P&IIgE2R-~LjbapiH5SDwAKdlXM#i280K?G>STzx;y zdgOdh&D!~Yc~V74xY^>{3-AqjPrXRY;Ja0LGV|@PJT7KN?%x#`_aV?XUWS>G9MR)o zW8;a@j&0=L#SioVfl>7=X;P5Q#jU}8!HB|3U^*)ptzL-Urot*DLe6_Nv#R|4I)-ml zjT+!$4)sQ279nZb)B$C4KfNElH&;@0>+hTxl%;{bDbNQI5>F(`_{Euo52xh7*S|}) z#hhnLf*^?!0#cm_|0=>5=eOqwLP5m6J*hL!vG&VX@u$}uWf``9Xf__eCCdsQ%9zZU zj@2xVV889NlwO_NmHWQ2GRS{NKN$*B*a9^8=O?=u1a>+L{0%ugKO(ZTNL36les zmMc?UT9(hP^NYxMRW5e|ZV=Uj| z#Rfj`fJn`4VDbxk+vO>I22kF+sebV8_c}vo4(NS6MMuY+j2}-fCrM&gXvl@VdgI~E zz1$|H!xk|vb(Sed0wH7h-WVQ(|Nfg%kxbAJU(#hK;<;pa@Y}Yu!fxy!4^nnF zbVVXSnv&=6^Dp}|^7QR{%*pHJzxsjIE) zV^JL>eUy26lr-U&UWttovBg(M@^i)AKtOrqb7eeBY$fCWhmW@min9sYMsZ7kV8Pko z8VEsyZ4%rq!6jJG;BJd+a19U^4H6uJCAdRiaSQJ5y6oG;A>RPuBl2>rgJ3+>-_tfVMB|b`SL)$q7bAqZ^sP#l` zZaR$Ew4LM!vSQm#(*8ldscpfJiX79dBDpMeYt^BY$|k+G<;faF$|trLG1v%;^^1uP ze1Kz?wJFvsvWuhqU4{|rx&7*qn6y>PRyp8mLpMw_!Fu}y;OJq^@UVXj)8Az*xv?p^ z4;>Ff@ew_Y%*k_JEd%qcbA#mf>>(S^#m|s5&0AHLJIOx-#jWh~_+;;l0%A3+TDNWr z7jJTqMFQ{A_7Wr>XLqe~0Oh!6Y66ih%2iTl>iP_^)WG0Y@(7Ka_d%hik`$|k2>qna zt7y1#6Imgr*$!_@@CV)!lP^d%?&_^e0$vGW;-j>$lDm_QD(Tk~AcFVECULT@I70z&1 zHZrzUfUnm>c}&hOXb@lPj9h2Ros}Yz1z8JUWEcrZu;OjM+5qA6wn&lF-jxI^SA9p2 z_$fIL2_-ACY6g3%jsEU^gVvA{~3#+Z3HA9cyC&HqKlLjqs3Acm0Xhb}|M+V%(S9V#Rl ztYXG!`EwpLEo3Iy!lJ_%dfkr31KH%~ejWgtgfjmaV1>Sg23+kj0iA@&qO3fgNUkMBy(Swwobz`-6F+((eUBn0fqKnJYNh$+31bfcAbQFciack z+|pYW2|<*obcWGOIj#xAnlWw>bG;(c<0d+#cv9ww!;b|SH5(lK=FKLot;d)bu(N<% z{Z8;+^5Is9_G`+SfANFQTeijqhvYJhV$6V&%c;}q3RJwb7|i!}qdb+Q$5<2H1p<^x zz7^j#UM50!B;PLSc`bZ@u}D;N?dxyuDQ5JuAbWtbGtT=i4SqSn4hOUgfEAb zH-stG4^{1i?rL)vS#JVOl#q42_kt#>hBQWt0I&n{FPu;J)mij83x5e4CzgJb% zR{b9^)We{C$ras?5Xc3S1%0g?Z90|Aj#j!t+H+9Dj$ z)5&9-LpDR_uNE{=!mp(ev?B&v77%)sSF980HtJ4x+%xBT4ek>Ndw(} zW4;}e2M`a=l{TdxNbhkSpJhw5JdLGfe2+Sh4d4ELyPT*Rn5>N32Rzf-9_`5jc5`4ir?vO8nyFbQ%uVK{;UM@2 z_txTB!|`m4k?TN?$rA+gc4|%4OvFN!3(spI`tM>< zbF0dN!JhpfXcA6Cy^G0brfj2mv?4i$wST%I>#WsoBp#g#*$6gds2Z2GhzCTCWD1>j z)NH^*viyPEqT5uI`6W%&jJ0yL2sw2v>}4QEYp<$S#x{Z*`&{uLHOOhY-}p`4rge$ z!I}9Qe9TA=4n++LSD}S<3h!eRlg_i$h5DH@I$TT3vj@Bz@=+N`-X zjj^n?EZu!s?MU?3NQ`>^y|* zwc8*Sp{m(kZlKLxLrVC#OAxQX%HyTYMq44BLSvb;tlI>C@{ia{)`^Z8+MTE*-iIfs zjlKP;q{z!TCi4sjAbzVf2sRRWK|KWtbQ)z9Fzo^7D-38x_K#*HLmVcSJ1VB4^n@o) zgde1yUFDTBOdWl@hJ5lNuJXSnHSy)TOT(PkvJ#?B_u__jN7mpP{ z0<{oSBqty!z>2t&gwC>t=cD9#0OwltOAmk@FE@`Of(Sk0qR3JthjeV9^JP{5M?kp0 zDhu|R`Z?J&2{M~BPZ0G6-SBDMWT|>hzSjM~E`A>mtD-P!;cGI3EgXDt0lOwV+f8C+ zjb;R>s0MJ*fr>|SQS8&E>{!mIhA&?)WZ)s4vDDV%MM%475{MKChl5H!-?2Zhq}zH7 z@hA@<^Q6~cGRf6& zAi10MmS^mff)FcDweksl&g*mtDt3kSmUnS*FfqAzo>e#El26~*)raHFqkWIdQ^b;yx+409WBFdUO_|`IdnRM!xNerSED|(qP9WB1uEg>??FXli zo0rH;FRN}UsYWt3gkO%g#h^Af;Ev+M%{|`BM>RlypFLLM|;`s)6U#YA6tH-Y4C0Lg3-Q_>!CJNOLjzY9u?K@>ca~l z$JE4H%bMj4E9o`NfntSt(-9MhGIC|dH+B`O2t$H|&tTtoF?DENnITk=TkNxh!*!F& z?LDA|S-jc$FNGHhd+Fk}K$%Qta^>#cOL-+r^i$8z#e8EmTwbVv*Ef*oHXJR0bXzMe zcpyRzkf2IIGg+RWquh;`P}jj-Fvb#1EAgak1j%b9>V@R?xU(@y1A{ z6tEplHUOmm_WvxHwHW;TX^kv)1nn99b6xWHA8F7{RX?$O*Z10G|G?JxU!XcIN-byaK6LG%Oix>86n6J#6|Ma^cOY7Dw4D!wFPXQgL;&hzNdUv{Lg6bwI zZV-q1qb<|M_;kysg1<)E&hN0m;W~JDG5qR%Oug3oHJ37c-&;B>6+vDXI^KJ$4pUx< zcl}`5#_#HPBUe0<>&2&h)q>`iLajWRT%(AG4&D~yAjsn_Hl8AGtrdC{A|+wSa@!x) zD;uP98>F|8ZBlF2Sk$*NWk+v9Q#Pc4cE%F0QAX{C6nHdu^`fC%p{Q^9%8q(MF&(Uj z5YL`+U(UJ!r@YjrRiKW>yT^ZanewSNM%bk(`-C}@%ex5+@6>pe2w%C}EU*-Rtn%mQ zAIPn=u|$@u{0^oK(#M6PQ5UAiSCIRx@)>Ss;r;ODW9 zNuR(OUnvGvBiA+qqm9Rg!^|u+1IEOQ3w{42aSJDw-*6t_WV6)N^dvaym(cDh4^Cn! zVu`*4YUQXS-8mJL?{~2?#(>>ZpnH=o34$il+R04RTuaik*9`R8fhqFln5`}B7)}+hXrG91^bH7E$2q2Lanb|rYt>PMyC4%uTJeBSsF=+Io_!n>gbyWeMNjM2*(&o`es2<-L#6@R+BO8bbr)DOwY4PpPN zAHb%M4(TlLwy64%fh+*50+rBB}A&n9JO9NWt)LTKZ1T1t~r z-Y>Cr{Jo`nU~d9KSPwkZ44$US7j!?YQ-VV{6^#j_U)=73e+7&5FE8NZEsyN>QIyL4 z1g4y*{}g>dy7izEYMGgVo|oGQIV7lTca@v^2GTCe-UArjUD8&)Hg?6u#bf!w!a2^` zB#3MAJHW1dg=>!W7JFFK1_9af8hSZYzyj;wPK4Zi8q~u3z*Y6QdpHEWtfT++`EIbO zXaVwO2<=$En*UNQyVL%HedT@*paD;pPgq=sK301TjYbugK8Gy*-y6UPG0GJ_koF>u zc`fV@5x=?>bUI9`z2AYDQz(HT~1RMG)moVe`Tt z%wN%pQNFM-Y>vhVo+m&%rs55}4N>*Qo}Kl7R*WZoj330V+xDS|SaJZhL9C0xfy_S| zm7M304}-w*#@VLjOEqXpiIKZrrBA-@Sh(Mei~4S0LM@X?f`L8a5#fKnV9w3?$c%IL@u!FdXV%*Ra8ix0v4NRo=x=^D zM2bwE_PFO{@$e~iwBUs4u^Ib!uo60ho%f7+V9PX$1If7wd!Eq=j<=DinFwLt_Ww%5 zRkcjYX4(+>@w&#E;FEUH{hdW8t@P$n%qOp;OBjeK%#szC9Nhuiz_+;0&i>7h%`jAR zkmbs9i9%U|XehiAa-HA4&;XAT@RmBuPPb|d)d1PO|MT!cmmIxbUv7F z-4^*bA9T}qJ~b&sb9Q@q)<)kq3><1Mm-Lda`v&YuaZnGdkVx%5{U!e^?w${COzH6x zK2W}NWQNgrl=i4+5}j)Hp({T-SCEH^Y($9^Jf!X@D8yeqm2w>YmJ>0!2ew6m?hBT9s!%bD^Q1sFk&>|DUcAs!vrs0WzEZ zXr^=_nh{dwoX*pPYm@}L7JtvL%(Y=}sN=(+8P6Y>YgAvqdfszSb*IaDW`z?!-14g} z>se011n$wjm(-6|jPU9nSU4m^zG|(>1L@xz{%3obj@;9!X-pN46tfJCqb{2% zS9rC#%Jk6UF&im1;@CuY6i4NZgE$#VJ#1!0szKmnCK&ReXfiVJiP0X_MeVa_1+f4u zYi*E6_rN8e(EF>yJH!;qhlg#~^ob__IehFS*dWXa;wsE@TGA~c#gz*f6VVCkgk-p z7KrX5xJK~U4N&`Ek)VM1Be>*IE~m?4Q#-htF+0%UDP}>v zeyTG=$Y&4KEVL!FZXk4AoJ@M0Z{9f5@L}=|7!wjdux|X%jqs_rA9S#Sv(Z z&}=^d+@qHRJdCZ0sJ*V|8xHICq(>Pv9!Ff)4HRo`Kc&p+&lQp;LZ+{?E)tsWKh@qV zMDOJ*_kb%b1e!yl5yQz2<+cDjW#FEXot{vol>3EjOe0XA;qX9L_juCfJ&_5-n1&gW zDu<}2kvEmP$_xxuMZ`P-b+LW?3g)$M^+noLJ7( z{o;X#G1tEXM_rwYG^q6@Fm(a+MGDv`Kx)gK&Mogeatar94(UlQ0WM*u4r@@IWiOMr zkW}gK6QI7tLQU2yd32Nu;^zRv7Dz|qFn7vH!}7+WA!WnK>uK35mLCtn@RnXr{c3$L z&bmISBW8(rxc$8?=r_Gqy}#WQ)3UCYA}0?XU6w%~PyRakKeE3qw5#ienx_Xn&$b-g z@XDN1(;<6T;9nfIzZ+-XVC3dqbpK@`GtVM074`=I%we`$N8;TNV5fWav}aei7E1uG zyF%ZI1Bq9Kun!>zunxMQFDt5TjSTaj|J{XiIBAxm2f-$aO8y875hmiIjbmKSIFyHV7RI_Yz zT!t>7;)mJ}+Q7ozgQjIBj2t1o&8x_7I5dz)4X@kTftu)l&sZ2 zT~@QWALF1n(d^8p0kdK7?M+P~#9+dM1YXFZ($0A$fBxxGXcS}TqHNz*JhC6#eMIG| zWgz(hJ-DCRv*$ zAs*(ypz;5KN+=PHYHlgd#yKRD!sqhqPf8{J6p~EWav#~5?nwqqH8n|m$B9J5xCec5QEG~#Y)~VH|L9qE zrD+IO)EL8rEUPWV6;aKO%|6xz5H;SCLHyM}h8l57ZhX9v8FtS*Afr-((oa5>9DLz5 z9J=i41UsMRdbV_^CxcJ&r2~MAfjaKPIC8X&~na;1CtO+^880AVPz{a0V>u@2oadXUm+hk zKkPgGZl=dL+VAM@jEICoO?7(Q23SCEbHI*)PSb$V$1mkTG%nMH3O?sa4<@i@ z7}@w8_=&=0`BD^hE1wqMe zD{S3VZpsX-_hPZPk<8^ZS_hCHW>Q*p5t;r)xrl$_a9-WCQixhN9tP)b=+Nl_j#`8{ z9Q-5+^4^>DewV>BuILcZ7-IC0V0VFoF4|vQ(xqN3stbV{WG=h72zNt&I@1CkQa2RM z?^Ka4AF;FLcbA#WuV4;qpt*(Qd->`pzAr#z$(OlMqr$&!q(WJ_kcB4~{5d%##|+~f zAkxwy%yt3?bI})z&(zb4rvMvU@l8&q3Z0*AOe}j)A(h{oFg=y_38iEd`@*==u)o4w z==A!KO~wHDXy|zgah|Q6^xaU{_0^7RfFc~~{Fg0h3>1&NO8(*X0xT{Rwmc2e9cfTU zMkWwGN!lac{fg7QaRrmk+d%yJFoHGyneZq8P2>ccWhxS0@!HIS*7M*alN`_!=+Z&Y z8`23TMXqm5*jSCtW2y6^qxc~*iW{I2I?}ExFtI*CY;Th4CDarnesQn)j^60k!&k!y ze?2Zv34!%Zp;sVMdM7fr?;>o7AfIp1XMS3ZIt9VGG40wMN2+|)`+{_K_B%*PsJOy_oOy zuVB;t43`&voD!hAS};M=<)4JMi|EvQh0itTsl@>1?^k~Z*eZh{MUQjzwU1sgX8So8 z|BbA;9hl*n5n`7wObgy zYxy8J1(n~sB`rJvmPkTH5>(D~)-QiY&OhD`vY;;nOGJL15B$#JSR)T{dx4C5?^r@n zK8%lkvO3mL*edMLZ`!;s;?p8A-Z_Q^_)l&&|0|~xtz4{rqu-ofi@WrK5fJSTb^DL( zLq>A9TQXpqG(#4=w7`;Tzk&7pHbW$zsSs7dp{*tuaRs%^oK#mt#8<2fHkj4bcZUj=I zv14aJUV#5RQ1$;4>GB((CjZcdOD4B1F#r8m{-!TbxgOQ5Kh#OPZ;g>JDz9P0K%S+z zn_*`gy$;N-&czw zVmJjsT1yDZBk)IxTw+c~lrk>yJVn-j(uhc;-82wPBe*@Bb@~hM45r-tPU}2@{L)mO zpRf!ccr`eXJ`1{?)sgt;tDTOE)mEh1gpUgA@Y6`eu2Ic+c#y}CALWIqXRLyBSNzd9 zkuKW&q0&Zx{P}%iu^1Eba_H0Y8^=yC`Zma8H(ptz21uB{tez5W_@M`YTy{+bcAt~XOLZD0qO=Xj{D0L( z(l#OUnGlAx=y~q{=Gy>)$Q2&vtKkP61dWVOA$NYiC#oYH_9giqFmfSZzSOq=?}h~C z;hyNA$_Z_c3ZwQ^l-Wc;x8C1^V2|*LP9C|5SH{kC;az)tFid$WPi)Eg{+PY zhhN_Y%*_Kx9kCol8FXys8Ju!5z74Wb3X5VrSB{T(qr?lzC2V8p(>&QzCMe_STF?me z*1N2PxD3_HI&u`ujH1)R9{!1Bs>akKjuJQxv+2`kFb^|iDt3Z(qpoStu*#< zK3STS$~(&+gbOEGe%CiD!xewdVcw~*9^rY)Q8&a^Bbu5ha!d48*!CHr zcIayDwXjBVc3Plt3{F*u{5~JOfCzoyt#O7<4Y^~$qyl_jX8_`AP%2oj_$(CutF}Fx3Tj&t9z5z6A`@Beqwit^)f&Zq*Gt%X$BkB_1UhJ|P50GI}%b8B!?_7t-bTyu16%#Y|yH7?6cvnJ-p_9s>>HG9~xHZL& zf_;Fi4>lK}Wo_Ul8nA}zrmVqS8~eAP;WJtLk8g&Ij8jvdH<(95Dpx}oOau*?nSa^T zo%$D2&%T|*7(P|rtz9fkoAKozzMZg0cdrE8I21Da;9qbr)FJeol}^FWtf-Lz?q5Cp zst>6}Tqf1|XW6-%14P6eO8ELff{rb)ly~pMc?q23dFIax$uVdzzD`%C`x zJ~#`}jLy?ar@nI-zqHu??%QdGS2gXGpnx(gyUANWu1{q2hvVg1j}d;99XHNw)6FR- ze<5F|Lc=UzWGinfGuwnhdoLnCjU66UUU?n*H*?!g9gL3UxgH6dZlYc*Ef#cvD z;#O8~43ck>q7q5mpRce@>VifkGmN62@C^OfgeolM$cY0#2p)clc-<1sn9l;}Ok6Nn zR3wCA=b7Hq|FY2Vf8BpvtU@2FDb*#n?&^qfhF_r*tfYlSwHkC1Hc0t!eOd(a>jMv6 z6PRj|A`Jr$Scl2gVqAR99++w3aIU|0V%kV=N6$-)iPRoGBA@YV3G z487Br`c0+O!1%Yp7;>Q6Xj9EVF&>K)zUpJN`OCZBBll(6BN zf;}i$H67;zxW?NIU?;N5S1E!H%>(+Susi@$r9g6@UFF7%1v!2G;Mreu%p*4oP`bFv zWldy2DuDrfdB{xh5fD|7V8TelzYhIW9wd!Wk}!S`uF2Gg@^lvRv+bJV-uZ<|daQvI zC**xKe0M&x@wvX3|1bZgW08&Vfd>)5A_qcmlIa%CVY*@bB@{I7MOAE#5gL$!`L`~ zYw4eeIc`F8j>H}f?ozVH7fk=YKQO*2>_s1U{~h{YE^}m?%U02WN8jWXe*(2{Q7zp` z(GV`@=lXVsB5GU2mH41LYZhf3BuvkewyXjmiwcqz{Gf_moH8athZJH9@tSgU*hnr> zXP}3r*aXmOgbnCa*Nt@&!XJoVlCf*CXA<|(zbK2o?g-DL(NRqsR2%DmEemD{=+Bsi z%k0|fvvUOh$19XP{*Neo_szeViFF8GRC|3a#LR75sjfHs608$tFYR)@RIL?HhLk-t)fJ3hA64*9j7%=~WEpSW*9cnxXqojU6ec7FLs z@XL5aRS+p?@lR{KVl4Zby=qj?M$CKhjn^3MS)H0<^(QBvQ%qvlV#2;ktvz2&>tzH= z(*mc{iU9$7cB-OSsSPxMibatY$?>pvswM$sc>4X!f@MjrBB#>Y~Q$Dtbe zH*Uq8A90}a#>}qmcYJ;jL@WI|H$}}S2Om}RIbd&nqHM0|O5brgP6sF)Xp)2#9~E|Rf;jW9qbVicQLw`J#FeR#+gHYXGPcMQyT^b1x)Lk-ZKp6{ ziF4KHIsf*9iV9nC_2lN7rWobBT69+g+jmn8u$!m$?XEkbUQ+J`9nu(|!rJ6>GT->v zY-jS3+${TZe{ns#kMQpH1uGJsLW9=^9g0)H1j*O(jLv4dRqn{^e;QYrMbUxNWM6x> zSjYG#c7|ucZa03##XknE1H>sp?QIIGZ&Q{rMfCPayRs)9AMN{@u=`$Xf}VA(Vgh!5jZq(-A&2 z{8;=ICESQwwWFh5qOEj-RUbV!H@*^`xb-PdYEo77&DCct35AhU{CsV%hMnN;)Z(9gU(;`>}HMu8~vL&oVyjregFzB@Ek z{s#oxv-baHkKXCyMYKU|MNX>6YayM ziH9|V*JtV$ha!e+AP$ULhHO8<8dO3!^xNGXmc%;?(jO%+-i^S(MyGV%I79z3GFnKl zAFYd)*2$i|odq$Ga}N^9TuO3YKb5bfu3&IRIy}|GoWj=)`;ZBiBD#%_zK)*1K(LM7r^ai1}| z=E2FLY$w5_*Rg$91RlqyYCJ9srIC}x*_GcRZF%OhGMyFJ;^yd@Me~{@npj;Jt8W-l zSVM{568-0H+o8bmFJ=~Mw%r!k8!ub)j!wIOAjrpUFfBg}(x4EzJSVL2@eX!MYjLI7 zB&&KQn0W`AN^;?V!3kd^5yD^CP>Jz2mNfzeksswus|!O$er3HqURbc!wl00wQBrHrkO36gIVwH|8 zWOb7Cu<6wn7#jafPAqD8rK9x;*aN5C;!K_85@gUh0k5o;0c1B(_UZ(zwe81Ich=F0 zw_uMn5~Kp{^}K|G!LR#-x32-kg~{O0lMs68*=O7+b-XLa52ZE1x;D-qQT1=brFkejqK5jlh_s~*97VVRom;^Kgwek{*L^kcFC2dwCPwXV9*iJJwYmd z2&wyry$s;Rr7|yPZ}53O6+FY zQ|ns=Gu=B+7y;B%=(ww8GR+ji8Xtqob7HRdlo9oLm!zbueAY>h>FN`sT@d8QI{j(m z>>qyZD-U9$hPhsF-5tzbB>zfX6bGJI4(H>dc$T(>iGd=~L1nX_(Ysakg5_T3Zy~g+ z&D??n|Ba1O5+ni$Zki08=)f?w!;A8CO!OwYHD00-~LLE zp9WBke>2hbxdIoEJP|zs{UWhL!!5AWv6q;$65(N!ZG!FOq&O>ipE%F0>HA#)#pZ$1 za-i-#@^@}uCN^PatXsNxkztWViwL}TrNUcHJ z0ytzV08SyZ%XML%dzhbd{g0sQfBf$pJ)gJ!uAl0lH7i~pgLs|>T|!@YytxE%0=rlz zYZTvr#I%<C+2#_7gU1p zVOVuP6MDgZlk{3M@aN-Wsb=zW_;v>!S{iBPXuv{u=HDEdg8j|le%DHXbz%Mlf@Y=xD7(E1sqaAVkyrJ^^nOMpCw+4izaWKkKCd zn`sIulYCh?&K5g1G@gys=15{RO=BDYH*X?3!*=4#e9<%bp8Q+tH*%aBc5cQkOrscX zfA4tVXG|!@IpK+qXT%bsL68u%Etbwr&KJTXc`n6+$SD%1?moI(Q`38Jkv>?$z86we ztk9j)PoF33bup^8_r4d2u7MWwFdQhuYFy*=2@7sebO;6XF5j@3^t%5r$(-B zeGD{geXH%(F%oUI=YQJUx&{hK9EItr`*{-bw#bNuGalXN0!uV@>J9&rH2WPYfo)k+ zCpnyNIeUc%6LW&L2RQ7aj3SNKMmT*@qbTkpX8-S2XHfgt_9|@P)oDk1-d0E#yip95 z)B?O_)Rbp1HNx?(I4#?E4qa3qQwy5Jyt8gsq(DsVQ}o#caQA_;*KvwZ_UgxyqGC*~ z=txLim^wA9;I}&*7v`Pd*M-#-laR_$N`8aY;kl0bS;+S;Zv{C7^M@?dA!+Vq&JH+c zucJ<#*}*KT-Gosi3-yyY=8w|FDHP4&vX8g<*-B!usqj)>hF3zIm(~x|PkKav=>96j z6eO;meX4juKF<1U&t73NEDn=x>1z>HB|v=+dyvR_y#D+-v1xbm@@epD^Q+yjP~`Te z%FIDW{6M}P2Y)Q8wEY36i?HrYO1$LB;U2T&EI_vd_NYOb#^CoG z-I~Ve1J;{*G-v8zM2{%*RqcBvu)G@0a<_{np(MftdAu8%oxT5luOnHyd+<+pHx?Ti zJ(^IExaX}Eg`mSA2!=`2mOj9Z2^>8?x^88FY|JbV->dV;f!c=K9j!;n11eRJluIu> zlCxdh24vF?I+Bxo2g#A863DsFIf+BjQd;rXfb*4|Sus!^dkvVW+|*`eLC@3I5Y|m? zF%wJOQH4X_$uzX^@j1DD2s8ZG+;J7mUmZ{Q=ugOh zU%PZ(2L=}*Lr}oG&^M9{*cAz40UAM1L+J4Zbv>Vht6xe$R&k`G zE`g-GAv`a@H&T(9m;N1*>|D6rn7r4GiEK(+4_`GFmAMdn*iKZe%Bk+#59Uzm2^vXv z+_P=(J)PS`zc6t%UWV2IekoA{=okN*t*5nb&yrjmE{ub!K`UD*_KFu{iq0fbnYVG3 zrc|OuggBr484@xRhD4u=9G84Xiu6DmA2uEhQ`KsiHj%;Tj}!TPnZ`#XpgXgNULVVu zKV#_MeOK1zUz0md_EimX{xWUA*?xXI2MA zly67eMnUn+2$q=MlF*vM<_{p0!DX{j-=XE~ak||8%Ms6}pY@n&*a;o{`6IxBYKt9n+87BsMn#C5*yE4`ijh9_!#l1ron+MePwe}fHT)fStcTuU! z6&OOibd#SwrRXHx2Q*3<*~XkB&gN~SifOW1yw{x1HO^g$`pyN~b?zulDj8f~i;|P^ zZ+_P%9{mSS<#VMd23_TeerLoGjM?mFynahKYN@Iwc}1Ip=60#E{*HLmSo-K$1Wnc- zC8EB5G?HA#cwZe-Vv4y?6H4LMb{bzB8K#6~qF{aLKi=sQ#4W#V0AzswtKB<^EQw~S z{STTyBlIW+_du3kaO?R1#g1x_dN*(&*ze)Rlbs&D@!@$);AiK0-Vd=9se5*0k9sf<+ zUnQr5iGk}Ov1F7{svnv8D63ba1^sP~u++F1U?Riy4nYHs_}9ZAdRODSuXdFf3FD6L zK7B@PhG-)M@A{=rAFfdNqHS^}dUj8329S5X|AIUxO|5goR=hHH759QZQ&3uVu{xcb z3a+I;U4oHbx)5)yd7DoBwc-|>vZ(&DhgP%Aa8-zAn3WI}`P>C+|5w9u-I9*lk0MAT z`ZQG38?670ecc_(>@Wc)j~PrEv0p?h;>=FUAb2sv&D~Cg_HFf&+&i5BWTF>VQK^4= zxz@95$^Jg}ZmXXm15b2sB*wIhl^s9w{isqwzHO8(nvp2E7GW(($$PXLJeYg}94VC-(4`}6Am6D;i zqB;wVsqMkm$(51A3vXu5)RmVQsV)|L2A6xPGcm-&LI?8|=6>$yi(@On!e!&c-+t?( z?^g1z9q*S_=S*kdTD6VxIjjsFj-0z1|HW54SnM$yh?Wy;x-5>9(Jei)vf(Pz9$Cdh zwEtpD!`p89YCM2Aio?UO34%`Iu_R(c+0brA(;9BtFstdNw&%F?-V!kA1FJ+91knkx+a7@pVpi|?rpoP zLS&S76NymKRT~wHh?HIo#M~utfK{rwbW*x+>0VTr<<7h>g=#Z&a}DRE@NtVExk1XI zVTH@LJSvpSkLmlOxfS%SB?(Wu81JoC?mB6kO^cIww07MqjdTXr*mYZx?3utLiaSY8 z+@;ug-$C2ida8ICXmN*-R#k-F4$(VGA;F*Cmb;1CQ~7Ux(S7?vl~og%U4cYvs-S^j zIwof2enDHx%jM4UlD{ELiNImVvDkESj8tyVWiRQDU8t-dEMTEA7nvT{$oeRNwRUB2 zqYNv)DZ+`I-jv*cIx)1`)4!K#miDf43Kz$k9v5|#LrA;qYH-v@ujMelcwFx+Q5Wfy zOBxM`Cq%hkU|g{bd!ujX(Zae80Aa}R>lY5@0kRVAexLD{rU z^&{IqzC9wRBVPuyj^9BnK_6WF{&~-wYP55GNz9~h+fEP+46o+g_mLqSr1Hm)j_Bg# zj!U*)0ggP~ybwYapM6K9)B$74@CUZLNO?eyU5&fbFM7-t2G zB+G|UXC*7sv7`3HQK&~EItbC5cx*&y;=TSYK$Y{MrV|Y?gXV@Ozsgyf z10#1bPqiKp^2AESzI%NarF(Ubt53_%4}MKNgLura#^Db2fa_kTN)>FfD|Ea8t?$-Xx9V2y zA3fbW-PO}ipXZ$Er+b#4nQhDjYsC}M&;3Rr zr0{>``}I=(176b~SNG^vCMqg+(`0G2D0hC=Z;TmiA0`sFV$zM2SU7((IjC>TSF(eUMpIel^N;BWJQNR8<5>@YRYX3oAWLTmsfI+@kh=OFv;aw9 z0Ed%^^CK|$33-X@>PzQg?!chM8gb>G_l^#DLrF7jo4^N`pbpTOqnR$$R9NOmC;bjY zzWx%350pC5Ow+tldrQ~$Yb(*ScYx)D&ee1m-2PQj#sRnLRT3Hl{)%rBIf<9fbH=N z=2~v1vWc#FeE3cb0UvHN)t48}dH`DgswzKvACYXs4D8J9I*O2CuP9RU$vQZ!P=jw= z5EpgtdYI&+z&jdJI*h1@%^rK0m6h(^D%KHQJ)s@pT5iKx!qD~7ETuE5e7Gmra{fm~ zN;AE)8r8k`Yin#8rqYl#e46*SLoq7y0=Sto9>+taJ;moDOxYkJ$IwrxqIQsiH+7L! z;U!<^Uci|z2ok!>3J>sDk;UWJQOb>@LkYLs@;xu?Cga~%W}-b`^zu!J3H%C3642^) zeHrN`AvL|?AJ@PH?jrY?60R2Q=setiL<_J`=SM7K(Y2E96XOiFHtx-tai|dy!3s{{!lpjQl)HpA+h13nsBP3k(h~!!1)+T*i$I}F_-vI|Yvb~T(3#5Sa0CWs{ zW#FAdP~e@k{F>(Sf!`_Pu1gkMgui()T5t{$@uyl?6v!B_2?>Xz|AN_1!UCM1?Sez7 z`K*v3LV0KDW;xt41)&%uAKJ=lbQw&mav%OR4K&L~FnQ}5qp?k?)lXf&Z8 zKJN?5{pSEOEj6Ax12|3n-Gv_*1wJPcb^u&G$-+vaI?OA=v;JcKCWd*Yc?M;0I&L@Ur@1 zuzUfqgPR?B`3h3K-qehG5H#5|G1^WeBfEat&Ic;ylG^h5h*yqEwc+wD|4}D~PyT_B z=Hay^cFSVjIIJY#xS@&HY2}X>#b>#Kc&Tn8oSHda=`{>ROhsQ>1u85wC2tJVp2<4=w_*?Fe$AK(ap60K~q!#A_DBTRKeOdG&!kqw!yn0V3Fr5*5=DkFopO z;AiHwz*8e6EnfQkjR3mxh4#;D+UV?vW$sN?q`+erT$dSYO}OF+{541=W!;8c_q9S91$;-lz?$_JyXc|6w`YzpL*l8_R+95`JqhG! zklG*Ol2jNv)a`;_Rv$U(K!mWTmqnwOh;p=+YB+;Exr>kVwn@IsL*;S!I1mZgyF=l4gx~`E5f1ZXm5*|*0(h@g!s{^ z6lA{?Kff)EaAI{acmGH<7zzb3R5UCAM?kp0Lp_LFm68h&C=IQf%PXLzw|2>8%}mSC zUg+C#re4K^*au$SBCdu_fB_I&YM}x$j2DI^Sl4G`OP@*}_@i+Ftm{(={EGd57%N@i@>CN513?zz+?Ic zAJFKmN5s|1ILJYKRJv^t3vXg&uAiD@p46=X7A9`Jc-dVf6aDlowEL|+ zeHs8iZ)P+gRyO7x#tzuV*327dz;=3XwExcy#&q*Ej98*)tO%PRKSHsC$tJs`y3p3r zw&-i~R!aiz?4Vb|qgy{sY6A{^AcWjo+Tb0=?7kcgXh^$%p+m}io zUv4m#vMxZikKmIxe?vykV5zB$oMoB$vpCnMHnZCk&^6@2wRyg@0{eR1MRR?W@Hr>90_sb@=W{k7)kR4qMNV>BE?+Izap;i;JS zpho+Sko81(MY|*VtdVXX84t!&`+;&rz~hCD!mo8;wgA{jV$@=PoR=xW|wenxq;a|6}Nm5&u2ig zY5;}^hx4ya3zE~Ja7s*d48&#rRt)1>u0Tq)2Ygp&&RSVRBpB7jR&i!k{N;G!)@z_# zEqsexuqjyp(2hph(@dMtW&MM#x7s}s84Ky2c0NM`T)hlN%Ag*j_u+v$SLQK^)23(u zhY3&|rxvrqBjF>n|0Oj55~{^|U-PU*79^AQfVdS=&m@TI-(yPM>*mR0>aq?`O1d^@ zy*pC|t7zf{QM>Q(2EnxT?`y!dTc+| zv=LDJ8_$uJ^b6l{cS>4*b~vMM`XA3snvp3W(IWhYlf{x+E;T}=NDd)@($5}yWZnH! zn-hr=TBfn_LUyJ~2hsl#(0_xN!FChHc zj(qWqR?vcI@r1hL^YbjW(4+>~Bew_HTc$<%tQ&=WT_s(W!Q#%2}LXjqh3h`palq zb)LuGxt6!E=-BU0L^aO3Z>-W6?cXr@nF7%P7(~3LOE&Tid)06g=R@4!`CdJLoLw|F zoNNdAhNO`0yW^?>{-J#stw?%p<~m2CQGt;z#0@iAn}KFtZo22tZQrC#^_}_iOsP>K zD(F1qk_HG@YA=D62p8;knUJa#uGGX}#;Hxormg;#voKPa$=<{GZ7GPQ?v9@yhw&_K z=I$wb;iEtP08_J4`_lU2U+KeF@LlqG@_g9e^kJ69a^|UmX~+q7D?D}H&x6*yNr@|J ztDjn0hoi5*g+yJ9dK|w5(_ypfGi?*t5!{}KVp~W{K#0MPm=Qi0uq-P1ak{Ibkl`Yb zBY)^mEG_Y&8Hv;TnKDe+lb)mcqkol$E|&6$LQ!wv_a!6Y#Z)COp0A<}yTQ;a9gC$T z(beP>p-j`hd*ZXK-Db#IRWnVlR2IpUAOd$#_{VAGR5j09E?4E3tjHGD>^+aSM4qEdgN}^@6Q>6*1(Z4S{V}YuOLaewOScs%X45suBnzLP_6b#~2z*bt@ zwOdBMZ7nz~U$8G@UAVF@6ei5A!|~;#zmKzdcy5jvUOEh-*l#_hg?RBVVSY0lTC!#s z69oB)ZCjrGR>xFakab;^0Fy;SlONxlW%?Y2hat_nzwwT3<`7`lm# zp&Aa`5&Pvw5%q$1X=Xf{8M# z=OEynI9lIUGXxSsjh6OtFjXOjacjpDXX*6^dbPsTS27B+FN{jp)VPKr#p1{*k?-6t zOz{W*(q;;J*9#-3Ah%<0qAhbxte@5HTHpx29~j1Ie;tP@Jmw#;8wp)3i2N2aWol2l z=zS>`M)p0E<~`W)YS60t%GAF5tL-cD9@S8$&DiBH|jQKW&>X@Hcom(!yv#n)A zM3*TO1FF!Y9B<+fgOU)}?aP6uWjO%_dWyArlu-HCej7w+l8$z^Oz;E zrOUOSjFjAXS}+WJ-nlg81(6G|$WhZC)ud1xNY3SmCIo$MPK1ku=e#2pbS8Y%X{8>o z!!}0cFf+K?5H^j*26VzLh?^B=0Or1Efuyv*+2{z8i*nN?>FTrfJH>DfH4=vuvOlFX zg=8-->053VnLVTPN?sJeSlGM>flxYI;F zskZ=SF8%DOWgyl7;imXfC^MN@z0Fb#E?5PugQbCuVHG*s`3&_ewGw0Y2CmVZ4 z^)71G^M9w{2-%{e+D{qqQe2;IP8x&XR6uB+q1)f|3trEP0Wb>h8qxF5VXDP;h<-(g zDLJAL0wWedY;J`<)C6a_jlE(90V!v0p9d7Oh_ts;;w_Pv32UW0@*8OhN8;u&2hCIg zyo_;+aqZUfFSC4k1Uyie>s^0ur@GJO6&r8}Htz^fjgKxuRia8P)&&|?q1+F}rCre> znBIZV3GY#n%;Y8@^XA>a3t&%-9t+XW%>BmH2!G`vkQQ29XJylvM3zvHQ_245%F#C? zSQo**w;UYK{Z@IF2p)K5?7u!K_w^>Zfn?bQIisJPJ!R@qn5xWzTK63Y)h_l+av3rq zW!xyYO++aNkweHZ9NTqHTGA8{g`g?-MT=-|ptMT##TlVKuZVrJkZhrtE>kr}_RUsO z;L~HJ%yO$80;8!A+LrLQHf9ZTLTvFre7~(aIg)*e&y^oXJ*v?&>TweBt>ThiigA?N zQDldhn-{`BXH{^t1CH}kQdYccHP>ih%{1r-CroHN+<~h`=;V&de8Q}Lns9t*Op0nS zNV}vT@6V63iPar|6d!Tx$4t?zaU4dvmQB~3WM4HX6t*&T^(nbg2!1%Bq&M#FKe4A4 zzYris9@oB6ScvdLn09Zj&N{BV5OR6tx%3kw4aWF+T5s z{07(YJr_T|dI0NAl)lb){a7oaFI%M#9p zfEk@FuNh$r{nSV*MOCb1mBq~*#LB^1d@a{Dg{H8Td22e_nx{R&%eF=amSh15dsht` zB91e7Uo!$!Ci08` zR!-irl%Y_B(&2V?^ka3PiL_^CQZJqKLG z#0-($^EG-eTBX9&yR%o{j@z3m2|rNBX(XSF@Efn!c)?hhw!rk;BsO?d-!xoufvw^` ztYqfK^U?_b_pyC}bM$gGQ1mZm?Itk$dU0?i%@j!}ti2b33Aa)8hmScP34S$Xg-{#U z9IvDcq986l!A`nBj+red-Dr=eC^fja>qO#$OfJHvWAvWxf(4}Ec|kP@o18Q@n!-2A z7sM9EmRme@Zeg@QsWHscnQ<^2xV-YQoE%|!Fpl+vvd1EJW7O7GM>m11QW|FO#VWK5 z)0Q=QTUek6%Ck6nvjl?r-zv|5BJD}ZClMyuRhvEw8B`f|FnB06TEA+e7P!FE@bweWyzkX?eCCs^PST_V(H)qEXxKHXZNE2V?}t<;0hc5nzV|Op zjLVa>wp~EQqn%DEYtg!z*l*gE(40ow>M?T|a@}yR^SYKLgx4`$;|ThOQ5lZ|@l;)9K=nXbU zxt!$jsG{n&5|>^=s-0GL%?~FPzO!82uMF>P9@B+JF=R^rxTB4%1EbCl*!R8^D zzbejiF}|=t8lQM8@oUO)GRE$%%Ad=nAg=t9qEU=i zr)x>wlTc{CeHc|^$m(oQ8`Enr#q+b5peMcQoEHPfdcG0AdiV;o^;R{;6OCnt<+&3q z^fb<(3?uGGKMEjOFmHmDN};NU&M0Z8`lI zYNQp?{5K+{C*d@#^u3J33(qX?7$$1Re(64v#?Q)ET7gni`&d#aNf7s(B-xiCX?`oC zDZk+vo<-I$9ZsrWb`;{_sG(i8e_>P%<5TSEhB(U0pG1i{K2f7Or$;)S95oQ|WCx;A z_x$;pha#cP3b0RU>A{~%V-qdpW+GFn`=If?q1E$*8dw>`4WZLN6EGlxa=-0>Q6RrfBOcS66@cyUO}W50Wd2@#?$OX+&zFz#_}4J_%4V-_^O*}0 z?b`Fj7f*jQZhlH4AcqO7gO>-tlB)5F;+!J=E}^&4PBys_yYrJa(4<{O%;=Ln_enU@ zKG-vr=4{eSR_);WtX+dc(3mf9#-hAD&;^*o+857$UrbM$h~c=dCjYfOLASYwFD;u? z8%1Vh=eQ8}td}4xt+R0CPxaW_ylbBfx=nl;>^M)`uUdA&)Z4@ssNG4wq{$gAcYs+T zNP85dPd)0+5pfU&uUPrh_{Sm>a35|=Uc8%Z8i%_7a9CemN3NX>Wh8U4zG1Q?Ai<>n(lJ!Z9$D^N5ZtDkn(xs5~(jK%Aa}k}uI!OY=K%v2~m!xNeo;T6!xdftF zANzqX0cf}8j^Fhbp!h+X{yn`aw!GUI+>Sw~R1$@)_>EB6fCap;Gfh+Ry7rUzs{zEb=;SbP1*-0gf-TBA}94ga=Kbg zUqpWsOeU-mivN~HgC&KdcXjbHgl!Mq6qB|#PAn4AWo;PVetMf9fg85_RvDqr{0ls@ z^@uf!h`@UOA!trnv@`Q>c-p<;L3g79_>s+6Zee1grc?`7L}p(ocChX5D9K17zpq+o z<^;C^Pq!TB{g`55wvAfHl6&HtqKTPso94YRcUC-jM77{z@kpw=2GQ9X>8?$VpcPRg zZQxM3&<;4AW7^C9Clp1r#!-Tajz{XSy^i^jA;0U=yMFHDi#LzH+emwU@#fcSxd({c z{T3itAUPS?ISu(q^pfZAad#P(4G3_a>sE{1Sp0Fp=da-KL^xGPQ%%5dd za@+h0^v^WQAvF`Mx^I@~1C`uJ*&o|yA?ba2tknAZ|B4axy~Xk$Ddx-N9@|rZ*vnjr zxP{~HWdanAwF2Ni8oK9H2I$mHRiB;xd2QK|WsYV3N=~}NKL9OIYa20K#+s4^JrHtm z(o+Pk4>6**k%t7PZAAMBa!KUyEy1#El>>J4Byaz zb?*?lr`P5N8vycapw@Z6Td3f1cLRCW`xt7kOiP7<1YaHzwAj-C``zt|%UsL1R`|ZTO#fm!e?g^GHm0#7; z&OWTaUk2cm?Z^Fmhh8Hjx%ryrG!R;ip?0RJnT0)vZJb0c_`!~;y5|HM*TcS~JgDS$ zz|Gz~+~BsS&Fv|x_TZ%Z>U@%?OjiHuG{r|g_Uw7Kx$+oa=EE;&SA28&GdK{+08sWY zFXbKI@4-p}fdrqAlxy^TJdh9Nb=}fyxudtAYyzfp#SycybozjItNL4thrAj z{$H6J2c|lY-7?DJ+rbdF?Ty^YJwEL)U~>QksGaRr3t=<={k&nW8;$9}54MS}zVZ5r zYQzK@-fGGmk80p?htM|!Stqh##>e32-MkWR#0_1%+`2e02()aY@%6;dfwDc$^Lfjw z9R;s>=RTGra2wAFCwcAA6ESbiQHg(SzJ8EWyz$M!AxstlTH~tSx6P%~akzHDcCOc_ z{O5!{amKgi$tPMgAM=nGVmHG z=Z3UXA?tiNl7xVdRnQt%NM7dD3B>?6O>pz>0?EbIuf=9LVj%vULjI z8lwy@~KZf0{SA4zHdWY+5kc~*o|Dv zRnhH18mc0KaOJf@y}_gJAKM;oO#w|g!W(jP$fueSq&;_q_txG&GW^UR(a?#>H><8y z_qV5Tuu?Rf{$yCps-?-R=+RL5k&-`cS1&F{I!3L zp`O-DyEw~Q2C7zXX)R>WOJQ>cacn@bLE4==h~00tE5XfH=g?@aAzfC;(mNg?LJ^h} zdpD_yhrQ?OKtprdyKStVCwh9+P%hAD`rl?6T%!r-);$1cw3R{%;zD&>43O03xwrnB)PXPesX{FKU zUp69k9wkE32MR~2(Trm`%zW5d5EI;|0D+l!oXYC-W!>3M95ibw5s5RT!ATIr;>OIo+ z%$;?8?Y}gf#nXYtbc0k6=d(mit(D7IP=a11aaS~EhogXnKSqTf&S4(!PvVV8qq5L` zKAspA)rj ziQ@h9<)tEO%(MD6qhC@I=Ge+ZJ1fa!s7k5MJY5tas^KcE%W-xi>=7YYs7OZPpwr6| zqWX=Wt6XtgjyM5tSnEy(Oe9c#{thh+PV9XMrwgSLyvo~E>CyJl1u-PeWy`7NVXAkL zA1;NzrRDjU8PPIoIQs25zr~XU7M;0m-cEAH!{5ww`-wJV`mRv%)hkdoqkoFtOp{@5 z35ZznZ!Nc!eBSIX{rU*I=S76#J*-2Z+!7BYHaR7D?5pp ztf^kQ7w1VNPH=u|z8rNU*)ZE%t?`nw*(8{NgZJZhk`8;C0z=#TT33~FTs2gu#Nd~d z@vox9D=gpDvpD>VdDm7n8^wB_5!}Xgara2?fkOHGCnDXAl<%;vja5=MhN_7gBeLw1 z`=k|Huy_ND@I$x$|y)Znu6>$?)g4qi>>-SNh5+Hma6~ zZN(n$AFNi4Oqf7xUU+hh644HxnKft;f5sX^azMV)7%wD2G*$GLL;Bb|bmLkdG|cSnBz?xJo>2%#6Mi?N(n z!P9E6Cs9P+0&*`pG9Vn+@={5 z?VX)ZbLB1Zw435(fR-=jsY?_(OM|rz6=;CZ_AGLwA532RDmy9uuCO)&e&9X-Aj zhnR@hl`)FwfoO{L>J9iOV&51mz2)E9hop4kJh3Am$&+qUn<(QAL$y_;eIVQ`dawD9 zy86n{G^>~3TH4ZNbwB4WYXlR$unx*;x7-yxHu=Qi3rz&fv?JVBvOCJERilq z?d4&m(oG-ZL zy}Quo_wbCGVL-Mq1@pN4Fm*lLApTi+EB+{6W_iLmTe+TOV#eCve>VyG3g<1D8jz+? zoiyd{p8$n6=K??T0|&VOPsl6^b`@mNhTC+}!%Cf6mMlFbO5i!*=P~*iLI^M2lqSM} zv%f8z33!@mbWykx>U0&aQr~jFnnK_O9;)h$&WZz^iVe8iZCk%G0q;DVAFfZXs*JdL z3}Lg&s=C0_%D)TFeJ;&8$7=WoVPpI49T`t+{h6k4_0WIcc^YG#c+6HFp5JLi2kZi}q(FVE4`D5yj27RUe8ILsX` z-QAcm%j#A5{$<o)BRBc-a4xR}wm(w3_t}00VaiN%i#SlkQSDNXpUl zGw8b#1q0pDtiBow3)L;lgFX0C#G=7Pt`G|F?tv>Q30tk&3wLAnw|eq&mJ+>|F>UK zK7sL51F5)sv_oSIhm15<%X zGxK8lE!NO}H&nQd;Gh-K0=s4Kapl|C2=k{*;Ed!d%I%2sAhgDs6^J0OP@%$0Ktsrn zx{QV`BCdRjNb<55v^v(BI>zM3&YJ98MjiJr55vOl{akHN8;7MhNJ_h#jQL6xa++q>xA(!*J;#>s8P_e&&x#f}tXtj>oAAx{@uYESQQ?f$4VCRsRX0HG0 z+*Qg?lBi=JYF;?0oZ)}y0hfrbl9tB!JM!=l5g{BQv7(`1An0aP zshBm|gntZSLu%q_gp&4v&F{dov_XxZP?0Et3`|yolNe1!EZ2Qy#0K}g8HioB)~b`W z);2Gla~4B6Z)8o)9%nDB#vde?68mi>>(dP7U)R5UpWAG`0bC8Qs?|Nt&)*T+VgqeC zloz=uO{vXpCr!aPfMhoT#>S(s_hc+jYg4i1!v=!=Q2{%TgtcX0bOZSILG@-1{|}G& zTh6azZy(kBtjW9Ey!gxd&~=mEFO%JWM@8Q0-TX~~Gb5)QK6Qi6CjUjZ7y0Sq*vE*U zv^T{t34f&eZ^d50H!BeAblYfNByHJcjl4;=>dPN;F{X#=9Gcf&6q?dA-J+<7a zho7tUE4{+tAE68!OhRXxVlw;WUIi;}qZVx8OwYF322^g$F(+DM;k)c_Yc2jlkGqCq zp-$SC4V$KM8KFm5h|T;?8vQ;}oNa2)@Du-|>nZ|}uVh~4L7v+3nQo+raO$#-#qYZ7V57 zSp{Y2Eea;`vh>qi$E18gpUmt8~DUu9U z3s|-3!zc=R^x@)|z&FN|`ml)*rTlg5`e1a-`oNo<$%$g(tUD44eOHQ(jrmQLx@jBu zT#Wv_zB8ivGD;;D&i>q_hU%Wz2wp0jZc6Kdz#&x<_zhi^4o)ZxO>v;#0^ULRE}%P< z`AB~cu6cQjlNFyu8z~_+t5RRAx=;XtZszI&%>9T;@LtpC0b88@`@-i<-kx&AwTO1| zB33z%)LA9NoBA4l1P9-gLI1HPZW7<_qp4RP2mPrO2aL!I6Swxdi+DH%*K)lqj1Iah ziY;?xuK6V%+v7FM=L#=LO^^-pWJpa}wq*fN&IN)<`QL}pqB~kce`JHGuvQ`A@I@ZT`X$2d#&`O$T(H3J}WxhfyJ51tU>c? zV{!T?Nd-+Ev^UIWnpAjH^2!YXCdc_$4{nYont|LQQ^dvuz?;)zWdqe|3hz2~c|ue< zzeGVp5@`*@@A=!NbQ6;#mIJphbtQJ!=CdfLF($Nn$HeR&qp7`sQGt8`s&Vi6MA83j zCtq0Q6|d!F~Vy+`AgG+TYl2{PxQZHLp_AKF2YU3;UqBTu4*?5Y3 zUw#>2G4g$1aoOO`RJaE3q5S(Hzq-3axryXT+Zlio^nz5XA9&lA?7ou~Zk1ZfAMgJh zUDU+d1@o9psl+p~+Li?t4b|A9+0=UHyr&jzjF0PXHG4QSHlIu;Wka-CkbL0;|SK-tAO&Iu1LKiyR^Pq^#10Q$vQU-<}C1 zFs{*uLNNUu-<@>7G}{pGED$~EMyKQ)Dj;OJ&B_ZFWk(T3mD%yP;9Ab_8#bI5BbrLr zEdCs(;HR_!pp~YJ-rkB`uN93c(?(GayFMwL!+&#bydo)D=8c&`jFC`-{lr;`vz_(Z zqhMs@^IH<#28XD>$7dsMXTH;yrzkR#uZ7*XQ56h$TFrG)_c+CMGRHr`nQP~|X9oQI zZmd(h+%^E%opXdUQeNbBjiCT5bu?^)?T>DpT?x9W5{aJ%JR&-q$K}1WLOvc>29oOx&dbNk)|QG_zzCGOy-UtWo-SC?#1#Lw zj<6}{>FG9BofbR50`C6z512vNeBo(5(oL$EfI6Yc95rLoLFJa?W7`9*RD$jZv)+@Z zKTea?j$hRo&Oh6N11Vd&7x2mGQWpJfFc8GL+s72*8K@KA!(K9qTDZ!!j@d-h0mV+q zsyzImbTG0F9$SKP+902(FlT6_>i~fLRz+`lX?PU12aHo1or8vqI?AOWgzx_SF9=g& z`+@$~0C>iEv{(<2H$vF>Ha==s1NpAHs4d6cN^1vZ<;aM7_~fv;)~_3IN*vTZQK9U= zf8N&!(j;=u16n1>L(=JlltU;7F7UPoB}34g9su`6{uNE9BUjEYeOMr)H#i89;H2e7*SivGJiFavL#q z0yHVLhvX1f97h(3Rrw9*$9c+=MsI%a)BI+?=c%l()UH#Qsx%gnt_~g2&6sksD(uuH z9xkoqFEq2Ip(osghnI`SYm#ijgZ1nhO(=)w6Y6g^i<^n#ycQ~$3eENldQIK+4%}HY z`QAU(HLbLy^Qlqopg8B;pm?fUF#fOp7L|Iuouz5Uf|BA>TZqCarS{OXlJ054?5XNm zNZ3C^LsgTEi=yI+SkGd31k^F6gd$A%^Op+##ne|+#2Wci(Q3@8l*g)3FZ@`Hk91}s zUMYF+$3|Iq=t#5Q4?^ZIpZZ8xxI(Yg4vzJN{$HK{e-${b#ChHd$`vSbYCbN~on~Fs zP6k4AcDF>MrXYiUs8m(YA8&N|5gBm?m|wXnP1{(0g51znrJlG<2Nn zOuUG<7)Ix+ZhHH#ToJMXbajtUy#o&}`7bvTXfZ&m?}A}9TU-A@1o+y7f;Hi%->(%Y0# z@x}KWQ*l$tY$m(wvnJH{7kZMw0;};D^CuyJ9xye8y@|9Z@A%WZz(IQ>{&SXsF6*H#K+JjcUfq+aBTbQGnZf9Rw@3S9;7iaO z9PUnHXP~K2+wp<-g`7nb4@7YO)7lARsX_d4L)va~N88v3KB{r4b~n2@?{wwT?%8Vx zG~-_jP|hHT@m1`n0q{kfrxA&|d0|;ZCrl|qlzwHwmuxL=J>5{wrJ2g#N=UEp@Vi4v zuO{u%CNYk8KYVfyW^rXaw>Xy;`QPfjI#!sl?Zf)mP$(1ua=H=|6T)2^8v8M$6*3VC z3BrjIN1QGD^%&2uAxcMT{F@j1L)<+4BhH=o|6Wpm4Q#p}#OlZ!$%E7=$iTb^6Qm=o zU*W-r$B7swE@vGBGl@5Cx(Y@^vr@QoTPC4J@HdeqtS?_@R>nh3r@m2!4Go2=!B`9t zB#5+%WoWP0(GE|#GKZMJwTAYCPMIx#DGGogicDagM1m%Bjt>z-Z&xoi&2Ec7F&u zX+GAu?B};z>$!L-bAENWJfR|%`|0>1d)q4(n@{7<<9|z$iohU-d&;Og<1lvWe>>8Hkfy|#u*Un>1g&; zSz|Z~RTT)iuvc3qu$If2ac&`$aRti2y-$PoBlNm!DPP>Y5K4g)Hc;1qY z41(vvhMf1~40W(w-%-NBj)H^Af>rIXA^a+@r%R%=iS2y)n7byb@jB8)(|;f=l7^{g z|Grd*PyY9Ir9H7JtX6d1Yr$ZuTt z+#^Fa()g#xi=hdDb&4=(1I7uHU$v6PdafAtc!pfaAcbHa7ahxZrl)kX$tuzwU$3xn z@8cT~b9SU>$fkdgOKy|1k9tezzq$BGe$D&cHG4kFflvbqJ5=Nsk+cF7K(fV9 zknU`!@nysQ<_8BJ6%12t{2dUdx?AO;d8yg#{+a!+KJD1xJ;QKVdhq zGj8cdp-B{`PRqG>>5d~S3j-rX@K(`8u+JvnxTEQhA`iKcL`mC$(CYqO?ZGF5o%&@q zpj@#4)fdP_3oXlhOdo!zkh9QS4{m0JG;Z9AZhweuqE982#pU+Xf^-~-kiQYw0u(&W zFonwCRzANWQBx;ee-OJU+L6%ALxWA=xh-t^vy>&pxC$H`GF7<%EM)s@h5i>ji_)C& z+W8G(5BGVc$x4Z`UV>27IxN~#f?xIYEOXR}ucHD%FaNb>SEV~*v@0_Zs)6-P`C#88 z&AEdls7uWKbaw}^j83$^8nh}2R1uaTR7~dE$7zUO8w|*;K?}3qIF{lGv=m0Y*}5|| zJzO*5h6G|F8pOO8HyiYHe->#$RzEu&w*LO=*AJG(fDrWe0GvU$Zfw)#!4(7Gu|kD1 zV479?e-2A~8Mot%yab(T5|-ozQF@1Lo#htGqWJ_uPx|mHf}waW$V~wL>+N_OIyVki z2&Q*obp?sd)Zy=U70-}-QAzMb?C@fi6j`ga>*&VU)_7FTvcDJ$|BEs;iz`c0Sk68Z z*XY!(d5L7Kf)&Xl8Yg|!4l<3=&Km;0E6emIkDZ|_K1v&+f!?n6WOss=mb=kJlyjz3 zU%8*o1+7@xW-hBIXKXKpVkGFBjxpe+?cmdz1TyOE04fk%f)^J zVOZ^MEvqJ96WMDloaf%KFZ_EayIP8|(zZ5yxbr z{vtp;MX_!Wywgg{mbYG@*v&QvpConX#%E~{>nR9UkY{4D*|mSxb)HM#<8|~Wr`LC_ zFn#MEIGiF%2r%leVBMh4k+o@hd0(S#iGvWErTPFPRy6Q(z=KDOLbipceHV_quMpS% zQW|GZTpw==!c+8PtBs8r7{;h1l0XXZ!sSQFC+#lj+bpzNr;Y;2IZ6AG_BX!bo8PBS zMzKF<0V%drh)KviC~PPY8vgeKEXj#FAr#{{;PAcIW4h0BpT^XFs0;AF7<+EyxbKTdzu66FY$L1s`$yT_}3IlVZ&43??4T0-DaCelLu0}v-Ecwp+ zV)!k!F2ome+Y<|L&zEab4>4l|7SceTsQ0q)M2e1`lXhf?@E_(+t6eH)R!+A031iWk zp{qrh{+#4T!sJ&BZWBO^esb5&n^TEBVb=W9(fA`=roYfHupjAf{Xh8H=tf5Y0Wjh= z!}h7;30>=*YP1M8`P?FOoFYhjuO3~6-R**bX?!T;l@F1Px#uILAj9cgE<2kOmPEgI zt3m8Ji!)*fqi$AKv$fz3E(zdVqPG&7-dlc(zjl9!iEXC&K5W>Gb#Ne!CV?{V|;;tjbpJ`Lo>)I8Kl;e;1*8 zqB*eO^e|$IJ(RRSJK3lYo#H6$sh1Twtxnr}H5zlbB^#qILBAe|_c2~&`64`w*KX!r zskHH(uDvi_PM&4Ma#0rbp(oPoI+qbaYLwe<-)zh3uKp)w~k^CcHxz9R5eXkavn0!Gy z!e|{-VlACcXx9gf!lO(7>{y1<_g6~#qeg9ErupinDV+?AUz{?TqhSO1dbpT90MYRN9ef7H4N6No)~@nlk` zRr_+&!s}J*h-cYv_~BYe!mM2iyNDdu^)c__!kGWpxQ@z+frc8DT=ttwUJvbikkE6d zmxtRB9+O#uj9TO=;;*!~-@W|*`-KrI+oR+7bh#T}@gUy)WVR-HME6fMCk+5KY>kEY z#$p|Kh)etxTuCnMjm`V#)JV)-6A=+&I!fuhV$GiT*opFK2waAM&=E~HX<+(`gdw}I zh#W&G+-M~UFQ|2R;xSr=bYymUfJD#pRX|56X$@cASr1Q@c_%T7p&)#9ZSEE53 z_uhNkfYWaLq39v+g}I)XR;#B{oZLgo0`Vc%Z(}YV30w>ZWzz};rt*7tQ-G|%DO|mk z|3%K^#|o(e(^9$WqrMyLS46e{1`$WWqgFoLPtYlpp?DiR9SDm!tFPB&toip`Zx*Do zKb4L=$!?i{!SD=qPx-d1l*qd+Q7ZOyk(;&GS9^2s?ST91+2Gf+U%kU$8ZnAUpuCFx zlU*W9aylqZ#m1?h5xOiBa>?P3LD~Q>SPgpk>@pfK?M9=8bl*jFCF+(%qQMuH6gM zJlvhS(b!Vw&Dt5O();|+kDZQZp~<8LZN|a5!Mj%_xD(mlT2&RnEVm0>tnrT@S{1(w ztYE13AjOvhh|~4N>~0OPYSS;=g~XO3dhyy)&2)im@6dY!GXOnsJKt<@#`@>e>5WZJ zfv53n3u^GE!<&-!qYIP3YQ`Ap%?se zp9456twtYoGMkTl*2S8isjLh7`y~-jNJfXS{H9lYbFZ%HubzPL<+}bbKS3U;%q%=xMYHnO9i;w*Ze^hYOf3`PtNlv>!R9!KKcu8AZ@tTh7G^dZl>3aFjR=zCy4t|sD4(_8}{-O*=5fUBP~Cc^U=AnY&jy7;tT+gsrC`lgG*_AkAE zvc;(Nu>Lw8)X4`vbBw83MOz|ok%&vcig5`#-3<{g{%C3k{wHrMrKZkE=P)ur5|^A9UuQFB*`{y>s0$kAT_ zKS030aG8_isP@-v?=vEWZ`Bl zWt75orXkEN5rypayf+xrLkuI8|eia}T-$D<3SxwCsXx3+3supAt zYxk?J1zW21gkF*^V3l$o*cb_S#+F=awXs`o;08^;-2WJ`=6v<-d^+t(Uf|NEFXe-^ zinS2o%oBl)#r_gYlW;YzpJZzf9Bpi}0~8X)yJ^u1dB0?u+lomnGQQLMehYGL`~HCK z!`IdRro7MLa1rmW7dZ8+*UzIL0Cy0;7A91fhZ&eoKRyjCmkfgAJ<8O(8~^H)nTesk zIf+3Q@OK_@LoS2(&uWg!J+=i)pdAf^aTE2c(lO3H_<_mJSAhI@g zRy@ctP-wR7<4OpO{t`Uag0;(&omxi+1!@oKbWJeT#$(82TH8)!_F9OlK zBy?XXHFb3!D0?$~{K?fS(gsc9rHNr9_~RjibG3!5)i%I~$DB8^8Ner2KRi4sBK)lx zLD0(#`=szvoHd`Y$?F2kXc-($OGiL<%jo>`R7S?$dpV zn?W%+NV@@KOTH6YXLlEBy4(*-^UjQPDI;3Rv9$IOgp8S3EEP#hGA^xy)pH+^DAXCK z|4@hoF-oLe*~Z1jTZwomkOLK?rD>PmkIE8_V=n1^Y8!M{e4a>}YLNxT3dd}`@Y*Ym zdvJHgd*G2!I*yoDSSO>p@sAxYr z5Kqvj$24b9bRDdm8^@-Ht)+)q+xibNR+c28?{2Nz^aDea3pbam$YKUV?YEj zF6|0h|7}}`Yv&$U7j{;%+=?s@R@68$MtyEB7E_Yp^^C1@z8zO*$uFzw!|Fy>wnGC5{#5L{ zb?X6N%pY<-ujZN226={8-amO6@N%;Z6v#;E$gRZAAF%c~8n-?VnftGK@h9q`5?mSf23jKu(k6hhKZCx zo{Y<`^9TA6b#v!!6O}nk9)y9@;T>a>f-r@B6FpIb?LzFqV&3Mjn5=Z%=d{)k_Uy;a z(*=u5MuW=6hvUZiZ=iAmtV2x?0gPKe_Y_46u~K6ZObSb$AC7h_6ZQ&8UqW8@K$)roEoeWd*>&8iyKi#J0|J3&FM zDD;oWt5(}7Hzb-}|4XYrvCO*5x@n`4%Wv=ET?23SDdl|8 zLX3rD%Oh3qsS+_Mt0#J~Y(8P*k)JW*s&vx5>q-H;pkZ`1V`&l4dn`f;x#;lIHPsg* zec$~JQ%6*{-LUltZU9j5+O5DZ-^+XX9FaU|5E6kcYbWqsS`cE*CS~~I9OlB`sX9sP z!3ZAT=as)RasM68&6b6-nf_l}_P%2YTub!U-=)Oa$kjkJr)eQEB8~&G*-f3s50;1n zCvQzd5U=n-j|@O+(rp^=7^WRuigeKtT!^9v;xFc$#r(Qz)D_>Ef|;6Di51^Agda#3 zYith^pkanT;`4S9md$v#QdwUaPAGbkiFpyn;Zn2{0hP9(LAuQICGle)J-~Cjg$VebyA*_WC2-ttwu_r&o980W=_DZ^pRD zSp;&=x^SQ3YQI|U!>)O>J)WSYu+$WF+P~rz(gnGbj^DJ#RBu4UijDtf5Aa*3`8zf& zd%31!#U$gbV{rdBnbjWtq9x|C)2*p(eh{>Fjrb zjMovX&IAr6w3$)Y*1uYl;_*ijZA4XT6A)4F?i9@HR!=N~@AE3_tKb}8&bEGp3b4>+ z5dwD>bB!74_{_AXQb@t_>2#&eX?5AB6z(_neK}Icxk*5KR7hjD9?!Y%WMW}A*ZJ@L z1Hi6bJWnF4$w9xVo>%p;9}K7Ym+Sn|4ziXcg|I)j&IHmg;MzF{7saam8C&4dsrRQAPad@4@ zRyUpdU4y|84{k$%-fmzDp^&JRvkAfJ2wvEJkbx%$8l?U*4g$9@ybE-1tzNZ{X8X|? zuC?%P$^|#W;M?V4kY0kcObuRW;|X{FOy83|iK+%4IiVqdcRT)q#Su1nqeW4InGVAj zn?I9Ac(e9rkNh!Pq8otlT26c(bj!;6H37fcz|4wi&9rY^@~U zMn#}V38L3lw>?U(_`Y{8rw{e@>HorUG(xifv1j3UN&lT3H3b4t3{(oBqM%0YVsPK3 zlTq+Bi~a=*hfs>ykHhr9h&&!Dg=@*uEKAem7vza%5h+s7++k#mS%iinTEJYz|j94FjG@v>di|P-I5p(3QTndOA1^jfmuTruVrm0>+NlQncc(*b$l*CUpjQ+c|JZ+eIrO zkCVc?L-=*b>xl>wUofwHid8xp-RInVJn88Z70a%=y&62UV5?D}n4jtVf+xM^OM_8_ z34s#m6K&U}C7onbeIA{1#^2@?{m^>uAd*z}tYql-*erTP2`OtqV7f+=}_0>R^NL@We0bbo#FrfTTv{!HJ z3roW(D@_`2^w|)?DHfdU+}&d1m>Sf#7Wo$Xg*mAHm}80 zqWq*qTkoFkDVH8W+D^eCXEpx!y!#RF(`d^GlsI&kd{$OoT62EQf$$k`o0vf^A@Y=p zx9Ynn@_BEhSg$KUhaB|2Ld+ce$lAw3{=Xd*}DIEQrRVqxvCaMj$qup(?cC~tyXrs-5`Um2q=N4dtMFexKlJGvQmWxxXRMWiW6clL&iS3aYK z6=b`cnvo0lOLs^I5@$4Ph;ndlN@B5*gY$YZe2J$?Kfe89)_cxeNDHe5hZZh4wmD0VaI^ z9$VX20Q+5-Y11=7Fl{FMUyW5d{4gd7fu&HA3Z%U>HQHl1Fhl{0!59jEft?(73Vr{N zbg5H#TxTmOZy&Q_Is989pzCK_nPB?SK=-y)DCGBElr>rAd&!TbFy!41xBGMNRB;t-A5GkZ-_@;H7S+Wmri281zzQbJX~1uE9fcjx&3BHU z-Aye8*E|rq>M-lHRrG%a-=yHEd-bjcJNje&x1=A|#{+L!sghgOndJW(cN?W%8SHS} zOA%GHJJIYg=)VES)t?1$)l6_=itjT_waEBy>Z{%=CPGiENK2xZDo7Vhn$oA0#s_G3 zqZf!yM0VA0<$O0~We$Xa48Rl-{HEg8Ja%2Q02nLU>AQT zO^a#ZLnru`IQAs?SK~H3f<~(qK;!=X;@DRcaMp2K0PGdI(E|pYR@?ll_q!Z4YW>zc zaK=rpQok!+s7hv69X;Wu(~goe1WuNODk>lS>p&1Y{(EiZ!?N7GG;Q58u-)u-#`ysI0XyP>TZpf?uga7L=CCPdYlfrx*sWSr*ppt6uPc+pefd!CdN(aF0WM=ZFpU51GJ$ zkIaZjhSe$V#qvWDg4ye3XTt13<*$TAu0t}k(ycCP%EC;BtAU)SH)J;50ziNZVKGf@ z8Bkv+hf_g;pr;WZ>>YATAiNH-H;miz#RXdJSSNLmF%? zJ}8UaO*}J%Ty30a%zRy~EqQ4r=0+kTK2I(Eb`H{Jyn0($W&c$xF}IEZ)Ql*1{*?83N5oH_Z@t_ILC~$aGiT>=7Qyitb82ix$at0>KR$DJ!Y1USkB};UJ|9vOkoMKnQ@efIx%s2*xkIc zl&EbOsw(p{5HDxbS8A8|&g%3!XL*%033bj85V97T2YU-}m%|ZO??0)5ioR06H+ikt zSXKEBRLuYw7_#+FQzHwP0IqyzXB)1@lf-859OmhJ{~S zypKrzunSLPrL#{iba^87L0Y1&-$;JHTkc1wi^}|q2h<)FTP*V4hC0>G2UsmO5=&WF z%jB(H@fIXbXU}+OWCkH?3T%qBlNZsrY+o{Hw+Q7+O;64;j^FyTLy;U_`(#l#P2MCB z_28QJK^Q@DZm2dFVxXs){$}|a>7uu}0k4jIxOV@DmU{(s#BJ4enpg-o-J{teQS#&| z*ITZmP_U&utRZspZiK^!+ejRExJ6EKZ$9}QtWpsiz;OuS6AYU-K=)=yW+jRE2}Wy= zj(R(%Vo)hb4!E;)nirCfzF_f>5}Ja`0zDV%8dzxepq#|yXzewsGc{-F0>j_X@-$$U zfU?Erc-@`q#ZAYY)hQnjaEqm_QDK!TD+fGInECPTm@Fc@H&R6ebIDwTw+NMb!D8me zg54_FbTc9%^`Z8x!=MLfKtJmp2h7+21Alcgi%ocS0>HCK`%p&syyJWuUY|lng&uYBPASlT$@8!r0ZXVQ42!GLmf(s*)fKTtCoPBP=l1${oarCCj}x6b&B-~E5fNm*^LK$z8sgtp>8^xOZt;#EWa zQv7k^@eL5${Gp*VI9h6e6^1!}{QOZ^Lf>>!d~V`~5N!a~Q}Vj5Czy@T$+|Ms-7@Cw zwIlC@weHV!_~fAA{7&umV5K&rb|$1-+!NI&I>l6gxhoubpMXJ8{akQ<gJwQla7sdLHy;UZbThaQ(Ly%lYixO!#K5 z^Aw3#f*M7WT^@ITwxh9MMR-bd-@+X39Lmq)b#PhVU+Jlu*$KEz+uzTgS?f|OPB-jl zKEnOk1Y(uSGsA|Nrjg2``Z3K%?|6Sq!o_@?o@oPaz{@)8-}ZxG3mxD@iPNMx25Y8X z#1pe;t1Vb?>#|2EU$z+iSd>Q_cpE{}(+l-9!+BD*N@pOhqP*sC<>+d?FE|PBUhd?} zbw=eA_0ba##eC%kCs%gCkW8Kx6Hs=>5ax_o_BZh)`(|`7qnNN1(v70NeB&RPn6X$$ zi6T(Qp;w44M0>9tejeX$OToHYfQIo$d(d3?2ce_9zIVkvpo@{7KB+upf5k1!6KH#j zV%H3pH)+1PKCRK!3TK;+{y-nQPeI~rp9O^V7j9Shdh_t>uj}%I0|6-)-l^D!ODK`lsz+gQ>bc=N>_b+DqXG?P= zC&J)8fus9OYtzVz{@m!tDzN@rO`+n}M2PpGNjPK+lwt5@aBAJ+OhURt|I_OP;m?RH zb0QX6opRiLGyc4zLs&VvG^*xbOmH8ahuVQXE`6?X(#i%-`P-}8l;d$6pI2PgA#biS z`r1-OlCMgoTKQLc41I0X3`bI*fq6{1BZAv%P+rEZgY@pJp9$NSoKccj(D~J>2tcTM zbK8;n?H~SAss&ZR&*`p)YtoNy|Bw?E_!DZ^Ertuf zHA-Q_OL5!_`ULe$Eo{_o#PBAr)=CBcX=P$Ae8SI3_+Iwr7KJNJ)cRBHkjnd2h~_(H z4Bi~@UT6jZ8(DBiQ37tctf`uRpbn-zQpb4n@hkt_$N=*tC`f+y+-T51SQj-Oax5rR z9O9XkzNTZ_)B!2M7zpOSsrPsKc?3V{flkX#tn=0e{k9iAVOx-kQ{eCId{p0$5+a(= zY(G=m3ss+i><}_Rur+y0aYcvIr`WvJ+@8ruPjOIs41RglE*j#pQ(MQ9x#W*R3sPyq zM7?kJY0KY%_VX{}m6z=_L(-Sj7|0j~WX?AOqq|_R5V8r?k~~{kg3Y4dme9F0(kAYl zQlA((g7zO7!e;sXhHHu^TydddlkgSn_TMtiGu(3uU50K%NkU$+8!@KQ@27H?^SJtF z(BJ-iNa^_a?jAWl0qe)#of&S;jf9fz8qv>*wKh7rJ&}3o?CxYb&yKp$ zdf?~84Bp$Wg>-Syy#pC0Uh%O`zD%cY4S9v3!z}3(V~|O;jM4F6NmMlMc8=||uf%rF z+Lm38r6`O+S?Q0ImDQwja%hZcpXHV;!O(T>M}q`mGD-eYZgz2c1JSG^V_7EomL^J=a?Q%mGvf<%Q`ht$`%|V{a3u zy@>7KmN9PWzIChd9UU(c+g97VZmA02hn=dx zxuC$%8suTr|HP8(O?j|u)wHvQnwXoZ;pz$x>1|%`lv=rQT)V%A%O_`PZtHy!()?}Cgt4!3lAwxD{jnUny}`D2zq_43H{Y+s3YiwOTj)L zRIn?UJR{)cXko(9dY$9!Q6Z`LyTi7kW*^?r^#JU}4EuD2rJCg^iat`U0t`_~EPW7j zT$n)w#(>X5x#kSaXsQy-K}&O`^+i?}6-0~5e32g8FakWKcVGXuK~7#Otjs9?iT4o) z0Ue6pgmqM>T&kdh@)YA~)>7hya#1QPmfZ>X?)bi^3Zt7`%NS7w)4yPWKU0?#jJq12 z?ei@x)3{8P*NibGkL*4@gF4W+VzwbN7QY2yr;;d_hdrp;Gw>Or$eDT6Am!S2+P)O- za!CL2XzEjN#A6vqv=WNH7k+y>&^}+CP|IU4@U+0Sh6q;0|1|%X#eO#p<@bK@=tJ10 zEc1{dF!m58ibvZEr|0or)~AGtlDhN!umNqOW)5NxijRaEFGgk_1!vHfV+_KL6kZm84c(XdAoMCWA7)|!9lD`{dDnLYc13yNsHscugc_sSvD5?=A3_GVG=#DhhS)F~)Mb9>Z?$!8KP3zd571!Iwr& zc?$c*Jco&K-v=`@a$+Bg{R?{_<;uA{vVrA;KCiT(Mwel2oE9lon(5@!nChQ9_ZFpu zL(KECZ1R6EejDa%10XIn3(Id-kNp>H(uiP}*!|fH%zGxXj~AL{1hdASF=>9hob9|t z8+KQY*#Ve{yT>b3LT^`BJ{KbXM5Pe9-`D z;#0xoD`JKaR?hc1rknUSwng2zewjaqh^MlX;(M=I!=z}=MusJ>I~+EJl{r43T!GH_`Ba!Zy@CfPk=?MLT5o){^O=8~*dv;?!WeA85 zOefU}Z=wzMA_Z_^(~sW_&op=>k3i4fbHdR0=kcYF(}HgLel*TAZIN``X*x80x_S^% z+x%ibutInQI?ZYbtR}`3xR-BHtck4unKzmNb~~qMfY$J_?MrdT<~wcE9N>o%oj6@PX+_U(AWT<=uR3`4?f!5nFlnbHMIBH4I9d@ z>|Ona>AHQ0S-gw47)|qhNc!a)5Lvi~u8}NnigAX`(Yib@yXtk;nZ0IFb<*WpqCbx( zwZ8PCbSQF{1hi#NAFVWnBJOGgHw_0qMQ)!$_a*i1nm$HdT(TYpL? zRnI?qtkc8$@zsYK2ei^TPp0}xoz=YBFCl%W0s{17QR(62Tnu;Th}FA_AoBlS%Y|Fa z#Fn`NpL-h*%jiE36)oy_*_e(uzJK`l+O{!9+{}_6^&FQuzEE!$s_#*kSEJ)?xTF2a z64S@x`0@{GS&cH0=bPhv=Ul}%jL_GXb~4A8Lh|Nif?ixv%)ey9+)b!6Zbn83WNrIW z9ho{b4uP`S21FU~Y3?cKI1OwY*ky^mU?Gom5PUp243*5qM9~G5z+Ik@(%7GO@SKwo zyks*5{8?{0qGLKP;U|v}HYW7Hzd&ZI`TljJlW5ILt@ydbH%xp)Ffygn zbVuOAuW_2R$2Sb#x)RgKNVOVLo==k8o*IE$rCy=9M)KO z-7=Y5lXPEW66uemQ^BoATw&rc6=etRw^!qnLp1cC^wn-^7$FK_IBoiVa2qVa{2UpCU5z^_-o?fawu;AivpbyC3_ zaR(ZTFMQaBp`z+fT1KFZ5lO$l*jF=~X%cw;lA-N-Vii$CDEvb0b<rYxpNVme*HX`|kUTc<}-!88YE7DE~y zkcDwqvqS%;JEV4=*tBIb9VCGxZ`b&CQIj=HBs=>(H#_~7 zY>@Uwim$%&liDm`?Dq^wXY*#E=|clbkRIp1>m1-$m*3rJ-a9jy+d}Ku6IXLro24Dy)u!amN z6~A@Mfd8EWR0Cuo_u`-k@T81AZ}|fiSki5@t7O)O3rYuhT@p+n1iXr1_(?= z&!#(N&b4AazU4kYN1>}n%k8BF5&-Bu^-aSD$NF77-);`O3^CAZh1vRz&0Dm65BS!X z3=Iql${jczp{!vp2h1LZ6}}j8vY~&^=+&Hr8heFrI-Yj_0%$^B-^|A1m>VC`&wq*1 zhyZV$-sNc6K3$`Ivx)!1JbCJ<7SGIY+_Okp^X1+eK$#>VgG+yG1Ctz%Xn0~D@Q$do zC?)iDac4y3#bQR01mF~f$A5SR?W-vAc^>kk(G7Om!Lyaa!ggjr2ehi(^|YL`|62E^ ziai^#*@MpO$P4|3BFC&=-k{=K30hOAGYQ9^O^iXq{nEGu6et+pJhR>c94?cmh5$_$ zHZ@u(&)$zn2rNpg2J@Mtz+(f@rV8V9n^mUVQh5V;FS^N11v3b`=zhpi?SeBa^wHu& zO{5J_UP&Fws503z`T6ZIRLM&5snc445PS03h$_(f(hOj0>uYj;YvZ(63-oRPs-3h# z3TUM)An;kBq}x*_qfsE6*Xp>ViKkeQG-gN>8RL8opY?mhljnwcGAcr_1!p#mIg|JZ zx`qZQpn}+7D?nvTa5@wE$mZ7ko=D3>yNM&Dal;l3%Drm*oi8;=W$Sy|k$Kg!1mRuR z4~3IRco&Yt<<&a6pP8QT|1%SdRqUcykf%O_vcI68H@Z9e6%(BJ)c9F?E!q_Ipe>mo z9ufJ2*ENm6ptR9Z539w4C%yJwdj=D8R`{Sq>VR^ZDluRfP#@=m8LOd^LI9yn?uH^I zlah;%y)c)9e^8l0zm--uZ#kgHD1%<04G&a}g>>vZX83;?()1n=3woQ1yL8F~##-7% zNoXo67B)dZeBk7$c)Rxh`6?R`WR8ycZ)6rzsr}zHdj${U@Du+VdgSOh!Y(pASGAsv zXP&%P`E9ug<9AA=<060QYv+C<&?443_nTspx8evR9AdN0xrE!GFR3os@Xuu?j2~(w z6XROlEyOS~U+on18`;dDxjd7I3MYjaibv7HE}<7JJvG1I*NO6cHQz+i!BSqP-~9Oz z*N074zZ!&rjEN!gY(SGCKOQ{`=!6*nZu0XjjS!Gb?e2q!cb;IgY{n_LUY`Jytm?#W zD-HGej|yFP%bh6C_&c2E(gzK-(wf3{eaFcSt1mAwiRSXOSFx_t9M4P^S1Uo9Oa zU~xc*pehSXB?+;uSD&cmh|61|uBl^FT9>KDMOu;mZ{0x8@zr7>?ZR_4MxWkpwTtIf zPR|nf?4*2o*eVs6ld~NL!}8P~a94dInPvY87rFtB6bF78=(xyX{}C#MAvX@)Z*H9A zA+^>jzcX&VS0!1Ada^tEz-mz-veY)Ko>PvQm1ec>->a1FN9WFG8txGiT=R)Luc|Dh~ zz5}fhdPmohC$0GLuw@u`iunLtO=Va+6=ym7Kw@`A^9W~@FO5!rgC{@kVBMQr>2BSd z%=#^0AK!%|ao)kSZb>5j#E$w$U-Toy<(cI9?l!fj$-`UoLErsd=+aF4XdbZn2`mxp z<_j{dU`!;5;DGceODd-NwdV#yO>!9}+b-k!+l-||V9Uqnl^7wr7>SQItGv1vK8$$% zkhZO-{{vdK)Ceku|G@YfPBFCY&a?4NVTKiOi#OJ%j@9IUOWY~}$HEfb#BlJ!prf2qT73pKUFn`t{K+4i8wRAdBk~%tx>?{dHUdU8 z-e^Ji)&BR=2m@8@durOkiG{bFDkM#Ee@|?`j{nuT8mXpow}CLLr6~nmJ4a{raxv~I zI5$teo1VCjLU;Vx<1NrzWKp_Pw)Jf9Pv;oJ!O8odKXEx-kj*8Kp~JyTbJTR`I&vx^ z!(MVw;RZy#oC!a1bhnZlut$pX_gYt22R8CuSvU2Pab(X?wB#+ z@*~vc=uyDn)yEVRnlJ=+eRGPbk`a3)hw9Y zML)zN*nE{tKh@j+`Tk)vWWqdgqks_EO4z5p_15#7e!LpGJ zVnv&+vKDg|2ntq=%SjrgJa4`j^gsS29SBJn6@T(hP4?hNn{NQK@VJ^-imtWeay!<& zI9XIwQ;Flrv{+S(2#~%>e60(NSt}gMzM)I#DSK^S67D&79Qla;AI~5-#KHA8?y#sBJcT8D62pIT6n@|cfKqp9l*q75vBIK2tx&q zz1dp7y`+2t8q8^%J76ypnV+!~>4;h$6CV-@#<*pfy>9T8zNz+G)HkAN<*P9Xa=Fgy zja0k$NH^M%tIqi-1gP70ruRjq$)v^h(z*!B9-~Aa3Q^BIADJ(41>BLNFXYi28c4zvj5;-B5D_vrF$kP&vLt1KP(&q!Ks~!24e^RHfetP*Q9Za8ridm;CO%z z{d8B^j&cz2s96ib416i=t=YK$1bumc*1hCaEUa@7Mo8MeMJ_R}Hmok{ixX+`0azbv z8=O&RKY9cd*-|$`G| zFF^FcOVP8gUZe9q8`CrzYuRSv|aAG4KV&js2=ic2qrCJvc_vwAm@oDKBW5MZ8IrLK5* z1IH5qj3E~+X^}Iy3(uu-nnpjAB-*ArYWWLg?t!tLXvyscuo#NX&via*Rdt6syJv)q&8mQ~RC?8*=m-&zq$dqDS z&odjimLee)izwMx{{HSe+w*W6m96U>6|JitE+Z#&6gPquhtZNI70}sbc?VJxt$(&B z1$;gu%Q5r}S&UEt>QP)u_lr$I)N|os_tPvNDTpu5xuhc>#aQkp3Xq&@xi% z4_oydM;lJTzknrF)>>REqF3(DJoe#TavHk8p4k3f`c_VZrD1vjW`{>_o5i6&+|$&E z+^=$Q!{lmYUbe35)&uQ7@x~XlIm|6+YL%)VKHBGIH1SM^qC|QBVO+~-{BTH1+z#`S z2^GRyls3d8GA~mO`G<jSkSi2W>ntJjpHV@lo`*JApU!9hjzoRewftC%(v6NueRZ zIR!smS|t1%N~Vq)xmo(}-?7s=WGkEqj~FZ~OM%EW^j=UQRO*FR=@`&} z#gXhW27Teauek!To6sdsb}xIeZ%=uCV=GospY74%pte=2ZQEX-8>=$|i24IpBxNYKF&G{zZ)0=onUnS*n zYn5R-198M~qZFc-ID$&1{rRsmz2ATyOd1Fk+2C%V7O}Rp^0`Q~FAP8{Q6SHm)GAue zJ4(xvD>+Q=qYc?+*>`drg~D8pZ=Y!0zV@H{A(;zLV#m+N5Qlr(nRj!{>i0#h^yc|M$=PX-J&iIF-Z5$o6@l-}6_s^u z=oIWVn^b=^WRw^I-$K#VZUMy@gSrUnd2mQE7N4L$Ar;u=FQK5EWqWxLwzoLlz&g0i{ z3T_(3f2^)S=xS>x^6$t|aopyvZgZ|B$J$s2bcEkEfo4 z`d)@P>1@#u2Yr%2*AYW_C5Hwr4DUVF+lT?+Fii8`2MCTU{-P$6-_$D_B&GZSMxfD? zcOA$V@%ps4Gb{!KjO#z^tD)YsEn&HgDgO3s%@6ILhk=NeMcvpA$6xfmp-^ZFqKxy; zbi2^89ff|{eNlx9v!s~2Ui9(%x_t<+S#%!D&@j|XC|bty#tfgv7u$?IiFY}PIpH&R(> z^O36VoCqwWWE@iuhEsI~`T6?mKAen0%$IDCG6~r*0dFN0=;$Phv!FmT^zP*fLQ^i7OA4KCSFi30V#dsHgptS6@@e!jA89XZ6}9^B(0@ zS=4kLe?10`f#CR*5&3$6*vjr%Sk0OGg5_%5f5lopapvDAXLdsL-PDq-G35r4fzHb^ z3B|7gqqUfVvpbHaaLnprWNZ!5&2k!~%~Gsg zM+cA;LWrta=d%EJ)8-2U{ z^^S$_B1a<HwheUe-Gi9YfCna zU50?o3^Xy9Nta)eZ^&}m0-sMqrAGPOGx3rC;vnk7Gr)Jc?|X1=SSpdWs{1ZH4--{^ z z%j)~BWmsst1B8&B1%Cai`ZeH1uM#tUZS!yFlLC&IPna~hm8iIcRNR-YajPy7Ha41f zIf-CoY#h@EPmW$fq(51u)#;Z*mJpp3llx@I4>a=!{UJ_t z00lqc?!^mw%VE0%$E=6|f2dncP0qEP{q~*jdViZ3yLTV|L{XN%0#aJ8S;oHV88C(B zD^hBLm&EGLTqmG+I+t>Q07~H7FkrG6>BPd1@z%9J&Hl|nJqUh`i}*jR{B=-VUGO%F zCIp9|N$}tvEWvHCpc$Ot?ry)Cg1Lfz4zTMmyG=#K91knK_l$cQFnYrG8Y@-AfI07@ZPRCf7@{{SCH9}S!GKt}KZ z+$zCi!E6If`fZv_00C-&YB5s7_Ob8KMst=xI?ots>hUEMQ2{aW!$54Jvn$W1fia>N z2MHQ2qx1Wb^D`v6ut^~B-NlG@kQUIJN6ZVg&f6!qX+zJd4l1$#e~RJLjsGv&|BDD1 zPko^mhrrGJhst`dq4Z~ApO$;qQ2tbGlA--wNKvx?`?W|R$UGF&mdz#TN*z$6ursjA zqVunh^(f`JOj_0vT- zh9PV)q=r>iqa&eK)HU?|bWy}P!Dq7nkwR21>;>ampy%XQDgiZBArgRUdpd3FEyfV{ zjJN9x;HR@B;LVKbF$EeB()|XNUcBk5GziYLX|uCs3#bEL+x3ru{Rw`O2w!!Dt1D65 zEN@E4pA_>|NeuZr@Ek;2Aucc8R{ECWb&nB4mMo2I334BI~3>}%cCng~V( z(_}xg;ozCFIhj4*gVvNe*tlUMrXdBNV90sT8Bo<9t*Znvhtr6X*B}hm4(Gl5Vw$*t zhn^c{CjVydC=RRKEBw=W%U-O&DINhY)k?TA{p5X~IU>T^qUfq~-bp)T!BLd&hwxCS z(4X+$n-wSDVT_iLPS`MCi%gkvN@tkX{_k)+cS*UXf0G!^UK#TA?mE73#yhIe;@!Cu z$%tQ}jVrSd>eF6whOd(tgIH`rGj{}tH3qHeWY3U7`U>Q{<3vWzH0rDASJ{#P%}c5) z{Ez5V50@B%u2r^tNKTf5_V0VIX`3D#F@jpUOE%kCFH|?@BCBS8fz7$Vns_pb#`tBG zw(5-Fc{Oa2ZsMM{u}S_kL2&sD2pl{Crl;vT$q7#RUf1=m@@2g>Y<%U};e{>Asxqu?(b+FFo-l=3gy`74P$d43SE)+tVdZN0RAW2F zxJ_>?ae2lT*`k33E>n_AD7J41VG1A5)lNRFoJTp-X{>4UUqjUk}Vw6tf0F>s(gB$`^bubssFadFjopn zn<7xt>V1g}XlMll;)*ZZxQq5blhjRto*SDy3<>@GjTLVMUF{0AP};LW16N0X$OrBV z4-)e9i3A# z)n+5dvV9c=l;s8>%thJS0{&c{kinu}UQJ7UrD7(w>0kqg?r5eC#?Dfdf|7&T<-a^^R|xhg>O4 z{|WjKTmnx(+P0Y$X}Kv7uYH;}jEm!ac}*QK@ue%UB>MZZa(r_fr}Q+R+qcd3_5M#O z)W1x%l1kM3-rBZFi|l(LMP6mli4^jUJAYDWdGZm0&}ac3?zF=}nh!ehZ6S z%CwY5-5_%Q6vYY7{5)pD0B+1$3SrVOFTzUoFB+Mry2ddw5v+_U8 z{J8&^`O78#06UG;fGhIb?NzgA=EjJ%q;3|rd)8F;AnzbDBM(wen7_XqlI)Z7k9T&G z~6-zea|BG5Bkmfik%yy@v+FEs-+WD7H0(3a zDxFe@?e6^6R$~SH?}+f@k#J38Z#jAdZ;-a`D~|dr7mbN8j#yN95E=13==Joq0Z_(T z1Dt6gjQ^+sxUcPIKr!oN=Gy?g1hO>ot#)K9>vyvB@0O^K2(H;s4>yJ78!7N5$Y}3) zLrTLH7E|{nDJlIxZvgnUiYNRx7aifQt|<=Q*ZzorJ%7)-I12a|4#&|Pnw*!KF^ybC zYk)Q0ceoYu(9Z09XOHE*?`k#+$WqGcW^mi5kUSE}G7wo@z?I>I*Em-3fY(T!%yXs~ z%6{bKV%|*LiyRec?uD`EXYd2J56OgB3|N zEKDwIn+XZ@f5V~s^oU zrCk|(sL{i2ib31t>RC22V>2LHw8O*2fud+X@Kt$ zO0B*Q?W?LS8!>w_TlqwqSG0d2J8at~2t2OHrj6*dX2$|WyhRY|yFMA;AUlICNMc?D_XS+*78or!(jS?j8<&UPo`O89#{&q zO|ILqoMLcA-hZJg;SCxX$}r~xC_fd;_M$K%xYvWqkx7}YKj_EH41A?nJ$ zA4b=x-cSXciN;16@?{w<0T2 zrzy>__eR@WrgGlb6Ck?Bcdzu94nftYw#}F0>7FjjOTpz6xB98PE={ zYCujmvS%_43xoBkz}3&&IPc}3hpSet+<95%0!i{@;Z7TYg$ukf!E$On~O>LEKQ@%z!b`-L0Qv{7-spp8iXuPB6n^@cz&y8; z#6Kz}QTqfFyVGeGA3?>1t!PPv7f=WY#kf2ir2?Iof0a zli5JUosg(1>~rra#;G6=*PmQ@>G}jYrK?lmAl|%k$+pe+4AO&URzP{*I-Kz05yAxqt z;S$@lHr7)|6Dl6@W~ZA3J9M6{2jF|f9Nc)@b+rbxh4xkaS8N{)xFK8Br%~BXfQ#(l zYv^nhEU=uXWHr81#J}=SEx?!PtbRA_Pq_!ZWY_;wOM2>+Xu$yf6Ctv9wq-pp!57vo z^n7Kvq1YYoW4T`QTauDb8ue;3kUK;&ogH{5l}U&eBhXqI>7|elekJa(eGhu`x!f1oUQwf?zU%h^w2pL!@cdqb=M55h|KF}r&WG@vx#SX=Af)a> zJS(_+Dt&<}Tuvht=|vDpd5!-r^Mcng+7#LP39x_0-zsGjL$~!fg)Y`F!wAMYp_u`} zO!!a`#H>u78c#Xrc)5CTEPdRjh zN883-j#FbxCd^((hZSl&>E*zxf=fp5lTEUo?NN*W;0+?pae7LO(%>P!9^-doDaI=R z62~T+X9Y_<&r$vJ!fHvRLwYNMR|qWd2IF;C4G`K(^$Hx-U)zZ*E~Jmb#%Q%s_yIeM zb8kvZ=Wl-9LO&yTjW?XmMt1QC5HJ_-Cm{|{GTExS%}iWS~Ac7UHkuC{0Atpw1(;}voXxmVm=rH>Tb7g zMO(rJN;~miw=68*`WV}${&Uj2Oq%!POTX)%%#-VQ>3Td8SBPI58Mx76mPe|gpPlsafLSh2=Z|FIl=xY_Vqx5-sICMSo<%3fAm{2OBkL^>;aDwTg+>Vj>p23Tu zI6j1|abe2=HT^d+^ySS6-Rp>yo=?-lG$uy)LJboA+>FAn#qi{;gP54M*6xnXYj05d z($HslF~?Z^&1Qw>UZ#H^ITR~g-{V93EN#9>XTQkH_JL2+;>|NJW_!h)57^eq*VNAs zpu>NEA3{58JCsQH&%@cb8oY*Vz4vdvf7MI`Pg9;8g@9f$Bg{=n@Zb^5u&v&8-nmPB z+DgIkYRBW}Y?i{M?&3)tcESXls_H|9AI_#KD$PGIXi?D{BqkRWZCzA+MMgPjKR%YT zumq?S9r~d$3UtTGDpqr-d0;iT4WH%6{S%zYc0>$P9dfMxw0IMIbgq21j1z5x7WryI zZQGz%R>?pc?Lni6`_@XTR8}$i#Xmt3Z9be(>o`WnFqLv0UzzSH_nVqAkSpz_bo z&+4uFh8Kw@h`LTFP=0D(^3-+__|_Rd=zhk|s5N=WyL|XTupxTY>!#z!S>y2y({8t$ zpKD>ILeI+(@8^8YNS1fsl6!!DjbMNAQTw+woVINgN_kc{v+qspM@AU3hO2gl1B{=N z1jbDb83h^H>Jv9oSrV2XRa9P;Sq_j zd18Ob8~;G#I|->PIR&-0P>PX@_8|x=({;8@(Uw&&`Bt{z9YgzN6$(M56q6z+rpyE= zLqakHxo;(H;T&!pO&B}!K55*}OUI#Sijndr%d8wI^}@Pb`Q2l7=BhPHNnUpdG3!WY=4Iu2iA*q`n{*-c$pVmC{2xPSseW!c z^oowWp_jN375~vFE}xz7N6r9+>bianuTqjigvBKE*h_|u8?f|upNMAXJ#S3N+KLGyw2+h*;r+N zEsI~2=2o*G1xOxiQn<3Be;HXl+v}YpM^-;GI%vvAe9B>xFtw2t8Uc~9LUQJ(s5$q8Grx?f&l7@^>Aquc3n(9*)QuA_->B7qx~t$zm8^tYTQ7>y)6f81(KYbzS^-gvEqj zb1k`A5zWKVhK->4kOs==ni`2>;K|c>8P+wkCtxLpoCqfD+Q@A9ITw&DVrIgPEm>Wg zT;VxNfIWV0Z~{*WYGSGS_ldY%r5!S}t40{A|88J2t{Ko<8EiN+p60Y~OOF5hBYuaqPz&ILK8lix zpaUW=%Zwh>wYg2oZmJ*VlU~W<${8yDbKwMJMz`E`|EQkxcQT8J8{4y{82qTQ_IYW} zUW3`4^_>(}(eKjH+?}ig(1Ag(X&CPyJYmhiA%R}u6QV}TWrJ;Od;vq4?Rvpwb)->#s+cZZj=Zw#K6 zH&<5hsTHpZkQ>oJ;p{U@y&V5!u208~;66VWx~qM6Clu{)vo!aiK2}NR8Jzi$MNToA zY|7CX_-FfkFD@9SZJLH`y`1uO#A>Z=j}ovc61e`^cN1tizjxS*t*B@|1M0$Ln`dl* zb~cv$vA{{5-q=rjnb$PXXQz}#1bxx-&4tVIi29q8w3}X^8<%<+aMf=e+xO_+k70Cj z-9Z~fEb*+*Wlaat^WK-T-p`|Mv$*PAjVqxJ zm8ovXFpRz$>v%=xeyYRZVE#?cx;-KT@3pufma%V={v88$Qj@2rx6{sh;1bITJ1+HA z6X!Sa`s&qZZOij#EybvmvhF+)_q%8a%gp30J5zoWh>zmQbcKI?HF3_MWbOlXf}6I2 z0`^#F*bh<0`E+>@*e6h0#-a`>x1zRX(~4HSn&d ze4+A)6*IZbS2Y;(zqwP<_t4`~NFe_%FMvzJ^1*VcA|fLx+3#Jsrp_$ws1NslRA-np zK?%cRv+o&9)y4gICFn>Y_iH^I;%W~@HZ zjh45bwfRTM%GwcT$HLlFi!A)bcoZjY2)>5C=UgYK1S0$G4nU5@eO%Rc=s10L{maQ9 zj(@POu<^Z)Dws%8d`HJTHG_)~Vv|%YJegg622u;4?IvCzYUGRhbJI$T`&~?_ z-AqtvIa8pkUmLW2bFKvHf{EmUkgXHy`H;$ydB_o=X>|`{`@yVNs==L7&me%yoP7a@ zQ6Y|mD~AzWD$d1a$j!*?`BuxDPM3^V+lGtJqFMMX?Zk}@6_ZQ5II1tg5;)5Xw}eNK zv_LkklpD$K3oOi*wb7Z`%aGjzEmw32K%i|GlknF)br2B`uBQqHY;fGO^hCGPpr|cyPmi zLjL;&l7z)(7LGrghwpp1kJhx#CLVymJjqj5qC)ScYHW`>eM>w6It!CE(7z2~QdA_K z;a|P&nt`FJwn~{y);Y0NZU zA$_BA9_)=SB1px;})m(^E{53R4DjYuE^k+fQ zY@6sNNyF_jPLs8j*aZoT&$LBa(|>kDg0Z}{pS(!z8bUPyPpGgYQ{=qqGKn$IE9yp@ zngZNdrK2`I<2 z7ii!r5Y3Q>AFnuyQb?D1rJ{B8&R!9Vr2-PGIL^n^A9GgVZ5-I0IuuX$QITQ)K5p=! zL0f(Uw)D_@qny3;(0wCL7ho=@6MDjkX5FhFA#C+F{NdODrT%bO`}+bm-Z8V4f`ept zsj3_6h9Ixw^0VX&bavO-l4HH{1L{tPB)-7U>XDH_MB0zIAbb3ywO7v(<`l(hKtCE0 zukOWYPR5qJ)`M{`BPXcWT9+TEw3i#0Hh;+^@YZOK;ElDO%?j$=cW^EMC$kve>-_5* zykhgCU>F<*jOa(R5^4Q02%t<%t|vr}-vtf#JZj*?MWuF!k4}p}VN0Y+_;Rpgd}0eQ z`X=OvEFd5|Bb1@~TNb1Om$91lrBt+LnVGG5u`nS0YlOAmWtSm7l-UmbQAmcIwe?r& zV&)Dc6%PAg$xOrd^D2LwOXIpLTV9{Q*O3A_{m}G#kD4#_U&jLG(^Qn$P!Ap?vSFbIlI@7EreshBLjDxMiE0)IjX55(Ym?fl*a_f5!Ts$ z6%vRzJ$N*BvN@fr0St2id-SEB*GCX-0bnyBWWx8l3qSbJp&#N>V2@htVRiPltKa3- zYz+_~TACXu0(T05zNhxH&+tdyGvUea&jZ!`jQiYKf36N8+Hz48VSh7e8K)A7?m_89 zQLKwGkdr!h#mz%~zX3#j?eGkffY(j_)fQmKM*9TJvL$8bUHCHgnSqbSOX|tK>t9n1 zVcT&`MDJJY7YW#&qWshOh&;3|1V z?SNq@am4l^=+hV%z~Ra<# zH1$Q9$*6#)pPnT8UOI!I7m*voBSJ|hok>(sh+-!GHKo&ik>%bzk>G&w>iW-@8H`Uw zf#z!;MSjd3Ptw)?#N?u@E`dQ5+=$e#8%sI>*+9+K{jTZ}B;ZZO6O{NFZ;~TkpX6B! z5cxYsEhl+#o%C2oiJAB@lg#;s)1Ws0chC%h6eIPjc*yug=ts;zo}_l*W2hb6m`TYO zbkG=CzLh$}p9iV+ZtoW;O4gubv_eLwGDOJ&8SfeQIc%z0x;Kqk)q67zsiO;#gkxn| zhT`ZbaM))bMJJGsp9C-(+O)f)3|%%$gR2~(=ut6h7que$GNRL-E$8vFE3-a#!YqBK zGW^=M8aT5gD%vtrOeR;XZvJn-@++4Mh}wDXmc<&Xao|&d@&k@ITyclzW zLvA>rP>9iuttT!ib5c-EGd|f5z41?-hhq#2(W*}x{m}QvzYpO{zYjs*1RMVR0k&h2 zw{5)_ssG4bri8mH)to2fAAWq%|HZsJ=b@X|ijD}m1m!_k0@U1^9=Z(9T*|2xtKR_g-4-+a@cb>iO@LR% zv5-sA$#N9rUICC6R%xoF8gN`gH7T7ZNf@=25Y}w*VuumQ6;;R&m`aiVjVS|U{J-i; zKvE5^FURv->Efc3uB*8etJ715a^*#4@ENs#BFX%NpK`(!2V+?oIda^GXAepesId7V z;ZNn}gPSULiA?qNa<&2mW*d5Nbx@u?%^vyxtB3?mN^%o?d5LQHLULYWg75#Mcne?z zpPVVlPZ(-fyr<&*4OjW6-R4Hcs~CL3ADPZVE(Q4 z$rc-Nx&B(K6_30a`7N}nPVvhGt}w1Iy$A9k<4duy4Vy&bv;q*@BBNFP-yl2Mm8uI$ zUPT0P9DQ@zVV>Gk>sq^|X$B}g#WIT63q(WN?#X_Y1c3Uzv91rUlH9%Z?}W_J$TA3- zG=$amk4YY_{+IIPV9PwDU}AmVQdgR*jyxFqos^O5_enFY;1j$`;LH0AVoiUUBPRLz z0!dUXHYW#Wd7q~?sR(Y%FjPig>hG&$OFPta+JvoXDirL*^<_d+z>H?dLl^zM>BN1p z$`1#LMc;gCvT7(`W=60cH616HR{%wZGE& z{%u+HoKX8Y_%BpziOXk-*qfE!b;x!??UeF_&MZ>{F-RwmXWKsx_m;qsyrrUL&HvbX zre``9^)1g+(O^Zi7vE3!f{=4!uG8-Y9hRYsfKY#WDsyRf)zv!KsLfKf8N;yGPdk-k z{WIX>_I;xi5=#WjPPm@Tq{&OqyP%jl5S)c%CEISEP!c5e4+%ROsWTiQev-K0hzSb) zXyahHR8ZmUjtFS7yG^J28;}x0zm1ng@K?(I>8-$x;H@h&BXlCo0y^WGSOr2i5 zD`cL?9eFRUI&B^_BK{E=dCn}Kfy1kum`*zsH`vGe6H9#)-z=ff)C~jt<=Fw$Z{X-! z#p8~Gb3YFKB@@OUeullks(hL~>`LM4{U0HTZx6-jB(Jaj>K^1iTNSMA+{ak|dbr1At} z4%k>0V6yr?I0Q2b`X@>JG2p(pCa9I53kU58ZtQ;UP@Y7i-3(ar!mz53#r+~MxA4eU z>Q>P_#ka-+-~JD!5QRtZq5qS1Fls9aO|#U0gu;hNDtO<9sFM2p3(1hBL(r3-qI^Tc zvICbECHhIVhJ=%7v8A5AvwlO-Pyu-R=aW(OuO5Kn_Tn%@L;v7V^f8y zjzpnYDD0FG^pakFr{H$)u51x!^{vx2W)hQIGn&0D<5t80L<;!C(c(Ul$s?H;8*-;j zN%^7PK^%9?2)3hH!DYp;cOe19o{S7jzCcOZd`vhm4aWZy&NByja!>r8s-dU8E_L%} z|JeOG*)=vN4rcwK-;2U;SZ!?#ELy~$_Lz?wig9M$*!+E@;3sE`FQw!~ByX|EU8L4> zXz0SA9J%h9SZyzg3OlR9%$+U-|A3r(-jdDVQ~LxHIhcO{qaG2ZrXp(Wb;vi=fkiZU z1fA8F!;(OYe;ui1)iTP7eoRbDSZxX8w3cCU$a~p|CK^HF!E9|<+`%zu`4O#2?=R=C+S6gVuXvg z@gCLilWO!EA=hkR@#`D<{b1P+YB!pp<7bG%EShdnK=czT+*)tiPazNTX6dP^P;bw` z=VHG5y=SggYz#i`&5?}6I`%!mwlAUCOP{fzDYZF-L zu-Y|>YPyS>K>cxw--ly>LQVr1Ie|tKe9^jAIgLA`u~8vVt%0Oof^>koTC;MXg`P|M z;W2?9(MRq1umC6fbPa9z?xMtY0O~Ts3=6-fcEsBt)B${)dNqM_Nk@0?9xpEFXx&M3 z{Z|gjTjSsyg4BbQ7g?)qOO}wuiy1i>txM79OJSf9O!tPZ=d$=jIClMk9pgg zTVo<)eet03NmT)!4?R;2BV0nHM)l+Jl(=nZ#HkX%S){J=^usXjHB`w)eLb1xe)%y0 ztW0`9J;_2~yYyf@?^@7?f@hZRF1Sa10a*|^tu3vtK0f`h$&LQ%lT7VI@?bpnKsXI! z-EA}-ok^6T%ty(Ceo}o0wL%>UGi}XE6G?^}^aa(U1y#b29|_YgmhHQlX>hHrGViNz zeOE7yt0~tRR!uWK$IK^`o?Wp-vl-o**)kNm^mbCOL_c_M;V zzG_I=?~^>?&I!#M7XJGlg3?;KD_~aHNw;^v@==vc33kgQjrTx9 zYG^R)R@rG599<1{syKe3cn7VhpA%8Pad|g`#HVl9jlNRB`=g^;9kL*Ag@d4HQ=Z@} zhLbe5Fd;yeVAYMR7AUUc*aDq~ALNKR;pe+5oW=U`RVGZCA1%78pDZ0bDx&6XxAbdV z9-ckUJSz?4`LN&8wte~j5>&8YqL3RDdyHT-<(nB(n%~3PSXe#8y3e%-}FX(JQ20+34uZ=M*IMa>Glq!%-Q$5AN)Kb7DUl z-||j1v#N|WSQ)RiR59W?D zeNSZ|U&sPNuwawFu8vbdRo6mFK;M*cdlJOuJk^nU0m8SulGUzu*q*3nkhH9O?w%a* zJFrJZwM;ToVEYlIzaT94h<>;|4zC@9%$-IpOhc4TJ%p9W=etQlkV&Z^Ns~QOx^uQ| z$Xi^5G=q`feEKg0Xd}g#n>V4x+*P>^@~&SZAMjxwhFxH_eY}Sf>_HT^6_tO|TXbpd zkhkR;UgXsR6;}N&19trL3POL|`%#DAW4ht*6MPnz^hyZKM-oy&otFYku!3{$9CU6; zxjK=zA5#@In~@$|>++wlf}>x-BkJiy;mmBA2gV7Um4^ZEI8WV`(~T5~?jbV5S_f?g ziU`|uof!*@-^Hf&8>q|cX~(OevB_<2~>pgi;-w3DbYH7lS+i#EDO_L~LXDhB_3$_{eYzQt|-nl>a){U6uGD z+Lz!`8Q4X|DcHePx;%iBfpqsAzMWt&UlkW0tPApE&M9>K#0n%35u^4WO^uxk5a2$7 z+5>`>%FaJ960i!*;j5@FDzXA9Ld2v6n_C8#hKfJq^(@ekD1 z#sCdkz#UZomt`B^8CYaLyM*1gq{GK6opCcqt;WwZ%&dqFir?VDMmQp5Hg(d|YnwSV zqdD+=&E6c;*53FHc;X6ed^C5^Is;*>J*UEB85+d)y}avHG!TyN`kv|>d^6rP-93;M zFu(a+5*W=Lg+ZLwr>2e8k7*d~@E4*n8oMhu{i%=I#rhoT;)`T`)PS-mORk{ z{YPv?fRMz7O3BshmK&wpv_E+YOPm!6ePv!tO^zBdDv3w(^$J12J?YF?G?u_^xu7JG zqSfm!w&z$V!l_)EYmm=bG=mio8iKS;nr>rY;}{^1Cs9npNDtUx4{rIG?WQ zl?r$-E_He4SZNnbA;wLIupMW*B8ghOM0Wmx|S zHaNMu7i+0xd#3Bkcs@tUf~)GLOC%AvFVR#0KDNTL&r?KXjFs!XQOLS`Nd5 z^Nw-S1kH6@RMKWNeg&>}J%8Rz*?@KcArQXgUe7S7Jf-UFI3=bcFYCzFqK`zRdFbjx zA(kRFqs{LTvE=&eklm`TAu1>E@R^q!WSGF1@g5X0 zq?>zqdsCnYbx*8bAu{ZB=peEgR`Yp=gd_<1jVJaj>UyLgGqx@rb!TlFP-g&t)tR>H) zNImhcdWG^ORL2)M8yp5CeoXrrQ=r3IAIq*b$Y>5`8+1k5{$m*B$nMPS6RM5at4)>v zv7l=9Tih#h*-IyH7T1ybbpcWpfy61T;4LTnFMONOQzZQI%ZianzW&$h^cFp{VmS++ z5|m zRjCydrp*jc_*T-3O{<=_4`8i9dfexKGJk}+-9@AfA6L>MRfLexNnSw-_a_}@TL|XM zA~K|k59%_OrqC=d0HX|c8nc(tjXQ_zfS~ zM#DZK%8bur*}uA?aaTVZc-`HCLaP026Whe#{~yXGv;jkl(|)6s=>xN@b~WgHGT%64s?`PaBy>;q`CBa5n<%2HjrJ7OIHQC)`jZ%ii!Uf=yWIi$liJA zmgE_iAF^d)@VQFmw^LQndNjFtEEiRidCf;=5()|%{8)VpT)xy0SKVSeFoiu%tbQtx zQ^0j60L+H`MgBxOZ?^bZM7OlS|nkhJ9v>kJkR}(AReSh41c|pdoaBJYX9v zeL-?5$tOGO#UAuX0{#4KGoUu1)u*a%8!`p^Rk1|+uH2Etz>+&9SKmo!NPf=KST8sd zhP?jPc8PS1?`6y>cb6=E@LahG?kp9JK={F%r#eyZefaxkD)``Rs!#`M%LrJnjfW{qI5ce- z@VAX!93<*Hl8()s{`nN-igTXW?qYehJS|kAmt|o)jzemdismjfawh+Ny*=-_FW`d9 z6#*fX9D@4By*vEm<^3E^odGp=Z}ZdRVf@+%5SAP$h}(d5>a_mkV}po-eHU zRXyB$h30G>zwkQvo<{6ehGcZceZyFtA%ko2qwu5XH5bKF$LJE*`<@ezz_7FkGG~NR zvR%LE-=}Z0)ew)iPd!QJUY3)X-J+bJiz?|H^UZpyI)7p1lgD@PL>j=--Usvcf3oQDLwQU)qro&+Z2+5!(aQWfKGX| z9-YVsQ6XY|9d@^3RnMtn9((B-{GFu`(#bnOsvB`qf@$DolF~b1(!qF_V9Xg=DR+2e zghoqm)ew7D&D~slnr?(Cw&^V6Lz)O!Z*a6}x55W+#0EVRI_q$FST2#EE;9X}oB9(Z z|EK~RxdC%V9K3>;4^uq&mfbGp^IZ%t?Q0`tV-FkFt)r%cj>nP;IyUVlHexSQSt=p- zQYi+Jb#3EiI>3~>f}p>N=wP@p<_~HqlYeEt*gY+f&09*bNAX0ktzJEV#?xf=PU#eus}oSpd7tT_)k(WfK% zH>dC)q3;0fh%5-&mMeD7-zk~^>~g11AQe=I>q0dtaaHMnh6uPg?-Hqo6>>0p=_}t~ z(kBQV3}J(1ImWWG1JjN|Nb&zAnwwZv?GI$uUx2)ZC2P(HGJ7UCIQVaM&{_#aYJ~Al zk5R-eCrm#z#PL`Lt?MQFX%)ut1UmQQ!Ux~gr~-oTpP)g5Qjd3;R3jZA!j?@7hros0JQXvMcgIvuqW9j%gQ41qEd{41+dUQ70Zrmx)$cv3v zqgIJ6b|0|22D^WYk!h9^ti|p?Lk5Lkj#6PhKd548T!IKKkISOah_@@AT(r=avt2()nSLv z%pN$LKZpDRopv=PvpJrdf_r*q5}q-N)h_xOH;#gExs&d9td&@4rCztkm@L83diaq{OMiDs9+&2*UM<3&@OK~N7kQza(!CAAE z92vO1iMrl$Y4<22790>5z#N<;f(n|+sPnRGFRgbylpsm#W+8|a$cGd>WrYvo7Jo&A zd_aOcDm5iPhGTyuv`+AM-XCu~g%b+!0u#{~7ejN%*acto zJ*aX5m|Eh>_HyZcofa6h`-^6+3e~tbY1zk=A``ex`1r3W=Ddyl0?09*t$IQp3FIy_ zWdd@TIVLw6Y2C*dkPnf$=#i%{L%udDZUn5~zEN%om|!NpHz~T|RaPj5XD|k-l_+5M zu%L#gWS#}BFC{PKKjMpj$vh1D65mRdZOA8aZu0EM9qXO@XCxQtyh&6CmEW&l;r zK_{RKBDfXqxC7a4)L4D8fX zmuZinf=Fee9aO~TXxt}pQAg1jw_P2}!`EQ+43y24UoU)fVhweTfGOIotRJy5;!`>7 zSvX$~aO6YCjdp%N-%Kxte8GxE3A*D)oFwsU0tH?j|B$yZ1_GZ0W5(+QSO0fc3Gyev zIWutS{lW-vZ@yX#5F*vaMbKkx25J;8%6ExxL8@W;8YhsQr}{_P_U(AO1TaJ8zs+-{ z;+!-Aa9(xcAWo%#2LOijTIos5Is-&uW`E}qr#O$bVqCmUAEr|QW+?w-&BLlY4Z*yp z(QYD45*Q~yw7PL~?n7h#mhj7Cdtnl4IUY^dCh$LWVP8I>B>_$20-(i9dcnWm*7Og! zZYcBTZvL>X{+kL{RUIfj0MvZA85+R`naddQzs7l*dw0^VuP)FHuzN#(Wq9@LZ7dwZ z1PY{H&2){{E%D(Da8v;7$wiH_JuQ~kkPaXDjvI3H&GcEoGP1DL>VK5r0+js!q#MXF z?cc8;fcAYOPZLvn%PYxXNZ;_s1?l1@mr zR^4^ai57V*yK$6^KZ%GwU6KSh;Mr){v_ zaytnjtwS&s**Xh)gFy4!3=K)Hv@%8kpUv$D+X@}4t^)MtfYNoil6S)p|NE(-J zBV^nm!#z9a(C$HUO@W;F?$M=hm~4wWk%Z%XVaRm~+_F_pv)d5j2@nwM3N8iMH4PZI zH}V5iY`L3tRp!&nuOMY*5XyYWm&~w{;z)>~)q=y8?z{Zf)wx5fg7uzfg+n%_Cfvz8 z=r1a1olBCDyV!W>I!dMg1ro0SsR=V7?`vGysV$SQmYVz;+@uOAAPa(d+PT635lR+Z+uge8(I-_{;u z56%fcJ|SOVw?g{L(NEaz4LOO&t9}*};o9Bk6S*^-Opok__wg%BElaScPE;V?b z#{NGyzW#o#+KR%a{r`&+faSYoJ&~^ z|4q%n809vEeB^qZr?liXvC~P|ipv0zwZl}%G>+>Qxo`h1qXPZ)b;8yb_59n1gBp(bYmf5v$l(S(JbbsAwQD; zyZ)SnVpZ=oXQH#U!)YNwhWI?t8(Z>Y;t zVgrXxrWM@jL`~SzQ%8X=X8dI}A%3zsK~7kwbq5^g`5Xhr&Vv>zDRhDFO_U>OiD>ly z5q~!KTNT^r#rBgeZ5c)nD;1plR5mjVzS)_yJz0GmR~-6&%cXv^wn{0sksURzVX`8U zIYse$2yie_>g-0(5J*0|gAAy0d88OS9S^=Z_=y$a_vRE=60<4gF_^Lk=YUp~o1foj zUN!VPEe12Fx5J61=uLroK`HVh@_N2c5FO!8$c_nz{UnQtKw_oU&xi=znxc8=7_5Hh zzphbp<#OJ@IrvvvGdwO`|)}JsJB0X(>N2q1!g_=Tl?cZ!yVedBkIa zUFpBoQhBS(n{%PEIknbR;?B+I3uB zyJoHVaA&RiZLeKb=hUfFRlA6|l&y-BqK@y?+URrQZ0d z>=ki{sl_m*$fE#{WHY>hf5Zg|q-Gvq(g}PYzu@!uEb%G!+W<)wf)4rM{OCtGW6z+O zGIew%S~N5k3#X&&9hL4RSiGyzoUcPEl|TA(%J!fz&k3FOwLX`5c|i}=`b*?&``s_9 zJJ}73eySf3VYwO|eb(x6KOHX(%q;{?JI-kPwZ)r8$_cWU@AutThYlJv-GZ3n1lF6_ z1r5gei1oCmdro_QmMG16k3}rQNecy~cUonglIMVIXYJY(GGduq7OxiwGV)51OKXwk z&g?lOCXtl^47huggtELoYVjf`L4W^<(ztwDK{d|iq2BDBZr6=Rrkhm()iNSs)gAts zEYxjw_4`SGd@l%4!xr3knWi}$aUA4uuuKwF8wa^hrbD_vz8Z=@314`@$NJ;wE+RaP z%`Smk>XWCF>3;p)K$QH6^!XoJ{{(6>;y1-bMq2wd6sG*^KBkW!xo(F-?ngk+P|2se zRhh~fvqLx5Fi4nq9y)qrCB8RWm?B5x>lZ9^p6_TqS`{xLtpWOtIzdD+{^RjB`HM|` z)aiFgNylB3uGxr`R(iZ_1Cyib`lL?<^G0R7JCO(!qo z09rw$lX=LHvE9$q@N%mAvPqLk zinNpM0fdwmwTjskWj*1hm{)@3Mu9z0g>BN^*QzLSG%v-7=6?MmhT(TJxhMNxaG#2H z@bhi#!N^)+1M_N#j*Zrj*DTe9rBi5ngiRewa^}r@-VXY?Sh@LHgKu|YA}dHtup7kr zm80j7@XWUyd2kpO4@zGFQ#>WByI)WL>;pa}$gx*Zquh>V=EJ#*5JMV_n z8iUw|3_As)p;6-`?VVsgP77O+QmIXrHLyh%L=ZSe`;Cmq%2yw886uhH6*MJvnnyHg z{8FN4Syx2fctg+bn+#Ff_`C_J&dT)PfRzSQ36E9M6M?;a&Nouf=!b`!a0C6SuvJ)4 zEOx30kqi6t*;NlMO#MqRku7nD?yD(QK1q#5OlbY|+u|~7N2XKomrn{gK}i!H0rQh8 zZ;RQ{8-Z#e-7~my1gewb1SkY%^0g*@4~U4>%0Wl7G0U%gtZE7_uoQnCa#H)9E|@^3 zEfk-2(9R#Y`b~SfevDxm9eBWQ#QjW*`jKnHilcA{7qY@`zi7g|09JBHz2Rf3J|Y|s zQt;+T2Q7xhxXteN&pIpRF3&>{NgI?2S~gk|cP?SuPjsGv=$jMogd+bj?dq3*Tk{S_ zC4(YOL@JJ*#?yht_19-<&2IsvK^d8{oA~)^EV5<4`~{DIo`E32ya1|7p$Ii$;5-N` zgY{67{W(hfeYJI(E1h1bt;ed}19E9u%X>3VX@vF>P>%th5-Y7?XT0nHKK{7um@|9D z`^=xxG;(9Jvd$wLzcIqE-v zZAj$Bn>9XaNk|_BmMntCZyLO>Vk|tq)!H*AXdtx_Fw99)R35U{GD+ai8cgT?@Awm- zCpFVd33U(6L#(qlw0COv=ND1fF12Cz1?AxbBjd;~9=X zlgIh*;3`zxbBc+j#a?HKg`#F$U&_N3*H)gQbQ|cWaRCkfbxOh(>XHS=;NI2YvR_Y% z9_l2ZCYa{RBEy|-BZo6G-ZuqHzW!iw@VGP`GzzadKH7wLT9{wLgI{vtRlUv~MRLD# zPcN}BPqg8YZA49Ie@{`$y)~|hwfz2$uLq!m4d%fs=`y>N-JY+sB`1`&wcbDYhukM}k1L(rztbEheH#}sI$J!S1n=%Fi7uTjS-Y1? zxNm$xx}bCT)z3{ybbIK?UXl_0=6?#IP~JT1+Gm4qNqlt9Tm8pLyUB`j27+}r(%u_u zLv?@$m|Vj~gKKaw-ho*)!rPGjdWKb#`8SZwQu#ju*>pvB7i^YW^knjlm(FgYN~b?t zCSl2W;-X6K4Nvp*(c@*~q7Xu)4BeNVrx}A!Vui~*zF_ZFs7Ib4Wy;a&H~FkdzffT< zx%RDy@8F*xZ#&#8%NyAV0 zpxTYfq`{-NFz4%ygkk=0Gm;G5RFZTF4uDiOGQZ~{G=qx@zLI=oGh*bvO;paynqT3w zRCvP^V{4Q8ngbmFH?U;OAZ<$S30D4NEsOy7z zxy;Asosyv0$MxOY12d>+Iv$F-yFIj9@Urmtg_M>Ea-o`&n`TDPfg3S+$ ze2r9BWITp+zlVD%j$JT3)Ki#Ofm$ed zSfTj0&Pm(E<=;tQm76S6K>I6TrVjpQsrw99=qReET(#C??FeEXsq{vd0LPq~Qj zvTvYWtn#i#J$qhd&O_E8Be^pUQX4*8L-+KI2XBs~H}c&`oCmlZcOOKa)EP&saW~bN z?P`lYHC(8VhHhzeAlr6$kRAkA_K$eRt^p?vz2ecbG!yL?l~W4`pHHV1RmYHK%)uPX zsu~8LiUm1vz$gA?EYItM#FoHWq}=cBe$T5rR7F4N1$Fz3EiFER-cueiMdy# zmU(BsW9cuMFQj|kAEJV9D_y_($&jK+H;Ie`VS0}0b^c>)L^Won?Xp#i;VF*Z$HBy5 z_RZWj;b*4pI@lmHbE9+ZvI1{bnKk#gLf^2fWw3M&DQz|NfZy}fu!nTwY7S+Icmr>B z>N4@x(y4_6!|p1_0$hj=F5)e>R}#ScVpST1GJPiW8H#mj6@u1 z9&?%VvUcMaoFBcM>{5K z=WQ+`vn9hU787dFE%mA3>_p>6nzNYesjGM{Xbgx{4&I&7UpDZr_k7(~lPNzKiObHl z(3}V3H}|8kvGlE<(D27ql{yk&^yJf|WF}8kvZx>TCR{0mYY-wvZ+Jb)TXQ-JclCZ@ zUlhW|GP}Zi%4~+dzL2C@5-BQGSM0<RA{vE2?ktF+0%_2|jvgM8kTkaZ%mUa*+Oceb54Kd4ZWZPl@R4btiF2y;Nd z2)jZhnDRv}&?zKr!YUREZ*ZR=v6Y{4V?R#EfT^pY%WwPKgAgD6Esq^Fko4Lp-y+)I zM6HptpKlas)3Q>iD#Tq+|J*9%iiQEQmVb}~oPtSIhwb&EO;?}Fjl(-A;Ro{9Bdw~4 zuk{(^%3$yzr>%#l3d8&cyMUF7%+r@$a{5|2fj~rmTGxUS1@L@@W84BH1wmySmQV~bl=}G`5p`mV&J{+hxH3l#IHSwbaaNkbzttq29YD}l7@o9%{3S+{SWl9;hJN%p*xuK zA$0vU%?V==oUr5GPbIjb2zGX087mvuW8B#Ob38efVi(yz${WIH?&M^$r1qL@*{O5X zT=xNPfw9O-IzrBnS&p7%HM{#W@2I5rM6Dq8sn_7k$QTx8#?e9x4e)13MX4Q_%-WP5^i3z6fzpp`nq@Rhw5k0Vm7hNaikH-06^#}siA0bJ z&c2%VMaDsO!cQxtfGTS27P~dacw6fQY+K{^kSV5Q5X4D#$l8y$g`puM4~zjtFsa=u zvz|Eol9T4+H)wCc!*NOOpI>h}Q}26wljFq%6VaGjL@)Nu5FV2emeY^5y?!owT5(sa zSuNyxJZaY1sAoBCa9g$gD`-BjY>{6ZQO(Q{oteN*B97a^;d>Y~6@s`l>zuOwT)nfk z)P2!;`khkU?ek~U`)YHqoEkb$RUthNT0g)_wv@+u$TGRz)D%#=siY-Z`{mOKl_~rA=lV?3-8V4bNoPCdJ+ODRp%3xhefI+M3Bk z#{%NXH5Y=PTdM}br;y+InpBEBA|JR09!I8oG|~^d64{I2`o6~MMpPAn`0Mhz#we{UqbduLZ<8eVjM~-p zYbqz0CRR^Y?e9Yy>V?h(p>Z^l;#yY0o~d_ev|*b|n1J?3Dz}L`GB_x>xt1{YBU`Ui zL>S7j^(EazGE0H0CWodwk^gf_qRXF^J6_E5LGj{(Yt}g| z%J%MDRyjUp=H(kfDt}ywXpay!kXDF7O zPP1SJ=6Sm`tnrQ^qZGy`g-;(&Qp9%i$-l58zf}EQMh8mwgV^h8&+$*Iu84cRAL)N#s9UoGT zwXx&htJkgHI$VX_a{1<67%FZ3zB2TM)?~2)czVZ_E`~8E^ur0#>y>q4F^px1P&7R* z`nD4TAle|E=rCP)A-i&8`{~&(9n6cu?PkdQ(2Ew*QN(&@!7{@R38Haac1J85 zv6tF?uZ05)R&HwZx#88Y@Dtu3Qjq7M4$FH>us)g$Jwiuws(3a9DbjjWdakX*WKfWk$!(1HmA)tnmuU0W3Gxsp@~ae36ZsYzXz8El6ilo> z)^&Ti8=$3;d*YGh4*RuAc^cRPxjRcvvaiXg9Xy=x)P96|GrKBlHuF*+e~f<|G?;5e z{X^|c9P=@rHXN_&*qhPf)or$Oqpsmtepk7U%XG{(=RVmo@rh!eMqG|4@ywdwEaR%A zRvC?+v$N-nd0Ke%ZM>n4pV<$lWgi2Ll^Z>W;ISl#5PYVg2wvyeMY9S%@)(9BoobCk zAnD6m9od`XHGJ`>Yv0+gBy+}46IPtd;RXxSCaODFI(0^Gp-J~~0n&tNUscU%`2>{i z^Jmkxwc-=Z4|Es-&+-G$NUfjQkKYOp-}U;#e%l4}3wq;1{(S!*FUu!hj$1fu>CodJ zA6&@6)VUjLYcir`GTIp+xs{~aJpE3)m;J8+_0pL$h}NR0k|!aYczN2pOMt9@a*2kV z+J#3-f3FKPDAGCvqY5+~?+@FQiur7Fs0jC85>%_RfMQRzYRlp9`4M zvJ*wh$0u)G#UxN0)iFsk*+s)WgOIq{=)3Q?o3VDg89kV@Na)Uw)IkVrw2FOD{TWy( zVMYR{S)(5CXHiWh##Vt0AayU{NV>jgeZQ+$z)T9iF#i>6E$&1 z*ejs|mP6uYJQtbqPOTa25RbL(MwF;ycBnR-?P6G5molE+q@zw|-{@a)xez17JH8O3 zzl#&#{iZGQNBeuv#t`wA6^(HUcFUE}~gfZ|2OPMI#KFA{Z&^utMBx zOcR3*Ey*iGM?cqruh;CYq8O;TK-pp%Z;d!*>kn>MNpI$$VbQt7obsI3j7W z|9b|IDO68s^?*l_(o{b`VfR^nkmL@6l@7(We3`*-XVpUZG&#`{!0A~9O}fz^g(bl}tc=&GN>iOZ ztJw}Uj4Xv!h|v0_0P5A8-SQVJPHing=E3twLLx$nTk7&c*Ka`Zk>Xpk9^j-1Q%>6K zwOgou;*}?gwZxOjIO@6?@9#h2CWMP=hqfat_6#r!&>Jn4ExP=Bb<{InN#>Cm+3w{I zTh&qDSc@mx^5j$kq)9@lzMF{t4wjz!=Pu@Wxoe+<;+Gd~sE_B>qL#nSp}gs__5JZrRo(9vto^S~GQX%y$_xcfXP!#g*bhVSfIKV$VcZk$r4+czn zfMOFB-!;a9oR0*T_PySOn`FZbP^X8_f$9$H_$nxxw_5aRb1cGkUN*Ngzu$$t)#a7k z+&W8Ids>)rAoun`;OoigiB#?;$|=f*(T#&Mm#mlv2;-^WfpwW@MftI4c(<0kCIUPQ z#AaXUh!d6=UzKz9GZ( zcS{X8hQp$zQ+;(Ky54zwxt^6*%9s+HTFL&&cbqH6W5Z~8)d^DT*GyqMSQ?*JTubp= z&Cr;KPF{#vj;B!k9;L-C;;Gje`ZDNOJ!9}9wB0TDaKvN-=2LIROL)ZSux`6X!pxbA zWz1ls3+~{1l6}3qYuv+_$sd!B9zvu6v3_FA!|yVf-kfyQMD9rXM^te5=JSmg@`R*p z!R>FQ`J;{yH{V{^sY_egM5J>Y4aI0cqC#Wy(gVa_Ubox*ToYuNIJ;UuOp<>!XoSyT$o?0+n8YSV7^T01{BZQPc0j zt{&BoEiWoj#X&8MyeXqpj)b%LDs)%U`jEmAY|qvR%#@#;EJXYh?+$fi3yF{B=rX#t z{8vM&c+5kZJO}SMT6{iOR0=8EpKr~%e9C%lr+*=#n=2}n`~bx~(kz8J2<5GRV^3W^ zj5%Q89aH^}ge;VL4j?^fh-0#@AmwqTj z&ibD14Ao1ECNRY~tgwC7_Ww==plxmTHFI(}MCm4u2AxFb73&8NiW6B;=e3un@)B01 z>OEhy!$~juTutUwcWXfB$JBj6!<&|D|F@NA;K2ZYC(6;ra?9N6%^tg~;0#YV@63RP z1*SJ?k&(*YlBF73JN#4gc#qwDyiHEcK|c3d8f)W>Rn)BhR}wdmi5hkiQPw}@(j5*= zy>3y0t`*^`s-&UmijM0tLW3e2}4hpOC9(Q$vMUL$mR*F^3Oh(eUQNK z6CMb5nxEzABM)Qp{&D7aQ-i#1TIfg^Vd~HL5l_eqrl=Qb%k?%F@oZJ=$;mb$?O=wc z?eyYJoVB%$CB5~~w)mTd%YM%Y#Ru);>OM%o_)bHld8q;+xQ}Ak2DnY@>K>jw>vAU) zBaisPvo!8|cu_=8XFNw%ZA4S;E|{KaU^O4apkg4EO!K^-8Gd4WA-WWzbsvqeAJP^Uc3}J$bFH0n_(by3iXwkB@VVJXwoOj z<*;`Z!p(z7Ng##}{B`!W_13YRhOOA;Cny5q0rym&wJvc2E{bJ4f+yYl=LQ=$6h{O- zn|MyT&aAoqd|tNDDf1t>npam`Y8X9nKatraP+rjW$P%DsDy5Xc!f(Z3Y0S9CpriKp9t zR$)EmFhy3%TKrR5cT*-^A1vaOhTQWny^`T%HiXLk*Xu@i;~DGONU>bUj=-rm8`c|Q z(#hpgn32H5FV2GX4C!T74V*3pbty*>A~tSj{b6Bs>-3q03o+a}5&7lwSusfDN3Qtx zM#;0ktcIxgMk7%{&pY)2H9});QjOelhrwruTW}dEo05@%9ZvkkNogowff-a1zB=l( z4M-0!VA$vW@l>b+$i#VXw%MPXT=!J&=#?^4jJ=3FK2Qk=pgb5kTuGrlq>ulset#o+ z&pXW}UPKCc5ij=agdvD9+wkK3SM5mfGW2T0gIg7gdtZGYgtm*Ei3;_!$8_RKa8mx+ zfIatqQ27FnVlm^;4heIbP*EnhJc6iPnC3&V=u5h}vc!l+9+I)Ph(lgWrPB}7o^VEW zufw*}iYDyq#3~QhSW0F4NGBr6Gg76;ex=_NL`;z@$82SN`Zk6guJ=w=8Yt+J5mSpjyG*r*^jbUWAjjScQu!(Ejn`gU$oXD}_7*OmD~u za*g0EY-FkZvUkVCV>7-u7U(;3Fm;GN7vxIwI+wggX_7O2wX{JaUiLFTsM+&9gA#AQ z3A=>vWUj3W#vNq79Do1&9o?PW@DhWlG0Q4)!;gUp31H}S#2yF3v*;MZ08|yz!3R^0 z!aSJF+_)fnjo8cmV&~He^7t9`e6aI)3yvHPI&wD-9Y-T`Xv$+52eg3jb5n8^N=~mw zOX%Eh0nW#a+GwX&7cIryo$X7(V-B8=-a?W-r@7qXN*%-j~ouBQcE(>D@Qhjvy(h>)sM!axKUG zo$8jA0;=bZH~Z+8IhVu;TN2*l*>hv(JEgBOnPex!ZO=q^kJHUC40F-fi+ECO`GJ<^ zN$(7|Mq;c}2SwV%YcUtx(k%IZxy?gB1xxqaP8_H*)%KSJR%3>NHk)|a#&n%txIl%2 z;_@Ur=Hr~x+4SUTMVHzua6x@`kT+r5L=;s(5vixX z0Q@IAeK^v8+{Uz5CW>GD6QS+M>xAp4_V*Wu5+vRtN4uolf8(@dISEDkFs$FPDG77YR{*}$6z7@iNsrWpVBE#J`d z+`SC(50c=~W^&`&2Z(6xU~@I)DU30s+90IN1z)mUgbv_zGb_fan0vi82 z&&PCpaBnvjcX|t^n$?wAaS5S1fH`7f+5ACn0@mXRgVpgSJnRY{+@VI7ax1;|$}G$RRSCk0q|CKVHp&$hI0?~G~) z&e*j3SbVHCP?gx-dRZa--v7M*nF<#%4{iG1iyrV^cc=~o+X}uA{#F)?-H?92qT>&q zopQd+b=O+!qTI_W;Dj{lPgixiRB`J+_c{fc%zvfgYF;eG%()m8Hsr7Bqxr$snz7&0K}+ zHKFND+_@5SA4)b0oFZ*GfMy;{5H}--*Q?@GuyCyDTMkR!o7^Jn4-&>i_?+db22_^e z%z1-AlXD=kcj6tn&wSNNb!|LgIfBgapyW%da7JtSk0F65OJ2hb1^L@k$-Sj|;0mdS zT}gk2tE9YQ*WSI#^W0+C8=aY8>_N10{+#Zxf3Bi+t~6E#)z(Z_gb>>mQ&Hd$aRz{-(BPzdg5C`DdA> z)2Xg5i0lt|5n7!Q!>CyRl+Dxa`T5$4#=!N!9g1@1pBmXd8*{+QPoXAW`b`x39jkV5 zV^94xVT&635gy1tYX{P#&|TKWbpJ0?OE=4&UFi?_?5cz(Nz z^W4wxCgt!r?8jC4 zprHA$RBy7vs{>dIqCM~ux_C7Kt139j?kkopSR-Jz)dgrj#ul&6j+r5YF^I;mqj4p#% z?CeSr*6lyc1GtNE*jNSZtsaXXmU0%31hd_v-x*Ny$hafOCoDWSey&ggiwHyMWh%#S zniZ*RQJtPf=NXJZ59-K~z|ODEbK%1xA@{LQN@m1kUUZ*^W)DR%9B?y4%`lRfV0M1v z>g!Vr%=Y1%em46HaOv^>0$gs++N+cO#u`DyxIfNDm0gQv$~s8|62wVyEQO&yj6AMR zi^=d+_Wold)%`fCrAoceK&73Aa;MFHaiEJmmMm6d@(zXlc_puTtMMr0Lf8?H91-d{ z)}O)<)&1Pm>Pj;6Nntm)wH0^HL!Ow6HSw!NX0W1ckzJOAb>PyDHe*l}vQdo8^o6^5 zq9N84l`SO|uul!PCX8*A2-goe<{U5}XeB!x26j)vspR|#FJZ#L4Eh1rOKe@aYu z_lLgR3MR$Uk(@*~;#L5m!LH+A&1=-b<&g0JW410h(($W$zcK&RtD<6N{MaPQ9;to* z>I2+Fy$|ZLB*$2G5PIq?GQGn&8s%7>9o)&zNAFBbtLEgZfNQhSMm-~6P#87-zH0_S zZOhL9U=Q*$rFYRTQ&hIm>GquGC|Xpo@6eU|_hMVGZ0~!w9%;lrpl{*u(UHQN_3mF>zLrJaUlsqDI9CI^_22b%=k`L_fiy zuD2?lVu=Iy+V3-9%p$%VbMH)a;J3^s2^q9pMmH*3Dt#kUMx)^WO5aIzUh?a8Aozm+?qHl|&4{LHTA$GeL{ z%xr5kyD+_#-af`T2H1joaV~bcb%f65>_X0e#Cg)4m}h74>B+PT2r*b zF5h|yZ0_k;o(7osXSrbtXH~ay4!HsIk^Ou>{|F+BwS*C(m^*%ObDrlagraoaVoAN} zGK}%2G>XQfqbW-!c$#Sj{+ZL^rO7erqRy35F}(T~MH&}XLhyu}i6QD<4zccHvt|+p z)JM`)3+LM45_@{Ju1}8p=6UjJrn9u`#p34Tp++{f+Q%Q6b-7loKQ7Rg}gKJgBflPLE82G7E- z-Warkf~Tr|T6@6e`HYrpdJbAuIZAt`#m-eju%Doc=0_5^$MmUt(K|QpB%u%NzO$5> z6ixdAg7u^Muj}5L>z&um2A0fHNQ_w^3$!zK#_?K@QAt&qoggyG2jA zorg3GS2vjm*70S8n2FSX&%f&QcMQ_a0&~~EC}dBRI?kQ0F$@Lkz~3fpeHB|NH_$Xo z9PmAupYGq#~RXQVn7C$E^C4oAL@ z{t#suHN<+WC4PBwzpEu^KC0{fQTJ=*ANQ3o8y;etNq*_w(a2?s^L^@H52q)+l)1c) zAaq~Ik0D8&heW8?ex*`a%ZxI!>TPCITkkaJL$(*o8$zhB?6pe?k#F^*pG&3Ot+C zfWqI3*&OzBQ%Q2X6Qu^uN08;YyRzJWNaKkw;R`Dk>&}2kM~wwkAi(8?oBw$*9bvudw!qrkHRgEsC*e!FYb(MwM>}OzWij?XP-oD0N zX}7puWXA|jN#u{Frn;;5eX4SSNYP6H-u4(<>U_tRV;kx5{U>R8D{m3!V>Eb0c2#}&| zW3!MrzthpV{`fS{E_xaZk9CQlh3*FLFYOV3Y#jhK&-p{IwH3#Zi%LOJeQuGa^7HSX zx}2+wXMf~>kpj_{IA}=E5v(iDOWGbp52NATcU(0WKrSz^BzluJ!hk3%#IC*BBIj_P z4MQVo3G-m(c3UQbh)tco(bv^lP3aRvo!V5`-b#Xc<|up8M={bq`|_WxJhAfZzI8?X zyMlL2FR4NOvni8A|J;(7{Kbf3x_rCyh;0JmG)&ck7(xTpglRr_>fCiUGn(*eG-##V zo~-&NZWjCFC(c$VWkb13I8Q^;P=_IL$W?*&K#)W_3XM`}B1Hw4*v$$L1|_YuCI2Gb zqO(ey9^tt*8V0U|8+J0)lyt`Q?4Bz$fi|yF4$rt-let=7vh9PI)&YV1JA95ngs-5m zA6g=wdV(xq{e(g!d=V}4XxR0g{RZ0ckmkSk+DCRDZ_&Xw8uZg%oadS*#uey3{N z0tBGZRlz8SVm-bB^GE(M2~l-I*gkw*XfOZn{<}+R`CVh^Pn^rI`Nt6L5V5lZYI(2Q zV@D8+%se*z{Kozl3m?RKZchiTKUUN?FIk+Wj$(A!Y~g&IgBrs;TOT(+f$5`jPpA>B zO-Wb=y2AQqa+mGKul>(M{*qwy^QmMUEiypq!o!%aAaVZP=wAM3Gs6Mu9j;R@=$vn5 zcHAWzk`dC7-9wFt#nRp0q*L<#a|&C!m#|8=uMZ15L1v_8=E0_s0x>RjJ5Z%9c-6XK zVjnQ-Qk6ZRCe^$PzI|?%>)E7p+j!i48QkQztUU-xV9SBs0o98z?7}*%yvQCK{r!5O z2-Q`uQO~e_9iLtWkDzwP;ysxr;ndD{F_@U$B#)>&k2Xa*)|)*)y^Hew{FQ95=2ohX z#K?UsgPU*IvaE4<%l^Mq{!C@j;{lF_JpYdZ?199HcoWrD<>c$L%k~}I3wgw!obajY zE5)7r_fJ)m!ZzWgSWD5Tdz;}9L=&+#*0s!SLb!AjR9+(RJV{b4BF}Z}dhZf|M zVKje%*>{{IqBU7CB^2C)Bu3`n>X#}}uWid%7!6(0zDR4gm$Bd+jmPgakh>jN-KdGo zjC5ly$QXY#qS`t1ocn_;-yUF51`$ji;T9N(`8HfG9&fl16tH77iCTDCP4d5vSLynb0Ea~T{o4JFz{t2K334OLH2>v!4L*#jYzi)D>!>)Z zJqmfq^L3$PjnAx_06TIoW_@T&_Uqj~)2`^Y^x!C^S#H7bN+-mEYbt3fIjX=%8>z!p zeh^S+9j1woXr^Thk`!DB$7gwImdWo(n<=xZS%A&=;AicKK|{WevG$6>P;YQa7Z+m> zRO9nuue+72T~{0PP3xdH$l1(v+Euu$dM8qwR47ovXo#WATK#(KcYm@A!OIXXQ&l|X zG5JrF?xs;UM^!dg) z_Pdgi;BFvpSIHU>lC$E!S7JR9d=eed$QyBS)nRugU0=&TS@Y-wW}kZ$MOD@qQ}x1+ ztOtYqP)6wzI~PTAcmlg5AybSTpH(yxZB1&8n*|YkG*GKzQ(nUuz#QgWYBS#U9RESM zRkeuR{4Q=06IDtBI_;t+)7Jheql59>EgUiBHe&(!vQ^!Dypnn@Hs5k-U21P7NGI3S z_Ei@~Uo{#5c(8XbW8$T&>CfKJioT?s5Wi|wNnn53MW=-Alv@xmbKbQJ2oAo*TgEi_ z+`G|%lwWQlblU_LUdJDV4P8A1uFOljU}^8+i=5Gm6jG}+>YXL#&c123XE>4{t^KiS zBoRv4Q)j*9F#xfXZ_B3NGe4wgaev4zarCG{4$~VZtA^<<%r~LK&qN~`9v&R{(DT~Y zAQ1<0r+QEe)ywBbuv&4Vpn6jkfiifL+t;1As*v*tPA{1*tMJ zq+}<^B$O`C5ipwLx2hD6#(MgjS6E;R=&}`qs0l?+XjHM5j8g@@=O7|jZrE0H(Jk0) zt3K-P#jB~IAIK@b z^|$;?FafMtq{qy~uzX?6{l+2sQL0;&|MP1>LHVtwyk+I3I&RIyWlZzon7isMv&O$_ z0@o>0HY;sv-><@G^QzxFp$lb5RDmuH)yM20?BRy-yzFyk-aW&py6B6=Pb7|*ojxga z#%K>vNGIQZMY#p%nw)A#;j0O>7)hI{A;g*v#nDo7Bl+VR8i=857QRAylHu495o)#< z9MFthq2rpr987ad4;&hL+?2i_Jv7$^*3y<8S;(o1eJPSa3FUD8zn^lrY39reNQ<+` zrQ;qOfXmyhgqQHP&bFmK*znpAq^14vk@6)R#=j;Tr=T+a6!2_ZG3^!7CN5VJ--`y! z;Z&=mlGULM!WFJ~$1o7V49ryS9uhfbG6@y?t|!^KiPU?H1STZTPSw6ml9ZZx{if|M zeSFQpmTqAN#_-DFvjgSF`3<2p*!Ms;k#Ax`!{!SOO?4`(K!|&x#_NQTQeb`iEL+xU zS};_zuRL46V|Qu}pN+@!81hB|$S7z@#i4_#ericocdCOce6zjtwu#SJ=sdlKB5?XQ zM%Byb^C|4e3Y#lvc0I?p?wV6*%O6Ahu}gXw9}y!aTl{{3y)xwbM>UQ&Mh6%ANLf zj%q7{AjDPdhODc1!3@23r&V2nYc*JhHERleHAK)4Mt7@l6pq}Dwcs))nYFtawM1P_ z@Eb|P{=}cCW>~?(>vJWIbkys5%c_X9D+JZcl(!KSbzAA^Vy3}BW>7fN<=Z_k-q>(| zx(oY9#Ppoj=sn$MTvx%Yr*s9C0?}Nh0ldOtjChh3pO0BjCil$f=>-GUe6UaL6)x#Q zW%ALgyxOzc+aN3td%j3jeAY$;X&D@a)^#t|KeOhINx@lz{d%sQT(r}~t(yfC?jEY? z+nd}i7IdByetB3&>4I8x<05ZE^zA*>8N78o5dO_?nDp0fmb>_6o{2v#K3&*eMrV)$ z!PE2waYa3UWl!uB<)FekE%bu~#KStA%cPQ%7E|Ns-8LsI)&p8OTDZ}8B z?sw*fF8_G|!%?s>MfHX3o8b36cQGzhNv{q%anxk{?%sL`x1^Y-{FHeX-tTuYXoqm> zzF9!_>9X+4t!7gbH1c}i`XEJh454dRElYNC;c|^Xj@kMk5l5SD95Tss5^|tiC1=|O zi~`mvq-QRch#uHKtgquBJQ##n+@T&x22*;)#Qh5o(h`HvNZ1tj)Z8;4#}nmq*s0A% z(KY%>971ASQfa`MHtG71J{emg%Y)s(W$q~LE)>D7^u*{_czh5c2l#b&@KVzB zbK{83#|~f+v~l;4j4BI@hA-B`2e@Qo8)-uia^HaCx25^sRsIfM|9D#scdEJ~mgr7S)cTHzdA4=+@cYzqSV9pjF;-bV z_mrzF>yqpWrgQ}RzOrN02{dEDNa}}s(|rK2o%VGBpSxw zHCr&wZD6AmD-(|0$iY?!KXR_wgmLzP*KZzDti!&xe8x7CP>?UCb$24U-CaX0^;NkX zDP7q`Uo1+m+aHo5+z35;BA8gWVc?j=jC^n73By^s;rO?TDEu|p<7i+hvUlvw?|Ow_ z#eBVO5Y_bi{+3iX<)5gz3VU$pxMU`S-pc{;0rI0oX$|%o?@jm{sS+#fjg(nnmURbw zi**YeL~C=t^_~{-TiF7r@I^bNL!#bQP{WAvr=UV@SjQ%F6h{B!M>8zcGnc83wG6nq z24kB48Y)FL2-zi==eHoClPUOibZxXHOxH6GcVFoY5@jTf2TlvMHCOVC4$OQMlj__Z zj;hohd0c(!ef2nrQx=UD869W}!tjSt9>ThM>dsbBTBn~NQI5RokKKri2SFwd&YZP= zb+($A76=Ueib8^s3|VrAT{>FeV>oOj z@}CgcL%r%t=}(#bI3qkxcp~+s>W-{l32;D3V4rKn8#5}>Mx(w~Pxxx_7-(f-wc|n2 zFN262Eqx_Bj#g|1NBlI4C2uVS5m2*xXq3MfKYUujos?}zO+N0!g#;g_;(#Z(n1SN& znHtXpV!H)u1@SP?%1x^a=d$ZnIm;{Yhakf3iV}nO@eFQCvOTrJxl{Hzi;TGDmY>f5 zuaua!s4Owa8zwfHK`L`K>;}HyS$UxrM)^Kt_0`W%Ar0C)q2werlP&*ZXhu)qHP_R_ zpg{Ex#^n3a>My0u#7O`1rJJ=GXWJ@<1QTr$)1KGIy+jLkY=20#@*mGxhkfsyEtKvL zT$O!T)PGS>7MpFduGQZ|-6-)$-<-pWe6zIKz%i9_mgt>(e*cRzlmETcK)oADdQ+*D z5P5;eCcK5Ihu&(WVj0P^_!(Tel9AMW;V8R(kMK&KV;2ZIM|PwU)N0z=Ov~aG?aoGj z8GQ1++%9O6hf`=c#SS^bJUbgs^hm3KK;^)PuFAxeLrjpnLil-q1dHmx%I&qZNz?Ob z;TK)_l^bh&(wb0~!GnHjmTE%oYQ&ZpW0$}~sZF!-BlEXfxD%jFq)*28svWbNoSroSj+#$u3Lr06*TRl zxCDZ02pOCZ91?;ACb$I;?jZyZA-E0_+$9hKVS)vB5AGHi+}&LU9cGxd`M&*MoSSob z&a*G}&HGeWuhm_>s=E8F?)Bh@|Ht3cfa$-5k>Yt2v(>AgfBQGI3*`SeF<)gq^E5P~ z6|&TASMz%N*`=nf)5mbxeuI~C=dxDcOM(lEj_jd~dYrmvd9#32?XoxV-G3s9`7>|y zw<%c!{f#>Atu~sI-~7V{{K;KGaZ`HeWX5^{txEAg82Si(J!6E^nn|{5cIzpf--0BNb*Z)}wb(A1I9Fu(fbLMB?F4 z%TrNt=<&S#o)h>XUrF$#p!UiF_6m`U#iP;t)|&vNIm~PH%O_}Ed+W8)e=aponX1vd zWLjg-G?U*+m%Ke1_2{QH?_^4$Z8!@S+AC*lVrpZ+Iolra#94>9<5<{@ zVm<_-D2+USg;y$4^0p1Ozz0w`tPNa zY(Jur02YmZXrxka1*;((JQAQAzeb`V#h z-p3Q3%AIuYlia^Vk;t)6w-3lF3^iJ)e7fN+6R%-8qvvj=w2H4~wFgvr4|~7~nc@G4 zfk5E)qU@fW5?({PLGZKIf8aff{gclOu}qi$#2u*`WPw?+>KimnZOBMTqJ~uMvGhWd zvBxd}(Ilw^+uBhblchEJvubeF9>s+E=iXMG%w}g=Mh)Sq<|BZj54;<>t5A*RAN?%P zcIr9+{wkzD_Vs8(a%jix{aFT+O3HNVN1;jtD>nYBC$c?9*;fZ27p#zpELi?}X7m_O z-NtYr${kj6kOF8u-5D3PubO)#^5)u59!>aJFVjX5m$!RZ^1?&~k3f>kJ9Q=xh zRpgG4s#g3>&Ts4Q9!3gycrD&nHp*r%wCj2v8EvKBcF$xF@UH8~rY7A%zp`sxfj#o# z-%;Q??cRa#5H^1!3sEq3?N_{!!Pm|R&*y7n9o<)EV z6JOm26w!$geQ94hSy;#g`pRoZU!k|?_Co2;Z>LRY@@>APavz)BWA3?MynQycNR)<< zw5>9A?XL@PX@$1L-pMnGHB($C-W4XD_OM5kmDLFM4UU}sG&$*LCb!S-^G0bZz$u{`%4qw28J?6PK+7+G^Mb5zKE{aCs>b|3FN+zP@LfCuZ3sPBUT{d`N83 z;V*YpVBldd2$Y6c`VKGy)JX@U;yH>SBsTI>UB}GQOvFwOJ~OBSv99yi~SM`ZXrVyv4wI(uF$4O^RYP)<0E(2CJeihl#w3 zGk@Qm*UTZ@S~L0!sEl8kU1gm$y}v>1?HBusg)a|=oZKQK;N#U6^@fu;HWiY1d|U)F zs>0V9FM=`8(y(oOwi;JH977P(0;boHSLi{R+Z3hfPnJmYOb$IYed^q#mZ1)_0Wjrj zRt`r-5e7vvIBfeIW47KQgLR?wi4e8TP2qlfJo6uDC~X5t$9c79=4^jAH#Z+zlw*@#PRtyXmaF-yi1EAnG| zWN5SKUmMigvjO$ae?#bBi-i_+`!q8stykUpUA5Tj3MSoCSHCfuXs9GK0QL&u1035h zoi%ebl=1LbE5eJUU&tIlAA6%8>dI2?|1^qh<2%q<_7w3BN)=-FC~Ak!R`*o(bPOl8 zC~0%?Rih0X;U`@12q!0tDz=~on14`L>n&O#g{}z z2XSs^EOTkevz2a8R^6e`!!!4u;;C2od00K}h?Lh4=#Zr*Qs+;TsLx}1P1CVUk4Lp8 zvFX|4u&oPzXDw}4IB96|!LQ=;N5!66JFx$Uvd5^O>)mrCTfwNM2h-TQW`cEPHMf9 zF$!A1BR!^)qQ&q($DM%gYrhq(twMu_%&!>XJ-RTCHB!$H=hWNf{%nu(`Hu)`hi;A^ zWfs$c+lh?T;-0HR--{Z0tP#_D5qN z8EfQeHDDcr!hVE3GUJQwoILT{<>znTHTaD`Ic9wfngY`$7@|`|YW@zicrzek1%bNK zxv;*KX2c(Dp$q>;{#VcgXOV&WNbc^|A)-Xj2UNB(?-chD=dNAaUK|M@XXlGqjf*zM zyZ(t+$E1?B99yM0p!~V8MPseP7-VcAOcaZSM&NCQZb7CzMitQO1n1}6AqPvl4FTj+ zz$NecRUxpPH3htX5m63#Hh+@4?EH+aVp~&xhzWSNa;j+_5bmzBT)I8C<^t}#urzTl z(dPzbD%wu=G4*xm{o-1Dy)nV&-(EywaKCLFK{>T`PwT8BRkSOWZL@3YSX~4#Bq(O z5E3_qXSf3i{xa7v3--J`C%PCXVKRKobpXu`x`Y&A!jw_K6(ekt{MKL398a9CscXKBB ztH1_vz@ZG|<{yqMK#@P0JRKlE$lZR^7UZD)jNiNU>DKDIOP>J@MXg0*!n=voHlGDD zcM9Mw-#0Zy9&3vjWQdjACX4(7Pcb9xYAv1)C3NW7g4*XiPWlI2)C2bd-(tI>=ht-D zvf=@@Aof9I`z6wvkV0D2W#$1~lVyKeOy0tc4p0&`9f1D2taE;(NI zMs(*3am_bvJNxb}2D+0?Mq#(M`XU;(g^ke0zkeSUc*&jh9`rSo^MSRsY_VXudDuIlyXtU&C7Da#9qp z=XQUg$D(xbAcY}7gvjXbF96e!VV=YFN#0ft!J zSMGNbk*HjU(fuGG;Sxgq_^29iAf?dKg5TL|e_-He|3lKJNx3v|6;s`;u}$_c6!!Fv)Y^k{>OqwpMB?sL=Nx9zpv= zlX&$@6U%ME%VbdpU=q;%W>ti|=V!%XPX0ULh??LgBwo_6b)9MRi|HON$tP6QJmIPc z7Y%;mpdxhMO>^t)l~s=-&7ErJygzkZHnu5Su)*%MAGJKxWMg^N(5x8R62JGzWyVLG zt#Q3>J`NlHO{h&KtDdZ%HVpyN z^@193UfD|3;QKknAF=GC3i z{tryB9|F1T-2YSh{|(Tq+A}0sQvbd`ci2v;Cq_0@*Tec-ju%pclq6)!Pw#L9*Kc%T zpCif~b!JV4SvY^P218<$H!2TY?xajw3vlY|nWSQU+*~Nlj^^&bAl3x6rSg6J76vdw zlPysZ;KGF3%z6F+D7mJek3Yl&zPt}-0o=-X=;K&9<(xfm_;~YH0To?M_A9XI3)O`~ z$V=t*7P+A$^&t^#Z6Nu@th6CRxNH&I#cX86OZ2TBpH_VXa3LDZfKbfX z4*(oqjbkaV09WMVXT85o7gb} zA6Z56nGCBFXvDZFXf0#w_=R{)eN%z-VU1P%wnx1;GM5hx@borCIYT~-v*-OXF;TAm z;;7u2uA((mANtAZ3LK@SPNbx|Yfur8a{#4O&UGfeZWq;{UPLM$7QgzdGgE&Gp^o!B zCIkwbcVgBUH!}@&^E6ALkPqYmeT?TJK}%jREV=!f5*49@L}E~;xdz?-_e5?euC8pn zUyC%@*6%Ns-LBn2xkiGAS-hc7H126}l<#c>;|~fUsx-50PvcxC@+9=#KUQeXrFv#T z?sIe%$H_ui6q4|^{ioV&`^R3*dTCDUjM(YCQzVH##;KIK+44}ZGoyMC21w)HJCD#n zZ86t5KIU?KV&9Sn9&eseU7vm8zPfElY(u0qQ{LzwD33;{jyOGo$1exq+TFF^dEeYf zqv_9FO8jn6n;x{-6`mIzF4P1cy=~sSjxYvzc9F&M4x0~wAdqtbW~yLIub#={1#$yX zu54EA>7)h7JJ7v}!WLTJcldLw9j7&e&2PKkNy;ewS-B&KZW80v)xytP)xe z7j>^3X^;Ld1j|vke%(fGa#vuD3RXqa$FIe3>Dm4ki%;>w!uCrK)WctWCOgc0VoTx| zk0d;KQ8{_8t4p4xXtqc!3;XUQ-uIE7UuHfm{O3Kxi-bfs6YZaLgf$mJw~VT&NXOgeS?%!dX`B`2gjtaN4DT|-A3kgrGgI zN|#jnYck6ZAkR!cSZ+;p3Bc8{_mSAVDk?f@d!;21=O(Q|=)pfEHbS&ZmKwct;zOC+nJz7;1R1t88o)8VPAt zhV4~_=;u(0oiD48ub_9(?x8g!aMualS5<3olJZU(^TjK|^3D^e$8e{^D>~6%*38^9 zk`={j^MA}8w^TN!1mY1L&8Uzt^5B1a1$JPU^+NpeWIDQrzr|#VnlnLM;G~J}I8+E{ z)CEt+D}{-_3%hiwW;EtY-`FN!22AVFD0u5qZ@lzC4lJ?KIH7V4jPh|0)B9(snoA+& z{lbT})8HE=Ll%Ag^Xx#;hD#lM1Hs+1mv3O^_N)RTpgPqr-F7@P=Swg07~>I~w9L3Z zHy16CpP`cuD=#gGjZ9M+q6KZ4Nxay&AD@1}k*jlIU~r*9x&*o$st}8ro05*pa?n@a zA$Uh`8@!P6wqO%P^R}F8^A;35(EIY<{tPeHbavv~Yo(lfe)U`yt!rcd5iPp85?{3P zVkEZa-JOj@A6L59b8iHv7}+PZ6}W}}8LqO!yX^1!UZV}~=)L;-X*Vuc=1S)bVl`5_ zn4)9pzLh&Nek|q9s)q2cFf|0HU?^^Zf^KGy9MgsU=WkXAE;_lireC_@Qe17nNaGYl$s0%$qUqv({x~k`G&4Krw)YkLe_JcD#G1F*) zeunpM%!SiAIsk2fz$=4(`Huinp?c$>6P+er1VP7Q>n$qGZ5&y9`khn9D0sUN@jnka zSx6-M;uHUqT(~@GHu=s5U-~{ihs+urmUV#7g{dUR&<`LY_U1OeSa-ec>0w;J?BIq; zo&+aYMR}4}SX0eS4ZQ8^G)xSVx#dh?uHcbDS7CjIW>5BqO!rt&Gk!~f(wHNHLMu+P zuIzg2(ni6rZ&+M)BUX;E2NYCGWgj<w ztu-KmEgQ$G-tAR2_FY~%Sf6G-nN#%+&97K8gj_58;odn;_v}x=2+MKljrUh-a5BYb z?dVk-l|oC5zVw!4zDA{Et^1mv(?=fEFMus+%b@_8iLQdMH~4SSP4d5u9vX5l=@HGj ze8D?GyA24kaDb}J48KhxUq^}4G=734(c~0N?f{8M77A>}Q+c$D9u8peQdn{UsQ#n? ziFCk0yCwnq^?(DGH*_fY+jn`&ydfZF3(9sw0$~Xlq0qVl@8o?ki$YgWWws7|jccYm z=T8WGCO_X+T__0%$DhH-ami6SKa|`gv}eH;7C03=LS)kbbxw6l@nBwgZAdi@qF~-l zvfJ#jAlzP%^0~vYZqFh>VFvKTqiVMBBKCl6p-U+b^duTDO?Q)xMbFt83;{l zu-Ka~w?2uly||1vm2I@R8qVw{h*0NR^^YL8Ikm~UIoBNgMUdfOeq# zprKxCS}G)0Hr^0#6*OzSN>Xr>pN=mD+BzVtglD5@{2!B7I6jrL_x-yEP?R0DcO8sQ zaFm=8XI{EtbChz~biNWErN=y%K4>s8v288)&ML>!3fgf#7<`sa&c{Pn=#xYr$*MF4 zNUh*mB@m#sgfL5rn0;!toJM1r)0t*VRB{d=pAtweB^#x>D7V+hFux$IT%BViiPU+A z-_sZZ-GvA9zd-!O0C!SLvyzWBul}@9@MWJ3V-+|~c3<}@S5-+Z2-Eq=nNd+$W@ zzW>$zAa7m$PwgVG@b)s`bRP1&6c|o_+10KOWDB6hO#51#f$TFhzjbvP+U!2O>y|To z@BTjx+dd?e6MfeK54yYd)a2K&mWMYI?kk)(6Ueq#YXD6HV75TqBI~_=7%KVi@|&jh zvT6_iS636-{Jj4c{=Ws4)BHMw?ppq@e$Fefb89WW7$<7_%9UQUxxVdb_{1Q8#idRc!)f@S|^X|D&T8Ht>z)nP=|AtCt>kFdM zvKhOJq=91*l8H7Z1vI{M!Z))dbp$l`8w9U7dIQ?aw#<^4mM08+;c&Fo`ntC;99C7Z zhYB-F>Q3!ZogmiKN@yVo(V9H`E7V&f!J~r>?Bp|s2F@7|f7tX|)3}p11uZ zpTog2$o3t?nMFk%V5mx2~$f z3aqoy8C1F!yM8$%7&Rk5IS595iA}Txp78d5$;T=txqVtgCPhEH$#gdH#Ne0oa~r9( zQ_x4A55RENmIrznn2vb;+)pJ>zUCHb-fe;piNiN>cG}1O&)F?^$VoA?j%O9wtNR-< z(O_azz5pAngf`F2=C2MwIRal957D1WWH&BaB9s7EJC&yUx{>xtVU0N?SUBq|$|oKsd9Z=ZW*m_A>UJ(a4<5Q7Gt< zVJdy64F45jTf~@7z!Pz7JDZb4L!@$De1=mDBV1wP+5H>5+l=We7o1;4ss0#M{gZ+& z?6e*~!=BIxp25Au9KaUH~`VH5h#ItX=lYL-0rx29zuYbBpUUm|*;1c6sQa}9@u)8yz zkNB`F#YCN8IBVNI|G-7WsBd)0O&-VbYBOVLICy;YHLTX#?#8%c^L(MD_Q~`|Kvm24 zZk{rp&uM>TwB3DXmk>LojP+l_Z|5-ozweCm($vZ?rG37Bxa$6>;DI#9;y>PmH53b) zPJvg6t=DI=ahYM>+S6cu3%#u#Uao;ycd|~f$0sG*{~YngI_lJH>Cw?f5uN^eVQDcD zmn}@@dxHMV5E6c{e4)WqBgU$=ep3{Xb)s|teO#meOWnZ`%1AZf>{>&0_xRo7XtmyA9U?Dhq8ZMF3Y9a|)=PhqB7XM0Z_`FGX)d z$-=_bJU28#Qp5S%6FVvi@*Skp$@zDq%G4%k;qzdK(aPGt_b=8!dg}Boy0oUKxX`M$ zSbjl*E5^^B#35r_P9ZUg9(R7<>yty5t?umw@b>vdHwB*tVBe-7*1nmqvI^t;DN6H? z$9jiqgJOndiG`QRo*Rk*cjPHi&gpBB@YBqQtQTz?7w4A8DHA>zzL zaI#lF9YF(QbfozIx<8YdFk7U{Tv^=)HC<%j&FhP?dkcr^U(@S*4uY}mm4Gx=eEu?U z{1hOkz5;!(2>;KbJ8ZCj0Y($UH7XNr(b;B32l@H`1N1(2O|1{H8eb7gN8c3Op?avh z{9)yr!I<^EtLMEr4;-ebx^UZ%Q~C#h?+fMqX&V;mGAt~S6+IpR`#x~B)qC@eU5HQn z)EP&-&#W0b4gt;1Ta@Jd)XC*E2!Ei8XnZd1qdSn@sw!JnTD>GtqvDXl(^`9s0BOKu z2Io{4Pqx8o_Cp^J^VPZ@vdxkv6``AndT)gPuo7q<(y6-G2DrC>FKzvJkUuSxbk1-Q z^KxKtMzuK`Q_Ma*k$qnc!S7|acmNqv4oW#{sg=A0A`bJH4VSWj$h(HL>kjIT{OR}9 zUUr80Xq`<`X5e3}$KPk|Jx>(;y{%s3WncPl`X@xFGp17ycHtXMdTg>BMJJHgHxe|v z_U!0eSE@wR$68u?kH<=Sg_!W)|BegVSt;6{zil7w5$EvO2m84tP8q2t0{dZ*Z?~lF zPV7m)`L5gHUtV#Xfk>Dig=WnF21#9(^V8slSzs$ECV}S=`d}>Mf-^tOJKesou{Xo@#lB&Bt(@pL1I)zZV-r-^hIuu$CVSYM|Dmye72^6FP?MUjg2=7xVz=gzxXik>E> z^B>k|YyxQ3>WZsoMXq9&-Cc8y#}+yDt%Q@v`h#^U8YLU*Zw%(YZq+rmdU_z8Zd}$s zVXvHmBoO;`R!GoOJ@ebe{G0Qm_f#Fhz||%`s2{(rrM}4%m0eIE3x4CeE{nlw(Wkev zEB9_8G}PTF_OHvvG7;5C*Iy~l__nsT^7|B<*Xn(r*c<|4l3XnVPmkyuAyXqd{ngKU zVp`ksPxreV-GW1>G72^?vM$=^U)=o?J3q1BPzent#m`KvZoQcau+gCYnBS-Fo#j?v zOJ-;I-17i(@~?n2-SawLUCOh*_Bmi392|{hVuWHo^!_R%Lxo3VM+s0J-?T4IxiY5x zOu{3g->c~cW}?fZ4?Q!{FRo^WPsU^Y@>M+7e9P~ycV#!9tpqIlnl_&+ON8O&cXQxX zn@hGNkDFiIgC_oTpL`;j8+T2Ae~gLhki_=5UKBf8S%i`jON!iI<<1+Db$4w1={~g~ zjm>a79lPznaSen1K$sMt;{tUyUMyq@S`6cI1Yy@I=}*K%^&?B;|0OYrK7o1_Uqm=KI(apBi^4+Z+wlOBtbpw3@^rEQn3454#Pj! zxB+J8&9j5eieP}^9HAp^)_3vT{F$gMK2+^X8hKW8@=%j2Ep%xPgv{?B9H zdLBE?eUamQX>0?bW~OE zF=OjY!0Y4Ver3-w4W6TPlQB3t$nw1A>k3V+N-Sb~2+DUL7P!d)78U0$sT@hBtP%PQ z%G7>&&-KWPS-7OIEEV`(wkcj`aJ8|q8aJ}NpxF`0{gd>~%d1Jx{nsLHBts);s+6Gm z#M3urdEJZ6+sLpUbr{$Fk6%7+8`Moc&(=5#WTdusQanay(apcr7x$sEtJw})s=EOm zzz=WVn~v_u;Wv+JK)o98ZCNpP+Z(H^k7M}F*g>mO5x){|*mI~_In8|(nqX-#c6ksS zPNnh#HYe&vPzSo`2Gw6Ni0%f_J6*Et`aFy$#{5V7H;g38fA{ee?#w(}b#O1q z?hgS^XN}qTBb8?}u|qIKd`Edvw?*oKt6(sCdrYwOWW~7fHBk_|J16Oa#Fx`GE6b}t z#8RTDt&eG_c?`uqGt)_ztnOON{y%TtNHtHbk6BUm$ozZLPoFzFihOUNc|cbv1!TmQ zy%C^Vo5;EQHN5NR@$=Qm!H{$KQ_>0$@`=+&X2A7+Ezo$;xBQrD+~Ru67a4w%wi}N5 zCg#`WnwMO6!oG$EjQ0hPgBJ^_DU-SQg!f75?Ch*331Z_Z4Qgj}WhJVEATYC^Q#Put z@mqMCueZC??sLSv#`9>u{jc61ZwpABpYX4m zwRbqUBuiblww>EfQ&+iGY$RcPg#?Y2Fz)2&3?j$eh99;lXe(KhGZW4*F4?wbpvh_`Z1{$VK;81`!i93xw_o9flKD*5xL#Q3-d0{p~Pc6?7R z3gQjpDO2UYuM6Qt89HEGmau<4sUGD&4hK8@y+xh8T$4|0Kb%g2f9^0h`>;-JX9-M} zJmda(y*y!kuTkM!W+x0cG4^7^X{K8I4&H zVYaj=t%<(q)JhXUWxl!{8>mT4eT;8|Jhl0(#Qax>!Oic9_=RFOa}%rj$f&o6-DX0I z)-KCL-rHva3l}sVk4kI-&vp~DjSXbM#n{bedD&-SpGw*rY0vf<%rrc|j^khc4IYbc zN-f~`A*_^WxA|N0XM$&2Cg$a~;+e6}8;?Wdl+6fn^p_%dRMDw8`IwI@ea8w0N z=>+|h9at&v(O?&iaC_5bgY2UD=cDxb+ac{HXMP)YZ<{h(``mr2+|NG#v4?vQS_s3` z(N37UuvyA#qc~9#xmHD4<<(WTP(mS;^%O25H zwj$x5z)+)Cc55eSz%BZ$W9zlZ*1$L40zD1-uU?l3@S%5k`rR2pGE3->SCe7=Fs0^} z1=r)t!~)rseCSor7AH-p0e*#hkQEuy7)$?xJC8tM(557! zG;ZpoGP*z6NQw%bcQ#cwJwFw@`=x6MryOpq^Hu6fjqsNDw$q=v`y9)Iv`vS0_C5Fd zs)W(S`w|eX-HG!Hp$*av6t?X_)j0PQu-@Jo-O6wDB&O-%0!QjAL-%^!`T8brlk)mi zA|a1owS`z1SB$jA9XIPgBnnQnf9LFtHti9+UJO!gyy_U-4q~?NP_A3s-}m}j8@@q1 zDk{1?V<^G20_#wDcw527%KJ0e(L3dT2A@pTUX4l%a37}sfT%Cl)A3F>TBXQ{s&s%0 zqT#i)VJQvUqu%SJ7Ili4Pb2}8)dldnNqQD!2)6#*(T-Wms049AaBzUw?GLT{c`bYv z|Lb^4f5y)$=1JA>r#{H<5fG8u>*h&fks#fA;WHMr;jws|!qauo9J33Ym7fsp zA-pr8-37)!QRP@pE_9kkS>aS9d5}O7xYwW-#?7BRS{~)L8))jFCdVgVZCm;(VV3lM zU5iCs6k4bpB%Tj8r)F*xNCsK-aalhLyheelcXG2*Xc-W@AKSKO)D>8DFK`JqK53#Z zdagU<=uczEC1_rLM_nwXVG|lR*9Ad7l3=D^kKz9Kq^;SG+^LR$@ZC9NTn48O@O3c0 z?E#RXT>Gtk%yad&Fs>A-(zU)sAVL)AqJ zdz=I9G8J&%*Lx~_^0a&n$`h)Wja#?{8YbzSIa5tjK^MXBB(P%+t6E%OuNZe`RtQUs zvqf2frT7eAPDT;ov+*AA71m(@k%Vu>>C5I|$Xydol0?KAEHgOCN9R6XiDK3OPY({e zBA!2NaVsctKYo97{`k&$Ms7ZON7v;`HbdoB`_T zikNRcu9ZhZ_?q9^GMU^CSJfm!{QQ68tc|WPpC|HN>}voNRs(2GR#kyJUd}g-Qv1vt|S-1tYdi*_H zQQFH+Ilas{CJR#fe#jHU2kipBJ2)tIo~MjH+`P>AjtV(iYQb!8^|ZL%ANZNoUmzuy z3^KOF1aPn}8i*@r2X88z9WP^&^v+5i+&5$i?7my}3| zA7xvLKOe7knHaC4+3DzLN91k6NMSUlFk+a10GOgZ*HFDCZW3*Ug~d%~nVNtFx;wzm zA-7aG(S!i8&&6bUTZTW{_UvL&8aV{{pxB7PxdzSMbpE_AZcPFZkQ-&+a>m@h;~Uin zkcw*Z4QdMLL6U8AW?RWS_@8VL2bK-&13Iar}N09DM$LHjfa)!-QU=| zs>;gkqtu)0$5N=_5&s%vNzs=tEaRe+GFx9DZ0;@~X|<2i(KsW_F>)92xh62MDm9qT zNpkif(D!YLEHgwb8CRB8rXjxVA(o6a1VGR5_pLqMPVldbtY6U|K9;X%Bld(46A$wR z;uRqy8)w!8qX`5hSv(eo|MLS3Ml zk@;NDI?``#Y>#@3n**elx$rmW8*MkBz6GnykRvJTp)@C6;gZ)HJWhWVs^7j)Iknvyts#HRGg{cUB@)~xb3>8}B0Np+8_iL}$*pkoDTF~#}-dF*23}0+k2UO7XNzod@35ZJ{)cP(Kprf;M ziaXZ<{*JcU!_WNJA;S#ppZ$B2L0WfRSxhkGfjJmihJfs77nv$QSlO6}5 z`Of{&ccG9;&)W~|fF?TJNli&%04xsF*l<8_rPK1&dQh7~D@?QnH_Ka65)BhTE%A;s zv%$7FdVKjc79jlN*8x^bB;*AOh8CNKYp^*e&T&Bg^|l|>m0g0%cHtG|Za)&yTkH5i zzyQ}3{LgP%4X56At=VA=k*srRJHr&Rq$I{-((UY=5~Kq#3| zF1l_bCv3z&1v((89XKssCd!#1CNE`p@(Nhyu2m18mMQzeBe6~-TKz94537XNVTijf z$CwF{sCiV$F^>JDTbtIITra*oc%d@TNe}KX(((SG+qw@nGwBxRD^NE7*Q%A6CAdaa z8PWGo19LYWMcnJQzlJeFuipnw7wn$F$o%;1>`yDO1ob3He|-`1Pnw=BCJL+yw6 zTraFTW%dg)ff8!%CslXhy?A8?ia+=8#{7#yVHGq(hHB;Iwlff85UZXtfXKls4QJw1PwFZ##An^JbsmPXF3G1?x3R(p-=Jml9%^T)3|@q>24{}fO2TDBW- zKZ6E;;LKr6eUXVRu<3nltD&eKF&*LFNUl-&#Bdg{;c6NxttsH?hTj-Qox z^8`=d5PGmSA5{Kz6Yr!@dT%wZ?KQLeorJ5DdG!h| z=y%FvB=d*-DASt$P?zr%uQv13S>P)RkMspBa|Sj3$BWm1T|A_(2TW}w$V0QbK*G(Y z?k8Z;1!gG|-7c5&m)`hT(nm_ZMDw984)SO2lT_VwZ7%_39K~6@8MpY~ubl1Ovz;&A zk~~l=g4z%B%9&Rf0hXDfaClDT-82#>LR>N9lb;x$m$( zQlqRUqRyQVIB+@^RIVHL%7YOG%b$sjOh9UNn&V+mek35}`~01ZhkdZ*H;5Tg)n8#m?}(y@-U}&ci+reo+U@QY`!I=c zsrJD|!hb2DPubj}mT)yQ`0)=5`n)}aVj(pWdA7w>whJxzZeW@+Y?2ZL&4S<OMH^pjXOo0tV~;6-|6&W+eN$kZ;Io_pyON14IWiS7p0j0TC(!yXvh z%eAOkTap0z)Gn(-8HOs~jEx<2wnvn*KHz025Iy zC<)m04p1nsLGBLm*Bhsjaym^`H~~UeQ;}Y`}Bl7_s54T|Es$z@)YD^JBXYff;ht|9 zwIA-|>1`z}8Itwfg2Z$NE3V4}Ap`SLu|a=1%H~W-EIYBvyoA^~@ujfe+xKShtb^1C zz=qX+hD>WJWwyPIp~l<|y7s;7S96*+nkNGGy?^ux0VNq+RI&@Uca^52)fP0B@$C3l zWhZM+m=?FlkJwXIQEHB{C~+-=sdgy<69l7rfu z4DrXmlOV)KRY;0H#YeigoDaL;-|B;v-#Z*=`ZLm+luEV#jwRl^um&RrWi=Jm9WMwQ zP>e*-P9ETP129^X*8lTpl4 z2NZ$G@cyVX|%* z+nihNcU<2Py{DDFXY-G1+)=JZ9$G>{en>H?R^+!hc>J-XMh zNB@Y1RF`dcL=P7WPmuoG7AA8uR5*t*U*k0yY(WTs#xT#dRb~VBORzY)38r@r=0G=f z8U2vYF=YGszhMJHoA|sCb|~$WQ<3ykt7K+3mg$~hlLyz=%72AmV*c|IH-;r7ha}Ir z)as84vk@Bwp;E*7+6uE!v0C7fRl}S{%N|Hn@#iL9WDTg(YZ2IznQTqLS+}@+7 z7F1iSL6FBRb2;)d`?z>&*UJ3h@PdzFZNzDhlPXd?VWr!C`#0ZDJ~7%}gJx|erxHzt zy*p{gsX^_B{yCgap-(aM39{H<>MJ1D4h}BF6Gmsdu<-u!oCHDkAH7tHKWzGOo-FgP z^pCBRv76QubPiH->Q&T=|D~bcQ0{e24EgMC{5!hSph0CTHyI~i4gAlAba@q^U4^VL zzet^T1aD>XVnah1NvLLb6=OSU`nruirpb)n&@0m{VR2LAP9RNn86-NbDt*}Tm%PK+ zETyb8rg!zl6%KA1>zck3Oopj@eWxth%d{_^9|1zB^G7)l6 z6m6o$q6Gt+zp2{-4X{zLkKh><$=^q2%-M?Bs-c95|NO}z^emmCW*P4QpO>~`)l#zv zdiI!PdNUvh?gNqlB z=0Bpn^zu18H7(VD^>jZfK z61V&vr_?wieW|pXbO|U@n97bU%aUEFPX3rK3w(6<6dJW*^E*2A#k-T=2`vj;1&m_k zO0PjnyEw0}A3dk%q!#NoH@6Z#X=9|p^~XC(jqj-`=&<}qd;HjBlr_48xa5aQ!c4T> zV!dF-5;Jz`-{%;<|5D$GcICat^S@;B?;Pis9v#6};KtTXA$`FsQ&AjTn*~9CpL?BJ zTBEnAO3;=u_6>ylRf62~ z6@iS$XH*seiI_H6`;X9(pdv9rN{Yl^;PqWnDxmS(2y8MZbh}hVHoyuCJs#pF7qpG! zN!33Kq$i5hCZ4GKDPnu(|)mLuaGYu1r z1*CAU2<|UJU~8#?c(g{vZXhHTQzBQ%jdrj#6rZg=`Sfb@6Y0;+4w$*D@z-9f@xL{LPopdda|VxzFCYS9sLz^8uI-o z@fK*n($-_%HyZd3|H6<2aSVv?$j??K)b4CjsVC<+UV6+FHWq08kNgDUv>@*wJu)^y zc>H^0k)-!QJeC&Aqb z4j~ZSo!~66EZ@HG|9t11dvDcNSJ&_9?&<02XQsPqcDIV*%0Dw%KA(ex|260F@YMjy zi&B^bNk5%>lWsgq>Q#>yN5h|sxR*X0VdmoU>GWr2TE1$*7|ZK*?&G9L#0N?*sHM9* zZEALQEz6NQxRSS)j>LfMmQPq2a1_N=e!oPbGc0px?vynh9fS5+UdDa^AoDfLgab!R z_=1EQgByQLA!_lSOohUT1JxTf9yl>Ti`wt?ZZ`LF23fxv?z_m5gpep7LOfmpA#!22 z8m6}^hqeI?Y+?%xshy$HtXN{F3Z?wdo%|N>X8kAk-wVmCX&{A_%A;(9xSR<|`t<1T zHa1(@QPKS9UXSE=^zP}~{lqfI8Zr0h@s=?orp40Q;;EQX^MEMa=)28$RZ($RRIde= zoM`8<7}c*{=IlvZXtX0}FhQvQ@C7YNLiqGwK5CAHor!neGWH~%vyVXa+4eVYyoee3 zBB9J1XSh?)+5ewFoL%H0z+bU`8zrM5sH8YgvJkxAkCg|s=NebV#Vp3Q0C8{e#H{AV zS%1l{9+diUnYd2_V(O|t&RW-17eYrL$YhIxe-(a#mtUUx1`cVv)+vW`ZD_-#Cn`|lKq-~3^L4m*sMmP(s{@8p{|PIHa(xV7H_1#_-$>4X9w8FKcW zi-)s+zNh_tA|b8;DS~lmh))xrokz+3w0 z%&b;cj9%R!rSfP}`hR>E1x{}J zMi8hxk_}okf*d2R*%g452Np|Z;>OY6Z7W=1(JjU+bV5-c@wNO2g9Xpf zcV25iCN4ra5U$-_9-4YNQ)bAN8Pr^j-xLlk3XqlRa}`Q198icLXoy8+)W1wwqriYm zOo~;6jt5UhPSEyRI#}enY412VYL|uq<8N~0ug9Mhhhnbi6rn38wZ^Noj`9LmTJE&4 zWBO;jqCDZNfYoo&h~`rKxgmgDgd0J4GBLsB`CnbXCND6NcR?kIQ28IFt{iXcFL zrM%r5<9th=(dOT143%RTane*PM~B4!t5408K!$zlD(i>V9`H zM)i-Vzu&tHLM9WGdsvbR3qMa^ZjVvP-)On_6anS`wRyikMy0o2JVevcR&5Ta9YK(P zYchSC9&Sf`j=Q_|iwsGAID(*;sx5s*b~zfBY))Y@c@lxJ1g~e%;%nb4tCopN68Ep( zgx8iyyeD+!lj76Xi6G#fywWF>E(ec(vLkQNLm`Ky?Xn+@H>>|%94ceL47){9nih1T zc<94af%Vv20?g5_pl921i~b_=43UG3miU5hp?N}69c^D?2y85+==P!17LxY>Q_st^ zA|MQ@T2xS&Hd`znx>GVQ`OZ-ea~#zGplL~zd5I*3b^b~6{l3bI?Rod$zg+<)d; zWytjK3$k3TTPvt%xa6w~LTBgxkL~uq$?HPIWOCWBjT>5$70z4aSeB--;Bm(qlw9wZ zn8fzi%hv^{+*N6&B%+XUy~D@x-}|-uhut`u-Bg#_7A|UGmEgnkv+caWGI)m0lW>VU z`{~{u9Av|40U9{0ml!I9*`lz--cVq&R-m&e`_FjSG0bp@*AFMeb+O>bSZl}6m^knU z=0>7!C{Uy0G0+Y1^B`)OU;s2tGThQLPi z`allN7Cl1Z8ch#^vu^QzQC>dod)_6kXiGl$uo4me=jf#Q8XWS6s&CxzY!J(aHx8HY zFzaH0Yb!zaJSe%aR%un-BB}V*?Yg@a42L*Edt+2ftV<~^fX}NUg3tSx3^KvTiS!b4 zKElX1Wn+QGXYRA~0~hYy>-KG>WonW|cpuHlBeM*VjnH+977`J?NIfQBAO;it2{B37 z(jG25m^loC&lf`oJUB8>D!i2lvA&H_GVgapiA>PHh4LuF6rTp$gd4NNWE7#iPlK2` znapmP3Hn%2o+1oB1Mr|&^R+bhfAp9EeVqT%bNk8>EfM)IXdk~Vh)DG>+teIHgNK~! z4ZSoW3$ogB5tOk*d|fOjt|J8>-m7&Lgn4KFgCu4b(JJ}Mq#DfteEF|5QP$WZCQ+cG zOq};{9B6#7jY7sBM`~Q!(KdCZ50J$te2qtdlzRucxs>b~m#*@gTkA#y{FEJEF~zf4 zx|vC*FBl2c{a04pmU-;I(3Y9D&@jSq15Ge)Z1HsIl#hXnjhM`btaeMQ z8TRea-q`)6ZI|4(b(~wf)Lc0teVo6QKN7=uk#~5P>i<)DM)o!%=%ah8=;6n&15y}U z>@o+EVXp;BI>a>`dHL3c0!mjikrXIUeApWV+1)BdN8!NW+guY4oCp?(cT0dx@Ah!i zG-Gdt;Q+K$687Pi8B-ER`!RveW6H#%7b9Xa5@^?}5Hprd3^M?2RIYy;;{zKw*GYa^ z$qN)E??~@;{UM6!^P1iOt>^W2RTcLU=I8KuSQ23R4ssWt&}P^Eyo)fYqV??PxmCQG zDHj*NyVGSGgV^VyPNNiuPcDFpY_7+db!83?q|CrnaBpmICfdf`B6X^0Z!(o#)zR`W z_+on0tV?b2-bCf*MeAjfLD7%Hd%VQ~Kd+J^`Uc)}*wd4m|gL zaD23HGEk9-{W$LfLKu@8)6srSAUMW&8sG4RvId<`?KL8N zlIuMrWt=8!YJYNM)){N+e@NVZ19O(-&ek;7GQGRb=hojnpihg!PF2GrZpPi)Cqjdk zV$H@!=sedM3Ky7a9Xrz01kRZ+6^_1%pV8j`6qf5Z7LiYaj<=})nReWh;IaudpEa)&pEm$J&4yropp`S_ z&oNwU!sj+I;&7VKp(r;e+DCMkFGPJXG zhTbmRHW7yJ)b*|oef@}dWl^Da)|Q+R*_A$91|(JL<=0#@w8)6GIt+=aZ&0{(lubHM z?iX}dO2lcwTeo8nKIcXHr()&n5T)3iltYo@siwin$MzcTR{rdsaFt%&u7mUkbLzcu z@D}@!gq0;$Pa|AUnuz<}`S-CormeW|E`lus9wjYFit3ZbA3U3;E)^Rmyilfwpad7f z=-Z%Bc;(hSBZ;TVhB~gMH4Fwp%eOOvV9T#Ax7)}4V0VeGHYUoK^auHHvXODkqy@Z1 zmko80VY18Gx(wh+yAP1Cm#?)W?7PK05{J8>@uqpso07mPxpgu~Ul;Z+cvy~FE<39@ za-`>Z`Iw>27uln~jy1&s>%Z;aqT3C39Skcc(!S-g%Az*a( zVaX81%()nTbGU6959MlFJGtTY^>?Dp@Bs8289rrRFG{oQC&>#WD=R61hAxlWnHa2Z zWq?!W`Zk&@R`h$)M9XH}Z5qG#wzqW6PBKLTZ>VC}+S+Cp-5s_-P1&=;k-}_ZICTgc zi(mBl8$5o0mSY_TY~uWxI{O@+w*z`}7fBWK<=14HCV!O9tPW%$q%k{-Qk50@ktUVn zLPVh6&TO}iu~H`vVYzV`C7 zutKlM#W2m~c+@8WD*5XxRvx^=0Q{8jD^sNBAxL_F4OZtga)!Wb0Tp#Ph>iu;cO<9* zn_N7!bs#tU<0b`h#o0N)W*TZs-SJ_fS8y`M=wEziyc;dcX*nT1c-44dX_3;-4BAFW z5)k^$s=FYp3_Q9sr7ndpL&N)eyK+hO9BHC{rS*HfAMyz)p9M-5m#nOsjX&#;f66YqAMM(!h7}Sx@@jn zxMG5zb(GCl?@$Evto)}bKGMU}u{8zt9aHs8!YMIkwYL5I+Fdd+locKI6vqkk1gf?) zh#IQOLF2(X+>T|z)>QDU{iQZ1gtIA#ewgPeSCfNXqT>Fd?_Hej!~OA7JZq-mMUV1{ zQe%uN2k0aRIfe{esLCF>5A(s1Ff^ODAqW z(w{Go1~#?k6kl{3UY=xC4ttaA$43O1TZBPBy{Td%RUBS-cfTIexQ}kn8N^i~V?y;t zEmK{VP%Ub9c`l~T9SKR&xF|)tb*W;m+O@P0F3hJG*7Dz`nRtg=_=;xRYu7`#Ssm zg{gCI*2a_CQ+QstZLL(bf{eYn8(&F^J8Y-#v6;`Vwc>i~--O%fJEM^Do?RMqprvcF za_@iF{c#BI9q#O#DA0tu=4cjK=6n30V?N<>-UuREN^xjTV7fKAT#{{d^>4993kWG{ zB56?Z*L}$&5(Wf8g{wGsK(1UT3*$?ULCF2MK%wnqp?-E_tf~?<<^;pX^RSY<^_MU? zA5H2|ROi-NZ;}xW@>s|ZH5PqpyUy8R{RMJZrF2Uk!EUmy%KyR|8Ir2~mC|z6Dx)xZ zO@;|-G=(J5Qnpks-+o;wwu5Ss$%2~H(*ZS^09_xUo%3o&K7;XZ(VWacP*EV$C1 zW7C%qNv%L<%xSU@;EiaZq_y~=V0z`}P}&Ty7s+(>v2U+nZ0%`uO>Rtf=@i2&FFC5$GvagOFOQ>Im_VLcI>QW2 zv-!qu0gyz}(DB}umXi5k0Ln%D{CFfVB)w*g*R()WirD9jlhTG3CdvfyR3cYSKHl;2 zkog!yU0`2aFH)!uu599nE?5{o-qXadlp2sk*^&R}?}cQVgIu>@MpL=%rTQsQP~!eE z$~C69SHPvoJgUSP6vl@lrJt_(qdqhG)v6Dq$}D`$=Ix29UXlLVk)ABfuMOj#zh*^$ zj=|5=tak@s(pXwh6>8#ya7u2Y=cqDsYfg=3le(47kLQa`#COB#9`EI)U|!0h=u~@M z4+Csmq(WiLd?lkuOqe!XVL`AK@d+lF{dkY-6yXtwFOsk@zMdsyF<2M|Gtt|dJG?2) z^S(gye)S&l^bCQrP_swCUkFIkPv$P>zkD@*M&==ysrh3PXZt(a84=OzR|}=Fe>B&> zL|wV!C*oD6R|3R`LzqZlejvARLzIwEtGtGvNzRy%-SosBClXnpnBg} znneqpHJM)<=EoJG)yBhbvZC_xF{VmqIsQXlS7T+{e?xp5W;-TIJLsa*ImBgRMVlu} z{{)A-LtaU=5Ha<7AAQ%X~8I6JmdqqNoY?u$N@{`=T=M+k&L4aF=G^*}gmj zFCsQ=r)yxWNjd26b9_q)C4*o1{N*0X2F@nm|TBdhg*( zuFDvmPWOA%&1Kn-Gu1;+m+q9JpM>E$<+sF#-@j#&;8WNXY&i^BUNT;Tmx?2o!xuH{ z2A)W#zj$9>^GTuv?Rfw5Lk1bxenI*08m&Di$+}0u;dG);q|JKRfAQeY50El=-|KBs z3IQI-2EMAk7-d`5EbKN?4m~5K2NM=UFyY{Bzam@Met*$i_5_5>uT04*y!jPOb++#0 zDJj#YwIUnGe2dJ>f{I#vzW$C-?M(>cEYRuPAthjKhPL{TUe4K;@1HOVoJ4($8ax9p zF{JKhy&bYL86r6;LNP*9qv#I9>dU~;@1IV~3#w-{kW_vJPn1!WbyIjOgraApRT|mf zO28X6s-E5B)N|NZ1&6mfk4mOmX(bjB=G?CnK#wI=SEfc{Wgfp<$3qnw4rykOz?E5< zz*5QYT9=`{c7eXy^y192gJRGX+se&KjvbIn6H>;uK^!3omfAUu0Ug|-tsft-Fw zqhA)4#fB~Fcr5+s#jFmFDOeqTMzUZ1@bg??59fn2sS3@Y(ZKDpRVAzha4@*gLk&Xx zdOs#(%~GQ#y3QOg?f>&5X$W4!O@X>^2ylW8rbFq!6AJY`;@TWZs zw69!$5D>QaR~z%LN&9M#r{)lpYfaaI|AAwpvTNtHF?izTe-iM&$R#kDh2uGeB#JQ7 zX)6+HU+T|OvVxYERQwOw`kk~@`DgftrXgO-9PUAGeo6y@JYYkKOdQ^emtxMM=&fB2 zRKeX}Bf}LMN|ZbeRYQ~tC7*1$Vvw-f8gK70ne)PH z#UY>%nRfmh*4>I45Qv*p91Fg~CsU%9WJD&`A}|cRvXHetEk>^OLh=5COHeHoShg~jviR1debPzC0ZPI z>HSoWh9vnx!r^TnLQb?=QeAjGD)1LuaV~tmxvO25`RL$(izuanMq7d7FcBLu*l|$HY*@Y!zkw%^=uCf`^T4 z=qQ%-g+x)Cq(vsj_PrJT#z~dAGlDu+yQ0!ita)FJd|ib93{_e18xy2QsXcPJ$s9UJ z83M-vXp&TR=(2Dqj=6!C>UFWMVP%dmGQwZz4#H#ByF4UYL1)iKg&*kE*ZN}+gYfkz z(c)((!Krmz;R!K=__atFywd<642l1Q2MWuZhK7d5 zU*B|aU0}Cw#*<(%y_J`kll!}^5e`kqU%r(?)tmM$UVkHlAou6OwSLikrisQZl_5I75c6}qVS74_cZckRpN zpMlgg_GT*mDe)^~KL$;!2P^9NtL9ucms%40?TlHGJ-K_U}0VCBPP3^(0h zF~`P-%7;nQ>$&jZu>h#b{XY6@XyzCi0W_^CWV-kbMgN<+b^CrnqzDDY)JrY@ulr0^wp%RyzP&aSeW#tx&if zgTsH3*XJaQCtxx@QL+*KBj_`&Cx5ZgOYSewzoeb)^3|Px>+`nB21~58kGTO5(4mO? z30KA9jb@_|mvnOr_aM5R$LXx^w=yX~tf#^dGnb=b#yxe7WtCJ!T7jCO9)iU$eWYTQ)m+A_O ztp<|anI3WppB8y`%R%Gax&8A(dlrfs%dFT-TBP4}suR2xW=;)HzgTpKR+Q$hQ6o2S z;3nA$a=}K!Sv7xYC!J5vrH)1y|6_IuM<`Wxf```$RlS>hoE- zK(7T(U#MkeuCmAM<=+rIL=XA)4E03kF>DdMXe{JxsWf)>+GaBfHu^C=GjYJ0`@G*mt)TATerb@tZB6~$sG;S^^gzBH*}qQolr^*?-f5XX zR4oS04KOEgQ`&K@9f>>}B1A!dLV5oYg{auo-T-P7Sb(LExnGR0wcOq_Y)d^Y!=ol( z8FM+q_#+BoNLCTX7ZAWE?&QjfbfWkmUmkQR+{EwDpore*tq(H z2PBkn;~EC6FE1f^vEDrdY&sV~hPb#Y!1iK+5hm)@y>@kSj{->Et2&UD=t9-19L3ED z%OK)#cQ0NDrBfwbkv_1~l)=vfb9hetWqt*5&r$WomNFS(05Ddh1N=^*^3Vpaed0PF zZ{YU^kvHS%CWm#(3$=K0bM?LJB{d3!Z;s-d?A1gf2Li zyX6!J- zop3|;0^7!=c`BgyxEaU+h)UJH%?k>tBhoOL^%bs!EBIs`;JZU9Ie0{O*cdOgKeLe$ zzATz|l8|Vd5^HLoG(70ot0$XfJUM~f3H_V2VWCT^=aQzD%y>G@c!5i@?oZMO8a-hv zdA3}>#xizYLN1qIc zp?Oa?PF~c%-a7BfYiQ)|J8PMlu~${IPPr5(b6lQQSF?UMEsgM25Oy5!sC{;LyPXuh z7`b1@bx#LZS%vS{(6W8k*ef;YW7Cs3#g&W;4Gyh7@H&l} zW=`_DqI&*tDl`5_xr!kA^2G{GZuW$G`73ntWacXr`nS6G{8eKSWWk)j{A$v~*Ls%Q zUAVIPUt}F99V^U65X>qk<`Q*-R({Qt*QaRMrRFztZGOO=$>3a~(~ID+)asMc{k>Lw z#3x1YjKD_+1#x5Pm*Prm^q;+RthJ7SoQ_|3q)}&I$8-IU-X)ZO$|AqAIqUlQ39WU< z(*oH;`>>V~xMUMZ=@y>7lhu7jn3{$pm6-mytJ~%8w>dA*YNnx8(R_@ne_gv%e0dv;Nn zams2>hdH6|NPvs?`p{Gm`ds4oknBdc?2QOH`}S0cEi&f!nH*IZ5iZvVXC1@3q}t2TY_ueB6ng+nqFH-~HwwWvjROao@p z-=v`#e^_k1WbmE-8!0@3E*yae{diQYQ8(<1`Xzw~QZ>G;VkU>)8ZO?vKp;@+b1aDw zf*6RV^se)ETLk2(xu{ zk(#CTl6I^8ZB#!+>`DcC(QS1GE1;}_J>5Ps$S3B12~ltJNPMC*2`_ThNk zpTA!s@SGFWU;3PG?owbOod`wPV90eL+Q4tXJx$Sun+{MUG-kmEuG`^lb8Q1{0)yag z^JS1aBU^(;XvkvcKB3@W5QH%QyhV28S9WW9`GHx5V@a@Whi4l5;Ck3+1Xnsm>+X4a z))Oaz{uxa;;z+iO)ATw5UOrz2L-mBpFLZ?9ExrrNT$Eo!J5n@UtWbm(hOgM?XkvUQ zF44bu<=D-*obop^8%?8#T_cvPppXAr|E-TtTEA+_xQs6rFxAq=upsXM!M!dIH266BRn^v(3>09n!hC=}$y%=oOiqKOerVbu@AQiD(#?on5 z^ZW5Nr=avds(_csO`a{+cn-a9XQrRq)0A+jjQpS#^v&8)KuuDxQX`rC3}CY{zaWed zHp6p*BBCK?Iy4eRVz17~z$P=K$c_1hPTUve3@Gae?qWqJuyu5YGzRzii}|2 zZ-@87wT=C?-bF=29Lmp}IHSmOC;FZ2+idKYwzA4C_-^uFq<@~WkBTY+u&_}{ z+^_M6OMx-W*K=>K3uoSSM#ivPjgfJr{9z3GfwZ6Y_w4zX>3|OKx*Q*u*Z;G~7Ngln zpR9e2i>Qv{9t|v;Wpe$*5jhp(h07jkGQI>HEoH|&+PmQ-+ta4i81L{^6Z+4RK8_l; zQa(|e+tU_%@aKgApC%yV5sIWnv;Lji3Sd#hrObt7>rq-EK=rP zNgtV&vQ(+n-oMPGV19aCJtO$?l2UEk&@Ni+^ds1A)&9T9L+|?icn-DfpWxt68VsXD zh-c$1H}GFI@fB{R9_^{i4uR^&#yV2&o7kmKK7rAnvnDh_>;tu zNp!#)$0wMtsrK(ILB>#5jBfz5~<;6_qz?-)9=VU++kycc8_O4ZXr5JG_BWu3pC+J>y6^J!6p ztiK`B#{hZjToO;AGms-i^T*!l6_}^Q&NO>NZr1x4wp>}C4s?w*j1zgbSq1>sT*ZIt zD8b(6Rg!+w%B2>Z0BV{s*e_ltr~cfl)0L`SFTZ3sS{*Zl5sxS5kkM} zJL6;7F8k$bfF&I+w?u7*&%O5R6PM)Dk2?{k9(={`hO`py=#kIR2bK}Z2 z7mU-s3B8Jtcf$Lco_Gr`KSc%rjdDACN@gEQ(}E*`9{}>{hS~;yq~xUcqtsN5KrQ44Kko-?RdejBFAQp@Hpj!DhXXN-h<2naU+ z4IkjfC>C3-+@*O>&J=p&8%nLcS%_BO%!A>P^fP zO1d6RzLL$fE?1;FQoIPIGrykEIUno!GaZH$$mR@UE$)0Fa8!0H8@SI!=y)!;v4_`NmeQ5> z)#7_fgA}m;BeD9EF?CJm&5s$PSxGRS23^}b%V%}?^;WLOJ)D}x;7s`SGVVE51eJd! zOyA^i*aIqP1FBV_E?P@d$QjR7p%-Nq$TS*Db6!vywAntkV9UF^-bs`=Zb^r^*mIT$c&= zc0A3|kCwc5=i%5;Ge|pbxw!4R%C&8!$Mpt7KYz4>s?6~uRNzU*;v7-mZZ*RtBB~~u z2BhrdFz=PcXv42lYH@+qCaj?TuWV@_IU|lr^rV0P%4etbJydp=7Ji76s^AI7IGVoP zRt??wbl)xlYL}7nCxUr%_A@ixxawD9+?VUbff;M<|3K@RT)xf{nN_+a$JH|CfVKKsw3ELTJpzB?AA*o$elOx^%Exg4kI-)Laosb-P@h@i07i6FA{KqyX6f)n%23y*QnT};?h5#dKE6s zou4(9$=^d20L?QW7!E*xKX@wj^e6PO^^pmPK_53Js_j2@Bmx@jMpXJ|c$)u)2rBZZ8@_I^$K|`^(off! zWa@KWu~4alV`_k?j&tT5RqQ`r{x~WwRw^zL1zhdXnKqsK@`>>k&BZ<*eiLnVnm*6- z>^xYjhq3>Az5gEq2H7HM$bS_MPxv$9d=*XK*il6b_}cK+>O;Vz6Z8@PEX~wg-X!o0 zlF2zsjhPuO5y=8wSqFCSHwbdHcLaTp%elj;{(bUde%9^lNBH|s!wUze8sf4tN(5(L z^BY-g4KH6U9|Bsvm=kRUmN4nQQ1}Is>bG=Tv&Pcsk$5%7f;=U@>gYLK0pEGo^+1AU zLQY0!e}fK#X9CZI?;#Cp?rmqef$OJ_l7M-fOCN?&uXhlHA)<{<|E6JRI!3JJ~U`pQ;!0)Iwu zx{bSygd&V@e}v1$!Nv=@xyy_#PkZ!8_&fnKcAHO1$DRcyN z#r>#W9P5{Sy*uR(*+YdFpHO_)lMz@EU?%23QHMBiTaLjA+!CRVx-Q#}!M*HODk{h@ zNR8`}!n=5R;MY@;ne%u>1AqOJFvl+a(`PG8^Hbw?75NDsIGBgGe~cZ_UrJzJSo+d? z952My-eGUv;!#xs1Po3)|Hh1E%>CoW-VcK#3{#V4FuRf%r|xOGEv{NZ`{kX~1$ zn^%x<`0Zd9zKM|Rwar`AOX1Q&^NNp|GfXTH&Y;{59Wwn$0v|^ zK+n*&BHBUyw9;l0K?8!D!QpS*>Av8k)M?W~CA|dvWBhL`XFKIyDskq8EjD!-T^Kkg zlgGEQ-|JnsIn7b8i-W2na-3os-lZyo1hRdRBbmv=bAO2fMty6k*Lq-O%OIudpc7z$ z^EC3Y_6BN+htE|3N9`JZ+vec)f){^NH~mif(Pm4z|Lq9Amj;~sj>$(bp~$F7;bCt` zJut5JL&UaepQ37wifhu#NbJy+^Qx;MEjpVsw~{t8Jo!`s66zP3!*dv#u<|q~*8M$F zK4jS9rMRivgIzl&j+N!%(fGH2e!X@vl_rCTL~EC|vS4%ar{}^A*R!QLG%O)oZ_PP8 zKrP~~kVh?IKFRmm!6BlIO7G_1e{qf9X_p{{!e=8gb#Y!6uQf(n>v4?3BgP>;{}d<$ zO{AzHvjO6dKQCHZVNAHj*txJ|aSI`FkoOBUQs$Z9b2?1bPd6c(NxsX`d&`m*ek`s# zlyV@~L66a040-tblvwnX_?h5q-{t!`rt-g|eGSv1wG;v7*<=nR-1KlJC0i$bZApl! zluSa)V{-mlTFP!gAJdEqTkQXoXzU!?(;?^cACbN6EDgY0Dh`Bu4008YU-r9UAlov9 zf(4k1Emr?lFhbb@BGBWaGw>ZEPgyaC>LfNZ3S{>gF0B?~SYHnTe~Z$D=#1XXEY6Z8 zseKj-H6WA8`(S&9Ax>WTp^)K5>u+B-09CApFhj9D3UNPm3@xswTSR*UeA(gZr9BK- zrI&xY-jViKD_YyY9Ou$5;+k}D+15YY+f;MwLUOy(f_RQ~hueb@ZDS%aFfDu&WFr{b zbB5}t{fQXoVp5feWT&XbP*?9M*pc3m`y>p^&)4bqU2CZbzIqL+1s2qy7UK{en$+^? zmG;WOtLA{`ar8=~>C*Q>%5Po;64C-dp-av=;AA#+NQK2Y<@Ajf=KIwK3928Mo|ZeO z1O3XWCJ;|v9n}o!N?b03vGGeS|GT3tr5lw1qoJ=jT|RKXN^`ajK$wgLxa`f41vyTP zRj#EviYng@@MyPKV{57vgOa zF~+7$gQ#2&QuY<+E9r4LTHOJSzxP&bzh}WTEqwWKdds(`=Eeoe-%wIj@+r> z+DYF-;tvNr3eQ?Z1M?UVuSCF59ZrF>dK`R{V5^whIhA3YrK+im_tqP0MUL3Q_FrKX z9+o80Vz2=%cm8}?6Tu5~uSFD=j9wts{JN`;8uGDuj(JKNrU06uqmVn%g>mk>s`Rl8 zn~h?PIv6X-#a>vgVvQN8#5haHu;CXF$7PT}h1>e5J?uH07B3OTRLdbeqHz`rZZ6&y zu9k=|jrqW!TErO~MolDz3Ab8AAi-ybzK%SY+((i2q~aGl*bs_zDItD{tzwr0%GYT9 zJ2)`IY6FSNjnMHroHM@YF2${fQm=v^V%o-3RdqOv`lwLfuQ-#ht!7lD@C3bLZpS>w zDB%yx9<}M&n5f4wAh+7|+LUiZmw=6gd4oX#>nbw2DN@G~ZT{#y-o?ZcJ1jwKFpV8n zFl2#Ir<9u;>KJ@ld&rVKu#{iFq(+Q8Q|`)r?G^Tj9>^jm{9x?!#vlT{gbpF zN<{g$l=l6~Se)@Fa2PW&b#DhW6K@AKMP9DGW2>pXyUCv`qYP#;7W(XZo0%k#lodd^ zSHE<+CJjWh@Fg2`2`|$nt9Ra0?#ck6ZtZ}QaKo-s(BE!4(Cve$nAnG@TY1$37he~n{ ztBlKc5icB?@$)X$e!`ei;_u6UuV`OhrsxQHjO>&nol|kjZ^mq!O+NJpw-uom6g8(g zPK$xyov;#;Ln)EuBp_wD^_E_IwhS$~&1q8^K<$n^;8C^S6u@#?88tCVHj%fh{Qr&R z{|+w%fX!f!YWJxO^?kS*#w|MM2boIl8p7_OGmKsNXjT4r6>MpvWz$;9JU!|u%S6is zqhE+K=NugRZ&`{$w9;60<{Nl6G$t>2rsJX>pDGIPju{yZlP&jZM|Re2wCfJj;@Y2# z*0-Dsi5!5Qmg9}^4zGHPa~6a&6TJ;|7Ob4^LtW14iw{`0!w0wc*FG-;+}Mgzi3_DI zV2L+SMHwB0;8t5yZ7UqRSkY$)guLIBU`v#A`kmK?(1$};OW&xPC3xHM2aCp-h!<5v zQ~A<*ZQHCI00xzTC{iB$h5aO|Qf<8TArq~|fXFC`5Ec=T!wMKv`~%^vQ1Fc=XG|R> zZ^M%La(!#GiVY#kVLj75xv4bONXd}I)@K*lx6~vyF-~YTzSYukxjaK_K`UPuo0zj1 z#w~OQL_jfTrurydmXsvC%+6skzb(0hjE@Nq?>D2)l@fhuc-bur8oBQM9>38DaN%1l z-h>vghezPRRULctSN*-W`ud4|BL?ipvdV{>D|kmW_Qva99Fl1gr{fDauha$@H(x}Dx$vg z4{Ssco;tUObeTx(T?qAD9M%OQyS8?KMC^(^f{qI*pIwXPULU!tBroYk3K9Xop|2z3 z^x8ypnr|c{t0K0zh6P1>sF{R}in25y%TW zO$Z)>nPo|}nsREH&9wbfz0hPFaYg~yku^jKuI`W)zeX>N{%Lf|*IPoIdE7%ch_0YU zM`W=)b~5k_dN3<-XOiVyIDd3Ls84jl+679sZ9meeVnYxd*q%i9A_A}m@38hbhQAel zb@rShZ9YZkK9?S5MQ9rwD~u?3*z}1>Tj!5nFDtwiJV0ldk(Hxf;~cJI=J@3P1>uTZ z&Y{=L_vS~yh#RBSN2JOa;Dku6uifC7%ttpenX1YnQf5wEk0qTM(vdv2VMM=2d!F$R zQ8@#j??fHMquy#sHjO`eFto$YMPB~slw7jGXu*Ojh7}$UtMc?~+>Fahtrddyep@4G zE+Mi3e3#`O#*<)mRKKwnM@(8LnP-{I?WNX!igtRgSTs-*nk-C8_nT0$eqP_Xm$

((iId4T}oW4A;>+8+a>B{7W|xmbhTBn+C)wdPbyacG33e7 z8@c;i$BRvilR?x0WCBvc5nFpVH{ak@$APH@&tIg*Q}weaafHr;aCo5S#bujlYWmXM z?8^)iv`9+VrqKwnlK^h@RJ9G3~;{glV_n z!KjcRXaez<>2jh9$wxTyr8eTwW)q~bx!-yZ&Idu}4&~X${8=1k#n`!aVXrC#D z`b3?wbz^hn|2-`Py$oJ;ryzQpHF}mPHh}+zsAELD?DgdF_}qVcK!;+3G%}=su8sS3 z?m>2!+#bH<2qX5B7kU}#9mMTbc5o*&14iLB2|T!+aE0w-;U2L*c$%{lW&p!ROn50B z>Xlc(swnxMAq9^gaC=a3D47f9GMWPFSFi}hXS*hTm~tD7LutW6RGCpeoicF8#@MAW z%-iDJ&YL%xrbeqodrdm1ICUs`OdZX{Y*$(vI0K)%2Yq%Lmv}xBA^4BdelV{jXurU_ zw^s`5_#d`{f$n-LN1^Z{JC6nIR4O-?Q@x()&-(k<((!hIjcW`JJ?-YMK*r(aitl#i zt1%xt=8rBmIEoIcR*Hj`v%_Lg&ja%pj15MQq6H3bN?e~vZi z-%4qC({MkH0P0%~ix(ke9FQ8N*r9(C;)J;NqP!fI-~p8soJ%uvN@iVksZ|NBdZ@7_6w>9hd!a?SK=6a6FeuH(Jc91G7l0x z0@?06UGoM%sS~~}GhP5*I8tB?IZQRa79j{F1Uy*szd2SCPU}hjEh`ScH4RV6L9wag z1pU4jEPD0}Q(YIN#uX4Srn?vqm>+tXyYn}oQ5d$FvAN{Jj<+A|Yw?wE;0Gc_&O6H$ zWIhUw)g03APE0p}qxy|a3kcWEO=zx{61M+f^MdL5X^(Ep z>P$e|Z-?g*>QVhg2~DRNqzecOl@*_TcKl(|Gub4+Z8y9hbX61H7*!uI9j1e^8NQOp zQ(Wdm=q#l>h!%FX!W)#fV0j$!qgC=lEYw108|?{%OkQa}L?O<(;+ zKy5{gahNu;y452FZqJ&-v{Dtg%SN%ev{F9x~lxp2wfL zaPO^J2MxDbVR$&6JvX>uwK*k=vonEt*f@xMu!}<^%e!MFx#r=_JRCFi3mZSDUB*SN zw0F6EEO{|9D(z7#5!VOv1vu?^4XML@@%*kFq@(W`c!Bj>G zxa}=W4&*Up9rC)zB0sfU1y$uH3&2f41Cim=e<% zGI;A$+b^;AgwYrhz7BcMqP3DDJi`n;`)sZweC}V+%tL~>O#uBhjecO7$%ZpJuWf|%-%)pMTfOWcM$hE3^* z);jpZey^l~@5z~EQO`;CJ7UL$-=EYcRh8EVh3sbL{s_l<$85P<(f!1#Ucs^3BW7Qf z5Hl83$INpd=otr8G_M{BeUhF`VZ`ca`J!FX%gQ&UeU<%qTSWfm&ryUX&LSXZMiL?W ztm%nfr^FkqijEjWY3bVe*~y(Cv+38`kjQg~gOTuG2G%}OgsH{&!RNz_$M$$Uf)kwY zA@JlQ&itEdgS>*->xd-)Ot@WsyMTR9uRALhVR>IIc@C zDd?y%chfM~mZr;t{FuAG2#>NETR`|+8zj12{hMSZN{DNC^c7%`|C1}H4L^JQAZY`1LO2+t@mN9I1-700 z(|F`TXOdT~*ZU7G`;oz!7Oz!KFw!ng9zQi)+(i9KpnKpA5@$<@7T=(Mq}Q;Jr)fp@ zSD1tS``1wuodT0R#^Au{_j|uBEUgY-h;OEq&Xo72Q6CI=ILaW+xj#|hM;l)Sul4OL zBE7|p`GI;nwww9dlkxUvLoeH3Z_xT%{NF$3IZHbjP(!4&Bfvm&9 zE&HluZ@DU;*L}0>=P1bZ3o``OGxPESB#Wh=R9CG%1MBP)#{jNtThEPoE{kO&&?L0b zjbtKvn;g{C+{HXWH58obEwg!7$q{62_`@#xz3VYUH^`=t!rq8uh~x?98@nQ!BWn*C zrsC<%+MEFA+ga?PyP%Rb{cXF3(ZOwJPt?8Na+EdsW{?w#xMzat(LsUNG!PGsA$(5K zd|0KunfW>a92rv6A$a@PGxQxLHYj%22`vMF#M)xY&qJk^lW7xOR8IImkJnMAaJ$1W zVUB(37pvtrHirQunJVQ(QR2KV5n|6xx>5YT1kF`PQpF?kmn9h+z7vJ-w#7l4ZQrH@ z{npW`N-)USSRJY8HxGb45l4vYKz_C^an^PdP)$*dLh<)`V(Xj}BH_=p-%E)7!3(=nvB2oe zbe!U(-W}(9ZTMtYh}HXRB^uS}PvvKHe+PWr8l1%_c;ed#z5(M<8j4gFGVAAs&NE31{>q`fNAPlubx1aErFF=g!s^zWJMYFB!F=DO zEVf%F{EmrPEAMh!I4FY93th%>ZA{iP(riP)L84RaEvo9)4=MH0E0?mkdE+`>j@FE# zTGV1En3m5WVA*qfbP-YKaqps=q{6^ zN!`i5tYPPO+ZUVMZ_TtBMz+XO5W}UwW%R)uN;V-p2-|Kkm+Ca+u&Z&i#x<3pK~c$Sv1+0IcyJq3x!krC zg@`mthvX}dtz~@V)0R|=!Ra4-==RcZqV|H2pZxiE%gs05$_<%~+{nM5#qZ)gZU~Jn z#r#$NRrx1ZZZ_#(0{@Ba7ugg7lz6r(`in>!1-_yo;B^Vg?;Og0 zYsp^%IWjq2{_ix zZqe_RuhS3xyunP+dEF$GVi+~??}&q^fh|ww)NmrX@7hrZyY~eeVsAP=H#I`Dbwq6o zOJeU+K%v#G(fc=z-nMpNh7zdupr9`@le*B#b@o{?e_N521Kr?^*QIw;7Dgnuv*w|~ zA>EelkQ|eh=gqH4FaDEzMHBr~s-vsYlNryOIdGlmf+nu)6$tm!XWX@3#P1v!ms zzng|Q-Xybpr$F@Zw|f)6S&;Uv@*j@1g2B(|Et(K6@N1ZRI&5?#Kr1R2T$&c44>Ex}I6%Yj9--jIW7J0e{a-oQcy20DeK# zdyW$*tKeZ5AzaTf(XuMc$b|<5<-jGhJ94}7Bv(*hT;9CCp0LC4yiM90 z-hOz>cZtac1;mr$9RT5=j79dR>ypGrpKBfuXD;Cf)n#ETTl&4vPer6NpkPb?bH?wm z=LAH031Wlnho9tQlj6~8#Hl|R!o$6C;yi$%Cr*huKVr8F3ROmpDH!b&Z|npHRO3@ zXu^Bi#-9CnPBYgA7E1FsG2Hi`#FHkb#~c$u6^0fA!EqZ};cuiR=CABTZ&T=9dcX3|^! zU~cK*8$VhN(>2ADz_jqT*{i?(d?0>zHEH+#I#l9mM$(#B5W*p<#5QH8ZhjHtxfe`e zKMP^2$N1)tPxk7~^RPKj&cDU>5_Hr630nX12tW7=1rDIVYuot1Zys3MpeC%BOW(Y4 z@ZY+!Nw6g+O9okxa>Z$=Y4DQO)+c6rJnfH_1jxaHNlsqVzjuzUYywea9Ep!psD>RP z+*)30j|a>CN*)j$h5dGG;x`RCM<_+_LUkw|&)TRh`7Y^uy&~B)3Kxds?Xw8*#A{d0 zb-(+Fr?Sz-jbq%)+gV$|Uj_b5aDgng72{6M!ePh?IMs~_S zM#y~(v>QzvwpD=dDu+Co}DR8j0ItvhLy%!xzsEuLA00_BFCG=ek(^`WQCO?NBDTem^~zp1AW;5B8-I@pvL*_n;rFg zA;%I2wYNDGH<}YlepW-`mbAg0BIjk|&L4}Yra@&g;0r3;!4s%cq}%EPfTDYVoOyxo zF^jP1U8XGsVc{jzk>Qywg(&L57hZp!=ua8ltZZem(a5L=0lVP z)Q5gUFmd;>N;xjO&}2dJB@cA)geoZQ=tBk#Kaa3GdY?KJdS&q$6yJ{9aj9#GGi4(l zwwMr5--M@4vij^`r3=5fHjWV(Jq3^R1L4XBN%Zkp5Q^!ZgU5BRq8!`thnJQ53f`Dr zg6KMEW}>8sd^$q{sISIhf6LHBMuR@Q_yvdvw7~s7sihK$)A9s=M!j>b)u6j^vpv=M zMV6d+Y$rlN>UrWGAk#ykDs8#-0UMI8w?{$f8y;k)f|a-XanjPK6(#@IckbhNUKC*~ z0`4gdk&$@`m-i{$eP4s8p$YNm%-)LRa@jQa)qf9nuqA|qoHbdSkaL>b(@Wm`l^)Sm z+LNl4@I$haa*3|TrsO;ScQiM#gw!))HL1!#U7}2-R`qV>39i`@t|8vj?L9`?q;@{K zDKTx&#r<>H%11hW#r&Y@n{(Hn)RtW65X&}Uco7}SZ3$BPA)q|g9??-!`(xB(wNyUt zI(4+nTEyu^?m%$I!nyQx`4zZn0|%9EEGm={1NvW#q9=HvE1!p9c;n6qiO4QesgRHP z+P)x_CYPz+a%a<(VWe9ItM-~@G;;?d_HT6DMU1T5cb6-#AwarYY?Y`7UkX_`M89x= z7611dHQrSOS;aaq6%=l{69=CE&q$5e5;05~QsrKO;aEb5lrmLO4yXrn4kxQc%917N z@e6a3H+H@i5^;*$!)K-MPFR8R)UjWC5rKsn&@c1n?Eibe_ zY8)BNwrD#2zJ6*Pj}<(+E;wTxR5RIwt)JitX=d}w_>RQFlbb20FC)=&o;1M z&u**N?R@O*Zf6HLqG#uvsQ=a8@B7kfl8`Kmgm9#y^JZGpHPyVc%2(8blr9>=ma+S` z(*TXI5Q!#%)*DmTMXJpdBO_bczj0Xa!{JP%daPbpi?B(|*{Qxb4YO2vnP5M$jwqxo&Ysg3)(_>ATbeRWD z>vYlbW5#{^QMy9uQHR z173B+L0KX$V4h&e!D^GbnL(W~-B`(ku6I#ZOWFC$=GQet7@9p+MhpEFJ?u{Zkz5Td zmnaw1gYxV4f;2q5`}6)XW|)cPb8bA5c^yFK%MGePm`SrK>cId&f-(bsjO`*vag&>m zftiWI6O7{WM#UItP@g#GsL*DANT?#|Q2lPiEK*hM0xQqNh5o~S)E{2yYd+Z2kPGX} zo7NohpfSGRlM5xXi0!?E!v%Vg6(FX-^$^Q^GJy$kPfHGGrN&IQ8-aIGezu*($G@5~ z$vajH%7Lf}uMVGI8!smerT+`XLlDepL^%M2Z^2pZPeRUVQNh z8YZIBB~|K?`B#w$cF0d9vp|B}*yQf8zY6vB+?Eqg^^Yoa?A=_5^s3OslO8JK3_AZ_ zMF-@DG2sQzcRA!fsQD}*c2JX9;FAs*4Z5PK-R8tN$`Yt{rYBuc;Q?^*dEL?;5eQ!y zJhyz;=h5qT_fxEXNgPk&Ty6T~FiFGv^{4<#())**1<^@J&nFO!@ky&&*Fqh$#Gxa& z5K{v%x+y-XjXN;%RHxACT<)h%yN6WwkqX~68kr+(kgnTCdbEb*VaDP zmUEQPid`C=NTTAgb2A0coCtxsZj~>`Vy`g4zln)TzEIzXzGmvW5xN+&gK+xi(_`n9 zx5~rd*kdq>KVa>$ee`-w<#;M{YH=Ne$&`+$hzP$WLisRl9vh%^LH)wfe5BI4LjTGU ztm{rxvZ?Ff#1GSa#C~=}_{tL;ONCGB@$zR(Iy^8=*Uih{14dqQ_K0reNIOO76KIlI za43K$lSKAPo!bm-J4h2243#4EH=*ArNAR5ClIb;{{(!`5IcFlbSCNgpJ5n$69-8*Uv!aH(wAM86V9X!S?s6%cT7Dc!U4aR<1 z94sJ9I$~v|)tD@NX|h;yA(^?L7zZmHA|v*4tEKB6x}B2*2y z)B+=744w9@wty4aptR-O_6#b%D5jJps@FzN2J_pPOfMI{L~nnz9-LA333P=YGm>ysOMHnMHEA!g&NJ4-iB_K zcJeW>ZBv9V3j2u*ghS)d$+itg%ujzDIU;yUULU3)v4{H@Aw@ z_yYm*r48YXhqv*^14=(PFYgt%w%prwX4hY9iDFu6n}3g2eY7Z-Y^8Skwa-3YReyp% zWls%%oP>$hXYbIbq*PT{us$C`K1$PV-FlS1v%_MzXDZ)60IX$CN5A%{{Iq-uP%l3o za!uOIpn7!ZkMmdBZD0{gUAX!A)zrGau(#pmoeQ$0t#tNuX42jq1~=0tOIw|I*Ex$= zk+lGYdZy?v_<4A=AaH^qb!#EE+HJ*tPsp;hchjorKm+27zbje@nD=8UCJyvPZ=S5b z0L27%|4)iC)PoQ=1N1i(v2FabwvVAR7nu3hj{_1f{-@W&-M=0`zfh2uHW1Rvce+un z7mL-eXA%hz8a|O8I!-E9CaqH9=La{bWKWYC1P60p+f5!#Zg=33!@f!Zituo3Stx*7 zVwik#AXV^)Q*8h5@0nXl3j&Q~N-PnYf;AyMBdONGy**V_8jhzcgd65wPE%Vb_~V%9 zTUqIy;;VpMwR6KGB%=j4hZbDXBa9m8A1M8`#N_Fj74vh?4}ul`&MP=E0hSCgADwWM z(bilRMd}0&{tw(4&oNnXS$rL>^kbV5%P2eaj;x>4o@0J3+^|0Pa9!=*{e{6UH`Q6* zQ=hNns9X=$FJ~2~jZyynNPzobhNAW})l;bkow0SRs*bovrbejP8L6|2!9a;R=UzWG$B zfd1CCyL4NeP{LgFhT8Yx?&AvZy{L8>DyxIwPxh;NBXqoRKeT!pVuADkB*s?b5)4n} zG3(%A1Q(?K`s@A2xL+t-EWiamZ+)Bz_GMBqIu-rw>d-_mcz9~>K)f=nZpX=*`!WYd zYrQBt-BVz>yW7m{%W|xunyJDY(~~2F-@~k22IKYqYclFc{`MB3fa2k5+`HSEMd4TP zgB<7Kohg zyrp(A-70@I^UVIX%F+f4v;ji$%Y8r3dEei;&UKxC zCX>(3%+9RLe)rnlDG`s}auNIF8+$^(1|h4r7Hk}U=~jzbe%9r}>`A{$#quCmFWUq} zIBU8W+f#FH6TleU7&ot1BX?A=@F(YJn{EI0bLaVtF%AltvCd1-;?3C<%smhu>$HBB z+M!;h=QZ$JpKQUVs40o3v^Yam#$+dWkM+6Uqn9=Jtw85dSE5BK0S&jzhV5!3yU%n` z@Ei$k^Jco&FItwV1TO+W%jP+)3yI@ufj{!v^yS)DH{d&zg^^$&IO=(bHsQh#JDG=Uk3V5%KAA+#L#9e0)Ut*V<* z{O~M=(I)RyAei|?K4}H)FnD)y2e-VI)fqIfRs9ZZH<8LQWOScmhC)cRKI^CYth5+k z&I|E{G$rW^jt%flq?)I;_@19W^?O^yDgh!rm(u9~Zx%&Dbkj;|6TrdD#KDUa3t)j5R^i5b!Kc6BVrVbvM^=J;=;K33LGv|h3Su@XlzOW{ z(Sm2Y@HMzxJv-zX&iXV7z6Bd86>!LG&)q=^2yy~HsCyESj-=2sj6w?saEtYj_N~tK zn5_Ch{wAj6NBocP7L?^!vtfVy%A_Lc0R20B zjfo1$bb1^O--OicLHVXq7LCZ<*=_W}uAeIJ-BgV^&vdkXub4r&{)InfSZHjY=bX2Y zcd1THg{(BPuH`FRy?&m8fjE?t7kPS$&t3wZdRehjZW=*ld-@c=%!SWt2!$|(3rkGU!0~d80NV2-5|J+$W>3F8hp z5FCSx(Wi*1^xZ#yyH)=6Y5$bFW4=`ec=G$gd9ItLEqQiIsDXYNNU6Ub(~>4q4d8uO zB0(_AoigFdrf`O%vwii^I_@%?T65zbJzY;C3lzL}IaPXc?<~>c*VOHq(=6E}Ra3N0 zPp!$#>)mSdTa)IIz@9vNOtK&nC=%7fA;=7&-fP1~F{byS=tXw1%IZ*~JTgNuHWzi( z3Gi1sG70t)N7ea6usNvif6g|xK;P`SpAm-y1~3GJ?<^SXaXPo~rM`i$qyv$cdH%r{sKEEIJD3h#ZD? z*nc^dHm(-bkX16bnq+t6UbtgdH+v znqMDS;Bi>1?MXmMpuZ}X+{b0)M|6GDFQdfs^0_iWPbD{fh-~ub_zA9S2=(PYo2wHb#-llFjyiy9wQ(5_%hyYE*tScom$Z+^cEO{*z=<;|J`S)=$X zX}RqA8O2Tz2l^2K94Z|$8hFpgw9kp&+-!+XmaEh5Ue*tw*A(@=34U{b6`Z?sl0JE2 z*|YK1pX7VS5iSWk49W$;BhAb2KBk~$i(n#Ea(Bz6uCSDB`X?O{+w znu8i&)6O|U#%lZB8b2_?QuzXA669?}lm(gu0JI-H)Z2GNdcNmHIFWQ}&nNd2|87nn z7n`QD2)e&&s~wTr@$rt!167!uZZ3-A&8BJPwUCU=*ZiJ{|Gwij&@Vki|NeVrM15Y| zM>d8ys}R74cCrjtW5F#mI$MUgq4qPiOc~NYNNx>g(PWjOhk|(JtJx{P7v1zb(yIns z;)PwE1(7xL=Yx8y8pN4gX8~hlj~*^VTTAGIH<-A|fcgQCIQ;8thFpl$%|g{tEF&Qj z{-|Tt_to+na=h7`#P?Hpb8{)w3GYws43MNnWs8M+Px^Gu;6!>sHn}m`p=!tMzw8d^ zE_r7a*^}&mnmIhNdG`Mj7^H?4C!pasEw+BDv#%Vf&<9q|{X-W7Br_u3){O?EBQNLR z=SH`>n>gn3(Raj6uQx@#hKOab+26~*vRyeWOd7)3hJ+T?&oAiTT;MCfCXE= z2(D=i@BL-n?B;Po)-inb&v01h%*Y{IP31ZR(w8yrk+bpl7*sp6&z=w%ykQ8xMc;6d z0AE2LzgG{5SVcm5Y!o+_U#=^NVPsGQoS8VMFFx6gZ-3VVs>-f{OSD_Rt+2B61-0 zD)f~6okB$0f&#`;Qt>#~9!%m5*k#+gv@Tac z{G>%(iU}oQ!Ct3#O)7S3zsh3ciLGI`x{)xcwCI_&jwt5WPJ@;IasNRYRd|*1T*T60 zvN$dc3Vc4`H5m4TPDOT)Iny4@pJ|WAfiUEat!MtILQBqr{7Th7#V_xG{x9fTL`H)H z8U^wWT+ocY2*yM0%Zv#RIsy8^(kMyFevPceK!fX)GFn6%J@yNL;YG@KT0KH72E5{3 zW-Sp*AKIU#nkAdeuSC9+=Sk2ijsVl`TdoL?mD5yvfj{g6S;W0*-=~8(2D~O--k{=D z#YPT)usg5izkh0Ug|g}~u6lB)%LmgiA$;tk&_L(D(|AGHyjM(c^Mwk|&KJwqS3ih` z0zcDQbApCQc`j34h`@W`a_MZ}p+NkB3)N3Qq;o+CkL1KGb;Ar0<%g|r?YmNhTI5J> zs(z!Gv26~7%tIsm0xck!kTqZN11#*x-Iu%>$5W5lqeGJ z9G{r|#d7pVOQBJ92dAg7oDdY~y(7u>A%xWUr00}7du#s%TieJf_ur=+OiIGX0`;Y< z;Dk>szS(PrlMHM{YF=-d#6%c}t;)K?hfz{m!0zEL=9nERLDZ%V^pJoCcLp}81ngR@Et=!`p}e}aa*Hn{UNKU(rAm+Zg0aL zAmlNUE5CPZsl|&gmcArL2kp&ZeHm@j+Iyin?GV^d6|F+mK_kCM_pihQ**axn#a^g* z3h3UZ&w-n)b?l!wYo)h1QS;qHXZ%6~hSzzOV|JT-^(By%=6!C?=m)##rqb`fxm}-) z{x{+MU#SnND#Bn=y5ktNE7I`6+9>Qfj`p)r`JR(xy_i4Bg7H7iX6>f-T~5+J-up(!@ABYyEBng)oKCPe8{E|i6Y|tJxC81+xY_9>;qUIzJ={x{ zY50OQ0LZtSDUTKURkmIg{(cJmR{B)zq9POH;s{P{MpYbTUby37wZb>?py>@^NX+|91&^it!4Vm6BC2DBI>B@M3-?!+5Ne9|`+ zTLT=$P#t~t_#FMack&K^Tf8>qk@i8k@Ic8jzJQFOWOQeZC6#diANC$e_fK5onFLizmw_FG8G8kb^_mX2l(NviBiq;Ws|>$;G*7VM%pM=m}cEMu~oar z-blPLJPvuA?xJz9nIb!t>|;}(P%<-fckk_}qO2ftz!%G`nsAfhQ7_o@uB&YL^w1Ko z4L6^s8NO;3I?Xf>SBbw1uO0Z37x?Rw-m4Wk14n zS}MbR_1`HF{iHvl&4kOUtS^7DEmd*#Rm+G2qpf139_H6UmKn0Wz-VG#qa#hV?kku%BUjwknX>G zT=`frLk4X7bHAgPJ8eZoB@(d@!^}7An$s?6lRFvhJ~lfgHhGa+6J{6@Wy+6c=HE)!*G;cKl+<%l2lUjh373l!kin3CGJ z7)H~?uV*Yj`Y02`mrL&J5gq2iVTtR|>cwIhwab(NuVUc&c|iBLlGC$v4W_dwym5G3 z!rYJ-6M*uGfTHjzRSGl9J!{S*!w4u!PG{04+!*gmwqJr;ZT{~!(AFNu`6`w}P+I1Kl-j}NyKQ;GbL^UR&c(UqR# zIlNw4yf8e}UGGM|VOT%3cpKuxNin&sbVb{gNXCo~H;|m3IkPhSMjI}-=PYrlC6o#~ z_)BIFh^)rt0%$gVzSMgqkiXfMlf7#JM`uLv|*9^Pc2h_o#&gBPo@JG#~uPd z4YVcZUl27KtNbt94;nC538=UpAfI~mc%YjOpmb6G^k0G)!;di!{iMuoNlLh7XHUMV zLaq6`=e^HR3BgCMmruQZlUNxLS>>gm-EcDfC1R)SI^fc~qT(ezzmzAbkijtpB=io) zMWPfgo9H)J&817~We-^!((0%cI*Ln+PaViNc`A1I*Y-G3a@; z`)fuYro?Jt(qqwuzOfZVn{aNVkC#R(0xS2$HoAKEJg{X_#>`-nzRIEdelAp#E;5oQ zTkw%J+fx`jP^N1La}>AyX>@0BNHLzT@T&C9H#dRT>L){f6QG&T)D$oyyn@29XXU1E zl!6?X8@>UT9cn+~qV+T3lz9^D|GbnJ$amXsKgFbPT8_hHDA?lr3HhoNq-7Q&S|;^f z{fqIvb2zfYx&mkj{i|7gYT7R&h3^qn7cQN8$SSKJYS;TVEWoR8QaGGr_)^|sCw0{P z>psLv7Wt#?l>){cg;Leo9#rB@d^#wrIi#6dMw&vFIfuSc&su%qmD>HEJm&Kg3iI(# z`;gjgcr0%aT+AC!Cd&2~>P?(H1az!r+47N#e@9rs6a}d_TPZxTU^g z%BkwyV9hEXCfqZNwE;-@5YPMRV(QZXg`O#9Jji^W;vAE(g=zCG{rJVy+$?3r3=pHv z>+&@3L~W3!`aCA9v8V6zHNEJ})BF4{zR)fjyV@U!UHh|r^tVVqe#@>FPI|^tpv#|n z>0PGprUM{P=H#=2zP`+{?r!CjL(jY^8jbE)uJvt#kL0%cA1G}WgV9da49aa}ewF@E zZA@G}SUvE1OYH~j`jJF!^L2p6r8oa{{CRB#T8!60+^jszNQAklRg=>H9d_6$j6hu4!M(mby&H86 zjM*BT*?;|=>#M;45GM$XxD2>Aw>eRZ1!Zj{_)V(}uk+Y0lt;uYcoePjy8JfVhI|gB z;oPG8nf~imReMDY?)&8}h27F1nTVQcK0|n})$a1;iQS#+kl_+{p^x$E6bBc@LdBBw zzRc@_o+?g6(1#ksvLfbe__56B#USvBGGf{L065dm&@vlNN?#_ba^r!^vY{%COCqUq z-}Yhnl2w$dwoY8tCi+S@x@wp8{)0)I9_ZN31#y%_BH&^{^ZsgTZQAcXY!Zix_WAOxC^c3)Y|fC$+OO?H8}z*UkC}>D3EbSHTb#M&tY~*-%ns&)z02;p=n0q zXbg%^@N4AG8U5Hd*40!#b^pbh2kk{S_gq~`Kwr{v*@SX)tUc_$7PfF91SBls*V2>Vsf zKn@;mVY2rd%vwMn$$dfZngEM1ps#0B`j;=-pen7rIE80fwxdq;iTObLIvS-N+s%0c z!JQ$NtxmL|P26ju)X-vH3cH@|sKc;PfyDTW|2#oQnMmi;`_x2fK})gKy7%do3OX%? zUF|jgeU?3eE8F)~Y5ugXP89TwC}0{tcrG#CE6~wwL6c|T#Tz!lrGHYVSd904?#fz&Xo8xS@w2MW z(4pU4N1v>-ruUkyTK1!7H_;FL3hgrIyPt(P$Pen}FQdb9wv8!7 zYw26c_wH0X5KVgcZ2W>3xp>iQRHwejwMud;aU*Yr3JRw|s84F#;0kSgFzoQ$hja^Q z!}kSlRFQWWnxcc|&`;{+hcI&r>n25pGH_H;N#U$65yI#3M|{}~V_nR{-FoA~GnZ+6 zjRD*dKH4{aJ}%hbUtUg)O6o-!Taj~&D#^{!KQn2>-D|Mihy0fA!Px$u=d*Mv7YpJw z|DFK~LDk>Y$6c`+nE`$nkS)?0eO27}3eQk~>_95yy5P0_VbGUchZSMl1s87Q>yZz( zWU7{;CH~g$;?u!A$kY79lYCYsHE0G1pcdbJxb=<0eiS$qJ}@_)I+?-P%wH5)8;r@h z9aP6AAC==AB*@8=X_5MhZ@q6x)|v_1>tO!Fh2K4%2)tg*8jG=sN+3-qx7Nqt~y*;68AOp_c zj{Q)$Kr|uQ_B_G5QoHM~#j+4(Y^14LYUS1kM(mU{XJ@(h0f=v6fhKI-xJsN}pd7#J> zajoqd*4m0YCxgHJ^)0=2XfB8)@tX@g1XBspMY19aPpw*2`h6bA+}%Yo5A?yd{sTxo zWq4$BVUD@-j%!2U&^5ViW0MKt-Hx4N#lGR7d3BMpBCyz3pxWIl58v@8DC*D+@3p=C z2W=Vdka-G@1Fd@W&ix(6A(PjXP?2mlPyq~p|?7Z zVJQ8kc6-{%q{kl{BwLx()tS|NK}AeMVZ6LKW24v*zR4I<*60F}=t;%~vcH}^cE!)K$ak#7pwD7GbRhuRo-LeC!Oj?-@ zMEe2Z8+<&U+)g*N{0WYKUjcNJ-rr1|Yvx1o-MuceA=Q~{k+PL?0yR4!J7q$K&QU;P zyvOJM1Z+kw;uqyvCxvT7dYO`oJB!fc9~PvEA<7Y|4Zny&${dzxkss5^M+b?4BLXmV zrAnST6A1gSr(*fyriAGh!m80LIT(gwqV{;Tb5~AnD??t9YMVB!Ucal~LYQ&Y+-4iJ zy+n4W+glEd;Fi6CY14ZgbCZY@~n?EFle5U7cugx$kv+#Pjj%#c9*3$Tj zWeG3b-5?DqpI7VF!M@KkTCCXQ%M^4=Sj~7NDWLzZS9+BG_tOuiK;AMUNm$%>qH2M- zFE7>k=p>U}uDcCiyr|XU*B_gz8sM5pJ)6NRVPbAywY| zzplf14l%9QA|I5k(lBSxqrN#YgbDH7X)JTBKmSoZ8%1l0$}^tiz*95_>z#6HuOJ?t zyIU1XvzZfscaL(8b&LfwM)5Mo7q}RlKw&MyXgu|BiNpVTYHB9+i9xMa!4UA^*H*pb z@Rb{R``y6{`C$z1#ed3^!dlX?$>sB*9*8FiF%0h`b+_dZL0;%wuUN+FwNCe$ogK0N zqd&#GEgt-43Q(61+8rFY|}Hfve-07JO8Xg>%*#5{Tx z4CTT)#a77aZl2_+#vf`g2*51F^_f3)=x$>mWV)Gb%CbEMTCmT5?HsX}EoT+Oec6dj zOI7D%y$RHzS2s^aS2zSWA-FV`+|N~i(>B{8rHd4l#@%jiulZ}Vcq?~;?y*Gjp+RsX zoBv+)fOQ{GcicjDsEw~vl#xL)ngnF?WkB8gxqryRuM9->*2E3uwZeV!?fO*Q4fH-a zC#A~@q56rTr~UcOBYNY?r@vDqGx5BuQ>G4xt z4>ZRI^l>T*x%5vq(-42^4)eVUC2~D=CP-#@6w<9IC5OQAzieM9+)vh}N|R`DnN*bq^2kx_@r5X}4zpHJsBd2hl9WSbfmA2e;30E8QY;V;MJmnuOEj1~=bMTN0DK zR`y#zTrXh(G263)OK+ORGm&pG7UvIF-fRTmgmu04$sRgEh-Pw#Jo zy)FcaM3ir)n}6x>HGgAPVokYzAz{8omL#^gM_O2l{$wQ*)^x14HJA@eeuiHAQhnp! zt6#%twYAUGHXf&+N}a3n(Ft4%)}f%?>#n1au^eY@boswMoBRF%@6M=qkedn z##GH8=I;gUB*s7~I4)C)^cy)rCsDj2@atp<$73evu=*0y8@o9r_+M|k;u&~YZ&6et50~0@P;;Od2 z;_JhV`aV2cs9aHRYtH@*d@ZX@*Mg*alnQ5{FEJ{O<$sK*bCQ-;C!TBd-%*D9@&=k* zdx<{1a3BC*?~u(L#Nh>_nmeQBrNEXMT@KcuLWU=e_;lyry2&r<8+lcc4DJ{MUkq+a zE@sukUx00?BA2~3WreNW_M+uK(jkue7)y*hDmpy^lP5njB_K<8>ruNy?$4j^+uS2~ zZy4n*JE9H=ik0v0NZwmz(D{=*_C{1KS?aHVD*lz`0A}8Y8}Lbe*+x5ZNsZyo)St!F%Y9_dHvju-$g)a zfNFWjwFEp?hl5)6^D8e7Tszj15Hlt*1}&u6<8l=GM5qH;8%cU-O?I@LJ&tWfhE%zw^2Wh-^Y*TO!l*{hxYQMXzUh~n~ZUv})8^SBr?$<5(8-T{?2m(H_2o{tGT-(3G_Y*?5l9a~OPok%igK;NGP7C; z7&V690`lj2u7;X-)o;XFCp|t^I~k?0=zuh%#Bw}^SZ1!Yz4!<3Vn*`IuVmP`kt%7l z67h!HnhfcB;vZ(X_1#;GG!fTVAzdnlXkZO;|3+jbe1ESvVlc_I%VC+NTL{%<} zcNCU%f-U1ueH|EAto9|gZ^$4&tteGB>({dT_$SEmJ;HEXM-zgqGp%SgtaGtS?1bwk zuhw0cLH+ZdG0ajsHXXkwbPnV{e0#N(I~~x-<$nI=MU7hDkJD>57^QnSM0mI@?1whQpG> zw$Jw2gxDw-)#SPvBZx)vf+WMOC59KUKJsblrY(&`&wX{{AK1 zO1s7O;~DW2AlJLh3N1W;l(Anp^hEu3#6C%c*ai!KB=&qHC9qGRoh6#})g;MoZ=*x- zVtD{(Q&C$8i5%xD0mM3=Pdy0~6A@oqSCaSon2|$9F>Y2Ho@DsC&r2}^J1J9k5N1SvA;hVvoQBN*%7-8%h8a`( zISKyIC*x>2vVNCU&Xd@~DnNVz`l)%h{}%%{+Hjs&8o^a}~~Xo?SkgoUP4irh+) zx`mq`@yigOR0@PURjXW|nRlgehsLq~aVvA4z*YQ7&AjCQ03H;6@$uM=QjaCacv^XN z-P{&-=jL{G&rNw`E1Z^XaAWefqf~b>{$Z1F%LlMJA9#%w@}pdq#_FAjA^dcv`X}k! zLRxC1rO^{jP_WL!zL`c-lcyc7cL2>6hV(ux!g(}M*F?_o?!dL?@S_689rj}dZumQ| zEBQVfKlhEDD|$g^kzk{o_oH}jYH)rz5I)Jt9^~tmW(8p8sdY7T@)F>WW{ly}bMC8K zbE-qjf^$o!Z3KsPQA>a)-MeZ5H~YT9oyk%W2HF$>3S6h3dk2BJkhwV5E@!r%)N>Sn zJ;Gb#XI}sh?y8=C7>#d-PKkAqYLjxTx@Ly7(sSRco9M9j}S2K$hNz+|Sxshs8Qy{&yNwC6*gpg^09o$FxXZd;cGA=USE z>AfQ#zX|y7+&`ctzU~`j*4s{D15cs~FpNZtrg#lX0GC-owQo#qe|y<4<#Ac!J{0c*`vxN*SKD`%F4>q#FS7GEUF zn9{1a4c;Pjp4z791-qq^*I#lYi-Pf>2c;epREubq7`$hZj{hdG-t#6cK@1gUrIs>X zq-Q?Lq4L(ZAiQBmmjnk#e0@^OQ547Q`_qyIoVYXn&`SiP?k64eUVL#ZjhW`nfH1R0 z1KI@+N)v~q>0`@8)g|{kX8t|4f?9s17-3EHW~im zM0SmVYD>d&O5%(6-vs`Iv8@i+$_>m@GtX$KYcju+_1>~!@8~i32snH^saS=z+UXwOq8-vS8Y7Y--?3i=Gy^?2OqaieKzHV|(g5N6Wrrn>6U zDk(lJ^r^%baB!I8Uh4CTA2e?=6p~eto*EHuyY{_ioxz{ww3e}e+fmz2x5_;6PenYq zQeGg)JvZH>V84i6;Ty9pGW0=_N`_5J@J~m&F;l@2VWiCGN7So1=LFJfxJQ4K;O9Ggc@nU!Lxv`QG`kp7s&et$tP&a6m1DY`Qam%upy--ZnBjM@2q zkFoYze+(x`CJE~qlHUh*7*KU6h(GoOP3UgAp_*n-tWHuotP+Ue3zgg6enHJf5^aO?HAEm<{305qxXwXFtyXb+RN`9J08$`igh+ zJaxi7GG`UhbmCy~b8GdV*K(f+j(uUD<84S1%-x?rsA~C>?$8k z7q%$lw&pUXeue`GGW1za7shUu&XMLGz!HTFh774mMI?wiR=_&Ti}}4 zu4(T|#)d$trDMp3=hjq-qi!>A2#_2{_9J#(pgjLI4Py0J5cO{&Ng7P}LqW)?yngLf zibUILl`*MEOXD$*R33N9m?)9~YISSdwaD|zMcgh4yb^pkH{2d4!WQ0DVa|e7#2Ew! zI#+gq8E)v&40#4pQ0t}UNpq& zwmTfBCByanm5Ggsf*{W=ctr}{B!X@1I$0?L7r@eCzf`0nSR+K(JWk6~2ryBW6sD>p z>1m&esY5{vhjaZAPKn_9Kk`u;wEn{C(DJ=|QTe-I5p}4P2O{aIZ|VRoSqT3H+3-w) z2%A4|20z?jf|ktG{{q|R#y$~jmGC)$9qWfWG=wV*+vg7oi6HSPzGKjFNCdM|z~2{i z(z4rY?#VYq$%#9t!u8IDQtN3PR0%`w@vK-$P?m(oApBcz)S+KZk*Y*8q#bmK+}#~X zE%GsEg!4OwcUy|^v@PaOcu@ z3M!pkeIs2|jg!C?WQG!K73!-0*H;_Q{U=kMo~|Q<2uquFJlfUuDoO`mA{ z36Ob-*WU(5MSR_3H?BT%>Auik`Fj2D)2;|ZxhPwzQfA^shCR`>-WX4zV*MxdV)u3< z=QZVor+5!Nm$^b0iQ&@aPFd2&+zmtPx0lzG$`i~gb<@*k=f@C!8s-roOEE=s9o?gs zXG-!9GBSn}f~$gMuE|hV>>xsQF|<-$%ky*8w_}B zHO`{??tU{a9r7nhF#E5dT~hsP-+uoCWMRi|vDZGrwAyI=a5sAB&}*5FX=$9mgj?e6 z{6H;r=!}67usJw4oSSI7Mn_jLwW~HZ*c3_0XQ6)@1$#wiaE*}>1?%w(y8Sem&j^i^ zZo_W9mJC;DTI>zfbiE+8*E8MHvO7%r-o*iqXS(2G&o7<#sr`!3*TEtu|5` z3Y5O$7+m-4(iL!aoXJUl;#2|32w9 z@FBlXQlI6|$Z-NcLQ>*@yjy zjeg&hB5v*zR3a(7O>M8F7=6T^aKgE?0M6(4De|{|*KIk^8N-W)I^a35G*`(XBa26v` z4dpzQdH24xHf8^JYn~)NA}tAA1AF>J$@A$hgXwI~_;&?{b5ZPS7H;XKS+k!?#ZS|o zAj@o2Hptyt)A8?Qrax>$^we6lTPT6TrinA8h~7HR@#B6spl`Zy;>qcPp_5}G*n(Oo z?zyVoebvO2%JIe2x=Sv|K0KCYs`y8j`tC8MSZLHIjTK zJA*9kVoKl609%oc>Z)qk|fRv?cdURUQ5+8%Wf2#LPF~GB?+wOxv$-M zPO3>y_ADjyaR9?z72b4RIQNwcckV6wcB)(Jh4Yc_=Ss3o zlJIsXnRpxM(vwLMt#9M*{#B+YqmMP&(7+v5nNx?>)9(#bpzK-FBH$vt01}iOM80zh z`)c=EkDlydT~3&!`Ipjjn^@2??j;tfGrN6v-^a=eKgO0Bb?DEnov0uTFlt~bdRxmX z=y^?P1FvpNlE5=09!$10t;5%Ivj#NC}zfLdf&`tqWkC9tj{1B%3^rCy|9Ytqv7xOO)qo?m~I{ zs)cu4Q{lXU%v=3p_gSD=tC+j0T;msu=u4T}N(j57ywgE>V}fM?;0l+F7L}_hp_>c$ z?@=`oOomfTeVl6lx5I9B*Ette=*bSmy>t6)R8zI-$bigwg*|&86HtwFf_`0o*-VOM zpbmx2l$&_hKPsX+CN5h-XIq> zW2`hr|JYHOdyO!BO2+x<#6xp}3y=lZc`qIdp?2Bo!q;;rAu?Pb;h9QE2>et^B{14{ z#Y__Hov+Pq`mZiKE^`}XJPXQ{ED zCfTwduh?i%<$*3dr@aQKHNSi2B>1=nZ29blRcQg9u;eG|D_RZY!;$AQmU3%C);IsQ7vDeMp2bk^w4ij z?z|9qkY4jkPU@h@wa_iCNZplX@1?lSCk}=00i5>WDAzF@wwEP+w>Q|Ae^Vr){9VRP zI4=z?7UHE9R5{{)LVW_R-w4e-pY>LUs^)EsYXE(06%g!OkhZiU37;}9hYs~KeaB2F zZ};jEnp*3m4@k6>V=|9bLEX#W;ba-H8=Rpk#Vnk^F)MyC3#As} z$&r?SB*;%?E|TiEe1FOEky*K#vI)Aq37vga4AXc)E_PP3hE-Nw9W5$0{z==I_XDw) zA;LCX^p!6U1q~1T1UZu9F5fBA;IT(0yFkUd4`>yJu?2Xq;rjQ+IqCR@3*<+TWxv>~ zny0e;?#=EkDrm@Q3nsnrjT74hJ92?e`8|r!%KcnRqk>K0jF?ohp`gK+qi_yys$|e} zFzZvzYfhOjk3*kQ&2u{Y7P~?#E0(rDwsQMss|Us8g-vaV#MTQXh__45>k?!jAf=g5 zvF2!I;WE**{Yl92k4QyZwuy}*0rK#E(qYxSxYPNUxKTZ-c~*k+8F=l&yCP}(OT_m* z2BbNf*YE_A7z)}iC`gz2q&y}9B}Q-kjoicb!jWyl2{h*v^N%KtbqGMNFLeMFiTtbz z3p1KdS*zzY=GP~Yy?cfK zeYRTOPWT4P9JilPFI5D$2e46^%Zc8;ZKF};#05^jKTw!xOj1h`I#qM zP0{lv)idni2tQktkLUU%l*7-!N#ZH8SD49csi=M=D0A<-y_oV`RP~rp%{!1I$|u4V+22=<-D}5 zJz_#*be@Dcc;R$uofV95$0vXKB8nzE+u^!Wu5pF0um5iX18Hl@%~-7`4XXwV zx&T(E%UWoIB9-1qS$q*X$Gg!C*EeC1?I7(uc}vQK!GC_3)?Ut|uNJ&Oa@d&99D<&> zXt7R|vBm!+el!if#ccjSLP5T>Md_o1Lx>m_RE+*n}#z7vx17z zE!u%7%kTFqgx;b{cL$xV#GVKXK1uSuHN_-gM*GX}ENf&3-+d%`+F$m}jL_)Rg-N6( zNX#s(DQeq@JR(H6?}mD9*|?Gx=gvm|=@UEiR)9X)(YJ1kJtD#LKqI9mv%@c!iNyaz z`a?xD{tsdA9n@48y^X5$-kS=cBfU!}NC!crcMy=SAiWdm9R#G8h#$hmoc-+bSlx%bZ8Ke8vYX6=2}*=wz5KTpn?$tvmZEj9dp@HTbR0tG$y z2b+Gwjk-9k>mx3dUixVURHaS)(3ndl|IFm(l#R~_6XSDZ_{^kYxTLNst;5gwERgDR z#x`G`7b)$i=w3t9GM=)#kX78Exa)@#@B7*JDwayL2IPqMn)M#G;!Rp1`6f4z2QH5L z&!3+gxF=fhMrbIdns+sDp=_7dkr{5qoWEN>-kWq>kgCr^x(QiuD)Hn=2M8s#pVBIA zACssWMRhrI_pS^HomEFep8digT6K#c$pAk7M$za7NoE$Ap3l!-S%f-SysTFb9~OTxRbY5W7t6xL z&G;_QwW|~LF+uuw3m4;`$i%;b$dG;81@$*u9zp%E7aQ=DC7Pv)Vq2K5I@Pl>&=>|0 zKMszoYCRG0PsQa$%swvi5dwx1YbITXH7VYUM=c)zERD3;k~T4fnZft!beZccXo%b-Fmp`19OM^>w>03cZTMp=h`5je-!um%@YZW)sUYF(4~MDEr)HM<8MnO{ zk|EEwU>7gCmZec+UM7{>pwYzqiVO-O?MJ^G4~JPiQNybZU%<;4J%OR~#O@O!|MuiK zNU=Y)Ipf=&z$Yc1ou)^atvW$6{8cvh#vSPE_6;p^*Vmc2D~5lP#K-eak+Md*!flaw zYSF^KUu{4k{yrQ(mN+LEv7=p75J$IH88`>HbcB4!gT?=>i?(;+Vn?4f3WYIrO`>FY ztBmenPhq9QZ-~?&))kBWV8|f_&Hq|p5x7_CR9*eRz4Uli`=5(hXThA;|GRqsAA_Dt z^tc3IGqpq0n!eaH_f7UZ5L+`xr4U=lcCVsmqYq8RgAFkfv0-26tk2_jQk=r*A>pKU z6+}L!&wU+4l1IIPd`>9dvSS9nNkL{1g`oZR1i#dt6G3w`3 z7eBRwX4t8w`(aC3G5f;XwH73;6@ffc2gjQ$8GW?d9iItUE!L?&ZA0~nn_{ZRDcuY> zzck|a+x(#RHXMaD{V}dF+UE+8%-;IeT@Ms1N4AYOh9X?7Fyn%3;7vZ2nwS0<-ZkE9LEs9SgK-k z-zu`iB=8a9f=WbUPRS(@y{_VYD8hVJT!Mr11HEHwDMg|B*i5*%fS zS+b7_GLMT~xaJMd;E1wI`qpX-wtX1FGA&3W$mT5?YcD2MY4)G0DSt{uSZbHFz{6k8 z9O$_4q(Pkhh3!Yo4@l} zkj5DsVm?P@di(tUV>s=#1uw^5zq6J^l|*c*A&u_v_8Pk|HVZmf@%GZS$h{Zk>ou3} zMeH40D#lJpWUtkr+#lJq5)owzf+#n+U1+9O)lEVf^x@+4$YN`wH`FnhKVxXZPoBl2uIRLox11<;Gd6n?_%YspY;|3;I6} z>oqx>u}}!Y-vdq(a|g=ZTEFgU3C8$;ZgwIcS8sf_JtBf=Q3ZD~=Y{*TNXXwmu;nj3 z6g%=Be(Un_4M|MFCY(ryO8+WLpkV7RpuVus4|{z3=2pzfjgy{zeuVgK)i~#eL&+@X z$Uup62v4IK_MdOlf-6m{!qR_RtShWc@GHfL&xNRXuIkbqhegK`>|H&=4gIAWsP5G$*hzHy)^=JW5WOU zljQg2$xN-(?d^=rk;)H>x|B&Zqd4WwuU#VI(BH;<`I%9^Jo4l25b;8dx7N!b3@O~$ap$fnVx;|yw4<4fJPx5eyi z%`b9>knmFO+huipbx{S8uH<1^l#re&(?`n~^H{}Ova9Gd;2 z|HIH5Uj{uK?n5;FmhB+~Z);85O|5GjcRHI%Qf0HpsT6j_`4QN9M5Ph00*;zEC^>;YL=Nl$BazSkRu%j)~cP zk7;?Fun98Pi0lXiHW2K7G;j66>H$|E$_=CNzE^Wpo&7mNk3L+1R`aewR;Yv&;VZ&g zZRgKWIhuOxKr-h$q!y7&ZB|Hq&m;05{b0395|kEA8*n#mPspCvfpS~pe5QXKfrZSDz?lBB%}lbXoN*E9Naq=D0?SS4@Y4FAuKEeMD~wd@ zu&f>jgTs=ImNFkwOrG(cQPEV4VT=tcbRLWhO`|sg|NmIPe@{tn&>LlTKfUPQj){7g zmLgUB&RrD^Tw}F}9}l-ao*<^tb&tflhx$76)5^u1#kG2@aJi;{r!Y2kpaYv#+f8mM zrguzzbFCl`Bo*U5Q6`$dNYsKO(H2KTpIU4V^LYm3QR?x)Zk|l>p9_>m_U|7fP_=i; zADz*^1BJk@scHpuQ+J7CR3e`kpLD`q!p{6@Hr|$6Vnzn2-*!_Z-5Ji6%aA`2n8I|e zdRL2u2)%mJ$nWqY&*rx-YA44ZlKgb|!t}{=;`9@A47H-ApM&7!iq>@%<{v}bT86>B zj>I=+vY}RS=ZE~9?~!oPv?b3%6Tb2{)O|(JHtO~sUZIN10m<^Lya&(I|FARBJoASs zZ1$Za_!MFxe+HT^8O5zH26AKcA>{G0+l9J`p>AKVHf)J9cU`F>h{O*mHpFB#(U~!8 zRfV(9WKU!pKs|7Vh)o6up>>+jb@M>|p5nOVMsbX!DM;TH^Ll=?C`(FV&JDHV20W&F3%)=>PpfNmb&hPM3}P zLbvm5$5)m8gjmSW5-lfLF*a*n%E{Vaurmp871g|$#`qWDbCJCWihrlD7~byj$Oz4` z0cWvKOvUj|`Cu8F3^l%F+XKP~(Cir0buBKagJDcRke1s%Vh zX0(2~dq6KvH1tbjoT>o(o=P@G)Z+XXWC7`V{c=R@&4|(9t1T#rjY!ABM^NJtp$pAJ z2JFCQ7m+Gu{39};Ym$LMhF}aL`z8fi@VQk3TrU*AV(gRY3v5(`cj-~#$np4P3p`I4 zJCI*+ir*KK5@``_Z#7uIBY$&e_8_6CE9#UgI`4)`OQDi4RP2`DJmv^H@YS_!jZk0S zcJN`tDJb2W5K48|1#qb8haiQIZ6t>ncnvkil`vzpJp8zBF$nz8v38cuPZ4zv;$=e! zKIQZdS%k=PlX-#E^k@*a#h((sSYZox7~o8Vt&w}+6iC#Ezq$5TuY{QG80s;`)`QEq z!;<*P8(~0b%HO_>VUKf{S%U@%eQL(%w1-BXSj$DK_E_3SueKk#FdEU3}QyYiELQvWlb z4&%U!a~=P!)fJp9FrqdO1N+@aAH=FqyJM#xM+B2GI#_6&8Ne89H4oK&1X-^#6&`4c*`FPu~UrEB@cUl%Wp*E$IuuA`IU$_k{$E#Q{-|rpYKVd5@fQnq4%Vf?dd;*pJRphFnwdI1< z&r*FtxmQB&$aT#(foQR)9fJxuyG9LYOk><0$;^9K?8ch0Ng@3^jkeKVf&|dDSyX!G z18x%MZW`Q>oXG0}~ z;P)ta4b(v%#}M8p=Lea8Aloir!OC_221oGzFWYsmN*1j;4lFKIi%*i9yUZf?CzFzo~ik|#zX#ggFHg- z;QTtjW+*4E@wTcRCU9j;kVu-UmfQ-{d5_&#yWPIISs2);th$pL`OVwgh^I%#>z4NR zvrfWf)F#CpUGTU<#Np{ZOe5t-W~`iCZy?a}AFJ_0n|G-bb-a(%M#9xU2UETMO(hBy z?I`9r@h(I1Ek0=?%emRgNA3SBKgYcPjUxsZlysoGZ2+!EZst#4DXH7;Ki5A5lMM`^ zFV*_N*8}%VNDOnRP5;j@$B+p0HmWGXUnzc){A!)^E zVNl`S`6de z5T>9d?kT7g4Jgi5iGpv%Q zLO?K|1V~j*U)^N?a6{XC^7lIzk_{K&mdeCwOyQk)h?giLjWgl1(oAuRcW0s*cC;mR z9jedZ&EYm(ZaO<8)5qq3Q}(g%fW9BFNjj6 z2$p$$Fa6>nU&rrd*1X6?!A)(g*ne=uLl}3+o{CjuI3Zm>Ha^nMQgX*y`f6zpP+t-a zfFTEeWAE6OsUlrNYCaRGS&|HZ)3zBBQRD!|LS);q+PY5>1GGlV@jfgyS>PW0OPW() zFim&wB`J2J>{;AcS%>U%BIh>RP0o4Q>)a39P=rFd-lhR(J}<%`f^jf1g+Y+|Bygj= zU6pbI%*22(-n7s}61}yJvZ%C(`bN)Z^8MZf7%Yf%J*EFvn3m6wDC1MhHs*oiQw6~v z^q*z`|GiDyWku3V%V4{{>48eHjY{lL0TI9B8C*?^)8vXfiI|dl0K`TjcD=QjHVlaW zjbIS&G4L1)$hdY+|TR^RsSZn;e~-@0-Sp1Fn_5kHZnbp#+KPr7BP zEl7M$xb55de+xXE_S!ubeMZ<-u>4nq<7@A?e=Wwnfh6L&cf>F=;x3|* zPrP2|QG23Xb*s-f-m0p;*Q$1UwzV26w#YNs$df0*yF5lLnPpEcsY;RDYwL|)F0iZ} z`Gl0jF#W}NaO%_1l0!JCJ3)kpI@n5;8-s9B7W(-;b$*h}Qvc0FSymf&bS@VfOAiL}7Cy+iA}tR|FIV&i7IiN8tr`a+W+ zkN%tLCcWH0rHWfo9JLykhF*U2f{XcP0IJ3ikEctLxe20}|4F3x9y_zYW^TksieI)? z$;sTyYK@i8QxmZpu74ZJ^F%i7MW#N`T>loEpK?hM7%T6;L%$&rU8Kc=E$EX&MI~9@ zz0EfG_@?M|VZrl)_)Q5X+B>LbVC-sQmKc=%9p(`{%Dc7z7nU0?<5&iY@@diwm-BE%3rk%fBPZ{ z=c-u^?~q4l1jM4(#j4`o0R;Z%k*F?A8)0zqQAv=6iTV81d;tqTsp^ktibSKMj~1UL zk|@Y{%#NH+y35yfq^_b_6`F&ZouCVy3(Za^$!U&T);FtKgNcKRA&rNg-q_$Kc&Jhf zeB1%r2D!1jw(4|gxUalp-CHIjmHdc@$N8pZkbV<5g<#>m0O|y&0wVQBX`gpLxvd2g zn2MS%KW(b0n93L>tkc67XbknH%*WOnjrKR+T;1lgpCqTO#5&UDvsgp$b3;BT z*(NeHb8WUC+oq$`DdcAOFQlU?Tea^&r;Wn*4s9paf2>7Kleu$wk56TV(_dTqIeSvE zRZaeBn-3%po!ok58|0KxJ0JG^n)vDMWf!Ip%40#B!T9?`-9q(&z*DCiqu$2xC70j;6bE=(=mMqz=WK0{Y z-^tW%e$crp+u=zJ#uL1umpEQ6z#Q}~YLwD<9NIc!{yAme)44qr*`M2baLAbWi^6gx z+G!JunfNh3Jh)shk0uHt!=VDmT=)7LT^Ra`SNrVYVnN%5Ol93C8 zLA>NVlRN#dB42D9^vXY<>LBVBYLs3XC>f_Z6a}1heq8+Bd{Fr6*ys8j)u`UkkaRR* zQ26Q~JhDN!AcbG?_ZiQGh_YvMHc(ZUIo>ed>+%IrP5-17qIX5h+Nk;7#O+Ji`tRL5 zbnCp~e(80iUFh7|`)_j}el;mTt*58599>eZUk7=(rwEc5B%zatULLbt!%qTBCpo3g zyBYrl-+K725mnXmf7xkkzlGXZooI{CuTrG;hP-&sSFlDEd3%ca{xfRR&VE^`t$r>| z?BmPjx~2mETuKcEjgDHb!STScyA13ec*79K;OW3+a{Jlbyl!S-zJifQVX0*J& zZ;EPG#(vC`3F!FtyJR!jHtpCC81Ir-pAU-X?gWRFaT4KJ?GUX{!0LZP3T+J4--cRW?(@95H+olU$ zibA%JR-)(Il_WBo_E-J==AT;so#${uyp90bEiAvPWR~OKnCL9--b7ZVC`f&k6Z`YU zf9&mbOsfCO+A8JR#$ZanF)F=aXdN{Ccen5LiEpH7b^zv-(7&^_88P$;`|`-=_3-*& ztfyb1*YShw7MlfA&{@S(Donmi#hO~&{orKl@KEO#Tg-6y3EloxrcQ46Zc00=m; zrBKArn|Gmi?Cm25O1Nf{-DX~T4RL^83%ozBDF5qMt5^LQH7$pyk+5jCe7^mg&@;L8 zxzAyXrvect@zB~bSR_M~s7!g2k;u8>l^LkiG}?A{?8M~z_ZXr(kKS;4tfC~a4nFl4b<)1b&}Ka#|-0d<8nXabF>?dEQ}KDPr&{=2mePdFPA zE_?^eI&w{d8n({$b?DozWP=Bg-)9PVuw-NC?2cb?(!y7%m>*B}&UcSxC?|gzKP$5> z<7b?lAiAW*yAPK!IQBb9CcB50?5)a&2HR?T0f`eycB z9+VL@81*anwD87wiq3-}EuVk=)2swtr;Q(+Bc6jl;yL-fck>IEiuTW0dqsNbqo7Lv zR>kjCyrTn~gT6~Z;9?^qe}Z{%b~4}g-=_8bIB8yYgGVp=V05I{_KPZ7o=e3Ri@hTn zkt(&XGw*7MzIJofkE|o4@BH#{k>zeq`Celh#_k{wovUHG+o7ynytigA{Kb4q_1V;wd&!|FS@E!%!O8D8rdyZ7D?0!F2HVzw zwOlse&AfCPo_YFUvVcP~T|u^cbDV8HaZ}v4PHEuc@l}rO=Fx-87# zT>CydkZX`43`EnFvLKQ50`cjm*1e}cWciea&`D*sbvS9JW%8VybriETcU7JC@H?!$ zPAeZ0OlY}N*PVo_ikBt;t+Ym?T%+(H4`{HPe&;cz;a(^)8?;zJyG%PCLFqqLWzhlE%Ny z?J>1e83=z=ii-J5*$?+y+r9uY4-hP8b0%`IrO3H~ee{aU}0w+|e%w(R5) zN{Jmc%o}g$11pHfEaL-R+bkycGe0_wgXZ``wz(wP-tGV0k+F4}_iPUhum<4Qjb~lB zxx!7A+k56Ie$suTs)qfBdm=ijf}vZdklD?H>2Q+s!UT%}Tu{ z-PJ4)Iu=#g|0BfNM#HiE@5B4Q z7Vxl(Q_;;tn3l>-y)+0LUv6SHCVM-;6Idt?5!%ED^v>$Hp7?w^Cc$=0{=&rM$4OT) zUc;xjkgiS{RZG8Kj(`*D|M62|TI$pJhD`4O%j@kT{MbKbH9Ys=*=Y3iP?{3-q_KYL z*NyA^rQRZIZF#IY4BGkacT?!$L$8Ov`4Hcne?hXvkQS#h%_73vH=W;+H-mh)Q|Txz z1G}~ni?Gm4UTd^%zx>*Me`Q}ftAT_wN1zyxI%>!(!*VyVd)4Le_X~{(C5<^=E(GlB zUGMbdZp|ew$uaWLvsRMEki2pt1eN`OVa-k>$0&(+ zMvpEYQ5J>u_1@_6b8*rW?VE)Qm;cfy8jBNVE9Jj;DM>w*+l9K3 z+j^~B+N!OcO$C=rz-uH{U6sIP%j_(D;E6yuPY&ogK_9qCekies5#tvb;ADf?{$1H< z9th{kiSZlh%=nt~acfJ=_tR_96E^6;kGFH4&MiuGf+y=lt*lVpP#^fhLwhD8)Lq+E z@$t!^>nU!Zne6auF@RfMB9djlUGMAU9B00}=Kbp5#JF}#=PYFQo~On*scFj4=+*#1}_fSZIkC&%8$d*(yNs2lWuffZZV zp`M+uM=fl~BQAdw`?~yP^%%v_EvM`S*&sen^pT*#Te33quJ?!Vf#9I^wxDW_p2~4> z%A1J?r?|}5F(IoKRqE}wmlbQ~8}LFU953a?Ve>;hLs#?T;0B?Avx-2<7Jv7eGlSq? zRMC*fq34Wm_LgJ}ydU2Gs$zq%^X5YFrA%hz37o=$1z#APJs!@x7Wol2Bk=Y;m+`VN z*J0zHlZh+6ci98mr{V64rR_%PO?yk3P0hS@{p(P1%Ep%h{(<}mgGX0(X@5wt8{Otl zdk8Tz8_b$7u^r7p_nZvK&cZ001g>pXSaZ++{0VseK1nkfq$~Td%jN5S#8TBC6#88HrNapx)-=!|OQ^vBT&pTGHCtnwhQ6-v--jL47-+rmymBk(M>rk)NKCw%p_4L%VwgLp?n^~i+?_CaZY-uFB?(VU+Zd*)kF{okKf zu>EYss3j+%Tqp0Wc3*FL?vfvcIKi6u9iKjJr_#TA)zwKm9dXteGC^jTlhU;V^N^qn zZ=6pX87*`DjJ#s+11G^yK|XsV5qZc>?2rzbY_$%S25rE5CuSD%>XAfmmL|XvlRX8g zWSlh?D@`%CCga=E?7@qGYK8pc$}8bSQ`{v)8<0_|UKYrB!0Ng~oCk6+%EN3cT?d=; zGi)|~?oL%Lg4S=SJu1tswcsSTt_q3T2jJYUd79qEGHtV_k;~U8gFTtvth&(aeASZA zXzwhlthQgUQ4qFmjU=fS14V8MBYzp5B!uSj8pfk zD|TZ+lBsrjF}>Ba0nJ03s#WK?yu2>%G0YT9ss27CnFY2J8Sk&jnPcROIDPdWn^9*r+VK z<*%u)0?J9XOwIZ>}n^uV+FvPmEpsJDj_j{AhM}>b;{a-FN8?cG|gHAJyi%>FziB zKL2%hXYeT4u%V3W2~hB{YUjDk2Y~yU$12f4=hoH9EcxjU^zKT?Uht{GLu&bS6nsz5 zTOnDnX{x~aa{dzexx1Gc-Pc0JPyo3wp+*p35aRFmc$7^E&>5^xfU;Mg$^eTqKb>-b zS!WO2l&wpv|9GnPcMR*_`BFRW@M8JdlARUQJ#X<-8@t|Tfk*Tg39683h#7NrpUWP& z&v6@u&d7|#&H$|C7mx8umK3>NVY!`zPTO{s-~?;zJ6SGk9Un%`?T8|@1YqzA31#R%=TmHX}IBd%su%QEy(Nht5 zSq|?Z<@_hvD+l9^Qt_@+{@N0&xLx;nBB1hU`(IcYK%KT@kYnfLqO&45gijNC}WAntfRZBS(mCxcg4+ZW{IX?+fmdVR-v(s1wLMU2 z*U!cq7P4bn(CZhQ@_x1cuCdC!-xt%xmVy*xaO|Sm;+PEUb6)ZY39pu>7_$nwIZS}( zp2?1b4VsSkMpHU|f|UwBtSNT%W3v14Z_#HtO4;EN1O69gGmw19^2nnLXt*gFm1(&L z7QUw1Wahqv<`D#knFlE&6F^gwG#v+PbqY&Mh2t#;YgGC6A)oQw(FT-3648*_izfYD z!Yqg?lBD5vv!IZ1X)Nbh{1xA>c^(?pw$clC$B?687eYVb!a)gaysiHSQiGQo9f`97jh0;vDOD{bEr>y;-Hw7uY@cjOac?xq_JE+~NT zew!tCdebwEksOn=C(t;Bm7#>=$)_^6p-)7Eiwnu{0mWS(8>3S(=lsbtorZ7&XP6(f8l_Fy&5JQCCOzKZjByGE_CG zcA=eG1&}~L>hBQWo|Dshkh*e=f973J)qDuOgwynQa%_$ELBO8jI#ef@KB;5~Q=6SE z37p!KdKhfB9EQa&bx`}fQto*4#@1pd$ea9SsfnXS4wbk4XpsR5X_$#YJXpljk-~Ct zeYF#hhIl;loz7M=3xL_@MnkIcx!+S_{qiBNw0GF?Z(aY!u(QU6Ni8ckAQpMLY7x>? z5c!2HD|JjM&0-&zR9KVx9$+G*)vAUmtu9cq76&Y{m{x_tV44bCp=-OHzT8cg5r08N zmS1v)`EKD(wx%|8t@YE5i?<-Ug3m<Vz7T=^KxrM8J z=?d8chxUOR!I^XB7`5-ua3q3K-^fkMWqvmL()=;F|A|{AQb}{W7j5+}M4*N&bwCnc zx6<@MH5!uc6>4FS;8}qkm}e*c$66$sk%ZD}vR>V^urzAlyuaORcP711eDFmUE?v{r zKFjbbWI170?VHO|eCIG{moO@hHw~hyE7;nF71)K!*lq{uU^hH>6MZ6oZ5r`Ne7}C} ztea_O$@MD*x-hND#1vPM_v0^+&e@Pkzv>VeT9a|kJU|m1d=mv{E&Cvpx1Sh z$^W>|U8BIPOJld`5+sRqb-a>?B3Ni^FbdglRq6j&qmz+t;#s@a$pi&j;1> z*+i;J=S2qf~x_#2qv6z>nNkr1XO8pOC)V{tj#RWL=e{A=n4>D=eHE=fqY~4vaU|uNRi` zM?;W>Dt+K&Q_2UT9=DRt%|a`%q`#-t+ z(@8o<`&&ru!D!uMmcuBBh~$%|>fVpjyYD9A&OV)a<5(Vrqu=~f#%E_FLnhei$A)TS zEY`=HaSW2$rv{5<#TUmuG?9`H&0wD&1!l*i@-Z9CKZ~ zp#P_Z<9P3`sd%%0f?bzSjfpH{Ajtb&mO1ejvW8$IpxR~Mgtz^WwDd_!^K=$wph$fD zL&doB!~r^x({Jnhu0%km;}a=5G{DsdK5V4x-pSCx`dqNjU$#@G!i<|^$cN&J{;|O@ zY9@ETkHMowUd8#(-8n?N+9bt2oe4IM=^?&U6~^Y{%q_&u-}SK5_A5zKuh3fowWPa%>{3?qb9qz~S>4%JW+*-x zb|d9o*cvLVG-`Ku>1Mx)Ek7G!j1`nE#Yt4nqeuCW;(tYA>=J-QEEf$g3VzqnR*Te| z^wzbGh=l;~?`{4AGVrS&f$$8`rD+Y)qC_9e$<1nKOg1%F;2Nv|*z~lCXW9Q(R28lt zkIqQO%%6iuU-VccVdm+nelm{dZ8lt`i_W&=8CrU-0*gh81`|yZdah3^qt_Wn!Q;Z( zR&OC@xf}tnaS=b2$mamsl1V~kzB#{}pXh!l|A@A4vEyLyW6V>ZHzJf?nV9Sh;>|7) z(zT&f6A>OliY*XJK)GIvb=#hGDSr#=Z5x4{Nn5~0>pjCfUhh$VJbc@zJ1TloV)at1 za%^)~8fzG=#_uJR1BpMRZ}&g9FXyXTmuBNQw{NSIXn#>SAvBF%Z$0k{gW`D={1tz` z-8({76qp7{bg;4sE)1ot+C+q!u5MF6P zFxKVk9@RN&H#U2lwX@CRF?CoO@dQX7Id?mgOCM*`@)N9ncBj?^n}pdn_kF|OasgnD zG&6CPvf|tes;t`uHdy60Nzd*5{4x(jBMpvdkdTXQd^x>ys*}~2J0|N;k+1ZlZxb=| zxgYiY{arqt0#Zx;qANitNKXz>%ZGIJn9DL^fRvp5a_%-?BY=?} zs&(Lmgs6sO-*5M}Nq)M1teZOJYSt#-7B>G5U;O!c=?Z9-M+qKfgyQ3!tS)~Q4e>H# zb^h=@B@-AM6`~142Tq)VAi8xx{>dL2Gz@9w(g(hUj$-~p>;P|40LEOhe=fHHpkly0 z$g0s@%Kg)fDagmLHKmS#?Cy#TfEHwh5;#?_&nJFLlkLf7wbgT-oEF z^8RJO1ZJjV#+%*&V|_P*8-V4!!~Cm{7N@Y{FQgnmfHOPY1@U$-j@{GZ^#OipD)mCF zfynFQkhOwwyZ~+_9DUQ*q@pmN34E#eK)?d;#Y*04of_oL|AI_Ury7kwge9L(;f}%k zDflKH*c2jqRMT-r!H04R`j$#h$FoBPA_tWM!13>f(!xUVs!}O;gLfTQ2HLZeC+~Yu zYRJ)I_RA9@ak4>H4=mJlA#5lb5_}SH*?9pKj)0QgCag7|(kg+?LT}JL--a@B*7&-+ zAmq(NSSVbCm$N(IK0wtQpbEIRCN)>Zr-u`EF`}AIThRthFNw;4R))SdrypI(F?7f5 zUMU!J?Z=TV2dB~!KpQ^Z7=~&1!wXhcod2=D42zpFWV<)i+L-)~okt4xgKeI?p@caNQE5E&~_5&>P@?mW`5!3%jHAa+Z4IuXK=++hqany*$ z^^<0v&~vnqDf6;&jS8_+8x9%n%_ed*B|^ z_~`PN*r%fS{8+y2-e@a4YSZoBFNdQ?S}xnYK7Bj)DlW zKKjCh0Xwgxu3Z7@S3cm!<-v9T5_OFL)HV=>ZTWn^g#yXeN>u0qs(GxqIn*Y6 zjd3X=D8!7$VWT0;PmqOCLva9HqM!Bx`mxJ>&68apVDZ*3CjcIn7!GD~;`X=aL?xpY z3u61c(~~q5|2{%@Hl~Ti=vWQ`Q&9h8O~l7QY-63oR~nT%1z{^`sT}ZB6XaFy>>0?k z!6BsN5Y`k8IrQo}H8+4WH|o1db8SDRprfxTPep!2eXPJDjG0?#5E*dM zr*(6j8-N40nQ<4t0QjDvpO2$BUK{$%=k_yxYIBLI*h9{fzYj;qP)#$46`gEglGk@l@gjv!j9YIVv!$g`#^{mJzU$aAIL^-#`D$GV4>9`V2Jed zC;O1E3?1A^NY~PfQoU#(!Mv%){T4R<7M!@=SUi$|L=m!=3{0WBK?9FD=MBa6>0spS7X)3Rw09`T;%Nue7ZB5HGrp zPdgwsYAlqIA#sw$Er19K`SWSFH&nrh`~tdmiTP8JABQ#mxaH+mpAv^fTv1F{wqjk` z43SWU)%BIItf+KR&wwqby&LuT-Ul+_X%Ya>!ySYBFLAo9f0~Rh)6ij6nR=EzHybP$p(2Z{?-UE*OkdZ*kHgxs% zknB2ii$;+W>FQuQ+=`(y-%VPFqPj@J0t0S-n8Ti(@(OvKJQ}7|3Pe8|#LOTTZ~7tv zvLF=zyki`~T4>cjS+J@K$Vo$)q>zv+6GQEDDh${%WV1CMsA4_o<_7Tn-sIwHkbs^R z3-6T%ZXbwGsWAJvvvFg!JY5Un;6-@59Vm5P?&AQ{d)dk2k^*ZA0wu$~b<0Cqo< z9g_@MJ(ZQ~0|zSVhqlYHBdhR+adqcD*{rqOD{{uBc52{uJW(>FvJQQE7{Y!EOUI_u zQ(;-0;05uJ)Dg(**Q1@+RpwtQ^f19`dEy%Y%WjwFXJ%J+3|KVy;dbw8w-1IB_!JGP zx5MP;LX>YHoC0YySg5L_0YAdtb$KLUaWo_+zAK{Njr$gCTNzOWbhI#66<>g%4zHpC z_`ex>e88_%V1-kg0d0`4@2e9R3%G`doR|Qx7-lwNEF2*?hzqXKl_mFUG^A9VSO@US z=%`p#Qm)$kTn~~ZF0tmpfEkERDM)TzU5&zPv54oXc(?E~OP2~I=q*hC3k}8(0biRz zL&j-(i2;KUqTd%#7O$6G)?tJQgLbQjUn_v&Oeepcsf_q#3VDKkw=TcS#s7X)@81#w z6g@smd{U;8>U5ab)J?H1810$PUbrMI6hzqDHt6J_nr`$a0dC;3HK}8!U}$XlKEgN8 z5T`MET=2f{R~MZE+{&E-t-SjH%a-=WDHtXQ*SQb8qF_h7b&ZA?1q*a{g{gYu8jRF- zMe;zd$#oq{Aksq`u?wYw>vIb->ED&J{+z|n!4hVtC7VJ3-?;(S}HBqyxT zwyDV;IPOLm&L<__I|VsNxtQGRUx%8e>ffdiV@Z7YvPU2#Ru%gh1;MF#fiL)#QTt!i_(`zQlMS|_<`c$_L9b`+M zf_y0q5BAn=jGBU^BXAu|CKx>9P@9M`C-s*90N_8{oWjHMZ~Q95(FcdHe*b=xbvk;T z0b%;~SB4u9>cv}&7>6o~3+Qd+dq;YVq(spreUAnXdVd*=_`k72)v#h*AMB-{0{9}; z#Rovaj|HiQ=Mg+&bXhkQQg%%#i^?TV26bRulgC~&i$9IS`Z-I22TzkF!*D6uF&M># zQJy&ki9D_ED@#DZhYKaoefQ)Zx2Wglm+0evVbQV-o;aSj)RF+^E~BHa$N+Ig8z>FzJC`jko$Cn=abNf=WN(7Hb5w*}jf>)tiryz7*^!!7Nvz1fa|J!wo%fB5ot zUVAsU;}i`4Txqf0YZJjbf-9=qI4}B+ib*}&q8RdugCsfU93u0Ka6#bHnE}_K6c)6V zJJ%1iOqFnMK$PX9xF*|JVL!2pGqcVW2GI}{K>n;LOHCJY9&iGtsla*W zI5^XE*m(!ZiF7snCEcR41?6yNm;h`VtI{KXZDyIuMp79}(%L*k!xXh7j-!F920o2r z?QoF`&{GHIu*>zMTC=Kt0YNKloo;oy9Vw7e3A|&$7^783U6E{$O?rhVDjHU|S6GsnyDWBTJS0z@Qek zqWs~f4?rsV8Mff3p%nq3xSZdN%@zmrng+JC0mL6%3(0jE0s(trDA#x}6%GQw#{|`^ zL$9a1QStb|dq@P7C<{aP3r`2TIE+Vf3!fe`Eu|;Ag&*mOSFKiUKu?;0Xh04{?!DEm zKn5TrAdwkT0L(z~nfBv8@Jz_VJ+R=t{KW67K*uk2ZVXs*OlnKN_5!eZRQawNxS!=` zOD3LP4M^Og1T!RObxsWehv}1)-i0$OaVJ|5^`)_S-BEV7}4HA&sE(7{dIwD6J5t)U|037`u~ghS?$m*Vd(fD7l5 zMSQ@(hf+0T9S9kFf&*ld6Io)C+u^TS2BJM2HNf(QPdrZCSfz5^f+JXb162Wj8tnsX zBqTV*JbMGs*>DzW!wP1<{^O4Um+<<1hytqY6nY_Bz&? zluD|qc^&$ajTj}mRLoYf55_yl>R2)C1Fyu2JZ}0j0=YaKlxe+#g(4IuwULm==d$Bq zrN)!{N7`5GP?GnAz_8x9{(tUE8kR?|bjp;qiJtuf4Bn2!*x5elY**q01-93;;nuzQ4}? z8NV1SaU=nev)J_Mc3^*f!W>YT5~}HYSpCoFWzjU|BGmQj9AnT}sEJmxlTbYarpy+9 zf0rLs{$c_Tr*mmVkmbOl>6c-^T*Y1GHK=^ESe}i24dy3H%!{Q~Uvxaj6UGh&Hmy!@ zamjY=8GmX#eQWp#ec?MAnarp*1LmAtdQj6ND@GH{YDvVKQc-a%XCq!gp(Dr79x4>9 z01Y3C)`((eu3#T4*w@WobO^|5rA+Xx!S-_AO&LH!vOI{>BouX>0&FO#-mYKgKc7gO zI8Wp*Plq<1(h~qt&WSn_bpKg66Qq*+>%=LrYAl8W;nmlmF$YL3R*DGWfhU{oioGek zAc~pRnF48c{c08@p~&c)W*5>ZmfoV7qZ50c(3W@|9kn&)4HF9Zk9n);n&g<_hoH=06lIaloy zkhnQE4NzA%i&2MZBKq4Qx0YH^3{G%knE!#{AX8K0%|KuTD2=D`C+e)zjkMI z=pCMK_IOP7qtB($9gq`gx4RBpU4Fnj(QsS;tg@|>8%O%2?tTsESEuR=t{&&Q(pL#> z$ds)D4iMRh-fXr3yAZoX6JxOY4X+*~6r_pKXUx(2PPW;%z)!8rWJVEFggP0{O=OEv zRYQXJ*s1XS@poP73~jhIu-HqjZobJVxh3x{-(45uDv0Yp1|+E1VGEHM5BkxE7JRB1px$5HO%E-TQXSIW{UbJt-Nw z1?4OVIKww!X+ROg4qmD7A-mi~_o)k$U1}+|!@&73zCTSb?YcB~%#49-hYE{-AGX7q z-h1S>@XC#OCNx7h-BS{oFy~t6qYc%%VuauR{CD`fxs}ZzoT9QMhGBBk5l39 zQY9h7z*!C1Q;5)mpRaQ}-xqn(n>hn7GK~EX*N%k4Dp1>mO5UCw1KtB(LZ)l5*MRKv zpz)2_z~%}xlm@PTMysUU(n~t}aY8?C470Q-uDBN^~Bwf7Aboc!mifX#G?lf%1FOEh~X)&8x2(&mb8x zdf}$cJRomfo~Ex4*$Xzv#cX>sCVyWQd3&3${X!*hP`PhJXSjPDri{38nS|Q4-MX&N z)($y&zlqv6fO2f#T;!l4w|Nv(h5@OtJCFp#0%7;xtgpca8;I(o?J#<1{)J&+CLv&C z0x*Bk^0hXhQJ*vN>|ymp!3Ykxv8Uc#1aZxe6a2|yE2zV5HhKj82sQ@HK|jLb^lSEiD2ZX<^e&^!F66Z@ibMSCg!fVq|CNo| zKcacPXG|z4F&7u*xH@X}N{8N!a`p|()(#^C;ZY3Ahy4}p+1`ToM*Taac#}XFOfzK# zxCMtMyZ^kjhg36}tf_#elK7*;>R7jYD3A`*tAwWh;?f6t0jqu?Q!f&<{$E^l3JIn# zCw>A#cEdapL#DMP0&j7rhRt$AwPHPmJ8MvFp^G8~aF+MdeO$)`qHf`-q5L8KYK@bHxkZc*P(98M>K5-bTFPvxl~>_^^Si^}!mX zUJ&+&!p}<^rEh*1`p)<0CvC4`fLH40xfN6m_KWeJE)n{7VTidk4(hQTPG;m8Bmc@B zIRI=fgjf>cSfm!)-l+s?b^kVRXnrseCcUGMqEwK?yG4m`n=#}~Y0gCIL4#_ya3!M# z(t$cD1tL84&1r_&TMr+(_6{=n1o-#=mLye6W z&YS&jv5sj?4~C+8GBhT6xEc{;7GN*{YofkTA7Z{-1k$-h89e@2>^?plK4LE(Sxo&WOCV9bR?MiUXFkj!9ccwr9{bo1tNjw^(6d|eccgS_>h`i;qFOyt79# zC%`INL-jc}@-|#yPo1Fv7KSd0ve5!POIKlQkS%;^0<#IN-?lTEz61?epaCcs&{T1< zg|iRLp)9vwYNwN3G!m~|B47|I6FA;>7}jyCKLywbi&0CZNfKsRpd!$(%J~4gMbJ=E zKXhNm*mieilUMn{6g#Dcwi7k@+}F^S!h$nETX`QTG*k zWm)l~8v=YMwDKFsQjg*qtVR&Us-;Ky2=Op9IsaNt=wexYp+Rt77GgM^gP-p{LVjOd zf4&lrxGa}rJ$-(sWE*mHdqT3)du~T;j}>ezJ zT`b{<^`&>{I4YP{)H_?t#H%gVkSOQF2OZV}5anuRN!e&SzM42OZ`fx+oto6a=VhW^ zVV$6S+xtc99Oudy!MbG^A8#A?rC5Kk$%e&!IyZz zCzfsj&yX{ZIJ%rL#=Omek=2&^#l&LnUW$5=S_|AxX4*Id0*f;V$^Ixp2&j}BlBjfOh za?PuyHDq%gr5B^t6-i$Uo%v^r?G>(82}<2E)W9LORB_M4AGMAC%4#9SoL}$SG(afu z)n6pfiAc@Xt2p{Ig_vsBN!5IlLX*0AMMLZni-xj~((BNTg{sCr@pck;jsqTxq@| zky+El(X`;P#jL*wta8~CZNY@I>4zm`r};fYznyPMe*B!{H#l>~PhGfr zD)Li*+IJRI8r~M|df{_LbjjT3F&l2g`GOaOYb{?#5PE#CzP`rG@oJneuENc$@()_- zoz!SL{Vus8$Q;>KYvJB{yB-}PgH!f9|868NrYCvU4nec%f0BFY&yo<*()xAQz1n-z zT}~qnGtX_#StHW55A4d-P>$H0viou95PONBdr7%WgesQM_OT1VwxJO(uJF1=bZzSt zVa2(|q->aRU4n0jYlW%yfWfI-urkq_w0?g98}7ifx2HcmX(f9g#T3vL(FwQaZ7d?RlK5X=17-d@sG#S^fubt1?pFQKMTIS z;f1>1*Vo+g$$!MiA|v-V3499Q|2^KD#-rzk7QjU}jFt8<1d- zt|wu#EOX|c%qO1Dg4+%IW2xP+##2iygs$jwek10{39Vh-5yVCZ_nM16d?M+@w^^<- z(GPS^!upCtKf+f#6CF}cuG{Y9#CZ4l5H2{{TqMH3*lh+cMdO-b}D%+4TO zu2uV#t8KqzHRAq>*U~8%>wGe)ZqQJ;XETV|;FG@UkV}3&rl@<&w{ZrR zqeLPjffMMEh?j)ek-XtwqqpBzK4By_#`iToaPj5N6xVN@4Yxp9rbo=V@aaG33=*D@m`a;R*dx*~!YYL4sniU0cu1 z8?HQBaJxosW8zfcYnUNV@$wt+Cs}I$PC3J9SahR)=$U}564D#Vp?*GQU)t!M$halc zO^>H1XHxLCE2m-z!`x!Z5v632zzm<(Kl5e!24_%9_~zrd*gJ$S)oIj-6P%K*7Mw_B}O>#%LV>&=_K+RJlJ z%;Xj&@UuD8Csi_Y?mngy~9)$SBGAeIb41a;UVt->&;x88HgKojd?;u zX6L+doU(62D4rz$m+q=v&r_~O%EmWa+dgdlz2!yoBWSR`xm6*!kpsy7cb=fYdHbWQ zeBT%|wmkS8!Wd^^-ZN^kH~DG0`f;&(+*=EzD)=q6z()nN_Br7-Jn0F(&};Tm)~ohw z?djl5x~JgU`mpXI|Kri4eI|>>?&du14@VU%>z#_xDTwzy%4@7!LghWb{+TWdk!T-e zv}d&WrVUC2%68%ZeM+2e*7#Wek)7Ui<@r2QiA}u9SLdZ=Q(MmW&&o3FjR+4Gjh$P= z+pY}*N`GuF-Ed{pRa305NtUQs<-5a@b(o>|_1#Q#VysN-dhyn|vWojHZl>CXdvZ6< zOIBm75Y`A=P#C>RDa0Wf(U*l*QxeBI6TFsh*KYA016{p30~72$);$c9{F82}RfbBwE+G&N}li>U+^@wXgMUIflgeFZ*kuLkh=bGiK+hmtE7` zWvFYLNq3)>D<|>qb~wb28DuM!ed(_7SXTrdvHF(xC@1#ak~Y(SeT^9610({k@)$@+iYoN}b=ic2r&5R-lJoNnM5F@yDQ zc!cK_bAQ&GSq~muf___zH0)Sw!(&r@_+4vTG8H1iQDwT2HN&7q_?28;8^(W_@`=zZ z2$v7-xBn?!AfNIPipMt#@IiM858eSv&62Vg$tmxoZ$)tZCU-oF2trx5#3tR%=u}Sh zyhs`xHuvY#j8m`oGj+pLuJXYF@B{Wq@Gt1lN%VN>>L#w|=+*bs05D%XQ zj5=WL)vL<4KHjODd?Eat{Oa%Pk;TxHm*;a~AKxwdnZxcFZM0`OaYCmc!(aI3KZwQh zaU1SX)2^*)@0;dGvI;xa$U%X6b};spB)Y=BuYO>bh{Pkd;%O(X5q}>em8)*^)Q>S6 z2#dyD)jlnWcvn6AS{6?3R=NRHikcN}qHVX3#KGe<|z^Ugr!xDFbP^ zO{Vx=Maa`H}8NK|0(^uQ`pPtKANfIjaWX|lnpf(n-u7(x{sb6SCAh`NZ$ zo?S&*_){e85d(sQ(rD64o&Rz5PO-EL0mjAB3yjwm|3^b5J3o@k&zx6B_?|yq&1fOY zd1Rdl_x1FkD6&4h;8^?|O08u`t*fF9X8ZWcmrDIPRLs_3(Cj!OR>*7x^FH-!{R$yVYPpT)`-P6Eza#aDqPcqyCqWQbA>$` z;i)hZpUb2>j0kyijRQ=jVjXr~{g_19`L*tzg`4`dmQ=jFS!`4h zP(?kUm0r*$o)t7B#NI@h$+hOZK4Xngj`_y+$<8n{Tl3?%E$u16 zGs80@5A!?q*w-(FO(G^*3`= zgm5kULI0QAT5?A#v{$Efmr~2sUU*nOX>|;ErQUH29pW<6tvZL&4!8P94toDmmtoy& z-KX@t&#a=Jv{)W~+LM|H`rDIv+Ukxei=-Z8-yZf=h*hZ*q50>}l+Kg&Q!bY%J%FdL zP4;=yMHK5?mP8HtWKNN`2^A@*MM9SMFe|fT9@iXERxnJE?!s=S{XVy;FG@KEI&w@sSkl1kZFF>{0OQ9iAC z#D6EhFt#9X*!g1#738=*^TAC1hrs@F#ecXvl@d14H}UE!l(~P4_96F|#F4yPn5#ck z(*R1Y!sm+h`kDG?ySay!US*KGxA!We zQtaZN&fAV4`u&aNF099ZyyFh)aR(+2u|aO*yPyXlR0rK)+JSu0Dp#{wehjFd^`tW|L{{o5?%@`Jh9%)l*g=`@X%) zgSd2c#QL;RvYt;XfmBkTobN$sdSP0;e;mqaEPLtPeH$Ty$jMgB5M7-axto29&@_9Y zO3)nXo%wKz!2_kM+YD-EAN*Or4@d4u9)*+Aatx5^SG9-o@zCC2DVqv-#nTEm49Jqb z;n5CcOmXPMW_l;l3GE_AEn`5&5LVu^a^h`yyAM)xk=9T!vdhXk?4#or|`Q z)^~coqa3iYiCi0M=EzvE^8~tnreNV%dLBs;;y{V+VBk%w*YCL?}-2ZQX#Rkmw@2G#P{=M_ieMYC5 z?P5uHpujWM>bZJ^CWYj}rOU>Ct@k# zY5`S5>h`h91(nC67M_vzYo>nR>l?{aDl-<65h?G<)3mphrU2BxOZ<9Ti1qV5;-Msu zY4#Bs7AF#Y33~kMb!ELX&bFOIgVADs)RO*_d%QagAnwg)4(eJD18EC6sSGEg5R-<& zLs!Wt$GcF#`rut08xs}YtHElR10n7}V%alJybUa&B;C&wf>bQ*YHQP0^*U^KjcE<$ z)2aj)W1+A1dx3lo+oCx?;j(<_5UuCEXJ~-|exFzjeJMR@z9tRMSC)+#Luw@GTS~+n z(17TBC3DzQU|0R&Pdn_)TVW;WZd8B)6)-LrZN1Z9;AGbUwHTf#Wz3orr$=m{EBnj1 z^l$2jhj#4T87!dpuONqQd49UmFk)f_9WsS)wh9r4`14LfZx;f}pq@W_>0qtpiDDAP zL{$4wC!uFD8sF@Xj!m#;te)X(LIu(~5*SooFp4ZhXEGqJ6`3x?TLv|;yT87wz4cGm z0hDWJjQeKKb1}n+<%7o2wNYHU|Ic-Pt9T1J`^mEC~k6DMOG!*dkBq>s9&;yC-X%QeGd)<$;W z-%S+iSH^!`d-zmuuVbSn_eC==J$eqsC@Ik($<4K>MK^Mq3Ag3qIOkz?U_mJ31^kAS zXVKFT1vM2lfj<(@RjZ%s7htOge;#9o_Rd)#~^3h;w!*;W+k8vmVng8Zzt!Gd>+2GUZi7`T zSkZ5RiU|Q_U~Sen6Dle@Jb<)+umdkU$4(pdc%W&b@LOr<98#9hSQNSX-I~6C%Rpw;4~>hHTm5-oHmzJ_Ux1xV1|J@Z=i97}ELKS}NJ4 z%W-XOxzw)%s@19||J`;IS;CyWX!;^e03!%ZaI(uic_K`VXR!^ct^6qMYvPY` z;~Ck&m=8c8zB$;)H;M#Vu7q%btD{f+HY%qKDX5C`KJ}GAHLrDtW+^@p>^KeD(M`rM z+bDLrwsstUYCd5-P2f6Cj zQJ0`Ai@H0oHb^=3e}9BhexXO4i2WcX674}hvQO6AT`P~ntjQb)R|9|jR98ndPkP1G znfU*rR__P@P=B5t-rgrDJK$B2B!bb^Ebs~ zW(5~%NN^9T1=ATISG~5cQZvo~Vdyl>)2R<3WQ~wK^23)01*XubwrWr(8A4>6+qwuUyT4ixOzs2A<;fVmUEt53TSWkw-7eT+X!WyJA%85MY}yboK=)mLF$g=lG6TY71Z31I<`aYr`Wj*~qMw^fqWz@TNR1d?@z`}E03?mvB zixrZcssA7|TSf@oa`Ay>{UGd@P_ZY3nrUg_1h1THs{9~9g)*=y!12|j+#q{Mr*A%w zRx4L)TOedK4%rZzhRQxQe)Lpa!Z;8i#`+O?t{f(CuWy{?vc5>fuG7YYg+yCV&{uzo6!Zcbxj6vkd?;yYsm?@! zVWhwGaOfG1KsQS(8Rl#SwKW(Nw0g;5d+YjedUMbDPN+2GiD=#qj(IOOCAyUDyuADU z*U;)17o>m`Jy+OO?>WOKvD{0#dTG~1NL2foxynW+*+rQd!;nqQuLA%c4VintwCJz# zfv1JmVau^K)Jf>3jfnPb;(!FRFCBRO**7$MPi}nNioXUKR-0%DiFJz-v2O(4YR{om zgS{d(5f49H_wOxj^xs?zct&cQe@0ND0}E*`!FL4kW9fcQ>+&b5(DnUS0G^6Hq^?Ej zwFV1Ofo{e9J10ZY&0le8V%zJ9ol%xy1b4&UR%HW|>%SZ@jMzGd6}KZnlQZ;Ms3PaO zv$)fBzw_@;tq0fl}3?*+oox{VZ$(27Z#p??HT##U; z?&FvZFhrOIxQ>1GJDi(@vbZ^bN|0TfVbV;Y!Fr zb}Bcs8^gb4csw&EOzR}kgZRGkQ2*aE`1)-i&VeY|=q{WUV~M#y&@cFFa$Tvk+MaGD zFgt_IJzYwEdcP+6uC+aMSC@+Nq}x^P$MG#g<6v0XBWmZd5})73b>wUb>jzFeXqRod zB{u9U66%iB-TZe2Zsqln$R zStJXR=W58>)w)a>9kaz3=mtUOXQOBqa4n_Mw=Zsz;wDa_jeoICQpfEgdt7LZWuwWF z_hi^Vm`zgaFICM~_GVF|S-RTW%MRi3tZh4QK;^FIyu~dac=59c2jwCx+QqD5*@Z9h zwJHn2f8u!ySYj-UcDZ(~??iOX3lg?*qq-a{is`QYwHb@ixfJ@4qpo?k)PJLe-!oq@yLf5^Ky>P*(G-&@HetWQC$4uQMu z(&hM4V1c66Hv6Ie2iBtdg##TYr>R(x?@2pyDqmL-X;OmR%yuew7RilDr@~*{70P)3 z(Bfr%jTw}%?}M{|GRsK(f*D`j6hZ_xsJ*|-KkWiTG?=PBfgi&^Pe(IhB0UrwT?Kep&z6sRki5(OArpm1%tM`7+tAc!gd6IUTJVvk*`GK)>s0rQGT; z`JQNK{e8OE+&E*x2fbbY0YL33l$p{mM)`Puryqn5lJ1cbcruY3`%W?0hhg(r0Sw_WEp>YAPTa|meY?#5nvc^+H{P6=N1~NIbmBX{u@U{aK+D>byl{RsulR)yUx=B+=u zq4AM7v~JxZachJVQIk)tr33_N-XC|9_v9kKDM~)^>+~B+u(A=(uls!1MDsphpzUSN zx?g2$l5frXDO2)~X0VEsk~{5XNs@qv$$5Hy(QK|(MG^G;0^iLkAXGN-@ne1e*MQbw zN%hxCFMP1^69>)vG_8N}ZdYunJnh^DYLwyL=`aXEg*Bcxh)&s^Iu+5OZ30y z!rDh9x}wse*WoGYr#WLAFq z)!9c8+hr#PIH?M#eQEQ(J^zB9(@*6ugJDaro|=*;%KP5j2ts@+@A(%z`l-255&WjM zP&-DQp}ogMIf)uHLNM z?Ff}yXxsXbaDxcn>AeXmQ$}W3pq$H;8eeExxa4R-{}5XQb-ZMDOrA&a(3zhZ5V*T~ z2jMYzv2&{PL9iBX ziV3V_Zf#W@auqwVQMHDl_04hHdeOYYTiTDcg)>1yVDfZM5|C7_Nohnz##H~-zds9a z^x@4)GZw>N(qf*|M{f&B>RqP32qU?~>vWmu0uRdh&a||1K|jL{Z;>u)Sdx%NSAJV% z05+)Y&uyk3VQz0P#c7fzI$#$!SdNv_vz1y@n9|xX5Bzi_;{g3+m3NU1G8e*-RBpA4 zM_iGomQmM@W#eLPTLd<^-#UIp`;g$lC4vu^OXz}o=)Nl0b<5lI)byIKYJO0x;4O6VD`{Q>0aPJ92DIjVmqvg^;{0FuaDEngS00O%`W@)dWoZ z#TsK6FS_eZ1&vpyow=JhWc|J4-V_GgYckEBq99-8h8ksYbQ(vA`IxVtG=ha}All)H%c0gQ5Stavjw1 zl`RZe1-nlYY)jjBIe%uev$_}uduq--CMJL5r^$PeGy2A3ZDYVHsNFyw9Fm1%8bxnI z()SFZE;GMpX5&-??jI+Z@?@YJi8gZgnV_5-MEB+l=qlj8)XsqIn>(=T?bdg zWAN^f4T-?)LCh_xhRqLbUv^ch{j@c{B z)Y5*%45lbVt%j##?Cs87B$t!Bvj!w?=8H~?4EU7z|4QGPb*%o7Ek%v1RM#~> zP`zjPsZp5YL}Yz~mhA>CkEk;R1dK|BPaqy=taZZT-^^9aVX~d>5xKvCSoW*e|XAnWu9_vo6pT6qqYyHs!IVZwVG`K<0jExkWq zVZKng$f8(y7ImBdTBx5#BqUVm!}uBs^e>es!X&%h`!FWAxWlhRY5d+OG2zCy@( z9_(a{&?}Q+!C=@o=Q-4*^^(#0B+>ucRF0uD5iT$K^r=fGBnq$+u|}*GmD<8|Rpy$L zORUDQHq5vKeV9dQ>7%g_R%oF+d&3uV0mk>I9G z@RuJpkT%H0pj1}$uXV$<>Ix-s_g9r*nBds619SQTMUoH7knETqqhPO#avH*r*{4NW zh};Ipu8Pc|vc)dSp1DV71CZW1dPJYL5zWOd5E$-T})SCj0!Nv zjpmKVKEp8){a00XV%lMSU6@vvzwW>TPWl-a#nZ2V=EO5W$0?1x?hBoZvpV;tv*-0sc=gFV6X$QLJ$ zuta66chF3SVc6>=OGr*2eobmg;ktxA3y!jFU&o^*K#@i&JuBoNc2Qf79vlN?m-obO z9Wm4X+48eI@_i&|TCAU{0qMQJqT zXL`OBS)8FPOBfJWCg`D>8 zdrzxst?Rkx`Q|Wow@zOpVQNB(v3rXPJ83XaT!Ir0d-h=L7$vSUtq&!;F1lm4^lfYZ zm)U{(#(Z8Z6>bH3jmx$HZyqz$<*}U~PPBAGVQOi2d9N``fFbVtq|OqYzm~AO8Nv06 zZMVW)v7yRYm*gr6xsp*;(Tfor8!)+ZCa$l4AchK#(cA}p&KV@U7w3)e!NrTyQ>_{{ zkTr@7spW?g3Q_KcbzoWpPe@e;@g_vBaD3-Kwz4INj#rfDKJ>8(ZTxlg=u}CSOt}v! z@I*OW>MU(hHdO!1xkO-!S)5vmIxKdg@jgEXqK)LfR%V?FHac{WSRg_~`9%4%0|KSIuqb9wd#HV#k$9vu_;%$OTi>*_P!s_)EPX`O5@R>!T* zMs!j{l~~o}NthLtd4i!X5cC!k~7s7 zQef~8sO6`+-mYLST%OysjOu4>NTkf{EjC9-7+OI>7aVPN1!XQcX78Xb`m@2nNh%Yx z>>dRms(X};0mB*j)BRI%RIK@&4AH+jcvQd6hF1vsjog0VrwAt0aPH0uPfVT0i=}qn zJx+5>N_$QBmbnAI=HKz^`H)Zodv2q7Qg2vnPzYWETfo@8*#)-v`hn9>mPo`L<{-_8 zH(kbH2QCMIe2b`Gzg)gX0S>1-?FNXzRfkS&B{V=CKIH6x5e zr&|J!B2>&jvlYRno>+S*(cT^SYk1kI%yyl8t{$02!6{XiaB39 zYIstDf|(#SvZ2Big_gIA(@P4(r}5+EI4oDbca#1RH+SmXj0ing6w)??ZNE=kq&pJ* z4`0IG4zhz`1%xvZ%O^KsH)}rde}?rhVk!>vl)t?$xdROL9HD>v-T}tmuYHxPgNXli z-y{{jYlE%s3_OK_z6Z12upA=F7dr26+&fEifAWRK;1&3aY2n9q6p*#2o{Q0N?#9uo z`uSyQH8)_jha2TIl)ua6Xdr96h8VpS%pJCtq?|_+v{(R?e(L2nGgj@^7=TO!t(kV{pJvhM=`EBeS5jEL|Vga!~f@$`Z&_Cxw z%c?dQ2HoF>I!m)Pq`KT7VyAg?Qw6D|V$wgZu42}Z8}O;ot1*!C6YeLk!OHdfy5|}5 z?&0si>S(2{H_$+s{6-WN`+8_0fsM8|{f`F>Qx4^IV4(*~-{E-&PL*gGtSAtZe8G;s z9~Wr{&)1QZE=%xuXxtmd-Omsi9jx%_ByskL;YO+4oy_AdCaFZA4on~WtG8(l&2uDj zAt5J@`c))V0!Uv=Qs0eCqz#JDahRW{j#Ep6I}Iq?%XqP!q;_Ci!`QIp^=-)h-jO0e zkCT>B%7!dOq1J3h>tEtt&=?mx6Coy+0dwX76=|q*+@3+NA!o9>(jh0DX}=N@b?Nxf z+)G3e5WPg8d+*x#TD4Z4xY%yb?4vf`vctS;qWmd3E-TVUQ5e#e_N==kMs*3x-aBbS zQ`UFb-hlm$s84dx14x@H;MFyxGv=C(DUcxzv=f1Qp}j6ps9Bv6an?`k+A!qMr+fm0 zH%>CbcK<|>5WPNK$$*Oedu7+H6RNVW1S^~RP{+ZG2w8a;7+P*557w?Av5Op2yJvRQ z1miL>MTdE<_?@I!G-fCrR>%RhhBzqk9EU=Q_fm0h-T}K4zz842WxBhnkdYR>?05DtVU^rtu*_7NTXxe zlmO#@S*l1|EeLP%11a$Eo?%)QZ|jAz$GJ=XqBK3$5!ztkPR1K5wzrfawmposHZYAz z)jM-A`$o3Aj)BIQ5FLX;1eG~=SN%VfLU%vhvY~*T{lL?`+$2kIWm9P= zVpwea3Oo|TO1B74BdOTXnRl-MFK29s!YH&c@X}q7i1ogCJ9XqWP?!>>$prl$(bgRg z^&kJ~xUpWl6dzdwHe+~@A|e!oAT&-*o=E1hO%aCFCP z{JrntIPE9RmBAmr-(Yl|+Iahn1i(sdtxJOeyu(J>n|Kk5ffN&znXT{zQkI?>Um)Yt zmnB5#sVn+nL3!JR7g?TKxVaT>>S)R0H7|wgZ{gutir{zQfv<(tvN9 z3;bkYQ1hYbp~8ooXyE#>ib^j=0Zkei0FuZ)N*YF_TNl{}Lhc4-%c?J^{=IA9du^144xS-F=_ul`_Cc1M zk>UbsQA2X^S^|&_ep*n30`JfRNO<2?dCom4D{%SMa*~QW3pm<0aP{bRw;mZ!)6c2c z!GF~rQD6kKGYCP)2WC*LlIRrm4aZ{=G*`7$wT!B9>sy=csrBW?qjTLd z+k|s^NAWMAfNk$3mVpi+x+7<}b5AN;83~RWqX^)YY{unRhFcL&p|kCm-t6a$12?bb zZsJ>B|GK6NYz3nb`?ZBb?D1HzvqP;Fv10|AEr?l=rN9{Y2DSrpGG%9F@4vAbqwMF< ztL4N3NXHuPayWHv_HF_6Q`Zd?-1cti%UTj)ROZJ8TYcFsz^kk8(W2w<`c}guzll8h z^Wv#i!4k|Qg1#a?BlcIa77gkk#KeQa=*J8su054O`_eZKc|-^?e%NN5r(qDsy%^r& z%4a|DhUUJvb?qRqvA!ZI=+D5<<(6M>egfe`2igv^^kfzC+_VClmMVCC9*nJ)E7QFI zZXKiV!O28nVTi^L#=@uAysd<(vHzx~4SP{XA42NBSy zvXgvFr@YC$#s&5DoaCnO{fm8&qeb1ABoyXE4NC#mY`@eO9ow#nAX%Wc=30s=gRtsU zk|c>3OZf|Kk6SZ=z16svK1!entj|>Qqm%#T(hBh!PtZ&=|P>v0! zXqY8EVC`Y%yQ$Gven-F2?2TAg!W~|)(6^d1iV`7uJmOl1E4U@Vb=L`ssnvr?56`_a zvR}ghUb3{}8AdY9pm-J8ASx0#NN9|T+^k+^SnTyUq@{d)YOA_98uh)6GzD-SSz}5cKtGY|_ z+$ZIKr@sKgiFF9-3aQ*v6#y?ItTmx(8M5P?RJ0@se|Tj`79d``hbZs{(`82P9gaN9 zP09rC++>5=rbU0#XrmW7ck$OUcwB(~^yvp`LumC=N?=M8?17Av|P zDx~kBw+Siq?pOvi%3_2I>zsD*jn~QenA0@OmgMP_i`MrEw0_8q#(C=z*hRS8kN?qc z%C>|5(&8s~^}-rPz%^{~%g{DK{$u}QPT}_0<5SeppIxV_;PgNf+j5K7(5D5@$6tZ? z`zE7vJ9wyY9b*N(p)&Xk2k@&(&{u41P$6(={spop?>BID)PdtGLF5mH{_QZ^6g-1O zo!ap}8I7hQrDjG<*}$&s3^mRzDb&x?y$sOa3*R&#VMYPPQvP|2Q3L`ey-g_j{`$WO zYRr{ZPN%=I1|`?$38Nr+6_b(Sj-ZE_g!}Fy-CoQI={0S40Zlj z(OauOi$1Kuu_J{VcH?4nqPX{h9zcJ3;^8;}zpRq-7Y?mVF|@fgr!EeP=rfGVNQEN4 zgth9cKnk%Xvelr#ypk#~cudn!6`F5h@U;|PppXE%TKH@pW2{nbB$T*YZ0h%T{i}Mo z3TTK_~#5qo8RwnJ82F)g? zIEL}!cWlLD`@IdW7hqi&TEMkx3fuFt9NKBj@M!z~+ryKJnK}hs*~w7XrsMs4kCUa| zfSP`CR3!57toDlzKz$DF2EJ$De#BTgbF2zG^r3hsj+XkS8b{wt0D9TO*t3AU&|BQQ zQv2FDXF-sWCywp812?Ml5S_!5=p7h>OoN!~{iWRpX7pkU8O(5!uw2jv~xx%K#WmOCj=d@NY#uIq_5v;*&r z#^vuu%=E4OtEETid(CX1l9^X{HIkNdK2|r}_ysW6>8&*nWvX8*Ao%t|fkLB~2Or(V z=FnF@taDK2RJ1EYwBRenbMJ(?Tb%AsJ!2^7hbLz6%-@5l8v)vAS8!DP68CfPrJ7&y zI0^a_Ztlep+P8|i!M8fco#}g$h!W1eLQlPi-VCLzl=z5~+Z!quOO2@WcA!_nH*ss1qBQfS_z? z61Nkm)681>2AbCg?;8u67#0y>F`Y{C+l9&yK%2n%dLa+&$gW&{3xM%+zZp3>z zKK=rG?rxhqri3H}={K`Q#pX4i{!6+}!cASb$GQIP`-czyW^Sv%KNT(G;yZ#vCcx;) z`N}xsFm1MI@2)^agnWwj(=&R;=nP@us>BzMvU>uInAAQZ*Oyp>I%;S9PQ#-F`IZC5 zbKmO4ZHL(F8%3mv<0v?;f9n~1#9wc|j$Lnx+y2)^&%`7Z1HmXVVyV2nPwE|S4tr??xxo@s_mi;tVL-@PJj9Px^c!Xz13 z%oT#4uUOJ>x`t^U!)D8nYE# z*4j^+-AUYcq~CN7L1V=gRm#U*$@u8Xd(-+aO~&D?#VAQ%&+w)jw@%y#^qjpqg|iVZI&AK_Sk3n$&PR~}uAw2DR*~5m!gGlD z{hs1`0e5$qxCYC(o-%7Zynnm#W&F1^-Ghv3YL^fDwy$B7S$UkLqRdV6-xEa#=|!g9 zufRn#+8<| z_2P}(G62j?)f)x5Yp}Dx|V^+wv4U&*pV3Ql(Cv+ z%Kc2}xPyi^+wnp<4arbHiLaFLK8M?H&COc|dBmxlIhpK(KipJ%Q13mtj{8bXZkf<4 zEp6P7Siu|`c{H}ye5Q-OOxP&x=2{cp{b&=gKOQPTfjw~RzjhC{T@Md~am-o8_DrhyRm<6?biXm9F_+ZX$9 zXx*d+-S2IO#%0z7w+e~Zi59LY-Y~r%CLd!T=rmj?{(M~0>%>*}YhPyxzi;ksAIIy} zUAiB1+w@=PX6O6GgS{5LVQT1hFr_1XAPARu>%cg766U0%7~X1rG`tbf{jGO{IokY1 zElsf6&r6;DwzR}NyC4Prvs{r$^mQWHr6?otxYSR(5lh&hDQv`mV^k4Q#WT(?B1YHY z63F7hJ>sYw*wc(Ci1ZP)o^QFV{V2@~(a*pOW)0R?W#Qwn9|Hy)>ke1YSq2|KLd%AkLY=d_nydF2@dvZ^ffM`J3$tR`rX__1nClk4{&^p@- zQoU*&R%?krbN@V{>LmcBw_KHvKYgw@pJ={)3!T*RF3gO?(xxe&N{*J0F5}80_@8&vm31W3jQIQvmBqWqIZ(pVU$Q%T&9FINV3tpz7rk8z=0~ z$Bk6Lwc{&)`e-~)J_ z_$F>UVRlCpDY?|4YT|1YvBpg&v?96z!>+{V_PSE5X2!X)icbh0Ys~IPn8q1v-jVty zV7VHXyjyA*(8N?CXO4-u5Q6TiCkAmpV;&dI%;fq9W_pSmn8VIC>%Q@-+<2^}$D1L3 zNl6KjQtM@XM9hvb*iUAn$7q!rB9I+baUV}36~FrPS~x3fNd9F$lrVXAWeldg{_Io- z$`vO4LcG5$PWPX-QSNnvj^x`+;@kxaCR*>|z6Y_UT>hGoY9xG+?dzi@#9@usdc5($ z7upzMb*ts3e;!K+k8!gaVG!`49&f2m<}1BiQQP5S;_xbyc#RVKV%uqxEBTQLLxxJn zI}&58M7fHHjkQ837j8jJQX4{&q-2D>niOkl)HvWHqamo;3+uD_&CES<{qx01uEAGl zjFM1er|DesY+^raH(7?+mzlQu<&*BW)=~8&MdR@ah>;9_Ef!?S)Cf0Ghs3n892jx= zIYQfU%o!<$T*PQLL?leYXw3lJD7O;KH$6V;#QhveLHRjG}R)PBUVW9ncDUrL?1I~&VkT&DJZ1P1;2gu-rfr<~)lji2x6fJIz&nMd_MHSCxt;mb{RNla32taruf z1-D7>m2}FO@yZQ z*J6vi6v&|^5eZKkc|EmtNt8Qj|Jb$@&7wTmFJ*jP7+_ zNd!c#FFh9J`Ltk>Zwzt`7PUHLA{>&*Mogo>vh!~7?MTJ{J|44xH<1F^r;`qx6KRcz#b?J@ulZ2LmKd`ip(hizSa!%x z#w2{aXMjHd7YILn@sfNEs^8&-$XLRL;Gi!LI3LJB(b2{cPe==8&{ZGhe<(v$I0L`S z%^kg~&Nr|U)Ik#)CaJfA*Wr;LJ$qo}ceUx_*pJ-Ej+5bNZ*EEra;>g83d6=0| z?&v(Lh-_+ zCziPp-(*gvyX5C$&!2w8RfIh+z@zcF2y0jjN$9IZ_{_ev+?^(hN1!}(6vf!)voC4C zSKB-!bTWVC`e*Wj*^P(19z|$qd;c8zL`sIyzmVfg(f&>jhYMIlfDEjBsKHD}QG#QE zRh+E$_6VvU`IW;P7N-5*DNEzrz^zG6u>$*+;~XOi{dW%o_Jx4T<@Y@y#HfJmw zAt9`nq{ltrTzvE&(NrODdjc5;1?R5Q;SQ zDjsBh;~1ljj-y@{Lz-^Ah|y788I_TsRu8+I%n%c4ucRHKdNE|vTP_NrhDC`zeFtK# z>&h5#fGpG36;1m2Lvi^5N3f^?nr$bIkVsX9F; zDPI9?NuP|*iMQlnD$-luye5EhMS&2ll&F%gInBOmHU6^lx zUq%AZ;Fdj{0c0XZj+b~@!Fm*rL{(qbm|HD%xgsaK>ZTkO@r-$w!O8DHt+!54jC_%P zFcDMx_4jvNYX=5z)bBZp=)e@Kq`ra-DulDuwIW28)_^P<(QTDrIBz2C3+3;*MOMV} zgv0FJT&&@_phX4@e-!vkO4Ml!<65FHSty@V=gFin2_hFMOATAsd^Sl^pbUYcXDE^R ziyKeL`J!cv0f`eY;dG15N`np~w6nvhlQDH4+g1ycbw};m8b^qR7ra=n2ryT(J+;vR)&gdEKT&V`W;)j^ zoMDHw5z|%ix`Pbu^sbyg1^bS9Tyah6Cvu6exA=t-Rz?xeAXc8b{I5RqY@m(J>%Hn@ zhZ!QWDg-bd8iq z??^SuzRhHE)}#jX)2>rKJ?~@`aVzv_zrZswUL-KE5CmU6ZIt~Lc+vMop3!PMx?QNh zz|3f!wT|h_Pr6jbp$UX>pJCr)c*41l3N&Rs(&@V;O!Tc4!p27qZmM>@)LHI{*rT0O z38@Q-C5&VZ1kgAuB}vTdfh>UuI9Ob~)`fDInERdRmJ|dDCm$JXeYnyYnKC91+kT%R z%ySoRG-Q4qVFrsgGJu&q@wmQ2XawKBeSFae{?6ZLm}8iE`H30D(PndVwv}Z^syZ9?-|eY(_h$X)jtjsUuz6aQ132sBTr7r^LQsdk4eYJV;x>(W>H%hDjN}bik_d(yvBgo-vD=y zp~epW@lgbS7I#F#3?<#|KG8nU3ihwcz5P7n7T)FNia%kR)zA_v6u9U6^PI+o2rDJT z`?QMl3~`b92m2!Nc9L7}KtlBQo_7{OWntXg1XRXQ6<0(;AO5s*8F*a-;?pKEzYjSR zW6NM1f&?NKJv}rAz3?2CnErfzD$WXNVRVhlc&CG-Uk>P$gcV64(p=w)J7XNK&OwPlcp%rG;Glr5K=19#w~% zwIu@g$PZ4hV|0#)@gAj%xtAw41ewJ&%rjGO`ORlsY= zOwb2Fa)EUrQSv9$Km@|H8eM`w9obyRd1g`UY;ZiXBYW}{FUmk;Gmr?2xZM=Jsm2P% zXp<6u%6r2<5;1<3C?T@x@x$8FL1iOdR@0o^6W8>;wIhv~!eh-EhMJ(8eow{8m2rJe z$i)H0aLE+-1saCYoi{l_*$x<)=%#$}+M%^E6L3)Et*T1@@?I0^{{^~jbA+P_+!7z< zHDZd&nrGj4He%AqG`W}tSB`>zq?m%vaZT5?O;v8vYvwlkW`8Zr_?L}xMWbVl{k7}l z`<&&q$L4~aOSM%Jq1N+)s+VHW(?&enrueKtrbsxdUrlUR5em$ZGfY`Y{IifK2-)%c zW`4xZ{r;EKYp4SjnEDn?&!2*={9C!GyU1b`5#}G7#I-$mRP7fR04(#t9*g(w}v@O<*5>L zPp#TK6gjM2X+GD~x!oalA<+(QaW=(x%rxu3FJ%rEV|`S`4p~A>WK>4nxC^qR;;k5z z24Ab3`HVGQnKBhpDYlS@-C=`ClYw?|iZ&k=qu))@=cMCzjXm#%20x&WE%dX| zOeV(`YQXhU4z~qnY7^Hsp#sbwg+(OCO_7+RcPmO6r{BA@hmx6E>q1x0GH&B3y1f?+ z@8&LLY$wjqWx@{L4J&=guxT|dFjT*O4g3NnXPYosuq^}g{WjYAFqc{DVNR48+~T2g1Vl zuT!`pbD4icpuIbD6Wc?Wr2JL^OYhLt73hMf(n1Y_q%|7gZ@TyzdiA7JqLd-Kf8_LD zWZ>B?n}ND{W`pg$_l&3*Q7bZG$PPzT)_)1OuHY}Z1kq9WIar-@&xFKAT{lcq`U2hc zi-5!hXdnylYP;-P$@72(GIB5Ff>w?|0&AGp_=PW=JNUtyv~6wr${U$1EVQ1%j?&)2 zyHE-Y8mt^6&-`tlKj)wDbp?Wfj?<5Y6ou$`jXsK=m(G~?qu(n~!NZS_#(}>#^9DNh z^XiA`^+T1A-j62+Z^lnWI>X7M&%01XHR5D*NY<5ft5s^>~V*j zaCt*Hn9Rr!wa@%_e?rYPUg`)$!;8?wmY|=LI6L@PJ(&$*aJ0dVXJ!LE=i1M7*l!tr zD;kRT_c8n|;lTM0bbYLa=SG%P%X+NrcKkl4+h(g24qTy>mDfNQ+7Kh*$AoXm4s4;;#>&%fL_3gQb!t~T=FWPRa}Ia-?a=b`tAM>jGaDT*bwKaJ zy1pofb{x>K?RaK1>o(z4zx(F!hHsXt3@Tx=^DkgL;C54L4bx__AjA+fI(o1h=0bD7 zD_Vlh>>~5emqYu@4?^0ahF^r!bWCZ{-g5Mntu9)$zS!!)qEOUD(L2kdZ7;=SZQ$Jl z6`qG;yge$H$$7vGoyjjfV}R;sr2xW=h8!>zQ;R;&-~F8&d~>>y@e<560!#k=3~&%MTj=%4$-Q|0X1iEE2AiJ zw(jbUU)m%*D>Jyo2RBD?>v<$7`^a&VkI?&ADXDJ#T73~IBHiB`|S9V72RWhNUTM}Xj`pS3iu5Qdz>ZmzT zcx?>&(PRX29C)Wdbo7SCcP}zNW3J;}6d7OGtbsh7y6gGJ>$RHg^xi^C>s}3b*nThh zE5I?-apJ~s)l%0>8~C~7kbt&n{zjPP?7uNWVf+I&`ic-z69pt_^DY^W_9W^Nba@m}=5&=PKKbMWu9+PUDx(`n2F~-YH`a3+v9^iu+N;Zq0 zxl3qg^G^V+MYqR*CUK%lkZ-HM&)HN2X~HuvG3;=PJQ~)6v0Npj0Vclmg|e787IU)? zACmCPi&wbmZ*@tK&avD?aJ3EO8XMk7C)BSIOmZr2c_sj}m(kmO**)3D68ALA)*3pc zQ3;7$uYP~n_O`j2m)Mj&7bvA&cR7}05u>3L8I6x_35fS%5p$;>g@NApA&Pj|w| zyhs*$ep4}^H3?i5JQO95_6A=aK+xX;laeI->aXYL(=D>suLxFulIS7sx}XyJJBnmN zxF&%+pP2YSjKZ?EX^BS#Q2A+YLuh{*F?9;c=p-{q-rWBtHI%Hx3XVQbMy^npTCvg5 zH{ss+7#I;ob`f)LCgDS1mj{l0TAkZt3`-jS+`qYV`rkwZ4RdJ%s4_Tl$3t=lFOK4e zqL_P$KJNf%?|GH+?1+B>CRP66P;;=~4g}oFBHzQ$yAGZDM zF*6$Lg zS0x7qZatN{uXOEfHaLTI4%jG}{Z}{*EtOk!0_|JhqmeWcCdhbJgzXTXBpgx8 zsWm|mm_Z!HyJ!xQ`B+vAD$7nN)pIL;TV|8l=LxS;OB3B>RTRy7W^w{4$kl#nQ0F%5 zN?_4#;>4f)2h^UsS4t)jU!JO^S*O`{7P4q%=9eq#*u%Jg3GR(%I7jtf4`%uO7Oe`ZhLF!@j5BiG4sfJ zC%>8}1<6VRoF4_(K74IBap%M$e6#bgLlR53yS~|pdvlBPgZ|P5BD?U7J%G%Mfqn4Y z3$J=;)6zb8tG-O-elq(BH2-3v5jN1&WI?w7Vygq;Hh)*p`|wePplVm#lnhL@8zok85$UyvOMl+?gyi`MqtcQOWg7TYQV6%qEXqoMHv^aIGn| zq>(@x+lzoLp`R~OlegZWu>JtA0S+bH9^8?FuSG9&col{|ns92!&^j&o6pz;&3O_F7qrk&sU$S z{e@@SytMJ*dC#lcVo>|KVIJE0GJ5Tf40+VQeH-`ONUbtHZ z_{ed3E&E9!!OZscEY>F^e7wb-1Q$`ro$rd*l}AB3;I+@k?Q(mawpi@Db)>-NANPWe zxA3X1QpN0e-;jo)am$;5(kVfgM5IePaeHI%65qOcZ!KI9ckj50fvn7`+`*p|R?fMt zw~4n*iQoE9uikhscm^(1=JV&0#SLVMtxp$_k?*2)G=&jgspw0XP^#EpQnWt5WQGR= zveuIeiUiCpc+9a)tDd1}(w@#b7ZP5XT7-0MRN0Tg)`a3ToYl&XR{CiEh?&6d*>H^y zn0wx>;^6#PnMBUCjTm%kM#9DAQ^sw?oZ#Y|!sXuN-CVSWPh--;jFXevi4s+6U$;|W zbj&+_!3lRJL}i)wA0mggAXkR4FROIQ{&-OwYwE8(0}o!l0*=35C&W827U5gT2Xcxv zZE}*gBou4ZC_*|~6BY+dvD+M8_t^eKnPUTtQg+G(Gy7=ga3Zek4!Qx$PEM&Nz2O9Q z!HFjcV^yrgIt1?tRyQ(?t6%*RQ6|C8uO_C8rybW-VdYn|Q4-BahNbT>qoWLf8*by^ z=;deEc68!=h`JJuuJmbL`|wk=S7=AIsmztr>nb*lCGe(hB2G)~!{4(mn(c)PZ9QD> zWA+~y>mCkjPV2ZuC}ah((`%`L=yB09!3QMG`}4*6Z%8{=wi8-KkJI0l+^9MWiuN9D zdY#MfAB-gr|C)}YA_Oz#v#dzT+Z-brXL7j-o<>ovcZDs?Zk7i!OGdDUOJeJJzlw9f zomULBsezn7k!NtO8LM;Z%2_JZX*J`;cRn%JspmM9y+4#8PZ6Z}B~jvT>RHmhY=>?a zcv&9+wiG06rAeK!RvvnBo0Xt|l3f=nN!;3veWE6-Al{PmkQq4=cqwQZ$hlcQ8m z-AXelQ?0X>xUPl8hWU4gS{4>9X6|*W=ltxN=8?^$n=SeOii>50Z+v~?E|^PFeLk{B zD7>Ki)PU?5L*nkiIQyrjN$>x=`M&)}2+<`oRF)VMUgJLY;bJ{^>P(R}PF1+IuFP|m zPhjx~tGQbARrk7|TW3@VtV3HTumW!otdr&X^@GjotNi)1(aX=fGJ#vE9T-D3ws`kr z2AO3!n;Vtp; zOrKIozzvSJX6!p^4R)~X&_4JptFFrMdu(#-zh`0}iLi5$lQS^F3MJzb_eV3BIX|9f zJjf+%pMb82O3=Dl;KcUVik>(c#eBSdDK2Jj{_I&?S6!;cMlKroTBgZ)!W{dzisxXO zF5GG}?0e@OvwY}j%6Rvc*XJsy@9eQ9zkwFJd9^-`33qcZavsBmRb;>Gz{H-ukCY!t zM@~}4%S|3eA3G3*{(SGKVUN-jF!y&p!@9|^c`Z_X2SZD&5ScQP>L%~OTRSJb5sqzj zW}ZuH_+tM#D z{+IrSYRb)#1w_gjWS;cFx>bk<%-+Z2egrcMyUDL%q$oNOpTBKmhvqW@YRdHlX5XsO zX5X#zV@ut<$-Nju{u1Vs+d7-S@E84p`kBGZ5BB`OC+6|5beW_X4HaR+_Y+5wLa;7G zXPW9nI#WE-q&K4#x%n5V#XDmIfy9=tFX$cC&Su22qME{OI5^+a(IG~yrf$4{RwQbr zpcabtawnG-pdHiEHOBj2JnQ4XjNwE-jgl@CsnzZrl>r{UM3 z<~S6eYGpddop5L2eCQ!GUl;D1CTdI10>G#Rsc;@v^FPx3p2H7BdlixP6QaF?*Zj$& zH3_U2WqS?Ee2g|8+cmpr&Sr^R-8HbTw>4MZq&TF%c7}p`V9d}@DEcGP73v4oa~KAx_^dWLFf`5H7Ykt|%&1@Z!L+(x-;yyiyGg z_-I)zgDX2whMU!qveM^u)P_}OGY#z|BD@74LTGxemulebeccIj-Kv%RQ$XE{HLM5CL^uAnF0NOGQND8TIt zj92-v^FUfc5c%(&2dasJ?qe6%tCHWhy<|xZ-%w)2^M0l94S=_j^>l8GzdQL&s70fc zY!Yr?Iw!LLiSKsGF&x?uU^KqSu85q0=gKEW;Dh(QGxc#wHT1uepj*`J_k?2;xa>LICilW`m2SsH)Iipp<{6#$zn& z)ngGx!)R{yD-+q%0qUJ)bV|{jf;Dd8;Xj5;@6{WTAQt2abOn)-YZxwOO{@#_efOy2 zSQZmc2$A2iJ%ObfcXw5cU0XTL3il~j-i=H40~c$T&3`X1cb(E@!z6=Ke4tV(+DWsy z(&!IcRoSMFGBI6V$}zW|l$@SZPg;uVWl=kNTlYjVB`3gMu?+N+VOu_}U@S@R@~-zB zXNY#zZz&YV?Vl!d#=aQuhBpgLmpS+s{kV8lSmg|mTkTi&y-zx8yK-1bl?&6k_Pzhk znO;>0XZiB5>-YkGwf~bVU}TX43roP=&g65}!?4QQ%lEt2JzL2@C$P|dmRpPlHE-+aQ#?+b8kt`L$Gk{_2L2pVNGkSm<74w zt^7o3k532(vtTdYjeg?H@gje1{xkU3F6xM0ScmKGtof3!Kp|enBV|NfuHSwr&v+0! zos9b!`AMv_0nN?>bopQd_q`1Tx$f;8fig2>G)@e?{nsizIpzDJu{cchPo-GXlf|x1 zEkUmG79)R|M^)@|G72eqQri5I<)w69#hOE4FpxOp-hRVOcp%DHy6p?P7o%@*^+G2X z>x0#%3)H@sf|-}x4e;9{oL;jf7|!fVi#bjlA?G)VrnCAjl$hyqp@pHIT1L z_u}fyUbgPQj4A}r{9oW6uS;W>q3)UWjB4MX>|(ND0bA9{awQ>dGe6>$yrUhMo81v2 z)M{IEmR0|-1cu_*F?7d;3aGmGX3PA*{d;1?+wG*>jAF=xez!DM4R5^!;9s`Pf&9=M zw9kFqx(Zr)cFSE;96Od#qaJUk=6)~lXjxhNbSZ(Y7xxBMiiOsd7!Sg#@!@k1E{YbR zu(SbC9D@%Pib6}#S!k+zYQ507J6`7PQ2 z6K=)Zq)oY%m6c-B1CI!|atef3^!soMgO^{hC?tk9MrE=-$v<%kSAZK7IHQ3kimA-J zVS-ZeiL65kcWB>!1)5^*FDB=m#a+1fC)MA-3u2#ywD}jcKTqRWX!YEzJA{vy^v6Zl zwEepQSjdl1HBvFPM9deNO~M=3#_LxK^O_&k$hBI-y!#1V*4gaD{p)TnWGqWwgp3T5SJlWsG_JFN6xwl=>#%k! z4zbV@OgWMEG``my+#el|bQ@))hFU$8IVOB#anOwAOQC~pu(=mQtZ)1}Xx$}5i7 zJ`1kc0_b|f{Qwy@o~RovGLQB*0~vJH55CV8zCpk20mYOA-AlDQtPXp_HWTT^+lWqbuOYVNnx^W2%p zKyPVQO93nAd1a!Z>4b9W=LfitkEE{lZuU&DH|izWU&&Nkv4IVJ^ktH1tBw8#zjmUdC3@(A>4=Gb%>{qXQY|?ijbL5F zSk~5o=6O~)NvzTb)o8}l!Pq^+&UZfS!;q4Iqf#!~4LCJu(u>P}hnfIPVsjNv$vhGt zHJTj72G@~}zUQ1oHIZ(9F`K>ZgPnTDxiP#7v^B>)lC^-=13^^_&BG5TAOE2k3ah%9 zeL9Nu+4oKgYi8qL%-Pkwes>g5=Y^uBiL^_{@msq*?saEICLtGk{$^d$FXb-Wzbj6W z6SyAxc3%FHdt<#zv1-yC!OSw5`XE}7D8}3B;%po%!pT-bZsltmei1r#@%li~+IuyF#fKO$ek;zz2 zXF&+J);7vXtzly4M3U3L0-qUpIOEn*BP3@q=fL?7pkye4O6Ki-k`v~XQJxev21nkr z>vJc|BUD$b7@<`0mOa8qw?q_I7tp*qP3QN24*h(X0qKbFm)zG|P!!-Vi7mC45g^Vy zj*>hTv8;zL@y2~wk;*n@i8_g@)}UgfwhVGfui~4H9)oTopcxjog+1+sB|(xN7n}|i zQlXLe- z+4xJGi&bdH47L`d_4!zGHVuA2KZX{9SoZkfX1Sz-nhsIxV+LKf92B|$@Og?i?rqIG zE*AEY<)|3+YiQg5=Mnai{))JPWR_}+NH+S`br^vquqUGL7-jUBYP24%;rt1yn{KMa zf%LLJjZVEEeAyiwSOW0&+G7uFBgAN^GOp(65b{f(PGE$aEf!*CX&5{ zG50X~-dXk$f0I$1%%kQC(DfeSd6oypFY#n?pYx%cXNp=4bi?~tq1J0CuA}-} z7#ZJ37BZK4bo^Bf=FwdBv3+kujN}q#V_$=Whn)FXSU81yF}}Zah!(tb;WW~0KP;Vv zMs}3iG?l@EOBhF;AI{+QzUIXw7TS{#v!{h%AptahJ_bCfDPUobybyVCzI87Fii;i| zNvQ|dwl)SVw= zc~qk}p3EqJVv{EWJPe(fp%8<1Cp)|>v3qPSzV zzX%9o&p%=f_s-K_YATB*3!vD9TZiDWoiCwnfym2VoFChZV6WZXf4ZgHuK+R-l5x?BaT(T& zL@0fepE2JTtFQQzZLp{Fvuelx@O7PSO>|vXq=`rql_tH5h=PD95^2)J&`}U#0TDEG z1OyrBUAhPep-7h^C4iJr1f&@{n$UZd&_hUOp4{&bc)w+?oSe))d+oK?y5`JGP=3fP zs`+`_d1Sa_o8OZXc+S<;e?Gn1-Vck3Y_o#lTCOv1(#{mzZ3_7r#*ok8eb1V_yyf>N4S7>DPNM}&{6EtX9OyHtng@apFB=8f)xTzBKt z!+EE=QQ&fNt>C4rT%pjHY?61_U;*gimpBu&By%GkOFfY+c!kdR!v^t%t?^67i)fem ztyqty2H8igKufcKJlH4Lb~k0$deD(>Y4_vdKXcWm8l?~I0F@hRw}RvgA^VP?x;Gc8 z-LS0@R97E7zWT^w_oODR_9uv})&8#51sjI+3(h?vsG$?;1eT!WQl&eep*)UcuI^iB zE1&AEjBMa`kx_Mx7C%ETq%|yC6+@4{kFt^~SX~s@O(-YH?3b=;YBeWegUUrk>SYWH zM)&NmoGd3004qiUT+>#)1WLD%1t^7sXM*aY`!{M^zRse;7mE4!5t>JPFh@GmNTqft zkbX+PrCIO;{nVp1+r{c^{C3XhRAbjiG&pIUjN5AQnIqfpW73;{Ry^XMROlZu=(sZO z0PSl)AoPUeQqy~2*6n7HyKJ!d5=5R%i~*Z3BBoQIrmyb;s9@x?zg#=}etX)dmkugC zx_#fEGW$+TsUimvJBA}{@lz^9W!6{fX8DEpKSlAX6KLb|C>34}$FL|s>7uiGs+9X) zU-c}Dqk8I1D=)L9k>+p53%oh^$*CLy+P$omu*v{0gcCl z=C6=Mm{VIRQD+@Sn;8b}-Gq{@2FD&0^wnhZNp5A{gEq$dBA)h!17Xbg6?s`7&3V*_ z{NU1KLfB7OM<}nS@1-VTQ7QPHd~a3|UK-iN$U~VhOciYN)MvP&8swh8I^#n5Z^waX z{mIU~2qEsb5jW0@oT-9G8eQa|LuRcQo*^AsZ%kC4;h+w2;ZMo-l+evQ6iJ4oK-%T!G(?=W+ zuu=COo*)>vJqFutSAhZVp^y#U@iZO5qv_RGM~JA|kGdDYIL&n!i^mRd_WSI61l>U| z2O);qC;sWBEDAL(oOT`$3HCaLKJ>XOWFC z+(b6GkbOrRsx3UW;~LO}UcXahv(JXi0i!P%Xequ*o`krYYS_cyHhT2KFb4RwGZsCO z19ILDa!gvqfRBIj4+6CuY#$W|f#zPs4XTpLm|4lMGUtZa8}~B2Re9Y-r&bWIdQf1mZO0#b3v$BoA-Ixh{I}Y0OBAZL2P?HJ&;#lSb%;;H8M+d)eHuZls6!gRi3kYlGVBa+I|BiH0XSmCkvHL;-L z5m=KqW|p12Cn@-%7nsIRs%M5kBc)@WT$G42y%XzxQp_C5-m233IqjnLInWOI^n!9g#6>~)o+E{o$V~eJ@v3%x|4UN^eu?GmR0kSYJ zqW;y@md`@h$g6;jx4~s<;Z=U^tIa+E4S5(T6^yr=`e>+Q@RR@V0VLNjP?kwkJ0Q}^;(`At#AnGof)x6}P|4`D z;C9Zh-P3xlbSf;9+ltnOyJvPkUAS(`UNU03;};GoWpHR;U$$5&>!X*^%9kt8p1+E` z6V>DzbbX2;jPKcxGVa;|z0KkKNX=0}O~F9i=Ro#vLf|}JX+bbrKYRskQi;1Y?#cK3 zS9C_`vka@zz-4tM9x^(Fw)iJF#Q8=IbGT=4^Oalue-(=^v; z*~jiI6D!s)(W4^{S-}@wd)&Sc?;*cr zy@I7#B0B^v^5p&$Xw`KmaJ?DvgEF66Qsml{_>K9mNk7dfuP$Qyx>5H9TC4yr~YxF{ysceygh9 z3d7qjFW79r^b)($N03ivE7{&eOYJ@jdO+#Uwf-JNJsX)<0{1P@(s7qX2=z`}97iSi zNuQW+Q(>>DqhfGfxdD%;v)K61QzEEqv{^u%^iG>nW>2bF8|=J~!Y__~r2$h&IQ*IQ zqnm67XWooBH-&)hgN+%a3s8+gjbw-s-Yhli!#JQ%%XI~xEcZ6Sk?IdX zx{eX=gN_yOphq<E6Dh7eQwP`j{8St-d%z3va6b*R%#bx!8WC^+$eyDtHJ;b@=UD%+*Akt^= z+Fa}v{PR;Ey{Uz?({ft#^(^9?_`FXABYCe;JG1hE8M0rw9AR-L zczE&sTDn*}VA}b`4j};Uh!D2tBbOIx@$=;^Ch8Z^+C8lAw>2pvMMz@R^llXa3(371 znS2vn{rict%+m}tL7p?JEfya3BLSCnH0!@c1+DJ zHV*{U78>xet(c-UyI|yv3W4prPpxCf5t!%Zjlswt)7o9gBSSVtTKsd6mdpq*qeZXj z+v*$fb!%Zotf!WF$Jk7fWKJ^WO;IctLBMlJn>DI72VKmrt#sDT`yma!Jy$@_1iI;z z!K;-Nc(xVo^Eu%hqTyvO=ppCw`wTflzK;K@b}arfj5bNd-ulf_qsrT>m4c#$EPf~Q^lk{r$jKI||&C_U!8s!_4 zISqNJ)p~cLB(CC_wb6aqw2uuo#?bdyEJ|04kmJf?e#&Q6qJYW84hiI_l9pg+5}UIO zLM<(G)7!eCP#knE#4LVRzfnT0;AUijb`z_CqOLb}_ zLU8rPyf!8Hv$Q!#Zint-*tdL5Hf?5nkXm=ei$9lQ-c}may(`>!G&9(N|3W$3zOtan zG(-ogNGoQe9d9$fiOxImj&oV>%$0Y_IMb^||6Sumc|(uVy{3DoY0JtfEIp>mFmsgg zUeWz+mi$g+XI0n^*DV;>Q;2?ic-H^Z2!~Ho#2mx*;lyqrI_u=m+A&N4RCy%Ex#F*- zI%yG6-YxEgP4`q@K;_5bXMPi5oA14S%*GB6Z!2g@&^k&K(VUK9KTj8;%kq9x*OG>V zJy)bz>)pze3i^vT_E{U0j2!j1^KvgVpQooOPLR;OA=JENePZJ$kz^;Yx9^QwIcb@p zV09iIeo?8$PH)?8$SQs-G~zg4rxMU95A%xu2Q*(;Vp0VSr!~uzmOEsh!lw-ED8#gU z3UA^+V96D4?ktmMs5ytKTKfTNCV$<4F~tCH%s3BFb^qL{zaER%6LywSbsKE{6|wn} zSMlAy$HqZFpw?}MPZ!sK6tRhsd9?KiORd|h-Kplt+qn&2*J1d=(DOH3|52i!TYp(m z)IGwCFB}XxEnF9wZBLH@5}(-Am1=moM+Sb~VFP@hS)W_p;|~*X|MczdA%k&Z>5Opc z(s0V~qCNwfhcl0wKqR?RR%u+>eO^%D7&~&fx>yyvhCaZJx#~Y9YifTt(S>Bv`*mvZ z<#_)I{f;-(^FNnYj36(dkZrCnZW7o!SRN#GW_}Z%aGu;+78HARzr*Br=Sk#YDciu! zKtSn}21jUd4P}2K2aU(yFz3FrtxM@~6CjTm_J?~>!e7yie~dxI zX0KK?$riy4FI#uv(5&nIQP|32-V`w_atM0B_$cUN$s0(U4M+mDWxO`g47-F;o@Ro%s^ z)z%lmN@hZ_fQ6E1-C5NZ)5T3s)miSMSd3y3jYS^6={B1;R$9H_MqNv$e(ElypN;Y+s_^h0ANp z%D=_NUZBOr@-r@6uPm}$<%G+ zjdp#K_|u$?=IKQ^oVo2Z_OS@GGpgO}It3B_mR@|PH;;auKdv&Wgi-*D#qW>fkf9ZAv?fk` zT`(=pK)YnzpGt*9JtM5;1aR)%1qI#HaU18i_PsJ*$F?ry^V`EXYYpGuPe*@oma)^N zgx4L}-wIx9ec|4A#k{fXk#v4a1-_gOKN^JSxB#-rX~ea;3OcNsmZ!>+6 zX@z0JjhxC!aS!=?>ZBdX3I=a0+-9Mx%CCf%FHBmbj|qF+t^7m)HWxK4bsEdqQ#ulb z9cgday{#MU*ltCpB5uS2m-+^{{vbcPOm`^dn|+SFx>;qq190o!5Yd1(`hQ#4y=>S5 zX-{<8cGwnJDh(MKjd{}vxk_NdD}Ay2|DGJS(?mkxkg47k3nicjczn=uAm|{ukNLN| z3uHxzZlOjx`s?yLqy%h`T>4{yNNEZBuR)+GrX#r9tmLr zaiG_7vlQod&#GEbD(ID*Sg6N`AMXWro2I-IDa7lSx4y_!ay|QIzUJMwxjcnELt7;fX7@~7-yc2=b zF<}_(x(h@cxZ!2bK;86Y(^66WZ|zEVbv>_C;(L#x%$Y!VB&^M3kRm4|UL#~FHwv%@ z<_jGNPEUtllpx%~lGu29k;>{Mgi^JUv&=(IM_N9g4%ry|la#|O&@%j@YIYp{a@-m8 z8j?Xn@6r{kvrZYZt9I&0)3h5s59CvD#D~9Pc^QhqiN*t`gDNA&A|z?k^jbqMnx|wZ z?so;d=Z-|!qhMM3BO|+ozx^`~k`zZ@<{%mmX6_dMitx=(25!7EZ z)Fx|$A6=oZ5F71sDS4S{iR`C63l++0uC5&Cnz&m3{m2oDw2GR{yej5Mo2iV`odCc! zAua;8@uHr6kR${L+Fp zR_LAXEupL2HqDz}OM+$Vusj|!zY|2^jahCuZdkLzIdF z2RSHBIQA>cs92J>T&B|CHAJ0?0AYSRyyXSCBje$PhVc*369I940xDxde6LZ1%9bB0 zu7FtH`-4fube2`%ZlWY~{52Ry3)x+si{k>Z9QKX#ov`5-zw;pNPdi5V2RB|BKO5~T zf~3eD+>TDuWWh`&f&!{787HAyPJ?-D0`VLQu}P6pshAG_DBPciWxm)%680+h?EpFF z#-`tw^pX8_5hN`3t$v2i4@eQiF^{?cszWkLx0ESq}u&o^Rq$f$+PGHXa|KJ&8C`V&OY_cM)WEht3sRaT(-25`xXMauD>&3k1dq z_xNUvof}1cg!VDE)jA1j5jBClZi{r_8u7bR-%2d-``?0_7P9GMcaJr)>7L@8$B!+g z{Q{QBA6*M?!tnoCbe`7kgu0VUl#HAf77QhDuULuLkt+%h?m?v%SY7=vBpASw8 zeZA&Dun8Xu38}DA;Ww?yKH3!w-@9kTH`8Jw9FHTu^YYSk#c%D1#sz~DSy-RT`xuME z7?Di0r{Xp0*J!I?r5J6>N{}GPwtz~$_LwXfoQ(5%wu=!!$CpD*_IY9iKq7g5Vo>Fj zqd_${ulYUPu+;s#C6M_|54lFY!P=}vX$GU8kto}U;(lDc}D2BUR z8#{aCbLR^oi>-A2yCecNQ4J@taOoAj-+z^9F39W%#wBuxSjOHkhe|)YXYl#7-@isc z0W(|K$g|KFSkp8c$W!u@`v)|3ixD&?3u-nD&}Yc09kSA%)F_%jj@){@P=EBn9BhMx zweG$7fP^TtvX;`$XW92(evzVNR4hw~6Gvw5>WAVdYpVUb*g<%=dBEhd^OY#Y$>n*$ z6;1BR<$}(BF_QDU+UJgZFIhLsNeDMPRt?RxjAV)wcPg(G5 zP?H%VuiCS5c9wx=x?Uy`e<9}OoMvRnlJu1_G9*Mym; zO!>(uW#lL?(Sh82_VnT$`ov@?^e&1js(e970!q|iZ?J8*v zxB>3WL5rD6*GF-a~K_(YjNEJ#t*PMvOA%sXJYIY zJyi!8%gEJT+{d&SFZgW0=shxC-^5|?`W|XYiR-}oKLH|`7q^jB+0HVL9cg9F>iQ;- z`7@Ce*xunoYY8Fb)8iR$Yh%(;R-?~lJ!>o8Kgj%)^|yKy2j}LdNefldk=ZJy96{Jd z@3xCxk^>{(4oZ&#y?4*272`XkBWlR^yb>JSR0TgwXNmLRd>a1aKq3bpI=CZ%0*hoirR8u~RAf*e+ z)n_49Si2y!@;Uek6*~A*qofk(%blupxcy^Ng(Z^JDX3c_80>@qcGE8LANPn?~EAhui+IuO`?J`0)2lzwYJosU$0Ub!bS z%z(a`^I(E=7MA*1i;gmZEA;hCt?7x&us&Tt9Z@sFOsQ~5Ph@sX@=i^^PY3Ri-ZnKe zLk@b92KVY3y4=)AM^tt%B^9}aKurn%0nOX@i;6DatNvGvI2O5 z3ZQ{&{pg*s^pNjW#m|Pvjs{kc;(+=|3&J}1To>%fXgz8qN+=~oY0Zj_a&m*g00)Ju zvks#h`%OQ>+o59RKN^VrFp$!;kGU;FbpYdsca#;ysLf+9w$-^&5Dk5weDSGi9VV!# zkPJ@ctixIy@efxqSzjLA0p~7f_nY45+!lL8$juyRs~KC9~1crQe-;4i%Z}LJ=+oOlF(;rG@sO32qd{Mu6yRxoC>ezop4|?z6 z7Aa-(?uCYEqfXydjK1yKRw?wubLo&Yw!bmqV#VpdgUvRj8DUV1_NMQGRXo-omtiGy zE@dk&7(vCHdS?6wD7|yB7oft0A8PQ#Wk|K~W&A4U3i{9xKTcjL^!-w249)Rq)fz)- zKL^i%)W)qO$_6#$JwLbt!oU9vc|%71yio)_*iBMO>jeHy1U+!Wzqb{CM>NP7B~_q& zzAzO*-WK4pl=RLTSQnB(?j*I^r@w|qA+ zUH#%^fZRwcfdFs|2~5V2um`FmXpeP_fZc3NO(2b+(hf>#(2sBU6E91g=cR8@J{hP1 z-;1+_)Z}{wU;cUwn-Hx9ZPqYiNG7UguH)^3HbB27F(&9|w%gM)x|AG&G`}3BM9Rd& zfB(HbDt#5xFQ+NNJ$ZZUZ^>|gs&mEO%AA5H)R}F{?&C-?ewTBgqe84xBXd@Xw*kx2 zGWkJtg&S;IacJew+tr7~KV;8$ z6Jwztq+~Xn70zDPsdEMr`!bnUX*opU3Qk|RBf3R!)y$1iB}%^Fn^PFROHR6CD6;zO z0C;x~;CIXq&1NpCL78&didX+e)dtF=jm5!JQ8Vw??zaynbZ;c zI^vN}-{&7pSgg}T-5JiRI27d#jEe^ixbuVO%a0JjRxetFwpQNQ$d6Q!>yb~ zx_=oDPrjRULedn){s%p|2RaP3ZT@oEuNA$Pcb?8rM}Q;FE0iO3QW#_Z zx$%>V%*32(Ju}Df;D;QFG`?jMSkzz3b1IhbF~OKA_OsMr)P^dQu3RH0A$OmJ?$3z! ziL*gyyPcx&UGIPC~vp zWlr;jMRSXGrYrglTirw_0}HL;nny7#zQN=KYj*b)0c@;K_E?)^@ubyy;^r4SXPO?W ze}!_uWm6aEUojHU{)=gAA?AL5QvBf?!An-Cene{-J@oc#=9%;}&CvOQb`j6D8jjI` zy^t``0T0IGf*TEA!RzV5Pn9T&Gcx-VgMxo!xcftuHOhp0x4f8r$w$B9NgYbL%h>}_ z_}5JC9c@E*tHby}1z`^4JNCN_+O?aQoW*&@ug7&M_{ey@VOHN)M)hpYXANCt*R+DS z7%|1$p0A<1Eo|n1k3XxNMnrBbd!6N-{MDM3O1g#k*0{`Sqi+MrVe43jx!8iegf$kx+9^rQ5r zolG)(7q@?EXxKJid zdL3+XBh?efmbO8#)I7=P`uu=AR)OGow$vfT#0%ph*4;U$zdJ5iely&*Y>9Cd!|oOK zVRbS^|AJ>x>4fFwClw6KvTvV$2UaxN>@`rN6EPanbZZDp0; zkN?hIwKv6X3PYd*(;x!z2ho;d3K()M%gDWd?Pb+D=<-g{4;QxLBt_w%m+aSoIv@7k zqrgoL!i85N62D(LfK?4SSMLQc`46RT7v7}@(ZJt@Lp9@u8&V6y6-eSX!I10#(+ zWXc8|$mr_te;=*IZMJphnVJfcB(?8fx(DpLJvf~qkMt7i zj7J$|-adWqS25;L&+jlC=oL%j`uI2nx(*jVGg^zkiwG{>AG_0zS;}PD7#S$#8IR{F z>uLx7EMOd%Xh7eA>?_~D?qG=$pIa5pp`||-xjNqug|;eXD((_|hACV!`3rw(pv`6N z)gFIxxFYB{bWTv1l&7hcO5kY6#0noVw;Io)9@t%`U&s8z5G*3&;%hptyRUR&SX>-m zmS8{UuCq-*VkGva06o9uF{xBA9Y#xB`Nf^PK|B`+^~)2nMy|IBaDR639pO`s;e`^P zl*#KA=D26dYJD`ORBlWjW?q54$pr`f-kJM{NN&YnnRi zB1XnLodbUYF6suuphEaDf8(jTlMeqFGte*1{$EVZU~#i+qr(TEo&}6$Rlz!Xu?&X$ z^gNG;Fq2^w`Rw%mgkqykA`@9*_xV+oIqDD%(=CkT0ppIK#X)%fr`H6*e?Jt5yQw*a zRoC-4Qijsj>`p^5m{0Netrq;zxsV#DpVQc`eXC_*_tr4jVPF-{RCjV`*2sq@^hQgEBjZT-{GPxr#d z!uvY4+V$HTu>PYfpn`pb?f1cu-7+&soDcfsqDhjQOaAF{iG0-c=JMRm4BvYj=dobG zy(h7+Gt=_X^a@dOcyN;OQyGEM2mk9)S{R{K%c0kf18Do8ol;&ZQ{Z-t>!mCrb0xnTb16-9;+){Z%XV9y2V_=8iR zZ+b=B&mppPE*F2m2o(TJe53U^%KeY8GY3`T8R@zk@@F!dSTu3V{?b3-@Lo1A@=Kc97 z?b}=3*#@T!+A(E%`w6j5?Eo>U@G@1x-8am|T+=RR4HymxW2#E5S>bkXE^<~=;Bmwz zmDOBjBWr%TsC#nUnLTAo8S1&Tc%EY1r(*KwJcWbyjv0F=P$0Y@$ne%H8>6Tm>Q1q*le{gx3Ah>~MHUn1tu?oaRsNBFd4L%ZWBw-yRlm3R<=!FYgjAle!C z!Eqj6zg6LYIH_vc_!uHvG?gVl=DZ7YC^}<0ITaHqGGI`M$Oh{^H53OCl8mXHVg1$> zb|{2KB#svBIi^)*B}z)7_ioLiE?e;<&VYJ%I*Tq+YG0gBqbAy{;wd{(k{0G{YXs^w zp2Y#T3L#rIQA=n^0!pDW>y2LtU>~SPm;&;MkQ9@Qqv1iD564pZTCU~4z z-3<^YWZkArWK5!6a}SXwz3{BIUt-IZi|9zwuYHYF0e>>2#RDXP4=P z(fW6uRBu}5>dw?S*dnrFq^ybTOeJ1Ma`naw7&%EaNP>1z?jF$mU9E&KZD)-!OUi>%WVasgzyo4U9I;V=`_<|Ea-|@V z7w1PZ5zUw07_L^!2F+#_97|f^{xMDJiSaz>yu_a75am5=(%JZ5J9DC>1 zBestTdPrXaeY>ywuuEkWI2RQmx&@m|OBGnHNWHm|uS38-aAddGw8kQXfb*XMUR)pI zlA6`Ud2r|Z9ZSAuHyWZPC!x6GcSJ(K2|*Y^L=5!H@Zrz^NBJ2d>O1>y6u2!o-2&!3 zyI|wE>XU7EQP@@u7j(HbbVXqPDvAX*c zIpk&1+ze?^uAe=AbU{b-mHwE~h0pj3);_78*I4&+VnoRpsi9P$5Ak4sg4}s}98l_^ z)xUOU{H8O`muPpup=}WMnMsa4w6)%nSSP$_H+z*Rd5QHS4hVCLP6dXO0^^Bnh!Bi5qkw)=S}f46Oh*}Vh>2Kc zn4kIhLn1$C;Dxib%adBhqzv0ytLFmzTytILh8@E#+?9oo7K)%J-RG!$hRuE$^Y$xP zEhz^*tGemay!f;e=J;Z3f8bl=)Yd8VPoPV`{EcI{T?EwsEY~_AkSO^+B>(!ABG0?Z zl>LOM<#c@D%(u=DkAi^qUrS}sx!~YDW>5h_C-TGt!$#<%Gy9m|f2Ha1C$ey;l$(UG z1@%6X;I^aSgtd4&%&m_^4DSZzZ5_3a1*eDdkyQbM9g5m@gZ5d z0wh6}l!V_i7r?UBxCS4QRjPeEOaP`Hu&vBjo zg_(n5J zC*l_hd=(ip$#u+}-;-$r3}`Zgb2-MsoDfggp!M~eyM;ee#l3Aozeo+2p<`x-+LyYb z$EJ)l?D=5Fsqg2F!dBxnr0%lTYm*(g>&%3=e(DxGB~zair_2!JQF zaUJtI5%bsRyMaPaF2&)T{z5bzblG`s0(dkAv`h1IyzBw)0Y8;5z}E%^Ww+;-W|Xqv zAPMo9^cpzMI190xGxV}kyv}fJug^kSne&RTz6=W`sEuFB)Y@#tZx0vjULf8YjdI@2w%}%yoIXtdaURAsb!dx!RB1!p?xgbuh|- z_T&ED{h5WxyyH5%cAMWS*`&xc}$l8~Grqt09ximM(vghL^+>nSzqX1^-GbR*={bv=lF( z<)u>EDqxO0bC@<4@=WBQKFapBW z%%nWW_mQMM#YBuN>xZ*o$I;Jt=FQk~J{%&ujPqR7!_O)oS~9>7in5kq!J%M)T2u`* zO(X3R0j5tZm-(~oPuy8KeI^j9PURBK#HcKPDkeMhUVFXB+w<9tnU$D}aXtqfVEW})BdBzC+o6)8Y6pvc)kz5<>vilIk6k&d|sI$E4J zs)%gi2^oUy#)0}Nhv;Dsk~qWc#32| z1pU>xL&CZ9>LQ+Mv)ig z_U8u-h%i{`g-&=VRQ)Rh28zUqQYxAo_P9+2m`a1SOWmS_?WGyn1SYC|iu*3@nd zsg!Mc(XPg=)1o85+sRLk9QsgJUx`=;oGf}sYPAC6lQ5)xmIw>CIPO8zhhBNF`#1&T znk*#|brU-Nd&`~j*KjSa{HA2xYQ;&3!u?C&ng<;zPm}sz6ZHy1f}u&#G$Le&EJ2l@ zanSU(ey!L6s(D9cPdRg72-2W8K4A+tGI>gMt2$c0$;CR+x2(;Az*prJpw%vyElb&x zjDESZYcSIlffZLZl40rLu!L=08^Fq(nUVy3+rb9j($}DLr~hqVfGONhd0({z4{b05 zUviu+SGv!$PbSci9j>ix0%_h=Z7-bLl-=O6w zir0;$zN^HQ%c86ME;D5@q_~{@39VKb3vQ(saV8s6UYNTb0!?~jh7ofAR@VW$J=iyn z*k!vtL5Yk*)OQmcVi19kHhaVQg(SB(5Ft_mng1pX0SSf+f-;--NnZx5I<5m&YhF0$A_$2Kb5jucBq#^VLp9?7NrBFL7; zhy1#yj5rv9oH`ldoWzyikFyL~VcXM1Yt~RtuONm2YDQq@-DWXI6wm*ZVA*ICRk*FJ zF!UnoJ)*mJRxA2`lEWzuT7WD`ZfC}nQi1Ly&!fJOQQt`LIzZv$c_@vo>6jUNW#%(P z!jdO|WrzbxEAU8fN@RiP1~EtGFFUQEvE)?xMLGZ8@asK-f=`M+kd8c@zBKJ%393Xa z)pRn&(L!^I?~nO=mS7!?jnb5gp}$0!0A)g3KzFzB_uf6KbGTE;uxkqWE*kPF>Oil_ z&8+%5&_+wR2<-tz%~RE+&%!d#Ki8uwB)@Tv^gHFFjL<(b+A1W zo2(Z`WRt|S9tDFF2G*2C7qts1sZh*1AZtM{k0QmvK#$vGhX6_<<9W(S5OESzFjz^7 z-eDjgwU7!?jUo3p~|+2Yk>U`%DBh5rxhk%BELxkIvzbsQl&UB*u<&+?bo<__yJPwwuan_ z-Hf2L$0>ErYry-T25jrF)Fb%72ylM~rA(F^0@AWbSbK(uT&QI7x5mc(Y_K|}n*)T) zg}Z>s#pI*=Q%#QoJt3Kg4O@C3(QA4o6nc`!H#4Y9K#llh-=Bp6y|mOiMOb*oiIf4* zh&2>x>Ffs*v4I;f2{Lt(v@vKs$EyD}ux=d2cLmt{v#>4y;1rl^{9tiob$XDE71@2Dk(5GHJ1F3RpbNoWAHY$I7@(b~+slAkg9FAHo{Af$x z4#Q|ISwqsxuEKyi2~lwdG(JE{`tKH!*D*nXfB+zwXc7o4PrUd;P1*eYu)FO*OYx7L zcEEPBP2&f(Tp3S;e{V*0Q=#?TIQa^waKLkT3L=gI&N2nb|Lt&_=Q|H#JSeY_Qidp+;n5?|+Q7g4A9-m+;-~ ze&O533Bs!#m2mCELhIC+33g$-_QMb4Za((QmeCD_;LqFOE$%zZ+~H+ zy1#+{CNv*Ug%+sis1Tk?9T#?B)^yJT9KAFOuzoxD3kXO`6<7lX_Lk$6owEPwEXBnM z_=a}VK*!7W0^N*I(HXWKs9~FfImHx&Z3BDg*msvhz~-1bXiOs~InD^;!Tz_`i*5JN z`L_I-SfD|sh!$|KIsx7&(3IinyaiYfykj28{9LAx!yY?kdF?r{$Erhz_KYt$F zb5&~=1)kXqWMz>#LFk*!NAiGBD4-sFiW$)d`pB1}P$a_7O{as{j+!Fw@}W-}M39ST1R~#JjuF!#KPlR(j}GTk;Dr8yEtS zX{ZRO{s1vR&cC7j9|C&p{{!Cs%(p*$xe)FqPQ#LFi~OGZ_V^MHam6^Zl$ z^-bzU(4^OwIPP}9`F7`h9JHd&C>soxg7?OLYzLCy^tNtU_bEE&>@eS#>O=OT|0LvZ zm&=}oo-rdc7$_%a5#XNuD(1fJ$FDJdZIJY(mEC0*fiG?0FBYj1*Wte{GQ{ik)8jzZu5zQ)Rw#yqMX!0an@jKX%XKDpq(E?nI8idH&p66Qh*I%Cj;=ebiSFrw z!O%9+HK7JUno^R`1R?aUbP+^)@1nc{A^`%C-jO02iU=wS3Kj$nT|hc02-1;IlopVH zA^YX`{gXVKXL5J$+?g}yB)fP2#_pq6Md2DYWh()(^1I8~=w z*Fw8rr+`w+l&jct?LE0ach}NBah{~MA}qFK2u3vgSAKjjYb#J*dQ{hMgzLp|9UV9y zayrt8C;n||m37K8NbM`uW$Zu7+>5GJdBzdJSl~O3i$-QxO}tF{I~S;0;NZcSK$q}w zogQ^7!ftQY`M6Ffy`&RY$5Sb?fW5!HdGdIo;$zF*co5@Z$n{x#j}#5ocJ~&O=CeRy z)k3n+?is<|Dd2db2sTTf==rLefs@3U(Qs!Yj>)gosUU5Mq#EZU)cRYs#S9vu=h0#a zwSUk*GJn7Xl@V%{;t5HVk^Z9(R2A5?&5Q>lEB+0O>mj^M*_tQ`%oqKXyhgma@ke&U z=o)qbjqg%U%s!ir|0@xst#G}Qb zAmTMYW0of*d__dxt?ITFzm_-mEKAry8S{BfSuBcWLa}eg9FbVJBcxya2-;7|>S+~| z#GIT$riimu?3q6ClqtOHu5)m;LX1}saK&5v%pw{A%p-EcznLJSGKMUv6`AgDM)iho zSVNnph^zLj_AEf8rd;Z6p;dPAtTnh)#ewecZ{FETW~?dV!R%8yQ14}Z{0MqGV;0_L z-Es3kj_|Nn`c6TH73$}u%v5)D1klE5(52H^fHq;{*)?aq@$qljL#YSvr+GOnA@3A| z-hX&f(KcM-j1GHq4gv)lN70wcawh8X2S2EZ-{>t%$Zg}>?d0aR@Ba?v6%0f02sgUU z)Rm{Yzj6AnZvA@SNn$5Z-gZrv`437&zxS*3X6$l7!*><^I@R!xgd6Gs7P`|`eythF zul`lH`4Kxit*hN2aIr?8pGo0O!Ays-w*h8r#jssCbh2(H+<83&4b)4j!gL^~g=X2? z?SJ1w%9L0^QL%GkW>>7XZwvL5`B^Z_;>XY3(zIKRU@aj8K)Tc&O3wzQMi{#3_@j^C z#(&QD{WgVhkDbec&$5T2aB zEo|1vBNi=>uU+ae{rPi6`BYZqW8dK+s&vZ2Dd|Enq1o#de7C%3#W&;w)A{s1)`yEf z>!qb8{i`ShG6Kq$~IkqQN7K=ZFG-E{1C&ErS;))z3BiKZ)dB6QI z_v_p0p)=~IR0?eCeEOJD_&%T4%Oe|Z+_$;ddZ+&SQ3I`St}42~81MOvCDtc?N?&3l zC(I$4rk#}FH&5cHFCF6FteQW+Q~dYBTU)uBNDV?t`I-J(I*{193p}cI!RA8EQ5-vL z>^`IVb%H!M#l70&7e{21-8~FS&*9%_nJNs&O}Cso&42xAK74iYE{9Eb+>Yk69qi^L zZ@rW`@Jpd$oZ($e$n&|)3-Z#71?PJs+zN8&e>S=-{OlrxdXU$hH{pHUKW3XZ9hHB6 zd$cm`%6UHcl*TdNTWEvB7p0sRbw_MBps#IMW-DLveJ#9>iCNwRp@}(}8BKb_obcURzhH5BYb#sFURz6fNfw%|>6iydjbuVF(p9P^i_i#7#X5 zSvMYd`T`zeEN5_`8i_9nwF+~g-oN>^T;%P~l~KuN`6*AlKF>#h*X&uO4#QxPl|Q}e z{!W(G)TfmMQx~Dq-&Er+ZY8?^j|7Y1IC8=1zgx0BWtrRRWf_$)J2oL5MS3rWkOJiB z-qW|^>tiJGwi!jNjQLDA^r9xc_ZzeXIsF4EpdYo*RWUK{@ORj)FZ(W1hj&Vcq-4_m z`)FS%2^U$Q{ii1dX9XQmb~8HxTbWf~$TIGJ0pC26Vs^sE1RiO-!5sIJ!_)(>q zConsF|9C0&3l8x95221D=9xk5_mxHj+$4L-o_BM%{W2`ng-fUmrQqaJVe{2PN+x=M z_rc2(xt-Aa4r_H`2YPt+>b;M{RpvmRZg0_zvyf1Om2+J5rvG--x=o@wxecBrbi0N? zyhq}sU)-dJ3(2HGg$1gD+`w?5RdZ1jzt+6UMu5hbQqC_Jwr1Ti_yV0=(Y6h7Viw%q z=RB!^_j(&&QuF6iwz3;8xa{v&gKkL_OL{c=_1{&v6#iAzw2nQ>odpDtCA><6vWG2B zf*^NBYe0gkm8W(%tbVxmNCsbeA~P3mHaom>vZ5`$scKVY^4YY`$R1W668a*D3AUaX zVc3U^`Fqr^Mo4id47=&|b}?B`Gf6ozyi<(ztb~Sk$_Q>t=14{4p_9dEOHkO#^oV+C zrVTo*8>_6H+6Z%d6d69FgYb$lA4S(-R~B7VV3TFAxKTAX$YjxKW=4x@+<}|Q2rA#} z^Ph}`%$rt9+m_A^CT1c`2&$Z8H4*|!VrhAWK%XbRALG{}mI@ssHL#U1Vr*^3B&74% zVc6|7W$m7&j!@~{rJHqn5TCdY(s3~4kK2q6_U^4m*h53eyA?yoO8E%1WW0duy3Q%y zjsd@v_je|luFzrJPGfGh<0XZ{!T2R(=w}l)hgDQw`xw#zy&pe3SI$hXZ(DG%pVGN%7GM@KiNx7)8RZIsN;SPisQv@) z;kfs3yOS6b51A9c3rcs$y`|=hG&uRt{0f-ruSikUHBsBdghjKJlFSAqpERYN^kUp z{37Z&7~NsUgX^V#tARB0T}eVNm!~NX4=rW5dV!~`O;}Av{Ba`?qa9YBxv;X4amkeIo-!)cMQsiq^4X!&>Xf|sE9;Ym@tQmxT8lw^Xr7LlGT+ZA z?mlR>wDm29QS75Cm|brF-_uhlPURv2RVRtyvbPZUG!vNq@xlt;17B*gq9%i7hfQ8t zv6zFK2|$M

|@!)^?w)RFlGZ%^1c$acV@wM(}3GyL>_%y~x zmbfpYVEnKmatom` z_b1BI=X4;Smn0xu1%+7BQPw(cNBs0g>ad$xcJFlaJKb1pGr)o3PV@G9&0KqVWw#U^ zCh9mnzs;3tujAK>@`~OWN)b^V-6uxH3}weF28o|~QUf;DSgtUqx5Vy?dEI+^?MJd* z7EoHp!-#4r=7W)!)nAM7UqT%>?5=Q5LGQ0B+%J@Bt4{Ezlu12abEY+kKzd#0AO zsy{hzvGW}#_x7aA6$p&_kinua2Oi{-H_p>M-^PH}4R>b_ds}UF@x_O3;z8MJ(+IG? zSJ>#wd&@>C_Z2?`|5n7|6p|c@$I}Z{SWrLhHdByhecq6s4~9Y6=G<=w3!xp5R3O!t zTQG6W;i!aA!CYvjXm!l0s!k@6Zvd+5_a;yYrv>{n+CEh^N}|IwtCsSQjO-%yq8AP6 zKH^$~YGvlt)HHVmQ!tIp8{Xasq=Ko{xt(soD;R?1v2?l^t6%F|P1BRFJv6=XG#Qt7wI}f3j;tu^ z=a?C%4!@8WG`V`4=aHJAZsDMSew{dVwtm^~vHDgkRCfZ`o*b~;j*a>fR#%6=abp5k zo1czo+~B)=d)F_Qzao9KazHkA+k#~a5=-a|{|$$}CgiP?OI;v>oBw4O2plRFUrOK}<{JDw9p z?dKZhrMzW+!bT<=zrSYwotzT*-%N}fp| z*udMT>|`B3cN#8PuMjNXpx)Z4Q`2C3=Zq5*vZmmJsytx(!b!qr#wHM+qg)8?TSCv zD^mXm8s0u7kp{?>IHv=Ky~mOeS)_{kwb_5~4!<5)a$G zZ@I0`3pr90WmWAt-)xU$w3tMzw|Cp1+pUeD4#;FIKSWv1HZCCh zBCE-Yvknmcqkv;N4R=}$r}j7n?3;rE-QSXN$EgcipyDsGskh%Rr@(z4d}+cNfz*&w@&dReT-S4Q-zVU zL)sO=aL09Q%-%YS@Hu6-ZfoR;6xf??Y?IggCjmqCuTbM196yYD!J6y#k~==8O1UXl zi~2?R^FxGaDGz3mcSO%w;#g6e2y0m0#Cgl?@ULShz+_*^PHDg%+zG|^hb>**X1X=q zx%(Xfn@v@w6vl5|)jiuF2@MpVx;zioa-D7O)tkU2|GgU&2Aw(xj`6Me`cr-V<1fL- z*$mIwGY+QaJ2APs;EPyX7kz(fUPdps9s7dH{Og^r)b+*0>l69Fb0c+?TD4?uk7WZe z&`bTI(6T$e1xh7<|0dHNt-#1~qZ2%5qu+G7J%bNtv&@Talo}tdCNW|;e>I#k!g;JX z4V&E_0ZSrGwaPAL41-6KYIJe^bu2CM)Cp{alP}9N_$Iy4U)htl^iwP7Kg6|q1M0Pk zyZi~8YV#Nm8`O!_nS%N%i)MF*Q6TI1hQglKuS#?tp=D@DHx_DWd z1f=8zoswd(&)+$WZ;dgn#$G?eI8N%Ye^kTL7nkKkMOhu@<+b;iY~YM(yb#QpF` zWLHp9)C7yZxk_=Yf0XVPo;+p^cjwUs3^ZN*qJ7`DF@wGkoWTPi6|3h_A4$=R@iOvbjU1Ar<5P#qo9m-5*sn(f*8jof|WsGk6p%D(ryI{;9X*cMwYG- zk&rFhgfrF#hDhkA#|2k@&4KGrPw}_oLR966^fNz{!5?N?+i`Y)0da#C=W}s&+_9%s zEaj{v@cqQwzmeYaeT7y#xOK1X|L|{FX+(O<-?F#DNND31u~nSNQaNS>JjxiNJ%-mR zSJ%PbjN?Sbwx+>RYV}JZJQ>g7Yz~|b-O_;W-O+xcV5h47XpkURGc4ZtiuviG=80 z3}#9KF%!>&vmJI_pV22?|Jm9(QWiX}+dQ0}x63-0fz!5Y55uXdmw=Z#jQ**Xfdbc{ z`#HP>bKrYHUsO*STrveH)g4VIAs~HudR|sB>NAU9pycAnQ7T!jikmA4uRjZ~WkGcTXE~PC=LOo8BoJ46Ibq{iN?ae-d z2>gl`Eu=RHUGp2YYRXjxs+8=Yw}(h`3uDcx%&-@{FC!N|(kU4k&V0}FsqH^ef0#Gk z8Fmy>Afv~M%TEGmp3>FCqoEQnOJKE?&rB`}KCDGVy+ho<$vEZ1w_{J6S*yl%us&va zlK`71r`NWiw+Ewkh_@zii>b}Fo0z%A{6HsR6!&Lzm=P-jLIm5fvE?4fsaT)GiW)sX z8bFF}P7eH5L23hcUV^~7|Am;RMMs~8epj6#JtSWTd&|L5R9RN1?elTmfA1u1Cvbhw z)i{ppF2jtSR#)$gy{HNie5v;~%g&Z2_9p2E^%8`BYeGf#;^B~gD>myBkGT{0jZi2& zIUf^A3MY-wd^_C=UM;&&`LmpV{fg)k0KW*~_b=+4TE_~&m$xBPpO(Foa95{ICUAsk z&wBLXr9}e$g;Vp&a;*P!=h*@L^5OU2B2SK7y()K-<0n5c)f=v3(~|>)5|4)B%YS$x zpSMEE<6M2s+*DuhgGfe<7JX-<$Dkj*$_;0-(tCaI7VmCn(O;P5u|-enO+gPY3%zw= z5i0fT7OHl`Q?f^I`Zlyhm9MrxZeHwnMqQICr{8)eJu=1|95%JU}aTWxHU+266gD3^@n2H?u z1tWuw3wtlyv9J)Df2H?vJPu8L`VRs+TMamJ(H$SK(#z-5o4Fx^2E7cD+d#$oQx4y= z1>yULPSMDBsX=8!pRduWg9d|VetBPUEE zH6wu2VxOFx{1Q`{ZxlTUkeC4j)_;L9os_|pqoL$mmrcLuY;T6h(J8AQ4j8eXhhUIs z5`E*s(76=mW5cMN)1y$bB12rgYqLgi*E6So{`vxx13y#^tEx*p&ECY#3>YiEW)8aE z^r)*2Hovj4brTQL271;EhH+ine>b0jEq>)i-SgN`E;!_)vwh+l>W~eXVyeuh8uy66 zRi8ai)UAee$0<1cCr|?hiD+VdkEPE;^)^VV%Iy4*smsMb1-?V2hv*LC!C`t)T_ZHI z6SNH)w#@x2+`&YsbUSJ}rET#oLN(6wi%em&-t@d5ZOrHg@;X2LnP$^8$yct_N?oY* zKl?Y+$QSg9^0xt{?LX$d7KmRzUD_w!BGugF@WtD*(QKK7O0VWD(@z}N)p?kHi3_o2 z`T%t{iCx^(jxHyayhLcwI!F6nt^DU?Qm6(s~dUkW7e$3Y#4gGD|+nRzlnd;DQT@ex%qgC#0_5FIU zqpBXeV|f+OkT3wL(s8iC!4@huRd?kC=sT=E+nor~Jey8T%>zD-c>2V3L;LBk6HwzH zNYJ@g4Z5*4J+DM>8h;e(7aeR5{y4r9oXuGS22{YkTCrRY?(T(l>fE;4l^P*A1muMR zi7Ozjvz~RIC`2yFVW*ANv6d=CSez`fb7dR8iVB}zLZh5n*JPOk%6 z@BT?QMqh*nAMxM|GrcKHSqMZ>$^EY)!$Tgx*}8D=tdbW&+R?YbmJrr?aD2JS_#003 z(|?c38=r$g#fYx^ckDsgA5V<~IT;+_et_@@?(K0RYc72vn`4=mW_$iRAhSL6`|j~7 zDKo?x^6NJamLV|!J{t(kYR5jD`!K*JNc9bQ7KAEW5}-H5?tBn5ULell=#p;7(zitY zj~#w=L%kP>VT$x;A{%?25P0*w+%XeFptDgf^MX)mDT@1WYJSgp_bC&K`{dg;#JGM2 z9-gd~CXAbb`i?89obo`vD*bgc5xgaZqf=&1ndQN2-@p)gNa*o1$AEIiDww78GGs_M zXocdgj^s$89~Wm)et`^@)NSp~vKrwd8ne@&E4jHxf1W(D8i{5g@te>K#SNh1V)% z*aIF{0l!W9MB3*gMZSIF6^y_wZJlHCO@{3-nW8=lM(5}b`VQlgc8_*BJ&-;xcmvSer}R z#ioEm%0|QU;P17E^3c8C?Blp@4U+>>du6F8Nc11*J>EUQq{>~G z7tBl#N*=@6V!K+wrgy(jaxw(=%m4jIhB!PR_2W07XEduc{JR}OPgg#)NX$Z#MjnNlJt(rb{xgY@$ft4rP9zx%pRwN zmh%SEOFkV;=EHV{g>L8{lU=zm`8vndhqt>`8C>J5+X=qv4wA>!v60dFAWoNwm1Rz{EJ{yQ1bfkXA$aiUw&Q01@CraHOMeW=U5_G4B zEfe7}zMl4gxJgwNg8y=eq{fCva=|Dh`ox#F+~s2maq-9inC7`&gUPNBOj<}o=;oaY z$tL&e<=n zmwYjs_sB9nRnV(wT|7UQSUwOo>)ZIGS1J90GD2N6K>JaORRWtTXLU9^y}m$hB>=k9 zBSQe#s%~p%>i`jkZ}}R$^<>f8R*0rnAN7j^O!u_nZz<#7^$m0573KAnS=8>)|C)@c ze^e3nkg3pGwuNP&Gvg3xCUyc%F~&D!-kYj^r(mTJd>C+B9wcs;Q3ZY|=Ya(Kpv zn5k>b-$i?rJP!vx!W?ZA6bLMuTb$Q`Cd-)p|8xMCg#HyLU{lo|C7BrTcixffseF&p zzFuz%>k~{lAlO~QKYbKQPtc3#hwP9;gI~88Azk0=|gCP#>(HdBTrb zJFO;6Qp)>LD;Zy}el2`_h9Bb!|bh&jrum0d_h2eW!`GMFo24Fuc9T{m~1=kx-KJd+UcN*|$WkUBmR8pJR5G(%{@^<1Pr1EvI zm~d!^_`?tm{`Jv5<}Scxk$$e42K{~}dyGte?y%u)2|jlb;(3H#8Qi03?1Hh>iUdQR z)hPT}Omic6R=I^e)eccQ z<~yY)oaN0MTO;!Ge2;cd*vL`qNH=yjiT4>UVYSDwDf7w_SVqvZ-B!S}e-^43;YTTl zimylVe5_=OBfos4w}jMh{HS{LvTo)EC^~Tq`LUK2F=;02Hc<&jf5MJT#uZyV>>T@^ zWU)P#?+x{Xnfj}bR;7yn=S283lL$|Jf57j|$j?3WvS3iTH#eS3E3|TAGq^TYTv~YC z$@{GH@lss>v){$@%mZQ62_^)!cirpCsbt({vX4`VZp+=b-He*qX3tp;=A_-JfP~?u3tC#u@*os`^sVFh5qG zI7Q4<{H&E49-il>qKERb6svfAN(>zqXh=ZB9(tV&KaaM^*;Xo+%1sGNmpT2F1Kno1 z>%sD=oX@MslKOb(l|M#!mh0I2Z0DCeRsGnya|D<2{_Iefb5jZBT|o7PCUJEOL)5ay z-0kx!#lWK!o!ePNi+((H<6u8xhgse0$|`ZRLmA}6`%cS1iszpxu-Y47YSz3& zSqI8aUOu1QjlEwI-P5}N&HAbee$d(R=ZM}EknL;L-Gq_j`?=>y_<137bOaELEILq8 z#y?lJi|TR$KIigyyuYEOuiXk^2A?!XA?JxAc~MA|e#mLdH9siqQoAu9!&v2U^+R1f zC;`slH4piFI}09gZ5K@Pa(gqLhzOtf1AWUp6_1#I`OyHoa<(H|+$WBlLij~YX;nAt z4}%;3LhG;O7y{m37EY9u-$X*L`(MlZSzLrIApfacyTGsiG#M^DP=<8L9nbJb10&P( zy-Z;GO2)}6AydE)vk6+Z`(Gbzw=#G|k5Y1h^gPkyKXm!{a2L1HLv9UIDIcJC`{URE z^7Gp1rZjO@`wo@!eDTJo(Mnms%Pgr5?mw+PxhdJK^bwCRqhVAk5Oe#h`Yhomh7sB{ z?~?;al84N^m@+z0_r>;hlk8JK+bS;FwFh|qkA|r zC2R1d`pFRI9C=h(TexUAhKh4FbeiE9hW+0t0eZo z!s2`e#TIBXzIo#u-$JX|{tJx};7B6bX2C!FOM-<(g_7|7%zOV{&8U?3g%xJU@ATZp zB(HjY_DAV&t`4xwy?CdA>R^YMI`y>Tcn3P-Y=8v}eAEa3Lgiysk5WU2?Ux{7U9!kU zhTxIL%P<%RU5XMnQ~oJEthFBhI3W`I2WB}z%v!+e>gR7OGA z4s8SB=P|9?q?lLm`~=)7_g3-`1cHi4Hi4>lW&rE7e$bIJz7dMdi+q4#=* z*?i)z&-v-T`s@e&^ui+HO^NPd$%3m$x^+-I- zPC8pfrw|-~*!S!AlM#b2nDqRh!M;dkk^R&2#CYB1?z4LVBg2W~wvzObXT;wX-^EyR zIL*E(M#lsCdj(=B>P9LAf}-Bh7q8}2fY3ax43UW5&RX-8JL~#0fvhD)>k@r>KQDdy z8abs%?=48VXsPnV5OCoN|9I~#^t~K9-Ib>QJq}gf15v* zC-5BX-EUj_H9nM^eq@=dD;tYOZ9K1ZxGKf}Gp z@%~)OR=FvTsa(FNKZf&AeH#h^iWK(Q1zLx#Y3KZ<0-cC`W&H2_Oye$U^=;lqSiiSQ z^aVOXS!JxA{_AAnxm(+5LrDoP!?u(zw}B}Vv>!Gq=HTX(j!$h;aJa!=^_;Po`CC&8GNS#0=l+41 zm=2)( zTpR+zC=VZgrOlZ_=|Vp0PwxX3jq8fV;(3M)$HK{{^-(O@KG})7WeDY+)ymUmtM9Jo zxWD^+{+!4-PJ1Oy&IHbaM!_^PlLEARlvGl*E`6NsDtpky-F0hY<3j& zePvh4*9oAQho>Dm)s?+*ZeQI-hfo|+rQ8ze0qtSpu{oWAgA0CBZA9#Ti#^SdTdu!W zID-iKZR~vjdHtlO{1U>z33z<;3yTQtw0XWq>ZasY_9%TiHCU9qKCiwEkQ;?y0xQX4gebSMc{q_DZvmExgnhosLfUGafw1x*+ zaE$Ot75%>g_&mc291xK3qr4qiF(VY33cplpBShW8?HE49u)ALy$=3wFzU&V>5^=@k zupQTtp(cMW-hmrfOuag(SpaV_aXD7JOXSlG5a+vf=|4qO$HGkDWl;~A&pw>}OnfQt zp3Rl*E4r2{`Fvk@m1;jGhiXz*aW6>~J*^9E{;XFIsuh{ZPw2Mtr9Wrs6KqDwAL+MT zMr9L${t4wo&7p*DPXT(9o25f>k=M=T-?*Hr{R*i%n%Li2%}dV%IE7q&eTpQr+b>(` z4SmZlv(Zy5Pfs=DI?G{GiA37yZH4`wM4D__^Ki?Sn zuaq~sIPu@V$K<=(^yZ;kOMSiW#7}ddzF#*%xBZc}RKRU3wAbq`4uh|y9?4%BKMh|K z$aaTIls8O)iCSnLA9V)>f|Ve5V>cGs)nz5PJ%qXS(gWE}qqg7MZnGeJeb}TO-Uh_gNh4nK``e1C&*59zA#qPSZ+qvHvvzjrZ=vvZym&hxEb7wE!P zwj<%woiFH?YX-fVxf|RWA9bs7>}8yBqjG$-QzI(k5Y2vv;P3NJo}82oHU%C*lLA*s zxf2)4wk?$m4Smnx)gC_e5aCBmdP1*V8ljC?`CnSGBqGvU7RBfX^ol~2NWb!m-FU7` zB7%P3f14}wlI&B;;nQ}5y1?d1FyBmh2tG~;^KaADCxug`1N`t2=(3>!=sQ0#76EsdAb>{pF)n`^*O;et_bDbCIr;=|f zx?gj_pbmW-;iWj4t3#yc7YDM4KsKFAfAzG?lWj}3b00hBRZ7!wRNru=j(B*CUqFR9 zF#Ahr|7$*6E}wTN!GZxY#ZTQf;t_5qKi?ENk!j;K&y?Et9ZX4p$2k7`-=;0k*T26w zJTrGW%bjRbToj`w;+gYb!fE{LO^6%Z@!44E3urQA1EM+MoF8QBsVarkm(~k?LDQvn0EU_pPc2f~-{O zIN=WsFBrPTK(QjFwx_ji)LR3o{0_;<>IEa)wb{o`zQt|~Fk+D?8T~@S&3wD7@E3Umi zwuf$K#~+lN*LBqM)2(U5!V)iKhynb2hN376omwtqHAMQI-+A6-#kDY0lu2jY%MJg$ z>I;>l!^t({yd0+0g>@&HTOZ~r+0mSX?zm{fnIZ2@@U+^Pyx=t-TK)ZZKd12kNp)f= zfdfqc{3{LAR}CigTeX|RJ4rF70w}7cy(L@Mpgwf49o$KD&-o*`ZRv3S+io}5F1G z7&iQednehPlj6|{b1SDhN_v<$J+^mFsMRMF@6C&={nrD1^ zD9bzo^dBeuo^)F6(4Gc+<i)F#(%;AY# z^jjJ@f(GD$3VG%P7}_O?r?g}1e3!5rhZE~ajbtS&auHiN6KHWkLJ*=x&xtB!=t_*{ z*+;_M_^p2|!D8NopK<*Ufu2k5Wggy~bV{4oY?S~=j+c&f*f@!@3rFnPwA?SxlmByO zoA{Ey>x_IC=r|P0;ePFpRPz9nSWfNNgRcqTPmz=M@T+VWqe7wdi^FOU5oLTHHHQ(0 z{U<01P3I66Uq>PQa^nq_!POkmSJDOzQJHt5ds@*noboyi)Nex^H@mq>?tsD4uSKLD@ZZ*Ee| zyt{K{EFUp%`l_;Bhs*3yGaBODU(b3>yv|z1`Y06BdhW8xeRmOF6u|+b6HLIAa>~+g z>9700!$amA*^bYT13r7m(8kC$86o8$WqJbmJq-R(dBnp* z#hOob=AP0g^*zif!R-s=^!=#3esnebXH&$ZL7F1jmFH;@^J#w6_*fKIhn#YOMGD$9MjyJQNE73tH_1JgA_e~|vywnT z_-RYkh{2Vd)1w|W)1sJGy_m1cuGAy>v^^YMf+iN~v3+smNe6OvUcUNb=ls~AVBL!A zOGXi5dNusi4;Oqz{++Q7X#2E5OhRN99!@-LVP|BA#cbj`9bi^@DEcimDsm=Kg-0x#zJcDlhZAoc+FL1MV88$i!+VPk<*87G$?wTiB ziqC-l5~FEp)qk>6T0oaRKP-d@jW*&EbF$FmeHF8DLQYHe@<*x7SZyUKC(KQ7sj*p5;-zgops}EK24y z&9h~Dp!uXEkk7p6QN;Y}kC1_E5f1iGy7aFH-5!aQN<+9w(J-}gSoe+{)c({(S9>@t<=ZrmcVxA-PfzowZJa7_vzYl5=&dt1oTo znt}UizjdF3d>hk181sa+JNNOYRnjTTn>#o=xXcl@2+4Y!c_eFE0?gWZeW@B1j2mIfn-?tp@-Xhe26eQy+E5L#zb#F9Z7s^o|IA zkgPnxD0))U6<rqr4H`%H)c} zIo%Q$AbgJJghA=W3u8dj)zTN>m}Nz}Fq=LY!aHGoR2ONaIS)$qdo@BZ%XP+TL)_4X zQQS`9xnp9L#Qpt3jU{&ijc>yGm*XiT)}rGjHon29Da8d9+P;bo3b=mvf+uK9A^Mt$6p7!Nn@tebcY|K@Ipt5ctp3@q)A93|cr-f7po2+#CL5$9 zV<8RTQ;;p^CmA1g%AEnXA|^~}(|*&bZOh7?1qICV(P%Hj*3hLLDYyOS`27|)^tHCb zaiJ`gT;Q1&wZTKP{dgEQ8$ApE(rs<)FrUk}Tsb z_1RZ7g)Xr8>W^V}ZhMzEUV#1trLJ1SudW0a>^uJC2#ft!j@ybj;lBE%$%^dR4j#sH zis-{T|05W&b!gz2juZt51=N}|nB|Jm>3GDT?N;;xR&C`atB*RaE?wm|ytBPpUb%{+ zUFXb)f7swBSFU1~Gh6m$=(lWVqq(6sqccJ=rO#|w87H0fOt0MEdQ{@8+QEZqg{i4#SZ!KbS;`9y)zmKyy&_hw01aAsr$JRf>Rg)ZLa41cZ=mfc4^rX}wb(2jn}HhfsM4sj65&ry404$Pj6Zs0}i7FCTJTrcfQ^z5}#YS z<&soB#|ynjAIR3C`WnBve81e%+DBau@YvW(K}FN;6D{u~y1!i+)AEF0W$ngR6R%Hc zCz}+DeW;o0&sz}@+wal)FSzChuDl(XK0@e3b%SBCaGEfQIq(Lona>e_MAPQ$}Ya(|3lB>~Wx9Wzr+nx0o(qSmWg(^bT0< zHu1ccChhip@G&1|#>j#J*nBmH>Tq+8SNYCZ`+ET`-Zy zX{+#uBf4ezzw7V3Y5MPWXZk8Z3RX#fcBbv({FC}EjR-HL z?XPEIOT>qQ=j*daN!Y(18+Fh{;^5yz>z1e2K&{E1Yy_DJ(eA|F{O0{N@3xOIWHj%=RW;G2)S{TR9qk;NLhHLA^O0vYxDtHl^Um z^XE8lIEWp%DMxaf3s1>5sV*0)Lc@>!DKX@MRkia^g0d;E5dNGyv~bBt=kw()HtUvM+zbe+O0Gonn zR{bgvuZaG&ATW3Ml9P1NJ))_kFzlcSqNLLL?aIfBVl$dXjb!VC;?({GR~dBZgoWQu zpu-S5e>^VWmL8CAJ>zb_pYowjr?;z|?)QVFr}F5r-!jjtm8p9JhhCWIFd76crSd#> zmD7&I^4d!n0ZBdtFl5_iDBLKc??EkNmiIgF^&bfYLd~mGDJ!1yI5TO#L|fyiR6uAM0y?EC5Z;~L;+EOzF4FtZ8ayKA)_ zA8syqa$54!%bBWYd|0Fc^{OM&m8+9-FqP{xjc(-8kYQkJks~8W`C???RX49BEnu}j z5F(yT5j>nqYJ8rSvQda2B=#2Q? zN1BXVak{s^MRs}KnYM`JyGmnhKL7;UrN(?orwvy(x?f1;SjQ$BGzn}~Ycz$(G^>_7 z6nO$SzZZnD-}TzEg{MBWncp(w_#wTODrB4$sL47q-Qe11Vgp=`dX_cdR=@LE`v+b- zTZFI7lqsjS&_^`+p=m5lmw#Lzhu#vY?6<*%Kj#D)121R+uUyWxKb3z6xZUb|ST`0( zd*Ogv@J<@8CI^^t74TL-hu)RYEoLxs-pGG%N|JOun;U)IZozBwlIeuMEcFwpDZ{}` zRmmfET6TPz?Aib6$+r40EHM47Xudc|DCWu}wNAb&iD4KBpJm9lFQ(G#**k{!Nl_77 zFW_nnvc1_1l`Jm(}(#fv*&ulzbmurMWj2`G5x<`ZZ;& z-nztx<$%{$g`|!eFtE;zb)`u)@qTTGM@JOoS4E9V!LL9%_#+}1amNXxi^Zv(cL|f> z@v8dNVyVt19N(C&%&n&hKpEY5ebdB@DrMZn@)vbuL6hRtRp&xyqScx%`{8rouohZt zZ$e8E^VN4EF#IjT9}^dJ{SIo0x$w@$D}xYpPFb_o=MjLtv4Bkr=J^Z^W;J}228f$+ z_?XlIVy>QY*h5hJiRZ51P5~(8VgRmeI{o*Je$0)Jb==6Ptq-_Twp|;)_V6>C@S&@$h7Y>L+-UyZHX4QAz^Ydbsy~F|C%dk4zlvZx zo0nlO)Bto*S*5t81o=%Hs&`_l*-wds2dMeA%70%} zeBHky?H5r&IYJrE+er)XBl1L`LTQTUZ9x&7>82#LeCbIopBMo*=|;@vBwgZGK+00* zhCKoNo2109lu=FB$cenTDJ8ZfY6=uOpe-dJ{@J#JW>w@rYj2q~#L7}PuL9Mstazkq z#>HC%ZcOz>N~T+uHXbjb8hUr7lI$VR9`7#e$Fuvf<3*h%KNsRBg%N+%ScJ`KHpfF* zf2#rSlINm)!)2dN73^0}T*Odgd3 zL{eu^m_-^uy%o(~YXUd2FQDs1EvB1ojA4c4RkGAo95h!5_?HTm(Nx6@D*uIA(KsJ| z%K4jY6aArMT~wLiz9&tMi+KIxsDk`N1;-vzQf3j7%K9q`=a)s!1DWX18DK zd+J>H8Q6QpX+}r|2yDxFHpc6iUw6a`QwB@4Y%czX=ja%X<*=`r8awkM`i5Y+Y?>p` zW7oJC(ttxJT5NnF)%*7%GYPBD^d#*!K5_g}yeQcw7`*s#JX3sLl+mj>?)<`iyE)K( zygP0V#O;hi2J>DQTLK@4%-9V1B~6MF{#d;>k4AYhj1<{3=znn#bbqcqppFACoz}~8 zA$%&cMH007I~W{VJoCj#oiDsiYZ68wEB9j&2EdRI`FY}!?gU)92-62!?ztJ>q#+YV z&5;g>lMohJH3K;!l~+*2vCtWgh+1QL*}q@j{*Zq$VZ6hUt^WMRSo&1Xro)^WM+Qut z)0AfO-C6uEr{*}J_+M$*_ash#dS87;Os}%>wMjrF_3t`=9<;)DKH(5mg9e%J-QYko ze`iqs+W=}l=jIl3c-kq^9~GPwmJgql($DdJ_8>-Kv#w8j){*2^cTlcX0-@7Lyd;-rk!v*381j2l<;QcUH)vfz$q2V#Vp z{!-&&EB%CYaAgQBw#I^=rQxO#TVd_{D4*&)k5Ag7OMN)B(>`w zmRzfr1!5*@QiU;JY-u+2U*ATlQyX;PR3VDlB3h^1-&XRlJLO_MROny(zMoq^DihWM zuQFTp>8)%A#^%)uY+hzqYM1Eu2yXQRqN8XkISmJ+)qmWAow&K!)?R@iWq1gX1bAV}aS10`g+(fg`aA*D7c7K$ zaSb7ydPS+O5!ZEj8OW!r@I>1UHA*WtPnP6%?pK4WrV;6YekzMNb?|Jg3LR#p zuwH%Bn8rE7=*40~wg_Aud!?I^xzOQp{ck))9iXXF=>cQ4WRjVt%o2q*9!sSr@nc9oy5Td36b>ew5iOWQ+lMdUND5Idc zo-2s7@;e86NzoTa$#U$12+oV#Yb`;jD-DYO_?%o|V(=gH_Z6LcxwUyM1K0KL*YZUQ zzkGyRY-2UVFKxPI0wKAdxF|F1Ptai~c8A4e9IrnoNjpClUNOD!Uvnw;y;S8gw^(4y z-!5*_!9$kXaLf}_-whET12LuP_x@JCQS}<;sDex(3$Us@qHl(p*0=nWi8#DPIG zNBrbbPC}=2a_94;6*MQ|yd5+KSM{}J&cSY9VG+X)jW6_DsIwNm_k(5Ud0VgS+y!)~ z{jW;Mw<`fUq}UteXqso6&^d;GiujCDp!MN>P%( zAAL>uSC-~T{PRRD(pyUzp9^lAb0phbY~>n5mrhx#clt9k_WD4w0A&a(X~s{o`a=x9 z?g#Q)H(}a1)J{^unxekvzQrbR&YZjK{xCW~_@YIybag(QCTy#3U6iMOC%S5#tsJ+c zmNXiy0Hneh9-Vh1v*k5hg3UchdB5>m=i#^@anLXOr5S=%@O~GN$=Eu4H_4`QR1%fL z;KLNS>cwkikz0E@s1O*0@#Vm6An7N=phZ3cyM7V1uKniq8-=x{y816Qq2wPrYxD3^6Z_9ADIP)>-j0;`zB{7M* zHHKkLEI#(|h@#_7>J+m_Gu^FfBJd4{ZgM&pqjGecq?E#?rn(?Pd-#4G-sbIIJ8 zUNSNFsb0sMMU|G-WAmd%|4Wtbl2W}m)rfdY9d?$y^XQh>bAk(gG1fe_WHW!lA0Ao* zRs2qc{;0;2frAJ>qRZ!^?W8#-l_j<5v{vMvkkyd(GujJN@0^cKpI_{Jl9$42^GUv$M{I(~xgkS2LW|tBLYe+Oalu__>Zn3NW!2=n<@ZzWYv0_b z=$ziZ;k!f}xuERcYMDI7U?8!dezuWY%uK#$-)9K&@!Gu($K*=*)x=ZAtsLaPG@+$t zx%iOG0XqCQ=TCT3dh&EI^+_pCxIyN>VStzy6i3%Y1zud?wABav>}f=50ixoNCJ2ji zZk;sWw>Bxq<<+c52FYW3Yx?d|Y0J-Ou8eYf)j2r8!y1C5{8WRQmA1~~-sSQm{q&evMI+Nhr}J&yjb-` zG$?=aWsIT#m|L*4)A<)seR`po@>4HMtM)Spd@2we(*&Q$hnrNuTOTH-AuPFHm4gEh ze^G(It3Sl1!kq=p3u&3ZqmsDIv9U#+xdlbSpTA_W80eAc`RSexcc&l--L=)WkXxFT&0Lc5R%*(S7KQieba&G%Z zH$6%uU(w4lQrAwy?x46ge}{>C@>8k?Z3PZYukglF9Y%@FXug81(M4o$>MqJhH7$iwqT%xK#D4-As-p*wA>bp9fqgUpU0K?pIcE9 zBQS# z!Uc`DSn7>C_)*a=*4?YEYwMfD(PpT0@fNj0J;r1$_63^W%%0 zq%e`$(ikj8@r_?)HAl$p__51DbG;;o(VZLJ_r zsW@RGkKzB-mGgmqN`8;qx@!*|sP+27*)NbC!oE%0eg#kYbQP9HK{w9~!an8T3uj%ZC`Vvc5;h(sY&4Brske9vL6St$AzjsNb*6 zeN4EN{9b+uJG@h{R4y?f@z~(8K~^S6`>M$tLVdn07c5x-=5w&`8-eXH!wN_B*Ilp; zC*faK*e2dkPg%dCpOfR@p(OA;rPCJZdtLP09P;>s(N9Lp2S0JVNiMoA#JHH37o>F31`yy&+mpD{u;Nw}WGbK!Mng z7JH}VtD^J7&?ncU<;h31Mt1Oo*q|BBh}wiTNQp9c^Pu6&*~kCx5dGWom7ecq`Wa1Y zaEy8;`cp>!9WhwAep{t2aM||n?fbR!ZixaH#WO`3J%XPIfl3pJ*ltUpx)KcQIfeMH zKYnqFx)-9C##G11D0?}G)089S_axFSbQA*7aNXEQ56R6$-;KJ>O>*?J&Gbp} zNA{NpyrTi_)IO=s}GZz9uS%4JF>yjU^is!2I1iFWW-E)`7g_4lba z$ZtCZ_DvLkVBviewf^4MKX2LU8+LOawbObrHOEd-;v)-3pMbv2CJZ%e?05p{PYfVDPbAFc~DEn?4OU6Yl zP;gvFB{L;LcruaG4xX2#nQnoNYk6a@lrlBQpw1c6y6yWhVc!I6l6&^=Z8)0uhdZ$v z&K7PV2YKtY&1pu zWa|$21c81T4EOEU7P)s*kQ#rTqzAvbDM$B6x>OJv1U$# zr%%Q94Oc;>7gLEGrmBpp;lIxHhk;q=ok<8xevvB{^Y9J%5aD*b%9xJZS3GNC$3t3t zbH#*VktKCpcnF7fP*ECK;)(6vh9IVvfD@yY<~T`u+Xk>>ukyMXY7pAhc`(9-4;-ji zP6LTbd>@D{N6poy2tODju>lX^<33HhF#|4)#=2sM1KK*c>UUsVji;z+%A_0U{|4|H z+G3l0i97FK2Ry)@&hOrARBbjLG|)5#DwcTORkJQX^tEEcS3X>>?;ez z^#0XiMNoG08fzyz1acw0F7K)*iyML=U;j;WXYNlf{K=m~GU@UvLa-0q&%*6j{e81} znHjSZ*#Biv=18{r43g9bcAh7+Ywb-Vhn>enF$M*=ePf3Etyi(sh`kj_V0Ap?dB*g< z4(*KzUczkliAWawCn_6Td58)lj{?L{eh8{z2cnu3CO~p>7 zgVp1db;206(PDaeR?3<#VFy+A$}{W_5%Yfz+wiE*r!fTy49%*noUx#67KL!gD=2fv ze;x1V*Ihb*lS%IE+9Tk?_*iena!v2sM+GOR=fLW1({o_sEmp(nqxPEDr3}QQ{WQ9$ z6L?DUtf>t+7y0rbKy10r=}kpWILM9k#qW3qzrI}qKUIePrC{GhIQJZA{&U)vmtc5U z$8b-S7>EOhfVkN;!38q}0rZy?hP`r|egq-3Z{$BU#3LPbGcJA^{0T(mq;&J9zxxTA z>rCQOrrgWf9m>13^!4-cs_7+?C#1ABUkg4`9?0zHOfr@;l?Ox5$NU?1PG7$XsZA4uAw z0RwGi4+&R^J3UFTWE~#jj;$9X3Gs{XA8@=LMplup9o-XBQ!^4duPkJ_*~DE`udTMBGd zYCMA=?uvgIeB4!TnxNTVGM#}g_z3KIQlhuAq}c>rw%h-9yt9*Vu{jPNEqs7L2|w0P z(CJqk5u+y}CTsYAKRccG>^4U{vBdN)cm-(?AT|8Eed%P>9RY#u8KCWa!}$>q_a%t2 zSv6vX3;J(#w<-O=AIJSW*^64l-MIXZ31j^KZXHjojgwqO6pxrY6deA`C%4Q~GZ;()TR;J86-6=@npH66?8W2dTEL(7 z{r8-9zA5+$mgmz*?fiRwgg=u5|CtT)<>EePD+az}xAP^RKBSQAGp!c|RhTn=tviRA zM(%R_#GUr#LUpI9Bh>=(FMU<%TQI(1-wVKgYvREcDS2iPByC)NUZM;HtKAg(GPvz} z@*K2TUrc;eu}nHzye6MHjhx_6m!Src%bcUytKgT`naek^E^SAunUpjTt1kigSt$jR z+`iI9y_nvQ@v&h_e&U%Ou`hz@RqMuivn}|Z2NQnNMIFNH+0t>|phMO5XB59X^83l2 z0nnQrrnWaNO98xPNU07*PBY3oJEqmu`&;5;g!!BaCq*U;6m(CL|LnxMh{$)|JUj~6 ztHt>J?*SAJKQ(_-lLh!bcv5UjaeRu&5JYlb#-A)r+<{Q!|IB<;TOnoK{qy!b?FViF zkjQ2J1WeLRub&3>;1yg&y;=l0^Hdx6hjkt&aF{$>KbwC;zQX{!8GOhWV}2V z0Wx69(U)x$%PR2|T#$d$zPvks9o=QG;^b<|k^iAopO=V;VX##IaMsH~YTy&+;)s0% zbr4n?v9AD}ar>%!sBn@|!NG;n^Fpr$Nur4g8=??1j!`vEQ!sGYC5h1t3=H}1RXm-s zl6QL*Jb2tdgcp;m?vk%w0elf*qB{;)mkgS;rgO!?Dd=aZ0309hMuE?1{5;{L$;&d! z=iY|?0cfp16}t)^?MJXaUIksor&du!*l+l`x5^H;`QBWf22xfw$5)1H?(BICFpB!Pz`(aRB*v zA7|*kRzl_0S;-APLESTO1pOV%0J8s8vukcN#0q7qo^q)Xk7i%Ij{$4X(1&MF|L*l%p-A(i1H1!vUAoQQ4EEpc3%fQ;@2P%7jM@gq=Dzltu0=Rbq?KacZK7yCD*$ofIl8DKHc6l0GB5oYLkZc+_r~Uo`4Rj+Iby$D{J$qc)6ASw7^yy6stALR>KS%lF&pKY%;W zBPpQdojI>1N;nyLczT(dd|Fl4`2kWMUvdqTG`93%_%NPo9q05z4?vL#{k=rK=o*?7IDX?v7qx=?R5L27#xS zX~_a{Est5i|M-s|wc-Ri*HHXsFiEwg@oaqxnq&Q5`7pL2WTM}z*dAL%pCN(VA>znQMSVK7MMZ^Ewuk8k#g((8NMtaME{0i$3(&RF{^ z(CK`RsrCo%A(kHTcds`fsIhvvIZg3p9L#57mAVBAzauez=`F^SYPTk>CIv-&q#4Rj zmx%e8&g*4UzJhl*Jk%z>R9J4mq?cn1Yxup?T3R85(F{d`6DRN9>@?dt!H}vJv#dXP zcq77^O8l@bRpA`}t9%I@m2x+KZp5$ah-9hdb z=`9z8J;ws&uh8_7r}$=14f(#`aX0qEd#N;vdT@I(f|H5b)yvDw$J`PBnH#A$0frbE zMNEkz2mWl#+4{o6(QmQ?xRlUV?ak}fj4uzIhXGfc@-UsH zo5!rJz>>yf@mBL^)QfurzT5|?1*a3*Z$2!~{->S>fk#4kF?9-+m}G@y#Rm@0e*!0& zi^W|!w>qh{zSOC*8t9Xfzr@nEc3x<=(d(DCc2n%dOlbDiWZ=}+RZZ$en%eoA^0e?j zcDc9dZ~jX3PrJS3Z@#jz?_{&0_Ta(E0uHn^y&Dw(bM_yL@SN~4ay>Jf6&Ch-ch#-_ z4Sui`?EN2h)7k=+OL;O}*|KG{TS)gk5`DIRTcy09a79*)0BVvBO+J6wIv$|4L^` zR&T7q@b8{>Fcbqe10<>>jCku$D2zsB~;TFEHKo&Z1n z!jVkO#JLRTSD2T3(Yi5V;!%G@102qp&2KbqRAJfw>Zu8PluG?vZ^A(GB0l=TrXntK zmA%G9aEJFuW^>*dU2$nyiMVE<^+dCntd*}mH#Sq+uBQcQcdy>83y(Vh3%tH-*8smQ6R6YrohI`d~!&22QM`xt&(-LlHtNHg* z=+Rpbk*kW{G~>({2KIoX+f#1WJ_!wmZz;Ao_ega{jtG@M7N^Wndt@UfZYf5_F+WOe zD8&eE$RX@FK%4{WOV?2pRgS#o_bxE$F)nc!v!r!Cj&#FxI?jt}zP==wEx-g6@_G<}xUfa{lQuf`EtFP%>;ya9V8-XncD z>ZHFcQ(Rr(Zl%|+l{97{?6+^!MhE8i;I!G10lX0%rW3!k$l@;;gJRxnjb-<5^2F9> z;{2p}_D6X4EoWx}`@ZtoPcSE*&;&N$P!2-sj4EGe$%lWa?vTokyJx7lHtlSOGBLL$ z2IyB;WljFTeSKQwfqSEAcTiv*SP;%6$I3<$a3VCXtm$m3Vmd-4^WvKJM}Op^aox^V z!xEX87Ov`W&+EZqJfJ>FcwYW%TNI3?$^1X{dXG{&I*g`YSb|aKz{hsWKHm(g2Y&=2 zqRDb#B>(!Ifoom#4{#IZuTB1B8a)0MtDdPIYhyN!hF%mpDKI z#dka-VK?av_uX~-kDYsq9_VZ@J=2^WiVfAng(?nv=64lzYgd1`kCmp<$b!Fgt>~u6 z-p=i_d>!(RTOW)E8sVTWL1X+I_B2LfL-dsQE{%x8_2;)tuOS!v1xi(-65U7$@Xr>I zCgVsL2yGA;b3yb?p9fi6w`%4C33+=GdpwtzFT>A~4`%AV~Kn-dS61|5gxsC$V%$TgeY{2~M3bq6 zirTFFSG|mP9|kVFJ&NT!efPwL`M*dXBLjds`)Y>IL`C@7OppoT$XtoRY0qQt9MYq3b zmqwq2xGv8=RuQGY);n)=x#cI2!>u04K$cJF_oYmieF=Q-5F0lJSX@g5FDp1$$zbgM(@mUQ3-`1%bH9+{SG_q z1qA23bsiUO4s6FY9-C|={^y^{ge)N(XCKj;;676w zmPR1Ek&qN6@NiV5c}M=RX%gP|ItQ0vF&a`EfAbA49-^rCWM@5QU2=K9u=80dX2O+9 zd@BZ>sw#wa3~mlo+YlXM0aMh_;DtEC$-$~P$~k6E1OUWxuPN=)cLqnfPqZ5=L#uqm%pXa95-t5oXMZdj@^ zxB^1H!&&07-(JBdp8RpG{D216A};=-Cm|U6Hi)0&S)sCSC->v11pSM?e)HiZ@`T2H7{6-JPYvp> z#r(nlQg5sY_q-(g>1?;!B}7Jic|QdQUMz_pbgWoPdR7hPFQ7K^HX=nE7i{!eK2$>? z@4Rldq~TX+-jS2J8+R-+?A61Up zQ%YCJs9#z52gQ581CtEibMD?EW?ZKpi8;{ir#J7qe9zVxyS{hp6Ye2-sND^Z%zvpl z@TUA1byEc0#sxLu8&&UWpefwtX`76-a@Y67bvc~;IF{`NzyMqQb%(d6hvtO4BKFL? zcV7sSI;WpLzJA^Y=UMZ`RY=;63R7pvc}zvdfr+}f#)iSnCuQ47Z;jU;Xsl^1D~Igf z-S)Vp+bq|NG_ekn-Xnb%^lH0#S9TvB;IQkW&`J=f59e|}EGUQ{+$!ebhXCof7C%v? zFzTqd%e~!C;J-$E%W}G#|E}U%$wkmd@F{tb zhC6WN3`Ule^LIZZqvRJ}d!zu@(70&LL_CDbZsZTK=6ZajG1#{#y+M|GZk zfJI&z&pJEfWTOgmvdiHtJ}v(V5|7feIx+Bh;e5}bnD1H-9cm~NnD4-vCL<$Cl`SQl zun+bLT+r!}0{sn;3{p#a&YqOeFUmF9Qh8VGDlX%v?8%Q*{K&WDIjwH56@1ZAb`1?! z(I8rFPAXvbX`v$H~zG1?{r?z!rYPYZGZQkzMhin z_u3qtlkkXIgOIw=I8&S3l<+|^4jqge-S?a2kJ{B9yvV~+Zvm((w>`U@`lM<7p%`rE zIY$qJC0G@_XlwkMTryivnQT+um%(Z9wTpjWx#rkFgXth~*4xKln**688T)8{(SX$&&&-tchJcIFZ|U`hfuxZ9MVzwtDE|vk zlW8zuf;N1kL!D;ID<@TiE8qE z^?tx%8*uBt`f6x=_t%Hu##^vq$Cvl?XHzxVa=1A(3?}epkktdr2+A@x%WnH#5iH?> zg5LANNY6_6u7qF(3n(i~EV+VXX*ZTF9q_N){PGZNC{tlcm*cXw{A`eStIDm3MPL&5 zkm7N@@Q{V|hhvvGSH0d{K5+1}XqJWN3_$`VatrI~$_B3eq)ugO-WgoF|8sK$m#HKO z@|T|bS5k)*zvl`Q)vckUb6mfqF-HtlM%=)HUi)CNG-H!0gks{=ew#qQPITzxRE)Ys z(dKv8nK9JFdub#^!hT5ilp5knL0jV(OO5h!4K$v|84C6>9~)zSX6QplAvfxU;3JJ! z+vy5(PwH!+_m_{?fx*<3ym@i~A99iE#g>(%f*+khv$P(zl~b`Y)vyQr=qQ9ugzCCG1W2gbGbx24l_Q2B<1SWf{tV^%%6}^>fR`BJz$$ za2h#*j8~rDR-fr2PoLC)Y|h0gTJoLwHT7z`bJmvT#~(WslNld$rz+y^@B26s*uB1M zED}FA?mT|0SoY_)r1`h6?OEn*KL0)Lp`MUyieR!vOIDL}ls5QcEUUveeDKA-h4%=m z!|!hK6Ck_PQK@i+S#cg!xm)-xm+Z3Yji1u&w~3g=9@?Es=N?vRYuFchu;K$~ef^rK zljGE3q`o@d6bFUn+r?Za^ehQ3!S47|oG`mwa^Cr~=_a660&%TgF1c3=;oksx3e@Cr zy~sUvvtz2{z%`EjgcADfQ|+sv_w_ddS7Yi+L^>3b3nJD}ocV25y3Hiycx_fd?V%4& zHhf0?#u0}_K*c#E=!Ows8pdgNrG;IhfF@&uAyWTP8~jrYRIq_D+`M_t1fx%h90#aw zl1an@b<@6G2GrsV3cEA|o`z3LV!_go9)g``UL09-U)8avbelvEToAP`SSVHTjF&vo z?Z0K!BV((#6%*UxUY}%1IT{usP||klYj=n^{kyk>TgiaG(W|+{XQEpT{b4^< z8nZis^Q9PLL5PwNISqZ{2kFKvB35#R{^OGK7CDY4ZHr?%qdS!~4W8a){C5I&%t3X!NOIsT4ji0a&>)NBK6&u-m2@Z#=Lp)`k?+3ZWH!tf{x`OFGayUzjioj>DDCu zjd8q&ggX%)y#tkeo#%)xA3cO97NZXR!^yh9zi%T5SrYzLV_j6k6+L6k(8@JXy?oEQ zWA22Qhp^MD$>f^*ybMmf1X^Y8n93#7f@`6dO9#qCH+!FNgR4BBu|R|b?yv@a%HCW$ zu!TQ>xs?te-g1k#a3H(SrRb6pw|$o~ck@r@&*od=$ozl~zJ>c@PdVa9)fz?zo}zZqmd@6eX7RrS zn49+bMtm?=;9l2y_d#zBk6Z3yHYTys=;%B{|ap1ma zSmPNaFT)EOp~@mD2#xp_vy4~;9;=04s9XhIU&gr+cfyE>)&!fL0cBP<0RF9en=g}$ z+H`5vlrhu@cY;RTv)Rh(V8A2UXw7x|?F~^>MTG#qh00u%;&GlD>(+eXH}6!sU;L{M z#2BdJEsx>Z-kxDtfq$)M>!Kp0XCv#BBR zr{Ns9rqlWANW_Uy`&k!UmyJF|@Q8jD0vD<R0)@z6}idK9X3`+O4S*ZqmNouA`EMydtdKQhuEPhqgAi7x3gXWu9z@jT_ z2r|+dgU-W%=Tc01Kuq=PgDg!fsM~i0|8p7SCP!sC zIod?PFX4a#k1~VDi`_)WHpo}N?f;-bfz5Al-pxTZGtlx7{ufg11im#paM-`s@{aug z)sy07d5BQW={<0$t7v=)jj+M>hQ29^XxcCO!LWN~S2`fX;J#;pJ2N(8F;v3ul*q1> zzrUJNHn6bcyHGZOBg-WlLS>f*%1uwN5@6L(21Jp5ToaJ2^5i0=ezgeeDMjm44=UBf z%ziQLf8=7!`YDI-w(fdw%@;z!3k1j$@FSDc2I8AZCoMFhMmsYw z4kjcOGJBkX@;rxReLlNTNJc69m`uZ!+eQ)#beA5BcBQ)&N_lg%z;5u-YK(xDWBknc zz~D2c2Q*|{Vru(gF>pQg6VAT)(aZe%AsOK1QCsv)QV&BNs0qVFvIw6Ek`TYNZD_%r zDSNZamfV9$UnM9B@d~*06&G+RR0fv9$e>>vbN0gki`8c_6mFxL5_|f z-#V;-*RY}XVw`;+Py9W+bxwL-c~qzUc^Qk-Pk?&K`>4GfzNt?s^>Q7EN3JZCba`3J zRWOI~ah&{EIDQD3~tUK)q||1A9`h8gi7Elhm6?=KWlbnLl@fe5^iC$(zdW*ozGklFhj4|{kD%V&!Duu4QImFFg>%G@lz0W|%m-5= z7jJn5@>GZWIs_I{$iKtfxAmVnh~zfIR$)OkQ&($CImhBhA`DY>A@W98;>am=N8Sx}Y_*{vT1 zJHA2eNK+wt7z4gS{d!A2U$26-!?><|CS=q%c(TEa19EZ7SvHU~KlMCCe`%^rpJKs} z_-!&Uy90Wxx#z*ac{@l=c!6zQ^(jiA_)*l|o@B3VFZVMD>PUnRx%lhs1}Ub8B=59PnbF{GQFYwHj=WLKrQ4gQ#`V;dssp2%lCpYx*s1s?JO zYyf=I;rDXzR{A~OW0VPe=9<0VNw)2a^0J<{@Hs{C-I8D+qj)|F=w^T}#DScnDfoaK z)b0VO870VtM@i>hofwDz{fztTYqS#!240d3!VcP$|0#~C>BZc!E#k*E{?ZZXk|R5C zWh+PaC=2i3B5%#%y6(#{9^Hrc=`C4|PS7%g$Upll0h{3ORB|#jDH9IucE_VAiziE= zHtIDS7ujIxhxCtWPf&J|s7Udm+0IPRZqd1u%9zqT%&w60m?1d%7`PsijevH0HnCTS zLqb+vz}gnfo%@~1r6B68@8l^M9U#7Mhy^e0iS8mXyNu4VieLiQzF}m~9sPHGlpV#< zyAw}hIxD(Gu!B!H+PDALh!NUtKX!DR5}gfO0TV^!V&+Wj(P~VUl#%z$dd%4(Pg37W zI!0M^zws31kaM;ZW8j$7vz?3CeI9rkzfX_!G+ItR1>>W8gcY>CYa1JUB744akOkz8 zyc1>z6XSaB;JdElo{=5q4v45h58J{c(=UTRS~vAW$j2%lMYO=5-)AAn#kR#Jc05Y+ zg@vOa0pifVV}P5f!t4(5I9nYe+@i#>gI$ATuNd)xw|3~jn(!(sOUm+;vqQzcf@>Q~ z5K{RVt}8btQw{7%3ew=J4!1lwgibkv_8ir*8-yILKaHeqVM+HQMmhP74$OP zE&;21F}oZZ%ZFbElcWrN0%jYl!xax6H59iw*h|y^?|fm?KS0^CUqd)Buht9c_|>-& zj813qC&wO1e3bbzXydx6%y`s36WIKI+jsE3aoZG=?$X|tX$5@O_fyt+GH&V_MaQkCV^O=iwTp{}-2gX0$iLl*;SneQ zE$_YWUy4rb-T8a(|LF^G?7!t_uWl!1JKONOtLo3sKYe{&)$wfah1qtR*BM0Qt!jHY zZ+j|e23$4oF_w7dtzRzAbInx{ypDT{ct4=>Hm{t_@xAu+BSfWJ=iF?Fufguf?Z#P^ zL2h1ZZrq!jrVE*;eFFUf;&UKwK-R5WC&d+jKWgOIT7=Z{hOA_&Mb&3L+wo=S- zhxkBoJmh0*%$w(`ZzC@J>;LOt)~3#%V%*BR!uyvBUUH-^3bnHnIir$WvYiC?(J#P` z6vwvk2qH41D^Q|{muWL=Y`N3o88QO*>;AX*OVI4owZD%Sb~8m^?~qY6{NP1Fsqlzzt}ib> z^R_woaQcH~QPZ+p=grFx#9uuokhPS1XU$r8)9b|t@N?fo6KddFbs*2J^j-IM-O5&g z_?fvY@V+~V*GG;&QNFu`^TB#c`Mv)m?n}d=dgHKLRR|>_TL@XARAim9q=*S2JK2?W zVlWtmi0ouH*|HO=F_xiJc9T5}lVyx;h8Tv~&O872{qVjY-uLtSWv+9b>pJKBp5Jr- z?%#bs=ef>kRP`FIj*C~?tIVukT5)tl%W(=LXJUaHe6!6exoz9=L2P-y+CN$*BvKbM zoj=d=7@GQDx>Gr>$YwvlF`MCBU`Au(ac+mK`#-h2u=NP}&M5!}F?+?MA=NEyFx2_= zd=|G%;f6*|W70F2}w$Kxu`rGh%PM#M_D}cZL93Y-|_620^igo_b z7xR*40>*OS7t%1TIa@O>Sfla$m+kW4w_-2)dVz)4{lu7WW)Vl)h;HW#OAnLpNtLFL zK5B}amOZIdNp|T1>G5^Tp7hRU>n;`Kcsqk%YHfe0OWa`wh2HBY`g9#@FYb!_1~i&U zwiok5HZl{^RbB0=&mQ!Dz#vLZj$D7-7A~ywwms%=u(GSyDJ5T-hFx~USF;auLuxO# z6e_IneHJP#SjaZHH0slkeW#Zm>SeI;hb6o3{zdp~=|5)Q#$XSg8lGUTHQlEl)ADxw zW!eR@5-@PyaHjKqb+IPgm?0(!r@T&Y=4Xh>OR4sc1FvSvgKIpaKLj6ck4e)>$nQiS z_n?=77no_t!X6eBtrtE#{3%VSW}N4}~mjbBrE%nEY2bl`L5z>3y+;dt`rvE~b3r zp4Mq5BrmHBDE)^~24UwzqIsB*#L%>5y=aBA-r0bSteg<@-!=jQfw{C4k1KMj`iHT& zGY8!8WqLa$4hUxv5+H|Ebr4Aa=n|$UT!enY>x)4DlTR*cNZq&_GlT!dG}s~R{{nxU zGl_R^oj`+^F~p%QW(vaB&vndK#?`8BR1vx_ER9Z zub+P#A2TKEjf#wURRoNN-{%MYxzkjf+S2hhXQDLNIsPLR zs@Q^F>;6`SuTS9R{-20tC;Cln68yD4#cWpz4Ga46_;8pd)$R(P`LdOQIFWYT8z+S^ zNL>MK@un#)A&wz8%Z(-_et{+{Ien3%eowp|OGA>X;7s%k_z2ep(l7yhq4 zlVnO;zU)Sro^Hfk0=<`VXPjmbuLFM~x&fWL=O^Ud5zoqcy#9bm#Mz+uIaFJ$8=b9w?T z?RV-O+z@&lIE-jBt=}c3g@;BJ$4=NHqtF~7~9Da*}uk4Dv15;j=0If6GV2Trdf4^t{11GiNeGWWgME?Z; zJ2+ZUY}Sb@jSO!ULmW5oA_hPEEqGtg7pW@9L;VW^_2bf1>D@qkw9FQT6x0XS_1yHCbB90g>4-q(+UYx@Z6J@jvjSTV1Mh+%zc z+yA(jn9L^Qhr0-Fu=wlQzWCeUFSiAs^mdf_@AGg_Ivym3pB2EG-1n&z=Z;~5%`RQ6 ze#^ul>35;Ofy2y=tMsw#hLV1wipncte+|LFg6#8HlV2~Y{H`K1T_J z<~JK%L?1$&8!I}PYWYiEU%z~ZTtjK>JiGmp{Uu+&FLIrp#fyvkUrRQ)WXX{}x$uc) zC3aswV;SZ9$AC~>gyy1{m@iU#8}l9w=_3zzvY&`GZ(0P#3+n!`eEU)bwM=Hs*m+cvEMtn%PU#{NQS)dT=p8(L zVc)Xaom1`a{$yxX3sjb2_0RiFB7n^h^W?z#!vDKmVRJjU207pljOB^CL4h>{xnOLpAzATSqKkigX=tsSkQ~8;%x}JAeEq zylARd=bH8}`tzow0#<*jT(Jdi_Yc8E6*tdrBTgx~N43ygw&4Og8lWS9hZ7Hxhu$1Y zHm;XEAk=+2e-Mqr%#n*uv3~o`CBKT z3uONlewmB@8ZP}yg^mhCxP(f3yk3O7g$QL~4 z%-ow&ZNKKlxQK8_0LI>fmB{dqj;<#}WOf^l$^<_w4Eqb*^3aPpoZ_&3)~g2)eyGf6 z{RU)h!uZcoLHQcv!+(H>S%*{cH6G>3fi&FDpmF+hj%cX~V0yUCj17sLVg$ejyqI|8 z?5t;6`HCh`z1%Q6E3md70%<$RoPOM_w4FkRNA9let4Q0syD%3JQEfqO6T1x+uODpS zKs(>zT<6l-)`3S=PAym_D|rW?LZm+{2|f||8&?@qaLFG;V)<5u)t0EPNy}U^H(z>6 z%UrOAdUS)p?@lq%O;whQ`LiOu4SxDM_Z;BKpk0mpuq$InrKGpx!}s2-;-FVDWOo~W zYQ5hT(H%`TVea;s^M~zp?2LpHpQB9V^t33x|WP$n}1 zF}}tEJ2-UXC_vb8M!$?iKS6f-A?`-0-vJ&1s?TT~_4(ZE~q z1adZAm2La@m!rVpyNk%#2(weZ$W#2=^EcI_XNC0T=;qhs-^kH_A}09ZWyJv_$}Ng+L=(!PF5n(ine zB+{Gi^^FMsJ5Z+pWjP;7Kqkf4FX(U}orxJu>!#Y0Mu|z>N?L%NxJV(rtwZZu>|{65 z=i`Nb7IyRRDI$UMCBYchv8vrEDn;vN`LKOUN9+wlQ!0h}I(Ub;uTr#8`MLCe9rN=q zjCQK>e9rJgBLCTWNMh$No&JN;Kq6h={8k3MY2C-cPf0BGb6(POuyY?ouf;23=jp4! z@)S{}ljUwZSv*O=5s4|p|5j`(3=6~CI_%ZI#r_S3twddDWetXZL~j+~)Rs79i=Lae z$&@M#e))v!D=TiPA3Vjt&Hp~xco+=Zl_ZWhB9ZWef>S_6zZlq>^-!WRV%HFe{Fo$_ zI6J!)8UWiJ>{JXwVq<6Xdx-m!sX$7{ZgNVs4UW<5@#fJnFPVFsT5@Nup1v0*reUx5 zYCpqczhP&F{onO_k_v+RnH|BL4H$FaBg zox~!Kk!+7r>i&<@OzuEjQw8>|??@1Qc9y3Bag4;IphUErgFwM1APs)p7E_~t-px`+N>0QYfWju zxk$WYs%52GP(YO{SBgC5fFW3=@-51hWB2}X+M*9T;_ z4W)n-3MQ}^` zY#xTYp3{TgflEYZS4MFA>fOw2Vck)w@0?3etr+D>gKTGPrja>0rF2lMA&#kgD5~)6 z-7SiqiJh^))P#9tp#X0?3Ap$C)Jqa&gxmF$1F%ra(HkLu0aR%=jBPk;L2EP@8Sr(nC1A! z7LGj(Lh;XXQ33Yin_q(+!ZiIe)o&z~)RT!p6~XscEKcolf4h#K`9xAuewA_tq9k8a zjZNfG8l|Yoii6A}doKD0#fjf}R0+#4AcK>!9KvsC(zvK_s_6URtcuc&udzf;%!@Q3 z;lf+Zt<|)fFp2`ouY6(6c;sUQ!bFb%q6j5eJ~xO+hV)#s4xPXB7wdRR7Ma9+XvA0y z{ju#-wgC$Gq%VGnadSBKMlJBjkW<%o5}vM}DDjL@0f!mTKqeoz56T9S94yR z8V4y6!W3_P(}AwL*;5+Mz^t`5K-`tl?^h8LW8K9(()Fa-?-QAhS0@YqMckLjDmKob zeer(~i5-XQ?LP)yI%o>1?12Z;f||(~DZXn$8P@y7$|_GvcHRACsG^fiJY^Nw+h*M5 zi=ZE{E%!MP*F0=j4Uaea0{W6;Pe6Sa>{O8FE`+7o)^9K+Vpf_BOCz|&A9VhrJj8W> zod@6f1U?i^4zk04H>HZUmySQuxQ!n+{cLz6L?3wMM%S&6vsEm`=6!=&E&57Fmg3_&T zNy~O0>64paT*cMZGg)7@b|Zq`lC?BU9_tTwM ziDa!4wLtvfWy+6puUvoGEcHdez$WA~575zQ3HkKJdx6bOXfy!c>o})js#wm3srUB(WSig3U<(Le}5FHrUG1sjA%}8eQ}kqkm`_Vy=pLVh1w%x)4Fyf zWfh?+?1M78N!e-K91RS^HNHY)zLey#BHOh*W~L96(a{TzLk~P|?unuFe!Ya#)}^4a zCPHy!K8E?ET46q!>}}tF6-RL2D4LRxG_wvX0+Wcr92@%kwN4~x^X9#AkemE~Mk3V% z5~tqo_KULigKNI&T>~~ZU=wZ)QjFJ;+xkrnb<6Mx^n=pRGImoVPmnVm)nbO+zYr=s zUL-PBcUt$e za0I{H+cMUCdtGy;IzC@M_l#7&TS=YV!`h-IIXtidYo&aYYbPG`#{=hM{rIK2!wsHo zY?r;W6}x)|+CS3rZg%oG_wp({h?a)UuvRQ^B1ugwEzP5g5gJ5V?NONcm!Vt_pN%R% z>wO#wJM;C{F~N_3g|c9w<;Fl|`q5JOvhusFqYnlRNZDB4382aUwa(QgBae_+lXCyj znd{f&ZY}LRRvS~cHK!;P$d^2#E#yBm?XxoIHKT*`U_!C)Ix!6Bn+m(t3QPV*#ItqK z&go!q9OTgzvjpZVXuf(4cji##StI3TwAR9pw;#8!Q_Pm$R$a%k@*bT7Lxnyx2*enZ zrjjnqf4DOO+6&00D_=HY%osO&-vMtA><4H3b2Bp>PX)$l97pj-|X*fNC?A?+(CzwzTo+E6)tEk{Riv#%O zq66c@y3$6hT8m5^k$Wbbr^$?*VtoBmf5%mlZHzs{6h>Zb_>Dyto{Mitw*Vd*27mB> zI0>IbIkoutt>TLJz_Nd=QI3b+*TN;_F6>zfp9~A!Q2)iW`f)mC6y*D-d7^Ly5kbLU)X)XNlcUJYc0=GY$ zX`i87e|erIYV3<}&$wxhD1V?-Ya9fT{JUWF{5C`h@ix^kstPFZdJpN}imkSrF(SiExinp3`9;6?^&bEc4aK1wms*CqqkR=aalB{mMtrRJwT~j`*5`p0G z?&EbOgPTN^KWu<}e`H*Jg5ucY{+-$ZSx3AtKWb^&bDh$pqi92#3!rLC;)y@8TZ>dQ z74zw6m>X4ls=xpLrYZj)QuF`AY36e*@(9W?h}6N6a}V_mJMf#Q;DfFnO&fmV+PIGNhLK+?6FJ0+g})+o*mrtNaY{bmyUx}gbb zhLS3ZiLa=lC0<#`cSOMOdk)F$IyBMm=McR)Zzn6<;%n>kn` zF{dNhD=5md6N^=r5>_+ufzsRGAaJl{^-6py}D*s<9h1rnN+j<+g5$*z8Hdh*o)DM6-V`<~*dc z`T6mk{c6ChDp;n?H|GLd2CYs~PXACF=X2wspomX*L$CbIQrRw=Wryz~ zMUI1BEJf`^aMd2PcyBmn!(Mf3eC>5&`G)gN zk@xE4_Aj~HyeV^gS^~fH-msoZo4VSak-o1LFOfb5)5XIonLXT!Wuuf&KF$J**g*MAJ)Uzo(`VW2V$<0W>3?WkDdz-50zz@yM1nNA6~)3LN>Iu zk9qmI2Ac-KO*J$Fzl14i_E}PUP`-fyAvTY%MA!^*tT?1g4=j-H$?;fF;JdF*|2Z1i zNz}V1M6y*XWL9h~c%Nh|kR>aZ=0oLvtKqq{DM zh;H_2c|GPnn23$2rEM`ZIp4sPRrtMXX`R*$nqsPjG!d>%e0g%q5sxtba`C0v%CgYF z=;42$q+va2Y;n1c`OVFXI)!}dKfbLjj30;={2BS9M0qww*|rCTc8pOzZ5}CID*nin ztbW_~?=4jCN_pgV2G!${pH25Z0j<`(w9Z%RP=srAT`zHb1>e2(3p+qc{}fFVC5?mQ zSb8#+!3(E(9QTnYJ|vLYs^poUYatiKblCWL)1Y9WQ1EO=&F})U!4ZOd z_BC=NeK@HdF<{Qu)cE2vWj)L2&d^rz)G8u`rk{3Z9OP2ceH}rq(EE~8+X@)r5B!xW zupO_ziZi!(RWZR_>Zpd69j&Xl3Wb9u4dc0o9Z*Xvlw%E_m@rkFdn&4M<@xwj?a9X<&F8aY5cT3ZLM1Q~e>j?&>c2xc*C9qj|v!*($ zMjofzS&y^!?3Rh!Lkt8hoBkbDr$`w7BGBj-CrXAHMHgS)d4sh*7jMU!a@U$F1Luke zZi&qU?Xf+vP=x*8ezz((Hilj8@l%>M{){qqHs`tk&g1M*SEw!_(i!Z${XPir&~-I3 zL(U(FE-MYs>1`m+f5PMUoMC}}7hA8#>>*N$*mZp-88AkzclXx{HS;f<>++WYLZ%^ zuxS_0U4E&106UNv_eA!eA4_Z>1S-Yx32be#4prj%I)n?8fiBA98macdj`r{uUl)o7 zi6LI8lb}2ptx2N+n<&qDAIN-nPZK3$F_W#U4C%x;^6US=8toSMU(e{o4k%cAUPk#5 zc~kAuut@p9q6((h4ZFe{O^{B1L0koIG1iGC=LC8TlUL^NOQ38X;|R)pxZ*7cS*qyp zEJZIyHEZI*O61%FJZ~99>EbevVB(WNKT3Un5^W_lYit$PQtpfrT0)l_KO<4R)k4?% zv`~&FHsc_3_L>>USLQjNQkEP4yeOzk6J3tKkrSp10ZTX!%qF5SYj7?6U3zpa@ZvF+ z<-FZ%UR;kraAO9F;d-BkM)$<7q+PvL++Gb795ZuCyTzLGVZ1Z92cw6J)ju2C3RvPm=BF_InWsaHK5Edp1eFCb&@=Yk8swoUJ|N^!Z;z z>w@!i#on<%TvSENTb4ic`tvOUk?0CDv|u#oxF`>($d!g8Y!7yAfL>|6tB1+gtQifz zMjYlZn*#w77Qv;oX~5=??fNT5!`r}Lzfrgt_np_+1CzF=Ld_eT3blcq>H^~lsBGDi z(#i{8n1_N-hr5OU+eZwWIfQ^!(qU~jMISf={eavmCwZA7JHl0)^4aw}KefacS&EMM z%51yop65Eq??{Gv{MbohFIs*Qygf9YtGnU<2LMK@bt(SHx$bZKwJt669y_%uO z`Zes*JvQAKJxMRKXtCY?U_JiqXcoBM%RC^1>mm)NI*;0tF+U8omf6&{|BTino0%@< zN}Fktl#)Hn=eDryBk{NMF;p-t?Ml-gD<#cUOomb+jad50U4}{D3rLGUEo)Y6Wg7^t z)Y~7S$Qm^GdVkYpVHmG>a{l5ZoY{VdClzTG49G8?k3i`6vmy|+s(hsjLfg7Vx*hM( zep&l@9d7F4h10#DEWoMEP~wmY4D>S}9;2l4T&9>ATJTgn32*oBkDYuh5s_~+0gvyQ z!op?Jna4?W!whLDTcsE^{8w^Bzm8qq;JwfQzlEp6%qb=Qul>F(r+&(&D?8Dh^YxK{>Ch`(O2Us0So_kn_KlBzw<@~?E{n0(dj4W4sH5pSOLpf{# zK$Y{B(H7Ck>e@n)ie?k=gA=(}qYH~dD=eS9)Z5Xab%auu9`^d<{B(f+$g077^g6uG zy_;1lDuE%Lh?4><4wvqCvB2s7@-hY97<;b7=N6^z{!La@Fs90@j-Zgu!&R%}_?k_4 zV~l5@lCRz_xgOGaDvGMG_ELt@d8_>!L`kWKZGdqfRtaCx|JLm{>Am8hX!A!(=jtq_ zD(A_+p(^+?i%gNT7QU*sz@N*MI)reCoO${cZ=A=S^w3||sXaR3QP_bR^SdOt-p^-p zknBR(Ib$6X@gGmwdBdQIZ&)RszK%Flc0dzFlNhyt_6hvv&LH$azvs}jntTE@CQjEp z#1Ss}8r1^t1SDWr8!QM8CEK0FKsgKZdb?^PSYDK_1m2<3IIZqX@;rYSuv| zwaaXcfJLXz5k@nh5xS7iw2FQTVz}I->=2EbmRSD7XILQ$mHkG<+t!KXwjUeGRaT81 z_;eFq8FY)Xy!&(|QXd%P)o0;FDN$k+iwq4VL^j}ymMeiPU$-#~wgq&M;dRtSB4GWI z#Rrj$vWhQ7XpBYl_fOGiu!ebL0vhvY_Ple8-%{SG-H4*KwbT6(ZFQPvEYYumc4JX( zmb14wRa>brD>fXB`5N{zW&v&OX>z8H0kY5s9$)xPIjr;)GX0$g(~up}DSRbIF=L7w z{KY12UQnml9&>0_eFqr49~KH2;`!vmKFQlDvqnLt64#3#bzHfJ%qnxR0P~26&t_VLkP&M5bEJ!RoH=Ba8mx& z(}l3NciH$}o~JqPURy3@?&cd_Xl6nJksu0Dopga9>JvW=vxd$;x_Wtx(QyBh5}}vDOk^TH+bXO$%xG_hI@Rjp9hj{{$086n zXljL0u~?+HNm+nZwJP~V$obX^+iJk#nCP>iMS{D^mS}nOL=(l&ino zz`cE1hxXPraduojbt|G;tCfs?`!m~wd5Q4@ZQUcHR~@bec-Y^jR8}?b_jc?1$a!Ya z_@+lMs6dp?7V(`N!WyL!bJ=l`A%q^jX`pG?QC3-G|KcE{$RlaO-%tO+zq^W58G8F7 zD?I_-E8?{bxk-3$$It>zp&@3MnRGLO(-h_S!hTyXGzVqSbNJnZRKD;a^u$@p36n|q zCVd?d>D~2U=`Egp2n}BQUIpbh+e2K&^3r`sv)A2(yz}3`VASwvE{JZK94(lzN%5kiv=C~@89khXt2He>sx~IbNicxoZ7s<%mfy6PT2#~ zRgqQ|6S}pGe7$;DJ!kV15+x61@JUglJBWmgFTJt##C1Bg5QykeY(< z>7iF>X;}hO*0PdT&pr-<(rEDItx`zmgcGKJ9scbwd|`B?A9zuG%+V$fCZSY!Qt>6u zm;|r5{=^|MSGmKK?HgyvQs;A*GI3S|$NKZ`sFV^JtSos~6e2 zJGA(sYf%RM&{PC=YXPBb|6rpV>t_>1O|cEi+|8iv<-=a;Ogb|Ni?Y@-MgV8A(LD&5 zB($U>v=Z6F&)t273ioXTvB?fHzS`O;J|#gzPj1+o10zh2AqMdB_f%}OKe&DZRIEA% zv=7JDGN*)`9|+weJ-j{9+E)lWUkOb3Ar$mTvtpoL^% zPo9bQAVL2iv>CfG-VP_0=C#4HtXfTr&z3QO_|*hKKGYY%GD!m+;UDrQ$$%y;qgS+$ zrX@||-2<0us}3?MfC|%6Q!YDl0IvUljCVde`UDfN4x*+EAAmV62iq|<1iZ{Z2VFTX0pe4BpJ z*OQcH%-Ci8m3zO4e#?9qHu@?jfeeIq*9$U06WDFGA;$i^>_z@|rsTzs3{3Z{ch_AH z5y&6k+g2bG0M$^tt_7$_4zGz9(-pYr2I5Gf zb>l4arY+|khXtS;4omPiOM53K;R*_mqOuPghRpN6ts*S6NyOAE^9jqYJsH39{CF9us^oL;j5QBRWZh5HmQ|oP|mAML`C=b?T^fGg$D7;wD_9&$+3TD=& zCHuGNfm)BIb7I;b58&58xwLHyY zq0}G%xb}%T=Ir@_In{6v1sIJ_-!?aWTGc-r%jK=GY8;9>SmU?v^N7*}HxorF_?HeoYsK-TU-;e-XFV(HCM zR`7dbp&c$>;Qdbi0T!u_8BZ1iGH!nF7MV4y7R{S#$ONbH{G4KWoh6>o4hU(aVkfM} z9ur6!aksAEJOpj^sTaTsC)N0iEDB%@m;@)#Q)jXoM%w#=h2Nz7%c+&58ce{sdUb4{ z0@Lyw0i>lW+<@fm8X&7E3@Pc)kSW9hG{^H_bOFqwit!Vs(II@jcg>#CDc!eGcu1+^ z6SdsX9AZxCHkrTz{XpqKk~%QHjpboL_F5BbdZ-R6lIsW$P}Pf9R01}AI8bY_6SS9QBjX53 zc}>94SgeL#5yeXa^Y^L>vAd2(3EEtoUBFu9crppj2#*5+9vwR%if^ag2{RgOyC3IM zD@K=YTt93O)8Bs#*lEt3a3{Hd(^$xig`?-#^MDz}4NnGS*&t1;mAUJ2lZr^peDTMK zl^^hdl6UQ{NHy?fv+KTfJflHuK0TI7NgVNfv?$Ntta^N7jjq5-;9rL9M zp6r_)R0&U}KN0;m4r&GPcUyJUF zx-U|XHQ=H1^B;5DC+jvfHoyAkh1t>+*N^{5$U#R7$wHz)p``h1TcP_R{dR)uUCH1O zlJ?f?J`se~)Ft8#Eb59Hlozan)Bnly2OC`ppSM=3z9Y?zs~Un(Y67F)`q6*J z20|#xFpj5s_#o6it+mBt%~Phlhv!T<7evbnJ4 zZ}<<(NLuCB<|u)ev^Q^%t*AX`(4Pm}wVX`RD0#bSvWUXF4-)DqfOfSU{(_EzWohIE z6-9}#z8Dj9YBg3%20T&(MKSS9>^ zgGf*vv+pOA-3TMZy^+VTR%>qJD%X~kyGeRpXfugiQ|~WPG%Yb-y7@^LR}nvQm!;_XBj|9H7ON@^o;d? zzp2cP`>BP`X?t5U2`QAz_^!Ms_Te@+7&ma%Mrr+h=hE-`#i}j8qeTSxGe(8WnLkZx zo#j+6F(%zP3OmGqXuqjJN8j)MehZ3MJ4B3&cy)dgrHiO->Z&K(uMaBn&B70Z)57`4GtC2B0V@392T}OF~-U|t%mB( zVk%45L89^nr>+s(c*JkvVpo=juZ3%J3X<(L4`Z*&Pr@T3F$4{f*>~{cT7G^F*c1F- zMRt^gHZx?>mp=PGYv2ZIiK5?p1G+ns(dem9!gfVV#N=osCwNyOfIdUO?vB)qLJf0? z6Z7ZhXlZzbuWJAybBlcb6PSM#n-v{8EQGd(!=|B?PweKM70OpbVJ3qFuSXOS*KBkT z2%7PYilG_nFSDAxA2OxQ4D8}IwyI(8C0-bJgV?@vEBcH)Ce}qU#*)k2EFn2y2==|Qru5~JvBe4?aV#)4BHuKE61G+(HGVc96ie> zUL?C`5Z6N!s4-?;qgO;?ITk&JEPSx|{1HuYdGpJ=scJ}1TGVNeb%fbPs`?YsQ@k%1 z>h~M;7zlob<%&$;K*D$NJ5ZA=QAMf>=Re%o3?kdV%%1x5l&Wab|NIeza6Zo+pPUc2 zDuuDy5a{VwRx@`P-KzlvNd4Ao&nmG)&u?#KW>1-{`0_)aR`HLb#~`e> zQ20yw`?4?fR-7?zEzjYm6B+Z<`~5=w_^Ry1AJ~8f5pR*9LZJdM1M=TX9}!gn>5FCI z70>l9U&6`KxRR2#G9ykA+d>ZNV9c$h;kUjc5vNA3VUt)KrTFQZ0Yko6((?_X8aD6WENfGDVxhh_ct;e~cH3!SRz8$BqC(_5 z4@+brw_H-yTo&T0|!PGm~ z4FjyaDGX!B=Oip2>d8`DRD=q!`v!t1$&vagca>v#2xtASaba}1{)m$c(Rdw{!J3B8 zxfLtmA&WYO9XZF8AjG!Svzk4DNyBq`UG67G4ovVug<#yz9<9*2hs|t^GY0+Np49?R z{Y{CLSYx}HO^&a5s=_DpRVz1w1FaO=pNguWYlfk}aExPE??(q2+ zBcRwzn0kAQLz6m^f?7&Sr>b&&-;er)wnfi@?Ds46FEi#OI5k!^_TDAGibIb*L`q-I zLXUlP$#*|A@i19{x;|P9OQ{O&jqqxfmaTH9)qN*XfK4ZtNK7_NIS^8f<%V0rJ*ZjDpaKvg(t+|2khV(F&@$x=kQ zBZ8tPoK+-y+z+l3sd0OIah0paYMMNk{eq4pr{MD*U%15oezGRy{`dPIgyetbd)SK0 zG?l$17fLD1(N8}bO4D#CUO8)pZM*`<+CyLN9NyY%OQ&Y`ZxiaDip%z_*Fc7Bxm)GQ z@6G9!D~dPo_#FtUru+SX_A|5T-POHQ!`cxosWr5K>6p_iM#I8xwoS%j-Z7l?u{8x7 zEiTd5UYyY3x1WD)|AvNz^$2|nugLfCs`R0}Jly|r;dRd#bxwMC{kTz=Zta7aPnR4N z%%tM4kY`>23Cu$`!N|v10*7vKNE;u}8vm=jR_0hhOGOI3cwDI_LWc~_)QD}gse_v& zR;I+}YHZsbI-!=nT&jV;M!8Jmyt%#Ha4yzT>s+UkJSZi^b@2$|smJ|V%P;Nd8cDdv zdt%cUUy?rF*<9fwWBZudr+@4Y&b7<3x2ix?hbj&Od#>&2ic<;Fn0Y3kZ&wF7OK{-Tf%M z)Es`X1kS#Zo;(QxA60EE)5ZO)^bd*GyiV=;?(Ikl3FQs39fgJkoZse;sL=OAc43D~ zp`Qbm2SvJP?pUoM3cD2F7gz&_tykI$=_trY=5C(n6{_M#tZ3qy8R&?k-%sYA7|N2U zkrWd*7(F1qJZyc{Vm?K~qou8MrE5^+syXiUC}>yxw!p_Ysa0n7Y9+Qsp3?46%(C3i z4u|GA3u83N+&bpDZPr4s4?dvBIufvt7`u3mu;|kz@D~{x?}a}9QHB1UehMP_HtiH9 zch;o|cf4c}sue>{jDSuji5ZjXg{n}?zJ@)Q*UnDD)w81O_Ao}-6NyANh$_TYd>xUZ zF4gss8M>Cx6g}reTJc*+hAjK8ocK#}|9Xo`2q@HGvOKzC;q2BU%AtwTZA;QwB7nj> zj0p7uyOUE53G+_C^YC1u1=1+AmA+4(grhGs>fzQF8nPN)u0eWqpWeF#Ipx}?SEdZ5 zZ@OFd=d>NZh}!(Bx~LIbdV>KnPHWK03{jPQ#p|47Lh3(UlBf&EX1&}C zo5E%_IGAZwwL^MN`{)ChE}pW1`BIu=#_>UsRG|y&RshPE)P){eCe=TidWS{nNIX2o zY7u=_wXN&$LiHAF{y`GF@AVAS=F^}#G;ssJ`nqtF^A^Tj(w$D4j=CoG$ zr{f#HvbWj^JFyx4l0UHj96JN_hwNo*e&lU3`FibTg^lKPNTly~?s8S!rioZE8vd2o z%I!U7G_aT0vQqx=XQGiS@Z5?ANKxpeVh|EBGwRzzdg}-`#T+U)A*F36+a?v8^+!4U z(MzC4>AIy`-09U+(PjznS99d4p8w3R!Zc%{a3+(R!l)5#LqdRa2Zolew~?( zO+Lp&Eq%bM;Q{fSl@pL2V?E@fm^C<~mwnN4629{E&b1yMDm+2zR%{#-@k&F+S1MzP}P{$@L77Px{G6g-4$G_V#&;ynhZhDYj%9J?2{yvkF*`+{$nLG$6KL)C8W1C;dFwg6ZWvLD7N2O5u>yqO@XS<*KL`MhRcvs^+fpG zU53p3fr(Lg{$DN zR1>gQgG~-ats)+XLCdqlEIRtB7?7{zO41(UEx-GK$7jIfZiX;JX{Zn;#l}06+Gcz6 zs3;Lw`~0Be^Jk#qsT8}cJ#bDOj|0;AlD3|R;;)BP!hi9n=-=AI6sE|0FO8_^8J{(p zf*Yn{+w-tab{hte9LCQ^J?#8B%Ovy0G zqwXZL<0z3qk(B>-MnZZxeCs9T^Qhe4en>-??zf-z2(;p#6wkrOhd!smOk43LOl8MFCrUF9CQ z*nMgWE}g2OdVm4)0GUEm{ujyVQ8QHjPD=Kx3OGL$`q%m5|KQ-v^LUIt$q~t!do8}2I{zf1+`EQnbO=tkS)))&Vjxuv@jilU1l1DJRk1Ht$(o_ zfm;K+A1$|Dd#rU+gwnPQpLIlCBI>-}{dj9VLB2Dyp$xhh(KTZ}p{Eat4g_N~L`ZMp)8G117{%3tAqKx#kg0_@MOVV}e)Rjqd0!Zx} z6z7PW+W7;=b{_KTR?gQv%r-e9W=7BrE<~*Nh{cPyW<2FzenpYWUxCwXr*A#h$|7Z* zd#wMt$OZg+R*n$TOl_cK!fU2A7+iQU)cSVqtfO5#rxlSkuuTMH!ybk z@$bGmn2X-u-M`n9S&p#(Fr@));%9{h{dJEN&k$xGl-2AlE?98c?8(_V{R@k`xkkr;&2yy!1o8-`2v;B!}*6kf>_$J=tP^MO0ibPVzL zYQFwX5}c%}J1)Cl z$tCPdxRB&up?{YI|F0ohMIpihHFVYy<230%0i~Jg)beej`x$jRGpCMr2>p}bS_5z7 z?<7x}bpaXAWUbEqC(eXGU*pPekit_&BuAO_#|Gu90u^XW`}Tj=k@*#FyY^3#Cei}1 znB50|mY$e8%z4pxc7P;F4`dQ3O6N7=ne6PhBbRrlwT)RZXH_D;K@2lXK!OEk?^R??QWf$V({^P=%=svFt3Id7(jCD}uJelw!Q9Q9Z|8cq%iD=f1u1nqo{%hA z^q@##ThiJJ?Iq4*OueS3VnnKlg`b_JY6voE8n<6 zTi&Wc?wi>&$FlpCk)UYxyUJTWgYdB&tGTw-72w^sYNVca)9FeHOBP$wRF3jvwKQdp zkP#ctlbI3&(GL5$+AD-}?-| zch)^>dLc+2_}U_Q`waCWmi-$fxA$r{hg_)h_~{~ zR4${jU$%mZ3$rhmCl^AeHs%#)%(*g)#Xw7Zvu4T$VDW%1P8l?L@>dqQ+sMw8QunBO zf@{~3_HJi9P%cNpR&scD41)ZVZ>P>Z*YrV`iu*ZEyZ^;&**ig=BAX6m*G*bZYO150 zf_S2$H#3y&h;YHUl@8}2sH#IAS($TN(Q|is#c(AWH!HLa{0R4ld5U~`RtSmo$|E=K z<-d?cIw0)GD)#NC9rK^kO~CsftykfHgVR=up-%+3Ar!0VAEMd0C`Fwo2fWPbVn6RE zS6gVy^#aujMx_8U)jc$%MD79-NQ1D@V|avszj4Jh4z<_0>YO}qTF(5EYBhGn#qL)_ zQgiF5bA6M@t6X%f7Uxhu)drvc%!MwOs8Fu9YD zA9+$m$#iG81x)6V?wt;q8?EkESv61+wU+K_bBKjQg=6>ah`|ePlOr2uk?cM*d=#`Z%JQS@LT>7g4{u>F%_+CYxq>E}N zBCJo*W{;I;A0A72AXkoJJ-$PM%3sV1>OA8sR_Jme$JLG9IxO$6XOPOo6+REMf}{nN ze%byRH5%X0ek<;BY8m@;;L^pEhC`Dt(Auc+wJOqPo)$*cXP#qh-3$K!xjwN)YD;f@qqLB$C{rqy)6ze8T6#PT0y>879ytCWCFU#7|F zNnBXQ^o(+%WcNCMoyJysd_zNTczA|#-Djq|2>>d1+zfi?cb5W*{AOv+H_c{V4|?+W zN)JqTx$f@1WJ?l4BC)h}8dH!AF@yfv+}?L0oM&cN&NI2baK&ZU$9=ck`-p!BdaZO* z1o_iUSvMFoJDv^^3{>(0OYUtSsqIc5R!~gOoFrLYS3t&|BHOIqzI^v6czR|jeQkMp zHa+ye5S3Z$K(!YNlOR}H?nP%+mvz?P4gJdhC*PfgFRJ8l35j7E@OGWSDI zw#_e+C-^_&ZbLYOy65-15BQ&+9c5MYH zQ)^B|j$=qA%?3C=t~x5*3Xa<~F0%}Q=wtq#`xy;~7Em^s9iUm45Arm~@aeh}O^;q) z1+jmhvl?Wbebnk2B1|0~o&&XOs>sTh48|MKYgzrTEF`$wXtE3^30{scx|9Z@na?ni zsDWEF33@WyO;A}pI0rslTnDU-Uo#wIs3kX+=b(>nbw(TLaR*)xWZ^Rg zBk1)l8u4RHyVVnxo-Gm*Z=E`M-31NeBv=m#PEtG{8yw@418JK>k75NQkk`Vg8Px+M)v&#*P}kORL~$wY$B z#U^>LLw;3H28aRokjR*C0%vd6^yEI2p2G1WO=Cp#pWMs%HfW<%_}1%POReGGhlZCy zEc$;NeV-CbbFSSBL{l&Mc>FhBEO=OG&Jjn zbDbE0i|dXZVKKmI+Affp7_Rgb;=!K+pb5RhIk3y5f)ov%o_Vnw8(|1c-ENg%*A_0#h0Iv9OoB2z9CSr*aN5QuK)i* zDoEpSeRU4?UvhW~0PWCQM4q0Y=OoddyPs*b$W5+?)`{mA0LT#h)lk=;nPCXw`yGCquI(z>&l%W)CB6<% zl$Uqajt$gE*mtLrsOJ$!Xs)Ete%{W|2A2}=3Fx-x zF=bg_`XW!u&jY?qBp<_1T$dE0!!tW)tZvXtLp9Z(?r1@VdkFjMs~B)vxH9Uryu{I7 zeuz|Vt;gZ9tKRU=YURBGvf-bwh3*#mC%NNwkHLdXqT~jq^96`#%Ok^hsU^r3e6+>C zI)@?gqcEX=TY)>#U2loHML&RF+TU89m18Q!b(dj4gzq`k65PR-*-S(gf9Clz!qriV zkQ^-;oa+%cPH3Q6=IvMl(4_~y&gU+fQ5J;ccXZMBIEk68T&~5b8@~}i@75P#^=_mD)%LY?z-Tg$B_O;nM|p~kk|+@ zg*=A&#B&Uyf=kFmB(MxB0gH>uG4S~A|$`e_xex&AAF3G&DS+~939ZK zp=oM*{3gF@vPJsxFR-5r*tZ0j?K@6cVC_hT{za_B+e9v{YF7JVbGH59*tR+q~-9VbC)oyD*1s=g+4J>GhCEx z(BKa0tNJl2%GSJiyFC5X_y7+lIS`F&j=T_;{fi!S4AtiyyM|KRWI+}~q$UpvIu{@5 zcRXL(bZ@up-NK6hd#+-2<)rOqz@8snzTnQUZyv;-10mQR1wM%EaD0XAkT$QkiO=j` zUgsqn@!I=3mZJLEv+Z>(tb7T|BkBJ!z5zxF4FVIR4{i6Ix1Gf^kr*}bj{LgahY|Zn zq`tDh_(ex#VSH#n=20Hko!bLa| zj(Tvci}CNegiAD@@GF_fyuP*8!zvs5;W95or|}I7J{OlJ;4rEEdrrhhbir*~10Us3 zQ;!53R6v0yTmZKpz~Wr@{%&f;VrCAqZ$+YTwLA>|F0U8g=<3_ZhSJ^L0#s9(7Imi zV1Llu@+DI?pV`gZhDHS)Hft*^Y0ECaG}i|@?+I{SJR(TdXWvvNSrWKgJ$4s<}`w=>4=xwXovD^DC2g%I`J?b$?-~c5eN#0Lo z;>D{mUxrjDfpZrlqSa+$58~j0X*OwTUO1@Of7^g-!f$bp*e+oh_c)S}d?>Q#oo)Gu z4?^H$W<>KaarLApJhQL-*7-xBpr=gd)tGnFkN4e{zFB0G?AC($n5r%ApGVNP#_%ZratU{iC+ZPDyd&$oj6vR_HUz0;fwn*b+d z3atl@J1lDAn4thcH4*USTAhYyG^8KNta)(?vFEGBbWn{^tIDx_MszIkFfKHP?v&f) zY{=#VYQC+Mc96NimK2yFY-n_`9UTDPAm0SKYf{}U0t~?E-d3QjV=l>Sb{z*dBKkFH zgwy`PEdkj-lppu_5m8Wutj{~xz?X(7K4o*BtS^al&mo2Qs{&v`OeUhJpj-dd`O*@f zSJ2@`_9C1`J&6a$&+K)+RH^c4u5RFi9gTe70=kkf8izAnE0M z+cn5&l;ld3v}^}A!y;Z(@<2kLO=?w#KQ94vzQvarK?Aj`SC!FU{f8%Uj)M>Uw>@sC z=>P1*0VzMvQF{uGaH+zu+xlT!JZDOU~`CG-@M5Yrf6!pH4(oRcK zH|Dq-H@9=ppKvl^n(S}NO|(+@tgW#{z`rtxQH{R zw*_TCjF3Fg24Q}8hcl(tI8ZZyUs=G{1M z?LqyuXcEj^6FnK8@GSBp6kUzw^L)uL_p+WO&Hc;yB5#AtDkB>(h(ibQ=)YNn8|cqJ z`rjGIt|fN&WP=u$qfF{F{D~TUvrLD|Ek8m% zLhly?@1W@l?)AZsj?6olhH?Ir5yT);i1@dYZrcO@yaAJ3M7#RVkW95V`1 zm0m4hJCMa`X;J$y2DA11OReb?!p%b{lL_y^+-Noz-~pX{16g-h`0V$Q?Ch&~U;W4g*k1ZrsL^LTdpckW4*sG)_8NPm0NS|I z-lbiV-CLoSi0L)p#zn83ZCj#Wcqd;$R^NmJeIY&sC{|55W;Pg*I_rpGJ)GYj zbuhfNQ!W^Awi1K6)^&!Jna-#IwQ9)Gd)Na0av!nUfOQLYr6k>#uJin-tLc&0Yd&x!0&uty|yd3 zQ|=L62$JFECFORx-EuP-C{jGWU`VE|OiOTVxV>Of0IC>@wC-A)XnuKnCKrckfIB3; zk~%_dA513*x#f{G!OUEHC#dK^OYrk+$aldRpFwE&ELZBXkIB)2e#LiR*_sYe^x-n+ z83EtsEiqpO+oINm_sJ0Os$I*5dxsy@xs|!ZEX?bg$_9W7j}@t2Eg|f+Reg-YF4Nl& zj=-@BgWZmRxFVwfh#XxaTZpqYEstl!)01*jzG5&s!K(DN=_mC=RN6O@da{!wwgF`| z7;aLY0sCySTkPgCm;Yw)QEV;H^uu5Yac6re&wdmBx)`B?N}(Xk#Z=M-Jv!goVa?e` z^7|0VYMl2}YHWQ{bYtMRzV0Jw{Mx1LlHH7;=r`2qN9dWV#D-=ccF|Q968`4P+dh{Z zbNc4@f*Vp~?1_c?uP0P_oEn3VQEyIvVEz;Rx-aRS1TH8kflq7iY?H(LssmkeYZ8C{ zbBXsPMr>R{XA~&%sU`{WSYK+*JAe1&*o=OG90hM!G`#5RbS6C-1f}=N(Kd*M5fMH| zq9O`MbJ{WIux))rcb7-Y&3_0Oc?y8uxB{|m7YViW#{@@*=)xVn_M{HR$wFB0(!=IC| z5>#irL>wjpdi=+$_UF;0y2zr8bLK?JHoq&)CIM?SuY3D;EPS0mJldb#323bMcY4jLC}R*ek+$|;%6?YCWEym2#1E-%cM1f>wR&ike?5?D=(xd*-*``Oz_|NW~gYCo3#me zTa}y3)I+A9(mSqi@=u1D&q^ILv%C?v@!Xytg*!H{PBgzwc6pIa%F~#kw6LS0;WaJd zishX?5$6~=Hk*vrDY|IpOj@$9?eFG>t_BjDLvK%Ih=Dh!4l#oBbu(Bno~sx9VO5%P z4b*-%awc+x@6`DTRJ}FV7jiZQo`up^kqt(O8?jmD;MU4VApT+v81-Jn7urf(mtI7a zWq95;Q{961byTQ7TjiUAYn^#^3?4I2zD--c0(D7)#VIYf-HF2!El-IZ@VCU~?C=BH zH}5_j@sqW66OE+~vn5e&q90W(D+kv~PRLx~Xe64ut>hWrP!Q{efZr|E@A8Ia;$E&` zx~Ib%AW29^@f%wypvC=Gt*CjfOK#4H=ot7&qS5jckjR9My?reQ-c|A2yT#lO)y%~_ zyVLIip*})aeQp{CKZdY5)q(Bkg$IZSB)f$Rk2G@uL9xT>DXR^BT`A(HZD%s=JNgl>AnCMG=zuY&uersR1J43{G?k^yo9TvQP zNVX_N0;tP4B=$`fV>%&O&2#fPB~FsQ`_(Oy9WUXvwF$M$_e(&sKIy5%)t}31;E*Kv zodkbW6WSYX^#!~C54k6H{p}%SKJ21Hq2*b#Kq%B^r;fa|Ob9KeBi?^~?!uo91pnyK zKIG_LN55;j;hYE3%$ppcm9oAQP&7P{m-f9Jgr(FHT5Tml(dTsaR<=Pu0Y52}W{NT)x9TQl0J;?@Y9EQM<|wAUU?>sU{HUG3yG& z=TOtTgqgBURo4=~FN0tE-p3`Cj|KOgUM}h)c+$Emw#k5&%a+0-MgQA4#gjRG-sur> zMwnjnN)6P(WgW0BW%|<%bNk8>byrSEjFhU6#W_#}Y>wxs^Ry7K&jue*Zo&gLw^B`~ zz}x5rnS3I=Mzxslsz-8txzF#DtYoe2!Q;FwxM|2+@OWIM9Z^ttQ2ufgE*Q3JL&W?Y zzxlZlCC*;T#z8WdOaaZxaBIYQ#Ig)oD7Y<1<)0mYu+p00Wl=k|I`&yHoh_JN(+U}th~nIn3y8}2pOXHxEhYd>1c2|4co9Ss%MgP}7lV{3yGGzY?>VLA0}=-9 zM+D{59?6~Bo0T}KGvJu|2(pZTOiATAqfGyKG|%tNj$KQNg%YWM9Yum%t}k44bbo+C zfKAyF928Xs??4q#zC)LxmsQ}?{CGFw@cI*vD{MKoRZtQqQ+fbC=s9Fgy9vef+YwKL ziRS+;Nv@DTgOxJ9961bC?lU!VkHC+dqd*)A>Lgd5;SE>CFi{Bg4N{0}#Tk+zwtNq0 zwq3-sAHOwQBoux)(zdC$X@@`2NGBLxCA)yz;IPxK154yaEA&x)xjT;&_<~)>(MZUE zq6(mP>@6}UKIQR|83Pi*)Qy%LD;{7#y}^at@S5eU766WF@0T=lj$H^*$W!tb@zIMj z^wVPx2P;UF?q39Oy7LG@^!sj|N@!5cw^@?|l^Z}Ic>WUH!3O$C!XD@jEMXj2)8<>Y zR;ENrPiG^5jqiGz{>0`nVW}m|mca)uNY;YM_8T}1u4SQ=DlAXYD3w22V>yuDsZ1pk z2^)0s&D)E3x<;lY^qB?@gLS2o0*k9|A!I<0^jm_l!0kErTH^W1TjGkVUknRW8ITg# zQ4Mkhs!)IqZm!ObGtdy|7=eSFpsHBNTwVb@Xn3edf~)eqo!%$(?5$iaC&9%MzTY9b zaYO*bQ-2~o3CjV0cdr7f(?j=YYSrR8CLsSCm$&R%=(^Hkq8$thcZ8~*1L;(=}mokWl!ntGd0tx=OeQf|v*$$Kt@DJ+*PDNG0%GBSf z#7%g`SP3p!NRmokT;v-rq=@RHD&)Lp?Y43ma zAcRiMSb-F+Le!kk0JVyNFMD@X!D3<8SphPPAyx_t99qT*$vfXVGIpSN<_$3UmZ)N* zC$V{pBZZm2g@B@M-MD(68+yTf&n+&BoPmA)-=J+{@zvBL9sVsFH2629)}=!j^^e!nA>SPAfIP`mwc;U?bSDiGr1jb#f3&6V*WZh z&es%rdkd=S*0oXwyYiN>a;K^4(xX$ zCbVG8&*l3{*z)%!m-YRMfMcCzAvei6!d~31CEyu|RtVn7Ji?cNWoL-LElcN(!<@2i zDL9&i2mQzJPxlkyR$5W_jLtf8+7X|H^oczY?4Px{r9mQ*?N9q5p=VZ|zs%fEKEs)) zA--)T1n-D=zl7l!`#_4J=Ud#nr*C|mms418LAH50i`eZnMj>!KhG0<<1zDM(nC^yR zsxBK2w}ddyUW^!hDKRT;IfMqL@^Baf z{ryv*026;scr^Gs#k6PVk`h??;XbO+`=3`tBgMLFxXyS$G|sWiZWF%mb;u9=Z~KSH zEI;(?s1?qcYTv_mx9^f6LUklm9G?-n8k*MuXT0zw%$Dj1D}ucpw>Q5Oz*GB=5mmUi z$F#|R_G1H3gMDqg0w4qTIPy;2Q*m=mir}$dp6k{$!r;bao{ac0X3+-E#6|qXI(+T# zE68QsE1WYDbCRE%m49I6Q4Lny!eP7ws=#6Rh1rMT67zh~Ks@M3A5(V{3DV8|SP*`4 zr3<0`!9NbKFHv{3$oEInZH=3FE(mQR~Vx2{1O`#oJG^`nV9#2)&C z5BTGWvt~{0&q&}I%%6DTO;Cmt%kPlE`^o+l>B+76toYexk-HhWK)3Itfb3aE)JO(y zt-W-AF}F0IEclMhnmBpkq7Au#`A zsRbcfw%zv|D|G%b{N4(2A9=784e#g-D(+mobcDU9D9-x*4H3@%v76)LX?YrNQh?Au zh#`UtdShD|sB;cv^_M_5$N!!gj_)nM4?Hbb{Z^mC3;dCQ@^Kg%-$HVA40oRk;&e#Bpo(P6KpLH^Alqid}=K#=E6Q;076Mj6+Mz%haYXleEv# ze^++mMj@rcxWD{>2zn4i)3)~fv)V%IDV`%LtJ~+gc)6SThmX(?T6=h366^XO-pqMS za(uf?F{Zj)DIATB3P8{#lU(yz_~c#ja-IU^ZT=DI`P{=5qvm94Xv0#ioI zP9)Zl<>_`{Z5>F`*Sd>GoWy=^j5satE0@Xtw|{XA2>z1$q8hZg$zRrmqY=A$*&GEg zl4S0eUm7KZ8lEfK5aGZtG9BzJhLcg^>gQ|bHYCTxG!`bFr$=a=yz*a{M<9I?nY&6; zrKOJe>9>}L>%porI+fPB858rQzgNNCsxF|~M|};$M;Zt(JA_Maa8fTplUKE%6r7>t z?KzKkILYxbKp*Uxx$=tGtiYCS2BJTBN-$kk)gjKz+fkk&8Yb(;2@RF|(LQ92x*ter z82Z@yxdAFf2X>HCrIo#!ZVQ)e(0Z{6qw?*-;!qY3R6TKV7}{J#KsYJr-R9iYNB+bR z+O=}3`Mq5eEp~oT9)+88ZySVu&qmdXdoJ!N-3~R8B0lQJ846OaL#GiBg4_4+LVw)y z11D1ew{b&WgQ^brn?UlyJ-6-5hSqk0kFisFkf9*+&LLEtta*N)BO2V``QSSS_`5!F zKyTqBOA4BoF$E$P153_4D zKI+w>c?nAQ$I?qn5$2Sy#A!gDTZ*}nw+XdE_RrHGT986I^i+{V%T|Gx*&UY~)K zgwI)Vk}+rd?n+gBIAl}A4@Deuh3$O?h%VBv)2JttZ6b^#h|q8sBq$HMQbY8K$EQGA zYtuZCeEW4PE_v%ichKMwI@jag3Zdto>helHe(#N1ry^*)g?@>AUI3X|H23%fJtpPd z@DklbSD5S16AFiFv}iYoKQm5lR_Ka_ z@rB5COf|C-So}QpjJX3bD?%>~&3b&fqeYzgd^|n^L}FZUQzjFwbEe--73qN!OOcKu z{TR&f5q~CtK~A^-awXZxK<^x9HvJX=O$+FBRc`Qs5@h{o)Te&{lG#ne8S(&%n|Rw? za%qIpSN5}?0TMLbK#Y+T(+=dPfWHS3llh<-*KhP&!ZMJx5W`06oE*0({l=ML>IL$9 zRi%u1Th;+^`-!66EuvLpvF=KAJh?6mzsGqOxD$ufEh0$egJ$-;WTM}?2C3?0{_+AF zt=x%w9pDoBb0j7p&(&{&91(@*sq)oOp`jLJnq?uJ;U(bbTL)7zW);d8d3Qn&-(00n003o{VsZ+@4Saao zAjL|@Cq^K)=`ZxD@q_{OZF)9hJd5iy!-os38{&L04Xwz-{-RrrJlg8cCYKZ0MTczp z186R8+&3{%52wAwtD~*1p{}lUg^jW}2YFEDUG6A+JyoCT4BEGb|H<4^u&8pZI>Aq4 zpCjXy)}gQ8qjezNRRn*>lh6C;s&@Y6PJqb<`a~$~=r~aRKqwTSd7-p^a}dLSw%0+q zh+fsp+&)6E64+#LG*&#zSI!$oL^uV$j1vk6Ep>5*+Rei2P_)DZ4~f4W-or&30u8^u z1k*YMe3)Mb6O6we1K{%g0f2X^Gu#IF=~;HQo&ogC`LY}i8J;BljIX`ZWRZY-I_ucc)UXhr`m4gU{zAnI{k))s;4@=q z=)7%-a*>(ZlXoAmcE3yU1jMU4bgp9xa8nErda0+?Uy}&6#|f-LvvRqJBFwCFs`+ho+j;i&)tE-ki)lxhJ_Y zPE84yLq4_Y(;KA$Gfn($G?>a4nQ+D9W3bo76N=^TkCb|G91N%% zzVmlhU6}eqisg(R{iQmgk%$rcOOe@`>th2pz(IfS?Hw|q^L`27pGjAn|E zSL&7)a`UIRRnmZ-!*-Hyy zgVEwu(0|o<<{T0gJB6l7mR&)=7WS(qbb2lkNIlK`K&HVn)wSndtlp6L3iA< z5Y7|&w$+T@Qf-j~NwXEwYn0ZAqqyM@i>~iOn_q7U@f*t;oqG6bU*Fr3+nAHbW~;F4 z^cHpRoHjRQ#Y8S|G_@?%24^T`Z}8?(+Im&fj357Q>Y%#f#(VO&iZd|S(9tLqaQ|u* z)4M?d3$yMti;$);m;Oe{d~-4U8lS?RWvK&;iu>j$*EYV1%b)sZuTAiSkHyTAnFB3Z z7yfl8NZ#+PV*dS-t(?Ok3kvqA{O0~1dGyajnr<8g-*)2VUe0E~$PxXW{J3{4lhDsEb{oh+yxZ zJG}(lvUmP#oK-^nxYIlYW9m&aUgq7%7JJ8}ujRw?U|EV!f&FI4Yq>w~Wr#ItE&tuP zWcWh4Oa7VAX{nab=8+^eUf)v!E ztW!1=xl6xB{LLhlP!N|Mo$O|s_VrVjR@UYP3BTUGIg9vBpZjj?UOqEQxMNvEO{bqX zsb0HgBs{-3s@~@s4rsDcUcQf8ec_Z^zH=)_C&0xhs>cma}jbzOEqd zt#k~X`CT{U2u#+Y*hgO6r_kRExwD=#C+BB_-SI@z=KF4MM$!BnWn+qV$`*93=(PJ+ z#<{xzxQA9Xhxy-BfnEPEf@WS#WX0`a*7f-&{WU9nRf>jv%4b3qX0cxOg(3A=7mS6% zNR8c&*>Bpmc^R2^JTBqA+vYMlf&!V*!1AmRRlqwWsOy<97>sIFBh;WN^>&$y{I{czv;zdNEHQLEA@;IH%<{S`BI6n`?GffME~jk0 zq#M^x<#y&Ne?^sZxqZ>0U=Y%Kn}CJ63nhP`CSv!vZWL2H$3$<#bkwaoLNu7eelQkE zD#4!Xj*RkC`U!mb-2`X3m=-sv-uKKWNQarHjDgW4!dZ*nh>K?)^XqO#&?wB2AIfmL z%ctdPl`9Cpuh|Lo7cFDmKZL}=%@>~z(kVxb1LU4 z_-9+Z%R*qkC1Q=GU`mbU`2cmbXnSGE*#oTu#opjVn2b;nb>MLx(@u!fh1x~_R7y6n zbxM%ySAYu3KDJ_q?aoq2Mz@QR`E@PzGLLu^3}2T1d4W$=PBo=ZP?e^XX9&bJcb^CV_rzbVQiVWG)n zE$O~$+??G+9?Ju(@aDKOO2-rmBL2GM&P1CmRrc-4D$5Zj)i}vhy82+g2jg;2DE$S{ zorh9ZFaxF~N~b4`t`E8?%N`zMLB7r-mWfnHb(*m5xLMfsh$!{NVS#?LO-ggrc7i+? zWmDV*?)@v#JQY_$VfY8~xdEa60%j|6w1l;lzm?R*H2rp}RlXFjN>kxN~yR}ldT%(swhRd&#dB}Q9ZU7lHqpuRE2^| zpateQA>SCtrx0}F(R|2dB+PiVZtOh_pPtY{kO+nidw5YWj(D>?_;1khTu~};FOa;L z#zxG=)G8USz=ffZS4?^;!L(mrOL(xUUl`K-4^7j5wRjz+zW7lIwy8bBdodssmVny6 zq^PYP8$un}qS~^z!_1`rfSpN-BaE78b768+{5`evX1EqjOjIcKX7DHOCfI3b`G0o_ zj>T8CE-X7T;(j|_7ll@TeHO3K39jz3~ zAc#sxus8UX{E4nQJF9V!|Al*X(bUrqcPPRxhqUTmNL%lpVOWPfBs4QPS;PSEfqS(# zlvSxV0HQuOQ*Wk%s3=W=lwU};U3N%)`jdBd2lruWdBSebBh{z6{snxY{VI<3(2#+h zUAkxr>lZ2OX#V+x@v2~j=X?_#>_^U?N(bmfMenS7@r5+Z(TvW7f9yfhZux8JnTR^%^ z|B_lZf+Mex*Q6A_kJaIbDb6N17Nq+(!5n9w4yrLXg$l+WFY%WZ@pU>zs%dLyQsSIj z3r|Mvazd1La_xJACBB|1HWlXOLw5UyC0mM8b%W{`(Uju9{vG(uSD)U$RB#)l%h7s( z_ECrZiCDiZB<5}-!3UkFP(1yl0psgRR}_rvIBN=z2f5*Q8gv6K*s%ZgXx4)L=0 z8~_cOSRPb;H8Nt--xZ7leI&W_2_i3;o03EQ=D&K{{IRVLrTsd`zW*|e_Qv?lcjK2s zWKW}0v>fM~yuxVuGngzN;IJr5T0VsUjA&*vtivh0Y+sq0kuXA!d#1yv5FVpJtu4@UYF%o)}@JY1@AIU%2Ew*wQKnk*Eh zy6Bkh@&&`yi%a=}a}pEr43YHR3{HtMpRZIdP~h2|A(6zFoW~SUZ|T|^1u8M@F9Q$f zN8CC?k(3{wmr=C;{O^U_vUXUZRxH7BiRL%US6%hh7MLOp5r6)3@#Zhr#V(Q3LFB9s zF5wKfdvk`*cYc~*4@d5{*af<~Je(pcC!WEDIMxfyP5$0(Xra3EW2BPb*N2D3IcJ#* zf{HMn6g+$BN1;4Q#PDYoHKyoZjHRN3`nKD06H%o!K3)^+8wl~w+3VOLNKg4H; zObtXw)66%Te~zIM7)*$c^a76x=U^fxL*t<3`R<$dz}avCD*Y8s{oP}7J~B%Yo-#KD zt@D$sEpeWoBNa3(`WFG{(K)Q3_-DN$Y5I5WLhnC5P>uWBwS@*$`zW5Mx<9|K?WoXn zy*lcr^a=%L58G6)%R`*IJyRWnTnn5^lea})bGM<0xe7_ic7K5VoqKnQgr+^9?hIC_ zd|dIYFn?}1%u@vMl7j}ipmqP74MROlp*5Rak7pQf`-n+qNPVtDF(jq!? z?z?BhZW+V@}ElIjXL6Ew?TCWoI2B0$}U7X&%Cjg zCd}QY{}ZL`K8lTnN3BDaWjfUp+U+pE$aLd)O597nTa*l-s6kd4r9#ug zD|_Z%^9$m!KJ@3ytjta20$#>U`WILv`@ewcQl0g1$y88jtFI=894+6LW*uRORQOK6 z8Dg2S&Bx??_aecuL7|r}|0X~Ip%GGOEbv}_3rRwsEKAUDf z!BUzRa@Xz)E;6VVl%n|{<_r}o_`&=r`oBCn)K&M;a1VY={RP)#u_oMbLqV#_$mpCr zf42n8ML+qF_QB*G{R>~Uv&_4iK7-gw9eWjune{$0Dp}vj?erDc(_1p^$bmecrF~os z84mPevkqv56;3xz2GXYEZ>YEvs3$vvQ^rRh#{>gs+pG7jX5WyPx_<~NmU`J09L=Y2Pj6tS!?JryE z*hnVhm;@_aeu}!W2v&6PyX*<{HEmYkNM@N38=3AC!k1G5tTiq7Z{o}4mLA0wB3}{z ze){&xhVlATN*gSKM+fFG$nZgxtVTh7MPCRXO;UjvJyzxnDShoEh}H18WPZpA5()dQ zkU=*7gF@zJO_W*B!?w3&jS{?n53j=jKS0304uLzHA+?8=zVjwR0cu7GuWY29TXIN= zhrsIY5~ig-n`PqnKz69pD#BoDp#y$X7CoS4^P-wPfc*>(9~Tbbe;4jA5b>F#(03_B zQ&uvWuXDKYDLb?JpOqAVbZ`URK%qiz;1x8t>m{>8(wqkUzw1mv3!}G0uMuS`c!c4? zRR11IJE_vs_@30xYGD_|MqncReuL@EG@ymU6Tk1mE(;+~6P#3Ey&!wT{Ac4X*=2js zz8OKV;`ET;940~J|1|sl_9c9IHrHjk6FbZF&mT`Ygt$}`M$dS~Gt^x=6nO2r;ZLKs z-x8u<+29;2Hz0+-0X%&3KOEH#xBzOQS4Zlz8r4&KA^6v?888>2LzTIR>$C#~ZCA@> zZq)1JuC~J|&)~~SVW;D|2Dd*o2oTYqMU(saSU|%ERIhqJKz})w*C;wRzVh+~n?+pN zX-OXC%hF(7>4g-_M0BK7V8S z+^Uq%m7Ua;b;0BD%5LpVo_RrOu2c9BWIM66M)C9XewW<;QF7i7Nxg3$cjZXUl{@!N zja-=%Ek|kPD$QKwNHf#Q%mImer+w&itkJhbXnu% zuKOKysFmK8IZC`@%+$l$cQt-jLv!sXq~~7~hkz#0->mgxY&j!06hv>}pD)~NZtC-Q z`T_F@csYL1Z@M|?!R{Q{P3M09+Kc#7HxyFGX{+nsGau2g_RnhDD(q(Iz&2fQ6nRTinGx`+a$wKej zRCemty^72j-L~hhauDYgVzEC}Xk$>BHCJOBC|+(!K}bLcPZrK)=PMb8YfSm&zqfnF z1@j}=T~s~ynTrE6ZmLPUE0MvVAv3=!ZIN3*F^fC0`SGO$ef5IDhMd?Mos4f4i3dqzT+ev#Wjb+Bxf-$lFKk78j^+f&AjW zl82ir?>rQIW3VLO-3~5Y;Aq#xpR?)CJo0PiS<~GTb!pqob$}p4*0Z+^TTJ8_Jw^&P z^}tC%Ml)P{7Y2;hVfK1*TKu+WdKWWfiZ%QD=cX9NF|%8KSK_SeV)tRA=i`E3T3VQ7+h&V$ z8lI3Kgor)10wu7`yalG-D*J1>ByvGZCNW%u2SZ=i2(7&N0rO$Y+!EW&Lwt{ z%U*OI?N6wTv)8ryJKj%A+GdR+BG%W%VgQ|F(LDxe#pyHlQ8fg^qtT#BgA?wsl2>^< zfZ6_`uwU#p7VACPu{U%J!K{2MNST+FVgVh=B^|$b`BtJ@AySNa1_EdKGDi`6@kOqT zD?DO`tcaC}JDzYLBRT#E9WZ*l$1(T>RJy+a=BnGO8#%>Vvk@j<=3B?Up?}$>DXg!hjNl)rR5&XY2cLcVYWQrQl; z#t!7R1k=M_04iKmxc(jm=jCK^pOgCa(zhyZufRgt8^ zA1Eap<~-yp76{0uV7){}c*>Y(=vj-vU9$k=Z1DlY_cp&G5sB|Q){tj z2EYV^5?I)q1zQT(Xg9Nvl7HWf3DAK4iaZowOYP|d>iX4RMls#U|8D|hH}fScfct34 zP1achf`E@okyh_Hvi?0IfW>Ze<%LnYg=}DQ4lv)rUsO8wJyPd{N zSBnY}(x$xGTeWqw)YJ#38;`nQ1rj9{8o*Urx3*$)2Qug>4ry*xd;Xp;pku z`JXHio>oc_?on3)fExP~>F0q!HwT_OA@;1MaiVe~Oq$Lk+Em=ei!=HknPr4y04Fc` zQ^B`c=~so4B+=%qrURDkJSlV322>flu^VlMH?V^>3Zs!{s9(*@w-y^ZEBN|LV232( zl8?3-y6p*H@jb^YWMrhbjJkCfP&hia@nRTZn6JDJJfj@u8MhwFvTkajP}kWl{xR^W zF##anO~Jt7V%d)_v;X19=-a3OCXL@_&>#AVGgdX-Dke_thKP{C>KmL-H0l-;IJ5jQ zueKu-fY$v;-OeUV0&0@46~{TVh%8vu=+A(I~B}IT@LuZQv}gjlFFybuuRxgfOOn!1)QwM;!wZQYC3r-9J3TKqEFR=ozHV|j8 zDNO?6o|EzsvY_*67b_*g5Ws;bb?R0;UmbzdIxE5oDj98{CjNd|4Fq#@t4m+2 z_1f3xC>7_cp}uxTuPH9N4D(IW*58T^-Q$yc4m1a8r8d5xJ45pMx2eBEm}oOEBQP1u zp&w|@3oQ0sE!yJ&t_E zP-)Z`To~ffebaH6=5Q6qD7$9@!!I>pEEyJyyDf1Sxi5T}XX{_m%i1?+ms$Zs)L%*b zd-sSm^0}Qd6I2~gKndjoeK_eZ8 zJ?Y<~Wb=bxYq84PvL6jO_MrhSD%?MkIKh(=3QbGqxZ4DYB+fNi9N^Pge9LgD*h?GC zf0JYRkHG3tyE&|MJ95>ah#YI1^DR`{9Owk#Zym$)tPs{?26(oBSmk`n8n?RnGB*$n zT4pz?PWo!{Mt3D3_#_k`j^1!7zGEJ^__#r)LIG@qku?W z>HbHiSrzYM#}Hsp4t&dm3Ac9`ecgl!v1+)%Oct4Opv{|7|@QL7@ zhTiK8E;Pv|aZyIxeO!+9RP&}NsQOXO!VhrI`}0zl*UTU*#U@T<2s=ZRs${?{GU9- zrvl9U;B7+AQofEDQ*)z^C<2(tvtln%0@AtralYV^cFsGIVD=^D|jY(Lq zgX1p_ZPwpLXPJQ3jf(p^F|dw=9H@{P z!Sw9=W*WW;WE)*7jLtI;Ts3F^{Z?OGB9;E)nA+^^_kjw}bJh721A8Cl_m`FWwM?p; zAvkWtd|nrajov0?u`zU{~C5evoa$KuE@LH$AgX5kD-vRO(V1Vt$ z$7^`tqU!t_!rd3^H6E;eM0)EWPDZYaul2H0#0}zOfyrOi>Ktvi1jTQ|A4{zv?$?+M zCY}d2z7=pYdU6DVDZq_l{WV%^i0C!7%Mm%Tr5&O-H}0(iRE&H#AU+h99eZFt5-^CC z7QZdTGK`3_&6lR)GUJ6PaW=64Rdgm&n2OW<@9IEAt#ig<*HaV*It-h`S7l8z>Hs_R zdafF)I+t{RRp1FM_+JP3>)E_!PI3)UGje7MmnA&`JW$)KJ|7_qrYeg%(p>5%7;y|y z_^avQ*0y6Z7#hVv7uT(+j8|lRXy0>g$YX{jQZ^sHKd(t_bHhTu50iom@eA7M7QgoJ zAt?S01@-?P@Z1?DHQ<4<1LoX};4reFxwLNop=`{$etH9WKxJBdTQ=r+`XRMJE(S6; zeW&&sXBjnpSL``RsKer!NZw1bp|aM%WjTYitZ3+r<<;|Waah)c2Mt0puZDQMil;dL z6}&}KZE0{Ga-VW`Nb(-<(dj&a4%!O_M619*P7d0%uU7f^K zQLb?=%b*r6i#OMje_5RF`9QgbyZyxydUJ#B(x7`u)O_GOG|JA5m4kWSEuFA!jkX$4 zxe++6_>~+b#PXv>12@5mq5Iu=piTC-aMFXAGwH)oTpMy+Wf>2hij9AmvjZ0&2n91_ zN@jo2E6Te&K7V8UG~e!c=rNUf{=po@p4y#EQoJc0;oG7EolFyuHvEkLXP?%WYRjq$ zB4)yin0B5~Ck;e+FX*8+ZElLZEVCm!!l@73U%j(~mA_Y&_>>FSq*zb=PQ>z%;pJO@Do74YpTL$r?=> zQ=XS~@wI4mIRcOF*@>!zykQEa8}vf60^Sq}9_|?a3+;2Uvv5+!z+*XvvPEg|{(h0? zZ|Hnpn;n(TR~lakpoVgfAhqhs`AZKamntt0Fk`)5T7AIYIR-E0I^Oxyb=~Xq)iT~U zPyGMp`nCz?WjPKG!n9X}((f!9^X+ZlGexQ#_gI^VHJ1UCcsna++&ExR*1*Y$Y0tj0 zpQMFW@GrEE>>nSswl}llM5{>pyLMTgLVNyX;xLjLQ37@I@lR zaaRwxHAY9upos_@iPf5DS-8dd_x8KvrE;ax7INO1jtA#o(8=@JAIZrs0=#7!S4*SC zM@V--7Jz@vKb3a6v@)l1Op#1g`j6j^2aT@z^*U9mZ4)d@>~wcttDg02XuXr7cX zi`-D7KKxq%n?9w~wH=^c<(bNMOlZ83T!4S|lDS+!X4k(OYXyBYO!^hzKaG(|e_m$I zMr9?6=ifS0@2`YL^<`bKy1E8T@s?BZ0_IR%wfE2dFb(oL?H)2<(R9zO39A6VJtdHp zoszY8`h%4&UzITM_<`rsDQKg8&&QP(?AYH+|au$E9ln8!5-&bL#BKbS)X>q7eJI*CFXGYCOb7?dz4QTuYnh&}+mc zZjADKjt-{XahTW&=2Rq#o${xGP2*BEaQUz`1KYj6w5hlcn;2b}f47o@OF^0Y74U(3 zWK~p%g~cpC9tl!){Mq0r*pPg?a&2BwGp zp!k&}$%Bdh$rU*UP{Q_(O^I!RtM!tnFwu#2KCxlJ#KiHi9d=qMNeS~xqIzq;My#l$(PFw_j_NO!|%uBcvT^+Y2l|V9=Y@qM#K`Ibvhhj~gl&xX8~fV7^=Z6w8m9CgG=X%NJ6z+T*2o ze_MwqzUJ6}V`I_@$VD+-VJ9|@D)n1Ie7r+3GQC%re_KoWMcf-EovPGk`GMfOuE58? zghh4&i!Rz9DCZUU3-H2s*|d`w;Ot@QV+N%Sc~tP|VRfDmJ|A<_;E(qz3}C0R-B(2t zeabtj4_$=WLk!XyV_|<_EV~D?H%*y$!`LF1tG;N-G}nrYNDLzmrxd#u7qUYxKX78& zbvS63_nu(dp*^JY65HM#zo5a{8}}yiZnXP7VX})7xTQQC|8;emyXquefNsaeRKbN&zEji# zO>!ToJoFIS9Z$Wo3fSM<@|*(}7dFF0o8l>if0i^VJUKz`_-YPGO5QZBSsY+u7;=?j zZ>D1InMBGbOTAD2BZ1pIUW!D}BEVk`I-K;~hea4=o@8XNE&Dxm`$~I#QUM<~eLx$i zIRBb5lTlsn%gWj0B_WW-6pYIhZk0HA z7f^}G#;IYz9H_yBVIjO?rO(}ODfAu_j6`@jbRDVP{^wDxZ`u+lE`DIZpC4_6EWa*e zh#KVXl{TpUGo-NeFq%H#*Z=*7d`6oE)THWqNChh3>lyUr(Mkc`>g;^y49+lzT?vcZ z#eHT_kGNOE?=rADb3mQ^*2_9R!~*Jg>%y)P5?wB_eI)Tm4rLl>dJSbU+*6ULRF?<}{Mr_XWk30=ov_0)^n@fP$2uqb)ZF+3H6p=v%wsH)v&S z_fxM7vSIFPOu6%(c>$9DR>$inkK={Ctyh*O|6=Vfd%SG_XnA;nn)jBE=yhFA2I!M% zKmSe!eI2=WvLvAFTaT)8EMFj^9;&5#9Sc;>&Z)ZGesrXR? zKDtr>dlWvQj3C#Z1`YH=s~b-2I@Fyj7072Kn#3)y;d~93d#hlhI;r4{JvdzPY=$R; zu+&$Ymh^ml={@kyrH{nMn$*A_1H%X_%U2v!_-7??7D;-j0slKJ7eFCyQ&AZHYOD^F zGyZ$pB3LK5^Pj)b34pLpYhq$lbwA_TWYA;364%}fsQ<9+NM(ZB`xcR_M1(bBpa}$WZ0eeofSxU zBexGbc{50kFY_&*U0mpuM%DtDRuHqiBeC!fHOImA7|i2lrbJ9_OzpLttTyQY^Dg&u z)9uhNCvf(}wg-20euxK6S;Ae87$x1BdnLLU+#Mvd__ep>sAsu;AQ%qSm0}PMNtNY z5I>$_#mpZ`K6ahoF|4<}Ujxh}; zZLNj@f(lWuIjL|Vb=!fz>K|pvNYKyz0_6NowJ!mdcI#!uO#(~;(QPMA2BV)nx+YD4 zi5Q!{VSFH+xU8`X%)c6d;#XVWQQ-*k`D!Zs6crOd?rMho_lWyg>Wx#sWNGlnwV zOaLxJDHwiM1og@0%fJGX$y1SL{{{R(CzxFUijXcpxYsLePIg^Nv{RIm)I>E zDPkcZ&W)U$kABZ&xY6qDc4GeTHU*$ z_ifDEL4!cu*elpZsZ0kT-s{CqEX-dRXRt=tN8L0Sd8?_@vKkpf6p|r&H1ha?lSEiw^?ZLjSQs-y+8vGGSae7 z;Ke*)K|4$H!3dKB&g-ii5q!xd!ER*x_r;S~$N~%til?CP>Yuthp&<{#D)P=fjRlCb zpDVruj29CQ!^H}E!PB`yXp9aJoGqw}g2hY_cM z`P`tCM%yuUS$X zL8VV$jl-mt)rVZ6C0xK7b<-p|t_+x)e;H!$!Wh6vp0-k31LgsQs`etcpG}Sn)(ieE zLPoMZ$xbUSy6d_XaKK$DwF+*;@;xMXA+>ErN`V|(VjJ@WV1U$D&G7yP5SLq+3CQPn ztuSzOk`QgXnhgWbL0IRk17}wFW=JHMaVBHqTNRi9sz>Tps`-fj;1wa)5z^aMyZiKm zgRmpe(?Uvq&ixA`9l+;K*iKVQNcK9=Pmt%JcDqzBZO-}PpE#V|_fqo=cP1_-O(cw> zp!X@~e`NYP@w!?}A47#L(^hJt6cs`a`}DpDsY$b!pFxgK-_W>R5?w3W%d_(F=QXI*fGT zP!K;<1q|{+7&e+Nqzc`*5v}!g4bdKO?ZOe*g)Tir<{&oev?A4kpT%J4y?>(r{d%Oa z^sywJkC3$vFagg71x*k7&h~052$&v7FN;jWZu!qw(BPl_RISKJsW*p<(^tsg7MEv& z7!7PUWn=2fGe(8n=1YV1_Bv#|#UDJn+Pnw*mC{U(qAm*%+XN5Tb>$li@L`^M1^Loi z=uqL*rK&yxJd)^0T)dw{U36+vq^W(SKa{PMX3V1BQ|*>6feEju^zFmui~_PS@MS5{ z<6We^TCNEYNmlFyuK}m)y#@BYwmD3PL=}%3L))G|m0ATfUKwJ*2oKdvV+?pkW-6`~ z18x+jm|GDSpV5h>c?B?!6~i^ay=6t&O1lFL)aBUHiN8o5jM9QyIhz5@bU_b}Mt2{^Uq^0-_t#WCTaMhPFCH-xlW;ftE77x0z zi)<9{)_+MM@S9pIc0lnBN3DYDfI*zg83GJDRsTp}m}GTMa(G682LF|hd@5A}D_Jx= zkIo^~eYsSePZ+WEiH6A*5k@ps{s^tU6R`7=ZJ~8VS5<(QgKh}dRUVO&{Q46@ zf|q+Zu@Juv_8#yHQWX;6Y+KGm?&GGpxW6!ktWzIK{NF)BRyf zxHa_9F9uVUdVf$tKfMWS)m;uTZ z+?|7rqrSShw+vqaM2|qcXkRMv;>^!=LSkebzncP-h`YK|#OHN;0xb{&5LsuK>9Yp< z5aME2n5PhNu{BtNhmMr0@|#mSYMi^AwI_aU-z#U+xZ5@>$*$yFi5@i6$uWkpca9u1 zczNI3mA`=2BDd$ccSnu-^X-I@32#ou&VE`R=@+5uRC$JYLfC+e;qXGV7BBogDj9Tz zfUKB0-M~sE&e$c6StsGA!$hu*H!7&=~pConBlFmtj?>nH4o}uytIQr+=KJtHW zidZKBP6#D|BXH$pg3a?~EW|Q`B#}->Dn!#}YDUiJ2erFogo6tBK##gZM23@9hF_gp zqI=~xm)NRXuQEy(S>;SF_|#G3^tnJL>ernY`j1%Xg`P7qrg|m!_z1|iE)naDt996X zMk@h;Hky!EWGaA@Q@~H?Wo)J{K1Ny5@)yA}j~x^U zuuSUNg__lFc2@$BdhO2h-ERmWbpE=KTLlm*|F%_q@)%rNhJ_p`WkP{0AE|oWkN)LH zhU4a&{>F4r0bU8~>J`%}+%NI;0to9B+ug?tP}&2uX>b>DfmlH2`}I&b1=Y5P(U6XK zhpAqeRw~r{!XJ48cr3>V?!RB2wtpw8pL<6KH)_krQBd>EnqL`!^d7`_gTA+YM~g0Z za`Q>8@NA=HkCrmS0v~;4#(IIQ!sQut zw|xe(Kz3&w>5Y40dR}R>vP+obHJerGE`V#7zYeg2!&T6qC@7cKmsL~k*)|m?TbqYX z6&AVGYiO5=ZFUmG1^4rqrB=HD$JL$86}fQ^adSQ^phh?ZAHGKY@VQ7ZjDEmXDK(8C zaq+i*y7A|JCzxc99(^;6$YkS#5AvJR%Z(la{w4c0p$wv;Iy8hzlX2nUf9P2WA=nrg znd?hAtZ@PK#=8s#cW-}^L{O`Qp&ffqrv3NS3YD#pFan|QRMX%J%ai?o=*Yv9&h}Ix z(d$XTJc6gE4A7}!X*5{_GK_oe_>aIL_=fu@dcg1}Qv0wshgp?k`>^VpL4{%TrpD4d zAv#6>2;2g|(0~@P5m2S?7V%kMlme2eJ3#~wHy&KL0CdJr@7RB(+ zXpX*>ZjTO)RJQN(TXMRc{`#WAsN!ij3h#2KyF?}#ou1=3)_mbu%E4#MHVpMDtCi>1 zNNaq^*!yssp;n*v^!7FOwoorroNmxsLSXtcltX+#p85}T$zn@%uO!-aD||U}d6W0+ z>}u<6bNqZ{#2sGuhb)IJUd6cyS{!f^AH6wT4D{fK^8#Lepoka=+ZNjjhoWF4aBa~3 z=zmXf#zgkj1zY5Gtad~KH!s3TnoC4ee7KG{e?}u)8y{<&^OzBAeDr#9S$OoG+x7L| z(c$kT@Zqvm8rFC0J3 zE!gHpSF+%P2T5eJl3WoJ*vB_p;Wzqzyf@*FlJ&pZ>TW-Je8Yx4$L7E^)dI99@-903 zM8%}4Ff^;h?aja-DWD;Chr1MQbaVKyB~dx+ugA+n@5dEUIC~p5zS{?;{weqO)}^CC z>BoEPp0u5%JNNe?v9t8ryo1MmPL}v0&f)jR4An&ix@q(%h%r<3ES86oY4zfn>#-(GA-lh z=5o%6G6PvpG#*{zJ((k#rX_j*>xtr(6GnE(V~-xsQw>WE7r`$mFqj}^=(v1n;N)c6{0;$Q;AP~#9|d z7Aq#HKgamACux>W>h6fhfZb=v3wgtE`+Cy3jd)=WXx0_aCzwCuoH!cqjVR6hdzTf? zZS>68wP4?+Q1$L8jir}=H@dSJY5sSq#ZFL|C3$MCqK;b3nwL+>UnaQ)L#)E-QJovi%*1oP2xRJI+Br1-SbfBhrlf{EjwE? zBj#zaUm?$4?Y0~bH-_(Y*(QlXXN3Me-6YB1OAqPk#IKgY@t^HZ#^RrNLu~gpUkL6X ztI$JRx3%w0F8^-6M%%&SL;xrZjglx{9y`byg{ z5$Eg-f&~p=e;VUS$4(Ob7x5wV0&ZqNV@o`lXu1q{go@s6Mu+>1;%J#C6L`?^wd}eV zC7N$17++^OreWXW&Hvod`iINQeO~tEh$f=I_qVt%zmxRX(HNx=io#p#UI_SW?;qyO zSrLv+cWzhr(&g8sdfM=)9O-mKf=lq9K~H&5*FsT9aUiADsyj)HiJjMNzUsJjx8-ZV z`8PF}xWKlF)&=q>C)-R^w=MzO1r>`=%jDdqU^(RA$)G z@d3}rh;&n7)1psu>E{x>A2}mLJ&TCUAlabLp%)`#%91x*yDRFLhYu$y4gOP(6A7Az z1MFw^M~qSo%l91)j_aiq?oCqVKAK20_di6YbUJ(*`=rO@pXginG01i_`zF?DPS0sW+=sa}?grj}PpW>fQxUO?MPBtwA z?V-qMIcci58Y6fZ<@!Tc&WA3+FKaIZJHlzf7ZI+#_|-70OPH27EtMbhdhnyl1QYf> z-i;t~gO`CmbgEdUd;z`xC72WVcFSGtH$L%#AUJGP{L+(;*Qk(hVXcCF+J8v<>Z0%)J^J5zEsQZ*qflO!2v606`cve1!6OW(b}sB%X>(;C(^Mc=IRHT2<*j zR}prC&?5QU_LPKgFFqe+DA~q$1g>(NjvhQs~X5iob(+808$iCOH2&1Q~BuOah$ukb% zJ_su{lx%^fwk{8zIB}_yw$p!)>Z$Q%SGDky+?PI%rIu6adx_UYZ+5?tjSkoU2MkEj znbCQGO6J!+Be21E*F(faJmhj2Y$J< zFf_BJcIg5SGq=9$Sa-9;%a@!vt*jMUOtn+^2G!XVowGb`c6>>ajLc`h^(_NspvU~d z#~G1tc$L@^@=ZG8t8cqv!o+X_Dze4$;S+YJb`E&wq2JW8?_$gIah&BbSe=|3#R-n4 zd?Jp5|6FebMT2nlm+dVmST9n}zY=kv7CqzyjlYSXkI-1=9WV+o{LB)ZEgFTRS@x$( zo}P@Jm)nJ?<%CTB@t3XwD$4deh_6IH?7J1fG(WASPpl-O{JW2tsnHxCrth3wyVF)0Y5#gg$c*ow{o@)_O4aA zZ-uzYa8=|lDlfywr27K3=$*3HDGlYhaW7B=la$`$1Jmo`E2{Ka92_=1bGc>rxG6yB zO@n%V@nZ6gTpIxmsC8b~{_k*$L}treyhDL8U!_VV)J3I-I8jOVSCP>w3zz)MsKvq+-$IG*1t#dvdg5UcM<`~_vp7M59Cp8B62~$ zMrGUUIg6ro-hTrH{VHni&Ojwoqr>eIZw&i@L%suZb4|XOM_Lbr$3n(GJxD)y7wklJ zir-McyL7`%(-{>M*{sYRO{@hteC+Ww=ilwmW zMFEDl@h3J)OZYQ1cv+SGA+lJdJY4xs?8FAC>mlE~VFtP+=g(f75W>jX;-W$TuiMIY z34im&cU~^_AE7h*tu@cubX-K*h;T_8{?7Vt^1`kq$Mi9e%*QC*&Gy~N3XWLrSt8N;J&(=qS zJeSO98o*h#sdsJy8G(;&kIe3cHuH4WCli`z6vq0HzFK(wPl zlnU>D4E(#qZVM<1XQG2vl8nOj(!4__jEY)ozE@_TIHLSVAeR7KdJzuDg)2f`xQe3P zQ5w8ObcoK3163z^GXSUw(T&XarmtpHzOyWVHLWDFN8#3m$y|mkZJ{d6DdPL1Ge|`6 zFka~^>gTr~O$~5nx?LC(#|XqbJ2XBT)ZemoK34QW3GCL^5@GX&r==fbdDq*<<@2V1sOT?chug&UlH9t?P* z{FY3A1|U#{EGr&^)O3)9HTSepmaJjo#9Pv7HJH^o?a2I$vX^MQ#{^EO&90Q$fAIA% zLIOJmEH*|E!j?(uk@>A+O}+lQ_*Ae^2Y$5(qPbR5)O_&F({#!(^lT_R&pzcS>y(W35iwve#OjTSZ(mU?)5Ae#u7mC?!Ic zrQ0ssuDF7>p7!`59l5CC?*Eu!RzNJR)S8%$j+Z$1X9|bRNc-YFdjGOZwjZ^fH*=SX!Oyt_d7dk#>SV1* zO$k%u4c~{Kvuv)_bC0HmwxJg!PZUeJ2rGk_Ds~~}s47unWBxhqxL)eRi3f6Qe2C|( z!N_6fef!cSzP!sQA*Pf=n4wN{=j_9a`=&&Vw`qRsMX1roEw|-N6?u$B1TOUI1A};T zpyno9WCSD`g?oZR)wQm@xFM6xE}mE06m!8S`*hG`GpOFLF}EYOA|C@M_N69O`qW3k zU$6AAO`mg}u}vQW^{(exT(_@4k zPG&}fTg`<_-Lqn|HcDCq^UR$ra9)eKOrVL}ZwHow0SY0n7w@gI%i^`oP@Q+kZO=R< zEVYoA@j9MS7A=VNlD^}^N521TOOdQ?I_yv9fX}Mz98Y@Qw<15L}9& zPzM$gU)RjYSiUDc5RQ+`*@J5aCM3xbkaNdkz+D@?klrlkS!*yaY4rG=T2G}{MHKw8 zjW?TS6kM5mP}>Y@w^w@KfOlvMlH(UdPxa2^;_LG3_f9!9RI>C#H zMZAZ9GhmknxR>*RE2zm1jwwvgIMO`2FWdXE1vv(8+hzDuO>s6 zn=5{Ss)Es#eB@WbM3cV#inM(g-Tm?u)8&R<3|LeyWOA75uLHdA@dq6oOUcI7iff(+ zj?Uz10Y|aw%k4qJ2((X2>S5)VLs;4?(b(Hs9pFYIwb(4r4lw-BVrBjj?UX=Ffhn*w zhyk8)S2^E~&p>mJ+wNt)i-fZR$)a+%l_lEvyq+<&5w93Pt^)q(CV_m1xqQ-8<*f{w z;4k&MH6X#B!8zkbLMA+qE23nQOj_H(ZQ{`&C->a6)|z)m6@w0l{HK3=F{C|>5T6ts z8)m?t*QJ+NI}&ZdVed)-L(kgwd*Xe_V0UpU4%L(ZWK6sb7q{X1aakE}#19zQqE=1S zO4`sKNt78p$$uC!SP>p=&+l*;9qvXJc0#luD8fax(cwC*Yb6Yvo5j80rZc2R7*HWc z0>rzDN>g91 zp*kH+jCPChKX{`eLs8x9#;7+gTo69I>*CgQDw{(egkwq zx!h^g;)$W-U}kql79oW+pH7~_$lz-O_B!z4TOJt!Q}~@} zk>*A~CoW%_p*5Y~=E``?8MD_%3tCs+Mw`&N`eBFi_~R0zh*$^e*8-$xe+v{E^rn>+ zL`DirKfL5gfP?FL?6MDG>?ofT)3B1?-V9Tx-7_$xzm}OcX4Lq^L5I9h562v?di)W6 z{q%lYs0V5xjIMVgbG)6BZ@IeMzan)FGtC(d;&6dth{sZ{$&Epq?)$LOmj0C&a2$xZ zju<3eVT-#E9WK(8rqm70cBf}!rF5gSGSCYU`Rt)Mm~k*ea6Zfr{xd`Hj<};X>L|c& zzSl0ga1rRkz+apj%%aTDn|64p{)1tt)vMwNzXQlsQVT>Zb+;FFB zWj*k%4}RV5^FzYu+vm`K;^RFG>j%Ji~sV@{#HDB;RK^YnBxD<09K6yBX$On&;Ol2PDPT+S2?k+z zq?ARX+m!0hr=u7A8Y&|>dc?LFXX$*E@f<^HS**Y6+l?oo?zFZk;2~=r{!)x!le7XE zM}MYerJV+|VkE;$v5o(6aPIL;{_h_rj4~u2Y;sm4XLDA~p-4hSIp%E6=Ti<@PD9Q) zpIVYrPGy7;az4z+nH<9+=i|og{_Xqw>)!U*d-wag?(4c<&(|OCF>D1I+kxZA6_5Wt zsM@JZApc7SF~6N^iA3!-_U)l{o69F`3G@Cb;=4m(f#IYu7MPk5X)J(w>gk2CVl@rf zU~&GLNC+jPmci}F_kN$nteylC=G~W6=#~Rdo&wkDLe z9lgg8=H*vaas!D1Q==`VrBJRy;7~ujWc4*S313VmPVw)5>{|&uy*>4mj4K!&zY-?( zyg5riie#z9X+K>Qv^aPXWYB3Jh}#V0j0MIDSScg=O8pI~quqJqgAE>vFVO1WxJ!R= zZcKSF2_GDP?lw&NUb2=wpTFpu%RWK=K+rO^En?8H2+IrJJr4_t7rN5{OyfI;b*sbX!{dbL#raG*$#MFyz2Qp6 z2fVjNlIIAwpFG;OxFMW&dSvpZW%us8>{LpM_KKi~Kj`^)H!4ILnGfCvlewZ>O1m$B zh4wG>1pIGH?}Q&gZCQS8i;MSJ1IEIna*nO`&J)`Xsp;9P=*Wkz)CcdmwCE1c=SAaF z*IzDN#6}7al3Y(Fb1bxvt+tz(_Y9`wEwnGf@G_e+XQxsZL0QcX^3TY)*9Hf46Iy_y z1m`6(F{oBGkc4lo<=P-sAmBgvCy+cVDsPbiO`dOnt)(%K>OBKy5}y8(KIKL5f;u{h zS>^2d9?nOS!bP0FOMs$7sc!g6;1o9NbiPv^Thh6^W6UW*L!OaeEKnu;qFHWLkX`?p zQfi)JGnfGUM_Ae7?slLl^}&z9<_-BJw9DP1U$-t39LI~Nt;5>rR=vKj&>bOvT*(a# zD)*CaTNM12UvnyWAc?M!cPbB=adb(nJ95PI9SnzLK50A?le*yCDnJ8z(;x0a! zc=#jh5ziWMr^|`{0(}}3q8LQ)F19`n$bg+dk5&XtM0 zV1ZHf+|zl0YASqjH`S_%nG|*^k)aZSJQ#8^PrZ|XJaHIOZ7gpY1PPiUNSGC79H}(V zkTQB{BJezSw*PlB`Nj_WuhINg-F|GUwLylYYoPB8SLp^&_Bpmqjj+hl`Yu>)QqsIxoFcwwp=HETo z0YH5rvzusS2AiJ8pVLA1H;XkxZ4v9QPj{C1u_2pv;}5jMku61pWh0~%@`*kbVC|qzTZsksS?_sKYX=oNbiEr?4 z8^fUuWA} zc(wSGhX?W!T1o=|)Z*?+(9vP~=J2@KgJEe6w;N|Q;W!Wf)c$7DfL^~WFx%|MF9baX z;_Pq0XwPbPB>HH`hd()LLv$7l>A3DqUnBc2Qk~V*u_{Ufvr?(G6e*?ISgb)K{Vi0E zTozok=N|E3RbOTP%WjE`w0L-IRSLOwvj{L;Mz7!f-y!Bo+czzP6$u~e0bhrd@loQG zO?70;E{wh@>F;8V#XX(@etHrb6iI~C1Q|3OzGiJ}jssqA+R#r>igH$_MSpDo5TS2d z9tZQgcMDqq+H>g!i*YPs}f5u^rTlQ6`01nq};D0=F>>8gevgZ1aft zcbxZcvWcyJiYl=c6LUw+wkSj(YC|s3$oC}#bioTVFby+*24)er)Gb#%uII>G`FiE$ zn9!EwGQ!VJNH$xWNJKgH_?}y1Aja7{w#Uc;Pfqo%$)zZb^$0Q~gl!tV`GjaYQ#jnGw^7OX)9N6J7gtspT;nRxcjlLOyP`wfLY}w?< ztmAY2-yAF~3Xzwk_Fh_!EgnSsC1d{Xdj`2Q8%8k`<(Z*9y9J;<*fCKZzuEOFm~@aT z&kVLG&KVb+-S}AbA!qC55a(<5xvjy`bQA8mt-H5%$4m-sLirgUV&)Zp!AL6`Z5nCX z+ZEczn1M>)SoV()J2CSpx4jf5Y?ltFG@ZvD;-Ss;ufNh<*NU=za@-L=6Y0lWF;vpTO{Ej`}Hm31dB$m_;IZCCB-?nTZ!QuD{o8p&}eDD-(uO%TuUz-E{=;0jYej*fatLCklS7`CP z)-jS-^-HO{zoq2p*jiwigS+(Cx2w{29P!Uy(QD;(cvp$C&VnTtY(q=r%zFfUA}7PB zb5hfOdPNRpM>{b$-k3<*_B-+ZeCFNF@HMA=ALv4?hp+6f5|^1AL+_g$G&={%r<*tr zc+P^C?a$qhS@u*jUr{N9j9IMSrpocBVdsVBUJl}oj86Cl%}?k1G`rIlY9<&&YP}}b zs-pPySB)GE_TZTnY8~&jgY8Zl6G0xjGC%`{Gbir|L6GCw<&>bLbK0kpEfq zNmeK<-RI-X!cRD$xMy(1Hu^j8pKcH06-$j41oL|?3WtJVxGlW`M@$NLp3%#tMZzkvxgk&a} z9xM`Bdc@B!6kvL*?9Cf8gI3@~HCXwio8bGi+mA38qaP zBTTR9%kX<`n}A4mqjY~S$-Ve`Wd_|?2I>*VK-8r!kvAGs7UdI6CtLBQGXbtH@z*9~ zWraTKE}e;CKPk<;8R>bks1r|82+0AI*15AXB(Z!C);h`t-|08YcVEyfd3aFojb2j1GKjmmbX>&W1 zq1Z*|gM&9_i-q;SI7Q!$%INAvg^cl5$!XHroxHUdZu=W>_nf3G(P;u`NH0zPaJ#X0 zMIiGRIF3Okrio)8$4njY)Ir#vmj4&xwZokE@Atx1L91U5r>w|k8#AvA_Y735s1qmP zG&^@`nIFLz1cMo(Wni_6X$Z(kdj5`P+tZwFZlZHH9RK$@{;>2rt>!6OR&e#++#%iK zZOg&B7v9n%Xs-v(AZiV!K~pimTGYKz*@oxOww&!acK;*6#DtTR>ctjHhG=1*cGzqTc{*ArcOs)!IfPq{~s&n)QT=P{Ik4|h8E)b*hRfCm7yP| zwD!d<5My}*)1tGzuvI^TxBkVb5TfJ(gilX#lym3HuFP&Fe=X7C`}LgdSTyZt819j4 zw2ROU^W^KFux3=YpkDDsBzH_dHzt%A@qs7Fbb-49w{~t2>CA}#))9szpXC0AyNQYdR_!f-` zp?VF+;t#65;#qex!liq~(T_K%uEP!!eyH!~NW{cQg}gZ+gV`lwKJ(*HpxfXd;UXxLDe$b+11Tm^5GZqD{`yezjlX3xG< zBgcKpG`HW_)BQbLTBzBepAc-{jw+D58%`5`?~`lYi(Ax0Chji?n?=T~i-K(EfLmmr zjMY~meO#6(EY)4YrOGyS$X;I;&o@^IGfZ(MpmQZq13&a~w(HDzob0w*?FoFM7sfO+ zPIlumy8a#TsB5<|?6*c)1t!H?erbQ>6~fG)<2g%JULei~i|qLKg#)6{hsboh%&^=}HzMDMn~JPe z&nua;l!pR_1XDic*Z;b7^|DSnprfWDwL9omRTs(BQdW*FQ~s?;Dcy7j@g(-j=SQE} zKTlyVRO{bb)@$TwJ#cZ?O8j%RWat|fE6QG-ajz=5<`!a?a}7C^UHt37g}v%|s3fzb zgK&+7R9T#C+J2j3^lgg@lb;AV;=oTC-BUk3`bjeZ_|CepUQU1F6qSGM3%{Xj$=qjO zDu-U^zY&7=6%KWsW?B5a1>5jRw)mM2ln1{(P;kMV2xWSW47kx%c{%f77_Q#7!z2k} zsz&d_9ZDz|lHJVyvVL@~^+@M+(Z_A#k=HOD&h-lSsM{g*{!ZVwX)H=I8|NC5Iw)MCJXBu*Yi5t9Z4#uut{c(72Wy z7=PkjzO0ke&()N;efkATT!GEvxyT}rzL_t3i)*ntO)q$#z&n90f(hTqS6hVTqoGAY zbc?!iYlvfUt%B=ySc?$b@_Ze%;1;LEr1%livSxo0v1%=y49m!=qn#|r21osaV>Fpk-`X-Kd1e>^ZZ*Ah$p z2lw`I=ckjJ_uMC?PB(jsLShc!?e~IyeL1WdooVNxPJ{*&jHh7<0)FHR{lnCKH54(g zR89a8>izrGL6~9M>;atPn&>saCwkA#IW&wbuuhmQ zt+&V@@9Yvt2FKk~#DP6HuRNpv;|$mw*LQM}c0^N*ISqWT?O9$b$+-5q4C2+w?4UWIoKgqSt=$rOUxdoS_w3u))f_f;Q7ICF$vc*d%%h)v8=%B zFy8>~Tnys#q2ElLX$8T;STFH;PTZXjVfm_}fRI|5G{ST1*6Ec5M&4<2*lU~Nyb^?Z z0Bc6hc*LC4XYAi=(KEfDO`Y<{vwM8`j#B11+4_OUPW4dbYyq{u5QU--`r6MCJ|MMV zyMyQiXp*{Y+M+7%ZbPts;~%-UucnG1#DQZhvS$$YIF+$HHD{>Or|alGtHv zRWH7$?Lb{%pQ-mwcy;sqk?>hIhVmi%Thi}z|B=L4m3?}xhZg>3Ri6|l#m%@L&C^q_ zhW;c@iN7>;pRA{TclZJ!#+RFyx{TyXx-LP5`KaJ(nsZnq?;jplTPvlIN5Zw)eo0Zg zRY1ZuA7EilH=l_mKA~sH^B+^Hbif)le?0?z#+GrDlHIPpKn&(a%gTfVb`_rc^bC#5 z*-p3dU{0$mHOY=xam+A0c1HJ+y2&xz9i{>BBRt<#7RiLT!!*0?PU;YpLNQiFsv>G0 zBAOlF?DBPOWvnAxT#6kS^PTKazGST-ufU$6gy~f3c;x4UO535K@BfxrVUKgR&pLcU zj7h#Tzo=dqBAP=#i95Lz(*d*?YuA$CYs|B`qD!ojWeP%xP|7rX-5IK}hAf#B^T71L z%V#VQCc!ntEPlbNoNM>Hwr6K69S$F=-#dl}7Nm$0Vibx7*L2FU-;1B71kEH|b;VoB z^ex8*p(h`!@3Hvw0Y<9wjKs>BEg9qY=_}|pRG#C|5IVjl>tgW$&Rs~3%(huWzUbTc z*k|#5b-;KGZ!HH{e}5z;FJ+!E6yVG@m?A!y!u$^%57>*yB5R^-D_mJ%EkQ7y49e^UQqAQ z{Q@NP0>@${+4{_bquZr8GQz92ACQwX$El{zFsD>$yK;otJ&nnjQ7Mxqd+SvlWw(Xp zUonT(*L1*_kRE;S*yRz3zM&P$4sT5WN=og-mPzqC)cH%Q0P;x$5Oh-+*z|sB%!bj<8!e*c zbb(#FFPA|o?S{7>{CDH@2014)E@e_Y;Vms%hH36`ueYt{s%kkVyShL!^J{Qt!8IG#nlcy3LU^Non^= z;yU1%T5DGhVeWW9MxhAx$Ljh{XFLd~n+0^KB=>Qii@646GMzP3krP1W4Y4YD#t7tX zWoUlZLN}#;{)W)psx&R*p&tbNbUHLvJZxb+ep(_ClE(}x@$jF<$TwWemC4U>80+9K@IX&Gx3R%)pOf< zt7byOKo%?LVa37ubOL^2QIBP6J^@N5m9NO>)hUctIRJ2e?UWh@qkS zntt~L8QAL*tr{qw%cQajfU&T=*|T;0`abVfyOSSW>(61wA5FFGP;H!qIyA|kzWP?e zOo*msVZk^slxp0!f_{1sF8md77-;&?4i(sKn#XE~q5}vNfjfqC!TtmeJjW=jQbzU} z?+>3zx8YRmL%WlE2VbD`PLJgj5C~MIX5OeB-fP5zq8itFnAKb{Cw3x%LPY>H_@Sf? zUr$EWOv`XWW~)ojvU$ZBA3fdikeI#QN%f`MY56u-BiyU(T<=IEGs(5NMSZ&zO2&7_ z=-I|kx>@F3sz5GuSLqV@nu71@@aothY*Pj+reqJ`*H_*5zCf`uJ1#ZZcWIMWp8ae#uCk$_i;8&IN*_1 zECFA^f92JCQ?fR%RktmIoJz*CvvuQ(k!0;NbRr^Q6n}iV?Z_KR9&kLu$>aR2-6k(r z<>7OI7G=Cu#RX)3ViTQKJ7%?rSE@<8`;1jAQfQzTZc*|s4j;B^;VGr)4InRW*^1pUfMFlX#IMOIhh23emy^m{GYE%Kv z!ZjWgASX1k^&ARkut@pCnj0PP__J^UX9a{5YvC<lw;%dLCoj7w|oNTve-~a=z@LRcS{fRXnM@Xo}_J;fo zH&f0>(iz`fytE^-j{8CoiM^vf@(+rcdi2AV0|3i!xjrlS(YSy8&XcB#_}6RQ=k}Jj ze_=_k7ZHPH{1_RSS57Kg(Wd3!)-2*u!W?;xTMuBf#g3=THF$ehwdSZFM!^rlh zm$P+RTgX!ZvS;`NTN&8F!Y0X;%8NP!HicwHS_;-yP^y)d##Mr>UYR%(-=WbI<&1q* zgGx$-T6#rOFpnuXraE{o>WNgoH>!N*@dE|dlHtZ$Fj;-;XO57CW5@G@m?j~lcmD5P zw&i!~btIN!oP+HNuk?#vch7|v5+NS07$Nw}M48U+ecb2~d=EYT^vmb-oz}3$H&|yc zW&X#oGjz;9#?08Rp?vWe4%d=A{8M#U2Cr?=e(PcES+Iq&o#ov>AN4B}?5OyjuTY28MaeSYm zHWPZ2HrVxI#>WHPd9!H>5JrM=NK4%$&~-I*s7ggb%uw9OVdBOe&t61+5C9? zei5=Kk`|GW(Z>p|$O=xH{(T}yg$v!O(2Q$EKEUnjt?y>E~?6Ji284h?B)H3Tq+#YN+8 zVJD>H+RIpBy^kaVu+N<%9F1PQJv8qa|GP96X<2sJs|91|+H+Bc!~9<1TPjo<5+Ucz zD6zi)e6uY0T3maU#XbOHi|Vq0Todp?`7jY>!EU(vDr2TFxoJ6ZGMa=Vd_6FgAS?4* z4^Ad9l9#t%URe2r)EEeB6VY-}vuVOlFED&fyfU3qc{S7X!U)1y{3?^n&+KK#K7NWG z)4r+fNiP|PtC|dPYF}6#tW@qTA)hG&t#KrrVFBx=!NnoM^3Md~LTDr}J*fvgrR8)> zS@0OyUwy+oY0}MuomTGx7sal{U!NfVr-<=V_Rrk%*3iiI>>Q~ys?7VlGiZCf)$=

IV|$Cz?`#rr%LrnUnC{S@C~QfWu?++N2fqMlJI(9Cut6 z$q1||^2=g+4IMlHv9@Pg_a?jA&9Kd%I(R5|NzmXh$BVek6d)yUk^fIdV ze)5d!sN3CQJH&7b{inkk5dym_E;|Pm_Ivv|Y~lWnr513wwf9fz-~j<0f7gWr@lMJp z?>NUxcL{B>iQK8oe|u-rZSwBemYjC-Oo)SZ4l!#J&E{D21p)gBBtRwk#A8UfYV1LC zAH|~Hs)qnVdtfi%O6!0Zatb7F1A4|#|*6gaB zr5y6&t+)$kvzHw6<#9lcd4MyQfL~B~ z+0%FZJ}~8~M^E&skAqsu+82d>xh2z?#t=eDJ%n+!x9|%hR4>SdW1tJ1)eYxtdq{Q+ z-p?%$?E$0ZN8jWZ^@0tmys(J$>}AXVq3vV$&|}#kdF^EA`*TH?9()>5KB`1VTt!MY ziqKFl9LHXyDIEvOB_^e;c~Zi=js!Bb6H)d0k5$4<%@03D--?G>PcTF|;WUY!POh z$vL7`NW_q3X$&o^C;&6ru}+vV4=eZPNUw$hTX8ESmUj|l9kuUN_lMhZ3nqJG5^FX{ zy6;mcR5bHfCh-U~{ zgl8ty=YQ|mQ*+Brk2Mlz&d)A3{PABAo99x-mZF|dDD+a?Xg4lSkv5_AYX}0GS<2CW zf4PLO9LTw@CT@q~gIi6G=3;|tlQxmn9Cbx!=qPVX!}@b_J;LQJzf%YNx!%y8ugRSw)iIls2HO)8n7p2(-Bk7a zr~Tt@3JPa5ml^$`(;5;F1iF40(k95nkIM-j*^LC@j! zDJ`P~=PMQY4LhF`@-?w*N4bb2oWx0}6r&7c7td6ia@Mw>FJTUopC<%ex!45({2_ko zygOxE{7w8s8UEwd_t46j${E|$lM{*om#*qBX0I?+TC*+fg&oMrl$Az02%Tz809%oRjC=;TpqO1q1Vrc}#$c|5>0{>k>FC%}|igCkQ+PCg@!9BsL zim?Qd<&hlrB&ek|1dlYrJoyinKQVJ^;KK3N7PUrw5t$p!b+GaUv000Vd<}d+6s7|g zyHn!-O+5Awy8q(|JvjjTlkrkvgDhhAz}TU%?Ip&Vz<1onLL8_(45V1Dgj5`*5%!U^ zZrH{p^nZhZ6Sk4k(TlI1{{^&d-blu%zwGUQ+B@98<}<$hRbNi?;?@V->)0peLx@==ZV$!4}F&k)Q+U{eZs@NFM*UgehpK%wI3pdL6V!?ul8;n zKNf@R72QKeby;-!y@=KB6~PoCgcr~eIx(=hYwzmaei6wqrpOm z5@KfFvAq@jL-3}i?b1EkESoppZG(lSwoAb;UqcQ>I@BzwoNs8p-}=xW{x7@9oW{do zom4O){#eEL3oBN^K_x6YEG>-WD+xc#vCq>vh`xXxA;Il>-9U1pfM8_Vx#GE^&sS=}}C6Q7Q7a+O)?ruTn2b~ym&UXSay}x1*on8^ORmH9l zYcqXZ8Sh6eSa8X5tA@l5N1Q3!d)Q_~e|WqjNtRQ}}_*2^^M#Ra28{@K`ETE%Nzkw6o%CDTi?SgOj z%nf_`+4q_C{#>AnOyQgzYrwAZ{&@ED!!@ShH?wT)on8F9h0k)D+2u9m&u3nK0vH z-afJqdv`m=u|3vCGJEpumlpG5$kKtfVE@hhghAgzt6Y1P{ue@#R2#T#&H zn-TU2up70UJ-jXK#fN{QgMH#twW2r}z}yGjkO6;hOS9lp0tb7jsM6n_+W4)S*KFq* zAgOm^2kK_J4dMl@?iab>p9vn&db;B6VNGT8M{J#5-#}LfrQ1ChoyKF}Am-5PtNb5rH1TOtFyxVsb2L3MgW_XLcvBvBMxU?XGtsCwWjQwE$pg#pt z>5#s?4Fa}EMo-=?-3pMr8x$s73EgCDO8Md|ipj8Wm}#nWUUhz3^dG>Gv?5bt^TsFP zB5V47f6D!+qWEymt_~FSg7vNRHYw181Y`pavLo=XS#v*-PG;;FGk(&;?~mjPqs_f!$bZ&xdj^j@AtC;0v<%I+!=?g6-1OJ zvop*{+tY_IMEOB(^iBsFiBv$8Vl9+=jKD1ql4No_{*nAKEZ3)%m3Q7ysWL~BRjlv1 z`t-FvP|aHUCBc&0VZdJ3$^DFp#)8f-VYbQWKNLoP{-8)1{CzsvfhXAkhEn{9kfsHo-OdobtVF{4N5)ydd01@-ww9Uvz@XpC3n2tq8IyBnm4f7d7 zkWvFtt+kiEA$|)cGf_(5&!&33DX8WAu;`z{?*uNZKSnh8Nd!o`XX(nTC%qG>XO;BC zKl}Xtv0S9pJrE%RozeLrPKUvp*}c`2g*RX5~zD#bye0EbF|853L=SveRn8U2AW^^CdeD>5IVnn~g zz*Qp$?)B`FUdPCS@rnCt>fW&NuY0(-U=~pc(7JSA@||nE!r;9MoCBMR81u$f{??wg z5y^K`4!dAT?(J~#l=&i~N%!@eAmSFWYyjvxx!+(?2;X ze2molYI-ryd&!XG+(M2a&DaiE7u10rm3x*8sEw;k{auW0cFaoZE-UKKOw+{&TNjEd zT{#Wfd=!gZNLFVai>QQvrUhaj6489HB7gmRAH%l-DCX(4&z%8)q6?2j@2Vh5tRK?H-rCTn^iLUJpIp!}ltTzU z%6L%}(KYQ3{|efyX5nvqlmL<|CYr4eXnFVc5{f!Zy8Jy91SR9YBdtM-wo8D*z}8<) zJ|R$;OTC3mCG@icH?tSSZ|7w#bBsEMi;Jc408X>^UlEiP^}1Axvoj$zx(+OueI?6w zDYw+oROSn6*@T37g)j9j0RE_C(JI{n*pja%2daX7ykeKkI=ynK3^LY`k7B5qHw>qn zBv@i~T_GQqMAmg?hjr+yf@I4fjnNiPjedNTK!p0Y3v9Dz<<}7EGD8kUs59x|SR3a5 zR{ssmDvLVB+>83LYY5)5wrI6WSAUgTXold|?v{?v@T)IKQ!}<@TS1?Hrg`ZEs_W#3 z+XgvNHb>f%QORnoV~+er)!bN@+d+3S;yC@@rjLc-t?0SCiuA~wJGWpV9R_pwBtX_t zo>@*q>$GgBF_Q7CJvhU(v^{a?PKz7AquG@qj0hSRa~2t+%nHt^i(IE}K+ z=18QQFg5~<>$3bfau8XM(>0R;#}KsdeatQNpb~!fs=sy9T|~IZ@!!#i{I0<}^fJve zGR07~DBZw-S2iTeP5OqnD_$Vj13cOaS^A;`nv0MZJvkON#vgUxI1%}XNr&C6q_p(k z9zT`rK&OGbFLHr>g-GC>n@?Z%u%>soSv!)<+h+aFYR`bt4pf65D$R&SlrbkRA5ceI z{qRV%BcNAXh@AJ5Vw_&r9!fR#QZSd!imz4GyHi`2PQ5t{`8ZaQQ49pq|Jb(CoLoXj z@AmTbp>&|0r7DukK#HAaaJgRE1u{(c_7dN}^mME&1MtTEQk2&IMT4z1wR@-r-6Ckr z38i8N@ySN)iR2Cz`twm#Y2bR7bURK~#VP)$E zzUojGMD18HsI{a&P73f4;pct{hN-6xJ>ON5VLEG2p!rS@iwt`0@Xuy)*X(ULj`5>v zTxUGz4MdC;G%72i0_|eAmmG<1ve_V2)Qd(ke4y^63apOP1WMcj{Gp+EfQ%H1)`$6!$B;0Dk4Fty$d~(w^s1ArJbd zJznv*i_>`lydaMBMQG9gP+t@nv)D6G?vh?(rb#!L^x-4vdN8af_l93;^E}w_U;Bx> z;Up2cXZNG!ks0V~&k~<}B=Yr(LI|OtS?&WN_jJkgeQX5My*E;XSSnYZbQKd;1^t~I zeSahZ`RmHFKrTDSj1u`LHl)KUpER+a9X)X+6TSEA?j6=AjZ8DIvPh)`NV?QPR+q!4 zs%AxT_o9EpWB$bBCHL_MgC(|*ru-i}Pf`Bg-IQ06!PqgWx!R`@AoJu*mb=9svOPc( z{qp-90m*i%40_@N6kwN zt~p*CS|l?gwTS2&p9c4<9ZagU&h}>Qhq{v6+Wz#Wf^43{S=JZJC8`~GK zgFB(9;VJw>DAK-@DE^W7p;G0UNBdX>;Wo;5M)fXJ5!oD z6eDBg!F!XBS5twYZ}HmD`!!=h4-A9=-v+gJD>iQqq!yj79>d#b${C)5fB)GRpO%fK z|DJx0#<+T}zUH&}7HKdU>I1cdPJZLWXVWcyguEj1`P2qw7f$+N5MG-|kGSl@YwY1q z0L)#cgx@k8dBmxX zr_D5t$%i#}Rs9|uzzwDzG**&(>ur}R-lR9$!ROVW&*6gp9zA$*hB9J;jIuI z_611gK1o)PNiDdJGUAP)FXm<}Mu*$D{iIkyeth)MTGX0Nk3M8*q4&5?GO=#~}gK0><1;0DMWT@eBlWwCmwQ2)A zR^8(=1wALhYt{zBDTA_=m2YDNIHRe5HpJL2)$kbY|NZ^rryaR|1#uDU?*!tke2xNA z)2zDNoxNHI)C}Hg`o0ane)NtH0qy!=;P}%$BcKsvhkb*{SZHYJH`^IvR-|`#C)HXYMXBKt@Gr+eu)weB|PX zXt@`_A8X$nX|*mha*V@=_lUjfx{ zlBm%dTC#gCzj_yVm)s{i5}^JG7d0ro%4w$HFHSaGt2|IAcR zm=F`iF2S@J>TdIb2=HG(`Z{t6L+DixQCF4IWU`p&mBc=&*U!{d#XjNtoo?!cnjJMY z|6~MkM?Q&@RwK;f1jydf?VaG`-*Syz@B%G)KQ$7@uOTU0vkP9X9;F*aojDFSzc4Tk zzuN`(HDH_8BpvQg!+5R&;_wtP#z{wL*D0ET*ACi+qPrBJ1O%Ud6N(5Z1A?maKJY{$ z@8t%b-IfW$yH{+(2I*1dO&RWOUdaKlOvR9{lfL!=*`Z^T5$U-{AbA)GHl+kA5rXJ^8O5eWeMG1Qd`;ww!8wgf3?0`+l90&fg_$)q!1ON${Mz6xhB@6(8&dNowZ^Qd7K5 z-jI6uuruX+fJbvs&B|RI{k+o)RBO-}5aJU;qeJRja@s~9BOyVjv))9Kvho*rH zOXpj)B}Fo6l7Iik%Y*UGmilqXGGx-Byhy~PtY1^Bv4HnS8>Rh+xL6gy#b%hH2zfGZ zSakar9pJZu)Q1_NX+OabHb+*Uk<6gnMRV_q!^i{G-yBk>R@<)10al^kGu2=1!|mo@ znp>XoDdj&IlwbpyXaT+^RD#DicR-fPd#3Bt_v*`<)o$xnJrcs6k~t)o%Y zAev>UBZ)r!RM-$0y9r98pcTaDc)YUKrqA_ob%Gh&LV!YGMmIZwB?k6B7f`agY~0rg z-@I?K1jiI4k%!&@Y^Z_g^dh}^lTUfl@(P<1-ZGb8|3m@%G3!S*59=lX1`mPL7XN7Y zLU!Smy`2m7$0ZS#ohheOIBogJT?I9#U8%g!U%`wqvB)J;6nk2a>=PKE!R*N~36_k} z1YPM&$znuMtXxFCDLO& zGUuOOcaGuyK3|MS-oYh2N$Q;D##fR)BAyS9unb5@e?%@Au_&h!jfYJeuKa!iGiqt+ zn_+qH_ug6Fe6!(nwTy&o;#swF3qFw-8H%O(LkOsjhdG#3c6}q)(5{-2d%W8;t2R6L7Ib<~(kI!h+d zL{yxC1lI)psgcp|XxOF4lKvT7Ql$g}xcaClEtoc+{A(ycIs_yr5_;3a1!q>!uZYhw zBsw=`b*+B}pqA_8y}SMeSP~jN9{^ZDr@wW6mL}KiI1KPZP-Ef=-Ba{5%6XH6lCOo3 zC=yPRnDY=Cyz`s_USp4KqK4acR84wasTMcO7MLBnPXE=V8R$NYZSP_ z5Bzs%Q3?IsvC2Shve+r%0!=%sSRBCNdWvA%r%(>tDnKljTQ>F^_~kwG8EN}&cr&fQ ztaQ8A4KQ9^`THRhHDUB+|1knT-LOy`K})0p3`y$yj?+|WOzgw;0B$o4W zaUku@llwquC;L5al_BC(9<|8@uGKr*M(KD})%L>boK}c}8>D@Qb*pb+KO~37IDG1^ zp#s+?a!2J(^9XR96_<8&nf2__Yyfd(?6l z|J?{=ZeL!f>-=7l?!9aj^}s@l1atp(Eb^XyG-a!z@XEvgqvSlp+5W;lo(QpG zmD;mfs`lPOX|-n2QcA5VHG`_%gsNGi1EmNxTQ#axS}R6vVr$ii8AYua36bBE|I6pa z^Wu4vT)*GR?_B5H_xIfQ=X>Q!YL1)4ZmR%Vn`#g+)iKdlQ)yegku7^1cv^51QA{ac zPP9orT>$*~e#k;}bw-iqb{AU#|?(VE(~97@f!={lOD)`R`&VD=6#-nj|B$6zSO z1j%#xA!iydwqQX1k~JMBKn2pU>#lqe*Xi$)37O3XD@cEeqcD3X$Oa!u2;YTUo>P6q zTj+?5IIT@4TcUr0Wghb4Dq{<)0G|LRCSEA7oxMNzJfQ|CrP%A`#Dk}>6Pgk^v`nYQxicq+tD5A7WGHB^Zg#{>GW} zBrOI(ne;Cy6dG=u7mzH^K53))Zh7&hWw}<}rr_)6DaYzhD2bv_^xUvc7T6JMX?+MBt0phUzk z<3Hv!O}mN=jL?ae*#Lz3XsUns78L2pHw9sg%~l@+?u^G7-8&;p*I6|1^wblhd%W!v zC^d)JKCKCOji9t>0->hoPoMdLDBk%d!KE`}^INbE-uaqq@6K1_64>o;c?o@hY=K>~A2}b$Z^jpKK(gZi#x3rP$r_?v-utFt+ol|oGM0m1LH?5ddTMtA^Ycxx zSdaBK-O=`hf^H!5dqf1t=P!^OUDk~HIiN2e(V<>tb;KEq0%z(LKHi{#A|(*BM+-r0wiBbEP$hkdE;>C+EEP1(L*2@u^CtQ#G+Y`YJazs6!XpZ%y4bO#tDi!7mUK3&)YwJw9BdLYC! z`duM5oVG9~R0p*a2KrwEXGmU*-9)!jU8*%DzdjYGRd~MiE98d$MUbS3GOFgTKxm(u zN!JY6jQ#td#tmg!qZ#u*bQwb0sG5q@!UqDSQMTfk^JGZWF!rM#?M?m7AllvYjJH-G zlU~&4a9mo6H}nNfo(;$u_~`GuKkh|ehH0*E4c69LXBThx+^eHo)@ zu=5=nn(gEYS|fGwgp{v=1D^HJi-l}eMl;uM3CL|~v1OpbdeYaD0IYkss9Sb{Axuif5t0`_@1-x>bwkCpJ~>F)lAllcn7erB7%K&sr&uYmt|@pZ@Urra~E0AZBvTiDAgbf3i;-sxl}6h zJv%$RmlDA8D?vMx8+Spod^B`dU4=-$fLy|?{+r{^z*Vz}l0kpoa!2ykR^}9Ik9Dtj zV4k?W0C_Yw1!Rka+RXSvh@wPe)QfK%6MG6S3ulFlh0^B=MP2sNMP7er%nx@ro9S>p zRKV_ZPgp#m@BOQq!z06ATRx?~{;C6%dtV-kh`jU7>b`)YsNXAm(1e-^X?uYvOa6Nk zYEDDh+C0ZI1L7y46OjcLaglG-Oaw5;X)k+?pbB-1qo}j%0)@9wn_3G<@2A(iww{V( zp8uI@Hu=-`6?%~g;k1fei=^F+j%;my9m)yl48nSvB{fkTFS;_y^WRnC8|T&*r@FqT zbsG9VhY(Hbts-c5Nz0y(TYu{mUVnD3^L#g40wr%oCJ;3bb@nVb;d%M{(^somPD4-a zhO}F%wF{)8eVFJP|GnSE=uR$-;Zc()$xOBB09`XZO zzpniZ={6M{7=A6a zA00NmepT0*Ktbe%FjrTm)MkSk9t(IVl5K`yb{A;4!jEiv0hL%&)LVi!3C+90De4Ux z^2B`QTG=TAH58)+-Sr(J#<|1GUx5D>jXUJTUEPPD-fw*I<^|I&@~^&{w@V_V^RMd^N3o&9%v-&??gU2`@r zix=!)3Bg1P`5nf}S0CoG*kC z=Fbq;m9DN${p$*85chktuPy*i#)+UdZn6V19JLd*7ob~qsjct5KsNm-7Echw>j}6x zoN>Sd^x3h)f8T(FUety(zJ-=Jwa80ubpo_`2|NIxz0n%Qp2y#g#PRwMX2F;%>!!19O6_OakBQ2qyR8N$IGSeCO2#i0M%!w{q(DEyOjA(KZ z|7N8Z;-BE<_lD5)I`1>h8{P+&bYA)q|IW=A%TNM$;}SsA(kNnSWwaG-<0VL6ImhsW-oRss?$zC0uAkGFS*b){;_TcwL;nglrg6j1bLFM zH)w&6CP{kiI-@bfOOL+bvOgw^u0FLc2Pc+Jny%UZlL}hRGBGevf{E(2RQl`k>l1`V zE*`E!OD0uZ0`6aVn+n}h57;}0vlwfk&ajfUXR!xzvwsOzvFo1?u6Pgijk$0IlWP35 z%>7Rzdc(RKH3Z5P6S0V45q-MKMbA~Z6>YKYvcm^9d8Q}78QV{_`p(1Q{252`qWYg| z$mt@G%leAouKd+O2XivvV>ZZRH~Yu?P|)7Oc^nhhXTRBEQ>w^RQP&}ft6}mf$g2+3 z2oAc-L5DvYHGtp=P_BEuRe!QKruz9U;4-oNGAf>CHkaOicX8X>w|G{V16MtnUO~eZ zbG!{N|p(0#8@bfY*w;V;} zCoWuTNK*p%gI(5nTrMe_2~nRK*~tYQbGOFG=LtY173sLr%d+c=%n)n%lyL{;ypRVkFS3s?qf)mvTw778d zOAHrBY2qX7Fa2p8B5#J@F`3H_A-&x_+*~_F@3fW1GvW*}!)@f=@*mGn&_l-82mU$h zzuQ0xXO{K>%|ZF17ohBxA0@O|4&l?XBV3;*EkoaVG4aO6ko}}tC}KmeEY8wh1o|~( zfMrp2$Tw6`R^FX1_%rw}gJ^5DK*Ly`E&(25269d%zn<1jE$=mW4UkHqBkz7aq*jes zKONXFQ{9=y)0Uc%vQbuID!7U6Np)sJvS9io@b%5>mu>e+XwiMo_I>!kJ8d--ane;! zXYv^xY+q>L5@1tn?m>gx4s_#YqVVl-L)n|2P&{@mFFC6zvr-9AHG@Clxr*qdZ*D9N zf33ET=-U&AefZf5YZ+KYNTmpYE!fa3+D=#EZ0tV_t9vc-y(IcJN}Hka#n*KHT29b^ z%Cnda$zju;eht{mX_!E~{FBy^);8$~0ukqmBzFW(8fxd+%f`|rIDpW?&j6`xaOCqN zX%9BYHX?l5mB%WHU`*Nx-z=3@rdHEwaau?R9BKi{<0tH@pz(kMS4;H8X zc7`q`Sr{}%$wlAy1c@kQsLQIlOvek%p>D?^B|c~@;?3o!uQh!`0X_1sXP^9&2s?jE z43|suG4kUHN7_vHoO9Z7`$Nj%ndF!HH5dGh@!9cYMO@ylXPR=;KO*PVpf4I1^pl}m zJajkq&XxJi1~J~VcBPbutRD-BXje&gUySR=iP(pWGxIut??E)U3@P}372v3*L#1yK z^K>D>9qvaYa}e6Gg*UAo#9s^;u+mvip;km!Tro!-@Uxr(+cA`E+aPqei0U?%Xl66; zbqDo?q!F}$a3Lrc(cB@aug}Czccf_|=#ZmM`ZN0Bt+@nSKDr*;ec4jCV(G>M;<5IUp6|!((6-CYZSvTkQ@A! z_z5~^zeEG@4o_;nx`HgDOpGTeK!_{g&q-9qp>4ca)PQUUxzU4LnI{05069P(oTr@% zj+{c079vVkMu+RSMpWE zZDY6sL&n-A5x5#+L4smG!B!z3_Ur_`B~AA)1K>L{hwfJ(^~4NDSP!w;SQQ$nd}RK) z7F5xDXi4!RTH40-5WDbDES(OP`T9qP2e>mc*!(IR#LM=pj5o$4mJ4kIl2&9B$e<}t?0o=c+1$CP7!UaHC%4#7yB^CT=$5J2e%??P?-G2Qe@X;ik zT9O!Js9>!P(9|`CHXB2%`7wQUEI7{DbBv-{%X6g^d_uPP2_DR|U+Mqf9}E=5`u3T_ z3c>Ts{jfqlYtHq14=JT5=;5HS4;0TFQ7`UWXMyAOKa^O0@zB0>XRQ%WQxyfDeNV{% z6z4B=m;a_Zmk}pZ{tPovs^AW_+vL`~!-86OJRX&G?nrmR1EpeS`{qyZ+~UV2Vi(zN zK!{ovj2I|8T8#H5#4G-^}?x*1oZhqw?QuIbr%};-$OO_go1k z8)$&Rg>b4Odd5eeHW`q**Ygei@??<55h<8!lRX58bXe$A~P|-%ICO)JRY-?T(YJnOM}bGHt))4q$|rW z{R^S50~Uhau7+fc_lA_^7jV7vWUg3wEd_`tR^)-ZEJRg46p>hBk+v`EuY<1=A@M-_=FbNWsn4vNmuOESYO*??Ek44808@+b$zZrpiDZvlM0$o zu?#(BmSFaE4h@Q^_@mP?ZI;5mKzPTls>=Vc@s{1-cD#U?8F8>(u2L4L#4%lQaAZRn)Dlg(!Tf(KS zHNvccY~QPRE1EAG-2rnqLxf76a~}qaR4)uozuQTHvLDmu_PBP^A0W*SaLDbEV~kK@$&9y$!psz{@dh4qOWw6oN(O5I$8Xt84en1vw*e9 zfvk2d6^v%^MK*l;Fr97|pB+nqSje}Q(eAl6?S=2B43`WJkX4VELy@3LYWZrkq|~Z% zoDm*;ckjBRO2jYlqZj6O;jI#el?o_JwSJi88KGyFZQ+je@4+^~cKDj-1WGnwgnR>R zC?6vr@Bs+7Vdo&B<%P#1jRGn3eXhf8ug-BiWV{by5q&a}^VWfi-dy(tFwqn6nn(wQ zbB9=Yr%U?RA;i{6$vd&Y%hC`^W}L#kUFa87kv>Hst%W*$=s4sj#A{!iPy?3enBi7f zhxEIGhH?~p%ZFFV3A`Bk107q?-YDMrS;XJ|QsxRT4D=|_<0rfxRjK*S`s|AcK^t;_ z{O9oRyuOZGMvV&KPJ8ns5?U6hY(uu8=j~&wqRZPaO0Pod8T9o0B#GGQUn6s}A%Y98 zl9eah-C9lf^{IfmTm}&RS|ujn$EP@hcG=|$r2l{4Wp8I5cWs~> z-h@(C?_k0MT5j0`(`w=R*9bNJL7_MBA0IxpzMzMxWF{S*lf(yzH zkJxUJ%7KY)>r}30q)|VF{667%2;Tr4I2dt})ed|VJX_3&1Lk*QTPVmphh}*0y37#C zuBqtDqBW35v`eWV_Ee&RLq~Y&?*OmK0rYuDr!ZBp+gn1-%v1eZ7Q8cl-mx0wrVK$6u7L;3D4u*S2sz7-^8CfnW%nE)8B zZ^YI@N641~GWOxB*vCxFycg(hxA|pTuGOJ7Qaq{7rT78(4j<%5gaW&J7VM*m{{Vd7 zK|QM2RdNxZ+CaN!+~?9Czq@@u8MzipMSS)jXt0Xz3*Q-ORChf={&${nx9j;cXHHfo zZ)?btA(@p5Fclbd9rRU)GYDFMgAOL4S_^ppOd(o^S%a=GY!_ZLzw&GnAhYXJ+ijCt zPNVKm_?Hc_auyOg>HHWxyBuW#dUn!A`)49lpRBx^kP9Tvoa!$~j zdP;Sk@@MH}(gd7Y-%QREWNx}VbeN0mqjv{6w_I%GIK!elqwn^DTCHf<18Tadt!in+ ziMc|puWC>kwKMcb^!1x$rUKK`BnR8%+I*nmjV zA-0sM*OY!YdLXWY9YW-%=g1oOF`w5wp8BhF-9la#=d0D5PziSfrRo&sPrwHtCUOG$ z%1J5-Q1*1X{pDMCG2hqj(%tr)3wlEXB_&@0#A}WHcd)?})#U??yTECxFgVHyCiej3 zjQ((mymJZfYzxeq+!vG;MQ?Lpj;n579vr8MSEi_+h(LG!^2OP6!BO{m6pIuRC^gP! zy(wH7HF-RvaJ4IjRy_;Byc#Us`uc~1%wt|cZ9On4vd(aTqf)pw!HCC8=FF{KQ|G`hb3D>O; zKudlObJi3fsCVOfMfD_bAXf@}21iXUC)2Ux$AFN(Uud$n$eUvcbbtbChY$hn`vuaY zH&;sei}nL8bbr8@%DCKJC`3>W-&-9!ki{8(pvYLi%&>QLCwK>jLqbt2VyAN#8Q;AFAEv87L~H)a*2MKzE&_{`1t}DC3>6YUfvcnKdF$-w(G@20x(y zUCTo_c0h{?zI@rG-+yrWw0h)hZKYB>L`$FJf3N1SjHAu4`ybZ(6 zO06J2cDq(>pf|s9WOUJdWcJK$!LOg-^J&;61<1kC>E#c^4%?iUIP`UFsy#do;4K!% ztfFa3K(5CVDrl5VW(oDIqo->c&Dg(=!W;6VMH^@+#~Iv+seF3gGt2tt1X~(oc~RtX z0SM}Q-{SxqvXpThusidYsy#uguiX4DLOo=BzvE~D!Rm7kI=HKEdI8OA^P(F_JwdSm z_CvTSfA8PJ-(Lm+`*Tt*ysn7bNn{b)4d>54)R`-2Mb|h0*ol!^@ebAfy<>a_D0g=P zA)U5hx@#x+-F>?RO(<|NmnNvo4MITf+m#8&9Jayb{?kG6%>reCM}UBCkcFx2TbktsYexbc?1u!lQwRr*ADBz z$-Ke?YCE~(zw-y4=Mavmb_hC(K6>o2jLiPibcpEcw2eA z3<9`OEQ7=-;Z_PE$R)I-mS`8KRoA)iAyw2Nhs!)%{ER@Z+saX($Q(kR{@@ z@C2=wGNk3fq)|-I4vhB3>>+j5aCV-)WP^--p4#IYGOHmOB-}7etF8#{?B+E#>g%vnBmVE z;L{s3(A30av9fL9-F=P~b0_f9xd5bMTxVoXZ4w|NAZuc^95NBZ?B<8N&+g+%V@7U^ zvQ?(Pj;Qk(uMmj3?5lIqEt=FDBaObp32gSpm_p-Hr{TnxhH_ZwHW(@Bf!nXp|0WZR zD|p;0bs)u*_wtJ|UBe`p`()5_$u7Wk|G56>j`}ouRpR8iclT~AsG3O{R z&asBXxJ%5lDqKWZEpyP=A~ub|^c}!og(_QrleL6y1lQ$4ls=0w8dPj^3xjC^B>J6u zF)a_LpBN>^bxI2j*_qMH!C%+aB{e)bhq>6GYR9B zAy?e}nsP1iE=)`GyH&R#Ov_%4Y9Nz9I@IQ6-T>ryO_WLF<&P$ZmTzmAGt9z+i75v$ ziE}eCDP-647X@*nLt`yI7i{lK3Bh*`$ zx<(Kx6ZNs2q(lD3Z01>E)x9Uio(+wQ{Ca=F{<%PmyY6M3F0+yEjLRbCzpG14`9&+& z<|5Pit6yi#f82^W#S5$HnP0CWAsN4!}G_KS}{ zbpcCB4T<&i{_L+wOE0uvy%ka_Vo+BnysCph zlCIS6V$Nn~=}q|4GJIL4_j-%Or9DF&SmnY%iRrrEt>DK+GVqz;;?J@73hf2cK1IaG zb^hWiVUt|EWoU7SzZxs-n(ULzKu`2te)&viTg5cq-=w)osY8J#J=A9{uhox^H5g6s zZyqMCgu83x3MJfC6%5Hpu~uxaVwRw!3(@`R?h$C)>nS{&4J6J5t`!8^xv?fJKEW+v zs_rhq#|#A<&2Abd1~D0GM5Wve`SbH)L)c}=!MIGx;`&G42ga8K^^n#_1AP|-BU&(} zeoOGH*-DLThmZE5bGspKzDEZ|_f}3i(4S*D*2Z^j3qDMP$z z+`f_GAOVfwgd$Pi#$$AFV^x$9d7NgOKg}f(89~cYrqaXe_(0k6+S??cfwU=>4#~D@ zVq9;C-_*Q3PfXNb#9TvhrH3w|GlQB%aGtVVa38|eM8XZSymxb&x=HDUcG;*Y#0t)n z)guvkq5>`uO&i1Lwp+dIF8;88zIZX>%q9HURErmE`dt>U;CV$yX0;;R#Y*lDI}T&1 zMNy69c~yZ%$xxeiE3Q19;38w(pyc(G%9zeC=_>NV8iY8GS&HUTIp`Da!8@4}BHqS6 z5t6+#Ut`>{nBLMf{W!Ro%`uIBTZnB|TevY)2eA`|o8QBP0{{>Gc4jWJ zDmpoS^Ah-1#|MtUy91bIrmr5NSXUADXo96G0S^=LkyM^?J=npYw{K&Xy-INUZh zIF^{8L{8_J?eBkFvPs^QFGNWjGnMZT|Mt#be<<$=c%x*=dFE*}~4H|}cZ;p6J&*Myo(R=4m| z35xxkv=CLH`JO7`%F-;)vWSY*7aY4b0^h z6H5(CF_8>+$;-yV>UeH(Gu&VZ#{OdL1mX><)_6IjhAy`FZsYgHw(6+gbpmI5;n}H7 ze4SRXsj)h4h4|vw8oG3=D)%+#z3}LOGKyqkTRVW6#{H;T?&y3{eHi-mBzb0$V+~sy zqx`glFh=R*Vcw{o`>|9w3eYxT2@H8sY?QwmKNo^WaopXA==d*L0U6NYUV%5p*S)Y7 z(CtZ!VK`}tBpef&j8tnC#9ZZ1%UbPj--ow(>pURZB46Rq&z+YGyvw^+2ucrGjP;eR zyS1m+=p@wz_h%5CjnkcQlrl97tZB6O30}>5^}_poz`@&UP+(teiSKY5`E?b=Sm70u ze+tcU>?2YL;Zb86*1VOe!YnABY>5rS!GGeNJ!m}Z&YL{WU+!WDaHGNGUWq@&gqj$A zHaQJ1U9tqC>`}$BDQg+Ki8O`d^4Z6R#mm_-%9ED5@2BGvELqD|ZGpMVeDOP#&lk00 zj%LLkQ!Yia5pHv*1)T5BnieWKKcRzL`Ufz*`fy`$jX%}bvEi$~#f>LLkehTjzx(sX z^eZA-@Xi*4>kc=ti2N58NuT}2zelLz-rm@=)EQe1vm=Z5MA~b`6p?W*q0M)hlC3_Tcv4*e4gZ^oV+rzJ>SAWf&1}i9NeGr<3(|9ncQzOUz zb+~Q#tOi5-6R!HW+zZIyRKCZ)6)K`t?!>)po2~34G0I9p@5Mf8ytH5^DW-kUd2PSl zVPRBhj&@jM{T@f>VL(Jci+GVkeqYP#2EJzUiXv(3-HGd!o;%(eCPuEfd7`3gXFyL= zJs@(_s;u*|&L{L=j;jPjEGfwa#s2=>RmlTZ_i|1!$A_1!G-E!Pvy&_DFBlo=dmCLV zxz1s*ahd6|wNVDE)uXXe=}Vrv#}P_78LmnO*pT7SG@yB2e!be<$?@;D zL5;+3W-!l-2;~ha-fh3YhdhpK=fpeQ?|E{W+S=AlJu5PB&7Zn7dY*jmFJH_%DY-gk z0RQ>3ZPSI>v4-2VSB>v6UH{~OFrSc_^#L;PUR{Ix5Z~Cwikni}y5_7o{Qs7Rn^#pg znD7*OD!jG_byQ`>qSE*PE_1$!Cv6?@a5sd657{soc>Nm3A0JsiyEYuodhgj!@dS3+ z_DMs;**%6!wsvsCqSejDA!qe{tjtaR_QrE?NN0;&(VKM+!Ky`W_doRPf=+ulw5gj- z@oqY;f2)~m=pt~r{bgpyZdGz5f2_}YCJU2z+>eL_gvs;8d89v&Bqy=l$m!AY8BSyo z_eedepc0a16`E!dSLgOA)=+AhA*FP4X#Xr`r&=JJ6nD_ z+ONUof4*cS69S_ZZw`O~qC#FfxyR%sOW$oWn(iOmRYv!Gyz(;>URv^(EhA-9R$_wm zPRt^LLkZn9j?lP=TSNFOqL8)}G1+(#a%bjw%5SBn%Q=k(kz$Qc2&$2=L_NDg%S|Kreb)lxtPXx6 zUfBduwFc+@jZ9ipIC5V8bX|@FUufwh(v#6xQ-0d^(bfAVmak1ISZNNf>?d8%n%Ry4(F*sB1)3z_f7 z7Eka7P=foW*d5awMsWk|jy()Bdo93w(IMnF8{|`j0zmTN5%*FQ)Ei3p%DY{Evwn2t zBMyH2CYD`L7H{*R_?Vm!t3nAZ8K=Wid_>?^lYyu&BOXuV_|2-yXs4TaGj8I-KHfg? zYwXFcScAi(I}&@vHMDMPgDB|(T@)Wjotzpae~!*Y#1PV!AOh9=sSNXU6R@xNcZGM@ zjPMzVxFf1SYdnK!9GX#ovxvCI0E25l+R$@cZmf0u(WT-Av4KXVq*ZO+7s!2^3*ufJf~68}zamYZwHcP)9;>Dms*yZ&AmYQ$~CV)iamJ_Z=zh{2gG9TcCF)dKFZ_{gT8 zmjl-A>kfb+3%ThO<0%Kf-!wfWXL<^0qMew)K*pa4ZSuyQO>f;3Zy{lYDd`-Kkr16f zi>2@0n1$amu+)1qzwO(~XWw2otqe%fnJ25|iEV66Bf{DrC=~(TUFr;PS58BDln%}! ze8F};+223m)3ekHNS1t}UPH(=q+YC}bP*8INLz{mX2Y&>Dv34fHTSw}jDJg}RrhtO zpPhYu((LXqJ~=rE5$@zuSXTu!-&lJ?y)~k4ur>=!%WHPbZcYzSq+iwTBV8KHiP$GI z?FkuFi7y#aSD!z}f3IRwZ5QqNYaIqlu-WLVlp~fae81;#g9yjU(26?{;ZIB$ZXu$AS);AvHhk4 z$kW&NVv!wrDk8*k#5K*h+qkR0?`QVr`4&&;wMn;!==5sh07qEa*G3Qt>wME-6nRHH zxHR=E#Ycn576fR>~LuDpSvTi?zi@I>~a?e{n)2zEj*n zN`PS(UHOfL3piNPQfIo-eyCVoyaq3(=#=`3Vt@}*1yw?i-vNG%(06?-p!MIxH}Hs z_Vj~M8bL&;Q9&U~g(B5sjHRqH0eJ6#g2^vR~WqZjPRIiT`t z=mIi7-$H_j4Vc4)nk}NqA>KqLN21kaJH_X)P$QZ8cK!L<12^DsEIfYf7t-9U#xtV_ zAz?{jr>3ZUhIw8Y@MKVTcheyF1zZwq_M}*rQ7Y5<0}>YpaRphC>Un?5Z(mW!`})O)$^oE9tR#i01lTI6!N4#eu6#5b%-GUAj-bENzr!z zeeH;c^&f+eF&|JYo^=~-+jIY2ixbf0xeKlu!EV?Si$j)vmDzz#cKw5$pS*uifX#Ph zcY$5xe66ZVUl$qQ6a!=biD)#kVg8Hx@LP@P%FK-+WWkEodmy(ngbOGisvG|NCiZMv zAe`%u{l7LXU%m}zKs9TV3_7z@yl$if{Ow5e;)8i^nVTcymG6`BssANOD7bNW=8m1k zj?>hoG;4N^!-qeLHF?f;E-KEpq);m7y;=*$jdkKXSFYo3lWe1oTqp!r7Lk+VrG6Gw z(asi6;?|=ilBKN3NlhKro@NF3x>jtwYbkVnp8m+}y#Rjbsz(@;(@onG4bxi3)%Eubn zcd;Kt?&NEeuzynaqB$s7`E;^Vr!F;lfHtJ-(Jy3uE;3>9#S#!DP*4Oo6po|~B4_sa zXghNRB8jlrMY9zVEa5PbQn~q1j<&O>(q9N?$PgcC zg=F0wKvJ?cjh+F`x4z%M)Q)Jh@h8Fo8zL41chjJf*)DP;JDsT)(QOJm9la_29Eg&o zw(%sn%G?PQ%(GvkkCyUbMmq0+vU#Ob5{F(7@($;CyEv%J&+4h1XWtuQ1WZcDzKTI! zcXlVzl5IZ4I$WlIpb?zA2?88R`J0dX^8ns1VtF5#f`#`ZO{c>Kk=vnTYb1rd4?-4- z5LXT5d_~qHXJ<)|Mfkg4$js^vPqt-r88h_Q+jWnZwPERM9pEtZJiY_*Q0rMm0{pJ zdtxf|#PUqG&c!+Ouh=^N?SH2eb?*1aqmqHy>6?8x0H!Tq@c{?${{8&mBMBQFXnFD$ zn79qHI&yA#GPB+}<-`20izq4b6N|I?X^dOD$8rCp#`om-R(&8=3HVS5OQ_*BKK>+s zJ!Tp0f~9QgUZdUbhBukGQrs0NIc5@T@C7!x^DejvF^W%#C}Ly6)_)-F6H+R@1HP9G za4wc2*?JK>-XbZ-WXZbUgPCgq?^u|EYe=sq$O*1&yIAHw?_Np~%kMt_t+^=Tk-Y<@ z{QaSn<24)2hBueGxh=`ws1Fn_3cvJ=^5qMHA8Us0Bw^r?r;KdAfa=wxP$fKk*;K`E zbGzPD_9<)qPqO5N>C>7SlRRx<0(wA_7@PL#p6N7z`;&usAT)}$8&-b0j}w`6r!7IE z%e|89Y0CEtW_gOTa%=2Oj#?dG1yv!DcCH&TG4GCevzhYxIRh*WaT1q609hfGDwx4& znI*aS&RFEsuxJ2<@TOnB+Q%~a-J9C43l6OY40%85Bi=q&Ykfw-UJ@wn0PrVtl0!X| z%^FjVH(85-vX0oh#rgE$zf~8)xhPWGaD;av8-QlrP=*5NmrbUTr+{xkED4Ku^ki6r zLmF9$@ZxHO&}dsimMFjINln^@GWQ5&pl$KKbrY-Xa&NvI8#U_VN=s0j)O6fdejqtdmZ#n{QQXJ?X2<8g-tlnQ&Nn=lbeh1tqV z837NVzKLtveuVq9=9(it1&)5QrZv)b{QGq!W@s!K__9T0b7Y9`gb%M?WpKQEAOrCK zLbmL3KD$K4q1=aTjvBYedw(IFbUQ=X8+UtZ{*$zFB(?b|ZYr}R$a2Y?)Y$pmraA)k zj(k_j!A>g?tNT4%M1%!Er#B8)KFhi*GBd`Cv(DTn!||MvxaUE7#@NL$rQNtkoj0{CQf zAdHLOmD_sUo-lgm6pi0zh(wTyZwn)6b0FR4#ISc1_|NC@R(w{Erdra#eZRoZH!p$p_QA*IpKwsv)f+qZV* z92CE{^Ig1t9G&=sCP@ama{>OdBwPJP6#mNNuOrULn9z-HRu{u~7Friag?gCYJ*?!N zQ>MgdU4N|3O0ll+`6nAmAK&>R_rVT(yU6H3Tw~O^$I#;yEX62+`{F*VCC31qZ z-fdQPO||}2(&?+up`shO?9d;1+JU@tcujGooBy*ww2X<$mu1Pr)BQrThP0QV?W%bQYCM(9g|&Ht|w=(1*LV+XZcjwkgP}oi`~%Owimgg4Wq5nM#0g$a{KBMoW%0Z4ygyRp))`l%pOio|&GOK=H0sHVigP6j)G19K$}`vBGDsM(1bmB!FmU=-#H*>fX0DeUkMEit z_;hFs-TcZk^6;=ec&euAFIP?T9Sh33*OzEBr}GDcBPLOHybj@XXNYtIk9KV#o{bW& z^)%)?uyUc@{KzE_%^;PXHx(C*w0atbYgxaQOg=_7=L?6c1{B2gg&sQ@j9?o$nN+Be zKYrz3SDrm5B%=0CDYcg1Mu*XiGY{?Oq9knXE3Aq#e0VS@+?L9tU9 zhyG`i`$~X_C(iKk5hVacK)S!BTA(JIBsPdt;`bw2q@v1v^m49il(J$q?{~mrM69jl zUAwky*Mh_^?zKB|cr00TWo!8uF1MQI)ee*?HIT@K3 znJ9s`W>l0Go5t`1peQ`i!DHvvn+iD0@#6Z3iEg0t*X?*OI7EptVxAqfBlufoXTs!c z6e~?uCb!Cw+hY)S_vH937}td7Yj*mDT7%_ULM!^6-}<&nKiIlskZ{I&F8^TH0!xR} zHdY&=Zdpgqoe7w|GT1V$TT}{MW+OxQf6tpO@`m2cI$W`Ucoqp$M$YWylmYw(&VzqBRlj!5a zt)j!4^Oe{ksKo8uP>w+&J;Ly@`fQ@|h1eYvyedoGX?0xBzVf`)5Yogqn(c?_khaH- zxK4X|-EsK8rpmB#Sn3s@xzD)(@0+FUP>#~!HOAynE5bR47rg8KSz4FrA}$VTuS%Tj z^vVIa#QFV-f|-B`%Qg?H=O+u3ZA|xmG{+W1<3x5XNHz3k`s`@vkj6%03C~cBvk4m~wx|q3o(lfT2mGMbO|UkR z-63J81CqDb-lwjG#%`~D>QCS%69ndez!o&qx5FUzDe9-S=GTrnmsvJ!ba)J~S$DG= zwlykuTRD>ulACgiWm)$vH@S9mB`Qj!Mb>@qu3Nd*xvLzDaqMR8{Pz3%V`pY(-t(T< zykD>P^LXsOor)msswMBf6AM}`>QfW8L-I9Kc7AM~Oz;(tl`@0(94?$=oAG(212H!Q)w z2uGjv2_Y}vbq7xyC&8GLNZW4KFk-v}l`H>l0$5W(PQbgJQI{r^Ga=T+Oe}+ns!Q0C zAGU_}RxE^V=ERezyCuKi9Cj`ffo#o1Fwt9&2oM4Fwt%1sMg=bLZo_D+Men_t;2w^G z55p*4^)))Cn+c?wpwJ%1}cJI317Pz4p$|e(_1edUIe*BZS~^J z^3$k|JxL5yi#U_3wALodf2}DV_h!ygdpK# z%bwu;qVXYKGY=}4fp>oBv!Ic0xx>XleFe?@@%)^Dybq=3cz%MiA1-{ip9y|*eV>!x z1CNq2_CF2o&+{NS0)JQt*vW#Za|IPUQW2uborBK~9n4m-@$k8yOD$n3q@2ByS*4Hq zK{1OhV{N+dlqS-7hmYj#_^gljU^~vB%gFFcGI8&fiq$)l2Cptqf_?f}=?(Vb^0YE^dnT1?co89T@NCPhO zaN_4f>5k}Rx@sQn;$Cs&0J|wUkCFTMF<0(g3&%n=x%N32q=pn)sFt4JWr57D^EZNj z!y-8v4(+Z~zIILy{`Y#$0@EKy!A-dLLY7shqmb9u-?-tgdY1|$2djimSk{WPwOG|- z;hYsllmDw)rLb7A%g&Sa+L=QK7GA!yH7y0Ad}ilDYYhqJtXBVWe*)}U0P4rcj0^{jhPE@ORhU4NC)_P{_Q^Pm~QueDOcJm8*JR^px|P7?4Y6Vtkq^Hu9S1d0i;Dn!N3LX4E*RR$)K& zAVrh7Ax^_d>&hlNEw`Ab4{x+lAQJD7EO+3q)!f%9I3|WrJXng(A~rB9_JlfypMdFkG6m33tZM?Zgl-6ezA?OzohogbZCdVu;4N6v-a4gHb~JqBI(hS1D;1Bf8y6dx86?3 zhl&b0@he@vT+22uW9ETwYbj`6VR#`#DrwOk49<^wIBIhfbnQrm4}Wx{@N~t9pctr& z&vxrtsyn~v5g=k~jL_rly&;JB6yIh|wd?Oe!e>*L;Osw)0^Oj;y^t!)L_#PTl2+n; z`7#YBw)7iXwbt#jwy096s-Yjqcw*fb6o2As=5MDjXOAKJVpjFs%eS`MT0+6eS*@+< z2f?klY}GG@7x>PF5@HxlGOpO|ORF>%sTQv`m#@#2N`Ibs1|#NdhsclqE)G;AK|D*{ zb6fUDxax)s|2h_crZ4=0hRh3F6*KsNV6l*A7~UC`pEa3zZ>o$oym`Shy!+2Ay)!n^ zH=@AxL(i@3(sk>A1^&mH?5Ie9!f$+uMsXu;cONzb=nz{??K)ain!Ae%Qdh z1&3dJvg84fy|Su#CswaCEZb)bu2Y4Q6V@u%-$-**YG)kU{kf!f6GpTTkQd_#?89LT z9-W!@3ujDx>-G6)2W}W&kiHrT7tU;XT(4YyX&BxNvnw`VZ1Mk_5TBE&de<7}{T2N8 z>mB1)+HU#+oNYyl8)YOZ5m=ugDny1j%zkjzTYrPk+A1Q~YG)R3fy%SAjU^>IrX^v> z#Zjc~P3s>Mta$Di0#iHu@Egcyt0od$oh1C}I z9oSB=@Ylm%&3OIuHDHF73iIJ$57S#75W09y0krN^|795mm#FwmROtkvk0v>2CQes% zJK=<`#M3YQp#|V@u(`=164$I9vL;n|8q~hDQkOprpAz@idQUq4n_dWpqM+icG2uz@ zoGzP{GpKM~v8{z0wu^+Dme)U|cKy>u&kLQyNS>id{AzM~E>_Cvz8$asHV9n)g4?v ze-nG7oHVt*Bk>a5ORQ7UHD4{?Vr-Kv#{wR)6q6GkN=FpQ&|O7DX9=7jZl6e}lMEdz zPn@VuX9_Y+aQ~Bfu@7vHYa_wUPTI%0fTc_ECsuf5N;U#~=B@9&iAL}$D?bQ^RpvMv zROtwy?`)a31?ZY1gwOg({5{Fn^n&fcC#XcP)eG8BRvEAucz}3=sV=+dq>RoRR>Ng#aqGLgF zXVzhmpg&6Gli-UFC;81Rv2z0sQL%e1~{l(Cx^)uy=x>v+d zGqx*~d7@R|_^jUr`Rc3|`7F>}5TDx^apVv{xT`fh2e-b2S2{T9(--ZS3YmdLqs@M@ z8#A0U+dwDUkb5AN;p=Zq8ka-W*~7tMtpJZ@tGxbRha5~A$y>((W%O+#D|r`ZY?Kmr zR@{a1N8L!~Fmo>IqG7dn#-M=v_xHW0SRP5l%tfu-^Ngh6{vfZ2IMTAeYjl=_gN^i@*Cfyz796mls+ct!G79Iu z?Pc2Zc`&&4nXOT6I57)046ix-zDhR%!?%UW+m};noz^KPcHELVws_P-8jz6st_Wk< z7LbNt6S>>e7EsXEk}~tuy=WPd@745-d$5G`3^T}Rusudpr$lG8>_^wVTs88diwtz4-RJOwh z75c-B_&RtR=x3&XWDLV=YqaQ3^&0qbmP!=F^N4M=v>1i|qIX1O0XLrreV_Y>ZA^V9 zKMW6jxl5<*PZ}@jJsc-(v;JhYmstH?8!@*jHxDCG_4Xoe0A1Fr^Ku3{2DfRdmO8=Pm+6FavMPE zKM}z9Vw2RqI-1>@cM5Iebyl2Iu_=n-Pt%wI#=ILO zUe{G6K@JOb{-C$zu7FAgM5)Q6k8s>7q{)27f;6{fgL~V;vuD(LbnMpeOkEk6M+_yp z%lI=b3@Qg#pC&+T-|rkycNSy=TH<cTyCB4FkD}PCXAAkyV#s6Gge#3KvtLm=acb1$L z(O&bm$UvV=f5QyiIj;MMn2XB)+d?LB-8`oC2DQ_T7J0%;9pJGc79E1QizK{DPQCnC$vgVk&%l5FGt? z#8g5zrnuPgv-DZ=%IR7x1vgpFcCvQiHmrAf6b-=zb%Sn-(;Zae(JR*2yC5p`(^vdl zl*|0%>jhvdE-pj$%T57vbL5Ft-4<&s9T!`x(Q{Fg5^oK0@Q5XqMF?_1Updj%?9zmEW(xadmuu znH!C&*?z&cRoBZ8$8xX7bC@cyN|;UiwB%}VI;71kbsL(JseJwgfiTkg)Wau0IGwO2 zS$0fuzKkVi&zU4EKleh8`o|I{{}ald|}WMeRU^b%g&h zZf0oT1|t1TU>EjL^Ni6qfgQD^pbo+@)o7wY115>%U^=5YLdLVkp#JrT2U6WOBlrf0Aq#C*7uuVE;# zyxFI_eClGEUzaJblRu~j%7h>CAI9)gy34z~1n!MlcU{Vv6O@B9#bE~5p(#^`MG66C zx5}?rcgdu;KD>U-&YRckmtEbN%vo6Z^DD;$3iZsR$fr+27gWC;jsOv|>rN4|y}mfr z9~GW|hy?4v9-6^th+YI5Ti5q0HHaHhfiaH zKks<;ZH`aAfAubl7P=t8h--R?t?C1u4R}J@N*XiIVTnEE#-5Rgqq!*NL7PbZOKdE?o33 zM^N}DxYN%awI249#oZn*BNQP8Be~%sg*-8D{8euko2VkCM_L;Y1MOUe`6@U|S?0q0 zqr9N}VT}Tc>Mid2dTn|@qJP<|>$ah!{uw|HKwfX^ngwDzAEpy+J%>duH-JkWA0)H;Xvv7N>5sU9X6}E71x2z+x0G<0@Q%@Ap6q9J~TfGCb<<%08 za>Wd1WwR9HUIh)P`3zn6`c)zWxKUU3urpqJ*FzVWigUj6sQZMU7K?_HHx$@vxOQ!t z#xL-)8sLsa%5U_VEmIC}1pnAk`_t6Ifm(uc+RkP-DmD}^-s*s)mhW6)p#2^~ew~hb z6boF`@}>*$0&6+xJv%c0{~r-8OXtEqgBz6B=?EX^R0}!KUX`-D9B~-2h;)VRw~mDR zwMZujuB;gNX;eLab^HC@8LtvYmN#Mpg-dn6tiHdr?sCpKLtIiaW7~ch2i(mZz2Hid zY*xhd7ka2LR7yMitZbcPYZj6>{sX0i9jY3#(ELWu-UWKRPNd6z(!v%5%38*YnIv7N zsj|;BaV-*WUu0bRsP4nojJ3O~W^z7%Wy;8vR&_I*4hr*GN7Z5Wl9TOW6$;WPywhb3 z?h#j*JH8qP2=X?OF@?XSqmT>MUC(UlUE>4NqX(CvoQ5)YD?AzMf<{}(12d|^r!s{P zhVgmDpSItJw8=zEu}Dm2cS5fPGpt8f5*xFiK6EdcsU2K|`;-h=r>(P4Gsaq<;;u^Z z5mfM*5R&5V?Q=v6dFdIi69F+)kJrzeBEq{i@T%CMCSsC;_{nQC1)yp-Qa2TTHOQW& z`lAxNZkU=yC-g%f(brfGe&^TAQ@HGll4)Hr7GJdbKIEKc_mNEvm7XoOFU-GVF{t`- z#jx?VAA6z0`SHShYEB}q<@htKYhoB|2nRareS! zGGj+uhy7gTs#Hcg|99>EjkLRLi^y&ptHiwZPmW!+J0@As-5 zPg+ak0;hH^@xMZxqgq(1-1!W>ZV|Td#IHz5Oet~M;_~MEdqbP6y^@Cq{FD`o%j$jT zCF+Gnr(p(3)*04Np&eY_-&8o&Ur=R$CGOLbWejSzyHX1BXbM}jS!+i4&G2MspiAe9 zhZfKHr#Jl8a+%u-a0#3u6Uz6-#wW*KCyoUKDDKrkDekWty)#TD4qxwjTyLcSzAAGFkJxWF^_Qd<|Kzr{qUwmWGpYhEo9x5+hq`jNaeO@-2{X- zMpE=xDtor{2PUX0%g5VwWojm=XVe4?0sq6tma-PFa-Lm~0};l5&*7S)&@#eV6r3Li zc%J4MNoJ&->xIOtI~oYMN5lah5;!n}+#pev8d!c%AEA@<&$$RpB*MQHIBNm^-f3c9 zA*&a!3VMvRzVtU0j{Jt8a0&ELZXkSm$Y>ga=_y>3R;n{u{y*9k+RB54LY{#_3Qb`2 z?|EX>?66Wh+b+!MM^s(5BVXwOWpoqR+acmbeW8y6~+^ier zLzGh-f+F`NWoQ(*Q|HVEa6NudAKz8dz*&g@P&4LWFl8?)i|_fNqh_+{{2$L>3Om?S z8qGoNN-&c|Tg$a{?NaHbqY*Y$))3N08Ke4pbU@6U4;xHH6hP|;&-}CoBIt%?#Wh{u zyOLc1f%h){(piK_tqv;1K1tJ>VjVsDaGPAkTVQZ`4Btx!0owM4!Wpf-p99T$IoeUM}AT_5KVGFF3O{gW_!(r<1Ij$~N9AmUyR!yaq!oyb?J&vxsz&4OC&Lay^ni)qVjXHp~0V@g6(2 z%bnIL^xFI)0n*pB`iauW%sH@4soP10E zBYf+0Vg9^OU*;6m_+SlML{5zSZRDV9 zhiZ!3QR^E28Q<%4M$)c~v~8zQV;xzHi+gs|st3tp_c2BdFK@{cvmqo=8LP_}qf4kz zt=QXiMnCZ3dpdO&f9-#N8R_C>o-~`TQOC3Xqaqa0tQWst;>0op{})mkO6sE~xZg0A z8klkVwSot3WCLGfg`^Zf9x5Pg#@VaT4$;xN#mH<%LbH0V45+D0$3Aqo|9en^umk)* zOV?V4Ey&JvMEi!ma*s&5yttlqVkKXI8sek55gm4Pb@|jlA~C7t1=TYm{ni-)^VcC%X;@68vopx8bH*sF zMB!!SHCWdY1>(;<%IWsJFn^Xad9eiAoqrxf8Vhx{@7Xsvbhu!MGMe%o^-YHJ7vtWf zx;xY44&R41#{$2Fr>@aC8Yo$T86H6{;F^BltLKg+&T>6tF2)hQ{kW)vM7~a4^SP^r z_pW}`x`>>D#dHI!SMl#y{#T9}k31rf_#KI9YX*#uS??W^CNNulrXc1W<>zO_J5>d= z70Q4cH_O8wP{MQ0h|{pTLRP5aXyl%X&epgJ8zIBKeMlBT=s1*(?G6P6EdWJo4vuDjs)j%Md-z|dx z^6tm5Z zFXb(Wnh@Px_+3z3FefhzGo*VOOIHD@uijBUcmIk<10z&>>z&lNHeP1T(AX&=@871@8`#L3=izigTE~Fw!}@|debEw z)4@>>rF8-`3%r)Cj}L}--ZxY;X=e7;;Gut03F#I3^8tN_9rF9Fj<><|oupZl<*cpJZW~sR zAyXbfDH*>ndr9uL<_`BoDk1HyyGTQ;#}nRNZT=5wZ$A>+aG%4b{S>E2#yCZ;zlmRj&@og@@>Qp^%B&QX5*|Z*yer25O7^g!=vi4k&Ck@wiIux;gc~7^* zibbqB_{yI`vx(58WsuwsIa-Bs5}ILIpB`z6qqeyJK{JCePvg~jzj|WoauXfXL)y}o z#F)uq&A0G81z;-Q-v@xdp*zF?Zs&h>t~=r4e4B?QSuC?3aR=Zi#Ge*Y^C%6K?g9>9 z@mJo$kn+S`N;KrrR%_>A{x1SYbDmTte){$x`bB)Pr;hK-X#h#hSr-A;NAmBDQH69z zG}|FgfJ{~<#U#ku=3(`SsESK><%%cGDP6O#k_gcj5V^%jCiXC}hDE|Z{YlM?gcq7D zGBJ^>V0LYqp)|Qmw^VhZZPrcUrmO{ej3_?|1A?9e-a(azfu8gKE{zvvox^2!rGUz`Z+|)< z4ktthToV|D@L{#>29Qyv!7P^ibRTgTwht3YL4dpyAe;%LYkleFg(4zO_8IJ+ z5y9*@_h_A>dpQkGhkQ~fVkh?;OhtTpm1}dy86)Jw;E{WtH_8M$_rk;#aNg>DZQ7( zfEoJ7g9^h!U3S5V2z$f=FG~dJDGxO?~G)3{QR{ zttmAOSGkk<)ubd}>&3GjDw1-eOV9z)^bz%D7;aBg>4et`>dgS~vkydm!%KMFumvXE z0jNpZkR#(F1=%fjK{h>t_vl{1PU{`ZB%h12>5Q+vZnQg^2FunQG*F?-R*tK zJ4>FXeh;Q+x{5W95Nvr{G%tY=j0z+*-MKE4RDA>CoWu<;f3}6z2p%K!yaiKVXy?!s z3xn*Fg?oJ*m*qlpYl?rUt9(6FC<~&>{er(ja_Ff5yP_JYU%(j+nLdii?6MRF{yhPz zhVLBqw&lHeyaqWQR2y6cH;*6t^BxPndx%rq7-glgtKkvgjBd{owEoZMQ@bV+rGkPd z(7b%&tWd+Btru?|>yGSD>!zblm+ip4??%6t#DNtYJg@dkhvC>gv?zf5M)h&Q!p+B@ z)%zHR(_|vT`5#IZ$XY=cUN@mA9EHp@-2I5yOEk*%w_m-N|5DY-knvZ6oCvXizslnD z9oxsnhvp*(+vG)P5hW<&~##bf7DB=Li zEJ~6Km;Iv0S>^)~lA2FW$bbfh9{*esl5O^;&lTGw#squ|@tka(_(L3gFT3b;*#cbV zek=OF9o$WRg#sq3+@ygAQxsfqw-hYA<~j)N+#&B?uamYqu{C`xU4jU?r38Wka zFYio_97{_cI5&tC=%p=)VnA!GqW%(+r&em2e;;t`hdpx?`3drES7cXZLYzq$)^mBPTkDOwF<6lk$(rf)ggw2QH0TC5$xO_St9pWSly$NE9gdm(%ISCOY}?jw*;!emL1dB@T+$#!zr; zT+j6O9rCZT>ATrWPX2J#SU56$O~VnKBZLCz(&k+QI}Nw+*m4~^T#j@aI7`EANu~|3 z(Qw%Y?JpWf$tPVbM7R;^2VygGl ztd(AR=&{Jq%m6z5$Y3s4Q*dN3T`v80&dmb7k3Zfx3#JiNe!+j{LGI&bZ48Sf|JAh} zpX?zwjt|tTt*q?QH_^PBLh~z);Bp?*zat9lFp(A>2|uw?Uds`tA{|mJ&hL|;0PvgP zBrM@9?C?Ig;I8k1{&&`l!8`F&XrD!uGSHAfFYubHnjY;Clb-`0^mcIdZ~jNuc@viE@j@rl7o=|dJ97D7 z-T26Eo~Bt(CH-$H*VsP^V}%5Y3IFy&2Z*ArbkPf)E8WI$$*~mp4UhZZIwwTaJf)w5I^~AMmBRYv2xv3gdF{tF#5J);~5ixw4I)GzvV<5k53m|M81Ov z$}^Lzmb<9XIt9+-I%hPQ_{^^a8eT=TmFr>$^`M9fre&!ASC;aedvv5P*=`NSWuqCZ z=_4>hq81^NH@QdUrwP?x4-L;LAmoe%RbU!RbGd6cZyxY$z=on?9x!vY;#Ht4naIk}-g((;K5) zMCeHHE}eq&Re14p9C$s{a?f7`2o2>HtMs9rIj%EXx+KSoXo|bCnnO1n?%ti+poaM_ zjI`gX{-v1Z90>iA9Y|PVtBpCaa_3-F{smXR_x_y6iB(qNC)Wb7bu!Nwd}#(CU!(dU zT@&v5YM>aFk!HPnf7qxgyWfG3X+ld+v#pJ9htx^Nhg_zmkFR-ZR7D8J!3m%UqG{KY zR<7m~gW3FdWR&{OuC4<}gOG%TflLUgf`XI&@Ih}t&3`9JZm}s1nha_2c73D(p1TWt z;9}4YS-?6*zBM*Zz62sid^@RnOUP@L;Q0(NK#!2A2omSV#90t#Pj(Uvvr%pyWezml z6OEWic-c^>Y~AdwJ}Dm@sDLX_XAPa;BiI}$k5cpITIsYpA!U2>%ny*yKbb%IHUcyd zNzH~>i{EtFx)qZ{Q-1(92EHl0_lJTT0z=zQHYkU6#91`L=YL=De<-~PTSJsZXs7$# zX*u{B`5Y4W@9(hK-L%Dk1%<4+bj<`|x3=6dSV-HW7mMexT_&Q1HuWA60~v@Tu`>1! z+XTYeQ%C=pOJxBOpL2e_om+=Omh+@+ar|JbC=*VxHdPKCW~Kr7ValLOUWB?+N=Uy0 zR|yP@JX(&Dc9sa3-pOqED~B|CFeafvR3!JA*(kC1ObF*q>_FIx6AU|YdiD6!@U*wb zv2hIqm$J01p=jWWnA({?P!^8sy@xmCrPFPyZ@ftFA-3|Ev3$872V@|}s5okZfr~Ow zo~R-ctgB}7NnS`>~)NbWk5R zbmYPsL4e78hKirVqzEB_KIe}m7{Iy6EUdfd`6Mvrba7N4SA=Btl*K^ry5AMrd`mG& zT8j~o1Fti^zvxJSc*gD2$`d;bcYg@@T&^h=48QKbXx4R$vYGX~hLtnEn8Nrd{`^x{ zT4UlNG;~3lQbt!$OOwohh2juRsV`mEhK~j*KcCO5HHKD&)@E8oDAs$~lSfApK0%9= z*Te#}Uu&8?5zxG@76lYMPJDPnrw;bBem18VFQHDUPH@!;U7%iy;jaj@kBJZSw;$X+ z&5rF_AGf$c^af`XkNO41G>A`XqEzME5=FA!LB;2#Vqf^i$P>rkc__td1Lj4;;Su3j z$$qxZVYC$fUCfUuh3SERCyx`9d#M}AllA6a5fmOSObWMjen{KWFcWb+P7b|7WUD`F z@qz_pAvYL&dUEuO(;<~+xDOg`slOsy|H4ZVQP{}dxo}u1du%h#~s&? zNb3OnI^H`q^KxFz9<#Sy4&Estg%pBJWq+h~LY2DMZ(%PCpi*uI!&%a#4|D!P(|>=! zdX3f)PX-LUN2#?JlfUk_9wc7-!@MEiD_>i@CiW{gV)m}LO^nm>0PITVCHIJw-{Oq^ zEqf(P0mhF>3-Z!ai+*t|86FMxGxy+w7uz%iO4`&+WEId!{m`#-u8ZezO|C0fLJYA! zrH1h)n4#45oKCGj0#*(8O6EaS=9A8RV$v8D>u359t1JzRE2FSgYKxv14tttL^Prl< zs$a^Tae%gQf4lKOAhsCaG~RNNJLWXAzdhk>_8$je?}zW(K&&(haT*yg*;3`>?_-C~RZX=f z60H2~Cx-DOg?OfKUiLpKhhE`k^?$xthms(ZV_JYM19*Ed2sf)udF+nu(z$W^{dxMF z;n`WsF90u=#od?fq_AGEL12I_Yyn(OXQFEuy^e7lxnP9zs-%6vabOvc>WV+=nLzC< z%6bXH=cd2CHv(4FKpZeWfcmsZ7+|Q^Khhyh6&?Eb`ed<5*`Fm1bsu$vPjCmMKsorC zjlVtF>y4F>mwlqArm|QezRdZsZ5+5`iwiUT>e@iy`L3lxd?UGn87t%|JNaV z<&=XnI+L{xuttx{UfsHb@QH>;^bVu-6=HG-!B1NLp4?uLgV)|x4lbqlWqk)CK={~S$%PhJl z$w`7?H_!BoibZ;Yfh!5S{>JCw36 z^ZzEO5fz-a@1TF`Pbcd8+y5BhxK^gn^~ecO9o)gT26n!kRmw5nNj^~HEOD1uaPkdx z24tjwF$2lQ4*s~}mIJ56T2+raF=Fi_|5VNM;O6}9qSIv`EJGYh20&hX=uMW8`fGD* z^rwqN5SsziJ0qo49o))1*J9|;bq@s_z1!qN&Bg#|Au#^jjvC5eZ7QD#o*EBPeR2}N z?G3q0J9k~NaAH?9`KByw34eEYM4xBtWWrF~U1fh5xh^Tz6L<+roaLxcHxaZr?tO?A zyJ7Ya82whzk3ZVqvQp5jfGOhYZFZuK*+y+E@k#Y>D0v3DEM@#LH3oohe)f%A$%c4Xtv>_8z#4Rw?t^o zD28-+*4=&%rq1fz4SW}Y59t8rde_|xz}sC{axw$lDiJ?qkOyROmEGRSZ^^)O6NXqp zezQ`0o3tpsTq0cG26I(GdV@O$yHCYwfE+lk@x)u)oD*Q16n~Oudl|Z8mfbdHMUSkc5qWEV-cl9#qhdZIgWc6BaXF;j-qfBQ+G zxpToojP%L}uPX@fC`=8Eg0p(Yxc$Ae1e2wq#)0lw!qI_|sI0uSSdndOK!o^hUH|Pp zi-%WIUv&c*I#>JUdEsn(A-C=nn@FkO(Ze$2U&a_Q_uMzGQ^&K;q&!&92w9M&X}T1E zcV`N6MTpswL&K1q7tjCr&6n<)bQG)m6!i~-b|zRKQ4LjDb-aH&WEnd9Z`i&}$h3Kj z=Pj6V>l!3-s84vsCv61TAIs0(U7`3=Xah4_@dCb_A? zEez#`(ah_R1wQ9nZ<;ftzGK7eNK)0JkNWx#@odh~-uT z14DhEC6+WwO0-5qSwlO6<7<#RUws)5Pp0-P@H>fO7RMr1T=}aK2NWR|zToFn29(s`Y7g zMH7)9#1;bgAG7_;4nbTJ{kw$(c7&kK&a^K4x9o*Np0+0)kkTARA31eU#yvrdc%&aN zxTv;C2k7B&Ry`}YyS=CE&4Fb~uP;L}?ugfWv6};9ub=h7DaJp|0*1D%g>_}Kpw5;1vLjyqZZQyf}}K9+iT3<_VA4ornH6h)ov@(prg-833J^-NNmw;gFkP z6reFc{oj~&#BHK^IUJ#1&HW?L&U2hbGXQv7$Y`W>Vx!AJ+TA|-$g+F9QZD^d;5@x` zoh6aFw-Q@!M1bfFtZtvA3HPh1{CxnlJQI`}P^-}+aMJ^Cn>y%(Cf;abiSX`(e#!bB zIrj@5E;pAzP<(QOo{G+qn<+$o>oUYCVz#Ep^Li}hvtdKGuE9?0$)^;#7J^Han%y99 zqvjF3%<88y5_WhBe7X9qN;Uq8jtuxU>!Ua-%!w9|Lb{}ecjaDUVmhunRQR`IL&}0` z=iS(CRdBy4So^UZqU;sHl%k3to$%oOK0`~03pi#CO|@F2poibF?*{%4awj7cq#KQX$R0Y#QCOo1mEN}AKi@$M zA}WI;3whQkq-cT*?qx*I88C2_{|3URNtd7n>@^r}vh)K}Xu?{-7|1~_SHp8EGJ|>^ z5kqS%w}?$Iuw=3PViRdxc2_DWQq%>&KNjmo!~bSO893KjspB`5PbV}mAHL})UMM1WI^d|X)*Ujjq7OlwrL+G17ozJe(j@dt`Ek{+%>CH1?jNzIA9&Nf}G{T-CJ1D!79VG;$Vjnu$$3Zo-!+)tkFsy8wzZg{j)CR`PEVC&mwy zbdgM-j9v)!axxKqy{z~0U$O!LLi&B{ks9bW?Pq?G3^8fDVjN1dkj&$Bp?!PiZ#n%F z_VLLJ>)fZ=or)jkqSgX=E0>`4jm&0MVPn+)z|_x0A-Tu)VK#wJUmWnTkMqiV%7RHZ zm+PiC`)<~pOq1@7A)1&@LTPpGZ~l*q>yD@D|NkYSBuUwuNLJCcx!EGgh-@-0*CtzT zlD)DsZ}y(oNOsERl6&LYn~aQW+{<;(_kMnV{{A`dKhAxh_jtWt&+$0tzK4DQ`GBnQ zUL1^r@993tPa0!6c^pHWo2{ut@_>F8JetrcsHzqY<5^u)uz!S1A=o+ef=N-VNW9Dt zLjtyogK?Vj?kWR6w@gTQd|pEGtw7H7KxEQbaL)6hQ!8YRGpyw@V@Uqv#)H(6?|y;A z?94m?K`lv3eH@beW!9f71o+M6wTR7GTQP|N{GB}a#NiJwg7WUrsKPc#c~$-t2gbS- zomo}qHX0uw135DjYhz5tI!0kx8vc7-`rUv`hhCnbJCT;LoqW{&X{L&BkwoIylN$Qb z{o>VIyWK#m2j7p<4JkF@J)e!H+`ErY%YW~0EpYn$_x(&Q3^}VWqWAb}R>r#TOek~X zZrACd%;3N_WC{b$UFdr`4ufYWb4W@X z0Z~-^Isgjw&lASN11_r&fXQGmIbia%D-OocaF;BrNIckd-2SWH8915Bsz{1ATasNGd6{Ft}{antXC zg;+aL4lkaz?=dJ{#y-P(nI5Dp!Rb`Zg8lR6vNx+&z<&JS7gGWZbCHaL?RodUZtm^t z#{N}2Kq|{+?kBf0{qeh}WHfD&mk`vG2bQ$$Lp<>Hqn@k)}-d z?*pi-n)ycbCUBoZc?k?Y{5Z{D8vt5F{5!3TSgUV^7}nmY+XP-yC@X}*)PrzfDL$hG zh01}O`k5NsHhYn0rS9(?L*Vhe8Fm~N&g%b`r(qpGX%j|ZF^6H^>s2!_G-u(z5)t7i z8$u1AF6|5jbsqFN2ky@KVx`UI?68J-1^+7K&ze zacm|?_q|@)xD=!CZ2-PIeKL@j*ZmhO`jTH>H!nC`PBasEfenW#Ceyto$juLyDRPNf zwO8WVdK)DMmTMVLtx{m_Hu%pWHprO{`uf?eq>46*qPZXNn(7;pOYw z1>}L{m=;mi)$N98E3>XV0U3WHj5l~!N#NG(Dm}C0#9xJ2cDX;E*pLPF+HUOL?lSxybIZt%KYj{K97kOj@sw`3``fv%JdVWj)1I#pJX@Zb&e_G<;c-c!($Wp%GKvz?%8w=mN zRrCQCGP^`d_ad&w=jlr;t}8l$zcTyNzxh!D>;I@dhz8i)-8_K@>mGkHHR1{hft^ie zHievu${Q1p1&Z0Hz4Hze>6C6jiP*r<$aR-LdB9xR`H6hyk2*@+CeHNV$^)#UgX|*r z5D2 zVn}K?jH9*E?gL>HZpeNBhY_QK9%EqaOB}Zdn?)UmCOBkr!2Bg8XVzl+_LVLOa(3i$ zN|)|((jIt_*V6IhkmEKb^BmH2?H^iDvSAJr7 z(a80Goxo7qX0_P{croF)2X=cJ+uYC~n{;XXJ4p2la8G1pUTy>I8cDt#U}mC6`vB`X zqA&t$_}4|G00btnc2O&cn?Na)nFI`va^IZFs08GVd`Wgjw&)W$9=s)D7iWVNFFU<0 zdxL2hTB4;d_2!&)k(~mlh{jP6P*NMy>(7N#ShXjM5FD&kMnLr#*>KnEe6wjw&3JSUKnyQQ1cDc^rO;w2|U}*K` z2=+8##5ou5|0Q%hR^yLJ>ZWVa4xXkKKPumVRpQM|L@-hP9dTag-;0M0wfG4V5g?oS z7o1qQ%_d<5ehq4}5e_3(9w3Lvy?dPrPG9~kIq4HPN%#4KPd@+>h$rM1xc^Ve_F@B*IK}qPo82SOBS>hD02e#`z!?_M%EVQcvRC&PvooX;9fBxTl zZPCuSc^nkwsme;byF0b4GCYiq_h#S{R1+71_X>!)NV z#{ukA)W{4WOn}n|Z1~-LFxl?Qoj4@F!g1}q_TU1rLsNe-*I_HaR-TFxu&w@g6=I{dPc!CAoWEdEG$ZTAlz{2$@2*L_#iF|GbCD zmU{yB2XHfW@)BUhZ$ueu`VsKXqE-TYxa0r3ZVS}#p4W<8J%Ba*U4&DW61%}j5kU?n zqRChOC{uvQHg%Qw_eg|v#kBBAJ|kc=3(lJFBF!Sk89%@j{dqpV+c^1sMFB7tB%i;D z+uT*GxkkXwwwP1mAdA1Hd3wN{Ql-GJg`h}{*VV48XL-!Yw<7oi>Q$ks3w)1sO#8IN z9^KSrrTe@Pl=R-kBZ21L^{d>Hs(MjTF(^H~J9LG=i`3PW%6}KHpH*F8GH=*WJ(r%^ zKkwXM={gJ4HOK5$b^N)MM_L5mCfWd$(YJ|*KPF&lX$on_{3liyEIrG>acaEEW!qU) z@Y^-;qG@M1^~(((=okFdtONU4m-=4DQ5?4RykmV!kBE(aA-D>{)|-~@p4U>#1H(g^ zNu?wGrk87>w}60&{97iG1o-0Afb383&Z6Yb7GcMldRkx*_zR|&l&H0p*fD#I+YC=l z?OzLcNhwh~+^s>FwATC#Oxo%*0QgR$3_cBh+cYz$q(s!W>?yLh!h>W}zhz{-d?z|R3k*k5k zuFp<^;~A>~tV6*$^CLdYc`f-P{Q>Oe8Os5z+d%MJ{xZ@Sdwf>u4vM0jtKZF$v!4N7 zD;aCj99WM4jQbzKy3IX#lTw+J{r5D@f}{~2Vm-S|w!BEf5nJm=Lsv!`Nd^ZdJ%#23 zrucZn{rXuujlCrj9zJ~7KDvyzvPRC9=(md%Z|y3<^dO_+2-|RK?}{P z+m|oKv44u9X`%`@_jEqG4gdiM)$$IOEnnzHf{!zOq=CQ_{vOF8AiykvW2`H|HR(|v zwswNGIA)WuGjnavVj7W@*P?BoYkLgTkIJ^%akvJ7+K914uv+3-I^~Qi0C#QpyuEnM zfg25bP46NRbAINxwrjI-63wpPH;IMK-4NZ^muw7NX4l&8|+F}9P|GGrxU%moD+%wwD%ibREcg!BDI8=!jB{-W zb~_Sw{$lh9n7}hiaE=molD?Dyv(+2Q#x(M`(LdswQPB-3OYvK8RKw%F(cN(@OI}*8 z)bXOwrng0si(<7frS`mh6`9k5ZglDrbA?)L%UwEa;#Y6|Bk%KHc5uKls}ugwoS<|>SIi+Re1~B z@hUQf6x4%;Vd9o0#x&FW-u4|(yYd-O zB%sv)t?U{asLB}Y@D&6`{U(;!?W4#L1)J7$0#?O3XML%3=pWjl`>gCWBBL)-^!Z(pq|OU8OM&Ttm8Av^DD*ntC&lTh=5q%+<;Z`D zR4;0;t28|`G@rfNd}7EKJQ30govR7NRBia9zlClv=9_*h>T9^IL4#C#%LCe$(6oj9{;OT+(hj+FJNNf)}#4$l+YIv9?`+k4c}Icz#X)J-_8QI_gLn@vZ&xG`2v zVfLdt`s5+vg6!q>;fCd{%(Z8~-t@F-f!JNQKlF=}DneZ*TcydvR#_sRG2z7vSHDU^ zK8w6(&9~H7sc^9p{yMI#Tt#kaD@sH%+ zJg@kB%wk!DK)qxZ;mn>7D~xChIQ2jsc!_OM}%(2PfL;>9kcJe%8CEp_G}DncyvX|d#QoKTKFYiE*ZQCF{j$Li6(e#Kus z{C&*oNwOQsMXqp4l^jjpCD7;QoNqiDelu!8+x7&uk$rvMMaX0+Rjk%Qn?pTf~pTLZV=Q#SG{RIO8V~Akhizd%8H&fIpHwtVb*g`QEk@%l}ZR5 zz$T$xOO-Z*x**|54_~m-)YN0cNsSMzWw(qf)Si;Hcl^on@7RzK-WBTA3rn4udaCr{ zFmQfB!hh=LjzVKT<9u9>zS-mF28ae`kjwluBh%I>bFE(|ivL+=g%6O!QYPR}>_#zMYUQGQE=KUsZ-S=h?MxKgb&75H3LB&}Qv# zRLea(^0jVwuh7;CYfSesK3IMX+LbLjQAI{gHCMpb$810zoV?7wZzTX zfl~Aw-bjev1nLf`K5|K)kyA|Pi7WQae_wGn-1gtSHu_m_lt%$u@K10twCULgBkxsO z$B%xIuN!A*mf_fHUm57p+`6CA81A)g_J4iW1`{YV6@lD zXXy8>g%<%TMOmG=^v?W7#x@dNlPBXNCiH_-y&hRXB<(-}Z${*iIKo30& zP>N%;tkbsoY}>?omD6hcYZp1M!pz;yfwH?k$-A5jjjuwvG&E@RJCN(JVw?uC(tI^s zPRy90;uQAw^U!w3#K=m>g`WWn)8x1LMKeEnUVy$ybF8fA$2RqW?!9t`uJ?OtW~Z>< zkO$ow6~8oc(>OEAMU$j3qJ)zCJANAc%`;vo7vQWdgNkjI`kg(mrUuP}J6pPDR{hX1nl3iwM$>@O&^gOTpb#B5 zF?}6B)a2}N(u``sePmuCrBpFf`IGz6%*^n@1D`Z`=3u1+^vzc(nh}PvkEX8anEQKN zy#VXKEutn0wz#Rgish7r!gfFN>gw32hTH^*V)Fu9g)}&-H#V?g96-zcW=!@uE1N5DMw~5}h5@7P$ zx9oU&eI3?&p~zPvhtMHKt&h{ zaaXpU{fjgm9ZlnL!LRND!COiC$wuDq+rM;|h1vwICTUfsjTn%{J2shp4`urmHx)#T z!)QT4XL^2&Eb>Z~G8m4_O3Y4vnM1sp+B+Z5Ck04ndeGDm5U1yr{!leXrN$ns(MFYspi zHHBpL2G32F?@(!b6xhLK9@l~HP1_t@{B5`YAlS8nB33Q<4DcIrI)XooJJc0? zo%|Vkh~ct+#Q0P{TaKO3TNe5Y6xcTYYotpN)AgW@#6pM2ZN`s!*930{r9z*u3>*xd zab|o^HS(_4tag_%YNP49$t7WEPzr=yIYjDIuMbS%ja_BD^6UYKwG)TT^F7U`I}dw! zH3})dUuy|SQX+U?bgeFrbPe4o%14#_8rE5QvnQHv8iuVs?X7qtvuV}v)fPlan2qH- zg%Po}=p8O}qxpLoPfgIgqSosWOd*9`k8s8HgLuXe(`v#&AV+!0 zDk^sW?j@&b^{$ynSlTveKcam9P8OG(KXWw}?ttgABSe1~LB)Krb z%Xta^j2@Eo>4T~g5EPEunkp54f9xzfFaMy@iZ?6XfP$vPn3@cgsYT)$7x+C^$tq=< zX7(C>hgIRH`2IrzmK)Nl6;7*+AB}`^4S$swQnxN>-05~~&oeX)Yp61c*d63Wzf0#S zn7aT=7EAh2-q__PW*VtkC#s8F;#btFe6LG8w%2`SXV_<{ZP61wB-3)e^Q)|QFpYQJ zIO{%XVIT0rNi0^v&}SE5@C~RbUCfj>_jkTSqyYy?;u79Q%wAHneslqDLY*rgB1OoK zdeYYnrKkP@6zsFIi-&%JmW|rZ3{a zr{zf!F-jH|Wd(+@`|?brO-OVPT;@d|he@(z72!%n>sCMFjq}P%g0AzXy#*jGqW+>x zpr!g`X)$}C^v7r&uoDV5@9qmJ4pwW_wTRB2NHqSOezS^4SHbJ`xWBF2d@YQBY#!Bw!+1r*WWL9SW;* zxy5;i1gTv=EknEQ#H33A=h`p-K0Sp=GK+PZpM$MH;fY3VQw$mGKk-w)n;XKI#0c9q z7^OjeOaFs^`y;Y(rh%maWWODoE)HTozNseft3;_EBHzTbXktu&yGCurqjh)>NJ4J+q)s62iWgbOJcO#Q*P&`*B?nOQEOm+;YRoSyk{Ni3Yi z_wgj?$z~#-<_hYP+qCi`-$X<0QRBiyWc0P~1%1^0JkMJ4no?Qd^>t$HXmiQRG*z+F zap+l}-0E^E+ASM%GbOj|s^9^pTM$`R%rih`3u^JKp-ez)ee++geI4#{4jyDiiwg^_ z@aU!VLG2Ghn`(b71H$Qyq>2Ji{bc!VfNEuHHQ9h;Q*yoX#MrWhGf&FJei@{IzuHLN z>#5~8#a)%2vSJ0xNL>t?0vHs>w+kM@0)aRDtUy0Mx{;`| z?O+k)qSw_yv`DFToZ51QjrYW-?Ea}lYgt{5Z8gwwhf-t{;QgmMqbmTRDmql4aEF{E z_8Lmk8+UGj2%K?CJFm(|J*=I+lnt6)|1InQK#0F^;Dfj*og32tTmAH=qF*|;mSD0M zk;XEmC-%?C{SDnO?E$*C=RZ#J9&KMZcjN&3DI!Hi-oNfR?);d^Oq~hNQGg0QD?glu ztyV-q!yBS{JL`9EYzf@Y_P4vW5cTUS8p`S79Urs zRIz0E+6O0%@=*!sAH;7}ns?@JZ5SgtU!sziIBAC-L!HpxmLvvxeSNQ8tNc(}Np1~j z_fGuEKGtO^D4t6#tItZ!;db>G_~gSp$J_%P1U25l+O7#ft&Q2=&@>`R9(k8v;MMiV zD+K3BYJ|r!L0i#w9qgBsqT=xbsugcQJ}CSZg`-hYWdR*jvB|_`&7JU$>8JAf7GGsYXIRS_ZhF2tCmp`eWZSNldDhr>Jq1v zBt1*V*C$_JT`P11nYPRvLDxKYCTgxPI3vqmh^)7zZ`mO)b~k{A+8# z8c);nq+~%jPM`|q5N<+xS^(`D3n<_QX+!U6Ue$8l06%opz#~seujRm8;g&+R`q}qM z)`hLw$PlVRz|tmg+%m1cml0p6cg-thmfo@HDF2G;I?#?84-c{|LjTjo;E~?aK!wEk z!?gQA>gUUXOE@TH`2Y^Cv0Kj9)j_SzNs0zn8+mgM%0ipup8c^wotAa^1gn>O!e{D? zEt^pkc{dxg%vRNDL&sT~=9NW7eu0O#O>sVH^)nW-zr*7>pFo#BQH~LIiavCOnHUKv zpg5ak3!4Gke4osAlf8d#93%~^$$g_(l9MrfUx|@H5D?I@jwO{WNv3%eq~9K3RwW{R z*0Xv_32+a;$K8$s+ZU47Qy-AsmxhC`rmXlNMsz^GCfEXeL>33#^JE- zU4W2Tqg$e%oqQ1QVbin>#IgIJ!oquU7rpoz6R8@Z_Os#vY z3ngvWG^=aV5Y3{gb-u(sJd>OYOt!^$*k%qN!0x6X&cRaDpF~YFu(fDRVY+BtR(5SX zvb4Ian3UErc9$!rWBwT+YdKtJM%-%L)~Id{TGIVm*#(jI?*04L?sMI z+>Z&m{$_ly?T4am)7ye(YWy88V6$?B1JwFNq;1^@p)yqr%jVH1 zQYY~DwBHHD2*H5^Qdxr5$9|XX`Oyzy4H7F1=bIP5&1!w;_ZeuDWFsbCE`)L3v%FjY z_0jsN+4KTVo8m@7B>?Z0iRc*)o^NV}6fgUGqqBsKZc_mAR@-AXH}ALy^6?J>Nryry zXSO&b!c3jVhh;Y&D{BeLe>Bcm6Vhd&T`7a7JA#E+{8i&Ug8k89R4vlGws)OQ4@yui z$X<;iZJnN<|4weoYvZr{k+K1tXUK!j4eloZ8}fZ?B-g)_vQp6Bb>q1wRB1(PRElM!Utb;x!i zk(@vKiKy#&X?F}v<7UxB*p4a`2PU%r(28NxN&*=-umXnv5%#*MuP-%^ zoZBADh*s7!?k_G0Axf`YPbumMDZTF(M*Y0pea>~mii6YT6c%Hgiio^~yHgBG^%0R8 z>sgV1fw|F3ci{Ntu=3H~Z|Eii?Fa+W zZ8Hy2)|-t0u%C9evuh&$0zcPhp>A~{Z=W}sI_*Yu-5|?>byx@hM&6l$?*svjugBZL z+VLtz-nQb-J9RpnR&GVj0~{ceE0wL$QXWbg(A_?cQ1v0eDF;^BI(7edyDq`tny=|& zXq!f7^@b&S<+5SJio5(Ua>!CamB6d^#eDxI>UTbw$>;-Sv>UO?=!fYjepD9UXnMmL zlmTTJf}^$GRJPvQUo(Y*NZd(s6W%X;E9Y_Inmj=1OSt++a}4m88{M|ubvvWEyKK*F zC?eMM`)H0dNlEf|3MZ3LaI)n_PT8JiK=TZ{6T?tBuq_ZdE_w<(AK)$`zzteV9~%iF z(d84`9GZA!wmth1tQqF;r&SsdfZ_V#b1>gd9e{B87Ocp407s4va5SkBCY|EuES({qM4uOb^3ikU zy#M`jEYBm`tzj9Mn|Aq_qcld)))G5{Q+5XEo|f$yZg*Lh^Dspld4q!bAwQx`W!)G^ zr4K7U)hFuGly@8r0y1var@spMu_J(t&Eu0_)xH_GQZ@rVp4WES=cZ$I00_LQ7e95_ zv1u||d`>QxP8t`-)2gAf;PbbvSe&*Fm^5doAT5ouVSIQOkmr9^MtcGF8%|OoB8Pvi z4cMT%lO^>#5P$?|bbrVXDE*WpZ8ivV6GOLU!a!f@2R`}85H|Pi*hy~SJr`-oSiE$P zrYr!RzH6d4%NU(?%+!;zy}17{%y6tLu-es-i$G5@>!Wf*B#{~{P@1GJ@_HE$OJ9`FN8=q zg8jQy3agc*mn`J(g6$>~;0_U$N<`h+=ST4}=d}@{jwrj;40->q9V9=rA0Vv&b-HY}UICtB+sViO6p9A_IV!%C#kYX{Drp*E-U*Fl>v`*Ig~zAi-9ek0EeUTt%Y) z{TnmOgq{AWEV4E5KHz&Zhlu29U2p>Vd3}AN?~Owm*t7ov`|)!PL;c)j>|O!H$rquj zjahOB^Y5k`vvyhxYy3<{bD5o{S@X*V6&kZHPP9#;2yh+mVcsw!Z}2Ge21))bofWTj z+yQ=GdbLp}5a;dmNO5F2lqPwFJ|s2CCF6@KbCgImF97+hu0!O5+&M&|VGBDSdto8{ zRVtOOr<}vG6+ob+HSHl%V1Dfu4-x;ZBZtJFrBZ{;s)Q8WS?`hjPjHwobGW%E0UoR- zQk(=JPF6`crp}*_D@Cv=W2!cIR!IfqPRDgjafmeQ3x1%B6SxPGvHYV)B$???PoWpC zww?M1Wua!*Nvt^;;w#Y)eR)EXr!RogVmZdjG^p3_J3yS=$pl8h{=HYNISYh>1g0)-HDy)WHSQT#ZPdIOwW!Z(dkgaW1^pL;p1WV! zDrDr%h{(Fw26+A>cNHRHg>!qcP1VZNw{2zD_ZMa_1~Gdd(jqkGW}Gu$Bi&~4|84;A zoO$Jump_$Mj^D2Du4KH<<`$oDi(BfO7&4N4=_&S7AERHAp9(i@ZnWa^5l#~^Tev&$ zMMRZWG!o;KFG-u_*VGVxd6l`KVmJOnZctPlmGSuHniBk27TUa9J! zY?p>}F?G-KKx@1F$e}UfNMJXoswxYQaMTt^cC;P=yZa{lLv#tE3wBYuzHAGDZI*gH z3YWb|Y_nfCT!HKRQ01^gq^6PdTCXs(;$-FI$_a=Y<6=Kuq+e5n_(+zEO|{8p=ZKfS zi^5#dy?3bVC&hAj=l7lEknYDkcV3_Nb&C~;+eo&9S9(Q7Q`=L97To0zYJ(Qvot6$G z_~YL~V6tK&zNeK##ZBQ`+L<&-nk&U8!A`!Xp_v&DzFA`-O`gO3Uo=MbPah%Z(<8Rt z3$|t6m8GG3`84-EW3X-t246s#&`?~#NFNP@o=DaSY`O3N4m5P>EQJq3BYjU}yZN65 zCvyjKn)p`=vq4fyvwmpU???Rg>W~{^FzDz{6F@9GW8Hm~k9m^%w&yh~Y9^=O-bj zzKpN2A?9v8%F7{N!A^3d;m$AK$lOa1S@GhH6*y(bbbAC2q9z|~*i3kesG>bJdK4r? ze=0pfVe(lZ!S{6iX8>Pm!vjW#UQda0Dw*to!n=TBg4!eZGxeX~Ad%yx=t5kn{k3V0 zAMZ|)#?7DE48n6SZ`7;azBREU;8^UH~>(K7KI6lY7k%)iVrIrIO-_UkzirNmnfq~a# zb1#|5p}(CHxC5CP3f4vJ&e`m1*_4C4_IA6RloelNl2@eL_IwX?Z4%*&UkZ1x57Bj( zA)fP$KLigk$i-r=%xSpv-dR@q=!%P$8F`6M*&K5D0#&U4j9@Jw`~)}gB=zX-=4H5v za|aIm3={rwk7S8>^{!y(9tB-j?ypC5IX5vXOau5sAABDPKwpN~R=6a&?}{l_inG2= z>T%!wyu^75s}Z3OXFclOJq{g-NNNas(H(@!Czowgv+iEY8FOQl&Q49EC#^1jG0W^l z?)+^pN|`Y?H$E(b_m|wT4;fL-1n(DI?`CyLzB}ldy?fVQ?3xK;nUixG zkP(pE()A58#IQ!IwD%ceylzelmp#wST^xoL;~;*!s;j#W%AY1v*2FfXvtMwLzk;Mt zeaJ{(b)lC38Ka#$U-W2LH_Gz1z1T`J3gfCx$5-Fzk>k3)JDi4T{WU(E{SaZ(K^Z%+ zO9)V^!3nS-yfrL@Vlf`uWmP0n2S?hoi_0N%GzWMQh&a#rcVB4IIy`db1N(Y=!2QE( zNf?vjYHgKIXO#%EC-#!NxQc&135dN+LAHXb`k)tv0xl)$}=-prAR zgu;k#2O$vI*?$>611{2p>?#$9ULu^n;R_A8T;ugC469z$ShM87F5C3=Z|R?l8c$(M zU7rwU3StE$NCIo|`+do+Tjb@}b5J2qcD%jy&}j{d~a1k%>1g9)F%g zObi#m_U@{t_PRu=itDoVW5}uQD3&9h3*yuvGHvk+KF=L`JESVejM6$jCQVz{cmP2Y zgCAkSmBU%gUZ+ntk#_*x8sv!d#Fl4!&gHun2IS?9owJC`+2zx*`>S($53(Q3&?2UO z-dT=zdyys)H8RG&KTM+4$zVe!NK2>LyY(Dw?YiwQZ`&lVf`M7PQU8LoWoDsE^vEZ% zu8}cpuIu&Ur#WvP=^r3tEuOTlbkvZ^wso9|d&y=Gj5auam794i*J8U?PgO)O7PuKy>=xZG3Kh_DgU>mzSZy3u5{6nMA7)?|I};m zCHC4&x1IdP{`w1C>4=@E|Ih6k?7e|krJAm;9)sQjs;BF4a8n`bv@+ATU}(Oy?M z<;x7Od;L_qImYzoO{?5_w{Vdw6Q0c-y|FVt!J{Pu+J+F|V#r|`zKA|wmxs*qk&wmB zQu2Fxhb@L_htK&N@->F~LWq_PGRt^9BNUQ}ocngL(A%zyCVTj&v0` zgTZ}}FEmLX?VLcQRy*NZ>^kHH0GVnLyP?Z~5e!Gv8dZ1*Y~L$Lt*{CsjBY zIP{KwgiX)>S8FR&cV1%(u(k3}30j(+DWWMcd*<`h z3EqKscU6mq(o2zFVnqZH*hB}>Tez^1vCIHcUO0Ew=ZuV9JNkJ0QpsV{%OrAWK@6Vk zQ|1zouS%9f)P;m97&ga&A$-&_nYsNJqznkq`IFzU-aEE#V}KD?M1;Q&MPz5fa&hyy zc{BhMCjSWL_Ic%4y%z{r6~gxgy6_SO|K2C4KvL2VJ#sF)9)_VX?<7r}vc|Vv#EU+z zOV1j$FNRBH)d9S6W&OUVFxmBOm!mXUuMKhFZdFtInA^vutlkoYSt;!rOCiDxI{^~$ zAg1YHtpX|@N>2N54hlLzBlY$R4|X)Czk7& zCrek-L>TxRPFxOj?cVm4+e$LXpBvl~pXTIS%ps#;v_34p({l;eS0@DEk3*Qbi%j?sK@8Kw z8o|49R~mHS3EFf7+hxZ9JZzBB;T$|jdu_rp?(roYFt<=NhRnjzkwbdTZ=_AsQ{u_p zm-@lw?ifR`<~nSF|o2LFu1tHW9=BVV5m; zzVptrQ~BKak^$NjK%%9%lks(PT+-g4BC5f_!`CD1)3K!AJ5IF0NPUc#PYn(d_5I!P z#p?_3$6-u8aXG$?Cnbwfy*qDyZZD|j0K_F@a}x9u{Lw6HV-}-xpnHqp;*OgdWs26A z8k9_|fPVtRf@ifltchv$31e>B9Z@8Lm_kzVkeaIR>&u&jog`iK%BbEc+0L7^T*VMB z@dM@ZdsM`FasS#(e8vjr6z~-=K<&4k>9kaiT#K!|wtoP*2Mizw`qk~V=P4`Sa1jIS z0TofZU>!)yFD3h7CWVm(ENsa$wPQIs-c+Oz;i}OiIr-P6&EWBG57}Au<`0Qu{H}S% z=3CNjCR;;0FLEzgsr>B{H4k<9lEn&kX*OMEpIc~xFl9oomhWO~sd30O%Ls*xOL)E# z1bFWX;(mUpOCFq&1)F!wxnv}Cs(akSgBOwa*oq08={HOW@cY{}ML5W}d+ixtV8x<( zMBS=X^LLVehgOJxaG4TVN{s5bOY`7x13cKaMxUG|G%;Z#QZ;TV$v{nL^K4ieYOV)g>`n}ZD( z$kIAAHs^MJ2(`Sz9FgjCwr@vzmIVUyX?{S&FXhkWTnhLvOV6gi{2&2ri;PDR;2ByG zIXdagui0~TUc!D$P2s1qGhRxH6X3+m;pd-cQb0h3CHI)0MtZ|p%?*ym4Z=?APKWeQ zyz$?*Rk3%HnKm!~er{kCEd^}z13S}1 zdXzoP)#=PwHN61Kz3Am#b0H%+6o>q#9EG3YUcX_yDrD9)7ecn}(oooWy&t1$IX?Im z@qD6^4WM>6Q(2$2ynC799O#x?h0BoQ|DuS=bJM5=@O1mvuW|^`3ndiTPHmV z0@G7i%2*P9Cc3GOk%nQn9HAR?ON?OD~Z047$e3;?j{MC*`a|UXbZn zmcJEwe{}xIOD2UKyW_&+{_v)6y`)snU4l1Brx#Ask?@lVcT*8~^p3y4?r_F|xXypY z*uFN^5*Dz;C+C*7Sd81uZa;#BwD!JjCOD>fqjDkbt)Oe7|1r3_jBJ|-=jND8o)jb{2{yb~K0$anyt~>r zt|O{6y9D3gWPC?F38H5tlmL^La;ulwM9h-kW1-^D_=LbYXB;@^g=2#rjdcyBg$$6j z_+-ki|1$#Cc$mxv)c&3-(dodb3O;R5AF?dC=Oam%dRHA26irM#f+bw8YX$M$Ra`s> zo2?e8)y`k};pv-Vrt5L<0u0+1rj`2%dkI=OxLED}#E~X@&vU~uGF#J!m_NS36iEaHNBIRlMUjZk2em$lr7|paB zc?Y&xEVTV^cPjq$W`?JvleVW;_{$uaiDRlegg!t?j~}zOcE$Nbd!08PaYNAeWrHmr znf@<10=$3QehPMZE^+}bm7k{1f@OL{319$b>$u^eM*qswj?gx^Pd>6{l+&^sXnlsc z$mH5G08>D$zkC!VU|eU7Lz?|LYqwg8Sb?A2nxi6#9;yjUbM_VP)(+q5Ld5J{+SALI zMC3Q~l=wfHecYQ??4_6PIhQJzoB`tNK>Z2%xYf6S)4@VAc{)kFQ)JL+F!!ga> ze*c=z`rCqrmb22Fud?T2)Sfoy;8DwP9TGOd=b@qJwFcBd-^CBVdY;x!^vC(-;HRw9 z4I;N(^zSmI{(>ofmm|TSf$5A44(TPThcPqQb`KD{ai0d*Y-zJ7Q-~wLqd8tk4yx-| z)-ds74#iZ`{}7QXUQQh(cs)6aW`@k_P3^wtN*rs3rJbxVVgL)1D8gpq_XpNj@JQaX zq38ejSbdTUEb!o2J4!v^^~&-1?m0;@?CF}2BxFd2&z8#ta^dL9mh|i+p!LfUjOgvN zPLUVP&l-M!Omq);6(oF!?C=tk9cvC7;)%Y5JH5{S@w|EALW|$!6qYB$Z5DM2pBW5c zyqS%kitMVnHP*$S87>V}$i1;nl+B*wtJAmuU#vgnAqM<=r$hY{d;*QsZYK6r6%MM0 z4f950YyYB)C{_v1b3K{(ZmKv#(0|Xz=N!!sB#O#21uImM}3Lr1jcBuj}lWjv&r?^~-g;8TSb{P@?^Dkq~5rFWO zCy@xVQQBSI^ZVjsZj?tcuo>#+e8eDqKTQcS30o`c6*Rg-Sg)%fX1#Cuy6C=>M^D}n zMt}$En=lb1+(#f@6r=4a1UOjXKX=w1RY2THbYI)>JGNRaVv@Lom!Yx!)gNRpvj5&r ziRUNsfZilOm)!Phx$?7S+vP&}#(wyqdhBRYuU+9u?a&@bXniq!b^FD_IPlkSKt2v$F4v$WdZ`YQ`jPY3HLNA;VQV=6}&7?*i=o~{o)R3 z`QGQl9n^5re*0askFJZgODidqGfr2dUN%hF&@TMOPlezuWPusDM;=kAcRUEO3V-fC zjpQ5B;lWXs00tE0?@nR0nQY3tz)0Ccv|R3G7qjp$!A&b_A@UQev+Sk3@-=Wvu9)=p z>Vm53N#3S( zJm;kv`GK33kX??>e@7B7yU2+m+V=@Wq?oaetI1H&7<;Q09?79PiCG)e6l1%bqB(R_ zgUqtbU*Fpz>}-D}URq4%_~F2#@x=FLzq;Kp>3LfXX=T8W579@*KpNNY&zl0wg?x(+ zkq!KSgM`cLw%&HV))nleDLH$0lrS6?t9SHSi|eEJr=8q^Y!@Lb42%AuEatg zj5w}Kk_jz&-vqOt**|$aHS&5D&P1`3>H(;R{4~N&?{ovCUUa!={L4Al@Q$_50`yWy zHndq8Fh45ScMA#a08yopNe`o~RB_v14Ez*+faEHU=Z$>6Rt9xZrqk0*&FWPa+YjPx z%8z&of%|)$

pzx3{OByqE@6Q?<))e~Yj?)9i zC@$^{57Y03Jx||L-E!8R+kZ@if2Z&}$r5Kov}?d@YMn1oQZoWm4~)Gy-&NRC`0 zjPB=Y<-{y&``@`*8~dHH{bvsGSc7U$-_rJ zU)!3P_mP9?*9;b@Z*{=n+b_e%{WW1E#_jFUDUR7+J;<){E7p|OJ;O&xF0Bqu<(Y_R z*wW+JVdF~My3ycJ#(iivbgV*->(`uIKfv~y$05+?>m2~I8rp2F2U{9X&BV+-KSdV! zC<}+B*pQ(992xJ)Af#H1mCwTS;QzGa^L0HhS=3yp-W)<6tTg_mxu@t)v@%c|Fzqx% zm`U}@H1}DMK&oaQMjtkuTr@||eZhUY!Bq_fD*Ph2qg?ttw|+D+W87p03PV)p+QFI? zlawBZ;Ts?yi!ZL;(|1V`xP!n`kU~pF@W3gsU7XiIwxP|o$j%0WPm~2UhSMrL1G7Hv zrJ9_jZv(I7P*|2@==A8E0V!h6-1tsGGjRCL*$ERyk-(!`8=w=S-t=$c)=2~HwHk6{ zEWZv+6R3Tx^+prphKQZTu#ES34{QB`c0~kGPu4*9zTB_`FbLils6sxJX{pGBj6~$p zkN0Hxhq+K(@q|3J?mFY8(Ghv=R}YI~V?S+JXh$`G^un#MUrGJ^ZQNg&bZ+4dqqt^s z1K^mQ+X?>f6##y8Z@n|9X3*a zm)m&O(f5ht^6oZQ&==AIzVt(N^Fpv-iZForG3k2B0kn+L0dwgmw=#jlXjwyI^D3K@ z>5&^~5HmJ+3@?%kIhF+4y4QsIplQ-GFltF>X~Dql;EkOB+~L92QgJhm&m|H7=bJI~=X$L%&HL>UQtw!&Cjf?B}SIjuPcN;jyAn~&C#AulT zGUPy+UyFamNH}l4=Hy~wP?B(7TEuv)5QplE@AnOB(m0kfv!CMYeGLbv^>=U1*QD3e z%67ECs)@`={`ErCox>|i_=cn1FL34$0j5=qbw`UX3wc-4b--U{&DbT2Sv6lzv!=nc z6fxS9zV|0-owveqo|3QuZuQW=HV=%p@NdoA@0xMzT$;B9&m?K^hP1XQ!3=Ud+OG;9 z=}n`4ORjJ6)MF92VkH;p7_r(ZlVZ8QJ`)0W&kG7|!RUm_d)}ahn?_-c6|(3yF^UL^ zU9&Cp{DBK`c|LU2-B4Wx_8xb68%}r=lDLq|tCf>sL3gUaQr6e=kj)#bI1*}!3Cgg3 zdN?`vRnLF7Vv}R(X8^;wvnjZ*@tqymq;HV>m%CRv=-*LrHFxY59pT?&C-qm05)QYF z#RirXkbicY-@hylIMI4^g@vsPoFozmXH?#p3}0`z5w+2NQUD^JMcciR%ZsK7dry7km(GxD`RF_yH|*UGpg`4fC{+DVOT!?7FhS|5(3 zSx4-jNK_jeDVkM54g0#m#VN%)))P!Z(y^{~{jaHydCBy)YS(bzAT04k?_9_>VW^O{P_Bb7oZs6Mhpg`VL_$ ze?A4Q-$Tyb^tU(5`D*x2m1Cfb6Gc6|+s)0U+y@8&wB~B+%To>TX?3>mX?${a^i$R3 zU2$r?ml!4MR;TB)ZdYGFDNcFBw|jjM%-kwUrJH5F2lvu?9U{B!A;;25$mi+$3CG8k(e(^5MyY7!~W& z#-Zpbm*^^en>8VI$iJL>6INku7M-&mhUzt8#D&%d`$xrrgA##;>Q>dTQ%_d2 zs-RWMs4y;?CvP_2_%F`(+5D7l-#lQ*vJOymm2P{J#yvA6ic?9%T{f$ldN3YmHHqup zJZjFiPNz%WeL8CT)@t$xYmM~2P?o%|N;_FQQ=WOKq#QT$*{ia@JH3LjM!Hwy_gx{! zrtkUVW)qPo4tZ=<+#;~$fXjcW2XQ2kttZ9HHaCo?G<5V(K7HIv5v}iT=on7cJYx}5 z(E_&$9viLWvz}36Ao+XlR^$VLEcrh#QMu(sEP^5;wv+MKMWURXD(o1GDyO|CBK@Eg zYOIPKkiA8D@LEOMv_=8VSQC@Lh7(Lfr<8kQoGQ!O;#)TlPCv@nDy&kLk8K%jR1y-u z1~-&$XAnnxxp`_*+W*6}>OlDccob>+_G5WbrUDfLd(6?*PCXVe1)kCGYO*f=#H(e> zrlw#y(PdUSexv;^dPzAxiEr`;`+>5uRC`;Ar>wXOhso>uSE%xb7Ye?WxT-Aa$OI&v zVm_l)9q!m=RWK3?lKUx!%&)(n-@A@GCGW@#^^`I=0-N*hjg3bP2yIz_@&exNgj`MXw6ulY*8i z8YKR{k7hWng`#rOVU=gProC71BZ<>%kN>Yzzj$Jb5m;~MB+hwrB8D%IcXo*OG3XP}4bM|5N>vG9Y|k ziF>@IIHxxc;A|WptGW=3{l0>8_pJfC`}r5>fE0!F3+JH2hknaJ$EfJQ7m$XPaJ}MB zFix!t_6OdgurpsRbt21>^Eis?$<-A0C}mwfoId&SjrPpHk-`gV=Iw7@1CI^p2)wE; zZ{vW+6?QKn+djZxswQ3^rSbdsH*Qz^>tv@xTk4x{pv%71S$B`2<7r&2{3T#vX0F+7Oiq4IM&35Z4o}yiXnBp=eG+%@;S>H2NR(3A=p5L*?I~x2n%VOu(W~ zc0>>z2YYuXagQ|jPwP_9MOSHlwFY7XAAQi3vBK0R)A}=}Ls>WbaYm0XSfq@luE4Q0{imp<7TGe#EWx6FDhCf z!MVVq_@A^w3?kT}K7V|31}q;9cPg>odjsJ+ z>=}kcH1uhGdC^r!*DXqvf0h~u{l^lxD)Qp{fRd|K*xsx8*bY;z$UoNg7FxmzEXQwrucHeHRu+w(s^R+O*n+Wx#` zq<(At_bS|N^bc3njP-Ly^@hl!H^MIc#sn@FobNxDHCs*KOzlz)tr%=1dxwAhFs!<+w%E(k3gVMs0_bp(?{`9|jy-n?YA+U1d|oW|gKB4SHHL`vX!)aDjtV zQe=&Wdn|Nh$z-Cjvy`eBI?Py}K2kV*&+x$ukLXK=6+O^ML{CE*=u4OjQiS3j6HJO?VUGjCFPc^Pa*M()=V05f4S>fx^YUamxS87oyS=@? zR!x8_>-iF|VIG3f21`s@uhbNSB8BNM$DVKe7D=qX^^4brbzS4s(yIFzT|gYwapOeJE+g=yKVoJ>u`QEHSuF(p}B8eTb{wa8erM*0u`2gPqh2*Sn^5 zU15k_4g8iQ#CUXZm{NW9^$brw0f$Jj+h$e!UUE{nmX7UG-CHTxmrGZ6Z_)uA(GBHJ zpGpd#pQ9Ae`iY8VH19WAZHd84R$P=oNB3ki`lR4S%2Hm}SC<#@oJtMsmR2w_dW`f| zRU{X>pMie(xo7}dofh2o=(hW(@>t+hO3PKLQ=FPZpy8T2BmLVnQIYlLc;-{y_FrCQ z58|}S`ATV7u8G53SyYfNU~QL2>}FMmLRT~T$jJ4<7Uwblb{QEztgTXMaJN~NYSF^b zO_!Q*-WT6Lr~FK>K>JBE$tzf&<~H5S#kLY!M&Gd4J5AXV-y=Hww0!;{g&gN-p0o$q zD4#xQKR4fZq%zQ2YHfkNa`I~rYB4z_rH=VAl>piRCn7mU=Is&aDJeH8X>Oy(%2 z3-mV3M^$<>S*xKn2ifr7QI*d>fLWRacX(wY|d?(Q^$d=*iJGA-H?{b6IGQe6}5}GeewOnQz<+*=as=WAaa@;zRd(v6B zt1I#2K->!ReGQ(Q^h!Y!D0Ehez-Apo=YMi9mLodSFZR!unszXMu3ARXw63JA_WN!Y-O` zQO~bkiQvxlCrx+&n8-a8<;#SVH zGTP;5_juBo9d%wwq2oQ)h?R^F*w&b} z+-*e1KYkG$-da7Y?`MyieN*XSL*#R36lyq;3$hv~&xE24Av*-3m-x$Qn#z;XEWQs~ z8V!QSQ(LG98Aiv};+QYIn_X}RFqR@$$S?HiBqBb&A;BF3_8+rh+>R1PSsW#CD&4YK z%i_3}B3B1Y!MM0)dC}ufLP74Mk9PpsgGb8y$fmol#h%!2*_km9pJ+L5Pd@^DLY3>W z8P<%?w8DBW2vcZ$@7|uG4_W;5HDfK#UwW3ny3}v+ytcfk=`YWp3~N!vOr0l8FiX0U z)WQB~4uKhtU+>E-V7gqh21Gk7e)K@#6A=7-st#3Iq*eWc-)PWBlvRDIt{TzNL;ZuD zbtbLM;`zm6DIES~GoyiI^6-&05#4Z3j8LBHGc zp1MjITUj?4p#)1vM(50@S1aSNh z5T~l>BY|^J$fNA8!V$6!##D^fNIeAbDWeHjUpg|4p2|PPfc|2zan=uJjdT-_p|vyt z2^gTF%M!gbc#*AujAUuv^S>-)kmuU%cbOaf$Bj$7Zt62&$Cm&Ea~nwZL^>%lVBDJj zN_sNFS`vfEN8sSovnJqh5iF0Tu&`7gu38zmN?g~Qu_}f>=lfNn1*v3!eLchy1149NuixB>+McIR0Ema_OEE;uA9OI2 zOS_~cP}O8M@ku`gcO!0W`%QcItDE{ViXoa*Fvm zBuZWJ25K#M$HGPxCgYVGH>bhNA+T~a^~!?;O++x)CHkQn0Bm?%W5zh2ZP7?eGdW&0)Z^8-` zuJ=sI7aPj<8V&pK^eS+_y2MGo3g^FUg?VJfN?bo=I$QFx#5Bdr3nWHN{Ih+*qr4Wt zm>sSw4YX1HOJ9}R1Oe+m${qn#$Nk%g;2Z61uZRAO$OWi2ji^M5&v$R`NK}9KeYYwU z)*#2&pkB7X!wKgmBZt4MEV*#9bVyi*Vcs2O0-BzC15dXf9Nu!8uCW`eA1Q>U^14&j zTf#Omv+@|)SCW#HCh8^BD*<5I?3&_p0B2=>GQqjSIuYt-SAE=?-z>TVn(0{|Ujwno zd+F0LGmvYziVR&_`+&wzPKuLBtiW76HftNu-qh1cuqLbbCSNd+wD>hijr-BGDo_R?9OX{Hl)thJI z{EJk1IN%raY0sxS#eA48%;H4oerRIY@6Fm6&!I{CiN*{4-Z7pUz~9rF+oIz=p;42> z?}V9x`D;>lnz~barXLE_7tOwSo4PY9!ZxVzbf<3Cb7jsB(fBmaoFOF>c*=M3t8JAy z>LLCv8cu@W8ZgEEAS34)vL7jfYu4m?QI-CeS+g&NB|0=`5J}myqSv@zxMPpww8-Xs z|5jh=-4YYlLdbk;Cyw{{H}9x6@GPhSA3KzkjZn;QJ>-^C%#MY&jI4tuvFzvZ8HdtU4q zqzXRZD|JADz1kj$Zh{&(p7f)JpC*kI5G62fQ=c_p_p1_8+gKjFv4{xYGX|5Lqj-Nv zGj7S5x|J^hw-&Z9Fg-mXgBj44K$ty=Wny#jMSa;`_nFyTx08zAkqE@)4Kp}lK5 z{I?uaLI(G5Jy5N&uW8{80=L=irC3Gn{9=73V)DR%?y{ha?jKoE601T0O56u5N-yTN+N1XoefC& z@8*-48(epY!8+8I;wNbEiMP5f-xhK7A5@UmAIU;yTbxWK3H8c(0o){b_*u1I`$Otz z@)d;7Cv5q_mz_=3M32(<#5rLkTTc-T6vH$2x_k6cNEaXJi{L0k}|mWH=(Y z=>?c9&WE{AHjo6=uAJS$UMQES#U-r?9l+hJGUjuLm}#Ahz!U6oWB_AY&>y#xh-SPS z{F=|ZN7&b|_^h#NiYn=dU=PprYV2tan>&T#g5u=afE-!Z1?qj9#LQS{%A7Mmefhif zH@Reb33=K)=OEZ1m5%`tQ%>jwQ6*Sgf7sd>LA0X`=hbSY=5X!QLDeR~CLAw~WvoH7C1MorD zIo%!t&vXj=kR87B{LTWo&pK;uabbAI|DVkR`>xJZ@ruIK)7>g;(I8NB5P#=FkfL{ z*ZTLmRNw%A>8H#*3pYGn)Oz}bCQ3dSx6a8n3k z@go04gS-vzyQGNcy&&p`_JBP7d@m?W;#&g|t`(SsAv zHYJmc@&xPp>@Gu+Bd~c8^9DLD$G7uc$eTR1y68(4sz-7RnLmy@1^uf~0+;6VC0?X& z1njzC&x-TjLZf8)N`UP+W|f=Q!V|2!7h$v!fc8( z?}m~b4O(|k1NM`y2s|k- zpFi)M)2EY1L3MnvUGqSJJ~4Re^T&B=ncp>&U6&0cY17$&PDn8I&|r7aHFSV!T4PuZ?a?U+&!sOA8ZE zZ$Gn{)3Zy(`8P}lYd?98cVme{jTaT+cAt9FWdgyF?CtBZ(Ar0jxEbHZK#zx25r+-% z&1-f5{C#>m>d9pSa*OtB25@+o81rxzaJo5*@3aRTx@lj<@@RcBsBZ-EYJGC4S3(1t zj%z;yx$jY^_t%8tDdtUGc03dk*9!{Y1^5*QI<(tF{bv3+^HUkohvk4FPeJGf!g~OlbnNWr29|o1#38TPRQhPMA$zK z0~U2P#;lmaaCbu_FML!v(GWBmTan&q ze80-@f5=r=N%&5|JKJ$>tWzNct6{9g`Li0)?2*-5%8A4+IDt*?R zmFtA>$TYtB$>U6fyT%kISsum9^}%7VDDb-X7j8~S$Hv!q5Ef0 z@J6`5N?M?v0HlAUajUcEe4(uIOHa}70kiJwoTMQ}2z-Y=DeB9>^_uB;gLFM(^wk?@!TH@oATsTo z-r(_rWOL!oVmTwI*t0+VW)pj2a@^jNH|y5^OrD2S7XRMQH=Q&gxd$~5S4tW|IfI*h zC!p?=rH_87?}nP3|IJq(`zM(+P4D?OY6EIFX=3$=s!H+(B<=QMPBLYroTSI+&g?qPSmH7HGM*@lF~S(t1IW>{!^s&*+wKv3IK5r@|n?G{TFh zn6aftq2&~1#9s&d$j!NsE0A(Be+HhM4yy-e<;qFp$EtR(=sfD-k?^hZ>EzHEN3F~H z{L;c>8Cq6n%O)?iVnS|~w^ncd*EVPVh(;@=;hC3DTy8JNQXe| z-BCrP2LAquEDPq6G2YpQ8N;QmcjEsFbd$BLy)){Thph>8o@G>l2Xeg_=fXT`vWo~C z^BZm85n>N_+}G$Bbp(u>+Q2bJpjZ5JU)NCY3;EdjFsAcL6^`#O!|$h6%*>S(n~i>= zwsEtYQQOVOH4C+lAR4o>Te2l5O}yT(3*$UqT|8lg{izz3G#tqwH5s z2nDgoPguHc46+*is_r$C(FjzPK3sB{lN4l~YB$j=+(qWwuSk<7o5hW7MJk&$nE8vT zbywgHbn{ff>(BJ>ZQveea=o+!NktKUuTR(tY15WJFInF_NG z7gI*aFZ{0Qt1o3HC0Ume@n8n>3)2Q)0vCw~?Ph}4o~$%rvAVW_imjO8Hs8!`2}U*x zUMa)+Oftnhd;`4!VuhCB_SAwKU9^2MdQ@)=D1-f0JoSO`>zG59>JR=M{zH}-X6Zi2 z>aQ+`FT56_pI7mfyqY*P5p3yJv1~IWefGi<8Ya_sFFTi(IMw{E%w)J(l!4Oe&TYQ0 zSIvZm5!jPYwKJ7d=_+Jy}(HBByn0ILsONk912wSz zAW<-=E7&st_Gdj)_A7TXIk$(&L9dxv=W%pZpAph0%XJesg&$uPg4?~lU<~`SMg6Pg zK2h}g=Dzd9A#y4QRBOP}kmE_;ggqe{S5#M|TYYYsNPGQjoFsbI$0IcY5c=yIe{3g$gIBkx%Y_6W#N`|tAu-&#Os@@& z-WDHxbg^XY)=YjyFZ!FFNcnzS#Z1piw$IsCYSI1m`qyUHQuL-h2u8?h{E3xn1*&Wi zrF&bZSB`_^e5A-dS8&m&Pu}D0=Tx`W+&j$XnSal5*>H%>SDR?<2!9)!e6Dc^A z?5?}In~DqgIAL%x@2oLOI8kuyBd$R>(*Dv}CeH^%Sa`)u@r_}RDg4Rb$-7q2RQiSd z?iOuv1vSI7*jx;-yqWy}Xk&v@lSa~ZEtP+V6&yQ1hV~~!c+4v5ZMxV0FMB z#`!TkpJb*Ei_6x%_>@xAnmQgg#x+M+T_ZB&fA4#Ob(Q=x!u#>hVk*tU&e@?Nw|(UO zM-4)}{1E=-vLwf+XBLx*t4upUKQ}GVpTpQKVUm23%OWH0+4WCJF%B?K6tsPbGg2a(HTI?UN|@ty%lKim!=zL*_la?B^MUbN(F0dh*hn60@@m=y4mQCLt#xh<|O)(@Z=zA(PbNW_JQuF=lVvl#7_SsQDn)vd7e4rZU@(!nvjU* zjZ)RfOlW0AdcwD?3~bxiAbU;x%Wjbvo$uL~4l`Xh!}D7XyR=!5@w$Ald=t;SOZz`8 z8oPfsRIDwGKr0an2MwNdHQ4ZHX;*iZdl1;U!3W;16 zoliad@C(i#?w~t(L+le5;pIXKX9?>{8YiRnoqnUy~P0xtpf|PFU5Z6 zo~(YHXtRysg=p{$q*7)yr-6R;GLpv%ipFcubCWDRvArROt!Y`hj}`MP^9sA4(M})I zzsD{8OQJ_*fEmmhJUTTd<<))Lx*p7{DI$7ep#0!04__&>hKTWLik_3$(U3CmPLz=Q zfURP;+GBn`RW*Z(t)*Ywi7s4D6_=SUT>ep5nb`w~f?N^>zf!-W7gzChdVME^&hlDa zPZdm`fdWYvY`$%x*2~`G9hwkf%aBm zm!(^=(VjgfApm*VSo}5b7-~$lw~zBZn{pG zAdnYw=Yy{f9Rg?$#G{rE^1|kTIhBT4M3|fx-E-GxrVSp()uC_AM&Sb@$tK7d1b{YA zIuUxGzs!(xXQG*t^a()YwQ7`1&t4M4-x%2WqE2X-N??%D#b#v601rtN{2K)k((?BFvO|tY_8y zg`WXS+OWa^8Cm0YAPy4)}Mf>CN4r9rk^fpYuHNj4IcgFx=@2eZ@_V>_a2_{J995WZ70%x;6 zdo3v+U+&X(*_O2es~Oo|tihqP=X5IO7$sKOw4k`C~BLMs6QOVYH_e$ z&}Q^8n<41q6i)zJTauzeb_4tN(;0NIbc<^ysZkVJq^<4i6xF^49|(-q70bZyFZV0! zvur5yl^hIY;*)$D(u@$kej6IFye5DKZ>=wJ4iaUlxL!npSJcXiNkMdVuFcx>;QD@$ zq^ z#tm9IQ*Cav+YazcM;ULX+YClt3Q%tNBRJL}@nyJTS2%@Vi`;Ft54BbQ=U>hA740D(4-p~M4*k4vB^(5B%xkh~F+`_~mX^e+C{87QW- z>imLy92zsZm~t&8vEr789z|5iEH)3IEQ~oi*!b^5<+3-iq(tT=)B5^TO)4Wo?eAySMjZU zR28@Kp9Fole@&WfVYjxu`R%kdc%JB+DCQyo`>b+!*j`?U)LDV8Dg#wg)ZdkBRDdkx zvOLC3U?9#Cn$Cik!Z{d$lAtMA%%C|Yr)AO58h8E#qP{6YmjIxTnv+vS%JR)d$3*g% zT_7|)s(1Ajl)nU$KUZnRBx(ZfcTmg*ERcqj>yY}DM8*db>7@e93jWGvbc4)PeOaF} ziE!wkaUt|`(j&;s;}+d=3AsU$Sf-};iB{O`0w@eKT1=e`%^hO2oj7a;@ZS<#t&iQ! zqfkpy6RZ+O31$v3c0}HKxNtwkOJdv=cj+%K7Ub2RRqq=xIM96fJ9(H zQphoiCK>wW^wU1)cuy|llUx0SnH~ETgZxoYJ8V?`U^q^KKd z7w>UA#CJYuIKdt^bfP5li14={m+jX?IAltsV0milvbZ+>LbpUQ^z+GM`-iPpx|~BW zDzju{<0wDR1hOwxKAmFS2aZ=*x;QM$4;6al%22VfHe56KA3u9)zX^DAIg_O@Zmz}H zm-6Ls@_g*Q$*l*av_&c#fU}dJo*rt6LcJ9%xd7U^{^#XwK$~Y6@1J=_Q;YSH`4lZP zkD|HQ5fO;WCkyVj-3Y?kMU#=~ek|4H7}D*qH7<%dWltr)0?Ma+LN}D~K}Zwo?OIl# zlcN`3pZ4dYAsou7NOF;0sSDWvPDS&VkNg9eKPwx+H+-}qN=Be>Xf{hIl>4XYgiT^1 zphe7f!)Nw&-@r`v_t$rl71N<~M^0r8dk{D9o3Zr~*KZZEoZB#UX z=gz2-gdQ^_=(_O5irWqZlc(^I|L&hdj2toV0M6&nE}}c~LX483uPgiS_JdaF(~_=3 zbhB`J`Zoxmr<6IVI<_2Y5&!Z%JL3UX#dT&&70B0?IELw-na&N@O%fZKhg+E7@_3T#y37&WByzw0g{!=yP zjXj=NcfH(5rK?En4OXUbdCWHAb~2Rz*CR<2D`*ILDz7XLuDg@C396FpO^E7|LA@!n zq0?m;>#42wh|~a_xa=o_EeIJb><)P1s36rtxGla@h66%yxzTiB2!bLxNu3cr1I!$P z5MD7dVT0kn6JZ@ISEa0o!Dx0>epcYsi@~F|vOG6=6m4=Z;48c$y9S3TGcrb+dt-Yo zW-iZ9p!PzfKb7SbRbKMI>iK<`8^Gt0;B!_mXLR20HX>#nR2ZKF92Uq{P(=MV1n9|G zllap1jZKix2xC)O9*ZFsATC{e@P}Im&U^N;%;vBm^%~?K+;9PyD=+tf^pvOMIleiHPoQiM`HYaGyrZ=)^( zKc81B-()|lD96No1JP6c!dqXz;GTW}51^5MI|9rJT-JH=A+_%k2dQ(T)Q)OG+5EwT z&6PCzoN3gaT#%_9-tO2`fh7kBPQK8;NSNW+QL4dNh+jM-bDm6S<_4O{NKykrlM-$)qkoOeP7%Ld`|n^wbN{dnl(4_~OG(yl|udWyCU+7s!q`7cG!+6Y3IZ`Bb!kgzA$xlt% zQxD)>Lrw}+nT6fO+`{;q4FKAU&Bj*-A|G!#84;U(!+R1h?oAp^{artv4!!)dC-?aH z+W{PODEvVb^`M$m&xOkJdm==X3Pw7lIEt5#_&o2T2 zS@H{*-pWgaiaHUV31VmxvAbLhjT8k*bpjq%Zqw#U-UN_s%R}OOe6F-gq|=`t#)8H% z(2>g2=lT-Rfxlb{K&(vL3l)5+`1P6dNDdv~Ao2}x=x}v%2uk*Kys^mOj_3iBo}E3E zN4xk$UCO{kDNeQdM-wH}!nTQ{z1`Iag+7I`DX0LpjxEm}p^)HDbgyZVHK-w0;-@)N zmt3A+G+{f&5-mnfLt@^8$*MZFxFwPwiw4CBxdNAVe`jn}?M)9%V;!#Ph3!=>s|7!G z3@u3VH$wVB@&=)yzGp?cW$$PTyX$;)dXI$DPBGm)Izb0pJ+35XiK}JK1FU^O+j6!U zg5$^yesZbsEYk|T(gVooV>esA&`;ub&uP(M|4O1z@7>PX+?At9T%G^ur9wiUWKnz< zMPCczMS~@+pbF+OQ_v%}41&SiNVO9%pNX1V`-{ zD^>j4Lj^k0f4#KZo@vE!=$tr-4yqDBT^kTOBP50F%D8P$iHzdP6dkE?X-+$?9#Tva zDm`@Lg1o-ScV>Ti?gyKW$qaP3`2mq;QQ?UR0hGru?~dI(3Z`JJ_QxG2zO#)VufC`U z87Kgoa=*<)=kJmaYTyRz8CavT z6iZ;2QzA;v1=I>k<2wqzF@g4xG)>Ispys~}^G@H$ND<908cm83+W%Vu@R><7o%g@9kghJD;A?PlI zThUC{fPC0e7-Qz|3HW&aLP;h>VUD18f-Vz4!H+W)CAINIr&XByZOb9#kgwaN?^sv? zqDX#f*-DSlNugP80n}wNO2nkb!5)9*mwYYGBC@872clUQkRZ8;k3UoB1K3)E3t~&2 zrr`Fz*T<-<;Bid7eA)8KM7aBZt@B?6M9khB=-QA@-SdAoq^{c^#Kcy{$jCYl&ztzw zrTw4}GUNgk#@(51pykQh>#`LO6PEqBL_rRB@;j3BKGIi`&379hel%CTkM3wh#8#8w zPlNAwtx(igO;gRLK@xN~xfDpEl&?(-YJRnDot@i0RWs0WgLyo8!S!6`3^afbbA1~Q z<91LiThTEMx`)mip{KuXPd;>6xh(h|H@cion_fUhM$08%bW*~fgV--_WB`YcpQ81& zqfYq(VtPT$Y1>OT=PZ?ij@*E@89gSwn@Lc?%;rQy*e#3+Q`h#t9O6L&uzrmc@%G!+ zVMFD-`Ve6zSc%>;mZ6J-s7$;gnO_#czgrw0+(96Q(19FD;(v$z~fPIQC2q z)j|QQ^HH zBroyMv>?Um@R2@C?Iug4)>l+4E%&qSDpM#Yek^n(f52X@G*o9?v_lu30NdA(y zO^^0E{{No_Ntf5ng=F35^tKVYZu~-lR%&B}^S@89o8gNbT7WtG#$N+TiizHxtkWO4 z@nKqXFFj9_-uVRedNbLPrhrj0I(y%;;}s<1!Tb-xqj5BiS-EAW2{&6dvC_gg6?7? zK+yZFW0>U@@Ia!c^8hYcbZEAZbPtU?-n$MseYpchuXLjJ7Uo}~N)TZ}y24equGkaJ zBPhbr;&0{QxkFp2=_<62L+*^pB_WqwGK~;6moi1U&26J4%w;iS z_Wthk`~7wHc<*uE@3Y72^}LVFO-d8?Md9N+b0R?=MDljK@m5BQ0P_c z0iY`;gR3|~HhM~~0Pk}{bN;(qa3X*!q680DQO`VU-K2=vWL+O$$WW;6J1TkG{fbm{ zICcQwnS#&`rze{7#Hds-z^f$Bz^HnI)LvkSy9>XQDtbv7py~A9EanSM@TMx_TyFr# zz8%0{7Ert|!?WueYS3Q2>b(0;^tX?%y0d0WX9bI|g(o2q!fJLdqZ z7|_EACIW~!n3P(Uo)&MzBJ%V)uuMQ!#TcQ6w_sw-4EBj({OcKC4NUqU*RCrL3GUX@ zI$nAyDQ|^t{@hkF(S@uU^oQDFA3oX-1{)_~m&>fqBKD0e{+P1STLOG_j}Cw5dlFw1 z&pXFvk3TIoi{0xX!&iY;ZYuDG;kW)vmt!43XizEjQWIKlyRSw)<$P)4DSEErfI6VO zqbiO6yNbSkvZ$x0aoHNSPSx4RO~C990^S+j5!r+B4jW*BcSd0o)(iVt7}{hU&T=#T z2&t|){Ndbwb^rIiZLXk&!%?bkVL@2d^BD0+{qIQ)2$*HOOM;hUiadaKYaY z9xRyWAaX2BcpE2TqUH^F&-%p&OaTu1_nLSP!F*V8{|Y)%+LNPjY@|7GMCe}f`?JAx zJPrJki2OJ%w}muITI)W(477^5Jpk9jV1JKy%&!9D9o>I`!AL|QSpzZMsBU~4=&O-(Wn53c7+A~Jz`sFi6 zleyNJfYXi7i{?^mrgR;bOnw?z(@EJ3{-YhG*CWVOh@!31X0`P zd0k?K8nvnhWWpg{H4KZo{{_U9c*Iej_LfMeutk=+Kh^V7f zVAN~qMX%Xcf|JY24Qa$tn{T&e3pEGjcsVK_QHUy zW>cgjL<)9fvB>(UhyGA-UPZV_k|E(~kGO3n(S$izkK zKyn#5WhYe{=0#OESKLcL>Az|k(x<~REeP*aM#aB4Umo4uiftGV&gJkaj0YjCznr~v zO@g<%LS=NsufI4Hzg>RLmES#&EXCXTMfP6B)!9zgdc@OR)YR)XM_owwHykL8DhO#! z>p_W9>Vo2@AT$lB!g@&xS)Dlb>XcO%QS-p8N|$XOHduUr21IrWx>)=ANk#Hil~zt| z)>Z4nMR)8{0o-9eIie_OfjaN|Tj$HaRad90PK!?l6*6FC;_vCv5 zow3K632C7Tbdl-_!@WkOrZW&9jp~!$4lOsN5+~E^ix!2!UQsV&c__-GD|`+6iacc2 z{?oOnRMX6;A%b*ARypRB?HvP;bg6X|=t*CFrJ&;U?N{;_0zr`1pZ2=OjMTF&M|;mA%HAaLF!Uw8{WIXn5YEKOaur#qDHa?U1S`4as=X+C#gpM-@ zj$?bh`ivg9ZQ)V{ICfVlL?m-^fs$DUMy=}39owLjh8o0W9ivFbd!m~Q_WZ4mK0U8v z1$!bO-roai7wxYWz9_A>?m<5a{K+P$qrA3NTS16!KzxJkr@fmBbNV8Yzjk6SNh58y z9%qN(R=R^pk5;HeMVbau<7^StZ|mNauA;)FyY`Vn7yM!w@pqkuzP*DR3LTRg)${uC zN-x@!720K^AF-Tam`{}U5`+N%^QAfZ)ZV3MAO@=)NthI z$3b--^iP1fVm3M!JDt$!QKY(G{i6M0bs-mvi_3ioc>trAR#;6i zLO_h>iaYjFc6M1My)K4?1rr^xo{Jq%(%hdqU7@TVwR162dQZn$$}r1| z9(ad(_%qr=$t&m|?smIN=*jt~;MeBHO8Z#!Q+qp9A|&m#sQ4=AY(lz6+e<^<3tqD+ z5tT&eDj!y!WRRL_WBzb%a@p|pYO!2+hy4M7OA_Qr`WVy znj??Cht_$Uys{|)?Xoa>OVbQA@~?zhDqL6dHIyh38Rna7?|yS3F#JOep;iFD@`iy^ z0`F-<2vTO<^iAUs)93}%IugH|5uxL<>LRtAjZpWBTC{q4YEM!uK|Y)mw?Ac3bbqEG zVDTkqYF)ER(p+cAowfBmck2R%b#$02<)0ZZX%AML%f{!uhLV4lfE}((wKSL3-yW~W zK*q8$lW(+VyW$RF)!q=TC`<7kOnfg!sJvvlQWYL(?9y3G`h;5vt{h~Kl~nxwr}0}h z>&so38WHvAG&~a${H-st&*$qNr<<~f9C1I~=lKmo&p{Mz-jr49idf6Jq%ql@SE?<5 z;H%ys&ZFUijmDMI$%2iGL?iY?@RsgU|1Fn0_LO*k!V3h5sR4r3XCquFm`%JDM*CTN zG>;u5S84xc8qur0E-kXxaveDKlwx}iV|g8Q>8?efdtT>*QK*-K73q?wKmPnFkSq}u z;d5?M^=!)owBmTdrL!&S$Tx)@8vqE@d$2AZijO-qhS?dAJ0-aDzdRCeVT&$cvlqph z+hkc_w&5b7*4M2O$(5)J@B|kW{&w2*XP3_o>hemXYpeGS#v!Dy3_hbU!m^tp-67Pt=ywYli=8WN53VVUd)Z%9Ww^=NKeADvNL zM!D1P`2(ce4#{amc+2QrNJEA*i&O#9guV=LjuUzgqOYWu2)i+03+h94{23qb){Xyxx(@NfOAKD1M!SU`*C#Io22HdKDOlYMbt)hX zpUXY@X^dun>Uc))d1CmH^G>|37kcKRl#};aEN+Bb&Si{YEzx{WlG(kplYTEG^1 z%(t?}P*dj2Dfx>95d12tw(!hmqbR+Fu%BTtuU8Z7su z4wP|K>%0+!VA2bAT}R*N4^hWhjBdW;PNc%-qjFN~sz|~u#?B&@1eK}_@zC%wA8pY~ zj?qo2b)HVlyKEm17!8p?{EakAXIuX@EsfQP%MgEIpLe2DGRe zBOuwX5Gl8>%irrWdJQz5)?&1?3^WY)<*ooEg$T`BRHXBHA9E1!_xC{7N6u%ApH0O% zE(lo3%vZ;CZk2}n`XDi7uPARNuO0P`AIMw?0tYFiG-6^5eBUc{OrE5 zqLT{CW9kLLVj$uSc7+q3AmFaByVDFeM8K{;l4vh}`>pg$32Z=CrxI0!T#5_DN6f!} zvTc2;d3i6A$w4=QqD1MGo#gAs-*m@fzjjA_Yfd9N8a2TUrMCorLKKCE zf}#WfHD87Hd;{Lk(+df4Cjeckf4LfGOHo28km$F>=8=L8gZh=7$6xneT=Y629fbd` zU>Q#64{%k3ebs`iax1(csI*;`O%Z%VX@s4ezk}^{_#`LEu@K0qwzp*BiN{NkQEJ?K zWrSL}<6UxEaBg$FiUtv64cTpWAX#g{DH0q}Mf~-_UJ@jI-o5uy(N`rohJd@64@&@k z^^A^J)Kd1dArSc)l$E0&{1k|+R;(rbeQi6%z%9d*e}nc9=VoH>VI@tkdAmzY@-R6O zfnv|FbJ9<)gUbk!+MYgn#T;;#icb`Sv^Oqr+<^QH2nNKBnd=N3VQV*9_y*#I-#NFV zc*cI*Q06AK#Q4>}Fr24^;k(?_#ajlvqx*uXs^MzHDbpM8fEF8HO#<@a>3^ZuHA=_)gpknhQtNUJNLAz7e<##6vHmrAt&H%P7HzM4&ahfJmv`GQ* z?=||Gi0QRA{q?;=>?Bt|>HF30Dx=j{mK~Q?przXRb3wp3KtoIWC_o6bu<_H61QiWV zR=y!b_uD^At0C@$x%o}d?4hQ{sW7Xj^=r@ftLd8UQDgo6{(Q648WT~H>&R`^9hV6R z2{m!H0xk%69J(>_wcmHztnZBdFi=zb?UW?~#Iv4_dqU`6njm2PCnp+oMX{OMrn&RRgop4sD#f3uHnVpoQry{bhO6{ULEA^2`5jWBRxEXlunt(Ny- z8XiD(K`OuXrG;d9;{Q9Gx(t=G{v1r={_GNEAI7uEz%sD3N*8Wfqw zu+9?2-nu>5lv2@?FU!C=?uaVDV~wtkjc!^PXE;+CMqJo7P6xsUT0?>CC-zI3xyYJblb!IC%A^1WEMXrxPaN}iprX9e4VD%^b zS!r1$ml1>7F*|N~03~3#pQ{)Y`4sLZzXx-%%5O21m^sRx2r*Sv_g2Lk@j4yqT z#pXBSKVN}M?Z7^Ld$;kN(x@I98PN(tdZzZf^~EF*3#Szm5NbI3*0R{olcxsB98SO*1*;9Ub`~0R~d= z)fzt6VEE(?kJb^gsDvc+0~pGHnG{ytye64%&T^3(#H+%fea>X?(Qux8BcdT+K)k%8 zE*0Dlv7Wtp2t-&p;jcSK-@Rj_-CmVv?m|ZULmikNfykop-Ekl{L6MEKkb(Q?8B_ib zb`nJXI-Z!chzu(9daXtZ$AP^&1*JpO#5->No9&7D;r1^h)6mpS4Y}zqi!LRHB*EuPKXqteme1b z8v+J$Jyw#?J?h(Yl?OyedTB89Yb3AP1`5A`;FSYTPkuT^N* zO-|WLGtiuYyJ|AkuO2tHates>P3Mga|E~ckMKxV>4wi|LADh=|DUF{m7e7|C3c0&F z@GB#aCrENq-1Wyl*z;?8$?AY;!^GR&gDila!Y5dWVsyy-X^cS;2F>*K?kkYMeO%o# zyEf#ADkT4E9vIT1G%6i5{(&1tq*5mR1`Tw?1%Scbz+Kl%2e>;3W+oBR@16#x>CtQh zv>jd2krN|B!Ac3JUL9N%L=t#mt z;MnpWDFV{#Xtw7(!(p1L{eXdcYaV`iik2@@QF(Tpfs=%t05XA7)cVJUyuZ&|`9CeY z0iNx;`YOFAWiZ*EcmILrRXj)3HQ zt@V!D3Hc&>VH!8&IOE~c!VSNw(2(}c`<60UuVyyu_=c|1r7F#bd1m4uf5 zlP#u`tdW5NrKmtbGx1f3+&(G({St?sIJV?8%o6ba@^T1z3DqY+54NlM%1LSLa~(1m z1_apVZqUFj9wm?vC-QwQNLQ$2HV85%j*xH_A}&XKmIutj?gRqWw*MH`5vHRd_=q}} z&`?m|!-Xn>QQ(lRKEm-Zi@D67*>Z~T6SfeSyS|!s2xy+jHEfCKPW<4lQf*FN8ZHnb zk^l6yNIz}zz57o5iDKI1V4s0Zh}jVvF30?FehA;GxwUO4{fd*v`;5o>zh{Akv5@Js zI{}1i$^pdBKofDzG4Z(4zQTPKWN6sYWYxV^KSIA&N7ob(LA=+|<)U+h%tv>YBM>iV z7pGenk2^SQWk8P*sn045$=9s$C(pz0^P&{WYnOjzpbPu&Y7U3Ks7rBd99$XbuZI)m zwvqA#Q3+w1f;zD@#gx_v1K5SVb|ABuXPG`ldUfpMEOt^j}oxzs5!2 zj5frtu(YZcs69ToXESaEjMI1l^wB0q zB-6i_8E9-GPvp}{Ng20N9&K9(baeR>4sJXAffA5Wa{pjIR`yf^3BL@RM=K_}ahahx z03y&Rd`A(^_dZ2VX>d4?lLH$!@zgiZ5rZRh|}k#QLAPk7Z{Jo;&Wd`(P!#o+5Vu6++JxvWMu7=cv!B4IB7s0 zejGDoj*4yhKVjFc5r5>+@{b8h+qm)EB7=_qPQ~G7=A%JgF>!M`#W@-19C$A%0GvSr z2#-XO`@rrh0^AGJ@T;ZDEg}EVNvW8WPX+sm)fE$hU~vXzd5m6H%Y~*1?vp-cGLG2g zW$~5_96k>w46OgA`+Y;v>7=&Ou z5NV_%H>2PEBxRF(-yB*@wtDj^z=y9e!675mJH8BtX2~ZVAw?WyZ; zsk#3uG-=?J{Id*7I}oCxEBqoq=9V~zy|IgGKa6piSM~tFf6!gAz?vq&8ltR@m4`aT zTK$q`ID}RAl#YDT*nS^_f1vpkHs>zEv2yB1f4%2-K}CfT>CUnK`qSo-`KOgW?_M<9 zvc7(81@J$0zM)&i4Du+&asGuz0?Y{8dFW42Qez6cfEm*bO@9!3rb==}vkUkT4w!M; zl@st@eHH#YmBYvfnonDrk=ug_5Av#fh3HoalRphSA{_Mc0}+nK924F7;2CU!o&^F$ z;g)uh?(I)Gk4b3S>Q8C(?js-@D={)a$Bs{p_Ai_p;n(*};*LPeXa@zR!E?T0LBrLc z4hB~8&ioFudlEqs#6X_iZ-J1GRH&VQ7c^Cm1Jeo((!LzUpc$Bg(+-Sa^-KL7Fw?G$ zhfMFm0`o1iI@6Aj(p4rEjIQ0wHC?)pHR{{z2w2S1Un!-i+~uSa*hR~0@dV`RPyRb} z9g9&J0+LGBvN^Q|2-W~CT@Q|=SAbKaIUE;K1i^DVpVdOZf{Jg44gLb-x?EhBv>}gg zoIJh^ZWNVG%2a^!Y9-y#Cb?$qgA;zMFlxL6CoKti3Jq)Y5eUN%N_9o&16*O6KpXn7 zac&N_LH_@XXkcdUXp&6s4ub-dTe$zOQ)lNSWf>7#Ch_==ir>CSQz*{^v`7cWD#%lv z>P}mc%XODqZE#6Ty-eMwy>>#QHygK?tI@*=4jp`3{C18gteHf1 zHHS~wZHW#{GE7XG{Ki}mKuB{#6;j=yqzi67ZMH2Mt5Q`~m%_?|NZHVc(Z^IdRllBl zb|`n+SdL-RRt8rAq7c#|^CZ{Ml~%?#x8UC4YMPY{HyoCq`pCJT6y)6t^5!b&w8{S! z)GZfpe9P0WqeUcc&-(*f_4HhyxJY=U|FJjEZ-JO0Q^8D%ujp<+*iNAGQ8~|4p6mKL z5k6{QU;pEv^!Cuzu*ORve6GUE)I2iFDT}Y#FVlx(RE|v+As=BST`Y2ZEjGwfsWMQh zXW6H%%owO~QON%6Fjrw{WIaY&cgTh=e|@UHwYhEpk|eS=D{_?_?8KTa-BlS{XVD85 z?~EI&Qcc(%TuaB(u#N0yAO77#=RNa9a(u)L%e1K}&)M*ANfDRsknyg)_cDa6 zPGKb<=$LAqUiiWabMHB)RmiO}SD}kPplw^nL9h}X^H2!>dwI$aFZ*`R$D4|yxpw2f z>AxS3)O~pHrM{8~&kpJsLk6E(8NJ?)MvrCGSE3c#ZBwTv#O6^AmOLix1TPnxg*KMi z9b__WM(_C{LesUU(kSVN@yBDE=e`h--gWYR-+8Y>OV0hAvR6%;n^P;9$$x)OdHm?N zvwNDbQ*1suFjYi{HVh(%o7Ym+45n)E)w<1MDsu@z-Vvqza~DmH&s8Mt#I?u%xbR16 ze#PQzjo5>jnU`%*+RxYJPgtCkA%4*99W$;CohzV0_kD79cs5)J{GC&#v_uSfF_3U|UEUZs&2|b z3V!{!O)I)7gs-eU!uuDw>Vno3JzQLC_4)$5=}5X`-g!7C*6IB9vnwXBG|Bp5g#mE0 z?yGV44whNQwFS3>r6&N%xa>p#zb)B@H%?EgRXZeOcGhGmp)IOLycsQCTQ6IM9_B3< zdYNm8b<={5?FgIz{Mmm6du{->*UF+^Jl}wE-4Jse9gKPXItbMYH}oiu{aW>ttRbAh z74_){*~;aPYpGl-+VH+;T(5UE`UcNbVzAsKnM%Kg91-zg&0`BZ_vMFiqV3gYt_QwU zTXp5uR-nalpis|kwwJ5a6sI&Nq~V%b6~%djz}-ybjT67l`5+!U(*=6+<+l0CsO8cQ z`rH%I4dcUShz9Q|4?5O4a|rjQ9eKewBia96F>zemsMv&@s8 zkMF;VC zNjfWuxaw&pjWH+Z(weS`mAU46dm>*!doZ4RYRy{rcURIi4dI zdxbgT{a!uqe=xQdpPouLJvq5tL+j^@(o!54vPM<5gzmk_^R5HWtCDusT~w5Be}8^! z09@$j1ErohRBuHe+MoLIpv$l4we6;ZQ&C8UrmD=?c`-(c2a|7A=PnBSyq_OG@OPmt z->I2rHF{!8R^zbjc0w<(%*IRw{AoxCtDnfVBipdYy^->8Htw71TBlxbKK#~#Hbmn) zn#(2I&~@DQwbYb#z`{~?i|MNSKps^En-r6ouCJkYUe8HYMGeaSD_>OlVeCRy1Eq#! ze6{=3JkzBck5)sFvv6cyDOG6LSfY34598%i2fFe2>ZE%sx#aH>+s_VSmTlRqKu;EvxL>U`U4e4Ib&2sJBhQPq_? zTp~G93zylp#`*+2Is|jS-41CDvFG4k(Mg&D3NfYS!zCOnfE*blckF{j$-MgOg!Tp` zq1Ai4fh9ZWRlvWctQ;vOuUwDQJ5>;ezc2Z@uAv-c+;NM4uCP_0n>+}9+Q=GBNFYrHJK=|Ghz}sl6Sh^!5gIhf9^{hGT$9$(sy-)jVvM) zf=)P~^Q&qv-~Ut>{T9yp!+2O~_4+|QL0U8DL`!SIL!FQK#hqQ2|6Wo(t>rI0mT2~J zIA)puuXG$)Y~t$8RjD9O9&WyZuijL@br-_8mdj^?bqnu&A3Ys_6jroB}!m+OV7tnSPW}2JGvJPnE@cLx#Dukvd}!I%|Zms_@S}~H7I@Fa(s!0^n`mt2t?%TH^Qsw;1PlBuS~x$p6!4a_Goi6%>M!eXY&NRcV8 zN;lE25vJ2M9T1KgU>o%FN?12_D5Z~q6A_fQL023OkC8`vjc8*VIfe^Zo(h&(6{;zd z+ZS@~J)_0@_f#g#%CY%_jhq7>THsAHgm}Y7yiaJ8Wqw=l!4zs73Nyl>20&2y+@_8_ z_%iZITa+m4I$_w|2P(psulc7-v=Ti`xt4o)W{B+hZV{<)HGR(6_Y;R6dC|eajN~f$ z)!=%TA!$HzBCBsY*kr|I^ToQ0LSg91{KNm=W!-QwNb^~KzVCgVXezMoVzTG>sctyt z{uIsr;IDAS`#I+kZK){+?y1STAl-en!Sd}{?qqJ?g^!`B5?4+DAxGcixL% zx=VJVKi|-Jbu|b%?!+|VFs?+DeDMndi+HtrfhCx(%meojMAu$jhVWP&L78qmH$1{ z_0RaV!~l8pFS)8^>6Q(798mvMbc&g?D-x7$*gWc11NC}8*!3j;g)mKb4X2O_wf#>ihJP5p zoSCf?4`bu3$|BD^@$WzeYQXWNZ?UD%7#Z*1g-nK*`Gog;JZ887a5k-Ric=@*notQ9 z9YLg%>L>e+)xCSdsxWN&>p&}ireJE+jMg~EEa3AS%QYO+KyP2F_r>=f+vCn9&+z?` zl4?@_Z%0@z2>&Mt$$g5Vdk&f>^phM|5S_82V*vd9V)bFPTcuC(Pw!l{5>{iZb(YO>)z)ZHw5;_dboHK!~L`pUrG?0FcJ2+Q@g*tP}-N3Cqb7F!_JE@d&{00*3aoQpTlcH71c zo?*KQEt!4C8efT?hYcoQ5T#B8EMFe|Udx2gtIK9=qD@CV(k&ERcli%z&)!akb#!^I zAvZ&rV{qD^2F?am-eXvwYL*+2?40_Zu#F+NW7=>xr!y6)Ln721k*4(54y{qsLJ7;F z&lbG!rtUS`$*>#hL;uQJKOELatsOR(mB5@aj0-C!RjRDeaJLPY&&%?E>8ipGEsL?= z7}l>8cy-B(^|De%0qe2FJbCTFYU7Yq3gkzv3s>9Lu)GQG3`g*}0%;dz0 zk)&3DlEhN*i`>An^;pU6Fp)D%(^|B)s}23?3y{Um19(SrE6;#;mOg&VIsCdUAGc2$ z;C+Ki5K@lYqKQbf9UgHN$RQu$dv)r$Qu+9SRyJO;gUjNx<~lGO!-)qq0t~%ibOC9V{_Fu@b4sm=321Nj<@jgQfT0fVXiuDkq$o=9)cFFxSxB*m#!1 zk7~{qun}u6yAP-`lpE$o{#;VHNdNrQr>|nM*7JhL69-nbwWDJ-5Bo8biDP4I$ z^sReKLCBVwBVO&mz!+uf^gpK;sOsZENSp}YLwUo&9he1B-RhrOoC5n)7gN}F$X?Vm zRB?pd;oRC2`%1Hqi95$6wa@SFW&$lV&y77Lz~M4)N1-qP+K)H9xnk0sz#v{Nys-`# z{V~&E)ReyH^`%}=@_^c37v;SBn~8g-L42-u1>b9boL?WP-9gJd@uNmLA?=tx<6as2 z4h6<`<{h4N|8gx=a2afmHW1JwueSN#Q~7e{HP_RQAGL2JbLD12>^`hjVS=Ec4&uQM0wgi_kn*CKbQ@Gl- z4vOlWhip2Uez`}Oim=9Z)$`(L5%0xzebuAF z8X679)a)a7r|Dh;B-D4V*O#w)rX5}wxB! zsQE(ePX-3e{El(Z)*18T;YW^@-&*_)m9$mmWemNFdRSk555&l_@%g_quZF64sus|d z*7VndN78NRL-+R_W`nTOEi+!aQuTpm+?X)X`r}gNfo6p15Q6txwn#`xE>HMVnwCaW zYdd0rZao1ju|Me$1PW&CMt)o4&NqDOI74;c5b*`7wedpzdadYJ*E{wqoFYt9@^f$8 zcczeM_<`u6Vs+ ztAXGTpQdL}31>a{LD0ymsPRS0#QHj5~C(aKDVdjXtzt(^GP&r9pWz<@sN;t>xX^|qDfhY3rERQaH z_Wp0%>!D@ojV>WT=D`Fip}B?=;Q@GOvRpYF0&2dF*0x1Kw9p_3M+6-?S@nyv1)ZQc z(r+B2J7?PT4#L6h!QB-!g5^$h{pFdp#OZ^lni)lfjgha9mP!qLjMDd-`4O?tQMy=> zb{LlF%#89v@eQn~UPfIcWWoSl~%eT7J zgGPNQt0R5t=Fwz>-L0DyjJ1}I8aF)k(wfxGq!r+6uH>0t!1h_eobxy*7TiF9_vG_K zM&F8tcijVG>2M#V3 z*Szf%Y9vItYY-@OjN)}Ffu(0R7L>qBq*k=Upnj1T$UTNSF_w@+aC})dQwJgf9+T;7 zk}O%hye-8!Ak?^Yw15LlGj6%qVwX1~9i0>W$sj(>@a(>)BbyJ^v8w|CS_W?xXtdP~ zfOmupefPmh(E7Z&QhW^UbSLReIA-7gi`df)(JTW}Pg7fch1cnRVJ>*j@i$WL(8&); zi_$IsFpWjUDArQn>nUh8erRSj;H%>x^$8YW>;qB5rv`^24%Z40|?0E2&G2( z5;gnD=XHVdxw+>;Z*PG;HWfuY0&CQt?oQ!cOP&%+ze4r0kC4qD2H~kNx!I$6&-YGj zKscFNE4h?3-uV`|`sQ_8)b?X!^x70qldaJvatKbcH&18w!cU?RNcb<{^gvw~c=$B_ z!)xELz4IfN9^HF+x^K9r-}poSF{XN?vEGxI=>BgDC0ERULK|Z|#H})(Non+B3)CVn z9*_jJ3n2MV=X?xCon5A`FTXMc203RE&%Uo-5Q8`~k+6&JSxhRuCpA-US! zFE&cn%(K)w(CjPYTU{9GJxb!p1B|{WHS~kCRjzFSUv(t=jnE4^hv1=Q@Ivdm0D%p1 zWJ8W~kvCOU5s5Ljix-y!3TPr5E(bBCul=aaW>%|Obk>aYsGzZn7yTKYL=BE1-X9EW z_`1vhxV|b9*syJwb7l`NTArTkt?}H)E!49=yvEFJ#zIWay%6+I38?-ZTB@@io8?Aw zt>VuVaVPT1ZNo6vlF{!qS8NOBHm8Xi6;pFrK(Izw_shTCsr@jMS98wZHlb%;0@3%; zuHw*$#vND=fZyu*O@j}B8xKE}G7ICr?COJu4%mA5;;C%I6YYp#yp1VcXQ$eDvt6E{ z9oDVGLO|v^uNk-jq#@*r>RUR=1UYCh3`CfmfsTDU_aKq*nKjK&$UpC0`B zjv9$#6Z64)a@3&S0Yx_7l{Ct?V5wqM+v%yiaPWJ} z(oo_&n>&v3-6x;8dxMj^Z zedRCE()FkpH6o8*-t;HAj=z5P;s}}lgPk*7KWwZe+C%Ml++A?Q)-@s*GuG)J=eK@A z;8n5pWfiA({XEuxWu4p*JZmKwVmOTSAn+pCQ2zi?N{dW6trd-S8iW5`pqlG9QW`Zi zLr=D&>v(7xj`q%ZPT2k!;6XoXq$0*hfN4GM02iNClgKvM+vi0dQ=?+o=;I^d&#rwq z-0inhW)-BKA(>?zfWc?wLUK^mVb{~qMsB{R#Hn-RtR*l{hM6#v-DZ-G9vvaeDKLx} z8ZDU6ofY~AXd%fvBr(9QpUwlWZ)>)mQ1rA3rl| z!l3T6$JPP=g0mONWqsMru;5?W0NyvPO!9BHg1r)fuIh@XMkc`SXMoSIr7ZsO@xtOrNJW4 z$)z7q(I3Tq$MlKFMvoHFe(-FaX5kyUcRIS2PI8>I{1?e+7ZWY;fFL1S<&0X_jH7rOq zeT(2aiFebr$LAI^w5rvQ$+e)Aa<5+zSABrOY8`g7p0Uf3y2B22qyH(x;58C zPGBs=X?EPp&tTRH4X3j?27y(q+v(uMSY&dq=PipoKam8Vui0{oSA&(z{I!IQX~C$& zKsC3oHkA@@(=qOsV7yS3ZiktItf#*;W7QtH|DI^P#Yx)=OJD3eEGaw^K5G&B2#uQ3 z$Fo|>m|k-)4IW(5I8%Ty=cRgNTa;mpf;ML)-XiXkiuz(|=sa><;}8vf{LACn!9z>2 zmZk5HX;*7wPY#SOY236J6}OOeOVJNyFW6tiWEI$a%g~ds_(@jy^GJ!qW;U1&R~e+W zTj#TLUCg4lJ(eXH)ulAu;Kpm}uAF+)LaO6|%b0Xx0V0e7``V=R#qD*a zcds|ng3|>NEv~1VIV}_rr=2SL5tth%1$?yFU55&;^d&H)QRaiA_VS+BIr~Qlp_(=OIpRnLg@kcc^;A3q*u$&ika(ip#BttB1JjZ?GB(cq>Dpn?_hC z9zSQ!)(O0A77UF6J;9CQT{zzOy4L({84w=-n@#t>vkPe zxuk@EIz5y|+sAyX558oZjG$IiW!jLxeDn&OU~f5{=V+U-9g z`-Z-5kCLC{^KwXS5_LwKM%NkXc;B6Iva#C_QFlN6HraQzD(EN3uA89K+zB zA9@Qta_ZX9{n)r?=RS<8FNYio;c16UJ#L?QT-*o3Kb;Lu42}HdB|LE1)Yv8;RMmYR zFTX=ej$Z<5rsjFqY*L~+Kl6X#x;1U5eW;@8aLHpC!<*E*z7t6WxoWBz-j0F2XUiP( zLPIy3t>n{r*cey(hk}%nycC6KCs#wHR2Z{R*50Q1x?O5_YL${&l_8r;+N!AjSP*Ut+IjP+rzNZ5^~#(7_(a(Le1L{nHwm1-I-^Sq zcfYuMkH-2B^`9|R&5VMmS%>-{bG8xX=gS1v>@=q*(@ju+SC{UWeCF402}c)qI8{+d zlE9+m8yBb{8;ZJ#zQYUGOJ2 zl&#@?)GxHAbMX1&0PwIf?`HI-T83jpPq=*!$0HYhn&&B~9t$EeI7*k@T_)c;f*#T9 z6FfM*^Q}4S#?3VW>G~X$w2>u@NgC?W%5>2|nq>}grOAB{#hsgY`xT!W$yQVC@^}Ga zDl`-=iM|odc0bHdrh=+HE*Foc11%5r^H475uqIs6!(JHM+GR~Y_9>gw%b^qT{5SpB zH~uugvvxn9iZNbI3tzWXMR!*wcIgS6Tca9E^980+XLY@gou+S@8jb{DdoVU$)I*E@ z0A&b@??`O!=L$N9eG*51nMB^sv~RW~I}i2u9B@3G6~fa{Hs)vACF z8DK5A2`E#~F~!~`l1rR%3Jj3d-{gHiR;KjD`(9I|gmbL=)zt8;Kk!eIi>+F-{vhkm zmgobg&#c)90`LF+N8UTRm|;a=(@Myg+Aoc+JuuvO0mD9aZ~CJSkpLGe4;3PLiRc(7 zY-?SUh;QtV?*hqw`QPozU>e1I zDk%x|s1@3poJNUlMp^RyD^^b0=3LPeq8%wo)pxIbG34g`SMRm4OVOx9rH4;0@*~UT z#`>=b_VP~Q7cW}g`?7}I{|=bPD&O=K4M;|QG+HbD_Y3arbawuU-I};|o44DQ!#(ZF z()zHeIp?gtp>i1>BtdroO|GlOrO8Qr4VR@~?(ui1Hh#69-t4x3SbrbYs4PnK5i z+Ai=OM7k#yq?Q&4CTiR1#Z6EOwS(=hmI~W?+c1|JM&KXc?lUb_y&*)Y^5P!Y!cdNK zB6rE9wQCH)e`f;Qp{6*0|=|A*N4IVqB9@**hkA?Q^Ho)j67A`DX~*%Hh&or$gx?Vu6jlP=>m@ zJGF4Uv{>OE>^`zTs=9o-p7c%N%g>LAv1^^;F5N$R!hAB`Jjdsiyio5m=XB@R9S6eX zZ)3uW5OWnOVe~hb1ud|(JtL}HGjd_1tHjZ#MO=#4;tjG94BMg zSLSRl-93O^iatBQ>e06rKn}#1rElUnz-xYJ>cRjr20R)eJit_L%ZEP7o}~}8 z^(BADg-s>hXc&+tC!y#k>!``J~XrPCcFcgbcAyiA+r_tenbS5iNr9TGY|6OMAE8aB3i=a{taMfK? zw4DSw3#^xT5!?*4#J9dz(}sSEpIFUkv2T9X375TjbMktXXG~f+y+gT>`8llpUqo8? zz0x+b(?6ZlD@}KW7u8z4UQ>Ccl8wZx^b_`_551@gu2q5iiy@2~x23Y^SZd{;tO*$W zfIFS*JGgYSzFTA`>fgRF0*g`uv)d6r{@dpBhjRr8#UbdsOMg9YVRe6A8xjj?0VYR% zu!QW&HXF8VCs}-%lp?95Y|ekPL{HjWY?ise9z}u?2>=sO_!9jDy};a=fOouJsNF5? zhpl@@y(%DT4m?-(8|A}t@R)ZMASF7I?=?!^&V_~~yob?{6ROCRNI!O&O^B7qroncT zp^lTY%Da@8@K}4>$NU4{@r~zoPWS8fCR`?Q64O@Dyd~geH39SkJ~E3EtF($&*&Oc< zWpS4vX1{PJgpriz5*wtkuMczB9-_1rD1!=nF7J-Bj$zmXCh~;)v6EsHg*{K&?|HS{ z$H4yNk83!atG$}eB;o$N(7`-btQdFHKSp@p|1AE%?8Q*LACCuFQo2Kx0U?9lq3aJ( zZnxF^l%&lA6yKQ2F_wRL@<`(|MVtiXFM$XzWhnQ}>-(2?YM}&a-Z&JBxLr4KZJH>6 z`XWu9oxv{_*}UcW63XPKBsX!K<4mj@1<9DBMWpKF+dWy%g+K7^-SH+hEO4r!Qb@V9 zb+jyyBjru*?}X}|YZ=26uTkKT(yQ0*c+gJ%7!v6Y)!-WVyZ+(=W;s{;HsBSt^%%sv z;5w9>Y>N{R(G)k&{~|oyRNR*U_E%V`6#2LJH+e;f0xi1vTaI2NOLGlqvx%^S`olwn zEMHm3ch#gf_c@(|mEm&L(%qpnjReSq`eJC(O+$Cln*nT{U@~JFcS~_Epo9p2>X*~q#0C8%owYGg zAvpH>Hoig@t&;6{KIHHfpF6Rcg}oH8P7Uvn^3X{QXBBCd<|R^@wLt4)UHxtLFB^Sk zkDH!RFSH;wex(22FiZX`=UD;PO#0e8`Zt3Ef3lTbOV*}dNay-^94 zbs(N_`Ph_EM6NC3+6Xn{S!jtO#1ko|sC6=&(!w3`+*wW#RWQ1!UsBj%S1kB@%=fEn z+`#Nwq8(*e+k0g5uJIrlBVGnms!_;2ke5>)6w%Myc{6k-JlT&(O-H&*kV|3dQ=WZ;nO?b} z%fxnswUbmDoa)!4ox!kU0`I!rxI}a5jm05eV_nxj;T>-TYVApxcdDXN_`k`%*}c0d zh%$4s=7?^!^M0M!#*iz!hs$IN|=;mk`nC0{l6w1 zyNhBqTG3URu~tbOiI*#u;UBFoqDy-cyKQ3x_P+xkXP-#t-A3I;m*5jZnqacMs#~0F z*$7oD6gt7>QF%f^Nqq%xYsa%NcaD#Y|K5%h1!!2ekFwy&&J7bUhZ>ANR@Wr>xbRb| z0}4u{0s?sQ*(|F01BWh{#^v>>r1`>0`Ox&8f3+FkUfk`+9udW$(D3@}vas$Wewi~5 z5h&mRdA`Z-#JNXzlfz|mta!qYjh$3G?NKU@VYhKMf2K!cZIcmi7H@Cuy1TIe3<=(x zx<6!gi~6Z|1L2Ry zBb9;Xk5|c1MtCDR%T=EWy(Ba4G_eN4OCTQIrwii(qI(F^@TX53z`7W^Gv3+pBXOWSir@WsC$T@25db3ed>qZw>j(q?h-xR zur|H6qw}Ex? zpAYN7E}M9_uBD*k8llxMxLsvhU<$jlx%QlcVRAbRw#AD*zH*7yRnC=vUVFUKl9`|h z6halHlYwk+1(Jzr7z^!2V+I$2SxC}9Cc*x4GZ<-3K!4%7C{#sVUWXCUcIa!9m{=`y zWLm+R*`FJ27`K80Gy}z-$prhPCM2r2yRtPtv+WWsc|skM``rE*(c;IVcegLQM&&H0 zIBj6DeeCyH4Q*FNem{(Q{`DyrVeI+%r2yPXBx@EGK63D23!XghKq5RDn!EGUFn8wX zt>27&1&6S%!kRAkh?@4Gn<~>wD7$d;0ASjK5{PG+Z5X9 zqD!c0{LFCg$E)2M3qQ^GSwlm|^QZxe6LS<6v7`~l?fMXKD~^=T3YEXj^+(IuylKz} zyJ)ZB8u{%@GsC1e!^pJ_lP`*(vnIabyYl%5Xt7?`K3FGqHL^Oc865c@_bEuemF&+8 zy`3};KCvCrpa23d!plg-U6$GsPPD$W@%k zeLk%T2q{eePabF{eBd#3Cn-Mp+z&enofl@7JH^v!zw1{3Dm1Yte#R?jcJHJ-E(PL! zoFZuN?XSW@0Haj_DN|4{?D;Li=!Oqge?Ka90%5BLvPm|O-0A$?KJY|f1SpA!u7)gJ z`wF#8M#GaI#164ywNqtds%=vXdR|3jkQf@TvSR)FUX~j(ZO-ove*zw!6rtD|NOU06 zgGJL+y6wu;e{}8@0UkuMd-%#lIFiU51tf&{TjCImlwx5Zx&HWsmGqMhz;ziI(gZZ# zgW}1rhwAch`AEjpf{+iuu#mYOp<1jB^Y{seA0E95coIXbQw#b~HN&c%0BssVWp*s@ z4ELz~R!LgH4vUbpOVPg+7fNy#v^gA~{9H%cq5dLF#9v0A!%|#USzWqs*Gs*G-dGcm zTIGb^jIv4I`_CS*cV;+n4{CqlbhO_J>YwcVW z-cgxuL7ZgiOR$eVh3&tmiajX`{I-4-yno;O}J|ER2}t;0XVOA#ln#2$$PQnO@eSvHhV-_cNu#B%|=@BWyKbap#?pN@ChZAluAU!3=xuG>q+mYY@c zBm-2q=c(b-cxjXtcEet)u;}Xu!Dr@QYMa{%7RGE|GB0zDQM)SmXzDU;dG2T70z6Js z9+;~pJ*R!1&!xy4mp;>RVZ0PrN>2D(=z+%KnGX6*_ih(fY_36=`3d47v&b*_g#R1o z;$z1oul>73{9v$!=RS+eX2IWdnXQn>gm8D@UXp^l%B1huC7SI{D3F48L@;|h6GM(e zrQ1%$)a6M~Up6W2z2J(S&P6_K!+HYU7@_)lZDK!6?9C(j+_Zv6vNa_e_cmdjD+cpv zM(m`sYR$*aSbgQ$-~hlw{JG&NELwqmBZ@rrlMm3}zw`d)H9z;)o$kanl+Y6p+K44OE#N%es+ih3RgU5sO?K7R zPiAjYS)9bY8GkF|=DWBwK?qIu{rySW*TAF?&k=D4MA6#}@t8Uwma^W$C;5M_5Uql0 zfN(eJmcf+D;fsL~CV;%x!i*e0lN$VXEfcSVb4p?gS!)ynr*tmhzlODv9 ziW#@0#~Pr2ir-2$$^kN1iulc;I91e>6U6;H6kI?k7psvSs+#WvF}*-$gms~@R#O!8 z1hG4env2Uzc>pvFB#sas0LRy#Z!+(6lJwz5JIg5pEvu^}9P#1NdLV#57~nHMUwV&4 z+{ctIo!R*zFuO3;xM@tOlehl{f^-~KwC-TT_}W@(-N~gphEVN(M|uTb^=x^JkTpWw z+0th)rFfeIB?vRSYzES8@|e96xqw#@hc#4FNIu**dUda5%|7sjkE*H2>HlX(GE=MLm*n6(km5F81Ay@0HBfaWe9fw{hR~LOd3L$ z&l={PTkj47NuLfx$)@2XP0NbesH<`iQ_4Z)Qrx#o%HLZ4KDOF`KEp9El{5R>UoQPf z*!;~Bo&&8LelkYHH&B;00R%!A<-J9qz`j8C^#u$vF#cqUP+d3aM5Z5&suVXfDI~Y^ z53r^Pywe&S!IU)L01S@1KZyWwkjOzC?jY^EsDGmmCKbdj{ zOZXd^R`C79qJ$Rk+me9}IMa~n?Z@88V9jzOS^m*Yi%Kgf1FUP_#o~CR00GJu>fmy& zMd~7oMS%NunMIRW_N1 zIkcs~C({aYE0^0QpXYo?{zI`y06ia1j>mzI#clU}&6{d&{u%K#SM&bC+`^xL@Mi4v zojf%nA`Afjl1uXQ_y_@B19ntmYC>UH7eOi)7guoUU-i$JQtLbT5ZMsDNx0!;iQJh@ z-}ac}sNbs~2kt59Rr%7%7>G2S71?zqGZyBlt^Pww4=Hhl>4Lf}n71^YLpFk-hOXP8 zpYOd5(%z<{GN)Slk?lV=&!mnHsf9S!!h%c%?8p8CQ@=^^3tXxd(yoIAExXWEjA!?| z6kk#iF)?mm1@R&MoVI6$n!hqtUT~O1_v_ScJ+UDl4PypY7F}X0q=3(!{27Jz>e_I+ zIbZA-XbQrL$>{|MX2eERGv9t2psg;0!uq{3wNhGf^+%S9P-DgHRALap+WL_2xG2I< z2o!W~XnIo?07c&MSZ9Z02Ab^iXVxv2EWH)NsVlBcBoL zzAl$tWa-d|WRBnlEj9`l(tOrH(~FzaXk^3=*WZA5DWS&gJRdppnE%Y2q^QM6*hu_D zH+_w#@#b355vtRSBB)cHvU#Y2=rhx`dSbHb*V?{g*}agns}V6#iC$mdB~r-S_x^I` zo>IA$MLBHO{a zB~1nMM!xZquXv_0Nf^ri9w$UqeU(7(SkiTcCd<7C`$|{6P2qzO%L_Jcn+i`dEd6k0 z`0P$$cFmt4=z*A`Ydx5+m`A3CV={*KR_5T@x?e|nOkQk;r0#y~uo(Kpw})KJwh1^E z0%>v=`4tA9ih2e%=coFD_%4M!hbN|%<$!-Tz`zEIhj)Ch_}puD(!Vz9ak^tA+pqI< zX(c^0YXuD#fqu9Brb`=eI^+=+MQQGjBrd|HuYA+Z%keELvRra5k|s~ArT41WIjuGo z!O6BLp6SOc5-@IPv*E*A_*bq1pZ(uYafaVfcSLMmA2NO~{>L?|^WmkNRmC@z#Jt%2 z_9Re`irVMT;ciw)oCIP|oNS&NZ+G{Q(M-%AV_mpXv^33LJ_E99;MsRUt~v1p=5J7V z$)A)^bUJij9|Xw{D6ObEhid1p2-H&8oO;v-B@XC(^eGG+;@ZMz5C@AxwW(Io^Yt4w zkiVJs%9DPteA4RIh0xwTO);l<)Pa)-y6x6=X)M%YRElug)BKWAPs4JgCM+ ze=BM$dU@T>8@kTVR+75irTEFoV6h0A7{;Zgwg}>G{n;lh zCPVd2DB&qg0rZ>FK=J*3-@v%lcdBM{f}I3cUpVhNZk_$2rN=aqUGnW+g%8^|0-;07 zw^s!+nxkU$$73#p{!@uL*$kQZZ}-uJqVxt*519*IeC<#5-~-CuW$S7#Zu>x(Dqm%g zTF*ce^l&c!86O7reEVpai)tfiFmV6NE4>m)G^SVBZ9d$q>)WRfQ6N!dgqRSG!iT7g z3vQ}Eq~!nXP6`^AAE@FYo`s28wz_Gh*%G@dxxy5t6yIAc_i%+@m=@9W!c@@;y`Y*F z8gDQ=&yo|ZR$S#~OEkrOaz~E<6)KFy;uO7*QDKXJPSHG+|k=i$56!wuj{^4~kcXW)cbf6l3_s85cR z^MZ~X+uvJ)(m#kqq|FYGDhH#&y;YxI>vT?230sNdfn$HDgMx>ofF!9|YF&{kT z8{8et4$3HC@BgaP6fJFx;2%pM@hJmeFeB3kZ%v;FaH_r`&Dd5jLDU?XH8**6olYt! zQox@K^cfuH2hR#8{nRiD2Xo49M;G5Ko?aIv7$=?;w%ph&v#i&9^zZN@7^kG9o4*x)FEaQ-(5XJ+bDdH-RWB;HanyTH#U~N<`PtO!R7*{ihYDb@8sVptA%yYR2<QYK<4_!(#c+S6?ZJ${JFTrnlaFo8sswrZ2)uE0djh zgWgoLPM^|anYqIxFtA7N_Yred13GNZE6zLz!`e2%P{`@Ul1+aKWK?FptN`@*{U;a? zR{Pj}lM7H2el@7HB9!}XmGt%kX7*d3JSBVH=k!?l8c92a57VN>1a&!8;|hus_VryU z@!yx)4krj&vl=Dj1&k_^@-06WM5FQ(xUtyX9mq`uDgHU~ z@|>(HjT!MNhnCA!6kc7o1(uGAcn286-Go3GN-yy-5B}29*>8}y(=`W-e4(PSW4!i{75>8US(f_hh6G-0Sqfo_lo&)@8B1H(T z9HTpO+rVydF!I_u33nqdq#2F)k(*veR*QteUBAF!B_QI_sI(tae?MsbNyED_cru31 zAG7}k1&R{BMv!aK)h+P5;xs1>Wd@BJ^t|763`o?@>N%U~2yoLSQ@pFkkkTVPSX=BZ zD|Bu%#9a~4Xn(|+!&PY&7A%VVs|vj1?UkY@N&M8Bb4y>1P8sZ)16i%11Cn{eDwo2# zVIk-`odv;3sm4pGMvBT`y)U!*s15}_{^$eM2w|^Ed$MvjUH_O}{J$%3d99RClXe~c z0hn6H0+4z21D*xJNb@)ub7c65lLS^2-6qKnK~-6~S~2!s)ai2afY$^f213ETAUVh; zn|&tU5l?GYaVYu~@|U0Hp5jJGgqQrV@%HUev3Pjz}yUaX^Z zW&&BuobunJ~pQ+a!8abWS`W#)D;go|W;Hk5rHJd~7ev<;elk#8SkZi-?J)!5EI@d1=ZwX0 zkmk7P$}vjO{|T9wV>A!hwe;c;9lm~;lj=I~Z(iO|uz{S%eI6`_Xs8p9dG-k%?Kam7 zdJqbv7d%^%)rrQ*6#32Tm*MI94I6n`d^PxozUc} zM~X&N)V3AGRis)yNE|VMwV{T*9+m7BMP6>3^0WM}7_X+vBL+@S-gUS5N3)>74ITS~ zXoUlFg4`xh88++)?)A2}a;{14-VMMMCX#LPQ%fM}eh$}|yvPrwP_XA0tV+yl(4z4( zo|hDtbRV|fBHA~6koveUrr+~~Wf_z9C@xAo;TisRv8T41q*j%b2b4>p=x;loKI zAbNLA~Vd3?VgI7m|UozxU=6jz}#5Elj`>>Qo4i{d59~+J*xR_ z?Adj{u-G}(tILh&PT~-`L4P0Sf^(n#(4JTtBMNk@)E@_edkfelp(3E8g{sA@OfD-a zQ}i*(p^@A2T-3MG>BP_`EQxo&` z98&mDchLrY0OVdJbDMKjfcKdKfdBXOJk<_lD{2PLvoAWdgi-1_Q$09WN3l}XS{l>O zg9f^kfwT33GZ?}Vl=~83(*Z>rKcK&BPNnT#Oz*_I3Di4I#ivUqAohwb)EIyYl{NJB zqg7?@(}B2^cQ|I!Sx_qYV=2G}R<-5C0^uEI&l#uzB~(rV&_k$>JKGwWfEL@d;cN)i z*0-)L=5mw)Y?R;fL5r<#uhMklNw~k#M1efrk<`Cz$fv$_(&DFL?Easmxqy0ssqb+P zMomD7I0KO6wsJdV>`j^9tB(@vdC_tTAG+^Q2Z24!^(iHM zk(1#C^$+%M0mHq(xWa+#ohc9EBwmO5aVml|FLUnj1CY>)_fGvr{N5G9Pj!w+MD&92 z_ARLI}_V%_Q>p~fKUEOj)oZ=b3^=Dv-r9FsJa-0mO1<}T!`w2B&&2vt5QV|FE! zjl{HyUMtz8ZPcwYs_{@;5WFTlE36Xp)`xen3l`2V4N^*h1V4G0bZA*OSI_vp>sse} z8fI9yr}GMwxyFl3==tEzS1vmFCsgH2*MOrXIldd$sggzhnMe;ow+0|nc>HbG6`MAo zk4Qt-sAp;`tD4e30SZ}z?z)@8psBay)h~Z5C16wrpU17P3bKqKR1P>!irh%YB3%;s z3*11PmyzO9wpgAmq568)V;HpfMlkZ(C$6KqP$UWKj0}+j3jALcAt)wiKUfXubm^-a zpTelNK~J4XIQwQ@{&WmI0}?O0e8Z2*vzP_9bQ@&Lw(>7AJvI=2YgzDrD`BbE_|*rK z71jb>dQmV6A%%Qh}0He zlxlb)b|1r~J=>2;LU~>DVtG-<3(o>JfzYz+6;uOo<($dOiH%8QE~L1JQ}x%wLKuk& z{8G1Ktz8>g$lux#$V}oc_kGoBpUtgq2Ii!|1FW9eTw^8gn)mjWDA(5m|u0pQ*jJ`=(ZLujUXRj=E!I|^F9_^G9we#aKX zI04-7C9A-zXUU*!$i*k zxz*D5t7=cXRiK@TWqvijS@9Djw(&2wJp1$oK`DghaT#xf*on#>mybTG$8w$-10>`Xymacfjh4 zYI4gwL_UNMX{*G4En~cal@zqx1CZl13pGETPc3A$nJT^^9%wUi&~L<;1kQx(k3Zw= zZAM4m>J3Xgx%M(vy!sTTX~xRScWYJP>HgYG+DGsHvA#KA(Y|YUI+WAA>Rl;d%^d9) zs4?jNultuBUkI<%!tQ=$UPHF2)@uv!hp!Y-X&pr$)h^Th>*s5q|FAjpMcM3}Zr*7Z^!CU2~WETAdxDA#?p~c@eTGc?$75xl%!ml@C z!n1Vf114CJ(OGu}FJ=WjUh3|PvrV-x0wxWTBujT**KW&h;0DoVL7rpD3cBS|7_S=E zUx|0rq3DEe?U51M&8q;=$0j*KM}QkmL7iN*0buGs?5POnXCB0TiV*ZGJkeOJlhllJM7X2-wq=k zN6Ft4P$b*eieqVJ;v_uIDySEfq}^j_HBX)d)NMvxgpg=S`-+ih7nrqpVeBlZk{8nt zYo)&B&Fd-%vko=@Y`MZj&47XU6*MnZY{1(Bm$6*P9rPoKYI{LwF;)8H54o)#g@eA# zu~U5e15Z*2{^B2k|D_JkD*8VmfdHM1m+898W&^0N6qRa6GoDm@NRFV&lB++%Nk8J5 zk3{!?sC!A%t0c=FlO?W75H(F5w)2@a2abp);^!X)I;#@UcH8>lWepl?68Wxj)Y!}|HfwJ1~ryjNxsJR0)o$d>;CbcZ(i#UYow+8rH zXd}gAXZ_@@&b-c>QppfO4yU=nF0b300GKJ^^DmeaQlG&TtU+-jqAmxlS@}#xW6ARR z3YWCr*i74bb(FnJ4eJ{M(Z@=O9cVUKv*!T(!4Y9TMF@*>rnehhm9dMF(*(Jfx^h^B zKo1eJn|13}!6NqBtRisAHsD@Sy|J_{MOLcAw{oPCtLh*#0{SU)Lce5ton*tnD>r!8`H+-vDZ|YCpuSPnw;>Sdm58^JjCntkmPBGRpfS_#WzX*iab1mxc2OuZS>#zB#_7-d|I2m3> z?FDS2XnPTf7CXydW_Jca;azK>98x=G!kSg&m;HRUl^g$YQU!YPc~ZOIvs1k6fAJ__ z@}s;tf{3?Y;(F2lH+0ve5oF=FSd`azbnE(Wdy|3I0bZoGEGvIH#pThq^d&m<@|Nya z;{pnZ5@GdEk8KJwaH9CY^#YzR*1&kVKJmm}-XY(sJnSG~6!fm|07&Z?leMdOZR+}nDEOIO^s zmSMhMw!UWW*O#$_>{P#fknf#q49A2^@s4(-h*11Li^1>xjqVzH-uc(r&LSxMSHynx2j44c?(yW7add0_7I5zmXDvnP zZ!t205SyvPovIKMzzAZqIm5X!D5ewvhg;^V+uD1Tm7{Sq=N^*F5-a)+V?s#s{u zMkM%(~ZwNdUF;8U+1%HT@ z91k2><$6*5TYqQSG7wFE#y&g2!+PHNU4higP|W1}B10M}>PhWlFx~C*|IQa~>kPy= z6kDc{KKlxLk_fut@%50eu{WZf1xJIMKOOIfih0u9KR;>Mzum7lpmS8_or-t#(U4f9 zRtD(5*t*bCC=?)`2az;+x*>s!$&RW)9Pv5XVhyCZSLqIdw zjg&xE_~-wSPvAd`)uAp}^+MqXoJvedg3t;pZ`ZTvbJ%`rk~%r!WI_$$mlxoNFI>j1 zH3-l4Kp#{um(N%_3oa3UFAK={5FxzlOyxa`BbR8s+^MUVm{TJNZ4VbvS@x7HggfAe zNqEPU`2nFKyko#(*?$(!HT@%TLr~_8WC{15>b00=Km8VxmI&n#IcV5Q@o0da`#mEH zU|&Nk%?2%ngtzHfC9fc@pkDQd7FR@b`(G;K+&xv z*oqt}^~P3}XF>AuW1}{W6qOd96#@%as)RkK()_NZj4_BpAN(m&C_brH%?4T=?fa^3_DV-h zJ&w~hO*91D>-uSlgQQ;5s%pEB@V{(gUl8PtN3^NLSg(P}y>#D)nz;;^e9HEOX{WOh zCnR2SyicH$eo1<+;HTBjRKBhrA|228VM^d{AW)H+*oK$-?1be2i5Oh%6`~YbgGWzg5XS~>gxlh0r=Bm zN!v|F+gwZ&g175Zh7MqseKhu-V?i(y@3;?|>a@J8{2-63(x9ct;i5y7=)%m`+QB6W z(gJR6%BI=-ed|?$ylJKHp6KA?_Jo>k|D2nYKK`=4$Z;itQrj9iW6{*`cJ7 z=(OFjP=3Hf`46GsJjV134>1nERbGTw1PY(i!e*wN>u?bEchT zrirUhn%n>0BK}+c_9f^&y(2hf2JRRU0Hr<$13_ zra=c?_Z}KM>(Hpx#^N2{R~dQox|RW~ysqiG`z<5{DH0$oU!Vbp?q)?>t<8Vxj2!yv zSDL=x82-43m}N<}kpup7@#LK1{RC`4J>>+fV_xy9u( zKP9OmFPIX>tG_4G{BZR6x|&d1Z1I3^An*~*V;TArX3qO2046!UJQlaCie#`om_*D< z{r&Y(nZ8#MnASEJuo4<{1=?4i!G+BCU~Q2fT^WyOTBk1CcU1D3YCsiW#FVCQY7;k4 zyd>T4Cw4t7tILVqwwRc$YIIZ|DKIgAvDnGTWSV)+3JhA%NLzw$E~iLBaOk zjSuQr&F0BucK7IULtHpu8&8__l#(l|1izY4BLm$aHiI-b#0hsB( z36TCgccy(d_}pG=g5@?|?k88|YQ*ha0POyE>%r*qyMwT^Z~CLK7yw(?U>J?uQ`6i` zx?SmBqezL{xS9Ajq(#5Z20Y2us{KtD=QwYM2wnd6VeKFzU~_~55TQ|N zqs^!eeb(846WB=>C(#+YyC)j34Kb1v*dGGK09?1`I!faz#QlFZLhd?%bp)yAI5c6m3C zt9*t|I0#rx@hDR#UKkd2Ubz;`rf>hI27+^Z7-b6gk|0g_{!z(Ig=3w$^III3FkQaCyJ53s52)_@9l4TV0Zcd*n0DDsNVm7 zoa}o-$Q~goB>OUisAP#kvV|0tHESG&M3y!o%aDCn6k&{vCF@wS6lN?j_Q7Cg%*;98 zdA&b>{Qmj z$wtlMf!xIQ-4W_cx7wBa(1Il)Gq*Rur%Tnz?1Y>eb5kV)OkM5-n8ft( zN7BOfB>{IRTdzv4Oevef$5w5z6zob;kbOS>BXwch%3E#;Uf^a{P)qZv7U|UI8g&a} z^8F|Oaz+ZYJ=*A3y+zF3q3@%(YOnw*kKZJq;Dp=RF}(WRe`8Kv<_p}7F050 zS{!Uhq?+pUo5ws&5hwf?l}3npUzGRG{P00;Plm-x1;i2meSG)CsV~;_!r1>O-U8l zUxM9Dw&5gL`|>HI*uTZ8Wc(W$qY4!8q4JTuVxiiXm+G>Sb-~l?D%Fhd?Me@*-Oo2k z3QqU3o}7B&{S9Py5Bt%a?s_rRJfk;%5iZibK8>n65tQF?LG3f_t8CgE^zA2daR$B10jVXjVnarmRJ>27cnk!P4RuhJmYTlfyz zKZGq<8>o<<x}S_RpW(44 zpFy{G6+P}BU+n{%0fb*om(%B)i!(R)uSq$Rb~#tj0^H~3J6-Q7R9*1yPAj_KAJ02HUqxhQ)4wcx0!%){?XZpA2p$3$^wFS2ofO&rG98 zKI7VdQ$-+gr%TRrzN1nT5iVK6$UyEeRsBaJmL!~pHNrIWfVL`pTK}$R7v7WVX?EhL zWDg9uP-dRAYu%c3reP88;`Xcm4!~^v@Y|k)3*U&qhrLgzkH2HAr7FK>a}0n{9)&y1 z`ADyMcxlk@Bq-qN-?6MmonVx|?nuf``U%ss)ZjeH!XGjFQ|V(LQPi0ye)kzSky8kzBwAhJ6A_2Cfg%V#~LEq%7tq~1N@?K;>I9Kt|U zA}9kPQpjAA9l*50%e^x#Uro~r^&2fIe9GWs@A;@>H#cE~B{AmWX<#Mp7$Yq)N_qNy zA@Vke5=VKRx%5ubL`SIr{96VEU%T|?ZKxAnF}eY z@qqIiPA(V?&dW$6e!OV_PXiVmwd6>>u`|}(-s}tBEAVGGdRr%@&TnK4!`7l+)t-}m z`R;<{y{zP8Oy?a*7m2{|U-uwxVgih|n$FvG8dHT#gBD=pD)Bb?CNdzRa?QF8ze=>lZI1%i0V15c}<5vY_P(bP1^$*@8 zKGbKjLSt)YY$DC4|GO^(9|isF_Tv+c;DeiInFe8h=3XCfoODwuDvR4mUx#vW_biWK z2@4Twts>Wk-O2^BGl0FqN)6tNP-qWq37NT`2Rya@wFt{qqQT?nz*r28Khiw!_>KbO zk>9y5r#9Ot`t7^B6KvSc(aYrYM*7E)u+MsE$F=5%apYmZErgPGLZdpal z>!Kv{d(E$5S~?<%yI}|Bs&YVbG7+&adI;C~kzw_veG?Y^A+#iX*lpE_WKIMQ$0SaG zTlX^l85-@I!*?>&=n~N;c6(saK?wTi>(3PrS8JBkX`n91=VUql z7TK=o38TWQdJ0er32k)4=9t#g*A>RF9c<978&G^OMUZvB2zhzK>L6E+r}keUS4>ry zPLeUsTC{m35s134Wcd?DceV2(+rB~^(wa6$MJ3F~-Q&nF{C{&zAPp}7SMcHB3gVul7kUoZa=3jQj1nNFNa`}aAZVmDz|8zLYc;7zxKGaW}o_!@TK zcN0Lf{G4=Jq4-s;S9Nc2ZZqN*dwc$V<8iWNF24_KEoUrnD${Rs*~_U3Nw+@P?G7Jp zq;G?^S`rYI`gZkYGt_WFL_W`T7RnCD#i-dG04-W@l)>o*>|cI!zO;WTji1L~dp0Y; zALdA+qTMUbDT3-UspnV9|is?kjV(Z~au8<&xrto=I~_E{kk;sS~rCOxWoGXU*ZWk>irl|@Yt_ZmX4m~J=oua?1 zGV>>*ui#Wj>**KC=vh9fL!@Be+yOe)Nu5E23{-RdhRp5Fu>kb*D zPXQ|uPPRudx)n!3g+kqciePO&Be@Cr>%}6{LGZrjk?bLS0kjs>2T8EER^g>z!IrO< z+CAzZKbVdbc~Y6@2<$&H4iD;w=&YUbT*wvN3UH%o)F>XwR`wht0Dmn6B4HFh2R7cB zfa^PGAo^fR2M38~vILJng;Vt(7qRQ)%tNfRvD74ps(OIuhZe*H-uwaNuY>lkn5aw8 z$Cy`BfIloci{Ka*FWn|8eID+=#+q>pL|IFgiFtj6cItwIvyiISD$GTol)Gy4 z&yQeFU7KV+5ri?krEID;;7a3ehv1~tw_mK#Ds||Jl>u~u_H73c1RuV4rZ0r-@XA;G zeh&;m#N7b|HD5#Jk)XHxc`e>!VyVVW~*ghZU zFzKdM&Q7q)FY2QX@M=vJpMn@*L2Er!cv^z{cOQwS9eB<3%<)f<-KXP?By?j2Gv6Fk zh~9Z#wwhDCm2Z? zeV5~3sZG46(1_k01p6HtdwphKm;=`kBhLN$jrhOH80tPao+%C*-u2UdwQHc2F`1Q*Z4Brmw)pr-H%DUq(BI zEJ0KYM0nlYRhDud?*a0;xC;a19O4n*xZC!7pX#5t>>;+)`_qgu0mqK`!X7`^qY{6E zZ?vaAqx#dg?qHNk52Z?l4yau60>zi0UPe{dC#tfShQBqk8Tg+q@@*MZr*LmB17`>- ze}h(8A3_VMR@@%c27ez)DH{^+yy7JqDznH?=h^1;-9+gMJn%nFb=_yOs)kBrcidY$LtYArocmJ2V@FLS z&=ZA0#v&v7to8-hWw%LQ9wi&&W&hJjQUarHFpgl|B}lH%tsd2#Lr6?d=<0@HUbIip zj5UNqqdlba6L(r5CW14J%%!j_NR9bDG*U3eyv%b5ujmBT*~jSV_Abk*j3CVZW|ucI zlIXyTc9z-6qGUSRHO1JlN7K`0#qIX*va#zcZiP}P{calk^hEbY>D@Bt(x9faYp0eq z>vHwH4&EcL)_F**K&C%wjsH~I4)thii>_yL@4Bwly*}mnSsiEXy#V7AwxK>mq+iad z5!tEy1UQ}ekv~OZTOGXQnGGa#{2N~Xdv$`XRj*du>kO4po-?6$3UVAh7XZG{0_>>7 zRUZp5oWGgO`jBd}5ir!g^*+)gaHW3W<=fn0I^0-*+R;7fP{7vTmy59R=PE}kdivC~ zHIUEd`9ks_lC#4r^7SzM>j}^n+fQD4-8x#ci-Ic7bg5+nqddtz&8Mlj<8r$^|KXq2 zj-C9`hEhBrVqO7S-wR}$9d{3cdlbp*tUzs+@a#Y-P<=tM2NuO)zI%S0 zP{$v#&$bU=pax1yp@deT!IH0=PNf~MpnRbV?Gh;Fm>b?N&R!>FnSU=ueKwEzqSZl(gFlt4kA~D(&bIN$V=8wUmNR!Z=a;o`M4z@ zx|@pgG@D;gP@DtOJ5FNxdSHn!87ZXor{FPtzARV<2Q*Se*HEr33~5uY9b~(SYVCzB z=u#H@7E;UB=@~jI|G>@dZvmlqjbn_Qqems-B|b0ZQDEL;5d`5T?GdjmmUnu*4_jY0fWy3D%MNwc*1mXN-(@!KkLSgmJ{#0CS_n%ed>?g>wce;3!y3@!4q6-6 zs=X71HW~H#UoBA+-0qJi>Hr>dIsbj>IaN#SVxIemte>IHq_d-E9^G=*noMsv{WkVE z5eP-M`G70~cUPcIwfe&tG8gxz@5H6Rzn^`t|MdZN?@0)B0M^Ut=~ifL$b_-Y@V?v6 z+G6ynu@aUW(c4#p2e>=WgJjtLf=)kxf9+y_lQarwJYj#2bCVK&M+sWAk!gzX;;9jz zD6iR*TA{Mk04mXZ;BzwqoafxmI?}MXCK2bBh%Bwav7?c?zb3~+{dmUhzmfxp>%N&^ z!oDy`{*CkkCqe|b3>IN`gSF%2@&OmwH@^i-wE*LYyZ0BM4+1v1q&i)`ALZ+xr5;l% zO&HOmFYUmeao<3w`G|htt8IZTH^Nw#$NpL!mo zq3!~j?kWuwoWHI>BEQTFvT$x0En|{hyOJM1uec!&FCPFW_8K6}<)ivRq+yYP z#a@|aU+c&8U9>gpf>b3c0-U%n`S@}tV3;LFb9kTKcF!yy$h$Q1?l?ah$tvZ)=T3#E zfa!biFN=3H>2d%KA0zo>24y;{rR-XN<+1w%gg@j{c=-z2iCto(5igQp6eSUz-$*w( z(=IS7p`ZV+2ZR{Kt%`U;2tK~E0{cSOp!W4Ur)p%$Ib{zddSIDpG^lCGazaaSw>CFG^dN^LcRh3(m9<6QO|okv}Rj@@xhPf z!6eP|Et9q@AVpR18OHggH(oTD2 z!+#3MX!E=ga!%rD{@$ZHg84m<`>V)z>xwT)1|*!5qtd>%<2y21V+|sIB*TK)x5`2E zfEc#S*K!{Zw~oyRcnZQ&Xo;D*_S|2OQh@eHTV0sc>&Qz&VBeDK;RIK>DkRkd8-RO( z*5Ji{!2kJ|g7>HEg22m53mN5tc1Lv?8I(IaEsIcbZkbpyq3ruu*i{JU_u}7GNIXYDF3Ld*SHa9$0!LLD*N|c>gc#{-xYgnLdQO z39s(o9xB??qU{5BMs_i;4)7suvKa?>6>xqCn_KiL_7wApf#eT8DDWEy46IOTDbNPw)!WytB ze5YJs1KD1fz#9nRwj5Wf37R(+=)q6^_7G2RWQ+Wi7YbIO<1e;j?okw0rr9rl+Qs3F z){vj+cACqs@C*$Of2eg2LO2NbdL3Y+Ksz(7!%-yM*SnE2e9XbO;ydePs>=9rcL;&W z2`Yldj)?i4(3js5J__^KNNhF2L=H8kT{!IsdS;StqeOu$6(vv$y=oa0s2no9>PLBB!RVo2X2WeeJ_i&lNy?a|I!Ec(gBe zZ4DwDiG@-5_T?N-1*Tf5Ex^oOq~jgm(GuPMYz_cvbUIRTj@U)Z>-L0$PY?QK{(Bal zqHEDBLIgOmtNzaik*s-zPiuo9oE8j37kn z$V&wlPudB5EOabZLlvK1oF1tR>V-@>Kynk>>sT*bZzBLLN9sd2rshz(lu zt)TTVhnuqk&j_XXgQ&n@1S|kBYWLG?F}^;H_l|gRZZ$`XC+9Hd+kd6g_b|k_KQivk{dZnc_;p9rUa89+{9W#qayA|OwM+G0yRC6*i)njNe|X!hOY|v- zXLpl$4R@o~8b5qLa~RbZ{GQh^KWy-q_3}A_%gl!@X#xwj?V-reN(>Crv(YZVMEnbh zrMV^xo|fx-Hg%Xm`E};MU3e1-o!Ak&MU>F-f1oOnut~hwdeQiCpL3fY*C)Jf_4nr~ zk+3c#mW#4Av#@7}NBarVH~3R^mX7~h!plvIy`wD$`RbtvlQ`k0U+2>ja1t=mZdAlw zRw?=(h9nJ!ZEeQ-u|BW&j)ta>G%^}&B3^r&7t1;B6!z!TM(&pLJUMLD`!=4$vl$i^ z{^P&fcEdrD!27@I>tA9)q&~5|fv{-v6K6HBs1CR#%R<|luh4UA#Iva$`!X)E&=n%}Mddy27ZU4w|Ifd`RnElO9_8GyL5W*jO68-3?seDJc zy_fBTelcyE+mtYPYp>MrSlQ!|h}Vg^BhS*$zqMG2gqk$$SLh-4EZVL)dYz^vM>nR{ zs_&+UO*f+1{cW;UBH>qP-g%0)cd{jNp$V{E2UzZUpoE}grB_jG3}lpm!$IwLq7L+S1xj$dg_EB`d% z5A5ia5c%fAnzq^rY1|0H#;ARMFaqmM4tGN|&_kf9NGzkl(;fkQ)-f>F_5V@|0QkN*ZuC1Qt(P3>c8gVondLZH8r9_O_Ddh602gAOdPVF-t3>zXUji_oJ7@qr!AMQA_a{LP>g88vQ z%w8!ZN{ti8@ksC$u_fXpJe#hsWY-dFEIg7`DRX38Yckg5kvoFm6O?R=@K3*7qAt%4 z#dmbus}DQx+aU1uC2bBrR8(IeO;#id12VLvpG|SPoD089b=90vjo(G91hdcVv@KQL z4&t>%NPXNhF8rB{_eD%PQe(J;)cv>-1=MYE_meo`tv~n#1}7sV%`!%XSG2la=%(WD z6ko74$-KX@p#*mWTlsnKWX;EHbeq6Jgb^}x&TV&l-EjO+Y-7a<&0F{HVN-S4munP6 zF89IUgK?tj6-@g(9OEwf#+UPU&m8i@;$LpGnH`O72)o}RmB~*~Sao+Ig%3MuqM9z z_z(cUKcF10S|U#Qjrhuv;npCVEz|grFG;->`m8-LblX0pDa(IeG)C7HPY|sC^-@jA ze>b7qh2v38ZZ%0Vado4t9z(9z^50G4-a~nsEDi3dc;?3Kqz}wJegFIdck1_{@2rVc zv!sxAYnur)@AA6R*NjT|tGZGi!TpC5&_Dz0+&$Qmq zz2?i+s`J!j_B#AQ!gtGH+W*yScb(;YNOaqLti_{Lw0Cl?wHXS2`(MfSZl5#t|EFGi z?7*-pG0OgewWIy{?l2es5C1yv(t)W{pT-b_{QnBHEwgWv*zm&PWBDz_%tZ^YJMsT+ z%zQB@Bb6mJ-eZRxq?GsJ>u@oQ+e@f!Rr*8bR z@ewTDVzibl*`beVal}9SEEIlq3SIhs?Sx`>oDfz4$ZNepj|Y zCd~;>5%0oOV?&$7*Yng#wPuyOFIpf^v*Se)yA%nITv_om5#wbX=BbB!E<&EV`Y#WL z{rN4k9L){@NtvuSd;CP8lB?x#f+DS$M%%swz6GrZ-m$Jn7gJ-b4gHNz;3OK4I$5{E zrbd$jdSG*!eJ^IA3ahfIYo#+*{M0z@zfr}|_(v<15@Lng*0Btcz}&Er$I1)kRE=8@ zPQBB1FwEt*>zkC7sFA!CWkIAK+q;Jw^fAo;Mo#0mmvZ8c9ntbXvW$~!ufp2W*td@l zP?b65Z{dX^10bA5h5U_Q`;-!_btN(I`_J~1NC zAtF7tgHJYsAIY`*$&YbuYuUGQwSPv8Lty&v_0Pl#_FzV?v@`gQwBJ`p2zA~iJOP-W z8Fe$x0hB;%7~L_Pn{xieMlAl~ zCeFvEAn8I7p6PPvd`b(S}t;S;Iul>FuM~9fskn^U+gx*G;f8CpPmnA9X zv1Wu!j0$tE7J0XI$_LbHi%>?JQ=ytJZUMD*utc#{zd4}A{I*d9i@$X)?wNO5Q z@GLm*A+{35t@oY>s}T(0AP+oZsi^XM{Q_=K^!~+9p#q~LIN!@Q$4?{VJxfdk1&h-{ ziW4vkca=2_R}tJtlGa#{mmA9DPs&^i8M?m@N`;7 z=T8H%VKVYs&?*0`1`ualj#y%iD0EI$^_(_#>29|l;iO*l8D&n8BQfV~!!{(6b>{(9 zeYO6aBB74G*=ePST+8}V;P^cRVcU#3K4{+%ogMF7dJ#N(rFl892WHKG{3v4=ZJRBT zjPa2jNqd^q8Frpk3L%{hoI&2GU$*UmrE+{16s8hnjH?txA#NGzkPo|PTBWHlGP89R zQIGYnijjEK{49eOgH)6TE6TGD$T&9>${14;b`ITW6@3_0C{cW1D8-ru+`5^eh(`8* zDD*#xM(U=OoLCAvq$;0oi;Y3Li2d;bOAZV>!)}p$JR1)w(N9mG8+LB1JM~!RCyW69 z>XF*7SH3f@z_>pQ96!C-)?JPK&~uCG>Z%}37f#=l#OB+|7T}@Ek0A3Qhs{>FOP8#C4yooj}abZgpnxUF7g-m=!$wIQmR9Y zB^5aA?!1R|Gw0WR=!hr$+fVuvgPho25>oHNcbj^7hKGX(|F|zKX<##+yi{7N^e;+QvaX)5csqjUkfn)1=Nd8-h9Ty5Cn!Ia$2`<_<|j9V>{J}qQQM1udaH4=pVk4WjWqA_K z^@6ykV3fs#jH%=_dXfhhxNEDgUuGNZVt1R1Z1hNT%J6ntZ9L z+^dK{Tj_;fgnaOvPHzqjuaHI6~oW`>$!@D*2MI%9$>d$;%FgQwIAj@v+_2ss> zn;Ps}JJhPB%V!-n(r!iKP(}hs46flT{pax*l^XU#RZukQa{mNg5PQ&X=5D=@MeD69 zWoA0>q|0(WjG)H&Z-xmw;^U5|0^$LG zXO~hnrfe-lkN1%$!rDOiv|j@Tr}52ZVc(FEw*92+=GZ-C>ye6!IsUFRvUTe7Q_S+r zB2hVlt>Fr>)3rug7qzxzhgGLVWt??!#pKtc07eM9G zU$XF4C|zTj1~!=!;2v}K7&~6j_Q7Hx%1lB*8+p+#KcI%>P#abvaK%2u z=OoBypgy&E2ha5Er}As8%b8BiZ@a$sJwJ%=?ZSs8S3lGCtQEaei)LSd>Aj6e#i~TF zdadXl*M%8*s$8o#QT7q{3ZIgrGkU zVG(P0&pi97`E(XKyE52$6$X-%D(zPwLA`Nu$BLO)!)aeW^=E>;K$oa7*}`1S{IH1}_OVyHoOW`2C@3mPad3 z9r3fjlkCC!-<-wr=gC!Cw3kus6}a{^tjinn6tm2n#LPxYl6Vz|N`*x42~e$K$BXHC zz$k^GAnMLErm^!H!`)oXdC=NTV2O0;R8S|s2`Mpuv$7_nSnK(dx&T8{ zIed<9U9B-K2L5rzZ|bWnzu3tU-b<4yp)c*LoyU(7r0(nwTQz4Jf3GL}HLc?;4R+-B zW`}Y+?}JV+fPM!197!nSVAx^vR}aj?b|gmSTL}YVyC?=p)n zWCE|dUi}v8xuNtLY~gKpjU;T)Oi6F`AyT*#ruhncy()o`1Z(v^{yk{5VkafTgyON~ zh)Vf)RvL{gJyald!$ORvZrs0296D3;e6Fbj<-@j-2UlXTt`nP=YFdHmd7ilM()gRI zd~8x9!g@bDSRtkqJ@~`7qRs!v^4*@E#-`e5=;CM#<36=YT*hY`dbA3pGj2oY3`dMw zF6l$-o3V)zwz~M({%yqtND&bn=y-lw*KDu-#^$+4P~z>$2y95i#k+mOkD{ko?_7vud?D=sTJz+5hNplHbxUi#PXbVG<7e&)GXjB%VG zggcHX5NZp&>?fhU=~q9?VQWza3-lpA-E~CPpUwwW5`Xbw{00sCz)@e!0%)Ds zQVF51b3xf3SG0;Lunvc|#(DU~9T#gT0|i)RNN(u&Q8WXyp1&DHwlEOY?(!quj$Nh- z>GhEmhK#(#FAfAkWd^tIz5q_yimF6pzV9FJ$F_AxSi?`=fhOXU!-iY+b#Hugl*;L# z!E2lv-;ijfN zYf~3IkeMZ^8LUP1F=75=LT4hZ=RS27?GNjH)DYV079EZ{`qLrtg96(RyAN7dyzSU# zhzWbWih4(+858bC`k(J?rClb^+HqvXO$A9KLmc)k#eid``E!S&ot& z`0bz@CcE(7ms3>+zy7O|(oy0{#^@Z|j-=11UIlnhbG1wWS5uXdnH7C#nSHw={0Lf4I$(wLPV$9ydr=9%CLmv(or?(v3c>6vXBmAts!Q5)~DR z3kYY|Rz?s8%h*C2=+50BrrvYKGdD~X%R*P0Vdu9l_@LV-it^sR-TLUxe`JqAK)p~>SFiWk+K$*m^5^_H0Hdp);*n@12J_2w~y*h)lV|-v?}xS zY!PQBu@L#XuJ;h;Y8$p{#jVs|-?+-~Pei7|A?q4MR}%d)n$;2c1YWB@Z+i%JNbt?x zcE1^_mg?H0M9XC20Lcy))3Y?7UZczxD-zE}Qm~ghyChYAQ>&UMf|cK#OknQQTkW7- zUW4q}Rs&*)S9@U5cP(mARGcT{id)j&$=L(c-0kJsteeFHhn!0IckKWj{8#HLSKQ)( zSo8~e?hGP(UTYE)TKmKGQ9R(u2z^=rJy+cC#g10zNGMeUvU1hMvNmK{Ql!_7oOv=k1 z*wvfno7CT;Yj)9&Gc8kJPxlwsv@lS!av)2X>x$de^(A|nd{rBh!|2y9bO)VeA79Z{ z*3S0Pj$fYS6RWz4fFTYfOZ%0Hqg0k!~rOz-xP87u57Uk}Z1{2C$YOt9SweA3aW}$GVYQ^R- zs&C%h3z=D#r+wA=)h>EQb|B*f?jcvICx~XaJn|QR!9>z_A78Q{`jLM3(iM+X z6mG6C{Ght#+SZ>Cf$Y-~oY#>3P2{X(wNWHqFtRQm5OuVIdw_)6^dATrQW@LME3VwI z9oP3~?fvROKTserpaDDZ%F{kB2}IP~BL-o-Q>&A=9%+AleSo+3%=JNLs^k|Xbm55? zbRT+w6Q9pr0(CIVCn*TI?+(?jB>-QQ9b)H*C@}P(l`hwdu9a{YLP}Q^c!_|kwl$F+ zfu-PUU%TDbjYT45PJ6VL<{A&3i(qru>Q?9xR>kYFK~%v6i*7*j*`p|N{r`&iEp6i_ zzhGS~1CZFL@1!g(WjOQI_-KS5&Ev0RSg{4d`w3Uq`$8@V9-9x)= za;Abg)P3KZrwd(UY0^=4)8LL&&M7;SB|* z{y<^{I_E3K=TC3FY)5|*>Q?++weT06ArOMdQ?w?Z>c~$xH#FNptDl$=bbvF?5W7IA zV;Qx8YwXRc<=~fJu~)grgwZ+5iVAePWc3s3vM3YN*Qqfk6^BQmmvv-Kcj1M57u?v_ z!@~Tz#2T@9^U(eIwlcG=wUGUDo=EJE&p;FHYv65i>j!Y|7555cw-6b@rW28Amox0x zLz7DMU-tP)+dj!ZP2VLIIC=s4?cmh92&sSkK0=1^%`MvzRB~vVQG2P>ZON1ulK{e6 zB#8pcadSA?_k#>;61su8ZFPoXe~wdX0bURjhVL~dz!k`37~z&EnhMW5n@gN4e1R8e z(NBHT7;3E!>6Bw#5TZW_?Q~P2P<0=Od@a<{0HIHoKzC?5_8~XCdN3KWi!e4qYtH+B zUw;0Ku-?dTGTA_${(uxG0>1iH)a_Hnej}+IEZ_a}D9bdzY+U5cb%x zFfYp~pB8hS9v$e@Ytc^s3B#1Ml2hjZ%+Ip^ z;qwaky)<9hkAtYF)UxTD7chavW>2W<7G&CY@Smvgi7Q{Rc<9)_X#B0r%vEfdFS7NC zkZQr}Lh5#;uszh?necuVBJV}qLc77ChNd*eL+EU@clh%aHwYR73S*Cmx8onvUAv#M zhuk3r5z=<{5>ys<(R7&3a~p~^s2F5fOySck=vfF`jckOqGZP)rqn`6dye9A_V{&q! zC4MBfl4n5xiREgHV}?w}|9eK!r`c9ZjR7EAjbHK5vvbnWk5dvlkHf*IN%f8Uz;>lq z?15X2UE99VXTX7Z_t+Y;_V|mFbT6VCC5Mj^nRBgEreCg%ul#8(8fgClS;u4XO_9}f zgV!!Vb9nbpqV3iSbcszr^x0!Iq|*q;FYL8v!`%#5OO9m<8Hu88o{P<-_;*Zgt`6~r zSCp7NILE8nJzsQGGjRWY>|p)UDr0EJ^`M_Mf+KiMs=wgHW~LpCXUGr-(;<*L@7FC6 zyLZ~~H*fgQ?zB3G(5Yt4UGktz#Kpk2bO1#_ zy1$0?lT_Af(O5>xM$vaF_Zw->0x|9OL!r-&>kMtT5qu-9J848=B{6r_M_cN13D-Kc z+?^FvWoLE|ZXOgQp>X7 zHXxi69}bDmCCxHUBg=JQ)80d|PiKcY9|S(*A*-DG!6;QAET=W!y8q(S&@#s_IlX6} z87kK=PqNxF@MN|(%y7=i`I!5wj;dZ<4g1O-Yui?y-tiYnxCEf zmt!rLEng^BlrXv96Q7_gG7xlyg8nk5yB^oWq&r?yDUSe6m~k*d)p z@a)a{E~y@vQ5gg0s(upeZ+adI`VHwQQ|_E@dg&ip~8=_5E0O@GWm_C+um8O=xUHli^RE?AW#&P_o3%$57cf zDBkd;*P6U5u?`iod99xqDHm(#ryyEbY!^x|&dSJl|Gg?sKMAdhxy4!VI{tO0;K;6_ z1KYgJuFaqq%f3M2FE74sL3(y-yt!6SpW;@KwNZ8UEFP|{TWU_Vq-+L;?EF+P%zg;+K`U;N8f!umB%3w5G8(_D5 zf&0zLAVUtY;@IHgD5G+LTcC6|I(9C!jxm(s!cp&gO^ZeAjSxNgi=flrah94`@YYE$ zw)yKCHrlL@jM%cC2ief5%gV!9AKnR{TMoB=khsjDyF1cRP{6rLOJWU@mu@n*J+H$3 z=E4g>ykA3QDJ8ELqx#|4FNr7`eU1{b1IY2)i@Y6GKK(8;jN#X)bF)%SVArfR!d0T_ zo8Lnf&j@iBe#x5HT1LNvp&Lf-&W0<~>?N_=+jp%z1+x(jtD)vM?st4TBb6jWMQw+UF{&=mv-U1YrSX_za)6Aw(VY;#dnW$b zv4eYh9r|+1BLa9a-UJ~37PsS_S9w;`yI=*zXsSTqO+$!b;NpU3(Ss$6MFLBp?(R_a zGHNBer+n<~<6*~NL$TgeldSh|ES-B`ICjG?Oaa-ahUPtcDVcR+qFrgkhU*h*9lYTZ z>%0Qc0&mgzrK~+~yGJt(y}5?KVsNp8INoI8%sd^FgQdO{ehy*4)T7?<8dY=gWHH|S za-LDtmf1`a%?a<|!(41>&t#jN$Ra7l$K@Nq zGzKEuBromNFGzr@45DI5@XN#ukqsxrNa!4hEU*WSxaOr|uR^F?JQ=~GCFFzZ!L?C( zrDksAY|H8$+J!_xA0-GwB!}?D`o%Mk1V|fEQdA+V)t%9Hrm6`n=_P}2W8r$=v79B` zwg@A!VsBup<6ejMJF>ohBAP@`>pJx{ukZ1 zet>1IY83i6M53Kef|VtJwitm|6*W%m+ETp`v{|+L#fbEbgg^62a)Q9w>Vpu&_{{7>Wk@cqK=rxUtnnPFzIf8@6)Z;h89rB1oR{Tt3o)+zUM$ z2SS!BxnHG2BlePb6`@%hag?x2!|T!mH92_O@o|R=2&wnTzIqar$&LY>+lrAzbBB<{Z>&!G9* zH%hfF5xDPwt6`$2jkV`1;^#3YPn#{^J|z3G>_B?5O%1w)zTL=UCF}E4eIYOu1NORi zk-=BRUK`d>F0zJCSo<4mOOuESu75K%a1y% z^e-k(hJ-|;pfxiqsg&9}tZHmPipM|w5Nm;Q!33}bU+c!Z47YgHm0>vi56S|ENv~(O z04Y~NCGyO^EW9qV_*%D&wCH)ViY(xxu$jrcg0^$8;GfIAZO@zvF)ZwW{v1Lb5dHV; zMj5>?pGTJ%lE1L=41m~Jbh};?6Dtx~4I=i=q(KFyvGB(Rqr=z8RQ*Ly#1+(tCA$4i z0yxi~$du*=CaM3z%oZ?DgfA(2Uge;N+$jF$Zm;)UY>C~EoCITry8m2^| zKT*}RsJo;i9cOBXgZluF==lmkaM+jgH(x?e;ZC|D z)jiM?3_C||-vjkJ`>gcR=+(oi$Qw%>KmO?7_b}r^+)hkeBf>0WWc5_@1&uH@$anv! zl-e&>m)auawO@<89UA{MIP!h<@cIdv$Qsh?;)2v%X2P-HQW+-@9Y@47~|)7+?;F#Cis2tA~)EOx{@uJYSp28p5y>afzR-=47ndXX`Y zsMe5{*V|rBIH{7VCD3MY^OiKYOVyUN-lzR)Bq}BJzUpwLbKD8Rnk42T4CI15DT3~u zjNW@h;0Mv6-86}fF(1z0jXF^UTQ(#f-@9&*8hGYY>t?dng00yWESPDvfwE^2`B4$x zW62EUY(MJX^+AC`=SdoKTarb+F(4{FBKt@`_njX{t{rYX9Wf%=1HCd%$8+F(e;i_J z%u9wR=-M}4j8IZ27bObO%%JQGm{ljYyXO`@TV0Sl`pZ#5RHDo)OO5WU>6a3wbV!tHSUb zstPC`Z=zDxlwn;r9GJc}Kla=Lq(N~hybn}y{Tp~1(#9{m^%PM^=;Rc^w1W#(5*H@7 z4SLwW&^~$%3EU=)0$Zcr5+bs^W+$^7%X!ieg8YQ}PiO!`O$@87pc zCn;xWf7&BQ2mx&e&fz!QO|f6aauw}fWsJAPc9VH_0Fo5IgUFLbxueWK${6^&!g)6# z{q&cF=^zt+VND@!G9Pep6!63<{0ry7;a834f~pJ)hZy+Z8=;%n>HYq&e8Q9$6<$m) zX@iuzXRn%+$I1BC_Q;>e>h{)yZyiS2fEJNA7zBA*wqvy#!Fs;;M19aG7fl_65O~Rp z%;Spb?LQ)>q!uL43F&kLUAycDIE-GIJ9;qiR5r2&^*#M4p-9V)8v#CoQ#4Y5(`=~^ z(ho|r3nCI25we?82^2@x{3N-j3{dhlWc3L;R8Yk;NyhXQxzfyULg9{#y7k;4M`bURqydARHXjJLpIk>mu4NntmaQ8BJi25L*NXkCMdZASMqxw zU~}nDe8g7UfPmB!_ubzd-`8tTEc!U=pe1Lpc1g7zDz`jpMF}Vq3ZAACk?t~+v3|PHTh0>`-2rmqrQ2~xA;;{Q!>UyvWW>F6uu%?FC-;yeKC-<3 z8c9!EAD>K8j_yF)h&+a4#j?4iZLTUq8&bZeIkhv+2-OpFG*(0nBBX0us`yn;V`!|)*lUt8w{3LTqLaR3S@NeS z@re`M@ucvha5|gtz%!e+-#!+>CPCmYTR%JQ!|fZR7vyb!7V1|?!X}N!ecqN}q3!3g z{zM;nqtNN^i&~B|+{u-G{OM#brT`+SjIfqQHNG6kU+usH_Bh=A#R=DhBG-&#Qgo7N~gkfs-P5 z_r=y5savjuJSSUL6v1?3)3}(>w96)x7DZsuL(|Vqr8eIT!c_utSQyDOnCeI7xbO*O z?Knu=RyZ(DTiz}4kn6Bxd5x~{DdPSJKM{cSTt*Vhx34cllD+KkBzJ>Mb=qSQ2~WbT z6Ar!~93M)NsR-2jg=3*cS)IMtNOiX@o?v|HnXL?fO-d(?t|fLMvg1@Kc}<-1-j8g! zkce+tX`F74XwAETOL18`OSOE527~16eu#nN1&=rgU_3Ki_K9D0v|{2tX>ShVFel3A zNqOTsb~?bQfa!**v5X(EHNHPfCxL!7^vQWB4sJV6X+Xx;JbxQg^P`q-K539HP>WIx zqB2aCFJ45LaLig#Kc=SIX#KkrUP?H>=-<5xEEavz%|dRg5XUjYL)RWBFh}6MTRgG+ zn3`Wxj!^qO!vERT2GEJiAQ?v@{O&m=AL8BiJ6O^-a|nYjmBZ*eXeLg zYEB_IMS>BmkxF#x#0cv8E%v$_`yD94LLYM3ay;TarR;&mY|wn3>SaitTTSF^c;3i* zzD?q_3DgpEsFWZC^!RYj9sF^1O!<#k7C_?t4Dt)pjzxwA4suFd3`$vTNq;XRA*{`>JXP?DUa;nCv?TJes3J zbP$d|CUrFqu&!a05T1BO<#JmQBozVBq#`N~6PPF;Vxi595>5RBVeml%W#E$h5_ z+li_1L`F)z#CPMInH1kVyGz%uj38ThK|r9pG_IfvH0ac(Hw=Ny2L4+Cm2)T|MxdZ} z_0dBDRmB7h@Oj+Bzok;yYMpX=TfgT+pwTLstegL-{ zh!u!k!){oBT4lWJY0H2<9YkmXx2;g4I}t>&y~XNjDXS&mxO_Upc>)guq>>I3eA?$5 zs-1brTj)CUZ{mF^FXns_%J=BH0m-C()@)N%aLvs`phfi~>yi%f_xDAPp6iHSN`osW z2K)+1`P-t8^R-Wo;9p{uwgM1X-5}mJcRad*(m=@Din0BDFYz|YX{=`jj~?|DY`4dw z>zF1_)C6tcsd^D$2+}0*otLiNW=uf|m^(5wz+;R{etqu&1qxMnUe+|zGr1>W%_^~u z1xILi1)o6tYh3W0niB9=(a3{$kI1(iap6ptJ2nW&e=UifC&Iw_mfnq&UP12zpZ?nn zqyjju#aI3v_k4&(y(`{*K}beJWe8JB16nXY0;)T5SDV;Ll{~|YcnB4gx&b!2Ir3pA zqzEE#Q6~J~_0mrL5f(yS__>-*s@witIz``g53$wN&xt_>k4JUVNmK?I>~zyf@Rm_N zqz1NZ)#rzf;B(*ekI%(;lxMCFEGZl+Pkl;1;=M zqgD2B4H;unM+lR@ccN<*su2FfQXk}(7ga&6nhI0z$d~!Vvda*cHII^s`FUtk3$3_p zKNBCY8UxZ9Hw+a`RTFZEa}jUaK0Wv(2l`9f!CwgcIQL%~EVBTlk2i0Qkd%`E8tu1n zTTIN#>1bp}0I7F7bqbNR9-ZqDTqaR~G9&E58r4LqNaTqw1o$Zp%(fp0X%kP3BO~Q& zw^0-fQD8m8Zos0N%T}#=fs!JJNIJwjJyf1Zm^?$F_H5KN`tD-thPs zd^`B9*`GUgOx$5^L)S?q|^p2R_xY2G^ZMqRtPpGA%pv}KRMfg=htK`BLfJ1ZQ0Nj<6hv@wMYeR|ICtaZ%o6vMO=uDk`%DzbOpEL8oF3>@FR>dEm_MaOFR@3#QsErKZ`Mp4(jP2L0On5{7Sm+VDtFu2$ z)|faDT`k1N*Kk&xV98OfvSZADtw2c;Olk2Id|2_;yEY8?mtPmbjX-AUXweJWPF0t5 zY%%`uq(2S(;hCu}_xaDs9c6cIJmU0Vp0de{xMRO0+?&W0Bi;?j$HmfHYL56jTn9hv z4WVnd1opod;1{zD_K)Uo;$5q_4!7Z7@z;F;w02+dtrKraax8#0?;CG>&4IR)HbD<_ zoQUG`C^bj#f6GBh=+aJI-R|OigXcP0g+C0VZi&S_$b@^8c4}|hoR9Y~;Pp8jm*!l@ z*RRIYP7T!)-=25!;h5S%7j|kI1bQ8}w#wq22g(@txSp!c-0xWeniU308QZ7*>w7F^ z!sZsk;55Z= zZpSBcfbx^&e3f1F*4E3^)wLQfqfZ-fq)guvy`>U)D%L8Yu?g_=j&+8StRD^sQSg zmX(LoI}XAh{w{g019jGuyuIfke}J>jUjOOZR~d*}1v$KY*-g7p_$ytuw84uq9}%8w zWB`WWmSqO?U6pcqzJe^702hriT&CB^cig6<3JUOvWlIqe|$uQMp?d9jYyOuI>;tH#JT-z z`yo39Hz^5jv!%>9XVA0788rmo6zPH9)@KS~#Opu5Z5hFo|4LEU`G%mXiLie6)zxG(zdNyTWZ3S7>NB$d>)T7g`ja?k@AjWI zhv-dP9u9i;<)!}ilA7ep(YtxSEJsy1I)3@!fufgn>~Y_SZfSOX=`H;qS+ zXz5b`0CHA|T60ric>h4%1U?lM4^VpyaGZ2ZY9@gAs9Itd`R;@G{N2+f2&M#NI?K8KwP)6x{q z*>t1*M(jBxwY6UM%jugX#h7!kA7J*ixSTWE~MBnzWaE z&z&^)&vJxK!Tn7W1z^vpMQe1?>k479ton8;T9j)k(m=p^GFY6GR|Rc7B}@h>1)C9u z;WyT5o*O^Qe$Sv?rAR|=epOY@^3&{5`;fScV8q@CwF1%zR%w@NUX9y*1&)Q^I>2Sn zzW#5I0A7>!?q6{dCNoc&ZIXA$&Llv3#WM9ao1d5XQcGm&xFU0Is=^k2Hqb5&IaM4z zLro44RTL^zNpcCfOOWnbKUi7uDAb~D=QsW9RiT_#`-BKLepN*ezaUJ7VH2t-^E|^3j~{#c&K&F3f9OeFGgQakEsmPS-QZ12^VpBn z;Y=J7=AIeKvR>rJjT=;LPLvmrAF^RoC(G@I)$I<+@-0c~EwL|_$Fn4zT(raO{yK*( zG3BNGhJ_#L%e4%4Pa0E$T^$QL$XNYxK6*yWY(}DgcE0utZVcs;{oay=M_>MQ!lZ*B$-7ZqeI-03nEwvkp%i|8af z@amWSyO`G9lGkH2!O4gFz%H%#@{szIA-FaFGAB;NPSLJ*oAh2~Gg|Qap?KdpV*&>Y zIRz(9J^o9BLrs|0$BwdX^w*)sme-1Kuxy=5>O|?)*{4IVG_RMfI|HzIFI~Ip&EZ>; z<3q6;mBaG?fYmq7#8P@vDlQHFXWP1}0zlPhEPS9{s?nf`>=(3Q?2nsL_jbK^sc>A8 zPs2oyDn-oY%L!GAORyqG?L;8nNbKKwE%(7(d5E3jJ=CiZ!^g4XT4!m8mRHkRjfqyq zM(1o?XYhe0?ku{s_~X%(*#9VRMXT+MIldmt00Rkis} zD0cEORm2{yKCvw!7T$=`s*N~8Tl`qWgzwQbbNT+Lqa6cw&bURp>^?w?n+3j2?gK9; zcrKi0w;8vBSkquf0=fNq($`gr%ywLjI17yY_j{i;pWo)33mG=1zKq^`-fbLXIfhQH z&0Vd{yV$BSwN5GthZ!=O%1-_Bni>8n0Q+{VUCyFj?y_7$3WRmOsIz^>c}H3d>_!5u zkLB1d$R^Lv8iS7(k>!_D&%wT!-sH1@jr)DGfh|})b^AKENE2%x9v90I-PtRy-QiE= z(GaW`@5bE-PFa7fa|yk(vc^hMYT2U27G;xYUOhr>`Iuhx+1kozRD0mDNR5u&7@}CucQg6_r?!vwtb>p~pRz zONhDK0d?B`m=FYJ|)8)CngG)FP-@#-VG!ALFE{*&wm(m{?TK z!jQ^nm%E94deJM_2KNXHc3*YDb>|D;vx5D5cjGA>C^NpvXb!`uMP{2oI-H8uVD zLKbA5MocwX&!#lQo;-2OkL_Rr@C?+x>z@xJTo@G-^);sK+xPpIr$|#8|c8Hi_cJcIYlLtDUT_f=kZM+)Am(by# zkBx@DTtcg-CQ=8~B4`<%)`G1KML#(u5YrtMG2$gtCj_Dn6W)Wl3U|}zLXH-o_6=j5 z_iy6w<(;jV3dx!u2tIRANW)WBJPA~rYv0%QjTe5QA1=z7(fb3M2bZn}y9NB}NUKSJ z+s;dr#ZDKFQ5$CjSAR`&4ft*)NEFiQ{L*XY8@T^r1xvO-j8x7EoZ{eHa^yb1R^D!3 z{U#y1z}17S(%w#Isg}m0K^J&jUTSxUE$Ml?69g@~gbFHrIX#D4GuCQc2!6p{`JFsx zY8*5b9?pn;F}bbfH~UOs?0j0%_69_lP=_oHpLk@LM|G0X&yKEx8?)wDLL28^2%lQ) z*E$mWyEg<+MbX@mE@doIi)E#6RjsM&T{&^C@-_^!y7(?;t-gXBZeJ8jArzvwqJ@cX zvX0P?h}G;HzC9s0s|{EAPDg|{Jv z$eFE%OvJ-EAIG`6ezi<02Yn?7l9`Vik70yTsG}(0GjlxK`R&^>6*Jz8cZFJwCx}*%X60vD1lj!A*pfh_BN1UYx?vmr^pk4g+3*eX>X5Oh zTiQAvLIKxPUrJ=|101~1)|{wSYB<7RLDBw+Kjn>{pnUDP){uLcq9mEDiqhL_D~^?^ z0kHC%@qFX)`48vt?Vw(gc#d+B$8!^wO28^UA!5r$Gp4RHx-{#VroJkSaSci5U26`* zJxe0O)Nn96J`yk$7U07eJhA(}tN%sOqy3>mR$;(Bs4_#{h0y)rAsq4UB z&ZjV9ZP^Ky1B5Kb91i24Grs3H@o5PrTL$GFQ*USV(6j zz@v$@!=neWsrb$%t;PPJ#ngJK3L%&#=E^Zm+&+W1Qokk4|yM)kT>RfZ`D)B6@rcRXaHVQ=sdZg>8^dH$2NWB_@y!4d2bb z=^RZPw~_`?k3rW|k?@aHp{CEWH617(Ldwv__7z8v_f?yCR2OvGyt-vObRS6JOW`y1 z<)%E7z`vPg;DP1GGe(nr&C1r?^#Zn%oVZQw-&f-9SsO@+a$ zG=%UuX%dk61?>Qm%I{19Z_en3;B5(FqCHSfawCQwQ{Uf0N*J5xXtPbMSq@YG0uK`&!@JBtd@mNBheQ0Aaz&G~J9;m+nn)QR zrw+ko4Bp!%XoxcWlONF4o!}Q8x^0{Ze?d$~(gT&pZK+ueF$fr%`qH=RmEUujc($wg z&1WBL_6JX2E%GcoD-0g!mnxkw7EXKyu!pjKgWqH8;t^N?%NeR@Ri(`;Gq<0QW zZ{UpRSy1_Ujg}Vrx~v}ouC-I8<7ij6=9e?zcFt+-IJX9u?TM9rv#g*gZE~q(7JN#UI>)sn3U!$2 z*YjO?AUA`I{$80i;BAZO1s*N-4mBtrj@mq@@@N+wl?- z$%DsTt-Ml+LPzA_O)U1>m|-UIZpDZ6e6J@|?=CD@D8sSm?_v1UZ$+Dx&I1+)l5SaB zHW=<&V(qwtviY~HYsm1u^`HRpIr8K!8`DizGW+kzx1VKO_H6<*^)%gZu^nq4Hkaae zkeDMh+vlF`plL&J@IN_t>JeIjwSID+KeHX*HbmoympZHRBgPGl8G_rztPhV#I}0wL z4ki=Dm&(P9$9)s)NJvn>>Y{m|@x=NycN$4#mXhdudDE{#Dta}gh3v`^GE8`Ho=*w4 zR-l0P7NdR_1;7^XZ?i-OS%H*s-}wxFp7R$^e_eNv$f|nqdrY)Zqor%Enw2&h)fI#G zy(Ai>;^wX(n2qhVG(g^%$tZ1W%ghSt#R^~nHyJKopd`%7xm5H1+VoI14f z%5*x;1|_+?Qr(5yN^lkSptEfln}Py!ylk#R=FYLfH>+&gynpPbpG{pRpHZoEL}iK4 z(wLwgX_sgqUth|qYd*35>J@@wBIE-lAy|d`?+TKfSyU0)CcpAuR!B@xO1r-N0>W zuJV79kX~_K+n`8(H+7HN6Mto=MZ9E9>E58$lpdiUooBTa@R9HBrNuvqmfhY63wLl3 z;t2f?e0C9fgcM&xt6SytyE#uAnF^nbZ^as3Lj=brvkCqlecFwATKD`XZPHYy`|W82 z7AoNXhXU}*4m+eD+C_fkuK3eG+Eufuo>zW1CU{;7c2 zAx8JEXKT9}qKobm>$_tV99=l)xrgG1Nf^0vc=Xy_ImpsNHxBgqPiqzM7m%tR#3AcY z43EYPbqi-H?HhE_BG-^0r!P-3ftFy3+Crjd4s5!Y+;dImWAbViE%n=|fp*zX5X(Y= zor3JjwZ0KUUB69@hv#sPZnZU}^2xvqtVB(8r@!nR|6&M5^HSo*r*aQsHP?m54A<%4 z%#i196_T~FrcRa3r!_qf*N|t%Ju#E7xLy0VHlgDKV*oEt{d+T0Q{QnnxupIj>92Fj zYF9u7rX-xR@;L!=I zPR4{Rz^q5(rfqJzf;3UaCQe<7FHhbzAB5)_$`j;9Lq0>*UMxKR#*NkT{^>@uM7h!h z!{jUgM{eU_-*823pyV>~61h{$288W>5m<2gYANZhA`4`)OyD5Pw*j1+tYyDhpO2d_prcwTiE-e($4Wa$7Aw0LN=^$$>{p(Q_sNjZZ|=WzG%Nhfo(C1p4iMf$(`(*&P4y)MmWoyb6mZ z3nC2KG{$}Rl@T;X$lF;kBvpKJ3FG@~3Co z^MQp6!B)(rqOx%M`)bpU6nu=J{TSfg!TKC{y;2W_EoThQ@WX!6RDDh3F8E6O=`DR* zzSwAPfBScMx8$cJ0@R0`dIY{yI)@9)97$SL>s7C-Wo@G_ueNMtT+3m2sl)HzvaBu{ zR9-zX^X~zI8tidJe-brl4I~?t$|oO(SX_Ju#zR_o3JCZRB)qv`=egZWlk7w_T)ex{kT( zpaPVRylDZnTv2x__(r2pxBUHpWEaTiKMJ_*hln{0=0$;V0EuQp)!j!{9zvdwd#J z)pB)S_YsrN+p%Ups%Fwp4@#DG`s3jEigF!gAbM3GVGyOZ5>Gn-|FYNdhSh4Q^)A2K zSIdI2-OdHmpY{2JsE*~*xFSLj&(C_jWW~TuXT9^VXd9)iggG&xx16IWqf>kfh#WW)LB`6^!Di3pDluFuatLiaHZ*B)U?N5ejUtmqBwRslP7SKJh%NT$64YMm#%Hx8=f`Py&<|(Al`Ee=4&6qDYI$t^ zca>rsBfbLN(s=T2`Fjs+Uz}L&zkB9C5*41nA5YQ^0P<>Ack(9}Xb;hm=7E`jRMRVU zLcyr9G3cI=HuZ(JJfW6$tZWvjPdr3!jjIg;7K(gy)|@C65k*NIF z831l*x1Med*}p+qLIF7QgLqc}{;`L*4Ry^Xl;PtzB)^(sdIpeOqW2h22W&;jfd^n; zrozrmg4Ym#fd@zw!SHlUJXqV|vPeB#|m}2DQE5?T$J0yiC3=m6e)m|DS3&+8nK##mYw#ZJ)u>tk-9)g3~;a%<&fpZcA2 zu4GCu>ujkE%`%picRR7gP*?O%(~!r5oyL03zo$vHJsgD9qE?t_Q@0QfI{PW67Fq-jy?qI~VCb;|jYN&5XV__oqEr5pa+}vt~ z@S2h_C4$HIET&uRQ%dLU_(E<1nNus}?`4o=d%%OzxK2YbT+dKuJKm)SYBPp8Ssc$g zq+cAy1VM)|H62!abi1AyttU<+;M2vc=GT6#D+7bV!Ua{Et4aMhq`u-9ZES?&*m zLD{@Enc2L&JSVmq;x^u`oy=K=aH{EIa4|&oFXAnOq!r5X!h*1GWtY*5Er9-0pe+6G z*G(k+^<@?dV`VdCeZ2Y5Wc0^Ps6H=2b}gNXyyw1P6*BAyT=c=6XyIiZfM(8F5YjT+^d~@Rc2z`QaAakJVs1>69_TS z6ccZpGAHlQGh4Z@{czq8wMfZkz*VhKutS=6-fN}4J7i*9x4qJReG)KnN zlJD=3gEZ_PQB14VW^mKs`K!!>#rsP?9oqu_|G;kMf$Q^3w z(eGHVSB6>*gd>L=CQQosFW9{E``PFe)W?vkfT8oTJ6p9JzH|IrwQMU$lxti1cnIRj z8bl}%_=2^5I=u2aLR{9ft?Wpjeu{;)W~h^OxeeYPA}+WUVLCNLZo3AxZ%Y^6lBl`3 zgh(v?W^a3r4DqSk48Fq-v@B)h?rDT9^3vqNyF$dvu&R)>73$RIKo0g+OoBo@dpB{3 zyhxjjT>z8UX3o|dN|K!L5XEx_P9gti^61m0MB|PbPdgcEvmRQ+tlf+9U|ZRIg_r}s zVLP^#H(k@=vdA50c86~l&bXFg3X~w%AKP^%-8IN%w5XpQ*Orp_uBNsVe>VDgA-xoK zJ&&VC*m_@-0=rRnsOfj9gO zrp)@|>|}I@EMxK!?TTSmdFwfyrLvAHQgf~WeGuN`yu*`)A|k{g(j`oUrtWcHlyBfcDTj(%^+tW~1B*Q3SO*S~WKYhB+oJUN_sSn5#4&pH zbaQ|_lJCS;66xL1Syj^AY92yc6*C;_Oq-2v=?vcC%09UB>xXge3bk7O#oYUsY4I)? z_Rl+%^Avb&TtH&R!6;pjtjq9idqSc$2-&JS$|3m2IWNPGCvJnbF-2=WB-j*7x$rFs6TOvs@+d3VPzMw~@Z0yYx)e zld~TC=hY((`!=n}yF}P^%r6-cqt#u7&Zi@FEU_Wk-h|))G%>9?InJS?1TZ5?w%C=* zjSz1aYp|*l>48@2whQHBt%ncqcmfe=Q>(Y6Ro5Z#Gs;=zi&i zwD}dw&WB{E6!kaN`ZG5I>O5L~f((cE51y<*o;L?h4K92ur|A(o*3l+T*2izC1{oYN zY;lc6Y;&yTbXtneSeYg3)6BdHXfXnY4%g9w_*Vtkd3t7@W!B>UCPYI1Semk zg=nl7O-RC=L4!CNwH-9Qy8G?fJsqHoErjHTX2?X&B4sjxe)`|jkVPKIQ0pFYW09fr znLj^cEk{rBmvQ|~7CvPCxR~L5TK@-j$Nd3DY$7^ah}s!LM-YsDcitAYH1Rm~gC1M|%){7fJ8!q%ua4$iR+{=v#VH{3RTUQz% ztMOXew8=$DKAzyX@wsdId&CPXHb8`$big_rlOw71QTP3jv+eZiZ6?ZC>fq~o5fRYa zbGDup_hZqwANvK-DO;IZb}0`4h7$SM2utz%U@`{xS))MQz&vEMM{6rk4orKsB_w2foz<>1+$^GQ)F0vtNu zHM_LBDd)8>I&4!ju?#T@pHpS?FQFR|Aq#UKhkPwU>Qh*PK6J8@HP_myWYgOW#KBwC zRM^KfBE;8amx5;qrr_=i32cc>xQQ<)(u%5KjklKXB4dEiLQB5o%ZIF7yt0fKa(?P&(7*V3kk35*a!b^ugF zEs5lv&^T90_S{u6`9+lMDF|_MP%q_MI$Qjy=BM)s!$lVZ6L)NDg8jYMC_aV}XJTJq zO~#{nF4@Jl@%q7PPn3YaD4VE+nGiW$Z7PKjkoIp674?LH_4#6rW&KB{b+aMps4&i9 z)Z>1U6X_l(eS6{VTP_S-AjQ!yze_7u_T%c@VYu4`RDM%RKW*!_#7Ez};0%|&10V(Y z2?C5X^I*gkyvt=;e@bQ_FoaIwkhA}FZ1k}A{ZPb?p+$(X3bAWL=rY8kt;t2k4=5|5 zfP@1@gXbI?m&b!BivC)xwY1v(kmoTM+gACGH&1Z8+2$^o6%X1Kx8ZB#l}1X*6_hM% zeG^z`vSIqffN7Puz+r0MaeLHi5d^TVqLAMj9vujJ(cC?q&_bZ%>V*PEjI!V-Y$6LD zO}QMdy1IMM?~G;>k5=<08^4BtA-!9COJj9HxH>re6TO#&oW*%)7M$VV8-CT{x_M6( z+;{gc3!74cnLzTUUCJ*Uav<#Gz6n%0^9bEb_G~q=e|^5`@#raGa4OjXo%Y8bHTPCM zGZ%q}%?q2ICyH)(dV}0~mFN(~0{eT*7AWlCD7U{0rgcM%OQq);3H0rM!t1{<_Y1yL zcX4_5iq-Mkg$r>3H^F!-)MWQV&*7T`2m=h9ODb(#g4BKFZ&=$pYZc23*x~$b=2bsL% zH0VX;(R9U65qa=M!+7r$f?H@w&Wx_^!rD9X<PY9nMIJ z9$$cp9z;!nqnZiM8CQ3)ZLJVgIl zwUMhfV@%B`>$+i$B2c)7=+>IIPq~PEbAJ$sl1*B;k=L^T+0p8OYh>5iyCZ3khN89Y(J7LAC*AqzG^#Z$sVl?@t|WrGh&-fc zw%Z5;xb{P3kQFycmjnj^cBb9;r=l?lDk9}rFDc82?&)e)?3RtgcJRw1T;+Ahk5vcU{Ov2PSUDm*6r;)Ix@KMrz$So_=2>mE*-ZEqouy7}# zvagP2@o^dH@Jr_+x)9nLZ-mP$M!c*SkuXC^*ZRPW_>7}>S zjLxNswV=L8`HIoE_L9*JCH|&6 z%)19b!9`;47)IacHA~~)+2dfhhT0EDX#{lfh3AfmnCk2p;PuF_*I?y%O)Y`JGwJs- zw*E}K-=OEKn2T<9wUGTeExu1}SvZQ4+HF3}`T`-gV)$KU^m~_H$dxFJvED_COH4yj zl5(W@$Xlm~!X8G>cDuk$*tq0=J{*X?6Y|)*g{cj$tqU;)m+~cItVlJ#>z-9=%iCQ{ zc#p|c;$5qKB9hg8;nfLz<2qK*0EJIw>*a%rWbv?#i`ta!zm*xDkxwMfz9cE~xg1g} zO%lU))r>`)ZcIc*3zYL31zAC!IwJ+DM#mBDvNMyn|6|ek10$?wmrROdU}nkU<~Ec?Bh8ok3AL9yb~4ygQh#f|GRkcqUTNs=HT6` zeCwi?UQ*Dw*-94R5oevv(`1wX!#E3Izc-3CY0MziJ=$qJbHC)H)W*1t)!p~h2VVOG zDbvYaQ4`chHB|bTF#kBeL0#K5{wD82E7It>{&=R4ef#3^u+EE-*GstE1UgI z{L7t27KBuqDJ$^icdu>GH$}e<;3m4a^!M|*q4l`?ylVx1Q=Am+g|j)Mu*8X2c|*IX z8}4B+^Yw}I4lFauyIhEr0Gn5wlyncH0CE!L8{SdlzoDie0Qg;y{oHXc-=;LIXu4RoRQWno^=BC@JA-}^Kyp&UC86_ znFixF;NkdBrnjz5M_;0O9x4NtSK5mH?7fdDJ=WNZ^V`$A%f)qSIUL8J4W3560XZT5 z=Lwh%Ysr4#6PH2_17(FV$>}c#FAthvDiUF`o{krVH?XT=J|`-xnCz1nO>(BfD8U3LO@EmzPyQ%ApB_112E zi3hO(rnJE`u`A~*NwfhA-+al1uvWqCB+j9&MN(PkH;DD}w5zm1a$a)DPrhSJtF!qs zf@CEa3(5WH_R z$3)S&r^D$;HO)Wcy?0D%?tW_7X$y~q?I`_3egD8x^v!+!3BDKOVVvC$L zXE!SP{75EmCHxb4jDEL-JPalp{_xtZGWK2QVjwoY5zHa6BEDWS;jfIX9e~0j2X1nJ z+W!JcQE#I{f4?EbR)-M>sp$V;%i8@$TyywKs1J|zQLUB-4a?fGq)i6u*ouRS5IuG5 zGhcPAEb8abk)38Zsb(IY8Q{3y6S;NP!nyL7pu z0j9;62OdphzM{647YZ*!2L}m4gUjL8da{Bfc)BGa@uvu>s6^~$5Y%bO(l8%5L5g@$ zBtQ0SS8Kvx$#!FFt$F>du>lQYze%l-BS`P>=rV~s0{I_th@wh~7*(ocxc9w9omo+c zg^<#JFyaiStmLdU3w>aQiZtc&A>z^VPfGlK2Ljh~PjZ3krL(haswmgu4^u6W68}~C z{T;Pl1&nMEbKdHr105gOIw?g(e+lX`*wCH5v{X?1w;7?|pWyZU9oYN!B6NJH#xmMY zDHU`_#rQP^__g`LO3~jx|FqBMoCM1VyV1>k${QC)UaO!l+!8nr33Q7Dlw(pmo5AOw z(x+A-)Ia0bH!hyQttK9ZW`sXY4WI2>S4N$V*UScXr5=BdZ3>4GH6kvvav4ReRi;;x z3E9k+2kXst`Eoj9RG4^EExvJ{fya0L7pd>RoD=jQJM*I30HC#gG>^uFzI_McC=UhR ze`_m}n9#?WJp)JV9`8p2B(ck7YL(BQ;!M;dc1|ReFEp!iFQBzAJYxZsEQYnxp@1hc zW37~pqT-u__#A69n};Nl!CIaYG5eEan5a`Md8}ooLoeS%+!6{prKffrVq_G$i<@{K zw+($0?$g`nUq%S}7W5yfm*gDIqeJubd~X#)g~ObyTM1_@p#w=~y9v~|u<}o@jXMo{zm|~?0`#nt(ls*15{|TvDJUdI%(Go%yZy#PFTyPB6 zvvgVW%q1(M0#;)e>}>kt$@ET_c%iAt=RvRLU@C?2@^8-1UQaoHjew(T2wx)$!n7Dq zLrEtosPQ~V+3cYVYtUpYS#XMlg@_5+dcgR|Xo#PIxB%Ita?A!2(9#&N(N2(6H|sPp zyX^enF-Bua;5n|DBrpJxr;O#BjhWGBYg_$N`cb;t7V(Z z+I-6 zkpS?4N{_NyIg81T-U=fpeE0=l|NlKmd_zB4mIVfM#*?2M2@3ILX~JBlvRPYEHDgG1 zSv+gKTURpSX)k-6nVS(&0@kJjWh2D!(0=&uc|DYr4T!VRozZeNqTq~IKJEc=iYlrE z;O>pyR-l98)8@*UFbH`cx&~7n?br!`z_VRks?eDD>MU$G&W~eB?xHp6mZ(vVG%BFu zw+jl+K|n+E`fA)p2ijwA2)nUoAxYGuuRrR0#EnK?aO)6oACkD=sZ!jVi^1s>^31z` z=g+1Qial#MmJuug|2rarZY^V1AwI9JAWdk3G9kmLL~Q)gji}UKMZ1SPhW8&a?sC&n z(flbhAE0zqBIbK?n8UxuJkMNlIPPmno8Y&n1!TAtc{NDj_-Lm%luzEP|=2$ zPL^nDA&*JNscn6R*0B{=kKwl?E}b*bghd32qdQ(xj%{;k&S;(lp!R`LqDsLsf~enz zcf3XfB7v@t5;R))epVw9ER|@}QxP=qUzF%r<}`4!Awz_DN}F4Z#tql3ic>_)BR0Jl zh*HN1p)FoTGectX1fnx~u3w*|YL+rqtYKeB*}-bQUjaUa-=bO?Z0teX=1G3M3Q z^VQiRhsF4>SiuNfg{>u~vCJY?ovDB4ALDX3H;r6kDTlxh0#0-3&o?rB4RxjQ2&0X; zXUq7!Fm0_l>4OUU{FQnF+Mb*tvYYxI5D%R&8F9f|7rlGz;e)`#Ap6 zYp5%9xsc_ONgi~pQx6*AD5V+t;jUniyLC`Z%B75@+3snE<4c1w@%Qw%*kSOVpcn|1 zuyGAq3FM&GA422c7CIvpdV*p zj--|8#0Emq^an;5mZ3Gl=03v5-z=03T2E( zZ{x#hvM7Hd)z>WXbvlxniq>LKSXzP~d3o%{Y3$FL(6A}k;6qv~)Lqkd`C8x6I$=U) zIlQ&uaFw9627M*;tC!s5okKH39PBh5!>hv9PD7sU6~+R1Xjp*g$}(&4{QmXnx`7aO zXj=rF)}_?@kV3`?JLUM*cm)86=y_$d0==N;rmsJ$r3Q{-4T_7qnEL!CBJqi`$CsiV2bT53J zl048hkzlhJjVf}E5&W_Gf|Jc}>g}6fI(b@eL`Y&J54J1bx&#(AMB-^_dTT6WBKjUJ z6n**rJU7T*JfuG{&N648Ct`pwXAjR}^9ofhxxUPvqvu;(0aeNsp=@E*ip#;eAM!qN zEx-o7(^g3j6nxFpVr%Ruen)ER8^lA&;Q3&C&h>7BaK zlm-bNfg2JLCWnJomdb4!SFiY-!ZvTobu+*(sfD^twV7+Eh0a$A#{oW*1QB{{b9N>v zA(vq`K4#?7B1OyRm2TDRJ4ER%=s3vS0EM)3C560IWQY4)AXtCFTeZ-^THzz+Ru191Yquul#2_QBF{(%(^$c1GeK+^gK-u@C zX-_Rwmj|13Zh_FW3Z@wExE2~io-6LAkOV&VaW|I%`!73@4>v0Duz|_b^XS+#$KxSq zHWZuw>9Z9DY9nrx6z_)flmMHkl*(?`CIXH@D)#|TM1H^CDLGl+$(zU)313p@#;X(% zcLV$paT>oA&+(>!9RDnfkrdl{-~Mziy7gu5(k^Z9K*m_O zT&M{RZB+=2u5;=2-U8kqtVUXy`_ za3zZNuiFQ-PSO`S-j3xLhL>|WDy;ZS*k8M5 zI`mB2r1R%N`M05-?8>w^arZ+n-zBskB{8*=%#VPxQw7k_9if44$p(^u!LFNpOy<04 zzh5Uohz_RCu9nLc!KwKNEA?tOMK^RPs#RUvo=_mh3~y_R|JsOX>t%k!Q#WHiQDBUW z6jLL~6JeCV4tPf~b>m9D`Ka(sLF6Q%=6BDohM&Cum62hgs2#V(M`v%?9Knd8x4xDhy0>>yE+qwv)1cz)$MWK5|&BUj4`bEB)4S0`kh_X>2=_S>$mop9iox22ne8{+dVeK-UlNn@ZJT_p%AQKKEbil${R`EDe=@yi;5( z*)Xp3qB?}jB{+!=?3=&q1ZH?D=v#-JU7oqz##|N|l!sT?>Ad@7 zjSk_nns&$rOjp!wT|coZygYWlv&K&0r3qPpr`5h0d;U>OWj8NyL5bZv9VpKv-Z^Yx zVhnZ*-TXX!@zoz678S=~E9-{VlyALnGBW#av|%iMmnu_+Uap?T;H`+8FF3ugO@f}K zs$*`U6w!ePs%BY$+Tw5KyU{Y8?
7cnMfnL?9-)hZc`vuVSzk`0={$x?>Us~%*i zdmc38{(k(QcN^wy&)n%&0%?&SR8eKDEE)7k8M$eG$Wa)G4ONj=mn{>?*i1X`Z4)(1 z9^0JAE9arO7s~Vy;;e-%3HyV3sIP55_J$|9UkhBXGbo|AHUmz1avNwp`Sd`>HEC@g07 zPtN1vGo+Q3;vY5^6Axr->z3pxdntkxY|)gTtw>Z+Nv^5x<ww4Zn9tD(%khKcjZ2_Yygd)WdTR{D z+?3MusEnnV@gCNRk)qwmGvMr%O%JJ7+Vkkg`|yFb?HF_o4a#Rc*uo=$Lp9%P5_9M? z3I)dR?h)DxUjPqpuh6kuUY{t4POHFekOgg1m(EKrK8b`ml_eV8zpQ+-6OaS)Qgu`XfMwS7-IG+7H_W)p|WY;GSUDQFuF zuvn(|B>@OMdKDKc619ViQo6{f=_*fxGm5i30fQ(ES~uT;DIX0gVAPGyJ7*y-DqTek z*^v)c>C*G|7j0znm%N!s@a~{SD*ECNgO_-AnAYvN+Xs}ywC`bWPeUse#s|o5LLA=l z3m#Eg-~}VavuQQ_ySASZ0_s-lggpRt-#E%-17-V%!148PY@?5l(#WkC@U+$Bn z#ROtD0kBMEdKIcifMs)xYC}gykP)}W+BwUD}vtVh;oV>PxHa0o~n_t`#f~_d7-;Pqq ze|6qK{<>Wl9umrkvSRg~G!!=3`Tj-#6)seHU@|vC52K*ri#2twCSC@Ezsi&vukTu% zJS3SH`Rd`>EtU~X8*exop1m29Bmn9f?{pPCo?PYZx1-ApT~D*LZt^WMHOMS(fjhli zrn7lDSAU`QE<+cjB#T>`c6*(aw{{hl5BuhEH%?YyVe2`gBs&RiZ<0{asPTrz6i#G| z7liO#KbPcoLJhUg`G0yWWlFHRzo|ftwy<(6sJtK2jkkRDRVCYbd)qnUinWm z=Q6?rtztEh1+ZUi_0-g&Y^*Oq6m?Fo-;?}?%c>(~D=^mxLPnLBBo+OL%iw?%4ZI@& zuc2iu0aLcxP@1f`cAisY|4EMG&;Q`|qTti3^e!X!=NeAcFc2}*Ni^g=gBOegM6Z-K z!mfUK>Q>}#<7MVc{IMk;*IC|tnRL%C#emD;pE}V^M&T#VMmZvR@ z>8(8z(xmq|Q3Q8VwbRTwKWHUr1ahU9rmW;WFk;s_k{1H+6njxy*r5BN3+9AdHLmyL z0FCGQ>n#M0V!nKFA<$uy?>dcX!}SWZvfW74sk4%65ZjX8H8#JJ!{rlO$_oE;&@PC! zq2o8cF@S4}vxHFDE34N%M<}J+*M`g{2p!w1Okj%8OalIU8~o#^?jeYV-DYsw9hYcl z&aEvRr@Ogd69ugvU6vrAIiG*Ynx7zK?czo?S^$^r|BGt<&`6vfUq%9myML)L{~m^n zGqwM2@B(VcaOD(v@Q)7W{0OXi+=4W6yG$_L=LFj&N$P2X#>$EKWSQu}qUB}y$reQ7 zMq$t66!ye4?B`Po0?Y9CV3F?@RK!F;V3s6BA$ zzqW{zf;agS1eLe9=86W+PG&QIDoG_5(y++>N2Cgc=u%~5x^sF`wS$uQojsNKS#*q1 z1?wJPHjy>|w%@l#3J+cA`e0lK1(ZFSAv;1co#jT) zc%Fe&H@0PYNuDhOPb|-XwP1>Yq%X-_hmrp`K|aPg^fGqy4U#@W-}&bwE?;wV<~YDm zJfmIXkCBi>=6)++q`-MibS5T^mGjyWAoQP`S1Uh7DqfA_}{@aT_eFStD6wijleC#^EbgbsPqIWG%t30FtH zh$N5sG&COVsf3nD_ncXS#H5uuO=YR*YtsKI4LkjZ#nMnUZ{UhXRuUkVk+gO12uSEK zv)R*}L%)BT1Z3c!79BJQy~HEpLKVmhkmjkUfYoTle~kX+3g&mztgTcn)i<|`RcPTx z>01-L^{7EU>aq}o+TXlzsc=!N0nb^#B#i|-p2vn>A+0;atGVhnTlVf^{z4bA&@1bm zLCE{93_m&BPn=BAnaUh`LKS8RaY_^x*CMoRT z@(`gxLS*sokYyESI?v#39AkzeWH!ji8UA@bh8=8xHwl}3Nn`m+>!$7vDQK6=o4 zC2ui|=Yv=){I(@lW{8%Nq+oHKcjACp(Yb8}We_(oD~z05l_M2h2)pmdGySmsJ#eQx zc_nWJVH!=HM}OaQHid$+R+YpWV<4Ik(v=B(wHrTd-&AuA9wStOOp8 zdqjiEMCikmLF|n=w1oZD=v*@V_Iu88IuN%MVG7IpK>(WE#-okmIZ&F4hVKSjpqGlj zRgwv4AlpKpfV$0I>;1nUU~?EZTPYhIPnyIEaoa z;M>!AZSc(Pi==O{>q;Q87k+aNXU^J$;7~{gZoZXoqild#jEl5d9eJ;T@ifHt@9>EW zb*vn7>*G|^jy>fd7qXWfb)0*<$98${%J%@w9{URjwHe}|2(*U0!|yg_7tp~D_mmuj z!RdsC^QnH<2)H{rqrkmCodjGwxN)8ha?5DAB#F(*5dwej2^P{4CH&DU+3~jl1szF0 z`|-MZ>9MZ)AIVfpWb+@0#&wS~K8}rezf9v*n|;NP7wP_-1H$q>L4H z31_5NByH#Zarp~WZO_Wp&t1AtNw_BR|9u>?OWHpqjCC0{Sc^IP`v~!Qsqt!NHEofaP$Uo+%Zp#7{-* zZD}nULq!xKdsj4wQ_z*?J`HV(d~T3$vZtbXH6lD@%7C3M1p;N}EOWy|mLb8O2(0f- z<}M?W02i+t&p?NhcV{r8yY=)`$9Qh#@_|e+apzn4HdGJ^eWQc5eb9l?68Y~ZmeC;f zUq}NLEzonR+i@d< zmM=x)5hHesB0M#qAqJm3*bI-n*{ARqV6YE-9C>_XO=3-s1D|(UxP0Ow+0Wr zvvv~@An}qjbDSQpdj_0+A92OngR|k>3*aomgPxsY|MQVXlT<8lyP$|X2ch!TuR$w1 z8rR!^ZNr>h&ix(XDt6yc=MehYlre&QPdwf#^JyJqe}jZ`3&oN=%dKcEwE`5f`Nhp% z=@CMAJHxTX3$F`Nxr9sZjHU zIjCVg<#b+Oa~-88h$bp+2Xg`rmh|2fig=MJM*u#Fx3&mV&|kYmBFOO0TjG7qH^HYW z#~3NLovNMBphG5`zm{kb$5( z*D;)TxrDof!vM2YP$9AkCE-nYLLK4-@3YaZI1)UH9tnnFf%HObG!7@@G{^gM1-SYV>bt(g|< z-;mOi`6mn`(D!ky@Am$xLIz~La?TqHv$P!)$asPOj)2gr)Uh6xj3EMPDSk-eQSQ1b zN(h8IuZ;h~^=G_hge9u0(&l!d8v5C`0Sy(U%PD~W$ZxZcUz7xRSU3{U{NUPm?B3aE zf5DFv?zR-P@B?W=$(TsQW#)#uUq8^C)RPF8^yw^sx*H}|+@jJudwXPGYN02rbPkr7 zKF<{U{IZ|GBKp0=VdOgWUmkTAh)C>aGeD1J?SNAkS7%`>`d7A}CymFM(@c~v;?EwiI8jC7co|B@lNvX5O%X^DsY#aSpx30(<)_(?Y3ZH}vY z`IUMh&xjXtn2)O4b+I7s?ucY>D=nfVIY>p1 ziQ^5VqK^}mrjSD+(`+jJryab03mk~R21No9@7scCQ8>(*_EBTsyYo=K#NY4Jf142t z`Ysgvz1tLk_PstAOcO zj%yk5on5#q0rbh1u8D$%j^!zyXewG2#QhNR4OM;N*aqBwu_9CUu0A+CWb1HyVr#l; zvfigUWDL1eK0R}ZA|`@elIBdzuc&G)N4O^}VhMBCfnBioo3fZs@*X}xRT zTg%#|Q7CHKWwa(%Xuz97S9|bra_cQUm_4cK*K|lb4|~}3b@Q+!I%SlBoR)^6YW&-h zLj1roGmi!~MZSGJ!1e)(559tWjk3s{G`eI)f2k0V^hxIucbEoq=2jyxTn@YDdbL6I z*)w0Xr0L)bIKQM+rYl`k#m8=!tEmH$A*!`kkC-eyUDQ#%sL4&sixVhUSzX^gvDhOiN*-jGH?MKXy7p zizSsZwBd(UorSM;siGg-@17s@(Ib)ox?eeKKFCjYt##fHsmw#xO|tNQPSF-*%B`kf zOu^x^V2UxvsyteEnV;7BY1#EE;K9yvi9|r-S_QvvrSuywotUA|ljqrT7sW{NkM$AV zmSvPfKwa(?^6utCvBfhm^C91tO)tA71^21|h?U2$*s|WW#Ze05^T+xA8g!6O&~m@yF?y4D7TTQoQ#5Q8u`2&yh}3D>?NU1+ zVic`J;hjY4WBO^`(+;)upT+f3A%b$)jJE3RhN^ngFl!)S3-uux}JUb~drx58Gf0}cg( zPfaB1U4MAhy)^cb6!=$FxdO8Pe8jUwl&bzDefe64TEqT_c?4TvI>7Py$6sGz=F`R- zD2ppnrl@f#&PZfih4P0yImC}3Z~tiIUUR^Y`VElN_C$=F*08zplr>xP!Od;Xde=)= z=`I2T+zSGEK}N@Lx&q^1%MLYDJ<@yQRMEGmR;*U8pL=4y)XLA9UiD|g)^+T}w3+5+MET!@iy2Wcu)_e(_be6Wk+SBCIVM2GU7^j1S z*|H6dtHsqh74IF;k}_Q$^DFbFL&HLXonr05;@y##=dWDQ+1i|CT!4ia{A=Fu%Q%&h zw-qN0{ikTVbe&_f-!GM&oW|6|`(Vil+25S`*H=8_%q+e#Id?Zz^bY567dZqc#rt0asNUqR zXFwQuK^Tg;!>a>I<$MC}6yJ#Sm(hiueOpOs+Il$ky5NjIY8U$u_;-d_^<9Qk1>pU`ru@cCkc4MzABF)>NYK!sr*jW6eq37G$o`sM*)Xzx@C!NaJsN?Sd53o z=EKnLHCBuB1l-PTOK%hPzGn691uV_qWX48JJU(i4k7;rZ`peA0s_@ZByC~bU$Fc8` z*Kg6CeUzKSCqq;R$<0Dg&3(Ng*D&6_%gC*+IX&iJY(Sp>K5%oB39LotKPFFP`{#p6 z!N@RBH{)qyS6QO=uQxTZr!DK=T|)R9tFTh{q~-3vto;D_G)!H2l zLOk`?y<*KsMSr<%8Kr+~^RSLL=!xBbR&!`{?f!!~bP!uiJMv-KCh;6}oQ@Dsm9EIi zc2099ELhx&M)|75y2=UEXP-7rr2J`Kw~SUL|5FNj$bQi}BdSLk5q^uC@nZB-oSPo@ zD2eK_S7n~|9@yCEIP=`FAwXjimBP(TS(^&Q1D`YRNl$71SFG-%D6>&P3zeEj7 zb-M8S;EEiruM*K>e>-AT@!z>FBpB>D&u|v6TJ5e>IXnOhec)ups zGl6s7bLjYQiyd@nWI_1n^%mK?NUKkm7X-ewZck%shbg@`S-AFIPHWPcik|AaKHJzo zwkf=rXSLuamZ7OuJ*Dp|xM)Un>kg2h(9(5x{xQ4=a=AeFMDs#9Rnw!oy8G_-$a6VE z>gVpbX~IoL3o9K^>1~ByzlVOE?;9C@uH!iYU8wG6pn8;DBM}q0L)nN{P~$uKFb^3J z%Oj(hHstQ0g%Rb?bQaL@Nqiq_=%HJ(0rC@Sp5&C{_OFDlipq1i=bKn$=pDp9An@;6 zZ(03?V%E)v$DmUBiSCs@abg!E-f-_jVzHaZ(67wk#jYo3rav_U@7J#x1VwQNuE$&7 zeLl{Pi=TSt0EazgRa+9xnKaOmpul|Q*AyAg7;QZiyE@&2#ML~>@%Lf8h@qH{swV-; zb?^Jmt%@BQtUhZTK@4+bcF{}TE|80?5=o=n-KYgsmSRVW-(cSOSU8%i5^Ea5U{I}d5O5|PCqZhoh zZNEm1$Am?yOvJxWF6gs|QMHSqr{T6gA2CFnbu2fHyINGh%}*c64RT^X{SkQfUw2va z@ri;c`x%gYa7DpEl#W7d5E}?l|IUc{+K;>X(Nmqj{5L5cm+Ga`OG&ieF@NFvh*LM& zkixf+KjwTc3FaANr*m_M=G%W?UwFj`a@W3zjmnP%NT)~ag0a?(yy_~ywcXiIG-nJs zkmuY)ChiSMxTi7&{p?EK%%F#Va@C7kxM$9juqlTw=iUlKsxK%;id$8e-PmWKQ|d|< z2-)MVCgfBW#v_`!gugV5m9@&5q}@GO%SxH~PZec4(U`z+vM8UJaOy%ohOsV% zD+p7}G(9wDp0Trk-9k?s{9Q8tXb_cH6QgfV;l&wdqyOkU=8m{<-U-^tA(| zE2s`G%1~&XBl6bu3Dq{pRWioltZ%hv;jL3r*}#G1GycLt$Uf@^L_r@F_=8XFo$nj~ zLJ*glxyKyLTrqowJmz8K#taHS=`;H{09|Y}SH>c1Rb;^`Up5{Nip zX5^kk3-~7A9FfyEgFj9^Wu1dhcBgxSzuz)I1mlgXtkgqc5)RcqPG657p)e}d4uA*C zLB(t!&U&P4DxpWZIyjZ-qybi+-PFll-GctlMrjVs{^v%0q`Wy>IQ{&W zgl$hry87&zxbL+KU(RjCt!SsFf!0OL{&WzV!FE-Sl5~Zq|J8OiuKp#T@@QyETr{Vv zY$G*HolsNE87o;x#w8{A(;$5+wT2=3*SXCX(|xhg2i*+y**#_CULFuP8h`Etj)rS& zH?2aH`$g^DrWKR*b>d-sB9P|mv%|=TFaL(~dVouxX$n0DO|Zk;0+f+bZ>o>N$x7Fz z+DfguMqvK=hbS*|Xr*K3eKuvKDtbxaH?qU{WkAA@13>IQNli$k<2ug1X=zjhs$5L` zjocCZew7I<9*(sIIBIPxmm$qfmT_c9{Pd@HPMxk7Bd1mwW%ASo2A}3Q`4*7jNuM3R zg18x7IqQM(O?R-pv&RM5#XHaw1bGFvIT4V3WowabWmlIjBjP56QU|aBhc2)bU!Bh3F1*u!*Ap6pLDOCzDgmIKLJ`Ms!Ii$ z1oVQ)0^bM#M?kp07ZKyOI6tkj>oed`LX8>NbTu zp}dRhxoaQ}q8tD}7DxFC$hct3xu2N?v;qOoM?sg}Bk}UqH!c$&4>MEI3>w#3H=nZJ z3C8keoT@ztyHt3b>=YXIGx^zN=&iof*@xAW8x0IkbM99!R*ITUwKZjru$xS^%}7y_ z)z;W+54<&?6~AIdQW0NRoVMhp=NEXPjP?N&BKl<;b)x^lKXylSq=TCHG3G&Zn^5;< z=-|J4gJ7(AB%mK5dJ{Q^h8xk6q>I|EbK=$AnBq)G`%=pECph(g{}_7#C=c7F+r#~R z5EM@|1)brAHOYgXJ*rHmm1-@8!f1#%H-vw&3%vWBoEter;cCd5{ z(6aT!6Yfr>B4j%6nS>*IQI^n{r+x{EEy5V}R8K}UG z72naL`K#Bu%d|}nmEQKGXy4aEsoT^WB91YG%2xhg#``mZW?oFyIjbZh>hYJCM&H)` z*FdwxOMVFCJ)hL2;Q248p0W*<%m$~VoI1IBL-5PUL!RQ9GlnGj{wC5g+RO|c7QW@)6*E2K z#L#ez33;1zXY=9Bs;X21p#N<_pQoB2KlY}k|C*Un3p^z_LC#n2@k*ijZ>#=r*Cc

A03aN*1Cr{V-)G=-LTDXbc16_bh2Vz_nY1tO4TY@(Nt zjF4yMY)!ucwbQD~(e-5=6tMJ#vhuo#1b^yMkLT!lX}ynd+DK<2VzNXU7F&=H$4nU$ zdXSidUr3zx)M_kwOuC$|zESS~ZZ*PT<=4RuvF6Xry+`1P@xR}Z5b9aDJI!8$&UC=< zP*$iVTqIl(CEXqcszj4YuhRJSb!Akg=s9V(Nd!X1Z%&7v|nzt zAa{o3+5BaZL>oF9S)9Lk3{P-U{PFeU@A&Dy=iI^>>ZBrX)i-nKX3;mVoxD0L%t^R< zQid{RWy#P%Y7Ij5$MS&}E5q#Q3*Es3@1YBW$46KzT2&3YKTbzU|Bm@(?55WcQBx8y zPztQXnjOQ<)4J$E;pfYzqh*ol)gb`M^D;@UXbwG}2gFZ5S;GDLQ;5r$Cag>fj$iBa z4QA6pmUIr&Lh?sT$rVpekC_A?E{P*Z$<$JUrPxB+T4Rs z8psfgEnFAL_y>{EAN4e*p<>(jQ*-bT=e&f=ZhDmeyrbQ6NLbfO76D$bCRx-NE#855y6A-`O*rk5dM5y3xj#Jiy+&ig7dJ<$f~UT8k?UOtBZg zDTl;A;cd6#4=g4-o{k0hF5ifF$O8#>z!?;SMhbA$PTx<9urHg3W+IO6&Wm=Gl^Vf^ zd*^@cQXxLe65TEIADyKE$1Mt=1u7BZUgb!4z(>7+r#ZFzh<_(=tRpWWWgYrj6Z?h# zezhm6lcezfXnU`yCcf`|_@gMGQ~~KyrHRs!PUuydsHh-_fPg3{5dlFWy>}2o3sn?^ zsHhkzk=_E4CLka^K{|wzgvrc5-;3Yfd;6|+)||EGoSC!tK6^j=*(+;Kp3=gON9v2n z)prclp-1niu#`-~)e`rAVUO_~=lcB>4FE@@1*3VCSNazRviJ3=W!Fs~`$);KfvlwI z8)WZCNt8_=I&MRpQhJlW6hVP@zlP2yDh9Yj#KU?ej@ub|_Y2}3Q(wXh=;7;Cc44JY z*y2sV-$7ZUOa3!v1rWU7(s2QT|5NLDdvJ#u?DQ6-O*fKXjlSqek8@l@i|Y9S#D7sK zr*scwYQD?4#t7P`D(;N-gu9d>$64NemxE}iH?~#`5V7ju?54#GnZ%Ym>-r!Pr7zbR z^qWrd?9LKB{5-}@=xnn^HH3{y9W*lK`t5nc@yz)7 zyK%Bj?&mfBRu+5zJY(87HNpO(!-e@p*fgPQ<;%1qKsD`XQeR0Ti9m~wx;=T2!7!z@6?|76nn zlqs8^&1be1Y8Jb<`}Cm^fcPpkfgjSm739VXlI;#$LjNYT5aDi6Qiq0Lzu9DsQFN2C3lkfYo$iqkEKI4=FaA0|Id zP}q#1YTG))#=lUXhW`TZF{~#?CY%ZOq9gJUVS8;Hgq0)ES6+0aAY3`lP_YB>G?$n7 z_ifY2)GvgG$m|#Q@w+K0Vthk0+#<^rSBGZU=7|66j=im~-bh1s)cK~pD#5(+@$Y>r z$zI1ZD^}fAs|~kXVClJm1u6=gk(sh!4vp_p_4xb*h_`i-JeOJ!K*WWDG z{H3}^t#o;*K3ph=`EOpx(9H$C_`Hxr&1=u8U%uN#6TUwPLldp8ghq4O6~Eh3c7#2j z>|f@4mY{kxzS3Oz2#;ZCMlRxBUCc%B56v{SE^e}VZ>3gv76oslqphlDzen5*9nzYC z^EfUUy(K28&b-O*SQIH54P)OCy0NM5lSVmVbE9aKFs7~x0tVxeP`vWL zzQzug<6mKTp_B}m0^O?8<1@C;r&mfB&or6rDr}`F&5^v}_ zZLJW?FK<;huJ$uYCx-a%`o_`S?=2fNH1E>f?BLq$rB~PXuuP&vulsejxS=JoC|Rk6 z7?ZK?AUO)JYKZOcD59q6g+s!wFHdg2c#h)0s%~jqJPLbUInMeb&hgdcPs`4-T(h}uTv_Tb z(H9C4Q<*Me6!k|9xXGiz!q}OW@qzdF835*?D#^aS^GEU=zHPLZXV50dwa66S5Ci)p z-4Cw{JbKc8zAi8@@9RV)Tc#DS7h+SnH@)c~^dOA6ViYhwm*=%K-U~Tc z-^f84yQ>TcXfIMc{aC7=ye9V7xbPlX!Cf?s?###!+Sr83@erR6eu*%F+kb>TWEc0m z)w+T-7P^xqwU`0>)FCU&dsODBGIM^=rkmONH*z|nwgk@e1Y{lZRaR*^7mtnbk*ve` zwoft^CErR852@F^0vPDQ#i&fD7qs|pX5sQK^SJWz9tl0Z5pebLmW=AR6QE`Nz$9ad z{2=VX$1glZ$&H;dH{rI4Ad_vLPg+Iqvc_sh+%I9#1_!>7`l7$Y%9Qe-F!OIaXPhM> zS`QxiU#)TKy_lR9a`m0?7Y?pnKz2CH6+e63cBG*2Vfb0Tp{}@*P;YjQ+IzFxd|bPiRheQDW^={~nwOmr^e9RW6TToTzL8v<;v?p?y}3OzAwjzTnoJMod{JV~ z<=mf`km*w12ygaek4{GR61aL@P6-EnpaAY}kHC_>LqnGc1xwCJT-Af$f1Lfl zQSRE45dW8$wk#9jp@BLrg}W@gN4Dn{&l1aJkrWeQm8&O+&z+P^djtyLcct5mwtI04 z9zr)e{-^eJPTr%t@Ffy%@VpnnKEHg4ZBXmgmx|b=3E?u$>cYokXI0}-)t088zNcSo zD2%;}?nZ{ir@kyrG5~xb?4euMRNR3Am5oVzO^oL+**ikg;2#4F|5)4q7#6O|Cc#iP)IKsni zl~%!m!dEO_nAW!s+5h>Pw6$tgyEr({ko?vdAWu`EC~M{JL-r^{_%mCKm|D*!Wk6d% z32ei)+4e1kX;ty6djk> z)x1;JxOH~1TC8aF=N2FE}yRJMVxKd zV%sgl2wb(;vfoLME+^VQ?1bNo{;bJY8q~!3?44xEQY@Uhaa**c>B*^+_@B5k31hCg zB9?k)JmyJP-}uNjZs9vmP6F|f$sCnX6lnNVVJrQj(?p?}aG;?H@S(Pgzej_pmPe3g zD;iyJ;!me#roye>(ur)pa_e8*%d-&wSXVM4YCW66pvi}7cea5mBvAi0L)of}TIocW zDLG15z0ct8q?zXf!pDHO!b|4Ha8i>YsfFzQ*h9rG3jTw0@5=ezY=Pa4bhv}n>P)0h znw;$UfPMQ>zrow6KU0` z-zD%Tm!I4F`iEHNMkSEXu!-C07df^XATX)bKbT&*^b3dQ%{VK^60 z49(nHb%rwxep+<7AvH>ZPJ@^zT;!F8zUCNQj+{2y%GH@t1I87P7;Ss6guv zphx5aYK+)*&Cg#XD61RB&@OkOq$wbN2YsnTUQt7gPlf5tHv%8ixp=2M)dLvC7iIR= zUb~!worQ#hvJPFecnlM{cEfekvMl`4avxnlGNkKsz9SGCM|e^}d~WpT1wX)8O)spI z1G*9mzZ;8F99HMT-_se+VLiY)P;k6shF2+%}Bw*I7KX}(M(Du}lHb%0sJ6*hRftv{Ox z$C3Js)HFVoGyPI!YK@Cjm0+))W0^1N!zjJGGS=hvnl6TqGKicP1-%{j0u>f#uivG6 zb^BW{V$r2RWh#tuFlLOuD7m))I9_l&L(GBB!^G$qqtK_wF2O#nY|yFRj+ys+=zZkU z_a3v*e-^SJydF&G{!HH?mGmr%?f-m&@mxH|?ue=$6EUit<ofl~$^?ZWXa~!>(Gz z5&J)j*b1`Yb0^ZQdPbg%#F#thR&;JU;sl#paqBTMm$`Z|wD^2%W;uhLlf?dyE$d z>2#4sB#{h*pQ$j>iH-)l6JThfVu67mq?YceQMa3JShd@V^V62O_`bW+gPx8(B6&h; z&wx{wIfD7bq%CNy%kSjiZoyiL+4z%AJ6LdM{>0oe||p)ckV2c7xpG!Zv6#aCR|Qf^ODs z9q1C%`(2M&Wk}-X$(9SkF9jX)vymaDWm*5cyw_BxKsg= zQtzMW6AsB;-Q7T1^VCU}Bxf_Fx%b16S z@p7wkmAiP^|9yM#h;#G5j44`y?+4+Z)d+moXxy%uY=y}AXBKY)(GNf``wGoH1Nh8w zy7Y1+L9;&Uzs}R;T7-tj^`My#|k-0-Ki)fOzyntYiivzG<- zqxM|hhc9vfA#E^rD~6Z|d5QWoXE_Do-KzGHnj=`bS`V--wRlI@Y453m?EwBYf$&_e z3-fWfCHJRRQ2Vb_MWcLCi6SOVD>GI4^)Hu1KGsn3SmGV^pwHK>A_|0B%{2MKQ|W0J zoQhSqP|16-ixTc>_$GgRC)zQN{rT{pvb@clw1zl>%g*seGW>^~>m?-Og5;y~hX0}v z-{z~bK#py1#Dk1D5U<8mh0mezEuqTz5^m8c%chtfMKTG#cHlBxGJ?r-uHeTBvY8 zGv<2MtC52u#Gx}Od~O{w3bBY6=-)TD0M>9c$$g#QGj=#HZpVX)Hf1)d&c3vEVP zvq?|+3<}M==t`@8mR5lz?AV9oPRHsgA{ky-wR3X~oraa6?C)w;OZ9ee&{h9~{+ih@ zY!Qqg!Q`$KC1)-7DMO1R?U<+bTp}4mE)r{z;(S!tq3j+2rNh^T7PR7VC#o>M{s-Aw z_7ee+Odhe{$CKSfb12B{r5ifj`|B=D5^X*!77i|B)gJ1ZnL?W)T zW^>2-_*EK3pE&_Ia+iNTn=a$TQ}XmbFkStfCsK*@>;Sb73|KjS zYGHxm8;^WseDRpK3(Jv2tLmW7EgkR``M-bfDFtSOcJ5X~6FH4OyT9lEDvh}YbOg;~xP&ZA z>c5+*X}B=qy;fe2iX+J~1|XjK*|nPmLO-4pUQ1#+K<(#jgKud;dMc1r!qqu@NXmxb z_9C!D`>WL)_jnGmsNW^@gxfUcqW4Sj#j z=;~Mr8d|}*tJHcB{AfVG>8apl(~(R}&O>|tp@l3h`(zEh`<3gpLqeC}DwpaARij@{ zBv?>kRS9u-Rs(Xi?hw0g3RQ)v%E;$t4n3*@O=nl3Nk_)rWK~A9i@$^?8XYCHYqK!> z+zxynG;JOVyzueohm-=ljOa;YHlaXA?lcK~)w;tntiDd->Uq zR_6v&4Uf8$h{9?;;|cefGjD%D`7%TCnIhvI1>!}H`hf=@GeWiw|16Eqge>=tA7!|s z)jp<4lmXy-NiUNp68uI!z*4EncL7VnvN||vAKDoNb}7Wgz2`P7%MeoqrONNKro}+q z*M>O=D@9i`EV@=Pu*|R13We`g@>0Z|$PXs(t>G-AYuvle4f%Sh@YNibUDCT&(GXp8 zK2R0>bobVb3b>mO!z=M)eh5JPY!74XpRddytpUd)E5b$Oe+duk+TQMtvaTpKt`)h5 z|CY`vSLPcdDMr8T&Kz=C+w6i}Ak4qQbdr0X2%a9pSm@2QXTBH>hT>U^wJgVYYb`7O z#l6A@+v-c=>Nh{qA6Z0yFmYN;9*4D6`Fzsg1~U7MF9ABse?v{nA#^C0xfAY> zeC&Gj<@SIqa;6B85@cdw{~VG$v(I{0c@g8BznlI;br4!;Iionf=oG;&SEJTrrdaEJ zEm2}@>n#cWVDVv#Z`*tWK@lL@7c2p~#2n&TuU&-ju#NY`b#l*BcjXasJkIf|ig6pH zwTc^?+2C>dB+qB7lO&yvQ>ucpIX5dHZ&%MxNmldJ5euLu*9|eh#}=7NwM4F<8~Y1_ef$`*4rPD1W*m=+r3pPc~nu_ zO{FX!f3o0*L8dlnDj?&E!VzHYrEmgLGAqax`rm?{vj8{w+ zV5BnIJ+TZ$W#~bU3V4eGORc(dYA_WKa=_R_>Mw71$@u=qN#D}G-JRDbAnyvpgK!^z ziRyvL8a)*mwm&gd=DJRgvM{1F*#($RZ*;?Z0>pWRF?84nZ3$QSf{Jj1*? zI_vo$!XdSA2rKf?f^H#XGxHh9dsI`U4f?0us)jzPc$pDIKByQweeXPkDad|K__==K zp}(ENk=p{)3$P%kcUJtBVl-F4* zOy_!YNz<#4^TQ?ZD)Po@{On&W?qac4XmWv2_D|@{(q>CWHdh;zn9!Gr*&Hc>1B=)8 z=qyckvRe@pNIv}HlUB=Fx4(dK7K`F4(^9x~)%bqtp-UvOen>c^K#>cNF`ZoV1aD^h zpxp+a8Ef5r4RQC$Br}Ak?w}fX)fcrIPh02hAQo-#h%08pg<0EBR(*b`M335g`4Qx@UQ*gP2jc%td7gxtu~_!Ae|6zmXhVWQO0G+U zdXE>^WsM$~-n`WZz2G{}y(ty|wH?1V<#&PLW=g?`!Ar+0E;uM}nC>9>vzN0mHdi+> z52QFcME*fD`Oj)eXq|x~VbSURj{AW<{4o?KEGG5x*t+6B2zOG(T!m>PM>i9PnTKvX z1fp?~C#U+5pC^zBRzR76LM z4e^Vw$EJSC3qm&OOD@waMA1g>MI;JWD|M6BE1b! zvX_vZ^%GJ5N44nsPXxGqzAdE&YIyX(sr+qt>OO*ZPrY{re0l1($UJJi`S9}5LtFNz z*DsOrlIVEvhYGnFRZK|hQ7V_sg5O1W>5Efc9q`5};mdlxE;WPJcz+>Zpy9Wr`#dZ_ zcxe+F%c0MnZiwUBB+_Re6F125EJ=j#3J_Ie*M(v`^5YAbys8xd5rwL>7(vPz+w}U0 z^WeXBqM3|1+}E|e=xJai1l%-76J1UqPd>um15bu~`@;=Gw(93M_4+Iw>FV4SkJ6oo z!6-i_;9CL-{mTqS6||KbwWh=0}z- z!(9Gh?~X|pV{Wi<-raXgDMjr)^vACxOToh=&}LZ}+WU}NU;Mr3{;mw7EWs^XFvENS z8n7f9G&HbRIJu7A84DL;dm5O9t099f-^$$n1E^gOH(EUf*&m zY{!LS#O|K@6QS>J&w(cTXTF`<#Xs?4e+>y=tAQ9m5)7sG<(EA61cWmPU?O|x6wfs8 z8$dyi`;6lbi;g!aI9%kd!N)IRrlFxd*__lK>_M>}A zGLg`ZeiVKV{H%~4zhQtAK>X98U}d;lG=LNCEkJlJi0aElLgDqYm^H75>}vU|4A8z~ zoyk1v*h4Ub?%OAtq`4v?RZk(sk$Gq)I4YfA1$xxd$pjHU)Np4Iw=hn$H7r%`J`>F; z^Sku_KAfG`*Y|i0={`_oFlmEL@Qe8uoCNV&^{SizO4WtF{Z4qZ%6RF2lnI|N^F79zWi&q2RUQ?-eD>PwVi2p)qY~IBc5TkQg`8dfRpR$+ZQP(fHyl z@1@%(k=Aq1#M@v?H^lp9wt5Vh0hGXsz_Ks{sMTlSL(JN0nn=a;2W0FcYIe7(C@>}r zyGmXfShMyM9vlS2l4YXP`H@t`ZkTvg-+_!;4=t4K<=GPWy&MHwU844J0pez}6#gCz zIeiCdormGn4QAm>l;R`9g4~v*%+-4Ic(fczsL-$gHKr2y^!K!VMC+nk9y zM(?eKFh-#R?thETzFgMK@R`51vJU1#{p_Jv|jto7-b!AjiUVmsKc5PE$@5{!f49 zmMqzun5R#M^>oMy&`>Gz%HY0RSI!;x3rGa5KgbjO`ZIv=4NY;X6f8<2_OGbARLfCe zGDVFVP)#3;u?Rl*ffgCoJ5v%8gZ*@I9z}p-e$hK(aX;L20X1I3t|CL!pUrT|N7!6y z`zcV;RO@_8={Y1*4Eyv^a|fmIw772`X1BM9Drj!?p-rf!(#6W!RUyOtrj-n0i2XS0 z#0b4$r&{func}0zj{QZ5*(jowEzL;;KRYQ5pv3kO`Y#D@$>w20E!|Cd`-pxLeYf!I z-)(uwuz+V5_7PcDrWr%Rt8{E86Yn*BW*XSUT<<~<41NQ(j<*Pv#*}C}BPm6=xp`~O z{o1`uYYG6e3n7=gpPcxfcLTLi*;#HTP&J37!0O!G7uBpbpcWuCA2LJLj2(1LyO&L7 zWduowfk`bj2iq-lCy`-7?Y2RoZT~yP)#)Y^_2@$KclP~l`D3NW!zE2A&(t7{6k6{R zyy;)Fb^|?m;ukmXsmS=r^@I#UT7$6{;q$n`opd}d@*$AwgU29Qim!G-Vr)8|oBsfqKLur;DUB?a$S zb9xnXTw5bq0*4o55unF%U3gO}452SlO5E}e{}tE<5AI9;U-E|E!5qGZ;M>CwPrTfU zc#A8Z$25ImLmTeQwA8qS&10PRan|85dd|PB@+A5bgggMyRe09nVIgvBZ>Y^WsPV=a2x!n zIIQ@~8G!h%V1Xx<;xziNieZ0dlUj6%zA~>1GYNfI`^wY*_-JOcLeNtoWDC^HT(=;2 z4l>iDSKw1*=ob5}H;)un=hC07X#c`)I}Zk3gx@!FnJ;OQ;;}eKI}SdO?WiKdPIuJ6 z%sa_MV?~I_#%NRnIVuHt!OOLRAiqR$ZW4O3n!*`JnW$WB4l{(l)_gvhptbLIlXqnA z@9`B&ecj;Ns$Y#2Ms;WW-j@9HDf8J!xPrrG1(521B^?soP5kj%lCnXYDJzvbc8n08 zr|q|*L34D6v;N1p_msRcmByYug*aT_k4-;;%6+$WZ#U7b_t1}eK~Ib-Q!6$OLi8^#L77og{1f9*-=sI$bA2jLSp z#mM^OaT1?%!!V_5(DHtfGZp%HG&}5Y(|P82h*edP^khKBD)D$i7ui(L|vkPz!M$`pK=krjZN(^o}WN@t$}ep7!p z?xC!=iZF2amKL_*2#u?j;uUnnt>}UIWhzYbTGdpygOH<>koUG{$WYhCpufd?r?BUK zfU33!*jtKTfa5jw!J)&Gptzzgk+(6x@UMPT9?$Vshb(ybsy5kyiVj=Ql0i;4huP4L zGU2|W6;H)0=vS?Q6Sb^a^)A|eQ0^5GkY{Wea-nARgc3r`*V4w!ov28W8qgz{xL+h+KS}Ytyy7Kki?KOvtO z4*|DVM2F3v5bwURARJ!03H_UXmY6DC4mK>x$yAkts`DH-5C5g(BBR4EInvOJa+2Cv zi_i_cl|&XUlA|zt4|;z`cd6gQ1cr+sZvqF~?kY^TV&-8-{R2=n|J;Loa6-T}`Zc0P zDz1NKs2iqO`)qGW`1?i>UhL*5a2 zao|LSD8s!3GSY$jYTafHG#;SnpsjxzAUXO52Frh^-1|n3*KqiUV3VNo6komxQP7fc z*B-kHLj3y%;<#-U_6iVglZcYLYD0!)=DctoqH+XWWp;`9E{F^`rofJvInQn&l>IY* zdUam`srY*#4R7>QsT{+6x5MnAf7c7HeP8kc({o>FNV-ARUs@swis3@>l$o0hcQ_A!uQAar~Y@KDVYZ`X=d{a(mu?d3D2OgOlg;c+3rD~d>KkS!n9febxpoq6TGD=2#$YBrqUJr5VZCp&RfpB5VdY9ECA_xa({&P6O zeprswNt3|pgCtcG9T?HpOnlN`XLrUP&BT(Gl58mSWq;E)*IQbKh!;%uv0E!SU0bsrmM4gv)+T;nICYd zAr{OPXo_)qAqGjnqnW=Uk1sBv-o|Xj@q<8|CnI*8o9(v+1%FkURY+t+wE`ZS{?0R6BJQkZk!!vh+H z)Un46$`9l3)W!SWcsZj|nWloR9qt*j&JqmGUqcK|+wQ+CB@-D3x3vc(ZiL+sNS~q! zeJ1sFe%;`-LHBs`Y^-TNMZV*PM0OBVAS9X)+;U1xWL7A9@p`UYQ$e=t5sR58#v;(; z$!|b3TH8rXV+~d@w4fA6wJQ`cIeJ|QU)A}1I2$0zu_Ri%lGF|JgkxDqeDX-GPQ0gp z!-p?DO@$nqssm~0@$@eg7LHVEdOE}akUp#G%h1go{PKe0lVW%KlS`u6MoXgiAT9Zz zN%Y^#cOt^Q_4l2tCz2C4RUV7cVw&Lnk;d;yOt>BvRz z)?-tdjNP`2s2*_EYbGn7``UlVh>Gqo1@$GcS*lSuM0kUN{MnmO7mO3z9=gHq?>_@ z#D{;S$g6ru>5CzaSeB>cW)Hao#y3PtqjuBXQKR)~)`_S(HIYKzHCO=YUrT280|&?# zi2I<%360fRs=@fcbELe(qj#v9K{r;as(5^m8L0YzaSU-0e+&^3F1pUW1{>Uxf?v^>0=$_p)$i9y5qCAzN1-^br|jYySbHIT6+@WA<1AB@#>EZ za%32K*%U7C2Khzb=>KL6z`IIQJimILfxHt3#CA{~2&ggk(zkfBz<k@4;>q!mDG8|4&~hKb3>wHZ6`&MgcP6Z9J%;baJMp!sTxTAV zPOEeRRI!6!4RoR;h?Bidq+4C7ZtSNkdyO;_DFfD5YpxI!p)5(O`!sVLqVd|7;Y$gK zx3}OU^lpu#`_m&GGQO(!u}4|(dK9FJH@nb=a3$xnGL!Nc!}t5mfh7YeI_5ce^)7GT zJHoa$40G#%I)(lJDTrdF?!UmOGEIq7qj z>PCdJPE>&%)t?=BsnY@Ab%hwp04KkFBH3Hjfut!o#|WTu!Owh!_>DPWyk) zSluKE!C$%ht{kB1<@wKiHHX$!-uNjari{NRrace`wLT@Iv8uvJ zGEX8V#1m({!{|3y2K#E)U=1RINItMn92EEXv4)_%O7k%ck2nxjNfVt^t9AaZUM^pT zb5n`!Dm!*#M24s4&y5qSm==l3wyrN@AiyE$OR(%&a*vpkgfD3>lQqU3qD18XXTnWT zv=(SwmXDB5-a_mSt)ih1MC1H1M18;bwL&ca@`_(2^tB|f%8?UhZYRSH5KRvj3f^_$ z8AG%BDmw8(uDHciIvenA6ue@c9xS{DW2_>ngVc0lE8pWACk{|f6-8Al_1_B?g?BijDwC@{BQ*ZTw1~p7uou^FdA36u@!5 zEW$#$ofD;C_DE_b>%JR91k>M=(|}_%Zc3btXYu;^b0>O-s@O`lrdvuH?#K+n8I(g6 zzJtk%`mDg{Yn(?J89z6W5Ics*ymA2`PJdW_ncw^qHtR&X%1E2=V8}`_TEj-I!!Vur z=qkh8%rvK`WJ3i&HEP#R{J_m~l%ROaienxY29Sx7PF;`<8K0Rm)Dw~ql^{t^eMoMt zxA)x~azxsJDq6UJ{SJqlW1m5k2nPbT9YVghsGWEMexAf5O2(7S{6M$}AvooUYcyrH zrTOhGdgpw;_uvF>n@Vs8S!hnFIy7>;sQszBli)u zzF}m1@{Kmh9140K)r#HGbn5{8ucSM&6M~_oXWAXWmaCc0JS^Dx1Lv@1{#-6g4YV=+ zABdL{@#e;DE9mr1K$p72=rD3@uEfc;OX8FgHi6bLS}CQ@tsJB#%d30k(1AajE2L2Y zjXw<$ria`0&;`+bEaR0`pR&5`Xs98I$q4crG?4KHNnRrKH|3A8`u>8bN3wBxKVi=1 zvXj}4P|${qi1%fx+dybh>?(pSU51a0U$RoodLIo=_?YQjhXks|2DFad^6(_hD8$gN zW^%)^n~9<}m^|4nG@Y1p)};emT&K%vYk6K3D5_$!CMvR)#0!1_`Y~V z$S7S98y)?yL~z|lr(1|C;-h_P1sOl^@|9>5M7pUO7;H?dW<`B}jrbk3+;sv$liTTN{Fe9&UnThM617{|3230O`&Z8G=u!#v5jorfZnfoghGk^Qj8Kb zzQ9y?R~hW^7lHdXP6z)7X{AuC^{HrwMBTJ~ECFjwFOKs$ zs^_8byH$m;m%H4-z0>KEH1sEN9Ku{?SCb3;vU9JSp2>woZv5}>z~(}W{FSiN{Y$0L z*D&RDNdPr`r%(5n)DD6de=!wdM%t>Px=mkI6MXWAURes;?uh~a+#0*u054pfde|*7 zMXH;J@L7p_pPRxf~SQBr|0ZM!!&m3 z+5PWQBxl6O^1L}f!)$!n9SpnwkRkdfLSO08#X0+{(49ob8&K`txxppNBdk7VR%~_7 z4)6cCES?8^|7-tAN%rXxYyds4GDD#eOm0x^sET@8j405yWI=8lHo4?Uj?9PR7GZoU zz<=aw4KmCd+&RLgoi7DdIjKiy8nxt3Q;(XaYdWIlQCi~kj4dEW=P|MjouQkN2;Ln{ zdO5atDxHCprw^-nb_xprY0ZIi8`}1grD^Tp{q-xYAo&GW(e_IcKq)a;H)9n`SI6um z^yQ{GLmNzV?~f!vbQ3U&hIAjfJ)vu*Ksx%bReT4*c!Btgt|P4w2F#nQ+lJ64Y-|N3 z$TC`K?fDS6$BTxPpt_Qo5u+baZZ};OLq`!ND*6R{sT~Pzy)yqNO1D!$Zapv683`#^ z9_<+@LA13X5oQ9GJA0`E34zr5ZkX2K_6K(E1C;N$!18MdATPf0jDm`S0vVCj?qFc* zQD_yEZ2fxSt`{H=Pejs=Q<{8mLCSG+d3JB0{DL)I7b>IKnNO61aK?t5A zU@!1G=RkI6ViEp0!1e&U73ze9^shHB9r@kbr^By?Pee7V0pe+P-flYGIYP&#BRNvuO#W`-%@{!_OVHzdy|TGoF`)h=smbQ*BX-(E zaCUJ{YruiZA5g?)XGqU91#tsfpTxh}n%-DGJe>9}TZ)fp*#ak4lHb4@<@lO2oi6u7 z=2W*~jWCsvfQFoRFI3x606mB~sc4S#in`Asc`7e{C`ornXZ*)ZWjl)Q!lMeh1Cws9B(&wDfI1(0f_y4;|tk)heGimPT#;T;^;DJ70&(2wUT z%I4&@D_srO(a%kw+PWQ7HLp-e85brDQfhjZK()IHQX&Lr4QMjV(06EHwC>v&!m9h6 zCD~Bue|X)Irs2-aLYk{4ubu{$bPMx^pf9$LbkZFfSPBlD22?jp97kwq{qX&hpj?FV zddUpAXXz%x2`EKQ!0r?P51#(GV3$Wp5hoM>dXwW`Bd(8*YfjP7MO3sSYYb<^uXixf zmQqA|-=ZXAJBOh9*)>G&Vqp};|Beh|A(Lo5lx%o|k*z|u* z$2pU9@Bgb6V@Yk+X`NusiezBft{pQ`M3IprE*R2@fgNJ3c-0Kr!&jc{A9EI`n@8v1 z7lF@iu2(;`5NG|8=;P;slQEP(uUDVnCgwTzMfQ6$KZ&8V#lQWT4THITYT0g!*N79k z-}-yC>z(BeUZ6EZBp-dWGtBbsNb-m|>Nn0QI_}g9{Sbo-$0XKadd6MBcX19WPJKHn zK5gVjs5Nvvi4@f>F9Mjj>qygovHc0M+e1AOS~0%#v0Ad=JgOzRYm)Ra@%yG zs4)`OQ?56MwNX}f4V%Lb^h)zZWsdq@kI+k;#yN@?{}+$H$os7-40=O#9Xi;F6HzYoi+P2m6(Srd}k8atoz^_aZr4BJ14oMSK0rob+~mY0KCxviG`m z(i{Iu(f+ZpFEKCf{N7f$68K*F$IT1JkE=EfaU3c!6sn|yprk~nK4L^YQw@~s_EvOn zwOwl9d%&?iD_>uzeK=+HY0y&Dt=f1&qBuR;L@p(PGc06owesnw_=C>pjy?L$?}M!O zl-<1Folqx54=YMaD8VA~bFT?P&Hdg=?yW-wr&rQBpxZGaB7Z2c5;urTn@LYLm#@f@ zDq>mRX^YzV5%V^v9Ht>RX)*^KIvCoJVstHk<&GWcrd-(${-S`rwy0NMc)!sTLNG02 zFv+9oX?zXX@aT1P^JuT`{HsqPlsA%(gifuHerVI@qlYFhtCuUm`m+;DUDvN8B&p3vIx=ep%BTJX-0`xfMJ3RLc)}LQmOtXb@Y9-f|ti64Gl9 zTS=ffkco--rFZH(cso(5j%Yyt*7gkrEjv?9&%V~+di9s_|{K2;Cg^iu_H|( z6^&@b=1Ov zmCmS#!cWF?--DEWEMud(L{to7`0wnb7mvtf)>txvrfr>K(C?YE$G^ zU?8GUMahk6jfpW|Ra=-Fx0T@2C07U?WD zixc9l4`c#W+|bdK1&5{OPS`{TU;(%NAVrp$q-;HKJ4yQb{2?TiK=V~k4hA=$zrO&b{&Y0)7a^ELrX$nOkiK3;qK=WG zkfZUjj%avlCpacWVq(3=NA58^RV`KrhzoX4rz^o09d7B%4q-j_{Q;buNJqSIL~ez1Ug+#~ zDg)dJ;%kEAf zdc)oH!cq|%%X>Zg6U{mtD|phSZa!Q(<>*#ql*LZEp#m-s-SZ~+;`g``Ww}X;0%E1% z1xOQx<^4yfwR{>OpjG4p8yh}QEdJzQJo~-*LBy|!pCaxNIWmHSN^L!WY)Kv|fSn+Z z3IBv&HamK7ymlh~JV{bp!s|ho`8(xDa(0oFbEeXE;z}^-3w|8eBydjM7r{y5VvSJh zB(bmh$D-hSXh%KYaLM2AhJvXwEG5U)&AcGQKr#0}Qb(zl~sBb!EFl_uWOm$884r8$UA-5!(^QU(;q zPv`5;Ja3hp-U_G*D+3Y(6^X$g&OGn3*Vjlb|I&2w*iF26+p3l+XE2=YMeEX>D#)&X z%BJpNgkRn7YL~}%OG(9R2xZPSk1`poDa1m^5--};o=`ryYsXI1asBZEmu_j5aMTxQ)-q8`!4 z%jf2h>dccZXHxS4v+Gq&tjog9m)50wT9u@G>;yksqE3=U>Z?VT3Ub}J#CcNmqI;wa zF5RO1tUhrrnuE9y-t(WbJ%yVtZzGkwX6PU{7Tj1?v~88$#9f~`Y1}H!48^=?_3O^I zI>H{dJyiKcN#iqZgL(C_IUlCft|IQp#o}G*h3sEc9%)wS(xw2}d*?*6U)xZy|0Ymq zL>SodK1YZz#)$zHRJW(Csy^qVr7ZRHE4b7ea$%msI4AwQ*yj<3w0Lv&hY=a3@+h|c zYZv*}U{%=#h5E!M>zfW1S<4KrH6K4&Zm2Spp_IO8Vl{hNuY)^mZImOixeP3*>hx*o zvj<1?I!t#@$~__@G}l~uGrYRk?fKq`=4ko_E5woAc%}$}x;`WR4BVU&o?s?LqubPm zX^Q2%KY8v_IxJ)pD27)x^3jXHqUm0n)2p#dSB9|fMO;)b^ozDK)_!@|$~E}&*VER* zE(6(jV28#f8HRTcGIL z)&`gt=oPGvsxI12QvER2$5afEiD%KjuR|13MZG_iJdW-1a1o`(s#vo-rQs3jOpkUy zwZjI-4DcP4dhUt~7e$+&=AsrAE%Za8t2x(T!E?@2V)LlB1@^UoTOx^8*>}F;)^-o7+^jsU2O?{jzDyZ<}W* zOK|R(QHWkpdb&UQWXa2PZLnXnp*6jV64lQGc+&z0^!|_zPi+3R0VfKU#g0CEQaB9% zPR=QbQ)P8F*w?loOZX(X?(O{2{H=nh=7fFRrHx2qO#z)F6pnr4SGg|9F*jr9)9#Cb zKHPBbUAArfcS+mV%6lV)-@}&7PXFeT01l95`=`AMc9QR+!YUY#P>J>SyCnQR5ofT& zG~93oEawY^I=C;W4g-Dj>t$dY8o_sL@;SbFF2t?u@6D&X+T8fn8JuM|!r|J5@5Pjf z`QUTA2AYNAPiL{778zmX+UB%8@DjsLtSZj7D+aijdArX)`iXl7v~6f0VyxQ@L_mnx z-)?b#eWQP%?&bqMS50iX-5wf&HLlEUfz#lwnbZcw_>*+OAO>Is64aj2%@179+tdiN z=e^_L^ZopiBH}0fL`LV^wHyEZTs-;A^1d;?ETwU*27gezQUsnNs9#NEE0}?VOAxVl zU{~iQAknSVyjoP9Xf?pN{SZtbXG zpTz%u7X6%jjE<4}7D?*nAUI6zzxQ*=AN!#VKP=?1HfE+Jhq;`3O6#?zZ4<-{s4>3{ z!+A%7bM;uDHNHW=YZTvR{gw{7^pJm%QjIg-17b%{N?}O|n-v;YPBkPS9T0Se7AIzN z8@D%&ongQ6jz{s<2rx_OK{;*X_r5piNz;w=jv`ud(4s4qmK~AGm_wJ@;Ce4Qn+^kS zQ+4ijr+eh7U*x!}!qdC;Dn9giAO?UNY*X{^A+)I-rVHZq?z)OIQA{<}?B8j15pRch zFA68G0d_L^xOvyp<^H|k4hlI93SEO&Q069xr|39&8Tz<{Tw4edpLde+op=A1pmypM zdq(#Trt5x>$HdpGf}@8r$6gXx-nJBV15)cDgyPq4wCU3)8&W4bUQ*1Fu{#m_`D=#= zCEz`W+$uC*(ftb zJ8TwD3-@Hfl4UTZaYXMcO%aL~Bx2onTN*NV%rt5nfo2_1)Xu>ALq}mE0@SR{L|MnX zFZqRW#}JBl*ao^B=~iv18xegzP;EvOwOa&qNxFauU{w|P%vi(8ch=!eCjPn6k5p-$ z5rB-?h6q|{?!o@Jlt2q9q>G%o9MoDHZsr|W?s2S9*fmgT6WV2l-K9FOT67blhe(W% zf0mch767elq%NZmH3s6S-k*@cZiMDU$57g%CH(oZP&phYEjjbe?CH~B4kG2OEd#YS znaXCchviO3nk=i#V!v&N|NVOQC}xGKwM6#?*|Rho=iT(b0y^rrlqyhy(t0Tl!G45) z=y~vi8V)waZ$a6ZskP52;m(sqBs>lJiI-aYw0a&qDS_n_G$f;S-7blcj_~{qJr@SM z3Fci;#yWYX)q|1+bIo$)GbWnjtV?@PHqkV$BSH=~H;t>C@GEHRC<7m{DId~h zc4u@~)m$@Je^2!%K(j$Uj@ecFAm0j?z#eaTnmWy5OPOek<#&}wsE;6EVC_vw;K1o0 zwvYLbXg9%I?BoH?mrkX*t5%xHp3Vll>H9cc(xU+i!a@@+nwWyusHiPX$+}Eo=&?e%jb0ck)CK)K0?{y+>?V99J9Br)ni|rArtaE=4J1N_O zFxdl&JEN+SWl4jD({ZKgvSyJ6nw@vRfo_6`hlszudez&Qo#c_?G}$f+C;Lq@npN+p z^cZckzCCkm4$DM*k+*e87y~wC%MV^2q(=gl59XER8hy^_9J zly|7jtBGd+-_K0|cfCr2$Joz~w_>ARDLVG)V6i+txtb+8Z^H-A?wU ziL&uTex=<{nZ$;?PvY_z288mxqgH{ywA7G5r+~Bfy_4z8oiIXuASsmB{Ao95N3V{u zhDux!A+;FYq)-Ub_k=bk?Avu>8mjJnemJh>*AWxKQgAj+Him6pzo2%{*yq0+eTJK}mP-ja?-vvewf_giz;1P;niZBiO3X^8@4epLUo4MV-p!pkY3$Ey( z4kzn9>;=RUOus9w;6()$WRVh z)x_7b1xXX%znsfMzs(5mb=*grZF4ZsQFgDN8aygjGrceiUS2o6HEH+_hoLPUA?r2i(hynWIO7d|#`PzU9i1&Uwtvx6ctcY&#~KQ2&1$J>6>0kYX-f8(zzQxG zWqs49$KMLCFs=|_cj?VOI@wnJ9@7t!s|>#IHfa@ibgvBjL*~X8?wj}d%Ujv^^0+$n zAs@MkWcAh03)47$U++ARPEQ>`J*`a;hKL zaQauOw1Trn^*HO#Ve#naBGS~e)HpI5wmoUQLhT6O;orq0c2tfGJjYT*LcwUe{uNyr(`Quc2bRXbGhN_aEhfS@4_Ife4(az%d6cj2MH)Uz^ZR0i+rh-j)Tw=(rrRUo zu9@@~NBDTKE@=BMu&XWYYFtG|+bz9gX-F9c^zO_(387#uiLX;6`~3CW(@)dz56is| zMG?2Q)j^fp`+raDV3SQnron4?X8SZ*X;L~)U8aGQ2iPR++HK=U=5?xGwV>`RkWMp# z!<84iNr$20Bhj>n+A%GQkZe^P-#vF18vQDz>s!SDumDRlR$TwIm!E%6e5!Cahai@R zhKJ_Pfchk%ZQ$$M zT9{nh5wV8i0Ueh{n7guNoCrA)ixJ;EcYhC12nRwk90V@)GR{w+a=T%zA-KL7V3bZ9 za~8aA3>v4&)~b^#*}G=Ilc1V6;TvnhC4G9e+p&q)NADSf%|fEOyF-71<5XHmH|DU} zB%Xa-d*;jX3nGBcM*Fzl8tS>GJH0N#m=Fs;059{}tl{S(r0@^c@FV>rR5tXK7`ktj zfiS~i$}>aP+WRNU{PuB8#Q+XFz)D9j(9v^#AMLv7!puVW=3Q`~S>{N*3mY=w`9|rz zE(pKjK?%>GHLu2hyCkr-2no~r3XWK*;R2w zy5W&ne}OiUL);O55#rxU$~N0-YIbWOU7haXI!y~yFBtp*=1eAPv*w|h_-0*C6R}cF z&94nKRh8=)H%-=YZ=Z_PlFcp=zUgysgoAs-7wuXsz__iO``+J~h}Z||M?AxTre{|y zGZCu|eVx{hN>+T4MH{=K*WZuJcAUB~e_DjprCn*&nN9ihtxCHem=5|J6XnUq1O|S4 zsRWXQ#aGE@$pJ>0i(B_UhC>f~n>{d4t`m3@X8T^GW2bk=WZU_NTD>9Ha448O57@W zu=t$zxX{w$ry9;h9HF}%XDp7IPMst*Oo(u@$C1ryG@saIFcD@3fIY1=%CvwSNirPF z8zLR3ZM+XJDT6(Cf@veN$~ z__X34`w_}Y0U4sxH4BFDkAr{c%4x@P>D&*}W~yBnZBOF%Lr&`0!XeLBFf+%~qR-z5 z42$i^`EP^l|Gur)JX-|v)SS`&32crU?0})<+I#)DnFDX(gQV9PH0qD5kA{P1mfhT5 zFF3Yc0PP-c<$`-jFICL+!D?%Bs(3btzk`fe_IDg5Ufhb;DTDRz(r5sal^5a4Y z@HUY2;=hPIutn|4&OGYBLaj}bdU|;9Uhod_6df!|Z^8Yk4( zcREvAL4p-+V{9YjYIXb|4lS3&H*Q@3o5jXEEm0ocTKle^wMzuMbimCkZ}I?XDQfl_ z5IQ_+PD6zez9ETQ=!zvtC0gURlVtQ6P)Xc7dF#yuD{}2x6^%oM*xQ_3bz69`c^!#CqwAEntyr|LqWX$hwpn^1cC57A#YB zzaXc09}RO*FatXRq;tGAo9)eXL=oFnBGlT8KghC%nwN(0X!K0;@77C6pk4U!{)(|$i?zUgm0(#kCw&hlHFznGc2X~$+7nNGl3iTMyzS1 zv+ce0dLne)G;(pVzAahD|1TcYxl7!dkxF_Brw8}8M)E`b56?v{zzI4}ldCw8JAb8y z!!Loq*)~GBW-05pT;C$|w%|>b2jKj9`UG0ep=@M62T%70Sc#)ZMD!IA^ymh`jJ8>d z?DQQ44lXwgNVNe@?(g8XCaW{Y347BYSRarH7t27S3ZBojk~Hj@INOAX{3we)AAG`p4Fx_uuY03dJ zf9)>p-uM&q#QfokwdcMK&vvUW+XXqQ{4Mgi8EK_i94o;Uet70^vQ2(bCh-`IoRt6g z4r5y>RB*ymz%6LI@nRx6fT19+bbos_nLoc&tp#f=a|7w)nc8@UAMpfiu)4N#>mVRt zmOaz_l2+iX$W`OwrvbCDal{!&lY^BfS0bay{odJEu}|9>bZZyqTTc3fCr4Ib-r~U+sEX%gs3{n=Bv^{FLQaVVpaYA^E;?aiz+i(}$DX&jQprS+{$dUY9T_ z)cXpZ^Ay-{GvItYQpoWv6IwW;@ZkaWd}FW{Wu*-wml@0D(~D^ob%2Z*O?j7(aEC~I zftISa;8?AAD&R??!iS?m=EzUb!U#*j(DR3re!6>SY#|c@7m{IZ7(wYlM6axqN`;DQ z&LE`9X!&);0|S;{rVnfnCm9aGm^PVv1pVkcfwTMaT+csSihYf>gskchb48-3A@M?7 zh<;s2!B6O?*G4*!zAf7g3m*{ZSv5vgNh;SNmTiR}CSP`iD_=yyIkbsP=VUr!2O;rd zHXBGK^SNoTmLP?M%i*VOVZlx1^uNq*veH9F= z$?MSFN(kTVOC6m6hQAYf_E(cNAM+$K7RMp!(J=8@NpUAj@Tm=&cmK70z5z(?bJ+p% ziF@JcmcP6-rLVO$&wJRf*d7`|vKzMG5xzd*fWDt`kQQrvzJ=D|q~c&P{8{gNEADm% zN_eq%vWf8$^a4xYTOVhkt_W!9P%Vp*bjC62#+L#xpnSmT_30~X_moTGvkN<%#VQ7q ztu^nLjct1xvQ?mT9K}6YO?I?ceg3W{D@I!&>=(DcTaH>)&%lG2q?1ZfyYUcURBXFt zqI-t(BE;okp`O!uEA4g$>-eOa=|3fd$(L)iiR`Su({)9)1GHHdzuO3PxnxAb`yHUK z?K}6Bkqk5S16cJr#&H2%a=nJ2>!5{|M@vcPNkjX?a9D0cT#1HH>}WKR;ZcUJJ*3`A zEXl#6kcoC<_?iuQ;n{xp6+>h|3RJ+3MzQ+_auq-phQe_#adWAm6_`sGQYGl~y|wnS zj43DO$es0lep8}Q6a*QImiKYC{@W^UyTSOST~EISGY5?qEUG41e?QH-B>=cjlbXIar2~DDo<>_y$(`*Rev(p=?r}}f0eTu;+z!YnE76H zKn)XGA(Xw#I_I12832?V^at=nS_`DyNGXt1$i$Y1E?B=S_yg7adBWY~Jkn?RnqfKW z`GTu@Hq^OwH4%)2C^jc)qe2Ri3^5wsMN8B<#sH~EXdk1pc~P*|#3p?p8a`rDr@S$P zwRJoAY{4l|TzLqtT8ue=2()1&y<}|3#KXkMQs^g@Na(H~AQ9@XJy_cy&)$8KHj!dO zzVF6(eGo#`ZPDPO=X!9Ook=P-4}N{!mMo-c{;luXNQ zxCiTRKV5HiLqBKFb}Obq1Ona>oEb3tl=EG+kF)7{mLDH8;&}|zd{Hs4v*7GAO6*EV zhMpt=vM?L#`%)A<_MB15JIrD-aAC<>wFSm-cAc?P7!W#&`OSIB&d)i(UBE3rO!sLz z4th%?&T+dXFytJzeU=71#{P5q1DPnit@?6xzl5F78E!X-;1&q`Ei6oZvcaIIyWNsw zUB;PX0PuxsT~G8Ermj?J=dix*HG(KFG40?RT`o&w;Ni%Rveej$v(cJ=pH$w((wsH2 z$I2NA3E@BpypvRlN)uew*N@F@O`Tu8Q-Mhtt7UYfC{AfF9kJV7qU;^230^r16^ovF z>8!?4`A{HB?B*}X`@7t282Q9*Gp|GmKHA+#<|am#TgL6YH~oSb6e=qAd&Ea;wi;$7 zKl!_&TN4b+G<-U5!TPQTGDP=lWfB=cRUxS{aSk?vhb{bz-8kqkh!KlOG;)nG^$uM; z7F$s;hg)o&20MWQieBefn>(927dKN9EQ%?S@SU;XiU>paQBbL`3xiceX`WjVSNS8Uj!m%;-c3G6kA1VAM7^ z1cu*j$h5iTLc?%g%TSCd5H)H=NwO_Eq4uhjj;^W72`bIrZeCW+*5PjbO^W^m6`1Ky zcj$)*1ziEjY<58rkh<~0F*$T3h7z>B1S~S z(D5&IMBIVu5`K4j-=R6i-oy$X9_ZQP7s%knt|iJhYp*Ne{Yj5XXv+V5KlO84Mm0X> z=gV3(smugo0qs7=?Uw5hmf<3W#SPaYXgecFx%2`ug{JLVf|a$~Y+A24JZ8q6K*+nCq znDIj&=kJOV%1By{7*eqY^U?Cgqva}^-)^^gG5;lddUJ`ucIz)l3Si3)wMF`T&h+O# z$!hX1`0(`E8-vOJGIFJwIO`%yg!wvg43+4e-PqpF7;_gCj&a00xKSSzZeT;l~t2S@nF;X z?ty}SnKYpX*^dWT<>;YqmsjtMmFNw>aAC)WrG2es4)MUnaf2==3cW{y@=d&t|P+w zxw~pD9sYaX|Mzh+G1ElZ)PU8*3En9<<4x&{!9;71mu42Pn9=A2qOPaNZ`&w*CGbGn_wricc9gm4r!ge zcS5xVIdO9dgOGx}kUarCvQwAYS-?K$WH|v1m`6XpoDmI+)kj6o^YXpIkHlWM26<7- zNR#KwIIlweIyNdmiD1-m)!J6xX2@%~B#CB~1HP4_*BN!(wB$@vXf`5cI|uOZU>jE} zgivu&Bc?mNa}EKmRs%uR8^lU-o%tAJIZZKX4P@gl8}mAZeb7QppQW0q{cOf@ytDD zHkWfr*5j*Npq&yoUzlI_hcZC@6P=OLw#(ANFzGJ>lv_n6E)`wL1?BJ!5uSx-!G4Tz$@FTNozCN5a=V z3U~Kgp2+BupWz;4xW+S8nfXwm!ppzIR}$cLeJw(sxd_@^0$TPZEgLSU3mWGrm_(YUA&PP!?dQTajy>++s6?52hw5sp(!{HYI+^cZA&5kWE_2c zQ-T$S7C!3GUMMGIhxN`};gQgAu7Jh5&$=1{Z;h7MRdx-?_qn5GDnJqC`{O5AZZ#-> zv0XzUZi@%(EOv)i^o;8veKZ&9q402vw!jSxQ%IPMQ*a@Mr>UY&7E(Ov=9%|mA6FdE zHo?ehGMyVe8eY2y-A~-r`rZK6(vJl&wEq_2fU73(?k8u5na zC)5Bn&D#dT*`G0G4mL&JVo`U|?PJhf!;n{C78r-v_lq zDI5&<3Yi@3SkKZ1Aj`ds@SnAplxC}|cQ$X=upOZoVy{7Zw4E+=D*bo(A0U%!`esn8 z*HcYFmgSZYec5*1DZ*L4PP5p}>VV8F*xS={L$%d~#&rHS*ez}VW%!d8Tg!Ei7D*LqL0|RTnCF+@5`@~UqKW9g-0Ww-VIBsSVHMt&wU>D+0vi>y zy3(HP9?z8`efT)B7Mi!m#=!X}A>4(cQo8Kjyvj^s0pY+ujyDoaat7z5VW z1?L6ohxgvdgYwd4{e=tB1IKQm3aMl>&ub8cofeW9isiygntO}>x%{ZRvnvl>dJ{3w z!n@LN{=09Gf&tPM`o{>nQG} z5?G#`Zc+(SINKrS9xd5458KMUG2%4Dy}HU~_pYD`leH7yEFX8_JfnddYuLIEe(eu~ ztnm@7|L}-0Gkt#@JT;GqCl=E7!bAc?=vvvF2=IM=+dH?83iJ!-dVXe?Valp7RSPoG__=5YFB;Z1$fE;^ij5HM*9zS%%y4FNfibf!w!XGc4SuS{ zl=V{T45)TZ35*f1cW#1=v?Wh!`GO~ zxU+B^m%+r=BICX5d6KfXuH>o+7lTNLFs;~X2E{>lVWK;!G@A`KQrHDBkT{`(OxgyX z@2Ji*e!+|ZSIKBppXnQ=S*&+Qy$}a?A@o77g8pljUN9B*bPlCRBQ2`%Kq!iH(aN)@ z+E8@PW&Yi=TTC1BeEaF|*`EafsK*=5y@*NoXUeJq zL&ljXSdxm71|uyQ3V8w6DRo~hJYe`VnEIaE_S6ql(62(8&0lWYDq4v~3jU8lFvA09 ztsRC$i${9$m*|$!THSS9D3Xj%L0r<#epf<90}3y^c3Bnr#iI{e=J>72=&zkR=W@aO z<+q$7%t6wzc!Bmd!{sM0+*0;&*J?iwn9gGH(bo+3fOvX^H5S5Ek@H#s3|V_K#V%T^ z63V)u^uy991uD)Cm>%_-OgD?>Le3QO(G?UEGCCho>8maG%2a;;M32tRXk89K~B#D4Bt zJOR2LNorylL8UePw`><@S7qjTko`2;WDSK**=2y0V0gM;kEyQ!ZF|e7C_|sQ7=-A0 zQ3y6?k@2Q9>e)+^22?k5XQnJ0GCG^|MT$a5yS>{P-Y;xKC9qpi0|trMF2hAlHX2^Q zDL4f~BpNscM}m&g+3-w8`MN!Gxz69d{aPew2H8w(z<-Q@89E)DHL8_*;4n&+Wx7ss z(NgjseCRI-xfjM)_5~DX2%vEWbY7X+aSw5F=V~~Ms8bUb1?8;=qUvYQ0>z>Dr_M9a8&<7HZQj=T_d-N2sk^nRuWH!o&*DJ zxP1>lrp#i?pO3qn1`h)luNejl943^D&Z5Ap(s56jbngq?*vCD9?XguT3Uq1=Uw;8a zqu&p-cU6FS{(}cB6GX(Rq9`zg7gaQwNn5AZI_X>JZYMLTxeQQcBUEezra|(Reu&*3 z7P>r#X{|8aP6o85L4}pr8Hji;7tb6@b~~7XwdO&8##$B4Dj(5LjR%`3wftXnwz1&C zHcQ1Zni6qozyPe~lvSdK-P?goIUVL8w;f(!4pm!=H*@a@37Brn=1vJ(tZh{?Y$Y}FkHZ1NqpV{IE_D2@L+2(3uHM@+e;G<;kE_5 z=AT4@iW8L@J|d9j=uII1D7bCN9`Zd6+h17MAHK&Ce}jpPeVu=*8hm8_ZzM!%_S@Nb z!1C9q*&3=LjrSZh7kpGU{AMwl_V7mei+QY|LLuE-HT4U$T$ARc);839Dh+gIvc_yD z+vqi2p_vsNxi!cYtN(dayMaE3yt961vUTssYw=L5wPp?$<1tRkKFCb>@i zLQO{=GGuyI_It@`%7YbzUe4w16H%C6!Uvq-pq~tvAf* zBcGm}Xli`TyPmU@1+SnIczDij7=l(*J9BwMt_Nv-Gf==P@}T93`M+;b??97__VWYS z<&Fy2(RiImw`J7cWC6frEz)|?B|PiP>$_)>49?& zU2;h|?y^m_x@E6I5hYGZYROqD#~VvS5_+32$YvhC)v?Anf&g_QWG$YeWp8bJK(V(C*Xt=ZB$sxXx?bq6lI zQmWE|LPl7oGG6<;R2n~vnjyxT(A&Bk-_OndeAgrrO85Lr+T6C%SRJa{hnMZbkb8k! zKp!OVw1U8M*(ZZY7p=E})))F#I~kVhp(megFs=x&)oOGyXpp#d+{=J1;4#1O%AFo_ zu+hC_BVms5)oQ7*CdI0x;oq7v(*r45o%Tk5aRA3rxzxbPB@Bdy&o><;*TTCY`XfMe z!D7N3#!+Zs3sZMv@#cSb@-Ts}TFCWM(H2xIz9aR3A#v^o*{%}Y-`~T&?s@l4!v~fd zoAk=1a7>gOg`tPb+UopLMJ;qv`K@8#jrANqFT9g~S>Yc&=C{>I6`Ids2bSLLLg?0I zEcQ2&gc92}`?Wd=J5L?sXS1t#H;=-5C=`eXK6VMWUTdipM5Y}7tI)c#f}?5nxyfw< zJjh-}1BHJ49xM&k|83I2{pD_pUH_9f9w-s}9R+nQB&0J6!ox>`cUdstQ1G{!Riz)O zvv9OD5iyIMt3NXZo@oUNCc8xxUxSXnx%K}*8m*m~f-p#i22(D9lWdE=7K_V-;vF+} zCy?tYoz|}NzSIIAt^$|RnL!elp6&{eKi(G0Os#z=o@=OW6exgQe*DSAmMt$87O4x~ z{+cz{6CTihUHBGL&;z}oFR-_dX!xfqGqQYs%R^w$d{}CYF}1_gLy(>@ox{h8X9H$U zTwfLLn1#A|(adB%BwRTO+66M$7Yhf@yHi>@^aB~JdB~3bPoyrgxj-9#?gFNgFK)q( zP`q@&Yz@taQfpa1>UglZ6zhca-WvuEc9rKbbystcVYE8Jcdkob%%FJu-4jR#p|9F$R;X1m_97l5S=08T`>%S3y6;a zXc@ax+_p2jV2+xv>caHRbP}iqaz=wgCyCgrUgspg1_;a;Juqc!`5X@p&3ruO4fFm1 zd^cXj!4;&QPxvZE5w$i2h1X$E6(SS_gt|wzV#&$%3;xw~5_dcS?~DgIu7k>nZYeHW zjw=av7lk09y8IEJ@m`LDHDvUsPk2VnpgYKCh2sw^zzUc-ibhh%88ipgKG$&O(8=JK z_SymD>;xtK%KiIVpSpcGHN7PwyMV}K{Q6JF^5gFK2C7f=2u$Ik)^;bW^!9>*G%pPA zJY%PiFqEbiY(<{-T(Xc2Ae_u({9*fWrpnB{E-H8OP9gv+1K1qE$`P}Vy2`YoQ8nUL zuhD1PvzJw`4>b5{j__5qp{i>C6}2cy)~BM~$WS>P*-V1Ub%b|Y-iCSNJ{{+;#`R_j z`2NUW0VLTL5rkZi!@cy~yx00w;4>X~XIs3@zr4cEa?!2NwTpS*&$Q5`lM1*&1=5xYSGNsP$LO$ zj*YkSU(n0WA*D}iOW`FmJltl?$bfDo>?AYqChz0ED!j0!VAHr=$5MZyZvO%a<@1$x zcjmyo5;Pd58htm}rO<6#MGW4F1062=M1qe9&G}EM1eo{zLC|P^odnaDoh|YT9`EaW2}(ow<(7nVx|!4Seew)7^@!iOLuC4T(TOS;Q@;%8tiTF z!YQLznnv1`HnsNLSRojM*K>tO(yF<2;w+0nvV<OiXP6c8IAx< z`IKkM!qP5UAt<$+sTJVyN-f_bkO^!I5ssixWrJZl=l5~lW4%`)XSr7(mjo1OtGQQc zk`f>(r|;qpSSfU*e}qp2*2bQ5+HL;-E2?42%HdsRQlZmcTlJ2-atrRZ~c@yTer;P_Rf530uQB_6DO%?6#Jz^; zrX@{8`jh&n%aY_>Kt2g*@ilwgm`O=nKbc%9(Qt4Zj9e;^Qp*tHC>qSUICy845bvj2v({%_d&}A-_q1Mg97mZh?rCm+wOFMe z;pnQ*Z;w!G^tyE`{`km?$iL}P-$ZX6{@Jm(uW;yK`u_M~O|q|a4|mK<+UXt*PW3K_ z%}`nMPq_#COVZKP8gD+^*BjsZ;9bz6=zck#U8wQ16%W49|NW9RKE`(tHd7?oP!@Zw zLh`~B|C)tmOOkN^g7ix&I`Q&mW#ZqLj~d+@FGxHOmp>iPp5o~5mvZ)ey&LW5D~90Y zhkJI*`qM+*2cAjekInArbeO*5cPS!PEt-Ra9=v7s*SqGvpna*?CW8;hV2{=A%S$fUx{s7*E{m`^4xfOjx6$FDm@KybcK)%}tsg-KJdCcCeTEx|6f-5VjAa@T!_o9>^n?nrWPOl%mu@cwC}uR^#4 zOT(G&4T`(bUN+y6?oB*DWXT2hQm9+_dAOT-M@=mME5Y>Z65=eK$hL-DPWqI=UhgJM zXN{KuIM0%0Kj%i9LBrMed5txx;T%6uO~E%U%(NRE|9#^B8!RhH@mqxI8AioTKj0HR zig|M3v;4HLAQ|yK&fKe{t4GnmFzJA*AOQ4m%NnuDi#k|lami#mtq1bz^pvf`!a9^_ z+`q$N8IwhBfdU`$)Y>41{|+RT%>%j(1A3ZQ!@4&-m;MG!X9X)#L}TjHCHsAAOU+F$ zfx~rZ+|u71)T4|+L`M4KCW*hVmYfd7N=3(RikS;6MVf61H7*aSrxd3-D$)e2xGsVr z3}-OESob-<+f}7G9+F(oS&Q0kvJiUL6&(ccXA7uVib8vrEk$Ee%Vd8ZS~WPn%{bBY zRzf@xh&^hlkQA4mH?;EVkZ0#PO&L0SQ@toJadb)*-BWh|N|!R^l0nitk(AZs{3!)_ zw(dCN21h0F#)Hn+x4Md0wE8G_;(dEAm0PkeJ-;nP+QwgZiWGUiJHDvZ!_DOK0{HZ< z)2wkRctnP)*wpjpXj;ifmt_av*@RP|yYKD=gLimDTeDG*XtbkGM$d(A!w(hbKkwW% zX1zJvRU)dNi|0!7b#U)Kead#T5@%cRVKS;X@1()IC$;o?=8hJQzM^IuCx`yD1RIE> z*k3X7)m6EvHlT8PLWb*Osj~jRQtiKIw9_71#eED1{sJwAe|){|`5Ltv?@XQM6O~ar zfA8&QzQ1ROqq2Gq%i_M;UGMNBaDLNOd8}5l#hH;FMcm}h;!`!>rTpt zB~6Q2IrMFFN0@CbU;rj;tng#C^foOC&Le{0zqNa6dwC}B9(A9dk-jI6sy2??jm~mZ z6p~6@WNrvLg5GFszGgCAmF4K*p)Qs^D*1n(E&F$}ro2@(aDn38P*5#c} z2AmtmJcJNG;RU6a4+}gW7OBbHZH#f;-FQEBY-C2itjD&L1z8jp z#LIi9z^3Q~4DiR(*Z1>ctPnEScfIgwo_vs0_v4y$-H3Tf{_K++^fSQ7!JrRDsI^7%r)e3BXrdS z6OyJ?R?cV=)kXA)r>xw&2J|H^vYhYH*kxsG=E_LtuvXZK_vqm+xNZvkv#dO$CgUb? z-O7uYn|o93ZlcJCbnEUviPLAZS|NdlLCnO9sgC-o9f)rlxaMI%OYQEXh%(y-N0(>4 zN%_CE8_SpFAJwk6<5q2J^JJOih1d! zKo>E@mKDXL;;T&!tY6x`Jry$y1Qcl(hZj%fwl7x6pGyP+g8lD@rjBoXT->}bNKc59 z5#C(gaS@j0L2&F;QC3?%nx@%Z~i>_*Wy&o5?*XB=}d%d=; z6#n~Rv<@LvEyM-8zdD4P&WMJVN{Tb}o_z~$w_R^s5{`~Lw%h%cJ^JO_w^9Eb;1Lg0 z_8do6>oJ!zlSkJ%3*NjpNz|s!4}lL7<|2MKSCbw599Y-M_{)xo<6T-ka~3uimRz1E z`iDl$z3Q^}-H-V>NTZE~F6uTfHOO|^V|L7LZ$;dPtvLv~Enc_n$+3jj7Uzv?1bcQO zq*$_AlS5yW>93)h_60_M)iq>gUsw{UBH}6WEWNI3>HOdLFRKZ;HWCc9z#jJo+bzmO zmZF-ZxQof#^Xc<ym0o5e;q5uu~Nk*2OMf5{rkCx1PEUK#zh zt1xmuamK{*2wJCOx5=-j!MntAv$5XN+eFlsm2KAB=JM`-UGwix$2GUkap;STneCv* z{$p-#R{ML69nc!n>uQxt$~N*TvRP?Pfw;SlswL+>Z33aEkzzyyG*Iq|CrhlUVQrZ1 z$dU9xb@dvGN#@5Nj%&|f@1)OSFB!$q`LAsW`E*VdpmlbodMPgWhd%Au1X6EUd2ySF zSD)6mF4quLK2V`mp04lyWoZsp_Z9jz@hG3p;IoKKn*MAhu9Eg%efzClM(aVQ0p4j}=n2FKOjMOp%yS`r%J$LWe zr;Tq|ebKuo9(Z^ywy1{^I(lTW8@TcdzTT2>-1GHA@M`BN=`yA{ghIxm0U7;7#cB*7 zk3QERF@VEK>*pQoaByL3`LG_1Gg*#%sJ_Q)t)At-jUm4`qIZHSbFVfsus@^5u%Y8h z*HPlv^-Z5S^@tmWj zG2PrNu7LFy3Muo;_lsujot0E&Tfe?>HJnc3g!6`gWEjTxcw~fK zLD_Es5kHdES|n6HRXDYY#@x0yei!G|UjED1R#r|>=CFN%(WB>X*m8>sK6Wv;Njrms zmhsDUx7_E~w(+|shu77(^LWASuM9nN%g7xX|1EHW&p>UCQS|w1wyirKa#uP}g^qW= z0~cFRE<19x#$&<_rX0E-@N;$kCjuB*nRY_LS6(cVAJ()Uc9<@=ph6HSd@p5;hh?*u zVD}mvUbbo0ijL_fY)gKa8iHp{P-{=QWg#zR}2} z!H^<+R4z7f@0N$i^GQ?VBX+T)atKf4HQ#Ol-ci;MeD7qwIbZ4;(2?onQd5?3{C6-B z;hFc-FzI@mS2fwpjQJG{^4AqSGlFp5HJd znrv40p%=29e7gH_f~oRR*~5~{F*baYZ3`{F-djd+XxhWAl>lGi=f@lBX^{UO;JKVT zmpr{`Kf|2i=&c8?@1B~Kb8UoD(wD7|I)t+n8`>-$>Y9%Hv%@r1fQilA35shd^`Mm? zW7Dny8Qb8e)R>iXZVa>mBG%@FsD8+Y;wep9o#wqqF-s=$`sY2mo&{g-y5<}2UX?z0 zdvW;SaPw?0IG4Z)62_V%=$VJ3%feT|ql>ucNI}S`kp<0qxOK*Wr!psrSd}4&OiY`v z*&L$y;7JB+PmYs=RLU~=Z^5mBKTGlE97pvrI;bV@!(^}XpD)I3UNd#cE9QJ2H2i?B zQFdpgFU7GGB=WEE89o+l9EC)$0aPm@0=FcGhS1p7P)|M&vu8}sjz5m?&kE*IHv-pv z60~~me%B+TpLa8#`xe?15hdbVHRP*x-j)?CgI=JS={3+#=#KBFAfj~zg}C}qu0=`M zVp<(9TJ}gEYls259fMT*eP7j{qa{Zu92si+Y6JnHIu4t67IenX?c?EGU1Gywk9HK zZ~j*Z`1LRLYp5XXSXS%gxy6W99YFuu6EKPd->2Q84m%S)g8-4#(HOJ`o=0cqjFfH=XqwaP~+6A z^B+`5D2u5lI6hxXRO=_qS z5E1E+(5s;&7-nw#eY($c|AG5ylJo4DIcM)ZYp-?Y%*+UAmLcTr^yu7B+kQqV51Idn zD<)y?;eRWa;3f*Qv<*DR!gPRjzb94|ZZa3u=(@eL2x1a>DFR_C4i_vp2Hc>~n7tow z;N!$KuG$%$7T(B4^lZn!;OQU^mm60lOYCo_%=e5>9Np>h$xN8Ndja5l35~z`4xAl} z%jMfg_USooSRe8qb{xJsa%`E3%Y8HSaW?VWxKK{_=6{|`*KBM@?=GoO5Hoew=iWfo zM;>5BF#0(~d+KWxOP>(A((OS0<93akRURQBMfrk`0>XLUJ%6OVZ7{5LY*kffxkx2p z7R;n6fVY!MFoIva7$r_CE^Zdge~l&k>c;(2NL$qWK=bj}=ZakVcvLnxNQAkmgcKGim2pRsaWOz_mWT*Kd%I*n#>*)QZ927ZPkkCKlW~Kz?d0 z%b%&jl2(> zrZWg49QSnO(#^RCWBy#pwH)cN2e90#;~87A+gWiW%K@%$Wg-Ur`_&i8^{uu}kDG8eNK^>x8KU)0p4f$L3W#)&Qm=y}77Z zz;vm9ZwO&^Cq+^!gW}rd*5k*(WSAjSyCB@>__S#2CIh)XZ7VD6_U)A5*n2l2=sr6J zF&ZMwxw=7q?^;Wm)Ij5fobJ8rUtDoN+93VBd)6*&FkJs!>aPHhfoON^j zyG5_$D47p|R`I$suR8ce+SlQJBUaBipLQW?5igtZ^O$c{S59H*=YG8T$PY{Ei=Y>Z zMO$GJB;QEMwAaDAnp0ys5@~jU*4Ti@P{PV8C0q$EoL^H{EVA+(oX=>yykQ#*qLRH`qkc+(Kqg+1wkN%XTo!zlTZ5D!RU zM!@*{x-SlqS7aCUXo={9Mfu3@AgRXAoM8hqvXNvfjJTPwg&})FrdR;~EUvCQ;97N| zehRu9Rsc@O&!)G4N}F%e&3j)VRZVf~oiKkc{Tr=^$i6>I4>mfQYJEFn%}0>!1JUC| zm?V>dbnFW0P?n4`$a?(Hz$d5CTwL0fFGMTvU+-fy6~W(=l%Fn-{=H4|uNcgQc^|vm zZ}GUKsc@q!0FF4*n4$tmuB`t)F6ICAbp|U@tX1pZpU8>X^c9P2o(6tMf2NB+FFXZF z3XPk>RBl81x2~8+KW@oI2-BRQyrg4a=7fM%`1zMBs0wqzU*FRisIUETfv(6PlwN`c zwxjQ}G=_q352RNekkT#!QJ8}^0w-;9z!HCK9;X|b^;U7})26~{9^X>;MHrI&xoEi) zSu;Oz2loP+-fdi)NMawe3^l$y+(Df?6o5M~)YJOr40SI|whOGBAJQoRTZbs{N7T^x za1pT9h+s0i3gx@H)CJSDT^tog5G2PonNvxY6rcQL^tvrd#Bx)QwxjXGkTEt7fXPe= zi@8JtL0^%>m!mHoI;e3Pm4ZrbX=+3Hnu%}@nT)OTxW&0U5H*_p-GEj3Q`pw718Bia zM1%J2)I_l(NjaTP7&duC6|otdFq=2r5rJXJcE)8j1E!)w`MbtFIn-oScvxlD;w3P2 zdQhhxxIiY9Cm)ilOWB9=@sM@)fumbtZ96CYP!&5y-DhAwO2;)Lo;0O)wDsZ-@)@}H zKKbr?hb-UG6KK8VMN^+MOTKbYQ3SDh**@tN^11Qfv?=5wO8o&tQQmd_$=fUb8vCJ zqupaPG0kf9!QvdlY3+jkg8@~`r%N|W0a^#dvZFLir&-!z<<>QN{;8;pl zJuri?M{H-EJF`3xi+Ds{o9b!6Xo3-V##2}%kqa){x{><vGgRg1z)Qw8M=2r(INO{`dYdW%_~S=Cd)D+<9m2DQ zUFFA+FJ~;QN6&#|7olY2Rle^C7cfE40Pz8d4Fj@;hT6tGLa-drurJBcQQ(B^SP0HV zTH4j8(8=v0U^=Gg5&I<;@KhvXzjoks7GZK5e9Hr1`~^T_CPF^}|B=#XQpqCff`hGp z0!CUBOanTYXQM4i+mww0$8 zF3Vi{;pPq1U1j5jI<}~_eJOf?d(d-iZB|R|N>Lp~O$zn5q6~Ia(U40x?VsW0r{C&%j$dT)GFvQ z)MnHN!(_PCF(_ZQF&C^g{Ra5#!#&)gz7IxX4x0bSW3+J(UelQy9`1=4hz~8*hu!wH zf3L&c$r6E|*9Ee~Pfy4Pk_fwvg)<(`eN(_q1-wr*I5Y3VRGvPWzuMN`&YHfWoi2%R zlxv_+&sJWz%gb`Gz??08C5ue>k@Io(>wlqpUh{L8z$_YY`Z1Z+w1NmSBDNO5u=@Vx%*Wa*eys}!}AnAtSfhiaNvZ=}4d;IS7WZefA)>oJ|+=Xwt%RH2DbobR= zLammqIb^%Ym)JnXK`jwI<-*!EOJct*?uXAev#C+;uq0%}RW@+?OJ_BhYPEYUA*D|^ zn_g5z^F2guf`ANr;})dOuSfG*nN)Yv1L#exPvP8iu^FQHe`8OYR^qSQRsLW4>(`r+eW7swOARnk*Ku{^0p?J5x{&(P_Jfg^)Q6`F5F=s~o! z?_noQJL~Puwn&Cv{R`mD(K`oH-#f5)-Cy;4xyo#pn{0(v%pwm`2cO1*ru%gj*EcVn z{cYkTUcLs@rO4f{ym;hubjkRFBZUEma(FV337*^rUws`EIKb`7JV8YLMqCC)=(kR4;5l$j zlY0YGX|SCy6WcscTnbdvgNKk{bzV^H@L)Kn`!9!p$5{w}pCPFiv3Db~%;w{)hM2%@ zA~upk?ic?@kX1NMf1~3(_x^IL6lI@=u#`22YupOJH}Y0g=o2nb-lLn*UOKi~@!`oR zC_#HR^;eGrQ+N^J>~obTg3o};RIfcQaDR{XN%cO`gXP2VC>U=5qMym=E3Sjjtml&s zQa3?j(Y3Ft$w)>UhEXXZ4E0k;eY*%OC1K_C(^jtUV^!=Uv+(oVYt&Vvo71}79AFZ! zcd=3xn5q7-O`F1n(`OHnDS#X<8@W$^4iC+|PSpi{BWp!=Q`fm2nzwD8zQ6dv#1 zOYpzkkx=xfSA2DE_WGrgz!O0aCKClBjGs-wRnPQnYRnt;@V_@(WnpfA!b_?^C-nCe zDOUFt4eD!mf1k)y0;V^h#UR&bG$&<4Un{&SlNg+p;`in`XY3)&jZu@|(8N&s!eWEk z`aA7w6z%&m=#-ZUmwf||y;%ibAlH9%8(*SMz%;dqpTfwAEoVqc)n#@X04BD~e_g7` z-t@-E$$Petisye8U=u=Ek{P2`UDF_lp1oeem-JjV9Q03MF@rygz<=nxNp{@Lo593w zBRv%J8Ejsr`)Z@Q5*)ukP@7;w{dZA(gQXvMm0KhNtVg^hKI5_HUPB>NVbCCr#_R8A z-*bew{=Rd#_t2)(K#m9&&itXvzm0i;x*~tk=Q{q_QQ_G*_)q@tiib| zXHjs!Nkv%+6f;Wm-zIfCiOcG-RC33Rj|_<=!@pbUf%9qD163ox71U;9H0WgZU^IC} z22U{C)_tUA{Bi7FA1vX{gquBz_g z%iGSv7XWNxJ?{G`5&a;kbZYM*!wTw4E7dE*i{yzJ)RQ?v?S;lpF#g>zvH#u~%0D#8 zl|1pu&1*viPV+GN#zF+oK3F;Fe4T}iyw5361-nzS49>PVmuCrle?!PV2M9`j8=~Jg zjeh9JMo(7Syj!}xk3CqmdPC8ubYV|T!oCBhM(lozENAR;1_B?nWKZS#w0y_}vm4Q} z8jY_Hu&-9jj`q%iSkx~9*c}^LvcFSWZ!c1-#j_Ze|JLvQ!_(Tvsrayf6_f}I#V|2w zqviEauuf4pKL#-MixI|_f)hq2cIDjf08_OR#9Lts)SiQOX&&uGAWLBNW|giL9shlr zY^`9tZ4?J$a~WU11#Q%fO;EeCXc2Q&>3e@hjRh%l>pm=1U->=0NAJ>-qxIey;}*aDZ~) zP|*eh+~_XA*VbQGfl4WXa#Q5QgH)uvNKXl9p35>Ov<|*M!oj}oGj-h92RX+}IYEmI z4PPpt(ih#j2!m;aGiZZaUwxppWPWzAC1YtE8PHJX|2q>f{az}wIR$|82vcd-s-vjZ z+u~vmJ!l}6R-^k40U~*ys_^^4fPfJu9RH@NFhBAVflGL*_3?bfa!6z%f}7vS_w8kY zoRBcKGr^UUWff|kxs#<*>TtxTQd=j6wzoD{K(eadK>Vx>&Ahuau1g#Y!o(kUBl?W! z1eQDo>BfepQ|y`~8{~3ACOSQ8<8r!PA`8O%+Ad{fs>Amg#$;$H!oj40NhX$4!dDDgX~cJJ9PeCcSK75d_cetMEKb7$mb9lj#;n2rCy zwlb%?p7k@rW++@uKgsU&nwUq7cZ*t+omA1eN7nkXZE?iTUw4x2t~V5P&nOBTNS$Vt zK25R{y|#*eD|3>gYgSlhb0v0)E9<=8E345TqW(_$1L^THC&N`9)B_l2ci-j~?~XQA zua+vj{KdjBy_U*o^2Cm2F=sLt`2zWC=TS}jKjfKZvUOBbs6Mo@GACCCL>TBz6 zyW8qgCNf16lp2no&zns6xtoo6ecm?mKu@UT!O#-B=MhWn9q3z{$t+f&x7Dj9mh@OS zM~S;{KOxY%ndeXLyHvEmYEv!qwXThqqoSXl;JR#Ok-!b;V2*9_uqi^Y<)KDl-prpX zvE5yObIgK~9+b1wwkG@Gve`fmcXeO&1MaMulY~DD*1}UCf+DXC94SzX{YLD$1RUKs z!t}mL)lJCK?0|93@p7v|>)Awz8Iwr&(X%wBA!*kg+i;_ZP>jYwm!)dO8x0j(G5WR`jdQ4f=JsuQ=n)l7IX7IT-WST~*@bj%;AyWi({ASZLGi%F z_sT;wJhVKuPhY~r24LAqm58UjbG_Mv(y=C`(4p2wH@8NL(4 z_Xr!-IPhvW@O7B(Xd|s*A5{|)TKv+RMts)!3*I2UYS|9bIb{?Y)_@i9&fKTh-YZEY zX+^x{%n6bEYlbs`E8S}G=5=HiHpn!E+lp)1-A;KPRxGUbaW>?@GwWv-E21IL%*WVi z5fvUJguZeE9-44(isBzOsD0DDZ?s8vs?+Nm5@|dyz0hTiX*`qWIF@Ey94ax%?nQ`0 z`11(Enn+AkwKIqGNecJjGx+nYO!4UBOo~ul6%8|2!#A0QzN71AHijMu(O`aCTm$k6 zN@K`~fp5gM(mx+yOr;n5VCj#LFKZPxa=OJ5H#g9cp7dr29a zLi3ZP!!YJ0{Xk!QA7YP>C|0`*63x1vDq4VRXZ@bIE<|d0T9Td7fi=x9>KrA0EBBkn zL5ryL&>KbgT~K+1a#HIC9Of;@D;}%S2l30z3>{UA{7CvDV_9;h?NHlc7Hj(#=qaps zU`){&&$hGss^z8q-rW-InLq22rDSu?rrWZj2G=a{?5TPHPerjT5o1F5&d1_t1}J|a!0 zqov%$Jt8`aQ1)D9(X$LRj}&K_+TL*Gq6f2{z0vGkzX;kqzl@ zkj59C^u)!&aqlbM{0LbpcPh+7_VQG11xq{w#gY;CMMWormV66%*!*8KM7A231Z{!) z#)-Hu#hc(cmnh)YPlyHSKOOFVTZ_*^8_l|RQpUx_sL}@pUul;ci^lqvei0&zLp6c? zt9J+TkR!hLM<)<{KbBI?;!YQ1AF3PvbP8uP|KVQ4i}%&=`+fF7X7i z6{!vWw9~gh;(4wwJG&!~EU^_v@tp2#IyPTW#HWq}zzY7)iXi^$2+&bv>jJk{+<|XIF&9 zBIvxQwX0#x9YtV32O+bRl7!@td{U*<;amaJ|jiax(l3uIT;-;_`*pByXv2%$BrkwSoX--B6oWX?B5%LxCboc2m zp{yDI7hj_g%TVxB*wJ``;vhmFNmfTBG}lSJ;7LMaf;VqjTa9u&(x)b>Mfi}&4YFOM==;UiE`WRngm#n3g5uU zO_1;0PPxul(0$I#cJ!@wg^Vj%oK0%<9Na|DD^R9)buEp@!j*jCBIJx;AO6DsXGaW) zo&Wc2M(HHp(A;N=iiHjz#f0`0Jd-EgnBx4bs73>6*B5@f z?qc3hfAKuX7})Y3M;t|aoHoWU#um^H9CdRo-NyYW=ew@g;w}HuZI4&AV2J2ywb14; zbK7?GkV!BZdB1xgiTm&b%ddZPDxv`=elI!jIILROpgG1CLsAtsU}WY5#RtZ$YlO~F z^>(h$3vl8;xB-dxqW3=Xgl#s;8+DSms$G%?ddQxuPW(W`R>&~4RKD)NPRf;ify@c1 zpy;CsUkeL)XFWt2!#o=Qrc&5o4f~;6lGnlc?We-AYkLa_Ib%dmCuMM&RGavp&=P274M-*h{`6-uw-J zQ`+_O+bG+U$*n-+dp=^0g;ayyfK3A8wKs_7(&gF!LsvfWv~%AJeAnLl4Oawq_O67 z2l*Umr2^O_$cFILQ~xf6Z&J-WY1cPJ){F)W#w{xEN^KD@8}m;9rmhbTTngtdK&^Jq z$`~qc*Zf<~N}Ze{(C-~(JLPfP#IR<;q)6}L;Li4^%a@HxXQ?+Z#*^Tfft(Cf^3s1) zL-|-vt`>=ALV-0V8mV=p1^!8zeQ-?$L~~0f;O|o1Pk*h`@@RHiComfEw=KB=zaR3P zi)kod;@T}vwt+@fzPoPWq^XjB6LC38cFZ{;GL0u6{UH!Fxze``U{9>pZTF1QamL$P z6ewDgS9AuBay}@y4=tJgXGDVw8fB&&K~($hVg#6pozl zqY)v-GVK{`KVZ(-TCx4#{p__E+6dfw*8$2lqgxkI^ai|eE2j@BYFpR02wGJ>lHF&* z2gKn&%2Hq)LVJHl9!zK)SC&hE?UzoClF7!M##1PGhEP&oBhFB_PkD&*lW`<2xIFt< zLHRt`;)TzD3S0&V>*~}JfN5etK&+WkUV1)pDBj z*Wmkq8wNS!0Riaga(Ua@l@AXCu5==PJr=$|HopNWe1=46R2G+m&luq8o6*P092npi(oX8)WN`fkTc=fd4Upk`BQ)gq3G72a_sE zh1R2?74&nlxZkhw!_AVJnzV`mr{d#`|1WaB|{#+qEC1HoY4q-Pz6#Y;%LYy&8h2z%yNVIAo+VW<; z?|IPf#V=X&9^@2(zq`Qe$VNYu|JwD1#Z|zRU;jHjm^2U= z&C;-lni!mv7(skiCBn{*5k)p#;PkhT?g~VhYmkZ&HHhXHa&>M|Q(dIqBPfDj%Djq} zp8`iv_AavLSmOL&nCPm*Wtx3R6v=Xt*FH)f^!V>ii+Zm%;@Og1O(bb5zfF*j*zXK| zFKjSvb7gATx5tkyr+bn+la)}b=CdywgQ+ytdEreB1QlC0M9)iIAQPgSNmKL|hSIJc z$|By-&_pKgt5ADobVjKSIr)YFwBS)a6#DlZIO`}71z^!4K_Lb>o&L|y(ix5fq3Vs9 zTeP5SbX6t;IWrlz9<%(rT+bu`pvu@X^~x$N&=?q#5m9}Gb0bdiN4K{+x!20)03u#tYz;6Lff(S>MDJsx);$n>wSH@N_3tJ>$I(51A&=99n%>PhR^#WPf(_YVg0TDCtF5kTlG{fDrumW?ytO zCl`g}?)J*qA_VJr-tamzuR44Ibrm9zd`twcmdEG8LmVHuX+AtPGAB2qyHNxqpUearuZ#(I|`^uJb}q62$4cr zgb_v?9p~?`oWd?NAE?Fv*mLvqbZ00HPkk%534?Xe?vokh&|%7ayW!kv?dx0}@#%6f zA@SjS_cQ@{bJbxS%pXE4*3JH15rRY`J(6r}_&yw3?IUe$CDNw^_K`K)Ybvy4l;z3( zh4Ua91J(a7cKiyg|Ed2QvDfiDAqw!+F*KAE{TPAVWcF?xySIp1fUP9HLSAp86O?J0 zAN@!}T>Gt{l}rbsmAd)r3axGXG0}ijHr4Wx=PqDcssEHx=nmUPXH>K5Wx7wEb=A*b~c4&iqAwsa|KPmiAcuu&H1%rx3c;>WZZo+~g2B z|9O;_>kQ@2r0MZlcq&=?$K4|&KnnPD&%FmB`ickI2b(8i&w3y0OP^d}ox5_p2y-n3 zCBLDr6S2H82gfx+=;MJR%~!Z!y^GMdp!%oJ<@@Q(EuBQMj`H*wgs5EYjgPao`xjvX z`zRVLo~BF)t##}IRZ_e$LEdVYiA~hw$|u5hF%qmq^i|(IXg6#RGn4KQ)Tb2w{+_{f z;+~`Qb2vxI12~H6f~f|VGDFx05A-O!J!=f1e+ml*REqvBFZ=P~;B$I9_+G}f z2iNt3P(*ryq@crNA@7!x_PJ|N3#yJP28#L^J|b5${##FUba`{B?-|%{G}aPVD9L^I z3S9Op51o#24wTe=NQbS;sY`Ya^V(VmFi>FADqdtgds zv`8bI{@UWpi$AELZp7Xi)3);EoREpCvgX2X8=X z(#?~Y{aoW_Z}P(FE?;H`g8m%wEThF5R^OrpU>%R=4;Bx?7 zNz-+k8Y1@Eu}qU0h5Y3le=?V|8f64DW=K;ubNuQ z)o}1;Fh9Qn9JJVmpcDqB&Zg3wIx2NKVg7IE1V4Z_|L%ah?%n*i?8}mAM1Z#?D6Lsx z3^|8W>N{_>`$*G!;*38Z8IR{%L5*IfC9nAzaA>0sPz0J-yegu2TDg64d(8p50t>X0 zXzv;aC2kc9xNc%3B0WMzJH`(k_)Y`sW~-;MIxl~aJ$qk8-oHo0YO^K0f5;2%tu9_8 z54Bzau$|TvJe9lvubcI}@l!|bQBLf7 zcf$yS0i$wqU1fNX?g|7HrtkUB=+L1&?u+d^#TC@?v(VSk!<|OFip!|;a%0nH*Q1g4 ze-uHkKTPuylVDxg#L=3fLJ~3ynDYmYBu`ohHw*2*dzOTNjdOd!U6VScbsNZuMb4t#z1yI_l56}rjYMpfrLY5g5Bd+>#y+p zyPU1ZZb0g}p7q8ZSxOGXgr}qIwpWWlJ7KdUBpYa*zh9m@```@K@c`!{0J83zK0tqw z1+WE6Z_D&XxxBgJc+~o5Ax6}WEQH!N=K+P$^JH_4-wy!T^*-1Jq`&%j+JZtoI2@|# zaxo)LBg$Y4+Wj418XtA3gVi{K%i=GXwhTi~xt^@hC4f#zavv(#yFkt>XMZcYuHeB^ zBFuQ;W&~i`>;4po15Ce)?Uz-kRIex_UsJ2Ki}c?f>EK!PavO5&a}vZ^DYbUgA<_aI z%A>AbxP1RqPB9$$CA2&CWaTyrkNR1n3>y5+!HKP@2fP3sBdo4CG>n~x``BwpJr51^ z;S3W?xhPVW;ogX6h%-qLw8VY>1G=-^8Z}^)v@bO+E?b+#WrHHY26^Mi@>IKD>fMUU z{36kLS^H1<=yyI2B?6u&mO3U{ZMpBc7e-hIB37r>>y#pPfmU`7R1WaN92 zu11qNtl{3Z4&gr7b$U-qy?&iw_iKE9TIseQUOhnI2{|(xC`wX`h=3GTPN`g*4qCa| za_I0>+Jd}hI~Av@n5N$Pyvz^LoIN&GWO+FnX_H|NL952$|9f1V#}>1j1F>Kz23Pbz zpscV~6y|ez-NHam)+)q{A+F^|zNKTv*~3TVOjrv2=gV*kYoeSC*ULgVz+CEl&?butr4 zR`*oOKy%BH4vuEV8#Cdj5=Q5vMICl(_fz?6&yR!d-+V{M9MG$K0eS2SPW_=<5mbSj ze->=FI!2WQSOKhGxPEi3!@g+&Kiq`SG?{?_;<{i0y=+myTE=+>UfIy+Il*k6VlD&-Ldv^%xC+E-OGS6HM)bQL%A7cQ}o-zCgIA+m(y&d&yn$<5`>9HoqK}&DV@3@$Q)a3CbHzL1w zk0!HWaUQ~FZzKJe#{ZNSDaf2s){ka^TOe6|= z2&$STJ|5TEo^)h#>&Y!=-q21WW+{?=na$$*6q{It3Nrp3a3~zI9MjSHF*en{7;92< zwj_n^tX!j*V*V(9;%DdjG2+jx%k-qD|D>v(e_(7TGw$y#{ZRQ5Gacbb)nNMc^>0Z; z;_XD{_=UIANw#S(mDi?el$V8GkG8uQF7JX5LT?FawOdfgJSE>u40>MPqd4};D`Zzf z=ENN6u0DDESU>cs^+m{_QDoB#7l(;Rj)Q{_^epAd&Estna09DZDM%b$T2pHX4WWW;a{kxO)ud9UmPR(DEm^y(WaBd zMR-~8lmQV|^ijHxG?#Av7$9_6SBiBT(r1!^7iz@A$Fz5a=x)U026Dj#L&4uq&t zO9?)wR86TKQ3iL24|CjS(Qk3SX(8S};cUQLt_wb$akZba`v~Lp7W-HY>!%J%Y3Es* zKV++3P&yAX2BH*biEC#W@FHavJ!gJv3Aw$^aWKf2D_=3LQfNAL=6bT-_yvyJl#-qP z(P(SiAkaDcrT67!IZtMvQ*PRHVnxv92xj-cG^mWBxY~9!bNn8s&USSu_Sd}Tr`n@_ zCZS~}64f<5DLsa|3z%W8zh9_or~x{;TdPhR`>C=cMD(q)={=@y!-A<8to_IINn*dz z?cMFI`=Oui5z#xppNAe^CHu2HO!)#jl}EggC7@9J9PU>%yg3wLrV=|gY^<(HH+;Q_ zwJL@zX6vrAd9>*fnD-%aLR{@u`%Dhfa68*N{rG{^tswqC=5V$%6VXs$_@d{e7L8}D zA|TaY?qP{&57HF%X99XIrqy(w_*Q+vo|ksjH+ zkMGW|PSw;wjtv(ZDx@U&OFrL0;{9|!)x>{>e-q?8LtW$;JDlr_GyGI5s1RL#SA-sx4Hld+h#KlIK%7nO!p`Rr5_!ruvmc{y(A^FI5Dmx zECyPhT!lExmn_MJ4k>rbRkDf1zSf^(2@lGYYiWDq)^FD>_X+t(uisr|!8g1QmZipd zhC-c*ufEao8mx{s1hSIMLtkulVsKTO#D`AUNhXnImdi;*a-zt>`)?hNsYTB8qMqbT zM?j^V>lD`S&DQqf)h4)IKSB4u=X_Y$?|^9OF*n9zYSm{2&Awd1`k@c+6mRZM370QQ zRj^zwmOlET_4kcg$cu9<(y}Tny>H7;gqK}9VOLI1MLcf}j&zYRUx2H5YE6d76Zfi> zDsABFA50nUwD-J(p7|Gfz^v0Njkb=H<6)cSv5JEX+^(JDdP;6BU+^RqtO-+hl+A+j z445N1Z+%L~eGcyXqA2cbv~lWps0R3dgK5Lfpe&HecX}c{Z|dXU$@W>ihJ`+8!CRd5 zlYgy$41V`0ayonOfrd|*4YG`X1@&jub(7>l?p~w)NA~yZmy|g6O-B-Tq|5%o?o9U9 z0IB?P?(5C7t>$vd3acSP4g>GbSqe$K9eq5JXD29LyD=KZsABpUuqYVTV#NG>3^b*sfaj@xw8yU!K-6RpE=-J;?#;{J3II+MIUKgC$8CYkAzZBPrX!f zh6N9IJagwbXk454+yB`rPCOj94}ou<#3nbeD7?F?{J8d*f<#s3A24$RgI7+*Od6*=^jD?DXp0DQx7S1-{Yx1lct07RtGgT%E2g z>Vk!^esv?4p2B>}-X+JK4bVYqHo>IIEhsp6w@1@etmx%SzH5&NwRvKXXfMKNTY531 zBt2&bDWQUR)nGYB{T`O*38V|~+1$w6C&-lObixfHEc$@74y1CA$FUgGA==%FlRbcd zinFWS8vyp!*M71Tod8ZivA@1@5V?DRPN+cA`ygIIs67SwStCyU?-ylgi0%5NXk^Xd zoM)*_Cyd7O*Pb(6mruXv3nc27=1_)`l;C}vQk)+Xi$#6!`$7)u32Bp$)X+iZ1$1-C1St(y>I`G>-Q9q zsER>?aXp2kxiUY$Dx~8TIP-veZ2vEx_5dXw`6En*QHC=_W@?IdJqo4O1=n`f!*2CI zF99!5TO+I~)!$P7GWd>^%isG4a;O#YE)4nO9+|%n2JTBV&iyGKjB_b)W!C*oPE?~2 zT+aFLB&0aZZ zKKYfU?)no4JxKPQxSg6jT>uN!ilPHHU`4}o%#b-9lAPd`<>pEqU zq10dL$o_AzkMoemcasO1;dn0Ac$UjTU*suSJe2%z((({-bRIn^)y^j>H6Tb0J8>l;zpXq z;dX)F^0DEd(|4O@<`BK1Br-~$K>#i}@bMC%R+5*4MKEruapo3Wr)OtZ%WZ>FoFH0- zDUFYTUQ_V~=)S3v1|^NU9vi)44(S%B)?t2voL_vS%^05A6_?7NrW^M;(?1gVOnVWP zL*~6)_x(NkaZjKkyUupV5mJ#M-6cAm4!~BZanRmo?iOQ=a=8GuzN)w6*vE#R+({9L zN&R~@pw3+TqQc=U*WGw|2;lz`s^|z0-9}5F!k&&1`yVmuY;Wksug*wt>1_8F=@?== zQlc^`mszs_P1G;B(Si*LIOI-qTPJKcf(v9W!e+Dhv1sf>vKYJR@^n3h&yLL%PU(2T z5UcZXE*kVYacAzgEx#c-LEO){U$&rF!28N1x-Rql4a5D zH)MW}vF+`#SsmSMmXU*<`Ugc<88P+A0zjwSSxfV1eS3aHMFH{GJN<`#jnIuZ<50JB{tQ z%fWJDO)r?S42BN)es6u408TvINb>IiN#2s~0AbKDG72PkTehQq8pX}&4kR6swF}eA6!WAX z=k$B(gLYa%Y!8r4x%f2*p%#WZR0Jhm?NxsC7^mK(4m$*8IV^ih(S8N+IfLiEKFg3Em^I6S3Y>P4yH3@Qiu#wydyDS#?tSW4L*fi_*9@V{;HCd_gXL_95`NM{T)P z7XZSWXgeB0t#g~MI3U6?X6NK4e1@`-!NilMW&(q_Spk1{``|Lr9A+}fLiSe|Bcl{O zfFq3Q-U}|4H48H8_EK_oA1xWhufGrisBFRHOGp_E-q9?^2C~B14?5oISaXm4SZy>I z>A2qe@~@E(5i1XwKh~1d%tKn#VDC^g_Ck`q$_-mGA+ahm9lYtMn(wZ!3{{1Mxn2!g z1Kk9VmF#FKRTgbWOJ4KUaAkV}YSoTJN!ZcoE1P2j8qzpe{`aUY{NJOtls48oN=@M` zBdm`&X5yG!;zG;XNU$fM6FCn^Op(~gYqBh5 zk?+Vsz3BiJeoaZ>4E42ChG`shFK-~%k`qW%hcxS8!-t*#;QkrP_sWM=^{a4|9Ll95 zwr)?5W7TG~)%bDmgc7ndI~r|9gP;Dqi>MJwb=}7nqB)MwfroRWo@tAyPZt0Fc-CyhZwGRBOd^>15=1i~g-i~SqdOjVZ9t3MT%L;x+bD9L zHYMiWyIS7T4bRR+a9sE2Jw)alp@Xos@o`39Td5ZDAvkF^;e5xK|6 zYnq`P+_ZpyRsoXt5+(Y)ksRRgUZ~p-`F26tI)lPB9oi?KLkk%w*xAKmVMxS=*k-&2 zc``!)0q@b>$Ex+h48 zjg?!L;{iZE;vh`VtZl^6Z23vS;D`sAN#cu{!vY{TN{B^wEjlcitU2-O^f_H>tJd)7rOcC-znYt zSg*GaRQMu^=@R}v`#ec?MZ)Y76IF|R#^lX5gBdir{{*Lx3!Z75e9v3^ep!lOjlEEN z>EuL((iymmP%GCbuu^BM9s{Dk(DHVAj*!uCWcr~|@Wh;gniBt12I0zJQH8|0rr#;D z>b(`2`!G^&F2c52oT%+WJd#f@$IC8(Swa)7;tlG4kNVYg`zy62O()~{J!o*3M3-?V zoUc?uiibV$>fBZ`}3-{2PnwN zcV@KQ79`@Xbidz8{bH8VD>(}NV7@(arJvQXtF%~Y7xXJEP27@BsLi|Bi}MEDxw>)R-bFk1O57V7_p=W_4`4d8J1!>^p8ztML#l5R@dCV ze3!FGq^BuVEz+a`P-P|u&^t&g6E=(jMe|NA#({kN;p_9MQ`%wt?>ng*+K}ed& zVO??6b_*Coz7$}u_GG+}1HG#upFRvM`DM7s1q?>4+HvL-y9OF7bL&611)u_M9kJ_9 zFAqLZU1v=B<-?b8;&8ix}ysC3A3cI+`1BI0hyC-o6x$zMz z#-f&5_MA2^Qc05!+?>TtpW=$`e`7JcdeO|%PO)TJv4k&I?CEeOKD<9YyaxIhb^33d zi>gKQO9mk@d?;q0b?}cjX1VkaGv2fC!ni{;Rxn>iZ(4y0pSFbNv8jywm+tFKxL3E! zqYhv_T#f%TjC+l==d;6pn;6w}yKa-7x*zvAgvkr~dc;w?1e-P4pAP%RVL4%^lA+7l zuX1_sG~cjvUdFRL;`BZWuqUT?G;cimqd#HT{6w^0fuvSUdiwLiJ`aCuD4q25{jG)A z7M3Ks9F*>}E|(G6f%)#M6+7J0d)s=io1785 z&6=-(4*rO`WILlgpL!<(F%9r#M}%gO}b8w51lVZy%bgmX-pWR^El@tEC<_bT;A zgpUM=^*$RF!$;nt?XVOZmTcrqiZ;^C$r2nvRtin^#2`8wp>V|qlzFx z{h==lGM5um8%f(+>C?WO`JB2QpUuP3g7fdU1+Gyv_^p>OP3kU?-s6P1bK13ellPe=8DV)%UmhAqA)GcDq zqg>^ML+Pz-_v|jX$7`W$njWL2oBEL3(Z4Spj%JMI0`(ciF<*lxl1BIze&Ysqy1F`! z)Id8Em3kR0rU_Lh*RR~?l3PRrR`%tRzwu2zHnS9$6ZRBt&J#9;Ywxpl$4`X)=U|9gyZde9UI(#83G(FXE6(0zOMVT zPr`W@{4c)T`LDc$A!@bHb9O9DpDRKq_LQB{$Qp`y~^Q1;x;!eR$&R*5_fo846oTFED%}f@Z7U4tY9PPsPpx8vcJEf zW#?HAyg#GmNKKik(VP=JmJmL+5IKu{^Q2-0>yl<%speC51}yB01$d9n zW0jxxU_w4K<+dyW&6?;+B`*7MyZB+IV2Qeh?gi^&()8PV_O}-FiRpo7spf`lW?{t6 znpxWGawmd2^fE;z^SJfOT=ZU-auIV5{+KYZYgL&n1( zqQvPw%H4_2Og`I313uuByF_yT3^Jif&bkZ~>VD2>zE{Muv71lS{oJ~WMc*KFfzHJf zDBUxfI-S{;MZbkA^CHViTo^y&EAX&1g2o5CzZezI41!x~{THoyZ%oCtmK4xO4Edw= zJNQPb7fCU!99SGMNjq5$XXd|k&2>vX*c2&cW+Sa=rqie^KA&j!;Uoyzo*(Q%R4hPU zo1?0VZ({6pLWLZ@?Kc~82)QcVnJpHEZ3$}dV>IWTytObdM!JAPWT&|I|4J-AK5NpA zo%h+7a@X!G<69ims!Z(d0{6BA7Ou=-oYS1`lV#Yw_gzunhLRlUxQp3FnwHla8rgoY zXECc(PKa4mg4!LT#?gCCkSsVw{+;Ih5&3h5WRcbm4BZqA`%D`3SYi1YBlUBBpm9g zCyxElJH&_F5c#}lhc z{=Av7ZU1zqd1htVHQsm2LL9omE9X*@b#P1QoTg7t!M6-|1|fN-=gf5Ms|_WyZa{h# zbD;$8Bu`4gD+u}>vlNRBlWsIw6wz5}o7RmVDsp^d#s227T%k*MtNs~__kC`85_G{( zgoZcwn?3ixBB->wOXG~9S>T9W%o=HgsVsQTh(Fi%9#4_i}w&YOv&lxWep%y^sk zTY{JNfhYp3eE}iZIhLC~qcD~rj?4H!R%Ce?k&5A3=cU{2&*(=J98RbSlT`gp z9qlOB#+_M7M`B+JM)(N$nZ0dHwiNKgPPO3b$CNv*UtQ)$;h!D9qi70}301`48xc;@ zbPXh4Mlv29wI&Ol2BQ#L4%y3z_raqm-7N6bt6O)edBLRChkr51@6Hc9;0LC7J%o~L zkrDGX{3u;7@uLPkjbz0zITf+uxGw`gHqFJfUZhOU9D2PHmtO4t`Xm@k!3^w&ZnS?n z{X2NSnN-gg9R(f5-R)oY!qoD3Nb`Qdqj;ClH#I~9vFKDFe2vA81l2+2#LP_7g+~(!Nc(|{TRvv zW8SS{n#!~4E;tZzxSpVam_aXjbU83_B0an|Yl#_GCaTDaJ|!>%p@e*02759dUBcCw z?)OKiVn9}UaT0x6vAC&cEe5jQqv!sOkgXqP#`_P( z620PsBbhk5o0VB^{w{E^$1`y~;W->`)KlfMW4-g^bN5E{E%WYpLHJ zzI;k}K9TQIe=M+lo9|L=%I6nEou=e;5mm})u&o}L3-*858;>@=%D_1``ZW217>dqR&o+56RE$r$%|Jj`N7M=iEW2Qbrck**Cfo&z1RYzXU4vEj@fwLyM(H|LC@8-6}Fc6LEL8R^Qyc50zT?U;4@L4?P{rz7`AN z_PUTygMB1EM%dgthU%KnI77M179vt4lj;(j-iU4q!LSS{d+j=N1xzRCb_BT zw{2ei^dWwuF`T!CF9lI2{jEE zf!w%5-{1O``M{P2h}7r3rh3>E^MqeJkU@oJGeQA?tn3ZIgz0Oi=-aDcbe% zaOarFedmwc7QlBc+hiG_%Sw%!_j;jY#dlu0-OJr{UoxP?$xn6sW(x(ICGpUewq*r2 zGKHymh~sH;N8k#`7=ChGZbGEbnXld z2&d{Pq3{OT=2t(k{62cWyjF6~;HUrUt8qIY(CmwI8Mtvxz*==d@t`;H$rFLfznB50 z9KE^Kaa`{*J@{p+bj)tjc4G-P1*boJ01qR6UZ48LI7@MP_}>$L@wgq_4UZE|2=yJ6 z$NOh!`j~LjtlgZrAVs|C3`V*ao_dzzD>Jd+;Q4-qr3^9E8$ zEhe~pd&u4^=RT2vfIAK*QlixwrH;ZK2AkOCp?U-CteM`GiVPWsbyGID|3My-4!X$c?1&xytpoG8(X z{~l8V3Ek-)GiYC2<@=F^{th_x-m@9R`$1pVwL}r0?h%1Fm!B3(HT>~tVB+ldAkuLS zdcIe8Q&EUC{-^jq0eUdn@=_P*1kUS0Uk@0O#-AEUEJV+v1MP>mEm)I}`B)z3Xl(|3 zx-TYSp!aejr}#5sF?j@rzKbb*Q4R=_wlV9{^`^WD`G1VFT8<^;?-eOX0@~N)t){?5 z!@Qkp#9d4QM>t;j_4#xnL-D=jZn~0qjB32$PmZ{1D1Wj{X!OiEVU(^3;-22~?csH9 zLppw&Y^E#KLsW6qhU#S~P{?mk8d}ckfUBiR>bydJfp^iCedCcVo|Gl%;$f4l z(Ep|lVyB8NRoKefJw~b?EEF@*>LV1@I)&RBR_Yu(PBJS6yhCa(feCq4hRL-W#3lk- zlC@qft8x=Z2vhJy?A3jwl^S&)7VD#+CrEmB(iFFnSpTChli`)}u=D)B&M{C%%YF(B zmz}xdLwVDzNv}O(3cI-cbQOBC_MzX7e-!eDjzZUFkk}CMs#uD-1Kz)N zkc~d&DGH+SH#xR%9_zXDEGO=yyUvL$C%S(+L-j@fBOkoS0Of0xd_04G*?`E^Yn z1ME{pGQrDP8 zB=ff{CpH!(jXMzV8A|UwH0khsbF@Y5DJPr(TxGDS8=Q3PyU5i|NwP$8mWedHr&5H(0Xaa*-IQi_$GuKM{V9lj*5l=q1-}o}3 zyUqP3CrCD8Z^=MO4xjgt%}Qws?0jB96+$T?LwVoUaByQc@qfAGs_8Og_&TL|P!aJ1iA2~+{h^%1;tEMp=H5aX&cMfEGc8*P~c{#~E z{~p*L@Fd%iqh%{*fhz$dP`a0HJ>MV%i}cRWF#m}@Q=|a#5#(9{4&m*m0$+>v8?HyD zRWgs)x|M{GOE^;5r!$Ik0LT3H;g(>55Im+whrS}p_Wy)+KA(1+$^$c?#k?mWdrhAwP{ zG3m!?!WJ&3oW?+(F}kjoWbGY`iQ(}3_XySx^f0-GInz0apO#K;lJ=R&fF(zQYV33x9f z1SioC4m)-jPl0(i8Y<3m3!S5%-Z@t1mWJfmPITBi9s<9r@DF=&$)NHIi`ueWVX&4c z#aRi~Fo#YJWeBj*-^z0`4nIWfTO3AyG(Q*JpA3lTzml(ls4>3!wC}x)-CM&>f8H=c zmn;0a-r*jP4r@irlZRolOf8{RMv0|hY}*J__w~VGxv?X${o)1?t{`%n7b?1{qDm)= z`lURj{V@7tZ3OLs(j}^W?sEyw-Rx9-HEpF2tf>&HhN!=ookbZOLa>`i$O&k5L(*Ec z0%4je?)VrOZww3^gv}(+(FXiK|1{k~GNMmOEcuJBj!Wgd5u>N;?I)`AA?HkPzygkJ z0e?|kDcr35Z2Le9g(qxesiN>f!PGwCcygMkXbs{z_7E!=9(lbCFztlm{vFpoOJGxN5SAF3Z`KXE`JlFFgbFY2Kcpd%OQ3HDS^j4 zV=BcLadQJ|KU$A#&(Dw^!ZAP7uugm#!5x6}jESZK9@FGloWz)P@_0Q1J#j-Je4osJ z$fHl2OIe+nL(ndtH0=bGKAd|aP{|^njVm!-hc3g?w1sJ;N5C0hJDUR9H@jP~yg#RjCZpq*set}cV(??23W~aP=JRU;takxn;;Az4 znF`F%6mDfN>aLxm00m0M0@Rjt&XV=Tuq{Bo2f2w3h#{XpRO;k_j1wm4rH;0^PMf>#tTwVlsY;uu$2|T6f2focw;|M_*PX0l8?t!f`?A z#q`AygPb=jKRn$6=z(rovvz*V73lfUdM=0g%7(C7Gi08Hap!HJXW3v7f84-b3P-J* zpnJY030uF>s$>P%(E12;joNthquCDzrGV&S)LF6ZVV>&+nb?#L7%Nkd(`F=sc)|8# z^##rxi{r23UV>_8FA7g%7uTWt4|(FzFROBUxv>2Cq}|9@s7+H~>K?;OO}jKaxxaqhHs>>2Az1+Wx=PpQD+ zoEOnj1g^5+o98_w?m$jNZx2HoWe5rz=-ITPxa=y$8mB^~$wpRYm>)*f|j zzpk+<#*+PW8uDyACmH!JjpQ`_YatOi0K1(nP(m6dSnE?Rhqr(<;&->< zkvP(q@=KcQN9k}uWA~jFVpIE#-=E=8fepzVr_>9H!noE|Xlt0w3N3!KkGn9X0W}U5 z_dCHLdeutXU!?b82}7+q+YNv`wG@TPrVu>;O>PQ{7M4=t#lpq~TteKzYF@aI%I@kgkE;1xA%2 zQ*L5VV!*OV5x&)be(b=ti->W><~bYUL4C;nX)vhnV8B`t2^?y`wghcsFmXeX&WpZB z@vvGMNJ=~d=&HdyL=qR5yDS&^OORHx%4vo(keS&>ISD!(-oBU){d%veR4dFzXWw-k zUZP!kaWRrvzohQ8VxP4elDfjFp*u{t_CQyY1}0U&$neB8!|k*uua0R|E1i58^hsN6 zQq~Cjy?OJv0&&J=kcU0F4X=0;v~p3GbEtsf)p*NXan$j%cpo$C#>(v$WA{^hn!sMI3{0A!}<9 zULJAZ+%-0SUmZTVX>whq2by)XCE({i_3^OA{)GMukHPQmNN^)w{Rcy_80gE>55G}( zJ6BmX{7C1rr^66_7%^B&iU_cT=|l21MT3(W9P`Q+@78L9-$3i7TIqAwS*>5s_&`R+ z$FG~ZVmOa3e&wNKz;2|?~>2e^nakv(z-s`WLCv-9Il@X<}7=_S@$ zy5`p%+VmbdTpY;*he8a62XJC8v6RiBkM{4bO+$yu2$R2(IXD-QntBdl=}Zt81#mL6Ps~CGBd%3_s9T24UVof~t(EhXQ&$1&I1g?+S`)|04~bl@ z)${Q{j2`pL*D^?8+g<6Y!zjc+oZ={6d?w}r&3seWkDnDBe8Vtk!S}A4=qF$r>RW)K z!)$ehAbSd-8Gi^vx9qZj;wgG&bb2{foGB7ft|Ndx~y#>$h2T1Kh*}> zrej1*I)F<0u1}&XP_#hp2@0CI9(02%`Ec@rBy}4(yZ4eXN6WeL^5$`#1;z7^oLjvh zPl>|2D-eb6)2>%NmlqU&zj`>#^jkV2P9f|_NJ#OqsO#n;AxA(AJ|UEAs#b5rgtJIa za+}y2Rn`h5*Es%3S^KcchE1_>-p}w)q!PV{gLo0g&o|C4Dq^WuV)TQjG>sB}c3JAd zmJF|)-Gj$`A&K<@?RtB))!$z}KM>ITo9w#Ng0itbv{HXK&ZX$$X%0cjsU(8i;>Qrz z7ZPEbZ=!i>0F%p)!#ctDk=FVIDHTwVo=$q+-QxG+zvYQ}^_eB*ip$|iTy(qz^AdLe zIvODNIR3kirH>m(6m8tV0rHCss@xNpqE#d z{Gs(VL`f;)zh_y3EDz1)=BEp{P>9y~`Z5wO%vBeAq8E0Y@4Ou&(dHo9ad-ev!1rcH)ZrA zeJZ$Dx(rc~CpdC{7;BBMuz+SgnK7ToaXc_qqoATFX(+)h(W_d0eN?$$_aAXvZ&OOvZ@^xI_p%LKKH02xDG8H8_cRU0S zx}p#hen_WeC}0HbdOwqc;~V{A{56i`a^lEtsCF!ia+zA`EafcsLoCikW}%`0tAV;h zyj#zKi`o!_1l+b^deGe7z0m#rMkk_)ojT3#v=%FfT%UXQ-%A<5p}M%&+{F8*9hu=K z;6e(J2CD+VK2T!IUsYG{use(VKHF>a6F^iAe>Vk|96A#<3q6HC)lOc~2w_o8p6yvM zKaiQBF0)y=N~3-PEL85ki{}K@%svgx;#9yVVi<$S{sp+Q1nkl@^y=%)5hENMZRMQJ z%v(@=Fk~*tR@N>Z>gB;>j+1aFQ{O-?ud?_uISnmv4mBHlt(}FUK14j=fkZ-Wm9_xm zd`6N0H{m%_6}ZjAp%7v;7MBn*n?NDSSDjRxY}RY?dr5_M%oLo7aM{w&Gu=SW288cx z{Cyw1TEi}7SOhHvZeRumX|pv4NC!~5l9#t`O5^&&m1T3YWf@CnCPYtsfKl%S+=4*OLM9a)rD{I|(tZICPOLI8|nd!!i zC&mZ+m8~V(2VUEOt25}+Zm#aKtBJ6hd+s&tKXC|ALa#|eHac5@k(L5n9n^^@`;Jb5 z&f`qJ=~S8YHTHw&Yk)<|YNC^4>u-id!c1?RSJBJJG-BM!*dr{pV%j{Ac;}Bkd&_NX zES@BLA#a!xI|7-+={fqg8P08O)@dP;E|I=;{jTp-(IUz=Z zW@)cqXyPH(eLl2jLVotDAY&2vVJpt7;H=rxhR$uX7ySq^y|*OacX9mQ&0S!svg>m% z@K>R-D~j!R@Zt5+&r3RM+X{K7j^{IjQXKbeI^g_^0k^ZqXCg==&Q~Q3SiZV-?I^P7 zN$@)5l?n>6-Ac!OI3yKH@!7B;gsb}gbbjx9tx@3F*69&eX ziZ^1nER^T{*=r)+u!v*l7OHr(8`F)84~xqmjwfDKLx|yJ#5WVgTHKF@0100D->|9; zc%0G6@Oat8fcr&)f{~)Jk6y|;-tI$)z3A2~TA7Q8c%YHB|5*_66*lB}yz=XngkbHk zPVH|&&&iU7>XCaIEQPzaBa7>K7YF2 zB2MpSdn*Cyhjs2WB`TsjMrn3D-;>liCS`T-3_9@;Zb|$IN;`rsR zjxZBG(S%Dk10rOiqTknuI3OZWgxh$f--a9a4N0G#Mr^bnA~6P5ZK810JbQ*Nm=BhP z5iPG?CpkAX)|C@=sx~XMM4|d&ggVV&Wn}UDaH&l=aH@thG}S+r5D!PmcB6=s2qFShl8F%em`wvpgCpPUWwMjhN?OpKBq6B!ttPL12 zauR!;`^HqrupCMtPb}s%a{kU>-u3j)VA^|Z%`NnD0g)+pWB+)x*iGp_x};JsXXC6IfP&zQ8!8_P50;@w_Z)0L1T1EQTS7&|CI43JK!GA z%67iMdl7jj6~zIj3>eQL=NI9A;TJL2mzb;XCmNrw|Lb$!RBzlE~A%LDm|@_ zA(q~Iu$(9>P&N6IDk@sPJNg*izT7!z2S*=cQn<7Dxjvp0QpfXMl`3X|TkoHvt#uI_ zw}+>i`mzC`WdxZYdSFY2t<~@lU`ap`@rz>&B}p3L9v_T4J@D!c#aB*Xc6SJX#J#5D zw&rp6n~9t{upE5ig##N9?Fut{=)pM`68R@!Q4)JuR9P8L?CBTFiBlK*QZ$Zmne=|; zr>}Fg>P6CSx)SZ$(L>v)Gn+HJCFn}Z2Jbp~l8J4S@TUhZ%ApV|Q{UDUYiMo8l>52z zDkRF_aa|5W{|h$d)+tgcC?~!=p%eO__Mzg9{7=&FRaWGx_dQKv>VYE!wUPwoRAA@6 zHmik=%9b?W?cQ7a7;!P=V5QT~8p0z-rO3Ow^XL-cHdkaAv@c#cxj1bJ(;VS^gqn8+ zb*KF^_Lf%XjjK z9$mMD{4fqRF}BmG1D@N9-NVVG_$n5Od|5WzOw{x>uj)wfLPfnLDYk7S8In@&2rIc_ z2!9^VVB{+&vkNlELj4pd=5H|5W87-&{|O>AoOL_l3Gs=^NX5uUjhT+%{5tXhT;@F8 zewJ2^)7(RbQIj7d2IbC0OK^cZLbEhibes9}@ccFNC2h;gjj`@1Bdy=NVQqKfJ=a88 zssPz+%OX(lXlg=mO8@S!0;J*wn(d96#^nI{OtGwE1vfaa2m9%@yrmT*g}9PU+nbAB3UtO9XbtSA z_)boer?%GK2NCb|*$F2Boc3McRZ+TCMUC|4z$mTuwFYNxwQG_)F5bw%E*F}CW6Nku z;A4$oh#zeF41Zezy@ostC!*RVC*^-yG`H-)XOI+^Op=o#V}Q8^#hw6V{Y5Sjx{R~H zRTEPBu}-~QyBZ|t|8SPx%OtEu3FNQA}>p~=^zRIOf7wbCJ%^=Mf zyDjuc3D!bm36s@DK4;kSF!-85PF|zJ)pIc5uTt`OtSNmOS3}ZTPV}R{F0&kAUcNM8 zkWKRK;D|#OC3(;l(e;?CML!2;Fq=l&{D(-c7P;qV&_TY8I|7_cwA0c#<@WSFVqLTu z9sZ6sqa{U9-cYu~d$U6Bq{}E(Y>`%dF6s3=T7%3OhQ$WWpsNzjUtJqb@Je|!pD4ll z)FF4hsSK=w^D5*00WK8yY>F}(>kssTG~P#PB+v>MeH{uG-mLIvjg^39=Ag?!b zQ#5#$dhB+)_zwIg(e26L#N+Y|;txl!D6V7oPk~-&>SW6^=#x>}ut~zenNKNIw9HPv zlms;3SnxwTKhFLxzNB>fVMMQ*9W%EVVbNhCilc4}U6f-C2MjzA{I`QA*!JL~_${?! zB4c(^^kx#xc1dwc&yB2TD5}wf@RPF*rnE#&0cp7_gE9)gveV;K1HD7J)&;JExvNddUrsC{ zfxpsxK#w#R-H7&9bHgVEK}0v*EJKcWi4&2lGb!;ecRf$uG-olUx7TtG)l9&C_7(4i zUUFP|B>X8n^}o1OrX2R1Dz>z{;or$v0kwGcdfK+6dQ(0rf{*m9($W#c25wDNNl#IP zwk&`X|M8Hv`%RB&Cgi(14s!FK!{d!PT9@wdPvh0&DBXD{Q!NpcuAUB?+o_mbY)T(4 zqb!!u+!|<0pKh8BxD~=-`^8#z!g(@2BiZXm1!87VY`^wUh8f$B zwVF8P`8k>SzMqwS_|k(+hF3{X+zu|V0tmrH)lOYJ@%jM0`NB>A2MNLLwzJ5Sy#`)X z-vgXD?;A>Si;BnAIoWf_^LO0!DK68ylkB*|dRtNgz6xzWrWqgn`4(!+!kb#@UWGvQ zo%TJF;MM#Ai$aL(d-Zfina|WHM}#%@#cCSS0n(o4Mnl2R3NO#mHrdRJZd(>LeNp7C z;lO|8t8U|VbO-MLMtZe^%MkKDNm!oOH`>!JwG#AL>!`}ZJEF6g7y5xYPi{!PY)wjo$-()KFqx ze|urycYV3hMAOq8N(c2U52pP4-8HQ{ GvqnZxU$z;uxr!+zMk6%`}B4TZ@_P%R| zq;cmnet68{oa}Rk(gnv*yYN25y5TsU1deMz2I6kDEmuU?gYzjs0e94EJ<tZge~KQ-bK0h0cpqnQoAFzwv|L zrdThlAI{NRJ50LmYS?xjs%13xo9_JE>X;#LSAVneEMLYc5+k^sk4_#r%2~Dn&9u@k zuiXQ7f`?!qR*tlgB8nC)9-U{SN$o+611h#l{&8!Gy^&)VE;FX(&VOA)Ov==cM_bP; z+l9?asiW{7jKeFC$kp#WEb6pFnt1_DD)WSV1+6=25C-`i7WvM1le0|@sg~Z`0S_Ey zE2o^|R_1Xbp1TU%fL3iSA(h4?+;*+6>s26CY-1C;G#<2UfhtGLGQP@j8k@_*KUsa_h&(AV88Gl z=p}DTyZ6>GSF-;!#w*itw-%c0m;9@++shpB2y+Tj4s%apD>a}W4En93us8`Wo*^P< zXs`_%17WqJc#@J%@YM3jEsH3HlbB6zPa&F2fF~a7}VeU?!><`gaPw?u5PaCf#ki$*4EC|pOfc1>MZdH`E7lly%btOrY z+9FXBt2hqHF=%mq{B$x%#X75V&fGB%thzl5VZ)!DoI$?_rHQ62loQeEy2snn5L|2V z3xnVHh~yC~kQI8{)S9elMlVW@=6a6t>RwjsR*3gLl&&cg&MYsClVYQ@M3CJe;^bz9 zCsaEXggQB|UAcs{PgU+(IS6Q98K zJ<0*K)D>DmDdN9ZIUSYXXJH=2#J0%WKHnqqU3q;}9s{DHORdvf;-*NlGT-_O08K!$ zzv2kd3V--GjHk_oUpt_8fTNa{NE&mnZ)VV}|w=09F#f8`Ops7=8$%dRvvb}o3qz6{#5UK4h z>)_UjN8;Lr%k6Q+&gn056xY;?5f7w=B9n(PM~)6hq%WA8_nmfF@+Z$ZkYR9w(RZl@ z#mmRbs}L6B?u{NdxAb%E6_2Rj*17&8U8ecWlGV>YJGfsu4ol3wiJ5UQ#}!MBSG9Iv z;#?mKqY$Ciy`BRA(IY8=%;`E{(LOE_QvXzb?cUpu>;FauQpRY&r6CC%Z2mi&57ZXd9yGEdpm8XKJ&5HC5 z9{FN1Jjp-o{C;Xn;Uydjf$qH?JxiGyafoce+!a*?keMPdb-zjUbf41=*vR_ z$K`_*IInat`s)+#h8FY0fi6qErGs0SMj*L}+Q8%UTnnGxG#XIgYQuc3we;J!BWWes zNAV>DHhr3XT)<}=IPCjr>})C^E3o(&ScHM6w}yF82Ljblh^ic5rd%UlIb|?;QE?et z8-MLuHQVo_L~-YNfp|2dGf-LuR6*ccDZmc&u}-seZ!gT@9tpHowVj?{qYVfrLs3WX z5AIp`CnDaPO^hA(W^y$`nq_UtAizdYco$nI5?@?N_-6%1KT>E0G1;886=nn8s`9-Ff@na_@}`4QHnkG^}Fye)ag6DObblOq$SaE^>apaq$+r2(*DQxa1QY3Itbms zlbBs+$mZJ_hs9ZbL`-(5ZYhx1_mbIO*U1w?PvSxtl_`7bH<6#A;>y zd7Q)Wg1u^kX|a^6wSQ2>gdQUpDYwq+L#wu184*MedHVl$o<3wuW+t(fDd@x{|1)Em z-QeO9idZuWwbAq^Wni1e_^cEb=z1Nhb;w}Vx3iR+Xmvvw4vUHx%Z9MREV2;e8}r~kYUZ# zjTkT17<<)acGhkhjRWXs1(2ldpBl@(FI<+kKuPQh*5wilt`n~2YK}E+t{>qfSD)pC_;&O@4 zstaNi-rKhW&(C}O_#rLPdpK?gdbMBgjRc2~U6{G_{)wk7OL;+YK@vK@lI(BcL|jxK z#IG;9ggVwZ_ozpI_!R&%k`0YmDMysTI9HPd4ywhYT_dR*d04R4K61M~{8oGr?c;28 z^~i5^&ko>v&Jh-xqw-I0$kMxuk7#Vdbn~p%%@D7KneKYW zruq#+phlP_6zt3vh^&Gn;n@-o2k~i+U%bh_1^vvw+f2piuln{6EtXphq-zkIq_bN< zz2W92dJ?}}PHSSLB4piP!CR zeqoN`+-f5{wmAi2Vod*)e_5X}y6E?jPWWBNJ|ffxtC8hNLcn8~#v*n+-m$6hvJKvr=it@gi3QXy27b0#K zy^`L6mjD4b?lZh3Fk5bKxj_|W2cuvbh`FPA!^PIl9Q$ynOpAMG{d2eXG<*ncwCBD8ScffmSUO^X{3k~aY`oyPoH+ zL#pxQ?({uN=H^9Jh;UkuFy2cw>SONRYaAm?o`Eiht2@I`N4W&cZ|7`NFDc_QydhSR z&Tdv1UV92rUmKyRx_lW#4xud(yKiea{Q}jB+ePdwFUa$rW!{4MP?&8mOg8!sEW+wB zu#J8uyT#bX415CLKk;DVGDsB*ikcjkhU6v1V;f`8k5joOYNIe{BPaFnC?xkiQCOEN zWcdcMhe`qS&klTX{ECn43MYUe^0WYptPLNoKFuOiMSgRI!HX}9v5EYlRiR&jxKcCT zHiVzjJ6wPT(QL)m?P2&?>9dDe=qr(AV@RHhi}A#GvUNj{Egi@5DEh^{1B=0y4!-=7 zK_Fslh0_B&8_wY6{uZ2*Z!^mQ8UvF*S8a%oUga#NLe#_yc-Vq@Q^G%``SQ?;7UW1c+`R~ims`9ghlrL zKa$QmuBrEL<044N5D;ku;a|J1K4 zRCZJ`i?X!5F0xQ?Wd^=4As33tPPa0Doz+}UXXSd(*#6LZ1u?pA>2ManQ2!sxX4)-L z0Y`oZj+xl6eAeM}fY=YcTL`$NWHy@=0|lRmR2g0B4ShU8)QV2UD{YwiX5n#eA?r|n zgl&gcL)pi=URPDmqMwcICZ2K(5fR<$6PzTF|c6NW-@*=p$Vzbvtfi zBHv)oFC1Okt*-Go$)X#4gf?|sGs)LA20TpTm_#@@9xMRzrA@K`RU(Er#!Ze;O)s!- zo1+`CZnxUNuPzTFomR5`1Q=O_2jlw?_`im}2NY>QUX{PnU^nge{@bY9rM$fP@S{up z8Y>W>4v3XV94zMIv+d1lZivP+kjy$?V ze`O9?6bHv-33X0IH{`sYPL@l~<6SsB8@NiUR~5D{R{P3;rKCjP4eZrj)FRu3Yafsn zX6=SrqsD-csg=QT;PgO#Z84YYMaVm4`K>X$@x_Yblk{b~5vGFjk+;IC-}k)L{(K3v zoWnrw?5i(QGYaQeT<2ZrC%w**6VH|W!=F8|qgxR%oBQGour z{MMH?{z;T3OY4%6U}i43mAf095d#GFD*$DoZNQvt_k#{VLWl0TvFy5|(~t3}Zy`$s z05OZk)m?hs@ts?PAQ=;y#~ZSgVwg`&gi~=(B8qPkdd$2XqT`nao>sQE0gXF@)?ANA zxd@)}=;YQ2*k_l9$WvQ{;}2&UG6psQU_~>SmJ#+R@!?(4RNX>rdDoonES()9>X!CB zoa9Vj*`?vnK$db(j5`1(QEq|%UgAl39iYNNME0pXrfO@F4@gybKa@wnZyDTrO(^g6 zVbjk2Nj~O_K#6!G;UzcOI^G-)YF(OA87CsS^ImcuBWv;aQ z?QfiXB+JAv?brFh2`#;36Y?H4cKeU1XW_3g0;Q+6@z1DXF5w}TM~@0p8TmN;2Uv?* zTJ4f!3`kZE_$^w#^ zu*o>~+5=d^5%$%&W8~#JWy)B{k}>e^bcVnd4oSt=7P-(In5$JcCZ;!=Xjbt2j-%}S z*$rmWXT&Fw!*C8?&^(lH;0L zN&?eMFZ*hu$vX?1?j8Gk2L;|C*w#0ZK+PI$siXr|hKh7{e(oBk%ku|Nl3^&-KsvGnnXpM91CUre=%GyAMIyfG6y-?fx4|-JHI( zJ4HEJuut&6TVb`B+Q9yrF0(g@czjXU_70G7;|w7oFTbt@h=o6vI{E?x`k)yiHzDMZ#pR-ZBJw`kA>b0Oi)V+W=^A8VJ#q zejs-?lRnJ#>)=mHBK)CRxeo{jLxp#h9j0ux76CQT!#FLmL14z8py_vsub-Ek*(q~p zTu=E8gCF5CoBF~N&D=}cI%AS+f9h5SlJbUAKutGP!ZdrTWLdKJl2{O@mt48QuyaTG z2~`W+Zc)^^j;j}9LH5_sZ#Mq-BY%beJj+Id6Y<^LfyeK?47~}9S{HO))yjl_oQ7O# z8_yFxrswnxIx1l~J#uqOEVgv}!M*SiU4B>8r*mMAoz*)fKb?WgGFB%pHNQh>?^*=@oqO3tJx(PDN?9FsW)=3}Hn@4c;NvwY z4P|ahXEB@FV*fSEZ1hWhZDiO-YC~5ilHKkEJ(h+3Q3<=y2?pHIcp#K|Y=58>DbaTL zSEWVuJ`#l7^n#ubE{wI0LW#S~$Lo*sD0tzRrI7I{C` zqd-k-o{@fpPWY{qS?o80m|{>|lT*=1HOKBdrUAckfUou}n z_A3Z&JrgjWgqQ=Z2s#}1olz_#0%knLQX)4yTKnl@lCYA4Ov{tax3F4}5j6Gs=riVp z588Bmd*58M!rZv~HY5L1yeP*%Z?JaDrKIGVHTGOuaq=ACfwR}X%L~_x61@@RgGnkd^em)Lavl9^r6K>m=e{*c%oEBu9Hw)gnGo(!Tqi%f; zj(~xT@AtHK6U2l|!*-lDV^dP<&#_;)j$SuLZv;R3%|_p7;qCth`CzLd;igPZ671LS z=gnyNe%D4=b3L(i%2P5Z?)gAymcCGsVdT%>u4se!SD@cYlt_X|A*xgCt3QD!#id^J z$8&ey>>v5^KVg%%uawwb;@i8AwuR?4vD{bTz5EE37ZxDo)$lJ5vaO&Aa{C~<563V7 z?(bX(CzttAm)GIufp*b%u8&%1TV=8&6Wx5s6Uy20Al`m>!1U;`t`SPXj{?mEsWVgL zlc5NEiD~yLx7UlKB_y-igXHGRy>R%gjUvF!!r_XD4tz3{zu_6GdIEG&bJ2L(Q>XXZ zioEC4^Vxqeq=1F6A7QmSHp5e`TaJ~~Epv1y31+Me>0>R;ruJDpH!nK#MomR;>h|lw zfNlx?QZU~AY97KV<9rI{X-v#vWbQ@0`uI@rM<8a@Q zBjWaB&(P^Jjv3dazJRb5*?~CNVkC$Jy}ExVXzWYZFTjn(Y>!Rrt7WjE;pniV#>W1q zXt;VYZ-}!_pzF|;DyJ!)&3RefCbT!kY4n-uMwQ^#FEM9Et$h81!ZwXt@{}}Lse8Mt z<5yL}65!(-7Q|}OI0+7u&o1860oN)kw@<8wz7~F2q3~%j#pO|U zuISth*nKJ4PjKQv$Jl$*wBG(00sFbk6B;ITkdgxWQyus`cDoj;g1B&z-Dvk^je~1! z@7KeWtp3w+i_C3|Ck_@@>FlA0W|6^nJWF!HC1x1xx>hO(|M2EZ$a=HVSV}Q(wdcFB znJVZ}y<&~JeGw2Ymp5&YmSAR44d&N8`H~R>CtY>p&(C*Q_I1{={&ov>Ge2MMS$Mx{ zP@F(vZ|dtpETf^28xM=oJ#)W$F5kULpIN$Wwakzd?-<3PM zud9lIP@=C^wC%;uC&ki}K^2}M%_D}JotOOcdPfZp%$GJZ#{ewt<>H?!HzjwA^B<&e z@v3--N;`QHgLbmfVGADQQ;PgKxVt?V4S{hamI(^*ND0})F0QC{OJo!pZa4G~J2kQdeY zAzLDpJZ3|kvf-=?t|i=;h((-U0I!Zt2s_>Dk?Bu9lj1X$7yD}f^i1)H>$%VV_Nb>$HZMtSyh8Bwnx?Ss_EGmL zXN_O|ubWOZzmzyjasO2XSvQI@y{OMGpm6^iwFRc{8}*pY=mk2tmIO0)O5R`*sbFED z<%!Rm{#!-e_Sc`n*VJ3CPnEyRqv$L>YWEUX^O$WecrVx&RfeYpuCBHij9G}<`1OaT z^roVEt>Htdy6%R*N7c@k{{frj3>!vS%fN2Z(34GfZ(XV&xEZ z_Fz$!uwS*aN=(Pd0Tts&5Y{I*gUb-Wya;!ny;9^>itF1Sft!!CUhg-74>60A6@S=4 z)D9*hVBt%{GZF9?8@honqTyB*fs0BG`np0ob36+Ki8$H@8rYlcz6MS6NaBLn&{g4fNx-RXxPM%B zkGxYGd8gfH&V6DBced0$nMjkm3L$l7wKh$lxIkuDY?+c zItQk4^u~r)g$VZGQ!q}O#i}{+{u6hxzDQY_7ytoym9W^Z5{oqin?g;eTtjXIe4wYP z8!T|!XrRho!PrtI%mgvVbN@H|NpzLO1>j&sITEP_Lc{pE#e?1~bvL?A1QcgjJbnvs z7(L^iTWe4K$dG~5Pz@&43$jyeJaR=|nW#tOy3tbXaIhDsB`Jp~ev<_=n{ zgeQH`s-7xOO{p}q-E~jgjnZX5th|gUK7B8k#eP(XgOwj4+Vp<`9O#)gIC5KYCG#9V zN>GfK&xOMILUmuZ4LtIF<+lj|tRu{!dUB2cwt2BJ0pH9C;7N&R8G22XU(OXx;t@=N z1nx72L$?NK*lVqaT2XI?(mW`|{&D>ByJ$rfd613q`hCXg353$-5O4K_0Y+uh z+@wdY`En|p^eYaCfXQeWJV-`Hf|?e;oS$W5X33@zdg9*(F4@g|O*)p%h=4r>i9EJ; z<_!|1PAd4NcuoDU1wjm!ANP?7j2s($Z}x*CnbdS zJmqxf+`iTpkTgFK%I4>l;JD?UKhY2W3Z&JoR*6x??Q_K@FPe>+cf3G_J(Hey{8+uf zsK?@3+I>T>hU~bn47Dpe{U51Ij;9=m((&giU<;*RPWkFslVzUINmu%LX}Z%netypE z=e5n6By$A2gWJF-FjOAD4$)_28;8K zdosLN^ke7S%ciC6Hp8=7#~qaeHA#XmHM*t8ft=2nI2K6jd-o`yCSNxDFt#0N#P-0S z9C$Y>Vu6_SipT)a6wMSc_w3jwh(FiCV-^-SQT)!ODqWqM-*xnvJ-$C0-eer4oCBM3 zCA$^A`tRwEGgKjY{F>rgf=v7vw+Gnut^a0LD7s(R_%8}n_OkB|eLgS|qU+4J_hfEL z7@jY`CJ?ZTV-fRMMy!HDg@fK)v2|zz|LJZ)Z~%8_OcgG*0oVZZl651>DLC*toaIoP zUxdQQzdD@c{{X-wUbpAH*+=#*?p>4v_@)tFruWDk_uywv#c4VawrKu#@TC_+qEyVh z5pERy3%&__Iobi{JCRf`fE!VML;MS#${@u;MWBU*M>g61K-x@=3fSzQx!^3rPC2@)bote zeb4Y1#p~o4&44q$G9UppnW9-*RE{ z>FoTgUiu0hK+a$jyPwxPWH{oG)z3@xNEOaAj2KYq`S|THB`>Iey~v*sCmNgaQ!u>w z7M2ID0ZMKIBcW1b8GO<)Y~;C<=4Y#*~3r4kq0AaWL)7j0@$41+UKb}h~jQY zA%MSw(=H6`JNnIL49^6n1yejO6Gm(aEURc=8p4RI)03+gc+(3F@A0ql9?@s*gr1pa z9GdeRAdhE!b4ii*PjA2@pC{VFYu^QX&$2C%U!=a>1Ye+xJmMD-giA_K8Ev$|G>

znQXIou!7)}ofakzrhr1tUw@&PGh5!9HWYZQz)tow=r8@H%d=mk1)rSMG9qu@gTLIl zHx%{{rkP%|P1f?QN?2(&?m=N*-FwVQ8M$9D!G2*6aPLp;{hh|TskQl&*Jgn73ZEXo z^7dW+536fPmL=ZBXf9XEiS20{!878)V}oh_e-*`7EVFyB&y@(61yAHQDkiXM*^fi%xrK4@KhfYhf9B#2PEG^_;Zb~ zb-hG7f+u)h*v#=jc7H53gn0UUh1a199g|?6HsnmQ%MI8{wkRpl9fYnlp7L?fr%2eZ zcPrbDgb}x=bG?bMgpk7#=g9Fa#Mw_Xp*4inr5blT*9Q}Se&%C;9B#CBD*KUQrFX<^ zwrj_4ADuMmU3Q7AB!S6b-oN5x4&uF=0iY1MeGMUt;1(AVe#Qr=1b#873Hf7gyMYk~idptd5bN!9y;(!YZ9eW!5X` z)6<&wx5$`GULP$gvgC=kn**L4I)#+Dm0Lre@0St7AT|+r0*%m~g&*PSmBjA`1h8tH z6|b)g<)mkfM?nz->GNqWJOv9T8OZ?&-#R|%x|Vhuo?N0Fb33&3L)%R0&i5j9lmV`u4!az%4%_g(k^^FEu=FI(+`2k-#WRW}aI-&z=NGQ{w}Pfp-= zTg)El`hWY(NjshJ{Uam#peWco1;2Ej@^YB(E|8 zo*#bXGmQde{r|lSpl8wfM$GBV_|N}FCV7biVw*ZhjdX(`DabPv{dWjCeFle^u z0?K-9FS#fq*E+ykfL)pGz@J@Ukex2C3rR!KmALwQd8ro#x;A zO%H*R0piM!FUyGZb^Y_t=z%$73F+SN1TebtYJUg7PC3rf1oi$CgdeA>!rO03N%H6Z zMjK~@%HcMI&_AsnYZ1Za^dvrlXKTZgS_lKbE1b!>R0zH@sIJ9S_ds(bpPS)SJtTRl&I4~UJ%(>pe5<>mG z2sDx1orE4c)uVkjpQ=0YsIFE)m9YHva?jB{y*$mb-N)*|5Ww?{DKHcKrFZHN9S=PCC|f`N{f^KvD~h;3@k7IfL8 z;oSuXDO$?_=(M)(?fbil6xJD@Cu5iW*>ybP?Uj@~L7uP)p2F2B4(tpvIn}|H=grf4 zB|hKaS4B&(pB67FMqVVlJbPgFP4V7J6QTbDjH)x*616f{tLYgYO5C4G%22=hJlw7K zes#Wl29wZ(?9;)URVxEia8cB=j_4#ol56emyoT z(GveVQd9a&AYO3)bRssm#PrPctYXPbq5ub3s z0g9*uwI@nTvxm@!z0!o9&0(skXSaww4RS9y&C{BClZ|U$8oIjw!^(YPm*Dwk!svMD zQJ9p=XODibv#gGA-2MhN!j_(4-r|90^{&rh)^;lbi=SXNXNZwI@dL(3>1f|?*~3=~ zbZd6>(N@C3w|PbM)n?fORK#U+(_3#>PgxC>uXnl&gWWyI*3q|$+ZVg6q4{L^7k zhbbB=7&JvAj?E>C7Vp&+rvl0E0HDg7}@C9Ez~;WsbnEHs<-#dhm-O4a;(mSk1{qgBU&>46hu z!|!##Do3ZaeIEdxhmQ`E~!(*8&&# zQL?|(`dH3FmmnPyri77DUOpz^6uHSuL~<9soxO)`{dIh}sCLJeoDmF~VtwHmJtEG( zI=-$2q9u&LX1#uPrXpGbihAiF_WCAL_KHea(ideQ#D=ah0`@*AAzv2`C2w*#kUQJ4^Q%y^+)&AWp}35 zQ{yiVg86OlRk8};G#a$Mz{_BSyH}?OJrO)HaT$8DPj$+KjG!n$vEGHQD+)O7ocB_o zg4j*D9 z9SJ&mje+R&bI$@|j*oSKak$+HO#jUZ#FiQZ)C%t-ziYVB$l*uEAF8B7V5_`g;f`8KVJ>`@G_NM-aXu2063+ zN?D+#1iqxztPzF>-hsJZ)_Hhe=*3H*M~kWX#F_HywJ6N>15BP$ok{?3tTC|RvmXT? z*WX71VeVJP_J_l^S}DQZa;f)^5#MXusNP_9lY;`5{q}^81wd3Ff@@MY!2u_+u+OP4 zj0qzreRngPSe}2DUk~J49af^cbjex^!4m^?S}d4>rbw1UV)`4!Fk1rVJ6|hU@R8+Z zou^_MkWV9Wz%Q#6YAV~|fs2MEZ!Kl(W&8skbu^_dBaWN-sZcj`hhE|=)DXQBU1UHl zk4pt=K#zYM0$6g>nJ<4Z3ncFao>~Za!fza%HeBE4AL1EV!}VcNtFm-yCRS?D+8`x4 zQYPTorKwRx3MmjHn+v?wa)C^yU&zQu@g6cj3g@`T){R2-^ECZe>V!&0?iKJ-rc|51 zRLR!5Dsk>fL0U2snQl$?oS9!!yPLjc8*p-#5x~fX6yHtkCbwRJXvz;2e`Vch?Bx2_ zwR>)R?SZMl8%i6u8}S*I?Xh|XYLwofj8ThR#4Dg9*Dmn-1LB(Lft0er^Bhmcf--Xf z=>zizyAwHeBwdp2(m}#+I)s-vLivMcvt`T(pQ%83a7*F#+92DFAKw4UrWqMjA~&M! zW=kAuju!=Y6E#2wozOi8byM*eXg>Gg0(4_8z+aM1h-D&{;q+dvRty$B1S#i_!Rj2g zevVCgWG2v3P@2Meq$*l6qD<-jIs<3g^`0;Ir0GzjNXNx7d$YMQfw+lC`(R;sGQ^y| zjgM!dKpoV*|6y;-Sa}dFuAYUuAm}p^5zN|ODhvXj&)Mg+h5vx64D;n^dNE~uSk4ac z&o@^WqI+JQ3@;>7$aA;$q$*#A%Naw-L4PG3Zxv1TSP*t53<5Z5%JFA*fIYT0QBFp&*yXMQ>JrE7w%NE~+|AMb z3^rYvUx?b#U43VSjA^~4=Yu7aJ1%SK|}Q_pn;x<&*#xbO-MTl!R+miw;og5hmPcG9Ph2`4Eg4tT}%}H_yqCTTbua> zIF_w*u9?AwI9lb#lA|e(Z-lfcyz?&RZY+0D+E3!Qi1vyZ3EBt$1LLA;}XqH-w2q`VZF_D z8JWvby|K3=YDxomZX@q~m)^Bj7iqU&FYf>wl?nYQrcr=ywTGrO><7^wo0RWQ=srJ? zW8)xOt%uVz=*?rkWlk%Eo(u)kpWIhJOa6Wx&eo|^xypVXH?!YBFBnH6jcGijPEi^ArVQVXtpL$pgjBLi{{f_ zd)8fV1#PzPwS2MDrTiBGFSj03Jm(BIwSUl*a-AGH7lhoCpprXjTb{+y7>eFmZe*nJ znuX7DyIW2$q=ZmX(Lu^ZW8!B4r#YGjciO6+?B-0LM#RDWRgX+=?C0sUGqsBPny~lWZHzi!?`5xWB?u z+%eC632KC?sttk&vYN@N&X(*y?&}Q9u`091GF_8FC?(Om0Zd%?M5plA>D}ep!BwDh z`t!Ehy^aR6+-~|3|GJXqXG`4uK34p1FiD$XVI?H!>hHVtgLXvBv~or1kK-W3+N;M< zo~s{A4^pNI=St>uB22%dR$59pla~AU1uI-3lkVigWJOpc6XlO}p99lBy}ws?ts9Vb z(QBDZni6LL7qff<yEedwzm6B&z8O) zy}aFJ7Dd`)e?Yi=-Z6UX=8&dxG9k&dB?$2+xU$Ts_R3u=7sPLg3x<{zDSbOA-&z0S zz9i*jRSE~+l~L$B*u!F!3mI(^y)2Kr`zXV#z4vrWzvA!b@pbqsyC`Z+8m!QiVbO3_ z0?oB9^W;^A+GCEzy@i{=PX)?Y_^rmW0K(Tg;)l}QZZI4=?>M}^wpsMzl6~l3l!?}3 zsGAsT>7uOEKWr2tPIR>-2mV)gYL(t)*aY;O0qmQBKar&nWgPBI0- ztAmsj7=B@${_^?Smq~rB$*RM%>+5Z1ez4@Bbyj)F- z58YtBKH5O&UzF>pTArpe%SNu;U)#0&zFpV<2p#s|FsKgox625t^>(qZlRYoD2DIKy zq8WUxeVn%WDIwk9cfqc!{1Rl^l-s~IKb)y$6ma&YP`adc6ffjseXrak8Z`8XI{ zz+L?g2h$c;wu!yr^!-g(q88I@NDTC`?V>JK-%y70!8;l8VnuT$myEk;$G7z^b!oRJt*_G zl>3aetrT<0q-xGypyl)Tf0(qWAW$*8D74MbMAJ|G3QOq*tMZxvzW7GRvAsWx|4skH zxi)OhDAX$Z7;3ty%P&%IS4sqP<`qpa5MwHEEWc?^s)H;7>tcvsFF=H`w22)OA&ir% zva8?51@9%n{MI8${kiRCzt58uH|eivS}fI$FQq*}9P!tBP@R0+Ak}2v_K8c+*siq# z>+A!@+HK%lVEQGkFn<2z24v>@3mo~HsAqA=NFrf?fs}ClF&$O2F6cqg z1kdIS(FmygcPqN$2(eM3mKF+K2)=JApxQotbDO{NK3>fmLth#& zZD6tc$x@Q;jm#6S%jXyDxCZx7QpZTrj|QYgs?E}DpGMkCVq4n}`bcQ?Ek3r#O(R%; zf$cScD=^;OM9z&rc3aHKz~kBgCW{2BbSTS5IbApzlG7h;16P!L;T<T%x)Axa@f`yYy~-U47R>`COKi6e>@0)hRibZNG9B;2*Xd3WwMRBi*Ex~Y%237!^8ShlVH zQY$^EVrC*eK#1kOE4a)2NUn->)x%uiUpUhmN$=)Xs+9l` zXf6ApPxE2Fp)K<=!Wk~T5do7t7tct-ITd{6vfAeF23zy|G$h_*5rY9P@$ykHciYhY z@MS~_Z+#ndV`4hVei=3HA*~}t#lH`a-d}FlJbR^tFKznxf}bH2|SsndbN`_9+(Q zaNQLI--u^v0C6{qe;^Uam;>fouVFLzZ;tnU--MI|L$73{X&%6ZtDUU&q!Iv~c1xtET9;PaSc1gdC07}2VfB4F`HRu0JJ zN`>-Ce{(Pr*t$O{kJ#pi?M~1vbh4Qdz?Y1N5@1&V4EcnU9HRtaobUbf=5X22h6efy z?=L&yvmC>ci)8(i^B)*N?(ltp{=)h{w@Jf^z{foT&`^ApA>Ge5DE{0SC1Ou1q+$Q_ zq97Hxha};nbx+aI`lK+6`ZZL#BXn(i*ZMi+bY`!Kq2%jYBK22vCD(1b-zC6t<9rU) zA=4fmux***7b*jvyV)wv2Pe*n%I&7%`hbJQB+3ZK!u1g`5-z zHi|S)%y&9@et+>r5fJDmnO4L)Z{0PD&8Uti6g2NOxhN{kmf(*zzfFjRo+4l&Z0B3T zb?M#;LXQ?0Prp!S2qo|}j9jCnBO^)kDHA;Gf^=`zs$L&O05u`EX1?Z}2{fU1Q1&L_ z54(Uza$oo@y~Qp_P)0hpHX9FNbikF*rjpS3Lvt1+u_4&klFJ0>F{>oQ@3n3h-DZlE zVNYuZyv1~%Jcsa_Dj|^U_yw~tmib_kfV-|R)&!dr*9OGH0#tIgfCo?QMtVQMSU5Q} z+NmdGlMh5jNG9|1#!yCPlie~&346emL#ot15KtGh+QVk-y&h~$2>e?-sp>ynyU9>Q z0H1vQJWoDd7y*CMw4`qzAN1M1veyvJ3n zX|w(DZ}YolM2JsbR3p+sU=SaDI#b2mJgkxY5WqA%HYjFwAir`0$CV;pbLJ)k`pI1wD%<1f84x{T;w$^EyWpwNJvxs(O0jTxg zCgb&D^7rTfPur2#;617D?pj6=SFf1EcQ#9Q5{F9%u&OTyFyl0a^(ozk{qU#7>r~5I zMJcG~3D!0slMZ5G?|Kvd8(KkKb4HUk)UEh|06BwTaqnd4W|F-e;G+upSMuo?8N0+N zYh1Pg8Uh{>GT&~3nNALc_(cSvH$>7_;Qx))q&mogurt$M6AUl*hL@S@j*P`NLyFKc zZXy`+GV#V=&0u4da^zARWAgp)fqfQdh;dyx?ne>0K3bbQc_x4X0nD#&zm*V7d3dXT z2lYdbj`BU5RbV-sys&ChjC$^NuXIY)#(4D_!(P*Xx~WhUslU$Ap z@$AR4RcV_*ZMId(8D`{;=*(5M$@@~o-5Oz+d*WvT;W9V#GU2$twio5krBTc|vsUOS zfla63R67*B2%NS-%7L1j!tsZ!-@E*nx7R|o;|UW~)dDh`hXY2>54OTT*NizIz$k<) zt&c?k`|)3tsFkxM>A9Hf`_9nViqPYYRFrvtm?CXYEZPJdTD8h9u1eWd$ustS% zZF#i^$UEuxylVDtOnh*hc~ckZd+Lq$Qjv;rP@WR7UK5D^XOLkg&@h^A+AyNL*He;> zgFT&kOX}~&^Gyr*SW+`$@-&M%hkeV*Oy8S(EaX$C@1B?7x#8nt8m#uE?l%t*OM~es z>Aj}=yuCM|%h)W-b$w!6T9=UMwG;Da-$*#K4cC>=kI={KVbOffPx3Jk<@opWaN_N; zi8BVL--@9zAAiD}Jzf`6nhQ|8+PFtfSXxc&pJnh!<3n@5lYXV6(c-m!8|1*&&P$4% zoAhXCO;LbY6c7fBzODI?6FAMg5&pal=46otxjNWNKb$zQ1t=kej+V9rJncx2#^0jH zy=pZ|!;)WY#@Id^>2KhI&vFrDZ6<^Q)odtr8IO7I8YG975X~-s0sYEJ8W_ zc=#EuQt2A&v&m}rXgvGs`q3KdV=VDCyp(Qa*OBj+N8iNFA!Y-W$I+v*53qksvWj~z zU|LNCN@Y<48LpqJ48EG)bt%UcJ1Z1UA#9tFJnmAVVpK30+iN20k+6XK2ACPZAQ!pD zu^FwsQ6FGx+I@7 zP-Si*SQ!YhlsT3)VJ~708+vUzZ@ZXvArkg3Q@F-62)}S{+pYBsq4DF38AMWNGj4aU zsj49QfV6z?vUl&CR&* zjn>U<2mc#Xc`$;69BNYm^Hf|^m12$>XHZzMeXM$4TIyKVNPtK57`Yy6@VH+PS~gd7 za2E#=9Tyv^bf=RhBSezEAjQenBF_s~|4}jNdmN*2OTww`v6rzP;5nMcr8^3H__QxH zb*pt4xpLod9%G{rbG>{#+ehpg=P_gq_SQ=BMy&unK*GP-(v$Uf?Nr=Tb|q&3xZ>so zTb@2fN@Mo?_plR4E-*2fqX=!P|8g&YXeAIqmE?b1A|2DmVw(^M?@Qa4GGxpy5y?4{ z==Erv;c8~gybKm7Voy%&-P5#{BOljqg4Ly8t^TM;;k`nw;V$PFdYo5)dw|}ib{iKK zn2rCsxeTliKsU0>ZztQfMxSP9RYtYRAse@WgG1D92oZVp48i&0^XadB1%hw$%GVCC z#}{B#axfB8a4>?4@r+AQ$c&)u>@o2Ey#Z}n87=iYMqYY56pH8|`yzLZ??5BAwEkY1hC ztxbr{NN8~!F{&O!NeNmSY5{8Av*X~lS08=>6w=o)!!$fo*bMq&*H9}mrgxUf_{cfM z=bAV12r)&LvO0m?TDLhy?%fyrZHR|Ss(GK^gw&BF@mU{U7ap2$X)1(jxQxbZy&v~B zvpGX7hQ?V{xi%HU&!(CZKVZBbneQ~Ki1aQn<=&!-?;~$dO(OV@RPI3BCZCoIi~8li zL}8~}t-{bRnqyj3NKY2U_oI4qygKK6*DOlrC<3{z1ZkKN>=OR6bM$AHd#&djm#ehrR&BN&S{fNN=}2 z`+o41x#sWN09hD#zl`|pUfI=JSKjLV;ZlO1{3pOl|MgooV>ry^Dh^`LvdwwF?-?XS zJ;sG~-u9O`puLQEf76$@_RVOS>*u<<(^w~&7px-4>(5O2w0Ma~cQ3g&*Bljc4_Rk! zVbhZNb7o`^N?gltsSf3)5HP!ivqX{-F3LeU_oO)X6dB1F#AoQBwleXWEatS72}}$w z_+$)-=sy$t0=Ul7(n0qK_~^UpJPLG8+l_b42NM+~3;w0!L}XgH{%pPFXahS22195( z5(V-N8u#fxzW!53s??HhuO?uEkBZKz{lB#4=Duj^NBXtqW{f2RGtt% zH-HkzcQ8`DFC@-Hz+@MbH6#)5$OynXWo-B?xC4q;%i{Uaii>QX$Ue%` z{F7w_jeutE3aaVUkw0g&)jY}7@H{1O<5C!L zy#NNg>r8C>{zUBrf`6d$V;OOSdpukFTIL`lAC{JL=O{567s)=rb3-K^*N5=yhuRLc z@@Vbg`Y@+$_#D_mJT7wZlpEIE0mKWd3OH43=mM0_$xat2IS;0>wtz3i<8yY_ zS&a(te)mQUHOA-F3J{U&-R$f70H!R3hdmN#J~?^+H2{dc^%!_+p_YYG+REHOZwA_D zSz%2gPr+*~XNbnYzJu*0NaSy33fSXs<~fS_Nyi6RB#;3<3sv(-&dSkj`=C9ZosiCe z{S&Z|SM@_kx_;N`g(?9|IUQjltpFTr8$ff3&`}$prd0TgxUs~3)})pAkoa#7`hHJu zlB*N&IM8eA08^phPZC>MK@sb&b1F5N$Ra`~vYjmb1#)2ibWE1x^#SZ>eKQe&gZMXB z>N&l*2p4$NiH8utwzPFM(2V}0| zH_jc5*epit0;mMAyauq{KVsXo{|k4r9I9SZ{!?j_AMWq33H% zJpgSObqvlJZ|yn70ywu+TKqzpM+oaSBjwoULX5)c2p|x;e@)bj^n z&H2G8wqNi?m))097t;7F=&N(2robkrR+bf5#U8=}e4M=;2 znwGp2^p>@J?{hBT9!Y#M2mR>0c3WjW=E@mAyI*j`LuWVEOW4W__T2~%n96*?%Y$Ah zqKO83BDyG2n9XINsC&Bj8Y|?YNFPU^-9k_Q6m{MgHH>4#&Q^*yj3m9?9Q}9j1&Zi@ zF&4zgA(w3hy`5+Ayh7viJC(#XQIDg*3*~;9qxKJ&o>5Oq9G7+(@J0Oi>C*Iw+QtXv zgnoW2=ccBd*TPSpi-;rNAdJ|(vK2LjKZmx^jt1;+9CYYH`Cdm#hFj7^7czAV0qw_? zPD}v$-+)Zy4-LO!)fBgqkSYWh?_R1c{Kk7Yq z_qc{jeVhSI3!;F5qDm*Urp4vTWn9!xMH$oZQEoxV!2{N~c$MtujPpGlX_w2Vv^h+r z<^_-I2`xaZL3k!Ne{$B~CX}219YE^ z`{~OTnxa^a+B&@%>!0q29i)g!o0A842B1x-H#`5h_>YuzP%`>&BsVq@>#yd|m0=59 z!-)yllH9nI*YGZA?dKjHS{JdjHap#d6tFW{=jT`Imb)Mks zN~!4Yw4t%H4+`F8(UlR7#KCMGT7BQbf8{wex@xAV^`Ho~Z&4FapG_?TMWWq)BTYsA zcso;C&e0J!yWjPc`7+PIiz=t7?EgL`N#ry9p-J^B?TE)N=Cs&lYopN0Zx~gY9iZG* z@e)cZkCm*i#@>p!$l@`ja@sx64{9GGUZEYzV-14%nf23b+#-!{0W?1|bZ7@y{@6#u zNt_zHg2!VJYMw1tYkz)QL;uS-;^fcV8(DWZW{w2B4QQp3nXqk|{{N{ZO3V=xo;pB1 zEKF~!iEMZm(I?KOU5;dJ?vyk6uAoVjRJ>TX(l73+q-&EU?2y=okUZ~?+Prc+2LA8k zRoHBi*NGEdQfhS}3qKkBP{SC}xOkX{C_^DHm@yaricev_BDcd}n4CeVE^d?u?DNlX z?Pn_Us{xJ46I9)~lp>;ZOYECcYS1m9EJPrq63c1WXh_S=&wlKL34OXC;65f3<{Hj1 z7f1iOVF)6cQx`X@y8B)y#wY9`Q}y803X4x+Shka9+T~kUxdFAX3@Ocri4xYjcl|ia z{b;#brSER!Q>BdIq@8!*; zCwv=f8jwR=jh&x#G9<2@DB*1jJ#?ruz6=0j%Tc`UEJxs(*T;a8tnVEfBJnZ1Jo4 zxcC?~41X><|97&L7L{eME8`x}UZd{D`bz;~BVVBbDAa;4R)w`y*JBPw>VfWGgz}G} zSw4lOY4IbALPW_C5Awi!npguQ_g5;)YFBu&ie0sN9SqtGd-{Fgt`+wiaZwUg$#IIB zoQJ-x=?We*r%65O-hni#?FjcoC`#AebI`p?uzfi$==VMD6-Mn?YF?Zu zndS7M4j`XF3_D4GoZX&1n&`}7a*`E9^ zr0hWAZQB|q?131QoDr9wu9SYFgvmbujS?W+#X@~8-hBeNIP(HmZl}f*jzk@yUskoC z==|Xk4pc8w8e@aWk*%AUpHeFaL|o6Y1c=gbfL0!vm|k$$$`aeir(x!oX|~LY#&c_*&X_sR@Z{S4J-XH?m zO}o?LXJj;=F_dw5Swy|n1s1fS5t*+oj7v-D{S#)4Vhy_akFMOPPxbn&P85t}Hc-sf9qDJ9BnXFUpFKqp)6~eCvzgXV z4Urm>dHUOdpbO-?!48PhxZph$n?gd+mGgqG|A_Ku z_s3SFsJF<{xY>aR2iJnjEVt?DDV=6?GO2v7Vha31FEfWjSrigBLj^<&oKzwX5rxo{ z>^O;eeh+aAs`e39T04s-VH{Vizj7TjOAF~xHfi@HQc0LE{hHJ{u&)7cDoe?fV(&k? zeRSIboU3Heqj)bk9%RLl%x6e;MatMOBDPm)K<3-1k{;!0c=^>OF(v|1pXDgYr-DyI zWCvnt88o6p@eWuuXy@{aNmRR*Ah}_0@fuPtvt-89Ffuo5P+JEa_j7e)e9A|xS1OM% zcbsk#4TI&#eB+18H6SWQaP(A+bfbbq(U!-i%sI%c4p+YcG8i8PEO`vItY&(!JiM&xkup&X0b_b^^rCekMx(hgG3{I71Cx!lCKmI=-I`Qsa@VRjKxs?yR_J z6-I*roGtSQ?EB_Z>56>a4m9uh7RMmeYy9~Yzk=wuo5vjnsb0KK$46(Nx4E>LE;-T~ zo1e#47L1JuC?qUArWPuD8%<817b5e7+)If$SOKUtJ*2AGx0p z-0yQ?2h!?0Xyuh8nhd>lp;|$-0NQRYh?1X{+?cz^&j0JT`gCQ` z-My3i?I8%u?z_uxp*n5S@&%zTZS+h;?>|Yki+YYN+ojiIyxxcjD$O*}b*9DtJKU!u zTd!SIvJdEmEbVe|pgw$-J8*`?K*Lx(WCfL|Dlmnni?($i-*~sc_tjqLyQeXFJ(Q;9*x4&Hx7i&qzP%ar3K+~Y=$#*EoRN&Wr19dM_#_EG2&{8kja%HU+lv*JxaA1k0y!{&V*jWPYwH@JA-|Wd0!`KC&1Lz>nSVnr zHeaQot^0-Y1#`7S7rsH?-SI!jwE=hES2%`>h|Wj$($aw8K+cWj^IXSHa7KuQ;WXUr z=r$FZ1=a#2%D|7-X{#3sT3U`UCaI;xik9GY4qYs}gRNhl7C-0&HI`*;5EdFSL7X=& zwvCa35k$+C#T#fX*HTT5p8{ zd(~tl0~y(DJ9Ar;B6r@e@%~1Y;UT6kA?6bQ-g=r$k3pD5e-ORQpa1`T94fGk94vOf zeu20-2;o!)F607vyetTT@IMTzu2eMmxebB#QH{+8Za$+C94jWh>)sF#%KSL zkqu|61))kmAA$K|Pj)c9@J9X1Ssj0w9lp9M9lz@M1Xw@l=6j67)oI$_{X0ShYj}tr z@1CFk{n_>KV`R{Izc%i(!=O4K2=84kdbS4Hl;eV|7LUthE1q=r2madV7%MCRQ!jQKNI4kl`iCa|H z2szvrCvP@UqjKt^^nLws)-vV{ZWU95iE~YLLKSFXzMG;@@=v5pVlUX)w0?h*yqQZ4 zYiq1sprw{u1v;C#uo8UmCQpDdjs9sE#t~YX7N0OGb`ey|xuvC}-bG7H;CTufD9#xS zApV{-ov-J9Vg8@o?5T1G%R#~LWh-ntV5lPAS3#2_cB7>%03D6_z3sY4Xr- zzY+XXoJ_Mo3x@YeGe4Pdxd8veYl&E&DyJhe|?cW7d* zmR<(KPaSUBv-E+rmL&qqkYJ~u^N+we2m|f+^Wa&zwsa766UTPN8645XyeskDX#~kq zUTg)RfcAHhn9E8Z@=w}Dowl(CIlmb|Eoe^er-_<2JJmAqfqX?iRfUFX zU2sUOzD~*#gXxj3x2$0x;z~@F@m&R2%QngD$%9Z$btPtA|UTr4JoXIaD<`X(l z=;MkN&eS3frO$ORwI4UONr-#f+H|IL><+k2HrgE^?;e)=*S-#%d9_jzK0$@FIXy-P zJ6tI%{WL``J1gd5!8a!G$CpSmyi0&b8M)9jrgXociZw130q_1r7!Y z?XSVsA+zdqEl|0nIStf|unV z?|4Rx1$d^Ou3)YO1hoOXH8&s8tw6+}D=C-NV|qcg&S-D;=*x}Ff4rE!fT7%Yy%i8I zXFA=!tj$EO=Ts8X=j)@ga4&>4AV7P9n8CTGcOSI@Da=VyHcE|m5qXkHl}dAw)&>KQ zX8B%-8iKp800e|$`|naOdlMteH&rUIKOA|CZzCjo%>7s7CA15}d5?XS#0&;pYDk#BI>w-W34oJoo-sL_ zWlTx{M(;w^l~_q~6)Qxe!05%lLPSW<6_koXdugp|N-0o*cVP~*0JppB?ibIdBaE#K z!VJewg``bnL)7d&3p*GoO*U0v7vFcx#spc|IY8(yEB~p=d>C1bv0=|A=G(`^22cc= zJQmRYCtVaj%J6PH;4Wy-VP$>e;40O0n!ihnPdXXS(2bo98?{VkZ16v7!Qp^(RWl=AyO%#y^<*hr&dG?GQPTFw14u*p^W@?wm?QF|%hLsx4gYUL zdB%71Y**lA|12IU=W~ZX-u^)>h1($5`5;Dqe)FEo!GpelA^ghA=xta~Ml}ab>RzaG!|%k2swADiL`9L2N9p zl+Kt2IJ31sd;j@C`9qJPeL1A~5JbgL(N9SL3G9-Q+~9CY>H~-9p>{4br|uo7db2<$ zFNi8$g7FLYa|73lnuF~<-N?r`!0ct@!xm#j;Ck-wo$gtb0$_As_FiR<@7xqN8SvjGBAs$r!9^woJ}8HBtnsMh8f54NI{<@-`B$S;I5Lc zvH`6;AcOYmAlD$&fY5Uix&tA{i*SCTp#t>1QjHkUI*=D@8}e`)uKzdhpaG7wXCQB` zMiL=cX4cMu-umX6VXwjN^4;#o(JC^BNPhwNa048C@X~^om!)TDRRkLJiVzqXGxU@9-b_y+FC)>)AeA&G!7LKnD$-A1P<~$eCET~A z6y%4K(BW719;J%Yt8t>_f5Qco2AMiGAUxvLlF)8{d4qvukV~4WX#-?89Y~x6Ims0< zmthMvQ}l<(Q~7iq(6`5Jq5>pPcAImO0bCJ(i;oY$hDGwSlTJe|*PgEVp?Ew%-`Hgx zKa{rx|C8iph;6<0Y<(|57*0m+PRYCRA7TPEJ>B;>D)^kkW*O?i7xyc;h^nLyVP)sTFG`BdkRn(q=b{0;fi0<2?iJ^>MiKRc$7= zR1*;sy0rF3htCxQ>q?JItxc%jl7g$_U$0uU64ox15O-JPAx{-<`5Yq6+Z4MZt>0$7 zJN?CV6RZv}`SZLVuGw$LzFaAZcc~f~F>G3k+mm?p{ABLqSK>2RMfyh;9g6qKI-t^0 z2@c8`G02d|PN^J>@wgjj8?-h}a|A3CHSbP8W zAPr~(6_=5z1rNE;X-xwV(Y2!F(xd*pTx`ijez~=4kHE7WoX`q{?Y$d&^gCYghf%c3 z-KNWTA0Hx9^sM-e>$^Zt3+qctGpmx01kHbR;Nh{+gifIyRX3!Cr=9nE=Bq!OPg@6A zZ)l9gU8LHL)qv*{)E!|E&av86-(uP&0{!;1=h5JK>fS0lvb)k5qw;_v={?~^cZMPf z(SviP?Y$2!0YNXz^KU>~=eQb?-cuR}x$WI%W%C+bYVQv3mU_VTlGYu7F8T?}Kv zk@>mp(%MO~XA<@hp`OdGv>0YD9>cB_*^HT*e3n%@&i<`e6OOk%geS^g2a$@8((*3m zuYEPT(_y*26k0|DdRv+-wo=8($7C0FA_Vt2vTp#~{$xT<5{Ig1-70NPr3cSGXy5Ax zn`I`R!NT(LFM=7WpxW~ZotFQ~UfcG~{!ngWxj&R#Ahe>tccV(RG` zPT?Rzcuo%8&I6d5r5n7lfNe9Ec?ZFz2QJb4)(X6CY@?ijm0v}2#K}6T+yAsLExo=6 zww_NkiXJj#_ghPNNE{|^|GJvP2ib{+kXJF#kPnCe{@RANzF-D0WkS{8#0mF|_PmLB02FQt=C_ZCjkdJz2vZ(jII`Uvt#KIq}`E|frfBIz= z{FAw7x%Rhg@v@b~t4$HYm)O%0B_&#VrcP`B3^{)2h+{-t;7e-dKR%-=} z*|aUpzm|luR54^{1s->x126CyHZIpldYGRhoG50~w_U9+U)*Joay-bTQ7Nzx zm{m-ODK9y^ToO0SY2g1FSq1vaUwgj*h=HSGgtM^O$iw3=E57&!)_>cm(r~6F zQG%N1g&R!#(+kgb(Ta^Rr6q@nDVaxhgH z>!|PO({_eXL_^600x!i#ow5)~vo zfeTrlUG0U8Cp`Q+1o_`0U#8#LGl;n%&#p7{GQ~C*qjF2AJ$)@ktfu|Fua5mszIUs7o!nQ?OW)&k;q(@(kggjMDH=iwCZ1n#kxf@f5{{U=!e5{!M z3JioEFLb|ED>-{$>Ax^!L2CMXdp<|&&gxOSSLKm;#3EV*y?8e_Sk9;^_+;bPuCF*+ zv(q<|UotPXB9q7b;HkFk5G1ag^UB{6I&!PO!~k3zy-($0w#X8ii=e=Q7o4cFtC1;` zTNlB=hhQ4b9)e^~_cbr`A&QAHS*gP?>k!pTlXf|by4mpJ2saq#waEA=i7awupc`9H)tF3#z{hM=yulm7mr}_ z+aXsT)XMkU3|Y4OC;eXjbSPhc`upoy4TA8T^D=)Q1jD6$`+b$6gVSA_X|Bb^OFuJG z#SHX*bjQL*3vri!>KvM14}5iQ_8h@O$zA3@Jqz>SA={yfP9GOvZfn1HZ>=(9&WeJU zT&TLw8%u~LjQNyvG@nt+Nv4uePH34E0>c~4C#M(WhkK4o7h#SE>y;(9Z(8_Y@%e}8 za(@jM!nmRh+wu>IE?yJHCAaO(d~@yCvg3Z4So`aaMDOJ;BM0*zx05o)pu>b=FFj1x&BK(NBdF3nX|YmCOz? zzkWmGW=$6M*~B&sjoBnPB+NP=2CZ5PBSA3*pgFAmbTQ&LZ%?Vh?HF;f)kMr=d4H6# z=1UzPM#Z%Lc@VPtvC)F@{mMEEAlu_Ap z`hys7V>Q6M=0V$2_pW%W#34wkpTKii=46bv+5IBb{YHV#^;I6oPd(3q)Rwyv?{T7h zI7pI@#Xg@Qv*?X|5$hGC1M1j1#_B^abeqH895s@=tWA!XBdMA0x)FjlC9WtvmD8@v?GY z68@_~I;6QGWWxZ8-ZCFEfth08^g}@7R1qfRj{)mdtD>bI?R%Nq(k^3xla(zJ`l~Wg zu)CcMSc(Xo5m*tXjcr?caBFbxM7KO~ZAQ1~8w_F9Y^)2@Fb(69FijvFI*D90MT_gi zu4M7i-gNISe0DtFqmzxB6_1vqf)tY{xuV3S5(noJ_ub_aTf3{NG2NDWNzR}))v%l; zM5hlz3O1t!_+aNw&jDu=IY$W6Fh0)m-t3Rv8><>0?{VpClnrRKr<jzrRI!I_#ODm(nvcHC zK82+X34HCPfMP~~NqVXnP%9440qEqbXKuU ziM?B^ti(x##4G(qPflKm|Mi_LDh%^8TASEmas)Be@}J`vz%4nGRpeXs`cq<>$!-%m10YJ8mRvP@oU)8u?1+2gtf&%b?SBba!LYwi^l&;JJEJ*dA|UC> zwsZp1e2CQVG-v&}-cPP{ND0mgwe`1pp@%s{>Ml7!X zL{o`t4p-5qY$la7_$h5%-hmt}To;-hU<#qP&kDd#>s<27`GoHxM$jh_7NcZUx&oLc zcl|>|sI{@3iy_|$` z?5@bF8Wu~|f~lQPsv}BEe*WfLdsmXfYz6yOp$3LJ)qrDle31;3yODy^T3G=~*M6G7 z#-Hmu$T7#3e6-9@SZuu2LgFaVzFq@94P$#wH#;C~V3j-!^RjahlBbLD z>s$m&fhkubIx%5NQ>37u^Q73tpjBbxlWDktjpAQwlUa;`?)OqK?(u7nZMZ2@)r-ZU zSlHdR&zTq?4?$p-+unlv?7F7Gzr`DFR)Fie=ye(_c&AS*vHjEr-}8BYgqD5@>{5O6 z0l3sFk!G8-ruCR5k=OMmlU$)`L zp7v=E@OEENL%)q_3x2fkoVwJXqJ!QmC;C&5++j#|)&h68Lyd1|ApxFM;bdKWKm2v6 zGU1IY8{FOsRIAsKkJ*P)NS09~1il9d9eRTVaITu}s06|ord9qb$)#0n#FA;UB0&F& z(~kS7AvlVI#WQg4x(*qUJ@0|izV$LlO4Ig*H9dSkNtD>41c*LaYZi($eOA*ov?rKyuj`)fpo6o(jjh8P!Ni z_#JtuJ3r9tqxPYEHf1$CkQ*OKy0#;;1%IhlDh%~RhT#|Zcf3O)jCTjZbpOdW ze;P_tyZ;T=XPQ;GPAh*p@G0-6xW*06dORt>IgFdowwOaFlRIzP9Q>a=EeUP@iyjaQ(w#*t>Y` zMCZ3J%c$z$8*C#OBud_^_clJ#EFOS0DavVN1w_PP9&aK`qELDo1_)ubBH|^U6D2`I)qY`WNZ99xo@t2wlvx>qhvYV#p2bK)| z;7VF-lQv~@qCK@h4~7q7ikC@-IljASF3>Eru~*CuRO{tjQ1- ztI6GmzE6%%G`FsmSD!50^LFyTC^-)14^J4B^K{*SX7d1QU!XoQ450es3b#`luzTYX zh@BlU&7oh&4j311YZH>F8^$FP4&i@w(G=54oOG1ct1@u#_7i#zl1s6d1mAgb>%YO? zSw+7P<$!yj@12JNerQAdxS*lAf=KpO+}j)IEWuf6a24PoLR^$KEAp0+p1Ca1gvBZh z^2i;n>eR;q#};1pcOFS$N2@6oy#RMb4fN(=L`8$UAK5jO^OiD<5VO}yQl{Y#qjhGP zuuMG#_-5fJ{A2Qt??ndruOiU} zt!{GF2qHYay}ebN(p;A)LqMKr&JLu)XU%F}ZE+M*X@B56zUQH$k}*ELg#RVW>;O6X zcNBa>l?^ijZZ};PGitg*7~w-drU9KGw&?;}@Z2s0!s$Z^U+|N6!9&XR3A>6~%oE^W z9&{~Lsc46JI>ms;Q`bK7(}1avn8B?t!Wqt7k5ULmO}-n$FbPMP9;>Zy*pGLPGVL3vvaipw5bv03_IWv8+2r4O~Va!f<(-tXhAM$v@BIokD+Z*4+m*6V2(`{C)#)%m@UfYZ%K;DyH;t6b=9YLlUWCmhj) z=bM z^B0c-tu>3Rdf|<;Y1s5>IBW7B4gzus#-XfxUH{PBGO*9S>VU%z!;61o2&33A1@^)G zBWdF`d_^@1`h3|BG5gmt^`X1@7REzKCdcqZp&=6W%9Cq5T=j%8-Ez1MrHKMhmmo1= zE&`a}Qq=A|r4KPIF=p%6(dq8puSpmUwnQ(}^6I*5WiG#~%_SGgn3Wh^9?U9UiWDyp zvVi^S=R(V)>%JCUgZ->rRoj7ZH#8)OT^D29CUDaGO@c5>0U^aGIDZx$tVt3kN>}DN z%1{C7qpp|YVPW=~_N4$91)yys6MD`WHB}FG9GM3lF&w8aRzls#?y=z-D4KFrQ-|o& ztJqbtv-3v4H5j?^$IVG9Wbi7o#^bl7+$)?jgg$aUs%cwP+)~Y{0-Y3Z-8kMLtbBuE z?(4i}o*g)0t~FbZK`z9Q`o4fFh^ASIo?JE z3~dgNumbtaU+CInZaaO?OU)oAcuNFV*rpVCXt%zr!s%6=6>!#oN-0NIG#ZaEN!LSG z|94+n_rFc^){>{idNAN!!u^bTyNZEKu) z8a|6yDNMrG7(M;I6}MilJWEWdQ7*-fETysA@?FyKym`RWwVp!npnE}W50ucX!p<(t{?J%+8RbCyb6Yx2y5z?taGAY{#5f;jd3pt^TLIK^CJk~x&~X3D|h=(jZ@>2xhxyl zxq0%l!TY3vA11|t*x7(^QKG!*0G|O72Jh6*TX4@?BJLSt#LaD$MQouO5ySTR=I`}x zB1DMHWrOcm(ywBx1GsL(T)W&R7&4q0sFKS^7o>AI+l+8@LN%AJfqyK#3A~!k6??akq zB>&{D^(+49t+-bBS@Yt*dnciIsz>bidl{tX->8ST2^SoSi-t=`W;BBM^(qD|ldsEL zoy#Pbm7XMs!J}p58;P{u`P8jihls>G3O}{*{XvE9Uv=>P!X~41sm?!&J$^nz4~rR7 zZ6emNWGk-v>@N3izF^WC`)>YJgV8Dy=ng9bFP(X7FxTutBp%Vc;s!8w<<=`o=Vldo z(iN^o6UGR~Wi$=m`w`qsvHS2iiyIyJoaoNn8mg$>;1nxgU54O(UoxyeZ9EEarb_1h zzjhcK9|XrsP&&4}d5t=&k-(0|tm0ryJvhB1gOH*$MXwG>p&^JI_xJg~BV6k;Ll7KssTb_5 zesjCVhr)xsn)Buh>>W2plg2#EKOYVt-1V$fTl5zfx-Qmk9dUvtyDKxs;EaNj-_w|t zwG~i+^M@EgI9|gY+d74163yYhkJlL-F1`r<$g(v(*@OS`LL0{G`R|JuGxfuWTCRt1 zZs2IE=UGy9!3KDk9WydTOKA!TJQ=$StNZ?+`vCktJ%gCD4SBR$18bexg7zi5Pe|8j zBOxcF{ne^U5?w{Xph>;=yN5>qqK*(8ZDA&kEfQ)%f0vP+Ef^Dg|FPanZ1N=8+xY!s zC=fI#`4q1mq-P%Sq}q8Iyt&V579jHu{+&3y%o2RYQ+bG}1_|hHo=2c)auQk)l-lJR zf-L;`-#dy0)nib=d*7RCIOg4itpLOMWEy(3P36r2p=D%a&|r*#Biof}rW(WD9^R9p z_Nn15_#@46tvZ%N3?IGKm6m@lW;5W9O&Mp}EYU z^RKSbIwn$+ZqkQEcQV(^ybi@0??Nq9pw+h{?Q`LIfUTz#6LuB0!W;u>7Sg<8vfU3} zcPg{(8u*)3QZPT=^$7fqlp?uAn6Q7ur!S7F(~(!Zi^0~3w&kNw7U0Sr<#v6-C%D+l zYqj9Cgs=FHQ4Q$CTE@KC&5oW3_RP|zCVKYQ{))KfYXE2dil1?J1+Vnk7P1Zh07F2$ zznFX6t>^!DjtRfWyo|)@T#pNu(~~Q59{tyCchDg8MC=i$=3Z@MhjF?sWQKmM5wtJI zXJP37?y!?FMBDx#w?e)kNq;gB+6iHOuj+*@5@GV!zIVXFaKz+~8() zuryO86+f_g@W)!ya4&g-Cms_c8aH4ovrvBuj?Nq(GVsnLA_c%5*|aztg1dQPQ9O;M zVD4k9M=#!3g0+1n=B7hN9~OOYs|Qm(D+FEL2Md`V@9%jZ-Ye|=8YI62DtKBdR8Q-o zw8r5;8Us}|R;j1k9k6U3Fp`m~?xWco5VqwRMp6?CWiumgEAA`Bcu1fZb`bc1(iHnn z#SUXcF)SZ0gE#TK--X~pSfT;nht0|KeK2C}B29h5C18$<`{l#%2Rb2oa@<`CVoS4% z7~oFaFzE$|cCxd*7cwcasG;{_t53 zlVELacR&EeYVq>sk~VE6-2;f+4w>1Bqvl}3CxFHaKCm* z<5>{;VNtz;7)E&97JOgB?fe<8D2u{<&YiKr=P1xr2Pc*Mp8Cgp*Dthbj*F>&?xhQA z@4Gjc0$QM7-IQt|LgE7shkf)XwZ5yTlQ8u3Zofg@lZ%hq+9Xqs%(*<#48XC9b85OM ztZwI~KP3;}BTJxr{Ia1oXy`fG=~~_W`ERWZKDVCdG=dp}g}6yghJ6#qptG?^-R+)l z7Ml12+Ch8Sh6rVz*I*IA1a}ANB{%@{^PTJ(fV;x7Fn=KFc_jWo{d=v3#|^;T&vrPM z-cP@b>i*p{1TufLXprG!IT4>?`XF_{)C1$Fe-HCH_#k=q{O%#(I+bMGyw3`2bgrqo(dy!+d zj+9^iBqKTGjN=E=GT$8S>SSl#+Htv+x*k&@Ma!v^1`lDa18{Gp+bZtroarxhxpD!B zPapwDtCDO@M((qjf0)WnbbkAl(P4Kr@;@aZhB7BSx)uLTl3+n*bs1FJFL-xgsD;omsK>O*#|H>pSj8 z>O$^c_uiO>i>JwSroO7%XH_e%ekDz@wHz%3sTCK}e$z(myAX3)>Ff4q6Z%spghNHo zB6qDwP5^N){|>D0-P6y)*L9F;Ez*VL_{{i0^&vTZ{DBGAPi&#^y|INFU)3Eo zO|!_QkP45Ae%1_CC2+3vBsEcURi@6IW-RS+`6whzJDgz$nl`q zBts>0SvcD%DwXOA=Tf#Z&l&iWuswPD_;aRse9w(r5~1{@rj?|%?*AyZdk@!i@Wl73 z-kEKSpPaAIKG}86E~<&{gEbvXl^3s65?tQ%DklNH2BCq`V5@6KHJ}(-!Uk4n6s7cn zwH9ghsv=fAyJ+^n`x1I`D01bUrtuF#LfuiA#}u|?!iOyOe%L`w0^dR@I752G$3bc` z_syh>`EOXB2*z4CXaN|Pr5_`rvtj;mR1y3YF(v&?*vZg8D~{fEUTPYyGty^*ah`_r zA2L&|@!MSWn8nXL1f&qh3)xj3fb#Fb{$uUrjCgFtFR!_Z82Ch4Woz-HFak03#nFrA zJ95hR%DjxV@zx8ceQY^^^}WEpQlozdZ(kGw;n$wgP#)M^>s+RJSJ4xsCtTy`Vd(N# z<_o-9sgtC=%L`;$wZtLed2f+V;FksT zR#Ws0Qkrtf@+$u?PKu!UG?SC}BRKXngEU$&zaw;qG*Sw^&s*E0bZz%`B+FLQ!&oc& z#M|?vyshA$wB*7;&!nNgKZY|dv2X#FM}elx2pmXfR7M;o#Ot$bJ8OL0__`xIBwTj# z8XyPX(AvL7c8_`TW(vz78bEt+=K{c@%k^WFI_9!A_yNGh_Z#xIuZQ-==U+W=MU>a< z2kV!`7>fL02&c7CIWpuT#qD&rXDTrsx#*qR5nd}bL%_PMculx8Z2iyEFWDxxEqhhS zmM~IaA+xt7A8EkHoF*j>fb2;m#fP|kW+fMRpm1)dbHIwl{FPmu2UijXdEKZ+McZW!3^}7xFo+kNXnCBZPBjfAP4h0F4Hv#3FfX@kgo)J zWw&^U&KaE|@j3WGVKScyMLgYU9-e^Fx(L_Pg8%}SYB{$m0Zwf*sQalXDXdXj^#gsx zVIp7I;c(OR6EjAA>3c?RLe^n_Ea9`y3BP>jPyBxC*YvOf7Cm+UN5yxCQ~Ce@ zS7b}F$tWu%35OCVBQuVXWTjHcICeI7SrL+zch-?TLe@dXA=zYRoMUex;~3{0=iK-A zeE#}fzkkknU9bCfKj-7|cwW~zciNHMv?4T>Bmda*x8UM?w_L`X(j=z=rSRiL)B=Wv z{7m^WBe>UH^Vw7)<>ycHjrYBUvgI!sbGfgT-vgfDR*n!YPUH5{f4+SZfY!_}rc7WS zj(^>fX;9o`9-t!gS(KcM?g4vR8)< zKR?)V;IqS+UPiSBv54B?6MB!c7T8+uz0d_tH{nLiAvd&dQq(U| z;J_LN8AraGpyA}UUhHkCJrNF!{}(fI#M1>QO9r#}Yvr=>x+^AwV}%5?DeAm1epU%-*kF+E8l%|AXC&e2?S(d9wNSfyf;@y!%SRW*A7`Gw^5>dPKr^z%Q!B zM3$b&DYjE{^GO?qA+fHYXWNE*b&(Y1&xD^* zUmJgZ9$U0}yR0_QS#GH4I)w$!u#nSEyV#$39%~Cf4beCpV>YZBlG}{kkPIKKdVYm;y37Bf;H)hJ6eWKt zu!T7VdMf@TBW;=cC;Z0!@w2>)V0ieUY;~R-F&R@?xirt4646OT5--9oAqM=dMq)ha z`3PQ^03`FSOBl-&cFAd;%*R&fX207dh50fz);gK3dITt9)+^ARcP@{wA z+(AO}*tRwBal?JYUJ4u>HuwSglvLXQPwKS5JQVjernEy;#h;ssV2St<&mRCQ_~P(H zvvVN{9FtXlu16_Ilmo4%qrPo#gr^O7c>UI<6wprajN7+5h^)kfVw3vhMx-c3j?eXU zg~=Z`B}g*rfguY+z&c0VlyNSmca^&5wC6hEKx<^yGLdHe#PDQI-(@BNlY7 z%&I=JlqOvwXIwKV>ptCiFS}Ov#19!gTEz_c0}Vgn#gJ<^H8{ORRWo@g|8yrR>&Dm~ zuJc3V%+jPJ`iDoT7S+>>Z{Ea+G_q1q1NY+W5fypZN~A-^H5U%!llUO?-K`9=Lq!VWp8{{ z3%~+;h9rN)AKy4ak<0^ZEJhHLH-Zl$;uXPmc2O#y4+d-RZB%~7v(p@*`eX0p&Eyih zH{gO1U~svarQ@mG)`q*t>v}NLwD0;ZiH(Gqv}D<9hyTy%<<5AnZlhexB+JRi&?Q}3 zXpL8^EMR4A{w?s}4(XucGT3hoD8*PFh%E}kxwO74%r}2$ZA#lqaJksBJAzWie+1Xg z;xvFKNtrvdk{kNe<>#Za6O6MW+p0SGTS@4S;bCLZ20`BOp46UW~by=SS2h+b+ zRNi7db+~43MFWg!RMAqm&|>#|27ISI%ssE_|QZB1=MCN$Jy-J46{|HcfhL4P_u| z#;h0T$Ditjkq~)_anTDHh5dgs2a>dqtr<%}POcmcKA}=7EQzptIM=%@Lh{5qh>cx^ zCT#I60`H#EdGOg33Bsc1Z~nD4BPmmq_+DW?T>*7_RhOe-fA*;DIEZ&XD~^DW`0&?g5~MWvJjN zY+;=-FkAZ%-uw2|qph}cd0McNlRF16kik^j1zF}wLf#Xtl zqeIJ7-1O_CYYQmxIj~w*lges7&|Cw<6O3!ec(nmUtMZIHrJIrR zr<}Q&P9OYrO}$wQusC>rUzBR$E=z_9)c7Wm=PTzp&@~{2r3LpbUWV~=LQ~y<$?0w! zUfX`|;kmk@rd$xumLLlDzxTh~4iEa@OH?epeth_td$Tfs%)?nVI zIC4yJCm?Uf@(9(XTzQ~!i4xKg{Z(uM1*ATkq9NlqWQ;<{e)-Ume^>^f zfVs`^Uf+BH!=Y(-mxPy;;iZMLN&vW9o}xAYKPst*jD2WU5aa<~Do&L}dX!PPGp6kULN7!P z&+ZL9dh&Sz&TF6O_;D|RUj7~@#p1b)0IM0O%8VOG-F91qI16e&%u(caFwH&29BmNs zFVSO^$U*$ev%=F3((+SC?PNMf@#gNR{lyGs~_ zbhh0dVLS8pY`X?Po$qQa{c1Zcq-|;)qHZSz4*j4D_J@v*dQmDo`JllO2G%NleX*IG zrUVrk--Fs?a^H-T9ddUG{0pco=00L2wDJ$`zoHv(vkSfpza^Dq+Z?b-j>)0;Ra5Be z>${R;z`xQX?g0y^6#9cL3`Yt$BMFRaLS;ify*ej{81+{ow9qOLN}AuORD`-adGslc zOnHL)^G@xr>B*>edVh9JaD#&MXHNnrXLO!3tR!Q^Nq8S&Xq9QF4IbKWtfdWryL*{O zDB=6QktBqxT<&`^VfPx;uHFXMxr9QR0KaNj-3a{Ma4RpwgDR)(_D8 zEy7`Mo)5^W(wGMO?_XWDY@wkL&+80-fbcql7~GPj_$CnfX?4V&D<67;A(w^$bQ~?+ zliDJ{VHI@$u-=lQOJvhgavFGVz619c^@CoH*k^IQOFy2M@0}h1z~VLG^+kAtChy*Y z@KA+OE!lVr{$i5QEfzasY`>S#{d+Bei&W+xDu)Gbe}@)cB^v8diWp1C<*EayoSQpv z-lY3Z=FLR-usthr#SfB?H2%Qn-g|3LDA2bmlNSX43WG4{`YGy^>Fb@ZAy_(91qkz z3sozNZVd6r=a~`XbO;{}kw>WVsFfl|7k!pO-*oGuCiu(c zN_h|Q+}_Y-I?blHz@3BRkKp7s7X>*AC6UXE3ZPDcBhWk#TEZYn2L-BB_ZmWdkAP+# zg;@XLf`}c1?h0-ikc+|9W$r^v(vMqiXQ6<-cG^m^EHj}9Ut~T@%tKh*BIAKOP;ImD zNSg?8$YSozdV;#B{i8Er%^DP!z*-HC-TzjOrGZ|g#{uKfmcPruwBPdv#=4M<{5#tZ zkZr}$c;Orw^>6{zOOB6$j2pR^GM3BXCxi0_^iTkp$Kn+M)zsW>`~)|xIE%=vT*N43 z`N=K9=Ovrc5FjmICntEj|BU>V&q?i2$sdJ8MX)i$^76j0ZMTBnfvzK-xj;g|N_XK(d%L)Qv-4k%f5RaMv-1NYhnO4tEc~3t zSYR(WE)DXcBl|h~EOOCTy(oqPmbsQ+ zfQM}5XcD~2`FlL~HA(=*P?ArN5t0m-zjSCR8Onb7&FW0zwJv!5^&h^(2gPs@0se*K zUM$w0LEu0Hsv|I2X2p8kn3~pzQXodFrZSKwPO(Y$MqdjvMuaw-?dNhlJ;g_ zUZ4K@=*#F4<7Q$>5~fph=*=nH=S3bdN~Na|IMk5gd$53NQX@Dr^?#k4ut5c z@*GO#O*@G3t2o0APihUF?8YQLf8DDZ26V|^3WFBH-t#dyQEt0i?+M!;(aQZ!!hoLD z51-~kTt5lA@eQ=Zyx>5;PXE;eXQ&-IND?LCr-K92$T7#kDyifnxhBJMsT1;ucugBL z-8Res_IJbN6*VbE{(SHw)E_NZb+RMX&LzlJvaf*#dRgrQmC<9GYzc5K|C4lEMl$NW zuQL;rHH4LyU?kOf^!%EQdOPcoWkR}|IS0Tluz9n;6yjl|MH5v z8eTl1hJ$ERNq1;3RAD{v6WT;4cC%9^#i*iq=npUYe`$_?_i4qw&OtJO6{Xog#O7N5;krWU$!VgWT(5AJ1cUYIE(6F;AtJ`~ zw=j?P_#?o{)lNIUgw8Oic?1EKrN89_h&EK_3U|;y-Q|=*@a+kI8~M`#IjME;z*Te~ z+*AbRODwY~l})_K(U6x5on|g{XS)~m8fLZq@qHW6oik719{~i{EEq|6Yr#U&(0t!- z;XkCxCs)t#P%Ndqe5(>SFlM$aHqLp_QJ_jAUbkOf;w)4&_g4Gr0*Wstn;e7cjHdd< z>+`OX9cn))@4d}HMp2siH6W#bL!_wR_89pY9h-;X#FQzOSvbBNyhw*KseV2o+K7IF zTYn2?u_}fXc~?+S9$wc^jNdN~sHXVcFi15!Tz}I`nxle7*5j+=fb|1?G5Wu77gGJz zEI2mdox)%E{gXfJTEFVs0hm*SFeGRX;^&kWAls@jw*^$jWC~Q>1!t@-3p&s(`(+}1 zFe$)lC?ZLwU!P#92Ju_EQ=VD5@LjW%Od+Vfbq7BPZOW|%ZUTbh&b)>rf;n(h`=l*( z(0=^s2*upJ+S`nujD!?E+na>qb#?A&U>idGW^p zGt_ROL6UQv1l(Ff4!o%|m8OSw-7nV_~)OcN7K9_}gY39>L zj@jyPT?u5RRC@c4lGlNs)JidCI)jk~D9{lGw1>8!CeTuS3yzUp@oWRY=9UELsR?mZikKzKaxLAB5Ro>4g4XIDJP|v)3b#LewT^vBo zk`nkoUW1!vU)})X)jOgYj!+VI>!bg>^di}Dr4kqKPj`E7-sx_>loMr=fepGX(Jrzt zygf3bg9QQ#?=dby+MjBE?hrZL$Ti^!&c56p&aQr2A>TV{)KmuAKkxJ*U`ywmM{cN&q>={Ca;&P6Z7sC@lvJvbeF?J zO1Dq0SmhR6nE zJdUey0x_DWpm~wE@$mcCJw&iUyaU|afx##eGEa}@aqNUl3F|JR4mh5IYnD*f@BF{; z!h`uMe&Da38Z&igpR5O+kdvlzC0}sPLj^se95YzO$M^xN*|*3jg~*2cz)z`=f^mSK zxP~Ar&K8pWo?C#l_EdlzsrC;h{KuEb0`>C+=|}mp_JC5 z@RRsj(D1>s`afIgPLe~UH1>|=Vf{qhO($i30coIiweOuahks_d0;U#LBt6zpTcDN# zAu0$(Hi=)Dk_BYwc}$5Mp?hp!glidA?e~`S$0oM!@l{8a_5PaxzKplBT20EKq-tM2 z;k24Kl8JnMa?EPPf&e%ATaQq3@SJ*sy&xlKRqGt+ISE+z6&=l>qs1oA%~@#~3qG;v z_#-ix!=Nbp3-~TE(k%_N5X~#HP*R)2b55xipmqhpQTwa8?uWh8XJoC9##E6}gsmCv z#VTw>?^j$A-#b*qm;cfcx0AbvzCn^p9wo3$xX+hr7HO^^2nLBJhnQ$Uz65ezxBtd` zXm4)-^%Gq3^e;@tCmKLpe|=6c|9Ey!%)8MEl4%-O68EKOvL<9AcdL)W?p-)S)y&2i zFoC8|cuE+#6x6R4K!LaX$mTtZVq1A@kx5K1bD-5OIHI5N{XXKg^Pi7(7L@LuO(>Z- z)XNtuJ82j>z;t$hXd%0cKEa4de}8b^`ud4Vi+dfPCmA);Gam-sg5;IJE7lhc`=IrF zBynLWrJb-Yhk#skuTxtUu_~ptis0y6n1p8K=$K&!@HaXjFVMDCA^9{Q9b+S>sq z&X~&tM~M#BUi->!4;u3Ndr&}&rX316Y4vgg@s~A*&7f(c+VeC9U?7zNnR;3G*a9lR|ThA@2iy&kOc9f!Svi$ApJ{zFLq{caBV(-Yzrr{u^^nl zQ=dR?P{9iFU$2}z4-v<2gvu?!8#*xH7S1w7Ozi@x@`Il>#7;R?Fo3d|sDu8q)n+na z+cQtvJ=&Zj=G_6*B8BycY}smrEVb5{jrsFXW*YfuG>=Xd+V#u`+B0gPbjX?b3262* zhMh6u4Z}#4+N!$+UvVB@st$LDpS%9l>P+3>$mmMg=ZO$3OzQmZavJdiQvL(Gc=m(wF4KcUjBFHc z=cVdqZFhe|SC(%Jt_+tJ zpWU?ha-Y2hdYo+FEt%*^aS69<{#x}|&0@;_rvi+Lw3FcI>oIm(_1Oj(5_T#dv z`K=V3ffm1yf2`@^<|;ca>(0gRM!*PY0*+Sn7XE?Jag8}I^O-i|Y|Y&uM8t*b(YVV` za^dJm;d-cG|5wq@7oFV})O*#1h&Sh*5_a(tXBlo`R^xiuBV8;=@!90G+@b~?~`(f8y@^QoeeXChO z!Gv7+J|ImesRbO>QSKc-Dj&lrBS~vV(s;pn#iAL-^4)+aZ=TE!yc|$QWjuE-!5wGGn z9JCrTRrt5=$1t)X;Li$;;t$pn#eqcMU%JCV%E{9N@$cRelIh^g8hU|MNl5t8IYJ`7CJVK>j6W75p`Yjx-k$=o_v0TDB6roRLeU+wypA8~^gk~8*K zM=1+Jxs57S3+QTdU~3@kX~yxPpUt9@ZVqhCn*P=CSFYujwX{n9-$~hvK)T}Bl&s8! zix2-*8I~Y>ERh0Ngt$&z{@FO!$5%c~7zqC|Rp_B}3EjyFO9AQiZo?2Q?|O^ZH9_Qd_AleyLJCjmJAn8J19^3%e|cz=YXcaWz-%T->ml3-Z&d zf1ZK0kBQE(0cf*+y|;AEK-F(-QG0k9d_~rh4+E|)#9SpvMm@0IrALPzm)#yAn&>^i z4;f~BU&eV!)89A9G{jB7f0-!s(UEH+wyBviNJ06St64vF=&fu#0L8CTHsQD{dbAJUH0) zd?0Yn0%t28j;t~`eS2lc(;g$q=T=&}6NqV{>`c+`-e#!QRUvJVRc zxmua)XEV;c=pNligmHA62^Yub2}NJsK_A@xE8=OxxNl5w7K(*3R%K&w;p4c9SuGT4kPc^7)c&*gyCk=pOouc;t)f9 z|Bj)~Sne!eS+}MnUrOh98E?Y|6XfK-uKDZ5$kC6t?nPh=AM(@uDpRjB8d;(OIoLd( zvvkCXJM=d<%>L8$Zi?Fx^?l|MGE)>6HuAAyCn~u{=Lc?Pb^Ut1q3RN=_)cO%<~(Jr zf~MLC5_-wlgD(Cr25Asg9M7jQnQ%4hScOYDGhfhR-5IBYMb=k{9-!T1UNGQN*q1I; zV8nMkm5i^Kpc@%*FY+?=mo)LThq$uF<0R)(?i<0+GO{K9E>*(TSR;|` zXZCpnxZIh@HAW%Y4+K8nWj9PmdKRz-ybnj-x!_pyU6xlQ`v{?WDbKwe2J2bPfPHg) z^~w?#+#{}Ixq58xyUvmcWouMP?_7(>Y^}}XZ}dS2Ts5IwCQbA@+Vz60%uSVcspP17M8tEK2Ojd zYf6(jn2vn*N%HdyYa5z(Vqzc?2Zva5U{7o;O5#JDbZ@#&I88nyOW||Si-L=2vr6s} zCVT7%^VCH*NRn4+f0_opdARCk__lV-V5ByfZi;((FFhLjxRCYE&TEScPLnJbk>k8$ zQ3~)o*Z9Vw90c;DsEhMM9jmJ--z+O+$7IwYkiAF=aUC--H9mNYB)xj=+emCmqwHXsBMqXeY%!cL6`ldGVtwn=g;! zs#y1iL^~qR6ipYh=mx~+vkn|AqCAl?ZfBtfnYyo{uv0?!kYr9_nNlzk*B4~l(tVIC zH)l>*cX}e`CZlVZEafyw3@cQ-@$cFXK2u!8@WB~phlU+&JB)Jc8ptO;%D#a5V%s9u zpXtj=PJZ_%}ZjfA)D4 z&P^Stx8vDE9Q`Y(Sw`{Q+MHpkZ?vMzVk(6RnW=A(x6UUnU)dq)$=_hv9!^XVx+`O|7&{J*lhbf zCAzx7Sx1d?B~mrN(Cja}vpy@y4n2@g+l+=QOfQlv<57kU-yJ>YqI zLzr&|zOTt0JUZEbU^NZ?lg=6bj0)pac_6+g6H&XB=S^Ugsxzw3%hRRJ243ph^d%kp z3)IizitH4?fQQ;vdNx01(u$jweF>81-DJue4wHzJ*wRh6CK^=gln$o$-61%1MN|!_VCtkEw(ZN~ zgU0gfZ2!It6LkxzmRhyCx|ZqpZG)Y0fEA2pb%4TnF;E(2dm zJ|Iii?A>=NuU88`%dnQ3TZj35o)>S}X$^@#n>&DUKBU&Pnzptb=wvuODl~aB%qy9N;>Bp33>cvUALbet&mGi z*4E2{Jr#GqFAJ!$8_Y#b5kX`X|k-#_3 zWGzTC^waYUx(7{evI2eNvTI)$%;Vl+3lmcBtPVr$Xr~zk?osCq_Yfx<^A}qBO*b?o zdQ!4(Le_2to3{@EPRMaG>Q4Ce!*^x^l}*9eG?>RP_AiCL5Al_qvj&N`nZ&M`;vR3_ z`%OWKI6DaAd^FhW>yKQ^(g6#}gBe-yww!LxUr2=tLSp%Ld2)qg6)~@FiX{>!wMJ6T z)X!cJR=Ai{TTt>7_L5wa`0i72ylzqF@=YJ%S&w*LNIOE}vX!zJGt_M}uZIv=p!W$@ z-Wy^S$T~U#q+!cAW(vx@pLjg#IwldO3}#|FpOVd+^$a$V$;G09sy|_}n!j`tvVlQ| z22(~tk=1bR)~Q%k!2J68hK6{aPl~SSrC>ehR*6sx8o6d4B3yrF4Bi zA@@n>+La(ah(%Rb@p6h#alA{K*ks$?BH2L>aPyB0d&=v;v6z4Mty^~=% zf4X>xESJdUzbB!92fu26BaV}fl0CZMjt@fbb-3^IjWeEmI)EWrIr&R~2{E->=PZeg zz2q|ILNMV~#_vyB8{;R!Q$@-Tw^>8)xpaf3de)aseC6}2g?}h0VCK=kpU6&oWH@oscutoD&j@kOkd2Eb&pdi^VF8^KOSHU5 zPEJJEYVXwWI%kBj1l@!J1LZLWZE)H5@?mtKmASY-RAg3bY!j$W7zz?YKK=8hTd%3rrX*^i-;ljEd93O9(qSz=^< z;Va~@YW>~^yFwQ=x&ZlnX`fUbMFaf5CW9PO(LS4vYH@ji8pxVB&+)MZh~-m|(H@f%}pkm*G&)m;<6UwBAo3* zE+-WZg|5yDEyCfthf}kKtT#zN$ubkhZYbOg&mYz>T;EaZkkzPOkzTQXpUdsXSfIkewd8L`peT?F%=kE9@}tVK+X0cFHlfDDBVHpQxl~s*>Y@)xu4%F{ zrrAo@ zdCcZ_4zAE>%>e0r|H2hbXx&1lk#Lb3Y8U0A`9@r3I`BvG&a*s{2TrZ!J=Ld^9-g?w z1$2`)zk`^*4S$#8#nI2`xMhVNDp*K~p$^Yc#?Z5>+@-YzOYg=cW?yz9U($$vo%OKk z-Cc%ePII0LY(h8DjBf$26(|S7uga*C&9wvHVUZsP>|nPFpqjVTo4j;#?9xj@heXR? zA`DH`PG|!^kuT4QZg6*Ic>qsNivH)bD!+@1F4*6bY{ zRF;%-Vd`Bj5s<@B0j^Ie(czRZBmoyW1l9`T>s~CNJ^^d?R_pd>XCDa#%<7bXLLd4* zv&XJz#l8519$(BGrDiV~{}a;Jz>X?FjDOM9kM*xEqP^Z$`dy-=jh5A)gB-hSZI6M^ z1Jq-t6@%f>oF5j@IFZP!XrF8WS|I1^ZZe zlCepUyh;|XZ~UKH@XR>vM*jQ%zCabYOfMW#sC;HE8yXpX>3zjZ9<(CbzsbN)Vr=lY zZHyAcy5OC;J&YhCwa6eGGQW=EdnESZ))DZiU*-G6+dQXEE@=8@m-WG_-!h{Un4nSd z4KF9ESo?Dr5yK0O5A*?O?Z7CD%~#jJAtTVfH3Db*z+zL>;+wa|5pd^ykj}S_``%v% zctC_oV~CSs*{M}ARM{k50uIrIUzO$=Dg-RasFtXplhI3LK#FQbRbJ%W7N2UD;#**b z)tx;n{I2L=5dg$oRai#3&(;1q!ZsZXEeO-!*Mzum)<-~Jt>QSjp9&P}l%SB3FzleR zrh(sMW#Vm8LwRaaj`t9yXqiu^2Htb_%{8rxhA{Ll1*fsOPD zfu1E^-Szh1sM@Tr+k#U@+CE~9IQipvZJ93QX^c-fDS~R^0}@#V^<6oXkpDE(vGooz$T|a2AiKYM z9vq4)i}6=&7SQ=m=O9=H3fsCN(6t`I$DpC{UpDON@XyBwZII)s_KlxzvkghR<5di+ zp^jCNxQAG?H4stbJ15~@M(lg%t7mg?1gL+ieDudr>-nWedyTIJgWcTOcY9~GE8A4_ z1P^!b_hyJAgKX$`!IuVV(*e%+Q_(56k*7U@{RQ5{p4FYTI+Mp%7A+3o8~=vj&bxN6 zY;y#E<}C=nQp|o_oWnzRHV}L{{PCL;j`(wX#f*aJUt#i%1fl`?d?adiC}(FQ@hBzz zxyg&4BP3~L+h4JADE-=Yyy1F+l9voQS+3c(c@(%Er!r%43o12O!@;OsYfPUS^iR`i zwfTGVdbpJEC3*16z`fTzdqaN73y?#MPigJh{)dLBv9yPmiMk0=D9%gKn$0d@cOc5PdV-%Di$ZB z=#0Tx?{?_6JJHwd*&Yx&)CaUaBW^u^HV9mR{$E!LzsX=u-|4Qa>pKo=JYz3a?`@-x zL56x0ng;iC1$vHM@{c9NY}IJd_gMnIa{X~NmnoR}jqF4qliSe?N@hqNG53HY_ESof zQ@Dn^R>O_y&^p3oV?55Fx(^*tKOp+jZY2^o6T`Z&#eiG#%RTmZK^-X>4_;Xc(EHg? zKT_DqoA5~N+TVb0iOw05#jfz`)U9x*@YL#{j047s(w)NHNYA?RP>VtI&1;Q{1tPe$ zpL1(pX4ZNxy>|}!s=@eHqz9&Mmprig)VHVhN`#q6q1t}OrB$@%AIi(Eje~5q+;0b!(rp&(C zOh7Qu8`|-Q_=c@YK)f?pW$bN?y9Y*kMTBkY+W+0;J!yH;?st{RpR~@;<>vA$ z#>&c&V4icGr%`SJ!BOz^OIMAmEay6DIp^8sxrd)@Z+hLm?(7imb`T%onqij8b?q!J z*Ps2r7@tI(L^#r!)mzDA#+k0Y4_Ps<5q`nZ+ve+V{ta8E?G^jmGdqI8Sm({^0!t>QGNasIY>NTvQE?z~jDb5QCnE2;^avi^|))*Z-?F8s}}wxX;%(7&Z0bjMXW zEq*l~gsmp=k45P=S8=H z^29P#JZ$v+GRFS+_Ft29s!z4t6XdAp0zWE^x_JvfPA2S_0DsVU&7{A+*ATJo@T@S` z<%87^i+-^RblOjQz{xQc>C-cxsX+;vvgnt8F@V&={#Jtcfelf`6hSz)4_A>o->2^gGX@9 zDej|)hBA}hozjuWb3^+4MCb*c9t;3c?{Ny zO4jWUq({MVVLUN7TM>6Z@bIZkrd+Rkxws1&RWj?_eaX&mxa81DD_*_yy-);lU#{Zq zpW0^zkrzyZA0JY)*ks^1ble{Xa7)*xwhF>qYFw9HUB{lN1{2mp=JaMB-O)C89{E~| zof}N>ehjR_{Bo6?>Wng`@8IOhaJI@p(&MmsqZL0U=M1G5ToNGKCwM)Rc>n^Mezwvw*qyUEk)gfV;1{iHv}V)>?LJ8 zH(P6{^Wn^L3$MnBYU#|H@8F~O{9KQWpm4*jV~>ooR6!_C2fe?p_C2o{Nee~e`rcM> z|Dd`t&MCKBJk5PSSIVq(N76$@p+e~%X2ymdFRA4uKt z+xh8f0ruBEGgaJlU4Et#Cx&adGN}y1zIDZH@l<^+j^~5?Z8?{6UYQ7-8VTGfZIvAF zdiL?;njn;73ed>Sti9q-NqR&bSZ%6ien!f!u@BgW zS$f#d+Qt0@8wXue0UoiMFb$84j6m)WutD_G?_Bqg9Kw$I%P66FQZ=&hsn#=_2>i*^ zu$xR8itP^4#5sfXp^)PmE=rlR&Xe|wZN^We+uVF!J9gO3I}Yf(pU!4=&42{Abx_P; zts4v}mAX3JC0F7q@ZHE}QHdRWU(SrMer9IPcKHk{0%@J{Ks^)w`YahA6Hj`JtlD;( zGjS9EOrq`@=DGps-*XZ25mqkqiH$k=zGiHmeSl4Ac7{8oz2HPLWl-hTd zeMyc=<6TW}OI~7cX1*>9zjGE68|sKM!ELx^Y`ez7$8}R&m+c*TsoKY%A=lQ^rYfn3 zFW?>Uv9j}M?(q@u5zrl1|B&<*9dL{r%^#!ly+)~`cOC0xxwwdKI=TID_YQ8#KuXrD z3;qN0&SeQ$L|@A`gWnXVyX;zo+UW^9or%AWLU_@wl1WCGiRf0kqK#$p2!Ue4fO! zj9bnJF9j;4-SeBuuT_rc8+h1kF6XE_E}#!&V4P?t`K$E+dLKCs3rE$gd_q3>57sPX z7pd|49)Q08s@{&n&6@`HM_F%*+ablW^dDq<)D7=2&2+9~HEq0RpbZrn1GcXJ!QzT_ z6ve6ac#9U=7PY3Trpf#N!Dfo@A4bL>KgWLej0iWANBU2f(qZL3;A|@&t^ADv5j;K@ zB9Va~imiNpQ&6tX8T!t;?Qr4xD`^coLRlQz$mLOjZUO7#INtv~weQX8|DHlNJ*?j@ zkw0CSc2PBFx{$6Dn(u`B{X}1^pN$GuL8;T$4I0gL zzXv`*%P^i97Q4;yw)$(%m2el9D3A=oX~Akq#A^ zQjP}c4iScgh=2oWNogb|h}7sths1~t#_qe{`#kUS{N8s;!L82es(yc=SMaDF+4-IIundIw3+5>uQ$4h?R6Jj6F>wxdbY;9756x)wYV`M z&YT}RZ^`c4R?Q3r@a9 z-2K)>@8&n{EGiUCy_e((V#31m4|mh!0ZhbqZ0%ov%c^@=yej%eAH#LKGTQ53rrezm zTZcNcQ{ce?mCA0Fvko1*hiCY-ki_G?1FE*XV0ILYQMGNweOhUg7u4!Q*OqrNIhKN% zk9n9{rXBD|RH1Xp6Rzva`yB*5QGJ4lKyk!!H9Rf=d+R$jJOX-J?S9TM6M|mlHi&KPWbwbTK;pQk9-H<}a8_d_V&7S}}?{_{K!2mUQZ?f#O{N}17 zG#wExG?0{6W3#mu{Sc=8J>aS8ksu1z>IN%B)(5JeO;s8QuwGz+AMK>|Dcl^$uHu_L*JK-u3b@iC z5oB*aocp(OcO{@L?es?!BJgc-Y8fez+BJAh9gFw}?V_DkP?XfoF$wsEcLclE@ntPb z=r;)rPY@OnF@&&nC@-5lf57_H;;x54K-qT`_4_P^OyD$@Yz+I?h&&|-_q*hhzAf*j zE2EO;J%+!)cPr&b!?Opo!)uT}&gg)v14gXh>;Tq}*1b|5?-!{4w7zgp+A&^Wq3!h* z%xgU}+)}LI-qr@HwE}uFPDfi`BQ;m|EKL!QrZ^`_5N>HpB0>Hs|JF$K>_1PkX8&cF z$@e{396;|KOu|++1&q!iJtgMiYJk^z5aH~w->m8rPI&EMF{Em8M~ZXTxr&25_k%+E zs1l-m+R$}=^*35~+3K7AQs0CfackA$*N#I4m&}S~E1oyP0Fz8s>zV#zvZO)?%}p-h z<`-IN!0xG_)U3i^vXhQ=(;H}`&-^(05+V(}qZ@81u)@bngd_idms+14oV(@wyT?N= z9_rPm&69^9xDyGi@R1xWCUiVajFnYN@B`~}dKV1S*)4-JWaEsVp!XK@%>&NlkKuLi z7Ms%YmOheAk0|&)s`_HUx!>nE=Zb5&Ul1s^51^|ggjOVQ8?<+Qi^;US{>=qcm5D@s zXjS6U+;v1jM$4~3WNA{toz2l*9g`||1I4bfd(Px#3K!|_e8g>e`n1xS^evNGI$AS8 zkywI`QqyB*z-#f*Z+*PAks=K#4%gTFi1W$e`4sS{Yt-)yo6<-#tB|MsjGhb!PYhvy z0HJK_5M`eEm48oFl&rm9f#yS88G28_cMIb-S^JADsb^ijAHXQWrhT_;$dsNSnoH&q zy4HdB%Y$?enX%9g)wT|@ia|Lv8yW;QDBUwhypE|2AgZDuys`TE_E3As^y^Ju>6sGjjwJYqrnr8{YQG#}X^=o0rhSJ>8kDa&EN5T9p#oP`9WD zPE`^rBLwvMBK-;ydKMdj$0pVr`Mzz;<>*J%5{}{bR|5Xc=wm!KTN%_X)<6Mj3m+?B zOZa((DT(~9ZhZ^C1w0Yq5BU0&qKNwacZCoKad&T=i8$+#AZ#FDh|wZo*|{#F5zg@SBi!XQofs*J>bP7|g)alQh%zm+`U zXs=)HVmLJ9{FGIIqt(X8PqvO1X^T0jiKA@$p;P{6u1S<;tV&k!YCcnzr)@9=ZO$>-PJ|CIKsdjQ68Y_hZYw6GR$S0@7 zS~gt+@VcQJ*4qbCJBQeA-9W<)@+Mb-Q2XC5>^?c0tuwEnkfm%Z&WyKr4_oc;DTD)F zF%CyCwSbAX>lj3(!~qVjp&AzlU^j1c6mY?)ORbKS=CJ_R?OWNcl_w^lJYAxnGJq+0}!UxGW-xZz}8@@@{|L@BkW}ItMy7!mCkciU#G5mp%SCOBV z!SjfHWYDu>&OQ8$bqF&8MT)b5!CFV;`t}?VJic4*3Y5PB#q7DZS=#b0*%{s<3}}^d z@(_XVFL&h!7CP%*iJr!|@2M~K$%|EOI|aOthNxy&kD;sEuQ%V8dB%(q0SjNHJfkznRNWU3x+{?N6-Z+Bh4mD#EE|xUYV_8<(3B!pyz;T6 zgz@aSMho6BB3n((o_+^Waetr5r%c%m^F0rwF=qqr*fZw0h-L`|bS^udK}KcJW~rRM z#hDLOfbu7kz^xC11U@;vZSL)+R6aRW0kb?9*{o+E!0y(eAOd6@nD zT&k*7e}k9q@!KZ|rkZjz{0ZXG2|cgI69hJ-5Y~9~K|$3Gx2b#`f)rNd>YA4UcS3uEMdge+}$6Vav2(WC^w`C0g{<<%11*4Z9N3}-?w zgPH;`Pqy{`>ae5+U;Nd&cd}S_9#izMbsScH3?-*ao2ptuBU6oh3_UoV8cOG}`ic6* zx(YTh+~hfXf8{wg1;WOwv1rz6Yy)BSp4YPXG>?U)Be7857i3i-HZzS{W{H$qb2NffI^b>L zGh_e;^bc70^xz7#*0aNBKC3`0V}K`kKBY8Of0bGfvz2#U!39Y_N2hA9>Aw;kkpXp? z=5J3-YAOSZfxE|$&C!TkG4=;QTqf$vv58))*YVNamZHmN>Mu0km#?AG!uY61Gec-vK$cV|6iV_A$c&4{Y$Yk3Yo&cwHno2 zMV1=idD|*b48b>2ji^4OU9#DL&a=02k>N0`11xw~ zcOhwbfb}}>i2{N>cDdJ~)NLIUHO+Sd4#=jTi~J}99@3~USD!)2WJ zy)*j>VQ*Hn#zIvi(xL&bunodvsF>QX;hGRxl8T!89=aE-+8O8C5LER$FI4Tt*HVz^{Lq#-r6e_ z^uAeO=P$<6&;^YlPB+}S4oh1HY2{37z>k1*|7Jq8m#6&Mcz3`q2fv~vkf`eICd7J& zEEzCwCN6OD;f}e0+zD`brdkDjE}tQbxPf{TDjnrEj6?zc7RcdDviZ(kfK9GbyhG9* zC*Ld|`W=D;Pl)3ue6tTUL@8X;91ss7s-$V}wgxJ>>A!@!TP{WAO|)_VA~&ql^~;qs za-+Ogd}}>Ai>;VEgu(b(^WmH<8df}PBLG&k9Z5h{!67t%emKqu!i5As#2oZ88G+NexEugxkpQz3s1y`;9~%DZLD6nU z1*Q+8K;dKPK9V~4YEmh%&IcQ^YViD@j!(jeUFR&xOlx^uH16*O7TU%u7k-l1#JzUS zx4bKJ0zo6LpWFM{m|?C*r6wm-;Ly z-tpRUfyy-foEq+KWvK)5;ScDYoo*F-*5csrz45DR5wKMxQ_`~!TUpI_NQInAc-Owy z{PJG0r*!TU6Q0ontpYpgj6&J?6P{GJyxp}Wy{(`5J<2dW(Z2$%V6& zsCGp>aHj3}>lQn(Sh z4{gD^omS-Us5ugjrv~g4^b3 zn=-+OZtsenO@)F#dL{^*c|hTK^Vs?oY{1v7kQrFUuY{z}KVm9rw(WqB=9Xd5W^48b z;GMgpk3;SmPw5Va^z4}3~J_4ubkUoHaiDjo6@`@7JThGpScqtfgFd>JZtM+<} zzdj#MlYyofUOKvIgJbBxm@P^Y4Ohj+HVcvcel-Zx3}li7cGj=uOnsh%(Er|+-v{K_ zzWe??wim4ZEWdI_DB^P7zxxYG?RO74kWJy&`%L`6C>8P}vlMdpzjqU5Gh;6B)^Fta^T7?2N@YmV7{8ph-|2s>lt|l_ z^$yLDhKI(Q81pyZCfdnRA;l|%O!$osl{s$rlf?e zk3IeR)cnOMJR&?VkQj@?i#30MXiaYH`=7m_g{f-b%FI|W4U)vODSr(3;3Q#-o%_!# zRwAehysDMOKjz&)#w&pz9YC;uj!TY3rNsTEu}a|c{dFzvppdH4%F(zGc?&?dXmE!T zm}S4S%yR%Nh$@KzBU8uuUq0?4Oxe(TFs_-_wLNl+i~H-)wumd3nFf*(7ffGnWjoU< z965pBvo!2Ld^>!zW5|it<3NYuwB@qr_i-k29n8R@*{3jmgPO32znvY2fE-(01b2qcO=634A600T?TVv=(3v+X;RBB-y;4oLveqTuSz;TF<}{X2ivS^h5#Sz zu*&Tj97vwMN^fbKL`QiasQb709-p`BB4!i;JNguN?P?ku(diM$o8*jAp>+T|s{Pr3 zlg)BJLs;Uai#!_S7d!*$n7Ia>Xswaz75iuF1Ekm{pki*D2#5Jg(W?;yQP6X$0T(yT zoElxhfMi4<&#$`>oAuygKjEr^{oXqPFN*&F;!5o^I`;w&2)3Znx9zCfR;I1XlkmQY z&N2UeA>{Ypwd7{NgjL&k)N)C!6$ekZJtsr&Iy+j16Qf`W=a`Eb^d7^JU|=Wfi*W#h zdZUSjkN`q6f;@D6J9kYu(wh{CL3_1Fd2f_d@UhM(cB}8KKaX*1drT|| zK7lE0HndgO5D)NWax2KQ?eNx*q%bXhe=>=?Bi#K$*Hd`VE11tO#7S_vU;eGC_IxTHu5LI)-O~upKUOpGBf`4K@M9YLDq|J`FMC=mgM_q%e z%u^o?A?sUwpH{O!w9EL!`iY(|yi`m*x3Fe0qn?<#({{o*V!9vuLH%OTmG`Nk5$pAE zyJOl~gWVErf=41%k$$@0RM_rHcl?-xzBBI0riVaMB@e6M`%@|xK}p{EfUAiy>u9N_VJ6jB$N0!H+AvE+Fn7^j zAd(bN5Csd)9*z8ND=c}xQ;wl5kB1C%h=M~OYeK#fFPcpw_Xc|0J~0MVqNl4FESo?I z9*1bqk>c8z*2N$((PC5BqS16%mrmVp&C!WF;0;4}!8&csj9U0%Q|X*^lH;c}{?%t3 zeYk_z1}kC9VX`~n_b!!~I)#(|oCZg?c#rGvUjZURl0~3soNZ8{kH}ZATYF0us<|6d zQ3G!r^j8y3aut-7MwuTDN{+O-q!pIr3drUx6o}~XUbt@e6K}fB9|mCWnijsgN|Ai` zBxddL#1FYeb==evGQinqX6{eHyT9HWcV!VE-Pwoc%{-dMjaE`RPBxb9p1)6Le8mDr zS@#~kT9BUNkZprd$v~XSi6|CUS(fY)+ z47O_A3Yk76r)6JjKelTmfD&TreEwQZlBqL@V|y4GrlIdw=zCF&_i-B^N~5t+_bM`A zDJ#;UK_QLTJ!#F0_O*PIZzMfjFXo_BwhlF$cIeRhuW*BjNJif_QDeb0RU32t$GgVElle10T(@a5GG z?NS^>(rwM`RIrQD^#VMmZg1IGJ9hgIR5w4MSK*)thIU5haEnY<72o_$T7TM#?Uq5% z)G~Ayc6(bL66Sk!DKSVa>#j<2ODt#BI`N0UiM^Gbsd7z!exW_iML43KyyCVUW68?e zS#$RLFa1Rv&9B-SmSRY=bk9cgXsHrQr-$eb$-5ge`C*!&p)V?)nD?C0;VFEKC3f5r zg0OljkHi}kI`e|D(`Ndd70akoYUE+^)#qz$-ovM|$|thn1GQ&Pp_$yOlx`Ov%j4T% zUuJC&>rOKAZm(N1G#Uy_zs8F3hbQ21J$k!~4DA*EAXG+s+#~P;HKhS&(|<_-7_fF; zEl;u3U$U}PZ9zRUQn+rD_qT_LXNJDyEl&>Fm7Ks!olbj57&<-p!#%bB zo=h=FOeE6tWRj!3;_}Rnqh2;SABlUwKR(r0yuRFcNQ_e&UCYzP7X5cqBN-Q|DA+KQ zl~zQ>E&(K0N_Py;$_o2S;O6W^j1Be|pKkFVlC*7Jm@gxFufkAooi}mkY%(5M1H)@f z3UAAWqVijk#bEU&hivY-S_&mmaDi&ckey>u_X+1Lx!c+$Ptxss7b~A0iSb=@H@>6& z@PXA7$+MNl2UhCIQ;b=-OvFroAM7dKGt+6Fc>MT6u5uxb>FeIQYSugC;(_BZi0=9g zpjKKDKOS5K*&7N}1O(%li~0bif{ zL(NsW2oDl=N^Km=Xa8Ep#k|T;*_&eysmpzrAzH>S2O7N>+TYez7g@2JZKJ#^%#VWy-jI z+vh^ITr?=Nu;k@Ff`0d_ba?ojfh2U{GW9C`X{L&-E>$d&povkyCZuD~PUXEr(SOa; zNpvNHp7@={AlY9S%je6CmUih(tiGnXT_MOhqCMCV=p>)Yje3_aaHLlky@+_~^eey1 zUA8Z z0c4w3Vj?9TIi}{VcH2|9i)<3NLS1DV7H{SA`j>@U0R`NqXK`vnk?E@!Ne3r;uIq&- zIm5J@^9%XLC+`ba0z(raIy%kNQZt@G+rBjB%~GGk^|k*pr<$Y5tq#8}I%gWaiEmyY z4a!W^9p{!M?E;{mhQ#I(6pzpma`6~ zsGIbza;j;dAD^-oD2kE%25J>Dx-2<~Q~$UTs?~Tnxpv8xA>p}T&S*L!a3Q~+3tG(K zO%w?Kw(fQ9OY(X;%<9JPKb?1)S%AL6D3rGQv zBu5dsP|(CGGB;DnPN)356i*@l4auncV*H`PvF|sd{;Ll!(J8F;qObF2CXJJNP{QG- z8LW*`*6vvDU(;c%7EQN25yOo=*6Z0V%M?w6g=I6fX(-@bgjZS1yZ})7xsIbqiM-ui*+p61ubNg(5N-)n zP)mymvfn;sp@8;7$%5S#2rP3FRce>v37SStdz3GN{jT%H=@T_a#Y0lG(eniYD_qW?P8>5d)j z=|`MNt9Ow>G7~Y@yzbfRkmSIRkHsYxR;j^wuOTwr{^SyJg(5fe6t-rB>_)nPU$}J& zmMzvRDA&x)g&0m4lK@f3nkx{*F398p4_MzS8iuAETKmczl~+>@^v5R1e;wy&_$C*1 zQ%?EQQJSv<1Mb+94CrF7Ot}4{mt*FHzO+aJ&o-0_!jjqPKI~A8f zj;=;r5lr_(#G(o?8EVXoGc`5Vmgej6PrLw);OL|A;bAGN97mOqHH(7m|cVppZpxtyh|N(6`uuWO4ex@-@B!vuqRb_Pfe&1>C`Patob}^o z=7v}2XDl;LX_3Y0OBjw>V+?z54C_Y;;-l|8i;USwkWp5;-lpV?{5%n;3_FB7fAaTi zi$9K^neKOxR(|1iVWl7W!Q}$%%~rq$(;CH6h0;H?D|BUCb8@=?jj)2{!s6n!kTQRg z)4&nbjYrPdF_{URxoj5qXe*ruyxi&`vl{X=vfk zTN@Yf2Y}5!<*^qc^%b}!{HtDc$cJ;H#UNkRZg_DXZs75@`6>brc*=C{rflcqvn&p17O3%stF531Mjg{%G~Y1f`GzJnBT$Aug)T!Cnn^m4dAYGn4rh>m5?+I);x$27k>|afdF*T`%>Oxvd@c|NCax#0bKf z2=SY$Bg5M%ifqf{`7Z(no)Dp6#^dFqos=k80a0N$bB=|{3gEiT5Cxb&Co7UxUc6ua zO<%g5pfUWvJd3m8>3y&II@gd9J|FSUz~`8tCDjskmBIG`*Jg`x+vnp?#W`^ zgqkY_1LSq`fWbe{ZEe6y9owl@2;qLM*ybs`r6m;ay%8(V1}uEaONxXZLm@&)Vqi~D zM_D=?AzvkQ&`@#yOAiG}pn}?=Aa&b5JG>VnSZGmCLj`}N#2rsSitFpNZBnGUg0e^? zPTtgsmKWHoOS>u|0ruFW7@B}1cDoM%3XzXUa^&ch?C}%0ps8SbqeclldgV$8Dock7 zBC~j(;Js7R7I)=9y3=i!3kB^;N*tE=@dY++UJvlP4OMr$=#8{vD+xS?CBSu9j0`J6r%zMPl!gs! z6X2#*f)`&MWszoqD-db<7;&RvO&IlyL8J%ce7*EFZLd5rbb`1E5lI$`|6IQpijx9s z4kav)Q~BZ^)ou~b{k!utF>Qh(j{D{wFE!fBO|<-e2XiNhTzKtWV`}|AP)mI-R~hgN zrye_qJbW2WC2#trG9sR93af^(0Y>$p zEe#W@q4U-131c+P3<^dY%KIHuVwpj41I<|@qfq|SPm6!4G*3?c4fX03CraS7sJDr?4)eG*V)e?=~si2 zYZL<6I9kDS{y(r^nB*&F@(*DMCRHW%UjsJf_>{pjpwiK3HgLl$RVpB!u5vMo*m$-*)zam zK*QZ-T*(V(myX$Ox3{A1xjNK(ahTKj@nN_E0uh4bIIZQgIOu*H? zWMK-9q4Ky6vqe6ki*rR~%#Po%L_ZLk@jPV{boAcQykYW^s?PP~NaV}=^P2Mwiy#q* zUzWFoC$O=-9hAQgedzFhs*W5Rd`}u!j9UagJl%W>K1fv=_MaW?dYXqRC=j6zeoo&x znAz4Ce-BKG3r*ZeNOI(HY3Fw<5U%)&g2IzOo*!nDZaJjz3_N)BbpVx{4nbG{X%IB? z4Ekcrowqn<(v_7Id6~yw5XbGf>r3rApOIc#$^(w9|Gb{R7}C{jXun$1@OLpIpD)N5 z3+?sVkwNaWp7^t43h2LnVCum2_&BJ#Uruew?gtf@=d?#d{?WQ)-826Hx#1_oZ`ATS zHIdipAfURa@a9#zKy~5Gks+hL+rabr6lL-Pb5CI*Dm+S;qZe;Y;DWTO;JuV7eW4mP zIzlBTR$C=vx^ZDT4#=uq=PubR5Ncp-vaO)7r0!qATR_}3!&&)jBDV@ZY;1Z`s^CR> zcc^7i$%0DR0{0Ci)~h@?wMPaGH;|OX?&&%;nQo2k1Iv459xH$ko%Zu}pt)EyqvD6` zTEb9KH<&}m?b{+J$osx>qZyI4og3Ib_vCBg3V4+1qOcy2wMiC%G_`fgUa#?SUKMxW z_w=K{M`HhgQPQTI{wj2bL)=1}-^YfL*_?d7hXb=>4pQk$d0wob$H#>6;MW zQXw(>y?BVJz6QM9Uvhw%nh*)8kYjpnq@Qrmze$|z#H!tx{x9>%USs4Pj?{`;iGH4x z?0K(?3-m|_66mT(ICXlZaN~?!AF%DH$fxCj8HT(^ck3EI9=|u}+q1k87 z3Z2@(NZDB7voat!En}Qr9Rtr}7foIS5+!E#oDp|y-0^19n!TAJxk0K7`22WqsB;(u zfS)XTA488)j4Ni@3Rha6>d8@fwh5!<{@^uYPPQuRj;k)&;m{||3EW@$8?5sI@F~b9 zy8L`(L}g(9kvr!Q5W2N$1HqZlH3lUt7Wad!l1Ciz)4ca~N!C9aS3Ayu^YUX@%%S*y$p7p~cLg%J{pab2tF4~^W)+0A+k)a31U> zEN1LCyn+l*f4?n}6)+kG73_})nJna}H6o;qH=wl~K0FQAM>FUU8h!@eQEw*qEbfNgxD% zE;Q4<(2)8xiF<3XD~dd^;m?o>uV}RDJSz}>L`T2ZqmT(+bY$ebhig&~OkH&a+s7Z( zQ7O!Lws~>?Kx&;p;Uq(#ou!U-khB;id?hZ?g55+u+N=+m9z#D$geVH4Z&fpb>6* z7v>mV4m2Uktbb(RYC>f220WHLhF0ghbfseEho>UK-!3A4Yf%XewbuLJbIPkfY5ws8 z=>b)%aNSGGw5HURX6SuA6KJ@woJr9ZzWnG6f^2?Nj1%Jos>6Lw34ktkvZkDc2&&HG zaew2Kaou0VCfQAUeXo9Z?{?{s4Ygb4Uzfgn?B*YP$B7&K_$>4%#(J(nFr0sW>A@Qg zQ)bGnJ=vbsjnY$05TfpP9aA;#QE4Lbe!cEoyCK&7UiK5aN4{H}ukkGUG(ok#l(z=#e~2 zKs#&o3CnI{JkW$A>#*s_TFQTK?T~Iq;B?a}-H~<)!@X zs>`2C`LX>N+D0Eji4Uke&V)ipK+^HdIg$PFzOTl@ zbW1Qc;yXmrA^(2*86<&ccZ=f&It{Wkqz!ivwN+h04jR|E2wp7qh8Z=w?j_3~JZ*su2Uh`nsi+yzPPMY@2${!wZ%!2_@sb?)?vsh=k& zRB)pAjl=d&l+=E291z=3K^S@n)FlZCqv479Ekd#b$U`A2$XsO{lo?kuq^EPxdoRTC z;1|SkOHpsaxV&BAw`Wkj`0SC$F`SF@zrV7MO{~fjKqsa&UgXPNSse6Ey>xJ^n8uU= zf3n7LCZNV=zg;uPLPQPifOBiH(-;7|Us%+!4GPLJx$D^`G&U}T7l#dqOuBD!!(Z8m za|!_SN=Iu&rvFP*akqS^rb|YgV`A2|0(S4}HYHx^&+6+ugSZPqoP{*e1Xr+-u8%Tk z5GRK2AsNnut@nX*48RDNN}uVq<~j*Ap|Z5X?)~Y0^+UKb1!tnOaL)W)uXC3eI2IlC zi-Jt?Zpx?OdAj4Z6b%*?rk;TJC^dr(+9=3y{$0i!XvQDIb4k~Ehi>6hfnx~nMiE8` zRQod?025VPGz)hJ&c7%|9-xOfzcnfF%>Yjwe$ z-{DpH_4#7eVw&sa3)YS}po{zlT8${WKR${Yi?H;JpU#J_F`z zMK^2h{uMBKaOBq4*z6lL=npdk0EP0lZkGWZqcA*ILdANw{~avRg8D!mHQxp zeBRo-&LLyVQ1gYYV^*3ffYRDE<2wd~+72sd5bRlnoeFS+NdR6!6(j}P3#@cR{`iw# z&zq6)tTvyS-WS85kc9UL{-a!lPN8>o6RUUxdb6M6#V|uUJa{ksTJ>BM9fCPqB>D`v z3myWp6Y8QA%2#dlk8e>Ze>)=dY4mMUEN53bhfE>c&XyhZ>Q%7u-_gQHo^1>l4?J_@ z(~rmiOu{L!xRU0OAOwwnFFKX@-_>=U$_X6rbu)YKgK`KkF#;ScItaX#dOi77@V-Ffpiiy5ju11!QR!e-@}!MQ&u9yUL#mn>leY(D8M{VcWyHUKOz z4k%r&tpn}&AwHM???Gyr_Oj?sqMLBjp6)2fL)slgEwmqGeT6!I!-(E1fW78jhnD=B zRqeRnX^8`QX#*jKlf804b!G-2+g^0LlcmZ2ma>^=8*TgoZqH^tOM;9=nyP-brsX1h zPwbh<{6u z8uESh>3t<8G1z&Bq0Cjw`O4fSm?wJum)ACTj`np4NA71kf{#BsZ>3at^~kwIk6809 zTMQn}8q`FH3LLc$=e&*`B=RIT1!ITdXl=S6pU4?tbl5*)yT&Va0y<(X4hm(a5Y| z49MEI5pA(mt4LXpTuING{_EvvTqWs~YzJ)A-0fIWr_Ya11814Nt1YNMn;bgNC7~&s z?IcBSHB*7a@P&X@+w@sfy`<`zv8RzI65Yf;LwJNye*VtQ7R|&jYrc9v-naMF8?z^> zkki@hZ^k`D(66_9L#|Z2M!%5~&_{k-3s-*;b0_%R>b{~K!3DZ)YYa;U>bZT2LwZc{ z1rP>v>o9*Kb@^%66$GkpPS~KVq^W2wveuy@5Tx6=46JH-5PlohHnL7B*E*TZC+x&^ z_{#TZblsaHkKo6;RDA|_d)zK9Dc|bHkbJ}09B$qX9g)ans}ben&m{>@06##$zn+F9 z4t22HeZm-iayP`1s;}(of%h3j<=tsBfJH683-px>_puOx2w^m3W9I!Q@RX1k^6*0x zlRB~ebvLIR$I&<~uibzH%}dP{aZ@ zanJs?Dubb%&5(~J8n^!|CLQ;PZn_}ifp|Jh?f-Y>Z|^q+fBC@<-pL(&0(R=(1-d|7{ z!nY)0+EO-cNhA*ikA``+xQ0f$TL74l;7SXeae{57p)b6QewFn@U;3S;y-fUI|J#ruDgcpL^y=3kC@o8S5A>Y<^v795mnZ|Z`vPOFop6Qsk`_7i=a+A1vP z)?Gb$6zl)mbyqP`_z^!SG%rYZ>s9hQ_$zc={#Ka=)q0QHcGwJ*NaMrvUuw}bA9jvM z`NQ=h&C`@M4PJ={_AG4KV?VPqc;hUGh=Qc2N6h)>xkiu|jC@9r7r| z6Ge`Q=)3jHp6eT(1D4m_mqDAXM~~iIEnm%3i%GmL&r(YG@*dtETh5$Mnyv3|Hw*TEMhJs&j)yEOpml%meIEt~p36d+eXf6BE`MqX0 zhQwR@Ro7i_!sa^dUCLBRHX!r7AV4VdmH$uCy6ClVG1ai1)9NS}l5&xTjqUH#vOOMu zzc~fNH9G5mO@(NZ^m=UHoTz3Jw78zpM8jC$9H9=k`=6Qa-JH-A=o7`!C;1Y}37`@r zYDuNVU!ek{tw%sdg#feoajArX)3Vb$qS`g(4}Hc%)<%Y97(~XUu?SxaCX&3!(vARZ zP-@)FED^>2Aoi+6jVkK7#i9|D&H09kNi5N|K-nZgGl9W9EHR!2`fSExr8w{W)YYNC zK}r#BLh!sfQZ-7&%KtQ4STNHwtKCjUhtKwonuq>p3M!Y_{5=oCt^Qsj(n>pZgXrj5 zo@k2x@v^AnwP$O4QX;qHSTEG9WQas{wx_-mePn+2VXBiqj44USu*{I6-!Oq>Uz)FZ zN;$nO7wwRthIZCp_LbTnO&zUB9eHNi5Y*C-<{$e@M7gIzGGEF5 zV*{0?b`=wus-tjkUPpoaQi{9@jV^5aEVg_(LKV*wPsI3G`b-`6h=JM2Xg}%g75|9U zg3C5oOxEusmB-js40S#|1u?&*giAE5b4zD@@*FW~KJOD@8-kw{-?1Gt13NtLPdtDlv37UDm5=f7Kzr6KwZ1X|! zG||t{k;T)%k1uwl-WJv00<8h0=bYv28C3)c?4;!Y~-oLjH zByjXGr*&;2Q_ZWH;B}fWVeaeXAQ)|sFyZFV5|If+%cf5}R3P7>A&UJKX(GKHf#V3> zO;#~Ig(iUw0qabIP$nrgj1X7zw5qut`-ueDh8o>FLgVJ3v+GUOlaGW5uPVFk>VFPzX%zyUC#H*jQzi7m$Jn5Xl2Q1SYd^ZjcJo<^+-Adjc(<+=*M}49neTTD&r5|le z%|eUv9R<0exnx|U+nUusc5!}{Cnnu~ze>lj|1|A|SM>bLm^7zok}xxNa-j%9s={j7 z9qz4{&HP8N-!M86Q|cjNqzLA*jJh^KXJ zNgN$>Y;&8`#t9hM_htT7f>5o))T33Jf2fa@^#$_eh#s+VDkP9xO=NVDG3>i@HhPkX zG*pes-%~mAoH&1v=UdV)=W~cET1Ujh~7(zto= zTXcZH^c$AfOGK|V#AKo&>2H$#jCA?-X@>9z{A2(EQjc2%n}(!(gaTnw@8ocyck&;N zb{=$g(iFKxTYskXHFI920h5qkO#Ey0xH?S-h%071)}b=|xzFrx$yMa4uto2etN5)p zmiL`+h&^Of?m$NAjwvJppaw&mV-D1cVxX zl;W`xZ_ES;&A8t@acLo??wAHd9e5VV1T$4FNkg8VK?1#^N6>o;EYb^4_HGV0W3K7C?c7&*B@~iPh22TeE;& zvV4O3xDSO_Tb6(@s3B_m=HQD?89fDzkaXLHP02lWcjZyS5LlkinLjbu6 zlPem%F1A$3e?NNtw<{ISvwc%lvUP1Few;q1cxn)Rcy)djj^5AeDlVOzxdu_wY3kb( zfJW5FLLu(BAA+P{s}T_T7Pplk-FFaMwBG94**y}DyP45L`yc!>ScV?8Cz~e9-7^R-IX=Qp$auh@5WpMp#?bMe(W296D_wPMUp0Ipp zU`}W%s?PL;j%dd8^vg}1N{+sa{@w4O)uS712DB8Zq@>hz*)$iHoFj*0rz_E9*Zc`| zn58{x=8hX0FsN<=jXiU5C ziOaQrwld?bmn5O>VXRc^J2&F9uKag?hWl@qID{JszS-8^<<6k%Ps~@#NmdhyCpEH! zkZ82|2L9McbGGlJ6C&M&dgzsr^VaFy72&&(xyDdbU!}wY z6PbHZ9regUID@o-nB_H;Yy?yk#CP;Mb>c#sNG^3hvWE=q_1ipx<`~`^%%E?sW*5Sz zC?4fCjG7a?ftGfi;~53_EvvoJNUw}mSwu<28PJ9I{)khHwnXoxiCWyF>MBFQs^hrL z+^D+t*f{W9@|^s?Co7<84Q%YkB#vLMK3}DXL%r~-YUrcNfAxOWwExV&KU#5A{~^(a zk(>|KJ+G}|kma?+E=d-FJdgOqB7^<9qxhYpBYM}@VQSIcE5^1C!vjP|W3(LR7`qc%&kwHUW?Jr5a)f3r%`ngcZ?bA4MLQ0nC z|KaGm!`XVizL>pe-%^E~(c-1m9UIq!brJ0r=xQ2EbIV)kCZhdAOZYCN2oZXM8L zEV9FhIKJxoZa~d7{ng|YqCw0Ky!1!pE#FTc@-d!EHJ@&>>g6#kl;`41R>U>h%(#~n z$-=Pe$Vg`q=KyTx^Z8N(OHn!^dnky+%I4Y~Q&CdpB7!#HP||~eCsHkFIDkbHF+zK5QHN|IKeJn~%r>Qy{N+q}&k1Q?F_uP2m zMHRNgnt)*1M?-!#aC=k9^F*E;Y%nnv$%Zb=nVIafO7NxG+`6Iv)HshvNS%88D&YN< z=F*?<4Cg$ki6{^?72SdtEp@tsL(TxiG%_?rOo(d!l^7-0mf?*ZqEt~(;lcxo`1!Rm zI>QcXC@avpK|8R`GENIKlHZ@^^iwr41UBKEIirH9WNW#V1TQLGwF)+pPcDro0n+Td z!VE(@nU%lg67XJm7)R$3?ZU>R%y1=_$QG&d?!vBrb>lOOI4-g|iwUAsk?{qkgjJCa zjN!{H7qjy3(fx%Vk%I3FlhN@X%2DK?*Re(LQaw!CxUC#zBy};kB8bsTZf9Z@kgcHY z${G0)+c^ISU$eKx>F2IQ)>d#U{|R(Ny|Hp+2RT`H&&a%cM#0Mz`_un6Fix9*lD9Is&Q7-BJFoq3gS3Vl)D9`NMkx~I zAzTC7E36aJ3Dm*MZR!~y^4pI+KK1=*oGAwRASN<}_Saezf*o{n^X+R%{nd3luT|dC z!@BdNJ1&@X^k!Y!M&5z6vlscmrlv(by@blTfxl zJc4u4g*!|(|7E(WcmdcKcuXVjwGuQ4rQuONQOYay4vYwn{F^VZe3qt%{E^F@>qY0( z7)Ru7(#-`OHZ`o`otX}szN%$%T?bAJ1@A}mxDDFF8`2O!$S}l}`cq%#4gI0qGd@Oh z+AlL}Vy>tv%TYpgYIa^w6t}aH@ynRgc-ziEt-fvD_qvaC*irry`W);^vzg( zi6<*6-`EhH-sSlw^tt+;^=`hvCiAz%40sz#rNIw|$CK~fxlV2$zpVORo!mStwUVw3 zSiqtb*F+}!2$9k^^xFb7mkfi4y~z1fC>GerCOkVI=pc3btX?566vJH=3fU0)ey9uz zRrzUk5~(VHxlUG^Pa8QAq^|T0i+DT?{ ze>vW!{wFip_W@mN^%78fqX4T0)RH{KegmH_3nY@Q7k>y3UDe&fwm`Dmoc1Db=;QE| zV1JPP>>=~ z{9$A4;i{C={I0|faxUd^r8Ru4aX;)k^{@q}is%=@=0*BSV~eL0|?TK z)RA?rrlwx1fBH!cn?Gl7qXI6@^!bvdI7Ejb-M>F8iBX-dm~;V$DY=w?6g)KD*GhL& ziLN3{0+lI(w~Wg}@Db2wvxG=z9@L$_pKGQ3y37}kf;DH^G3HkI7>0{Bq1weL_yR$? z5cq?j^VkGTa?9Y?JxY5lsCxtP=?0QB5>=Y>9HMfI*VdoXvfFIq_u*Us42UoyMDmgk zb606i;)2O&QSC)qK;55o!bziZ8zs0p#@kQ5(aq`|Z{ih1wzlX(Q{OXu;nsCs6?kvY zLnBrUe~Ri>m(=&?`QY87LPtW@FFTl zzwDkg7iS6pWj1PnI9S%Rrcd44svzQ_3b+m>^UB95RqKiWvXu7{T!$r0K0rZ47Ozr( zDmQ^_UC+w60BB|bVS+p;q#m!4@IKQ2vUxiLJBDHy4v~;UQ+_AZL-RIb`cU-}F9$YF~Tj+ccN0_x(r)0 zo3cT*r7=^qlYJS+|J8wR`cb_$RHx{fosDk!$Cg(yNOMY}imv|uiO?yNy6l*1)Q48m z8YV!^@=cR(98^y{pUfUMt?o*3o+Ejds||SMsaqU40+x!+$~)zwzZ<@85~o(1Flm;} zohNNGN(Yh;byTR>Z4{5-8((mSmf|Q@^Msq|KBlhqfLk7ueY^a{($rTZNr#Dl3^))o zaRCOk>^98j>_yHn&K%;R@wY^33FU-;`L>2Iwpk|AO_qrg{1fjRssce)kH45qc^cAw zxf+vB5vG&iI$rVed#WZmpj!4G1_}+*t|UBC@cQFd#rqtT$lsn#6cK?+J z)mAF<@7|>ToQw;eyoo@V(XD>XS>Wz{PKrar0VSvGJg9BhgaNQQw|P#e$vL9#y8zT& zEY`X2YXNLdQM$__`1S<=1^yj=`@&|cG&Rt<=4Plw6IcTmo`A*yWhGu=cS<+*0h zM5qa|uuO=9U%qccd$0tmji-FI>#OS2xJ&n_NIQ1fQiWR7^+8!=bX)jW9ScFK`Ifp0 z3FW3oDOktDrshKQXnMXl^%k&ssMO?1-D{bm98#M=Mq6qIF)$XHbC>4PF&4er^^%m} zL4`U?C`r!f{MRP-7Kn-n5+flwOho0KSfr~zpeA+z`O?=!^mpoeUAHkpz}DLzLZxwV zC*A&crl$~Y*aQ_B=4~WHLcVz=iYQz7E-!;12aB_fR!gAPo)ca@ad5%RYj-S|xB`CH z%ny-1-j`f4JNcmny8zpoh;oIsICcFMi6jowIOn{3!z-C=R726)JY1yQ@(oYTE(38? zsg(sCApPI9*IdnUur7Y~Q;1sqfr-Rn6B;MCu?dve=gCgNH+=rQ6<3Ip{>QK`MDV{8 z-`o#|_$f*js(0>uk%<0*o1hF<*H2L-ZSvgc`xYQ`MQbZ=iqx)_yF1z zx|>>^3H$Ioa`5sY3UbV6(bF&DuT1WwB8kvH55XY+s?qVwk~p!Y%YG;iZ<77xOS9Bd z0Cr29ixKnazn_2%qi**PX$1L}AbE4eetvDIFCx%7*`bDD~4qSeFLObV0vLDKSiL2H>FMJ>O?S@10=M1xQQ?>7PNkr!`dY$L` zMe8LN8-Gl!^QXR@?q*eGd?M6DQBgd)FS?0!CV>V0_dms92z{{^e-O!NW1NyKcEmUT z7AX(n$Sm6P_GpXGMc%urK)w7$)3=41$1kJXg5ML*ErJU&E-0}Bu#o|$C)c-AtoZJ7 zZvQm_D2eq*;;Ku7-;}`r2lBffaUy4eU&H!t=$BLy(e%Az?-&UFhlf1^7Xe~=U6sPV zaP3{2hQ-j2Dq*g-q~g0bO(y$D>_ykB(C7Q6N#|3FxKYandhPpxQz6o>ZeCI0ZFg}iKW7P*%VN_30dqaqW%W&TZBbGJWko=HAuUCL3xh|$$G6xfseFC7|t z1#u3wNuY0hvKuZpO8Wh7=)===iM1`v#ZT-?qfuBxVfpyC|uRFiKpRU#Gq^Fw^pA3o3X#;$4C=a zF2CAB%HlycD|_K|fax*t1W5jMpigdDfJ^PKygr2-Gwfi*nNCk?7Q@o1&KHqBCZh@# zz=(lM`Rq1`wvw@*(kJln5bq-8OQ%g|INx}>$=cH!hst9VJL9KQP2Y^t>GxFm4LqJ| z&Hiv9Hm8|~05Qs_rL#l>utMH=x_#59Ul0&#!lukWt#JI)=jM^;84=8RzBqpuaT+lN zCqG+6(5>v5ez_*K11t&B0)J}%lVH06G)kaW&rjb*r@52E%oXCoHa-0v7@PHJ2X;t) z*`WQply%CNf3N8T<&x3%VWV7(J&=%-->XWDLsn}d48<`N1uyt8JyRv+LMVF|-Ts04 z6-nwIDrff@_;on!(&ziX`n+?PCuDU7Cc{!k-e+%TCH! z9OOebdbk3#i#l!n#MxlF=1YYa3Y$ zea?aR2}hpj+W{e$P9c3?YZ>;@b$hsBN$M+z9#Y(2OIiZ1?;hE47E~*2b*Y=1y$Cr` zQB8#SHp)Jvbs&E8y#&-eGWxB70D5LE+$g$c$Z-;Urb4MD%N{-i&=RKFXYif6qtgHb zqz~&|j)<@c(6$g=d+f^^Kg9)a9)D>3?4&)Aq`up%&dZL-n$79=@gS*T zIrrg$#lJZ(CoTca7aMRNPcVSdk&Gt|Np0P99{61pVtd#m-~fz>AhLvo%Hc}Nz)Oi+ z0S^!i=h5^e__%Sa#NocbQhwcmSL86mMnS7;Jz%di1DJ02ef;oQ9D@uk)DGm4-*D6s*ZV{(EDWl-lF64+Kiz zTl1!x)G$ftAtBKC2MHZjgrv`c*hY^!?isFR=0o?)qS{= zEP(vL3X5*Emp)--6{DUDMHGI@GBNgY+S8ISqumIQRnaY?h~h|j5n!|PQ6r1vBf=sv zCMw710GXN=H=d3n-NaPw0+A_yh-Z+38;qS_66%*`Dvvymc&GZhuH^eL(?4Hk`hJGB z4?ZLSUgReS=PNw#NA!Uq&^2PYln*4V@Q?aC*#N21_3s~B>++;i6+Z&rpBscsN ztyGC&K82tU1oB>h5-pyDaip*z_Gf4bXW?6zjwXtzL2lzx#zTxlC)0$mLPxXRJ%&dK(1!;Y-Ac9n63D0@ zf@5j@lP67$TlDl{d{w>a*iz?jD@L&}X#a4^R&cFa#GjtxoH|g=l&7BCl&EnBbisPk zr~?z^bSXq(w{Wu5P59kR^mth8!HQIe35%dkahiiNR3<_t`Kl8rB&6xbgrRez7v_S! z1Sf-4$;y>@c0~gJ*b+84a*rKNY)EIi4h<*?ssKJApa=G4@dAvMn2>KVRnI;a(G2|7 zcF&EcNlwN0#0&U8wn4)@%vlA^bZz26Aq4(AP|oEY2y_KA?gHYq6nSG6!lpQkOlDwY zDPch7a3?Se68ObRSniFkOFqSXNvlxThMGW2pR$x4rYplgmo=R7_+g;gO;lKQY&e_T zF$xi2{VPb&h-u?ldX7zG zpmgGqgS89T$7k@FKa9&^ddgu8jP$Y{M1BRW%Uh0uSVLz0JgocNVU)o^3HmpieuoTf zLzK76Q0u8}!b{CsKN*W_DOz8qXGw8~0*{c-RJpK~@WK-kO>)_^ z{s0}%PD6g9(K_@>3eOjFN>!eVC(1isloMZP5lduc6R z$(a;QOzojz_;?#s;~nB))_+UUo_p#1WF*H$o{UZP!`Qg6e5FpDJ9%w%Si(Ra>mNcv zdRsMaQK5-k>rXCXTdx$;$Pcku(Wn+Sm-8$MKCYki15fWTJ;Y@30$=>T^5 zyo4&2BKV8iiZq7O0e||fFJi?rmTV+{$Gg!K$02B3)UolOFN}gIsK(J;5W0XtN(<$= z7GW*)2A*}Dx%V@#CWA6KN3UzLlE*SW<@Mb612?icZoHz${#-b_@PbCIkT1jwGjk_F z5EQZ{{Y2(+m@*1F!%oBcEPLIcmbJb z2dpl8rJN$Tai5OVNg-9%ufI4zHOY{nkwIv+i)80ZL-S8`?Te%beS^bv?SD_F6*^!- zpbSqR+PCuavOIJVkuit2r8+9E$YjOAHHRPd3z=zEE2qQ%a6L0BW!$~I>!Y*5=hO6U z{HH*+4iu9)PZ+TZ)LL0g2U_aHOI#Jg_F}8!d96pe2D#;6FgK3Dd^JLu)Mwl&W!ttpKqA#U@7>n8W+{&SYQHoHHSgBaAVluCOY z2VclU=}46wDx}!FGC39ok}FghIG427h~?7NlGNnDtIZd%@qICFmj%M}JC_zqassLs zz6OYD3FEbFC02lT@vs^k*qwR6I9hLbmvJh#yVJiKaEu++lq)0Hq}1L{$Y&h^Zt6a z%ko-KskZ?0&RFBKo0ZL=w&%tp8YP7zW~_(YuBb5efSo^i9kDY-CfTWKD#kdnHH`${ly!rMx*7 zZig|va`R|$z3fzXqXLb(0puF1S^VJP4XXz34&^h_R zD(`p2n~WI|Kl#}O)x@_hK0GXJkDg+_yl6VcSDYN^otDrOH+!|6jx1KsYo;|-*lBck zw{7s2&oP@~j+#@cW}5#*;MsWjxd;R+t}69fTP$H&2VMeGIm>#&vm!;ya=)nj6!Q5? zATNdXv??zfLc;_;OKRoE%=@Q4oa+PkM!`=@g2G`xlBQe)b}OS6le(QA-3<#-J1?m3 z4%v_OdF&B@{`KJ}kKw~E#&hJY@GbR(#~4)F_pQA*XAt_*H;VB-!IGgd-;&b{-?j;_ z@}xrUDPr>~d*b2ZqdzVT_hyQKcZA+4&2{bUls%MU6@=cj*!pV6ZpEC)zkVrXRI7%8 zg~!z86~k)trDOWxsDQN_4b{1-C2G0r^J?CJB+nP^?oAv}D7tNZ<5VRQu4+$uaOO*03`vEJg$*_*}rC zmFEO8+5dvnpoz#W1Ml)5#d$)*m{`qJ8Tfe1P}V7!+mLe0*umxT_Y4=Y@gg(K7g&A) zX)Myb!daT`cROs;Td_+Vx}sw4`!rpQ5UHf@Q!4HS01FV$`PrBDS6RB*X|KZZA)l;k z--x@%?5~-)eMC`_RhEHl1}+Y4Wzl5+EtN)ZX6H-40WPpsiQ}huVPqWiG%5Y@#v`>@ zMX=V=xoT4a8Qo(tuL2#F^~FFd3YzWcS%^%VfAhi;pxHCOqUf%6o{K2ajwsz(yAFz9 zBRam~t5S{Nh)Nd+6PTSt<_V_kidWbV zu+^%C$T~9aNEboVf56ISRy6JY!G8E?fQI44n4n~8v?`NxYhXtm^JWr9uaHa!*4p_T zDe)3|ZRHXv?5jy*#;D*DkBrtv>JU__Kcaz8dFlGPsB;u0!P$2hL*9I;(;zZM5T7Ej z<6F}&pjVQ@AZ3ggEPa-m^KLxo`PS2lpy;4gTDde zB~<+f(jB)!yzPv%GOfU%)}`W!-)XO3sJnN9On=H?kP2is=8%p7B+KBP(oT2x4a}`k z&e?b$U{;KdC6Nk{?~&HRii<+^AA9rd!lm;6?HHwWZa*?I*pmVS^I#edi&AJYSFRK6W_muhCneY)p(3YmJ&!oWX$Fk*?h;3J^d z>($|QoXO6^(SPtO3_($n&1_Z@vKFD3YtWQF{w4+v#nH-(HS|;7UnvlHiExiH$IRy1 zn+r^rTH5gj4{8H062%klJttlrwmbxX%M`k+inlsyf9l*<$L6trzHb-*s%>q;(->*Q zsmRooXLb!*9>Spd87PJZDHivkpg8NigKJO}yvYJ8SUfO)rY-%J9($CTszSOO5|4>= zQvIB&8AOh`?>`Jt>8D{Jc(Yl_&7bVN4*@s2GfH3WGf*yx#*Cd=1ZkOS-qJb_Z9$wP zWyIWd6SwYwc-?#94_r3H%&Eg++RfSw*=MdY46e0YVh37H#>5kSx|!}e?sn@*Ax zV%fyOVkM`@fqrYT4(y!-h;aO?5;?6cb+TAkOS(F_cti4m)}nmUW5AzfJW@zY^u#q0 zyPKH9W_9=%UKU1@I6%yohNzlhk;6?uq;oTZ<~wvZn!ZTvVupz8@U|!v^Livwh?a%v z8fx1NGkY@sG+uyRFcmwF(R1_A+UbX^ zyf>xn9{4SRRSDcUc}^%cq-Myu+ktnC;t1?NuoN0gCoZ=CCehP@8*iTQ7 z3J;np!UhQbLFs2Z>k6eghddIS*mJ3)`f!Xf@ulxd&uS87s=+@QHe83$?3g^WMbi;;iy+1^>H7Bx2a z_e~GM--O8z0X?<)p?fYe*|&9N{80#|3WGdYyRrP5x==ntfIn>sSHU~J9AHkVYy#xz z^Pd*H9aS5i%sm(&&$>4|^JEuxz$)T#1&RrJX1}Z;iOB6GJYI}ngX@?d!b2v9AP+bn zg3y6KpHG=!^n)LC7#83L`vH}J2yb(`tA$pn2@@T1zRT?vxV6u?)wkV5&)oy?D8&Q0 z<6Q4HzOoSOcxVZX2wO4-6#;GJnn$pf{@fcoCEi*!)P>`@+fE5=9baml7AZ-0k|VXm z{xSC8PwK{r;~w685%$ThI_f|CC%5Z$DX_;ng+W@aA2|kO*N~z~-4b98%3)jF5Iw6r|3ybz4l(Y&GfxRlbiUIalBP#h#1#ekeNmPk_Y*m)PkOCZxOfn8>l9KbwdzjCGE=A) zz32oju6bG0&62IuVuTIF-%K z-qHG>*ypnk+f0HTjvff=f2)iO=1sb(?xRn4+XZMzYA1H}&t`K(JGA*}fj9YdG-r^Q zWHv$MLz{bd2{oUR z1sIY@JEf1@lm~aHTNfM9u0h)A$=c0@i&W&X?^)7NRho08lIOPBWdbA_EXJ=#Mhl_y z!GQV!V@Q%DpD`Y;hk4>lJ{JEsPv&`LxvVizU8szqkjve*+*O`2f8S-E#qj z{QV%Y3E2j1 zmLQ>yqr#seLqM}#A2mq?i49?b%>Z?g!Qy^*$xncom4`Q=Ej!G!2+I}$a0Nq1aw2@A z9i^~^l9ChJt0h28id0LY*X=9628<_REv8uJUog6JkPmfsjN>_Fv@RQ+Lb~~k)Pmx% z<0IC*0v(d1&B>L@688S|gag(O!4I@XM&L{oC2{}IwHaYK04Lu*YW)tKHmCzD1e1(f zHJ)Qb*b$cBq19b^H(vmPmdCNGb*?|YVb=#R4vZm8-<_+l<1j_LY9LHb#3PuC>@uri z-xFCF+c$7}{_^fa(30XKGItJ!@GyJ-1i0(R$h#)efSYZDtW7$AB0fQfEF{heo4;-7 zMhb~B`Q%Ksf$I&NeFnG&&G*KL6*o&PQsNb#KOJkyQ9RJ|kZaGtTPq@=53oPfI~d!U zT~&{RU+JcZlEKU1>183)7r<+pLcI^4V~gSGcim-^!lsMj5xEneCTfX^&wcliZ^QQL zELO+?n&!ZWg-2sAo0eCfrt~LNwFn(L{iD~{#aquGd-{;p-ofVVpErK{0={vG92^@2 z@h62P@G8eIkT#tk_=-aVEi|MlOnVaWNDLY{vm%I?QIP)XbH~X1IUAo5k^+?KlyW%` z8O#5=&=X4Xdr0zM22y%&efre)GIC1ko{Trcn#fD!Ptkv~t>2-I#`Xzb9jYOd(0;SE zk(a}tOPxGAha3mB2riPaxbi2MID-P270+FLz@JAVpdV7oXW&CVj+$mqn(@|Jy!Rq~ zg%*I_Q)spdgp9lcTGRWLs%z3(Of~#Ky)y-?Q0IMXUXz(C(DH9-xp!!2+OJXq84mX0 z*hfmcbALJdW{H!~dA5z&9dn%Y7e0&6u>@1kL)!pxhal09kxAYJLs+O03oS$TLFq4;A@$-&+kDwrB8q2ac~TwIiG8BG2hJ z!G0u8A8NzBGr?0Oxgn2CY>7t9YM}&M$ByXL55CKQSXc397N*Uule?`FN!{#e;fX*` z9=kfz=J(9Iw9My=-&5nuk4X~-9-;x1!KcbeJtzqBdg0)|Y0HPE|E`Jr{e;hOf)s4{ zLIfb~xjb|o&wEvF;~`3s*0k4ci-TM-!rK9l^dvryGj z;%~V74@4{dn??7n&q2cwy}^7|k(PfC4sNw|75WXxKjL93(f#iQd>_6Qj_GAclB-!b zpy3!?)*!_>vGkL+AA4cK5vXb(^AI>@w$X;!xl7rEqA*#8%~XN!P3z=Y;M)`{C`s-r zDGDwM_{%Uk>qSat0dHs+!U;5mEUncJD(=~2YbhbEed3q;g`4}xO?7z{vhAtTMM15I zo7o?vpwVIIN~$uN$d84< z!J~l65ENHfS4B6n8N~Gken3P0CmU6bU$1|_cI4T<`Q_pmboVf@eFX^`?ChKx7R(Rx zU{GX#vKH(B_!Hwo3HGtPROdvS0GT`Chez%x0Z7NX7;MM3vg&1+@ZPD*2?k)PTcSW)F}Z( z_k8PulF>RsX$73)IgAL)qJk81gD&jFF2M9Shec3}T?@cVija>wRU28te112nFSVI; z4r>VAn|ZINBaxWnWwPLi$fftoy8?ahtPxM17$DTxwRTT^kwjd-28~<#HwQPl_Uykw zXtnbbRR&RL9TNeaS!j#+(H1ygZWJF*!5_02|Lz8>`_>&6L~d))K6~J#1>i5q&bwDi zQKDhOO*EA9iBc1%#6pT;Pm80YblOpO-`XM+yenK+R43Z1TGC)=L>HoSiC%Hu{`BOT z+vOht;sl62XK57?-FwL2NDJQmaK(PjWW3hgq~@95SZ^hqSEVtRoLRc0HkybXFC2t% z8e&lO*RCs6(X>OFZNW?tkmzg(BL!C%F9EuTMJf#+^WFJGuHT(0ZF0zZVJR@petmp# zl(Y@cSY3 zNOGm9AW(l(2fRBzNs6K=Y$XRj@&EoZJU82EdrbD{Y1Of(ldD4&)}5W;KQ?FrgicTc zZ*p8-WdgZ4R>9F2NgH7P;$hQ+7I5ZWavR38p8!!?ubw?ss^q0P-gEf{EFo7ApmXG$ zPDh$4;6Ecxin{@{m5F45r_=kcZnb`b) zlMpna%a$RuR%Jk0stT=D#y7WyQDUL&kCI`CrQOxXVli*5q z7!i$=h{qsfn}W$Lv{oN@2bcNUqCY;EBls%;dHe9`gTINWoo>^^g#^^j_*MLZ#9@;n z4I$FAX2Vmu8-Jk3d)SnJUfO@W2V;(~_jqGhS7zYdp}PK%Q6m{^(bq(&;FB-cDrl{o za8rDw%`zXbXydC3q_Ct#Du6i96jYi;BOR(R3q5&%Wz&eU#NFK&^Vbn4z+;D5f2)!w zhRF2+ubW|fE$OwTm4VYRO$s&}$`msI_fI4_Q)8z@u-3a~4O~YBysyG8i@5k>C(r5< zLAv)@Ze<+Ywb!)^HRgUEO*xM75u3|v$SEG@hD%jUoA(k&0hnWY#HC_*+ifH1k55;- z8ldOV<=93fP*Pm|$BvO!aX$U>9s{FWHf|bls7oO_pXrms==af0KjOi=dgYTJd1VHd zPyY7!zm8@2XhWzeX8PZfCKiZq|ii(!|$m zeNUUokSKDSsozDeSwQ=a2^~wT!#T2ugO;UAhAMiWjLv%Rf*gs|;g9ICDc~YsRDpU? z6?)KZc&csJqY#g{uRU>exT)t6vD1C)01n>K2Uhkkn&`{|b*Qt{U#p91@kxL+TII^z z8%jL7GP%jN>Gy^qt8c*$@hB<{*Xekyd#|v2q8QZ6e0$Lmqd)k5Boi<)C%EQ2o+G1; zm;}unp7@iAi!a=mqaIJJ3tW#5K2YjA87#v&U?iIoP>3fIsU`3!n^2p(2+Ot>{_(XV z(#+^m-@U|nN9w_4LrSf2JN_k9Br=`08`H6HadwoVq3vvB>u&<~p|vpN7N9$Q7Y|Me>m#W_*`NCg5QIb?I? z6mpCH0LVpE-&lDx1MtPhd5Rsuhu=Cj+0@arvmOW}lF^CG3XH${8qv^csif|?lGPUxGrl+TRCopU~&wx$t$#{D}HhIpT9m)P-Bi`+I4#WqFI=;PhBYC{I7h$XY>{RRPjUx#Ij}1=eVsLyvS*uT&=Z!;D8g3eDl#R zYNFfj9+Mq=LBxOm#pqf%^qae7tAJPkM9J@2m=a|M@Quo2VRZ_7Hx-C{rjN(XIrDML z`Ha$yi};Lz*7XsqVxnreg+I)BAg%PL!XAO;86I0cDXvCu)ZX*DmZ=|t&Fg=s>OivZ zzzHdV#=-!6UjH#j5+WCTCz9(! zqMud^EwE(OD%|5BOFX~!m;NRG?hK7q7A>GIc6zDj_89oF&4<2q0FfDiL3UW08P4%d z#29^jVT@NBhM%b7%`iD4S{AC|Z}PsWIAjuga{Oa|&d~vc);BNl{8o+pFiCWJg>HUM zpIEWw19iN$S((TaJ0C}U6PWP!dq9lv;X~g6u6+&}5%ihy+x%B|*9{&shG)P2vFSRf z^ei)Jogv5nqe=9nT1kc&rxZ}MN+X3PHM|%h1ynAxNxbdde|1-+cy6{@mHr%AE!%XX z?`KQLdW8EU^>BY!ie;=De=FY6^~%_xN#k8Ub_4v9ccj-a3%uZOk#xLg13&Ech{xIM zoN@3`-9q(xXsslq{O)6yo<%Q`y~g;{N2?62%*I8&@*~;pY?&|*SJiOw(L1aakXO>_ zc6qGH>GfB_t7?m+e3F|2s~W%c!Ir?rFdo5#BLQYv0|Y-{a!bWk*6Kdm$-&-wO8 z@Lzc;q6%y5Pq^%5fK!6iq`6~*)ReT`A~g^;ulr=?E@Q)l8>_nQv)don?eqkdV`mGL z)`u5bf{!1(>sfoP2OGD(EcN8?tA_y-o(Zwk1y1y>a2jLPH zbTDqd3GTJA>slS_$p#rVdg5dE8)HkaidfhbM8q5h%5p9}KCav&Ze%&Gi6jJhCBS6I z`SltOAd|OD?w>zb-eY)5prO(gg~eu8u=Cbr>B95At{9?a#twCKHeYwY7d#iyfjx{; z=xC35G#q`0uzMyHP%jTZ-=!~)d9%(zlO)79!IE|Sr^(SV4NE`A&*S-Q(sU+=eLOk; zUYo?Bf10j?1*CgR{omx24_ZKgL6g4z}#T3Zoo07XE$zuQAX5^AVlU-%~Ih5SYHbnqP#&~<~m6hPJ75yh-(5R)!@ z;B4-B&bH^AY<^ICs*`Tx0qdS-8y~4`O4BrXTi;y(F;Bgw)ar|o`%hNnMyB#*D%|l{ zQ)S?yhmFV8HHF-41`(gyTRJ-vy7zDM0qQ3#|GnMX&r`BYFsKK@(tatb2Z8~u0^N_j zbrf&-1Byc6AC~jkIw_B#;Rik_$jDDVpw_fnH5?4YtnCZhT?x3G@Em7ZRoP>hG^g~# zhu=@mMARoshW)9)?T;mCQb9txKFRFQZ z2d(J~lxSXMbx!dVsXPyjVsTOZ41DvSQFLdXC|rG+U7RlhaDicOhG^;$*&|I9>L_Vj zcUkR4;cNHo% z3V>m3UAWR$%6LL<`^UZeM78kIqFR}z*VC#inPLjW-?ynJx~X&g->aLf_fOD2d0ViG zcXnG@{#lg`hN(wV=jYgX$YrMtHmK}d>i`E83VY8NFy6m`eX`9A-q|wNYEtqLJ__zh z^toxTM@ZTaHygIHM|h@t8n5>oEn33n&yk@pp%h)h?9)s-}Thu$t~DaDD3KhOjfhK>+)z`N9=(>qKUc8fke?VGHrA@^ESQb^e>G^wLn~a4NouiwexkfH_~NtPD{s z)y+iM{eiOtD}^)3pI?IVGQ*Xu#W?daUmkNFQaUZV?i9kfd(fwS%_l2>PqQWTE{xao z)BITmU3pKVjG|Amb388e?Pa}=Mk#b!HZ;ob0`Ny;bmHI2nG(o#G}xCkSK=vX7CZcw zuRi3-wyKRDslChS81CvWbq$mznYTz=c~Rj1b!31+0#xy>F1j)}rME`UTA)f09}Na%3RBC>gDOGDsR{onmXNLz8ir+<$K z{vn+5pVjji{%|f3f4229Kj9V9|0$J4#@ zvW(f*XhF={QxS3G_i0i4V|dxMW{$Gamv}6^GZFvN;z&eaz9Qy^4G(YKEI|B~LgGuf z1o&PHWZ&=%PV(ZVbo8D*kGA!12EsB?ZR^PaVP5OGA4s?SMH5F8=5#pS=`Z4r_k{uw z?$qQ1V2JvFbrt9>f;);w=rA7tplDBbfl{61(HfnSsx!#kYgvJ8hQ06$*LcQFjaEg{ zW&oEl9hPB_ZUfkz&>8KIwX7V?~d-oFXBJ=w}^ znSn(7xbasty)pF6nr74mzJ$J~n@rDjL0P)p8aN5#Rk=XYoNmEnUkG{Vov}<4liLi) z^-3PAD#S~WZhYhgIqkJNS)1q9Wf>7?v=eK2OxF7t+jEh=S5+6s&>D^CQ>yzH$<4E) z9Jk4KMgR4jzJj*s6SYETwy3ii3Q0N0E95ksWgsai0e_GQ=l@9LCAlse!j6G{NgpV~ z=i7hW4T-fkSF)raatGlT$p-=-f+ki$j|M+~Af^_<`MyZpAgNI&y`?3!jKP}_*$gpF zC`bnFFvQrE?D2s|6PRtYFUN%jsujrIwD&jXgqBnptW?H&x99*#kioaF_FC~km=#q(rn;HHO9GDNA z3deP1DVrBK8s_UFN$X9k8=Kk`b8xPDa%ZNjWb^oXBN%Z~r zi0Nkt197A#ey@ZQRk(f5XyW>2ih|H+I@x+Capu4p#%BfsHZ@f}$1>qJZ-0y--=u(_ zQp$(E{X-Z)U6Qj*E98c7jE}UEh|jy2KppW+TvN56Bn0`rmSRT>#;o$2-$2u`b^t7nHBD?`%z7FJOA+R z4+|ncIc0;VT7v%#3Q3l6E6cnAB(wi) zxK_hm+$=bWum1gbHrsj-4Cs2_YPeobT@kg%Kt`)Mxqc~lyxw0};kr_t>N4Nd9k%jB znasCSkcR7<#QP0u*&PgEC@9R9`uKGwYhHvG{oA7YyA!)tSp z?i1Z$7{CYI9#p$aRIN!#c>fs`^p*e_jFsAQty-5Z!YzNl>rHL{7Wl|+8oVIfG~~bnAv;0IY&(*X@okS< zKz0Ow{tYK_Sca%`0-IWa=L<2;0;RC67iJMB-llJ~Gd{T`x@Kt1lNuh5h|9w@$bmMJ zj17;A%g$?j4i;|e{8%EO}=0HBh8X=PkZq8 zY_w70dy-UYfv$l1)hvKB9@q4t$4#-4Sb$Etdf*VOHuURVdFdR@y;91Ey!=6Dxry$J z1&|+!_h2Xc46|dlgpu8+ZN;}&K07^r%?;Enm^Jo#$nyZ}?LDc%iM>W{WLH_H9eaxR z&Jh33Ub!y4h*uOdRtUIgX##ADgpYrsov)UMi!6W2H~mbg;aIGi1v-(QV*Bv_6t)r< zL5tCc$qFinmrE~ zGVq0#18d*dZt(+?ylPb0x{xU;DIQv&)1B6DpC%r+y!C@{M5%kvueGZNNAL}`dz|F3 z5%4U*->CA3Kdil#tDvuwgk2XA$36`Lyu%N}Kn6 z`~NZaol#9Z@!Bd#lO~{aBGOfe6a|S8kt!fX1VjZ?M4EsU6Oa-CY0|r>5I_(V1?eR~ zq&FcHk(Nj&)TNt1%HI6nbKXz)%l(o~PEK}zGxN+-GMNn{Di1VmMt6Yht-D?|Y&I`J zBD51u?=f+(ID$XKMPE4Vt!a8dVmZ5S$Kbqd`AqCva5VlG?P?zjDAzA~o_;Gk>5M9? z&aTstQu@vcC{O3H6ZVeRTU<(zy#F`kp8c~2e1UF7&pF$qcV}-E)S8W?WO+xr2G)H| zjQG7Lxbv8ezrcXpiPW7)TqEz$_ zQAi|qF;D=m5PLo5mKxh$IR!mCAM))zN_-|m^2I_{dGFUFIM*5aT;$N=Y3gXqO8a>k zxS$YxDJfXyU{HoVu;k8{wDY4{DZ7{`T}UsJ_!EcP2yW=a+~-+dmq23NLM)Z&y(Rv%`ds z;D+aW)6gLQ`^ARb4K26ZYe8<@Ggi-+e>=eh==o27+#eAD4|~01R5)sack~^W>!=~v zth^k%NxLqSC_NXi?B~rM1{uac_wF9*ZESGr;rJ|sA%Il0({xAAj|HvNFYl@_Wn$xC%)_Xc zS6u8zEL(Lx0=}Q5 zn;U;o-13z||KV2WRxY{K650{{Abo|G(f`rlmulXSGXm|?55>2;YN zSo3iWw793Bxb+(KZ_0I@dX}IJRsz#4wfl@yr?9&8ZBDONH7zI_p2yl2pAG}@2L~Ux zzW%n@L4%Whcf`QSpyw`)ONW)tnp(4))X(|@*tZ?aY{;#+6Rl;tn2KeE08&PV2Gm0DS>{%Gr@jsTYGB6!Uf(&upCh}xH}?OSzimE1 z*xJ_o2{&onWjrQ@p1qxWN=_Kus7SsTx5Fb&d*zlwMYo1rlAMF|FW6+@^t#?Zwpjvo z&oWrS@Ok?6mGt+hP2Z98MJV_ojrHM}^Obh$#W0e@4mjCcJT9|>y}G^na%p~xThN); zPTV=oOnbI+lzXpLMPo#r}y^ANX%xH{EhS#FA;_iUXwB9ZC?kh)+^0mbyZUZ^82 z^LW8NYVh)ncxx)L3}`_=SP9E;f)eN8r@&&+OLVoT-;_S(N*i}$xhWNfN)=kX&P2!Q zXm^FSQrB;co|nQ_jcIO!4Z)29RP;mIxKm$V+0ztrib})|8f9}T3=CdU?=qN$*NJJw zncn*w>DZz;1Qvf8K`Fvp;skGkF1Vm3-b-7V@XsFs<3C_S`oKCHcxdaunu#Qa+N`HP z1kYUBUA=YkW84rb1Y!zai5N~?!A2=hou-W+JM)VbM3;Tx2rr#Vm zDjikO*b2JHei9$X^0-v1^6r%v|6V;aLM`P5Ap@XrI&hNy<+BuY@4;W;E)a%5BEK$p z9YP)VG|wX)y4I@=8{qO$O&*rO1Lc>74cJfDvM4gwfn6DD$s1jc=}2E}V~%FCnHqaH zVHUN^5;Fc&3~9x*38+YP`KZ6^_s`f4?mmUkF1G;2S5LTINW^H-VkQ2K+N;aYWLYd} zs~fe=omJ!Om>bi<{5{8q$r|P&cS-PKvb;U3xqQ*IQgYQf(%}r&sU)o@Ts zclJ9@9l-AqrzPN1XB7#Vub;k_Hgl~4Go9?P;nQ7HFM_Oo&u3;g#_8Dh_<>0ZcX{32 z;!8atz=v*gEVV=$+yt5~<|tcOH1MoQw4grqMzV60+%Au*)sgDzmT zO6tBO?Mn=8`e5GfQ%AIybHiY|^LBL_Q27_KC}&4}iBiEXGX4Qt%zxJ#;z8ksR!?%&^vBDZI5WKs004kfSZwfV7z!hSX9Vd$DUrad{14Hl&nKCJo; zkC5Y+!XF+4cZrOWR1P0A4smtf)UU0+n`Q*gPI^Nw+fPF=X8t-SO z;pItL+8sgBqD%Dr3>hPh!pkA}Yw`n7Jry?DdY0^#yZ@!jC@TXSCK2#LlNvZUYN)Z+uaGIMV%|Dq#e)N=Mv! zsU-=h;6)(0?p_i71t@VKpytn$KATQ7i_r@;-5T~}x5H}i(JL}wy;R~%;+>w5@F_x# zYoB`NmkppRUc>nl@iCG_!BOpQk~Ay?2pMhuZTUHP>T=Tw>ZqOgmP|LCTFO97coL~B zv~m3vT}r?xpe2LndrdMWp3NG-hePa6`zNLLhTQCi>krwZzyDeF$PZ$phY@nx;0Fn5 z@Q~tda}M2DFCk>pAiPG@mlz~;nw{W;YhH{b*dIJ=Vr>P0XyM>jjKj31R>tk}A$z%P z&6`LXtL?1JC(PAXUpo+3&hHG>xLn?t0ny}cT67ux|q;qNQ*w&u~5l`;?ES}IFpsoiTp=d3*-n$F=>Dt-OpXrkE5 z(a=U#Xb7X%tcB&7p`<{!R(Q%oAVU43KTm{8SX~i*-2LxDC}t)lZ>qoENy?H-sSgbE z1>MRhlmdKt<~UHMyCv8qXHuHf7PYMn_A(=jobjh`F_(UE zlPBTUGyGO1o<7{2BxJdz{QeKY|W*PUZRS@J!0v zEI=j*C;jS)9?amq-tU`$jg933w+zV+_$5cOL@_}ZwV@9!H#)_fVnvWEZ-|eULPHx3 z(_yfiep+DicsQnOl;F_08c^G_jBaaZ5N|DE4 z(v5rg!2Gd0-%Xr-BY3_o4c@Oz^f9br^=Nf4 z+i>8V5{g0L&m+siSp|v~wIh$-% zkE&kX8TblTJsL&bOob6|2P}Vhu63oty>bn_Cph6kfYHwS`A9UfhH_N|eIeM>>H+Yc z8^qfrnbI--USD=2v@W4zwP=S2d{~ePGwiBGYyEyob=u^Jy}*|{?ryj_e*_QO8?q1N zVWO`N;7&ShRB3t%d5dI)CUNSv@z@semOjB(KMImd9{UBBou}J?ME9X z#KXQ-i(pcQv<&!mR>uNV76(-ixl-#RC>4sa_+l9a>ZfjTY)>B_^5+LjUaHj-jN_K5 z5m{LBo9>X-Gxc+?LKv_YIFO|&d zepU;3VIjin&!dlQ{>k?0k?9S57=>>|qK-*toz63y8}5tVT>U;G%x3?y?&ws@Rg@sJnEfx_!t|Hr+L zLTuMEOBG!nvY&Z&W4Smeow#Pd0AcG=qbv7h=t)GPM!+itOg4XLkOK1`yGvDu-{-#1 zK(?+znLFX>tl&?e&d0^|$`RneMQz{0KtQ!DRw){K^Wn(?QG&8o=VQ!HCQCf(zVO%! z*I0r$7&Kw3XJc7awV%*hL(_5_5$a+(v#=o*jj8@}dG#R8tcy|j!%)^hm$X>AIYpWJ zdI+=&qU2G1A!tJR^>d2ItFBMXQYV`4wPUyVs$mY;@!~5ST`yy>K3+ z?1ws$w^;jK9`nDw2@>V#o#CVQV>i4m9KrR1oQ%4~pPyC?HwUQR9>OLnQD>JfjFqN} zFa34xke*aE%x%baI5~BkP{Z{pMwuX$XKUbi~%ygvmLW1g|e+3<0f zrUsqcACE_wI&GHp{Me{ZnbhOAY-A@$DxOPWuwuSJ-0lh8qrVfX2rk zjgmtUNK`DL{1+AovXWMNhkAL@T#0SfA0MVu^+W6X&>mC;{%^WD9+8>|dy!Ws#U5=< ze*R-f@hiADyfFRLjp0j@PmP|+dEq*w-4yTy*)22vI10*$6V#J4YAa=!nbtnR*Qoj4 zeL1m2Z8?SuBt;wf7TvhI$$(qi44GKo1#ey8mYv~jTz?QoLw|pVI=C--Tj>`TG5D-H zqQ2NXT6t<`lA!$JEGEk|DN>C-*|BcQLU<`sUd8M`{!F=WY0!_aH-e6HX&8_*PsK_k36_SD67{{H3MW}yYesqu{k;k8Hvh-9=Yp0!jE2> zX!Z|*zxM};lWQ%9Skx1RoZ-Udg={RxelJ zQZXEZk5+%5yy@V(kX!4oHfB3oz);<>6H})f*FE)un>-r)!qXn+Vg780-Nrnn*EEQ% z*7p-Ujiw2N^U*esX=qbXM)=?#oL?<72AcFfo}FqYe2UGgZJzt#ya{@-4Wov!DfPu{ zyJww9sd7Lb|2FQewr`Xw{nm>A3xcx@$w%NLYEyGD3x?qg$IDT$V5P-;$Sl~z#|-+P z<>|0Hkchjtzqn<1+g)=$eN%HI^d(-$TzMKprb{l^8JY%Xd|QOTKXJ$>KO$Aeo0NC! zX28%!`!d&U(^EYRXgyD7As)rEH1y4tvX=UOwospRf=Q7<$b+;8!}gGee~(^LSEF*{ z4qgN9=G=e5Fg$Trz3EmQp&y!~sAkfe6ESKJPy6K6dCVhV&3cTtMq=iHw)FzU;Cvex zdv3n(W`7RuAG;;6SJc!0C(*a#S@)IC?SEp=NLiJ}pFiR=F~Q43e0-)qVvp0f)oq`f zBnc;nsJ|U#V z8+Uh1!O6^aIV%D*KodJ2A|K`D~toS+1UcgFl6cS^4^T}HtO@fjB#7_x^hoW zd#~xG3c{9aN;ebuYj)n>BDbC?4U)W#suVugM=j-nY$3nYM|8J*l#m7Wp$6$<94v)72Uci#erSxGAUx@bKmmu+VTES&dvsNogjc#|5D6(yS#4Lln$~NOL z2bW7bjyqwXXAIzpakrY5*yw?}mVG9!$hy~ILwEL34xT2m z^W5COEaP`A7PvWP(Y>1t#@rM5bX}Tmu8e@Lt{Ma14&f>wA6<0;&EqvfC5|_A-2jQ} zEs?*ll|R9Tg_-!LdcemXBZGNv2kQ^Xv*^`_MHlGiq-G{a?GWIc3@hh|rwsC8U=ccR zCkQe!skx}znc~GU^-PtkhglhwbWM!l?jGv8o^#G|DkNzhZrjp~y27~7g@+%F@vYrf zFMhkxt8^yi4?=uqmOtE32}f-#W}C^9aoDn1);EXakuucR|3YI9$1h+B1h&=9H&?;= z_R^?XGzmz`zch=MYk4}sMBSsb{?o=}Wfm;|w;ZRL?-)bJDQK@(jy12f8@YmED?RVA_asv!>OxJkb;JQgzNMp`jsx@$&&dR!l~hl@t{>F z!!UCewE;!Fd3xt3ntz!-H*W;1LaqYVI4q}TblLD6&H>T!*`|S$&eJaOL;XMH(T!Vz z6Q`QfiY-^Y)9vC6ZJ0-hS3M_^U= zw$Xvku`nT{-+~vRo3;P#zw#O2MPBQeh|Wy@g%weyS6E(J9Q;6Uf5+@gbi1W|a%nL3 zj_=7X`(4Fc%^zJptM?D#2JfQPTb_D8JO}t5Jl#gy?sO_jB6HvCO)*9a4q%X)XGTDF za|3)`O?TJ5-Ltdc8v)XOjamyrK(ELQQExB^i3cUV|_hkaS^61;rD zlO2Iw>R0jYqOLcKx{I7#c9cDsdX)51K|*X&b*eki9lYVsJMg?C#5+l9O}qdf_~zpp zy*|P@Pw2`~!Z&lsoOqHMr24*m>Z$LoJ+8d;qzu{JQfVr>cTeaU1pkvj54Ky^xY2Xd;LUtY!J|3(HNL( zXH3aNy|*G#TT-h)4255+64;Cj@~zIDCSUa%y;xxY|7iLZv?{u-rBWQ#gK7~iip*ac z)Pp>6?t=f>o~Ugne1-z&9Km8Xn(cS>U|(YNdR#`%1ns^81a!I5Witgts@$xl5R|Tn zk#XWyJxv@VBP=u2iv#*N!2<9C3#gX5dG0ds`nkkDO7nq2b-fAn2wwcJTBdx0_9wMA zDUouuM0`-H6Q$jVWgk{D#dH9jI+$nto4TN3gZg=3H~MtMrNRYn7D$Dq;x;OAD>+C} z2-!>SHkR0aSWSSUoj7}_2q^6i!!k>=Y3y`s)#zS$&8Tlx@1!zn3bPJOS1-&xxxusy-Dk$%=EpCCG-2f zE3HYqqc2X39Fo3s86Vg|nvFPuR@$NwZexgmR0 z9@msu8PxEvTfZg|4SqbTBiQ$D0{hhUOI<>*=*~lHBa;OCHfrhC+2;mrsKifVRsIO1 zEE!(z;FgS{zs&K;3Os}tRb%=$K_k5S0{Vq-)(BwMpVK`nMA03W03hafZAUTNvM!?LYcyPOC$TBM(1B^ySn2f?-Zqf1~~>CY|4kvTTn}_uh0{G`NN-R3yR7;af1nCGJWKYE4JK-HvupdW3xWSI|(KM%nssrmjJCf6H18G+f1!SH>>d9wCcfZi}sM zh!?KG`44+!$S&Ch#tL-vI;q$f@qc{GIj4M!P-tq(wU&Ea)3N+S?a;Pj(EX}@@AnP> z2>h7S0Yswgwe42upssj?S8FhTu!wMBk!VLd${dkIP|YZ)#mg@lR5(n~`s zBR>B*MxSF%--d4EASvmuk8p68sxd5O70bd}0*XCjcIJ?(=C`62DP5mA09}3HCJ;HdjRgxege-q^}Fo z*`-$IeqcN(vA(`x1VWGMfDuDlvJKvYelp+BgPnaqnaxsJyn5MjFo5Yds!#sZ*;cqq zU089`F|No6dWaX7SOzogNsdv`P(YOXC#CvJ2R(IGQw|nDz3P4roS)2x@ zw(j}?68S;f?th4|2^}AkPtU(e|*}$;=A*D@LrYS%los@;#-p_MFz;=N~fHD1)sXvTK8BGjwiP8IoT0Obp#-wI)1;SpvXlhvKF3H*jS?=}(+@LyQ^iU{@+ zQ{oeV8BOE^dJ||K0r10G)LKmSjcG#AS0x31TkQe9T8y3k1!5rPLjiB?MDw=W)$-M- z)oFsj82%@MB#VB=K22!4lfY?gj^yS^2AW#fPp_2aOE%5F1{yW)IFnw|SXQ+m8uI#! zgMRDL7MG{Qld$}&c4awlPWPE?p*a~ z+`~b>_`CsjBm)k9U?`u20==_=cEMAGPs|V!D7eY(TNL;hIW85|qyb!{`mIVQz4K)z z6FuVDsH2|*7SIoiC{In1f5=3MqRY(mLl6DzU;ef^UiVT3MS5~5w4UjA6NHc*i)DY8 zG2jt(ulkP2=LJZm8zW7fCm(|%koJL{ryI=UIQ^4MR|i8^wHabO_^m$!KM5tsIW_um zY&a-wc;2QyizW1x^!jc*Cb!Q5!M=KL1#d#VHRb_-sii5VV zXaLB0kID+lnTEX?dzE4sWhA6lVa=kE0Ci+;z;y=?D^80GkLFyVyA=%zt#*gBupp4d zHFNa=;k>qBIpZHwDIJOv!gnUi8CTG~Vp5{7*lJ&P$f!8Hcngy5? z#KFZ_pzpulQp(1NGe#0!`rsQY)$jQDG0?OG$FPE=Lz&Bj3}`08P)a3`G*#zTkV>Jp zGaXHdk4e6ZCfI)<#sJM@K4|El)j>V3n=y}&q=^rGeD@Z$b$_bu)xcsJne&1c?*PFU zF*Dqj$EU@El{Z3RNsng5Z^~}tfJpZnaX{br4FLaAEyBI!v%r)%N`JV}Yju-5VzC~a zlX~>U7}IX{K#{taRFKI8axN0=?IZl#-y@8tMP38g$LRQ>PP$5G{T#Xre%)G(U`VI5U^p#RNCO>3^3jdL*>&}UAh)GvhAaW?v# z2U+1K;QVAl3_|#%qZM%jDP9$-Az$%?Ql;tQvQNbS_Jj?BcGr01@qMmPbJO z#rCJP@-Ug~`Q(}{g$8rMhXg+V@q$IK%{0n6y0R7_x%@QMDjwey6etXuu1`OXjHo8q zV|m{*Q&l%OmHr@*i=ltyNQbl_i5Gl}+M^c>tiNuC_6}fdo7&*IM;q+%DDsm?)Dv{9 zB6~>tbcS@QkFdY8zr}-EbXSn`S0^?I2?)D zakM3ufxKF)NvL2(fPD7FLyZuzqSazBht7e(`yLt2p)*1d)#t6|(8v*E>!=t~mP`rA zOY3q>I>JSM>Py6b1!Ew-5-Vi_-=@WNs{J?^j3>@o>K^)v7F|Bc#lT_|n&En2x1E@| zFf&>-O=bK9cLPo{7`CW=@}mGeg3Xtx+g>Wv;!hL|?W-VE$PoRB9hy-*Am_DCrdB}x zz%_A``|scHi0E#2q*`d2OI{gxH?B~i5C-458X;B)D_{9_F>#_<&3QH+(d~HgF#Ur6 zxV~rbW@7CraZQ57qIOmx-KxTTfWvNausHa(ksM_5dB_MTIndi+BE-U0o$N~ENcOBqg66~Ddagb_nJgjvNAA5r)7jx(E zj4sWEj3W^p09`)fdr893?GiBlo6f$l-iRxb%B#nIty8@I)=F7TEP30^t@TSDMtfyH zBsvqe#}%f_MfLlnR8FavbVF(=WXvcRiY8vW%b>sf2jeIa-mti=W7MjdBMwhwx4M7{ zr?Al&s;g7WM1VsvamZe`x6?awqB-5G;A|kn9*@3wz7>uN2r5r9SG-9p=54J=6MGLl zF1|8wqBrCQQn7=KO@$L8#HN5wvzm(RIDFu_;k&`6D-w@ewuTNMEt1cdZ<2%zswD)= zb!c)QAHL*NBz=M13o%}R!0bE!m90qdi|+ml)$Eu7^ild|2;^j6j%pDsEu~})E{E7n zjWmxXKp%Qmb%`fecmhtHT7MHuH){{d1@bYk`Tp^5|vlSj$ z9sH#1Cr!vc{=}BkKEIuV@-ONBA1fbL8YlF1RVFMgK-{pz-=z!wc(Ikq*PKl}f2C|G z=Rww=FVEU|4Eu9(($&&TXVGsD@>s_YAaujm^D|em{C}U?%}?!Bhd5{zdzi6cVD*lT zFKNY++1u?t3F1y7F|h4>2qe|k6(d;iB1*bO5?NNV9Ij6~hqSUV9Je{(TO9NWMPOrK zKO7VjP#bFOJS*9_hyqFE)h2I&^Qx>E$37wCPbus^Y|`G`wA9gXl6D9A>zpEKN9*=& z5}Oo1&&ot@ssw4qu!l3r_1%gr10_;Yz@tHk!}om7Y89zEDwVX_*HLzU3RQTQOW6oR9mC*o(%kq)EeGLK zFGaqv{XgnWmo3~3T|XwsK|hQgpERn#sD_(C7OF%RAjOp!sqpu<-QxE|AI!iXAzg+7 zo?Di_n$*=|ef-Dgc~CyNnHFpCvsIq~>cs?AHnPi%32q^@kls#;aHh0(e;nRnCHHT5 zW_9ZYKy&6Pt(E8flTw-C-Ug1EK<_P>r1r;|SFd!Txa$XnNG~D%FCc9S2=6m2hV4b2 zy?HAc2Kp&E;ltRq?i#N}&xoWrll@bC!Dkn2O6#OfzSk8;KGJesMeVGHT-AO#N9JBQ zEmctHVr%XfT1E2IQg*Ze8-9N%aWeYn!@79yx678@fB$KsQgpT`!13yNl?lS?wTu%~ zRr&axq%tu6F0H)t-y+@R?P)yztPZf3kz;nxmuOSf37n)s3r>XW1ElVHe<7W}tV}im)gl?>KfrhT>99Sa5wwnUWdy4N@PHoi!Ysw##o=egRmUc> z`4wmgxnz_9b-e+fghJSZ($vpE6AIj?7|-`Qgx|#JGb~m2;iy*6TpB}Ug;fn#B!XXo zLJ^q7J!A^ABztBxbG```{Gpf{9N+#v9LnzNV z{gpD#JI8{Kz8>LrHuYlZ2q;yd70&q)KEMM0az`$|J`f)P=Q{(Xc-iK2Ih0R?c{KOr zBq^VZE4a0Sx*`my`9#Zk=g=q z?8RMaH%wNWF@%S@=haH;l_M#5C$FJ0AdS_{3ALs55pP+svtbDZfH>k?jeJCt>EoyLGLD-$=+WfT ze;xBB7N6B9QJ&Auyq*yEm{hQnw@BO?q$Wjv{1Kj{vHh4ePZOLZ37n0$K=@ALo zlcv{0j4~%U?t#6Yc$t$yN|(q z-*4TF>gXCJj>qnK_s(%&Mo>JS{ zc35wyRRsCmS!30U2@(m(Yt$(L1u(z1?%|q$1YLXcByAh@G5zN>Y^6+neEf90+UKyW zL9xr}ft5kzYfkI_2724l}~k?I)H=4FkNWXMPkhr^OlzUr8gq{*VCs3%XMPsp?tSf>Pke;hV@SuINti zQ1_w6sk1zBemPPij@N_92UE)x5Ki}`Tls5nb=~#UL%(B<9!Vt+{$WEii`I9unFiy7wpeEJvjF;i|_<;JU2wC#rEq745W&ihhACIa(Y37GKI@uVXBv_!>Pl8bpt}we0K2 z@W;aQO1Bk}%k4p?49YOK@|pULKt3JRqp7l!U7qtcRIZ{FgSdUNN-7w0;+*k&1KCCG zrMSB)mHSxJJGAnmcE|JYUzju`-D23*`31ne%d!M~E++X;R1-+uNs>w<*mEPGZ$aJG zTTUSe=#J8>u{L=AJgW>9O`T|d`VCy6k55k9M)t5AaA+f_HstnIlrnGQm~^z zyRN>w+y(|_67T=>g>=Cc-#<%)4f&XVvQBq?P0wmzrV=IN^D^)ahie?-&kL6ZOPOZT zqV69zHXSOnD96BU&M7%6diVmLsw&vPXa3P@7M`@Md*R6OyZQiIK%~DP$YUu2*!YWy zo!;<^$<7!3uPMm662ng&MOnX)ID~s@Ba&+=o&!Sc$TTa-#1dg z>H`VG4F}mMG8vxH*AOO4y?5vhj?&e7lfMJ~Y>XykKR$ubsGxi|*w2sf2TR=&$;E}w=lDqe zF)PgXw@Sc=7tFkPj8&eH+Q@LpEFW{~>lfuKD0}t=bqH6(eW$e`hTRZnpX{(-)&^ov zbc?^S=Stl0QlVGa`DDdPjs>jFjZ#Fc!E7s(5Eyv}jD32JF^#I${w-W#xG-HqyAM?z z;u}0xzUV4hF72A)*RkziGGY8@o=P)evgXJ7QcoeEIhi}|Z|LI#qY(0j19yyqZDxJ( zDdn&9v?opWO}_8eP&*J=ay0m30^gf6n#}t60DH5S>zO@myg(}%#v15ycW(1c=b1T> zd05)v)lZa?NT*5CGFV-ivX7k)WXU)N2I{D`8i{}cQr@sfsq_9fcWKyV~Amz%$3zy_Xw4LSX1kB5%|Lo<2Os#7YF&@8m5@6 z!E^rKEmB|E4SxTgPOwj~3VzjI*%2L;m%bT%^=i^{f6efV${UAVMh0Y@nXhv@I9axB z9)W^h>p4aB+YJ9W8jL{ZMmy%S4S>qwB1t^YEU0EFZ*^efcP*(dcilPwnV8f-j^p6( zzzZ&znX@O%sE6=)^|czFZP&>C#gsFvSwQ%pwObci6*e>^Is3=+Jt9Gw8F&$ zR&_v+<*&J9u5b1I8XFGm-F52Qyy8o&cQ3Swg#zmD-^gYqVx&pQ54%Dj>1PO)n4p9I z{iQes)e4yHpecH7fY}&IsdV&RhkXkT9t$(73z~za(QSnmAiCWld;iQtz>R146SVO# zYa5!2z$e*lwA@)s(V9-wpa3#k25|e=t3JY??uQrH z$0nxB4uA?n@pth{ij|wz@vVQq&D3z@NdL`&I;<}V_QMmIB&tye@jb_|KbXs{zf^NIf3P3 zpKlSBUv+y0uA}%!i-M%vwclsqIF%J<$4`gw;KdX4)!o&*BR*NW5`AA$TX}fXDP1u4 zg+OV_3ifT<)zHk^C#SJ`S)HOX1Yeq zC#$J>y*}ci@V0qV|MnKj@<$?!tsuSk=~udKq{GE;^l`ZX&;5Dm5Ng2JV$H211fyH> zaRiI=zH0Y{a2mP2D0WP54z7A5cn}L_oWNif;F&vhn6&zcv#LgWsAuxw^<^NH`M;l3 z%L$zx_yuE#yi5=)7&v7417HTF`EnLL7)}%nx*JvR4aZOOR*KhHs8LeXn30pW6gS4Jm9d?Z@DBb>zGRSe!lco~iHnHT>Rga|0 z8SbemB-T-R4sPq}{R!f7fo9un?Jnj?^8Lt~T3@{9+>^mcH-UH0waJb!3dFOTR{Vma`&Q#GOQee&f8Dr2P zCW{kG7bN4RyLLvHXyXyqCX{uo?+6xK#9~Z@>w9Dj z+j#CY9rx~RC<4o<@yxo)i_H26=qq9g3T2WYv z_!K#dcA_SBLR~4XX~6QRr!lB4??)@X_n-@Y)fyD0=IQ)|Z=t6CoYeqefg&5B>KtI; zi!dilmVgiI2sQ~i6mUPXz-t&K!w~a1OXov}3<9AI;a8c2;q)t|+a7;AqY>n}Twk2} zc?3J7@8idx0jXRDKQ2psDFek4WK;=&(yiVPDpQ1P4p5^BjiNGNa}}eaKBqg!@3+4- z7XXQ((Lqb$nN+Nq*iEqXIL{^v=+Z}XmUmM=a@^W=uNI}1n+LKel7j6mR%c)qI7j$^ z07>oJ&)9Z&h@IW)j3UieL1wkl8Z5Qf@6Zx{K&<_!|9NG`uoW1z>-fIE)dwgg(cybn z--8ayW&%52Q+i<=fQ0F)kohOXI2r;?4uEh$XS2=HSf49E{gI~&9 z(+~_gy?{2QqOY7ZD_vAJK5RVJAOzk+AYZafUINk0lA%YWfZr$tG+t&a0aZScxcMS= z8g|rpdv!*2nudU8`8qm{op-pO7_*a9iHV@EKR)T^RRnTi#;fAsE+1@aOmreqj2_)| zcp?ige!iy!l{rHi_`ymhW|1;@B@oEsmPP&h8=+~3#}(+fu?yX`KY=o)X0{8J7ng4G zScB}oF8{y=56i6iKEUgq0R|J+H*l4emi3RJbyD(6UGfXT$2&pc_GyZ9KUjPjcg`7E zT$WE15w$d!wr+-E}0{0u#l%P=X z>S8khRuJs761|)+*1NAHcyRLOqo8zWc*}U)PdG?>e>{6-W+thCK}wz9h5s3W>l%>$ z*C_XwA3mG`&jFNzNYomU52R|0>3Hykp}yXyl}q-ODS4BLKQBly0vOBgjTnkqQz1*A zDE(SP{|L5x)`uD7vIfNs}6X633g)Zha;?ko7qb&tAm^(z#CO;eR7 zc{dPPL*E~i&2}bskO-A9+IJ*c>&qu0uqFXwE7+D5=gttv*&3~K1h)9Z`pkkol=TP3 z>3>MN@_4Ae|DBY5X(2nwzSM+JSti-3?2&CMAt{QAvCIe|OO`AV#+ndQQL;|R9%D_& zK4UB~V+>}-%)P()e1Cu4`+A-Iea`zi&vRe*eGWrQ`|{Fj{H)YKvtD*ch&pl0Wp4I! zOS2a|q=&rq{|R8VIx!*}aym2tpLCNlccUZK))W5oq(x}JX}P)Ac#nG@OClnUjJ!1@ z&j?G%&1Z*k5#ciC{i zyO%IV-cpsJiGiWT#N++(TG!2fRp1Nc8GQP2ZkckJy=RU51A5nKE<9)2Bx%iW?p|M+ zB%d=_jOUESRro-~&SSDeE-yBo5FfLkR_Z#=-vwB8pDcu5DQxgBeH|Xx3~z`IR$o5 z)X$b(rQg4KH*CMZ0`GAdSY$ILm+GQtKDB6-13M=e-!A~iPrhYkrdc9-bp&=%Sz$jI zx8`#7hRLGb-*Rig{~6BrD!Ik$S`1%?Z}M9tYy9w+erW;l`{A!y1^xPVT=)%xr_n;hTlSlwvN&rIXTXG!Lf+jV2FTa#=`p6wU42rQc_A1}JT z4&|HfxO30m_SRe^mETUo)1ixux8%5LMRpIw`Tb*9kr_89L=segh}}>|XID-+Eugxe zSEcmGW=_tTJbLnZ0d*C(q|2(b>$T2hQdYhG?;k^^$}%mV2aok>ky7!t@@G>oM=Zr? zSbHqUDCM4iKb$F8G6!T3DA}CzX|{%DCpJ6yU*USlU&nT3^!Tv}$lCZ1{MQ@wOZo<8 znqAs1tZcUKg#3VZwMz^4n=3ay`7)g8JkV@CiA<30>`F!^7tzs^!hNX)EulVX`VGq} zl*7xX>hE$L$q#9;OQniYL91$-L+WO+|maa+k!~3WtniobuX(XG0+y4 z|2_&CeODAbQVWtQj)1XmBZDKZ}$!MN!F9FvmKBCWg=`@Y@ zw!-!s)!Y*CAT0f{i!$65x`B_!&u%T@w&MKPC-=w@@66Yhg-?dcUhQ8}5fD_a?J3-q z;alRw3FJTS)jgY;ddKA5V{9ijJu+id>pQ1Z!;H91>bBT9mh(gm%5@GuCv=c5?r?w( zY2>%!!7;V@!S-K*oUC#fqK>r}6f8l1eYrEH@Xg5k$0I79f=R2E=N*xG0Y@I4HL~=+ zUbe|7(aqrMt&I()*= z2&a)s{4~?usy99`T;Ke6GSapXK$)x988{G`MOS&X<#` z_CE`bkX}xW^s2Gy{mcSMpC|U|4 zuafn_S84w?pt#4+mc;6-a*Rgw_VYE?%Ax9R`p5y1apac3er$#_v+C?hHc1mGUT>@d z&qp*JW}Lj?iNX(bn-Gr{!iz0EBtql!ee_izuim$uV7^S9rS2y(-x^e=hb;5c38N~6 z8>Y0=Zb#@aw3txOX+j&Z+qB#8QZu)*%vX-Wxx5i_trTTs#>epqx&PUsP4G7jN4QSk z)lwDo^DpW|tFUtge{0sEN9|AR{ryEH1ifT83I9^u;&UzZeS%OobnlDJS>fF-oXaTRT>w5GDQ-arlon6&z6^8=a%w~V`Ej>QDN)=Amv>5LJ^ zI}r=cIZPr86`FXXZ_7)5Tdp~Hp`ML_M3JXkRT}TW8>-lE5LpJkFL_gvf{Akh9=Vc# z_3{W>&|!?Iy~Vr55v0+v2@DXNYJr$rqG5H949xqAT00Nird3x&xwP>qxQd;yC4W?w z;cy~^TvI2W`o>!K5l48At|2{^B|7}BtwYNrLIQ8i9c|v?Rfpf)mC;nF>SZlxaW)m>ArB2tXzS?8Ai3vX6V0P4qo2EuNVa5)`T@cU&t@6&_Gt8ner2V3jVb7$cz!b+w1q_zV|bz;yL#^6n|vCegP7Ht1! zHY8rBtyWpaU+U@V!L3BHMxwbo+d%U%YqG<2?M7R0D!+F|R3563w7n>lfqHEmHuGUe z(FNGTo~9ii$Y}`8ANtf%{XWl~*3hEnbE}OnpyU8HbBY@{-iHnMmiemv0#{*QVG+a< zEelt~k$f34Gb!&888RocLj)&6dxx_z*3(n>OTV;K4=)A-_;<dhklg%V z3RNAZ*eI>2M2ej`*Xog5$le+XIwiCEz6J?9Dx^cW%J2)M(cpVe49r? z29>Ry9Pq5`hR$dP0WJCT@Gk1Q*Hy+@xfyN2vO&Fc)Z_V6douGpL<hvB2gq z!y)Yn?|=7DlwVhT!_5d@3Z5g#(k?+|YNrjrXnfMwUO@xFKEYerA!cbyAVpKez^}p- zp2eUp%D}HQflkp}erifLCbtiRvGCe$KC1(v)0SsY9m@D@hRlDbE&gU#)m`vm`it0* zWl}kx0^hZ!gX@dovk2PNN-py7!1HD4+l!O3v{!OVW8rQVh4W{{my~FhSC|!@n(llI zt&+Sz4LJ&iN90Vj@Qq#BF8wW!>4lEjEiJD>TL9(~_C6Zid1+J~+EfhG5l(uB{UqZT zo{pw-9(-zcp3Ly|f7$zx7M}5M6NB^0%GpEL6Gt;-g36-F^qPXDfLzkQNpKWO<#quy zue4NAp#eMI@dovK-?=nXhL`TNJK`^ugxe~n(`}R+tt6TNhxfTDmk~$%H~2U=WAMcM zrp82luEh5p!U_MNLmyrK?uxPhWd9VQ>jZB6;D|FxKK2!#r2805T1<3l%On-+o@AhI zM6`8IRqwZZ^SgijsqxN=&zG_DzV&%F-A4KVrY~X(=C+?C@b8u%e`H7>`s6^^{W^#sz~0zb9S&=&oJ>LbKrcuGX5gt!99u&ig?_=(fHQFY%K3MUIMo?Afu zHp-+()6|Lm;ztH@x?)a!`!XUbaYoS4WXggH{e5`L2=vc2Qk*X904pt59`K!|#>e|C zv>eYMOf#Z>lDC2pS2XU@c%3aD!F|1v@*3*AVLIJp+GkZ+~cD{+Y2@ zYAeqGtSoW$Kkr@vMI$bn(#E0Ws1QWDzSgn+Gc-FU+7Do$+)bpfkgtvNix`tzqF&gV z#+-&1BEHq7*n_e70W%i;fCGEaUg;kKd-4?ZCXHEco1o6}CDCk5j~Lg4W<8!baYW*J zVx-UVzvD5Yr*-0Rgz*6PWIFcU?=v-LVbsFyL?g0(;71nkik16JLM``3j2T`30m3UD zbNLO-o2BD8Ktg5e`R1QwXUNbrIp86#Mght~#QTRBV|X>~8?4c|Q84o4A~P8Yg9>%n z&3(P1?&ARVk8(85cO#SYt_n93y>ikmo#l*ArzADY!HFzdqavAr<@#BJ4@v&eTd9}b>VFg}+omQiN^eE{>joclS& zL|VTjA(vQtBJE`FvVt4;DnZ;Z z$kU4~bUG>bU_ehVt8;62`Kd7ceDtaaNVnTq~a)=GMbi!&Yf{daCZiQS_&f88^JugFV``$lQ zyz$}kD_nf4`_v*TV*#q0QewFPKBaw8Z)6#^>7xRo$z%a_gg$p`>6Pd{lt61Nd3ttN z#>@ns#wgoC7Dk%nuX~0KUEp!9H`RKmtMv2F;Bbk_#jrJcYC$*tA-sWVDdEQg5`j@_ z#hX7jH(_^0c2)g#zy~l^<1*s6*&V?jWPCVhqXPN%1FlD=xc0wc zc4cxXpDYCrP;A*`DT zu_Cm9%Gty{F!CfJJ>9Vm+`FhD8At61YDIjCxQz*p4O&#birjQlWT&4&we3@N+Q}hzI{&)1xo6`~TjNfWY>vUdRD$&m*%0q<*F(})ppyc{CqG5Nao!5#69e<- zlAqtEk#A|uwZ?)wM~R*bpu6XEyIHs>=w2JTYAtQ9PK?{RxB$i6Rep4{mWI9L0F z!2?(H?NiIaM*~pwGCAll=4qNf8c_@nkjL_X51*y2fWOidDwL>Dg}&JtU)oFvD(1N? zd4V*2Y;+ej`QM!l3?uv@CVG?n<@!rtBg3egny0afJd2ySpDY9Bmb^fELPW22zCdoh zN|8TFt(d#=-%V>;*1Z2lc4bQD8vJP;hfJtgPM-86nPk=;qtDkH3wCnZ?PKP=uINZJ z2(B%5yr3^nr}P@KRk_(Y@N>kXQv0L@&cRW=@i$8%Mu`+BrRnbz#A16ZKE2bAkteel z1jjvA=dZP>hP>Z}jo=1r0JURaoBkD3>gGZ9gSK-5BQ(`sX2e7qkA~QR`v2+kah7&|&ZLabGAfei~ab zsQp|HT2zNA1v&+LdywaZEc-sSG^6kMq42YGMa47NzpN>Fp}5NHdQH~; z)DnC-Pz-)LI|R&sBN2n2&3>#&Bc0nt)oXlvI?!Asd6sy!rnf>+J_N6$PGnGsOx(|B z64@a&)SoYb$-f!otFy211u$H@LSx;`mlo?uWY@P-zlR^W%$09o{;^Wy@C7KKbjs8*)av)^1I-jVb~)%oEB`>VcEnFxY6TI>4*QWZn~nFl3^v9K^#%-(hp^;H zh40I(W`R!>*kG3M7Zbw*yw;yr1_=yL>~a*ivpb;w+69=C$TGW3Zq}ALdyIDBANEw) zw#`7U&_%LJYBr?C>8-dr>~6s8B?%U$|Jr4^v|V_&K1?;=z?gstgL{u33S?u#voBki zOOyEeEy3C|C2 z{X4t$F+C@Mgq%jCMdFN%8on^j7E!TPSF(=)*fCTGdCSvbZJyWuZ++VVw1EPP%<9TF z@sw6ER;0j&PJ6VGl*(n@&Nya9#Xt` z^oHa)o0XC4K3t{^S3tG=k)gbtLS03J+e*%(i6VSe(*z4%&D2uB!MxbyfBh&sc^ zwigL#TyGLQ72K0o1&GLN2 z4#6S@dwCeBS>FD%{a@{nM?K#}PB%0RguPowZfi)Va~AlXCO1bc793Qa$EnRWorjBo zE_8Gw!)?7c=H--`XIofKwG($Y*eyuxxhxFyeE_HDy7bhYq@tUiDKt?h1FXT_nlq|b z!DkT;J_CA~-T@uT_it|xnA}d_lkZwIP{@5;*2?>`U|l6$kj6k3GF&eGn$;v95(xIa zEXK||Tw>@<_x+7Dt=0)$eDKnt#ED5TtOh5H@0h=ec#z8uObx%XcWBcQ1wRsIt?4j- z8r#{>2k_2wPiYi{TssEnB)*M#igS4IIz4B0KxIJhB{IwvfI^|ShuN=JWFgW#?{`Oc+FkK*P%Mk){Z#T?t=QJ-vLV93Z zuQ7k6kd1mkPzVFRMIHG6gnh69lfUN-|NN`wOqSMq(pQX?+C*HxFy=0tOTh@%vZrvY z@h40nZnWhT&`7G#tYVEDggZ);Y_3FJ-|S1i(2l;iuOWB822$A3V@p!=`KghaX3=hx zo_QG2_3mhU&kM}ozV&P?ZIg>b~x`5n~=Oq6OX(6Q{j5KSEB; z-oMKHN&u-Ev2n(*7XLd!xxKmrs7-SPW4qs8?B{B8-22o z`<$7o81~C7s|bQv@{m67Lg^Y5n~s7_Z>n`$c)x^wHk3eK`qJ1AVUC{`WS2GJX$z{% zUF(EutI-bz3Yh#<-x#^uWt#K36cMXqyS+t_fWwi4UvAjp#E&Anj5w^n{yzJUN%18u;8PTw?FUJ(N)i@Y+ zicViihWP;~sigr1bn&yk>^8T9(v4{I+%%sBXtjZ7AK+60?Y|9U5`35PN zjpYNLVlh!jMC9jkWZCUg$Sn8+$psY9{4QODT5-*w@Zt|BXT|>-aL&R)3XcL)KbJFK zwyj*3=1lK@W4&Ew%RJf*4R=l3bN2i3E~g4?2@UnB-PPVHxd7NPc-o3K&=oS%HI1!Ipp z4K6`^cSV9W(IpsmdR|6pg22+qOY$18T&Y3ZQG4h-m7qIbqN6JsX_d}R`?v-9lf};8`D`_585Ma%(|ytT zPap5%{XPz_tIL0TzKL7GSkmKD>IAFgxdB*+{;`oaqp;I~ZdB_Hx*+U-e*k>8yOPiW z@s0`=m)U4~Qoj@SVt6_c_8>EA$cWDSEJ8I@mn!Vl*~`CUBURZMbxg#XSs^`goYO-v z8q?n}e7GG}Zg;Wqb8E<)=AHVboXb*;&Q{d$AHjE=$E_8Ff0Sz3~Hd8FjfIR^C>?Bis_77BY`>7)O4LuV6lVrDK-(Ux3^a zW2pT@gFdC}`E-AU#>iys-OUlYc$;hrGWm4sQcm6sVOm8rFx!Mj;B)e^ACOtkjQdwk zAw9-=Aoyq9^RSoDql~d8A=Kr^&@2h~B1EL{!gF;w6ZUh~Re-~{8gGG9x?)r%i)}T{ z9c$Wy8nmn~(UUt7Hd4Cl9&v3Q%FwVFe+@y9 zfa=TMBhkNE@^SR7kXsRYQH=5?rpXvq=Ou8YgOuYYYUabgTSI6Qe#MNElzd$QFiQ{1#~Z5MS<I+7QnV3(YBI8$x^hnup<)_p3X`<*e8@sGui;P-5J@fF7POPv zMC)Mz*Y?A>?Qgk(*J>zFjUcZ@n?UbJGQ%RoMXv}*1P`D(!<(+#Ot;ayFT)~J(6m+d zi1m(bspp9H=41+BnuPK^QGe1uIOHYr9P0CX?scfgh8l8KonfzelMQs2r4K>v)1Yy5 zn+7wJ$?eM0l$LvS@!1=VW2NMy4-GV)$io&cZFHG|)rYY;K8?Rcb}CF@4#`~Psgp{O zU5ymNx7efM5kOEx$~ZFdz~`<2H6D^p7euvFAJV^9fsRm`|7$9NKZ1{*$Cmxf>%Zhv zks4?-FHLhEGOB7<+r;h)=Le9_JIz5{Q|OgY{Yo?*?FNxg!|+grIlk@^a>y0OD&4=! zfaVcatfix(Zonc2kuPgO(y!d;>X)$oKOE@ajugyuiTXs?S?b1?N7OSn^3k}HclJk_ z5ROZsU#C?WT9o8nDHam7xlpttp2+n8+Y$mXW71f{&ZNrD0vz+>> zE$;Y|KlKk`!P#X+&=uCH8`}_o2<}2J%=-K3?4qW*58e9!KJ>a0_vjiOc8k(Q%KIV@GZMukhEFhpQ=I)r zCV#awv@GsSJ_;;-WCaUeiLyWKz3A+I&NyJvc^vm9o1|8Y?X3iYgK~@y2DpKAi}w1W zw^Luac2SKA>!Pt1?dRk;XYN-f3Nb=f#W)qN!XkkEV?oBS2p^=7=@r%^P`z<%w@VQ7 z(Msinii>`R`w5jlpD)0F!nQ7Bi)4nuFg0IfmX1uK);p3lZ#tovL9b35Jj$q>I^8e< zx}L2%UI4i=&ST9%={hO8M~TUS&|wl{a(CLUz}{Sp7EF}AAOHk~XD>L6f=e%bVj-~G z;|h)N5s~;e=B;;y$~K_S6MMqT$VVU8tZ0AWvW+8I&vPk^ts|K)oWfX2`s=^?u_{nV zi1yHLHAzH#T=YIC6ODdw>@3zEywg%l(Pj#2&Gnvx-A;;O5fI1XKL-TfG&xT^?>~=# z3QIQTGdqx5>5cEPb!c!cf1FJcj`xpBiS}B=i&w!y3`52+ zTvp2bp!$!>sdBJ^9}VU` zKK-7~>JdwAJ$bTt9jY&IJC!_{@*Lg#P!)Lz5bAg$3XbpRVFI6Bo(_azweI_Ju!<({ z%5WeTz%os$|MeAq3T9>?F(Cf)OEAU_d9UmeyP{J9Kj9D zP+7o^IDJ{h349gbcCi37KCUu6$;VeNWDWbmLuBu#pF+kO$Q)aQJcLqc*KYwftmLpvNeFZGf>_s9R3+z>2dg?^;?wIi({8^7FHP|9{+ckc|EkYVC z52}lRlR^cL+b*;Q24b2_jPBa zvvm2JA#hPI24htt8YEahlI+?UkSz+|{vNuR(sXT4J0|}LAO#Txui4(F61#O>X zKX1TJ`yBq9O!sn4r^*zqP4trQ!XY>D*pEi{L}=f0FQli1K*Tlluco zX10$|>}Ci2UG6NKn++$6hcphog;Aaz2GbgUzleCj_CB48PQe&n>N?MJ%2(b&p$=2a zOS38+@@a>5wnIF2qklDYZfq7iODF@W%*`{4K9PvpnW*P&O%{7LdRg}0WPAmt;w!q+ zF8x)c9roNARf#F7wCmh>x}Evvnm3}b3#Qu&n(7olZrhfWqbpAz{x>ASIr#*fx;jryBO}B`W%F9gnm8 zL*ixF=>)}X6({o1lFzOFkts&I19Wm$cHf>S^4}nR%gKPY;oD>KJ(mW{c|AbR8k%?*D}UD~v&1{ZZ5F(5Tkrqp)|)thDG{TPuaU5wI>}h!uPE&sU4$IJ zZmU~EO05FNLsS5pY&ur6TSs^VTxWkMN55RoLP z1#i-%HoACR`!@kCK`e2Wa&)B)7|?x+|1t0@y2pJHBO;u3lB)OY!KwC!q=^swbXZp} z{{h32vx!N*5YcfnT5OO~0`J!{Vld8&=&@jvGnvjh<))K()M=e?LM7 zjNi6txPFn;xTW-`lA zt{K1h=;A;o2}yOR>vLof0%r6kkE1*n!9RX7-{2|86R&P8Z0SaUoqs;a#`d#B#l?;- z?mZupv#j|ED%Nx}^X@K2+OcUvpFUvVspt<@x7f$8jKH>DeYtf26V5us%|x>sQ`sz; zEKBi%=cD&%)z?;DlmR;?>VC-C)|y^5Cb~cPSsTu7WYVGyw~IWt`E0A64y$b3L>dTN zs(b(`)Xx{nGfhE%5%dgv6M1x0gc$JpsLKxO6_1y8-#;{Oq(*1UF!V6A!V zYJu?%Qnb)(lMSFX%jP<>A5<$pi=|foyLVb1P?dqk<=i}|KA+LPq|QLOEy#F)ew}G1 zWCctDg;w2imN#WJ>EI;))miUDROm}l&!R^Pbn%#~ift=7+F@STG>Ki*Ks?p^O+-2S zL?nG_BJ!uo4F*0xGtnv+XI-e2_u{A-6`Pz8H+`Ruee?%^k^&4P~1o}vvN%@Y)Dt?H9!57i$B zu)fKU0y<$YC0$_;l>z~>0*$RQe1hQ{^JPf}133R|`MZO1)=m#fqF@qS7mJP}v$joN zxlVt#|yiE{cIx8z|FR9bf2Qqbb6LSkto5#Z!S26E1`B|#1 ziOZc0q;VR?4zfQuf%GuDem~X^&>Bg!AS2iqJ3Q>NT}WVeEPIeiR=7?=A(uE~^AC@R z%v2h9Dp!JhI8z<@U~uCm{VoYPm*tR)Y&+p4zX;VE>rLq{S^&rGPkH8FdiAyJJ_XvecCUM4 zRXPb3(;Fifx0fSF*UHc*O|}MQyf{XO*=guMS%6F;XZTn~qm$YDqmkE6L~d&v#@763 z{jHxw7c3vwU0yqB9muf`6_DK#h<7ib-@>OM0Q}*fuJM z)|y+0v|ynr;VrtgsPXJR!&jFM%Ml=*<3MKFg{x8-PT9}wmw&*#JGrD;B-NSs{Gs%S zzJ7QI6i&%;pjnZe#lp+lurTjsWeG}GG_uy68&$RoTb75p?P3tYz9Yy-1wt2LDxAnh z_6O-hAHnIIU&~iu4AVeq^L@-~HwIzH`^$Qgpvs$pd>Xb17?XS4R`y#&#_k}(hY_5` zw#2uZ1S!C=#Ms)!RCWN15y@wxbP=|ebKQ&8)}h49%EuXmed*iZ*?{0Kfreiq#{g_O zr!W{?IVysBurDTfByu~Ez)rAUk9c7{Xt|6s@eGdzcWi{88P6_2_GP4%xPTxTf-LO7 zhJp$B;PAkZ4pTjFCh*3_4Sx?>MH+lawkQ7q?P-T=vlqbng%>}N0Szwv@(hARayqM` zp$qoAsKB|iu%vD5Fh&@J!<{8&>tBEj?Nm7?dRt$BgstbyTIL}i^=$3i7=sNXTL;V` z6pZTc>NOPFpj*0@3I*{`9GQd;)XYa<@~*Z31Mhi#+1rf@JxN%J0o8$^++A1|h@awP zPF~}bVh4Ra*}Elrox_$qq`7E^WpvbyeF9f8Iv2F+)-glpqq70$9CWB838kB|^5h2y zuRFGUii9HiJ7!-ap~4?JnkWCFqpe$Mh;Kze=S=gE^K_aVFg4!)s|sw=&`hOGmmopr z6SqT*v9lR}hcSZmvpGOe#q<>UAG$g7zp$p{59?zXSg2hJJM#d|%W4Uk(X8?CICGNd ztzt=mEw)gufOC(;Q0+&+)sba+~^^jX!}SLoKlt1+d$^Q7CCMOjQHLDT#vrboZveQsO*_S zW*)813r=cX0DWULiyHZv%`yPoR+h$CpJ$S6Alf%rJwi}@pCs%#-b9EFCkq1*d+nzXCRdwbQ88<6r2oYPU2I z2BAnoKMP^hcJH$|hYB;i?}qUKnV%aZd}4v0S&AGj<14@kG_i1|@LF4h0-{bH7KAt{ z@zov(@NKr=U0)fZlp-fMaaG%KjP~GME9v*xWpr8 zAV^pc(avrJ0uu|zrW5Jdx5)qESUQtF9PRF()F}pCUE}A~sRJqQ6K?`ny>M!qD^H4$ z&`*gytS#$<@+i&9betZU-RodTQH zqz(x8mVsfvY=NM5E6}uk81qgQ*0<;c1PQP}o)Pu5ZS+2Z=N1Dgy(A`2J8T@v0dI?3 z_H5{YQU)qKY(X*G+Z*{Su$wISI0QgbwIqKXB)@m}uB5a`D|+Kul};d)QDL`b!5?hv)2IQXNZqQvNM#k9}{xX0$s0$gvel#h#S9 zAWh3~7*Ji7?92JWlJ&umpmg5bS(kcFrqc=Emtv$nr7ns$VTZJOM-vHOg$7i#tHyAK*ql>bsGgPK8auu)*Eq9 z-(Qn^CE&zuQS{&o8kVyT4-LU>MjFszF#e?%!=QMiDSxIe5PXGsaNh7{Bf11|NE#8K zzyjy;2bE4_p)WMZLq#Zl*fuSMGr<~&9U z088ej8nx$34T&KF4>ScMQfK=w>7Wr1K@3lH1&*7;PT%+;j(aQIvN5ZipaR)^jWG9~ zi;=e>lv_+!Ql$l1A|&BA?^i_!-k{7RemejaHv&q?Re`0DW* z(c{Vah<$!}I7jvYrdjheej}$;*^@3s>>$O*$B)mNYpr6>kd8xcZ0MULJ*c!Ohr_gOZcGoY1b7_)W^>cG|V!^MgoiWcqrU5GkVa05TN zK8Jv2(J}xZtXyc`%fa3k)+nf7K zWtJ>z9OA5coz`^pB2naAq#6|dMPac8I(Ry;X?{SpOMAen@V0Lfk$y5kZXvwIE&oFj zmRLT+;eGCnE~dFDZD)O_%}6DCen2?OBJGk5p}M_+{DvSAGv{lrQt}zWe0}VhGB#`8 zqZ`3_RLSOe8z*W%5&@Zzxu8kw!(a)_bFxBqf?+7oBx5D~p}MgvkrBD`E5dnm-`@S! zcMFVHaMCpdIbx33D|nw)S2h2n5uI}4$G&LP4?osimE9X4mcO_>q^O3;vFoH>JP>Di zo@2DnE5Q5XOO3cgmGwcA`etnK0CLh~G`>?J|+ zK+%LA@LwW2%LVIvms|=FF_Av@^1YKO#<+g!5>5rZ!WoO`$@1oMABRUjSB=kdAz`ge zh-{lbh~8=0X!=^go5hd@$fej*?ZmBWqvYocA^R&HpBwkX+gn#AiLd##518rm*?d&a zGaG+b)^L`@f7ythVdDr_V~pGrC~;>sM5y&o=R7ed+JPlGFR8FyaNd4nqae;{A`kC( zJyyFrbYQH!t2lHpuXQG|9gb{77M`A08#?#_mHjsTXF?2&Rc%eZqo0{FpTynf-}_S2 z5Yz0!ksyaQBEG>D-Csc2Oud6FQ&03pi&D${V+&d(}(o|wA)j!t8e6VO8HQx7LfQ1Ave zUvTo}_K;%m`fpnV#LVWGO5fOs;l)OLIH`AXS#P0eY`nwOgT#tmq!Ubv}F(Yt(8DGy~tOrFqa zfJc{6hN=LE=$21)6Nr%KRvmn!5zcU0U~WYzU8-*E^yQC-+IqNST%!ceB0~5R?Co4n zv{ifdnG&^p>Ql}Ek$(vx&e4;Ffu0UwsG2vuvGajgY;pg1m`se~ouK0RZD%J<-$PbE zh#_wgJi`a>?&IBN;-Gpx10iOFikc%d27cz?o{xcPyv;g$wUBI~SMY;=7J=a=3)wI1 zOU$&W`!RycOm*Ow?~Nfa9OcnJ{;4tCCkn43G*Y(jl~EZQ%e%p5KqHTh}FAKBtuZu~H6 z{6hj4*YUCjuRt!l#HT^R$BxQRsC4=%SQaj@pMJo4cX5Gc} zL@dhrdjMwMSi<8yP)e&+c!u#>$P)N^y>JNfETCk5#A(;HZUd@2LA-y-vaPOfOzfwy z`Hq2AM|S6d5PqLLl^U$#U3(oPlIwjkasztU?ONgMk6ds(w%F1vW>0n7$2;SQ}X@%ojW!N1Y~kG zBu5v}2qUod&+5J-vc*dc(NkYKHkxfjnmz6tiI{9VOpN@m$_w+Q7f2u3Cxn1X+TZ;u zGu)uV5PttI#n5Cpj(9wId}Hsxi5edv$ou-k`}1Ka%S+of5zh-3ECTF)LI)mj0oyB5 z<2)(R|NWGr{T^WNU3y3{bo?dJW`6OXWQ6l~`vfG%rO!^MWLLQ7q~CMlDB-yjU12Yk z@q-BEN!$kzmsO$O@EHY<=8u~9P{!)FcfNCcUi@1x1b^;5tv2P*t#>JN1jwA~1~gvo z4q+v{Hx7LOyJIn!L{y!;&$1`imGLlw)uRt%em7cEuW-ZK;vban-3kC7sm5am;>qCuW9&PFn&`r|MZl;av4K(~6a}R!pmZYA6qKMy@6rVU ziPS(=Kq*QQr3eBMk={W>YA8yF(7SYq)X+i+$?liuo%#N~Ki-)$vngkj-E&{}Rg#@q zBR0GuCc}M-;oC!+@bTO-KYGZ|XLrOX29Q-eqk1s!)bse9=5PjNr(v5Fr0z*DsGEWs zdFyIR{TYkJZ%-|@VV>nUd(&W#Pe5-P1>cu9AyICtOWFIku#%l8&pw(8`g+x!Mns4h zza`^BIuF0gfQOt1kBi1s@ip$WU6nC9T(}FM9}%xx^Y?QPQzloORsylqH~(`!ZA%*U ztCAs9vI_4WUd1^sar$qo$g68#NT+rF`!n;#iuUpt}WiB+xT< zUOFERs06BWXoLWiE!M0p)Rocqo(rcsav#j&JbwQ<8U|*#9#W;JA~HS21i`9mEwB0V zLtf9Q7NqAR(aY4GODw!xPzqC4o6$8eY5%gzyJS+|eKse{Is)R|HBV_j)P+t*GTvWw zK9QS%NxsfUlw7(}`}KA28KQEa@wpA@F3ie6$2!p``=DsLcI}lk>hP$2+o((5vRAj{ z%oZUBT_$o1Rn>mo9m-E}B@HV2T)EvJ)M;6V?t>oJ=bu?gh3ztpmsQhgZuY)_r|z|B zuTo~X@Q~*m?#QNC$gPizXcYfseF1t}DEhuU7kZWL{<=@G^Y$e1(bT6DBP!K$=DBtq z!9jRx?wwlwYPd~3}yp>AB&;GOs)|chrKnzioeg;I*9==5%M>8{I zf{URzk};ssRfsn`fB*fx=vci01&!PZeM3_+|%WN zU=a&7c}k00K4zhi{^F_nb3&viMYHI`t(_in{B-d^sk8U{Io_W>{bP{-ScDvJ-BkU0 zTA7&-jaFkH?=CIa)y(yyUH>N&bg&w@k=lW4-NoBg#>UGHL@`d+A(DQ$6F%r4*m!@k z2Jlvgp=!DawlruGM6qOA@>{>Cdk9)!Ck>n8!XGOgPQ_(Nh4h=>#QG>Rxltd0gkn>k!ksj` zGvD}>mIKOOphXFP{O=|=Y=r4XmqJI2|mOckv zh^rDqaLa(8=jE5@M*sYIxMR`Ts4Nwuv}kc7B{LMVwYgc^eYBWrb4u;o`WeHa=yUuz zV^?xCM(Isu6P{lHiu|-`zlxb~<#YAuV{khfXl4F*R)1A#aq4bm?=io`FXS{Lai7Jp zowAZY9t@1bx)eX)lc#UVab((6`Rq{_ccfkh?XXdIRgkM>jUB-543{>b&xV}t=5edn(9 z#yQa*nLvK`y!XZEK4!NjpOl7YAz+c2_qAtgLuMOo!nNBPMs&57!?ao{M|B6wffB>t zI$syku!~$OQjJLa3=uTT?9+Bwv>vuP_{GQScFJ3 zPjcW;Yib7MrD)S>vD#v+>TUd2F4M>R+YQ87au02hVdxxH@X{%;H2AEgr0!u|n?^T! z9!Fk)UYYP;j-cYiFOc;I{ZN6ggQz7(ucjD2QmI(`(X(I=JbZbZNTumQ9S3<^|E7f0sM=RwCF z-9yhKC^X%J@+ofp8R7 z5>L)RYF^rNEdv3v2im zYTv-~cP|#Nj~)ap-_zF|Ut>+N`eOe`m;RVw5UfR1W_M?t_o>umvl%D3W$`qzf;Q0V z_*P6F6nD&&LaAF=fIioDN;32HlNP%iQF0}}aN9L5cs<6mbX;mFEZ3P`DuL|z1;eIw zbU}LU4*$Mun?qI(xj)@sUjy2ow=fHZ&?Lubl*9k|tkDJJzQx~$<21{nBCe%|^z{ys z+rqW=z%OD3!Vr6&9+l72g^;oLEKW%v}G zp-$xQ`%T3_89QlG)e2{yq+oqqEOIoc1M4*xqe_2F>p1_4I6LnJ>Aa3JA`OhiK%Zi! zJC~MgjpltEXLa0c=@pq6Wxu40*+PYgfrQNuD`gmVmbbakq~wJ#DvkpY7uH_0mKFE{ zvtel>&_1Q~cUcdMFM@I>Asjr$^f9NfN1|ZOn1pBrU(?X#2qqj@9Ljgqcur z7UQo(Y+f}@@5yH9Fu~dK&6>ezHEGgU>`lHe!|(#gydd)sSpc zWX^KX_O`zO>T*^cwB7BQ^&l4SFK2gut1kE%95*G;X@%c-bL4+yRR?fPep^L~F3fw; zh>ymv&Gzp1-28Gt%w}%9Kg;QwIR}}mz+K-3ZPl1Haek+5*qyKte+?-e1nS;ZEKvVA zgLr#U|47*dRUmXT=u7n$Cbry2aJtsr@33O3>MEa0OgqMOCU_Y*^~SA(g2Lg*x>W&R zrCn*YbEEE2U96{bllk=0y6)+1ci4WK+$U51MqM<|*-4xFa{3Dy_dN50 zm<8lKQwk4szo8c?tyM^1q|mrPLXJxVa*Rr>oFXh-@GmruxVrC?aAInM17Z3g3l#jN zaa4EGE%(?LKNLGcii+6AF^tvyECUC;;sZYSD1PR@Zb2_FHbu6QxEhq2P z#%dSl^l!kMJ$t&-w9b=^qy}z=AT--@*9cd7+!P(-UVfT7uL17Z2%n$e>m-eA0&&wi zLR%=m%lMiE&yxd`;tu&kg2$gbM>DSrU3H!*(z*gJ>aZ6S*2*j z1jd89wIOMzKQsF0YVKf1jSguFwucVOweDkePXcx%&_e)ksus1p`7NCysbz8(YhF$5 zpM0S%bRaA@T)rI~#p8!s15A8H!$8qXkmCr4J0yB7Y});B_%uWhbcsBl_OG0Pv)E}u z_BNl2r%W{O9Kf77lC7qEex*qwMex&jECbu-zsxPv4c}_YO6s|12ZTmFKiHR1C|Rud zGTExe%3nnXx&{xg^iQB7lCH77Y3r!@p;uu|8-d>%&7)yQ+W+hxDVJtOiA!IOzD}vD zb>L5P^y-Gr%_xnJkc>Q74hdFo`0Ik1w{fU$=f#GGPrEf=mhWApSvuUFjoTw&o=_3O zB9)tT)ChYyac{vr?r;f_mN^>lZ0**p3g|yLqTTu2CHwauUhMl>ZEiZzFI&|6O?wtE zAY0W1lpBv95)u>1KGr?9yERABT2qyi`Fohyfy)D@&p@paf-gM(;h#=1Y{krQ_&re| zEx#_%tOj?!r-%vlk~ZF-7O|uQqcGb-5gJ256$h4_#5T4SykUo2GJUlp)$r3k?(f^x zHN|Kf<83>=eyA|h(;z>Tz0?%}13WPeCpT=3afOL+tm%z#c66o`u2 zessl2uczWB{fUn3euCW4KMKlV?62&@^y$t-diD?^ZJYnVPgT_v zwHqzp2c`8;9Q?z8Y2+stjy_Ul%-1XH``f{5mgA(wvRd0Jnx&<}x4(h85SNfswragU zF?c&|f#KYLDtPyN@igfzya*3fywNG`1BQY+6gLMHt->`glC5U{)cH*_nl-7>Mk<@_ z&po4Mcl}AOmCk@Xn-T^-ZvatenF^1r%IZQW#ky*qn}EV&$jkg%RzGQDMjW^wX#7gXgyyR?dIKC!oyn)wk@($*XJ z!QBV^iF!6Qp!@fv?Nm?`Qlns-VUuwCAD$I-{%RBIKGjHG#RnWBJ>s&bNeV1*!`gE+ zBRA?N3#jTb&B(qGfRDtFxrYPKY^HyJ7Mmxw2=^Mf4=FT`uh8jUr~?K?ZZ8E+U%RYD zw1|FjhF;Jk$A9I{3DfW2R(ZzuC|+Le#+b_Kck^37hb!^Ncdb&~g-FM8EI)vMGY*i2 zjw3ru%8x!sf9|O%pnTs5cLq^s|69ck2KT9f*6Q~*+x0>A%q!E%3M5PgPZPcWSWPm$ zrc|44YRlDAea=P}GOtcEF5@3g?fV8s${uLxRx&S7zt{%+#fzQ?QHsl16coO0p%U4v zWxlT>&jkag`|(DG6v;+p%%6q@3%A{E!29334_h^iLLA=#x(lWk0PB;KueH!tBUadH zV+vJpqRUWpz-_7U8X)o$o606a#ewI}&s@3%tN|6yh=hoS~bgE!4O217zUEd5=qn z$U32>g)1qYwIjiKP%Q=uwBY&W8%yztVf|f8a-WI~l7X_Hus)<^7uP<%L$->t{O}^u z4>p@<7e9>k+24=leuDMUk?yzW*#_9$PLgMl^dkRN!`D=VOvC>97|ktMB<$7DDRpR^ zq_Igj=nbG^+kvd<&I@%5k)@tjX;^5zNoE+}*gxIfQKMAwoqV{8|F^$w2W_y)T-Lx( zBkXum_mxpCeY9d?+*uWO{Iry(Tqc;mQxfD!mswVDLYCAn1RZwS+}35l;`i09GC-ZP z&+4)G9Xq}~{4i_b>#KWsmFL&PZWaP~4R6VA=tj}$u)_n|lwdf$AW>nb1fD^w^O}X$ z2ZY(NkY2J~r<3}9mki{;JMc`&|I9a=#R>3BRCd(t&f!ir-g%U?|El<7KAmLUr!u>1Ne;Bz z&=`9*6?^66jRERQ)!ty>-S13nHSGc{S=Sv@X9pmhF#eYi%_z3R;?9|(q630%)tN?Q zG*{hTw`BTe3_J6_=QeJUAyeWTNISEbsP_S^d+a!eAQpA!fA%?aaClkSgnS2GxC>2H zl-k$CP5&L@0 zz~j3tM2M<%+y#e#eoaA8Nv3Y(D=w zvHDYa`U-NYHe0z&EEWoIS00+Wah>Gc8gSH^_B^GWiXpf5ShJ4EPy7TeA6wByuSYG^YFak5cDyrU^`80 zv5Y;;@k@R!V(%0yNXW$0*x9eKFyG`!2M7VdF7Ib`YUlV?l=`o zfApP93Z(5?jV$k*m1aI<{SE&de&dL39-ZLD@w2U4j6S>G};F zHhJNFCGSsZ7|-I}l;z72a${2!#D^i!IYQlaS!G9uH;t)UEpt)68;gVH*&ooD#kEKN z8|`(^dmb=N`gypy_o7jJuZy5J{0|a%{d;N;wwh&^tAtUFPSN2&F?2seffb1zO>T)CMh zZRKq7vjussgjv;viq5NHG)2>>cjh<>t2ibFz9jUUlZ=OA-<@l&ijPk}@9OYSmUAe% zGH(@0WQt*KGYu1y{w~=qW5vb(WVI@yy>%aTfnFLXvpuKqGECRE3gp7A!kEDXQV^pE;Zso2vq6 z*u^LKLFX$cVrp&I0NUom3)5htHMa$H>fW>HNM|`#;_rJyr#H_+Kd>nLGb<75PIm1x z7e<$&MTt>p{6_I;XFVBV;`FadR_knR?X!xDV61zdy8)(0z2Vb)bQ6Vuy~jNOoo+D< z>fN1Ol}aTUlbX9e|GYW=rF2viZwtE!R*8^CmuuyPgCD$V`DT4SNq-G6c%fN}H5-ER zN*???aJo!I-`|P94K6dZE!V-Mvk)}nqIBZa$29fcWV9J3y48>y^X z)jU&YK)-hZk3xXpSQ*Yc%8t?E`OZOUm`nJ8DSi?@AGV6LlDlOj#4caiipM3xh(9Bi zw&`Jt^}_(h-cvW)EoN&rrLE4gXFDZG`r_8^WX<9o37CSd0*9MP%7r+hWhL46wh7%h zzP!w*CF$vA%U0R30$j@~d;Xh3Sm3j3yQ4{{OGiF_LOn0d0ll!PRk|^U`;$tR$hxg{ z#MDoWy(K=awU1|#N1f3KZ9&3Ie`aiYNBQHEqk**7}v`;yTxxZ<>aUFbi|c z$ahf)(ixYPo`h>52ac~(ei`%WaC_rn z`DLt>djBDl0$t}{RW*)6pU{du5i+T_EPQ%R#-`F&k)0K`rrYuUyi_XAnw%RFZTshg zwVek>mWr@Ge@VpCEivvb=Ib7h+rP$X{7$i^`EMMu(`s>DesU^bo-)hU>C1*$+t)S; z(aC6G;MPRPu|RFyd56er*24WFOR1*37;hb&`seEzcX)Baz#rI;fa}VRug_TatY2b3 zo2|U>O&HeOn)|bP@axAb`})|M3XR}8H_NF4W^DE$vIOrKujx`0VuM~^E)ft=ar>Hr z3Gpt(rq|CPScr#4_~$+1Yaz$m+Mq*Qv+D^T(KU{-Dmk=3YYV&JTCWu&3xyo>$(2sx z%ed7~Eyxb^lKq>*Az)_vVKvz&vQLFSjZ88VcN{$JRle&gl_kD*PCd>HJ*Y3)f5uW$ zp2%72xf0&iRzvS0K&37*oO{XLc; zS7{{4*!t{4|Lq#kR@Lz#k_1`+# zm+t8gDP44rh01>=@^Kbcsbp7tTLaD{t_MoneD0@z%0GF*>0UdDhWfTAiA4~T#NEcaM) zhC+4i?3bz!@1almprD*KrsrhzGnl~+u)?5dVXRo1-eGQh@pK3f>ddclWR=f-Y?o0w ziKX$rpseaXUgDTpl@cd`^Ox3Ht-_=SV2t$lSBuw=7#g{gA!_JMl=C3Z-z@%PTPSA} zYp2|av;*O(QviXJK_?ly5ItC9QfWhTOnb}A{-J|HP@EOF{say9t*FR4# zF)=diAp0JB#oj^-64mwTCEG`6(x7FzjY5FeC$Xf>@1`8)GCzd!SFWPi2z7T%(aKfc zHD3#VBjI(N4d`F#YbI;Jda6I3c@KXPd}&P$Tm#xKK0kbLX|bwOt^M0xKbrK&otACE zJ>A8;joT-;>uh|b|9MOKA|L#NNgKn?0A6piO%{xDq zMx|k{1-0(<^V}fJrXl~uMgacpeeaL;`SzR~=f5kH#1f{h zygcJ8zO}w&L6i9H5p%+@pk!~FbP3cQ%M3MCVf>PEwsnqrm_OUR9%%)wM!rzW`q91q*QqGgxMSU7 z1PE30+c?|=jt>Ht_s=(=eeHvf_QP7`t~6Vu_oZchZ+Ggh&`)?yK34!K3Avz~skgOQ zUY#?gw+O$Rz}c0OUq+i9BRcG)kVQt&!)_Ka4A_ZW*mELq1}Pu96d?x^0;@s&6KN_} z@5H~D;fLgC=9EcEb{U(>HVFQbhH*L3&LxwQhRLuy;dgtam???# zHw9&%q)Iw_y!N`5qycmz_9>_+@A5LRkBX4@uMhxb$%w|Qo_lonr)X9uZMg|wMOG}X z{preOs2hL&Bjd;_*xT2Cdf&@8;7LW@X~|-o4|1tMoj0uKx4|Fb6qeS!e>?b* z|NC5sNw6os%58EqcmucrqDzd>7O2DtcF4q!jdAg$Ms*;}aZanq_wc`)B540jZg`3W zr3>QTp~Mgn)tm1&*Vb<&wDE*N2%UGI=BP*{Uadj#Ee^tWwHHYcF!R1|G8$>J4CEA_fB+OcQJ)m190h^u8sTYo8SNG*v z(qudT#PG1qR|!v>%;H+4wT=@H;>$Nt13!Mj&d!l}Brctiy*+XpCxOeX@?O61F9TN! zGAeaz{5#6^3XWMj^|m!_Y?Sj!mi8@xU~~Ai5L&s`8mITWA6?v<;N(kJDVj@=<^se+ z)5JCMyn|ANPAVkx5%;}pL%!&;eu^3W_iA^vvg{3wNBNl9WAwT@rs%f<`Z-^Tj1?hC z)+=Q5y(Qg&UBEufqC~P}M>7%DtFvO~kY=m$H8N0DDHnGD2I2c?79o#Wi~+`-=0&=J z1kSklsuJd&5rGbMzcPxI=NV;3g~1a`HJ*+V3rooT^d-|R+jt$BId zblL2W!@wyIAwREGw{5!7{fq?Vz3#&c4+)-E7X@Uh&y%e+)o)JD;mO zbocz;`9u;enfec3pU;Ip!s5CXA2{j`(UWArA)&E)JVSZkE`ExG>pD7!CduNQB3A(I z#;c7kDs2*m6~%JEkr^}OFmoB}HjvEBpLA5)V?KKIdX%t4H*|wRwv*e>PS=^bGoYL> ziE|~sk%>UNf2y4 zRe4-x9$BrDy=z%5=f4qOTPg4AzY!&;08*6qd(?4w4a|l!7}F2LLR9fc;#3xQA8_c; zcBuiV9|nznSiifthv#CmJ45gM#f!*3W_>7^lMg-X9i;pl8FfM?OluzJ?k^JW_LEq0 z>;^r7^4-;rHcuUexP3>?_y*ex?8YJNO>Uoz%pi6JUg;2wpc0KFI?shri3w^d z9~iurno<%*LV%?TFKe>sE)U1a7Zj_yh5Td+!L7*H6kVtsRy=@S>bPiUFCoT%mo-4-0ztW!xRmaJLsxOJgPTvD>dV`#}2Tw5sH4LyJ^iD%_rUP!}?9cW<H0(#|wrnaFF=)9p0H;Mc~^w>FdGfDsb0z)i>_69%ICw&n@ z+kB7;-12AZT%+qt+2ul?#u?m$bh&%9BP}RW_2XWOvuv?I|FV2}U}z##uKfpX@2J~o zFD0A$Sv3FL79pbh&_QA$4=M@%+zs@YKwCRJQ-=IEq3p}bjt;W%#VZ+5hbWq#9&BiE zO+otRQZ$B4VmI?ATYVXP9#;s2b#4MF;e>U0{hN?q{&rFOFuq2WDxDWQgSb@xo!Ipj zuNIr797bST0~WU@xdW0Y#flI5`;u5}!t;GUrD3)O)+~#*;2uvU1i-3SWyj}az}(^U zKU<3kHelRk64mP-5tJHCrgXiLl{cD0mhpCoj>eJt$IkOjK9 zIta3iQHh)30m}Q|?05RpFzj)e8{|1$dfCZ7bk^Xyfp`W~ZB*goHj4gbC(g2)vaDB= zt%m7YSb_>9yynv|z`-8={5+GP=Hvr{GDG-AAKgQ7yd$G?p*+d(90b`)pXnUEx5yqQ zAwR&tfeA~(?6Y|LueW^q(e6K4)oDkLF|f9jj&@*5hb{$WQS=m>zl{~#5C9J~TR7FI z(w6`IzD>DxlZwk_c9e*$dY^`Y+4DH)UZFj1;fMUDPjk|7E3S{WlW?LPjV6^8`>!HZ zPYX?!5G}uO2Ftq95v6*&H_Cs*%R+p9QkPISGAQTgyYW{JyhX)%dG2!k?%MM7+*P3K zr-bKjKN&IqAAuN2pF7e`N@eK(G79%SQcZ5NQh98{FTANtt@g^KmRNvZ=8mUo?{9iok)-=NByA*IfZv|`w+PM1j#WtqqIFe9 zE1ErI)v3Ed!@QZ`pz5n{;Uw#??hm-&sW`N7|NES-t)X}PnTN>cxXhIH6optpw7Kd( zC8=zc|84?ot_`tKAf*-OTe%c{HU~Eczl|43{DM|tU?8UN>RaF`O^b}*EAq&ppk59^ z+uJz7_-ip6z@xa^;go$f(K-;RZ+GS!7`Xr1lI_)m)V!sGgmEj|dM3C!WcCA8?nK8N zI2Po6aHfsy9!tYC_Z-gQc4;nLHR4+5v=%#-UqU5upoOlf%XD?c{Wuj5rN?1L+g2I_q_IC;-O(@l z8k^X(*a*toIN-UBRSkq%yyOk!?pkTQ{Vk0QSnur3t2>u zSOgMo^gEJ0ArIo~MEj1tk9HXZlE6gvZ{hWVg*qCM#5C!Pr`-Qcrw5!mA8Hg`qdx?g zVh#zgN%)8qwC&(p}VuVa`9#P$%TYB`wG2i--tv=Nh|3m9p6)c z2!AvgXQkq7JN=z#`4iOP)oXj?yFrfqyE>|}2$j2Ma{bd(-apZ?VZTel49dx_RLOe! zCP0LtPN|+K%tjz`?Cq}37h@>5SVW%wW~EOO>v!e-(A<~Dm#3GZa=%ygv|>Cq;bjgq zUU@HyPSKf66NC0N$2gw4HfT&rGuko>x~5gEX7-fkIHXsFD1@r{=3Py4R*;DWP6JT2 z!74ruf|E(^riSSV>;K#MTGIyPh!q(jOsO1+>* zsQ1r(E~|LlG~)UT7dcb&C4*1JBc zI|teM(mbL=pLtw8YzSjqZhVBy2^E;K-?6d2eR=sjwVc<vPs8|H1}J@ZV;fkqzg}Q=$-jy=Fyy0iV(AjPnTNY7os~ z%*m(^aT+~1@tt?2`{A(ZBjm|nYZqOcDTLHB$tU)jgPOAtGcb-Pbpf9fpAKd7AscEb zs`50Ck8LpUMeK?Ow~B|7*TZDEYa`u7iJR*^3uI*G`vLG_9i8 zYg$7z0uP0u$Vph3#R6)-FQ95N0;Jq`J|(+~RMgB@iOGXxQar0+#U>iayH#22 z($}q&YHh36ERRNNdCY4c3y;53G!i_4aApns7w zVXSW8pRz)YdQPQr?Kya@WO?IesW!hEFztQqXMpp$%^JzpQ{7yMCLMJ_PH z#|)>AA2#%bnacI5Z3W?6x{sTL=0Vh*b-t@HFoz3JK<@KiF6c+*H;Iq~qGFYh7NoP< zhKym2ii5b=)o`Ju1ZNj?XViNpr1kmYe#n%s=VFd3nw#rX6I59!rl22bOk}iphF^UvD3mopOLyPi11a_wWl` z4tm2Y3`%s2WKTF$QhO&L{~yQa+F-ntOEHCOz>uHn-khE!V(BZ_<0A%e==jfJwcN}4 zg<;it%K4B%c?}~u^c>_$US#A(E;!oGZ4anf#J+eqi@SAZl^6rC<-K%kf#0lw$W*xy zKnAg-ybhDg9@>4UQMQRPYrL}tL@5pMw7?xdKTQ9#(0sAjAqe|=o#Op+rmR22@oM1W zJ!c97_9*%ZQn?*8QVf#$jm+rzx%&rc3OCReie6%bpMNc9PCqb_d6{&?%!xuyVT6N) ziSNH0rj5DNgDjvq>WhswBwujVG-LyFiwiubJH_zH1G+aCO{M@~8 zfY6E0uPDA!#*-2&o`_<+ypk__mDtV^s00W;CJ z!ZQ%M(^j4{2+pz1pdik*7tSgVOzb|zNSy&xUU zERF#?k4~V>u3!ZjKb{bJl(gEbCcaUqHOx8L7%pt7ODAikuTP^Ns%-X@1p~f>HNaWf zHvthXrkwaMY|`L}!C@A$j?FkkmX=gu)=eRjqpRo!?i_y}J#cG#G*JjLs_K(ZF?^G48>#TzZVvLtH=C0Eo zemwI0$!>5Yxut3|ruN&eS{geG9HB=k+6ROYa$+ydDjj}(c_Hu@IvtWNFri)z+4GB8 zDGwf7cX&qiNV)&)*8HCk)l!iVOs3cPzx0`t`E6k0+dL-uad@@_pX0z(SWZ|n=eu7$ z1(4Et#tax#jJ-P;0tvwXiB3IVI>gUD_1`>^c6=3Swz!tLMR@3UcenZ00qgc0y6==k)pY{8AP^VVP%2b;4?0jl7=@9BXeN|2Mws=qnSH(bC;DbdW4Ji z)W;j@X>vMQRpdO+9GC|f-H0F4w_?OJ|3*~&;MFckczvZCTHdKkY4MJQ%I_IE&%;I& z2X04Daqy7Xp2QbDwy|5wvmU%%ubd5$PFYR;+S-Cs4EZE&|HNXZRUhuf z|Ji|McMJ2LE_aQG!vNcz`QqT@-$>PxL-SBeB@5Z2GMXe&u`+C;A`Q4G&bW_%N^5Pp zs;W-61K_>)$2cczHs&#(zmjJ>)h2&1L2(tUd&utx4XV}#YDcp?%nA#S>AL8{pWvSp z!p$_{giIM5cWobmbMn7~vl2Jh7r$LZj5bO1+;FE6*%J+t!>+Hxt_v6WnE(5caFG2p z2Y^;tXk1I1%@Lk1)C+9;f5+i^NBn00ewQZ>tF@1sK-T)56|Oi8 z?h6rojo{;HUJfhoZu+s>7cw#^-S#j2bsbL<8J866akQw~)^-BAuKbHVQ4$iaRssgI z7NDo-Iq@R7YH+|RO;7OWhc;NfG|z*8dzx^2{s+C$dusIpz4!O=Q>iDvP;nS005?F$ zzrHf}%p)t^Hy-u9aK{p*ueW&U$}jNYH_VO&%q52@eH{!Y>m5?~JKGfC0_PtD+|`7; z@z|@achLNMr`> zSazq}ndPv7dyhjt!%8=z=vGMd@G1CfjH9#|LL>?_4xu%mSLmM9_q-ey?9C`9(qO^`JJ^lJ5y+%q!d2R ze8kZ4{M(V$aSV8M;K2@Vk-F-jF?7Jsh+i~;PO)j6KSWk0=!@CALpn=WX&)-vZ&dJ6 z1tr(TC=A2Lm#TX|Vqfn}EH(l{U^;^B<{QCs@B$|TT&KXG0`awhLqcXyZT%eIRmXuo z_6R^F7Bi(o9WdBRB$PoFAsqKiuyVidAnh_??pF_lvw1tdz*HYQ+<~9uqp$u0Yql<5=TNB=3f|((8pG!BJTz(+R z%=O2PA6fkYeSZf%$b*JL zr~e80I93~O59LoYa9Vy~rFpFE#aDelva%mKB{Knsv4sF)DN6xI@%}j{8dNQyT%Ymt zRDFrl%{{!$Ptsa`Aavz_g(;M{oFdOig*@c_?Z*rT?9yBJ+UhO%@tmd*4zgb2;!Vw( znm3gwUOg&KvPrM}na<#8@$2x_t*Sh&7SJgB8jl{GsQ6_e#~ZlbO&!%zJ(^^`Q|`&% zeGhH~QE{Y&i^MNs_Mf1EoN}pKNz*U?v~5zn{PWKx)*D$30V#3>pNFgIp>2IvpAu~v zAAm7|dy*_6qRz;LC=ocz9_OWm=w~#d-Svd%7j*e_eAauBcN>S`>4H72j?IJG81p*3 ze+{d}-%UV#?fM#FP#B+?!K?Uv6Lt2~OO3THoY#5N5Y)$TEx@2AyXee8w& zH|%PjX=twT8FtH3*EZmu8J${2Bd!nsNv;JPlYDo5?xKRN%2*(7Rk83H{{Fn!^Qd#at2jxO` zhHoB$=*$-F<}cmQW9;kSNDj;Sr~UpOW3@f8x0TTg>~hjSC813n72)`uFZo zC%x5>S{>h|4v6}O9cjY{uz$o6s9FWtFrp&x;h7o$$jz&ih9U z!_lgtOHgEdy!j4L{qIR_d-L%V7bb4PSHUt5^vgRGNd`0Xlrcy|@cj-cJshC{?x3W~ zHh>qg%PkZ_6)jfbgy>I(RV2rGImI6cYp+pK>9`t*ZcqweQxZcW8~!TnH*5k-ZXZH& z$W~s~l5=fb5cP%{dcDdwmLmDpi8T`8`6x)ai*fzH59?Mg3A=FIN!j=EWI%anjWJYuQ~WL@Y6f{# zpU*$#_80POS{i?dl-h^@6JLILRtVR!tDb%QvbJFSP&Pc2yQ z6Rh;#TiQ7f`jHjvl!(O@G0}5>6s#h28Hh?@(}P=XJP#D*gZu!yTE(3_dyAe8V6lVo zt)TM-n#X5Y`T=3_{r8`>U#P^MK0Uv)pRNRAJHI$Rc>892-OJ0O( z>t!M>rQ~KkB;1#Z$U3r`y}#Xze6WSb(w@iu8XkCNz^k5n`S+RJtu>$M=Y#-mp##Sq znM1K*U`b}6;x2_@O7w!GEYfAPYCd5G=lS@-DM>K&mXg0S@~*u98SBCaN3(M0`a`J* z(eKMmvH4#qR&)W3W-r6e=G^YZw!QPuA1=c%avXE9cupX8|8Ojgmc=&N|5Gg(coqoVh zoxc~Otb^c(TF*_au@obE8ked+(HlZ_^D>C`4Op2(Jazd>3pL%VSg;w>`3Q(B0d9N*6SbitS)-<`YD17{HPVPdk~VoBMp=_mo(VuBcmwo%M&A7Par99nOU;cxfCoOfzCQ^xPx;(Rh|5)@H|a0=B=-)%n-c5=jp2su5aGy~O*-N@66AqmWqO@-v%`4m~I?y66%xiBj;{NxJP>%B!V)C)rHUm2hKxwRu zKgM@B9J7_y!DvBRq&IZ_ZXP(`N#~OD=Kl2P(i5IrYcK@7yLv(O3k7~NG5f{}z9;TY zO&#N^@UYweW8}Mon)t$Y6-BBPsUjT(5u^wRh!BbhDn;oa5a~@tKtSpuRY62hsX|1W z^lG6-iUJ~p-g_~WNJ}UQ$)20v{rBE)=9}3`HgmFj-hIpSKF^%w42;e=+%w*|BMaO} z>Z^_31&hELk4pQ6uk~Ga=#DciSGy=GC@WB1Rwy(oc_Pqw_RT1|M`?Oo5xB{1BS|92 zX9JTeMDCTiGMv32@`~@4FM00{djT-Gz`rKfNQ zcIh_|%M9k)^6BASUvlt^sHKJlDxt^Z{eaV6HFd4dImr4s@a%1s#>SGiQK#X0<8O&d zpa{+HPP2~g0@YOsn)*izT^&RvO|bEnpA>kOs#9)u6-SAAAwKAL>>SS?qg0A6d}}lz zuOlYHyTnCE2o+COe7~F3_dli?f59+K`u@A0fI*wl1Je@3CP~nlZM1s8 zO2iy-GvGvENne9F%voyKPLa`Fn30m1MlsmHin)Ja$kqcA2Dd z?yWzbuRjURBd=U>xt?YSq*l50hOQrCuCdnLH=U?`1vh(MoOKj*&$b`P&@T|${hUCP zLUl)OnIcuVSU!!M7b1y_rApp_@jqtcXB`hQ%`U?0=u`07M6-SL6sRS^jlSuU){=~9_-5Gve#D_*c>_*)6&?>4@b*H3lLXwn_EZ?q8 zsCe`Bj*bk+=_sh-9r`9^Orl)C9`PgY1@um6o)C%`E;KsX1y#%$9TO#Dgk*4w+oa#wTk3ey}FrZsp36em4K{z3O1PEh>v|HVc^P}<}^DPR+M7BIGi+q z22ox~)HD%!SjOai4`@_4UNjK27<1<(81bBlm>|hUr(Z7 z`t%v)4IKyl0IRAv&Dz*JU3w#2YxKyUb~zWj0GtjHrRQuO04cxxc(tzgbjR=DA?Je< z{?qNyrr1_y1$dXHq(26Qrz%|xq_{mk7dOUQfA+r9F#6kRI_F1)Lc5PTC)vS87?UPL z58tCy^5`Vr1Ski$rl8etW`ByII2Qs^Q#*oWd|*L15gLwN?Ax_n1%4}ONBHd%@6?8Z z*l2wDtUe0)PDts{5YRHwd@llxjJ3R7aN)_w$BQ9$<_*I#*kqCu9eKeg?iLjMB)RXU01VlA0!!m%QB%O(L!8+K{!3^nf_6mx>c|g}_ z+jV;CyxHMh?VK|Cu-6m56+^>fz_-+KUwMP>qVi3$X?gCv5KjuV5?drR1xfL@^#>2# z68}gFk(as_;-pKjQlL}|V>&8Wmh&DVxlT?=i-Drr2j@m1`)LW`TStJ?gQXn=={?%| z6P5mi;6pdq@D||fU-_LQA~3&Iq`IJ4sioEHgs{kD(iK+G&le1US6z)G4_TqQb@}ZK z|BE7+qN>$j$qpW(E-$%LEU~;D9H8ktTJ&Kr#PT>h2Kr^-cQ8hgwF3h%A`%MLjpfV~ zBgXvQ-MDhH8GCZP3~;L75mtODvYH?*{XRiy zE$9HpJ@y>xtA!Ox87##{2sZLH%nwyR?3?9dFNB%>8X`VfL@7Hx7$Q9J?7jWszR^&7 zOpSDQ!7AnODeu78$f=HdHU!;)kgsd3D<=y)NcdlubE`Z^ep55VlyI3h&)l`EA{J)Z z`rSt`j`XV=JS%0nI+{h(0ZHEU4Oi81Rb9#_Ise<>r~Th)tu(r2?Gv#dy?J>Ub%HC- zEpbF{TI`|vyiWa|%+t}-W25&N64L6=^LsH%(Z{%)JB*039g%hWVNAu z)ZmX=7BkbsqQ8D~rF_Za>fc^yT{46kD(Hnt@#z*H*B0tWeVr85ZO1?7x)?wftby}~ zTZ-yUi@T#xTi%JQ%~!&*>{ra)3=(E7)8cvy$8x{aP$jS5{afA?>`o(f@W~?=N>M4t zJ^MjSBS*3~)eo~I;|yNjK|f0xw^4|XaesRA0_Be-J1$7MdAAL5+z=9V`<<9fP*Cv8 z7t11Wn7vuG;|@0r4(EPxK+Vj%7qX}GQJRGa)gvUm{HIFc@6CItfZyKjNd$#ocY<4F ze@UEFr&g(Q;X0z+g85`>uui|;XTE!n2aT1ied^XQFG#A|{fVpD9oLj~G>sYRb{VuF z&E%ZKRmyAk^Hh1?7!o8jm}E@fPyt@`^bm^Pr(~oMh8(otpZkSq0DDK@e4UEVQw%Bu zmz;vr4M{r;S{PG;zmv#E`+>&E*#)^zvP>^m9Bv>HCG?$~u@T`5|DfK-3z`RyaFC(D zNK`(`EbyWMJxL{!Um(){jC!m^VcSWY(B)K} z@gJr*=wDzl+uadhBgE(wd?;_J^eEVhoA*Nu6nn|Yn{^TO79Dbgge)Q}*BpMpJla?4 zR;B2>XW0m{m#kU}LTE@>5O&ACH4K=pOW76B*`G%T785%P4vgPy{Z#?u~$y27ibRF%72G(A!p9}fB1lHD(!FOlXjQY z8NnYzcnndqjZyY)d~gFT`HNS&C&sA!6a1 zL3A$((=@g9C3J^Xn|V!@-sg(Xt5rMk*KG*%xX^*iOn@JrrF9}H^mqjy<~Px69Cn^E zgko7!>7>Wphfu}xFY;Ey96Cb{Q9Y9A(?Vp^H>=Iv>7lm}jV_tQc_AywNaz;=Eueq- zygE-x4%016Og>h7(2x3%Kz6X!Zc%;&1Z(g~-X@GZ>6qA+D~$V+g}o1oOB4->>4G_2 z!Z!x}>aSg1pngBG=So!FYUvHbo^8le(r~CQt{Y@XEWIH#trBT;WPK0Ne9Qr>Bn+28 z@I9l1miIQ#F|4? zwDb^4#AfKxcG9aqqettO6M*KbpYvtXCAwD+gnH98=tMLk*8Cn12|X8`83$+HbZq{# z*@X&iRDO7}?-)1R{OH78xJTHNJ$o%;4PU)a`*QXrB>m3qi<`hLjQ{gQe(MauzEEM_ z9pWxzgc}j_8baxhOTSV7d<(b9QC!dfrZAlg>b zH*3ZkxjdzWt`5Yx4SIN?s{vdP<8daL=J6P!l$iv@K+q9?z1t$HuFvLk844Xs&q1gG3(E zo3vB=ODq(9#pgCdhZ6A^tET}aY?jy)ej{Dz@dYYuG*paH!{yXraB29u>4!3NO_l4L*FTf(H?_S}#gE8C4BEM%pdFt*B>65cua z?OphX!y@Ni_k0eZ7^Jh2avGfdto7Z*N8tk_815J~W4ib7t>-|<<#E3Wv7thYmc9f` zEc{#op58swlRLA4m@53%wTP(fkF-S&G%|mOPq>ZS#z1l_@Pqsz%eHOE8UD7i=<})& zQzN~1Zx5w&P$&f_arq-rO$HEK3bl>;x9VRZW4jIsQteN!JlkmdxW;arnZ5|<;p7Wa$KbP; z*XwbYMhLJ8Jf=U=Odkzh@;VSg3^Xd>$i(~q_AH3p>>>fDGgQoP{TLu)PG3keR3bz;I-Rv`~Ee(de!^+^mt;=28I(v?kp-fZE+PHme1w!!pw zfG`LrP^v43ldOU=!Sdy(%6De6hL!)c0w@=r8=%lbuSqSr$S2uJ--Em1`x;F8HE}uB z0hO!g&W?Q+9uiX9Pfpr?-N5u+eXb?$l$c%aSBO&Ncqd#Z3W{V?vd2SwqkeOVs(l`N zZFxC|I3H(?)=P;j&Gd0o7{c9`vs_4#6*$J~-;Wj10`sW;bPdG62nLD`hUlm*D>`3T zgqla?=H#gTRkLilSNJZj;*DUYU`2{#$4jRh$R7t~U^t|k(=m36P1JS5-#Y+R*bWzfa}Jyl9L0M z&%?zhwh1|61kt;zmhY~Fh3_1A6S^`1|M9Gf@tXt}2kyY{*&T?z=SbUn9a$@2J#%Z+ z^)jj9(1aThr;)+Vc7XlW4NKe?p+SP)69IXX!Z8At%Z2n-its6R5DII0Y(Nz@Zi&uy zYkr2~3NNXK)C!+%c_^eL0UvLf?^5dqlC(#+=)Q(gb*m%K=nx_3=a2rf{z|Em+|>KP z-2ckMObu^7QPvlGsFg1#wGh{m_D}-(ETmZ;cLw zleaf5>jq-RvRmpGY>6$dNyz9=XklsMEoPx#y)`Cr-RyF$_!aquf9EfKPKFNj*xpwxTO&DB%PELV`J;X8!T`3 z07GNB*)MC=$X!ap2+r#(rdWgeC)bN$skAkP#p&OKp#F)$CkjvGs8#zx-<2VmQfU(9 zth6-4TDX=Q!EYU+2-lC#J+EhORym?tG6Pb)zVC;2lQ6VQ2;YD8iz^7VGOJGlR0}!A zrlcz_QY_adfl&c6wCj+ab1y^!;OK-D#VJCVwk}~G{Pm4(>T-ZOYEvy?&ZuGCB zr41JINgm;yU9AfRr%Fgkev+tn6@dBF=$TRm*vF_C@wsdfT%@Fx#bg38P4}u3=L)ZG zPX26p_Se{}xDu8!HD^2SL^8d~{J23>>DlFFU9Mn0+^#`MIM1)U;srYB$Tw}Z9z`SV z8<*hPp=!5(Q}Ti-mbp6U4^K8oNnDO;p;S1@(s0{D!yh6d3Ng@)XN6xm=+zg0Pl(EOcx9}rKYu>LVQFH?(GE+RdNEl(&4=iS6Yyi{Q zU_MS|tKvXWvsK+aH^TlVPw+i)2I!$F_Qi1S30xT|>9jepmNT%g9wBQm_y!A^%Bc9O z7zy}Zn-0nl6qHVxytyBEgxe8)wxRKUJfDDM(o5u9+unj09-bA-S`J^e>(F~>XQMte z<6IUq_99UcEg%qkjqbLo^&#Pn9}^XjQ^fv)}G(g(j~QloG?Lw-~j=Bw)E|p@5Z={$Eqz% zaB}ZsTQ2X_pBugPf5LMQ5%>TaWLCeal>6he`7(z;Z^KINCJ9FlmBM2B8*@<$f2Fv? zjBkXNXm4v7x|x-G2~E4_>P5g~B`o`>D^I)fEd+|eZwHd}x7(DX+p3NaS9om0|9-|N z*!#MFqsy*p30GFQa^?MWdwmB2D_!Q6ygEqaNNyBeVd&qk&UQV7GxnQ#`|C34KKUxyN6ipKg@B~stCX3xTNf*QckTDw?>z0 z*fi$vOFC;%Ete_Pmxl39dJ&^1*q_i^G5q>YiNl5CFBsHyKVAR$#e44Wt25v6e}eUb zisn&#zR#2kFpn9#+YpC$yq;jc})8fKmxK)qM#J`{-ZvlMbstd5ql^y1ZDMtF*{| zxrpt1_N0S{u<>rV#?Pm*;UlkZT=ObB1+4V{Dh}ipB|hK~<{NU$us!+0t#;eky-rFs zhUzTn>u*s`zi*2ml>DYW37qJe7y3vy!kGzdX&%$uXb0^2+7bF>zx%Quojy}Zf`Xma zaBfl9bBE~5eBj2hWnGsHjr-3&++o?lxL6y8cSV7>z0}5KI$*GxX4@{_)QX^I3qEdn zWd|cEF-Wt9ez=Q@6T0|bRhiJXioUj73bDTSF&-iv z{K^DW1yj`EeIx4LG9W+Upj1lT{?+8mxq*OsR2slu(W|^8BxFkmir(#Fc6L!NCh0H_ zPEw{7k)>f?h&WBwMn7F$b4ya&CJqgwdpK*X0deu{S$Th9LnzfRbAaHM^GEnS;~6a& z@5O(21ATIzD!0Jere>aHHz2lp1*;g(d{QQuol>O8j-%_@I{SxoUK!q^YRwG&^FfBJ zmW+|J)Xs?c{y-aTRIK)gMcur*{>j-P9SIZP^||+JwbY7u2r7>je4)o-` zkprbDo8x{~yCIBuRPgK@bYW1ZZWu;)l4 z7OHog#~|N*G)`$4&FWhCegRmy+prfe1Hzb;sao%;(B znOE3)2$->dT?ku|d6z`|V_2GByt@!bI;dc3&MmUzioxYGHp}e^eNk%S4!-~8apOs5 z^vZp}o}NoYIBa)`Iv*k~^Y<5G$_pDF9AXm9!msUZaa=mHyh=*?FpFh+J$P6oMK=s1 z7Pg=5ZG~M^`)G)2tng`#&S}XNooyxsE_0DCobaTi9m}OzM`Vp$u7PV;>u=X@An@Hl z6^ax~0e+$;G4?8K`G*J}BzpZ^umcG>;?k@HBMP!sHld9EA-Lo6ZIBZK?G8X;1*e7& zixe;qY=DegRjzO10J=6De~yTuiEdcIfS~g?Z&I`Ph*_}BV72(8KN1KnGgEA)QMo)4 zw)(m71|hj$8pc<8YNGm?g4)Lc$)=BgpzDPrC{@C~D*+!gWbcx~cz$j4Cp79f7@$yG z=iA}2H+k_<>0PiT9byRU9BVJV(XOfsP!9ePxe+&FdK0!drm@>wqTRkO5%_Lb038_Jn zJ#Lc78~htiL$3a;?yfpZ+M9j6Qa$28YlsTm;Z_3HuX}6u-C?}TZRyi+b-#T-3k8bJ zcK^2pYLOFD{FE3b36q^Gh_FL4hmv~Zwvol1xHT4~7cyH@h#QL&}=x;gcyt#XW z^JMgAefH=5?2=fUd(!OrX*W2v?lDA~C_F^*# z#Uz1EKY>|QKi|GtFZ9->z-fqS>iji=-Rqiv^9Arms@_R0Q(!KCb2EWOy*zjPay{8i zwZM~X$}OlGOw^X$bbb5ziXm(Eb{(st`fx7NvM20bA;!ExqCkb5yJgB4ox=x&Xl^Wl zrRZ!uKy`45Ik_;-kgNG;bhuCCx(Em|Ie$HffP8Y6ycGS7L#7kd-xtvA1`L6CXLepv zj^+?H7lFG%5$Z!fkmYR3-9vUiHYi+Mn7KpAMdppX{PYCI)XcKjK4k-;7yYjd61G)~ z^K|Aa6SRJb?QsFOSJyJYO@P9>oKwi!S~`kVC+Uo)_5>lk)m`6UZL+k%Es`j_!CX zzajy}I(8LTzjdrRE_!%Z>pQ)IyYWE0m5^49| zS2pnvWBiGbRBWvGh{WL)8z(*yeCu7IC+X3~sxCco8*&Tl>MlI>&lxNY_08ql1(|Zd zEkM`$2ae2`sk?d@=jhG!`t#TAH@GSPnn8}g@y!drZOVTd8Qpl#2A*+FeCs6)dN%1) z9}vX&c{1Uo(3`0raugejZC{Kt$P57=(aUEA_o0>$quwC*eP!(V&KU89+Zsd11zq+7 zw;PEyA(Ay)Bjv$Fjy}w33F1u&Th-(d_8fW65IvbUk@2`jpVk7}= zv~ak5IiR5f<4zX4lT_Aw3Oo#OIQDaP>ZDeu39dPx!;*0ZBp2wh5w^t|zlBzLu0EFJ zEj`nB5WeI1CK|6^0{s5Obl=9EKg33yo4K{$JJD{8d;lcuUpCv(y8-M=wm@N+J^dpV z8Ob?h`|LHoGR^CI@~0!jfo}VYW>(H0h%aBpi}My!-K$LzCS2oSP4bwJ%MtM`&oX@dMjJ0Yz?Qey&)vk;6}HgEK&QYL-;YZvjwfpYzJlY2IUyeJoQUkvlstblf?crWcN11*_5ce=oV+_j$rThJBn6YV$P%*uQSt zAJ6phu!Ax)b@xPQGrTlnTrvc%ZEn@uXkxZE`MOg zcEs-92dJ4newY|>CInUgy`_HTj*5`p)u$}oE$D8yWwrnj9uR%A-AEL!*NW)ZUPOI+ zmNUkH4SRm$`8jI-gr)6P?qJZ%3ZC%FW4UABbv4?lgwz0#OYeN?$(q4ga!Ce|HV*bd z%*0&OZ?t$9xo6ZB;&DD_dWell?M?WFI5N*pq}CHlv}M7>&RdzfXOXU*@{f6I?wB5z zw81HgRnL>X<`ttL#fgKU>}_Ij9C73haT_xe`KWIEX!zfD0G7OQFDP0+pg7cDfa!eW z({qzk&++lNl()ONDBNzDG~2yQ+?1hkB6kSNh@}JZ8@E;b{kJaF#iJz>j{
`jym z+ln{@a}2*sBz`b7?oKM2*opDeT|$*sT(Fx?r>G93Vwk#H#7M^zr@u#k<`fUT_FJ^z zJHFD;X3eWhrfRHVH-o}(W5}_nS#yN<{2QFfx1tB9*7l^~vvx0LQOWP4bP$ zzotgxLIzKxCKAPdk6%0=aA9Js?ea_7Kd;$S72mrmi*!k!1mhwGD#%ssMLPo(<581O zjqU-a3^|X9J4v?9SJeb*Ue((I2{u6X?wCD}uM8uZoSxtjY&h=n$N=Sh|2EF-f<h~1-1MaPT^z;Hy!9Ms$QX|Ew9Y=|BKujzBQ^3ZwMJ)(P>8NP5%H+*{1;6iq5jm@ zds=!<+eVX zi&hD&uQvf%iSe28A(`4mNL`-!-7V#Ww{LIPc)yNO)KHONRcoii702U8h3LfEO=P++ zQFtWoN}d8Yf*vz=Andw9P|K&r2LNkWo<4d$*$Ki?RYg1ffEvl9>Am8mpo zWi09M`->K_!N%d_x8>e!v1>%j=q8KxgQsEWF)_gj4()dhwXGselbb)^nB~kiR`P}A z4!Wy0v%b(O5j_gu?((rtts8HQ{ZL0BdwnardH%@AcJ7r_eFmhO?EL4oWe3)5@549v zI&Y@xnu(O|zAi5_DAGL!u$lL%IPxn6cQmLJw4F5VW4K?=8lthGXqDnJbEQDry0P_~ z-5c8nux^ca-sH@Q?iS7vCiv{;uXRbz5Y%7+Vi+&mijXevFzT~WFTHu8Or`2)n^gdv z(!{s3IPU_s?v}Y8jyh6MN?2!?b^#VCuk_|)@9U$$Wk=V)$R9Q`u9-lL?y{iKY1qec zZ>9J37I$5HRoqLxsT$v!Giv|lm0|p-nsUnNAQ?bfQv84kz+rJj=CnwO|OW?fox?ZS~UXTiZ5YCVWnstlnTv34^}8xqCwLH%^@L z_g0#6BSS2d$@yrXXBhXr$dQRl8JGJa8ZTrnfB>C3)uu6*XxJnJY{r`Hf)t86AD`V~RCzoLUx`6&oO6ifK4P+!b1b@Vp8cLv zZRW@P%(#*=Pgc+HY_ zmX2Q^*YV#^jIM`9xOT2RK7SWS_$45J*AyZBM6A&YW-xX!;gQ7Y_UVPHg=-8wm*>b3Q1!MK;~&b9Xky%|L#l zHL8HLBw-y@H(s1E$nhUVJQv9Q198z@Cixc2Bo=NjU-PKE%` zYuQ++%{Z{e*xTLgMU%o8A{j!`B_Sftk2xg0O|Sb8dfEDu*@c)xH-n2cE7pVfN`FIX zdU(x3qH(QD=NYBXMZo2sDyh8$wsb=thA@XyLT#EJZb1XMXL+7qJ!_XvUw7ia;<&|l z+Vc$f@jf`!Q9|O%RGUt1*4suZ#b7VkPAd5G_^L;(6A^MG1=ewbptlhGrKT`~>2kpO zZRHzc0gPHKaXgc_L%QP$16>Z|?cVoG<#DYdhuF?i^YJJy+7VFA39s?IUI`cc-t;p;obkk5sOh+XZGt+a=dNNGiLGiHtwK;PT zNA0Htu6Bvhsa3CGt5fdfpW^GrJr;v#HABhR$HiuzycBs5U)Y zkhPyB(35?Y>^5$;?#hfM--$4=hm+<}y9_JuR0Z26`y}JM?YAz<-fH`76KEZX{->$^ z+kd7%WXY+3s)qE;Q{P@~3@WlbUW+)#9Kx%{{7gg$UaPc{ckoe{yDK+ZDj=~6C)ND> zv4v2HRY5;H8jugK9;!kvvv-{iVJj=xF*c8I8#@3e9D*MvU z$H^r^XQ)Z;ac`()YpLi>|KAlWYoAbx(1^&8<5obE&P|*^|io$zE-FMRszoP|-=TM+OIYl!y75@;#(6JI!j9l$PWN`E#&B{{?-3 zEe(*vSK35=JiA|2U+UR+t`@EQ(nJN2tY(=--Np7#g*Wqq&O%sFoA`(rZcoc7%Pfjv+7K@hb2RpQliJ`EoKHIfY~zM` zYO7^}3g8pteM!awD;mu%NCugAt!Wt%n3>GH55)>vVaK3ej|b23l|po#;Iavp^_a~kNysXM#F#=ZY4XB;mi zgWCMsbEiPr8%P(%U@;a_$RFK#d#wZEx88YjUI+!3 ze#23#_j}Y2!9H|2_*m@?D_JyZ9T5?d`ReDYmAB8PT}XtSZwVsD$x`WX0$He)^BPWs zU)bkQy1@PM)|*O11a_SX7MT-#sJ`;3MSFX3F^3#F#8lYQgL;0}t(#!vklOQSNGZn76TVD<_{Xr9iTn9^H?*?aGBAUYJ2-|1s%8j z+4c1Gz$Q}$cTWShXzz@3-N~18FKVfL?d^cKNwZb=^((J8`pk5RD6{5#0e^4M z!SXaM_`c0g#7G@ACTqi(D&(hu3^K|Ty}r`DI``a-zS;;!nJ$iuFljovT-w?{vtUp1 z(6DT7SE-y75i_+Be@6H3cU{ZkwSsoR%o7tbhiw{&G}$Hc#)hS1sH-yT@b<|uNcG^+ zog``os&pUR?L~{00l|% z8*8jWqmSdUIr%(@ZnQ>`sw|SbMk%P=&c4dC3wJ3(F8dNmZ!(~N0|g0fh@jrb%jHhv z?Z2&_O{(-eEADN1Q2V}M|5;Yuc_FJ*muL+iq#Xh; zxxZ^sLf-${ZGSpG!W7)LgHSk{l6;>y(Q85Ryaz^Z*q2K%uFfsUkt)yhp+(*x3!o4qH?l4BYIIq3JVkAN@ z&aS+nX~OXF_{c3T6@%I(<;S17r*Yq{T$PtfL<+7RAx?8msgB`Jv2X;)fVakR3`fzF z3?{)?yGPt7DL?OjfRvJ+Qwg$^aGV2HD9PfXTdZC4Ow$kY9Y(>A@ST9lcA5OUz#k`} zYyhx#v`Z4I!y+#?NLbI4 zy+(JB3g#~>*nF&xI@Rcqh%V(T!|AMaKD-%nk zC~8!Q_iMtZXF}x~A`*EiyC|ykOtNX%jka6CgXbd;x?1dzTNGQPoS&>A^dBIRKieps zH$T^6MxiiP+>2O-BXSvEQwO56Nh@TA6iWBgP8(VGg#mG~L62V!M!4B_+5D#@ z>>LFtJp8A-K+rGkv5qt86oI#_Z-K3d(Lei4&=mDt!LAz$pK9w$NvG)kW>qAkX5*dG zN})KX{84Ho>(9GF{x0vuLH+%PYZabkyq}->w#zQc*xJfkyB5t+bK48Xw@mRrcfh7G zG+r9U$s)(ij{&_)p0Jv08=KKd*TC`2@11 zDy}vhpe|j=fVhTmRy%Xd@bOKTpLxrOv=}|3Eh-T^Ct7rn*SH31HMYnJFv3#G8`VaJ z;9KKFNAtT`lu8FT0Kp1(v)SL{N6uhLq-4V02j)zv?Pf``!PV zCP1sNT~Ih0Ar%>#I_v|kk;|(@%%Ty?^#UxS@^&^Ek6I~t%@PvtMNMVc_te8a=v4Wo z20%^D;-Xj2R6&NtmlGec2%WEgp7#+BYbzejTrv0AVpKjdG)!i493P^fZi~!-baDTuAkTyO z(>Pd_GrYG+eIS1=BQeFJY^3$5paQt{o+%y+qY?eMv2AFKvk;6rr0rZ72Q#K&l%!S9 zUHyId1tn=BsEvhpzpkp6Q?i4CXwGALtR#)?zVF zfm2gsJ`BC*lK`6HK2@bjx)R#L9i-EYqNe`1rMW;)|77x(X6X3hu z>jPonvVz`XUPd!N7$2&cH7Qa*5}L|qp?L2p$t`%RwVdWr7nu1XUAPQ)3TcQlnKKg6mJ}(z>~8RA-1tEfi`vC8nKz)6 z0t&nLAGhT=hbTMDF_{Yp1Q2gOjew7?IHWz{1qh~v$VYW>F;z3JS8YxBK2w5~Y1q7j zk{#&gzHf6=l2)E3G4CQrnZ+j3C`o+g=Abv{dyh zTg%`+!Xe?N%D$^05dU^S4v)I#TSPrS5EXP%0xJ&l$CXnA3qY^3XQ-Jr$O`|HpzJ`G zdzU0J1FT9dJbIzCyKk8!Q~_~Chj^tAYWX1ZLVp9T)}1}*Fh>^@Z0=wj2}OuKPOmCz z_Pm+Djj30l6QkW&1fh|W3ghn4H&TXfG!RNLWWidbNezRAy!ekt++1@ zKtxW3PaL|k=#d=Bt7~rhw-85h8lt# z&STK$E3e`Ba~uKek@f8T^FZua67&@7y31#@z3iH zblq#rA*9h5e`9%^N6dqB0{awt9ud;qv<27f0t+v^B~o8QbLi%JI=!k#zUEXl*dd;` zj`GDKL~`@X-6?_6PfAbtn~fiQ`KDee{qhUk*sOW37ubaaQ8@So*U@uzZx*+p4-GT> z3#gl8JZPW=C+vb!8p1)0`f0L_PsrJQf)OS?qDo>3^1IdNwv2unti@KB0TfMnT>m1t zA8OsWhWl8ZxxY-BTZKD+y*v(gx_nCRt}24QoDqJC*uh9V^0{SyuNPWl8b0*iGp4`u z2|n6jbN(Hr1G4@mZdhp@${Ld>NlPcY{}cirA*~z1Fh^T*(=SBcNgJ*nQE%)rup;&VU?w?ua9~m58Zb_{WF>=`cn&=EXRuKQIh>M|}%3EAQEZ*xqZw zTMIQ^+ebp!&)3=WpkS>hOq3+I5w2-%xTiPCl64*WkOUkFb32}I@T_eucp|8u(I9hF zn()Q%gCV$+a!o~UmzR<>WJjHFYFdN*wql*L6R5b44-J{<9f6;v4`UY(g^~oN$NWCP zogtAs$U!G&ILH_SQL*HJsAu$`^Gd9GHqo1i4*YDyLPSqty9e`#=&X}kUhvlM%;%({ zu|_r^{r=hq;9RIL@DIX&Daj+s;-~1A5nW;rNC?fC^kV?M7#B9wZG}fJy08|(A*&W8 zip+ra8S*yr0dR{SfTTc{MHQx|t`DL&#|y_&E&s0S%{xH1l^Z3 z;X8khMch*_1g6}d-@C!>zkl{YpZhGDTM_NgJ){lco^<+%TDQ000hK%TH1EEa2McXU z9|q6C2@&jo>?ZHPdW7&5{r-UeT}>$gwd-Z%W zt%lXN_Gu@Wet}=lUeUE9cF&T4U3^5R&_02lVbB#`icT}`$828NSB&@^5byS zGE~6bLZrm(BA1H?iL5ly%c~>n@+8pw4hhbWsVtA(S-3?&`+RzMVV=cbz!i3a`v@SU zcEO)}6)2M~YiP%UjSgq)kUBVo5ha`lxOs1k*B&TH-Pl3KPGz~N?9k3#8v*s<)A|&> zE@5~L|NMtwIhg&CjdDm^>ZC!{CyXM#>`p@bRcu-EJ2XVJ{>Nu4={Z1?$QB$5L$!m) zcA>24dz?VGB?nd>A{|e>C9;M-ZKOtXYo#d?(zeSUyLzn4Uf*$i#n);6H;j3e=Nt8O z+{M^cHM8oD){US;*(c{`4arUm+$@>s4{OXA8;Pyz_u)>?>5CxpUd-1)XzrjssL&TL zWIGIPUHCstoq0S|U;O`zWGSTxC5D77Q6#%tlzk~nqL@U4N+Gg!EuloVY+0tVMrBL3 zm_$*Qi6Ofo%NR?8vCX)1&u>28@9+0`{QkN3k9+Q&Iotd7eBFEA_ePGc!N;7{kIUvmkUJCBe zKj_pIEMG84QrP{?eN5Mcg3lTJuJ#;NqeAb&Y9XLqRux97qS#Ky!&c!xBif$Ay?^Um zPc8b?C>SvN_dj1c^)Ejp2AOH}G91!U*;aR7Y3^a-PBY{q=%jY{HDuwk_y{i!tekxu z!!k83g+T!WpZ&tyES|`mT`8R44v3SC(!U)Y3g6ZK5!r>GR@n`9_yiH$(3W^GdF&ff zs*S2WTF5;lF$9GK=lli7I_L0RDu?!hcK1)HtblzL+ctDxm{}7=W;Okr3VSHKa&mve zla^^{?qWWR0(1$&KY%3<#|C_38%YTFb=|dq#0;}5^DoL6@xMVEqU#tpm>7E8A1F%v z;*Eqq&u-jR2Em~ zJg49mdU})%n*>Xn?iC^lbiy1MAUzV{%z_X7g2P*$I6AZ+=j)}F{V6BVu=F+3Sl;cJ zds|9&bxFn`>lg3OkplGH`HPS4?Oz1qoy%Cg#Fm*LpQz}mU%aFHI{)HXXPVrYNh+Fy zLvVONI|;R*gy7UkisMUw9WOVfyAK0(i}ZbGK{DYI=&O~}Zb?P67&5zEGmz3t)0bom z$yj)2!&X@aD3^1xQ&a>}6(8K&|f+neK*b&ycvx=2I`dC99 zt;+`HY%NXl1IE$KkfHl!Fy`ChOP&_BM$(tvcz3sRbln2(hI}ZsIHT&8DzIC?uOp_* zgz%0TNFV?`wDdyn2+PI0<4=HBUlv)w^a0F8&p`W?9+g#jN4jUn4gK z@7?AwN%k8I;8HI9CbZG@(uRbXq5-NI<<#hspZZMe57F8#EC$oa^0ilHVg|pQ=XRNc z6$;#){(vo}>$^_+zFeVEBx` zyxxW;%YD}^w}*SZuWrKGMqZq(Agq;pS&p|0=(MUBxUa%mKiz9fy}(q*9Bx9HrX=$= zR%qkHHgv#vqgXPHXZuBD%+ACA-8iv62RN@p{{=6!I9wU)mI)t8)D^!EPR@tVyt&;$9}w8X&KlxA=UXAFF4u879g4KgPi=<2hz6lj?s45u47|sy z`{@7lAvM$T0L29t^I852U`lC*+d)d*QvK{Sj$)H{&xODmgJSp;TDgk^4r z-tDCaC4ea7cO}M9Wax~M!sWq$;vk<-psSUQArr{70YxV643JvNzfZ|KaS~j8q<$Zj zg2tbOvacg_oW5;J3>|l1CNNR}r8~!rzSepwz7O6*HYtEt{W}C|n^lcAVOL%U(PEIc zp9s{g65~e2SK)yE;IETp&^hhi4gxRhwur1L`kaHgaZop+Dk5x6prE&OTuOW9$&OZ&sA*Wy(=szL z7uI-CybD#6sF3|-EGCf|DesO2{P)8+b;rOv1rnwm9N6)88)CbLplwwPvRK{DjP9KRbUeF$qFlM*0_eg zxgAZ)HvJc87Vo8d1gc!hqd}58>y+dG&8q+TnWtwFjEs*D8EgM5@~xH1JWgHw(-81b z3`@U@X@<=C1;=!An7Cd4oErWd09Vzsgu~$J(-NR)QvaM?`0-A7Zap{R>>57PS7*P2 zH1Op*)!PQD^}3yI=ib0Nft(wN_um`P^q%dM3kfJctI`SKm!0DeBCzVEpOtwMEwGZ4;Z$VW#LICFy3Szy3}*8K$m-vKoikPZPn zOIA#P0LyHSNkikyaMRoRp$UvV_~(YT-zK&0)ITcBl%WweYx!&g+Q0N~G<6Yk*kgag z^ZOZUEwT)+_V-qi!hJ_wkhLJb<^H;0 zi*TY9sC#H{Lo+Ulvd8oPTSLz1skutVY9#gT6w86vRcIJSx-&4Gfp)DxODacJ8E zIDuy#e!on(FhO8L+zsCi3Z|r>tc$Hy_PLdR9QC3Cju9o7X4agztYBKQDqEzpsyJvSl=2g9|E#i zec8-UDCIawH4kR{NPcNd#zwWIlmTL88o5Ssb>?L5dyZ5PI50yelz&1!)O?uUzkbvt zRg)icvigr?{tIl?1%L7UzDNIh!EQdpIs=?uC)t*s3L?zm6h2KnuqMNwcfF`PrTdaW zZM^rWH@K|LV<~4&2ez2O84j~zW;7Tz8rro*Bh38mLiWtHrzCfS_gw9>s?&C3;IiQG zEw5DTF_KV88e#4iYMCdFknyRxF-%;Kw~C&>G7U1+TH!dax(&HlhZn6PjC0?L%vx{$6INdu>?j1f9I?p9n^Kc~fB3i?(Yt-18A28v#<( zz;&K^q?C=OlnUOh5Do`^zY`BXekhT&MV=7EgDxxEL?*6Czi#a|O4~J|zDF4#H;ym< z5%gKHe0PivDIUs5dk#;WqABf(spaxOs}%Igc<#MHgC7{Gxv``@)c#ByAi4C1(8)D& zoZ;7zJj6Q#z9kVS9>buPz2ZW6l3}L|V7>zlmo%f{PWDaz$frToplioGK7rb;0+dl0 zGjMMA3L%M*>VE0;ie(R@?0~8Q2GC#FG3j{Q8l_}xJ*X-sW;`85e9LCRC#N0R`G6FWCe;o?C`aUtP<`{} zlhY7sIc!332?SgXOWXj@_awurje^8oW@r(=YTfTS`oN8_zhjH6i@!XCv}hdJhHzYF zWiRgh#c2s0=bXT(fLFw?a;z$5(FcU-7ZpCxFvpU*1aG?k#pMZFc?lE%6YpD&i8tdkRc4@4v3UwTjvIOHL;Z!F3;2%s?%#E#=H)!GXe!?yC^Tz2)6Wz}lJX zR77e_GElf7`;(1MsF3-`uU$xoeo55JZ$sDID7{Vebh3og8J4n4zYUD^L;J64v{roj z17GBk<7$OUe}cnVl3y(r@cSvhSVd!2#x-=o)bM#EVWyy;j`O(<%U%L^5x_8PnMD?Q zT!xPrQt-KV#dlQPK-p=8R1hC!ubtZH&Vb4XYRkQJ$k?gTophW(KZDfwp}Y4L(AIOh zjYaDYDVN%I?Al(g@zyw_3lDwYQ3uEXOX~H+N{!ur0a{?Add&Is#TZS=R?jC(qsI`C_t$0rp zQhK>^|BK6N*{&L!{lDSh?YlYN+$g2u;MG2fMhi|XAa80Q&QK3{R1-zZe#&G}RFZNz z8pY`B*JjoB#ow$Uf6y&;kY#Qrbg#Tg|NDhwl|Wm;b;u$#%!TVIN2M1aIzhFp-oVhf z0$BGyJ0*rG3E+7q1={ABm({I8Xl1do(@@Nl!$9Q8VWu6|!9wy3FjK(2MnL&PXU;k$ zs29vq4|C1KQ+s95<~=KKj3$ywyV(yI_rY3ATpG7|k0%(>a5)*H5<9?PgPPfiyfti# z4Vnr8=eVw_pBvp0!v11>Mgwem;&GLCo*i(;mrp9X}nr>FhqW^81n< zx!?FJF@O3?E8UJunYMrzOzX&YFkkQ1m2X26FyPZ~Z>s1bXisiw@o=b$c2Zb!TaUL! z;SPrSf$!u3Fd$KpnHTSf?KRUl9sl51X&87Q4(QRGA2JIP5B*zNBhB6boLkWd*^QUp zc7}<0*Aj4d3f6N;`JwhE>pGeDq_*TfhuSDEM!SXkqd(Ey^COLxhSDd6_a8mJ*u%tx z>wHj11lf;0-p)d5lsoWN?P&TxtY7pm{OS~LJjn4f6X#F;CFhvJUwje!EfDYBcN)20 z8*mY(-$aX~uq_(ZS7cA13B(@%$5MjyEheLs-OJ#L6xeClD~l^ z%TcyU#cW0`v4oeUkkP?QDs8DK^x4|G8;@+c=lM6~pZe zsY0)owhKJ$o!Pih&}c+_96fGMeE*T^_b_+Vj7XU>zOTnh5V^RTEe?AX-8KVS{ z`gp+7r|j9T6rI_w&tUyF(}&R$cUGnT*|9?|J1cCeAL~ zD6t@BtB4eo0~aX#XpDq?;{4nI;xaXgl=9=}FPw~rk4>(xz|vgkPk&vzJWE@udC?Wv!nFn3}EhEC;M1jpe#cq~pp=JmyJu>jMeTHtFMznCW3s`zHGl(V+6?fC=UGhB^p@a zi^75sjX;9Sb@qayBJtbvc+|-@J=GkvJu`i$?b8Da-`=76+B=Y;cKPo6ZY`jqT>oBT zH?019XEY!Ni+x22-yfQ*m-PO7A)mLJnJ+!?p9P7#5H*pYMNR&K%f^El)94 z`@;g}?#~T-QQ26aVwPTrGWBk0DDv0*2D2K?Cx14t5WJSi2I6m=#lg@P226h&dOi(8 zirHk(fTG8|$KOk#_R{B`qH5_cxcT0KuIED6+BM%G-VQaZWazp!W)8Og_rLTgkHU5c z__zfeS3tx6kCp#>Aa)g`*j}f>w+=y9g+ak1R839S=>`~}TeiFm9yuuoJg^NO;dIMr zpk?OoGmp#No0Nm_ty>1Hj5WAiZo8<{6v|cGMu3Nk+rLR|f)w*tG+cD0@iq;zHpa5ycjh+m zt%zENF*Z9X_TEWgq%~e&6Hr!e7o#zYpt?k2o21o;xHpZyqmIv4cy3X%_AJ05g`Epb zhWCgm$m*|*wFZ9vb>i*w0aoDn4q4j`z=)|-e&j3-WZ8y%;#BFJk%zp~k@P!dBC11R zQ2XR=H{gwBA6kUt1|Nk-nGpAXWN zz8=K`bHj(fUxTiw7Peh;u*!_z-HtZ!L9%pn5~B|6WwuYQlQ6nh#KE~uXnJ?usWem# ztK=s%0+ev*$QEg@E&2JkP4Fjh3vbIm>5N*JChcM?1Up-Q7F=P$&|Ob&k|mR9q!BcS zPz|S>Zi6fz)sr~1;W<0a;WAj;O{W?Ffrt}e1 zW8sVc#n7c$}*WlAaRPxzlNv#;zI@u7$oFq zWWL`jc0*th7Vlh!^(T^K+<8_AqMpG@8fX&E1Y~Wx22=l0;rDhoWxHU?Z{63xnFy#I zryql?<3rHo0(Xp#XN|)bKOhHia;vE6 z;Y(EQ@COc)-=J{JbhT^_t|~UK{!1tI)!F@VoC0db^Q>r;S$Z?^BFOTru~o_i=91M7 zBFs&LcP?8>O2!ghaGiNrA`V2{pj$?;gkt93Lz@+>Ixyawr*Zxi;`@}D;C1hX;ys`* z*Hs(---Dq-!4(2w)vO0>F*a0g5W!aawol|795R~|tYN6B$E;zY^Fc2}Wp9SbCtr5D z(htKoH-IEDUTO_o)6)CBj|p-W^|{PJmJ8b(S?{wn+$C4v>tN{2Jie}KB|OYgVW!9p2EAM zkEMRNm;+w2n@Jz*MMxzKhlY=T5puBFUa)C>{(>g($Q(7-0rqANVl<^^@aPNa(S8o_ zL`F&hX|6X4co$$m}ot1Cu*Oo1dz1C@| zA|r7%XX%zdtj{grvwr@2c$gXB#N?kA3ao6rA8pro>rV1!VbQOvCkNGo)3%Co{TEJc z6*->it>*(@a5!+#20YpI9aQTDRe+W5MjxTti*%;(!c97145==EboI}FB2{f{Wt^E zKGty*z_C=FrXAqo$K^iin+lIm2lj4Sw;cC7vF!>znc-FLW8hZC%51whC|FlNuhj=D z-RapE4Hmky059jSUNES~HuA14z%SbJ-tJKjvesssER(-HyxedTl3OSmGX+!Cl4Gan zB-l_gd`qUXQvC1-_*Hv<`y6i93#+ev6Kd*OO>**LFG9!y2tVhgafD;tQlwMT{yHMb zp#kXX2W1U_C@#WO$W}&4_T)7RUO&-z@;VzGkDJGbCx8V{&i!P_(cO|m9e4z}KircI z%(U8N8z+HgPo}(ik)7Vnl)`~zNY4#ldm=b-R9q=taTTU2s}r!|3{O-WrOP{QC;ga{};e)y#U}djCXbnWB`rf~`Ab>-ErntgekBV(mv{L?o3_C9& zW&*#1VT(SP(h`zB7KyfH;k<+<;Pu-+~jD6S{QkcfV zk))5KIZRxiPI?DO`{jnRkACIQTTW2X{@kLTAgTbaF>O^ofd{UJ!f1yU?Zy_K{}sn9 z^mS&6Rim7;v~N((S&Nr>p*sOoKQ@P2X0>+E!ORT4-RCtK;td-ZH0<4K%c#4JcD%l~ zZ;npTk-z?E;ruLi-=X&xnVXU_wKpkr%R`Sms9p%jZlB1i;MuV!!>@psMCQq$B`|QX{ih0;tQO1y4cRJJR6s@IkC^B5{rIsRSsRMc z!0TT`H*oTaVc!iDHnuT)BF-YDgu2}ngfsB_-23tibkchfa;wBgk~FwRin&B-XdEzUtR>+T?dX9(3C*jPc^{( z3gon)gz6dt$y6pTc-(mJGN4@oeyhQO-<>4Dt(c=$8bCH+%YxGkow2etq`^?CCM#N_ zIi6A*knVCp#~SR1*35(ZC^1iB4ya8U+CMo1HWx2pfqU(!cnV-vi$sG}r@gEIbGGpW z;pGNc$m3d1{|O$~o?Nt42ZzPoPjvz2l*67rXzY@Cci}Kl3QSAB=7n|$2c9@`35a$@ z8IYkfPfo^c{#$ql{!RX{Uxl|Hch49v;N+%~m#ARPN#@X=Wq5&R{~J{^zx7={AEdAX z>m&W_qXBz}&}GoLW^r?#RE=H}JR#(E8m#s?tNl<0tJCKN`%twASR0(Ly&;kVLL(14 zjO!sPpSb}3<^)b}jZ_b;!*g05E`s;f%hnQGoN0fcC3ep?@pzWLNS2tMF8~(aD(3~= zSK%o+wO<6l8w(47bVTUJ*bEJ`>+_Q=G(mq7vjiWN-v43+|E8+YoJq9VbC~VP9N6qJ z>R;#n3I(B6`MzQR|Fe={4L;2bQOe)J^&cpp!mgA9yFM@zV*aH?RP;fiD>wIkS(pO> zD(PZN@aVyf6}>Gbt0YE|&Ppu!tEavVSw5+aoWcz89YX8egRv8&d--!L=h`#ZNgpU? z=U7hYfgZk$XkNwSi*G}upNH>U6-R##8-uJ%xn*ZS0r&FvTePzOd043$)rrr4JB#O9 zOloUS|!oLp(YCJo9{f#cIENuR?kuz#Lb-tgSR(9t$LK`??zsa(y z&Zni;ptGo2=p$?#h*u!84I`);d*3ks9W<)^=wF<_gwZ+;uCZfHn&w-pm~5luf*7kQo41 zS74bd8ftMzWOabgG3!5KvrzU>$+ub7H2x;JZ}BY}tj+t@`W?^}DhaZwe3p8JPAUso zyKn=oIqLce)jG=kOc$xb=SV&UBw^}Ec(#iRyj&4PW@ zg|H1grRF;oY?i?2!vyin+-!JBU@(uFVA1_W&*$@hkoR zJ!u?XHDl-@tdV?Guo@f&Le7Uw)BnsP})mJ<;`v^mDODmB$)>D2MAB0dMkinAJ3dl4@2T?f$6Hjb~f4T!VH; zSEl5VvA3Mb&Cq4lIT*>rf$hpgRTiH&JUAGjo(Hq}3x0*lKcU;Z5(OhW1J|H?j_e!i zv)E84Ve~*8%P1UfHVUadgBn*uA)}PycSyorxz;#Tc9&CgZ!5GE-65IJ1SQo0=V-82 z=jGa4{40dxih7#hyr2Y6tJ@el?i}o~IU^APf0R!Gyz7=185n}bI|h{GJYJo*TOwub zNDh1`JP!N;h5a0+%Y@I5HsGU16K?+$%T^&O`$st|kfUJuJPeUn??gKE)Ma0lRr^Ml zA~D{t8Boqo&&w!G5*y+U`MZe^bmTig$6qVzh}B1_Xo1RscEH^WxZ|bJihJD4zy(c> zD*Nwqys~gcoQWb_+0FV>Yva|8LxD)cTEmp=BXcG^#A?y z3s?%?sS8+UAl_lU5yqcDS9QeG3^26y<^mRV*LXPH%ftl?iK`%(JsQ{JmGAjP58Nqq zjXSp$*1}uv@+_*mwQm|6O{0Yr;((So9>h=J*kieDgwdpj^Iz?*7|mA*&dUpq3p%Jg z1rL5*XVEt2Dpv*YeO#@rr!rrP5Ry1&GSAR)@>eg@;J-VY9N1sftPZY>HCg(TuI9V1;@Ll~Eu>wTq z4CLr|$qhrvRD|RkGqXFe9eDITS8Jcc&+O};K3+;n8+s$D|EauU>C@9blcL=>WjQ@s zAuY+U9~U>^%ISWd3GnEC#8%Ov`?2aJo~5)uxIZtpY<4~say`U?b05*&U`J}i7U}3Z z85&iw&4s)1^L9!XfPlLi>p#&!DjVs&S}-5j;Z>l#4Ph-lL_et+0*AA+z;EkDhiq8Y zSdLXV>1Fxfuv@CZYi||w*tZUkfj8*L8oaxMe@%r}*4>fQi?*@xaSVE(RR6y$Du}E9^ZHJ%;XpkxWn619ygOFm+X?^y83eT&bw$Byn0`&Pz4-Ps6_yL{`A>HfO*~!Gz{drN=3xwyo`Y$Zt zr*@s?Go&Ca9P`;tMAZ&Nq-TO7ft%kkKrR+ZQ3abudzF2d;AN!0 z05ie+vB^aAD|k$^;ERa80}vot{eJNebS911KS7P@m?GOEl=4va0!rv@$WQzMEfl=y z#=pwDa)gBfc8dAHdIvB|*Bl15xArE#1f;)OvsVb&c}#D&>2+|;Wo8!gD+FDi>8oXf zz<#{QAPks?fr?(3aA&No42&1Ak?ys6qE+YS;6*Ut5uWDdg<9IXUdnc6K!U5?_{Brh zM<_@P-=o-R(9V1F76b4eykm{E%{4=5YJzFyhOMY&%F;ui=vhfESoq&bzU*ki4MLkX8(Ratb5DU7_fPlSLVVja%PYab!j`GbEj;N z8(M9fzHbdmd;SM;ubb;2Ljvtk87BVDDts^4^EvA|*>$B8-%x9{XA&a4 zy92bv0Xd2BHH3{b2NsL4IhgOJdT@E;(tuj_i;q+to2J(wzNjrY^8Kl1h%^`KVX=fZO0vKp!2C^D71eLmlY_6KCY{dY*QWs9{Z5LdUZee zgJn^IpZolGxvuHH`g{>OWH=Mk17gj-7nK4(KSj4qII;F2Ks!$YWsqZkX5 zdQ-1){DimA-Q!hN*>L0R1^&-`4}trc^5jEyqLIF%nS60IzK7>57E#7!Gx)`+-M>Q;-%! zkRbu@`ki~6rAOx0cRk3KB2#(VLIifhXGz823av~8qyJ%ArJ0;lyLqq+7BtfXSA7gyu_IIquS$%YxBv=013)=%I=Bk{klY>``vetnZZ z7ZOjmE6v|WaF2kGRq6_>*ii5}m#^Rcga!*QzZLN~>4qN}cx6uwPwdoVCe)qqL?lfK zxon7gF%nZEn|s+X`$dzMTKkmV%EuKbjc^;OTA9yZv72kQF|DSiJrQXHil zU&$L8^zNaPE&Rx|w=0&=YvXXF^w=ta+Mmi!#@;{7Oy_Ulna0l0@UnVf{qx87>!b(e z&t7*SrXMq}JH^94b~aocnub^?uP?j}i8$!ae=Wlv1jx*{H##F{@3BDjl}AYNUNEGm z9Q@1y*zPxyG&i8(X5Pt7z(>u%j4gtD@!ZqgeI{np8XQsp=jD9ShbD5qr+ZNtzhfqL z8p4)|*060u%fbK(jHt99VhL2zcSEC1*y5*^lPelH<~L1-+^Y!A%L!54cPQFzbUU5F zFXvf38ga)y{tO7b_;61rp5a@uUUi z$Q_5vGhyA^wxS&qn3>o4G|WG6wV%~E1A@w+cgwv!O{1IwYi~swaL~%4l@v(~TLJ2fjZFL5xX8dr%J_oKQkO18QTbc`VpE=ezu#C$Q?%>64!qVKsCpHlI ziVsup02mVBjfBVh4WEOHpr{fH#vVOwS%ZBhLKL0E%_Y#b^Xu4^RktMY-&<}tyICEc zcj%zKLl@pAxP{daRi}quT8z`G>@U^0)(?FxPp#T*S8< zZ_qf@A0F_Jr2J2DF@)^Y@dv>y!P>Q)u)((s7sx&@0UQxZwW=ZPzV13e5n_;Rc8Z-4 z1mqlDOw93O#FDeJrpd9cVx|L_&DgodVQDeds?1qqTkBUj3|=bVf3o{1Y~0lV%(S`& zMtnjSzORL#YB_dAjc;eMAFlrb`xUnk?~*x5lLn6~Tr2y{&{Skb6&Q;9d3`I~nGJKd z%bo#~^$kB^e?Rm_Gzo0wI3b5KRbqw(X_!Tup~km^52Qh$!J$WU7A8p`SS$B=n((v!Aghdf`p`mqK>+OlNds6CzWtn!=la;~`&dt3+ zLob~8QYpMNiFkcEIz9yK2VSDE%7W{A(E{m1!Si1Qk(34aOVkdcZpzeNT>tQ1XaBoi zcjZxEE=}4?2a7KJW4f@>w2vN2(A7xeSOAwm)8B`kUY-(ircrO?(1iWK@AYQ!x4uzSf_(%$eNO37B&w#S(|uB98(U-Me}y`iTt#$G>%YC#RDLI! zFCO5cI$IG&S$19@K6teFCrVf%wgXDZ8oQ}j{hollwvtH_V;$R7J#~9_LNafKIPcv3 zzJ}8=emuPgKIx2-vl*w`bqx`fi(!fLKtVB8W14It2PVUXaiHrwb0T!j&&<`zKLt|o zGGM;(;hLK4>%?cP_npQA(vDRe7+e^@V=gl>QLst6X?zezETG-LhD8HM(zjI1K-;&D z*1lK3m;F@3SETelN*nX0(;(|L2Lph=oH{9Se)K;&YmtoJfYP>=FXqBT@J=wYFSZlH z$d#qdLUx;E=uXeodM+fg{OBq?&UJAfIR~R@fAm}q%A|SpKeogfOGV*2D8B& zp)J06_2<^|{&2+j7@@cCvPX*-USWN%98NLqVuGLce|?p zJM5W($bW$Cn79iqpw55{8?@RzJhTk7NP@3Xd3vTc9rp^T+ZH zB%{n2wln0RL1lNBaIjyu=+xOTsJ7O~6hotD%EU*ZS)#Z zzp-rTkch-^YRD>_0kK=|6|bdd;m4=6jnMw2!*U>e%iUd6XN0F6&+uL)1m2PcM>o)O zC}Cb`4pf<(I!G&P%9a-P(4k?X{Udtt`%~|-m;k-{?vjo`Kiak)=Kj1DilUHI0X={B z)6d-yzJr^()Vz-Uz5(=qBecJ(IGO)>zmL$TNdF0>Lo1C=PKjLvClLMvt-$@g?#*32 zaQx;9+{F6}aL)sxwZyS*{O7I%7yNXXuylKJ9-UzAwtE1!ZOr`dZh`gZWT0)=|C0dc zW+;L4>#zgYxqQ?Mc{Ry$^KXXUkr1>Es*gt<$?)ejIxt5@P&|3f?nY5rNvlwu2{obc zb^#G%^;CGI!pXv(TxjY7#SXt$~-KA;g5GxDiCCiXVj`9&V83jUfwi^6;gl+}B- zMe4?uQ5sA~1By+b1>~P`J<%da^F&npnhqmJ&SopW=wW^Kw_-mN?mLQ@?n z{a!PY+^WxAX<~qNQcCVpnTe=eW4d3OrY6OwuEe_1IKhNrag4b6{0v-_mwWJZ>iu3? zo0IIBms+7|BW-JrA9-p|=$|=1TYBl)9Zzjs2?gPFp$3^3kLmgEd zkGT@Wj0%X6m1PV0MPkHIV-nTT{WQhOJm`wx0Hp?prSy4ye*f6D{{3b9xjGY>6uu-A)9Qs8iiERY{@vsJQ)1C}ZA`^B=caj(SQ>Ax zR_In;kZuv4!JLlhv%p)uzI3%B2VG?P)hEU8Eg`Pxb)oumAoQ ze@b)gQ|7TKzh3JqapUw`u0NfNT;83N;Pi(t6lJMe8Oeyo1*G>;%93xM+e9-=s8q*O z#-xS(7j6uTBuzhtxwjL$tRExvhKpf_v1LA%Vq>czE@)J*?ad*V6H8vJne(5yq+f)u z;~i_i*=xPPZS~ZInqG2ck5T>FDayo%MkSYQp(%YfVBm9HFJ(zg>VExc6KY`P?jB0n zt%Tiz#eT17i$)ci&|Gj@hOb^PO+W4+tc?465<~1;pm}S{QE5&$^$t#gB6Q#Ehfn*quv5oR5n{ zQv5V3Zx@<+za!?F6y3ElR0&l(A!9X0xvH6xlB;Z+(#CVewRdJdNXkg_cS^2gJ@HMd zMFgx{>1jpryI*=vL@vkw%pW~Lxc@_UO`T;|}%l?gpW@5_1K|Il1?a+~PMOg&ghv>!O zR%nxlkMV4)s8L&8lX4?fp?8=fonm#W#qYt*XAs3FC!*J^9`Twv*CsXE7-0cxIi9jm zD;-n6P&AbwmT;aY=Y_kIA@8p@OV6p-&*n=l=0zP31ujMS*Cs`xi+-Kn^U;@spu_Pt z{+ezBNm>Odxq=T8-=s8Qc8lHL&QK&3w>;p^xyK0oo*D07&wa7vj>9m;?Sn%B=i+oe zSw_gBlDN_lvwCaMFGRzUFDZA;FInrmMO#KBM{=&!GZ-B$-PdLlgwqg~EjP3bvbkb# zz}Rx&zFvzlVfg-Wn@tU-=;CX(oqoN|=J(Dx7fA*dk|?#$&x!bJ{$MH=Df??W?7JB~ zyx;3#Z}``V+k^4X;1}=ui`BQtm(uoB4N|r?%$*Le5E2d;-ClI5&lZ)l6ErRCK)db< z7gBHHi7Q+?HCR>rAd<0m(#|b8*R?2t`$qjBI8HHkdYHHN|8m3Y%ZMumXX1z}bDpP+ zIeAAo+o1ay7BNwB{m#j`(Zv^y0yO3R)Pz{Hu^%M&`CBqNMfYk#w~Bh!zguc@X;KbvRa?a^ zOw{q$E#ty!Y+|IG{k zi+dLue&n8?G_mHM<;?TzEGf8IFWz|~M=w!!|GrdDpL zzrM#fT}(N&=)1h>da^~3h0uLQN6uPnsdn_+9P|_M0e;`UIBkQ1fcia?W{k&6XKVI@ z{8G*D4>ZMm%GhNR4uCr^HO(04w8y)=)bSO$s5j~sZF!s>>qm=-D`M|llAEA(Jxz+1 z&W{P)CMzz!$X&~Zr!TC?=$ucgG;$0wq`!6oVn=9gR1QRQ~}>F zhwdDyV+GIi1!_wgv`x54`cV6E9s8@9!V_yUrC~k{euRD0Y$fR*c*)^8V;ppI=BA8} zjm)9;3Ebq@GilPqTV}h?CG1e4oGw_;Cy#4esI9iFM}D@kilf}OLi4O<=*NjPlOlJ^ejc;>mxMVqf#1b9bv$xw zTDWk}B?hS29PhEUrEG8c{%g#?ZKQlbVy!s*b;O19L3L@D69q38$nLs$i%+6g%h1qW z*vqEG>ZZOx;t8w#z`J3w5j+tB&ti>)_+)veH~*2x>gpuaB*aP8k8#yo1XbTk-@zew zsym))eOvKOPOi!Q;N_?}KkcD?p(Vt4yxIOcB)(CuqsanraEL>e1-DCw6ZMV}Y z$Pv-wZt<*HY{x4LU%2!ZRD0fHx+7wluT}l(Dats4*=u6F_**vP^yTB$l*Vt( z@zY(Qr<+Bp+@6KBb$?p;`_>Bkk~=68@0dF7Z$2ZVGcDgZbNB;jZ9W+WPds<;Z7Flg zPtup(k~#g*LFH=$7${l!H+Nk*)VQ$^wJRCs6Edj}c}Xbl-9A#wzf1ZAvA{{9fL~kn z-^UxtP2m~OYWA)*sAvUMgkuFWn`prc*zrHGZprTFaFas5pWGjc+n@wpbTxgy!jZX; zgnzR+?NNE_xMqgD0$QiQVeiNAT35g;EWx~-dKpcnbgfUQ^`U=8G_AAmbETIDpJz<( z4)@a*%$YDQ^8APCA8z`y;;H{%mWN>g3ow@FN3wDEo+!<${LzZv_r8*5Z3BhA*=tyxXWKX3<$w17+2 zqT8f(=A^CEDLy7ny*15{@}=8MGK;G(%>1TEe{Tw{Wh7|(CT_{FQP_f?!v5VlIlb2& zn=C;7uF^V3bH~0%E=f}+OE7}8!^)wOl1WZs6up8P3arsC&i9u1K)6U`pXf(m>)I=& zoyj6ipG0%{Vm;h9QXir>ku9#u+%Hd2%QgYx5eD9mm) zZRr6@BOFSu6?$*Z@fDE zn0*5v+5HTqfd}ykGJkJbz@K+YfhcuwTQ+I{9l(9vjS8fQ3EK20nZEMN9M#*mEcDaw zu*eUJ_v;&z$2CRmiXZN|F75bYRQ0SjDx1AFX1V(Q{HEgQbXt-#rD^8y60Dlm5nJ}CD9h{*oVKrU$q|>`;Q-g5TbB-6rGr;M4PSS9TUEX$LZu=u}UmA|k0^;MFn9CL%^V zjdCo&_nzrzls%wHE@#e9L0{-befu6+qMg%;LGX&?VBiKyuShzbRv-?X4;xATKa9Nj zJ5=u>j2u%8MPd z0)3g;PvPR`{bHP~%NCSfM@mT2SEuH97EDj?z?qj2kU?MB8*NbPu4qB@ll}8 z`_#{d(C2rkAB8C{*>O41rSCLfaXalJIRYd1FnC0nu9=*s8C$Z(!FwUj)Vc3GKCe&0 zlXa$|e@%s=Y}V$l8$x%)b4&+UT)OQw^uKy#3ub;8T3HQJXUv~PLkl*F8nz?swrr68 za;5u`y+7~n2}MRr!~^Ewb0wfL+q-DEPpky+sXlrC;w$X4+Y%qFovcy(KAn}-P#4i^ zWqMP^X0Zr&E;;n(A%{0YNgC|KJQ~wr+adDzEbkb^dx5i$#z66B8X}-Wgfc>4ME3`rqp?riZo5w6 zR&|!h!u$;;r#mDJiXvl@zodlkI$mMeDViSuyfpKdj*RrUfi>lTJCCF3$obAADOa0>6E2lU>T-5Ai`;DdYbh7PKf=C;`FegZdm9>)9qd-2E5D4_ zTcN|4UoLh9H+UnRUb6tfcL&r59a@7oStPk~`G z0S8e07-?T}GLM9&Hj2xx6j?aeHL3meSVisxd0iXieg_@dxzLVW$o6z{G6+1+%;p2(U8Ceuh>Hec0k#X<(pyW>knCH+QBj1AG{9=z@u{5F$YGy{V=*>bZXb-f<>w! z?(3N%AUG!%;+W9Ct^H#K`u0-W{xDbKHiX~Sn2S8LkCYjUy|5i0-8U)7Za{3UHpa1n8_D{Ybma)4+ z?4TNnSjhZB>Z#cBzk?oT`ab$^*uX_IA1SJ?FR<;+A!jhuHUP_MDjM1kwf%TfVxdi| zRA6GLspavKp%u)(`TU_EdX-y%uGu;i5Ce#lD)eu;t6FE0OKs`-7nYk8*}*>bY1gNr zhVgXPHKh7yy)@tJlZS4Lsy9M?Ap+Jrg8N9j2jd2Cyr_*uK4zifHA8sqq>1rY!Lk@a z5-jTl$RhWVKE%FGzU{@60|H)wsf(7zJ`#|Zm_fG-Z5L_bmq7&IP>e*UArdh&h_ z!!Tr0u};ap0{?PthEj2&>LYAvRPVp>v-}0V+h61EHvTvYV-Fep=#k59fA4!;M>>Di zS1m{HZcZGl668TacJln(L<_S6v6x|8{PyoBq9LK~Pmry7MvThVz>WikSCdOw{;9dH zBGc{4y}OVzS=skuWe$}F{99i|zR8XY^7jo)k@KF+ZqHH^U8#Keg$!bTN8ck#Oq=Np zS&OQ#3o9@%IC#_ER3RdzR0&6i)#ZJMUTZ`EVNQ2?{tK%*OE4)`D|@vGJ?XHWkqUx+ zp(jJBE!W1N8|{}0>%*5I{|i;)EWPK9X&_HG2*WbSgG*$`fr90!$r$A-ok#z1 z?S#%He7@(!q%2~f0&F$D@4EQ@*uU9dJrkUNQN~@q2Z{@mI5fD4W(Bn1WVWJR&;3Tu zv6V7Y1-Wjh6ntR1md`*^mr*aHuP2=o4Dy#absmL(1GLIVkMozP)v*b&y34c3?N)|P z|H*P23Xq&pSx7EDzJ-pwD0FHcE3R1k22C$J_9!a-`U)*7J9YUm=n`Uh@dkDL)YZB- z)UpxJ*H-3W1sI?ILXQ5!Tc*2gHnY&$(9{KqY8L&c%X#S@F`wP@63Fjw8cj32^*sfsrGmxq<*l1ukY4^0Ht(h#k8+qcp3KY|RufvzqRl{oEF($6t3`eOJSTr78UFT&tQ zQhJ(CE)6|(80j&ze|qvibD%8i@qyzT4RPW9mr>#$xfSk-{*S4^dh|-urrcCD>c8EJ!d00nI)neNs#M`gYN7m={ zbEH=uyUzS{&07BP3;7_l#{cSfKtiw?u>4B^!zA^|Ku;mCJmK?d9%0n@fKeLvNykCy z>-{%Y?j@YLpjot=_C`L3y+Aean_tZER^xGa*1x~wv}sTm1iRj0af-MhBMb-NBKu4N zWiCor-Y;~#!2(jEuhC^l>+;mH=hZ@q-CiYc7m$b~uafK0YE8{PLMMUisT=(}LN{5S zk?dOX2On?6gR-_Dbv+hfa_nT{QjTX=5h!|hsHx)FP&iNX<+L^jnOI$Cdgod#%{AM>2Yvbdoi8a^lulXZ(pi6<1x^r`C?cZ*QJH zc%24LF75K%SfQn3J5%q0dkK*TqazQaibm_D)3XT3AlZnWpYXVMeYSKZg>BAt(Ze8Q zmHoM{aOAOU$xm>luem0Dyvp;u!9z`UMY;;x)~HqOJn~wGZ|+k#tUgpU$Ie6mPMZQh zL|+}NCba4M5y5E^BJSo5jO_R>CNPxU>G;%Rutsos{h|xh1LH7l{y)A_q=|N(9*n~V z^OuC0m-nUM6Y_=luB2r4_iRwrG%@dEb?EtpIo{)kKzoBp)KS18#~nhc6rTZ z{GNKWprXU!5!;)4gEs<%=iP_nb$?XQ%PIz0llp_Ie4kb+&Ov3hC~xet?i%D{?9l3>0J+B$jR&& zdeEESmw0`EC3>T8m-494FAz1dd`SbMUjKQxw9Vr9mIqJtw!6r7PBSng#jStWtlZP^8-=mNWmd>$(S1{$)6+SGx($Xg_|n=kd^rp$7ms`M}3MMR~F$Iv6m!GR$eW*^ ztbZ8pBLV4f-!Vwk`T@JL?~nRt*@gqOP&|^k7c00zd;51WJFd4zR(M&(6tK=1%~Zte zn!S#aqYPPs#B`V8pNk0l&y5NPcF2(I`{v-fCVvR$MY`%f(mD9T!hzipaQP?p7$bDp zI#vCLrYtuZvQq6B$$ak%}SL>Rw9|YL%V^xFy-i5KY7MeQUV6sMtN&3Jp>(AGgG3Za% zH%TH=ya(+v%_JE8I-UyW-}z zNJQHYj|OSW-Xz2{nEISM@V`Yqddp3P-^lGU+YQ+ymD!`(R~Zbi9b)G9gKMJgnsPb~ zjbzc;;&4e8&_%WgsHk*@LIjTCUE2I)2xy9ij6F|Co z{BRMtI?lngjYd#zzB)xB&T1|rg%M=-pQH+bM&y-T*~;5 zN_O8kTcrhK=WV*X9@F#n30AMX3Lv{Y?vxu1w^W*a&S*Mn)l|~mv@TTd-kFn?Qz@S$J4|9RblvP=;@vZp=-!x4ypss%(ikzMVRMbO-y-Gk@W&iQtSl{}Y z^yiD%u&?k^pY#Yw`Zx-7N((ELFVf_Tw(YiKrSf3+u24kB5Lk}SVI>hmGpI&Fax5St z4|}`rB#QSZ6??~4691FoR3x+Uf0Oos@7SB+SGvnyDu261Ka$Xa-nhqQvD7jHu9p8k z4C4}uZzXHU6m-GkVj2^9k;}fE!+AU2y6ElFcMyKBIqyH+Wu(651l1^XDBt6xDG*@B zWZRBlaDnl`4n3fL#SNi-AHX!{!A)cEs5pG;@KsRH(ZWZ3X`m8+Q&g{H)wj`f151C(r>6lXHMRY%NP=NP1+XxgwY;zH{p_iKrxD1WvgVE{_h7gM6kWfoN z>Ij8C<)WcUB5P2$$5ftA9robo3g$*4I2m)5Im_2pQ;?kwl_JC&633w2^K1-Kk?0*u zv7oG;-G=@goBIH;pO{`NRsoV(A_>t30-in;0QIj60YjbqOhkZREcE&IU=zUhYP@9c0!?LI^2Ii z&-s5k0o*Xi^lrl@o*|!~V-R0n&It@M8cIWPS5P`N|0r)A2*ct#CzSm!FkJYS(1A6D z{(xw*p}3b#2JO^%p2Fo{#Q5&C)VS{}9?3cQFW9#@I}KR#%t=_+^Xw)6m$VQJCKz+^ z!sHm_^w(#BoQ@@@ejFu=p5CY@9rTs~!8xO*;ezuNY$gs(IN%VwWzt-B=$HEPZ(`Y9 zTDZ?LIUVg0Kn=xO?9&P4pySlpYzYi<5nX34o$ab$VcdobD_ou+%G@Th*r}1`#BK9( z95EOgH9lrIJ&(C~ZrglKh|6aD936toj57VppHrPj;YZLhI6FyHKcg<8Krh0j!Wk&x zvoB&WbZugYMXzg;QSy4opHS$aw*uZd`{-c_pglrR2K+f_@GYuc=xP%i+T-y0E@RCO zkpIZ(b|BHHRn+RRS`!U_kSocKG%CBd4oUeB z44a9&hbdxNpTb|YBAYO5W8P3jAn{r-ZZj0x4^_NI>D&wvPV8$a+&a)BBdfaIDb2BA z--5Ld?6&of(czf6pj$hhjp5I%ybkPaFQamf!i8tLHelIBnTGadF}AJuw(Ahn_|Dfa zB!uRZZ)yWnqbwd)$!C-p5_-^i9*U=-^#TIj=sJdFnNG*BLEcS{n?}KDd}K-lv*)L| z1|e(1fe4ut5dKS^)F4YLkjZs>e;p!z-Jjz}{x1EyC!*~D+`sVf2YkFnLf7a;c5Cy+ zvfF(J)e4ss#XW$Yw%U{3xv#J=r877({!WmiK7HBo%z)}M3e`39!^OAH7@Tdp8PBG! zPj@{?cX|DDf41w@`}Gj1mZgxsA8a7(7*zwiBLh>swqM zY`6dYbs*83gf6G#H>{<|qBONqOYb8cI^&8LTq>^oq#8vY#c%R@B{wf z$tDh4Tn3 zIsB$p{g>=Z_&i2}Wd1%a7+lz$V9;mWBOw7>D`9#U+VQoh+oe) zkr`qFpaXjS+VXjtkBH*3*PH`5L0MU+!5s;J_RrT8-39iA&RIayMD4lte3=Pwm?M`P z_GzVJCn@xt4k9In1|J>?^0MnOTyVrA5$cCx)KxJUb>;R=`d3^5;O}P!|`?k>>e=E?XJS){8E4rrP<0ln3TFN4(N% zA}896+HV=1RjxD#E0ym?V^{xKfJA;7Cgsj9DD>f}#68TP7iX&~8Oi0VOpY6USs^e# z`vVSL|7D0M1Lt47MTbv0uOYd*5|(V%kT{diI9!kxcWCM{-k)%~{{1r#SNvz5T3;}4 z@}GWsiK0`ZOPxM#G-o+!y;LAz5E=Z$BCd2@Q-+ z9>&(_)i*QnJdgy7VJ zCV%I;3j%5kclAs|!Jh8AtTW5*RV8ev1u6`!=JZhI>V*l04 z6A?6YWV2obBMag`rZK|cvro|kFe10;ybWDVYv^IxNAgo&kZ$e>3?U`|gWmlC>z3|a z+Uro?f!SfXNx&PBr$YN3RA7njRNcbPq8iIiv_bvLVE4M=@q7XjokX6OBd+c{gYTIDns_K~_; z7GUM)ROL6&LhoS9B43l>t$}y!4_6fU^M>3mcVs-;3e=>I+7kQ+UBF=ey-vh)8#fux zC|Kx~0EP5p$Eox?X>xb*3N+W3B{+1rQBrgp(l2)GAsya$*&8YF3+b~Fr>(YyWwtw4 z1kSX9!Y5bMSIM=F*J*f=@=a2;zZMUCo5vb62uRfl(tVfgA4sg^Q~FI;+?(`yE5>*ln!HG06lX2%0~2X!Hm z;@VgBpBimgs(l;%4Cw6$#K7+9$yhL1t`Y^w5q94-Pt2n?P=O+~B5=UN&4@xUlWlRG zS2w3wma;r1XU+gi2qN?1MY?BAlg!PiP{o|ip0jO#)T_wz)ionfU&`|lVp&$q@Edwr znKNH2@&p1V2$N2IfZdy{A0iT6Cjb?PoVe`@+yg~eo)p@X=4>y2|4$Rfc7(O4ex$h; zllOW=dS)OCFsXKiqN4I2kkpcZ3*WU24DZvwBm?V>Xv#bCLGKdyedh8n zc)WH=oXLNL_!7gX*uII5CLL7L(6i!yEM4hw$yd_*?14mUw8!LsBSaUiOHH-GpH}^i z1wwy}TEdk)g|#h4Ft#OD&UhVS&+=#0sR`}ej8R(wA z4>;V3?OPS(1mN?sQvd9MoiI8PuYr3-_7?2~oQJO<1&YrW{JhG2OsNBU{kXDQdOA)^m z)Sj#&`Gu^dFGWN4G|vZ zi9L#NBT!?qS5eHNY1MOJfG9Jt94YcepPtXe0*(rO!r@X>#Iq~>DjU9K1d_e&qkrFe zw1~!@{<%-@?)TZQV@3>ZEqt~$dboCbKH?U=>@3zqn$U4{$_HI$`;OAtT@0#j_@rOF zgTc6b7EUlZ2`t85&!aO6I=&XM|EkKc;JianH(8C^;BaW#ve24i$}J4k3S*Fa!ZzCf z@H6gP6y)RW(ulhLZPyht`QU}?Q!aYhM>ci4CSGx97Q+#wPZx`VSOm{)Kn!^mXKqYb zjMWXo#L!N-`^U%LV_yKLKuUx}{&*kOCh+;*M6bc$b_1SaqD&ODu_(7SOr&sxsoSV^ zSs21tA-M#xD}b%AF}WlE*Vi9rIEl{A*d>YUyEyWjZ~>1mMR0NAJOW|c;Pb*2gs-tr zc}~3vg^qQNQL%Q^w-;!Y9u>h3#h(+qsaWddOw18uAR-tUX*R`XV>BN{ji4COR@Q;v zootCi4>sdv9*~H(z8*|wn?{A_99h3N7>TN=E0!rBvZ(xC?jJ>8F3>%lKAimnlCf{6R6;xwQ7&@LYB~Ny<0(T-3=RE{&S?>?Z{2HCn@zcfanhnSTE5n>S~|H6!L`wp55 zLD`CsHN3qx&U=~l-Cjd|S*i~ygy%daOU7>dG3P$kqR*H67y;my5X0aTEOFRPt*Q#H zJT}pxeV*(ElXM@G9O1eh&s8gbA)r;MKTeq1JtfeYpWj9>79PAEi{B6b@* zN&xHmxW11>hQd;al1IRkxwgZfTf49ajRkW@+rYk^!uF<7HH%Y<7^pA%eTu}$^e3hs z1NVV6hG0KezwqQJ-BS+bB-p14B&iA{1aco6UVHrEW#GEY)*-v{H&GDe58;*opELO1 z-dbcOAdO9}lm~EDyp9+k5_5t+Va)+dAN?x$KZjIXmJ}8RHHV5&^+`l4UX}b6i2bcF zST=LG91YQwVVzUPUzISJ9PA}Ep#0b0kq&o#AQSU&c5W0_9$t)NE0VM4w57wgz@Be# zXJq&9pbVTI60XEHyuZ&Mx(#WZzg!+kLlZa)pIOP$WqQG5hxcd_+U@x@n(HFk^D_%r zpZvjSp^I7{?CD~eogW9G(U%l~50cZj-;+;;?>e!KV8}r^VfKy;?}Uw>NxkzIQw0yN zy~qu)C;`EO=ide~mWCOb5=KCJ;p3clG@Qi`?DvsIcO3N%#-P$ORazL#pK#J&Q#jt* z5cj9yDo{SRLgMcQ7I*awvivvNGp>ay_8s5Duw0Mzd~tX#1Zfs)By*K0U)k!T4IUL= z13GlrGM4i*geVogS2jnp_L%y_1b$xpXS;W%|3Gb;%IL-Fr7_9O-G*qlak{EdY94!E z(1cWm2vZ@0kMG6g(aS_zuo{;c>#&yoks|gV5cZ<1jSm;`(#Ic*xwb<~GuLPcGdko_ z9{ZR?Y`MO$k1cIQqK%}%VYBVHKEkj~8x(i9@r(3~+1tb_AadS~ZzgjDVs|>faDEl@ zMmspPoA8F7`;r>DH%LNp>d*#>ieE}zP8r?(wJ0$UE@9_mF9rs(Ee z{x$^jeYoPe;yvEIuOvOLB#Sth-FI=}t2j;waUqZ~rH;pLlk9Ary63d$<27$kfe*mw zO2B@H0ijEl4L|#ei&*hI|#L6{Gss8n`v-q%fB$k?>3-3(uYGN=OtDV>D z6BS)_w2~Q#$H-%4S4*AXV51M#e(wm(?N0`1+_|sW-)bjF5fh3`H_0N4Q$Rn9Ico0M}M#svEs(ak6M7LXAL|mcF`-f4iX6RqIM$ zCM(G!P%h?l9xjEAkFcJ!Q2N>aprczrcXT>s9^IuC`C?2q6f#_BWGe=%zY>FzfiZ8v zBPHU;wbRM|cj)QG!nmE4$NG%#={Y|fMV(Z&Mc%0+Iom&i}hg(F{AMz461#|b7 zznrJ@Y#EJO-#fY2`Q<8Hcp=qS_dAuWH{u;nC36uLk_4T%;%~ma4-mq&4{6MeBI6c+ zLZmm$q%ISOMSv^*>sGxIL)GjfA;?-lqTm1dKJ`u;|fMEc%d&`t*M-# z9$D{q_@l0F_380Yzdnk;A0-nhmF~XbA{#H2jtnfUp-W1VJ$jv4S&gPp-z!!t%V#R3 z(%l*_$vVvc@(!u2{65o=lQNy(Ta53z@7dW|Q1~;`YGYrwp6)2@9RCiw@w!>W)v?}H zL`P<)&jjW4PoRGWg~g`n(LM2a4?)71N>O$!!PdM->dUmNXT2E{_M89j2yXqi z6#3n!8Psgr6^9QjDbq>j{V5r!UPaYv{J@SoLUb7Qry+CD^=AEbFBcJ=aASKbY0`8K z7kupF49ZybVsBGsP0zEM<=a) z#vDudgr`n0?H7I^%)tO>wPDds;xrk!3SSVU#7BHjL%CxE^pq!gn+d3Q_`v-WRoVMF z^Tf+rMwf@gr#NFX@E5oYQ3~?|4}_I}{O8?vTmQv-nVSWH1^5C%f~mGcFF`2!#Is*7 z5Sq8AP=%%^2{-cKC2TU^^+Ot;TkeXB?WqY9YJMWX>6;x<-E$XN61adz0GilCHBz3u z!>jUvp#KGVU@WHbE`?iWkEe?72Jr_vtxeS||4bOvABB>jbl&Q4xUA#lZFLD&e!%3Z zD@A@1NahgH39)RO)$V>k)I}|QCj3U9$+w~{*+2F(?_H?XeP!IcaH&Ns*sK+4csqQs zFUh*b`lac2!s(!gH!gk2R^a~=Rz36C?elat34QmPLxYK%F<3DiO6sLz+ih8tWKy%@ ziq1gO?h+5a^_kc>x8K%3r>7jodDa4K{q&M@J@+V%SngH1w7&m%pEJNfPl!S=Q@%W-oK^l8~+uD3T9z8+z?tZ%+M1@~tDMHd2nL8&|9wSBCBQl6QT zoDALR)e~!-N9$=_sWoGZf*dLYn$5;-PH!1C@UZR`5Plz>3_Jr$sEA4H&uTFlEV;+N zfi859)h404E9bYCpBvRuK z!$a{+dnt16pR7Xs zm2;)R47bwwGD_UV6!q_pD5iv!Q%BSVV}jWHk@#q7_7a<<(A8#`Ildjqp>ZrWBZbLl zp_;df0$>zXNi)fcqw zOr`;BQ=JMX~wVqA7h$$?yo&pMhF4@KjYEH0SWyFfs;VoeW)?HPw2R zv{t+^FRR^~>4%C*yk@_OH2h^e$}8_MJ@(?~D6}nJ8&yFOJMH$r{R}sZ7UyVZ){yS} zzZV~L!s%}_6rf(dMKmWu_vZUW^mB3Xqo+VsR*P1zZuI=ChcBPR zd-Q#tskyIH&q#AB&IvxR=w)Irr5u6#@^x=9YI#JOe6rir6Jb$3;mEJCHQ>ulZf(?v z7jb!M{yl8=jcs}SB!eeJ9I{fTi!F`RQc)7qdD_zE-Z{EG-BU9cUO(GQLZQ4454kDJ zzx4kY)tzL%v1K!g`Z}vn*QrIdyDLJ#|BSA8egY)8;4HN#*Ps1nU*lxk1dz79k%j+~>Uqq90Ef z&WE0f6HfZ8y$E%IXLYlETU^!kGlQBy9~$5 zM4&4E>I;eLy@uusf5GDrpTeaaQz=&`uTaq{Sa?EPTdR$US=#Ly6~NZVAHER<_7d`d_>^iq@WI-Gxe&+sTE%ouMHv~&gph@ z>FTASPO~A`9L@oyBUOnq+(F{PqG4C>m~*(-6HK~E9o8DZsMxc93(dD)e_KQ!z5M+X zbx%EJ;z(L|{2-q0Cm6v9)$`ubig%f`jMZ{*X{D>HX@cB=a zgHxrtt$P0HZJ*r2W=pG8N)paVbF)?`ILO5it}*ya7>9~)(G;!$OYX-o^JuGloW|QA zBNeGg42xWyV|CWp%tdb8lSAUQTwS8y^hCzj@KXXY69VmLZ+8za22%Gj*CF;f32XuW zN;l6ielj zeDPcTRIHlvoT|1X^MTz8Vsn8gUSC?UJkTRv0!}8mj-X`zHyM4AI&zZX^pD3a*VEd* zhmjAMEq;kwg12`}nYynBU#tSoGM4!dJukB3<*zDl3n?bbS1nC<{d%y{!LUYE$SJ}#XvSb(=cXYPbl)x){jo%)5CMcVdIUCWkq}lh6iaqV+jgh zy*l*ED2DJm*U;zfGZWM^F5hFDoY_A@9F3VV7hg*z@G4bte@qMd3Rm)+_wJy}Sd^yl zJ$rV<+sANPrN=Lht||NB?L=I`p#0f4B03w!>rj)H!-p?3mKiNnY)P6-i!<@btz9!sh2a8Kd4z^%De+z2JA> zvjfuK@A3ydQC_;kZhJ)^f6JjC4P_`>?YtoHao1mqdh#~$L^a{`7&d&%FA55E7B&#C ztFKEGq?V9G3Pb(OLmEx{-fe|2X7%AN=jT{@q+?#;Yaf@j8 zu*pB^8<@I3N0x0}JqR&vO|ryRQf|yGN~gAT$~{J8LIV;8FVytV11JoB-0_N`#+DJv&t%7OA9Z?|4k5llFDb(`ltO?JJObqP?|?s zzVty+XN-#N7`j`;&-@68n@~&wbC4QMOc(j0f`D85fql zEH-9sjugt2!U%|50SDH6`AD$zs|bJ1m$bS+>NltmgCNxG$6SY|F^x%ehS54NOSEN80kl-^Q*t zmd-zy(3|wKE7FY6uaW*`H-b~D3uFeko&u8BnIEF)rdg)tX>DePN56pZ=9+%KY4frY?8AdRkA`zgX7;GU=zS2_WW4%TdNqr6-#6CG@hd-q6nDnCk3Ff!H`a zl8AHAR-i58{L4Jq6uY1svU@e)Y$qyK+pt}*Wr<$sWZAEjks5PI%C(tyS`Lf)V<6mr zH3H0BDuajV!>;#)?+>4RBBE2_XR2k7U=WbyrJKD?tJh=CK9)@jx;>Bv_@WOPM4NsG z)fRh`s|+RmmFj*m#`u#yOU?Oonaq#*Re&U+qt;3_;yXO^-yZF~8PJXE zVFXT_>nQ2^L0NiPoF;82`y=@r7I4)uhJ8EU^2`MgyQ$b`w!iD%kkJ0p6W`ruQ$3kr zEdRHe;Vrb0tMc%P`droJy}~=aO&hILY?=1(7fSv=r@ENzsoL9dvX=eX=C_YFg(n&# zk0l7bcEOKyCtgeXd?CaqJ(G&vxLZ9VA@aXVZ<}2{0H%18)p9NQE6P`$CCO?T&+`-c zLKKJXFQ-~VS259vCvX@WssDiR(lyja>j;ksNfry(N76g?`;5C?8ePV*ew_CP>SJGY z#g+5*77Lr<3;Rg9gl#loS7`lzdc+)qr~VDa_8d#xJOOW=hj%XMl`edzA=tPBOc-8+ zV@6lQ!(6VaY1$Ot&3!qN!*mobj9}aw=NhvE4Dl(`^9w=Q`BwuIQ+7Q$JMWfa^I)Ij zFqeIPS`dfIS3$R}4^6pu+Js|ofmdwYSxD8h()-BgR_w8(+MWXBg(tZL@3Wa7PW3)k z>H!uU3JW6d5>F_m>CB(w+_1c{hypK^eM7R4^?afzMggd09xdm<$wpSDRK5@u{r&(; z4l~K#CPsI&)}G=L2NwPN&abWRJz72AwLg%87Zc;{uIA=+S00{ISf~bmw=`Ci!70pZ%>GrWsip3m!wEA^(m7lr}v{j1Bp5|coM>lc#nTl5xfM?dC& z8=5WH*OYFcR@H@`^nxvilUxc?t`R3*+n$R6*0`wscb&-1*vDY4oJXQw!VO5!e?aM| zq_$0Tr<&_!f8P4~Z+BjHlf={HH!n-T&%GV5TlV|DT!xLj!Z6JHSe`f?0TAG|e7`sI z86E^*OJYs}*{4sWv)_t_o+ukfft{+!@a;9^f>6HG&3{mPD?wl+JS_;L{8ExtdJB6| zu{N9$(zq96_|=3J^9gZR-+}PU>1l0Q+zUh1Q+t@kzvok|?F6c_-Oxh4dxIyQC<`C` z(J{LYRiEX%LmPh1!DiCi^v+xLfuR9x6v4wCon*aC##kl*TKmh^2fvUOecblKJhtC# z`QZBAkw0im62-M4Mn0Ba7TlZ3c^b7~*4t!eoKT%YTXtpLI|V#)(;H!E%FMaVuAu5qhXE zT!MTh!G$YUcTrN;!f0qcR?#z6d=Z^{7`}C*h=0(dW|7M%&Cmt{=(0%>{@*KdLIauS z-=?+ES%g{lk)X0lRDwVW24vmn9Z+k4C(@jiH!E!aewYq47MFG)*T}#V(D8}$w=1PCv+X} z6U5c0_2WkcgB+1kWccjc7<>GmMJjdZQsRj>yLMI+f_!@PaRkTQl8F9Ry!B7RQ(+TS zqvTWFBqPNF*}2hm=*}Tj;CDd6gm>Q=f#Ob!ABV+ru2MQ>lh%%VpBRNCL4EH_SZYTo zpU>lH8Aw99yy-}>W;RQhmZXTGy1JLxXZh_gfAsNmj~s337J9Q%cXjN{%z~DVw&B0o zyD@n**VenfLFP&-{wYDUiH~=2KV6G&`1i52ZljBTwLrQ8V1MYMvgQrc)x%pAOQ(ST zRb#0XAneU(MFUpPiLSni)L1;#&lOb}#`9kx_|KAqCuZjNI>QV-4v0uvDz&K-b)`xD2YYSM6Po6wvvQI5YPMm6Nlfx$az*{ZZ`W|K5d+4l=5QI=_BJ zR6L^Yy8AEv^Gv)eBvXXNjzCPGUz5&uhUf{>DFu3eBV{j|(p8aPQdf|qpS zU*sekdIU2&a-ZLikE(i8!A4L2C(39W{;>66`uO#y{%1u|Pm?@uS%UheQ*a`Z_?u%c z<9f7o8gB4kn>n{KhuZyDVBI#g6J%-`{kc*M-rcSVc91wi%+YQuHF>qB{$`zJ~x>5jCE*+9} zWhL1`5#NB|)r64lPQylK<;KGs$%ICAMJz=pks|g)IoS?~=Rug)XMgzzJpio1P6u|J zw>t!{g_bq6tXU|3d0xLLoIn4a(y3{rsXB(z9^g2zvvHkfkh3=6*j>^fO^@AlG8uI- zm|;ei`9AgbO8_%9;xn@vyEFG_l_Dw{<7(g1+8!-%kXy8+yG~BR>8(iimxPD(gU*|B zW!7K8+2qQF$Cwr9j$mv^>$Cw_OP6^i;2cB9Ed;bYm-{s^b5#sSF;HnoiROXSj)p>Q zPtVl1_iplTVcXLmEqC2O9nUz!?yekcbL%et))f%>yeoh%lV@;4%0Y)j#4wk2$BTn~ zTl#7WP&+&!F2k4&vPT||1Y5~}wZ+4> z8C``!B)c5tid7sVw)do}1emAK{Z&14}@u#mU+v3ms%PX*WJ;4@neP^R-bRHd2 zaqak>xb#CH`)R%Zc3tCBCD@0lci|^P&{zr}Xi(G8Q_cfFax82hX2wI_L-MC3M^@KX zpf`@xItB&mv@sY)RQn<9Bl02W3jCe6%oDX=0^3Mif~zKAu&Xeiu;;mtT(>ocE19A5 zj0&A;ernpF-1omf{V<^G)a14;P_S+%3PF+ek#oS;KoqntBIa2U9{5||o;X#&#z50= zikrrL5%=V#Q5)FGd+uXOPxt8O|4GI$qILavokqD*70hi!ABSX2@|(_LO=EjK^65GIkuiN?Keqdf2_v@k8F>xlgtp%QF&@qvlHs`zo%oQa?xx1Kzy6sC+iOq<32~GTAE7d)lf4|z51$;c-Glq9Iv>0do7FZLX- z{Q6~mY9XQ1FYqq}3wufrX`|+^o!otnn)G$HaS4vy10nN-V>wJ)RBV==&W{3$aKcm# z84^?h?B@z?{rWjRde1+gJ6G0sAS$|YM^&y$Mav3cZxl?pzJr`?1Tl>H1)dI1J_gQm zer%T(e?SCRz5fGuqCzstdNgNQD?Ht&J5uXGtZbS~PCPzaOW=Ui2K=_USpHO30njU+ zAeL}ko-gQn>9VTZKk)L%_UCt6EYi!NCvP)fH}QhVZiO_WV~dv-fW)|I?`_3TDN-O4 z@jS;BQ7MZCQ9A8Aa6zg%d`Cyy0jn9iQy%!sZYh3d%W<*HMeng#{J*SWd@n2!ME8KVUh+W?orZYv$P} z?3aD*-mW;73tEXdzXWEw+@xzJSrBGuzwseUJM`lY(m-sjw~bmpxV3Fgw}6s- z7)~R=QpHb;WSoZ4qT*j^5~BI3_%><_cjPtY_6Lc5Au_o%O}ateHvwiEk8X0%*mS+E zh73O8nwfaHuU60$q|VS>SxSH@f8i6yq|ZLhO`!WkRvqtO+SHQ=Y3iJm-M~n7W|2RB ze|3#|?q6an%YYs`URkcPII|rED7@SXU<~;L?6QOuM!>x@rna5u)=a3@;#7FX0p64U z#PJ(T@4ux6=u)5zWAVnvMs+2}Og#Aflc^`(`9Wn<7R%5v#4>7Co=d%C4oP?hs* zEPSCoEAK~%@a-Of_guV2<5w(x_K@0j49G}%V3aG%fkxa+P;t|)RS2QwA5i>&64jK=BUiwhtSh2eWz7aYx)-=OAa1_)*co&u^83D(*Lmn8ctPH#Wa(ySf5 zZSTafjg6hII#<{L439Qi^|gi!+T5_xLVu>88PjzgFEUzrmR+|E;1mp{S>X3Z?88C<}z?C+Mf~CTka`+tu;c(C4fR z!D}4=kI~G{wJr}X#u_7rS5^oXB#sF0E;ylN2{~)oTcsv zS>($z7WQ7~y@`ubpdDkk`o-(~dIlFZE==9FCTHsgz3RU&reT6b%(TZp? z1jjk@TLV2x>QE7WQJm|oxKkkB-X{`ECQO|FpS>jMdlCAaR|MeDLgx_;{)Yw^m~srn zK?2TY+ea4m(&(oWbnt!=Dp&7nhB93dFfQ6CxN(2!4I8d@%@O{dE<<`e7uH7TgKnq~ z^@W1b?dh8#8inG0%rg7gqpRi_R>tu z%!`m$qka{q)MC}wg<_i=w(5Iq7(fZtFK+c~G>6uF_pfVwrKdU*S3P;<)~wxK-0xi6 z)699lMrm51Tvr2q*Fu}(^kK|SrEO7<;Q#+Fk}fX}DtQL@&Bu6p8+(ZD6W5{aZcx26 z_$rdD6h>SHz0gyF2p_*c z$bY)9Hjk!|L)khE`o5;Ci6`NUaq*U)qfr@`B++B9P>b*G6Tv&g%fL9pZmGMhj9BuD zo0(zo$UP%^w9v9bKzF-7Xzg5m5hi0ZFv01H7rNj5#K`_uQ@BumZbtV=hWB1YVoX#` z#&bCly^(~2%JM=pRCr5?NwIULH=Pqp%uth`M(HSLR6k!5waow7E-B(sU9A~W>|A%M z0PdDP)Rl6ZQUW|K{VQKXW76BkTDP3lQvW2N`J~p4%M6wLN-VL!A~oFa%E+s(`K5+J z+*5YW)Ki`yhh{a(Z}9X=9eW4(C-CZFIMS*e zmDKI7Q`9qndgLP1qQDWyYe+xFhBCx_ zfl8m^@+|>|N`@DNiyy#Gp61za?YAbNa3m$V7}Ap#lb~~3fV{QCR_e}9X@OWkr|*%B3vf@xb9K3)=k&8j zkL!njoHl{hjAqC~s%X`7*-rJZWs# zY9+XeF}MJURAs@f?qpN!9P}`E?CE1RdI2X5U;8oUkg5-QeiXq`O)o6C-;<$H7WM`p z$;#0TCTRWZw9|J>%qR+9vA3u1$nu1bDkR&%TIgjUNO#+KLxouY5k8tBCe&LFSbV%s z6r(o{KUl3*`@JTUVod{dQ?XkN?y*NbdIG0iPz!ki+_6X%9E1OR4jarNNV-JMrr(P|y_WY$4Y+_JO+FX~Q2l(uP9BTH;mI zqVqOZfgbKKGsBhFDeZ9mrs{Or9ND! zGG!gJbQmcOFfDN}tZlh=(ZT|1!6n-Py>9_kY>A7^-WP6vbB9d$hJ1T|>BCLj?~I)# zbFXDh8^BgD?z&%1>F7gmWaD!&ekb`LrP1FtVg~OGK(#p|Cb32448v!v07y!7dqj zZQa(bKwN@W;{*TmFrijf#Pt$|8?}!#L+~&UR$7)PJ-SR;RI#3}mg4$u^jr5<;IMYB zklh9F<$<8J-PD8Tx28+IVbos3nXf!P7f1~?8S{QOaISNY|4g9De&FKSe%r2A@1uGj zy*u>k{5~5d^|>PwW;Sv)e0+$OR&rVz9d&ZKL$Cec=BvEAOrA{8`XmePf@&5Hal-xJ zzF6YbSImG#g0YX{OWv70C9@Mp%kXS^hz;v#yCk2K?uGN(I-&1s1nDW8&^y?;bcR-z zQ&=0*8%9gRew&<516$2BH4nrcS%u`$n3~r$)9L8(El0Ei%+)O(!A{Ayc!t-x73kQo zm|I1$**S5H33!g?-*(m%|kS-ZNYM(qU5`&d)B&!u{olJgZ3v|j*ytxB2R z8ws&KN`k`a$ zU}626 zh5l4zOv}W0QiCJ+TCP?vb-|R_70twm)A{?z?WuP%G~bHovudQ6(4A6lCa@aNpd#CA z;d6mW1=|}6 zU!y1C_R8@p!&VW=svRF{Z!1AHHA|%Qf2JDb^N!Y9!0=Q2v2WA zwGfpQVJ8h`6k6fWwGsU_$Jb%VKdSKlbrZAVCIJU=<4+A$`h!u$%U#8FNrvEq%{KzJ zxfXvVFvB)QUSdC8Kjs0i1Iw$sCme4;Mx2xhsIDfE%GH_-)w`F-;@v=cpJ5CIIVpdq zy@s~ieh%My;8UEZ+|{$#CAcfsPlT6d6P`Zl0JD1&$V*VhTT;0p3fu%>KPaxz<8*?C~ zFT_HtSQDN!MB~kop~dQ)u?)S+%T|=NxkYWlSASU>qqxz5lkCf`L&MHVkziYrQc&jr zMfq;vh{{X*gcdLYQRT!+jNGVQLTX18Ve$)VSKOH_knMq{{L zQ9R6(mr8(&Q{3zD9@9qRoRo9 zc5x2<3rWv`zDcBAaXL$^*F!W_i<1k}T?;IKzlaqztot%Tgb8oj8PTu?6v-WUUPl{t z#Ipy^2XdSJRdAft7*z@kn%*V}08!nej>J&Gz zO0-xOc~Nhh5ktVBMH8qZOjjQl8;juQwvDedraI@2iLG@!%N^T6Vjex% zYu5JwWsNX095jEt!t=?F+uVZ+@iJhZbW1-r7TeUc15AtOujP)7LCfVBV)PSUJij>$ z{V9Hgl}6sWLp%4iV1N}?`-;KuZYZU#x?14r{uxwWBB{=fGNPc~V{;+@U?%0LNBxTK=u@K~o?tiLi!~;K z#oj*oZGwK@Xx@0m^`yFPxUAbh@Fy`zpiBGt7(A-HxtGp?C$=tzxueIAS^@skWbDft z0L|p#oWoPLvE0AD^cnv9`%GfxI^FW|U_YUn&L6Ddr8uEMri)mS{z&)>{4_R}6`N&2D#L7Ml7=1_%n3FTs&cU3;*D{0ly{WWPa8a*EzZ zc1er8Y@qWT-ng8EZ8Xmd)+uB)q_Uj;F*fDbwhdq6KWNTi?Z${JLd{A$&iBz^8F;a-VaIveIkZ5O58DxVcS9*% z4y|OG>5~-LEKKM0#g1MRS~ca07nOx9%1}$Ijx|_w9fD%p`@SM6gvwaPEHv=iQ#Qe+ z6j$*M&wQm4KvV9>iXJpSO};UzaS7-)Spp{&f4}0xq3bn>QCE8ph+&7I?pY3Po9qdb zGJl|F5JI!V&~fh%N2Zy?=V*K+D}?SOYt=}(EjIwwW%o@LJAhx@IQpR3^}5~0>w=ZTcXbsb3m$3dAgC3VfbS;Cpn0RocQYATvM2X7u9W=Mv3m)i16- zu)V;oCrK-m3>o5GhhO;o-%G@=mjj!<#P=O%k19S|DP*qbwd>xBH+tIvw!F|q*wXWo zqG5g~E0ofT9TfePOn}$lMACf+6osP`f6^f>M&m z^IJlU!Zo)#;^{6V;KGSx+`m6inM`W=1@li@`By-kH)f^mNsrcxm!!7thzow3XNK2?A-{AC%)5 zP0hnR(Gr<^CwEJ30FX7#tF}kj&vVWBF$I$RcBSqv3^OyOw7yomnIjyu{LbUi2y}*R z;r(J-@p+10h~q^?ftRMNLDX``)%g){4UkY+)y1jCf=KYx>uR+J z?2^+z5q)Q}mw~g;jI1vB%+y$v>3DeLEvLg2$IdRpeZ7!AC>aC)Nx0PX)YN;aV4)#Mnx+INx9(U2cw z>`J><{yXn^KaT@`^b8Twpyk^DzR8IAS7UG?%^&}*uzU~CXv!*TrO*k*j>27|CNGvb zRI<<1HGJt6h}riE(JVtxa>vta>fMNxrGqjv;`~1sCsgNY@ApY|CdEy5ber*Mwl(w5 z=2lkBz02kfEZ6*g6=ivs9xMTW3Ve@%KE`qacN~1K`dChGVY!oz@ajl235-i=lyR&8R^$gAC+8IuIVp_`Yu$={0GDen zoexUrqINeN)lN3eYz|79U=Q0t;*B%K@c`vGDeZz7uDMGju5J{O3DH=eDM7^8YTmt{ za8OQWjLgX^p;3R-ix$uftW50qXuBwlCOR-`8ZxHw&yE=8U_`>^k%B&0OG|9UFpFX2I; zekXzG0vSMPdX~Q~Ct%`|jt^1Z%=&G>xd*o#uD0|`PssuS_FJdRq}~JOME9`wu>wh2 zc-yy~;A+<^&!~_S&R;HJ191Oq?@r6VgTG}_l|RDnyPa`x<1cP1-olA_QWkZs(^Yi#^pFklWty++%-beF!w~w4&3@=xysTDA~caJltc9Q#TcQ%^tU!HuWY05+O zA6$w5PH!|~eIPqPHV}hinuM0`GXBDHnPv{>bh^2EfyBf{INC{YX535q51GDHZ~?se zgh{iiRv=N&pBb32x!{As=_;X=wqWhpk%m|S7NPlTKT6$Qxdd(~(`oe%%l?oZfOz{N zT}ut>+^)kCkGsA2xatD!`MB5z96}>_0SWKvS=dRY5v{zuFMr6eTWY(rfFzVznzGhp zq+7=05b-5P*u%a*{C}E-Eyaih0|HfN(rI|_pa~lweeauw=)YNMp@#B@)RTwH%P3T* zr+?HMd_(dJ;BXHJv*V0rN za54VJ5jG+vFc^5na0HIvIFD~`ciVF`8uw~Ham)ZOymjHq_;4g4sN9K4 z5W_^}2Ur4yllDeivspf)fe(RY1&Xj6i0zX^*f3e|yJ^wsf3tdx<98i^#3`F+Aa15| zL5v6bP`_PT=4NsFB#LD^Pc!Z{U{E$c`E1XzMTo&{66y|lO-AGTReyz*0P$C3!O8hf zfM>}}=XsrkVw$xE#q`)wj}E~(O4Hpj7tyzEeV!K4mMF+<7k63;VeYdrgUoaB^~(cQfZV3ei+mP%dV z^1G6JR$VWvtrqNReQC27=;}tdx4?Mwg9lzYo?O&&4Qd0$MI6e&x2Y912B)UKm-*M+ ztFlkrUA?jSsRV63%pr%e9%*Sp+Viz>*N?$L+?2((V(J7C{G`))bX2{4!UHmDF;FB0 zE#d@(F&-}xFYoGrVDdrxzo6mO`QgZ;!+5)KN5eM)E~cKR_P2T--6rS12|rGXoh(^y zJ*8{-i%Oxt)u6976j zB@RQ$Z|svwjvak5?St{Vl9a{HiksB(2-C%RYB>jU@>)Tpttbt~LUm|6aeT;pg!NqV zt&y67+R*fgvu=m=j>5LD#@t@u#A?Ebt}2)#s;A)6Iu*rxww14OY++NyP@q3qtRO7(<4 zwvnpd?pU`OoYl|_GUwH^t*!UXKgfjDU%SgwJpWWtAiVNjStfzOjr~Wrvr?^X5{Jn7 z4C|4yOD3>y^Sj^2M1nUpGvB}Hw3qXAygG3nDB<5u_#o}xxHyN~`H%whqkxLWzL>)_ zTVt6R6RLVt*KTH}{c)HDPV6QcS5Wn}r)J5S1cYKX5#=^s91antz zt6bW$$&Q=i-^R`-_{sk{pk0ydRpVUBQY`5}7$6Ao^HWxPBGHJDwAtkM&g+<|?GGPX zWVISmD_M;FN{^S%izc%e&v<{@WvKkwv8(zIML*J481nsg@5VuIgc4cnODOr}hLzl;UUgGN1=Wgf)7E$pia z`_mxV449cUd%QY+EwBqMyFO0(S)}@yGNI?!IHqV^rSRVB3=pCMKXRs%hQ z?pal=h@3L@Oq{$s(5@CD+w?4%{Nm9z_Lkr1odeo4?d`+_8?quI?>>RiPH1Tj(=vtBnY)|(D%LMEjKH@5H3_SEdB!cIm)c78-4LL%VZ z1-Ho(><4Z<-IzCOhZ5=9zx29QgYuFC#khT*=vrlDu3?msxw~wgSKctWii@Uh>uwoy zdHfDUrFpNu6(Lw5;?^*u*5887m!C?u4mQ|eRCV8B&^O+ zQBan>@HRH9$hGku+t z0_-8m8LM_QAdedt^)IT$z-v21Oz3$dvE0c7|1MCl$j9XUmM z&<@7aB(h*GihBEQ{#g<*c5nrU7mFb1LBB|=uqo%jrt4C@*wxfG=5a}c|Lsk@h} zOu)f8Jr(u5$w!H$LP0jZ-rvje7-7Q}%#e-@-#3+B*%aUXujCqv7%?wxC784o|qDKXDTV%gu zYGdKzR4GDGsj8Y(`PYLf^`Q<^oklJ}@tNT^hn<-jB~9M(EggD^ZB?>^SvcNy%qHt? zT9T3!yd5yHcQZ3Q)y-ggU*qo&>_?mdIS0#@N2}C@j6!t=y^6nBY-9%NNsRd+t@n}Q zzt&scSvla8B4OT1=M-JAxdVaeT6VBE7C6=O80-BV>Lc`hUa7BY;n`3fF>3wtZ`8r~nxvj*ncfUb z&9#|={mWVUuxZY}Ppjq@HSTQdR=6Ej^>G1v+>P0)bq(5CsT;jFo*X(XjH4B{6Yp8d zayExTJIKl%M@f^8vi=P&L?v#@eI?e#tFIeh({t2@Z(#TXBSPUN+N}|FzR|q3KR;2u zx~>~;Xo(1^%%gF?dmMx>P=)%z6e+tsBfZkbUG|47lXTs>ZLHBicr76MpSHFn9W;G&FJPJhm?=>gjMb`7AjfEQb`#+n9eGml1yK` z;uJTcTnf$cTDZEbbjBQJar##_JKJC5fngzD2dNofvaxXF!d2dzIy!?JZ-i45+m5jd zk0MmhRf;2{6i%L*oi#$fq*1&;^<_|4{u`cFIMF#`Q3VWPH5L z_~_@JgJxhzT>O~>9>VL9g*P9_+OBuh@}{qc{+|+k#PxyS$qw+hcxZV%d_5lhM!2>cif&=)NbgZ6`&F zK5&;kd+*gfdRX5iy`WF}x-QQwXajS|I+Sd1X4^PHDgrL}bK|w2(XZF@u;9&&r;F!; z*^MoodIC3J6W-rJgtP=)pM_FV-<~?nRell3sv!j#GV}SoA^(;Amzf?&^n8a<@XG5^ zyhANNbNx3IJO81|&4lI0b`p0V=Ik0Rtwam|&6p}7KVuj6ct90%M(K35IIJ~c9W$&M z4I;P>)jDAC45Naj_0h!hsuIyfr(UrYb{u2#jo9Xxa#mJLG(=RJt_%#b^pKFx8uveB zbX#Q*see}rJ`AN-(9*O;>VX?ux<6KeFF{y9_{H)w?w2P3De2;pBx?DvIws*cc78+N z^rsUSpjgB#d}}$X5&->)?Kf#$lvRy4g(*IdJ|()LnpLH5s=uqKr&NJau0yF3 zGBevs?y?k7g`3gotl&4-7d@~|_~p`(43wMSSp|eb2_O#c{z;aHE#d4NKiuer8uj9BqVGSgiTP0K)~ z%Ci%}5X*bPCCV+w#+jFx%V76HHqj{?P59__xaCJrsVV>`+Xjo0ADaFRj?3glVXUa- z_?(yC&5O86*f?20r_UkT^A$B;o{`$FdB{^mt9}#7qNE9#5$cXZ6CPV?I=LM*AG{IA zdkhTsHR^gMG9SeMcMd~xadKeS$H2jEkob(i!MckA?>%f25m7SzFzILdKJxBXFQ^1~ zk&g9TRfXlss1iRA;gLa`zAZMns*}fNEEwHIWI#|tE4bqGvxnWVpSKoEn;nwb+A$kF z(|s?!_uuBxs*BGKyoPR5CseSBf}-@!nl2*Ka&_N1dqd5LD78YyDTq5P$D%$URBe#>82!vZj5 z)Oej933RTVn`olG!NB=NDRU3KECs@0_*Tw-2Hi)kj;gT`x8J> z4vP$*4^>d^Pdck57ay@Dx}GL>YIqLWKA+*)${PNCC1Y#;)DHZktZ~Hl)Ro`pXfot` ziKfDnZr7F3DMgMi3b3Myhg$0Nh9KSN_5cqT_{V#qy->O8_Zro+okOL zC2VKew#F#}dSi&`{~-L+RnB{P+t}(JXM{qp<9kH=t^F9ZkO*x;96tne2Oy1|bHZQ| zvP44G-#A&W>nP1=^*&fOf%Fw=>w-&XJ%nL(Lyb~#5yk^K~&op^n*MbxTQM}bc03(Ez{4g zOR)pp9g{HQdEumY<<5#U<5(gX?ZNj)$1|~Ce_Q?t49Br2)qU|)1iE5 zN;K(xxayaiJa1WjCWyZjqa=0_`27v)#6RdIrWVCFSjz1gH!281NzU6HH2Y^V>?-0P z1a3-F6zYgiBIl7BgSQN=b{0xe^x;5?F~n#;sNqd(f8pY5SCIkC#(pB34mLf3+f8Hv z8LipjF1y-aXntPgAUFvTGTWolpEbE#7LNkyCH2zBcCQQ#$zL%;0fv_b&>AA8s{e}r zE0+iBwveJV_!Iw`tiP)iq@HEUDn+Hv^av~@_54LHCalQFJpJcU1rBjLF- z6w@DRld8I`+o`?ZAb{1}cy=iYzAh%~Is{uGZwWnytc-5I6I4}x52q#k)>hzkH!1D! zkIYP1&Yj~K04hnae@FYxt{w?+0j5uvGW#Qkx^-c$qG(PUhYa!`JpI`C>B^`068ZdE z9}o8R=Q-B^q=Op4PocFkWjL1d}BP))u^X? z5~d==&xKI%!`i`a9ri!4!^1nH{`;(=k7es;c8zipw!S$G)*86E;SAtO@S2r^n-r!r zxZc>4+Wqo>+xoCQ)0a{L11VWEkSK_(D<`0M)rc<1L*@e=5|Yc448GItBDvk3U!=@I zp-pFw;ib2lV&eE6qJTvme-`Gi7!y>a$SOxV=62fINfNS9wU-F*S-hQY-g2d6RS@}2u6LF_|6 zUSS-X9fgY>&<~-KX=(IH7TYx61=l#e$Uyz z;o??p>5{bI!*gTO)ig+V7>FcoU+q9%mwIggA8`t9wyOE)9FE?ZKjE+vt z@UQv188zzO2OY^Jz!$=cwfB)?hnMD6OB8th|Aej(aZ6x(l}_AhG)I>2b91To+Pj%) zoSEi9+N9Fm6abbZ6}p(Si(J0@3Xo=qt;I5o01Ir4M6h*dOCGd&;{8v>E!HZbj()G< zQE775SIfV7G?=hRCjL?ICOZAgr7x;Cn!j~|6P%h?BG%x;=t+Zt{dzyr%GfijwkZ9u z5bPO|&7}v6p$8`%kzq%xi1w$sn9YjsLbH|DAw|bnzPVc*WAx@0CS(6B1Ca5=p50J| z9r&E!z5GgQ?2T%G@utq_cNc`8m!SgL@xEP>SpNRHsARpI8-oHsH{|^I@IKPKDvp0( z|He9J#&J?)B(7aPf9VRn^H3R_IXZ6C$Enm(E(2N;FOG`um z$Ok5fHiZxiSjX6S}(pObxnhEH0aD;J(uedaE?jSWP4(xKPWr@N=)p*S08m75 z2C&Ny!b|ur&CMhegcv9Acb70fX1uBn3i=}lRz?1OyKydT6 zgaz=t^A&USE;24Q@%2g|>l~g$OQYbf z?;!6JQx|~ov0+DUmuSn(V563rKk%c+D}e&*(WnKaf@GNe%Fp{jgDvhqJ`TfsKp8;j zl8d4PdoITNH_H!A$eBnZt-p@MWubrr3$f3tZV%=Goe_sR(l6gbz2D%c@?G(S_()#? zK*QSIySOK9QWsc8_rPatZp8EqyEr7?Am`iV9Dd_!cDh?WjA)OD;gfWd@A~H9BCm|M zFei6ah5bc~w;GtYrP7=CPJdSL4~NToIlY2)Ci&QCrfb_tTGy(HGh9B_jc9=2q4=sn z+Hqx+`}B$9%hM~4ew-fw2ieNM$wEKz+?%TFsn!N0>H;zRi%{ zzdfV|df*=ry!pibD*-_n%%d@7Mji2sZK7{|#+VE{F0#(iemJ~d8Kyu~L#9jQdA$n; zj+2%{s+5u(q_@NYRf|Ad+$=xdKXUGg$|0pLD`1`4!q;{gg^5B6s>>aLGhHEyR5{4h zO629rS-UAU(J^)BPl9=5>vi}W5jbwV?H|j*7p#d~@v{%rjF)KhhMrXGg2DM&Gh7mP zbj2Of+AwL#o8Q`Jh6jD!Wox3|^jf|rAQrZ|A6i8gyHw?j!sU?+E2CTXO4Reqd`o<_ zTXbm!CXB%&rBuD#pZq&~vwQyP+Is73&eqADhbS9x92vr!JW$)X0$4mwe_>~f-2(DF^w(AOSC6btN5cIuQKkT{*^i{WdOV%tSQ3&q{^w22V{}9yYgW9QIRyH^9u;SsV6%&=PRh ziQ{(-Y1Q+<)DHARR<0nONYP{!Xn^D8v@cJnu)crI`EaOE+Ir_RiGuIS40ko%LgOW6$dB>Es7{ z&Fg}U@&kacSSHA-9eyTJLQK6IE|l-pbf4VSlK&E`*~+R0GheYtpwIo?RMi>Iz?269 z2hC3tTPu=j)J$nHQqs{`dKmlyk*ht9K*|ZdA|TI-ES3x%2dH)`+`02f6(M#q5wH&5 zfC`4g`yn2*q{mra6t+^};`sfAuk&D55iuK=l{O38;K7UHeAs!OJ;5=ALK@nc^eB+6r>KH5#BnYBNeZH!DK&Ypzp$sv zd|nK|Kv5=Cg7OWsbMD^$uzQfi!$2_&SZ^MIwCaco8DuU1f|m|4Q2=Ux>lO%r5aYh( zb$sm~0D*EKiJsW`Ucai_h!B)uAX;wU5&qLjo;S?mJo;`4P*TYM1;gLnCkWK_y{qjP zOe+7xOs^IDTWmq>$mgsTv=Lv5LQDbRsu#=-e}HZ38lSmyEa&8wl5U<4f$59@AK_zf zdjx%MjL^NIafk0lVF1U8#^lRHym@N=TcSgTJk*qrn-Mg z+U*K(pUO*elb5IWq!aG}?L9~K4F$SGn`_WIc=80mtV#056JEaNv8*Ay6ntU({?$rb zSWwqGpL-&BD`FmaH7W+Bp5&5V2Z)fmZ}h@^KC{YA4fbzmbL|T^gsA|kmE!x3nGUxp z$n@ZQJMdoXa37~v)HJpN&BKdMeUl}chcu!2duJ4Qb%7Q7;nv5P=;KiCt3%ZC9!Adl zh>#*0Y6emYZ#L*WY;jG*8*ATNn{C1a>Pr~BZpaGWrky^C_OAZIrM}CwYyfNS;9CP=ABu>5pi!AHHpE%w*9&i>0GWI^ zCMtc=pu9YmZ3oHXGau$7&%3l2*20vk`pf$6(-3yx(k3R&x3qTp0(L6i3JBj3v`S|+ za=ny>G5HSU?r&AQeO5w*Xc~+H$9jl%h8^VA-x=0Vf?5@aT3 z)_{?KN7Af&0DTHihtR>L1)-!xJrUO$YA*ov4}(qO_K}myp2bZIdt0Q8eqYtkucfVk z&q+YmLjcK<-v?5I&HL5XzFLX6*U3QFzBo*D1#a5%T^E7fWoB}Sj!P&7>p#tKtf7m{m)&A|# zT6Pc0NSOHibDfnyx&<``WV=652$%0vKw8HrghgxJ;pRiTV!crsnDOJEwj*#JfKIO~ zIB|>s2rNg~d;z;W02dg;^v++Bwo7f;f9p8`=)bLdt^yD~BSxEnL3>v!K|y)FqXM`F zNYv%FJfW9e#6nBqcfwPQ0UC#B14}B;UU0bB0YLnV%7B|R7!#)F+7Z~=n(#IyebEls zk9E+Y?H(2v6AlFvJK;u+?B1N7#Fzu2o?1Z+r`w*V&ms|94IQP#Ljg&rQ>s~gDPJ}*WK>WFas z56+vhy`8$0k+toK+Z4ii{VW%aLkno}CS3FS+xUHCzSOx%dg-MiBK%*ijH*sxzhWki z9;NHo1D211?nH0HKP-VW`y(ADI%#-*ER! zFvI`pM%xNupl6m8?kutU7)yqPca#7Qk3T!Rjlo?5d`xS2jgIQS6au5f<(SP#G3kUA zMr+;gNT%0{yyUo5yO;0awS|c`in0}_)Te=qPS)Ia{hadMgBI*!1 z1;If_f9LmwuN2X$L&=&EU~O{#{fqLtoIpux++|)}684yrcjZ&Bd7>2%+dr#BE$`fU z+a1Z@P-T5dwjb_V9cu?vX8!}fPT(op@ca$dOr8JMP1$(jw+)IuJWK2OY|c*vKgpBP z**i{p%x(!x)O94pbMoe)HADs?gM2=;fumSqrroUh=|yt9gwteasdd_KurN=~>8+it zBA_vcvG+uhGD6*-nLqklC6IocG$+jBckl*v?05yWHP6_W{q5=7 zE1N$)c8tQ8esUt3ui{uu_M4Stg9GBH#_yj8;=#vpp)yN0x@*mcWM$r&cb@lo_OtgS*|VL<%IUT6fqj+m&5-zGYh(5f7S0Be zpJ2{q5E!l<(q{JT>x8=0VNQIASwR4uQwcb_ZL4th#zSfamN_W@(a-|@!F!1Yj=G=9 zQA4Y=fMj=Wo(Tsm*qa%y6s{sgnYWU;NzdRv^>g(!*qct+HpWazKl^o1b82B1j3|fB zMC`Ld6!JaU`a#%s1(iM`j9QU>4lTq474d^B-(zhM-d3Hj(rggAr?g$9hJln>Q83f! zUe2c*9JN%*%TmJ=x7xFg5+fgkYV^)b*B6KLh{XUZZ{{QAtTr~S%~Bw_QPw4fYd4(^ zkzu}3h~L13b{oUA$w!V^-B|R?Y>Q&JqjWNi@>7Gp^?_F?1HaSR3xni|fCm2J$Zn#= zW2)#*$6ELvTKDtaM|se=T%!El%QtFg>MG&0cSl&s1XEk6ZqGvS9on z?BUk)5g^3${=b_nnP@H>XyA}Q(nLq-$^yL~Ol(mB%-Z9d)uDS>3awk4 z@wUU9??rz-N-QJ&$m+~0Z1OZU)OBw3R6XfyGj2Y|m#3e6ND#}sd;A)(=-w}{F^~|& z>f8){w2$47Jbi=0H)|jp$8I&pCm;TwJjdZhj`3bM;_lSVTfs4m=EJIe#Urw`_+v6$ z69-lM+{XEfVe-@pudj=A7olE?$(5u{U<%asp`5?@ zGxk6x4H#QuT*|At#`J91O#2>w?gGUKP2m$%%d*&R`s3h`b&ngKjX33yW@>}Djj2CA z&B!$kRTEWn1+EM1C6pn5WSJX14jWOOy6WOQfzdM^8i>a1pE?D(F!&q5uI!SBsPJD> zUZ?G;MXyL>*DbckyjmYq2@}Tw(WAipGoSUlQnf?a0L_mac>_!0l%aqk9vK_v1J%vX ztH`amX}-Bg*8E6|R~L_MwPEV~ACGMX?v>ME&+@+uofVM}Ew-b(O^b74-CoU2Rv9qc zWi3Zmlb`-wNuM(KKJrtzZEe|xeN6MSpRk8!0A19ltU2V#>Vjba

yS^WrKw;6}| zf4h};@fh3KI>WpFYU}}2n?m9H-t-81dco^e%qD}2D}~gdboccF5^W=O2Gwggk8X;E zhKe{6D3Xk2bkVPVu~Y8-{kWt5CyyCv1q;Sd2(H~sU~)iXqqX@w--!FaBfXNwCrD6I z86I%NsZr!AY|b7czesn9+3G*v`MIx?cVxT%?J4N3)z4SLz}E$iViLB`KZd^eMR4Tp ztEvdFxJ<_2Uko^8Os=+#wn~6FsyB9}3ldli@$le{N1>(-9U8p6gHYT-vGjO`o2b70}@W@217jh5}q_YJ; z^3F+?z*eMfuTDZN(EM+gjYzBWp{m>ZJ2stm7R6Si1YJX$kZ&0JTyX(Nu?iIW8$}_A z<)BTJ8Jt{%lGzdYnpJW!sK z!h~n|`-wMjI1E>MO_XULJ6-qf6NA&;-dp?^l&t6Yeu+Hee?{jDltoa`n}#*B$e;z{ zG_A;Y>9dQf`MZBij`5d*Et0PpI-)O)K^DT7V{Ae^cF+3@5PZM4G@USES6jPE5K&C3 zMq6xq*KPlFxGOBIz6kv-g^bIKtxv^)J(28X;D^o9=8M=otpm z?rHqH+1Ud4eqc0bT7XLHkTtvOj#SNeS1hLXXl-jMrj84`ak-K3M+=^9+5BzR3;7u@h=Nk}Vaz&sbYl3={#|VO>7^ zJhJsvanpf4A~K3NavktjD$u*`0HndqbjfpJ~)e#3=?kI(D=`T%f(=oq?cup#RfEYTMe<6D?Rt|H*90i1N1U`cy zGy$fX`k^CAhuVsi5$)-2^$-8>d==~4CGUL+iasQq+7Q;IcBpE{Uk9tt(KxA4Xr$!% zrWYsr_q$;o;6~O3{s7mn%5BesRfF>~QVE}5o`*Cse=m*zpPQvK*{v`c?MPzVWVVQ7 zciYF0epKX7^E#77wV6b|sOhny&j`GMjM&$)&(mVfiLrQi&x z71~n{8!|Q_%5#&viLFd@rQ9374D1nlTbOJ&1YW%Dx8Y`&`11xYqgkWDifV-d&SCv%#PRUzo!U={}VsBTaH%yL0R|`@g-UTmUhV=xd%BluOd@lWx{qyW2j! zO!*uJ%vy;tt`0D6q&4OE)@h$ejXMW*#H?JMM@Bo^joDUt$lJYjZyHwbKMiUyN`1jG z=7*0jjlM^JJ#c%=xx69Y52A+f{oex%g`%GonPd`Fe#tCa}ags8cnm9eMu5*tGR-v-zu& zbvG;^6Xx%3TiVdK=OcGzJO815`Q$X;;T+av@DkLK$(p*i`S-|fkI(r@Uw>K}2(#J$ zcmpgshnDPDf<56s`%nbd9Dl<+>jk@-{!ua)#at(9) zSPzO}5k7Krt4j=w;JO~ahel3y#w{RI3zPEqwoskd1YtUuN8`8A?S+nwyOECZ1{6Ly zogfcG=t}qrv!?|Rz3JJi4;=lBP`S-Zs-{ZL(ZL%B3@VML<9;9gN0W^l|Gv(Uthjt^ zyQT2yPY-h$)49qL$&AUbjOlt{MapW_Mr zVP|sA#DTDkt~cK}h=pe+H-B0lR6~18hD^t{GiRQ~&LUgvs=h7+*UCJFXN%;I$2BA4 zue{a_;ZaMCbe#Lcb%vr&P51qWJ+ndjFCm@hF2XzdCZ{skoepvN4bXcEVdy8WF@u7p z{uz$%RI+iUp+px$;|bCs?zQ|;#bpQYi>uf)YK4Je(xJ1#y}v%PGJELDhqU~ziU+Or zb_`dM^3NTiH93xRkhznAG1bi1o(SxB7JT4%OzjtdFGT7(=A}QUb+NmguPKCVH~nGB z(ho59(`TIhO)m^L$P{b0jky{F#=v(+? z4y|lU-u&)S!@<11b*3zXLB`PcnY3b}$+NP?6L;Yb)=*uVGu#?FYhv1@4Pn-%PQA2@ z44C`LfRD6Fv+4YL1Jc~q&-vKAhO|5E$onZL{c7$}v6Tz8=;p=OBS2bsqC(c1Ib<$c zs}9E5gb={_o=?ZNE%)Rzm(BCSENeLO_yuz%eW-*B*JIvvUMB($Z85)rE?Zh&{5H22 zsJ@eE`G{|m;Tq(>Un-|J3UnjJz1&9*HAj|+4mVukcS1EdOrfq0fPIp-JpFKVMN09m zEjxW<)S+{m{MVcH&5fQ5_!yuZ*P`$XG;=GWDGquE3+QS{T+-5ZZv1m$WB2?c+jr?C zniPHvErp>e=z9(1iwltCc{Xbr1jqEXtzc)D5B7+2b(kzCWABFlvJ$k~s$01ikTY!I zHLs5n-IBSW2w(d?i1U=s0sk9n1%_Me*p5or*WRu?$enIH7BGRvjHJE$=TtIQ#Tp?7 zvv=iCgpE=eq;%#GwgwnxrPO&gx*I|!uQ|L%WcRTNnzMX*<*y}wUval{h0qqs<`+hH z0{6SBP2)a8ldib?-cO`MFRURwrbQ1H?lP?*-JQ3e;2~Ce4lbzm=S}X@P~4n^6OA>T zybJ1EX}`I2%;@^cOV#ahS>uzHt3A!KiEa@i*dB$9b4E~Iw%lt9x?eV&nHa3_pr*|2 zM<9@UjoHon;O`)l!o?3e^gcpg)q>iRz}t9#h>nfh1fn1`W1p}?l-I0f9+z~IcAYP=X1ytoJDG< zRS@R2dv*+czu_2NDh4;v>{^0U70(4TAiYyqjL`sRN)M0!f`)vwjgPRT@nand)o31Lxo-C0+scx&jD zSi(s3!?^xgK1Hmw7#-wayVWiVf03Z%OSHv&PiU_G+I(c|B%K5WYBQI>VF`nW1m@s~ zcI4s-jJZd=$`m3al5Gm6C<&90gf%y6jYEW+d72j~e63vYJ2$;*=itul%c#4X0u;G$ zk~#@c=x(vUdDr81lJRo&y9B-;Z#AkW81`Nk@Nwq6)^ z2FTf_Gg0NF2nf%s*Eh^o8?RcZ4_ zm@*g?I+5Fi z4AAVwfG15D>@F@Zy`K0qwjB& z(*YOxbK5Lmom?3r<>X$NAlv90ldsX`i1idPwYC?FSjW?Sjz@#VCrj_Tv?X&ZR-G9@ zd8}SBkMQYOoK$0EzScG{>|K;LDZWrU`tbtAJ3m2A1JxvX_PeQ7;$!upYvp#>jb@9o z+F<7hM%R+sTpv~bi)^DDS35`C2h)So>(n|jMZtG|55&sx@8x-ZmdCXt_2O7-3Wh|Z zrIY@|t>`!}&<&s}B5(0?y#I~b8Z9H5`%F$WY5Y!b%D&YTTkMow6Gip2i9r^Lx~PEHjaco1E<%#^2j;rkp1XScewbgg`Pe#{X*!u&R){ zk#p$W?DDON@ipL!xr$b&RHL5AHnt+=#MuT=GO-0X)fF9AW53jd086J%>35?jRKRbO zA{JviOEo=8t_iJ=W%DTNxx(7Km$FviTB980y%si;AM{0MQ=G%ocI za6$+~QHwm=-T&n)25-}edL*JwK27VI@=PD$Y~8i-o3-g?(A_5rJGjnMzmOk)4vXSj zRYf?&t)XQ1^#@%%2gV)h+c+r4_Q*jdAt3S-yzA%#+i2QcgXQhWNjrz^TWe}gJG2jE zpOX}sj~`Hr3TM&tya(s-!qJlpBB{#=`+UZO0?`*o|G9Cka<$(-x4rH^%a?AY+f?A)ic60|T^OXZ&2wFJl&saDEg05M<-eA07s>?=(F5H-Uv$^cVUNVvM%#e^6!# z0R?IJxQR&Z__fDuihN2zVy9!gV_!^Gn?so|7_zR31pP;cvB#*VgN2tyUFTT?5;((t zylF#{c@Bu%iM?WFwa?FAw^C}_GM-Qh7%?AHmQ)VOh`8+V^{OM!LCCB7o&5OkdlANm zCAGnreB${5-t^P5s5tzz?CFg6M~STNW0`A>!XlBp9IlH~i?gB=eTYr-&SIP2sFxpP z-}ba1HBh@3{lzU%3mT{Q1j|5H^b%6x1iVg8>BsK;tS^kNpC4b#7?Ako@zLO|MG>At zQ0QcE-P>59$1J0D)kUQlZmjp?4_if^Vq75_+kW%fcd~)$(`1#C+fD2?3d&p6+G;ZYJb3Y|0edm^TlvL1n2)H*H+Mgk-fl(q zHS}z%Q{)JHDhbO8a^%ZDJgkF74L9z~hSKBt@$BzRd3fL9;vYhppLf;$4>grH zb2tgtB!iUA3Oy`QV(vpsx-kWoW0MpQqBGEG`0|^16-o6H?U4Ns$mRc}#(Jvcb zuKs_1QU>;aWcn!*h`r@&Nc+W(^nj5BIXTsp)G`L%i3jM(2Cs`@u?1qn#+NDfc5**t zej(2)f0xW;TC6owRQRdhIm-v*HL@H0fqiGsxud_zrk#>DgZeZ?Z~Ax{LGnCTpAvJc zdb%n7=;RM+Kk<&+8wjD%Wtya5RYys&Xw61BP3rCa)M@W3faYWGwRh>HU93%48=G1O z6z3N4dHCM`fuG#;?#s0Y{o4)>$tdDiqbr)byY~@M>!M)gwpwT_Ln~JY=gq2j5j+R7 zOx}|{PE^9D4n*Bw=*VPWPP>X?%o87fSHamJId&~YMEcSS!YUh|i(W~!XBp{4tC)Yp zd!FSv7%WEAFaKB0TMwnt^Yw5?q1M#6dhEggV+iJA(1^csY|Dz7<8*uLG-;7>-|NH< zZl(RFvSim+8t*^nrOkT)+ImmEDX*2jd-Uz1Q3R{t%s@rXqz3?~tkwVQ7z9(0V6e$&FHX`P9ga=fT{ zD{-e~mdDa!+_jnKne6}#@!A53|I_&HXN zaPOqQ-lL!o_~ITTUZ#MiSC7eRKtin@>CD0+8N0&PU(RUAvHR=!{cUZ0GKn z8(?nSqXo}yl;NX?#QsyWPaeqR`MTe&CEK)f&t4w^WEt*`KA}4QR`{j$9(DYbu48U~ z@W-E!l|3J-b6t_}VNoq@#}BB7OJ zg!#V3_iE~!UZ^S9#(fyER~E|K)}Ki%>O3ci|88NTn8ihrySqtfdKw0t;s?ul4sL+= zs^2FsN2Tf1XF&aZzHf_^fxgat^6n8BQ=I0@GZN z^I9!^vg`DI_e;!_F7GVm!N*%n4+)jxW*XNLh^t6X&W_un4`2JaB=E|2^~!@Conf%F^gJh&o&mc1VUu^YLwn{&fpHh1|oD6>zd? zG_m?pC(G_t@Tow+J5wuMQ#U;sITK$y8X84g(_M^|^yRia*@ObN(Va6lK6wSqM8(h* zBD?~QC41eQ?=Lz-YhMT{c9%2{{zTljD*nwtHwC?d(?gwin4nXa!29uiOZ&MB!fpC( zhP^~Ovy~N#b5Jz3Vg~8Yu7RSm2g#g+##GIoWSMRGwDTGqGWr2-`_(QKbT*P%988q@ z_Z#L)2Ri;%q`b1HyZ}gU5?hxvMeD^^l}(z}D<{93TK4N%#rA25H1i&4C)JwAvzZWq z{w}e-Ghg(|`hWS~x;w`=;q_|XKKknWRCPT;US}naFu}2qwE67P+*^d)a&ch3cE*?Iinea&n`i42f4ddfg@m z*SHu%L^-sj6w2NA z7ka;`%A(YN%fEy%3i37*h;s&6~@-zhvH3}JdUfRJ;gx|Vc zw1X?A%HE)5d!i?%sbVLwG?b@)o8|)ak+N3h z39?v;(3IgDiON}Y?Tr}>PRg}2mnEKC!v^Zn?8ww3o$8z{=XI_>D=Y{(ciYm()ZnMp z1f9cykQp%#*@Tu?tu9nJ9wBF%caezhD`_pnb#%OCK1f9@=bYDO98-)-E76>~O6-2R0;}Li@$@D8kbkSuURXE#d zwBMfy?2Fv_qZ#RsgQ%Aliuyhsx9>9GQXt7YF3ig2Tu{cI0v11A2Cw@amCz#boW;y4 z{T51Fd2_1zemjlAR6Fp{Xr$O*jXAXAO?U}iR)QQKQqcKpzE8HlS7_Y5w!P-mme?=Q zrQ}*g`xyN{0Mh_)8<>K!K^Suj@csJjx3#~`LoTc#->gZ8peD1lbaEf{0qnyfccS*t z%yMTUZa~h0gg0JRRb9ZQ<@tRz#$bwD%pj2CH5cNWZFs18*JTkga&LhwcibjPRGr^`kwu3 z8By)gwEo0ra_u%&Aaal<-2^lZ>M{}nqJLD&vo9c{H{~Z$A{8L36e>T0CLNnjLYe-Gol#QU?izC3h3#rSkk(V8C5q(FvydVCp*Z0s#uPyNC{1BdLU1-ha zzjF}z1K0Hgxge!?1}OK&@E6pMc(+WQdaxQ3<9KOwEAE_Mb~3!ZwIp$B8!iI*t4FST zbLSo6dTJ$Cz;taMIR3i0Qw{e2a({!Fbx~y(jIVp61%s?N!ih ziutRbENX6cU6A)+6YR`iMY6F$(N}p#0FkJ-ZC4kNHw*J#@{n=1%9Zb^71oaJi|sA! z6eEol?3oLrRnVTFONs;nL+sb`Fgv|DUKF1R@0s? zM{{USo(nn^xmP#xV zcW+)$HAndAUau7`jgZ91p{{efdnJ%ykd{PwIJ?C4LroIkW)0P8pLkMp#PURU-uN6% zBH!y-dLFE3Zs_&tU8J9JS(eatqW@E~>AqIfWBZ3Phedh#>D%uFUHqwLGhbGbG8p

l&>z=?!eS{oG4h4A*K3vu7 z68^o1{$hvZP5_cJ?t&x_)UU;#i2hDPu?XN9F03i+(0MnF-I!X`vZ}P}b|c0}?-$bU z2ri$pgcMqC)u0R+GH;;SN+9uNgxfYwWcqVEQovp?wU&Pk(B`;IA=SIt4+AeU$GhqG zY`4DUK4T4a7(l^Ko6AOi8bWQY^a%1e_rlqmk!EMV)6k|R^}+&V9DU!aHpeI?#%Sgo z{pL1*R*aEflRp1PGA=C)`{)eh&{sDLa`1?tjAovN#K7b&r6j*YCTjD;`r2*=*HNUF zr)_&!reR>(untiQ@mNZ`CehfN2)%Y$Xx(|=W+VssXx(yCQd0{s_vX$%T8d2>_|bceN^ zQweqVcR&B0b4xJ*48`^U`;O` z5+3Q2krTtfW=fxA7-aMitp3SP1aevbHXp^0U~h)?)v6UhDcJYt3!rhE=r4D^fg&7V7VJ0tmGx?(nLtMqqPP_`1=W|vf1h)NzBr!yF9!AvKV5VOPA5B# z@=zwc_2qef<`xbs2$~B6@?Wk~3U;@MZKt5OIwan4cQI-@PY)vU9Mo|Z%~R(yzr@+k zpHnL)SIm$c=an5pl6pUTPaq-S6O>j;y^MV{TyODY>qQ6uVMmT$?9@@o=Ch#w>{BWn zrGKLL(DK=i7SNSB#CjaIR5w=J>W)e7uN#nN|KpnsJNgYp-=aPDn5%Pwz+*KdI40?Z zjB>hi_WkTIDgDlgn8Tt1O=V^YYQPn!Zt(IV0`L3HdZ4Vl>HTf79E9|?gfVpgesvA_ zr3)Pk4c(en4mLsdE`=ze1g?Nd;F>$;$v^i*r~Z<7BYiWIqYA##B54oat)kYSH^XD_ zRq-PcSVb~URB7Tpl%4cx8zjm4#p6a|b0eQ8Hfh5sl|LPFo{;QBkApp@| zei!jzb%yU&+Un4usK`~L9ia3a9$}RM!n#o}QwOmy3n33eXp^TnsLwiM&G$&l(Sydk z*>&m0n$Cgd#p>6Qvm(_1}yyt&MJZ6@39mr3G)q1Cv^oXRf2X?cpl?Mi8yxJoS26 z5m<3#5(HQtW3^xRS>579^iJ@DW9SnN^1KJ1bDBb>G8a(mBnmP*1{K+ZjYAnqf!%+}Gv5WCvNLm>26n8CJrdzkmJ zt2MfF(RO6M46lwag~J zPjZGY`X#jc_E_AWPp=w_RAk1j{xq3AOh&$U9z9?z%~*Ja8rv;Z0Fmtf^JhlRk}uri zsAtP!is6fxW)|1oV`gW_9d2THxjLB0 z`i*H9OFTq|%xnV=?*|NSYW@MTztx-QsnHtB(^qW26I$5j@GDco&y{2(n<2nk{Q z6AGlf!Zjp)KxbcEO6?d*dPm1${w?kh>U_4Fc0%k33M9$W#VsK_PD++tv8NIkVI1BV ziYn+0n;kWPT%$1o``Gf5(|kf=aiET;u_1*ILklvUOZ&HbG(>HFz7ks0%})w>`X2L z|0Ag(Cm&e7JpNG;<6Z+LKIX<)u!2 zF@g*gp2ze?Z~>BJk29X4>AD{y%f&MPk&&qk8wRiP_qHq@t!1edwtu0N83Pi)B@_!v zdl^x0|d6n*HCm11vsJblFk@I2F{hxhI~<=gL=m zouFLX=zw`vK~0%MtLAJG#n8GV^D!_#j6^7r5SE4$Af&^*FsaHz{qR0VZb<>sHv`%b zW|yVtY>MmEqjK*%a>(%qa9P)7Hvj78#(jSx6u?bVk^V%=;c?@B$M0&PI8(MRwnZ8B zV1?q+af3ODvg^+?`ZXPYA(@x^H6fh!^=>rbmKMGXXxk8|pyC{mZ$1b16DX>LhRS0{ z`9V<#Q}y&|1@$&DBtzbMqp8cwxkE1^duzPeLF$SY+>T&Bb>XhEF<8oX zO$R3^i^P;ajg$c5MpcHm;zlvqq{TXyMeiwPUUhMF&zZ(P(m2jL*l-DIMKXr$ldC_? za5cI*iaP1)-wN7HWVoKg;(v#^*F%rca8i0Vk4Q+&$)$)wPtuTWiZflIIlYO~UXOwA z9}kOY5u75wtbKTJOan*lj})cX`YvV|%^ zIn%=-xOA0qn3%`7hI|?Xr|nazGLLqbIdkUiCt_X8Uzj&0mD{Kr4Ihbrho0pd1q7?v zVe{yRga+tl_ExM)y&J1f73eipKBb9@PW`eM_kD|_pL zYc^niUGgFM7QYxk5W{GjTzkOY*YMMs8$QgcW!(kPOK~)WZ3ktqaO=d+43# z3tyWns@W4e14Ta!?x8<#$D4qSp^1DtF zdAv29_vP={FXJL39hjyP=+SZ-6JI8uQNlN=5whNi*QP2R{5e^p&8DeYGOpt#*;~?Q zuZ#UfU=ed`2F6rTG|BqWNK&Hv%6%E<4v8hr|6^mNuR>6Md6Ef|tSk9{+E{5$tCR)m z(Jd@HW^Y|#`};weFOA{*wWT#V`Ejwos!MBtXyxd$?5)j{78gB+kC^(4mMIMzS78`> z^Ed6}a`f!^9Qh`Rvr*(q>tsit5d)r>>+NLb;%xT7+rx|2rh3nmh#daXLtT`=n6Q(pM~S5fe3z_$lgUq& z%~KOY`z3fiwv*8%gtkEH!pfP&>PFl+S#Km)+NZ#E zkE=c*iFschg6lI?cLYA1ErkLfNSFFzvV=2IFSrt)BC3&}gl1P*`_b64(=*1=-Qx-J z-Oj&+tKwI{(e2lzPp6>&LN0Sh0Sk+xd+6A&4K7P-#V7Y4@-uHw{BI+xhxY$tWVPH6 zS6D-$UznYcVS&_Sk;eP@*$pXKuD_Pqqq=Vn%DCk(^xWV2-4EaCd~Kk@O-Scs&|Epf zaWH9C(ir^hj~SeV22=Wku)~gS@(TXcOaQkiUO3v+e5e)4sq*rc`X6Bat@Z64oFC3u z)IPkM5b0HJ?_n!rYxF-pRt@HR-2!*$3swO0%MY8bQG`wmgxW_h2OeDMy6l5#XpNV| zxV|SJ{lEOI!~x8UU)OC%BdXYo#J{$du9)vNcsE~OlDISV*-%B4-+u^YGUII8N(K2he*`&geC?(S_Q6x}0x+HNGF%@Kr%3 z_sa*m3BC?}Yhe1tTU#Qa`J`2OK9L-y>E9=%1`mvQ`1Q}m;^=g97a@M< z3&*y6-=Zs&k8RT&h6elLQ=OJKsBpPFL$LExVYY33*%pe8^NR)9E#LP1U!j#i4k1|l zLlyYicP+iGaoS0Pl!HaUG1t@o+a8LtqB}7XJoAuj|NpUvqD|6cEzo?Z=EmzNfXN;+ z-ik!Wl|JPpaFp?`?TF>r_J$A&C^6$unCo-R5ui83GDH4$e!k->GX6A)%NjyG(G>k# z51Duf06jp$zqpw|E60l;FOa^CEo&{ZXnrI*|1*gyrubK|HE<{w2bvxhxj&g5zeYhb zg&MXbsFUYq`h&*)Egv|?ZmG4!TEAKI3|uLit3`iElcC=9}*zW ztnm>E=p8hoC(O9?!-K`qGqrCxjL40+Ye9G)DuIxA+pZt(HQ*v^`$h2jNGQt-s^*Cg zzmU8|ibN(M^sO!C>|9qib>;L&eX=ch>zw3!D zhi(kRzt2I9VxtCR(vE#&6a^g~Z`ZSwrcOxT!=Fm|ONV-QzUk^DU8xQ2_{(A_`Z2gB67G}O@C7Pve3*j^0yX2KzkK}B; zMT!G8B1j%L2po4!TaouQWYGoN`r)CrpQ30KL;itE?MwdOzryWGp~<5s|3%TTrl0x; zRst2%#gn%3SO4*f_+Ln&5N8IrmC_ZBu1*2)=}n&eGnDLFOevAQ z_}7F7Onwss3|<0v47fmC3i??x*JK1un3{{(h(Lsj@wt9Kate!^ldi{QNmtxyaX@c` z#Satin*9V5EISmJksg>t2oxQCD+G1Kd^0?GCM1!cQloQ6_dUjzLJ zwh~jlVkW3}X(`IFJ!+h~LZ^g%t0+bASHemn z%4-a1DLkqy`&)AzD5)h1<^a0HU*?MC&kIRYVhk6)k*1Qi9C*ah2aOVV12ScxK$bZT z_olq$EME?ATprW;-ihiQbBjMG@VA)+8m2VEHU`D|fKss#saX8KWXOAJVgCdfX?sfn zra(rTDZH!ZAS3N<8mhIxM9ps>UIXRmhYrNlhv32Q7XQdn~R@=ZPF(fwC?O(Y?h!PV*L0h> zid+ZW*J%9LFuIsxF8DH>TEX$07_kGKdj6Zd$#GY0s%v7U?#Qo@5|h|B4qE7M%4?7t zE0pLLAmO*^k+f0**9h_WQk7l_(ZsaH?=+;6Q3Z$APccoJg%byY&jzg9#$k_~pU83Xc!4&bs z(12$RN=>03>vv|7Mhy7Q3O?*6{F*7dRO5|I!@)IZ>l`t$eWA$ps8hTw4cQ=uys^gzE70OOCW$&*z5lQ2 z5hGUTbzg~6+X-1B8d9;6YXzOo`I%8gv$c#U&UjZE*FAeZG-&%y7@?Uv|Q*K^3pCO>AVps<>w+O(^;N+ZCkNy-uGN_2jPoV9_UNIYC$xj z1xR#EwzS|7iFBv8@Vsc)?}x(p3oTPzZSE0)=S>yWCsX?e$eW5B;;9fRh2O=J1Uv^6 z#o{0#Y(_gWo{lMil>sW{rbct_+#=(=i$(tF%*C>+>M`BuuJ)$yXTU^$7h@rakUF1+ z3j=x;jMbno_x_6`^(fKlbdwK4`MgI-jzj%^Rnn848ni>U8+ReFjiwHetMiM)70Mc- zBVJc|<~(EWe8_vqUg3L+62t}SyXx`uL76Aq2qCnFntiLkgPcySC;}XOAHCBH{=B3n zhz?sqX5D)flLN3p{!^vlQy&o6b32^~pJ1}L{)@jp;EN8Qh{S&E6nl5Wr!<^j=cYPg z#$OY{?oTT#eSQ|i2jF^>x~>V=#XMOqh&o}TeQhvuCrq1B{S<=$KmTB%(8f!3^<=>& z)*d$F2rD+VRV(#arc-K9=`XIR$DYYwI^HBqdCT5&YL3`8U;7c<16e6;Dvnj-rg6bD zMLvrKySlUJja$rzNM()^)GUQa3{@NKV(fSNwT9FRejo7S0+Qv(_8yLe2m|!$Za|!S zd~-gEB8eJYzo&205w4nO8O_53@vn4FVfJ$x9Fgp5R0LMZbP~p3)P5`<6vU zia&jrZ7VsH>&IkfPccgFjj3xznjgV6I0vCBr@Nwje@nPMZWyQlBN-eCyI9P8Ps29w zQ2E2CNI%@yqSNpCeU-zDceIXdMXwDVl>RV*WN<`2(|p`}lch*ja`8ka_h@uzl2o^I z~*+O^eN?ST;*DkIl3WxC0u1{8wGIN59mooFs$ z{s%a1U2(r=0!b!HzQxU8u4o3m`2ZRTWRQADu@EIEX9p*ADKBs&_X#KF2sc&9ESG-R zk!`+M*zgW(;(u4p1p37k2ClhVBN}6%)QN@gW2O&KVG4_ZI4d<>&syNeS&C87} z)ZFIW24XF@l3Ad?UECi*qp#U>c;vlX6=WRcPUC?vAg5$&s9sRGhbM{=Yz@}N%NW5* zU$dORJG+?xq=QH6#4Lmo>(stEjgupHSopS(64X&4iIv2l(`~Iaw(8;p$%XE96o}(p2SeWQ6=yO9xYC0-Z(_` z3mw5hYX<01fc~KIE>`fG{A+OS#*ED19EC_>+Ox|Eoc$=g0;FBbP{Ly=dk07v8xn8{ zthAAtg2Xkhs7$Bc41sz(8a9D4U&C{UNbdGfp^s$T0;h}69Fhe*seg^)?Ak65&bexe z$x2Z<@Mtq$V~cJ5w{M(IK&@scx-?U;u;j7d8cO5qydPZUls`eYi@od+nGZdY95~(Y7kBmY{%|`Uki!!~YM?PqSJCx(@fUJ7!wTp7Sxa;K$H-cBEyb}@!_r3X`- zWtF}x6ul{0G;Gqs#9=DTUoYq&ur9=eQ7#Pz?19;JXaS6W67!R zF;w9%OZYmR_Iq2A(ag@DXGX<==0)fN4oPTu?^ry-@_bty7Y!Yki> zma2CdeBjdRy*qh1l+3`T-iBM_bp<9`7gqdPE^8de^;b7M6%chGNL)vA`FNgT|ag8@5ETH zurG>!|4l`Q+ge_fLu)d{_z?w3x&h@4Dd2QfhX|WiD~k3nRM2-N_`_w30(OF*vx&X9 z9KdmUNsO&HV;m>(!B__?6e4FyV6%Q6XA%eQ5Z&mC%7~#qhPUls!r&_n-3UG{RixOx zP)k6SY6ZLj00ucA4 zFdljH-Vx{%veP-klU{98QMiM#T$xdswHPY7PP!>dD(m`(mM*iSr9^Z&%#k4}{Zof>iw z)VGZ32p8GfPbUDDOY19l>8S#AmBkOInW62{y&LKIxYXRW*DI(h#pzgpQpSDv6Af{# zXx!ALrtD?CiRiAuCsO_ZL?{BONyp4zkOSTqLidT!MM31w%Nh9nyF62%&;X2;>~a4L zexVfgZm&$c`#Gpf*LM9N(GO}ao!Roc=>il_3+qXLdr<5bvaoDz#h=;jh{{i5+5E;r zpQ?VF8A@1am~Yp3LcT3Q(lG2Odw5Fe38-Sb1WVbG2RNQ5)t~+fB7Qb-<3ADgUgzMh z3O8N>8_Lk_7w^QlV3Spk;;C14;Z@orQSUv_e|G%R8XeyUIis1_%I1f0^$xD-)cyUieWP_J_a$5+xXrKPWjG5VMocy1SG!3-yk4S078mxiz zy8c#SiqFHXFb!aZr%U?$iuG=sZ4Q@Py$@={Q=8gykdjkM6NZ%2+uJjTdajgpHMJo; zuO^abVXXN#FQ)+{&6miccOt-<-!N+cHh5O}2o(puE3|1p6G`JJ?s`}85*9X(HbquWPL>RX z!wEgo-7{rW`6(d_j~ZywSMc5G6mT8p8YCL$U(2$Ujc+8q8Bir3&nZ zL&T(p!U2&_yng1YJR4b&my!m2JjEYGU%71+N~>b#jhBWk%LW`L=2d6m93&2uBM1$f zD5`UQy3YbkwSwM|v3e}*l3R?xIi=)c5ol8WoDo4ym1ke~cL!mM+Mb~>z{9GOzwJNX z4~r}C+(0eMke{Dnuw_+3_3spQE7FHow!)SmU$!YZ4L4q>^n-itl~19LMzoQn8$Zxv-^AC)}d z^9(p4r`S6zK-D9HFO^NuZl=VCEK&&>DLINq-8U?ne1~L-#a}fC2z!Iv{ zuSykPg6YhfkqzBt)~W66Wo##U>6Cy{KXFKuA);fcKl{->UEqVp8}R+P;M4))ojn-i zLIYCYxQO`ApOpA~*xze-X>M2XG`RbJQY-J7(()&83 zKmG*JxW>_o3Ul0XgcDm)k8W!n=cKQ7G%}*b+rQlIandkP-bXM~8AiH|X}iPT#}=KZ zF01y2yc6&}HH5eXr`zkXZM+#EDv3T9912hWv6bi@<2epA-@(BDao|!fW2p|^#}}yW zt%j;-VUb)OSEw%tqgG1Izhxnjk0?c9dcSARqxtR>v(vSV$J8#9FF6zAj+XVKZVY5d z5YMQy74tk*|DuZ4Ychj>eS7mpTMjNqIR6+i&z5ib)yZah_BD`v3`>@`p`W>Fo0Zi@ zz5G?A@q{uWao4M2o#PfXGLRwKd{HcktYrOMUxD28`@NSQlPX$j)(rk~z9L79LDzn+ zag2X$IC2oW`YJI)N`0Y>Rvh^aq6Wb1dk)8xI8iRnX{m-xM{S&Nf=sixc-9N-r)p9U)p&O|aKeqY4mDMBB#&mJf~F@1amH|M1| zeXftKe6hEIV<3a=YUU6srol;WMNKP&)?=j>vDI{fu#r{x61JcNWsL~$p$CnaVtwGz z&avRFU1iIlwF>Xz>H(dgIb{Pw^ViV=XdUvtXE@5_VhIGv*b6I z@wp1d-{@+^;ZlJyP95JrU53)He{s?21oyK5n(2A;DY|)RVR@W1f@K5HtFw68mW39| zP9}>Po{Z@dphvDZm}vdOPCu&r?IsP*TLaXNH-<0k#~F^_(&a9V80>9=F4fftyMP0& zR#`kklRG+LeEsN)jB`v}9W>N#Kz_+9)WR@8gQznGygsM==O|tHeW4@oHf|Ul5&IZe zJgwm2JPX}hWhlplkU3n=@|A|03}Uv9V=!i5;`tuz2iPS=Ax63y+`+iwfAGg2SdY{r zzc|tA{i$AeV*cC~30EU~2p%9i)Xp%nQl7A%Qy<%S9eqN>>)G3U*jD^1kdRH?RfBC+ z?BFEDO^dd)n}Uj|pT3txovoyTi@h1TyhT7`%T~WX=_f{{-h#$PtIv%ETB*su`qA+J zve(>9j&%sq?T2;!KwDjfD--PZBH0fn(RO&WG$g@0&%)Mv?db8jsOQ;aF z2A$n3=9?2}k6b|CK!4%jhXx2&g^ITBH6Fcat%JR*`C_=wT7Sn{ZiPqOQV0L{2MZm< z`_cPhDIEI?k|UWo?-$k}v2XojVaYR*a;yg^-@;PTHd-K~kRb4vrZUS4t1Q zqq2Div&=;EG=o#z3@DT;_qRw=C6;8`a$J5Ciqf*As^w;cOCxBb{J$RewIY=K^6%d4 z%X6X-ZM!vW;ElH)D<hddU`f#)n6W<76G|q z=NWB*dDrypvoId0(bZA<75$Q2Rz5^M4-C~d2#^#F!tF==VK6c{%kSo`69e?d+qDoGN`q_f&hDW%Q|7YN z=sRzqn|*8q&4)2Ht|KKo=hmMo>3l-^I6o#$i5-I|m3U4G=H}pR#v^wsB9!8!clbk*5XP}%a07TBE1Z% z_L!mX)_B9l(-+He*{E*S=_yV+d8E@>0Ufsud@+_)qc>R)PXs)?N4s2vw1Lc%EyI1R zF)oM~JFG=pGT(#)TEUgl7kAq^p(b9wUiO}J$A9f$<&}?i^kl;A34ro$$1=Uo8msov zfJPi;FM3#pdS^w@K3kTScg=7QoATOG+g(et;t9ks8L^{ObVAEJ1?o>U--yF5cGLS) zc@grpyf0)W_O=!toZI%40~EF7_5RQ&JSOFtApuEi{$bC=8~;XK zbZ$SCwry1J5V9Z6>ex8QD7Fi>#0-pp32Bsj`L^a2X|s~ATyM<$UHdk5>#tdig6{KG z`~N^W@A4=Wtwud)1NSj@gS|{deA}Vh^;!^N^dpqkll;=|VnOrngn3nYfVUX_du=wl zEYf*t&`B2Y%=@MCnSSEUMD}?|ppJk>^&yJq(fyu7@vhaR)YrC!$QkKF@mzcODAvrGx>gZMM1l5>yrLuPIYKSWvd9luliB zHeeY1)m0z1i>8p@K2eJV98ik3ex(DRLL>EAD6Qopz1Ui7Az#;0`)Dkw#c3KGD&dmt z5I2Vl1zn|+ul415?*7KUOC+woXzJb;P5-{z95IiY7_*YWw^h)|v*(+Gj6nP6vaEY( zpC~;d5?UxtfI194T!kQkYyZM-1vIMAYuuVXm_g@LbWzb0LrPj?V6OewIa>YVa>AvL z-H2aq68-4&x;|;ionW}Elw@FxK8@(W*2DrWhlA%}+cJ_grMbHz25A?lK7Pw{Y+Yd9 z-3oxzcCmqmsCh5$n*;^d{NNHaWp(i(8$a>873TJ@ zAo@A$4(D;;V-^-PVl+!qByh#=T9(p9L4m(0pJDB8vxgWR((R@BKDJhb8ZJEX6f6fv z{wom2Sv>}@_BYH46LtQx;1lsg>S6-HnSBLVCLrIoNi&F#l;?EnSu-cJ88p-F?$*f_~^ z;Gs_Czmvqgs7f<3@I^z4oF$ELcqyn0&Qtp?(0L+r$v;|F3e{Wf1ZWfCMh-`Ohan3& zBUe{_Yo*QR=);BXpR{76TzBe$_4T8q7aG?|_IY|buE)Bp$iQ1;MOyu=kk#%-A4>KDHMwQexffGLhOt*LueEy*JK6AAF-8Aay@p#tNKU z_biWHq#jEt4vYCP0LZ+^JK(7=exJn?vC27*5R_eXyef-))i&G?eMq#5qi={Py|&}H z?MLls1s^1yd`N$GUU&luvFB1dmG!fDNQYLTU4X{3Zng63G#|E+_V))!&2vZvnxj)x zQ}$6huI36+@UPpazj28p=2GvtY^Xlb_}1mP+y!lXovS%9247ON^^S*o6BRD=t-cnz zEH>WL`8&Ws#Oh%T=);`9l@%7F1(oPqfYck6S*xyGN1-YX57D`_F9UV{M&4g*FG$xS zzC1L4$lnQ8ehsA*Ibr9ZECTN#X2W z5W8i}=)5+e&+2vQwso^?E7-W!e9+8xiDLxn55kahRI1m07wj8=Ty!3SB^muFnDFR9 zKHrD+y<7c)?(=KMx9-E|CgwMc!9NvGn;({-6mKQ(i?3o|{pImHqmX#OgKsOyYCKd* zCDlch9|1aERf0Jr;l_PGzE;HP-*dVX!Z0%GnzkEqOd_fb1_}4&jVv#;IA!seoZ$Hb zcxgd0agYm9d_UV0b1i)tb~O%)<4CE6)B$`{Mt9DF=)~KkXVm=u!aEzf9u-z}nZmFb zeNtsujMHBnaKq~SBFgj=QVXJw3<-sFn`|y~h;day_fF1J-zWu}VJ^HA(;ujI?{i2e z6O#Om&mkV^ymGUyf4`2vURZkwTY|-mUaFsEqsH~0JV3I&)suVV_tdjLKCwImiG%3= zou1b4)uJvSOx=^pL?;iN9a5?+5Di7HV2A(wt4QH`aHM*IyFYVFIDUvEq} z|9AT}rb@w6paw+z(o%nYwru@SNGi(pS<19RQvJQB*^bwcMCICW{85vz@ zm@7f`nG0lwP``fZ2DiT6#+`W|%@D+>2_vYh{^qw-Re6&Qe7XENb^3z5AU*ixh!8*% z>|icNtwn#%3O9tjs&d}%ZiPe-VRcn^x28d3S->%^T%7@q(fs~OtJr_z($lA6p|?fA z_HY?MOzOVqpo8b5b@N>8!_P$A-A3~SwAxl>DP@RPzd1m1@jnJOQ<#o9u~9wzCkh~b z{l18XSKyfgrLuSQA?%u2Ov=SXP0l`RL08GB{*^pxr&KEa$1SC`YUs>Lrki!3Edz|r zh_o&QXf6xjI&#>dCdA0$QJRwYtt_4tC&R0>9`nBHdv|F%Bc_3E#arz`F}95zB^23k zTK!*Gbzl&qoraX6BO6(v$jKoR>k|ltpE^PHHY2q)>RNdtFJmng`^;GTtS=pG_v=G` zQSSDn<4`0r{a~Mr)d_(%zFNiD(dygs9_S)dTOr}k45}Kz;_om&x-WZxoH!HGwuqP* z%7jp>?B_JupvZMY#Zd~e>!xZOARY#nkb|_GNk40TQ9`e61OG^y8n@2@=jr6htGb4Z zs61x&(lCt2?~|)k9kx(WbA~Ew;#-WM&<{ym&kk_P@Fc=-RZyO3t3VpI-L938nE;)+ z_5jo3vjcGaK8NvGMD;Z8N8esfAP0O9`5?e3!Ph_ETbq--yvm%?m5Y$7&X4}(P<}g6 zs!{<`r4^jQB04Uzi>>8p5?@7C9U`;&RaCQhrcX%T66pk2;D2Q77y^$vCRI*v zC@_5`SVkvjQ&ygm>)R3FeQuHMniRUDG?Pyq0PrS;mI0b_EE8(0*#7G>!8hdSWIeC6 z_q#~XpF>#I%B2{Yq44oYd8X{+P>=a$enqwtr$cwIHaKdpy*3 zRiHrS&4Yb}S8NC=JG;4|mznOXec>!vcxI;>y8p>u)Y4{}Z72fwyMSPRNYu0Fta#hp zxfS@2KB1og{7U;ODfU@83mbW{gigMIpR1~ty*e}3rLIu>I3hH>Rz+Fzk>-|({5paG zvd1l!X3dHKHn{X8!`a1zXX1yk+1F%|A~~>0?^)-6)IO|2Sc+b0LG%7{A}TX(!noO;Z?= z7|VE0ee@|liWl?Ik4okNSAw0Oojt7hB{?|Q-tJ$qj#!&~hlt|XgOz#RAgzdgbmoMr z1Z@$Hm#IDu5J-Kma#i|;_jL5Xrz?f0UvGY~?DBuH9H4WNYZqc4!&b#XKZ7s`C?6I` zTjsEUtadpcj(tuGw!%CW|LW%2Ff;T@tAv6$+jSQFXv18*a?Naa3wm?}+x*`LF7nKD zi5V6O)a)~?cR7*%Gs62i{yY~i;sN;K;;HAh6a*@wWxh=MGW3sP{!hLQfr>;9K3r0| z3}O1wZq%P^-#_dhWT6G1D!Isq(T)d7KPe?{pGc6F{upcVg#}1Jif2o&4APFiqV1H0 z>bqH>R22)nJMn3R)b@~`5ME}52^n=N@`(i=@OsHXune%40iw0gDt!XcnaJOVpIGeU zDuwz-!1TPw^RT%%cmo9|=HI_2Lz3qVabNj>7Ls(%r)t)LYuXmNyrd{^&Lq##bcJ6r?=1uHHr=qQIC-u@e zQdY5~Y&J%LqPc2|y8!xIzC_o`ZCuT5BLr!95!IN6V5T+z^7p(8Aj4hlL{@r@sPQXC z##(r~e3YJ##K@aqa_OnZXh4MLJlb|-xsBPEmZ$&5BJ&q;(Y9yu*#%mjnkg;M??{Rx%JXskbaU$_ojLcfXl!Qhw|Y$A{8j)U`i==%rzUCps{8#>~Y_S>*z2p7=K zKm+>#)KtDwG#$E&Y2;m=DNeoJbQT@2j{os-mUwFEJbc^D=MwVOiTAUne}>=X!It%| zyC;cQg$+xpBfYOb1TzD`_XgxjPd5_Ww_JWj0^>SgB9mO~IyP;5uzmZ@n5Fx7e2TT3 zFJ`P@(;9QG@Bv2Xx<+OQ+3ObX&fa;29fw|7y|4p8*vH<+<$syhDcN~Ec%H&?Flbc2 z02S1#0Ke}LBqIR(VijP;m`?}s7!{pOiPCdV(jYlpaj!yHE3cORs-+B9s+2$ZWSm_5 zZnMmzhfjxg>n#j?5iarE`WyLIQ!-EBb8EY7kT-9#6iunlUF?VoJ(CI!@ccB-wtgWC zW(V|r;9V?t3e)ZmH=o?h3(&rrX`SN-o30K z{@AzrUq+SVuAq>dN8k=w(!6N+XE|AnegNO3$N1MklU z=c1{j>yurNo^9rUbCwQA#y_MYX0^yI{!|e*O>|!>I2Zf6*&=jj!FHlp70>DX|C=~P z^MCP|{$Qe510QWEA^dWyZgpyVs9;O%h9Mgt5`UE3a@Xs^g)fJ|K+_T-ng>3l20ybr zW6tw!^DQ{{O6k;tC*WM`(g(&u0uo53Uka4M=TP<8{aze0@NdC{XqF#4iXS0SbtHz4 zP{wmU1vZPoxnR^cXT}_ikpWwFuKyLJ_CKf#{oKimHu(0AY;1i|S2NV9`yA0lS-Rt`!a!Z{8hKUe;PStdI z991s~4Qgdu2f1M;Zuq7)^vXu6;SX!Ku3oQ0gGoZPu6~;H9e9IV^LfUgb7Nv$1U2#xTnppQ zY!&YHOQQeVNr~~ZjBfqKsO`enWv?%g{27-zja!k80|)2}hU5pB{X_+ICTRD!`hV4} zhyKhW2%CKY&K(Q=d6MM+b1ITiuqu|9{u!Zi7W%(~;?_3Lw`*q9@rU*!wjUF$XJ3PJ z7NI@8r{ST8glIEj)&Y)bk9-HSU!*-}jM^Rk8L5Kzt1@UZLS3&3{_n})M)uHhPm8|{ z?7Ken&d2??#IK}Z3E-UlZG)#7?W%a4ny~UaiyK9Q`;?Q?w0aREXCYJLE2v$}F@Ibe zitwg$dvKrHH8I%7xiiB5NZ{cKMR@-dDnnEEY4M!V*nle9>9vNG3L4z0>tl4it!0qY zE|$Y4sCql$A!EwEf=7?(g0|7!7ptM&sTvI_2k3*T=^#5El>xKtZ5+vQ={dWLHF+k| z$3xfvPQ7ohT33F~$V(gFSlz|}%XxJ22KV5f1$$YF*ww!&^a*#iXBa_Ry`tj1ql}h* zbd;bb?l5OVsoDxqC_e8@xz5H1Q=vbd7bBkA#S&Pb2lTxj+~*2X&N@n$cW_JyNIHjs z|CPK5*DkiFRizuKRzuWp+p`Q`xDL!lLUPys7)P^y%u^(Vz$8ms^ z)gb?pTHN?3r{_o3D`OMtI|5zbY<(luxVW(wp{#;WeVwB-3YD2X10?01;Z-E?|LnWb zZJiTq$I(%{<2Of_XN*`4cwv_crx_ABaKnM`t#v!BKyb8^+PuKW>e^WJmjjafy zo{k!5`98+5nAwdGBk+~dvzwsI;XRnR&9=2{S@Z>Rev(4~2SKf&U$vLPYLq~e z;Z>kzPgd$~H)6uY!J_f~$V@^T^h8JE2}Jf~&ToEL0qv!d_USRF87Rt|m+_qLh@D^b zyN3B&d<~UzluBOkNNFwKww}pt5&l~g3%V+q49`OXU%QWsES67SI+(iqVjlLDzaSPU zGHQM}q>uW91f#c{o=*C*CQqXXDJPMBB>yR1LbT^kno=A=Sv?jUD0sDAZ2glrv5(#j5>c>abTZty5&kERL zC^R`4hq7LP#oUF@u`iZWJh#fr|9<%h-}wxqkJ!^sP;mBKx0Ye>&D`z5dfDtv;M$X8 zSN5@#SL^&p)5wo+?dvLsZEzj5_ei5;M zhw4!^`i!a5bozH28scxG$V#o9wYE7q;GJS;l zz9NZw0!8p4XVz{8(=}8jISJNf>5ybfwQ9+?8GGnr+$$6N5SL7mI7QE!CVAhWuB}_{ z=?hET3$jv_BqZ#<9ur-m=+Bhbu!%DYA6s<_l8qfI)0d<8V7-2CBX+U1YV<&xQ)_^K z)P!-(Dliyxc?gfX%%FYs8z(;7pXAo^6Y|(xVy`X5xK02U;+T@_Q&k?)iv!2xk~3lm znQri4)cGX?6}*m!o~>S)23|^m>lp4^V_LGt8_7Pu^z-Z22_2s8&Q?hQx0GhzIUGl; z$}PQE^!vDWz!M;~TBG>a7l&BhJX8f)U=LLlM&JHQI{t@y5J)G{NQT2ufqT_T^Tkpq zbZN|Tbg}%r18Q-j_WjTV6{h2L$N83~#=H2M=CwZ!_l$o^U1A$fKeCfaPg&=M6@UEs z{QWwD`sO%&Awf-?xu^>M`|hN#6;0%oqI#W+b0PaY7xgE}*pEMx?#O@%uX3%>#{hgN z#71Wqd%AH&))0F0!?)Wk7EqkwH!H-A{S!ozJ1r;#)-vr$ zqdV}Q@%66H{24io@>9yc6`@<&coH`23GXm0hd$@;i@(t2%Veq>NP1SuWL zczba}uK*4FQf4LO)x=+*3{yO1?EM&Y%{qzP2QrT?-hIgf?ZVRDQV?RA+er=M3~1a& zPmFyrNdyN{M=Clel+pR&{>jfqPnXa;js(tzi~;oa-txBx8Hcth%t5>~#~6z?uFyD2 zRP5U;{DkPgaPJG;lB5Ev3|JdXdf`|?@sE-T~{8>dc}Uy<~O|5SN# z%Zx&4W9IRtIbMU$A>zTg9933*6fJoFhQ@*2IAHa~buj>5-XR#X6c|mxl~FC&jZmgkc)L!s0K z))KyTgs`rD!1rp=ZCnN0l3C%_p4osme8BK3B{n;&jz00{9%FM501A-B8CQ+duG_#-v(kQuPAO zamJUDvLf~F#4)I6?I!#mhJ$dg=Nh478nvamHEIXdaD{$*2W=br(aDZ1HNQT-N*sB# z_3_x|Jcd7=%Xyr?MQ`QQ;xGhF6x1u7vf0R25O?KIF9UzxDUP$vC zAVXOfEjY*vPIXbddOTr+HMc$Osl-|a7T<7M+`x$}^|pkSU&%7v!~QXbDOt=}t@njc zUKckXU9g{r1}>(}94c(1_l0)PEAP?A|CxEgQ7^cLtfw61 z=;Vb#s&Mr`z=1;SzEox(sQ$t51;eQApvW!cV|I7={i4lka`BZh^lF}WSi8tK4$_Cx^*}f<)<68}6oS(@7J%4mq zbgj}EfbjLxB{v6r#AOzjT4cAa@3BGK355emOoRK)ViPs=i85g!Co;gb{`8bI z)yGv|=gg<$RO0QcE;+~OYFq;W%Mi$B~Rc9{1EU z-GpMGaPRFm$jb()+-n34O;F@dj2_f;JX}9J23YNUertS_3j4UW%@Su9*nV*>$+tl|4!d=XmUWG!YfvtWa&H1d(4JeWQa+Ef-n|cO!jTC%;$m3#<-pH=3#TM zZc&n~9gafalB~+dRlM-u`SnXIRCEAIK(@bST6silR2-!1DCXM44FfKuf1*EPw@!-B zz}fq!6?Ig_u3E(yWp_{7NbM}-fyJJmmaO!VK6C-zca9Zi^ zcV!qHXJ*?jc(QP^`a1{u*6{sw!wqaJbd-J+^0||LpAhXevy6ep9ReNV609#3aMKRc zl{*Ao??4z|0+98@ zoeq%adUdi)W&AOMgGb``&$6Qrk|wqNbEXb_gU-3L7I}dFd=%d zIH$dkcO9{kay>WcUqXP12B$NW%V^Hb1ZY5tES}>4J#8IY;lxYxAm_5bS&SK>y=2U{ z;t*K}o=iNz?W+BCD`2A?7KYt>GDM0$M9aNWZy0s+!U$$^z)acE+CxdlmQFLW#XC!8 zt5f8bHX}b;hh`}BM$xJ|dmhp?VD_L?M1EEhHEp~o$LEp$H7aw|a05?%8k$3aIV!K-*pCC!zUVs<#9Brqp-xEXoidBVqGtVaT?Xq1t%ZLC*&yxgCY~%Lp=)Un? zfcKem3!F}bj(^l@XgICkisT;q3E~;C4_r>5(-J?hX@BxfgeF;-p=~j) ze;V?C??tSc)-8HcJ%FBiDB*y|P;@RT2b{~7Lihbz`iUjypOi#pr-i0K!OtXiPoOs5 zb4e}suyC9f$%BGk9}XnRo9tnQpyjHNZiF&kXJYNceL{x}1<`}x83yFZdSk$!>$eJd zOgpQPa)zGJYK^M^Ty{ZwJ!D79fF=5Ld=NnEw@$L`jI9U)|}mHD42Sb*I8I zr=x?fX@CAvqW)I11{#x8o7KMsVf61o=0N>k0U8@h@W*`(R*_Wa#P2i`4~XPV|Lp(rPM!sA^$qB;1oM1FW3E^ zWwsX`1T^ll1^hjUd2jt-XBzd=@v4hVaRT@-pFr*DKt*ix-M>m-u&0Wp0e@!06ct~u zBh;C{A8)U!Jtb?6fYF;CUcYD54m~B5D@Rad))BSP%60UMGjS^BV%MaAOu2X^UOfTP z->CW#0aSu0m8b{y-Bvh5&&YCaiKPpU?0=9L*q^S?Dt$y#3h$by`iP6{p}VbA^*zQ9 zl*%?GsK9|(*8g8PZ{A`xr$qdHe87|8B6y%Au2oIfs|H;W@@{wOYPMQ}`3WUZ)Ki2M zp+j(F{tD!GTe$ z%7=+i(P#t#mke&-De{>|;OY(Imu?sGkb7U2_%WvWP8tB9@M+v4IsX_llk=6zQB zAY5?%&Q@IX$Mz};H=dQ69Y?3tN+!oaYO4*z*bRKgA_D1{aCQW%6?2K(|#2CKiBVCAz`h%F#f%Z&e)ez@3%}o24=%wxNT#fBa#D0LPKoE?adM- zWaW6SU(`4#X)#@F+btbDR?k@^*ngqK=jelIK#8n{i-h? z2kT_ne2jHJ>vx&8U7K%{-ECJBp6v{$Vp{scX}QS+986_I`pOhr>}k83w|MHJ?Stkm*~Cwu>N{3c=aa+q zhvxyWi}%rVu3ys&qmQOoHXAST&9D5u^)>K%ElU%pr(-K3Iic!D=$9)A0!Ia12%bWL zqic^dUj5WsFwHL4S&rXmYW=4znN!TiJ*=JgV*a8X6mA%dPO_w#rh-0T7MVbxeMBddNUyj{PiXBk@6jkb;sdz?GQ~S)r)##z)YLZ6HIIr7m zKC_oO><%{zCzO>-Xl1PBif3lTgx;*Bd$RilSVXCqdT! ziL26+!4R7f1CM$IAMtlkn zm*g35xmOY~EM;8p%x>yVZ^ldgaJltpZRjt+Yap%k`*?8TlZEOJCz^kcK7j|}``BK; ztK*M)NIrk7aUFiY+qa#1Ydte|BKmJd@Azc{s`N`-&`?pf14ks%>Xe!CGs*Y|`%XA? zz#bwV5k(t*(=fHM^v?D2@5`4z2Jv{kHSSeIfP06a`-83__T|NqQbmK8O zp9FXB?Fw7>f;BD|HSwq^!lF$=X)b4T_r^B*HMoH5&_VusqgH{8*4bE((BXjm@z=zn zTQ4Kx{v@|G-@Qfj zAV$`BTm$UoMtNM6fM+?H&y5C42OOox25CgdP6|dChZws;u!q() z!eQuBRb*V4%o>w`f`{BS)Ld?*z@?qiOy=&;IwrbPtYYv5>sKtfe?4L*9vqe&wY4x1 z^d!iS&8%9&w_cM4Rvuj74Y2B_}uYY2g_)U}Dv@5|Yj66WD1$K?tmzJ3z&s59rOJCgcs(Yl57XVd1T zMxXcZ_d99>m;3elJFDy0K1a4Sa%HbQ6F^!Czn?iin_xRUoQCiH?7_#(Tf%|}->;$npqKnrq)@TK|I|I&OYTbD zNSwlT>02q)o$CC_<$a%xQ2N@5>vXB=z$KlN%GtUYoN$Tk%A&*{x!3t#pQ-POm->&U z%HliZ{HyUAr;E&1YleI#9NkJy&@+r+8#tK!!@k=`oFCIH;(S28_gR1H^t0pXv>egj zgT;Igmru`0&iKA#iD^v<1mo@FeYV`|qIaruxHi)-S7ehfJ$PBGfA!!tLExpvD}ITY zfqP7xufEMU`P?@uxljXFWLf66Rm($#zpM%mFdGY8JPZ-M2ojtLokcpU0(hn9AH(3IoJej!p> zVMT{gasAFZixWDxS&09QH_NgAg_R8h)fe#UZ-*_n!fsQIH5bFi?jHy(8JJ*K#}@CJ zJ{?Q$8M9@+lOLPcl#TPd5bF`dT|1@wW7q&xfWNK;Sm&?-Pp&@sn=g%_O@8I>kD$A> zf)38UGAF0umKYQEc(`Vs{s@Q&;t{L``js~i-?t-l`==SD>THVOU0o8nf_s-?d{)>< ze)p}L|L+;L=<1B&bH60v-eKNxy?!|dic|}aul(UjzC3usOq3f+?fLrzk8*BZIp!sM z(Jd^G+{pGHBUL~f4c%T~sN|l!E9owpNNW)a_-^0{%`s~EB=tw594+`|gbT}frz_Tn zHKOvAs!urWI04GafZa7gq_sOgpJrUnr^%xN-Fm)_X@G9f)_>8SDu~CGfC3 zt}PnsW0*ohU45r{*pt(TO50BqFG$4>v*q65%|A@a3=SBGgED6h4NGg#&zOH)$zsgJ zu$&^p{dIbH^a^sBT>%6PC4oyj6#<&JN^%^Z3PuA0#ovq#L}=hW3BWKDP>~?xV=Q5T zKG<+asdIdH49olBJ;hxLVvE;*NVRc6Ep%Wgt)^Cj*7dlV_0=I{!K3lakquVVSnQY) z4uKEZ061rb>#*TfrK|nr$I*DyNi14xI=20PHq4g>DC8P(Jk>Wy3_wP)^)Cdx$4!{c zH6H^i`=Kd)v)qK%)jFsS;^-S&bn`eh(_Ze`Zwn3FdeUCTdo=fBC}y0cNG}4!Mh%H; zf_~$FlWTAS<^u$sT>^LVu{0{?1$XzIIdC_uU74ig_t`kxC?z;ehR@FhGxQz2WMPw+ zoUj7ahs?)v!spT!vQziFN33DbZi%pKT8VQ^P7F6_{H$4ok1y$IC?POqycE-A9bdFu z>RRYe5G`t_m@~jYslD_gd3)AQK--xY6{el%>i~J%_~9P^eY~Wso#2^HeVu3|j%}j1U_NwKu)(Z# z6ZR9qth4MA4Z`p?OM~)3XP9Fbw~2rsA6$DHsiyv+*z~BL-xX|_A2i4h6Erb+z6C2C z#>+p@F@pMygFl8{w4;VCvw6V4lCqMd%R;EEZLt+;fVmuLU#5MV$@uRhPteh zHbm;}Eq@TLQl}0ghNuL3uW|=`PS#Lw6%*GnJp5RRjI(fxIm-2P~qzEhX_ej6AZiVQ`%2X&DMSY_2xg zUCd8MVkdKEvV(z)-T}-++2+(V69h4pTwxJ3+M*fKqnu`%!%32_<$4D1657y5ur8Wl z>hzpV`8NXB75(QTny`7)O5hTr=xh^5l`=OG;hHTy^=T@E?ds*xMWDM!RaAq2nFI(a zv=99Q|5l-cf#-;$u>~cm(QYp@n~p!d`9Y#0T0dq^rW0vi z^icn1di4<@!^3nLwhkkY9>lIzh@l|VNE9?dea^+>ENhui#4;k-2zA!f;RTv*oG*06 zp(|LImm(*^;dev~@swt`c*wVJMTI-&DheYbZ)(5KCG0gOZT+IjQ5NLSg(3ati&q3DAXC(f(WOQ07mP)vdw%WYRh%#TffBxmbiT& zrp*Z-QvMfbMC8IXFr;@e58Wm9*k%-aHxSh( zcZw7Y%{GpK0bgB}Fy~FS#G#QQXFl?kcAst79AJ^G%po^ zlSImXl^cn`rx}1&f0ddW6z7xyn;!l4ahs}ud~OZ@U~(f$nQp`u(6EwPRe_JxHH$N) z@XsI2VSUyUKfV4$LP1PhBOcqrx^8N0B~Yi^m^t}^cD^oGD7uK#af@LBPD>4&)BNF_sJJ`DWifJVaJ~`@8*^`>x4)%vVcagou}%w<6wO8Us51KCEx~ER6$^zVtR1FP!%*gvr+k2A94oWd>F= zVGvL|I`8_{g5o&1WaBC6=|HaI*96{p=LW_?RmRM*Hxn0%*ZWY=yf^{xcq1*MDbC`t ziVVf6xrGqh;waLez3kfP{HS^{s6`F6iYJipQL7iWd~0N2M`cKr*NIgB8|STpE>6!y zr=KIxnQwsjo5CF55MD*M?jPG+Kv!Ji2%OeRNJjUuWsBs|D>SC+~JA`_8@rB2l3MEv!0VMA4b1MnAxq>m@W&c>K zk6l_X^wrJNb0U6?!>Y<7S(|+qFoV{xM}sp3Oo;5Ux*Hf{%mD2$Ymo8~V?dBlbQ<=p zN|}NG8iawTA~e70SNlg+4UA_HebT0ZuId>n5iups_%qjX846IZQVcd(%W}KxNo&Gv z{^dFPHg_ECNF9M)<3Mt0FzZBw5GRN7w8X>BDT;wkezt{%ps0UFo^8$lg_??UapC{A zwY9ELd)ytL_N$Td!d{2HGtce+V`gWhbjFUWyw^NVdCktvf#pr0ajkDY723BB)+leZ z49~VQroX@_H+CIabnd%r{pXlY6C{QC`R~8$z(+LiXy7gj`V|5ZbC9j;*njG*eM7=G z=xjst;O0L;tS=0nErVHM?WpOAW2OT|-jV97P@t4LdqRtLtPKId7{+C5^e}`kOc?-i zVy@@ucb*Pa=3EFWgdERl+Tg~VD2^(#T}+MeuN+D70LmgBJ6U@ghYQz$Q z(4LjnU+I2n1Y^4%XyfZa&AT5tGX?q^*g90yP8(>p(Q~*#SGE1B@`Dt zi|LtM1$acrFv(o*!-l=+tslBz!cRp^<3pWQ#yyJ9OBHBr$d=Dsa9;6Af$W znRHj8vi@;&oMW5L1UPf8Y74>M(X=u!j|j38BeL6m(_c6WJ`h8^)mSwDF)@Egt9G3a zm;4aiCe%!qL!v+>9XE)#787>C^WSk8Gik`Aysl^PKo*fZVNuo3rSQmqu;K(RVu=#_ zs-m)B1V;LO(Lun<)q1@#QKOau+Xi5_d4z!kn3=WI3H;Rr3x{T7< zz;>Ze4Wja{~|pXCsOyf_of;P0af^ zF#S&*7gVgl_HNp>{!Y3a6@=4n91zk!Q^{3xRyPn6ms#5H?M^_k#Db|R6^5FvswbYt zrqWUvYxG0ot`{BJU>!7rUqMn!@@-se-4YyMN%Wx!tVvKPjDKq(w-3p@7iM4nx9%K) z=3kD9Bw|Dj@_PvOw^upUtFDw?+h1TRj&dCG5@vB$B~;*<@P~+_i}C4f)8)Ix=M-kk z{u=M4{0hh0PTW)hjt=C-ib^O*J%2Hk0%##+_4zw7?!roYc$Ph?`ov&^SKCO98VpTROwOkB+ws>`s^uNQ>Vc|uk_Sa6AO29*Ri1>XPHSWnf0Z=?sGvg^TXX%Ry)-$#Y0<6=+y=T7s50&j(# zcmm3$p)2V^kGt7~R6G*B9Fn8i6N^veETp&!c2g(4kd4|AGv;>P0sGRwA>aklyd&a3 z@JwIQAA;SSUDkO|lV)YW`(g>T#lYjNO9A(&>S>Y7%5e5JGR^P126fTYffb?6JpW}K zS^t%n0%F3p8UzrIP ze0mg$%e&uj*(ygh6)5~XNWjtRYEQbsVqb*{1CBp6<8ZGv>;x00_}efzVvCI5Z^k0B zl!d$q_XAaPaK9dfj>Rcy$JYA;V?^Fd3c8djVJ>mddGiN~tPCa@OSoB%?As05(K!v{ zfj|G+eki6<`I5B~XTthFE}$k$n*WwGO#xnKa74PB7s~mg-m$=jQrzvg8vN#aljYVf zQ?_*g3o$5E2${6nX&l8au~VhH3wx6z7s8HH7&Jk>115cYSujliDR-dv4YHiONO|KK zVGI|IX;=vcaidTmKxuRoCHuub>n9yz)(vv{BBA!y(>e($PDKVs!zm)H7YhR$ibPqZhIftfHG|0;77`v2VF9prkF==)68zh6GJ(68{kLTvRAm5 zwC+s)WZRdq|G*tCw{HWAAY&8E5P#}wX|PX3edq^Y7kT$+Dh1Dz0(UW^NKNU+B(Mqo zJb$N+fL3gt1C&~g5v_2aviK)v@frMNHf2CeVPPTzUVO&t-`ol8Pt7bZ2u!2i0Pzzk zw4JYTs*Js0iOp10y)WXZvTHh7I#~Kg&^&yq$Wa$aBq~%ti79&OR{!XfCqaPu+Np+l z)`~DpviaO92}A}WR;ghXFM4w`$bxkE!zY!EK}BrHju;I39YAAm=$)hQ(hSl4dVb8! zfW1l#aweY@vY$;Rp!CiSJWi`LvPvtxE+r};OpjwKUTH`h#%%pZml!GEynQOjjjS@H zQXJ9St6Erk+A}vc@WYn&C5S*PT7fTIDs&`n`Y(%R32WKgeqc;ckw*H?V($1qc*HTA z{Bi`atOO>rYG!wUY|kW=@P`2}#}jsDF$YS-;ijQnsLL)Um_)?r+dBW!QK0U9_+pNH z9lFL!Fw)+fGc`W3-3BNm`$Z>b2|c-=UENl@x8Y%>IGc*(9#ooGIaPo2Ed!KT?4(++?K*_)g!e*DLz2icZjz@1FBO{r`bLkGj&GR-8>? znJ|r~e>>P^@Pr)|_wT!amZf5GGQT|0T z+dtzdh)9wCDzOwBTr5e#Yp{NxUQ??9(e4h3?4{}WQi}9V@A#Dv$>kJ^L<5SbaW{qM z!ew|fv&=RB*7B&|SULA~xv@(Sct6k+oRL&Q+adR9A0$@6`%Uv>wr^|6xi{7x{Olu4 z(q3CFIAh&JAHr)Dr!9Rh*b24Ftp0G&G@KfFB%X#|YPVZw<)f%5z4mnk81ohG{cW~a zuv^A{h`^5bc$*R{!4wFi)36=%u31K9uBr2gp)6)xr%+^@h#}OmwY?GN;A{fV2^WK- z*HBnW54e4%+eC5_`(|0LWS)U8r}a@W!4xr;zWmZqXULhw459-RCZODx(RTV6fh{`U z;61J8<%z+4u@Vk@o;_MbYkIQ20&jGuA^2D@Y5Y7O&!|ao5y#?GUB|S|#hdKw) z6eSKM<*YXxaF<1Band*2{EZ{8zYYO|F3iwb3j1{|de#W5Zi{ezyLgRaf~c0$fv{>2 zi<2%c`7X?|uqr$(PFUrN_rcM=*PjB{X=yFW)7@qJZJlvCM&NkgMXXDC2R%-L@x8r4A~Z#xEf*!8FTkmrD#fK=iB53G~CPq3Kvr9B8c8D({k=4l)>8JX;(g33vZ$4 zi2A#Yb(yLG89Vmo+p{+fnO!NBkQTz{;5s)4>S`{AW7HbH8{yO@oI&cuC{_;Hr@a7+ z-Eey$-5{2U3&lE&E5b}wl19|y zd*Millmt=Iyb>EMIbg;2F%goHf&nBuPEDtBx?2QLHqS2|qm*YuHVdG4)1n_%u_)<| z1+@E*XX(%b#=i}Q6zvE;PAucZ^KT_>y`-b7JyXZL4y$3uaoK+<@pnv!ms%WNiNAjC z*)2Tr6-tpDJ5@7-(kVEYXb?$VNNl|T!r}m8Ip1!IbI8 z6YR+b$foQHSqIfi%ja}KMnpj1F1BGXB@whay7xZ8lsR*5r9!r%)JWe_&se5MBFkq= zP|mH6J9lHQ9-SWsxmRay^j1#SmrCV*4fJO{#CfjpqwsxffT))!;AFP@31V)}9?zfW zjtU5a*fUltDkpdzIl+gwlnN0-V*xmS*}i&DK;Bu(I?7Cl-U8{-xTZuml0!3&`wVb^ z&ul9}9;8H217n`bp36}^vkFL+@$rjupsjzE%!B*+Zn8wr|6*P+xfjzsk}t}-bLS8&ND6+AHp!)p7| z>EHe^HEqpm5PFnt!=mBMX|xmM*H3^O?RlzW9RetK7SM8^ore7l1$2y$n%KrVqJnC9 z_F2j*7tMEkIjC@2?aS*t@sRPTAMGs@ItZ^r8`#TPFz?f?;?HJRF}GxQ9_e23!~f#sI|MI+vA;?iDBK{!2BhRgGTOvT!pF3paVgz?2&MuS`6q;l`(~HF zSRp>r#cf!(r6Sq*z+x5mb3yXY7M#&7xKun$wFk*^yT>7K6v@vES)XHKf zSeYvBMEs_;4hvk;i(%%KDYx7fF16)Fo|V|gq?Nc-ye7*q2d|NB5~bQ|xqrgE0iR1< zjgpIo8M90?hzkCAMMd8E)Tp|>S_9uJ(WRVUk)mxvBT_G4mEB(5aV@Lh@wCqUI12$c z82yM7^HBIMPiYGej)`xJo0M(ukjo5K;zOO45kx%)yXV8WkBCmbYxS?_tJfy(KY77N zN$-?Hq5l?2U%P^=qW_(?Oa6 zq1M;^s*@9D;ABL2-+@VCsMkH@_mU+1q}w6#p2=S0`O^1sG}Xp`HoFWz-))Z7cXi?t z8GwCr7`z)f>fNA12EyqdkgE98evc7fKJxclFhuY!xojV^(5`OS?@#7Gz#v|5%d*qy z2Y0o=*n}6yh`gGc8b882_qA>U6FR%vJHl_(g(3)lvfUiVKasRkPaOcdL(xR~C4qA? z;B@&l7+1nw8^t1W>^bfS&ThX7I!{7iUx8i?&%wM30m{jdCJdAP9Y~m9v%-@!og@R4 zgI>{Ke^aXxDu0}`I?0%{EC<&%?nvib_}RIqVH1lA@ni`-^@vY-#+qO{J6ONqsKr5U zp^n~{t6n_#==)3Px%887Z(HumfM%j0LWZ@^_mMA+^V6Z^0mrzTWNZazx+`h4a78;F z?VQfTo!eQn5v&WsaTi^)9*NlGF6%XeE3hX2kb(ic#?A@yV>Kvv35On3=VwJ zNN)Bj`(x<3_hZ6zO?lM-jB_usV))uaRdN(Za~m^6YhHH=O(0YJwhjp{;*3)l(32C= zd-1-fR_1d!?OP)D8`k0+9V&o({ zV-aoXFFEvNf3l>;wrCO_%ZE1&w@qaxd8TC$#qrF6T0sU@nLLWQ=Bib61Q>FL<1My zIAOw}ypfaPF}!C;q*)?g_*q;v*7_)qO0HTYqNR21^C_yteKX9Fm4(fmPFhkA01pY( zRN+C_=$M|`*9$pTMu1o3^*^gSW|soX5S}9 zTeSIw`~Jd+ZKhRBH8FIM)7pgUOn737SmZRcp8#MvbQk*g! zYYCPh=7SR(I+PEZrO(TO`eWepCLb1s$U|H;&uSkZig(Vurn*$yQv(1-mW#zj@`vH9lleC!^^0-^03=P%>1VKrAgoP=?^bS0u#v{#x4F$>V zCBAr|Ys$Z8Ll||ZB-E-};%v`@S;y;1<7vWvCEbi}g&NxsiJUXK#{*xVk}zvasrBeR z0koHuW{zi4ULT|XJ_0U~6@16`IE7EVI7LL$KNwrr&VsjMoGoRQ$9{!miC9a~ei3-8 z-_HN!Xv+=|TrGqF0@f|h=sZKu9fWI!ff1I)pyi7xh2H+t(ymz?x)5;%I_SoKi8YIv zxx03{4t0nX2MRZ98xqTg5B-l(tP6Es*ov4jh_M`A^!Uy3 zyH*E6%|;^Z;-n1jjM|EoZW*sZ16bNOmW_jn=hh5@K|jpQ>KuVs8YE??a+_`o4~^@^73-0FSq)ow%7 ziWR0?Q}=u(x6UA7>ZkR&@%OQ@_9B_r_ymev)%twM&P6-%^hGd5(+i=Mjb3!2aAB8S zI#zpsONn)VKE8RNi$+f%8CQh64xFfHyjArD+*QYx7xiRO8cQnhcnq!m)0K5~zUkz|uUksDEBxCI^Du;|kBNUn0vFs{v=&w*zBUP?0h)$MxXjO4-` z);<83M^9F;zf&tyyUH0?=-V1GLA6V^L}W9wFD3ns5CnEC!DRhq(#|c$n|B>)vot34 zoEh{gs8LeY^0QryM!&M;if*A&F+!F6;K%5>Y_C*xEm?#X{>`Vgrxb3)uR=#FI>bICVL3KL<^^O;<0`a( zMA<0tdSqo0ml4Th*s^WxS@H||TOm9nS+;6yxToQ5yd7@@#L$Q$zUX)Cctk!5d^(Sm zxh_0!#@2=J(Dv;Exf<=uRl%xW2yU9e!3Bn|)UI*USp1XnXr3!9In*X-fg0&s;6T_? z&DsNJ|GN~b+r;_(t0?aH_e$$r3Bd10b{J&Z`FvI#$#ra7!bY2877;RWDQD*rWrB$U z-=2|}f8(O|>9+x?O≠=@78T64a4OIq=^b>gohasHvCueergdy*n@ei*FrHFLJdz zfD7ZAvY)m^-(NycLWwOm|+ z2jDiB4qX9RBrg9!bNWXd+jCV;j}^OSfnRZ@#3CW6^%j;{4nii!OmOy+_@_~gsDb|@ zu|jv8+fz0Pbd~;0D01&#BHMXwGb2UO1t&9VCnem{xm4w6cU^fSiR3nOmGAS(XZHQm zxMBK3aUr_OGaBq(Z9~U?9IrF5icfH@N1*_PaE-22V9+dQBEW?dx_Aq129;9BjD29S zyX9y=r~@hbU(L9_ZxPm*JD|yd!gVF^s`UV22H#O!@ME+@NV4W&5e*#q;(>CV^sdx& zx{|U~s)^uLDy1c4rBp=6<5F1=zm4TeCefK2kJW@`=TmoVuAbQr*x6hjk}7&0Wu<&$ zYE_H_DHi0SSJ7kcON!D2o#o3tobid3ELF6!ZOb({U22fxYZ73u``vl7?Y3TcOrfQv|}L7 z(^Ayv%FR_1aZVqM0!wAS`otVfX%UQep9r#c zGn&1!1o*J-(HuD{z(ZQ<+fX2v=QB6h60l;lR{Q#&38Fe(e#I3Rv&XdwZpqrd>3&ab zLI(cqN|V|yzeiQz9^OEZNcM1i-^z`HnWU}Jj6$MF=JotH9J7)Qq~J&AaHPF4n3J|Y zpTfM9=@WG!T+JC$VG9E3w3r6L{gggyrRYti*L-6>Nh|(Vf3u)PYr{2I{>d00Zs7dK z%}&QlA1xx~lo?-V`$>!WDmNZk2v%&&PK)n$W}&RPmey_WtKF!oiz7<^lZQCARKLx| z-&=E|h|g7@AFG%MrScjZqkp>}&wo^e#qV*8a!-v*pL;vD{VD=z_~uCEtDXs3KNCE^ za_X9_9)$4daw;WPez}bh(n*Q-`E7sWxZ>RHd`f0C%e-u)-669+ZbUC}WYO>WFQL^*Kc;&6KO|w1^r(c8i-{s3o+Uh4 zu)DqwLbsOndgT$C=dp!LS@}i5wtiK_p5=7wa}ohLv{;8NAuo=;8wzr=ZBE-HV0PKt zvMhjW=A~iB3x7+DtMm{@*BMBaSJ6dt*`b2Yqp6qr-oi6X_QWei`E2mKZGz$(gb2%Y zI=>i)qWxQZ`8J>PH?aiJtcn=*Fg39g@KPRM`>;x$tPqfbruU$5d3+!dtj;}{} zj`lTjF_aE#JTXq_6bDKb#)&8)K@aZzF!@b^=S;hqKh)?&Hd<=eN0R3;AFUVIVm|s> zX_V;;!?*@sAi|j>7_#j}6^Vf2B}gGL;d@|l-n-dURzwf=$^MbqV8)cQhY{k;sMhRy zQ$I?|NQw#J&YD}K4cLiy;O|EXx7%{Fh6+KDlML6fYYOh*4@>X=oHMCx`z4nXe5c2x24S|Z0?zJ8 zXVHYUxx_1?KdT7a>I+5q+l7K%_4GE&OBVi(df~s3E4TXCQW@(iAq2QD3g>xST0adB z7Xq03Pm!50aroI_=BT2ZFM2z?uBX%i!jhsoxYI<(E>*n_Dd$JE5lMN`tBF2mGBlwloo@UWx@SQ4ds3fk`_|-Ke>1CWE*4?~T z_Slt`Tu*#+22nfED?)@OhuoZiqIOJ@o*m`pJ#wOX9?v58)=vWhKKX*iDr%adXq)b9gpdV z6-hAT53{!YV`PxP2c8Bl<~&@t+m@&~8fl8~WUZO*a7JV0UlJ|;=*%e&z=iZvd5OA= zp^?zMv1#>^J_;(~D1QKNwIayzUoM2AmDWl7Z#*-eJWBN)U0I3JlpM9M?olO-61{|r zX*+A_uBxSlR<&JUqvDv20g20JH@(ZvX zRgJyQHBFiS!>Szn+d0RO~n^nM#i7dE-#as=(OZvTOEbGS^vzbice~ zqm6r)1*@Tw7)Pl0D!kPpl6M9IZ8+vtZeozIUA8i9I5l(ofg||i-)uLTmk#$*{8urR zWSRZ&CJnrgsj*a~_WpSc%#QI+Bw46-Tu-f9al3yE3{Z-$gq>sXr>j~v@j0`Y`Uc_& z2vfdbDCdu3wqbr4QwmIdwt@Jw_sclVCsEi~Z!22KG*ktUmz_@F8eC)hKzb~N0Z6`d z7!}&YAIw#8X2S@FymQRk$Dx5X@I-e)UD`SxvoWQ(d%wWUDxZ)^5HoCoqwrmO65sMI z4x;ZVew=F=d`DA0*->=f{|s;>X}x3Ez>C_Mrk-kg+AxMl z!;rx(H=h=b8z)9RxWE<-aJ#dPgH=IkR z9{`B>OpB=HW&AgI>7UopVk$`RsPlitl9TQrLHF&*9lPrVPnO`WEssS-a5a7}?kKtqQYa z;{oMfqIG^TE8%low0r!Prvm3;T!vC8lf5}fhX$)`T{>skTjIS+V&NUeK4|H98oW-} z7lXL38>gl}VLR_(g@k?_sUa0XDxU(MDo8bO7;y0ov%wFZXk-Ex=<8tDTrhv%Fy@x^ zFdEGgJCWEi*%Yp!vblc?N=yl=O~1CY?EkW=9mvXv)eYVVIr$w>VH}kdm%-6 zeLHI6q1@;jP4YW(#G-FK8ZQ1x&`y;uF(6yWX_BxzhVflCpTP~iE6SL7u_C|9U|2o# zJz9G|zhCx*m`_#2G@!}p74QPoeLk>ss}|?fk}?6S0B8oCXf2$XnTDmLC0yQr{kSpb z#ea(fg?vi2Ag2`>kr*16nKw8_+Mt#DAkA0P2N{b%e6rU1J=DWn-|}F}k3W@qPj_3@ ztctSD$}GPSuGL7WO)59=pL4fK1brc@(a<%np-hv9;;=!-b6c0Xflp;wW+)p+)Vib6 zBKmiIVaUSYzPg;eII=#rC68Kz;t7V=g`~Twx6)=5qA*I}4!W18-^zh8YFqX?iR@>sgK(S}-+!NrL&57v73v=u3MZ&4N)HK@8-|73 zH}-Kj6)H^*HVV?|SiJ~eyGDkeM$|oD0Cc#HmImZB&4+8Q4D4Yuq#K0ixq_iTlg{;b zo;_;dKFq?vGVeer+Kt_qKP3Efm=#`GaoHHlx3f-|z}sd+eW(cNGUJkDLMC|^c2-dV z>+JR{ngpFqcQ2mwK;$Or#_r=&!(ZFEmAMc|fEnB%x;lt#Ev>_QF>4J_GV2rV}>gMBUx1a_{yN$IdAHjVT8(SiPUR~S(K!Pa;EZskB6URc_$ zUA+6<&+^HVzX5+dd`TdtF$@o`=inHuGNdH~8bA`_*+W+Y6G;KY3hAD6f-U1ut^~Bd zLl4y1k|kXV*|toA>vpDm3XI+8pl);107)~=uT9Yb{eSgXNuOYipaT+kj=d@Sd%?@r z0i3=_iq{cD!o2#H$NVKQ2POruH(d=NBFpve`k`SBeZ#qBwCH>-T!eom%~+AS-24pw z-XrjLyHFm*9`f#KQD{V$PaHw{eTn&eJlsG6dkkx&)bKWBctCKUO}S+mCUfqf&#~uL z8htR{7Y`LI^VduPwCF5AvRb;f&|Ztf$;#QEmd%x)qFN|REgTgu z!EP1%fqFz#3evzkX-A65Cc3F3C4RO$A#vBmM+I*^;h|b)^^*)3CnG-^4z2V`K76H7 z;y?09Qm*$zp3NI1H)P=H`d}t~j@a&;h}JefdmN`+0xx5IKgfTv~{=cbhxD?F<#0(Ctv*9teZKgBOD4hZ&}&{w5CFmyjs z5{NxlE5fl>0K{0A%0jW16$-H}RfJXE9vMVu4+5HRAzWc$#nwq~b!m`M*4sm6HCP zG|sXhNT1DjgRIf?$8E|GR23ZNFH9#}S17m;H||k5Vx|UvpmiwmB<$`@&+am6_?)rj zxq$~qFZU-#F+VkTHJavZ3a8Mw1Y;Un ztND~|OY(U4`f#PboJBAzc1(vk4@`s9l^*nconUJz)h)vKL*r{MVwa(#*inCu6_rS* z_+oibc)|SK-j#=HECQ7g-f^Z}42&V5iF^{%;a8LHIioraE!9>uqaWkabN28{s=>_Q zBwnfyo9N3vq@vcnfyfnxD~_?$u3?r8`M45qh2$(k{L&wZ5~qXNGX1RXQ}>M_a4TuT zsqP1m#K495onYO$H?@Qw1au)%SSmv*u20-g_d_zf_z|*x5BCP>;6!|iJkaRv`abD= z29y6NuNpOY3Y(N%1$0Qb&ndJFR;bUTrnQu3t!0&Kn0n%X4H<{wq}q-|WIhs|&7f*p>QqIJjc)Xfgyxi70!VobuN#HScwG z%bE0TITWxSxhQT{Cn&tj)b7u+!SG?L56>}MnFP5@p`f%auG{b=X<|L<1f)@wEp7m_ zQ!uN|MGF?%#IjZFGFSZMfDSM5QF<0Yh2`AlFXuHm=GR$`U+&{b62O7Lnr$#s4{VI69-u$6k}w2VHir7@4UkEGtx)!RsAge9-sH14mLUacBN_gUA60=d{~FaL8sBg*0l3hM0^=D{!o+^`fgTqmA>-D%-!hEwc;*_ zV0ncC*d_=;*|pOxeM&0ZvQ`#Z<#-+m3a(>Qfo_Pk0St2E$grIgLqf}b2DjaZS#2%8 zw^%Vn2`md!Wzmr~>yEcAu;4QFbcBE)I4C?L5UB3-xlV~ONdOjVm&6~Vg)@Cm)^JeB zZSv%~YXL~SEObbquNpd}nQ_k9?pKezS>9Klo6mg$KFq3&`KcW%A3v{^pXy2K9p#^d zk~{KRGB9ECyOw^7HPdWVV7bt#zDG*(EtDxk1Kge3dh(3Jlq4c)TtQ;OT7T8P&(8PW zRcH&I#>oN6as^10RYHm5AHv0f4JWSU$^s&k5qfP}C1$Cxz)JT_{>os|+2lz!rRPur zG854xPGQ_OzJYv=jm@esQKp7JO5~{f=Uy{Mwa6Dj&V3A;?(=@B(C18R_iU8!3*_#) z)@lQLE9uJ<3rKn1qQVHPW6uhSK@{7|2*nd`%9d~<+k=3YK+LyVpGm_I(Hun?IWE%1 z1fc>%`+s6S$G(yC9sLNj2z2y~p7!`dS?Alqzp3vva4Dy(xKxLm(g5Sps%urOX8eWL zL&o2`cymINC3s4$zOwwb8mj(BYb|OOi2L#>)&IXEZn!qZDp&enY3e3 z&AKCq6kh#Pjf`v|I0Y7qfkSS~3PPv*nuw=EFZDHK4fc$h!W8bvUQXD*7kvY%cKx}< zGIeHZC_!p3+odn=jh|PcKTVL|rR#xx-AXn7B+ut6zaybGE^u&HH~jeOdYoMx9!$qS zhvU3AGMbPLp)xaBly)XYkkj*EK}NU?5Tj5L%v18`g;cS{T{cwf5`OA*lmxZVu1~Qd z3fKP@tahu$LPJoS9r~bTcRZLEhVLq|sXt3Q6~2=A(yhJbgr5XWkdJz65r@TFjd8;2 zlP9@LGF9-7g$_eM>J*e=*pJDZQ-W&Kf(y(eZuBR(@x>I2dex4*@6aK2m=%SOZB(ho zZXj|9QSdP9(#CsI=y2Z5^~LiMs1r3aqp-Y3&O>=EI6{Ys#x81jd8F9{G28KsTU!G= zd{5&>?Z5nK-+Euk!&iZin*Vsv3HsD-i*!@C?GL z-~Rz}&(a+p#D*a5SPacTVFi6+li%3vIhJ|bj@n_d8A^Oqe`G+O^ zIo&vx>H{T2Ftz{)$!UVm?su3wU`_%BLWvICC9~=$lsO{GdXic*9a@AAa;Q@^MHmui1;c5S zU&l7ktf%s(96lXkmekad&ug=&J8FZgpCSfa@tlcUoZ)>s>NX=AeC-%ytfbSm3)6@@ zij~Q~eLxSJ)+ocFp;s`0t&u4ak_W1kg^g$gy+SST5mqbU=rrWVjZ0AhXXmnV=o!x8 zU|S$ltBJO_eU6o1hw9ceO6z%enSVM*$YIA;rIFuHF=;$it@l!M!tt}YLFP~xsz(CL z|H*9C>73NyE)=EwtWHDnIwV#!Ly1jkfvoR%-a@IWZ6n?ES1pOx)2N#?3Ck2!7oPvZ z2S8rBo`t|fwH40Yx9VZe6JrFZbI7)?I+98kPkJEK?(6gQI{A46_wBPAX(-;PNr1tLQVI;~jg!XV7P23r@VK?jVuR$* zWh|6WQEDK;F5>|%Ym35DZHtNri(wu^TwgW;3T`1vgZjIl6Gj{5JrJ0^*;)w}Pn6=Iq8UaG*&e*NX~_#elXODwFuf-#vLfuH49+e}a)GOTt9 zMlE-PO*c=i>b?OQpVB2j)ONwwS~?M`g7KMQ2#SLR(`X)!bmc{;Ii(AP>UyZ%#8ak7 zDB8i-2ImM_DpQj|!<mty*j45zOX{hb#U!4lhr%DCnr|FVC<)>1ibb(}q z$^u;!55-65M4vBLuTsIL?XiuzR;Hi|G8Q!$1$+AA zfG%>04mPH^xbX*+0kUgx#CAl~z}mM>Ivo=Bdp9q{Qtqjg9l)u=`@q~tW*H>b;ZnH|xR{|mDEBXiAQ_BzZBhbl~BZLhny2`i5o_^`kK-L$*C!R+Zim3L5LW#Wac?_K**M^z?~omAC# zZjGTTy;B)6X)rP>6y(qigsNZy)inJ#V>I!qgQ~0Q$S9=mVO74ClvH^%$^XxuJ`k=~ zg;Y{g8CLEbTki0^pC5Kr8v%nl^H$B0rp%uv7;d8`lI;I9{ki|slApJSR#|+nUp*_v zMM2Ja!DwQ%)pr72)%5qD2&(*`asF#}b=F0|@1+(=^17!IFBl5b%ZFMf&BxR-$hL}+ zaB--`Xi{3;sW(U;oU};+4nIaapuHN?LRGfJ5zAY_O3CpFd)@nHkW}E`Rp#IE+ADw# zrH4S-ZmZcuE7Qi1BzEKEQg-XCbpQ6jUJvwXrYHH{H~h_bTwm>{?CGQBNSUjjp&T#X zg#&K)AVDd-P5M67xkasEWD^yf85XX#1hX%jM8e%L(ah%t7j)h5E43(nijtf<^SM40 z@0x(07ODewhg64mHP9 zVsWW&VL|Y{oiF&+YJy@1f?HL0#8X#a9V@o?t+BSApYrapuZf!}WId24LrGUl0Q_*p zXoQsxg$>?XVUfp{M(I;p;x_&lwr;S}wT+L?MXqbA>5(Gjko0;wKO>2hAZv{ga2}z)S z*P;#eUy!T!zUj7lHKT(l*E_xqniK1LDbm)1uZAJj+ZXH`g{q;0n* z{^bWKvK{lC!#GF_$p_AD*Et z;OrB--lq=H3AwwpF35-^Ud|8oe!E=fOqrRZZk}4ndW&)zQg>l$ca-GC8U^tWxZt;~ zrb+DxG)^*Ea4H^HHr(Z&6`ZN09CM*!q%O)|h2-Y=CO-Wc-BK9>%V;`+6v@nrl9FNZ zSp=Jzfg(>kAajEv6w00USL<`j@f7W~K%;K73YZkR4JO{;IU-r- zl&c(MOSJ&s-cU-gK8TnryoGsl<{5NY?+sh+H_ZSD1)+z{XM$kugNekcfUUDRgV z`m2ZM*RH388NO1C{d(JoN8n4u^b6 zWWjw4`6RmqV%PL_W*&k&tM%YbKt*2igG>htJ1}W5V4hKgRJ#H1=QK$HIg@N^i1!F_ zOoMq8TYEcdy0E=%S4QBN=Vov;)N5Hs>r=6V5={@6a^N`6HjNt4n z1jrA}CM`0$NH|`)X#$|+1GyR&X?1lID<=+%bPMra|=x1rNy=#CWLml3m?H z5qcJ93jyF>tmoj@2YM46&BAX{GYJWDoUYT#^v(f5)| zf)E9Mv&y4d4tDp&TUOa!Qo#|qCLEdwxUGydD&PuTenFAF^bq zzZ%uS;adp9lWPgdkUbGOUOa|+VsNUw8yq{#h`>LWb^Yt^O(zIz@Kg*AgPLlYmTS<_eQITw4YD(b zHhC6BCyBkjml@)f7zD5UAfs9owSit~C>bZqm4W&(!)UL9vhptXy(=;*CifULK8%y) z#sFK8e)1&x$pccAvwMnLMrk9uK+P~dN&8J}I^|vDXz#CWX?XDmYBc^GCQ#hGirP4U z)YhC{O;N+I9)sfrtP!f4unN%4)L6u>-fLx~J!DRRcfiTzG^&u#I19{xaxR6K7)Mb*e%)#wOx-Lu!g&XI zPw-G72iXvdiL9R+z}o=z$^NnUAMO`|0$z}~A4kN_EGQmpr#p@w+C@pw?SSdh`8-ts(ugUM$d;A${naW-^ zo8Pe}6n{PYJDl{l4n!ZtJ7L`8@`` z({JG<^>!9bFtfGoYD1_@pDpLF3n0H}yoXrNS5TER7vRr>P(p}8EYj~yZp?$zp-cg$hNFjJ+rDIwY=GeD4D zrZTCJN zRzcn+Jg-979JofQfgiu+jDeY6>e=s)s)X_i{2Lkj)JsQ_9xb4~R~rD0M6$MFXXHdl z5-iqMKIrGi=1IM65uVCjMpIxO)pIBH8-M-Q7-Bc%f3Dv5q)<8-GP`Ri%?k>5W90S~L1Wn(Li| zenXKx9qCBY_`x9kevdv(`*b9A<45N0Eu?Xx49ev;O#NjR4aBEd^pMv zzctO_-aj*tCdZk&T@?U{(~u5!$qh;~{r1XEI(MCwohTo+`-yoj=##WTWSz;V*VgyC z4}QkVx6Y{lvcK=~Ehy9z?n{1W^!)9pZ@be70e9Q0iD}=k8L}Xizp}Z_=TH(U1d#x; zHjl2)bg9mRkfw%EXmUz!2gh9*OB!nlu=0JfaJF3@;IpYpl3@_gZB%Puz(P!+ReN=c ze|u}KY#FhK4z<}JYaPN=nZzsL$L@-FsF`dslk1-}-W&-NRmk)Qf0Fxb3x6`Dz^YUP z&gXI=xG)NXP86#Nfl=i;W!OH!3IH_J{V7{oT36YaJK)*B^UGL8#wdY5yNoz7M&dCb+8IL6r3lQ{iAcP zgQtjr*Z9Zx?*x+gYwrQd?NAckp--kN`LC6POgMu5xOAQ{yT@vT8(i$w=)W*`8y>OX&>-KogtU@ zeqNizhxWY{PT1W@CXvhWW)ke4$^9_^>e+?^Rii)R39eF)V{kmB#UkiHA;kX zXRucAl=qUrc~UQj7aElO0St=L*p|v@@{DesUX%O%EPn@#ln7bh*Gu!Sm*!tD&Hr~_ zntyEv`q~ckwH@edJJ8pA6aNc)69w4Z6%~H|CY;ZX#=H{TCRz;-w~G=yS%kr8eu_2% z&emhYVaF-G8iH0~q}bTs90iwJ@hV8pvM9jW5D`3!p=Xs-@MayPIL^Ny+s+#Ds@ zNi9qSoU6q&hQVJ+!LPbK8{yV*miK6kCg6)`2aFkoxz4>nqyOLGn`!$$-}=_KgO&%! z03!>x?phG4qh#Eldub(@-5-kdBu(81@m1?OuE&ex5w{{P$vYi+1aVpesL zL1k}!AlywhMSvHeXw`v{$1x=7^CVwR3!Mk71!htbzJLL6ODS5WW>e}_pOf?l4EGa5^xj*V{#y6KJ-}dw-AG;ga1NV4~zPLmGJ9 zCueLrU#}S*(sH2r9H(>}uieTl{p#My?L*@6Fpv4q} zT=+Jfsy7t5=HYL4YHWTb=z5_LJYctl8ERg*Y3j-25Gvp6rsSMij=1`s4)4Q6o+t0z zB@O%J+3&U8e*QVhBbm>r<^E#QbS zm^2Ua{=k}b_d+pnCG^}ozx|bFGoj~Q9PsNRaZ0b4N&8;RTjBg&g#!i#qWU~Emvnd> zgj2eDwS|LWr*WEhypPYd5ePhNTZDHcyroHk<_?tsipW;ef-IW%1O;JO9jRmC?n0aL z-?8HByrwzZR7T%Bw5%^NW|zk+cq4u;Zx5|BC24U~ z@0%{MHTe+U$`TLl++=-ZFvB|P`y0zuKO6VI@TeDj9`$d0!o=|^cS@gT@YW0==H=c6 zb>JxXNld$4NzXDz?!7pRK&R9LSx08D646rBZ@IO1+1Wk~? z+Urw{MvGP%A|P+AUix$YeG*xRHo)_p>Lkxn(1WH@YXyEXYK7P|$zFI;#zB`7z4Vum zXn`XKaol|xjLcAcQ`OE|E0TufUi&|*@9EoiLA~KG_jc8yVa`}pBsdx%A1CUBR47?S zx6;3SQ8~?$Su}Cn1Dg68xMDiPm!K+V6HCF4IJ!wASGb&xF|6GS{ITWdvKZgn$ZCZ6 zcKubs{$dgJacw4CYQgwRbe7EJlc38$91Gc}%p3%6u$pu}L%N_mZ?C*vWbrSz3nj z7?c~O)3kx)?W_KU!lGid&aq;z?DTw zxB=6Bk8KFCt2wIb?+|MQ$3>%+$K#|^u7MQOWee$0!^;ty^1oNhBN}J}`%T#$D@O_q+s#8rzcG)YL)e zc79U(Wfe^8IgCw(aN@3Wi#)V;q9w{vPbq5)r(g|ABWf%~k286go55jdH6a zkew*^m(;ZRLR-+`-BwhhuN?ZY1dqoYiO6mQgPCT`S zT*|8+@&&$%?hCsnHN>NBAIV;@i>|6mrPMRuVzs}%(zo>PxjVqjrMn(pKTfw<+)U?! zls-yx<4FP)+-NLydEZRN<)f;ji!wh7&KXbnt6&^nvhQ~a8$x-<`j)#F_nhRW?_3mA zV(yD)6m=UDMka5CB{Fr8Z`L&4tcifxXP5|>v)W*UfR@Sxo)Ys?Z~XXe)>r<8@a8hI z_y4@vx2>k&nj#9Oe4-72DUFciE1S~__~QZ2(eh>pQf4t#lUI*@)4d)=?Svys z=s?Kci^P^5itzrlHY;iLX7U(MC;y1DaPpUW zyC?f?8_Y)10o@?X9;+J)5`5B^xXInagWNNiL!fK8R)8T>#URoQRDhEHtOToCTB0E@ z*uuihKA39}{-Tx`mh!NW?D?%f5t(2gjrm~UWh(}gzYf&RLoV+6Fz|6Arv2NO_ATgK)w2|e z+t>oZz7Y>_uqQI*aTR!lr~wlM^6d!zv%8~UkIKU=x%g*GB4oRnKqyWZ<)D^OwGUlx3WAImY;6eXgDPBC zC`)|ZCiZok*ngWt%4L;7n-e+}ucA^pD@qz8faeH<Uiw3A8A^oV?SooPK5Xwt&mah~P;ZPE$_Em$# zK-p%TxOdtLYRFrFbM8+&D-5>tnK*c~R}&77p9~^wy<;-LsB!uz*gX@UvdgAddV0i5 zkXOpuJ+tC9;H<_&mgHd!${WF_lKPJ{QuQH+BL<-ksrpdBZl_8Cw@UC_yAn(q&!mh% zXnxof18bL}qzG+cY7XVGhA0qEYt?XP3}R?OZ$&_%oz=mkMtGHu!9pZ4p-O}4tu+#i z&5{K`#cW;X|E_(MayY)mY=rZ@X*~ollx&b$WK*km6b|XWg{JmK!j#7P0Qlm8k%OUQ zB@GN{%!k3^kL4OqwMo#ttjs=%0si-ObS!M_Mq8L^E!>_TMn%ewegR^R(gL*)s1jgk z8484E5f&3TXW}KlaG*q5nUBe$%>q+PxYb(V)YQlNaAk3q<6$tfL}P~w@8Rzq%~i3I zu9cfG$!AG94ZJtrTE@uQeA7yTU2hugFl&U3_f(o$iHUL%F#1G_4lX})ug?#3L{^~| zw4K#KaAdq>HcT-hhu65MUk)Jj_$ZqVj=)u)TqqOY( z$OraRA7ZRFnv`F+n8V0EgVOKR4itC@52PH#+$z%db-Qmkg5{%n{$JlxgQ;M$&wa9ER!PwU!4i`lbsga8>ysPm64$7x?V z0+X|Cl3;`EjvBkQ0RT}p`=L2EsZ+a6H3mP>P-&b<;dfCFk-9XMmNjM{rjgPUAenxJD z#jOzhMwRRAWQUhP902{BLo1y)jSiF{{}zu%PHMAFJ8gypeyMA@*eXa@wO|``oT*d629i5rs9ry4Wy^3teqn_NmTDCMd({GF9 zZpH9ki~l34;4|2c3CFbsqmNm z2QA2vIa}S4PsH9Yy?r|KDIw+Lx18fsT^+umsZ<}}FMn%!r%mJQ_%GZ!NfiJa8YzuD zYt6T&;AF7Yt0BU)G*+~LY(B;`?<}pLcTralGur&SL3rL1%d>=hxc5l(dB*;%v!z6= zN^^w!%SXPT$;!GM{ghhs6ig(Ue|}6Zxf7;wW>Qa*4K7wGR8a>&a74X4&bkVuDhIqO7C`!@3rS;U>` zH0s4zQ7CjLjvd|X{)#p+AwQyS- zK-~a3473mQW?-F&DHrRQc&KP?V<9a=2#0JNLdBtMeK^=}pu%3;6{%i*2rvCo70<)d zlct9M?yZzlo}@p(S*ky^|4egdx`raF;q6c#kw%izpJyt;8v_vORKD{ zHZG&67M{2_q4Ae;C+FTV_Q7R^cgxyb%9A-K33fKs3938iw~n~AzpWegd?^pp?#gjL zuB!Z=sj+75g0ZpkXc!jCX@!L`-pg-dI2im3@_X|FNbiehnNh%6#5k!Ov-L7c-sYP) z%G+@j?1@8#;##V;LDvsaly1%!1o{cMb4_nR*j<$bW$ol28_}fl9fQr2aWE0`j?+rWYme9Dyoi-PwrftzpCeUi6Id39m+)TMjDzyQ72O!a3pF1p=c9ph0Fyq1=PcP z19wlTSy=wNKJm|bYbWG(LSdOXQ8Gbgyj%&pFB*7ouhr^d!>1xD*l$rIJ|>Pa0sC1I zO5aBSb^lFeg4{4V)g#QP2jb7x+OS=(5Mj|>%*TuSlWvb~tO0N*LI{ClAvpVdwxVdA}CqvB{tq7GVY7^Xjt~Nvca{P~?tr|Vu1SUQ=T4DHQT?kaJ;#IGjJ18Nd ztF109{u1L${F_D-OrUKhQ1-G}phkg#)80knfc8d1G*p~7aI_bSc59o#2@78Fnkk69 zBlB>OKkTA$FXOR@^XsG^}lJFoR|R8v~i=Fy!3os!506<+4z?G!O?v zejl3z1M8t)=-Ce+SFsZTX4h7&#MoQAL9Y@c;OSEwTUWYi8qEt<2gBM3#Qc&U1&h21 zafm87sYSxuthd3z!&*BW9VDSnTd%f(y9kA;JwjndxTB3QvN_pZ|FxBeb~_~{JlT(u zHlc&OHmvQ9(v;VloR(XK-X7aW-SMwa?ed#4R#)yoWsLE^I0o-Zr%!FE(DPSC3}jnW z92B)z`dWGTtRv1RNf~e<3sF#coi~BCp9blwnJktV2n=tGIihv3%m$02P@kM=hemNi*U?Rlx%ca-9WacdrjXTQcYd$$ zKhj6@lsg6LEs@0;UcS!w<#cI7FIBOdm5b1~y)0hMT1$$ayQ!+5q46B0$waw)dczRc zBdvHz(~sz|Rtr^dWU>&QNT*?}@ord}MY+fcikBV#3PY&Re26Un&TlpErh8?V4ZO;` zX;5c;KAI?qsmlbKi9rk=wX8T(u(sWvdV^Z_)>lfeXZVnaq=~<~ua$q*RB)9x7)71& z9d!o@v(syYKM&ibEW_*enK`;(j+e1PRyJm{VP%3Y)|!VQH}p-VqiZeFSb!ktk5`GO zqmLEIm%aUqEzg@(3=6+eC#q=Ws&~(y9zwyMRie@SatXWFURs&Zmy`e>n8V78Vagm? zK(oFV&poF!2Wr%!PkQGypK`6 zGS=)%&q(jkm6$FI)@e)-)>auNz5KQERmix<77_%#F~SACI`I~m{ZKA6pkxJ!fqB%q zjRXcsZ_lmoG1M~O{!!xUaqIE0KNsGq%Xrn!p?KBq19MsVGFbIc5sA4)O&FZ87PbUg{GA(Pa87`zYw;H5sk`heLf{`(3^S6Asx&sI30| zL^MpkCPmfn#eSynzlk`rURQ0t+npLg9unm<%=zurqj#d5L#QVFTTKje=F+5LZKSX! z9Y1?+u7?oN*mZfPR$|5 zjq?{x`uv4F9s0+^?JT|8XPLuN%!+PgUqIs`h0*_hE#w2Uqa5q| zLNr48G=mm$Kf#=z$Y4k-zmFugRU@+le*gpNPzix?b}f#DufNqaf`HGo!H`e3CE7cw zc&IMN|G1$vMZ@xyG7D7IH6%d91G&V+&ZU4mh4}kookaro3cVbfd@Kux8~X(A0B>c zIOZ^+2*c?gn`nu-i7@VMq>O@X78&Ndb(|e`){$$V{v-^q^|~m-z_k|h`K}L1F?|X? z*Jba?O^{*WIh`YprruuJt@P76w>ruotB)#*6y{6tXf!4xXIEt~xT=(D*cN~(Ik|yK z0ViZy6D&Gu(!ipwsBZ2xN3HGJg;F6i+1L`wzeWOB#b|3o$|7rsOe2R-p4ci*kupL1 zP)Y??zL|%GABtUJ(jZyg!XY{Zq6@9tb1PuZDI6=w zZBQq!#&LM~mSE+5jdsqI$o9)(2K_b|Vv=lO$apHA4ds>%K3mHGQ9!Q04Fxf%>ashr z9Jh0;`GgngZ-b&S z7Cg*&nE?t$@K%@tsNr(5O`&N~WKsGWW>Me9{y*+UQfVdsyeK^`|1}W*`++zXs*hmE zsvK@%UjZ?O;758ZB+g=A0kH-yMj$2$1Sqm>96Yv34pQ@1cgu7AbH8aR-SQrXGk-!fZcY(-AC|8;w} z0%E(`Y>==^!NW79Ne^NjivoHtFr~v_UJM4^DRT%EoMIE8XB$BeQ?8=MogIf~`+l{~ z0XQ>)iCb<6{AR_#=FHaxuw-et!>ShW{5lSvZ3ZqDdY(oW&iGWXfjpf|An{XA+%k=D zWDgFN3!6=55t90$ck%ErWXNe*4Q2sZfqJ-GuyR%+~^?T`7lQ88)_?E)JIArtWi z2xIW3sH>Fe#3(l!k{2*~PgO=L+-ac`JxLd%r3-(dW?%WiB@XTlpryi5TnOAUVCGr7Lt%qtnX(O>`;j$*(3GnOH{4P) zc)ul^0t?sCJmikyF%8XEXjJw**MJj@>}wbvf)Q-cWFZ1xEJNWtZwcUOVPnboV~IK{0O&RC$1xIbXaqq zcr+gRjPA5vOI5;>4~$mx|M;DS`9Ex1A8amc1FE`;OtF|mtDX4Y!S@_oqHr|bCyVym$>q%z~h9}xUPVbx#$o;JR8X`C6r zjC^kZK% zF-9LC@0)HDYVw~w8BQJa#4>Pcl-Z&=#dO&+)J}~ad=~U_1Io!nqe6#tJbvU2TKV>t z42KKi(#&=g9w)bCI$V<>xkw0R+C{iPG{_Tu-Wi3m@#30HodcMv}p1?;L8g~C=!4GcEnR13PnYhdO*fraUt)CrK0Bw}3aY_LP7 zN9&NCR6}{0F7Q<^NjWq)i9&zP6a=T5_B4L=u2#%o_Z51`c^jig(Jzv*a0$v>n;FAz z_ELiZ;x6GNzf)TyKE0>maYjP!tqeK8YA!QWttFho|Hl$lf076QftCt)_aJa)dSQV(+@dj zg`fz4<&W^zm-WLWV7||Wz-cQA`iu7H1^(DjFDH&**^UAdJOTyRAwX<3QOk z^uA<(yzTfcD-Y79f6m4HO-Yh%=mYJqYn^0ce<~C&!vs0gj6hn_#-IUr0jDJW=OyrR zLMoIMaH-(lhsUVOW0IlAEfhHe8fao6N2ae01((FvyAA)Z-);DsNdJpOn&&Xb!ptGO zp1^42yjijqkljH0cjiok0n<>iYVTS(Sgf+@;L=y>0GM{coJvSD0vbIx1`@suo7WxZ ze02dT5rj%3w9zS(2ukuG&c)|scwzJbB+fDw#;xLZ#^!__El@oQ)i+`za`1KxkTErQ zIAli`*uKH65LAirx%&$~&{KX^0TU9<_QK{y6O@h9y$P8=VN`XV(8^(V4x(g<(H01S zca##$9*t7GI73^Py!B6qmHlV~v_UfTbbcW~!B)KO)k;Qy$~4jf?|{fdeu>uciW3Fc z|1}0~hXIEX!lDsix-0}7O>ofWk5DKHHimg}16X)r!Y?AJm90JH8E^g9FfkD>Hni$J z1s}>mT||dK_Ax}I(Q`4_)Kn=nWQ8>bYIQ(#eK1%N0nb(_Gsr579K>~WA`uCC=6OrE zr&m-6{9c8!A(_a5(l3=8JZUt?EK za*e6F#7whfLXv2Vg~au=+99wl7Pc7fa!w`4iVa3~O$gZUg zhpZiXMFb^lu_$g7@%3M;JacGNynC$=2RA32rTd>)(KfJn_4&=vJ`e9<2~iT>w)m{_ zT8Roft?Ym4$hHxPlMH3`z13az7Hjox-*oFIx;MemzDOG%xKO%Tz=4=d>AwL!1a=ZM`~v`$PB1n&o)l%BQ2;XV&*&micb%_A$4D zD+B#S12H`tQTKm2K*Jw_<{%h$h+~tF`}olPYzO$FyCoP3&gzxX3tRf|b?k2s>r?&9i|LQebu?5gfz#AcPe;1mUfzLu zd((Ice$K$*QZ!Y@*K4wlVSa8oADK6I$-Ua zvQ*eYSfRpfvqE+V(oXeP$O<1#k<79a3=~Y}g9>%pgu)wIJE)H0QTgsXD!6YLJ-EUQ zYL9-94F*MF@IbWWm>tMhrA2fyxYx@x%;bqOO!|Bp)V^5rC?YYa^4!JTnDae zs89p{CTj|OKUEAE-#~3tK+8hm`BDTs*LnB>ObJhdW z&T}jNc(Il?40>*1aOi%{HUhVb3q!?@3FOE&O&Exca7V+c@U0?@G`q(sKF?GigiUzQS{HL_ zsQwPgy3KE590Ya4sF|ea;LCAEm32t_nc5&) zE#c%LnPJx+x)#hfv30;XlSOi0hPkqBYh&d9)Yvh2Y(6Sehrg2Ph?@F1Y0~zQ^y#%6 zRXnNTn7fca2&azMq+p?BG_N2|PI}dnCbSURMDB!p$8=#Z#iKGnr)Q|#_YTsFP?65q zIeN^!U-xnDVY$00Xgo$~hwMKvjFB{Bs51xJQ|{T0rqC;jO)$=l>ljyz;FL|P=C|rd5h)Jdtp+^%yE+X_Ybpi8 zp-$RZnEe)V^5|}O_sdB;-BI&YDIknf>cx(lyn`6qNbi&4t|y`>Shref_pCf);E_iu z@9nPi0D9HC=SM^6awZt=#Lxz|P%-$|BQ`JeYOhM=cJ=U%xGtmGEbUq9*)um7`s`8H zgWJW(f5V>3S;!sEo1h{dv7_$02+n1{X$?@`NYVL!P>j1%a3zcr#re%lCdR~0CbsQ~ zZF^$dyc6BnH};Kf+qP}v#>W0@cb|4^tG24^_uQwdtNZlBVZ^5si|S&|0s9UWqBk#s zcaMAV@~u$Ge2|=~Nv=2sI)N?Azn<{+3`938-ynMEb2TGcc`>#(&xa+ZvwkDtO4z6nxJ^qP0z|Ib}hq`XF25<3I=Y<6B z6JmsKG@7t3OCH_-_dg@ftjD;6N{8wZAsP&O9iyys?-gy8go^7U@Y|vg&SkENJl@=x7==zkq(@A@R3klC8Yzry2MMV?XYLDO z8x)x8g3+$O&W0jVxdVG>9cGti=3Ahse7QIp`0o1qeTUVI&zob?7U|KZSp9e2E9M=u zB*S1UMrJRMxz@M2S?9fTR;2t#xI_>RUIml(`5H|7O~-tse|Jt*AS0asPh1)DvT+u< zrbPCdKvdZ+P#=8(crQ;@@=)o?H{ASkn1ucQk=*h$uZ3@_b3HrYF!QEl-#i%0WXNTT?}A*aZVpanFdQV4ugxFBBgLA(49P!v}=bXDvIevv|5d zK?>H2{D<`K2IiF$^8Zzs3{@WN9L&vMQP`M+$hpAZzZvqRn1YZBD8krLS;6cyaA5X; z4x8biV*X>V?K{0?pAUb?1OLmho+NAgrcWj0wZdonNrzD{r!W$^gI6T5v7Hjv2F6)i z>e5ndvZ>6iF%o z9WDNr;QUo_P43jOlES8kjW0_ho(sdkNoj+~?SVn*0I9p zw9&{WDN2@|U!7#{iizWW*TZrce(TCm?asm`L0_c!A#*edSj7m``F`c0>(pq58lVd! z*GDx>_h74x)Gi{-iBhw%Y<%?8!j{>Mqu^pe2xnrpAw+fE#?fc~uYn_cBs$%ymYT0s z6YZg45*X{k@;<}i?(dSaAR<{dS`K^M?1xGY{n4QVspej%& zZLu+uD9a(92#b2_U_A+zUpM)qGLMgTlQOTXjm)eW4?jSG!>Z@)Y)Da|txt62jC66S zzugbulwkvJ^|sX%EXe8FZSr!8QCO`IW+Ci9cm8I}#6eqeBCAFaIS>v4-mI9YZq=`- z_0BC)oVIOd(^vtN^c&nom~23vMx+0hFmF!Hft#U4m1)whm5CX{GIH*Jm6-`%>ItEg zq(SoIbhg%H{T|QEB9Fcu0OWC7;S9yB-ImYF#i$Vj4Jvqf`yXTe`}`vUl<7*z#x6f8 z%`#s2e(L>gO8ab#7DOx#JtLsDvVeqXbdEavC05eD>ls?0+;5eMTBGC>d2yWtCrvi; zT28qziUFo6$hvj1!WL2*C?}|oymUNjaSv{G5%`5~%{-OjFeO0=j@+ zTxAT_wHFlFtffrmfgZLT^3sMB5Ztt)g1S=(=MVi@njA{Nq7d@eop;~D79}4Rz+xJi zef@7=^ls6s%no+$XBMdX40s_tN{=Sx;GiAC@#YG!&Oq{8AMqye?fWafWt4F1VP9rP zZ6{sqd+L;dX)_%6?bClG6V8cQ+h$i#WKaAD4VUH|wq=4GZrq(4l9WhA@*lQXbt`RL z-Jnw z75V&ZiQqxRP+ZXC+6RX@2r++Wg*SA?ME^%*z_ex z4%(nh!n_=9OdbF7`_vQ3W(aQrW2_5vPW)%IS1eJ0wdwo(Tm9Ud$*r_-7v}A_D(=zaLp|SY5lB%Z zy}@#(Ba>X2mg_`WnB2ifR-1g*yG-Z$8OB||kHu-)ZoKEBF&U~*0JzknZph6!eLB7_ zAb$F+Yacy~Q7_JX#0Z(2`5KEuFTB7WtYqdJ+>;;~mQ8kWV^wX+9n{iT)nR&}ifIoQ z&}=wOz3pnE0UN5@T=)O)fvS+3Ux`pmJa|?tqxS?JtA0NlR{dc`^{^X>HTR+R)v|oL zo+V$p$fTJWHP@Z09G`d~R309B?9;B{&+GZ0SQv#s1?Z47UpHfIFpf5S>~>pt`2NNC zTBn+!zxKH-FCg>wTh?{ZL|{KnZwB^w8dVM7^wmGA_-HgHXIYm_x$7-DO-17+{gfY= z${VWGV=OI6I`H~46p~5dN&;wKnd~X}6t!U3X>`-(qzBl$_M~_rdO1o_;wA-hx3ofE zwrt~Y@(LbS;Y+DDwJPi4Kbv$xv6W}FUK2qzETW2bI1MBBkVn!g(GK2$a?y^jys%~1 zPcIFQgHH_kTTZ+M>Q?eq&|-`q>O^yD?Y^WBH>+3joIA{F>M}_x$X%NV$3Mj_z>LnU z;Y|E@^GO`5az`YXUV>Mpwj@tA)W-*jl-7TqLzPlZZ-m#(Ch+!#_cVuZCJAF^C#ADY z#RL#M{+QD^DT6_9l0-u5`Q#x`hlYwbf$)oXzoHFb8HXuk(P9AO7B2p($ZKkKZS+m{ zGoixequ8zv?U(oFZ%07Gkc@iE)C2Ll@gSXzzTH0r^{d)D*XwQ*s(p^waJqA`O8&pmx}J%)I^|rXAc~ zl?TNO4a=;o&vnRDNC#6mPlsY+U<#zhrj|12fZCyHE2LdHMr&i6@-$-mSAMNyheULp z-0hpTX6E-LO&vQ}TUCnuws7rGnq@66y)jq4BpE~Ep+~gS1*e2t-yn=*@54Il%X|f` z_F`$PMBS~ALlLh4Xdn~|T=qk-SR{QWW|GN*!ZVjHbs5Jmyq^(GHcHC5>b*8f%Z~#x zVpn+Z{QJ=Z%%~6+tDeII?N!} zQTBRik_nu7U`D@_Cd?}{R3a(QO_m9pA@J%}E7~;*c-xj#)xwUf%bGYpXxD*1J~rmI z-HvVq#0j*VWGW?Qz1lQVtd+pW|VA7B`+)bgy6JNv~~!C8>asn`&RPJa@zfYCJytF%_nyu8Qrzgy6)2KgPj z`&@Z4TQnZ%fn@oLSt0pa8m5^r8~>3T!=7WHuU%WA_d0C=uA0-CGxUq7^ZX{?@}Q3~ zH9^U9_HSRuBhXZ?&33wHO>Hi7X&mlGgK3txrIkKLtOSo_C3sTolTiDQp&8s(yW-pf zpGO1mPM3k`KJs^-CHE8w*t?Pq$qW@EcW44D<2O!eQARz2`8gjJ%0BV^#O+icx=RAi zwo7?i64(YvCvS6db^i3?%8A@6d4a@2Hv#5kF3X95TnsWNwHJTJR0(+6IAPfIfqeq0 z87`aJs~!3-%h)>VjMIA;vdyu5-(X?kAAiZEO0;np<9C*SycS-j1zTts&*FH8wmY=z z0gt9>45c_7sh$AoQE7&8UL&<(&?OI|=6Pji@HFE%W%&d&)C& z(DEV(+ zc^mdJV)?z?KqHAvAw-ub8u=?!atzakY(b-5 z;cfrEvBBa>s|4xZVbz=n64mRc+C_`qRa+^RG91W^ta#*ygU*=?onyh8cB8QGNc?e5cSFs(6jG=FOz>0)C*LfbA$H2TgPw_lQN~4Ho0qUES(SnBt=I zHsMC4A=J=ta3BF$G4kLUHo?uc-css;v*ag49tVi;GT$4Vt`X|k3DcQ-p&^m)>t6gd z#T3l`!qJX%v~`(r-34~AO-tMmHT2Zh-c1$^qr-o(!h8v`$8}!*(>`GZk-a9J) zXdPPpw4^L#EY{IP($$?mA3@RgCJzjGfhN{{AXqy{Rjn~8$i-BI%XcVd?b393Z^8C& zUd7c)+`y|8JeX|JsR|eQ^wSJhp&7w|jd9GstHxuaO@!4X9XiV(DMt`K0?kqieAe}6 z4}KB+o$+g!`LPvAD**N>L-O=)V<9x_whD`mImz1KMZvqQXBXQw&7+wrawN}7LBhSJ z`6=Mx=CXLqq$>bxuB8J-)D0@CSOBkVtRK-%9d@3)MIHRv;*U@(w4;W_2=^s3JC=eT z!?K)5^-IBw+lchdpOz9B?e&%U9YvX_jmq{-uVOJE6V8_$s}h`pLAiq0lT%zcvLY@Q zY{Xsyv|H|DdMAn<`i(Xv_*x|I7F;0IhAcBuH-5W zw2MSQs0uZo*A25#jEZsn2j}7f%asTR_@=O{xC4K>T#jN=7bBn#zVL}E=;B|#T*D{M za~$pYDUE9*K~0 zAmzIpwk5mhp91&L1+v3Aibrhu5Ja^&`6pp!%Ku46HRjpBF@+w6;$Ta$GU2D<8)(V? zwZL`!A(x->XflV!gS#wYi3G$l2^8Q&<^@j)>2Lh(%6t^U6J++vCwrX z-u-T)LZLwK&ev*-;w!peMA^g_r*S)J6X7du5;H=3X}=QlKTN%-GTZms zoHPyqa~Z@Gx4pWhYlK~_23wp`u|?RT^(`{TtWH8)J@$@8GP&WfGfTN8h)l{p9>7h(xr+#Ub5kU{mlT3CQXS=c4 zf)~;`O!~ELTG2Ni9fbxzlu(7_IhOF>J^@4NwrHfj%bTqDDyHO zUi~`70%@7IS-9?!$v)GzcRlB5D(csLmB<$A`6LtIUBJ^Er;jD>x>3OZdZgUHuW=KP zC8uA8(Sn(mNMcC;x&=gjU&HAHvbxnH!8j;GdqFxF@V2T)eMR6pq85HpTtRFp&EPw2 zB1O(mK~ri^Xp^u%Pd|KQsk2UQS1?#cTtH4zas+nj1x3lt}fPTrD-E>=t)nDdgK{MD`DfJA6?tk^Nh zO~4BfdXo5?^?H}~JvM}UeE5=P)}Xqqsbj;6u>1)}$Xols2Z{jsI zAIdXxH7b~_0>d(6Lp{6+8PQ;`DD(cOKJC^^U-r#Y^mi#gWa=C-U7BbcsxqSCniK{+ z8I21&CFf~{UC%N)EKG9)cdt91LOO9f;#3bZUB%dwoAJoEV=b?VborAwLqny!>tCl* z?ogxRMv|Sl2!`(Hgb-$dkWMVpmb0+yRzW;TO$OV4ynHy_G_PGUhm6JT`;n%)W&#sH^a{7T8Gs#IZ9Xs4x1A zR}Mv&&+BEmWd|^GOfOid%k!xGk?~id_o@-B;3Df%TIV|k>et3Z>M#7A%=XQpXMw?_hw?1cmT(YrS(!RRLxWZCjfpld6n)oyA<-P|Aj~0PBHn4TPd&?lMk==T%Z>{--Sl3kJ-_8C*o_o%7M=EQ^jr!(rp8Qp)PqNgr>W}By%^}(QRjS@| zlSm%)Llmg1I`rikQBD-)+X|HY{nkvp47NL4ml)xv3}BX@<}4T~KI36?ifnisQBVSMHHG&TPd@ z*+hu!Y9vDEF|f}O^0g(@c!x3WPm7o0YOP1u(iC-e;_8rE%BevedQZnGt7}64(JCSt=P;I>^lk=^z8)M`XPRd zo&vuLX3@F5X^@;P#_qQKg52_Mv&X{FlJD|zaZl5l|1~TCG8g}FUfU_%C&`F|dc1aZ zr4)BRRA^HBowEL0)gTsDc;lwkflW(&;lcf=Vk|@KVExCs;K7v$XCZlOkJFT@Q;(ob zNaiZxTBHyO!Mck9Y#f&plH<#?y$o)#R#e2E>7=C&s%GUr2=* zWltapvNwpjxQ7i1*u1~Icvvr685aD|8~jdHeUmT~&}yM{2#7RwYi zX-}#Jo3Y!w3In#XWb>-)0@X=5*bcb+9dK#Xy7$hz6iyE{R)-b`B+8D74lRxaB5q~Z zY&tCUseNY3#(w8t!)lj&*+q}JMj{j8SE~86+>m_p94X!EXFB4H?m>v0QtB(iWy(@N$zkehs}1tZUo*^y@JQWO^OI=kqGwcTOHs^`7b z7scv1)U9=M&Ew!06fFB^PtD+EHqi)zWBVIVwbtXe6b!qQJAJTQ1Hm2E`(*z45vpt#@bfXudtfq_Hv6CCIx_>e9LUGF$(70dqQV>-ced z&&(d`rw9Qik?Lo~aGW1u1-IjFf&q{t* z`q!$RN$xHgamK2Jy)=%vn&&%5!5aYY`d3bQd|URLSmPscn(WIAyWXEgc)a5dMn-(D zTk(T;b9d6^;-$aOj_pW_O@|fK#1M9}g}%}+-A;ISn$g2>L|GU$MPs*X-!gTh5ni;% zhD{8@dLP+J7ADjBy&VPTqdI!t&DB+M`5*1=`!l*^Lp1mn^EFR?vVnC`VnG91Xm263 zByi~BQz4*?grC`#bC8XIHmvI?HnY6Hq{xpyJ<=W?*0w8JJ_ z*smBEL>y!9b4>&~067!q$1w7IKS_fFmn1A5vc<`N+orf#J~MoiHl%Aa$LeUcw`suQ z9f)z1Xs;W`A)WnJP;C4B`iCRux4|@=+t0JRGXnZ_GK4{gcH5GZ;&4|3TKG><#PkYh zp^tf?g}M_ry$0yAcIm-#kTLRhJ*&+AY5M3gPv17}q4cmy@~+R8o#>|xIsM)CuYBfH zI1g?j1xiNyRP%*Aejv8v>-V3^qcUF(ZDZzr@W~Y0TYpN$&Q05T71!c)j}?lSyVi<1 z#cI!5Eb95YhPUH9i%OaTW9w(fp<}y?0J_N!M#Xd1#b;J%K(OHbR5=gFV8Vw&xg}FZ zqHo~nI-CAAsd+`4LC2izap<&XYwsO{K-Z^_MV;aR9cCWNZ+UXC{D|>4&fsEPy>zY<*_03ZG>FcGq`$9^(T$bn5Kn5>8ZKiIg zU*m#EcRpv&YBG`z2BwRb;IxDz|HlJpY5WbI03#9JA|SWjus-lj!Zc9X|Lf9Pm;?Lh z4d{vZp;H_QfmoPi7rZk*&o9%zUFftK_!YiS_7>V3*4+8F8JsMzVchVSw?a!taXvjO z3Gtnz-oDusYd;qEOHiijtd|}8wlJk)t1WFmT`6=E3wKLyA*cL;SS)vn(U@H?aU>J! zn;8enCvD^7i7oo8tMiIwU&_N|PAZLczni10oYE#G64!V*&`LHU*AXxkHzcMFgG3~L zFeNcOc3Wv+jct1N3X#!giQbb)2EAgBYiGyfbD|*KGhH%=;%o8MDKbHfpO`0Z7B?s# z3Ugv9$c8Fqq9Vq+RX={*HL>Xvrk5{#h6Bv)WJ*}faOEZRxq_MBA_yiOD22|aB2d?; ziCxZVhcWy18wbD^0{KMBcRX-}jBD#6I;Dz(SM>)YdDODpzUXjU)xoEAE#msW2*^FZ zk0+WayDr*_j5ZIE%dOLz<&$dYqqGuck<#MbSVzejvF*&>rcHs|yNHQ0?4#acX!UNn zCw}k0+vrz`WnTa4A@m=|jn>|mwgYb~ah56+v^ZTehn#rSNyh~Sg@3~$x0WC2t_BeF zx3%0;Uy2@dl|;EZEs3_>UmIQ;Q6u5%?ta`=Ahoxl360RHe^2mdW`|vvU?po0bgeN( z+Lf9$${<2er3bD}_5doOx4t*C(%fw6OZdMP+vTgb$`GfTX~G<^&>mEBp$iVVyb2X5 z33~U%VVtn(##R=CG!PkRCIodHFJLdy2_xC$r4^7vQDHn)3DB-g(Z3($s34p)>5lbG z!NR|{KgJcoY#;Q2KHGPx-?RUF*=4&|;^5NaGhfu2^i`#ZgUp)mp<^FP8>uIE+pX7- z%ot<@XE3_4^BaM-#H-b{K41!^yi_G|%9?cxdPbA2vkdfF|Ym7IRiJ=VdYejL*^r{R_^dhn&14x(jjm})i{4`dO+pv z>GS#e?dBU0VD#mSs4FvTU8%CHy{c=k)*hGv;`s0QQF3e`C95*}cZxQsE_PFw<&(s9 zB7pi;^i(?IhoSCC$p&lZmK1;?))-`blF-y7>&aHgmpNv64L$9lwnRm5cLcX5Rl$ee zO9uX{-cB1sr}=W~SU}L>4>a;`%(c#C)P@`6EiDHDq_mB4F9^=qs^Qi!&h1h;hJOJ& z*1faDE{e==>8M};7WSD!xxIT`A_(yS|BuKkrH>$&w3J$}8jiu*DHUDij*E~^Yt7Hf z^08Oog9NmwMOf?ej+-cfzrRA!@~_chV=Yi*pKGH4YEYZ>5+J|K_3u}sr{}&L78G_V zUfy06BjKqwxsuVcYa-qj{cOyf7}ZBc z&hi~LP1yUEa1I=#US|Gudz6e{JlFu!%G3js$r8%aJ3%%niZdb=tui*yujML7s-h6* zirUMKuzw=T-h%4)S0D$XJFx-)?a#J>Hb3mSMAugNoiw6o0WXdB-jsL*i$xAA0iW?3!)>7} zqO1lT(yz6q;*dj?X_x~MD$8k!HD5@ICt$s`SSJv z)8@MQ8I=q0s*>QqPOQ?xr!)Nd0H(}?#L$1m`Xoh*Tte#7pyeAJhLiE>-wd3&BKYkN z)Pgs>v+QM%_K($zTA)&o;$9lDq8QoXtcm*@n-_TX=YQ9mY&a}OW1e`b+6SkFfTu>v zP#D<~(EYpB^Wk?#xOCuCtTLzW`F;YzN28HhpCHz1Qrf36A`3hu5y&N4l_sB6jELfE zNgbwM3xQV|LOL@vNk51ib0gVkIDW!Vuji!eS*)_a>MyU}T<6gce1;|{Y(4Ytl(ssMG&Anfq( zWq`bYgqe^ik-S+-@U&=Q@v7mabnK7IkwKU)YTTNV1qPbckl+LnKU!ME1g(mD%vQl3 z;r*wiQx;F3(eGf$4B_i5`aQbE+RatDyMWNZS#SK?lH7zL#UR;JUbcu*D+88l%*=B$ zAV&%?YblR)fVyTLHb!ODom{0D{A8AdG-s`VGS8&Po~?0Cfa9EP+mV~7rhC4C_uLsz zq#hmCGp4xeYPqEdifbML3r|7m(-Luq-y&=$*u5vOB6>e@2nHEdqh%>-i*7tF8hfXU zMoP^fB*OK%gw9CWI!Fu{v`A(J8+R@SGw#!*`{<5TDsQGj?~1gw3U{g00XC6>D6+a%}Kw&7$gy!sch@ql&+TQTrxcbjDsouC~_PnOfA>8=Rn7Zc61FS1|js35ZCOi+e zNvggwVLtZjZ{ZWSnS(g|iSNJpbRzbayGM*H_|FsALTd^?Jcss)*`mzgnTSx{9r+je zyyY~-m`JO&3hG3?rrnY>3Fh0`&Lf)4dVk%*Uyp{a`q0dGU6Lp$Y$k~ma2Iq!)>04BXvfYVezfz@UttpSE zzD!Cuk)#sxOIr$=ozOxhd5Yx#UcG>;tb6bN1T8(9zt6)aE*iXZzkrVaK{b3z-$H*8`5I{Y(*0axudOLdcY!jR2lVHu-c(Jn}o z&OR&kmaS(rtl>Sx*nDB{S~nhCq~ZOXwhHCnDI;FLTU{w&zC5kHgcg~Rp|bO)b2CM) zFc7o2!8Q`pkor}zSuX1QM`kZ$+GvJK7&){fYGF_loK%yGf)5neG-Y!(_)SF6HbI{E zcIA#j`M408XPYh)`t!wfEBljl_Y}%`$=sf` zvtyPQzfH>y4cT``4iN)DMslI){oEHGe9YD~t3F+WVHq@w)z+$q?{vjlh0Ckhj5c;; zIq{x`l9a%=7BH1}>3jUy{~B1R2)_8?!ykysE6@j+|3x}fd+HK}u`GZ-X-cbj>q?}I z-fPitXX1_?K5!4mu(_>2#;?Mt+a9z{*&e?xhixDmy*D8^ad}R#pECDqtT?h0=&68R zm6f^!*d1@iWm<|hXG28P#Q4%f+xIc|n&SV7IgdE{8Q8IzsFjpo7%6-P4If+vl_cNc zoH@R&yFt6bb-`{g1x9)qK6G?SpKo{EYsrusytQAMUpE!M#+42lThzg?lFMHpeM!)c zJ&Ps(7McjxD(S%0Z|h`IGJHm%h_WN0J*Mn;`JMkO&O5Pg*~`dcT3G>AY30&-Yhqa%

jteZ~3to*-?H`F`wHI(TZ{+El=`#4pQob0KQy*KuN;0X7EB9nnQG53Yw-hN7@V z;aavxO36kNxNLLW7I&6G6-wO;T^){;pVR9tqgGXs>FGL%+BjV&wnx(up|an7rV{V9Zro@K2C9 z0$30KeD&&QaTWOKb0goj+3^s}T1|9*_L}(}MH_W( z^9BwtaNFTP+t<%v7PnJmwoFjmv51vEc`>eyTr&acy_H_KWl+#QgS9Xz^^4!Xq8W6` zjZ$+6z2c*2K;FrN3toB{pUxqF29m96;u9+@2bue4OMk{^D#%Xnx0yv8`^NU(bsZc$ zO`0=X{dTf4!fsT^_~ZUsM5b~oP#r_>j{ZKqz7sbQRrdN+By2v-{u4)^>m5-dRb6>z zxrE$xYapGRwWW{v(bCq6l5!+d=wB&II<=q$Nf1uT#XumNMwTEMUCzzdVD2buk85UZ z=KeC=8Uwo9={U0g432op1@edG7Z-6Ny#i^;xLt{}i4uAeL2pS@$9%*6#Z)-IOL2ak z+r1mh#=%1c-!NL!0Jr`Dhs`8%gS{~o-KOX_)!S=p*Kvx7QGxuaC>&hL)P<|g*DP#7 zmMa0X8$*bGb$d>$W1JCCr^d7aBWSJ_gRx0I*8(#XhlbYmml^p{LWkjJbdr-j5}fn- zr!|90ws8`M;>`6M9EKl{t=3LjuQF^A9!F!YGDpCI%adu_^=Q4AAZE7``!jV;uBox4 z4fJ-@Z^#9|9TZqLOIj}mTJ1!JefYIh?ao)7*d-qiEU6@;`^6igvEFmG}y=xE`s|gva z5|hWi&1qhKZeQAA{Mg?|rN-sJ|M!#y_Qw7F+ER5bZmu22Mr+g zc+2$m@cagwj$s{cA3K1lSl-yLT~cJr9!+DplRCo}S(ua9sE1aWo?BFsI9ipN0sw^m zi>%FD4r2{C#s#^QETH}bq}f5udJIcGtpHL$t-lW=hixS(+i7=?RoC%j7^bJA2H^kK zb6Cx3iWXq?dF+7DC12hR^HTuKXv}ONGd#5lJKy5gev3=372E;gLG?=FuSy0 z3}mM)VsP|rD2N$wF87Gix$C6 zOeV@z(Qf}fW<@5zs|mfJydsZh<$@M!#{>_9?dgbs;*85wu^;Yog5|SWx9T73uAv@A z&DUwSAhj_wilQNwXu;P#23_ZALI+Oo1yB0}gs-crUEY<7ph@19_Z;Mm*%3m;nkMi6 zHcvm;Z>yR+-1#(Y^r2KM!`YgzS`NFs?9k+O!0F;XL4u0c*RVaMlI&5!U$lG z0w|SF4t;nX_Aj)eJ1|d9GMAKVQD(4uc4XpJ9kbH)zofVZ?XYxo*AfFEI~tjTke*e8ao6l5pTJd{xHEa@oKpZxzG#--QY@cfHo+j0MHoM_ zpU<+Q=+wnPNxG{o28;fxB>AmG;DgjBLw2&+ zcF5QXCJ`M4P-BhS^gk<#eIjC&D6J%4y}F1qt5p&E>~K-5gcH+p^NUj z@Q~gehEVdgEM+%>tpdsf(cVUsVeJ;nU3Xu`LQ_o6d=HxPsHmSaCkiZn!1E)uTUDAT zay_5_o{L>%YQBXE%{0Zm;B`O|O%$t(ZWeEzJg{K0N%f@>;@(A`CKMZWF~v_YJfI*T z`mGfm9rW26o%8_D#q^xEN2*EwrRGO`@U5w4PfuA%L#l#=`38bhJ&~x%ltlKu3No@> zoM0Bpn*o(Sif_%Tm&Vb(C1X{XMJI7fW!WI-EeWz){_(iwz<;U}mNs_N6m@mmEfWd# zk^;-LKdwK~UHTaF#S!p=6U}(5@ZOPQd$NnRROHZ$0Sv5IS`FU9yxtS#ZjLURsU4n9SrS|Z>6R4fp z7ts|uWbvv9dN>x48!fUaZjNfZRksj&^c9<}Zo#%lcyBJJ4dJo@ySO*rZ+Ez2v;%9M z0(`x5x4p-a2T+P&avr{UZGOkJfv@oACzzJSxS-3|c5!OKf%Y?5+Kk(+GT?T3br%jZ zLqtmbJDIh%w1zi?P2d)8!pi_@|7TQ{{R$HKD8Qm$10@33-$&^8!*zj<9Rgct5qX(v zqY#H7V)IlMEK1iqW8}9Le3`F-;9~8fmK)}KmGVxFy6g2<+2W*RsA{X{;5?5GP)PT{ z@b5FTRoW1``BoEqUXfinx0CwPfGYj7-Cb{S=;kxEblN;HJ5#!_BgX7ZK8+R z4Rrz=pX5H7;hX-D%z8BWSu*Muwp-*2<$eT+aGto-5TRD46zh=WtVT9(`~;Rg(``Gy^IX2pqA zmt%3Ry}$_Azhn15V?f``HiN-6XZH3dTNXXk`0?#bm$4V;9>5DCPqNZ+$ugw@kmz3v zt9Fj}l{H7PB&24qasR15+{B9dLTsxjL^MAVF4v*iV9Iqv87;Us`Lh4t{ZZr$pzLpX zpH+2q%UX*DGmSXS3PV^?SXyzmZ~H6eHdyG%0$yOv)Ip5dgy&&P&DZK#o1MiSMbG%#s&*W#f=BI&p;mh0o%Zwk9OYcP5|GAc9mbDPG)~A#Qd>P zU@*RAk3{W<>cmQfq>7bmx&u%R71*mj9XYAjE{ZLX&8M|sO^UN<#FDFsL$wkPK)?_dqQbcNNDbu zoZ{G`@x*E6PMBjvZm0-1+&%4A_LW+a-iQCMhy*xH6Q3ue9}sJ5{!UGTV1z9vP2FM1 zY-Hdg%1Wu`ykI+SjMH5|IEbiqu*+%yAp6lGQwFQft_a z{Z{?UhRTix6R1K2Ytp9*3QBRiR#wPNT@v;${ZujDE!`OkPp^2rC0xbu|@A;(>5cBW5b$iERw-ju+XB%|Cq#^h;?s8 zT~k~Rxg$ta+hgYITeUc(6@cU*O+@|Rdi0idm^>|{oK ztPY7fNcW$2-g86&{X>eeerSwU6(!Db+aoqKZFV{?*Lh9^*2744T3j9%o*erG2iryM zPd5vjg0TZRq~&^r`3vHYdj8mqhOP&uCJVHhzY%pKopktSS}ze_VpgB_jph5P7zB)s z1K%;BQ>G3WRvl|Qf&u7}Z`ymiA9B^A&RoiKR(djt$Fz@Z6}uD)W>D4!n> zw(t$G=s8m3@*5hHt#*m>X8ywAKb9!7w|+SzV$YKl^XAXv;!Q6}E=F0Pyp6BEekL5V zD_vHU|H3Bfqw{MXK49+MzDdsdq&f~u$2e_)FPG9yWz$%%YhbAth<@mRC)un=-eSFg ziW80$6s{mUE?KCyqGliExsQ*MnxN!+l^rv4lsWI^LSsxm=`!;IWw%2&>SJ+rq8QF! zARHV}=!!;{299v#-_Yq$5z98l#2PLWrB8AhPHTM8KlXBNnD^EEt&I?qMTTCrl&9pH zH>5-Qgv8*jES-oJKe|L@`nA!u1~>6zO8#n#8La%6ggE>+^0C}}mkynOF_7cOQ>IhV zKokS}+nX@%5=md@X)a>6+bl`x7?Y)V7>#D$PPHB{owFpWj(tLjaM!`EFc@P=1=jQx zjfU!c_6@Vhl+7K{|0|T7w5FRj4`+8zl-W$lp=>wv8UdL0+zJWXh49DUY%^`3Hv!Gv1*nIPQ$05%M~-X}I^2ll12)NyXO3kJoMWY$;iuy2+Ms=eo(~|J~$sBlzc0E++RU(9GLtAauDzEHHl^+ZVxniy1q9w1*>;XFc#Z$i}=5 z!Gvs;;HfQ36IixVMyp=R64*mzdu6*QAH+yU>Yw#~MO6HzYo;t|bs_7NtqID<>m<{G zG0GA#jjE*A=2Qa8k^&n1vR74V?2mBSWQKjV^!Zl7--etl|Z@JV3nyU90TzbA}V_s7ox8(dZ2|?5(t0T+>3Cu#4m!kX@_> z9~O|H29tFzN!j3}VOALPNvV~sPq~dX4h0?(5PIDpaFh{`yKYT|Zz`u4UyE^y@eQnC z**9@xlrA25g=U-C$LyRyDfjOb4%ulsNn41WBc%qapN~?=ymL!g-0}B|TOGu;;McTBoqzDQQ}g))|rN~2ntYRW{XV_j3kDZ?SdJ}+ss~LcsF04?Hu#d zvF+W`SS+@YP~`Piw!!r_X*5fPrO&Sk^A(p1=_Zqp5}HbZky|fp$#B?nMVnIGNYLoJ z`oDa~mByXiw+=V3Qo_qNZBWcQ7Px=1V zde&NBAh2Aj5tEIKPTIzb&~kOG((m|o8d0L0T!qPCU*P2Y)8YD$8E%f}T=YL7M>>;+-WZ#@cPAlR$Tn$9`)=vD7B1 z%AgLEUSZTF<4~PAGqlOT|3dSyetn7LG3vIqRbCcqxSxn#ywvyhWhWh7z%^s zGMfx0G{V4<(@-wg5c-qO-;@mNxs@W{1J{FgfOZ?IwsY?(S_D>@KiY(e*fSN4<*HT{ z@lW%?a&K{{3EWSR*}lIfM8fh$Vk$hJF?7R*chGz|{&{vfX$L(sTC8f;3$+eRl@zo5QqOm1%>55KE?%n$JBNeUnylhizcm&)bm{@68E&pZRM&>91;RuN z5*iX|*%R*vU2{!6$v$|cG(7BoHqis_ffh5A0VccP&Qy>VGw(w^X5Kn2dCa^{Ic_>@+gTg@lZ&T3 zRU25bKnw@%PiT-nl_1v=dO_BQ^`wDvR#pHB6WIFTT&;CNpN)F%FXQ3RJ4#4}oKFxH zlA?GXO5Z@c?YjX1vG+J-3&^if$)T*jJOO;5&_nVRP6f?>F=s;VTbS53gqdSuL^97p zBk9!I3kfvx<~2>arG7hi!+&AIC!=A-Vlf2T-M3m`S08l-{P2K^0QRcf1kP5x4)S(j z1Y3}zvBQD&DiH>?)J4D|0Xe?(H}ub93iZiO1wutpn9L4g<1u%7s!#&%_u|hgoTgAH zUWn@NO@jdu?knlkg-{-HKGX=%w?fB3=wgN774S#x@|1D#{UQN1Y>-mUTiw1_XEOBA zUr(i=&QBf=5GT%&}V_o4z>`b1!hbW)lf7< zEmt8hCY<^FEv6iD=AahbIZKR&i62`H@Omz1fb47y3oXZ4jIebZvO(xjvyx*-CF}HO zKc}espr0UEA^!l1*PA_UM$pW)O3UlX;0nSN+?NrzH1A8B;OGc!7vWp%rx!y|a;*f8 zk};0_nL^Z=9B;D2mD{RlZKhNLOA%{z$YT$IHQNn#W?hU6QvRlhblkb*`M`*uc~S;= zW`)$J9|lgQsZ9iLVEl(=2V@TTy-X+YUo>(P&xT%XftJ)8+6{bJiXi58ZyKfs%c9`* z5te$8=@(5(!Rgp8?YsdiqBLe0ULk6{o!mL#{GM09qP3g}CT~SdI(DteJVNBHr8EvLN>(H7jes6nVJu=c+`w(Vh}% zOse6Jd51T6h#m56C{{a5gqh&QSnBIys0-Ekh&}no3<}iRf&~%|>5cR-HBdcu*$|S; zn4r(=f*JZQ#>WbsX9$D7`xqAbG?E!0Z#?s}DT%}P%&_ndTJ{cA(x zb4cmt2)t(SZKDQ!GcYjU{?()d*BSgjwcRX0ekq2O30v7Z@c46!9+FBqW63O&9GAlPinCo5w?)tKNP`ufc&fHu`AopEu29*EKIze|-ZJ{?gAZ4E+5%Q;F9v zf$IYD?9Q&%DCU~c0lsuYWA7_hlchQp9x8j8tR=aO^Ur=w(4Lo1vm_JOgu7P`I^#dt>enlm3!ydG5eK1uqN_yN9md$Kch*DrL9DQn7asFD7QeNSJN~X znqc)@`tE%+cSDkPuNisq=2w8Z%qHlU*7dC=q&ubbpXp3UdB5+VsGltYD)$*Rx;_b*8VQlL zS(QM1ZyZ_xw=*nBdd`x2uBHKIqIt;{;(K)Wio}vh`n#O}VwePz3WJIG9Zjc!ubnIz z7V1qrMZuj}$|N0;DT)972G3XI^05sZ27%o>^mfoqM6bJ$DOV`-j?k}pqqTl$lEI~G zmnZofjiU%^kBX(lJF;CKsJ5-eYWBXZ7JD5Pm) zOILjN9#6Gc(qbq^~trSAx!Z7rU zE^YJzGb7RkmwGVKuzVq_&>iUHdwsPuHQod1bba?NaB03F2|frH1g2Qp@a%*k+Utm8 zb>_-%>LZd)6Ig`Pha+hSZ>vY{cuj2*D5Z}0o=>J8QB_xvP~paK+NtTF&bxLj%CjJK z@SaH6{T7#`{bHS`yuX~p4!OU+I}K%RwDmw&!Kz_sx~vCO9AtFBKcl<RR?R@GXs00O43P&3X!IHrUuq;n$QjhYroEOWiA9&gTn|7i$Kd}t&F>VV!Z2LBX zTrz7K@j%%KZ4lBYsa(-SaP&Si<{DYk`7=c&oZq6iLg5^% z&YP3UYl*`*zM4|ZjOlw}5`lhUY&aCksR)(E>5n`xuP###MLn${(Doa1v@$^g*v$wA zbA!!xS^w`apj@z9gG@vgJFJnUKQXidN!r*(?8BN}8 z>BPC;t_Vsu=p2Xz&YKDaKqWSLSEE_b5Ly1CR3BNGhL8cp@nSF}>an5zD0?<)DaG+G zIOK~R6Iod`=;!I|;QtN4X~^? z>W$*|6z>fQL5dxtW#QHmn*#0)vV}tXYn%dfi$$BadQuqd*o@h~8Aw?2SZ4yu zC|N_Oo{qAze27YIMvR{F>@n&#qC>gHsClKw*{>-y*(TJ1@-Qp*>q<;!3)?F~U`@QB zhx_L>CTLfmQz%in;jR~Ppn+9{)C$*_5Y%?N8_ zFu*R{BsiqiCvrH`9ig-8Hm`x#Ht8am)PXeoP{D+OD~Fduwp|qrImV~qqV71xeU}qkxAN6T2 zFwP$jji%`8{zAriFn2mdNcY_={FBxlS>-e zZYo#}2?O`#N(V{=47 z10`mJumvHQT$awZSor{XKATA0p$>i4w+QB*FvLP~p@QQ~asJ3w^GKJ?L-)hb!I(E@ zZPi#|?_<^kXIfa$FInMLX#WYD4mAace-0mQD_Mywb7ZzE1cd&m>$?qNQo+AWrlHmW zb=HVyFIvN2SV5GeiGZM?g8)Q;Jz;G|!sh(Jqx0 z4(ySc=bx2MzRWsqUh~4ydD($X%=kj z(em3YM}#{h>dc_)ig)1pn-S8&QLa`jF|jaX2bBqftmrix_M<~z=z-Q9l*_5%nME&n zmrk|8xhmewut-lA*O=2_b~BcRR`GIl?6U|tE&EVe;G1Goc;&|g_^JoqYM7ok!?lM< zn!1UyB)EN4+BzmdZiU2m1Ose1#j;*kUSL{54fd;y)ms&BhAJE8(6Mu%(M3o!u(prtL$Fo zJ|UX?XN_Fo_tIv_IjE6Gc&LLc$>-g!G1nNn)7DVp8YjJD^>h2Af;+zi-to>-y_u&J z_Z|rO(jc)qNaKO*ZGt(5&;?JeZqL5Cc-aOYjB`S+u!h}1^i?e_}){S80j|o-^xrzMCoVJfVUD`;cXcc(yRdC zV!YWBIxa+)8i|kJW|^kA;~T9=q6+)Qb-B%=jA0iv*tT-35lYX}^5%`~PU1VhriNM1 zD~mg?>q2@5zFF06^BaekX^AUQ>cs8Sx$!QSoA^4GL3DMOi?OcqFJW&!#6hgpoPozzf=m*%_|+Pg5&P|_8~-8#Rc zhz%pY0;wBZ%ll0BP&dyDN?qUQPYJ0qLU9gng%z|aB;Z?3zEnrToKs9B%-w*L8`M&f zz!RN2{o1ir?N{>BU?Mc96-gM#%rm0a z1ku=I!jSYzrP2=jUg2V;AC<<42-U3wO`w@8WI{5CA7EfK9Yuo7sfW;Zq>obVE%jR< zzmLdUtKIbb#`r#mzS{K5-r_o~m$lip`ELgo*cdo^n+=93D}|QKd+j|io3f=q@GVj8 ztr=GjT%B}ma37V9{Lj|zA6h{LL+h(+4uAQUB-t`+{@!Z=z7Q)D(C zT;FmW@N@BBNKY&*9)hiSWT%ydZd$z5v(}6XK0cv}gWg$q--SQQ>Pcq-uwL6O0d^=A zLdhfy?k8?3%_dZ?ajkVE@Pm;xs65rYB?xOvaJGUA#UJ2K*j8PZl@=k8geVAn3plcg=> zYH0Wardi=C>SXmAItz;4#)oY4+$?x6K4f9MNnrrvGlR4;TxIY&qSdhI0c8PpHFnCP znN|+-r$Ga`(fB80W(`n7F}!!qVH5a{Ak{B9z^Y;L34N?-+_1A-M_&19$KvrLF7BGj zTC$fk7!Gs8RMBwpJG~k{?SX1Ka_k8kGY}hLoiy-8dw55JKnOP$*A9`(bU?h-sKxr|)t4q$9xig)~ zsq+4N>DVFnOLp@~nt}W1=+`$6wUNt*7J2PeEno{Ji!%JqT}>tqGL^_o>9NGsbUFt; zX5!t8ztSkcHGx*>j+yd}F&JG6?wbh7%UuDXCR_MNSz6smb^a4Gy8GyBYLagm(Fb-$ znXO*uO^fbh`Mi2sj5H_K$w~gRpnBe3S;+$Z)o7NuTm^T?KBzBvAaoUNfnjyzTIn!) zUZ8aj;@LBcZbUy+rvBO-mQT|rv-8yjYO$`ztGU%e+FC9hnk*F4!T+(k0VKY!RKl_` zsOz?0R%AfUWJMb2tkSV!VQMq%+)bsz91}vna~3OyovYDH+qO1FSxTZx$0eTqa`?q> zXT^_CI{)>!+dm&(cQO&eHNpV+xjxkZru4)6uLu8kj188Jx5PZbgv><^J8%xZv!+mlsUP|91y^|5~r=?K(_gh|2?OweoOq)aA5rwvng_q>H|J zN+*N%phA^=%)C3CMDg<`ekT^fJra6!>=F%&#s(t53j%&?>luX;FA zN_9~A9=4pT8LJ1~7kUlkKQtJ?c}^RqB}#Wq_VcHYA7K6`@X2&l3b-C4_pN!|nhMn} ziw-`vDZ;>c0{QVmjgq4X^ctdadM9Bl>(|MeA>cbr*9143q)Z_Jn#^5_j+kFY^RW1D zd@RH#8O`ucf+-BjD=kJi+=xL!->j@{o|@TQJS=Fb;cG!uznDpb9f>Fz4)RtN-2a)4 zh4=Saw2btj|2HwoJ1nubead=bCTyQ(;Js@R<&QkURP6tW4};qW1y0JK!JimHV8d92 zh|Q4H_uH9qqXz|=;Ybff2t0daGQoMlqM=C1Kp_;WGtpj))u@lk-cZ#-@Xi=628zpU zYB-dKUghwr&0b3@r}f}`nftGRo-_5B3@(kK20)AY68}k?CT_f>(Sl~DLf{GA46>IN zj=e!iXBk57Ijae>KU7=bs}__=DecpQ4~8tY5*xXyjwKXy?+c3B=2%f7gwB6zlMX(m zRB)t{lLy{?X(KTXYP?3Hdi%D!S1KiB^uVyQj01dLE(CH{P#ipMDj1=r2#NN`<~9Qy z`9jV~z$#fy_b-NPoxU3e>owsOrcFF&+}pe0GTY zx5xt&)n)TQggQz>tkrLBx$IfeX#91J|L$k)NmFRj2{~g~zSbDJ*6zQbcy23o`PfqK z#ka5CsPFUm-e!%w8*+cLnxXs`g&Kk?SW`G_bw8l`)iZh*U_Z^sVL^m%?g0BM5% z=>!5ro7HysX#s+UYqltZu=@&Ka4q(^ntebiI;~fdU&yRrME8eiymKMvQ@SqrJ*+H_ z(d0iiAqBs_hF9i`L;j#eSGP?l4(!rVwuC4?9~(@7nxIRM<(=dnAuB13b8?{z1Am}T zq=(8oi64!>CFy|bXN_G@rk9!j0Z`h_5m~5OX|?E)vNI9wznv~owsIu)Xwi5>IGp*C z;^T?zp(@?ySx=^~xZ^-z6(2(efy_!)X3 zq&UsjZdoQ842+qmPlg^()Iu~-7I(dUftvj6u~yfu+WiNK_un&%(Rf18{tnDfWQo@k z83OC@`yzhc>tQuvW`>37T#rUwuJt4;S{4Qst97W(@96~r8=H@4vpPTgJIS9M(D)5y zD3pI`4uWho#c5w?Lw8;2Krq{?MZOqi4N^mn5LZqC8f@r}`- z)tD>K{66F)8F*xs)&u&FsR&4HtJf&M8BbmMVNQ$yyPxVTFmr|?li9P+3P20lBSF7gwe* z3K*4u0P9d9yZ&YWgzq|-(!qI%PG?RUJy8D45+tGI)VHLGv>uq5(=Mr!;npwabZKWu z**}~|ctBb*aj-(u2=;2&G^mJRP;DNfqW;q@AYCw`Av%n61WGy2Pg2#mhG_!Ddt`{? zJ7_fbWlDUaA1DV@(sUaDohb@@wDJ|GY%I@!+22{yVP`v8I*c8RfpFGVO)zBJk*hUd z@!dfcfpQ=$L#2c>D;P8Q<7Eoib>6^(rwp~yN~=By`uI`JZ)8!UP8p0ZxqJpK%09H! z{}jrnaaO3@ifVPo0%e0Zm8t>K;$KQFW?FSG`THe*zvSw9E4pq3b3n$Q^4uxauX~I!l$j$STMhAz}}p`T!j~q zPM2QxCH8uYF4E!ktMJO&J!%nFkLF`wRvQM%vKejk9vP{Iw)a?0IuFPuotshzc1f#Q zeUH^DNV`cJrA5s)sMKS}Ep0Ask~o|(N!89~I1?fmOIAv!IYpX+U}gi3hw&3EdTJc> zovgJo?_n#a)U^ad+p(Mqh9n9COrS8%oo|e2>;E22C;vM7#p-ZN7|b%s)ZjZV9ltY; z<>A&%s;+l+5)YHM>4U*ngQ51IN(a@0I44Xvgx_)wWW&V&mPIIPIIvbaFnzHgL=dXk zoaq0s00rK8M-vYf*Jv#;mql~%i<*<#`jGHu?=#UIrFEoxX4)w4>@OTJ!YY?1|A&<) z0i67D)cHpK_*3O>rZ}-RP?XVxj4n3@%1?}cFFzqQOqVUgZ#E*(W3xI0N)Iwh2#rUJ z{C1kgY<+R9%RdUzV*RI&nd;kN(j+8UCkSExg9;S$3-(1}p)e(?$u;t6YgU#Ic@w$F z|ELN@r3rDnF27if=z@`33_-AC2O6_fhm!W6*P#TS^3+zrNv1rNu2K#cCM$zr`)s4o z=-;eF!N>Mg;$vI0)Wqlcw2ids#8J04s?on!j#5k5J44)_L5C&6);u2VflYE$Wav-A z|F|+G=*4Xx|9+kQimlzgZfP(RKgrS89G@PrkVO-9cn2T-;y@2`nB<{dY=J7!3)V z;v&EO?J!G>!vz=( zDz;E+C_joas`p)t=?6lr*`WMY?KMH95yYt})4FB>?2(2z`1p4<63;b!q5oTTB0P6T?LXTU<^NPU2mNG>9g=p*1rBpS z;TPo8rU6VT_>|^IZ}w@OhEh8EUf52^wG(CSF{;X(n%AhH`lHS%{dM*g6R=N3n#tcF z({o9pUyAZ%RmTm|k*-GY`#3bdcA~&vy>YTEF#4fzKXmqoudhxG_l`9bZtbFV`oeqkPaN@wdXSpsWc7#|=&)0trb~z;7+{F(^D%M6)n;KD zq;z7iLxR{~-6*kIMBt+Lm5784>+uD2_`zz2(tO&Wi{Z*A+0cy_H{?jQZA>2Mcj@g= zwVyT!^7IkBGy?P^)oJS#J1H4S+rbxYGzN;69V?nr8VVurC|shAbzo_VLCZFjKYaQK zLAYNZCS2MK$N<}qm^o?hlRIGsJIpCororhth|KSA)Q5n3B8oZJV#e^Y%H;pWDib^O zEH~S{S09H-M~tbwnmsmHKakSEwMawVz+rK^RU!u;y3s$fn;T*wN5Lv!PoXLS4p$j9 zP}7xDGhxV!eL|Qx`1YC746R?|B4ErC=~TaVB1gNM%wJ$k@s`Djro0Y;ge5rx)$y}- zTKDgDyMVqe;?0N+*tHc&R#D7c5Xi~S;pYm-ulqh?WvhptPKBMh>d-$I6tZBg)ThAB zFLX9o_fV};Iep{u35Ht;{C6t-<61CLuBQs|+6WJ&GzS!cJg z)!@7+;v0Wuvg?p0QU-Xakj9t|C0c<)8>nz;MMX$~I%DneHJv^+1O{6+^kOs*)zZca z0}t{_SaFn&fE&Y*KKU62R_wJf@X`g(5k&HiO2jkk@fdIh%R}MhQ)F)4S2`7hRZ((% zHFMy_K3-L)*#e;^3my$nhC<0xeUORJOPYN!3V{Jxz|+v-YqKbw0%|E00(oFEGYQ>j zNV#K7g6ajB1Z}tFaga0?<4|ucK1P!)x-}Dw9-&-D#la*KuYs&>m?6h~q_#rMZA@Ry zr)WbVwIhnHLnkonJy?qxIWrU^_~8mnqNR3)hT#kj@C`O2nqS3?aEf8Y7pIKHL404# zkJsO^*7~kB!2f@8(pWucJr{$${id`<)p(Nv+9Wb6xGB{Q{xrnu1ph6) z5BFGybKWMYMk9QiiV>}Rj?G?Dfcd@IW_3_tuIA1elvlb^ z5e0?p>aFunJk>w%6_oDhPbZM;5(@gWo<98T%8x>Tr1VqiP-wGBG=@et#8^H*j`@7d zX@jec#E9h+jDy=PZCpDOVAsonq3oE<7~-LZoc?lbFd2T1kgF0Ky87&x_~`!c!P3}V zb-dONvlm%~h`@lgCjVK`xGJ*cO@#8r(hUn9+m-u z=0qT$Tu&v$68hA3o7+=CkwV%^5LrpH4;92SPrB5iMw2CL4_vCmh4@#cm+QPPr3y|x zK;4*?Z*>|3r(@eFLKF>(vK^j#*C^5riUwPAAasb_LE+OHF(55l!13PX2a516_Tr3q z1n2Ejw7M=syOB2n05?F$zu{qcA4AN3|V&RI#q1##NGSGAn0|THo@>*S_}+A7CoTi;rUo}gSI~j zN_9Zk)(&Cs{u=qHZ##<-p53O+65y89VS-@SVF59%*n|;Yl-G_=CVp*Af@W_cXWjf< z+OB^vBBEc*W(SI1t$laOXfCwPSLK(!`v>+#>Z86jq{Eg$ya{gK&=|bQr_|781HOA7 zH&X2Kr&gGRM=1P| zbVi7V??%u@4WW*G_PT}m8{Mxu37Y$SFqED*C?IZ>!mLr}Kh*vF3c-LabHTDNsBn1i zm`u>Fc;!8{v(WTrN78y>Jt%$^Ghe$1Q!rO3zxVqSi<9hr2mQK>whmOhLJ9CFTNMgB zE36`%q-A>j7E5&>LLL#>@vX_kHRah8Xx|e-uy7Ku7rLtpR4ARP^dV~EN&5^YhUU6`|gBYDK;xqyAeq>+c2};9;Q^o^D(FyN~YSP zU|be$hW(2~-n(qN2z@^?Xu#gps)oG=rVbQ;q86oMZ2^`HkS2f+=@4N)w5Va@NsSra zR_0s;r9U5aV=ACsMCF^}@Pnl;!#3<*~6YUr(EkVZtDTiY1r`mn_6j z6IQA-;NloXI#i}8Cxbg5zv7%uCCeV)lbwT&&ny2Q4`LZ-_slfiH2@GT13$kGGnEEeT)PI$kI#sFmv@g)kU z*|ad{BW&vYarozr%PbEOyZFK29;Qo!+hNGz#a~E#cT^q)z1J`q-ptbt0tcny-# zJ@tX20q^W}4dKQ%wGhbYO+}Gv&-+wV(}>hv;js?3@zVVsl}Ap&EqIf*Iz_V2w&f zQ(hES^qiCwF{1clP_AwdjngsUH4?%WJZ<2BJ2QC;jH|0NHqbg9YdT6TwEd9fjrjG$OOS>$i6@URtrmwO zW}M%zvsj3dWEOa+??03_#r6F5H)>I1fjAmL=^9lXeu{^>cSD+<-u0ZK9>ea_tTFh- z!zT36I9H_{487zvJMq0GHJiuY^Zw8?U;H^@IjQD;ma9SS;lRb2 zXaI*7GM1b*?4^VZT;W|$LsD{=p_hYDq7%D9e^ zo{wN^T{N}w_c+1vDpF19m}{O@)Uu}2)$Y2IrMVBU9UFgg|| z=AIG5;GihyL0d|_7;^Vb5gM07>uO7ZCX`Z%e-6+Grie$jQ#&M9=ReZ6fL+7rEbuJE zn6i42PNX?AjJz)ngfOKk7A{EBJ3;3Vj_MRDOwgx~7R5weJ{Bgml{PiFX=I^(U6T>s z`2bye=oX6z7C$S(h0hQGYwpQ-7;@Rb!}IQH7Itm7GyBG~%WgObAe(Nj>?35}E5!F7RcdF5hOMML4qw-D!IUs-1?DMG@+EL-FG(!|1(pI?`az6pH|> zN6Ac(ZQwbr1ADB~x*j`bWZdXXL;C^BXlYj(5q=-1&H`t5MJTv#D^##8hzf?}Mwmx_ zq*2x^J#JFN((cG^cbe;iz_k-odQLK4J!cfsfIp2^l;j99_;51{>s8YgMsSVAXSGQ) zFRj~C4wyZFH^7arFja}eRWWe*3U6UvK|wclfhiK+-h#inpG6zN8KV+mxx}!q^rvhf ziul1rN$EwbMIGK&lv%<5p^b+SlU_Sts`0ut*ldCcpJRWEVJagm6gU%HE4Lb8&@@9N z^gUo^VfKAJHc^Vk2scgo<}l|LO4b^)pn=CRiexYaThcAlRk42^u}vE?PUF3JGFdwP zHv|0fXbU4cu@-Qg!~AX+*eDn=34^q=g;Ec$!xkB=7-De3Iv%s{1uIfc`-^Nm+!}~! zv2GZh4(A3~>VWRNDiMm>%i6#P3d&g08goYekJ=zu-&~UpWrdjT?A!P#>G*NH<5z;* zd$CUv9Goc60GEwPhsP|UWpGapvu_=winba|5 zz^=V?B3vI&ne_f9lv+;mlrHl9`!va2p9j0UDvjRKw(4-Hgn+Dm^KNv-l18Hq#DDin zd#dSsG;#Q2J5r=1br^I=(OJOrTxrpr-00a}NTFDMP)5kxtP%9T3lVTA*o<_g(j>z` zD`n6nOXXmwlSc@WrY1_b#i*z52cWR3$@X3quZVF=)Bx2dScDf%9?iOlPBE*?TB=mM`WS!r};|o$cjJci0$; z1~p+A#^kE#Q^^*vP~JuIZMxQaqMJp|o)?_Hlji z*Ol2oH$xu=l}e2kj?@^t!id>Ov}aR=D7g3=C5P&FxYR#;A^~1kpreJ(QH6kipg~2& zK=oK#q)B1_FBJ#8u=-^59Dze940^dGIehm=4S0&G=Ls#hy?{YBE=Mg9!dO zw8!KL_y&Ck3vw8v$8K~!V1glQBb1y|b0rLbZ96`(?VQ+F$F}XHW81cE+qTu|*tTsO z^VNO7Q#CK^C#>4*W#5N=*u2pwJV7oY&D)MH2MoD08ZmgwwEFo!dX*^$AO{(&5@G7>hj~Di;Rc zuyE=70@&BIJqGj7EU0om;fXO_d;YeD8DVnU&JpKe;Nb7!6XDGP?EQ_JaXS7ctr(KY zlKl3|K}gG7Ko&fjPY?mpUqzGT?)CH)R59dfSUWszsfwH#TqFMhbo)uvovf3277Hq$ zeco>tv8x8f=(EVZtP>jAwrOk}TA>-iA8KJ{k%{AtNq&1om8rbhaDt;Z#l zIZ)eMh>hoP`V2-fD!YmPp7e~d@5R%4jyCCG%;vS_O@1hcdNk|RNQbtGD8eh#9^&wm zzg-+K^vn#47AsIh$gP`kjsI>5rm+G6uf`&tfgo>sYIH6htdvfLGw|k8umW1Xz6Z;Gq@fwby~7iv$`H>pB4ul$ZnSo z&WTBm7~1S11}AJ`38EF(kWN(EF@W}}g<3}o5IS)PZ2OGBvwQ}2r|BAYVE()FTPqC*e?X`7!oTtrfWZYv^im{Wtrb9(OUF& z&om8ww0O6inMpIU7J4uRGp^Z%oo*(4Bv45M$((7RY5Am+v`pL(2N2pS--yaR3M+GX zr$>g)zbL5@JavyycfSTq2^!^DT0Re`1~ooi!GS{e&T7fg#^0{o*E3)c3R}Jr2$))? za{i67ry0;w`pLDaeWvh%YDc43CKPj%h>)d7cT8~@;RuvdqSEwVDyVR5J!m}bh{|-; zZJu)wZgqM5lsOt#hTrMp-XcLiHmb>O@1*C;D#;Gh@T*l9I?l>Bqcb(}JY%%@?7^E$ z9W3*rzGixmf3JxhRnwWee%`y7_GWA{_YDF-cRKZ{ON((cnD1tSN}Ffh_yXr^8ZcUz zhh%-fa|?3v)$<8I#EcMd?ph^-30yeYSZfUa7-)-UnfLy2miU(vtcWdi@AB1KLTS}g zAa(lEFm-H`ZpRXbu&5l5UMF*s6H;m=E_o3md2hMOs-%s1@224_C1qI2F~*%JVB*sR zajY|NT+U?*qpFH}>?8d}l|gSWm&CQ34F?q0ge?*rDW77N?>*T#gWEDlRK1T;Dnbt| z%dpS4L79 zVTKV4cL-RD;c`5Io;FUi63(IwWgw zB4jjQ62+)7o7eKc$a_Y;0pOL1vhcfI=94aQ+Kp=NpFkRcBUUh2c0w&>54U0PSTPb< zs3z<${+iVJC_j!3{<1z*IGy?P!~t%l6p#ojDxgkxl|Wl&#lGQmRrarLaf2SG*m7d) zpjpUN9C-NMOF#w;!lSZPeup+XoS`r$T_ZBGFhh>`%v`?;|ALyiDrx?q zW)%mBpzEF(tOh;23<9u@lNQ=!1ktKG!fIyEm5aq)&eDUPuIWg7%cOD0e~{>~6Jo$K zG!RTM+?`VRU2V~#Dp_%tg?%%2oMSjuHvjvZ_R!szy_lR5^273@^h{c=cev9xRPLM^ z%7qojxeruYF)SkdzJm_`$dY%jA>t9LP^c+Lu=C+IbBuAQ3=@TI(7nsJMmdnkvI}j2 zIc`$Zc2`d|D36P$srZL?P~I41qv`4AWCaL2Ys+5V;(KcwrM|&7;m@}rGb=bBtw>O z7|-rzo`P?=CjYY8DQ2RoQ8f8`tD%jwnK`*3s_tS_;%cBgqxqceCfsRv4Ha9^jp}|D z9{cRc4Ow}G!-97aBqhe7k_GAHS%H)nMitnPYS!e8*@Z~~<&gF_#&(~+o%Ks|2J?WfB{?nlnRZ?THpck^G zO;EHFb1LnDHOW?T^hIeDegpPp97(#g&%kSD;fU@tlzF`F^QTHKEkKW(;PTyFN`>7gjJ*s&<-I_%lqcpl=1JUG z$!J&T5P`i#-{AL-8G9G9IV}&=vkGg=G=F}Y4S+qpoA9xkPI6;W1QVgghZ<~%Q1sX7 zIv$y|q1&V09`!#dD;x&dl&~OfS&Z^lH~HN`!-KMMBIr{k(~4PBIgONfQv9K`FD5eZ ze1A`((7)k$GHTuwQqUc09qf!42XwqwfW_Loiwp+wFFFY=XB2~fw*WMuuI*Ip5K6bKIp?U4cvV`c5LecqR?WrE zB^Y(9wEtRn*7(I@^WQt z2};3?PasyOU2P=2c$E%mO>o|p3^?Pv(fkfpKuBqAf8iE`qjXKFVqeKdp?1S^L_52> z`z+KWD6ZfHWx)QTiU!nGkgeOyc@;-{ZSzI1{;9XBipiO+m* zD)b$8`~;i)XMXIwG-=Ci+FzgR(t32QRY6R$UhgU7{%2b4D;fHmp z->a^BUId29m8^pM{Ol=6beO@j^R%&iav|<~vrBtO zaKJa z@3YzahM=g2EI(7r=SGCZYjIJbXQ^n7oNg6eiH-hYp@A0+3?oR(+W{90^mJx4vHBKN zxOf*RaaKL<1-wL~qhgk+1?E}kogYbEpZ+5GQf6Aa-h{AEid0Kf7r6O!;u?UkgNdIM zaF>+rCphlZ!$E?r_@-Et*V?=4#IT*He%q=H&ix~M)a=h7+dL&yT2{jQ@Y zE#=R8Sa)aVgn8tax##EUvOjixgW7XgOCaj2A8hn)n8-jD6E0v-gk%w5RDlAGh zi;6*}aUAzfdFxP~Vvea`m10b~o-(%ztk;k_r87KViy_tXqE#0MOs-`O&Id9{!UG{k z@8y(vE$`7ShUgig=sCL27PJnaDm`o{A^zYTJS&0M%^5RiXQG_VeuPurF^$eYB*Q$@vJ~Aa+DJFP?YOM z)l%%&Y3e@%4q1u=8+gXsX=yR!@@kzJ+`(GSxo>&{%K3i^X26RuQl0A;pH!M{cNE*1bv8Tqtx->6Ej-6fhDn|4j8jR})I>6(U zOOE}n!+b_;Qp*mi6h@rh`zb(2`;>vk2voEPLj4ga{d%ZnLdr(xJheMGZ23MydQPcZ zHK*( zTNw#MPh(`=q+x0M{s{m0KzbFbMg(?7iD}O9l;>1Af6C{K1n@S|0;r1=5supySdzsp z-Rvd8gH?F{h02VJWmwd?!kVcJWe-{GL<1SFcEbZLWXaV8p6%zs`GOX|jX`4%O$ybj zbT$IfgtSNTZy&SC;Y=mC5_I*83O{Vy)q>|uRT#zPyggqyPo~FdN%P-CNO^v)-R}%~ z@Kdc^Ak9g?roXbq%XOS3S^b;1`0UQufK^sK)LdIW-UOYEG@@riJ401<|W&Kt-bU;J&K{LqlKn?2mO-YRZL442_0rUHsYr!lLNUk7Q_RzZxK1?XM%waQqT zqw8QM5WMsvDD_UbN)M?to#*t3ah~zQV>MuP^8T}RdObQ zkIykZ48V1}OQ%&QX9vI|f2AcL4u%rexM!d}hbxmqzyhzb0uro(l=54*CMdDgnd8>i zOfod@_t-_i7o271e(E{^@k=;b!D=wZ1l(b~@5D170s082(PQ}J*U=yhU%u>b`M==% z#h)6i0t}mtcDxPzZyv_-Oi{|g7^k`S*62VhxhIHPn0bqW$v*OiEAfJw^#FukxAX;r zD`#g4GXYB7SrGApNNE4w`?6U4s*Y3@+FeFx#suWV+6BG^9o)J}WdCN0z7I*)uHi9IQ(+c}5L4)Di6*5Tf=@b`K!BBYr* zzo;-s${jg9H-76%yJj#z;CyPMWMdUeJPIl5dk@-UGpotjh8XOe% z>T>mSk^Y+kA4V+PvZK8@MCW-kTBh#0%0EFRT_zZ27dY4?ih4^674Z#C86r|gF$u~g zRsFVSvF++j;wk*25+T-}xyq7S%j9$}lv&_I>-UsWG>Ektzc7XBPNxIy*L|`gibHBAY65EEv9j^H*iD_ z6wncK&Ucse-4ug~=#@d@*`?)zwuzgLe9V}Xau`MUdJC#<47X1USaF4;ts(@WY*CQY zJ<9)0aL2Rke3B@GGL%dmPolkb^S+Mfxs)UgI1R7j)c#C|2|R=3CMeM+IIe3N8D7|{ z^4brNGmhIaF9uyrx$_Ras6z(d2vRcDp7-gwexDN>D(p8~LsS&fiG9ggZ4|H}V6nH#?W1f@Tf{{sxl3yL z8=4ZkunN|v)1LadU&E~I6HUt28m`F7DS{}L!h$zICed9P=3}7nchw_6@RWF?AR-7a zOc7ygehi$FUX5kenyt{<0hqar)cCYzO@Fe8DL{oX zKP>9~&b&km_4HnMD`v$-cLrCof3Btwg70QJ^h{>T`>9gdSgIFWUT1ABE%0i_JUhh$ zC&fV81sQ8Ey@q`*ObP08=15GUrx2c1&-zL};>WW~r^ z(G{kg)SzYYx55s*JLyY=A7F#?d5>0Qg|8D8^tW=%fLDS#on*WT>7}Q|F)S%K&!@+B zEp9Z7cNH)UgLr`VjI+wafblEd-@8MOf`~!fqJRk-zKu!6Fmwqa`E?&gMCo`%2Xfn< z>M!Lm^Yw7CQ6r-{>wd$)Z+qop-&r~qT;QQRCQ;9@@9xgCsDOFcQB5#kto#w@w}?C7 zlF=x%?4|a2?daT|bAL-7k$7gpw}kr;I)<8r$CCQ+dROD+R1{;5H3MCLd-q9Fbun8m(e zmU+FZiPA_rl#7OD-sjss2rtO`!M%aTx#3f0B0T>7rV=F0+txg}!uB!}oUqx2OzsQ6 zwEwELGa5eO9<`=DG%pf!xnA6q%B1r0;&@n2k-Pt@|+cC3=f+6<}GUGuJZ@N_X1 zOpXe@*6pP&%3mO0NRsy&Lp@CgyG0KJ?s}zN#P-V~7IH-s|G;5+sXwzN8D~?U7?@(0 z^iN<9P0P}=2GdX_gd1EZ&I5UsS_kx@$eEak_6W>y&zc?_G;ddY8NSfSAD&&+E+U0r zh~mK!5mW~%sDjN{+IWb0u`OPnGeUGj)N4zcvj`Ifcc(d zt=)KTHLnAv@6ON%NBe0E&efvDb7BWx?nhB%P(UX9+9KW_}ZzQ8u=IT<~6Kqm0Yf;bII)M*7_X;zEPu*e87F?oO6zp&3h# zqCaBb;aE$;QJOD&t*Ri!S_Zh}Ugdn)$$bO4VN>fwrSvIm!OI8iC=hRk0wL zHy{<`9BwM*i@_Ipj zg&u=1zudX|)F`aybhOi#YzhJO~-w(*}?_xzK$#1p%j*D!)6 zIhwa7oiVFCHFZ!#BPj=06ml{>C7xw;aEg~GayG7NGr~`cknroK$qus;EJ&ZZrfAr} zEUls)Y57i`As#KzR4ICsX-tuq|922)R+&Ys0kMeDEmvUXV^?!tN^5E&_BqsRn^PRQ zG`wlpJx?}%&D>euCTUr3JTd}FSM;j}*(*e@8^+-4nn$1{hht?N5rQ1XEfs}o>f1_F z>u<5rd2|erMp=wZ^lp~QQdB#=FUf)R0i$0$o0D2G!PYWy2;qHcEFdh{g}f=qy7K^;v0un!miI7`z&dsZUMrbXv>$#dCf2#IlsGk> z(uUMWe$ebkb_#ZkhuxItpNpvjBYNx8^c^F?OZ{~3sbA}J+)(DBz^7Mg2V(NsOJRu;t&Ov%oL5YOQIO6$0luAr;937`BQ(B?n?lUiVrtNZ`|wr=_?q zCRJ?%d3vH$Y*%=E7hA)Wi5Nqfpn{aBnL+L%%2%rMWlRJ0y7RKnKG`6peMvIgNH%zz z?UikVqsyaJ`;KTFaZrViHm;y0y5Dl+mWLhk)`qiP))JnyR`j zMWn)^pUtzhXXO3=^o6fK!*TlB+CD}^DX4zsx64Wvduqr;KapBq(eI)cdwz)Q>v{)m z1IOt^i+5SOs8v#;11Up2IRa95U{z_)#p>KlsSP?Y%vi^ELroGg_URD#nBeR+7EqX_Dn4adW#JK6n*<;2Na6=WI7UlPrK!cKY~x18e(uSE6yI;F+mL!~^xA1qq~0CmXt36h#j9 z6=NhVvHVbOKZV9F^J>aru?} z-mb{P@3Vszv^Hi1+h9fv;p&6V^HL8T&oefv63aeE1i#h3lT8N{ovxiEg;gQ%)S=+% zFmU(QK*xKn4;Dt+q(Vn305OwCjQjEToy}~r10uaQ`gFfcTr3=UpZ!gu@>F3x@sW>;}tk#;6Iqbxibip%nTT8w-l10c5jGl z4tN0JyL>f#<$S;`%s*SylKy+C-pS2r>)n|x9?es{BEsR}o~f|)5tPo5!+aOh;gzb? zDa@>3c9m_?Dl{$eC^D$dI-*=xLFCms!7EKXs?`RWdtzPBi@b9e$StKBF0-ℑF;n zX7Q58zdh@~_M5`>)0AV*MVzqn*E>0*;`84{ZR|QZ^6|bL*>zw^T#g3T=u|{r_k|e= zLq!^BGj+zm0dG)gW^ukS^xV=dw4m&W|5md5JF4VrQSr?tuQ=5Sshrg(E=U`J{T;gs z4<($_r8u-$6;U@A3iTSl!p7FuY@8xk{sJ{S-X(w5kwb9OsC>zyQT2dFu64=;O^v+`+iTV85off&k1Ow4F(TF*y>B7^|h3`?pP5of0igFD7Z!sV!C#?TAu&PgfHp z-x**s=lMp^meq|5(L53LMU$J;XjGNbMsXb@9|-J|B;NC{Fwgv7lC7MP18-igai4`- zB)z|45tOiqDAXqe_gv-Bp8^$8dq1fMWyjE;ul0U2X?N;Nj;+5)V9)vbG}jR_?Py_k zs4Hj+q!XhCMU5Tj)n_(Gy(~nq&&zrva@!@=+_ZD|zeuOm*gKbRLuzWJ#*!PCW8u6j zFflrM-O#Lo-*x;P$bVoa+&|;)U-brLSi1+%uZ)?QJ7R!TiQWT0%zTMQJ$2@|z+0MK^ z6JGh;8y)Vd^r)(O7N-RP-P>t_c-(+eO9&+C@%`O4xRf8L^!2$@CRor}uGP{5n7pOw z_+QVmYJQcYlh!F<eD@!f2#LVT^E~P0 z))Cxpyl@W2&0Wx9IcCVSGJZ4yu4PmEvl24VeUBY5a%k6AxFyGtDh>zT6g$9hZmskq`>w7*7<6>3u$<<943UoFyJ#+#1qNToD4A4C@p+@Jp^cEvJ`cP( zb|!5UKCv=h52*;}-`CCy*esLQ^h=^hvrwO2HDHxzq!|=GSh3WKMIp^b)F7q!rPlVP zLKvVu*?`2VpMi?qgIZJAa0fKVC1D(#_s79s-23vKygEPd#Jq->e<1aJGVlo4OfLf7 zL#f_xyt98jgLE+Y;NR#n`Er{7UWgcu`qsfb8KP)QUprnrliOeGF@h-_QFf7`J9&rg zL#zJX5pT)FL0KC6kxY>otSP+;zS@wdTJc}-Zquc?#qUfnA?>!CM}|zQH*3v}p78Zl z(;KJJB)W`pjxxXu%><@hhpA(|p}W7QEGJ|BiAIwU{m6Io0TchUz8g?Z;R3u`qaj!9 zaXvvmv?DIQX28$SG@7NdIL>VFnxWaq_Vl zq7O8#%v6f2x%nybd-uP;a`tW?iRK+f5F!m`t7@1O=7)`L%PauF-lIkQxt&oM;*42h zpJxm?x#m;!HP?7apBFR5g9$1741yxaS=@p40-6{45+)m9r5Mqh(KN2c1igzXsxZ8? z)D#w@>sCyI2fbp*dcU2gKDjsVRdE4a+f@v2k zs+CAhiM9>OVHJ7yg^vuV;c;9QaZD)qf{Iek0~UGVc<^~62;pAZ7^pdrsw)YQX zJZ<6p@l9}&-fz(w3zLDDEJ~$4!9_1edyJeTME0@&z~qj`=h>UpJN6_y=HbwLf)(J2 zAABrD^llowZY}W5bBvIZr}{y5{+_5$-~}<%G;%{>o+Od*dcN>{!6+6{;&*Sz7e3)_ zhK0T`L2X|QGq$T0qY<_CuoGM=M^mrn3c}VKP!3*1GAKlv`UGdTjfj&I;2jm}#OrQG z7yezb;=g<7VBq4Y?kSCc(b40NqVRT+tZ2*2qlBCZ|yqf}|A=SRD#e@|ZI>9@)xL%)mg{Yf*9P!Jl3>!xgA- zu2lQfEP!Z3)kk&0%`yCzXaXhhEv^R3A&g*BU=_!dItF>`uZG%j%0TICHKvWvi4-~K zp1ig9s9%PBek)e{4m*E?wck<2h5PDtTB&!2m;S%w$1lEQV@$3&EQevzlyX#PF?%zw zk|OUJ1&n)50h2W6v_Ys!P(u8{DP6-a2*dqw2|}nk{JH0T2G)M>B%`^Pls91RE9JTT4!^WFHfW!2K8*;3-ywOd4Z&6gxfK#cf^yS3>Ax z&pX;_e8r#sZ7^h$3lKDfvbnSrYawZrzMCBhgN4kSLbP(DFOH1~wJl)P%YX?AQvgX_L(D#B83(O|u2TF^~&HZRuP zeYGxZrW`l1B*%9&vW(fV10(HBdfp?CQ++KWgQ0ZHg5L~Sp6i8R4OI;|n7M~APhSPr zSB~T`$)JP~ZJia^=?$1Cw?P^nE|sUJkN|Y33FD>3I~8Km0q5VEvY3OU%qW}uRS>IZ z7&U$raE**~Vt`PHya{VYC7YVql{NfRiEz~-nq2hhfBk<=l7GG6YN76`B`El`j28*8IEJ7j?sII7n9)54LPtH!5}NCv^U|8y`$WSY z1340#oLI)N&{LbO1Jh+G_n}BPd~T*HdZ+lS^1VB95r-(8UI}!F#-8x2%!XE!^wq-- zey*LQt~w0`Sw~}#k}c}r;R&JQIIQ3^k<#ogeF7CQA2W_g1$aU}&zjAhxf3L{>qIhG z`2ELyRf?bNP&thYJa#k>Z`lW^u{MXoaDPJrZFSbAd}`W%CgL1kOpH zJgakp-|gv%Q|x5moMA$)1xzH#@k{5f#!H6{FQ4Qj_!`~aEDOiFye>u%XCKkXoA!FL zD5%olgd(JoXy;kYNivHyng~tNdjCeT@!iN`p2!gYF>S;?R zw7o|zTmMv|Jcx}zChY;_5^N8rYeAxlB3XPZlPA7`i27uzm;qTVZyFL8X3J3BxD7pF zKRQLu$8&Bck2@zgK;oTZ)vQ}pDeO_+=lIAPFs|UeuOe&@5K@=r%zXbvDuWABE<&l% zFWDKX;0{;=amZIJAuX8f5|K!ZDU#U1}-}D0T_T zdMf7QdZZowr7PlsyD|;jOn%=al}wiwf38mX#cE<$M1qbvt?s3}%Rr?v6FLCn+Z>#E zs)WGW^~PE9z%WbjTN5jIr5M69MpvC;D&+ZVVk+nEf?L8S+vjYV5KOlx!(0(Ms5b69 z+PJjGRQ#MN2(`Q$hB$SRfP>QPkjsqnt@9)?sLfBIH0$m4^*2Iz-fjKwy|kM9?n#02 zSB=szCmKb;D8Kv=?wKr^Sqlo34{=^aZca67u!6@#F^H;Gdz@3x2n47j&5muSDuy5! z30%dQ>X{xM{G>E z$^;x>Rv!05$v^`sH~z{iiwlw%U1C_V5U?&aLT_|jMA$^e(%myr=;#$i( zm7Oh=f30@Uv~YJfih{8-X!BPD-I5qq;nCuL!;EnY_c(Hz_H-ID(6<*f1)B1%ba@!F zwFdXwtGeOP2+r5fM;u6eB*s8?)d!|D+kP3V0ACb>=T(oQ@Z@E z8J@1K{D+YS7}X3HX3cQ9<6|fqMb?_in?bR@D8ycW!Xm3^Msse>e^-{e9W7bdFRqY+ z_-POSUH*Q5rTpkbg$>M<@u;b8>ZNQ`>q3|1uS0cc3?2T_FLc1*p@|XG$h3(JMLGLn z-${;$XI;cYaQjFQGf$1u;d7~616Ld|;ijZqw&JBojJCfDm-1gy^2z>`d6bwsx91s( z78fsYu)m1Q2WYSI1V*J&3Ect8`9*Ypa@8HJ{*0iJrW3Z6K7H>-pcp z%>g!wh4rpt08ek8QK0ijDrnVq(GHHHzZ)>{)~Q*99PFB_Iba5U{YI(Acm+xw`+2~L z*dLRQzT%j*$S1D7GT4l&8^>d;z9QnS*0j9}VVD5?D@(wKmiJV|6{D*AHO{66kFZ5NY z=Y~0UA7niwm!S4~8wp&b$UN0XpaSYp|KwMeju9=&W|*7^8(x8QfT>NBUkLu=DhjCB z#+V3=WSj%1n3y#yg&)z({#L1j%33VW-m;CYrJrOPzyhU_Ej6$Bo>Gn?TU~Hgro3Qq zQ=Rg4LM$+;K(+to9yiD5S@Zwn)dglV`IV)1LIuwpAUriO@LfWWqeE9?$@!9u`N^ti z%AP@NP%K7)-h0~JfLbzb!9#FYm&k&{V;$0CX|HTTF6ci2z;DJ9~ZSK0lZ!&r(lS4=moSzvC z?|+&iI$nioyBFs7_Q`ByB0%-|dT;dsufA}=R3Kwc9TL<+g!CfWu+*kk9CocpM)vy3 z;$g#aCMfr#H}|@{Qb|Vy^Mqocb#!CMXD`K!t@xvqEpgt%g_BB{Vbk(Q`-^ZNSJt1O z@gHmRdTnwQVGTWu!)>Ulw4a$KvsNrPX}qV=>4~j$`LxgT82eU$*ICQ3C?FkqoM%>I z91^XFW=_{340YFjD0(wY4M=RJg0}6{EDqS1)kkA+BZn;c7a|VwoI+JLoOhc(9$+D} zkaUuV_)0@MYgo{l_OW(IQY%TA%22#sX-~bsDySyF{t!C-2lgWt`y@Q&Rf*moR)WeS2zYE2-gkfD%D`q7?07hQkx zz*8@xRe(A8k+#^@3inkoH$#>lquayGi`%WXNPHaYL?slykEZ(B_32*DSUefw1p?N# z=71?YM+6cC8kt6atzEB36mtva@3n8f_YvGCh0q!7KNxS0fha{nA}^=k6q5Tsx3w$U zyJ>l|KV9!h)rLEi{9jyZvYt}8lkk|`fc5(KOeU=JNY5gQxpK6_)MYM{a@42p`ID;; z0?ra(U0WKg*nwjlVMoe9!+`>2t6uTx_g6cXM7j-(o|Ylod2HJQOLg=C{yh0M9yQK! z)+y0#?EGo9+}QR@99_&KYR}Cd9n`RA{vKI+$UAnku<;1Tq_>8}x{(NK`Dv&U$+A{T zVWap{uy@oXd|Uc~7f$xbU$^2!0Oc)I=Kjl9;n++(2)}B06mF zX<5oxDKu)ejYaWI{vK$BV9XVP4Pr1u27i0mO$@mmi&1DO8@VkJQbYxwmXr{`C0lU68O}B^LNAC& zMZXT?G#wS7;DS=~r*~0z)MAO6NNk;AlfYV4AuLf+qboBu!IxxN$Oxn?@K4-b6%f}b zdG)zv8Biuy^~87soQBD1J(W@LM8Z^AQtMU<^8g?&FF>MZ%Q2PbU&L98ETy&AI!VQ+ zHja=4MjyJg5Lu9+BUlYn?SHA>O)@wS?Lhvxp(Z|KnDAOo$Cxba}npYR!VHr<+6pNCqXd zdgm2az{diq*=CP$oI@2I;Scf36wyeDyYatPQHHjRoid5u39?@wDFC`OQNh_eQ#=c? z(!Y61`2N%CVS^Hch+a%|M&~h-;O6f>CiK|nqBlnhpHm9wz5Rs@3L28b1 zb`S7X7PM?~n=L4@lkF*fTVhxEF6e?OQC{bLg2klZp>lWLal^HvmG6pc1kqtkvt&-cfO&kkEM(MQI8eQt)^>fJ5CO!QQ@C z35-Pq&D=&t_nka~brK)LGE7e^Z5 zAb&lm1s|frBCsxlsUiq^LX9v$_WFpR8n2ldNYV#w7YcOoK+n#$V6^PhN1+jP(-iS) zM9^$>DTi9_6|rjHPJ>lqKa(CVeEA}m0us8EWdXY93k=q}&vUg@J zob=Of84j^=dD=Y^ION)a3e1l7-#`y#6!vcH39YvRbp8ME@uMDW6M(Q24j26nPWb1Q zGXuT0hYd<_n3@c%*T9=EnpG@ZT}uO?be>3TQaA`RP{Vl4Swhka3pzWu+%#rWU&Ph> zh=&lVM|e`h|Bmi-h}7?!aSy7s5;y&&!%8;PecZQEckTnv7xZo+-((=lhoJ2gDI z4;o}#RWTE-gt+qyG-v6QDi+Ar@8m$0J$+buI|F?M7!yQ$SQCtD?v-ZOK=;y5tcKJ*lFIXazU7QR|2*Zcft3YElOwB zxO47P@}n%c7jE-;^|2UBGfn?W2B8HS&);DOOCe`AuO~0jGo}E0!D~KcVFUSccn=du zFyV@YUTg|O)=?pUnKaP~3a*bl9SLGF?Hn**vP4iO2Lo-mY-^be6E=6X> zck5~rGs?an5IQ9EEK48f>8BWi(>eNhdD6?+u40-<^bj*#f;O%-{71wQvMu4D{DOY` zT`{<(j5_U=z4Dbe{4xhYmG;yC&u%ew-apWgbUXA)3W<@#ZPR#I7om?(MvUGM5y^1g zlhyuLSKuX&G(WX_fknH!BGD60mDUxm9L24FWNzE(qpHj+sg$sG$^T#tmN>qx(-j$; z5fu>Nf(PkM0N)a99Ut+jnu_+xCHcZeR9_KMUF|(JbmcDCWW9%Cz))25{Q?xW_~PIL z#15DM>Q{{CqFK^X!DV+7CeL<~^Wbm#)Pa;>v6OEi)H<6^J3YPdzHOIp3NNLID`N)D z<&8prHH1m|%PXSnG3iX_-B>~GyWk_5GAjMWX9dC4>hwfPf@x}_KcADqrvyXV%` z8YAT=Dxud#7}f;LGE9R##;xIvQZxqZlm|^aJXZYpIu0Cn-GqGGwn>p4YH|;{5@YfE z-fnh42-_Vu1nqpehFasw3;4)6)xn-Qz$T!|k{jH16&xLHhq1)pCNWUV=WqknDL_Vh z#Z^#X;Fyp@c-mr!uw+Y~!_W%x*^10J7Z zBX~I-HdDP!H2KjbY298^u<@qsfOH(%M!yw;9jqT%jjzF z0y8{uf;dVYd5d7F_MDZq%GRy{5>{sKV(?)JHo?42dl1+ZE*y-st`7g18>?^>C>mkM^!V{x3P z+{1GanwVZ6W)8;lyf>%r%y>VjiVX0(^M4B$_S{qr=>61mFU7eXz6(h0bENm>$hdZzR?G;RGZrNHlE!^or^ zblXvPQW&SjVvQi)Y#Zj(9~C_B>n~lIN@>-l$D|*g3EzrW3cf!R>ep(r{x-GcU!WUa z)i#CEO6r20JGc8hv9NVhQt)gW@H{GHxMqN{J20SZy_2pW{eDyiEG-YLANBDqlHL`I zqZFGlds>5A!v&(IbEkAyswdCN!pPQ^vAJd{(OXw@=CEkd=1V%cQGTm)!06QOhw1DU zD&!cIlDHy>+g!AN7vd(UyClszLCdr9^7up1;(P@o&c3!#pZ*UHoU=~h1ZoLKLMj4i z1jmNWlSy*lV)mE)>S{p)?m!coda@kj1$MEfp7F2P+F<)I+;6iM?xjvekV6}NELIYY z;m?6{>ZOSzggcB@%oh5IM*Abj4=S6TWN##!9HP>PyWHmH-&_-Y0RY zOwW2;r(b_bD#8b`2#40EwY@t8g1GEVaZF(G{k%V)GGx{MOO#bWyd>!%$$L;{W3k4T zdf?>o=@=9wOSUf;wF!el>))Te!gEAq{|bP7QV=XrV_C-g^Z3aslL5M{^uip0xB8&^ z_9(*2=~>t)oD8IrjORMuVi1NeIE_H6@m;3vwi^>hpaPf5H)-_j=|jWeRC z4=k@YBPbHez2zWz#FJ?gMmp0!6$P0!`x6oNXNY{Ju@`I&y7#c{*M$=jX)j1s!|v?m zpEN2FF|V!EiC%OSRok+uhgRy~Gkp@8k=62u85urMRnz zersV>7pbgJo@_Mu$%ym*yYm)Ci97MxGdZ(q+(#Y5=P^-5eL3VRUhEc7%TcqL6_4jO ztNcEhB{dgcc+5DD`@i`|Npyax1SNq^AAySz+SYdrE(+UW!!=@s9G6an7Ogv^azvz1 z!QvL9r=(WQm!Ae4uDVJMM7*FRu3UihOa}|i_8!{boyYE9W0Z|&k|hE8Pk%JfD(U`v z;m$`=ir0RXxveo7m|w~SZQ*91@IHu#bbPAXRsSSSCSI!y=~Z!6{Q2POP-XG;f}c7N zV=rJ?m;YR-L7>VYb^h6lg8Szsad@HunTuySuhL=)=}+z{vPQ|?XVpBe6YXECiTEdv z@46{U>t?7)4gSe z8_xwFYx6UwU)1x(#H_vOkr!-Qpun>CKD+xmaL8;J7U5JZ%&J1 zSAYF2KCvtd(>yOAY(XJ`oFaR-5Bljy%XsrN?X`<0R%+{VUTRlJ?;_(_4*MqzY39@Z z7-9#OEY7!5A-VVtQPPDR3myVdZe(AZHIs|$K=f1xkvDM>Sn>e!{s=Sv&l9gMeje;! zyYH+cYs6yD?}q1ecteb-dH*&AH9~W#SifH;JN)_-)V6g*{g!juL<9eAsydlnusaTh zOc}RU$BC-(TpV)_k>kw_0PpV0vZrp!c$C|3y@2y1qEJ|g-F$s_gKoBA)-%5mUs8GG zx1dI?0c+M5_3-9dE6AM#qe6ALz#7Pn<1Dk9?Zab6wt$hxS!>Y3&QR3D=N!&I^1`}f zym=QB1{7)rUqfU_MU7S%9#q+U`$-+sLe-e~)oV2suY)N4mT~Pm5@3iwX^Q6IKJU!- z!fsMIyGj>CRJ@-z{F(u%+nGjQLi`zGWBQX`lgcy1{k;})|2rZ_a@8V^PoZAl-GL(v zor!m2y)G>zURxLC>`fy%kdYsLa_JIFMV$158(BBuimlNvR4k2rP3|wC>li+ttSn=g zF!p}PfgNk_-EZu0cW$$`4{Rk5dFu`oHoXLlYJieF<;ZV8kHE)Nip>A*963C3eGd0sf(i z#EwsM=uQgW{^8abut76WP)rr=eCtZU(!x#{KIi<|EQJfoL7>0o9MIP5PeI;f^I%~k zW=qvjJb!QP_t$BuvK=Wn!F%G(=n1#HF4Ac@>N)} zX>5jS8MlgD!&%U8xmBSe;&kBhl5-CIEN-1)zOpxijV(OlGEZ2s>BvH3jsxL<(QF`}$7JEaZF{%Te7`}06B z_irVAEO>P@hiKrkhLWH|2y=eT>&mCs+CnY86eUJ|&383s33vun7LfSou7yJ{U(I)` z$h~7J3E5c3k2WY#d~&HIJ-cJY-k)Hv_S?~U_g#KxCAdnt@8+En;r6?oxT9j0u6Uf* z$l{*Og@^&^o9U{?bm`qS{wYrxcuEsfXN43AuVLxd%e!ykWq9{tnz4Z-uf0RBD^jD( zHApo!zs>a#sej?h^kBx?4pAFTvGJ!heDB~(iYYDjHQP3fJg==}#Pz4HpQ-yVYs^c^ zF)^yNsn)^>v03WP$AJ)A`xWUwjjcgU9_8fTE@C{rY46yG8tPtap9H-rjeq;{cwL^0 z#jMU1@SVefVq`h7XD&31{ANliO(ktEd8})}02BCS^ZCko32k6t4<)XUrT049$=j7* zT89qVCHj#+adOq|LWTvHcE62c{tk5x;9TLR=@W^UP#nM-vr$_g#mDwF$lEaNDQhd~ z-Y{3_X}FtCW4@7#jz2f5ohXfG(09!wCY4rsof+M+e^4kNwjiVu?>xC=qy2`e)gdh? zWVNJ`NigC$W=T;HQ}A~X#h`NXfp$A2+g&>%nc*{q;DRwnpH{`o3?cGiUIt;Vl*Sow zr?#8x0eOsVKm5(n=MP$2FZV0{yli>ni9a|g_=z`Eg1eDf!A%>^>>L?z=3PcZO({MrOGqy9Mah`uRG#2`|PK~gZ{S zH|%n5AM2ip(VfI8;C4#@G9pd-c0RihSEyuAw90PXFgOFTlqxdf7^T*K5DzHX422{B zt#ic=nEuHnrD`UM7u*Z4VD@O3ab#qC8Mr@Xf7%`Mo+lw?9V^ojnJLZpFL-2xspn}G zAoP8Fj)L&#mmmE$L3p$0HK^{+N@{%TV{(iToxK7S!yynA*dCY6$(J*Vn`D@Vi`m%b zZJW}O)9Uh$AM(XMU15}1A|*5BF0X}rqNC4qPU@|bt3~~xtU7z~OMU=>(2LfE*#o`<_YW*)(Z!}!=LUYI zsURtFaFNPu zkM0SVT;mq}_A9lSfuAyg4p=JBiH*;jj{j$nymQPm_=@$6>CZB@>u!hL_B zR1zUjZQwi3P2OK0&ZD6<@R@n($)?Ug$>sxZlAvQJ7e*_>$*H)~T+|01hEHl)VFYtPMo*MGm;{UXa`=~6JAs3m5gNTG{3bZzQzS+wF~NlPfi47riPu10V7 zPbv8B^R;7BKt-1kJdpZ1@M7gf#097xO2?=fahyK)?8t|Qa!)BT>(M;A!NK~A!H}YE zv1P@H=13e&L84an)Vi6(Soa`Vn&n?GKW;}v!Z8u6V5pBo-|rMo88{x+t(`F4RI1aD zYc0f%E0=MK={Lo&qU0gmsZyA}m?$a?C$0hq>geHJXG0&Vvb96*&01Jrz{S5-J|?P2 zef2Uj1PI@XV7_tvbQ{u<5suTq5|cotZ}aj!H@x!v4Bm{VUXi(kAZBvcmIG*)A9bcit<=VfD z_Pr{XQa+UB16Ag5`&5nF>ou9Sn5da-N##b808Z4ERDi20IQzg&xFJFkQ-`=}A|Iw+ z659lE4e-h#cS%g8_VNY?8iMUoYrYefs*KjupWqM(K&>;Tywxo6xz6($Ardv~yIz#MlrYc>Ch?&S-`;(du2Sc^$HHL6jI-QTm8ufHS*pmmp*H@? zqJW#MYn#kNR|ek`@<3f=gW#%|m)nla)r(QX^~1#`s`y5__B!H-iIQx+peuRq;J!5% zFEHuX_cs}C?=O-f$VJf?dzEYKzuVw9@vuCQ`239PP*|@Q6)7B3i9Ti-6^PjnYD*o zD)^aSv^SAWc{{s5DL>#W>+Fxi@XtBv{%BzE`!@=5zDA9WfkD9bSul=BhK7K`Yi_|L zu3N_MWVv&Ll|9QPeYumh!RFm^gRn}XWrO`d965QoNqxcN?Z*C;B({9s@IliOC7bF*Qb2Z4WpH8b3vrHQsExeh+gOr?Pl@G&p#2ZzPW02Bf+h--z2#J{G8M zL;gmjA;#NT=`B#J;jX1_o=9s?lRy`hi_DItSaDt%VUE{V-yU{{c84jVh9%gr1rHNNa0%V2#romEd!r# zF*NPXEtMa~5^V^%*PT#ZUYunK3z4TkQ}zQ8y43i@sAG-!$kJHrr547Eo2+@Q#$>DR zg|2b@VR70yB^Aok^1Do`sLo09Xsz5Sq;X35TO1ird%8S#Ug(*U88qPbO3TW z1%uQv=IE8qU*@i2EByr#eHGm>R0?P1x;nbFl!+-*m;U_@T=y+^nxQnu>aR1AftBoN zd#FQPrgFgj-@}0*V;xr31Y?J8=Z2-YRvq?N=sDuc3lyBN*skN-5=UK#(4g^B5XWBA zQX=vqzJj5odeFwHg2KJias(9plf0ABdS&mXMungXpf%6&qi3CUFYVl?_BmFKH6P-= z(Y;Fa8~k7Wb3@}RKls)a-B_UA0hcG;M&4V(CH|csDm=uY(w(gjB_DQQhHg#njIZ2( zx8;!4=nOavj;?xnC}+X^MOga$p>KTsuPvhUxGyhW!W*wa{vg!HDg&d9-`am!^#y|0rH~Ih=n`9t7ZBW+qBa36H!k>*$9#2Kr1ccpzutfTS7bB?cGw7&&TY{ zmR4#XmE0<}fm^)xdqezs$KZC5KJtr}u__WVoA);VD1X7md^T`a*E#SH;N4*Q_Y!lD zgU*^4$U$!PxT+vI-dX=}&u{#@Q-?n^Cdx=Hd#^Zmiw(fqC7~Y~Go(pZeU#_LJX`1Z zjE`FhJFID5a@97`dDJ_6?QXBQQ>QdetI2r*Jh-h86Ejy(rGm2Z)t#zi(S~V;zi4^K z^!pZZaOaI$C?f~a_8hOmlyFk$MH(r@xo&xbHLxhJb-g~4|NA{V&{V^8I1rXJdVZ%621O(aqpW-=a% z;D6iBoGy+qTp<>cuw1gSS}d3*xb^I5{G(yHR^9{|ps^Tz;`BkEjxNWtOE5ihGKBvf zbzm{K{6XK`_sJpzHtYF}+@fQbWHJ~AV)aVAs>6NYc<^^rb-V{0`MIc67?}?R3%29% zmdp1_>_m<|=?E0DD8!dT_rb+l16C9?UjrG=dHDqh&%utvBSAgX(UW@M(^LYZi)SKh zvdo4IKV*kw|4kic)U+n?2nQ}BG*|y67#sWbZvJMyNWv$K9KM^PDeAIW`{oQdr@Ooo z&mQiJch>pgAHcIz2gL?=znHrRT%=%hX~Dwqzq(TA8$6y)=V7`fjlHjFzS)}A1&8kH zm;Uv)_h>ub-eFKJXiRUuj!iZewfZD+E1JwDewjJV5d`PN%6L0YGg1(68NkVttzPpn zk)|@od_HuzJk!(An4{+0?=_|LpN~Xpe^VWKPTAUnyg%5!#d7@pjdI%BsxoH!UBo{| zc{ZPr%#aSnE8OIx$R|!;g*qZ!cWUh4C(hQ%y1mno`q|tUagm45^>MmZy!TV4M@CP5 z#Wb5vN&Oc;$NdA!w$EFdmz0kt?Y9eN)7ToSB{P>lj8(WlYHQYzmRbeWu0gu&eOjGi zQ)S>y$;RIDsB;>q+DGrI=Wgc=*9ILt!#Y*D7e~7yrhq;t(qQ`$T3CQX6!h(gIcsnP z%9xSsi_XuahrLkW?x_W4+nt+$35g>ufen|2ja3Uer_AH}pniHYJHz7fhdr^*kl+H( z1Y&$Koj4ejV01tha-Ul|PRoans@09}oMh@8%u0ILZ{sX_h2xq~;5EKbEE$~R^zjT>;VZmK>S0TQMeDU+gtYtKWm<*_}F1B>qjL&|sM10V*bw~ZPv~RMM)mq|Q zM!nTtA2eklo8e2(=p~+xU6XTk0VSMW*qN+kST;gF4p%=1cruvle^ejNEE3*+-~TIL zyKX2fo!dOqXN=+Q=qDiqd%XJn(BN@V3=8v?$OR$XN8{pE@T(s34!T5olbha!-jz-N ze+^_ZqTR34tf3oU|Bjtp)njdzVj+)g<-k%|rR@xspb% z)rS9Q8Stm6MjoPjw}T%Wr1|~YU4}jQLrHVVNi>n1_M_R_y8hYRSEMmYFv?`i`j0&C zf1j2g;IUYj)5e@`WKl%IpZb&zRjC?uA@^A$1Kj?kW_U6n|F}r?1pXUtOBSd_)Hwgt z%Q(+5v-)iUnk?tOo2v2H!z!IMJhxsGesznZ*>dx5eaF!rckAf>gV`+m4+>pRdh4q{ z%~#6RPxhC5mnZw4At8N0KNkjy)*x7SmCZzOnADJcKAd;Mngsol`;sZglK zTtKYK)(@8n1?!FxS zTsefo;jD{7OL*H2lZ|I%@$GMZc~6S4%_F2m;oxnAKciM?g-F#<^_--=MRL=sGt@O& zMEjybsuPr*C~5mE=?`MXgWl#QEJX?-b3%n|Bj%u+;j-;=@ae;oInVu8=b#D5x#(>M z98c9kF|;Mi?501+?1vdpf~CFClqg4p4URu%*tZ!BiHZ7-K8zYsp` zEm#SAo)B~m{EA}R87|`dV(~Ph=+*sF%d5j%oNn)LGe8)=Ldiy^9%^Cm;Tu^m94EE{ zjsyN?dcgQ-E5xJCtAMb_#n}A^Dh6MubMcsYIrh#5Y}%;l_q62yzfAYK(K4#2Hy_w#_fR$Gd zh#*&hHN})uu!AYn1>la?5CrtYWcJWaatg+YxLvHO2CrfEI>GHRj~Fd@U;c-vqeg*S zr2g9j^$>%Pfq$$3pSR1u(!w_abRRj9f=j?*%D*ASYXK(0LVIt!h{lVy-fA9C_i;Fq z!g2D*a?IELj`JUGvxpj@;0ef9e!*nsfarV_$_5HIpU zo(&O6x?+M#|F!8@6$LL_u?f~s`2urar{2B(8|B;x2DLPx5PFh3F+a+F-v;TCJHJMw zrrg4rqqhi@v}ZlaQ{^f3I3TX= zsI(I z1nSE>sFLXRvR?AJ!@}TXh=2#9`TQ56u~w!#{zkPJIi+fKGb`R;Z!6#%RMx&$8dZ%cQj> zA$$JQ^`}Z+iVHwn8R7GtC>?YLm=kV7AH9#*1aiDSE;Xs%@z1J2j~ticDOHS{CLdb} zG4iGpVaN05-syi)i!IaOn?AkS|CBWyGg>r3O;HBJ`!Wy>$iI+l{4I~D#E=q!1vB7G=8+dmNs zkgKqdchey}(S6Ecv3}TVw*Zvfrm`nFWXzeFfbaKsa~$XGC!br(^}_deHKlsV99IV5 z9O5g*!PD?qKRQDc=ZnKU{m3Y!>$hS3N8Cfkm9nHsXGCPOhUo-6w&$o4smD48EQ!TO ze3;N$1UAz)AYBbeRa24o6@=ZA=(ok*f20e*J$KwI%L5T`XSW&W=G`?}*-pjZJ1t)6 zGr(pk0nnAX{Q~fFbL;#@#cmYxXQ(WQ7(x1@6RECkgSbBFt09GmT{h+;)jK;fRSJ;m zkkANmmI81gUtVy$CNHym$F(zQH*2!X954;ReMmXbdP(b2J^OfikBMtQ7k6}63Wn2o91WWEze*PE> z)miYhBu4CHR3Y`Ap`qh%Isaqj`FU(E&sd0`tDrNH_&>dEZ^ryR0 zr%H347O&7{O7?S`TqGPv)s+9o&q#QzN^bWmCKQmwjAXgdam&`KM7rV-BUtt;!7tLH zkfs~R{>U?zI4o439yhAM-O?7RRHTct*bpZ}ljP5JuCaOn*v2XP$0!=GSUjW>rDo3+ zS{#w)6ipoZPn!)whcGzEGq!JfzX-(mdBn(Owg`-kLmhaOgZGS%)R}RhI*HqW|9p(Q zA5R@I9eXUxH?&30KX|(?0kKzigw@fokzM9p(|Uno#wWmDAS1t|z^g5!9I(m`F=Ue8 zH3JM~?EE(a3xQN@-U@{a}ar^+patBjXKg<}N!&Dx1{f)Y)neq!`({ivE*)fF8prDgq)3%# z+FmXo6KTCHx^=rkyu@)>4>B9Y-(E_6nJ{gF?nQ^?z#UID(~wsbO|nDz_Hi0ftlZGpQBaV4eR=yNxPoyf}V@C^r^ii`g9#N&_( z5tx_>!=?w*cK;oQ`S&I7xx9`o`DdX)e~a4g|Mufp{>_qKblKLdP#6 z(2sT7@+x{?NgLVeCKzH~x6=w#v)j#aGF;4xcF)7jJVe5&PkF#bs)*4eA zTe_EcC3f9^u1mGq=#lcCAGDBm^<-}TjZ%y?-3f*C@Sl6078R{za{n~QYUh}wrOG4(eP~WtFB*O9%$!ZLG?}AUZDEU2T zw=ypq*~8AZ=k#es?O)!a4J_7J}*b%YkY>*Qn5eI+FgNSvMny!qUSo0z<@N){KCH3j*c z?dte;X7aaGg3e=dKTHtjLA0JUaVxY9WV?_lBy`}(Y$N)bMyS!^JK8xU>*p=@T`^sb zG(kmI5r_LpwA-zKdu1_QmTBqw85g5ThCP9M4>p8fBXP2u&82gZ5eWt~;OUWM@ZV|2mpj*qOm)tEE}-MGO;VU|i^(jk3mYO0%jcqIR0mDj_oxp!Kjjy4 zSF+i;_LAa_5poFDpP{ZCQVp|a0*U4l(;F z8bf8bJZl66FOfojU0!me?yo@qU$%N#qDs(1% zi@NNyRr&+FxfLR^zmQoZ5lQH)o7xmzqkin?5tq+ZxZ)k#`dG(*A%Y262aN4@BA-4Y zxraraE@<7O)cav1_4IrR$v_juKd_X{fRUkC(P2w*`F~*sJCPko7>F=b@3(oG5KP$; z?3IfjicnD4v7a!Gh{B@D0~}AVE-)wP^t{VC;7bHV;qEAj+`6@5CIaG;D5}hA&b`0F z);qL=dhROJ4J!9XclrD#*ts1aK2T6Irl(6R!&pfiYQ-cfuQ%eNy!MGt0&P}eqr7He z6BWy8x&$H1j0?c1_*qy&aE?klRbp~q)w5YkQF-z}3wC^# z1Rb`V3v@tCqA&Xxzj?4B-WG7!QlYyySIcGDEx#*r{MvlvVR{af(D>KU593kT7T-|) zWq?Y%302;sBMVo1#8B)FAc?yVUPip3YT!1usSHpdO&}k_0hIDflE3lcJ-V*ERZ5|tl9$JZ)4QHNq`B>cQy$Ui>r7~7%flBtmI20)Xy z#reb#wZ&;0%xOm=eKI>AyneLJxdendZv)~3zvf|X{<0&udDm6Gr&idB9u}~BT!)!< zngYm_2}fYEcLksN6vX8zF9EIURE%~FMF?!4HL)y{YU^~z>9&ARgTyC|;47TPM%VG@ z@K~!#dxA>9^hBX5bT#usWfI?OrpP68u8Va_(B;I}hUnNkNQ6islKMIHw^fJo67bqZ z1&!@^LGRrK$Bi~31TmB2w!>xTb>45Z03;JqlY;mLlN>QpGgO3w?i8_<6z+Fq&~tP! z*ao5WsQs+R6;um^ECDjU(ho6G0NJO9$CIa6C>{oed6EPwm3do*aeCs;YGqrhwYeOqNG7Ud1{&v8k#L)L2XOa? zGBWm#&TVXTmu=&ThlmvU+$-TRRkeCbmD$GbG1V<_imcdMLP~DW8wa=#IzL8}#%R~H z5J3U!g_fu=^#jHgYsm}C?3e313z{-RL9=@k@G^1FsSArR>jj`Eh7B?v3h>}jM9nxS zJl$l}{gv-uG?+K-9!RR%vMIM6>C{H^*TZY8^U}_);kH-eH334&>kgfp&2`Ndn()|z zvzW7C3UEW1A~vG-1X=4(3{Ekj7b;5AyO|up!1THl9H0 z7!s5Z(P55HX2h!fGlp>c#Ni|7_Zg5?Ms2t?gSrSU!=%lVer_r3leMCab zm{hl91n+w9o?EIx{SJ79-SGdrQLU`x8h6N{tKi@h>O&BMtqQ z@&2<=dn3FH9$0cQc2D1PnD`zIEXh=$w{8olU;G5(!-o-?NPNK~6TbPa2J;ag$$~|2 zq51p8*6)KSkVX+$;vY7cPTI3MpuT7L`OF%cDxpl6n-kzGbd$XJ;Wv;B=E>6wLS=gz zVT8Jhym%~QPtv~!Bq^e^L^V|n)l9k7rU*!;3qLpe`K+gka}IvSd@&gZ4)}1Ccd-Tq zVD-r0U3Au#Wg3QAe~v3c2IPd&-gy|J9KLxOUppeMrAF8megQE;w_iqDCdCsgPRgK8FXQqol=PA4kf(pV=btzJqRws}q79o^+4ej^s ziY-(XSP?WVJ#FD9*!}QLJY}iz#a8jqvJ$Qd1$HC%B?YnE#>V2p*QrdpcS?%yEe}mlakfIXjq61Z&ub;KYpbdQb#8n z_G8?V%kx#r>8{HY%fHeB<}k^_pXs*NM?HgzeE}7eDOvG8p)*9Y&i~BufyXvZedUjX zwno&!Wx`k4v{Px!zTN?E_J-nZUOV+AU^=x*8?Nz6jY(bTXWmeI&a|~G7-E8+V2R&- z6(E=bF!>bmk1H8Ljh!(*Yj70K2Z*6|OuN&)1y zzE`G(RYiqXa=KEnc0@k48lk%vyfeecebxDkhar0B2;%b0YHcYjxW=XZ`s5#cTy0M? z)nV$x5`A95zV0xNZ%fusZwhzx&3V?t^gzC%qf_M)vUPzuvxBP!hCk~o%49gB+qjKw z7UD%!rD9*rk0^NgC};sW3sV+dGq+szgmcHGMhfO(l10Ug+seb#z8|M)Ggk0k3+JuW zXovovSoJ?k<422)fi2Zmd{zHvW}@y2HhycZl&eYfj5aT%otFQe-zfgIYDU_wu8yT) z*3Hte7?3(zZrmc>T4n5R&gNSY&%9nM9v%qY5ErYG=wT*CseLYGUWYm#m>n^n*Dt2Z z+Wl0=Gox=;!SPjNPhS)7uZ!pM`|_j?kJ3;lD9Mw6pLwT}=*FJn=zQTK?RLiN!ITl|BVQLekuBUY-eL^q?l z6VMJj8_z-svAI_Tq*@BI7`SW0KNpg{{Nbf;>e>7U?&barUUDe+7tXmFeqZCDv=<;w ze2e0TmvS$0juu;%q^h;o)b>k=(TcdRXwI>EIjVI9L~}L{cs8nx-^*^ogC##dm?7EkJ9pr&mkEZ{{ft6G4@| z7@4SawRViiQ8(Z+eNhIRaYjMPA$V(9FZnH}rIfBHm#IF68)UZl7v!spej%EoN8d8? z^07^GX>l%E@~VtyjwZNQ`#oMW%)eLZhLez*=FYAaKdVV{0O=PJh&?IV06jp$zi@m8 z8po~r&=7qkU1eOz&gS6D{~_(gN78}~%4wpFgur!z*fzi*aGWeIJKaa7oQF>)D_u5Z zy|`v*Z-cn%X?8Jdt8`|CyM-00S#C&L_am7~H-THyU30l<_gk!^j8iR*n*u9#PyNQJ z@zs-8-JKbkAd}n!Tw7muue2AOJh}pEEd~n@&l{E=3)R&(_4i)XWvBB}XlaZbJ(9W9 zo6{SEWR5JAI%mz6rKWV-9sexpe{|iZObG0Aoh^1ysKUQxXLHik0}Z;voGS*k zG>BC-li1npEncU+Q0(@oTCPc=ex25eWgeED{_L#!wnKfCZ?-t9dMZmhiQS{tEv+>! zv~tmu{2wAOE8V}N(x}YmiAmHVQ-mF84jW&B%T&EIC9An$u`k&h>^W!jMVj-8n5)b^ z&rx3>CbW>DxyTAUf)0S(9XlJy-M5#ynpo(2(0;i(UIiiA?U*;)XP1qXX7o8;ZnNy} zvq$*@5;{COMzMnY=D7|zGpt$ZGf}8_`hTH}S?Ri)&2&bpd7Xe^RKrj%I3)HYYU8RZ)>S62o60 z_}aZmJ5dg1l|&<7jx*e+{LUA5hrzZ=iAv%}4e?LPX6cS(nu*&_3J8`+H(!TF>Fd}O zq)_1^b)ZfJzbbn4h~gTp29FR|7~#~_igWehU0NTD-~2RVNar{-P6QQhGzdI_CR6^} zLfwI~o3YXx3m0emjyz+r!z;n%83xdTHd!?kZ#do$d9haqd1nrZKh@36F@4vHPkDVC zRynfLx28?jLsT{0J7MAAL%Wqp^(OONpGS#6fz2oG7S2yBO~$e>DFSPwV)J6pphQy)1A3z$2rovE>$|SeE4=P>-$)QErQGwfyls^}nd_ zt1SZk@fLtp!5)-VTk|L0S&Hj-r&=q*tZs8l4HUAkn(~w1l8E+Qzw91C7cZFB ziamr2Y6+hou^Y*>4-3%~s`C(JV4x%Snw&-hK1_g`ukWG!`%lFEr@BYirFFr^XmI7> zWj8@p#4FJXIh|)tIhEjx_33&QxA0I(V zg83P}GY)li|4-2tI2S+frI6U*X?QSr-+fxkSXJ#7N18;Us2aw$R<+x$a7E`PXPp~0A-0@CHGJ_`hDk#ObHTf0Si{kbh zkz8+Nf7gOvp~p&>2wdN{`BxVaEA#yy6Z`RYhJzM2a-oFQuCvAB1R%XdVE8lblLF5t z&yPywKIwtp+HDh2LzjjF{DhBkldnyvp)t>s$>FiMb-2EyZyRDeyj4`Xqw&y%-6IbF zIX~%4rd&@HiYA6NM%klCyQV&5)=p4t`^CN7rg+$U@wuOf%6Mu@L z4fm)7YA8dIIT<{b-2ox; z$Y$Rq^bOjhITDu1OeK{6UL|;h!z#Q&WriAPxpu;Z-<nfE#kBpV;A}7v_dkKl#Ggxp2g*Ob9GR!rTA|@}wP1BlGyD6YP~GF( z--Y7aKH$9$LtgStTpTzg1|Hi#{k!K|1RUq1O&|Dksuz+GmRzLnk-C<>(BnC`bbMl_ zl%)GHbxmm#kF?K}4FT(S5J zK6<-cP>$yW-vYw;tl-Q=HBH=2ziPs9L6bu1B5zevBj1YoI}`IFz=D}hYlU+G(W>Rv zjW(EbQ9!l)p@3y!o%T#Rx`PzGq=w=YSLyyI^sB=2it6_5I&9z*N_MiBanEH=rQpmW zdqIPtocI~4U_I)7ntd(mbsBoNGGt-*Rd;)K*cMRq2rRjnSoG`&N*K_jp6#NaUCT*> zT)e1@0gF9WZCewe++fB75E>-HdH|wDaopoA^i5(g6zb0Zz2`g(ETfr*pYGEzR0lb8 z0B<}+CDHK~ATRA79~ebKP;$sqOTo*3(2fLfRixphPeT^~3FWd^M$Er8TMrqo&-u=O z{uw-d-0@TCwTwGtT#yxEL1m(x?4~o0!bD(t{Sl&Cw{s8MrBW?{>bSHt3nNxs%LK-hv`6aVhnUXYZCC9W1*t*b7P^x3Zm zwe2^QgSdVDNRlFibFv%dnI66UC;n7QGJZGm->79y-5`Vc4VIH>(kpu`4eS+_83BUB zL)R*KR^=F6^z{7-e>KlN-G%oAV9D}s;_W)$+QYkTpE)i$TspZ*`tLn@IfttYES1dK zywSvS&i{D)bRza-7>7srJ%}@8$QBv z@P)N%_l4F|lH1tZy|V8K;L8_S_r>#}fvBR!M~NpBoQDki3NYgsCPfqp3&S{!GE+sR zz_wz)WMrOSFFLl&^2@d!`!OS2;x^5Q0CUydW%WWNix9=6VUmXi-}RFX z5iClNl0%Rv53t>%W}8Vo6yUw(Q4Rj|bT!^AoLO5M>u*&0p$uQn$oL>V+TRRq4Vy>J z=EWrNrto}QJInlC`HlSdo}Q#Po6UElH7QG1EoY1&mTd4JQi;|ILR&{9>b zg;EdW@y&LDn5e*^cg~NR*|VhR_A9+lX$aCx8~0|@OL=S*d<8q;RZjq`Tjn@CYVK#% zil7QSMr*;ZYX}KAZ%0j7Qh66Z~sNn%js=fm1oR(z-$E6>8MjFYq&D zO>ABRUIO^=+7o!lr?99{3_c3CaPA;QP6AY^QD68-A7j%Ci^@4B_R=;c6|6QE@KDm1 zjLH{`(n&%GXxufbLteosf#>{KITv+6FIW&?$e`q7uJj94Ayhz;b%XiLdzS6Zo|ku$ zr%_-6CF~t*xKUi%1}ljnQt8q^12qDk^6H0#gQ6hcQOg&Mo%l#}xk6hQZ?^DuYt!!H z!>oWhYKv{G`A-CVpCakAY+Kw7Ai8Gpjdf;3&SBNx%{eJ3KoE7$frR?_F3!}&go@o% z6<8A?+zyPDAT;pBnRBoLB%Q=Fg$HV48YqKg{D^yn`gE_CAJ7q+Mi_+1umz}5uM@42wR{}hL8OE z4Q{Dq=$xl;(Jq(U;{Uv83;JLxe(q|X&-si2$~H`T zb63TQ^$nf_+lSeb;|rYu?c^Vpu>VfsT6(u}ynueGPfiui_0Z^b=?Q;G@BE7{sU6@u zJ`{~{y&3INuZ$7m>F0eWYRFz8(=`+Cp;Wn)j|xU2(KOOF-w3>sl)FHKQ`c01s_c8z z(s#memo4p+8vg#zJ};i%(Qd)Wlek;0uhQ@7DHBran;F}#>c?AwOKw^Mzt1O>jAnCJ zy{27o8gX?r6t+@q9`(`27VQ>D$ID&oa!iFT^{BDeHLBrAG5t-!9>$lpSsYQ~5D!05 z@Om7-Au;QOnX>tm=fn^>G_7Cijs;^uT2H_I{`75~qSb zHTE07O}02`u6axPq>_hq&^Hxc84JoHlEZYuVMC;;KfuC&e{=}Htkxr?8vaf~tYil% zdDbJYQ{L#8VE6&@!=OL$mqAoFg^`7` zYIQxFadL28_3nbkjT@+4FJpWj3o2&_n{QoQ{e&7v_m0RZjhGbUtfc>qPbOL3#je^x zykd8x$s1Ji5LOLmA67a<8pBDGk?C9|E>4f88NSAVdmSGap+#`dwULA3+k{ZbmX6{S z{o8khBG|4fs=u#^+%_giF$Two@hyvC%FokRKz<6P1_z~hqji4$_w@R%a5q%7 z6hkmmm?>d%me}C@f#GTMAp> zxjfSQ&;-)pbI+oxrS*@dk##}@=p#SpNnl4@gzt8t;tZ(6(G;Zjg{Z5U>~(sHli8}5 zn+M+DbvvKtiP7SNsYf>S!S_5+$7&z)e3mFji!?(d!~1j9>WpyWZ}b#4Mp7@H%=>@a zhstv*d>CIz>A>_)3FvBy^fm;|i}S&XoD5RQBO~}w9vXjDzXSi{e|q*M*4X7d@$1`| z6SsO|XzDafg^u8f*0ds?1!COs$KM|5AwYqLTwmG;Gr=S^5Sf(+La6ckO+#OIISZdL zV4-&+pdW0^oEibb$~oOPZEkbK$qnIShlTUu!g9(uE_gkwHcKX9 zR_nb<5oqZ+E3);FW35(uHpZh_aeLl=S@JCF8CINotzBJgWNm$|Sg}i3aIs`J{+OIg_o4Gx>;^Yt{ zM9p%yO=or<{kOQI4Xmqlx5Zd}*2(|;`0ttPKPR&@9aLy_F};j_wx9Fb-Ws9l)BJqt zG9m5c^5J%%-Iv#x!F>3^*B0kLr_I|fFji!tEi>-=`ipS9JZ5h-JqF_hC|-VlI`9DX z`wJrmF!{v7h2)8Jqr%6o%SgqKaRU$5BbR7!59g@Nc?&AnYs@E^?@5V`o!YW?-~>>X zJcx~GP}dIRyO~qHxQM#? z2{S+UmW!5WNG>6djK%_n4}pFvA+?Yta-4#) z(63P@V+X%Y`8+6kd>Q5S=>M0@|C{{&Ptzsj2F9cPX^x?iOzYvuix!e-d@(s{f4e2q z$8qo_-^TaU)l+qz^`IpHZr`<{2rqgY6HajtrRTrVIZGJ*dgA>-CTRPb#=pB7hA5>$62CAd0`cbFU{of_{9NSvGdq0UH z)^6e82blov>lQI9S0WFw696vN4=3IluX_)m5eJWq73G}Glm-F7YhUKJ_ZBC-m__Dm zWK5+U5(vx06AZ-l6vw_DIIp-f#3qe2?m30t>8HSv{ttFKAP}}^oxgU{W_R^|t(<&P z0>A~7mDA*2U(oG3elDeX=|E>*sL`@rasBtCUUqVw^Bo63F1kMjqn@nczvBekp6$!r zuiCNlxLCPo9Kv@VsRe$aJVF)XRprN{^cyMn!nx)&NcE)VVK};{w3gly7}ZkS_)%fL zdS$sSp5A(FQFWUUw<~X}77JNi0}F=d!cf1x>iwKvd-i1RZ&Dd3ni}hB)#_&2D5OVS zHoaNunIX^a;*>TG(He1E(-tO6SZdb)gbJMP8!1fVoFV&FK%*v3pSkZ+Ev?-Cpi-W_ z_rB+=j|zx2(fOOW*MbaX<%GeO^DebETzbdRf6#JdxO|Ay_EbsWNe4SdktX+8B6Y=U#=qrm4 zNC*=Ocfpmvpqxj(s+GVC}hD>)e9^d(`N3&2GMO2)Vo}OnOOGO&cnp(-R@+y$$5=A z6ZQLT`+TPDx%Yx$nfpT+qh{C>$?ou%p*FM7%X|FJzMEN!w77ZC=!YO$sr z2neKxSnni-TFp00JUJ1W^}}pxmrDT zN_6W!3c563V$yDD_MMY|aj~Y(q}`1hevqaIX}4sKMi(Pw!8=KU9L2Vc3g`a1_%)tJ z1;CK&ZA_=OPLK7-XEaAtpL|_Z%t)Meppj!7{Gg@an&l1LV;hxk+&6JqtUBMgFB{Oz zioBRY`MUcW$!@|G?6V|&n3SKli5^wu{L5F0gSgAK6}!_&aUord7&3BD<`kAfN{ z&`pl1-B;{REc`&&@W5~q6i`PVdRFB|`e{5NsZ z#3G!o1i#}XB^6&guD_N`2E;4xs(jm;u?4-tK(UE%(0Uw#eKrU@2^wr(8;pm0wY?MC zrXpW_LS3pI=n$^iLV;7!`(6j({&|moaW(yY2B3eN5#qZ`*HJkMuvV~JF#D_!ahnS+ ztO}bU$B~@1dOlUxFly(*6C=eh{YcpvP(@6)-{1F7l#z)lKVXyfSDt!EmXi>T7W9^4 z?lZn^{x$#gc{I(tXne0mVd-E~WT;iLWKf1PNb=e#E@hnZPP!T4_Kb0;)mnG4cK4Jv z+)X(%BGiicjpRV){=_D5UT_9wI>fu z+qs?*3noMQ6m8<}SdSyd>W4e0?6Z@*a7op2fCp)6tg5@(w(QJ})P=PWQVz>v2(O0O zG)PLqw4L>Kl3_eNkT&nVz+z`jtRX?V)Pc-WXu)X=TrP|K@KtCAw08KM*)@4r?(hA6 z&D*Y{Iq&3MT2gQJX$&u+dOI6M^Mk8~#*rrZ)RJ5T6PtS@oY1gBYK$qr+6#ju3Bh!~ zSZo6q>h>_?!U2(J8t@wcS0Q(H6rA_xWt6SH6!Cay?ID%vWUR7$?iG!%RFixJMUDGf zx#zxhO2-$ovH(@Lvu{+wP?j(gUS8L2R$wCwW+34p>!g=Q zpK!^faCoN~0r8Z;c%}M7EHh-=!uMD zcgQvKGRO^Fd+6)Z$$FQf8{z4AX%U7#bAtYYn0k?I39MwYem#{cNxS<4#~6nEE(Bw^ zQ(W%i+k6A{WXQw_bKk?h9_7aJkRMlt*%DG&{s87gEe25)+inkGpjWdWM@H;e5amkz zY>F{P>&?=!Vc!1hEyVafAJMD0$tKluTb2#0oij`D&TLYIZ*&)_&n$>C8vm(uIVT~) zg{gSi7OBs$eAj9xJAkVSwQEQ>%d}H?SN6sH!~vUn_oKLp`qbvku~42(fw{Er4RPAt zK(o~t;s)2~CLlnOZM%iL%7mX22b_Ot;Rgk=MjmYF@nMEq*e@w#Ouiv&_D_%i?he5_ zbIE|vFDU!lXvb8Wb~{^_d3u@XWp(hj>~BdHwc>-sPAU%3G6)-1BaUG^yY5>>fjXRb z1|&AjJ_B;JQZ~Ys9s~I_DoE|?@~NLUQ{YPPexbNaB%BUnEVgC2&yV{)CzWt9&Tk)M zu@k739=#}RE0^gbi>7xCZdfrUuxjy4*nh;Y>XJuC)Gq9-NO>5ILOk_1O!ocYSriwt zD4N0$nNU-;ZfWIZ3E!LHrut>|RJz0!asvEFfmd-vniKr)MGj%MI4M+%A!5cTH)8?Z zTM{IYI^593xfim%>u)E11jK$ZMB4IzcLJ^TgsWkwC(6!(yTFAOV|=<@pY@OIi}9^{ z@$j=k*K*rl$oI5AL_=h7?iNBbAmUGTaOvHFzd?bNn>WH~e%wbT0)=}aY17)>_s(g; zi?3nsbM#}knAL*Rkha`IB~!hN;#)-LYHq+K=fsGI9s`Z-7RcR=RW0tIRwN&9mI!d5musMko}JPdxxc@ ziNf;BwCJw7jVEM)JKVEH1NKX_=(Ooz&1|RYwHL1WOr$A;6@EP(#Wo6XO{SK;2lVo` zV?I&+;sSczT*&*?D(u0_G#3j(+3AN=n6{wBoUow;TSRZqdYxS9C3GE<~g-K@wB(nO>8t4CEP2HB)CB69}wiU%j%jKPwK z5whBsn=29Xl6+2|P4B3w=sYkVk^E;A zT=hcDocpEF3hRc~{%6SNBWtkN=18Bz6`l)Yw}mC_%JtFO$JleG|5s!FKQ$c?=3$b! znfC!RSBe50`XzH@4d%i16`8VP)FL64Zdd&s1kE|vfQJAV|MBW!VOs~J=a6U#SLwPE zfVnoU9YMHHn0LqFO$fEWuYYbYXs2no3s0n4SR+r9 z67+4KS5N%|5?XX$Nc{v&{!z)brZ$l(nqHgq&_vK*tMIp{{l9^P?EE+DOsTt?Wu901 zY9NnGOaOspkCkknAv#`ZZi+~6;3*qrZeI5)y)Tealk~@Hqzccgwm0-H2f01Oi{m2~ zM|LJ2yw=$plB#(9<0e{rA@x@m!um?Kj0K=mlAdgTC@6(pAMEnLJpZe>U#&lwABm1( z-T8u8Z2&q~W*Zbm_CLYhk}r~yTq$3g=lMD)nAP6#4Xdu9p1+9PR;z9II%!=a6$=E* z8VuG$co4FjNa5<@B2oTl+|yO*IPt63kQ+&eGY*>~n-uErC0za#3PlNu;{Ig{-DOux zm;QctA$59!8Rw=$aiP{E!vOkh31~Q`L)@$OT_fo_^+i(zmU*r|=1h*2HLyiMVSvhp zc2>g^t#E@WH#k=%5d` z-^dvJ8$5_F1AL%l?IInajZSp-=T5F*mv~%$qgn+}xo1NDSWC;@|KVEUhKmhdsf{q` zzT#>UVohGiRRdXR=kC8=QLrjIfz-t43s2Q;Is4DhlCg8uj7=|oACaul^^5)}Lx=5d zKfOrSZ(i$av$d`#H4p-X!BOg+JHXYG#4#^N#C%Ie+qKd9N#WpSIr^ID5Om*v;D>9TCbPbspP-K%W`cRDtT{3NsXpUP%zFXzA%qO$FmjgZ+bM1izit?K%chbF?HlG@H zZjq^50?Z+wq6}-ts_q@=5{kK+Dym)mwaH(pnE|Sjv;dXeNB1&0WDL^z1eb}nr!(&y z095AQw)imusJvNAfrt;$-_uNtn|^IlVc#ct%@~f%J8z;@VrZ0sV2$4Ty>s<{IndH5 z^NiMn{634m2C~yo%P}Si;Xo0Xyp@EIT1AvBq_P0If>{E~24v-iOn7ou>bEs@=t`#l zz9{+gmPI~lmq+sz(J?U7EyYfCa^!R;BQg^JGYz3Zgj)L)*#yxbE=MCnpnaDmHaWFp zP2nrHU?^q_Q+Ou~V^7RW|BeYd_gSlbYp|_oJz1RClKkY5YNyaFM6owIM30WLa0J#O zVIKdZdMwKE;Ox3`=bgOYNr=5eAkeDcgq$Ecja0*N4dbcS#8;Z|k;!HReLdf41|giHN8TXdnmoY&jX*_ z{cj+e<`%y4(4xm>eo9VPd{{)bZaw#(Jo4V|3?r@a&cEX4IEQ3QDK6gDOza+P<3gv{ zOlp&weK!Z5MWPpJuDv!k101WB5FLD<^;NB|n1q!=TTPv*D($X*%3~^#2R6(&)z{&D z$Q5yWoH&NhhU>#r+G0xibE}3jamAMoc~pv2{&W_O)hgNwaWV$Lgn@3em-SSO89C+L zFPr3f>qcu4({zH$1pVp7v{+FX#2F&p)L6h#R>;b${m0axpCwML5>V;;NhaI-g8N_B z*w^3{AE&_yw*_Sk=Wq!NF=)cWnl)FDY2ziefw!o;~mZn0s=TPT`kaLxyI)hEAi ze%iWvpVWg3F0ic}67s8xMk|#=c^dREpIBStiTLK+ZrKz+8SHdKOeWS+Lmn`+Y!e~q?At;w0z3L4d^I+f8 z2tAn^9Tdiv2=+9ESd#uBOz7t*FGhK4)}v`|(Gu%;$L7FxXlfua?;}PrgHu2woXQnm<<2*KLo9Si zz^?kZT6FtFh_v#e*eKPSzND@c(G2I5nXWaE8w|uHDE#!9DV6_P0#pityD~NeYf*Q6 zK3;`S1-!*eJE_VOl2%JX;9AkqFsOO0nroxuQDAnLNn=C#08^vq)dIZmb6 za}soYvFw?8C%5Dc^!1sUi`+lOp6Z?OR5Z<-RreZ*uHb{l@0bc|D!O7%wg)IZ#>f!6 zGs)nQTLSZM;rwsfg*`Yt4Bvu<6FCvG4_P(5Ve{26IOz8IoDl+qLVsqn)xkjPjFbK$ z=x9#=|Ku{FhA(Cn$dCalcHpVO2s!FHjHZPi$)y0s%-<>*gm z4Fv0*@vrM~nq44;|5hRgV&0u$&oDY9S&;fGI65RfdBXH*k&O|YpAa0Bd!Foct9U6f z+*zFHA8FZ`zNgV4viU`enng?haZlb4d9cCduY}!-;%_D0D*3ld@`2)gJ!eD?{UqoS zQdfKu)QE<`W}iDE+_pe6BloHOr(h>+$PeHh;vbxvTIXGX6JRl9(>OIBJzi?#WYXN@ zm6&<1cSiR}o7ibg@N2BVm|r6l+bvW-6DMwivq!%|jSy~NmBrTtxvW^i4TL2P$ewdL zS>fQ~$!9!Aqf87+Pd;V$DG#1Yc`@+#E123p{a<(I#FK$Y$DI>kO17jmv9AVVbMp84 zJZ_Mx{Rb?2h`$1}y0uuY7+egAA)J%$6uxlV5eV>$XzF#T!{78S7 zG1wW81cE992_*&>4g;L$Ds0F%4+$-0uY}E(-zz1S>f@AIW0jDlr|5nMS*KbeS*W~yQ3?`(>E!?G!VN($BSC#U7(TLL1Szn+@qVXfrtS75I@!R*t>6$G)gTylSP=54eM{Qd>!ltno)FI!={ ztxw4KAecR){_zuj=P5y**y&=22+bQMF2!6FzbMi}Zw}<(|B#1vSa8N6zm!>>?p#9V{tIQw0-rMfYnm;7r9C4J> zCAtcgd~4}<2HhAFABv{Ds|qs+bXTL%otmDAjP3F(x7_ADaBt0-LURAM{c5=ld=CYC z8NSzWXoJf+N{zhf+5e`K`nEM<{}slmD5Ld5?@N87>L|$p%$Cb_b7A$)#Y1YYA>=~I zqiuTc@3Ai)R7YW;J)W6}lXi9=7wrvR|9B|r>B_L_?u@JjO4)e%2u|9$FUQ$)wsyj-*|dB9xqiNl#NW#$Sz!NBGq zCWLP!`^)U1sccJczM;WM?OK5R?Ei3{X?jg~BrnKYm?OgirmA0kWbC$N=YxT>lBIwzhJSjw(76+CJ2U zoz+sv{QW#k;SvERNT4DS*KQvS?pa|UT&gZ2-2rlb`!4{p|7sv-GCPb{)bkSoH~wBr&^RW=|GBhs|O1?P4rt=k=AKdJcxri)133Y?bAp=~ZS^Dnsny zPA?ZlI8Q?8tf1{Ah(1O}PWiNOGrrmasW6#P&D2bzuD78gv;BPWI_=UR3)BPw$NWk+%{09-sPA29+F*=Lm(B*vuD5f`xa1eXUx%kkVhB zp0~ZL<~hPntC`Vb2Ps3;X9P>Ze86-P{=W+Te&lDL;3QF z@0d@qq_a*(t!Ey|t)jvmQh5E|11mo;QM2#IV#sjyugqx7geRx|fyi$m=!*!tU=ZpbmFm0Of5GoqF`GBN z4bVUb2v%4@wDL^g3AH@zTfyNNUW-md^}iL z@2rDlRXVAW7HyZj!@wDk5TE5cl-^`WNqr3jb0^$bPa53*G35JAr**;Qmp&uJ(CPbW z7kJY(fmmMc^kr;>o(Yyy`eq$G_}JCpbZH;VKwjiEa`NiTL#YCxa2@ezi9{yoow!h%}UmP`;)gZZ=OW-NbRFIqDmi# zB$jE9S=2jy<%C`krTGs?d_;xda>xb!*m396NS?7{T-rIC(}iczQ_RDJ5Y|Y!A-<88 z2qhwC+1rI{g3FkH!0-5H_^IpbhPvimR9K(wLRta>^gF}wK9Q698zQGYO@sFhVXU+T z!~p<$or|~N1I3gIS;vdV4@+KV9T@*gP|fn9`|V4#QEn28l|4fC?mD@2Ll|h1$0y6n z+dzycJMceY;Z+Fq!`1{Tr;9p?^=)hYbzmnYFCtNULV6_d)d(vF8p2q#^U){kz~@&c z6WWH){S7f52r&6Zw?|1KbYR)1?CPDSq_rz~8@ZmNIs|NH2R=67`|*aMuGe(Oy*^@Y z@Mb0|cnN@DabG#~!7!XzrJ$U$kV0`Z{@$L5CNq5VZBQe1e5OefJ%rPtauNrMQMHUyRlZ z_Uh9aD%cpoYGjk5e|uj2yLcu+UB)kIJ`)$y16*l59W0x;aCVuG#|3R=HZuN&lRscp zWB+oscCJTdy0=h>EPX!4S`8D=@Q{O1FyBYk_0_TQ! z&%ZJgkk8f1G$t!r0BLwP=X{=~Kqe z(jgTawj~SIkQenPk9*sw7AezlsrS5fI##Q2;G=qX&^nAj_!)X~2&<||49xd+y`iP4 zeUn@9#xw=X6%LB$@2aOweqC{4R(0UzQB_SyNFAPXt*(s$gN?3Fm?)~<{LXzZRoc}ijnBx$4}Xtx1y72IK<*v z%AuC*^Io-=yw?dc9}E*9Q+z`a0Xh`nUSU;uKOL)bshvt6J0I^mi~f7OT+4zw$K<~h z>eu!1UpOT~rcZ5uiNaH5zD(8m^!A2mYy1HGd05--)ak9B4xt!64ZMWWULmks1(g31 zOlFo_KgT^6ACa+s`8=wf)JBB2vM9vfLUcb4HOYjPh0$Cg6x2*0J|rhdq<}6TS9=&* z$YtN4kOidhyCX7>b<Dr5eIo5*{im*KHc416my>(|FWMTI6H`oysFFD5kOR+cpuGvy+=cl_F{_DbOr{l{Cgk}i*$qOo%m zSv9O9hX-rZ)zO{~=}>LFzU=dKDCyZRSIOMSS@@1mXGi(`td@-RXibS2wxePM+8D-} zD+fDs{^foQ2jXdSirGbie_Dc4G*lxT zcP8IHBJvw_yimh)w~>|x4IRmG!DOxF0+|0cN95?RS6y=-=pI?VaQ;k%wQO5@pMRQj z-m+8_;-Yb{nUZ&R#ZO)@*fU?Ut_hrqI(|+L0*-P&`2Bn=Fpft<^ihg_jEJ= zfD4jgFOMi;`%$(DE~n3h6C(FhbQy~fn+rQv`HDA^Ek4l%wij*uy+ge6CsUV^-X#Sj zG*Dt%Wwf4bA>uhS5koEgVepo8S{4Xq9tpAad}G76vaB~y+}_feQ#;7`oT4$wubo7r zWmA`N#aRe~c>T3)itM=)6OrqCB0q#P?L{dPNKTMBpY83Y#`_=EM?nDC-4)ya+C-pP zU5*PRR51!i4hoM(KqRvCT&C2A3%Ldg^5a7cdPGL915Qd7yZ;=Ou;NG`^O)M+sq>X@ zhjbKEF`dtD#IG!CU*SJFmuQ_J^xSWsus!sGv@Uc<*cug`frulQ?Ioo6}>RCwuZ;*2z}Ge!0I&6Py3$krtR+$NIG_@YgjkI$6`V7BaP#2C5@wV4<#1D z^=(rNvBwQ7Mo;`F-b!!dn}!8H#Q`e;Md^9N`A*e5jy_b`%LE?F>ws~}9VLxKh&6FX zUKbx$Shm63Y1?e)eI^Xa#JKxhAZi~)zGXKgOjycKjn)XH;SLs*H{D*?i3Ji^Dt*Za zW3!9hw8^$ew^2aq9jPB_;7TKcU&-pAx8Yn*x3Lj}{933OFJ!&oVSxErJqLT>uQk&^ z!r0k4R%b6!{6L4JW0NgZfJ(lrJ}bm&d16Xpe*GM{ucwW7FsOwM=7(#_?f+a0G^Z!D zQ%vEs@$WijLp|<85GJ5I%zkp9(jmr7Pzaosy}Q({X_CF$pUSkz;MLIqRIy+3PAGQ2 z@F|^3YY;Z=+hR8-1P?BCHNu~(OvimtC zFWVk)bIpHdCUlXgT3N4;cU_Ns;CSqmp9KnEJ=*2FipM!bXReP>i{d>pO7|xVo#o^r za%mpTuq<}v#VBbUJ7cRmAWc}Xq0Hep-H!e!$ z_eaaLw&(&0)cxX`Ys=G&F4$aJ89G8c7kjr>x*1*&X*fL|m9(4?K$s{hO?`0e^r=#Q zKg*2j;7+n^A|$XjWa>S4RFYvQihgZ*u4hOP!X7zbhYjU`39l`0={I~Y;=1-|Z8`Ne z+%_Hy7L%~Szq`3XlR8v*qFeojiKraeBK!GSfun2jI5QzBsv_@}%+<*(pr0QPyB{~R zpQ^{*>-atB={A59@+x&?1Sn99<+yVZW&V7idkxNRO@fuZwm0!`pc_QOHB!X2WOt#Y z;pGSaq$BbIp6USql<1MMY8wR5#F4)Ij@GkRv>52ti^7QE_pUd{61gAoqiKGX0@ZNh z40{va?A>pJu(~u42VNTxU=~X=?18RIS)h?u(&qS4wNTUr4DzB#zP%O!QtaS_98w3v zKSibarU${jnKwTM!@XIrrQVV$3CP5d?@f4>;B(07)|ZdY9kG*pP!?8<50ub?UH zentY%;YYoOJR-rv$4;#w@dBbaKjgJ{{$+TVG)rm%u~XUKxN8XL&)~#nMY!|PgL~~2 zs)*~O(Au(V&~Hs#6pwCja>sRWDO5+DERo#^7nSS!X2rItgbfmf3FtjKTj@dy9UldD z*Se|=OPM9KFbzu;ecs^-$0mlx9UXApy9M&f6JPhEzOMLWW;}9&ROGV)UbixJl}TPo zohE4_${w}yUmsy-5K43eT)wP-Wh=&)QvPY!q_|JC4%^l#MpMSyubW{BACmj3H@~YJ z8j}7*8_}Te;jiaeE>ehksy=d%=vtkt6(i2kG;NJP{MPJ;c@f=+n1Ia!0cU-2ON36t5yC8&2&AP{FoZg<=$mkK>kO0BpUT1<*%lvl5MtxhQ z!}!3Jg`MhVOpmqSua^v=eS_CI#vF2Zg2I{;*In~aFquC$myj&z7Xp-H-*B;5VNvt+nz4n8_eqAv2@E zKm77ugW2?nzEJ=_K)}Dus`rsic`U4MRtoJcfTkd+wnl>-M4RwkGz-AVe~*4!urZCr zuehU&#k#HNeSe!jh#YZQ6}rM6K8-*s}mvOCPHI!HCXr%MPq{S2jl zn17MMu;<4o#lX6yo(qZ)X+Swpa{3ONc`1zzm!cYa3j%nkxuPECXCOqAf?~t2Tpnbl zPK)t9vo*f!bSFMUkNaZT_3}$_r&a`Ee%3H*-l@WXAy?LVsVagRcD7NqU{=l^@dkZU zorbiA$}?76I%d91cO6#_u)7TNp{O=o#J0QJ#|$&aGQ-vwOIsVf%ZM7x^|PLBwmH9J+FJkget z?>88HN|E)JaH}#MkfhXUo1pAt?|I(Wyp8nGm$%Yr#tddxcmLcN&~`@$Fu4lOHDPRu zY|2NdGs0MBr{{y;SC;9@mr#8RG*k0rcE!%5uhZDN+QW`;cUuf-SVDqqHx5_TwS9uC ztPr1H!x&W(1IwP)vU#?9HHZcgx)?M)Zv}^o!p>C1I~=CF4L+R8Rw9l+NS{?e(I(?H zdo6pUoc`*!#3RH_bi3QA66GR!_k7hmeuHR1ov%Jkq@ z;7&Md(BR(3Vl%D$c8MK>-yiUfI@65uJ!axhwDLVpesI0mIYJ|82)t};pC6V&K!y^4 z;b|~JmgML@jfp_*QuL7fWl2TRK(95bK$OnFX*}fKukU4LV^}Dp&V&MSS(!P(t0Vsh zV)fz?^JkBsqFZSGO8d-$UowRexTSlxO1c6v=O-2i5EmqT{SL$$J3i?tJ#aTmqNqeoLZY;nsZoU}DV-G5jrDw3&b9f9f#njSPr z{8@CXXsbLdd+e{-;1*YLpf^#sG$E9%;2OE!V%9AQ_9m+g)iAt; z^^es71lYw}+kHe_gnW@+NYKt#WJc~ac5sOz!)p!0bK6)lithd-Z``{s# z!~NVk@0Bh|?tZvtjttuzqa4uAPrY9LsX{o%0x7pKqmnot8nvN+u)h1ngGd19$oKP+ zxKUqZ)5D)Kri-;umPp%iPgttvTNdyq z{>l@BlkA@2xX@ZPpC*f=qbi{kD=I9%0lC)1MtKs?Cj>{T%!+;_Pj!=lfrijI(H~Vb zcO?n}TEFF^y8u z!bL5tj$FEf(c}KeaJkGMXquNF5g=p~Ss6aEAhlQ>o_Va2xT1gaEsNF@fB?ULLbL(Y zJynL+x(#ohh^6R+zx2J$HHrHt37~<+n0JxT^8u0LaLKM5HBJ~l7UKtHT@>>!m zX|_=qr-K*h-mLC$*|h4&Ov!o;ABmz`O@QM3u>p@nBtrV^yjdK|0v;aG=it5~i@;-F;)h8%A4 z;k^$cF7+wt=lfG6gHJ`8SQhULKCN#G;;x=BE`V$XL2!ivzb(bT42Q_(w% zRV9XWmuKKpi98n)&iCk+gZkih{ur17a;!uXS9)6W&f-el$I9tysbqV@>Lihc35`Xh zM}{MdzNQ!ziHDu5|2Rd>5PDpB6s&;+a#liKw)cN9u@>3CDF;+_+WRiXy96H8d)A-EE9_;Lm|``GV8uJ?GU2f{s{^2Dk;p|-z zZmM7}i(AP1mik`W@Vod@9l&TShDY8GSGwlbR2fzN&ww3Km}%zYgpjG%0U3lM{HI4q zi4vHk{9W{<`ElVkxFvH1Ha}|7rmbI}R1awNBLD3+ZCvT!!P#|pa6EVBUe{gIn@D5R zS_XswR$~^_V%UiPbY})6AgPJt`atazis`b9(vOni(=D&7lzfs-VC`G;jn@qM1LQY znk$zsr4%4%K2R4B)!7=wI6;e+QUGqqwq?3NgthD{xW$@ZWAQn1g(|+1r|!97H4a%7 zQ^}C=<`7sexDLB~Xidm)lp*U!&~@9?KklMIC$|&T!=U?&S`Z*9eZE$WVfcZY=iH=rxvj9f5Z+Sp z1~IVw0$0H#Id`756($|qm?JJ#73qTLD>aR{k; z@1;~sE8li!N?!F%2mW^;5D&OS?u ztx8;Om!*2Jbk1XprtLTaKBZ}q@&1YBd-Ga;2XRxT5fbV_@e6e+>iP5BwR>d1t0ec* ziFzUE$%^{k>`T#6z-hU9SeV=DKya?jmGkhHs4KqYp8GuGJ)bx6cGDygvkw^A>eU@? z{dBoYW4WBHK)Y&~Aj|vJkYx4bX=cV0Sta+1Db9v^yBx~O5%z2!F7YumG31hr$L$)% zGtaK}i2hltH=YFTudfi^a$cBY+Pr&rTW`ds)8Ho1xOw2!G6!_Z`8?IYH?~}1Af^~~ zGLSs^#8>tFi;ZP?Lb@OK)d<71lXR@t;mcBi@$o*M=ljp-3#5)E(SW$Q{rQZa3raT& z({dBt5@3hJK#o4B{Q}ayRn(l}>(n3SMpToD1TOoRVAx|x`C4{%QYv=OKc9N!UZ_J^ zPc$~NzqNi!K3KnwztZU(7Hx>{5?k%?@lXRNXN8YlMeiHROV_K6Q z@OYd<>kG|yy(D9aJ)HA(<28E__S}A(NK$-TjA={csZC#scPkrip zde*f7QsU@W@$Qo=f!}rBzFJQ6N;nqvbgN(iY9KnHxdgT zQO}s$&OcTdtqgNzVPpc%`y!n5Ui%^m{GR)V`kcN{Rv2;h|A;!;q|P>3Sm5A^Y$N4< zPWA&ZN%u}RPjn%unYgEJjsp0N7Z!Rgmljsp-CT*Wmi^F#9;;x*j-7e`Xcv$6+xi?^ z%kRGbEHg)TWrLAl35G26dtZj%wd(P@U}{N5tbLxl=TBK%aGa==T=D4;EWEtJeplgK zQf%SIF@GQVOEV-;BzwLMf5q4Q@UWi!eprSdzGFUDg@f7inmX+&>5>4N=9DFBS?Bz< zjMzCDv+cvLi6ldogH**YhlfE7s|iU_^?B;Co()1ymB=+Q~Vc8TS{ z7TxemF8Y)kz?VNX6JnJ*Tk%(B1&Q$-C*kARO51bVBU`AFa z)fGz0Lh_F?~c3P|y-;awGTOf8nh zhGqEP(kIWKzNEx~6NVofQ~eH}=ed8~f98Gd+3#9wueJ7GYwdY|n+ta_zyn)ggm~hs zt99G`YWB(7gF{N%cMh?(L9NOomC$hUX5t#w(G@1qM^BrO0*b-R*Fi+*OBdr7h&6J1 zGF^P3zCfnTHGNLp2y0buqAGQ#!l`>qAXv%aATF3Ynd*rbGsSu23x@+2x<|AGMabT|Hl+}w*Q`m7+pNB z!Dj0oJdjO^be!U9@T$jfQaGzv?{hha>_7ZnfIv6SjsAAoI6E7+jpg$xZC0KR@{EA@ zoUVolc|1`VS&#B}C5Qb$(;RS=LOjbzb0oWrqH8cgzlMD|r2Gam1Q;7rT~$rw_BLHr zD~3Idc59D|UXI?nDBopotc!ZCxcV~dH9)vy^_d}wIlPDdC1y4Dsa|M=GV;xL7$KmU znD*bl0rB}>dple9ooz@uGQrT?$fZ9wW zOWw{ddM}hA7010GZ#71NHhKhpLw`Taf#ua{?X1+-(feraMw_$pSHV|KLT7qU)tW{I zx?KOiS>*rFZg>^gQ!?r@+>MA;j4koSLu5|nknThQuGLlK3;f>}fN&+-8U!?c01*PtWb_z#A#%1{A;9x!n*`&0T5+>NRsZ4IJ$^bX**Eq zb1oy~pDivAc*NwvQcN7(p&hk=R75lXKS34la?JpWZYjd`)ck9LWO<JjC@)X!C(oF2a=XcVk2W;`DB4gR8o5Xv5OixOqV^Z2@B8h!jnV zl&$FGhz4_WI$|M;*x;)V=kw>j0aCPMxs^FD*D9AtWInZig(o!fGNPxcUDa8TIRHSJ z{T>Tp1>YwWE=6Cz!ZTju(lEAjL-63;oJ@Hm5Z?THL^*c~90hhwC)Bca>gp^7mZlf{ zNOt8Y?Y8h;_6`&ODu~O}-BL*1b$pz-DlEzEg`z#`#}f@=-c>VHKOkLhkFe@E&!i|rK!0;O{w@8<>8S6n z&fGD!}b|z9~evqyg6~Gg79;2Itug<6(H>O@ZW74t-8@P$Q0vNoWjzx zH(N8RD}(lp{4UGLK}-8F&c?}T16IAOxGD%UI)?n)<9G0O_P5UOS(11bt9=HH z6W=hlBz!M}cWzqi@wPBZ6~w1RF?k91f!mPvF&5c|s0p(_AdxS3XHL*CvhX6qX# z^7sdK*_isnyzJl;{g;8QKFfV@08{T_&c>AECP0e+a~i&)+g1O_B65;=J)_JRg|Fum z!~eE2knM*WFDZyB{}W4JP_B^CugV(THK9arR=V3|wMivp?>~{dEQaUQDLZ7(MNkRi z?}dUPD+LF8NKU!h@6(lf+XElo3!7Fc&A;@OEYw2gyl6dToB>?;%s`DT6dK+Y+4I{j<)OM55)E3;%x4O{X9ICFeq20xoQY_G!Cju47oU{TbRz1R8 z_U%kJm*RmknVTrio099$mtwYloF-FdLkDAY9ImR)2%qL5JC z*-*r^8Ine0_#(iG>x%Z7e78!qx*@-(0Kx@m9rP?l?IdUE({vEtAm(|EeR6lVM)Dsh zgiy3q5_2~Cgw#TfX8=s9yE(k7KQLLtwu94@V!q%sai>>$Lfz8vs0{SEl= z4fa*!o#uh1bvUorpax>+#mDmpk6_7Ic&a>@ompb`kz0*n9g~J_yo@4v!1VPh{fP3; zp?1mKD=_8Y#gE?~CffCvC|4g!2WH7Nfm#~<6;bXBUnas2PI-uTV1@NVYD{$Ms6gI} zAQ3U|5>fy0y?3~A?GlzdO@A$eMWK@E-X-Uyrlv1A3XPNRS=cUhxi;}<-Dn>Un{Z7o zzAKn@4^2vXz{8)_!?!t_Tg`tyJr~BqDF{#mzuvitsqjZNNd*gDk@)WvTuqO$J!QAj z&EwVJ`#XB5nyhWt#LvnJ6Jv_!`8mbZ(EHn$$F}!RFfWoaA_1Wa|EA3bC<9jSjM!fq zAXrRy%smx&5_sMKG)MB>J^f0Zu8hskHk?KMOBG7~{?Nb;z5-kL1n_6Q<$&o*gC;(0 zwub{z3)wtfU7h?{T3LG{A2i4d32GV|Ab}uGfe}^$S*ruFt^T-|SCa$!_Du&V)i|Mo zD8ENA&0+BLB4|WTk6=I9gn2)kaQ>{#dW>>1^@x~w&BPpWgxTC!p@i(@XjAgUjUB3a zUFqiFQ^g${Ih&_wlU1Y`OL?OtDlY5 z4PpfGNRd~<(tzH=W)yEA{|?o7uMxLAIL|$M#8oXxJE7WMJ!$mH`+>har97C${poM@ zYTH0PX%3igp1au2$ms1qwn@JW&wbIhb^?XXzqZdv1zC^p0F-u=xq!qC;oQLfkz&W3 zC|wQPp5%KV>C?nGOXY-)L$??g7DUqJ6bxL6Y)IgIsyK8 z9;k4xJDZjY=&=TQ#uG&_6>^~Pd@VCsEcbVXyS!`2X_=y`AgiuOZ+pIgo2y3$Zn^V>n;r>q1Nr|4F~(Tc4Fa)MVM_M5MForx{W%3$4R7DfeqTfq zF%-|U(4DTb^G2b_{mjuSs&4&nty0(ZAr-t{IYE9WO+!qbheG+4IZwEtD-Q*bheE!1 zpFl(|%}5#<%E|_(BQLMZ}WQ`Fy zs7M;6(d8Fy{yK5PFh!m2W6p6rPj+uS4^KRg+Eh?t70Uwnhb~Xd5-ec#>suwFXa0_l zs3G2X?G7c7w8@jaiOY7CHe79fJdf%Rb-DG)@2t(ar80 zK+_cF!T~Fwa)b)c;64&2lHMK9gM>Mm^vCn$wzx&2o@AGOuH_T+>Zc*IDGG{_%3Q&m5*|nc6^r>*SFNe69BvBvGD<_;@lfBg70(KYtPNKe4 z*&%YQ@>{o863Du9+aHgz6+*TL1FJE=ujzn6kopLq=&i~sNK)P1B9fO(&;8TOO8A;oJsDd=Ti%gyZh^ZpkIrao(o60b1!mJ4Kw{pp4w_juJ zl+?~AGj^Jr^6si7yaF-Ey7CNq@(Ed_5f{TI9ro{fcd*!! z60H+U=6<IUnkKIL zyF1Ax(Y7bYxy1&aPBovL*VIZeTRbNn0I9SphIr}Ia0J7jsXgN00AjkMbo zK4w1MAw9;1{p$_~T#TCzp=mtf!&eI6cuDb7Np(@Wfx7u8JBE~u#ZPzo{%yYhO4Vr_ zxN)hf-8Pt#==5=eet;XTQzS;v->v=gSKuR)))b&9SFGimiAWjs9Cwlmr}q<-@`7j@ zqWw?ic%na^bQ6&b8-*;tnkf@47y0VbPuJ8mX8`tn^d)JhTpIhPk6`vHgc06|4>>K? ze}#3c8I#rm`|?Ftv+(jQ5%T%luGkaaIF#HYSgtOhifG5NsA{4xwLhobpB3H3*!F#r-0@W5o+B(o4oEW&sh$4T#5hD5Cks0EyL&lifHm5>AXm9=_3zII%l2iPmx8F$-J{5GvJxIcd7w;k(}QWhwOXVbpGOiz=xT-O*`!1ml!> zl;GIyWPB^$A#gyKn=@g*4~%?Xjw#8SIxDg!B;^C!F!Yjw^`{Azx^%n?56gwl7w);U z_0KWZnms2Q?R`2Xt=YOp-MG46A7yHLe#WcMBRMmH1jk}Uob*lsq|EU`4GGUC;pJ{T z3*VxcNV$>_pA0syn+ycw%Jl{)7lky8m}SC!p{`7aN#?TqdLkq_SY? z?QK)@gkU5@WtSZjwaYFB?E3&>p{R!r;#P;aycWWR%NGD5ERjZUosM7+@T!#TXy=wb zFRlX)rz49%8TC4-D>*E8mpyX66VWDPYQEU8x=TB76!2%Q&gF)*Ny*tYnQ*7Kl60gg zENwKuYO+Y(&UR@nD20wz?&e|pUuP&}@o(3D2rCleVSpYhV;|>Xb9BwCX=ouDd%RkJ zWPkFOR&IMfnn5n-(Ur5+%(*zP->?w6%V`Tk?iJV2vO=98WWT6ze{n5{ZJT9$&3Ei` zM@kh0{>kxycg|5vW>D%kYJ|@7gW8(h(R?Gy&ykw34h=1u*x*Oo%}zFlku*(i3)AaL z8VWFvNX^HUi{kInj>_BI!X3~`&U0%Xca7a|W8nuUUwWyb#ut$4{Y{>9ibWNWnbsR%QJ;TnVcDMgHa(vwIS zA0?&gF8kv+3^`G{nK{8bn)$i4?+Te-Kp8Y!`$(VJ4osx#flJ~YQ!Ad$>o|wver|`+ z`}h2?g_^c83wQGUArVV9^NW4Ep&sd+`dPTM+4_R>Q!=r@NRj-)7QXOcpP0?HjjB7P z9y+sH%7doK)^JwG7Knr%N~M5wU~L2yRLdQ%)M_g~KX8wAWin`YPXrw?F2j$7JEc-5 z82pI>WPOkj$G`s~3+`tqO;$yCCE8#!nWzi7k$Fcd1Vb+Rh+EtHsUiLv^V9f0t{tw_ zx=&^RZeN1l@3Q|+#gKnTKVx$xZtV-xkIeI46Y^nOVLBhuq;krU=%C9$8er>Ja-xqM zHfc*r_j_J*b88zARIpNE+(CXANm}AabU}I>(aWtZ98@y6iP^gEAZ&>L{%UYwTBB+X zY=DcBnTjyS?_J!+_QqpE)sgJEE3eyc1oW7mIo!PHQ~J0Y^mahUu>a?4)$cX+mOsNL zI|A){MD*}qjCL_f=*Sf{AQ}(;SuAQP_c_XvwMe`PBU#-U8_=vcP|B6r}?^*hgCQifj_sdl`ZW6h-}M6b!+>pl#>qUy0nzYDO~m*L6MAe}Jyx=G=+W=6#5N(QnV)PD&&rJY z6t^}M@=a`y%^2@ztGvq|Q3^?Um55Ozbj6T!Ya`3pBGuma0xcM(g*%arH?PZPF!J4~ zDJX<2K1y9j1RsdEa#nQ=+l5IkRxR zq|X2V;oA-_Bc(@JpjB;`(Vx`dHM<`9X11r8LMc+_Ya_$?$lV<66Dym;PmFJS?QnMu zIGA0*z4GkNJ*<&b%YFNL`=JMzXT`Wy-z+!)LU!RuC$Z)pruMzF>I03BOS?kV@VB4g z%-13YzEmrzyX;@AoR4m@xhgh3|E~7^aQnWecMke;R@^Zxh8@S76+wEPj}y54>9G7V zRx?&IeVn$$fiO>gWz6rF<2?3jePk#3Ij#qWCstAk9I#BtoTBH;J72)>iDM<-irMYt zUhVJ5Qfb^Cc6tMWUuBBjrnq=fFLtp6OQz`@>zf~KX4W;7^*D% zGZ4M+{HPHq%Hle);&uOG3^4+?ngX0`y&)4S6+dBb5P`^EzxCac@J8!oEC8I-vb8yH zbWQIFQ3WhdqQ}l9FD%1Ej_CXk_ch#BFY(?*Cog+=<)UB@AhKn*;vO_`g zD=V?!Oq+uULZ6(caKJFZPWp;JAdBOe64DvKmF#Lp z%QY%x10~)pHnCQotpM%?Ug+50LXTNxE3u2R`<>e}SRa<(?6%=W*sX{H8VQ7oNeNRm z18A#)?k=48B!6^VzAmh%MtN?31v8RV^du zzp))L;`9ZCxDin6!ak`c~WpQ_UB+<+6f1u~CxHmjRgRqeQ3k?BT{y-JcLBpkt3oqfg(Vfteq zkD`y+xmE$eBz}jon#i02<|9??x<~6n#=V-A+x08=@FP;3MInbEw?c}|R@J=er&>>H z$AUUHOkCFgCeWd4gxXe+^-aDjwt)*woyxM5QH7=c)SZ#Gf&X9%Ol!71Lz8>y4Kc4> zBRy67+s?)#vGOd>?HKh@2p59Snv2usMXu(Kt+QdQXFhRqG zt;t_WQA4?RSk_}+nDUZ4vUx?#%BH#H*a%y<#PA3T8#qh|8pCbdL($#6B`gt)ZCImr zwRA>eN|AcsKkG3_Mtg^9=4GypU*X7)K_HnQRd4GVkshU8r6Z=>a~YhKAq`8xGn=jh zu10iIbc`=~)s`ZB?GQDSc9lAJo_aD@H~3JShP`xS2Rhe3d3XM&KgK>!&BCZj)5PUb z6q5)Co7Wz;`A=i-dQ3A^(WsxbaY?DDTSr2XvehpzGK3W6RKb=Ad0LkjVg&mTBjGr% zm=$I4k0;4d%lj=&M{xHW#*PnFm|`V?Kr1l6@Tr-hl@b5N{RUn8C+=HJPd#Z;*u2Ed z4Wzn-F8ai*IN)mitg&PHS`e{=Z5k>bHm{db21_3(pB#9v{Svi(e(3GMMxF5<|odPAiMh#JqE{#Vm(6mPvmsM4R%DX^sflM({trcKU}e-+(tomcH8;g zBCYWEJX@bdTE8<{Hm{pG$*yXLuZu!dXF4Akl1udXMXz#?D?$&4KIM_~=P~BjI|Ok~ zXg#s-@dvzFJfdyPwdt`8qqXUQ5Z4f%G+fD{enL4aKgFr_1Uj2W+I7p+l8Dhy^o@`E zV*LD23wM}s{mOki`#yF0J9fffgTB6Yq0}X_>3Pq|2<&I=Ns@7+k20+ZYrLy#Tlq)dedh zHq%t~=MIvT?d0xQ;>OX7RZT-Y5rQC*3?ZT;Dl#qCF?_K6!e*bOQ#SFm(%rbOi=r;7H(6se)EYQ`o$I`8=KcK?tMG?5;{MLdIrGJpofQ6 z3$MXQ!$AVK%@jM-b)!+%pmHs&;#@ zg#Tf(*SO0wa<R9f!nH?aW!0S91Ufu1b75SQu> zTRFOC-A?|U;o7V?J2yiC9BNEXoLX8!r(Rvmy;-_k;1!)7tTU>=5$TS7XVP`}WI=X? z%5_F`@y_j*U9T62xSK3ru%Bm8CN)+E7}g-QQBbX~uIPh(jEA8HH-Y3A`yyOV@Fpdp z)k@F{i=z+jT|_T1k8R3rW3q@Pg1MieZ`S;_=#n6=Kw{K{e_3=X^#ru;HZx=twCqao z(p$tXqJ4SUMhA59_kNCWXd>AgXxwH{%>URq-Z&G~*Ll;`-glr-$y^EvE5!ATf=$U; zOp@@o7G#NZ2%1N5d0|CJt~RQcFW+?O;a7|Z8C}P{m8#|6MCAm^0=oM&@6L12xV^BJ z!M23Kti%snbg{aSQ>B*jwTQXQpsd|#)Ck2$0)||WFihote6i#K{nkhdzTzG@3NdZqmbt+`aV%j9huMAflgo%klW zvKBqIlj8pPfkSR5#}fKeq5Z^y`V1fqtcQ*ThkGi5rz%A^tw%2Kn9kd9L9Bv&nSy7rI_MECZP`NUUQWrH=|oaY8O_z~pg3J0QCk#!$uM1!bC zJ(0i#+3l+3i#H^K|H1EEW2U;vSN`~?SyuY#TI1K#wK7*Fw#CkE6tXOAiG-Y`S4Z|W zMM?ItC1CP^Q*O4Kg-ZO%`yi@ED(cmTIlKh=Z5x6Im_Ebja|bt!VLv^M4?2Li#4q?2 z!OkR_56_6RZ-Z+bG8ZH#T^CZKJXi7+-MWB(tX#x0NE zoySQCH0?ISO^&5?WLNvREV?|7$IH^+AWMOO+V-`oRIht4^JjeRG;CG#l>s1M;h&BwK7JdS3P}LBah$q}^sjV^)ncW<-57 zhAwD|JI9U-+a;f&N3Qwo&*gbMX^N0#_KWDq@Tjv1v}SftS}E$i*4~R{`*!k#Qq=l( z1v|m>E_}{`Gu4IkeYj4WrXiL-W>(e(n^+v2C8D6k9Q;|`e=OiQ&z~GT*RLWg@Foq?Fht@>YjLG`#=9y#6ANnZV0qU3 z#(wKqVT0MAtwit3?+(^Gj`jmvQm9Xa+S`eJp!q(q^_)H9+sVVt&Lt>b`#A+xNI5jX z?doTAJ`8)d)z-{jETZA3;oOycuS%O&aL^Eb7+7WfzG*!9FOAh(=Zin4mpvpW3O-9k zABshK)SUsS&l$t*eoyY#SJI)k4IkUqCSP%MrY@--1vmcW3#CmLEGY>!ixrExU(}_H zd5!$5KkPFKC}(EvgNF*L2+2J&7T^IR>rh78XPZ#xtecWm3*NmmXp z1Y#bzsTJPTelGFo_^RZ66RkutYvF)?Sz$b+|Gdy(`IoGPLyWVTWLY}q#zsI&ShLax z*@f++ryDSdki87Iy-Kmui%wHymOpuN;yNU4ec>5*k{8JI&t^c%&v7&p57aDDEJTX>BfJ8Pn9>s*gtMxh(o`xT7snB z*_YU|Ohvmz&BCjwIyrtsJ;|JZ@eg>KGNLUWZ$tX7bZP6-C z!{Rny_nz}t%dG9kiP_!tl{qa%GU>OEe<=3EF(=pizlAUK{2_Q{>DB9h@KcPmXOrtp znS?OwviOwQ_=WgAX4~@5IDNkbpKu$NJQ6|X298n&a6(C{vxg>Xk9P~H3k+kvJleOOMohQIRnVqrjSMZajgB1t|W z$9C8bi>qQ@SKK~7G1B@ob_H15|KqM49Zcwhezd*C>Z6VA*y7#ZvamO3$&rv!0H6?V z+goTJ!dIq80hWRa0Ow3M`wH|mDp;PF&oiq8c=|^XB z!TyE7(I;r(f{w*8mMzERtl5RKsN<%~_5oI7JY2&Z-wT%UTOEDx9BSpt;49-D4~h&x zM9Qp(a~F3a!}V2t{{8#flcwLmJx4=TpE-E5A68$+)z8sK-wF-@_P@^ltONFa9aUNl zstXYiwTYTt$P=#07))X^rr7Y<(uz1ed4_+w)=}9(0joR}Xx;>6%<$K(IB3XrLOn5_qOA`v@V!cwsa_j+HiqECPPQKg=&ESe!jO$qBU9irLv(G8Pnp6Lp=`=8@6j+yXom8~`YQfblE; zRbU@_>;yTU%!ib2ln$A%w=?Qi)9XVd9hnE(k~w)S(|uT3wo|6W2w%@rVM+Cr0< zuOEtc=7mz2`Ct8_6|ap(?Q-w*j$@6yuXY`Yzu8@r!&&UMVWWv=K!?3YB^ctxKc60v zo?D&1D6Kp=4i$~rO#N6|0IHVpcJbmYYZ#S$yKKmL=6%;-6Oy^MYSTmno zTBmomWpBtx&$g%?dm50lBEQA$9PYFPp@VfQ*`0%(QTtNX<&{oa{xy6Or?}8*?Eemv zZ9FnOw^sOn!O2!+2C)1WPImUf^S{B#ZgXGq{%=6pBsj%AKU=vEDkqw^wfn?RCpKf* zhlOP4f_|Aln>LGcJLbna3>)a#{tf?m$fnK5>k-9m$z|1zYyV)~`0DnoF{rnHCQJp! zmvHUe=#S_?@O0t#cT0Qr>-9?`M#MXv4gZPp*$@yO{N}B{w4?K-_m{>VM;?FLfy&v$L;RUV27KT{Vd?7m8;BmT?CswjGz; zbNw?}sAvt=EUSEws4(kHu`P}7w!*v^K7PNlK)mzI+JMl69^c6S`8%yx8U)pjRNkZ; zyCMoVQmIr6^(q+m+G?UbDHU^wYHQj0wd($5`jxR>$&@SlcnG%t8hM>Sr%h7}f0`In zv z`es{`)yph@1%KJ8kV7nYh1E;q)rDTC98lg=?H?9t`ya>lJhoB7A}N1Em!%eB@-5xb zk4q>2H^6NVXllQoo1^mo5pA1hA$F2edL~z3iPq6xSh{N{TVC81AKlxOm|1PGJhx6Z zCW82%K--*;ul$|q)C|o=-nW5sKA^}X7_#^5&s7XOGAt$$P}HN8ukUnGfX8OErR#P2Btjn2-KBXK@Hh=S!LGzmK3ARB$bI zyK;ZGFb!4+FRRL=OX{WbvV3BMj1mi@*OcSd*UTze1bZI&IM+NxX8C0`3u|W4NEar0 zmXfS8z$sgPK+kQu8Gu3x_l@Lg|5g;n#FUYJ)3|5I3TVG2LzZrFlT#$eBq+kL9bU>1 z$y;VxdEq_9m}YS^fT5J6x&2sgAkhTbt73>>-Q${j)_$wyhn}s_%5616(OwA#kJ5tV zJw0+p+~uq|;gKIAnHIysxh+t856a|L__+R>9{EEdG|eUvcQ~&YkWN$HShB25201BU zj{l;B4eb@*b)Z`?zy?pepfjwrCQ&n#08t{2En^upkWiW?Wxf<|ep9@zJd$0AU&P~! zpsR?73pQ@9FVVhFIh^;ye?e%whg0YmEJ5rJ32*>Mej9Eu|1MwCi?7-7b1IAaLLTC) znQL=>9?-utxwyIZ*ZxoT^~fuP>Lb52ad{9q4ryk=Z|aeEA2`KJBUumPq4sY!*gnIf zo1yfYb-&G#{5E%+O!Y^X%NgHaRa*CW3tDN?H}%pcC{ z@<@^^oVUh&rNBv&UKzqRB`Vz`h%h)>#ba10v7?rBk5{AaMO$Ge}ncnK?Ewy9PIoFdX6zZ`{zoJwXpPo&({6F#{9OnXr7ssR>-ub`XN&D0;K; zSOGR)eIvb`WnBcfR!6L(ZgjzV%zBd6Umu1dzPtPfn`ON9jh(PwAmeYM(Ai!WU^EK$``SGRk%T^dfZ$djAbv(tAuB|pVebV+Z$dz`Ef%Pe^3v#%%`&MZYi(V`tAp485Uv9nSR59L9VjEr(H((4CRrSqeC#v788 z2w!`yG7z>52D!S+I(1Q>HvO7*bsd@YU^5zGkmW3!ym;SIrpm;#pW)U1~Si_#wgg?6|EUxXwWpjY+6oWOQkox(B&z6stzgPak!R{ zGlaA~ns3K$v31pFlT!3`jLd_)gNV@__Oz1O zO9i|Wj^y&oh=D#;Kfjp%XivZQP!;~L&bfYGyr{c`iM>el zyi_4$f7+SI10ZNp+KX3*0|+)3QNtg+=v_Jg&tvZFP%%x`OU&=*&0rc~TE2klzch{X zt8OcnX8Sh!-6-@yic?P7G|btCp)cbsKz-=#e0y^_IJVe=UbYiE-jyk^;^mB$XREbC zc^7?or{(Jlj^N4e)i|DtV@v(k2zPU0eR|tnO0+8MvWnJ$35$Us2UD+XTDYtTn@oR0a~D}xssV*E*bk6V z5i*vMG(YYZ z`|0NA*dfI_h4ZUkFJg!ARq^d`zHRlGf< zBo-?6;ONdO_HZr7l+eH0Wh?>tmO}mr4$F30tuGU#cd9x*;`t)#p?r~Rd{IK$9P$tA zGPI0sB*DvGEFG{Vmv+jodDEelaU_${NMSV11@yESf*z_`DX7C@OOP^6u$Qx5N z-gl;<$x;m~OJ;fc@795vZJLm>S51{jcB=xHh@=g$1PZ8S6FpqM0)OIAd#E$$-9mZCHFV{G+Y-*o|xYFp1rWha@Hu4~6nAH0HnxQClR%5%{wQwNFN_Ry&!7rip`L^KiDcR1HzlqAa&dSIO5`W$tYut+4 zZ(Ek9r*9~P!rcl7Hqff@%_z3iij;}<{QZXwaEjS6Jva3%R;#iF97Ael;ix8+G%u6U zv5t)lxpDU2de%h0gilfGv61Ph6u*S#9=F{{_#Avt#O>gPWR5a0OE?Owe_i7y88E#6 zUM4gtqq2RY95BQm>}!5t`EI&CUsgozH`p(83`w6XEjBsZt!GP_5hHkYA6;xa@38fx zs*l@FVSiEZlINt7z(f#WaSSXjxgrM1god0AJ}->=Zzz5C;AsjId;Ahkm6FNFhMiw;>jbt}K?*q9VxLLLaz{rw}Y z3E55^8~f4|OUGptx}}iUu2f$%Gl-CB8gt2ZTJQRaqu&ohy@<-90v6YOpHi=sKHEy< zpi$PXE+gmmq<(#GW<4RHWnjahh?-OdNmRuV4nF#5a6dtr3e<8)tDSE|8Pge;2#t3; zez=uLL|zY{BYP+$D`HaP7psXuownrei)$lW=rmdBN`vL<>WBF_&5nUCr^9+%yuWOB zTze!uQe;K$@p-%qFQN{IOO-K34G~u4OP(aja106r|D;kBbJ}7qG2eouFkboDjKc(} z==0{>tW(frMLQGxO)UHbh`dqHU;3Q$GI`U`vB^yruUSZqOD%&`lrC)<2WJI0E6R2X zu%vX&&*E$*a>mDhcv7QQm#;el+KP}ljMmQCSGv+6EjXV$yXshyVlE*3qwWgZay4i? zlnLM?{k+Dd=Ox?kX;>*H!U`NapNwEk61_*2IvZR!Hp)@3vw_=WlJUO)n+EpMHF$;bP|b zQJz9{L~W_Q=JbfOi=&lei5<-RT)Vl6bFd_fc+NaUN0Y4zQl%$oFC!?)eVTn#EUU4p z9%lR9+y_+ZCa0NIRN6DeM0-CN;udnQbJ@=kFAe>BV%<-%s~$_)=JDRwT<+P7|F)5! zSEeSQ^7B^TX$uj$xR6u=y7dJ_B`fkg;K^?!`@|*BHc5)#yjN|*-^rj$@ulXuhu4v6 z!WI#JrlKj+o<>pN@mJ)(q^}g5Fa93=Bm>Ir>o&UAKJ-}Es?YlQwBli{+cv=(>4T(i zI=`F$DSC-?)Xh4$L6S4a-LF{8)|ro_(;CM1M!;+#?mpC# z__9GqA`tZZ(|4T(TH(geaHQ?ZW4#Pxdp>XH9Vu*6Fu=s@MLi~&w`)}>IvcH15$n6t9WK=$} z1G0Op3%~2%={<7)U84KN_B)X$A^d!Zzt0!Nq#WC3*YVIJID;5X2$N6&UFULIyVv*r zpmo?n2xjxv-m9?u=@&@o!KF;d+b7U8m@fmhS;?5A)1CW#jce84bts2_s4ZC$Ej>k6 zhNuxjYj^gCeu}-O;icLqTc2rgYz1ciz248UFi4v6(W$|W#fR0smQlJC>PKYgED z7ev#HNm}SnubX`Xd&p(VA!AN2j3Ul1sH66egq2{@Ir}7|C6$cE_&Y#z{6`rf0LvBI53R1Uj^Ii2T-KeCUOu5o`9W$A?*kCn+zf}r~>fx=o@0T?H zS@$IgKt#3-oNGK}p+xq5m{{K(orIqATHqH)o@-BnoLyp3uoewn(W(SCE05${g#G*| zQi#9+ds3I}I^W}IBsqvCEV`-xvyXGODlEMX=R+ym`ICCRkwE!L)#D*s!Q16lj}MNp ziH=D}IsN`kJXw`l$%i?m;C_LOG1aEO^)HN#DUf|WiKiEjX~|>+OCc_B))8B7v*S6J z8eu3?f!4(O!mPM;KM9DcOewZU}&<}CPy1gi{ zd>O6!6xlbHmWz(u4<-Hq|@2Yu!m09(oMSz21+V ze~r#tdg^H;NU&F$Yu!&mf6Boq2{BckR-pfr!yN7+WUz{_9|+I$-Bl4=otVkdkcJB?)SdJQiqjgBsUI-5`5sW~`0wE4^M^A2(H7;ED`$?h z5>^g)r2|<@*ZSw?tuSNE{y^gMN$^11rKP6d#K~@m_zgS`&1fpUNmtNcX1PU(9e={d5Ys_vy%`_C>Och2?W|-qO{6*r34zUXK{ZJGnu*o0y#{ zO2ysc_@&tzbtFYIVXIRNuhMm4S=^(}1iaQ~Hy4|bbA{&?x;XBf%7(5q^`BR=KqxF; z(VN~mPBP+ZtgphAq)WJKQS>?CN1a%893MT!y+}=)xao!m*RNc(n!F^!i`VX0^`{l5 zPUOldSEXYgPp-%!`_0w`zGN<4-w;^b`M+p+%b>QtsBaX97Ax+gMT3=MMcd#MC=%SE zc%jAJf)prDixw$?;7)OODekm*ae_MplDzz%d1l@__kO-#_MACq?{${1-#WAAFueX* zy1@>Cq7yEMQ*Igx4IWzm(gL$PtmDt9w*qAGy~{dPKdL)Qw3r9>{5Le%)%m|egF349 z7pIku!k$w$-a*W}J0q;|FLOF8ky_ZSCDr7Augj`^insK0stdZ$W{)_3Xk^}L*e%<2 zj-iNk5B{Eb1aKZLhb|>Q{MGz~&UC1w{ZWB)svi+WHi$fP>Y#P4`~GK4|H*kXt5f6s zROpBMnOO;3!Hp)HN+hcy<6PBVm+9s4QH)=#OXw1#sQKX7g6oCnasbo+z06iOFPd4{ zzx0#{{C`~j|Mee>|IZIK(bVua4CvMW8CtEMI|IA-bsqoE&??Ha2Gt%oa(DAu{UsN8 zq9>2A@|x|g*VDQT5Yl6=p&fy`3*9if{1i=Qd%B z9y-5i_y=wq?a{0I`$6=|gCs@C!m>J1KXk?l8d}tBQw_U}S9Xv0xD#U)h01gujM+-J z*x@YT&(@c(V*tm`>^a%KyZu{Z0L(>HRC>+R8NUhd>+C;*MO>a?++-MXSo98akuDJX|C2xCjCEdQewdOb@ zb5+eHMvOM=x;ihWL3ygM#z6S(bi7RyDSsDV-yQs!SP=LgAWvkw`+o;f`#QbN@ofGi zO8n0tYN`Y)8w~MA8)nw)$@UlYS*}L8$^^UyDS1jPOvbk9W9OJLUC5@j zJbZ=9xgLKjR(k$)b9X1A{F@C0)3Yrhm}Wg!27M7!y7uUF`%f+kA^*Qi>b4E>CUXYfH&vgIlykAk7mLv9gUa*;sMM)Dt7cdWHIL$r9QTiuUoW zD9@{u2Q~kd-cUb{Yn7ADZB-rPS8JJM40Ahv8*BGplHR}R*;q5CS0VMUeW#MG(K!7f z1KDu9@XfIx{$$O+_4X!0uhisUO+KQWU6^x6ybuuJ@k>$CoDJ66rkd><>WtD8^$la-q+Pp<0LYD7f z=Z&{?*`vU6P~EdC|F?aT(U6lBX@QM3f7VjlvlAxF+)}!^SGTNJRbiwq3l9gt z|0Ux~r(2sWVUzfQ)V`#bA)Xq&x>=s#(ARnMc{%rX;ixy_ktY4S*xMUs*)xi(lgF}O z^|JIcJSPju(CSN-S*Y8jQ-SjAYHBT*?Y84Cn3cK8D1NPjDL^DnCf6Jrulk3xy%amB=NdY-#YT(DlQ;t z3q4?m(dU5C5*EwouMno_9IUe{Y^RbWEqv1P2HcZy2Qzxbt>qNzbc3~y+covUScm9L zu$ms6C^}W*qY*p{Fse~>lH@)$_dAZS(|0?ocp)mxZ(_XorP>eQCeF?SjyKttqr8l7 zIw09{x?1oq<2sWILTP)x&S3BP;zi-rzk;Q{yRbRZCJNDWZK}Vu^IQ(A?)VD4oxzpB=@GZDI_P!b=w0&# zK6yPj#-c+I=6bY^&ILMWE-ha5cMI*w&1eoARiN8B6U$h!B85%OHciv4wovKaKy=+4 z@qB3k$>{MT2_j2m@7f!q72y-4DBd&7tDtiM6osbq$w~09!|4Ys55mXIc$Ig!gSXCB zaq@G_ef^6FJ;ov$C(n_S!<&uB0J*OY#zdM6+Ifqs-F9rHM;K1;j8>K~Anc{rWT&a@ z71HfuZi(inmQOmM6EFvDIm)RAtZJh+?Y9!(F&_RR1q~_cciLa>-u^^qRmR- zdWJ$;PoxvWTKAkV2;-LnB2si#PjuaR?G4!3;@9)@c6rX|^(nFMX$wOIk? zK&P|y#f#4s*g7q63@RF0K1}lOiHa47D!v9p-r107E zVzcxPmmd>Vit2_@!XUSrema+=L)_hItpd z+U0;3YL=xq4X^BRe1Adc!eOk#=WI}Thuij{ZcZ(~;y74g?q`(WlG1?U9G-m};zpu- zJ8clU>iXpYYaSHv*KC-!)%F%nXJv6=Zrn^?oM!%de`jsvDGsV#ADNsM1*uB#Yt1lE zysF-z0m6^|z`ZsbTKG!mHTlCKfm6lx7lb}@5nTLLrkXwk3xa4fGhFm)cITGv#@YdF zHU19QY55_voKvK;OR1g7cqOE%!f6IH6NlE}frmSh%`yydU!C-GuXQeHo{-IeMpDr0 zMxlN&)hbB;n&F!hk=YnM(E~t{T$$vv&z`LU{~abCfQ)O;#yCEKS9Je(xU?8jmAfhx zQ(B=n8xwO^?0tI5f-74}@8!~d-;!2IwA_?@jLf)uLXL80nMf!VnH?!B zEtrdYX1@*ciHYklza?!3r4}#Bnxkpq+q!W*kkpL_eN>p7eONr=TkG~~rYnkBo(KG5 zQ8Zd%&L}7(4xvDFbsibwN2ty+;vIc^lSj%43-TWU7Hqqt!X651jaSspj8>Ahj(0_$ zpv74`*BQa}3Ica(MO9I=$J(ZouWF5r(rVugs;BN(UD42T#?N>yOOSgN*Z?XZM zL(e(>-^&_!buc?cBXaD8T7dQ*wbWvIP~VuoH_hjDbY z>52a_!Z&&~n8)>*mgI!3ph6->aVA{-Tu6u?TAC7wuT~;*#|hsB++9r|qK}zNh(A0kX#uo#)0+YL}P{Eh^j2X~jw-?eTBkhe}?^%Ng5Y z^D$|wykfB7Vh^w@g`BQ0>j>OiwuG-;DA!LEOjr1im>`DmjI@BNz!PFxznWz7O8>gy zZjbC7EmdnS_Me)o%<_#QiE@DYpq8)ifG1kjrm(j3%c#QH=>WBo{x5NwtDtquK$`KO zi*T&qB?b2c9k>2CfH1O`P%~9?b>%h|4hT>Lp0H#I%lbzC^Zhs%V%AZrWj0#&^PJ1s z8x<-jTtK}WvkyQ8rJQfPqaOAb+0k{E@fFQeuUvY8-JafCpZu6nzns4%Y2ra}(zq-H zP4+Q7-EOCO-mT>zP_>9N+)8&*TC`{lrD-y_A~h)2Tz&tak~_OT=5y5#$!QJ82pL{Z z5R8_+BtwQtG;72YU4~KixK#=Ql*fVf`}n4hY-u~ z^VL|PcU=5tX0pD3UOKj9Z$UG&T876Lxi@8K#fY0AWhgYu;W=u5rY=cOfXUuO5&K`x zuX~q9v>bc~Cob{Ty82~(%YtKc(ZSR)D7tVN6Sb1rkrlSSHEn_OrJWM*et7?&DYYLGG&uerpGaZ=i9&;E(=Q|LgMR zlZlRnc6wn%q0v>P9TfXL&QLd64-1%xeG5S6w4u*Y_nkYP6K@S=-sdX9r(8J%x%8Y^^+LdYsj6Wv6@)R?NxwuIv9&-i5Q@9KW}cP@ME+B2s)ghFl}%Ni5I%Kwl}ETT?y#r$&xddp_JCFEk26e}PJm4Fnl%xP`<>RJ zLvsyJ!bW|qagGL~8JpksRQYUPJ|AmMm+(xcNv8^O{?sFtBZdR8Gm6+i?Y=&G&BaMv zk`c1-bB>XWlgiv`59(}Jx0}ju37$@=Sr`ES;mpI)+7Xqqkl|K+L(v4jiy9ulYWsLl zqN~Hn>RGhRi7AWV+}SqHDe=*_pMFTk$&|#yZ$-|tIXiZjb%i_H3~T{wcM;3Kzw+Dwt zuc%4*+IjaC4v*_yDC;L31flaQOV#eLh->txpaPn}ja|CzCA>oV5Vd}CLlR(Lbe+u*H z<9b&S!_+V}9ww?A=AXUI#O*?rb3u_1RFeS9=4R84^G`~b+hR|f#8|ChzL7I_pIdzb zSL;A83oX<%--05DFUW;nhwi*Ock?3R$?4c=B0Qda`0&@!>#H>Z98k?uUeC`thgxGa z`%3SA+>Ed49-6CN;w@c_&S-7=QsFH_%yvvHM|AlXe8q+vOv~0@Z-K3XFY_x0aKesk zh6?xsHT?01;?Gsi6I<=(=d$okEiTIj`x&cx7>SmjCd0P~I$bTIA^jnw<`CC|)A!Uh z9Wz)>?g6=y)^FR=EAo3pYi04Nc(z6x6T<@PlhND zXe6nb_%BMUA=Cd>NwtqMF_XWe(}jQ``sfdMHGF^gj5-oA^=g8K&(qsI%AzU*Tzt z|0m)l!40{8WTZ2vL*Lshwz|q%72g0;bmillQIS)kqYOGtfw8*uzI+PIUMGhqM-78i z3_=@I+e|Ix3Hpw|ibQD*tgsw=WAJwaEd!oA9cCW-4~F(oiIc3h`(6zX<`tj$-C(qn zuX+M5gt5LDpJCmlIx^%1)I6+fC^*IkIhR&%oK8&od?-+y|L$YKQtlBImSJn7o z;;MVm?_*K5QV~_sIhzSL)jHE1FaKm&u16%T8|2*2;+?QsYGSbPabb^?S$DqB`Ox)! z>Fc>FsiZ#k5wL`^3wnA!WB!%R%^z=PkO_0t!Y3@kVxfw~0Vn*+aL<~bb2*r!FLx<0 z=EuC%X&qwk_6)65clN2&sORoVs@>2JM&4hM)tdXH#t2gqooHY}Rbi~JchT!r%7qUq z-&$rI+v#|j?B)ji%>!Y}k|QiP>YWl2sHqfFE4Q|T{I(#!gFh@VExVh&U4-gVoTm50 zpM{qDmMuEZo1ll&fxRchHar1#r%%givY|n^X8rK@1h)-MEL6N}+g%UN&Ot)$21jD; ztpR2KM!%1x!h5&BziJHFeM50PrC%=C>SiHR*2+n9FsP#0H2Mz}J!m6#fcn_5cp`^}H)+;3rykc=aT zv^Se?1B5B#Mt>dMcdK&p#|JDPt!s>GpC#v9iAfX%ly$i%9O0@uBzW^&H?T-ztEC0j z=J&>QSnSo^3|caoUw77dN)L3iE6j#0cqNRuUNn>lobGDK5T|xPDqbNiv&sM59aZ1k zElbi-NOUrs-NKzrs|!Z%|JXHZ@H$qPk3>%K{~KP7k^ut-LyvRSO{r5TZq6#mqK99X zoo8oioE6qQITi824;10^d1dyzv)XBp0v{4&?`N@cBDM0FJ${sTZbLt?sUlxcRM=sy zPShi$l&IyK`@>DHccx7C*?0ZFAG)J2$6`0e^hkN!uN!wpIuO5m^{yY_4Gb-GJ)gIy z=e_^lugncJB*tX=Wh3hI|LUR68$#&riO3oz%OsVgnq4i&hN@%VxDvI=S@u?)V2e6l z)Yq6VONsq9{mCp?+H(X0Wq#&dh%fcI*_v)y^G^eq?yvq>qrVk45p6jSBRnv^3-f%R zqLkJ^JdGPx0Z(%DA2^65!_aVM%sTTWbe05qtG~SWQ?wWO(H%~_qEP|R*ZyWPar5?r zYrSF!<{{tfzPX^A>2=Jco`HqC(`)t%-22-WaR5m0w-#T|Kv#N|&cW7~iBr>gXIuUI zX5+vB{RH>lpj{8=SJ+ZrT7k9+inS`6OW$p;KMC@emzlo;wrB8_O5Dc3#oD4Cfe*A& zT~=fBI-k0iyp{|+6Ek`^AIqY5AKv(?+&+GiHB-eDFP3XD){I8Sv8e^Jty?AO`O=%=Dy>VV8ikPo3uc!$0wTDWU7T_^bf5Rh32IqkE~eOwO9)cNF)R$no;FWsg= zJ_X6TCTPk&Oj88@wJUOl8UOuZz$XScK&s|Wtsx23O`+ZBJS2?_S}jnsIIK zy!h`NWhEqjCPfxW@#m8rJ7e{B#!F5wr@&y&fvXAFh|iaoNbFyeAmR1+!`5QRsR|cA zbH4L&n#|2C#UQFgT!s(JwQsc6F;D%7j)uQG&T7Qya7BoOR6u(VFzJ7<7>d}&xe3jE z^WG_a;&Z3x`jM(fK?_Qp=}Lou#3%mpKMpMi^}XKvFN@=9jJk>+ zDJRn>U_V7_Khqd3fJjpzB9BXE&EXTkx#m|SxBi*L?UiQq@w;=Iv(DN4os7Xsx2NNC zBUO0fF4B%^JmOXOYZ9`NKH&!d`fryL^D;gqv)>FC&{^O`e5HV(6DfGk8?lXP&TgR{gM}1_XS8rQ`x>~t?z7|86+so*~SRD}fO0kN_3#khtrj1B? zet_AcI_vxvB9f!DapI)tGc#p4Ye$gR&0&O$jM(~;{$GtO1is~$M^6fKy5gV#&ZHKm zI0tcS1@^hO6a}H$rc7o*x0P;3jl}**oislUF2ng?PcNO~BhMc0GO?z}I@dpZNw_`= zL*SFLs?LDKEIJzwB#voA^i|@Q{~(8!hiw@X4|2ShT)J{^UP4h%dN2Ht=lzmsS_YCm zW1CN1>rRl3V0B~|>i`rbUQ;abla?svZF}34R7Nr4r}<3K{nv4^jsCH~>4rF-P}XDm zEBJOK%bZJ}GD>$dI-Wk1RgeCzW^eDqjY}U}E@banOk7-`-vt4&nA2qpK22lfzuS>tL_2_n~+W>C2FJtAa!PX+F$jEYg2nEYM(wDNJad?J&J?Sv3c_BV^q2X_FR$18}N7>?&VkXqg>sY# ze2vi-sw<5_(_FEQ-k4&@^`x2tVi%?lW<2fEnFh_dc7=-!`Y)`C6XDAvwfz@Hl*HB@ z{}c7#Fz9~cpACf3W#V3R0$@@Y1>cOkIa#Gs`z04IDm~?Q!CDEqV(n0Ua(k`V&V_qs z`a~AJ9wmN78f3@d{+LXvPI`v1E13*5F-T$8O$msein9f#Lfm;yWLP9P<{)QITr97p zdMvErKyveIY6B^&=#`WVs~fVmNGg7B7K9;-###Rg^1%(P6(b+xqO7>Ln!29wTmPC( z^3o}1<$#TZpkG2N;ce}A60)m_aL??s|5BWN-t;9YiBH%8phb%_bvB#!Pct4#(`)#o z)+|VD-N5SEK7dhk2eJFjuqqMdd%scGISB&1-IFaBmcka<^ zyH1hes1n{E!$dETy#WlUxA|L-u{cb`H%BD=XuI@DP*johUXK47)_%^f-m_h%tBI(& z-f+~Jpd(_?V(wC7m!&~7{2)g__r=ot%w*-XDjskjn`~&0I*f^?B^1t4PFJ(j}sI zwIQR5*?dWIdXS#L`R9sweH(62?rM8LRt+R#5~Se~DJpiQ(WHPn_n!sThuY9_JVjm{ zV4m7FBkV$}CP8bcdB4HY@4<-O^+Px=r2&qB_xMqIeW`d7TlTD2UU^ygy=o$kU7ilH4{ANGxG1XOv~&_OamzOdVm3W zz)OuN%5D^N*jw3b@e^uGnd${(30=P8;G+{dR~~pNgt{_H47|B4^%8;)S*`!Rl0*F) zfyajFYn()|KS5f}-;(a<$jAJG@=V)USP5Gr<+jSoaLMZROr={xuAsdIY%RzIqlNGs+frr?^jId56uVe4JETE5JkHM z?l(r=jXa%a(SMAn=YR195|`}W*ISv=?Sg1T|EW0%l_f}Ukao0B@bs7;WeHB7o;J`? z3fGU;>Qq>#Txu9p@$8$R_YSkEi9$AsE;013j=73Od#l=vo{WD(87ehSD|%%;7F*R% z>Eo`RQ(i9b*7pIf>_zm;66dv@M~f*Mx1=V^i%XpKv?!6frFrTQc=4_`v144x!rNG1 zq%gKKcRfiP1pK74i5e5$<*M0*hR&-?H5Dx*Mn#f zJjRH1yY*tM$L@QHIm%XPuReLI>X@?5s2gvMj_}KM4gki8(HV^`XZ>)NtnyFkrm%}QiVL?iulnEO=f2y{09S!D~&^KvEd&4H)d^e zzoo`lUK5!JN-$qdOmZfG>=A?4HS>F)ygr0Yoz-0b(Vu2A-GF9E0M3&{KIcY+x|ZCys`o z$EIy?G~7{5rJf5U{g2*C&HTZLl*mRh|GnE;&VxFVl_xIbAg7k+EaAGn?`|T`1SwQd zJqc2ynWNnWYsw&N3wxbN4L3n$JUk}o*~(_M8PHB?$3YbVvNo`N9klaD#*G;`fxQjE z+%A{`=B>^SB)OmK$77fRoHQP+{yzF zFQyE8fYc)5IKYKS$!J1Yx)kl2Sr&UD_>+|CnqD*qOso^@LTyX2W?;yU)HSc~hnFSl zJTPaIX*NBa0_fcS6qF-Rs$`e}^iT1i{Nmw63I1ia$P#OL$Mak6SS8Ay(lyWe=MP(a z`u>yHfi4%^7EJu@i_Whgq5`9e6Lg_0|0|t4>pVtH+G9f4o(%=XHUcxojQc%EHZ4j~ zV5DYY#j@^+W+kDQ3#_ShE+WL5@&iK-&`QQzlSXRKs^SrVZM3fwe7Uiv+^u(+1+O|P zg1_14d4#wn%>W~Mce^-;dp$x-pe6vWekk};3zX$TjRn?}zf+vH4D4A*LUJcu^TDb) z%E{G|clWT^eFB5sdTsU9|s88yu(VuOLt;@ zavpFcvAXGjqB^}%zY!r$#t#@ZW-u^9U)+Mi828uXkGD$BwZwFrWKBSXtB` z|Ksc(^BtU3JsB!lPPj{9!(j$3m;OMRRv+8~5s1dD{{rTfrLb|ujbl+yzSG&}ES7;j z%)e+m+C4~l%9Aku{lK=-<+=}#*zaU3dgO(G_DR-w)Bvej#DYu;Ne1Fyx(`ODnt z`WKp^Z3FWi!HNRoWwun|pFiyT2bDGlu2Zqxn9XmMAK(`vlJ%mwC!7H0UlgY*r8>Qx z)RaC@Rvk`@$K~0H0C+^B8A5xgY2YXjNDt*uy7M-b`(F|R1|ns;7|!}V~yQd`fxK*72_6ijzOHP!JFL`w=pO3f!c`@ zV8e)L1z!?YyTfR8UgLbSrtEDYtDlw@U-ZV>R)cU-!{`)*3$Vk%pF%1r36gVARHh&0 zFo(xFESSb+#1Xc{$rKG6iZyiVvQO#ybB4)2OiGK#-+p80uM~sg4L>n+*L<6{UXQUY z-}@xF(LntVygU2~d|yQIp{efkM`M~V53)+N1u@boFiz6z z$HcM_8)Pylot<|C@;H_Y>33pL=qLh+U_vA_PR{FeBD-G7UuD zK(0u}pE3j7aMfXYulUU(c0;ckk6uBlhWxYFxFYP`Y0ChkU^GruGx#A)Q6Bc^{?hqN z#U#)SBo;}4EgIdrr_vnDV$cL`i{jLCcB*)EyC-HvytIq=vx1DT0{&IMwbKt&6%O#} ztRprhC99`*@x5Pfr8!lR!4r*^ryT+w8%lATVflyodZxxQt7~9PUnhd|@i9Ep;+awc z)8bwBn*^$UniOGf4z$a`-uakfZx=&4Vnu@sAp~ZGNV!wL&2zjNPjh5(nrMZ+@C-&& zb`z>QR|F|n^>;Lmh0I+z`f4R?$>$FQ)LQz3E^cFfZg)X`gI&L|QhCW@Nux>z36rdC zAzlFCiM4$RbHiHh0g|t)??AIKZU!(eF3J@9R5Su5iLS)-s`kpe#6Oy>Ppq~H9BlXpo!rsE9b zBw*da)Pe`d0xzE?!EU=t@V>r)`G{eHYAj|!N_J`NKFCko0(!YcIRE0c#&QRS?q94w z*L*7g8soH&!(~DNedL|RXCI$FF zx&K79o{V&=6vpJ5$+SW8CSm+_SLz27^fPyAL}a>cn0*#vz$B-Oi$ zV>*Z)8*mv#vV^2p7TV z-Gz01R#jRB*;Q72Gqu-mA&0p6kW&#?6ZGVZr;pzIPtlc4?Ug;{4GjT+5||Klwq6dh z#yy7ygF5iy9#J{#dK|?>9LpP+I~?%j3-|#qK}3)6sgHF1??+ZkDu=ET%wO3fA}u6v zai1;}GAIc}i#@07=Stb9TTy4G{Mv`P3qu)tf8^=*P(!cNaqBjtUS#~-dXAJM-FJEr zuXr~m=J(Fq+N(L*frC)>#C{4Og|JcG|{4RzxylML?>MID(8PwO_d*=9PqgR3W z3!uJ^a8z!j2MbVoV$X@mSjSs2dr@q?z;rVZe(amj=z%a}rhL9rzMs?eEb(!tc=0Ob z)2EY+ePLMPd;0CxwFOtyb-YoV9*aV`nq`ugywUi$8T3r3n$MPB)k zYxUS)rLyR{KV>LmO=g|mg<%a@xhCedxr1=zRmKJ>JU%^m7}HO4Jz&CpnT5i)2{CZ+McyM-zHV zZBgE=j)ObDrz5Jb*@b!ng!swLfaGM1N;CVeCJe)VTl3@L40l4P?8^#VG&^2s#-k~d z*Mx(}DXV*LEOMb;{N@76rA0J|qAcT5E`2ocrKY+Jrk91+m0^1Z9~P1?aJqbzFLjPd zZb;Gq2e(k?yYV(2uA9sq07NY;d6l5cMfxbGHYM$Er|dNmXUzd8JPFl`I|0)h4eJiF zM`DOUI!cU&ST!iLt9|CX8Q>rL<;2UAQf0H4Umr^(_le!T)$b9GUi-L0h3YOKJ$T@?UhPnxe)NLPKB+L+!hsz5i`>iYwWX=f_h|N8r_7io2h4 zPL>nu7KdJT#-67B?OO6HqXCno$uRS_o>LYt^D~bvW!h5u}ZY zJpmh+eZ}U%Vb;Ph3(ac{lC> zoD(d?g$Rchsq*UqTeG8O#a6To_Un5`Nvl8&E$)H$ukt}oiH&bZR#L!MTO65XF+mSO zvzlJE1utFKBd7TDHDr9#6T%L1#uUq7uW@l>qrfu7kky@#b1BWA_`?^{zW>wT_(G*f zlXw63rYSTl))GfE&p{b=Co%+76f}R}BpEK61m&UxIJ{Q3i=Ze0bQSqF?sNiwadFl! z$bQ0qDVMl?M`jU^l}%T=?WY2>)wSNd|T*f4l!E1%As=6?3#v z_&pzTF4q7#K*qoK*NpRYGtlRNWgif_z+`KO1<%onm~{^vS9uAu$4EnAmOWO#CVm&# zJMqiGZ0~(A`tV#IrgM60ocEAeDNi9b9n#tzo^nOO1`x}Nel{7 zJ0CQB@)*&@cKZ@BQ$Uqht-(h|U{8AvKt@&Zl8Rg>!=1dhwB1!r3Cv!>v=t}~dl%(4 zjC~=mtN5&#yvY@LL3Or;`caohqQNTQXLWY_k%b;NKq31rRu#WF*gJtE_i1|qb$vD; zE|lTy{mGD-r%EW`v7|ORHW`4GHZKEZ5Rs=XLSWqZMAghIn%8tLTk5}Ec4I>n)#;G% zzcBnus{3(!lDdM(&7mNw6~-#j)nvCT(mn#DT*(Edb2B(U_b$3+RlcQ6s-qlm)<+yU zo+qF3ErOZ{zBl4!s~hutNlciu<#SkUJn4yd=&dOt7>&=N9c)aiU^$L6tSbECs~VzcQ@v-y1f4%gnmGwF zKGk>z%+V?#!fl!Y>Y1%%U@f;6Y35~NEn6xu;Wj-xxpD;lPZ#9(?p@NLGq!Gl&s;wnJ(Gm$ zcBNi0o;>&3Z0^eQ{iOb^z!d)u+Gb$dvw<@VBbj!fCK-;IN`~zNSZ*GrB=Ol^{~lk3 zXV#lT`&=_oI~5~8S}3UBz{&jA|HVurn~Rh=T;ECdd6;3CjO@x8-J&M_Sq*8E9;mE4Hk`}3r2 zIPO{0xvk@au%=V0OnI^6hDtJiAMjo}`25&{;K8W8m{qf*;n`OzPcy+w%vsRb0@do% z(G=iXxqFkijPHNMQ-Kp5*s7g7QErLz2VYRlpI-gQiaIZqX;_TL*p_jVS^yd4Eo+!QttcBuwn?{U-pG5uM(@|T*cMqnsznJE6% z#NC?V_vBLc;Xpnqa6F=jxS|+)8YHqCj^!TWrd-wN!;Lfi7|iNV6YJ%hO?RTs31!g% zCAG6Vh{zu%&@8cejOWPO)}41l82x)EE6PEB*yJjWnLqO?idm5t&p*RnrX1uD+5Wl( z7DEi+ZmjzN4rTn$41ASJ2 zos0IbF`c}x9&6CGuOg*pCSRq4Pce;#C#MfB%eX>L>Z4BaacA#81A!LRrQ6R-%AAY@ zY8L?w^0oDEMt%#j^;RVDNEU{hg@X_O9hn}zLkIs}6qY{nv&a+gdopbNFu!k!`|P18 z5vxSgIl^0>ILP0R!RTT$c$&FdqA)8H-;Sq|94&rJ6JMRU>aR7c!GPNU zXI!Z4VEEyg{iFOX{9e>0;fWtoB8JIaLBo7?x3bs@VUg5}n8W$apR8Wb9HzymFE8On zu^mYvjB@vQy_a+uZ^uq$k=W9Z2uuFegttN%KRpXf>h{U-K~SCt38 zm}&LszCUn=A2HhlW%vOgX4BQKt;^7CouGC~jqH$%9#_R*5^H__-w!;BKZ|nwc)Ag9 zkGEXT0cvPA;sE(a4)|kUK*GZEgdSJruwc4{R zwaY?-S1c+H#SzoHSB{UoQ_>@wZ3yPXW1*jRCBRVg`T?%^%5Y6GzHd zyEpYlDOK{C>mK?xZcduehV~9*>Xl+Z)WrHw0hd9sTI+{^ zJx^N&Pose0AjbgJ&O-NUs{?Gtua3d7|3EE@5Fui32rz8MydQF z(OU3aCRzAm70@5mJl>4=S7>pj8>UPAHTMP!W$!A}?*q4ZQa%RYa8`<)R$t!?Hu?$#|le*R8sqKwV4R7N#Iy_L^uTf!f># zbcTsPp3cFt)A2hRGyf9|x?v1RosUPu`vmGtni$^@5O>QKSwwj#=?ojo4rRqg-!f5Fy7{Xwa83?_+z%OV$Rj=5iyr39SRy4MqkFX^Nc*#duUz2a?02h+|RlM2St zccujtwqGlKILCac^~D#q7@&52u`iwyDVc~l241%$c8?mj8p{SqxQ#I(z0YEOKA zlrc}supVnzn&b(iuE-M^sx{nVRv+!=n^!BBsZ3ZSaI^`Ze?DkBCt+~?pi^7Qawamw zViX=fMOX2(ab8nQ;6#uF^o55vRdRqA2{siurUeL^j|q?qRJDqdWH}~w3N`tUaFA7$ z2CuQ!|8$Afh^gIgj<4C*`bYbb*ew^!L0f%v<+Fo(4c^_IY50DTa$Zq=s()1bnjA<&~zH@e;o#4CKv2MhIJxCBn<`!7H98% z&?c*n9$=BVzwwUsgXzkby+8it5_^}f_NcSS4X*j;6as)e&+23@D0ha{vI%z(mw8w~ zrT0S>WZyz&&?i1hWXN$`^*JM}i2a|;3OX?A{I$n)(r{cp33cATwlMa{Q@`7*KoBB8 zUD(7p>=V9%*Z`A#Ks^byaK~8Mg~*T?R3ckaG4G!xvEY=;BdZZGE2-u~$Gm_1$J+R+ zG>nOPx^F*h<@?rg$M^Uy_|!!|6=v)az`B(Y6IZl)oA+0K6m%bPDn@o;w##xP{+!Xu zGtcMcAi?W@w60L&T`$eL=skDh4!j?G7oQ$iMM}rSkr$g^PQFYN?Ez{@w@;Y_U4H5=fv~0jX0qau`W2z-Q@6a1xv%6~mu0#62nQcg zSQJ!bs2EZ?txZc-!D0}O*v%}w?q0#Xnn=i&E|=+;e`4+)?OpvW`32J5l2hh~(_j-S z1Wn5a;4XOSY?17J67ZtCFeo2lri_;^KgB%@(s}6~tu(U#?B@6dQm)U`qZ*2;Df_O9 zriJcDM}W$%im2T1l@J5Rlzc@bsmy2Jzv`_x$v*j%3jv0VMct*%X0zn!;&g{wu?5Z9 ze$JPK=ZFW#yLJNP36PtSRsZ@P*xCz)5>jWz3eN4tO`nrhso#D{8D4q?BOCSJ;_QqSb+_%xKw@*Fl-!SPlf5k zXB57+3abrUNxJO$eJxSaXoL2W4JDHjNQ};jP}l>|E6sq!NH}+vJ~IBJ-7m?0M>L82 zF12Ng>I*{c(I|B$3IaX>`sC~2oVYS8;?cb!Vj<^% z(eml)QMO6O&9x|Tc0BrCblKFo$$zKB;;AcoT!#dxOEctz)Dl>0U7#pJ`8w`d&(#9{R%~k<&RsW_+y}x9HTk)FkkW=yr1m~2d(sp&S zUy_hA7(EuaNuBqeul$|)UITGliBJXsx;y%|BV)U&TMBX9omuRYl(p$@F*Yn zYQO#+dh*5ms&W)2eIh5`z+{-zs*HXb^L)FH&tn(@GLphkB{A4R8ML=K3M2jllCW*H zf1OdO&h82$`wIsEr4h;2yln+kroC^v!cJ63I~aTpc}GN9Me~WGuT48RmDb?HuP73U zz%P9IXvR#2VI4lBN^kqVwhVPuqJZD!vgqK*_lQXENiqTbI{W540?Uqy&R@ipNB1Sm zLm~PWIPPCbB&3L^WQshifkt&E;CNXeJO?c2FQBZoT;3K|`K=~5twKYSazC`!w{H1wHcAc4cM@#+>DPI}YW*3BsODPQmYY9@^ODSF`!3nM{4uuMo;uM$Q zTAV^#B)GJ=6fGLGxI^(^#hnmHvibJxIeX6TzuiB%@40hl?%bJo<~~n;ywD5E%d@@> zsG@wYm7gbKhq6>gMc9QO|NbT5-trgiV9!*Gn2-yOOb;28cH(#}UfgTS3?CCdkz6IT zpK-a;u!U|O9W5HW*w2g^MCHD{jJ*>cVS15!1cR2!RlUGhb>?MO;NbxSexC-hX0t-jK0P1c^p z99mA@1e<oJ3SqeY(gWW$BRFtc5Xi39=DfH(3jp2N|0QV&Z0@u*A|WWMUK1(NkFD`wDEC5yT3_=FT1wVOK?-E4?8u_bt$ z2W<&oGs@%QmRyZgar_F=3hh7OxBZa42F9_B&X>_#=u}xU903rK{__x#o%!(LIUm-h z4re?lPG7L1Rb5X`QiPB2JjsIjQw=w+ps78yv3LSq+a4GtKoJqPZ2G19KiC|*i1bU2 zZCNv`X2GXUAC`TonA^}(oO#e$>4L45{VKrmUYGybQK72A$py8oaO+A{eOL(UHSn|3iKJB!sBO2D=k--QHe;zKO+6L5MM z{b2fJhoOI?H+InhkrJ7nNV?VoWJX{`I~6S-$bFIrSt2UYUTys$O!U8=7^V?o@XV0T;nmmb?+ zbg6u+*%i0iV{7jPIXPC(>tmlrx9=&KO!sl0I+;n-h#G9xH1G2oC0G|85>{yxLM@_~ zrkYDf9k7bhX`L76nMV%^o_6Gq+BWe2?v)T?j^wk;+xot}#b?x>!)#?1lySCes9fox zN~D!+WT=C(i);+sah1)sFNj6$#Jp+LKY`gzJ`ydpO!KZ<>#KRBkEKdpbr5USqL5_a zhw)q{!4qFz=A|Gu%2y5ym027hbb0QBKz*u9KouXup}-V{S%riD+MOL754YSo)~*NJ*s&Ujq- z`FX0-UOT>!(Sy&5TSu>CYikde@|T&pG7Tp3rUh)KZ$##Ba-cPpm< zXq4qPlcR8@$V|&`huZxURB9Z#`?v-MegXC6d*_Z3wX$=o`~BcJl9c{vr*n0tM^y8c zCn$EHK!M^5A)iGC;1Ye$tn|(vbQkMsSEr7(SR9@#dZ^Fe8IYP4ZBaiK zqj$s5nU!~D9(9tHDvP_~f>!baD<|6ct-Kau22^e>QpmoDrp95{wAldo_r5^29uZ$Z z)=#?^%(l=AMdm-J=D$vUCuG#_4G6NW@-#IoZL40fMQc%`cwkJ&%A2;C^gH>_C0-jit zv!}i+++#hPzF%blsXI1jvRXM&gXd3gu##Wf(A@uqT}V9twj}(;yixa^3q;)&El+R8 zF9SBD78lujSVTIWP1C$6t2fm0kmCk_*i@2oNMNMq*c!x3+#T^oK&h2-j#62b+wmN z2HCj1&cc7T4ArH@**HCyVXKR$ieyTeO)YmFwr2^@p3NdOuexXSQ{NJ7edf*&gF@(J zZhu*GwgPkKC1Ba`!ODw_lrZ7R6eHwBg}sQd(K5N~#5lH6(zEK}YqNj6lVBtM)GqHj z_7oVkq>dlrK?1!!4c8<%zcA|p@MVfDb6&_Qe|dO<+dxEF5mJy|3RQOA^P<_aBFDF( zf$0xe?6N9;7hoJ4Vm-|Jl3m4#uIt0YMP^THohv)=l-|{T*`sTX?DsECznkIUCsW0t z<{_Orc(#%r-~IHk^=7Lu#$tw7hsHyv#q2D-lA=qFE{#j6yy_|+6U!d5zuaYQW5`1F zGdLcnlGYWFL_hE};=$b!_nEr6Cmjp9%1&HlJD|VhKdC3{6m^B^ZAsc2nF7y?n~$Z z5EbRq3^AYKJ;0Mr^k~#;AdY^u!;X&^3Vt6Yo?cRx!R5o&h`a=@8;Z^ zUk@R3Pg%JLn`T)UuIRqD<*FTgjB|fo>!J=GzW^k=)LeZnW|??#%?ffXmjTDiS+cSx!;()kpiuBxxH~-#8)jJA*`Y<{a`@k=055X z;%nOz%iP=V7?IA8Jltctbuqg+9JpmS0e%UcBEFh_3Az7Re`u|L4Pbj&?5bszY4U3~ zrvOM_lPRq&K#+=6_ejL?--2Ed-@j*48(&_u`A7K3^lvJMC>*>`;7P}s5D8tt<${`+ zyX3F(r?H7I8(-TvUa0NDx-J3hUU|@y_Ak=hV1+2&mL2;}?|gq^QrgyUQ;GD#e1Lpb zA#Lnh@Ub`=v7NwEY~%i2TZ$X3qif-FZSe@DD1DhoWLdu~tIsXfYAo9c7rBvq|9kd@ zWWUQ~$=9rwok-gH*rP9eR#(&kl4F-70&`&9!q+932`^D^uPTVy^$=U1h;Q#of8K*R zaA4=-i^Sa=0c#>6?&}nVv@}>b`J!`hcAK7ekTG+lJtU5ks&7D=x+~Ki*7uoqww0mHcyTi~hmaSjK%z;g^gb3o(ds2R-lTTbEcNL_)q0b6t9 z{Kymz&bDXXL{?UEHuKGBu7AB_Zm6+WKQk*>b9TPAFmRc8(eEt^NNdJ?*x`snGnJo872`y9dAE+KS`sho4wu0Xi=_kf7MXv~anfI*n zTpRnK?N{6C?s?GqBk3A-O;GkB{|-k@BA~I%l87kOA2|`uVI3oI73Od-yMO-KD0dE_B^gh7PShJQam6D`eK! z|1E5MuEBpUaN_7&mfBXBW%sAUR+f9*T{?4r1Kb|-L@NQMh5!pI1z)~XCmepYmqQq6 zOC9`??~|d_T?kSaza#&_Vqrv!H{qsF9#l;0Xq#%e^?6*p_c8ZgO53R)@4xZ=1A|5Q zdF=d1z$T5Iy7b|D7P;M1p5@lqbWY*BjO*OY z1@F-Jh}`JKh8h~*~NGv?^a_VWt=%4|~vxLN(Ij-_V%hidoc zBWIMXL{5L-1PCRiqbEGWdX&=_Eg)S5TSh5wb@q;R9i8cz&GF{jHe~_F>s2Ce`W3J!xkVzCZ#ci~lb46hg;q9tG#r?2c30s>=%MK0PZ&gh!;tDP2OImL18o zdA9U>T#aQZ<{{%$XOXERi~JxbXCpWVT=9pZqwJ^7zc$w<38QJ7+dj97>Z{YYUo#Ia zzxc4^@#`j1y{s8Rev8(NY1S`!q4+lE+e&_h1G|lf**k>Nik{69=t-E=MI_*kS%KHV zlVnxng3PTOYVz)lM@4OO#9YR!h5N~wiYnXR!ad%cbNMZHt{?CtRRq^l#>(Yg=IaQX z{K`r*EF9OUnyHuGgKVWhV9m!b(j=;%H;&%tAB{h_iQ}ev01a??8kI=d$@;GhlNAqm zk7FX#zc&M%Kl$|pKdi`TqhK42qspd#cwTugNB2OWv7CKo-t+6GH6xawQ$(pKxqgUy zDhWBbgsk~0$ijx2-<;?M9p%YHyR7AQlQPTW!Fl-!6TAJsffWUnSEv$)8M6Nt1}q~y z7NmTYCNx*{?*$X&XLzW1g5bg-6E2tQvs!y%l%)qY=-*ts3>q~o#Ktn$1p155=A}_8 zr2E#}s@rFg&i_biLS`tsx283nFYt>N4I>oRzY}Jf35=7x+KsBn{6Yh~bU!k(_`P{O zR&`h!w*@kBvdI~=%Fj4gf8CT1?+htZ2~EqNDa)vnFPHS~nTi}ZT%pZtvEO#IP%gD6 z6cAD@;&%y`9KH#*zO-;LVei^Fi!=+5^&2c<_*rq8lJPdDwsm*`$-S|o9deywmtm`E zQ@+bZ+Ed?&qEHF{F$##FXlcvSb`tQDwwCK0+iBuFw>oP-;vG+aYW`PqteK zuz3c{E$vRT4=v}y{mypd=skWX(#=NKco%J`mM#f*di`!3{T+vNJ zF3g0$4zXXhF+-OZn#!)o^+=u$WJ3LB4kvys^P6p;8w9jmbCi{4tx=%PV{fDClB|9n zNr?07EZAtYZy(r2zmlXn-Nu%{%WkC1AO)&Y|D>?L^*N(c!7<>qo$bLA=!mQDXsPpF z3Y_GXa&@!hfi`66A0>XcWMLY+9U?CJ37iVVs$MbaC#1NAPn{JH%jcN(lXjC|rzDS| zaCsO(*z)R8ShU{E%R&nY9X^LVC>eml1p>+P@8N?NNr8^1S-?c2eIRNe11gGa#e0$e z4IvS3!Dq|!Na$)O%_TrVuo9ZC=Z;f&Lv)n&^E21Ln7?#uPlpX+xVH~Y3{rD}+}Bht zE#0tke67;-;8Hhi`%2#et93vyM6sf&)0lD+DHN?q?jrR7TmxDHO(N)=2n8-8+mmx+ zt#n#ExVHqpt+F&HOH}_jAT29#Nx`sjjxHemy0@`Ggu1bM6Fp%I`5CE;X0If2lpL3s znKFJ|-R_g*EP)JYv;BK$!$_r`l)Q*BKrF2K4Mt$tB;;0qW3vS)qT`{50P63O@VOI8 zS>1_`M4;mXTZ9UBw67VH*OK1q&#M@O24ssn{4?Rc7!&uyeA=g7?M$uR_09(Q(8t$f z1wbVmR~^---^NWsf87)uZ$%-_dq#2(|0rHAF9Ze9Yh#VU#EcFF&J>{RNB$6*bFZk0R}#U^g0CotFw)XSK$!erqK} zPzoKU8vebm|6ExcOC+=4Grv0Tv$(CGyZdTmb!h4Z!=*YOiWn>|909&p1ABzVe#YZ=QZQ^%_M?JfImNzr^Hc~V~R4e*f*t;{(ua^ ze@G7qv+k(~^AO^+LH*4kU+|kGKi|sBgXch)ezKd2$XOOj{7c=KtF?nI_*^s@U0_|f z^$@XYGWTlMZe{xKI!cU7=Pm?E$)>$djcYr!#+;X0_JC$P zzc(-cm1s<|hU5tA2y z8q{xDkbJq7^1$hX$@!+l3+o!Uu+7g&nHA7Ml3a}GmYYfS>4-D;geInBI?0ZZqeSTDT+V`Dcb!PA;Yu}S6)WH*zxzkc94D66GofZ*sz zAr#pzSO{%%3XKO(|Gz0H(?%K_x1LvW9=7$4a|g_Usk&%br%_R^~D*ymFbwz zD0;=l11bqLr6-rDyiAd~Oyn9g-bUFmxs{10DT90}o7dhV4kg3-V)AJYVt(D&oS8z7%2aknSi!j5hBt||FWYkbJW&83XqG-q~nPko^2NiNML z4!veCXF1-r%XR1ct#$YFRob2gqasOsKQ5e8H8*D{ESps_t4vVZk}s1tLKv_yGC2-$qW?Q+9r}vfH9QWA$XKO#iYL9S1vOE#;Y)f>e0^`YAq5y1}Ef zEP>~Bzh4GYfrI1r6-=J4m6)5ti~y=J5-O~S*-KkQOM_~1Oo!+2!(@d-k}@cj99ll9H0&j z6Q!i@@451kQ@2{h;Yy*z65ywm2yg}nx?o;93j;smDa5HY6+}CWdV3POOC$9b)#Vyt&FA#kzTR<`GWr+13 z(=ecL-!n^qm)+g2Y)_tgy)e;O%po-li)FP(MLljdQP4tf{8N{(RP+_au#)zJDF75o>iuA`PO0Md5}$OwiXml{=dXt-?9C(q|WDJi%J+40OxfMq=&0YaDyN z@@RxptA$E)x2kO3?^0R^Iq%BHk@k7eQbQG5Cm^Wm1{H%@6XbaC0wrxR&qPQ3m~oRV zrDKlK=wv`Eh>IZi=Wh7m5bFLo5QPD89$_RogKCwvy!h|}4>SW`g$x@4wSX#_)B7VJ zqi7^|yxc0oY>XP=|NLmz$Ie?xZ(;uNS5yd_w%J$-$~E&MY(O(8?=p)z=jrCG4PiY! zDrz3Z18F-8l-(8Lh{ZotnW5=$e44`H_Px|r3dj?1^C3**=cGjTeU)Pl9ZvB$ZuJ^|Y;%O7)Zn7{fPH-uMX7M4Y7lxdO28J2W)^XBHB zEny*rCM3 z3P;(k>Td0-{r5qXd_(NR&Y{aO1&411Bwq?V)c=y95iUR#gwXr&MN3NlH$*^d>lN+` zT~!fnY}%EV_`re~=MzSwuZpOh5z+C#Dic1%I#Xa>Q|B-yVMcE@H!4d0-VitRxyR@T zm@Qg6H~*9aTt{X+(d0rf2iDm(i6_*kBqhFXLt(DW9i1Xj4vzk%yDz z^cy=;S5D>IJvx+?bld5+B-TJ|6tsCkePx01aQHT?MGzAvH=lb(@`a0qKJIsUZ6tE6 zdq_iu>W)V_IU39K_t$6{ln5(gVtL17p#K_3v5ne!EKr##j0Bv|Tmb0ekC!I=0XVi*Fme2oS8|;@Jk)M#&OsEP}l2}fI^X)Bu z$@@Lu|JX zIOKsLWKkz%i6)WGgVx8t!mwVg$jJV1aKb`obJE`$L~?TGC2GW^QQ>o4H~}D!=)^Dd z6%+x#dlLN8!g#6m_4x`OZ}W0pJTTW&a8OQzkaIjHBBP3>atPY@on1_w2*W%$v0A;p zZC^!a%yA8l=DhHqT^h4n-ESACq}i;tCni)(ISkqWUrW{Ypr0x9BC3Qhbiw-d-R|Q+ zy|1LN4mRLK%OQ|vkb_bEWEYA=rhOv2kU*Y#*qYJa^%pUF>1o-zZ-ATAa5BH z1~nXLwm&I13g_zlD1sal*09z85IPiGxle#nBd_61RDXz^Y@~5y<6=`GpH~UOQiuu= zaPIv6Qzl<*~H#S|U@uazhj)S+T#8%b6YGk~c|NWu&>6dqpBYOfL zI=S-L@T*Cf|NNEt_j|Wd*5fFpyKgf03~#C{MlZ*#X0MSbPrQ8pd&1+)QDsb!%hmq1 zm2ttn9KSIVX|n!CU4+Xvb3~a|`F1?biFhD*4|bpD?fLps*gkAt=Dy*-#Xjt7YmD|c z8Od(!_&2l!?{lT)Op^;=+@^mWd;sF4)mANQCkWS4DK=9_ZdJJeoB-RgoY}}wyb+z} zv`*CjvuCyI6f3J$M1qICLM5NBC#_0FVw!P>`}MNn+4YB}Z%tRQ7RiyPbfQ|l_YFmQ zjE4WuN97Id7YGxK!3^jDupsxdBGV+*V-_)KuW4%CvS#%?eb_!dgxcMxOE|Qy|Cyio zJ4m{IJo%$w%s&>F_y5in+d0jC17moQfAD6g``IJD-y0MD#eI7Nu@71D&10-5zG>4l zWbqX>6O0MqJYd*n(99QF5YB+lDBawC9;3*%ns2OcB%IXkRoK>$QXQjc4b?^#!BgfX z6`sVbTkSpwhl8J2iR>$-hju6LX828~%pbq0;WTJ3E*?*Tj8Q~cOVNZt0uY%Z32&jX zs+t=pk-M?u_c;+GXyYh?dC?Q!b+WCee}2fUy`TEYz$R$dxsW*$ZhbEU9&82jL(%- z6ulEc#l0U*46OPwk%AnvxAbRye#TXUsd2gsOR}5Uo0}heJv|dUn7b8(g4l;@7jT`@ z$gz#5AYZEKtKI+kL0yaR&nvEoBp=}*H6_{j@PxK^I@jHIp0htSYDBsEv6uUT67LL# z@_V|4BRJNl&9adY6YDoq3(=VYoLXaBF7GwMGgZ4IX7WN+IoAE2jKd+2pfZ=0+J`}9 zt|9R2fSF2-+G5Hu!jGyM%G;8IvZUtJF;yVqUx6R{)MAodv5|u>#YBMYz_I#pfjzOY zs>C+xKHbHas*g{7%k}~%g?o*&__~EvAOXK;9+K14+J0grr?@Fe3?vDM-z+XgilfzT zcE*K?ZL)75J{rg(ep&0kCz}7NnI0rr@8NY!k~-RQwlPtzha~I0;89gn+&HcqUDZBz zUH_8(n)QvhXWQ5I+jE|q+j2SHJA)GN>}@)#Xf~cMh);P#SCx^&rgrh6J?8;99Xg8} zTD*MzK_HN<#PH3GSc4k~PDL(zq2S0VJNxus&po!&kSJE-7`S+Zs$vKn{E>ODLKU*~ z{h*>fED3p$ZY2klV6D2F?fRqE{hAYVHOLkWxhrC>NDqk&DvLIQ1WX9`swv8uZ@UC! zCWb=-(7>1|1u|B0=rWR|+v!5{w;}Pm1gdDSJqsT7G6bG%HVKhpW%iyFjv%Szkl9Ie zj)pVjr}PzO$$Z8L$pNjTe$k-KZ8<;hkfw34f!F95$*DChpL^I(ZMHxHgtK3S2Jr>S z>lIoLk+U$87bG#AC!bUxNs&HP>kyNzb;%^n3CKK)c4kO26&w<6p`p}a^c-k?ruiZc zF8TBHb~j_=JaF_ja9lW|zD@Y1C`mCUnTjusiTnn?UzCYF8QJ0>+mW6aeixjWZ0!Jo z9eYGr-_uQ2?Jl8qK2LUCkIPI>$Bk>Juw+%-Moc!EuQ`cWF3XjUb7hUEfwxZV!{m;)Efor4or4k~3{3ClDvjD6c@b;1sXltLp#A9o-ZX=Pz% z%ZsR2lyVPXFU0p`D7Rly%b9B%umbzdbAx>BZKx8M*4Lqs5)=ZZGjNwrU zKxWzUic+Ctnn4{$HgVk}k@=2Dt$WyqcLtnGfey29uwS>*Tt(c7ewUx9vUj8pDvLns zT43jY^@p{yR9f3a^h-M8l|f_28JLZ$JbqwVNG9@feAzJm9sYkv9!0z*mv33xmo+0u zdIG}+fXoOwx4^TJ`>b5a(9eUyFG=x{S+9;GilR;_0Ix3tShlNf&W-P+?8yiIbRjNG ztzVv&i(V_)3@^UvtRE`?xo~mOw(kvO%Mp`da?kCfFt?{DG`q?P0Rhe~CT>rmD zKR#6NBXE2mz`-}pI)QXOgE3*L)n0_4t~EcA@ZQkI!W->MXcAX}b!-GDPbiAaBOSqEa};+!oGVy(H0jxaJ-?CmR_ees zNuC8(tD(yv4?0fqL{*XKp9pd5n!YNaY5n|FdDr+;u|E`=cxS_3dUWQ(7eCwy7N-;M zHoEYE6L4?8o2s>r?f!UoppeyN5c&sAxb>b8y*Ysjc^IxzcAg{Wy^7^0E>#_#4?9ag z!f%=e-xZ_Ysq6AsdSE1<^x=QzJ^TCkDNwO*WIz5Z`wG41LvY@@U2CWf`C11?#+2Zs z&TAc>sQ-v-1m8pi_J=wK!8NrB`AX(o38fV%*ZMR@GR%M#1p*?djf?FoLU8p5v*&D! z#fAr!osFD4*g*Bm<;&)L^iTigJpBx#)(eRSWc~_TDxW$x>wj16aygIwqbbk9iJ^Zo zy2}6^%3fFDwE8V-Z-71F2+~E7#vEU87bL#?1W=EX)(uqbJB1AhCwSRUgIy#QWG;9o z!MKvMS3pHs=TawcGN>i3e7@Tl_#MH8XPou38`ufpy##=B^PpA@5q%2jQFj}VW3+BkWb9f0v9=i~ zV%Iig%kJIu-r#4ps|UHkx8IUWMVIgj;@dNIDBu={6=K#&+a8_7nP(?`*A8+^BPSdb zPDrvaBS50qF9335=-|7JCJ747&ewN-tR;_u{Z-yE@QEFvBq{z!HsqZgDO_u!0dE^| zt~YIxaV8jRu)77x;{-0km&|8zh&O(?XG2p+Tdqc>iENe^a^yNI*4iNx5{7+{w;lAA8#R!Ox?2ojtl866*(60F!+ zsi1mI9V@rD{CE8D=H`XbqDOqD!MU4evLHPyE&E+nehOwmPvs*%KKlIf=da!G z3U?)+Bh!OdNEM9^)m_2QYi$46ZJ%_YK5v3O1% zkuft4d^+)}%h;R)QzxV|!l{Y(kM$yMKO8PZpYc5x-SR`j!`<0yWFX*GjWF7t@jO&Ih_wOQt#CESkb5I(QqXrw0e!TGnG~DRV>5x zA8pNA$3{OQtUizn8Is$Xkr}$@8QIX+-;OXx4p`8yYL^nU(I)+DM?Q#XxG50 z%qs)ImySzH4uj|>2l+@hh?G#uz~_($a%f%HArbq26Um|$|c@&=g46ylc# zLym3xG-e7W!=aC9(o8;*u${W&NNE|@*%r^rHfqP#?b}gVZXy35M}dF6A2@lHLS@2X zKXLCGi z9{lliP_-kLWK~hIC&gxthQ4EXHT~-9ey#GXH*G|i@&*f4(5a2&fbi&*Agj5J4>k;U z@R<-R|8ZF+VN`hDzYTN+!SUb_~9p$b~dD0fmuC~XqUdzb8?(vpw zf)!=llaIGkqxnOzyuIDPK}7?_<+m{w!ATa6PgO-Ct(SN%$CL-5oIpD^V2p6R zW(RW44Hw&Xzw$bpM>wn?Bp{eP_aHNEsc^H{ZAmO4v(Ifx%9*iW8`1QE32W`lBxx=+ zLa$R7G3B@}7LWM9*GT@)N#zZ0pk#7W3|yPjnca;hCKK}iIE8gl+<4#C^hnsbFD~=+ ze+z^BA$KQ|NhiT6YRj3<5_z_NdOgf^^vhor_xE@u z2I;X^bb_iFkG&i0!x%~JEGjMC{o!z-Fw{T;(d{dm_VcYcMaDkb)A z_wA+?0Z(BZr>qAnPHw26hd#=uo+wv^VC<)!`?O(LDNF)Z>|;4Bi8H{I)nxjOM^;Jb zI9P_-5w-4F34M@Om=$(x>Qo7>y#FM)dRF0X=kfc&z(ZFx{1@Umb3z&&BB4yGyhrv06Wv=mXxjj*ZG_byBEK99N13+# z5^OboTP7Q?1*+Th!!+fZ((D4h7Wp%__Sg7}2E#ef7TAMYpY~{FblqI~`iWus(A@vm z3xmYsaRyz@-|1f9cdB)la{hkH<&^bd?j^u^=`JEMe7XtmO8o&q;z`zgM91p=*fI34 z{kWF8iyGIh>$=nVra`ipozA|sK5Z^@!5G1|$8cSc-HJSL<(@-(b<=SS-^?>N#BkPa zjhjM9ez$eHyepzRD%JvM>LVecbDDheMdNO63f{L4sCL`i>3!jFv(IEwCVzCNBk&e6 z&G&}$!2B}@8-*_5dMX9^_8n%w3NJpKb!JSN)2cavH3ROLJ_YS};2oxLSgNNThcqQe zkTNnP`2+Bmq^6*%h+#z&Iiy|+l(N&t{M9lqbm}LABUqT+VX2%zvdm317JB(sk?ctf zENEo4h2)h8aD_ozbAyvhh5B#uk-vdDOLMCX&ho)!ow$-+FjV536}5s`5G9spMxEj+lencl|gp zz28O>4a)|LI_xbV55%5)-eFx_>x=j?-LzJ(Uqlbq(Uwb)JvsB%cnWmk2Qj=dPhFnr zh0NPL1?w2T8dmFeOdNZz-^G;-Xl0FxFC}NR24;WQ4^4+JewW{HYf+6i9F63%yL#hU z5&|!X%RXYl&VBn89@V`Z_W*jUYj-apT&+WrQPNw>c&~L?9^MHl z^=9D6bIc0Wm~I;Eo*}#RcFZ#C1_!Akwi}{M-9&SFGYXFZF?M5k^0M$c4afNt1l?U6OFQOLpz=AGa3M zX&{WV&qCRxUotChjE&Zeh^cl*2p{GqR~IP7S$-qWI}DB>0v@jq1SMSPf_b!P;ti3a0`y+yE-#t9Jl~DR zfo|NWVptbG_fw-tp^>WHYBvID95z6(h#c_wuVKem_{mK)1Srz!1_4rj^Oc|L<8zfQ zxvvrL|9buuq35&|(>dkH;M7AH;j@@IGN}%<2OWt2xiO3~3%1YtK{?;_$V`IV?Ye-* zBClen@Ha&OXF0a9i!TqKvpqb?$AoJvCzSOe`)NWhI1_!b;{@&p*;z$#O<)4SGm4yY zH4kKGT$-Kt-XgSP$ILo1KUZWK3wMucws;a!p-9<$bKNYcQ9HeA$YceG>&&viQw;xWf}#{T5=j@rQQ;>^NoW534VZu-hq7vg9WT@+(U z6$5*&K_hgQ8+P1y$HNprxn(I}P zOf^`Hqr%h?n6CX4T*HXCJDRTxIIs^6`^dDeXJn(COhB{offLRkaku>`n5I?fugo+4 z`1bJl%t9y<{3sq_lccZ6!Pb4nF9tjtWBkWz{&tAWA&VbGD$ijfL^Kz`Dai1A!XCx3qmOC(tLCC+=|yaSW{7zg{r;w(=E$Uc^E zc1QOBa6A@y$c0N*@O+c>PY}I>bw+&V)p%2@;>J8Jj^dD)h`)S%dX3<|)5m^Vep*^y zPo(uZDLt6@S1{(3k^#k&!q`t3pBYx ziCKIkzd@(w(Epm7%?)$d5WEko!S&F01#T7*6OS4DMW(xO9scVw9dI0Mria)zKP|sy z34wQ^TEu-duG5caeE5gAo+bI%2R3MbKr~G>NrOO=J@?UwSqFlDu$ZwA-$UR~Q@A@_ zqWp%SgrdW4~Q;t>sS3g7GGAi3?6ea{l3r#m7;tA%p->GpJyP10N z1V2jRtyNFAP7FLVAskWbmSvsGw9qtosT&YC6p7KFnUF`75L@NG=5+SFS;AdtPsBx| zvD3#+;*9;44q|((nqK~j8+V2AYiW0`FI3r2pTZ&`ZN=l@oQ2ptsNz|7fc2qer-Y8+ z{wB&!B;RuoI;&QCinF5w^Rf@CCfaP;lP&|(PR;_mc~EQkURJ&yqgvF6a+XIv!=f ziQn6BI{?=Cjp4^d_v^>w_r;veP>b%wKx5#GqPv|>Jok0jH1T*D=&+bbR37DP&nELX z2=0z(60EP-q~VissmkkN^aYMcA&PtblaK(4QDBJrteWfmpP;AP$TOkFdJm9yZx{uG*+%4XS0XcCgKi?3&_TXp-t zsH9M8Zq`xUuG%|_C44la4!E{j=Ev-FtZ%Mxnbh+oBg@?uJ)1Z|l0tbC4et=e^~#F* z_De)uDtz62Bi_r=Wd$#%!GB~j!uC#*tBFdr*xkyY&y*3x8~m{^2Z7RrtaZatUvTvO zFe2mICYXvFBuR`dqo{Q=469-z3PaHU9-u8!h_Mt)Slt;QC;bp}aYu};AwdG@ofz** z_)c+W7IDcxJ{6m2BA=yx^%!$5 z)$xZD`+;?j8(lZHmyFtJQQjcWccUqv1dh+L-;7a=Qj-8B1(U&%ZfJx&+GD_39QW?` z-8k;v6<7wWO&qT_a&EJby+>3V)l-LUf#SYWCa!B%lqWus`}h~5(#nsT-_KB@U~X0| zHEQY4d^Fq#tMMzcl}=a!15;$-Is5Wu+6Ht ze}p~M`nao?Co<3>L_IWap>VjtOzJFG3g^>G99g;FUub%< zuq@Bf>ce2+zb*2e@^;(Q-<+V%#!Y$3Kthi9b%w3K3FAW1y&Dz5giooK0q&F)wO%jh zdjlC4iwGQ-Y=z@2>y-a@YNEKg?R^ea|_2@7eL3XRr01RXv)3>ic3t zf#z^)*E`2S5jFQ7ISM8;pE|)?dV{H5Hw`sQx|I=N^5WophJ#dFu!$4&DB9Z9hy2a* zmw&O=4>FO6Z^{AiLGpK^51fyVcOD*bF!6TFw7|WuB6CSINV|i8J<1o17-KH#J_!F3 z8Q~rHF(|^lEJML`Iw+djmSj6ff;%YEJ5fG;_}+2;%Wt8V&!xglr((IGTcf0oQf+D# zZ>X6d3B{?J?mhV}YRe-Q?&HYFl;7?M8cnGNaeI2U2Q{Xce>=~K1Na95d>o)F*?108 z4?pnJQ!Z4_KdvCQbiepZ_4rBkp3nur*DQjLMqfBpDe@O~k zp_5AueHy^CkCB1y>x_2LY|B=7qiw4%GDF+_f$Vq({03S)yI2w!TCyC(jQeJ&&k-v4 z$ekH^iJl&1i-)rzUV9rjLA6SMyLVC~ENnKxDf94|*dBuiMxCWz)7OhNs(UKE#|cXz zAYc8`@bFRp5+h)fA6tqHvtqB^=!xWex#EM6+?m+=>`3*yJ*T?64)=9BJLtd0Wdx6# z)dzWc#y=g(SI*7>_@ygnj~leXL!&c%;I^dA$dpFbntzsWu(OWXWO{DvzisW^#m&Ta z8X`%VHmh)#WghUD8W**zM?^J@o=(Rq-lEg)c;TnZh!!lSu@CqiPS6iZPXj)xNJK5E zI;Lacr(4(HvW+31F|p;pZSv8z?xSV$B00SMwD>kVxqwzfJT-v-*lVRzvVT;#;|&y| z*R&YAQKKk!pP(hhz;Q|btp=T+vP9MdZ|Zh{`d~|W_AyDFJZ|!t{L|!P$mhPehCSYbG}IcvLv%0>F_%z5pz!11;(sBHpWUSyQc$7PW?9kO5?InuowlYS;F z+8Al9`U6a@L-_+Zq&^bRtnE^yLjV}Q=BSEx-y(SY3+Q7PgwyHG?$xhDIPtdIN+!&i z2pMJeV-}dWO}hL=Y@0M_)j4F(;BIfy+1+Mf^taFpbkBIGhbCG%7ZuQCZAI;_pz>Q2 zJ!P54aWx}K?qaEh&NP+`#9tdZRvH);7Lh(E!CnJ=pv7hCvzjOcn^rcWs0=#||nq%()T-ZSm~5V=hHmAnn-93GbMk=m{)O*4ev*dIXj`jZL|k z3?{Hgt>|W6%>y%pAS)R_0Mb!w>XvU=gp74yR0Xar)jLl|m2Rw}CUwGB{rO8jr2~}K zG(cflCY@`J0|aYLou}_ z?|N=)OURUlrunNPH+jmXn!4hhM#I}bzZXgK(>lI_w}C$+Y<2c3+Gr)g`&}4 z7afhSe);i+45!b=wof)Mx3Sdf9Pc2No8Yni{&X+6f7KVsw~Rb`D=+B!kO+R0w@7!q z1GMt z3d@^^b6RMI03V(e8S@THMPTpX6?s(ZI`^~w8!_^@ID{O(-_4wmvl#My1hO7)j{0(@B#b%si~Exw12G1Sm4a*g+16q zm=FcUD^TRo3gS#inLPI076|6N2i-B1(Zh)__fi_C=@X`HBx(I?i=_jkl!;K zN6j6Bp;dqR3_|fg46a@bmoC*jXS1g#&#d92&V(6_j~d_B)7U|G%jl-)?=)T~*`Rte z(aVXrOlDS$aEI>ofmfllV#|oeGdeeK$K`iUP>Mw*3i|#l1DwVLzCUrvmcG-*n0oIo}-ouCDSnDv{l zseCfx3qb2!J3N%q37uNQdH_!55QCSU;J%=5jKs0;!Cv`b>V`-m*n}w zOr(kyX-6g2k8L1z^OVu3g9ru5`-7D^r3NsAzxiS}+5RYq@Y^gfzVb6_dyE6n8;FHQ zn19;IKg49^glk_*05{fQNmmh^^Lj2yv33s?nWYEO`*`$7juLKm_7VWtz^kUAgd<6B zcPW_SlV{qvvIh#BG*cw)^JYA;!85v=FY}sN6ktP6QI}e zsT1OMSM#4O&MDsqV+Bq_(K6NSP~kcXO4{LOwVSD zrylspJ19~1R_85_k1Q_@2{H=Ei33B-FnIFMMq3g)7RARK3^V=b@*pT7@wUJPKb<2**d3j| zW}ZHl{le2HgTf!##1H7=y1kPiGmm|fzDBbtUkXvJ>L)O zK{^lqpR1`yeu8`?jn;K({?G6M-37wXB6qOJ%2ZiC!%Q{aFaQ-vY;@V-AOwDn~HvGa|V&UkKL$y7O@a(K( zg!_{GmnWwYbq~OZ+IY-uSGX-qdIbN8KM48z3w3`)mJH<%3m1kR9ZXOfCPN5X(!9J+r7)PLa?gw^|@)+&@J9`y6RSxYjts+t%XrEtOPi);7$G~Fg72T~v2 zO;Y9LihDk_91C5nXxeOV!5{K!US3tu)7AwN>GQh1qOrW1gUqT*P@*A~sM!H)_OpDUWJ?ccMEElVS zn6<6}fSpc4Ee$MSImYDpawlfsE0!e+%Dq$jjrV{Z^I?8w?D17CbZ8`jl^%#WCrc+X z{sJd4-3J6!MJGe0>9dl(7{(054JB9u*Fd4&RO~z6(x4u2h01XEpfv^j88+Z!;4BL@ z+|$!q=FzIhEV>H(ibO*>YqU8QMTDNYf;jB6z1e4@xZ9I$5+>wI?0aG)NG4|V&?==* zD`pf5U0Y*~6j25#goVF+CE=OD4Sj*`_W}(MneG#Th9^bwMps|Gz?k=X?l?W(l8O{b zo%J;Kh%G2iTZ$k`;jTZxA$v>ldNPO*gyjz zx51&Of*Y{eC!Xg%%cB}`Tk(7TRF`J&ZtRc}WVK&HekAc6Z4P)8PK zhs&y^VdKp;?xs6nXB(}LIo3wg^$1`e&^9wuTj@XZrL1`pyXRY04QwOQ!@Q58N`)_9 zWCnVqAjot4irAs!fm!pkBxDjY;PXoT-k5=eohffzbK~Q~n)gq)J+=>yl^q-vk zUVgT*k^dSP5p`5DNB;KSj4Xael4_*3&AQvSv83CkaU4b zmMwd(z@@=gk29<_rOG+G@1##1wo$fCPOUMLJ+H;qX}4E*%#LU3>-@=wXA^9_B9zzek9gv=7KU%#ag+qqyU*FokQsv4?9 zU&!ZaK!%;hH{JKW%E<~-nAQ=bU78@~I($lSw&+v!@}S8(0jU~xN`rYCp%rFssu8v) zPw>Ad#ILR6Q*I{K<4YKYZc=4e;$9j|yOe=b(N00zUk$l-$gx6_p7E&w!u%z{KQU5D z#>;)Nk)N*V?kW|sk}bU>8-c#R1Vypkq|M@Om~>cAC`2w8ZR#>ZOwCQ;GXq~%u2 zk;!o4uh~NfE;^WBH9NV_NF&=FfKI4T^u2#GpB@4WF!9dIT_1+=@JNeyB~0k)dOG-0 zXB6Z<^s!~IB^NJ7Csyi@*~580CX82~5@vwQi#GZqXh~JVE>yC)Shg-elatlW~<@a9q#OEpyJ-XYQ7Q4MnrwN!R{^|x} zb-~30hCU%@{RNzyr3ur>RoE+@A}U{@EY8IxCzVUZXHTjO4UC=W3)Mg9RK;EZAL2ED zc^0w3#~*Y;POYa-OXUY)wVkNj%!sv+uMzE)ut{g=0A#d!L{U6qS6?i+BTp=mB()IR z2=&G>(Id4xHQ5+3c0!R+(nwb|=3d($uM@+fykA1#Q`RM^+n@};byb7c+#D?xrYgi< zk|%vP*6oE-e;qd%g@48RCG#c!|2XOvOwYq+YvHCqb1ib%#i8Z)$fSoc#x!tjrI21~ ziaF#s@Nb9H*W7p14c1CJBL)9va@YOH9enTG<^xX`I#EQEB$Ri~R&{IjUvcMwo50`C z?aXEmZLrgA`0P=%s&hXmxkh%_vN6;AG&{^&PqhBH#lu#y3|tYo)i^fBTk^b4 z=Jpt?+H4N;%dob-n0?FL8Pux%*=x6J$X&bMlcXuguTprStlUSztM_6vN!r9k424PXe z&mV5Syn#1f4BK4AqlMzHG={F-&1a7b)X0dfeowK(^avyMf#N(ABEo> zh?zGUzB8KOPf2*)g~c%gxmJR5J-0vkyuyr+cP75u4&MtVrTDS^PfnprQYLm-vtkbbuz}v8C;Y<@%gWnOUzPvaeQ1L_G>t@WgCmAPEnv zNax)3mx&XM+#ew?J5kke`UGg%>x^&EqUX5^r1q@#=WG;z71i&P`pn$z{)v*|*v4@$ zFfIY-QtzSk-889=0QALIq6}x~3Me8v}KkMj9$F z$=~Wk^e?Y^ZYY~g>2>@B7#Oq6DjwX26AvQ~`;~fIPho>O7HMSeklk?hEAn4H2NgSU zJi0bGr_6(}IEufNJiutNQ-Hni6Kc1p&F}DMZiRqixVo8f%Wr!4soAP+D4FOiwXy4s zoGmRh)$i`bjtmg;&!(*}w=W2dw`x1(CEeA>K#3@JH>h$BV^HSlQ)aB}5Wz`VH~DbJ zxfpV~6ILm*tsztjF4f#;OID+l!VcTKO`6~d$L-l@_E?O=x7*RiUcvo_Rk;5Aj=+n- zp1Plga2{CM$T1ZXbxlvc6Q|FroB(}@P4g5`IAB_e^+V!#p6W4EgA)n*W`T&YVU$A{ zB{vUSguj*DdB|-V^;_LxtS64TkE38|d$sucL2h`zz{N}0CiB1?5r;%%rGqYPj53$m z&$)#%<7PnlnBy~;4lb!ncNbT}aU%M%)GqK@bKf$NnhZ}E@=2uMAZ%BNI{_A`8-FTQLlFQ{Nz2%!-<|)3U4Ax9-7JRho!)ylf@}x36W8v$zr$sC{G%U zha)tL#blR&vZV>gk)Q|{-B2u1L%k=|CE=E^S{bjjMu+dnP};TnbJNVVX*1dCb%1Tm zpMc&U%Tt*7{^Pyc%kdty-XAP>=er%_W?eYC8@qJ30)gQZN8jd|q9?lF8An~tu!-F| zCfE$ZApZKPw*r2uJs5+(zVG>R=e$OsrJIwxen@{SAZaY5ocqukmuI=pMre$-qn_pB zS$aAMvlTwA6^?sI@mhMafyR357&c`l+g9B7co?RNo??QBmw}&p2y&&Nf@Am--vk=> zYU^mYM(3QHO06hjRB8lF^CE|o9eiuowwZp%AC*7~MRGbYJqk2`;OqtRd2Xmxt6TKM zryi<|?(f1+BPLpV$omq9WkDg7R|e2Cn8YFUGgvj0TfpA8jfkDcOo1vR&MC&V(XRJL zuX%^12Z6A}@oOe&&A63RQ)JhIafl*`6r02OhAO&Fq*hzEp@te6tm>{pwT3=#`IM4 z((;O}v~0R_N_QC{QW6F6JeQg6LSZj`FfhjerxmPLC?5%vkMDT?VH{js#Mq$Wwx(2f zv@G#`*KSrxi;*FRKC^M~%HS|$fpb=m?98I}#_i74)s_Q&%fMKiqo<7eZ7;r7|Jmby ztF0XX)#R!Yykj!~^(GF(wtEZdrJMI(Sqqk<^4yF+eoOHu{2^u%pnw{Uyx<(jVglz|(_wxvtc8e66@@EAOKBFenA zz-Rl~SGub}Bp2Pu7CYOrFy!ZVifw0VX`u-0xs(wcyYof*IZ4{E2Ud;e+zTm4zXH3y zvK|YljhnIg&D1>jsWy9m9Q-KP3Vu@X!(e0Rl}1qZLUb?-UCiwe2>PRzq5KziJo@nw zf#CN3^i9D&2_o?8_cQ6ru=X|l0@{n+$GXLs&csRQNE7=j(`EA>N}`iD<6&>uGt6Ow zkmc3@qL)@o{JsxIkU8d-vNQ_qMrZTsc9p8=-1M=oC;VNJIC1?#W`*yqP^hGAZ^MSk zen|{se3yA``fBDUli7+|5_&e4PR^JaYa$FiL_t?#vc{I#c+Rql`;q5tPT|N#Cwn0T z<)ZU4BMtQy$pLKrXV|7C#Tq8G&W6*~8^_Y1y{}#JJDvMxt%D{W&5&!wFLP_mCcLMH z&#@7wY`J=0+2qf=%0<9}{%9+)PwYgE@D8ttQsk-T@89+k3eXf4V$IdM6~O1|xNn>n z4WALRaz;$>oJC414jM>B{PcNHM#i9SojU(&Ab*EGeq0`!b>%jvr!R#)2+NL?lvcl3 z+qKl$2`_L|Czihb`NmNMN|Ho%w-=Pzr=5yQlfsUesAr6xVB&8mmv9w0XG<1~yx}0d zomuxVEICfQ6|kr0w^chVxhE|M$46uK)$SpL6^CC&-|g+{T|4oPhtiT?o(g%S$4{)N zVK5=;nO%l|cBhpgIiGFAJI*Uw46fwAaLs_-@jfEVIQ`mP#=$&1n#EFt_}-K-;)|3# zf1uF7>w zp1|%}SEr1BXOGdJzRK6LQollHRVu@3&*qZEf}`B4!Rr{<_!B)L-)p@gjF>TC`E!Ax z&y)SIwPNz$gc*&s3$q?OKGcwcp!H!W534l1Ghs|m!Sny?6$*_RK7TotN-OC!^4LQB ze}=nY6T09IO(f?N@-be<<3g?SqKjP*NlM>DIE7I%D!(=MTLA8BpGCaQ8j`UtlnUtT zzPg9|1KD=H|6*SPB<}c)6oyzPNUOfw@k&!G&OSosw~QN$C;cX_-}@|bSDTtOy?eP? zoIh%JUzn|v95Br-(C)Vrlv{-qclxq$f9H207aa-SY*Es6mq0OoV!tj$$^J<|PhHQH zxW;n`(w&{!iub$hO^usquT__Rb*WUn{IK=uPUz(C!zryGP`i7D?&t;&*TSEkbHynI&Kciqb^ zHGZIRT&P)hZ>7I;N09Kf4D4B1bWy7bb}WsXU$GYG&%+TcV!wi9lQif6BV! z)@zbWpYNlPUgA%=inCfzuE~a1@vt zsxX{Ni?SSK4X0#$J9*z8efSXg%LqNcT#dV0?M@*m#pq}=v5@o~k=9xQ8Rwh4GSJ|q zp=1R!QtL*e24FtotUGcG@BUO7ws4(s#tkd058}kpYPx6^BzIptp9~NP+b=&d!yh6q zfvYHd5|6Fxp&(at$Ph=nU>%mE<@l792h_wqL`8E5ia#tp5npv$r!~Qd1Tq#RGkPPb zP3M`v+C*?Whi1VaR~02*LU1g_fq&u)k}3U}uCZnjLCbvtpx;3VeGkGm;|QMBYH#uU zIseN%?|8j=L=b}o0W8RT>^Y`5KORx!>V#kcNio7#g)V0zr{g~@8I2Riev;vL!ixwd zj;T9a`bT$1VD`yT6@A8;uDIpV6Z7}oU@EOuVaVxNY8Hy&qi@DP?D5MOluE^Bvg%8U z+{QA@tZkwe+OD9siC}9-=#t#FaQpBXrxUkO_?X*fnEYP-Si{$5Q+GOG9)|i<`Cu+d;{5TXjuj z>+iozB9{i`@`Mn!TQUJZq-2W?xgjBB3D}#W6{0NCR+MrVCpv8{{hjlYe8*>Hnvu5p z(LqUoN5P<7X~4Ahr?!ly|2}-vi;eP7+=dHEIN@=zn%N6k$%QyA4GJ)Yi#Nq-B8`<4 z#%O1OR1imb-W{J2?lTz@c(iN$n5|WcNe`)^yfd;`jO1yEh5H_E!7$X;a9rEsIlAhQ zMEP!yBYAw>RAGL4W$8Rcm=TUR^V2RLhJKX!1>B+vVy5ZfCV?Z`tCxE$^!G~&XuslP zcvRyoaOm#XR{X5fddDY8u&Z2~Lr^%~g!UsRtCy1Fm0GEgulrtf(-M>j@(Hyf)|!W5 zO`Z5^RbZ)PM{c`To@fLi#Efjo`pFV73sCtba}LuNQ1jl-e4-X?w^GoEtrKd%*KPM} zzcp`ub)otpddw@fwt1Dj)tDv0`sJ7_Fl^hWVw&Q=lRRb^yz|t9=pU+OvVFHwUP5if z!FYnzd$oA)Qutr{K)H;dG~V%&o|oKfx~?tYRdTAV-WndHRB_8e4t?d`t$$o-&HNh> zI(dKG@mMZvGg+tLIa48rX(Ymc^H>I6qMaW=Om`#O=J)5T{lgM(rVL1ak4)oLrq`dZ zI@v>OZx)b{*6}XXMA)m}TdYzE|FDhCLq_f=GdJyCNkrk!Ko+`Cj10Wp=3J} zKlz+aDn)PA9&f=34GK5x-x`c5Sb7$c2@$2+~Rsg;q$^ z5>O0B%#07Dl0fu6y2V<+>`HEkci`WpC?%E*+A}^Wp0gP}11wM#Aq-|7#A^{B>CtH}HbSDQfgg;drO~4UFZ`ab@FR0PC^B*7fZiBOT zt=xk+gb8{JUesqi&O72=oABGqKw-PFJB=Lp-ex>&(P+@4ydMAILT*PlE@RwDe4GYI zD*T7tQsgI2QYLhcQya&opl@I%IS}Kut-LoMz4nb{)G(j0Bwb-dIk?A#D#^Y81DKGWfVQ@uG)DyAi z(REe-es87HS{ez()bWY@0v5d_l{e35e3TUaFYN*xq2zvGNNf?&q&OL=`Z9gnjk2YD zpyLHvC@jzR{J%?F8vkA5(m@UNjmF}~)e(j`p`+}+yNK3=a|kxtU%cU@l;~)!mizqF z>16R?jAaY>2T{qBi;kv7b10l-++k9fZzK}klu9!JrnB40m<6Udv=srIMtG00C7je4 zRWRUTF|IuTrdv%?liyqN;rGglt;;M+&8mO>MRCg;6t6X*rya(W8T!wK@4Qv(*|!v^ zAmE?H@`YhotJ}kjS74BLR*$dLDBG9c0@a;8YL^tv8+|~KkZwPpz*D1V=Vr7YulYUSKg+qw#6GGmfm@qU z5>Y1`+;W(shTxvX$zZ@AJ7ye|nqYaAyUzsYumC?RKy?b~F(mn8eT=svQ)RJSLPNUs zog`D{BR=wi)+8JZ9!lfm;sJ@S77Ul$^e(&XmUB_A zHe|Nkwa+5tFgw=7!paCFUvXb7nI~nHyM@r7o2Ewyev)DA{ZQ+!kUt)ikXQjie4Yy* zFIRWS)&1+|e_DdtHFu>q-kBsEONomwsbXw8C^3RFhCZP5@)l=}OyGxjYIW)791dK` zj7G!l=F#@3c^e&dEfi_zBfMq%o^AZTED^b6zUjnj8Z}c9caS*xEuC9b7WI#wLPmE0 zdbwudsF{|bj@kGlUvT4iR@orv9sVB}YpJO(L@bTGUMuG@f-Q)wa@f?zH*1~elg#~z zEh{#hTkD~*s4Z&ITUm)Iisw9&xp8_kI@6nS#G*qkbBIr$$!v|K&5YqG>xb6Xkhc-J zR7|4sw;5t-Zu7Ssd|j_I)pE!C{*pw$G~%Q-ooPQAeNsZTHsbc+7HyD#bEtap)j_*( zJ@n7Q<{aogU{B!x-x-jiX zL3LhjrAA6ym&epy8o`C8p%0FQR2<2CUAa_!NJQ5wc3qfNu5j$J=ETuI=&3n$~Lm769bq?-DcmI=f7Y@2l;9w%Z>9iRG!$8nHN&Ipg;U zx$eJzidQ)to}e%<$k+93#nSpdR~N*r1Qd6_ z@cTHDS7IK9q#&PRzd$4FzS?cIGTf;4XM*!O#yiuH!Y4-OfBe>d(q)DYkQk+>_c4RJ zGas^feAieE&n@%EJv#Izyj}xi=6^km&n@%DrPqo6_w&TDvF|QjW=}>b zjF;SVi2W_)NP^0*xhgOn2d+Nt9l@GRN24k-%DFS2PgZU^hmu>v3PsOz&bxM}o{kD` zJu2aom3qlBD2Ra9dUh{LL%~@F3CB`(ePs{z2@0z@>PCg+qbeulG$RRry(El;l{5B? zKZ9p)?`>5kbeO*esh`2?}RtdgP<6xdsz(A>RY`*+%dWU(5pKqQhtwdB+~Yvcya<2Ae3axXeNX8 zGt$pG>?yWpW#CjKe!;k4`=!}yL&>L=UR8ao*W@0`ch0GzzcnER z(@NhCcA<>eZZ)c?C886EJ=TALvIu93bWJ*yz5;TT2~a3Qnmx@>3flOcXd<6?uLh70 zZUIm96Q13y4}!?RQfZA(LIEGqMB>~#k(flGOxfA&rDZc#WyM3~ubB}rgu2TEXSr8F zaRlii5#~|pmX9(O4rVfYo+kF-D`(jM2$O=H|9l57hYjfnK!4->0z$>#%8PMlcf3c* zB`Ta+3tw5gT>1PaOHv6b(KQLOe#9fdpvEH*nw-jxDZ2PcesMY{8s5917}UTQ4gG{( zs@ZYS6ESPLjL0mmAyJNY4X1N$r!PF3G8a9jRG5hk4`_ zO1+7^kt;pw7Z72Gp|m!680J2~ zmKCnkUka1`F+HV+#8!Y!Ntd-U<; z|EbWSBTk?FiqVrjfC?0T@IZb`KR&lwf! z+d3XHvuLSkU047l9w5ypE3bXGM`2$7(J2uCo;>h<`|!=d(b-9I2&D8Wp~8+PI6PhB zt)s>#luzf^zuu%*GfrHl+;-bT+LNp=>Cg$l=4L>6@ft0*Ch_I3pFL)ed&yVga@gyy-$4^aHH+A(X!j6S>+3k$ z<^<2GFZ~sb4Bqi&DD0zBK1xC<>cp~$y!RH!Mym_E}cj;u- zQhtF>rj<|dk9`Oy_h#|<=9y4)Lc`&WH?Uyo!vgzF&a^P!GLi- zU(!bpakj&~QWztn{v6Mf$O|?9GwRWA+(^)c6XoAJFNNUap*;*>u#~5Y@qe39L zEM4JyGo7&y1XK#Q7#!u;2;s|3XEGJCLN8~5Y8-sd3{p0^pU77LPe8E0@*qwZc(TA^ zE$lJm=zlgM6d}ha#r!qqT-+V+-%|ha9#P{t*VsrFnFK1WG9$2V4I0_D8UN;O`tUs@ zr+&9%lM3kFpQ`!2_ky3B{-M&Kr=lK@n37>O#a1uJdcQ=1Y&*Rz+!^~#x*w8SlxY%= zTa=U%t37wOT&D&7=V?GbkMD1g=MEgb67&F^GPju3{nuP=6j-SAA~_D~=`xiVGupa1 zbR|eHgnHiks){&oT+P_jm)#K$E@R8ev`Ru8)@d;L?Pfw#=;_9foUyKb9{QNk&|vpT zyXD~8!^Scq>fpM3SzZRnT2Eie0}-bA$40$+E1P|q33|mEa;`HoHiRQMkT2giL$Yi# zVuh#QH*Ve>;zf~92DlN4G*?73YEBjqI$-$)1TA-CbtUH znA@(v(_Y7~H)h=xD}H)7<;{JXMt`#FoJ~D2_Pl)*U+tZk(RvgiKVhk(5w_H3cW<^5NtyAY-OJPhN0uGQ$<$# zV_bID%PRDbfW`1#_%1xD{kM1n*Zj}1=XgI|gHGx_LKTEa^3adcng{oso5##WAU>fA z{~Rr3=ZUy$+7neazReANZE#SS8luS;)A{jvhgnHFC-7rcik_*l>hFI~QsVRwF27_L zAdbeS+&5y;@Zf;@i9|_vVGLW0WfKqKIuU=V+cgb@<7wM8r;$6x(7tnEJi;Jr$9)sH&zxy>QnY zIuD@EiWpOIs+<44e4jizKYS-648%;Y%^ml+J?H}kVfpWvyYs(w-M&Y!%a&rxoJpl8kbPmP3|@N{9Z z3EIhV?D;h|_(@_$$o3ThT@%HQwa*DXG!XE?>_|{96xUG!CMlc}mywq7I|#w&d)^;Z zl>nz0NLqKC(l`p8GZ~3MSv3ybs)qUb8|ZbIXvD0BJRN3HI_?`)WJ*)4^;>#vAz*7^iLjrKD_!1M2p{2 zP27B~5j{rfM$V(~v9NG-Xpi!BBAlXo`sHJg!bd3uy?^3=3F@@k-O1(;f(aJ@=Fu9MU3rS8hWz;h*^+6hciB-bTGvUKXuV>pn!v;BctQr-;5oakqQUry-AnjkvikmZ39W9Lf@;V8*vASXDZ&Im~suT_&W zjfrTysCl1!#HREH*#hf)1kvL`4uRjW=8a#RIS&WLH#z`Q;vUiuRMGQlITf=&&3a<6 z1r2ltDDz(C;qJKLol2j+UJ*jlL19=!c1`aY>F-27V9Ek;~DAX@Vu2ss%&|E{yMl#`OGE znf*tTyBLd?tDE-(erZYt5+zlLV zSEl!m=Rzy_F3C4^Fe{bvsiE60VTqe_K{CWcju=W63dfOpI@1#t6Tt+R@hq>+$FG?M zBEr@3s}MfX3Ctx^U&53ze%g7_obZv2H*R!@XS~vpO7X|1hj2zr!Y~lTCJ0cLlIK-< zlkPybZ`c&#r9Uo%;^bcq2bt%f6C74dUPmp*-RX1kO#{&a^Nwd5$Qd4_wK1=f7KL8X zkYsiBg?bF73w8zEd)8~;wa=niJc_L^ej)+40qlvvILzK?P?B00nd# z=#xeV<#-hya>|K}D#oPe>L9yXZdI!NWJL`q70U6#Cr^BAQzv2!3Wrl{!()SD8UU;_ z@C#{3Qj~;KA+@YcRTlTlBlfD zV6rWKVq_EnmqIUo4tJV(T;j!|l>zZoi# z_UUF#o5wdS!lZ?uP34!+6*;TXAzK)zIf2ZpY}W}e_LTsR;1%7%`xZv62MdZ2AP&KL z^C>=AHhSBangz&yP+~ziM2f^=za+O|ABJYhL-EoJ_7VJ2A2M`d@kh?-3Csp*|FGr% ztF)~rTvz14t)Jnj0TL(o@@_%uF1<8HLE)mq5~;a&)|7MrU0A_CU+O;wKcZiK2AluO zZj8K~Ny1Rcei!OL<0Bd{o^K?IAmhC%{QMWLT0tSv)$PX3EU;4k;IRi)Iu`b;3J{?d zZnRjAhFTsDAiB$_pODa=%x{H7SA`14Y`tZ`n_}B^Kvp;D5ifN2? zFRgY9kODMd>w94{f%m6eS%-&*e@^cfx;JkGIFXX5?3{$28d;0O5`Hbr=`P9wDd5ee zo&aq;I+CgFZ;LM8)P*;GDH!J+b0OaLu_v#(dlBwQ?(Mn|{tw&BF~#+8VP>o zkULJ$;+@tku@gBDlw)=FYWd*;0Jd0&7f9m7D3*AWKhp1<|BOYC>E%&D*3P4z zN~CSZd&@1TDC-rGY8>*M8x8fRU(|@DQL~CRuHl%xdzz|PN|X(wH8SRXQzurilD0cP z3t~N<{#-FV>hp_r)@;iMQSdZP@R$o-ZQr~yFP4IssrY3fzRG;XHemF#a@f_+HkUa~ z^%`pcOxN@Ig3V5Slol^OjiXA!K0F^qyt zW1B=qlWu+1a-Kg=;EFl3ONnF}`Whk1L}=^@uriUE)i{_3|AO9Y3(gnFJs$dSVCURw zIJ-+1gbgkVL`r-8Uu@WNRluRWf7HdZwmckx=Ffv#<|jKjv8| z_PPwUo@Qa1R2D)=2ksXLK;rKqz+?#?#Ar|_)p)BhP<=&N(^1gqbn|CT`nlJkka9?P zo5qPT^Dr2{1hM_;u*P)@K6DoNH~zGQY_+1JJ|a`RGRV2EkmQn3;bbq9QE(E!?q~}5 z9iP>CNE~E09DIER)N97^tgtowtoXvI4mSjswL?yEeqL6#j7UTlvusRnW!@yLD2Bec zam#@ULXGAZeU4jf$ov*k9X}Sw)=wbBMu?O|TUJd!MsmVC+X(g_G;laUSl>6Z+g?yz|(@P{CImw6;ScpHTdAQ#|=pT4;C2Z7PTk3g;0-8aE)?$oTe3=fOCK?$Kv( zd(3B3LoKuwiBA=yDw^wup1&O0#|4s*{O41rxXOMBxggVZ_~W^G3!OJp5&h7I#^{6| zhtzqXo|_9d$D>GeX6*@0P)}B4_V2Ob)8^zcf1o$7mT`}7B|>+fj>(OTpxY~%PtNIs zVN+s|lS`Up8R5$^jo1oU3_k?BgsA@Qn`R!R6f&|mp0f|84 zsc2t4C*wo^+Q#cF!)|@B5_OoZQhl9vcJ=cV1nC2b?(q6g)2K|3x@*E$)EuQq?M97Q z((S=C;g?MP%SYjm$1@Pg)g&*ssegZL>XfR!cP_oGTGz#nYn8U``BoPnE7sinhV{`; zi$Mv~W%b}}tw|bOt`ypA{=DvqB~agvM3C(J=7W?C#<0G1`12$6JNk?56kln*W?#qV zDV8`_>&3wO%Bz)#Y&Z_*R}*%gUbF=IV!Fcr#~-U5(Oks9*H5=Jzh~vmN1Al==yHd3 z5!ANwr9I2$%)duh#+S_t;cM0d0`E^MWQ{&B&q^qjiKU1JeKB9s`@P0``zF1s>`{6` zT&+zPbjp$m#`^&689 zs}4oRs*U8$K3e(HUcYrm0InSQ;b%?T&_mK*uK?D>#uJ9h#)s~XIy|tHfm6dXUOzB$ zZH4lftbUl|AQXnuu_OJionfc)i}OEa{!NSdATvppxFEFDSfK5e>cD3z1WpTGt) z6`oh4f@;bVhb6xZ%8Uzzpr;&VfXN_HE)xuYHtyMIGTyENGw2eEEZz1E4Rh{@1wQD} zEjfFzV(oqm9Dyt+SK8F)&s@#iJf)7boQe(-q)mNI>_g)rgkmhI(Pc*(hpIFyuh$-R ze~S~#b)Gr2nO6KarF3sg^0LfM)@jyZl38M3(sFH(1n_FuuWjW+ZcsWn&BA@|J?HYy zx5Do_l-bq1s~6G&7BJ89daJ4E{S`V%j@Zm4nWh)AsHm6{P+%buc>p%Wcfrm}_xJKb zLRY#p=tt*yCZ2CJN+nx3a#yV3Vf07DKesTF(U5ffp}^lY zyboXOXb*_a$aksZ&Txmh9hnPqQ@->DRtLv68pGbVZ+r%od*ph6LcAO>PJ>LF+@5A^faJRYIrh4`=)qIY0 zp9jH*N#^&00c704hL}KARsjYcTC0H(>>qSpr_zjJNe) z#%9WY%v~+<9fJoB>&9BT&2=`Gz7IQJ`IwkjC!&o-FLgu7Hi$UGS9(euS{C zrycN(&*3-!Lb>yr_+>Etg^4Em>1fRV;_W@-ntFo2Q99Cl4N|2eog@eXN|oLfM0(Xo zS3n>E=_M4Ajs#G;(h(4(6Pomn^w2>d^b(TG@BZ(bdtcn=dG)-Rb3SMH%+Aiv&U|P0 z6y&IKo<}B1ld;45^4~kWFtcI1V@l$|cfBv(?vSZy8NA#a;HkP3k?g;342RS?xieb} zI}CE1vgG20pE7+Zg4m`S(RRm@k6 zwoRpF(jb&Zhm%JKjbw&nn>sKYKrV(0DX;$(!EyC;E96XFj<&ork4qqA^}MZZQfKDM zYQm!TV(qd^U`N!S{zhNn=wIb%$qGYcSWvsQ&4=6ZGCVt6a{p38`GQRvtoCvrQelhD7Ah~{cYmXqvUJ{)l zXa)~eRbBjKYI9lJp2B=FSoo-fX4CNkpvz~~)s@iE8ej$}lKI@Cfj!fIr0To-tb6_I z*%<%P20Fp8CH07N&uRneCz}R?)o~neM<>|rC%)vbW#y>8AF4!JHqSdqC`bZ6zZ+U|VFN&0bRg{#09zMFqpcr3h6I)9D^ zS=O_&em&0IG2vHeCsr2ku0jv9@@*ZD)ZGigqSBd1rqq>{tKw4FUgabg#F>=JbSEOb z%AT{h{FEy;plz%t$RR^BKdSnm^ptnz$^{4!bdGH^9Ee5;n%aOZz z^9dTb==wf+`7>EAveDqcm%xxkkV04Y-4WZ2wdFI>;lqAUgw;kJ1=H1_7Q6n3aJFUUzX4D+xT z(6UhQB8z}W5rmr$(@Kft1q(Vqg&M?U(#~*K!&$l5TOM`-rEVP|?Mox|BQ}o0drM1y z(=#rsuZOY1tU&8Rik^~vD!n#3FVY4&uDDB|S2?qUK^`nFHvZe_|)~|BpIQbzh-{NGBjXIG!zrgG;be?A9Qry zrz$!>d~n{c7yLVJL6C(8Ld19L~hZ@!2$ahrE}izgp3aMX8k(em=k-KUj=uR4Zz4U{ZaL-PXH!ix_RdGK1@HZ^F)$$kbz0CKr^z2ZnStE6&^FAuNE&r^r7cdhdT&ev4usH_T zQqWyyjoG@9kHi`&nG!L0L@cy?){qj(9&}{0L%k+Ja~%_L5Ned#$Uzt+rVa1 zz*yEMingZ$GWP51O7g$saxO;_q_Z?$t^b&|`j=QO4s~BscKBh`Vet|7BE@Nrd>0{F z%#T~W?zN%^CdeWBx)UMUCiyWyW+I#$fnNmtLd3IC7!s7L_fA>xS0Cvr@dtCSaiylR zqKM6)dBK^NF{wA}!!fK(U*oJY3mQw}TI6cZj9K{_g4z?L4|bDOrM!c7k};Z+UOLB0 zu|`H5SIO>_02;DQmV3aMnpDBWcG5pWFLxxa61RCtJGuv54e=O()?4qOp_Bsd??}@3 zQ*Od)NLegRI00jWT>_*i;T2h#`tsiXRz zN~Xx*=KK-Gx(L7h`gz*em4PypG#~<9B!S zs2rqBTqWABxm-Oo#S5;2xPz_pd6j8We>lFl1MZg7jyq&YuX1!}ls9$D<4#A!Gfjv7EE1%!Eg2s&HgHEJgLfx^*-FRuOT>PN>x_2LF zEug5L_M?xLDUJ7%CSziZq3SGbV1yXVQ3EOb-InsJ?;<Sb3cplFw9(dVZ85HMjgXei(4yuMRb2ZTCN8Qj~PSd(%@MNOA2 zv5A!HsyEOiL{ij5>|Hwvy|{Xvf-~)Q(~Cux?KX5?&S$i$-5?e0t)6`lt!p3|4S2_0 zjkcU7s6Eo5LS1}Ji7gqB-Bnb7L;6s;)Py<`{oKxSw4>nh)S+h2O@JL=vyai%1_Vbg zzH$Y*=lL3UOfpRL_P%ooM)4zJjz}Xg+Gfd7+(Db>yC0pc_F0mUr?^6Emob-%>SNVH z$^#zY2JwlM_;KUdnGR%Rs}Jd6W#kH+lreqk zNrQf7vJ_1lKigvauj8Oz#64JG}aY1$ny6dt2%5N4y* zlQx&oE3*`G4C{&>+G?japhLaBz#fo5zgD|q^wCRjSp#r=>uUsa&?4XMG3B}QRYI`5yCv&4$sRRjS|4yU> zmAY`d7OUw`_aDZUB#EThls)wJ5GlAOfq(G^uA>_1s>yimPspI07J6R5pct;|8S5;n z?MOycf(+VQeS8 zIvvwT-(ww%)p4xQF^>gJy&42%;c@{n1tgm))h#j z9O+^!(=(i66y@D$68!hXN}oMY4I+L@WA7tYKrR%;Jk<0JcBD;q8zvX4V}uJDOXl({ z7*+0D?Sb^!WGDfS3B{Rf?T!%+s7qh|W(9S~F$?gEb63lg5?I9-la<4938&B(lgjCm zj?N$U-pxdb5bNj?E6gl5mbK5YO0_~CTIcU`F7p9;ZRnxGMAs!SMjy|*p`o)x;x#Rt zty=V`a$$1d;iKc0<4Yfp>=YRI`XRFyJ)qasGan|&ZlwmXm%Ys0{~5oBFkVq=x{--z zJ#4v{)NzeEPr~lRobPk7yf<#dmJ%G)Dp$4`bc1Dhky1n(oyiZiH-HGFQFr5gMPpp} zGLsba10r)I@`t3oVj^z%Nfs5_FP>69=~|G7ZB)X2_1eXpQ0RK2T#L4L)f#huCw;9= z=6q8QCnG4SN*!0SmIOyZqOJAh0mlsqv;M2}*lDvM>_5-v3Zq3r_tTE&Jh8wpaEEr) z*lfHU~%0qVdxxGK4tqS62s_3C*Rhr7HO6Ikq-5vxwGeMkUlM4 z_SO3&y6ou1GhUH)Uhk8{!S5JZBx4tP(dR>mk=n4bMW;jr;=9nea`t8N->srLvutTC zav>39^ic1=B%$Bcjd`9^zS2E(d7L`&Xw3`)~zk=r4aJ%^M})M;C~%2P6u67ziqs& zcVTlX38{e99P}HOQVuIUWEx0%siB?|1ng$gWHefx$9M&2QC&r709#dFPDY^Tj(VXm ztL`QVC5J`V?=OT|`)ohn=R=}*p0FL0taOEaK4X*`ex7B2m%dc!Gl{;U4XMnF_Okjl z924yY9q$B$EL;Coe&T@#4Scx$l+7FS$1##4?VWMY0NaCIJt2Dznvy`m%R-#O4i{zS z6Au;Bmz0@w*Q5=O@pNq7(d7{!Jg<5oUQImlxV|eFU}kI|8hKaj!C_H7p13UkCIS@F zHkoOv)=CiY&Z6walW+QL-VI05|NV7lSA8Atu6Y`oB)HFQ7|FT(H`r+V-F#%9RgdJz zEjS|K#$!=mZ~Pw3dwhTi`34hNx&7}?EsBz{eXAj-Bt+XxqC5k^m^EQoheU$GRs^<# zkGu_h&T$9pwQ^2JST2u3W-spr`P-_+6Kb`AXC6GoaXSd!rqH?#(qFsSww!PRTfJ1M zN@33FDQqHwA1OUT2D9O~KC4a%YUfgzV$pZtH<-S37hvV;F~Gy)#BJwwiRu2Xor#o94|k)b--ThWl?Eb2jKyVe|P(bUP4%N5x7d1kda8Q zl_IfJ{JtD5oGJ2e_#VP=#z3e14i$&zN>s33t;rn>vC#DGw)o*IO6qKAxtga3qqhOd zJ+Ljsp~&dMB)H(|PS!iWH0`!~e~j&p_HQ(023265ORXc2NR#6^I0#5-Voc%lL3;GYjB!#qf<@%5nOKtW1wEEci#Om%6<9r zXTr)S>`6iH(R;_U%jFpl664!V8N2Ey+y1SDKXa6xGQy^ z_Hbt>IQo!oAcSajC1#35gfgj8`;q3u=p`xU6?B|l#_#DnK-r&QuAOC3C21ACBhkf>?Mim~S>O<%^AOHw~hv zNjdsV$fY0}|Arf}@_#oDu7}3I!Au8ODznVlRaR^5#)g(7C24DUG$NwsjwoR?7vJ;d zXmN8t0HRbLSGL%(?+R54TDM^hpZwAW)>$8S@WT_IIY66Tqc{ePn?faD3`f?R9Q;_O z^!qsdS^Z_+ta&X9p*>128S)+(gfcwgG0a#1rNjKv$gh#||Og=xa=z~$TdD=kn7hJXimJrm|_%X{11Q;2_ zY`D={h%klrUk~wR=M!XyoMkUV+fT2Dh?sNTs=O9Jq`W%(%(ExorKdd(aueBv@C$4T zeUhx(ECzB5)2h5GV2qiCosg^d@LiAx_s}lcH`phVowMw)z%c$XTa<@}tjO1k@{-re z)gBcc771#XhbMrjM8%T=)gd^#2Fad$pW z6%ur2(oTsDlNWws@ww1j@CAID2uC(N-ZFP7eM(}RMqZnGgwQHB0X)1m`w&0j?tfNJ z^r0G!=?$IrA8=sTrN0$P&D~;grT;*fr58{DM+`3OL4u0q=l%^L5ze9A$gNQM;ZKQ> zmVX@3(1j__<%JLjirh{yrr*$oZ1_mS;I&@d#2360SQ+vz&5RpsZQ@(58YmL8N6EZs z?2UulK&kF5qD@6wXG6@IJMsInY{Uc2^lh#N5YEDVymlQjOgmO7oo>Soa_;xKm4R|B zgb#kWEt2CNcB=B)eNJcLt@ot#D`Nq+|9i-11;#2Hcek7IT#~YUh)TO&w|g&w=KLkgZ9%E{A|y-zb(9z+kgJdb%FLSkU%%^ek44{+$jlKa_Tu|v&Jlm2JL*Lu#PH8Bez4P95iMmegWn1;gN zb|l}hS{5R<&)RlhwUq`Enmz{EUJq#kRL1p>e{$8O9ZHPE00 z#H6wgZ|q$#QI-%Qe)qNyX#l(Lc+%m;#c}cfC1_l|Cr6Yjw2%_T*tbf1D6GVB9_@8P zq4p%3jQhXPsPhzey3+RGl!gnT^AU*4@>;NG?HwzDciGh@(x#(4!EJJ#g3|i;RjqO- zd>l$rXrd^s|2F55G`>SIHoAcorz3FGR8gq*sI(DL45AWu2|Q_R;>+`R(m-F;e5bAUeRCWBS6{apT`hF znTkrc{+zgV{5<-)_>ie3%oY4Sp~&0;9BwMbv6fsJJg3LTueW+Yxf$(|8TI79C1dTF zyrB*Z<(}cFzX+!$b(B9hok$dE!Kx=w!O<`Es3DK=e3qaiqYTS(O+P4|+NglCIy?2q zssj<5NUuTisR5sF{kW(9jX^BC_k1gtG2ck)*y|3Hc6}mEJsT zX}ImN4e=Cp9Jj~i+%Szv>$5JNNb&$u_{^>YYpY+To%WZR=gOv6MRpu4*HMVQ3}4v< z!MaclSB}J>#$!Dt4&Vlk_f=XVsml0ExkNdWwMW7`*!l#IlT?rE0Z{NpcI)vF&VJPY zl8+cps%SjgSa`JZGXeJ4b)n|n`*QI_T3Ppi?mRT!7Q>4BxzH1oM3X|S#oX^Ih8bnV zbe{wndoSFl8#a7_*8}mkFIgx8UAnDnHpSptSpHRz{dG}vyWkKSYWs~ni|?QNhsysW z>e_N~0qr+cQu?#rC$U##*p%ddmO!p~1)406qMmFfYJM>|!tUooMl*gIolId9}II=z*I~+ zgwH3QYjMVlyTrid`@l-+`CdUITi0DK+wFJ3dfLK0hvB+dBE^&A&($~>8Vd7&OjZ!o zVn^Q+zk#B|47RY`B<(p)D*tsFi@nG0M+RORTR-}YPd@dfE($M1pB%~gJfO_T zEcOb`03R*!@?0qM$nj|$AdJ|3ue$Z=}n(61J?g{*qucY`tJUYGV$k{IagCB5c9k7!$m34Mp+N6$B3}{M?DfNzLV2237VB2=zj0PtDp+lg#?8je(m!d5} za`(n3KEwZHP;G>%b|3-UNMVpw$d8Gq;{$QO``oq$%N?MqNX||dRY|5OJn({LWQ@98 zo!k7fwzKMCJ39NgW3_iv&4uaTSFrO>d>a5hi0P*JGDglZ2+ z)RRGVC~haIU7~xF>YGDVZ>D-#KePMsr#s~m#jx7Hab|j$v}Vm(2CgaaP8s9 zyUOjp_Ne_yC3_U+tF`e{Gx-U-&xZZ&di_XvJ6dJRrm!7dI2M!jk|e>VY{0oI zqdI%>dXYV!=fcJT9wQifR)_ra*J;Jy23OBa`c_1#yn8^e=Mx+diYjtV$)HkJD-b8_ z9jP~clm6K-=KtX=q*<3u(96JnQg29-FvC^0@ARr+6mk!L9Lx`5+T%BMZ=`l92`87J zH8scgH1X=0muTdi@iakyo9&&XoHzYm>s9NX6FWdoEUiz~Bd5h$thD@UN>bkm^H;%h zftL_;g@~aqYc+Eun+jy^0`n90@&iBwAiaX{-9C+4q5R$7Q@n5!eh6mSoDH1&7B<_C z_P!wFXvGl6jXgY+mKZPRK&Fq+MA%= zG5q8l%W3VfXWLxs5@3Kvul)*m7r{6RVmjf{E70rN_EAUqAJYW+Z(a*h-@$;!2(MTh zX%&=z>hR^CygAJ&?R>A;I}lfXvIwzq7tsJ(6T7GqCmiHVsk@-whrzcNyO1uu`BP@CD&YerKv{3r>A?KnvPD%y3Q0a}#s6m;{IQ zoQ>>DQLaP!Xna49L`Hces`rVXCfZX`HQ22*5`V*XDCyXvoFpS&em3Nxuf!}m77`+u z|8~5eRF+Y&BNH5s?5l`|U-VE+23gjNMQRn+h_#Tzkf(9p@r3&swjt8eM1Bpx z^n@ux>iWV}?*xrQI4&Qj4D)!j_r4eU6d}5rB;c%yWh0c<|JQZq*4{D2%szLOFF4Qh z@Yj-hp(C(G)DvS)_?yx6eguCGVPKvY$8Y@0{!jd-7^<)k&rvlprUHl(z}x5x%U%ve z`k!fxZxYS%X$GQ!G?b1g+OV+Aiix9OI<* znO{VQW(U2kOEZ}4_UXcRDN@d?W-1$p%#w<|i}e2icUt-U0uW|3YD44>^)_>G77^Nf zgCc?bj`ztDp7xmtd>hUr3n&WP0i+0hebjMCY!IMy7Z<%O0R|5LimU)~J>_|g!-jy- zqAco7wGyJn(#vD>7L`bkg`9$KV~`hd5nKLDjKuOlY#W8ysxEiFoR5gI)OvYZx`31uUv(TrS+8n&~JsG ziQLB_TG_E)z&eG}_}yRp$+L}LMk04GZ!^2^hNAMv=4*kb(u@d}^h;!Qd~A>j)vl{p zI@5PpL&pF<>E&Nzjt9LI;w#Ly%dJij-O57Ym2Ea6+;YQ*a zxb$!ccqf~_72V*p;%p+h-GE{qV}m6H@zTTUjqJg~ z1R#|0=cPaQ7OZXz=o*ECOs_JWw-HPr4(0yC}&uO=$w0bE4MBZy@EMI2xxONCEPo~^!>IY#@0ewSb zAgo*9=}NLk8E1h_7l=6PNDs}TksWJ;p?o~M4Zs!f8M|juwRQ5v&e0|_T_fAB?u8Z| z>8+-IY88g6Xi{DkT~Y+50h#!&hgfQP{;T_!ccW;|lfqv36ppy0>ma;F@&Udga)V4{ z=pgK^!rS~oCVi`JCTKE6)f}o3+(*)Hb4Nt>K8v8-lr!JRJ3sz}=A1$Yy4YQiYkG&w zkKrF%OUM7!Ij7_=50=ve%Ku(ak7# z2>&gb9&NRrl>5Y6lV+!64-q25Ui<}Im4J#lXInd$`5|3mEBf%CUqo3vXTCqM)klz|W;&t8 z7h%YRDWpR6o@^*JdNVd6a&0&K^r6@`ptKS|BkX2o=@|eom%i)YIxxR^f|{{h!)CJ9s1RgP&OBY^d3XBV7zue>?~_ zah)&BjW`(3OI;KwY1UA^ya zdcqb=<8^4#ysoz>@2%m!+r}6n<=4iWY8&tI0JuCESn^BjKYYH3XLR}Z=!H#JJNm_u ztdNq&^=7bW2=MM~IjNM8n%lo3U@d zJe|r8WT{$Mb$O19xdZNJl=MngtJhmWext!A7Jj4&ht$uyCfUp(O!*>@04|0_P+wlx zAYdJ#yT~R>3sgnQrsfjjs<{l_<&w3UxvncD2!?(C9&rcr6c}}GyAoAf(FRG?KDS-W zm-yg%>d|*YN9>)TR(Y_SFHx%3F-vGDDqx_L>!Iah&*F?_UPchgCjVKm@pwe;z_;-_ zkAvusPU}{<(Wcq-?pqWY9C@;dUK8Jpl)Vynk+GXJc92%qo}F|TQ1A6X{|n=!DIoG)UG z+rHW4kY_9#KN)QAmaO)0f9v+J$Z>hWs@g;U10nr0oY*RiU)9V#pU0nVoeyAB$3;@3 zw@&qf5PuwycZ1w$mDi&&TV1A#9Uqj4IyE3H=hM1Nw=->OCY4PJJupH6gw$acG-d(4 z2cE8H-K|Y48oO1Wf_=J?&v1DjR(GzZXLnN7Z0RuH&68Bu*)Rr9dQ~`ELxfvhwq9!W z63k}R!;t=Y1UQTg6*iMwQ9E+IKzJ=IV!qJOeJ>D4YiBU}cnEJpB6Q0hZx&VKT_}=PU@+y`l{K0)O7# z!q^Y{1s~V%8i#hn+IcCSyDnQL9#Uby?m}96Nk$p@M{{%`irphlKDeilg&Z-}jRfZX zN1bOh*=kPfChgyA(VVOq#!S(0nDyWyi56~ul|^ft>rR_cyj_I;AnA*oM}W!%mTkG8 zv~!muqO(T)aX~0Pi$*c4>ON`j9V$G5xl5@Qm0T6j4gy9bm@f5nW8qeq7ahBwiJ=V+7|+j6Nq1 z2K#&{$+2)jJs#Wqaf5tw>*Ddv#05olMDT3ce;mgQ5v7qlH3#X3CDOs@nFs$ z@fJhCL_&Uh;#^f8j&meiZoHTJ^6}GSL}jL)3rY$@&|%1pm3i~oyb~tY3GjRsIY5T3 z1*@Goqilc9jdR>t)*D9C@CG8l^SNgGWsZn4UAf|9?9Ml^FHq@(tTihSbbRb#H@8f< zNkR*V3Yq@}qNjLX*2QboSg->LG7`jAtXry;4A8Z4C{R8C4$mCQrk`pQVD0J)LG*Vv zP}!5tP(sxWvz&T<^Kll`+qQIPedK059uI1~%J2H7(0tqwLE*lx{M?{cF~vwj{4#M= z8HL=A5N&VqA(WI<|G?*WHn}7+2YSBTN7dfRvB33)n~2< z(q9TO&N|g7^Io<(dsG60T0ee26ziR}qReZ0fING7-qQ0Cr>kvX65PF@`R7pIG6rTi zPX1W$^lNcUTSxz5kiDXB+{nlyskU7hGE2g5zPo61LO(m&)C1HMl;w{vsL^T>T6KgWlol|kWWqeZ4+v%3&Xj)Rt?RDo)Wzi1I z(W1^GRYfVN!Ljap&`^DcVm(osU;RwxDoM{Pb*(oW6j#*MH{izn*WJaWaCtSn|FPh?u%Mn{&O-oeUg#Y5Y z{%11oPPrur3^!7r%9q=;FWg49_OuIXcz$CN?A-LzR^L^CBa*tUx(%nMXRUzh-%?jT zTlwA!D@h&8$k>-yKB33dpb_WY{)VKrZ)J4d8ffA%VZ|{5B4sZ#KkGt5v*5w`RZHTp>J z!H*UR2k=C{-n&dLK_bzz*|Rs^#nlJ^M?kp0)%uv2YG-pg)*5S+#Q+dbn z*af%my`EWXy6(K2Tql3KjWl%Q(LYSXGRk!YepoR`<$O>#Gj*=Z7AYes_<{Kxy$X@E zlJASZV=e7QylDEEeUITQz&*XO)RQaZKtN(Rd`sn5zg)y8^YCnr6mv&PmxA&#frWLx zsH2z33+t*Be?ND78jr`ePJmSqtPHQ#1#o^%%;MfDu*JXb-JZX+-?KZ6$0|u(mmZFF zXY3r(yC1W?e|X6WGV7(?5rFX@S2p)+hOG)TD1OT3@R9zOy>NWb-TM?p2D*B(LAS8( zLrb^rtxBxAz4|4YCxl1jqzxwYey|9iR*W)}carZ7?6*$aq?oZ(y z(mU{I?DvVtZn47VrZjW)=vjZNlNJhFSf?_=W!&ABVAtQN2UX_*UCrV?X6LZEAKzOQ z0KK~$r>xNQOG$t5W42!pngq93x8mb(TsbrkRP6gE|G>B|`#^{C-yGKKM^&c^IyKiq zuFvFV@<4%k)B$|;EG~x5z;eGRZe8$;FFM-oHG#d?xU$bL!ai1gs9G!!S9HtCWUGX| zVgmqOIf`0-`O0s4*?6Vwpy>F&*EL3!8}B7MaDyV+G7r;&MH)+vq()y9I%}`>aea`Y z>tc&dl!3GAHwIl5_rF%c1#9y5cmiuxvqqKsTHQd4B9wMJ1_`0;7#m;frJ1ak)G-%j z9e0@GC0UjJzU*ePh=9gIB(2%Sa7dbA3c^?Hps4(v75)zE3hrI3uL*I;f?GVZhlWRMdp|IZ`Q$NVI=`)7m9eT6Y}L*p{*Br)Iv_3+qatrV+gZpOl;b`!o@LGu!z8CYLx;NKoI7YII>$L0QWq znP;jXvrfJmuSsPa;S$)Vg*HeKgGfVeIstFQiO`ac6)yad!bARB27DysUQGdclPRDB zc_m?@z6Dhac)JCXNN>5Av?ggHq<&8VGV?Wmf|q+Ue_+TmqfAM1|E7Xuo5>6V?$nEr zjTL!BW4lgn=8Fw~!dp+u`8wKNgh97#{Nue7yAg4P8N<2wZGf$IpvES4JGL4$_Q*x#ksjjb6D{Z&l<7-zm+?sY)^a0DVm_9w z_aKXkNAFLhR#lJtBZz!n?C_vJX|Ul=(@OWu(+#7vgKeC3YHm0*3<_x6p_y|Rw!$3*=zaQpPi9BVTEXFf;zJC=TB z%5d>N2MF#(G65K_NS75|+bYUYlpw(+yYh2mmJxmR#f89&) zaI28gTNxxAd!}qzwN5t+-r>R;`;c;cKis_TB}yr)dw~{$Tv8wMN+DKi& zd9DS?dCz|G=J1i?4fAm1%kTYBy^A027*d(_ZvzZ!nD3vFf@YNe7<~;mxU=1PxA+GN z$)YLuOkU?vB-oEH>{sB3feE!d6&lY+7+*~26P=`pU#B_=)LGzcY*9m1`6lYim+52b zEai9^#xU99h%d9~lK-?2v{d7MTE9|2f{Ap-Rb|olG8RAP1r&cl&eqExuNh68(&D8L z5Zk}+?ODVrKypSRStqPvlS2=5Ev2sP#=_R0P8XV;kLqb|s z{02fh)*<(xftYbuAV=pRu|3K|efsTV6aHmW$FZRgL4AKVdxd9K%6Qoy_4YC9(+ZnT ztGQB68cuB{&NeOuo54r;4O)-kYduJ1_ic$84WEKq5uoWmo4;<+OQRMF3i5-PpP#yC zG;EK&R6fX!o(C;0iO*<^jj|QT)Jt{#c$nT=>Z5$JXA~lw6EzKQqPK+jZThzkRH;po z`HEk?J9+tDn@3rM{n6gkxSFa9Tf@>=GNvoXE?R~SbNp+ zSrE`F`0BxB;R`u!9-Co_g4%qo)qSa#0DcoN;0<|U5-7RWj=b87`F7t|mxr`1()rHq zrN@#b7*A>6dFpl20~T##@EhNk66|G?5eqg5Du$^Sd2>~VhR}Y?>c~Fih2yf6gy{lT zoA0l#^uTrZug%w8jh~s7)Sxnpnm#6GAn1Xxn#;)yWKb>s?uN|97=!h=jr%O} z*uPot+@K2F2QPa^7Hz(HE|qjq-?=vZGHzqAa70T^4prEzgH@H<)qj9l*4WzBAA9NA z&RsHX20WFn(|(Rv;YbA~@n@^Da1)^jpMms?bp$+g4s&Kr9RUqvpjZ zI4&Y`IQ;H%+vF&l-~U|ZyQ}V`x|Q#J;Pwr6Fd6ba#|DUyBFN-@6t^QP!{Pe?qdt~zQW;mpw|?i+*#9~N*CAb0wRtKLUcYQ(N^Q4( zjZsQQv^>o|yY%0Pl8}tx@W}t+GcUV(y~Qr7*kTtp;0wgc;*2jRm|VCf9l(Pw2Q$&*fLBk7ot4{5lr}2@cIu|4?*Bo* z`&{~oZgU^HqW@?YkwKJnO^QGDN_gEIBE&@|c#&m)mFP$=h#S+(q7n^xiQhK5{eY%T zWhH%<%LKfv2E=xfj+%x(%8!Uh?3mUqzTXXbK}Ggc2|zHTjHp^=k!n_Cb5R{AL1-@F zIeyuu+72X^7>(^9-9z2Y-AuXt+{W9%B)CU5`K!=^Uyrwbb(Fj*ejVa!N7f~${)-;&*pbMrtv#69$35pI?u}+Q z0jx3<(0OJMM|3nbEz~1T}=j+)DwQpy0|M3;efJ7@0kG}$>%sm zeO(ujv+94Z={A1Ad(-T5%_h7)LqGdGV@O&+J2>Fv@B7i;`2McJKq6Mlj1fjH4pCz; zi-WkloNqN9QMxd9$}S{&^bvJ@+w$q$p01dPGx zK2%8rG1nZt`Q4)LRZuh`LMfBm%|1Dt)0>Pv*a~}Yu z%vQ%%kD>Tu`D5h02)7CN%4Vn54(}VO;$5=M0jFXw@_b(%MHoFtOE|z{L@sC(c3?}s zRlU+Omdw73xEJz`BmFN2?M^ZO!+?O4|8^axb$>??=Rw_HE4@*o0u0=&-C&ZazFhu$ z=b^gwZetbj2rbCGH!ROn!z~jXYp4(iq1P8!7q?cKO5iU(P9s zA;lmCZ$!?Vz$a|ehwidHF6F49QRPo?gJ%WI%+oqWNnrs5y9nnLF3^xn9OJP#G!vU; zi^7Z$0g9Q^nnq;iIteyioXD?O@TsHBUu|;*-mf=_!O}s3R6wuos^9QNER2-)Gaf$^ z<OKd2J_u{!mAj?V;Ew_*d+e{|n_;IJ^|$^(`7+2byWg2uxTFkNjAfEvm!f6WoMi zaJ5qR14Z0Ji=w2;7~e-z)lYG1Du&?BBzfHL zGNG4$$cCSKwAVmC!UKU>2Wl~l-p)E?bW0&r85R}ANr|Lz7#)}x;Ib+{uH1)A#@z$9 zmeWgSKE?iv!nFOus245Fg!bEM6y+RH&6-4GG;7d5|D^Go^x24S`2JELs79~e^01uW z15&4bRoJ>yzUWrMEl24d*VIFkzKL`^hg7|S2&b~rHpE!xr+AF!HbV6F0if;rk&x=$ zfP8pxGUm6?jr{xBZ-%e~cDbf|9d$V(k-eObajEQ2mN4mC*6a?b{T30C=^$l7eU&Ib zXR-!Y!=~KnwIj!kc`4*4tCF?4sz-b0hx$A@EntPPHKurT51(@tUI%_WrA9Hhn|}HG zNcyX8)9~PIV(gfq_;37=INKtQ!X(1LZ2kgJ%?vZP!jbgDD8S-k%GSV*7Lp?|E**jS z4iTD>*?>em;T-r}YkzrZ%k8oLC)qLvG|p?>7w1&O5zTl^%Y?*3Bjoq4{(4V2ZX*Va zj{IY=$-?WENg%x6$r(>8z~fKN98sdb?(U~d7QUFHc5N4@01m6SYVmkT5ndU{B`s-D zF+btlMMT`c67-GbhhL7kx5+Bp0}iLZO@0bh_pnQ*+Sr`;!^CRgOgoRzmQ{qm@&E3X zmU3UxJRjzb2x^>EzKhbgIV8A_BlO$vK;UN^rH&ds`H%40Bi=BaExG)ioXB=@sbi9b z){f;gdJmP)Cth8EBp%CcL!26jm4D8kq9^~(`mHepWZ0AG^|DA1H_E0Hpz=`c%!y)= z1yf0c3y{~^MSxzrz6-~)y{QC7W=c`X*_4U zv_!8{rL5%%W<)9}CoyJ7Fk3;WRM1U#eemxa>rm5v?(}DHIhPn#ttG4s7R#E^5?jp3 zXvm%VnBqK;)L8ffw~KB5tN0l6@U<7_W^dXvcMP0uxdlm#&+v7=@Z2$QI(3rnBfbrl zyb^pa=gcWCbwvV;n!2}sx^PPUji6fhzrAmheHF@Y|Fiqet*i6{UR?d&gv8j5JRwGd zJ5`UwJBuNo-jI9CyhSB-_VYZSY9I_Wms9YKOn11`M}u5foS<|ivw`kK4^Gg$6XxSKh+xOl}YJcdW)alU;XBYpfEeu{jT7z@SG1pwTXjE0M9;WViHH z#>+q!LiwwV#)#$fi6M64Pi>4@ST4d+%%^o}d1bKnJM)=)DTg>M9eg^i9;PhH2P}i@ zkD9Fn9>xr-{S&2J-_IIePJ0%Q?6!V6G^LBo5F(UU=MJ!WG3zwcm`CzXHf@l6y6vy+ zMhGG1tuQscgS}Qf^V|5Cc_%3WdI1{;ks_7hum2aPZW$L<^tO%CDcvC*k}@DehzuoN zN=Y|}4Ba6^Hw-XJ2ucbP(kU%4FodMiJ(SWRJ#)tY`#aC`p7%Ll&iS-w@4ePu*L~e} zt=U^xOrT#Q0+|Hp41GY!2U?Qeq_^m{eQ$54VBvK7*iJiHNm~?GCO5LY5QlciG~If2 zR6pRXT=Lu3qgCy%B+&@x=<;N-ASM7`W{X2}8Q?{N_qz=H4K9n6F4~AZF9A&tCp_=! z`!XV=Tu3}Lfw(qo|LbDr?jxyCvIs_X6td>d%|$%|x$0P7jRRBeVFX1WytMbhcp2@A zco~qFaW6Q6uABVUXUC3=1H=3qhw=Cr?CmIAours)bY74@-~lynj%908siU4p77rc( zxYF*}wArY52m9Z6Q83deqtCNnL)NliqZ{mtwj%7gPKB9iR{O9_0DN5Gy#PSi1MA%N zPEEHiR)ZsIc9ap-Py$gNjts>*htQx(m;x&eg@;Puz+nHGT9;a2OtuA{x1hiJkCA>r zdUv%7GflqAN{YQ*OxdbJ^a@ygM3N? zQHMabLmcEDL$#_v_?JZyc6|=#SRB)FJy$u`8L_a~&i?Fpb*t2b zy9dtG+3|!W3O{J#E8r~l8`DhmX*Y4*wR=ivj!Y}@%j?T`-JQ0^SwCF=7UI{phbgl9S(g8Se;_dhVhk2Kr*K*c)43_50S)ZCFUr0K)vyjV zp90dwf;@YH5eRO@v61T-D8|Ap)_J??IlDJD^s^xPw1~XB7zBo^oWq~A>Z#ow8yETS zd&?+(VjbMAlljNLdDg%5t^aN9jc51xCE2sL79sK%#mE`{(T$X%AJkQwv+V z-E*g%`fkiIK}>r?ING}|uoe~^IYvEO8NO;o!(K^?jc7-8=9~Zk%u8y`9Uj=sG`Tvx zONaTH#)|heB#3sA_kOwqet_J8hq7XS8Mb^(u`qoup6gWY`CW0?7q zEUwK({PYRN1%0JbO0^%(HvaF=*VWM+2XYM%(^u<--dMvm=B+5x@gojan(ZbbrbuAgkGH#-5@OFz(RY@q4i`wc)f$-5;y%ynH|u#~hn#%(0t54tGC!IZ6jH17q4 zd?WY04{7(DI$HZX^jAqe#!&XJZ(h^9ifz%qal>R2k0^;LWgZRRQJk+u3voCf7$>ls z=)NJ9U4J*9;EvXWo6kD;2ZgjH#<`1GF~WDWF*-eXp?jR#DB&ruLP+NECMBOoQo;e* zb57Fjc#u{EGMifzj5AbE41+$G6}`|CZzYDIX%?x@mFUA%z9SzVYmE52(A0*{fQAU_a79c;d?!`Jsk^7{LV*-pm=ma z0V?Q02If-o_Z*I&q+pu1g`#?Qy-H_<)1Bjre>E zzl_!iKc^tlW~<7yIptJ{KyInJl&y`Jh^2+%z52zH9qOJtuoZDWe#M!Wey z^qZR(_bTA(naBGr+>hnS%Y=(C`77mC&qJbJ-M_6(lPwns#(9bEz)75br*Y$) zIN&rQAOgl|DQ;Te1xCEZ@Nf>b{=v^p;C*lJ&7)cwkFfvHUn!V}38xFkxxokb5}Yr0 z^DXoJmc-iA-DxJ)2xtp+Z2F}A35B;GnbYHEWKsCCL?BHUoFWhjXgO|Ngnf?40LT^9 z?6SY*G5Wh+&)gQPbv`8m`TJG(mvG^JB&^ajqd|UE zC}ooMzA-IMd;at>i76f$@Oj@NDFQhPd?uarhrPLRb^Wx!mZR+;RY#%QX=?Jn@nB`TX!vStd1dVGFReIolZ&wRvP{JfV@{wigrK&v+-RqG zLn=s7(?@TXP)l9f^!4E~?6?R?M!%SG+D{QZ7f{5yFY}#q+TZ+ZsOFusd2J=O8GSms z$`k>Ddt}=Ya-iq!G#fQh_KAMW)Ie+|b<`L#I--Ro6=zG$;P>^Sy<}L#DFelRp^N9o z_q+K$PvQix^)hd&r1(q{6NzCGxm4<@T6$V9jDK`{IxBudg6PX3JbWh4IUiaFpDUa` zzLN1UqDJuLycuqy)j-kY((k1pdg(Q?S>I(Sc!v>|zBiua6Wh>#GDGkYUE8^mZLz?k z_JLKEL#F!Kpp7g+Mokx9yJ-jlf)i_X_8pnBme_;*K&399**%{1l>-peNCCSQyGHA4 z|JMt(|9-c2jTV!VnFS4fxXOMd3J&g@Vx9#_avgcWeD0(CKF%!4u&_5yWVhn7H?C!W zsn%hS?)*s15mGs^5VYkR0r_n^x#&XJr!%1E^{lO&>$-3jhU#5qd%TPk{e7_i0l95o zL~QmW#Fn@qH?(IGr_om)wY2!&OWUyEKFjiH1g53SJ?tDmD#h+le@Tc44N%wNIS^35)9!3uv zztMl>zK`P9vEK*?oVv-5cR!sjrNd1xn9hDxNVw}R7&bM)0F8FM{L<6?y~dgbbD_u6 zmS&CrBVX{ywMH2N0H664LbzKP@4z zicdBvVf+f@uci5bdGVql%X|e2O+Q=>2wZd!eEPL`Po_W~oXCDaT~33Z*4mA;75_wu zPPlb2fNw*dOr9^5ERdvyQ+PM|rVMw=ZPGl(*Zka=b2$@;jki6;mzLbQcMDOj*A`||q75sFy@&!E}AH_>vzSVgw0&Kg}$Hi)I25o^C z{^YnxJ;*5?D>lsS7q(A0`1HPntN!reD}=b;X-KunU^F`Ed5j4a5F2eK%cqvBkdS*9 zRq+^q_hzu%2p7Xq@y{m0T>XmP85BqaAa<=l|H-8l@dCXI#J2IMA9kRbtI8hBiBLlO z68{1Ospw~#ZSbwOw3Z1N*mygQP)rfQR(try6~FUMc$TChCi&9rTM?_t;FngjpbA-L zb@Y>7QbUK~RgU1NIb-`n-fw0W+kO^01)k)ay&ZF)vB7j;7yfHKvVG)(>3B3)@FQI@ zu~bm&2Y|^jdjtZZLoP6K^QslKH)H|kn#arJTFJa~ew)TzzUS^exoFz;xskX8oG-7w zm|=<_`CZZ5Iq$0qj4%l(GLTc>&6nvt6puxvXImWSL$Zs^#OnHY#J3Y~BFMlr86}5z71G}3YMRo#K zLf#zVWL5^oqt6}I(PfLZ)aeD$VCiF`0l-5pp~t z%55g0oj_VTNt|I0r9V_dbkUCt!m<459BGThRp9c;*fnw7 z%$D3SaE5-(Wd-HUm*uDV9UP7n_wG!6$ma~zgE5u22wa$Y^ihsDQgqZQIzXXrq=CC0uEeQBPnC&p%o@`pA3 zfzi1g0m?T!pO^T=Hh(esM1$%eo2zD0e{Z+cO8W zPBy21IIVu>qtDYNr1|($S=afxqeJuvIM6xiQivX;0X!nZQoiL3zWkA$h>U?zT<#6 z7W}>a2o!CYc9QG-(JDny`I^(@wHqT@PMhc9qF(H(Iy2F*A1uX0R}(5JoAT_KwYJl% z7kUnu)fN^09rY0i9Uh=TyO|kCel64bYwCr*qb7qDv$fktdwe)w&B3g*o z;xk^bZo5iq$Ud)U7AKX#+q$Yd5eJdVq!U+?fzU#otKt(%x`H(BzQLP{;)iA-Ux;6^ z06%Jb*`9aMhu7ftg$qobX45uR%uzJ%LA1^>dy2F``09cCL1{lp^X zY(ityU6suv^CVLIm|;U1iztbdh;AitX}Tn`C)kX|N=CC=RpXKNFqVsxqoDZ=Ao?)H zpUy0@Hmwp_4`OLy`3SEojp-n}XAUShkA1il8#Qu`J+Ywg{T+9~|1Zhp`p-Ebm8g{B z7?Mj>YP(lMmji82zLJ(`-vg$IKjbzVaa=dK66a{e#8ePI>HyM$<;*ri>}%EjknYoM ztk;j6hg!Vzw)W^UTNHo)ZiV<$O;EV$myErYJ`uynlsM^0rIj~t!bj!}?deakTQ944 z+9Y1PWn0;>@a=T6$p8Eh{HqSq@ zpE%;3=}#X!_z^dx!PG@oR+Y#D4jG#Ysej4m)}0K$cj6NRT4^ohjLk_W_|9yB^M1yD z)F!2K0VdX+Z0TuywN?RIwOR+CiooG6ZmrP`C*UBu8Zwhd9COkORfkzA(6M*Ju6?)8 zo95cURNr;f{8iZ&nDylQ+lG@aJ{D*-5sRMqeBkTnbAibS_yL{$<|H3Wi&1n6VJ_I> zhL;7z2JEJ*BzvqD)fHj-B`pMPHKmCeh#Q80_mOsm)wHR)_A%Qy7 zy7wg(K~_y?vlaHjqjZ3xXbBdm&(19>mxDzQ|6>62rpS{@O_*j^q{kaE;3uN&Mx_f1 zRj%WOS||c8gpN$1cwKJ2`$>eG#hS3Te*9ayGqfP(Ea$OxT$3I!}6p1>A@%HVvnB;~8qaEQxMprNAzPJc&wy z@aBM74;?qPF4N#VtJ|kPRquTORH1&^Qds7OUH4z!7HF*-?beGX?vz0(?50S5#h#zI z*Ln>&Y$6%Jp}DM-Y*=g+5RPw%9u(Ft*=`q!RBbbMASHDH`G1JQH z-_?2DDPFvaxFNJ=11ov-Xm&(xpDj?PH@_}aE*d|BVweE6_ zC>Xa{SJ^{u7Ie<{*m<+ylwNkpdO;kR8bWzSV5oTR8jM>2x+LJ3TKO?2UEOECJ5eWM zwo6@Lfe!z1L2>wGi6tBg=wsnfoFv{RE08*Ni>dR{RyXaRlm54OnhWW@G6Pb?_ll85 z#Mf7Gn`~DqeZ`S&|^Bg^0*<1A=#H!|#Ihl20x zofheJSOO++r>(ZetatOF?E-8+;0BI`kWZW|W!b0*j{B4w%qE{+_6OWvGR1tzM|6bS z?^zQAS6J(2{G?Cte}W|3`}{T*$ImGgo_G`%-;@zHPhdN5QeXq-u2QAMnt1e;fvEHE zX?dD&_1c#=BhFu-{>jaH|(vi*D{}HrKenC^Ri!z6RjxC z3aZj+2vrEq1$vWHQggOXDs|asE&+R*!uQx^IOCcoUflO(825-|81q1FHeDy3tc;!m z-5*z$^7gsF;hTovGmOoA?j`y^+4FpD#bfb^+GCXJQM(w=_}a2gAmcMbp1<`h7ElPmQNf z8uJYv8(KeYeA`c5H$@{R$tW8eHf_L2rB{Dj=$h zq+Kg=R0`AZeIZ|TfWI` z9>KVIyI|maAePxsz5>wQ=T8|EdR@fV4>7z01eV?qQh5_jr8oi`2=Fj+aQH$mXiD=8@SNN_<*%tq(M=iPOtm`k+z98K zAoMB2#~uQfRSIZ2(GLhm{Bt5PRA0un@n1^@{Xm{XHnDFGwhR|lBI)=-2>2Oz ziDd#Wy>$)rkh${Vi21HG3j-BSM9+NdM)bFy^Ktjk_Q;hXffs!gUpd-d5})Sl<@2VC zf>09mAWz^5<&dE_nFzAW&9~bp#?wRLzTK902!fHjo$~!cbC`$LYXx#YWGd_tMut!( zK=Cr4s6p(};W>aY=YAbF+(esA0yLF~Z-p#zc1)-%+ww7qLp4MbiTl_49>#r60j2!H zr9%x~O8rEO&Qwk}0*$aP&SW-e6_wc~sA)%BX|LmkV zfbW_f*>a-AW^6gVNE6Kcy+%ac=LbaETOAo%_BS#u??4_5Xg{8?ZvPYO=Np%p6Qloi zqPB7A`kyw&k5vdf@Z@jqa3wAqR%l9NrcR+~Y6pJ4(NA&gBcl z?;Gfk3KHiAYOG^Z#1+8s6rMTp3)Ls@#CX`^!wyIm2egPph&L?@Qf1oXD%h;3x!h{f zEbrPJcGj9HO}mfO%57fZbYxmC)hGQ2DvG@>slo6)Q4{rVcA0iUEu%t0w`gcS(WHaubSmrd=lJo zX6W`6Uuqvy*syw!m(jmtS7#}j=ZkuzXk<(O#BqJ|sX{c=zG7+EyFD=f=nYFI7Foyf z%U54+v6in9OTn5bY_rc9P#%xfD29;gecm#mYt{rUVs}V;plJNDS6a13SMk^dY_Z2F zz#ne-9t37^zKfIYD3pb;)iW=a!O~8&md6z3;~bd4%+&RA%2aTm7=cm)JV4EM&> zI8NLl{6`<|_{Br!^tSxHdLh;PIK77NZTQc}KgXq8(yghmtJx@KZHGB`2;eQ}uOMTBhK9K1<}bo}rSjO&|51wi37bqn63+r@$R4?b0JJ%1N(`2OC)&aN((7nzp zDFvPHyt(y4kO6w6cs}B)?5k{20rH#-J(z!ZtY0?Z#iD3`l>i&T>QNxG%*qbzu0i)SAMn;FJ&lxzP#43c|swfEANW((LtvLn!|}c zAvb-EKB}A=YyN6ltDU3Zcbm9m+VA`TEbmX(T+AwN@#r1#h9s#1bR0#gk`zCS_D}Zd zG}DbhYy!2}{ox|R-H6ZmBj({v>&5jxJ>R`9$+!D^4YT$l=swyD&`DVIS6jO8yCCg|6jR|f18jVSsiO+o$z5AqS7M2Bi zRMnv=#9hfIHH9Jb@qR;4wH?nb9>6Q6y6=L*tT$zWQmzBo?-#hicj_bWh3}EK1Sk`7 zq<|LcjS7G?VzPkATyPt~n~%k9mWgbI7!h|)L>jcgS3OEI7-BC=!lHfg(B0A`b&DV~ zj}(9C+}3T5mm;uzi7xqar)ixwc#{c(No;Y8Bjpdf#kVTS-|P-OuUG`%Tb#G)qUg0h zqza8 zsezSS2d$pIC|~N4rNS&?%U{4E?q_TbHxlD<_fm-Q`Eq29sY*Mm zIOyR81w4b0zu1yhAmFs@HG|sl*abyb`P0D*3d?joi{7yuJ&PQLn)TXZ@CM1V{;BpV zOLD~yjEJXvfo6RbL0X?6WZ(r6aKxzNierD7CyE$v?#@CshPz!ER)uo>y}ra-hB}nm zGO!cT)h2lW4>rb{Nah9zW&Bn|XZ>=+YEXNli(4D!s?0b1Na!DS#002+0)!E9diusW z;Je3#7sV+FLV~Q0G$wzaL}7<+XpTN|bG`0=FP1Ki(Y<+&{wI`lu~-dBpjLXKJ+2n# z8|RoRLZbH8Jeytiyw=k`0f3jIuzJ!~&ZsEr^4Sen zjmtvkdMfpcq>vb*#w5rwV|qz5P~Pivm_k=^=mtqk5=QCOa01Su2gxFFd->(@V>ID) zC$k?amLqsmyAAYU*_L^F7C|zCbJ?vMGt5}|t&wc23DJA`Fms}zQq;QtNejyQDbNf5 z3rgl)O962jFOs)0$@)sQetxMibYpm1%)}_KRa{OEP2Ar*F0F;HAuz zM#TqjlI+&>7iirHBJDS>GIRB{o5x+xie&G9`86ha*wdIq#-MD$9sT8AsI&0WPV$tv zgOm$Y=R@-C2a2voth;}`(3bfhm&ruyeO@HhF@Cm-i?sBKb+{qh&NI-P#Akup_*ote6_+MPt5t6A*)wX zd>|?c@u%M*_l2M#x&&K`wR@gceSUBVVx~l;S--Du-5{j4x9o^<`DK4E)Y-rL;U|AC za{lU2O4zfh$6le-wY_-(2H_SwGR2AJeNX5=M~(@hz;fZg*)e;rY1dlFhBQ+^N}X5j zSd*jS_x_W^*BlJs4rQ7qQcvKg0*%m(JatxzNZaHTA>A(&`30!?l%pLy7EpaGt-8~@ zya^DO*HU&X4$^O`5&ZpN4&dzRTiy$gi+gWK^+K@Fst}^G(FzX4Wfc}(W4G>LPWB1I zXBp3Zz@VT~9-i@Z669s!c;{@_TA_=J1U+`|WAF_FFVWlbrkZCjeMi#swTQLFAVjp? zZM{ih|GzJ-23%d;^L)MIJ9>PQofr*O4!XLEGM-9&-)P7C^_7Ou|MTa6_5c5p1XpNTpp`o(=!H)H;JKb@ z6v^kb=>^Kk$7Wwh))}+V`Y1xpI$$!p;X3W)kX9AGv zrIkPEZl6#xlG)Q!=Tsx53!dbM$_Ob>CwzSt9)++M%vREM?Tg#JET5Hn?AQ1g$ui+R zd=RPkmDXD*=06Ofp|vOe9vOpH&E@%TXKb5ibdK{068j2`e=(xxtp?21HtA2*C8zUK z0t^i+ARK;Uq$EE#c&PnXa4G3)QMz>~E>hp))ZC>f2ChI|l9-2ZqxWJkVz*9Q(=4}8R7gB#441^wQ- zp-eW~n|D95>$T67ppDMsW=CKEalH)#-~DxH_4Eu14D&<9pZllGmaTRbUQulOPWwg1 zlV0mBny9rtT4SL95pi5&*=E~_*B=?h1R!ycWNcC{zFAIjIp;X*LBB_;A^ zuV!k8O-VMjoj?(rJ7*L7CJ9fVZVMX(i1MWLByHP>4#w<>saigTIj~NHDiOLG54L&I znZm%KGYV_qJojf1-t@83fmHbm{LUz8^t@!5ds%I=pq-d=yvfB5=q>VrJ12i?p1#7p z#5MKdGp6u6XVn!RX;M2e{zvl|;@l;Tj@gYq;tat@n(b(m3P;U%HMd;BC8j*6ZJ3&7MTg=3wWrK$Yipq9Z<~#|W^)hw_o%Vk; zpLJ6`pBd+w8Z44J>EujWDG9O9>l>U4l*%2sOaqc$!j~q`$Jx0|_dC~F%^wRK;8c?B zbxEDSIe44&>YP^9g9KQep4u-Nd6@DnbNY{c{fU{pMDVaskyMadpw%$jk2lKHFBwN& zm7ivKo~(F=6_0+@&pk+a96pu&TyK3vn%@O)k)^IodIoo|XYMLwR?;q)WA3WCho~JR zAJ33@`Kv2|okPhh@47Q+UY&0T1wX#3z_s5z`*ve?dqF!A(NUItkU~!N%+W`YED*O? zDqw@SLUG&feEU$Q*-#*=!HZg;NNQk(NG`adT2(f^bkjqX{AoDg_)E@CmLfl`mH10b{!En7kT+OAQLTk9t4Z`FP+_J;OiMogWBJB;6wx zEU&h63k=jSd5!yu#TzJ!KUg~In-j9Bta@HrUAHhXc{NHNyg1(ClH%pDZ|X92beMvT z;0k`@@~Y~0)p)TKw4?ZQrkcm;u7DK_+dMURCRVgeYqM#cC)m`93J%X-bgAUu`SGRz zV?dn0ZhL7qN!>xYN^40(WHQ%&U{H%2S9BnemLNrO8+`_0^~ zsoGybpFw$?oQU-wB2o}R*>AEdpIxQ{&$ki3asmz$Rz|%s&&aY%Pf_OjN|rBw&od-I zZMf%OdcPmHdp%ZB5)j(b8~%0g;miWYgE%+7j^I2skFFP%fFdb-?jjYpE8;`dN)G6+ z*iLGL&pQY%IDA%FpRJNiDy2JlGMCKu&-v4U_pPrU%EzT$@iTk|%07RrH9X?Q*qQVE z#s^5QIE^FCGPqEC)W;RvP(iN7$O5ThlBshHHdVOj*!KKABDJw7l4?Ngyx9R{b&&eE zC_(-1n8|MkvW}?9Tp?!F2j51Dr634+u{}OZO%@Rw98PGT-Wv5YR-N1l(?yb0df_hV zuj{>T3?0CC%4R%?3u(sPx>hV)m)n!M*+CpdQhSNQ!42}KvhOdFTlyBo!{zhuEhFDh4>{>;$kM^}3&%7(xm>|8(V=3g^u=%Y#a)#r0^QWRUyXgr-JV)V3#+TjJbLh_tw`#nJ6mUhP~xu6HyaRJrv?ME ziSv`d3zjK-?%%cl#x)BE`%rjab>Q2(xW&HK>K7w3== z-79y`g;ql-8H+tM#X}8994=RX9#jEqIOp5GC|@!#vaGK3K8@aE|NJ<9ffBLpT?!GK z3;&b9eohH;zA2-4*1wM(66~Vpa?){jugk*7LLoHQHy`-mOPkENx`fay@q~-pBinQ7 zpP=lju~36X;DgAyG9V|*VE?hspR^Qi5o^%KmnYj(2?Pan`AhT<>MZHs7kLP>zBMr6 zy{et=5X~;Pi)w_Kw}gKoYqzV7FlG6UNua3_t|BQHH^v;2o8reZ@A`L0b`K=aBA6%F zP}_ZvHQvXm!6U+g2hB-FE9PyRS8Uyv6^=vE^JjP}Zt4uIYOcx^klo3oU7g^tXG8FY zeAVMn&sTa=&dWnliZ`EED)(fY%ycHoQP--ctZxJG6ju#Jq`s0#8QlBSvhPN{Iu9i3 z#1A~@b4oL~Z>U0F+V5Vw4PQ{)SD7}k?^ROpPH@lp+$p{d?EP(XOdfb7%R0hKnG2zP zYmfs7pUf@(Y;Nulf|8#@^T9gZb4?q4WEVg|F}2L` z%SU(v6P^G5%tkJ}Z0^Ig9*cCcPyLLuKk7~3l+b@-RSC{6=jtV0tuDDKBQ1J81G=^f zHlk^j3b!(1uO!RjKCBLS9n%>0;Qn6u4SA?!yBR2%O|M}`j{U*6%%`zPNMwNi%Dc+< zB#&(guB~ZhIG;&Hy}K>NnB(AiA2k<(2W3+`eCxC9dCJr&z;g8=C0d8(F>gdo5JKa6 z%M)*{5g6KnD2O=cq1`6@mNl&@-mpuAc(kJCQAkbSnGir(f3ndYNC;Rp{L_9;{%ug| zv;5x{j?XUNqCUl%@b1Fu|e``Vg{d`G{WtWLuFze4xl~Dpxs)oE z!)DVwY2XcHL|j$^$o4?y3#2QIf??x^kQjKTv>Oqe8{#R^-zW3H!KkA8vT296ltZ(GgAeQ4#g0mn>T4Lr|kb1d` z+f{hRw_g!Cf<}>r@AmRPJS;()u;efEDz55GcjbTev%>02MP8WVVhRB^Tc>H(r}7vP zQ#bXRh%aOSqrBNzqFPLq56heY#rNl++M+0cr4Xley`;Jm`m5=RIB+Q0iW2Nl{5~q} zW8KMnqIo73_EhzyV}gm?s32U*1fTnc7}22%J)s9dioj2C>lbI~DXaYxIU30+)_uyz z6)dLRUww0d<|-Q&=upjBTkzc;v+qY*hfU{&X+L@Z5G7g~=;MJ^8Ylsje;5UiK^1np z!r|y&S)xPpS4vBtnOKfuj<)`ic!c99p*b9!!J)F$U^rYk-9`eq61(#7J4^T?wf*K# z$^_E;DAU~`cug6%y)*a@Puw34@9xJS2xt6OFah#a`~g1!+JZxgQT>c8?9t}XV}b&3 z9n%eAG)hq&Rkq8XYqP0U9FVC7^#RI1)vP^%D&w=PmNA0mdK18`o!HuFEa%PfsQ&q@ z>t`ypC!uTLQ0Ygw-GQ(W+-wvTSWX*%n{0Q17axujjGIxRGbxJWyh)=tpSTkKF8}$t zwAlTbaVyC*_Qm+`D>nrqJn3kuwcfNmqG^@_`QA=!51)JP=Z!YtJn5wfW>qMtL11d{ z$y^|ev$n(n?eKH^Chj>@jcEnT=zwgLW#U0pF&XfasE3I~UWyj6%r4p54gY4&<7iY6 zJx^KQ!1=hvz`EGn)w1+gmo}+2u}cDNn;U+Poo8_P7LiHbaM9o*PdP@aNLk|2Kl^bp zt_8sOCTemMiU}|>cSYi0+~zS3$AUS#hgS5@te&YrC#2C{fs$}I^ney@am>D~F|+2J1b^T3A8{VEJ0^B@ zDS~cxp3)_=t;+3rD=5~+vr%}Gdh{st!WVr|xXq1NZ|!Nw)M}z50uJ;?IrPXb#P1tK zw7&|5ZWdaNC9^q$<@~JKf^p;R!?FXW13B`^K&B5HSq5x;-pM>z4QvlQ;%2@~8LZ4f zN22ivqRPUr;R$_zCiLDC8=_e~cgq$hYXBp$E8%T^?F4~1em~%D?jO(cCG5@sBTd#(xV<{-}E&gwgw_uZH59^HQR?I;sJB#Q$<=v(lXkGpw8Rn) z&O2eR1+2Cb7=#r`#(2>7SnrF+eqgzCZv1uSXhmxqrAsaW^rzL|UPfeiAB`%j&aAk2W!(K>71>`)Cv={D587D+3 z7ThsD*m^;)2j~cQ#5FP4>>J+7+Q)Km?!Z>OiTjsA|5(E zIwG6^ofZ1x5D2`S+}B~KkqHTKtrmiQ^V-$WJ655=<7@a+)@f3>?`M}70rinoNSDCCwG$L~rT&L~_F7pl@Ya+`KdG%;pp^cM z#3zmZy{QD+T<$L5&rRgdN^dz`kD0EUlc)6}16w(~2qsmI&p%>7oBRYU=iRrLHjdeb zq~>M;>2z2GNeR@c)D&c_hY$4M%aQSp`=4pF{&*{S@qH8VbVsM^E5rQj&}}3W`&LR7 zG?)d34%uJ7n}*AphQC-r{UR-#g*bT|3YoE53-nZd_{(!Nj_XjIG$NsuA(5=$6yPe? zq%QMtp~^91?3cu!oS?0k`8%xT+vj9QMB)5_dDu`aPxK+qAx;{Iw(JHtC10g7OXYl& z32bM20qgqSlWyQoi(Ecy3vGI6QAfu2>WTF0KW&m;V?>UcF;uo>gubAP+fysNJMGe0-@|hU=u)04EW>;hp(fLueDEri?97_aZ%48 ztS3~j@lh&`1hf_gYSX&!R|8kS3~kyy```|bWDCr@{TgZ-7tdsHE|cWY;4}x1!(`6&=4{SB+D7c?sB>r?J03Sdb{_=j42V6{USJA=t)t z=|tw;ZOo(??8|Jmv3h1#6_Oi%k(Wu+FcV`&kdqaB;Ztv9)1d@cI#uzRXukP?GuMAt zvN2Os-=8sh*KxHG*^e}=wGyi34)-Q$wkf0iOkbAyT7mV$s|PbLaOmL^RatrYCK=f| zCRNz)a0K(g1GaKOB5}eJX-}`^;|yp8K4>(MVr7rOAe-3X9vFLh0BYXKl;aNe?q3n; zwgJAQ4I~bhdIUj-2S;FDF84gYOs`jQ9PjKlxhNGMme3sDzz!}`VBR0t9WQ`0%+BJl zUpgfrUk+$4zTl(jEknyg&%0fkN$Y6*8P0%wNALQjZ2NurA8jkd=k{yCSzT2uQ!3iy z1H1@VOo-@U)uX2vb4b&i(kCCKr!WzDxY^nIiujAcII-Dq9|@4aK-d6mNthQlD{PL4 z61GCf42#80Yp5cc#)gHFXf<4um}0{uNCjbNQhL}oViK%x%nBYB7~F{Ev1b_EYRqTb zUq|&Au|=HmyY>kpZ7_?1v0#`*VZ>?t=@SC!DU9T9w${qF-oE-CV8{P+X~6%vG^~mU z4C5yTz!-@sV40*3VSdEruq{Fw0v}&TPiq(0HK72^hJ+n{%l)rn^hHyOBOHW zkUU!JpKt4;AxaDVNz61MywNy`|68RhchBi{c&^%(up@Uh>A*Jv5EOTP?XA=7{3O#D zE3$tB6L8cqGacfc9alI7MRxe+=Jw>gwi;SFnsp|*17Y>&Gn~0XV5UHB&*6*!tsei<|v3&LP1~Jk4GBD5l zJNgr(kNPj-^Zo*izdk`{+ja{FON)^}@?W%ujHA;ak(;VewqIi{rq8pnq_pwRHnB0$ z8Zrb)gnZvh$-*<`N8A2fuIg=uQDK=66LI|^j`jP;lvlX({~J?5F>U{oDVNu}7@{1W zp_O>zl1Gj-@4I-bJVp;H+@mzl&sNy+=cZ;Kpw!~AZ*n8%a~KPZ>VPT%`#)S51b%;6y>>d9fqcNd zEFOhOO!lT352{su{@>=G=XONDcv;U=_vl#@k36VcT3)XgW8;E?TCc+Rgq;5=jxj)# zX#Bf0(En7&{~F;o>U^Chsp70_g?J$q#N4uve5vcT68JFY-Cc*_!v|RhDRt?rQ&=K9Ygqmhetms+&styPB?PmkSK(M z=N*FlTM(CXErTi`vh{msZKf68Vew>);>BxqMUH;{QtHkAJ0+{h5@xc$i%- zxWBMjCe@yh8>@DLl(QN`v;+VPC+2+gNvuc>?nhB^Vfh=$<@F!Nxomyll^%Vp%6Zb8 zN`n7VM(TQFS|Xal8Yy~XLP_H~!o#|V$Ky$lHMEB{e*P0n%RYV*PcLul8(PNiu1-IN zjDyDhm|lY zw%NRh>5b9f{9a8{ToT+xOvXQ%l3}LR7kExcjaw2uxeQQZR%HCY03AT$zuh`Iq?@6D zFuGI1poB?Dw}2?=WG{Z7`aZwseXn|gJK!2lEtz`+0#44}aP1`J^T`Y0F*K%oE}3Lv2X8VX>b0QRrH z2g3m<9Du_CBpg7)0Sp|#{`F76NC1ii;79<81kgwTg9NaDeG?cBK+ymk4It3~8Vz93 z0QRr12V(#z27qG#BnCiZ01O7e{`E{?EC9s01N|zVNfs(4u&DYFfE28Y6sP#78t!$4tJC=7tZ zz;GB84uivCNH`1)hhg9_EF121Uc*Xc!U= zL!)6BGz^P|0T>t<1A}567zPW&VqpM) z0s|;0fPw=kB!EH#C=7tY0w@5C0)tUdFbWPvA;Bm#7=;0&uwWDbMS-CxC=>;UqL5G& z8j8X|QCKJnfTO^06cmnv!%;{$3JphL;3zB{1t3viBnpZ|!I3B=5`{*hFh~>@i2~3l zFd79#qu^*15{*KmQ5ZA|i$(z$6c~eoVo-1l3W-6XF(?cMg~gx%EDDT8L9r+}7KOy3 z&{z}(i^5`20DuDnI4FRF12`mrLjyPrfWrbf0E`2JaZoT04#pwDI5ZfC0pqY>900|E zp*ScM2Z!R2P#hYH!$5IZC=P(*z;GNCj)TK-NH`7+$6??&EF1?QabP44ip0T@I3yB> zM&d9?92SWK&^Ry}2SwxHXdDuaL!)sRG!Bc#0T>(@gM(sla10KK!J#oY3-+Zcty8KfjJxGt>=xxP#F*HTUEKcuH>NRSK`yV(Pb|x#d>eW6Y zly)8hU3=KK$0{O5^>**dQD*~?uU!-!RA;oG4Yam%kl0&*1e_=xqhbyi_eL1DwgDJ z=j&bg1?Ha?(K{!HIv{OU_V(WwT%2e4hwq}F;s)@}!Vv`A{L?|FyoMa^)O)AzC^iK* zzeYUtjzU;-SyJJXFx{*uaJ~5M(1&=K0JVw$FMf}9e%-UN;w<*!ggajR@<&m^-QFk& z-SYCr+{*_Sn6!ta*szs=*>MK2yZYtEnrjcQ_CnIg?<6xXBTInD-SDun)YT%4Y9q6U z5W?!K&zG?g^1j1i_Et%cOm+@#%xTl$`??V`z#O!-97b`Z!t2}Eu9Tl2Ow_tjKzCsv zfdAZHK(GkW*|1eGM6Xty_q}CUFlg@gHou!rdbCrTVH|45uQV@i2i~HC%U<%}D6b643UU#arT!7$d zMR6wpdn>u_x%+dOJ6Ewqxdr7H4UqYda?EJTiur#cb3UY8Ei%b& zEnawpM@=C%4r#%S?Jc%_ixjH!d(-*)VC*%G6RPK8ahi96kcDD)z+-VD0*{Eb;Fy}2 z^{IO9gOrV%W7aP1p8kEU@WW7h%G{IiL5;vX#DGz}RsD%;yr}!)5l-sc*YLsj*{wTu z(~5*fF1bF_dh;|Vtu`sYwb~!w6mVZ}q%c|=r7wH-bzN(MWxcDsjHx!mVN-IU<1tUp zuH$vTH^+PREQfIeEy|jbVU@Qq;+vd5Ufc=qpA{-t_;Nizv~*%6kh3*1TDfPBa;&o% zRQpX(##!O?quKZyKgx!Q9Db&%({56g);G=O@P<9~yIbOUb8R!e2kxMdy~BkL(bFXF zaHf&{s0VXDDcnBL&cuxk)cf7x3MPim?|r<>ynvVJndfU`4@q*Ay+;wF^Ud_ygzqn?YK9=maqER! z&BWvPSwHxSQMHD)XscxqJR!MTDa-w~I#y-k`pYNMy;NbLh1$;wvr&pqIF?Y+O{Sl) zo|wT&%tSgCywQnpVt<}oA^ix;N8)Jw`qPw-Et>?NvW_B`ox-iSFzIoqxC}Zy}=}0 zo*vK-Rq|g5i(M9>_6WjmU7AOjYjh{IQ5t(zyh|^E`Bh5(Uxc<Zfd%|Np8=zSlfgG%ac>} zVbAtfVBLzs$#u>?-G2L)CmTXDEds7?gn=8BA!kLcpTDX2JTF|>cbjROh21a&|B647 zc-Owpd+XVZ$)YS?0mH(2^}`D;i;F?2_L+s7zFobGKT$JXIKo=3K;H!ms*l}rk7{Kp z6O?>fZ*&6>4nlZEmVX{O7qm)L)ta~)f(YGxN2S_DB))9LhFmF57y`X zfT!0$A;OmCKXjMc*Ckze$^1vyzZe!T-_LTYGH-A$Dd*92dsdCgBr?nTXe;*6G`l!! zaNgTjT7AAHHMmssuF-(xl>A$bavdtos!H#I2j8g)ev0U8y5~mz279-5T?cjtRbRPR zU!N!_F5$grxaUc>IaA?VJ_?tWT6G+m{yonN;B0c1tbXM+Za2(o@)oF^ds?nU`4-W6K5<9dM;Sma4HyTAof+@+3_12IXAODSVGGK z`Uv0OeIoaN&Fn^fD$aVo8_Z>iNzYpwW1rneO1%mx#tk;NP7-Jfkq*tte!&01IN+Mo zNjN8=p{?EUTg63cbFNRlX++%I^W-`6qL~j(>2~hVvN?yz(Wjp~jTn0wJO!TvB+V#Y zlo(sy#qPeme*|6Lx2O=f5=2Y*6Yphs&PI@hf|ldp91py#c%FkprIC??%(K|xwvS-O zuaWr4BiFecE!hvY;&l4vW!MG%GRaz zZCZNkxdib<@C-|p5pi3kNu0JO6IR&z-HVW+o61eq$KRc$?v?Se*ERhbUU`*4iI!sT zOev;5cz)VG6!^}s?cu}oN~WOLp#@j;yP%DmH37Ie_ZLl5->{iyt-kM{%Xzo>b$Y81 zzud;jxxaWfx6PX>)O{isavm66Xyfq|?HAa*>VuXF3``E(&~d5d3&`w65KeJ*8i>%I zdi3!S?eY8fFg!mz^*fn1WQ`HjScS7r1+noH;S0FV8u(LgJ*ReLc|7GWbvJx!+I7uA zB`_s?_DS$c8DdymloKQ<8yRY|dYf^Y9ka=Pz#o>s>0YBE^4Bf3ztgJS=imH*7y0&OxGc@=nF9 z!cBPV{>_*>Y}{Uxe8FBmY+`;Y(#_$%CH}oQKOt+bh$iKT;H$yiEt5B>m&D+F)c9ax z22M8+J6Q<3l@eRf9GjKOHCgCWH64eJK$sBg#)mytSs`YKc_8P{8u=5{mdOhDchRaOuiCDXpN7w;=xc!U8@#9kcw8{$ z38hxCq7^Z+;rmX_Z$wSHOe05REip*LhvQ4;E~FtDqgF^trd>`^_(BVgvQ-X9RVu^t z6n9YnV*fY`Pm3un>V)3a!`@+<=;j!a8zRm43y}|mtg)NlQ{E7sZfC2S{-n}gZ#WI+ zWbY=(;5vryDwg4ENp2;Gr(Hm(72v2po}tx~N{w)w8c*}eN^`SJ3;Ry&stvNNWPa1l z{>G9#NhT}VC@YmIEA44kx>45VkwM*ZkY;IQ$}}a(QY_mLY&KBb-R7pIwbv*p@J+}JzUO_HbzxYS1G3WNWcvDQM9?iK&FH>~koRQH& z{oqlC2UdD&6B_`2u40N4VbXC##VMy-#Yl;>03f1sA>AR!Ok7RlJ+&aaei( zy0kzxzTx+$5e=J~g)7g5X?Z=NyxtfkMTr8Y&k-!EcUG;kAIE5t*Zbrax$U*Ye4u%X zk&mZ8d&NSVv*O1-Sr<6?<(?cgUKEq_%;%AiEVDAkV60JF%iqV+I+uaa$%-aV;|6%a zNY-FNPB*4_b-Tc8OZEpX%MO!kfM+17;MRwikEwnRn?N|(#hURz^H`9u`*{@ zG+*#X%Z77~xVGQ$zCKZ*m{y^}%aiO;zIcJU+h>q1lq=4xR%(kX7gUpuE0#4bo}Vf% zi1miVy;*NzM;z~bo|P_aTNyH1}8`oT9@+S0>|U6 zd~4@|PuXQ+ub$=myBOlS%hoEDo51{+$hvNT06Cj)+^>bjuV7^6!u_b;Di%0W6WwYJ zUY$lya}|ve@T3`+4>4aubQ}Ddt!aK5QfT;`bWVnZN?DlT$=A?ETHe~nXMPXWihUiO z(5O~><8Xl;;7HMS;^>}UA`ub*@-*#Wn zsv^u!gF5Vk0rYb18K$ohfsis)ZW^%)3C2&WHdWSSBjIqux>go1 zNHnJFX!q6eMbzPU=`q#B?~E_0k%@~?hH@pNYwg{sX{Ny8i_eBVGnw6ZNQV#p#AE)Z zY(eC#IgfJaUoc>eNMAFjDwzwm_X^GT(roa3sCx1&{%u2CS`!%r0Ri%Df$$mY!Ib~Vj(Q*I$dcVp0ehBLT?EZk# z{r+ED9gSJqczAytJ|O{s|Ht7|0ZV||Uk;y;gcVv_&=XEVD|9E*uCOnfibpMn@jK-} zJmYQW`JuPP?_P2~ihGaoC>c%>&_$UJ*OiWDh*2xYs@9i%cqQ$%Ge7(m{3#z2PR6E& zu0UidUpf2{bme5&gM77IwFcDZD#Q9HY-xkbXTZEqwt=TR-7txnKv8Zqoj0b@V|65V z^t@0SY#`A_E#BrUhty~sDH9#SD7zy?{KQOjjA05N#7nTHJcg|w6 z;+!Z3$x|;O2%pj?v4voch{MPv919e0|K_(0y=Lrf8AfjWvjF)0yF7`_4=0?)LTm3`iDls?Cc_0Ku(sk7mtAEn~eoS7o5|XCFkJY^ns% zI-=uEQDL^VO-s}e%=9w!&B{nOj}qL-v`u%~$l_#+wzspa-rjh{*+9CP<2~)Pnd`54 zm)9xaq|#QggEyJmB*^WQ$E@pI2KLBszGAD`oq^0rH&)ZRGKWW2IS!I?J%>dXyljeF{E6>ya>6K&L&CJ+AE zbHelzAjjp$Y~pCsfnt;O53=ifO|@2Gh7G~%DBU+*JMBGX-Z#zZ1pkzB9n1As7zB~6YMC^ThfY`=#bNH#Ah^&}vqhqA-DQV`` z)1}ekcaQT6i(H7V>yBxrnnr*XfHs`KxI|19L*U8MMet^ zC{qPO+yv1UWNidA z&D$XGY?HUUvmF+h*j`F6Oh)2S9txHfQ8sTE6^!@{e-VBB(`lO;>3mu&AE|FySxR9a z=TMseB7h)RCXU%G9wy?v4N|_3^Oz_O3qvf*X)$uy=afkte#|njoJGD7+*Y2XZe=gn z%6y#3^tyI#L1BQ<>d2{^P;<XoVGcaVcb5Z(DG7Zp4Z^Jvi8tE^3R1++c)ahNz8Ce8u~=Jc+-i7j zI(9_bOjU6E>)H=*$yLTF?K8rTg2~e@hvP|hih(K1!@ChFa;m@+Uv34@jugF5G7(?p zdH|l&c?3W!6rxO?VH{vWyHD_#yQg1Nnllz`?QU^rq@voZgY;z1+L+u{YvJjP-04@p zD$&WOyW(qFG~HmY^Z8+_>T|Wh9qirB=exG&Tg9}QejesEK5Q5y#~tuiQQzXP-tr9p z4dX;9bC&Vy(S=`IPifhL6f4!!)8$;HY_+%x`yTj-bw7MLG4;vy1KoOeJA2b|o%&{j z*ZKf)N$Qt(qJHZL)_pm3-q14UeIn{yC{-v4ffzpZHQ?kR_>6@DC$e&VjC3Eo@W>Ls z^c3kM3ihJ$CiM{s>#4HS4*$^uO}RMkX;Q3o;(Z-Qs0&T+e0z9%TRh>Gc{36L{D%Sv zt|(v+F#7+b06OH~Dc}$Lv#Qqp7yDmPz<=1kzWf8ngMYC9Kl0r_>|d^3{BQPmUby1+ znHtN!IJUM4;xF}%lPI&%#y4|K9x3>)XSb^Vu>a1&=oR~Cg{u@}9Sy?iX6bCq2D0c7~O=+F2- z;@|b0&p{t%EFzi64nIXPDl;a9;q&4_x%qGUb1F_-8RSNDn|tR!JO%LICw7BD35lIz znT7E;tM6LEgvmU(#dXo6v=iuaX;1>cF>xh_A%Pza`jFCT}(>p7+R1RZu|9AzuP4z&0+&SytL8qFoS*-SX ztFK_R(y-ohs`3hSu?AgnY%aQhLERtFso#Mc{X=qyhXZZiKcf1%sUsG<9(!dIK9=T+ z>b;U2mm_}vk{m4(@*=^a{{r2k`1CD1^!8*#8_9jdtkXt{g2`x}MqA^zxn{3j{flRQ zj`!XK{vryn`D>{l%#YEw{P9*hSy@W@`MK$`0N&oso5(ic;BghR3SEepJrERYQCI_e>830M>suG~Cg}av=7qp}~#^mMWnfQvYaZ4t(ud@vDYb z#|t4U3R3l2tqGS>Dw;RCn(2w+A!KYL9=w13w$EjB5pQzTjV|M_8d~+9_8aE~l(*2- zYfz6f{35mLs-caTHH62AO)?k>2I&RTt;WLqC^YGtGX%c0iY_$D2{c>dcYf8d|E-&DM}$X1f`EwP_cxAdd?VFf(71M)RPfwXhB)aV z{@J9K&-pn0bQCk$8s{?=!8Kig6psrL%O_M~tsvYe0L0(QrdMLnaimg`w-c8hm+P88 z8r74>tb9V^u5D{2;lKeU$*SC6Y?F^eGjtvU*dMDYhSmY4X&P?lCi1dJgxMx;=hV6> z>atnu*+x5rh4M6=Fy2=W&7EwMO&*-vCeyvFJqTD!IxVOaNWHE@@+4AJ zRcb27OdnPDunQAjUix+U9kstLnPr*7kyhzCsNp$5KPbv5-6N8hXt4EYO9$MSLlk)`Ps6>4j)bf6qMQ z9h)%fHxBd^RF(*8V%DU4t#HYoTHJ8}zB_j}7(b>89%-zTW+oLdD%>OuIcz?iPkx>) zRWq6+w|*xfNa9<`v#%!e)Ng*4wXcxgT)(s0hG{ z5H0Vb7yt=`5GuYxSESTpxDZdvp5RKsyczf}r*S1e3+nW!`#AVCXl+t4*Pm`P?LfU!q zvrO9Vh%=fF`Wl}}QF$6H^L~!y`ok9L{h@;g=c8N(ml*n+cCjh(Nnl}$Fla}NM@ACe zBE^j&b<1kpn^4dPKVi@LTh#^(QD!f@4`@8p*7|MUYgkzk4G~fD(8LsQ!mXE#*0g{w9oapl|;O zBi^W?e+nZBtj^erp{p4sNl;wJt7Ot8>*lRG$rBSH<+1~J^eop(V%fI5$ zg#bOsWSP3(iLJCQ=QA9&;CK+#i6oo`QWU`sdi%)}DN`AWfA(4CHk)!>5snjV$V$hlU${V?_np^jM zM@HW~J-vMQ>X-dsc8@zQPtzRH7513}pZ_LHYhbL#Esg_hOXudVpDB|D568?eVs<+= zlU%>ty}@MLnss6tki5XtZAB+Dvlzj_o$SENrk-{gudCv?nv_zjvALlY{nVi{3&x5-%PJM|q7hr9*remreOT&FBuQ%z7lg&dBCj@Y$Ee?NY`Hfu}v_)D0pRdzU4j%ocS`q8mK9 znLiBnoJCR$+b_imDto6yHV60uImp5DEkR`=?@t+KM4++^UIp*1=yjb4HqTQLs_h)@ zk22*UoxI*xYX1<|QkieXw?gHS9~PGsbnN6yPD>0WEl*8mw<=F3C9D%HvkHr4KqOu$ zn<$HH=*y#C+LQCAzP5Hk%Ebc3PGzrr0~Q0CHe}0G60%QRYLiC-5Y$baO;nuO5^W-4 z_!^_gR-9nmV$8A*bxk8B6>)bW#~8-m;7j-^y*fj>*zIy({A#0+NbQ}-S^Uj zO17J6eq}}2Zgrl#3Vuw49n+AJe`{Z?a^C~-QCH!MV#mo%@b}?IAW!4b8!UqL-I-a_ z%ffnwWBu)VY83U={%R>CoFmTJ1e_%>+XA((f-@LlhKef}=WD&4Bj@o@#H|TGcek=b z{Sm?EprT{kf=e)hM|~lwK56ryEApAGm&-M&cOI(}@s5jY1YB}wfdLk&lY>MzPb(GC zY{3CV{&{-7Zi}wO0hkinee+SgzF^x5dMksL;NaEE)7G@-bcDuA^J+K}@On{x4P_!o|{UyO#_n^Qo3$ z`42N7CLpUYR!8EJjx1?;e8b^-2Q9ptUVrS;s%AbkVAke*G#bI7Rjlyg+<{X)9p9Z{ zK~21J#)7AZn?$H!XAGr#+j`yjXT?ktcW52`QtFail~+q^ z@e#jRd(i!BNci*pcOTgwpiefquY3QH^gzg@8SrLg6BtXOtP1VFlbQa?lJ2eL33-Sc zd)|*o+GuCXF&3Qu)w92!__es8{T#H>o@knsH*sK9s5uMeH~i?CZRgp@#kQ6_i6EI1%BE?VD=LE> z6Fu?mu9zh)`BJ-83gS}UxkTAg@T&h!Noe;y93W-p0A<*kq<+ATTn9Zi<+C&8VOA`G z*pO+LCBmNB=0I!wtruSr-RX`H6$(|c$)-Kho48k#>pz?{r2w5nus8B5WXLx)9LLd9 zh=)(_%_%b)m5jAq6SUQ71puAu*4;eeef`8$c*p8@f4*;Wz^8Vo|7p|2I{||s^N=@< z?=djd;e@9ak~U-p9|cvagg)^cT)X2KM4^EmlVqYu9_tV?qqEZ7RQl0Grt%?u@V?Qv zFWmzAuaDm|g=nB>tY1Di_+oP_>SPw)^sTWny;Jmb-g`#WchYy0?R(v`Q{Hcys-#7g6Lx~Zzasa&skaO7M!`Z4d5>`jsz4)!2Qz5E9n>-zJ_)mABHm1)wpVZpIqbw9MX;1Snc z@1k2g7FEbOt=Mu4wrmP-t}fTbot_&t%8!e>i_3q=ZTq>p8d}t`c=EM_=-XUI!Y`I! zY}SC$aklj}t0wbcl$<5R(w|F`Dirth4#)Eo*qJ6RiNqmyOjeJq3w9LE&oihf<2r|Eo!bD!7La2MMt+uwbzFKOjNMr!HnRe;|}J3 z_(f5L`S%DR111q(>k@R&^2H7cJ?*^vrg;OkTit%Lvs=?mB)nJK&)~B6c8ITPukO9% z?v~ug>&{dWnT!?G8sV+Q|ke9;w?W@2$cYkC*Bo_VoM2^z+=lD27 z;?D_l!4E$tUG;bGOzGXq_4=Ia8h1Qx>i6w<#xg|lY3AQmq&dKOg?iclgL*z}smIbo zJW4_d_&-$Szv4AQo-kHb&*F|KZe3YZqq@@3IBxz2bi+rbA6^kVD9l&VmVbJM)B9L* zUSBa<08JrUa6(rO#FD2G{iaW-u`fav?1njodM`b65c(B-bqS+wPn|+lu~>*ooVSOgM}8?>TM&?_COyMRoi6 z(F%_{VMBE4#m^l~O;g?HxPmwJxYyCRPal155`_c%zs)SzH(!Px5DrQ)59Gs7_T+Xi z6YV%jLIJD|#sb7`m#O|^nI6x~>ioycaH{HZ4sViNKX%Vu9lOjZ{5Yj$C2g_hv~c+( z={#EiM?kp04r@9faEC2D-9zv?+S1zbO9}y}6rVz_zi7J3ygP$E!piUz0Fj9GaFRae z?2dTHWqIp-%pis;%W);ykdNsaXDC?5D1pL^aS3sK7W6*Lv8;0S7MLCLB+TnWga%Tn5rz4vc(#(Y zB;%c6@oGa()6kVRJ=a|_D!fPAY8ZF<<{h7!`0neh$?YPYj@tEt*YAzV5gWAATa#_l zdDxZf(^36h&B$J<8ICJ-M*a<**RkdQ?Q#Bz&Ip`>1Rc{v9yEnR%38Wv;!pTXdch)n z`yb&iomleb$9#y*CCR^YxSgj0{|KM#BSBm9LiI==o}`0G>;*>k?b`TsjD z^hhlK`DD$9`9M4ag?)*?2ao(q)_=s(swpQ#Zl;yVJnEshv_|$YPnK#zr6@LXno}&b za&G@SmVT%_@9*w7S)zbb%|+?0d@#XxTeuQR{8NQY<#B0b<6nCMS!3>SYA*{aCAM{Q^z~el=dD4udv8@iAmVHHAqoDZ(_vb;o1!GA2BuY~hwPXf|N3wP zu~0PznKMtz@KXVg{r6{xaN=NfIufV*F)JKM*6uzbPW- zySGSFYFzzTdU?=9-_u2E3@h_9nvoEM;}2|I61_ERz~gqa`

  • )nSG9N~6H=z_zc^ z30-2IlORfuMn(sM`y2VeWW&7E-qv!dUvH`3(Vn5EWL<{P-W!PEv|}6B*HzGn6zCRa zU`)b*P8kItwk*o0%c@cXTjuf8pBeL_h-LU8YJ#LwsYY^*Qez2s)CNtGSZHF`qEv2f zQ^j+v7)~P{C@<^r0E%dEhVq|W+NNkxV`bpV^Hv4OUaGO&Aht1 z!<+dCB4d8{U^3^e!Wcf(R*^*7yRBlLyZr)mARQrODKW3ksL~tE#CCb(RBVqtC(^h$ zL+V;83R1-vDX3uFiIYLe*K^ZB6e4BOI@KiS-DSndsKF=xBm)A>N}bP`Xln_!hZgG< z6xyub=!;iPzU4U1w0c$6k+}+yp3KM7o*6XV__g#7+HRJ?WQ4{;_u$w!}eL4)V=hs?+$MO)w8+UzV+FvY$ zhe2!o2UfE1AgN$sq+D{pNnPIq2GutJQIjH@4>cb5!TTThwAm@ttRjp4T*9_m+>pLk z37IF{AK;pocPOq=ZJ><4TlP|$Xu}21f9((U0M@)2GJewOb~Dakh}H`JL`~aTZ;Ouy zZ75sJpf)j2AXSZR<9+W5@_>wqG@t#QZ%c#t`%-z`=I<-kT$RPY2LAWF;kfD>$^Vyi z&ay(t^}jlLr8|UJoO#OeKkJ;OTX_8HDAR56Qb>%5&B&e1iA6{zzRbt@N}6(wJc!LK zIzS8fSOC4pu7NM0Gnt`UtUR@SU|wt4UF%G-@hN{a=Sd$Q0iA$wZM|^{BmJl^*L1zd zGXBLEhRTnjN`$koY#Cf6EJIT6MrJnce{3TLbtDSk$XtGhf55vl%J@xZNP?RDaaIL< zy&e(f`ocKVayu=XboNzyvyc9JQeA_o*OPhH=mfm64iXb*F-R&7AH;3x=U#aN=2z2a z@c*(}mPBbv!C2e2f+?Ly->JKIkT}-zXgAVd=Ke&@2JeXGr^VHi6Vma%{IA*%o<|QJ zL=$)*yW|N7smm=4bsT?wdU(Pbv-FH0tw~$B5h7p}L8lo*M?`GpuMPbCEryK750$1R z!9V8)!a-_^Me@Q7C0yD9_F(I{I}_2=D&o_OG%vp9Vt8K4J|QSkCAz-$txAVe3jW)uYaMGfgxDHdLJ4TFMlzuVDBx%(!K& z4Q)`d6R+w&sPgCUtgme*B4qoi6@#A{2`KQZ3i74J8-vU9l9Fv5O5U@BgL#RtkZ{T(W;-^zX}%SbRv>9<L62&ocM?wNhuN}voKW`ma>VYWx zgMX{7>(A9YeP&>7N#}wUp7xWm!@GNg+nqwTf6p1L$_aTAb7bxoyfKOlxGlvO5Kn42 zoXm9Vea`95M*Q7_y8uP@(n6TD>Tbic))l!=uLmB8jp;61ypE%-dTbnM?TmMjy~cno z*l>AYS5F2}ZEOD8Jz5$!cwg=reD_Az2F~r!3+;4(|MZCPV0P&ZFGVY}j3v194Qg|o zi2vr&f$s_aeW@|M1L-9g9x-sYAYjTTL;u+%o?}vo@ZF*O2y#-);?o+l^oam!1q^2d zW1WFn>{4!wb)0Z|g}EX}0|{rO#I)mTqI7bJ^$X$On>;TeL|?{Z6iyl@Unr&D$#GKP zC7txs{AgEehY_V%i9)C5@)fw!d;! zxkOITqFk4iFYMk98+y5~cz^qU)-m%`WyuNVzQb3TSOzvJ;b%iJB&B*b@o7fc$5}*V zKD(%_D6z&kRaOiV2qK@A?VwV26DLky%0b7UrM{WyaGsz5eHTL3v{py<$oG-y?HMOr zY&@!cuX%?q@q^IMSu{vTc+BPP6S8k0v$f!14*Ft&if?9AUUJ(W^C)6hZsVn!bOK5k$kB0TRQnu zc??nS;i$?D+qc#D3MKZH;gaVfO8-rB#6S}8|up^nRcP1 zx3dHaR>mCOIyLegWss#C@>AG;tnpq^_}+A7*tiu4(zUVuz;LoL zSK;)v<6Yn}d_hTNZ@5KMgCx2Agr--0buGgo%->-*T={h%fJ(;cGg+PIdq z1?S@*#Mjz|AJUe2gq}7t9=>?+JA8dFVA1AD*F_Rm?OU|D|7SeIDN*dP)B-hzNkG@( z^bQ`o4*`2EQ}FdDJq8~Yy;B4p#r&jY)NKue!dPM|2z#uomA5)A1KnSXk8yxjxUPBf zUQN5TeB|jo_e!Ab;Lo`vP@>^l@?%L=>ZC&%!Sz%+o>WyW;%c>^*X~d z9aU9XV(JJAGk&Mw$OokY3MXJ{C^rHe#u6{tqBJfyK>$iXb^j6enwpdAbVbqJ1W?YZ zjS(TL7V)gl>hhmaNV%HE3+K1p;9frY@brxIqFNu6OZA&Zp`GMULb5D%pFPQi6+Rdb zZdo|7sb$K=>Ww-#45Niui}&zer2wG}-jtk+`X;U_VEbR%=7e&=iEIZq#+A!4R4pbEg1e9=S3=sv?O;dCzjZUlCS7IA)qQ24@CTe6%(+Q z_AUz&PIl~vEDa^T_jIMiGrnT1TwA3R>+fFvxr&IjJ#M6-YA8@I*<7w*3U55$d=e$+ zAFulHtxb7#Zd>DW2a(6}#De;1-dy`V4#u&erbFE}{KdO?@MHN$}yYN~XN%s6E(U>WiP|`1*NKVgJUkL*B*9Hs7B+yf!(v z5W(1)Z*$}kA^F0%`19>1ixBPSP5HiE1eq3LC)Ak?&&ia}^25zlD>x$+vZOc>4CXP* z;ban9%hB|{jGVEY*_j-1Iv>hc;%Ms}$KCN`7z+|5e2=YPh)z3jzm(l4S<~I8&|5=7 z`6|JxxPj)i6m8`{IM@7=CwFcYV3)2LH>Kv%7$j()Y5T&-KFd-6+&0Dej|?t62mZ+5 z^RgV@4H=z`XD6g=dBH?NTLlpe#s!5SCB#;dvlEFqTKc)U3N)tKzd#V46(n|hK zX)XVvG}8Lj(UjX)Ud%`rF=}zkkBp7%$bKwOF-6GEk#2LMT$@GmE@^}FR0t$QwqXQT zrkAfP!~IQ7($4UO-Q+Whs>W|BC4BWyeAN9`FqNQp(+h1>W#*j->dW63u%)lv|9ytp zQfK$>^7>-i!@b&uZUf>X^^6y9n>>rEr8xX+pSOmB^xVwmn`Aq%c%Xy9$sn^s_8Xx2 z5BVWT(~hyg$7@;-tt|wDe|*)Lv3Sz7R&L>~VklMc_|?K>gx>_AKg9l6lySmvf=zq@ zM0CgFWf;Rpy4(mJb_l|sYbRn{RKrjkfv3#sp$K|hkh!9H(0&w0NVDiQ8r{7hpzE$E z$-kBa<&$}W1i>o)Rcd`fh;XB2rCqwIn`(;Q%`f#ynaoM&FS8ue?d?LUnWi?fuN;qq z!Nd1|9FM+{LtceQ9DgVe^>)rF~9V)$|HAUdhA@B?_XK#*0C( zZsyJ=@oALc2g}K!=mkF4@P)du3Ddv+g0yYQGgE(S_aw||xGs0@DcqpJ`-gr)yg`jc zOwg}OwrfrHmKAoS`K79d`?~Mqca)5rrl;Kdu{an!zZ14S2L{jr(>OkcU^M%6GT5(Q2~M1=)&`>?;YSLCO9=AUmF6y(3+za;nat7r;NGuLx6@0}o0iRp)x^8@9e(k+j(utQ@~@t5W_Jq| zAKI;@d3zZ-ECw2NC=k-pi|Gu6tc#Z9ny3QFG0~!STa$)($-e5e#3dC)c*;MCsckOj zuLi{({FG~Aw=X=Y?bBifAb8hV4-s+1bd~^Z7div!&VD6nQtlFYnMQr*Qe^U1`oa%9 ze@A*eDyJ9~fB@AP{RD%Q1{`+%DRxaz<2r z%*Y{*ZdlqT+ex{}R5O|vZ0-)i^V1@XnvHdIBytsqO-v{fr1HRr$ru&ngem1eKTFCU z7xgL|_K(PQr~~lGxT|* zIGIYvXz#^H15q(>Szwwu3_R=4(Ix%z`o=S1cd4{bV!Y||VOdFPdrvOX4UHS({`(Ax z126$({s-M#2TL7)q1*8AD$Sc!{cjLib&_uTD6|%eUvm|!RukK z)sB22hZ9yIk4^cTikVbSMNc5QMLZg<5rRvfL#qtmexxJw5qKX>%pUQZ#B6!Cu3=MogeSMM|#bh?uk z`To?cQc}niJhPeq>S}W3@;a1rWRbUP8vY#Extd&`IBs>{JOA$N!R)Gew}$R!>aJ&Pdzt&s0ct@krGaD`1eSC7~TyevN(xtR~((GV;@#9@26 zqSr|Qvp`bk=A~$X;2}$8!A8f$I4v(eGvwR5Nv%$h>(WA2>`Ril`))BYMNO%;L zm<4r^A6zp_;>64|xs++jws_*Ia-~Wjv=2zvlNf2Xc{6ns8CNpNe?u1&*gXT*((PLr zc`}{Cwl;FTg(55xoi_xH(*iW7O0z=Cj0EB%g#J4K4M6h0KI8yA{+GcX|A)cEmuo_A z2rCB}B8vzA&0xuw1SR)}Cki?u*@*|I{IyCoLWNCs?quq8KYT&ypueYz-1=Bdz#?=! ztYE?)N!f_`A{kH*8Gk0@y@$IvrIiM>r#8oTF$@|kzeuf&#mgb z=IyYhD?^IFu^;Y?=Ib-JX$!tHj1KR0iRlOnqRe9uW*Zc z-P)cR7`nT=kra>+7`mmqyHh|AfuXw_q`NyM21O)P8tHDNl$Q9=ea=4abM|?^?|Q#~ z;5W}T&syug*S+F#8u8X1m6!Cj%#9e{Ovdq5?X_7K=U8e>kU%xmRtXTyEHjIjB(U8` z;KJCTePH2NNy3_9R~E?M6)q zD_ehvbb^Bs+c~e!cT2NhUbk=L{sCD_>E{1{ENPbAqGXghCWF5=#{B)Q|LNJXs;fOe zZ~eyO_ghaQe4JyTt3vSP$JO;~bsp-?@Y z=|7+P_39V~_s$cYEYD4$aQQ<3xVA~*>LCo+Qj4ZWPg(c+7@(2WI9vtjHX-s#Y`ywM zfjjOQhxO53<@s`Rl~PAd)r^zXRNM2o&NC9eMOy8iX@pqPisKN>BVIX!){B5p%oGmm z?y0NqbBOyCU+RCrAVEG&xD{OotsE;yG`o6f{!qwzI8VDzSWo==L#+5Sf4b1@Bh$Yz zj)W}|ie1a!i3$ymjmE!o6t};L;(vO0 z+wm8%C*X1_Ke_j} z84n|x*7z?5o|o+#Bdi)cqwqEe{G-8At%?>oj^Gly3E3ka6Bv-R8z=7-PL6iI97zh( zj5N;YVj2_Lw~{$fwtoI36~-YZb0zjMkK7piPclE{B<76=4=1JrNHY)@2%B)(CMVr< z%BixpsMuPj=oppLB|Jf!EQT1j*i1*KI@;nOP~xAB#UfpqY-NdiPK~KLP;4-~rW$Lf zaeO%!ULr1WyRH!lWzbwtHJ?j@r;1`Q!-^HC_tMXn1A&k@)6Ga^IiC#f5_rOA3d}n| zsX&UH83iQah43*~%mK;YhAvT+GNvM(hXbDJ1Uip@Nx<${Jc|L4UH!r3obc`u*Qkd*<Wl0V4INK}2~o+KPm^l~D7hh0DLSz1p`iDsOwOiiCptz%6e-&<_Xpk?*b9UxTv)!>Wb=;fR2_9l!Dwsj(4;0)H z*y2cwg+z5OsgH7i-=(d=uedr8MtE;MLWtv|8po05*GdO~&6u7`-|r*3w(G@&3-ni9Ph}pC)j|# z+eU45yrt(+Ar~q{%daq|-%}}~b&CUWH91wq0`;4NXZ;U)-5 zMc@qqViwuN3#;z%V@2Wo>}Rmd^oC}}d1R;rz@IQNWN2bA?_`^_U#TR!b__^HTT+}r zfmXNgR;_Hrr1TOm+*Pd;Yzgda#9g?hHcJRrcFOhYc6IPwvnXJI)KZier^N2Lp(NWJ z>O2t1Xaph{0QZ=9Dyujb1HMG7uZz|q2J}Gy8k3LJBv!4#_)()CU!)_myUcxsjDqs7tUib_)ucEVyrMp> z2?8Ko>OcU$!zvv>8Nl;ki@tx_BD_EKzxSK|zYpU49K)PLmDecgDd6`1+8K75A6KGM zC~tVh!&FeN)0kLQ@oNyLo&`-I_3LPZjcY9U<*6m)uP0Ik@IIUf`mYn+|9_rnyp!>d z>Ttyk_E_k6sR}tu^vihWpQAx~P1mF4)4xW86j~lB_*-E=QyOlms?Pgsf#avk^8Xzo zNd)u|{C+4`^S;@#Em7Q%+N%Dyhc)~Y786V z92^2tx_W6ZlstQ=o}tb{xzG+Q-fS-oQO%767fn>o240|CE70(<)|CZB=41;Un@*Iz z$PiJIe^JvV={z5azSUSen*#cV;ukq{JnQ4o+!4a^g=?{7@U66CCSA2tXCP&1H5lvJA-U z$(&lV#)gmTEiBUXO8BvosO*@ql1&9Ci(^e%GOVO@34*P(CDe-7G7P#iiZT&Cl|Qf# z!EI}~|Lm6G{_EwU-;rLj=D!O?phnTzsr%jH;7R4-<**+tj`9e-DeS@Gl3H~L9?7}U zyi4O3YFK#Sm_x52K}H+Q&*}O@WbeoDhLZDoQsi*xfr*kPW2G{TEQ=jY%oG0Shd?;^ zDhJ&Lf*6#W_a3bJwJy0JRQN~fx^H@MkmJ3%Z1rngR_AyXyqtt${qGU1GK*?-Ev%BJ z@WVZ}KK4ZrB7%9^Ty)`e>*ewDHyZtAR-ka<>yJsrCBj z_+fbr>GmThSeO3ww-3b5t5mO?MHCeUJ>6nc0_2uzZoe)?`*F<=Hu)JAiuy$00Objy z9j@bQ9aBZGCYQV$g7yz$a?UPq-fvK!FDoeH^N%VsQ1dQ9nIbgeZEv}lCOJ~5FV&zp z1sXV^#GX4p=TVtPLupcxBqK;>Pr@QWvkhwc)iXHj(cIIrg=*&x#rQhEG1g+KNTmZv~bxIgkAQ63}$)BUhHs@OsnGP)xV`)x* zRQ;JDX`Wl<)=~tA5o)=9nHDm62!E9X{`PiBE;;@$w@V)@`4CS&7_%}K80c`93}+Ea zp^Vz%KExA5Z^TS!VA-wz{f0cX;a{0+TRev=AJurE|W}+5kDjmbpvZoPr z4h4exQyn=?OnM^86to$o3t@@PzQcsF2JIfZZt$lq)us@I`#Us)3x{4G@GU)8C*jOI z#0Y=Yri6H47rzoHE!8uiOq0T26 z&X%nizFCiiQoR)$?UV5R-tZw?Y%=7QEVYj{8?E;O$cxD1$mx#b#-Z>i8-n3Hz=+|X z;7-H6{^A3@?@D$f#%@;tqe*=zvui1ie8Zr0p7%n?DYg>owS{relUu@k1vsO!)ljlO zZg&UA_kIR9C4x!XBhD6@fs_(OkHTOU_5@9%5b}hEKR$$Xwr@>B>{D4jruffvQI?^6 zBT1dWZJqOb0!oEgr3Pm=Y^LCkdP8SbzJC1D_}sbaR=LGtkxm`8E)8Dg zL=t8Q0>M{#e~0wSe|h$-z6r7l#R8&kz4oEnnB)~nJl*c>{4P0dxfYvs*VM%{|49fj zS=Q?^g6RmHXe>g^nYTZyIVUXczD>scncjTa7v+|3)bEDti-A4W>~v~SX;&eV|0n7) zsvv@x52nv|kS(agK_>jK0SK9h;Q-VE{xMNGx&|`NkJP}W2yPmUOyLN3?MxT0^YBp# z3>?c1X^fJzr9?CVI{IV{En?jSfj|8V_!F;10b&5GhieiC!2Nkm6gQ=ufOnz4;kDjr z{-1dLjrK$0A9zjQpTzzPuNmI@fZ;&De^wY^55Rr+tj2Gj{bLXYa8L1HXuw-ahJVoj zGfC!u(g5T{hJWsd*pc$#|Hlvg!^}7z%#8lwhl2jvEBU|U8|teu0is|8;oaI&)ohVi z7Q<_9=HEt$26ae_8}Av>G!xr{h#KKyu|vzBHji6 z^;~1CurM&8=ReQIf>&2pJe0;seTU<9G~Jc;2z&=3ZY-6HKkDEt=Pnu&_o47*5%wM$ z_*I`vw}X|y{1qBKUzx#RaOE15kJ8JVtNMM^CjyIK4pr$#T@XP+GWi^}i=4lUcfO~q znCnm-+#@Gz0Is}4JjOA4Jw@GPDZ5rsfN6X-JcQvf^E}_Y%W64-QCIg5O?9e~)u0fG+-FJGYunR{ zBF^E$Z>A@~y*~GhIqscT-*+Y|mE)RI*e;HjhvRy>@75^cPEaR%uz9>j;h8;CeM#if zu)GlTf;C-mPGn5H$mlOuAUY8r63jqWV5$(14pE*brT1jETKTQ{S_Iv#8nvA4>~~C~ zK<`T&FUIL~>=@pma8(>A9Vnaz!?hJJR$$J{Jb|f3c1*aJhZv1hS)?3ST%=5cI#Gb2 zh$sfn%yvx*9f*UJYypC^#M#s$%&VOEWB+#=e}0*WAqXLw~IBSpgtgeRuyQVdY% z1xULm<5@7kMHfV$BaMgpd&=&lVPnfN8ALqI*e-#k^Onb!$z9t@oimA-lz2&Bkrgq% zC$^4Z(W)e&sB+M9kE2s^tFSI5N1@7>%MF~{*DQD@zeiW@G8#QAZ#|EbE*-hblme~LH*@vA*Sn56QrXE@hjVFb+o^efCe98T^c zxhP_eOdege2gbx{-hgN0yrPgn!i6##@pwZC$m=5?eOYTWmJNzSDb{J+5(YleXQvMz zFHG&ZzTvT5F_Z2%S9ecJieF19j+z7@AW6J_~*$qt&66L{nA>ZFSY@fCZAj_I7qKqB?JF7jnO@uS&P;e{jgcl~VzPTE6MpH~`zqex*TmL}vNcx= zZjF|=%l&{ddIUr})x&xY5B`YVl!T2o@iWElEBTdT7rwQsRl7kx{#=nO0y!zFqkoaZ zp(=b?sUlcizFm%R@>Nxd2!$y%Qs9G5+WT7Dr_<_0H00Zclp==l^&k~TUYQ0#rA=^CKoQ?Su=EohLr4JN{h30$ z)#_U;Uv58#w3>~fJmIA`>F3;5q4LFm=Zaqr*ChibhGW}CPxV@GPnJ9!UmSm3CRZ}{ zv*8%DdlT-32=|%H%63)1o_n&IROhVY6K)_@dhyC?gJ8bQJ*7N_sq35A{Ssqx{T0uP zS>%H3>~G7CZ=HrwHFdiL7bQ2}Gg3*cf!juHVDN4WF9!kKD?v;NjASww!Qc!J7#&E~ z$Eu4F8b*b&VQMmdt3_#QE5?(ld`Q=@VnJpV2A25ltzj;)ETx$vsd5WXtt2pnFcnYz z{&GE`IRRlTCOHIYISKxgY<$8;UqTB{qWC`iVC|Qa#d4m>@Zc2LROV4~#b=Y*0Frdu zyetv7jtr@*|$g=3b~%w2-|re05$|G3*5bDDNVdWj8kX|1OuY)Wigp9{r6ma?6 zSYN+-{?LV9`iFTcH?MxWXqsu%eAYLSaQ5y|;8%yXTO<6p5Jc2um#(+SRVnX|yUz%E z!L=GSkj?WTXh&Lg^EBH9^wsd2bh##x8$#eM&hsn37tp++k3@k^;n1(u+X(yxjw!0D`Dpm`G(NQ#~6anapap&j}ZLPAAXB?!{;4>@29jx??<0ZDeZr)TGYHSF7!MFH(Ef|N1NA;iwFa@p<|#^JZ4j1OAJ8AK7E%oUJ^JTX1; z>J^+vgvaQpYo6c8U$w4n=HHjm;dJZkAtdTlKt-)9pA0_ng#!JI{ULC}xqS8CN9HFP z=dA5hyl^XVec|C0lSdv)UAmI`DNSe(YgHbh-9oW3JK?&eK4#@Afkx}YyrW;Ox#HbC z>*oM=2fxJpf!K!<^!Q5ZnW({Y6NPIr*blFGi?;z$0@D82Qo5v ztobqkz}pr@7f=16F;q*hL@q8xR&RY%6grMVRxw^qqo+J_2t<~c`ZI7Al!t$ zyCii~3no!bw|RASC}p*BjC3j7QlL>^(wTH4aiM7}yl^m$QnhC<&N@F_AC~VueE*S{ zekG3Y=^Zmt!Sm5tGeki?wj^FvN{|U@^1=0qT|~D$ca87q`MU);|L`p(gFJFzp%4Y5 z;X?Ulh~X^~A-^HyY_Cb;9mZlO|0jJAOz6Ie9@=?TfzSC(O{bk}#7h015!Y z1Auju0RQ{@t1re-cn~KR^zM^&mhk<;cg!FR=}VI!tt(nBEX=#NRKa-iJNeEiB0RW3 zwCqxA!Dv8}5e@>1Ei*ts01=G_-mUp+L|8MwX#{UTN<#GG3V4`BQnT!=sb;TDnjNO_ z7qakY=SX92c%L`&V%eYBPR zF{SMCoE-S~t*Yr-s#;TQ7oBd(nt?dw7UGV#LAbfii(7O`Iw!jX!3J1k@IZ9_aM}?4 z)WISvoBZ+z%9)oLEKClRPZ6%gyLW|EHBq?qS()?dRg~M}&?#}k0HBX`PNeAUBzrb>Du12xR}wiuSV#6tX{ym zC09{c?m9eL*Y{1~Q6;u2@w70W;t%#}1b&ybl-al5>D{Pu*z7$k*I7s%7}Y`(Y>I6@ z$5pK*R+H)RED!wm?KNrD5E`?6_B+MJ30M~(zq zy^G?(hw;yqI|G^Iau$Y9^^?Di-ctN9*E#w~#x{ZZ(boK3MLW`om#D{negt&$RG>1Wbmr2?o$KrLE?6lf^Mu2chcEkWq2>0bVk+8+FdPuATgLQSy(3CZ z-FXYbUti9VjBPvOduT4mZf7s-7ji4VOFP@rK~MbjwwA_jDE#(#J8#>PS;LQpVjn~2 zO+dVIfTen_&U9F61vG^Frji1z&ixb^=@86?jP$$5b3AxF>wkE>ET=@T4;rn!#g6ko zvOA&^!hd+Yo9X++KRuqw7lwxH`p@+C8{$KH>)#@WOj9;ih$xJKPxZY}rbwj_!vN)5 z++?MSJ(WY|XsI?(t3DE}rw*--pqM~8wzRIA&$h;{kRqBk5e?{O&v`XWXbOM_{hW=Z zqQi;}0izuLXo<<&K(bD*z2C@OkiKJO^}19pS7h>L)P4`M!Qn65U}QR3iZtK6k8()q~k({c>y_TP);p{!p1WVV(rSGX06ytgKGFvf7K3`vXe4ODypC=#! z0005qh<6xRN&L=KwFf_+ybyT4-2=L5e9|T8v3K8Kzx>&-;Nk%5Nb^n41@9Z^;Z&o! z@`#_XN5~6!x;jQ!E^O|N?h;Yk{fbHGDYHc`_dz z3_`v>RSqFuIvs(Mf5Te|rk>_8jCkr8zV?ia$PthnEXo|rDSE}dMB62G~`J#%}J&6X|QW+IAV1ifq={h(?xPdDF8(vv{x%`|r^)kynln&uD?q~f59vEoe%$*$@@S;+W+A`{cbFkS84-@ zEq_WpCQAN4;nnV#;o&}Id!-cp!9?=SLQsr^8{EsEt^zQQtjGeGMk zRjhdsXn$t5Li3hht5JJ@@i5`8GBa$>?QBmgVX`|sDDYN(#(wRC@BuyYSA5lVllaH` z2AxDY+clXIZ<4<@J8Dk^U}E;wU9V7%)e#zfUC6Mf0cx4=-+3tm>LAIgjBrhFyas)g zmR};IBE|2;i$sSV8crA%teEF2EcPpMi-A1>`Q|KZmnh*uCxElw+Wh3S*p~yFl2BQsz=u4vG6d4BM zcetyDQN<9_NHZn2sVE63@vKQOR9F?fA`vVqkY>O8jv7R&M+fsoS#%#3mRKsF3e~1} zSAQlu7fu?a(h*)Du5#<9Oo*m2Dv5uwpuQdfFEzL>YnJM59t`?PL>*dop|}!(v+J5x z@`I{T96K zO=&EH#Ry&i=o(t?XC=Q~YAxSa5!rLO^&Ax!k&KJE*kLGxwlb%ige6x5`S=;QmEZ>q zYJ51fUyJ41!BoZp5{PF5EsRUnHzT~aGnV=|omS+acdvt6QYiBFle%+|SKBZa z>B_6$Q2vOH4}%-WII{xsLu>_E{A zz^x!4;wm#;y5bnXK0E)M`E&V-&=o_uLjk5x`|3OCE81FsQi}mj1Q#5yc?k#ik1;8t zEwV9hF>5>bk9V1LDI*|g*>fAwZF|ON6!f1?b-`k6$CEl?2MJ^F3Fw~kzt^@n0l){- z*ZFVEQl4z*$V4jOg5jU0?;r4o>0{obYcR(b1Jam&hO*~*vH?%*|09$sylrA!$x(j| za(giSW@>dGIRO2*-dN6Lg-PXBCn1_joe2yCd_2rkrC&gZ+7HItdBoXd14oaIby00x z;CPVTj9TTgDrw|+XLyrTE+pl7vII^=!^-Q9e*B|KXol1BU8Y1~uVObHs_|4EB4%_lbnL;4^s?s z@F^WHB8V3U@B6b(-dsgEB401U4%H zNT9P6anuHDl^|s4KX%@k%G}p&-~%t-9*!+kuA44IF{*kE|-_?hrnW^RxLmb z%i9J9OjWNRqdj*=F72n#QEl6-W04&+$Pw3ZXrFBnIKr(cTr7ua2EGvzbnz=ZQC*9Y?YY)gLhX zrfe19j1}fM_Fuk&)>U-K75BfVsB0J*<90!<1);vG>E~H~;Cr9N^?|rphI=%x2hCCw z)NA=y4&W{p3RIgJzblR^5qCwaL4pA)RM@w9(H#6bNfBngw( zh>S^nYv=*;*>61J8D-N1e~0|Y$z@2XVxhd@;8-ajmM>pVs?B4q#~8os89D2e5e~#U z$)Fv>$ed3$wLKsZGqw6iZ_i zMwYb~FA)&9yd2j?=}}g^mB*cyyaiAf=AN&c)QvomzNhp*HcDI^B0NHe*M#^4e-25? zRUT!r{yHnvpQhk{Ckz}sX~~d2=+aiv@70`M4iupW4H)JE3)6iZ5)e|oP|w50{Lh)r zO}H->)+fX9xP$Xv5hs2ie{p}(XJ|psoW2(L%=A=00K2zuIht9onKlmMN0$^JMu0pX zCC#8qrck1*fvY4xj;J9iEB&)dXjT4m@_TnP?YA26zfFE4L_ktJwrmS>K&B8Zfa9^8 zF}etuMp<4OFmlI{3J=dlJPZZPRztMEe5YV+mTD9;-b|hV1of8JE(9eAjuyh(vZxeQ zL=GjBgIvDt?Y(;PQC#&fig;M>(J(qlmYATSgl_Fz1K2D#?d z&}PRrs3Usu47a0DO_P~|P)oQ%P5Wl7kAuX$>iifb2r!DFeYi(wUj<%N!;0k+l-v+M znARWd?tc8rLrJ7fY<8|Iw)<=T1Yi2!+?ma@S@%=>b~;j_y5d>8=dGuq*Mi|=q6BUQWS zJ_r@TqF|}h7dNR>UWI+}0}Q9x!t0XI?0vdiXvjxmtiE?)kE0F;W=;$yN{~Hykw{MV z#_GtWm3WAVxIYKo7CV08YdH#w1|x9a>eLv7KDKXYuKLmd`_fZLxkVMdb(iK|?oD><#doybrJOh2bxr|CxGW=guj#M*dck&<5q(|K=}H~Cn(AWo7U88;HVatWl%Z#F8yD8Xgo zktiz?;;}~8Da8fflFI?5EEZv6xhQ=$WyNprd3KB7sIlxFbmHgi#OjuYRPo?|LfC+w z8&li*RwQUugTG2jN-e74R2-yRkrmRMs-@s4I;*Ep(Xb-EkMm$m0Qh#`YyFyz-GWMX zz%k0t9rs17hZ0z5$+eIlC|Ci-tv}tR2Fan04Y5`2_$^B7x43ub3r#Dc*b0_QUFepG zneS6QUMYjHQEh4a;l=WE-{0p|ebZBGe*JA&ZAzPV5Oqus3vka_>iXIKb}dd%kW`04 zdZXad28rnDhpb_y`*v|4!ZT6j{g1q(c->Fjs?5+Xxv|le5-@D^MV2vQq~(qQ*(v zJYI#T>l*#rh$mt`+Y7vtB~O-b;1KzbV@HA!jm<}@r)03&rKTfJpF1`_1s{Ii_X7aY zACkY-&4Wj+Q}f>!Tz3@to$rC8jTi+#ic@1ohwUO}gfdM;7B@e;W165ISF}8ShaK!> zK0Xhj(_zoSuJOBIU0w;4!KyW_eS-EVt&K!Gf3x!)%7%o=6$XI!OK@4h(Hp|)o7G-4 zz7bHZ*TS{VJJFePX!k^@y$UkH&1dd1!{7A zn|DpDylw_*0D~;Kr)o5AqB5EiT`^N&zrFJJ_uo#H%_-Ljd2h{D z(6_m-m?6vmdnpnZzzJ{yu>4yK(45>KQH?w_{9pMz_IhnBm4wyJ^Z25Hf8~E*(YL?y z`S2%BdmB44Tc3ZC;0o?QF(i;Za_2={`9z6IuDl})ET`9BQf#U564hR;R`;D*6U%7s zOt=X49h}do^;{&-Ea$!a3lqZt2h7Z38;_&eAp4_+GeMn|`5vs^t!f`_fO!}A)AiXt zXYI4$*vBu>(b}rF^)HMTe|YJ@KD6m_AsS^9KbEMxXv4A<&+FA{fNP~`+4pKbYDS;E zYq+Qw2v{G$B%4)Isb5^K$|fy(lqZV|hZ#%C4;3DSEPM*Xx$b{6`@(w#d2vZS=V^Pu zi%`^l55q>T(;w)!ONDh^!ftRFD<%3maf=uFF7S(Q;eGDNTgje(Cf?EaLO`B@dEWB2 zX;_QtT2Z*;F%(g~K(`|_epMDhzZ#q&aQZo9Hsum#;h%r;I+#uHboeDbg$e0%fwigm z$g52T7?A7w1B|WCRTfWY8$3NF^-IF*r`87IBDJ4c*h>TwSxD$>UPZ#q3A>E2A*pK9 zx`3VNCbXQU2eRV376rE4F{hZZocu!Jmn1~QV&y4Xr6lf^8Cb300}59#*gesAW)0z3w|0vP`^2bY!Hz<+@p?p-|5zsoxo zUjKM{0Vg4%JqA6OiVwV=+W zjwhFC^I99tO%9%%1bDH;BB?B zJQ}!WW<0(~F_>7wbp0TmMiCG@!wz%2kzdJh? zfCwN15c`X>Bi^HF*kGZ){gv@vu0LeFEeyZf#*$;H(2XALn48E2JV7XtWL|kTkOtyG z0+=X6q>cH&m*1X0X6)66Ke`E&s=;`AqyW3)W2tP?+jm?!>HN&6T%&1UJM}A1N@q2N85Sf zgA%C%!HSjTTG$JE`=zQU(ACl?>^bbVD~kDvm>)QIuV$&j241RK&AxgpP%&?z2NPv#AIIwcZr3P#qKh<(O3Xu5iA0Nb~QJE2pXg7CZ>k7`o4bk1C;)n z3*wvM?1so$M&66#l0YRglRj@v&R*Sg`nhqs#9y=0YD?cyj4Hr0CFu0s-t&dJ@(3j< zOCh0j8JV*mXiZ-P^HDk^fJ${i#AjEGC4f7IwGv@B7OcW0<}LUfM~*p!Qaxk?+YcOd zVq5kxumfprF~>-|M^40+JYOw)=!)kWWeeyR!-EIY)3uE~=OY?}`t5Lx^~#Tk*a z1c(Zfi<5J8+xX#|!+um}C0=YR4s~q;NgdT6ck7x;OCg-A!id<9;(Kw|e2%)Qxn9@4 zyTV9z>N8h$sFy^Oc*P}qA>H9Bb8dHjBHbPHQRnTcV`%N*Sjcr*KdHp4N5e?O3Xg5C zBVRuL42t$405zl%Ls|5-*1=(!^&D|5+*QeynpGFDYqR|pLS-Js*yj(+I>i*Bumj6)}u zh@tfJfrCrNLz#XCbqt=1Y*^=3b2cgwtFX6t4Ab=Sf9RIe!|)N7gpii4OZW@~`&W zzCC+ufO;=vFx9YKl}Z2NPO*b=Yt&($z=1a2b@vN9t+L@8EP(w?nB0{{zfWyM<6z#* zs~Tto-JVc`2xK1{6Un_25`#mUo>n&AVO2R!fI7`l!QumeYkefI(liV59LB-mueXCk zOO7&#JmPm@%x_&VW7RqI(<(}%-XwipR=~{+i3Djg)a3Sjt;j8QCHRV!cjxrMr4LX0 zqgpJ10-|d_+wVmcDgZ8^?jbV&&5kHBe~ka-T*t45PFa2Fq=K8mCgwr?XN7?lEX@_~ z{o&xKnupHTC@Br!Jqu=Oc_dy1SDtAY<-$7O3P1HWM8T8YpxyIL0e&9ukzt+RmuYkZ z0*=)P9}fhgS8bK6{hGuQHP>ElpNtW?IFcP^nKfbA*=`uwP29T$p3f9+ze_Z$7j+xd zDn7+35SvG_9VS3Ap^9H*zDK5f*4cKlc2{@T*t)Rdx`>)@-zjtIeG)ewV80y8^+Wn- zfg{k!S=RTJSM1ivJH^*VT?*%i0WEPoIo%^y=lW}ec>?!?w|8Jd6QyJHa9k)tYOqNl zE!zN<%q058N|3k`(UK8?T3Nov5PeF%KP3TPff0@q(?mE!N!gkwQO%CZ@54P*57~Ue z--^Gh;{JbK%k}dwCp{p?LvARNi@e7uX=-aQ8BY4wU+yZ8SC_Ot93c7}<-M$$$yiVi z!JWbt2lGU}3i%z5tujiUBqWintF38Y4agQGO>Tc`LydtrN2* zdv9?DqL+s@9p^X)*)cZH^}=%v15@f^dFZ5SXEV#9&FwdRwwDILC0 z>DF*J%Ez9wb8<5MlCn>NS`f zw5g7GPbV!*Z_{x75tKfR%5!j8rU!E2gNY?5`b}*23vhZ3@J|R!s4c zpgLLXC~%IBdf=nm>D(CZt+Mr4GJ%10MfN+UjXx%(|7d;Z6Cgl(_K)QrMR8!$cDiIO zihj$7Y`(lG)$Et3MKf#qo2blg?y}!8a_+grv8~Y^wLP^~aZh=sbpnYZsF6D&&-|?TUkwbm_*0 z|L8xkXw!^uyxz|LCHdA5P`CuAzt&>36Nml^I%M`R;3PvnQ@A=Gy>9uSoJU;4AcH+j z=n|iYxm5B_BijSzz#S9(B{ixFh=B)oXgzx7HOi(KZ+}!j?78dIgSI;lYd*v1Tr0FI z5h7|U5sAWr&0shHP>_>Wx-eHi>H*!o>az$dG zceH+8t`R~Vh!>)TPQ#M!E4w<|p()By+Nr`XTb(!GM);&Gxg5G~cHg|ekM+$G_&&eD zRArpO)gM*aA6wwq=VED7RM^M!>$b`mPUl97N(>%xP9sn?m zMlwC07uGPbfrQGK=}2N11@KYoKqxjAqgy#$3qujeR#*M01o6$1_cga?h46{IA&JB! zC&9Me?MTHssu+^RDK;fu!KqEW+vZSLgk54cNs5dOIGc)Hk|3Mj*=@*j7r{JG#=si@ z)>xY^6-78nn{z{W%C!UJqp0g+8xWAwB8VDd(s$6*8|{OWA-Xq#!bx#%&#|S-gXv1( zPWqQ~TRVTQLBKPCu@=o)@3|~|(!`UfzU7{a)xQhfhjd6sFqfp(59T`O6*H97zRk~M z5kHr!eC_rHC%HO&9%nriZLYd7D1D`ff*qGP}U=>*MiuXp&{U5{9S z()EwK;dlp*E5hc)Gm1M2CSAo_3|`gS!cA=|i{haM17zZa8RCbz z8G>@@)G-Y3V)Il6?+Z`esDy(v=L~$FmIX)FuvQ7HeS>VCPd-rxwS}qp8Z!nGM?M(? zvWQ00?_&uUSvNf3DB7!oMhv_c4P@A#uP2~>cG+=ZL>Kr_WKBV`OS`4?WFIaQ!>0s& z@qk}10yX}}*W&MoLUK0FcS6!{)Cqp;ho$)MYa!qL4gOHE-3jt@{|oXJB0we`uge>X zAQ?Q*wQdj-N|kd2gUoK^@O}aPgD!OA&BwD z0qKOPN0pp;^Vurbi>iy;{=hoa?hrUwdK7iBI@f05cwT&!$~@d2guB$0l}@ktT1Mbb z)`mdBJTTC9u08q*Xgfm$A(Wv_7rH%CrruaZ57XwDVgjNiE9p}`c&Ej?X>VndqeY^} zsCev9{;jBJO+TgBv1N&dkEW!#S2K22y`RBWgva`@^L(Xz%=??}lvgJK6CW^pY?2O^ zT4?=eeUv@(zyDZ5&E(LzwXs;)$$nsna2Rby(ne#1i}Dwze7VAJ9c7k;HL%8lLw+z% zP3oYc06_Bdi%IZQ@_>9KKlCm$81PYp7kpIDg1TDh!PRM63{LMx6h!wG6M%^6c3oKr zhG$!paQ9I=SL`$KF(gWb0`i^>Z}x0ijM54j$FG+Tq{u1b=be>^h{3Z07D zZkppzbK+8&Qe?;qU0rEmg|Jc_TO@-G0O~K0W;c~%6;R>>!dPaL`qDfIkdde?DBKlr z*-D5vvGEp}@M94Um>_YWfE9Du^HJpJiptRTQK5A4`>WC@#GUfVtTp_i&2q}wb;}}Y zgrs~(D`&n{vb4=bw$qJn#1zD><&-shvPEwr2ky*M8T?3a);j8$UFl4y*&?l~GdEks zrd+qU9%FiMc-9W&kx1Tdl#Ca(xfk53_mWYrUC5;9IlQ?`k(@?meC`JdmG^9)1uUcd zp9m;WD)rLgmh@L{MU2Y6=Yy7G+vu4Nx&tnMJTk!Wwb7NOS0X{nHj)PoxZqSUjtaGQ zngw8=Qyf1Fq8PS7g>A5>*o09ESdR!F-k>OQf8XrH?NI=Y%D+bdzNa+WGW~)Txj-*i z?}ewn;GqBuK=i-TwJO7xZ)umzt)+;FzuW^Zwzi*L6a4gryu(WSqAE~l-jR1X9ML9z z3I5Y9uXQvne5<(K$nfX}z^I5@YK~#)%kpe!cdN>~%%WolqQz9K#tha;dA904r{Vpk z;kyoq1&DoY_8jZ0&k^jN-NMg;G+00eDDB%EU( z)51M9kp`5(BI)V^tRu+a8D)D>P@pe2lT4&|n4rJE<4E!M3$qPuID!Ay;VFUR4-m>n ze=dNosLo_GJ%@rDr)Q0sShef904_ObOy_Ps9wd562~ z5P~Y9hG8AKObg)dmgVOvgE^#Cf=S`onxnuL9b-w~H*F`MDMO6i7lqDGMjLY#@ILD` zp9NIVqjrj@97t~V&{E#_Ts%EJotz=Y5gq8jmpV>1m*EgGzLppo0Pk$YVD&iP&lfX7 zT)Jj8mjYH?H;f&DH1L4uB=4>DeeV~{^M$ijZRUJYSw4_^qVd_pzrc11Gj+`U`IsYw zDUjNlF)=k4GIwi5<%CR9lu)g>F6?r8z5g*fvW29ALvJ3i` z^_4RbcO@sJnfK!hGBO_(!{|&$3Iw(6*-FpzwM|@m>DDb9phezHDcW}|wH@?MmAj(Q zjk#jFlOevjcJl(;gXq50jFl9%3jNo@7c-BUsRoJORMsNUzF_Qb=XFSM!a3Agwx+>- zMfm{@vF#+|-Krd^cb+T5a~nu()cUqXQs7t`6ITJU$E=dNex%V z^WWQ0jvyo4o2Z%w3iDyDpkc?R%i1NRUqTlnc1#Pgo6VyB4`FZpQ03lqjc@jbO?P*9 zhe(KSy1To(yA<}OySuxQPHB-e=n@c+ZWKv{FL=&<9?$)}@B97b`UkF=S!>N_X3b{G zcU{4iINqv8=Tz_K*E>nkGhTAVmoc1gOTX&VHfOadHNU`)6-hXhJXjQxt327BZFzI4 zgnbBvdx3M*aw(W_O5gBqj{TDBUVf3%5|t+b(K0*H;!;Hvf*`@?@`-aWIK^M_63-P1 zhggz-hdp=uI^k53&$JJ;JAV(=SvP)hdxVP}|5NHxB_A+r4fSr0r$KDilfXfOu%Nj> z!7hMvDW2z<%R$JPWNX;V!wL5*#c;dh1vk%Cp|4TT<-J1{%YF3x307S~VG_3BW6jw#0i-zmjG z(ddqC)ZW&>!~edZAUPyZ=Ilv;S|MG_*XfpXv$G-mphWmKLtMpgXw*wO~7#cfh@0N+$z<>?%}We$8ceQYMnX^w7yi9kCh1hDDeIV zY4c5>6XY1X1V_;}j6m`f`vi_9*!|$@tE$#QgEtiG9QZF2Y8T$pBcs-b6Sx=$g&YyL z+RWFkj2qo<$By_oj{)Mbd929viq--v7Fnd^b)0b^kpN~AeN|7p8DLT8(NC>Ps~~k4 z98l_|fNP*1_3E9cew%gIXgLEaJe$ssgFdLj%u3g_JFEa(@19es>t&G_-W%`xyw3$9 zbPMC@i10GLPCr(B(=YbgYhAw%j3U0*1HsJ^xUW!OVOnX}rsaspA~BhAU97jQdPhw% z7DCH(eMlb^H};nhe7_75(c2d9WEeb#g)$eHQwF^A7<)N|Kk7TCFFrp?cc|msktK|| z-qM3&=tS<4mc?3Ed?*t@itluUXRy=J%uO1Q@B@5A7cKR5!Nikx4;(cuMXi)sB?_ry zE63eZ>^eQ;bLFn3gq&37T;AR*TSjSzo)UU#3o*W3+0Y9HwIXl+&WR*DBZ=Z@Qf|G; z5=J<#2Nmne0xd5vv)KqYM=8+KD9>72t-N3UN^Zo+alc)jja|7_Rb8>ZWM}tUs4AuO zRqHHPNdkVWY7~s8gx(+u`@r4=EEt{8{0T+hq~+q>-h@S;Bnv^kzhF6)lL~T4sX^QF z_c4)hoS9>jOToo5n=-fGb5R8ZWky*#PNKaF-{r6o@f=R!qy_f3ftbU5jRIIaJ*1)soog43{drW z@+1TI<{v27LDzPDZ(2@4SKv~N`y>Zx!J)d#(7{Rm63Xev3EYF1eW00DN!|$xf5ZA7 zfhoHm@k3HobMr>LE~vJMp}(fvE4j5NMK0OIQ~~{9?LPR{Wy6dgqODa2fOnY7<^-)? z8{lIxk6hWS8tg^KHo0WNp}99mnsfYJM4rbJ+X>|X>0^FX$LR<7f41j19*{ivA0)>k zGAV!tfsz3_OeWOOFf?+pmrRMKZvn_$V2DlqKS)kD5_qRlrI7jUi(D#)?p@PU^)=4q z-|abEfMWV@B-enl=RF`f``%ls$#RuY6o<~{D$VNWhIdgYp7ttmPZ91q_>3=E;C=P3Ip0Z^j7bB~_Kb>d5frv!`(*$wQ}oK^L76g!)%98TOst;H*R+t7qAW{gM~c}I``7&)D2Me8 zI0w1)BP9Msb@$y>HL(NFt5<6#{>pc|xDzi6~`(VeLZJsq&TMwxp{afaHU zG?SKf_)hzS9`~?)sUBm^k{;T(QKs@5HyCT1Ld_`0Fj@8py<=D7AbCFidj(T(J zO^n7Tv3R}nNk*0!z9IQKPsp?s<~iSttfJx%89vnW=2>{0Bl>ymY!AQ!Pz38@(Nfou zf9bZ-F=^gunEg_1VB^{4iVvouo0zi}^PPnSr(!>ekp9HRu@ zBkk@rv%J^RYU(MeYO$9oGkR;c${>j@Y)bgM}Kui-`iQg zK45$df8i;YUYZPgwqZ39$J&r3s`zUbPm|Tr3|>~v3PJk~Sdy^tE4EI%-6>hwVl6*E z&ySzd4WCN-X|!qrsvHSt9iKdP&TR=~e8=8pd$7S|N{c9W`~6#PPrh?1$0X&CgM}vh zFCV+#T%O?Mw%y}9XFOB=?sz2Gh1l~-?_6s=9f2pI2XYHM{pwALU43&81Rzn#)h?*5 z!5D}usDjWqcBujvE3FekxJRY2d?M8d=j;Bhu+IH?)&zZAA2%Ttl0+d0bGFafQYX z$SwL>mv=KPg59e zv>Z?6FdzGg<%6AfFz>CE=4KTlU7pMavM_4Dl>HbE@{`E0faR2H*5lBgaEO9kc& zLeqDh-qDvLEA9HL6o#h+gbesud_)|2Ui)j*&iULfYa_6Oxp)BKnzJSAE=hRu{e{CC z%yWOHPI>PGd6(=+H|DbrB}#vo_pxFWp-Y|o4I&}{X$=m+V1mr<+VGqtP+`KF+zW$Y zWJ*L4UdIUfLw6^`0Vl{qKJ*O5s)}7K^p64Z{87~PW${eX?%TfzmAJ|SIkeUTrblkEL zRb0CO+$)ZkTgsA1ld7^UOM870FLED3VkcH~=N?~K5$0D_Rg|~1qlb7mIP%3B|DBM% zLJ0pLLpDlmUTc#Xs6Cp zqQOX*C_gRhO#orYfB4C;J(OGNc#^+c3Q-u!RR*Ix=40s~^`&UpCadvm0e6g$3M|vH z$QPa{AKip2A9IPh6{kfA^B zwtAz>?3ksz_Lh9R4{UD5KAmB;mw)FQ`QZ$sZj;3i93z8o&PJtj#PX}VI;`5o$-K^V zI*Z>_`?H8|lOuFkHzq1%y}XD7TMoi<$b0i*XAUeEYi$pdvb?q5~q5j?_YE*HI4^f7L|o}QoQmhtaW&o&)NI@ z*V$F!XAP&uHwTxSs!z1_06dx|UWnVn+CeB)d|41ov&{*GXzq4GakLe;_^{sP;3={D zZvN?i$|!2+$0suLB9de#eQbD3<3_nnBhH^j zM2~^v@<-s65mox)I0R&nrl`A!Xb zwb4tuvoB#FL__p2&-a3T>D0KAr zReMkDc5>MSj}EG`Ud7J}s<0n_2FGZ$x9NABwD_PCnYcyMYaa9jYTh@fe3m&zQ3&V< zb9%p;(ndcoQxr<=cF38Hzq@rZ?wCzm{lT~f?eRr00caq|6eu4EYODk?iP&I=;At)B zs=%34tV4+OBGkfF;MUfqYiHCpA{okjHmn|-8PQxlOy)6jE8pW|@0T-*;{=xkX#6!y ziAof~aSJ9AvwJhyZa6H%8}T{p$feQ(+=P<~T4O_Bl+mA@kH-nN5)pp;2 zF@(^o>W8J+Dgp7&t#iS9Gh-Q81p8_L4>gjC)F>&x)Tez)D?S4J|C{t7;j-}ka|ux3 z?C}yHvRTRBT?UMwntzUrCUT)?z5HWZVW&R^;pCdk57$PjfER&=Y7CElBKp2aiu|g% zQZrGi)Gl}bba=K7hudkzPlair32!(KKSQIS4rVkt#$KUSyV&7FJx8D2b~4ay{|m*c z8km&X401=;sk%`NVo2lmgsD2>l=L+BdjIqg(BA@{fFPx3`~e-WA}on!k@ z1);LH<*0bGx2yU8!hXEy1SpCK5zu5H=Yj%g%_3-SKvdz>=sF1=qwqilF3uq&1z)Br z4?HohIHrxb$Y%9o^$2_mOdOu$c3LXv&ylH@bV*1>iG5w=`AJBXG6^EGkV;D~jU^wf zB)wxK^K>dzCORxhr%H<;3K2_XE|)Om5I>7nyz)dr2HkRb25Mr3h*1D-WS;u*b)v_3 z%h#faQHXa@A*1F+eBc5;imikgHOWdDP}0X%=9=7M50pn2MINjc;+y&2?#qcNId4}Wl z%%9uP-VWsI{~nZCocr*GfYZ3i|32;q>s9sVfya|94k=vb$#(R2er!*8w#G9At5#l4 ztcs3j8Y+be2%W1;l*KyT4QF}U<>es6Ijm$+?awpFSO1cH^W`Xx8X0S24hGu6-p6@g4aW~9?Df1nbWRmxU?tJ zMmU8T)4B+#P+%^!9>#|s9VuwFAqD%u1POgAM`s@GtjNE}ehIjajzXQ?#fy>h^}@u5 z*MLZXE`M$;iU}+E{3SZl^eJF@qEb>3=8ugaXqwG)s!)J>**Q(1{>s~}90QWmkLjS$ z(XYDlQ?li77+AV0sTivRA^9E@swJ^R_zwa@k7kTM^*A+TQxQMTqN2orB`P6L+E7Sc zN}^{PQ2v0HP)HDpgbhsGP1z5S63r839@^`24SlW8<079O++&+iA^ICGJ(d8V7^jd8 zVGIY`ZT2>-8M`KoRDO%N1vLq}L;){W-LefgJg1J`GGT%|RBU^*Kh*=^0Wcf>cX?CT zRfdQMV-o()yA9u!^WWW`$a-4w-}VE#lbdzHoVsH+D*xJgeu+AJl^+iC4{HuH~ny(oC;f{Zz*m*&%*iDJwcW|95)`1jU3 zxyWluoF zSOB7kf*@pyg|#3=loYA}!e8d`>=!x}rhmm)?K#8bi~D79vq;7)nW9L!%nF())-t9d zF+k4a!M<%9+EAiVBHBpeq}@-E;y~YG<)1vPDkiIEQEFwhe>MK1Z5ekbn`1bQ;-;<(C5w+AXjI%;B-mD$_|C21*<`mFAQN(d1n>0H@^xM9OdT3yBSo z@H|;n#M`Bs31Wtm4-jr&6A8T7w*$8dRF8}%vyv*XOsuAT@Dj|&x5xFcnM8WAEon%I z#g!*2dXLcMU13vpk>pbLO50gI92SU@h^d zl|)}<;c9ljKHeUnlc)04(v1P2rIbMAIcDgv#Pw#P*pC)?EK{}hWG&fsl>{wQ(2kH8 z0tKzw*OkoW+H#_8DHh6<{#@{Ye;{x20I@&*p=dA!fCT7sDHIPwq2+L3QUxTy2r(Z! zc4DGbmp^|sJ#8$JNn*OM#2Zz!l!aZw|NJV`wA~N5H26JgDi7;*KxrXHV3JLhmZ$M# zo+_6Mx(Jhdf+S`F4eT|r((J&mR?ky^Ysbd&y?QavY?`F&rsZI<+-^3>bE$Ff#Tzfr zq1ewScJnp*N7ur8(?{>S&6L^4C_zOpkL#Wb8bs- zIAo*X)6jGZKJ^>OGsKd-wk_1WwfnV$TDh~34?{pLd!PI1vl=hk+%ulY)3r~6j6!h+ zEZcqeILPrU9p6va4ZOAP+lb9L-2FeAgT5RKVmWq$%;N|nOieW50(^mI7&UJWOkfdO zOE1ZowdRyAnEFXP8W3G^0Zvqiwr&*I@(eWsDXwJ2zJK9+o)XP^JGN5j)(?UxDGa>O7r^#~VkL2> z1v*g~j?BzRZ6V}oZ6EG;mdUIe<%MZjUd(_LwXlw)4EcV9(!?lPF08E3MQ}=%PhROZ z1kJD{0TL`^ksuLKWg87>h4&!`I}Az9fQ$9^p7XQ`!MEiJI6_;lZMl`(r7Bn?yClhh9) zFxQ+`*;9n@)S@K6)E@^T!vBfMfIl!90D$l73qnN4(TH$ahXi9H{P!#c$$ez=WqiqK z5|eV3#I>@pKn^o*$zPHKY3&PV;^Wmt!IX-TJmJS#3O;MqB=JI&R5_fo7m@4; zR=JDK)$?^^aFAcjNi~yAR?~OEHmn`DpJ2XKq`iHmL_)E0$n{71UPbP#*Blis`q&Z93O^=P|| zNtQT*gwcmv*6qSeAOcA0r2v%)eIWJosMMmB(e7f-T2vYA`t`SwjGzsH>h5jsUvl@9 zy|_)=fs2tKVI?vZ!`3GUW%SCFJIjgsuMf1pJWt%XhiKyE`TW4&opgpE3>*6&R9-4O zLi!mqjVp<53ZXQf{Ax6}mEL4L^KeX4<{$edOhXBcB22;t*mpOqMMJ}8eYtK@He!Kx zJ~}w$of&8$aIKQmaS|vqxUst(3nn-x-68Alk~o3~iBQ|M7ef42IN=NIPg7#n78!P8S@RYwOqlL`zoul1_!t|z zwrUm?62vhxso1@H>G zD!W&njW?D0nQ`0Owc-0&q*y=?WVR9{sstziiR-|jX+nrqr3tJmjGaF$buC@jBQsMA z`GSJ=njN@6eN5S7Rm}|8^qRF-KdHJ+!~8afI@Nfw&3fW=b8&Fe3qWK$?Sm)$soDi* ztT`P(dwO{K7E7A#Y!EMbtSJ zw&E7Q-dq2xO&SCaZTk3SCeqBavmx)4s~g1joZIsq zGD%BL*2=wXNC0;yB3(ROTm3A6HdxK?n#v2SgKbs{K_du-JnhTcomKG^l0I9_K1kIR99s zV6$;HQHH%hxcV&Hyj>R}jc_N|(O5B^Cdz&H{j$z(s9eWVZ^iD1*=&u8{Kcjme~oUE zbTtS4=kuD+kq%$&UdmsZE;l+J>_4%YUG4REL*8CtJq!QP;@!(ZcF4!PK4^f<%V$8? zxH*nO9o_ZR#YXadDkzWw{l-zMwe*3#ck9ra2j%God$LWQ*)MWrZh*;6=#EyL1yN(j z`C47pc!OyqKju&mZ1cd$5JKyPejLmpJn!YGf9jDX9e90zBT3<+`914?HEzK}#J)8G zdpPzaVErVaeOkS=Gc|DYgc~klRqfjfLV=dW(bH7T-RqqKnTAXaV_Pl!3a!w){-^6O z(iO|RmnhFGh{D;BedYu17Tn`SdM$JaBH0_Z@b9V#^c(t*xz~w=iV2 zP_66ah%`MQ5KULULpCWxHV2tLcZ6>tHILG?V^+j{s&Xq-ANz(u_u~Dvb+Q9u%yt`fsKeUe<#NNSL^o*Xk*q{+<}*{q-zD<Ls)HRPPeq;;x0}F$R*#r4Q_c8=^^xzM<#=J(N3Zc& zFVnDWiq_J+Tc?g~3W1Y1KNvbzdO&B@xr&Dux2xT0@BB~HHyNHCR;M}&&A%P{a9-_j ziLB!K4#T?UY=~R&{h2r+r#03PWxB(+Vc*i1-(;!dyS|Nc4cAP0WAih(iL#H^eE)di zw$J+s~SQtBHR?g@Y{a^1` zM=?_>*(_uct758L#Stacr_O4qoG)7Fr)EMmq)sPRDK-j+O*c}DbI3z)ws5-1<<-!U zpcltjr&r?kY@~OcamA~94t#~H%7Fo1yjO zhU;kGWRNsQg?3=cPtMc4ckwS4OJIH=#8XP2dI=|$>kRaZ36X=nw$%p>M}zOV!Rjb80MNLYhz0?%NfdPcobL*|fVWt#XrLXAC%-rRUEIQZ zrOIBme|xkbp|^S|Fm3uFi`#uaWfaG3JRbpeOeTcK`1eR@N=a|f6G@bpOU-tZ{wlS0 zvmM;XkK0Q`D{>c`-H?80r6MX)#*nf3_*g2^12t*3on=sG(ge)Lxwp0AWoWb4x7xws zAAWvXuT^}}x}57E;969kL2ggr)9SYIHthb_3L9ZH#6OaW)G7MVX}dC7mhlJNnS<%4 zFh-0t_WCpVX_e%PpTRk8uM!%&1m!sg6tT+&Wh<5YS-mO zvG}sHL@~VIGSs642s(-r=(w%u;(&t-bV-b*&6_WJi3H~pEwKeH6IF0pwvr4nuIV$B z=OXBnq_hMrQ!#Uiw=>ki#M?41db-;=^s}Q|xlTPD44FRL*A}_nU^|?gCnRz^1+Fan zYWZT~ zmNCum;5EkH*g;(%nDwxJkl^wChv74`dPr07a~aiUG!C`PW>aBs6fuE85!7P@h(t>)?&XY_bxz;r2*S}Dy(cE3qKapXmeO7PHrNe?RvLW}nCwnU zlpgoe>GWC-(>~y_S`l14{6c51PauqsWItOeh)Byq(C)sT>Y0}7q&(_6n@cYIie$2i*ClI}QB}mrYaz7Y zc)nu_!;jRV4S)JBOheB`q+~Ttg21*!nF`&PE?x&gcG}~mC^IxpzOfuDRvP^PR6wi0 zoK~e-bI(#-E`V82j5(2cJ5wi=S=HaBmANp@@wbB$2a-vynb}_lBCTV+?{v4a*|Aw3 z4@8bxE5TSgiwom^p9N)^u_=|Syv8cY@Op@wg0OwO@}e}m{mP26s{N{()+meQ*F3_O zRyCTPh%~kBab=ZadRv_Yl|5iqw%6Sl!c`3_M^T53P@Y@7`j1kqwZ&s{S`H->O4UcL z{#wxv|I{JCV;la(KhSjYCp5tWKmcBV*5A{4K5(pb9ZX1*55nO(bDGP z=rhcncnQ@d9tDOXL9%@NAwCFeaSQa(P52z2ardr&GyW{6%NrKc{{-2j39ZnDPDl$j(A-Uc1dC&l}F6A*7llm z-13UGR_yBxFRSOtd$QN=1E4QYknYz;K)VCu^3N-t{)Btym`RG(!+%uw9`k&tkG&dw zCe^FX)Z=lZnA!-0#y3M~0sP=Sp>-kvCAMu_7E^JFTPK*(_Cq)|h;T5)BMZa}MZk{4 z_tY+ww2HJT6laX0FH*pnUHBTOm5cw!O9{@yf6D(~3w;=901FdwI5ct=JOEYS;NOb* ztAsxMJvtt9LJ8EL;`bxn);lj|#m>h+Ar(j{ijp9xMk|*M2?lQoVr?tMKZ8Ru9K);l z_;|cdb0KbbrdlWXIeP@I^<14w?H%p4y~$#*lwux!#P{;P7TiE+uD@!@N`ueh1zUS6 zmC~aVA~0oeRW-V4VtHxAe`6@Oz77y58-j=lW=fdd%(vv= zWM6BCj>5y4-lx4LCbACt*dYJJ*IRfqxCL#=Trt3h>obC5#i!SAJnol~U{A5AepnJj z_}(7vdlg z9G@ma;i-Wmu`Ix386M8Lp;c8L{*hEb*flOP*C>^3yp0HpSFYieuFtyYs;9MOa)E4S zqO-Wy6Rcz%Gi`b)+fZ98`Mq49?JSE%W}0M|&WvOukIu}s49<1yojkE`k2hhBHu0li z1W}1;VgGZy{&&i2+3l5PSXJ$n<@}ZMoqGj;ru_bXb^Q<>%Rx=^Toq1+@lw^n>#mFa ze?!F^h9Y|~{J%MdF+lG>frDgH&>!@-fD#8FAy9?LOqjk61pY+stC%brj;Fl$d@ta# zDdSJ|5IC7QwDM_8#$y_LJXQmVJccAcpET`A!k)e2nPG0O5G#^OX8G0OQK(b`bmjL% z(lnVZ(k>5!uWG5DtTB2!g?FpUqFZe}O#&Z1$Ex3=v_95y<>j>0=@r`ZF8e9lmmb4= zVBWsk=~`X*cM_$T3|H+YM{;F!zm#gRv7}^TaOxyE*Ij}t7443gENb}fNia>bw%CaJ8F!lld!&Q~8zMs`4w zT{t$zVLTND!RWG+5L__4{a_s;PS!>hvLQ+ZpiT!$mxPCMx|&l$J#rxS>x_@RC9&JH zQjs6|YR^=~sal0Km3@YygNNB)9X8w`nEM7LmNmNrIKoQaE&1>H6cK@%gJ3c$5vF!J z(=bM7?v9)*nfbnaRP~QOb^W?iw#t|1IZJzXL3aTw%SMXhmyOuu6~=tK~xHOYLcK~?mBCh z*y9f6KdIF>N0?JJ7;YJyQ-zzs&++3iWlVFMTC|un=Ow?ymF`^OFBFYcy(|bKGi*-^ z#>}HLi)7N$F^MH&n>14R2WGNGvHNbCE!mK zJ#Ol%=R*X^Q>DinSj!7bIHM$f@0u3yj+rBJ2i+5%iD)Bv6Q7=R-Zs9`$qJqJaekcwlWc6iUoA^(=DU35OTE#+ zr8vWP$79 z#AaiY0%>tVVn4#;Nf(*0qD^dPC=*o?MmLjXrED!yR7DW~@ieqmk2d{Kq*kR@+?BI+n(;whj41~njSam~ zuviId^h-%j=zrUP#}Q87eI8HW{oE^hf`9_9;WCBv@!u_Fsk1BhBwPxC6^I$AB>RwX zIdT*V-xm7OO6M7L@Qn>rz^FOUY|KpR+_=KmRP7Kou*LYPH>tX zK}9`Xsb@T$94wY^wtmTLH#ow>Yy7e4)_rp7t?)gb!>@L?==+JGZC3p*AAec5Y~|Km z)7_J#m4l}`cS7}ayIUdoF1L}4`Br`KKqTE|?!h%BLv$SznEQH|4h_s_D;;9?YDZNM zSBf!7Byx=a%^YLTgVKV!6)D+5c5kaFV!7}tA?gEqNOBnOMMm+f=U`$g7!QiAg^@JF zwXQA@HdK<-e1`OoPo-{_wwj$_Y3##4WzfOAncK=3%v~hUV!xPMllak~yTpDuAzPmF zd{rC)SCLBX-#@PlW_q*bG|n~5#w#K4@tLOH{twIVYS`d5cNLtwpBv_%5*g0r z_WIM@hot30D5DiWo%2nrI?H%W{KUvHmOp@Ld&*J;T=Bt-gGD?SXQN`DEHSWupqO-ShGX4BY?Ss?-)~lzAoVVz$YaxVj)@%5P|0_up{>5_s4@q*5e98!+l@2Bm z_xhhCsZ2B-Pkk(Wwv$Gq>0n2mnGP(*5BZS58@$u%^@-@mB&j<8CrMa7q$?o?k8RmHACjbk`hGN5zOkg64jKeNv6w#i+{?3rQYPQFAz}iBGHa1B zWdL$z;p8!%E#$Aopu-S=1TnnU5MRdV9w8qtDgS6}&f^-HquqIe@U%R3wGG2hQNjnS z(FE*=Um9$UFGzGD8#7hH&ft$-MW<6Dh~H*12i+Ur&DSA*Tj_f9wxP72@IFlEXS><= znBZ=<3T6mimaMV3NrK-O;IB9mlJXb&$kIiH% zy@AZXCU%H#iAfa+Z6%@_LGoA+sgLB-m(+K)NEV6kL5dN%aFE=H!MDsv7L|;ZF^{CZ zewXk=${Dml9OkR-_JQ7R}QrD<9{W#<2wxiaWTviNRi zc{co~@M)CF{w{ph;>O2|Il2J^M7sqc1bP*@VGMc7^jX`g7B z`g((quqhONhYA{k7@3drKW~rb;{FukRt8K{CjG%o$Ypebi6!!>^ybfNh^VOk9H0a} zpsv;*s0#@E*BizE-VFD}rhW)M**m#*wYAhoo} zZxSkW^E|$T*?qKMul8ERW_`G6<1$U>_Uu9MU*KAUN#L4$I_MlPjmI+Yuz#MgQ2tOr zKJ@p55g`kHyU^ba+}tTn$LUsQQm5M6?U_|~ksWY=$Nt!t!D$>v(nN2!r2E6`@0Xw4 z#Wz23`(4z`N zf^KvsV=xd%CT%IjNkEq?LaAkgscgE$zm*Pprj~XtOCPqjDl&-*J)RI~;}zEN0Jt zrswi$sH#x^da5g^el3k9djQG%?K8A#cL-90zhV`?JlsEg4et7S&{+NawgwI(r;|pNjp?eh@fxY9KDt{-CwX-$$MkpzJY?bEsWlN+lP%gLh79Tp%e9>rw`?P{57V| zcprPykpE+w;miqy26*-wa3WB^+R(Nvme2|y)iy4|2}iCNMEPZw)Hxoc@NJB(oYnQ8 zme23b4kf_xZ~wgC4BZ6k|7Xl9T(O!-2H!LMl?5Rb;!!ghk2^Jt^3UNT-AcyN;CXvO zWO=PVgz}PZca5rV^yb5PG2ksVm5&vR2C_)%H8T%Jiz43XRe4lFQdJy<99f#EX6uC7 zqTXHMA1*W*A&#->Yq2l2!EvbN-aKJ7XtrNei;X%v*6NnOzVWQWp>Ae8U+H=|TH@D(-JtDDD?G z;>Kvd(kv@nLf-Yg@9Qu%#*2?`?I*h=tV`WCzr4H;SBMGjIU~59Sye0{u6ADT8@&s> z;Je9o#w--^|6X?!7Z{J43|7!E`67`$`&C&u5m1Cg6-l(4RoMMV|2L{i|HV1C`W3f5k&}huZT+~@iFW{L4r6TUm+I~aPK4N;snehg3i5A zv1NyBsr}Hd-Q#IA(~yw3XizLRtxr^#taE^&z7y`iby+hWdY&pPCzX|f!2&swz6p6R z7P+km|D=WW{EQWKgi_fk+12Fi72eUz8X*+4;!%gxAu_8HiV3JcMx{^j3tzQJB;kPl zFszsq92fb*2pZCr7q$FyG*mxk;0vXRZEd00h-9vgAVnyOC5oFcR*9x>BKy5awJ#A_CyM6G`b%Pb;`xAbwcAal>AA4V06N(%7xOMf>16-e zCzlp7y1K)h4l{;is>5utf?Up|`e>^EXd1XjXn)*w5c+oy5|>5d%Kq=_+CD+?fAHN1 zx>Wj)nVvX*j4W^QyT|#><;+QAiTr2(@!i0K(HNot4nXn0sQ*4%T;;*?y9e3y&xInT z=Gmy4#YUS?Rf^gCQ1&E;`LS62IY<2lac8hhr3$BLFXBD>X7v2p`;c&C2B`(50+Zl) z5ZfHd;#uEFDngHlj84tk`*>bl2euBAx6>d-WBgZw0Q)wSRJGEdbr*F@?FdyEkj}d9 zv8UW^{5QmHXUqBMPW(2LM!)At&iuOk9OuFQ=@@IDjOs1=2Ey+#@z|i<#?WHDwg4C2 zFPShX0F~ePB@aLqGrY$z_x?xl67mfBD`Q7~eMtr5di}ck_9Wki}$~R|HYoh04rZ?J0pnXAz`>HEn3PU=`uyR*GUa3CjHnd-Si4 zb*xh3TZxA=vT?vXG+YP_YDRXNiM~1Zh0Mb_zS|#UD1!{Sk(tumDLizn_e>F%T!JV{ z(&L1j=$A?mT7bGH#tmttxgbd$BC%9QlFiDBRtAFt_)N^PeKBSXJQ`8ik9G{o|4cIe zzhcDsAx2P1#d2g^H@c(#w>JwJebIjm7hh{IH*HI0{eD7iqWzfnU*0UtDuGJ>^r?Tx zi2Zn=N&hOUu$F_4zv*-l0_hs#NVCh=F~!gPj;#9L-#1pO=d0yYosfRm-Y$x?!jZJbTk0p;mxU{t*`$fts*s3Cewlsg5< z2U9CLLJ%BCFYct3ochLceJ8nJ4y6)ES-Yw$!yScR9J}c zRK9KCsAl{tQ&azm;^9CHHr0~}#1`80+6g1P)Ai7FG^zpSt+W7#h87ju?+8p(Tyrhu zq(KX|QaXSmz`6-^lrKF?^+hFNdc-X?Jc$On6Om!aJat7u2;ow4MAW2b#`|zd{I*Ol zVg!MF_eiubN-Nxt2&#^EAv*>h@@nCA_%vj@pVsO^aB1!Ed|bc2je!>xtI?S(9u2*Cs0D)c=72!@$64-V zv;GXA@1;ba?I*H{@5zwL{-Y--O0NqToKU)PnqFo9x$^wF4fltMAc z$b7guzJXW*7BOn%rXA%TSP(skAcn6V5-EV#j=u(lb^DV$%)Xz??~aNjJeWh@b?*1q zKvVfB{L&2$5ts=st8FxUL)VwUyxJ|GJD{(4HSp0tig9^zVKGttHSqnKLrrQf#rZLDzF-qJac6rNPhX@hWmP_7^8|JZbhHm=*!bUP+45 zn9T#t?<^{T+K%NEA^M^)m=-=~unV`&+gDMAH7r$DiDnLvK$VE8H4a|1)11p!)koPL z)D(ne9n?2npiD}<-Oyu`YGOF3X&mZVI&A8dlc;T+m9jmoo##N!XnKP)UR_R3PwLc2 zWElOrW>Tq|qnWxp+rIK>iN(3|(oes->S8SVP_!dY`0(`+&ux4k+$a08Ub4A^<2Q3+ z23$QWS=XmcjTBY3z59No1nq-LWM};cBrmzUXxv8{1}I|nImd_PvfbYDi1RH-^b_Q` zS3mjmzD9xLwq~kF6!nA3q!i=1{GSkq_7K*g0RG=${U}NPeTszqF8cQzKrfcnv-UGH zXe3D3@2_LZzah@8WF(DCH|S@C%eI_9yz9`iEY9ZpbbQeb1zYDb#XK4Msb_vXTYZIa zISjcZiF=T0ELX8v8+f31CWSk&v*_6)-e4BaK& zDX~Es>F!2q=uQD?W)P7Qh7LixyFoz)l~R#ZB&EA0RhSp|-e(`rKIi@3_q)FD|M}y0 z&9m0M?zNuhUiV0N-mS$BLll8F^@hi?R1*K$Fzp3w)5=7{Yd1LP6qeNZ_04X@g9p=9XCy%k_^izrdHqRu zO*kI}xuf87!x4n&KqW*9D@KfS5u5L{y7G||GN4xPBujETHY4Fao7p3BA4&qz6s_-E zpY=oglUb+ieAU*NG@(`wddj>cleTi9(rHXu%cUKRYT9|q$XJe74?ZVIL%9B0UMg;^ zHw!e=x#x> z78{H`wU!Dc1SB=_hr@jvgJRucX1c5sR^=?ya`pLpEv`n@IM%l0I2qN`X|@B}S2O{; zcgtJ}6QwlHmz3MRnzt-whiA49)B3{nSJ$M zGJ|Ufd7q1;|82IPa*wKDF~#t`R^6v{X76stbQbQEH#6NU3wpt8^s-cB{n|YE{z`Z2 zY8svXkNozyYLiX+xaJD3AyJAsmQBb%-L7Q|3nIJ-9=LO@#UZeP4#+OpVHZnAXTA+)x>7pAc)`?E)^fH=?UF-AbX@d>;X@Y5mRh*Kvj z-6lwT(>~~(^IqYUepEK)()fg@CAMWkS3d5U28wUF=-#261Rbq=S;LGY8##o^2SrRL zI;vPmDvF6o1Q&bf$sAka>;R}2TdXKPkhN%1lvvBf78vpi9+AGdqj)?l8C1zKHd|$w z;&9G{pYj9?yyvHs^P{e`oYu*S9v4+~G|Q zud6KIWw_3?v4r*JX|&f(wg-IOwv3&+-Ox>f#)zicT>RV%N@ucq-_f{|O$fsLmj@AC zfLegy--+g535|E?_3NjT$^HAMzpCLhU>*M1jN}I1=jeSV^WWN~{z)`<#v{=C2z<#; z3#JWL6RxM_#y738EO_SDrJK!M$xTn`wifLUX*H2nJ3Ww4S(qqq7171-QE zQ{VShS0GChc}Hi#@&_HRVth3-M%#n}YCMeB7m$u);DxkZJ7U7p)u@Xl6X|CN* zW6t11gph>r0vP`6&TX0VZzv3WszX39hR~sgQ;%F}zGpU6l!y^A$&nw`3m}&DyGy*0 z6B0bXIwifC_aG{WuxkZ3TE}BBC(ilQj}ivg*SoXl<;m9~hASpb2>;6;&Q ziJy?tYN}!@QLz@rCCP~&`=gSWa|RYt6QL9sRhZ?)5LXl%6@xw24f=D!mM-)9ehe{xlrl zv@6>cYPOFrHO?=GiZUS34kZz^OO0np_2Uh)UwzBOW%rEgm)R~RfafMX`Tw`+`5%yl ztN(w2EJ9Z*tyM-juJx{;A}l+N@o@uc%qpGzjWxgf@F}-rNARV&A?xHiF!tWs zzdK%#jffCkmKMwMQb#f25*oU96k9)x!IBBzZ1Z-AOt@v;!ZWpB8&bbN1l}3-?2GHE z*;mRk`>GrQ-1o5=hpT~m8aS-FpFfAWO+5w8x%w7dBrih}e>~sd@qIo2g7g{^-WdCE z8H=7u47S*t0)Vqf{;=+%lJG)!mm6dT629=+RiW*|V91o}n|yH;(*5FkMVfoxf8!a5 zAf_D|GUkgi+7B8mvdzTokyQJ|IAwvB22O%3W@3W21Z z85w=63@D7HnK$T*+b0%Ya9N9u`3Peh?;uLV5@w`v;LmaTl5z!(MOshSs-pN9jXP&t zWUdGEkzReukATXw-k4eV!i7@^P#<=2s)h1urggp8sO5tL-O8sanXN(N*~B)L>Cf2K z*v*uKzCk!`XnG$}JvE!}t1w7vG8qjdItrvt!ljoWDUEfZ(YFuxdA^peVY|Jh9Qusg zzNCueFFWL$fDpPVH4%R;%J>&sfZhlYlSJBP!`}l!+^0}Fh10;QBnG)$pUQ3bb~TV> z{ryuh>Tf)O{5JuCQ3_DQ;2hFM+(B^jxg1WRLV5wfR2)20AsA-f?fEJnjs}=@bu>N% zlU<|EnbehzGrj6_RcuO`lEx9QK0@u)^V%Fd-J)V3V1!Onmzl9WzdGS%g-?K%>P(uwfH)|n!&3(o$u@BNVSg6%kJ+p z_-hMc2c+Ux-!E>>3=>*!$ZlyKZRq+#JKv0lV?<((-Nn1p0gL1k=VXc`5@KZtB~_;a zJR+x8EI^Q6FT)-)IAk%!-lB6}j^VT_nUCga>adA7<10yzVvbfVjwkN>yqHLPIanY| z+ZHpQFn!v=oFc2dTKJecPSrk@N;=Rs*>GEwCBw}4jYB5h^gxM$b)@)Owqsc(du-3I z_4a3HRQe}@KKQrJ=$};Knz`f1W+35j3C$)Sf`{(RRlC1B5XFEVsu^nLP|ubO!=Z~4 z+>|X=OyxGJXm-&p(I|e_r6#b+U#?Y85yjnFtzKaII+*SSiHm8Y{lvW!6{7XtJg1SW zJdGxnPu0#_+xXO?2Maxv-zB7RIs_COBA?P&@e&_c<|VK&P|yM@R^LIn>=r^tt>>z8 z6>hyZGi%uzkK!|z8P4Hx8`6>M`AfY0$C$cp8Uwu}fl{t|oytQQ8`{~oOgRrBlIgC! zU+FGw(6cWRLZYY!V0#nEl@~G|f=U6J_s|nH0M2rr)31!rm;v zQLE*uq5Y==29o+@n~U(Fn2izsSuQc7NX$-^n|h7F%=2KNU>%iZfrt?WsflHM3FfI& zv3x{fe8Lc2!Pc}C4IZLgQDpNKCNe)$Jzhj!vVm3-d8Q|pqgUu2+oxbwaykgW51rOh zNw=JtV`DKfVnX(T#Hxv=ZHmvHoPiCjvTutImE9Y!y>wK#C!si#8;G&t_)LUu$Vt5p z{LAVivYUhL{_|kb|A&M9fA^jLSca#XJ`U+92P98w0UUv+itF6Q1EH&oR08&iI+dN2Ie^m!XJe>!C==) zCNxf6U|16C88nL)0S#)00ph@-3lWq#SgJxK%u^OYtR#b6x3L&thop*G+=dMz24?V}m5;Iu*H;o7l=5Zsd*Ftg`1*?{Vq>dbflG zNCRL1+P{OY=I6J@0CRj6`CQJQf)?Dr&Q|0a7kW%Tl-dJHG|W1-dk5gb?U(xVY$1Uz z$npo#dO;-)rVpq+x^>C^{vl z;G_K92I{?g0z!ibO&eaIjxjLEQ2FLamK7Ws&H%;+*#J^3N!*eH>nAeZ+T>E?RSr>S zk!xUxbMv~Q8-tStSBIZUh#m>YY_90E*_f9G;X+KyZ#%O=HvD*a1_2fi&8}uZO+g}_ zuyB8;T-r?W&v~~z7`|8-UFv>mro_=P{BgDo8>;1Tesg=O*s=Q=;bRJ&j`o8y75sMX zk!h>I$dCT$`&RUs?aoB>`X3@MetF{TkZv%CI{quo0iFNSnh?G(8KU_xRp_L{% zku3fHSZ=Bsg{}2xVu1|)Fo4WBw}r@7Hm?l-OsrFC)IvQ#1sZ%J_A{}F_xRF+yg4Rs zIu!k`0g-Z(+MC2`8UEeM>QlvhOvagKFqslvqUoZ zp=D64UguwJy!3o@7wdo^@|*@`9kk1WbvZrR?Zd^7^&wwBaxChgiTx%KpxbTp*3;ku zN?Ut2&O(Fhi-(L+q>h^@n&L->>CueQe;DB`$8wfcEXOg=jK}@48cO(sm>r<4;bQ!W z$eD9-QrnN>WTYaF3t_SfP3&rl`u66kN{0wGv8Rr9M*#xg(u_@)-YB3n!;FB8QJ>;b z+f;@#4W?txb3`Y*BL2&vbOxIJ_LH*g;Ahi%$*M!*PO2t}W}mF|ab`B|+R;=^ zS}EadZ4_u3WnePK!nn!hgCySQl_QdLuIjov8axuH|Gm)~MeS{_iXO!(_u4Mw7R!bjV_a4DbaC}6u@@k+se#e7BC4EMh!nwOjl%_eWlmPAR)@WNuizAi(<2Dymx#7&wOr zIchW*FXg$Qnb?tlLaC zxs-cK2^>q`*39ALpbGL|TM$9GU9-*Co<{bZyc9=uhn!mvGkigl<6D-Ud~GI{aHPFN zfeseR7sK5T{e#9g$M`MczoUpR=IHd`p4dkL`5Kp7?esnvjZ=?+bKM+F55QtXR&p71 zF$M0Ec~eXQpr8Ok()lZ_MdH@vavQZUwm>Vzb3T3)s8Q&?FfT= zqFt*M)zv9$>YM0x)*JM*)ssu!$6(71etDL??Jpc}kDhsYg z2f6rj?3pU6ul8@##qP`CZ3z@AMoaqR?5X;$i~`PdnT>*5wgw_}3)j0;P8VJP1oK>< zTpZg>R$y8ZKXWV~jm?p%_;1^$W8$YkrzVP)lEt z?-ujRrwtsv=9&bwK4B<0@vOGiJACzKpihr-;-tLygRfQr(lU=P&Do z%FgO7)No3MQ7F*zSh=~ij!cQ?1ld=DAAQt65FBtVUGp$8t_?!@?Na4Yeif<31?vMk zfj+(_V$*S(Y_nB!`F1}WIbukMQxwQJK_%}|30pvcElwvh(%JONR~-ZvyFf|)06LJy zQ6?%T^!3c$8&{4`Ev5g?v|;RbDvV3P4YB+Z#rnTntcp2;HQG z98}3GIJe0Y9HhiYhp~q}KjwXp7Y7mTUiiX%Pn#*;BeZg90|Do71Xk|m{t1`tIv>#E z`x)QiKDSvfd$nV$x}2Z|2rE}C$dz2^U9;*zw`=6I@0@}bL&(0f$u1XGgcgHQa7aO3ylxS9p0lS&lRFic%GtOZ19;&qRD5ELChZ-AE2e& z#xi?M(s9HD58uFnSX$waD6L6tKAF_RUi2J$UW9(b+sZ)R9iRV-tJ?+)ObN4#^-r?H zzb3ztSxzH-@6v@8<0N%4;BEX@{w>#~vKQ6^mOk)kZoLkL)zOrVh&84LYKCBs5r-V)P?sR21uo(ZcR=1XS^SienB`^qq! z$@qIETdPN)NgzsMWEHX@J=bilLpPPIOe6a&_okBR0nbS4SclR}0agC zuS@$}T5HXd6e7ZT?-|y8E)UFSpMTo6IoksVf4iz6DP2l&CEc*Bp>kXth`=&uyG71z zSE|kA*2W!FZSoqM_}nA+yl$d3MYQ@^Ncz4>Wn_=l6-{efPf7d!)%bABo!yTpvv-Mf zWc-^eMWp5AJZJP^2!261opg@J;m$2K>ubKna?5VCr#M<%b^!$hK@wIWN!?-nh3E3O z&+wifX*7aq)9#{gkcU0}$@&7EuTOc-R)ZQl&C8X}8#8D_E)WGOH+3XMP!Sl-xy?Qy zGqkbRdQ3sSzQS%d^`}of|FmK}|Q*;m&1J#G17CGGDaSCP(f(zF~HnmV=>#;mf(x zcVytt&QTBX0nTd`N%AxtcJ?^zw)b7HjuXQPY0*b0o`fQiPno4JAMLowm2t}gip5(g z^^5CNw(2^4l_$NM#Gl*f4V7LLwT#E5?>hRhu};a&wyhEZiiN~nt@JE!IjRh{o#;34 z$5o`YuFGWHw!k=AXY0(`>$B{7&TU%IGo8(LSnDvI!{7BonO(O%fU1EmbA8ltb9R1h(> zqF+KlpoSHhg;dq05EiQdAsIWF0*lx#%g(vr6<&$xNUy~94PPwMZ`(`&!QZRD@=OzhE-3XT5iKe&{Jjj;%n&x{}bq;c_c8?e+tCKYv|azSW@? z+Y6NAlh|mf?8yWn(#n-3qznD`C2c-WPDSLPiy%3f<%M7AYqsk{GXY;)VnS9Wrtr!* zK1D`~^1(iodt*HDs_IP_DfwSkiok)n;1_tDgt})?(^1#H`=ccE-m3v3y(xkqprJ`Z zGnCMzNk=2S7p3)G$LIF4wKtllx6u?0NJQN`O zcPkVQz~KNI4q)H_4i4bq0O7yqp=bb(2GD2#g9dPD0FMR;|GgB70pJ(_jR7zi0EYqa z7=ZBKm!LQRjsws*0D}W?H~^0W2>*Q^iU;6$0F4JQcmRh7@OXgm-VP#7EvheP3UC;|=z;89RK3XVsi@hA))g~OxpcoYGT0thH5 z0R<qSTq!ifnsq` zEFOv_K(PQE3x#9ha4Z^*#lW#RI2I4b65v<>jfJAIa5NT;#$wP|92$#9V+m+1fWbmB zSU3iY#$YiREDnRkW3U7a7QkVlI4m58MdPp-92SSe;&E644h!J1P&^in$D;9A3?7Ta zWAS(_0gnX;SSSGtCt%S8EQWx^5wLgymO#J)01*lh;Q$d05HSD|2N3Z9kpK_@C=m)J z!l6Vol!$>6aZn;2N+dvu0GtSg6X9?o8cxK(i8weB4<{1fL;y{MqKR-c5sfBd&_o=X zh({9%Xd-|iLNP=*hKR-xF&H8aL&Rf<1Pl?t5urFD97jash!`9Zha=)~L;{Wo;E7N? z5soLK@k9)sh{F@{cp?E$1PDYZfe0rM(F7ufK*SM7`$SyyfAPpX+3v=^vwTlPp_i zFXzRxr~1H^k1i1RrZ@GA{Lcrgt&`ZAgw?9&k9$4Zvrb-Ljhm`jto)1X>C&B5E2beX zEuTCXc=%W6LH^d!_phy>E~3?j-_PUEhi@ouytWAn$$GwQFi^B6a@iXk!|Z3p@)}uh z9wcHC5YObyulu#wH^)%C9uysl9=Z9+%5S~e_Qn0{&&E&5NS(6++qg_2FH5G)=5=c6 zT29%0f7++$}h5=68<~unSQx9$TF|wi@9Z5Rd9eKq87U%uG8?l zz$N#Cl;}$4m%s0;9+_5X+T_?y;OE;kv?!j}d8aBJ#iqT_s_|{@aI@6wXp?GL|4=g}WBjpb{A}HC(ISLZVbL~(&R@ep&bM>Lh9%2i?OB&p zz?yF^pIh~Hj!8apd4<^ZrF5{<_pm#nJxlbdS%h*`Tor%lPRhkbN=|?9rC63!kle4k zIz( zAH=S!4@_WL1?4QnU2Mf!7m&(uj+*i3XjE1#A_T``uNSJ6B7 zE{2*y#|!zu4s&9EL$Sf7%2Z@Q0T0e(XG9=I3+KH?D#V2MA;CsAdIhx&)5TTcMN|T= z_c=4K%ZKAG#03h72Ao3+^-oo(k8LekIbO>+J+oLdnx}gLt(`@r-@7MC#i(yz<6cZ{ zgVtf|EAZ3W{rf>KD+*a-L!M3cdpI$2a>vhKTlSvdSi~Ar1y{;p=AA`dDIRpfe&}2* z>|ebte_t`Lc6|36$48M9_Z*nhTv@r=4lQ%l9O>BHULOY;iHhA}+wBeLVp}}WV>6k# zl8zhCGtanbSoMfBJ$^(^)g|d>y{#@}Oj+e1;pIxA8O@W*th}2sj<+VMZ|7P}g>ik$ zhd$H2l^>)RGCazUcjQ(Oi`Ac2^C=j)lW(4=WYGGFwrJs1&7aw8y^=7w;-3Jc(kddy zv?Vitz1&%uWRQ3#Oi)hZyvlhY90No?)fXTSgRAi{GsP%;hcLvzAF2y}jH36g;4ZIw zsBMplQLie4bhxU^6ntzlgpwfYC=qw>eT+By^ppzZrl-waKh?u%8$Ojx7tl7JG5p9= z1)ae9J&>;=b}m4(N?$)sG2qmc!278x&6Dz;-L!CX&d@$D>YY}Fx6G}U z2*(?(-2Cr|Hb?d1iL(WtG%Z)Vo9jgcIG9nZ%3sdgZi*Fg%(=Qfl->C5M`E(Ozqj(f zSRB;x=52}L@85%O=o9K>?>PCc1}kY)KwE_^URm9-=r7|wYYp>~ax%2H*k-G5AKf`Y zOI%y5xhM4V7CkG6xTr2-O*>%TdzZt^8$>0FuE_*(CoA$M?E5{e@y%1ji`S1ceLECw zQ_wqkuca*u{F1$lu;%$rt5gM#q(1$~4d0KqmQr7YuI;A=04qS$zr57O-%nf;t=HUr za!teb!+>%S=66EsQKWl)-@64jF6r=n)xqq=V&7fO%TnjQt&`gagCTTDjyDu_@bvRp zJger)dCp1lwLL^F^o#bQRoWcIRi)|1t=fX+6K91?L5Wa$-vWLA&*~oPq&@~X)+(f{ z5^<*}(%q-%W($?_ubvMG$h(reQeW~h>T|iBa+nR?qdR0}9}=9~OL+KP+ItF~*~z)W zUH%?fmeMVD&d4v}VvlW(e=j}G>{yQGL)-E^emZK<+0a~b;{1{oFY74o)S`z9rv|+% zL`&Luf6nBcpXzw`Iw7F))YKf|Zo07eJXgNf*x*t`3pB%+>GLp;FL(KNy2NbjPyBYq z)e^#Gfg{Gv?MeB_?Lo`a(ntGnE0Nrf&~dGbj#}>4U)#_7mKN<^I~7%O6W{G(fBY;u zu|BiSAt+=Md&_*O5YM^X@_lIIng>~FC%hbf?wNog>6_jXE`GAB>epg6>kUgF?(Dr} zfvz+YiuW%I|8!ClwTHSEy!}UqyLNWi-hEiW-YPsFmnY$tRkrflvf&*gHTe~NV7X)$ z#Jl&6{d$hn_wuV>XUksk&xvRDj@JDI5zWn(sU$5Q))m6QpW+;xT;9AN(JCjf+-u8? z{a4UZK$~A2;waFD{5d)1MiLzs3h2{ak%Xc>EuvU;02aMVISejJ{0j zyhVZ!``O~UYZiXD6$HY+phCJGKbX887UbU3Cu>ycFW%yiS+xf>|GJh;To@ZHv0?l~5aM za>zez#!|Mxe28ME7!@N`2(r`+OSBJrjtrFd3fhc_UA-&V|I6am9(&HbnnXTtY?DBdOU?*TMeL#$ z@7miq##)hw=x3OUFDB%{*=3bj9CS z1N#T^JvQT}H}hHHa3bz6av3-Y=U zXu`UJN7kXJ(vE6)U{WAO`rfzDrxd;(cT(ir9T&&qZf(b1c>73q@)%Rb#fqjSa9W-c zJ3Ts}p|+foM&WVlT9RAw(68MI!&wP>s5DnKDGC3~XOT8>PZ9^Gc>&{jQA+s%3A`FwdC@(txo&w? zm|RC>M8++XMj-D6RUTq6JWDD6c`Fqrp&)fR43I5ot0vt)v*R8+p?T&d{mb@8ScNM@{LtG@JHsbu#$@<)B~+b^m=@hKt~ zvEmgm)uaCN@ydP|iINH2$_kn76xOa?S+Y4fU|f!eNy@6c2i3V*#)SjYzuYX>d-}@$ z17}zQjw32gn#Iy0Zd6=R+r4BVS@&0n>1cd0G`W{0^}An+TX(2)kk&=FJ))I)WdOwVXOKn73TCRhf@=aY+vU%e zU&oPuPe7z%BriyqJEK?8QRyjb7)6jwqUbG4;m{S`Wl)MMt=nwQ;DaD zLX#|gFU^^09r>k$Fwc)99C5MPPUQ+-W$p8252(e1u#M4X3C(eA2NRNWaB6?I<{-D` zIe0{6eRJVf^KYGq?r+Uxy40Z8AS%1&*z0DR7L~oornLhYynNOm`1d6 zJZ^sL*38S(%Dqa>^tcU7!|@m>_>FD(``g8UsZBzc8o;!HS6c;%g}~$4Hs+AlAcX?u z$u`lZ5-wgVX+lLlhcWG<2Yjaj1rl zn};2ikC8H!9VW`1CdVD$mOGG-<;b%;43)cnixqh*cX*_eGbqKs5KgO;nh-lbN7l}DYzzWUzH+FsrI zr@Rf~Dy8QBuR#sk)VgHQ>Z0~;ujurr6QZww_fiV>h^@rR;p*Fu0*(`F)b9D57ler< zLc&)3hcT5ts6d4Gq=cw_Td(2Ft(od8ftTrsGhbr#Jk9>l8^V#ohY&Nd2>a*DBu4EuiTqMKgn^cnj_fg3+}Zas5=gb?VD; zfzbLfABZBgG}B4!U}11b+d#2|oex{D$Vbn?AAZYBD*de(4ZUto9krg%$xyv?5W0p^ zrMmO=VE5`^%TYj4VO-<7O@RYD#8%BiY9HVHz@N0Tt(!=abK-d^1yk5t}T_%n<0 zj+~f|daH!KTh&N?7N-3E^^x~ovS%D~@UftD4mstqI3k)OxsM8H8N;ACtXinpWC53R zDqGBW-mCF^0;qHg$m5+TQJKgKr80$)r=U3;JjWXJs53CMYcTpY$;nk1{jMy1#}r)# zhVp)ra;4qaoa^nZ#j%)IW9#`7=|vnc2q?~KY6guPUYnXO8k>=v_P-b#vzkgS8jD?< zCR1*gXPJ?ao7tu<-V2@Cp`AGlok1ZjC&p`h})f0S3FB*_1Aj>y@TXp^pQGZ^07O(}__5fOy_4)g!^RO1Wd9#re)}AhT zcZtS(hd+ZbULn3{=4LyPd>8;e2aKmb$o~b%JwZ25xK7UJ?WX6407~`eqe|(ig?W}? zyqrsvX)gYYlAS$wN2GP5b%^KGGGXM#{YxS1fY~_>Ngtmw-=Yc~r7GX2uVlT~w`926 z_?GUP8gh7_Mh{5*U!T;7s5dFq^5_=6OtD~{{rRu3%B|g<_cE~U#%EvsKJB(Y=&aXhy%PW6=1K3SCWR?RQ~Du;)z7(wi{qX3r(f zmL=`*HMVmG7tE6E`daujCW?-797ca~&aexttEAv{j*MYn5*u~id@;2DrJFPq9s5*^ zpTHyJC0-*rwA_r@m^uYaW|uYZFVLQYYdvX9&zFBrFE~rVRf=f`0dly&)_pq4J=u}7 zf^^!o^}XM)@Tu6KIL&6*n(T9c&6A;dYc(v4+T)y><9zPcI$$chKXVH&CH-m7Zapq# zovUIUTRT)SJmgM%TG{iL6N2XdT)YRKC5I0j+>%-;wF70mp!TPqClB8={>Mfl`_|H7-A_Z!`WiP!8qku*ZZk_@+j;QT+QXsG zgN&4_bca#{Uw^9|nk9;E`9GE-e;fT|PR3_SOm*f+Z_aoSq(2KBORpTiI0t!>R3oX2 z?Wkoosho+kRMH;cJ3HN8ShK7^i5oN7*#Oli;q7Y%}Rwy0}le>y(;>5ioK*rJjvz8rXQey##KfKl&3 ze)&H5<$t4SL}lDWrU+v-Ew);`()XjgeagIfdbVh4n3jCGXbLZRwLUeSN}S%{UsCsC zpvD}Pm0V+AFm%DbpsZKQQqE2J<{#Pt)kqEp$ZBl+qSlnH(|{@>#XD1m*8mIowe=I7 zb`C;QW981@YuUd*^(K`%_6XLe z;YR%X9M^AGf2*#UO%m}m)#q)>_Azfx!%c&*802r7E`}3&Bsm!Q`)FVdv_(#C_u;nG z2f*-Pv-u^6id*t3Z7L_4j_=>Onn?0o^pgR>L{24TN%`_@HJ=q8n1dxOP{vEXjPw_kEQ zA0i(^|KY}dltw36v1HxhxfB6H?3a7l3XReFt)1bXfBih_$b`Di`g;RL!i}owi}J1l zCO>gVMBm_K9Lt^(ZnYW7U~4c$NBY}Glv%IlNQ7tK^|NO^l`D5mu65}&R{dr(^f-!` z?|(G6w^=_F%^>^Z@_1cBgpQI&cJ|ASgCseU_k z&1CqRV|&V7DV4uX@q4o2J7t7$^M>lrzpn>#hihK=2~I_j_62^ePGzs4tc@%oYj{P@ zGB?$q3JXrCj)ecFx_PknXH@>zr;vE*KA10MXK@>KqO8bCdt^OXfzbp&!F?4;O&XK< zFV}1nu-uvPxGS%cEEHp9Nb5UZ2k$0|SKLJ6SRXid*s02^km6 zV9D$KY?;PSOcGb_1$wU#O*`1J(mNV+abFwiEuCS(;2K z@pgaj%pa}l2uqH2XCPtNT?vQ81|0aGZi+}^J z35_oTj~dHgtk^fH=0xsdyK>A=uu&U%tVY5@PAuJ;Ttv~z&*S#c=wO<443Za3U zjV@QE-(IxUuIJbEeJN9W`ILguZbtjT-*3~u0z|ySsf7K%Pe0lWrHJSw)}R8O`pqc! zYnYeK?u&nwAGZUA#!H1V4@PE1y*P@3Kvn`WpXbTEU{tXs0O@axy<~}Y{7^j6SzW+y zJ0XH5iRPW=cb3OcH6*3vN3zd%HGU@ShcTIy{*@+EN6}3Q{9PS)v6nAUl^+&KMw&l`Vw{7c){~4C62<1&b(VQosG0$dpx9*FEwsc1$c@aF~)&-*`BtSXOvm zJXg*4tw3*lgV0uC!j&B!b6L#b4})nc_jcd6Lo=m*owvKa@{t}fhKyq~LUw)s{4N%A zE~3tx!c$6~Ya)>T#Cu!k(Xdex=-+VSc)YbBVLQ!#Zx%zBKMkqANYh%;;I^gRt$W}7 zZn`YpE=zL&2SE70n{i2AP;~t$#%maUJw*}k*kaF1>=qEP{ep27x95Qo@$VmW8a}YY z+!Amu+H=nEMc7^mzB6%-pN}e5oHO`$GaL7Q+J{(k zP`N&Qn{@mwSNewFIDIfU4`f&wc<4Ip^Y)g?YSg%|mHlY+7L)3uzJ%KmttqHEhwRm4 zTS)j&!pkf5a+z)L-|+B9!QE$NO+J`z>bJAhT4NPA5E;?bkqAelPwI_97@+4DA~qpI zg4$|EzfGO~n*bTpN<6s)!I0qxDDt)ii9}l49J|G(G3#Ep6CrZD*Dzj7`AOIp^FFGN z-lpC~f!;%@FK}J92x(<3 zc1Ty{sA{naY03)(f&(+prlaqjp#@c}Q5g1ze!><{;)Iw4==94NxLW^U&HW0qrJdcG zGw{)Opk9w;Vp}1AUp%YeJVZIc7?;o6=!QpeaQu|N3CwQLZ%f-G)-d&cnHl$O=W&Tu+3%(6 zVRdpHVZYPnAl+3ld8!z5^I$3`!T`vc6NK$7l->Dy${@C zSi70UIQB>HYAB9)jf~;-^?Z!yk!}Q}*fW_XHpYN{DpwLtYqHL+y2GhMZ-tfhm*HuZ zxa8>)%Zo4P#cw~=I1h?s1T*S~YSj|+VhK61<=8Ln} zwt1#eiN%l)U&h`x9=S`dsNvxECk@#O?s@K3O!#E^jCj?P%7-Lu2fv#>zGIrztSf&& zvOImFQt(BQl=~a9LW&)MDX-g~@E;-}A*Z8wyE?O<*fio}pfF _06Y>bkBVAhD*y zDxu5V<|9wA^)_v!04>QCS=|@knA^|muOE;On0n(Y`Re*Aj_*ZTp%~btXisdFA8!c6 zy(B=yc@>}*t0iZVtQnKqt zR2#3P;A}Ywy68f# z95A1GR3^s7^i-bnoyflYJXN2>9k)jidC8SHOLLj}vQJ1%N8s&j{n^Tkmx(4-6q1#|IUw4mEARxw52;d&OKA8k#*_tZX}$oh}NdV8As(X9sL zBl>;K`p^*r!myF#h#~C=a<~Asrb&V%%@PRDa?8&~a0w#GpI~AY^kS{`VjYoSJ3TN0 z2?pXL^^*Yuz4!EFk(zmY89scjyh;#Cz>HiArKx2+_1*;c&`4#(U5^xKi4o%&)oBMA z0YzRWqo(Gg-mKdG(WCy3+D2Q+e!Ok|XbRt~(Xci0h~?2h-mlx`T1qnQ?NhNEIO(%kNLQN4#rJCm)FIj|2Dg<$REIK+V|0dr)(Ycp9 zS=X#_|FMP2D;4YolQ@|5b>!AWFCp&hoWk|CFoQTHBQ+IQs8rEC}b$>gz5 zUtze)A(lf~jRi@Hrc-l}!JlhGtFm0_w0M*-Q;_{k!w zfiWsmog?8cBYiQ&`=)OCsCtc@DE%t3^!X#)@?Ocy+3D%;0sBYw@`)1*~lG7GCC_oP}r2klcZtj>#7X>m1p8IIQ`_=sp73Emhl<9 zTZ5M!C!URV|o+<-vG3nVn8h3j0Q={Tyh?Jw{6wAlK{{>& z0kN6gz0y72-2G7LdMnL08PW;w5NFRQdHWM+{o-4W8lOWbAn~ufC?}`HbT4r9f?AId zmWDr|oL!C;`cWTP5F=(^n*s(5oV@xu$fQ(Wjs#mySixY*SahC7FkWDrI*#(VSmj+AY>VVW|@v=(H$`_Sj~giNez9> zUS8?{er85I#ze+Wo3)*rCF_&kwp$A3#HYv?VN}guyMmJa%=msxS5GqUwg zO!XNk)Gy~#GA}+$GNj5rH5XmboVfGU&?JS5(}^m-ktTNNL!2nSad>FDZ=rI9RB-*K zkU-x-q_m*m%ADXz^^Z4&e-gBRJb2S*kxZEpuw!AS8}DvmIHDr+W%B+ns~)`+r5|?2 zWLES;w_y&tFb*i?z2fX>#>dEnl{gH~jtW-svcBr+6onfy-F3l77fW1yr=?$$U%e_S z^|)r;DvmOv^w8oUk!1mzD|pj#Np1xny)2bHGa7;#3R`)>l=P0xs1+I6P_1$FTEkXX zP$ug!(bT%pKE(P>vUN0U{x1_8WQm$WCuf#GluN4oxTF5+@TK%<7^|hmiENAZ3$yzMjrhUP- zGscdXKG0ub$9iAkr1(S2r05?LsqcSR$u^=3{LGwN#UO33pJ17e8D#V)J%h^?YzvHR7c+0>+1swjMX;*1x_m)kn)%7Qh zhrP!}35F=gpO{wZ5xND#C{70hn@fWcb7+1xzX z`Uh3u{ikTq+KDY@q1s58`zt2!@RRVu+w=Nap@VT~4ZHWjT|NWK%Z`KM^UY}E&H6hc z6^*K9PdHnqx48XQCS+awN)r6}InawYpbIO$f>JV$bCWmO;#xk7S=%wpSUq*O7YN!S zGFWaZz1m{^c`&R`34!a~U0-#IO>q9Vdc4Od-oIT)m&SNskN=-bGZ}V3C4RI6qoImQ zF+g!9+j6mfZjpESvw#1E!ed|OjgKkzNe`cxa%}~@|H9=rSL#Ye@9x?z|C#Xx<;z&c z@Di%$C94spbCK`2)MEkCsZC+AS%V+Z5? zfoRwMjm>BGx5J$`hof93{vN1Mf3B^t9mjue2OpL0o0vvSi2G)kPP=;ASd+P1oAEuH zwLTf0aeg&J88P5mPYY?1LPtvGAs^rP~XS9|nf8 zW3Rl2Btc?dQ-8#hXeXoq=YJMUGB97i3&&r~N}{h(TrVC-5Lw?X7ssRk<&?g>_%kE)7}hwL4zaGri-b8X_r*j7R6%;4(}a*%71bx zE=>=&f3BTZcTSuN9V;#+E;(PYrksBAt!v$XAoaQ9wdyH3=@&)6p-A(cjLz*cdau48 zUMV(KUnz6XGTcDrZq?puRUK<9=u3vKnau;=4q@hnP}Qn~Z$mjk+-A-)qm*lohreT< zf1nqs1f!Gr?Ty}SHTCPr=6<`Z{uXVss=T?a5_ch|Z?tNJ0v_qjxn)R(H z)|vRfZT~yYhxt?*`FJ?{M47HUb~&|rLxr8$y!Z773qu74eqnR-WR@{0{J$-cAaZ~a zkc0q|KqZj%-!e6S{rY#{ZI;vq8`b&tNU09&lG2K^{Ah(MONw*>vUij z8qDj3dn~d(o5sp0>zMb*k}Su#u;?I;)04;2s_^UtGZSgG*QlG4*>i2e8_Blq)clfu zg4^%nStM+xx?6WgF+#c}dOerpZ*t_GZA@{A9n6$*y``z)Z#E% zqW{`o&8RcNed4drov)jNQ8X`Ce6Jpfy;okm9lsoK`C0BSU6wQ^SPDu}tkGN-raZjQ zSf@dBJqxq5iwUFqa5Hcq)GM}wk=EDXr$fLt^*uMH4ku$GvCwlUflMp(VuY(SG~F%| z{K8vDC;{^g&rTlWmqA9(Wfu-J_AHO#G(t1R}MYm$I@k>O+#onsXw8jK`kFu7ot zXpUcm%H33zqritu6y#!i7&QeoO30Yr&qpC*e6ko424rics_|Lgt{6S#HIS}$A4bWt z&$JiL#I|?!K}NZvepiY}!0$6zff3{aaR4uhiCh=_~%$3aXy z`zQM6U9^)J66-yfim z=a}QU|E|8A2}YCfO?3daQMt!BWdmaZYu(Ea=1K^lKA=|)906(6jEVN7q~5P%qNcx7 z1+z{zTh(l4^crMB6?xMP*M{mh6sYL(yXBKv3mBr`JNyG!YV!ooBq-1&JFL*cp7TS_ zx3lVP+Kd-1((<1WYrn`*B>p`#25R?6EAPh)JS^U};<;z{um)CerADR0{A_R%%~754 z^si9*_RXN}&3887<1f*VX}a%yv|OMZLdM@7zp3cbf6SJuF3DXfjgZ zR79lK2Lt^+avvY+tekO+(U#j$(AfkA>!2_^&}+XB@^V#XntNpHT_IqI7uGriI{yQk63zO();cRRqh|+xSVyW;|USyMX7-3TT_} z>nyf_E1F#Y#*r^hOl{ejK%Ub{K-fb*y(DFC4CJHt`~G|JIX;SaLR8%129I{f!|$HW z2uVoyHiP%4cl>jr;?^H5CHPS15=TnTo@eD|Ekqq5W*8?0dDX?EihtQ1DwcVv;K`3W zq@P9kNE5}eurn+D zBueiq4enWMskGTUB}?L3)vQX*uyannA4qn_ggOdx=_FcBS9Zq14z>^R#~*i@HGAy! zAKo^iSv<@UYSHDU@_RyRS>P$(l&tmMaI#>FCD`=MH65LB_0Z0`ju2B4h@8=0WtW>w zkP-$5kO8lD`N`M-wXhlcYpMvh?n*xGPHp9^SB?RcHy=0E15d31n3scG2}NMh!HY86TATSe*z zXI=qZW7(_NerMAprc)XHlh&*X$agNGUDsEww3-`?-IsrM;o;>}mana`_N_m{$)NfOCiQ-Aar)s2X zjVm8cJLyC`vbB4epj!Czgw%yK^x_1m6FG0xNq%EMYcfa7ik4G!D?BIyJd=C3bc2z} zW02VPY_sI=Qa_nCeLV5TqY@R5*s#@`TQmPQ`yX#Js6af*`mQ1<0s{_?%_5|Lb+=zh z;l#kZ8%*DpT}nuQ6}w>If*ox(A^%)|@rO}-GAj)u-Tw?x7j#?X4*)xN7qPg6l9WVJ zD9V0RbQe&-1fMKdTE~2K3mg)0S^G8-%gcT~p;~e}TKW=3PAL#^lmC2}t5)=!KIum6 zo!#h_Yro2nu6VN>BvXENbGn(Wxm)nQz<>hWi zf5lX>4Kng>QPDW+GIJ-jm#z8x{$vW*!#vG_0q;d55(A=P?0s#AurLI^CMJ^md{E3K z&-*=_Ehc}H1QbdPo|Q@Mbh8WPY75*9^q;_nvrF79A zOEBLfVsk-a5b$}7EE(EwGukre8(T)=2v$9K|cLbBCP5EBD#IIjD z@3|z3n%{w}AO~WmvH_R=wC!DTm1KyT#ca z!j5;y_X6V4$R>5@4iI#g?6+uu?TrMpcr}7vZKf?FGJ@3Qj+zY_Yk;btNQILH$EPzr z^X~?r(h4C8N*#eRG8Uebf4g;b*PRmn7Y}-(+aDk0t^iSq&YrH7P&>LX?O; zE9-vCf)S;aZ~I=2QzdGXpIi%{<&aY#AS2|QES@n!Z*3t2wu9WS(cZK2Fc``C=pz`D z25sB4!kS)t%AG%7AEf-7i;sljWMQ3XQUS-UNYXwSAo5Da9_+gATjCSWs^+F3xBF4# z{!p?2K*~sZT}G)@7p@7lXQ_mYh?;Qajm$cZ@kfwobw%ocfJo3@zQ( z!-Og~D4dg$vSDX5qR2j#iDyu1V;giGI1X56B_*qnc0^$eg`_nTWx(I4)%n$2yO1z4 zFh&gG|4}IOlMxdIX);2mlJ69F>kPHPjBrM~z$2z^9`8nXu3ykzkW`T5ZH**2k_u$9 zQl&%o?hK(FSodb5)gFFi6sx)v)v&N9$?^d$^zX1Y*X4&&BQ-hQD5z! z1J5Aim;%9w0fZ4M6WG&x`8#r|#KZ{FoVQC-a2_LGG(=rImF->U_9!kK!(uq^{D3Bd z^0}ruL`2nIhK7_&kY1Bo(lL>%>k2`}8bMn9X^e#Gwhjm_#N|f%M4HY!HzvA>~&zAL5aq*yq>UIv_5IM%`Nn#@%#3s87jgKajcIt^*Ors!5YLNC%XjK#D$k`17W zM_5U{rS@)#DtdCQGA`=y|D%jF)YDqwqExh5C zxp^RR=ZJ`97@$sllK1M#>L9hRp#sUQPNLP>+goDH!)%ai3b7hw>vJNMlyhjX(Os!< z)h^VNg)LLvFv$LcDY>&iV?0dDa+1RcUIx3qC+Z?!_;}hor#%7yiyea3Pk9%_Gj2)4 zCm=DQsCb^P{YLkYhQ}~;?z7YDGoxV%cr^5v;QQ}MJ`%#~Q!mbY@qQiwtI!A^W{#dR z>cU7i*|k(mMPDU$R-x^9)vfy-|2Wb^288aWe@0Qm{iRMTF=Q=9G*@tH!6PWMw)&LU zq`k03TLS-tA>3xE{VU)3**}5J#@QvIJ^8N>y4AKd;X)1nTu1#HURzIhT)hLE>Ljpu z+a}>mJ3s8tWVX-S!r+xSrf9w}wndg|^|xEf>%P|AgBIw`(hqMlWk znjLZXAQSE=X~jzSC^6mw?{4fE*h@bcGma@MxW%-6pg^HsoX8xG6)8U^o2ArG@rX0n3WywHmDT1mwP;G9N@pc}%H`M(tgGJph-WCUb@W+45)fI{|v0Ll~f5PBhI=mKN-@P7d1 z>Dt);018AS@&6xCqNw@)Pe9RqsrJ7WhW7tYK=It2DAw(2T7Mn*?Pz_nt9kQHI2jGU zUU$p(yZGbZJl+4XuA z%~{}wA4e%;=7j@nke??hApTGEo{xWi-IY}ZT^Y?PdPfpz`QsJ;Dyhv=v@!U)s}APr zJJv`NWwkbFufH|v<-f6mzW(#;jE=R;-T*Q({V|*Hy6epYB9Rzs5Az4Bz#yTm+3G6~ zPNC-?q8K0k$M_SAZb3phCtTjUbwWtWA28n^+Z3U<3xq^~eVGAG(pFasU)pBTF{8%?aJ~IKO;yt}!3k~q z$$vt^wC^*F#D;P&Eg$67s%2qpP>-o0>zJQ?&Fwbe}Geitb{ls#(0ru@J{4km&Qn7x(-1*nwOr@oZwIrO~m{5;fqR>5>VmyyHC z+$ucO%Hy$=Y%ld#~G<}kw|8LjLPKJ5CSh)gi0oT>BO^kvoy*;kFCF`r<`2-K?{foW5tHrA_nF*g% zN=QQ8sDjR~9>I@?YNtT-fx=L#zotgSZea;5oH0DO#X*wb^?}UlTg>CoCw3He#m8XK z>*b&s<2|_Msry}luNDYj|IAp`6E*sm(A|K%$`TmfkfDuMM%lK76htIFr?z7{417w; zL@UDq@=q27sQ%+28So9WVvLOAoZ5nQ3WFc%G@p{It(%I=-K1kLBWVgE?WV-_fAjhp zLOlP!2z$>!xB_--dm6?VjNWC8(MbeRB4UiejNWUaNAw_Cs?kf3F-r8_gXlpd(K}J2 zCt5@gLWm^1d7e|A^PY3Q_xr#9?t5Kpt^3|vgZir3^M^=2z#0Jib?9q{C>HUl zNAyA^9s}Zsg(X>Sf{;Cq&LKQMNa|+AMeSOq%PcuiJ`y(SJz`TVK4mad3Y|wCQ?JS> zw&svPqEh`ZWx#n%qJ${OkshM4Ocl(`#!qY5H9424-#7JgMLFZ_d#>6~O=ey$Xj1rE zDTPJaYaq)^FX*GH9Wq}0$}fSI$yaHYNkM!tY(t|IS8my*X8IQ366AWh<~+IR7x=jo zPV(eqE_0Yqh6&aO8xSccJ)=oo@@uSsZj*a2p5OleVhLR+wyc}+CB zzJNvPD9JHz2BxP5tO|r<8iZ?SP2I87yQSTZ8detNd3*KPYqtT$+lj7zjus#&rr6Pn zv~kgjY7A;&RdEMGSxt(i5O(3JmO%9gNnP(~_^J_iG(wrDP?gqY6pIFPh6)ck#G{$y ze4{KwB;&n18m3I14yM97o*|30`|Md&VSlFx|Gd7$7ISaYPsT2_p)WBKuFnvgvSfZw zt4sr$e9W{@c;+% zo5;yTR#?fkfX!n zK6UlQK)cN?7#h^;5-x$}i}tS%jb!hURG6#T@^M#%RTbe2Y?!jY2A3#K&vmdaH5tfx z+A&@KNJyNhr4R|*Aj*hJSAlc4*WP}=^nQ;{rO(!VK!i&nXs)a3!Lhmxcz&qlCsSIJ z{;m>1cy|1zd0q43Td2o!^tUY91`U>cHwVG@A9muqj+^fhZ#uSO{LeRG zMXj};5zRw)ZzkGJ{9c#;8ao_LoBW@p1Q;L=0_X#R0CxXG2CDx=2EpG&)l2Uq5d8nh zG*-GyVSgb~O9OtkxLk|c zK>qWPM@zx;O+oa`P2|_rZTS?>KG6)#cz`=1`1xA}rX@bVr-wdH7)o1u6%fZozc*yD zsx*-L>Q?hYW7~Q#J;fZq(vRttM!k{NTH2Xi=NVduiIfMHd!a`6&hyR{%~wAxYd&}{ zVf$rTtml2+>b18w>XKK2GcH}UmOK$s-E`wHuJI%uJ9K5!!>%Zk-y9gdI{$Na_~lBl z*tKDe&pI1C-aSH8PrWJblY$|ii7dNctiTlVZRxhO*NsWcoG|n)BZO{YHwGu8|CbEp zz3dqTIc*^n+U_;fY|@_C9zm&3O$5|95=W=NXIW3dk*TfHCX$1EutbJFJyc_XJJ*CW zTWMeYl5DJ7+YKN>2;<5>g!ezBI;IeuI5bns)>GF2;Mf;@E=SKLfTYb2=_kL91|V+f z59mo=5l5qcK?bvaB{F4`p2ILUS~x#GrC%ra67`9mL`3|9uUW3-@eSn%KOWGFB*oyI zi=U7Qn(1X^>J7{o{32JUjAfWQEUU@&?0LzPIU_;Ry#e)?Y9)f#iBFwUIz$UY(|@{3 zRlv2zavK4o<>fhN>E*ipz9Jr#KXN%;N(i?G*IzRYuP@h3S#OKIa*QZ(@QhBC&eaW*7#C zh)j6Z_V*WNwo;3%))x>Mjdj~8l4Wc}s3Y6y5PiL1cNYXl0jZW2*@n=m&(L^7t&JS; zQlVxun2jo!v_dBU6r|_?s1MY$U_7!7!t`UtUW~R+#$n)%pa|JZ$v{c7S3+)lT;{Qa zNI>%<3e0`)3^|!mXY=qM8)t4^zY@q)B&RY$YpeI! z9&G(VwhwZ%$)l3Li9Uv-@{K(5ox%3jYT;?$4MCWG|C?fy}MNaNV$P6vh^4m6`@feu$ zZ@az=bON-Uuo|UNPhGa(0W+)*F>dqHP}ot4_cuO>F?IiIn3uaQlUJUJ@vDh6s?c$Q zzzob*{YJS)rST0Y82U!;>+zWum4*)e%-W1j|L(kBe2(_VFvk(u6Ov+jzfaV|0^ItB zES!#bUv;i2a^@Bvi;NUi>Dix+YM*S!sLK^Mt|Z zds{bLzU97;;t7Do$-9=0Ju764pCy;_G@GtvVv{*dq8WoT$-F$A-ADpq8~|jZaj&qb zu}?&!Qt!W<58fpTBx(gz-EPJ-ym?irzFp+*= zQP#pZs0+v>D}3D_I{~IoeJJ4dx~sbsNZ^|GB(Am*`xGwT{u(0mqN81p%evF(+`y;^U#8M%gM=P~05G6u!@e?TT*5bPAS;M5@2Z zx;n~IB;ZT$+J;w+R|Lr(hHI^a}67obK_(?G7VH=+*o#jB1DxKUKX=2WXvtwh_ z{xuZ03ZQVrt(X&W{j=sP96v_|3*Avsk}<+{Q7;C?QA01H(TPz~T<#1^98dxXf(!;Q zea-Myr-lN+?lt|aMjI|gDu(V?`vJjFLuaH)TbVP`U$mBj<7&4ri8;8UZ&+plP$r@< zPfElwW?(dX1aINbN>s9nU~t4lX#hgR>h}fzc18Wj*4HxXYmU#A5>!$wZTkU+Ggp+4 zySOdR?yi-THtH~jmT=XlodX;x_ zw6Won(RTLQWr#Vd4bA8n@l6320~r)mbbUKi)F9Ikkb1lc_MNLO8h>HAn-S3CUf z2Eu+;AQ2^|Osio2p61W?_uAJm9sMC67b#7{AMQ-E<~i50e7Z4OoRcF^Mx8Wsa+G$h z%;Lk_+XG@$IV8wklq|J7dMfUN0X6^C=Qd-M17Tm0B`% z6^}Ne51zM1xy|pXG%DlXOzxtIrxkZ@T{~G}VmW;-0Ud(DYDBGYwr z44uoT?=9gK+;Lt=OE?|YIHaRy*?{l7Zy7tVfk30sfgciCMC$hu0-`vY>{h%LSC*^9x$yvOa_^UK*Ti$E#}63h{k>5K>b1M5kr_^+qA5?s`f|t_tbt#%V|VD ztL^oPwD%A}Ms)&-_s3p(CNh*wR38BL{hsj&E(GMRhDh5+Gh1N4mX4+gj5g#8QxNWE z14dznq-EEQ8+VlmH5assS>3t-lL$n+en<_*MU96`)ob7=ISlBWrhBx;n+pyzzd93PrIK685nTaH&txR}D zd0`raTg_Sp@`CdiaP^7xzCM)sX8h2#P#PEPP0exEstVJjc`F^8qnKpf`XZL_P31<7 zj~2teAs_{tV6Su|n}eh0m|g2Oz+FNJaE*H%NNw1u|Q+Qc;47q1E{G>qn4W;v1X6dM7U)>KK_8Xfgn_A|yYHx%%$ zp3-zs8m9JLJ5gjs6y5=dqnCiyR_K}=7*>A}Qf01Lp@^R9_Xv8xH6qNoRuMGKpamCi1*t$7=OY%voeegslL-B|XPZ+6PQxN(gqEu0)|gToWvfC*u; zGj{aF)M1hXEj!9mZxGb}KL9#`-D0i#(MI*Jso8aQ)t`n;IM0OgWY|W21cApr>Qx!Z zr@c~C$R*kH%3%5jYB*Zbf=z2kQ?FW;&t0RJ>xZU%=?sI!Eud}Vx(>3xlzQ>y@gfJRtdF3D9OEgrj0V{D*aY#lZEK>JLMVaIyT18KV)h*QJTAeEE|0eE5!) z>tfA~?l0*j%5l#n4Rr6F0W|QUD=B9T@r@heh7x_c$}L97`p&|<$9*PIOJQd-met}< z+7Rp{2sK&-_V(PbI+(U4!Wsu_k4h{b#p)Z=I+jrXozE!A9oJvifvwDVM#gc<;OczZ zA_Sg;(TgUawPWF%{E683Zj4!~J)o$xq}7V=p5bI7e4f`v4bR{ie5CIP_!i$bKlUMW zWmxIRlR}Jn#Ou+-qxr>y03|*jdI|(O9@3%)aF3BtxmU>DbF)uLCUKEG*x3K!BJ>b@ z^a6W{4p%^fwzKhxabuA>4r1R7BHRyhtZ66je zJSuA}{;B92_k>FHsXFKQJ#DtGEA3lJF`BJu1c!Qy>&pF1)GNKCeZRL4?KIHjW}A%r z4=faNs6TXmz5m?z37;w(DQqsUeloYUP0#=Gaa-o=InBoXm+38ucZ(=qzH%4gFT&BgcMBdeTy1?9EtrsqEM?6$p9V&n@qo$0s$oCbmL%3w*N4vz3m2(i-P z1FE4oOChq_DV`JARqQwy8~iE2!4!Pg;pPaTPzj#WV3d3ax)ibKceBwfgoiyuU?xPG z!G|i(=b4e&_fQ6zjSz8m1|aZXSVsBRchCJ#EV~u8&-P!jOyX?tzeV*7b6RCyH2;X| zd3}XR#o8f;p8pZmQwctpfuUaJ;u9&~26(L!Gx^O#vmSm=CZ|hNq-#p_rCRNHBhAy9`p;m|i+rt|%UPMbj&0T8!FNv{dFV`bYFvC?Ic$3HoQ~%mUv=wS zXDZ5;s0+2lhbO;IJwi(^WnCy=SYf^-n|)<5XD}*}S;~Gy*>&<ar5-hOQv@-JU@M*ie;71eBCX`zb?-ZB1HD@V$noJz%mncH&*SfZ4(ItOZ~fr9>CaIPA=&+YG4MkEt(by-A)YYsDO;9bBSDHz z?7aTFeQd^Wp{z7n?42)JR0r&qz`IRSZ6d7V)47iktY=Gkh&d{4nYC(ZVm|jn`9;8_ z4m&D4mD6#>nIh2p1BXt1FDmTP+zV|F`{oBZ9^Sg5pW@P-E(Vgaso3p^EJ4?i8$dNq z%}u3ool;?B6#n~0#-|2Ro}%U3nzj{z_8fC}Z^Nst@9#%*@Ra$I3#?E)yuSor`l$Dc zNBLtP2~|>L+7AF;WqNmmi#ggT^1-@7JJ0&-SDh7i@rla;lUa&cbz7DeDCeRlhvA($ zwbM$F>^=RFhZflbLj{(~(>7<@cLhl;1@ou1&z9A8&zCY9ni3TKn1U z?_G_75JgMbICO=Fp@^N>FJ1pC6#H_?E5skeut-`4YVzX7GaN*-sRsjfTA(;*duVnX{U6 z07zv3ouS#`H|T%+ti26^&`YmB%hevAPAuSdDss zF^GBGj}n2@stIc1f}Qx(4C1jpKzFdk>I&ki{ZTOukdMJS6aS%f1rvgCRb(5`4>&6A zz4j@B%bS}=(K@Neq?tkm2R)pjs5~JNOM3rCaNCGq1mqo^U*3s(V_&A?qyLhjO47l!% z%Cj*Gh&-5}dxV!paBm^n@}n;F}Z0KdD*5mQ$l;XzP6PBST}9MP(3D+i>8mn=cG z*?|gDPnHH{k4m+X%tnmzt{TdGoMYO4_W^t6(fIH9er?wKfa7WnDf<)+48}M`du<7q zz^JF&BEsITA0gFvHO@ubNA5Ff%b;#`jF?idxugG;?AkLf@P$e7)N#{e=VX#|CBo_D zBmqC2Dm(7GVFzt0P9i;K~RmMWionqSxMKDn1Q{)o4_OAcF@ zyH|U3Og}MxBJYm8QWwR=b70Sg6a4+kN!EUL(^&k5d+JWY%`nN}$7by}f9?+M{VInz zq>37_V;J8xQj^k~x`eZFEH9&ES)b2H@#_N2#*gg^>+@N5EapV#qh0@A7sfmS& z_i)t(2wRD3vbn`F`yh>9o*MLcc>c&kOIe6`c6^cQS6gx#I~^bxFbcd#Ct&myu;&!_ z4WG7Tq;niz4858IJ)HVYj!Wvqws}#t&9%pArYpSm&5p^~PbXOM0#RxJ@}Fq&30AlJ-Y)g*X*#Re1a*U7}~yr{pfINtXo z{y@ZItFL-aiI3{$>(CFWc8u&}#TCizJWTZ{-r9t27TB!vbfvTKUD*X&0~|6WdV@PB{LFy;z>m%y9j{VpR^vaiQ|z-^i7ZQ1k0 zmpI3f`_}}U2SUvwlD|5aKrLzq{gMvTz!bHz8qr<%Ks}jej>53K+b!6iB${8X;@6a~ zW1I4i!XS`;Pw+tiapAut(e&>mLGf=S5qyE5`|Vf#e+4}Jf9v7WnB8S&;iBTdB%!rMl#OpxI5IB8Rd)Fz-EM(%cfo;JovI6 zdKJ8-P^6Gz4&1gbBY=L}-FEX!#28vJl28rNL&eShgg8`Z^s^AF;#+ACuQgRKnR)is zpr*?6tr|%xnj{pE$UKmGMKAk?U4JnDzmY_JFh~tT0TLu$+wQ9lZq(X;)wny~7?f~x zALHvXrj{vgIoT4pvns+QIBEHdZEwBd+t2LFXGcgCj!4w!Pm*?Gb02a(Ur614(!xX) zLP?Ej%zpDIVK6wX_s=$u-xj4Ijs5eDw?TQJAz2%b^GVt4d-S)wyFKZd4{#56Yf6Sy zUH{x&RvB89ox1-c+&_%tmE@6?i@JD77NFlt00BV$zjZ+Q_|`_gI_Poad<5kD{)gbg zG@n7`*QZWh;7(gAycD=*;`1}SY|=8Z!Oz?h_~Q3U%stAas4TYx`9d`^t0)^XuyO2E zFIbMO(Z&Kg5>EfeH3CAC2rWzkw2z9A(6`<-IeOT~|9qaPA*tVerIFnhbI&Bbj}zHO z>Kai=XO=O|?tMTdnDaUzdKFmG%GY+q=5erdh7hQZRsqzCbMy{0-1_W8ki2loNTkd> z=;z3@9-9Z6)}}-O)^bJLWI$aaqbc^yTiHr4Zki#VudyIKz!8V)bAH08?_w%`A4LG! zETYdASTo###M^7dR(mbJ)Tmh0fH>nfsb{TX+Q@cv;uXnP%{{^e(SXD*SrAgp)A6sN z4riuLdb!_8qj963Ik>bq9T04*t>4qWzwO=2Lj~{ai+0cNW~bDv?`~HS7&Cb7OF2=~ zD>HrL&_VqE3k>c`yEy5!e+YbN=-T)BtH^}%q7-8V8~MAcRSoOjQvJ-s--dPShCY@5 z(e$XFlHt!X{n*P_#A60|D~OPzyc%-_IaVzlFBtnU7AonfquFR8HhfW>h|w-NtIGl< z@dM_=z(#)Wb|xoumF8zMWzDi$){4OyU~P#}Y_oY~vUAvfa_i82-a1 zL2|)f(m3BSzkXAX6CF-!_msc?YukcE%T86JGxu=xjd%UB<6XR-OQ1B}8~Ejru1_l7 z)*h}z%?2j_jY8%6@S_5~i=?eyn=GipJgGJgExQ^$F)Ne)+@NXdexdA{-J{<>-v=f_ zId<;mrJL-!*)g>VB754R)SeT$2^?&iSXeyXj(gRJtz`=fqi%wSaBy^DT9y(9>~MZ^ zi6$OBR9NqD-S7+cOMfio8JPUhLF^CmU>3siZDS-K_t`Gxi_A_m?>U11;eAb50RA1n zdYC_n0mVRtyDYR$h?0J&+x*zMJ7Voa#LZ=%AI$PNi-Uc^+hI`lyCC=-_Z9C|BtEmc zH18|>)wma-c!fgA=%5>w=8i)s&+;e@jFm-?@|7=Wpuz^3_r>b<7>cR34NowK_g*&(Bve* z+`dXku0b%R4)rcVQkFZvQgl-ssxrkfMw$czf_$O;ml*rwW6&@Y0+vyhk2#*HFAKfL37(h}Wn1K})pn{6m-nu?Y^9!-R{p-sNHKq0`YFF(=n=h)}0`#C#1Z`DL zX|RPP=qfGyp8s368Y(Yf@D-hoYlw2c3bzBA1SAv!{Z6Zn03<|Vt~04J&G zaE&+2AHxbI^fa0VFEsU{6lzriW7h<~MwjcJQy=k4wF%2n)SK!OlIw=KJf+Bok&H{jN0^lgVjpSX$C$4mDuQR zC6!LD=E97iM5+VBZwj0qm4%@Jg?NOvd5m2r84Vm5Z1Tog0jf2#Z_IwO5ANr7i?l>IW>aPKRxuJ2d3&*pa~ z8g*J9jHb5=Bt{6Q^W@GY^w92qZT+Ra)%Z~BVqG+ydl$Fc`tmEs;>l6aI#sc?-X(!k zbW*=j{1r+4KNmk}M|tmjvr~)r`mQ#~L4EOAL)Dfj9tV8TI0CJ7=S9=yl`*h;u4Ec4W8Yt>ko-+~hsbh!`EIQ#2HE4~@ zTGV&TW14n}Khc=8izna;9f`xq(6xCi#N;8BQwm5T0*>tnr$(pTkt!KYa$Tnc=ui~- z&&to|atN89z$Z&EC$VK?y5Eallw^SKqaz`kcS^yKZRZrDQ9$(qkgH zNY;%)y?d@s>0 z5U;CYMFdmISMPnk1hjpIy`1o%StIDrH6kO3MIX{$)oG0bQ^+8XH4N|xIMh6T_1CvH zJ?S{~6%Z&avE&lE9m9d1f{Ca%7;1`Ti+c@-|=uaGdGwgoH{^A;ZbnUgZ`B_K77*M58_L zB}#|TTJBk#ELx-Wj9g8`TzKXD$ejw&|Ld6o&0B%D6mn@k%)S4d0p5JE19eDF)z`b0L| zXL}U!A`F@|fmG)J9PLQ~>grc=M^!gea!y+7d<$82t0_c6st5fsNBOM78-k2>&tlJw zP*31i!u>`A&t_1FqMwm2@h+SbB!?@POh~FI!*M%0+>LMjo66dUgizzF#iq=`IzdnX~Z8d;IDsNk8&ChYp z{b+0&uJjBToPSj+EBJWVBkHN>!sl_>|18I0|M*LdK49tJ<>Fr#U39;Ts`EO6=>Au^ z`2Sj%R%{&jhg<}oKZWt6GZIE4Dp6`^wfeG&&;m-7oklr}!pq$%^5C2%Iq9mSY{DF~ zzq^H@A@k1T0oIbRkb5k+1V-xA+qT60t>HnrG*%HJPz%Z}PXM)fFwq}ViL+Nb2Jp>? z!=a;%L8;PiBMYLHD<7_Ma5CTiR<$|d4*t{N@m+jN&j)X&R{d z4(shu?uf8O$8rXF3)cx}*2(*|ex>$mn@WLfh-l$9-N%%!J%}fqpI)!Tv;MiuCbe2$ zKvZ7y63Pir8v$tL|Zl+Zj{~h7@B*VizBhgmy$j&u+44)sk{D=Qm zU?Gm7_<_+Xp)jse(?*zA2(iGC}hco zK_zReQu|!u6$CjXqw!>ms-g0qdR1#(P;qP^PgoB22Sh9JDGlu;-Tj6Es#xaA#qa6O zX-~vP>*o%g7sHm%eq_|9=e^-j$f}*MdO(d%`t-UO2t+Hq2dciRN;|kNA+;uqj0U~k zQ%v$`oK5slVWTJLTeku`MwAEo1H8yTU>fmxZ66D)6zR3#Ox+yEgOw_P)o{4Tj<&gE zv%F{-M;lCL>rnedSpP@`h?BXGU25!%H{|<#P0C5L>L>j>B8YQ!RA|=U7dMSc@>ad3 zyrVaA?nu4dev`h}O%lOh&^B|c?&x?CB%p*CvblLt@9hv3E3GMZUid^DX(R;wG?hge zIz;sAoV0)a9)QuG0UdJ{y_Nyx$HpFf>HZ$xS;pnNcIEBMvjxr*rTI@Fi6{EQ2_AL* zw_A0x+|oxmIklzs@~RBSYaL44Bp!aRTysxR#EIRDi{3W3&OrXY|0=jm;;|+u)i6dZ zEAXEBX;G>{%n1#CdFyQJ_4}IuZ`ne4Ua`uh>xhh|R~M}5?qu!QK4^qwV*sO|@iG9a zoeN?MBvyyrp>dprKrS^=10>b>GE=b#O#?bZQfDag_Y##DKBn3$n8^{t2O|JgbP(0( zTlyH->KcWZSNe+y)OnNIG3ms9gFam)HT6Lsx%TALP`RL^%Afy=3Y0zl{dmY z!*FSSHaE4aZ$FZ@Cp&dux1F)^{(2DWWtfe88QpVY8>4dSs;(-Yeb6yFpFcF-NWEX+2zqFp>db`)-AMM!Fw%`Y+*_eHUSk|b*FU_kuPHvi9`eOEG62sm}$!ox1?jNYtq3{I(<`o4Sv)f4{5BtC}7x0=}+Da zj7hO$I>1h2UVQv|3;^(5p&7R4rS$C00Q-sXgxckun3BckrzzwoT|=)AQPKW0uX)r% z5FQhcNK*ELJisp9Df9rm0rGGl`=(hCoPm~y^L0{J+r1V?3mRt`=IiV3uh7-U^z;=x4&Xr0JIPuGO4ZFzr)5>+sS*3p|m+N{31rv%}O}^ z;+6WF9p{ei4^z_C73l1`K3^jMq{!9^=#dB@p@OS!iv%8M zYChQYM+JmK1BIcx7r zuY%{{HFeuBF5KfTFV1{c_DM7Xt43ICZhyY4HWBfd({JxxFg@7DTFTBH*~vBk<_UMI ztV>WR&~)p0b3lA0HSFTv??Hvlm8hPCL$hyzwx!GO2d8JuWgUSDE6=O zk0JJx`eJq0A}&2h*rmdwLClfdk^?Nx%W92i?@Nx2n~`ZU43x!L{zSO*Qg~E%fPTnG z@`<^3G=Q-?D6OoAGe-XU5q$_3dEL%tP4bSO;`pt`+wpj*upG>d@%4O{U*1fGNfU5& zJF+=zlASNDeTqfglLTk`6oM{)INHjcyw2|G+2NE<-Y8zQ305moOLu(87KJ530bPR5 znl<6MVV->bg8fm@p{Qr)*AOCB)e7p_gU6c)W5x#QgrX#S4s8_J@9P4Y)|+!kWsdg9 zqT~~wDk)k;* z(u>lhhUxstEX)si$LC5rGj(w)sfI5rn>R#%-mcwj+}?X5??>~k^?Q}3_}kOSMYun+yh z{jguSq}AI}^!3i+pk%N3_aT{S_wU21ezD(2ghGeDf51>l9F5}GJdVcjLX}73+H$)` z6Z+5A?oS$p3Ucq?G{c-l5n;>V2R3y&YO&Lx#UX40#6f6ROJN}MeCPJ5gz z-qCyOi}^3hpa8(Fzr5@CKei6~s_okU-><}9-j(|DKIMNN2y)uMn6DFsNB?;sxFy|> z_o3qxob0MEHwkWMK6^93!p{+!(FmUb5w&M@oL;+g-Zzy%`m8l3tjfONy4^D0NDJr+ zcxCgB2mxvOv+dpL^zA2T%Ggd$SCQB>IfMmN%tfpu-9F{td4`rCf`Pme>Gx1B-mCHR zjSr7yJtv59*-BR9jv@B^+c%xX`rm(U3kAQXf~Fp<_T2%4QN!8$UoZMt^mLyze_EVj z6C~0td{z$n5ifMXME{qm5BdQfXS<8dzk){oG>F+KDdcYB;+!EV8beFtsW8#0 zr!0-P0n;~KtAj28Dx>Le`h7vH%pAM0*L&VbZSUx_2jW>^7$;xIxbR%b^XbY5Zf#5^ z}*C+7Dl8v)m~kwIO))Tp{gwlzy=pcz|1-4 zTuAh@PNm)D4F~&1%nBBw;{#WsvNdv+?5V)`dMxsib$?M&S$V zhu{=(W>bs1ex4tb24!xOqE*3k*10D&halmtI*`{LoSt|0_Bq^Yo$%f*bz(09T(tj1eC`tiPRZSW^JYE3E z$^rRObF$KoH~k~@z@rg#G-t3(?juPbU$dyYH|Qn%$9RHMITJL+xH~TBmVj7}&sSaHl+KSJk&|bO5FIEt(?9*S?t6=J-q-a7Pq#7~LZK`hAp178j zh$hX(GBZl<+>x+X4>=mEue;ILz{w?B{Z`a@{&dGtQ*rLy1@D?-2Pau-u6^FWNvj$J zqZ4tmvg(SFMLo@gN>CWS?2#?u9UFZSjApa4eo|W7eVZ{GCLnI=>|j|`p6{iqHNPWT z8A`t1=R>_-n@a%S(4XaSeEEiN@2?G6-;Md0PlD?7MdG@f5tp~cZmn=aGJo#&+q0iV7ss>N~9DUPolE?S%Gya%XN^ z3hMp+bzg^)L4SU7Yx<%Y^AP-MXp{P~319()tkMy0Eaqc2ZzoXFFa;2C=yOLk5X3F& z3Jc-;XZ0T1)D+8fGaUbWD&@(vdi1}fQks=3AB6vBDm8JP`xH^{k#w%9s2(>|zct); zHSyr_8~MTUlJEKFJddSCntKYKGjDzPwCOnBsHXcwp-E}D)=lf;l8wP{l{J5jI4*s? zpSRavWkd?;y?5}N*x9OiKhO25ewYp@ck2nn32mwPPCyzE(*LK#f;ggLwiL!`{R?tfY$_CF7WbfLH0PVW|;j>Ty6JX>S*~p1khL81YOfzxV+oolkBbubZr9G zx1203kkNQlH*1b3aH}+j^Wz4lOTiLZl!O8e3EDed$d;BqT5xvT-z1P@g++k6e=$*6`LFQ2|kOIWwQaPS4Bsaj%6t263+wDn+jEZ5~qTf>D z;kkfZwU$`~W6QlqoQQeIXT+pTC3;oU#pUCX^29|b@>_8~_1(j|;kPw|@(IG{mWKNC zw-gBZFC`8KxY(2irol{jDN7e23x|{jJyPHh(-iDrUn@>(%Z}cEhTqTMF6W$XTsbXD zL;ahP<)AE+@|{?I#&LFl!oqify~RBrTKKTa4Ov7)J&Jlbk2Um=k(%594&Ge(1fpTY zX5T0YjOSR8omr-LsenS)(wDU^Zn7sZfrQ){JD*AY_7gZt!$)AbF|^(wC2IUD1+k=v z_#M0Fwh(*6KnuN-e0f~(Mro8VJRUa266K-2`ZNF-BHJfV63ItXHxluX?)o=`CkAHJ za1_U`Gr1jrfrSzrZ7KQL0mPVus8}>VQ{GMOgCZ@Vjzu_CC}T0SE3t3}1i(eI)~3cY zxe$n(_+ZVxgka{{d5Gjddz=p`G^J6zH_KX>IvCe|Kil}z>OfO2q-{5dZw$o46P0Gl z&UD9^bM>6EmMKjCZkkM4ob>ZDVRu0%hb#Bh0IBM_`*^5p*pZXAY)`sTi5t)AE_8yg31$ECg&D#OxvU_|QBxV?3ft`Zua}r)*TK%JEKG$@Z$jnI zGM%mJe~t!WCJQc{RpqX2v3h`z(x1*qVfge^nVxc%w~j7W{-9K}mp!MuR;DgTue_SD zsEO1rfr}EUXt)4J*vvh%eph;#wlSDR#V^LMIx?N4apN5m zM~4L+jCaDHR#F+U4w?p&Mr)*0G=G!FM3_Agv4Ij?PWL+4Tw+1cbaQ>5;;VDnT~|WU zbyQOF>~kN7R>H2dlOjBqQ&m6GE!8%LN1HDP@?Zd0nq&=_5l!_A>!6P2!P_D}1bY^} zm@1YcTP@^*LA<~33$dlC7{y`+qnFwNf-rMU?GK7fW)QmvAfV_YHPFJtkpSTbt*?wm zpb1hNuhE8zeUBs}?Y zkLBN6ddRkd!gH6vikK8Qn+; z2!eD;w^*Y?Vxv=uk-+j(~pL70(&*Slaey;0vU9Z={ zf}R7O7tYsIwH#{1s|8fgxB<5bI@u&mS4~cRj{y+@N+|tBvGpp|X;(>}{m73}xc}}{OQ>Q0= zW4plhRzYb~Kko+NHEnQVjZ+^Bvc3Mu8APW5pcK}SqvCn1SpO`1(F6W00%|B;f{L{$ zKGd`%$1_`8x~d^=WEuPJmZXe>0IU5$SMmK6l5nFYApFR>pkH#(%I95TYM2Gz? zu~Cw<`PfOspgGtlXR|YH@ubTB?)0g-J;vlWqD_{e+yHwHgDth6^i$u2Fa2oEcKj6f zWVsngtLyDdNu64oT0mNeMv~%04Um|H<<;TyUmwO(lK6(b#5X-Jje*zG{0^Jm3+Ioa z7lGV*EgAQ6V?3L*RXD|$+1&lKYdd2+m5dc<>wui?PYwg7TP+R5g&sB|Ms$HIQn9DJ zAc2Jyhs#FQd+irXYO?YP*naTiLu|}7ZDQ8~`e_unjV_CGxUlENLI`n4xOEy*H1 z78N>I!{y*pwV#tvpLtVObAA>*7e|og~ayq`Ltl{ zq`aQPYpUA!p1=J*-FQ!68?rR$EG6>X?!viCQdN0}7?0&^-~fU*xM|uVNg|Cpu?tDo z$54x_rQ!^gQ0hQ%d<;O>kjl(IEZphBMrhQ4itYx+Tu{`smTCst!D~k?9B*4o_+HWv z4u1%9`$yw{f$$5i1JE>u5Mp#Npu)hB+GP<8O=_hC9FOyRJ_LI1PW< z-lssr+7Ha#V04BTatmBgD=POXmR82_g?UFzIGL>|P{I$|3p2TTs0&Xq&qS=4fkHa7WcwtgP zD%hjTZU7)h*@_!s(MA~+CdI(@%v%X#;1V>(%z`PMGIHzM_%SIkJ`!d}^2oe9jv0Vw zKsd(>>rD^G8kOD4I(qWv!HydrsS*MAvFMHUc?jd#ebxM@w3RkKMp9k;Gu+kQblPuy#6CrC>ig z`}@C{pAdCNxVJESO42%(g+Ck_lL6J<;W`hr;7U|l=xT6sPQ(_%j0(_|<^1Fcqp7)! z)}Q(X*TE?=)5f0m`>tvt+T3em$EqE!#npNM1{!K+a!226n^&{TDc=RR9};~hteV5Q z*DeKpf6!9?ULgTzI?g`~j-B=`gW2@dn;FJZP5JOIp*f9<+HK@X#HVEL_XLECL4!kQyJb)P0oVv|9XOBEs4ooX76UVnTz)94Q^kJek`L*p#6#z!H+AKw{+H=Ll7Q0pME{3>;!qH zhFmTx!h!#XziC7|CadV@-v*iZBRr1dveYtK@ibW z+GjleJ-z*MOFXG<)LH)|d--x(cA{-OsQRnL$!+fc>dXQEuD8<%`244PP5QsxYhD(| z|MuHJw1pi%`nN9rZZnW9Xnf~~47PY!k6fxESLm2`NDSymHL3O2y(SF}ZXDt-gym~e zzlGJ#7K3BYGPRiollk;3v}l*0dz04Cn1`J8(_llRkKy=cUbvbYW0s!Br~cJOKAe0Y zaG}X=UjODp68og)4SvjQx1j}ek$u38!`hT!PLe39HC!fvH%}1GH3@@J0va_L#EKhT zCyZdEpnf)p_y>@-V<>+KHZ5p*LHewldv*1h&+AXF#8fu|yLlTDkdu$eHL-dHy)$S6 zNx>`iAs@j)_ip#FYfn~{Vho;%yi3GNfv`qVj#^xt;SBPsL@m6@art zR!D1YL5#u@iCiN8JPyS52;Bv?N%aZmik8m&@c!&kinLH@+ zWi~JH$LOAvx(_y%&<2LSBledo-~eG10Xk8`F1|k5=Yo&MNqjgn(xzpwdN$4JG(x8? z^|aY5<*XHGukYC_xutB2iRk4Eippa0HWO|*VKUY%vkNGZSPfcC2FJ`_fyy5dW-Y&> zW~#`JUU265zxE%-^3cj?lu+(9om(#@AVhb$+ z#~+zJryP+(cuk-xGPGl-FjfqRK(`R$KhVk$*5m!ghrvJ!2tA0l?h*8<+WdVDzC3`H zk^9t-8GSE_SXd}ll2*pdIpMTnnU|+)vR3Kf0TOJfC;rjpVP5aF{#_BYxiY^Eea8?h z+ry`!fCkFdAsq4G3{6RZ!PCHP?GHjJ5zlj1gY17UNyiD?IVTWNVg@FX3gwWy*&?}X zimz*eTw=eqwVv;k@d%uLlBip(?imZgqY*g}uL1fI0B6uZiUv_^nXo99gdPii`XzgwO8waQaFR z{jCt6=kG!=ogxm@2YXzD!Ki1hgYQF9Zm)WDKmIB58UXwF<}|HH)nSOvIA4mrljJKG zxo*A@S(w}h+@7qT{R70H1eIwm*y5M)y3Y|wAq_J)1g(^=R3W7(EVL7v1B{rDnJY9L zy{Ug{oL5NnSMI$G=na1>qS5P48l&njqWB&33U)gXTc%kPbQps5u;GHg?yt@t*3|Tai+_?)qq%9;|Yj`LNoABVzr_fd`R*7ogy%aU* zLoAu)pUz1V$>PJkHsuuj#K(^S@Q!T2nt$?CznGutz;x&a#oP!?Bz7v-91oVOX z_#KA2*aP=cDe%+~Q(k#jGzTpBjzvQ4k|F<{u{y!ZH>D~RLxxX#8uOBETojTQ3uEd0 zd2TkP`!4Kedtg;4IgIVR08?>Ao`mcA)DO{0Fe3^n6To5D!t)0XCT}bV`IMxt^5Ne^ z@79so%SydbyY*}43*j*J6HxX3)-7Cr>?8-NqsFVGYeXU3jGif$=e9L;T|f= zx}Y3q8l1qfwXQBF{__6Vk?y^xv{f#QJ;vc(SU8Nm%FS$K>Eqk8rdqYQ;0HY&Gekr$ zbL|f;b+v!kHLFq@mVyJg7j!+I?rr7uq4Z%?ZuL;&^%hB}LO-WwsD|7qmS;6^FP=}` zvWtYz!YF_=P^+)n=kN#jHRY)+)Yi2#N(}+Myu5XkuI^#)&icMSBQeP4g;muN550hZ zTEn>`DPwnRe1YZ<1S6?^eN~yeqa}kavUanM(nTX;Vnk^hy`wy}$QNY>Ns3z0Zn?NJ?!S^@dQO-hcn;ST?JES3f=y1 zpFvf3RgWA!{IimawkOS;&RUrs1;W}t2nh@uhx$BOvKQnPcl5vI;8QaVa^j>WT%TIx zepSfra^sxd`ga3LiJ6l|Ob1jLKxSPmmazMjRW>XPTA7-=q??zOr~r`p4vz5?r6b(h zHK@kQ&Z@TP&1?3dvhyq5_kz41{>FE4&o>rEQM8`qrTLz)f8w}2*Vq&D#vQb;vn#cz zt%>c0dqpfzB(-Xu#X}-FttxVdaClhgd$Mi(zh>70q>XWc8>LfW%@ky5kH0&qz5w1LOB1#+9x38xMEk$R z!sa7Q>*BzZ!+ZS59L+pE5llR+%H_@VAKCah!AAU2eP50$P^ii0L!3hJiIZWN{@nYMd_sj5R9dFtO+Y<4Tm14m zQ9C&X5HuHyWd>>%lEQPzdUi|mpWm9q0|7=5_^YUjz0zb8_0nQ62@yD2*)Z%TQdN%x z#{;@2_x5YL064))3Z<Zi>>ec6?a(E zv#IK?WUv@s**XMdtZoJ!hD=r;xv^I@Z@qehZJmne8LMica;N~{p{X=A-9W^U>{X)t zqj%H_ZL3=QZtV$g=Z7l}`oBRVXp|cGB|YS?{xz}Zl^64q$di9Gd=SrrRv4i}i7U5i z^OW%d8&!^K<#%-;^_>WiD(q_VKm^&4{>w*s6PTU5xct`Xcj*MV=2ne8=g31>?|HTOJ* zHH0Q%bZN?TLOZ2NSE|eAUw?my%`Zk=^Q$gKus$1;gRoxb_qp2y{TcmvFW{zaf#6fS z`*Z-Y=|rElL$_r#?no7RTc%Q$foRlwZgxPNTJ+<-TZ|P0(s#Sv|PTKkqZVEu?y;i zs{koqaUULgRU-zHg7`qq9${ez`*?l7^aKYicM>WNKr$~EK*|FFgL`F}ALmuN@}dnT zmK8^vi$eijg5h9wc(VlU@)Q-tk_(Ti1p|pmNUe2du66BYCf8l{2$2ViVZ34R1Xt0$ z%zoVx6fa3k&kwxE`v#^`yp$fC*G@7%lm;q@q>8My)io+ALERrLSM_fYAT0dbV$VAg z9gx1j_e<2-!CnlN-WgIMLeU&aR#xnVJmgTV^x6H_?|~7hL|_T^517;i=&*jU6#&~#B#abv*7)msmLi9tzsYdjQ7E#_e@7ZzDc z^k_{eWb1WR4})@LQbjl~d-I9k&c+H*1#rv7Rw!P^4yk$gRcwA=HkR^D#)XNquB@j5 z4sgf^uJ^dB7u=K!g|SaqYMf6YXp~8XX`JBXo2BV2R3k}DJuuwWxk^O+N@;4F)TEZV zya@K(*}OcKypEDJPDyi$%BV3ErX^eMbayE((EqGK$ps_=f5ZUn+(G%or6}}hE`@`q z;rROy_mvro_D-W1)b@QNkDFo3Unfmtn(~GfLmuUEs`W**(?Y#NUZyO*LbjO;)B2Q0 zu&o}^y``}|dTxD!IH-^2`S6T3K*xS%KSYMRE-yJoTV?s=4&mD9-1wv6{k4O_vi4Iw zs84}?($_6BZa`N&hEaT-1l0AOs23+@flD<7*mOgjY9qLH(s18K1Zjv_M7+AC8jWH43hmH&20d{39t^( z{lAH}&h`Jc!J#WiYry0M3a&pv3(}Gei^e1dfhpu>28Suy2=;QvD)qWYoVZ@98 znX1Dk!^~pBT3Xr&v=f}ftif*1CADn6#)gmoLZi8IvHlRCMZ^4eT|48UR3j;&(&`Cu zu>2kcP4%bu7}_i7f~V->Lq~=jJrS&!a?>-A&Z~pD4Mj`)Z%cenZq_V0fNG?#vt-rp z&Y)?%tZ4)@KZkYw8fLD|;MiMkWxXY;8Pu@-BJY$6oJ!N!8Hjr*U#;O?gPCtDV&o9T z-kMq^t$V}2Jn3mMkneGm2)l{N`*A`c#sgme$Q(B390{VX5zjn{keS*iul!TF+2DHG zl<)w^5CgeKcGSWZNJC|F7l@D=WCtAkY-a&*9@AJw^18J6$G`>tB|oaY=sdK*a|ceP zKle0E6w%8ASntX2B9!mZE4rN0nIUa68P%R*CHYEy;t&e^6^k`Q= zXb7&niL>Me)4ws>)?A9yWTP;b-$ zFYJ>VhX9qLZ>N}7MB$i@{k!+6Q|68pacus4ooqgXg{`soC?YJknt!`?fiOqq=YjV{wVodW{R0M>Grlub2?Fxa53Z8{xdQHq#pS zxWqgqmgwkjbXq0rhyu1aNiT#Tu_jfX?Mebd0AHKsfM@+D$eQ2}up91v?;Wl7n*rX` zChxZ>^>VF4AKxA0m^k4!pFB`PYw7MAP<>A9z!bkr^e*E%&|w|kB#r1cpR{{lo7Gg& zz}}lrAwSptz&)%v0s8&G8T>g5sl8oY)OM^W+HMU}*i2%RuGTrv&r`ep1KY+)WsAOT zGeiWE&>)c1k8c>w(B#b5gu>PtA7Tn&@6l?))|*r+*eFzEousSohnF@4)^lDsDKfh* z-X#?P;1!1ywfY!4Py^xWQMI^5hL~VH&-GB*dN417L<>M*EynbN0ZKqE8y-YP41+XP z1c|O72E?m1q$QmXpoODJVlV{?$+{*K)Kjl~rG}+&VX2gGwb}(k=%Ji#Qs9PIAL_3A~M)nRu5zp_H~!QwTJxrJa-KW)D@Ue-Dg&utb{P zJHmr)9iYn6drpxqtmeOqHG1Vw+_g(Xib1v)=47ZVPDCZiSeqr4dT$Y z*J`5bRDn*V8ejd_(Q7{RUThlx`nheH4{g;x+Anpi>PheY7+~19N#Z`z6;$7pyZeM~ z9e?t)OW~w#f1FeQaQo7niB6h#?0WPQ_D&(`UE_`MLu5hgQx_TqowNr3_$b0)cjA4 zLwxV+iEN*z8TqFPzO?@$UDS$svQmgAhxfwmRMNifZk2a+2_76E=c=4~i(rc%tvcW8 z0a0BrU~3qN?ik|^MO6K9ALLI=Y01#8L9hO#MThztuj;1rc+b3%W`}&#Cj!m@x;7Cz zsZ$NlpC##R?2W%D(N0Y|`dU#_Orh2EIL^{!17^0c4I=A~a83*N0&+7Ub#-J1agNm- zKJLEV(B&E@KXK>dMUsXu{kYo1Og6e@kSBvIlEfdMz_c4l27vUc=5s%3HnIF&6x z*N_ov2%Huc81ERb$vP9zy8`Wj!fdEyGB=s3gTsX$*v1)OUcf3f-70N*3FZkCoROhU zNUzdi;gs&oie8^Y^X@pKg}IU9{H#Az^)vj?-pAmf4JAwUO!h~M{*L{5D>*5@@1XAT zj{8|7UY(3Ohl|qnwC+njZyKJ9Jh(quVQ;k}=We>5BZKWQwJCE8-2v3+ZK?dW$=n*SAxruYkD`+q>J_b(7L{BIEJ ze)*p}a{ou*O5lahO2ohB(M(q_|IVWmp~=`ql+(=kE6bEOK zBJF2Fq@n3zT2S{zgyYCu^}`lz0J=oPyxOJRX%*R2Wtr-Mzs%8TxnX7Gd${-K+K;MF z@8RE$<}A-%E4+<94P+8Bf$lW$er8ie^ zj0$#Vite6U=`B0jhXa`7)_i|?u64!|mDIdz&X%))xECfzwE;>yWcCwk`DSmm$NXD# zwLDimzHO3&f{E@-65T4hl9a^s)wcF037Vovt_!yIXGYL3rfXh7XTQtW8Du7dHX~=T z%35Ed(C*QO&SuZ94x}DnTK+gQ`cv4rF!)p6k-_>gJk5j%C4bqQm<(<5m&N+-Dx{ft2k=QRDfp|&&fAvvX{H3T#Rcnxa*Smd$Q$Oi zTz4h7D-moI44<=F~Q|p!aWO2MOv5O z1=M$Y*9WF19^}7 zSX2QW5lo|8Oa>aQD%cDW5l8cWH!*lXBeSdi07i&p%Hn^5A;}H_gTuz_t>3^*OtzF&>@mUP$Y6M8AsFy=`}QfNyQ@hbQAkT#cJ9v(haIyoEJm)T(!X0^vV=#y z6DZanp~MO9j1}Y1^T2P->@Twir2*u|KUy)oD*E$X7~1+l-6a3J!|CrrF|6mg#EtL8 z;z8iARk2uOdf;(Cdj;*+Evq6ZflKvz=Q|Qf@ql&pT%QTj z(d)Lm3}b5efO1^#xU@Lq>k=v8^n!{uBGv27*{>3Q`@8bFN&AnLOp){qe&;Ko3s#C? z!!A$;vWHQ28hc2>!|={+pn*UZbL_KgJ1wrOaA}z2;67@)(v$XkkAc)+D2d~2nF0$g zRMctP;IfkKV~w)`?2A3Z$PF`L`7ClOBvv z+m2r0-iM9#!YG0@Kn6jzZ)kbV25mIxmp7Ug^=N{6>!|xV(kbBaJzSLw^Ggu|NKQU_ zI}9cdniGw`wA!OGyXfI73o0kKNeWb8l0ww#?ND)~hMZL)6<0RbuG zV{vfGUJ^AywD|o6WO`F2g}Qiz`T{idE=hNIe#o|D5oL^=j~bN`79iaatEY8QVZXx0 zgi?|kH#l|6q3Ghj$%hOL)wQ7l%l6IQ;hvBz^M3W+KT|z0j0jNU4gHr73lKI4d*{K3 zE}PNU>^8V1m3q4omNBr&PwNY7K8<87qRW?S-R&1U&&D{skUpc^slE=inmz zm@m16x_sE^CNd+JjZ7l2PUEf@Qn?>a?|i+Nugtt~w(9J=23<&n)j^+Xx^QnXU-#r0 z)X>n^1yt3QDWsH6X4ayOe*@Dw&==wVIM6Zm>}x}`^n#PghcJA(ufmwZ!LRgAPaevq zM+4{P$_t?!6ZN_Q!sauKAK|D8IMb~IG+XlR1|n@ahk!ZNknae^V!ms8jJCye$b3o#7(k;n;^Fb)gTzK3y;aNs}GCVE7$UWazBfHX( z{$bmQiRpMiemq><>?izzbVK(1%Fs<}2ZceZT`|q2;qB(GtkHn+hF23EJ9mEnoWOX# z6a(sgQD*-&8Kd7=cx~*9`adAXCN| zccU%L#)lYMMRA+;nDqfH*P4p^I(lp5B^O699OX)wfF!K3AA5e^7>-Ti2;?vk{xZ+> zb+!HJQoToSt)H>N{a=_p9=bb`O`{jmcjw=x$fqV;%v2e_uXgAcWoXlc!-~xtOK*P} zn(Cx~O|T(h@SsRQ~vl40v(2VcR z(5o&e_r3`Jm50}$VDa%D3&uWkp7xX0!B>RhXN$_)<% z>8Iv9fIHG3eB@~QZBdAm_0=oJCZK zQ3xtSkywWTp4Z~voLM}A_1vH~bBnvGxrW^P6a)rddQZWzn9jT2HvqxuCwE0N4+g49(TlF*ui_{U*&_%HbnVY!*H4 zx}imi?!CG+_8a3SB6n0-&sQYjLuI-33U{v2(m<|L)B4I|Xl7BPqTJ^rijtq)Pv%UQ z8zfaU89eHy*n)H$hGl*mBNs6FmvsulpFa4CgZMxJ+7R{Ie!a$@6HewVRfr~c`!|5! zrTQ^~DZXp=%QJ`T8)xy2lQBeNx-M@;g%)iiJ*(3J)2~%cDppA}nuyQX8OZ^y(v`OP zYhF6z(yRQ&K1Rot8jY2<4=%Kb{ct6k*kJ{?l+e$Et>P;Oz`h_ zuIM*=q94#5fdm@_z6!Z|%I}yHPky5BAa>~YyWKw|DyTZghYW(yPn&*T69}h}C(v~H z?>DkP9QM?J=wa^i1{5s~ATp2#s9u3PxZ!4Cw^*A3z4z#d<=HWF-A=%p>jw+dM2d{Y zay^f5u15nSWbjVVa|ea4<>fVW;GId%Ev{=XR}G)?H*eEc1)P;>5dR@6t<9k4a|RXL z!8Lw}Ywc*Hkm343Dg}~Uy_YKi6h_KUeq&qD&{6ouN7>vMCRc!mL_6{uO>;+FqJEsFn=w$K z`ftfxesG!~2W$|ZJ^M7aU%DAb^O)(#TZo!x(jg<>v^hBfZfs9zj&66*wP#CBE#l+S z>WBu^`Y@+(wt&k~AG>wz3;i4pTX|$`w~cs>v@kGU@i8}}lx(9xeuy`vXsEkA+o#KI zK(npHUe<}rHvT#USQqMU(U4u9tO|!{DR>%sz1pQAlkpTH{=@wbuU9C(2&R3z2}gNE zP7pN73IP~XY)=6`SWw8EdnyJ9C;+*z!GRJORX`CZK~O`pje9x}o62#RsSellsg7xe zJrrM67FV@M^uYWG*jgu(^S=dn>N%TxXG>5cz|F{lAI7tQ7!)+!z?F25J&dEGT1gaU zEH`ID7)MfQ$QdEH?niRpALBfyQx>S`$18#BR$|%03>*3@6Xkm}O;;Wn^WWOuacJ{*B*GRPpN-*S8B})qyP{saw85Z`b~K-s{s@`<{K?@g?Am z*9>*Tas7Gc@iUvJLc41}mohphtG+%furvE}bp9SdS8GGXXS3?v+@<|rQDc(7_pKfR z*8USjAPSC)qIFwPgV0l80U%h+Gj z*B)@^Zg6GHfBynRa{W>UmoAY5)GQxq-Gm97laPYQ+>u)&8KIhwMt>q^@~+6D&UR|* zcZbsTKp@`ohDXtoU!T#jEnSDtv~c1s+E4AoVnNfB-8No zJrr7Dw1ot>9L1Ws{bjw_f=DBV?Lb?ykEJMjvQA3-k}MqtzN;!cK2|I(t-Nsupd3BN znZYVw*-K90*Ya!t@isA{fHiO@d}hbFjd{ON?+Y#dZ|@$U8Q57i&+@4-QW12*Nz6%cn73Lf zgR;|tg>3L8mixIbz$PS9f~Om+v%TmIre0NqHk#HuRSvbEz%fIgua%DSCA5uK}9LWEwJ@yd7?!y-tvkAqt1byfK{vO#`M zKbp#J*>BG5y{lS=zJ1L$9ELzf;`gL6+21Egid+pHOM^F!JH&?XkZd$K#vZKHjbGm% zGDs&b0K0m1MXDMUh)?4!{2qYx>fDv>#0Wz)j8(V-`qdQ6z!%(p<-9WUqAggX@V0zC zzX6Yifi?*udGA98kjnQpCdPMa-A)4zjj2mM^H$%Pd0Yj;!&(>y@kya1-@{H3*rDF|#A_956?i%+wyrg)h z?njm%7&xRF_Vh5dTkqnxPZ2wW*N#4WquRLrw}#w}6bc*OFb=8g^mX7xeZkmQgl)y( zroTPsgjy^Tm%{=O`q&v`Q*HP44VzqG3e5%YXX$d=1cH7G=}#+fzVqL#(0Gun4qx{( z&W1h!QPK$=*O}xk0_FxU*WV`BH}EB{zs&S>N$%qg=Qy64l+NVgJB;;?(0ih(rSGB| zw<{Xk<)WwHG*CJOP-fJA!o@-)5oI@$Q0~T!FD-n(?x7hjw>gUokuIdgL^{&m{5+lzp^Y2%(O zI%s00AZjtbQmv`!P0idSLO;G|yrU@r1K^+i^j0F%eG-|D_1dTPftn!4BYRMW? zEp#~DWQ0rF57%?|>8A6(8p5*DI*?sD~lLZj$R7~(Kfb2VfdH^jxrJZp`W#6rjb@!iEItb`2 z4qKenXO2}m$&4(HhnzL!(^k1?=qycVpEZ{KiM26i7ig@!Fy!Pv?C2>{E`#u!%Kc{gT|6>}5(E5iIUH@%2LSyTH9LkE*rLbYl(zDbpG|BR% z#v~bm^{E;?*jqdZt*agPQ*s)?$rhe$4aqLrlv?_U3}`3(OsRo@G*d9Ni&@C5!EQC= zbgN%XKue1XSmq8#JC9V5rV0-6i`h52j0wwVg^8Jwy2K8*)O-pxC%I(TZS}UTfgh4k zk;O~Y{v%n)fh9w=TiqUZSF59P73!Nkk2*K9vWF+6%u#S*yx5tl-e;>-H@+Y1N(^R@BS4aLn2H4$P<=GzWT zVAVtUs2i3>!5`Y%M_%9LXayX9eUWfC25#qPGeM#}E`88sy7(l*QR7qC$uohaaQx!T z>t{exnMpT_tN{ymjOL8KE1=mq3sA(s{Mw7ABnB5*F-QK&W1f`^tOQX&K~z82nHK?$ zM49bG2)X1j{#|g%)tHa8fp(EFy2eO=e1+gvT-c&hu{Uvr$6D3UNzlQ5sL}W?>}nQ7 zg9E&EW=W3}8^VUYuan)%rOe7;d%m$Nz+mUQ6uoAjOaa3yOb?R9GtaYC_P?LtkTV%OtD&XJ4D}ObK-k*rRy^3RcHFr zpe~7zXGl5+%#&w-)IGRo)4ZW>rSt*G#8@@I=4K6yJ$vKglP67}ul^e7}IzFLIi*omAplgWPXo$ zbY28ffbw3m_97p=ep1D3sqb0R#{q{(BNY$GY&QiYM)l}|C;)KSk6LAcub?jPIyE&b zNZ@@Le@v{=Zsn<}LGIDJm1-#lC4xef?2me;2sI)1>lP0QM+Pti&lv zp{_uv6_fu-esaI4)I21miVDW2JOoK?K_m#EJ}Vi1OtD6~ajyK7){5G3`bvEy7_#E4 zT@?@L%W!W2RM|%qJZ_kOr7K5pk;Fg3JY5sdT@V@HEDda#N%W3kLH%4AVjTP~s(}v8 z6k-Xb=s#Fc63l#jBk1(a>{kndY?y%uoFGW66leA>^zta^_lLR2z*4p)34Ye47}fk& z%vl!+gUE)s{5T`YZoqYXK*|X@qDKZhXN9*I2N}bOJr4AavIcB(?Yd8Y@qT!3u8&~R z)?7PqN{R%i00#H+f7&_sNRkwCo?#=phqroqT7{J$kL;YscNlB{-Pf-GK>8XNm6x-F zx&SINk-fTD5123#jU@vMmG!T>e1)?i2HLUMc^_TS!VhjUAbvx_>yZex0|0a%O=&(IFXJ2*>OvP>uhh%`hk;7n~z#qR*DoF zpn7XY2*Bm0N-d$3Szz&{-$4^ib)Fh_=e(RaA-3Si2&2rF7C{)0z+l|&T4ncVlhbX$ zpB*4lti!sdNxqFBimLp;ULq5Q zO4Ia~=-p=*(C%)ttRL_`gPSY&I*vfJcGAW0)*m##@$xehcu|McJNTArkNauj7SKJl zx8~JL-8WMxhWWp03U-u5fuCikl-{L1y{WG724xXhOfr`}pj@{TszYnx-5iu`mo3F} zj$t*jWu^&=6hPf%6;at!>%I||pin=wetByVf)PHexz;`DweW!|P#2}TOQQZNk-RfGtS8u1yBO#Mu@g&(d4nZZumM{L|(1L*l<#<@O0^S?^Bs&2nGe^ z9-H8s61P(r^b&ot9AE3xfMuW<#;4FY#!3fAC%jQ3><$d!MiOj1m}pLyuq78wZ)szV zY`9a{Fv&dLTVaNgqC7g^m{l_qICrl(lXuocY3S5ce0JOr!F?jRNXKFOiTx}&4010G ziGXS08u9K>;(s_t2Ktp?rwcW>` zOGI32 zV>+-=(_kyCuNzN=kEl?51Q&vX7J8C*D5d8YB%%ZxLmiZ$S@m*kVk8bGaMovMWtuSizjLu@T&JfP*JVh{p# zoIiuOo;)QVRz|h@Y;?OB7B8q^70|-(BC@9@SN-0W6mkc3Yf&vZ=L)XxylER>LmjVJ34$NY5_13u?@o#%Dm4|mj5Wv-j20q=oBZUFZN1^@^EzyJUa07w8p z0|4gNN5NnK0tR4U01gI_U;qsUFuy((3;`ez00sfz5C91Q&=3Ig>+8WV00IMGFaQn% zkT3uZ12Dh7DHskw-~bE`z~KNA4xr%x=GRXGBLN5!fFS`m5#PsLZD~}6a#?*Fen%Xg}|UN7!(eJ zB4JQ842pq40XP&4heF^`7#s?RLy>SO8V<$4p#TyJMnWM-C=3aOBcVtn6pe&pkWc^( z1*4%5G!%x0!qHG98j40kF=!}&fr2qm2nGtnK;alD5(7nJpco7k01#jR0Ra#&009RO zNC1Hb5EuXffDvFY0s=#oW z7>0nr5HJ`54nrVe2s8|Vfgu1m0t`n$;0PET0f!@ya0D8Tz`zjz5&=dcAV>rZiGU*! zNF)M{L|~8z0F3~n5fC&2hDN~A2qYSTMk6q21b{(+F$f3-0mC5R7z7f7Kw}UX3<3bK zU;qmNurL4%2e3#0iw3Y501JSzU@#T}#=^i@I2elrW6@wN28;zDSTF<&fnZ?}EF6MG zLa=BE76ZWoFf15`g}|^d7#0r0B4JoG42ywb0XP;6$3oy(7#s_SW07zy8ji)lu>cYa zMq(jIEDVW-Be6&%7LCMWkXQhX1*5SLG!}-&!qHeH8jD6_F=#A+!GbYZ2nGwoVBr`n z5`#r!uow*X0+w7$MPE-zQCmeyToQzb1M0#t0Raf$21L)raV`^p1Pk=R=<2L<6kiY% z$J6y{n=6Hq)5k{~!50*xXt-q)!`&=ZV(1mUDEf5X^aaxqJLA<$@j;R$P5YzO9_N2Z zlX0m8U);$a%vJ~@<5Wi$4(CA;)pWW&gN+?oO zJXE1Lj1P5`{;FMLHQ16VQd?@==(v97s~%c3+Zq7Ea-#Lh=R3k__e%PYxIcBmFKV2r z9c{`ZQ@*Tp71ckUYRtJ7_T$<8vbE7S^WUbC?vj=nyd{VzR^5Jz*_i1B;XWpwa$JsObYcA7M9e1!FZ=KX{?Df7CF&u2V z>ASZ+O#W89<(AWA($`IIP5)N^&ZWN9tbk2g^rOICK9zxndT`07Sl4waWk5wo& zuz)L`8RbP^3oXoxzx&ZH5B{8%)hbdKzt8d&Q+JAGf|_i^Qk?PA_0d;G>ULJC%<&N` z{;IduSCU;qCzmtKh~BQG{nUvli17Y)KRZbyl2tI%jp{)`{PPpmnbgRWWV^i2)ugL$ z+_{7WlapSmStP=571|bH<;XT%0xTXBzDs0#>foFE@-us0sXKPUp%|(5U>y;hAGuMY z^g(^oG5d@ytE^q!VB>A?O8;gh!mHN4!vA|BH}VUuhFHLeRHI|nXa01dnDIxBM&%1< zOwRQ*Hb2EmH#auK8Yyl?m3o$c{Q0^5hqJXvEyj6syK#L!-Nj&u>RDM6(S%oVe zSEpfxfU5W8t+Raw*aez~YFHgEWY)M;y2Y~&+cX|UiI`+#=i0@5Cc@Z9bn~3*g2j_D z;9OzARY5NH<#oOwarnW zl*_yQQ}o|4m*2bjgY}w&^=pt;>P7JmxHn`W9w4>NN1$U4iF^R{~G*?$)UsixS+L z3mK&HSI7}dVP^N#c}d!BA<8D+Jr6nrS1}L}uvBd!-iEwx3&FS$uz#th!Xa`_gas{t1zgBGK&;A#g6& zRbp+d$-d=fIa`g#4h4CX)^Eu;9qySo=pPbZ&A7^kP&PG` zdq>BYDd8yusTEEuuy|Z(z$tX|28+RSt)kcWW9;=l$wIYWeG=4pUS!lKY*=;6_LYe> zLWywsllhBJ2x3YyU2?=&v%^6_m?sl-lP$5Xg|Q&|l-OV|4c%XPHHSW?*h&*|zXGwJ zEADG)KH4xoQqu*0zWmm5uWzu~)uJNkt+id4)csEEqom492c(h#cqC>7_$&%>Qxp0|!G>0p{tl%3~=$weEXsk>vG{BH`SH}@02S-DBN z{q;%X}xs+c2-inKmb*y+L(A9$0$5pyFbv4>X_DlUopO(i8zosN1 z0d7EwQUmvZDB#(zDTznI0V&UE3nHNxHjx>cZS$wNsGb3-$nA>4$4>v`bG-<>WKoLb zQ2V|h2e9AExVnuDa`S~CijDCpI7S52YK7E z^-nn`)Rp3D8_wDp@?bn zJNKRTcY0-6PP=wipX&rnXEe2cpmOVjE@K>P>vk5&qu5O`tWVk-Jim^6YRTofFMrB3 zHO)Fb@@kxp^T$$4oHgitUmJ}on=)zsd9YnH`|@6E%jr*+FN=mZ{9CbDfJksIfPl$v zE|7%3aP9?#^x9kyjjG^$FoS{Jd}a3NB(W^Ex# zvRiO5`pUT7VvNj6p{2@crV)dy*hUyHoU^ZxPF)f|K0H1Ethf|@N0ER+G60~bkh~qb zHk4@afb?Rr3BqVB$&6?*R|^HeZDIxT*ZHIFBYc>pZj9TnX8NqWTg`g*V|_I{fGBc4 zL0H}M9$f?hErc!*Wf!Z8y{SIL#Mk0A<3$*AJ6DO7)%sBshf~2Qi-79=40*SwEX&aX zZlAQnPu3YaZK5)$ln}jY5O{LHWgY)6z$U&*t;rs0DFrHo@<&7^(!H@t(Qw zEqsWecXmEKF)W``3ug50zq#8^5s)iEc;>L2k9fyzgvIOx4Xw3Dw+34r>~-`|mk|w% zV-A%>#;bHl@g<`^!KQ+Yzntt`*ts|L!UhB?zkC!cmh4V{o3r=TwqmjayX?1gX}1nz z5T}Gqik{i(dtFUcuhh80^&zI~>ZjNC_ooZ)Y0@|Q(xc=iRWDu+hCN+-SXSqW4WVqx z+nFGuRf;A%komqkq<4{m?i`~wQLKyx9so%uHa@k?3w<516mPbgP zLap*mUxJ_>!emk>S0hQ%67g()rHegX>hT8HekCM=S?R4~?v4VVIRXZ3q|-(pLh_ewDb1*JHR z07o~Dpk?!srW2o=$q&mdV>Pjuk|Kk(#Q_$l)7@80RjaYjo&z+e^&airKufM_-Dl)x zV2oO8mgera)-Sd8$y~)W*PN$cl4qYC4%E8NmBwIIwTz!QuT;6rUZ#?hJ=hq3ULZY9 z(D-c;NRh;1$y&r%^Ji>bE}`d8c!3l(1+6A$(BosnLHc3^c*OJLqmZkiiqo|FkqU=| zh0GwVke$(UQdy%!KM-^$OW{mK@U@;Mqs&ks`-|!sMP{i&^Kd1_a0@@e-D}#RKq+J_ zfbfYRO}Ie6QDTJb?P}5&Bv;Xk!T6JxheGB_uP?>ll~Pto&{h>%PSiKBUw&m|^KLoG z#A|&y*(_9OB_*X=btTnKHPteWBeicO-PMA4Ce1C&04J9GJGNOA^?f+8gkH7H31)Iw zdmYY?Sj&x;-dM|ns|v5bNi=x+${~(z9c~?wOGF|W7(jMcF@r^TST_KS58)|tM+i!y zo`sTa8k&Gl7>nzOE((>_-Iir3L#;IwW&_b=_U3tPB15s^E4{EPBB3wu-i{e)3}=no z6n(9k^4k1bI~O{@!Bd0}K;BTjB?F2gcE;CqKT;|$=8l|NQrA2a_J%o3$g#P++3B_` zLXpa`+XWw0M(ijs$flLoSx_MapR7TD?qm+K-PQiHj{|@L4$Z*7qj~qpWv(oQ<_{ld z_-g(enoQhhY}e_2p~=)lJeS0NSz;RblqBbMf4a=$+nD7I9GWguLt|=4JFO&f|1;6O zf&2hKK)=7?0&T_b&lC@Wql|g;9HwRCmL;yq-)oZGyOX0=YB$P+`P3g{tS7&4c08|C z@~{zH(S4wF*fZG}5JSbtQo9q23LicZY?7(4%m^YroBP>wXSydPLVSh-rL{7sathQ@ zG+gPge1nHMc5FDfQxNxX;H6Lf`qmfbJ7-^-*=x5m8=hdf{K_x5%y%YG-+cM(2mf+c zDz&ZVv%}_vEdm z>$mmOT6)$Ud(@F0_>FK4t%RL=92IofnV(~MN~=r=qLds)(E&%7=WR1cuB71MU)uQ? zmR~`j0K$(WLCJAB1oK*$h;{}<(tQj!q9?k^LqlKdQ+Uzlx~hzo#T$}079eqa1%l{1 zw^ZrrMO0XgLI{sZ0Ql_(tVVIypRz*Z2`XfU5-vqX7{n0jsL{N1u;{kV@D5ZHcwpr< z2~P8NR#EjIOPwq*FlHG;;Q>5kOF72o^=}o;r4R|4j4X-yZ|p8a7AmFCL0@Z%a`6a* zIK3V)uwJ0@Ag9lb6fS&s&mQYRolrvOrlBO&j-X?HM~2+MBdSB>ma1Icr;4QxCtE7- zq&pa(qr5^ZM)+PjDj|%vM{@~e)CE% z#9sqGsCJD~=e5=yv8#$8Eho>;JEkQEr3fQ7g+waqEmEo zFLug3L8HophufBC%Dip*nmkG1X(?r5vr_ z#_2?bv&CmG7s38QgwiIY|6p+hq2cmrOMUz;`DMk2bw{88r^#B5OsakRm97Uj{C7UH z33`i6wUQZ^DnHpCEaMLMKVn^DE?8|LbwZK~EAjH;5TDglV|eW?G0h_e0Y}L?f`{3zO+1c&+-%>sn{c(xZvi z&7jxxrcEq2yuNl4`(m+FXBAV^!X>d1++2-MC$=PF1-MQ3pB)`=&kj->p4?u=5E?NE z;7i%EdF;%urkKs)t)ZC!4hW4s1CtSxfY|HcNA^rDo^wxVy?Pm$38&kG@rXFg2tZob z1#`82bw(j6w}B`V+VHFh;(Z?=O6Imoq=NKSNy}K}02Z(on6A?@QeOPnfc_2_Dmc-V zlSdn*Zm~WBC$?83uzbYkxvW45zt68jxnGB~a3-+1B3ZZnYSe?7)PxtF8>zIcgiv^a zLT&L_K{i8-xi&osXJp|W3Se$6k7-wJ{Z0VB*`PuYzV_N%PPf;2kKFQwQSH1M(@lGZe^mWqAPnnCmMPiIRV#mu*svhOYB}P`sM1@7e*&akHWfZ z-)1@Hf7DPm)hQQuM6jE_Jb2{3GDJ@`fw+0Hz4WPMF{@{~`aGPWNYBJU^9C7y+l`kPx9?b%-J*M-C4_(KIq= zm`-wKCge6}_$vj*yCeX8Ev(mE&60&+E+YNX&V?xLdK=zowkK%)a9pI$)6*=eThyZ# zURsEjXR=?4gYdt@MG@)srGFnqtdi_;Q6&G}N*b(YeI@-N9qDR@$8WiPa@N9q1DD%H zUmPx^ZaIalzy?bqGSiIt`t97eGVSlVe0+VKg$E!2Dew@vS@82jSf#Dsvm%SM&rHK7 z%l9KBR6-!^ZFEUCi|@NDZkFiq25pu$wUb1rflq}+fw-*eTV;;;d0Umy_~pfwz5DiW zD+46cx2k#(q$E6nfL=+Od?49fo}9GnEN()%#|d)HHuWq;!3GR8e~uLrILuQ2h8b5= zt@i7GZG+_FB<}9!^})Rbi^Mj%{T!oiQ1aTpc8P9>Dd`?OW?;fm)GjINQO- z1PpuXz=^&L`-prv%b$7U;M?FCkkNAI=izE~eN54}lBW8{Q-iU0vbCBU4mQ#BpSkMf z8V|SUI-;=LzhhQpI`|8-3YW-x+>P`6gdyV-GPCm~sGo%YLp?NL&QKsy7Nli8y>BbH~uE!sj6yJiZq*LL2NtPUh-AiQLLOAK&#B7M!=gVCyK`kh$IVn z^^8YXY2DX0o)YSx@v3PcDOCJ(M+5*t94)W^N(<>3`;o}MqlMs!%eg7Zi{I_5d^?aV zY1t1uDzec`m3gcyN{UKiy(VvbRkTjGKqs%&q*-QnAte3vojmpC>en4PCmPTCF!skuEs>B#FzW!bIrb7YT- zX%Px7aDT}xjuB$3fyB$+7&o+p*@v7<54b2%T5F?p%|e`-qg|K2d&%@z)ENiEod2C^)f|nVlqv8R#5KyJa$opn}BpVL$5T++H;l;$^T&kMkq2f6MeR za$V7wQvsfz&b3d?NO?s^nDDjMl{xvcsqb3BwQmkKh6F*!xNxe_FRUHVE_5felvtCE zy|muBf5CX3mb_TeisL7fcUONJm(OtM$#f;@bz`xbDR-n(ZEv8+w>p8{A7b@u`9EWRk}hfq zk%|5*5_jz)MZSm@RmmPc<7U1^EVe(S%ZwiFKS+vFlo~6t12AM0~}gpb}iNv zVj+3zoP9{d=sTk-KaaLgloRp*%7IhEYfe4EMAP-ueD$8OvGrpAGRu#_*hcwI2aXiC z(%HP$t+Sn7(6`wRJuJFC{g7Gz#dBJXv$~N*?so659Kk0(m9`OFPC{po2f|_l{caWc zu^#QOj}=)41%Eb_csN8P`E#iAbqU)QRmHJOpEKTN z`t@lCW2fk0KJf7q?Pp(Eec@`oOz%W}@GtxlBm|Mb}kgBgry|M zZ~qr1;9};rCh=N)Qd#lPO)0BO4HS6s;oWR|4qK`?q)8;(fIR13aKbaG{tL|dC6%x6 zP?i0Qs!4HuGP6Vhos+kcs`{uJN!^A*{M2wt5GdInxqI#F{;0Mi<9h|YJjZ)<3Iuo% zx`0B*L3vaE=8=I>Rj(||M^$0Ys$(Gq_I2p-=*^+|?y0I;T3mwoQ;y)l>EYd9^gw{U zVA5OuA993Mmuc%+&XvAOzJUieQ=PDbfwv#hIsEINx-E2(U%To?#wom;#7-{(PdfID z#l=-(|N1ApE1jg*Y8`}w`=NGv?;yTuc;(ea_cL(0^*%1Qevy|y%gFpfw_IinOnp9{ zUnHJx6<=)dB(HPrYY94KdqkltY0?#a(NMIgyv({UdgsiowIZ~V>4TH$=4B^820M+X{2whxG!~`>r|uvr(56Hj$BrH<&*~B8=rWL z)s{KVQq{@+&pG1v$q<+RP3fSt73^|$>Zv@b01){MyL@D{%TYOl6K2ShiId>^&WHSr z5L)%MAtvC8p!N%_PocSL2AT-atDQ=)@Ve+=6|2A?IUEqJ5sbFb!oPHBh?Zh1MaPfu zDz1tV0Vw6rTl~lRAQduYWj*!txOz=mflsZ#WO7*KfqSQQGUHXDp=2}hE1^ob%&L>- zz(%liyNu_Z^?hOhms*s|Cs!1#fcp=VeAzlLxKU0C(SY2DHdYipU%p$ro&Mw6T z{eoic6}Qz)O8o{~zd=0hwfyk@$z`#-u%`+IvAknra(Oc=skTKrq{=EF%{g_vM5Mc* zc93SX`@jvM1(hO&Jc=dis2(;j@9lhPk*~-x1IMq`=s#`F{u@2GG{2|wc!vZz{4_y6 zmbEjfesKEuLiCHxokRRT#RaOm$c6bF5(?*;`R$7FQ=aU3@&x&Mfv2^y&W5>c`k3`Eb>%N1KCC zZuu?I_q}mjdd2tUsPug=FLIX~X0m~ON5{EMHOWsVACwnS1OBVu8IP1M7`OO?sK;;Y z47T>fGBXELT;DJUV*PdU2fQC!_hnsVUgk*KGZ>7g&DHwAHm^7uBvG5C*;u25u6evU z^P#bJdlnT$&aKr{x4YPzAY%HlseW&DC{N?J2pz8jy0^`r^|3R2rYJh$TX0COvXtzmt4dw@TkcxPfBpO zK+L~h&0#?$)$}mO;knR3$ON;vbLNzK&6_cV+6T;~6sXykN>_<9hfkw1@3?)im>t?$ zm!7m%=hOC0+wIt3KYC-2$l6`%F8I|GyPMZ5t;4hk&jgE;x0eTL1)_+|h>`XYj`xb3 z>&B~4^5KR9sxpnzz+kFVPyh!&_`eNwf1-7^)lFA61&QgBh#W1P| zCyzJ$B0xvaF|t1f23r3xgp}c;zemM?GaxBOdenkwx*5mhjFy>RE2c9fGKR3@&kcsq zyv9XEsw;B}v=ny43%5z9l@#%)P<6qPB2=9!iWj+Z7HNgX3$;Tlrbh$tsMl~)a{gFQ zJM;@4JntngDq1OpUj>%&lzFZ%D-q!&pG2*8orc!z_x_sy>`JK++{l@0EI-wNj-Wu! zEPh_wm?@vsHiNBH?IOKDzU)wj1D%5)a0f60K-C~ZjcH#@B3tF1OpaR>z5GR6 zmG7lDx2lFz{})3LuarJ^xR?{3HEG_YXF~|#J)F_l4r#^*em_qOJrgNM5wc^p3qX8t zB-`yi2h+7NfJBC3%INejwQa7fI{&_hp=NyFV(gv)E8(l9VD1+0!NljcQRr8t2~Q>{ z^C$w9Rj>2TEC=01(}jv$+z=ey(HRLI(AaBIj8ed4jv)XrA(6_f^k z_k!QYAU$pjR{xWF9>AA~{!i4i>>4UYvEUiK`JH;QDZyU{NbCyTQ?Fw##gN zkRImLxNbK%#-CQguc^Lm`F*Aav4Y9*F#8liQ04ExZ_cjRpL>s=e3V1yb%)4`43~o zm^w)vTBjaI#T~LRSt+LJ;2PFF@m*@ThLi^Y+|fVQrYE@cdO7*lI*Cn$3+G^o*JJf-SfO-W-^?#zBpx4{nzO?$G8M+tslZ5j0_)%KM) zoreZJu~mOzo>(r40w|3uz%Em9rd_!(rr`;7VH#EBimLvVjiKOT2IU`o5%|r$qaw9U zkFD%2UrNE{)rIf#!E5N0m($+-SpVOt7cr^a_-zdS6m9%I? AfBVZ;GRbaYSt^eD za1@F}XtT&AuDR zGQ0VY?IVZPfb{A!Mzw%_3V`P+Z>J)b7 z`tC~-h7wv$GK(vgedqgu44;7*b$W?@OhXTR^?kNNkgn)n?iw!h5uAa_^>HasoAAs% zh`xX|S*OY&^SJ0$dRe|-`03u_z0nf8(FgiRo8uA{(tbxrZU!aOjnv6Wj3PDSGpzy1 zAD1;IB<4E24&zS)>vkWuxnKO^cK5o|=v#k4vh@A+lGu+q{%rhp*OBX~d9S48KFH>` z`+9B`N#TZ8S4LH`ai#_+mu;%pO~ao((NOH>!m)GCVmnN`xe`XSFj?V{k@1E z$2)~&YCGX#Q5#W$hUsgvYg;?NFW-xLT)S9YJeedpKmbJ}m;d|S9@B>4q27y)!532n zy)kc1#S)_VCEKDoqU*_c_haSM2MJBu8BA#>J~j3$m>s* zdAvbzywDkSO?(PGrG_jV4&{FY7Bdf894^pyYe{+99uk#^L!5Da)K)jdxFa&;XfRui z*}AyFXLofXyTWzxGtBZJB#PVl41H`VQaam;Gk+G$;@8=ZJ5F?%J=^kFZm(Y6tgXso z)s|p#OF6ar%W5yRus6AyCaTtx1qnZ&ENVz7$ro5!>TeB|Py@Bro1qO)YPX|mA7i=B zXlqdN*mwbfkCm+nd)m#vQGIhMVAe*9ybu+bJvyzSd`w7V3RzLk~Lbo66q z@$wz&fR<;o&D;N_!~dK6G!!2?7gQ{p1Q!i5-V4TSO4qL~L|>}$8IY4dOG%5NnpRR_ ze8i*HC8kVwrCX5Td~>6bc3#5Mae$2 z4<+`axvozeD4bMCpo8Y{)8GA03M7NBUK&b!p-Nzr8yz}klSg^MY7GznYwB7$We6!M z1Mc-8E*pfmv0nHliEJb|3b7oX9odM_54mo@QDv5$}ZVKw^Ggka4KKY@4=T+3Gc|62Ud5toSG zsEF1|w$A<0YbDHsA{eP#0E|FuA*WDIE$ghDJ&onlx zoyJVP`qO-dSD3Z3ocHiKFGqgOHjvP&lNs3^mgKbKBX`FeOG|=*8tH&RdH2~k(Z)mN7&+F@NArt44uM+F zMOq6JuG-xg$Wxl48-BSv4&-Re-ulkdhxDt9!iJGdzK-o(>|qY2ob&v+-~8_KQu9ue zSp5FHMvo(TfJ7sfASK%`k#a3l7oQC>XSYcjJo^A&T+sXlo_ZZm2-$-|j!-tOZoMEN zAjLfFmt3Uy!v2rv3;*LW*6r&H$yUMk{5rFllQhpQAJ``-0f;^t5aanxiPjX=6@E%i zARzk?d3~ART3EmBwlSfA0KL0#=jv@r!Op_07z2m(w<$J=_52L4jrD?@P+?)~?P!Gk zyI*oCuG!cq5+`dX+bl+oJ8YKJtspi_n|^Hk?-Ri{J`!K&fzTrnzQg>~^V>Oe%Qj>r zl3S(eLXvGL^0JKh3P%aaN*jBZrnV+>KoV;aN#BS^oKTinbBQ4LE#P=#v5o4^U04U* zujAON&t>^Cc=!Q5TuRTZTfv6?26TZ0c;WrYx0wp%u&Aw)G!C-6^R+Y6u=R(s=^Q$* zuN;Ei6P0}XMu%Bv>s0<4bpDJr$5opg1t0@(oo~1AWC|w!Yr9=k3gqT)$YPZ*4p76!-f8h3aU3en|w>=L!WFOYa#J}ShHh7qiFCX?OCV9-V-QPK*`t4D+ z!+2d@{QUR!4^fw|eM46q^h{)%wx02Ar!^iA#kJgPa&-eqefO<#x-okzk)nWELJNZR zk2`Vm`Z_g=T~a1$@}QRxx6B?qs5$w`T3;u$*ZzLH#LYT+>dB9JTQg^muF514!im|d;g$O#__)@jAhIj+VN|Zx zQUEOxiy~eLt&Vm8s6$*aMK9Wcjh3F6lXmzaQi}az`B;+SIo3b*K(w^9?AJ z>M9%6PiB!9;}JhGcEWu!04|wgBis_Z%aOB<8K^+3ClmT|vo@&#>CBKfbv3 zh#gLObRgCKRSl@$)i9-et_A}7_(_6*v9`bMdWOuDL&@+&Z=&7ueqEbfI-<51NXA<9 z-$c8={g5v6*rJ#e#if~g>Eib>Tqe?qBD{Dedqi$G^cFviD^vZIo!V<4gQ%pxx=1HW zvijq`dQGurwmEk4G25r&FSTrz($?g)r4Pf6E6>KB)|Q!;Tf9Bt+^RMkN^uONdt~C3 zV3PV6`{C0-U8PQF;QE6Z*eeOmk2#(eKZ*~n*G7dn9aTR(;8PqavCO-odB4UfMop8v z!Vd@qvh5*N<9#*Y_4O=5e?1n8iX`kG^+Q z`(yB+uGDAW2dZyj=E!Tw#eIjk#~=L+s-ZvGAK>0NGb67j`;LS)Y?c;@HPSqdDru?3 zxb<)sv7w9WcOQL$1@C^|QEDr@jnr?KhijC*pp+LKlAvLae$Cwy(DaRzziB6w-GIxe zymtG{t%F=5ajLtEh{?8_TwrNSHL|Clv)xsm9vq42q!gaE-B^^g@IjpSnP9r~F@UZK&f)$n)O>2g6q= zpOXiO$^V|JNSSS#ep@q1#G4U49d~tTq*N=QR;1AjE@=&jKP}ZX0WVvJ({NhuY*#%u3yj$_U1YbiQ4tOlx<|m95 z#B|j-8`Yg{ZA=f;ZtJxMfXKP-)}254MlP)l0asXtI&*7e-L0=&&P^7JH7BX%2{KGo zisI5daU}%3rF8N0nR~p^5rvmS1e{!*H@-`?c54VIacXDu;$Fk}Zf_*kgt6I^Zs`sc%YEl&ml)b2fj2Su zp9Lp2UkeYptW)^wIOTKn1x0_sE4&HRnZyCqgyyNduM`}VX_8F*2v%M_ z3Y`S2-#UQ|QTvy_%TLvi-M^E54HE#)V3uB5q<&Ue&yYoPI%1RUjZZe5itKZ)J`k+QUf3;@$f@Te6#N+`Dh@a^Wmm5{X=a+3Bg8-n*808sT! zkw#whQWevPbwwAGyW-bw8l^^>US`80rrx%uO}k_OeZEZ{Vh?Y&`_$wN-yHFls93ci zcO9hoYn@idy&L{UvgBMZo8GV;mn#p9SCF?&`3HGHada^x?tWX?!Et#rLSRP?M^3I4Izw7!H zkcCsg`2v`BhsG1Vv{2R0VnHsgF(~Z<3f>tjKFQQv;B;R89E?1Q@XB56F znjpBCvdwjk&v>Z9DkPkn%bw=gUAY+$H2yKQGZ9ausjX+G@p-h0)j#j@n z?k6SNu@keL_0Cn8xFnN+Zc4!sZz8{IBmPaaf@bbyYKJ+_tw*=BnvX`3MGfuBP5NOVv=i8(v_a?ThWEyb@=`SE55VhaJq_b=w|n;^oe{0rRIYoX{S*FYvR8YUC5vB=bs}PM9enD{j2$(3h8$~=Es44kU>$>(?M5>4JShl%< z*Qso^(`RR=g6`AL#cR`5Pr@j`NZ<>13$POuRP#y@P7M^U%7Mn-Wh4d)y zNZzlM`*~s0*f!^k5>kO#ga62FfUE7)NM+wxe`) ztE`r*@8rmQv4@k{dJ5@gDaHSrg4(>-%i078T8)HvLZX->9!e7J$b7fxb~dTMlCJqIs=NF!MIVgEkkXo0r1Y>axfa2LNpmBTcy zLg3`n5y`1hQ#@JqZ{?$0Obl??xemH~Cc>xYFmn~jH?OQs5j2QCx-^+W$KHxst zn-D80xRnZ90TEG2uA4mAD&wO}>&zdLEtUsv&Q6a|26#G^)iVFUP$pg3UF^NjcJE`; zlg{Oj|9>h(?p2dO5~{Tn%b!jj+5!BEy*mEnydL-N4lS?p1j%vwWNQFEcBXg0y!%lS zc#n*R0{wu1h3C9f!3*3}T{7dEKF34aq5EA|@)WKRvF1LbkY4KHhJ4=b5%b?!gT{Hq z2@c80>)+>#0Zp=u1-`WJn~*m+!=;jt>-MwM{=w_kMy$(FqgS_#`t(vkgaQnSz+9i< zJ-;LOFfmAj*$Z;#y4`^^Z+8!#m%f}wJ(S{2+U?{{JZWpHt)~m!r5*pn&qtCG1WK?rY+R*Ez#GvqGUY!X>QVw4K+Hd^ z9Vzg~BPD;ZdGpBew|d)o-E=Fd|bWVI8WPu|3?$ZccZ^s zJL1!?e>Cx5vN`*wv`sJa=zU$qVh{Y{Z)tlV<+9_<`$^ZHk2$mn6q*gl=|MnSqExW_ zqQgWy_MNk6qu1B2GN)N_8Ma-C>B{>O%ef47TjSAzAet3miWyI+@S$V>D+T z4JGt4dQB~l-s#{JPn=g}cOKej4FiCnDFcb7U7UFnq$b4ay-oRSdpE0miTs-46R@E~ z-A;S5_z-x}{m~Qqx5;t!^O5YZG`dN#)=ts_`#+XFwz`(z)&(BSJY<3>1`(y4n=4Ip zZcIo4yE9?^k3qQ_KIZkjvElp-%XBzxY5`FoLi?{P4h@_?`Tyx)KfwPl2m5wA*(y=Q zguK5W>|cMx!TM{w3%mrUo-@E>_$V01fMy7je@PRLfj&(_Ym0rgUkt=aOG?6fo8SD2 z0p97tUkp&Pu3|#efD4jO+T3z_iSc9*~?_2-9OSaoAA&{7Di?itm01!p7 zhBbFB1uS>Q@p(_9=f30dre1YC^)8B7CGu3UoVF<6v+hN->E3b;mJaf+Xq_o?J;qB2 z{NZkU>bcDfECN|%wTPR?-z$5weGdfwAFenUYy?ttK0Nn+tuPE-qxbwo>iJ0zOeUbG z&x#K~ctuL>7PLT~b_}qPP@_7)TP%|nv=9)ovLY6q6_ zh^xo3TK-4)C~CF7!o}pCW$Ac{-eVvjNf#fm@wB^zw$U|lD$H{jzOWyaWO5!Ce0h(^ zlYP(5U2bakP9*VJ&RVY2H`^rztmsJ)r96*%j$6F&=fc2vUBp!W8LMpucMJTagIgeF zr=8t-%ALpk#mtkPc5*}w9zrwL;2oS*B4e(IUHJp)jF@xA&?dR{l*711Fhlm+7!Ff#!8SI*gKh^%&SNcugpXFd2cY`n{>+Pt9m-SICy@*p!$L_r z)SI^WsNZi}cHr7|;mskQ*T2r$i+#%4`a#OhAI=@A&sfvB=A1^IKx3xyrapz$;olqy zY83(krOK52*^%ytiw~aD=+ybf`P`ANIsVC!JVx<uf;I)yvbS8f;f+HRrQe>tgXRPrak6+eY27r$6}`rMIfmnZ{?FPP~j~YMxiU zdhSxYGh2zb|LyrR(Y@6V{&&0wp8da~(ZmyG&cqXff9X&o=nNRT5F7wpvE0{W`^#N76y<;7{CC7ZV*L4KvB9Ilon=S=mtSjx{*{$X%!R8du~!R z2isLu?;u8K7+l=^$~IniIgDSao7~5`m0nfQse?w9^%wu?(jWd4zW#C)5JY6Zur-Sp z2o}*Fhx;fcYzU9U&M_5EWxW5CKn#n-`%l_Ina9eNeSUPrxsi?zd7j;{pF852>GpDb zM--M2II2IH^FMXO5j~fgwx|^^v0INAkBznZNjc#`aAeI94}u6X7#;*?IwBcu=RP4f z2zV0tY^|M|32wg8Zxn1X+=B;!FG(Vbdq;xkhAAg2C;h4w=}3r=7c#x{_WEeN3kRz& zvpWvXJc|}FjT?mge_;5!Y!ky}TBx3@z>j8GbZn%rLs>xTH$6HC5$57dATVhdIdOz} z7}$3kA`Y%CO^5>ONH0YW4n!b-O!esPEaT;g(I@nj_QMh{sSM3pcnY(QSOs07Pv#9! z$DVkv*#6-TBiDItW@``GAkzF)A`Tq;Q@YAR$mOW(Ju8o+fHqflF_!P-JY7m}-lu#+ z)&Y^R-YfEyB|RhPX$Lw_n;X@3r!$f20n1Q*H0kkgma z>pSK59ad<{_fz%DWJjWIK4cgbT=i_VUJ4h5jkxzf`O{Fr?9Kq1 zBIn*DvYW!XWk9uR-#UE`GTg>h#Xi2X8_mw&lHkHR_jOlGYC{!iiWMbw|*kM~#jEBp!mZ2y2i;8*x#{73K?_8tD#o_&Wu zX5ikFKM2{UVyabd5mXz3=ACm*f0@jS+h52pc5z9}HWRP1wT7I*qOaHwptwe==Qy;HwIG9laA(7??u;7yFF?qSe1>t*}OUTxxHxW0_uGeqx_i^2h;E5VLODwp7!)o{g8Iv zj8gSmfrh1dKohQHT+&d)Fk^E4MS18QKgE(`XUelPK>_%a2g}rc^VKFX#~1#>=+WxPr|UfTJwFfuoT-PhE>&C6RVKk==5% z42;~n>ed#&b_2TlV&aBIotbvi!?r=ImG=v8A;8_&4>3k!bFkf$da~#y zNn~)AHl#}3h5@c!ISPtc8#~hj|0##U4yXoL05;r5K0EqI9b{`TP7EDKFQh7kQ!qbq zM|H=tO3^X4`D{) z7N^7NW$F~m%gNpa`q`!OOcKf7zvaO67(VNAn(ID(QR<{&S`2GX3hN%{trSn@L$;RY zFVinmPf^Ed_|^6HCx9zdS$^_VXaP$g60kYKltTg*^*@Wezqtr|6#eplN#re&zt&n$ z?Zlx2!DOEl3fW22cudTPWqmp*gj*_*5?<5MxkIZbMk+{`%lkko+Q8?FSH}Z)&<@i1aL1K*9EWl(S0~u3nhbRU zNxQb61yf)*`Vi}Ugs(%6WvEQbQ%d?oyAiGxLqZwq_$Fy!;)`60_Xz;fh;K#PTW1M8wmze$A~vc{b})K z$EVjhQ$W=6l6ytpFWZ;jFefVgB@=hJrkG8P}RS%P~oNxFMDoi;4jd&pZ5imBdgeVy^ji!*@rj$Lh` zMh<7@lg;ItbeVW!v;F$DPffR1s&4rI^D7+xs-aY$@z|T}fD?e2C`Z0wFdM`D{L2KS zW_~*aw8)^vL|$4A;`s$Gf4Y4&Q@>UrABZx;WgnD3XGUtS^2;XMD|b(ns|`lAKa{5B znzkmDXSZ%bsR}h@4V&9TTim5jloGCWOgR-f0$JZ!W-eXNKK&H-EXjMvbzICo2d%n>4xk7As?q8P(wucJ6C<4~rN=wUeML+f(B-(jNGKBu@KSYTQCXUgCPBm+W# z=&!e=h_UDYs04SUOb8=Hj2B-dah#`g3CWU>Oo=oaUdrccXJitXub;n!UNC%~fxfa(B+SfcS)Bon)Y#u54ggG5!LngPrA)pD1}EA2lDU-oay z_0T?_AaFIulLycU=2cLsBp>$RbNFaN8|Qc3WtObrR#GLL08_L1EdH7?X(4eo%VX{I zvo5PQJ#R#t)kLV@2IZ1>ghn%$nv&uQ@{~nTiFS+h#`I)$TGe&_V+UQ@%Tm=;*b5xr zi1f!9RjPML2enzH`fm{l?0sc60TevFFzUKZ{S(5b<#31gmwVW z51N(xW#`5PXtP@z~2ECFwmaI#{$&%thL@>w|0CJbO2d8CwoO8Dm zc+VZmYTd-7s{zo5M8cJMEi8zHJO|uI^6uPH$RiRMds$7}|ta?Z~VB5w1}4I)fZqeo!h z^l*lPpM;x-d{RN1!M`P@vTCFg&{M*(?&t|QYy#;eNlO0GY#D{;>X|;p*B77B?C_dd zhSRW2UrH!XUqL}^Yt-UA(8K&D)33ZgtFL7N078@(0RK8{nQ;5`Csr4Fmq0;TkW z-1!X9!88rm{NAbl3`ega9>6-vHc$Re`KXEYow0Q#8xIitlD9u-UF-Yt2`QbuXexsw z!_s^g>E%)i<@0xzeDGPjrSWe|roB7>J3z$0B)6hD1aK%r;cOIC46zF(KBXgGBI!L3 zt!i2hrl_^ifZ#cTZEY#_mR51qB<)G()=>wMy3y;D@4riiOb@88?S3SRC4yvP4UO$8=1>q0;bY<=I{#UNsdP<+s}S(iUm3@4XIMcKy+g-ms4czuOo^3_<$ z(*641t8zB2;jira7mnpSczNw z)Dv~ok=qhZJ>ZjjC2a4p1>zCaY(a3xz}Z6hj()^=5*fKId-6$HU#s8-qFtI4a1s$* zv~#Vy2qt@O7D4mY*}ecyzYoaYQ(-c)QL@z}eF+b17D~v(it`r8cixQ1fy!a7F4BMx zuFgZTymX|ra6*EJkgr0!I8BO96v_uM8{v0$Fboc*hjdY8G_Z@a_wKv*4wvWwM$UfA zG+r12*B%J~Up;!1zk7T<@pM(D-uhffcHa$Xki`*};Gc2v1PSY34mXfY7U*;+^zgABiKE$<#@cwp#q7J`V%1cHY9$t-5pOz! z7f0L_^G*O>98rqQT6Hnt2#dHLC(uEU!$ofQs`1G0xOf>OP!l}z2QJ!Ebv$5CqWt>_ zn0q%)E-A!a8rm4vrWSB)$THZpd21_kg_vXPk=3BZt0>ipbWg{ilv{n@2}{$zamD#( zT>SR;6*#jkK$o^VQoAI;Wrrct=K@GLODCW+9Hnf*Q@+IV@L>+0cUH`7BM+<&$8V2$ zGn< zLu$s2US%nZpNwQ(g7**!CzLb(YQV|j5Gne1M5@m13(ah_zpj4|M8fq7Mr+zzg4I8u-&tvyS5CtdXR7h|F?SS7erdQejt!5uR@@M z^e16c*A2K{e~b%=cr{EP$k1(QcA0URnG(kCB8|jfg5vQcBfpa#4&E&Q43 zc`c)sYo#p%p>2!VahosHk+JhKX*2PZ1QvNg@?8xPvG<*C`K`&m5ONQNBdKFnbhqS^{be3MddI9 z4i(@7n1VwB484J)tyI-EzPXXz;6k{n zdGD(V``V**YfsleQiz6aG_Bgni7vn3U(duU;<|JAPc^9k91{O(;3)n+lZarUz-pq8 zxSP`VnM8yzMO&7}DZZ|>Ge-uF?K^(*boB8*8#sO~0)OFdnU6rf*W@4$^j^yJcdQK% z-^O8`b?Txhzc0Ja+R7&9WV14w{`1~2oPb1?h0+A$(bwg#&&ll8G0SY7-Tpb0`k-0h z<9mQVb!f@Aer34E^1sBW$-k?~4vp!l_-lt>LxdH=@4>lcNX&xzV^1et>Z_kmmmw8i zh*pH1e93hi2I!dEi7+nYa10(4jL9?Xhu;hX=Z$2Ewqw4AL*&o|Km_oZzdxe$0QrpO zA~rX{c0ZkEJWhGvs35j?sdSb`oj&Avyh>a|fouEY>0nl|CiiC?j4nneJldPgyD2Q6 z7So`^A}L%eAq1O=ZwL`y3IS_NbN{%D$>zz8I35;}efik_ zsf0+bq+QNPo_s+I|JvpzRQSp(K_0HUb%@Yc3K z-lsi_cn;@eoiGY-%zAizioqy<`M@oQCCD%qSH%qpOEihIMD~3&t2aa({ft5DJx!}) zUrCgLY`m?Drf)>AK~u<|pY+~3=&FLqaoz2MwqXdC3W5IfI!{-LZ*m6Qsk`vGY0CMN zPW$FQ0bS##*y!~pnnyiE0ziUfzu;0)x=r7ypF5}#Pzvzl-*);JcC#c#C7}R+jhOmz zc!HIBoS+J4i_|)^RtoH9v-oCL40}8gkqHJLmzQIGlYXmYgXRr0^z!)0h}$KZ3>uDD z9^ty!=n9vcm_0DuQ7^jTbDG=L>CgvqwgCqcWb!31jKNhqeehP6B8O^mItC_V5F0M@}B%)kq0(Iv&Sh9C-bhz9b8{KU`jDDDMhJm~-FNuwxeu)~n#>4xF zOnx_L>XaAJ;pKTV`CrYkpg)FtJ{AAyAZ?T}+oU6W9oNg5zn^V*X?*Ft@aZEXp8CtR zMRh{n*O}6{C6B+a$<=%yk650KSTE==P`}ROl<>GrFG`ta=bPb~(Y788?-S*vMNnp}2g~o9zr2Z{Zj)!b zbIePS*7_ZX50V0gNtlMf1O>Q(#;sk7Xt@vQXQHC23(>K{ms9g&y=O@$;$%kzZ^15- zH;;#d61%NDHC1<6JRxnQ1L0SgdLTd>rYGHwNuiY8LUd=B#k!PHfM2PkDDwFIih8o^ z_gx!=in3mlkuGGvkvKQ!pso6=%Z}Hv;lJuwa_T?TJj0i(+<3Xj$MR#u5{4siM!vb7 zt@rRF7W)ko@O=`N@%&ehO0L{%Ohew;!yl2RcAn;~WB7~h`B5mn>bI4RIDGq8xb||+ z{ar<{%vbYqu%EuJGqNg&_(EOmd~zpeuJ68(VK>I7ZocY1%pDu!$)NN$RRh6U$#xq4 zxW*WNf8gnMPW^o$Btq0PG`Pv7O zth#mb(uLDN%mX|syayp@G;PRW-pEg#dhTx*SfRVny-gZ)fcgCFV1 z)SYC-kMy{&Pyc14_w{(eZ#Mh>I_Ez@65~j!Nr%>tv+xP9`<$k;+b#~Z{PJ)Lq%r7Y zhi2vEjDISr*>i0Yh`3t^DdJf|xXV?acyennO&X^6;nr~LK!JwTwTrZm916=!aY*t_ z#Adkuf1n57k))((Po|+GnAbRiIo<7%4;D4x4N*O$BZMpLSXpKBCINV(eIr?6}iW-MDILSO^yKE1PhFzS-CS5uDr^Sh_MF= zGZg$ST(q$>R6>P`#KE>H30jiPv`O-#g2Q&uC&+GGG4;|}jvoCT!IM+`D-h-n`g;ER%UzP=U1f^YjSVz8WUel$& zTwQ829}Q_sQ|G!&Lrf4+_vC?5zYGb=>RQah+K-NxfuUoc>dW8gf8ek4Wq+U6qENx$ zHGaQPOlj7yr!s7;)!5|eMWR35{`ko_{x6xR4XWC|(QWeEkwHRJ_Wk%uocw<=J-*ef z4IJq32kkwlosMhGO(m`;5dN?FyA zB6B#tNfScuyfFl=YBB4kaA^uMGk^oM77=&wqaHR7be=wSR2n|jI*Ifa%n98^A>iYa zYY230N3#ncOD&>N2ZNeXu1#hP0QiO)FTn2-G0g!R|Eh)nkKVf`9(RB&N`>X__NB^U zHlAI+61aS|LzgQW zn`e)Sx9ntyRK$;siBx09J?)fwc0LT$v=sN8tePhPoe&@VY~!FC`-=2LjQv}p#YWBT z($V_qGd5zrukuUylqM8~ul{VpDFbQ1RY3grS&&s&rGd&Zd7ilCuf{sy`Ur(Q6}Htl_isKs$yyZQSK4K}ig)m)%as3NBpf4PW^|jtk_#ZA9--AC zK1=%ApPX|xHwZ0c)F0AeBsdw$;4nfPMvL|Cm)qf`Q>TR!(nW+**3R-oQ*!X0h#?z2 zrDHemYxMpqPk@?b=xO5XgEGJ7|Kc~i&p&6W(JUN4u6t4be%$~EK={9lebVCVRdc)p z*4yG!-KECEW4k>5ieZK&g_!Q_BK|yh1SehC#ecQobwzT2+wfl3M1Dw5>u;{rH~9fd zK)kEt%XQ&ks*7e{`@~00cLNXZVXCWVY7akOd+W3(IntXf;Jvfz`DkgNusM4&Yv0lI zO^Wed%F^dOCM9lFF@0r+>ih>!wuq-rA zrx8Tt%gX|0eSYUaa4rWF-z(Z=r04t*X;8DPO-bdw-zIOvb{ABsI_G0k)p#R)emjLf z1Kcdc<_J1-ea_ygt>i}lgt0rpVlHYx9UlO(pOqVRR`WsLuj6_=j|lsJ$6KxQ>sRz7 zO@3Qz?!&y_!QFGr>^GaidtM9h{6_xW{k}y7h_Az6nLf9Li7t)gz5=~;a(&=ezn603 zwasB9rzqPUyTJnORK^2}dWXtPmi)^HH=LH=)%TRYh`r!FGTDY(>$j;=-`6_({(s?i zLWl#Wape73EFJQ0Bds7ZS{z%W7qN-s0`cNwyR30?V<7mAiw%ZIh%6S(69@wR z#sJiMhmDiA19tM{WvwEoaNQm_4qI+j=1(`rhR*x9p;3MCs!=thlk1ak+iXi55}Ei< zGTYf*|LQv|gIGSxZvM3X1)Fp5kWfFR|EXbPFP-xKvN!zG31D z=LKC7pBfr2E2koV&IkOO9eWE0BwF8XfQhGP);Zz-S5KXR&~wY7(abL67-7G@*hpsE z7fc71^P!&@93rtEx2IKZ)$YuC9qf0%$U(q*uqVIOujR-!U@`la>m_C(ay(j0`xP6| zS4r#)FEC2AT|E~L3D!3dcV=5iR3OT4^lGOe>O_YsizM*Mf?S$;Y6o!MgVFK;;@zvC z;&V(7lLVNT!SZWUy@+qeFPwd{oj;VyuENQ20lqpI8cZRT5wPS^xoB8UP5LW^*T(&$ ze@x+-IZv~EAARDPJMRBL&hQXw)LAJda%h+9M|%A9f0C44-X1Ub{(sK52>kdzj7H|) z{|`|@-O>mA)(iCoI)cE;F3tfsNFT1I>Q;1e+X29}4y-uB-as(WMDe9ohMnn4S2GFa zaq%u>)tD&xm#cJ6NP>H&fsEKEB&;5@EdV8Z+r@s#g?G7z4%FID9xuHw)vGXh^N-Dd ztOJSh)!ld1Gr&dI+0EjZ%!TJC+NC89UQo=Xmgz}t1mE#onc(~8@sT>abg%DSPCKqG zALPiTx^f(DW7WPvEY^Z{tfyo#AtSLbOkfQF*Ku3e`oa`V-kxM`0NJQHm!0y#?x;HX zo;;(2+@0jXFbBF9NGib2XV55n4~29mb|++uX1g^)Z+mL6X+Rl7ByA9z>N}$I=LvMm zANNiE1_5dQgn$}BzaXH{&j^_KI|73L0Rl4N5U}Fv9|%bG69S5k{(*qgKOvy%ZwP4l z8v@!Ka$hWEdNcydRYL4wuHz8UT;WhMc5InHtoTafkl&Nd)hY+)Pi+r8rSRSE=||Cj z-R%}3t{W0Z2q=K*`$)kIooupyl8u5&@te%LZdL?7lfz$>hr4VoykMj0KI; zN_~mxT`|lC3Bb(OG%1R1#c^D3jC&un{pMm$*ug>WbQB4XU_Vr(%FZNO4ozDS2HFeD zjy+qBM#@#xqtznC8%7d>_{nF%Km(9#86e#z-GA?bqFrH9Ab<>qR74IacByS$;!QBAV-?8s=}xx!XK_bI@8bQI;+2VI_{p{r1nvUvfX^#j0f7kN%EP4A{N;jp6DT97e@R54_HlbkU8&VH6*nS-_&rXQrF8 zzsn)_oL?nFp6D+)^M&B$xvfy$peu=*RC^Ld6jD$urhngd=0i1?tM6>7=SsY>+54!i z?SeIfdN&a!LY;<6(O@9Idm?&$2GZbM|CM9Bzu*%jFmPnwk_3wCxt%AQ{ZPD40TL(V zh?VG9Q|QZ>5uSU^x~bQnH?#kBredajxLQwchm^g+sjJ>`iZkoe&ga%v%fmSK^9Pn-v3nIK{CEj^bom)JJ^7?G_w|Y`Q2y7l%^r!b(&H^%k=^tqt zzb_T4)1Ml*PhYM>UPRL!5WpjF?F7Ll7nN07JNxn;nJ7KxnhsxYDiL>Y2-zm*J+O4F zep64tyZe>*2oI#kn@XN$FG-VFO}T(^H`Nd?)Fs`9bK zA>F$h3u}zOgR^VnpBV6>KwV-OdLiGWugE`<}Bv+>81E0;O?1&>;2Sl z7-8oPr?NX`MhMw9i!*5O>h&~YW0O__uC?oBga>v9+tpv>+O!z=w;jF2Q8m=YY_eH( z(5cQ)=FDqP8|pfr5`#g1R1pi{2CxDgxKDTm91?tI46frCPzQd)Qy#H;i_B2kQ)iRW z)5>b0Y)YD2ivzE#{Q;e@JL>D9tRZ4p41xG53)PrLeuKO23-Ohv?byhy-q7HMc~w%*XRwIw@{i>KVGc;(V0&J z{5XpAdmQa>2aNFOYdCS>3~|x5>o8X^%POZvx-~9mph4(*(i_MAW&}EaP+v&E(9ST4^(Ud99j-s57r}ZORZjQBzE?DeLC?;<^uymx$L;us zOmfi%i*7D2Lh^?1W3OebK@Fkf&%N%QUps23#m@0-kWr{fmfOtE4jBr1^B}2g$8(k0 z^(G~zV29~5vyI94o@Mw2&&MSMxE;T-5WWYgg2l>x0k=8=a$`Bn5$9G^_cmZ#sw3YX zcFMJ3Bse`kpRpbAUfcuVqOKpi4IBwa-Dz)-mXW|ApuNr!xtKF0Tw!faX=%>W0|L_# zHVW!fR<<8|@{OSlj3yO_v-`!B{VU&%4JZUZ0 z^Jp>{LBDiAk7~p*fv7FCJ2EX~H|@B%8%OvC_e~$Ixu6`X&+g-q z9u^Gu_Z1)L*&9grdkVwvk!4cyQVgAh6O*+* zX&v!$x&8;}-ljTE81u2(jGvHTm+@lFXZ`1(Tb~0ARgkEb_SFNw-ReiSEjxtRr&3QI zo4hNn@1AuK-(F~`x{&iW>PGEv92KSWu@$j0B(0CkF>a5D&i+50H{#BW~&Ns*RHo~zvWqA}lYJnx#cy_s$MX6A^ zW9n00w+~Q=poZ0Lk|3fXZ&Q;tG&46q^_D-$bnQUpJxyRJ@if)d6uTkaKz%fYgUA!U zYaUe(m@;D|x^~`0)sGsTFrnE_V|0q*ZlVMJycns#gP+MtF!7%RRhKE#_h8Ov+Jq3E zEp@!0qNRwVjKm8nCbe=aGMr#YwW7y<+jT5?mno09Pw4BK`>sGFRQOhgyY0Jdxpd73 z%Q}36{)d>W6TREWR%6q4>gOZ%4?hS|veDYt5W)=}xjDy48l6TxZSrv)a<=F|RBcQ(1UBIa>;Olp|YH$ zu8pd>wgX)43vFY**|6Sm!Byg1$8n#D8TTmzyFmC$mqj-T;gYw=b6(}f#14n8Jslkc zo0kGx4=$&?ci^YWIBh1`dEr~-4sbVsglYAl@+QmU?5l? z;l$16T@+$1@Par+^V2gF0v{H;py%lA-$T84(`OU*&yDKF5Z6x!vqD=?T@X-^{hU?< z#v)o&E@PvcLoWH1Rj{Q;@{EisX#Lp(AAZIX4MA-10*x$Y9hEH_OSqV$y{uuU7DVh< zB7ae|cR?P8{u-WRp}9P8`_#w19vZWK0=Vd7`}6jS+8H+nX#li*Zn=Q@A#XV(P`;kw zA!nXoIkZ7IszSwz+jFIwpG?cCR-I!d@tT^DR+8S;bA5@mg9hxW)kA*{;J+Z&F#qqz zGuM=ri5(@7j}O49QY)?n?iEE_D8PaD?mb^X2uiL90HOY}qcF?k>0a4G$*iefLit{f zl9;L{vFf{zo-2Q*bbWD=#5v>L3h2bvfV!UQoxw{|GL?}af_&)f%TtZhiu&G9H)M3R zL}~k{`~*BDfrK|=s@T=!201f?zq^5(T5n~30R3~lM{Bwiu92VT$&|@~b5z9gzv{Tt zxZRQVQ+RoYDd8`>V<4|H>bu==Lg_L&mC5(raa=7QN9E_9M;`_rWS)DJ+m|VA>;gj4 zbjSdw5ks#v9t*yD06yGfC?70+U&2Tw_CS*Xr{ich98NH34OLu!BuBzl z=0<}WZM?ijR%n4#xW%uH`uCDdS~e`eZxK|Jj-VSgo!;8iv@O^6SJSS#kZ`62Aa6T9 zGVb15CCd#a7`rq$*{vCdY&R~mg_v{QZTcvxL1d72koih<@@?LIZp8X)p)YI<@}p3qslK+p*Jw02Mo~55k%(?moH<0w)OW{c(!+s3GDItC^~^8?6@x9 z1ZcSeK~yXg2`w=8O+_I>*7CVtI3ZK1ikz0g_3PxA!^Y1Ku-g7O{R~dW<>NoX_LZJw}Cb6&_63>Sgh1F`+H4;!At{8+#EhfZ0jh>?`NP z^)9{_XX(Y;_xs%JPe=&6mrh~963@>zmk)LWZj=?cePZeHius-}v-m53A*RnVQP}kb zKu90+rjzjAdU*%obX{L#FrSDJa7^v=0O;2#?`a?axbqhV^m|SFxc*4zY0}73K(OPz z3HtnY{OFCB^y{Q(s<>fqR;2sSoCvr)cw93XB^Zor(MEo`3NP|?3aB6YR7VmND`#s? zod19nEz*jYp3^*Aq+O_;#%bnO@$gfMGs)T3y@o3F5J0$;kAAr7U#Igl-!=pSWbRq# z(pK9gTb)fO?k`kW;b*v9J+4fgClk~-*?PD6#!fG+0|DNRt+VVx?uIZ3q_dT*FZ5{) z23~IU-irMIyWSp{dj8hu!HV~XFUU+R)i!4PT=fsG`0uPOb7xmqUBs$L_AGrGUOax# zc|e!AFnN27!0sw+tJ4+Ptu%8$Le%B2=3h^w(tx7VHim#A5!-9i+! zGUO3#AJDE`Zy#KU#2o8=yi}P*l7F_zwyx}BIy7@Nru^~Z$7;zO)fDoEhp&k(X~di6 zK0Y*#%%q(3(~>G2CMTI4n&|7F@8kz>e&wL7CHRLmM#3)Mbo$YC;_N9G8%1#(tDu%* zSD8qD+>!k-y{&@B?T%CKcN*CE<0Eacar&WGo35YhvQUav zkJw{j@2B8Q*#DCLgMajaI!fv$Tw-Gw6-X57qYIlfPn(2RWw~`z*fd?84rpw`(TGz| zSSAASw{plXK2exv7g$9rvO>qgd19e0$Y2fFEb~mvsmYCS4n1rhDgZyvy)XCp8t+{d z_dNuS+~RJunuSZ*alr6Ncv6}bQzLgE45Pb{44fY2h53`~7efNt3@vU$1Fz@MfCVFXE&eh4WJurHphb8Dm1n0wngP2@GF>727ZvGo{P-}K!Ud^seuH0dPcfMYwUZyL z*R~!^!O!)<5@Fw*V4Vz5?otF$KN$$al^YIFU>IaLG1$8^SPoikZnRX)AeT)Fexs&f}b~%*#gJq$$W`?;Q*`AL1|gGj9a_ zCcP1eNnJS@R_n@rDMn91x5t8_LlOAXCc;G$vvh3Z$qz zV`ez=D-nw z-b^w4NPg2wJA_1T=Q#eUk!0b+3BBGAIC86ef%>#AK?k(d=uyUFfrO|a64Ue#A2k1y zn7%BplMKO>ET1ZcJk6|LxhIladE{;e5vMJQwGMY|67V#3huLI)MON|`v+J8wdPQpn z70H}Z*UmiCX`Cg~_n9O0vHb^-EraRXoB3<=|M{hq_J9KUL$-~c{%aszLbD*MTSZ7H zAy1Qz2yQ3ku~Um}bx{apt#`4LToddMWphB^Lm0N2fx`Ixq-i4rCzZGiNuQW&hKP-C zJc|uPUqz}mioZlcr1sYl9HgA2aH!NIx+f02M`D$v%x{Q-YJluNDwdxyf*>^Ufk6gK zr95f?Je>>HD^WNWP(C+{$RMgB>&OJfHkoFE_CRosKtjIcKu0lOYb}p%jm3=6X>YrH zjA*Z5ZggDaLK^o~2YaP^CjXG!miv%6EFo2Ic19bSA=nyIhtJngTdCKp*vGu(SN*!hjg{(Yy&XsR8;kbo$uP&8ow zc(4^9PY$r!Ui#AzO%@3HcRB;?;WyYNN9@0(o{Q*^Z1+0T;wXPe#nXtNX73or`y z9EE1}Ch+|GF_JO;5L9G1gNUlVk=e`(&^Z$#an`RguM$o?l{B9ZWtB>YH7|{izJXSo zfDhd_r(Z-e6mv(*qV`qNd&6PP9_HPl?9A<0<+n2Xp^?nq_&2FmhFgQ)9Zq8v9z2PE z4Lm$t-%v8U(?tTqHJS0nIGo@e??Dsz;95FK^0U`VB=`Ke=!dUB=SsYkZBW^|T=r*uKvQ4W^M~GM~ z;Ge>2>JJg>{qLT_IlHIzEAbP@VM^8?;sWxo0XnImGwJ)9S@&p*S@!b!df?WK+;Mu` z{(bJ{Y@@o7t#jXd2zM4ad@{=S%>~Mm) z&z~w8Vg?^>?wc7*H?~b@dvb^8p%Up`rp5~fM(n0h0_q0~5C^M1qWs*z;5+55&T32-hDDg6@2&E?TI zva)iQ{w~LAn;nAhLHn-i+gOoxyKi+y4C0qxMXf6sQ?jlXQ;NpSUDf7x+E5q?_T6X! z0MP3mPKal*&(UO1fEi;JhnLBHorGW!@t~}`bycjE8stF8TFa?FnbhLItv?P@2(B~# zeId`tBRKJeEP0{?SID5Ik~(YU1Tjngtw!Bq?TFK^%NNxjExgSL14%QRP6WXk?n`#QHMUBhju zxN{?($em!=h+1WM$}1 zI|f^`R|QhJpLGr|KJ%#f()v`pX&Yk7fTgABGpaTAo9ZMx{OjCX`_0Ro$E7zB3bCeB z)Em-A&GF42%@Hzi!GHbg`==!beIFxFL0>mTzQf#SXV`14fgHsGkdb_rnsRP$3E~R2 zUJt(+00q#&_%rYw5c;}#2pqtv5K-7j9`|kHBGL{#o1CVhpk8oWAoBG~&} zrTPrb$$?`9*5a(sQHfqorL%4pLCGk2IRM$`)w5wZ>-J=gr3fP6;hhfw+{m)rh7W3I z;@Za5i{+KVc5dDt781zLnzB|CNO^nEzz=ZLc`f+6gYXl4o*{nV*OX!%;1Y1=XobK0 z%q4YDm0?E-wK`l!*%hzx-d;3rg$Zli3tz=>T`5mC^7#AqqCelBk8dwg8;c4hm!SLL z1WDmeQ=>=Ny7RP!9cNv7LjsJu8u;vR_CW9M!4-=gFzvxD1sCi?*8Ejb% z8N&8h-lXSRkC4e6*4we|w!@qQMm}?Vw08?u%M-gM+n)+C4y?gun_$d_mOdozTg$F9 z@2rBGMc5Ch+g1&}wioahN15E!E&W_QoqeTUR(YD_=P4&Gpa;kSVz~dc1+4w_zarL| z|6T5FaYs6#_;n^btogh4fB*^*>EiS@~1jQ1PHyc5IzC9Rpin7?O6{^m9P1E%$%!;=~H8H&z4$3 z^_2f1J;lU6{xL3RvaXMg4HAHyF8Bf_TBr6Bgizu3=Z9nOs$W(c7VK=Fud^>|a9&x< zFy7UE5$QjL6}{`ZK3PGq(0%S=^Jc^w>&du4J+VJUum8-O3D>odt;Qq_kV$qc$p=H=gm4C zla4jXos~G=Oy7Fm%|@-#>{OM;&&?PAS-iE=U`?mct6)>}p>K|n>WK`K&O_h6 za43dYy^72JR$aehT4wlkVz$5D`D5MnDVGPCo;?#y{uB_C#It`K?sily zk0%d&p7$GJ7Q38v=(ExN#Jg0?Bt~jux@CxYb@=JKeMK3}4tVX2)N*%>7qd5pD;E;D zYWP9)uUYDUHRW0x{M1bkd=q;{_@ejEwZtknVNnRw@Zuj@;4D^(Ou`C!K9h4);@ z?H8CFlL!DI0KMX>^fPRU0Y-o^VEw)Nej3OBlz@b)4rUhlu?>&^w%1q{%5U!V#gg5F z>QrTseT~m#7I5n=e0AHhFY8=i;i|^z{FltoBgNX0lc!R^(=tcD6nz2abWrMBQ@9S;Z+KJ8TA46@%C4B6jWqPcMSffwWPGc(h* zz+H)K3P$<4SEs-fAu9_KHu<-d4)@h=Y}&@c=&&0E@a<~TUCo@&K6_?M1BH(746fhU z`JCUXak!JlX6}aS4)=dO!Cv4o*_or2gY7^4t7_BwAFZ_?-ih=lxnCQRo|%%o9ifl2ggOV1i?6obeF9VIbz`cTF&nz;^@OqN|&U$AGXLz}8H zR1?@hi?^Ho?NRZN$pYpe1pf1h{?TT>GcsWp+0)! z8T5R)c;7gS> zg&m(2Q9&`X6*OG1Z?1=2{nX^#M?!(zQwTJj56Jd{YHW6qxC8u*A!f(Y=D}=!!WN+% zA*m}mmyxa(p?rS9Cc)sDU76^6{tcrsw?fusWS6jL6CN%RDa9qoBa@61?JsIvK4>mA zLz1MU9z(LEPQ&Gt+g|!?sRH3j?_G$}y#Cvp3V#~i`(sr+JCll9z-e7viO=p?fu49` zRqmsFpV4)Sx%o=|z`8-P`#o%fTrVWWgJCd|LXVRv?E9-P&g;o#dNxU%6@Zz_r}`%X z{hUm5-y5R5Tc2{q&@f=lv&fRq(&H!kYcGSwl&i|$aut7WBk&UCY(FCQCS!tAs5-y& zvcXgpgkz5HHhHt)Niub4Tvp`|M<-Y8B2w6@%6$s#>h&{o|NJ;q5Rka2kbhrk)*g6{ z|7)D0v1M{Tw?Hj|LPyAvE_ayX2}fKm8DW)9;N)}Vvio8-m_lv1K!8_MC>?YDd48;9 zyzsR@!7neU$YO3^(3h?12($o_)H%!Fq_Ejtis?{E=C+d)*K2R1)Ctlhw&UJ+);S1w zMn}5v&c_gcTHxuo%A0OA2W`#2b9SDKqZ~uuR7s7J3Q!|fygdS|?!rza;jezROO^PF zh_Hqz+)MD%sEdBkXYc2N-1TiKYpe`>u-zlef@c)Wk)k=4=P9Pg^(%a`! zJp~w!6?yf~8@*ieU(&xJD98y|LguTwzUdrjci!73fq^$ZEF|1H>&XJU@4jq!jO@z+ z0Z3p$8V$J=2wUpVWrAUxV8HJZu!u29?dsU}1z|BC^!d>UQY^x>r6^Wlw>5Cbz)bPD zlHqNcn`9w8O~Pg>3=k`=XaV3a zkD>)RBmxB=01lu1-PFCsDkzf_G!&L&KIai!mf!?1Lp=1ak=v~#o0mbYVJ zXsi^T_9*2H4ZvpPm%4g=y|qx#={RsJX{!eVBm7p~O;_%F(G6{oeXe?onHN+8`N5v< zjA}e>FJ2N2BcOO(uHj}<9-(nH@ajRtWnrp}#w++mGyn8Y{fXSQ+#=z?&p?Av6~e+geClT@TZ*#iHc~KG2vEA zm#{S{bp-$zcVdZtR|JFM$SU7GUZWQNRqXN^Q;2Fy+`C>!y+*z}n~vY&`(+wk)mu0% zsbS!!_l;A%mV0}uzyOjW8_)CeFZmVY2KP)LSUU9c<-=^LQLw`)sUjIyXDO{_DDv8q zrNXpg-*S=7n;Rh01n5jSA)KTeo32Az}#suIFvv@H3`UPtPunjxhH-1 zR{E$oV7lx})-t&U`-VZJ>eh0Vak&+HhlHOA47TYsR9&OaQ4~0EH}#Y4FS?|94GoSZ zt<66vNlwm5GirrXl4N3DNO|x^f(mk%NuTZZ!=JpZZ~#ew>-)50{|^gBIBK)U^)S-q z&QMZgz)&B9!pc-u~L?q%G0;PWOEkqgN<+3R!T~lb|H+Lv06N0D?8F{?5KV(1AT zs`5?ik6-2!r4gw2pNU^>_lM54pmJkx0+BlJsb|1?q#&C)`pn(koZ%cn?%Bu7E7KqV z>biw_;W2JL#3C*j7ld<~zgGx$5X^ znNzVx+t^;@<$fuu9foSeVddZvn9m37RT6%F=rL8|{`zwQBR`TePnM6&fuZN8FB}4P zfV!q6Wtp?rt5|$MUW~W&T zkfl0aipOU!dHPtDeRD05Kt6ma>2AKwvuJ&b#%KO|?p|+GBqHpX(`3@q3Z$j!sD4U| zLKpNc?GFbCK>s|?_RDlbnDd{iR4c^BG#|ioZ!q(C^fkSZ)9g&%Kn+I^wUbFE=~hkg zXaEBmq4CMeMR6t_oGrFjwKPEe1g`QZ@YWFY=BI(`krGmoE}aur>U(H^YZh5 z>FIFyX<^O^n|4nQ0zu2G77WjrJ+)Rs!g+f=4wpoXg z2`-7Z9tvl|P-KC#NnhbNa(!#AXY4+pA6%qE$(~YW=gH2G`TYKxuU)p@TE@dIs{6mu zxBD^>~z`34pDj zMqE_<+x?nu>~(fWib$hPTvM*&8q5_>%SPDMl}-D%f3)@0#4c{FvUMq(IPR2tNAH0I zj5F$a2V#yKni`iyDl1I=kTVRK@ktuZ?Vr!|UZqyA#tVxJB9~YDR)?6i@0eWH;d(zz zLUN~)P=wufB$J}>)J^3xi8yfDoh)&1-A0RW8hQLxrrO~hT)>_9PG&9%&}U9iZan9* zVtACxflcpoSOuL)6K(K`#gE4(q`P*0)zhfHYWrtjGT|~{%86Z4UTc)3z31k98$s2DQ7-N)5XzWygNPPeYo!a=<; zQjsRn@kwxatkZR*`GcjkymB{sRTcL`n@MQkHskHMyB$zv=ndp@=y6VYcfkXR#z!yq zw+So_`uwtv4nF~p_s=nhB|L@$2+rgA&Zi!ACHeW4neF2|DQy>ByzOzkU$Oym zM>`Z-7ujL+ZWBmXsdrp%e#+$2hNKC3k ziEu4ET$V)i1jQ%s&QmKTEi&*cwR?iVW8>maWenOX#Zj`pQdG=!_oijAraXRF z`S)NF9vl48(jdmdmgFDj(Kl889s%PY{+x2t7|0E@?`c!b?iEcom4p_O)m9ru!? zc=Wei@1+P8hw#r+Kl>E`0Lew5z5L1B8Wm`qy9-jGguk1HA~PBV{+PN5p{TWaSGoiE z@Y*upz9>{n=5nLD>s3>v9(RcO#Q{-MtQpbnceeX=br_T_6RXUlti8zx0#~#9esD4y zDKT?$$K36kj;&EE2r>4Vmad#zmDz z`AHj)>Eq4*ILbWNOM0xe5Os;i;7u+!NlVnE`AU$fOek`Ft0#@xDSctGy!o)h@PZPL ze6lzM!RrIbPghZoksx=`so&h%&e3WUlkRbLYAHr<+ZZ12fMT&%F!8n~!!k zvDl}S;@rJ={nSm>P2-~EPY`GhgTo*oO>OspV~*X*;kqsi9b%KCQHmADHP;?M z-yIGnyHl(eL}OB%=LooS>W6>=ddx$wXmRF;vK5L^`g1dG=3AT_BUGY9>%}b9IeIsj zV;)WxYeq><)#dujQa!bdXa96RFZg_ob0u2!j?R+D?R>p;%{zV`ixGxnk8;Dz>DHgx zXdyRE(j=Ib*fT`WLgO-MIzNmDU`Cg$C%7M$EW7$4C$0bu6#L4#{}a?&`Ip0{e?+aH zN3kD#Zh|9gOe&-vJ)x_B0pw0C_2`L&{0MBBZFxj*B%hQ~cf<}H{*eIaKzz~X>VyL) zTuco%Vw}yATpzd6DejjGNMkvD-0Q=XBK*?7hCL%NphJwcH^?$XM}gV|i2JIZ9S=_P zeBo>Cn?pT9dn+2P8xQVLp9C)3J@+wS$H5A_UrKiJv1Ga; zAo@jQ#?QOiNC0C%{V!o(yt~-p^Gn!U{Q3gnBJ6P(0xm0&d8h|FR^wcVd=cc zyrKD2IqkX|OE;pbXgC|)W$SW;>r!hD7^jnMaDRg@N3{Sy2v=aUy+G-iJp*Fn>R5%Q z)(+l2bH%nL!%Sng+gq17EPp-RY;)s4_t zkQL5twYkwx#->3~nrzjyf`ws#3Owbo5!>}Fi=+7wqU@j@>>=2doJnUzFKRtkd#sIk z&I5QqFsyyN6P!Wazer=uGbaR4^WABIA$O>|6FV~Q)k1DHyKK*XMTg^(~0f?1o3>{|{a*;|-v$$JSvR3gJ4B>2m zGRs@5c=azd)=`oc{a7;Ey{YRFhLD-{gkSNMJx#(($JmdY<;heu1ii;H(?L)%L`u;G zdqdwzi{>YDQPK-6d-0b;wf(1Nk-@1)+*8zq-<|9FanA;SX|4$s45f>?akX5HD|(Y7 zi6%yt)li3MKAm%E<*Y&#lMr)ZU*g z)vI&HBX+!cHy7j&ZPbb$y0^>+0zRL(M%GsuB=a&LnLH<*^}BH#Z+spaTLuAMP$R}h zuP;`n(9@vznGF>iwPELHdaccF%j0@GSN8V;54U`(=(|j2?+R`AG<#S|;KS&YBZLOKEuuL( z!U~i{KjS9E+|Hh`jH4Eo9Ep@Qk;RsgVK2E9cgK6eDoK?E^UD7yY5q$FX1}JM;g)ut z%W$dbBBjFE#<((JKp<;bIphTvp3iwG-^Mxiyo&VA(fPriG9)ZGP7T%u2~6Y;pFcwS=1*Wd4|FVJ&s;N3jw|NFlD& z+fpieIuNb(*ya}ETFMU75MgKSZxQ0mE<|NFCNtu0w3v7QY4{xkBmxD0j>~I^ou$^` z-v^VSXo#a7>b@8vfnRd0KS(>|7RvNCnG4*OsZ_&0=uZN|U-MLd_Bu!Ms{M7b{iPgJ zMr~~4!Mjg+0Cxa^KB8?mUZ%%0vG_BclsL6}Lnu zH)pV#g*CIj$@pO8+&|}dGaK!7FQ`wU==3<*=wNiQ%Ofq_X6xHmwS|{;?O^@mW2uKE zl!VvevKT6uH1j;FmzS2%h$f11?T+s5E}_Z35;3^tx^)$^7v5VT7zTvL_N4o_QE-&_Aus(A!$yp37ab1d1B`k(ok4j}h8{0!9PPYn9+44)dyzoK>JKn)DAw|KEiNz@>wTR3kmlV(3jlf9bZrHpn}{(&a-9^pmr_Q3 z-=t;0VVd_t+!38@3K+gbHGraZ8ath?9$Sn~0+CeMis^+rmlt=LxV3-C3ucL@mQeWx)Qc*i+yE!IvjJ|hK!swt#MkC zMC*`VS}5VKu_LV`xfU>GS9rmjhDgj`+ zh)+F6ayl>a5BuqrU(X5`#@rW6c|j_U|IElP(rID?-r|_HBzw-RD)|$$HKjAc3 zyYNn&gfDUpu|YkYDx-3{c6=|VJyS{HMU?#>PG_t1itL!(-Jl&q!OfH&mdwd1T zdUis)v)Wz^h4vtka+qc?5+X%2={*L{+yDYWz1#pJNvqsIOf|)&i)){|j7FqKvm`;} z;0at}Jn>K=XAuEt{S7^-S3(Hqq?rvf^~?(>jms66^>;kq(+2*m#IRoQfUG~?8O&X$ z-z_COL(rl6f4MVS-Fm%0z&Qt>)docrvk;L(HEftrb`@5GiR zg3ceGj?5d_uXVrRPOzg*E76Tpfu6cVH@Z8=sQQY>&cq#k1~r^Xgr4+GMv5*zIX_1` zE896NXqZd&e7+#JFbIVZBtf&;&LW_5Ei=&0*QY1^S@okl?wPv*{?@0i?|gyu?j_mw z)GwX1h_G;7?3Ut|*&5s9T~V`wuBiUK#f^iMB!Rh^y9Z|=eD%bqEKQ@1)D}?-bUdo8C+KQO+ z;47h<-{vzdKyH7Ogi|;;7ZcGsR5!(l$R>4S;SlZ>$)cf0^-zV{lK^m(wG@m^y1b~% zx74@|3b{5ifc-~;zy$CCUO@cs0c+OrNBx*smh+DwABHRAcUfs{wKW|5yaZG7`O~2k zDw)#E&q}-o=}=HgI%QcaEYGK$;iM3 zK?p`tApw3}vzbcMagozUFC3KXoC&h6335M4)w-{{Aw^U%kuMCEzfU+TU_0 z#PH=20np?zeB1I{!)^<^1Ba-O(Wi{b5@X@%Z`-<$=ErLu;CQ?Km#fD9@ATsWGodNY zIelDZ}^Su#4B&@*WB zhhA$bs~OwEd!|~b_P_33n+)goN$mcxZoB(;I&Ot&WIbMaW|r;6Rc!d3Fy<*u9mGmR zA_G@pNoo%O4M6h0DFbgE3l%}dM<0HG3hI}R6_CZ`Efsovu1}QyV=IUT;0I^`=8H&Z z{H+!ABNAkS&l#1!M}q$)2AgY*dn_<0lZ-dv)h{ecajfchC}_UiQ;G&oo9FD3062!XC1qq91XhnXIbB{2)pA86?Vwp5q)?z0;Z1@G;Z#dE+a-!Vv0G>!D=+ zq0okfIb(;W;*ES$KI}h)H3c9J1OtITwWYBy{Xr4$-(8aW5RLo-JGf9X4P-X=~7xgP;t1DA1GwG__7md5t;lq~`VH~+P9Se4PJGhS6PRRk?_M2f2bE{koZCgNwl^s@@ zhqNQt7#y}qCcEOcr_L}MzT}j~F~9uIat8yPtQm=xdeW#lOj@c3(;vQ{DAB>N{Y3XA zQnXm-T-xy4qt6rQuMb5|3*GMD>r$9*y>jE945R#5p038G;UpgqYhuoenJM3A$F=ddVz|ozmG%yL8z=$sXL_Pluh5%=`*T%u z>yLXK{yKFpHex3A>oav!s0RGwP80U62YTs`8nd+t9b)ILEKeHi-Hn zE+L9|)Ehn3PQ))lzM{(GE57`(W%Y8*5i$4SF^=SFCw>)g_8cnqQ_v&}UwN&>4(HH& zFrM*0j|CNhRj#~FUycRmLNOs~hGzB?V6thOGgJm`H$8uTTJ|0?_C0~l$<6mdAA+cd zuB=?#h8H=`tQsf#qZhYf$5Wu~RaN&wyEf8#tB3Xo5>+x!5`1fQ&z4ck>m0dm$maSNd#c^*R& z*Yf{=lc+DUELcZjGfz1?43%XE>1bo*PYOa4e$KLTTqo9lU3%v8^Nw%7OMmWp(Ibcc zQk}3hJ--5r=C&NXh4=Lt9LoRT&D?-cfpmyiAoeGvTH78Uj9`4>!P*jd4v>q~iWo1^ z$wp$cP}=TX(aE~kl^-;?J6u(wQV?P6Bgic!4m{3N8>+7y$nk^C)()jucUAg)!#^yJ zE0?SZB#%5@-yrF#h+q}EHMV;J2Prs6Ozx(yofYj!1VZ2@pVXV|w+ZGb*AHwLtl1N( zZ(rVKrfY8_X}zyI8j5Ice}1);+zIyJHFdu=S7QiZW8U8V5)z$71?7*$UbwZF8pdG~ zb`D*ZU+(Cf52g2XUJ3D}=FOMBQZB)MplhB z5`rFn!sO1_lcK8Cn9{b{Pajag(yp}t*VZph6- zXCh;+trDArh^&I9QmrA5jNxrLpuqypnoV0T8CDAA3yCLc-(-5cQ&u1!n z%X%HY1l;Ulp1T$mmUM!8jYla;ZgzVZ@z6UDKL2SJWQv#))H>+i1w#p0F&Ob4;x{`2f6%f46!0F0d+{ig0kdn>pz z^gNVUvd-|c@wqP!-$Gxk&y&&n405~m>l-Uj!j#y_6%)kn`bM@~`&nY70&E#hgyBQ>STU|cz`_$Rpbeb4ko z{r>lIUdfY(YO|?Nu^mgVS|oXMl@v$5@U-K{H~0Eb8pVPKFyVc&?cFNR)5YgT(fHyr ze3!iKs;CgWkJbL@7_jHPy6GRE$8;W$@Q=tFQqjiB)AKx=k`~{3{+7;uOVS^}z~x`I zpeJlvlOEhI=tMUo0klNERVLyLC|}FnRyJTNJ6_o4%h?-Kmj79I(YOdc$oJqI0B8^o z%o6=^$sgPgxc1=wCQXONoY`6Qe zz)-ULj6_fPqz&EuW{xd6n6<~M?Wo?L793_xJ$IWIHDN~P5f;W2r< z8gpMf%8M%Jwk81}3K=wO|GE;ACEoc=`y2$0CxB6Js^^BPv`KXH7>%`yxpbL!m=?}& z-)pocI63G0$dN1iC2d#Q7|>&%hD}5Z3=-HhE=SNW&vP&aev0lKVk4gh zD^P>qyYL1_L6QZH-On|oeNv*0UA_#QDapRDq`rAm5#D`+i#x-5fa^ht8RH=d6MT2@ zfpkzQo+^J=?+KT57-}s+lKlH?K)j|mb)|a@E}Xr~mBo03Tv>gEPCaB5UE)f^XZ1$+ug%?M4SY&%ei#ohuQgW7c7~9P z%9b|@NcYBIJh(CggRFIDkTK2>!XFF`1%p8j0B!9eHV)vxt~LA6ke7%XO7Y16ViQ!N zjo~Txvfg5h%a?k}7gpCF+qs%hg1bbO_&S4o_+!>Bhxf@x8Zt+ZTf7Yg8?bJkhx!4Y z9y^bAbeNn@#UAY|c7QThCO%*;++HqfyLZa=1k(G$T%_fPQA%kdh0~UjtF^?S zsO_vTVLSPVsQJFR-jJ{@lPi5d9Ud+1n?v(3T{*;X5G&q3LB!N+_S``Jt)~_q*z0<$ z!E9Qd@iF4{a+YxdJc|DtuEdeSXR@}E1t5gtay%AXB6Fc@H#Q_N(w0MLos!IPa^2gU zI|-7SbBi*=?%XUP;KRlIkeL+XEu9(15>bqo?>@_s8H`-o0zn%;j`o(<(b%WkqAaN^ z;X&&ge9*?NErwvEMY$gwv+F^X#hH0zH5jp%53cuaaH!PPk)s6QCrsshP{eUb-L%1h zMcyWM86*#vh&DBk?|n0OJw~*D5)X7Heg`Ca>fA@NgdfEDV%-F*qRdCv@Kah?vwdLo zY>bbc>nn61>LowJX<#rxWN>lA$QL{~XbJtnP?k1rnQNajUBvT#_ZNg;UtOPe`1)FG zqr4`OEb#+`2?LGDjUNDb_A`0E;cmdipRWyrC8~~TRk&GjVXbZv`wz2}2*iqVu>D`( zR}oyiI!h&9RNZJA3PL9S~%*s1`Ks;ES{FuZOfLMblqO#sxOzWwt1bP#!bcZuKK~E`y=jq z59XS6Kde-Ed;DiltAF%7{oNwkWS&}7G43Tba_Veol0cO2)eIWsTQa@SfN8Dy0-d?^ zHsB`(?U*Ne^!9F6Mx&<-%kj|^)A^|u2|SwM2-4DZSTr(~V+j3x>aw$P<~0q*+Ci^E`V;YT;{_ucFUi0BN&JS%>H5A9A)O@{6ntN zbNSBtZ=Nc=U{i{}AZE3;V^2Gt@LNW(3GGtDlF$8gK5|}Ee$is;$jiz})+70fiGpli zRaV;1Gzz#LbJUgIWp60A*YS2LyO(ysyWw0T3)3}Ru|O_9oPD-|rKU5B{ss5^OR6q( zPe77);hzqZ);Vy{%}MwM{1|Ti%7p&mZ8)6Iwxl6djQ>Tt|NOYd%kMjJKV7M=D0DZ= zJ=mTDH&C=FJaczshxl{X9dDbst!r6h#1a3f>l+5$z9}y+bVcs`ySg6tiUN~NJC^M@ zsQo%_E*L9qFN7A7~BGFhumrZ)2e_lj4r+U`vhlEFhbskRGIu7n*v?S(X?n=o{{tPI6@41U90e;jn$xzaO(jaF#Q zIo?i1ViJ4_Naj7jSz>Zqxe2|Jgo=Tfu@ckR za~D0For-N9G;y{ZLI7@Fwur$)HyZ-EpSpy`N}?ay&YWbEg^&F3J{u2@V=f&) zTzV^II6LtyuYKQ^%?yT#rpN+gL!01>Bjy+hbl|}vYrk0(oMz8eI#7~+ zQe;vY<2_zbs-?SKa`TyYaZwNp=eB#s$i$W04t8-+6_!Nd$C?rQZo}GF@~F8X>FW=E z%nh+%;sUENk&gTrSPA8x-}h{%p}*kCI&2(|Y5jDoEUWzM`~FO4B#i;1zYZOOqzEoO z#3vOH{+2Hf$`$&t*^2ln1MSppqrBb$MsCu+<&bMHV2o~tyZxLpsb+tMte?X5&ye-6 z!gV)SDqYO<7st^|Nu86+_Cm#)39?W|V>-bPt@)}2x@_9J)2)X0YLw@j^_L#IMjatTC**p0X3O1>OEM2}9Hk z=S=jQ2gYQl;faK2Rp-u>M&fwOWVe{>2fm%-`qyyb!<6=9*g2N?@`s}|$9HXA5eJV) zOh4a;aM4KoCn=Jy17|r%rQ>KQn8N062>s<`^S(#4_-y7<7opi1&KsntX~>?w*-5~4 zyPF=s$+BI=6A-jgS5r@OW{tWbxZlSIra0~cL2C1?cv!KOMl9M-HLc=32))b`?@;ZB z@@ONMLM4%Av}-Zw@yDSg+9zLdQtwddGVzZ!$aDND_b~tw zfPXXQpd@TP3MN01oP$o3Zkw!hwoCx*z7_nS4+B<9iVTSkD3Y0!>sa{*9@D9o2p%LQ zNn0fQt=kpl3icRkWmFz~D9T$&pQ)^q4}MT-k4B(s(CsR`64uAkEms^ZU&d7SScLm; z#@r$e^5r^?>>Ff`6R+iLyV5(=YAoNha2R`&;nV~n-mh%>qvPnSNsARKEV>3l?rV=CcKI8PAUTUn(M85Xe%dWC=zmuIBa9se7Qqi=1Hp-WKbNEE-EBS-9=tYBl7Jva>w~EM7Sai)=oI_4ACl2q!|E zN$O^MF?XUrzgnt&(2$-b?dbdRQ)?M9p_twnedV3(p80wcVwlI5zo_FLeeFQdIkgy4 zd?c%VFr(r42Z7Rue!U=2d#1IT>gQ|&;N8~UPCUr&^rBM5*-8i=Mw>>6KaTtL1wJSM zb}M- zee+~5Q1Wo!c8tXRlsTVAiS~H(@tx-GRaH%C*|AuRV*tjmDy(lUb*i5<%GU-g;YjoO zDN)F`+v?}>*lV$krMk$=j!985^o&W`O$~4W77_a#>3&o}8;Ze@S}jjsc6-uX?i=6b z4g9PWn8APDrKJCBmr~o@n~V0TH|B%WOje+ri3q0PlI&LFZ_1lUD`tO*7mqTqKjqeM4gemvm{6>X8jtli)qRX!t3qm?&Mfcs}FW zR)KsZH%WT!@kP8yu%CV9wtM*e*<7lN_WL&n)qI_i=jWA_2jSEk#w|?k*H@K*m988! z(}TrU=+6YuzWQg^eEzAX6XsR`YS__F^Gi#Tz9UBrn`fb9GQkFHu=|A9v=-|;9I zYln8?>h1yxk6iU<*#1zWg`Q)Ic>bsDh{k6f-?t-{w}19KC^@7`#DI~gs(rga2BFYT zYxB+4=7De=nmJ{kAtH-zYpTElf+n=Z=b8oRlBfB#^sjxmCD{e_Kib=sS_Tgtf;Vt2dzZ0pfi!j z+^y(l)#grkx*Dr!ZabnVdvo0~K~+R4?pKW>jVwU-yGC&*+&aZ9zs@T4YP6D#l2@9S zjgoyOZDJhc1>XJ4har=CD7lrI*86YeR)b6Yaq=fu^5W&UHYJ}(dzVP2p|i5)XF>2K zi!mGnHlKzba->L>MSRC2;@{%y_Klw$R!IRv;2yyKANOJZ7c2@wA+}2`ogpZC!rJ(C zsWp=xipLN^&e@xTG1O?cS{TLfRGwq(I%4HjBsH(mOeWnyemFHpbsKC}C;xRm--6L= zgW8YX+3js1kM=B>^c#rIPfzwjcSuBG?;7nRr#EB?oY z52zm%{=+Y0=YMAzX^M0`=u}k~Z+FFa z_ln#-A>Aqwq7xj+w_u3T(?{Am<%Wh($XQe1R4`s$dqndy+T^9!b{LMqruD2&Ecms#@d*9c?bvdWZr`y`c=DI7?^ zjO6NV)Z`be)cbq&m_q{lRf6guE`6fvU>|dm3RSmU^j-8CR#KSrfOX6vK>9wFW?@P{ z^%X-CUcz^|$nPnNw-2U7SMkcqTH?_06C zN%$LeP_mVP=6eS`lx#(|qtmYx#4+}Z^=IPmn5YXi>9Rk=S{mkiVKJ`Ev)&t(A?PtN zD0h-yIp5^z+k9&;pemBu3!^L3uHiFYN0?mm!{Zdb#qKz6WYa;xmmDx;*0!Fq$*tQI z%JmK5()UrnHd~wT)UwRg8jOCHr>jO^y9sIoV9#C~F2#<(&wuG z;SG*Fv~rda1|2UMe^Y6E5bHm+f!xa&doGoTmmO|tdLDbQ+mu0vl<|!7xqmnlPSD%V zcoX`~^9+Z{LI{tB7TQjjN)v!zYR%+C!s|I7L`24EY>%ZiGau6Pq3?hD46Dg1SZgrU z^a<5osI}a=<5iH%@xjJ3Sp(wQp43Zd0q^BH%9Y||>_x7JWIUX*qj+M{YxkEHlqXuj zX1?#6+Rc1kr4P#v=eJ5XbCS^mOP*!iHsH**{2XGNJO1;e11buf|J{n}_82Bws?%VB zi`z^G%`?ISEWa}^*$-sPuQCL5R+98~v!!yxzl)cgeWsNrshai*MWaO;JPWQh5AA4+ zG;8dw5f45@Rza#E42Js)k~P{Fi7fB#wYLr8I>^!EzUT{vJQox^*|e$#WP7hD@+X5cf4YN{E4@}N0nJ51Glh-h~}L$IFo<$xRXaEbNPs`2kuoV=(5l?aCw822NQS?iC%l>jf4EhmZ3O>NQWOz0 zYgbZ8HMs6GiG`1u3*)?GPXg4uNgrX)h$%6tw2h(u13xzd)PWm-_aBG&e6Z1&R8dh9 za?bcW=9f3_nsvs|3(FItoNXi8fl~u7!8Jv|dM?#V`+ba591lmE*k1ldWG4NkFKZuG zI7Xr_`5xYKd1#>!gsX;FyGnY0vJm4U@pT*wgG3$S&7C**ra_fPJGRDY2_L(_^&`k# z))7;~2IAKzlH1#{GG$i7_=k}x7l}{QZ|>No(YYsuk3psO;<{Q5q^cu}Gt-vZhjY&n zT5sPnZ9}}YVF9&FP9u<>u?r&voe;q7T|#dINo0Was8exNWJY;Y4H7T zidc^fqn{z-{&b7~aN&9S`G?tdJWNGttM2CfOex+ z{+iJ@Rgp}1yqL~IL<%w1w;2g(zfkv4^UoJs(SBi666G*iyDpSy(*tft0*KXKIZaVP zUz6H{jj0@QS>sgxS?!W;kY0^$ox!MrZn?;g<2wEx_tbMMR=?4M%rPIT^064&OBY+*&*1V z|7pBJ0r3CJwE7QJ!`zYjC0>cn&Hu_#ZXpQZIi@Jnf*aLI^g@uqxx4}G^+Tovmtt(a%`e+9qd&Hg3$dnK||FRB|z0x7urHt&_WxOa*`T6|k#^1!!gB?!A zUhZBL2gM}U^@f^0$VHRv2rK%+Ix40(%&BW}mU1A$g?3fFYF|Cp@oU59*gtDF$ghI2 z>8QoBn)=~w&|_puO7%hQy9OL@WSj6~&o8Q3MBxb)Ad>mHC@32AHk2iXLSp(|>@AU_f>d5xrgO(xk`2*1U(n&$;d0CcWJ?^t?uAPj^idAi=)eQ^qkZEK1G z&j_L*MxU*0w^Opf9>~+{tDQOndAQKiW2u9_jxr6rUW(0Cg?pt(evYs4filSwo%1;B znHu})I&7>H7S4m=3b5Pq&AY|A+;c^oSfytQG`tg=lh)DtWg-RO(pe5YZzI8{Oa?;{xU*QQF*tG90E zOC^EtpqFnw+Ubpb>$Zqm*sDl}^`__td<;6GG2ev)8s@=8!tTq?Fkm$=w1X01&R-=# ziObJ{EzRN##|&hN=uqM$rEVMGB)tkOg|YtR;NT2M0(9SpZwG(95&q}I-){um6!24e z*ocaDr;BL!-0cacW-@hV^CW@y#q!E(sVefb^^+5;Bw(7ctSQ7m{my2NH)1nHz{Ubr zw0pyEvOHXsgZf9OMnQg#TQKj=_`^HD73vIX-;$t?yoUYv9gV!B*U6Rfq2pe$ZuoLF?v31(V*7aa+bg zShS+=uQlx}kCTu@$fKH+lc7i0KhYR*ARdtT`wLI{>xHMGMxs}1n)Q7rvu`z66i_zF zEr?%rIn+o?K*L;W_;LQLz6{Ago71|l!J4ArU@qfk4U0GAe7WgLgSGaPzryD|QN@Y~ z9eXlZ1-Jn@x4hLEbg$HMb?Q{Af!Cj*U9TR_R za+N~F(7n|1Byv!cv9|tYz8rCyRXV=+DG%sTzL% zFy8KWJTv>!vtzN|nBLUygz?Mz$OG1UjJ}5twVNpu%Aj%#yQD)jmd z-0x&|L5ez^a}4!dZ3T_C)N_;yTX=T`jc|@-HG-#Y2Ie4c@K7B~lq}COPUgT9c8L%O zUz7b#fwTn$783CtpROzEnXFL*rtTm9WJfM~K|K}z;ml|keynvBqB}*^3$b)4kp8mV zEeLtlqN$H6B$4p$12YPeq+qUXk;`?o24vZYpM@o1Q*Zii^CUo~$Q?%EA`~ z2;Mp3qMF_&5qC0bbqTC1pUeYpySIK_eUoF@zLYPla&%?vJ4g@JRW6o+j*-<*oW~ct zm~W8n&t6cG?l_kPG_%mE*bd;m{6dtrz{c9#XCTNF!r$Zyw51KZFU;2Nk4+Sq=CrxJeE4ahD-!-U ztybt>rSFOQ0L4Q;dhq;5WEz$kOpi!0SEqf>EiwR~(Kg8(OS3}~L-l$!rYEt@%mNz5 zrM$d5cYOb;0BN>@OR7$Ujq<1wK9t0`X)ytWUKVs#iKCqF>_$^XiqRzy5xr%*q#%tK zpPLKZ`<`UYBxE#37|aROc%4%wdhx1d$95n5i_r)Q%!F&RP{v%vM88+ZD?zCK`OU{S_GQNlv})`r5gTkUu>7I}L)m`eMbR*V^m2Pr zZuk_uGI&NwzW(kd>o$H8H+9uwn&>eSA>+KV*hSNdT-{#9QHTv=)jqW*fXDhnk#i@j zDlaxdMRAbLA6Lr;#&fhm`vxO1Tho*>B-e;u9(H7$yPkj38H@e2h+)0Ga8`ERe5qe; z3;9I+hU;8c8iB*-mP6;QNqDbRA_eX3ez4+%5zk9_0 z_BOAfIlt3M z?PM8v1=-k~?D)=~Qyd4-XUpMi%n_qaUOC8$85x4nHil(&9OGoW*`{hzV^HbXn;k z5dFA}F;m*i^#K7%-3@?QuN9d>PtGhNeEAw-Iew(<`yk-X8QBtykAh}71m_V)4!t#M z#_w`6VuY_D7ZGX8UO&U^k=VppAl3GIGs|D(w;HZ*8TW5$xEpK%TK|h_^?&syY;esw z|9iJ3+m0mI4;?+K?Eho#EyJSR`!3KMh7Ku_k`R$n1OWj-7*IMCq`MmlK_rwRq#LBW z8+GvWNI)auV8WDz;~tCMiw8vd}LqP7Lq@{{L|F`wl(`txCW@DF^bv zg#@@!QX4xAWyjp}ip$u|L?${H$>MB7#R(PW43UN1s**Dl;Eq}^m^F`kV zRk6Q|0$x7yb?GZq4+rBIk$MG>cyCi0-NL_Bi?m$upa|>I4e2SbdQT=f1Y9vezlTgf!Ma z%VL85W?%-Qdh`GGKG*|kbsMp7;F}vB_q$QYtK#>_awS`Vnosqb+~M88E8j|qaWX%F z&-F^=g$s9x8LT>7N!4uB0%GpLfC!dXjA0l!T%PD!EH6EnNiDcEO(}7>g0Z~w{a_fa ztXZ!OYuiP2a4Kbd<81K&wlqKXUouoKBmrIfS$&~4yNH;7c6GdR&*#SpWhpNz1Yz&- zo9wPY#;ds)3{OnEVsDV+Ik|E(D}*DStQFr;T>kKqUkS-Xd!`T5?ii;c!C zbJ}wW00YGzzI5%Zss(Z3GM=G%7N>xZTm<*=Zs&)4A9>`E3#eaH_pP}AANji6N&k}R z4jgT}vy-UR&oykaN8Luk9kdL;_xxCD>@C44JBM}c*0g>)$ni_>NG~$klg+rS$u&>U z&#c!VFe_Qv`Zit+{^Y8VGgTegU-+GyMK%1cF!8j{C?(ZQKS~9{dVwi;=5bdOqmxB5##rt3xtoF!C({*AZ4*? zP`A&3`q5t{&CdVG2>4=1h}_WJ&FTrp*Nl2ux0@y$%742M!=7kCECD!UcqOgOAm;@R zGBP;^nM(OzL3{Zn{Sd*X4DY z8c-*7$@;vMex)oGe5XCeb{qyd{cvE+Z{G&Cz^!+2KzRWCS>!|M{&6v-Leb;xD zESHw&EJU!<>Ls%Xgl~^JV|A3%g0A#iLN5uz{%Lt!6u1W{|D(No{PV>xU(A0j`itCn zr-bAPTp>ve8SeRak=K+x&JM~<F3EOjJsK3=yJ8(MC*NIkF?4mQt;QgFrg45zo7o>CrneBANc z9I|qr6;V1F1f6R+zjbKW)og*dsB5EPJk;n0dF5){p}+*yF+k40=F4SBD__zb<1SEsB_E1zaavg@QFtMX^xU$ls4YP`7=FsHHY533Z$K?% z&2Ckt%AQTcY2BY;sMf{H?5QQ5sVdyMU{*d5IS zB}Z+tfj!_O=<-^QpK_rd$HzO^6{{UyTf%1*ZU>yhW!Dfg=ULAdD_lTFDf0D4L_)QW z>Jma+a~tD-@bsc<-zL@!3Xjuogb*HIGtUAkZnhPjTW^0o*|rP}TP&)}3EVQ%i+6n+ zm@`x#-IzCxZ*`tWr{{P+a7h^$ab(bePLR#M1TN)IfJwWV$9=r-%w(Vu*yF_cpp(yd zq7e~`X9nC5Q8oOPdX2Tpz8tHBi!fuFMBF2kF?lFSSEn2$yVPb9&39{u>D4XTe%LFO z%+H>p*Z>TO)r@t*C&PYac`e>BO2rG!Og3Rju7=leDZ#GkGb~wusNhq)Bj0vdD%+%E zUV?Z!?us<@^|Swa7u@HMuBIS~Uy<1PhG1c49yc})S1T5JZ*V3$839)Z)+ctZe#v?( zgOB%ULhK9g3EimdQ^OA_`?+ggwD~rE>R_X^vAl=1x}qB6U2#`-cd{@(`@4yPLsnM% zA|(ap7*6dWETLb_b$bs7mqK(tG-H!EZPXUbl*zf)JuE+n-RPH)09%lEaD@t*Q zD0bn*N2`x=ulXVa#6>HWJed+b1zFYgL3d@yE#9@KnIk@6ARBYFZ%xt5a$pS2e#>CI zTx4VRw#%0KWFws&+VMfwb{ajTElL<)fKd}bNrD?EcE+Jb_||Z;otr>kYSkL~@e##h z7kEvJnNn+}#J$F-RZll|hX@=%h>-1dnThp2qUOr)Pj89RFR5}9(ft-Xv<^gEg1fxd zv8nr;z8rn0jdpvUl=P{Y5hCKcJpMb%N6?9v*!sE8o7YyUY`|7&U+8&e8x7E132sNE z8N^Xtq<7gLHCJ>aWOC`}EU8uUqoHIMS98M<_8Yc>zI5r~yeiA81OeSz$_Tts9iMLc zwJspS2UNm}#HELeQb_2iG>P|}O-d$7(F-yp;gUC1Orl7xbS;R9-#E#>C$++#bB`Ry z1OVivU2{aCH$IAz89$g=h(AM$iVcPxtNCPQ}g?-)paa@v)-?B zIx0xUarq77%tF1O)PH?hfDRSQm{3=!JTBfTUu^VeN4{)A1Tx}tS*}e`k>x#174>m$ z(=5;K&yv0#RyiD1(3P#2b&ud=ApCQY1_b@sr>f|4vR>JEzXKdg^mW#1*y}NfK*#RvXn8p8<4wKfTRT`qCSI7ADP7XF zug*^aI<;Io-hxB=ok?lWk2AjpZjQC;BeLO>Z;$s#6%_MqCBLBUQ#d{;gSYhJA0HEC ze)vx0?BstT@^okUyZPg2gwhu3zCc42!pgTmyLz#K?^W@Y|=vg`2#N9)_J$HT9a1ln#ISzJK?#wIoL0msY7+-PpjJuqWIR zI(AX@f9Xam11|vQ-y+M0h)~kz&tg!{pPaEf--asSVc0)0?1}k)-)x~LQ9w#$qKiAP zFNVwqe);Y0Ff>aV&HkcGbiQ(sbe2ep6}Odgp#}+7mD}wi^+0I|$29?mFY0-B6a3Ci z*Ne5|h&tg0r$g%sVUEjVCqDhtV$i3SiR1z`44vg(jhCxZzY5UP`hzKaE~71kaV< z<*@%AS%<6(f4LH@k96dQ}B53IYYLm>Gk{UzUh&q#SyB*4nmkOJ{c#>+?#%D+xLWw zQ%Aa3RK#}LYvyCHx^(V~sVT-;AO5|TkaGlj*YGqxaH%#L$qp-@Rl?bRmU-UTfW4?6 z1*tPW8IA1-%$;_q5>uZLNfDr$SCqs4>|ra6I!7u7q5<-i0$0_b^34DlA=fRmK%p@c zC4(bk%>|iXPp}&xpN0K5B{{sDJH*2R z-a1k~T%COTD-1uq#5_5O$p;gDk;E90_Fv#a(N;3i&AW=r1@_1?;L%3sj(%}i^pr*XvwYjJl zgEE7u4z?qCv4T^)pVJ=LUdg-vYa7i5paL!Tdnxg$1_h0n|xh6_>u5G6*Hl z&6fdOS)GB{!bTWWkC>!h!bwl}j*A$&{HV0Y+}QeOrQ-mPh=l;z!c$yfTG`mqpeKj5U?jGrR=LdI8dZpQJvc*BoHk?R(7o#<&5%1JxdM z4%2Ig^C)MZuKO>8gKr0i9d}lQ(>vb_*DHCw8$JMDK-U}bteOYER^puMX?LC;1#IC( zIwyX?6y^_-cZ3D!(}Gd3HJrnAT;hCSG3;@^GwVB$r>+Tc5jEJ@v~jcr`%KJw9j$8ege9^v!YRiK z!vM+4N2L3`*jOCkyc!jt_Ormj0PKL!Z}X+csQFUXh#A;_JT{r|CpUagiqvl8f7=qJ zfYZqG<|~DW2h-^aI;?Q!$Y%-43#t^V`pKl}O$Kol$(0&V!ExlOZARYULK7#u$ugi- z;1?*WE$ykcfAL}R>%87n78o66K+QT`=W%}N@T8S!vKb0LdsKhSJPk9Oigy?Yda{`G zC*D%Q6>UP>aY>)Dd$QFCqnGJDvKfN(O3Oo5hwN#t#yBu*X31zY0YamM1 zggF2BF`ffX04*T@FZ}9v44Q-RIE^N7-lTVg(jfxRq#d$lqDUp+g1Ecc?_!zNTTJSR zb71-yxM63r=1Xz_L;#Q6VhDx`d#$zMRC2r(&RNf|*C60H_x^r4KPJ3+wWU3f!rzxs#NoJIvfQ62 z$-X!8)_h|)nzV2J$_FkS+)(yzU-ENhoTyG&=zLbA9)!r$q zll11ER)xuwp6c6+hDvoBSyS!!d{$D-&&}5qTpk%l!-TEjr@k7+bNbc52pVzStxI9CwLvo4)&kqLjtg+motC*m8kddEu&8R+vEVqe2K6g7UfA_rS)VU7 zxF+zMO}wjd`zEy;>{~q4&W)rzSsFwQRH3ybGx*Fmq*Z+Ybk-}%A?I^JsFqK+uX4__ zU^fP&X4Ye-!z2r2!&6fazH@?U9K!u!CDVv+%A^l0P7*V(;0X|$M1VhFe2*mz^W+Jn zkq`c+nRB8Qza0^QzoX>Be4~+j-y*XvR|3|?wwy#P!<3iI`bnl=?^#Cyi};JTY>G}0 zW(*t#F8~yf?#q1fb0m^a2%}~*iFLaz@3j*C>RAQ`;sNd7AmECp`-LBS zGr>(~y5A9id)xaPsrW!_bL3r7l@`1!Zyl8Va%lpv99ky2D&~XJSUgaU^H^};Ic;KE5S8a5aKxxDUt+taNK8-y<@t8zM`_>XBK~gCu8;#l8!-H$$z5`-p_{ z5F=eeuQ^wSFK`OM(dwP=rwFdYUtd?udpvGKh=u)nd}nK(WP=@Xg{p4PJ9(Mf<3UQg zPTgP;4c&tX_rv8TZ3rUIw)X6J&0RO`e^6M(8RtBjh?-TqL@j{{A|(C@Bc*jimF==! zL_xDLtR&h;yy|m*O??_?Hq)ou4r$kkG;w6Kx*VO~ymh*&@-zTiDEOeuNJDtJ-d3f&J># zMk9+pt>kab?imePTBIVRyC<13$=$M3Bc7b*<|eplWoHFHa?Q=L^b@o!w9hSEDzvP| zT3Zvf{&gY-WvUwR+v4zVN;TifVM-K&Y6{ETG7|T{VcdN=kQ7$t4%jCqcU739=6P#f zcRJq7rAbH2HY|4ZB2%622@FN$0@WNMO!$*iWO&F@Zw&zdKLEhOv|oRZTgKTd-+rf> zqfh_-G|tnV&RpFZ7pd}V>saicQ6618sz(Mx4FT7i+FZAm-qS-oq~VUdZ%jko4P|4$ z%y%yKUD zZVGO2ec%J}#IID2pK(~u)jvgNNO|%Fq8Vq5csNm?Zr%sLJ@xgUH`vAz5Bod=i<}9c zXqF9hUWeyDpZi+pKmD8wvv%AQYsg>U8~dJ9gbP7zZkh*)o|%q2`I@ug|HNg3JAlDg zLY9{yN`|kJDO!Q?<$xz#ls!3#i>8u7JXL%{4Y03Pp7g*{@huK>OTM8uOY6fTs@@?UGIIpW?acy3^ zp@TdtKW(0@R}S!5n^l$myi!$aIgwhl`Szv=#=AIoERbh|$0oHngII%gp3Kqc-b1ME zl3#l2QI}LioWq5?=63jR3pQFBB$C%!t;{a{R6fK2xRCi#{ESK0?J3b{ z6eg)i7=JAcSDvi*mFd@bpb?MR=Y5PTonU=Dzm%_>z}ed%Fxx`#F;7ETR+!|V{b8JD zt6TBuoG(6w2(Dfudv%F8nX*2NaILyjlHauOZSCIt!)dTHh*KPw!J;4R0XaUIwPw(0 zj=?hTjcvL3x6;21ooqWNDP+}Z8BYeM2)FWZAo@~#K~bt zyrRo!1QSmdIpn`sF2}f1i?Wb;G_CaAljGvLMn*PqXU_QBRjQa+3|N$GQve%RvuENp zuI$(#bFuVTypXW)NH?|ug(dBL_5yJ&%_7mF>Vm%<3oo{bTB1ZbRJ|^aZ3c)pDg^Zo z^kRob9d*kEv{>|tCrkZHYNh6mF4x>C_00*h0u=-PeY5hJo3106x64ymBPou;kvl%`Uam@H0E$ z(I)PqED0iFz1~IT59Com5?sX&FckAODju92EVbd6nHfwC$X465rb`dZA?Me| zokv{ObdTR;bAYQoj?uP#D%V4IkW=Obrt4&Cux1W97G~a&j5*+* zxa=in{_LpTIVDuTK83jVaQukwxzNe(l2=;agzmZ9*yi;A{`Ra1o8LFQ zsVm>?V9k<1`DBD*^QJQfAE_7sVe%7<^Bf8@@$+HVrv^MD32>;4LnZl$ujY*lCTugs1bSw!7sf^g2q-3L9w zHJ%z|Ho(;VAHoQ0s-+E$i>@CCJP$<1o?CRVHV(|U@$(^w$2-(vx42=({(sb8m?Q}= zzH5(GF_LQYl2<9nQOCAnOGCqfmy3hsxnOM=*zl`!{u?{b5}zU-GroF;={CIN;SB$I z?SvPg2TcET3cJj@P4`oRL-^YShfR^OE1Ky%<69#!5=+0^bk)pW z>`I4>w0709IJGC3(umZfadKju8-?A2u0|1y5x5Aptjh|6&>tl4Z@clWOJ?&9RLV%)Tvvyg-F z>HjnD!5>WloKb%hyLK@gaZ*pw1PotNH1v=d7?8E#4Z|2PjlE>X#>{0HMkODOGbWiU zYVC&$Bs{)(N7>TBNFy)#kHI#UBnTenJ5e7!%3e`(k>eO~84ObKS6(&+211(?d*Q-S z?D6`(A2q2{gp=WWDZWW~l`=jFj{pXq6xJ9R99ho7Yb}9pfJ*@&xM#EJ?_bQpTwaW>(E*eN)sdCgGYNZ zC;pLw(~n(OizG)Wi#aOSM7GLG%nvlqh3{L3#oUlkA@Cgh zM46{VHisc9{ z5cc8Te8p;*IXR=qd#%jyq7q|$6KU=0jQ}&{=_!Ti%`yngYme>%?Ai8An?F4`HTmet zLSRxwZsS?a{&-)8zMXFUq4Or0mj|M=;nd({XJtTpu*9SG_%OV$_5W>uN?H*L+ThW~ zcv^R+!u@3M-B_FnyWW)AF4a4q^8c*4b{S&cjJmQI!_AFmOs3&{ClgJt+(k456(16? zc>!Y(R(e;PJV$JxS>l!gtNy)jBG!=`UZ^H3Szr}UODf17>KXBmmgFP$b^n47Z?ABHW~sAE0f215UuO`b z_jQ}B72}(>mVhs>yKl-_1bx#dAK7ULxQ5->6;qk1X|nXVu&mc>7c!S0*R6awvaVUJ zUY~x;=0^Tgrp*dm257(SPfY%L(IjHF%#UzO?6M1&VY}D&ndQ(Ve=@SEC&qIUUEB=k zzD&o#^T8A>BIN{ZJHtt2T>9}R*OcbEeW(htjs4~Xe`#s!TW*>SdYBM*pMl)>IpBQ_xMbTk;qk6rcXhEm8!YT z_^8|%^OTaZsz?v09>Y5W9j7e@`n2O!Dv$2Y`1#kv*Uttr&HKXHQsEQZbIpTsV$r@S zS_i9RwIo6x6yi}UEZSs0?|EYYlt{5&@au<#xbmBVfPkK^@$R(bA!DdRQ@`fLf`L5pK)lCzRfU5-(9ODUpAi_` z0zYxADjs-+Zs$dSdx-M!foE4*R@?SQVj@QHu2(bY)_dMTf8TTD&jCO{*ZJ=3QcTbr zP{^g??v|v{bFo-6pFA{NexI&%0Tn~VDl?S73R@^=tPU0$^X%&+-ZmPoeH=V-UR1Hw zR;LO=jC`FIo^Fn&In#Y!`(y$N0jJT?o;WQ+!y!!`lXt!iZ6W9%>Tyr){@)$$N0M*T zvM!Q*17<=64_OVKjtAl}g;S{w(*)j+iXY@S;&Pqxw@?xl|2FO)De@QbZx&d?JDZzCn0{$an6TFs0M$ge;FGxx$uOp3^>&G;QSR1afK<6}d8H7FjA}Y! z=X_qs36wN9ddk2&lmH&W4b(6{e96wpAd`+K&B%<-9Z|%BWx^9tOmEb+gu2Ob{Z;*U zu|BRF2m=&t-muVF7w4m+&3~z27WrCSLL#zyg8^nBSK^C|4kZ=LSZD#Yb5RyV09#6u zZAmq5<%WL}P7%Fsgw=6&J*e^eR+IXm_MGdljE=H8d;Kf$aN%B@vCj8`Nu01eGDvR!Xap~m%g_*}B=(TF0dMs4- zcFO8pUs{=koHJ%Wk3~NX_{-VOp84akdTNKMf$pQJa4;A>tzGWe_z}Q|j@aJGdTku< zL5}SquQ@*i3!zeyfC>&w|FjVDO*2>!es-`ECQFc*~c=G9_szVgp}K`!uV9mv?Pb6k(@G< zMAs~E;~WkS&`DQx73G{5Qek@`yEx-d+s44s4+0z2Gk!&5qDX+27W2kxZ-&7t*KI3N zL-2tM?c=3kho#s1vu{RW<8XNadz@j~Xk;PHFkYeg1|jF`S(YmvZV|#5znYj315bh4 z?~pvBK3=BqM2d`GAj#;-ZOdKlVEi4De9s-c7lJ7mk&t97H<2M@)kICFWXQzbc3iuN zJ7C(IBI+aQG~ca2MFa6*J>QecSI(%xf*TcnM%7R^(KL=LiyjEc#^LSDI_IhZGOc4{ z-POh85m=aT>BRjK-55}b)`3s|gYkM#+9$flWnAO*I3!m?XlieHR^URe9M0BqOSF1( zVEYlAFqpnizxIIkyy*36e?E4iNGSFG+F*$;wBvI9?w3zh9DTVdLR_=ZMjw1=kIRn* z#;Ml=|73x2{{Py<klo~lW&Zkef>`yAHBW*)xX#kTyyxucFb(bc

    R!mTNNHj(CX#na zJC->s#-|_aQqz)~rC5vu!epYe;FDxOB_Kq|KS4m(w|Bqu?beBy*Kf_wzf;l0bFbe_ zGOj9SOM62t)-GF;iT0(4DlJ{M;xg~gx@5=98&t{Gp*|mj_7U=+;sI*Q6`zN5 zjB8N#@XBWjmQO>+?M^nMHAjH`fg9tOG%ttgHIB6o%#kL@X5wV9(5H%W)Yt ziaWyN zMqHm6LvG)bz3O}z!O!;Xd}40iD2_B^8hVsvMgivipC@edmTtbJMp=_w=i;AoE6TjS zCTktib%IAw#)KDOSpviGF3L;y7F-ieg1w!{A^n*;h>?E+e?SNg2Lon0L+O=OYft?v zgf7$wF!}z7W=_Y}-yp=-z~4^~#NAXyQO6uKQ2rupVEY!Ch@-sDoIc=7_Q)ih3OXUwI`Ku>by=V`o!*8!aUbh z8^_B$3BB@Sv|hPTo{Hj`$C=U@_UrlJn?;Q0d!3!(Jxz4h7fzOk7JEXe{3aVl?M#3} z40!qTQKHphLa8`%i+z9kaWLHnKf-Z!(}~gzfcQhbK*z@pna<1SoHgMiU0L|gMHDyQ zPky3TE`yi3JHH>If+M)c8t?9IIrSC6O{aw1rj|zQAwb&JnpI6O^SpDV9_c zZW{AB<@wN|3;lg+av`fD|YDpY!M6LX{?w{ZB)C9ia^tY zPiKb9L?2_=Wbk-p4#Gcs-F2WlIJ%xoUnC0|>rOma{hS~Qne1`<{H3wl!mISLrL$3Y zJ>YSSc6y)Hvl_B~p;3sZv@yeOy~Sy7?tLH+|K_o!)n1xu4|>mB!{Tv!AcZYV4SsZ~ z!l<}ZFJI*2+|nWhs+ab`W_fR!8+vwr{!GK!pdHfNAh6f+l4iKfkVU=sB-E<^T{Ne9 z@6)`eP|0Us`HIFF3^E4ZudUs6b2?ibslauPV-;>3+Rht(XtSl4vvm6ZdSnWieTKQl ztRLirM$F`ScWwD906vZTBz!c%J^W&I^lJYl3EpOLa%Bvp@Wys>6$RaYiybAp#XGx* zMlgXs6Kn2`J0na^IdhH7YRv)#FRDlJZf+W4X$;g{#AImj3Jj#ohqYmRRaGIjd?2XF zo-5!t3c@Mc^xi-edaGd0X%>U(8;ef#TK;Xxu^OYk3umC3Qsrs@?)nt}W9lUI7UKAH zZ`q%|)st-N?4a}EJiU{@x2n8rwgJ8G74<2Cz3_1QG~BO?@`-cw`RH4seDXDw)`7ND z@ooci{?DUfDcRFeuyc;l6iQz3T@qsAJFcEm7t34T3mU3D0W!UZ_ zv);_y`})bFq><8zTI#~ahIMV|Oe+QJaN~DjLR2gC|AAKiKVX*#5iNbx3J6;c;@8a| z0GT57{IqZ<_5j2mi6-i&a?j@{U1Bz|8COIz>DW0JGoY6fpW+ZmZWase!*Z7GvZuJi>zq4Ou9e~Q_gwiL9Sd4 zH%Cvbd2jxqNf2#qb%|~Y7z?fxyjwb*#J7vS-Cnaf-O30#KaCuIh#KkCzsnwBuaX_b zLy>J`zZDL~N_aAK7B?@hkLm>PvDoS9>tcp6zlbAS90w@ZFUi**iRac+=^rX2^_T%d z&#(};k_0xgw6eGlGsf!<7OM3cK6=derD;Q+e?(NKj$RQHV%E!k_U0L<>zJ`*SJP3^ zU?3UigWB4IshU+UXss8$hbiwcz0nSdN*X>v;tUo9yCpB3W;I8h!HgsdOVf|!!V#%| zLg-STKf~!#h6)g&fqZteT%-T?*{K%%)S_ORBA;Dm!am&fZ_jRL_v^D8Jie2c_8j%> zd{ol^a^p8hRR_H>V3{M1>eNgn-&S>*Q(3^^Wzj`X1lzK(3gwI_+Jzfetdnba9bjeH zysy^_2myBhMEuDWpR9wAcs$9D9RCJcmmBq>@CU75PX69Um))#us#{|Rxjgc`UMd$K zyd(qdmr85co7xL+IvrzE(|3XgZzX6cEtcsRXB2I>5b%HDl8y*;X31w;!bK)f1_T7D zo+uYcN+RoPO+sj(-gf4B%iV68>8>Z;H!Cjp!iWZH9eQpr7w>(L`_#MFCmM|_-7n4) z;`ovG5f*CRh4@|nQ^~&yB~FLm5bFw%USlVSc*l?(P1>)^(xYxgAQ zh86g*pT*>wRdx-A1__!>+C*cz)`K(h^gxUAzU#F^Ej~*J}p_s*#T>=o-7pvxW+FTg}e! z)QU%6Y9kZ}PpxN$Ds;}#BCPvUhrI1(TMi=2%ch{9)l2(Vw$p@bJjg-&O`NXT@ep>BR&%1--z`;*!Ix%a_joO)-7UH@@|sV}$dUftSh z6h6sZse6J=4;O!4n?2%#Z3k}}SmVxP|~?U?_^CE3#}WLT7d zCka8a6%5qMoiQ5?y&=Y$L8fu?Fzq>dpoxmL!nMZ{^>q@LVoY?c7pgzR-p*i*98@Ic(_b&dj znMDLd0o8w`8TNmjsB}6;Z5boD;a^o9GQVF?u4OZ9kEXsI_6_%5NrLDRf!vEdNdk^; z*$Le9@+qP|Yj;ngvlIXd%P+2aILWK|6n^4W{5CauV=)e3Z4mzef=ay}Lt z4cH#wsg;Z;-5Sk&A$OFkllFA4nu+Y_%Y={20iIuNV90QDfO>~-V(l&IR>UUlQ~5h{ zbI=%`>o>YID$IRNz`=A1Zl*Jzvg9)LLL>iTLh@f72}At0Lg7CCLkEKQ1}7ge(-H{So5M%8pf}G7{_>>eZ`}l)HaP$M2;OT}Al<@se z0bKC@OyN${-j)DevuN;*^Vwkfrn6aW|I6MhjF0LbJM;@H%viwCs^p-kqn`Ynu%HDuin>|6?6)0jDPXVDljY}4>=wmOap}I_fUxlpZ(BwN4S{Ch zxr(CP25GxT=SR*X&tP@;$e8O1?urgCZ>p_3cTA~vl~Qa~>BT}He5aeoHJkvLm^=JKm6yykR)smDW-r)n zqqgupe(wXtbI~IgV^G9~`AixZ*v2=}ix&aF;E`Q_dWbZCyjbCRdb;=N2ET@V4acz$ z(I`och6E}5%7`#RTKWorfISopi+LHynQDk zosl%ECtv6Y;feK{_rg9epQpL|)`7{e@}qN41ZVtqOK6m^>r|{hW*3s41s6^KO3#*| zfg1LnfU-FQOcYC+5r#3qVgegsV@>zMq7E^}!qhTP*P>Ii$nZpuFi(wCVrNVKE?k;S z41TOEH@PS>0025r=*C>m!cfdETs`R!RQHU=zZUC|$r0y5+_-V~txL4Fm-vQg0^W^9 zkCeu}u=-*mCK#-qMQ}?>vS6f;$s8$Mgea>1V%2Jc4EQR4>HpH5<@~Kg=dwc&gkMG? z>JBkeR{-9SZZ&9^T{AEnFqp*xgLXgVm1CGwi0mvmL+E2eYnW8Lm{GU;)b zPVu;m$8Fk`tD!ViUv-fBAc`Xf{&jC~X|ULW=8@qi&kfR|EB<3ASNk7~)w_@U_Rx*n+;%_)5NxN6xDWI6g(2>7*YpA$)S! ztCcD29WIx4xOdfQZm9PD8q?RTnpJ(l^Ch}X;?1%t5swkZ4-Lj*Q+e*um{Sd>;^8p& z@neTh2`YZ%ze7|%pv(M#GPC3CYs&bbzcvpgUMXul_cHyso(6#;I=z`27}6l;F^hiJ zXJ}b3dwoG$p0Og3WX+I}&sI$ntC{4Ry^rnWA_{uNQEPaQ&;_+`7BVu$As;{D>Bp|6 z?uACj9XV%2(Xm1tjMZ;}0pePdo)G2{F@S;OMOm4wp*KFNX_H~pG1>Y*Z6AE4VYaGc zsO7wRO-&V@VKFWmEeeAnFkXh)nhp&;B*`|Yc1ah8p=_SwSdP`3YKI=3Xa0PXsy7}& zts0w#eno4=#2tMmFEjY&p+$Cyn5?BoDqh7re z^&-|P5~o9zdCjmGmrF^i=EH>j2<3|JHA__4NIq=NUauhXB`2$*B=qHw@L;Rsh{GCO zjx4V_mDsFVS$b;w7H{$0Cdv@KENnd$6GpkWR8;o%kAA?Pt1}6J;vb+x?S1=5QBKhO zpfNc0G3O7^;W)^(x)WZY7Jw~W`flIZMQYsagH3b8#_|m#ipE45)D0J!qXdyOmbj-z z7LQF$oFlM349$Cp`9MeZsfBbrfxH?y4*A+pnL&;70d@6_9*|jG*y3rw*;tJ|nmlIz zV1agx!`@ns5Fh(gqvwG0`8+->Jkg5`yAgkIIIUF#IL;`leKOiYh>)%7)m%<{mvSG% zarY!#bO5pxr1#=cg-I9K=KRf`IrI8B+jX_G=A>s<6R1lT2ALb>hfD}OdeWI)qx-hxkG&7UET5HO{9 zVt-3uPQrN7CsUKNbVX7Bf##*GkQWlyjGx?@sx>x!8)-V~2fj2q>PbqCRGIdxl2p;f z2beh)WFG}`EFS4mX~k+iIeZ1yeTr0>0+{#zL1WAs=Z(Ln-R#)XX`Zsnm$(38S&Q`h zv|9P$0cO0Yd-91US*#M&Y&jf+Rt1^y)s<{qAQZ${ie*GDShE(8g%Y1akMjZ=Krr^V0I1qRfDC|B5QN7SJ)7grxlSLN0ab!K zK6*@w-qkl&`PEW;xuVCH7&>IkwtUF_1%5A{Mb=rL`9p9boZ(pi|VnN*9sX z!oV7htMuX=*q(pr*pIGEVfJEgN_h2_g7wi(Yafw*y#yF>c$s`ZeDD2ZyVB?9D>M4b zb>q8tgm0M~q0b6|iB97nu%wHx63m1f{ileY)gmPj!R{qWdE(5RDg2A6q8TP!RemY* zECtw}xSSdmvyT;B1&DoFn?#MY7=n;VinedZOo(l5^_7Q_CwWfb6UoEN z7^0D0DK9A)0=c>tB9zcT1PqVRLDy7N)m#^qG-c~@lyuSOXW~q3ElvLs0NsAc+|5(X zH&^pC){tU89r6UpIL#rN7UEne#_?n#c^4imXwnCct5#nXUsJ7KmH6`FG$JrHNLHyJ zS5Vxt5Z}n|Dzrnm?C$1pqfd z2>^eG4RFlzx1;;<5$P~}+*^2+(%)g@y>MKXjk2Xn)JZ5`?2LkH=sC&k&2~r8Sg+Z* zM9;%gM2QaDR(bLtgj`1)qi_3UNM^CB-+Vmw`NP9}!525%ttweP!6NZD_q$b%lyZQx z=j6xck3W>(Y%M(4YaOB&g1l>aaS}Zf4uv4@zIYp#DUl`uuwaSS+Dny#NkO(>1acPM zDWFXh74H_6fop8f?l>&9(eT#@xL2CL-cYAu(Q|P%_A6ft^qFiJdC~HIBNjT*7N*;A zXcyG1-!Qt-aqJKP2PR12Ep5SLeO~4$dEc`#)f#-`KWsn_?aTeTD`gI7{&bh{pO|E^ zz{lS?7ZxA1x#_fN6(s=rQ0jMYBQUbgwL}d z!)xI1R{xaj)P*14T_2gFy6@zfe;R!gAibIfIQ*aSr>&m)@VO@#IQXe$q30{89gq7sg{Rq;xK-INnGsn*-A}`jQhx{?fzIb~cK=He zcIx2)%X6(X7&=-Kz5I#Zk@<3Xn*QYa83J~;=r0NZ6lXMm-QhSQ;CRP}19H6M{{nsu zXM~8-$c4e%=OiM-|NQBJk3W^)JMzG*bLN?k)Lf4`Irz^UT5vfVW;@ib!<}a6`TH$8 zXv{9hT~+8jpX<(@1YEU(USju*B*XB8f4c8EN)% z%qB+5i7F>S3*&Nj3Ll15K5eM1)qUO6!nH!%gl=~6f+lSC^8$;Tg&={_q(?Ag(}L3bRRy7x;Nqdr-Cb>at*WA5qahm9;VPtOl1(mjPGusyvHgZz4y;~Yz$ zIL{}yTWN2;n)Rb`9syms_eXUA%OhQ(V}KJ{>4F~2iL;PmBTzTff-m~kVa^mdb1@FS z&&WF$b?*YkRns(F@%dK|FKC)Z0bc@gqP72{ZUsxqn_s1{IbaD${QK`c8>L#pj|6Rz zf@bay)so#4M`(BJj_97Kk2{y+8isutngR%$a$-!TGP^gC&*74;+YbH1wA%>;rr+Ka zm&)MZX7rYX7VX-fV4&3JJaD8CxfYGIWU)7y z)|NfK&dDP1fpXd=fyT`I;A`U^zjHTK;mI&ImS9HR*1JG~s5oVuXJs1aS-oVF&WZg( zbINrNfpaIPyr#>fM7m=~;a=78S5x*=?7v#=O+)=DG=%w1rU3Xc{OWcm{mDLNQ|I|4 z(f5p=7E})~TOE zqRF5I|IURw#II2Kmlzh_uM$-MSB9SX_x7fwv(4X5?QC1;^u%3$h#rAkfeeL6!6{&b zj8lpHK3I)X%`h|&Q9MB2y>I#fq)5c8F;>a`jstV+c7U5_Ge?nusz~SFRSQLz(E8y<-M(3jmg6CNEQBiTmF5e zCH`j{_l0=xI+-K2E*jC*iB_Mt51Psa#Fl%-;9+#-*Y2(3Whk>yxz~3tqBmbf+<4%T z?z~0P{7?$9d)Ga*GZe~6)iC3Fv`0Z#mPXHnfYYC|B)EuG8P9kD-bm4pf76A^SspEF z=H-QUsaa_Zy<_5+9g(?7ew3I;Q~_6lq~|0U&0LJ`&As3oGl$sMGKx2k?sSG+)C@2V z6Uw}JRZdvTBs@e+a32JK&mtGXB}DdHqdm@MdgK-OF3v}Tk6|V*gJ^)fnE0bylUVS` zqn!KdVRZ|fHmCf=T=s4!IhN26%T}ZuR`jE>ow)MI(jE&hB^{q(_Ll#vBaQ}8WSVrhUS}lTWG)~ocM7AtB+Vu<%UXhGHmzI>xI;} z^qRouaLmi`D6|+#R@Di7Lks`o8+R7x^by?VrDy*SYhM`<<+_FY(cPUxNlJ=f!2lxN zB_Z7)2#83}&?Sw4bax3z2vQ;~AR?f2BPa|AGT<=W8Qo_e-N$|6-uvtO|Gn#7>sf0( z@AK$1*pCWI-NSvYBW|1N-j;*A#PG#wf=fL^aO_%lrSH|2 zZ0&8~8|~hMn)l9IAFPf=QK8DQf}(-##s}x5Gj}YgUAuO;-|MH`?2fO{HJLtV2{)96 zeZa`c$q1hfN|w`{877=Xu@pY*)A%@yyPe!RO3+u29bld+v(~!0`}3+%L5dZ->stjZj^s%})jk|APfA<%(<&61VTJaszj-U$fi7w+CY-nwchq3xz#- zn8~=Rf&BWuF@r&+xb%m2RGzKBAjBo5_wKTBdRNPP@3PNnW6@0G;1!C@J%oUy>W4HCfoh(0SdsnqF!`jnCVJ!QLZW7`WaP0n%2%-Z{QH!dSp zT9B3}hpbOucH*A}$i9bf>RyaD1FL*PwjRzwF~U>k{MKj&#y#=)x3oh@z1GNFsVKS3 zLusECE{D^5pzb;-aK=)h9z%>A%K4YA*KW8~>=hNgS(m=`$nl__6D9KaK|asbNEudw z)wq_kH#rGB-1hw-QofH%o&xCp(iqiW%9(l*xl`$vgTwE|38T zSVXoBWx6&c6m@vH)~B!#5Ew3JtN0a2a47hA_2gxtSD?{&HW=SYRr#LXCU3D9a>X(8 zs>dm+`lm1Vl_yD`^i=Rv@$<=YxAToVRZ0gpZ&wRLJvUsRj%Mss?qu^__sx$Y`Kn!%3 zj2j!dtY0=jL)YTc}f7Wn=d9PQ3K^ zX@=PQc&d@fQ6`a*_2rZwz{Ouh0!~uTi!cd=URz; zE_&NajR&*A$_)aLuPQNM7A(65BEB13TQVpKvL3%oxH+2621t`X;Zgx3ugsAtQz}mc zYReI)i2*E^9{n-Z*T#(iomk%qL_29`p)kNkpnHx_3B$u}x=o^cBYYbW*XKSl zO!gTtIi>)@^gD%$3jBOGHHLo101iL}Ko2mYjy(6f089KIYBYuk%bn)u913@~v- z2?dRN)3sJR{YduQskipWY7@B#oEpF2t#<{Bm=0wrzu|%=W4aGDmL^-j3<3iP_@&8> zk$f-$0mG3gQcVUIoH{5cZ;^q(clFQ0#485PEfkUUPOJSabhdLXt*bhlG0i*+6WHX< zyD_czzAO|wu7u6EI$^bC5H-1twMOEdGSUCC;zWn3!#~3mZCDpL&-rVTBX1UIR}7QI z;U-D7_#??Nq3xtmZ6A>?dUs^^%8!PaSp)`ER?9?HqRT#&%EPH7fuyKP#P3SHVo}R9^yT=Gn6%3?*(VuIaM~*b?})$9WX% z+0G@0mO|Z3>CXm74pa9`W5WO4iW9NDy>VMoHqDU?3I^VlzQs<7y#DaBcu>>aSF{yr9m36l^vbpgu1Uv*~S|)VY>78&Gq(nK6{BwS?NE zxHvV*f!)I)>n^YDKvpzvefo+FH)Wb}B;F9MHKZJEIt^~Tk|7K^G<+LNN2|OD0@!@= zK|fYgckGwQ^mNMCl&w=XZ$36LgTGAG=T2t}vIyiW3cB`3==KoZYmv=Mg3Q5681AKr zazp))s9Fd`^`|-H6j$fv0y|Dhawr&#b8>~VFgEir-CG#*Kbvlr5(v8uIRph@Ldz3r zr09Nuuezu(@cI3t(EmnVRsb@ejE1kFxI~x(J`J10=4#B;?n|K543xzr|Wg!c`po(549wT%15Y5^PT-+LIi~47k?;mMbT1j;+y?fux!gPmZxL0tljjE7)#EGuf zJO&T@>zV2dJ5{g?kpOp#Z9YbK3@6yX_;&p|VLZkSKxD>ac_{L_Uuu@op+id^MTT^> z_ZO$Bhh$`G#7wT%Va3kzfz`7Y?Sg9jOaghnhk@o0ixSt>xjgk)5jSLNc}R(xj;`YA zPOjRMbEM1p-szs%$(PmA@I6*(f8WeZ(xzNtq?FF-=g*6HGJh{3@SiDfqai75U` zGSWwd;KM&da2y2uEOdzu!IfISW|4(9!#_e$?H{8phVLOr{k`YK|K9V?8MzY;;^LBG z+j_*5qA)StP?2&2#%PgN;WqYAO4($k;mrl7o|k`@FQkpuZ=zeZLB;O6ZiG07a>Hsh?DxQLn$;iURt4K~)&awC=6i8TqjE#Y*#4#>q#F{oc|0Q1PhS=4O$LnE( zYl*Yj{W*^hiw7^HsFK?7B>CYwYUMm!IN6lNdQY__r;pjn;2Caw5+fj-zTd-iZ$Q5& z-)nv)9F>Wj6bfX5386(i?QX$LsH#Ti+1aAaU+Pkn^`X!d0%+}kYC6dID&*xhhe4eS z?k_8aQ`jrxyw37I)ngX>Ry~qNkh?T$8nk|@57>4f*LcIvR2=|Da!iw9p1|L%1r`v9ujkyLgtQu#*`LDeY6A-yFADB|y?!CDA;W&5V;^aOSAg+QQ3Q~@ zQe_vgoUN$!7^r;CYPmDjQj)&ouH1Kz=f2VVA;DjVG-!Wy?|;%R7K8yjf?spyOF_at zjPZdtKiEOctuR_DJPl@Q%e=%QH0?s8T`?bum8oijYT}ZM2C-z~>pm$rmT89{GXj_0 zP{^!Mt8f=Va43USZrG5dXZ(0Tx7y;RaU*|Il~FhzP-N@&y>z}baDWT{QaYcC4br#3 z`fJ0fRPUCz>~|63;&XT*($wVpQZ5l?RXq{g7!BMJTO&T*vk1;r<>9V=UW>(98_|S& z=SxY*g#!D0&Z zzgQm#-L_Dgc=B*boip;r*p=I3Q90AL_=gaNmF0yF zUd~&DL->!PFKv_+DiVCghk^ya8d}FnZ{rgKinLK&B3i27T!TjlgV*@^<*<}}SIWS3PO1fB(N3R0kgb93vMN?R=ZZ@UdD$`@l%a2+ zW;Nvr8MF%hSr4Y(l~)A6!%fe-hg-g_dhenHUe<$81-y(N{VZvP_m+^UpCgYL@v_{n zE%?~C3p6#>-uyLZDgqhbpV|CZdx7s|J888$1Wj4K3R@6NI|h1flp8ks{Pfbr$6I?# z#&XeF)NKulfqN;4-!vL#hJ&R55>Xy%tkB8Ft7mnPc~@~elcbT=!fmJ0DA$T9%XIjl z+<@Em{SeE0o298%Z)H}XiwzxuXJQ0@wy0U-tbb+xTwPrlMKgb@Vbe+3*Lf1mzL9MV zLn=Gb;^s}CXJjO48ecO>>{h(EcW%cxckH&*6E7V0ni=Z8{yz}Z{s_UneL6^#$*1YL z`aOKM|@uXYY2AC4sjDhbj9P}sKOeijqNyAQFl1UOa76;TFo!IBb z%^%N6K_pHi+(kC*9SQ}8jumCWi2Z)y%qjaU;WSKCIt%a^a*W)`=T9KRFVAq2t)f(4 ztMF?-TaFAy9Vmu8&Fn&iM1Cr}$z;vy~&Fh8y+7+_u$`lM*I* zA+x07&i+`w!{+y~&$c+5-Xvu%HxD`eI@e@FMOfZn)nC-$_x>irjN2v0bI@(9&|UcE zZB!GWKa}-k0=?V3n}mL_-Vc%qIb~MyTAgzd}G-cK00D@$4r)jT0Rp~t@ynniO`H{--)S3Hc zd{<8dr#;sWR&_j+;JV7f#YDb$;F!HRx7EH!OrvE2D>!Z|uVSWnbcSp2zlBB7HQZP9 zJp1r$cag|>5my6|NA$Kxlg?ZSmO8gNf{w6pu0J9yploxu?6%1{&Tk?Naw5o$cL1`4 zy^tOd#F>8^ih)Ny&?|^M@-z=Txjb$j%rpNM2s&ZGXIF0)?_R6 zF!0G=VFq(&e{QrM1Fr`9?PRz~3)rSQ`$pMjZ}k{)w}p^+aYHwAjD&%0{=wahhXtzS z_3nLbRSCGmCT=}=L_WnGUVTh;UHC_;Mcv*KY9e*arQx4tjBtM`!)UFN?0=0==f83Nj-p(YdQZn+Wrgszwii|1_kbqv+mJg@oa$rRJp_-h2CdiEFL z+im78=O^Z>PvAVc3|68Ib%X&E7z_+VIyXG+mZMNi}nG)C2}r9edyu)xN;w4oP-0noz$ zDvS-@za^(%P(%x^W8F5G?82~M?dlRVRq-UEvUFvzd4jK>z)fZ2+Gi)%ES_ip6F}_0 zhvht~JT7G(ezhcr^sK%&Mpz?(C(oC_(<&?Se=}@xL`Nt`@k)t_7Gp}*W5TfV+B8h( z^12t{TRPQ0jM{vC9U8_J1U}1(X`o-WHxZ5$MYCHw>XKk^?zS{@T-$9)&k=BazHKmD z)&d2ngxk8Xb?-GPo`&poAVv@Gb^>^h_#=L;z`TGZAoH)$N7Mz5AgB8w78-TV>eP;o zzG#Mm2Y->G7Wl&NdpK!HrH|j>Ur?hUkP5 zO?}h{fq}&pRqw4@P~;vhDUp6&E*MI&cw$Wk?W~OnK<+<~gvI7{GVZ>T(# zS?x<b8RqXI{vDzN{si5h7cxTmZ6B+4X5Levc^o!q@ARdBnuBj_7~7L^MFw&(FI6j8Zw z5rc$zOB;(w44NCljLL-})E?N&VW?ba8UFjH5m$0hG1m<|LNnar)UuuxsL51JTB#i0 zjVRC%Yc11NP!THUQdcXX|D}3e{H-=(8I10)xq)K2GoI}_3LklJ4J3Sj(gm9ygj}>9 zvN<_eMdb#`98a{EZ`5VTRfKY*Lw9!5BYyqoU*dDYs_5PL;|-YzFyTVw&}CblAV|I3 z8RsF}NR7h&^Fi%}w@KUaO8Tb~OZ2f!c??_bgu} z#A8DF;1RK)W4OWsG1Ug>x(x*3v%e9>B$>(1Neg%W1ft_M>lG@z`fTwz>3NF}+Juqr znTuk_QOU!5A9}*5LF-(HJuCY(DRR{!+AJAV;(Nu=%TzJPLrX-iU5|nCb#bp|#Tk`*E;usw|6AFDEL0NE_FT<$-baHh1CZa1(G}wvZ(@4$YOSJ) zs|ARDvBKdbw1H1+!g{B<1OXZ&*{fvAsG~3vwas|ytz`aUZ|ZUdKddmJ?B|DDx!J-* zR>eb)H?&P4pNJ2qpH=B3azn^doSLe~8i7f|P<~j>Y#aYmCq02__JvNd&xC8&$7?LQ zA~jDrw$SO}HkE3hx3|q55Lje3W5lO$&>SoWj#Fu3n;k79`*s>~^)MTZsalxx8tua& zjkM`-Kbo#JX$;{MA2dG9W^$cNcjs(B`B25E+3wAL7W!qU9(Jkjc`j*3 z!}u4h=7+*$WlH&$hmmiqlSI6+Sup%%^Gq*)9 z_fl#QsWe47;u_3kVGujCEMoyO|G{R(+8*~48YAhTHHE2tWUyo>CZUOVA{cj)_(=wB zusGz5-q7OAHxwlj-*8-4)?u_n!6?0x3%1nsZcq#|5zCDy^iuL{Sl}SB1jOA4cr`E~U*rt)&=qPix_d=qHs|J@Xp8lvn*Pg%C=whC& zdfcVw{JCh#cz8Si9i~Th(R+g!L6HUD*|NKsVN6OT_obtE8$AtZe@SLnejjEDOPvwq zkjwu_W*VXR+CQ2-@qzv2?;9lqULKb!dZSb|us=Qg`uuc|1&Kqg(9_fy49l^Ex{sZ!J(Ur@OB7(-L7N+<@1R^d#e{gpU|4tAo9e@Kg8A zSDAFKbDn1E`)M;^$enZ_%avQ@^;P$U5vv=o2o_!4a_6gv7@fqj#+S`)7Dd)+RRZfMDXe`lV2;SBmUmdl+oXVgKnBUfElW?~b(B&g$j8m({yC zwNzyB=FM)_^y(=qmozHNO0mCRQ`Xkt(|v4U_FgCTy8B1lNS(m{XGhu(E57~&pP)3E?-x~>_0xb+j_=mSmFc##8^C&vdsFjk|hT=KQhxE2QN?vDzOjg!->+>`L z(TBc>W8E!<0;NR8XIq!bCyLe1&pd`ic=bXxl7rhBMR!1zci1UkL`+VB!jU~zI7YhF z@&))1=DRyyLO|;SrtL> z42<>rYj(tKtrf5z^JVl#$EL@e^=1Z|6E7FXr?11jRbTQtghIQ3*>S%_GgTA=u`<{T zrNC35KST^R=J$h}L|m_W$E#rA-A)sq9~xh+1v5BOw!d0~zQ-a-^)U7OViwYX7g=We zO}BU`A54rix|TNFZKA#;k*5mJt8NF7YdGIFr?rCPDLSNvmkgGNfls^|h?lf_b?G;C znN=9O;@_5Wa^CTUTHF~f*x^9f?l+Q6l4bUGT>P>SKU;o%m(|sc4+fG5oYPfr_mKr> z1f0RqhO&~bS{0e%MunZghC&wRfkUK$<7 zFs(#IfS`Xp#08tKFlGG~T6FWxPPOm2m85Z=N1Yg{UhEmK#yds*Rzq-s`fZfx2XW4r z%Qe&!@>jn%?$;!5Couf_ph^btLVy0%E~?v{^dAr*QwV*fta~7=sN9X8w1&#vl5%Q; z^|rl9)R&%TIZ_m<10sAIMT1|~UuTn^pMAT~=rC4@6-6uf+#`P^NHvz@7Nfq?%hD^A znAsn6cBaDR_HM0zcv~rx@Rc%&kN0p&s#$$-wCa>cXrvaqT*R^NA%R&dN9Bi8Yx_KK zS0n@S2K+b=BZw@0PF3yFW_*w%j>eSQ)u*OF8pw-PoqTO0C?7*jT0P?sWDXdd!oI9( z&QC>ZHiW}pD(-=r+=zrvb|UsaXps@&`G1%OE%xmpD9ZeORv@(jh>rnuS?Fgh^{L5ZEj6tHrdmD?6wM)jEk>ddo6^lfdapVns)Zwrd47?RoPw_5@5uAUy90r;g zcq^<5do_)HX*+Vl&h;7ZA5AK^D4))++-g-+t_%3*>J#*{;RgwAsM1rb9tb^&N+UYk zg?9orBlHJ_%AuT>a_$UHSgAY`yxkGX&`|O^Ncf)DPoFAH*5!`6RG%uY&9o#>xw3d%TCQo0sIhZ-DFL1BPHsMIn4XTmF1*n0wyMy2 zl-wzH?ryme7DZ?4Z+;FM+mFwr;~8(G(Cj$tnQdMu(!3Y`;@dZZy)zG*d7K9bK*BSI zX+B^&JOG1R>9#%LhtNCdi(+{ftwS1yCoYm7Ok&{;P&MDhw+LgEfaOKDx?mVZ(2J3C zMDHd#$Qu%8;TL)wYgu$*xQSY>#tt3ft_3uoT|pgS zTrB7z^>cXIO&Rm4)-TmYlQIaK0>3fx!~MBuRVz-V;B?IKt=vTpMxla%pYyK`paqEj zb2|hllj#ob#Qh_zFwa>~VZ|XIMK5YRXSi>n5X*d1It9K{IFL*XwiAz}EFJ=WL={Bip_tY_f4VH!-eZk6g<#SNFY3s@IH@jBmo3Ie)@9iT|#GI zDf-3gWUf)fITC(A@vfgFy)$s$cNTWrxaLO{Bw36`-907$71&KbfL=Io znZ96UFh$^cN|s^6P05h?$ggvfwT=@k+LdOnpZ`y9$4~o;YpR>P1QPt5EtX!ksgnvL zulrc6j+OFscC;U%is<5$P$%+p?$~}%IKR5SofF^JOaR2^?&%f)){S~ITFV#lVUm`{HKP~JF zWswTJ0T(Ee0eBP(#GmM_tHfO(GGiX3lo`wvzm9y0v!tmRgKe)te6oY;>7`Bi_0tbl z$bzN!@;x1Q_@E%o0OS)NZQfhgA@#_*XARD><&>|ly&5ton+_KgB<$T@RWxk`5c+q0 z#%dO;B_2<`a$4;dY-SYRor^vUW9oNZknSaVwu$cGZG{BZp1@`FYp z&2{f;p8WA12(Pd69En6+J*I#_;XKSx#H-S!;MbyCOP+EB@TE{{jdUFqC^^@%uMCic z&BnrNKL+NEb1>0hXStp1Cg4O(r+8eHPJfeWLT_2&=fRgJpa-x6cfbFA2nId;w;(YD z=MpBbw|l-q919-uRI#2*IfRWbL!5zfeJEK_vO~=WwjiI(d5-_+1ic}^I_D}HgcZIF zzJ8$ufdS>~i&B=9$lgw|n~YX-`H*6qSXdjR20>5=96G%T!-T+3x~JmCqi;W`xBW&d zqtigtj0rpH?BvZ`Qicx$S4FZI)DFX&5 z%A@E?HOV^QI}7$>LUT7<@pS~CqdA&ncdFf*WF(8{T0Gu5HOXhju{fK>B}f8476X6% z`-z>TW=8e5$%OwNEh5c#S1Xi#y@e{Wq!Bh?=`@x3evr4ZXMLH|_I|MJ^rlev3Vx1p51FAi3%L(WjCHg(posTfuWGoRf@$mt)hXn=6g0ls% zfyFKEPWa;F-d2~1y44V|Q)9VqoC&#v*3i<_+e$;(zPky}DsM;9VE5Qq9aX-su_2dW z@LsB#Z3aFxg?(qe$!)I84>#|)tvgW4+*BS%Y2Z?!&{eu*h{M7!3mG|N%! z><*pVo3}-}*AdT(Yz7G`IL|ZH{k-}ibocMaW!HOso;?pd*?8g61Y0DRWI0sNY*E>G zEh)G7d5-f7IE_!EHZG~P8Vu6Df4)Z|)7}M|z|#R~vI4$eAI20wK&;aJN1GpKAp-A5 zPrzU}1)CTA^E9>h<_Rn$xQTv?)(b2T#|kEup!pyZh{(4HwZ&M{3^PJ+AAwUhEU!+) z1bd{0W6KiR!#Rs^rjnNK72PtL2R%f>pS*iw<=$|;M{oSCcLu^ zw$*yyBUc-vb$vds!L?NR3wmt-b4J4;#lM8@?zlb z(8p5ZSlr+24ETqb*ottmpw+2j{U7fl<>`3sOGwfx>yoxJoP7r37`D>tCBBu^huZzj zmb;5;kOcvq@8Pre+sgA>w%j93vLBwj|3+{l{XCBUa=NF*+hVOkImCsd<<3e&BdgcG z2h&p}ddh(>d_C??lydZwHsS%F`>P@T81PHY7Bg`skhA@pTCS?karYKp!Be)$LNJqS z5Dq&cP>?5Q1PfvxmJ9Ym&R`NJRc!V{0wX*2jS5XlI~C4bmebd2{m|@E_EVv>#QIy) zfh-3#8{kgY8K32F_#pV9Yl|VWhjTqB$29xOrR;S|DCXZF>D2Q$^;*x9gCBiknq#wiKQ3GsJ|l)f7u_O3C31Hn`8~9 zNw{wuDmRwBG6xbcTX+sj-jr(5bWp=79dygJ8XSGtm(avI1?w+)?#p&Vam$8wM4&?LCXJWa~^IOlN zWWUU>ng=~#K~E_~@ab?c8DL$MHOd22bhx6T;}ncNwgA8x?ywR?Q}EYuD9A%%!`mwA z+SjeZC^QaFz!QgAE0U1=pl~>IxpsnzL_&Ttk&E`YlrrU@OtPLCb}HN(g z3_Qbu-gKL5heny{B)!~Op3;{~<6!UJCqU2GAD8Cf9Nh-x5feP8@`P+V@)m?jOXlTW zyUe&%5Tc?&4%vG#ZC;YfitSh$hnt=&V*|bRv9yHad0BaB0Y9JTo&CqhWVul#(V)tF zp-AV-#}bS?HUBUq`t-**RAT4T_k~>Jq`|OubxZUIfx1jm&!w7q(%BsFbHeMnQnR~+ zN*p^i&VtQ*eFA&!zo_O2-|FIGa_j2Fw1#!RVtpi1)5%CY(DRc2uG`b^aE6A?nG9gM z=g(6%gB0-a0olj=@~qYZ6{%N&l%I4yyUl;n4WL9QJ}u~tp%*Z|2q!D-4`t*dl!rep z>P}!$ugHQ6tPQ1Mp0(adr!0917L@!T?kG??kS!C!-#9%!3JN-wA;i(TzoZRT75>_I zgtr}EMy8DXw!9iK4Z_c~${z5vn65F4BIWw1byqnQAA&gX^}PEsrE}EdEOD}W9&HOJ zy1dbs4c2?lAM5K3Bf2DPyo3!EDVH5)eu#AcvLbIU*-1UC58sKkAG4K@*T9FrVw!d} zxPJHU%6`-&p=dnvL437CXT9HXs``zJ-Jt%Mf2Z#h_`fJn7c}1r?z%P2cArt z+7(?-F>Nm0U}mot9S_zY6ID2iNmtc7c9DxsS4?9BrInatpe?*Ck1lrLtHoZH*wmE0 zf{VH0AZB)481?+uLCN3jQqXSCoqr||9OUr#)JntYgk9&(g&uNyT~w~WfF5%9`5U3R zYZCVfsred8H6LFMnLuys6!Ndd@|zANtSe-G4b8AodBzU~W4mwkIzFp>2Mz?_nBIfs*e0{1w~2B`&v2A8-@ha}Ptibz!eO(+et2wICq8`_ zQ8~wY=+Tqo&3k7PwII}yAiHt?oIS#|?-P2xG6;Fhj>8%nqCCJBCY`ALR^7xjGGFKF z`^(D_E@vKNf;fLRiueWUbL2UONwA! zQ%B>y?qh?#P}@{~m%$4vW`TomFl;2qOVVle*~p&&Sy3DE&Em3lPh30rw6X59cFAOW zNqC|VVa9uV#p9)`JQV(<&&Rom$-A+sbn(j^6!zw<%Po^tdX-DWp7nnW&JA93EW>kA zsZdO!OxZ5Si|wu`t5r!8SE*tkXy&iRZH;!mRk*t?pz>njbmwo9#~(WNO>@Rg%FT^m zpI1qZVkSklZio?$JpWWNy!(6?b7HrCuZL(CojhuTF7NK_bpm)7|B&30e~OK7ieE^8 z&87?-h%{*HBjW3NIbK!u4SQtmhjYZiccLf1&Wa=n|2kq7L&a*$Z(?=12SWPqN%I8m zC>TJ49^o}y7|FasGc|{1o&;UDMKVY0mUL!dyHJ{5^ROE?S1ZgRuF$K{Pk=aZycjwv z6s#88d%!aLl>dDL?dLpGzj5wfFczfCR#h_pb!`U{esZbV)wB!~;~br9mO)Sqnaev5 zt6ukn6G`SU8;U|-hBxTg3D9oIFhvl>e7=)#XjjsCr`2tB<#p_J$X!S2*I&^-Y$NRs z`3x0~AFRDLfEUZbngc^6|W=H$wQ@o>rDP1Cr78sc-&~+aUKz~ z1s@(+6Pt3LY#$^;Sx(3Di-=x>ed|5cVvJ&porqy0o|Wi4*PiCtmq63JG;NLZ{1jP{ki*FyJOJxRLI)v?EQrpDpZd|7S(=2)TUgj z5(i(oaVAs3M>VB&-k4ZR0gxVwoquX1P#Jdooaq@6azkw zxkD|#fPu~W1{&!8S^SOy$>ePiR`bQpH&ATWacJljf=P=oHTlS2_jTUnRQyd_Rwgv` zumU-o5mYL>(fx`e4*&DOYCJPT5G$4pqdW*W#(?HUUzQ7dmuyAzaSeQYs~+sI>Fg@B5uymHqeq#jqjcg$`By5*@O+xFkz`WKmU ze6`jwfeaz@&|+C_c#NSgCi{zIATZ|GXTE(itaiNgxs@0Sk)^J$w8k`sbWhFtsMzHY z`-*N;(GmM%hL4`C2=_zqG`%95WgR2X4lmX{-_yW}WsP zp<`zVQO9K^Ux1;9#C2^auVz;cKTbaBfDGomP&NDfQsv=#X5mXMmNw<#?9%D`%Dhkw!5(J@E(V6tv zZ)5m4#5rRx55dQxIM;c#)vlcKuFA*~I&!Hi(*7#gc@x#w`~HTBvE2C&CdREOG92Xd zA6JgZjuD;(P@7f6nyRB_8MU#zfl|n0l;v&a>g5o%MHSqv@e*Z+wF0;OD61* z0T7(HwMshn(YmXf3-dE{u&%DK)s^;A@dMe86au-X5m+P!C-D^!;%HU#r^S7hdscXyUn^0^1+n zlvP&>ZG~0_LLXh<*5{17>Os$?Z9+e}8Y{}`Go7H|-v>?9kbXdhgUbwKq|!4unN&FT zakRCzN_8xeH_iFKT#^4^Vw5{X%laDHKIRT9FH2ilBP&bUT{9?sgev=$HLj}sl}Zgi zgla2x3LSU<J*w(w_2Uy6k%)0@7!4-`8ldvde2pVkgy`a5FIb`-MkG@kUvK=W z^}fP|9hF~GIv5pSfxoo||4$YFw~fDMGC72B9lp0%TOb?3{~)sQ<3uS@CT*C?ySx_> zIt1LN&-KgSR7p$HsC_sppR5OuX@s3_^6Hd1y+Jmve|6Cmv+ksoGCo?CX_ZP`X!tfc zn%e;c&bK{-F9;V$;3(AiYO>LAkfe(nVtSWUgTKQA2GMsQ>A*6iEX>{h?G@ch4U%sk zSNkjV(`sHc)XOqeI%2_I`)vx}7Mg4EzP~yOG6`g*JN6lljiR$NYjnu;J{<$Qnq{4n z?`@=c?ny&lFn~KAc|x{GI_iB8Xjm}RuXv@??r6s@YiS`6??^FE2bYysFPQMO*COPG z6n?%2r3SAq6;8HQehAZsXIBJ?-b}tK3H~8Fz!7IiNry8((Hp^5QdXcs(&Y_ciH&+s z2&&D{7sLtcduhgsP$qIFToCMr+Nr6@FehCk7^Jt>8uc`Ld`T+m|8iHvs|I5s1W7?e5rb_li@tI^eY6$Ss%4|#(9Y?y8}+8KpB54uU0NL_%%tXG%2=v3hgd8U zAo>sbO{z)$zypa85sfEpOK3bmR5gh3at|IT?6hD>^4+EDPwaGKp*+4HG?7RY!5lE* z>nv$IQYIPl_S-SqRsb(5E$?KmO#At{>yh+%JPmc@B7`BaI{|{7BW@8a!al$ze6$!f^6#NGN17oAeflm0jx>>pBax&SS}fI6GfKNJL^Xazy8?<8`fbLA}j z#3lJ?8dTwktmBZ2W07)gFB9RCg)mN3Zr0~O{A@Zz<@tS-SJfIYJ>t2nAP{5#X1NO!Mnig z<1yNndWmfEQ>*8?pRRIRAYyYm9$ahAGuq~RAUp&1N5~Mf?SYD(v3xs~y(6)?WEXV) zMx9yQqovbzv4iTH=FX2^D2%=|bvR5JOdt{TNUFn=(bSX1G1*sT zNZtl*);XT6^KaV_$p?KKnS$@J03DAWBMP>trKr+u8jRJ=Tz{QIp1dFx@v_`7Mql2? zQGZ82ti-_N2fA&xwZ6++RJcli(766c6x39HyTSF9Ls6qs#!gv7@HTUD`ycbRyQ$Sb zn|KQWCV(Sg@Uz9qzf;V_lZ|HnK2+hRNV>x(R;-#Ta{akDLj$*hA7EF}ICy_$@JXjH zVQks`GVKV6R>_Say?nu9)l8La75)5duzr{o4cyv!s!%iFZ2wFD&h32E$r^s+ER$)r zcdjt4ln@%rlYL)v*?GEzr0D_1+>Yu&ngE^ao%UAu&3Hku8$o>g4+}42E*x^B`T_Li zApg8^+kT7#Rez+OKQe)o5w-ccZ@kK!cxt56bax|kXYRjIHpj6HF#zFLwV&)`zRp!N zXB|J?-ySZ+BL~eOPQUz);^jcc28hgCaOFe!v>I_s6o}yNpa;XV5v}PwL$1S{ARO(7zg$3vR`suIC86wrDVk6FntZ zDgq~8v>xj9K3E&cyh7w#6(v@xnFR^lM<{!)gMyljr!EfZt=5LqfhDpT_szONg5^#Z z93QhyRa;bI!bD-tQ*{n_s21<*SbSkeTuHV7Y#No@VW$-EQeo5Q9WzZ*-{?Ne0uWez zfYVc$L7VLPobNY-8tXdOxyC_6zs3USk?_3$_h9jlrDeB51Ye)ty8X`5JVEFT)9b@j zF%D~!`a9IE>Us?$_1n3LU#dpE3FN-Yi<<&k7Mf~ptaj@airBgV7? z{Z5ZQuP%a*FaMrRC#{UL97>HD#L6ZM9OOx@JlM(&zM&ksq(z!ElNUuA&b<<>t-Zw& zLsSi4c|Oyn|f2Kb^q&~4{R0hD%l@n z+AdN5p~lj70D9kG|GMBGG8YDV0?oTanOy|@33BE~Qnqo>z7DZ#U?y0? z2T}z0DMvmlN^0iev(G`62;}KC@j5K} zhy+7R7!}A<-%WsA+uedAT@eQs5fsl4hnYCz4Ev&Zb8amW-5cLhcyx_`{*53_hI>!U zm49`xkSmh%p|R0XxMllT(=@7B+t^i^+7h|7wVoaurTp)J8zat0F7)dOc^)*QoF^aB z!@DIDiP_+A6M~8Ps3=}6)IsF2gnU}T6T;pMnWhIQ;6}zb#D@ZID%xcgP0z<^f5u~M zsG9i@ko!qB7z)87`DaX|c>j>iw?jODE!INVWS!7;voC>{;yXjKJB1xB*W%75m(Jn1 zJ@+JGxjzBqLrAkt1O)-NEJ;oIlMCOZNjy<`_Yq6xRk@+!HB&fe<#ZBrd?1To;%crr z7+Itqb4d2SnH+MqzaLU#(&Ej2dfJ^LEij#t`MY;Q75Yw4SQyO7-u>RWb7TD~}_v`NJ zu;kT%e@}B|Q^BySZ(qzcq%>Rcol*NQEwc0%Bk>(<)#P1 zTdPPp9La_J@-Sh2FpUt6H@++gqwp|VE`Bp9_UiA{V_m%OqweIbAj%tTbr?F;q-2pa z-b5Ft8VH`uF*;2xS91d`tgz^YZJ7{85`(*JbU?^6d1amUSMpUexW_XU*vg5umd}iE zB)RUw@k4q=C^S-TxT*H|kq|SJlNj+TzaWJOetdr)(=OTKeC^ge`6cxXIS6s%rtcT;hKB<0}Oi16CBl`5@;pF$L7vgHu%%IoVzP}V=>)?4? zj&MVMk;uLsrVwgKpp?hJV-uOG$5QjZ+iVSvLk#SE%yfQjUC^sC(cg+$sS{DDppMGw zRuq+>(`2ORa$h_(1uOLl8S1DA_Iz#(dgP%;e~ov4C|0G;W&TC4&PYOesL^p8SH1NL zjiE=yyX5P&IyZm^+Zcekt;CC0Yfz&a>FfM9hRdHYxSM?|s#RbHRLvyy8HmenTz%mr zb7GOR^=o&+ZT7dbPlor5gO3^MuRy7URLubdP^UlsbShYZ4hxQP?IZY2(UaXnQIQ9Z zIsSCW)BOku7oY>l=JIJ(@!o5|b4Pl6Ct`orGv<^EUfR4rQyU_I{IWB~G7AZ3daznT zat!wJKNp3X9lj#?vVn78dhf9P8(l$shX@3TBYPmfm`m4kqdZg04%?wmuNk3Z5p??k z-T*;BzQ13XrX~O{aP7BD(=a4@AnaZs8%!bK)c!DL^@aJdIRBgn4E^FkONfGR{|D;? zf3SY}1tG4LSgU>mDs)9{CqA{-$!g)=h?02pUWV^xAz-oV`4>sBX4U}}?yC0nv_&0CtRs<2|~uTt*1t3mxqAm1aEC-wbXl=ts@64I}F_DEE?tAr?>++1nbOJKyU-TxIHiY!X&?KjB0Yjp67(cn=t8!4O z@hf=$)|(1^5|u-YuIXAobtZnCx~7YAIVMz$7x};DRflJ}B2ismzgf(K2+X>p%_$HV z904_Od%;4wBYti!HY4r~2h`t&aqV6M+ZDFMO^=Brg;4>cJEJ)p zZgOo}biXGwhBjHi!}wd}yBMnedHNSh(Mz=w13qb2Gc!2oMlm6+zGucKpChxB7-+@e zJE<{rc6Zdsq(sZU4Siz49nBGC`>R;1@;_K~OaH=M{GQP%u0(5nSTsSWH z588Y}Iu%~i9)sVNTkIP+Y%phBLV6;pIMKOp6)S=!xf%usl1-9cEja141}%-4?_Dk% z0WNLTKDrK(ey!gagk>Py5I*ewgGDD{Shv*wp-mPOQ^FrUwxk?%0h5-PUSv%(TNOge z^3ij%lAX17_sxmhrg&?}dAo*b`^lHq$Na=5Cx=$4g-`J zV3>IapVIR_=e*zZ)Auv~!QAUwYwhdWdoMEx%t&$Sz)qD91{N%VZjj-N4 ziS}HoM8f3Q+d7%*SB3546tH}&MU0+G(d+>^JL=@;87wQr;7%U|fYa>qav-B?kX)?5 zTeB z@K&afk%G_JXRFq%x;c-}?A0#PynwPvvgWAq)0+y)*GOkpCcqeZ<3DZDi;*wMIiK72 zV-Jh*iV|Vv1 zlRK=DG?U$_3xM?rKpLa%JsZrL79xPtmE5iiauz@CK9YIT26%2M=k9cqo<=U&?_ zxr5a0mnjDpQ%L7_hr~iLw_{2Jt#X~?&bq{+4B_Z+Jr#;0T3k^Pp2|d-dJ`8nC>1An zG`s`}s;tWSr}@@k@JXdwYF~jHL<_~WbEcb6xUNSJb(6i-3{SRC z+BURXpQ!z)F4Z_CtDzYxGnjg(s;sHZ#;bd@JUCvZ#95?J4kKqzi{fVDvHB1T z&=Ta{^>==B0W4kn$uVV|uu*9rM}M0lU2(;iHWSoHjd1h8O%p|MPO?bVRO?0iYD z*Db<(Tbu6p&*6FfbZSes@$9_M`19u6h1eUo+e`Vf-Y=SCsARj2U+&Zmkh&0WFBu$s z5pc@)@fe@%r~WP*I&;)mszM|7ZeK7*G|mD*WQN%;c?93lV-yU_*K6f{Zi+rfTassO!Tekz z1Gg`sA46HxC8kKDUk%=A4$WT`rO*iUUvwpK8~O(g+%9<;&iNxj7S)k1WRQiYffdyV zVUuU4H?jTj4#(-4L6aqcNC~rj4(*K7#dj%l$7WA!O=3)QF4!A84E9%BJdtD#j8uC6 z$l@#uAf82P((q5qG`kh*{-l9|9T8O9KES81yS?Vnia@9(VbJGGR1pL=^*5Pw*4{>6 z>^Z?$Ua%P}I>RM?!r<2s-JHXb6N2bcc&{2Jf%l@e;OWL&s8DwHz=)4MALr7vw1K6WD*`1J9s(`Mfra;ft(y^lDK&VI9VX>sTd z@#@q_3Havs4ho1(9(0qRCwa%gFTsCD`ROOtL+eNR@zX$CRh zg+n$}Vj;Ax?^DeecVp6YvG&B-_umk%TAPO~69bAQyZxe9f?KbvT^fv<%_lenijzZF zNjoFCm^`fVy;*u^@cvn_9T;Z;p68&tnIx-Kr zkcW(aSEaAqNVw_l^Q%CdB17Syhs=IQ2J@fDSWrm*0~w+M#P?amyBDqx;?0@%g}~(S zFfb9N_8uIICYp{ z>0xR^$k|PXP7q-9;eQK6@~&XZb`N}?;;cG}kkL|K!Wl4udalp%%g!`2D*s){g;$vh z>_nbH+$N(TVIaeYY>oC~4{gpi2|aJ}d`sjtl=$%UYPdaOYmQ74RTa^d&d!l!-s-&? zP}$>?hDgq!)E3ifgq43 z&Dn3vMy@_5OFh(5bK4PW&9MR53bA`a)6_7csq8xmta9a)7qH^X{$YA@hOt-B+cD`A z{DLe7{y@sgeb}%NjYE*uhYFaZ#PWg_mGNPA8iBL$NTA32N>1Q*v{p5&Nj}c75Yk>z zs3rrHRER$9nq1R7IwYDAYnn+F1lqS(7IH@^oo0h#ywsek%-qdOGYME<)#?r#T|yRT zNd|d@4a~Z87uS^$ThTnYPW8XjtBn7@vqr2nX?w6v^UBP#K#jRKLjjIa5qFOlUj(Pz z@97NRSnZ9KllWM|z<&8Exh0Q7Yx+FVi+gfX)>TSx)iXXF*1dpZ4e7B{trsjt5k!5V za{k=zA1vrWBu&S2y&qe;e7s2H0~Z(j?ha=K0-a&21L@yYneg^z?JYaH=-jifSrG?p z_Bh4Pg{P`-4ZzRZy70}uTdxO$`C|4{OHUTrr=>r)ZGuXs<0Cd>n5 zf)azj4(y@!$0$JomG;R}nE%LnL?YJ<|CA)TOEMK|@LhBSUgE>EllnrI!GI!QxFRZ{ z9>*l^R&{aI`(}u+fQd_ZW>sF&%ZTRYAXw?2 zMhR>Svv9`&^c3P@Hz%`8GFPMv3Vl7R(hLKqRQQ0ZU;$)mo~Hm!{(~!OjCs|hT_QPk zzpWy-G<&9={n@l&TjBmo%j1p3-^TuKdHgGPdeT?e(LJdlR{C>toDv ze_QNIxJq5Ip`NC zL=c@Ee$Cb`)wH3T$bIVmx>CO!c`eu!Ig#7}$O$cWl1AJ$?}@H3wy%reH+zYkCGme~ zo$NW^z84@-f9<}dCH9Hy&imO$1RWuj?&uf0Is`)as(fnfe2+sIzct6}29=={;#Hd* zri!kY375t06TUYOxkfR08v4CRj5!-}xZAJ0=46BDDL6eARrk;{=Qdx&84eX{g|ab#V8# zj}h+i-kLsqNi3(NF!;>Gfg%0)>EcVp3uy4&@BA(A#o*}eh$t+rkb?>G1XV))X?CI$ zRN3fkdY91_9U%%Ht$Z)4$HBb>>eQPYfy!0we~6WmM9f?HQ!InyX`}tf>7~YMN-fA; z{Hc>AC3mjJ1#oV_ENw-##$AYJeam2>7@#wtY-g8E=^({##ZpZ`%``;@5O*?4arT>b zzeVe8nyU%o<+MOVXVRH*N;IO(W|UUw4Kxl11)Wr9zq=WBam1$-Ih;~^{!y!gcG?`@ zhJX5)(_7mbD>kEAgU?#k$N*X~!Y#YU!y?&bO^=^dKJ4*p<-{Ku_RNHCfjyi2mp^w2 ziyGgf{FQdy0Czz6=j(sF|1&i3_xqA}=pPxU=C4d!YPnzMOWl!aJA7O@TugMUsqKKe z#we`g%dBxQHEJ}Kq!8=mh^V?KYRzm0cot(lMJnGwwC=i;~fp+kyZE#gakruk8fRGMGb zmozrm6|3grgOgvorRd{Ma`cY(UX97<;u;>(J)wreNxT@ZG}MO?xA8DFf)PO*Q9I1cJ=X7Czc3BRUmhs4_N0uPvHh+7poE~A_bCX`*4akdT3}99vNB}X^d**f;YPa# zTwZTwK0ujLmi8Wa+O*h^opWIUM9F7Sq8u;$00Th$zpgWaSR7FZr)QHJ71gz$1j5%S z*Gkn&6$W~vNqu~E=y}uRiX9mxecwhAv!w7O@CWY>(OF1z_Ch3G0(jMfBr-En!JJc$ z8W*$~{w38a7r6Ly^mh3F!`F$&>i>G1_0kUTUVN(M&D~ujA2Eh`*{eiLtj~gB(573% ztI4oxVJkmML4@%BC5y-DwOUeQ`fcJn%IkRJHOA%E@NR3M!P|i3g?SgkO2BhA^qf%B z*Mv46oL1X}>SwDL9E9wc?_1Oe=JWco3gzFj9b+jeq*|gTmmb-LVrOVMS?dSkH6}Xp zz=HQ$68zpt&PmU2!|AnVimzvm0NUY4vNA;>OEHCt!h{0QT+}1Uzn$)y~>hmUt=jVa#L!Lc+}uQrL%)CL2XPPC17YlFUcw`?&$l* zk_eS>!DWpjQEpo9Ouz#a$#ws(RMb+G1l8eUunb#mMbzcd|t}TdkyQ?{tkS49jKi9rh<0mZ+At&xkZ`DFE7fKnTiovjg5S zjQkwG+YfD449#h9M<$Nykxp!e4Ek3mL*chDEqe*#Lp8zUMW5N9xXtt{vXIXlf90|p z6kVDIID$$#>hUZ=7xGouCAo^Sdk?kr);aCbkKE(^w8`qrOi#gxgZUQt9k3ihCjdtA zdGgpm4G*raC#&phc~1u=mx{F`HtF zWG@-rrCGq!n#t)kj-0ruEcd0n^Fy3#3k zh~({2Dj4V3LL&rUnqxd8r!21uiYMx0!UZb^3P6@B(mYX~3=*q0yti&_e(`*7R0^R8 z`pl0PB()C;8_OV5t`S6rw2a}(X(!D}Y_s@p&S0a^T4<=5B&sr(aBi$O$N4cSqd47c z`3i{0yS1ud@#L|pCxy6#Qo%jm14j2!M&x=Kx9H${c}kP&T1BcpF-2wPYl~XbZ2!EC zs@9-j;hI*^SZ!@Z5s|KBAycl1aTUK2h=c|)aD;T?Tn*l;ZxeyoKXMVR5dM8wM4Bq$ zZy9))@=E_F=k!oHG8FHema}erXZJQIV~WkaM^?I)mDJsiJK+y!Za?XMYO8Q(KX9qUtr%xq$0Z+pebY{E zkS_LmD9*uHSQ zvuVT0g1c=PsT}aNSJ*^o0dYdNy+#@u1J4-jJVUYgT;|1lMCgp%=z(1TE#u?}2;lGr zXu(&b3zhdN-@zgR!u;vI8C-JaVAf77^U-8x3p}p1`10I~Fjb<=n-*W%ClJ)PScn57 zn&tQ9US%)YsPZ%DCSU&XPD!w8m#Qhi${4PF)t{W6OhnEhJwcjg1*AKp%x`aCI|Zaz zD^Kt!9aymPgRP>2KBO?(DQ_dU3pazh8&!6G2wk zKSnOQ{-C0tmk9nolB84iJIdOuew`Drz{AJjAK?@8AI3mP&y@YGWWwLL-sH=XV+k74 z*M^HOMm+X4iLS{vhDC{p9-i)l6|-&8diGRgz}l-sSz2WR`QiR$oLpkC2et7_&T-uy(Y) zL;S=w9L|bkT>tugt-ea(3=weg$H@NKlY_73p4Nc((5g1$po1z-4zFI+iv(-W zH*Brn9;Kv8!P^TO_cq_nQ@z!v@f*@3Os2;b4=dF8F0N#cwKprgIGTRpX>ySCT|VeQ z3O@V15yusRkvv^GscYKM8v+On%)-e|`+ufwh#hoNKJ#%#&}Z@4L`!-O@MymtT{XB1 zHP*~mSkIiK!xvX+kOW9rb(t1|!gT3A|3eu=@ymJ1T%En5#?g3tVkU!d%HohV9J9oL)Ck&5&_VA zR5Mz#QHaZ@U81EyKt?@MkutPS)WDNrr0av0_zQ0Jlqbx;G((vFBxV2C(DdIZp@*0L zff7;V{ok87*@qz|-(Kc3yCPOO$i{o|;J8m`^Xgm8IK6Ip_-z_C`SsO#jET zb2Fg<2dkc7z$<(oA!YOW*!kOMsE<~Q=iRIDJ z1?(eZC~i*DF%kiWjxkr`Y8Hd|$y9seZMxRe?nstB|ODG&`9{xiowaGV(=zo3Tz4nnJ?{g|2+@ipm7WRB{6fn0S-i|ySyoe@yz*@n^dzFE2WDt3K=a!WiQ{^zozZCp#Z4pE#JHS6!{@f z5gSQ)W3jSU#%>(ByA$y_5!xL3Qo6ROz8@O+eR}pT-J`@N#d@;c{$}yM->%%lNW7Li zW8D@SdZFkZt?0N?jh-fb z50CwCRJ~B?q59FzDRk$|e;8y=gBrB|*86=$QN&=L_Tc6|+wJt|WFu_w5!LPa=gGyW zj=Rl3;9Kt#2~|Syj~Q4D<2tt}(kDyLKjzSie!|JKW*nuYf%uQPN8zYZ39wjOGNV~nNyuMa7llypQX@L0|Q;Yn+ z`%Lb{#gm5}5Bov<#*c~?1X2))!Q@*U7UGBd?yit{DEtK585wDcAfT1A3Il`2X99oP z$;YKh!#zR>Mntnw?apav0d$eEEXGXv3Tr3j^EWD4*YXqUx>&AjWN|i^z<#}hSH)AT z!#^Jq8o1KGqgXCn>EA+p@pmEC&LCM}ep>*uc_u9NPOlb!NU|_1t1si2#b&t8v>BI0 zWetk+a;+M-R5Ky`Sa-6$NYi@TaTy5tzQ=oJ>xoxUHQ)N7&;KmuTM^JBv;?}3y13bY zn&LGJD!D(UOQz_WVdbl?9Jt}!Cv-i&nCH7g)W(7k0|EewPnI6QY9=awl10t zsXTV#lU)Dt1$s^yH$(rap&h$K9nh)nS`=&I7jsT2-?VA2f(;)?k%>2^ReobCubEk1Ku@%2G0{cMZ) zPWbxUa!p3$Jrsf9aD`!NFvR<-`FO!aLG{^JKFA-^-C$`dsC53?cp1id>n&%q$`G>f z`Phf#=Zd{!90Td~c8LyOr_nq9j%E+x^opN!wyA8P!Hhb6=!B(W#efh@gLLbYz9EMY ziGAH^ZpUz$cBJV>kVxpLc>8ux(eq~y`DglkP#?;2TeJTaN z<&w(8Z25A!ovL6R-P)BLR%zI3E(0k1&)(ui4WfLLTGq8f$|+Jr zQ8Q<+YK#@3l%e<;k0~oU1O!A<7U+deR!EMIs|qBS_C}QA&|nS~ybx@FTx1_31=w98dVyWR?d>V5((6!y8&J&t%L}Zml1$htH@GSVtz6e zI{|uW^OSI|(6up-y&1}u1YyA2L-WNl;x?-cvD;aC*%HoJ^b_CO{Hq3vUv_uOJtpid zOq6HbCpHbkaL=x)p&NzA6(Ri=kcAue=2aJXu%nk!>h*JFfyvh6TWcyE{&xtlml|%( z+cg1vO%qp<4iDr>(?ZZg-t&(xi?Va&SxPw+7JVgSI$=FGKUPD3#Lpq;vx%<`nv?!{ zq1R-cIsLG1-V>J<$zpic6>pv0ksRFA@f!7!)i?r(RHy!Wa$Ou)ORlpkW@o{JC-H_K zyxh0Baz=#rnUAw(o_RjZo@e%+J(^k?GO`XW zKnrlg4*b(V000I6C;-3#00{sn06_oS3jzUP5C8=Ma1ek50VoiF{&y${41mD^6b!(@ z01^zKzySK+`5-6&h5}G10EYrdD1d?j=zoKP-~bp7K;Zx!4j|zG3J#$ET?vE)z(@d! z1mH*ji3Ctc0R8XsAQS*b0ZK%xN@8bJRiQ3wct zfB^^;fWQF=5`drp2pWI@AP5Kq0fQh=5Cjf_AVClm2!aMd05AjuhJe8kC>R0K+zC58iGVa zP-qAm4FLcI2ta@V1QbBP0R$32pa29KKmZ^F2!sHG5Ks^T4niP72owl`1|a}20t7~Y z!3ZcA0S6v@yAPfkE0fR755C#syAVC-u z2!jS;05Apw#(=>XC>R3=V~}7B3XDO6F#r?;f?~i>3>1oiLorAw1_i~Sp%?&;0l_g~ zI0g#Gz~LAq9D{;m&~OZZ#DI_(FcJerV&F&&5{W?}F=!+PKw&^A3>bxhqA+k228qI; zP#81{1E4V=GzN^uK+zaD8iPb*P-qMqjp4+Q=&PHWJXJAN7Z(x5!TM)Iao*wL;sETx z!~cNeKQI9ZF#r}J3n4%_5`|ALXfxebF_u8V&8BNXQ8AuGEI_sFF1D+hhK)&N(dww4 z$>RCUqde15qa|)^GYf5~o}Vw`rAGbfSz`ZCCKrRvDu_H(0%Wq1kU44|#>QeLo7t+z zWgi;NIvqC?&4kgb$VCs2VHvb*|y^|ZP0>Ug=; zdQ9W2adWS&&{pw%m@$u6?u*KMDaN>*p@F}~LI!D}eVcn9Z%Xw(dfv`#tmI+O*kxwp zo`V{!Fc8+kfmfIH9?bYcbOYeiwY7eFBJ;B^NWSk?v;uGYBK9$21RRk|T`{OAUSTt; z(A)x3MlEv3m24f#_yZ0t@2C}fE!vkBmL6v9X6Z1Hpm`-UNH0EC_}S~j9hV9*nPRWY zk;i2;&MeX#_7Mu-xSu^Un#zpkqAbYK62H&2OdX9Ps8rf32;79duc@kKCkf{ak^2o6EJMO4hMvFag>Vn6&;y~9cR5Jaw>|&<9uT)_|jOT5I z2cL0)O=U|(ZY5Y~6kH#dWwqk+@>w+fmr%@BeN~Zs{O<#w*iV%V(RIA}Jayer^W6Yx zE(~Hl`#@N*CYXC$>T6Bm%v5#39iGrC*!_2vYkE0ZFRvY5Mv83nop@c?+pDV+TRXgVK0h2fB14={VJw79de+T|#d~6+g<2}x z+rQM?2@0P6m}kuzZ!;yc?m0@8T}k$dtityYoI3BCZ3rm7UY%>gNq8>Qa_aNjqM&eJ zo@G2C^|kB7q}}Q3$@xiv$6b%TWSXa)i_U!F<7%5LX7@@_Kj#iA8~xW%-E$3}hGsA% zY=d(54`qr}&gWc)G^J;3hjoRtz5UF11H5y^-2ysILNDVVe-hD~XJ4Jd7XJN#nLVxN zGZnA;PcXFX{mWTc3)=Tf=g0$({eW+q+aV!6`H5y5Xj8h<+R>lz#jUsd2!l6as3kzRF6VG$tOV^K(IeJr5 z4VTM_GmKE`dq?piCkRafRfwRYF`|3r4;&a`%@{=EgiIaqUrmp)9(zY!0m|5zSi)Fb z1|yip!yU@he1L~6BT=7*i?WPgUC^pM$zq_Au5)@qSWyfx(2c}}A5MrAE+pB{J2FwX zO-kkdNNTeBdJVYJ-+b|cULKXXMy^t~-ew3{$Wk*N;7%(sb%3TNoH=ohG`s8x!o8>E9Y%i{h^TZSMf%4znb=g6dk-zs4cOFm zHL=Y~sa~-^``f5_jqQ4P@l!VAfkyJsf(-m2rWfrc9r2;&jIuES&z~?@7HihaT zsG=FiE#LXDmB##lK8qsw6?gRg(=y?A9!mbIiwd>}G~WZ&4P&$yU;ougTdKA^zmIz7 z<9gZ9cE{`CIQ1@gd5rCfBf)GFKR}G zN?oWz1owrY({(!OM1~Q^^|FOX0xk+H!cnr7Mv47oy}D(VNG?g}z*8;f_x7KE_hq@NlgYva` zwqbzWepKjKjq2I)&la8~0i_%4rQtgq#z+g{=U7;=t1G4jVvv?w8pMk9#Y#`^m9SU3 zcX9|=;q4o!H*2W-J`Ws)YR-9r(tVbNT zgb>lza7DSPk1Z)xxZG!f*-j7-BppL@YHYe(3gCZ{m&;vnmYOHuc1RIu@iuIj_O0^b z=FeQBoBVHD$wfsuld%P~xgM|9Xi^WkxEU8k4GU^WaQ(L=F%_t=OW&S9>7~J^Y)0W@ zLgCx#7i?Qt_^=;AT8_=t{qI5E4&SA)bVW=?>0i!g;KQR-$QtjTE-vB?KPg{+mTDE( zwAXYzR~kxnK{t#AzY{5ESB?w$^yjv>{+T^N2}q~@&7gMsW{Q2m;fHN(3pMsB%xirRdqTq#40p{^R4}Xqhh^G3}U){Tp z?nNjsX?JSXe_Arny)E%NZH0aNjWpH5=J5x)LLCmg6IOzCqD7n6A{ME!j%(mTS2F7Pw#mfZ|Pv7_`}nu9vnXY3SlnkiyQDk zf;|8!bveuAa02d1taraFzMq6pJ=Ez&+) zq;1NiX~BU*RNDD%+Ic>pRgQOIk&cqV+l8ec5CR+gY4`bQ*A{8cFzU$(*5-1VHBP*E z9JVJ>ni{Xgb!rq_1Ki1AaN`9|18knDZIa)9EUXO~Y<(G~hvfHNJn-=_oFHo6TN=6< zX2mG(GP6MZ9G1fDM^l~fFPMx>2?ss%rOY8QBc?#39W@28zZ@A6Tfw4&OGh@(ZYIyC zcrlO2Me{Hmzw9p~K8kJl|z1%-_$9b zp_Re?BPUr(jvONLm_9JKATkJ+-F_W&{v$JH-0Q*Ff&p<|Z-05*?5s@70#BR*TgH&tpiGn3QI)v`uKWCd4KqpW zO`Z4(i+?JO8S%Y1qHn-x$uu^|H{bb_;}m_VkgSjjKWfkG706pLg1_}(P4J*k)GCHU za+(T?Df%KQ){5El6jLpw@)>g-f}CffO3Jj#zh`S!1?efu29@|0+V5IOBnC+fl#Oi{ z7FLxK2^1Z!ig)!f^>}3eEnpoKsEEJIuFWW8;+9h5E|(Z9UXv~OP#^{JDd&=jT0Qc> zJ1z$=%5BJ2Oi$$Y#XHPJSG;$~Qq8PXfR(;3kPH;?eS{NjN>yg~gyq*v zqq22@3}@|(=Bz4}*(!S?vy!_6)xzRnoQka}iLz)h-P!6;F)D*6wXXzf%yBALu(H4R zNp^ZvZaA~C_0|kwDl~PEsNxD(MF?w6`tqk`*(GapEc9PBB_q4kzhh(7EgF$bu}|A8Qf8U`80-DU1^P!3Hpd9JX#s5t<;w;5 z8}YPnsML4AQ+H75Yt!lTR?V2(rd6x76L8wvdh-`Syr1Co1NY{y!D+jKc*>E@`&P}j z7O7`e&44$oOa}Ji0(PlQeDwr&Q`lPF)>;GrP5FsfLT$LoiTHnD##@O4U}BQbZEp|T zqGgDkMTz4rQd(tTG&=3E49O3y+gaZvHRi+0@`0{fSkrJ4lMXTC95Itzl8ALmx_e4Z zTTy9O0yzdp=~4{uH7?u8Z2nYDcJH#z_$)IfrcpzVL6eyVoQ&kg%8?&IYEX6=$R+4L zFE$8iOv>q+cWl%>F)8}p)grC+Nw(XefWDhX|J0P+<4yN$Mz`6{v)Yc37dkx9BK=n( z&fRLA52JFWe<%s`Mhk+)%4(x&u60-VN=*`b<&E7cf^s6Qnm!$rV%0ib!^3lh`phzg z9F6Mv0{XV2nX2)TW8(cc=GD?XCW&(_W6Q`#NnygD9tKwMc95|Cwv=!9*5zX*1F61;3;^ir5rd_ghL_&Vd7Ai*_ZrwW}!7=Vwk+#pySa5PIDDwI) zj_iwyo{0`?6rvgldicbGe0S?bmWD;4J`<~(W!KvYKPJVsRO%G~iAijD$iEWIcT~ZsOI397rucfKP+;^_XQFd)G7BF!XN4>g z1AjJb=#--PKB|Ec}Qmso%48Y)m??Ue3)%@i*#zsix?;G@)Wib z29iD-dkjp5Y)@kIDSd$x`shY(Sx-G*)a*^3TCSY(JT9iFFF0fwZ4(=fl^6Q#J-v59 z{+YsdMAT!fsXKRX^qP8XY|33SWk%|htytHjw3wy*OZ6UyM`Ro6PhGMzQPL(JVzbAF zC|dvE*%)2&%2_0KD2UShU; z+)mk-piIOA^h70ug&ahg@CUPibDb{vl`^@u?#L#ZMIKxg+dZ<67OT1#KRa6dJN%v} zlirM;D2MtW&>G0636d!%alPerV0 z{h)ox`mU69wRdGKNY6p;?;vt=l^|!iy?DW~D4bobl|l&^9LSS8X3?9WR>tG!`VTB2Zo3%*O|hD;QAIVpA3Ktz>gqJRGIko+KI&K;>I#Z#N@ReF-L6{cuCe4(FxykH zWnu?P6@7hG!*Eq9$$gIOJqc4Cme_r!JyqgreS#`|dpn(5mIDHbef@iKWo3K=k^Se2 z`&Q?BCQoU6MY=RNsm>Mh-HedFB4WN|S=`}ZVNMXU2yJjO)!^?!Q_Omp_U#4>(M-zy z;pE9-;atU#aDLw4=CAFn0UBg|CT^eqx{haq{&raV>9WI{DK_&hH!A^yo$leH_lCaA zw1L}$LQG7hGlFdFB^{3x9SvKdC?>vzgQC)v5mtXxc*?C=>)8{%Z5 z;w&ZSyZB>P84p>?2F8h&h^hOm+6w(j^O=yKmkJ>by&?}ky{vXBrqvoIF3ek%-Z&c@ z;i!H4X4fvG@1(r+n7l=?1k`%|h=mTdR2)qG9WC*lkMPIyALKWf%`YEHzDGv=aJl-I_o;;`?je+ z@>~c$&-U$ARpaPSEuU`>jlN$BT+$d^?pQmRu6^HZxpbnuFy59I68U-A^|HI>Lj5~k z+$OWghs-YvntGUDL{nu=p2rzj1?Ts0gyK1;(ob7*vy-%+EX7yf)vje6u49QVDI=O_ z%PNM_*3IS~jVa#9V=9J-d6bp1N%v*cl%xtDQOhk5Ybgzv8o>3Eo4ShfWm=0GOQl^A zZyFWL_tWmIKYVVzhwyzpANfF~*;*NuPVN4o#xteq@z>QqcBP}3xAK}27Xuln+@dJo zv5c&@jhmO`-DM|38NuZkayioDKa5vC>33W2Z$G40C6UX4p!L8)tGQA%4ha>FMQizJ z_}lxxIF56hPm^dde>-j|oF-H0F<7_KV!Pl>p3gk|1;#w;#>QrE=Vgqv8WV_*t{PWQ z4&I0V^1H_Mq~_C);G#zU`9^-Qn7|(LKbivhYANi9*6FK%fn=9fUw6DsB z5T>n;UH!;|))srE`^%cwjcc8dF@)`#+`Z*`qlMXx+3ptGLTc~#8_8EMN2_qL$mNgB zn=_4{>^cft=lHBN#gT^7djH(qo6e|vGWmNB>eiqCx2p-=*a0kH6O~PV>6-r>aWnt2 zTQEfMIabd(t4v2oPfS3z*VilQqDSsS0_sCf-dmAa-^gJWl&~kg-ZeFL&%*}>t@JG@sn+M}HHEbS&dESY1TL2bb zFUC{3N;VVbsab$PSlZUrnyGW^NgAp1K5WlKzZrtgGLwg<d%zoJlv<@KUWfO9> zu4Qv4x=6siGq;Cc6}1>5w5V`!R84td{u;>>)Ar6Ew~lmr1Po7(>HvCBPOD(l%_)yD{LSf>aF@own_Hu743OYHe*OQ#MGatBy&GW z=so<{LCf;+E$Nj6{FIHAc3gR4@U7>xd=4yYQ+^}PqA!V&Z3|M>DME3U{^2}SE!O(n zzptKMyW;8mRLi-to#s&6XxOXGrSG11vS@#%XZ03PW*3aSn(8xJ=0HHd;K|j4WK89V z)oWg(M^}dgd)OA_lYQho3H_|Euhr`SH$ce0?-=Sh6IAEEp8N6_QW}^{SCRf<;bL05 zi-}}E)l_(b8c3Kn2yPV0J6|ywoD**4z<JJZYANqI{bD7KW_4v-VItc&{7;{>HCBIDaTDO){1DP-9vjK1Cx?CNkc

    *22uAKq$UX*A`Kf+y``8aos$tZy>-X+%Jr#?xWmollO5>LQs7T?t^jQ=}j% zLf>U>F+atYMl;UCv>6B$2s_l|*V{Q&-=Nr^^0G@FJbEbjs$8HBrjfvq8SfBV!`ZMz z{)})uWmdaNP;7h>dHOOG${EA*=tz_#O)Mkev|3!#y|S9nIX(T?Zk#Cfw7ajbzqp#S zR8JpRl?*qpeyn8T_s_X`Ll>CCXf^M%-gye%%aWhOF6<}jd9JqKlKP?hbZ;0jI-+kE zIp)WB=qzVuAM3+|6`WBA$5fg+g~@3rG;VS^)9@-8v10O!YHx(>hZCZ5dIGgtDfM1( zH0o#3*p~{GK=TiyrX9f)Jnwm@dgTMfGZS$#_?|q*n-7gjcX_V0&qy>oEWjJx+v~-SH?9erw$*W$n{0bhXd|A%vk6Vuv>t2q> z8~wE#4=3y;JZ=?Ik^dIwN^KG`pB)4m;M8OFHR|MT@{XO`5J`N+6ihT+g{|f3Ea!`O z$+%+sjk+XKGFe`cOsx@H#>1Ux$4q;YC#HV?p<7XqMIRIH<8GQ}Wk+j@HH_+bps{b> zT@%yWceaoDWbk%2sR-l;-$Cq62uYmlZuqiEt7(eM8U1R60OJ%@4(Hx;{*L606Jh{4 z3R1_W+OfeZKOMkPX;qVo+Bgo+qq!jc9|G%#4} z?S+t=h4LdZ_(SpVHWdb~mCwYHzBojqY*=7d#>~DlmwPT?1lp`E+=rRbJ9p0Vc(_v_ z^QqQ5ytDLg22~R039lquxl+@EVU;=G*e<`0EF+Vxgk`7n4JbJF`gX+1kg?i?&c2qw zqFJg4Mk=cBSE@Pa=IX!KD6zk-e6D+JHq@87^Lvp1i{C|*>+)maq<1Cj+{(A3;E%&1 z0$K!I)_AkM^Bz*29d!?krMTzbwRHbb%XnC*>-h3hqeUFQnOG8}tqO^okypB(;wC?~ zZY#cmR9bC5_a_NnAg`EUJ8MsM26i`Z@YN(K^p-9R3qP8^ypgI41<~p7FU#^i?%NJM zce6vW@51rl{vah(b02HadqUi(bEY4c-nQM7c_rl)HXZ49IV*#iXj4Erd{6w!%s8y9 zCh4v+yte+b`&1;TmcY)%?JFX)rnLT}Qut>d_tAY30$dKR3{RhcFIi8?Ij3JQH~N)@ zE?Ep!2{WCW{u6<6S{auEo)-B!fB9?OQQaG#u&dx(mS>^|GRt54#!DVo?_cA`N=+{K zhT@C}e15vaRr6x0KF0evm+J04J-c@{4l=niLxv+$EXn7e8sIb~jg1-^;3FAF@qXx+ z`hfP25?;G7D(hFO6uWi?@AP9}8eF{h0skCA7rhbv;*{wKzsA2e32eVsSVzgQ0$CZf z*|&duxAmtqW;%;^l^$Nz-(R#&eo%4Ud;Ts}B8FyYf295vb0ZPq1^jasGqKBWoWXPh zy31}}7>CYiGW~vi+0Wt7OdxAP{(u8;j7@|G!GH%=%H*6hR?_rVvYNhC0Um$t!(L$f z1;qznR0p38lMTZkEQ*pnQ6bxdd%=duKzn|#dcyh%B2J86r}72AxKSB%Q@t7{i{v2v zEK2s-I|i#B?;h}|!3$fuiLoSznx+a^mkxNn8n!KEb^noUuiQ^NAvCg_2;ah}u{Qkk zYB&!pTA#~zZzYlx{QM$5BDK@=ewcXIFv$JTIepixV>bvJ2tLk*PD@41^2V-E`Sm4e zA_xrGq};pJqiNiu)xJqF<2!Kgk#q1UC-uUT7{Zt_{t?>SF_i?d*U}1dya6p5PAAw3 zt1A>o0l|L!#wP)>;+H(k-LFIGR8xABj&ogD%-q;0*~;^v4;Ydpenl?DB>cUAV2vlg z0l)O~rw?L)#kdQL#UwFxJ5l+=EW1rLF#9B zL!wTqhF+&Un2eVO%RxO=ih8PeMT)QyS7p1%PgL4IN-Fh0j_rpq=}P+jZ8~A6jTb)^ z;iUaK4^U-qdrqtA<7fNe$ZiutQT35nLIq#(z?R2>q8vg@3c-IclYzaRan3_gwMDZs zfe#d90u8jGxHjD*nWA@@rZjZAxGy8N-McZXR+dPed2Bj*f76UkQx)0x^iEUJr{eXh zO4@hqE+d&&`}Ci-)3hAC`MAK68K&J6TF_~ichqK@k_xHqmJUa(R_>})?as>%o-8Uk zWKr21Sq8tkRIK-89Db>45E(|~E9m1HH{yF#yHNPU9zMUydCuU$#hKn4#51zojZncMJTq;3AvI)R5?R7mo`E;AwpZ zDK~MlN3S}T$mXPdoTUAgDT-4-#pRg9C#0(?WBI!3xh7itz}ZF7`@~SNJ^jlgUT=XS zgp=|xz4r-`L{O~S74T{pjvF~lcKwm8K-wP+5F&TYvd41K4#nBV(bBo4m3>T|387~f z;@pQ>o zn*ew3f;`d8*Q^tyvw-T6mMJi*-vOsdLOg0VR8f8>mi&kUz6T=SE zMR{sTw4=(iV=H-L65)hJ7!Fc90>ZtI@>z+Zr(F;qSyM&H%1o<>9ZPHM3|tQ_ig1CL z`)?82bqyN4qEdo|TBc4*hm!e0QM&F~4^|AlndFfD<&g~v^V|i04N3YZ^Jb4k#Ig(- z#&t?ydeQqT;$m@w7v^qpkGlI2$r8n70@Uf3k?=GLo`xFG%9Ghej4 zh9;pB(wk!)osi&T=K0d1n>)k!U{p#-Cz&@74D`4{>6OMhge@Q_fS26$jm z$Np2JK#3S51E&3TtWLx;hC%y^-qe|1LT9svz~JIxF=WFk5a@U88V#a{d*IhzZHL=`zCqV+jvBT zW#Fw$1k>rz^c!u~riYwO)bHL5Ev|O+4~$WdkAgqHWq2^gL*H%uMJKPz;nnTL9Npw! z>yG}pHxQQbio;~jhsNqQ-7HN5S8w{TIY!?JCA=4UdiObnR;V5m((m>lsghyZGE~!g zexfCgI(@q1k?M)_%JlT=ZAVgZ$|C$V>>%_`|6D;s?&9DKhv_q}Z%}%S4<=` zNO*pkUZ}_v$HkD+Xc;TdWbv9LolbYu%;hoG5dPwetbJD~<8?egLtXo!OxaR{lc%jz zsS2Dfs5k4^pcG;dqg|p6src6$HipPKS!sIqCVJwyakCZSzI2~5I{S4Y+;Vp`Oo(3K z%Nwn;hK;#^s`)~zAH*$k3_@$cVbnsd{BisnQ$Cy@Qdum&@66>B*`>~P*?s#^JjL}+ zH&~3ZC%Z(dYtBmG-D2Axd@|eKmhD~_rxOjGS>eH%tUP=#+n}Bw4Mp40X8NBy538oC z{>YPBcD=40sd?^pcY`n+-71j4C=CpvCpYiITiHbVx?Cb_bgkg;Q6|P4_sLi+uz54CYxj6M_7SE3ADx&U5-8fzN z`XBGaZi=D4_`c8Rsf>DC;?C{!YEXJNoF$33G~qV2dpDfNF=02Og_MXrIx?)s{uS9M zUPY$784Q#JJ_hgmE8L?T;9M^oY%@k}3%H^jJy&;{1AkAXO=n! zGi23JWW{8FY4}z4YhmN{Cjq9Gj@w8>C7g^oL?0RFkB5z0Tyba4r156C!tQj0TZ1bN zb=g+1tB{SMarGJpxFhWYu93gX#*b8*yMv7ywH`P>6MkoOTI1fWY{emjeHF8vN-w3h zHNE(RW^*QHn@2QRd6HrM#g&74t8xU-GRaf2&L@WB*u55?KU}*!U@ufFN#@J_G!yV~ zl#ug4Y|-#@{i*_W_|fi&xe&BN;R?R{SZ*(ct3}0?Xk4OEw6^56f~vI;1v6_zXx%Ky zU38WLi~}3C8I+6(TTe2E!kOR8#=(Z6aAnDSrnV-S zJLio1^`xea2nIerp9WNe_9KC)KGnZu`u)q?-OsNklsq^5hNhcZe(w$8aCh%w4+sDs zKGUJM&)&PtpZc?O5?`D)i)E)AXSR}|_mtaI4OH9NNjtVb&twnmbiaNtwM5Cqo4P-2 z!8Y?}oqJgS!}$04yDaqQ1oO}P4qso_=SgBM&pX`1;*a4O4?K{&?8G*#6sRbXjw{`l z4D++g^J5P`-?;}NeT+x*dBq~Y+~LCfRl@azS&CtJ%JXNSrRJX21oj6 zBRSZQ8oYp+JV-y8XnlQpn?NbBI#}wV;ixgUmHJyn@hqP_%<~V+uDH1>J@!+tGjRDe z+*>KwT|YuW$A&IOii=xo!4E&yQNMDeA0*=A3kGzSQZ$C^AwF_!HPr2BxM`hsB^zu4Tm3m*K;#C>k3eM_wnnj_+j=aav2T0yGjUd47=k0*QTqoB6e z4pNb!g+0s;I-h>xoYvbL+&COTsG}NkP;Mp@@{9s}ZmG#jz0Q8@Z#S4Zf$W z2)=V-UW83@_fQgz?|G#goN%Ajo0XEq!-MSl<_O6b-$t|STXH|A?YFR)DF6sS_rLr? zQNoaANUlJH8J~Vo?XDrfIOB?~Q=E7SHf343TdD6BRuplgb%t$xJXFc=gQm(YlXc49 zO@^h$7UI&vHek|*#djCwnGH35o6(PBI3EZpbjG=OuC&LfWpkz`Etm(lFXcK4S6v9K zF990M3%I5*ymw!3p6rP9)la>OHA;V!Tf_EJobhR| z(Xs?#9h$My)c?Bq9=7a{`*_@)Q>RI7FU#?qNqAr-xfoq>WPq#F^me(6wwgoRQ9wQ0 z*jtq&jTR8#5&~w)GK_0vtSj&x>)-pD0r9?ITLB2Zz6>3xI;_b#2SjZ(v)4sO`4DQ| zm{Ew)tKf_keyCzqVeR;Lw`Z8N#XEuI%QHfA&bP{Wea`viX?uwr_Uc)_BuFab>DXS> zyK^4`)%ch%DK=ZL9Rs}($PUE}j=jfl|J6p@7^ph%)f5~|{z7<$`{Ph-{ zICy$R;Ejit@6ye)SQQrWNtLs~r#9!%b({|QF_q~yf>%nEyq}`Y6)6Q`7(U_gjvAq_ z-R#Fz;n4S~dZ{7*PMF{=9;bQe9FHAe>8D1?Bd^f>njHWml_n)jpJwrsDWc7hrs^YZ z<}V>OnbUq~68@(@v%~dJ*_NFb3IaE@4si{v=@5O3c;7&^o6h z*&kdirK^Pw^wnjJ0Z!DA@_p*$M%CWEpwHu<8$)@^%%C4=bBb6l3Y>b1=4q0GJsTBeV}Hl&oQaWnX-$M`Ke0xO14#@=_&PMz~rF@5w&w}*P3EiXSs){U*q zFWMw?72mG#YEf^l>i|L{sEKokh=q@bo5YJc4gFmiNtCq`EKarFL0eiU<|d9nC0b|N zPLI%QKia^y!q49%?xZgMTqD-qKYqq7M&>Y9wOdyn&qVav|0FNo2QwK-A`|=IsWPFM zPjg=EE175%B;lb{ysY12!jXjefVidHxIcA4zOU$Q5@PcRUX4jTD@cgPJxs`8aQ*>$ zC5xYc4NlE@+U$Im}6qfPUON3^$>qL5yt0rK4;bxeF95Hh=Lh) zenJpK0R^dS=K6l1u~{~&gMid@HiKB00Z|UFilY8nl1LDxWLhQ&O8Tj9@&foG%-uDBFoe< z$6m(U>4?fYNWi}!Uz;)aY&y@SAb+VlU-U9XL#b{^eSA*oMcl5A9A4JD`{)9o&lwm=xE5RKeA81D`=4|`p`$ws#)~VrwE=!B}thv$w;4}b`v6<{$UFRK*|5l7DPEC#zy5l>(jNbJt7rYbUNBo-roh%k=p<$Gp9F-DC-TJ-AY2IdonZ zh&(H()%!DMeJ6hSwIC{j>|FZnY?t(vP{3?E-G^0J6?^H+{I*AXC~A?fEmuD;P~#&J zn1FYW-`qv%s=t(0pn$z<9enU6zN8==z5eZk^&mk3=DzNYEG?BvNu2psU}=rIW%yb% z{hG}?h1GR?YHflIhg&gNNlf4Cse-_kDE>+xCaLKPXZ$F$3g|;S!i_MVLWY6}$O70Y za-Nd^OO!k>Lsd$E<>Nflr;4)O;&oZx^Z+G4Kat}D8@y~CW6Ai?3?Xv5L_DX?{ zg;lP=>xJBXRSOa1R`d&Xtxrlaxz$cNe#UDCg>H1e=G34}YI$(@JD&QCpuumInGa{{ zsIC2%sSQ^HLULotAJ;dyzB*>{nLd7Q^d|qWqP#Z`$ax< zYDd+hwjl?l6oc9y%R2Cxgks~g`AaHw9W9!TXG=W`s=w0l=k{Li)SfD<5RQ91!;J~H z7$KBitM6mFn%PgSGt*cZ&o_dK4Nd!SQS@PZ&x@(BlhgcEX5!mxpC(9c64rfwO=7|< z!mhJ3kF%glF`2wj>}n-5C7+>&4U|U$SS)TCdDyt2lAVguXGjef3Z${V>4|Apu|b*T zck2gN;_y}b=+q3K)g0l-HCJ*}tEr?u-dfR1C$|r;Evrty`F7S8{lj&BU`n!G9FtkQ zmBXn)x1(3$Hip|NH^L@Ho=bmUHzo?bm3bQX$nSRBo|vX=uAuwH-9Elcc*YTK18b*8 zP>U$w|YzXW_xpxC{A5gLB5GWS(Aocc|M(jLcRnD_=a4ldQnuahU`Sw5*0_V%N4 zwpVc^iG9UOdHzsxhDbm|l;Tjqo+FBNkW0Pjc!I)_GJ)&fodCV{P2rX^CIqG&d&*Bg z7!#Q4<(Rnw9gv_5;oy!s$B712k0_sWs>EV%vD}eS^;-$x+Tce|M`9!iQ=9=+rp(A$ z{5}Fg(0nrZc(iP@WmN#f=<0&cSq|7tkNcnQvyq(I{UlU<^Kj&W8HNg`N#;0$V%W1{ zjD04WO#tu+w!tDP9Dqv@mgN0$!tx)}Hl`gGdR}-4a51GI?}tRc+f7G`>=2U# zlp#5euF2) zQkk5o?%=0P>)tk9<^1aY`jngr)boUxXeQO`c)nkzC0_^3Fq3>Gl!2aKik^IvRXw6S zFY>(HV57{_U37Ubj>lTB>UY>TNxPRplbUg7=SGvPK@*0;ovG>0lGeLg@TQ`DrY&*Y ze6CwvW~EbHyb>stxc;kL{mxa0kgHE+rCA_C*GR+PjDoLJ&kb*TjjqJ3t9ISlGlem6 zy~DA&(6S4m1ew+9wBiy+EjD6vPMA+2G>CkoAMgm_2v}tR-PtM5*jsALJHhbGoKd&1 z`9`377T)~3ivu|h4#pL=h@#1Dgz24XnAVEN|MKHX-#^rJxv`jVR6{T}G`UD8Q;w)) zcPj1{X{vHR4c4p8vo&b*Kys|CTs+D()qmJynfwXCkGZK?y&yy`?_Zs?`$zT zM8KCKz-~~X#WydR;-x6PTxZzJly1(AOSPVUCT%`{O^MGRUf#UfB5e(v(%Jl3cr$RQ z*BZP!w|U)vGkB5O8v0G=>+kiOx3}+F!*Axk-v7QC0?67T33RuxnQw;)^}QX#>g+Gs z==^Fx-giv84j&a6MqV_FMpXz(lZIG~rT%S$Y0isN72S^W$G?D?%sa+|uZJGxdMN|{ zuZ^bYpBnZ*Y&5|Cu+eB2gWD_rTkrZW8%+}E*2`WiiVv@vWu~KME=lOwY_V2n?YknG zSI1vwM0sap6@!QegF5zx%AaPfOCpT-7itBn1WSJZtpC)g+u-?PhDKumS#$3g@(_GQFdcpYMf8M8Nhc*Q~dTG2Ju^02ZkQ zctQwwf^S>*sKj9$OHSNO+j^>w|*V00VcT*xSHy zp_^)YhcvEthr8+Gir(W`x(u*t@HP4{lA1PfWRDU;lenAB-bKHQV;-u)k@85~Cq2?W z9J!nCHSjZ8Ntx{;63dot?LbqBZYvTPx()n>WBO%jGcUx~?_2uN7n8b_xZ{F)b+NU( zos`JRJ7qpB|0e>L*k6pblb(fg4J$meQE|j6b-rm(Bj3baj^ng02hRz#4p_uxw}z`} zW8u;mswI(H?0B@Yc4>a3HJ4vFrLE;kf`T*Gza1lqD5no%P|*zK$gfwb#o3F$slTRf z)E2>e@!K2E-&{-zeINL>^8EYY?Lp)Bx9IDW@4Dyu%jp0PhQw&(Zxh$7GRZ3FDBuc= z+cH9p#~nFA6`k<;>q zBf=-Ammib~gnxZhqkr*h0o>C7UDTC6`?X}KDSWkTYX0KtlZ9LJ)rt-EU+-0WxxrtH z$`P#BYwmT;*XyQTS@6%$C7Pu!pVb%A`Y*mT-)x57;49E>WZ~ofPbruKs06(JFEs2& zBJimNX<#4i$Kw81X#Cd)WPTHb&3~YgC+Z{;dTaO(G`Kx@n+g`bDwknL?XYNP)Tvcb zMBNu28t^aFk>TC?_4^=}8;KjF;%0M%KDF3WCee!KZY!5)wT27x|Jt-{^Padb(plv< z?#9_5?tOj!>5X|En)oA0kEmt*>l3Ph$tUIpUz zTaiagkoUr$Y;H^4K6ot=`#n$6KKfi2XX+#5Cw!7yqCjB4f)<@HFC$}-9sJ_=isXCS zJ=^d8;L}8SQQ79p!eRQxB`-4!yHeiHqB(B19RJqVEDrZ)W{V6q#o909&dm~~3KT!1 z2=r|(CWOxm%&c&}HbB^Q$4JfdV)j`dfBZkJy>(NZU)wF%H13VN26qS=0(9fI=e$$rIWzC6I#V?t;NE{+RckNV>#CbolF8;PYX^#} zG~877>L9@#I|`oTAxV@GJf*(sDCgfiZE$FROeL;ggO;Z;-V}*xI5M#J>z*B^A#QOz zqnea?f))E+7D0Jahj|<&gY2P#0)*mN)isttx%Xy@j5u%~fLAdXQ?7WBZL4O02~ry@k^-q7XLJ9}UV_jWAKK*p`s5Z- zW}@qcR+1Rou>`JBE7cSq&|R^upiPL-)Tdt0)^4t01 z*6u}MRuOAC2UeS}cC3Tnu!pvgL(s39*$KAPI1v^sK<&%kXx~*Dy&V<;S?`g4X788R z{FKTCNuiMdlT)C*!1`+3Bkh4jXi6mnYu!^Svj*`7qkX z!BW$SwgR^Xq={z=Of@a*Tubco%E-k~QF}AR)^T|=%{9(;JHxx?csnb2Qu|(U>F56n z4aT~=Ma6?$z5f?9e6H@-OrzKz)-5xg{tFt89aj%quH)>F+mZ_0M&SQ=7{LKh{9|;I z|C@)Ae-Q=610EOmf1%;2k@nwc2tRE)lnooeD*yOjX!uW}XsMVh0EeMb#NPe@mjIJ- zjaYE2)FWW|JbCBym0ED>sPYN&ZLHI^#=|VFQEhC?CGc-CMh>b~OO>#|uYA+(&RfmS zd*^-j9lv(^g1cU>6tt-~27}K+DHOFD3$H`7iErnJbOk*6Fa1vC z;gV0v^S8Qq)fP&O1hD!ty{}fA147%84P2a?TyL#r}RGRrBxbutCD(pw?mXLSLm?f+DWvH2CStc0NQ{ zNDyRuQ=|ojx?(=7K-K(E88X!f^vcCGl_kqk^tm_4#MRJ1jDc^S!8)7gtT&0(^5$hA2Vy*Ig$t(jurkM@GH&UAZLt?!tc3Yc&Z!WUu z5xa)NBWh9)vP!5rf)T&%d;VNhL~nKOS(kR|X?$13HYn0DcsWAy?fdN%0W$t|a0|xy ztlHomy;S>G5B9W&P2n}qI7XyTHxUm7gNl#~5}VfFYM*c4+=@=$URqy}De`Z?y<%nn zI7CCISJnY0nAU8D1l3#GMG<9HDP3yXYcakJmlSYoH?cQC#j$akB-a(ccFSy5)GV$h#*LzJg&?f&^%Bd<> zoN5(K*c2f#=h>)&G1_F#H*2bqw7D3}Cd8iz)KbGua;)t%#CDrzP@vgl96C!pW?5=6 zt8y8Eb!jZ&;zXJniYb0Au`KhUtV~E+u>=qmhnZ&tDa3Mu+ENgOy zpB$6!M^Hd+iV~X;mxI&1Ny%s}6RP3xh##^cCo`KVha-I-pfopoh2*b_Btky;vyeYA!g^Gz*gK`E0$%S%c(u;Xo(o&CBQb$)3zd`!z9z9&AdL|HwN{DR&IPP!E9m><()1T1OL^MC zq=Q~>(njcxuaKH0LQ7ke{Zb=&i<{;n7R3!v(y_w8CE;)L9{%eiry^u@ASb|n3AL0Y z!>S!L!F>EnsDTN~&e2>*Rw{KLxKgY%8N+n_u6z&>$}2wkZJ*0VB_xgUgo9-|LMa6d zOh)2yjnx2C!{H`eT5%ObPWcdM*q=H?@n%hFnNmZ@ifU!wiSIi0U*2h$2;SOHxk+mRIA>Kvpl~P+zz~Xi+i@lII0VT?sF(^`fzfWZ zfSGk#WqZ7p(c5qBFTi37*m}6g<$|W<;7x%VdSwq13p5nhwJ|?k8aH^bZV0xCj~WbJ z)a8D^5R%1LdyDbYN^Y+Ua{4c#_iBUCr;g+r`SHv>HrwTu$%PR+}OPv*6IxN-Y+~_^% zi)60xmh`XBr??}|2>d>p72$SNpe$76ju_5H;I9cXNW013Z?c*^F~X+u1EPbx<0#12 zQq81{4EXpu7BFtOT#)xFKFdF|n%oV~Mk?PGU6`HH!>V<(K1PcLc-n%Iqan99C``Zn zj=L@IzWMw;Q44vp5rco3Ly(M-m)DR^be^QGUvWuFJjXVf*_0_CD|^=@NNotexz_)k zPa%~0GK!j-0nXRbb8<-=j#rg>&iesetOW9zvCS1UPwPKwD)NaL-?J}0Ne#jf zfG#&94x(Aw8XSH9h^y)1e(f1QtYoFwPi30vo#-)%pd2!!Q~bY2^HewB8?Nm= zJMFWc${vsMf|euIs1GxJ_0aT6VHX)91}!bJ-Fl(`M?|~lvEiSURd^yF1b+B`OYQe; zHqcE~!+u@2${!=H>RGir>Dg~Me@GLixb_I@+Ib5EeSF3{TRnP7U@G|i5qqWC9>#N| z#SVa$ydSK@6iNN_(@RNdcED3-tAyd#Q^YQF8DO=iVQj;-am=iNUQa-+xS;?M?kxgYWnrf|Ds!%PF&y#PfUa8S-vEP|$t66I> z9gd|?Xs_L9v0bV*nr^S#>To@n%2ViYR!T?uy}9#8n-end2_BtNv9ob+I2w=Dc&4-I zU>y2RAz!hp`Di-py~FpJu9lOzLWwv^rS8_VCEg$kO(*F6Q2@POalTT(J$Mbl&7?M+ zm|r=d*#W(SP0G}5se~{E7i`}($Pb@~9?a3$1>splWuWR8?1Tit{$j%LRRVj1Z+1s3 z#>xBp{ybhEL=CVFh`!#z1~oFM=lg@6<6~^*-A>+4r~;v=vhx80BWr~rXuuFvPT<`; z^$4J2` zxyMK+;kyuBocwpbQZ?4%?Up(W>qP(>PK;KH8}WAc-Xg=mo;k%h@NBFEG7aZP^@&s4 z&N4x6RW>tQoifcLf}ga@Dvu|yYA?~iPjEiT6={P~frv2tu!yidhE_b#{z$(>b?EUh zNQ3f_K9kqD`Ya#lM`vBI7D3Dub4#yjQS3nbL8tJ(O%|G{T@LfFh&M(XvC|d#D@2(q zj@1n+mMP9*5?72pvP?@EV+N#~T4&j1&RfUMSLpNUGt@j6*)d1PHfLAMk7kS(RIX=q zs^*{W$DEjqv1tAf`TaG(+ILT14yVzneZ;H;RV^UR!91hps`(-Et~yVe=a5r7A%({E zlb@5UO^5tV)oLg903!(0yOFa8=B~c5)@PeUGH>6T{F6U*gLyU|Ay=&RtIe03gAJN9o(z%%2VGFdCRfhsvh{ zPBOlu=pJ|^d`YSO5nJ1T#xD1SU8^@%`R?_$M)FlR1wBntcE=^@SJ42O?9Qt>cWbtk z;#fciFx98mf4SIiE-t^XwUp@i>?)MI1f8`tQ}id|v=5Vcit%4aQj8=Jhf%R;WY@i) zbmuGr<;%*MqH#c`jRO}=2M|xN&!wYAiyR0e zR&#k%*hjwY9E!ceAr1y32wg5Df|YS_qD|3(No%7VA3&jasN&;U87Z<4^PiI&qeRLp zrFF-w!aOevdH%VtMe>yZAqbMRP}TOJSFUsqJDQ@3=;}77A0)cE(xepJn%~ifkiuvsdJznnhS77R0MwHlAGHqa8&0>W~SZ! z7!5kc--rokY^P7G%aVP$oKW)%<)pzgv*M}Vpwui+p+Mq|6&YqM`f+nbzV;>=su{A8U--CBa&B?Co|2^W}wXBPp-wJhR> zd)rLro|#T<`nNu`)hApz97KcGIbLX8>j?&bsD7EvV7l+<$K-IOxvets9d|&QB)C&- zZfwv*$aZFz+fm;H!6732uU0%45b}Ru#dRD1MIo&2|E7?rw{ui{=9~ZDDWra<$M^Pp zXS$=|$3XD^35BTq4G{WpH?&x;QPXbc&;|*rG@8y=@?BCP9J^!5nQtY613S65^rUI2n zHJO9yC;k~V3kFZK)pU@c$M>0-rSi!Fo-nj@xy3?}{C`IM=F?!aP^4P&t+lPye$hmz zDd&~`pHWv_Aw2yNSgqE}H(n|)+i1Sr=&(OPLuAdkH0X4{ox%`r`?(XOpKQOPKy0ns z5mZNLyq(+t9*Ifk$(Co+|*y*#nK|%!^ zz!LpvadR}Qke3J&(Z3JxLPvh>6ZU?x8}AuN8en~Vdy}H4@XG2UD%9D!LfF>2Kl3cq ze<67))B%^aQUx71=|G);BBP5M?R<#4;mA8tG%-x&y1BA!Xw)fS0<^p0I9}Z4;#ecC zG?0Pd!|8!I81ZZg3MRo{P2zekvD6FU%U({APbpJ%K~fX?M*^6}6o{eUlB;$0)!Pk)xtLf7HB#gi`tZx)px7PFw`BFcX4P;jIN#d76Tj4&@5W-qWS}eAgk$<{TKRyKILuZ z0r;a2MFEj4yv;oyGhc~gv4tx@o!mrkjtwNKvc})gn0s%#@5t#~9PY+83#XLYYuOjL zd>mRJ{V0P3fr4b)^WB>%+qQX_$i$xRC@a}@qj&*VJT-E2=!f={nXoQUPu>+q!z6++ z#mOFPQXVuT=jQyT$IfEYWW!SQ<^6YWtiSJLT&DR!24)J0+^wI^bL=u$y#HD%fBvEb zouiU+nqN+3uI~htew4PJl6w{}|7E$uG2e(46Ih5Zz;3suHbb=jnw9P@4f<`*KnGX$ z4sFolewmN%@KF*g8~2e)2^{E<3uOh6z&DA_GXMt%V*cEqYeoV}z8Aork&QT^Fl?Mb z!jHufA&T~B$dZ(8Pve6}+l;^xP+Vi$p)i{NBvpT>DH5|S_K)XcJz~YB{F^Z`4yh1K zzn8)m?Me>9?nX+M@FJ14<-tf>no#0!v$ftbPG_s@Fz_-67xplMb5Ip!1|K7Cqdzec z-gg}{TWpMp34}B)iyKMbY7zb^VOc)1#LHmGV1#po&#HSpVs)^lq{~TBVf&OLxUdd2 zrUiZ(LH)@{++Bsn0oTF8eZxe@{ziPD!TJl%o`$Fo-iYlNU{B zP1)A*Z5J`Ro91TJ<r8;WpDf5G00_dN=h*I!!wnHujGo`^MkzS>((Vz= zY>H!{{_*fQTe5ZbWg!SkU}b8g2rTihFGIO7(b?pTX?5ZAri7rE_VoId%MihruieMV z^}e2~vWQR}v91|Ua}s2sR&KQO#1g{xSUH`Q;t@%PhxWrmSz*l;#?;29yfmEeU{_mo zuhGW1ZFj;?A*V?X96oMHeph6(7AbhN?HK@P8h?lCh1tH9kDeo@$S^H#6QT9ee2n78 zpJaHrNR{z0EjEel${`ZvCYr2VS1j1mFwSGi2rc37D7-}zepkE5_l{=Neamd5!Z%}_ zqSvvgp~G}u)`P6lhl%&Js3F826aF!~Ou8X*k9#iEk9g59DB@^OG|yB18_e;@Jb4@* zUHYc^dvR`YM^dx5or*g|p3{a)21tUi1V_qf@~b^l>&@p9_Z80biE(NAQ+0DPEZ)t- z=A}E*(vkv2kDas+P`cVqr$>)WrycT9!7n$7Sz{#&g>uk7Z51uj8;^Vj_ET?ymjxT= z@{bFK2h!8L9;T9NTkDK=4X7{WmI+(c#IiO)d)X`Y$E(mkuP47qzs!8BqfuCnvsL-< zr-jO@6--{&Vx)qy7;LUx#hl;>)iY2El5{Q)>*G{5nH&-X-h(A)8Py$pjVk`Sw&=WF zLcNo&#sLg?N=$9ecta#`Dy`V$GZ3&>eL}L54d*Z;;6QIt@R$)80|%6cZ4+ zZ$If<97Sm+O6ogK%PBNTuauNeMLJDk)U})}aL-EZk}!g!SPQDFZTeBc5`+-`b1x|Uz8**_QgG+bhDL|Ar#$D?7@wg zh<-QeeR&BrGW`2WaoQbOu;~Y~T5$F^XYZ1~GT+dDk08aTRiO7Jx_zs7ad=yaFk2h5 zYjFD^RI2iAk#XrOi{u@n1>QTp5|`%`PJgpcZL5LP0-!oT(B{6>W-v-L2$dun<#Tah z6_9NciS34nV0}Dj5J{aYjiDgZh(H7TZWntqSUEZvqt%t0;N(;3Ee~oGH{}yCtK{1W38C~Odf(VvE4c&ag|icYKf z`5-JfpgfFhjnfW-H1-GyMsjssXP#L@!e;_(25WCDM+^pI$PS`)6LGf_M%k-{KD2+1 zJPm3KrLQVg8l+a0f=$Q+2C+JLM2|9q_ahCz9TAAF2f8A0Vjf^uA8H(O8P`~<5h5a7 zx^gp)D&_Lnk4JhJN_pBvNB>4vUW6)LUSsD#<2r1$5((JMCH$eZ0$e5Rf5wf>)eJ(O zB5@@^c&ADPZBfBcl*%XUzDjkalaT$#U}dIIV@NckZuHKh*f1ku2MWq0wQ30g$X+6w<%&V0}M|LV<#9F2Ib$ctDI*EtY(^Q?HdXc3)$^)3L}Sq zTO`4H3OIf;fJtI`(Was<#9kf8stBe-I-r&t&}wnFfFrwmfRMDNuib5WXK0WBDqg=d z51(EN8-leOR2XQc%{(c+bcnMrkd+3P=s};zw4R#h9hu1!kDQ!V?#Y{+8YX8UsRox0 zwoQ$fOn(s!{WhNd9@XK%Cn!fYgX&q%ypw66nDKoCvondnhE*6jDjm}${+xAH~`N7>L93*H}Q5|c12Q|5c7FE)xp!yf6SU?R(Pq>KMo>e zIs3MJE*l&MeDi6{7Yq0sPpg%4XRA@7nz~7*r(U^IgO$2Gwehd|r_n$p;zfGxdb5!e zDs^mIok55r;&SSrRw~UNx!?YM(~qZH13{cGFWqa->qGZgTYR~l1^R=~u(7-?AD8b_ z{_m6GXGkhnQ(_ikMHM#L^i1+Jrb}~4ejkl~GW(;YwDk_E4ey0KDb=`T@y`4+hzfKjf!t#C(?wQGc~ zy`Jq2{JlmQOlN*l

    gw&(Wd=}YU>|Jsr7|8A3SH_n;BWhvy4?(X3n4-2`)B^!TRrr>>hx)|>ED@l(L-jA9`c~zhN_^n+ z?QXQUsBI~`2SGSHF>Qzk%_D~UQmi`YWtk*Q=YD~ zpm}G>w(L?~4cHq#6@z+F_9jWh=MM%7ZnFK=Ty|~dhE}}kMSSPkm7^E-rdUC(%SfC* z`puW8*+{5Z9DphPC&;UsY2^6Z>U_2wV@84=Lk~N64A6q(&79b#D>(>I6cJX57|0n)_Ka296}E0c z7F>uSylEx?wjDB@=h6LT%}dEj@a^*x50SlmF=kq}Yp{MHfmQDu&+s$s{vlVf0HF;bOIaSDZHeq)z+G*B-%V=McrkxNEdx4m#(0BN0!7 z(n3ZNlCL$Ck$7Mw&+@PIITx&tQ>ImgSwgw#O*GUaq+$DuxoRkqDy2m~f`Tm=#YU59 zT;etGWmD3K+oV74Lo~%e4>IQLlGZ}M5mgu_@k*bDiDa9Q8IVIi$Z(8AC`zO=X?d}* z`QtL+UegVtHOoVxAvC#yD9rlYP_1^Jj7|h%4#b-&i+goI1AgKR8qdrChDR1E2n;9e zt_UX^Vx%r>#jPSYn{=q#W;Zle$5!?ZDf~}JJFKg02RB|7?t4J z2>x17^547-TCuc|_}c~}^n20Bherh#;4VsL8RF5Ewk|{Taf#?X$0>KhPhqb(3(o3y zLAY+O!qQWCgFYVA#XJQ z!A+&k$31|pR1IA1H5#{_Mk$PrQSeBA@sY$bki@Z6Uh73ggXEA<#WfgiF;*L>yH|x| zhEC$oGFR>GL8Cy5Jp`Q8jZoB}-t-$KK$+Hq_uZrTpkyURvZWJ(q6w&Mnw~}yph$C% zZQ!wru145LK!%a;!sRMm0D9+(o=4JJwK3}j$Y=#R%%^q-aTDLUvSG9!tISHE81$tD_ffBS|&w|&4Q6J+5tpQL)Kp$g4Hhx9O4k` zr6Pn4BRZMmV(y&<*?UxvDF@+{U57;i_fUsK*Dz3%vjg9rHI{!`nL{vUDe~4RI#Jfk z*}P&x&|MKVw)ZJ*s2$-*O=Cl!yjx4_?4^ZszU-STg#AckjX&Eo3+Jt%Lz2F54*HFX ztRX;CJnQCu@?rPa194crhY_Utwt^w=2cB&ggGt*B?(&PSv9fPDH5PrB zcx2&wh7s75o#=!W;!Pqt#sdocbvQuL3~1=^Pg|xWca}PzmDTum zEREQ25-;jqv%R_kiS#eF$m{Cq+f46y5^+mx+#DYcsp=rP(EY?<+0MTwP)&rW<3p88 zT?r$}_=aCBS%M=qwl`y|))G6ozi#*=5g|*EMtLp2;FY=s^~E~T{8Zg;`|$h{Bqh<8 z{YhqO59um#sU`-*JfCu|n;B!PmAxDW)uV1=ljTC3M+$L-0L*0mfQOVLThT4?`wk_! zEGDI~p_|p~gO~X5ojVouV-N-P`Ypfvb1FzxMHiMf9W-g%Hlx zGt#Cj+n!1E9d60bdQP9WelLi9_%(ciT<=KZ!O-)UDn>>`hBWE9;AF>?*~jgQ$i-*# z7v;2Wl5P!$9)STyZcpE8y^qx#=*L?A^TxNgKY`e9-s%G{UEPP^_+KjdoFsqH+(co= zMM(k6=YOAfWul*wu|8F!4;ZE3qdb1DdtECKeM8h6m|Rj}JwwiYV=Q(EhEpEbQPCu@ zdFuE~R$=lXBbbH}@N(jeg~GHZ5CU3PL7Y&7A$atj$bCpbKo0lBMENW|;?l}%R(9e> z_!RUfGkADE;JtM)DT;Dkg)-}ePDy&$k1;})^^j&CV|Ai%e%~-j!H}>HrXw{G`7F@U zg0EDT75-y53PF%?2%e@b9t$t=a*ChKQE((M+^NVNm_>?#Opl-uIrp1>gpYhPG%~jc zf}}w%(@7!UA|<%!&dEb%F+t&RO8$657jR1Mn?+an)r?>K9% z5>1e-8`pVC2$qWNT9500iu+v@c_+?!qk}!X9{r#DNYwv`&4m9huI8eFhGAm@1qId2 zhyVRDP+2-Nmhea#Nhf$yGWn0#e7R)|E}PEaGMS3k1ECHDA-k1tPu>>xMclt4Qn0i# zEtE*Vy;}RbDQ%8cam0wVNJ>{oBr0tBoT5;@WDMiiGtcl@oOGS8pYWp{MAh zu2%09`P8AUptsu&1>PY53CVTA10hjASfqv89JCt^-$@`{TwD(7TW=65P786Lj23>T z)Z}Swu|1U)L>snKnqp`-2mq1%8-bp{xkqhyHP3E0hWLY9JZ>S=fL%ohe^~1=bkbw!Ty@`UxeA6$Q7K3t*fuZ=E1-}% zIy;A5X&FfmQo}3qAQDC+i2S(fC~)eO>NIffMI^|CWF`A#hzxq37DWXEnvD&B+$YrX zL|D9;!Ang~Ka5tb)|R7qr=n0mHp#x`zK3!T>q&`TQtAE(X5Q=m_td)Y5@e|tGqi*w z8Ftg)w5u{TP?^t)GLSZjkIWgWdCmOrK0KtmpsJ0UMbR+q&52DlqAI4EEVi3_YL;g( zs-ZsfKvRNIOy|TNsBNivf~kvt6njb2=R=qC1hTaxDZ_Vi?&zVBGh$o)# z94_UT{*sop^qnjLW(EBL1Oy7Glwmk`hW!M4s_r9k2R3f#M`693BkGWN1aaxE@44t&Qn+6ne6Qav-9u#3 zyLdddMUj&8EbvKuOlIyyq{_x~{ocdD`_C;l9tMb zkgI^g^IZYx>N+O3h6EMT8GWBg=eO`W_*}PdHoRy%46HFl1#i44U(174El@`iF}h?V zb}%hJQQ^n?!PKp@ZmJSKmp&U~3o(J7;Z*GZLKEQmWW;xhpk6w6MSo*`8@HAZSh#!9 zkPyMKl>NVg*EBG(9*5rV&CbWy#_b}{5z(RfTg4-Vyu(A>g|NBRGN~V$QR-w$tq|~d zMZHF$t$3LPP^g#D>x&+E^T{1eUz{N8HM|i$}!k8^W|P5CDNh*ezsGF z@urrAYXNT#Sk*D3^*g-tW4`!PCh<{7HM_VRI*Z@wBxZ3_!g}f#`ro=U=cshF9b_AN z47asNXwYBn^n`H({=eD@7maqOF1Bxf4>?-nOufvJTE zz|RtR97Bm5W>E5GQs7!I2p(Av{PcM-@VkK`ZrbT2oc&$c%;Xt(U1ngl6w9mjV&^=( z`|G(^swsn$tU5%05nA4mmv(t7{!s?H)g}&DW-Zs$z{T<7Np&ejzxb#lr1?FYFVWy? zOG6}P$g`ZlAGNDm=f_KOwm3nerxc190nb`5VR#jynlhS4%;?&4x`!EGw4t28ZoFrD zQ$n2;FwREy342SE703^iL}S{a8XZl&6W$b4-|pO8_Q_sR@T0XpOE-Pri~ z%nr{0IuBd*Y(?{{wOx6bEBG8PR2AXK!ri=_WXYdAadmm??~vkWoL*_&%Iu`cDm&`y zRh#r)&FxE#o+Rl_&v#50pRV!k$_X#eVuc0{x9RT>4~OLp3zgCpsyt7T=Y^%dwh}UG zYSE&t#Xu&*J%FM#51}%(lj;H(mkp#2()|kTWRapV+(jaK5+W{d4ESxEWb{VIA&<%( zEv#ss{)yPYFKpiXsbYY_ zxa>}@yY5CG!aUHAE{9OK{VTM$**LZgy8~4xnW`4nOQlfkC**y7qaJ8G8R%_HK4q78 z5O`G0etO@hxT)6}X!>ic@{GfMQ?X#o%l-CMPqgN0t`h6cSi6ozT%*G?ijuat>^9x< z2SEeS_b$tCmd)Npr8E2CNAB8ZGM8V@8^5I8t5n#ps#S%jIuf`Z>%}506;e>v=no z$G2nso7Ns*?tk^?b=NBYs+G1b7~zadj_KJb0y6X0h0G8n28`~UdY}Km+P|cLkj<;W z%D)w&JDq-5Bo&b;l^^UOx;&k_sDwXrDxFk1ToKwn86#`A#YjE(= z?71TJ`qSHoLe)k5pt1Wijd^{y^g6R!E&KR2!gGWMwLID22jfFZ4|Hbd52;g2;UVv{ zi+crvpf}FC<)NHwE`2AB7gjwN!78K3~e>jxErWlMzvbfb=A z#(|TPpB5^g5~%yDn;V)1E(jk&?=q?#SGPuB-~2*zpI*8v@vS8b6%|AD)<>F%pEn z_P(h#I1rzdXpwaO9*hPdGScMjYm)kU>%ab6YTjJ(OOq7*lC(`2LOs3x{=Gg5lh3_1 z0o<~w1u{=Zn6GY_hZo*Q%5@Ko@%Q58Je$&iN<|ig5|lX$Bv+5jL^yP1yh@h~Jawb) zTb6`(k*SoQj?3N)Y!xkMa?A7kXA#Q}?zlCkWa1s&7oZR@8 zA)!8qE)DbsN+qG{>qKs9XqU?r9q+O5@SK0_3P6MdS3=4x3Jw8>+0>Bs-S8pg2w@)9s?t`u45HKQNo$xY`;ofwp&5yq8L$(eIsp{QEU%02;8{+8r6u| zxUue<=o~{)O2A?4!yXb4f#1Wfyy3#aolfVcVOmEd>_wQj+*GpHoVQ;Yf(Lyaqr*8X zzyN&RQsMwWv6oLW2IG++PYR;iM6hEgM#TH{%780c%#=)U08Ier58BIM23r3JF#Dq0s-V@WJ z=*59)5(m9Wjw>N%N6|~%4vLeBA1Xyi&uDVilka7zo}zIIveFAef|PXi0NGLh+~M#= z(80mftzZNLTv{Cxf%`}XkY5_^lldbg`Fz#xyoudG4 zhte97#9M~#5rldFgj9hTUo;zKA~?0K8M6*O?&~3)l>F0PuVk z>sdT|mHhG^twJ_9a}sGvPi(@*uQyvU14zc|MbkbSl(3F#CE&pH&T@zF3ke+OfUTED zOXOG7Ohh`p&?ae4TI;c1RTD9Tr%0<5(VZE zZk80-6Q$rAQ?}k$>W}&bqAY>rq_Kbe{CLeo?wi@Z77;le;vUHw{a_4Ksm9+GCt#!&l$QDy3^}4W#70K4rLU9a;@Ore5|&doR-PX|v|1IJ zwpnxJ@bNc^Y~AY+xzXgWf$lVLBRBna)r@TxK7B@D0Q{#h!uiJ-1^J|6|7T_D z|7e%*|7&GRi-9o`)*U-*@A5w?Q?t5+W?M0jtGKW9|52G5Yz+KwD^s=ju8?U2ZjJAk zn!uAeLOvLUY^J^LLaa&!VWB>s`EsR1DkX7GmBT!V0{XnjW7W|QSSY$pUy~NA`j|l< z{i)0DO@22YI?-PTe<_o{r{7<0c7x&IzoQ$%_VYCF?rw)2SRFa8rxCBOvGH9SBYFW3 zGAboM8Upv~bdV8zN8n}1Rd;c)KAI``jlhln@N-3VjU|FZBBt_{795-meZ&kmQ9IF! z%8$m|p&cYZB=!x)K_V_H4o2$?gMJoid)N)cF^2mN&u0946mLhXGCYQY|8Y+_pT_fD zBJP401PKGSgR@K$)A7Ej!xCXycg2 zkAVD7QoZhT1Ug7e5MkN3G5t(qY0o)M6;0i6_=@eDaClq!uSH-d+=zEUcne;U1W5RU z0Jx5A3ph9={&@h2mDWpYp<%gzO?X2tjJtrqrk8e)WS4t>(BSR)&f3*YS%*TZJ8yA> zsDL`I7Ea2w36$_rPT(|bL~hy8@G*B^5?b&;5=7hSOg2nM$}9;>69#2m$_ ztJ0riO>Peqr(PXSoX?+rOOvo#xm(Z{Ts8xkSy#E0Sy`8xEd)3-Zl?`aX8yIyN1-(s zg~NWgAnSp^B(A_{l-a9)^)Y)mj9%?(CKOKI%Y=^exMdCg9zJVD6?v@QyiZnSeJv&V zs#V))V8_E6*!8x)!B0`ut`vYQGuEy|*CI1zjy0}S1vt2tYUcb*b{w}&|NNf+5Kkm< zd6jb|Ovo&iFrr{bLXIu?w6q1^Lovg|1`qc^TH9N;Va+yef%m^UatQx-=#cpT3>}bx zY7_Ij;!&voU`zMfOe*$Y*y{XEIi3W(QhU{b(oLk|yrHU2y8kX z+F<;zKQRC)SeEm}l7-*rC{CpnH~7Du--m9x*lX5k70H)n{}s zIhPR6zI~qlBy^l`ha@eAy3ToZQ39hq$$nxt|J8sw4r9n7JioUc1e75oS$+{~DjZhN zh;I+@y&djgMAq*ZxS7qjqC@4H0LfBQN5>ay*i4-X;MiA7t#8`bOY!AH?yt3yL{8xj z_M*;dxRu$vl8F~wbc)LK&8jgGHr%k9M|Ta}`A(>$H;JG2q9y=96yX&$%My=kA=`OV z5v7&s*tnfXBM~AjO;VG^OWU%4PH9w5#>ul<&y3AxH2gGSc1#rvb7Q~}`75hY80A8$ zOHB)&7q`x|%eOnUBWPZyi!gb>qlb{RcNybGx8Bb2U-(4>pd@hTHoiWj8`$=K*4^2* zz~m6ss9@-Ak_b$;#5iOE09KF|nv~kgK5m0`pBSrzqEdS3VftsNXqlr%7xx7|F-br3 z={T6BYn^9h=kRpuL4x5n%>0Wosws29jW;;8{*Dp36qvp3(xeg;ausEd8G&6OAA_2${A&6wlBG49URPY zhMfdFdc4}l8K)^KU#LTE>0+SdZsNoD42A_y#|HCd-&WRLjB?R-&t`b^GTWd}{(kKn z8U`7M?yTQG5Lv9`9&W%GP3Ew=U%gu9)l0c#qiS~AAwRG_-yPiR3-URy z&myO6!XN3}L&Sv{-pz!;MV6M>1MYDK|DCj1P6WXI6o7#!gXj0o)C(n4GPu;i2mn)R zxZpwrL|h1({9y>v*FpgA1O(ClI*j_G+=02ccy`)6bsY#Gr(HQj*nAzq5pQ-NE|~1X zJVB-6mxskXCgEO&=nF$B2En3aM`%g8qj}RN)b()@ofaa+W`SElC^rMtDO0iHz5A7k zx{}kKNzCu>WTr0fJ4-+a(Xgio&lPe;<_{zO%I{N?$<|8gUWBb*QlsZTy^|}Ko3ih}P2OOmmVw!+g$8@j@6Vi>Y2T$q z&a{^Bv6FM|F;5oIa#zj|Rg^NM5u{Avl`$SMsydu63P0QwzP&$JOMW2|zj2MiN;qY2 z(_bo^xU}Jn&{Ip8l2zE$V5#4-6cUhIq^>PZVxGOwW@o|87Fr~_Yt%B$e7_!M+z|Yi znpFp8E@QyInkdMNhD8(5GcM{)?Y(oX=x~mcp1TrW$Zm$3sV+s5Lm0_L7yU{&y(yC5 z@(H+yK!TqZ&JJTjX$S5~k_f7ysb99^cOTWcn6Lu&-Dx|lz$naPGLc6q#a%O_?9L|w zS?3IvTr#96-ssInP{)p*ACBxYqO8W{A4e10P^SFMD!xBZj%IZ9!<-w;4k6Y=^piL@hADwB^5@y#4&^E52^W4gc`q8s*KKsUn!z6cF$v z1W#f>fk&!mA}P~Dq}In zr$&C0uhxLQySg1`|M@!to_H7F%O4Ck6wpHEx>vcPE(X(gR$wyeb)Qm4FcdEF8t{<{ZVrnnlo{)fD3#4JRT$r(wG{DST;q|rKFF7(NVMD>#{fJx^N{LgqBTHfGp)|jI*D?9E;6qD|#90Ex?t7T6lUYwYwhY+k>`*|_xA%(fm}>lOgpbdW^Y^L0(H5G)p*}m?;|Hg1eK8f2_t9=@9JG&ifJ3RPFIm0S{ z<;=hpOq)8UP-fEN(#0pYbXtBG!swoBk_FttVj4vF=@v)pDZGNu6#+DQQ!XMp+?O5^ zJY%{^H2Aa+hbS1f`2)wLSGW+}rJyZYS9Cg!J3v+4x*G2U$)E00WkNdB z4k=|I@b8lFG%l-cM9igzlr)SmI`_pBEGyugf)bXM2DyQzs2$NA`Nxe&#ybSNsy%)M zj(msbhhQ)_3d{%6#HE3`ma~oH+K%PgNQBwWQrk%Kf@|Pl!nHW;3N|1X(g%P90+9Nq zU=uL1S%h$>5&_&p;_WE{SMYGyot`5M6?UCfE=}r{*Tdvl;Js z6z>3v(Wa5trA}-mcNBynf#tIJ*%65yF^DkFf+ z7tCxpPzKD0WGd`dv8;&JH-~QpCWfw>;8rAV_fY6$F;ve*_d9u>nJF1xIkwRxO~)sM zu1b|ja>kpbc=hl&Q90YO8-J}#VxSYK2GP*3X$EWsBveQXQhF2ur3hT8wz^!x_Gz0A zB`>a>Wkr+H9W8R{A_{I@g0Te^MM>90($~NQt6x+{u#y{VOak7ee|Mw5$4)BWPneb> z%>;0OCLB5B%cI*zfa9};OZ zWHKc3nA}XrZo{}(V?huVa`T$Z;;#q^EaMNBVP}U47W-tUD`>W3b|@gTk39JnW0*|Z zlkIpF*nDTaeSnh1qHYq&Rse#a-6HfRF8>N+`ZKAVx+7bB#$ekIKQ}`R+5-x+pdF5Jilfk zD$*!7Uy!-V!`2vy{;)5I5Ww3BB(XWh91_JR(|_-!Hx@7|S*2ZvDZGY}j;)}fYy?$n zyvG%6URu2R4SWDCPfyv*nILrIw79@HsSN?h_=4;QTAU>o2hLWguvgTM@7zRBN1vp} zsKpy?#B}2z0%s$^jFGN6kj?x3w}L2FHm#~t2~ScAJPbv{9dRW}&48gGftHhC(3U9y zB5*M7nFxxnB3r;yw6vf3Qj@B|3h_{(xQYg4wG*td5xn6eIWr{FK)w2X{ABl1Z~#R| z59B_c&bGTmQjU}E0C1kh=D^EyR_@H<%DDlcf?5O$Ym>?R&=jme8nR7#)?P^mBgF3E zi%HTO2#I$Nc=Ab>wl|yn=N4Iy2y-1DV2cagtrDzmqTf)@dTD9sSz|9=t}Uupn`GGHEKCV*03})v-5XvC$X@jjbQ4AJ&Ao)NH1Kp`c%%-IifhA?eUADg? zX_4iuHMSyM!$q(wwlmJiIBC4<4C>> zPofHxz|C!Lv~=+^_iB`~Rc3Yfr<8`_sUS<4dU!q(Hs!2}xa~D>ZM2ZarX3aOAIXA0 zOG{2mxu`-qj;C5yX4HTJs#FM$5g7Y?CG)2MRY446DF33l6KAZA32P`-8|ed*{K15@ z6HG0y@>tY8Ka<2S)27{K@o%HM{l$DpQ_Ft;oRt7eHmG)kw>KB70t zTPvHq*M*9Ak}9r9O|bSY%}YUb&NkL)c^bL22U0&Ag2nNW?{I)x2Kjk?&G|D_YlUa%zfbh~eIG`G7vH}(AAWIOOU-NPu+^_l#`v_}a=$H()g-`rtNRMt zEy+n1^5?n}@TT~GOx5d6n(PIw#bNK8ooSJRL%rv_{eh>S-VXKs{5%#%#Hz30viEf^ z`fL%*3cyV9I;XyV!|R+T^%z%`inqGCo5+iIoPky)^{hqq^Z-U@oBbLe1Z*HI69eP0mUs&dROBOxxIonv^bC&)UMNmIG?EQ?FGux{NRUTM6ca zI47)Hli))x(|a zaxw&4Sad8rmYVmj8wJWg>8~b~7wY8YAM!|FJUOa{3188iw_K|$2xo2B7q^Yh`!lxJ zW1>$l$5>*+eq_-q$=VJH=tou$KY1?zu@b*G2}phTLBre!%)fszD@*37GmgXcq`ON& zS>Uu=P{#)nxS6F8xGLRhDYiQ*l#d*Ho!{^7p( z*9?I1RD1>dTt;Goh<5$Lmuwf~)jQ%&v=X#qNUI@PuaMGoH*@HF9<(#}c2I1a5~FwN zEw=DiLc^P<`oNHFXWw0Ta!IketQKRPp@@o(?n`TXSF-Y{ALb)pz$`YK3`Z_na(%DH z-Se)mCh{?>7-#vJpD27QB=7&jmgeVq|EH9JUqD{omG3~m*VZWRJWZ!@j=Uqe1{aWYPS9VcGX1h!o0U+uS%UPt86I%z)4RKc^lD`kzqP+(Ubm@D!v0$&bA z)$N`Mih6E`H^7FhG$ur(#!|CVLGd+sm9YRTNNiblfCf7l0Ae!$9y^peSc2(*NAY|v zpctHXl}B_%qNu_g4F*R!*%-}X@LWcdf)+T59l%9^RsF_YLA-|VpNH_Vm~AF&t$(d7qPKs?q4QZKU(qUkLr8l6k^(pb$d^U|r7T zIV-F-U(ep8YvZ63T~j3lJSqHED#;uHeqlL%n#Y|~H;5Do$f%^KR@D~+H_=UH&`G#x z8P`hW-^8)hiBUg6)D{>FPUV?-eu0)s&5Qa@vA)>d9{Y0J620OkP|cBQ-+V`JvfyP7 zY>!tC%33rcM;4JIm{&W4=jQ(i+Fr-fsvRUw^W$~0i4s_d1tw{y+$=#@l3i!pd;1&d zt7(wzbEb3bm^4+*{W457h&3rmaoR@n0+;=vnr+ZFXy4< zH+1AprytIxy&BxJXPQgMcR~zQex%GF(auRfhHB7tPDVWsa}~x}osc9Ddov|DrPU!$ zKPe;DU@|Zt@Y73E5sif^T%rPJ(#@J)1S~Zn1FLxNJ(bla*!$Vryr!786m|IE`EobS zW13Skc*pbEvD9V*Ibe)i9)%?PMuqEQBcL)793Vm6VxXt0`eR27YsHy;7^i_YhT=`B zoVoi`Z60|5g4hATL4Q*}%l)`xs;WB4o=R1#WpYBU@!Z6Yi|=HM;a)mHSE_dYZjSd1 zmR^U^jw{5~`XY9ty7iI^qP@iUMeS>0WB6uvDI?>llYOzdCj`cT@tA#v*xT+Gsx~!~ z{o#*$r=NZOyeDj0K;o9hiD)U6;$#<;G6SE<68Tj zB>5OXS~Xz2N~MVH#jOLfsC(mj~U zF-u%43+0W3>VfL>b?#5M#0|cqe0e89yu({h&>`J%OlMGuN=L2fOyebX7f@@cZ1=F9 zQ>T~6j{_C23d*}4letRM3srN5lQ@O?X`#|j0uhWI4^GGxo?cYij4jXSZ8`Ps+5=79 zYjIgOino;?3+7llDb;IvxXZY8JLZcF1UYdw*kF+II%W^f@q4!S_8_+6D72=$@Gh9Z zZa}U&J7DtiLm(9u3+?byob~&U{te=GcVa({OuE8ihm(`OTXzl$1z7`lh{Ou*_6k=R z-eBf>QmN9tr+2DTFk?r^&%blfj(lo-3mm=v@{EXek9-dW5K+F1XtH70`!|@g)CEa< zeZJPehYZhTVJl8EMRk{{xCr{CDw%@peH@S?3~FktNj3U|T9n}`d5yfQpf>UCv8Cx8 z4UrAtk&O!;*^r_*O< z7xF`Iq$?{b{^|o_q(K~5_9M1j3SUvnId5Fl!Mwhx!;in-+(Lxa_KK>bxV@f~#~CWm^ZWG6B98*w2p*vv6cCpYerGA z2^QOl@DDuTPZ|^?Ta7UU?US-9LtiO?yjj)}dK9LEG_tsK9_yP4x?S;haej6(Onb9h3MjXa2nr2mG4{U+$Y4wD zO?TP+Fj+zuq_|4rcm*oN85u8xSLM4dOsxTN}0;dO#~u$a*bDU_KwO^0-ZirX}d6(McB zZb`mC7cOL!0{Y)RbYTGifad=*Zo&o9_Z#kwC*ZQP5W@{QCXxTSN_9bB95musagoFW z9S~Ry(2x+yW>^D{z;-NRs?S<~Q!EqMOL_6aVL9(<9K+vv@$Y&NzJC>_|96zDD@`G( zE$P18gz5A;YzRKxs5EJEQneDU*?#%B(x|r>(z4t8ct+pNW6^Ba_}KmJ?cnccpIhNw zG$Zono^8>I5gVt8VgAlD$?WpI@ta)Z(fFc_>$)H$4?lU|9%_`O+!N$5 zqJ}s6ivic=IeTg9xm@Use8+CeWL2>(2km4v+C4`d?R*!d9RkQW3e~f|BsJzjK+G|h z$^k`vE;QwG>ryi>_NJ@RI7lL1kT6UUo+0>djhlY=Ona9Ns4Er$#0GKL6Y_p5fpK?p z7poI~DOb{wQ{SsfS8rU`8r(3+kg|3=;SnjG!>v|#Ht#N9sZf{F1A}Ia?98V=#e|3; zV;)F6yR6x8`tqQfdP|c-KrE1ekUw2@A6%tzx1k}oS$2Ci^(QBBvjM--< zi*!F{A-)ewCD<(xHV*PRIDkTZwU6`_zWDr0V&$tW#lNnuF4p5;aCc3l?xU8UR#xTc zfMmcC^;2B3I-tTIw@=?v@xIt)-0AJ_mHf<{=AE3B1*oN}{dI@dO`JCr#8gjC{4T81 z=KaO2+|C0&KNd2Y`m>5(*I&?`t15miLhze7R@j{xC&7TJ2s6Vgqt&1aS6U!m7@cJ2 zTs!jZ*#Y4e%LtK_x&;oMmf)YN^B~$Cg8-fH2wm~{VR?xvxCU$ zLgMdzlbF@gkL85|ur$Y@>YYm=gh!nPw7UeO>Mso}w;Zlvk+F-li*kb-hR>_rkEwmCN1O`}aUC zO;4r2Jrj=vTndO&UH+{HSl}_+tqmANPZu0K@bARz63*;*)PeHTV*fWpWLx%bGUb7Sbjn1tOYiXkqT;DYX3oD|m z)1gcl3%>p0TQbwK zn|t_-Bb7vNmDRbm&C5?>XGt9S8n%%Jzdp?C_tn*P`ow(3{q&L&L{BXA_N7koUea1f zz~~6K3amsy3F~fb!|o!5^}~1~1!`Dvi7oNj?I#H4S^8*-%B6Q*qz8kv)Vvri9p~%e z4L`fPw`nd{D?dtL4%Z=%;5Gj$2P-y>L_cp&p^IRQ!smSE0p}Rml~S_8HG@8#R+#;z zsdLCwGgsZvW!lBvqw075Wx2R^eY##o+gCsM;&&MBj?}7{qIqWraY}3rShSV<aXc{4T^i#h`|g@nyD3FzWM0WaUVNLNrPj%RAEX8z`+Tg361hEOq-i!aZ_ z{pQtkBsT=8MW|^|KSGp?OgS0GIa@LL*|RtXYdE2y;}iwKdUHn%DpqvQj$j)Nu#uLbRD+9_dhjcDXUT?$t3>k0SPJul+ykWHTZB^<($IrX>uqMx%=9{wZ>2)*N8vCNtlLY?!SVQ+ zY9C-^S-*6io4zW?PUO?{9m6=1&GjS2P}X!<+^t4hd07-Krg0_+$|0&Ffz0ww`^ zwnVL0Jyps(N%U4zq?vF~CK`nOQm{c2oF{EEzE%#b%})$PhRWKf<7nYsst~D=-HeeA zRC1A3;r=s>+x*2hoa;cHoc~*k?oSf!Us;EB6gVl9^Q@d_D2uNZyK*g%{u8~q56p2; zt(1z;?zdVJ@fPX9O2rtMnE77h_Z2sLY{|838Nj7%ldHHsT)R}1RI`zbxri8EPNFFR zy9zf6uCe4}Fj;(8ynk2>lx0itmnn|LIql*z%O)1OWS_ID@R6%n_Jpiyv6!D!o>Y=+ z5E9P;K&t@xlrEkj3=otl5`8IMyo*b#KB4GnMGf^TRnb;?&?L2{AhG17 zJRjKamq=u#=lO*Qijt%@uFjUu9>!p$5?mWX-Xts9B=5!gh8;}$j%S{Yw%)2JG|Vj&ttD-A_{k5>nq`~vZ!o(g!y7Vh-l?rXa_eJ*=E`|Cu0#Mt9Q&D4 zsg-q0HxLkpsLgDzrIWw@ES6?}hv=-FqIB@7DkDi>IQW3Fbuzh&>#{&xwC(!1P`$cz z-IT_(xKP5V&G-RAP;5cevOisflkc}~6F1A)Dv+CQ> z)JQ%q4JmA17;x0}rji@VRPcUeIfg4r3|K)yu4b9#t=}d@MT+3OqB( zWX(9hu-W79S|dp{=o!JDcbdozdzy`F&ai0b?G0f~@xOI2%*iy&F&f$?AL5l7>L)VF zPaf)h7`nO88a3)gC*N4ACxu*XB67e7aA zJ;P<()8YA5x0o69kukq{kNcav5AR+#*X7{Y{{|rWKlL&g^dG%U0MwjocTkQap%$}6 zp`E7wtNOgJgvCmqpb++7AAZjIS3i$KTr3w#8dl5nu!I7?GS$DUfxcqFPiG<&`>I(c zkxBS(uG`~XoqpAt%W>y_Ly|N*Z4~u1ZM509@xRhuv7c&oIlWXk`dPNy>%0FLC-A#$ zq(4+6Yx-!-zU}3|`+w&aFWMc(;;~psefrrQ61>ROLyiTU4&w3&+-SB84%+(5J|;X$ z4@w;MM|i*FD#?{}nrL<8^!*sXvsTI|e@e4L@Awn(nmnsPXSm;gph5CpkV!_=c`5;4 zo|dVE&F9nbhW)3fD<{+T*yBVPs%J98V^=r=tO`q|Q30cl?u!}!avv>6y<-@-P~CFh zQ(pgp;2NA$TfUE=-8M;aS`|I|5(vKirK9|~r)sH~*^q0d*i*to1^-F_uy<$ob@}3f ztB+z+k=poP=_4}A?HqsiZ{5H(|08WY;KsCfH)HD(1VQ6WyL`#?C0}1i#@|*c2qJkZ zTNQ}>x=O z^0(lLsnIdIFId}=M6@ExqNeABbo20glCNQU=S{wKc7GZzbau>8c zYtX<>B7eJ6=T4HQSkd2iq{B8&r;aL+D;oa(6i3td4VBc1qMihLMYbtt z6EEAmf0I;yl7Su+zOf|~C!QO==L#wtl)w&M@q4I!*$U8K;Qd%=qO9dCj8OCD$v5g& zXql9XcnMw+eSG3uCrjJ+%izpSF<^vbq?m^du2~r*@@%>36gvDk&{Ir80$LjlCl8fe z##W)b=!*`R(T8ksV}?bxlK^p3GLP{Gy)WJhNqtnS^-AUUthXb%+6hRu?R2r@<_wX~ zw7tJNbp1D5Sbj^c9lT-w>^XPW=gvo$PcWe;X?C{`uE#x5*6$1VV zMwcI0ztfd_PrK2@#Rgp&tdVkio`=4va)0%)?7otv?{x~$w(o03z!WKm5hM%eJ}GFf z>S2@hF(@b0J*Z=U>iO9v^x>zZY!rtHi?>Ta@2e>vwrViSD#5kwGgv*Sda&@)xUZ`e z_SPL1M|wtTobu&PKj@!F3KY+jUgB6?{=I+R4eEC<0XIQ-jtNVNj51lwN)_vMeI|4* zkG#x*jNNx}`!@z1*rTqO8oB6|s@^`!KJTFuO+A?Eu=ER-Nv9E$!T9#WW8d%r;?S$v zkd5g1Oj!@SHC`xkZ^;~gSjr2{&YoNqJ|nI{e%`9Je@xOh)2$(QpC0dNx<6nInWXDo zWd0!kg`Tss3Rn3;5rOmhLoLTV68K7&|%=Dw%LkJWmN`NWFA<_A8hCJ z7(j`o39nPU;S*NmhoC>{RO1jFD2COUTHor{XzKF)YNvbGs@~c-Qeq4kLea$&I@D$2 z2#9dWt#AHo4kkEX)y!_(G!ipY2T@a*t`?aaKQ(Q*w_VY`%#*1kJ*dgZCbDqX);3zB zs&nbd&}YQ{iP-Mcz;((K?R+PhLW7Ac z|N8?OYQ8LD)R}!)?N-`!tZ8>#4dHgvdiC**QF-%R4wyNJr^!m}Lpx-!#7EogRM%6{ zChLh9w}hxprH%>@{%R8UVqP4dJE6dRsZoumfVB~!MYnY>2|>{4+(z49aMwB$$#>x& z{wW3uUjSoxAy}>OQ&STuv)J|PO7v9{5gpssvJe!j*YM0#yhgF!{I-%iD8&MYaZ5u6 z+E4NYF0iG{?X`Bhq1B$6B$0r~jEJk-mV9Zp2WwxupI)-}m|h69!3 z=)z4KH8aZ+P@8$vqw5|e|?tA@5CDt+)Bkj5~G={dg%Jr1gzQ zjoyp?kSlv{zB&w(Q$y@}PXys6IEemjb@kz2-3*T*uGx^ODGmreaLq292dVFX6BdB^ z-9u_sN*`qsGukID3^}jLxE?G z)1QJhUIIGg63WO65FL+HS5?}tQazwBEUvR-?I6IuieQ>mW23W*yNZg0~TpxdB7T1CaV134$*u)dB91J6KiI+ho zzUfP}v6KUTiRZ*ic*T>Du53w~C4U~wA*QD4Wf}4O2qk`$I6{pIl!*^MdeT6bJlgZD zaU$v01~Pr!LS7SG|2y#OFYEn%(Zv4bBzG6gy68!7BBV>fIhZ-r^2uCoyvv^TxLJ~4 zFg=(&=>?ujd%qMAD|yEG89lpqXRoF>*3%yaOzIFz<-<4}6XV&0gb%?=v-_ze}uLHZgHnRxL#eBA`dkVJDOe;3w>?F>*;}O3gy7;Imr6%## zAH_Tk&H*{;q+ID}CV2t@-uRlK(lWWrVR1~3xz$Jy;o59>M{nsu^1-QyuubqHm0awm zQ++oh#*v!cg*}QQ-|v{GAE85-tNW!}mq>!IRE_ChY?|XY8!NgzCH5>jf4=}xv*+Z} zsY$fy^aj}MIn5g$dexa^CTee0G{`AMpKt&^Bhe^Sg5K{8ha0#7CCq0VoUYF9;0cAJ zOk?OX3-K&pi-DOtH3E=@1gnAv?2LLhj4?>+XSY_Kd9Lm`DmvGcN;vXpI@((<)kBwp z`U8qDh~$lVacM_K4&y?AvU&-EoAZh>8)HP~NQN1uX4kn<*@@5@k(jEa(zNf4yqzZ>)YZUGZRK#=gVjh2RS$$F&$2So%eaUcq%vhKOLS2yCQq-U@eo%TWn#eGx-=WBZ+_EfQy9s(xJwoEp6Il~4 zxzjy)fgAbHV%$V>B~dpu1R;D__vD%@T$=&4Fb*EC9i$Hfa+yF{Yb}P#oXlXM0SL!= zMM#WVskd}0B)^RNJ;zvQ?TR0gPz!T%-c!KYq3Y@6QtDO(NbstPfz|J=?)j{{4xY+f zH`y8s6akgmIax;hm8o$VUm0;F)}>IXll9GWf={a9VEQ>eIU$(!7P-J(CybDawj+r* zoEA6|o4%69BTMiV1&4%~+GA!*VSPr@e}ue7J7$$$D6?pjyxKIdJ&dlk(u^Z> zyw(4jjv@m5a6?|a2OE;*R(D}fpA5ACAMmi)6flK98@i9&n{WL{(D>KcLHxeLBwsmm zwF|CooTu%9%?);>r9Yepzt8CjZ|Ipap|>bb@xRN7aHLUP&9&146E(e%B1{8{^7#r- z0{D_+37%=PzsfSV|6l8a|IwX*%>Om0WrML{lM|Zq!YjrTv9T0#Kt@nNGA*`cwI!Mp zmO=ern_I+hmPf{ZV+S29VVf-Sg3ela4&J%q6hZdkd>NGeo>RsYx{A=yVzEo^uEdEI zsJhM34sP7Yhjn=|Hv~3~zXT4_{Qdp&lTw7sa-*6KP0MbNm)*-(Hl1Z_iT_^0y1mhD zP#d|9TCUR9N zG|>gG9NJD>&toy}=Q@YKe#O!2ger8iDEP19P_hktvil0pRG!sWil%(mp4Dc!VV zgGBII3PAD>d^~7HtOf3h(4NDM<@nEbmb(eS9QT!&Bfi@dxJ?H(t_TWH#UTcVIF)E^ z?9o;!iDuz^eNk4S#$otxWuHu*ACC_rLS)T@7Tz7f`;BpLGI81foloZstTFNoCiSzTJx{vN&(K>dOn2fSLrneH$J(7^i-)Lw#@mUT2~6- zhDk~Yrz^(NAR!j7h+b7%|G4-iM4k2(bVRp^*g77IG8K&i<(Nh-^zL}px#tb+xw&-n zHNGfleXp$r!wYn94{a=%@(2Hr^nNZ%!FyWxnL6kQnq+@!)4;XC`9x%%D&i~|B|@F2 zSzO8I7Y(*rEE#rR^?t?yXDZNbd}IncLs&-=(6|ELHHv0;z^Hz{7_6lhvl^z5sGP{^ z$OgrZYpT$QAaDxXl%7&ZQgmY#hYP^O|Ga0Mn7wKCv*8#*Ppdcx4s5r^Gt*dKrN|9UNQ`RlX99=$(HHgjzR%t>HxUS*4sqn& zeFeTs6*F;^^NsRa4=8=;fdxJ{2>*g%d^GI6ppU+TpZEMSNQmB&o3{}kTr;+t_EqYz z|J^hCHm~d77pN~M$!~j&f!gaf-q+A_U{zl!4k?n7J@b<-#DjwH5$8GHH7g#0OuiH^ z@snPG`GUbUJhUH}V~7&1LKq7+XNQ1D$N|K-mE;Vh)d@7R6{Ml{?T;5|q1=ZueN4ig zFb6{iihm%PqG@8Pln%+GM4YUIcB$uQX=F{f(ophHEqvlju4K~u=7E?K%&P7$a9&|V z<2t1#HQ146%am*hK+<7l)o>^vI7Xb@DHbB}rDB?M+3ytjyt2vQLuBq4^ zC%Y6uj`>tZ4cMA3T`)o(8A?kF%Cutd;vury!Oz`N<5mxB*HC2|q^VhoMJ@;5?s~oX zzNKocavJ_6pG#Zy@IoUYGv&b7PW(&ZiCE>7`Sf)*ANHwN6s851E(z_5sxuSUsF*&_ zy_K=<3{-Q(exDGVZ6kre!r_I=CHf_kB&h1biCxbXw3e;5#nC@hTu*%3D|%LJhBH-?47zk*G-hlPCs#N;T?jSuj9 zi8iXGp$6z&wC8JZ&gv*h{cws%*K4H(C%J`#8{)mXim~>co(l5;jQs?;ELQ6ENc>E& z6~-F0`*HD9cxxqc*Gk7z4+eb_7vx8O&;a}Yb6xMixn;f(;_;5@k+s8f`m=fcf)1rv z@3r;+Wdu@VQ5TIV5#RjG`cND`=2=*L`Y;H! zUR?W&8hCVWT?60h{3u$&Q`qq=9B zxIPwFN4fA^th-f4h*0nsf8&Tg85i(BLZ#F%UdOhA(BYN8n*D?&RFYGK3GsWuldm1eS)wdJwm(Y`zhd~sp)))cQQ-U`Xp?1`y>3%{okv*_Cygr#Z zv6MA*xB}`OtRTWKTR_h%CKgHpCnx?MbFFv$mkBD_X zm9cM*Ucd@Ta^5|Y$Vc(}>Vj9-+$qEp`2Lyw>Yfco_fLFx|COQFswd9pWYd*`%c2(_ zUOD}_;b4>w65}6Anr-F&v|K&*3CfRlB|Vv zYqpNfXrt20`a#>zBVpzAlX3CO~@$-NS^Knqhh_qi*XMDciGbq3-M(s1CwkWd+1<~cyDG1T;V=0yZ~@eg4-R`^Tua})cu;)0zi7v;y!t9z7%CAP zVf{8hRRt+3#48&BLCM45%;M}21cYMBZv-P9EgVW!a*@zu!CG+1!<3a&FCNTb5t&9S z`?rpgWQ|!tS@z~Y%N*DBvyqziR_Fq!GVn_xz%ryF6j5%T?d470raWI1xy6eZB3fw4{|p z+Juyno)JqCo77o9iF>xC9DB+ukC0woDs(;-t;v7X4kR>yZP$UUhmBS960W)m;kJlQ zQxo8q)XnLSd;Oiv8qSs#3qrM%*qDm{EZ~!(efbnO83YQKt6;b8&d8d`7&8o836bz*vp7#|lsBv+aH2gWH7ST8kc>MO72r-4wr_T$$aW5eUfgLjhB9v3`8c=&UGJ9S6xg01!_uWCTEC z@1?058Q4f`+1zAZ@ni$ppO9qd&bN#G{VIGZ1O8>1;)8^JyHn9}q^$$u++2_mQ$qME zbGFA(xCE)_nq>b?G9oJ~lKHsi_hh?YM63!yH&9Y|633|sYrIqnl}7=~c9AJ&p) zjR+|mAe!-qjKV4huiCG6F1YkAytyAq8(W-JPOw%^FmK4SKUS0}N_C}~f)!TK2P$gD zXT*&KIZ;Zf%_q`d+Z$r|(t|w>)b_Ib@4#ncqz7Z9`|YLYM$nsa2U(Lcuj<%Vlx&Y< zF3zQN^**im6ns*)eA|>um-l5<1uGwW#!Klu1$wq5PtWg_Ox>5%9OUKuR%Ety6*k}5 zk7T2IpviAEt&NYWH=BWQ2ZcE9BrCPzxLJGIn6|}3d(adeMxWT$Ac7^ zM(WY8!mr3naZqI(>epF z@42pD5;Yx11=qOdK^DREER(+v9ecGz$Fl$E^!_TZ3<#qj>0E zC$}CJvQt)85*jt&H8n)y#a!<+lRRXHY$n5vC40rdd+aboQ72HxN;AJKD^uuMr{N-& z;}1eV>-r+TX%%f_6B(@CR!5ECOU{>nj4WCvl;n)sdXOTy-_VW7vc1dh_`p`Aa?@0f z%x+`n)7~JNd-GOvL0!g9=L)5$9sb@-1&%>*Okb0dO)Dgznly;Ch4owEgpk zqNjWNaS;ykp`J$}qQL*HWy}g-0VD%b{$JZSnG4qc2N

    b&Kal)lXhElftHw>_fL& zG4>2!kH%EXkZUpXzkpGC@jrZxeR$wJ1_-5y6Q=(i7^U-!A((#kvj*Zfo?HBaS*;ls zFG<3=cJJX-8*qicyRo-gCOX z&$B{OJ-pH3z0Y}w`BzSw@R!NxFo4fR1lTK^ zJWk6`3%jRm3@qh&JIfbvvG1&;9kT4J6>oR=EYM~@(H3LWW+1~aVa=keWA^_VEOz~K zj6hq;#^bX^San-pmj#+yLZG;R)xuU^0_;(S8XpNRn7whS;K&p*kuZ4pQQ;gonbu1> zGoeN6%d#Lx;|w%O&Uk9yV|!p09s2AqO?>RK=ohtgEhi7A0HDzKLoHyW5fs}z$bVU| z)Ye}WTh%Y~Fo{oW;3FLP;n$*~)9B@bV9Kj20J(+3^!sA2jm-l!B!QnA`5isjncSD6 z5u$~q-ECq1)AkZsYhimiZJw1Zv7_Bj4>5DKOGtLq&UrfJ^s1w0%mZBC_*8OJ*>dx~ zCcC<55L^$Q&<5KN2IP8m%yBoJb-s~0ZPu$+5IFB%H}^j8*>Y<-@BI*RdfxXXLEuOK zL7w-Ifuq`{AA@JTr$2@+W(0l?UvGN<9J%|}^mFv@?P;5>%isSAjB37^AeKJ6m;`G) zxtyZ1@VT6(b8o(!VG2FFoMlUV(%Cbc?$bGSQ__63z&`TQXyHksb-P}9hLu`g64Vg zc`^dqh|rV|UZaGl#wjY{8*`aYl)@dh{lEA+%ceNE1zSU7jW_NDZ7jG$aBJM%-QAtW z-K}vC?j9hxg-?UA1fNC68kJd-Zr>=h?TUgf{b) zNJ;qF(m$)Ui#1A>Qk3rBuQt&6Td_N+OAf|8ysQ!^`!#O1J1nFU)%rGUrd3xURbN?7Lc=~|(lK0c1KLUlI*}mawdy0svgJ@BmWxOt{H0+Y;p~ZeYTr^66 zV(YEWbS>JkE_-XU!a_Dn_@J`!{rrH+1jo0OeEVnmOgXe%&rYf; zlfcF=PMlFu#%SQv00zDHBn2O$dlP*R4u3fRv16k*e0&1OE2Jm@QCM6;bOZ?!BwM<~K3o7{qfmA|hTSL@)-@yN-H}yXVIg@I zM!Kc^W6fX%aGh;P}~=)ufGwm_M-&zgfJTcP{_2&BoJqevdzme$#;L#{L`7^a?4YRn7?>LC-&U?{j~ z#$^Iv;UCT=wwR4R3pBP);>U0s*BN1vOBDo7Sn^=^5GZDE4GDwT0GkAEpNW#>HB^oT zD7H&-0QR5{E$E#MeDICRE(r~eS$h9R43MLRO{v3~!|3OCB<}^R1+$aL0RY(oa;Gmu zB}l3-m7f(`;MINs5kYyjdsH7p@kQfhGa&d0OqnJ6X@^ve4Ir=IW^+?uV-5eOnKg%) z;x{Vzb~?h7olBqSnx4(WFUX853qOYGvLB#3xjP$u-~N%nNcZHv)r9NBr6`OmnGfpZR{BAQrfei%% zvwnG0sw?0W>`p+XwZkKgH#-i7OVVOAWR1Fs$$L)_#2*7I`pCj$SqfM!u_^^K^(Y&t z00YM`4gJK|~^bY?H(k%PCJFQzBb!F{(Fm2<(EA9I8%Q-CSF0oxrKm{Bdfp0X+<& zz$y8H==k|4bH|sP$vEW3bk>&`N`9EM+TXD(Z{%vq8KC$ls-FeT}PtsQj4crh-O9O2LFPsFGuZCDSaiy#NIrEO25d5Pm5OxWt-KxeS2g zdZyxMp89~3%V6pUG0UkxijFzI=Bwrj6AIElW`Pf5OHb*M+c7+|qgD*jhGZm`feGZ-k| zMH)dEB~VKIx88t?Vr|EWKL;1@N0K`8)w?g4Bf?T`K~5-R4vPM3^uoU>-ER{9j)-sZfdI5QznPx zQ@=mU4|NpnH=gudcJznJtLCGbo{XSwErf`7@)0zh(joOPndmT=3!WK3+-*PQc4O4= zUS&zp=Y1)z;jNZ)bVRH^N{s@qXe2TK8BXWTn+QtMY&H2CKDes)F#1xy)I z$K!|LyMU6K)``L($=DU2c;4L%4S~l(Zz!^y{ycAS^$C2L_PDZo6hWUxRn!L=637m5 zJM#$lxO{PN=FWZm8a{t7x%PUnhlOlJ3)4wg6jRF&2(G&eQ;O*IEkYrlSoJe4ll=Lb zrC$F0iEzM+adqyB>dUsr0aP~)%YR)ywQIT1VmA9%AO(i-PhT|nLdV=?qTtI=gPONSGF~&AF?K+umuVrfb z9XnGmsPx91{#}zv-|HbTj_DE|c_%C{q}-Lxl5uL9bwK|&ycjC+4!yYldY^jM^gJFC z&dT$AR_e>DP^-w)@_TSL;bWg)LpWNy(eW*wCRp7RGuC}9su2IhkitWUdp=Fd6JN?NHGTxTed~=+biA=i5VEr|?SN-iQc!iVDu+3kMoy zS9z>+Cl9x#N@gXubdYbX$Lr>i-kw_BK~lOM@FByriK8jq`zZ^^ahKzI_ZrFQEC`UY z^zKMtVKuc>xppHY>L+&wd#~YeL_A+PnEdfJ@n}40$x{ccs~8loC2j+1zZaU7$)u);Eo#l?nK`J+p-SsZQ1#cZ2#BA&Um1q`CRqiP783m{htbZ-99!5ivFR1)o#^(ud zlo4dMEY>9~Mb!)K>no%Xf<2T8=b$SPLJMML<*$89i9K#z{ zQuBM1Q1mrgSwR1N+~vQ;f`JibBhY`3yJ$#C#o$uN7xi+{NhhIE{1XedRtTmReJ0@I zsfq~yM+*dM(UL)Feqg_afvTioZ1x2^GaXE$;gZG`lm%K2nz^EetY4Y;-)mO zEG)BSf9zxWnrYEptMw3IfmID?U}1|G!RtNnsOze?U?GE~t88>_Gg&JrZkFP20fG+B zo2hL$Texm?v(vlG_CKDj3=>i7T?99ZrsZV?k1G!deQO+zQs%sTF+UgtWYG+$elYtO z)`J)fM{5bT^c_ohzy8{lJ}CJRqAX}XsO@d{BmdXJUsyLhzenQjrb%nq< z5=(wk6MBkHy60fzDZd}{YPnIQ_-*x2J?;&7120sQ8`$<#=z7QwG)v}o zJ32@dI?pkazAh)#N;^JV-ANCAWzven3W{e4`;I*xOURS~p;ZM89ML;zomb;U@`8KI zBB>^0ja`PY#*K3W5Ca*lNqO{-3Z-0;4~o!ZPxqsY@C=HKoR!wCQp25KR5Coo zSAvTv@xtBortj;{m}GAk(4`a|i$Vd9XYOWJ-A8$fiD!8*%`kE&r^O zzS)U_z3S^S#&-D>^@p#9jg!w;R%5zMi;WbP&d!F|p4)c5iz8e1WiJ8^^j@*KiG~4K z+^laurlF!$wX~*gErreU*VPU>G=H+|#jV^v@#SUC~WiQAd&CSK`WxIs*9J*get#dw(DqlMlwO2VJz@~al zYM7ck&MQ8zUR{#r!LMeDR^Fd00|g4$zpgv!>&kCa7Jv7Fu#RQp?f8y;CIA5(;8Tyz z;kbkmd(oJDz2D-QIB2pHABFK%Y2-K}jRmd3ASCQwC((A#Q~4Xwzb@?eYwO(N98x)3XR-O3I!FFgfxGgr z`EB}lV&9FlGEtd_*=F2=dg#i!ocKv@u&1fNYl8xYeih$<(jzTr*CvJ9-+Y6rMm#F7 zmeb{nK)L3j3x!k&t`nn(%|DG$RcJ83%Z-Vm??p2+p#Qgxb&)S%S6UC?CLcq3cVXZ%7@+w zGP*^_s%ct98X+mH2apQh96j}k9dGoC3`GP&EDaiZvTW;^=osV z%y#(a1}vh=2JCL0@69h)$`>Z2Tb93C{<87Au{vA52O^oEywlU(Q<_T{`8iEtGo-} z?j1^`iWC3ww_e5A3- zU_dO_xKbF+5pZL;Td9*49>tMGu!eEHP6>+kr9{rEG%Tq!5F{N8Yllr~oF<(b~iWKB3DubH&Rh^8YbQxH|t(8=lty2PK zA~>7qdo>Gp)JEHo&%$Qf(TK9_n%@~Bf?DbEI73-1ptKt%@M_+aH$L`GyM0rTT(1&V z*)s>%TD+j_XsYG-Sb>U+oq3#YS~OE@#xF{9U24na-k;4`I~v4Odov@uf4RItIhciV z;kSL4{sC#K7s`rUEMb3Z2ibVo!PWx57tPUk44oB;x6>Xx{u1v3(t^p>@CrvDUZLrJ zQ0|SK>B$;*vflroTXKh#(oNO!Bf8MMi-&KtRn(kHp-7wzTUHduY5F?I-`KQTJJLhd z*5u?lZ^Epn>!)i~s^IrN>X!OuOxlviGQ$DKc0te1G1GIcyd$va?-?luOc_aIFl!#p ztie3b6^+X0rG)EGW)D1xqgcm0QK{k8svQNw|EwWN!x7x6QCNg%(}mYz^8jX~5|jvg zF}5lq&#I2H-zukKxLb?SdO)VjIdq9P{fo&QKGH%g}1_^v)IjLFWTz=tD8&_j`v+rv& z+lz~#0^aOY^Kez1>A@$pM@JNck=oKwRiU1sd8kWYPs>!ciTazUm7d4f%|Uf9+z^SE zlLqFOP?0Uq1|4YsqccZ~OB`D25$Z)63Ej*s*66!Wq2z&Qy-YWq9VnmVerPXmkKNU( zf4hzU`g?CN^{04>*rh#Z@aB3kcC5RYlUZf(7ZwYV#?!&1y4y2#ObG@(ky=R+lBr{T zMTGgZL*)wdUO_>l#f-_nl_Y~Ff^WhfcjJ8F*~4h8Ipb$9#|jB5sU5Z!l3(xAM8wz*lV&8EUAZT@QGrK7GZH2Q4xl2oowfHzvEqWi_e0D#y_XE4m%`AdL3pR_%^(w5 zL1eC=&lHH{b|(*|C6W6g@2Q^0(25jI_MT)wi|T?zqTn{dfyg!{$9Dpew+tX~wrY`Z zPo+s{%g=24E-y+0=)TG?BN5;y4Z%M>_K| zgVd+*E{0JvK6)fBE(iqG5j zD{^wfE$GQ#fifynB?b{~y0#5*%6D|6mk zsJX_6YF|laoxp%dLI7sf1ozDn{_WJmjDOTiINS@0MHBI%C7@SS)qOBL&CG7(1wk|j zo-?uz@N%@#No8j=jB9K@Jdl|5<(T!^)OMJQZGtfCKUZXIO}JmM)Db+Z3VMUocFzrjFKem z5QWqvl4gb{ugFTpzMR!d2ERp6mEw!$-C^F70CnVV<0M0BtRofu-$lPYDY9bz>A*rF zfCV0TzJ{o&XCAt6CY8n$$s9$BnoRdkBi#1;xR*4_!)55oHWujM0`4hecn%GNZ3r_% zp-2~=EsRZX21gMAYZ{~+QHI@D$s?~&nBiOKHI82mh{DcgQ+E|EiU~CYgVr z4wO)y3Iln`qui##u@YUYT9LliP^n`?y2)1^YpEatY6&P)=U~Kk;?^74(?XUbXz!Ce zJfrP0#0M3~;0GDQb(!C3;#yCrhY(@&tSYMBi%2>t#dQ@$JW+vJ6`~W;R;*dqGA(^z zWG3WlGjJ}A5IJokn7sgbYDOKyce1>m7#dE!Xz|YXz>#2076jP>ovR2Pj-2=Xc#7&a zuI{}>e-u@<3s_>6m;|BaZ&B1DEZ#rR7z6k#bww2Gs}5|ʎwOUGKzix$oSy|?nt1l^uE^pQHvVv@RigPpk*dCix z`C(Z{d^sb2_ts=PZO~k9i@QA|Xf`k=S|=4oF(^PlC}=&vkp9e@>k;)PieE`ck<-D> zq?AwZQ#4jze^xf)JH}owdlAv`db@q@_9vra#)cL@GO|E23E6SjotK6qiO=M86WM!V ztkMSW-g|-keUG0(KJt1-9kf+X`OHV$Tvz1k_ZSlS32&_0)tWF89?A4ulzbRHZAk3M zjLz{imdoBWa8DR{6DE+m^;k^Ry77JTmV}tIi3l2BU?SXglT-y_R8>}wrwQ5;yZ|XT zDlO);gXZXiA}FFFDSA(cd3zXvlXEN3x44V1ettf zA|DC=t0{|s@CKm{A?<%1B#nANl_wO^l$?nBca;_Iuq71-PDNOELqGjrDL&w2h<1f4 z?0=#7K%Xgkb?6ll)l0#}PT5r}Q2jcmSF*11=|pK`(&Jqo0~84f(HPw!(&M5A(2@q)Y_ATNyEUGWTOY?27gGwghxmA( zn50|3QUv(=)`@j2Or?-X?fPnJ=+!i5-gf_76}Vjsa9q@OgSn zNFtt~sif-#owt#gsF8fmTNL{kqF{!EQ{?rDpgW}-IS2qUSPaUk^DFbfrdm&r#UR$% zC6bT#pJa32@rL3dsO_mK4x6r9sxjoo=OVl&06##$zbHr|wY=OS2UMy;Eg{r?4rRe2 z^><4|b5x6xi1&k}){$mp)8whD9MhHF$%Aym=?LU~I;rlAEE}0*3Dckg#XiMdCLGdz zD3N!l=ZqJ&fmM=^X>1is<9w?-{n{opxq#z0;h18A>OzTKOMjiOUKfeY7O; z{)Wy<5^W0Wo~4Ld)}0wThTcvP=$pX3``+wPa7< z0Hm`=5CfoMyI%mw(i$v)VRru=pM|eVUwtm3Ww#rX;md;C0F*g&7DP;q*uARsWr1C{ zKcS!<+spZ?4F#ve$wbT^4-VhG5NU)X zHI}62jt<^byN7^WXnqH@4`Il2g*u$sQqh$AskT-kAzQ^E(l(Us%!{Zz}*Qlx@->FnQ8oZ~!VONix7quWP z1&Y1BSEm`=D@NGJz;cz9AvaH+#L#9&D7ZO}(vGGhdJlYOzURR#GEjzlLOR0Yo=ciAI_d+wWGJ0$(Gjh=jXx(dvT9_@J- z8w1#j#-wE#Q&94Z9@$p=BFWQ|df(Hye%N!+a3kT z1GgrEO@10dHE`4}*DJ6)|45R@d&bO&fHZz1_Cu?|B=l)6*2(6WcLJsag$#SOzO{SH zP1<~z9F@0@jLO6FTRDmWlHNXNkJ{%4@62b$H#zjNp+_>32{(z!1k|C4B8<$~tW;8O zThxN_XeV9pWV3DOs##FTM_)DbC zeVI@y-}_C>ucBw@@v;UC_`|?hl`AqxMvZsXsqrlBpP9^Xf#?h>{_j|5tTR#l36#Oe zs33$J++bv4QY<+DT`#&)@{mSL`giB~-YxR{PA$>#7giJJK&{AR`m7mH0(k32tLv|p z1IpPH7r(Pg39EPRp$}LOPh_$X{$NcRkv=Ywdz%)6Awd3t9>@$E)!fYd6(L(~!X zcHEprduN~|TaAaV2$8`r>#AU`W zewLHeepR>}YEj4MdY6kHb42EK{y9b#gC859pdFtA4!`&)!w=dqeW)$3LM>RnlveC^ zM~OLvOCD|cgSfX(*;G1a^GaBqqXW0^aqoUmzFHo}V)fBvuoy+6^MBu=K+T zd~8orT}UxAeY&9fXrH`iu1vtJHj1?Cv(3BjlCAge-SAk?b70S7eHEEi#;ixGOH8I{ z(x_R5VnUP(iaQ;5exF`AkGGRuS-#pzAim6$9q;IV^={7UD|G(phbwGg=;~8dNVN~J z>{qUFt#45lJ4{fAF#OtrM82G{TH%TvBURFm+f&ni)xF{SFvynuV`qnl`(XByiCavD@QWofrO=mX3b{HG4nEjww=xm!0=C7(*)hTvvz1Ci)K}kp?qDYb*g{ zKAY_m7+T|ks>uiMj!F{*E$HhT_2Pfai&F^*daL=an$GW16I$gDHF2P9hzq02rICw* z3Rx-xm8aubXrbNs;r37X!BH`kn#|Q;%EFC=qWA^Q6BJo58Mg7J)40O-F?rTmkRs{0?)qDW8>77dLg2#Sb+mFSy)G)Dby~RRI9Qiq^@nebYQl@FrgUd#b$8JAGK=-x82dQ7@ zO(Lz%J0Bx_uc=goP7Sl61Z;m#6Pdr&Ye3~wRDL(BBZiSUe|u5h+Zc6($56YH-W-3a zMWN=H>$f+BR##P(U>lTQto43f{zlb9^mU*i3W+F;qI$PrAd79{^G>N7qIR~jeSg94 zy)l5j+AhaN(8bBlSQh-QN#yb8uL;X9gbQYu>Z9dRlk3~m_s38>(YdEa|@Ol1-h=Fn#3tjS{cias+%4usP=v_n6U!#Mw;y*-y~Heahg610q)r4B+dc=b;3~Y2 zjV_4+SoD-B`c1%{Hl4Dh?3}H7W{eCqy0&bOAs(sY)~eH=0Nds^Q8{DX+E_Fq9 z{O0TJ_%|HFHc289(mnV$Yzx3yro4D+zN;tjU0G)`itxosqL(*YP9fGolYc-9f_`lz z!rP6(LGE(IYe4=VHKsS?uF<~CKE<1h;0U91H=dpjDq{m;y55l#hKKD!L^Vsy@BwM} zAW=yaIC4N!V@MPItg^*nG&o&5VIhFx9Qd`-TMW_0-5rG)0l)Dn0*4wQMEzw8x!8AT zt?5VdoyhdIlD_1hWcLv39SF{EFk<))U3{Q50$!bcuK?XqM1%?5BC(L*h_D8aXHi9} z5TQci-~nc7(kN>96ar`b4~8zb@@ox2kW=}MFB>HcVcFr$)@O|@p7G-7`aGU*ZHfry z%#uiri!s4FqXhaw(n#^cF(-H&Cesx0x5%3V%Hk+NVaa9p})}1Yb6~i&Hct`=Tn*6#=}6Cgs4typqI8i=X=LZvw(qbZN-e z;q^QG>B6`wN4b)GQi3t9jz4-w2?&E?*g`CWehb^PG8whBGIGX%>8qg@kWsSCNlV+H zd_u)_9(sWUhBYs`J=B1OM?~i`uDTg1Y`~IRuDUO+atT-IgzRduYRY-?f#X_qx^h8P zwY009LRdEkhWJ7{f}G7K2nK)ZG@-{t!A2JknV3McH+{SF0mS}<@?>MNp_!#&$74GruMsswm*666Tn`-B+FEQ!$%GNEUzy12h=Ffw){;}3 zc+$c^({&z->#igk*;kaU_6o}ZXOvk7`j4oxbmN0glB6F>t*C&oJ{9-lB^l4uP|VKn;+{6e05)j$t9fdPNYLJLBY7y?f2~ARyq0wm-V|le6zEOYUSC8oF?GJO=8~7{2O(5=yjv2Tj zGZ`##V36ji60(~iDY}AJ?pX|Yx3)ZSI`#wSI4u3$I&xZYjo`B7#|G0~I1SngB3?pR zkn*NS+Q1I(Y4%KP+}4`eJM#6EAy>|yD_@Pn$(9#`^xKhhu$1mhL+m zY!-rox@lU`kcGEsiM-0lLELfo+rDAeIY?J~jx~#EWSll}NhS*_ZnW@aerUXFvNgpv zO`JdkUW{jQCJhHx$JA%k+v{-n&5Co{Ps^QOEgh75SBp4H2tZ^YP?F2#P?D2Z*Qi`S z)GSW2dfiHjNzF^Zd9X*i*`IT?3cHXCr?L#Yau|oqPje|&Bxb3b#zo%2ThGQjSAbA2*1TL3+0*$o!ju?nPkJ0{(QA#_9=?gH z#`Q!Za2nzH-tqUN_YR`z1&S%}vFXR2SY$9ykDW~dqWR^jsRO*mIvp5)S#2AhV~($9 zn(tr=>rGgxw)4buJ@1NHB=*5$Hg@(4nD5e>nV^=)Oq`!^L~cu5E;q)fNt*Lb>Lo(y zw@o6HrbgLM8iIgEfTUF#7*!t0Q_P?lh*3>fGJp$XHr{Zm*APw0$m<}HYBcG!JmTuW zsy;(r#nvf2hMd4haW@A!LdJiao4t4?mHa8SzRg#qO^+pq^T9#?cubqjfad+E0A{up zS_W%s4W+{d?Sr&EP!R$wSO0E9@P*$VxRh=KG5j-4L1YL*vj-Ikq<_{;-#E2yeaH~B ziblZ;qeIVRa%7-Z1hK&~-}m7AM<#L@zFD#bfrYb#;EF7=q@<3^+QJDAAL+6Lbe+roccn^~({#Hq>l7E8v0_D2-%Tc_j%uGB-LBFw~IU#-~vGWgyg$ zb`ueI!N_~ate$`bv#2CnLRq6Q^6fx^ju|q95Y!?Q>D<;CH7HMUt!8oTd#Jp>_TX6| zy;R9}#UdfcA{|&d(mu?it*AMmWM12T;Wxv1;J?l)704*Q?ZY$erP)0*`O8AJUzaoA zS4M195PwrfvQi}%Kl6|6-V$tCp60b zryxND=>QSphhveb$g>U7D-dFV|B0;^d?w)1LI6M+QC_skGA$H&09RfAZXp#yAkP$i z9Vcp;l7XmHK^^~mc%!dT^mdd;y#IG$!?#`n27oYro$b09VPFWoztC;j-DtP}vr6xr zdu7zwT@+NTEO@2`)mar^H72g#`lyT6Ua#EyQzhKxa|Oy}x2Ji#RcJMzN?*H_zvQda zw-0@s_H_nbM%1nW)d;PBw<=WUg1VenvlKry7wPlvZHFS76HT+uUhaBFrN4--xoq|H zQazrZVD@^qT%r0MD~ATHECN&|cEfmmw|+F~Ccz~NwJ;uk$+RAROH^=Mdp<_5&#^0y(?u6N}ZrbEqinF9A zxr(6WBIqS5(B~@HjoTJi-ozK`s)_J}%TB*5tW~&kpPw4L&~0V?g#&bSDawmP zZ0oAfEbo;SD|u||5vnT~H0pCS9n%|7M&OByrTjDfcugkNcPzCX?7n#j*0T(akz=Ts zmu=iLBwyQ+ifN8%%Y&k#vRhrb~B!Asn(te`D(jEQH+StlDD*bjJg6X_(UCsiccCnjy&B}#nAnrGe`gTPE+ji4t^ylT1OPkr~R;+X(tv-FjLF!39s=U@^EkTJbZOP zLZ-<;hy?P)|HR6c)b=O+ty4)dOQ!t7;<_+Bd2@eyk%EKMkAU2unO3D+d#<{sGG_uv7O_Gy=ewj#>-gMy=*UMIG_l~xJ1yj zL{18WRvFUXHXpC#d&UzF?n6r?ZxVD`HBNFXsp+3VRYH7|Y8DtqB@&X%wU{>`yx|RV zV))4LnJZGPV#6PQ-{d8kx#nx>Bvl9nzng_LO)r=OaC` z7I;a;$dgQM#7d)NKg8)o%GSXTBg-tA^mokYqE&dN%S$GC{|a0&{Sp1?pKF5$gjnj% zhRotj=pr3J&d^PpQYX@wDde*ypU@YTErDg+AcyQyRH-_;;;O;ughodpc9?=_TA}2G?=%iY2to6RhHHP+OPELO_TbiblW{erqw%(X7%@#ogArs;Tjbd1|R-!jb zA}DW6Vnh&Bd0a}(mX(o&RYm0!N!syVWZ-Nu7z4O)H&71@KW9F4B6F)dUd$Suyc~A5 zrm(w&({8WBm%F>)NWo?qa`D=$h;_TLn>}3FjM1Bpx}+$ZU9vvxtOp_6YKvV+Dp^v0 zH`4E|y;$(=G2Tg|ym6Aa0{Tx~>>y|O4`f?j2U-T>?xYfnbzon++KTRO%nSB(QDBOX z{TX^%>mK@x-R!v@{54lrmwuJ;r$B-kf^ zH5k!{&$cG-EEPR#A*Sz7N!odYR3+O*QR2H96!x6oFGz;MAGm0vp51&oWnl@~qKTBn}kV5Gen zX1!9IhOTSV*v{FJA9ajmx>~I2 z+fW+S1q?D-q%7m90WRa-XGxXAod?w4U393TE@OYwS}}+BOojWB^my=qJ?w<4QahbVxLu5w0t}|=N!*1LyCP%JAqMabT{rq?Z5MQ7{+`5E>f+pQR z>@E$3eBzpQ!qjXUH7wnh^&=RPwsvKLD4sgKeJTDXcr*w@9WHg< zx3RfSy>W9TWZGFu^*u>Td-()%zw1g~^26{6nI()WCy?amlhKwy57jqR9X4Kh!S;ZZ zDN8&5&_@M}27AYJ9Xlk%;O`2-MSjp1zu@#5hxbvMq;yIj`NPR->@W4npW<95rkH*C zME~~qUgrh1tcD;zgfL6Xzd?=MW{RYliU4MYW{rmO>8bDOYr1iZ9_0}o+nav%w>lRj zVVw5l(3bI_3y1Fo(C9~2-9)mE@sHbwAaUz1=XwKxF_B-aBO@X8nNeemQQWnGgXm}xBvJQn;Di*&*G+?z<)_|F< zm09m?IU70(r(G2?AtCUc!wmbGq?6)29*l!Du%03ONdRGyXoB@6wR z5Rzw*+bX_bHd8P@)Z`#2F+O4cyD>_6ma32Aq!jOtO++Z4v~+;Q2Y>$ZqkpL%4T~A2 z=}a23PX3uYVkAUjI7sA0&V|TE^EpPrJ6h1rQN@NZ@uoFfpAh zdnb;S~2$Qhc}!9oj)z z1X0e=CuGGX{tqfZi&q;SEf&WDN?>^-7nltnhrMKG zztzU=@{#+%D?5x(@;fe816@NFNRoU;sk*OJJp%ytGF(Jqh-9RKPzc4>%(V3M5?%6w~oe253m@Ghj>U zKt?>|IFWU^`6a3iOKR4Mk`%M)s!ND+at~ukVtk}wgxJP+@)=N1SuF!bA^>&l2zN?z`)g&o>~%TYq*Rn@}idS&Fjkm&=A4f zoNls46qIsrDcjdTP?J;MjFHxGCn2*8Zb+7B!lEeod5)W~5J?r7pH4kRi;{`FfR((g zgOv#lPi<8j@+S%_0)!1G^F}GJ=d=m;7RfV6yMdwuNcWHo)K%^ zO5VOHRkh6t@+RD224%-EbqU)$D5)$P5=@b6A4*X^!$^ONB(fbgpC9WC@sn*DF|gLE zrV-Z0NY(1IHqRZhEchl*0P&BxOa67?vc(H{`7B$I3K@yE@z8&TH3zludbiC+RofM` z0#{qP#t9(W9Ye#sS{J(AlB$NFed;}yQ0ur-^HxI*I?~%JrD7xh@@LX<{9}u|p zrFcz6yUi(jep?9z(bvk+;=K(90g&`VDD?YyQtp;}Jov$RL4>e+m!-92AUgi)6JFd0 zBEIS51EDvvnxux9`2W341VTtbDEZ$}(P-gv(f{ePV^$3^OyEM{$t?ArEf<(ww7 z|3F12g5A$|`Z6ulO8z66#r(NcKFiIfu9IaZU#(K9SgEUiwZWjn>0d&%VZFs-ERjmN zyK%F_;Y+>gr|zch9*@J>V&$IZuOI$t6MgDw+5H%Sgh{Q^+qyp%Przn2*V}eDl}@i* z@-LzK{Zl@#^S8Nwsp5sQ|BH%FeGPo@{b_y?*9>3dWsj*s9_Y~Yx8uL1-rMRQWaUnj z%#2{3qNfoQGO_SRT)+mX&p2fxTp%fN008l z{kC8C*bn<#bIrLH+APbqY;&` zuD;&zjS)XnG>+_|X0Ys(HV3)auR7s^u3VLG5kj6QRpdWSCRJ0X-_SMUA$pqPUsYjU z(wv6j)PvF+iwU~u!ax;8<7PVPz0Hf?@%Um(vRM8rtInmOtZaHd@j@11W(buUb5zLR zUj)cl^32-WQk_I{RjK$U?AH4i+o@8Tbu|JLjK+wG3f|!3bE+p_I%kZ8PhHS#C3shE zm<3^|f1DCSh?o?BXJQ@fgkR2LpB-m#X|M4^_u4F?P%!zOh}J8NJtdA1F*C%Nt*t)6 zp5M=uUbh>6rSO98#4sSs{#15&HLo zsQsG8&(`vHw$tQrA6AYz-4*QQ{uBt?}rbIo?Bd_|NEuF(}_& zu4y9YW%xG$k$hkpGzaK#$e%N+urF1ur_A%A&tS!k_q2Q3%T!0XyKg#3V~i5AHNR;z z;>s9yDOaN4M!U-!tL`{NZq*-O4&Eu?MXp zT_z#9=cML!unEVVsf5U;*Tt<;Q#-n=F^|} zSD$LX5hb3g`LN)}A^S;>(R|EYikNiP%>sdR*)^>8GX)wQrZZkFrQf79h4{lOXjMl+ zhYhIXFYQrb1#PrV*Rvlu*I~hl-G-Z>M3c|&d4>va7lWh;E-eI*W zc1lZm8`?0$UMy84^QEGDF9z-?2i2&vrQ$y8syC*(50^Ar$epsa+^MbzlSZ{p%oHlUd z-6$0~xem+I{yLXY`!2?!UoNsTKCT_ks##yXERYi0r&SyOs$e^@p*KR*S^P9jZDh2+ zUff4OSIO@vwPihF{Kjf8tps({`M13Ou1nb1LTM_4Drz)3&fk!9d8R5QFZ9T5-9Gfn z@-f88DEFQ>EOe2@A^+dnQgCG7oa&_Y>!eAPxnRG->lO1*g>;?GiVP3CvqKQZ{6-&P zZ!-NA({X@l&9Gk2OyyPn+djQMmU83LO0%~(_-EunbR=q5gD(o^cO>#r=%LQ6AJX9F z@^e@FVB=|!-E_*H>xaHuW!LDaZ)-0C#2plW^-b>LOnNN!zoP8Yo9>e|sD=rR{^Gxm zC6ul-Ps2#Yv|#aGC;$0?ObpK7qN8PGHJXX*8Rc|sRbZgdx|>@alZ$RejJWUCYjV@l zvuVv}7dtxET>_sG-uR;|nr5avD$Pz0GzRR7?Fl($4N zFd!vVvy8@svnO(*eGBZ&r{#>>_R#j;W)X|Hswuy@!kt{E7w7tNGd$q`xx**ne+%nc zmkfm?BY{+2F!Xb|N3 zJ-CP*7IGlYb-MDkgLq%3sN3{lt!>G+s=BYwX*JC$N%uLc-ut3{r2aiv2;X0}H#zgc zB&7RibEF=YYqP?Z`G6sOEVV$vm#_PW`Rnlce&jmihK@}%Ik}K42;!nS`vL-<8wl%e z8B&3|{K2w_lF8Kd6%z_Fz6aP_dJf!1mBo2_FUJRIUCU#0h0a5*cs37Xv^*JE);5ei z<&IJjX_d5dysL*2g-D9VQte?@zMwCfVakn zqCkWJrt-{><7lW!cPrWImU=7UIsuobewG2>Q_+ZFk^VIsS)6F2*j%ttG?h$UaGH%h za%w3}ku7CkMNBw0`hhK^4b{_G>*9ZwTHSg{Ovw^Zq)9IR;C44*qin`d`+Sp5;F)zJnV>U{G-bbOIy z*5W~YbT_E{&4<0o7YUuRS>DgyKD^hV6IVD9K7B^A-`|OzClso+_NkgQ$D!kKT){=U z@!;4XpnHR%PPTs7pqy?$3j@mcwp?lDKBBR8mE1d9c|Ib9gmILe?ZTJTI{l{AP_W3Q zyr#>=dy$-m&B#}!dfQd`bD;vh3Rm0NxW`o43|h8`)Pkl978 z9(nU>*JPXSSMl9jur@VPYXKD%s3!W_1&9}9?fUi$&gB>)qkz-}Smil>^k-Tb@dCr< zCGJ+L>-%$ye|Ob;|M2~0=eC^}M%+6$BLNGP_K;NeAL-R1zt+Yt$NW7;G4Kd>pmDOU zcJI_Rer*hBWCNu7et7%*?OD?{8=$wz_wcNeC&8{=Pj1c7^-|9mQde1v)7`S#-yn?b zdo)F@KB_C$=FOx+gQj#6o}SX*K;X}*t ztD$3w00=;D226zFAs6Q59&d{IQ?3p=&XeaJD+d<~)X^jm=)={au!x5ptVzXWgsZyE zSdl;EjH#i&8uLuhEyAXD25qnoYif_eAY*!ZZyH4 z6&GB9h`25yUTK*xFd7t*=|qXYsd0f+3ai3%w)Z!M&07p&R*EvBv-H+Y`^sR-wnb=D zM%Mf18ZpWY>gP;(p{ZpDI&v0mnYnbnTE~h<^>Jv7en-vK0f@WQac=vzL6Khjh#G|~ zD;}K=P4lpt6~j*}+>wcfb0u$e$7j?eE3={AnqC{{bb#XRtU-Y}Li#vwx?qklbz@|9 zUrk%1@rJ+YujRIdLRNQu6LKvxE}a4;SypshW{TUGv$FfcBD|>+3*OWP+dtBkm<9*g z7=G4#|9ySbT)Ix#C>M?pmbE|cr_n->6q8!SMx(Vvoprz`%)7Equ^6x0WE1Zby^*}d zR?Do8pv=j52s|f=gKY795Ql1Q=ElhzPIE@xj_DJb(NAf*mMHg%x{v4iTUS}12Xxp1 zaYR_XnpS(#*d1+&D=5*wclpykhPa6x*xKE3s57E?#8FK&aPZ zss+TqwWAFd5MY^g^-BGN3vYHJO_0@yZ?}g_^!Iwqf2M}Y4%14g=z3g;Dzy?Zg+>k4 z9AETEHf5>>4Gy(;68grSN;w>@fjT%Pr&a|?JGH_rwVtY2bI&S&v7)kCpGKw`6Q<50 z*v?zehgVd}MQSS$o1dCN!y>4fLu%qa{Tz5ktiHb%0M7b{R9#x;@VXp%R2FS_`s_ zM+2Ue*1W`kTEkhmNi z^RvEZ3KTUig;yH~^!@lsdDoz70@kSQ(`_(Ktq)>dcK=1UnL$q3@v?hq;_00aWI#8F zbdfIvxj22%$^#S(sN47d+s0(u>=GSt`q*u>h4KT&IbBFf$nIU+hKD+&N_quzL;NY} z)c758Q(?kLJiPUe*fkNQcfq%^{p_F%J9uW0gT?gyk`@W4Zsuk{$xEj*^dh+H> zG72k3^qq=RK7p=6c&~@DxZb+zLMGT!6bYa3gEh>tEifzukd}!oapdIHduTpAf^I+R zCiF-*ypNF&M`>%W=~57T?+OyR(nLdWu?i+*r+{b|DekT1&~qCe1sfV}J1@Cg{CHFm zcp#FR6|Gu18h8Y@q6xxlrz=a1Ga&JH`V=de$^lPclvf62~jh zL=?j&UxME{{_Kx-aXj>!B#fJ=sL(&WTkbwvyVBe;W+|ht*ULK2(dNEdBpL|1ZM=;9 znzVFPjN+yKRwo`j{J>BcO*pFW`(e=FBT?*Mrnh12cOaU86&oTgppnFMXpD@8o(lc_ zDFw6E*wQ_X%-x3=4RaHO1^uHiX-SI^OD)WuT&W*$(gl5(j8E&1t7wI>)C#lZVsfFO z8hXS7=>dQvJp(4`qYr}CEo^lQXkV?hx}FdPR|Nt0d#YWkf{sM4 z8X1^%FpOG3U%D3oRHDP;vlxL9o42nUCALC-FBCz*KrMHuXcMV=Aq)W#;w2FjKMNOA z=@&`!7a|X(T5Doh>(dEqmR2e91a?MQXSfG3#-(INS0TshSc|ym^HCWPC+emZDETkv_$y5KXHRgApit9i`r-)tF}K(4mpd!j7svgO9`>Ou`_H`( zQ)16DC46d;2|gyY#@soi6|(3i6V=@a)ElOX4ASK?3p)Pm@o0F?C_OISM>4A=b#D`? z8yY4$mR0jsD0b}gXpE=@O!!Qn`pgR?F6uVKiHJ?cIaT$g&*s&|>48lYopUt4My&RVp;Q9M>0K zx8PShdjNmCk*Kafd*y zCKJYDGQC_f5b=ii7SFQt3sSGyEU-P@X(tY6)*r{8BMSh+NzIf|e(Q4iUZYVgLCB;B z+3x@Jcg{P0v6%DgT_6(n)`ySH{xSGJ&SQG6Exs_8ul^OxX*-;X=jN^YEX=DjN%#7s zyp?>r8(c|TZ5!zCtv^qDsc|ot>}@;V<90q5L~sucxhEj!kL!CpTOQ$+yl2;JJKV@C z_e?7K$LpC=j!P|h;nsh)x)osd{>~Ke@%Y58I(NX!<3NfKL^_yiQG4*CTQavd!i1DfG8oZ?}&&Jn(HRs3T9qPKqNb4GAd1 zej4PI`O+J9?D%6n>rn=HtRxPsBb}vVD{UMktm&gg!B3R&=(w5M z@uo>9j4*c*#2iZ`qHOCleJr{o8UNvX?R<^5S_3%+9g&pE{n(=)RUr;%r`08K0`mr1 zln3#E>LM2Px*EKWY`2eS@4)khwpFF`valm`EoS&caAVprt9_0i4x)ZKK2x6Bgw6doNLEnj)Fi?-1s?bzX?q8N2;;B=LlvI znLIo#QU)f-oH28JH~Bk&5r3FGryt_2FTMdlzy!r{ewnqCrp)XH$Z36I3S-2{X^|3l zLO|r23dI%y#U{uA;{&W-1rv-AuI}{p2avM_LP1N9mxG|T#QFN*=q`)g&Fpelyab5pu&4=qM_FHm>s;X9O)*@m4#4f!oY4L^THLM%yfYuaAZKd(c}|4ZUj(rO zI}&pO2a%j9-V+}l=PTFgfGsynX!j%I3o` zLwX`HE{b~fK52oY8mjKTH?sbUKFolx+3e&ROSzgbDWv;ig}=Fm46>gTGK)P9gMTWV zy5KF|J*@Igp;^$AhwB?k+^2pAP=i#6A$oFd{%T&}9Rj5Ofn{2@JFY!}Bf+oM+O0{= z%+xrSkzYF9VVst6(?S5Lso!wn6t8VMJE&z;U!P7ROJYqB0<1DLK7J(6Cz&9TB_vFZ z;to$71!g(Za$|h?&42qiGr-e38lC`5w%Gec6>v&kLOPmL-JgLyrBF<`dnEK-)~bq+ zw1+viHgc!{i{O|(r7Tk;YbT*fd{C&luVayVepgk*t$ptAqnl~w-xziVt7Jd;iK4p- zMu}sGcW$H4>G`4=iI6av`7pAMl85I6tlnIr&@@>%T8k(dBz2B(ZLcyisZc;1@`qy&UCm&nm$sJE-8iXjy_@DYDZNpJUMl5hf2=#Z8hX|g40|? zyF}N?6feA7C1P5;m?P7|0j#4{t+c{zfbXxFi>~q;ce(O;nTiXkUX2F7_cDk+RMs!i zw@5*|?zy8zGK#gYuB|r)4#<{<5zm3P$JPD3QjJ2&s_r`h-VSW12~llOm1-ZR&-+-7 z8dk0!GQILOj-Mv{5NgY>nOFJUd8gEiS$ z%-`_R^3cQz7svS(A#2^VvR8HB5((=@b(a~%8kH8+zaBmRvqVEA2S&qO`1IDOWf$H2 zEc77nbh7O3x+FPP-yScpNYH)lT{=yeETAh)FIKJ;70 z+DNc@;BLT@4U|iA${V#Nw5}tmk;Qh>^Lv@&;H+$yCb~&)@7tm1@2tUED9zi9N#Bdc ze5Wc4-kDrmjEYyQxV&i!I4>@ z{|DIi{1>k;-(DRf6%2Q+>Rfyd3mroL8L?K%3yzpv{s}o}*#U93_Zs=Tr@3t%FtRB; znK2gaN1tyVE=Bvz*u1&=O&D!!vk1$Z&0XvK+ggT%bOuH>8I*a_fdK#a%*x{b2@nJ% z00;mM1r46$e*naPFJ;(+fbfY$mSKSM&?F)mwHHPYYsHv1#0q0SO@LzB!+-GTiHkud zOGxKzrgyRUdj{qpGuW5wM0!s!`qsva!J3dt=vx56_74#TE|0L zgJ%2te4S2SmdjeV?m54T$C90Rh|k}>?fUL!$$$`QrOY^OI>nH$FReN=U9I~QuDcQ! z^y_6A8Z1F~e}6Rp*q+N54Y{9LZ*TVRJbQW;>1Z2ie<%2sq)OiAGOYWLzk)2QdoD%; z`m}Y?E%a;O|59!gtMl+{KzI4C=o4kYVRQ-736*cqzq|5H8OwL7t6nqlzpsnMX2n!N z;@_(d`?65~aT|`iTqeT5vQ*uwALZ9@u$ zGfF@0U}b2sQ>gL%bU4j?`03A(#b_Ao(;A5=^Hq$X9Kx%60y3Is<|hxvVbQ^LJ3g&2 zQtBi0BM^rNk?y&vU?Y}5#t3t=GfJusK9+iS>g;-bQ1b%6Q1{6O%-G@_4Be$7C46ra~8tV zX6IM3Bbi{w?y}q(BnOR<5C`6~_V0bMxe$;9-<~}ad|p}CR#VkdyzWo%Agcm}`Bm?c z03@{McEz@~|8eYmO_r6E(Omh}p4O3PUOxUT3kzwYZlLkzu;>Fq7<8rKYeqarACi6v zV*ls+XOL{4hOMdi`h9PaT<5J8hUn`ThnpJMBx!w&zyHNe>;Ph2iR^)d^Itejt_6CB>K=|DmfgR{^{{Sr9YJ1P=`ehpz-C~Zg5-5wHs={?O0{b0D zU@c}yal3*P;>;(rqQf}9>8ezoBc9KP1!H166$ds*lk0c7vsT~EAsi8XR90yo4CcVa zN>IJ`!V0S9Ze!@PE0fJL>dqDNar^sSsLaZ?8iJ>N^3WRM`mfsW%S<4B*@@D}@9fys zo3j0hcWh6xwdE2!hq#q|)9q6LDId|<4?*b=r4+czv$KcdogJ*$*mLJ@Ci;Z;7k9JeGNl(&)@bEn&Qwq*m9kjTk$93?4;{k{;8-yT=UOWI1 z&}GG6$oJ0bmL=dm%K&%+0Y#?na9-@Y^8M|JvP}q|lLlf$37?WAJK#_qDh@gCNl7VmfI$$N-bQ#)3LGmnCbu&UXa>Kk8V9g4Vv4e!f+f{0v zE5-O4)*ODMGJ5tXx&p_8?{9Ij?}cfPbG`*=IOLl#!_Fo+wv*F86)7HOZw4@64lH@6 zN97fCCaJ}csb3PiQwybVRmxafc*=BbTs4n30*r)N(BNZcrlCs1X)xn}n$YZ*6+RaL z0V`A~(t0AP=b=GI%a18R%?oT%i7dDoVv%D?Pn#jt zU|cDkMn2RMgFMUd{=rq9se7PWC^%}YTdxf%*=y=dQRSh(1P>~~XXN%SIW}(XnPl*O z2T+YMjodgWI*zJ;zT$6g+fdRdd(Nm6yQjmNcGfx)0hWIJZ*1~@#rF^ETzb!w(*X3UUfDdsLJKc9%tni2RLUU~c=M0h%>_(`b! zvIe6;{Px=&;Q4C@FN!JdkVBvAFem+|JU152f-13d*|PPAQutgVM>`_(hpU_Z+u0W9 z0=~{9`1=uRLI}Z{x#2RlJXn+9UOGBC_ zTnZ3@{KG)Sfg-fAjUTQc# zqRz_0R!ye$dP=m&cT8z2%fq8BFfO$qvsb2i5^K9uVo&|F7U5+4dii0v|M<}5-FHpj z?7bgX`t;&SbIBp&7IplYnW!=zWy;{pKO)hK0l~_gKsG=EG4Ns@5*9&*U=3 zPW$$C@~7Ux2DNBIU82Tq$~a7ODm@L7hek}jY258)L&Uc`DDm8t(UqA zV^luslNJ;QysI`2Z;EsNA{$%&9WB&i`Q?mo<7MT1>cQRNl3s`U$1gxKCUehrUD}7h z7De65;o52A6%ny@2jlw%)-TmNHW>R?;yw}~U-F?bOKe!J9o_qW2j8*++La}*qRdUr zYrcIBM|w4?pG_zGY#wBITGW}~is{wNY6OsZ@Lu2y0$I?mALhGwShT7A{fd*vNBp_h z@cPmQYqeG3;2=BA#M&7(gA#pvtYYaGuy8vi1^ z3V3JyDV^{xj;Zr{7UMNFtNPVbyf*XRjh?=P{p(8%>-+O&2&jN*ga}a+SuH~vU;_fY z6NkPt0`j?lC{T3dH6Z#4Fy}hd`A;YoO&F$o=!0Hx94$}_1c<{IUq}z5v9JnV1}>B% zkxl>^gu>x1!im?yIUd8gauiDlBlu2Cg-i9A^&&)iBE;4rBpxFmB$3iWkrFTuS>H%E zZVhNpq{><($$q3dlcR=El(v2pjt0B#WQ0;ql<``W>0^{RN%Wnbnx#-=uy(Y4PW0po zwSG;s6_m}DB*sH1#!E>3qil@fBn@gzj3Xnl{AA1|Jv9a`Ec7H?K^q3kfyMQ}64qeQ zwzW{q6IdEkOiE`=`Ux51r`Wp~zr3|rEvRdeuXd?WoDFJVg>Rg>uWrqWX=i6#4U$C@ zNqjL={Nzep#|d?pZ+zccL`+Y7L1z4yN1{QJgi+s+Ed7L*$b^EWBIj70@m zD8#+7>0&b9>`KBhYGqN7v{eGKIVg*)YuXs5^98itHibRS`}5h%o}GlXnWnO#$-^CM z>AMS=_{B{Jx%iA~b@~BdcakT)#5N% z^PwIY)1YKIxDjd4R%o-7fXB=8yo!9Gcwlf)3qz37SI808;)Ek+n%p&b1YMUV7WV7D z+rpE}=T}YUJ6kLtseVr+>wkgtF#a|%`z>5%HX6VCO;0RDMEN9aX|?_R^XlKu0Zamr zGn|_Cws)?Hj~1S>i(P?Fy_|lk$E{b?vXye}3Wb-#N>u^mBiiUs*ubkUiDN|JnzAKJ z+as|DJV8$}-~fOCOB7(h*9>FDdNg-zlcNiR(KbC8MEt5;E%a5CKGD&dvw(gO7Xl;9 zWQNL?Wf-c36EDWNVO5x>8muGP$b1=7_>~vi>9EF$+qh<&sJCjs2oj0AU5<^c)W?W} zm)153FSL=f*Tq2JAyVFvvo|hROgg3z9Ryi2a_O4J!m=EN?X$8G!u-fnt*->sSs?=z z;JB#X$+J90>36;oz{QsDX(L3oy-Y5o^5n#UQ%Sqdd)@EJK3yeE9#^<)HFu^l6pBYF zN0GF0=#Di@PzJEVW+m3a8~AY?=$iJ=&YWPy|9_f0W;aM?tq5=#2J4%KMR0Xv9Oc2P zYEtO6=}B3g4V8m*T0bPuOk*)Yi!%%knh&G_E*?v^=j@;0OE*&7zkJEdi#PX&o)->6 zZ(Kiz%I@Rjs70|DPwwXtRk&2RCg`+I8WuG85+p7;cf4~sF^-acK+OLxfdRo8gokpI zc8k!S73Cea<$|ra1xK-mLxEVFS%aX`St~J*6<~wo%~Y|<>$s|(?K(2WPX~lwln!#* z4DSd~@W$tSSmB~GT57OLh!5|au--trEZB|G?|>_YMP2fb(%)q9f=wA_;D(n_=}g*> zgkLf=B40e+V$EK76{*bT)T16YP*H&vF(y5LTj`bdJ!MJplrGDt4X47o_+8`H)rp-J z??gnS?oy=%Nxl6bPe72hg^47=#4jg0VW|3(tfm@6qk)@5v_SU22ZG(l5&~h%=6pBw zu893(ug8t2j%E|X4UgoY+xD%9eQrr;;@UaPdJ2r?o+D1;-AdPwrAVw!?_rF!egoo9?+ry2SE70iVAAxqpXyk!wjgD4Kv|kisonoOH0Qz zjf`+B;FGFTI`4LT-ro$WxyAHOwkf7_Zdo=Cbw9FAG}5m+3`IF;Nu^$}*^Iwku;pke z&Tn}F;!K$c5^5l=@=B*K`Xez+f4nDeZ9PwLvvEdG_)q zxD3ciByC&NG7Xd^AnDF&XA>>+4idICjB94>YqCX>Zc)VDA{dd6;J`17e-?_Lv z^m5WDwM17w-HdIfiV(76N>b~xw70{sJBLlPQ=u|KD$epmStU-_xLTuiG+ethbVEjE z67TT#nsHt$zOV54uVMmZd9d;MP+X+$d4zBUIII@P_{NH(4hWgQPA@{574DtxLMhB? zLh|K)t%L5~)VG{1J+NK3$V-{Ck_hP7A)5Vn#CQWGPntl3q%E>oN6w zsg5OIk;Cw8&0M!1()#ge)gNS%sgL+isu`FS1-Yl{u{|eP6JsZXnJb8IXyAB&W^6hr z8D%1!B)*}S5IlA61wXBsz28c+<@VrSxkkda6)TBq4>`W*_8U$Kle~8s&$6Xf>%_0$ z(vmI?dh&G9B+8RrbyK^Zd<+rY`<&t4>+akHDZ(gpd{SW3B8Zpi=-YOFUSVYtuk}r@ zB_GM_X4LvtPP?RBfdMBl@Yf6|;5w_os(4V@3YYYn$^AKc@y#1J*nRt9;~IcEDBGv%0+*px5N&d6z!$EE0N$Z(Kw)Puz8d=uei~?xDO#B^VMqJAqd7` zBO14q5xqNA)7Jz{n(J}I8xP0gp7S5kj(C3+Qyadmp4jVo$%%Z{j5nOPs#pHBQbAfHX9xk@>$}R_(eXDI%aE+Vnx`?$?4uzjCh{FtCOXd7Pbcux9rwXB4Up9oa`6J*4l3Ry<&(>|z z(4A7C`)23se;k_aDSZ$V5-$^kyP_#g6#gqR>^reB5RZaCXFzm66dcMz=s0q-IYEfhb& zKIL_&D+nR^_4pL*C^>ThX^(+XYrq&NkQG-giVoQD{wF4d#OD-_;QAKtn$G`IsX=(8 zN~)09t7TXrDk%?1Tx)4~MGw(DgGP>T?DqN#_(0acv{~$XLpFFMZ8m_U*Abw z)ml8gly94UqyuWg7g+qbZ^CJIf=VfQX*Fm@i13GC!myD4!in!+Ug9CI#KYClEm-2` z$3zxw(+;Kt7-NzuEa8MC;;1WW=Qv@cEpZQ;D2p0DB@}gq3f;{~W<6v#woe|1GC1kS zzpN!+>Zfem%b}7IyR1a{>_&}urbyf4*?Tb@Go=C|+(~*4S9XeCzHU2I)kE?x{k(>0;~Y5>M%n^>jhf46pxd;4=V_ z`~T<|L%9H<03fZj7;0$QAOMk-=f66}3Q_L@Kgmx%K&H|ew6bJ{+A2d+;bo=?xI7t# zQm8%8jA5RKbH!-%qZNuB46`Bf>F_4g+9d;F9}0gaY-H?;SG+fYJ7Y4>SDVaJbtyMn zP1(b27m5f|R*S!yS)<&GzzA!<_CZ4|R`QA6e)WV3mfiX-9)Ax;!ovUj?a|Dt$){K8 z(+dudO->jut$=>=j&WF54bd(A4)4$`ym~fK&w2->G=u!topS&E`HZ#bF~O|5x7vEE z(^63(;e-A!XF(r%j`>C@v;5bUe+{CYPOHVa)P!?|U8ei)rWt=X*FRSy@ey_#%EF?MK}1w@@WtQ{oIr%A=y~J(JACRuAQcm}%%cvB*!Um+ z7&bvD2oJb)v*1Qz=J%Y6AigbrxK;Uf20YUMmtar zs8-E3lpKKM9E5d(pcV;4=ns@hLZ=<2$>QJ$6&@Beh@lc%p?0ztbCLr?yqJ!zRfaxy z^u9t_CZc(==QD|%_==PrFruD5wmK3Q*&hUT*7e=Yr&^G~f$0asR6?PTBWhCgL>+sT zEM+-lrKyL<#O~g)^2lWvjr8;Xd0CPut2bX%P$;zIoj==%1k6{MhRoqtAF*!oP!<)a zXh2_!`Lz5kW(pcJDtWmaiwTyHs|M!USvnq-y=dHz>RN zL6jOERZig$-oKbAbx>kF~MOMO>8GFLiv9469-kEDEYCG&TH0f63 z^@MR`*hdc!t9xILfPydK(~;8hv3lp5pnKExY56}{_rC=AF}9CUwIX=%B(XbvC|o3A zxRhSbV(v%>VnD{b%)+KQd-yQ8nvCHKkZ!c@wXqUnQX8t4n!+|P_@#qXx7?OcJAWjN zlUHM@@X=S=0aMMvLS9wo9(7e~nUtfq!K}IOEb|TUbYLN{BB;!j{kFg+&l97&$9np2 zo4m!0h1uG6hV@r*>YU3*{1snUT9j^oPpIF4s#dV-ly3P9c3m?=)(! z?MR@12Iq_~Mu*GM`M~A9C5p?D|8`zqPD+qEhe_+#pn98EcJj2dH7vGmb2x zVK@ko;JW}7DoIMQzkvtOdaOW_1W@2R5|?b7Mw9~t2mvCu$)Y>baVXMl;`>0mb~Uc! z;2k3TYmPcNJg(zSF|DMatr7^}X};H|L~-eTVY4+Ek>vRDFp3j-!Ji)^ZAZ(r#E3f_ z@7(PJ=?by`X>nksp-VmBrRAm}DrCARrrx!T!CzV^#E-Sa1DDB&$_KL~pRdzH0+ken za(mM;BX~d! z4dcxy2?N)i zET$37tWFcFQpLAmimp})9)ayc1 zU?D9Cs9w^JP^6UamG7FFwV}qTsJMhH)@|dZ`TUCxq9aLVWy*LnU?yCaXhYGR z-N!>Hq*Q(0D=dn2L4+p>iVBr&LP~LmTlf(LN7Nloe=+juu|u^nvz6r%8h8?hio2a@ zun0ru3Jry6JXVNAFF4!%65Yw9)RNMp(6(zMr_pt-pt6&R7GGOPnsSw~+J~YEqi7lp zIzEi_tys7{WvuR~K{<{uziz$G4%?z6is2=+N=!7YoArYeJ^_ul3#sd6B(_KvN)l6mGmLI-XI zO*(Gc5WXkfdosh!g_H=}nn*p=zCUOQmRUzW`X`p3i9JtKYrg^_XitMapp)j&cM{`q znXjF|m`>lAem)~GQ*lWP^K>F%(c%&v-O`e>ojh8#Utg}eHnL^z9T(K}2xE@GrnK62 z6a@EFOQ(6B9(3nCF4%eAl>U-BI_=OGAk1jW;<4pH$yT;&3q$+th37szSSU-MOwX5V zq!vEh{V{)2<0(o!xb?y$IC4V7&n(+46wCS&@Eg--Qo-95T|6j?gRE!}z*K5IKAuq2E0|Tz~z!TID8% zd^GMq|NoP`yxMO{y#7P#Aym~F!PYRkA-Z;ns=0q~l-)fG$I6F(^OUHbRmNjHgNpti zL-A#J(MZ(Ct5%n2w5KR6`%X|&dl4rXLg}8%7ZUF6%6ks8YCRo)BU zIloV1YO-n`7d;FGh?VFj{~Wdqp&k1zswj~Ba;Cegv-^2M<4l3gU(!l=@8ynrO=@&%}h`5`^Iq1l8(= ztq=$5{RnL74m$Pphu;lKL^RvgivXhvP#^fJelo>O54TFQJFX6kr!nu+vz#VYZu0Rd z@AiK62`7M=vM@oP81WtLggJ9029Sfb#`JXw!v$f!I%@%0@Sy-E%1l{5usyLWO&Hpv zx3*2>$4SqOj7U2>>cpWC6EE;viRsO74DMcx4{~Iw56)h+UwgMp+JSIvkGy0LEa_cH zvHLy>o*R!QHR5L)NnnO^UQTR5O{}+!lAyk;&VppY2Ily1}2G zlv?yn9MPlKE=^RHqEm=WoJ36;WlB2GPdd^k4EIeMzfK61iaC2sGCN5c$w~U7pM39| z{0Q^D%}G`nN^+-3ej!PL7fwMmNI`Z;Uhz$_N=-puPr){bqu~?B5l&r(rV{z3VxNZo zgQcPx5Ku7#>&8=+hf+?*@pHyf-Vx=g@-%N}3d_lXT zEb8@AW_Qnb=Y*e}4caUh@>Gphvvit15{A<~cTr2XTKv8D?CU=L`8BY3BVJjkw7DDM za)Vvp&3u0pp8D(dy0!~vtJ9Cw_JZsNqcBm7cp{;8TFYe3e@A!ZIi2Sdip`o%K%MM9$gqe<*U0q~a zPjXE2Hb+?YyHTMn-MIRyD|_1RV}4NMzQvnZD0yJ$eNWZBCYpsKleJO;8m1@;ygk-%RIS8MRm!89#j>c_s74< z`3PVH7+*XM5t6dyk-qIGHC#jO3>IVu3;(nk14huoBkihG##1b1I^CCdbo%fj4}P^Q zleY-^=*2+4>?)kJUNUi8GHaVg#ns8Avd{x1#;`$9X`=4IsR7Vw1n&6&M!y($p#xy} zFAa7kcQR^`LSLOUJKAr*Dj)v-QniFfp_=cVsgfWbV(+Q#`D1aSv3_$^@Aqdazlkcy z_uXhJ5QJ~jxewlVjG|>YP0_h}9IUeJa>tgB3Z_-AuXg=%t!PfP8LaEypbNox7ed3$Xg>faCK}SC_jM3g!Pt z*IWNZ!GC?i!_et~fP}!%L${Q44JqB-E#1;F^w2{}cXtTVjdTl2NTVQ)$n169yZ7#X z_w_vgz~}SZdB-{DeQ*U7X8vOF&MDS^;+R7H{3M{usd~sJeYTR<;&biNv9}?-CBG86 zmxvT>3k=2;GfG1n>n=5L-1`6@6x`DFvg|*2D>#Pk&yn+^_;1Ts zBAcXi;QV{Nr1!p8vH#_8r^muKtKO33h;NnG)e4;t3a@W=IaIYI)D~_o#$E(?oCdlY zdWc4G_1-`IGW(#UwQF*P2c+DB$P0$jk1y_zNuFZw`>xl}=+>`F%;>lUEv0GRz6qCX zxhc2Bw)oCfDkwdZ0S};`esMafhZy~&%*#)YMJc}8Pm%f#Fm<{GkkID~+fp?pp&SPy zUG#_WiAsY-tr8`jeNult%!a@FM13;1N*FK9WUoT_WwH_~LRsJzU}juIhz%3KAQ_J1 zXDEJZAN+KFToOP}#`7v+ItjdhmwkTPINeDJGQ2@d);lwVE5f zD#~=9a=$&b#HV>k<}Uv&X5fjBm2BCx)B7tUjGs>5v9-F?rTqP{Fq9?5f5^;{0b#+T zMAM@%6NX3Rw_rdMnTa)Xn1U~nSWC^Nr4L`+a!n0-OT2Dt3s<2|%W3>tNQ-)h9OxBe z3S0QZEKct*<8P`Ni0=M_4ZAopoZ32@Wa=eH51mHX+Ack6UAc_I${WO|VV4S?^1HK? zhE&I-Qb>lWl46cjxbzGw^gdPDndqgj<8Ooz<3}|q!QMmEJgRwF-EmS^bXf^yva;b`H(@s*Yb876YA$1z6YSnIfA5bSmiMPd2iU0G(E$S`K^k zxOZC~){HA6$7X~lo2%4Ukxg&i&v7cqd29_MUB=cb$hVg zIbFqYO~WI7I)4aYxsgvTxDpBW0(UqWdTYKG4Bo-WW{y7mpxgSpfcIv(r=FrE*VOAv zM;}&nZAeGex38S~fR`yhy%ca$M?;4-&ht&9bxww{Pv2mTwVj2funh;t!ccTI6AJiKz7t;)ii<-X#wP*gjKW(u7)4 zb9&0|4oi!=s_qw$$jDFMSe>#b%?Mf$?{|N&WM%hT6}+-sG~VNDd0?1vOKQnGR5>h` z?E2#OrKR9b1@7~Dcs7WDKbJnXi>YUQ;{R^I0stugz~+AfC)!_Hd= z{GW>m7^nEeEbvmvM6g!1@k5gtR~V;7A2F!KOd|8&I}d&>hms#fBXD#aQoql9l!zxV znY^a$O(e@z%4c;epU)P|U@Du;aOy3%dHO)2_!~0QVzP*FuGenL)xkN|cTUn$t=r~j zuwccO`AaWa;|k~bKJ&mRH)fA68hi2V`xw<&bN<`*U+cLLS|?Zi>jUEv*B`-kY=svG zi!T`UOwY4i2Kpb|XD5R>65 zeC0W#(e$P-vGaDVx~E|bnW^^RD;?H2g8Mvm&uDa49Y)|)AMELO^6l4l4VO@$)GUA0 z%2puCN-!Yk8A+6v?*hO&q4%Yi$+Pl1*x5$a$bUFGUg6Sz;?q;IN+pWtt!c0MWSNX-e`eWdXM; zUD_!`^L9Vf^`(h-GTC_!*Gw?i(QMU79hNL)%?j32xUy{Y>0PKCkC_}@wvXo%7_s{a zMEBvS5YYg&vJ{=#e;xzF0L2i|zg`9q=w|~t0&WB_oP|4w$Pu2O<`bFXo#YzF!kY&y zPM7nKRw7i)azw^emuGw6;1#tI-^ABY=?KH6d%1D&s>?~^p*5e>O5rv)OIECO^Y%pe zb?x#RjAC5NbPmn<2C@+D3G?b1iSws4ytaODGlx!JoE>4_D7Lku*cssati*`?dked=dI0ka)ilp_JChbSQr6 zHiq?JW&SC}vG`cx+1{P24r(u>T#9}km-U76(Q{F29z3WfcD^J-K!=cj$uLRD@h2;+ z7nTuQ9`6_jy`+XE#Xo!5bWbRb3jsd}1f@ACTjM zpECuQ{f{e8wI?O5U&~HP+F)IG18QFI_W(L7I71mUvE~AQVW5P-(NGB+{#U_PTFP=I{0MTjc%FVoaM}kktckzP@GbyI}pztLjf;UTs|F_MWk%$Uvve z?Y|Gf`3*>~x~MXr<69!YpYN%89WX|{Uxcbasj{9wb+#d({crbh=Fl5_tl(cSlYl$Y z18Lu1NYH$xF?+U?K=DxEDP$@1onDf{bZmUP7jN&O2k#~wN`?ge5|kW% zVTLrcA0#POk9x&#{`ZV|;FS_L?BTRf(iuwr^V5siq7=&)QTMVZpp-ZT{$1=0$OsG_ zGf4f64&SMA;5AcYV)1}~h-r>e7=&y59~O&!X{8=7$yY$e=QR-eL&JrNidsOM9_*2dYd?j5f)`l-t zi;RqES;!EsrY#Xwn$aT?)@sSb~OC4tDr9_I2^`BGNa6rxf*BX}$ ziXM77kE})eKdL;F3u75^6q}2!CzIIqlzK zCsRQ27n{Vexl%G$CN}3p_C+bGA}_LlP#jN8LxjolmRj$Dh!?~>^?7LSYp@lTF&%Za^z=|c^5A=HQ4;#cERrsYY za^81+x~_or+U?SC6h;5FhXYj>!fp0O_6782ZqYmfIQP=P+K$6I%fB}zy~rE?fxal> z`kKB-E!$?5+&lnmfW&baM?0 z4?K~uK#CEktHdSHl2BtC48g_qu*;rVh0M zSQ?l_iBY^x_Qa3tOn~Z|qOlFPN0QuPnYMeaDb^a!SQhS*&U+0#G}bKFOkLNmZ_?Lu ze72oez=%kRGhzldsqZ&a<$AJ%Y@p5_5lXxob4SN)fIq>X!_UF;cm-X}a+TJ^NGuYj zM1HzUQ{&GN2FI|HOheWnoB?RA#GYFo9&rCDtrP(RXx2xSA2^#&=x4GMt2cA%#3LrW zqJ|~Tg>docl#Aru{eoFvdH0{HimFfXD_y-CE#Z%cLFf&9E$fJ6*cqD7*@r2{cI9{8 ziS9`^nI+pG zo90RWW!#i=uKxEFXqu}%OsKNcZZw}5z{Q^gY5+=b6iwKRRnvBT3PvDSBH3yLly1+K zHgEAcG7nzq%lwh=ux?qd3lTbHyjL#?ZjS59i!Ak=O?>~Ju^Bj$4ifEWPP?2(#Iq!t zsPW0Ne-tf#%8^<~4}O#=1+<(lJcx5#o_C~em}8{2r(;WU^n1pMk#_Q%%Ss!f6|G#Z zEc4xA?Ym*erv5t%U5DLIecm`G6bgXp95K*7htBVYwzj0CI4b|Kgo=7~^iTllDOyCw z`&2Bc0k@6H&oC`~q;NEr2i#hyAs+8M$o>s18v}xFfe$5;l9OdXA#|-l7h3`{PLREz zUC`K`Kvd+qgLixRPhkJgdY(Q zbw^cUGt`w%!syUw{va59TAYgMXh~%vI-@w&iWxq$t_z3XAEhQe{7C+q4nON6Oxlh= z0ASPgMMxe*l6cy9&!J99E}-R(pY0(@AB0if{03KNPZ>+{W=TxG5Ua$IMw3%)Q!m#7 znDN`TEh!>Xz2Cgka3iEjc!~gXv`s06qJYdPT2gd?xU`>WpPpGq*}a)qk<7H-4oW<6GpW?WRCF^{O1PSFd5_l>VqKHL zZmghUoc>CA^b0O+jh_pKoE*g9HHsVv%MhVZGC?_4``Sg!2 zGt?gv&rFY@>j}qf?CLGc$j;yDozjMrPqwuBEI<5&@h z&YvPxh$FOx>m1)rZH{Z|g2lY`ZDVTIlXz8{tZZx(9Qi(W;WGdrU8fS7JsT%EK2E?1 zRCHDLk&qR<{nt&dbC6d&A=)D{R(9mGdS6 zHnDTG4Z5Bll=l$*$!iovwc01kUPpf?V7aT4|4OgmsD?s2wAE@dZ=Z228LXbbpMXAy z0csgvNT}sdKmBS<`BFr}VMQPY{jcL%ySiPxQeK)rsxz*F9Dppi%5-)5K=D=Sg32d; z-#qqtwZMn3(^TGGn8qjUn24!5Dc_nRaxx<wohm8F!mm<%R5`7V-ywkjQyu zW{4)+U-i38jlVxNYr3auGxt8Fy>e1e&R9v%bnO=>@aV;f!~=`Bly!Mt-& z5jSX7c$MD}E?#T*w0Ks3i1_+zUtRW_|2bHKv>|P!EkS|6&f|H56EfS0f*)?QAnywy z_pIG0szlVSZT`aaxR55?_ubgY{mSIPER=mnnMOv^oRRoRo9W~2)@HzK7)!qgUB ze!;7WkoR~jHAa|&u?4rS!yFt zUby$68!M)py4FNkVJ{Fw6&N0*eqnODUX8u=B&<85c8S6CR| zUHEylo;uW3(LPcSW~^8lYE-E$QW<7uZ+V7cMGi5ys?;|)igJLtDXvB5+FGlUIk-V| zSOzTYup+#%qNyO!f#WfP_c24!q0h{)A$d(&dzi!zo!ClP(%3x=t{-!g6_SF`g5K#` zbK@W&07NHt?tN?qSzJmd@mmzCGC!AMsH<6PTverZb97uCS-jDB^kcD1vtN8CBEGvb z{?j<<-CDdhZDhYt!mxeG5y_98$xmy^&-cjy^8aPZ zk0~T!WW#Z~;GbY{ya)Is;y(_q0RQ2($^Spx7JwTDK?9`F$;OcJQ9lX7OB7IOBpivq zCznp7LR2aCq3YH{aIjTUo?;8rXeuA_c!~UFxgvt#Ur)7DvgUYSC;iU+v>_ zOGnztTnSzS^?3|Q37RH2@dg9DT{1QM`el`FR};(tSAJ4C}kJ55*=6S6xjF=FE13|GIst!!$7O zHLT|$_^FfZ#1;0}l`CCko&Co=B{@O4UbCB`88eC9*&U+G;j33hIG!pMpIbz`$txyI zJoUSu6G|sYrZq#0^Q-B_l6Uvs0XGw?xZ5)0BwZ2#H+}S|qcrNJ9XS)a$qrxg^%l%A zI2h;beuEQg8Ec9#)Z6hrBP^V`>_B{73}d)g%0H}hd;W+4G_P_5a-prZ`vLnwqP=g* z;nNBdRy9Q=q=iC@^ZbG5YgS>-nPaKpK<>^T5-+6JbQFDVhZsU)0nHN<=aK6s{Nuv6 zmYz332+GXLsLnYpwgR~=E>5B-JZb+_K6>tnxrf<;Ct6e(e^tJsx{^bE)k$(2cE3)j z>=)!jv1FZ>0o+{K7At`v6bA>&|6G^}f*&y&vctxxOABlL)*T(C_96`qB+M;^PFk*0 ztiQLr;7GnJ(}WME49kDx?I=;eouXJWglgGoDsg2<{|EIQ?}O^eLtA5OK61X3rz zmRT`AT*qUUd6|3Ae@NVPup3nJQb)C^Kny`OBf z5loYvFM3+rUc>g<{8FKnw{>3k&M%jPjcU71sw6~Zu=#%_^pn_$1eGE6$zpTw6>z>KPHUmYm!&1?hh*SIcvl&JFC6g5+3hqiXHoPWuzj=Z|NR#_9u85lCvUKJui} zNFWrb_;W;-nDr#K&43tBF-y?y<}6OcI*AK_nyK$!%_4=qqz&#^;eA7g_mCsG9zS7D%O}=}_<^L5zc*w3Eig^i620 zA5>ooBZg(o!v`N-N); zxC-EW+*3v&TQJgmC>tW&9dhKU{I}In3b+Baqo^rNOShFa>z;F}X3VzSDrc5a)IQX{ zF6JSrp<9QR$z8T3YUJ-}E3&fnnzn`4&pVoYhPgM%P13t8 zMh}zvUz#&JUR%*AuTL2hv>?ph9`NQqOy8sOr7tHOyyGQYLg76uRr)gd31oEy) zT9+5aiK*ne?Plqb1J`e7se9JcCw|I_88^EnEqr8%shm?t!?<5Z_d)#WgU>?D2hI&w~o6O*rvhAhOP4FhGk4`FZL##LU&ty zhF1#ba8#(n3N~WVcL*K6DMpc9VZl#?Ay;U4qNtz+XlvXzl|5j&*NhQlvDOaGNNhDz z;$#}%Rn*dnG15UZVO~z8HuWx6M1feXOmeB1lEx zMa&OYhXq1Bw{u(}6#EdVH2FwosVsu@V|kqOS1efsZNO?t$ppu0H2@^zq)I_)fK@ue z2e}ljxX&WHT_<-CcUs>?Dmz(2awAZoV6O4X&|aWUFPUqK^^v7=)@OgqZAlYBp}!{o z$I%t-Mj%6KIvgvs(Uj2{Htt;0R+4InE6Gl;E)ipNmP_ z?jP|(6}K(;pirYI;YAJiBe{cgOuX;pI`w&#lat@4Fc5jAT|{ zgVcO&?mYg`CNOQB)Y7r8CWsGvgM0J!#lRhY;zn;9DMxuW!3n+!-3GKpaC@Hpo=Hl_ zO4V;*w0ETK03$I1i+ci0k1u?gn1r>Dzs%p%Msj$wTv+eX{7Gwl1mX3aV>ruXpmd%3@bF*8K&JFpn9k8N zr22ujrC|vgDjXpdB{_bz?IIIor`!$n-~W4Pj)nnqo*F3i{jVAqF#~laCzOOZlaao> z1teXX#e1@T8GvCyU&%iT35U-HNwnsHVr68@gKB8MQB0)1DW@NK+Lxs`H5)t`mu}-` zi#OyqlnPjDXwPmkHIvHMwWjNUX~zW##OB0m$5yF!m)WGfRblUFXAkV0<0L^_k`5T0 zn!NW=CQZ%aKJ?}az<=*p%epKvI6R#zd18Z0pVKl-uUaps^_FYMn&~fpo0|7k?l#`x z67MUBwd+-$2BD6W5D&%Wdof_(afR?@L@i>=s)R$7`{| z`jD;oQz2Q}1jZA?B2`LptY?r9RWoxOsd{YS*uZm=fw87y6N`m6APOe=*6_PbL*5i3!GRp!i2k!#kyL+d1uD+wdPIEeYz4I4iK9l z`QRfdZH$S*Tz%zN8-d$Se>_QRRMAZb4w7!KMHkbWF}*YS4kI+tkACr|CegBO-Zf>r zb_jvd-vzamsbw7APEPBaYK|6xi3u*(HPnBbS9=XJ8q9}?76G9o{pvQZmZ- z0)U{_H6?itMgzd1wAhSTLPa&QNijMMxeF8M4!>nhaeugEPV(zZYiD#! z|KJSFmWz7&T{5h4z^e?IwpDA%Sq-A*Z}CS1f5yW4f7R!Q|5Ts2K^~Ux-5f` z`}ZLLG39>&<3I?y6vZdsIWxs%RHMh`}*6o}O%v4bZ4bh;4v_WoLG*}Dv6K>Jgh zR=Gy{Ed~=wr*`3ZxdN{A2l6G;xf=Zz+Z$hj>ea*>M{GP%-C96%pCePPApCr--Hp(IZ{+hOqW>f=>x>PpMff z#zt!uL-+l@jCU{df?B;C>ZNFEk|uJ@#a5i1UR{s1nM-UHqTk|jX1BOj8=ik*RQ2-F z4q!T+PvbNJh=oNvwae%SdVJnY9STIgN}g+T8v9~(J2*oCYP|dTvZ?29ug}KT7G}Wb zna{y~xd-U?$^pFp)$90#`Cq*b4K}Ro>wJkP4Ix^;!DBBH=R}ty3-#9?BN;kqK1;LS zXOPO`@ZUR3>V~4;db2j4Zw5am3Rog`q~d<)tm;gnOz6d8lXn)!jf2w?0O1domLcjZ znA?7DF)9(TM8a&4Hd9wcaTc4r#ygukY_ws=N4cun<(FaW_Gn8L)Gs@akihOfq$YaLQ$;Gxw+*F98ggNM?OwuIp$9K# zbc7qWK@fOR(5FQL#5fpPyy8x{Vnu7@xUIw}r5Sm0U|G)Vm0jB3`6^1{9CAcWjGnQc zZLPlRXI1en>fp4Y3`3bb|Ci!bvKambi<bUx%8H?WunV+v8|O7*b*ZmS++-;8a+1 zzctTS$uCU9$DeZpl$?~N6CE4d<+gKGtm?vR3i-1LjBC;l9y|6Vt*<9J8?h+8vi0wn zY-Y#%8OUM*12;XHt?b`A=SN@r2C0i)PCD^%&m+9HzkPiMtZvFN71{a(AYVPY)i zr&9!bICEUXyvY}`9MLY`Q{UJ(R!U6PR{A`%lWctL5$MaID2kb*qCkEPT z3fiPCEh*|#<|8*OR;ZBpGygJ1|d9v55`b3-m>=sL!oLW`Qd#;%G~SOPauEiK)Vkn*HGpGy z5l%K;h#@#WKoD>djDEJ6WPCk9RQ$`AmfymRv~rNN`yz@vp$N|oIz+K?5iK-bMBqC< zME!UXBZ*fGl8u25(@|W)pen^gS>wY@BA2o13B|+}&=FR{%Q*e%V$zQB5l-n8O;fxQ zva!m6GurtCYn2kr`O48Z-SZ*N?7I>`;D7E>xdY(;Sy(WEWLOMEGLiV`^2FRJ=5o=% zSC&EJm*0CMo<(q;eBq^1X>^#;#w?+X%HcFxKC+73)~YdYULL3%H(EJ=Gh)w1VNPTW?N%Pb@I_q`n)R@5%Zj4%5Xk|Dm3yK zF~~+?dgjezTwAW-`;*T}cNkw(Hjj2H4Vh2-x;LA38mnGpn&1YwI}Bl*k~HW!u}F6Z zmvrcE-L+X6A#D0;akH8MWwA5|NMl7Vh6Q+EFzK7M7a}%ONxow-c~tC9<37DH9NWTG z0FQnEk$LDuHGK%Fb;FXnEd5Kt2=^@J^%rPF3y_T5Tq^j6=y1z$cp=xf~%Yo&kf zucZDUEIK;+YK4ttCbi5>-7EZdQg3Bxj8#jqf{~<9?Ph;^(EJ>1(dk9zdSQzEqlZzf z%I)o%)K3Q4q+BqDXGumVw3*t$P#~^2jsEBdwQVxD6-5F(UAo<)9?EzHPApn6PIHPJQCDPwM{3!rp!V zC_igW-77-nT;HX0DtY!Dr}mR1gB1tSMyYS1(q1zKRV%c4sv`f_{Dwx%#J@+5a)W{@ z`5B0`<(uxPajW7icu&{$`IDk5^zD4!xY(b+c1{*M$HTz_ZRjU#*q5xB2OPZdi^a?C&`M>(~ZZ_u;oVrSUwY(&Gi{#83v182G&UsAm1I_ADGa~3FUN4275FZ&6GZ_GQh zT8=x`aeYT0;V&`bb<*XN_Mt5o(a#iP>zS{oxY<9C&o@}=N57GERA3$|`yJGL>wi_% zZk)gUJL2=R=(Jeh&BaO4jNVN7KxP-|H6&MvFns3A&tb_`V&Lo3r(k% zhy7DnETfU?heB8A{}z^A6~o;jCi$Znu_apKB0u;yDR|7C5FN3<&6(&|j7Gn5=1IhHaGB~iy-$EXs1TujLGu~5PdBAcmpfC-{#|2BZRtLqs$Dat z0nL()l6*v?xrzQn)J$VdUBM`Io;LWpjOb-JF$H0rO=?e$Br(ghg-00lt$zd>~w zLCevUG|}SwOpR#^c%_2a_HWu5+2Us$x{TgR8=pEhODW5VN&4oN5HxINBH`2q9LC5b z*K|Ab9c-k;ht%ePaC*H?+aKh%kDh6-vk#;mgsTGNnTkY9ZG|vX(cJnx6JDX0vC&wn z#iHZoSx;9&$#Z6-78W&*4VNS*=cXdc|M}}}K2Kt_*vQ{6g-j&7Gc(| z79JSpEQM3OHX_|x@$9v;^W{*tlQR{pgkUq|(@!5zBFa-CE`#t)4HZlhGR;5jzG{JN z+9OXI+iTJ3WlA=tS4Xd44|qSBV*boVu3U8TmlnNGnfkgvpjXn#BMI4#Ezu1JPf|jgt^e#Nj1OZ zxQ9R>1aSr03A-eJWQJI`FQTIkGP>Pn@{D#;bvv)=XKbU1M|e{ny@E1mEXG|=HK$Bk zG3!OhtJGfXZXF2GRe0OX@VaOvS#6Lq5ZI9obFde{R}XXPf4$R3#cqCF$=j=rEJfIXtM^*wK!?GC<2f+ecmqOv_Rz{~5b?c)~-qMB9)O%KXq(KvgqSERa? z{?o}{+;j>CIf?&_OvnZD=Icg)W-FET?QLH>U;%~qsICM|SF~V?L3CUe6q=+h6A=BT zQ`45+8sRRla+<%k_cxr7R3}zdlxUC^3#I_hOL*?u4heOl(42sYPr-(jZ3O?LSclq2 zKeRytRe-c4jX!JSsONm`*0bn`Ar)uHyD$_E;ugo8`SySDXlP}0>S zg>NH%l94|!hM;nd7%ED%K;Vg)r$lRbr#QhbQZJ021xX?k5fg!fUzbuBB2#@rYl(}k z!4*t^z+45xR@2RS-Ab23vwSep|!*m6_W4G^N5zb#ED-~TeRD3hx5P0EPx0f`Ne_G18he#bHn)mep&?;=q5j92nUN9*|JBI~a#sy&c!4$h0Gv5AHY!Hg+f5;j!e(dM!)(Uf05!6`kl7>=>JA zFGa*xV-ckfv9WLmC;xb@=}xi66Lkx>*R6b2s}Y)W8^M1dC`(FXy!hh1KKX?9~BY+*ftxZtlr@;T$Y#T_u^;_z9M$X%Zy|Y zTCzV_()o z79u7TIH{lSg0CO5fBM`AJ?MH-=xO~P=Q!ZaSJ_mF{=k_r@p z`yIY|vdxr}Q{%^W-Pg=?jeV){Cg35HHBUe5HEp^Uc51g&R!}xv$-$Q|#`pXU0*aj_ zx*Fz@O%Yy&02XDuoyaI|&q`hAvYZH%%g$CVB~o!fus@`yt|PRo1O$(>>E=PKoz_)A zM^gXn>UF}qHe%rzuVbR@)QOzvip&_nTwX%9JO}bB?7Y~$oW`yU((IH`|4bmaJCHAr z%_@1O3(laEzm}c9(Ul)pmHSngtn?v&7yH=gYDFIW-zf`d|E&({{?Bze2mmk;hVg<_ zyHlMO6p2T#6ba$B7>XdpX2L6UFa9@|Km3VV6Ra4;EVXTQlx#B*#$bg5XGt@k%o6h4 zO?XIVRz>hdJb#K^pr22XPJ7g5y}NF1W3>9jR_}n$6TC--thIt6p!h z`>XS}Q}bVS@GUdwk>6US!v5++?6hfH@-q+x7sRe#Z14#QVyln4ciU=&AL{mYcQmi( znWI~8zt*VxF^gQ9OjLPfJX&-jzfJ8lv~Y?gvMm-j>29~~EwGbFTH~tz`Ke3c9WjBP z^rzJr3v?}ig8 z$`O<2e58|M1T#Ap&`>^qNGptE=|21_Ynuq!O<>Hd)YYZ!wTeQ)ynxXX;P+UkNRJg(W#&Y2+h53%G*57jzMzg^QFeZv3z$@apiew8V z?o5YDDo2{!!h-!cS+>j#%QziPJuAbH+p!r6Y$nJZ$&Vq?{*pxPUX=``4z30!rXS=c zO4A3`SWJTEvae4wq65652r*x4puDs;4LwG|>Eu%{Ky_g)uo7qV$+k#IVoQl*fBa$J zjYq&BSClgNIQ3GD+AR)Xy+Bz`)Pc1`)FFrB#L)0VZHXaoP+Hi=0&PUMzAMcz-U3xl zJFpIIW4pbknxCMPrK(~j?dN!vA^Gvx8~mlMmu+w6CHvYe#~VJ&V@HvGs>-%*ZkWD( zvS^Zgsq8#~`608kbJ~%)|CQYJ8UW?JI?g!?0U3|fs0}CEvHPU>Q<3fI=mVXU^{3OD zn9pCY(34e#B+F~bC@%ZQ0E>24hW+&q+|U~a=|v{zDjM={w=r+f28U=&CK1Amle(a z#X7LpsN6;B2Sb-fehoA4e9Go}@Z$*n_QI3(I==a{i0^sJPx})^#T+z}5{DTcAw}vd zkHMZWc90#6C^m@qE~Qy&PHMRX{HS>=kgont0QFK%mW-4voWwZ-vSe1n#0+@aiByw2 zMHoQ%hCWRP9IulM7q~LqZJEMEg$Q5`uy86Nv>6h&Mk-;a2g|sEuya1_YSagW(OVj zpJJ}6ES5Sg4ZvDRa)(Dyfzes?`AUr)|Wed=wLvh>Nz*4&IRqLaVzC*WEwy~`* zV9T)8Cx3Wt6VCWX34KE1)#PM!u?IU-KI~0s=(4pn36@b*1weKj6xDsxfY;_=C};U1 zq5Qfr{4Gw#hQ@NYElEu*;6!pa07S}hmTCPrS7KU3Kc%)!djpT%UV@jPM*OxlbS_#p z`eZd+q&4Gt+{#WyQ_Tvg!DoSMNJeXN^$G<()T)~i$zSbgXbzyG`2F=%s{c`2VMo`V zGvDP#c4|epc#?ik5{>&zB&m>rt_G%F7f_44MuN`Y{xkKbEKZ=lNw;8H>`dWLwGo=Oj-Re*< zW^fv9B6fU-^Sw|%RP????KsIRC4F0@}ZQH@Fb#>E4JEs3+MFl}_~dZxeV zBXZl&W0sA+XVa}Ai%{zoM2W&+K5N-j`v4DLqd~oG^UhD0+uUiZf3`P8^bpfUF=j+N z-(Q}35kI-RM=b`<^9i1Gbtn8R)KS>_UEIuYHW$q=FLvT`(GsYLapo_PIjY(Ln#@l4 zZ7bo$9>uCvq$vQx)-c>NHuXXL1-WS;z8=gy3k-md`4V1GNJksrSK{OI2bk2sH!MYJ zn}RjNK&=Iv{(th*IBC|d{g0HHVep*tTM4aRD=GTB>s;5U*v`Gc$YW7!r zqA2*It{oF!TQSt-{v&3F*-lV zKKB{>)_rnh$iZjvXE)&I%OB1O!qa9+ve}zK2ZC8HgAUOb1%Hg0Z9wCYHSRcx)bN+U z@3JiEUt&v`d-0%$XiFbWM=M`ipJ=`UylJAo4wEzl$l65#S{t!ZB)iW}1itmSy(WC! z@&WulS*4d`*QayS1nm70MmqX@wNB9Xt6i9qpl0b1mZ1OD&~O_C^P?{1&he#V?Av{s z+55b7%Yd(YYJZp5?or~bI|K=XelT`d&rDQagV<(T(beBkb_ai8P~1D3?d&m~kq=i* zCE&Est^m-T-x+pi&&naC`(*_@OZ$TcY=5cj-rayPUy{D{HIeQSrP=u`+*^azIX@88 z%~!jkf;53Tpq?zsy z1LlZqAr`g7!UbuCn?ScQk>HR+?oy@NdD{y zW9$sM6SU7iRI6fQ{yrM;YKaLAqRfVH`<)WnRjOC8Cy{;VD1gOVaH~!@A7E_3EYacj zs2Bbs-GxNpjciy*^J1`qedyetp)Q%&6iLL5l*H0)#M_wA_Z}`9yi|BFU&GF*p)po6 z1^{s-L<*suyU65V9}X=RpoW4PQUV*_gx^&}LJ@ofrQzCsAUw+`&SnTsQP3rC#K--p zNTd*R=9>>McS6OEA{;xT1xduuX-U)H1l#+8PAar3QKDMYq)U%ts_bHGcEL^aSZJ1ed+q5b zE9oWPBa%5SUL1`it1l%39+H8=a^eh0I>c~}Ncb}XUWc%{4OPjp(SWrYXlOd*wn=pj za+G>Oqe~TEv)EEsDaJE6i{BZH`lJ%ZNCQ0IL)KGsO6fw!O!?pFhU}@$RzwIP*Ue&E z)8d!X-a1%u2W9;I^ln=W-YsM*@mU^va7y@A+hX7fZZSAZz7oz+(a zpXR6;);(o}-L1)85QNHk1M=)rP$Eowl(yIpdTQ#~p%yUv59nk?41jYGeh>@(Jhw52 zp2NW#LLN=y&Y2RBTpMQCD&)+*x}$7r;@78$>^ zUd*c-KiW8^cyVUF$&!4j9^(+=d~0X(U>9DN5fX;CVPpC7)+TR&2$F66-hj+H)J=5V zsm35}c9rrh6z`>vVSgyrCfFf=2VyB_^-el1JRpaN`*18jq6T_qhe_>AHHc z@HZSg#>N(g-4K3TM}&f~8FBCi0MjGc;B zwcid6IK%2ssx2ObG7uI-z2p@9|AVu);EMC>wmpl&tti~xAtAUED69&naCd?yxCPh3 z3wL*Smq2g{f#8UuEAUnJ4_HRRlw`u&pty99I9dD?!r!YE?;gsX7TX~E z8i4iemMD?{`}Ph?CwJ-hF);L3mnMWR7d_nw z*Zd%~TLG=3WZE@SzN5&PDrTAmx3$Bm7v}Q5N1mkHL#4aPtFy_NI%m4`>uINKR#+!> z&%SOKs%_5)ZqgXz7xgOTlSX~=f@JPb!GEb-cs#pj;^;TmJEIPie2x2CRdV;lJ2rDv z-^#n?(PuE5-wdI(@5>m zpLSLHqon4@kAEjVPPcu$3;AfV)4{fBlOEYDO5;J1HzefT&s-%ZEy*cWr|4+d|L5i9bZLT|3UMigp6E(7*#%R}n7MLcXCppGb-M$+#M{CtO5KnW14`YO4 zU~qO>9fU`obWlB{R+)8KW48R_UOiAwhX1Ldn^=7JrU~zU>Mh&!?OV!&Zwi-^kgfMi_!DU88?1U}@Wg)Kber3GZ$0hgH8yYrRB&zt?F|2z z*~g3XxCmf#10_+_S;Jt0sP_L1d5^xiyTYwK0F_4-qKY94SNm$8EZsN6J|xwNMPW|t zse5=}C@tnwnxz5S&jD!l+~*RV+5R#{HCbOHQ+^nI(1^z-^h<@uZyi!7b{Z5{q7XfJ zOp{)z$)H2lIb#=dyC#1bq>3w&lmq-{Kby9(rv8qQEL)8?CRTX($tNPX$|3nNOWYpAE|)rT?#oksZ?vXRA)6)kt9%IyWV z93@yNzey2eW+U2(*dW!bKFXC5*AI^%)gP1OxO3{tbPHGpR4I0J1i+GT+ zq_^>v!n`(nfC!W~JT8CK@O)GlZ#}n+&OKULI&a@FtLmhD;?aC{d!tQo_a}`f#;_ z{~Tu)PrD$v@^@gr{V_k_+-bP3%5MFj6G_9GatdWd-Rf+xLOxi2rVpPs!)YJy->3Y} zyFgt$Ei)mbsZZ&8ixg2Rz40(%BKpE=Nh@8%}Cryk;y0g#@C43jC2hO<$l8#j-j+p zLdv3W;HSzMu|W^WhE1^Igt($dSWOebsBac(j_6n%$W@VF;wmZc}9B zIUOXgs+A(eEsDXjOxBZPQYn$=tsD&*+dYCl#`B%0lKd{C+KPPkTuX+Q&H+B_$?I6x zW-oKR(vhnBvHHy`PL=)Q?Yd0K(3s@xSU#-Jv23al@<+Z)GZ}P*2qky#AINg+Vi`2w z{xGCgaQ1W6DdjWtYJ!NfeXx{%k8*4yK+z-ReUa>VL^k4~iCrSp zexHH8JKy(8a7yxkEmy2NE>emA2%8gDsIS?#bx}rvAE%@r`^7LvBb0|V<|_<>K|kA8 zQC3tfXTsIFmO9w+3F05m48R; z5?Tc%WKf0#w%nW}RFQSVXryLEVBXe9KX4Uoa-_V{v##5~h+R31Q;P2VNvOdcy-qq{ zK53B5yD6I0+2XAGQ@|u?v{(EK8+LBPK0OxKZJkxf#2n4k?p%Ee`h|@G^+>;BWBuR3 z0>{$sb(W3}y0_`E4@D7UOV4CK+B@o_Up{nqj_WPAen>crQ<7nGeZ|oMb)7Z@RZ7J8cliZ^}F??EX5?;J3ZNY&z$&!>( zjB>3%P3(1~q)$c0{y8U{T2~avLKr6Q*_^K6$d6~Z9};DJ`l5PU!D#n zl8gMB;Z;}Ef+`JGl88P>cH5R@CTGbqE+8i+KMC+vW*}KnfrR$2e-^F{?SzfoPmGjB zXOOzL)~|(+Z-w9B1WKNefmpE6hIq2AU+htM^h|~zqRc!t_lRq3sA6jU&-mc zSY^U2vdfEiL&{MVMl}mondorS_bk=M^B0V#IB(CMh*`4dzfGOy82^yumUW80YuI9-$3$y+k>oDoxBU!5 z1BVd0Qn2S)qCb;_G)mDp^StUwy5sGrJ?L{eN8xc?N4Sv`)W;b2Qd7VwFuiX6TG`odzzEaj={+o83ZU%*he^`z*N z(BpRu=M&w+W6cRJu)RqbSR4|NnH|;m=aOjU@QNxAMYua$2i9L{6AXhry&HLIz*(em;d?D`J_d@{V z7|m=rZt(MZy1_(8-Dtg=-oRDxi=Zx$nG*=~w2-QjxyHE+wFyS38izuP7# zSRufN;J-(D19O4Gmey!!#y#l97{_{&Y@s00P{M~`9D8ZiWa2w;2pxw$EmoKZX&604 zh!gE&P-?q@g)YOx56C|sdkJN2Ga z76z7u=>nT0RCU3sh;S{@$fge0uITrQ+TfreLcLJNt)0lLYaIx}v z#auLT9$lHP2;n3T&G++>S#7~MkN|R&*QmwOA6sBL2J(TT2p}?amp_`=mS)Bp(K_OT z_AF?dbLjoFNIWJNb+Q;e8E&fhpb{nfLd$?cBQ)~T$J*XE3fod$OHd}|LCTg4|I2pl zd=YjS2YA*_rI*C)c~89(5o=E>nsa5BFz$Ob=JkmQ(7;RRuZZY{X$l^b#v2gbg8gpi z-b&wltY6umk!T8IC{lXLpN#nTq6tz9CG!7JJh&xIU?sX4Br#&J7Pzy~jKW$E(T}u5 zjCzX}FxL0XxiQ{gRma|kOMl{!KPN-lMd4{)QGI@pimFry zTIC;JLgEUFV`5^}4LX0v!b1#Njg31sj7vr9IF@%I4`7MWKOj6yHj{%O+6RKrXT6j& zgIJqrXjkRmAr1qCy;!ok=one)4LUuTmlzpP9MNwGp3hsuL z=@`kx;|MTV@jWFrLSll9RE%hz15m(b{b8HHMa){WC=PWf;hb4?s?wXiBc6{_j@R&Y z9+2!rlxPENt$~3RO=RzH7bWE12^;-Zp{wGKgj7K^MH%e&pGYSg`lM zOm`_1gblKii?HVzu;+{L1>X?d6v?6`XEdh8pOq%tDaa*vsn|qeC19e3TxGwc%pyVY z9p=oYNzXfEjQk@5I^WKl?e+?%h>pzgr68BS^-SksRj$xB=!}r96p15!9#RM1$g+#2 z$#aH((IE`ikFo|ruDT)A0#Ov61+2{yQ;2BPrKsO)FZr|FIsCwli>aD^c^Q1X-mpUS z81=2md^cGBXtzpDdrB>ZsQec-k*t@E^Qphfifhvl@^0`?0%Q)|G7PdwUxf{DQQsTy zN2>4g=PhuG3zPyGONGS2fsx^Ot)-NL{C_`cy+{`@8%hX;5q4{riTIYy?+G2E3Ju?u zi5G)AT!r+s%i;P$`pQBxx+OT42*A3D(!3T$c{%n_Xryn&8^;Rtc?=v+UKkE*oGt>J zBE+Mn#EsR!r=}$Hj?mLr>0KEs9Mh}pGlf6HeEe7P5L-)?Xiu{OhtW5;dn6erGhp#y0tL8 zuUXDP>ikNZylLEflpl&9NxGGOfHr;;r_5NCX&q`Z=WH6uXsFA4Cw!(sWj`(HWnGA^(2{E67+OL2nnb-0k#fJ*gb<$UG#nx$HUR>!TW_S$~{ zcQ}?-rK4`G)qbJIbh@K{qtkV7D*qqA-RkxIdAc>-+4${42r4d}YFE?FFoKxdY^JMu zZyd?2QlQ%1axk4G;PiE-yY;9}Q96!Jt>@k1mn!XAvrj$kXDdzC(*aBP4Tj^=tM_$X?@gp~o6q)j|2RVh^O@3}o&ZFTxKD|LSIeS0_w>*D1=AEe^H zhlip5L;wGMLLK#g49}r|r51%VjUv#VX%`XyPlq#|$!k5OYS?U{hNMB?&zo#L8h-x) z=dTtALlv|X&*#zPl&Cr+MiZTbg^Is&tQNJI%XX^WX3qGYj)LO-woJITy7XV;f9QX? z!Bu~t)oNR2t3{rw?bKekQoYk<59c-Zo$T5$g0k-;LhyLq+?lbLR@HEy`xOFTQf4M% z$ShtTCC<#a+jFFd9LHuL&LB)GsT7)9u;sc-ikU6K{2DnzcD@?ofNSnMIuXiDn)PB} z@G%7FgH15NlOx1`(vhM>aM^RqFvspqkoQc~L?LO#+KHy_LgC+XesKl==Avoi=Yc&3 zK_vhMKlyB_32GL+wJolsdmj{YM;;9HTA??A1i~Xs@xfP70CEbS5L87~B)xTsExCan z%WIC+SN8#MePYlHZ(~D6N#1-N0teEpFg6f$kz78{?KAf!;FnDFC8Dh`TnuzarDG1B zlQGL7-Ino3gGXwbU+^JpQcMZc3;yqx@AZxYW();o=0`ZGt`3L80eP--h3%INPc82S z$l0f3O&@nQU8@4OHKlwdB&7{U@#}i_1><q)(oZ(x)Fc-bzz?{B9Vh z3a)E-MPO&s*pE-JSU1X$u<=TYahUt(puh@VlMMaqzKq%Ds~G`T0%F~%^cT`L8lV2K z)VEEj-UD0uP|>e4XB`XrT4#Yu@Ea%(HJ-qb1vTE?%pT6Rvg|VqFY7b7(X>OZ^{uJX zxaYI6&lj&$XOezB`sX#*N_XZ(p-%vEo~BPId+~QhcQ+S$bHH@ME+~{Z=xTC8 zINw{eS7IsU2Mv#V*>r+cY1SlbSDPtp>?r;9PCDUo^QWv3N<*WRPx$%!%hLhVEoof| zp0#5-{Gh87%i5q;0DT9k#Bm98()|3uMql@gMye$EThTY}kmY5O{T~NW5=s6;!1llT z=L^QBPqIDapClH1_nsvdI0?)py%}3TDVn6B2j!Qp)++hwRA{|dYKj0Gj>7pLcC9lG zNv>bH5vZZeLvi(ue#cJ?@RIwU_K{1QpK`FxIv;7A@1G;R&gfAS$Yllz=d3Vn=pH93 zWWr(Z!kr1RIG@E=!RKD*VPrH0hk3$amH-fffQ6nA-+kbaQ8NOI7*_XJtCvWeC$cc7 zq&=ikn*Z)|j2^d{cVaY>`t8eTOiwxj@`rBXFm(n(l2T6&SQl%?wKXLT9Z6}`c$f}h z9HWkTImL?1sYA4e{|&S|Pf<}>_>;Bj`Q{=_xFPD#IZUpaUVbx+IVjyJIzxqZjjg4r zpqt;m!8z{*yJD~Y*TmQ@N6i6GY)%`m6q~{GGF~yxNh=}&WWh8M7}aosYu;XJE}%lN zv>L1jiZx!nAxp^NOtp1eU`-(&a0fQqJ;H}x-Ahxid_Abxm~vth#AzNfVu4 z2Zf})%JVl1Gs?nf$L|;My=B^(XK>uo6i=4sF!H)KM`SIIdjgQf%z%=m21ZKChAUYM zaXejsvT;}KIdUp3uH%;C44Qi3JDQ%wB4av)5M;IE9C~~>dL()k3}TL}=l*;~{JMf_ zw1izux=B-xs1$<$`8NW+_%-{wQpI0t;pSChRX8JMA&{3A#tJ;4_RNG?=!*O{xLESf z&em~yM4a&Utchv9h>3(cpcL>GJEj){YF-E7K*B0b*p09u%e3LD_VU){7njIqMt zZ5*hhjqxM^*q+n-N^7T9O5F@m{wx`Y7ywB$bbh@$b6|t zrVKEDXRSUE={4hH@J{f7+&#$&l@t&jkgUN{vIxAv+iF*It(84sILR~MQrN0?O`i9> z0cg>n${sJq+LKkw+S1@3IQnYM%A2MXAp6-O&d^3?l|e$c@R35LwoYlZStefReK~i? zqm}mn16}hQq1=f3W!-m}*RLo|yNpc}Ps;$R7aE2KNk%ua1N&FwYygxEPUg^O^I1u( zxPvP3-j71X8Z`S+=$cDHP8p=nt0SWEPx}NLw^M{{6CUJ$8c6638Tdgw37Gu5Hj>UX zPS;nWSLh4Z{3Crb_<+pn3o(;Nt#DBv21F@k@>1-Mq@3F-O~V(Cj^+zB`*6|}`D?@Z zX{T(%r6j6_-E>%!U3ZXlv>40;yF;+fr5JvU_nfhlJ#kso6c z36;i@{S3_ggQm$%Q>(Z)O2|z5uEg(JjI~a`Et1Gdir;dXnuc0FtuGG?EoK!PeEh>; zD_h02(8ye-qcv1`Sm531c`avZPoJ-kdY)VyMP7C=)w7RsCqyA|YTwu4771 z=jPl1HbBY07FkIR|7oQ28NIzGrWPtFd2YEe8^rLujI}))^?j)H$)Wib_j4vyza);d z@o&ER=p>hOQ~CJaYo&b2s&7I8VF8j#%Ja3`z2Ce(UuMuwFqm(l<8>Se){y(cZmjC` zMPi`tDYBMl)?h~{cz@e+i!>y=96zzc8%z=K%7Ygb9+32dXr|M4{5}9TROfaOG>(oI z0!9aFhW=p1f4Ss!)aHK`#=|G)8;tg9=ZKq*!ZAUS1jMAz&tcq1XhnvFwZ833ZNQqG z`TFMy!7V#y7N(0*n^eM~PsWP5_h=X^8-D2(!9`QV{5GyqE4cE|KYmOKvEuFdC3FCt zCsQ^whZOH8DYlJow8nD82li;NK_rkN?2T^N6FYx36g+$6d+QpJ^IVSo9$_LQ6Fv@Q z+NccqmKlTkGaODDeQ*zfEjbmE`|n-ZINuTMBc1Oe@f!I}HL}1z*rUU+AY(rIh%PZo z51-ZyZag>VFBSwp_bd=dyvr0NtIUwD2dgH`Xg{I|Lme+5@4wojNW4Zp<2Wc(2^GP_ z;xjVN@weUBxFYW64jdHaJRyUtjjn}S8vb=9d|)_c0L^E1pQB_1v>%c*wrt5OWE$iTY zvdc%LlR2!)iDQ6-@_Zzi-UNe87|v)*Xjom}Ts#+t{2+ zu;cgDhAG73CoG%zT^H_4|gCsNMI?>8Q=El4O# zWXN9X@k|ksyXL%F{_3v82#uXo!Z#%tEjpw{Xr-OhGfH{Hfk%Q>k%}EKtDj~#;h#(X z3cphG(~Umy&g%%&bXj4~*JV)DXe|_dizGN}@3TY7V>l0PnU6bho~f;wB#8^UN1k?C zUV(>vmO&wS%@qvIf5V(|F`0nVrbUd9{7UI;8Kt%RE0I^ephq`%wxj@^FOQV0xY*C_ zhmy|BWK_LRA?7Ij(INj&O#XDM7T$VMV5yP%4#8e_5!X7wAt%Y^e$mxf0oYOeYBIY_ zxRCy5-tQwT=2o}hTPnJoWZJ6yRMJ9p|1w}E8E1{t8?+2XWyjx?#a0W20s_S(nR%Yw zd`dP-pviD4N1JaXdFisj-%kqVa|kyLOW4cHu%^lm)?V?5=TB0WlgpRK$`_Yql#}Pg z%PDz7vn!-?{B<}4hS^j90Enp4InK!F_Ps_;jl z{-$ay?NVC7N-e^i!4cjn$C_$Kf|?xOnjFX%Lm0#+JHwFE$*c9tfv=^-Jx)cUw{Z;jN_q&RduNwk!O952MqWRDXv4XBZ8~p;d0HnlFW=a+ywD z(=SwNmj3Tx^x{OPNHyW2qJxw-(_$;uC-U&+!X&pCTD0fQac!QT5~e2z|Yyq!g!Lz67b)llPY1Hi^F9kcx>_ z=623VOtw^BoF-GLb+r}KY~Njj{KWnF(Vfh3l*gp^Glc#ozCY7rDstjMCj==NpGM2| zf#MFF9_VYSwCUd>IT$kt*D0(Dg9c7Yqj$=?k88H_VUE=m?$21I(zb8)&+>J5Cn^Dw zRur{yKy2g{l*kIY{Zu`kGhihKZjB zzgXX3?0VEQarSP&(ZuN?;ublpUkHt_trz>gYCTSe%ivt|WE8+whW^KRxn)*Zi6Ko( zInKHr-?&zwf73m(#e-x|oTs)XXZm`sYiqTk{klb^aB=`tPuB%AH*x-eIjKrE^l4o} zGwtsp`(CA;z;8jHwIC8t_=H2&rhn8jeqq*5LT)7@mg-rJeg=4QNdJ!a8ADV@h&#(S zZvd;PG|~9A*Xcz@^8Dknq_>Y7jaq3&r*z4TU{0H|;~qd$ zs@2nZ(}%>T@3|~REf>HpP}WtO2UEy3&`mb>dZN`#v#}3^CUt`be-%PP-r&6h#iD7G zf_#IPMFRFD-oaz{V^##E%g3Z) z|B%m2W)t}Fr0^IAfn2&e4`)hSPl!OIV}W>?hYce+3>ahK@m8dJZZKkV(J~7D10!92 z2BN1>3aZ#6_!j^M0~&U`Y>TyuHw`RMc&Sy<`}DO7pQVT$7z!kHmM;p@;q*oYThH)3?B03LB~UdQNAwhF4YR0UW$x`@aw<@N*EZ6 zY4)ng8W};uA}N85u)n&>N%XMExS@Pj(#`l0c|XG9_8S6nAmgXpV~7?!KUr zi3sP54&<;~U=dYme}KmmN zBtJasQx5}3R~qOXH;f;FP>wN2K{ZldY)%txJ@|q+S~csBoESNj!-^trp=ymJtas`K zc}4YY)4Yo?@a?HDdHbH8xztVJwmRwMU*UsC&B-aEz80yPK3f_IF{V1#>C!RhB)({S z6}#698B=|S*8S-Ff#wCU2TA@T(Tz0@Z7{3=Q_C*q=GQE4c0Rv_$`LaG5&KE z=>M7U^Zu_cbU03A%$&u1@xO$>t@{5<`2T;n&`eK!i<_Aqkk5f

    ZA=4j@`%95G9JXAu|C0ZS679E}3#4Oy6gzhcx&zu(RLS7{!K)XU5 z*X^{QBb5E(+D~hZ*(97g`Ns~OGIfY&BKet+I?a|paSi({9{E^JB}fu=B`HG)nV?Q* z0cbQxzQ)ixCP4AOZWv7uvSvV)aSx$;Kn)w1w2rWW4bVjX1?9SjU8;vTG^0Qe-1X&4 zk0?0G*B8W%>dKE=D2VA3{iUA`?2+AzOyjEshIE~Qr~l!5bYd85^sUzAuE+PHaQ8`S z_I4wb!8>eUcgYE|H^n(xP_$rSd2ABuf4*j>zr7~!Z}?CilB?G4Alx~RD?BQE^RNHU zzkaPn5wTwZl8I-5k8Aad9e5f zK$zU=$>|yy_u#|nbRW2zdH1sN{*v$eN!rUegdfWquVB(=+UFYaC>uh5&NoRXym?j{ zUK#9W6&$L*k^Ht|XHXSEwf`A#q?jri$MvpN&riKGI}2nL6-TbB`iy@^rnX{8zyF-bRpRy=^acg>M3dT&}38)iB3X1j?=hl8^<)*J-2Zx|()re)A(t;He z6yKg%@&qxb=8TjGQ)*ItI4{9SZMH2dvGtB@(-WI&FW@!C9+p0>HLxthuOsYp$X7eD zmA%H(@hxKNRw`78xgPR#zaAxhnie-4^L4cGum8|fV`p?d7X3VcnTjH~tfI(;87x+P z@=PPOTbNkT0c;UenK09z9~9q;;dcOMZ{hy3$@mSz-q*0N8W$%l8;VAsB*8SMTS{eX zfmS;Wdl?RgPfo_k>ZDW5x}oH2+fQ}2e>OVQX3kYQX^&JSh&4E`d)1hxx3I%Nv>0fk zF3@t12I!3fAv3rH5*Kr7zaBN@a`8W}N2Eo*v)R>eTM{0tR8`W9 zE9z(ul;jh8`b0d{{ zY?kmk6>*)(s4uJ%&JruxQHaHB;u_}d*Z_+4vND82!}GI)L5X@b-}^_GBinFd;h^Le zrRfZ8n&En-e+#XKWIU8m4~4~L1>CR_CJs$yAvKrt@bC4D z!#7K#6w5!q$%6Rx+@;s)F3ef14eJidP_xX#ZuRNB+TrYZG>~BwJ@Pw5I`r?f$IA6w zP#Iwq>-zBp_>K#=KN7tl8l7ogLvcqbdaV&pwF!)4On08@11*tpB3CG9$J#Z8%xvxY zZGe$1X{v8;B5&r5N+546TV;tI|Et`l9 zZfs}wJ$N~!jueLsdd}*pQG={$b}7YLtZQ}Le}rGW!_mV!Z^k?uz_j~{8&P9PUJ*GA z-Na!dlSokJO`LyCoDZ+%Z+#HDhYIj-%Z{>G4A1=Jf4l6i9>s z$t~Cp1$3Cgc=-7NV!XAj|9yNKiLVq@%I0Fs;%zD>x07>{deEr628IVc)yIXeR$!jD z_h+-{R2py! zu8*^YcV-~r9IA&Z{uMFy6^Q!sblkmbln*_bjdl%Jvy@K|%L5{DEf>GuOIkxZNLDs< zG;-qnZvWcb<3jLB1Qtsq_lu~@_J-A!h)BZ6R&(A0>5m9vRYcT!N+6_!P^*3IJ&S!H zIhzy5he|Nnae@r>?}u-16eYm#%xo&LspN~nkG|PT5_cEIcY6KAZ>q^P@oIo6zsVOA zfoXyruzNJ5Ov4y(J0QW{)x)9be&{`&)1O!ri~FbNdvA5Zs8vkARmH$!LFYxqAUsIG z?=Mg1eIi-|2V_OwE4MrdT?@~Iu66+dr*Xr3vn~ zXJqv&8O6jgzuC&-=*izp#@!#)gwUlVcW*Q@D&HECC79i-ihRm+&`s@T_0ao7={8Vr zS#SD}H1)aM%t6&Rw2A)jWdHM57y2$&uj>(tKbPBCh*H84_UB!TBe?{JM2@*iMmq|^ z*#n8;DOv(AQJzG1tYSAL^(%n7Ig*+iKHbpYv3eH3ygD2n@QO)wuAf{o5Q;;I`%J94 zCmTn|Ay3hDxGSGTDK2u$G*ARjLGCwTz8V+^rNynGQIP)D9FpZUOR5*#Sj6= zSN`QGcFZ(tpy987HqwJ5Y^W*^rBB!eV$YyLb4};vQ?|sA*_5n39$La+Y?%%N@H~ z!w7E1K7kCtY{*owy^O%`a}LeEihJl7Os-Zpw8<%h=QU+<6b`UTbW9GFxUXKv#Oivr zZ~!kzC_JXVi>#z0X(N=kCGNXrD~;Up~vrq&g#Tz!t~ksS2TC|&Iy&#h2*-M8h7I0dZ$1_eg*2o-<>ka zgpK7EvCiBJQJ4Oev6wYAWwBVz9w}2vz1T#ZHfHCFCFO&3`fz@o? zl?zM@Fw7bQ@4H9xe-?dFj{mcw0jS;pZn=I##tS-$D&Tmct6j)abQ?h>PIsNSW({(i z!svf~n|nVq;hHdjvVA>`jPh_=CzBoUdqUq%+z!Jix=$`m5EJHuWZU^7B=Iuc=lhde zR@`$s9MSi<=87@+r~Na2NhUf}kSgHhFzHnxuKI7ZLtNXeYuRV(jqZT!%glRG{c%99ak zy-VlL^G<2hJuzRUdZepR_Wz%1o%dY%|4^+@wb)f!Sq`@d9cYCZPuR-3v8koDIs zfl$C~flNH(Q&ZbXNWWUXS_4`8>AZ^ZRFP^v&v|jK<5^=c`N`$a1{Bm|@moH}<*~#m zHI4y+>%xKZY$Ei(!n0cozQ#;>ltOnGa@p|+*TOAfzk=g3qY6=C287}dE-6J4*`jL)*Kv^S z1yd~ym)bMr`B}#C6bhM2%bz^#yeFB_dv7n;vqBTk!Z2SD&oeqsr$FAbwiPKuM3AAl zu)bdi7$UTum6YEeqD>^`_S?#$J3Tf^<0Q5Jks*D_R4$)S$&w=Dq~%|cVi(3ztzDqf zHX-SoMmxZaqUr%IMny83W41nchguu}tvD+4BkGbT9AYDoHRC9@aq7KY*^gV4cm*<*7Ahd2a-46jgyC4MmLm)SvJ|=!_0zhWlifS zCPqDI7Ro6TcaKAD3%AlBvVp4Zf+8Cp!J50q_aD(xN5syYJk}td@w`LQXFV|MQcSm_ zY0UM1e2c<@4IUj@>7VkJFDwv=*JJ3rLA;B7;I{&E5^o&2GiCD0i~C1d`F!iuv|qhs ze!iG`|ID-rZjR~wYEbxnlIXSJB)s^oz*i#V0rv1_Uwewp#6i=P*{mZ@$wMz~_Pe}9 zL88yU(0(L&?pdz(@tz95DHPbr^tQYGlIqAdcc5z3)Ry<0_lx@|&4x1Y_`Qd!sVHu7 zV$X^7XZ36Nw8zU+@G>vaDUoZoGWgQV&aCrt`)1^K%mYsuKPT@kbn7p+f=F1*nLnhA zvKu#oFG6xUgydjbhDVV6t4%a-yHt+L_w2XM99PC@-Kz~;0ul`@V{jzQtW1i%^3C0| zko8TfADCql69XJb$6r<#5W{+1gP9tZ(_!Y5YMM|(f-PbOfc7w zh+#cE_))CjZJ|7ZZP}jJbJaLzHz8X6jy~xDGmbF(r0tM_&)W-Y zbH~RvUcFr=S~A{X9ju@pD%nuFq5Mf2V-J%&YYCQPeIPm7{AH-PJqmm9$M^Ae5+fW}KnoK} z7bhec0Dz1Z$r6n+IO<*`KS>|vNmW6#0334T+H@^?)+L1Cn3_{raRP0mxwbN)KKmO! zqps#8my)}=D6>_i^)|+GqVP=o`?LVf9v*L+4M|y>`8;g%3?i`Z_O>!xt>&R=hQIk; z&4RV=Mn9s85bRLN4qUTPqbk&?^e(J3wg07Mt+LVvCm9oPh}VOGy`>}hJm+%ALgQ31GtDg>zQ#pLqkD6y&38xvJOh5A|_`z2VV|SWSd9Nw(4E(>P$vloX~jm zjtYrLDjv~*dEp6LK&#HF16wy@6@8pE)+SV84%&hxioms@wwd6|l;()Fv%R*Sze`mK z8{F%PiZWPtk7EL|SRly+=^-%(@_>C7lluk!4_FvVqOWo2uyanb5K6QpGuB4hk0c%O z{2mH43qQkp<0#YRBKRsov1Pjq+6~*$0v)G!?irEv(~KQm(!VtPFq))(P^tGOocbIR zcZWy@i#wN^ z4ua2REROEti$v?l9ensgCaC|tO1b?GLp-k;S$d9z|spe+q{^T|B- zwz8@%1_a$_RQ33~W%RC6(ue^-U2aTY365dgE6fe(SK{rc=>!6-X38wF);h9c_r2a(JH)Q`Go$Gn(-#+ z1d3zW%~r;ur9^*%*E1&W(sCF0JgD@Zc!0I(a(LV#SWUYXik~+S!O9*A1kNfZ)Ch8g zdA<0dC(JCSpWXD0<$DOW{U4J`2hNR zW^tLhwYo6$A6lJ&V5F~CN!lbkNV9@)WL7C};P)z>0w1(v!<%@0`0K;B&KQK3*aYov zxb`WqEAo_D5``Ci0OzNKBysJVlCx?|PZp!f`-j&6q^;L$R1JWQ`9~}oQB>`~#0&?` zjG$x%kba6^(vcs^4xmsnF6A>L01E21-2++h8_ z!o+6|mcIfONwVgS$*hcp+cAPo#=On+4fe9x76v2ndDJ6NeKf6!3}b>N!niBjq#pHY zm6)R~SJ_p^!Sbom$3)&BByXWE4_juYIs!im?iek1vCB$AWjpe1sw#RR1MdTUe^m>$ zZXyriPVk4ESkw6^-_RH!`2Svo`ycA)_Wz`g2;5vtB*KxgFm)8Ub_f2OI%2+}9*M`M zw(b4DA|v$RTiXBG&I7xU5>wP3c$f1(BcrS_z3V%IQ#YphP|4)KCjW_yp%^7^3zgby z)j^^mhW`~A*{1S%hA_d(KK1LZS||o4bsZd=oi6)Ew@RIt${~JdPYat(`ak<$|2{!g z{F*Jk38K9k|BuLc^ftq(v-L3E|7Ce`y8C_qOo?bbZsD5q!D8|YM-fZQ?cu8F_20Ah z?k2m<7q_#j!hC`E4PRL~Un6_`tk%ZU6{abA{Euemq!h+P7rVSifyOL9T}YeyW_#UN z5#<+xVo%#Is6dUtzF?ortFvD(^|>EC|2NZ4t0m4@lj!Az`^)&Ld5%3NcJCyhhPEH2E+?3*0I#2F z6xG!XZHja+eKr$U-Pdy}>o{sZDOhGbZIvvP`qx;oxj3B4ypfJ!N~o94s}2BKAOem* zwQ({XH`qcQj%(JIPaVrsS59gw^zO46tXrv$&YG-}-hF5+yU9CAn&QN)F2LRUO{0p!VM z;~yZMQJ;r!2}tn-OVqLT{#5fyPbI;#K2RkSX`h##8Z_K|@4nGmY}R*pub=U?h7qnu zkn~r1tOXbJXG-A6#M78!d#hO;oZLa2XzMV})6Q zKBxJj_Ag~t?eq9GoIg|be%)4`adpG3ol>AHBa}eU*ZlXhfO=yFEr1P}g78RXLq_8C z+W`Vz!5Z++vPa2{i#4JAN-fz7$&U&;!m{ZBbcFJQS|~{@Q=h&b&Oa13F6(q=bx*HCCjDG-z5D>n#I4y zHl||b0w8aJN)aFNPZX-e24P4!>M*?eZR@Jew_|Q6VTnj{K!*R?kgc*o`*c*|DbZq* zr56$4chYhhM@8P z98SH2aJp{Mod-Apl;|`$ByqPG@!nUqFs6xW1Axi};2z~placA)MMqjhuDm92gW(&^ zSCY^*?Oip-t6z@^j5Ety$3Vcb24idTwA^zgthB7dW?1|&6;->MOc!f(Txs0zxU(P- zJNjaP({A2EO@dM_89I_ba|(@5Vtn%OJMj)AkTU8}S{CR)1eYZ>;>6M95J)q^7c;VV z^+q*GrBeCy8?Mr8OM+!Z1$8+q%@!;TrA>eV!f^A{Kp--NJU@nm#$a69ojyDo`0Fd#B-C;c3Wk$JON?7g*htGd z0%LwjQYorB7^U0wOxX%{4$nBJ=`fuGqNvA&l0p)~lbd>LKx`gg!qn{KOQA8GkwfE# z+sD<+rQ2_jQWG?v!wJ$-P^m-bbX7NKAwVC{VmzZl-gTv}yUh zbzT~CwVh8hsyA;xS(!@X`zQh{nJM?JHOF;U5CrPA^bRSEIGfT@jU92|(p@Q92#zX& zAk;|c9mh1;2*xtFRDyj?hsk4d^!2`ls66)TT5jLKWmfqBJ3z$0y`!xdF-GpNNODX} zp)oo{Et)F}lxOS}pnv;n;70x;(l&NLjT9eYjW%CV-mP zL%*BXzKg=a04CaXbf7yXHJ(^xjhkg&5E+lid9824o^Mk0^cVv09&L>WbcCg>l1^FXzcgOLUt63MyFJhH*N|*0n2kty=L0r*qD4hU3$V{+I$Tj zg@_>1Is0xolh>Q5XMs3wbb23rcQ@IkXE2Y!FN6o)b$+<;u5GP#H{Ay!$s!c4DE1>m zd>j`p;$LDPq-G6i-H0T!*4b^V6_FT>z>*bA94FB#7MQ@UTS=~&6M3eic_cn^gYVL( ztrYYl%ov|Nl<;?2D$_ffY_Y~|w^Nc!dK&L*BEMZ#oax<(sB;Cs3ljJEYtW zPeZpob$L{<10$i?qx=jC5$4hTimQfKk=w@?Zz3mZrfA*QvC(r(V-!P~9!|tv<54A{ zNAKE^?nRh?{`W^ioy&X>1T7wLYD)kG1(W1_FlQsUksFq-a9 zX^E6sD$>JSohWQZTMap zh6&O9rtt$G!(p>rP8|i@2ZnwrzlQuBqV^c7-=sMI*B#3?`t_&(Vm`|U;@LKt zx1UzJw|_Y6{L46|`dm0l(cUNBcZe)OTjpr$S-IQ26KhLdqW^LpBiM73&N)>1=I4nD zu~10W`+u5-eP`K-m+h|0C*QEwZ6K*Q!Ybu_kHAp;5|%B#X)2M@Nv?Bwl?=UbHnVnl zU%VZ0_-`eAD$o}y!G|@%cGSs%^zd=|NGd5iC}ov3t=bG|;4haReApUN5FOlQX;rpL za?mM3Xz5*_{ZW;eit6BFD_Yn>6*1=#{F8SeyO8JGGG8uq_-3|P>MzNcqrgyNpGbY5 z#5Mwh-5>;U1P*~D{o}_?OC~j9C)N+TJ;d(9$1EFYjI~7gMCbwd+CppcKI?kAD%atS zH;&16aL54@;~2z<4nvNFy=2L8-j6Gmno zP%|GXa7{d@#TDb{s`TI#=`It7;HIh-Dmq_C^vC~ zt*Aq}+Igwk+BFg92S|q^tSL(Xg~uU>Cc*9xq$~GJ0m!I zF(uNy^CUU~KkRmfF zxf_o-RFfYXMxu?ME{Is?={LvY9g0`grIlRgcDJ;WoOjRBB%^$c!}Cf|UQV2txBj+E zc5jY1v+FFh;#{)pY++$l{DGSzHi2wkOm0^L&7GF#o0tGi*7QELYz~c#Jstxy7e}Yl z2Xv_^IdZWMlfcyMyb4me69z;HQ-Yubs*cPFcT^R>f`_8QfE|5#iivt81~7o-Jy1##ZP2Z_N+f5-qqEEJq8;w-=jv2bq1+SDP@Z#3!2fETmi!c}7a1 z-FLsPCO0=AA4WWa#RRki%1G_HKkk?3qwVsBxJx?Tazfm5AQryy-C^DzI3^!+@Oa7I zuLd&=`sc3~v<=~=a$)AdOUfnjHXj|L@Cyr$LS)8GxiSGf?m=q2Q8gF|Gac|o76!a) zA6K~|U^C8yo4R!c%K47?W+n0BSpH9K8P6~tRNj1uE$0f8C}wCu?waP=UBN7dax(~E zG*{3lZ)Na63Q7}MIkEg@L+Ov6KfP3VJjOXV=z~UsLGKU_WvF=1io^ET`dcYs(Vrr% z%8wZP#DrB`9oadzV>uT4WY46z#XPwZ3S=03;Bzh=CU32cfy5k)QfOTG?{>cBu?UMl zlH`*VP6!kw#Rt}l2e9}rym@$;O+*=d0Mj-=l?&4h#fx0lif1UQJm!=ywiH9mA+*an zEzewS0F1sViEDYM5R-x-B`?%2%S;gMsbpEoyjLM!BfLq9Az+cou!grwO0q8u7PG}W z)RC7g!Le_}IRc^AnO9w~co#l>a$pW&v@hmR3>n3MJfE z1~l+S3TMm{t5w1}PgogcAZOQ2*yZIwcrDnvk?0t5j)`7+jdyWX4~L`Y*@&*7aJ(W& zBPaIj$YKhARf})g*Z=(Fx?Cj*EUlczG=14IV$Q_2AZ$8-V_QdnLgGxx9K|ron4FHu zDQF1pe2drl#PU3hzxjUlAq(A6fSl9US%E;31GoscGOG%lEq>_3g8yx3E}nZSw_bq+ zudY!Z@w}j)?;valBVWM6{gQj)Kz!wJJMha4s1?Z(lP*Q?1mr#j%z5oP!5Ip1*+%R9R6wacG3V=6Dnp1ZXm z{j>gz5j!7h37cy7jV|7s?6$;me-X^3ZnY6T$n|wMM0PJ(>h`GR2V(rC}d zbnFtAeQRGUgD-EI5#PQ4S^gIt$*^%k{=YZchussgu_9d!Cx5+P*~|#8v#CoPDJ(?p z>V!?M00lt$zmhu0sn&tD64pF+m5!0hFr|RjYzwU2$zHO0?TsPZi(mbhM0`o4(%dyY zbV#Ln!eC}6qJO%zh%y>^M=N^IOd)}xZTWN#Wd09W3DtV_d=PIF8j5(8CJy%y-8g14@l~n2dQ+KrXBX^cra^&uiTrZN6BNkS6P{w-4> z9xAb9?Wg<(-(w_;JA>^HY(ZY;fsu~dSjpQ9E(IH>*qlW*)3w!o z`70S^1+C{Uq}4QUOouWF61r=_?&`RM%Sot4~3N{}Biy*&i;=9ctjDvrc2 z1l@})??jl1S;1>!goM+6RX13_{~I6BzXGVap}rZK=DhZCn}F>io=EvAWAU7c#5=0; zAs7>2ocR>&2{6WaiXP2MCiq2f_gOshgi%lvjCPHQV-CQRT>g(d>4vEbnNbdJ9VE`3 zv!GYJvii;rP>4h)X3&94*7r3JQq8L&ayk32FwrzteN6u&G`d_ZAO-t1u$XSOP8Ab4 zT`%SaXsHD$^$et-clx(z=PpX3_J@itpp4KpA}xbymUV(x{aygtDYIR?__{*wkJs~w z;*h0U6=@VrN1QCi6UlcQjGpr|p1Zhc=73J;RfUpGe`i!(&1^sSde0-C=10lV2$gXT zc?}J`Bl|r0vhROL*Z2#qq;%HJZX6!mu)6xzyAwCP5zZjgIShY#aERdSx^FchJHSdD z5)`*}fdjJY8v26P^ISl2-mB9Q48Yw1VQZjtO0N360OWU2gtQ4U6z7tZx{en!``^8@ z27zM0z3+QQ13dJmHge~q+Zrt`9{MOLJ;5uUSl$-^;*{-&E6g9kfSqj=GLDVnz|VIU z8#FlW#3Q)**Buv$pzCc2EX-JTkL})Vm2d|GnDW!l8Nhx?z?B)5`p7(;nEiQ`yORLt zC<47RaLY0RKv(gFu^XPcHl6Q*x3vv2g8=|Zt7xzvdT4zgYkW@%LDnx|zK@zQB}5Zk zIvJp!=Idk^0-^0&jQ4KKXl@`Ja}>Fct0)K!koO9KvWsh?5`3wUS}A4;D0K*jV=E4{ zW{@cl000)43sczh(c^i#%53^O+aU8}T9;LnE0o5$Ro;r)zh|U8Zz1ZU@8EMC2srkI zIA)KRBMs3e*9=X+IVS2AsMCC}=?Vi^17reUPxuJ6kNK#3N)9ZiQDyOU*a34RA&0Di-bg`=keu3}&q zvV>9=qiSiStZlsK zPaQy}=-O21stc+fc%Cnq)rc!hJ~OZ&A7MALi~4bK`(-B3d_Pmy@Y~r+2oAJp(tI z1Zu3oJ%&vfz1*ZiDO+b3)|+tt>aS8joBAGqK5l?Jl7*y18fT%J^D|AmarN;c2HY=k zDlaSQkR=Jo($D8Q#Qw9^x0fFhNk`+qUA6kqAQZ=DeJ9kAlVUSwsn`D2BlXl|u2QRh z*F@X-p}!%z2T}0=8p$1rP#`)LyJGW>bQBKd+X#~XK^?uXc*NQt1_3Xg{v2Uhk0;Yh z{bh~4v6cvVaawnLJo~AVhp_(MYyX*XrT|m$6+}i*CV?cINJN;oM5k4WZ5&*G^yXCO zlkTVY?WcH*DoE&{zyHvDjtp9@^NZy2^_+*zWLJsWkLrp{dOR;CFZAEI{&;^EaCa6D zWr2s<;85Q`>9rVj##0NtPG-34)n|TQm@+2x&>X%8|MsHTD4QKFQ`L4_Z*5u{xN}gw z*u-o8u^xgj6L{WAXEIEz4(6qI=Cb%U)1S?XYH#Ro+h@>?31_+gbGtAaoXsCI*K^`i zYyW`qQ!L1-fA1)YUP(yw?qoCMC;rFeDuV}bakc7q6IRbx?SGE8iao7jR!v_QJ)~2r z25Wi24Ox$WzxVdO4|f~<1Y#S{y7N&2GyQOI&}hp&;AIzm5WwL0Ke~F$pg5py-L`4m zyBpWw?(Q_&jk`O+-GUR`-JReL!QEYh1cwkj5J-Xtg1x=(t@loy_j|3MYgWy<#y4t8 zzU4iA?tW=- zl^lznQ+*kB^4M?Hwh2=FH$IvC>*&6BOA%ghL0;tybvdHQnBhH^25kjp*5z~_NfWyE z@d=>*qGqc^f3*mX>khlJZm55X3N^bdZ-)TUauXX@f1u8=fnZqXghjCC)wsmxpMPLx z7!d7MNb<`xn`2FXb+7yj&kYP_fYiyqQ*J`AtsKp&yQVac{6Vd*o?J)LPK#bo|1eJM zml9~RAFbS&UR!I#1T&0x%^bH8-RvW+CR%eSVHR3!>8r7>BsBj$QbSTpk2!zW(wExQo+ZuiIHx=+SRPdTesG6CNuygyRX5LD|4g9SExzm~O=ah#SRrwAEjdY<4N7773MJlvFFBs8sIPpI)0`EnN4gQsS?*Haj@Jcz;I_OX+a@2ot#&jZK{hI!7egz1?%D`x^ zU9A(YIS`BeZ+?Z^G46v<2is;e8VCcPqi(a^4&0fDWNqazZt}fpd7k{g`IVc*|KmbO zalZNI z^H3)uE5T`Uv)x!F{eKnczZZv7*!TTXOu!1_obhtmI~Mv%n*Pl+9E=-UGx zv1iC8UQTk*CQ(_;)Bbv%p%3 zMjjb1FOpi+gRXGmWuv+NDK^bp(Y^v0!3Zeu;4rLBNMm0;tHpXjJTenoeiJ-N$$Le& zOsiL{{ZaD$O1B~_F$<$Upu6SYnOSR6OwFhEN`s5yNt!2yT2Ir*gLZF;I>)wtoH3`$ zy|n6I9dDN`#Ghj&R`S~ua!F>1@1o zZE!x4iz2e*h%JYeKZX)Oz|aHSIKxdllv*=-pCWvs@7l3PTAT*ZbkgdBaG&W|t9r0% z*#BB0Ih_Byb=U-B>78p)B4n?DM3T6(yz2*AKgCN9a5<}4s7tbcSkd0TiC^U((Y=a= z4^i{~wG{XODc+ zi;Lal{{8)M_Qb>zG?H6awWdXdowX7_~-d%f#@DTO@;S6?(Ll3!tI^1 z3~A^!qBINMHBD~dlA3kZCwE!pQ+A`-ds@7F6F`AXCHz#0XjVp!#bY~wyYh07 z$QWZCHlmBy3hA+tfSrA$XA=_TfAUhp$l(*TS>P~SDWL0fwhRry*hk@-k}n%@*b5R! z4jJh?@NW#%;p!?`3TyiAi%<%Wglrg)H*@|coqP^{C7((k#7$mkByBTH`cfFYMG2%x z;!BgRCq`zS7>?7)XpEU{(HXESPxqbY&ipX9K;eo%I%Zh5@!*q1?)$LR_C>cy@n%a z&hk_YRAe}-D~P3TJ(vuw9o>Y%Xr@sVhnJO5WDGRf7*7uT?p+a*D4Bb4^-R`=FJfxi z^FAai#dkt+?6L%4Z>q!WVSHw;w5wSw-`N~Rs{_9L7`d2hy1cXmI<1DY@9A4r*+3$ZEDc1+B84Or4Vez3< zAr?s?wHb=qywG?Tk)jnqvqUelq8RV2`2oL2>8Fzx4J84Z@IWk9KmsB<8JRtwNa=EW zi?#^Ie5~Se?M$u>yjH0&3+nA}5c+mYPz@!Xl$=1&mQ7eQ#Z1uSajexokQHE5FJ$4( z3@|3M&?yvDu|U)!)^r0}lD%!9fI`5F=9L5$*nQMS{HwaU_ehPiEUus2?&-_dzKZ=N z*P$w_>-(=GJ*14}{a;5Kz>jg*#VidL*1#m~>K3+V`@)35b8Eb!P%m!jRfW0hPlpJK zbt7O79Q#&+Lfn@VN$)6?VOUTvPgGHU+a7(q3_d4+8Q?{y0294bU%(SAg(;7ST>U6 zGyQ59-6*gHkB_luN5G|f`~1+RS3pGjKW6mO7yd2H+|ik0t%=?_Gk#z0J7`PbE(7V~ zE}!XfAKUC{iFD}VC~bv!L*=G$t7>$4BrOV2e7Y8ha)6|@z{ylF0_@vkn5f3+});ok!izZ^+jOHNW0nc`Q*ma?!9h;WuEs%{hd0) zOV@2R5~W)>3(BqRAK7&om{ zbPch8hKAYtbp27?g`|eHcLM(T<`3K!sr7fTX{JQ9?NZlEDqT!or?p3q9gK?R{}zn( z%JO>TA3|05BNxWBLp=B}72Hb|E7^Qv79k!IwTNUZL3`bvw!t4P?C0b{im@CUMzxFa zwp{#be|b#_g(jG7$lVT{HAog!h6<{dA8>MVGg+-9C`ZT=OAA$r43HB*d~dj4e!zYc zI&Bbr6dWcJS*B&VrvFp!j??MB89~~i2WbpZ#p(>je{eIpRVE52dw1o)t|3TH{b9!1 z?vEuYIT+sVCJxgk-5rtnd_v7?XQgb7N!Jx}_yauZgV!H!x$I3h@9l2-pZ?K?Cz%2Et#8s77!JLI|`r1a3Dm+%Y&uirgSG@Zz0XKNVL?UXbfNXU2mIGA=0(Ua2~u! z>jzuVfwd`pY~`s^3>7wjQGh^GB&J<9bRw31io|In3J8pT2Z@pxQ%7%I|$+-E-jacE%H3MB=AW z#dbFC{Ba05e86ICx(aLw8f;q~P3XNqF^apV4r_Q$vH(qF6;Gs&Z+vwW=qv|qs0sZ@ zJ!K6U>-~eBB@<=R2TI#hDv^hju+)URl4JsEa9)Sv@~|b4M0wd9SMv>u^G1aodZ`q= zN`_aGxYMPn`Vet!$uwCNXl@}G2RF|1B_;ABesz04WS;n+%XgQ!zua`805r}A-kj<% zO9RLF)U;ip_)loE&l_&l{NT-&%;7fd?IuCJ+@z3VY@1fGhK=M#w0M?WuC{9BQ9Ble za_H-&&L9%nT}Tol0rKr0rkA`mK^Q3ZSii<2)bAR%`GAORGi_@LG)J5+P3&@UsT50v zeF0Jrp~AjlkPStQkztOpHK0;>B+BI;uNQ{$%UfD-lT#-)1nuIi~+|mkCwpr zC9>bK@-y!XP~^DI@AY}YJgJCTKLhE3bYSuy3ANnvrd9=&0L$Ah2LUOg_#;0n2^?2@n=w5Q>clW_Vw~!jg-a8Y(F3T-uc{ zI$Rv$Cr!IT_E;I5GEMnYfmp2!hUMkrOHa#u90bJUqQK*N$fapqCMwLORV5{M^5ru; zWdC<%-Hcz{Ol_;3<+N6lUD=*kaYbsMji>t|1Bsj1j+9O@mwZBonrMngycMJjL7=U4j8Ztoh8r-E1 z#)6={+M4cz8X_>VZcP6k*Q`G%T;CbDJSX`UUD8} zCY|mpM#@o-M6(dtk6LmI8i3eA;qR4b+U<`luT z{#5M-Od;-MSCzF<+$=HNpRgR2v6ZM35Wtk0O%ZlYPb>gYqNZ{J+59GCzR)b7zn0-$ zGRbfQqkaXs5qwh`VR}S0A$0E_SoZu*gyfdXk~L}-rG*2flY_FQX{0&^2@7b68Uh4eJnM$2bkHB-koI3DFe zluWgB<*}DAn6_K#!1f&Zl^cpak{YTUYaERRS=k!RG_0NGG$0ebqDH?iHhv3Ooq#)Q z7yVKjr%{ahY-Od7Lz!j!kzMOvYkT@*oB2~W$$8g1u)U3A=OiRbXf?aOSp`-HR$%QZ zzXN^1Z0$O-#P?(8<}t!8X|i)=w{n%{3?#6(NoQ3LCtddqJc?2>1Qd&j!c1CHp^0q2GRrP#^3?Rq6p>CG}@ zZP;ddL!WvhG$p~LgE`}g+2dWG>jo+J+qSn!h_$*6^25@sAcO%K#YUXE8X42>)?I0X za3m({gU0!Q9;BVHlJ#MKnqHDJ%68GzVRu<3LA^PcMPQ5BuXAd=tT1BXzgbJ5qTafi155T9|neY zGAOqv>6$g3IVg&at3C6YbbfmLN2DfFF+*toI6MrZ{*0I~4%XHFSg96ch6LI3*Q-0Qw}SxqEY+FQ?f)U zZH3F6Ye$9AW}Gnfj=LfpfzI)Yf)Ug7JWr`1Qc+gSq*07(QQOdj-^eB0VD58BE_KBA z#hb{HoLhK5P?^eI5zhiMiguxuUg)+S|E5ngB}_EoE_5w{+!e6~8-bh3Cjv2!4z1w> zdjFzR>f#zkLKF*Of-*?L1JB7~*5L{?*PM&fQ&0BQW!R$#*^c~u8k8qYDD6rr^rvad zIFv75O(nWXyJWilt<+i9znGebVU2kTzh2Fx77Rr2bDCPhWfD^((JB2|h&84e}2Vub+y zI%AB3+u62usi_?LAp=n>U7|aeBaYXY5~bXky~N5>=4K2pU_K*r!*=K|>aQK@lOM60 zZiqK*mZ4K`yqV_h@w*Xd_p1S^<~3jr)pb4R-?gL?~Srn%R3 z3u3S}OUIl|0l6;jp=OB4fJ1X!TpX+z^?k0K0GwVX4PDA@?Qh2MYu%a~f?|bXFToQl zINl04Nc*_`vfB6qIPi7%j~?LjZ&MWg^{Tnumj^LF?6Rjp$QPHm+SBn=x!dFj&6K3n zSeRQzO8F59OsE`fKilmO4!%aFCJi7V{W3oSxtP3TfT-YbDAW`7H6*0=aBz6^mRsr< zMH~#g$p-U3YQaO$>W2o3Dxr4@TrF9dsP5nR`dP$&A3iRFFU&3WkO1|2xcOY#gM2vW zP{>yju*1z!uqm0_I%tTc#B$oo$TFc_opd?Hi0qK%l?t0v0)uqsyOhI-kp4-i*NMHv zR)p^!bmfGHghK9tIx{gYmI^&1@%<-`L0Wf1i3Sgm#)WqMv~%=0K@(k?)y(1d4h2lB z0Y)rA5ncUZ-|}A9w|^bh&>a#~zCT*T5n1c-4EW=f8Rbv<`x#gtuhQ?q6C@iTgh3#? z|klkpCp z?oLbuyfP2nJb+eKZkh^QBOnH~RvnRcvYNMly(Q1C()Hn;eejHe#X z^03a7F3S%4(Ze(P_uD(jHxwHRnB?pJ)CKl&QSK{;J4wI5$#^?`Fzbj4RzKiMt$64^ zFBt=xqrU&YVZ_+RqFH|;h49DQjX!1ZREl5hd^wuOidkpEcYnP5`Nv8Nf$c>uC=LYB z?QpJn_2^AulMaT(-=F*NaV~yT2>+=3)3tFp%{R=O5^ohUxfiXbk?I)XI&2Cff%j8JZ~ePPI|N75>1e#3x{n_BZU8ueUs$Ll7#=%Z>6Dz(vVJPUQsbGVy87s%re#{-9x>Of!djqOMa zao=yQY#KU^UJWD=r`*m7rYjf9R~Byh123bb0n*5xmvz1?cl!k^=kTb)($nfK{qkMA`$_^j?ST4GZJ`({kgbYAdef z+T5H+1K*!E^dSU9@|JQMWVm_=^4T0jO@{M&F>0EL$T~>_v&suBn%Pv~H73J3emaK# z;f81Ka)vcsy@rc^wY2D{mAkz(>za8o!W(~eS<^NVu0}mrn^0B&l;c^`ut&b9Hu7z; z-6%|87cj6&{SYI<`bnHu-~LRxajT%QUBJj`I&Ze7W#9v|vD+j;=Z^h?oS+Fw0A7fR zHvtB|x(p)lf4=4s006+=|8Hp$N+=av2n3fwz(E&DU;-h?0nq-JTB^C%ANGn=Zd}ts zKby_*9z#~M<^QUsSdW+$gP)+`7!;&!$6x-BTIz6F9)g7UvU+2y#-dZFQ)Sa%%l{La z$DI4$Exs30#RHZ1=M`BEgYC|Tuea6<_1}7ZJw@1s+CS?IhP+`k8tOqC`eN`{f1h%) z8+E+w$$R$-cv!XDpZ&w=?P|5{3kXJ`WY6)^`%)lTXBxDU4P8p1pG{IRYG4l#M{cs5 z?-kVi`XZ0Y$)4FbWXKo9YWBL6JvwnmxsJ&};c0SpZz!*~6FINs#JX!j=tib)oFpTe&dI!(gx z?+GdW4ob53qli_pJ)}OC_Z~3a7 zeGm>b6g1Z1{B1|4Me6NamUc#QMrRE#9Ow($Gf?ohYjo9 zXy-m94SI{EhzLsRlBO=T_Qhzw1Rt+9s5p5gy8{=?Qu>T+#I{b2MKs zw$<*Peke)i|LXJP6hnP!U1T6ynT5hIGj5yqj*T7POT%8W7Ni#LZcT)$`1zZ~-cQ_p z(OS>*T;hDwV=Em)?_JTg>r}bda4ExXae#J5esfwHM!tRT|Mv~=2lQ@*c4!&Jo`Pk! z^rYdt<~NbTheKc2iS0<+x|t(4S3K=s!}26Zg^8yGLR5Z+(F%my#o6+%Vb_%EkX_w_B!$$nY)i4XCiCV64mLrcxb)E{rQ z)$F)2%)2flIh4nOQKH*g6PhCGufcCq(nM$4=pk7MhHh^;>q!ej3;qjanlO2ABoV`C zYzQ0)dDD9M_lFd%&8N(tDLLh;K?^Zuns^6R;VGl=amvGx5>hi_Y;UY`AH@Bf7aYb3 zv@2zQVUiVMZOddX47UWAY2z!lPPtk38oLYV7$!pV1d-yQK^|)%LAnqD_K7=H85bE5 z<0cV80TMcuaf0~Fk5L8mKhzdi`Fk6~aC0sm$aIQ4YwEtD`2{=epo|7{xYdX;x+a}J z9k|?ILGf58%FpWwmtOb`R|$tMl>=!^;NmLTTg*L`X9s?llUrM3rQx{~SVV-2B#*EQ za@UoP*+W`R(p4O#Jc3Wel%CC4xmhJSv}gioTi;|{O*E+j=IAp|A10xyUsAGrX*iI9 z^LC0_fCgV`_Oj3>8V#3xw3Q5ASDtER%i9!_u_6G}eoP}tUGp&mS23wl5kWt_DDMGP z*pOu23eBC0J`N+|+DFE0B%V3INnKG(o8%Y)J6B%&grVDqFpLkMc>4jB6^~lOT-}RS zr_U1Ju#8{$z*(?HR+TgJC{y|yFA0geOb`zNO+lT$66jXVATdS}n0MER`0Ixza0}*H z)GmNjP9`lpRjcAxP7w>0HQ9sVm?I&Awl1fBM6|A8@!+KAA=MFx3oom)Oq3%mlX1fW zq(gJi>3NYN92}?N?Bx-ZP|pTBOTx zdA3d|0{`MWs&^)}<~bwvv^d#Xo;PgC&qC_IC5Y^7#_X+ptN6;rJKa8x*Gc>+({oRy z%X!oKW7fLD9RY!t>={%$3akr%0#vnR_SHkh6gL{YdD-Y*5)?vxDvPoRX;3su0MCTm z__fJ>owH|--f?vszqCIqpOu8q8|L;SICzY#I{Yo4{4PYI-a|8&aD-%b zRvtd6d4Q|(fXPt$F-Vt**Z)cmWHK7kaeFGIcTTwcIi<$O}L zH+zAsFv^jZ+wDmzpnY}$Gg2U^dSe}@S5SQZ`%YmkhD&dArnGIgjBUZ%$?MT`AqL5E z{=;uM*D~FO`o7a{E>iXcd93nSX*yVNH+RTDgUOejmfY?OFCOOH9%&OiQYGCLF7DeI z-8D)r#|vKar#TlJ>?K3YA%e~?{Ea>h=cYd8_=wENjhHkSp_d}GBOD#9AwiBdoJLfm zgy^lvT+Zgb=o@X7cl!L&7~gzFw-GJI>-3i}oLf&Us5L_XE9{{kdh@^iF3Raa=W!jQ zo1|*~*gjBroBZ+b6VWS9*AOblWyM385fbXd;7h}0Fkj$a5pCQU;{b%~Y1W>x?@NT5 zuPIFU-=Qwbqm3+%apLXM9_kxo*9>~nb$+=p?Mr3l+*|5(=BI)aAxZRB%C?h%+voW( zX3yY!jBlS;y7;_Ndh8Rzjq*i{Os_g$Q^1rq9$iLP z^X+4n-Iv;V=x)?tcfz)84Jr-X&j0&F`8x6R?dHqh{`cS51YyEoe!b#4Kuj-Jb{=_f zFI2u$9E||a9voTL1uk@v)N~gDhXo;7XT-TT$RvTQ6kjfvId{JE|NXwE_jb32YJ2_w zR@nL^YI->o<(M(uZ+t4_V_s@{BfzOptH7Q@hemZ7-r2L~IB+9fn=?S+?s7E$Dj#0o z+DcUM9G{!ioCuJ-p`~7U!}`qGbsqitq_*%+{qBcF^Sd#A*?~gqBgVJgZiacuCDnC0 zr0h?tGlAKuBpdRX(Vvt`sE_q(BHlw*RY52=YS!YP8Q6w~}Wt5X{c<*RAr07ZXB&m=4Tg7#z(_ zuqLa;QI5{C@An$U1svvfb*E8{5oFY)*S|Jpul8X60bXbYPv}K*z(D{TIR;+-w@@tE zQ)y_5LKFi5JHos79YOS{p2l!D0xJn;T^UZFKzF9-ymXn~)o4~tAwDan*F~JR6?lSG z1d>!Zumo@wE?DuFq1MM2Nz3iVj~cpH<;|QGnid`H*X3s#6FNBpzDvQfX40s&j;N*< zve?I9M2j2MfRtzZ3Hkar3J4+J!B%}h*n;uROm2N>Zmb{Tr)VWyOTbPVxXS<;ZJnsN zHnM_!Qrc1%*DE1fhRBm2A%zbCn=Pu$_#)_XVL!4HF{X%0J)rQke zNn8PJZKVyH za|PN<4T|e>NhCHrhZzm;e*{JO>28FPfP?etn2gWlk1S@<$B4&`du$!Dg3d?Z#=!S`dg+nM}G8BY&cCmOd*rnDTilmfmWH{$3zzfiLWde>9jZ8wz=0Mg+l?$Qt( zM;kYYIi~g@n=>oDGTf-oHyk}W1C`rsPao8~Q6Rrr0JtMnM2C0kCy;`LU~skVChY0A z7^Kw8+|Mnli>EGMV^q&!@6t{KtR8*9@lL~H|^q85X-Dm2J_~;i< z+|>~ZLHlGQZL07`_DBN|Jxy@O!#f9H+EevM@thL!RKAbAIH@4)ykcxa^`hf!cTOV7 zWKG=)l2~jET2|ZGY&4;!$Jp)%zk;enAY&*vrffwvf4zcwNR#*jzf8c0aFh)``aiSh z+=7j+WN|4j{c*mv$sAle&)Kus&j^KEkA#G_IS!MWBw3_TcZzOVf(mI?n=PUTZfnY= ztoDsO<@XvwjJ~;@fsuc48Jp2VZpwr#@$!k?oqcWJGbfQbn7U?&E|Da8$9=SZBOy@J z)ebsLR+gs-HpHtZB*>*U>|ADKuu-qJPh&IyEmYXsa#so2hE(^IP5r0%OOT_0ajXJx zp=t3ik<=E=VGMtZ=^?cD2c-QN3i2X?@k{_c#rUr^9;ep)6Nl7>hkW)d1o|UhHd4f) zC-t8YDa3DrGu+9f<*jLXLVcq{@}doa8oWPCDAyV5xgb27)jq)Ds^dIced^DK!vrW{ zL=j+~`yO%VF#pX~X^9X1=5^(hZqxHHI?6d1jTP{o3j>=KkbBgG=nb;d2bDol$U>Vz z`QS^3W-?aDH*XLrYYWO!GlM9Ikrm{xh(SAxk4}XSY-&OrMmMGc@x8GEjGNH;$y)jI zTSb*xMYfx1N{}`7YEdlOsHl;Vnvv1B+ibYGRr85#1=_3c*(ZJ4X+0p{{o8k2+F!kh zXf=r}0@^KEdEY;^Yecu})wLJ!Ga1abJ7RXe3*}YTtu$pMkn-;InWf0{sqbg{1RUb# zvjtshb|G4Hb(2QpqWUrl`X}XkjMs)BDS@6zt96Gd;l?4^E zl5ESVmoTPau|hq0-hz;;_KY9+WUmJxVPZE}P5)Rt$YiSm>CM8uq(>bIfrL*xfK0H= z*q(irIbI(Fl37$eq&VBmRMOO z!7=P0HbNb83$!!8h_6E~9^(o7flVS~bVz0_F0)^qFpmeX*jRam*d>O;1^#oA<<=F!P=5s>gRYoT#5BrmP=` zry03OP?B`7Os_2r)SUF$iSJb^?B1@3E13J3Fgc2q@b9VMgpM*_$n`*Ennt7h^Ha|| z;P4~yFq2e>u@J~;PoqUgUpdee@NK3b+crtVGWCr`qV=m0B6tSDWQ^;jpySk{VlM3Z zqIb2GrQKmAlAzCb%cu6SZ%ZFU%g#%hx*()D@~vPnM5{ph+lnqbfOx+*Cre4tc`@il zYt(5~T@-Y1;1J%@ienqQ?N~^l0*UyFFD|~ydA~%m%AEULst1d;8gBb+v20SfEQ;;< zD^S7G*H~Egh1m<;xG%`0B7&$4flE9`#4o)Qk_p6Tu;eLHs{bDMp?os-6WIXMsMm^g zMsz?|Zj(M%zbL^111R&HR}P&@BuLOfV1&{hw@N} z8%3HGKKZxU;_c(tj+0N0;{`fBr-^A{?luzm@0cNHq=Y))&n}Ih18>3X{>R8{v)|XNuVw@! z^;iQ3egrn){WJ`RF%8=DeqN3b=6yRNb!a~w@ghzer8sXBm1`yZxxM}|f6`BRNkeK8 zYCzU{Ll|xZibk-G;`wR(*UzrqY)x>@bBD-D5mFRnw?M-dE-a-jc!bI*e9rI+!}0rJSP&v5^vD;`PuhT@AIw7c>emc3 z7dF`taj>Jx+Ev;0PX@v4fwvm;ROn5fc|^u5b(?BquJlzz&BBv%0Wx8>$q31ti`@kk zdRZ~Lk|f5cpN&SS)UL0Y@W{qjrn1+e6bG+dOfwlOI#$qvrl8kAs>??MoMD)D9vQRu zI~t)f4u)<8t%h-&Q%9+%#KOw*(#lve7#JI>t`v`imRj8BGq^1NNAz#fNAay%=}&2- zmbYX}@F4)e5HRu+Lfha*jaLwMrMW(ZCWCG8-}HwIC&R?e{b2L3nEB#*5sA5{n4(ck&v1KaWAfAp<1^&?N8e9Rn~W)o12o zsm*%QeQnDxzXeZ#hFXXu0y6d*9@CI1A9N@U!I2oeAqx%BzdK)5gNmEoV-eo!)tL_Z z_seIADWsJvWHPC2UF(CUV0rw`gav49#KqnhQ&My2Xc{A-$nPniPydu_rYWBb(^p@e z3t({So(Q*bqpnTK`w=tE2}(D9t8Q@s*hw|2aR0ou)k}!V6cDm~&pmO5R{o z&n|{h-9>%l8)g#1*}1*LZk=AdHpb2;&M?g>Vkf6%QxkJ&hxp56APo()jw_GjpAr0UF7TiEU~OBawebQMmIeiX2@&@rA98F9?jC<;#QzbUHQeZoREe?4A_Wu5&x+h z%hH-0g2g-u($ZRbS6qo$4;F!cz7#D**~EYuW8 zSFf+uegi8T#cZvJc~2mqkT}1V{CF4TK{H8(gX(ZuiEIR6Q%e+i`m7evz>=cwD>Jt; zly@YSz|mb@AJ^#CeudACi{5{u*;-WU`L$ZUbm`N_jS>!0=P%AoI%zQnS(S`Rhq%+U z*l9Dg<01Hay8+l~5hmFqw_lXfT~3&+%91o^ioDZO<@21wIzWb@F0ylcAG6uel`B8c zwi?MOl})};`ESIs(m*$wYra9a?-r>1C3{d=B}}3gE^4po*`z+Avac{CgV;}h#F;0^ z+v2f1ce%z?cxP|dOc8{Bs zxSie%e@f@dpO#3}n5e0npPM5Q=@*Z)Cvhj{W7roV)ve2Y2_^L}mh!kurntYGIq?wk*gheYK5KcEMjEu+4z*6$SP1t-sxKJK?xye?lKj6stCthmtPpRuggH#B(IIW#NP5rQ2i*sAfc z%|SqV%8$VF;IRCyLlYP0T5_fnK(DkQ;bONdFC6HNxq&6Y->s$b%R!H%1?8WbLq|1+ zlVGp(X8%$j3{vFqrVWzruxZ&8PyfQxTXR9Vt_-=^M!(zEYt;PPrAAOzq|Ca75Pj4B zSHx(&qgoC{iJkQz^72Yoe}ggX%SiPwGvHr5E7Ipq{9l>%;ry(0?{vpK6uw&e{)VSq z>$nM!e<2Sg+%Uf3`Ru%?P#3)P-Xv&3huetj(pPCrZ=uvFcr5Y{@aC=RfR@6I{_NNp zMXG-Bp9pgZj?oo>11bpKl_%HdsIoo7gH2p4DfUUy2r zE6kQWsrlnBn!AykD$=Ze(ol^Y_k~=`o8>IrQr^vrcQ7sY3q6)+V>a;gpm09ELXNjm zs^x4y)H>|NQc(qH98Vql*ZM=FTbr#d>G+29C>foU-0iXx? zkjDOg!%w4uM=|Namc(IOZ^1sW#vPEv=Zybczzjn|BO$3j32fWQVs~9H&c1n#(O@B4S zTU}HQ2`^8Z9&3Ok_Wzusd^$?XL8=%kq!Ml z;lD!mVi%gNA)C$t2gD5Q^ds@@^Kf?j6H^)a;B5Z`)62JK)@Z1>b9?nG@W<@ZE11O6 zZ>?t7BXkPdl$JyCwXlzV;1GhD&q^oVg&rd^JHOd-q!!2}dyuZ>*{O zJKezP zzPb}n*w3R1bF2<)uI|3O${nNrS2}BrXf-bWa7P}7e`-apR?pr~iD1Vvv6(vJIVb|{ zFb*F(ZrgA~oG@AaFy)aHkyD?}2uJ6&6!*2%yXRUu?U69Us))1@`iK!GW9l9Z|qW~5Ory_7DTB zg$e8 zpcwngP)tLs)(C%sPCLOwBtJkXlD-+`M`T(>-A+!&Z zAx9Im)gfxcuSSbW)<_A0gT>^oK4-vt>kvSI=G;RQ&2h>FU=`5s~?aitBu1KZ5 z37}A3OYb9HcX7f{uP(8;v-C(p_|2d6jwF~hU1JG0h zE|{=69i_dL)IaImBD9SIbV^kqs}$>c~5QLNnD zShX07)4yI-GIU-kUo}B?R@18Q7vJpQj_76im!Gr?*9lDx598ykvp3pP6wBlNUJCM7n_-)OYXw?tQ*}#`wlK{NM)= z=A2{Q&wXF>nR97rYbsoE909C=13uvF3jzQz06+i$3IH$wfCB*h@1H^-01N^kAOH#i zU?2bv0^ooDDFg<1_j_y0RHz#5DWmr00<0#!T=ZyfWrXz-@`+201O8pZ~zJiU~m8q2jG9t34#Y; zcmRP1PEw57&w4~LvU~y4vxUVQ8+jT2gl*y zcpMzS!y$M$3=c=(;V3*DgNNhra6BFk0B8t+h5={3N30q1`e$FDL0$O%9=1vl(PVdq1lB!CF}qNwXJ{W+anSnbm{ zaqJg}`b2LzcY*m6k=)vKCE6+Erh?R@@j^0Cf>VF4c755qFz(yyQ+-U>cLj3a`gRY# zl?)e)+7sy#MOTiN85QV|lo;{TheGQoH~Si@%nP7lo~0Bc&~Q?{A+-KLP_=a=7DcVq7LaBuw29rt^M_Zvv_ zL-AY()4hpC$?C=zlF*&h~v#n$F?fend7G#SRMOy>W0zH19`7 z%r+k_$I?3$aY6Dj4+*syD&6SkaQFNZ)=wB)WkaC|_>)WF>gr)5!x_X$OQ751`b`SA zl0kXXkzZ07pLiHvEM`b0m$8Uha`oD-g*iP?6JWEjpX@V>1dk`VV8J%(YZi{wB+#Y2 z*jsE&{ML1gA~G?=Dy9znwi&{AizotIZhzH31Y$-#8Eu?%;&n^f)cJZLP9@g+GBG)D`|s1EP>+mv<|8%o6JlFW zB2*?sZkT@>Er*`4+>NW8yP~s&6Sj_<~M<5zE_fr3Xf-nT|8n{7pv%M zF&D?WkoI_4(g8nxeHcfrQPFK%ZTSPpTMIh+|6(R&!gt9$my9Q#qZb_?)uO2lrg>xMlEj43gZD+ol&NKgkV`cV z8Fq87vt}90NC2eAe|>#*|NV82wI300V}F)W0eYJU-Lop-+iF9p+5erLrGcrvdjH(W`;eYD{iYy2#F&!sSb>p0l9paY!HMxnX1B07LyVE^Q`O|TBtENJk^82)Om)p~ zq+Tt>Awa6MKU8}zEE~nMLmW7N$TAhA4ome9%=h2CsQTi?1?27AVxA8oz@XW3Vk*U= ztN10@bx!}p=^QSfdm|m{Gc`OMJNAMMpWeb)HIs}77KJ22UK)K$N)b_oVAVav0i0@B z3A3rVC8nRGp-$-qg#M0Q46l_A*;$4bNaob*F7Em(ba3m+GAXGTK2k~ zzDw>-g~%!$!Sr=*hX!DHzliSj`50WhHyS`Ers8?deSIWzMOS!#Fv}n)*)VUd&sE8I z?Y(g}(*qN6VdaAmf|mKj=(X40#h1G*4g9?rGl(0O1v|ChMFm|dns;7=a>Zspz9^eY zFcVUo%`(0bG**DuE5CY$(He<#j-r@m_^$ zocM4pWka@myQuN#UHv$sy$6Uu7o*wMp%<;ml_m||dV7=;&l8f-A$#6%eXsG?ub66_ z+Qn&qh!*t&CK5Px{-G+JtOJermtw#L^gLrT>{H9P3aBYEY)-jHdoy1k*C$* zj%ii?654T#L#v1Ip14fP)<;%}Yq(jJx#$&(xO1PjAq%#+C7(vhfA&{8S*`3 zw$C|XOyTutpli}4|HVMYn=0xI{#+STMQkgVy0kybM2o2}(fYJJjRRMwN_#FynV0m* zH2QxZ|owvoDIQ%c;#@)pc>$!1{CJ$k66`{r}kqR-FV z)Vmt>A|~dap40b)TGu_hM*F>l@3mu(r1^(H{awk+TPQWUmY3?ndtk2nUK{Bq<5jE; z-dk74uCnbtc+nVm^ZJH!!Bx)TwZVqU6mf5R%AE1;(*`@y`?M%Cy3&CpVXB^ZPrt(O@B`{o&gLJb9}1wZwDA;m?nA|}6v}qb zp;?uuYUfD%DluM06*ua=^w(=QfEptAN$yq5+C;qv8N1M-?q){=RqX>Im|XQV+K#X` zP1We|38Y86+?GLXB*?GrrC1%(RQuuiy>lyhpXrTc95OF}iS|F%x@Wow5q^{(IIJ{W zp>cmT$QgaTr@+}cb)c!*B}!Av{{0{qJAaSAD^;SXRR9i@M>nZUwCP_r(@8DDXV5M% zypWV=J`>+PyP=MhwqIoX50J4+RXRoN}HWrav#q9{-Sgr-aqjK8@8WNL4I=BLeIaZOM4R zPGbJ}b#2Cs&E0pR0{$*EL%F6<2#%XyCWt*(+vkbsP#j5%sFC0gnwyErNC?AQ1Epzp zNFd79z0`q7$bp?7iY5J==ECV>Voxcc&6?&)%8?T9*eQm}<-aL&h2vcxUl zVgx6mdN^;fdiTovqaib7a*VrXfV`p+5uNe6sY9}x&Ikke7UiT%p%$=np(mq1echXY% zpz^epe6l9&LQ88V#Xn0{k5le$Hc5Z4%!3Ui=ZH70k7WD2JZYzr4%vCnRnh)*BF?Yo zrx)6ZCgT$yQg=mm+G4wpo**&`y_8dlml+ zCUX>SU&EBwyLZ`(h3xEV%PtQgfxm_U{~4wjP(B-`4oJas{l_r>WY&KUb5;b}{}|?1 z5oE|X|6$g%Va}MfzWl5R9#Zn?U@C^sia^}9A5%G6Zd9yO%Dy-=R%KD|<+|=y^|8jG zE0#yM@!6+(k5RP7m7CR{n{JjjKB>ArXVV(|V|{wy4tDw#5=_Ob*Zg9>E1F5dZm_xb zOE2o2ZoXbi-O@njC2!JWY5u=z8T4xzV4&`=S_c2BS%WzTW z-`D%K456+^EGI2z%UJG=A}6dxR&JNQD}ZYB%Qn}0d?|5}NtJfwdbA zJBXJjrf-lKdt{z-`gFYk@$u>D>mAvFeT3;Fz0duR6jLkpFb*94c>{M!C&M~0`#e*O z%kbt13W~j)I{a`XNu>%g(-sE9r6+{$L(jY;hAxbeYiJ3|tc{d>IYkCmZfQv(m1)`j)z>d)L&?Jp0$ zvBt0wn6C#YcBjs<(Jft5YIn=pe)p13dH-X(^}(C#F4sG@l*QF$>G@TbI8qt!o`eMI zg2-f7=fGrR8FLQbS>})r*{{98N`KeKeN<*w^uU5h6}D;bklgi{2s)`m9 zM4&>@gnW8L{UMF4zz|4FA-EeVHiH{iR^r{J)dWQbBcU3c9OIID%^ux}?7W@!D%Lud zqh<$Y-xt8wuMI7LWbazCqR7}Uu;ysV-Zaec3JUB7YhFFlMX6luEO`*9xi`Rg*q&?; zPRFrTgn=Ua)Dz)tYPotyDKv8)v7`H%p;+=|+tQpUFVB#LaPjrZiZb*%`dQ=pdew^_ z3h|9+;*Qxy(0_>hOdXtOPTvSf|4SYJCUP#Kn16^oGT85rIv|HtPs0Hkql7bcENO<3 zZP9FsqTG;R@GM809IE_vveZM$X>wVpF^UlNOXLn(xuhwakyjcj$E&W_dQJB;mwg0e zTqkm`(0c)uB*Xr;17=&;FemWawv~N&XYj;Q z?nNi{@aa?Kwp#2|Q_3YwM)#d{7i?iUt?Uh1l}TKhw|lUx(F+Id%}<0M?kxx3Kj5DD z+1I@IqC9pnJX-7ir?S1K9+t-s248h_Leoi-AgwiqJ9yDOaGBYpsPB>vRBoS^+QX0afUl(W3BPe_It94UO9Wa%chE?P)5 zKjxwST-i&E3FLtGEX#G89Lh>hE=z$HM|u5GM~dY0-|FC9cdDvO*7mM$UPf;`Z`(f} zs(wW(vH7A~)oN1*1p1d`M9((#Yyi!fWJG`y683+W48_(j`(41=vr^~`{xXz^eIlZN znGq4d3tT@l?>};*rg!=KXDPQ-0uDDE-o!GAoDO&l%|a82RCwDmj3B)!?0lyeL+XUo zk|1iN+iw}?HFLmzq(gmYato6)ZJiP`Dj$lMbtgrM-Vz;tVw7h`y}dD09clrdQley7 z@2IhV8H88Z=1mhai;}73tQWF)L3+POD0X|MN|EGD6zCJCJ=~Go zMt0Ve(G-*B>nVJx24d+p;rN7euKgM!+3&myYZk~fdB75y&seJ<2~GpN-p z{7j+6V=kA;5B~P&`o-lz6wm(5QI1r%b@1Md@O^Cb*MfH{*CX*>hu|I}-Aj2=(~TlO znqGymxdj{6yM>n}%C2ta3Le)yoNr_%K%Y7Z_qYbk`fps=rD8Gn3`}EbG+wgwc5K0n zE5l_}R2XR6Qs%st@3UHkW5v3yqu+;H%=$`omatfI*9D;rl8Y9{;%Q=B&RG(rt+2)j z3-4wp%94sLqSRTBPzfSJmr=d)Ee|Mj$q z&J5x`pz>b~;+zgMxx+tF(D+M7!teNCT+oW4C6Y0+{GEan`R5gGm$iqJlFnQ@^mZ_O z)a3Ll1()9zsGp_a@wT(U!&BP#seKsq_!HBbGlP)B_SV>SQOb&Mdm<~0UZQOWe#}W1 z6Hpfi71QSo8iPMZob)zH<1j>7+*2Xa6ou;0`!!NB+>;69k3VONr|bq zzI;$~KhDAX$xhc3FMYlGjHBM0|CIA#zT^&Owkd@}Q1Xdoy`KY;v!G0$TVid>QTZGu zgP48UrQ;#b4ZAkQp2}XKk3Qtlml(rpnF0xSXa`%5o?LFoN#D$BRSAu`IIGSKT$5St zeA;^bLx-4eA7M??h(GgIy88Xxq>r-^Oi9ad-!2xH50FI?|Iy&*7b+Z8mG{h;?41R) z(o{2f*s;M`LbvEl{C%3|9+W+LV;;{yIc5jE*B~K|cn?}wNIb_GFk($d+d>~lBxGU_ z9N@q(#_#|O4wP9x`Mmlap_^wE_*U3D%|S3=#9m8$F)`KghS;QvM_SJ^RHjU<97Xj> z7@l?e6vmN9zVV#wEle02e!~S zF!5q%!uwnM;d0ftSNQ#nZyf;H`E4vQ8)CbYT;$pIYwFAAMcQe(nExh6&>78q_&d$~ zCzO$YXZ!yFWqCRc=<#1rHr?OwA5dmHAoEte3Gt<0`MAbmx|LRP*|x%*f2Ku>z{niG zQ=p6$rjWpwH@i4@Cy<@Cb2;pE_8gENkKWsU;S!!NP=znW?R38b9bfDDY#y-C|3b`p zN>2$`e^X?u^=hE`&Qk9^)7h7DS1+l}wO;0Yyw%inyZyQ)@B8rN{nVV9DY*R178`!_la3XS_PN{?irRkDYoUHH<(>~BW1qt+ z-Supm4ilhYpM6N7|JWgfoD(?%+8Zz9u{%`>9=_Wie!%4{tQernv?Mna2_@rpoMFhN z5jOG`d?hj)Gk+z3E{?42HdQ<+<)VYDz;eZKBoq2Aag-+gm2(LNVwJfrN*7y<+z!l= z=D{-8Zq9@7pFEbdlp`lkq0nGC&MK{f{nm{dEU9j4HTf;^&004OR%Ah=Z#tw-=7`eAq}$-9SwjaY z@xOwI|IF6vfUEySM|A(<9+%a>xCbxvTSt6BvL1eO${D=cWx3k7iwA+rDyLPybQCK= zMS7x*S;gjRrihnvU`nCopeWi%otrBoRoAq=?u&2F421F|3ASnPSHL5jYSDYRVLL`> z%v@t?w#>@=u;3pvG&h5)XJf4?CUr^4Gab8PkSh8TMs7x7ch3>>5J|p3Gd6-}B9~fw z?2(dE>?s+CpXW&MbyK_c?Xv3fM{?fRsLv<0zoPzeI`Z!21%nAJ%PWkC(4`kHkCVv2 z#IQowhe^d7GzQai`sLaYB&IGkp@m(u_Y(=#u~v82*H??E%#vGN`R^PsdXp}u*8Ttv zjP6k9H3b|Kh%8yyWRS!U)$b4!G4ZNNlZaXd08cYgth0nt^mHlud*<9nz6)7KGE5eK z3I|bJSTLGB8Fz4A$+}GM%a0ZswcN60MJn&;?{e!cZtXe7Kf90)LrMyPiTQu}yfsJa zs*;~WVH?!B`L#$F#u8*=_BmW^k{g#YjQS6)9Vm`QhNNj08pnM-PFS%FjwYmO$`S?* zL$Sd-(R_*fZ(43x-EaLgWNCPO^zr89h$BHGu(I1~F}<>-Xo-kWp>b@g+hn#Dna8@P z1}Qc3nXRQ;(KCUI#8GHa;Uuw62q(d;yi%w=Ww?Z_!P!>DrBPcVx^eMnJm2JN-$)Ix zes}1`tHV)5;oD>N)u&YLH;nSF{7Zc>x|V^g&ix7IteB~Z#V^U@`Er}OR&F_>=9%fC zBevJYnWbuMrx(V}7yEMpuQT$#s$=mMr{drS*kR-OrDsP8;(hEI+6NF5-M;6_IE+wb_M z@r79GFb$0l*Z7ROnIvg+G1>t5YAv)sKah%v=kgG-HkrrNgIFz$TK3h{d;? ziMswtFpNI*r{M~P>oh-(n+uu@;^eyk>N@awZfZv>^9$-{f8r{)dvRWXgEshvAIFfdW_*+Y2=Lwz!CrLme|LCD^uvg*sRJ z?^_wtvyvYC@1&youcp=8q`ALHWn{iL`ByZwe0NdFklMO+wu2EK$WgF-&3z>CD;jiL zciGRPAzm%H>(l9%{u{cD$j_?!Zv*+yxP(!Jl=zwWY_I5^4o%y{tW^_b`2&cq`GtiK zP<)>(Rm-kMhH?YZsm77L*W*)qiF_JrlwC*J?IyEn#chYE{bWEbH$F(FH6c)<<`C25 zH%RuF$SJ_%#dF1*5GD#w68dN|dTBMESiRO1y6IiwfJ_)6FE7!Ek7a>Ki zu;yE**=puPw`nf3&=Y9L-bTxw-bjZ&<6%ceJI;$532u=5=~BTAuTEGLTYacj$h3Dk zNdat;zy&K*vshoAz2$!9toBv^EN)*i53wTMbPnQzZ19z+GSi!??#t0`SJogZ0`Ds! z8KOFGY;yhA#Y8o=zbc5@GYbT+KINWX8)AspE&WSA7 zE5x{IBx`G;(79gf6UL7i2|?$K0~`BSuP1-6^bTmG)#1!w+2@?C0%d0MRVlf0q`l~e z?@iWFsD2cxZ`F78sqKB;QdKu;@{%q2V2V+ymV$K>ZhaV1^;y}Bei-?ZI1fFZCf_J5 zS-5WXt1Ga(@eNGIza~_@#{n~Yb;H&g2|ks_V~IL!ewsWtx?(<$A(sv`>^kMVi3FWt zI_JGWtK8>LI3(UY?>)JT0I+1|&(~^QwDsJ16JCZGlhoh+3{;SD&r6q_cXJg@Ae|Ah zc&11eX!wDw&6%k;^?6KH<@^dt_ioVoemNPN#xAxlZr^z;h^}(X1UmnZHmtP<^ zle;tQ@P@DGm|KUtzqQ`;lwj@O-Mwe~q`k1uFEOQ!HSpWX;bMm0>CtLA-|*hL8uPzx zCeEF0|1O=CsPErP^!%YTwCm9sz1bYz{8OSBi(AE734H&gHwO95On=Bg`u~>R;Kz&} zAD{lUdnGHS|%db&StwWIW(lm2aOV4&(Z^hP; z-xx(h4Ai+puKJg7C#f{D11MeL-38^QDQBS@8mu1KR@(1(r^6hs9ijXWj5^gx)Xh3g z(`lxoTr$+2v!CZoe zaCA+T4kbB?)b$P>Nw`0!XqkLt%tpmaDYL~eqEE#XGVS}>Z%g`sWLQ0s^7y!<9`{Xk zFy|igDg&=#&W2yBrZuhx3-A{yBZq!^qm&fKksz|Ey5usl7nT;TFWykCL|y&vV#tUh zdeoiLvCkRbDVj#*@oZzEg6Fkw|I6Vd7vx;-PIH_hs{jhmjRFs@&<3z z)~#$s*(}Wm&tF~@G)COk zN*383Zb(uWAIm&1{YW}!vsy$&0 z>*4Wksu^h^D%kO%N}C^VCp_5N8)*KHb&z1_b=_#L)5!_uP&LSGK`*{ZSCgI|l&PHg zkOGRcds6$Ote3JNh&4D(-sX*o-_Mf4)}}rvX=j-vjaA*ZFA|q2`C<;;ZZ8fQ*F{oW z?)_L9u7A0-Nq^Y+iHxp`(9qu8xc`HliRL`}tqVcB+hRmI&D@R?AHM%!($KeTI3+M< z^_m(ivnZgDfN*e97^VE`urIA9xrv*{p2++o76F*GDyP%x{A2zo(S>M8x`$IhHO4gncTktaB4tSC{PwFRE2%o*dD4DyI9Nj`Ucdt4GU z<^MtQ8$KfA`l%WD(k;VlM`M!A)L1NFYZC)Wv0pAQh!|jV+$XwDkHFm$pv8-+>%rcZ~;IM1oq4(1?@95@|DGYVP z!+38-J}GkRT~vB-Eo{4A((t5Dw4$2#4iyuUG8N76m?X>FHDkuKvp9U;V*?aS8#h&8 zuDrbpB7v=+_c!xouE+)9>$X7pMwa6Z&yu<+(hh8nXj$|;w0(nur+*l*sLQCRvm|?s zZN7YEYq_MXwliSWs83r`gC!T)98Mv<@NBz_-Z1r3H^XCA`hRZ{zyFsef!~^>5dP1D zH;&_hKf!S6R7c_BLdO#Ul1AOLwA0NHm7|USG&0eVp>W%&O$;T{og>6|Otqf4Jdmb% zmUf&D#iLnd1$Da|Sk}=h=}N`6rnI6P75fpSK9hAU6ZIa-*V3|wJalqAzZlRsNY8(a z77d$z8CX+;wf9Zz=g|vx91JzfG~+jUw7JK(%SJI0_WYBg(z}GKVF3lDLR`MPnxi;SO zta7!iy(E%0q-?Z)`oUU{mYd*-g*t54D!j9@Pegg&i96tASGc*69)d$z&bWQxbL@D2P)0(omWo{PwFP*dd=`}Mr@oU4W`k0h>^VHq>$)R`0V98cR8K`lRVB2`Ozw5X3CpH8X zPsP@dtR*dB5UFC;p2Oo&D!tbiv66z%sTAhdTL}!MDb z4WoTpoeoS;HxrRtvspw@QX~xEZe`QEvBMHj;7!V2n;b?~Q86ElWSA$><>xrA2yl9K9}33MXI}TBn1@?vtc8Ogl-uc zSTIJ8Bxn(DQzh0g6fNa&ousjZT7p1=p4%#maoM?%d*ezpvPF!!kqZzbL4HMdXU6cW z%c$!e_X2fTh<*{yw>C4kqFmNA6M%x7l$`zvMF^_60$qtN&TBlLD5Ov!puN-}pBOz$ z-UlwPKM&dSGK%2`4Q%90ZmBx~y|h)Eb%XwQ5q0l0CAKj9JcUdbK@JkA#<3u8R31^d zH9aFZN%wj4%xjKJ=mzVUZ|ioH8AA03sgYmO_X!uXEIj4z`MP~6%!EE6GrI;x5T<_d zoSLI=qWz-EJsB8O+r}t_<`N?i;4dgEGc2noV3Khc~sqmB5$JwSD!mx!>C* z{LF3+NR28n^l9#&w|K_~dh$c%&HKUcs$pqU^<6UO^iV@;iFi!X0qWkasszW8?gwk> zcIGkT-oYCmbqS+#>aOMY_mHHIx8~rUq0_t8w=zx|K5I;D)vNliH~3t}QD>1J4qf^w zd0ExPU!VDt`+;opXBTrsO;o& z$A0IxKX_Y*-QRn`it+57>B{s!AZoHZHCVa+B;3;S_)HjX)T? z$h`z)zHDUJzRLJ=7No{V|MyBJKC9zD`wh^4o?Fs$oT?s5|Aji_wF}r7Fd+tipzDH+ z6~2O&Rj70APvJH9Ri_v&)y+~k`7{-CN$^GZfz#Ij%N-ZP1htT!tF1ERDyj0`jV_$V z&Tn(BRyv&zN!c)~W=+#1;WFOTucJTFPCI7Y>WUF;ze}^W?>LgAxpA?i@1uu(d#K}j z$rU3>hg@B{JG(jNwO_h8B{U!CacqAm06&BWDWd9CJ{YK(^WJADQ5z+tG4(k}y`?cu z0q>gd8aTK9=v41#j(&PzXSbJYIW0Elu)nmCS(vtR!cJxEjccP5`&4ki7Y}aO{Or1>kSH?n5C5Mkv1UBM^?;V&_D2=nr{QQ;*I0c<5&gDbuju#7Z`0ip~Y=i3QU~ zA6>|tWf0w$FA6#1Xk;9?MPUvET}!i3Wr?XMh!>Q<`9WRzn%g4SYw4qUdy)|+4^Mpa1RM!EZnbQ@}Bh#F{}FEcJ%$sx~i_?z`jX@;-zZKYVyF-^}A<}Xhb`wEOh0$O{R>%NMxa1gOkM1F>UtS^L^1&ukOR=km2CwTYN1`Hwrr% z(B*Et`K-6x!zHM$zV%T$NCqSlAJcqIS@vFtV$nR^dyh?-(#qx|v}{W`BG!2?5PVxt zXYYQvTuT}ZFT-vvHS$GD&&`#ZW1Wi*h68W6q)%9rvj$KxHD58BRHmCM%L%zc;lX%I z&T(}}8P2|ql^8l%DDoxacc*t?wTl)AqS6S`Wg?5b(=C09WGlW$oR=Y}nRT+v<+e}V zIHdDU7Wvud^w1x8(98ly>4UyjFnDJ!mCUPLZ4X2;(6Or&gKRjBR6;_v*GxktY>$k5 zw64yjGtja-01T3OZf@ES9m}1aY_=zG65nKu-Z?~#k^}MsoAvL7&urTauCt~6_w*+{Ls8Se?kj{7|A%K#|GsMG z^xB?+Kr=>OJFOPab%}J2JZVw&i7?-!j^_nrdpeV){B8yP7wH&A@7Dp*bqoDToz)Slvx^eD`m0m_3rP|r~A|RwQC$c z(+*zQE*}cN18&QCT|$u}?!Lx4)+#54>*{!Q?ZiA?OeMaFk~>o`r0!F=cab1ZNOS$z zLpY7%Eu)aSaIo1$ri8m=_1?8)ek7vzv+Ew-k>_Iz8u--U9uW~X(%w{97fzPS$9z*J zL(=bgUMOFb&o*lcSc{q&YjF&~I^Rzl931eBIC!+t2zRs!?FPFR1(<(JU01pr5|Kb= z)$5BTKNxWhZ?eAMQR_lwM?5{+eM|G|1tT9a?vJ_ZJC-G>j`>1mnZAe9z0(W~PA*ta zEj|7*<*Y{{`A#*;@S*SwWzq!@9n1v@<}Z^kiY0raVW)cZE>Ksbqp*2KnMT&p)Q%`> zCFlBiVd8VPj1+r1Bw(5Ck4kkMV-Pd8i>Xe4>`}_-9UYb{%!xGmfvIuC`V>>yd9*-&e7Txk?^2pXope8y=_cGuWC`3BFDI~)…rrjN? zx(*^aU*Q^xMS_Hz7VhRMeH1nbJN0eXeK`>hYSf}^E55=aI{#XpOYnhfR1JuXjQ80) zYfGo@7fvLFW!C%VPQs)KvL3Hd-zU#SP%+8pZhw6#oajeE*lXSjpQ^lbT|YQZdy>xLRe%G4Mg%neMf3l| z#b#PrnL?E}kEm(gQvc9A+oY1h$Xt6k8R;uivF!)`bq#oBU0#$lcf|9`|{6=VpE-AbfY7<^_b+m^Ymr`w+=kz7&o!2jR zze6HEt%N$C1UkJ(bfVR~JqPkz5>KamR_okLd;hMSUP$fIQ_FfHRpRqmUmUOzAdpS^ zoMhcYu_}XmErCNlNaULb+;7oc;~kVMQ5&sy-nHiOg!)o*(yL7Ln%C^V!xw*mK1tam zk>!$A>+GkwJWIP{_jE`{>oegCwPM!RqXSI&^Z4ACp_px^a9U>SKG-L0q1JA%-JPEY z04K3c$iaOwP8my~-2pqowmG{~8i}ZYTA`2tePDvXFv_LneyNQ;Qxtis$sCPCt}-cS z#+*upO0Y$wKD$~l|+ z%C@G3WiPof6=e>``g2g*z!<5h0*L(LcUP&)QL}J2CO+)Guj*c68 zLaFrlf*lq+Jr-!2cSyaPo-MSw6nv8kj)N_dk`O24-&051j<=HRezC*9R9Q67&`1%E zXa8aIA^+0MMf=GZDB07Nl`?VqtMHu#km&HeB$fjdh?d$s$6s83#SRuj5}lzb$|I8n znv8KH27en|$`8#=+M9{wt2}_JmyP#CTJb$#f9l71;{QR?^sOjXsj1CWg=MCpIV+g4 zmDEL(ZhB`x5>DNz&UB-VggNwD)XIFSFvC8*#w#d?P#7D{(Xi1qTaK|z>+~}p$h@*T z1T|!QwEXyIyI6THl)-EzH+cC2LEc}&f6>Wq{6-Ig+tZLXcGiQ^NRT+gii)hfMk zL$B|mP+vH~&W*ZnC>@$5d0TMv?Imn?0zH^mLm}YIUrx!NcURVn=|ynex~P08;V}Zp zQY8fR!xu&iM0tdHxR8j>G}}>C60Uk5y#iBZ*0#XJ=h$aNKF4M`>E+f7vsPto4sZD|UPtSSZDq7(M^k6Ajmb0Y6Ysuy#7Y77m76Z^iB#)8|9 zuRWEM+KcZ^l(;JX;^*?6x-jJfMEdum^%S2k3Bd!QE9W{6Z)5K*GL&S59C!v;-361p zu>-WGN7Zj@T}7GQ{$ZO|Xqr4f;b?t_Cvp^|$?l+z6u#yFB{`UDC^AUNOE-2GI{u-1 z27}uGLtWaQVEl!Js5zUwp}k>=Z?ahR#Ov})^e+3G=3{q3ZK@;w-)C}+Vv1v#+iX1#dBOPlgWq!CIA&_OP&$yocICB8MbhG64$GDPtIei` z@8jgb&zRW*Dtb@~DX;JcDXuyR9Z|I0m6&ytbmn@+c2W3Te+_Y{6;Dn*ziHVu{h8LL zeKz%na((@elp3CYs&`^~#iF=9!z>JsGW)JLv^v%0Mkc7iHODqnpU|MV8(j4k1SY%a zZ6_fEfp-@&dEeJltWl}d;ARglJlY8t0>LcjJdKssAWu6|vyve>X_C8l`^u&D* zruZtvzH7%em{nx|0L8iRxwzXqWj`j=(JZ_vll!y6Yh-xuMAm35b@u(|22)K<`Lqu0 zD$CYQ#80m~*xnDFd>gnjB$>WjR^XRS(K{0+!_mug@3Rr=FyFD>n-IKPo`Ct${I4l9zPelI>*Vk@ar1>l+8< ziKv+al~crREdrcPBZM+^Y>!PH2GYfx4W|18=SCA0z^j`&L2g56{x~nm zfyU}Lk;c>gJRvpbtR8x=tdDaAJ?k$F1#{s!=PE2Zqx*XF2ZOz4dQhWMn_Mkr_KkNW zyk`bEzAwH1a|S!V@}bnU%I;ljgLh-3{Xo9{%SNBHa=(pV?@@f4rM-PXP5yVo;{s9v zma`Ry{rieh4iQoRj~8GB2}L%ajD!ZFrR@82EiVYa%TZvO9PGQa(vhe9upcjm5!El$ zOi5NpW7f3F42s<#;tgoc45MnVM=b%~AD-ED2@*>*ZbIwa+MV;KBQ%AZFJd#ZPM|+shV3ph z$M2Ky>$Em4CPmO}lKZ69M<&85i4qLL(=90_&;nifSG-ml8d?InOY7pFR;V1V0|*f% zFM7YPf{$v#Ip>zzp-5z38rR1HqsMOb2JoC4-(K6RvJjaNpZ_}b8tS#PIB+1)AKDqg zBX6T1t=ALJBxM$CsO{pAyre@b+~{i3V99k^j}zv<0Gu)=(b~!id6l5~CaH*yy`4g1 zmPx-HGP=5riGj2eav)gcp-v)@1jP@#FN0Miljqv-RTe|-0!{>FFS$Vb$GrB9zP$X# zHhr4{Ruk4~4`v&P^Aza-y{rRZKc4ihvi;fto2kbZ+)hVS$SFfPLJ0I$_pr!)<~}*A zdbg4N(Yd0~m;iEX1D4ljTnMDGyNM6EMKH7OP8wvvTrh}b)Jo`d$@G>|e!zsz^O01=%`kj!W}b2dqFLEbK@y-R}A($aKnS-s$y-yAwdRjjm#N&$*r8^E$r?RY7vg z5n8hn^b_e8u3?pSeV&$bSXs~IGc@x9Z@69U6Nb>4TGC3ZZ=i3mpSrE>yjKl)h>vhg2vI_%p(<3ou!tFZ-V7>D`R`=h|tB**S$rzJ> zXb{=AUn<>RmW_&QBu+C`{JiGuWro&IegKoNHY&B)lYFu=JAUEa3;hR8?HA^vsqzmO zj3TZoOom*P3Q29i39l|@UY4yY$(m(8Q2TpJ1A11~Nr3SGom(;h^1oNLb`*hiMyao^ z6q`yECK<^AEmO%MxwUQ+Jz8prlBM`{^L5dvU_X|VCm7>knypVEET{|cr(19QXq=W96SK&?&9ZO?Hrq96_hO=0 zHV*vkdq`4dSc3g1Q^eKKj3Dw`&)pfK_gMtF`c8Ew6my*U6hB7V38MBkIi^MzfhfwvXYTYd<2ro2CSZ!pL5T z6go|hlMPFzs~;2w)Ktp}7zM;MA)aj{bX79k!d->-ZSJFYfwwY4BQ1+8Q+<;Zp2ppjHlv%DP^biSWAuDY{n}1+!wk0%(d; z>f3;#{wGGp*V51!@d5h-U)M+Mg&u?Xgx7)%Iuim0;Rj_hMXyUJJ_3FWs_~P?o63p& zU(%;PZ~vb@t~?ydw(ZZDF_vs&ERE3EXOc`swuTYeO=G7lgKW{0ii{y z$kIY6OkRpMQ3;Q-hIpuuL1n&C&-5~QEYubWB7CAM#S8aqRy7@beKVEig$Y&+{=kI1^T7=^H#k2lesUg0^NLyJdk(e~!Bi!j#WX(WS&wQ0zb!gCP4fs8 zy^=xR)Li*6Q0Js6wDur9r*yLYnt#!9V9{JnTiIIx3z)bG{{%Q4EVDA;_Hu|dj9XWF z&xu>pAfvESO_7{qdU0L37k3>Zy7`oaihKH4`8&|huU=&p2FosK+AFCCe0hu!x2K33 z#ZkB&V$|Gu`cz@)^g__bk@LOk+Lu+*FgmC;X)2~gJ1tQMWsdh%HGwM`yJkpY?U24} zc~?$e3Gh8iDsB+TybKo;&+^mdEhsoU5HNFTRi@RB3nrg4lfqceE@x<-MlFIog0JMHH z7SP`pZ_KKYN#yLOH-^b#KitM-X9`4oj{xEG048&%W|Bbtk~_f3HQ(OfE3WiNj)G3x zfpc`fUIoadT&uxgpXXdcksBHWGT)O5q4O<~Gs``oB7K?rmM$xmPa<5ibcuAaN^A}r z`A+nDX$4ijeNhJeBD3;nv<}olFt{$619yCX%~!yTNlr$jRATiC6Iv_gHmqRBB0qU> zUYIQA-#Y`PSjjjw`9aV9D>YYL1%oBbPCHpnyweL$1*Z3dto1_R1b$BPBy~UHDEgyG z1ZsZ<&cATNv=y&2hr^*4rl~qL9$~qJd#sOR|4JqqT)&Dq(cOtRPh|Zi*uj(6 z7)4$$m}mdl?tzF(Sp2LqhGk<`lF8p>{5?!JftFpTFmls_E&=_>OC$cB`KJTU$^LCxdwVWK*q zCvR&aa0u<&P0`vh@U7d9ek9(Mjrx+v(Pjl#H030(_;g3LspLL~D)DMdiZMgfxF!3; z(_8)(hgwf*EDU6?-fs*o$o`<#5YpV%Sr8Ure9Nb7Pj<_#+JQE+8u)CE~}^{dctu3>sa-RXrQ8fI{}U!R24w$x=yx2>yBFWhJR^x^qFY(Wosn z^h|u7X%7(k9QA7r+m*nvFP3fO*q9~?9}XqwMV*wLP7)?2yAVa=5ebI}=!_$>0P|`EiAG>MEt3=Nh%-}_2bSin{>a@=?^Mhmw z*d*arUI7FX40?L?!gE??I8OUTIq(ntdgDy^rfp_8S^zSWFxE>Bq)0vXqw1wzRf+FpLESmRBpplVZX z%@6k$OjeS^6NysBS@eGMwDQ@i0df^jE>AsBI9JOnVKOQ(5OEf0x39L0>w_ZR#A2N+#w)ZG7UzS={X7@VHQ>K^*4azT>!+SZZa)R-OwS<&Z6zyF>bvx zPiwG=E5m3}tus}2)^%3Q&U&Hy=X){B`geR>R~4aVqGZ5?PqbP>)}m8^<7{etDaroj zaN1^*=%0}DM6V(C>7i-|$d`z_BatlG1=c8;CAK!E-QblxPTO5eoDlJ}UYkra-RFth zp0NLf%=UTHb_x#+g1=r*-{u8{g!x}JiO4-nMR0GDv=5Z_q)EezT$^txKFJZ)8J_38 z?fW48?&5pto|npI+0q_1o-fE|nGx-a>vzJddngrnnVrqz z+~Q#M(G*sj6#CQa4!s+`D8|CMu?Kb_$w4XN3kQ0C(>cv|-d=BxWR~)zU@=R{3Iv*k z@aB%-I*YeXI-rq;2=-!4EquFrq5BX^)4X}Xmx>Z zF32VM(uhZtcl?-y+ZrJr`A&mjE`?JhB=3@M>NnVY;a`GLZs8p_IJ}XSCP7n7HAbLt;?q?r&T4V?MmQ;t zS75?VnWtYo=eu*7en5X}2xC#|(rbYI<}IAx?b`HV=mw<&HJc>)9Yo8|RX*E`Iqn-^ zlbMlmJRlT2lxdbv){p`qa^3R;ic@e8PhQWI=bKF=`}UL@(0CdCh_Wa?WU5jReEO#N zS+Kx?)e&l}0gE7Ig$P#GW@MO2Jz1>CQeqTWlL37yxei@n@;4dZb%MCCihVaQVk{)I zR7UWUnicretCubOusw)wq39|2rQOA@2cnUF*%?$}&!9ibR6+jB&$JU1&p-bvK7Z=E z?!e(+y#4=gT0NKZSLZi^q#+e6v=!bn+bO~dx`k_adV4%4yziW)%TcMj4>}eigQv_Q zK%7kZuFcegbg(Cccj%U~d6Jel1P>>dJ-f#3BUQ>5Qqofh+{Zul;pK!$*-nMcf;~3= zFQ|7HeVz^1&skRyDtzy&tjKaJ3+CE?THjH|i6K!**{EfSbP#b_R(q)$Wtb&u*%6nV zg?N<=mY#imZ%V=_lgbzPv}KLv4*Do@Bt@X{q|p_Yva3D;QAgp4@^+j$>|8MqxbV6B zyEvilbQ?s8OL^q@QeuMBnY zBJG=2#>XoAPRYYozaXbbMiT1`tfmxd3TPqJGnLt(i8=`b*_ivJ5Bp^0VNW!k6i?kj zF;C025Ts35?nvxr;P%SAw@ts(6<~KIP}csMr>fv9`;5@zW_~2x`?ZQBefeC^C_hJW zXY0AQ-n}IzE^khlAl>Y*{e8XAU%k3IXa400p+w(71Wg9O1kg8XNi*JJCMy?83^!LC zWDtqhtLCd7DDKCnc%ZcWV=E<|?5ScZP7F}o`R+*@;fEBs(cRJ~)T+D3q$YE1Z8T5C z@yq1yL7Us>&);OyD`(HWy|5~kyb3T>v)~(|(rTHp{zdih@8@jfA3{DOZ&jwa_+8-k z4XC{T8WJOlX1VY3w01pNtLQ+sD%s84)|BptPW#tFRcC<-5g|x1#EZsyEyJ%;RkI4_|Mk;?|JV2iMkX7 z$Kj5tZ`Ef7hb;q^6n^7OJP)b7?bn;GFSjIMk>%OLFw^0({=u2xO*ry{GZ9}C}S)wA0_bWBt6~P zk1&oEur5+;N-wp@Q4-uN;9D-~&SNk3O%a95O=<1H{MIh88S`SZfVRT5g8}_zLJLYE z8zDZ==tn5>mjez4c2+s&MU|ZrD>b1Vt8NTJ+W1Tt~oJ49b2Xfj^SwZ6>bp%PRQGBWs=clQ=;Lm>0OOmBe_RrV^>+wdhg!Z+$^%L?{Y zAzbuoEB<`fpCAXxSI_ir<6_9_JZ)>b(H@>|jQ9~pn@HQsri4fc z|B9=p`8V%g3~TVVG@|B#ecxGwcYe@-QX3POV@A{EE)lMBsP|c2av|Qn2{*S7!Pj#V zVYwf)B(yK!y2pmy)e^+mwQANHb@c+|Hx=h7qw)# z4t?7F%V0Ui_+Eg1+wb``eJU3>={(~JL+qHSebUd~^btY0$p5vd^RGqOTZ_^=eqWUK zNzRliYmg-EvS0DHqyO>E)XwtW!Naa?c&Miv zP}uzP_TBY_jtHRWp_6N>*eyI9ibbZCn8ff1>wa77Q|L)qwapNNGv8suY`IWeialkf zIs7jCUnt+X-2%7yU*-0gfuDfm=bk3mYa(hV+-rj>RVv3Fz@`utfC}x_Q_zp1_ zTZ)a$D)>rgnL(KoCE&2Xu$glX_PG3U{->oK9y~A?!3@#B-e$FhB*7`sLHAq!?Bk{U z0%WXlAVqE{|79Tg9uUwFn~W?mj!DEvnb?vJ7^RB14O@yL`4HF;c+%mEeRQ*06**bK zG1e6G9TQ#@|F)sz^URA9%WJsSmM`;y$6&AY)9%g3^t8i>Q>5XqJ+bfanMS6#w|1=T zFx6F55B<8a!d{v`bLJhU1O8#H{uq%A@K`jk)(;;BWcZCy zV0)Uz$x!PidhA9V(kmJ~9X1*(ajJ7HW?bpfH44&;$1C~U`V7Nld}wx(q9yfc*U|0H zsfnX^H~`8~2KwKcr?7>7^sgR72yN4R2=CdFM=^Y=n;ScsqmR<0wsW|CdN^5*5Q|^t z6D{{&-pESiH`s{KUYhL6P~OF``bA7z_EtTi1sWR(zg(4E9-7ZsWZ@jn)>Q9%(-Bh5d!Y7LH^|fG|fk5&--eC;>KqoPWH1yXDJ9Ucaee zlEP;I0R!~@2b<%4LcGxtc$|#8XIN8fw>COM3BC6sHS~@OSdh>|uR%cRMM^+G1O(|F z6hiO42b317bOb3Xpdtb3y(>yLkb~=8d%x@5`+Vm+`H>&FGLv~<*SO0VV?OE{YI3)1 zJ^|*y2`6y%f&c&v01yCx1OPMuU;qI7*QXE&0D}Ms2!MnDXb6CT0NB623V{JI7=VBQ zNEm>I0T>v7{p)%N0)Qa^1Oh-J05k%?AOP%NKMFwtFeHFL0!So)MgkZlfc@*AAZP%F z1`ucfi3ZSU0D}gwfBiiK1Hdo<0s|m102%{eFaY+idxBsA7#2Wa0VEbcV*v~n!2Y>W zI0S&h05}4GBLO%XfMWnS7Jvf~I0OQRLEs1o90`G=A#e-?j)lMh7#sqF!(ea(4331s z(J(j$2FJqS00Is{z+nhD0s%)N;AjLKgMecZZ~zI1AmK119D#%*k#IB;jzPk)NH~Co zL(p&-8je82k!Uy?4acD2STr2Kz#$ko3uy7<6 zj>f_R=rK%#Bo>9nqA*w#7K;J^90b6@02~6qApsm3z+nI!7Qg`z90Y=cL2w8N4hg}b zAvg>KhlSt(7!Cr%!C*K942Oi_&@dbZhQq>e00IX=;9v+G0)ay!aA*V$gTP@CH~@)* zAaO7x4uQlWkvKFGhe6`7NF0F1LC`oD8izpRkZ2qljl-aESTqj6;2;efkmCWccKsd+VXVfDoW@yt?q!+r-L?-IF`+%TWlw2z@5J#yqQ?`+5Yf~z9dweh}`F`7s%s9UA1M{OZiPIk|5jG;SFR-U}{4GJ68(M+wN88HgAy2KW%a_FNoetC0FZ0YBSmYO*q@=n1 zqTU`?8ZlR_KB2WXhoU-os4vxebCW(K#e-ih&smLyfS4M4%@3esl{v|TD#^UF4lA4E z{X+O%-;6Hz#Xh76I#frXEsYQQWR=#gQCV5c?S959IpjCO&+sV9I0Ro?hP6Wb7$QYo z7DJCq9rA{s$2j(4n$@s7==)Gd{t5+@e8cpkRt=h}XuXy!LZ@ zf_koNMNVx)Um0Tz^3R@(RgZ@`IYghYf17{aVoUy69)34!y@l$oqiElH!!`;U+j?WG z+Tz$$qG_McQz)tDhDb^ghtOFfNX>6`pb66NbfC~n@M~Hj-ytbtphHgsApxb#ACp40QD@`uJ>L-dM8U7XQ1-iPcn43ekeX9t=ca7>XWB z8BHX*_3L%dp$mh*$FC$1?IZhb!8PkFh53i&K_THUuJ1kQ zMo?r>Z(wq8G>FEgZmUP!>FN)35C7RY7;1B2>%DP`=# zFI{7f)R<-EBUg!)Su_<*ypkEaHUw^ukXhnAVPYPUjCf76sBJ=&zf$%inj6%YhXZH` zc%;i>U_!;Ti9;%p5Kx5Ni((z-o^<}wfjERB8RNGGReq#SlyOlJ`G6mjkH%}Q_HA0g zT(L)H#QCXY5$XLIS@oZL_92hI6tVU4hI}wxXUxjc<^pMCDN=e#KM^hZ)vC;_CNq}Q zrC~q7D4TOz>1Q%|!zz#bA&<6h^3CMc9Kl~@!w=GEwJNX%!W!t2ON+X+hi?)@U6+S{ z^fcUQLFkGnHIA+vddYRau-Jdd(rx`yC81W^QK~xZ-of~KR!%7&pX+qCo?2t_m)vqw zC817r|AFM(T}NK^st+#WzjD2e9dA!Kt3Pc{$wOav)LY%s_S;U#*nd%RdnpDt`rLB! z`8xe08E2M{UI|YMCkQHTN|=p=Nq;mTqY#j#(9{gh{#0D~wo>@!h-L!qlRU1#M=;qb zgMdroEdCEpnzAD~F`A!q$lNNm>bJ&SyU^wN#)7vvg+Av=zs%SESt%AdGnSU_U7Co$ z4Ua)-l`sD)=<;?ko){i0XFV#Rw_sNZ)cg?Jm-_fNgA@OwnGdhxnrtnbs=`u*>6*dD zr5=+T@aWr9#XLvFPn)(a;O+O13(>XeFI^2ktbD3-WKgh=WK*Og8&N0SE^}EEvI`c5 zTbr018c0bQ++Mrv3Z_ zP6@eQvO$q~*@zRub+PZyKHr%?ikhNWO)h?Z9S@H@1Ry{9P#dU-w7T4H={_F-z}i3X z@ZwK;HdO1v=-yjj?)3o(kDFL4!g|q8AHDERQb@O%BJR_eh=P9D1Ah3s3`UH5KO!wIVCspoEAe6 z+&UmpF6tnxcg6%C{F&Cc8^)As^R9ijUN`9Bg5_yH(`6xB_SEzBpqgpi2RF;czy-2( zRPG|dZAyXT8R7Md)@IwI4_B^lWOa~T^iiJXdqz>a#l4d={-IbVon10QtGn(W{nT#g zZ9aUju6#Qvl;hBo^dd}|F7ZurI7z6^+#Rsto^XEi@P@ky!*8{oPp|G}XiL6Yc{g`! z46X4itBkJGbkw|lqwQxtj%IsGF=d)r-#hP2z%3i5PNtj?#;D?kAUj%Rlnx4y@3y@| zqYj!fU2ZCO?%R5P#zThp^)a|EWIgHW8=kvo_z6tdsCzemo0Q)3W{s}eHgt`7E6^;# zs6O6k?WE@E1=Z8vw?8u_Y6R6&e~+}|!tcZ($NOn`vu}i- z^!Nt#vV6Sb@D=-}`8vsI`y-yc^HXI~eiJFAvi0{g^@umGOeMqSeuRk|n6E&7p1zu) zNb!BNxZ0oZOI#=Uy_oHSo6M#Enbl%cFgA2+wQTr9aOp3j$D&i!`!5wF z4!0?P%X02dMWA&7r{EJFyw)HxX@@csmy2e8W+J>awcxB{xuJJ-b@=yKDnrEx9Vsx^ zsQh?*96fGGc}T4~9i#=!d0u;y?zb83;(kIq=xfpy8XQ3wl7tQ;m%4F>$c>TMV?r&Q zddUAnuYzW4#0wui@dIxN#?|P4n74N%(_1UO7{1wP_V%BFgv%DsD=n=S#GCA`&LFjwvKqmZD8vZd{@fr0Bp|%_2M(}iZt~qJ+>!gAG8i5iY0gSt8rAf zWmZtC@P(LIWoLTiRoa3;e*z;R0e}yPvI8VQE^zBAFm3@v#8-jQx;Ge~n)Tt(zXK!B zzyne-m)W8El6Q#&RN5eohSK2_K_k?|;fAt*1xAf0C841#uG`NuuVX$IK*QDULdOfn zOBe!qEJn~(pDXlBH1jl@P?Obv0^>+i^;E5O|JV0I$0E}W4r9d@qv_S=O>QeAd0H*6 z=id1J+MXM2sa@y}!6)O;ZmnDDkD(K>eBWBX@(#(Xm9O2_ur```UEv0*?boCg?sA$; zEcUe7Ai3T11dwO&WvZW^g|JFB-P1n}m8T=p?b)i{n=wImZ z|Dfl%LZ1!DU7>IM7y8AX05F~KKhgh*%h&%#|0gd08~xAK&~!;59`;Pp@~gPabZ=;w z&hA4#G@SIBCc1K>1ePNFKhfX19Q_NuZcErTt>&8PMyJmxi_zv+-&#EW7kbw{kmZ@{ zztOApCS0LMG?P0dbh5|iwVK)d0|(bt?9T+0{)+uqIOG80-cT^rKXFKL|0fQtY)C0I zh35*#84aIe2Axg5y+$LyYN{O3$lcAc^09(HI9MA?4dUbpXCAZ~FUsesl!U#%)#x%2 zY-(YcHj=hj8`)hM6nS>!__fC3^ZX5e#yyi*w-r?04$^+MQG)GWlmB;js~T^7W%*}J zV!fS2qg~|jX`T-T?hyW-efHIJEk1HhlQGxKeKp)izGbe)L}Wg>BvD-D?1uYJ@TE0Q zSvz~@)5P(<4BXtb?hi_vi=}6E}-Bm*6dVo-OE{-{@U!b@mjm5cxrj=Z9WxI z^@F$sx5@SorFTIKml-*$?pxuVjCy&q9Ut;+2-gMjTY{cdMy@$n85g}=zxJ?$qLNFw zWh@8zB#eXJ5=)p(@ZSH~6TCZBUkg04fvCy5_~TM+ebsL-OSmh&?Q zm_?NFg4hZ2%t2F!(xXJ}q2vUVh9b8rnH~xO!Du(js>XF&*At+RT_=B>X^(Ky# zxFjCWP*~A<{&OTKg~J*s)0OiOsUg=`+NvoOCWY}l%Z81MX&25w>F^BvBP9B1AHca{ zrFRnL^Q3P>o8=4-Htw{LUnSbk@|n-sGqm=4v@gXSkn%BptmIX6it(u0?$dr@)XKT? z>*k@4aS_lE2+Apzw$o^!Hu(7MPVIgITdk$br*C_evEJF6UE1Z9)F>+UgGrFuU27;| zt$W9Y(rnyqC=Z1QzBbRcrM>3}QRN#H2uxki)FDkqug#@LC}ktv1Hp%>S*xJkJ zN2Y0r*lW=m%+tl>v8O$wBGfikRKef8x8^XF?L)3Y9U>{l=S7)B<9)q8M+w?LT}{0|Pm`4XsB6VN_uX8a*kf1Sl-1%V^d<3ScB^FJ zRd!J=YMCwK22kMN`uK-g_5WlR@IRULe^LliY!k9<~5UZUBl9;!&lPZ6UzHv(T26Cmn~leb@*hk{NT8r|x^w=ek?-)zVWv;a_by-@p8Dhj*b4Y|1K0`*n1Eeu(qj zhzxrSe0!3Wc}@Rz+tD_@yVzM1Isl*l-m7>VXG|b4(Xahl#!ohI@AQPij56wR;IHv; zm1&NO0&GR_C17G~8A8D5v>5t*i;q5RxPBmAK0xdSS2(?iUurnj#f_p!!w>4nFm|ui zB_*yXLCa@1%7&)n_>x--V^v?ueU2CD%QT4PQ>vtnx}!OSG{S{oieFx*IQydFU&Y7# zlA;P`ud3@9u@pru#B7h$Q?uuT9q$sYrKvfdacAN*4y|dSg;g;DOB6R2B<#T?8)gekh(WZ}kTm zm;F$oum_IH#~!~cpdl|bowdYtL@HV(yEHb^o*4et=jw&@O7w?-K}+(|U)jXls`p2h z(e#SU+LBny?b>F7XDs=x?T+uahY6Gk;9Pif%EqrGjZ#s473NM-?`Yh7@}FrCB&&2Q zC=l3}fB&epkSs8^+3;E2z0@|fT3E+mg1HUVpt?SDrQg>LJDVn%ve=LEA5^?LoA%t?lb%nD#%RzA%;G>-qNVTmQ>`v?qp9VXmfi)_ zY6TI*&$rj^EjU2U{Faj5VLWhLnHIqVe0+TlG#zOmm+A|L1>5+vN3`MO=X6SpEKc0k z5*=$_6CX`D!`P!6%>;X%_^U2yNi3Nie^VTie|kTPb#SoQ(-5{ehFow)Mk)(#^X2Lv zUHCD25j-QzU^H+F4JEz+)M%X6Z*HxciL;~c2MF!OEM=5v2#xD#?m^b9_=$lvMVEZh92+$QZuV*Z zUVD=K<<$=T&JSoE(c8-p*hn9ku9U6CI!1Tx*N<@wkmDX&S|VkmTnr zx4JGq79D>f0uc;3|6JYwRg3**%!LcZZ3!&GMj7O&gYfw)7sJS9HW$OG)dZI!=uMoK zBAFd4xuZbtmE5*?b3`0r97@EtvHN*~DRDwYtz5yv7koz1;&Cm@+Befr$6{|$wk{{& zPD_%tM~`AGLU35T)nrZ9EG``_wPz_Q=Nis@X?W&U4vKeeZ)GQIoimkXoV_Gj%d~Iy zElYJyc(<S5Ju!BpavZ|&zZ;M%?ZCS-*lIxgKeU6G5V+MU`F}H5TCDG125WO3 zukkN~P0tk?pI0ov`%d9q)mS+ZEs7K36AlslT%sQ@p2vEn9aW}j=5wq@mC)johD);Q zwCd@UZt~0DO{s2Q(N<$4VnQanvu_|Nj&5-bwN?q%XSxiiFAEC&5jf%b$}g`*JL7pD zMUa~{ti=V0Sk33wiB7+ZR(`jp-Cm{mE@8A!?8BM!Y`cirsJpX&^KMT`^w)W}fM44S zo#A@kVjZmqD?@>t5A*IeOwEnO(mmGg^#8aqMsttIC2!|2^!=MFgY9OJd)9^_`gxZ# zYqNG2ufH*4C*?P`d+G7q=ip!|CVmR~325;BYu@p6M374Bu@efK8r>D*42{>Gl{QNB?D8gKI4?UIxx=tJ{V3x6;=U?~lH3eZ5RG~}o4lYJ zmz*r%*uDYvFyF3vxhvxj*c1!@e|EW1dcyj;u+Ce?sI>_&_5+~|;&hqQN zW)9f1VRra)iEfyJ#^sI%dQ&|ZKS@~UHc&P3r+s6POWjtiz(-NW+-j=UtRd+AiGH?Y zX;z3I?r(9&;MMz3^I+$j{)u?uq}r?=4x=Y4kUrheP{gsF+(PTj~e8O$fC< zlFIR!)Tpwv`ynwsp+fH0;-duHZjAOmpsT*;v1Bnj@zQ`P^+C4JV#eL?23s#MPe4R< z;k-5G?S5-{_BrhC<8Rkpmd2O*81Ck^Yra1uvF9H33jMpJnX z^V?Lw|5lFGbDs!+-(~abx%`1Y_Pz5yd*Bzv!B6s?)Ak`W(a-Adhm6mu6-L}STlr%1 zn*?3(NL%tNeMD~((X^kLnd7*aR*$iX8HJ<`Lrm||jj89(MUZr;_|_nSB}K3W{uy0dXErSr}5e1G9HoYto^f9!ST!A3h%Y50w*0 zuc#sPd9_%}ZvOGUTiv~mq#Il1WCVsEWkj4c*qF9F)sG*(uYrb?j_w)tk4znmTqC5_ zrj^{8MNmJPf6|~YbU(gP^$D1c{QbhnLn+1I#SaUgsy{w7;*x8RkocA^J2?S5I$AwkXtTwjlv*qf6|S3v3@Jf_A>{tJ2m9cO=-zT@te$hu0L0A z=k>EX0m9Ktsozg+xtF!Rz3>%D%$i($)dglX<=*Y-+2QTZmR)Uct#fK0?Qg`f``2C( zYMF`UNS!&uSV~r&2F19XYi3j+fBUB;0OB zhCkgDY+z^HbxN1!Z{m~6mhH_7u{Du!i2RZdj-B2pP-XOfXbS!sTV6o?A(qxQo;9xA zOoHJlLoubnOje00kIj~Osa`(}vFCmpCkC#?JIG~3(m>?wT- zdQ(tOxdu*e`xvLi*wk9KgN)~0HqUH7cTja#-JsR{-bJl3{M~D&^GgDK2m}PMs=lB`Pki>>u)I{ z?-oZJEU0>a?X{m2`KqzNy5=-X$Z|X$I!mriOZckesN;gaV)2{ZG*0VyKwRLLuXM5a z`=$Bk_-SlcOsp zdP^S!t-sHyo~NGx6M=>{8knA)zi(cdbnh_f6Mftf3J%W5hs;%UyiH>>c^3MJ|NL%q zl~wJ<@vQIni<8AEcJJ|H=(S5t@mvKHFi#%T$Xy<#*JH$dWTa2b+s2u6kQ+YG4Ref{ z?srtJ+&xm`E=?{_J*V({Gks0Wv%EgS9azA!e<5)sc}+BkZ5XM6c3Go#cm@r zi6N4knOKoNnix+sf{T<_+4fhynwS(xeH^r=nEv{BJkGhEn?aO|iLIDE7-kb=h~=R! z@f%F|p$|jgzcgwky@MjLrA4(B){ZM6mK0zjAE-f9I zN;lr%hrC0onXE*v$jAj%x<^S%HaS{q@*}gbv-h!~c?=Ju+0~q zNb&!bSHdnS{|G5}uNCE$kjDHd>k$nRC}Q@p08&O1A(*%c6khn$igXiNkmQI*28r?q#j1XUqU%? zTPiLLLepiU-G=Cvfttx0k5CY8rooSN!BIV-hfstV;?n2)$jw%`!c^$|Uol|23t|qGD2&q#wN~dX{Jh%is^!()P|-HqlDJ~E3Z_oXCs=o z)^j{Kwk-cMcY=VizYUQHpp*OGhM2V69mh-rF=&mrT9Bo=&V)UxNPc!M&RF3{e9m}wfwjg&?JD67zLAe zri(qqyjmRPrqCKO z=kidM!MxwQ67@|f$~9UTxsC9m-;}2mxrffx5@&sp5AMGH`Q7GC08ZWA<8?Qryb7PH z+(|}hc+4ieU`)T}w55P50sPeyT5R7-AUCdLIl#mgNG$!<*x}g2Cr9}hvdm`Gr<%!O zzc1wd(1Fb-eKDHk!;xMzyM21dK8&xE|G`7Db;5FGd0jbZq$C)B zNadr&m!)(a7t>FkULUI9im}FNtXvv3W=YR|dt+CO&BaM<3St4;QUuv8ABXfPQu7@1 z?3dofmI9R2GkQ$Vs?180bxFC^n5E5K6j{I{`sGFLvFw#x2e_lEN3WtZgB&|_AU1&c+0n5*T+?-Gu7Ws@n^jfyD!6d(swvKJKi@j;l1yT&!4jt1px}m(Q0!Tv&s!QQigm?gZicWP&hD5#iSvm*>MyRq zdu49V{iVWdLXTq@I+iG}9Ejt%E-J7{JGR4!`oPAMo0(SelBckf9D)per<{}wJXMP& zpoVip3mJJ96D_^Cc#F@j*{>BKZ>YEPBP494-mD&vVGX7a4}0{@HTBI3=@v7mK5f>NxeZ8BQdT zP9U%Ia_FO`VLT`EcwF)!hg56cTZnd9J4FGXAW!3l&3HVw`sIU>C(FYTE&GAkG9w$! zIPI(kgs?FeapcWMUpt;wzsQmHb#Z?`;#2iO!CiMq{$P0^_Bvdoc}~l~O-WXKleGDF zahgG_dGH>*2F^8=fW*h-X=UxdioDMtJoin;Zgx0)E4!`a$EH$Oo|q4I1?6r**FB%v z_PcLDB7ppvm+mzwe4Wb2o7sMzJ5#Uj^y|zL{}??SNkC-%8aN$&A1~{^-?oK0kBpVQ zeZ3(3V|$vcGPOQ!!o|amv;*1U4Bwlt+B&|Kz7$bw)@oW=Yk%}GDh>8#?H%&5;-pH3 z*=wvy2A?`^^>QPsvdO7(HqMy0knWhJybSaR)GMkbvGBWLJb zKX(cKBHNnn4xv!9_{R7>zQX`tgmI^3QT~D70jl$N-}{__Oa`3iQ@Oph#ymZH`b@dT zt%{a?SdgbHYOdo%ca-rr*FJ;?agO&TQ#NTjImwFF!BIn`p~^7Exy#;=m>}s!a-3UNgk!GS(pH7m zL;vANdNyaV8&EAuQqyFghO_vhL}B5=qF|nFgW}|$-U4N5<@8SF5~@9roZRwFN6mmR z((}=X{RQ^4BCHnMql&xCb;CV(~i+gtU>nE|A<*)cJNOxbq^6^LbrLx-ZHiW>> zMA7k~eB4bK$7;9o&+Z~@`W~dz)c>65b6%Cy?f=t6U-(eqzo%;xOWiNJiG$dw8%q9~ zuHDe6UjyPquG;?TddR=pe&Szi&;TO!dssnq#dsbpWgShDw53#`S)%#D?=f@|#hOUT z)&3MURm=E&uKi?Zsvw#Bvl`?)Q3TWC!DiC=wEAU^j1Mth8m9S`LNk8A;5F>=SPtDI z%Ur*MzPIsK_R;PaC)KrUq2?s6>F4Wp%Y%hFY4jcA4O^o*M(Py9r`PRd<8)?ePvgDE zhbs)K=kvZC#L^`Mc86G`*{+ARJ^5uB^lPQoB+jBk>d6k(?tpj#Zw(^b-Z%Mue=Y8C zYNS4%sU>L6s}UphHEK95z98}ES^RZ(Fk0xrb|Xv7Tzwf^Ac4u@JoG8VUe=dUKJB1YHqiUvynodD)c4Vy;(ggp+)#LYI7qD%iXos z%DOL44>G1foNrJT-|8GpGmRPMpTGxoEYQj;4Ri5_Q{QTXdhcOR-+dsXlh73m@+=oz zcxQpTEJ^t8+pjZZ7_5v%lMX1p#jIY{g9NpC>2^dq$w zXOUU5_2CQHQ`sNirpR%vE`iK`HtPa{T1>imYls=j$8*yc%jwgj+27$f&Vpna=a_Et z0;lUC7ZOI>awSqf_A1#F1U$f8E3gtV+^LTXOMP=uf z8@U+Cs3XIo5;X}Q332B#;f{i6rY=A$C=N%x96ElCayqsyfddcIoOl(dbfU)}Nt~}M zu{`$kT|u7k8QU2_uf_5vX&1x3nA%B2q(PT0CRfu4r&sCJYM;hBq&(OcSWifBfTy}y zg$)~JBl1&u6pfmNHgdhwH%b$zip)0hDMNcU3YgBElR`tJV?j!0caI&5q|ZK$K@&BF zkcGE}hMk-o<+f-Y^iAw@@-uuGEGkQ7;9tre&0{N+;gsVpmSsw~uPM*$SiMUh<)6`3 zWX2C>r|7X2ZzJ{1Rgx6x%|o_Dj7yd?3+l)MSG5?Kt%0y6X8VxH*P}ltT}&@w<)TW= z5&GiIT-G#7^g&NWb{hQyecak7%U#{Uou=FSx>;VNwMyn^yoAl`A0N22{q~v^jbnAk zjH6lXcM@OEM1@2)lT;Hl`~K8}}&h>@{)QgL>bZImI93 z3G?2}4oVh-aMZ8aE{(lqDhP1;=-T_H^3mjscH;nxO5C9lXO5%!7?HzmdXulZvWd0|&6hkSAF5~__>!I}GRD|o_36AUPGM2w(9iQNb zz`riz$CgoZD~27D&bz`6xccMpNJw7@df2MRGUFJq0~3p?kyJ$B7*pDkfmBJ0!9!X8 zIK~`SnmGEQ9~lZ(W|G!`6g`1~)x~rE!-^YP#N>LhniMPXh3Y8^7Y@Y2-j(-hE+{Se zHeJahDwIlCpJsn9k`Q99v){abQ+$Hl)c7`grNibJG3+||0Bs+tuZ3$>ZTttFmTdhc7ydSIkof+Mo_4yYM4Ime4ByCo8F3(2`o6;Dr`xS_EbFYCEMY{ zQMy}Xvn-3Hmf>&ps^7CbZyeoVrjXsp?XZ5j8Bp)4;1~GDH-fL9DUcp_Ex>nq&Z0ax zsH-DjnQy58r;f`QS`~Q7O9ky#2t3A~2`mP4QwUmXVY%%WLda<4*nPp2rUntThAIp4 z;F*;LKWb8pbvVDgA7><+Im|YA)+}Q&7RS|!3_X^YvkAfD-<%7S{#ePDAj^-oR90xO zD0!(OQ);F9_X_lHk>mizfxCZ;q>3~!1cd#`)UffzzW=$#S{VrDxIV8XXJ|c){FB{c zX&|t45-UrDfWG5fE6FF@5wX|Hd-AC-D5!nK;s%e-*!q5h&3_w=`~#q|l|1 zLT;n0;c+@*pGt|nZKrFUK1bPae|Gu?Lb#edCza)fr{Jq;!bdY!EaJ7eqpo=zKkDma z480_~<7Xq%h9pr|sj2f^K&MjK-K~Z@&-F0ib56p1c_u&mFkN=OH@|5;R$?WQ-`VOu zI8!`e4mpqa-B}Q_Sf0N~_u1Sn;^B^StaKpR)T^q{7E21BWBDZrhz*Ck znbYAgF!~`;meR9E$?DSICX>qcUetNlbC$0kIT8Nro{`{1MDtUD*${%Lqgj(~%>9M% z6NmAf`}iJH3z4JTe!0P{?oQUx9OA2%NHn8_4aQ4#~cK&86Z6 zUi+}4mX_L$WjzD?5LK{Q9Y%}!&_;^8SfYWcBwB}$Z=*y;x@|MWkFxMlqJkPuDpf+^ ztGQrgTv#^a%MVgnh0iXrM#i-|c$N7o`*~mN%G1t&34m>2Tu6DBZ4{$ltNgS`F7CPP zC)-}|n^$_bnl&>lEmV^t^$j9e`pOKAUH2Ro=7hfa8@k-sE=dNj#WS*!16E*SCu&07 z?$z6mt*+&>eDPXtm~!z$$<%OJV$MqHds(@9iq-PO*x6H#hq7`}BYeju)v3;A(>C0r zB0paz%|UX!>fH!fKT>UF)n{El5im9a&E#gihwI!n8z&=@bDt1-lKQ~zp0c{sXUVRk zJ*6tta~fM%d`#1j4p@F)37&WjHN!6S;rLr%B)j2^$CTU2oL}SW;mc>dxS2SyV2NF& zg$M~mlq=Q6FMT>!DQBM{=&WHU8=m*JCO;l$Kw(wv!Af?ZB$2FWSL}% zEvw1g&qq1C7YT0WTDBQW0{a0(^oxw&^8>0M&u1KD%n}>?>OdVXoz8?J~^FQ$D?AIl}SM~u5y>CI(k|9Kyj2!ijy!3-( z6k>=XY>;QPc2|hG4Be2}j2{6$vmzG({i0AH@mmdPoy1~M!lPM_L>c0@3_&+SxrKhJ zkUswbQc5aeFvvjmw|z2KtfV z_z*zEkRFn0GLzyAeHP9#-}C-{Yi}l7=rS#%BmqCf!;|V2+jRkRa9q@$z$TtEj3rt; zp^ywBR;ItD*=kt+RHsbX(@AGfqa*M9umVb;${L`;0(xNTH zJx1z(Kl1sXa!f8AVoe)*<*V!-$p0Of`!^Q5udCw|u@|5lDG;tpFip2KmZ`48YC$Sh^|3-)oi4?nb{i29q>bSk;<9#Yl~Gd zq)VSCiDvFP;R`Bes5K2`$lF|Ximwp?a$F{jUB+SXH0Iy@Frt_;1jmZ|XNv+_)(=WSoK7`Gk96D7JL>MS#zEf1$SUkhe& z=AAhE;}t5*8G(FSp38)!626XQ(Kk=`ld}zA%xF=UL+4B?ghqw5_3oTjY+)1hDqjbTOL*@1ab8C zm}*-(Fg>H}Be2oJ?_t6tpdA|EWt?VZPu7wc-fCxT5 zZD+mDm~hTBAaQYJIK&2rHD0QxiudFC-8W?$RrAFT$ebQt?@_VQV9AzvB7UZHSi(&w zra7qh=uoTtzdMN}L`BmN`No?#MJPVm%JDTzMY22^gYLs*MS7ZCieziIcafW%UhVpW zY<;$=QrZb=&D2($kJk+5lU%>hZAGgQ7@F3|re_!>6``{nNYnEtG@goKuA2u<5M}N4 z2(;gK1cdEH!)dp6Cc<8EQej^X_*zOJP!?^;M`L6J{>P)9kKcTrmu0v-T~&_=Jlm|O zy*%G-d5lrG%q_@Z+n-wSVL8r_=vH|Ox^Ic$&l;}ydH>!Q5R%7HPRpF9zXwf@44F6L zATF%yd8mwFyI&C{sB$sj_etdn>dOQPN5ge-I_llAa;6 z>UY&eeWMx;lfKGKZl42ZmS3DMq&GPZpb_ypHA#lYzJrsPh%kpN9$f-hi6m)UI7f}$ z@ocfWa2IL(jgIK?T!EqHi`96SFD0jVf?^A;^)Fr?a&S9r()GtaMnpA|TuPszvn2OK zBcIzEy<$(q=sxuu; zU;$zlnce6_zR|Rt4qm4hD8loEk(t&-5U#;hLHywUvek)uBu+s7;PuI`%P3in+-r9R zK7RYHo5*h7&=+8wDarMz=c#J2xk}MhLtDDRsf-$!%JJZs1x51-I3qvB8trcZ@7qfYQh`o0y>yhOT-Dj7% zo9j8c$G@gxFwf4`41FV~?Nl!BR$3G=-cx_5cv`f@ovkB>%(T<%A_+_7h9VUMy|;zR z4acV2)?8bh)mISt$4&P=S|odI@O6ld*R#~MuNrf%5L{BIg803&&DgtNv+SO~S#D!Z z==s{T6SnOzB~bX(97ApY-MDwjuRC*iDGD!%C=)F5DLYc4vyyAACd~PEfx)z*lYk28 zdKE&YEZ_O9MKy;_2u;z%t?}x5N8~_hSB$3!c$)^Y->UEu*En}|Np19w+xG`O}a)K zYaAM0d5@=k5y97&c4ek&iSdifE?!m43BJ(_xd1vN5*SWrP z{`Zf)d*8LHR@JI?t1)Etx3aGA<8OcI`nu|Fwj|U@y}9Z;Z8{PTK|~ERm?LB!8hc}V z?|(IFO-j2IRTlPC+wp?UNza_vl=eo^Zp%8u`N0O0Xg@Aj4Rj0?-lA}=yTsRUey(PG zZ2j!@y8H&oM!M!oi3~Y;ehc34`6OlxtfhvQvv)xyfT@e*T1wMr<@|Vz+w1O3eZ&3! zYRd;ic$2gt{2-Qs4?yPieTzbii1--I-YC8u`0N-8azZ&apanzcA%4ORF+##lN!s<; z{?9D1u51;5lwBaROE_vp3>W0*&C3`+H-Jsuf||S)0y(n>A@c^v#F@*$C-H_X<3L9t zvJ24{oG+jlOULl^{UiQ09IFa?I9+(j%Sh_+*j5^4ANF|b9vl3w& zosGU9BAMmO5af6^z;Alx*EF#ioxp{1%lIT!zqcZwtUerEX5RBW2$x~gA{f2)DqJ9W z1=H4zP9gI&_<3>%lARj@zB5z&taD_tpgZwZ^bQ{84V}wibLeGGp`azqm~%1!mGKcS z=ue;n`~W(^|EglgM=|&>K<9VIeLN|Y^vHi+S?bfKSR$Ik><^$b>+->bqTxUDWmJ6s z1LzKR>=hQu zRj~feZf*54{RSHZ*L#`iowg=FBqBZQWSbc%6fCppzMM5z8G|gpsH5~)vn*M_YE$;v znmIJ&AjNyHw%RH+D_H!k9O=f|h`rU`Qzm?#xrKJ_r_Nq{^$za@yRmA#n)x~l5Gsar z-d^lGZcTr(D(7onULGSdv3;PFZ*9Il-5xuV{T|)oQS5}w4(8Q2h=aR4SX(M@%WAm< z_)Wa>#l!qK;k-y-q9w9`%0(_x3o=*qM6DsJ z$5_(()yD@jzZ_qPKnO9<(E<0kDJnnQOBm^}?=mX#O97OOX@XcmS-a&U1d>>XK!rtE@@*Xr+#m}%n$)PPDRiKavA*ungK z9PP5Y?pKeXOA{YLQ`IxwO;UZwPu0U7ok-0o`sRgp*{hjaOyER=@lDfF7vd08A8IVH zX*7VG1@g%DH)=xmehsrb&c5TQ7$6hN#t9_}UT{GP3XVw>t<_T`lp>3w8zZuDuY+@Oq4BQycZ$~Hqz!~DLq zxFM=XDh11m?}cxv6nyD%dSmpJR2PJzdptohkHP05=qGhzAB( zHKZse*n6-Yy%ahfpk!6HaBZ?>T5N5y+JjHGN=}DfT#`)ll|>MIPYk znd*&JIyf+id6TEIX7xktNZbr6D%X0&v0uNPF&11bL7uQ&)Om9zHkOA?!h~|;Xqg67 z^Q@Hxo$tkoGkS`f-(K`014-Du_DMJWhcvbQVN;E6GC%}dOjj&U_m1JR!luCienY#1 zN0pEVJFJ&L!o*r?BOL&QPoGhT9$_zr&Dvag2dD^&oweZ^*u4 z05pwn5SXB8mOr_sDYXzDGuKEEQq6{;uZBW+5I#a)YuS_qb6~D~6uf!X%;Wt|~W*T=<$c z>1GBYB|$vXu#*&$=N>F72wVM-M&^h7Ncdvf(Kx|X%c3XM+5^=jN~w%)mC{DTJ&3gK zZbdxRLby8nz0YJ~rit^t5ppJEUTKwFwei?bTUWR-C+`l&!w=PVK5h+Crg04{5P-#8 zZvdi^o)JbYj9OYGRI^=&le#~f%_(>cyb9-sq#M$dA|%^)l716g->Arif5s%-=5_1` zwu^>OAys8-peq~)@)ehw1)3|cusyyB&#|fQFcyirk)jTwYsEPvZ5@=cEBLgi9UVf_ zEmh{982NnE9U())&<~$LyQCd)qY)<2e?3B_q3edtT>Ak-=xIdgkNAI0W-C3U(URlDTyKd(3FwHuOS3|VOk2zSs*}`Or`66UraCxdHS#{CUaaq!z9&$usk=DTVrI>C>Rp@qz zX|ej_YtMk2?-X&>*2p1=UvJOaJ&Um=36n26Ru&~81GL&2_0W-EY%Ya&%^qDjZj>nT zUder>t&`q_4{}noOQm!HDNpYAd75uS9Q_w_ztvcc#V#-H2_*ZREZqZe11Qzn2+Tpw za7Aa7l2RB?K^Cau$s>NaYK$YV4wyu<11G7kCe%qC&B-lkw1`y{A7@BX$xtV5uZBlP zN@SS;hjILhn(>^W>zMr4-{a`a*C*!He4? zaWXVWi*Y(w>$%CsdZ^!`pQ{C!iu&KznR=?wFwu!>;997PU0ttY>wBeK55M>A=_B=1 zo=u_kbG*(K!v~9IL&c%uBzlUDHF1bkN%XiMMWBnP`56Js{}}TWQl;<82)wZpPSj#yhq!`2IeJN?l#g?+<+RY(5BP)n=ULN>`=rVqUMbt;wc?1_TsWn zg{C;diRIrtztvP$8q4F(!^g>%FIdWq#EChhLB!f56)l0V?NB(vd{bYsV$5!>lDw2i zJ^)pd-(HkJ$To>X1v@Hnfz0E$@uHlM7pL<0#ffO}gw)oS37zk?%!X^*r`Lu>FUI#- zGM-1xor<(dDmoI91js%iq(2$})x7Q|r5JqrDCVH7X|-9Sq6z5w{q^Z|zz z2tLcGu9h*4tS!cTjdW73*w$=QC+(kKIr&_j%MqyHyN}-I%hGaIGTsWFE0b#0wZm0p zr6nl2t$N2U;>bbjh}e4etz-$o^qOcRIGBN`4j%4_FF(|tr8(~9+}9>C$n_KPX^+%& zA;O%vzT3$oIa(r_c3SGldAv?3A8y?gclPr47d-&0pt-V=yWPg)Q%O<=3DJgOBl08pSPa@1^b4jKc zvq?YFQUU%Pa!~=Q08f8m%EMo?F&vomADPcc2#SDYS*>wQ@e5Oox9#%!py={HGM^G% z4*>ES*Nbm-O)w44e$`8I%&;#}(4SB?<-|ZfS|X5HcNh2LSS0-cM~3t{ZI6$}C;Ei? z<0OG{z2Y$P0fkNRQcaPYF>*2tlAgPJiEHnO(ug2$4?#816Y!?FGu}hxx?S+f$O1apHS%Spoxv zLdIBd)StH+D)Mb@3X>oQD?o1(I5>KLfv!YnWUN;i22)Waw?m2h$|5UUnqKD*N-;C; z&b!YmhYwj4`A#u3y>4+Ik^8Tn3dTfnkv3Grp!biFh4J6qE6tb$E(B0omr0$)I%o;? zPS~7?t>{1kp;R4yR7j3`E_(#rng9*nOBsCFnV%+#lKr-k7*%YKDgga!e#aYpt&7!n z7$4N*BOdxxao^dxCka7X@z=povN>F1Vl;0NC(Wj~Elf1!3z6od4Rv?uR4=`mJ_?y> zR_DD!yyo3Zr4Q_*PqFLyId^{7V4Q(d7`l~-1J}Wjv1hR{EN}t38P4*D6>jH5h>sg6 zP;)WwJf{3N3GZ@w7%ryK@1%oygo!Uzl`9_oTf$5Gvov~A%@h6c#{m&V8|OR88-}u( zppiF@eh>&C5~8Q;uV(7W<+7mQM|)B*R;&T0Ki?Wyf|RRpklabsT2Fq`aDRo#t|}9c zp~w|Lpj~??U+re1*SyVw4#c1W_@nF>FR6z(`^?Z4*BdN;VaGN@fkGXZ2MilGz3S^Z zHM62c^6$brj`Z6L#Lfn`-8qDc@eRXUXKTOj4ETQvNKP9u+iQGhGE9MWk`&(tuD|*0 zOJ@&!Y#r{ZhH}-sbdqtjP3JjY>cROQ71{zqg0Z~bWC9ZRX2*HAahJY6r zK&@~Qw1yR_Da8q6`SMH zz0@3+jGUVIsBx1%V^L2I_hCre-2>cS;=6<*C3qx?WGWPeXoe9;9l%qD)HdGXOL1+c z357^HR`nFW%lP+BFhlxrXN?_gT7r8K4Pu8-?cKL9@^LMzL-YglfpN1!)9F%?_SWMb z^odM|gO<9xA~wnTiyqcgt_8<|hEqMOQn2ejdR){4+Ej^HUGH8ey6v4~FzNT!W#K$D zqVx4OtU%`ha)x<>PV|m7U-2qMCHqXK*!B`; zP8f}L;t}1?7>-^{O4n@AiK!&Uz$JYVxx=o6(?+$Yr~4kXKS8OPaPCfx3@WKHnI1)} zuuZq_Q8QLQpEVk{IST^YMLhKyAf1_PVgCj?xo`#B)jNFoQP8S7NtD*k)~_+1bm8U1 zDI;2J?(h#ageUvFEQ})yP5tK<=6?z$ihmEKwlG{D!(X8kO!QOZgS*}y&iI_J}5^^ytHuJ27Q=w&wKhNMU(>YBmx+S~=z+Idq3(pPcdb z$lZoj!8}n2gLI*_d*!W`kWbEg!$d*DRKl+T&~A0*^f2W|qO8YJX+-pT>B;x25t1#8 z9Ip>|+1Y|Hh=n64u;4q(%Z*i8$}M}u_VKC=+ZZ{)b%(9sd+)`edsl(U5T>dJwbqjy z@tNxEwvd-E4G;IG!4$wD@~&eL=m{;h{7dUT&ssrbUF+c%gC0_l;YYBG4f(I8T8xkx z3}CiohI(P9D&Tc>R&9PEtAFs2}DU*8#M5P6i z%Bfpm(VR(UtS-kCQ<2=sD1SJA?HtJs?WL9zkF~2;(+1zY?S$dL!o>VreDh2$Af1y9On*NA?Z1|AB_}h;X>mTM*&e7|_$nanc6^m#Sp(L80>!B67 zLTN}2Bq$2LS$+POzTe2C&(?->5qt>eXO5SI^2KtcBm23ihl&v#Uj3uXfNfNI?UX(F zvuEbJJN%v>`Y=Qv&{(@%-c+|tuR4%q`kdAm1Sc(8%QTc#Yn|etycyY^_4-!2#~c?_ zF_6rP9BmcN+9zPEmkyGW-52i^$koi*? z`?`9E9uFQ+HTk)z~(tjDt2N@ty4jAjd)FFf9&|b$P0f~J_I8k2Sev|K( zGr2<_d>|_&dRIhKTAlxfAmC(_eXL2L8yw&ZoqXv^9@I5Y5a~1_bLxIFjr}I*q8}4A zxMS$W<>NY(d%^0b1Beb03kx3T59>u=tln^apLB4=bRwkA!{A=x*xP%u?8<3u+QS`n zkdFVtY&oin*kI)iP@}mIH1Rl&;!_EIrHx0A3F*4ow~FUAjY}Cfxnjf464YG7R0c6_ z)67p6FN0O%N*y<#oa^EE@6Di9kj&R`jSJnRODx}_St7>ADb0o*qDjuuk*0}_sbELV z-v??p6a?P!U!Z**yxK-^&V<8H6?wSs!Y*yT8SkYbUoMz@x#qU-3wy462VYV1Rji$; zWqI~0grKIEV8GP5Bdz(dq=0_JiaQLT3#ip&1!yd=dz(iiSg3 z$5umg2hV>4f1?$dmfxW8d-K&jdbl}5l=LI0>1Xz1Lz|W z*07A*@eWaSX%h+QYZ!e>{!F+^JKLqMM09EF3`a>nZrc^{#Pba2t? z@RLe3GWRI74P;~m%4wF-9mC4Y?x3sg+y3`7r5m5u}6Tp`xl6q9&-K9|gN zw_5t?Sa%YMEm5C~(2Wc}&6$nZU^EZiEG22D8FyD^Ep0lk%@M#iIZ+`o4;~p1uz@ceO1%*X9@u)8n=emgtM=>Zc#D2A0&Ofg)CeB+E2g zjVrByFpNMr!UbkcR^5(72E5$Uo7|4l)+^j@NXvL?nb1fl%ZbLskc@K6oDme#Gk}X> z*i4|U6DNShvlj(X7*E|VpTuNv)C3#jaJIH=he3g!Fh+&;2j8nIO{UA-3P=l88q#|< zzxnlHZU^S>9@>HQz%M|#7mvqOuDO&1v_!Mv5UI~rWqKWgE=j({5^AaCTZ_y^XQ)CV z&vhR)wBVP#N!yqXq26NCU4if=X5f63ojbT(31KL^T8-d2bdxBZ#=TyTRfk<|cvD0Q zDR{qriBnI7_%Ep*4p0sd_z^-K{~^_5!ovTr1`0vx1^=rK;6M3gx~8oD_RHithySNr zsP*WgU`NDHw@~*->;1``R*Qqzdq0iG{0Ye$KkbYOdmn~yj9qL9QvvB z8vkuywo?A?=UXv-qx>Il)pz{pzl}*qzEl3cdMn*X4S4u}-8LTiYuorgJRX^w-QR!u zYheUOA>z!lO_?S|l5 z%Ba9JRRvq67q`ABT>cWhcacHHUA55;>wUH9+3?1oKb}9l742+tuJ3a7BgNNeEVK2( zE1ef}jTb62X&bPbuXBvbbxsaEW%EsL!}AX0r23gk%->fAb6t&78+JAe>^@l45jRum z2Kd&6KR}4%LM_6aS+E~ow(@_jJX|})fV8#V-(H>WEwu@tJUk$98G;M}gaz|rGvMAt zm5EVxazC`U;_1e_?nyY_I8P*&ZQ*T_v(>P$j;VwmQ3R|6EPI>Jguon_>otnLdkZ&7i zuu@wLq*|F4Zlaqu$4ptbeRY`>^rmUVQ~AOF7!qT(u6Iz8{TQ|jr(z0sRO;qOx6$ja z9+K(Pg8P|>?7d4pv~}Bb+=WRI0Jq^x)~(CYI?&A`7FC6X)t=PLyKnuFp=Xx+kzOlk zsvifG_}bB9VOInEy0s^{B0e7zwT!r3JSP%l5`{}iHB&uHl4pdsY*pLbw0h9n?hJ!S zZQH+;MviCKLX_k)&yn-k;!ng&K+X(votmny_Juq-A(eJdUrHLJqtS)ze9;a3;Mo0U zG_>j!cJ1x=K`t4UoB{6NGfAC^nwPGLfv7^^li0yY(uhu@5(y_Jy-x|B0iZQcyyKJ9 zJU6SyA*x;69^|Gr2`c>0b_M{gU5{PV^>1^9@a?QdR%g`=y)9;2jiDWj8Ug#hNI=az zn@0;OEh?`GNd$i?Vc3V<2w~B(SBe(?6?JRA{s16QLgpo(%&Xypr@%z*R@XJEvxDm` zW>!{}MJnHDApr|BkoZ2>fP;6B--cfS67DbPiQfv4Z(J=tEmP8{cpWW7h;y}uPM8QO zyHjDaJa(If@!Gnh^y!z!>>&hZxe|gU-v@5)gzLwT!w2HP&;Cdi*t;4iuRx=Rm$ZR| zfgYK_s*$!u@;n9Y=uScqLlyk95q?BV(s4;HP)^_>lzb-n3;T2oy5lhi%dLU@avca3 zdThpU7J`Z-Tuq489*B5ifJTH1`qLtf=m+Zl+}N1>kGq|J0>FPoT_$$aANa&DJMbHy zz%caxFvjKHC@=aOpJ>bfpIwT55zy(G6PHIiAg*X7A=6vvoRN<{v)&0i4u(arE_szS zm~0L~n9yHfXl-nLgYf0*EvRx?jls({zuYM|*qEuWSsZF%r}|TW0S6cYDExP+(WaX;mE}7M=us7RF zp>iOK7xfd6=5S7{GQR6TvwpT7EWwfnC%tT2#`n>8*MXs=k#Ylzw_6EYL}r5#E@IP& zJ2UP1@b?LC+Dm!j^`o61X*wQnX$*gVscb5kwg6ddwtqw1cv%SX;dt_{GJp{sj0o*U)q&1lRde+Z9Doay1N}uczH43NwDpu#%tK*Yojafr7@fi{3G{ zmq4cVXi-9o^%zMkkZ7zdR=cRDoH&z7R354rb+j_Y)i*Wm*$|UNgCOvH66<&K#AK z&k29Bs~hRJi)xV_+aENY4h3q(?6|t?k?a(F+>pS@dXt#KSTwv4ZLwQe0jx7C z_VdtUDg0#ASZvWkxV~4eJCUrc9gVw}q44=zzO^ZFt=nex@Qu`V%>+DhV)cy6-huFo z*z1G(^+c<~hV7tGvBpp5oryx0RpF`*U;B%vP;IA+%d12)DvtD=V8!58_+nMN4)t>u z=1x8G!DCfKfyb4}X!7%_Vlb_Adm*sZ{z*K!PlV$CwEy_OPsriR&;7XV7V`MAOb33} z3#`O%e#IlM?vXy}7T@!0|rZmU;31nvWwc9J+u#_&ivl%b6M?P&SP2`zh?hU@USm16rjZ{uL z^?t=~GWIo8OukL0<+*6Pt^?(R^(r*gjB*iSm1+*Gx zb#1=sWEI8rVDvznL1Q9pNG;yVpM^QiL##4KdH|;|qyB7&iaG0h*cH53l3T5|s!CnG z;Wq!z{m_nz&07VareL}i#K%z^*XZ&KF%V%D_yU zTWR`4wnm8WU5j=kVGhGnVGC!s+F9Z5gXMQ7aWZwPuUbc2E|+{+;#cRPzisEM2nFEuQ+pb&>&;0N!7$;CEY7A*;>r zca*QR`(Tm6XizxXyF!RLNigiAmph`!9_fusRvCYZCgsQ!MX59rN+OX+>|%Vg(gNVI z8}jP=MAze4cX)quzO~wwOREh#Ej-yQ)wY?hu3Y0->-!*c#6({3X5~E(!%#Et$!cII zQCq;%0I$$3(BHgBuOC7p@Beg$69wti*F35?VmCoFt#H0x=;GXXM#-#LpC zi0Yn7l7VMaQ4GO&(gikKHzABQ_0mPo(cvKAinWpIaLW+U-RPZ=Wn1W)botg$I~EO^ z)F6nFq<3EtBnFP;#VUyZGHXf!5vt+N$N>IU5dHX6zCS*d(jTAd7a%|U^q)5TtdI5R zNdV(NXsM1QhoD3x8c*1ZZM&?N7;1z(S-vNa1`?6=I|tpqQ%EQ9ni&OL?NJQtQ*yva zrMdDtbD!ZP;=2t*rtnzrqf|4gm?$PbiNjL)$KrU1w*!4am%e&%4qZ_~HYt&Td_abF zf63=+3nWuHG>5WzqPM#9Rc`opiY=aIFv~}El}kjXhbY`$yeESKrhs5)JizHZky$CB zuIGg9+(!rf`5WiGNS%rRCVd;vWA#lu8+d!z#%>q5s&aAiJ9*vB|vJt$IDr#rU<}f z-$mv|7uiZ(XlkuaDw}X^!;-_>?}D6LEJxNX4709v8^I1zt+|fN9i|LK&qy1dD4c&k zdi*u#bdT?Nzm^wx7;@kGA6uW8aD}LDiTg)WH*kxHm=B>8$Z=n!85XC<)U=za4Km z=#ui3izDF=K6e%o&*|~!UjqCXTXv!_6S?45ZFs2^2H%1N!&k^Vt*?V};4`q|{HWZl z7^B6xNw?F$anYZ!eKP_99wFrZ!1SX*)Y*tO@kcI(pIi7A z#1(O{O&S|!B(IjAMfA_GB9K54uq#fvN!brG5#kpg>pM@fab)928?UYv%B5jGRGX0Z z;&~9C{(@%t#CRT-bhs5``a!Y$B*WGi_>myjSUC$z{DTBo{S(8AQ?3brG1_{K0kO4=)Ocr4kYyS4`@F0|HUR-{Z0@OYs-Y|=g zt$D~pLtzK;J_&s$xeC17&D&*cUMr4XlOG zB#TGSPaCIPhK`UbxiU)Z3Dz@UMGQV};Vz@)vi6p6q)O*|~`T&k7Ig`GTHqMu?l8CDa>A8aONtVH4<4APFL=MkrWsM`E_S z9?=Ij(!`!G7M~lh8gpRg4s@-}1*0(p6z4yZP1_QAVSTr8v(_@r$dqz{lA>jnQkH7l zGHpN=1ll1z(^EP;p1>7o6{_t7$;i{?;Ln0Yufp0g`cann^4s+g%t>*Qeh@~-V2#2T zZ-`PVk9O>N(~~<8LR@K%v+?wyIA4YOCW|81#~e$y#>dzbG6UpQ3^n-3V-ZqUjCnbq z&mb{3fggm>7@yrz%7Uzt)tz9JZ18Pa+Dz0Dv(n=k`2{c`^DN1S1=mKIDV}fk(@)Cx z{+G@`uiOD~S0XX}u^V1nbyW9q&k8dyDHmkc5;vZU{IOM%god=jCaKytr)-&#BV=DH z3E>^-R_nqQ4~DtY&aEt?MUDjL_(K?*OR*FVM!GCrEC!WC%EX5@K1l`AbIoG)U7g5* zYhaZkxPL8^jNuf#cGjy`%Kt<%V8&(x{9=6;4tF3i*{JYELvIqEqgKE_hgn z3|C2s&b2P1)~8nN5zTbGxHP3@dIhFwOt3A+r`tt8ct~r_7QI>ZNppb|W-P)R;H`RF z+!!r)Z&{`q&DgdgKFt6>+0$F`rXu#0n{KD!xE=zsb8aVPRNi4vTK`Ai< zTmzpXLg@03oCc@;c!k{r01tzv9El6rKYBCb+4b>)OG4!X5#vHvx9B#L_6GLK8$DKN zb3Byj0oV+!If&1#FcK6p!=D6ntk)&DZY_?gn7qIHmCT@d`_k$s+HeqmVx9IgdVyI- z0FsxkcMu7sodN`CgOWcFC>@Al+6J~i`VRsit$gurTb;8&!l0k@mSl??G0B$+Jn#*E zF!Jv@Qyf$E$+6QB>G9gLMI*7GsyhVr5fDU)RS7kM1f3-9E$j5XTK-}We8$?zf(xy?%JeR z_<1oh5`YPD@pk{=qnr%X+uQ2$(>hCwvFLPV3+g$+W;-|L3}iOgIsw^I>C@E4|odd_tI7tC*uk32m(CaLPI<{$2v$8VY$=|6bNN-sYzz22;bLZLLS;$Mbj(heke z!)!29j!r`k@-##mS|2||m=8~@O;@vgfv!!{uvl5XLoRO;Z>7`nWrARNny2jz`*rx7 zSKcLPyN9xx^YW1(iB1hi{P{S=3KV&DlBTUr<6LQemkG?Des5A*v8`i-r_k1veen2h z>E%Kg;Xyc3Rqh^iLOGZyPe0aAsoXA-7O@cGN`pZcCPTtSTeE}-H6)eFh)+XU z)(GElZMKw@kERJtv2g*Uf0Ij)JuRg@s90UToGh|lq5Aev%1?zKUK@eAv{+JSi0(1b zH?5;YB-X6=TeMt3S4?@M3LD;4d*q6dR?ZsgWE7h;Lvk1eg!=*Ge^ z)GRtp6c+=;S2m1qP6WM^k>BR?K$@XeK7Zs=teYz6lC(Qnwj*DknqX6vbZp4>dhp>Wc?JsNBKmlgc z;zeD;EeU1$ISw{n$h6#5?a7q>=*9>Uk~<;IF&YxcQ8{;PZMC)#kP@E#+0qG!fzQdj z1+9Vx3-MP73Yce^sw&)yD#0kxiRtZva2Kq;{Kt-H4yUH+s4daK+p`=dmgj zUR+nm{32G1{8-+7=0E!`5L>Vr@hqiTD0h)i9I&=X3yJ~G*wlP-AIV{M2U^1Jd z@&)()>mRK-I-puAduwrgQLw1Yp;g8Nl{IIbX7xwHoUS$$X; zi(m4z^bH&HRn)}!Ej{(|nTyH714~3RNAH3Jh8k3f+>!Y#Y#`ylal5xLF1k5T7cX&p z?B$3^uXz5XrAe*w+z_#V=~EMFN?}*vUWO;9`mZ@>=QMZHjp`=8DbbVT`GrL}nPAm% z>)`VBSS@Xd&d7^|s;!LJON~!d%(H&(?ym&IRMEG~CPUta*+~}@3hlgC7RSsOL*{IF zg<^=;vsQi*TIMOnJxXj8rZ^`xD_PTLCg&DkKb1o#Cn&sdW?G%Yb9>}YNRbn?yToLi zD~18fiY=N(v1T!|mPl+=JmBYu(eE%!&$KGIP#MMAqGq>0OozFrWkJ?$On-UXobZ!D zNkCcIkB9+Nt9Swx3nTud;@^hU5q-P_nSvQ3cvaVgd|Oo#m9(43cu_|Z>Je%9uyx|$ zB&`mz4(sI0=e#>58kJ7mL&MTGEfZtyZ!f^nC$ICSXVNFseL*(6d=BPw?~wXM{&i>M zZDB#})l^duFCx+3(qk!yDg!f>u-!Lbn;Rx%GBJ{lRh7~oLc4;Pctwa3UEoy(J!w`I zV#FZd<$iZoMRkP891|M4tAy;TaEUE*BSF_1IdB5*a&*^?ZiMdJc;;6hE)ghg$AI1} zrJ2RRB@F+t$Q>ej8wZHdSvkXSbZ4rFJ^>ws8xMy$G?#`pRD;IGM2gu`$0=FoH`qLR zs-&JZ&L41fiwBGQJ~pu;j*OplTpqS70?>YX_L6*E@9DbaEb!Myx7+d7^UNADp=G%) zpJk+q`8nOb`-oe0$%xG6zL=Re4+ZE6bvfs@luk4ByI%>zPMWH%JYq>LEo@fTDp2xaG^N=+f)*xJspd^hsS>rB!ntLtx zIFv!JQ?OMw?2Bb+>vEcl4GL9A{b6JFRvRiX*1xLlc*3k1mBtBs^{?FXgw;z8*3EZ!cRtkN z-(ut(o)ux4p@Dj7*bha9*+MK&3%u3IqU^57{~Erz6#{$dwersLMrR?$>$TUgtrf!+Z{()?6RpPQVY z4z(kS?-!EpXw_FHX&EV9aK*rUkLPFuVbhV(;`=ctk^4xEw2cW8jtA@!5Uon)5GUCT6o^ipx;mGi2HKBSf9OE>ag1Veiz5R*j(q|mh9dcIY(qGA zgf47O34b=A_HHpGGn_g@z&P=^%_Q9V6l z5qgH#A~5;}B7Qu_k9n-!bX->i()6&{(r4EnCv4`%?=wVz%{m9)x;9o^?!q_y@nTPZ zz(}`IDh=x|*V`G%8OX5^NV|8%q@h5kG8DWbgvGW&ctGWna{QSJ^B7xff zlN(guo7uO+1()~W6ZT!GIU7}AteK)%1f}@Lk}ud%#NiO1o(3pqt+A^@!?G*lpWPGa zkc6kT$QN-?R-T3)kUQ~mDrc`qwFX`&?t`0cje0PN!>c(}KjKyvb&<>zcJk4SM2QRE zk(8mE&H1&W(^!9s4i74UTM*RwD(N1_e47xe9Py=d!24&r{ewjP3(Of>={mKmsS^s^ zgS^n*h^}+m%XTO;!H0=%6#;Bd>eGJ1Dt{1wMVJE&z7c0euL$>5J2nz?jGy4suIhuw zxGFQHbA?FA0kzdrbC42)Z^-%;D>2wjssU>7ivajkS5y?@-Ze&%H^M-SsV3BAxNL{q z_xKuTc@j94JCgzp-otYV>Ft9WxZ2mLTUW`Eyb#DI(7_N?YP9mIhoXg4P-;U=(_u$R z&L5%?8wMHP37L{SPge>CerBjiS@+*0PN%)h@UoBnzZ*%FS@z#lsD(*3aUd<}Pw=1w z*k%n|UmW_csvPd-Es`8;LaGXTbNQdCTGO@>LI^z`uIeY_ z!@P{ih6f_PE-@8s{G!g%ItzLfF3Fr6@Cy4W0{*uXh3^%5ui z61{Fv$JsmRWqna4lO;1s9Q~@$zHi=cbG-BvZp1rlGa;?~t^GqO^yp^mlKLn7RUj=w zb9M>5iQTMGkp;VD4AM5Smt~53EmshcpQWv)_7bbsB?Kd`ylFY7QgXW_|K;A8UM-aj zU%k}vuVzsrf5Od+=wqlf4a-8h!*xfu`iz(t-_psqwOx<53g!R^@6l7^pL3>8nko91 zx8KHL`XYXm9O}r2`>i2%e@_=UK!!9^Z?9NkQFINKF&sYUF!%E#yHLk9&Z(_D2KZ># z*~bqB^d{LOZugiHXJli<1rbcAe4;39{N~(!X|2=HyQSYxC)y@9oF4eak4mRA!Lz}i zN&tqJsi>g(p$I1=hS&rSk}6Q1AoNT~l5uFN5{tB@QXX z9XXJNpAAG!4op@ElhshJFGt-qlog~9b{pkxX@^TN=-P-W(XDjRT-VLw@Y(&3RNP7P zr~09;CgEm*#q#nQrB~$xuIvuT;3ZgwW!=^ps8=5XqMjAP?`RiBUeg&=(5lM4U$0Dd zqiaBHst<}G_-&IK~9d;f*#hoDdHQ!A@LO0CCKP+UFr zk4r*}DU?9NXVbe?qSE-Qi#h8IbZtD{7n9Yp%OhC*hd!-6K@tcSwFR~{|08)AT^-~T zZ%V647Oh@5v8T`Tq;in0+viJ{Avdp-!;lUP!uU;9cGj zNIbGtj;vsQ-y!`db)$1W2MtNlEsyIfQzurR(98r3*N4~m#r!yUV*S9Qa?Q9SBy$^J ze0BZNH1OuMDN`GR+j4j`(|IYy;CjkgN<$(4eA7_(t;!RoZX^1>YT9d0x{+&IdT(nW z!>Dj<=(**46bpb6oCm;8Y7Z(;XH+Y1n={fUltM$iy!D~nf`=J&-WkUY^%r?)tVpIw zoyEtqD|Wj12uV&ve2{e3kG2F>JRkKc>Y7hr2Lki6_k1$R8%pdr-Z;9DTfb4^oQ1Jd z<~{?roe|1U*Ui{RJ$o73$Pvet)_Y>I?1*o~c8zd=9XAFa^($nV2>h5sLkZ1tyLE^W zq-AE_>+{$Js4jUfbv=sid7YHv#%nu?$^Px`c`4nUkL|$;u7({zKF+U3(0;+sON$Cl zxLHLh$AaT*7^`l%_}ZT=wr>fkv7r?~e@e}%H?#}}C#dYM!GvDve&Mz|MZrRKpIUcf zD^%nNsC40;7K|EYp~*5L`Gm7>hsM%|Oa&q;U6=uuS{qP^#$ES8Tbb# zU95V-T+n>x56q^7V-+f1^|}G180LOn4CI$NEuO%^J}kJ9vrVth!b4j*z9ERcSUdsQ zzB_pu8D>2cMaq@s>@jaq^S9+0zL&c_B=?}5Te0W+{EODkuzr;NcgA;mA}};w#H+VURy2%LiU&RfD__E zw9@H~UzgZ3&$syY4944u>ShbXQd)kpxTr2&a&9in8(*?dv2-RgFlJlkcZY9tyT;G(gt}Z9f;S)@;_r> z2UAW(2k)S5X3TcyLIFW+!CgP^F2&5WtqLyE^!yY`;fd~ABf!_3-aL2-*O%gLX0Bg; z+T9%&PoxvUMXOb+q949LKxD*0)sm(~Y|RHpgMrz6d;_w)5vNxIT{D;%03bm(W6Th_ z$?Yyj2+A0lUZ`AV>tEekTy%>I0DvYTc=ig!_jSzl^h^x$wi9B~6EZZE3p7kql9N(~ zYOx=yV?Bgz{0E2_4rP8AV>>ipOfm#gj!#NUQ2SBL8QwxMq&(P90{8ShqMx?XIVEf= zgD7E#CM8`cq{sazv|Wz-z7Y*;x+2HJF-BJ0HTlGFeo)m&Op8~G*99w}lBSZ<=ggwh zE}28GY8zEd}N#!OGBtSu=x1W zA{Xj4s}m6r9@4DtaTg)IM$T=gzyg>XgOl6>p>@pw5&$;dqClcRVl7a`o#N-6!ISTj zL%-ol#p$igaxNjq)u6=1dt8qp2%iG!e%oG|VBP}FpYDrc^bCf8Fd-#{>E2Bx8U9Gh zhXhFOg5f#`lDTHAPSj_b^~lCaJTi@D^rwO>WTCm{u4q2YS?yw14Yx&V0Ol89H@85j z{#+{1I203E0BK<$8>ED}GHP?ZLw#V{`Hk5w|DsP^5_r20Z@&vh^X{!MpyjOzqzNSU z@iVWc6G1sNWYOK7tUN%wJ+4%{RhWUEo|TyyU(v3g?TD2!5_P797)!!m|WQNI)w9mgh*fc`9v^m8*8YGcK-T5e>qD-7wdR zBG}>VNHazg*R3pQp8pdDV~JbdvtDJWMRGijMo*&r~eD1yk^?;m%LVVS$<=2b31WQdjmUj27ps0G+Y_z-*GD@J1$8hK`kC7J5Cib zDNQ{hGfX!bK_@OXB0f)5V#H2=y13xmX&f?J&^M!Fb4PppRMGGe5Qv63Sp@cXBcU;q zW24kzF^lhCxU$|)pGad=ad@~`R8qyX*d2(=UI?Lp;>(GRQ&E|0HavUvDRm9~jI)=u z5V>0ES0=8iUKemI3SK9PV2q(L1bO8pVapa$Zh91%9&>LSY6R;JNh zMKW^&ru~nW8OPSlJMV&D&g?`}l(=p7YW*ZNe_M{AcIqrzY4MVl@IhPmgT+V?<>UXu z@*u)%xE%+qOM|#?+S|(7-uHl9D^lsKG2&5?agc*9RLHH{;}5G@VgcBZcQGE23M5{ zE7b3pWn*E%cQAHzvZb|k1N_qnN&XKJl6x^}dlv(Aa*|k1_G$R<=9ny6bjNRY4gQcZD?*am#h{aEPBvA^sLf zHWuao9LTA96ivs??4COjcx$rvQ)s0HX~Oj^i0o+#9uhT|Br+F zljBVgZP|}zIF`E~ex`j4%gqw7ig&j1$d$NqNI;gv^h6jatCxCl|G9v0_cn6tQ*@@* zwv3dA--*vc$B?j9+FB}AxA=MXd_-0!^VQP%XSs7dgAhNzw~o7~gqb;f@45_ET`?}3 z3_Gne?{G)!g069KF{F~Tyt#ydxVJoato*R;^XNo#5n)V{`xE9qjwMHzg-PcYKuFup z%|zTDLh$uDGTj@)RFZzp6m13n`|+-DD#sNfIEiqWnOW-arFx;;FiGm88*B%>i^^L2 z;&ECPW^;dIwo}@5#xgcyVIfH>M?Wk-Q_O1`S8-kKuBpe>^7TIYSUQsxA%MX-wDK6t zkNy2nB?0wseIR`xu^FgyVWiyLu0pH13$^Td^44%yr_-Hr)u^Zj);Dj>+qu;!VsBU4 z0~K*--6kM$OKPZfLh+~W3pMD`eWfn!9SlkrfCls?2GY_qpoVJ(SC3<$nSbh#S83=d zv@#)u`${R(w>+t{l+24~_UU5`Kw}2%&e?AW{<$*r=J~MP%SyrL`6suUYqIRFejnCb zbeK*jU}PJfu=1Z@&sa4x7Y{Fm9AF*Dv6cf(Ks(Q83SBsYk$5JBggV8DaQ+eI@*~wJ zhbr#o8M}d|sRxe1N)l}=NC02}r{vo|xA`!lzn+-q=~ntyK7z;lN!6UYbHYRqO8#p` zc6y>3h{D{+jscDWHj=SVhVtPn`g%3jmG>}!s|P!v3hbXk@xjMTT<5jwIGXo|k92g4 z9pzEqbeLN(qA~cR+u-y#zynto4~C{b_{lD3nfPCwcslY;G~}9d(l5PA1iBGN|TAF<8(Eb zZOTKnzk*gcQidszn1o7FRA@{(XqvP~l5DiM@0Y0iFnj57Tq9{A%e&OCt!j6-y*l4n z3<^l7BrW{QVCX4cH~OXWd?6d`755Z}%uPjVMl}>x*M7YqE};zy_yk4v>o=+He||5a z^w0FOY?T7L8waKC9Bg-VQ-iv9)=TLp`wrIVm07kek)X5yhoQZ>i4A4$#r8SJZEHkk!; zEC9zdO3mM)IaL4qhWl8pX8I#^6ML&}U6#un*&Ai7=~|h8hh{bn{^+kc_X#(=vSs?^ z=+dpcztMbm9B#BhPz!%%S2zqDyi~X!HW`fL2Uvv6bt|%JEQ+beenmoJ3j{!NoC4$o z7-9ViZOSXHP>;E4Yu602^@kvXD+~Y)pJAKMfE0hG%>O65CNB9Q*#BEs z{N?J-t;$L+A|wC%PRKd7!ajP_a{@R(I*L&9JvhMs|KOx`8auRkD99WKj!0*v55CLB zEa{Ga32HDKvi*O9|J<^UVriIOe07{&foEHX8R`|F650G zA4ijq&Z=-WRwZ7oug8($)zLi78L14p0NWsNZF|74IRCeHUX3cwsU>b`CjQJ(!`^@i zPz{0t{cDxp12#<(4n|d9i^n@DK?@t2sLV|jdSaW02j9MwpK8J7Te7b#q(dONbv`y{qp2Ndg2_b z^KijYJM!#5*VZ@ou+x_p@Bf^k6#F&m^uDa>%&NtEkec_tvQ20999s1$t^svtzi_ZP zv1rd1mF^Y&O;t!hbyp^{`gsyM#EGuq-ZY1tRzg~w`kl;Yl8VrZ@M-Q%na*NrD*oh_V8(9&IdoRNJx?fH8k&%4A&mWv}^qWE`Qy=8jIh7 zC5oM4>tA+u`8omrO4WN;zAZBqekPXd2sj9j1Kk2^-TJLD`2ihoGnf-Fu_kfT~o0Axw917asEK_Mn=Kw>2XgT~fkvKnZ2=4NCG&};h#o{teG@o^WE#2_gnMS?-6F|e?34&AeDZa`XMD{oNA2B#(*Yc zt*dCVPnpS$pd74RAW8sk(FlpPb@+I>4VeE4A?Of=*EJ(5R$8K04mHRW2Mj6MwlK$J z3XHQ(j=9jEH#8tLRCwoZH|>7x>F0WKAm<_aTDYbi-dXojnBxAvr{})ssn@xWRFj2K zS(keAwNV6B2^*rG^IBGM~22{PeQ{R?$K+^YDI1&N|cG6C<~SqR=vzm8P6); z%O~?Jh9;7mW*0v_*}k}5)(d&Z6rastSAc2|>_c!Uh(VrQzBM22Z|l`d$6sQ?Osg$g zl$kRc{iOb$9NrCiFG45~1B>QVkcnp^3<)}Mtc^Sd0m7hBAc=WN_>55n6HEF}rCwnBr;a8q91Ym-n2<`}^^-#n$qm}wO+DZUIl+RnI zMbgZCqR*eOa;}ZBJ#TJ%4moPL74gNOk{I zPki#`g>FHQeg1MH31NekGVrFTQ#63^ezHBIuZ5ILE0jVZs7XbX1-lp&eN;w!6sRP7 zg~jFtdr618{Q;=fSre|ANB}_?Jn4NjZPyI#m~M$adcpNzh z6P&_LNsf@Oz+0j;)AAtS=mMg`Q{u+dfkHUO;DXX^s`xBtv^q&W6bC|C+Gt~ zM1HN+fyNBxh729_2s>gaFW=)^EqTQ!Eat)EEHgscH`nb9we_ylL+plTG`A0&UL=_A z{o^nL(W;yMh-#~mu1^x56KrNO9GQNCwVvqwK9@E8dcUO9X}v3J=ce)HF>dO%{=)BA z!lk=m!W5ToX$i;6Z!$Wg`aPJuk0c?(JJ>SHet%@Lzn$Fs8A{Uv-jp!_hcNQ485W3~ zeJlh4Jk8v7#2md}4EW>py%p>{v-~Rr%ym+AWlkc zzxCY)TY&P8_Jt6p7pl*FOln5;#jk08XU#vz^RsOCf=VuN4LMH(G49wt2;gfGBK^(> z&UYa8IwjZ$!VxLfcN^iTRE%dvl8#~jM~o6=YpBzxyAnr}H4qVy6swg1R7zP|qdU}4 zipBY{@!82``G~xa)CrtbAgI5wX(S_~hrC$7t-uFj+ndK^h*ocfrdI^`p1{jwCYv>! zM+m73x}YAfp#`#oQT+AGJATdVfq-ovcJzN+U70=}mzFv)(R$MnI`wkejn*>t|9}OG z(Bp6Y8#SE5XC(Le*YkNV!Ztc;GtFBC0k0V#fCrNhZy(IpcS!s_d8 z=IG3cxA0GqlQ+-O(LA2 z`mqdb&D)#2K8f4$*+8sEfUz`o)SiZzYsRh;kIk$cYSPa@IW{ZgIZ~OKF6EEUQlHb* z%tL0F&F97N8<0qHB)}F6q9lnp8jXt`VPU^f-$UXV`V zJ6$}X)qwj2R#sM1Dv8sxe3&(mnWeh0QQ~mI${<5RxgElOl>6xr-Ow}t!PNlCpO@;G zDmmt+SSoKtTLQm^CD%KqJ70#7qsFjQa#Eb6=fhU%(i#Eru&Kz+Qg`y{2Bq@Y*EhMZ&P1R%XI|M$B>0L&7OhYTUBV@Z0pyFO>!QYRLiju1=jH4SL|fk6U6zg z?0f;5`e!~@qX|Bz7>l%rcFtSXacyC{a$vlq#B0*Qo@K!!O2|J*%G#|UqEI(8cRbHi zdXk5qOmk)r$)jlDb~su~;eYD-Z8M5gl3fh06=kC*>v7=Rhn;b4Fw9tyinjeY))AaV zovkjN^xX^9Q@b!VLX$iMBjU{cgy7W)aBVRBa)KQo=$fH<;KK7okcCXu(_5kDo!%U& z7k9|vsT%z~5pAwCR4#`(x7&~}!wyT`NG?<9mH7_81gL?(3_~^xIXa>8X$j)1=CZSp zNz;99Cubt*U?-}`k1%?K#Py4K)#HFUsgcL?VZ&=2H2b>QAGe+j@Ps`9V&cks66Hy1Y% zH(55(7NLOy1PMYT?E*>PZ$8!y`*c z!{kQjR6D!L-r`Eu%G{XEXYxMJN7&!yy$Jv-C7(oDNVd!apu`9_; z^__f$!am}pl^5FFJxQ4Ia{-oDrpVWEXC z0WP97O7 ztmyw1+D4)GAJEq@N4|Y>3==d_oStt+OQYo_I0_$Sl8k40EX;liJshGuw-!o(Ia#{2HaO&8;Y zaPf!}O`0@IO0p$5;KPO=%}-E2wm)_3AI&FLRNFT()!;6s&oc(~DD*bS1iV?vQ*jb| zsIPoBfknfOrI6{LG;s=(h+EbKQ~ADPpznbSyc=Kz4F87#8^roip@NagHPN`pIKZ|p zZ;@AQoJ2aRf_1004;F>A(21G3lKmNuU0WAg+@oZ1{KrpcK%qZ7Wj%^<*L63)*l^g* z)ZL?32|W*Co7f*g3d4BYfj=1GZvJxDQl^GIoGKOVhpLUD%1W~R*GH8S10Z0h5X-C> z-690T*3Iv!vJ z8}(ct0Kb}W7jrTGt-g@rsE*i*tjEDx@|?!&s$=cymHHYhx%u^P1oKfcG5R=R(QGu% z?F5KzUebggH~eT|$1flI1HzHAGSrk5LzCj*W{=eR>E+yP<-dM#SA803qdxfrtNaxE zpZh{TKA+QD`Arvwq3kFheSUo=I)7#CffBa+o7O^#tz@pC2RDOO+s9gTk*+_Hw7TpGzKV0 zyWKwdk-cs+8%$KKIQ+<^a?qm#Kx=)H6Y zj*adp0Q~#5x99SZ!D9(?plcIAQ+5Fkv4j&lkiLFtT<_p^1VsF;*`dWzR`;3YB=oD1 z4NLkcNHFmIV#Xu(Fw#wF3CZV*Cy zdv)8^1{kj@m3)Yd@2!b{j9b=0<+TNpuNF|)RB+vz4ova0E9x}UBuE1JDNh(1-kkvd zM;CnSc;mbSo|Ug|Gi1Nq>X0m@rDFa!7xZ;qaY74h1g3is?089cLYvoK6o0$$`cyfb z+A)@`fblt2k<35xhyg_Xw+m!MhFn@#hm8Xz1Uu`B8CFECI)uMns7GZ+`D;{ubF!cK z)@X3S}XnW@?Z3Ri2ReIQtV^u_s2H$BO z>s9H7B#{y;@Y)$_(LoghxM(!BEC1K0IGL47P!LSQ$MzxA8NM?tF0fIxqsgj#E@z~( zG>?bZ+f{_kUgiZ+*E_|@rIlzh4F7A;WJsbF{uk<*l@?QLz)^d{=C;9pF;qUBogbz^?OhzwKAEKJj zz{|8aeST&1fjg%R07*_G&3Gy;Ri3TV#n7;LA3E+a!WSNLof65PjmGrKH=cN%C+yo# zH;FXbHF7{Rcq`p4f-28EmR47Lcyk>yGYm+fiz zRH_$_giQzoz^g9W#q_AgC9GgK1NRD-{WL)6TrQ|bsezXk6~fAXGg9csS}@;+8OPaD zsk1kc>bFl@ zC_^Vs#x^GID&&Wg3z^}{p;_STTma|8_c_EFI{5F?p83lbxM;FfAXF$P!O-!*Ie*4{ zZ;$=*l!JU|_gpAa7H?2c#*CE>T{O3c(~^`V=4+uo zE$~$X4LaY6P+*)ml&U`|NChsN)}QERLuoq`k+IuhrwmF8=|?c$l5Q!o67a9xCGM20 z+`B91y^MR+MrF?v=#HN4CV4bMZJk6lxmXRQFR7 z9<#Y602?37&+bn zW{H?ps@K8&%OkwXqB8o+N4Q|k)v5Dk$1w-Uk;@LWqfm-)?Bx$! zJe!~CKKCKN#~k!SVRsLXGQTQ+m3&)v`7RRxt0NUljy3exqUGwNmh4ti&S2Ma*q>?& zKb%y#VqpyZuSUYqpj2fg=zmoQkiVp~yajI-tB)%2Htl|=)%pCx>>q$ ze8yARW6nA>BR*rhhHLw6fr_?W4qW)LzyHy=R2Vi`J69_@ohiJ}g4JUm-TPLic4-4b zK0!A&EjeXauleZEag?pt9z<)*V5eYW_&DW?C+I)v1gPjl-L0I1qZfmLtP0%Zv#N%- z($0_npcB-zXj&_%e6C@RTVy4-dkfMkwG_krO8$1a^qM=LM^BV&mVamni0>`4v9)_q zF6_a`%PAzLeEph=ovoaftQDV{J*ZdVYK^5i1&@-z*69V(aiul!P$!C&S6@%eRGgwO z?mDt|JZ;yB=i}|Ed9Ws3g;@jU0*H=G+ByOHslMNa?i+tIL8$!D0N0zkw<5@|cC?(z zl@`;MW!-H?RDiOU5ARa!_pH2G>o|p-LSy$l2yMp3425_h0p#KjlU}>>h^J~rSLSAw4 zd|Bg0CEK&pYYQm=V+0B2-)MhQXX}#FEy)|LrdTEVIE9Rpg+SYs1C%p~CjPzo{@mTA ze5SI{DDIO>aZK9l!YE6cA}Bydgec{IgR&py#>;q;yv&=ntzN7?IBR~SeRX0%0*wD4 zYWTOi%#!Bii?Z%mH{D)aD$$Wa9$y=OuvUWul-CHV{+ilKQFOE^OL;xyD}UB(Jq}_! z?pSjm!vH)*;hvYF{5xt(qljBO_)C;+hx%-5(|M)MB#f7R9{k^@@X6YWdxaT#$5wL~ z=WZsc?soFY)Zzba2#@|E9MZH{!Mbm9KMcqhOgXVzJ?@lAL^;g3h(K!xv-bt5lf z(ksbj)cJIGg=nJDUflK2fL%Oa^q=5<@6^27fAiJMzdcmR|23HUHQ5w|xTVkPIZL); zkwqdC)3d8tpD4}Bxxb68z`>sWA^GKYMpP&+drq0XXmdsES5Zic7-72f@ zSMU@;Nk956c#r$Wd2OOXQ*2yEM74fTX_wUOIm)3qY)H~~2g5=*0XEqFo{zZ?GV|N# z{j&bg(@)Pork`j4K+<;@lwFYDq&s*MVdhV4V`9zkEgvReZQ$u{d?o~Ky6`)H#fORK zMi!yCSb_rJ`r*~L5E)!Eq+=vuKp0?yKXyyY#%1Cp@@vD?UBckzB$`2dXrM_b*&Gl& z;OEY!T8)kjroJN6)j3~wcx_SkAb`E=68i>0Y*gqmz^`AA3QKt-lGi!M1sqnGLzP$% zPTxer4KcTy49Co`o$6^ptr9&(^wgW07>@KVey$lsXH}G`s5O1ZuGGfY@1A`5T#emS z;Vr$vHx!(Sx2{2?E8AVOuQiHj%RzPf0xvq$2w@1@*s|E$KtK3DcKOvz0M*0x_siGd z=>cL~5rBvcMA9DU@20@$+&DLXJlmgBBQ}Q5?TjvYfhcUTyBTUq$g$Z8oD2+!)3)2r z*I`vkqz(*6(yaRRU;wNH6qtRWpV03T@Fv9UKhK=oqws2Sn)dbCB=A-v`n% z$i;7;{|gQM(EK)IZta=di3;2S2mH3r-#aKrg1qEU}FU@b_~nu zn$bGw5#ZCSiB1b{9P~%qn!9p1Tq<4#1_o<-8p0id69&S4FBpFYkV+4NG+}+i<}38F5cZQ?jzez-b<3& z(`(qH5vP`rr3r(ijGmlbKI!8OK?1bL(DokyeWBhR>3&P}`L9qB8GbndIOjvF9>Ed6 zhjOGd+RA@G$O-b*H*YSnFaDVjU)xMWpOb-iO&$n^91)=yA%UuJJ#urk{lVA1AEvPp zCRj^Li=xyFhNQHg>5*2&Es8r}Q0>3k{kN{4uiK**)IF zASWjMoj0USjOwy7`b@Y}v%QdsFaVcdg-6c7f>?j!q&~_`0M&=@|I%d@eRc}!$QMSt zK~@rQd@*X;_o7e!Jy*;|&;dpU|KUM^%$*_XJ^&iSV!$2<`10~e=>Vww&Xoh6DQ^Eg zY~LdR)-+JFMW6ouf^JH3l6qKba+FTOnEboXxzQ!l=r2VKvQ&C^>y)^GZO^ek9^Zg{ zqa18lR1!EW^-RSFFM=sZwi5S9f?_*QHM)5e_p>?bkD|bQ$GlJ+|wU zVA1j5iqY>u;U|?hEMdfxHC64@Qw>&5KUMnSC?z?}mg$vN8Z}+8y-)0&gB&~RuZ5%S zseNm&h$&HN0rBzSM7wkfzy5ZPw=V&S2FjkUv?DUkX9i!Rt|BDR?701Y2XNZ8DDe{siB&wByuPr`U408jw9^(oLFBpZEyauQQDwR$A1YNT&fcR=g|KB0}Ug!kzbbOm(XVP`-5l}>N3oiMY4r!ji=kLZN zo=Is;OQHV~UldG}o1x|~YA(72@`uD1o=v>wVS|*n!XNAuylic~KA543r{4u$r{m0t zjB&kptG%%7@3)kzj9QY)r(XRMykX#Nj$*1i-MeXdMFW zUD8S`GCq+HA|W34b>}9tpYpWfdq~IW5d{$y0{hJ!0?#4CPuC?k zd)H9iJ2R4=VDORNO0e|MByt^@pe>!X58}#;^2NAVLdNK(Vf&r(JA?fJ;xnEeu}b+AEDkDmKJGKEnp1eg-&2L zctf#Wqo7l8ti+!BcG@M;ErAnmU9(364BF$xwl(ov6)!SYc(RNLp^oW!4?hWEX^g`( zbLDc7b-o`5 z^}Jc8Q0H!B(!&|>-EE!W7z5W6$y}1k2ty50saL&YRGfUt0k}5r;Do(rARQYq`0q*m zJBp=r;`>3%^I~LOvWLvC?li&Hyoi67FacW7_?-ZQDSzpn)(F>bvuJCme(Xy9UZYxO z$UH{+l;R&Y%)ag$FQ~1VGm642UTRy!W6?bQBUyMdcF1>Y~_n=o5M(CC<~*@xvRVdLQdc z)W;&|2w45BMNQ2j@m_~1!=0SgY0Lcq@6e0Sp|YnRzqymGoOCR@uaZ3AWe#9It&5|s z%W|ce~_2#bC5!{usu^mRn# zBj8)83xT~JABdirYGk|H@L4&UDC-lEWS~BRM>}h>^)c0;Ov_oWo5RpV`q8^w%Wu9| zc)M`#*^tXDFDIKDl3?jFUHtTU&0aGT&&Rh@nkZExEa_6@`trG*i18)YZw0wV z%eKD{9UGi!U7&U$yjGp4BBSHKiKBKkZZU3II1@6ZBV4Sy*C@k?`C|H{T9WM7=EU>m z)1p>Ol@79zIIEx?Ukj|Y;H2y(y7YqXBzq~0ncuG_ia~dOW53D%9y1F7ju8ZpnV->r zH!zmrB)@)qiZS(D>4nLUuqVR?k_AA_o`Qw{Dtj>iCVrZ(eXuC{Xmw)Zf3GhUeCJ4j z?+vr`EmjO)tx~AWAD2&z?zl~Y0EwIcd4J)*iZK;T8(;_D3|C{)XVFLoQ^dj?Rkq|v z?&PW3;|BCJBg}7toBSCe?SuV9qXXiHX2Cbf-8&uuXO8ucy!Pl^t_g{@+XQJemAa(3 zQe3{h@AnnAaQ(oB%DU&~iM4kG6!;UB5M^#~SIC$(pX8HQ`D2~_=;%ed*XVm&UclrX zQqk|F5KLmwH8hnRx6h`Gx6Rp3qgUlCy(fVHFf>Au0|;N@pI4Uz)cy17EH4!t)(91j za@HvE!i-HO%MuQB(RPe)^t5th6+V;+d>Wd&G0^;4@x4Dfx-RQ4zi9K3xZ@c3;U`>t zz{Wq|=Dhg}ry;toc61W)`UDvGDoeTHA-i8FcCDvx7d2gR)zd+(P@oU#Qz1*ZCulO05D9S|#IA1&=_4x^N#T&z zo7OVvgaY96Q@{H?1!l5ap?={f(A4#Q%PdB>xi8n;TWaQS8{2ycsZENo{(qSLf6w?n zUeSBU4_Sg^*e+xZx;nRgR)jG8XXnQYq*RTod*)Az0dvUd>PR{L1-??NTlqDcK6iyt5aZ$_oGfD^w?RaY|D$ch z#5}lIs(zni)th;9D-B$%Vs`-lozH(s2UFU4FCOctw>o#6gq|6CLh06$aj)>tfbt7S z6Sgq>FhTjM%mYd7PwVQr=Nj{;+PyMDaytme5az+n8yC`1<{Sx_t)JA%jC2_Ku%BEq zpYX#B1p@zKAapi#x1b90n32i#a*<+6QpjYAq(2c!OTSwS%P9#~X-bR7*y2W)^TzaM&qvq0hFr*G$bNA$+e z(N}RVVJMetU<`kzxxp!N2Sz^6sK(L{70i0k!jG1-yvMY8==m~O9liqCu169-fuQ{f zS{6|Cx(BxRtMermC6?M&AMdftcdpdKvqaUhfXO3Tk_p5VssblaV`7KtQ6pt(do88U z(GVCN$pnhkWT{krc^R0t_7POhI%=$+N5io#tG8UvXKPeQsx&Xvqvv zr+c@#=~cOo`l8igF>Io;J5hFA^pk~~1~&TXul}1qu=ynqq^hucCAYuBwAl5wX)7i5 z7PYPD8;sETNzeO!q@i7Knm?0lf1*?6i5du|7G7*XHSM)nmpPVxThbIQ*#RX&Iule0$0_nb>aC zX1OfNz$prtSVnX^{l#SeJZkDaMX=}|yVxd@0XH(jnNm&8mgSI}_|uXuRU^v(!H<7b z8~>on9;kO0M@&_Q2uDk$vEjn=`%u*ggxgPIuO}g!o&stAj0^vyVCiW59N}F4N)&Qo zHAQbM)Q^Lt)GA2=;440I^3fmLD}HS;`17z_ky{6ImXsyKxAjpfYm&B@6@q$k!x2Q1xiKYwUFq@b;yIb-y+!YoCpzH(Ig) z_&srPe$Pmkd2ME>SjO@S3`K?P?i`P%Si5Jc3giukp~1r`hJ|9&!`ijOVx;dllT{E= zjNOcBQcyV}0w!LltA3l?mMSYAOZsD$N5&~MVxkG`85h5AyaAk-mfz>YF$%o@KuKB}pnpJi;TgrUM z>gy;@(g{%TPo426fcOg&DVhY8u5hHc9+PigYuzR!ETijdmeY}Le$RpBFfiF3%Alfo zye9JB31r81q!HmGfB#;4pX|M+ zaNo6(nY(2xN@7C>;H(kQo&f#c|0_fZ{v6Z0{!L6p2V|xd?~NN{6BY72n-L@8!U<{TAqpZ*vwqw;G3clOrXwGI=UbiM{D}lKi0Ka>ggK0S^@~{ z$3EuxJdrBh7pD>hkX{QC^_wvObldvnAz`iJeg9z z8aU&YUS^EKcfc9bw`JoX0p6*QIPU{Be_kFNu=A>^86^g4%L+T89yZz>iDndPZaCB5 zo-C{?vPDw+DNdhjf7)`*@v4MR-RS?QN<}mXB>_1Pp+_RnkqNlg86d{- zm^PKuJD5t&kda&9pkHPI2e1ae<^4TDMX}H*B5_P}--*n+Ra5*Vjq&i3rV0Z%^CJ*G zgsjDKNqg*Z`0V=cQ;ntp`~6@9qGJE8l<_ySjXu-czUUZ#R5qq|31^OgpY?uCxj6)z z`V-_1%vg~ZFQ_zR_Lr>iG|5{gJfhBg=LeMdB1r$-XUtFzm+*O>a`ax0` zQPIMZj3JtSH`_@%7$fBOxJE0|K=-}9m}sJS>9!bDM6H_VBHtqa-x;wvLh8QX-z<}m zFHXuP0$D`(y0L!FJ2$5C{Sgar%8M)Zn^Zl#JW(@okrckW9#+VH19oy>|8NBkfDk5_ zI)n&j`g2#$A5>)ry#)CX@`dd+ijhE2H*5pb zB@xfr{-Smj*;4cqv$RZ2EgkHYr>gN{WqC+t``g4#aE<@nbG?8S_9CA1;L--{NN9H^ z&NsF#C8|hbRc%8R^8G6_)k`vz9Z!aR?+VvfOqLJn1{n{R?Ak{b!pR9Y1ef1e40k|P zY*@^B4|(wMe7YAD@bnJSu?Q#YV+xqEfRRqKpoD0Npn$%elfK(QzqR=X>?S~pJel+F zk@JjWwRwMal860b^n=}VVTY%K5oaYh0Lh9l>jcRW$Htc`yl6b-cl=~t8evcbr zKn1O(4<5OOf-(Ax)`p{m%Tki7e>?HAT}p%U+uUf7Mo3VB!fe!nvT(Uk5?<|-U|fj~ zIJ?HFFN6B6JejzT!H=3)8hVRk(dcyUGVX3aXIb`tP=d0ER)UUx5*(-RjKlGRfr4{*_g77Iz2O5`00aKQ*Jm;k%D!L$o252RshQGnq?X5>Mo;y7Y1c!Gh7-QIiXMTGlQVQB6$n{fV*G9 zK=kk^J8qui&${w%wJNV_t;=%oRhSWbck_?2ng0Ud$lXl}{#%BbUD!woOQ<}V3XI_S zF94k0aAwWg`NXrRd~OWk1zdZQIIe=c=$*gEZp!U9VOse{q}I)QMu*i+t%u;|8_s{L z{5qwndH#b)?9gd!#ah29Sc$F4_wB`hwxPf}n%DTuR_1UxFu+Kw zd(pm~&h8EX5L6^K{XIs$(=x-yvXXfr51zeg{5T1RAb>%lwA8M?IlYp<};Z!;`~IR86(8B)_~mP7wXV^6&-vzPW6 zEB2dw7k3#t^IgS~bGxL(rF4G;dVu1sw0^G_dBSh4O&Q1(##e-qn~%-+=zxJE)~zFm z2(mxZDt}5}YShQe0FvS+WsSCW7-FwWO8m3f=tvz9&soTvRiH#pzDlccX-`b0)>#5` zED%TsvZa?IQg3b)6C)TPNWY~-)n^cC*0CdI`MtLu$S7;{d`N~cPuwmQ@>94qsuzJ~ z`CGA^b)ydWZ7YBj2q~4}b!qmBztddI**)CLr^DC?13-kwANaqBtQl20n#Vs7pBPU; z_D8PZY3V@gU_k%?gE7Q`zsGDGD{j(z6H{TZ_Fw2};Far}dTz;uhbR8gq4>wVcWbzBv^J_5i(=6Y3TEVMaEB%^qFQZAD_eV|wG8qJdvai`q$98~A@MD!MOJ+hE|^ zE@i*`F+jgu^;tWvMrphk00UB_L9>UqQw+miF*0;-=OWk|nZ$5*ttg6(@5^YCBa*ax zvpGd$kBy-v8F@%0{r-RytM#ztL=cKKvKOX^N2E9)KUt)>so0gQ&y5V748eUoZrj47 zW7;z)zbW>;*DB15`!!q6=cicjnM9OE8Dn;l+bNnCGUGv&ygTgQ55(vDKshO zSnPI5X*;fMZM>^v?K_3kPqw%btFXT&czfxUt@oyDFYI6KBNG3$cE{dt?*G+X(_CLcj(_Q!a#PHVT$OA zic)AS_MzqAGwm&^=1AOw1QA=rfEc9#b%OU2= zu!~DOrAx_B*8c|RnoaW$Ncp$ucv)!whfhECZ_iqwE_(`cjBv8ku@!9D4jZnlWPnE& zh}$CXFI3xMKGLHdxsFfTzAXGWQ&*hQ3g3{f<$`V!og5M(s9>gNYNTLhWu_xhr6VB% z+q1%A$cg)d!I5y+piJ9BXECMY5JH@8VE>CzGll2%y2RlpUTGFT zb#<*6^_b)lA`tmk>&@h3cjKF}w#}N`3B)(gLvnv~q>=R@1h}ohp|&_RwhX67r9Ns4 zF-ojDVgf#gI6CY@`TpuiGX*5PX8oF*mJ<7OXO^ymq2(X@d76PfDQE z_PJV$F7P!NU*tdX8Y9%3X&$AO%LnS8N?8ovV|sWEV<{s6#4CtJHh?~#7i9m)QGNN; zZFsZx$ELd6KeAM!-=TH?>z4RHRtuar__V7Jjd=1r&eMuV3mzHXFMt?Sc!iE+beA-( zR73eifw`MkexJ1yMN$q>p-I>qE&m1+Dl|%+Ji0D?zZQ2N{cIDe6ZtfvJ}F-uy6=jL z7c1k>t*433#(c+V%l7HXJS>l?O{!vQnmS#k$n)S{tKcszL;wIJfCE4T1Rw&i0G%(+ z=wQGs6aW%13I%`&WJCcxuFAuJd<20UAbvsMJ5VP-@W}gCXbk{D>Rbn)A!V%tppZVk zx>G{H@<#*kV2PpuIIth105=`UL?(bxs5KKn2-JiLz#r=Cs|<1%Kpp|`4Iqx-(+|Kx z@OiEzhr#gC2XbKW>jU3mbbbULy}|?t0I)zlT|g`}pKKr&l#d<|3*ak2#7GGN71SAp zf`tRt!~ano2EnJBNFECW=P#ruPKkgRpqIr+3xev@L*YPx3+h^DB*hC1>|wOQ!axF# zVuV4166szO$3lY@BGJPlMF{&0G{J&E^v}_=d4~`Zm_5mefDgvu?$HIdb($_PgMeq;Mt6~K33jjbM{w^R0jXxU*g5uW$f&c;n zM3j^ekU?2dC?FiaZhm=n7&xCCB6$!f)K5r3oDvREL@$ey76jR;fWiR>C)Tx2Ns1Q` zn9FDb!uSFfMG1ojDU!1$4nhkNBvHU3g$ok~nt&i8{Bjg*-ob^4W=~Qg-~@=|0ubQ* z1O%8UIiLc1vnH@WQ2yQe@&F9TuAD4&5MG#|zJduYoWF=(6(yik0DuST>;mGUb!G$c zeo_aB2j~(Y;-G|p4qA&s!Nc+E;kQwTLGsxklE(vu`U}~LQz9Yu>1A=yf}lHXQ8 zngUZo5~e+2!CGInC{J)vf&q=t!7p<5Pjab%k9a=NMnUxn$Rclq?H1(fRl|GzvDIr{ ziCm)q1e`V9A2s7;aSW@7IAVXe(xxb^ii_^GT9C_i>&5ys=Q{NZ^txn0b^n@pApZ6&D#i!h&s?u>cF!V!FXap;5L1yZ}j0>))uI_VdxL~oq zXTa`|e22+$;*lzWUrR;1cQz)iHYc1(Sf{_I(Q0qww7u;&t?`{jixDk1tw@^o4k=;pF(Z||-cZo5y|m%chO<4i+QpLa*#GmO zEDhJeN2^{pyEFXLLg~TzkkOA#aQe-5cn)`_cad_l{>0QT=>=Rk&|F9H<7pe8GuA`; z;US6A;9ZIRy|kNfN)#>ZwyCSXo=i=ZulJWvWu!c(sVc%(tT5mp+cBt`B~ts$6S~94 zL961*%wQ+OUN&)S$O<9TG#JDll}WUCld%+8_j7{|ZDY$Y8Cvxn>#6ueSRm@JD>@N7 zlG5qj*}mj@BCJ5fdcfLUyyk_9U0tZrS`OccSY+=23^Z4FuV*8JPpOZ#fgOwSlHps= z2c-TLa-Toy=P!9f`Vncz=b}fsf672_2>sf_G?*I2WSOJSxA}J8243ZT465DbToZfV zcgMN2N}#9+IEt~-+gS;kDC5YELf^czLc1__ZYMtMe7aPP)G*Ou!-9nxhWMvHt=wno_!kuZ9_rpb5wl|=~d?&KD*aR-6y&SE(6q{-X05z zy%AZ?2fleKPWl^`S`OJ)AU7!N(?dqt#Tk)7)Q)k8IxFI!{Wj)Z1*h}dT-R#e^~YPo zkXM34rFh8z@0Qhq*R&6_`fKd6ux>(el^%UQ!cP+n-!z{^^5O7A z6+37t*1!wjFHd+dYYv!wT%6T0nJnw@Ber##BRvN7#`4`zxHdCuu?0-u5*Y{Cp7XP6 zv67icC*E9*`rQSB7vnbyyeDn+#KM=TuZ(m12HHP4|C&5OA=`Wa=N4_{dl?JKdpc{k zK^+K}fa41RD?v~lp&_%1EAt38CEa{GN4t))OL{p87nY90bupSm5yZX()lm_8D2S1H z;8Wd5aM@e(o&KWfn?bUyKSMTQoL=qZ;X_owiDi2%xz-~yFJ$X z^xykK;tHTMHD%nqw4Em5fVJjz7c+U!ef&OdM#{)_E~4S}y|D1%sqDd(IVV|%=R}10 zD|z}5nX#>4+=7|}V%sL!*wTSnP5p#jG&dM)Jb6J~9BxUF8_hsav*Y=V(xpHiLVylK zz@Fon$%6N!$TmA?Dza(P)^Q!#urMrjefHkRP*$=>=x#FiiziVq3-H`9;_{f!+jmlytU5XEu#zED^X<|K+>n+6T;_O&UU))6V4wZ67(AD z-@|=9$Y%xL?s>DcWGa&T)=gHi${ND`%~(rgX6=6S+T+zdoToTGNBUHzIv}LQ^66=eWll zhN5XRbnXxPDZN#nlVvh+C}?(iwR6E*7TU@x9pQBuap*VsE*0DG(^(>qT{Ln}#+)P0 zZ{!@P%g>$N-HKSr52(~^J&w6hoxJV0PlghC8dEOWS6h?`b3Ap+e0ETUH(L#z z7VGmuhx&5v0^d#c!Sk6JQDeUKx73!FhT(1ZUhDj486qPK|K0Z)>aK~hti|5<_?xV1 zwBt9thK+rIZ`WBDRRivD+3~3T)ywT~5_|YgZ6Y6Ik{+DzG~OiE>s+fj1)NsJJ>x%j z0j*?RZncANoi5MLPdlOA`8IoX(Gc=plR@6kG?m-i6!|Bq68b2t{NP-5?x6a#+Y}66 z08k6yaCgH4&$@_w zhO;mZ%{Og?Xq5LK`25~pWM|$L@~A)-1@gXaj2ccg;sAg)9^-k>a)v`kco>(y->l0S z!=9J~cwCHhaS#Z8y^4w7jt}Z9oM%Wt8HT>)(asI^t1V$Ji&!2sNwz$Z>UWA%)H+o_ z>=%7_36c2b2)9Bf+CT8<&j?hCQFt4x>XljkuJSqTj0f)m44apw?p#a`xFz*PSp_D- zdfF^>c1I;QgEZ5EkVnQN`V~Y#SM>bNW1}$bNqyvoBTa*`YsY7HgvU)f0mPd8F?Njw zEVO!3FbZJHaEU7W*1`WeCs=}jYr^}Wu$dsw14*HTv236Usyfz({*x*GBq)RnZOYC6ECX0Ik8EJbFQ)VKXDWVf4)_h&HL#Th;BtJ=^Jpok~MjymfwuPR@4{R2t zf%*t=H)8k1%?k`swpKoQk$=K4m`zB#Vq5W!u5~?we0n$M?te83Mz!*>VpYcK%+$(3 zb!9kp>7dJ|qZm_xz_p^pfkE(+8=9(etY#!KoD)l#8NWCM%AOnZqBQ_!)Z@}q;^G4R zLI~|ra3QkG8i@b+Al%VCphI zuJ}Gm{oO~UB{{|SR!ZTSz?^O3Jbc;=3aV^d{(;f7sy)@!|cX7*bn|{Jn|>=F8Bvtw13U_k~0P z3!7a2l!9Zl7NgTn+w^91GCssUT!_iflR z3j*dkK7YP-*TioFY<0#L9%H1Ds;5rl4;t&I(Mk6-+~lkk-CudXnv6_%lAP%|(yC>x zOxny;KeE`PNvRjDLD_$=1Nh6Z9=46zS&7greoYC=aQNsdA-9YWl6m4-iw(152E8P( z-3`fM{$ zrPGRd$UG}x2M%7Hxst`g>cYiap8Ww!Y?ut}zU)%Fm+y*f{NSHA$=}yFfUc>?Q?cOk z+UsURHfQkg?%~Nq^EKyaGes;nSyX~4bu0Y3sa&z{X>Az3ROyABMaR+_oFs7_lmPrD(0!lz8B1BOxU zU?O!;Wt)1M`0#2cX3k2OJqDmZXuUk>gUsG%3{Xsb zm7b%CuGm}L2Z4KlVQ1PU7J(4ndLE5als9~ z)h%YgjpWEp7yVk8;`V?z)Gn;q7*`{h7-5M3?3v1JbHA{i4l+gAbKrGdSVkGtqwLVe$zzl(rOnygK#Ox@I#`g`=O#rJDtc3Nr+A zS(y~dHH&a&boP?G9)ufZaWcEEEQeYc9PP85SLK#h_$^WZlcT-;L_Uu1vvMwnvGwHCsFgMapy4n&YXFsx-ldZl)l;$xxB#@q@%|85!f z1^1#FdX9jx^HhA5K}t1Ihb$6iYLfC+;6ZalKPc^ZkvC8?A{VEdQHW3mdat>f*93a2 z9G089R|-jGd%>*qYor?ik`kt|L>rz0n9Otr*}y_UdVK`KmTdTD-=wrrPZdcp87{9>+w?*&d3exzobuDSv#F< zBZR;@lzWebA*iFLTPbF&^z1CWH8xgC;t|;y7d8eSG__S;GBkucw}uKycoXf zRjl=l7a%)76TW=p)ABgJQ+-e%u-A{PYS^o>vVglSIL?LZ$C0aeiw$;`(P0ZriT0%Ar^WNW}*C+Ua zbRSORR6Z#xW`-gM2%b^xAh`*V1Z2Nxjuw?cx1O9k!WEtENnP(*j1aQ%RT&qGmVA{A z@s>JSp-~$TOjRn3!m2T;V$Cmgj)CG-isU;%nF-zgc0{l*x-QHf9;B+HlJn6@K6efF z1D~@)xYHs>u9-xfk^0zA4LZ}60@K5x!xTM*OVU`5;uD9%?5(WM(I*FlA|)`cK(%Q9cGbH>-Ry2ATodyoaiHjt5xg(6Fg z4A~zdLAx!rr-V(lG3tm>7Qn20)zI8axi6O>BydpXPm4Mh8Gb@2*~(cR#^n*`wDf3r zB7I?>y1xq;lk?5b^gr=NTDixwd1V~gvtOLp9iYf`7st}KMdq8L$0L16`|R=5XTlcE~;x@B)fBVkx(%ek9eVO{8YNEJUTF$q|gs;oJqcuZh3+FaL~T-5T6 zaR2^QIDd`Qt=235sFdB5(3`hHMaM!GjUc6fiEiqN8kY~nt7-IlZm5^hHYe__2PF^h zo@RYPC=L-9edfHD&oqDg_bSDpr?fNSHeL_M#5Uw#uxkZW$dg7qD|?O7nSu=;hO zO+Yya``VIkK<7D^U=egNH^Ujf=?S;vamH4-Mkg`Nw?9p!(XiXm zI9u3$g;vuA}t8({jUx3A5#Cid;!>CBh5npMYI7T2vPFT-~+ zl7-tEmD_f1T`nh4>DT_$kI>=YIxM6F)ET z?d4ovYLI?=ih(5&h=YzAY!{l_CgWD=q;hK*eoa>`u?hmqr^OC)bud1|x8%=vR@VdLvATB3DJIHMRsx*}+)yau6 z3vsrv+HK2*-YMjns7_Y!zTTt-YB+A|cn|-+L1w%F_znxSHot9!Gi$F1Q&F?*gQBu$ ze#3ykb&pe<7faI*b7;JNl``C6`9cE?%M79A`8u-g>m9*yyEdl_TxqZ-ka*17u4GbiIJDL@Yj(|aP{k!1$TUNDD|5Cek(Wc zeSz8@qg-khnuD!~ofj9P8zFsRHhMp*vVZ$EDXcAnj~PWuOM<@M(<{<1Y6fBGnKIvt zG(tyRoP|83oi|9dxuHE~cc%7{)YA-3{qe_VWf(bP&i*ny;`iJt)bTJtIWh}#S-T;n zIzHCWYBJMpg;_BP0WRZ)Ga351cG4p4MKM8JWb0rPh7;Ss?)`^>yhZTFm;Gdk^9%F4 zsT=r^G)HCa_isC<6%%isI3Z?<`q)V^O?Z>qv{ej=DH+(QbRZBYS(9oh zO`Lz2qI{uxJY>85G|q1j+s>B|ST`dnlPDyLTM)^IFy(%>bL>=Wq}3U-VIPRFNfDS+ zNEnND*O`#59B}m`ppcVYZt0VK#mAY5*D{;P>yT5kn|6{Li@*q1o^6^pg(((tC%9pb z$9eBAR};2t&X!bL+e~im$8z6|)u%az(mB{Qc2l1_d_B8T)hRIst1+!TX+z(S$=v`G ztchoMHvZgO_S^ZyhW@s$~U( zqQ)lOv8OM_f0e|`7V`FH<+<4VS@Yf~ThgqvVt|f5Y{M$*a4NH^! zUAf*NJ*Z{!HJRn7f z`uh}wu&$>hV{v$+gI#87YD}neo?3!Eu0njwy|(LseqWakDMP1iR|I5X^Bnh9VlJ_` z1_L){2uTLMCE5Y^@lPbDgD(4xd54$KgGZ=wU4o*o9!*Zt@Iy;2tEBSpD0t0#g@ee5 zD5S4PrVB=S`_tl>Kx_h^>s1TO8CW-y3YTrjHKFqpu(es0p2p|1yKcd0;Fgu}7suFG zUi;sg?t1Vb#ufMMZ3e{3wZ#MnC4E65Dn1LJhLA$ktZwH!Xbm8KPhOO7`8AHz6AH26 z%L~pNa{K-kK^v_huBsUu?t)?%R{Gipg3mM|2)Z%F_QlNVQFBdR;Ut%Oj@RVKJ#j8o znFd@hM)9gvFm@clq{y|e%I;KrLQ~~jGYITL-LeBLT?E33i&-Pm6{+SZj>*|#m72Md z(~g-D@ys34v9?wc3TTggQuMaI9224$19v#u(UN zSCr9AUS;K_i{RUIBoVciTI2C#lA#>#vbb6p^ipJ?qs&jA#*!<>8mN^ZHdcH+fig3V zH17YR`7x;gR0K|$Dz&3TP4xjuc&f0vX)cU?{s*Id{O1$XqDt6{#y5_!-ZlFgt&7d) zYCE<^aZ4#x*zw;;TBG8NzPBAHsoXAAq^7|mhLfwz}w&*+kD`&#xhfWJ+GU2^Iqr<^JXhzJE;&qdDN|-#FMQv z6q(Ci<*L|kd8hi-g~FG$bD{&Kr7GbO!|BUuti2UWbn0UK*x7MYH;j?YDc8~Wu!+|z zY;K~5EaR7{FS@FeEmJ{<7bWDk{*nVwekybBDNG8-`I2&8*!kC$w!xaUt6tKy6$SI* zo+qpJXY*yOp}-T4e#ImBCGEU3z($~7PS zSJm?bMN+zVhB3L}33eL`5!1LQUh0YW;g}Sj3Vb9sYh7CDv`ieFW4|Q5+MNd1>qPW9 z@dn`-;f= z&EU>?5FGE2V9n{7gejU;$ep!C-sPK2;?(Aw%(vwQ4C`;D%@HuB9J=ZTPLj{Z;pL5g zq|Vr11`&S{*@|e>h&C=5wPGDZRLnxXpbL38Atxz;4`?M^AiS_{U4n_b9ncUABndYj z&QTs5!t!%D^#@iSit(mSyA(px}BJ{m{evb1TzS+F{ss=#Eh^q z^dmUc2pe`7*<%qf3SQnIsjB9jZGM&Kx|Fxawx9VqJS6ettwD)ktLKtIOvA^TQ9S(j z+26_~dQOcR3HKOKG)W+Iq|i2?2okJ+ssOI=SF<|DotSwR25xT)XYoAk{`iqoO_|T0 zeKfB3Lv>?mqT^ z1AmdZU>Wk#qj%+4P^CkMxKIe2lvW5E)E3$8l`c?$eQjwuB<3$ zC-1W8Ev|cd4(;@N*^$o~D+HDfSC`NmeQ zDlTx;ZCuap;e{Q4p~`G4BV4mHHM#yVNrJ#IgcEY#E8M!r8pnCn z{l**Fp#J)9gMeD;{LU#htqB=iYLdXi1PNj;7TRjbilV7DxrDib+v5aLQG?uWyIJ)h zk5(~nHp}R{Z2iDGqh^C#VPaOHXb5+2chr(FF@jKt-oM z{tt1@r9{pm4$zQ2GRLn>)EjN7)qxz!bkEwcGiXHT`&;c)`r= zd7?2Ts9C03)?gaY(lSJ=UCnI~A`xJQQp}vmeM)512PVHQ;jUs8wAQ6T<3byyH8x+( zDoP$;&Oxm#q*NEybzg+dHw96RWeFTE&Af@Rqk!r28I=Y|8OogJ43hGTez7`YumiP` z5=unC6rRZF&TjTAndFFl%{7eaacf#Kb(F)}YUsQ_{(PkOu9CJ>Ni$9bdZa+TJgNYq z)y-r5jmP=}rQV1%pQ`RWYE>^w3n?={dfpITl-Lp)=vyOq1c$>tEh z0?qnz@LMZn?gmH-%7s+!BX$I5iZPL>Z8FjuFrrBNZjw<~AyS)q<_5>)`FDo3;E~RE zLKi9vCdURZb7~ivH1a?+mw>mV&k`SZX!97Z}o&~ ziW4!>TBxn`%#13x;+q``u>uIu#nN zoZi6VP-QETF3Y=L@fXgn7ifbG3d#*i2; zRM>H!KA6x(+bI77wbqr91O>XePCAZ|siTF9*+3@4fDkm?qRnT0&MGO8SCm z!;h`S8cFUT`RMr~G!i^%H-guk)3jf2taxtDMMpPfK7h-2Zyu+7(~^%e^CGsoJ=i38 zD~=voSppYc=&|JTzGdO*T>NmbYCU$mN={YFdG;PaaNzhlgTI2Z#;3$@lMaDbH~ZBd zD{c=i!d`i~&EvpE1QlA4cR_raC)FJq5M2BJ01ZI$zwlVQ$=!deyLNla=F9-=RG}A* zS|J665RW-pj>&}G66s2v2J#lcKozLq_cS>rjJ%bnh-21LX&`V1ux%w zMFQH>ep^*oh;(^YlA(H3OP<_E_a@;tSG>Qyo=)u4I|kWJ>}varHFq2FQ^!L}!YGYk zvkZn&2o1~t%|c-_{%p?1YA%6JR?g?@=)JytyzQD+9LWPpTSxj~#Q`c_c5s6~V{x}@ zoMY^`*EkuATyesf{n-P_s(um%{0?3l53U0zMw8U-L!syc{iL?{gn7~AQ2ku|Uf5Jo z_=>|at-{NCE$jo~$ACGt3|gx@kBwF~F0o*5kMf&F9T~JFpC%r{tn?Zh2X0_qI()e$ zrOL_pE*1P_e)$3NP`upqPFQK4;kb(spA$=*EcUr%Pq{CqNHDEByIK;LFp#A@zVUKp zLy*0zR4IxCx%VXat;yTAr&{=`t;CxD;!R&C0`;&gmrtT4LYCU&F51lWD-x)u33hU3 zNeai|@^aEy_;$=kZg#R({@GDPQDLf8;0$|52{)S#e%fkhC6>&`Szx|Fk46$XaY{x% z1y;}f6OuocA`*b_YdIOC(>!^1VFRyxcXx#0I05HMtuw}=@H1+gj=@Vi3S;f&tXI18 zd5X7_K)y}VWjk_>vf2p51;wv1isOFb?j+R%ACb$faDQ6tY&>;FFn4104gBu*)O>Iy zT|-t=gd~UaNrH(-osWk_+A~(kNvAPWdp|}uNkN@l?(jQ?Fa}0klM18S{5}B}hTk2Z zs2<>ulbj2viF%v7ANv(}_f_ZqSe5ezGK3KCPdD?bPy46uNPRt$icgP>s9(K>i9olE z9|F84h(QhT3$ACd*dO_-fKF{^ary2#-|p@B^M4PYfWQE$V-zXam%ly=PyCz+?911> zd${_8=!NgutMl=gbCSA`hZF#wATZ7+DQ$<3H(9cz3G{;(w1_~E1mEMiAM08OnBo*E zccslt;p}{2rUhMQf4q6C**^m*f|IE5bHYs5pAkS+7MzWs!jD{*HrR=dXLmx?4-ziP z`-*1@A{hN9Wl`|kbovMlN$q?_Alq+8uVE2gq!;Aglm?Es$K^#ZD&#v&wBu%>d@ND^ z;*3%UaXCKQP%iy;Yt6=&xUk-mCuW3(Kwd9Yhb4irL{R9ZcaF;sQkTuY^BPW&5#@^9 z2V4jej!w?_berT)FQ=+qw>{bVY~Q4+6;j#<_0-xZZHLu#vba3^hWMOPpJ;`=Ad&45 zKKK9VW8wbkYL_w*EO-@T5vYmd%itJ0E$uqfKD3(jBbJCFl7E;pcUdTZWMozAKBt&! z=_Zr3D!mizoaaL>*)NlKfYQ<`e2B6Ay@S=lsD&eh!50lSlP{W{KKk*^{&oHAJ9TZ6 z>l85H;GC<8@)mV%Y;e8OsB8D}=t&77G!6oq;P|*F4C{AN_=XY3g|*AQHOKcg%p1p! zRgJWLe%mnV2Sqx2c=#x$nFdcYUfZ+;Me*T_{E2Vp=zNj>;^8lsmqc>E-hc^Dep@H% zW&0l0#j(pLxxyRhDV&1u{DVUr{;FL>X}_nVFMcu4`J+T@(fZ$}2Jt4GG}B7EPXANrUUFu?maI7)?2WKYr=OP~(u^`9E==#YMQbgu3?= zKer^U(-^%s=`L$CY3VU-S?jZs@vy5Ys+SY{J6=Q;9=9r=_gOq7&!zB@OvsC)hwIiq zzgrLP9BHG&dmmg!z;5*H?czT8`SHn&SXn36eSSE%19`O_+Vm*ljp5}~-dF3z+qV(V z0$QY;n7|#6&V>}oT-(&;l{zp3-@o;Pnp{MS-S2m+}8*HXIt ze=Vil|JPCqrW5|Jr8JHpfjuAE#xuri0Rk{!S6)OKy9z!Fc&}nAOVAu;KlSVAUDO@d zwvx+FL}VdxT>syRj=J}yDtt)AKP@$jRnRP9+0!5E0>!fFq8Jji=Lbl0uII<;&c>NY z$!BXa=X-8rd={a1M;=u5Cg5u##w)$N{g&R4dw=_`0uN{i9;C0h`c9*qny9}YKc4Mw zds5)QzXQ>K%WpekxvRJ{+jF5hKu5CsF6c*;G0S{3Aft6$ zV7x{!s(+0pSy#Cp+^AK#&vbK#->x+5pnx#d`7)+Icqt=6jgx=9I#XjK&5LIA+ z`m{Pzj}(ND0O3J-5D>h{x>BMgA!^+ap^Dek02hEqoc+nOkzNHSu}Di0;@a4Z7otg> zRm*505RJnS%(jj)+KF-vU824Q0KgWV#1ON{x7I1q#?|CCo|SvinwtcZ=p~+_Z{7Bi zdZi^`fML5Ar>Z0-HEv19-gCLO3OA0!=M$CK#m}6r+vO~RzXOHmyzy*gIrfoQ9VaC3 zADgy2AJzL8AZ|j!G|kvPxw%Y9J4mS9)J>mBY_pjKauTorqHrV#k;BzI8n(~wjWF(war$1EW0-L<)Rc#O-r8_355Th<> zTp!4eSg48lmhv_r_sUZu@@muG_0p%`l$O2arAD9lkA97P6>$}{G2=%nF9yl7`>#5) zcnH-OS0`S*n-E$s5z%!llTzDFHx`e5bKbk~`~5uOq z25|wEvG+mhQqH+CI(S2-@-MR~{lIuS1->@lnJv+{L0Kt}ug!=@vuguf&qRyxoLPsQ z4Dkf=G=;~H8S(w%C!_eA9RYRLm`t93ncYp}t6x)A{vPa1E3jGCo=2D@|HP?s6^Xe=ck_JjCrWf5s&I1yTr)TzFgv z1Sav+=r$3+1NU5Vb=0B_d>fLv2W^6)8#s$mHhiz*SX59;`udg~MH^~YQ2YVH zmR-}UZ^Et`9VtY1?(me_3a^^nuO8d3Pp`f0y=wMNZM6Tax8BPLK?RQ`XJno%ukxvq z?rI1CSpwEWvR+vsbB)eSGnjG3x2^#~U;c9H+cfIRh3CwPnsoiD9QC~JcWr zUwhDJSbAPHbHRA_nyR7vqc&HyTG}RV^ZMc)I}$6Q~{@c zoCJ;C>+vkm*XO}opVC!}Pam|r^76z(wD2`bWiB0e_I28DHR#`3IWJ< zp5Ff%EZ}tGnTjWS3jq;u-~ZZ5ti=4bmA8?f|50LPih$Q1dNbc)2?OHtpLKA)VRm5U zYYc_yABC!wk3KmXC;MHOBBrQ-vldVL`RZ}e#<_Jj$5-a~M{m`nGw{keccx5VtoaKn zIy6^WhSL_}wNX%9^t^wNc9455+%5bY)raJ_Z6nv&kxaKHqc(|?mi;#G2(==0X?=rg z=_BMX2pnDE)e@RiBeh*-#c$#weEwk3P6}DMuCd>#stvXW19@b;E28x-0(3Di=|eCb zGEKU}$~4Ao@8LX7rGIR)^8(pI@QYjB`ZRfWD2@gW-Uj!yq}oOTdmD?jk|bku-=hbw z{jV$&fL%kk(A{T!5I&E~UldjGvjux#WBE_i&AwI8zWvSU5dFWxKa=dR2h1P3#~;l9 zObf>za#8h6>KOgXEW#cM;*5Kyt>%B__F+FcjB!lgqr52G!XC?i=$U^pf00vle*A)M z^24K2-n%0U$>u)}#Z@D|vCMgr1n)QPZJ%X#yT&ekyEYQ37k0XRw~1n9+Y=MD3ThUv zW_Nf5g$7J0uNTA|Sl@O%V_ZNAi^f_cA~kvFGhQrQJ*G`Uf*OLMf7qwRxuvWo_uuQ4 z9%Q>@zWP!`e|a=C9p4)5>gYYNwXe@Zt52(AaGSt(8v5QNDmP!v;k_DL#e6s?O-A{1 zF|sX8y@*b-{Sm#wd%h)iTdglV9FkQ}hHvr+M>sRH8BZ=S4+~qauCvHN=SjG(PdZz| z-tN6F4VW-@Z_k(jyXQErrrlhp;f>rNxvrFU`Kjl+>TVxE+CN5~FzhMqHjm=OCGaa! zh|VGs)+V^1ff3%H#MB7DyCb28AaXorH(YrbC_1QdIRy060paLIjeI#G^_orAz&jF} z-RbjX0-6Pp<)757k%=Ivl|0cvgj6tVMKa5DQWID`5&6X?r5?n*1WoUF0(3v{(hCr>21X^OkmQbnan6%M z_ncy%S;w|4QgV)|Q9F{6qgjuTAn$$tAzG$8RW%l{l(89oQo!r2faf?#rfV+3YHCrA zYC-QzdG8_TUEEXYo!?HrmhoBOXV3qzi{nRvXoApPJf|*VHqi`CYJ!U zG?l=z)0V=NL*JZ`*b$PTuj3$^n>^TN&r1OIHf1<|COOINy$`ne)cs(3@4*k*Q0l$3 zLfUi{0{Y7n=0paV@4CoxeR`8qDz645^`PzV1*FEMI)TtVLWL-1YYT!5BxpZ*x*e%X zdKnA1sW>c)NQw9P6uR0@d zy)EyB9}$5n&)^bea=eVjIU!D!buBVKJe2Bpc^cV?Y^i!i{$h4YMYMqlUfvW4hAb#b zEMQmwf%a$~C&VdRoDrU>~8hK+8)|?X{EJA_-}y0 zZfs%$hRihca~DWsUZGRUNIwgDTK!ztqF;0$b=F= zu%R?$=al%|QG!EB#WW(+n`KO@rCeID8xl4r8Wv(jnqQ~XZ+v-oLIIsUfY5~Rn*t5x zIY@Ofh^g0xNRdW+0WkChwD$yvk>w1|_a^)V*11`z_COE)NpsX_w96|IBxWZHu7}P7 zQbVN(HQ?DfrI8{H)j|bxh(L*~3;w7Laa$Q3cX5j<-tTi#Y1UG0JwX=Ua{CGD%5@QO zPhd4NPYu810jgZogIc4DnsSp_+>g&-0k8n@>J|Wo{W6dTgI!)JJsqRb>KS0B%IYx4 zMppqOLtfx$Dv9<0BLMFQb#Z)By%&mFN=rO^sFY3;?~yZ#KC+6vp@!O*&q$({oU1{` zvpT?rp|HOCXt@5_KCk(7MULM6C&Hyt+_fzxpjgjpW)}v6!YaaxTJ2UKdWn$Iq^i-g zt_UOI^b;lLhiCGOEIN#j$d}L)&$}E|Pd8HkV2InqpRA=Em2E4*eO8LtZrtPJ&|%J?ggO~b>;4Gejf1<_FdFX<8d zKhoO+w*NoUBL!9d&*wpEV*CVXTjf|Xv$Urs#F^*NI~>iy@@4!bdJI)7F&vGaA6b7^4P0YUX!rgmrsyyt_>SFcA zph*7aH$xDWLD3Vr+l{&XFT_GBe_nmTd=n_w##5~=h41}yA+zqM=q?Kdum0YJUxQBe zVrgn7N>$OH;`d`2DP{6iy;iLr$Fje!uT=< z79I_Pq-%>Dhe-^i+#l+?WxC|rJWKI>d-_Z&fJm6Xgu1@xbF2eVKs6+Glb5f+O)f<6 zCA_!5K;@f+>xYbJb^3#9k0cThSK*&!qtckDq>8d6sZFD7CtI9iCJ1G=`m`uneSwk( zp^h>!FI$3rvB{YiGU18wW382PB5|EbAL$)OseCiBGb(kHLtCK>Zaj|abr^;CJH9lt zWK^L2Z_Vw9RtX~eT6kMy-K>WTnVV z31+20XpFJGJd<*e&G9O~uaef({Ijc%B=`d#2X*N3OPRzmUZy`PG?T;3#*%`*sd>>9 z$H3*s*E-AsjV1Nt7z9IAbom*ZYIEi74HIAEExFbw%3d>uzZi6aiu44x5N(IrjtcJQ?$d%d}9Hy?%0N;;R7jPI6D2)ykUHs63UI@jzQ z2Vfh-H}bcSYFy#B7B+;te^uY0&GfjwS!|^#O`Ctg@|&T}$hKZmi2Bp&shS*Lhr}g+ z`tiowwz}xhWlET9tDEA^w3{tZxY&PyBteFH^>y|a|Icu$M(jl57a4HFc9u00{VeD^ z*R*YB`=HFDKds4j-{p7Zs<|odgh&xzqV>&3H7SSXg^e#~9AUkey^lIw*boRnA?>J^ zEJ;F0p4&-HlaIpej|&@@@*bI0Jn~b`XmxxQH@ZYwl}cETSE3J6sZXFHn+r5%fhE1iKWHyxhX?Tw_6uC<*yZN-QVKA)RRRYBq8;zRw? zn#|H!l4BoQy?#gSKsgaUSX9k)yK-yJvkZu6>PV5%e7nF}U(J1Q0sRz^f8Jp63O`6a z&M7mBYm9wTBKnl>)^Dz$j3$AXOuezeFsneJjZf~g+`Ux!OvDx>ae7#HJ|?oGSXYuu z@m1wqQ13;Fk`ueK4*ql~KcDNhCdu9Z;qEP?qI$gc?=t}mJq#`3&AbdBGS?#AkqyI3aG$5e&hbd|BnCj;$F{-XRUcRYn_>OW}m%3 z`?{`k=BOW^Jll_bdF>MzURWI^wTr-NKF>KW{_M@C zg~s(w_JK;yxJGXSGdiwp|nF!E37p z_!oImj@Jfi*|zMLy_4j;HYY3Y*G6<#2tM0QK?~NnTT$v@Fr{=4=1*@tH%f?@t(}#; zM&CjHBRnIZEshiDvbZ3iRvk0D2YYO+cOqHovhv=%5j=w&cc(@c%Tu^5x~r4i`Wmht zrpPf&3SUM%cgq}T5OVB#myGQxQK5yQhh};uRnRnKH*Lb)^{=eXs7UqcQMl2xFwZUa zdJx@=W(`yVvlCW_6X<_Q6=J|`8)hUk9Sw$0W7CM1^B|?qgk};& zXqW_Bgvr;hI>g+6f88`U~(PXS~bm^4iqd`ZI1 zytoLHYX*_!x@(5_L}`T~chn=w3Im6XF!>BpCSwoomqt{<|!$(osfUtu9fH}_M%@eaSbsLf~G zyPrdAXwl6q+pRuo>+;4MB8CA7dBOdD!Xj3&(GBDf=V{yHV+)dtt4esHzJ#1_kd$-U zCP`(G+NHM`_ghvfJE1{`jT{RS)OuSR0TsnQ&~7G=;^1kkv5z>hn|Y?##dj|LZ2#(; zf$V)U5gb9C> z>1o-M8gTV(+LEqc9-mR3UhQDdULtJnHrE2xR}yDp_IyAM$FPabFUNy17#p67#(0V3KV_}m2m`-;{Zhr^ntgnilc3?Rj8O`XwkJ$b&P$P7*&)W zgOonyr4=Q$18G$?@YS8v|6{m6eMBb_>~fLfx@|<59%ZOCZ)GE^n7kxMzNquM>Prqq za0Kf{qJ&P~vqr2d<AUFAmPXeNjo`CI@T_akR)dsWNS2%;^dqlQ1?{BJ zxirvptK56HZeLI@MhkLk4bvcAVtOv-Y?=zkNJfYF!*wMQ#t_3Q7!|5HI@sTdX({rV z8tUn$4!I{o@g~*1P{u|{%?>&?DLvA6k(O2Wqc+8qogj!OGY=;H9obFdX42wjo8#tZ z;ua+N8cuaUpnozc#cw7Z|1aGKYefLgp#l}Xh6W%i6pr(KdIb<3k`OZJa&_U~?gL^; z4u)@Z`Qh;O<#5L3E5T}9y z`{?%2T6SVNG<_X?Nlm+N;v3#Ye+P|%eR*P7wF4IAN(`*^FMPg+fFJ}T=w_d<)3oui zEH4&@rCrtrGm0w7_fNj_!7Ld1ikLD4p&cCsnh(!nkXA$bcYlRlIEjY|FiNcBpUQN! zKZ@XLa|p%rf6-!nn-LEaCFb! zT#hC7A?T7Nky0&GSIh93)#{v~UQ0PSv!QgYWO0Yu@GRSh*@|DNF2sTa`RD+UQVJ%6 zitUdl9rzcIndkIB!a;nVOThvAoG|oD4F7YSW9I?mW?m$BnPVh`O$sX#qhe|q3%TIi za!Wxx*ec5M-QId7$d^uEn45a#u1`>O09+kz=d@i8cfpro%ZVExI~BG|WzHJ4h*e&# z0QqOw3NUe7d1XreEH?3dhUtgbsw8%fwUBo+1G&TYOn~}pja4Bf8c=m{-ltRnAzy=r z(2ACq`p&vOfR#j=b|XU>e_M8$DQ~O3MNC!8;nA-Iq1TsfJDQ;5a<{vLD1)dr_rK&GLiO%+yOi~%J#q1R z)tMtb<^J}0`*Ge0v!VA`A+QuLy*INGFRLxbazhG(KG>~OR2_Jg7Yu-*NUI}bLGCB2E{wk~_*w0<83)w;D>)ew$C)FE>_Ud$me|bBfs>^$mRFg<6lc70+C5B*Glr2KFGv! zWbYz~ZavGTV5?B;2Ay_Co1$~+8Uwz7>C?y{rn<LC;Opa9e(c4_PsV7r{>Yg zwI6Oe)o)XNei&0ArpHDMpOItV#L}78aCWoPD&*SYLh3z`P>vyY7ln5Oi6MOVz6~63 z`(RP|&Zr9AkgfSU*cP_6hZ<~1?N#3AH**dZM~DjsP!ojmw>)pUKIVW0BePKC&v*(^ zD_jMNY9VJHHw=kaSylP2Xm?+Wy`iVcvt^=eB&wEF+BHM%dN)Ihq6hOmgDb2Up<$`x z;Pz@aUnk{LmBp0_8ILRuYqg~z2b~BS*X9CoGmrP$ocYXaNFvF%pC~+1ZG#rV04Wn| z?f&_yRPZAnDCbN=Wn!F*g{Bg6HXVP}56aE}_%Na8=tj|6w$IOL;-tA>Ni>0Mt`I9* zq2TYu3Ap0}P|=%qLrtR*{3CgDqr3%Ba66}qEp>k2)n#DpX5vXPOCSdkG(&tW;Z};T zu#on~EE_p-cZ#>raH(UTeoO~Q0U@egx(rb1Qxf}#;xxl5w%Ou}ScOY&XUED}4K&)h zC?cYCH@MPkl7il5w-Lcc4Ah*0x;NVJfL8zb5!3NI8XYwkokT7cKO;(VSL5+S4 zq_~iT(fDyfnEK>N&`_Jul*|%#artOHxG!J6OP)O&iU0;(Dx8?{Pd8@N5Riff)8@_@YMXcuixgs&Dg0aMnjgzMwv0r~uirdv z8I7y-ex><&+w`<`X1LO~Y4G!Yz>Ali7nS}W3g!==7PYVYXaq2ce*E4_FFKPHq9tg{ z2;HnA0e$_-Dw7g8PW_99QTIN*l2Z=9&4s@V&!8gheWgr z0qbCSAmHmAeS~SSHQc}iiP%JSB2Y{4_s8AV-8vByW*ih%Jl7v2R!5MF+N5@I+E8g? zqF7?->t@bu46P622al3iA-r1X%lEOvW@5E7lBpbni77?RtHGSX2&}9QMX|<6DqN*_3C|T&ILTZn zQ!w%c+*M!ni;};+;}(pT-1on}wmDKkGpPf`Glp2(oya$$z;<`!3Rvc2 zUux<|A|V{ducs4(ZHwV#{eBL*~_x30V)#iZdW(#SCvuFgR_=s zWCGBjE4e;`w%jnO&ix3>w`rn7>(qA7eg+uXCzpD?*z@GW(KlO>l3yqZ-+k{MQ)QM- zuH8p^ic}PE3(k!@SBt)kGFh6i{M-;kBu=pzFZcK`+RWR2-$x=w%oQBgu4sIn@&1@9 zuY1hBcAH%*k&wj+#A+K%W$#1Ov@0tfL(slNYoDtL3X5MYtxlWmTU++)rhdUAW2Kox ze+wU>p76PBPJN%EnOZe^6yvAle>9=20LA74Tx3=N8BzfhT^$6eehDMjg_2c6NvxCz zBb5NG>iB*623`3|cTz> z0Ea!9xI@UJCavm06R)pq&n9EZ9n!A?Pgi&<1yhzi3+3x_Yo5AaM-avv9IC>=XMLn5 z>&0ttO=+NFCvweJ&>zuAZ(Mdnaa7YPnY zkc=uG<4xg3H~2)wmPBPWM`fQyi@aSQuIoAZ$Piy8JWJ_)Wk5u-$|%S|W+_9@FXwkDi~O0ms@E^6o_9)eu|&I2e9 zJCB#cLz<|BYsCqniCiiNgd{|uyWppIwL}yfc2fcC*#)H?WqB8Debt)?dm3qfbVNzk z5+neomlI7)6pb&^PmC$#=sJX&bUl5)l)_AvG^@-f%V^JWYJ}G+Ibqvv;p24(?2pne^7Zei)Q2NPj!ycmzsl(6?0AQD+Ae@O_iV9{ExS@ ziP1Zj&)|3x@m4xe*gYyLO1DhSHC^d%+OneKK#ct%(x3 z5EWQ}&vsq~FjGg+HN89AgE&f~Q? zgKyB(m36q^GZb_mPyc0pdTn2V3E!G?@kn5Q({I-H%bUkZXX0jxxs-VqzQL1AMAZH> z--)sxf!BFP1HQKq-3t*PCdv*bBBVlw@dtA=23>afu|;sF_vQrCs1Ws-76;EDLdnHX zbE0o#&n(NG0N?`6y-^h=D0+S(URI8fcon-k?`Ny|6hs9>VNm^?NNtsVJ5kjEXrw5P z6RahA*X`4N6h>|3_H#kDX=Y>>6cU@^0$wT3bWBQTlgGdQVV~jT`HA~EJhl{@NbscD zK!6p*pJd10(rBM-uJMCAxl1>NU0xq9#%J-9k+Vd_r;$%vbX?{q_e(;yo1;%aZu~q7 zm(4AO+)t+Q1#1Nu>@2n{qEZ9<@`-|#gXs53h+bU&QNMr!#(wr?B?%wNdUEEBVtk25 zfFWNU(T`7C`EzJoe2Lra0JgTwMsqa3%LNRQih8W!Ec$}gu(G9%V)b^p6F-M8JA86t zwczK;&kgjzxWGgC&FQx~(tBE(&V^(x1wfaxl<{Z-A(0bL3L3q&RxmEMssRiNht;&a zMvv+UbS6U926a@nJTlOYbqB-b*XKvX)mKCDFj- zY2(LDG0klc;dk1O5fX^5Dx;_Y{}1Y4%uC(Yj)JjDA*;zhEmfgs2D9*V2;A$M z0@m?GmJ0U7W<;avBPR36<106$^d9h$qz;;q5abKPuz3;i86|;6sxZ0{x3B^R0zSNo zJfS|rVT5Xm0$9!CE#tB{_9K1&R3m zVo}=cTL#$g+K9LUJ=?2Gef>HCN6EQlMZBu&16kO-loG8Xe&hZDqxhr5%2+NHGCs`+ zwWXAX_DD{zepL;sZRvKhV$}Uwg5Xilw21Ee$|-jTY$=lUKV&O)Slye<}R zQrC_>i01#SwMIjt&-k%A?|BxJ8XsJN)?HjQ_4yc%_X>i*zS2|cEDTLEB87*$i;JYL zm`A>Z5oq4kunuW4$04AoFpxyJ3pbsuJT(>u5H^ zJQxKp2BKT04`I#)$aR(A`Gmn}nx5j)b^_=dt1u#F+$&*O95sb|7=+o1Dx_~Mt>h@` z<(2P%@D0rEm#wk#XC&an0e&NcEz0btY$YmBaE31fCJU!MOQoNnb*FwBR#R}4;v~zB z=3tP8e>intv|%umE!FWt!R)G4gxS1ewbK(E`|B$ogTu&dirQZoTR4RgnuL%g z)qqWYk${qj3t5)0aG6#e030$))TdYlf&C?dT46{fL3*(x)iFT|{>l>Vvzlkt?{7w- zkt?s^xdU9x$SH0UYm(09_X0gmlOHV`QX^Zy7B9Ck2JK5Eb`=q*an8Z9`0qvZ{@o^5 zab)GuG50FY6{yJ;$2qD$mh9fILPDu#jf62^WnQ#l-oTf$D`>ca|31G#Z#1|mpO$(v z%+>NYSG9+BHMzeRFlUp|IWJCizN^5P;gt~N&2V#BidRNNg@iieOiIJw_LoTu9Upet zpDV1!%Y10CK`poEx<*Je>9*WO$+V}~x(APiOy2IM|D#HmYw>2@D_0a3dsC%ri#4`Z zxCGpL3r%ud{X4*)7vBdC}&h2&9 zOPS%KhoZ&zCf}mEzqxFPKB0Wu+B#Jofs@7_XmP^|Ng5-7N;Y&j$-xy^VqZN7Cc0}a zn#)FU@J*Z&34bYW_GvV}vHq3zo|7U@WqL%MPqkvtE%4XeB1MzEg3!Ka=CAopou<~d zw|lkc9ER=)cXLRSaWG3|>)`b`q)FKD4ja;m+CS|9hAQ3Wx%zX5BqNPFr_TdZOC7qZY8-MeV+b67Nqcib{2#YKr}h8D#QbmAWDy zer^NOs;qM0g81u0Ee$&r)()%;2kZCJ!4yOKuiL4+XUmAgdTAw38xB{3cfHxf{NxN_Y#C@_n5_JZ-1kEy9w}wxD|3nXXna#8$`F9- zN?`#gv+i`Nss1;dk|x>WSeh+iwVph&TIsqoP36{lrh6~nMwa*VgNZ(wIL0%q9V1lFT7gI>hlDl@1nC2x#ZOIBL!pK;YrXEdvoI|B90%_ zNekSx?_M5jNPQHcJ^e^hOYGcxaJ!Vs$a5$wpxOTgFa zZ6&1l7AyP~R*`rEN@>@JT=HN9=b-=;!&!N@9K*pXX@Q4Qlc7?K-F~d$WGbi3@xXPd z^<0C>N5+^i)}NbZ4Qnl1FBGv-+oq2!Zz5(L5!zctO>X*A)%4#MxBFcnpv7^_{!R7{ zfjd7}_hOSDthIFJ$mn}iP>{jVF5oWlv7lQ|^k~2ct!1>!@dwFfMUkJUDi^mNiLRhg z@!aoh_iz4EUCN1ln1Q2}{dQ(XoU%CHa#|D060~Bu)=rPhpI3a0H~6v^GKNmYhSE&i zW^eG%By|&5U&+LVaVx2HQ^o5B&Ry?cqU*jklTFUH2ZTgRp22lKq1cp43CI&g-?A#? zDqWw?Q^31LV#Qg8O%;@22y$=_r_)+E*%C`p?q(0a>0z|VP(EE#-#g&|J%hua`3&#X zdPR@;97z&w7GF0gT^F{)KpreHNO8RvS7EzvCkRmpfq{(9){%~6NUGZ*H-x)Akk9OF zwBD?%g!9TL#D$@G22kocy0v6kyy0A$*9f@W<&mZgYP@xl5qi3L)Req{@_F|F8&wIh z`C=Hl7X>%9za|Jj>v>UMU(~-_3y`}LFDKm@P?%7#s?x|u@YYI8RMe+ZND`Jrl1JD0)1F+GGPmu&RWx)N>+ zKO*w9D2w`9H-Q#g#(T9o(0&+?OqEHN8G4aK*%1JZxKmQpFVp3q2C1wjKcv=EJ$UEA zEy1$LY%H@uyGyoy)w)|>n0=k3Obka*X+=nxZS#U?al^Wux2(#%3xx3qfjDwfe+KFb z(8u|Tow*us7#2y4Q3}eqAJN&Vhp!&9jY48`GwIl##|RVxv@l&v?N5TNzwUJlG^ur$%)54l&f@6h>eH$nOUFEkEmG}C5o3`1s4*4_ZJ|TAN zYr-)eLbkKtpB%O%hoik(c0OOaG{vf80pQymNxq(s?x)Yuq6jvXyvTA{;u!@2wQGxD z>Sh{OUS)wKRJ2N&ud}N@2mNycwenekB(uJ3jKYVfzuALC9hp@Z2@M?cE)|;p%J^tT z_nofv#!4^}Xkzqj_GEl2K<-MW0;F|(oW;6QLK{A^(cZdT#^b>e^muw)lIQx*3wrJ}l^lAjX1 zl$#Jsj~D}9$LEwK-GiMD73P#KV|N}Y8?d-17>j|xUnS8GD6GP=NPkY18^!eF509f3 zBVb0GTFPO!PFOTDoYjn5A+gAlLdDPfX9cZMl_SY-oE!ECHu;yz8IBbSvcGCKfW3;+TcmN34?%|pew5~Rqf+z@k1>u?IB4`VrzajKl*k(*X*c% zt^9*}NCCm$rwaJ+H}%l}8}%$H{9i54+Qm+SKP*qxH?OZUsRe_Ebpf8B6(N8M{; zC_~qI;iumck8U|@%#7)u;tKgalE;i*28J0@502WGfpQfHWi_40D9JiH$m)3ev-3CO ziC6gZoC}{18!8+O!8duA2ZX=8$8v~p3LLu(XPrJ}yk+%=<>}28!>WCg+p1RY!sv$X z;&6!Bshh=8bMv*Wx#rH<(3W?C+G&o?m9EZE6tSR5SYj|XJw!(Qou2p%&m z#}?dhFpaudasLj49J5a6aBZBo&@P=oC7I?r)%r(~6Vz6WZ1o)<92L{WYAQsfPjWTY z5>$K;1h-I_jbCMWPPUnade*XJ#l z2NUblE)Am=nLy(_$s-{RaaO2n6Umc6{pQu_$k8WTjZD0)@hv;>_6^3#k;Wt8Z=`8z zpOTc`5Z0K{1BLZA?p?4Z1r4S`EGVp-oU0D#lj(a>U$V=%}y*(!T z#ksyyO9WYgR<$vOTPH@kGpq73VV7rxYE`eYI`n!@u2i_?Y55HT6XO z^sc@%zT>3#W=w^+liu*&_oW-baS~cGZ!K(Q166(e=71L!3wLj*-jUFb?sQq1xJkhC zQ|^hPkHHc?Dg0EgN8h{kDfmO41e829Kq8kzaJA zPf`>~;Dy+3lD(U%Y~eJ2Brs3rAAXfR#d%+Sb2i7r7DdA>LZQWugI>SEAhtMH&VrE0W~-(lpNdNYTN+ZwXL608TW4H0%%mmHWZ4hj?3KD3s^KVa zO4L53vC8lBnFB%!RKZ(l^ZDuJnVVvvk2MXl~q zB~LKAgqC780H-jD@uY-Eny#8yB7RJY(3?%FuPfel)I9_@`dOTCQxTG{$B2c72{OVI z?miWHjMP$MpI43`_@xlU^xhU~%J>Afl7eKoeBGIbEVc(qr& zzEs=O~n95WK)|_|MenT@+XVv{eEopK5@R5I zQb_E1DKR<;?b|xxpVY-ir-<0Ep3Ph^pRyi@uA`qm(3|4Rqg6=;w3Dwiy(pP-R7X4M z-K)(26hQ00EV!agBIde!yEN3edEuGmp-VhMEpa+;HoWBi#jsf*opIxKLM7?uPMBcm z$G2{i`NYq6t#Zdr-fve`hV7-B3dPLeX7ATzR;VLpv31jxBf_~IJEpROA4l=Cuj_KS zA}y2!J{DSvfRM={&@j#UB?@B%#snvpvNN;knop}fqwp<@sbfny!Ecu|Og`z?#C#?6 z6%o~-XQ&?hDB!dZ4C=jXAGAzPnRFsecL>}lr;p|GDTac+q9--)g|#6Zlu^0Vk(*h`80n za!h66M;NTuEzt3lLtlb-&w51^39-+#R@o^vOvkI;W8-2gDo`zb5Pxl#Vlt)~-3>Ab z?DNNicVbX@7%{CCngFM@&X2h}q>aR+O_Nbw7s7lWTy+{}B(4L9K6CKOU&B#9P5kgy zD2f|4=aLTV6AVRaBK?m)2!HRpN7 z(Q(2Q|1`VPdZ%UN{ZnPC6m4qzW9^AMZ(~}?-GdeI;JI<`<`r=w$A=>OlOCRwEf(Ys zenr1VjXaaOC2=(c@8eZ@ zVQK(0|1T2wL8&nK;&*5JsajVomxjXuZ`4vtRLgm`{D<2H=q2-$EB=dm{Q|+}Y-bZ1O_iTR@OBa8q&F%+5S2pK zy9XzgOK^t1 zs^F@9AG{R7QvdCxnF)vD^!jVZB}B}U*67tef_fsu@-Gih7Y@^6ulZ1GFR&OcPp5W@45L}2>Sri}*v9I+yY za8IIsAbM$&5dTKz#%B;Ev&S^HR2xpCaM@8(s6ebWS*Uay4Cex00kzQGce&A*hybIp zIC(&jk6XdRdTJoevAA?S!=>^6rT+21MglG+>Wk61nxPCaWBWZZ(%rhZdeTnmy+zRM zdfrECy0yKj-a9WoU!I!2UKXCDK#)Ib*ei9Q6B|I=!f2pUdyC}{-&9~HMEKy79B^qk z+@%`*Ljgtjn*`(s-rBiLUO_g8K;wIKpWV)$O%| zB0Pz};I-#hps21znWGVo4Hr)o&np>h|G2J6ve2Suw2*lNi2dk2hn%*j0#Z?kjK7v%zxub)%Xu6@5s#;H9D{v=!AU^=ntQ-)q7QPM#SL`uNXjtSCY>a2SzKGIo-ORy zfY4Jo_T?CZ1ivuMjx?Jo4<^m62sau=@FR_4vuV@3|bwWSr7r4Fsdc}8(S zov64YCON?1bW)evcOAV;?_!Ddkm-cu@56eO9~F=QT4_eq!82eLV)@lqxyO`fm2tLR zn?AxSf<#qrPIY*svHa+afchmrFt5R7e^NYP)Oz&1I<`pAL49!NbxmsX(Aul^4Gqif zag+Q>E&(P4{O^qh{D+Q90XzpVfA%qg|Fe(5NiC!P7wqC!x&<298VzMrzx$Ze|7jOb zlv(^IyZG~9Q`MJF41|{Jrjp9kDXpE5(dz9iK`yQ0BxL8(hZub5vl(1hg&fJ3!%Od^W;k}`cCB$yp+@#177!J)kQ z&sifK3uxJiO|pIS*IOM>CKPm6O8voL7~fp1epz;$IL&K|9mbYHhWwZ$O3_lYO!w6a15I|ykJPY96H9KCcFw_M!P2sz4GI)| z9533h^uNZ2;H(8CR)y^HF5hz8qokZXQmFVjwOtGi=f`|F7}iSvkVechGfXn|`qQ|U z#9r$75k*R!wUTspx+qu+@`;yZ$==7dq3G=Uj2)h9SpFaRW$+&({5F6D6#g##bDrwl z-)~d=F8$F(5rCWt5&LK9lkwQJpY2$^kNczaAHmuCBe(^);WH|1r6HM6Fj6@pMO(DM zZMKl_;Jz|3P^gxO^mf$CR|vHsJE;0y`nL^8&NuaNa=`5%c^{|SPRg&%nr`)+p?0S$ z5wsyUm}pm~UPJK@)}pOeOkY9j0^&@JDh(BKa|EbPq^H^0|VRa)4-~l{(w|LM9rsu`xuALV~849+7^}`Qv z-kVF$AmC&uhW0ZAhS~@&u?eC2u6i+rFp$NNr_ZplkH&G}BvqpQu?eIEiMGM?3p3UU>b{bk-s5(}_Z5Rz8Df2s zfEsoc{aeJ@;;UA~>`5e_3Ut|7Nfr})I(_^tMSDE>^srsO*IlufMO#aW44)t+SPWFZ zFw>a?8U70F!iM2Y2P+g7D-{zeA=TAt;e46cTU5?kb5K1UsnwLXYCJ+&*^-a>U_HZ<_RsqlA6f&VNiy}wJU z`p=RA{wk@?2sp9Ae;dAGy5#@8Ms}w#nvwXc)fEnt zos1Y21~yeLM0x#68aC5go~@Dmv=A6`vM7giA(&;+JY&Dtj_x(D3PUTK8Gl#RU+oI=*Db)fzqLoA|1_x?!GAce0Fs&j zin3b7{H`t8o9;>jdw;#OFK~JHXVA<~R26FATK^Sc)e6y8N}1iz^2;?RO-d z*MJh=X4?+x@cD7y?Yr)kALoJ#4mIv96RSS2;nfOma1gksb6`Bw9>{q6``p}t?1odr z75kA6Y5Cq8rR(?wp_WL{qM>C6scF0X`s%{U;HBU#hwaYiR~ZtmcIX0|pg8U?m`oOj zOl@gBcbYr#qjJUt|{Bgb5F_#gjt6pPh4ra7yTHDH$Q1D9X4c| zcF^i(Kv*iKlt8MbB`d}Odz=K0fsc3#WjkjS*^NS{`f{zv#jZ`rpYHN45;EH2+~-EG|2Z#(D}_}cz$+WBio+kl;G6XYP~ zYv<~fv%<#9UOSXLPmP&-D?a601&tY6(4&V&Q3gjhVIv4VmKBC>_7K=pOP+qWKxg-aT(o`X@r;5`1wJV+*Y0-wh2(vy;d5zj%L8_qY-#Hf^%AyC-KZSNy#mL52hY zb6WViII%g$`K!pr(+UEohZy3NFB^^bDT$s^27}D6HY{^e2Lh;| zRY#)B6JxY9i+gaS<9mSYJ;jBPZs@G36$(=mrv$JC{v_>%Huz9Nh2gO83}`XDc^}S2 zxt|q{eR*Q&`1X7l6)+WGNlBjuqt(b~6&m_N+DQ;*1r71Kl#E9>q{XhjLoryUM2Ho$ z#o29sh(q;E%ZB7CF@ibZI-HAm3_P;;d!h##P(nh|!w@RYIuJ)o>3Ka0m7AW4dy#h1 zm|O+dilFdJ^H;15+F((2l>eN=4i1bHNUik(6-KQat3+{*TAyE6vmPBx}t6JV196I2%Lyr}x_B~qalqLV~hWGn`}OqK#@ zqwP|m7X>Vmw$nlx7+tKP)y>QHC?U6*Ve_I|Bw+<_C~Ng-S_6CjU`B~rtS$qDL7taZ zZO+z(O4WL*IQK2%mVlcF$rY}85~)zfI>T))C-E-D;2&zVXX>=CUB&J;JW|zZG1Li$ zzv3_~`CzDX<#jwFtb_;!CLpx>@VNJzi+$c!HJP_TEvt9oq+keS#uLcBnIfWaWBNGq z%7mE~sMPryPXAOixp`@|6!+87@XKtNgx+x($-5}@S@)eN{gg^YDS^RBr`L@U-s(Mh z0=l-TA1(jU|Nq3*pxyVTGYZ8ML8H+$%1C?&2@C(ErQ$PUIYMHmD&Dc6&iG5R zpF3N+1*A9V%$>fM8b;wUBo$7y&dQbY1m1Dcdb_$|#+`KS64*+h9-m>_!SiBK6~vAQ zF__*Fx?B8&g@T*J&&4KOuT&x1{Q2uop}tcjTByl;#$6datruCRHQmT0@gmELRv#y% z2yLpi!Hl!R!#3PEXail(Q)9FTI!gS)`3C&fMF{(kJ8&r*?ItEvKTxNw zaH(eVC3HitkP3wZ^s~e$Pd@4e!uPqeA{e+5VMscC$K@#6;*SGym_UJ2qaZw z1~>B5uvXeO zMJK&{y437DSd+b3d#K?_eFGP!joU}KUN8a{U7!PwBK88Vxe)=wUTlpy3LS3t%hp% z4|@06CKE&{={_j%3SMINlv-Va)u*P*4|rIm>;6GNTX1vq=?Q?$K|f&=Uj%81p~Hb! zV!3k=Latujz{d|bbHWHE7sG(Bx$kr3M`@upLHprnA0l;_`v9*i!mG<@3Zo6XFvz-J zLU)0$=`1v<;1{NBW4ShZX0 zw|;Np<(9*(4LcvJZ3lC-+ZuoWrd~S>!);Apzu-biuj;fnAFlNM?r0A<5UsiE2Zxr&DJOpx{ zPlqA=1fjz;wR4OLEm1k9g0#utIUy7oK1vX6m?Jve=>{&9x(W8}gAi5z9USB|QVknK ziM@}0eAz`{6?^~RmkJ7?85sX>mI@BxcJ_aq`9vfp{{LrZex}a(%ScDt>DR@#G5D*u z-?aZY-kd1E_u);)uhWB-!I!r?-(37UJKp*7p|cZ@2MBo=F;F_k#b6kB$zlkZ_-3wK zC-h)Ywn^8~I-GjpNCSYJuPi-71~?9?rm~?hQ7|Z)078ozvx&J=-S_*)F+?f>z-q}# zyfg(m-$stYj#3^^cA!B;#96wUq`@lRCEIyUUGz-P=*MjV_4Dhi_fHu<0m)XZRydI1 zxy|iJ)|@SCX<~O@O=|0iyYi6+8b9)c(CEhd3RRC8UnLRB=yGQ`Hjb~`KjhxRP?6JE zW2mf*+$rWRS@oCG7KXy$f;or~ z{O11H5a6%O4HX8Bgj`oRXrk(F6Cq=~qFb^!-5GSLQ)d1{Bv=-P7aKTV`#N2Ol}r%y zKaE)%$iWK^F!LYS_9YUrB4w-ppXRPS9Lm0J-?K4eof%V>5VG%4Qpk*bkTqMx5FtcD zikLyjI%JQCku2GZqKuMkg`(A-kYvfeWWK4V=XvUR-{<|d&E5%W}Yt4*c5nop>zK=lFdj?z0+kiU7PGJ*x=SZsZ&#B=QM0%QGrkQ7_ zrTkN-EtsH`Eki}fV8gNzUFxh7A7TtFD_YN44WI-4yA42$Uid?|q%5bY849@7{tM7b zLKB=r=x{ODDROm!iiVN$x_!Di@|t|Mhm{MsG5_*&EiWn}99AJQXN$n%fCny)7a7H`)3d0RLe%01lL6?S3$A0A#cxCl z*_u!QAkcOfF$H=$6{rTA^_q1-^jRlemC@rxCnXU37rzs5T1yA`08(z zpnWIq`E;Ri8scql7ycV1&?se%t1m>5nPQ1_O4vWvB&>FPdm}T?qMuF)Y+sA2KXzt_ zKPowHfDtjSkqu{ap;N+>Dt!lqWd!A=VFMAO()jjlsqrIP*!KQ4gUS;P5>^jX*U`9J zcM}yEPiID?8h4y|=gu#~Rfy@{K6c}zdxZ$P$rRsd-)V*)O^ZgI%Xw<+U- zW&iRr6|%NCsa_r0tg-@aZC|3`mK~?8J&lAP~C2S4CB0JGo{`TH~7% z3}NSu&PB7Fa?965@ta^#touf(B#cqu=;eL<`|_~7C2JjOtJ%Jt9B^6d_hc9@a2Ujr zy=Wk|Q=XW2i3@#joCujvZjn;H=17HXB#D0@Yib8-fr_@w9&@kSAqy_ia6PZn#Vprh zAu#JoD5PQwObHWth<`7)J>ZMTsR+bT6YQoB6y%DUA(~armISuS)M3P(LhKdvPvoug zIZ}&sczq;7JWrhxIzkTM8jH=p6Eh69_NBQ>UIgI6RHTfO2|veCRv047MqC(?8}674`X#mFDiRO_a_(*jo8C z*F*+M&r_kCTZLr|VrBCWSq^M@-DlI4UTDEsc`iKWaxGhEeVS5NHL(zlJk(hCz^ zndQEtR$j~f)~>gfU)r7fy!?s~CA0F{Io)gJja$*JmA9UCpI6@bJds%)@O$O8Iv6;4 zYjr4imG*gcn8+v_7*Av|dT$NMTrvRS_s@zZ#4lJYcdSxZMRBHwm~je>AjLDcm_Pym z2P4-46~)dqN}M)ooWg=MpuJUfG{PDigg^3T14!;L-8sZMKUnyX&g{pGLhHL-n&%QL z>XBj<&LwbkB?cRC2=F7FO1Dg{am!EH{R(P&ChNYYrSn^BF9arAGXoWDQ zfXxDMnMYBh(q5OyuwntBMOdn&e=JZ*A<4Yyv{+0O?$PDdAnN-cJz~Cc^KSCj@OX7x zmpX$@Ta4i81d?C7p*{HCm9Iiqq(SipGtIhJG9!|`wsgggAljvLl9PpZF-dAl=ns@t zJLejTNNYyc+khw2IGN!;du0G92j2e8R&&>%Uzdm(nY(^J46pP});PD4MZbSB{%F%B zaab@$xu%fLVa(e+YhMTr+1T*vnKzX;dJ6RLqJmfGvL>02b?E)&o|i95t^Mz*)+^Fw z4GV`g%wR?0MYUTK&kn4fM4pF3!#7!zJ^$){4vj)yUyWTD zeY6syy`7$y3be*;o)qoM9MuByG{cRMSxasJg28B}i)Tz-E)M0p73$kywf!z%^hLY zgDV6N^J3lI{xue2SJhD0&N;ePOE~Hf#lbg;E_(uA_N$>&zI2~Zliw!-(10VMiDB9zJ)ByW>)2F0y0}HggiWVU#)?;9BI5+V=>&Mr z4{xoz8jPE`>V3L+f}ldOnMl$RmY6^>_w08)&R6yEWSZ4ZeM*M!z2&0g0sK^K7P!&3 z8{+&0pY9MFvg0P;6>~R&Vz1mwy?#QH>T<)spL+u2#vJ)QLKdE_4XqucgVa| zf?$t5nQ!(yra!$@sRs%%Y&_G=aenqqNu>)yQWQid%4VrLRn%m0Pgr3|23vOi4ENit zgC}zbPaWz}E9Du%g4kL%ZWQk;46=|}hA(w+FpRrTpsILK^WZ}GNJ98$Ww#0wxB%VX z!BiJoQ0%;W&phTdZ=}0;Ju(j(Oj&vV@JLlJ89bL1*LQ=jV(jHn`Liv>0DkJ`mw&}! zIX!Jg|4R;A8;RgNy*3&xR=zeCcVKqyJy}r)0R6uTNy3n_ZvGi;~@#qNJbR z_;|CZd}F$-Zg%5Snc7#UH$T@;mT%5BuFh`GQ5og7=36)u{VH8H3%6XbZDPYN z({CVM*k}L(mw(L}5%-E_sMTkyVw4ySyz`bbk%!=GET5a~fkwDpzt7Ht1OZ4aQWLJw zBCxqyp-u}&Y9N802ZfHO1d@{w02af|u>FX%QpqUDwHK~qSQ*AyOV|T7MIiA<9Q4naC`_?ja;Rs&;Zs6hfJg)Q32dXfsK~k3i8O zh4Il0mU&=yzc)?RfQecn1`P<=d#D!dL?ySbX!YuOVNNua-+XxR^undjcp;9j z^?(8cI@(go`F#FOibLc3ovVQ`-gNONB~_IPoz3s#wsE4OM~QGtA1vhM8eo|9<#J?W z6sT)iRFSVE%4XBBZA90wM*!G={$a90x*GseE;|)}H%^ADh9`3-Lfz+X*Rd&uJz)c5 zcxv&nBYju01hZyufx#aa)GwL@fjRnL-l_qd5J3kryP58vX&B+a9gL9JiS;S=3USSiR3U0T}MqyE|UA;!E<& zWyd4wx6;+g2_J4c|8=@QsBXTw%J+57jGaA%~uIDdPs)uN;g? z!=9>fFUm7dydC$Z_B~X~!eFPDjlTP<_~!C7wTAsBudbb2R5IAy?LH81%6>rl?dt0n zRq44hA-wBOgK339w}LZwxQc$1VF5Wn~ssZPW1#Zf`x$;jaqST z2kG+Wjm1dk*HcwkEG)+icn24?W0gzRsOX9=c&bUAU2i^!oy&)qB=x{LB~cZz{48FI zN5Uo6VBn#!MV(!~6U5TGw*_W}Y0po#(paX%Xl46$%Rmp~3R8Vj+s(<6Mr3n|ok-$X zhP(zfCwpJcc_O;5?0h%3vL(XJ%#YuMDp7az3@}U2sExtwZWVUfhCzYD;~e?;EC4!f8#U$u}7EY7|LG}yJCn- zhw!mL)4dmg4GRX5oNRj~7`kBrfaGUz<>w=)!d#?%_)W?NK z6XW)Ey0L?YMc8v?^qlJyb%CvL{tL0D-PS4yvVDzRukC803|`q=(Q8R3l>Nkte89HT z>vwkC=bnVF+I_hLIA*RWXwH3AzvVnvqc-T#^_2&5c%eT|b;DrT2{XktGwSWqvdds= z-+QqaeceRm7rCn&tJE!!Pbh3gptBaO#-LESmu)7~4h%ee^U+ftfCGe+ZRx1dF7ebe zf+isvWXlL|*=~;LV`4E7&V<^;0LWgZOUZ35*S~7WX<<6MmC?jJq;T*-j*fK}A&tyG z#+GnES_D7=33P@3jXO|(-8ONiOSVEyu`}BmYO#i}?%nGBt;=!GH895Fou9kO#s5GCffr=-TYBiDJk8j}kSu zBvw1s^w&iFo9Datsm*MRKS5*<{p4{#68+?x{|FH%>wn?oDsN-|Jo#ebEt2}Tl~hXM zEz3J^cYISZk3x+KgtoLRQ~UluPPUt_cAdrknU7R@KKOvW-&b_s^C8e2`p3eL1*H)z zj%u#BOwUJ)^_lPako4S7rnFsw(TZoHe?fw{$jCsu3DeK2?w~BoPoPTx_5i|w-1q8) zK13%pX5?zeQQ*ZDYf23)1`wa4Dpg9-GP z^o7aR*}-%HFJQ-CnZrQoA!97Amz2)K;l3nJDI+T45B9tW`m zE->vV+WHa^yz|-A5Qx>fnF@FhoFzr8I33MToJ09^N}S{s8wP#-TIURM2o7vbI{Nm& z4G3&sWlpd`n;uHH5K7enT#xo*QSaP!^FXW^fhlNqV@t2*emqJWfU=CVa9Uv{-6<&{ z*IO`VZX`F&Y4@d6U5-Ay>1BKwC${KD)&<8BsVW|`VlEiy_k#S63OPWL{`!uJ-wG1I zvi-Vo){b0Z5N&KT9jQnV(6sd2Q$5}sC%QXiU%i5hE+EUS@4jQi;VDV!ER(0}^PNxW zbtv+ByhgmI;1&@q;E!<*BO#K70p#tHzEVq&UHrOn$Bj^WL9$+Lu{0#nl52T7cY1=a ziF)OyVoPM+Rl9nW+}@nCdJWwNsb$H$BI{6v@ND6s1g#5RezjDpJsV$>!Fss-qIVZ*y_T@a?bnvbOx?;BQK5duGUCFt} zt~IVpz(CejOarDG00IwYaWkQUK(nk5IB6sHc@0!yv+~`wrx#wPLg;ciy=laLD9B;r zQ)0z2-?IYGz1844+a&TEBlX}=zeU};>`7Zdg6c1V#h#Z$Q}GTcaHHGQsOB3nKJEe1|f zVB)q?l^4ii+&VanJ|6dwNy3wi1|;H)C~i0oc(1mEi=R_uVNPH#p1Uw`6ybM_C50bX zaLlh8d;{a`V9!z|gp4gy;NB-$#^l23A;HSpDp_9J`WET_z}LA{?++9D9WByytn8i!MM1A{3{ z76tc9tk3wWo>CC_nZmGuDBa%yaGf7u#gRU<{nN)Is!k9s`G38<$fHnMUEdhI`JS#U z`sH~uga!UQ85vNg2Nt4-KKxr~0#4v>p1k;RvY9oHUr&}Vjaabyo!fzLArI4;eb4vO z0>Lp{ExJ>AM=)BnkE|j*#+J`?& zFd!+q3Z($GldT1iD_ec-7{E9e`3`VPbv7uqq`W0Rj7i-grFa8wcfo(!@Qu0K)2xg8 zi3Y`&@o3?sWG#)syMcSeZLky^h!MKQBrA}CfB@)xr2m@n$0mBO(+nYu5b3>6Mn&8= zc1#^X5srPZI)|Tlk%lOZcwL|B)~+CTR44xY*~utfTvJHgm8C0hYGpXxN1oNZA5?HL z3m~+Fr_1G*XB^JYRP@NYcTRofB9%z8YBIV{+gJh?=M#)~-Fdma_EP-~E8Op}`Qsq3 zPt)=Gh_4wZ$A=2JM_x$*MBO3K2fq1Nr|S7QGeZ+4tTI7X+0`VOI=eG~KjUkq9zDc| z*5UA^Ggfan*n|n4YF2<%1Z1jeV&wE}`vewLmp%4a%{3z+GpfF~Ha7s+o){E!9`uF! zFdZ@((E3y*=IE2Y?r*E4tGZY5)61y1k|h# zUzFpozn6po`zpGyb6!P$mhO>jCuB!L4UuHVIw zw7AbdV6Jz!?xiN-?RQ+rKOuV!H~=sM67-MD(Hr{OAI0eA@yExVT7q^eil*fKfNY9T z_4uREKMIh?k(tBEI9!AUCZF; zCBypg|dIqasQT!B^Y8dl7Be!u?85};HTLU6Y5Ae06**$rOs|1@~3 zru?~N#yuYrk`cznl_6C(YqtlQ4Y!tjfjABc_{6W=Luu}gshd>;ROyPbnKg#csJGDm z9um8o(+%(+#}P;ot%@?p}Eo72}8<&1cD(ayUqTckfQYJ_}Tp?0K|}tW8wzaR*^9VCi2uBRUh=;B~+akBw}p!uETT?wO_iE z^h7T^qoZ&wrr>h%;&C-y$(|TPUWuNB6g!2x1UE3$k&?WvmkAi``n+E>ownsmNjD1P zj*#(`JTIDY=GtXvH&1XcMT7UuzR7F;Lzy?O`3rrV#7cthapqj+w-LU^u6EajLX+kZ z(+zRjNWAVK#3PZ-?j=MOr2v(%lB?wAl2S^ehMqpywmzC%TErv_!r}EsOEKszo|#Hu zM)EPXL3Z3+KTfHEQ%C2b1b{rK)-j`LWf_aQ-M`&;^Ug@*W?4P_XRn$_hn4(yzW?2; z{>t}TVpU_zedkUvPxuuAxmc zs?S(=Y`m(C&!=3v5iu6aTc3Yf1fd&|!#*uU6vLd^C#87Q*vXcv-O#A`z&5GDRX;Je zC$)E*Z7tn)(OG&i>9y`oURsg)xz#R~<3qFtjjfn~ju^@NR~U-?;dJA4$9&K@F^A_32deRt05y>er^H1a;!H9Q<^V#9^trc`@4nAiyg1 zG#>Ms95_x_K-?83XFBAJ07pQ$zq?wm1bG#U&Wqn+(gMvU<6`o?$j@U6$ala*tck@b4z#KGUJwYP z>NvzXb%|;ECdfqa@lL#6QUtWP{Ou{{k`kFE6Fc^*RwUA02Q#d(7esV627?Hn^fg95 z_E69gCLJ2OIW|41 z0)_OetR8#BLH@ZSX?;uq= zp^E|nqKMK7O=4e^U4;>;PML!NNH%3Qa;1iA+tZ~|vPPym1d00IC|0Du7iECAqse+q>H zFere40w^eefdW`4fcyQYP#6G%0SFj?f&my9fQ13L-`|HK02l&5AOI8sz#srD0>J(L zr%)6CLjedBfIiUU4nW~hC>#cbBcN~; z6pn$yu~0Y;3I||tC=3pR!4WVx3I@l(;8++O2ZI9$I1~YgA>arE9EE^m5O6F4jzho! z6da0z!%%Pp3XVd-F(^0|1;?S_00s`lz+o6T0s}{3;1~=Xi-F@XZ~zO3V&O0>9D#+S zuy70(j>W=pSU7-#Lve5z4vxUVQ8+jT2gl;zI2;@R&`u3;<1lCd zi-uy+Ff1B@MWe823>J;WqH$O>fI~xZXc!KSz@bq%GzN#p;?Ou88UXN601pH32mp@) z@E8D(1@Jfk4?yuyC>{pIBcON`6pw-8u~0k?iU(kLC=3sS;Sn%A3WmqP@K_ig2g3sh zJQRV4A@B$U9)-YT5O^#Ck3--A6dsDg!%%nx3XekJF(^D1g~y@r00s}m;9(d%0)t0k z@E8mpi^1bCcmRusV(~C69)ZQ9uy_m>kHz9~SUiBkLveT*4v)a$Q8+vXhsWaZI2@iE zPku+m&_GE+S4HBo7#MVxP~1ll2pHf5E}a+0d6@ttcp#XR4OU;!6H1T>5Omjb16))W z6=M%kMSU^MS6ycM8;ai~a4AKz3H?A0Bnuj#&7A29HN$$4eGsDc{CC+h_@nWGtL4K% z&`>gV4NQezJpSv6+2Ah!P`NHOvkVsvO+6G@@A7WYyUJLejF3QY8dE)7>@ zWO;{LMr;hJKHIMSZF;uK*j+R0G==PrhHw4COFa6t{orA__8TphPj;Hm8&`%dxO2?K z9@f7DQsTzpu_d1|h_!Kd0f> z`_xPyv;_+mxmM@;P@4%H6D`Du#T&APQqZN(fQF93a$I@iv2;Gbs-*ed1KkE=`dE0R zL9CW^REiFMDiqFxguXP1dmxUqeu8@bh#~C<=<1|?i%jAe?dUU-lw?Hn6ipN!A~7x> z42&BFSs5J|`>cH@%Cs?IoJVS?vWclYWN%wvEfSX?gyj`!FI%4?cC z(Ld31TmSB7u;s&6!TSlSi+G-Sf6Po;4$1FC^cB1FHlxp!K@9d)S>$!qezt97{f}eb zd)YEx9S$ah`g~9PLuVhA`#-(zcDFYB1MYzRf!}`I0IWMQcsGQXl-~bq!Qe-?+DdaTU|OL)D$Kl6nA~QQg)o448o{Nb=gshJsrrp8$TTU=Iwk` z8mX|cL~<}i&dOEd))FF)tC@024${M`TKC9Z!F@sb06X=_NA5+OvHwC0VX)S`*vr_{TC()xuJFAs7S4t_O!akb5ff_{#`+Mi>g%~xT9st*7iB*ygZ94M0Lhy4 zG$+?TmHH}6+iueLI=La?1NvT_#@*UmUX5w*stulJj5gPPsP$c8SAXL_Hk`CxEpx&z zSFUcHUWCC^+Vi_C_#5`)-;X!%^=`>|%fhSOCRRNtDYx%2Agu55n7lj;ZwR!{F#(SoH>qr0iNm)q0SCU_ zw=dTlK5f^f{FKqJaHp32z;<@UBCU@`wZRmANgmi`>XRz{p{5k0eW{MDnd#00^UmuQ zqB_@`1LQm(uf^E0$xh+liAVg5U3qIMO`g(2^z32iUXG&_{Y1ac>xW8wS)Fr|*WXWw zy6`f+TYrmv#aM&Hl<;bAsGGKpCB7CYyW$$7`X+unn@XxeHUJGk^1sdsUZkHcAJ&ML zJCRX1z>M6S6RKigvr<-*d27Dv;w?C=Pep!8}RBCQ@ z_hq*Avq}4AVoEci0MovM*9V}XTw`;ChY&HpY5zq(lMKpdve&v`^{5Y50|F&;z8T}xw6N+ z9V48dJv4@IdRqUS(I)!nHrAf&6wW!jDn>0lWO?9Kp<}+OzIx|FZkTI+fLWo$qiw|K zr|O!kl}%1M!c#45d&OBNUtNu5b3Y&LtM=v27(WV_wOrn-9g-V8rD`TTI2A4oNLjzX zw)&2A_@Mn0l{J|UR_z$AZY6cT!2=%IVztEV))8ac3rKQzh& z*5gl&qTaaN6VHF9Q3e`IHNylg2XoY$%HD;DyKl}6G?n+|+&a@JgY<%<#V`Zbz4CFV zv2xww(-e(o=nvs{>Re|1B}G2g8D)}PxIZp}Z93N|+28X&wRkKI#@*j$>U;iRYIDVD z+tsoq2v5lI)!0L+DS%!iEhG8T-H?4oP5x-#M*E@EEBEK#K5tqZE>Mc$&~C$Qj8_;` zS-fj|_T^)}?NFX}d-K+()`wr_-eudZ&UMogL4Myb`22>?MVUOH@LQDq{71ex_^-8Ba(M0YmpMDBzl`CoVJZ7R%RoD!I24jR_2TQq_gnp?jc z5jQz16iay$E*m|~NM)RRJZ?>w0dk}5G3_3>)6tqBBjr0<^eOrQrRIgB_t8_2xPoTo ze)MvG29!S;`2?^QBRYs;bKX9Y4^T{e5hQv_csa1+^`c+g&5F!xC-hyP6E%o0Ti1-L zaRqZ6{fyBmH{=Ewv3Ni&3h=8zm`5E*qs84(N@t9G}gN z@`4EA&5DvV7qmhU8RT;bXJEgbJn;9FfB?4vt=}mDB4qxn2F&{F&r^csRPpf2Sq+eF zThEX*EOtkM@lm(~O*>>X|8=yaOr!l^GKaC7GU<4!W~R)^)J)V>p8DVen{Qrsf7t27 zEBU>?*(OG!kRpA**^X7uwm)oBdf_eL0>DzD5^ag{riQnrSdLQpA@2<0#1XARNJOWd{e|Tkn zzOCoGo%0)WunGl)(;NcopFp^DjsuVCxR`*)ZO%UA^s_AA=V<3fr9a7jS~)-l&n~QS zx_&NoocB|c+@bOr+&DnqJA9zlEsA}n+IMA;&oSKl=j>#$7Xyu6Urgz?)HGl+xyt@eKaV1OE6Kf7?_toA8(+|L3{NKI$>JJ1IlJSngT zHY!o1gwwBl8~y=yCF;Q}S0%KZbyw8GFX{aQO8%R5euL`n**e#Luk#yY4igF#f(eQL zOeku4Ma1EKtG=kej9(NT`|&5IEAssS@9oRDM|!0LDR_;nj`W6tw;6a(#+&MT3;l`v zNdR3|8251Mjbx5oUs`4^E8RjOvSz(XcrrB$MG6a`%VX8k9YV=S0RysVyiXekJ%wJy z92DJ1CBH=TF%nUC3&FxwfvpcdB_M;+xV|cZq}t5N?#xYCspHhCE*W!KhO0X>?v~5& zz6-Dnjg!2;Eiz&EhP4>1nIIv!$m+~bEt@Y} zvl8Mr#q8<06!~7}Yi>R_%+*TQOE>k8ld{V&qBz0m`w_8Q&4=Hz#wMClu@o$XPdurh zd&=XvTqrLUv8`zbC(G89g=VPdzdKzIe-hqeD?Sg|ITTD*7IMSR)esg zsc2a0`zy?U`0g`K8%h_m{e#o8WghKAU*MiVMGWWtxtSe7%NLo~V>sATFPo257!<2i$p9un!dvLP zcSm132Wp(ZFm5sIcLuct;6ppx4_CWG%r9Eb$sF#~)jQ5nh~+-tdwE;j^79;1mFLvV zD?Fa?%8Q%ho3ZbzEQeoos{Q!3G?cFs#C3dp`16a}?4S`Ie@aL)k0ki!Xyv^BIJLl- zlw*C~u(9W0K8)sMdHBh+oXSFkqfpqKE5Ve~Y~;ZMCmvtEtO}cO!YCQ~NKm(+O(gGo znZ;-sR&9c78~Z>W&~Z zg>7(B8BBDuU?q)87gQ7{9rAop&heI2Zzf@f?MS-Y=9@{m3(N(jw1I9%@*pajWxG7; zMsITZyV4_J296d{0!75%29WtJ@}(+)lOhHIU05=Ux2+9IFEpMJ?Pwf{SiZ@)dL{58($F+HIlFOI5Y@CgUa>Z}C+y)p zH&Jh^6L+It$UeFGZhp$U_G_b-zDjE+Gt({Kmw26t*VHq%2a<_6xbp8j;jx@2+-{yT z?`(e*P64CFFK0?ucCQt5RaxeHIKJz^HVr!JbhS0@ zuPi2JW+c+<^Px1yWoBU)INn$ROco{!5m|kG^B@lX3%yFb@%;-?DpEZ(u7b_$i!owJ z{7XtD4~UoIM1(SEjBZXwEF~CzkmN)uQ}wYwDr3K{o1`MGP8UpYVpFVe?B1kVa{53Lge$7x4K&k3f7eJq7ywgnbweHsh`Fh-tmV_>me>P` zRZ=a92qJKpoJAS&dNCwwjk?%xF{(V*pP-SiB!_XL^up6`QO>1N8O_e+$fSJ+E`OdL zC%GtWp^yj~ys;8r0cJ*>IkylFLF_A7yBz$$Bwsr=ZNUGJLQCENobIQi}7+c zi=QgdU{ax-raSCxbv-}Ct9M?}sJ$IYp)*E2$bnl_e}7GDwEF!G$E%_5{ZL|(D}8)j zB=&>c30&KeAT;f5}?>zh!ObKeLvC>hx#&_x#Q*0{qF#%WB?&$@lv^8?-cVJMb-a->NdiuBeD6hcC6;$}^VI)-0XJ>c1wxXxq|>MZGj0cG_4Ut1j=ZF`CR zqj}19o>vQkSr$0(z~j!2qsTn`6i3D-k|jWPcqo4uHLyOZoeU^s2|AABcL@cp6!90Z zDSAYwPiK%6X=-We_~)zBUM$)q7Qm8J#eyF>&+>Exm&>8GLhS&m8j<=1dJ1iU<-`F$)!ZbQw6k6e(+x;z46(WiP3j$o z%#lcnyVrTrnZzqsGFXKUN1q%7tFLAmtCG%V>mI++%^~&U=g-#G&*V~kq-e*Nhoti@ zG;lpu6U;xDNm`eKOR=sO+cQdy$XyWA>!X2_l<=1z?o?U@7fA3qDz-?9&s8q57gl6s zdv4lk%ZB6cSv|gJeOH&BiW#{HYP+)&lUVEHQdZr|3KLL(Z?#o{4+z^=L;6&!Mj}&+ zT)N}S8RwB@@Ean_DWh~>$`nfDKg!b*2+0aN%<)9!*fqWuu6hc7zJCY2)LA4L(Ox!H zn}3~~vUfjlmlU{eU*H%` zHIi3m5dKVDJ$5(Z;%ci&!qK+^`koc{D)TPZ@31Fn?9jp@=@g8DU6ts;jMqokgIV9T znuEE=2PInvpONHV-t)+B1AA>@#MfmOwL)_b6Q8`s9cFYlYnQJi2h4i)$2kTr&EE#U zOIgml-@p7OJN+_roijnBZBu}eyeX#c6^GSIg;GHC7I)s|o$iiLkpbgi&WpI{!q>!~ zha$EEABVmQHT$|Sob5)_+ihI8`04VmSh>AvcM|s>1rgid%Ev#t1%AcU;kW7?+OFVn zIvb?7P7c?uBz=$lqGKulZxZJ=An{k>GTWUME`;`X;r0Yj@oL;WbT8-(W)QF`>EHRN z8h$FtiV)bb>_^dxIIu;IEvTo_yGD7nN))}xgsRlekm_?Q2aATjWHZrR)Q;!7O;mAx zB>iPE#l!c8*T?5oB5#8+l9BojgF#lWF0;#yE?uc^`8MnC*j#8BCi^Nxds??bzS=Bf z-L=uXqPvK4CFF!JL+EZYLBb2+b7Rh%zbo^#3Md7;^Z zw@VnhRp8I0N0RK1z^m#CV-pwQ@JhxNO*tP@_3)76099Z&p^6p*PXh%{Nl3NC-iYzl#t6 z5CSRuk8rLMsdoAA!uL;kUVZ5Phdej*v-&O1(KxC4Ql)$la+B>}<+<>QH|FC*E_j7W z6JzS;_S^@{eP14g#}1QjzC$08Zq>6I({y!!k03?4U?RMQqIvgVd#RZS2gqV?Uf;UigtR2*G9Y4QwZN`~yGP)8v! zJ{>XH=J#=*4y38d00SOvUcc%{GFv{;HLkvWYxwh)oO}-?*&M)g`(!A#)ja#;YjeVX zSNB9^-H5Jc)mmuB~nIZXlxm~bLpM8=Ba==(nN;F4RT#= z#G)*9VCJ!LnIFB$EwWyojN7T#XmoDUR<7m>HmsMW*k(Uz9u+{Zp)ZSa%1@i+NIn$_Oe|gezaZZ$+q8!3kNEN`hIpnrXGk z^Ax>=Nkm&lg$J{$0?Ag!>jH}IUV6TqrD@q@fA=D9$6=tUXq)Yisv1N0RS^$3(z~^; zk!;0#NcNxuyxhGq-JaF+`QFgwU_;4KfVM4~^m~mbfs} z^Fd7DQsCqD6a`k?e9!(ZT+A8q>;P}YTC@nWr>CpY?=SG0ja)gPVQ-qC>Y%S3(_DKi z85vS5J0fcc6Ws?B${@X;v%65*_PNZyguJDs0i;H+4Eg(%E^Jj_ZJS_N-MvF7)+O-} zP}HQnH}>drhtx+Irsd`#BIfTgw{1AZ{1#Ihn|-H!{6%T7XS5PVPw>oa7wly7zM}s| zf*MVw{K|^N(9P?*$RP^mJ{qG48Alv`_;W${U_E5_#DI>r0-|far8tHD&~HlxBx-^@ zJ$5vqbei0xDtpxoZo*uDBZLb86+j8P`tRN5e^qi}U8m-MmA)H=kr?V_0J#*IAxn2& z|NkLpZW`ov?#!?854QRK+?;LydHM61|ESs%X{}#)6-h5*@fIt%@EXOdnWuF=5zn~# z|DxNR-1fF6oloI&$LzmKS1N!Lr1-x`R|x*-)jy8_h~4c`Cc}%)2%_wf0nqJ-&9+Z4e(tjP1mdV>a93_4pryB%85di6qw89kQo-D4pB=)x$=7<+xRw~t zDeQCfT(bWyjcm3BrM|?Mk8*mi@}7*lBL^`LLnq&vmTk>->kxB(6PNtHL~qY)dAf7$ z=Nd(-Qoh^B^}R3PINL6_C$iofbA^xOsqt(#k5-n3d9T&w`5y1@oSXcvTiP@GKfin) z?&`wh0V2V9B!tmvKA42RVm^f8x8YJ1T+kBJ?l+eI+suLkfB>?Q2 z*#Dw)@*n-3Oi!M7*^I7>TI=rT57qQk#Hnt6BuR~SpJFG*Mm9t}nKi}bg3WL#sQyMt zw1=Q^jJAS5rC4*}WC)nl#@tl*zGww}A&u|i$HKZWYnEH3pPpIG)ltV#X=SH6Dxx12 z>*J&jCp%5IvsRz2@8#P!Mp8EN$z+fx4{N_iOC6@wIS*Bce3WXF+^*swreKqi_|=na zs`VidjyN)QnRxs>kKJUxCwQqLQHNAacekgws~AlDcqcblpx=4umS^ye(#E{@MSP%f z8S*Rf^eCs`7_$CgOs@wY?;A6KAlA;fje3IXf7z*i#F&fh77zO%2RAz)d)UQnHG?4b zn4DA`vzGpBNH#kL2~)!@>i9HiV^NP8gu{)1ZA3=SZBCv?GjNhA0wzNyBbd1xgx#ta ztm^C39ZzKVohstNNq8ob@pS2v3ps=CMFUkCDkyae0R1 zJt^L}afbd9#e`WU5WTzfPP{GXf;TqPGyJK55=YwIjN3?|{hWep!pm&2a^@^%a9b&j z^kV9xMK-0(D6do_G2!SnGv!+X7O-o_QMMZ4?+wvfBZPeVdexH~E>$)C6csL7K5ZJh zHP4za(EPLNIL;J?2q6Dg)ydG1^n?(fCx~uA7eLJ^{YTY-`4rz)=<>b}#$SlQ!IBD8 z6Zj0!U;D?n)zXE7C~;Eyt2F`F+-vP*Jm=+vwOAqAzV4-3AuTyA zoB8hc0PZG6Dwae$lgxviA}O97g@rO}_^nI+b%Xj~YKTSiV=Gau;3xTHURHL_;Q^Tj zL{z%WnDrLoj9#Cc28%VdGND&T%)MQg-hl!otmR8U?_Za7lBGNkSou1hQ!H>~oNI;6 z8-1no7Gk6GSRK9ouBs*D#pC0>9g!D0-pYSQm$L75Rs9L zg%AWg7O0SNwCW@Kqy&?~g-+-TBG`;tIYTHI-z-EjMPO;PnWLq6o=WPOz#Z_ku#pIn zr1iyEmPwQ1Sf*7sVKOZ@9}!dF_~MK+2S9@ z0{8;~z-PYz2|19A1IyQ2^#uK+&-ME{(b@Ggp-Dml1Y^PSJ7?eiTXqluKmT+k{|O&| zVKNoo`Y-q}nt-87`u~IvJ7iE7>@<6-@f?$lU2+Bt9NwQ}vNv;6xn8o;878BDRE(x; zoni8gA5MB1vXy6;yjQ9hbhGD2*z~jNyM;QPWpYE!u0!RlefwUTEw%kswgIm(LoIbG znI8B7mCcrV>xhTB9A_j{aJw#H=?5FO^@nw4K*M{S{h0x&M!sT%eto`sH|w%Nvedh* zFSel%$f=;SZRaF(SJgs$$NSr7eP=Ldo}|e4CBtxDOAC6*pKAu$^7)s?EEGRaG+kYe zZRI)$BgMqvZ#(63uY6zU4K(h5-g$5Uc~#Ej-uW2#6;FQr`gYfC9MHV;Xv7~BrdANt zPb&E7359SZmp_Clf&fpp+$>1$}jv%5DN1oFrG=u zO}lK+p=wcn_d6j=o`)8G9L0--^dt&VGfVLj-4vT#rEl++!xZ*s=(Xt5L!#%x07y|} zPAQs2*dm=sTlh;ggi^9HyoGL!qo$L`>AupRsRaIuY<>cg|4X*~C)u!_6#m~BCb)eh z^7*+*`Rj_cWke9usW3A$H?q^7{FjQBvG~guC{j&+-wbiXREYSsSgnx<$nFABQ}!7@ z@%fOhn>v?$O`fe6?hCiYzdhP>3TY5ewy400Z;eH`9(dI>wnF69kpRfj$;FcC3942$|_|cCM)Y z{f`meR8q{&yR=@za!4(@NQ4a;i=ckPY%+vPq2h?Y7P)Nf4WkYr4hq>BM)@ zzq7)U?7{R|RtV^1h&}l$D{d3H4(zVTXQ~%#u#-;Yj|Z7F-8svO5_rhnXB=kd7!t1e z3e6C3CS-&6i2@lsuG}}TKCO>`GW1AfB>jHF>7ZtFopo(+)9Sk~yY)-O3SdXvLCn_D zkW%42^C8bIhxiPoah!+=!g;93AYrbR(Zl3Jy{oFU_L2A3PXg-W{ASE8?y>Ubn>A1F zA>V)Qmvb?De?xqIc|?IMh@N5)~GfrYYzPOofjQPNuP(S@EM&Oy}*~H zA!APC0(jaU;oICZT^AJcYhceZD11-a%4v!HOxK+f3espAy(44@MSV!&3O8Ue84V{@ ziCpmSnt2!#Ngv3}u6dxVpbVxk1PY=CX2-fgl9PwsAejdw#V&+0R`PMum&LI0t#KsF zAT?m4TlNOs0F#!WzV%W9@rC980#J1dT?|1Dm{KO!xtA=2^ya20vBjLnpajGocCijca47uVOz*13n^b2F zRX>^Cx9woPY%lxc!mg10Bp|USa{fza4t`@!ZrXNZ@(JY&BGnGx2SFJJ-%k5})h>Qw zD7;4WVrOqhxjV%EGY3T!a+%8D+M>JLk868jrvQVoJ6obx#M09C+@S@q{*&aydHRl7 zK#+VC@+0c;_e5ouKr&DfJKd=y2}wXN?Zibzh(x)m`T@5f(p#P8fR*O6b-B41!OLqF z4$Fylc@YqGdJ?fskT_ddwwyATJ72@=B8B}(5-kY`=*ME@-jugZux!$D`t4u{SRNK! z!K5XoZe?Vvm?-+xE=kBNrJ$lcgXBRQuyH*(YUU>}`tJ3DE+DQocd)NV>>hEYl^6zLN zUep!LMCefcuN-pW8oD>~JnW))p%l{duv1;BorwH1?9ZMX$d4Anl2vgkD1L*omZCX& zF}M}H)TD_F<@aDyl!{VVvO(g*TutT0XK3x?!T#VD4|e9~&zdifR7nzE@Bo0hPSRTHzF~|F3QX$dufJ#hSw_9xe zBhQt=4&%P=P-8Dbv0GLF5??CvOG&7 zR;Ro$qQ0_1??rXXoTxI#2{oj}lweU}1Vl^O58n2nR828!L-C^L@Yg}Lpf$_*||=zG?gSPwoM(80>3vXm^UQ6GLfjEdU0IA zkW3i;Ow4siR3lfKe5-XXP*d+8R`|cltpDI4@x}i#v$n0-D)e~=&j95Gt`t@F_6$&N zbhI{tk}V;LLcf3+FjUV7^0P zu8E<$11uNx_lS*Z%!^5f15Z>AHfFm7uJC;Lm3rB{JJL4d!>=m0#W%cHJf*5eFB$h| zT;E@wIz%{PGL-@*zC5e79Iar9F8#XKgq^6T=IPY(b6dc+QVWZU&Ocn6>8{(XeJtJj zJ-nb@_;%j(_Wc$kLR0&^>!eQ06A4#_-*>hjgf+4A;7@9EA0wI}@J9QK44p>@JGA+X zG`9lRkFZaEgF$es3L#er(cHtB&?Rw_42?hOMt&rOF5X)cOdKY- z2qqr!vU;lWKu{h;#mu!BNxl=#bC>#sZ#V6g8$0i#u3lE*X5ek}1)_-SNphp*2)s4v zxzFiFnj;ZvFw*&I8B{kzP#R~dJuo4v6?VonrqRBT1dA7tSo!q5k zrf329|2X9K-|N^vsE4o53T)k$qyDq#ZPstHMso%^%*#dE-3O45pC?~S0`%7NepO;U+^iOV+FpElF>kgGCJ+@mHx~x!(s8MmUB>M_5?O07Bz`ICQGnSP zd!Ta7A`Z2+)|DeNtw}_Zv1Qdw0E(%uZ*q?eV3(0`otOX4*Jah{#y0WDG?XrzNCQIA zvbTw3yn$yIfBs;+Pg@r6+tPs>-a@qhYFevn4*0YKs-%qJkZ=zZ+twB|zuB=tK=t+- z%3*b5!tBY0ik%K!Hx#Vp+5%6%7@V7XkfYop_Mn@V@`BiibLlLTN$~Zf{b?h#LJG_^ z>Tqc>p!K@Zf-O_L;D{V$fzQxWE@8FByMd)l%i%OrO6D=A04-8362`3MzW;z&FXAcDKSkCb%%=SXa(*6!#ksAlA`7Vnm% z;y)S+1b_4lh5_k6UEu#L-2U5^$P&CP%_`}xiuHgdS&%2MPGd9M{Pe}7f zjz8B;J|K^AV3EF5eRP;3UX z$q3$CR4F7TTT8D&TY*RI*Ly8bzaAxPKWX-sPtbFCI`hfn;Tp{rUneMyA_Fn~+BM%I zzVXQAJw)V+7F0H|>(__-O*$I^4?TA37##C&%ad49^K1|{eb#t?!awS|3JMLoeE^L# z4XAJ?%6WA0g|vlV)J-C-2XAUx=GCH}yOk2vBs_#ZyJRce||JW#3ouTG2jETY%{6h=D&{}&_b($95J{T~q}#GMZmrZJYJ z`iM6>>7T;f7g*o6Un`!Cfvix4ISpA`yzmb{O@&=+c+)3Bc=4^UZv9F~(tXmKk3B@! z-oUBUabVy5o^~Rf*{6w=1*1k2&O6 zyUgQ@)R^3U)HsfiULl`&7an!MpCo<}Kf!jf&}(C*fgI^-RNo1epBF*j{C@1WOO78f z$D#W{U}84fd9cn5Pq*di&K{f7=!vIg=;Za)VYzNYr^4{ldZtl7s=PP%+==bs-C*LB z!~7@%5)|B*02eVEDn@5)WJLfZlZf@qSJ;Zds#^;0iQgd}l_k1qU4#TU9`v}8h?3Z$ z4#QWexfu2RxOpxKLsmkkq7Lq`5CVrlk}Zd>GPqq_ea_lcRR|SfHM*X_f5RxFI2Q>e z5KwY$s~i#4f~?Ea3*Q2)qSP)e+dVxL7hbk$3Ag8?J2fPWOl3fS)J4T!hGOO57mlK% zyO`x|ki-Z9idnl!zIz zA!h!Lr0r!br~|<}#y@sGQaCIK_5Z#y+yr14NK(&$4T6~T;m<7&lrZc@60Q!C zPF4qr=PcJsX3o!Uc!Rg0YqqX+d8UmPZG4529d}FHoj)ovt_`~cS3Y(R>+V_c@mCCT zou19}QRf|4E!F3h_bpKf-o9^UB}GVB8KYv!Da8REPY9S%09Y5thbxC ze~6%zzt{E_&^OoliZ!+NP8+R>d9y7f(RLI@^15laqS5_r($0sv3zhmjuYW}|owE5E z6schux7I%U-x7ip$P*LoBqNEW4_SlAR3i&QDFRF8gYb44g<;1GG7Dj>5$p3$rGlYJ z+T0G2MPYc8NzSKSWf?YM1V)F6u@Vv(PNdAf%t%PxyU5~rUb5@Zr~E`y#gS0ysO50Y zXcOK<>xIZgozFs13$bEEBz!4a_n20KY}`z&Vr-;gWpNCdLZ1_u?!i{Gy(XR4a(!>L z^2bRwX0C-o`CIk>K41B7-=H`rO5Xo+^vbOeOvNjK6A%YfLYPGE-8`%-eiO^ebo}dD zBX3VIGhYbiu%UDS1=1xFR39sS9m{0yk6RIk-bs+WJvYQcYc~{p5J}U=T<&{8 zu`sPE!^}MsPV?p1@a_3g%-;k2DKUi_hEJszt~bqFM>GrDOjb6*CUPp)r^#g+`$t>(LFzf8J!s**SE z_6>u{Z%)X_=lLHAiOb3kpS_`(8G>z!Qs8GxCr zb$cG%)MbB6KC3pP4rku!6%jwN>Gs7IHghPoNCwZ+b{!hbd8=J89(m$Hxf>t3dSq+k zw=CtI6-*T8gQTTq)+++tWYz7HB?6`Cf+pp5vS_K8tpHlV4@|TkW=?NDiEK;is(5(s z`A`#TA*Z@!I|y|%Z1_B#ixY$S>}(zNU6zNOQB%Iy><30=Yk7}f_O9B761uDwgfxfP z=F{a?Iw?3tT%h^a8T;^`flkK#FJlk2F!`@f{%S0!>>8u^&KnDZl80aB&qt_}IcNN- zlYCGi<}8*nFCQWtbk3ji@6Ttc88`OM%REW?j6c^#0YWz2QkUNDq*|x(w8w2f=S z)H1&1XKpU;Ug(P_V$G4@l1E^2xAmlXQGSDqlB7G^TZug=-iw!R?pV-+iRWf7$}s+v zPYdUZX*OkPI4k_qBUzgCz46idQ5R7Y!X^;$C0R~@t^LvXN5b;F)0J(|DwxvYm&NJN zDuqwfCam)T1p}h1*wMLrNm6TaMA}G4Jm=@+mCou58HUWq-;A7o?4(X!;Ox3JP`B;J zY3SC~r4oEf*lHL^@Sg`4!eUwYM)h{}d}*Jss4laqfS>FwDDt|jTSie9I@zcS7a0~s z3b9TsMw>CHTE%j{H{p(BI|;Q`u=MU9F}NXoVKJVHVW3!tr-8XBL6;{Jm8AItlMK6T zPf{GNt4dmu2)_rlNmtWTr*Q%u@`AY3fyCZ8OP(Y0%v z88?Nwv?cJzrIqHdow+O36*az8QvWg2QK~4+<~gRxjpZiiuYH{4T%NATnnF?-b!h8a z^)x=Kx-@W3n78ST+b^f$XNq`Md5}8?D^PtcDZ*@KVJal4No~*nI!#+osw26y_ZSxmT?>Z<)5jEW93Wg** z(cu#NYbx{|a;o$=n{Fl!@YLKNE<6(qo6YP$Hybm+`S+F|L!LxWocqas<+5g6(^I+6 z*Arm}P!Nm;TAouhY^M54vX7FMMGtdD!ce-@GG@a|^ii@&8>c@ia=R6a+s?h9X6Yj{}^=O63v>41^JfBHeWXEyx#2 z6<4iCgJ8*waWW5i!}+zD4M${t@~fAep4!hwK3u5bC#GOy4w&yrHocYo?PdkWbu5}P zTT)O{`fI2VyMlv?|M1pyaLXLKjKn0;q>;)(MT>OQbx2|H!hu1cUvkl)=Rr9L5;ZkNxh>N@3**ESNV=Zy*Jne zg4~}@=U*9PmUo zFcjX%)$lIr3f@lwM$xc&c%?c7xr=AwDdwvQ#`mX1C)p+1 zdaEEXu@W7Z+p7v6dg6COzUbTXuH5Y^yQzhXL-rT%tF~Q3-qsGDB4DHxOth{euCp=f zc)q7U0K*LsK{IP~p5ix!Cp)pv+|!Ml8vh^K-a0JmwQC!mVVD`ZyQNbQP(Vcn7`nS* z=#p*_n2|0?N$CcWkXB$&8UsYSyF@?{5Qz`m``)+iec#XX9`AR&|L`Zr@tbR{b)D;6 z>pZVEoT3^tbf+r+OWv{1yXzvwUahmSz{O6){Ko-2r4eYBt*1==uGN*`)ugC%la;uO4B+RJ4OByB7f0yBY7OAAAliyJS>jF^Hc+% zSb>qSQv>Z0r+Yxo#YhGdUo{fyOrLyrCia88NY);^Nq4xPow^IA*F=8oT^g$bO>xNP z@FOV-zmYIC>T)Y3`1Q?|WOZ4t)g)2XEAg>S`D(K8H{I1#Lx1eG^xr_wReCf1JG}va zq&K^AL5=zAhv%469)I42%o*v#%N?Odw{M@4FD!Q?Ujzl-mH)Cdki@Bx`h-icw^Dpzc|19-)*bNUnceW73P=W`jq9 ztdk5+1ybwYKd>+h^`(p{p=J_(Rh$ppJ~;B@wF@p$+C8H@v2~2Cl~#@A`!e2UbS)(L zOgZX#%?P(lDO<4OvJ6iAEgAqT4Ffc1Po+W5pS};=^I%M{3 zR;slCoi8d!NEohS-=qwb>IOp? z-Yr+PAbjeC*#MwxnhEK=M)x=wy1_S|Tohk1+09ZU)ay8Rm#QOUSAIn4%N(QiEIZxIO46z~mbUzCs8 zQ44NGXWRA=$p{B#I_fc!SO!GxAj|RZ(;cRuTaAn5hRx*2(X60DB| z%#lER;R|B4IH9l~rnG0yk$&Q>C3_Qi)IoZrMAr$nm2|49K@t2&+Z+w5M)z7}K|iru zXU~vFJ7gHRi#?TqVIl~b3n5xt5z{BFTd*VH;S!)M-JwkuK1m8s<6r?#VnQCq6(5#z zr$~DHd!extlfIB=ZnkcMDv&nLbk zSC|aBvhW_k!cv8aLu)1ZTO#M52+D15rnr9>X^<&DVh%k=l9m@vr^ zxeGz(3630b=PG4~v!ez6&$p(R*KOKi-PkliD!&R;A(vMvE)e#&3z7sg?{5ieaX^V= zk&F8WZ5IcU@P6<|g0h?s{7g_1>jBKa5>!#s?oH{>-8@&0PNSn5iv}9xwnPQuHG7*} zm;&fIMhDJIj>f9#Ok?nzyL+DVEDJRD)Z#~DwOOhIh|Co+{tO6GS~@p*bIhz)d4x0F z9f0n9fsPY$4DS;G81xh35SqSdso#7k?Q)}hIxh;=4^UVgyroBoWbJq+P*%yzHwF;-wqUL)$(S{Z|++1ZG>v;(lLFisqz zz9KC_Ky94FdZRocSuwMD4kp7c#`ajlCL=Xb{!3YLk{$&nTZ&oE#B#>HlRj%HtJDnh zbh{Gl(lqC1Wx9xtz6_fjduN5x1TQI3>nHvLE$rE-Oa<%w*t(3e6n#N%uGEx!I2I9M zG#Ok{PKJu+C2QULmo(+6C>Npw$%fJR&DVTwK%Um;6CG06u9O>EGIrd2v-5pvLg%5$c z9~fP-m_1<%unM&@2%Q2 zeWXO3T$UFJTWyZSIe-`@9#umb`+z7~&&-d6EnQwvVZu^Pa3i16LY-uaa#&ol2rAEf zcH3U$2CHdmhV32&W>XtCA6IerE zgBkNCb|f?EqwKK!S>1ej3t^s17^{{7qg%!qgKPbY?p zuf9DMkNG*XO7`Wqiz|?_%VbJKTz)$l+xwEsZ+F_s1-2FU}qxlEn39)|fDlArWlD2Gf3Ugy9ie*@y%(k`hZm0qY>ky30V1$o~20txSC! z?}-oM3N;j1y5&_Ac)F1~)-c^zG2OfffmUJDP@u3VgMiviIE5Kj`;mSuEQdAOL6FVr zeMsm15`8a1ZO$2|M_?{C#V97QJdsYSmw9--%I2#48XEx9JPHeI48|k1GMskbV#NTE zi!C|>ugjiX%QmFx(uf}Ymcf#9)svo1J5$q;U_@T^q^+aSE1#p*k9w&#{)t7d z4*nda_eY^<>1cl(*3u`MU1Lv5xFfj(4M#6nhspvPM2MmTQ>V6k9B;Q2SC>;9#AJBw z+_u(8Vwc&L(qF8otetDys;uyub&qct@=xE&?4z&el|FKngo9*+6PzO2OegfiHc5Cs z!J0%}(J$rb)OUd$GnOBpub+F-r!{n@&D43W(jq~GZz^kEX@8n_DdNSTv#zO}d%g#h z<$ch3hmc<{(+!ITzE=edbwpoc#e^@fR7KG)z0PhtwQh9`q0@ZIZ>C|S7$y%w`T1Yt zQKbM}^n{e>DGmt8znbFO4seYdk(WsQs2uImPz)6s;09IIwH>mrsSW8!k5cyS`Mnsr zqw*%JPNMw@-8Z08s`!Dj-G%R{G=kP^KSly#^#Bgio_3;9wujOQ%0OLr2Ki*l$?=~ZeAEjb`7);V`(;A~00D@8G6TO}HXHm* zze`qF{+6x)0&d}rpA^zny6!W2fUtGlQ<5I&94{VB39A>ojB40i60%S(W!iqOo#wP0pMNoza(bHzBLOSzq?>_cj{6QlGSR^5*nwQ%VAUGv0a1++3x<(}L0qdK&4iZ*x6)KCmJHJ{i& z=Tym-4EXTU_M=r-7D3ke4;`WPo+M32!C5gfE1gJf$GT5A!WEj4w*nqLI8L;gh;Vk& zWi=Gu`d9_*Ng|7`-JFiI3>jNB_`22EBv8!#>3VO6X)D25lK0$DHE5<)u?$}`#ghh zd{AfsqnN-Mx;jADh4f>9I5Gq}{5qnfG^uTAJ79^Bia2f0?&^H z8AG8T4=2K9PnY6>+w{ZeP;PqHeEtq9;4Dyi=2fO@lll!~)#{g|hzFU(K6IoyWDDp! z)y4<;@S5R`WGfQpjE^o_jk1aNL6e7BX;!KBC5jin&$Ivl0^mPV@9#HdJfi3_8G9em z{&iE5-;@6XC8(Uo7!VcpqHft=p#n;gah-PD}ql1&1Oh9DIz&p9Ags{UiIQJK|TSkEPLwB#cY`Dj`eH699tzSN@j z5Pf%$7zNwvqO~OW+@o$SG@IP#g8{r|8r!MvRTS}#dT|)IX-;jDVBE->n;89;@?$C{ zxWUQ|ElLd&I02KsNhHxl6#@x7yI7)@yvl8rxWWxqEMgM-x_n84j%Ic3ET~phR>X7? z9N(QSy$6Xh4557kY@#)7%?T!ljH$vYVN6aKq;Q*WDir7^gq~JtBvkmqEvBSC^W(5h zlYf_ie(ss<($jHz8UPb;;yG)UHF^Ci-_YbLFGwZ9V7s33ESb{w!>yGfiF2mgnGQG_ zqr>9$8fZ8;q0J*EoUFUX>f$m<{56#*@1G#|BYyT~9@%N^t$zTwazq2?#M zuPkmpDYC`Fg~P2ai(84*@ob;&_t2?nTPgr~X7>Bop6P2yv6j+lIx$y%64dp5)b)8# zIEKvQ&CT^#1LNCA#@dbJ-+4XN*a@9H6_NB+_3{eRGY22kg~atgYLkBQ9Mw0%q)IpN z5l{3oji?&ZH2bW3h_>LbVt&bWNZ~o9@qDzg)K#zT+l=QpzE@{XkFs9s zZEfxAAMGHi{A*#Ia^a@`$vhA)ojjO|1b4o4YLe;hU(k z08PL^k<{-dtrPQ!#v<-eNxaGuYXO}*pimW`6i%ci~&_j>*39-iHo z2V;)AZO3bsyX~ik+q)eAJzN(w`>&hN0N~Xhk>A6QkU{=m_1*K>g7j>j!+HVy_B`=7_Tp*?jxt`nfCPKVYjTMF1*UDn?p@&JwCT@7g>t>(6kSH+@T;h|)scd7=9DHs!K$ zj?t_)xv$9bR09f>a3=E6YChrS3ydJ%#Qx4`0U_uPc5RLB`^Vjoi+5!KbH?<5 z`&&^dwGSUS4FQ}Y8o_JrZ(sOG4w*7yj2}IM^$CcKHfhNe zyx+~iDDdGyX`&-Sp_B0e*~w)SzeOx+#Pr#GBVPOmU$cSS7F(y8_&MOlRL~jz2ErE8 zP5v4Qh6)vDJ6Mwzeh-bZ2-tcLIb zVP?_rQoU-!&xj=-B)?yi5KJEokKJE_)hG{=Kq#yhjG~$_m$Yd})BypW&~4T`d(R|- z6RdISpAtKCCLCIQQ0{LvpMIQsjf5IES#WT!uOGv#06eh7)-q$K}3@(WGlD z`F7`iR4k697-<&n-&^did6AelnTJc&>05=(-Z(J$)s5l5RrkM9wu~5q$3wqYclLi* zcd8Ezg9(h#rIbHN77*)q+kL&EE0X20JlVB(*-BTGEpe1eBJ7#w70J?F0^ZdjbKEO` z_uO=(LdPLxqoe5}2t{wFZ9a=?`n>u!;OO&A0}}Kc6Sv6~&}3OIU|*xpq+j=tMd_YF z^Oxz@!MJnezJvX!_qp+O0$*^ytc;ZL`_~(bF;8F+z(Q^z5dCW(n>}cS-?r8nzwiXl zCF2_&0HM2r+ot!#(mUtc$ferl!j3_cHDN;YLASAX$I?6RT(J4Z^sirbP$UxQHgSi( zRQg`VT~YK;?w5t3h--|KzfM1lRD-*GRgFtZZ^%sD&r{8wOhBX4Vs}PrA@%;rL7_Xu zZ(TjcX)NPo@E7`2QW??Nvg~YRYs#v$G}yroAu&dY^oaO5{B{?fO~bf>!y$_&let_h z;8w9&s$kDKoSBz;TZ0Igr#Pfofnl*#DgIqUA)&Np%wfB#dCiccy!CKvyBZDQ`Bc+I z;rOYx?~m%f{ptD8FV&rvuX^e~Rdjbb zi1WAs1+Bis5aVN)7^299{~}&>&V{5KgU`}=2+nl)R<4yrSaGTaw(cjLHlM5esS+~Y z_Qe1&PVR%$!`#;cV8&oNk6|X7 z;in|qAflLJynPawNVB zsvCXb!o>8SOYC3I=WFmYQ;DshFnAm(>yVUBaILnmC|#CqtOP?P?GGz4>5WZjYH6B9 zD7ws{#38BjX^-{ENQ_Ls1CNXXCYnf6euJBcxe2mOk`L+R{%=wCpD@HUU-4Gs_U9LS zf5Q;x-Oe2>;Q!U-`F~oH{BN=Pw~C1SkBSIp|97nZUJ(V7PoE#-{-mAn(M1!js}`#l zDh5V&mB=&cmR{mr9p;gTN8I5zH?&J9FJl#h@Ag6NbZ|#I0dGaZXSC7L-cp03-m|F= zq*73zk|@lmRy$o-op*100-eAT!>T-4W!?ea!MSQ5ZQozDkKgp`Hb+RrbxY{>9f8nX z30i>~xX{*gGbqJWMa+4&1!K4SobCh};*;*db3yxt$z=|QalTOSRPxh!XPP6F&Prr^ za6XIV;$?T2M+iGc1mncJdwa_Raf>l7PhUi#K)XFX?Gj{GKQ7e8+0>20z_0})l3A^} z`&#qL#GR(AinyWaj@1xm?h7Ti&a!=e)z}@;55er& z*CJCKcqr1BU}k>DNI8hU4~(>FqpV2NnV5>v7+--3&DXGM%?47DZ&^tdC_grq3=&@* z7B9N$*P<``HKcLl_Ya70@sHWk*IVsN)y}#pIkE^ALoL01APz~XgxG8w`6itTW`MWX zep<|*BzPilK5|A=-C*5JO3(bbS(0~BWbTlks`zU~ynd#pF!ZmASh4-=<^Nt02fsx9 zC%o%yWl%M(`A-LwD}ss=5HbUT5h2kJ&vE|pKsv$2`h67e3&HV+NED2r@b_t#&D}iW zfFOZdv~vgQ$fIgCY0=o}8UeB(4lkhD$VZMKk^ygJ=d$5tXCM z3|~jT-H11zjIou3a60a;bTqkklF%zPIZK;?b{g;zyQXsy_{j6Z?~blWqo7?bf&^cS znI??0pa!KD!Fj;9V|HK)r+PtS03!sS>tIK`IZjIme$w2aH(1l%s087qs+mRuz(&f6 zdq%;osE&J7nR$D`y=u~r-rl`$@UGcu{gAiiS~AK$m}UO@X&t(e&V}M4=uVJ;@w2xh zN)2*Z-g#4&L=+Y%=V3W^s`Xb(imPB(=(z!27 zrX)MU#dR0Ll_rz1n9!`l1ZL4Q6v|AU`c^qA=U95?@Sujn(NJsrzQs5mKP+WIC*Z4z zfLDM*WjEr)I%5O$<~NGO#j3`oqr z1+oG|O{G+IL`tkcb&wCi*;5stWRlnbp(IZlbh5qhzRPTSkK8IAtDCW~Qj!YXw0~bs zbcVrY4p#_&W_g}5y7Pj(fh|qi*T0v_yK*UOsBrHvYt%rtn60J*v#Fk47W%?r#t@Iu zc<(h4M<}d|tlGvcaN5Rrzvm9MYeFvuiVoSw>^Qi8iDplcy}a=+1qTE=JhNfrqi=@Z z+*rR)>A$k|>3Qdo?BUlTI3>Q3T|Rat|7%$hCla)(oJvxqwvtmrphrf`1mCXl&P_Oc zP8v+Ck|4e}ZpY}U1cKJ@91d(dJn{Z$O}>DfcBb7!V~D>mn-#Wz(j|(_QNH}pG!Kt) zf=%;CxWbkRL~8_=H0EEyr}2fa5zV2r&cEGJ;XTL6>EL~IWSpbqvOFeN{;Aa$=ch*!#V6-H>M+3H*-63w zx-$4vX2C-7Pa)IIbEQ9%mtDC|^J<2vc21)&^zRv_#$&l;Zkfqs#7M*;^c!va#WY+DJ?Uq2EauXQ4_>!?!6_ zG7{lwiFNFHgjwBNPFAIxZ0r+KmegdA#RJbCPba8nl8t#1_m5Mc?#q;~%In~JT1Y{j zHO(c92Z|Ae84J9zdBqNj{}_B{efZOB>#uztmjA_YUM3v+pS)g`j~#!T zz_XmwP6+-ofsYXfv5F7W+p7md^WE#$>nlc|v_qx}Y0z`xRXDVCr{AjC3{V~IINAXX zRnv*u+g>UdoEmv>?H2A&Ey;Nu!N`jsdOGCJ^y|o*f7~X$m$%6a+w}a&I^+$I>YEl1 znKG|ycOE@xa+hQW6+^I|80@Vz0JQ^hRE&5mdr&mcY5KlwbUS`O28HP%|K3!et#m_L zQLRVMBR?K1^`T>l?cg(PO3phiN3Rb_fEO64MJ9SO@gZQGlH_BT<>`(=IXVV?!3reK zIt4^IV=g=c8$N1wXyrZg0897Vdw=u_yNq7cUJnC2^YJ%YHzl^RYg%u`*31C;;KEqu zky*_l*U9c%#>f(<4GZ;CabGp2?H!6Paq7>-+K9?3-vtyu#GX{=g$B7i;4{ppND33L ziy~f~8jCJLO4nYC`Hsfr4dA(xvMhZkUuFy!eLjLIRfqjN%Xl6HQegHc+sg9PQm4%)vgEsY(XkG3UZKd*(^5Q3%AG6`J zlH?Bei-M*$ucEg?O8eFcZBAuRDt9svJj|U?1zd=(eERa8)i$UelEqVl3EoerZDYqx zd;-i{`y~11RmrFN>@|g*26CV+clv}kZF0k0_~5YHLbRiEBj)3P&#fDE+t1{-+Ije! zYTh~SO0UdSR;&Iv#LepfaFDs6x7U>RI;)u^zQF>Q~l)DeLn5? zGDYxy>0`X1{6+a!cInEjTI4UYYCHfdED420N&uHqzY102shF#riM)W=yllRK=8q$| zBXuCRZwg3eCyx$oc3+i=U^E*@QP4!F#L#;@Br6ljSBi&8;AB%LacHCf*_mH*2^Ee2 z)Xe5y_&9F0ndylOc4LumbyS;0L45VX=N-TraBxAq?Kyz0+;eM09O|8Qv}=G#5Q zprG(-HoeQu+UdN`-j8yb?KyX87AtYQ&A#EIMO_r1q)7S%`o>^ea_D?hlprDfZj|#f zAQv+z4-HB#Wu$ug{oRLk=Q#=Jy&D$jJ~>cyoUnWb1N%lNPduZDdsiPRYM&n6#nbDC znPz29TtcLfGBZN0bEI!BgW_-TMo9IZEi1`dX-I;JbzWMMDIM(8|Bc|x*-eChn-AvrmEtcm#riz-}-YmJ4IXf&)w`)WADw? zKVAW{O8sNY*Cyp&I5p9KJwT$+1#}1&l!70EA^s4 z@$qe~P3c(u-PsQ(&UT~a@*GRC@?r9%&&7=%o*m)d*GO%mp-WdfLIIzGs)9_m$Cmmr z2sS?4+(V#-iyn-(&-gF&Av!5d&x|wD`3D2E@!mB2SOC~U3AEe6+zovFAN(2TECib* zNl_1JS&S9@9h{~q=5G$4Zs{V;GQk%?`UbV#qaOHFD8A#qm6w`lGP6A54O*c`oOWYx z9{dPNzAZgfd`rcvxw{7%d)gXt4hz|XeI-b_p6&T}!n0d7Jl(U=-TUijX=T4olrJy}SA4F$l?21tSw zpC1(&ddScdne5P1+^7`~xtt*M4X$s7i?U8I@L2`-nS!wuR#Le9 zMor*O)?ze|jVY+neyr@Q3x1kXVU}rj=T@wecGi#Ave>g*5;B@<{fv~%U;Wh2^sDFe z;lr2mf+)C0fZ%J~2?aobfwKIBcSRe(NZl#UV$eNS!=firAG=_=;r>IV-d+wBPT~Dx z>O2s)zNzw>TNiL!>BcSxN9CQF3La^vu}#Nn@aV~E4Cw7GUXvH>+a=Z5wKT}eKIgJe zRfZ>C2{pq6JgfDkHulugEe!P|{6{mIpT(EmY0ES8lGb=75vrYVa%t_vC$(F$t#d{p z-~CORG0&FX^AV^!)O4=3``>SO)r6RbVwneg;wK<`um^l!cU(>0kLq|!H~N!`?pZa1%Mgv?qqf!SZsQVPepwS2D)Q%YWE)BWqiH? zGOGnFu4a8&|4>=v%-tF*k(X$giWbpTKq;No5b??R%yS6RrX}hY3F% z9ve!TN(2YI2gxFy2oK;T*d5K@^4;v)kt?`sa{(kOnh#1OyC$xnaEyTSi4C-(vWfRL=c zX6(Q0PI?}@Fezq&9N+!AJISdO3{7;ex2-Q9iMj-AmfV+st^5kG6(?7sv(#Qkv-Q=R zjf< zgh6_ZZfTeB>k_Vb>~ zc@{xzeM&_nKuu!CaDBUbAIzEL6r9fiu65p_9DeU#PaZo{P&<-bsk+a*@$AJT8{ogby4pArz*J6_ENe(*c7mbbP zzUUZ;{=lb7huCd{P-$$ps-C@GZW=W#b(9hjTST`V(+J4qo~q)kMFByP>QU!2?(cze z3jw>tke4H_rX5mrwI*+vz0@eNXpwtB8PUDk*Du>ESK@nubZ(GRi40*XQZeRzA?A-+ z-3>)YDSwfYZyO06sMbG?0TRNh+=o6rjIwAToc(~=1EJcBkr1C(`lhvmV|*aJt8HFA zAYXino;Wj%)3=R|>$_j^{T>2dDm#(r7}o1bTYTSUm+R1_-$d3a5ukb(;X;grs%qD; zC<*hEl29BD+#wTiD_N0Ip%tj+Vjo=)ubvt4BgYqZjqeV6c%3X9VNjV<)4g{6$4aq; zvj1)rh3DhE7ld9%v)$FVU8Vc(z#n|M=hdKDJ|t(>R`k}4(drP_`r_r`Ld32X0Gh~u z{(a-QRXbq2Dfe9ISI`OrFad4=nEr)%_8jv%A3`p(^f-R;)t@)#5;`2`2F-y;LLn@G zUfPfjdw5wgismK+UG}8_mQ1ITtOt2K69|x(9}013WOUZj?bi9fqw35L~;K| zYGp_cXaCw;C7%Ux90+)5T~VQ{*-qmqZ`Fjl9aF~ zZHat8x}=4o%hCK1fu~Nc_eG99T5OG-)9VawcG)-Qs{et-YVfWJ(;SaqR%z8Vt3uvjz zn2yYxxH=D`P9m9+xi~>@hp|MiL7UmBMT-v@iC<0M)n0dO74_msxu-7nb(`EPwL$fz z=;7}5_v{y`^7JpiHvlc5M044717G)+2j`7~+RuI*f2H;82x1C813)QkM25u!T+}A`0va-;Ite|Dk5p+6 zC!{RiZ4Qeo0P-=JqBpoU<Re40Iwo$q72bEaxj%@b>!-Mx z(CCXTwE}u4^X(}s`n8;EYsa8!{Ai<|8`^G&A1){W~aL4IpmdWIg>gQq@C$s4vr@ zRexa8TfR&)!jJuPnzB%e2fAM!B3}mtjU)|7L5Vy|bY3ohaV)LV(TlgPE?HGW!}KX4 zM>;V{8TCs)Qy2o|D&|LQ$@kimTc798t}PDN+lJbZNU{Ru*Sney7V3D;jHNI5HAvBtK6%cihTaVOBoe!oGJ(kVtT#}?J*k3KyBuoP zJ`@;sR+zl-=I%~(z369C@G5w1cI3R;=95-6-ytduIcC#E%(R46a#J)P%#pF{glo4@ zqMg*x=|XBL6#|ROuOe3;raCr1hedUq&bqtiT~-BuHK8%(7q{N$LkS~fU#25I6n4;o{w)Wv%2F&{=r z6R5wb_XH@5#hUL!k?Nc5BdJhYV4>fp)$R8Q+mqfO`o?A{PD3Yr*v>Uhc2>q$907M5 zHR(k^R9-Bz(mSM{$dXdZ_Q{ES#^@IGkX!JeTq5Bq7e8s$`--go>9+W{bQ#I5zHsK zk?RNsGMkZ|A8Wn$_`(Oxl6q~Rm*&cRVymlg$ZB!f1Y;u9B9_i{XClXop--aBHf3@j z$RB&eqZphij6${+P^#h3ap~;BUI!=B z$i5*iT}Fv2v72sEy36&mL&HzQG6byC2)`-XylHl_M)|6Vki|{85B64EOU2`b?X5)I zm9Ws=5Mz5Ej1Iw#%x+p}TRvJ#ZO8X1?MZr`7VJr9EIL1n()z z-lNE+xBA&|f?LN-AxJDHgv>{5dwO|{nGu%GcXG*E?&^(Xm zc|j?Va0A;)QZO~mJ_~O$nOWfcc%-0|n*=`X$!?lH4|Cf}FXcaPzL?4drrmBTP9O%? zElW%9BGN0dWpJNf_v8VWVK+PC56m{02A?lZk{;T+0E^vURJsj9AyBO4NyO!&ux8 zewS zj+bJgt_jPra@WNY5^&YOuO|N{Y%;yPONIVwKIo2J{%k(}%p7*Os78Sds^hxDgN^t? zSTvC6A#3vu^PWhsRyc2eO`%c*%SC@mOVVmvK$G*_3sDanl^A;PT;Z#t{v>XU5N>6$ zuX<{kv@Yagu!b!Zi&$FFgMn`rz03qpHE+5i`}qT)EfrzhHwyLZbm(7JHaAvw2axtK z2v)!Sg0Sp4*2neXB7>^$Zx3XSg$*x{OqMXk2-{^QRKj zxI~EBo;#TJ8FCb+2G(v7_gcjZh>of^n#6SU+}r3&<4x96R5-ay|5)*z(XRU~goVHL zdM=UA+z+pKjARHV08c=$zs&+{W-=?UFNp)>`aXTw>`OGXHtkcHIQrcY@7L}5CyX6u zjkv{4^H+~sF{we>%NC>oydM`H644wZ0HnH9=D*H+uXvXc!2iUB{*!lcU#w30_p%1p zIjyhYZ<{0a-b4?(25${D3?aNA_0 zPH3uY%|M=;DJ1niXCB}PfD0<&y}vQea{;cv)sL14h#Yzs{`N)ja%&an4SuS!hSpPv zKhH(8APlnE(@7D)2Uj(|gop!^8B3f}9y==U`?%ilr0l=~K z#+RV%@$b#@`9cn74spHC$kT__h0MuKbs<;ATgsB0i=kw?c8g(DHf4+9bdWH!2uM4W zIZWu& z>|v@^Qpa5bt7H*HF6&q&P3p|YdaC7Y2FBr0tLbch3LhSk+@oHNm4EiVI6;@>JFBF0 zdy8JW-V`=Xnmc!f`W^ph^3^97cv~FVpfk!ceZ^GKQYqy1fwI`3ld`fRZLz6UsVYS4 z#{_17zw%O@&JD|=qM#E?>B5rCP3f{$!V|;FI$a9Ks>j38juyW@ZzK*7xC|A%pU)dz zVAIK~BYt__=nwz}&;oA!5kQ#7BI*dqpII4MgFu?$(_SZ&uUSF|QwiOEB1&x06HgWR|jXPdVMdUAkXHh<}X^Stl`MxyuGD zz)WC9J^B&-68hBJ35It4LF#nNyawJF0%CbX%l3J?#R)V;cROki(Fp`{-y5FWXN`=X zmc^cAa7CP-5*uEa2Hek^S`V3dZu{d@V;7yrF^GdD>WPrM_#Q$K>Yn$t6SHtms*D;_ znaKc5g1j@()nsfEMU|o>kbGC4NzS000#1Tt(x1{7h$7EG)G6pwI#90ZEUEmyluiI( z0}x;S+~QY1_n$_y1lqm{Os~ViXOa)(hZm|Z`v?nV&Kre2|50PkfQc0R#U$$Z5|9Y| z`_)|eJ<6;9$b3**V=fg=fPD!+cA2^&scxJPZn(1aKEf99Kb8M)s&$P1exdx%9DErF zRf{Y8a6U({>|JixqTXO-kT#j`eu3;mH7IvnMg2m584QXOw&lL0S!S0N9VnV`?RE5d z5kL`xQ$wy^J$MluqrNGQ=tKfB@OX&?!U9NHy!3Cvm}EjhP+SOSg2B_M{p;tVzBT6k zVT3jevo*DCtmq_S2!)tF(;L6sg3-ZYmOQk=>Dlu%JHZ8nDml+1-+j-mX!#=c6ME$X zySKzI<|y!ZkJ|d5WkChf{WeS^v_LOIW!~O4=x%-U_WVQTcg7#}1uQhG7VjDIrbCkN z``n{Akqhqy*CgNRSGO^cckJv36WCiNN?<5xCDGje zwP`?H=HB4H9Z&yJnyq}T{)erK)xO~CzuE~z4bQ`J?}p$!K8|4;3YJ3=m<*&WR0V)h z>_Fk0iSH|cDYs?^$f%eQh)xJjPs4q!TqFq)mm<|kfC-4m%pN4=ovuc7Kf@UNW|GrZP>8gb>|vR7H{p^$TeyzK`PiBxzrcjm1KbyFD#j}k0iAPSz+`>SBxUt6O1Wi*OlGKD5C8|E z<|NQH^4jv^v{<+w9ek7op1j_ECjI{r_tk$yGET(3h^EaPT3T z`7mVe^SQZdPsvmbQS=1MV-j|JzhZOi8K2A0eC=OC^F_0bgb*V?gb_DsDjW;5k3RB! z5;<8E@5|J*DE{Q4MTF}~K|zdQ_R9D*F?39;M*%bBOK~>|`^n?wIqjK(RQP>j1Jw3( zrs80Xiq=WenB+@Q3T{m#NKK)m{M4Hi6NQPk`TaV{273wx8Dco>wlQYI{Ub4;C!!_U z{7rVNDFL7x+IgoWckI&PS2xxQ9#oFAN=8au(!4e23*{c`m_I1Aj~&DU-HLh`e*vlMB}D% z9TB0eS5*gS{xpqx7KTp+-}`2M*DNw=ygoKNb^npYRM;5^`(@NU>;zWj(rKc30t>^= zAV>>~pI!$K89~yI4&2c_! zbG*PdW8b?9d*=f*QpDv7b^UHaRqB${;1Zb=6ZFcG4XrP?)zCX--?p10^j|8b+Uuyi zbF;*j#&%wWbyJFad77PW*$U0ys)nuZ#qBHkOo-19`9ls?=@7s`Spr);at3_cG{XfW zna>fVW^DdW-%$Ru9^mR)TGy#`) zj-bNjSQFtD5kkGuGV719Acit41zN1SaSQU9qUA(mPec4o$@=YU0%l_ju6~fY*wlt@7{l=!+|T1^CCEyQ{}+H4OBD#9Tie9q;vY8gSo?EdkhxV`(uqh- z)EOcx?m6F5cp46#sK%j&?+s)-mNX~q2(gnj?vCTm(#U_(_pu2_v7$-9=R3bZ+a9XF zcb+uxaQZTUJ!=7aE=h#!z>BtxoV?)>hT`n-*A+WGu6LgBGLvo)Ajkyn0#S4insmKp zq)`Gc^olh;Y;%DM{kn8!6^g`D1iP4CXy6ifc%lNH5Qc#Z-r$EU7nK&R^a$yZ)%hTT z87YW$LyNQ_SY^;V-~2#w6kic_?1pzS)r5f)3|W_A8$au6!h!4@DaNO-MT#_{DGc=i z3h@+(6tm(oR`qqBrJ_U!$`BUGdoM5f93!+cd=J)3Q~=b(YdOKBuZDAh^yQ^_8oxg( zfuG-VAAs)so|FFWo>|Xbmw`9%l=J63<6h^l3&a0;&k_8W_uO^-KfLE(rT^3j>u>xp z7jnb{4OMfsPG^-kzqKu6&rg~PAU!|ny?;At@ENTVnhbA#_hpa?A6)FZk3r=6lduSR z8DwAqSD{_e{{to)%ePme1b@ROfBs{M|KI+xzcT9l_{XkkMfy*m|0OCz0dIc375v+- zj-B(bTurbj5l9C)#uBY9_>rp(2n#y`cuzWscWYK&T^w{@gs}Kstl0TMn`?{J@88;a z)1to%7&<>F|JKnXPTh2^^MihSU-G)rpdpfmiok9F4?yt0qLT0Ih|50V@l*rO8&)(U z&ElC@w}p+FjA~?OIlMD;Sy`k)BD}#AO9Shma;7Vmi3#oXf$cy`{PpM7-`rLf>R@~h zv`;*WEs$nL@uuPj$#3fKSVq~Xe&<**GjPTJAl_KNlPN=pT{-Ak&X#bpJFNOt#eaTD_09e(Lb_Im5mh3CArL^2mx?@GX@-R_vkdD3%%<{)0oN#mY{X;v4|iMW=@=-!3H8 zHVysbgGIS+OTvS3Fb2IYC?Qw@UAM2i0!0295GgyK0f#Rfa1EIU=%EDzJ~3-ACvjuC zU=u}<99|5I%?D(>_8dMpMo{b1halHWp&YV)-+wA?tt))cY2KHxnR^j}NAJ5C7fAV9$LuZSwND%MsUznToLtor4 zvEKOH_T-B*M(h^#ru-o1_%8!a5kPHL>_V9(j+N&0cB$*tE;Q#aa&-+ZJl%~FK=6T` zzH?U>+A1e$wE=vg%Uj;_jeR4-TxKHFg0^og_vJenAMU5{MzGh(DPtQv&-mQghihGNd_pDEoq0i-URxLmn<_Rd zKYP1=6_fr;1ll+l?>4*#MNy^Au_sW#I1_ZrEV$hb(nm|TE z<|I&0(2V`?^iYG!vLd@?0qE2n-Pg)zgiIYl01G&8M{cPFAFH7IY%l;>zX(kT|BCRSIFXAt0*E(#2n%IMug_4$D8x|VH}dSa=6_=n zTp0IW@^JnE(G7QOq7%WvU>Fr`^+y;pFwlZfgul_B&9GOG@~sU zP=!Ihn#qI}+l@Ru1cw8*nFHax^FBg0clQ*%+4Diu!)RgX9Zd_F&1?*ZiwM`T5rkIOyP@i{>87-nxh`#B<7^{_dsMnkMh?egT7 z<_KIoz0a{nweR4QIEdS3)=O~#b`Lw7?*x9I3=g>;Qs3GoOr+uetW+X~P81INAsQ@V z#9~5hifx&P^NxLpM+e_)K&)G^Be)6rxd_7+g_h9WK{TOk;wR5aF?3sU_8ghwoI9?| zwp(VYF90ncgC04q{YA3!L}{Z%nvS&aUeaV{>OIIwbgX*_36G}pp4vX)z<_3jyQHT% zPO0p!VymD|)YtW-@pPJ{{c%He)zzu&m(0vR{McJL*-IDu8h#^cj=sn`b{@esVp{(f z1{zyQv}$!{?BJDE^SCWX7c+Z_LT# zqzPBauJ|IxN4I|GMeHoOD;k>o1w222hsnG@o=@O z*6uywO^4WR#rQiPZ)akVTs6E8nn|ytN^+KMRs1|mtnr?y=Y?AtsIB*q79MfNsH~Rh z@uNItqUI}D>L=%r7XMz}hkRd(L+&%ZdvD3CQug|@-uiIxhcX!Z+arLf$5P6(H^PgN9fxzQ07dTMlM_eo(IeV_AS_aTr8T1hu?`fbs6T)SP` ziYIaTIctg8jGNy~N6_Y$>{yoZ9?n-R|9+bJD2cuzQn(cehuuVLM+Z3x=FS?O1LH}t zF9geH6-|QWFbXH_^#sN!?5bsU;=UAWR%c&Q%NbA{!p*8yRg43fRP2nNZ7HDiQK8^U6zpn*RyZhtJlt?-Jp{%Y8hvo>GP%udT?BE50F)ggBSd8gj8|>Bxbv{ zbpz%cCTfPQKJdK2t|EkF_W84xqxBTXnPN&dsn=Gpa1dUJ(nRgDmvO4j8H1K}(+Ysi zQLV(kg>#h|Hm^*Wnq9N|^?i+2Hr2cy5{ypm(hWqTbu|P~jgZYaygvyvyRR{mM8Hd| zXOJg>#fUcMi(A<{rsLnb91KRk38O~>ukH@mj9a7V{f)AqDvTXy+8-Zw(l2DFUTIS> z(d{R3e-{DCwnKm$IhG0@nI3@^Ley`kDuhCLjx2O%%;kV_GRX!AaI>UsQJBH{(j!)3 zExUqr1ih5CB*jC;Vx8|e=cy4S1na&gjqbe&bDl-1haD26kC-6F!_h+3LdWR#y^EpoX!u8u5g6bD_d`qF^H$1SG_W8F_oKkyKm$ zG6ey^(&-40=iSwh9}n3ekG3WyR^i2TPws0#u@aEieo>NN%ha0iOtY55KSaRfM0n|m zP~Q4Q6%rQDb!}0eR0y$h)>y~gh*3A@x-&M}=XSF}GO;t^ zyEUJlfS^~X4c__tj4bCjY|cCoX*n6cLLHYh&J z#Nypmuu4X^PE7HdKS?&oAbF{q0+hQ6mx+~oY}+Gd7w9pW9OY4#s#eI*o#P7{X)-17 zJ6KP${z@V;i3EvYE{UUB_HZF8xG>v?8YqN9?X*OL0WLJTB$^BxinK;*zJ@s>(d^qo!0ArPl>3 zE}rOHM?(TGN7|q3lv_0{c zy7p3`Tyaa{oGR(uWsEL-Hig83FHW8{Ek0p{S`6HBDcoCsnL#JtV$j{WHI9zUdt|+H zFg)GFpmj`RTp#51GND|KmTa*VvjdOZd+=CtXK^U5vi^@37&jW1=zyww?iNp0gxe-nKVj(e%`v)6-65;U38G_gMEW}ieMXZ z(Ky^^-6zD` zr+`{UQge z0;m&Lv$P<;QP?$}->JjjAnfp6x))(!8uVC}CFU&~9uo5@wHZ;riR41{Ubj`;eqvcVaBet#XAR8)N6>OnA>Bo2t{*D70Ow(bg zE}J5S`l{~dn3_sJVhn}&B(98ulVvZy`EZkM2w1MkjS;R*lH>ULh+Db|h%hO>NnyYX zi&)`YnQqD#4@d+86HF)!=ttxm+N|F+(|k;q%2Z{;z(u!43qiR!y$1cV)SsvPshb+ru6<+$>B;q?(fvo~`23)bIs!^9nGa6nOU#ejc9mU)*Ra?;eXI<2 zfVN=8co3Z){_wDAj6j)~0^SXlQ@9h1&}-4A^$j3R(={CI*_@Z=h7i7f0D5h}gfAJR z6B`c1a<4?lHezoVubETwJj0P7R-yL*B8@^3XvT!HkL7&diT|t$+zXx2TsbuvLF0LmOuU6b+_|O1fva;l6su)ETNBPNGS8J^t18+a=*HTeL%&)Bbme?-Zu7OhG~?g z=9q!Infa5iGNVWC0&UO-CoF1~?N&l#cN?}3rd$eWP}A;XZ(1L&wb+i ze<~alK3||6@SkW0t?5s+vn=}q?ObR6-xLlG&Tu0n+t{mTldtBhg1okGMOV09Rc}N4 zav+P5Tg>ywnF@5kb&qd4{Dw)rw)a?y#6Td|o#Wm#KGil-Y1E8@FWmeU(0|* z3lsziPkNaFMxKyfq3-}-Ug++CR-zj-UzN)n}0y(nI_JI)*h)k(uo zPJ0zvA>C^3vubf;B0(mN7g$4CV+8_J`C!%9XO=@n7J@6-YYj>o1^mV0i<2mWbBtus zBZDPS7+6-1iYeXI^P$8G!LXxS@NNR^t1o#PPC11cI%-YHr<*(2rLBx;BX4g=t%#Dr zBBN)m0g>|D0FI}=J*fn}Gy!OqMRfk)B)Wamhx99pG)(5vG?_DQ$S!RgP zqh2Yd8Y5U>+N?8Oe?nBpvB6j~JLnrH&wTCem7$pL>R&vVy`SQn^m6P%isri$6vU+Dx%|f<=hLGhJnqFTZ_oHO(n@3WA@=zw)x@b-luqNt1 zI5||TkjR@ms?hvpU^R*9g>Jg zjJXKzA_-m$RnBjsOo>*ja}0wM=81K8JK52tYb)BjCdy+ne0K#+X*0=Hn|N~OM6&so|dh&f^8e#C7xWG z<~ccHUtw!+WnvNbs(i!z$|6>##KV{A_6}X7HWpO_Bh7<%pR#T;TQ+L4J$ck%%H>!; zi4}FHq5k(Wg_WVvxYdq6W$QljYVgS}% zkCV$3zObl$R$f>~JSRl&(xOQCMKIV2s^R)b7GksD^MW`A{Vy{&gVH$aEQQI=9 zea-6*7A6W@J+ewNzR}N=tQ7x&C8gaD(2~(EIG8d#+Q@t7K^gvv9FY|6$oCXSSz?ixldC_4&p2+4$_OX&(me3 z+$b@i^tCG~OlM!sD{DIz#MAUeece~B#}f(gcjepOk8&%Xxd@t4k8@UQCBeN&#B?XmMk0auaDQENM@P8(RUi0gFOOCV&SjXF!%o33Z-fgGv3M-(cSD$ zz@_aAthFQMA*>|Tb)^@-u1%o-=|30*2a zwKjnQ${Yt8uJsnR$DCrn7T`5bqkS;DIG8I%#0A=(nA~irU5=7JIhlG)@o6EYDw>*H zbrKP0l%icxEaLri^9?@-9``&#U>$^>nk)D1-m{$=72LM^)o-+SC$1*6F`XG_B=P>v$wxf%Yp%VCfFH7V(~Ix#L2s7kdID|OiHQ=M*bw4O6zq(uzL?c$_Hw)m zlLT+ygrtP>-WWkhjy1j1@e`bR-(yFLwM*|UcaxzbijP$EgOj}vdA?10zrdIi>M?sP z5S`j3m}IB%FelC!i-g?#Q)_sh(k)%l+iAGet}vH0kM&}Wz|Iu{+v-l4C0G^r=OuAM z{=x1juQrZS6_3#IbS-U~!cjD3-cA{?U+Dfoa*$qgJX|?)yjbG^aAT?D>TS*q<#C?y z+*F=r8C^;GK#`4#ew;Hj#$sLuNqb3TS2>rG;<8eLV#6oSEk#RQhpieRlBliPE1Z)= znJs)=+cHpI9Yhw;jDeZ$U48Qy&C>svN8L}V@S4(!|0MRzOY zWa(w7?`N9VqxO}xGpiy#IQO$w9qcM-0pii|@(?4T)z^1|v-0(jK3Am&FQPqh z=HckJ)B3RFA@O-x&l{jG!&@-2wGDY&_I!`mmsmVJ_d8ABTg<1_W?!ez&=u5C047Do zTZ%FM>u5Y3gp2495p^yOMSQkQG6tY;cOzcf;w(!T(C8+RN~Lir11> z*HT}>(S-Xtaea4#=(%$tbo^cT>AS&ULG)PqeO-@4xtLg@Rj{GIKTV-N4*<-+t?Kdu z5P={(M&rKSIanBo0X=rDn(@U2$j%_&fb2SAQE=? z(Eb41mR%SA8PJ=6XZCiw{?kOjqq!?b#`A9$YMixE1FfQ~_ZvFQI_Uhr(2_9XkPDET zWvx8{K^F%k;JH>i(c9P}yu3N5L6G)3;rHSvG<;^jolItWxs^A)b3nZARRkEs6F$VF z+C2s%0+D!6iBVj@B$@+#bNdEff*hb+BKsUc!;E%MBwpg-Jb3EO%SE7kqC0|sa`v8? z4L8MSKK|0Jl zPF0yiwf8;CLa{$Z}`K5)KEaG<`@B&c1Wymi2< z#0VKcl~0mdvpd)pu3!F!i?`}yR5X!trKm7Me1HUpD*LtQO6xnFd6tI_8c2spdpz1|)=Y8Q{LuHvW_D2chAVs?_H`3l zeB2zZ!E^7!sgwK}of+Rw(8lOPe2Lw86RXZSbum2k)RqS!QahR*G1=ZouIWy)If;d# zu%QPSw5)?Xs%T>{u+I~(4g&lK^5YU;SV|crwP&;?U8~S`qk^H#VRZM%7CqmaQq7{U zi8@8+(cDHM=Ky$z$12Zp@!7bAKOL0e^Z8Aa zd1gJvz8Z;+hGa02cVK9mBpiKyG-Dpfxt8E)K~WYHM3DHgr~<;e0dorxvX@Dh^~lZ4 zt>4hNQ^_hcUM%@kT*t8dUC#v-B1lN1Fynx8tGZiGpuCEZ#qbxe?uIXCZMx2uy2NB+PKaQXH+O!>C-zl9wJ2}J3Ii5cc3&R!XVd2xp)=TL}coL?J zN4uD|WG-;H_;)DwF+Y`G4={B4!CX?x{Daq`t8)aKB)e}-z=VC|7EpfnH03L9D17(L zuU;MCbxcD!h5%w=r)~yEAz2XPjc1 z_NdoVq?N6Lz?o#)?pkMgy|E#)te9<0W|5%~7cRd$VT0OMJB_`4)e$Rn)L3$d--@}( zr}BbMe3tlH{n3RaiQ-r5tiUwT-Q7+Mw#U4RPsWFLU*O7L`P@w?-?H2DsAps584zDF;ba^(tr2eFAe zeP{QE=^zvwh{4-OZsN`x&pWW+95-|Q>!21mA-qrj`+apnIVd=KLK_^fFn+Y+p-wN$ zRC9gE4*bHiU64)7T@HbE8~*OScCVkX-%jTPAzno*o%$J1oNUM0FlfQydmFZojqNgQ zv;cZ{rnjG$H_w6W90J87vi7VATgDg&AVM1~U7aF)ckA2~R%=qG)`putM%~|;Xy&>r z`TJVR4-#ATUwB*uAEPr6`Zs#3QkOI}k3cXt>av=Djof8gg;E)G&{<`2#N`|7=40X5 zQdj{f7V9!{$1~qk!kLKU>?`@ByTR|OrONKfHvZJ^y^s#)drvr|~%k>zDP>qgDM@IOG+R za}&OG*!6t4+A?Dr^AciHiZxYF^pPM_tg-1%HTJ|AAJPz)-A6n_g6>rj=G@Bi)B>W) z$-Q?sM%6~4^+sp5%B{z{&7fovp}?8XaU)Kv={f2rrYSEMA$NNnmlTb@0bZ?;+MycR zegWogHWLKCR=sxML)Eno0w}NROb~%*c(NZ_Yi&0I<=qP-+C#1_M8aL|eu85b|9$K* z?v?OlagI~6xqut!O-ZVnG**uQ@>pe7dJ6Ljv60GhX(WnjN|)4pX#_sEJt>H^jD0y+ zRj6qUSv~DO5l^RHD5zUmFj$YJJ72D)tgM@-`W}?wr{)p#u;loY`r;>svNp%&juDiv z6`aX%t`{WIGM5&{1ZQ7~H*6{u`|l(cwAo;CSNFt!E315}rgXQ~Ae7R+diWOhllXM>sjU)v!~2dg zoNCWE8Z6yJ2pbmTzpvEQ-zuxjh^}SXc)sbYyc6%)Gqu+GLjeftH&hHFJ9jsr{GCUI z*j&_`FEBnTR>u*n3fldk0mWu`<(k~;&%M`2!J@bq-(l=k8>{&Eu@(K*UhXfi1;*~) zE8>{DV*d6D!Pr7r z0u!HJ<{BSNKAig;Q(|h)wJMhiRcGz?Havae|L)c6~hL*q2V510(<=ev#8-B zAJpdUCJN{^>q9c$#b6@HStcc5>0>cmp1()2{`zdALIQuXLc^9+q$iY4`=iNksOq=p&~mVdX%UUC^L+_pD`FYfK4;LU;k95Mdkn8N6oJn?cJ#b9 zI-F;l3Zx2wLFn_(-nHn?Bfg`cS7JZ3YCHf!(dT;us^?RDlTd43Xnv3u%i#d*hfp(r z=*{_TFz##F_dKeN7ij=fVM>hT0!~tFur2tiG@e5ZhBhxW$MfQI5zN4U2=L6$^6nsD zrhx~CpavVXWaMsE7f7^|chVE-KYKPp?9sY66B5k9Obb!r?T`WQczaPfw(UY_Z3U3b zvyuX*D|ms|HG^l;Up-r2$%yFWILDOP$t$vdEK`C0dYeB(QV<4^-nelv2M~r!p_Iw~ znM^hi#V(haN?;y){|r5pSVRk^7zNTaO>A`is05@l`cDfiX~pW-z; z^%y41^RQ}A3Dl5}06ygKOV2tiqD*>A@s6*uM3Gbh%g>H!6u3y(Upz*Y9}3QQnCtg& z0f~aY8QfiLF=R>RZaP9rgI)0innG$8yNr|{AGGu$FqN92jyGCtlRybrvU5ZJ$uN#REJ5z@)9tfOR+yRz>-81Rh2YHUZV7$RTj%j<~^-@wlNU!q-t1Lr;Z{&%uvq4Mbsm15?mjcOrhnkQXG|L;%Y3Z^8Z`*+$=C z_`3sX7Dz<3v-@uuxH9(xz&N?N~iHd_(h`;KH{`#mCAZYsMN6-G{qkx@D zV491Jq36JTTzz+!2j^Oe|`2+gVy=C&jz+%ewL{> zfL5CDwFqH>WGtE{y2koY_P_k@KQH3`4P?I1KKa#kN@2LsID=}l*QBI-{%q3QaR!+Rt-FJ+X(j7{`fW%_z9=5X>a~r zfeb;H%$mCG`%IK4?iM*w;af9(k4c$`{q^=z#fOVG@5J9750dJ?<^Ju_9Y6yS!|gW_ zNtGX>sNyOin0hSlYUDSwvu>164-;c$KMyCPCwd{f0HX6 zMf)Va2KF99p<=}^eRpFFnt~qtfTDgoyV$WgiKoue=|zD<@BT@>$ksU$XMx_jz4l|w z0l4WkU5&w+89`@$kopLZH+H!$3IUlog*GUzC56iHsRycjxU}-uy548v&-Fc`bn{5pzGP=oKjR@*MJ|s`wi?Y$^ zNxd%J0u|jQ=)j$|r&XoQ({0S6gLz)5uXAWsapnU~=SIBhKnGMjcw^-H&P&y;{VBzv zq*F!8eJ;9+z1h3q;IkvKUBlhU-Pt=e|4pYrM4rBoT88TpN+AgS z>nWlDc%bS;;EvZGFt7=lt_?P~n|T2sl6kf=7sw^vGbA2*AvojKT3{xB#fE*vhtf0k zg9*r>C^wAxa;8vRx)|x()N26Xe#!nZS|LjaKR{rm^kWS31Ur7r&8vNyZjz*BL~(NY z;a2hduT~Q@G(cYqVF$cT1)2&VyX6RR4~pevxl^WsH@*f!}w+ItmP&gKv%OY zoFeUuPfmQ;a&o&GNK5V~Z_pV?Z>*P=huq*vBYYeywQiMJ=PPojl9JJOv#RBXvW*y1 z?8_g@Hc>RSXu9inEq>o&MmrBgVwdzL&c6!pK*>)u{}ep_zK2uB$M83Wcm7$-dEx!R zcDPvdMDowM!;L+A$twQegWQmz~w#dBRR$I|M+n^V$z~1nQK5)Qe@%H7ks&#p$qk4$J$H&%J3xkdyVf zQ|(54rabe2Yo=W~OQa(k>YJ?%_IMz|aX0!8VEJ8}TgPwVBNd4O_9R(J`@?~zOmhVm z;~n4^!6D`121L8{5S=cWP8)nvs4KN_ux>)+0`tsp^)OU4)A%zX2KJ^+@kHVBfA;KCU;{1l$>GO z|9HYGr zr3_~-&)=BdSJE%4kqnoY7eMYuHF8Ney}p`*K<8Y`^?b$7{1>H23?LnV`BMMl46p;) zCCmLYMM;OD(Ou3yyOHSE?mRBtrIU}MkS1+8jLYwcr@|~(0fk#L#y2z#N|rtCAQ+*b?-DHjW}(dTiB`d+ENZV-<= zy_*a>bH+*C;KL$)sKR48;|p8Hyz< z#b`muaH~G?;Bs*M2I&$tvxa2aeHuzRJ9T(qGJ^O?WvcJOIG_|~^eRWSSJzAq6-eK4 zu=ex(#p};XmjjiRL9+Tw|D*!kBY+RuC;UPy9G7Qpt z3t{G=os#Se$RiH}Mg*2g>6{NIt8G5=m1AR0)DVtsfzMy^hZ~Kew@t4y#cl(RCw@R|)TT5zTV)CY0 zm(@Nf@DNS8{;^@28Z_ofs^C&*J!mXDOF3n0*bIEOeuCn90%!pNF%=dCcxH{lcYvo{ z4V52z)hlRkHaGCUZ%KhD2w@Db^5Eid3e+$TW$wd=PhEOxjL*KaI>-cyn%ehTZbOFWefmMn789HAWSRmmgBB18^|nfHLmauVV2lssZhqE})5#}hx{cNqY9$rcqdP?aqJVb*u)O`XkTm;>2UrIF=F%>aj*O9n&i#lZ==stR4Xxn#(JE zfP~GPRb-=BTh-_$id!{(!`63eUr+WJKYf0yoKm4dYwF{(TbR%v`e*V^97UszbM?)4 zf9d^hetI6ag7xouzfkVo4m7gM=9+O_J(lp+zlnbAW)HYF(*~SZ`^+iPnxPCCkM~}g zu3Nt|z4rWdwg>mSD7QD)X1O(gACF%QkePw1e@PKg;NK094L;Ao9?^^bdz$i%05?F$ zzZSU;Qw=~Td3l+pvLrv9P>SflhXC>c7yV(5fpi>)(BO@_Ki6?O(owf)6FEmKCZ1k( z#nSoCU#6OIwy0DTSzkF5Eq}JKzxvJ;5Fj{{pUEj)*>OK~htrh%bF$$ZU90}#<$uoDluSsrEh-?-r)iTgN&?)0@|0*ABkmt z+WMHK7^a8I9V^~RE$u(^SpVu`Ro~Mb)jQV|KkYOov7L4K5i);WbgOYNYbUzVRfq_L zNY|1P+$&&l^}V7@<7C|N?d!`~gjgIKJ~#$4KPEm42_kYn7%!R9rYBjY^K~XDWzfDt zo`~U83^acy-Wg7+ztmH?~aUo*D!mZ%I+Ndb#P%S`61rOvK_C!QnhFu=21briy)B3^OG6Tk`iR? zrInYpCNY)37kZ5stbBm@{;B1jvKgIAw+58F6j-NmFT*}>_*SJ=H`xFCmnZ^WF(r?A zDyPh^!^E0j2Z(RCzMeTNx>uD& zI=`1rH+_~N8Oq;y!R2+ zzhO*VF}Xwz#EN(j5PpF%ZTpzdPH~7iV@m-Aw=Rb}jBi2(ByTzXfiWrh%X@=NA+t+L z^^Ye46(GliE*0CY&p-hW#g!Ybj>QN&yV_~wMUWs5?!I}Z+~_i*FNe!+dbgvG=S3{j z3+kKaFhiGB@wgR_#qy-xpFI5xc!Lyw? ze7ZTN{BcQwCa4w_JHU%`ceT=zj>M%Z5(%wxno)T$-@IQp_@dz~nG4=;!SsQQPO_D> z@(|Uw=*$Dn6(gN{DwI0K@uuHwlF~yrICAqM z)7MJvk|`4<60;*W($YzT1lKD(-ufm!c{EIE{kY*q`R2o9x$cV6rc*4I63Lm!yEQMR zn(mbO3sDtC^h&jpAPc-^(k4P?X5jODO>JE{=!9{;bDy(tTIuXBWAD_VjnGB%I zx9j7v7gG@AC*~suxF%8_%#8D@u1q_(Lv7JL{QA;|iz$dTR`={^75qgy_3c=0*-X@} zQ5~+`^6>_j0M?$P%~_u3_SlQjk=@Y1D9%xxwwrju8if#4rI3u-9>P(SBa zpX`eR-^tkjgk&nq+wr%Jtuk8)3b3ZjYw;8*-dSI!&AMY(l0<4#T!o4 zBz)rMfKGyvGr18Rt3xL&8v1C+rEJftH~MpdwQ*n7z@@G4)19C<>0=$qK*`DXk(jL0 zYP9Bk^WmX0U1)c3$#-*{yp6r2sYN#qanN#GIo5vgQ1UXA`J*- zNlVvD>V_~brTjPz7_-YcwG~;KH!4&F?KrvaU+TifoEC4wo-blK+%EgZ`476Vp|}5{ zVeC0(|GQH5-)R^he7DPa@Xg5bcfr_yR4e{rmH+m%3C#sRyz3t-@(I3v$pq$K(GvP| z^EirM482MI^B64W0Zx7N1TRkH?N9!5dNc2PbgSo0Q z2`C6F7VSx=8Oo`rmF=Kn<~ifTm()h}TjNOndmF93y7cBxbW(LRUTfRvwZt-+`EgfQ zS+|oolZbXSeE~AL%qd;;bfyN9=}1R!R3pIeGI7XL_Z&Rdr;3o6Ey5(vmY<@J)q89b z39T|^(zLvIb6h|B%I`r5ZnVd|IA=ATx93dD_qAe#w*1-7ef$Gvo|hE zq>`gVng8=)ip=feCMpN*9?{ht>9mx2NfOYVHEkc-^0oYEp&P?@j72%uwOWBf>qUYl zJxRqt7ZUptAamsUxm9k=+^N9vE-8CB*?%E0N%eOElNG}yUGb#9ZK@xf+l$a81=KHl zkqnf_{iFShd(3ueFI1=LsTcM_ARKs;-7U90@c&Tul~Gmi+uAE=>F(|jltz(7w{&;6 zpdc-*MTgQU-CfeXkOmcy?k+(>X^9tfpMBi>p0n?JKk&&h@IUy?c;=kXoDY+duKIaY zo~RM{Yf+6GC!<6d$oFdWykw(Yk9Qx4fgP9G3OEK$$%B|xs3oN%<#I#5FAA4vhT3x2 zRvJ7shWo(~k(iT&i*R3Fe?P4#(+g%w@xx%`Bc_%Nv%OFMB8p?S#9>-hBAhRO%G|Y+ z6>|;KRJQif{Pnxjx~1q?c>-UJ{G?}@LG>tThT*(iYk{XAL=a(U+Ghz<7$RD}8_iQ# zv__Ld3EruwS16R_5n9NW?Waznhs$S?_2wFX19;c;JD99W*el*nBqX|u&g0E` zrk6po_)LlNZn`F8z!}41vL#NN;P*9T8H)VT%`p+TudlnIajXY$p*T){l#2MBMBH1H zf7ttYJeWY900lt$zXj14DjJGT2z}JSC5Z#YI#$nk#@ORFE9!YC!s|w%ks@BdEji;% z@X$P5lxLosBs5~XHd;nZ>S!ShS+Y>wTNgFX#Gy9L91f<-8k>xNO1k!0QqR=-vly8s zi)Fmbg(YLG87uyVs9jxhzPQPg%H>qj0iDg1LDV=afER9!RmSf?P1LL%(BzQt1uh{$K+LF_61>YByy116>ZJ}~JBu?cVrt_M9AwZ&{o!;T ztoxlt-x_shcZff5lpI}V4(3l0p1^WOWjvaNfvaer#9IkDDa5Q1XsVO%b6}7+WS%vQ z`3UE&eo8-C_xv+#GCRFgnJ?40BMRjxxUUv6-CJFvkgnwwV01PmYJSkj@e^Q)4t|-Fq7@U5;Lu&@P=zpI;HGBWzfAFR_8t#$>eu`L{+{?_dV(sdu0UjDZc~iLSq)Vo1 z)uwlADw=njCo=;dN&Y6ug8~%zM@gP8@PClx`3pA?Z@J>_U;4_dM^v{T=0Qr}&JAxj zlFC4`7UC@$FJZgFczv;NIGHD36-JTrEn~Xcs=wHvB)SjcrFV1w!)o7tzR>>r7mF!l zYK=lmpon5~eWi(CJ<2D^!85z5&L}Ky*BR}l)xM;%htqO`9CJO1+~H4V5{oQ43WroF zK3=eH0SrHWs-0`9)2K;99IgBKC{Hu^8RGRG`4acuaBJ-9C0PQS_S!(sAo-q4`7vND z&+gQNd-u!NB^9I_&b^wJgKwuu`B+2g>yBGrS4%#WYqz~hIN0BDoNie*YX1Q{H4vB9 z4$$`jG8QiR33pgT`6G(e5jU5KR5_b5@n4ym7?^onTb*^FsveDJK%^C&%G}i7}F2Sq&8Dz zP$ZZV)o}+FLQdjH81xmN2o~m?wxS~FhqfeJq)MA!SOYrTN#GeL?r0%}_pXZ!xkM4z zO-W0Z-mpaZ%L1Y$s6(gZN?fH?*9|n4)Fci+{X5Bx=E`_6y>e*x#B zFbF0{;QwPVv9{&-XZSPznEbcF#Og;sX+5;Zx_9JKgd5qk0sPyd@qr!XI4ShY7~r(Y zltP{`f3jMmJHcRZB3n2Ge5DxAc1$^5=l)&wGLQ#eQRJJ4^t7bDa=DOiWyzX!hJ|Qb0&QK&(Yc*H63e|TP31B+ij}r|4gjJdF<_6 zt3C@4t}7+cJaPOK?$0)){g{3crWh`z|d$a&UCsjB0M`~k;3P|Ba zUJe9LuMh>1<4TwWV_)M=2-inh(}aAp$)67Sskfw=X-x$K&#!a=f1YT0OsSk50%T`i zeUTTq-USwB^nl|+&;-}q+FNqWB7_)i;E$BvY0a=%;191xAX9rl@i0@9);&*oI?TX| zP=<{N=~R|QVGZJgE|zYRgAE`knPt+m6!zs7Yimo?J(7S2J+FLSI*BVD~j^@dex79XNmHdzN+! z&IM~t<=GC}*f?AEng;h;&(gbg1J3IRv46L^TOb9b`D@rhoss{G)rFpd&YEQojHF?d zbamWkJI1mR{}SpBrZiKnS?5|E{51}HnA;T%R{BPMr|zEkBo^s_Bs#wbMuvDFpZIut zq(r;gGG4GM9+HKiG-Fm=TRL6M-~YtL*2zqx$a-}kjqK1!q1lxRgh7_MR^e5@cXcIy zYjyABMDKT}OlE17MC~h@sKR99+k(067)d{Ga`w23a2x76ISvEr(a5!1XW&3OpuOMx z?Jtph=O((tXfV7zYoth8`DAJ4IL1hgX4ddiOZna=s-kb}Dg33~1QI#Qvrnx*MCXRS zUfgv4NLX4eK4m6$A8rylL`EcMXWwu2Ih`NIc|zzJ-VdY0#HSv1Dw(#*gf(;Ukvbww zi%mRwPjN870V3!n)j+oFBX>c4$Qi;+1azpbrok?}8<3a?+zD zMPN%O)DDrjGWsP^F*ChG_PT{u^au;zye|^5%4e|DWiLcZ0Mmv+lhfyP$env7&)gqD zgTTU-Kg=ZUb^MP;JjOvxfc7dMB+bD$~%r4DbVzIDgX?wpftg4BODyTTG zDkj2IOM_yxud8w3;(|p;jL1HMtuAQ>gwDy^H z{h`cO2W5NKuokiNwb4?PjIC*x(EKZ;|M%+^!nano3i!8`alu>t{s&_t7@n zEp%jA>JLqc)qcP)iJSa^H%!i$k8OrlMFT_}?$)7bp5}mVuD)5@EvZIG=E#+iI5U0l z(_lI|B7*LWRTx`nev~=6O^!Nb_g|4NO*HF?b`GTdIJEwpX5Fv5YZG}dno*&!D^I@R zptFiAeuLMm_OMe+5>qDqatT$xGl|=*HQepcv;@78(8(#ll zSLP8ASxSMQ_(RjFR&%J)=F8Hk9X9SltjEL-O|sy{;SSIZ?_hK*sjH z>+zFu#q<#LFEOX>xHC(nELo2!_{??E^(R9i9`CXFbRS6H>%s(0#>$17M>Fyiaf zK9E*&Oru$Pn$J}P&y|A~zE6GfMz_NK@mxH91|2jMdkT^j%ekC;iHn>u zW_y>Xf=8Hk$Sp6zAr-Q9c;YJ$dKdl0lN0p(M>)m}EiknS0k4n+jx;^fM44BVX%-mM zJV8|i@JHy%nouAgL4mO`b%Mak8;4}el>nMAr(?qVB*p#D;<_Vst=sucXq~FLeIoqD8hyqcY{8x;u2`Liw>RE)k)(_wk@#y)Xv&rro%vn}X zgkDYv-}T>dFz7v|vtAW%8>heiLTngohjcO~Yr$vBl9s)_MJLKSW4`4l6m-rkN?>SP z>|mPLRAf+4ccE%pM05E={}K~W%DmyhaJ0ot7)9}yWU@&DKZVV zI%PYD+9|^lhq@`byUvspZ%SBopxrY$c=}yuDqUUkt}|7U-WVpd-*Q3^3~7^vdO9}4 zZv!$Mj12nFeh4yk;zHmiSHP~JX}RA4#AA1&UVq4TSS|}#t?Wf$$~oxSM51-VJvlSO zqGrdtyx%`e>Y~&OCq@Vv2<1Vi{4i8!&xMV|>wS>>_>q^=C(K#(=MO1~F!tcGQ*j}1 zF!-y>hhR^8Y;53EY;_-wk)plH%@JFa@F@Z=0=22{4of?f-qdmSW4`=p{li6yXBrwd zn5CzSP%6t)Bbcw9<6kF&lmKQR?VsxM=C_#|e2?}&5LW~aDSyX@IM=M_jwXX~zlPe3 zeTjT`nfm+MK;Sjc%C78P<%8porcMesgk~zF$;FWZ3dUjT(YEsg1na|rB6LB#XWtg5 z%B@F>G5X`o1}lsv?~dgz-zMk-&z7^gFYWEn)d>oPq>|pd1rL+YaN~(!P4Fp)LgTj) zJw;6G1#dXCyWW!Fri@j&a(%Rrzm7s14g-Elqhg0CcxG!}{%SiT_wdjSBt-y>N%DXF!O zoBU(@V5e!L-o3V)!@U#_LV!Iob^(>wHABO!34ygG9DLGxk||vhZ87`u2TS3yjfsP0 zfcqLRmw~tmVjZQmZ&~kR5If<9F%0#5ezT!v%wM;)s$V_-9qrXAmvOg2 z2EplSOdtgx%v3?$2`fezlV>Gj5^-ELUcuD{KQSL>L1tu_(2wU0R6ZMfi*3M5;@X(Z zV&!G>$)@0`^?hEIRBv6ujWjYtn`sXyq3u>y2xIY9HV*2*ogOmlcCO{M1(tE5u5H@$ zIM1@KJzQhI4#|D-(dCkie_P2LAA@@h4g+*jueH*j zDYiU}v7?Yp9CcB*ULO@E;0a1lZ}*s-)jZ{@LLflbb{aIMs!CK2DdP}ORX?pfQgf!3 ztT9la43tuP&Hkcy#7`(0Wg7NSrPiRIo;+d%2*;-} z{m}OS-X1|M{`8VLDpxxCudM>^bnYsY`rkFtU7 zNcrdo`^y9VoJzksBVinRPsjc^;9!Tzk4O*A8u+X$`N>Ui>3IWKO`N z=msx&r|z$JfZp)6jITjAhmg@_C>1a4<_F@*+J44|;Vde47R6@1p437W`L%mLP^`iG z_V=jDD{c2yu^-W1UoP=(O+3EB&A9K)WdWNQeAv?6KaKtg*BRK5o4Y58OMx~9 zu9G69i~^dkkcn{?+%aT=mqTCBG%Sm-+D=CKf{?hDfmmEL1YrW%$Y^ESqID$|hQiqJy$1*z3FZT1nvwP@>foo|2f?G z@|X4*8tYCeQL>C(-tW9eY5+ap>L+PB4{(D>{GS&N|2N*_KP()|jZlpiD2r%@$knoT zcwso|%r|N5z>yRi!pa8Lb{m$a2|B!<@_=8*kTXq0hsM(0Z27na+?6Fp<`BvD6!Zctko#XP(3oSLlPZZ7M((sSO}tjcC;MKkRPI|!v17@48qZ~l^w=2 zVQr=xDSVQpBlzB(E;QF9z|>QO&}J=K9YdBrMnTGjHX2EIdMyrJMJ6gn-7*5Wt;Rl&_P{RvsKIQQ#Sr6kMsqRk+ex6_u% z{`u}JSwZQ^ThW1p7h92Td|wJPV|CAHQj7_YunLka-_m5H>FU}RW*f>B=N3)KkfvMK zaeC%tY7Pjo##Kq#?3B0YGFgvzpDg>wh%X*bY_R^vI#9-AClXy20x~p?(Nz}`9RC}?z(P7H&?K0M3g1#O9Li6585iXC_ltINiE9< zGT46Dn#4E`HPeY%fh*_1CLY)-bHj@Dqp^f)QUH4GP!8nPbDSKFLHY+!2uQ z6nG#}zC1#`rZ2}CfFaSLxl_X&isVacvB<>O^zLXgU7kBDqXgKe1YUWfu1MHlVpWlpH-PV-6)+^iscEpO4HihD+353X zuvnzW0QTf{O6{+K+{v~qmlCp$K%>oWycfxW)`*M z2EiP4W@S^VC)1V%Z%$@x3xrQ*okWgb&ar+`;D6z|NMWesdFt@hO;LFBihn^}hfl!O zmDlaeIh2m!Y$fV=gV0t<{J!pLnhquTMw)PWqkh+O&9lvVmD+@@^5}(P)9#`+-OaM* z2pt3Sti8rvZNw)tYYmC_n^szbUA+#X4i#H$bEE5y4nO$%tQ>qCda3_=U4sn-1IPiw zf52IHUgckj>G9d1ac3YZ3FHTz12feI4K*+V;r;5RA`7HCVkS8gjX2mOe`UNon?aRjVv%G zptjX3>hJ|`(syI05b9Z3MX%;Y=fLia--L3~D@)w`d)f9pQ?Fk!xP8YPT;=OuSo>Pu zK2seT`LIR)^xO9erJ;xI*VjF;$qe6%Kq9#KSBfnv=sYaZSRPUHD8u_0vEj2H@e0A> zD@F(FC0%r}z6Iw6h7dh}=4<4U=Le`~%V9qnWei`drCF`Md0Q1?7IEWb5fdql%e?kR z$d>T8NO|`h!~ou(&jAFu^Z)T2msU|(V*U!0a;eP2G*Dkm4#wBa!|7s?1Pl*r&IU8! zMwq@p<)Y6^WdU?I^Rda#X%t=yJU4tSgJ7cId$YTYFQjTR=*#?oQ8KJLPYD2Vmdhjg zw$@Rk&9Z?2aVZ`NlUFfBXZyUP@EXEmI5w?b&{vBtY?;Pfmo*#8NCKMw;aH(tiIk?B zX8pjivpoclj;E-hYPC0tPCi|s@r7xWGRDJCueHrqKW1{=YdPhrfy7?N094bgzl#@~ zKoP-6GJKT~_|qoK%R#zKi8`wx%sBe$ol$*%mI91Y!AxVAOvMq=j3Mr9zuzOrwCN)8 zIe3_wS+9W?pL#*cS0L5~4$9>5I~9@AS7gMvFsM^XM#OaP%UReNk`dQ8;Q+no003+j z_BkF2Y7c6(iu8d|2T=n~=c;uB*+>GWMX8H=W`bd_G&F+!R3v7?kgIZ&!dd*zwEwK? z{)rxg4Wt8*{P7^Sz>PU-2@XKSV$~NoAiB*^IFt%WT-n{h1VpN-AqN?~kwgr9mone9 zL?cO2(dhee*X3SQ=)F^CQO*94!iy#Jnq}C0FanJwrska6Y&iL`QX3C5caePf6OHTd zHarY++0^2sah|oN8hPR-4;30l^ED&nlhY+roOfq_18$oMty`RqodB<^CJL{}@>Y5% zHWSIzDLWSgPAiRxV9!B4&_Qp7Nk|Vd3y-0a4mOttk#dT12mJHflb8pIc)2AQ3h z3!(u-1Th{EsTCS`hSf5ECz67MzSFHGWtRO|Z+80!98}I}HFmE6&E8|yrp^Gie5Ozk zC{^*2eaEOa=*BYlK8g-Yv5_E2&-c2{v{qNG_!d7pey&O6BbK5D%9M!pAVpSF66z0p9-t_yw?b-cfX zA+&(yv~Q@o59=%m#%Q5-%9Wd~$e@~W7X868I(v)f>xXU-WNVFl5heII*US zl0bszL{VLft%=AX2}Zq^dDdVat&%NZ5iPB$vm7QuDwvl*4=Tz{6vB*SNW#4@Yng2A zHRi!IA*rehvEX|=qpR#7e9V=og(j6 zI^X2MJ+{VJ+XqO#uPdb)bOzN z4NyXYF!P9r>}h+F;^{#)C?T{k2-AqiOB|-vNU?DjC0q&Jz7@TpugGgLqJqeTc8FzK zNl%|Vw6Rc-w(Lrd)HB>>kW;p;VF)ow2vtpxk;BLS>wp~nZv%1w=Kl{=H#|x_%ozB! zxpzSHw+c2aHV=}6wU1*kP-7#Sh~Ti$#6O-7DFJ2*%?}W438SHtauqqbWPUEU9$jg< ztqf=Dks~gJO6tnzqg`~~u`dD2^}xspA1Dfs%a=Rx?;$7c5FO|t@7J(O-DW27gwuRD z?*ZVx(O{;yc*}T~gFhqR6(#n^x4Z_h{o;mK{JhXbWCW)A1&7HEVMZr|i}=Cj29w-M zc`5+Aa_!+}!ZIJf=>?_c)>O*pS|Ui%>Nrp!<)$sZ)qB*a4;d|g;JW7gRQ{My-7|7P zn-Rmhb1_mt_~u54CJ^5s256o6bdsWQ>6af{k4RDn;J;PRgfwXsE=#dLXTo-qgH6o` z(I;4&s_9}_6Phr6>s1Y++D0-9Z*_BDYZrY#a|b*7`@Boo$R5)J*5HDfvQj_$p6Dz-6_M%c-8XXT z2nUzKoj_js?|L5MGuL1ep6ynTS(bc-Ni#OzNhbUnkl(Ra<>NTYDtDeWAIxO$wHz%- zu$7_xUK)P8jTJ(-KY!?${}k3nM5RT5-tF+AQOU*0aG7?)@GzuuLsXW!L(sXy6sD@y zdfrfy&9F77X14pCO7XouySFYHht4q&AaJ<)DdRCO?UBxBHsWSa9ml zGmn-Ay<`Cvibvn9x(jY>#R8iu*Ekd4pFwi>*s-}_?M-Nk=LDtm*|@%C2yrOdTK__S#*FN8!j3jUjk8$l};Cp3%y=YE>u=^%-A_11J+JB z^*e(LkbNQrYKtMIP~cbLJ^qW%Y9-;YH#ehSmjIBt8UCRMJ_Z+$&KZVZoVXC7D5EjR zOqk-$OOFfdQV>HOi%^lNl{s!}5Oq2O37Y2*&!t8KAXl2SVzMe0!#Y&=Uz$p>$DFMw za!**Vz7e?FHGX_+V3812dP*Hp!elys48*@th1qXo``@X;+5d+omN$xhKHQh@-ywju z%jO_)_-_5XMk~7TbAfUWB@gTH1YGa5(v;VBJfl4e_4zf~4i7}9$a?hshl}EScMCHu zq+9W{I`80$+@ zANj7`_Mfed7E*zlWO?CbT3CE^PrSR}jevBjh%3nix7CRv|9eXACE_NXy;x+8E9W~_ zBRM2k8BZSf!{HJu5}xj7S$^TbUM zzzHSF(vA~eB(Z?vyyZQ3>W(SeKo-Ebff8>;Kq;CYn1Vx-ZfFVH$_&}qE~Hm&i|Xyy z)M*`s;DJWAmPK1F8rKqJ&%euS7O2wz2o%&0K5?uGyCrC^KHtt)g65HIrrNmXuB5ov zb;WTQ`Gk^1W5bZFMS&YO%*uh3#mu%wyrhuq@VgmET-{n>PP}FDt>aq;e1K$FcHsx7 zC7|Kr0yB$?i%KsLY)UGBZAD3iZRa$!>E2O=(c&{b5O;Pe$#w1)`xsNytq1sKO+@YG zHg{{Z4f#nOF;z>-%tn^GOY25m?N-y4-m?}qe2&(8R`QhOc-B3ltc1PgqKIw3^yn?o zers*Hj$`X3>rRS4)*hMDyF)3vgASA(3%X7W{|2tE-a7f*R{TLyU7bbC*SkHG!9N%K z?yhDPlztnatTzoGJ+(=aYMJ@Pi%+vbQN!vQ~0d zqp2rz)x$0rJiKF)cwNgPqC|nUxjBr$kg3quVI^ZmY@SM7&^v}33;>qlGAN*6p$=`p z_^{xJTp5E>?%F=RSLo1tb+X`8`6H|`VdYJMxm}3irAck)0+%!hB~#? zZv^>RC@r`k10;kCb|@5};;~N+?DUbbh34SYhmCWd!OV}+Ph?aD=MtQa zA@~E->Y1KljSfm%djg(K;=yfCx*P`b(6I+b4~mS)Y`XB+|xBM!KPJ=yRTwA43$NMJw<$Y=fwTxjl_dB_O@e9Uef5kT%zFR`^84q0{?_x_H0 zWGH&yM`|Ksq+`oWwc&vqooFIK*O;*SRJR0p?P_aH2`bt6Hjbk4Y%QMa<#D7rlmlhN zbB9-II7&OAA1~Q5e>*A8y7=ezdor$?)CkMZ?e_wyJk&8JDc3y~Ynfh(LHVLB3Rad>gQrEaxk;AAMY;@VV>SSjXRO824iR^Q3bdEz#ibF}*p9M*8(ceD%{p5) z?Gr!_8+8*1PbQ;%%(kScJq|^N7LE{WUjZZ6U8@|P!F&AnholxM0`VrRc-N6?$637l?aMO&^s;zY z@jYg#b3L)6ZTk*%Y!~?+7V?Yao*>jTHph z(@vtlKi;-Mjz_DDJ*G5>nz3q^jh}Wm1e`$y-!r(r0UX(?FyCWk$<`F!8{-TJ)0SBrCA9X)Wv|BbGU z4MYNv-NKtc8ClTwJ4P1y+0fE^zjoZ~W~R!lZU6?AP)#C&GA%@Q*G3%$1-mkq2@Jdk zc>w7OLS$>C1Uexzn>b9?9tp&AydZn~K!lHFc>AZ`0iV%g@Tumu7t(-BF{A`k217WL z1?aW$>;Z19{a@G%Yga^HLy#yHkmz^c8zt6fBd%!V>IX<7O1a1#mCg-dVg)^OKX4h0 zZgAdNc=NViViE3tkE;A|0mDlZ@hjep7R@ef5tglgwbtVQF;$q#4Wfr zSMgCpNd^0i1C|NK4As7q$Yed+MXb*XnTWqTq>?R=E%j&2`=^5R1ONi?{v6Bw5e$tx ze3Aag`~efk6#s5j^bJtRHr0JUqcw_|DuVt@ezgL?X-2y1{@D;00t3iiBt4{k?|UPc zpx-#0Aq?q&GAM?PWr!Hz!{pLPA#^Hnp|~vbZ+S>!(n(6Lzgwvm3ukJO+(#rtNz$u_ zKFzl{duJ=3>2|aJsxihgrmP=|Xo$uE1{%izEs{sNf-t7d;rM*r*E{9pg!pZgAX&&>DNXEyqu&y0Iz zLVxzhGnbz2YxTxr1RZ5?vurXsXh^YHW5YRYjS#tD^b#ySJ9KC4>iz)pI zn7Di@TSDQ3z={KCp71Y}3nBp9Pky1*pO^#m^Hdtmdtb~y-HgU`uJ3tP4-8wxXIJK3 zPAiNFBcJ_0O~0)a0-yYMwE-*-q`!<9i2yhN)Z71?_OFw7j7RHy>FFpa+mtvn;G*BRjcmyrZ_dLRRdv%>q3A!@_@E=a8 z$fH}BheJUNc?;q#7`aKJbgt?vRz$kFpTmHuNh{%8J=)me><^i4BLiOmy6B^}tram* z>HrCVI3WonN{})jQA{MiK|kh><`bu``SWCH(Ru=(>+ij(yT^I}Q1~@9`P*X!wPRm< z$yAv_y$HEUQw6wI;J(y1S0EB&TG==F;>C}3V(8=o8THEb&%b50Ckxn-BJdrg%crtQ zCyX^>TR%+{^2g>8RLhskMtbHw5#Lo7zY~P!@MeSh7N}sGzXyztlZJ@QH}Jbj)YkiSka3> z+J{|wikLDPK1(vzV+7Bs*jJ!UE?{KB%i0(WzU))FF$!;_oo|$gUzS!yn3v#uUVimS z>W*;Z%u}#Aj;j**^E|FT;D#X~!v&MrnDV7`*El73kz^^u1&jCqYI>^f1`R~kgRjvJb{04) z_OVU##%?I8dkwr z>2oWrrxojiLKt?0U!1ICk)o)$eYg;-w@3Lf7TH=nnT-O?99thbbydN7A}LzLJ~q3_ z)C18g7#@3DTfpGhp9vh6VAjY)d%#o0Rw2oWx2QkJytw-9U|}TuHxLr|`6}ILS&31* zle5$5z4bQR;mh$gz{lrQP6KVeHy^z%h(Ob!gcEPfaEbE#B zdy)j8`;(jYBtWMzx{bv2NUSA7r35D6Ary7))4_p=^@IU9FWiQe-;ls)d?I-}vcn;A z-Cd9u-NczD&WK#Ze)s0Zy0uQ{%@W^;YRZnggrMTnX<>`L97;4`7e&NF_wHYOu9=TI zW}^FSa|5D&e{B%rB#6-EX1n2mr>SSdtC{+op?_R1;ETr6e0CqZ%955>> z3-H{^d#rO2rJqV0yj}3b2JK6sMV=dh)FOqI#*5rK&sY@^pPPXEk{VU|BJp}D+s{F@ zCl{Z~B6Q*;01)y3genlt%Fa8+=tP`caUr%w(?KxN5td4kxLfv1vb%0m|S)>BZi)FIj4lXV54;qA;381fz37 z=Ssh~+KA$3Lm#*`YXQ9T&|iO)p`r*CzT#6q91>M~lqRA;mFE0O2po%2GXlEg_EtG% zo$OQv#G-!^ntfe8a88Mk>I5XJ)13P7_HEzC=(ANNVK=K&ip4k7ev0)|5_o9=(+Wg( zShO}&r)DAhG=c&hNQkD+pHR-6l_pBx>zSL(dX!T1Q7e5~n37m@sD7wX;!~qxAjok( zsL)~OS(1Qd04+e$zo$f-u=yt^1b_j-jZ{#yU{(P;7LI<$1PP8A?G975T2Y$pKGUH#&ZdydXr3C2>K@~^CPJRIG@n_~&&1H^SmjV(I^wx7((|y?wW8*STiu`j@9SF3_x_jiCGtiNKJ{>Lmyz$S%Fy<`kfU0OM>fqBU zFoZe`fGeX$o-i5j2uAIrQPiwDrq$#c=Uvt3^mL_jQR?WUQM$i}V)FYG_uK7*Q2lXH z?+*kAyZL?kvo*X32`6MhtGTO>0K~kyKPy*i*(X7%)>zn;iChMjABacjKN!QEkmNxx zvPF{M`XG!WmF#wUu4v##RQBY7kWhWvjwI$njRXz&&{pM*Uf_9mOdjX25O4v!D2ul%)Ca2{c{UIS@oy=R8t`Wd1pG>Yj0)M6CY^rQ2s6%qq(Iu_B%IaufKwr%OM<1< z_ShSrSC3`BtqjC4t0bhdJk0+X%f&6#V)Na!5BAuhIF}gRWH=Mk{}SIl)@C$JKI7YN z;lq4*@xwgv9 zp|LmBS9a#fgORE2r=G0_7;+<_a!pfr4nWbE_;joGMH~ItR%%gBtJ+5h>53^sgdjVw z?R=jfJ$5~|i@uh*ZgSg6cl57eiD;7!lntcOKv;+%ThfdL59d$`ysLBRrp3$^=XWyh zs9494wXG7No$uJF0Pm|reQLe(2Vn}RdV7TTA0eS5*owgCmRga9=RfNKeuS!1-QpPP zbQkG{I7;v=NhZ}#93wtB&pi*s#qFEN!Dz`zl03b{od@IS^icroLvqOon!iq@5cIP! zy)|Yiwz|VH)w71LuGA^qOBwK0WMrE1>7$9aCq|qkXBuYAg5LDd$7{3NnEfP}ztOWj zU;o>A%A`$|-&y8Jw+_Pyqq zN}lfQ_An$aw8M*9R|hupj}^SlH`74L-2@~HnYtaMgt~9mjN^ebKBidK3g;x^1@SngHCeL+LNMI zvX|suwNh0`!WfFquH!HO1(hv)OffaA%Ai1&`dsZ;+1?tAEBkXA1rf5f&At-||VYDPv5A!F$--}qn#B2R?te%vI#Le3DE zna7jilBGV@=0azDq3ebXVqVtg)6p;ioO*9ALL|_Si~(^h#86>aQE$2~NSJFm?5U2V zwi{enJ;6eb0gOQHSyf1<{ib%pOxy`I00CQOEeK!t5hE(T7V^h>6H!jg_-Ipq78?r_ z@ifA82zn~-3gkJjYZinG%^V(Y4GPhDYrj%sA!%2~!jyp3DzKT>>=3w_u0Sy|oNW1x z#o7}31Yj*H`Z#VY^L10rh^Pf0YmwZ(`kI9V=ZxTXK67_}R6*)r_~5yYVP!2>m24f zfO};U3JniTJ0s`@2@18}tq&$~Xi2q98(Rg4kZ^09kt|yld#JrQBGNVK`j`h+518jV z>Kd-!TmRI5kU8d$?x;dN43!O>3PyZ-w{_sWDUJE~{N<3=mylLuzURhE9;aZegub`s zU(cb@n!;vMdiI2Lw;*s*j>K%Fbr2;pv6 z0$L+p0~kjc&Pc>G*q+G5%2_WM`h&B;1k0y$;BpF=Az?(ys$~yk9_(4WLquaLA5b!J zjz7||+X#MRSpzMB0=-m{=_z(U0-gr74ZXC8U?_=$s^x^XWRZzYo_UQ*-CwqV?hc(Y z17-f%y#sFO`Tohh8@cFyy0`Z)_qMtC>E1cS{&#nbSBJfBvpX~EVPd-x(8uC3Sc7_^ zT(L8Z;L(`jEjEGDN6Yb=Os0$9^ry`_?@Tqj!)Cw;r4>ePQqe-W@=3-+Eg^WF5xCeXug^IuB%H0r%Dq}ARFx5$Hj~m_M_Eq@Rnf2jNT7$r8SJ16MEL_N z`_|tQE%a@q2)vSwPH~zDCDz88DJpOQBMon+Ls|1S#QZTyx3a>DJfK-2e4dk15mLpY zYy?*<+j%A<|Ge&@^rw4YnA`lzz5mXN|JN-q%^lkGiGSbn3S!%Hl6d{=mRGsb>vBD# z-!T^m!1*&M08r$A+Sncb{qN-XpfdYFa+Y`W{jrRQb?@Ax%m)2V(AvfXzUSQ9*gf37 z{%R{pFwXCHun&U~aSOz#k7+ivJzR_H*0dqB+q3qbV0e|Db0_Wwo5?N^bXe#sxjq?88n zM_b?eo;U;$3BGlnqs$>+TyH|sBdnjbt}*34C&KcP@)ST;&`ZzMGQ;GU^ht@4I1y~Vvm1n$maM~UvS zyD{{fb1}^m>zmGSVHuJuxnS*1iS*9X@gGemA9GL$GU`5hu?%@Nd)=dY+Hxi_(-X^P z^r@xs`{r;glZSwC)61}vayzBu2d5|deI$lA*H53&IxXQ{z2V$%X*N3q-mfs?jC$g= zJdKzwea7YsxIu25%z*68OY6hVF^E&ezIAVA!iIon#-g~mYf4h>bIyappmC++kJO`NX3J>vRxFG+pv9At`YF*!+VS=H%!vTht5NSk* z9$@GOfgwc%32C-CLxX^HN=S#4(t->rU{TT_4PtvxJ3X%4~WDHbv5S%sC59Y4O0D&DYN9sduAez(;hOltWIlEB*An zy&}4ah!cD71n>;Q1 z60lyBc6|6L@K(0$XXEpQI%i;luF;j40V!@|S{9FJB99M-;QJi(uPRxK9;K z<{LsWi_=EG`=GW+Cr$h;-fp<8oqEc=j1Z*4!uJ%fd=I?yei&x9V%&hYLY+dR{hp^iUEb-8rQ?9J4y)n9==JF)5DG{OLnV$? zCY3)Q7%kthPA4QkPGMXX;-w&EO~3lHuQD*}&MlGX5%!wR_aVTOQ=eGn{;DsD5YFba z?;eh~A#}$q;6WRNajd3^^833d-1T*IqET4h96v0Ano0Q!x9@}=nN@0<=ce?0sMmp> zZzyQ9hHyUOg7-Ba|3%TVMAvVZSaCMKZ9+;g>aD7JUSv%jM(lzNExsj@0)iI^&C~=FV(h}qa->z zq8tQe?06V{;E!4_C8-3%g3q$?^2({JoH4q`CrfX$p7e%|${VA2Ej6Xs@l=I{(3hMA z3egdzzBs@)r6pUZHsrcD2%3gFZflrMoWXUenHQdxF(KQ~Mw>Lk& z^EV8jE;g-5{PsjH-b?LDZ4Q+QGrn76E?98u77=mXx_k$?izRx{eD%vsO3A)rMbyvX z9-+^3+vK1IvQL+kq8LwIxjNWW?;(Xj0HwANzeL5$dIG_1WboV_HV9f2i3Dw4-y`X| zKygvU4a;Zy);M4PsHTfFT%|0TrfAgAT7lx?hPt{AKb4V*w`;nR#K=WnU7WG2L0n2;rnevaFLXIZ&{3-Yc(W_)S_HpM>gz1J2Z+Mf?)xGlY~%U-p=7$ZflFW6AW z>A#$;NiMS-ccbcH_2HeBX4_m}VOgdB?CCwHM*p*?SF*0;+(^9Mdr-Y&CH~rMWGbP} zvA*<3;p;4pTkNAU2_=f@(ro3VqO|#y5qnUcvw^K6VXT*VEC*I9B#`BX6j?Du?Hzg1+p57=M?zZsGds9PLTPzOq<)3eklpCAv*LQqT z+MX{=u90eMj&mRHRPmFnUCUD8_leUqU2qx{FsTaVOJ8lXsNVdU__MiTra$Py)HC0` zn!W)CqY6m@(K5#(gLFeCeeMr8rw8wm5WbPYe+yK{@&S|ArB`}`N~%weBrG|MGmfq3 zWbn`Be*nS@UbdGSNc%4k+$G%dr$|3JIXU3U(6TS$vGVt}EFKHimY!!1JoPKe_Aak) z*eENpnX)Urg(ShSc<|B%%CYa?6T{1A8+C6TK-`g`lr3vmD~7RsT|2(2u@pt@(YW1% zV~)iKLF{yqi}GDU$;LK@rw$=hr@;gZnwvCTvUOdFe6Q)TkOGe6JdEtK1j&~q7%fss z%(>sMWyA}iS>2UETCaBIfy(9wOt+P1rLuFOA8t^@tJgu6@A+%IC=PdEE1Wl^tJx!V z5x&|ty8{+cu*vL~2OBn5kdQ}NJ3se_Ug zWHGl|VQZC3RKaHIg9fUqTQ9hWY)6|$n|U6Kq>}AtFK*gs{!n$DS-IHaLUcN*+DzDb zZie(h26LVur}#8Tl@U{MUsMc9*Dz^r`rdh?n2$;W2L^%Fa661>QH8wwBO1pp)9#H> zewax;#p4xrnk}Ke8_x%edL1%p!_EMB6OAxj`#{{v{c*#yku3BppO=Murg|ztMxKvv zY6$6gQS|rB(!4?p*ZI+07nQh}Gc3!QoNDx>%>Y6}MyZQ)>HYAyJ`-jvZbXek0AYi~ zA~WmAIQ~&$#W?@4&9QCQDgm;%7xGvjdF{~qt5qD6a3m7iTwXUMU_ak ze}IbYM&o&MzewP@WGYH%(*O(nhDLFTY)Vkfr z@};EXGkZK1a-W1$S2WOhdSg~F@#bVx$0`(5h|)8D9!cX?CnYtBX^saq8=#w?C>7gr zbQwWq+~F_tlrFR~@KM&FK6UeCl3diS>B_gWXh?&lKdHtN#5cIgReyRPdD0wy5wht% zf=}YrCOGiHw&O|;jJ4Z7UgzFJJtKeEmp!yGU*LE`c15O9?oG$l2y2RWPMoP`vk z5Qk*m@2J8UlfycLG}CZ!p@Ky!6Chn9+&?MK;rN?omX>N*aVETDVm1556VzI);A~Wh z;his&L+LjTKitUC`}KW0%kcdCpWi28`G+YYEecO3^LBsDO9w5uH$YaTS9 z+@Aat^-BBM+_e+4A2#(ES_Hb^_|-%$p7>fEqF>*kFI7}s_6Rm| z;Z@|NlMKYo#HE2$Np+}-zv$}QY%!SzRwfMJ%W%}44TL%E`oN=OcMZz{(F)z-y(Y<- z(j-##%413mN_N$nZEBRujX*?$0UUXwjcx_1%Mb1-1ghH-8j!*C zTtjxfcxxZa)w>1nVn9#V+N$yHOcS-e zN{9$mH2O@>LoU1Z#$;?=c$a@D#D~-XZ>ZKvVv%LUI!_xRJvcB2{nXxRgTgm65OU@h9O`#G`kDa7|lV(H+Px9vC7c2KZuNB zi3rH-9UZs=cyU}C!tS&XN#T(KfU_Mo>Zyz-tS_SIW>@25M8;rw3MZ4L1-i;q%S%Y! zsHUZOW%`>4MMd_qWz=~Dggm{LvR``xfA{BajpOfuDR};7ew9N>E=i7-4(0zczg`m_ zznOZd={A>I7=Z=UPv$GTQkLMU@bg_RR7k&UC-zBZa15VI6!qmgv;NJVH05l`ItIM8 zQL(A72IQc+!las>jI-_g7y37kokmoZ`*!EYs}wFP7CJv5JsTMJ(CFp(qH(nKy|%FX zU)``KD&%}tNAppW&*Kjo*w9Q)N)-^+i&if4fTX^>3(U`@leWjBo*yO1yDQg%xUxt- zk!i^ZD`WC!GlOnf6u&cZl)c~=?|&x&DpxiJeKqLZiJ^vkOAX58?Xo_$lCZIS)L$K# zWhOFWeS6}dceqthu0EhV3P)HSG<#UobZ6~ry_!hqnfFX6}3rzpj%a+6+>y}U}V;4hzHX^#NFT8 z>4i5_>br7tQ|sFnGd)sNfErmTO1>^$eRym5W{JA?kUa~^bMC_$O3Y(_?e|oZ`iv63 z{H61RQS|hl+x~se|K(;c?tjGeM4bx-Iq$6LoZYk3v0Rh>VjGSd@;oNvN%3nOgWjr( zx`(Z1pj3$CGC%F&KmprLV({UUQ}ts`AU;w{;p+`FeF}9VuCEx6P^&P`Mtx(a<1vaf zq$=I1m~I6198=5DyI?I;$=BucO*Vz9+(AlYnkJu=dC7N}*Pr7YI#YzC#4G%3S2@Y2g5SVCVc?T#S_4H_(geCK&^wZFxb8O` z7tsA}s*{m;$3rWEHvfSHz zIoq31LUu+`xO*+|=WKHt_06GJ69#{B8}AvPpLAPzDsb!N6fAA{S{5!`W#_%^PvAti zAYFJQl?|igzOG6&y7ryW=1~uThO^gc)Q@{gYDi?$utOQ7ERLhIpz`Nk^4<1Grcqbo zDT_DKcQB+!_p9(1CK~35JEWM>8zRiid@?A#q$gMY%sH8E7xHA@{wrrl@LNwS0Ze~i zxpj7bgY(s4C*-Q)(zO_sU*D^%4NQi`BkJQ7 zv{})NH6sxZPkt^Q(Ux}@XtOg);}5Ywkk6%HFcfAgifM8?c{$^;IzJR5A3>&dWnrL| zFZ;Ub>{LAtNU$9YrtNF^KHU)y)2(0<{`#i#X3zzmnlUG~1oye3Fi_atrrJ^_ch(-rj5SNRa*l5Y&P#_5dSz=sh+X1^ zxg#&PP1dzzr}4PZ=1csxSmgFNS45#hKbJzI_$gz)Rr%tfKMZn8s-wV3q4K+k%NImZ z2ayY@!aq$}70^7Y%ODlTyfI)olhcaAuC%|!DeG6t_NKdSLhFAmGrR;PN1a-RMiYsT z1n5kx0Pc|C$IFnrGxS)bGxUxHb+J_y<501=)rL{HRpJiz76r^65KKr(`B9yqNkEr# zuPKT2w%Z)BQCfp1EG^BNeTRgOVj;`~q^ztqEFUfIrvr}$BqjzmsZJ03r3QPw6sgv| zq)Z1+4xrG8$g!uO#Gw>2n^U>}z6ZGgGO^{93 zJ;Hk`51n*vRiYt6vcF9b8%MA>pXZqgk~#dwTZ2%FG0lM^*vhfF3Aq1~8MhXU#Q3*koEZ=gb z8-W9vM{47a<_&)GhplR1>U5k&;P9W#@ea@Uu!#sB=?RY67v7Dl+1Du-g}4aCF7!V< zNWC7z{253LeJp-+*to||g+e?IJ{DYA&!GRQnhgyp(1|G^o) z*`IXd_SH=xo(u^j7CFr3t_(OvgUd$3xvYFvVbP3rtyA4LIXX>6&V4w_SRp-btu##*t>A? znjYYhDW^c@K4sF8>*zo=&&|3-wQ7z%owCrlQM#)8wjT$rWI0N!>IjGEDq2E4%S+7p zqnw3J!3kLvC5NhEYNVO*2&=w%K0 zJDP<9?){R^*#DHy&U>JHGXI?`p@n%Asz2b?`FqM@AQ|j2^=Dn;L+JjV&ZDkCLJrw06S|n5Fh*YbI*F~c-UsZLgAj~~d1s~V zafW_FoV}@V!nrJdk0OO|X_n49Cj1;dk5;;2S@X87cDS&baot`y9>+zPuPWhwe_i>3 z+R0B@+3i9+0ePN=JO&R)vbgfJob5yP#z-kSDN0^{-cqmLY!cd}rm_J7k&|DEG{<*Z zwqEyTK3dtNODX{~`~AW4z0anS?#^w?(k9G+uU7u^^23*Iqc2f1ci0XD#LYlS5KWbA z|L>?lU9u~b(%xIQiz>$B9=dd6R{-QOS6pp<_ctnKT#LoG_mEg?Re}pnwnlbd@1$J= z3|o7a8D5)gws`R$>ff-+;pdnu5CuDh1UktS_hcoP;1iFz?G0}b6)DI60VKCi4izUw%H634#D zLMGAEI_s_JyHi>lX*C;5*-Fe$;roa2T7H=>;sxdAFhreOB(CD!+tjlCWq_FU*~PkmA_hek2e zi#!!h0;))AEMieR63FOht4dR;YNTW~Y4|=C^?eM3;v^^=eLx%Qm7AazP`8XcA4fX7 zo7jKyF1icVzwTzg`Vl_*_kM^am0$Y1{`P%e5>UWanF(Xtn*z3RY?T@;L}p5wiPia1 zF14qKc`LFq(jruI6lGcvhkD!{Mf8v8NjK{FUOXi5A4YG~I%`yk+%tdYSN(VrVA0{@ zb?AY3T}Rq8@HzcY$jebS#))peD#w`u;G@bznC4m~{~+1YiKTO6l0hE890nTaca6|H zH|$=-{%jz{=rGlcl!721Cnvwo#A>tZO1fw4CR?MV1ij>aA1fq<=@I^>r}F-CWBVWH z90rm^q}Zwjm*K(URJX+tV5?(6g4M(5Q=@6e47g zp97K_H=6phU3E&3XXde^_7t5WM0-*QGr&P<{EXNrf6O(FsLMSxSVhO#;n)@0Hytf zvq8m3P5*|)d#oN23)iAbi*4YU9_#Ou>AlsbQ>`Sw4&)jrNay^)RLM{2`Nf?928Zbt$Y^ECMB{f@bV%9k4_wM(knL*IkTDV9 ziC{szhlHMby1sP0;Z(HhBeV<7&_s(QRZ#Gj!y7f-g0;-)+tqlVx>2Q_xCXrzdZ8b} z2|uQ88Ma?vAo;bE)cqOqlJ{08E96}B5*EPsm-V=Jo`PXdB~-7hCC~}em2~$< zondEG+i8&fmxn>@Z$45DWl^%-L5X42aP-`VH2u`J8aTL)^UnS)2jvVQu3%a|1CBQU z5lg74x|h?NhiplQSD5F=+Dgz;iH3(;YnoY<&7wE;3)M$s_{)xesJxrk&g8ddlEIGw zra{o#ciH%jT{`kWLUSY-yux&&OG#gL;M1zv7#Ibs|FhjnoNp)zi}1&<LFUy7h zfNv`XYV>cT1zNaNd0bfh$N;9bH=ag(oQw5)Ks+7jd9VH|obuzx35Uybift1pTfCYT z3z+@NW@p=8W{qX*HXek3o|0rl{PY0-l!We@vvlwv(vD=$M?fdtOMEXD+WU)S^oacz?>ll}Lqq zhONLhD)|w_VkXmQx^(r*U*p=Pl^S^5t39}~k%-S?zOsz%k5Ikj+b zYv(#9a9tC&6|zJ2nXCSNKa zOK`~kv48*b*mKueYVSp46|NU&e)g6Ite?<0^IEIylV=KCmZ=$W}2mS-70;C3& zzb>{C-^e)nnGN6U`-QQLCQ2mdn{fz;f|QZcZvq*x4tWG z%I-oSAJuW{ERj9JT{mKlVokpz80eJ?Yrpb3$+mwx96CA&YRBKcTg@IBI3 zo#2!>^I@+wI=vMNq5xY^ z#ycW9$S#aM*7urNu;W$|&aUB^Lqu!_WZzDMd5A7d?Bx_y(ibVL;i(}Zw>PC~ka+Ww z7v`DIIy3jcwM_22gGCmi{-33cJ*^Xpse*(c!MTp=^M!Dy6Xj8pKqd%HBrrqKUdKqn~UAr96j)&unSJf7?>hyg^J?*zY-0^GcBn44j3qJX_^c zWscr~dD{2IkZT1TMd%l4rm!gW7ETOvc4hHHd@k!ubIq}=Zx#&|(9j)TSNE{f z$|O0VTZ!5!?8RL!;OU7LbL@>{xGZ4FXlOh16j*FrWa{JE8EN+-&R*T5N7t%qDMjuHj^eTgBYtHtoSk#hmGMMgcf5za0 zfy7|@k6T$i*Ch?D-i1yF-GWneVofvey5>gkgvF}Fc0UAX+%t2aU_qRZ`CdqVb z3Rf`dt~$(Khm&S1nO2iych@2Y$ZE{&q(lrGlGCc$)v#(uOqhpA28G&*sd)PqgyOH4 zTKo5SNZR{zr1WJ1)nKFM_X&|q2{98x>u=>7-<1@pFf;Ey@~~uYq{NWM9AGmD5&L+@ z+7(}&&e4@LtcOI$%N@ekvN5ClmduS<$+| z*N@zrp+QmNXnzdOamP9Mb)&{{mHYnx=$^EbQ^g- z!rpW$suKSe)wA)+I4H%bafzR=4yR<{iTSkwh=Iel&d;4oEu0;4C@8(f(Vs+3r7B+L zZr}Srlws(HBj5~M(ONTUnKbksPL zg(L9jUdEKZH{hG1F>i+0-Bjcq@8kGcaz^dg}CUO?y^uZGE7la3{}2ukf}_ZhoS| z%esDDZ@2M9ffrm9ZQ?q6tX^SnE+ zIz3iF0R5js-p{5-A%oAdPm%8bIOT+hg_)M=g24ad+)mD;(6a|#J8wDX@zzif9DVxQ9slms zred(Ii#UB?Ff)^LD4(c(ZwQsTx#2$yAK*dSN?ETcmZ`FL41sJg(Fow zl*Y6Y3JY<2Um(QPd@YvCR5y9}0Ns|-s!33_7MI6^^IIF{zbW0NtPbHm{T+&JSf1h+o4e(#joYX}se# zz#T(dG^x+_WCD*IBqC`S^~&2zwZEf81>s2i;sC_w6x2Tjj&4|zBsds1x`zO1(2S_r zSh2k?Ve5u@^UZgFlebs_8w!EZm-oo#_2I!C_{1>w>`xziYI)zuKFmyYGl<55xn#(s zsxS$nLV}}NQX-q-K@grO!7=w|%AO&QR9j{5yvu;tfi!7d8rp{O1i*sF)9cq>88O&- zvHFShVMcbJ2a~Ig*?$C7-pd-b5{vigP?u%+`C{wSx1_T+HO{@J0#eapI>-g|>t@Gc zm1t8h9ntIjsCdKeHds6UrJ60O1^10rq@=Ro{YLw zTxzGp$y2}dfbu_Q$4UGv67x&@6;3ZgfEETvDuz)oup{{N;62gAKW%2D-MKm*kcdcj zr`HrJ$J5Kuw>s$-Ak%=K=t9f37e!KGChRVLLp-merOASo%czT0GsNUQSu^z)P$|GK zVV@?v=G!YwDuvO^c&G8|-NI7~%r~TyEn92VmLb8X6ejlbCkd345svZNWNytEE1zmN z6l9NO%DXt%)j`eM!YffT*d9MC88h)N=US`p=9m_1pXK|M1P-|zGs-erO#i1bdD4rz zdA%ItL543S-8=h#5|G{Q?b!9=eM*XOgm-r-L>NEY`p|*&$8mi1 zrEpus$bfX*2>*x|L>J<5{7+d)o7`heg`UJx)WdfXmuPI--+o8*41T|Bl!|;k8^Q+1 z?UYdJEkjHxgCMgsXlmT(vPlRzvDX7?8~_-}`DFq$ZK%*}WUVZ%2Q}2Vn9-MD$R+%a zN{{Kwimn1rnUhT{)2Zo_`+uMqU6Qr)On97pyLDWX?cY9rjTjB1M)xQY0YSw;1`-1V zsZr7$qoq|EDP=TBHw+|4N-8OeD4+vrX_S!0!G7v}f8u_A&-2~>+iSbd>;1Zp_i-M_ z@jj=jrXqXY>LSPlIN||*yjgp#TI5K%oFM3c#WO916hyxhM<`AkY8`4WQ8g77gIg z0RGQPFf4$;0w^qi#sXL@fWrd#Kg+{#00IY~Z~z(yU~vEr2jKtQ6ATX^@Bj)Apz#0} z58&_s{`ZX{VE_^VAW;Ak4Ir@q5(gmh01|*9VK5{DhD5=TXc!U;L*igaJPZjSkT3)i zfk2`VNHhY8MIdnqBp!hTP)HaGi9jJyC?pz%#G;Tm6cUd@0%#-*jYOc4C^Qm{Mq<%O z92$v7BLOTDhD9QH0M10XOk1O|b?pb!`|0)s_ha0m<@fdNn$7z%?xVNfUx8im24FgO$jkHP?G3=EAy zpfM;k293sG(HI;WgGXZkECzSV+jC`0K*Xw zI06bsK;sBl907+T;Bf>1Pk`YG2s{CWC!p~JES`YF6YzKfAAw3sSzqstg0}MIOOnK( zUx4B}At50Kcz}z)o8xzz0OSNf!4d$e@J5Fd({sy|YcIg5*&sj$qOPznj#b)uda$l& zAc6NzET>w1u^KN69q0A#`jX*vNt@d;vJFD-vaU89|2TSY{yy*4Bg(Vu*fsfNMDhi* zH`t1SVC^FH9Cf7gXfW8?dFGTM+%VesveFR57iC;;{o%Qwm?ZIgjK^0MANA(ynU}tV zGfZ2$^IS!+@eRq3A!-Y+qZkFuvoi#vAgm7$zMKrtjSQb(u<5)t`2(JFyCCXZ^v1^8 zn71&vd$_G}^E2b}o2cRX?iWGL?u1v5+?(BIJA=O3PhQN8(0{H0}kq z-_TE^-cqo0JIj+#B`!E76k>87P7mnx`=!fubo@MGRjVm`6fz`UerqpiuBMCdc#&76 z?DIonsH&yuj$pav17dCjvj$Dd>b(53Q1tBmp9?I^QLO5++9Wq!Wfyd)(j2YL>Yo~L zX&h<}GI~O+aJ-g-eS zJ|%h<4JB-5iI3zf!gZU3Tc0Q=vSqh?%PhCMnKx<*bsNVN$*g*`!O8}t1reV-2bL-~ zbtsX=W*&ASpiZL7?_oa{uBm92ai=Jd@7G=Dk0hJl;H~bTps->Ls#wx(`toVh92puG z5_vz!&aw(_G#y=C$Y_vZ{``ls33JKR`q8r-vS}}lw!ICf?LKPV<$K@+yHaNf&Sqtc zIF%8W%6OtOPNbLCtT!FwMK;R_zaRAvNM2}LRY{|cDcO0!?D(;`_pFwlum)KnH4Fby zU$PQ;8r1`@w{+#kQplSQ?@XEr``#Bw__kb}TLW$!+m8(lvDqMxdQtFpV=isb#|kIc zX1(BzMkyn=1A_IpsV3UH%|r2FK%qY6y_*|vpr=W)ZQxtodP^6PI#;8Z)E*qI2tC~x zwn;Ae9H(GOH-f3OFQfAR*Z@9HZ_}NrSMq4S0+09z^QsO~^E@kz1gE)j8bDLI#M9sC z$^}(pEzgT)EtMj^x;5^VR{lxL`?RxDo&$52<9hL+ z>Yh$syzHyD0=H8mPH9HY2^`m-J!ZOgd7Nuds?wWh%P8s=`%omm2q>QfG+Om_@O>!j zLr9TsWK)mvGS9u6>1jd;MN0l}Aues4p(6%dsRi#91Yrf_^vbo=;7b9K#2inQ_RDF% zuS(rme?~^wENvj^-4WGS&~)NvZe*D#(@cmf93O2&hrxPz4NM1_o69A97B0He-QqXr zcci~;&>QdXnZs`Uxu~htR`99|1*eP3z;)<|kbiFs_iMe14eDpY>{5%$B%=MUF_0%x zaca}%^@xmoJO6}u7))4HjiMI(H1h3|62!6}?Eg_x(-AJBD*8q@=|@V5EfEo3Dsu!8 zC7o3lFZQ$Njp6fAnHJN+JugunO7lLEayE5_1h2u?9o-A$8bVTWt#|29Bf{QBtQ<1* z-x1ig1+YP?ZV!qzxfV*ZzThKevXXCK(X9<#QW~ZCJk+)l<bMlp)iVG4fz*Ia|G+ zZ*wJ+@ILUm`x7hDWT7FObf!$N?nZv07cv*lT%bP5_$hB>ywncq zXu{TJ@Fe$2xjVDVeX$##lS&55oS#-2mrZizyh$zfx2rVYGnsfetB0|;VsAjNAy5N_ zy66j*>#~VWmR?G$fb?CoWxPL9c^evT^T^Un`o>f?_|pqVt}08tVUw1o!t&J|P-LUle=9eOG4$pwW{+JKgNN2)L zKElS;_ER~}AhoJd(xQw$(ayCHjT>K+p<}%)%yrVhira&RX5_6n=) z$fpMBvsmDA%OWo!yAaE7-B@2GKH7;W$(pCWInhJie>utf)`~!vQQvlC-SZEe7OXd} z6W*b6-Lq{_E6jQR(XW?1ReVsZ-1qhdjwZYj?iU@2cyw(p)NxhN&a_&z1`z_KgK6&QrFl@g+Cgze|;oGc#t- zle+u&G!KYxc_sgx7Ct^?`}iFUa{Mtf1|s4Ok#e^j{W5zz(%g9>K{Vsny3rouCA$21l zTHm~@{gGDraCy-;{P@BR&l$P#*}b%D3_#sp&hSl&TXd)I!#ll$2wzBoTlyr@svnd} zoUv#=IUOrH{~qpe|F(MX4SE;1Yv%THoA{q(l3^h7!VInq2f`4}W+b~`W}}Qlk(sztJXiQZ@M18 z>}j|eT<6EYP8Z+f%~6_9hM#z}cHi8fb2H%TVS2y0FE5Yk&EVjN^R1MAp<4{$w{JZ9 zR9cm8fwx*uXSpdE24N?z+})NCdHpRxAkbxNfE%9BDJfnW6t9^q+*M0DZ2ZhG(A)fn z08uajc}x=zCwVRWqpEXc@duf7@ylsJ`JU;L!hk*=vJ(Ryojo4X=zyTcxb1hqq(?+0ytm^WP=gPGd%E3RRe2i+?r^h^AFX)qLCFtyl)?TXGM7CT47?=f ziD+dX1~oeohXoH`J^?8aDHua92YkENA zAE|moRHGQdXdNLT!VxCI%uNQ|;-feKQzvQDkv7IYBWGDWd_c?)lhG2ex)`RY&+H(} zZZp70jH7iz_>LHc^pjukqz`u+4M~!Xid%hTYZRD)2yOTPXA-$RvUckTLav4;f0TZ6 zW>C0ejPeo?(>dT(%Ih4^9G1MwIeHwE>&Q(&lfZ!o+%0^4(uzmX3FHkAf@>4oB8K(G1-%^ee67FMSBXEwZ+bi5YmLGQa?;_xj0 z8b>rPpkCNYhc8<2)@Yu-WS+NHDQ5gU=TbbJ0leHX z5g{T52CahckX)dX_#CoWouMRMg=dEivIz`n)=tSN@1z?6@_8)2&m6)w4BQr)(&s*< zOVqjrx5hOYT6wTt%;@Kmed6AOb3bazK%Znduz3K6ZViacbf?T^UylzDGUQ)fc@!PP zckMYJ3*?V-CnW(uz(pcJ5?}@~{8H}A05LfyqBg%LoSg9j?AX2F^%IDIdiKASyJIci zzm@w?823Gl=}>)%M%*F1^r2cq>ANhxV6JgvPoa@KcsQjvt9Q}+g5S!0DzRd$Oj{UA zBYCj+q1>p!emQ0r^Qp%2^`-p_O;rZ`yMstgON^onkJ6yh4ti44!uU8mM@~3j*-<1QvZ9o{r@yD|ellb%f zUAgIC?LR9A4-M|lcSZ2r{HyY05sj#Q$I-lQ3(|Ws5fwXM-=xTTZ%hw0@IrzF%qZn# zv5r8x+#k=f{vrAj1$VL9ZhTJ`Z^Mn5H;t8s;X(}jG#)jU#t}RNl%_*XmiMt1U*F`! z>28dN+iinrcDA{ZE!4qQc52PF3$N*4UzpA!(dHkxv~T4UfnA^pMO zVQ6=4+74mR%H_(?HvTmU)WV4H4>>l-Qef^p;IU_Qb6kGj|J~3!f&hHXP~>GN`5G+Q zG_aD}4_f~C6b_@0PG*VJw+(qHbed}}!iHHZ>VHI%5&_ns%+5@ws1Pd79!V`!3WO># z8`StXI~VFw`fXsMdzPGmyn5CPn~4QPD!3PoBN#C)-;Pk-s+V95T{!SJSSW#uzYzlg zDf|Ho^MBgpA7D|ABrr|C{BMK%3oLqn+T^#veUtsCP5uH)+50@Wj~lf*wp=HSkaPiw z-Yp+1Q%$|lc&%1+ywb>3edf7L<>wj+X|EZ|rYhs)u6{4M;U;m_>hpev{ncM>5^wvp zn)!IMdamnk^T&yw?{kse(ceoopEqqwyn4c@ar@=xxy9;a=`7?+!-mx==WUdOJ_J0k5+q&XL%y2knr(*%U;qjokXx! zN83RS0GQ4_*6~F~Hn+qP+{rl?_IKCDDtU|_cLXKs&vm)i@OJ(PUtQjutC^D`5Ds^R zMxnZpmPj}e?SVyLr)4Mz%nK*ygLaa))f3I_u0ut{*VJ3a538Rsg5bDDgy( zoCuJD1yooCBDDRF`QpBuFhs_m-AV&N>8;VE7P4-u2~oWJiy9=I`(d#72ckwWpBpL} z3Q_0BM5N^O)^&?1uO~m#B|(lChb6RmW0c8>Cdo>@nau-Az~}p6ec8~?6HYTHdp6}{ z0sH*E^gsr~f@dP98IfS)W5cyvO54cPNYV$;cS#?^*X`#zqz$<-K+dV z(ch`aFK#VT?_jIn=zpTu;A}@}X=xa0K^8Hi-t=wEplKANIqdu5OB3qdn;ebiYO@jq zc!?&q(ef9!y3BGMAI)2ao)d|gQ#Xjb=6xdZ-12T~y>cZ3f%IQ2N%Fe}5WoU5_yc^3 z2wwPKz-QPwaxb_OMsp?ygCrKd`k%>;B{)on)?T!wzW84?*sH~m{A3rPkAW(=hi z$L;U~S$}HK=Zh+32^LJDY{X1d8p+dM^iNppuQq-i%cb#Yc`QkJ7-Ke^CXS1MUnw1D zZGJKH(qFl>!zV$)BAsx4Mw$J|diHfR*aANIpH>s%D>|3^VrqJ!BH_Y&)i#K zPE2>M+0t^b3ES4GaGmd2`$BpVM|f$Je#7P4+L%jwU#~B*5J_V4?XbiPE9+?#^8;0# z!Gz{*8INw_ZM+YM6CFf`Talr}=e;~NsnCb;lM@F%Uw!)SSwC>^fY~E3eH4%KN2DVz zQp9bA(UAG!QHavC+zwQcro?A9Ky*bmTYs;wc^X9XwuRnBcEb_o@Ri1&LxGIDyIcG{ z7dFcfvfRw>PD0_glxA#)7?sf2Inb~;YG2h?V6%5L*D?267cv&RfC#Z=x~M?P$lNZl zgob7wt>oE;!e`ULR+BnbAwo9{NEi>mpA~|*Ww;CBKFFfNj3yxwL^MMnIxRCQR=J*HtVk6vv z_We&T+Z|*>aoe5v!_Dn3fb61Vd*jBHowif%s-0KV*S_t%hNxcL?PV%T_vmAFFinyp zcyMM6a547p4)Uj0?GCXCD*zCU#6YPxu(LP3wUT)ks#~tCR_%@4I{LQv9!@5{KdQvy zy8i(o{9^wj3)p(J>TZ{l%D8@>smFwo!;6DYrhYkz`Vv8!2LERJUyVx(@B(b0Ykxpb z_P?_I?p(J86PbyAwA6pt{!TEl_Va(Sedn~uUyUo&?x$VyM-(+qN(pE_C5fZ`)woWY zDYtXfbJTqePIP>R=F`Qc6s;G#%n?OM^q=IXt8%gzpZxvsijR zeCLHcxbiS+k8^a z>c72XUofm0e5e2?5T4@N0`@u!c?XiF4K2F94h_iNv$(4Ng|GHlFBgP259?rFN_)8H zYA_cX|6N!cY%IyDU;rt;%BuP?^~ro>tilxlYPJ9Mz1>+zdS3)ZqYzJ&$fY#%pmW=9 zbEs&aYHq0LCq$}^s1rS@CC2Is+_=-#xA3V_m~yhEl(mqdhX!=LSRYGM%opQI;h2B_ zp1pC7p`r8ovbHQG6R)BTcZ8si%;Soc=cuM3RtL9!(ba6{G_}+=oBdl=4FO znSeXK4A8X#>C6BK1kS%9qOH%)v|f~A;JjXvVUJla&Gy|`FUya*uu)!|=Dbl+UW(bk z_<eAaHdLyxpPmb6H~nJ!>h{B5Y~Ov^mfo<3i2WZ#&-;t$um1}#B0D@ipUzA8kWi7eTYGo(Q7$k9#~`f=uXpcS-lPim*=T&O6H6``3gPGNU-T27}A zP(vG2UpHdPa8!hqI;dPq8I>SdX0BwDyqGM_^6bAZnDyXL6rH2DB3hnpa(cZi-B#iH zTD}AOhH$>Pc-va+GXwRC;-nwOQ6&~cvFpWTpf{q=LT05aD#CP4%Gru&*Vl_Po9;ee zsf@QbDXDr~$}U!I890P4)L*F>dqGt9rV@SOoU$yJHYTttq381H#p<_OF6*^Ve>i>5 zj&R%j-kheJDc%zGSoKndJN1M9R`vh(Dx&}&pbBvR0+{7rF3l1M2LCT!g&kY=2Vfjt z4~`{(H1sbefr|F{8(=Y>FwN(@lGA_7&u~hvU-I+6l!SUB37i4)_o76<7li-{fchU) zlm43p8SrI)7G+WSUr;^h@lVaG*e)z>JrvOeIc@N75YUXgNdbN#D!KYPSH$Upoa_Tf zg<^@P(3Dem5&cAYQd{O59%Y7wL|(CHg3v`|_{}0C>^1FmiTh1cI48T|3gG2sLQ8<> zN*_g+#O!oypZ@%01G`_Frw!50TvzpDyS^_!(&bn#nLBP(oShbQ8J zjSG~H_lJ}iJ)2r|qC#v1MYQ!&&-N4itPX^jbu`48)1H`lmNP4GBz^uA&6SH_v=eG# zrH{O1-dm_4Im>JlzfA%E{H$qb%|`t8VMY@%DnuhZ$Pxv@E9Dku6mJQwe$5YClq zZ7s`=c8Xz0w^&gU$`hh}O2JpOno`Hj^e@``0}!5H0J-^J0QrMTnZL6GDE^?T>+Lb}6V3x-d5 zC2uKbHLg#`t8Jd1Z8D1^UOH2ZF{2DTm3pABZ@9f}x%hraTwF^~UFWiWuEONuReQzDm`SKWH+TxwvI6&!m z34ZGS)_{?4df@PLcoj^4-c@hcuNkl-NXWSExv2b;WlL1Zzzp4zdF}h&%U~Vn$3I$s zHXeZ=?z949!4a(mb72x0yF~s(XNxK!Zsa>p!nR3|K8RCpHFG`rp1G5+*sYnC6E3W8 zv>@TyE}2gO=^S71JyR1Yj3!?;(0w?ovt%C0E;Me==C&^ke`5cZWl7>i%RH~&a%N$% z_m{H;+i;2LA5@;#`(@D&A&+g%6vs)^XUr%yPuLVufofCfZa0)iGr*}%5Xt$9bu2@##^EZDp$c6T6qq)k7{MQQq%QZ{$ zzkbe)1h-`B2rK3?hb|VcGZv~vE0PMC8S5|T#%Ly7I2hbn*A7P#oL)SahU-P@25(Kb zO?-ulIy95Fv+4*>TG$S2?MIRlgCTy~*1LU8eDA9~hYa2(@kz|SHX(|5#M&Y?4)A^Z zbS0iIysMlHbnO|(hdd_g#NDV)D#1GIFpGinaE^?|iSW+C=&hq+3-CpMDO>uC=lLt| zOa!MozYH6aSaeXvZ4I-w%tzPl+Vg&U&%y-pue0pOn0Q#u(fkxmdNM!!SwiGZT?C}% znAf!AzM1@!LBGH#YIgj=tIKj-?{a3&4xcNqD4-Prba+{e!6y4FOdf!uA(v0%L3*Fc z7v_D#kXaun+~>>@5(xzx3~)NrwDuvjgaaoQgE^{YKg6(1$U@W%ea6irZqdqGTF5^3 zw|S=YiODvR_PfZ35X9B>rDR#q{AkF{&ge12+kZcElHY@S-d{cy^y>|flK}rQe}8lh zKfw1-9SkDS{{OcfgZ69aqV5RlGpB#29uLU9tcESf?pOb9U?qf@oX`$`bTHScB;VW6 z%XSREw;jD?ZhBw5YU;*7_+hT|?})4?xC;~Yg0j9ux5jY{W2Tmd>1w)3c53>|z|O>b ze3_jYZkFf`LW1;|@A_Klcl|Pdn(9}oy6Y^5FXY~xtelMq5ObMrH)i2j<0DrhE#dC>j`owiwG+yF?;r0fZhimie)ST%6aUliUUv8UuC8Fb zeM=8Iguv`LAG$l#J0|ZfcgP$DF|mCG*hL6`j3l-!w~F9Q+7DCgI%c$vnsQpTd?KPQ zW5q#gCR=!q^^+mwmk#zHeIjhF%&o|O)j#zqoK}WUfios0RZa~^Lh!4xwrWQzs9Xo0 zkwiM)Tms2ETxM_yGSl4S}`j!p;%2du)4z{yGm=Rgwmv(Tqq`Xom>K1 zy#g?0Z>EeK3HEJg^-5jQ$PLsuM{b^452W7 zhxX;E0r%o1&!z1y+uZx3fh3=@^NcZn7j^zIg zZp~rk|54jZcc_UPIOBEEB}0F)%AQSE*trKK9{HCTvR)GTOALjDXzy6`1gZT#Y|_a8 z0rN|3&tQ1N!m^3Jb3J&|ST)%o^(gdgd3#R2>DEG+@^4f|o%=sN``ol`nBzxGH5=$t zM${b5D8T);#nYfZcxLtvif-3-@R_vB?KcK>4sWyW#8poo)-S$^`-Mu^`n|=VE~bWD z&0lIK)KxI~iW!^V_qlF?Hs6(xr@tUA6d>mIW*Br}MI$!zpyh8C-#j=JxYsBt) zlToyXo)Q)7PInQVNDzRrra3`5eq>+O);fy^g_iMIYH#bZ+VZ`6zQz22ioXBqXb60$ zT7A8(P%x^(x{gn&L?OTlL7jLv~Q#+g`3)oI5U5 z%6Hm&)OE10%2X*fOYW`-B&Stjh%V<=8NV63x*s>pu38W(-(lPCc?UhSw+m8cg-ccn z4jiY9hWAaGz0kc(QVeAxqiX0sv`++eQJh&P)V+XaeWPdU?muv(61&T#%;7+Q^i6e^ z)AUR+oP|}jXf-qNS>LRq=5e5i$3|GdsIzxFbvlqVJHX!?g1@!k1P zyI+0jbLA+H!a*P!N6m|CRp}*JOK3Z!==^pPA?P&<5_n?GXra zc|gd!mu3YE=1Z=)MSU;( zgN5M%Gd`CNikAB2VI7J7K9R;^wItXzma3OWv*Ke%2#uw<)<&0Hj1RXN%2UiSB3JWc zDc%Rw6T}oP12Ip&XdEh)&}BYyOZv6AZod1vIQ#zMbnfT#pUas+ZjFo<%!KRkTzn$2 zs@50Rk2psD)b3*TZKygw>df0V{OAf_7IvXpqWk&xVmndTO{giLTB@ zFj1x`gAYEG6+kXc_lpSJp^fCVc*tgr4wbL! zLGnjur#?td-J!Q+S%|ppWd=)QsQbfvX=sFF(}w$e2tZ$sPZw zJ|o!6<990cfQoVH`e+3FLarZ)vYd5=gK?+Qf}MQVaWV+HZ-wqtjAB;K51vFiXlL3f zl4QcSWqKBGmeJG}$0GZg7IO26gw`H*95aMUbXcz0X=Xh_l)CX>CXm32p&e?uAQjp1 zMNlhcszrJC;e_xbHq*5P3GEO#4>aJysk2X$@(nR}^GOoUxQic8U^R5}`wXe$V6m-x zd4}VhsKak4SXyss`=X>!wmufC-3^27WdoS8=wnHzCLMB>}ACX!WD6 zi`$r~bFPPd+x2wAMaRm(=DV*}DF4)<`EIOcv%P(63&$YoqY)QZ+lfwbe^oWLtJ+xZ z#~Ih&2`?-AnZN2#Ie-XI*(`FsZA+aAsY9Qarf5g#@0JhV6%3> z{Ju$m%B7#*4WFPGx7v>$D{gm_N#J+#+j`Kuy~6Hn48)Q(_`S>OvYF=5D0o89)L_X z5eC@_aSTv9halj)iCg@%w)rGa_~8@`JATl&y!TLLJs|KuCK~jBFi7pMiN^mi%Iga~ zzx2IVzkuylIDHx9kJ z#n+S{h@j*dHFIOcq>S+4=Wx6D=gMAMYiw5$eSFe3DrVZ-s4Y?|@qzl2OwX2{h0+Ex)fXm~R-sQU#0U?nZ!InCRF8eV8p|XxwM5rI$hsFC^vd^< za8Q@e_Uy}v<%(hsFZan+2Gcx25P|^?;~Nc}-TlP!h)ErRc772sK+1@OSdki!6eydA zKokr|p*je@NNA%nvLN&#NZ@PyFNqp2@1j{YD#7$|z~+%8gEpsEvy5j5h>X*-J4qEl zSgRopl1Rx}^(lzR*=!f6+=v;*zsk!KWhU8$kPzLKbSR~uhLFfQEc#kJR%7>jl(?)! z48B#nl!UmHXl}JF7eH@EVhMSb%Brnq2i7_MVUT^$!IqTiIEAp`r_v;MN;rDw^FH@( zm@_;o%u6QomjVD0v9ek!*rl>95pWQTtk{4gIkn;#q37w()|Q2)->(;4&lJb9? z!0Blp8ajtM(9aLELTmv=$|xZdI}(0)hITDn-I?AdBtNFtByB`QJN+?*n6P3(j}H~2 z+SO4KUDd4srQpTeL1v1bwjdYz`%xjvsW0k(rYCs*kgZZ{$qiuTb>+WoBRMRFcpJkhW(S^x_oYl4>RKC-ZiU@k`D~iGN5SC=m%cvK)5*VAhvNb9>^Ar#Oly~uPiqUHNNyAP(! z;O-DX{RR26fKkH65rn=~g-RzH4wZpR?)I+o9Od8pwlda{prj)Il445X@m{Hcyk%eRrv1t-U7scH*$+Zw9&lI?gIQp2tt{h7RcZd$yhevC8RrL#jYe{V1 zQJ$gdoP!^#%N0buc4cp)D8s;u*-$n|8F$-3;w!k`Fny_5`~95c!*Y7rSBc`PuNSHJ z5Qv-4ZJ$M^EQWNlCc%Y%BzBz|szE4*)Pt_`V`wDgctO5G7wHHMIphNhJ7 zQ`k&@>}xC56?^f39P+S{LoGK1C>%FpIB-OpcK#5H$Mdj_c@MLxyooCpU6Ckbc$+x= zF`Yg(g?AVc;&4e?EIC|YJVXJ)Ur&E^QAg-OL`!ucm+h0!)Ngp|(i^VTRa`=xyp~wG z&ThAnWWcFGE;D9KV178I&L80!4u7`;uFn|FEq3C)J`<+>SXa4#;(Cd~`Fs>hJ1ALL zHk4|bm&>J^tdOLx^bU^+mg(^jceWIdg62*bBX*zE-a|o*9Qn>0ypt{P-GMk50r|m8 znhf_@S2hcb3yo&dlHJs}++~j>_%VhI*2ZpJ+kdANq);o!5?TMhbdo*l5JydV`wFnR zWuCbDZGsbIs5F&PR9N=)t%EV$YfMPqTvyc1)`86VeqRZZap!MCrDIa%FOp;Cv`IPc@2Zisn{5`;uW0N;);U%v#1DwZZoj^Z zu!RY-n@=5yCcm_qrPO^%d)ZHwU`NR*{d%7CF)W_KJa*u=btiO7{1v58Y@`%*N>6t4 z3%Vh9g;`0ow1>agyIY=ny6uS}z<=^IIe-%w`~5GZf8=TZm`jrX&pFd&KBA~YaH-7S z!+3W5wLMD~IK3;%gLxySBJw>%W9-2!Llv^SD3O=`Z$uzuckPa}r9{ZpiUVopwAN zz!a*IA>`Y3UK!q{X7#WXyuP`*vgeRedEORSXef1{#XK^TG%c}EQ92jKNOQ#FO52k@ zMSKsx_};5hs(j%NE#-=Zt&>^gxES)=A)wcm*t`(yL$PFTyZ%UoT?t-i-%T$gXW2_k zCbPp42WMAhaqZJdgXpS;lr281px?DY!?mDLU%d(JsQBG!ulHi0cCUM+cj2O?Xg+Jd z85S|3U0NS4GibP&_-o#Ki7rLD%MGUZLy|# zj7xR<)`nOO_V+NJeTL~S{9k_N@2rF#r2Vg~MBx9*O8#%L!*w3_N9=GE9KZgz*uh+W z)00E}o&J~D5q%vj%?w9+yyqUz(Js0%^M-k?qvEg-P?u@q)-5r4i80gIGq1PoUhJn~ z7|hje^51KtP9bNVZt3Y4`l!+3p;qn*IyuWau3i{GOHb+Xvv@4m9r3%&a_;i*_Nh@` zqD*1UkT1^GtGwh`<1See{!SbdAK3VPz~oF641SI=?ITy1=I(^9k9_JzgEX#6L)Ke{ zv_A`J&Dc#nc|~{Lpn3x<2Y(+bNbof6xOKmzd*cGYzjHl5Y^>q<@;yS?r-~#ky^eGd zwZcszJHxpU*@?7F-9rR@bg+W#)iJFxdJ#S?J<*+E1r~GJg(z7znJCFs1?}D{H6RG%hyDbz1(}};;)o(0l-ZnnOARdSm?z@A4A}L` zg9A+*l7Fo7Ls(buH0bwZv3ZeNEGE#w)9I-kDgxIGH7>(XZ8dx!Fp3!a&+{hF`n~ zHlmJGNMuY`7wbA#xv1Z@thirS`G)X{LYQl%d9dPC!M3kCjk`jv`_mpV>) z2<$rb4NK;8bNuGjEseSg!=IaW@>lQ7$4PX-t)!uRN=Ue*UcK+ zo&@Gb z@8zAptylaV8lP>i)&$!kKn(-}Q2d^X$WdtgRze8*-FzaJ!S1<<>d;CyUJ^h3D2gdW z7_>bRBt7^$D+%Xq*DO>-WdHZ}UX<`$N-z!AyHfQfYU*DRe zE1AvOFp=H)8rh?(JaVf5ITw|s!4l(rI=@!-*<@~{>_QTP4h zj-Q7+b6GqIF!4K7AN){T-2fkha^MU#-n@U z`D}T^ppDD6=9U6@ziy>o{YS9&dj&9*B);k5$&-AA(4?P9 zp%=m`FX=CI4u6GzJ9Tt#BKj`8o&S^jinz+Z=X~VA#os5%89~N>fJgr~@Q4U}+U1f9 ziX=yn|JN?V#G1I=od&V^*PMv#>ctlCzve`m>)!u4o}l;qTx2Uk4ZSh;%ThDonVEZP zA}m;-9mh2;#!rY}I-3n&EtjvhK1j(qZe}?yDC#>~l#>rXuTlaR88MWcJghuKtkT?dS8Or33qnQhBK5%LL-Eaq z*>_*&U0O4i+0$CaxO-s*p@PI$10V2?cNA$P$lbc?GfBq$@tlQ?l_BUHKSL0rK{f5~ zrCM=6!$7>Pj-(!KGG#=!dX61#Bxr%TI)dMr4*8m)z*dqvyic!0MaUmY_L~L}ltnTW zU%3iFed#odKZnOCk;XrMt0rP?`B`dz%82=T6ED#tQfU*0{-+`~2H(yYb?!i&WwqnU zaZdnV@}&VzH1eHev91Tr6INGdCt0PsF$3;u?xsj!A&Peu+idBUeXLkTK&TMLKN03>4R>^K+pV_!F zRbEn1ST3eIx^QFtVHeaTxv*a-PN&u>XQfI5yo>pU{bDCPAqTo!DX#ohQFTS(qnntU zf>rA2H-oRG_eZM&Tb z*I%tg#LBk6s3FhVkPU)%_C80~qB`{=%dT21C4j|$W=G0*R87~nmT`x6Gio+N`#Sk; z6BXn;Z3`s2Kt-5+xPYH4B;mX~o0RB`7osp~6fv6jW&M;;6v;&2zIktw-LXKSdf|sz zdV{VbJu}n!zCiy=-Qtdx#V;x)F;Mbj-2^ZKe~_UO7*w?`6UE(}UL18unQy zr>=v*Z&SXT7aP*Qx}lm)N!?#=k8QZ!+Jza$wJ#oA>e?{OQ`x|`Anp-dDv7Mz&q{=H z*E)`f75qO^d?Ao^d|(m`g+rF}C>cL3Hp@%aSr65S#X4UNb~`vc+78xp#^`!d^Jfw= zDU=1Lk{(h!3f+9SsH0 z%e$Ethve>u%_~r@DN-EtN{)ZfwQei!V*gPZflDl)Gv4i04BM}La>rKVxy&*9yW`j> zwRp74T~=j*gC#zFC2MBCvGy%NmAL!2?w!{n`T+)#SgVu~EfunU8QPUt``$c`J=^~4 zBjnF)tCHaMcB&kQo=Q$P)ZqX1mlcv<+Z^ogjQ)RjLBH50ocxmWbjIjhPY~x{Y||S| zrt{o%sBUKHNf&Lu?Ju^`jP24N(O=swMcjud@MBjS%5{oFk}v6Jdhw1G@tL_zQ#4{G zitg2XBAL>51PV=F$2OLyI8BDiQI!g|_*N-ZKq_9PtT$Dwm3on`h-WauLppp#z3kTN z#1(2OQYqiEx9s*eptIOzIJ+yqMtYo_clgEa$#rW`opZc9u3wM6@D6Ng%5HVn;@mRt zsx0gG?jZdP8D8Dxz0s)R@b}zXk}eas7GVy-d0`f78f9^cairf_VS6m z8NNzjnl)()C>tC|PK&Q)0Yi2c!a<=5hu@}1p335KXSzcLLB*sD*5rH+mI15ue$%0^ zc5OMpeC@(~kNA}6=TYNl%s}wD(`xrkFsOe%lp-y9HhQx}g~epYYKI%T8;Vwt57fB| zeVnROmjj}6Jc85HLhOw7M6NnohbbrdK|(>!d{zsrFQ;QjFVSJmG0uXYFRx9 z<-)Upsu%pNGsvEmWt-n%UebW>+d;DA`MOWI$zOXe(<7So(KF)B!B-8Yu6^R=Afidxv{WN*%Mf}Z9}4=S*h_Cj!bm#U9H8vI_QH<>>&Bzv zUG0jYZY28i6S-4SRW$<*PsK*%il4?Mb!*M|YFn{aQCZ2K&p$K~n{z6yZ(3*^+8a;;++cN z;R8oh*G#jM2gkDu^I=7g7My(3BR0mLd^tFnkTs}2SrJnnIa%$#J~g#W;Wgn>lV3K% zeJ9$k@^sUcSPK2E_OshhrJm@6SKr&eJV?7!FRY^t>0q%CvM%PR!T)%B&BA$lL{;K! zf85ah>;UIbb9OlGw|#ar|3u>acq!fe|6}YcwK8s|HCW>4$a+LZv*B)>2?y3DmoDZ0L z@dOM+aOI!4_8{3zne-t}hN)6bXYDUT^^C*i`n}<5d<|uDP{VH}dLP-C7aHxBdeW`c zS(aM$W#{{4XI4ZyB`?oNnAF(Tdbzv^B}VXf*9V~Q1V6}WZEp@GYv_L+ncLj{nkD4@ zOrCI~(=%V?6LEQqv)FXSo8b)k)`r6cJ~$kdGbqo|KoKf<<~{m_-5|UV4q1_I)-jCi z1gxNf8*ks<1c_PUeZ$LlMXlxU9?%5z+Pcqy=M)$ubOz#Km#h-jVMdXULFy2&88iJDo$ z_+~{D?*Uoc10*!QbZ;WdIYX8l>C|_l!)U=(>EUMWC&7`V9fnkaz$JKcCjg2?P8>aK zn96G;Z~ToA&5kbe2n6|0&!d&6)kay(@ZEceY;Vs(h#)n1q@FaQ8pg3ksm{Yh8XYIY z88$-}P!BFIfldgz*L8=8_NH(4I-00VyGFf^A=uIlbq+(H1I- zyJdD-fJE`0yy($jW{mBk|N3Z9q4MlTiCMp9FcBAcv^cRd6yCz&-Pf1shQaGYQ>^7p ziovK>N2%MMXad*Y@`n;Zvl*`OtNuhT-{&@?6%kXJb-Eec%S6Z6-H)`ruc=9`PF?!8 zT8#kPv0j_%zx4929Mz7y1aTLS4JK=d>|GJ}8rY0^j8#e9(9pnAl7ByztT2}_SdYgt z;eVCW_V^&OYyWq=y6I;3A4Wj`#H;jDQ3}5mB>y%73Tn^}B4KP`nyWSI_tq=L8I_>2 z*_fWhFWaaLbluvQM>MHiDNrQ2M@INl>p*|;H|T%(8|2>?NW`BUiR#sV<8K1-jdMi| zFdtCsG<@?u{o!v+!G22MfxxiS8Ih07zrWO6)(7k$3Zy@w7Ks7J;BxNOJ}Jw&jCtu* zvqnqdswIAZsUL+IZWB)fxmFlPM~U9B`r%}hnq5P>Y$MB%fw?R<(4#JEEaLqk66vet zBK@rUWR9)Ys#UeT&Umo8OjC_%AxYdcV+v4_N;sA~4m-?^JU*E6p;%v!@8{M`)w_V&hJX@ib+<=ebnA|_Y1WjsSbX#!Dr#SfF!01$dDasy-baeVFgkQ zUI{?EkCU#N55z?_2)j|7Z zRqQ26ChF|xJ5qJOod^;M8dZI`*59kqz}!qVBS>aewWI0{Hg%Kg)`#`8PxhNdmxIH` z)nLXy_M6>;(m(c_lL5v*_M5Bi(n?7`jLG8;71D=a%3Vru+;Hq7>k>Bk7?jLnZPygO~g-F%;L(f_X4!UJgj zn|o5kA0PXFIK^!UkEhrN^B5I){Vn%T8Hio$VS+P4inpxJ`^PD!9rgx$NdsTjW=RD| zBCnqhK8;+7_&sz>U<#-FG_s%Wo3#kKVwJiYAA9&rt;Iw^LP^ZNe7*g~klJ2E5p1g+ zA`OlOZtg;}FC5%G7u0^xKo0WqIoWMtrDS82E6=if}K@P21%?$Tl*H(fVNYSa}CR`h(q zmCex47m}pg*6JD=YW?lu*=rrjulw!gZ&rS_xBm(@LJVI?Tg*J1gcFOXD_qr&-S47; z8*!ci)y#V$&}gE&Z`}2+8lc``5wl6nZbS?18NtihAi2zDC~A&L#0rnho$Gp=EX95H zy|py|7CuA$E-Qr7BWdpaCIxs*+3$Rs2=jincfprv+QP)OBusMsJywi_2_-oIZo^kn zT`SGz2mKzpS_h!G%d%TH3FxDqIGAci%M!9EX2LK^6M+P?P7~?qXuh3pZ(5*fwMBn1 zYNTellLp%j#6SpZSt~BmC>bow&%`{HBqYzL?#WD#FdI@+*A`7KD#&V49}jbhL8MT4 zYfQLbQr56v@;amAf#IO6ZhBtdq+xrA(5U4?JFTSSiMJM@fEs@6n?|sy86mthuN|%F z)32Ls=qrVt(bqX_pr6}16#neUXe&0)L8Gp}Q9z{BBvpLC(mZSY3)yI{!pScR$OPh~YKh3+=#_VywJ8m;Qc9vU%;Gb?4My&~T^k^>&ctC{ zC?~m){c78EbaK%awf^N$$yXn_Xmu(ai0L^@sql981Dc?sa=4^7>t>>PMRQrQWTSh1 zln*Hr3|tJ$O~3g5dCy)p($GzbPtd(bX}V9i3i`y?ulq|aGcq{G?cY)-4Z`NAbWb8^ z8mPRxp4*uXz2`tI{!wG*PWH2$^zyaikGYT{VH475uYzet%d0EbV4aThT~_U*n}+%o zhRc&3>W*73ZE+pj?;DvN$|oPA9%@|Muk;`;aCs<>L(eGzm{ji(nN;pVtmlbREmv1*jR5YMsD|zM9*7FUydPDAX&8QF5Y2_@2T)-u!Q)tMkVvfEY&PUr3N# z1m>ya?y~x}2R%LkD5ImRy1I1I*ici)J<@T``O>0g2Z*hB+b#KfG0z3 zWm;UlAHfmx{N5_zN1-lH1#b_N`%X{_PoZpqDga49w!Z1usKGFu=_F`{i z@PU$*l=9c*Ci{NygTA2G4c1&`=CN1%^~q&-5}7A^zJ|WrOg|pET_njca17Rog`PFJL?4L@{*dd(cc23NY=_>#ZR0R@8DO6<4=GUBiA%yM z*^%gx6a$^{hi^Jxkmd& z|DS~*v0yraTLKcI#>$n{jvC)Cm>f1Ristye-pCb|YAe1dI!z=sBz_?Ig)8U29ou?$ z5VHBMqAkNbtvCFSQ7CD7_Gy8wlc6wtnETKf<_odY}xByTz$dh zb<<=HtgkQ~Q}+fY)@v9NB8(;?XhG}=TtE^{2|D!_*+F2%NN(~#Oqb+I{mv0;m=My$ zt2hmCG0%FqZH&cw1lz!Fb_Bx-$fC&1Q?(Ng$m*lV5)&AEH`82xWf_@ zKq6X`BH;;vxpB1)JLp=JC+-Ia+0BZOCLU|I)jNa14r@Z~4hAsj2x%EPXT|4VQXB zSn}AXu@DF25(d$5tCmR-h2n?#5|K)5`>yh)A_o;EN5acS zu4p*TgubO6AYO9jwo{h_TlGZ{lf7son}aU^6`OptV@(j=CSz**WtG-|vL|}w8?~%s zMv8mK(vC)?P3oNeA%2HeJUjEMDg;KRBaG*Ko`IgooLIX(o`)^+l~RprfajdfMc8cJFj%N<@1jGIFiI)}uv)3p z)A?YB0-qvliDnfa+H_f;rxb|CG4@ ze_m2(fF#4e2zAb@*In&Uc!@ug9OM+@c*NfgidrJo^+}$ufuJAKZNl$K4$e_z6SmPr zo?@;{&^2~;r=JGM45v!XaH!a@H#}X_`ALSVH!jyy_jbO~{$1j&WFyl=i_^ybk`5ND zo+k~c$&`1-ZrRV}nn8bYw9267!zsrjU@|fpL~Ya;)NQWa@&EXeb&){ZY5iN#H^Pe@ zd}GV8;x`9ew-NPgevpI?LFQJT*{KJchgwOGwW*~xFUHq5r7UBM(L~?gD)8T&9jN%N zNQ|hT*mlqMiajvv;=MASr8kR&FkI(Yv1qcDty*y@1 z&l(1)Xw`QWBi=AzIi+zmkXt}%RQs7fO?D8rOHo`1x%z-nC{>HbdKiy@^OQP%&aSa6 zqL~Ra5;P_SjbhzKn9u;77Hvd7qk!3jT3o9_V<0ps6dq{K`I~W4+yiz9Q6A^3L=w6{+8ln;_zOP|Pv7xVTcUH28 z#UpNc|FBu4i_obOhXa{-=S>f zSFJBS!kR_ggT2yV%*{L@?UR<$lHL85_2k*$%TJ}BQb%`NS;-H~MLSHgt-$^_SpFUP z{-YKCk9g+*x$(sEq;QR9Ga8(K@tcEtXIt$2Y;l>jz<$AkNT^xat{r`=)ee^s0ja8l zu$I=WJL6vUfZ4W>uvARrr=u-6z%0hEIi?|U3EBKM#@^#+OE#0(H?!4R7@9ZT$K#>~ zpNa|W;$c<4u+lLr75l?3eEUXj5~DV%c6kx;*{tbgn9R0qtsa8)gUAA`Y%bzF-;rS1 zXWnhwG|HVxo$a*~PAFAwEp)HzOuN#4y?Ob1@h}Ry^;Yy~LHa{&=~#=7K*hCla&$&s zOcDITrZ6;hc`fC63(K$X!wDa_51V!}2w~t(6-LEdH0JD2`tVlv$McDfmCuGY2wGwO zogVo2+xNc^%#^<2{~BkJ+#nks=lVhaYPz?2XyA=np6CJj_~)pJPFD)6Ohu zdH~5sUNsf_sp)>Zt$Sr>sOF~uXM*F#Q8H6$m=sJR+fXJCMm|PWWrUVLmPl)cyPpSW zDv`g7rnS{#r?1v;IAm0>R(K;{G@XfyLXcx};>zQ@(3Si`_I6{tKLw9gxYJGPI=hmI zoXuA=3ILbbmilJcdv`kuaV%S}jZTfvNc}UuXpwr$v$iS z!DOk%V}-41JUh0ttA9Wwxw$wTHPaV%dZpWB{~cY_H#75E!{f|$e;MA(^`LvA!csUz zh#)I*-uj#xe}>sMUUHrlg^zC37)*Vki4ZRyqX9ltDq0QEzl&KWdJ9|)7B=;D=|P-` zq6)GGA-SbGSGpstM@gdM&NFv5i6#;4v{$LXaK}Q%al4JiW-JM+g0oF|CHCT#bH}(e zV6vK5^+YBSu|@HT!@KZs%jIddQeJ0ECM$2WD-I;WV%RHMdx%uccxs&oZD)wzjfb1( zh3y72qAxk8f&u5$qPQ5u8dRCk;5=MC*EYieqNH*0F{!_bxlQ3tZcHNKc*2r^(+q1A zhUI=KXUs!5x;9vEdta)lr-V_t;b@=HtmdA9Nz?wBHIo8JEddAP6F0v~HIT#}S_f9r zf!2RD<5$rDIg}nYKGOppHJJd^pv|fz=g~f7)v~slqoNvY6SAIDTu}R)?3_WYT{ZU}8)D3}ys2aWa3Q@ZKp@!{B z%%LtOdpCo0cX(Rjw<>{JyrUFAxQI#dMgLxtHK(O0Tz3m&E#K z!LP_K;KX?qV|5;s^~ZLyDv3G1yPALHokm~kfy@90K={9+a%N91PhNb#kr1+$cb!(` zsk>NkR4Kpy435g1X2uEhif{r5vs|12Cyt<(lpFdy=LmaW8!=!3-@LZObV&KPGiA$a zuhwZfr}$A%?qD>GlUPQfe9C%xQARl z0qw7u)JNv#J~s`7t)2MV*l6wAf~Xsu1`PpnZ(3%h1fPE`k6syGBzx}oLLeC8#+*h( zM#h2D$ix;M3|aq-`?5<~;!Hz}ev_JKa8RQgB?dx(Im#zJq^tl$fD@q>=`)py6RAn4 z8bM%uBR#B535;)mEHJ{MqyIa;CxyBEqY?ZUbbyum-xOaj?)k2U4LtGxCPPH#O_HS}jv5BN! zwfVx0qdI?-V79kD-cgWVF?Q?{4uFLX>rF%;c>Jk6JK`baMyIPCsn?+SKFE#N*zYr< zcRGsT+0`%O!rtYh;~z2crvxTsUWIGshOhPJ)I>bd9-+MP79#JcFSdf)Y7jqxA<*Wz zWvC{ofnxGQ;9HRfY&ZOZ&v6)NHoMy7(_mf}v5cypov$7b&VLfv*#rV^|R-rZ$DkdCj&z zvRImt#^p}AqbJEy7t_9z3q#z=B)S$$)oLUOBgzW(o}A5j<}f&^9!;vMl@sr6Vv(P$ zA7oyT#_eobn03@gtt>^cX)+0auO~bfCFg%XCsHRc|H>>0ki#z^Za(=d*r~1QMB_{9i@%uVJ|A?y8{(%#vk&)R z8k{CoVHH2p=qC=Gp)XOr)+(K0a$P1wuUFxgI=-yS0N84)H#|+L?I-+j_kwYiS)$o( zN;xuOCUZ2;VF8Jhcq{{^`Qq#ur*4d$ezWCsS?{vss;yxZdZCWeX2b$T`)!c~whMi}mZ>z-iP;dfmvdS=BsjuAtHU zkNaVS$s*E&C1Mu+52+<-nP06KmE*T>(BI&m@|EPGUOQT6h~Ed42sZ#d3TvKGgJ!F+ zVzn_UgS>zR0{4o@hoLq1rx9iKGMnGs#=~1dxF}w5Wsc z_(~QOLAaeVCa)h`D>wu;1hU)L#tIGmOJsI!=%>iovbVA|EJ#^zMZ8R4V8LVoz96vX z)%v$Wt2<+S@^Bax`)&kyEMVZX+@wh&yR;MkuY4a*L>W8V7Dk&0LSaNdlfUs5!ELWX zP7*>SiezIOgu(jkzBV11B9EhDBhgQr46QHBn2r;{m+?uv_1B@sS`7X=tOST(&y9F|XOv3>I z+sP@Je!ED$QC8(Zv>@>Pi8yXg5@^PP2}4D;ZwMX##YEFF2XnLdSAF{kon%oDTl};< z6#T{*({2-3AQ`xiCnkV`f_*nmo0534XjEX7)}?5rQQZ|8uc;fgu}KP?1NCDvG4vD+ zgn@bJr-n(w_~wTADmPd|If$VSpPa>$uh z{lxv;szX}Pt9kIa9cXbPvq@}4_+BVF*maNhpm`~wu|q`Qln`d}QQky5zz_7{b5sYD}z{v^!lF6QK|*I#+aDS484co%#-g~$5e0HSZWNWOWu~-nE!e`55%P- z|65^zhe7#cum7L92c_)xf9rlQ?uo8syDANQWRUO){`lkKdC!tQa@aa96YbjlCJ2#C zuReW=c)_Yb<#j5$HTslVxb)bOgyqXTb|hO)Pu)*p;`>uGdX;=RyCA*chPNX+Z{r!- zJ=BU*@N9qVFC9Rs2U{JNvVduIj4QpN7%Pr18jF4Vqp)N%$}Ybx@w^2#VQA9qg<-{u zMO;{^?3Ei-D6|y+T3_Ftsq*^p)?+-+H;)D7)AE;Jxl12AkNUloBwUX-KGWV?IoB1N z@>ImiWn zE=*NHl@qub1EdsD=$9IgWE{Vwgz%8Yzlj2T*xV2jEO9f2@}F5e52u4IHi1$zKsIG$ zVZ-ub?!>R717wjVr#@&(%$dfh_FU#BC@B=@hMQ;;6nI;fG%NbOO&nZJJ?bVXj1{Hf z+)g(fw_J|Z6h+Jn0ApC`i0H?r6lVT74Wo@S=VPGG{YhV9VemnjAv4{TySOmRwWP2( zN+@;nbs~oI^>DG{hxy%Dy!(exe2;=$f!WNGiEa)ZOOGa^iEiI<_ zb};gCfLUU^br`+X-gnQ}HB?)5(jT;}YjNt`-fAK{S}Qh?h4HZ%>~dJx<`H+a9<|wx z^Ov+;TqGZN+$zt6Ge3Y)PP$<6Mobm`$3|{N zBBbvY*W4&~X!i$S5n#&=p2mES5Mr$y;_$}NFA~Ql5RUN^;y~s+9XY@D2Z8Z@6*&g=a!U!1#fRPGg!T>5bx7?tWj~$3RsdKeBeA!2lLTpZ zq?n`kn4mHK%GB;rqi>P)uAcW6B6X}C5DH;dWm@m779U#pKZl&;3nCIeXYTQ{_j17D1^ znp5~;clH)$<7$~?|L_>>4RIJ6YoulHgul@jaAsBLyt`bLJ%4vK^iHb7wTn1j5b}qf z0RClQg@<7PsQf1bt8u^2zccu^T<9B}KAmjg|1hxP)=94Bwa3xIkH`O|UJzO*=HvXW zULd3%Qq3QKx}Z@OHG6~cDz#Qigi{PE6^6axL4%y8GC6N1O7uKW8HeKTRrySx@-J%X ztdStrjg>3)`rii9e>GOERd7Y9e{pZBev@yzhUqQW%rVz!z-+zT=Ey$QlqMGO%(7;6 zbF4^()oFh+bAP7lAEKJ3cL&Q~S{sj7x&n|Zo8C29%z%$kRav0TuVbHOV?h;8>KtY# z;muZBCt4=QJ%8#yd^@}~zc@sd^(6a7&^|JaRLpO#SL5w=5fX!CuYYgokjXFYn(7N* z9%5htzMoi;sn~PPv;;|5tom8Mk(tGWF$yu9CdJ&Kb)RxzrgW%imr*p|L?Q5fRuVZvu?O+)#t20=Yl@{6xg6c-@nr zc5gf!0Cf5^&e~0-yqVkuGqx2uMZK*RrF643{~NxfIi^0rm(|Y`F%pzoROED&oGBXO zVKdv>or?im31)`w{)k7-d6X&L*kLH-p@y2oMuBI}yPCK4Ydy#=i8A}hrfqBUvCMP5 z=9r&PAGGo!(L^Ox<9Nky=rw$rPJ0Mgn@+$0pLmCS#NN~a;&|-2=QM2hS8vP&zF`)l zyIF15(bTreI7bk5rLOiccfPkSu^wXntV|B6`Pe6Laf^r|_D)>gqS5MZY!}gNabv)# z-0(WMg7GD?T>c;(d}#8Z=0|idu9`4bjjv0XJIIZT#xl*e$@6Ki{eW?S0+&>n2;|+N zZQmo|{F50&brnVuNFGv)()w+i9Ap)x81N7PN1+4-AuQ&yD*=l`OcSgwqu+xWw>n}N zCbto>`y4NUl!#e=qJ7C@GL;GJUtDg-O9?876Kpa4sH)jYVQ~tN8X`rAFI0JwtR;zT z;~yO0o25enm&58f++Fn;ks;*WY{kQv4Sr-D$}PXt1WC4}MV3=Gv9bEDWcT?5$Hsy) za9{3_Dt)+iOnm7gH|d$MOeQd@r&^+nP4gOJYcn*_R&6gS`7-G zAD@jf3VNKkVQ4DqvQu(ii7co@XCVUP*`XP1n4?+OyVg!Emn&f=Mi;=Jf&#xv{K!aK zb}wB!L@tQJTOTlTR?IeMxcnE_N#S1^j;yPYMRFqg!fwhv3`s&go%|5w4MlP2;5)6| zLH-yJRB3&sKse$x7%;CdQgoJZ0D6^hI#Ra9O-RfSoNfSvsu2>}i7)kHuPBy@R^h`m zZEuUzjxWu~8sa+(Uk#~moUj~-6&35nl-dI*8w+S4$52rg(kz%~!U(uVybwm)Z6LT= z97}H2mj>DwA$gM8ZfGX)m3NB-Rmy0fV51-I1WjIILfuSQ^HY!14xZd-HUPLT4u^i^ zL%lXkj;JNo8p-a1Gr9&BztIoxgzgyYBug#cEB?dz6b4ms26;lYNCeX%US}`*8QLl# z+$-TAlqS$3RUW=l&p3Ms_Gj#8ZfL;e0Dzn@0-&5Hoy%w|skj*)C%FqqnOBp%=ZUbi z50W3g5)+I*Pm|Q2lP}*b7Rq^tuvew_wRdxnd<1(^bW!Z9i%}rxSN3*(V9SD}n!(K1AkF4dcj>vZa7JBeL_B*v@WTw)JnZw7gfnY32Fc z!Y$w9rDWFlh-&EXtxV(h{z5gRO0+}2QH|{rszH^%m~Nx2pxDOMKo%~kwIC)fUL}8geYe#RWQSR3FiD~XG!RzAVm%xo1|cSb zX}}F#n)6=l4GbGid~W2OpG8)b@UjJYG=v2^FA5!;k`vH~L;og9QsU+42ZcoZtpru| zxwQ~YnUwrQd~gcY2jLP;XbPP_fnk#VI=&L6)r~84iev|VVVX(};wID;*YbHf0>fNk zoaD`q=U&m(H_t<{Iz)M<`nt}9;EA?q)e7s^5QrnZL8PW z9z*mcr2{y>B`ALayeryR0wFNMz`x}~kijqk`2SX3fyX4Oq5r`$sk|u^cn>bX{%7Jf z;#r=2>K_74`nNdHYHWGf0P-XbjPDvOc$&QpRb`D8KMlyk&Xa^H4Rs{STI$Q@YBUh@ zbW7D`u^YAfGSX!mZAA>=Jc9s!eyj*=_-t1EZ#M`kEdO54d_zj*B0?DnH&T_&UFt-Q5@_ zqf7I(!_eYDKf!7Lb$$(#@rbX6O;=@w2sZJwWVS)Ot2XoQI|<+=arW5&eEr5mur>sm zX$!hbS_=)Df+v=Rosw7&&m)0f^r5Vw+t5Y*YVlkTiJf=u1JA|oMzk=PeiI78O;J-* z>$lj96#?~b#wk(2ZN)#Q77<2AH44iY)=}Q0R;f=(&QZyInYS)%{=rf))nEV&(FQF+ zM&)0JX)dHIYz@*#a5WIDsCW)ZkEJ|>?!{z#S4dfiIc?KJMQPRg%>tO|-Hmg0)`|;q zfq(!S(T6|=D|@UUX&B|k#W<_SKX2S+%?WP;k!y)r98dBId8 z!Y}l`VUQNC-3e!s5Tkb3DCAD~gGu9f==K+0SF7!!D-Iisda#q1(>QKhj`B)K3hL<> zghQ15KE-I~x@Dw%*>ap>eI$2F-x+ng!7rk%HeZ2@HGL+<%ejm=xp6cv%X0)hW9+kU zA||BEsL3o-hD>|DM?#fPjh%`H#+M6Nw-=rGI36 z!!O5OE9Q45|Gwt$)?ng0@k_q{9`bSdkhX7P*aHYYx3{&=bK<29c@Tkfnh@Ny)P zZTQ|_+1XV9;-4&AXjEJ%Fgb{Rt68Y9w!idivcjP3Jsk4(-cRc_pO9`+NQ!IK#wXIb z0-UCcmCXP(iboVZR@vS8h@y|a&mHYf6j6KJTk_Q(%tSDIcBD0Pe_w#&e6xvab+rhh z(;5CL|LVki6om-SKwtgHJUAsq)|LN#i*tv_TZ0wB>6e$s?@I$7$X<7}xrcXuE2ZVb zGWzxNBq}5HsUWnoGNm?u^VuoChf$8#a{dOgSQYzmNINP&G{!mM2`}M>>5r-2v?`3s zVP+IGW}28J02FVe@cn_&w+h2EhXo=JgNX%((te=J6~3T?MPd8Ah=4-g70UDGJv9rL zFeUeL@&efYhxOq4^wpTQ@&)EinaiRRO>dD5KulL<@<9TRcjS>ONeb3pX5S-AW zyBG20T;M{bXuRm>@goK9f$YKp+%g0sLGR}AmQr7yB?k)go$W%?{PgiQgD&jd-;k5T z`iFiP()$r8f@yTCx#LSMoLyHNFvU@eZ>1@Kqq-WJgox_gJ)oV zd1t)HW2ewITCvm@q26ix;$a~C-M)EX=Ml4pt=TmqRt-U7WxI?0DmRP1+e+nwcnXh0 z_64!~5BaIlf6%NLU1X+o=KcX}VP?3_q6GD$%(hGxzu8m@9|wThnmvf{^C;P2`V(o| zH$CKN*Y`;?TrS@%Q@GLOvWK}bnKFmy_K;UUm`1AzO2Xa&){K5^MeeI(Tm#(-#C^-1qQhS@P-)60 z5SZk?PZx0A6`?WSwy=rn37g6rV+PLUSmjSzGOW%XCsceEzTWGbsN%#mVf^j17>?KlZ6xrSlUO-vzg^8B+Go=gu zP#Y5x14(~5AUFJ>BDkZ3o_JsaL-d}e4dWO;{5(V{weo5Z3^T%JI8kL2!?{1bCi8VJ z2!ZZufu;G_%pMODry4h?X(2dCr0JfZTU?IPC5MMmu?z*2a*6T!{#Wvm_>yR-#R^@x$C&O@{?Q$}&)ntE! zvP`Dc(hv^SAfDX^O0Mv8^(R5NMBCIr(O9ucRH}P+!{ddZ-f3}ur))P?<_3p+?>_ux zArBUb47cdr7ku4CpdSo{=9qVxiBDIwyk(wgkAE) znYE&=k8oN|O9Jr36#41O-1gC7Hp>^=nJk!_&v8@;$@8;8y!NJ{1gIT;!YXc4tmiV-Nd<7_xXiI9-`O`6%mbx|Tem|--}*>?Y1iez4$ zMOpeCB11XiHNHNmTrf+eR5=yNCfDI>iAfw9fgvQFN{h%8VU5NbY2>!n{`f{5eRDtqTHoyoj~k*|-LNQ3Np_k*4*!`s;v34ZRmGE=XO=?u&W z7u#`#U)+n(76I2P-kHOODc4*j&`y=3%7Uxm%Litrm4HI7qZLEiXDJ`)#P*vEwK5T| zgz^-QREIJ2N!>P58wvR8{C8e?m@iUz^GT*@d^!D6QpwWVsQO9p*Wr))Wy&MBe@cCU zPrKa!Fbsce7ysyFU=J}AKig;b2Hzt<(n72LQ}~N)020=-$UQ|v$Vb*cQhZs`yFjTLTHN`n}sNIA1!Be6m=-|FP%2pqrGY zM|wGz?}QWtMh)eZ+C?^h+TEkG=`lR$_X$eI+ia>?y)_($bHuy~!(sAO)my1guElA# zUE*^2#mQADU>Zu&R$|@KaG2-${K!i(sQwr}<0Uf0tF_4?gXfst{d?BuN>BL6SKfw( z7V{gkwGkuot7Lwz^3Ps9xN&LR#O>wQml_B;9%Fx&KoAx>x#`B3|Ni5EC;6T<=~LO^ z7Lga)ck@+0cn8oN2JFkOR1BO&ENpKCbqih95)7CH04%}kDgZ^%VlA|PNp+Grdyslf z0EJTtKa9;RY1tlDzN^mwLm7AD!z=|(i~ts)nj-WSu6v3GwyA}xmISvJ;af^R`Yj{r z3)8K5O$J=kMBQ!i>4dW{lUvCyZ>6_W+%Q6()|fQY#Mus{%m^(HTGFaGKikyHaJ|^u z$s9N>-pPVRMB2@!Aebk~iDbyP$c1{>?c_L(6YS*^Z53O(CTNG#OGgh~%N8blXxYij zEa2NK$!iF=D#=88*xxU!n$|ALsBN%fEPKB{WLel+(qfU_49jm*RT6tsTJwq0bGM*^ zMaQ!CD=zlF!i)eSQ+k1iHgjda{x9Z?xgh?E@-r5Eh!g# zmF-a#Ev!$wdrrE5cs3`$%|lHrP1S6P(&*@Y5?IH1+r0*Qx{7SGYZ2@c>$nN#O2S?4 zrDea&ez8ZZ3GfvUF_Kvnh|_(tA&kSrDtuQENNP>T#{I@69vXDL9zBNdAcsD*)I(Hf z8O|^n0Y*1L55i?b&a2bdyOBAZ$R{|I#l??D75yEzF)M0{UD07^0G^tCERg<7ref%# zkG1TP460U+R;j7ErvIOwwHKXB@mR0rzN6bWe?!n{P# zT=e2WAP9sytJ?7|y)^d!fZ8*6hfgAk4?SikR04tgv}gLqe|ng~{fz7Wes>-V0< z!orPZT(IQ`2SBY# z=rFSX%d@e^9?(q(rxg3IfXum5jnAR54(cN>6$=$o|8hrC2m5jBFkm-LQcU>CM?5J* ztj5rR?@u|e;@(VDD)wqW$sgxxU$aYy4>y!AWMAmsyeg_)n}6D7qt|>sEzsGxJa@bG z+E5vm{8@$C)j{S)pyzYhjI=R!2oi*l%>q5ie60uE%~Ai|&T_9VUZqsR^L(|G2L1za zvR9WcMki4Cxvoufqv>R60CHnVtvv>Ooh1=(6;lM=Ts=iIM zNU8^(6NSUwpmw5P-Xrbpo~f5~quz$A-GE>9WC*vqqbxT++tU-5Jb}1ZG4hRU>^gAr zx9`xrYf09YY;yq?DIQ$}MQg!u;)_A1!eBL;gr@503BKG;JEVsM8C>2vqXiH$|4az_K1N?;k1h=XV5zbx zcJpfob7!dtKVz`;?0<_r6^-`uayKunp&nQKRbRtL})uKGNS#tSPPTsS`r%<>LgS@P-un$qtzBUY0ZwWEKn{Q7U6Lw(r~|AXfc zN!wQK^K|KP$ISk~@te6sg^|Jb$n$I>fLQ;wi&G?S^-8_w0Dn)TJ^!TT+VW}>hT7%}PoTHLj4sTFdF zA#9DtJnv_KNJ!q5mnGzHJJiI!0j}Km#Cj)_`06zJBOWbV=I8GP%$wh`L=;Hdr^$6| z;l3E^r_c2xt7M>sWXS8RX^qDdvFNqjj38sNa(15Rwl(mg)We zoYiNBwl?aZ0rg-YQ3LIvP{I2`p!5Dd|8+FCfR&#?nA4Pd#Pu;=<8F)leNNlMH>>P= zRpohVr&Jx=<})24Am>Gde|Oa7N}!8$;(7@nWI3CY%bup-% z;SIjgg!<#ks2YMvgafl9%|NP>VOCU@x$z}|MPdO2@s{~%iSbH>S>wQ6$SVR>tD+*S z{F&h5213jX$7c`mJNXq){H}U;Nog5edH6vk)m-eTP#vr_Q)Sh@bFzL1=H&P4QAa%M zx~P(uv6X{XF@f09Hh@y$TxHpEp>`7hESTI7(9=SG6>HwKzU-likN;RpS`Oxkf?BJ4 zh*&FRgA7@5@rxo@zy#-Fidy&g3nfB?vy9bUJHar12tx|PrM+D@##J%a=P?|3N2_vYBmbA3MBbw9uSXmGh7*2pL7XHJTLTV_PlNt@_)&bAB1V zx6Rb4tuHoj2l|DA3dF{ceMUnWDK)*07`_nPiPQ-qS8VL~7${kOf48*o0Ut0Krgm)ZY&wIIXU* zE2K$5d}Aa0m+q$$K2#uj7>ooPi$%8~m7f9$1j+F>HhZwDkQ{|-Fvg{bB0v+Fpi2tO z_5j;0YDlj5j{TfmM=+FXD?d=8$?|heEDQ>{F*Jzom0WdfDAb(&l$HLaNNNRefbN;S z5fIRa6j#Ln0b=0eZ00~hfAn)yY(nCTvq@0TaF~_Z34{lw1Qus|nBOXWcoP`yD5WC0 zL9oaAiWpZU7BdtH+#BT4oP{0}@;P|-I7Z2a3UCkqibHw84`J^);CnU#j=nOvM4KU7 z-K=!5n?pjTlqoJH~vVDox-tV>0qy zCWaiuWJnJCBSLKaohBg?#gF(sbbj{ptOksH`3%X}E)msW!BOB!}eBoHG6yweF)z`9#C>W4cDmEwjZaC>$ zt0kGIs2hV$e}otcy2n<&cKJD!0;1VyCMhhRgPz4y!deEAA>b8-KrTHip8oQ!VR9jKS~WyH~`&0Puu%i z+#*}H6B}nGh?s?QG6X_WXBa}2XptjLUVxt?Je9ti6M|@Zl@rD=O}8G&tDK%ED!16T8OlSG*zfVEJA*~Tm`Rj0;2I={s&~+!F*pC<7TQD z;OFG+WE77u>SP=J-R+MKj@!9`kS~K&3C^Otn<<@F&rLHiz*8JpgC@FOLj1u!(Ash;G#mT9#7P(Q`+`J6QafuA{X@w;+xusPDR3$5kS&wL;JQ~0b zt!P@m@+WA&kSvvB2A58x52R{J)S&-8haK(_W@`K&IqVPr=fRPGEG)mf(>@>Jhr1!^ zlNsS$!N=~@W1JL_b>WZMep#0VDA266%BQQh9xXTQ4cAK^E1jvfmxc?z($g0>I^uVBh>p-XYNwgISGSpPFl__!we#SJnIyIan zh8I7Km%BSrG{$gW1|gnEC#9TA|7VJf%0^_1Uh0pt+ozm&e+JKG}+Ny|?k@@nOSj8#b5A zvZJlU72oGCLFwCEP#vB1VvYz#4-TefD9n!uN3#ca6G_>^ z(=}tNkuZ7MfJlQ;FyKk%2g5(4Q!r#yP2==QsKSowVjPU#*)&Q9fX^6X1P&Afk@!2B z#nFf=DSqHn3LF!X1mO*)$uJ_xQkS9y=PHOBl`+O@D2G5nBqOAw(dV*=vOF-L2xF8K zW7JBz;3XYAlcq|`ELq8NY3waffGnEpZjKI>wvGN?NkQC;wSFxy<#Z zf;qVb7cy-KSeC{&I%jsGH-h0BeoT-CMYE~5SSsdlHTy3saM6a7breu@t}7J5HPI!D znWO60s%7`lJ2JH{8rWhMzvc?LlR6ul`KtMefb=yTDU_vh)QLotXb<< zrppj7#&W}k_TuEOcGT}{-$|52e@&Qo7m-kqv1R&>ayxK^!1Ihb+O&%1yy_DXGd@Ch zEv;!fyoHM`wim(-!V{&ACQW{#aCcJ?LIiwpc0;vH4@fssGW+;mhG}Gd^m(dI0s)i7 zewhAu!jiFWwANy*k|M9^U;n{(r1xFbbKE{t3vv0JXU~>6(exAa!+)d_^ag*) zu3L%Ju)kbSu=sqrk>dWVW24nCfws1gjqum9aT;`YG7`ahugkovyef*HsPVsYVI2U! zt`6(*79$54jPEtGiN6rj#rf%V^sZk1`Rj()3|gmy4os6OV;ZfoXanw5kF4fxeh?i^ zKL;kNMs#zwXn!kGc&;%O%{X;6NjN#Jd#X`;b$S%P@%By!v46Ke9yi-X%o7EZ3ZZhO zWj@iI(~f2dpwby%bK~G!Kx!!OMF&AMK4j!V?JNJ_oYLF$!*8tIz#?iLN^td!n!_c- z{w>q1UDS`%kBe#S#9u=xCcX!;JW2y#!X>+Cqv(UgU8Meb4n;x$BP#rt0`cEFOJ_KR z&n5oFz+&q=A&i5nhYekzU^F(vg@fT}z@6_^%9X?RNS0+9op$- zWy$O()9U_Ij7s8%^tfM52Wrv%>3jwRdKCY8xs%}sxEGh!!K8^MGr_0#OhaeD9&EF) zr^FQeokYa29Rq+F_!fV{-}jlHut8Y4 z;ZSlziBO$i(Mv4(RIgUdy5EiIdNp;FeCHl?RaLvbvjxXNc`w%Kw33b-Xpx^(He0ZL z^7<@*!a*7HHsC1{)RIN@)0?&SoJHT@%Sv64L9@o3nl&F#U&p58Y|IqX&2m&egQ9lc zH3Ry1%;kgzsPE9?xwWEXeVG~HBq)?17W$b1NU7z((z_f6>FJ6EX<0gH2=@vgD=Esd zkgi2@bE?w#kaUl$s)Rxd)@5YOszx=n@XAyZbmFYcMTOITr|@oXBy(b;Zl<_QSXl%i z%oQ*}JrA}mr1{_SuBRi({X(28u@(_pp?uN^iAYWX`Dq#(TiH24YTU?-Xt#!8|01-Y=ZvGkjgobd@sxTQpNhS&C6Kp zB35z1Uqd#xQa7#|bKtIaZ(2S3$CAebLmYCCs1Lv^q1Q5TrRLr{I_xV`JIe9|)Stm9 zB#J)}F;aGm8fKhlvwW}3cXoturgXTP*NG0Xp($PiKYTUrW{um7@p$2g$dlKQdK7{s zSo}A1;C3gM^bJ?39Hk2eKFXtE2)?z`T!nT;RMWcsz3~QF37_x`thu&{HewKEPJ5{L zsk^1~N9K0}$B5YcP6I+H)O#>+EkROMqD(8pEaaS<+8(UHY<`F&{sLjUTt_3w*c|M= z&)o@pVdRe}zNh*bOh7$b;yTN4Z@K^{AT_$c?+LZx!q-C5XDoD75~~xfRdo8^aXxqa z0a+a~Q@JW_;=v=PVxZqmpzCqdUw@u0@NBCbnnz>UpQ~P;g*<*Eu-Dcume$(Z$~m_G zk?8x=AGjxE?xTd;RlCAlv^Z-gZAdqKHP5v$&JIVONDBQV&zZfwSaV1{I9R{lw!eb4 z9{NIlRC3+lD7F~(ys{@*1Khq|Pyk<{f+_C>k=qc-4u5+WgPGfq1rOo4;pspo$#!(@ zLtm1?^klO9KfH@EGf+9UYQdRI)ChmqwW9t41j_&T-kA?4h2gaa3{5{jP{aMO|I7tY#{t#$nIMlY#=1upPE0>wb&3R!ToDRj8lG~7;2JxN1pTR6iVvkS>t(V%OvK0Ps zp(07NPS1E)3KjZ?<-5e)8rlTJuN6A4?&R&ei`xk}x2V6YmP+Y*m=-PsqlExcpV6#nMDE0)z^ZQtgY4)@a{|$r7T1)%|w@T z3f2=%L1@d-ca~C{kYh0YdYW4yX`UZL#q?&nR}IN*x*On<$&V$@<$Lr;#(tADBU;vt zOtuy)?TzT*pWAe~#AlwCnW1l4ta3IYSu=BTbMFh8it`)#spE3-q$A?Nc!fnJ z*wWK1`Dy?G_N2D}n~b=~iO@``H$+9YaN{0y#(=yeiM1+>k)+~RTZC6^wK*NVxuR)r zd+F*?8T)c1cKiG7eL*zlT!^5)>&U8;h0T>le6J9@R{kED2f7+FyB$56^dJoD+`9t} z&}T<-;Sj%BaB&zIvp#i<$D_%4lE&K_ z`Z=EAoZS=;hMO8J1I1`I=7AxsR*k}euL=64pD|NWhalh&j>@A%1H071f-xVilu#$z z9;NC}spA_e;I)Ng@l9&dxjvai{(#;)AypO~h&tWI`~LhbK*v9A-jN9jzxU_OeDxc5 z98PLcI4?!`B&U#`TXGb>N8N0wrB6M?N|+*iKK04g$59`ASl6^XlQZ#n!F5ntygUZ5 zLP!gx#ce+tXKhs&%D?7Hi;P&RKPZ4N$~z4Azm{3+$JgZ9ALpujeTF&mf#+-vNpkya zISlpf<^BF3&(&V-C-sp7uD{&8etoC72ml0>e_1R2+r;N9{{J4J^!G4o#ea3rEByuB z|KNuoQb8a1;abPlzo{7pYPnv$dy6D-qZYw!)E@lnI~`0&Z-#$F7Pl`xcU*?wGo!qH zp>?w2owg2$dx2Xf!~%;#kS3V1U-O2;5RriFns$aj2!7cdS&0?kBN9V{U1MG5T=V!b z=%Dne-AsiYn{)gehXqS*20jjfq`Pb702=*$1w~()uyq=JBSiYD-+YHJ;Le3(t`@Mn3avuck(mCuw&QLeRfD$()F-sm~-6q$;OPl zJG59*pLy-z$NQ!w7JkZ2vo6T8s1e#xjz#O9wkS^Ozfvgxez^RYpSnXfmYuS|%ASi{ z?p2)K&N)yFZ4qQst?HM)F0O_YSh3a2z16U*rAGc2$t4rwlYfz1()@c$Mb|IPCgDpV zZ2#AbqEG)vytx&)z0&~bxRU+Hi(AV3F@ke-s&`0|VozdgzQnyprMtIjm znlA-39=R&u#v(=q3!U$ulNsRydi7u=xm%CqV~{fxX^YF_L}h|FPPL3f=M$1cU%WI= zgoNNOec*w4Y%RVDN7FR=DEsm(OfA&Z9mf`~J2q`BXTF^+a)yBGPf#@NM`VDTn!ldt zp2L*kj~(J4t8Lj+udacfVYcq&ydl7(0a%d&r}!M?Z>9_1Rz@rP;uUUXhCkie%8C$Y z-Oi3tx8BZ4FfZKBO>x`V&QmHu+sV(4lg3TN$*}%?v;2^2LjTX=+}|#T5B_k~jM97M zy*$%Wo-$i^hza9?%UvZ5QORc{DDc(c`czBpD@&2z;M2Cak^eiF=5{qPrcN$dj zM=9#Muvy|tM1>#xX8p5>j5@Oa6!A$nPrqK0h!x5E!&orh;68-#x}x|2GKk!gp5$rP zQjPyAMdxed<##o=38M%IBWTerAE!Z_-yhwM+=%?jA?$nw{raj8LYlk+@ef! zn8DQ^hf(?v6fGwNu~;yCZTR`MvrMBgO^%#?YN|kZa^b5^mc+MKz9LMDhT5Z5>^^M6 zMDqHRO^iNPG0f%$SHE@zoGSO`#-k!MGw*wk3F&_5Gzmjocj>+H-Uo2T zh!ENy%@qVQ|E-MBN8*~ea@_q}bWm=h%}JrNDI_BAfZP0R2;Yt*zWbRdq9iSZ9-6Hn zz*(>!ayJz%$=wEw#!P@AsAO}!w5zB0wSzI^(n9=Mn7CogzW&M9g{0b%eM8$!38oDu zMrkjfk{aVV`=ViT`+$%nspYXvnKN)&B5|qB#Elf!2=qR(7}r zGXZKTFU>J)0;c?~NfjDm)?hBgzk!4LJGLl$$pRTklC%^-Js!0bWX8Vaj9zIyMsb17 zq`(RkTi9w1aHaPFZr1CUp*`Cxdxy}-zX;`cxksmoqQjn<15leS@8S7pmg~RorMkl~ za18&wS+Z-H{!!XoJQbF>0JXrDu$)iz+Ki@UXbUq)V{PjYGKk&&w*ycKc0U-TB50gpnpvI?ZvWb}* zT{?@IC4=>^p^O4(7q@wr2Kplr&oESX>-jb_br*>^F`t$1&X(4veF$$nw*ll>eBs<} z^w42M(!!VInC4v^@5wT;<5qlZw=h!q-R#z()V>cX#O9t#fza(@8dMT(P&#HmvyNng z9)r`G?|gRATKN4tRldg-!WmP#irMY6-=LNnD)DG+$rty`v<-qUZn><`tCw#id)i&K z7>N27btRZS&T=JWh9!?Sls!;&$&;gpIZ~S!&+$_@|4Wxqt{0&Rs~>^mr(xt~bw+{Y zY?r9n4iBD?q+BV7A~6ngYv13?ROGA6nTu}c#4t83^>gu=hpft}giX?E6^fCqa&#wf=D4$7grgH&3wnBW?UlxBJy^}PMo0`!q@eQ>9y-a6Q7wlKcy!o}waMYAx=kD|mc(DjwBT`)rrP6VI?8NIQF$J9Cm zsRu}6oV6qP@&pc$RpBZ&)>Uq%2va;)@1@^|w8gRc!EmAIGvCWhIte(gXmq`qA*wBy z02vS_?NuM8{2TiHVJvU^@)Uq5>HtkexPWd`!_SZp?RUo##Lk)kW$ugn&sqvVe_R?K zh=)&e^R@r{{l3^s+{Yq3BpnwggaNjJr&|7P)84?ANkY9DK1N(`)ajEj0#Z|&Wy2QJ za&n?$!u)0LEIU^Pcg%KI98z4-EgUUvQ3xr}c+7M$kQ>kaQE-hJ(XwGPghn4!WlOaB zu`^(T!d=^GGr|aHnjt292XQh6X^L)QM5uzMr{KDBsx$$o!H5ANPHHT<79t}*Ga&Ad z$1o7}(22x|m=5?hQcU%$<~2PIo*5>&oLP;jWCU2@enRcuUJdz)3(Ni#4!9)CcrU&~ z`3Xt-H}+v2$xL!%CC^R^WPQLs)`i(AX@%Ai81ROjqNnxu=nvS3j3vEhgLTZN=@%Z7 zfrGAg%Q#|L*vfmwO^en0p3Z{fP1?lN$_828qAAdG->RpX6AJ!~Tz9=)Xm?+||fIhz9 z?l2O3QU008ApEhK@N!Xk3En&Q<6ipsV(T0%WwYZV80Hv{Xj^e{8HRnYtTiqg4HN+v zNDqa5<7u{h5?){|p{PLfWX*cy)uumi1SOz&UL>~P>Su;{R)O=VwAszj7zU3EK(xbo zer!}W*&MfDiDv!`N`cQ^GvU{vcQj$nWwtmD=)3R3^*fg})2vrsO*&VG5Xd2DxALZ) zLChYx=-|QVmFt5;3-=X+f}Z3OYuZ0p>G9!HyFj_<2~erq5b$MpOPF|{y{U4Y^;A+I zKMnd->4&a=IJ{V>wSvt|lRc*Sqzr}bc{BF3@qk_9jo!Grg zAcosgbrcf`ZRj5bCC7$f2 zGUxI}lcIs{fl(PMj(qQ)_wIoRsn({BK7jl5G{rB1E}MSOQ;=l@8q1 z4itb#lw;jk$4@w-_N7cQdD}a_Vj)8jed4G<9jcP}UDV~HA#m9YJ!8f9pd;43dL*>q zDuB&$!`C2T2&)8U{K`HnpB@2<2G=y1NFy$Dpip6Y-`Gv=VG+1gj$(#hE^%_GS?hp@ zY>R=I3yp}836J@j`kZBE!L5Ur$#~_2vE%xi+|6b0JAHo#O*$QsB`XEL3b8aUH1*R* zg81~892r0;NX%%h>l=Yf7SCS?J%7p8p#MBX{zL70k9t9etF$cY|E~u<5}2UMguiGB z(`g8GbZm<_D?!~T&3CvE9bOYNZu{VGq!oIU)@Fc|K?bBzVMQ#`%;Nb?OT>!^D^rlv ztaYW~f5)Z|jo6>8bzG(7U~}4^Ep~pOC8`)@YU$sB#W2P`L)+fqv%Q>oovq?$MD_0G zXDN@Fui`dY71i)KwJcl1ekwq{>EMJNU~>a9w` zOFhyxVb%u#C+mNEVWQ2U#UqLfuHtM*LUx}F{3=w_7HmS}{tIaxc|-HJO5eIUr#D}t z$cwT$$hW?2*C&CjeaE2{4IC%AbY1W$i~W*hdq0Pi>>DdU2zEvZP=g^I)N8;HW#>c) z48^B6jjZI&a68`8?GTF)oPZ&sHjC|UI%a1C`C#~(6aBalA(D;qL5&I|MP*G* z5CTQj+(H46kesd#+s6!VQHW!85Shd-Sk=a<;NOaieIVYO(g)+9(rbl^G+rt%0U6WYCQn{Zfke zUG!!eN%O_`*cL4T3#ih?1exRW7H1jJeW61bB97>NOk8Hw#<&8==}Os}t1fPf$U(vB ze1!{B-ZLU?VcSy+x@@#DhI>G|OrC+VWr+|`ZGKIZC2pL7h&V@PWg%WhBePV2ET=kc zS6e-mDjtP6x$r?g`mUoOJ#q*;@kDv;>i` zDf36S{^ctx&~VNKb6c>v1*dyFVlt6EQz;vPvCyq}+lSMZHx7iEMz*h&@hK9u3u-CR z#Cj(J1k>`IN9vR`)tf7Ts_F0^So$-Gk{B-S7ojsfTGWGiF<>JZ26!KzS>eD(xi(Qk zSykMmF>g^R8)=>P8iD7GQCt zBC`z@EU4ziIwh9+WSfzDS9$>!S?}0lkcqvQvE=KY{6-+Hw^zHQ68wUDRg-B-3Mtpw zjr&alX(S{_e9sfr6~K$vM>$l#V_|;W=+|%F-5gaS3jL0SY1W-^_>ZwLch`r1Mz`PJ zUTl`$-`$-5_%)4UrX!$8wgLWJ7ZCThf&OJ(;0Y(ybOvBPtP4&JNkt@EVE^J?c8u_c zM)e*ISJE*Z3O@(Z>*g!Mgb@EruBjCdjc7cUfw(^!er8)Rg61C_Ti+i( z_IWKwOO*`b_nZ&zW#gIZ4;*AH<8Yz@i}bp5O|jMo=sVJ1-)~7*xy5RK>UFH)?7O>z zD?h=Q&x@_n80alB3ET#4!T3};p66)`J%&shs z=c@=4+9i~kj~1AG!A>f()1S?BxG(b2u5tKoe{WUOcv*WiEgD8F{OyseV`p(dr=7eH zr~1lw42{JDjgsxg;r7?}R9w$oMu*o%nXuoM)n1*;SK6|ewVq#flOl9hUhZl*~N9f#L^aQLz}St z3XO0HlIl;%-Oa%&-edadqMnpC6U48y363T@#Ll8PKsHZ7c>%G?b0KnsYX}vFYt{e|kql?Qzm<+k#t z*cHWOWyBdVoR`a)Lmd_{FbYXIdmTvA$*PuU2mN!T_>FWI2TumuGKX5*b-z$B&rny~ zF?R3|AysxLB0gc$=x*gL-d30}jaO|Z9If+0c5)lm)j8EadIrh3Zb0GU9{)i3n_ zAs6EjqLIR1m>>NQz7r2o;BR>t{|n#Ae(}-pOAytwf4p8oW?cbAnywGrD1j*~mFE4w z8T9GHc)+KZp1&D%rHH4*4QmN8xsW_`k(cki9xy+X#f(bXPea9AP!8iy4@LHgO3P6} z-I-XM?gUf)4S$q z6&H#G=xGw*hMZBqmc3Q=a9o?Jw77{+zqH&)M{2$Y+Nbi(FVBqT-D6@(*}WS`y)V>r z^D;Yz`xGsHy4AMA2ao!9t;3q*&#ndT+(51zRf_V#n#&V#CPbCBD5TkYQIH3U%HU+n0rtc+< zYcl~W(^j&^x{7D6P0CKp3xjY+h*nZazPwI#+)C_tDQq5o4sQHE4-D$_~)AAD!} zpO}BAAV0UTP?D>7r%*et<(`%8RRlOxN%GCN!UAtVB6L#Kk;58n(C}F>C8tDs%(y&v zhh+^??P@CWii1*;J#6p;Yh0GQHGf2E?c7ON9r`7{P0i~o!FWf^mor1s__&IQ_*jOG z>#+Bwl?sVFQ-2|xn1VOphsJcLA8=6Jl&p}x5 zVvDG4LoU1Z_ylctH5E0Oujmpw_a6{MR%I3v%yxV)H7qU5J^ZWMlvVxQOOhAF%VC`c z-Q>#5vqk;jpb++gecQabUaXK1&f7}VE)^0*VQsyA`yov47ubA$zGnhJ(bSXH+d+ z9ySnup}dwD%VWM8Xy)%9z+)6wGPZ7AnI$yGV6;%L=qPR0`^O7=KK}TU`9^uOiKZ_B zX!x{phO~T=qwEWv!QQ?%$j9o4NFZ#naR0_khg5Y!O3n9*OK0wZ4fGNlARI5 zm$bL2&gKtQ6iVd>(u~ycC2(V+^aI=&cX^7=W$Fe{H@L|VQnomMx5DHt*9&H+bX)mH zDOI6-$FboN&vhruFe1ag&Ag^3#Q9WFvjwZbRGadMc`{N)ypJdFY(G1bd!I~Rfd_V{~7 z5MG0Ey1`k9Wv0J)|5oBh_0X+s(_^cGoCNfSJoRKUFRMIb`v#T*leGSA2rQEKM=yQ z(|;1e1efwlCGO`0b4kBn8s|M00s8OlDHEPr8Sh_&@J?UKFXR$%Rl3`1AV>I=pZPRi z*udhJMX_jUa-!BzZ$`!Nhk;%xawndt&d{g82P^R ztGMcrCpIn2F^^5&d6BC3Ao%hI;>Koi&kr`ZzWSclE`AyWuQDBphMI0|Hq`!TDieCH zx~mYjJ(92F*A@k(UR)e{@+FYsvGjJkCC6qD^^o1^k6_{__Y4PwE|z{>d?{z=Z2ek4 zd)o&jFl%gXOZClah4|x^xLB&Sa@)1EWX(y;l%a0WU#RNc7 zd><(j>Y9MX%t_rM)Ou!LQ{)D#TC6;l9Mc7XhvETwKUXk-t*3MVE7FV1m5=h}9*YrR zPo{3z^ka7|Kgd%nf-s!4ya+9A(F{a_siT0*Xx(G_oL382rg2(&aoeNHsVs&h2HZsS zSrCDo?MLPYNmiSw9vcsb3W3`bpNx-LER`Lwtu1v?Fk|u|VU)jjT{}@mQAyJqqOFQ# z1W;FOF~#SgfNg`EH@P7juX4RuhYE_&O$U?(y6?$`bF$qMcBIomf~<*q4D{9|i=?>) z(2Vj1?V?`FqGCX4p+H1gf)yHj#M^!;cC9J>q7q0wEzr{aS`H(-T&JaGreMN(hqd?{ zI`3p4`Y+}jaJ`9c&U2Y|0u=?6q zuF)~xT$1g@v^%$msOva|?Rm%}^;-CXavLJ(kz*Gkw+(MuPfMcqLI<@s`g+g=-Z*>8 zyuk9W0jmx3^?(WiJ=@ozqj2J0+R`1DF_es(;^D_8b`Uw>*FhI^kUQa-u!&0Qe!Ysy zyra=~Wq9=j4#6_4SpOYI>K9q8%hMX=u=_7#Uy#=u#yDi_<{ z#cy!oJ`)?HNqD1Ha?!d0Z2m}(fyXM^pSOZ?D2}InYDqjh zA)Q*?h&UuiqMPF6?RAybX0q45g70FVzg-glQFL(5Kwp<|UUvgSv?m=p2IE@_ULQM& zQCNgstccB5dL3N@7X za)scZG3v3(_KwWDMg zj#My%t%N@nmE^tyD=$jZipV=kzLYBB_(-3nO)@R{>K5J5rCx&pnNZSiu$xi}L6lZh zfH(}zIoiz&Dgr=6HAJ8z`RrmQ!W<&~T#5}>%EOG?aJ%0WfySl!MU0T{biZ*A9xnGt z7?odQraUb(dTBT#yC+(>=$;w%H`9%ess&Z*fYo>#DIj+f8m4d>G}+8Jn&0m+v190H zZ2Q(cWDdsV;){aX1r)~{bB4(pS}I54Xypo38Uv8|pLo_kvY_3d?bl?ai<;)0gz$BK z^_`-_y)tDMDDlMAmRE%$BO5as^eHybug1c9vyISa0zW#3FiR_CFV@;}GPki#r!Jj!GAJ!TlAoHy_>`bE%d!`@GL5*5GA zu){>)fnE77d84^(0fHPCm4zY&62B!a)UrWB$o}IjafVZss*eNZt_VyYxT7@8f%4pW zIPWg@jo^^_K&ySJydB$vI~bZ?L)VVV%3uZ3<3h2i=scYo2r-ZNVExHNAIp!BmUDYL zgF=HR_d3m9Wt)>_i+2srt|v={^RQ+HS(Ye?8D)ccGo<1^uUAuRRZeqb9>u#L~Y@aplQi*~b zZ4m6l>7~1xN#t{qlEEXQERCoN31D%~rd@H;%KG7sMk>4bAr7^R^L`-=o*_~D(uR#S z=N(4d3Z>SVk&d|9>g7PT^E(SCS#p(fAd;-=;kS$@3IX^!XOhfA?M%e)N19R__5#UI zcDf_D?wA8Lv+OaN0dtT!<%p%pa~r@$LG1$)n&ScA`(;dDwN z>}JV9oL5`f@%C5S`FYqn2!*u=SG$3BjL!qgChftCHJjg*aeqO_#)CEsf34|u;zisa zt5>2GM}us4%ndZpxSFB8!h*-AQ%t2d)_44gLTBSh#vPylyTpm zSuhOk21pVeDIQT9?MbMFJ3xU7i@{F-1$5WnHc>1>WU5=)^#DUe@$~I~NfMG@wGJ-AbT7I9msWP<(T}5bPg+7)OwmS6=SSzx)HRh_ds3<6$ z`E}*K)2;@-3<5BQ@K z2t`sRS%L{UStuf6lfXXtP|6>LSvcLs7{Piu%qe`?YZ73jI6YhGe!1ah>L$#&lEYl* zQ=3^n*iBE2WbaF2vp#S#7OcHRYg$^1RuJT$h!CMD(1C-RI}v%oWGj&?6 z%s`pk?QtP{e#;IFhn8*XaCA>`GH_`>l=j+j?oP%s> z=ay9;zgJJ=c@{=4zi=W$+EIVCk>IYbRxG^2<3IiBoiyL#C#<38k-5EAp$F}5LKj~g zt%SCEBBt zP;xPCf&6xECZpMQ%Xxh!#&RXhf|F<#H8FK5b4tXO_cc<1Y@wx=&1b<^$zs)sl8{K7 zgPp2L`|JI>&Ck~dP3H&KKiW{ZZVu~{p*KgJCLeB&U;mtR7lS?FY8LMi{&?}T|2gHt z*{J({T^fp<&G76({HND@OS8!^v>Yl^3$GGS`@5UQzHjcnzMhQ2w;LMwiFL3(eLxff zDuB_K>qdqeDt|~xz#;kuLkdIo15SYVPsZeOKezokRlpG`;q z!jLX1Cn?a2AkG5=2jc-&VNN2VOY=WoSUPyzWy%ahMa%y}D7Qs0Y6td2=#mRK2ug?o z8vC~MFTY~t?~HE(dt$E9V4(D|JkI?>8A@VAiR|3@>+cD9r>jpOB>q{B5a^4NVMvG% z^CwgO_110ODtdzX9Yn4;#T09w_LlohEhtAbD~6FQZ=92hemC)jWj`unDEgQF`M zP)c93hFb=}fSCRw(j9b0BMK0g~|8Jt-(#52EK@ib}OxsfH z8ytMhK@umuZWNm5*sKB>t$yc}PhSptF0J=R`}pefC-E-(1z_c3?~kRy1``?VDft^7 z+Z41qZ+`B*_~bBK8yQEmqz&VE;xJbSpVC_TvG~mj<=xfh*D(A^1Iu@nGe*msE$@Wx zKVkD#Chkp*wmOc_(_EZ~21TD=?xvmLPw81ME-vuCISl*CGy{#FcMe0Frtw(BK<$|r zdUr_5xTKJEyCUw!Rd+N%wQ1xIdfgf_spH{CF|p|Xa()HmhPrSZ7Pe`pbJ2|Jr)lG#6-qccoE=(zOISng(Wa%)04G-S)xtaY>(&z6=EjMp+2*kOPtw58BS`=oeKFxbjle8? z;@FR_PfiBdOIwA<>5veaBVc!g4>uogMKyh|=8r!}e-i6m@<*QD)D_Jh@1lfe=R6w< zBsf;1k}e9cTrnYBs9l0mU~jC10;+qLL%jPgXhbd0x+!!7d1A8Uk+(gJKJqS@M+P{x zZ_&PzX3;XWp?)f*WYI{&GA#6b-UTCAp(ZXzxb&6Nd{WF{qC9Gd{Lt0e#NbrvaOWI(yb7 z)v3M_h&6p`PbVA&=XVopfVU!IS(vIc5p@!ni3hqZmT*K6>5J$4)tW=-vt`kq*G7b8 z3sE_BD8<1C?aGNo+^Dw~t2SJR`pG3wLawtRj!;TN82GtBKQo;A$ypddYG_0N=m#d%eXX64=ZC(>vPgZ&RC94+YrTHq^p1x+O#Pb?1+1~6TlZEAFJK2N7sVS zj5|Mb*GSu7!N-p|zW>i5$2MysNpbx77>K`3E!X8oZ+v_uit@- zG{1fbepP`df0=Uw?*DBcM$9_cTW%umP5EaQRphy>Jw53^*$N^cp!_Me{ok6D7+eB- z0pC!et>?oXKRW94(PH<%G^uFwNIy9;H(pd{)2|OrD)u+qO74L5yGg|{B8&U{Zc?B; z`R<1%Riup*NWgzqG+k-&Oq};}cY&kU8YNEb7g!$vp-f+$R6kprG61e#{gljsA~UKT z-0}!aeO~>-ve`c#LGvmG(@d_1OL{9!*drPacc&zbnZGr7*x-_0S>4lz zU&9Z+i$5|ocsE}fx7riUgaT(k((=$hdVGP5z9{ z%H@`d1o?JuG6eMEFp<#V=f--9Fj!GhviDEIR>X<)Vr24NLjHg@ zL7Xhydbws2g2CR70LE&{BCrb&WF6zWvjhi#6sa}iOi>WcQ^PWQ?gjCSv!fbtyR_)M@V(*=7)rEh68-E@Np`=&R5H?I*TbqerNm~PpK6qIpq4di-as@-$>kxl?h3O%RL*DLqU?1q75l_ z50p87@8YU@4MLrlda;c|>-v0te7a7ZPl zmLGlZ?_h!g7%cx>O7VOI4JplA!U#t}=dhU=tI0ufu~WSo0K}8UV@IO#;cllt{{E$p zSjcXVn4CbCIyc!zVGPEtVLLcq>HZ|gE~vEpjQ^8x2(N9<++AktV!sj|d?}GuI=lS# zu{6imPOb5%xa#hU9yn<2Q*~xXb56kzUvTy`vQb{up&EzZ;s_N8T2})NQ8cHZ|D&1! zHvKqdi4j0Nog_1df&7CsYS_7*W=;xkg)nJOGYv^Z(eTQHNm%3XBs@gvjlbyQ0yFW= z7#bPnkHmaT76`Yq3BN!c!t}4^zP@A<!YewnXr}qQda)|rhUwx=v}i=^i*ZItjU|xhWQxo3 zR79n%WzxP1No634h(QvWW>)ePSFcg_Pj$&*&m$SoGG?IuCpo}_a!ui%HHJm?dz~kULZ?n)ZLn7VXVXC*$^DT(dv0bsEXrKe08s2yRu#->|h>n(dFR1yYuH~e` zYWEvDnV0C3OC=AkYEw&?%S+qR&_3nhH z!%sw{BCMuP@59#}L#|@8j4h4zuOr|238NlRvx@}wi#>n)t@--PKr%V+y;4HrN#y%! z((fe$bB)*Mphcs%hdXhQrJ^S*H=7;@M)z**OiM7pd`Fi#h=e-AVbjc1!`ZZim<&$f z<*tC-R$W7FgF-K^^s0ZE&{&@^&N9gpn7{)VVNw2L^?24n~t7kgr^s2#QEhg%=Ta*wJ_o-4OGCX14d0&ZzQBC(Mnk1fs zHE<)rR-AA+Ik0;g7s_nK7onraFt(BQO-LZZ$>x~FQi~nidh0cuve44rvwVvN>dDE< zqGpJxS^&|+UCV=p$nO-W^0-}X=N}+Kb_ytj&}X3mnL9fQJ$Cn38Dg0&4bjE*?Pz(B z5@)Z?(*15!+mAKaYHS9)3_5!iN+qP^rJVrMy|ONf-EE^O>}*y`6s@p0*K%bx`-D}z zfnva?Y5YRTx|d&Q!#uwhnAfhSk#8$gO3kL|*3LgwYMk6=$*%WGvx%ta&*dd^KcJ74 z#Dg%?Mg)ETrn4VRbiY%QVK}WENS;&tQ>NPrdpkws3?Yk$rwy3J!|n9V%h7GmZf~d8 zz0RqFl{(<&gLUhDKYGTq zan&vRV%Reu29a|e-T>+++w&j!*E41~t3#NVK5PYR(EqShQPAY8oW5HDZ7|LmE^-;P z!UVh`cHORKAQvWrJC7)u9rx1iM*8>7=SBk;t4?kLbE~AO%2?wfUrKGbc{R>}G*~fH6ugF=S z@wCpjVU?g}ocQxrZ`2i2pdUUw}XmlX{z9WTU1xYzf>5D7$LwW{cp35QTR4Ma>zT%Jc+M zO1=z=N*comIdJ}xOCTObMPtf*-7PKKKNx)lSYS=ekyiY69EY4u&FS7Zps#Ian{#Sb z9>p|h>^6+qBfA0hH6)N{jiyBvd(_0?KCTH#w$?$g@7ul8Oi$xE1d|A^w-l4Z&q z>-*-`|1I?}JWuT{v%*v4zn%2qB2Xg`BXa+B()Vu|j`@h{|7Ke2OJssw6+FDy?)}@e z_H}OCyH;#aw`ApmpOh34U)^a&Ga|neu3brJh?gD};d=gErbA~9?cGghX~QL5MtKiM^SbkLwR!k~`KwywhLW9Sp% z8&{k|v{p`pu-rGlFkBUqrFqTn&_;<4<6Y0$4Q=X~kd<8&??q+@C$NwSi zt^cCly7%EJhM|XU9J)ggK@l0cOOWm^329JZ=n{rb=?)R3L|Ph@5R`5tm5@~8E1q+E z&bj0B{P6q-?|IGMYp->!D{zF`E`^VMWg5P%hwGmBc?~{Wm~8NS7<=9KeFvEr=iU;6t28w8nMPo| zC^3If@RV&l)*HjxL_>tkmNbrViYY%%={{>_yu@vGW(Bi|FeXX0=>kh}HI>am1?tO_ zwbamu@}JXeib`3M_+D&UsREkWuU$Xxun>7HgdyR{#^@d z>u+dOwpkQsHLzKnvn6L!68>8@2fM}vt~1UQf1Lut|Ah-gUXfWf|2_rYd*q-5&?CIV zwcQl4WbhaIy)d-6JnG8tV&-g{Zon6eqNA7p18yQZd}xmgAynPQD=2l&?|6Ro)lwQj zQi9A@xQ3gwD&$jIYl7PWAY3ReWlei@$%kyQ3a`rKUvN_-?<*VFD>8^a8|5wKrzJ;U z=bpz}^IR&vu339&t;(<=%?UBB*WnAs4?-n(Lc!{vyUGNrj9OC62eU3l$;WXiB#?X# z6p~~&`6;btrzQopGUIJIjo;IlS|zmPNQK8Q6}G{3aLuzhp9m^v-J#963Z83j_Wpd) zh4{Ko5Lji@v;dNYHzNpxDr*@|*His@m)9}aavry;u~N9-vZSYi!Y8HY#qZf3#qzS& zK`29bM1>baY3QUA#LU&!jl^K#s2P2Z$s;wb^G9wes-Es_E74XH`QjvmHtL2y2htI9 zUikC1^bj6pnUy#>A1MtIs%DaI-sZ(*aR|`~V-rDdpDkLIQ6PJ(=o@{_u%&g!xU%HVu+@eV@&L&tJ zr!Dd&WhyV#j3Ys`4RBXO)qvR(orK~2u*~9yRoT|_xEEVn<>ZA|18cHcwy^EW#;-yp zFX}#T*}o{Dve>HXA`?kO_A;Oys)u-dkmAEKQSGnZY57_cW0~52sdumGXAxhRKd4l1 zxjW9$u<>1xrs~aIi>;S3MVwB?+tYVH9M=3*C2YXy-qs1 z#0_n)$br2sM$21tkEy48KbW1kDed>t%X?+M=)+pA>W`XAfFL+o$JX8=dMie}%2C>- zE!bCJCBF861jzL7|LI5t~p_@H&lX{&>vU2*Zz=wdtow{2Z z8r&XgIuTq4E(9NC69$w$zkGY&0a)H0JbZR>)BbH;{d7lZ-4%}9C;7=W0tTaR4oRKZ zGonsL&HP#Dio=_D(MoDFZx4q{C1PrqPxQW=N(SuB3KD??X<>X=t|~_pKc3QQj2|+a zHYq6%x9a3&l*jO>IxcgoBRZJq;qZ1$VkHLXcB@YzvKn8V52f~gVmN@!$+Ivhrea^b z50i9=X{^ApW$5T`)=ep7DNWx%q!-uzDbb_s?K3i3e1F*k>*?(#X4!(>i%;Mj0J~Yz z)dj&*%sY#Kzmd5!656n?$=r|xoLN3NzSUk3Ln-3F-~xQnxwzJHZ>gqVq8OTU@jWRr zXb$$mkQU(}+#tCwhG}+BnOddac=YcAq4EDwAWr^SAa49D5dL`Qy$1`tiLB;y4r&i_ z#EgY5Eub6D$r+(252mnn$vcKdFh1=(^*2!LP?Dea9YN?M4C7_&-%od}P zo%DWA=7gFNjjz^+87OBSosMUdwLJxbp(y5{cg;CJ3&bO}n>t0jlp^P(g;n;M4!q`k z{b}|4E4^u32J;V!&4QG0?M|ZtPTp;nzG7H--{R_+!mV0!b#7#^#LJ#AWrtL|&8?rb z@@6o5;pq7Nb695BQ?I*_#0Fi994*)(sf~$Q-m@dLymj@%hq{}#A3vZ-v|#a1J-MZj z;;I)DQM~Z;u~^s>%f-Q9vZ9h@sm8iPLrLmFCN#GRlNCp{rlbW|RjW<&M9|#Z2Yw<%wWf~*(-Px6^on%p)9vav zt)z5ni!*E7sN9i*}MEt$;^w9hqBy#j*XQXslt&%=_8hZLQ!tT}3Zw z0wc7et65lW#M{5xX;$LhRHRDKcwbDLJm6?OBUOkxV~^E!(XDj(p^iopD%iP84;E{4fODs{@XYh3gCd-l zT5y6LG$pomrgLYJu#$cQ)(eq+1NyKNv;uEIMa=+~FO9wib4E00BBSr&*Vvm;LnES& zOb(qmVpo}bJ>Bo2#MaNILbj>}^;h@26b5IxdhWkb=bK<3BwyEoulU%15MRR`=I9LJzME?UYUaC=K@<@mw_5r;t~i zBX5q2o(%I^o=e=GaZG7?5oP@TnXvq0@SV6$fAmaYr(>hTqjlug`5kR$#aoBeg*sn* zhD)t4Z$Du1z4Vczm#tl2_+xv>BsNk$R!{eRm!AE-MK=+l)>mF$65~*u)%31&l%k|6 z(JNU3zpql%7XoKzrez$LrXTVMtV$NzVujP^u#M5Lj%EgGFUaDIURaS)Mp`iGtKLEU zRUj6hK-hk%J;Dd+ph7V6+efiMxRB&6WVrZU?$}`JRbp&v9WEflUNGmAg{lowNn92n zTu2FGdSB=52{Rbp4l{#Kg9)S0qhE$GzQsWh0i#V|H^V#RKo#glqy>PHvK`h%jLMaK zq`vQ|S}jWyVA@T8-z5U;e0hBU3}a=<)ediD;0q|}8SZ97nJdZ21Ox>5bP-0rEn}Qh zBZ48IjFG{F3}ix#eGzknz+*KjVghzJ5B5z8TZh%{z*89rLOe=bsY^emq@U~TYg`f} z6nSbx!18`Hs?b6E%!epAQ0gWjeNe%?;Css2&KQ`d$0mLG0_h3247P@ZOCvxTl_ zy@qllja_SaxgsfFo4q^MCd+%D4;h>ueReLF^lHAWI``bIHqYR_G2hsEwB}z)PSAG$ zXlC_GhW3j#zx|_wwUgK<=SM=>p5L(g<{c*~y%zn!JR8G7@MNK-5E6blHUCYFPukum zsY3MOi@6PO-$9qKYN@hDBMQSW*AaiPX{+Y1T`0%Z(l5Lwj^F>++Y^w(^Xz&Tn%9C-#hQ)+ zv0>#%i3i=jb{*;PZw~}gb820?j%0K;Q`JAxMciD$j=r|MMPuehDnoEbCfmj~=4q8E zm<&^o<7_{#gUeh3F^>O$mP5|lL@YZ|l?%rl6L|>eHP7w7mhp+V zRMzd)9Dr^@J^?hW;%S`{EYT96J~oJOy+}# zxBSRnoJVCPV-j+;fyG=v{eqG$AvLtbEbqCMk8KI<TY+Z$IN!G_ zYD3rUDz|?xHGi!$=m2*B4FAj;1p_D>`L53g-gjzYb(3xYIV071vAT%}iiY-bfNf8^ z_vi3Jm?=C;*8||ur*n(oU+xWKQNE8$8Bjcox<_30gq@E?DVouE?4FOBwQ`^k)%m<`zf)R<}=dDK#1O6_Dq8%o?~IjErlGZ04A5`UaDN3#?A{M zKQDm|DU*Nzb}bs*$+j~19cb2&$0LBgE-Q;7q@sHLJ&xh=7k$cAivid$kwd0%_3}r> z!WvO!Uk`jzo4S>p{ z*J}xz^tVV<#ge7QlE8Ot)sk)M*~ulWn<5G`-H!Vvo@*uRu4g?lif9-2KW5)Za`7Os z0ffmyG=NFaKUPVE`!p{FkSH{>zJEk%YvT12*c5GGSLdWF-C7O!7c-jXZac~BT&8PBh?sdUAcKPKvVjbDghq$Vj&@}S#A z)wQ?=YOBoiSW~cH3JCxaK<>X0{wxOqiu##uMJvOYa9tQ#S?!~RQjjD(Z zw%+wFl|!Z`oX$LB`(_XyFU@l?L&m&c6z?umXH7ZT^>zq{lbp}6da(CZH5KF*xd0I= z7z}J0EI)4AX>~5eWTEw2UVM#`DwoxrTlPL&!-HZcOCit~t{5({78+P4)EYNXJ?eb> zSk9B89zYP0M0d5IrXB&a1Uy(AB4K3BN~--(1B1bW*cO#o7_NNVv~SXacRDOfB|u#m zjuMy>Hi)KKVoG`%J*vt0fcWz%nMJMW>PqNs?8Ox`F&7!82a+nb%(2Wy-j+#vD(vyG zsy%F$DHh4>NY474Qmb?amT@(4r*!u9WNm2s$aDDLnFx?yOa!(6+tn@#6aQ}m#b3D7 zAS-AdDN1g4;0N!-{*U8sj-!2bNVtE&ue84!jr%X&n{Vls$>O~r>S7NnireFD< zRId$^{FpZ6(n1&v<=iChP(bdq^c7#ktg(+6ZG5VFnk1ui_Gw}GrK>sJ56^m?)~28^ zx`DLEwo85AVyr|mc5Od6>a%9Y2yw2x569D3@Vi~J-rpnYdUc~MY15@XslU8xU=^|b zymx`vt)=x3E2%KNhO5gfBsb<_fFZ5;zuYIw+y+J9qz}=-AVMKRM?7rVTo2Op6Li>a zzYsc%(kh%@noR3wQ74UnC5tn~v&=`YPxR$E)A1Lsgu^c+Zve$wZAL%g<0Mv^ZKp~sCx)(QPR+3}ZbkJf{ z$T-wl(EYmnC{Qli+;+BMun_A`%FT^kZiMDghHv|bwtUFuIyH+=?lO!S zA=o`6P1fcFH?`>W`ouZt9zJfrpKSc ztxf$sKBS5&Ag!Y%%51+xGy%$}R`i0Ox}eCRshizo9(9*jZ`UPC9C~?5+|7GovB#;Q z1AVYrgKQ)#?LXTP1AJ0e0cqp38;DZ zy$GsF(gVHsR-4pi(G)xFo_C)Z@5N5Fsz0OSsDAb$&h+|mF>*mxkkadZjG*pG-)qz* zDp>WkpUqMm@6|cyZPouvJ`PJ*z6lrj-K2r9?=u3x_@~Ae6GpS=l+zK2iAD8an>0qr zU~G&)rHy}4`uh(o#LB^G4CajHH_5m-`f2A082ewIZ zNZDNz){x@!KV*|Dv_g-s1HV67s@FKlQa!pRtcCDmB_C+be$dZy{s1)$7b;@V9%=cyG<$)b{+joV-#eEx9W zmr$!uO_M^xaw)XnX2Cr227bdx08O$mVQAawtE>pFqQeD1eZi;YXwpJURiXeUkQ7mW zi`+^qej%wGpA?7~mk$L1ug1eM;$eJhblw&Sq)p#y5_r;;2_fh4x=_U0bc(@59%;Lp zCJ#+84>cR}8b-G>-Bd#aIAg6zlGJ-IJd-JUxCW4C0bA-#Z`v{-^n32Ki&{UXEXrk` z*Cf)Iw6i%q9O z;0nTb?8+M-+o)!B&qRzOtcU~P+^0AB{f`8UEi6nf?DhWTUxls1xMv^U2?btj0_{82ev`3`bHX_zt>2*;n`) zh0}FaU^$7IB_^=lx^jiT1&vsvFtcJoBBTVj3k<%nq@VPH1q$@Yum1GmG+N?0s>O5vw3XZiD&u_d?_m9l2X9#d=-Ch=fd=ipkVZN?b-ty-n1`JI(U zqUpy=vI?@WH^bbF7&bZLz29>b>d2(qmB;da-P*1q3}kaCZ6~v@@N}Z_wfMXKAO7=D zQT*9!{4WitbcpPK?KLQ?`@~te{vgwK8)NKF{fa+LxcXz^FX=Dyn@lUV6|GRmp_mD} zvc;p)WO#rf?s7>LHBmP8it!D{%ahnIAH(I^u-(jilqrQy{Q9K_8+f}b1AwaM=AT0% z$ir2ey)-^nt@L8OhPF0LWKRd-LI*^L&O*$(3lAmel&QET+XMU!Z>OdqREERN+fZ|j zwU1g$T&uLah9}m^d2ru)e3MulpXn?8e4n;4S4X4Tuu*+4>~wu~vV04=bDLi2lOsOn zRol^aa>l~c2QkB#>0F%!)@ioX%fTdviLGy}7)r0N#0H#hl8lS4x1hBt%}Glp9lr}s zCYTu9>|Kux-)Yx1nf6bf(qCj+oUOEv$E5JTmZjJA2kURkQnkglKuiKMgT6iTf3H7r zd6L0kQhc~P+M@S2;VlYFvh`0^w$Qx5o%p2|lo-a(`Bc;9{WUAw@=TFwI372f=NBtG z?B+8ME}!d1jz_*HF*43x?WN6!vJkehg|-GKJpNiC^b6!0_k4f?h>cMDg*_PO>-VZn z{kLY8GUu3F6fb*YXbG~E?riEbIxh7v!%~+`53|%Z-c@! zsIFKk^yv{Z^5>wV#i;4}Y`>dt4_=O<_289#LfM4zUwlK$R{D5xI~9 zms20f!ASHF?J&rVK#H+X(}5MFo|(1~4k3EOoI7N&#~6XBOOdv;q1j&GE*3^f6eJ-S zi5%b1fu9)kEypIjKGq1c5G5Ik0HZxrEdl_WOqSSpx=rGKjx!_i;t;c?1Oup+aTtLu z=D3SIq%(yV#ZvvI8wq$w!D>E>#RHvt{-pBr=0PZHN=E&~e^sB&0rW|v zOvu)?`-n~g(Z$2Wa&;8jt7|I%jzpgHD4pP7eWiXmb5_L zXoHQ)by#<18aZP$ZSFMeE|7i!;v;A_7h>BJI8`Y;UVmFBCYAgvc;|h~o0HvC6z&xX zLF*Hv%ngEcPp&7n--?N5ZYH+2-tSle?#;y5Mj@0x5Y<^P1n~8LLd)(_CL&OS7pysw zW75LtV7w!1DhW`0Z3dEnKh@CJbks9Prjmk*!3C3hBiH>@zF|R<^{&N|KYm_#t}U%c zR|HaxcnpAxfjl#QN&I}b%3HuzWZ3As6!q3cOU+V-%z|EAkthX% zFlXt^wLRUOWODGEGwT96(*f7P(Afig>1cOSMscV@7!h9pfN>+IeKP-ZdOxW;GSivP zq#dl}OiK8`AUWAGxxOn^3DFMuuvy%}gfc^)(SM&2eKv4rm5)CgQJ74ub!As7E=RU% zn}Z_44v*nTkE9}v?X>LzKciC%0E3of(ZMzKw=>Psk_YICva!?civySf;ib)I*wfs~ zG|p0BJlLvDwF&botMy|SN`~y0AZU28oC&P2uDxrU+p*ymV4d3)1MZOj%}`;+0aJrR zt=5EPwA2}SYOb2o@Akc*_D2)+Yo^L95i%bdPCcgw)Aul|1kd5H@u|aZIq{(&l4@0y zz8`s4&*7~k{LV5W5_-`f(LV}bco`Ho`4ly1a-~p$5cq5Ax`%O z>6?lpEwyYDD6vu(kg2nD3M=#V215_m8s+Q?i@~S{?Hw$eK##%|;WlGxoghuC|lJNH4#&$gLO-b+?d2k0GAAr1h5hB#rf$zh|c?pC4b3^*CYfkvb zHTV6m*ZllwH@n@p0Xa|nv)hn;_L~e%sm#zQi;@Ggkjh}+!}?RP)9rPz@QeGW(=^^! zmHP|OcxVa4R7#`rdRcHyqu`aE4lZTE}>$UlJ_zZl!6Os+qFveg1jPcS=UG z?Y=Dj9Dsry4Uaq__qxlCv-w1Soa}Lq#i;(%8`}=E9SMfoS`aukxhy z;;Mub^FgK;uJa)x$}mDz#9djmER)JAeW*lQSbHdWwa!!ooWBnq%@%s386x)Va8^_L zCIy3!*hg4$4 z`pVd%(_&*N3nZ`{f4OeE*yKXFOY%hZnCQic^kHsv@it8-HmS@PSH$)ePcGevl0`=C zh-Bes2hT;@A=D0h*H}pj=~*BUsuKW!!VtlL9$pTj+ML9y0geyREO)+d6?1y0e#x^O z&{50G07>pNwL{`z<$e5zI8geLC}1sOmbyEWlWdg(y`pu;w0(7+GC>0J3EL(ggkfJ% z-ARVy)QmHc0VEr=iWY}nAktUEu{d{}WKU8x`CotT95)Z6znXfElq=D%`1wv{z`F6z z^1Ex`T>4I}2xJ^etW{~pZV-oiU_0l{i&^4&jBSN6$u9si2d3kn1%yF>ILcMB)gq1; zD)qy4f0Cjie9cpq^ZBE(Do}G}0ps=^$7tfRkNBVZEw1n&nFduq1SKzRoSk@EEKpw1 z<4|Y13~vV7r5TqfVcWvT#j@rT_?SUbZ+GxpM|jPgGP_4iMFt((A{d_Y4`TE_+LKn% zb`(j6Y|;wzJ^evE^tB^EzoaBI2!Im1I3>bZLHc*m2PavH+?b#4G>TvI@aWq9QpHIk z$0whY0z-5#frsAZ3AnjHb#0JtPD$TmoU^MYg+jSz`vQTk;TKlc^nO5A$Fe0;kv=O13?#%-XcV8%tmOsN(V{D2q5gj_#Yh z(x2{i=$1~YyvZQkluP!3wT~xej8KPRso@7@OcavBA*xZf_c9G-_ ziwGE|X8}MISZ;O)4qD#*mNrt+Ck6ujJ#`WQVEMOdLh< z^;7U^?Q3_gjsMNgi<3>q>aXb^XU7QLAjyXxpARQ(f%AVE21ftk_#m0!@Lo4 z1G<^Sd6~917bnwn0(UuPgH5`J6ZTGdEb1&BAf$5;Vi4EiWg=*ajcEv z?Ki9gPb9)XAuKC4@JMLBsWM2`z!Y~v4xqn#%cUJ$CFjuvRDDuPlotKjZIF7A^7RzA zqty4ncqpl`fe%($*rE-VwD+X7=3U4v$=JT;5|HA)GbHm_&(*-#4a;!P*z~qqqIpmF zs1-%9|D|_o^ru|S%q#&0%}=6+8C0DAea}SIy(&yeHiEAG^d{by^ zT&~7xdpR}J1--+bThsCWbk*ggl5+&2)c6r;Atd{hP%eFLX`i~)GK$|v`9&t$1cuU! z6a7Kfl{s#s>UOM#zfW1;cf=1N(sw1!S#@_;Ub}jJynr?LFgf1u14YFc@-F0- zL`*l{=?Wcp)`yRlU!DsupskHY8=RkiXY$ zZPWyOr|#+*|N84NANWQ!*q?B%g)_u=2S{$5wtj+rV|GcM9G|@Nz3lTRnb0y8_ag$+ z5X!yieY(|?wi&mY11B8J4|3xQH`EXIZn)k5j{8XY$(ctrPr()uzv}44NWQUzidI=T z(=X2%qJW(6=~-w1{gebr0Rn?t4k7>)A%Ok>byNwnsqiR#`T-lkHvxbsIsl>AU?47k zbqGf@3=HGrlY|E#AQS+7=SVQ$k)>+D6wb4RtFDje0stc+o01uJ5D;==ipfL)L*d!V zb;G4{t#+yN2t-mvK=^SC0dl!m59NBKPWvU>5E)`f02Frwjw9e^4*k9_#$5voH~qjs zE*T%pRJIq)feS@|fJFO8-8YCz7)Y%t0$~=5fj%Ur293pu#%aYWMAfnsC}41cO?)&v zQvz>|i6r5vB;U_5iF#MK9e$t|{%|UPJGg|V>UBsg2LuZt5eQ?xJ9_(a1WEE^DGGL; zM=or^#Kd?ec`uSn7sE{sq0=Y9lo=e2f`zilM*#6G=)uLoD=)M`{U7@R-OLMhw;i=#!9dUQh-W3Ir(MWX0UH81#cOQpf||BT#Y52%4WiydOZq zkH)|XCJv+GTNT)5;$XP@Dv7d|So*PifZ=pdNFb8}T7@X&&UZ;{2-1oqH%EPkd0m6< zg6$aT)o3GOk>UYXK@bUZa|aM7QRJ|Z(FeXK$Ex*G7F1E5%6o~y!Ureljx~af5O7P; zgWei&Fe$SzGK-Ri=TI66Sh}Z|I~;j)QBCS@3VjM>bKsYlE;E|2(`ekRy(bPSkse@q zbBgh$3OppQJX#|%F{|tw@P(6;@Xg%;kKvlsx_&c(LET8TnpZh^l{T`9(_{Eu7HKVO7jnA#soVu^oZYNP?Cy)G_SONQ@ zG|g&*Za1MQx-b0P#v>760p@p2g0clA#Qs)0Kil0P1n-Y8+p$h-{wYp4H)8fy)?-~` z%~{@O-ZJP`$Q8U<(8tcaDS5HU?k^t)RiZ`Nd58t#vEO16RUXwEB zd$PreLT&Pt+jf%jB51Q3)t&gTYnPq5n@-20mn*IF^(`N2=BuZMSgA)yeVrtVX~YX= zk`yDe)J}k2fat{qBy0<$MCXgl#hDoj7813`5|>l4&mRwSGv2N@u%<{#wMr#NO(h@= z_|&C|wZgR~+zCl-Rfz%s>0ODB>&FX2Em|LR0cB_wHwZClP@m(y3t`hPMDnEpV#fdw zj4vAm=HP?sdRsb){`@V#nEEX?;s)52=7#@T8fKi|cVIk>uVZh^9)WC68;d@fH3yXU zHEp$vq^EsKd4lJD*RG@$^tj|DF|7{)DfAw?qFhCI@3Udsoj1%8XUt3GC3(kGxCqox zA{o5TK)33_k;A!77k8FT1l>lX6ZY~9J_)NS0OY-D|hz%h% zr9QURw#F^T@?M>{xvKoy-}%3-*=?J`vD(Z2*Z!_@U-qBgxBvj#zu)8k7+b-yG~?m_ zxrdLphznBE`y&=Qz72c+%`bFH@Cx|r9>-k9{5G2SKFeNLHj#S=<8p>Wv!ts4JNWat z^zH45S2ngj4^4KB^bA+W{_qgGg(mJ~eWO>|3vMN^=|jwa2xx9-W`GZ3Iv8*}oyXjB4R1?*5&%}2 zeLM`ryzwYFtUCQi){q22<($C@-^e3Ef*no14~7{Wsaobg?A>4R-@lG|Wx{wth>J{j z>fI_k-kWlz$~_@|+mq=FAXbxPGM%zosr<~XIs0l)w4vD?MQYuu2RFD-ar8=2;Jl7` zwK$J#JW*0y3@0MHj~-5u+$c3dD;PP1{7@LM+p>E>$JNUaWmmCT>$(|L!-LtT)|UEk zPQ(_sEv@DrOn%lMxp}abZj%sWt1vIbvFT<)qFZ!Iu){Muw-^i|=Bo zCoPrV9P15XLhK$6tS0Ntv6Z*D2Dd3^G!Rvm! zyLj{dRaMsX8&KBLhjO8Czm^{-B-U@URsTx{Tg~T0am@KobkbO4Fm+!#X|JpxaKEgn zD!@QEP6N*lmS~7j5grSah&xU8hMav^lDb#7K`+HgMzW&H(p1L~33-<~;(@VWuo4B4 z-^dFFFa)C|1j~|`(74Yn#ynJbUe8C%UT&bHlvpgtRWI*htp$Hiv%cQX1A`erOOvZe zLZ1lB`q@Y)y58?64AiZ+d643mUi{f-jUAxG@KkHd znh!bojm?+1NtZQ&vS)L%_#C*rp5>D|TgR5R1LCG%m43%sQR#1@EKt*`Df&{;J0|pq$e#fc4mw7d zZN@q9tg<)qcT*Kqzx+zh#fM^u(OSCp6tx8NUPswcn0ROv6V|Nr?a z;Za%q>8pe@W{ms`E{6U5nj3${r~E0XK^?=H3`$FwwU zXM|MBpC8g;3`Sv-L_XaD+;#Jr(mOsCs_89doNDEp9p>#YpUc2fD&D;uou2dv`@ zZT@YJFP`}xF1O{KIIgv%{Csz79hfg&6(>K-gg@C;H&sK)C**lvDP$>4e3=sRzu1Y_fu_l12=}roAmRF_bk|ke(en8wDD1XK_}^56Wz{n8|8a zV?!@qZ(i@D?G)4~nUYG1>&rBYA(U5ZTNT>;3fHSdpK>M0ewlq%`dJBS#asMdjiNV% zKkOX`?3%X3J9!2_7{0r&U0G}8K3#0v+H@-}WT`IVOMYjZ;#i$Ey<%eh%I8FQc}md~ zqKWkB3|tDgZji7B+CfFx^3;~)Wiz0>Y*!Im{ug$=Um*jcX5Y1#!)L9Q;a9eZC6kv! zcacSfK6%)oH1#+^w7nqn@ikx_EsInxo_D=jd`2BX_jl-8#z0!(2M-7b{Ptk19TG3s z-Sa?kxw}bT^r)WdB2XNrZw|NxGm^bYTrX6(Q!~iXK=VjOyGoY7_>o2>;U`Ly>Th!t z79+zQFvL}^+aSlQ*~bht5tH>6YBMxK?Uy#y0uT{byxTA}i^W)8z5NVQ1dps)tk^LV z$cwMTIXG*grDZNL?18G6z{kGt@fed7247X@YRv&#Ri}|zL+eO3rtf2Wc`B959fA$Q zA1TiAQJ-24FoFp0#Wv!3_iE7LyYvY|O7l7gviDsQa6%1%&*K;Co%*n1Et`hm28I zun9Jb=TQb&P!QRWDIDrL zU^+4={p^kmz!>ABm73*+=twsBE-4eWdu<355y+-hIPLRZuovM=`}Fqw(=gItYA3u~ zacIXD+T1wn3Bo?zNU5((2wWzo2>7%>pLloAKrZZT22h)~r;Ooc$LSY5jv1aMrpByk= zY?i3W^Q4JGsSs(>P=&xHoF@;Ytx_bfU`6!eCz|aG9wK(0fMP4u_bCVQ`&S(HN|!bx z2G9&5L}-S4XBwfIRk9AX(bb_1TPj{~s*Qy|KQ)KtrlCod-jstn z1kWb`IkVr>#m%Zmk<{QnFH}&w7f2!jyEtWF1va;)d_{Ui>KJ=rPYG0rV=3SKBMQLs1TcwDaj!?hW;YXfN zup8y%ak?WqplYATvUIMM$DQ7`g*zYgL~4&`>Eo17=DhMl^o&P7Uwt`I7HVlp>QSva zKUwPDe|EYI5%_w#a<7G3c-2!ii)<;;+OhU?=C|>NC04)rvyA{#mG7I%vBEdXP4d5f z*IIpk@qJU`)BPV`6#L)DnYEpp-u+s@nDkgJZ-o_o9k|*iK?R&e{&MuNreg5$);9F(F$WSOaUsE+=hBopDjT7o)_{<^JF`Y# z892^tnS5C_ae=otpzF+pE^l$3W#7va6==*QX=@DQ9MP6;Hza)gR%7a`!4e2^{61<& zB>(Ez;)du$JjRBNIh>(I(Dw5^36BIuvZas0aa>ln=tx#)gSCZYpD4$PBxWp+K>I)4 zd|=2xIVd%x)34%lkdT90$R=xhPoOC_u|#R@2Re6vj0FV{r;y{)Ka;nB{G&Yt0A>MN z|CAE_eYn!u>+1w#azFU@qD6fh>*WKJ!NlwK(0gRQ%M;{8I!*~#Z}+|RvpvNAw$N@Y zOF+f^vpv*^kx7A~zh8587)s6N1}1Wq6OkC9Uli&cI{j^~E+X$HRZ8Z!ou9>Tt8R(; zTYdAGu-i3WsHMHI?a!l+BtzZa-!g0#sxG``XhtBNAdr!tbCWX_3 zKMxd3N^K5(-(2}*S4U={#>cZZkmzLoHKev~H|M3PBuVt$6WalLKTfey!PD(hd}i&B zi@c7Z{Us^(s++=x*0oPb$F5{(+EbuIt$xxSzI3mbzwEs#GW;QWTQ7J8V_@RJ^ey-P zflC6M=OJN7iusGvQ~reXr!c2WNR9WNwx@7Gz#{{VNHa_1(D< z7=+>`;$B!}9p03u{GU|SLL=b%0hJpCewgZMnoef*l!f}>JMvv2$d(jlH=Fo%mRN@v z-LW*^NZWNO4-C5!HN7M6jf5ajBswNa#x7olRDQZBoxrrf7C@o9#V*VH)NaFaS;^p?0jh!umaAafFz~xA>IdaLcN5B0P4%9?VA+qf%ang=!TFMqS&=olbYoG1l2j(Om?UTLFGQ0&PUeF7(){dR5^Q(mKh z+VH-X4w`LONnX|2J=ZxI61kQmhM#>93Lcok)f3^TY3NNW>4w+iMJ*eDXPSSJ za<@_eYh?A#Z>eV&Lw(Q9xID1~Aa^z>gU*-mcmtNOtWE>9_kt1B7N!qp_dvwWix9CV zD9Yja#YpNc^i7spOgvwhBD)P!TUpdAo4D>oEcLTmeVK{(X>z2*4Rz`WaF*K8ea?ym@VZz5lNT)60hwvgfq}4 zlr5AHjdN#FSMvx*(s-^PJD;GAlfaHC^n*3T!BxltSQ*u*$q5m z_|2~W)BcR>KNE5GiWe6;!*EIeDW{cS6kJ>B8JP@~XKEE>ul)w_iDL|)u zw?eDsiN+82)P`kFFs9j{he_Ao@K~d(5Z;dSu4ro<{X!9kdk$)az~Tm*iOGQT+(|__ajk3p&U3AEovR)`-Gj}s902bW|33aD`u@WE^J>R3>?-%Z zjBnd!wi;UkM%xkr42tO6U#_ga{<=qGT?i+r7L=*-rNZ8tjKp}2v7Nbr?(=O zebV+CE*Zd%m*wlpw6hA|M)e^;^C&UQ^u3N-kd2g5J{o!08LwYLFI~{tV zjlT@!?-g1GrY#n4C@V)3c5z!KxP;10u*)8(S357i+!*M8v97^_m%exBNb1+V_xewVbt5Xnm{7b|d_iS*# zQr4VayBz+v8rsQtlY4W&=6_iW|2dJ)dH6e6$HUJLu8q&Zqe!jkPvI6uWHAfSXryCR5DCaCGokBn(@^(ma7DsvyCqam~@WGj#$nt3)Z+eZ=i&2SduAkEd?%qSx)=gSr)QxyXkKHA8 z?HZ=-SiMH6U7GIAx-eN=SAsEJ=qfxG-@MPbu=Bq^xWo)>At;h7ktf?+Yg$CU#QfL+ zx!lSyd5>qx5f^YJl#`+319F@wM+{3djk*6Idw*49p+ z90?hFYN)`$r&(^78{;la_vO>xBVKK*UmtOl58fC%p}F)*IztZzFf4BfO!mKNS0l;k6rVqZS{YSFh;lkm?o zpnu>I4WRSS4(`W)3y(a&R{w?5{pp^GwX*(y>7JdJbd=InKDiN2;GF)iuH?_}zCGAP zbp+n}468O(srq37k^d)PGXNc+3Lx^I@Ep#K4vazm?fs$h-vN_es}^gS{dG=)e0;el zgx6s3qw@33{&XaTd2#QW!ruWi--(M5S&n8a#SeCij<0CNtK_K^*gt|-mgyELIF;6y zb!KRyF1?iP)i6kA?ezT}@G3&e)udbU$6CmO>4gP`_O}LCqxUW*DaER@F66Dit83%d zB#Tv27-I_pGoego=v~`~`XK|pReoMC4iY8N+TFCkuX0aTS)}*+dHF1!sJCp*c6wFb zy)-A~@0_cDJ9r$L@!kjbl(rn{uI=7!!nfsW z_3coT{rUb+%*P$jtgPh|Jx)?I;`IHi5ll&TYmsb!7KR#j)}r}(FY)>KG0;}u3vLy!8{$1_ zS`TDh-C2i8@r13vzh@HFjv*xeJZ-ni&TbutIksbE^xS4(Be96tCno9637t9F&_0#f z-s*_~EY-GfAj!l@W7o#Uv{!laqZ57V2CoxEK7Hn6cAKIEumO7Uv$sn=Y%yBXpRMA| z`9-&KrL6~w;u0*bvQxEe=4awoJyr6Hw1T2wCHWdU31Lx%QFg`Iy%-dwWpyU(RB{6P1gru7y;~SGel;@kvR^xVg13#2U43QV~SO)PZ7B3?dq+Ua1bgpn(g~z-u=1y_o}Qk zqK|)sB(-7SQm3At9Zjmc-WO#^RQ7dE-xtW!Qnvo8aTdm93T~`&`jz~R6wA@)m&TXa zHr5rBek3$A`wsmgQti3T@Y%qMO`(+#W@}awzbRBU`Vbw@*A$_2X{amSU&uYu-m<+U zU5m0dPMY(%pCw!gKy%k)h!(LBoO>k^DzE4N)jwq&M&PW32h59PS$}UDt7mCKLcngV z7AZ=yY4>x5Su;r*BD$H(@o1IEr1Ip9!U*FEyp_g5p30K$c1mpe(W7)LFMV^WWLw#r z~!r`j~_|9O)owDVG!aWw4zS!Kp!IcG81Omi?&?B@dJuUa^Tj zT@qjgMFP=fze5G-U#{VnVEAl=07+@hq0K}C>UR|N%F%&6MZI^E$WNbxm_=gzIUTHScYMDz9lP-UU0#Q2m z#^7$}pKWJwS?&vu$?#G>29K-A{wBusY)%MnHT?x?LgdbhF|+AU`4`jq+}qi&&2)&T zhYUliUh=k|5z3Ho*}{6-27bzg_UvnTz8Zd>AaG5-3`shi zADv2R?Z*9_{*e#?m}hJ$d?obkX+PB(E_XQ>W;-7ID&lqXdWi~m20t7h%M7JJ1P&%D zhf5Z2r&?u4(!o=rGX?sR&_v}p$RlJh9s^prtN_THe&DHh$Iucp1OS8=f4WfHfp#|_ z(E7++_I{iI6}zs=gICQQc|RFuq#0!V-M2@Q1CHXks-cS(tsxtbgGzFpOJl)iK_sGt z2h)V2>#zj_E&7@{PfOS?I}uVNgt2vJGC&Tl+$)DZx#jXfIs8|Q@;tV&IFW^%Rw24( zBqj2p2wo<;`5e0U=S(s($J%^^=uG<}%6m5vOe*NijP${!8}(Xo84E+zxkRPTq20zM z|7^lOZv0yC#|iZMG#UPmut94Dx9e{x_uE-Jq~tX zTEiys?^JxnMC{ryfpuZ`MbP*pSonoU0Vjp3@Hf#B-5Z8{8u&Iga^!Q2MPSx+YhkmA z*hubHKO=zq3#AnYG^!nrmFmF}{XfV>?7U=(%6&KOH5-z|5n{4D$x}WTi@9tGgF*WV ztk~mr#E|MoP3tqT4zw-V_*P}MtKl$J79ISrjxY~({XahYf7NwHOkApNyZ?<-o|(l1 zc#RwWn(|>Sqqpk7Yyla-qt!okJ^UVY_so~?k2q~jZBIa2r{rNz*w^!c;xJ=@3geOK`GsD4D6^u9&jR~y+J?_5wZU~N^B3s2 zz`VNBRBMAI9Q(5IRevXH%@lts|6y08EcBAIYX8SkP_*Se9PyKsXpo*{h#RwAauo-o3`u)c(Q=T9&So@;VN+t~TZ8^h>R=)Yd6d^E1D<*) zJ=QC#Mi3n~2JK61ZHDJX?~8NQ<*HbYPz(!1#@Nk^ zClNy=Co?pcyh}awCeDxYoli{5`P6S;6r>_n{0a)tB!HHe5G1gcnbm0L*#HMX_`lf9 zN3!Qz`|n!vmFVj16>$FK(4?ym;Mg%r37+SQZ_DuyHlHByD=Pa=j*q z*q#ZMxZhp;bv9#A4+XwH(^JJl!6uWN91?RsGAXb@Rn7bka|W0i@|kr4Kqz5Y1`b!0 z*kjeZeF@LYo{;H9rwo%xgu@vWhID!{Z*M0mzMbtY$8-NUo5J*nm`(4r8ja^m zUnd3_@*eNU`&R_PkYvH&%tzk$;dG!znZC9%OH*FmOKUdrDyxY7qZV7l>ruU|4{2?I z)wdzBJ#X5rR#l}K%wWi`?>1QR@YPb!MkUeq}U zD?a3U@kjmrUQYaj82@VZOa}l1P5_$!+3H#8WEl|s$FZ||T_K2!{htZTa$_4~t@wfC zZ|y@2iRs>?&mZ>Ye7NBqv)z;Zz}}*M>}Y0mNbo;cbPN~CR{T!bDRYe?{r6tK*&A1X z6_0tE@sQWvP_g(b&4Wb;qL%syo9%l1({jh-v7e}FO;!CL#cs*#ZJVmsLn%ZDLopw3 ztPdsQA8AvK({7Dscbwi}G}oCAYoPJ#IyMWM4}G!}#<`lR{r;`#)&A0OOXJZp{4FML z`RgY4e0o7&$&m$4%aA&$U?&a#=B_oMv)*!mj;}-TSjDURz6xV;$oEmq8inFFpRP~Z zv%>YVOm5Dus#cGtN1Eh3Za70ein)D>xVX_4u`X&jOi{@EuC0)A}a#sC&l&_YCk(1qyUeX^Hh1nJ{N zGJYVxoX8AXvj3U;I^+br0l2d=vwSZ=gDy|<&WEMISK$Cr3t1qP+jELqH`x&?S4#gjR>7&KGy&*571q?z$j)hPqQM5oTL8`2+cWvQV z0B|CZj$fL8p5<*6Oqq+ljd8h=Gjtvuyl~EF&r(!=YWTgT<<#_7`_+(SBjQbrf=I{0H%Vs0u8dTe*hZ!f;@ogU=Qfzo1xe54azUH>CZVEcZP+&p?z2H8_ehC$ zu4Y<6ivB#>?66%x5ti*pNb@P3YJq<32LTZevzd^NPdJlTiMDgqxSlw8EwGB|MmJU* zk#r(+twwizlL8JsI>kf{otNF|jmO_=m_~)v28Ao70^7llTSawCSBFyCMlv|-YNdxT z;z{b0Q8ag%^q4OekeOA9xJxmUoaLju%0^RPQ64T_C2!ZgJwif8c_&h}25&n2IERK6 zwoN7t^`t^%El)#_WOA!0-E!g%%J5uea z*ZbZeRAe;*d3pkHnX*9PiZs4l4upi@-9QegzwkT;)A{CP7ZA8$A&c%Df_~JC;R7n} z!)>Eb(u!gM($YB=yi4qiIiDGNMSrHZ!OLA|)=OQ8*jGU3NyvHBC(YSM?n`nNucQM7 z&<_Jb4cX*TU{~>waD(EH_Ut^ft6khPpCPPmknoCyY+`ESP)8b`!}{DdQWf9zuF#+qtm-r-d=2=nlI_DBuBv^+sT#fr`g;s zcfa;Aaqn+1v8TjuFZ-tM?yfGf^fq?v5w|yr(kV=iP*k-QWt>?$R549{Lofv>$qmCk`OCkd2lOS$QO8-iVLRm#pvudSE1yK^t1%E56a0&t1oU^| zDEq(7F8@(DQ0^sfm>ir;3ejZNG-qf6MN_{EhwmG7B3noAPkgo} zFMjQhKCg9KPrFe}{xpySJlMo`8`32gwpr+>#!9wWY!2J!NVx9%8vXe9e;#pTb0A$3 z@;KVVr>S00+PmXouTMcadyBO=o{g}Kfc z)42(dNm*KtW(6B}mz$CFgy?U`u+PWlQwo{cjp=Jzoaf7I80#*`>jj*RrkJbDBpPS? zpE3HGHLzR7x&cvspr3JoSWe}YZ|`vPJ}wTCmiGj9OqXIi z9oreT0HN$Sg-z8RE&hzEN!J&pW}W!U$QUT^s<%g!Ku#HI&1WM+#nfGAXvOaPy>J4^ zhltL)gKt1vBl%%JdMm#)F=3llGCYRGW_M20C)K<2@MLrg?#m;=xW;l*PMeXomag5A zDUN4Q#WO{@M!B0|33tczYO_j|+C+Y_r(iN}h<5`=p zy*I6%BJx>RKFgJ+0x?ALt7t{6#}+L1q<@g)5tx=#Mu|hAi+mSG7N;MluqLBJzBQR- z5Vosh$S#CF7H?4~O74xVwP6Es)c_Z!d1&<5WO(-q+y1qz0R?gV!^miDE0{7;=voJl zYfLF4FG0woDY;Y^{nk*s6ilWxUO} zWSvz=pw?PTb26P5tFOjM?5nb^vCf!)zjFP$fjbn4Dqyb6w7U*UOiOjPt(^p%Q!RU6 z8ees@tde*2iH=ziFX8+n$dNYmP`nYGaMTW19<1?~M#J7I??5>pB0!UHha%dIl6b0|-1bgHNNi7W@EKX>d?aS|u;qenqSH2aMvDd;3FCFK$@fqR5b`%NNa6 zIA8-i?VA@!>P)xXCgO(r?*0)!BHbYIZaY`v-|lj~&C!e{W>-_UBex{AxwPc*)xAIH zwxYhx@Xd+K5;!;r{=mYXMiom=auj~*L=fMNmd4){A0xoMH(-n7f5vimcyi7VtL&fj zua1Wg8<6aOZ%+MtI`wCB3g_Q$PW?kl&VEYMX#@;t%lk77ll*&g>d!DtMT_lW7^dNo zhRA#{qoc5L)7+^<3Ji#3R?;b(t5v$C|I17aWhysAwQ*-g%En|z@+9~uz0wni`7?D! zZuv`?>TTSEgK6vw^Z?1PGsmf4L}(qCUQ7@#4#-(vLu<<6 z!sFZf)`VKB`)eP2$H-#wO+cqe?)VEvOZ1)=>ft{+rKoqr@9limj;qT~2BUwo4> zy}73rzY3xJwmdPz?1rdH6|QLaC3$bs=3o086b2ux&CoX3faP@jFNm9n2+@#cj7)qY zbv_2+$s{g2#A(@?hmi3I=1A)jhb+A%4-=UTV@N6{jsR-vBt>#D6j(&_^|GzM!=QGD zy%*T(CsZWg-C2LCe#4fpCXQ{-q{4|If{;?C+0B=KmqRuKaL`fSOnhonvYEsuc$FLX zBFsn23pi&}m}2YMpODPn5V@6(htA9zuij;!l!gU)G?DB)u3VG>o%TZT28=5gf3yaP zB`5fK>Fng0afp$DpOjqg6vV1r6J_%7q^>>G&?z--gMUrNhc!OVzb9G$B^m$y8s9&XG4hnU(+}vfn*S#mp9KC(GDiKCjJgY{ z$A6p=8JyfSi>Ct3DsVhX>&xc~EU?2+RrwC)8r`A;j8$ix7h1giu75SYb~SGC0b-I2 z)E_`i&GBcbzc#rK)Cb&SoNYJNY=(!TeJ$XG)ou-C%ThA&yB)Sw6*0ao;ApPj{~RUt z_I7io-gzd0v22U-XM^jQ3~}uu)@7r6Q~uQHlZrW$qjki(4bNrr!Rq)}t);aaUiqtm z`7gHj8FzDT^J{Aptky4Kp4;hr>zvHI9d~nU39R3iaTB==?*42AVN$!nf>2m?;*o%l zidWuBSEHF*Uy&y!NfLh6!}qoH6v>gJqBvVNW8~0Y^E|+Y66$aTh-imOh0ZR%Gaaz8 ze9zx(XKu|Z#AN9L)CT8$K+%cFkyaE+V3MU!@SySHz%HSMT^l4U2kQAMZJ0lcW6led z2hB{$O3I`Xq#TvV<0nFJP?)SmIU;frs5f_ILb28Z+sFh(h^v>5GT0-WlY#+8%V1<>(uDjE~=d^{R&2B>{uDB!`}Q=iKI~OM3@Vj zi?3)9c#IX1MI|ie0$S6R-~nkkY@PIRUcHWKW0-La>(_ARz#~SJ$aaF?Q=9-s@ScU5 zmts#Wxc#PZ7EXp$7Siq@8$BvyA_jfI281l(dxFY_S(%q@=W%K^Thy43#Wt3HO$CtsFH}=;&%d$Oy9V7y7h5S&^muL>IJxv z0q$X5VWIh()HQz6LnsyiTlQl>^eOh8RY;B>Q;zun9FNbtkp^uvOrCRi0{Ny7>BR)P2)kyuKp$h@IAbW# z@tw6;abe*(^OKY)r62{~Fkk|AP(f9A!A#kFPvb3C%XoqbNt@tUE^M*|3kYOBu>w`Y zW>jL7#CSjb9CT!5QYy5#2n{4;@VIm$Xvb-By5=Dn2g_<3s3?p6g&>UCRbC529oy39((8FbwP0Kg9Nc^Kn zDMkN?0^G;UD62wv}NWbY0o@{geL6D%eP0*ecd@2K4fad(8C zJ>q@8Yj({(kv4x;C^B54@p><=Nj<7a+E^Jcn1T#I)m^MQN3&gJNBS9(P#pE8NM;yC z=NHy4DpEdvQB>Y!>1iK|p|PC`CbDiNa&ehDGpiAv>{L;jptOkLJN;!bdSaelI^zPz zH-q!kx2utrzVC4jRcqx(-5ynqJp!uui&06ilT`1Cc0MED*X|$Ot`51~eVrLk0yoqZ zSGi4jAt{(|O%U8$rp!@drXE6_zvas!3i+AYVH>2b7{!kI- zYv$tTE2CLKXiOtRO(zP?Z$vr#F;B}@+kqhQrLR5~i*2#-*FUVuEoP)Tn%7<|kk;*w zR9Ui0d>54ZXkAFkGV*=3Q#l@3~dt=u>hISS~m4_N1uO#c1r{yqOhVzN?~w-bIV=K;yXGZKiXHIY!nHZ zeYZ;T6+f`OorMEpFuOk*)E44xCf6}*u1M+9#1E)_ssP6SB<9&t)yXv4OWIF##UPK8 zcNO0mVbFh$;BMS;?Bm`}&8fDf*s9|rW~QVfb)G-;=8+-1&g(284?(Gqwi6*5WvHnG7=mpe27iK>hkhXUT3hG2+r0Qdu}3=bYM+N#U1!`6#KzW~L5b+bkWfC0Gwnaw}^YHMEtfq~eU zo=2nH?@9iN*;JQ)S??#6>A4)fA@b%e_)AU4u9f$d)Kw!z^aQlSo%u&icebh`{(7NR z6MK4v&e>1@|tC=FBXvgsOgYh7pooCLLw=73w7HpU7=DEV;(_2?V^}ApL#A407{+XDZSqqu=5wS`h2H41tSWUKDGPBJ6|72^%bi zGN_PIK@LZo2?oG9%QOX^G+IT8M4d4hJtqlaQV|S3(@WH{5wVN~9yc+`WQE#qKyE(w zZzTN^++~4acqxyk5Ds6i%iD6Ck;dpP+1q4#&D+7aQAATVA%cS?2wo(Tgq`;o=W~cW zX1Lu48r0@ZXxB|sV zH<9DutJ)zJf&I}qA&MDE5YLZ-_h~w%J)7J-@t1>h+`>ha_ zH_Sm7e5KP2H*Ts#RKYa%$N}e>* zwCIH*;eLF%4`>Cxf!2A%_`WX6aDcf%wJz6Hcl{g>e$3#}k2oCc_NKSVHVVZ0Y$aKK zs>27X1h(=TXlBTxjM!?u8lEDh3H;f}TbF-eOumiAR%<&`Kz9A+requ zlxRr^svwGO8YX4wrY&5vk1VA9!GIjgQqDyyG_Ms|@4`%)Sgc46PXPiH8V3;Z144Cf)!yoro67D8 z)O|3Q(?R>(M_y8e0^-X-KGHOxbZ^X8P=<#Z^h-^lkgul3(Ud{P zrNm|^ufQt~Z^M{mV?7tM)ZuW)-Cs>MA;!-`=ZHKs$gjz+FN~0gHZGRaWjoQAHCMo6 zs3u?J@Fb_8J?9G*$1r5^L+@OB^w{=u9g6@d|Iofafx(#EreL4)2v20~#~rmIv6+Dp zujX3LexD+VZFbtS^6_uD#Yb$6Z?AD%3=c) z82DYR(aku_dbm~p&6@nWuo0?irrKE5j9(}-htO)%A*Im3vbu4EqP1#87el)muetyhqmGQW!jI_r140~m(9^T(qTd97i z;{q5fLtDDPJqKM@)Vc(804m$sCO?;;`@QclTAel>eOi}u*!qL=0$^fIW|P9+2vcv& zzvzi-u(hnptRk^8@OQ|Cav8gvL4Gvu8>3Q0%knAi zJ(3fvo6|_r@}|Yu3Tp#cm-JttI&PJfW0aNU2KuLbFRB9VI~BueDD5hd7n!vy>qrE? zH;f5!a%UMWzKm%?N@-4%np1F0X`BymWYx)SlSTba;K$_a+uL{dvYHRRxqLbtd~mNJ zdL!3~Ms}>~p7~^-=)>_aaaBQwzzU!XOnD3^C=uPSPF=Wi-0!wp3#mp@_*lN{nY}ka zt}gZ-tBUATGfb5RqOZ1vj&x@inv$tEqZCgU_}cIA?x^B6%}<^ zhH?V!JC%Umz6!QLiw>rmTc-lp!=nq03}Pf}p{VXTcc^uaGmV?i1w%rY%7QnRHm*^4 z&3+%j)^BFAWIZHQBs1HQBojXVkiuwcOa8$!0F1Q=5_IV%bYT1n|Au?3CbUf_LdH|-d7lSO@$R`#TXPVqYHEc5*XQN97e4EJ_!W!&i*QbImi(}`kBfG+$gXHD z+Iwj9C)e$}%wT1oSELk8k{?@hoA+FL97OM~zc(J;{W|XB{o_Rw_^fNu)azgQ0y2OW z@C6|HUqB2ZtJ4(i9($4vwho9P`s<%n4xxDNa?^jB|M1VYYqS4Z2~b}$YKpfR`Y8Pt zLvjG$YAo^L?u#L!NFka7!k?i+Y&RIFZ(uWQ_#4E;i;Lr&h1Zwsm;7umHCDynudHz$ zq9Gn?__P@E@;-LB?DhA>4(RCVu&SoR$`{k2rRQH8+1L7vH+!k^E|#snz#elt-^A8# zjjD;_g4D(AW=CF%`|=DM7Ag+tt-f?t|7GJaY2&!hbLDG!_$}o%X1KTEvE6(7aL(}A zmfF1_vSd}#xt8m*eQ!O5;SX&soonQ=zW20G+gq=G>`auIj<&b`x;S1R%AytUeI8AG z5M5L`$8d*qkA7wfy-D0Mg`iP*$OL5_c&upPAbTT%v4X?KxG!2?;)ao(GiO^8GuVu2 zH1bKy5vytAtp=flMa+h3P|>Zw!==hu4dJobk^?@baSMrgXw!!yYo)B4lU=i|13_H& zP^2S-CzI6$-*MhM1qr%5BwVspKr}G%5wlq(AAKHB{>I{I%#-W!wG?RE^PqP}3k;eu zQUs#VShx9q^xy$CKF|bzkOCxL-LQKLNX+azo6Uc>h8xds=oS+ouW~(QFYeBsC$By@ z8pRPIQ8&LQ&UVL(oLTAx5rO~RU~f{{w=KTIFnHOBV^vm_{WyPa&3~g590HC)Ic& zcY*!bgn-U99&ANPoHJVJosKhE+E6)}0fnN|P+^gM3z0pK`Wk(aJl!0?*s}2XvB64L zN)T$ixYto#wL?{N3d}Y+c zY|;*2YItAYqh}fM_^>Wn>4%0~50JHXgr8Syd+f`rdn_C-yec;2w`iD4V)Wm~@W!IE zo=Vsg3_}4(pFX78?~JvLvC_%i5rN=Z&=gW7KFGp^2}E=C-1OBOY&&zAH1foL$)7}p zY77~I35h0*D&YB0HVT1pH(hK17!UeJJIR)%&@K;RhMpitrwl;0c$O*JP2w7b@0s_W zu6|=h0heaiN?)2no>cy|zioazDURSok`+XoK1q>gIyD8j`4g7%oJ}OJ`fh=jSe~V#O@MPIfSf%1%xg$#sz-PgC_yp1tREZayrkFWD0BN$*T@ zqM>iWa~t-MNptSF`4U&wac=~sEkXBIiGR>NOIac)RdcHXMN60VSt~~fp!`MR4t1e@ z^Ywz_#R5N^`$C|s{2c;0J*j+TcyL#`^_jFBP)=a1tS%pY=vyhq;#$aWyqkT9k4DzX z)v$6cpC6((?|?8z6@h(f-lcGd6SVY}k+>3ZOc(MxQ=7qod)NC%NI0!OFw>Bq3i3kr z(%cb}&yW4!_)odYd5~~4UsP+8ehEh?4G`qGpkm*=jA=%b=&Xuj&BMxiwlo#JuLW$K z2x)GVP<{LbjqA*DL`Y}oqhjTgP$EJBR!Ml@-V-xul9~}#_MqMfs=zZDKdxs}Xc7qB z8KZ)7;<5`2?fQjZVhDzBq6b^Jk;f-Gp0gUmX_C->Tq3vV0q^Zwk5PQ2k5080N_0Kx zku7r0P`neB9s)JIPPxrktq-v~+3DJ)r4;9Bmb>=&eq+n!jn13pM7yP&S=HO0e%V3- zn#Dc-&zTh+U?1=wC?`%J_6?dleNPlBGsXWPtVjgSKM)puPcqQ#Z(%eBZ{e9RB40X- z-P5;K7*;HuClb#gU;Lc$Yao><|4q;1_3jV%H6%pWu>IZwvoAE0yXQFrr4UO8=17u~ z*+${eC-2#-oMf|I9*qn)ljKVUwj;Wzo#MHc>q!p?K|Bit z4-)tNl4xTw&AF^8 zFgP5I23a-?v4c9Rllq9Dne1$nS{$4eQ0XQctrTg3)J>2$l3 z7N62icWcqX znal|og7Q=4OcI@&U%V@QFQ|#4Wp@t?MG^g36eQ(t5k(dE?n(D^zMzy1EB;9&w)+Gmal+23q|KZKIfN2iY#H{X3o09PUDHEXX^b;rQGSIr2JF7YKoV&-dg z*#LP(oxU(3mPb7oFZcPR_P2K)Q~I_Qo+>|YCOmab{f6~stYVyhPTNtQ1(Vx19(ZYq z2;U0LgQ>JH^y;*d9ywEt-b-98W50J|?w4ofY+Qh{^6;+$GP9{i{5h6{v>0)@y&)gO z+orre&VZmBMyh{?>W{@D=@m_?vw)Li!aFXQV(`C*>YDA|le|Niervhn?(pMoDL1)Y zH0u$fAlUtf210DSKsf${lx7d-Tg?2{y{}o{Z>IB}O5FfK?_7Vs*5B2Pj6aq-EMM3y ze*v6-p*(>fgCxvCfSkD6z4~s!rr%6DLsmtz!5vlMXN9&@L{rSrbV9nRC9R;*J@5%v6JW}n3=x9fV8^g-QkoF*Db>sb@r67qpJzIR zDDbzhdv!=Bs-oEB@>w|BU1WC=lxl{sB;a!*Q|3}q_Vy(0xW~|UVJ=J-CyyZ2IAh;{ z*Km~U@6MlEgq71XvMLXxX`QgSl`(z)r?32v=3E#CHq#x2wP^L}1GFld}!9zs8P?h}}N^=6C@ z_CR926N5Xi52X+ak$DTb8+B(qj;2wEfwu=8o_|y2yvouUuTn0$3vc3gKvcgPQaE^h zY8Z6qfjNpb@3ykqe?*YeJX8N;(4U(%>u&mBO!6tD21Jxt>bt615>`TBhd42D@GE#l^}4OhxEf?YEmB@g%kp#!!c5=&@blR z`hIvSm$4pT3R@%JhDb0&Fmc#aW;mA%%Bmv}5<%;SQEk00ue<3z_Ff9 zwuQ(#Kv97yGnC@lzd4*XKy;YQ`aOxs z#%TIo`DYzN$&$+<-oXCdm1t!^9130}IAK#pSYvui_Ra}Eltjxa2a#gO-aGCOTzwtQ zZQV2$MrmRq3ZR3Ba5(V6trT-hii12eJRFgyFiFVF?g7E&WLD5nUKj-viE+=E=qextvN$+`W*|;xmtVEkSd?w zVA$azU(R!K^F?<1>_sqN`%%d_UFFGlqKs0=5Rp42R&|*B3d0LnByJ8{dYe~~A!vl& zxW#b5qfAL@RZ6T%e-ty_< zWWY5SAB1V3u++35^qZ`-Jzgltz>ZIvcO}eRtgCsg;@p!LKN@oZ#!CFz44N4XtK$mK zm%DV4n3lnGl~G{jd1Mt(H1UF$pRSC{`@otmL0RL)F|u%Lkl#-`c}&nOAGo2P<|nNC z&66qC{Lk+si4!7t^UcIw%y9reG*epL6K#pb-*gcJ>yR=RRzN-{S~z0%8-DOFd)yU9bv9w6))=8WYjUx>QCWJWG1rsD{`M7wUgArqQ>ZDgb^4MTHG3yRv5 zeu+Z<_((NVoNd&)SA`*fNBB&z{O9QcU?I+M=v$8o27ri?h$_QZ+*K{f^1%WD!fxhs zDh!5iFShNXn)>B2LSCB2+Xkq0-z&XMp-Zq*$$U~`DbZr47=4(72y=1A4V$;m-t3${ zxgLOpY2~03o5>LWw3d%>DaxJ`13$rk8I4S}lFudy#W&m;i-;;$yw_GjqA-J^<<}Kr zUG~r!HK#b84~9vOm#{v)hB8vY0cd=CNHoVd+b2Y#6^^w1x%PM*T%ILLlJ@GDE6|K5 zI(%AFj5M~rA!uaTNQLA8mBj0*4Q*G!AahkUN%q>;+DIk#6o6;7zR>-8Vxh_PDT6Z) zT$b!fu%a@D3byn>2l7wv>{_gM-+iAkQqi#*O6~!hHuV?lM(7xO@(p9mepmg1I@WqP zXU*JPFOh=~FDAC^%bJ;Cr@g5{v?^p-#yP_|u$PrOc|Ug&f-h`ooklxb##K^P0U@~& zk7XvKkd8pG@(2r;dlcwZ`VsT2RnZ4^qf0OM8_!bJtwb{IczL6k zTuG13^tm;iy1KKfCs4%^%_m3p^=9|_OPFsf1Dg8?S@i{) z+>Vb&ebG4OuYO?Ny|F&w%n-A^_}0T}rYGT?{gG_=DsRF1l3WPp2uHnKffCINBzTg` z9e)H}S>q#<;FhY~WN27?A;6GPIp}Kw+ z12Tx93*%ixC#KSp6iS^myR*ZW52Gk%8FTI4)z%J)65TC$M!J?_5r<5U`YqY6-GkVZ zp}|hm*0%7Ed3Pk9CU7OPCKsuzm zyHi?Xfq;}qcXx@Dl*ASH-skN7-E;1Df5ZFEImaAxjPbAuT#@mrX0`gBLl(LEJ}`=e z2)#w`CO%nKNRcvQW^1h7fv1T6##OuUXeJa>EO)L#p-M&P&^m37E#*>kqBiozgbmRu z`-lQ8*~w+?dS(Ge<73WEmU0K$wpZtA5>tvVUKZKvXe>Y%6R69bn| zq^eRLE{g9TQ!hmneP2+y!T!QGn5OC zes=I>q`JHU30rkDf$3F_ai16dE*p>OV%1&s(-WrC9-(5*%!`s<-}CR0rv8*Rq#M>xRG_xA)>dcuu(XaG$>vcKCxmF>^iawq{uxh%}^p?gB00mMcy zt^(!eX`pYC7Bna9u8XIX6rSEs@TZh+Fo^OrkD3B*%ArUA!PPdpjH|$8PFR|J!Gyp` zhZ(?2694dh=)-5*${vy$L^{Hlgl<`hTny$Ww%}`XhA6>i*pjUFZQ{E9pz_f=5Qq2> zkp6upc`U;O3qTRy`IbvJ6;Ui6OK9Nd4ntKow%1m-$$GSdVKRTDqi2izib)S;VYJH= zi;G$*_L!n0>ix)#G`=(GEzg2To6sGw)AzG(V=?;BRL;$h4Cm}sHU_})ib0mF z-hAnWJpbj8C3tXV-QH^wxyUJH7d>8(lvs-e|MpacHX`}$mQg^%^|5XgI~3`anYGE> zFq_=E1wulJ64n)Bq(`Q%JOi1ymk$IRL2fVny30P8_UuL4U;_5p%hQxrNT5x@8wGDw zUU$SQzV|3>!$D}J>D)dTXIJjElp&E{@Pi`11Jealp+=%Rk2in2RS9git+M)N%D9`K zA;|U;X@mWi%UuZky&ycZx*SxvQjKntcP}Q$?f|q76GJ*7JN>c#n$AuGk%;HxDy8Y= zTS^%?sB}x7a{hHSCR4ScJV{L}D8Us*NXvx?tJnL*PvYfE>XrV)UuzF}aE{$v z&|>j~piW;@_Gm}#GWiQPnP~P>WpW?HAk!%3wB#mT27Vz61AXYMeMfcp7J&Cq#MVd< z`T4pp$x{uZ3)h-Pe72`^qp!69TG3dq4*f&;2qcw~sOhT@131F7G0mhtk@G9W?0z#N?3AgL}I(xYGopon-Byo-`%nM#pA8!8}@#PI4UG^sO(P(1Ndru9)+ z&saQhhg6SMdgN1kJS~zbre~wh(Gm7tggdKJpMUn2zIRI_tvOYxyQO**3D+5+082|{ zUzeMZv*ghr2M|S&e2cWx$)ltAkS)o;850{vhqy566EWb7&9crbr6TE9F;y~3yUu$= zbjfW}>=<8gO?IV+Ho#vJlThcMi}q%sS7&on>hXL!dk|V>-F9SBuUr8qO7Flq8(sp! ztAHI0`n%TkA%O9}O+6^uY}lVu&+yvyQ1i1dC6;3zuS#xTEJzfaNm_|sEP=_WpB(k+ zhMey$`XX7?2gd#wAq(_Y>C@G&Uu)NMe{DP(zu?F5GUEMC8^gs~A0Aj+auyZKEgvvp?3l=4z82lMrAb0zCnBc}6yf8>#= zFr5RJDojS|`@wm$aK-?pDh= z9Xs@g^gjh*d9U?PpGbDjV~p+z2*Ck(R2(4C2kp12(kQsv3iz-LTohvH3J{}gO3UfI zn<&+ph{i27@U+8eN+=r(5KMzWid_x$1*{nOJrOV&_P~)MruN?zWVP<%W%TZM$Q!cSWvoV4vK@1a9*dgf2TULuHh_zs{yMe}b*}`!$Kn zSG#Lb++d9|Z%Ng--dF|@HCBpvdr|3-x?BfxLS>n&qGKX%s^hq*55aib7a7G_4SNlv@v;CE9=zVdP~+=cQIz$+UF#3P55s$wHJl2jwy^g^ay#U{e%M= z(*jHD8yw%S1@gX;?&jlji}cADp63a8{9B^*+<<vn3c$SsQq$p$gAoS-2shny&ouU@&jh0 zR)^x8=I$w+@2UaD^bqfhN_&Impc`hrL9HB_Pl0@-Y1ex=^TB$8{Q+LCS*<-;qNV8r zmxnrjdV(0&UiUUOn|ZgJ%kPN>b5VPn1YbV7D!YdhYr@BKzZD|;6l4uMEh=D0AEAc*8jYdt$p4>GJsvH%P8{M?lzq{az78b zczf-UABHLBau481Th%<;0}~BWIPy!20%x)mjU=ZK`ECN$_QofR?CrJ318j~lg>>+T zA}*`R@aDozk!W?U7z|r2?is*isuWyfqy&=fyjJmZoK*Roo^1@<7K1{8@>=*dyPW!# zva!*w*iMPJ^dpM`)22JWBCPnPl|XlroH}EXl4{HJ;%Q{_BqtREv2`|=INu@+rE(pq+l4LaRGUcLH@-<+g39dm3)%J$KWVX*uy^63pj&klYfi8zON*RJ^ibg; z4+_x+Ss(X-TcS+FUsH~ACNX`Ue?Ub_jXTE1xizoSCw%F;qnW5%b}}p`Kt}8=#>iGF zzE`pb9h8yORWCMtBJKbQK=!}vtR!1rrvh1U{K|6fhDAP#XJ_Lg#~|Db!2Pvmc1$Q_ zvblZ*&#!g?-L`HJxG=$Kf%By&d?P}ODCEi)VD%OnsHQgfpix4P=>0huB4-N+e$TqN z*hGE`<6V^Chd10GXN9zM;yW&P@mo8Q7P0Yzd4}qb!7pvfB z6+e%V*&GmfD8x#t-%!loU!Q`MmE1QNCT@i)#$DN!PTUg3_!bH0-y|O`-rXz%^44rr zHr#lC{0BR4UiJxG1ie*6mWz7jbEOf6PLQA!z`#Z2^Ybbbl8Z@Z@&?E2EJ9{!JO|2^aRzuHW&|0ClW^)JSAs`cIXKN(M-KN!#K ze=?qbd7ebBpDll_tOWV0Ovu!;j^?5)Bsr+|U{u#J8-wroZ zX|*z}T#$y6r zA^Wxk=P6D529nAb_L0qfp2Px)BIg0cW4vbAthOv(mf>G+3y}) z8KsF*G8aXMf*u)Ud2Y^_jq&YBcD4($|2sZkIV2?oC}VaRX_EU}EwzP)?L zoMYY=N?T|nVrdn1PCjdut8>7-qvbWpY#ovEyw0@hv+1khyg}}dMHQo~MT_|pzq3Q1 zpBXmrfjFZY`RfeBnzl1aZJYNiN*^~mbVI)T3(4Ul$Tf^=7Pa9jbop&)R^9Bv81E$4G4QF$t91rkyrR&+$!<^`p7@q0rq!d`_O+MFCU}|+ z^Ry?PBO&gNI#%er)uQKeRY+hlE`#GQ;rVHn!o-ZHp|}uYd&HO|IZSL293aIt)W}S> z)C(p!gpVl*+H?VB5oSLojQL(!R3tW&F9^US1T27imd zRU|${j3M!II4mlEV`COQZ|UqB!{UOCjiK<{^R9Ua+A0=y__0ENH6j(^&B~Ub%N0`f zFYX%8ziLaC9xr{4B3qRQF`-3?!BCzv^u@6n zcBOZFYY@AvBrs{`n9qe~>(}d-o1)=;rfB9*2*}MrXix z*?bTvRSJ){JjHRBakZ5d3L$GBeAJi^=Z*(E#oq0>mHP49p_d+~n93ysTYHsf`#%w& z8H@w`8}p{yo2@8zFz;u*H`K5wpw-DUUVPBS%x4OYkc$&lgEizNVvA;rU~*x+%iM0? zZwB{Lf8HV*<+CmbCKP-zMq|@4Uq39}N|3lH`rLbR&6a;+U6$|*+cYsmON z8oj0M?jL|Q67^l&z&K{u;BYgWRUV(-)j0Z_tGRNmt;#R*u?ZnddPkN ze7lJG#t0{2aUz+5Ll~Z@q^(1FS74NpLC~<}fN)@ddeInD1C6xlihKnPV=S%3=FyOP zT|+oXjT|cv-1Q1@^9y%Gs9%Py8Xo=dqe>5@uqqQclQL9GkDJkA%MT$)P4ih1Uk@Xc zxQg}l)8=P6XbYu}Ml}gxeYTU|pqDzVqSRrfq~jK`6AAG%g@TO0Csq!k51UIh#9sz* z6G)h7yB+lsJ-N2-#-#V6>{;!8tz6F$%X%VidL(%`$os!(L9<58hNce_&Sp$2znzU5 zSK`B~*^_Y37jkJj9%u-!-3j0UP=E(e5DKns)Jg@`*;i8XX1iB z{Fj>wjNOfG$muZU^|VB9sLBjY!1HaSkcy@OdGEBQo!U42h~GM|16sD1wVwr?r(@x7 zr);k~wHJo|Koz@B&Mm$jjq`AgvMG>f-JIc&Xa5wguq)=+pGahn!o#xmLE!G)Uz7QR z9(s194B2{LHda>NfFb+$*JL*eVHeJ{hkcfk=TDaeh!HQpk*ok`pCZ8F%kXWiEKE8y zyqyQc4VoOnp4d#C?EBemKjaxrNY`PYC1v2Z`-8Ib!=I~L)sN%5VC$*|JKoj)VEz>V zBv=_O6NgHu5Z{`acYNiDX}JAXKF=Qq`YP7CF`JJ0$W3T`k^QB(hDZ$)R^uPb(@&*I*;&o`>3TKlg=J%Go`}a`9P(>_OwxwZrIFJ=D&~pNs5p+7 z451%^?exL_N6*>=Q2)1{HT&g3VkqEN2*(^Vvy+D9VMbgGY?n>j4SN)?MO1xVlQYWPXTbKb~ktirRwRoq^bemzNg1a`b^8a1K-M zDHV*sj`TKw$YG!hEPdAL2Lz)|5p%+h=(V#aM&ejrZ}Q!bLxH~Dxhx0MbN z6M%v}=<0A8I6@aA_^S5vq7}7Rw6Ew?pX5WuPr%q-o#h)1N9>KBp)Uv}W2tWic%fl&%>o z_Y#CAf?taTC9|gLbpvxlyQ2c>K@Fb88JgVjKVMR4zjy~E{DTGLCkjjyByl1O z{H(5NE+g8_iyv{eO@v8xzPxI_@LF@Rc+3$DTghivwomc3(oT|5dW|*dQ6ftxF0P|; zpI-wrPH4os5a&-dSiojt%mmF4x(A@9e! zD?1&PFaDv{y42rHo={S%rq{_oKIKZ+47{t_cJ z{acK{_~j2V!s+rK&^+`HXfE-m7{P*)0QGdLRj?sV<6f1ZT>jePmxu<9M?wzxoRsh(RyS zeMQD*!q?|6<;Js;b91T+@3uOgxE7;3Td}cmP08_9Kf(y(;p4-}^kD{tq{gGpr`s>i zAQvxs^On~q5nmw;Lkq_bHHUPk?~Z;@)S~gAlVQ%2^4n{F^ZPi^ZpnWdM%L!N;DI+g zn3{HkCDi5Ndufs-@Ydj5QSam#(*(OoBTRzqwlYc0$oO z`{$-8!bjlob3Y6ovCmP)Z;K^n^g;4wtM8@VH~h$i7LF6x1e<}{VzgJAq>*xjKQzoL z8VzK$8X6BpNEDEzkkmD*>$8lxxgTvR8DqgD$?UE8Cmd1gr4x;L?_C&VVg<>YEmDMcap{umK;M28C==_+tOkH^bFRZpb# zoy0XJ$Vx7-9L>iRFWD_kioQeLE32Kf-1{TepNj&qSBZeI>{oUD8SB4aGswNS|KXpp z{^d2cf5rM&Im=emNvE-tyX@BQJuV;1zKN1A%KmU`fkya5cf6s_Fb1x}zJr=fU{Q@Q13%Zoz`G1Df{3`A z0*6MmU&o1rB^=rV9s7B=6TLDKa)q{}E6zf)%IknU3)9ywiZBc|$IkObQRGFO z?YD3s&ULxk^0!C$ba>i4%cB4bvu|!H->A1|^^87h4-2Sg-SkJJ&{h0MO$F;cHNjDo z8)l0$d)uX)Og8*nMP$nB>AO6;Z!DXhIPIRFqCPX$2@C0_Hwsz2KI}h&<7qK{F+iEl z-`0Nml433Xr?fzFY+pGvMAfbPL;zpOclKh!=+NWVe%!v)ai&T= z5oeS<9b@K$WfenLTm-!Tp{**FOkjOY2qK*7JTwc*ET#sF(3BlWlQO`M8vD-~_y1Mw z|8Wnui27Ht|7#Ct{PM>hewy{qV*hgwcfBP=y?4qGHQZi`Yrv+;%*P69&m-C9Diw-p z_&tUrNO(-SI`sYl**Ecf)!hou1)uT;qph=XC8`Gq9sA{tczt$5~(>#nvCS*RwIsvkh05 z5>@JMQ!EEMpR7$!Fq~hx0JePxTXF6#en2kRYHYpRFEU!+sx=mZJtHDLS9X3oWnR9# z3*$LJb|cVfTu@7P#8_~W^n zaoRruOI~;?wL<#L$a>`XM&O@O$VZNUvtaskc>Z>dM7jL<2sYPiIPfJIOh|$6YZHSl zicdbR@cb_Sm;h$ud^IkPlQjY?!3%$^Z-U=S5vhRp109dzg;E#$S=D?Ed%1!enJsD{ z#rO*Xr|b8jtRMMLgy`@}&g#1zQT{M50CI1O%9TgtlOa65z8K=bVWEflXJAi)^#(^f zNy_mR3ZJt#p9ztymA-Hb9?-962;OW0G+n*$hbefnb7C(RWvQ0S&$;>p5Olzet8JjJ4ar}5w)h6fhFI{@2%Dqm4QIOSjEOKbQ)$`}0x z>pJKkx(#IY)&C}YWW3p%)q}0?e4C z92Ucqizd|`&q}$yq7FJLD^tjM9kX0)Xk!$^bjx2YPqsm(2pwPumv10Q5~fyZ>YinJ z3i2zx;-~upYmm@6?!)&mvyUyG_P$cE+cTd4GphxMy%C6an~+ao`WHV4V`s5aP`UKT zyyq9e)8J)0UjipZg6>ww0q?yTh9sY2@xm0tkY~o3Hj-6cnLwz=)STrai8+j*fud^3 zg~4Sz%`3erVFK`>+||&Auw_;MRp8|((%|(Frl?;KV>3hMVC8(^9kT)|l_{#cL!NPr zvQW3s)=mP6oxpRk88bO01QOo4e^Xj0y9XCVGLk@}h8}USOS)Z*1&l_m{NwXh|r7;~8Px zSO(y-&t)L^gc~b)(QsGV?faigBH+2!2)pBZQA`^B+zHR#qG^%S6C&3Exofj zUg1ot=2(fYY+WANFJ$Ew@v{#9OMpyf#=<~-84&~0;+=q2LiK9urk$$(*V{Gr z7kWN7H9z;TH&-zSnC-6Bvf9r85VsNb^IqgJL*$HQh#|>DL8b>tf+<^s-d#LS_q-8T zlZ3O_HO&UhHl?FBAaA}Jg4#5`fMMElD63(c1j1E|c)~y#%A0esB=Dp9LL#gI+6V>r zcAtGU`?-Q|_+Bo>Elci>4T(&cxrc!?n1{P~!{dTLZPZ7Zp-WrUm{58?!CZw$FU7i^ z4_6i1{9Bvzs8D5Org$sE>+BIM?T;3H)rFrCpiyfo62`DwBuO7XzL6URlvzxRP$?-^ zkHsE0NiwiyE>VnFHG3q=mHvFU1jpGjx}>b0~`PvtQlbT%o8nMjngwvCS}?6=|cj^am=}pVM0GoijkRDy^`J5y=9tX{V)Jo zwG3U0Axmj$ntg}FvFq~DngdR^I_z^iDPdP}2k2ybaCQ+}(SQ&;{aQQ9%*RoBHxt`c zg9ctv6c}~;X&dVGjJvuFZeGgPc}6#8%b~bQWZy*ly*RFWfkBnLZE=XW#%Q%ij#2H_ zKDEmygo#X!LwZ|GqwCX^+@mBgD+{#ojIPFJ^yzz+6z%Nit;K?8fYlwqO(}@zg)MX6YI^Wh}lUnmcy$V{x)teaETJ zyZp2xjD9_*0_&S_gwgLHg_k8uafyFiDX)wk7u8?w5f$@XiUJTBMVs}C;hqP$kbh(- zU<2*|GXJkCGwdQF{J&o$Qq1~)2*_`?P591w{#mn$6KKZs%Iu za|M0cnBN7Y@zI(Mp?~E5!{UnAKVQI}xEkZkcpeDV=Iqsb zsC@qBt_Ta3YJ=>X5MaL*5C|mf34kLDFAC>V`ao`eP3x`fgY( zEBdR-Gv&n$WxM2x9Ajl^cxM_+VYp8tyVFN}f5sTi{)%o}5wdBrk{|vwhf&G1k)(-U zI449S$&0i$Nfx1&Hg{TI)g+Xqv zqq~xUkK@&&H*tY7*{(?%cg9888qUqT<&dUVG#PjVJzGMxtIYdV4Uof83{z++t8^Q8 z^S%&(9KT$6j8KhD^w%<7E<9&TXYpoR{1vs>(#tUQL{I=mndzo!z(I3hNv+mLJ>*Ff z*x8UpR122wE^DhApYv}YNR{f{X$WoE0WgUXp;1gQ{a3i+$KsH~Zgd9~&~cSZ>C`TkCi{tSB!8?$dkbZU;M$vzVx4$59RZmH!wF!L)$`zh6vf>C3$sBnt)I zwu}p44)fRG;)MT3LsdrrlZ(NrkaTIR2^P8VeIeZTY1)&)UaUUOG2)ajztB+l^$;7V zuUhWS5a?)NI%0n2%Q(9%`C%6j;&yoklzDo(G#@A^E<7S$`r{1TF4A~t;XR?-q!J#9 z`DldpwS8pX3nP-{j9hTQF5q=!nZ^h$Pe(%=3ro6*S-|Itak_JSyF&p2jw9I*p)_5( zcF`E@_-){M(e4N8TqE$K2}!4%=~2eIpAwWo57j(BsMnHbT6;tZ`VIqEGI^OreOgu`6@rsOB4W^ulX+fjge>x?SkIjdD@%e~p%wMU}Qk-<0 z=5yt10#hv6`pxzaMr6O{8yG2c9+(?$x=v>D9mEoY}ZN7njo0Hky_QrsI z-~tsV>=Dj&^UvwDNOtdSD?Zd?Wyq-`Ht5-2`KMO1bgxihd`eWUbwXc3l!JOo9U!_C zq4SF`g!;M|3mcJdwX4pPjPtRa6<_1ZNF^CKIVT#5HEk653bw+s7Q>dJ)pJ7V6!MLH zkwv)F{4gbex#ra7M;HO+D@De_NV^xqmMbIXdW59LD{2wakI`OhDjc|_DQ#Hb1`v*3 zy5PrjS`^cV5-T{FC~MjlZ^qPoo6Z-L#L6!;YAbkLkZCVa2F00?Q%q}8D!M76r5Q-1| zr@$kU!kJPb%)01YjAJ1Zk?7aXBa9P;S|1ALi`5O)!pJ^`zU`mb(F)}dsF-*EQ8w40 z9dVobLUVa4=!^~^1ENFItgr`T++H3NEqn^3xN;)GOfrH+0D&J~G&)LuNYJNTOsij! z5%Q%9vn5B3+884TY9+n3I`nSMewSoEfSQn$8lFOd2hJusoH-W&#(JJG9v zy4rq9%N~o+BGsbtnYNHM9lC+d`II1A-JP|P*wMgztH30%?_O)Lm5*;CeNwzGhtNc6 zQ3AU%TW{X?F;cyeYC205{e~+!FAq=K(&7O3Z_UGOu!J-XQ5FNDjBjq#|WQ&4y@pybO zu~1&-Vk^Q4W?}ZKyT;uzmEz)pdm3poigcuT*Nw&TY8TKB?_U#Wpz@4NZL2}wcdCEs zghPPIg%AUiJen;lV4ihVDqF-ll)*zGcR)04kw)*==gsoOC^m8PDdA!h!w;H}i4jPZ z`!9D>otXU47u7%uZz_RjIaY`l)@QOsA6Zl(RO+fj*+#(vp5Z(%@;SEK)koW=Vhl8X!N^j+f86g-^wIxKdXX$_3dKt9?3bF zpLPv`uQsh5t>qOH-ErMg@O4eW(@^@Ey=f%p537x~w`#Ayk9@;;RcqI_Vzf0}M~F$~ z+nb>8CXBWmNl5HZ_0WYHhAZ;+I>bfG6G1_wBeny4=dVQ3>9Q>9XHxwd;&&MfB>d`}5TP|0-DGkU zCXOX?p{;j*iMN%i!OWd&?}-MLzhh zV6doe^-cU8D2+TP_cAq8rxd$P)z$Vb*Q0p=p_%bvH}jLUqwa_=NJNDP3`KE$ zA<+eg)&M)!=mGN2tS5sc@^U9%c(-)av|0A{PreGFu$>Nz5!sxM7}D5Bjk5j-KON)c z!}>9f`>&|AKY#JwKW~En`4|5yVWPKstwW#kfA~yg6VD+%_`Dyo+JgvLum>+|3ctos zVL{}1ET?Y1MPzMkkX-@D{zR8jw8ZHBXaQ^ZEvJS~`e1H*{kdDk9(hrhvlCJ;Gri%_L;vi{3@)t61-~0lH*S;&4=z37b5R4VHEj$Xy5iC>)xPKVGM)_e zHJ^R-0EBX65tWGyl)?Rl7NP&Z2w(siQ2l?w2>-i{Sl*X}V6)|(Ye{NH-<)^n*#7!! z3-qoHI?8PIQ3|2t%DNA&#LTsv2*tbK0-ta%G(Y>b1xh`QnYfZ=o^7z#uSh3rsL+kK zzC*Sd_))RcPThyz^7*{D!xLoz8;0KKVASS;LL4zWS-mlUo4NS<=lDal!7Ne6&64`E z?XjZlmpA0jPj{!XvqdtnKGv^|8uOIaeIYwM$ZmZ%<%~gIeYj>%Kqfxa(sZ&loGY8D z(AxZCZ~BAP687AZ==UHZ(z~t4_Dzx};*87*P2Myf>EV3?D16>t*O!#}^`E~ktqk2k z-ZBio#ldC5h3g_;D7^(o1VP=v9K?(88)qaP_k;YzWr@&ljD&$?mg_5mPz~bfAZWbS zgHQ~2l8ADkzms&#%RSrt&5sj#~Qp1j+GhkE>#Gv$68VWbEw%&su zAP59*2mqnNub4(2+OMGW zm>Zt)WUq{Ym~Wx-5ku*|!qJGi;)8ULSql!h6dyGH$o=)MP-s&TGK88x)e=?9AEETt zrt#3P9H9TYhYbIixx`vj*<~djG!x=4csu>Bw<_)N$H@#ni;orT+R&7a$XjDLayAtl zIP%c81y+uMbZ5|0#t3C4Y57607Btl-&wbk6qp>UA$94>frFXBt zh&Y9jO{Oe&vbD?U?qM{O>f|@z0##T!aDJ{?P$7iYA2*PDrgeXA3lHDzW6!+L>y(<5 z$>sFCOP?6=E8k!ck&D=~;54E>^-{JAFr__V}EM*!lE|M0682qKV9A z-mrN#L1Hn=6+!h9@wLwjy?11Ok4{6Ik6nAXS5V@K5IZ~&=6#>x*+N#qbk~Due5r4u zx)e#UdaBawhC>x2g%rre5GRB6o?7@QfN+_R6VsRqW8qydg{=XYXL^N*>8a@TK4VGs zsR>QE`vd6c#50&Uy53T5m~a7pZmtCtI!Uh(s*nW##?30W3)4HYevXbgheIb6G1P`T zB~+Su3v*LyQbQ3|Nd&n;7q7j@DvoBSNH&^$u<+m#RNc56RRrk?Y&_byleZi1ivo>& z83um7mP8m+HI>D&p<(bg1O3}Re>ePGaF6PDKwxHMk>+8A| zOp=5LQVB`f`Dn#Xpr@;xT*0`nh@JVO^sX3ve*WCh_j3&MY+0rJB9_j!5+LRORyH}| z2;0!M13;tZi$;LTa2{M7i!c~_GU(I^oa`GyAzw*Gc~anqa2e|1%?9rCBXCbI@8HZW zG?B-2y=#~keI7LV$yYjCV9YEfF!3@)U)*58spAt9EV=}Uc)tK~)mLZI72qD%Z+XyT z^gK)yo!1iIkI@;*#(?4%0RqRxilQS!??C8FGE2wr(aTOKhRB3oo}(#AuW}iT&J}C_ zDGyRRLXpRTjsU86ie$<$kcs=1z7L-Qk(8soW8%ikQjkmbYUf3!EP$QFbXlMuR3k6V zH7Yrz^7c-#MW<%`vsBQ8crf-q?ph{D&#zStw@4xU+Z7aupEh?+JtVgM$D6oiGSE`~Mmgiy!X?Me-f`?E_vegL@pneA9T6NcPmWe}1O; z5HVV+^?~zxZL>QSk5Mo9nS(5qdb#;n!NN{YKnRh+QYf3cTKU39nV`-EjPt?88aFUT z{SKG)e5-?98pU8^wM>QA0ur2jGGJ{WNp{ikjn?vJ1HB$G|10|=p|NCDvs+b&Z7QG= z+~~0~<-a@E_~^C_x&44{DEVWfgMa%$2SvZmB^nPrQa1B32di2$aK432HL>}8 zzTGtS&iyE?luR|Rz4YroJEC=&eXwQ6gT)0TTAX@efcji&@ zK;_$cd#MKa1OR?E7wtM|DZuA9iI4D_a)`tsJ zL?tVTUS}A|ifs`WEXGi*q$EdLydDc&Cf*QK(;%Ub(;`);k8igwHcEgTX|2ZTC4|r= zNhcH+#6m;+a8u1^wnxIOJd$Zs#*rV5Cg^haZ7AnCOXsGd;HDU)n=XbhMq9>MY~)~c zgixiMTp4Y}M-^N$#=n}NQeSzgoA zyB-feNU=(B^|)dQ{OJ8?zv}NIi2)U_(IMi$9q*6;Q~=t)$M(WH^uf5KfBr)cnZy6C ziT%I-A=&7=Us!V6ZlyKgOz)m@UKz{~!kM(@eNr@%%iu%vR{KNo*Zh0R*@I&R{`$OX z@uT#8Y_E_HHidm8dMvYO$vip6eP#*09RZ(E!+|fY7A|4HKtrW|;ImZTD%mLnkippG zR(CLoU8FN`kJN3dvBso9ZJ*Q)dI^t4;rm?EDo2_(I*M)gq-icIj=$#yY3t7c{quvxI|^?q zPH=@_#P|8V)>he1sW+E%)=7Qrs|P20Q!kLOHOjE43evri9=PGSsX<4OfLP>>8ipVT z&1L^@@_Cd2#H$)(^5ocCD_RdsnO1}85-z3`=ks}$f*J1Lln?Frmj-^bXzfo8d^3Sq zlo*Zk9DCu-Q!-C+ZfRw?9RmFPSTza%hjA*{_H!8OE@-$3Fv~{;!VGQ}o1#k0#invn zJ?0b1_Myp}DF%ilTj`)L(pxDQ4k2b~o(IWWLMFHFrdfAXeOmxI(pmZ(-+sO*lJ_ss zi?iZEGKE5C*3UNq0j4tfSx)tRMMd^8yE|D$9`3~{kW0Rql1OFdoyeG19v&-Y!U@gH zmFj+%Uj_0&o@Ti{8hr~|1`J_^#l6pEOY^?S{4}eZ1o4?9%-jaoaC*d(&B=+|@kBMPmcOxt}`inQakdaLCRXkBe)aL~5IC_^&(4J(C z37t2t&IQca1f`2KaINTrHRd@H;M;PDo*$EjI@#-5KBOtzmR72HfX#G+Hn_hvmk{1d z!GU0pqDWee5;E9?y&JDXP8NQDvl{3!yKp?frf8d}>IQO7N&|w*3sP{!0x#9P5=JD6 zO&{M*_uJVFU*tZTYXxeeD&spTBKJ`s#(@HqM?xZP!H<~a5(F+N{c)OB<^nVUNjG_rQw2UF)Hc1SqCMQN;;?$sGO@?CFB^`#@+79-{ z8-S_mkqG3}j<6F(>&YGz1!+u1>G_dZeJ@7aAt=(M$q8VsKY&)ItEBS#JX22Mtq1wnbiUHeVVG!hHe)XNY;qQ5rRUA8e!xwYnDzu)oXuq?7Qqba*SiH*agsZ{YDFZ*)A~it zzS1Ylw$E-y2kWRaU`2RXE^bog*lN0AjIxv(>)5_JS@+Wj?!nx`(Zd z(d-7sFi%3Nb!@-M#~nYvZkm_eHiw$^%w2jJ_(uu<0`XBVAk<<2TR0@iZGr35S&SA{ed%)1B9Fg0d8)ng`#}7IW^}J>W-3eej+BD zr&qpEw*mAa9=@@d{y0b^2#~^HP{HnsyaU*)|7YzvRDTc$3A0R^{=7cwV3iwoX|h`UwQ#+_fK=Ie zdCsFNZ@jQlhte$l%z#{vU|@>bhbdnNNAvP_=i&#x(R7khFqHblWUuT_J>1EDAsK_! zJRD>~rvuexWF|EQhTjew5)O{PSQD@gb`S5nI;{B&7V!_f9Tvf_%@RB|33xxQszuYW zD)RAq+tMd>@)LTHFu$#erVL5`(sn^pLi%xhN}eL|>zJnQw^#`(we++^lDPF~{<@8r z1snsW{RX5bzn$8NeS@aYKTu1#nHmG^p_VcyDbs@tCbfNEY&}j=moW=@KT8ierkwmc z1o|ZXFuag|i=@hMG*;kW356g4YZBlgz~MhIHATY@@2Bf+Xz<|@A_T;y!d2w(ujxuB z9L}IxLOzYV-s;P4gzdx>wfQL#?#ypvF)>3#J+GA~iA9V}a9=@;?@7jN zUjW4GE=eQPYett*M{|*$Gg-8DCYRde`HhCE&9?7f%jHMm=>&}_b^Ij>UnvG{C}FY|;VA$jf4aJ1Qs+`Q!mr%s1YgV1+AHL&A36I7&~UYNvET zLK?d4*9Bib&l;l=$4kL)(c+R{f%t<0P_?C$vOM6?3@gd_s||E6b2i;605Wu_)cZ>6 zcfAS?1YK0x%BaJ#X4Go4V=8b<#?Xd^81eun)7Jt%H8`3{WpSyfw)LIIRPtjE1{iPpz`5MOI32&HLTR}EImqu}A=5}ELZ1?Ry-NiDpSA<(52ihy$PTx(sGS9W z({w%hhK0$;5yqbPYmi4=R_+d!&>$NU1RIpxb5MFWTam0YelI@V`=IxPO6~7%IZQ`Nm(cFd|@&G{dXu!&c~9G9zbl z_q~p6oY)$&Ih8yV5Ts0q!X2!l2;=Za7R5zZGXUT{zh5nwdtgCSAn_RKw#LmA(Yt^P z9beLwbxO$qKS030rMH?SVa&f_p<|=I2c$_$$&(L4f-8m7$Fv^ZkG!lLx=#JbI2<_d z7#VkdvArJu8OsUh*I;}iM!@^Toj#@k-vNcp(5-?{8J*&0yuwv(kp zFO!Q?_2?|pC3IxWO1KqZlNRZa##dWl>CRVc?B_+r1u-aqet6i6UG$vv;Hx#992ZXE zyz2kM*jGnI*>3I6FfepCNJ=TKA_5{qcXxMpr!sWQ(B0DAh{S+|3P?9YHzFaa#1DMm z^M3C+zjMy_{li+UVOYE-W`TSLk8UbMR?zR zV0D+IpaO92J|G781IhlW!!YRPuap@1kNJH^z#mdU==bycG#J{XxR$GncN;$^s zk&SPU=13QWPV6+4DQBzH+q|1xrP(&(~J@${KV zc_vN9*Rtl*fdC)PFoZNEtJN&Tc=uI{FRw#sbL}gf8v!%Bncf6e!#}xG!#8y8t?saX z!nZ`JSr_hsLei19%v>EWFAqrdpQ-5{A^!|Fr~UO(@kya~@b@KIxAtFC?S`!6>$tv! zJN-(2KT6#(QRJR+K}c8F3rPaO)M8YQ$O|UDK*2GmBaY&aO&eRa{i#yt*k^ zu_)S0V4g1-SwN1;{gZ@Pu1!YVD^`byC8_94hZU$=WX6`Xg|0iQ^a;$$QaPW-Tv<0! zXGSlB#aC|khI!qy4`TUAhh3N{S$DMs2jy=wDS*2l%s|cmp4Z-9qM9P2@GNq2;8p@{n6K8*hH^>Z$Q^n!|ywbB3MBx2+)l_j-jx zTOGZSg z!|Y7m2z}%(XhvpE(^Qfw1X_5yglW;{JjH&C4R|Q>F5pzrZE$CE6&8X1DH|f7(~~4j zO0csoe}=6_6fGV39FH)LG0h}e(SuJT!65X}W~y1T^=6t?QR!y7UDM9yec7d0)rjet zR+j5#=~i|n5WbbeYoEp-?95Hfm>Vw9Gm)k8J=`)M<{(!1HZhQRrx2gpXFxDx(Uh?$ zx9r78t{RteFP{3hXI3Q@&!0`Oi$A!f-K#)-P+J;TxfzKUE&l>iUIBgpnJaEJ*=0@u zT~U0i0Jo`rtJlAG)eRJf@H$koBZz`@6^4Dj9HLvS^1jwyjjHvNz5L)A?H+raF)M3) zQqk6aD|-ZH1^(7eq+^v_BW4Bg8tE6H{nIt&u3+O-*&1-*38bPs4{Td0bfZPdM%cL^ z>--K$;V7p#66edd#QMaFJQLr;S+))Zc<@>_q6b5Tzj83)E} z_lz-Nr;PeLyk{I&lDwB2z8@devaB3M*YVtUPgRtaR#JNO>gOqphp^IP)G~^Eeh8mx z+9SoHrv3RSID(gV<~imE_H`vJj7Hu#cS+kh^TIemrIn!OPwv8)MNNeZF(0(f<0E9N zS|C1}3hOIwj3w!_*}R{;*3;Ks0@e!8HEGww$fx)b9&675Hrus`%WpVOywWr`jucvk z()O5q8<1)_uEI(cJNyo6wwQfqhE*ls#|}ii89ANQ)da7T(J8z>A7SNvb-BBrh##G5 z`nDq>`t16R|5B{|;{@iOc@V{TCjWQMV83F}Bf*m6ev~_%}bAQ&&wxN!V}jv7EV$bera8tj+gmjl%sG z+mx>JPj?NQ2bd-qe1)C2?+#&|?p=K+g_HUW+izxP998N?X`)?B?UObyn8^4# zNLlKOp3gWhaRy#)Nl-S6okUT2G^~WO0xu-99Eng$a;ch)EWx4pv%QP#o{6I((A>(o?LoIB1{Ht#-A7Lf zJ{mIl3#wVe7(f5^yO(2TfQ;uTn-5ZJw}x_F1&uxUjYSFdpSOkG&liX~ANRZ%SrRS> zSK3r!+*6baNz`JGMhXYA_`B|~=oH;ts7gJFS!?Yb7WET)87ne!1ddlUH24iNGo}^7NG~u(^;O#+PxE; z9+|%-byktQIq|IMT)up6&Jgu)$kQPQHsC|TMzWk&BC_m@GlWn(dS?T%7x8pi19&bd zBVYJT&6x1VO~GV2TDY}>-yrZ9?D&|{G#zxVKhnqIa&H7;TC^yK8ag!KG01*AB+m9!foCqsR1yYOjVh?R;RN!U&?x1f&FJKRdmDGj6{lD~EL+~Yg8Kl|tI z6KbX9>Ht*CeRwxVPH0{nTt^1=xdXQY@p=^L4=_2I| zx8nN_>0`Z_+8uZD(VhqRAhq?56Fv;y%>CZ9(6Y`s+s7&}9`4C?Q+R&+11 zJ=WrtCqcE%$*KDR3c_5(Oe{BSCR(|P{P5ynz6DkZZD2t0&1+ZHT{t1=cB`#b zWc>$9i>}l)m4&il{iFsW@CbdI)^WR?cfoccjbtH+%wl>@6`O!zK5)dG#l!_;l_w(% zn`(N}pY`o_zReL(%h>Z7{n=^^?DXBk7%Zyw0)P+&@;dCvs+HN>GvSH67`j>4^#oVC zsk9g+4eJew8U50YWId0ajTEEJlWvIVV1X`Jon zv%wtIh0vA|=RCEx6815VU=i-hbnDLZI@TQ5YW~cicM2jRlw?HiOE*OrpN2Th>pWRbc`IheFgTV3=zuhDo9>3#3JO0Zk%7rI> zvK5ZenaAbom4nqlq^u~O5D|LpP0lJiUAtN%PhT*!fxt}>boGiW5%?86gx7E2=^aIh zwe&-lymO(1-Cb05Z||Eh82%apqQHdMBQ%+SlTEcZE&G&WtZrMQEyS-OBW&c$kb11# z^cQP-iyC#5r#B{&pT$E3#7oapeQ;4^VcA7p)lJ8tlpI(2u9TH-RHI5gs$Iy0ShU&i z^PBt;%{81Ym`4)6&xeaDuz+K*vL4S}!5>roeEeeCD&()yD?2eLwXp;IU!^3EJtJs2 zs4jcma$5bEuoy&_&{+g0lm0uY@b%7-lLGMXjF;sfjvR1<^KXoILF;HqJQkf5{5SM{ z|78QfKZ(EsZ*d;0c{%?6GXE80ZN6|MA*#uPn$i6JSbRzyIjLhK5g3cAfm+&_L2m*Y zYIiV;3uGuA?^F2Gg}@>P6hoEjh0u zuaPo!ko0K7))xmHQ7zB|0$kC$v+6G}Y!978pT<%N+{fSQdTMqBh!}*F(?-(*BX*t|9qbR$)m5CAtkME*e05&_+8f5(pp(LfP+`VmoxmYUN8Ax|g z9|Ih4i0afs_k2oQ>lA_>$17gq$LWe4aJec=1Uezmx9nhdySr$J;{i@br!W=&8}(pn z$BD|65Tn4ZusL%3g|)t*5@KeU z%85$e8>44qfJ}3kxMPPI>>Vdg+Vo729yWZ25|^5MQeBT?2R{&0wWSSK@>x%%bZijD z4`M!iLFfOacS4f{zd{q!l3opUjI*%i5W()*vJz@5nXRRr`4D?~*p|W!XoQB8PCClz zzhD%KS;EJn6iIW$m?-fQJ1y5;xlbc4(!~e)U4G<^%6JK8*E|KWxt2FkqO)#Dm|7W^ zw2pFMLm9u4e8*3nz1mN^>HptE_z;p4tZBwWx6*r=0J*$5O}UX!7!Jc66YMMd#a>^aqbi<<(z472_ZjTNLuy_jI+pzFr_y z#T;U75YME(h7G=A)qgxHCx9J4szNiNB%&#+jyEX;I#z@HZ790s?!LPL{zdc{|EKI9 z#zKBWapAZu7Eb)oX2FVU4G>Gs_nSKI3Sq(Jcy<|Q-W5Tr|K$OBeEv`*E#=}kpVQW0 zD35>y)=zE*DOCeyS_H731Pwdp-IJdFlLe)h#7;oe_Xj52UR7%}q|-|N93 zT&Fh5I-4+35V=I)X|~%;R0QZ+;D~#-Z7Ay1Eot%zh)@D8X3I$UY#V{5PB-7B)+KIhru{h- zeLqETKNM{f3>Pjh+I{Tj#D8gSrCQD5PF{$FqPiys+uya3F(C=ouIi{m?n&k3Q@5&} zquK~1A8zepbIv6{2_xssgzYyu*6{5szDz#~ZCd{w+OCjxni;8B_r&J!W7%A3d5a21`-x3dCB$!cZ;W%s#`ZhFKeN0h9b6wg$?0Gvd%%9t0X)RSPdJTYot0tT z@1M_(bL?k8tb*ImuT=UtA3H|7NJ}Fpv3G=6{OV(W`k|))?(ptc6ctj=LcQqsj)^an z4lReWAMG3H_eN1aUVXz{U6tiA9bkI5HpNDa%zGxK&gavVX;@9AJO?KTj-Cf! zORD_c8tzqc*?AIjz<&BCTekk$R_@SdkG|Ur9Scc zv#^}CJ-Q?kdlq^>W6V4pAVDlqIa9Yg5uli1gZQHU?n?l>VC0b1$>XVd?GI)8A9yNz zO3#@zsGRl;5Wdc{yNH4Fwa*Fu1|q?-A<{{7>Z|Gd11(2$rBU`t&Ze8#eY)Z}R*Qbi zQi#sCqEQCF`iG)OnH6&17>jU;7RWzvnzCC4Bmli7_ZDA|+vz0^T)GFQ?b6yCSF=pL{EC_gWog)^Vyg%7c#D3F)$8SB5j8Sw@oGlEnT2%hMbuax_+=p*{1IfWiU91P~B9EdD zI!eEu=3Ai;bT@zX z#C1aApiHIF&Ska=TZX6CyPtboCtN2bD8+t!mEmM>np!c?@)*?KS*WkkC>T1Je@Xc6 zV8CLY&~wT~H#cK^?v>(x!z`EPP`t3;`Ss&ZMjwzp7f>3bRT1tq&ze>jk*HgQ`Gq)K zHc)^umyLS$ttU$vgzMk6Xhmaw?i3UYw2UFIVR@|9?&IqK7tUgGm$=&!mx=fi zI3f~sH7ay{tx=&y1I?TxU{rqu;jg{v978G8?+msh+vOZ`6kLaZ6k5~H)TjN=_=#)T zc^x531cUW%GXd3^_j&-QK#;^Yn{0XA%^Y4%ryTF2?vRWgBu?C`jh=u#)Jp7^xh73b zKep(H+D5Ned!n9j=w0hXD}CQxmoR9)^?J;PVs66EoA$`5;JHqax)6bue z*#^BlpT-mxJQI}sv4CAa=Z>m#f^cJg_I$`Jh|H^<)Rd}G-;33)Wx?lm?b|9kp@COv zoZ{Qd@)Mg%@7YI^*gUJrwsNOiz5E&O8_JZYo^u9{Ut8N^rWflMnyMGhM{AZG>>4dV z2OhJbKi+x_c4Nx5i29DV`~CnG4Dt41?AXE1CVuF;t|@b~-<&LY7#=JQI?|+GdYC>t z8UHpfaIzcU?|6H?33}^4&~tEbdkera>KdelBT*be(d2TGNI${B{D%Q}(YeUXQe7B2 zhk;~+xnSN;T{u36L9|$TXp*0*Z{iOBiBV+*YX4Qh1O8oB`~T<(VjsMO{*Ncf>kPUD zFR*^*`T|Jd(Ef9v+1| zuoL5{88p6T**^{yN8!aqj?i#6W{0_4>q(EH!(f~8{>099T_}n{cKc(X{M-~j#R!Nre*y&3FuZ{z%y^Y4X;wzv9HS?|q*TZb@7 zSYCbb?Tu3;|xcUsXaG%7wO`=OI>ss`lEC$*jR{Q)Gye$_zq+>#p zEkaQ&V!i9=#B|LDxmi~AJ5vLe&>g-pJb1mIhm5ehrAk)vMlO4%8;bJi#p+u%J}W)X z=BriH;5FjQL<o zMV@5$CM4ZiIFxtY+5f%QtaO9+bW)gH(d=f9D#BIKIgXeg?#$G%tr1;YX${lf*wLm= zgKF?6I-9qRMw&XLIt7uZHA-X$)Fn0Jg`=t1?3BH@9^3Hf)Vke~U?pBDnisb0o+6JK zuH9vob{6#L$1F=*5rHA8qs>6aEAP_vc=wsv>WL-_b&C;>er;j*&q*qJ%)LAY)sxAx zadqw#kBkar8)d8BA7T`;$nzum9NI2I-Bg;8_bKBwFTpPvkvTBz4#j;wCsoK91UTc7 zph~hG^>iG|s*(^{z9cfIx+Y5G6g++B+(^Qzb_`%1Fsl}B8xnZpanQk5@5?=GS6>a| zQ2R0T(sSY%qIAY~qI82}eVIB#6kEyon&a70l$Qv10`POx-W$jx?vKU z@{R;U>|H0!RwFvc0W;0pyf0>kRTM5)9Vmg`^Vum5@8`kM-_Jl`Dg*xM^Ju~6Ur?8a zmi7uesak|?kZ+z73a@?j%u{{AR5FIG7hZQ5F2uxxFON3@H^m$nx*B1P3SBvM&1o1r zA0VwHI@c3wetKycW%KTITfpAq*I~co$**I|*Ke*bwyJM$%8NdA_6c9)08pj=Po6-r z>KAR&-*+H_e{1YM{V$dtx6&DjbdB8>URwY+1PN2Dzko6b$1&36pZp=9Pgvz?T&kNy&xXm`#K2xiGG1}ORY!|~@i=Fl$OotCC^b296KOXzZC{?TlnT_LQ%UzbN z2I+3#;C;aR-Z`B6DDc7S1C1Tu0#cHy!B(Du9(>iHlZ{vp(fAukSkuoF>p5Sd@Mm;m z%R5sI5XKJ%U$6ENJ*}Lvg&(|sB(vz$SqWbqkXjv#KJ2))aa#Tq@M9}B)d11f?prw8 z6=q;pLrQ{2b{I4h+m+MVJtAJ5p^I)`s0_r(EnWu7Ev%~Ri?%bbxX7Zj7>PuyhY6!m zZl)}~QC2$73;}&n8{s@p49f;?Qz^p*2>Vz}r5Qqt>49FKn zQKIf^?;z>yW(GwilHjRmHLylO3|iqvED5zsi)jkFzSNrK;!4j5yL{#rgb=CJY7rl+ zR&yg$VuvXwYb@YYmgK@*DMN_v@X}brn0Jx%)ZH6CQ70bpcQm5cb$>m~wcln8StGi6PjKSA#Obk{Fo6;lR2RMgO-qKH29<}2=46!pL@|+(D zZH${6C=Hm~Oi~2e>}JYjy}0;dt-(cS*~8YU@ld=Y8oxq_%qTB_gc?MKfb}!A;w2Js zbs>X@bI*(hq?+H2TE28;brdzW1_1lpuaFW0m_y7Grr60dPo{D5PRJEZ$JNwjY6@SI ziuPec_CFf+3*y^--Pv_FU0fX^Ga{?>I%!ydimHtxn06UQjghB*vTHeLl3uL2(uiG* z`agY3ww`KPb-9^2KY6+3tcmk;`(28|y->{CMDnucMGAtdevV(?8m1k7?YC@I{rcW^ z{^vT2!yZz{tD{fM)mO)Zya!i5MkG0}PbM@SuTP&SycOL4`+{UN05yOPNdK?KsD@@{ z^fBb0w^`H{bo-(T`}|QYLAlP;<2zguPazd|x$3QBl1)Xie%$jJzxZq5?YG&5?Rd+9 z0wiY&9<;ihkz%ESs>DZ0EPXlZ^)~xjZVW@!Ab*q7mWF(lcc9j6d5mXya!E)l4q3iG zjI;x7AAS*M(PYsN_Wan*MmdwK&%cM0=DYOY?+ z3iJ{as}*yr6*n#>0<>V9sZ`{U{gqQqHx3L*Wk<%hQu%Li+8)kEN+C_IMS=J(mJ{Fr z6GfM*qn35ov08B|6!6QMoRpu$%W$Z)hP9@S=93*PJW@)8aX;T4KU&3+?Fp+|K+|Hi;q}7s^f03d>(bu2s_9Ea!Asn^scnJWd zHnNroA}CWXir^?y%7k#;mJrb2h+)V*!7d}R6z&);e3WG+au<70$wrrQpg%78`M z__UZDWW8=N9A57%O1dq!s;LMV#s#ooT2LTToYi+wYkdDV$0v>34{I>!EEFrcAmz05 zxZsbA;t2M=(SQ{?A4{$CZ?iP=xMa1=DD73{t9F17{54NL1L%loBOu{Db)`kmDYLX) zcwyaFf)-Ga(0R2ki()$nl@WsVr927Y&h!{xWQTwXylkxL>S&{1?FJoB87di2C-3t36s2l)-+3kf)~-JA2p zPX-Jm!EY%-GiA*#zLtEbxU}vlwZPvflkD&j%NqKgg=pHU@~!`HZ;a(zg)jcu7+=Z& zURIl&fR$;C{!U(D-9ch1pwNF6g!$-?dW~IU`EMGB7)9P9abpDYL5c?cF6+e6Agxp| z@(+PL)pyS$UfrYLUVklanZ~<3STk7rBTj?$N%?HO%*V3c$fnBqcUlMLNl}waQZo?E}T z@;=$KuC*1S)HOceU;Y?%OQQ%Y8uA5=CKwKPT)XdmUlX#Y?R?YKOX`iQ)S2TV@Qe1^ z1<#KfodpI3QzL5`G(p7(p+ZNJMw+*;6KR7n=w6I@Dx*))hLQ-TE{a?oZNVhW(aB=~!bZNKJ-ON6f53Sa6_A5wF13u!x@<>ML(qE(_i%bMgl-TXh1OjO4{T zLQEz3n?A(}sQ?1e2&ArN@cf%wl=~>nW>$td8Pm-k%1F|qD<`OZMIOxZLIR<25g4EL zQ*ve=b)yG$9D+0C<3pyB=;%xFeGFD3u(%SKki_0>XB~y2hY-kEJyMWEHPQvQnf^bRp`Im{q{w z5G6y@S?4v2?I7hfJlBL(%gFndiO(V>*yEiqBy4qHU$1`2wZB^H19T>;ICRCde{d4r z+Lt`Y>VnL172+jHTP0SB?ex_l;a|cy!M)`P-49h8ED3l}0>ZQ5b^=?1p*IAFR#Kf# zRi^akY|8B*J)Plq&bewb!|Hp`Vj(`{_p!Lqg*+2+jn8%jfp3e+F)*rDO& z{Mz>7DE~WUmHm%hahuJMi1*OjXnadjoG=f=R*-sqYM{B-GNF9jIb_p7U+J=dDFw4!>J6-4y>R6m65)eArnigAx|C~YW< zNw^ZW+G;pUQHe@y0tJSnivgsHitOyYVp zAN%{11YBWhga<7-d3>VOlQcZJ&P**l7{(6S20)|8T`9`Pm@i*PZ-0wUnfI~V3JiA$ zJtG}+1${n)Y~M02$tPWVGo>9Ra`nv{hwV|JXcBXl+~|017AdQ<=V}B{qja$E)Vz|CR7u7=&mT$Mj>KN`h)z5a<=N?XOq2z-{cf zYM8ho`NlG><<+LIFe|OPyVGdupnwE+t_7Qc3emA-3j3D0R*lM1WObVv?K3$SHm=WP z8CbLtVkivnR|BIsOFA_omIMPDFna3F?B?xjgpV<73ffn#_v)J?eaboyW)sjV$mnYA znm&OG90Cd{_}E2%CC^tU>)I1Jam_yV#igAAi{fR`u3_R=(QnCCCVy*7aY7#U*HT!_M*AZE1pirC1;!rD(36qje_iUzxKkB zxktNYGyGf&hP*`mO#R*J|Z{mt9MQQL8j-IfC7cO zy{WGpJm26sZ%2Yr=y12^zdq+Y`rIRF7{5B^x=(*PsTuqFY{B#3`g}R;pnZN-7YrBP z;v~lVJ8T649{w-Z+4KL$OD2#KhMM)Cym8_uWm<^<9@GUfuf41;-KT#J9|U&3y|>Os zAG45o9{02mmW|)bAR_L0;|g_7>cPv#irIRU+tR|_8etK`8^6;a0m$RI4_?hx2mz;J z{or%(&v7I=(L}79Hz*`6FRIu2h0kR)1V%OQ(mF2FK4<{P`f#pcS$lJGgO0z3`&+7T zIm&-}vDOS>D-s-fvCjEV38I;Dt$c<7{!eWpu60~(H)iT- z76t^G&BC?0jBbUm1`NS$v~V*x?P2rT`6v0?1Wp>BlMZB3+mqrv_f?jgR8mSmWTt6^ z5MuU3Pn1GDKpU)0Ymw$e2URi%APan+?y$t&u;fN(2At-cyyS(Lk{5+fdD^jW!-Q}x z@`UxY)xx!WA8&e1o1Ilm7RX6GZZR5%8R{f&fowe!*`&XhnStI3&X)GE?_ad{S|-5sLRJBmtbxA~qWLb2Jn5 z>DGEzNaeJtEVK+kp)S!tJY%CBBW^ieows9yUt87jlRlv7EB|PEDbp`Ddyj}n)oOo? z9XpYl=aXzX%Ni4v4J5BKNb026%_?%|5aAUKoXNtEpRs1MV!%_(r0vn@}nxv{9}*nuOP*rJmfX9SA!WcUhBz$-vBL++Rg zI2jg7$=7OP-)eQ6F<>7$5s9R73Ldl_{-!!ePKzL#Idg^W;Waq}#{D2C<*TCC({7ts0@R zn0Vcd?r_!Jir_@MEO(HUjZo?nJ1T$gfSWZI48%o#2^|clf&YRxn_-`@gFhU)B0kvc z`KEe%f6WuE=5lVqM-r#1-(6^3b)^Kg;M+c`R=ri)#1q)yU4n;&jIa{-$hAqJzEtwM zU4JO0z<>5M=zX`515__(yN7-=wO4QkT4AN`a^65k5M+PJMC8+o(X{frH{ZTB1#wobqH9)Bvdj z2g#>Dr_*m9886VVw&)bz@zBf_9wdB_A=xcz9)i-geZvOccZQc4e4y zogge)lWA$XstsbJcA=9eoX=As@g!cSBr#Fz@U*51TJ>;~zmUyRdO&B6fWEA4irw?` zfWgn>6nFz6!PmQhL%{O)rBi4CS^zDO_CGo^wu?O@@gUMi)K|Rt^pb%Ogm6Ax?3qdg zG7H>8jXbG<8*qo6bUjmd+H+Tj#L_>&{}h&Wy~&rP7mgw?AK!PcDxti&Vt!KE9&}D( zLEJLA0mxUu?UTcBHIl2s{4AkA&{Wx1W4PAAkmj)^Q{*+-7qWM+oa`R4Yfn~xDAwiS zjRPKRsa*>+IaJsc{QFHWHt%6|AQ4jV1l6Df&ZIw=e;P@ zp24|Rz9&Wpgun>0V?Xb ztBrb+z1c!wa)evv%A6}58cV7xr!GVANg5W3hEUcFT%lfOtJc!)JTS^dZ_7Y`PAlGg zh32bID(2W<&)&$~n8{6}o+HX9wU{r{4Kq^DN<>G6c zyO8XsiNXyPl<6QX6n#49ks;i?JhQ~vxo{l%mJ=qe3JdgcWAd?Dw1&mQS;%_?&tLEs z%UV!0t0miG)fBRl8xYM~ggiDgEkRH6$|)?wNt_h4eL{-m8{K^ymL|7#nP8=__x*(m zx!>*%RlqxN2BChBv^b>;zHmy9rZDaC11`C-;xzka{t2oa3GBlxDr$^9d)cN?^-57Y z94;Lj2zp2T}C@Qs%rX{As#aP#a zs*BZVUXJt(818b>S|VjITsQ^S;Ga6b(Ev06#P5bdkN?#9o!4my;@0@v#LbiZ&cIu6 zHrs#JL;e?K`Fk??=Lzv|J!JczRiCHs^^miy-R-_Y|JJ+sKab-4pB`nU)Bm3z<=+ke z{I^GmIYVmm{VmNbN@;YDNi-ABJHBF}{rxEz0J;FIyQhEv{87mMLmlm3)jFi3&lSp| zw;c-yf*u)uQ6qER{POoTnD-uP1MuHG)WM$~isv8yly8jY&sS5Smwfw>)xz%gt6|FH zTok%O6a$~4f_qtrSiau9n#3U5c=m9yI0_CsT0+^qj12#?&+gFi*e9rRlQK=;)Qb)C z028CMq~a@ify4HIGbVclaJ1)^8I&NcZGR@; zZmcl@bVFM56ffrtQ0pMy@Kfls{iR8cjQ0EUq~-(0acMh(2+hf?pbczxp#cJ^G?@-X-`+dLBIm5JQf{IH9N7vm0r&?Z_%xlX*KuvIaWq z2A+;nv&#M+o7$9FK4&}s&?No6e!~MDE*$lDb@BTvyzw7biTzT5I;2s|#yVhEdkp){2FGW>fD%ZeDZVp6cSLfMu>kpxeF zgnoqQhVoLG8+3pvz`1^u3eT`W>yYAa|%|`or zN$q-r47V@qMg`2ER{1MC(*liKD#=ABQ`*l`XN$hCez_d&DSC1ke3;K`7b`pmkC7W4 z@LBK8j0bJ#xd}Li4hIP0x*SlOpX~nlIA}jXb3CPu{cb{VS(-%7>L{sbDUiF;Ez{&P zW6_vMGNZ9NUbq@q&JEMKcR7rTVBJvbyExGzpkz{%g7#y;cD}5!`1Y(Z63TI`%Gm{o zjGpt1Y)t>Gfn?26J|OwG*OhvUG`YE!KpYpWuE%Jv4hvHgDJ@hu7z!UT`VzsioNT7B zZZ7tUD$zX6j$++dGzkA0t(E=MWkHtL9EoMN-#?#oGo2GG|1Z)X`5&H&?jIwt|2XCc zUa>1(p_qU92*#UVyfD+gSnxdgl{<{#Qy{hCAD$|<3GI2N_2^<(A~N5q-)~PvEROSx zFA>h^k9>ts#pld?f6Qmm(f||nFsV3<2^(c*V6n=_NGjIaBzU2CM7)?UKHHr+_p1;~ zspbw(f{U|(33r%YWF>;3S9U^R8u1}cMk8`` z3pt14E3&a<;$MYD~VDyH8ET!3Pw7^(U#~4LNkEwXk!?=$eUtLOm%9I$yRU<|m z;k>>sautVGoEc1DHl=A#F#%ISW~4C5gPDkJLA+i!ZwtSD8$PC)DY08Bc3>X)NS5-o zjg7FHv)ZG=EQ+v{57;WC(z0*d8X_na`v}};W_#8si zc|tE@NB1pkgdKS!Ow&7N|18#K>(X*XGbI2Z0oJk_Awi~~w_Jj_J=z|&ioJYD=TF+S zog{=B2oSa+Z(*7@GGlK@32xmeQV-z(zfX*O@a5cqv%j4*&h28SFc+r8k1`npB0-}> z)4s^Z!#BU&P7PKL-BB~sk#brEJg|YGhz<+#0V3b>&8DQfk%$|sg;+xhAr|C4#Oi)W zlWPVslh)fRY>YbiDt?zd3wf|2vt(_xQ*?yzmO4MIFz$Ezu}!e)AMMA)$|wwF6-|Ha zWcsNdG4+}kT8X^!m{ql`=@Wfah)2V2-owa-DPh7zR5H^Tr1A>kw9S@1>5nRXc3v?k z>9G2d?{h4Yk@SoXvPR;brGZOeO8mE(W2NO8Xu=+GYA)^NRc(j#YHSLd6p0S4MEjJ5 za7St}F3~2x{{1)QZz-o5kCCG7L`IwD>?F&~{b21S*A#Zvn7nrh`@?44vNyXL?~Xii zc!^jH3{QPItetoyXUB9mS8uj->Xa7 z>Jutr3=Y)uw1)ygsccLIzS@HLYWIsv=4{I=^Ix-3zC>7QD#y?alDcB5gGk(N_fJ8x zeskXtTE8ST=HSQ)7pq;g&gCmdBr+G1^4I))gO?P3!$3(l#QW(3n9qB$$a;HzF6hj- z%X)z({(hY~K}Rx*D%1a6HGCv;mn`)f)-Az4?s3HrQl`kk25u ztY5$*OoC7OOgy2GUg`*~j7R`C)PupvzU#&E+INt^)@78%sp}{9RH(p1v>r!2Om!5x zz!ouvhv;WDs81$H(2Uq5ghUc34%NC}lADZJfZt1^v=j-IJXi)c{aVSp9V8HKYv-hE z5;0l@#G(0Kfbhb#GBKH@xRE5R_+O^cVbc&ZKg3~omE19$es%F496lt1{4nu^6@Mz0u zxLc!(O}joEF`PcOO30zOnPHnZPM3*P{Gt|R$m(kX-RUzjt`LUXT2GI0tqql8SvQ^$ zS$!X+zWxGzQ4IKF~mdjK2je(+cTFQu*$x_-IeiM$+%fIqE1TKTcBH z5YXz)E}JlZ8_9-QxLXQ^*XCZWmQzgbHe9$e@uqfRn>&5XpnW{0l{8H2&-D>gfor-o zovy+@hFE=FX1b#Tj?C#4^7A-drD`gK(fzdCuv11WirDR!=q@`-7lEiPLicytNCv@Zz!a!Oy3UWk}U)1Z5uS9nKpX zowfE{ewXUD7TE{`B>)%&tWlx zGz<#(XY7jpUw3NUWrC4^B|cQQ0dg0%E1YovJV3+04}Yg$=6zAD$~cKTzrzt@GTUe6 zcf)BA*Sm1!#xN9ej>FU&9?v8fBpD@>C6i<{0ep$7*N$=Y4qk!Qu5rH^N4!91T!d$! zsZ!V9$vCj^qTX7m+3UxYpN%8*em({$eVmv#`0E7)u2@n49f;||SbsQs@(Z>62ak`w@Ob$SKX1>a@S32dM){Y|wtZDt z_Ojc2cn_CHqGU;5Ct2nW1fwp9A}F7V^%@9zA~dc}S+;k!`{zVsDmY=lGo;e&Z%VFq zD?;OYqzK9nOB2Jwl8bMUHyKq`=9NjJt;1QYX5_jq3FqxlG^8fHK^NzWc-RJ^J&rg@ ziCSK3>=r8#-ldz1fYS;v8ZF$qaYmd#CY&zv1&%pFNP2b+mtKliMDxv)=gDhvVuB&} zlP}Kcu}A@0G^P>KEhO<;n~6!@s6&m(Qm%;9jns#Nz2w3&7ANFsc8d+FDZ~-8o0-(a z&+!81Ld3SRDL;SS%K0^m{@04YvOo7GgwZ>}-m(9s`2zeSD@6d1?oRyCeX7gi^uMIK zpbJd2e}*p%|I12A@etI117AX~c#lZC{|;X?%+e001cI8~e!~}$Y?L^P-zE2awz$Wy z@6+`}4)o_zKKBJa-UIB+$0O8t@FmBB>acph)9~eIl7+`gyMX&z!bbsk4x#}HnvdR{ z+83(@H9|h&F$Q%ShNFfjvEJphF(FSW-)^&FqG9C-nZn`cckqR7cYUAuQmjDgR*ezNsioZ&DEwuWO`i zulT}}hg!SlXmKuN&=O$-?ZoZWHv&xF>d@fXwd4d0d?!4s_6ve8&W>;+N|FwiokDev z#-v36$rHB+@kp2XA>lu|`Y8hp`#!6w*>~M%GXg^V}*j9Yndc^+%+nyqoVI*Hbg< z7IDeNJAcBLf|vLCu`F51L^-t^-Ua_dq~w3ekNwBJ2-kk&TD<*!)7JgH$X@;Ke#>Dm zmlL$UR^y=Ud?VtZ{nyN_3eO2TH=kTGQLb$#3fd!QKW>g@uw{{eoo*qHgnOkgIZiRS z3!k*q`7_)_Sq;1^{sycHhda2Ka=&6P?B!t#;`{7Iby+o_^iytwr{h$4;|mDQe=3{L z%Co3ngq$UwtD(_cj?&}sD}xjyb#BF*rN-gas3?JR9}jQqQPgvs?79WV}pxv$vV6pYzp3ylC-v;2Yeu~RV1J#K)t28u9Hem6#j=tLE zO$4+oBus2i+ILgF5mWDo<*%o_n3CM!VryLv!1|06TfY~gzh!P5$@^)P$8-4GE9#E9 zWAp>oxymL;+N;IJk*tVMF-o0cH#ld!i3~zg%$g}?tZ8QTt!QCOrII1CYXh-^#i4d= z&Z`6WHUszeA3O7d4Hb(Wa1aiya%1IkcPKH}vzv;l)xJn3xpPmKs`bG{F`b7tPD>m9 z2m!W<*GDIkX&9YkP;q$O?nI?b&Ad@l)g~~HJlW$?w05th1&i^h?@889K=E6oN=l0G z{$>*zI-3{46?_%e=|P+Mrb+)tWCk_T0?*>wDy`73i1*pyq}zdol2Uv+Sj{fY@oOW; zYQDSsuQ+#%MZIGMs}!m3KaUTGmjbOS*&^YfUh5@^*ra-6K2(~%%s~2fSQk$DjjK_x zrN8I0KW3a?!{4S zAI^wTPlu^R$=by+M>J55P4ha6Y<1!6w((~NqP|SX)BDt3X2{aKD6=qr>K}*oFvrB3q{P*JFBc}X_w~N!aT|+|7?zpF8!K-@)xhC}eRm0( zhbUuP*#2E4Ci8pNtpWa*M{mriFMtRXevNqO>{xV_?L7LqEavM-TkTdp} z?foxDr!HlorpA6R5)-bSCYB>*Eyy>^UR{9(8Jk8_^;m)Ld0SE*tVVXD*3R7&2anIDm zzThI?bHWY6&0v75JV1#^oCI|d$+a{e|5#so`yDBao7WK1bq9q3fxY}a16D2M1gc0I z2_F5f%j>*#U2jYd2tA6XGELC9$NYSZj;FGj?qVP}&4i)*<{3(-V*^L-yoo5o6je(o`}0}?=dD0E`g!JBBDy-CU~06vc6_cr ztan>INz9i1GZ{EkT}d*R@Ux}?&carnr7IcLh}XG|(6IOmLO~{l9vAir$OkGrTPbCk zxI9Sm-D|LF+C$EryxeD6wnc@CGF0!2?k868F|tR~zw$l=`D~qn2n;$3->sm&=aFvr`%W+ItUb``rGg&!EY5W`0lcSZTfKPM)S+Om$MELBTVehcWQ& zC)3UKqcgDy(_LaKr3!#(RB-g0$w*Tl7VCq+RVvi4otPkET)s|zv1ymLczco64$G*@ zjM&z#&U^~E^ao83y5uT)E>NPYzEGQTNA)vAw??+`lbO!~CPbEnNDVx>C;v}{>?Dy=htN4;WlxHf|llWfoF5sA1#pV>h# zawiF$avi4xY|~r>P=HdxlUTOD+(S{i6Yxxw{hhyRF?zmIwl#U?gIT{=d8Abj@yA{I zJ`v{C>QVR7k3?vd{zI_lTIvF^JmY}!FOOWGwkDCy1YoKyk^fpsf1hXwcE)IF!#KMl zAVFzrv>oG<_kN`d*VBHn?QyZj6t0AaFNeCdc=rp1J93Zk%}gWD6N&w{w-?^Me0#Z; z;B)KU@T)pty|w9uL-1Y53G*rwX+;>Y2m`cu>ya^pOHJhMBP|fz6JRjzj?( zdO>T+dk6^*OFM2$S-H!E#90knO3->Qw?6a9% zh-samTLePUpOA(!KsS3brJ?%RO{dBTO|*t@-lV@Fr1mWjdRC(FZwRTpd(Poi?NpDI zL~>fda%%lKt!>}rBau6-bZA!UCJqfm*3xlfi9kV7M#p`a6oE7qy|TOV+vJ%Q^x<>T z$4fhq<-ewJ32j?F$*{QnbOyL9V;PHsSjqV7q3cgsQoLkmB-3~~ihWhur_OT7;4x7U2@r#2B7&FjW+ehM z&1`w_y}WFYw*3wNslj{DWp;4DjPc`SO9iVUH8|UjSETckB=<>xCK08Sp1EBSt5l)7 z4XG3*Ep*&3m!ly+nH)cM(kcjR%_QwD`hbbPC)rM3R_HOSb!ONff#vHg9Df9sYbF>M zbI5^`g}3rKpU;q}CwZT86#iQOGLrc^?=@%fY8*10F{^T5Ypyt`xp=d@Zi9_WUi-{o zMyG?-B3@IGg=;T#^rhCg_^?#Tq|#k$bE#>Of-bi|s7>>a>!Dx`6@8zVrVhZuV(AJ& z7etF(>naK8hC;t)a9xe!qn_U%c~BnZ9L9!?0Lb`I~1WJq_}F22qYN(z|f7sXEr+9&4ILY(P(h1sx@KKXLzu9spBQ^;YT~sLlBxsCU-+< z2I-P-nU3$6+OAY=eg>mP`Bh~OQ?cfz(~*Ii*L(uf=h=BhMiQkzzIk< zG5QUsSk{bYUlpzs6Z4LA$a(`^)Yspza}Zfh9S#WqB*2XO9fSxo?#hJ$hI3t{_fac1 zePH^24%4D5TA$+v2-W;D882oQrs-111Yr;yy=qElYEm+k{WHF&b0?k&pEOZBf>iwl z0Uy40{J>U8@HJCEFB0k%K%D9y%Ig&nukq~rA=#q?}@m?;`EjWKB-}~mFMEQOWwX2K&;ToA^W;WK`l!S0Q;KP1mMx91Z zx*rLV|4GfNl&^XSIN1#F@^XkvxnVi8oThyVe+iH(;1Fd>(AR$@!1Ct(I3Cz6Ta4XVE9!)NXDul-pzaBf;GF% zSpcexHuyl$=`kn**z`}{#EZXZ(p@}&AD-*}#O zv0wKNO)2@#Sc%&Y#?MNPRr;T`iJ8uJ8mmVPQFl$+-U(H;rNm2<3*DSq!7A^VBTRm3 z9T_!6=nt|Qkh;E1sCm`EoIQ4?J&({o{!F7?HC~|+$c4nzVw?D0>8EL;BA%;N3d&=x z_->s2Juee20Ew~K)_Qfa^`pdesJ-p_;&62^gTUp<+xhqEGaVJKEwU+B3d}mDQ%41( zz7NnE)&0?h(U$;g&x7WJ-V#pI%Vmg?WJ%vu1uX?Xu)0w9#eM>12&5l4U4mZBjjn{8 zHK9)E4&`}8D~RAPMM#KAK}W^4z{3Dy(nzfpN%NxM><J&} zM2E1$^a>cp8Dem!CvnK=VUetuw&uexq=bcF{_%#-Dtn|AX3iOsaF#8J(HQ{68<`Xq z%Ssw7^REzu1_I6NVrSb|)EV?w-K^+=)mSKG6sI}E;v^K6mhl7@`fw0fA)hVpInmlD zvH)~s@pdM|n)N8pbBk#RF)NRvbuL_8=0&AXwLx@AheebX`qN*;0o8Wqgz1tOf-nS# z7eCXo2qRHsFE^Z$D=O-7CpOBbKfJ(`IK6O)bEz1>Y5FK!wpFJerXv;Fp`BLf`h2@M z$qwd6O7IJ<$n`p5nnQRY{w!CT!HS(;OcUiH*H$w2dGUJdM`6IK)FXVb)n`UL5_SX% z#(T7yAGD{tfJrNbo_;K>^Ri0P)d+Cr)_(s(MA2vN6<)2zLbp0fh1N#{w6{dT=-;g1 z!RQ7L28KV8+((3vw${WBr#v1&(fKg1FZS(k_t>d#KUoBU*AhwNEe z^gv?F1adj%IxS^wLq*a_8GWNyeymv1O=G%yHLiZp{CKsIx*IPnnc?~Hi6;VA>(`dx zd`YuaVZ(m&TU%An=3N6kuRY24wru8FtQRURi%GgKz%aNk);=Ew9M1!W622 zYa@x1T_N&7aAb!Fr6Zs*JBayBC*Wmv2Z{HP50^6L>${~8^2?Zg z%O~at&;Bytg~MiYh}I|>PWY++bG17m0|fS-010m=Oh@2->o&Kd3QfeoW)4s}CPJTI z3eiX%`iRM_yG0U~>x34<03$KuFar0P9!5s1ixROt0#U9;eSOo>7)WUZ3lXH1iz|s$ zndYwTU0`vIdz@=HAHS9(&xoBU+8zF@l|5>gUK7%io4(OYB)Tac;~zu0Ak;$il;sJk z>q?P=FB6;YARKBl`ZRKUjsBObOs)`!13~O}JbruMakjqM&(Kh73zQ3Bnbok|L7T z<;Fp8Gg^XOhSFB7VWwM(Z76C{sBNWRX+!PWdI@G8{YdvEpGE(0`$s;}(Z_kQ2B)4( zYp!Q!I8g-#z=nMmU#{bf7kxoYNYb4OSCw8^LVD}f=>iEWX}s1^)2R7)-=`AAbQ~ff zi}3)g>Qgo}FeAMJy4=P#nTtON5eQO!3Vt=w*J0FcGJlfYy;hE78A>EiJljP8fG^s| zI3$_ygQ$@re7;{QDwG>!b!k(sL9diV@LLR-DQs`ce!D4JY3Jg zPmpZ36*+89`6x#PJG6{9%=LOv;uj3vL||pK=nQw^hq0?bSqfOsCppZ5C>N{bBn$Q3 zy0C01Qk5`bU&XH6b%+twAs`g=S|mLY68B{xOPbV<-IQbA+L-T*!PS^A>b9F&3S|#@ z2`B8he6;%!H}i3BU`{}>rj#6K@klmE^p`5Q*zR3VqK)nAB%mq<46ceU0%j*>e>qK5 zk@1VFNb@B}vysfzHx7vSFCa?4nvJkqj_Tc-mx*~8Kb*(ZRcNbBg;%{UJLOI>NSz*; zYVk=Tl-?iHGj`N-P>h5R0U*Rn9h+&~VQJ=lZaiO`pW-4jMyaR(APRbmLD0jkV{V3< z7v1Cdln|CnO75xy#)~-d|3+4;uf<_`t%YEs$~E{Z&}TtfCH+d!HT@?`u-mBEs`BYK zHAc%h?HEprgnB$x(#dhH6xRBFT(q(h(~1@9mmE{C%HAq7k)1ALUDB}Bnhf2R9)DS|kF7I5${Cp3V?fBzDIF*o=RX2j@H z74{|SH&Up*v|sxY$ZJ^s;8NhleWI^xE0*i3ayps+&I3g)GleX2!-27|+QJ_NV#LC4 zx13sKqF)u^vLXZax)h5?IO!dSqRm$W2%Y z%MfFu1nXPc7UjQ^)B5JK(q>WN4v5q(i4ysRXX-&@jCb^ns)%bTaH&L7;H!DaUhXOsIA!?mJ9@ zsIk?hFcA|$Sn#5PmbUfXC8!^8Mo(nUe0`L#8jbJX_rmsWiueoUd@N9zB((4%MnTyl zOif2ys4owCetDuw1YrhZGHT1j$C;Y+hNZ;AzBNT^1nI}E#k-zgMJJ)Rp|Zr)hPl7U z2zAAZ4O8V$wlQ$R522FLmwA$%mm|KDZj(G3WLpsW#2~gXGu|M+s4)K@Tz(}wk_=kv ziV_)FM)_eX<@&k*O8m@&=e1C`G-bKIp&X=jc48w)Rs%Ty*#N@T41R~{Y$on^@-VS z-iw>%UL4lerOIx{Q?AGsz+<^G$tkMRZSjsqmv=b4sfU%6yp4=5nL6W^`LC!7ehqdr zb^uRs8u*}9xnu-3L~T60N#*^FkM}D3ty_;cBuO`K=IcQubFVTw8i|f^9wxdB(m17p z-K2%iJQ#Xa*}R>G1(B{w6IzSTI(A&hAN8$sAs@DQEkzu90WS~N2e1TM@?a+^B}MEN zv#w0ub^SN>=I7LSEt%(1$kE!I#06nvuT&oX_XX1)4TCYkC?*TrvL1_Bmu+|{F4eiU z93QBLvXSIM?SjbVaO!dqw^x}7!JF4URZQM-)1I+MA$>OUH? zy_`z*19XP|TN?eb(@gu{XmoN}qQ4y<;s5?-N}%9>>?KsFng7;p2>fAl?1!{}QU9Zv z9g#@Ik2gYxlV9F(GxR=))VMzl#F{Xgu4d~^6A@2;qaRt==1+@8tDRVF`@KLdi--kF zaCxlop87Y4T33#Eq_4B?&9@Dxu}Z*K8?4XN_t#GAV)^~ev(?mntYZ<86+IQ8mo%u{ zk%#cgz@MBGPc$n`gM)rj?u#8(^*7i*@|jDWhRrr*+CHRxe!c1jn}zlwv3CLf z^70G%Pvb?+pzTm~j?_CCW*9 zrty^VRjlke>Pk+a2oaVPZ##5JiFZe3N6jU@7egoZd2TYNQeiPR3_jbgNSv@ZV31!1 zIA>R{pt9o2m1u|Xb7R7KL=D@{4c11)t`@8j3Y-R_TJ<5{*_|7nkiMT)Njchr@jYQY zhVc=df-6dPR;-$I;Ei4XUuIkC;_|ArhrA9L(yY2XIXtj|+zHjcyMl709Z z-=E=GCY?o+tLnZ`Vim4{le)S&NKT||j+QcQPN})njPv>?Y`%4WDywTYdoefd4=kSFn2CeIsA;0PKis#H0B{wtOT z%O9`{yU3akgR!wmaINLLsb&s^{fXuAo%*|Je;vMqkSKsP9EsudjMzX7Npq-%O%H4F z`!JRFD;`g6BsmO@!u7splQr~ze&~G^0mIFq=#%diNIrURqntxE+21SNeDuNjea?eT z`MxJJM<2aX`_WGl`~EJvzxT;u{r4BffWN$u+d}@?W{8kxbNx}P75;Q+A>4)POC8Z* zb&tPVqv2f(c})H{|HrllWBn)e0~sLg-ak$4QQaB;YK=S}nTxe-yGlUB*kexsgNP-f5w_hh@v1e z2b5K-A68v;nKJ(%&nk)#V|{j*4nfD=TmuSb1+5xmT45Q#RP^{V76+vvoFfEXo*+DA z1;H7v^g?T?4&Od+t*5$STLdM_J*}PhP@Y5Ch~#H%v`h;MmvvJ;;&2vY&)W~vS+zO`d=*StW*^|tY&n*=`#Pxo^%2Hx02XiKvCj9! z{nn-KV?0RBu#_(MXsbL^iJL2ii%}$(n#8TtSFALnvs&H9V-s|at(oUI%sFl6Sm?`7 z&uc8@PKS9NzwUM6kRaQS(MKK;Gw;uRaXlw=7~<&>l;^JzhftHGNS#n`YdrTVap%ev zl6TxId$889CVq+|aPA{I-_H24P71PDa!yC;$6P~wmA+8A^Yx5-GY)t`DM>Ixg4RZg3oSx zhk$Pk{Vr)a3L~$>ey0-M1h4V1!1aL?38Wt2ZS(7s86VyG&`08yiCDKT&zvn@->#FJ z#7PWEoLXEIZ|Ba7#}f9d8(I;%5s;$88&ZJ;B1-^C%ya8y6MzW@h5Kxtu#pSqd{>a` zeNNfi^Lh8>@Q<$x=ikwPiA|F@sFPzdlAsy;A+F7S@tr48)C*w6O*X9;vV&DW;{NmH zMP%q8rHlDnt$pjYI7J41CTC~De(I0K>y4z62ZAH8Kpc+AOjn$`ko6R3md^YKh4cxF z=iTFt>ES4p&Ot!g&^0(ATxW?gHB>-GKm9ejjk=w#e-jJzjPeLz6+{IC5I1ngZxu)x z7feaGS)pb|gXg8ttW(=uRne4fkmx(5S;DUia>#ElKKZ|TsW+^Bp@x8x%7|1>>(#QB zXe7jrOea^MB(0UaA6}?qdQQq&QkKTI{EJ;DB#Ko%ML!NFzF;jQEV#+b@lkGR1#`hoZ#kz3i~HJdyiz3C@)Ti=kV4x z%%E(~(n4QH`OuU6m@?No8Q`igUK-818c@%#He+==0R{YbowB;Bs#?uk;ats z$(kyeIXp`klVXi0fJ}wFtCpD0<#)T9rnHb0A)|-;OG=j&?!PXj?%=ueR?qjj5h<))kM(lh=@qC68{)73g+{zb&lELGWk{I}+M> zl+HGoy*2BE1EKRR?KAOXlp6GUjc^e0*XZT)`ELDKY?H5gg>bOjCRJ}_7TtKb$}8ao z?uiT%+{m9H5`1Z-;SV}k>lt_zeQV<{np%|>K80|dQ94>Td|aR z_RHkEQKo$hCFb=BTCgxCUoqPTYky1AMH}cE4f!teLU+4`LD&)by`BG%n=to!0F0uv zets!>eQoLYf&|zGA(J9$U$b5^C%^g)10MmcP0cZ6*%Kv5>sU_6K8P=5)H0Z_Uy0CmFPweP_a6z_=nA6}P%_)KD0#;1AqgqrA`P&LtiGBj z0+Tr>R$PE1G9Jw(Y(2pcdx<3xCLn5+XmQuSo`hAEv=Odw`zk8RCj5#u(keV`Elr)Z ziA~b+7%L}3_Mz>td?%*eYHARXty*Sy{1kgy)UgOho`;V2i)e3+=KOpis;h{cR4MG8 zcX@ZO?LH>>Hro~E(sDPgC5EAC`z1Gyi2S*ePGy!{q$waPFu@h|Iry`17S}T_J#%DJ0g*3Gyu& z#c8HgqqCdA_#|KneoZr(>Zt;eTg?16;5@9-MRQ1LB*~)qM zTkx_%B`P`D?YO(ue4s61{`_oxg=vBE2_vrPBc{m;GXgpjstGpzM02EKy7JYsc@oHR z{zvPHiJYG^<&!(P=2Pj;IL(r0^LiH3?`U2S9Ik&7hE^UF@)cF*MobPz4x#hB{N?bq zuOx=c?>5>?VayQ2WWAXsJf7QtdBM&#^EzrD1NR2myhB(ax0&llX!^aqO<{9ql1+IQ z2L}`PLUOJTma<%djt|F32NH0rCmKp*ZKk?JHEk~0VVAS%*%+9C1A;FtLnXu>DRlc&kEfi&8K4R;Rs>+S;m^D*;V5;n*TNmQ zVb!u9AUGUcw_wLb=5g#DNw`Q`1@=c3aEZt|=qzGy#igsG*F`Gh*h9%o6le460yZin z(`+`5#BtRaQlKir4%O-p^B)vJJD^#eX_WVFi0x0^h)I?BBq5;m*TH-NgxH?P!|YKW zshDVvb_klks&;KULi) zdH0-D3}=!Z-u1n?44F^3CAK@)OoQ{PuhI1ZxpAQ5`p&#hY`JOu8xW2P)p>k7>B~Eo z5k8N-H9`VeAYKzw=JN+PY_Ak-C2RG}LW%b$^eRG-s8_rQXqoL{+b}+ErmGz11BPZ{ zslCG<5~s&wpGjxQ!GHfT!2nF4{Qq&X@PEgP{-0(Cf72XueKiGjVw3FFaewy#4QG=U z3YSgalcnNK6J-!T(BVp(nDfD0hv`N1H!-qGvqbM}{Qg9dqvbx1&V}a3(?7*4?{1_P zlt&N5IF*aJx4U#7ZT?_%5)wuH!rTR+`;Oy9G@tjH;yq~feYw<0G>Z^lpsJjON39%S z;^TjxTNdZUpsn0*o9zerM5Nfi^NRT5&xD2zN6z2Ay*mE6g|SeK_%-3YvQ~xjHL=}S za=hXL<_&&J>+Q@Y&ndnS!2`BPU?r&+z~B^*e$j8+-#y4)ZtV=>Br8}+rVo^wN+l$w z^*45E<98Vhq)nfINvA1qto-hLK%I9S*#%_xT@D^%z8q}+p;Jgc2huKrZsBdiqNDpl zE}v~pl1un?u4#>e?RuP^RhU8SfbD&>*?qjIS{fgm(vdxuH7PZzbHmSwOpxNO5&l)a zmgD!wYO-&vCO0!Ylk4&&-|iO;>{zkyvJcVNRno~R1evivFuo#ouUjk z{My2Ntu(&E^z{7V9m&vW@3<88;s8$jLKMjzJDr&CzamN-JU(yd)g0Sx7q&gKw3F=o zZm_4*!)ZTM9EWeQTdgT0Mps+3pRArg_VBEvdSHhGTs5zgvR8msXYf#~>TY$YX}T;W zPOsRGdK0iSz%`!H*;`VfJ9B4978oSDQ-RR_X+Y6sr5|3Lf%|)XfW}Zgaru{39uUBN z^cw?<0d)T@eGxqcRgFDzSt{uMw`4z@Xz?cp&1L`jTXNjt@VVw9q8T*F93V%NkOa;b z3)D8%uy2VXXm`gAKN2J|fLyygM!-q!(YaqqpPDPF7bAEqs&c>ZdF~nU%UU0Q*y2J| z6zB^ao<3$o!2~nVbBFY=vCjcs(9@si$z+BOnn9TtTBGSzIVcc82*zph_-Z}R6)jv> z+i`GIfvzTU_DjeRVyM zFPV8qUV)E6W6=`5Psx@bAA0=`F7w*brM5|a_<{-f!A`8V0vgU|J)ajLjxwk}0x_@9 zCW1l!GRq-!?!C+JNT|T@P?Ruv#!z(k{1pgowqB%p2lHn3^G7;*s~_kILsu2-e@+o9 ziezj3hQ9%NKcbikinGt;exG*Qbd$($Hf`Fh2GqPSB82h)%E$ zptSlZQuKN~iOxJ}Elwc(Dmn#HoVzTi_6x+C>eBprRouvgoLTZnh|)Uwz1{ZeM|sAp zt$VE_hiQFn^mcx-$G~<$y4}q8Hwk1UJw86zL1vPj^nEk(4>^$786O0XZR~SuY1b)2 ze0lQCN*MAK)re#ip_vNssi|FcvYj1=@~5)JT$N}XzXo|aEM^x=QP6;^2Ko=NGAdEO ztqt&$PpOB{Ocf6l#L18imO{p!P$Fm=S`T}1ZjP^eWeL%$=+ZG*aH6<88;1}6ykp@f z6C~vewJOZbu&2%4>bgH;>B3zkbap)~Rv>Qrc-wV2aI=7@r1TsU1%nTl^(`vO40Z>w z*x276&^htl1mF-7Mii#Hl37$rHZ*otV)3>{zZ~XGgw*z6;DsGY7^Wm;1z8^2_lkRn zoDX0*HNsFDfL=bQ(U#uNi3S?xI{E5f8eOB0vCh4te)=mN$uS&jkZNbDc zI*C5?5R92}g1o^Qw}S2^Lr4OtW4~EvJ2X<8EgsA9xJN!YUPZERc{kD;`^Zr`jhVgU z;k$hs>1(&sH=+{8whv2z`!Z|CG}mwpTLYO@)qIU=eyTsO_7ys?`z~iuSj&Lb!Q_?xD_GmA{{<0Dspe7w~`9(EWjB!j;i2NE+2wKmTiF5c3KfX_(I!h^A_PO8CK6 z>z^w%lpBY;o;+(CY;am^a|Q(l<4jd7cWY6D9k|ZzUo?d>DTO>q*wgLuPBBUQ+Enwr z;kBRH!fsRTb~Lmf#WJF4-LUdpCnNoN%8EvT6J6igst4~cis*1Fy1otA9cA#~aGQP&uG(;O{$eeFrg zRr{%J|m5ZVW92M1UsM(*@z;P(oNd|#WL99g9z4Z zqEE&A9C`3OY_;B_De0vslWG&iEz`Abx+z5m0kZ z%Ax^n)t(Se+-Elqfin{w4xq3R9v4Xb-rm0APQI5Ulk)cj+Ysm2gFnzx9|pC`fsGO2 zjD1?J?qDD;CZ8&ZNoeexc_2~(|AksjH$BW>Qi9+E`h?qQT;54rjH^1Jm4M_e0(3;? z3`~j)Np*uhO?W_vqN#+$b6)g!uy>ZcvynXcYBc$vYg%kG&sc8`Ys)%Xxn~yf4oc^Y zDqX*jHk_8TZ!aK;Lho-G4e{!8r*@X#1%Z9OUF=^Nq_uRoSO4|8=k*lO zdZMCN6rKL*>CS2tFS8a{;{?42@Uw8Qp*p=>N2I$&_DD&B(&1a9UZ99xPCP@4o((@yILNwr#5gUIHLT0sz6U>i`+H^fP1@WY;xy9!K zdG%c{Si&=GN^PM0Bb^k-92l0LdCoCuJ`+W=(xY-)o*!b+gZb-VkToCren`2^1mAU* zz1E=1NInwR8TI$NNp>LuC!~eJIq84;UU3e(PE3N~Xb}1^OjN$w#2}&tMfyjY^dR!k z;At5(&(kVpCwhTaMiB&yoL2x47ZMcjMHmjIR9ff~j)YA)1EyWy31r5H5^_SVmVanp zk@TlLQsjH|$*!L#@sQ}#>r$XLJ3TFnD;s^;Aw!iqG;yEsJu|8!V@GHJx=jL*+XjNu zzO=U5Kk;S6eVg^8g|Axqwi}RQE^1*J%=h;D^vB z$EpvdLiHG%z{dLDeCkOyq+El(1vuEb)93VCy%BsOhZdGH8nI3ub{r!X{>oaXDrSwg z8dT(pV;fw2`X$JgoKE0_n(ebGF48%Vd$` z70JI{RV%D5&SiRDq^LVE8sAL9qZG02sMZ`wj*Gy1s1{0T5}vW^--G&8Xpj^yn<0B6 z%5a(lvvZGZ46Q9lQ9@e2rd^-QNJvPOCp1u>%3RdE~wtF!W=UAwiAzei8eLE+gz0@r4`$AoseMR)=L4Tl<3PoNc%EY4padIgY z>qv>jWi&QKsnxe(w!UZef-FqRl98u!FY%VDlb-*?Q9eY`?cb`Y&DtF$3D6BdwnA%) zwRuc^pqf+pGWlB2BuS$v!@2gQ>J!^iCnl!ZL{4VR&Z|Ib6UaO{yH4}CxWDAn7NZ`( zzf)h!;mN!#Yw7oskDredMDM)@{v&9a1SQmMa=Mgdza?x63U}P&lIr@e&f?Zb>xCw zx8lC@X{ymxenGUmW%_(c`#6SLGn=jUUE|MFCU@)^)rs%g^a>+*7km+tZ3eS=&`11Oc^@c?; z;o&KU9F-CBn@aokrT>MnlZg%F0?4I3I4F_)=&FnbRH7p^|E3+8V+!m;5aEL97kH2? ziP~mg&I*K%T~Uca>x@el|k zRKg#@T>XHpUYiz*e-xUd#u^g)Ah;v+H)I9BwEht-!MLui{7jgVD$pAb&a9$iHO&-b z;8DaJujI~=r^<~V6dleLoVb=uAS0R|`6NJVO-fQ=jb2lJgM&4}vHdksGUMAOOIaj| z-B$X`Uszhk?juEd@sF_vHlssyp4cY7BU_f zUB0tkSMZQ)OdP=+V_(ts(f)Rjm@RX9;Rl70A5|Uqm1ugx( zl)MVjXrsG@AaTR;iXbv?W3@P;rJm0nUL!c&QyqwIK=f^ObDS7$=Bv5cLVlFieZ_be|x&Ggww&^OZd zkvLu1hgYAoBFYKl2=u?R4$zX~IPI3pp>FA0dIb6GBQR7O|4z8QLIprEArhK@r@R5b zE07cT=hBD~L;?8|m=fJQhG%t$fGHO=|4RivP4tJw2I^zyKlmq?@w)1ZoFCftI`h^p zlXL*F|LC9U270Hq)lmPfiU3tCS(!h%3>stR0ZrUaOUxl z2|ZagDt9>9nfE!|?+UPd6o2b}ptG#23NlP>K3uHui@Z`yPdzW<^=toSOuO`{s@F#= zQP38aRbn1YKqGaxdBFDlUF?g&eEoz>GF-}Ri!Gz;CHrv=m4QwWUs??<6h7rf;3xcl%8W1~Z(5LFd=>^lRu2BZr48 z=oI|vzyq&*HLdUB8xP)`3T-U`a4CXf#3wC1Gu~2e1&xT4P+Y))E0h@7rdUPwuvcZ` zh1r3`M75vaGQhzo9^fD~CKYBXHzndepbDWT7VU?2Jo7h%ejNsn$Mu5tkp-Q{3bGDjH?_p60T+Z)FbA1%i&drhRI-3p4N_#gv@Iu<_q)0|BRaR zxc=E@9!1>=P=8hw+?;`jU?5fo4`Lhx`&hxKq zjh=ljIP7L7h)YyURN#(F^imO2kD#oh%J|AMxi`{}N$8&-RXyj=jUNr0C*Va7TNON~bKnlXo$Rvf zqajqLAAI>(%EI0#TVBNpxpqE(i7l=tW%J5yTjDk73Gb4@i=XO=VXsK?2j9PXS-gtQ zFMRbQYF9k*MX;`ltA0*&c+}Ur=@?Y70w5amD@oSJNs0Q{ zm#>H56RnIC(ySMUH`Q6{gf|gfy3bPJk-bg69HhzTk9lx=*aYk$+ z&Z+((nxhwR16_o!lP5?m4`o>v!2gtaEYEkLL>@-5g1sR$gonJ8M$foFac~(n9O8$` zvc{q=FGJs)JHw8UPx>+p8<$_RdPycY9-%Ya9fsr#n%z7^k%YMo*@G%gFNBXH*3+`^ zE^NAaeMh9oS&N*M1{NojUorbo*h-Q|CG7-ZNTxRK`;e*+0TL5?&STBq%LUO&4qws( zn^h%$i4nfqt3og<=Z5=Z%7Vv4;2R%~_&>QU(BNWymc}Y}d(er+2RLfU5RLSSRc74p z@%jX5_K!(38dVL~s`v<%c_Rgc66ID2d~EDOb9CB$%7FC%2f$Ryszuxs69FFPn*3xgkDCs`DrP$E2ctzR&eOJ7W$p zd+T$JXPWG?vf*e^w(2=#K|c^f`PW9pqZ8Zdy9(MbCD(_}FwaIyW^fZ`yzMMvcgAzZ z1xuY$&GkR$QwcQO~PRe$fdf_NdrmjjHC{H;9Pj&4e444w>?ho9FLwqa8r{mr-u2;d$D zpa66M5C4rmyr(FsSBwo7>~bk_?C;@5}h7R|k_FIZ#W*)6I&BM9dfTp@r|NSVnU6$~eX4WAZDw87g6b zK%jI8wM2jvuqqnGu z5)#Hl??msti(aD)5ribVAkjM!q7&su-g@tS?_J+Net&<~`kZyvIs5FrpJzXNyFefX z53R_C`}w3|u_}q(LiZ|$I*oFWm>76v5Rb)e*!H{lgHNSd8jR-c4DPl@PM`l zkcQd#vBNXmL|9$$p`rvd=q|Mi(jpqDMgC%W%<6EC8s&esm;H|K%ukJ$B7n>~N}hR2 zdPbiiLVK8tc3fJ+?;0Xn%}SfnUo(vHWFeH7RxS(2l_oz(^`#un@}fQuR&a``O*@+= z9Bf-KE-yj+G|gjnLU=&A6Nhxzgjb{!DfK{ik3CAigL}mnOw=%#@?<$_#U#+TU?l@p zJc7)GVZl|O``LC{Wj{M#;LLGnq#)0YeUv0`JdIp-X>D{9kB~gWQe#%HC(+x2^epUX+o zLgH==GQoLi1QG8sqdepFPIw#^qII`4f?BFjmoNPxYhZ;XZnqeiv{C3s~Evm-Ke} zr)jp4IrSO3Ezt@~I4^Sr64Oza@T5)Ww8PK~yU8!j(rij*1c)#7D&Tf2n|pxK|)zXn9>5fQ5fUUiZIc7!eTy45DHUA zeAW0NBBK^TMp$9yqIXKO_==P=A=3a)t!osZhsXp{Jhri*3RevfNkZdalZl5DQX%LA z^mS_~UYK@?^P}(~t5}>c#E16(l`PD35UlD3q5bKqezBqO+H~MPW zgE2H@EuUecd#%8%3U9rT##wf~2%e>|o?ToQ!fE~5`vN~hD!&V#yTpHTebuzxrg%fS z)xmt(+>OXiHKz@WJMVdUK=QX&iLP=Xfbdd%B}z9_~cP+sve)ETUrf z_9!J0IV#dVB=XXolOVUb#Bqkbf6!&>p0c3pFVrcPX*p!&I2YTk$Z^47)O126vb%fZ z^pC7O{vjTpwRrWTA~>M=dchR}kl&7VZW?n(A15iMk_>qqCM$Nbr=%>>SKfd|y0qqJ zDGnBr=e2#+I2`k7!K$-QP^k6ygBTTpn6`d@MVYCvLgPC zOd~)JcWpxKxRVYOf8nGCo(IvBANH-%N9tD}MBgWJev*z&iqAAh9xS@1xd%UH0$nBld5@N<|%OSvTO~k}jvcTQ;Hdrj9$#SlIfs`eS zO7~OCyoZrfA9HP$3yzt?k$3qzc)OE+FpM#RNU6=UzwQf8(F*xeMzP;H@V1M#-@bpL zj9RF9nY|dcoF>%|+0TiHGqN@cuP{ z!(TJZ_ssF;6%da6)0pg}K4FmkCf`ypU}abWpF!83)l_eC`|C7MfVbW$qsWxm^E` z?3*gZaz*=Fr>u%KoTG@Fpsr!cIb&u^@c@^L(}ZSNi54Zc5x(#1JnOkO=ZtA><2+bh ztj0z4e%$7!{mSK+ig?Mu)jKhSnbxClER&y|wjg&>_XSINbg}};^%oBx<82swg@0NJ zZX#&|awo4Kck|r*tUx!}vH(@PxOMLRm7u{D4C@l6XIkzyJeZ>5FkjP7L@vx;OS;Mp zfn60;?81-UK^~K=(`Qi7R~;tvY-e#HC6kQ9rKV@AcIaQcp|@9Rp%aY)1FRa-jtAi&5}z^7!x8k0ixi$4Tkgfn8I>#aNObB4e^OdCuX^{! zk&EKmoq;TyAoTvZb)mg1>0YBN&})f;WLxPJR+W}wpPjoc`wDEAQpX_{1T96g3>whP z^@1s(6=J}|A3-Yp3q{sR{=8J)pI`_$BHZs#KXjd_{*@!eDz*-s&PW2u-GX}k3}B-^ zuGhRKq98KnKO5cFr{eT1#L7aBNIwHf zNE$9Hl(qfht`!H3nUwr?C(Qty3FYHitm{y*w)GnR#iInmyWb1IovuMuWLyq^LLfNVa$0}j5}a+p!cpx#np$Tz;JksfMk_1H^k{yB#E_dfND z>eGB-Nk(IflgxOtL~5{@<{TU`(op!1QfRriKp1b>f;|7akR%I`4kaRfzqoqxrZwTx z5Z*S$U-7*1@r6~Q!J)as)G?L6uTy?vQK#@ZRoAVbn-;3BR+e->Uq1_=a1eTN>{Y?x z|La7-vMg@8m$ff%Ev!=QE!*YZ`8GXo4YG@q7;WxJwkje0FsWU4S~G8AJzBZ_fGYVd zCItdarQP&I#`zGqys(D?d`jl7+?>-|E)r-Sfux{4LLHpz{?WpMY zinEa|^TZuWPFOJ8BOn?ie|%_AvSfdBlH~>}6ibs?_O1W--#3vgz3tj7sT5i5Eij#f z-_AICv#Y)D5!7)SY5Xu4(>g2Wnrjy{;OIu4;8vQLg}PN~A@MxO*|c4nII)Sa$LdP{ zNmaUbWTCMF&Q}HxLS>894K%l020qU4P|qbS6tLPC(|bx(G671-*==g1Oul9j3*mPz za^+thg>)o{UR@RtHir_F;(rBmi#QkDkm-Znw@MGI=gZTn6R}#gBz+tilZ|BOC80l5 zv{{S!E;D}ZUcMH<#UE?lS2iHKF;P!&h>ti{92S{zAXm8-=w7*^=Er^H@Fr}G0XChD zpE=GSks4lZ)!L2@H;QJSV&A%B)!)h5QVo^|Plv7>R{V@Tg=D%}0tRW#0y=lHUiSmt z^3eRh7y{X#g7~wO@D&c(qp6DwLtOmV_%TGRO0B$YU>1%3dX;S_Dcq=gR9)>4M_#zd zoxhU7t(iBfDPHZuiI;lqI^l!8Uu~!i5(WQdlHGPGep~S~O*5u2dZ+;Lj_a zamQTG-R60(>&@r^;ms5KCI(rwI+Pk-M!~kx-)6bnzA_Rt;Lq6Qy|Zc6*w@khjcM;D zBq|u__3(8V#p-q^uT-5b9lUb~$6yDjR?M5b@l>@TILXL6EZdIsrc3qN_^r>II1i;ajqHWc-?<{>XN1 zFRBkrK351vt(y2{aNw%x%4oC8!4-FH24a_>0mCzK)p)mF%f$B`HAGP@~Ei*@_T}MJg?m zkYk_i1!W8Jfk}LHB8$iJebHm zG~0)~`8}rOmVv#{ybgao7CsJY*TIcD1i2pC@R=3(oD|N#JtV%sU|ah}&e%kU+#Wc9 zP8ZzWJWUnYZ#^^WDw4Ai3|r*tN+=`3F&7M8oGHq71e8nO*^Z$-(`qw&t;jZ7z6T5!H%|6 zU`8y}rjC+af@P441u@>InKTb`sptB_XzIcXW_coJzP@GNWS$s+^o*#zZpx+%Ohj(& z(^8!C(hyA4g%hu7hGd!4lXE~QZshkkAhnl7oo*AryQr90QDQe7vvW+09Oqts3Q5Ds zpv&UCef4-`b>qk-QTT;d8(fxvv}jcrAwgh^NKzw{I6sSnwTHc&1-E&Y85@W_jRcgt zwNR8Va@jT9HlRSY(+w}DJaNHGVTEY-M@`;Tf#gfg@q=+(2cGXw1{H`b6QM%h%bjjP z8$RqU?^JeBykT*5s(#vrwUmy3)iB#xTge%_5<*%vvd*Rq0?5XM zsffOH+n8;j`&YG}GVj$s;tFsI8kN995!3eMUhN#(olVK%i8ZM8NS=hACiNsZo+a?6 ztXrgbMe|aMiKlz8mcY#l;#Zbics-aN58y_VsyBKXk7g6@6;(qw4{jC2(KDdIVW4uu z?O61DCor1n+&t5ek#%yeC69?;T&!n@i+8N%x4OFGCw#wHk5)(*gdA&IkNChaZ=;0k zET3k-VgPkTkk5)?)`{mfzeTF~I^fjWC|XIWdNUzhJ3bFlg+C4-?Sk2d2qGshKPH@y z(@jj1CAwFD<~{RR2Q7{1!k)slo=BQ71}?F;%q&L^=I^d{c{Rog?wa}`PW}qYIb+0* zqQjgqm(Khg;L%blWGJQZa|Pr0Pv?%`6Pm&rWYHrfE%%AgJ`r^uEe0EQj4D~-v@}Ab zZ_koWc=VU61tyFiLeQ6-<;_vvJ5zd6)^pJ}n`J4_@u%^1*GpCGmEPAmVd{!(Cl{RD zNJNi|@%7E3rd;f5xwb+=`b6pqFUKvH_%XA`gJbj4tpFWbN)vCsMDvNz@cWI>FZfO~ z)JDyccWhMhVH{TZ1;x`gOQ^R32TX}N6eO=`qyczBXT3Y@SJsK;)>Y+X|CVFNO00kDq)lFQb)h$(1e$-3N@=6y^(Qf1_dS!-wBDhrMii1TFt!Q{ePg7m z-u##BiVOU<(}WE)En|%9^gQRB8%=4MO4|+Lp1+q38QICieRYZmzINLAOd}#`a-7@# zN-RjG97Z2}R@MEcjUgUHDU%_^t;Z9y5v*)U{u4A9E(7H17hoL|Tm9Ey-r9?2LF`f4 zB(Y@OBV`NQ7)~t|<}Zs-wo0w<4b`{fM0Vl@3>bjej;R6*^fEHOP5yea5;r<^w<^<# z%KejoDldpA+~2qrR^2@i_&1H5!r?9WN0en{eCV)LGxU$Ne&rJ?zSZlCgxjf~Ccp+T zeB6!l&4=KFtq}(sKq0@3`iI8(?X#O^6v+>T0MbwATl!E=`a2-$F4w2TWCN+6dQWDI z?sb!7=2wlzpRgmQ%j?Ip*}qa9PU`v{a?tOAJD$2iT7yNE1wTDAcz$4Z*nLjzF@JL} zKWprNytDhMj<=?L9ilsc)vtB9-BPS915^W#rB8nBq90fUH!(pt66R4M%ehwSdiSVI z01w9=zGsi*E!QS*Ui?RF;|B^s!9e#_ z)FrR1Fw8?4x0s(6`(I;}4^MnWY8C4yhc=O*5Io*FUOSlf7Zf=~LSh2MSL%7srwztO zA;%wCUFizE`Vs0wiL=&THxh#-$+@$vCIQ9o#8%p)^^F)`p;}n2LD<@yLVx93C~aau zI4iWT)ii@&DhdS}IMH9?*pH=k>RCI6KEnBVA?1G%cVKUkG{IA4ckxh%Hv! ztN3r)K{Vy4+|7Kaa6YO}%U3};aKHI$-+b@aE5=^Td4}I$NaGDv6w=Rf><3eiGXlLw z7pD>ma(v>BcU?N-eu)S7a^jAKl%&HM%%Y2i+_JkBc`zI;9)X^=J{pDUrE3Zi1*K}M z3$8uJypyD{PelU}mT;~sL{x8^)F!1mI~eZosx}I_3AIgq-V~7;NimXiW{!mQ*rdU9 z{K~gP6C%Ry$}*_2;=wjQd?OR?r5Tr#M>8Keby@B|`lQ!WOSt#(ru2xOlv@w1N5uOfGf~FFbaD%OuzS!3hU~{9C zbZ22Zjf%3~=sR6L3~U<M-a~G*sTjhU>U=#>cAE=Otj*b@Z2i zhTWU{UHe4eBwEbnDm5*@Pl-P^Z!hzGj-tWH<6A zhiXP0uFolA2i*t(ycqf6^-S5n+0#W!cD{?x0|tw;;PKiy0-mpqzU|uKS$6i$%dST z)zB{+J~)a4HR8II8jWZBhEJ?<`tUw&X>>q*EG*%!C=(Q^@$F6u0L;qd{XL!s(9Is* z3k-Z0V%{HzW=kD*!sLBBtMNlVw`|+Wp|Fm8PE1hDW|n2g+x6zAg*a1sXg z#24y|))l%Q*!s8!uUp(Rn=HfK_UH)emT7t2DZR$I`CCsg**W&WxuzEd+u{)v)hoEg zS+30e8x9ui&bU8DPKgFm-)tlOnv}Ur_(gh0Sw6gr-5cHYaISz)8Z>5sMq(}Fk+h@` zDbzwE^AB>?I>85WR`X%F@~z|Wmd#_Res`R?*m&2Up2wi#8~GlU8xMVK1_V`Q)X^?Y zOF-p~)^*^;@BT+rj}4r!XvL0)T*cjWbrhf+P1uknLJ03#Un6N<)pg$k&mHsc`itku z@l{(2ge*z-i($`JkpXV~iUlI&r{qkXouTgRAe);wz_ zx2MUb^l0d;DL!=)s@8cxZ@)kthg>6yI^jB_8It%1rBRbNYdZL^d0_lESo0i*w_H zj6ce>M>>gHNM@SE*>Ssw|0KN=^rvmnPs!{cZT9z0SUkR&MJkexm@d=lco~88Vdr2& zsy}>yXIX)Iq_LX(XVcOcpn;+hEWG+8kh)E6`=V^}T5jc`ZknA9*@DD}JwWS32f?w= z`_q-FUv+PUjI;RLb!6ovwHF-^bP1JUd!VEaO$Un>q3;hl&jLY-P&@lnp zjq0(Yj5dY$5kyW5>`eRU8%_1&d&S~)LgJ7(3BUL+5KcN<*6`iNZlZv7p4cY_i*uPM z7dd14C5qRvp_8S>n0E6lk|T3dxQmvOff9-gs)3~sbMqOE!|&W#G)(X26RLekyN(lq zin!xdB|M)-S85h4`o2&?Gxp<-T!V#V+*F?2@MC$KofkP7OonnTOGEc;w#O}3NN~cE zh?x+cY>yG@VmdCQjzaf~g(i-eoamkw%YMhXC(s3$_5V7W*R7Z`cKpTVK5q!FrSGva z6NP%Z$$(qCtp6C;r7>mNjZEGTBwPgiB!M4_w_jBNikbpdD!B7^D|PP3A7q>&Gq=lz z4M{`MJ-?>W4+5eUCm*4B&U)jrW46m8Sl-NhUeVW~fcIHMLpF4N&U@9+P?)3m<;sfN z-7!B0wloBT4KfucD$T->LDmU7hn3OzW5%3 z28Yt+F<`ycd|>C&+$=S!Wxf>7#$8RGE51sNz(?OwT5hy%v?uwDbf=#A)D;v7=4+UP zPUbHtn@b3T8%||syL{tU(`~U#97am#tZ*j--s;xxJ+L<0`+lKVjo?HbqB=Ed|r zl`*7-bQ<>C3`b2jQpl4Z_)gX#?*cD)NtC3v&Le0Iy1rWUS-tzpu~~2B%ktgn=%Nic zd-*!I#dDA9-^|*Sifl9OVZD$kn~@m<3F#7iyDq*LmARI@hY3RkoG&SgX9t@w{mIt3 z#?A5;*=h*vY_m;qb`YpstggBFi!Inl63YiyM_5|!) z2o_6!MXkAKIMWt5ACY8*)Ry<-e01egksRpsEtKV-ynqvFo3aptH=R-$z!LAcXg;bu ztrF)ES(gZWaHfmC36>BaJa9*jr}PTTU)#w%5T0ct65x!9uQ(Pw5?a1?AF8tJ;XbVU zFvCy`QD(B^XAZl#$5BmCzprf(6G>j;N&8HWgE&5fjhX*SLv)?5%kVaVPYUj8$WlIX zEGiI~?DUP*ggS-3p^@Jq_K@U{kupq(|K%p-RtRp3ADt*=nFA)QuL=f4wZVg8Wpx8e zjC%Wybd404Ru)H&@%h~noHgwUrc;SGf%Ge5wuR5+)$G|SeZOW>fGXis9~ufivc<#Q zuA1;cH9aaZ12g@I$mX~oM5Nvhnmh_j4MFktpYlj3c-OzPR&G3zdW)3F+Us}NU@>W9 z9m}Y+p^upOr@iPE6~CH@m_=H}LVxc>@+9}p18$@QLTHd41eK?`)=?+KA}1-$k))c?iQ z$@Spi@Uq`vJVNgd;i?HNS@{z?4Bq@b*EMIbwi@o|+todHK=CrvEjJ)*QL=Ttcqm3h7E@-SpQHLA33p~o+1 z8rle&bSf6P0Dd|eX+4{|qGI}}`?>Y^q?3L7Yc%E@gXnw&eu@%y;Gxaa9pFC|t#I46=7c3LvTu`=H<(I-=0^eVKp|C3%# z5t7?uSR!b5sg@Da-GJu!i9k$W(uY}z6s5?3Vqal$3qK?YO-3o~G=4U4iFF=gdlI)| zh&AUCrIMN@N^v-T12%jz2P)sPbG2Yi9_RsTV3`5YRiD9&BGxlwYO#%J4cBh-bhD!n z8LZZV?B{DSBDiH=kYIbqXM7O#sbdPHC9Y_F6dzm7=kL4eG8Z!A1OBy_kI|1=QD3M? zF^LVWP#9e!&E5C3=SN;JHOB%@en zrbE}MU9rfrvWK-G7Bj>^h@4}~)*gu{cS|Lyyc;eO?BcKoU5|yJl}w2{j{5uMg8)&r z(clB(!v{vi2y|&<`w`%+H^+Rw8&3H>3C~vDBb2@KA<}qN=**P^r?wLZn#AL{rg(;Q zPs|DZ`%4^fD!cSlQmb;u9uLjA05N*yONFRjiWG{*s88>Smz1y;`$P!&R&W972oa{q zkm1)f1Ik~t_P|*YXbPBvVk0_rXM$`ZDx#+@8}Tq-I#re{3y3{(0Z{9`;mgmWdREL8 z^quso8733V(fvLB#C|c#^U1A_SiB~B5ey*V_;Nx|8swDE)BB^Gx@f`|pc-Dbfyp(P+i*=uL4*Z)8;|#f8J*Pq7M8 z^(w6{y~Lfngo{OE2RoNij}GkdKBJmmbt)IigKIh~-an3p%cm+0%aH>+j|PjGDSWLf zcWy(uLw-29*gh^O#v+GnPY-_0p`8X`nRdT$HxlJPs9b;^zJ~tI5 zN93FyX8+vS^2Daa__`QJMP@tqMm?(EDtx#(#h}4#lMVY{XC&o67ssE*Hf1LKz{4~E z*#WWh*Z&~-Aegj#(tMJ#mu~O8W+4SOAOCPsi?eALBh^hIlHEpv?p?(>(Mx)R?|O#1 zc7DInAp?+3Z+YT>hML&J-enHLIdSN8wP-vJ|>e%pTb=}Or*cUAnkiV%vDWAj6 zi&(`EjEy=lxVgUTiF*%gC3?3yOn#4GXoKL_8@#v9$IJirw4+|PfSgtIEm?J1ymosi zs2NKz9~&iA9fp>zLxgxF;3j$5(lhKZu5(*_J+Q>M`74uO9gG+9aB127?PQ&UnT0ul z?093av2a}_8Or_KYJwM}Rc0r4hkwo5{d?YuSxv5dB z%jJ4N$H&Xa@?PoXsFvNs8y03m=%>YZIx@Y|aTjszba@QEj`?cln2$4m=sDx346ioxw@={Lkrr zPEqOjg^H$x^qyHDFVd>9Z8Yv8OML$MYx7MJMiMw%vl7J$Ksc{sY|0W9HB!|6WzVW-sOUnoeju(Or*C9L4x6Z%EesoTPFOTlQov4YyyysxWnA zl5(3UsB%eUE;m{n=aSQa-vrQsUrx$O#ufSB&CH0smF|k0!Kv(nN&F#GTJD<}@=oul z>mLW%Uf1;dPjalL-gbP>p-)W-!VTAW@w>NcIYL`UHa;SBJ2VR>yfHC%bsfG?jwvP+ zcu#wmTn_dvayg_3(Hb`vZ)S5%GIhsR$Zv*c^~V{+TewP4@F1d~dW9>3XDaO=i;@r3 zk&W%2JBe67g)jLr*{!zfCw~rbW^_aNGX+W9X!9!d^*=>so00yi<+#c;Quevn95o?u z=0Q2L5;jQN4w7)ikicBQ4uvV?P?;bUQp)S?Na?hvc!|u)>BtgN-sB}NES-^>w|6nBDkod=Gv)|()lcpe~D`OLQpIqd8GeS!91 z5WGErayiX-nRvO^Q|sfNf#@)@*ma~3So!YCxbbYxs5qd^(Fj`~@O-+M(qG-f^&>}x z_!9>wUy3;tP4)U)S&hCyn?jeqUzGM0gzRCrqSX}u4DbMqA9s?zVh2PDzxn+jx!4e7 zTYi91?ZGifi0oNoA`5;8gnd)FJl^0BedA{vcG_QpMeW(SWdUgdihlBd^MFYi=;_&5 zSO{oM&7I6mZEPKk0X|wWj%(jwvAfrbmB>sQLB=B;KER5J`Lky6HHIT zF;yyVVB=%GQ4!61rP)avqC3HDTR=}24Gey_cPz}4HhRcA zsxqTjhb^`9=ZVM0OF>&BZ%X9OACCLjpl~2W@}iUEUrf?`Z&D%;qxY!~E1*{~)EQVk zC7Oqf!W~`Y^?3x%Woi2f1gDvL^X5LP2ha`&RySSF35u%#0Hr*&Ng>$3ys(YjQO@B) zGY7vmgY-d1TIYLqd9`){v7joCoQ0&GxR)GntPHgL`gLsc>E(;_d(T>zCW;p)qp$ec zmII&+L}XY3`k#537!>}`yl$*7^H(zHwH%(VmNitd9rkAua<{@LAO!N_(^OO@Y2P1f zvfh~Lr`ZwQEW5m~DQ)jvr~tql8*YzjYz3=Z77Toj9-o35C;;jOnD#GF0ESzZW?z$^ zxnMKoV&wAOtV_xI2#)*88FpKAA&F^cLf6h0Sg6&K@x!(w*G>}&zhlMM?>p7?w7>HH9!E7cN*%Ne~mv*&D_<~ zTQLu)5%^NektVR4hnGSZ?)O9jlVW0{Qe=d{1nY`0Egc^%dMknH=)md1gexd)Od1dX zOasTO7W7|(@HLmv`L4Z;;oJ3?&0KmaZ5kp%6SxW}s z{SVk__bOMf--emnCxW?Ib4@0NG&g*b9l~7TFn%1^ni`~9B zX7TCa;erYl`sOAI7B&_-67@O~B2a@nOvW788+6Y2r$$xURyvy*y_ZAEWrYz+Qj#!_ z(3F@~d`>#EmJeD@%uaIe7lVl_qh4typ!Ica#P!5X>HULdKg-|CU&ESB2dGz!I2F~VIMgXIZI&BEDc$XaeNE<11M}&Y9 z)}(kTdQ^Pby2Mb<3g&KkMHx$+-k24mQp)()c;EZqD>opc>ni*$YV?OBmpT ziSya>4?z|eq-5k)r>CSQPZxk0{EY;R8B={^eLNUaeI!lnp;A?F(Nh})`HdY!4HRfo zbxb5J3VKpfdQx;mY~UH$nAjiyA)08}Mv`t>4@{YQ-~bS2AgLOle~9-VPn$JG93m2M zYVs^mJ*D;WWY&*o&)bT)I$jG@Ld)v?JRqc!rIH*o70L&wT$vIH$6o*?@WK7N; zZA0`{@ z08$ek@eGu~E$i}BZTg?+R7W2hMT2mgbpQ7wVA0@dtEg~k>U4MWw?wA2LDV<_0Rxcz1eO@~`uzgE&&3HftoIeZQI+06?M3 z-HnG~H-&9G8fv|9nlJl*e3kHeI?riY+}Zv|Z@!!qP6vFhXUt2jpzB;LkVF+tLoNO`-?VrQ z0n{||`|#g`<9e9(=T-aTIr8Qs>5acg&RI6nf&70&BkV!0TKWGnQG~)8nwAodV6`j& z1sR0s{~ykYy_57;yPkshM-cO}d2)f8PfpI*4jf=&M6`DR9?tb&^C!T4sGuT1L?H9; zt}{ocs76dI%U|7-5jIkA{=7Ji_%zM9M2)Gs`%KRs)5UJ($ZDJyxujmf3U8q~5L5OOom+B41W=s%v5 z$b-AF2OZmp<(Adi$CPP0z$JTH@OfbH)38(!t*eRYPHFIK4>)r&4Ub8Z{q&TSB;_mx zb7+)miF{}j;^?%vGpXc6n%3wpb8Dtz5dIL~=vTJZ1ZCeC%X)-zgo6IPr;0@X>+9bA z)IaSPD)Xrszf`{_eP~#BpcEA{_Ino=Myz;TU|M9hZ1V#EptD$nuc7_m+_EyKAHqz9 zSR;k)+)X4LLgH=G1-!ZhqXoUVh5L*gm?H7H{x~=gRfZ7@=IeNvWSB%7FiVrjtDsGU zOgi{SL66N!xl+hXr)P~3@~G!3C&gRd%b^91pQ=BSqBm-cJ1zr~QHx86Q;yJq&Ltn9 z9n+Zn;w}zJHI=%DrSfT20eW4`};c;U6pZdZNy`5q-stXyUA?H z-Wd};NQIQF+{`33V8tedm%S3G)@znakH;6cYu}8|6e0BWme{gZTHr8nBhFoXb?cJM#WKWQqi!MQT~iiVGG^ zMk2#ND@I0O(n2}(K%r%yxRTTmkj*^LXp&Jl!vS(BC>j4vuti38`>~5TET6q;Yj4IZ zIzP*3l_t4k zFBl34SSm0GkqXGN4G<}P(+;N^3vcXxSdoy_1_9W)Q5gMyIMH-8Y&L|uZ3I0RpZI1k z-Qmpc^MW%#l7=YYzvuJg6my7UqioU6$Mu(M=Xh|WDh{{obEj?`2w)|QB8jsD!-?tbLvrH?_z?k!fg^(fox-YRtf~qKI>lfJi?BBs*&0aFM)vNNP0yQ9?*H zP(LyLJ4kjs{~@9|H&C?LzC$EXZeRj}JW$XqKYnCHGhkpT{yiv2U?3m>2DzU<2nMmA z00;)PUq1*2updw!201E7s6m`OELKFnAcGru2w)gs2a6ixKgb}?4h}b}w~xV%BjjfY zAO?>Z>Wjx9Mh+A;s1t|93;#{P0KftfG2AnQMUECIWKd%Uj~zjr!vIA38!*^WgNG02 z)nibj1&$g(s=@k=^50>w;{^^G&bh&&#r7XEfO3Nq=;eXJX88#q)0+Wd9C}9@AFmjj~ zFp?C(0hBPXKe;&QrDX<&t~}e)pSWa*_=q$y#<&*VOZ=bmx!&}e?Oj_Hqu1|`HWel-*WC6~Nb zhDD@`fzbrouJ>v(u0wnB`EYZ=j^b32Tz{hYZ+1zbD%bAeb?qXYi0}x7a2XS;L93YV zOu^S_>=QyDs|eqvZ8P!8nJk=u3m-1vkCtlmu15PUgna1sFG#FJ^%gh-6-GK{?mYe} zcFv4HUQydVmL^E6{RP~3zT8vI#IjyS_T=-cr~1HIU1DCc)}E2Tw4f-rNUZF|Mcl5Q zKP+6`pCT*2TEaLQF5DqxO)t>RWQw)NX!@NpkZt25UW2&=XH-RCSvXV}6B3NJD81Ha z9aS+f&}+ipL+}rEB(Ztw8`bivWy?$RBuK7MKPE=4fA@#jIRB>7fIM_c#kf8(Blw{9 zDVIL|(`#R||K=aHnf`6KUfi}{`71tH)G($!zlw?$d<3_npH@cJXXIBU>&396uTfh(GtLuTobYn*B^T>i z0ygJb%>sAz^NrO|IZA4kN~J7K62lt;?hDD5QRG$5XCHY<4UQUG_m$~HzVaFf33qX9 z>2M7LXvZi zla$8_arCG4nsOK!r(JBitQ(}_&Qo$GJndLbWzWvN?s&z{b)8i{U&>MrR`I68&>
    3~pH2OX#2GyR=WH4vQI9{fbb{T4Sdk)#GHAgU99h$U-3QuPnDK>)PU`uZKG*cm-l)neiR`L4r`mi;tBPgss(ID!Ds-(hyU?)urjI}G z^&Ns$xvRH>Ti_iMds7kuZm6h9@W>xJ@nuYO0q|DsO``6%$ z$3$F|Xm`gW06{>$zpF^%jv>+n;}l4QnhAOIWPWQq^D`Kv!x&;>h|g5-Yhbp9QEJ7? zr(BGIW(H4V!iNRV)x(A6Z6k%i3~yJ7bUMz;ZVTxCeqCvp$4PQ21;o(bEw(LQ?qGTY z`OLEZJ?<^k{h=?csh14h#j-!+<=~s?PQDQCa7yUO{SL1M^q! zu>J(vCwo+|^Wv|+Eu0PZnlp|49y+@cjaqBy%d}|OVFgSWu6Z#9JSDpEIi3RBe_~v< z(Irla?e@Xd2!^o{Q(C959i2EGRS%fork#C)xZ#t!@N7`M7Q9~Pn<{Kj6ef->L)d59 z6!k^C+{{EtBhJ{1(F|)*$sMc8c+0K~M$V}yyfj~ycyDzZE|9=m{a{{qH0XO)SszoW z=@>3^x=@uFqgo7BrdV9C{mvnonk_;FYEw;vt%(I(TGHg%!KYFFuop%vSg>G-{^jo* zOaAJBwcr^6keCw=e6O@QmPY0eWOp2Fs|3^QCi_I^dOY=`v0N1&IS&G|P_?!0BIBr)}LP5co7o0j8NJ}-}CuE3=CHh9z4 zZ4(_RqnHm0fYrh3slHN(Xq zAi~B|3cZ4y!=$r^4C4z%SCXQ>qPK}1BNLoj*QK?>z4?xX+X6;J#F(sco-E7h?80m~bkQW$=9Hjewh@d8;eVU1LqYmHQh$ zeCg7cxFG)brMv;fJV8ZvwbNi@RY(7^B9e53a|m{4RlF}nVBk>iT@tAR~$H&?zQm_*OnhPuQdt)*63`E6Nny&;Mcd|WgtJ|G%KNO+scgOA#|L<#z zb{ZDFA6`M?SPLFW_j7ehG6q}IGS_RLM`a!0^f&Wpl3k?%U9$U#P0tAL0?14SzR0D9 zDj$#Rv(4B}xO0F@ROsthPLFaXoY}j9b!+D$|3_X7yLpQY=s|Db*Zto|+1qiG8LfjS z*I~h{jPaG+yNTmkv3q1<4m~ggZ8&y^LoZbuU1-u1ASfTJgcHqpT`@PeCxpPM189ApS1oV=(e{<#ePG`3`0evR5L-}?d6U=(S1%( z=@9N)(#F8SKc@9=65@^3nwi3mk){-5I=n>))`yxwzaRw-z}fi+YEZrA$_Id?yTml#Ap8E&hMGh z)Eu1t&S21fSPe5cGvV@>-P=}DMnN0wE^ODdZyo}O+fs3;y}PZjMuu@qtbt7`8L>V68{!O zQ(bq{*tOZmPQ0V2QBci1@loP;QQgqT)o#n*G0Gi}T)6${W>+o)THa?Db1>Yr9(rb4 zsN&^X>z{=JFR`PR_hOGDL~jZn3F@lD-cj(&k(S+#J5E!Jh0k7HFD|T*)mPQ<{PeMz zBp3#Ua%H)&XE#VhoUjvXx9q%>`L|>^M+@2U7mU;@k`hGGMT$p>rtclBN^GPFVlg=< zmzz>d1Xk}ve{J_ZhnnlwmKoC`?!{c_ESkNCHEOFRS1}Mov?IIp)&`~llJ%EWn#I9K zcv-UgUa;R$b;uk-;7NO_$z!zG~8ERIHSey4m z3h@xW%BHrTL56$F(@36Wx-u)8QP>Mur#~2<#IKKR%aq%=Ea`8v32*8kA5WPSc)`+5 z18K<#;TrH|38Tlos2-at;93d)%7?6ds2)gbJxG&>Sk%+AHeO|vx<1HJb4ZKnw7HMG zf$1PAvv=Z%TH8HPL}TSB?7t3gzKr&m@BOF}t$N#XGV)sN`C_V5%d{qLdtPWc0%=DF z(ekH>teow;#*(U6%!xn?R~|&!1+OP6w~JlcW2z>J7L*L9$}lqeDYEku#&74}*#$&+ z(w(E{3AZMx+c{8#`W=`spd$NS&NrLm0!L&h|B}M(Ps#A(<*x!>XYFL{*co|A$#`nf%ub>XOo0yedVkz3o$w7Cszsk z6_HEIH7FpJFEQx6sC{RpxIzbC7TN{YYR&|aHbdc^ZUe+?=BzW@b5`oA|y9~J!z`XK%r z`UvQE>m(*wZY{l5fBm*sQzLq%x8D(Rs_G3&`uX#bi`{PnnAoVfym2h_Q1dAcN~@}S zv3ldPc41A5KDb0`MQ{7CHN|@~i5J`AZ4#v$cOdEv^L;oO_5sfMR0;&lFyW#oEzF%_ z;tilqFD|i=X=3WOaL3%ax?J5$wPh}X^2C*4iJuIbn(B4&z)+>7YZno-+69%thj$hW#22H#jT@Y-c?Kkl9IDsHs= zu~jal4CK7KlBMmRxQeN5TwSfH@9CEOREvG*{^@@6NV#MPa*+p8nQ25Noz_K%^$5x5 z2SdDKSVyI9^)p_~&h6tdTb6_D7o#7|ov_Ga0(LgE%89!QPL|5_j=hkcvG+BNfa}j* zxzCM1i}=)Pr92JxHo!PlH45qVhHtmJGnvSR&t;;>sXv3B{tyi9Vxqb6X|iqez`Gah z(l#B-%iN_U2cQTu&%j44DcBic#x0l6e6o)yNMcEa!580B<62oZ!Cu3)kQU{^+HbEs zvVxrk+5f1E3;+#@DXac9u)%*}%1fw2ZxX};g{O-fClXE>+6&o++mdI9qmu^mR8EkG zZxz$gnuYxa*Bk=U&PwxeZ{e=v?Up{}R^3c7rpVauZzR8J5#UI*i zn$VY`d;R|3>f*lx_S8^kCy@wHiBE@9aJWuK(1=z|N75@TPe(Cd5}%1aX6kxReOvzS zlnkf;@=TmStoUp^G|P21LA;`BmW1Tx*(7;V#{Z2eGu`4HT>3?piV#uloa1Wk+-0S~ z7w7_}(3-N`8atgf`g2$H%&wbFzxV=Vg>M8Xq@_5=Ci-r&JJ6}5rP9Fn@G zoOL)jt}`k)P^_6Dsmk@Hs54uU%=@a&+e$brAza|8)~Y7@)5-$rOkeWvB&v`!_B%Lz6TPIv-%jMe3}@c0H}cvxDYPZQ5afX0RV1)I&o!6 zecQxvSmxV_57uc0&eaQ9mi{XXcWq-O7PB3*g3@%=2b6olj1`A3QeM_uPrWnm^ur)> zdwv&OFq<+mrFZsuuc_Wysxf|UN_9%{P}whAnTFp_wL0D+Z#5~-S=&f(3rhDKsH*(T zP$^MWmgT-$T~&ek2N^#62N^D6KEX?42TaP6zTTBEx=Qu7CaA~XbgD4XoqM#9`~+7~ z32I0AiyqZ}M0#jg;4U?ANP$fmt-VuBp#3W{E$a`#;Sf~vr=%y#8*!#)YJ`ohj%8~M zKkEECZ+++5#*;dxcN~N+1~rP>h1O02i2}vi)?;~TE4K%mR*eXev~rngJ|WiwL&`C!Z!Kp*RhfPRX#>Q1v=lfs))}V2hLp7tNm44SvuOtg_W1sc zVzjVsq26i1;v?qEPZwEwewXsyzFzBAKdQ<&uQ2b-s>`1zR?T(Sd@|`KdU|QETZh1D( z_B<#b4H%Te9bB`?+hkZZgaLo#J$+Jwyzv`iyvhHZyjp zwGq9%$99-1;`qtN?R|?ES*S#9NU4Girz%OsNa2*;Q2Dx)tskinSfl+Cq{Y*Afs`t` z1N4g&S`eccvzojE?lSHzSX7)M$grIifVh)oBoKc_bq zs)hJ1heP%-@69^A&fC=e+}jD5aw+=nwDLbejKkcQCA$=ncdo*5XVP6Kd`=&jo0y$D zC_S!qK6$59OPhSXHkhx%#{=kWP%^~yyZc*lV~1y*AVxqG(a7(5Rnf|I$vA0OayMXk0&|3IoSMkTYWkzfQ{fl4fsB*wQ3Sw2FS7rx zmOwnDqjpIE=)8!Zuvy@)nk4P+>r338;;#5;pl$jv#jRiRbU4<-6D`^{Y@hTqC>sxk z7OIzA2QsVXzzW|eXJKq(4!UQrANxS(IJf&PNsY}joyTU^$#l~Gy{en}9nh~a{gMDO zk_7bqYYCufg8g|+Nk7h$#^gI+m;RSAIrn5J-B>-h<3$=1;QfO_C({8QvrYBm_%&>` zW?x$+6J_M?QzXvy7fBwMap0AHCr=0;T0T6j= z(X^&s^^}TSp}D`#qCzW!fqrqQ-XX$}Htg2XnC^m@kjWp`re7V7X~I`wr8Vu&fhuXC zKfSNN3QD^Cr18tg=7QW;FP|s_wz#ODjoilr>nFM_$f=>{c3;fm3Wg|3I1SN5BM4u9 zu16vMnNQMChY~>ODqH-I&F@+E9@0WzZXAtZ#ALmTp50aExauSUXE$@??xPe--`~_l zh$+y_s!4e&3nWlX+(=5^F3S@5WqllJic@IqPYs`?JV7>-1{bZKPuEjgna|KqwL_;F zn&#M2QkhqmM^O6(!9dJPw&8byaeD!7P=TK=n{^nsa~^$m%e0Ih_5%=1BYN|8wl%Au zScDPl*5EM)#VhJ;Y;7e%fa0X_Ttq2N?qbx(nFJ`BxwY9TJj=6jrLuUj((S(*nOA=_ zG7s8*tbVG)QAvHS@8rMm18j`?(sx+RdyhUCg9rN$~F3%W)loG!cAA^wew{8>zjY{s63)H$xS5 z?8(NnS7L#Eu7;i@1si5Q7@-@Kz7RMlmD!mS5k$Jw zd}p7Tmrmzx_^ce~(D<6talWgEF6X2Xk?DVw?`&;YWwM6@32Cp3-Kq4&nf1ANRQUWT zJ;en$WA+>KeVNDZ-f@3A(PbH)4{sb#dch|)eM54fxyXv*jxzvg$qP(#P@Q-S_LW4W2y;wD$VR-2e#RbW$|3H+WR2WIXrp_dRecQVdf~IJH6P`mY?>(4ya5WIj z=3P?lj1ru96Hewlz&*`HIS~}j5~l!L3g0&Qz0sg*FuZ+}X5%kWlu|ew$7Np- z+sa>6(bH4~tExDJI61LOwBq5ry=ygOo~G{MHDke(X}B}Zy!zO0?=w_s4rT{vk4lB- ziIR)FqK91TL>T!Lrm}C>1D-Bir!uLzBsC>>2)b71?3yXLDWy-bU=(2^seR&09A3t& z%Hj(j3jZw*f;vXSZgPn3XKF82ZE}_6KX)1O6cki+nMpa z)maw024OQm4)*VS@Ho*y@xyPeM-6|aKJDO3);Fdws9wX(KFh;v0bjnqw~c#w&0mME$s>%n@Vd>|F)EeW8+z}18)Mz-vOOmJ zL47lR6fqJX9>ygw#Pdy2vs&+{d2znCGvs}wJw2W1(HA%$*I7_JLz6D!?>GAAqAGQ(1yw!6%5au^&CIU)I2YNQApMG2yi=8}(8Z^y*YHA>ev zJ$*M|#k+ieq9riFNMxBq_I+EpaimDI!q*QCdLSw17PDeIz=czngI-``eoz+dFiG%U z*MQ|8I=)uy%9|?VBgaSDo4DGi^nx%t^jhDgNd7YJ~#TlL@7VnV2dw58Cxqi&DvhwhQKXDA5uf8qFY zM{SZgtf}V`9s5+Kk~jwfnNztv$=~V22DD0C)ETQM0|i!>GlHjX!C$BA<@U`7^B0N@ z>uitTNXoj#*ex7uB0mhUF-UI=*_=$oQ>b6;?!9-1;(ic1{w@+J(xiUlS~OUN`m;2vMftn_;;o(+F{_UeJXWPI$ZqA4lsp*d{d`SGa?h zQ^VG7_>W(2=Ioa(s;!I3+P6#UIcLVv&@8?AxtHlqfwX*V*8XTV6W#NqX=|e23oJ}I zWSKfy@1)h^=*kBdzxJ-?n0sal>L@Ru&lIF6V8;1Yu+hdH6Vr_!`!CCMdFXQdDKrCr z--Cm-Nekiq=AR3p08*zrMjKw%dt_|g)***-(%~n876tMYNXB;8qxfYZUDM~|I@smQ zBx76Xm(Z|yrQEPo2&uDP^tQ&N+?|ikPtdf6;<6`S(pKS6o#UJI_Eo0Zu_n#DYu7(2 zNsT`ddtyq7rU3*x4L0av6HamiU9r?s&L7@=Yce%{T5}4k22fM-Se-KM;i?8nlKG z%vWB&TzV0F^-xZdG;aUrb>Y8R2yXlC*-)Nq-@(@atn!SR?VTn8F{_zn)8zBl`ZRC2 zymT!~rK8gvRy8|u`hAQNxp#LQg2oXC0&wDj+Nu+8qT|;>6GQ+#TNSz_!K{SET%=^+ zKztoeqB7oRgjOt_e(WTg48(H9C61bYYbjcX!w+|lT3&uRT9(`!%ofi-j4RNropji#Ju*t&q*d-qY`8Ld>lt-;cy&mhyv9uN-U zn3Xid>7Jio4}sip-Drc@)8ksq6D;SFCA^j$`lwq?%m!uOaLrv-@y~8}rS#m6KcJebdL@ zL3`a}O87U;fo|>-c?xw?IiXVreCJrDXS#xEc`m)c3m5mGPs^tM7SN((3v++NaU6R} zkist$J#puCdHnz4|IqrpcF9Y5k^)UcT9$bbo3#2loe(Y+Vu!?6=5_rP;+IC$^Qht&#yg7>Fl zHYHkQc+$cS2sj@YpI$aNqUD5rz34t|kBZ&{`-7d|Uk{j8YUa8zrB5wrjz3jd?Dj=R zvNG$7bu)L&(yP9rHuRs}gf(ylQ2y8b5s_afAD%Yb9?~T54&{1S(tDe2?TX!SxOS8X zK!~vHADOO51w2ojhHb*4LS$~NQh3)TMC4xd{(5{5{X#UVDR&erky&LNqs7o+d@ZA2 zCflHar`vr!{9PS`6muCjTkX}<6FK=-N$b`iz?w|nHt&?=#1rbN*_S$0b{Xx5tpMSn znK}qDO6)=!G&e#^{YTs)qNFvWirTEB;fb#uwexB2$B`95xz9PPAvu(RXcwFY# zPaBSE@ts+W;Z;6EbbFD#`&mX=H1vshY4Ael(lCmBh;H+C4ktcHca6UN=cG;F!_ErZ z$wyBD+B&bbG=J@RUZ(~eAnRQDTW2FXKU*4vE#2lLu0LEH_NXRX?}Dg7ce+C9PKgQV zD~T_6F}wlz zT*WC2teN(qBOiK&H6$u_bkQ1&3Bz&UB|aR}yDZr+r1e_0az4c*H()`%O0#7_-8w9x zDBCg1ZSmThKQ;x2cd4PVae~hSK6BhIV^P2=hXl)$_nX!#Hbu^kuDj< z8>^viH%Ku`(BKD!4z}tU3ZB7uYGBmGpd3zXgw(mYq?##y&vat%hLaKMN=xEfi7kkE zy>2h&dJakDn#|u1#N^P!CR%T7CKAe<)?V0B?C<`oviPrsj#74`K>eF4ts9yJ*KQ)- zRf2teFCR0Pe~0a7{2at%=n-M*3yNXTYxa2`$snp`{Yde!EPm1^S+H!h^!ESuF_|OR z33x$7sOn$&!ae|>;wt3Krq7if#!kLInAC*LoYTlS4ZTfU67^uoFqCM9E zD?@+3?>Gn0cg59gj?3r(_gkh+(r*u*a&8_dp|8c*?0Z)Jr z;3a|N%@Ih}Igda>sPvwBYV$xZJ8+{X| zxl}p*Hu)QA_h5NKsK>!Ru_Z}rD1fm0+)$*)CO}lbZL{|sdgR7U5{d*nZ=amLUWnWPVvqpZ*Z#!$S4-HMCa$YNAf7U-X`XLv459)86+z6#Q($S zJ;Wb71p$Bz2nM46#WL7Em0|p2yO}-Af7ouy5=8#oZmOjxYoA=`%vHJasx7BMOyf4) z8be$Eo60K@C&3&dyLHvZk<7PsR0D^qJ4=ik=%iP-s^8UFkSVqYv)5jI=#VG6;A<*b zd@($|?Yh({xB7^1de0Kx2JvA>(Au|`ImZOwJw03o=2_2w(FynLqF6Vcmk z(>-AER`uZ~^8SqO9o+-(NjX!6ey1Gr0=L3cyvQ;rR4=D# zsn9=myH_sC|Lw7$r~^Ckr1w4xqjJ_UTB9Y&IsshecTCL$D^{HLrYnn;#H@Z9s6SMv z{|W>D-}Op#()bUy?sKM!7H_5X6?``(&n21G*+i+-Wu59!X z@6bRj#NihqzDZ~JEeNQC^?wTjx5#-^&XAHZ1#x9E<>TVliTIG-(fy>R^_k4%jaZs~RGfrkp@_fr7uTVq9H|qSzE> z?#1biKPwJv@(#o$F~m1don>g-5n^fD;WSE8;vWtVKQ-2elNpx?w4tOZcb&hW$v;}= zy-iSI2I-n^B}s|pZ$$EXn#WH#onq|1c&Vw`?ind*8urL3xWA9Sq?NPrEMKgR43YP~ zee?BEpHt*~Y37PbWw$_H!ai5p>w=CN0v@>}-KYCI;>}0DvX7#Mus#Z?>vHdyMI|7@WK7G); z$ZrAG=^TGL2Q&wi4&U?o4Xi5bkl(->=xwYKMa8X?-#0Ay6vZszKEAVVdofi&%>MZ9 z+WTI#xC{qWo3&gwT3V|kzhwk6h(4fN^xFmFNG>g${MK&GvP^^+SikI9FJ?DX`|$Zw{vZBjUY?$}8|ZF2p{2oJxPDms)^akt+3^kZmo z#Gk74^KLKa+{Z^j2MYIhv_9h_Z%DQ2u2Tb$o7Ax!`Yb^fi;5?O{62EYrQffmR83?c+P`vfm^O3!2(v&A=; zB%G&YA5EtHuPs`{zYAN_Rq?;L;%UOs{d4-9+@H-;weNIi47=Ov_|=iau%&qvwqD0v z%z%un;V8Khp(1pYElL0*rX;u1KWJN{Z4B#qbpRvPe@=Xy&p zSB5)wKHk2q>^0#=q=i^}vw4?xB|xTH(*1pBDEg)4_VT}PeR_1&qW5c>(jiw6iQ}?` z72E9#|GOKbBlE-?#KRak61S`#XrOaZbtg$mH2ZMh@ruJiWCydcv{D_0RLRr_C9~jj ztsCIfO&?cuvf{VEQkVuNAj7-!T<@^l_2%M|!&RpSPy{X^b(p=`L^^oimpVmqawCcl zRh;7(Y(qyotF6|}UCFdfyL3Clz%(|Uq%gxvOyqaOAsND{%d_Y>^7WO=fwcD3k zMb*Ph>BMy|H})o|GjdIbv(PJUiDq>=?snZV+f#?IQ~IsqScA{sDHF*`9C4!tI&gQ5 z*!Zt%xu{!NZ!^|`PLRQBkH20SyhATqAz9u3QY%^uHN4Ub=}_dB33R|lmo|P>2pqY~ zY%2BzuB(9=k6O%1a*10wWebbxexxYTw$oAJ0^<9YCl}cy?}Nip@BFl-q=4YMWtgb) zy+M7lfd(MjleqdQGSn)Jh5lmgwCXYJ*eWXr)hC-js~Q)O2Dts*WfOVg|Ig2DRVq85 z#jEn`BiA{@ZP06{ozD#?=P7te%T>@%lL)8ShB=>*y}mRgYCR6B&waEsEULnEabM_E zIEvgyENpHdh5I+2>1A4pXx*ROdrTODK-7vqrZY_6z7N62O!YkBw)1#LUgYwr>Ev*s zW$*Ua38SIH8HmjXB0)U8zwL)VrO9; z!1N%b(_0*x=z_4FiM-f{AC4`hJPl28Tj`0`x^PC@@#^+GN#CR|nM`FE8c{iCdtcVR z;_G0&!p<#rKRz$o9;01g=^O@K$`4%RHlw6oriB%RQRP;8veW1=-;MSCG06J!i}KTa zVtvzO0V;yK4C!WnrlP`tO2^Im*O~7TQ!xXE05|D3KvPGWKgGJoulcL2->qSDv2OM; zMhiZ(@YHMb-Y9ud7sA1GaALD$K0%pMJ@(kn;L8L~xtAt{rgYnCPL_%0olR<7k&U9v z*R4tg#DTXNLI zaUHDq%Mc*^3q{Xp;cF6x`Cn+uo7WfVmAg-1?o^u+3mVpgq~1-ORmQ{GLP;2=@vXOe5Y8mEC>ik2s5;gICcFB3R5g&5`c={X$PXm3 zZS1MRSSxNrAy_lpYks|4EDnW>`Z?K-pH!RcNzh{Cw``fqe-UvOMNcr>TuHhI!X>vq z35Xx>Y#{Q<9y+sj>`Nr_y@S1zDE1>JXkhc;8=!$(D(T? zx$#Y~5%k(}EU#Dd%wb-=vIt>M-$_dBB-kx7ohR51`m?3S-jh+AzgiJ1BZL=B5ItaV zDZDYMG7Ra+tX_s`B)Ar*C_Tt=$qysgQUi@D!M1m4_knPxBB4<+6pCky$%sfGm#hMF zD50}dgkGC))O5Z+?T*sUs;a35-QSy5)jMvBBfBJDJntO=Js_>J)G=(A0Q=|(4@IWBs(G zHEr#BwlTU_T1B!mJAR3YRVu3v-=03p3}SeHu|7<6?<_MLlhtP%Hg?`EeUdW^&FL5C zls0?@Vo_J<760f88&Dk5YhVG6=_*j39D30$hc|Vd7=C{$)ZKeKu+8@7M2&h?1mW&qOg){I%rE0P0Iw*0o_@&GAJ)yD z+7NvgkL}#9AThPWk5O&KL#AdLgTnn>Mk)<6SiiqNNn&c@8pcuWxBLdRwp|)}=@iVY zFl|DD)u6wi0PvK~6sa3<{ZWL^r_~9IVUej$!%cGj(3=wfEM)3qxyeI*FJzo+Z~pQW z*LP(|C}%RTE*LveZ;P`x@N75pZ~-1UO;xC>1Ry69Zw*nMH^1-ow43@aW#l`*Z%?gD z)r$0Qm5r_53~)Hipo<}szw$>=dm43x+JE3FxFBM58rUlu)2HBR!2n@ofRqi^BYB&+ z%J{;W=(du0xpUVwxH~W61)^51s%B`9hiMK66JyNp@rt#fOSu(!R5$_E2_w2#xi+6J z2#WA6F8V#0_n?g2`1++xwVlJfpy(!$B*f}B#ALfCAZpA|K?4LT6Q-kb)wbsEL{sjs zQ)k=y?+Qk^f5$H7+OASS^FboJy)wSPWd^e4WDsAgxQ1@anGNULfv3aFbb>dB?aR%ef)w1b@V`Y*x;bT8Ad+?|G6W-4$Q&^afK9^aC>d*t=i= z`@s}B{^Www*w3czPZn+AF_XH-l8YyB@pe??hAIyoFgm-!;}?4oetWOQLI0hY7fDG)8|P+P49iwMdKj8>~x9(FZ=&bB@b zHnmMC(Rmu7wZ@`*Sr-@?9Y)5U(|X*-)rAm0LIUwEVCmiHb#q#kUnE8mPz^D82;w){j&f zm%gfPy{T?{bbp#B_G#>a=G@UcjEPMsxmcOQi{%F=skI8(8*ePWBs&-P^;q`zPopWZ z?VkI7{&xo{$%*7jlY0TB@1w{z!#c!{?)G1+qK!lD+C3pR)ZPg*mOWU=wE4F=>Yh5= zt;hE?^U#85D~KtU)^g?!+%R4ov%}0ScCeXrP4?ThN-_1&0URI0WxLN_b$PJKU74yIqbv>>h>){x^aQ4;vRP7)fS^muGCs^x& zFHY@F-a@b+1JwL1>Ar5%dCYPh(kW2u{p@R&KNID*fx+&wALNFyyj)EiaeOHJ zqmAC<)S*{PtQWg8_2k;TH3k9*(DV1=j_tRT%tG?9jF#4U0 z)m8gx*Yxnumz)1~71IB?3X0xIN2l0?9goAHrgm_)S>Z6KF*GLs3TlPY)eV&fSum%A z<2uIx1K^|!hn3D2zfpyu8^1`EA_kF!pfr6<9bTgb=+`CQ&wf1J8Ill5W&uXle$2p; zlNp0+KYx6}q?=VX$Wb>H>(u$m$zr?CE;0H&zcoSL$D#)vLl=6y<=Zn`6B*vdvuFjE zmvItNzp;9_{;xX(WzDYN4s$TZhpP}UUr*V&*OS+}<6V^i*!G0xFLy2t=MgP~)`~ip zM~w9^?+`eDd{(`OB2pI_6E-7{_m|eTK57rHah#kZNdHft--g}MghYSzfuoIW?ugGO z(9y=tCQAD%2mwtv*JIv1Mo%w-lw~&(V?_?%SAp?A1yZ;=-cX>aj*}Ah#Y-x5>P-Zi zc`}C2eRd#YepvLKW2sjRlY!&R&Y(*3Y<_ifa#{dX7wRf_ms^QeEQwc1d#NCBv3jX6 z^vB9lQ8<<4a`A(G5*aEhoKV--igA4h^E|jITDC3U?^t%;jwdH*?nBv%M}A>IMxx|i zPj$eVmG6TPpngD7eex!Hd&T+QD@+=%Q?p$U@6j1S<1xoUxwkiUazh#;x8e?8a(V_< zyC@~29!6lTsj6(LWH0qp`BaO2O~>nT(?F*AbHmF>-mkE?)_KUVPQQ}L(ng|cw>!0#C6-I5F{~VM#x3MLOz<paAPw;OAG8VcfnI-U6DY3w{;^FU_aOgAoA6THui7{fSNQ+z3S5-nH6z!5=On1= zT+59MFY}cH7{9bAKD|sqthT;CcGBZr#>viu?`yseqtz~N8ee7`uZDDnPkDR^ZuS{T z>OUsEQXCEvrq7j~v_~i0Skm*%vHrTt2Y@Kby*Ad)|G6V)IVo`~sXyW;HP4HSo@ixt z;>9awhM*ljT)2P1TQghGtg62OB_(TwfX>13W4VSJ8ktohDzZ1MP*T5(u~#RNwuR7NGOn0bTJ9ht#+Rn8|Dv(Y6F5~nNZ314X$jH^c3{Wq=Yd-LIJb6 zf^A@w6X|1DUgQOWdcWufx{o$ReCM$Kyaln2gmde1@C2;S51#NM?|BhawP0Itp%W|p zTOQv$RrCAFRQnrgx3Qxe!Y56}?>eH6{^tRok3DZL>2q<=T!r;>I}~&%2alof4V)%! z`$+#U6OV<6RtJ1%jk0E1vg!&uZiT%3dYDwE&ivdnx;sarD%6$#?Wu{-IH5~xAq~={9Hi(-aA;$;f1f^gF_STiQkBO=uaZ8`xiI@z-(A&6q{vD`EsjArx4E!JV-uf-7 zzU>;{F!az!w<0MalA_KK(kRl1lyr!Mw8GFS-QC?S5<^I;fFhxEH-m_TFuXH(dENI_ z&wD@5b9{ezkK_Bx{sZ=2d!6e%KkHl{wX~jaI$>hi-QbJ%+i|Hk1O$n}6bN1Jzi7Y3 zzs|y;9RSA1V=(l73e=Avz_QXfOGd&6`>QB=Pzv&%jO zo9%Z$#Ic}uV4Or4aj4tRArmo=^KYOgq8CJURs8BR>3Tw0GB96`jcpxk8{j;K2Xe2a zrqz2WQ*^xS(ebmPzEbZtK5_BsrK##K=Tl55Y71D3&g@O|LQ|Vc<1Zf|_#VXYf8w7C zK1)aYrjaak` zyvhtl%NRuRvd|kwO5GpPbo*n-GT>|6hwO0=Gdn!&edKmYY#pHJiqQzAf7@`O z6)|lk4=oRw^d)xiER=7z&h>!jW)NLXkV{LwuIIzpvye{>O2ty`7rkT?lyQyNuSJp5jORxgLcDw-q6#7AW;|u+o2I>Sp4wOmmZF0uEn0W^m^V~ zGW?cGV%9-7f=o^qSKfqB^LqtbG2b7KbpBLDX`VUAw8$?y+a5`M|Lpn6>!}XU%=_M1 z#~(Y7ox#s%0z?fZKUnpSU(k!6_te)Qe$>~7@l$6cGUy4MeI4XwIbbKXhZE(^>C*>l zP1sQ6b81?|(8j4c*1U0JmiptgMimeNuK(+#c81^Y5bxscO9OhB*k$lo=DWgZ=p8S+ zMTYoA%F?t9%Ga>Tabq*&J$gKnHyGBK!jR;iCpiE+K*Yb8C`R2DE4~lJ>WTZ=U3Lpo z7y=RbqGTH!djB#sape@3r34P5%;#<$&*A`ZfFkRw;?K%-11Ry3x58@iFCCXxm2hT*@0YL4-t@wJ7Pr0Jc5gqg;lniZ_Ll5SFVUmJE~z#=F&4A-CWlTR zg6|elt}D?lVgAQ^g>zCacuk)#{)Lc>e}M5q|8j6{zG{Ywmi8%PcD=HSvWE-bzh9QMWUmd@Zp?4YpkGYuWw!h9$US_ zW!f;R`HR$%?F7l+emILN3?nl5#K*rgXAs@o6KcJB{l8Ni{bxPdX=uLr6E#KXZ z-F4-1qpq0e9+4K>-j6?)5FHQwl9JtT#`tBJP)ptce|LpldAg6QK#)&$t-VGw5y;~A=y~QrJ8nWf$#m^&T%|;HZrov%Rf^qthuiWA38I(X zK0#}JH?DdDmShYG!&+-=Wbu3v4_&g~9M* z!0|iUSLWG8@(rmPMI!fgbk69szvMRFIJl>Hi`*HfXB=KSZm{P`WV) ztsAr6qIBcUwBK}N#qv!pn<`Jo{hagx(qVY4aMRMRirNNX|g z8;2pkqujjI`vIjJnbw}wZFOR`(ZHUwL8h2+y&IoYdH#slKz+yNq+a2v0 z0_QnOB=1hETD`|BwJWdbpM39`&hPS){XtEzs+s# zXfi*;_@A%P{|sG7A__)_E{sDXIRr75qGYJ7mVV#3Wpl+{FUpCdP~*yR=$XG7m4K~F zJAc0&>i`!pAvZN}H@LYbOKVqWHOTC8YCH&a`s`TXMw{|mW0DCYUy@6jHa-fYXK#sG zBZ47SRnr0<9|dK+$QWMB^qom&_iJqG;K&Te;f^8!pdK>;?R5)*JWpm&{YQHrH$FTK z8jv`D+j!+>GSsZcb*6wpO)lGEFIg-r9xfuk1%2e0d`)HO@gLic`@ogo#Tgdj>7VxM zP1i`EZ$dGz-%~!aG3tFu0`|St*FV?l*{C}6!}+Ocr>91eP}HlXcqKo^-Rt7;eR)4u zzD9H<@0s|(TzUE}5+O_XeIrobtq@gg@s&AVA6!6wy4eAx$sS+`u;|qDb|>JZ zHK$8B@cwzCa%9wNsBSJ|DcE7s6-jhHt(dIFCS5~#ZZ%VuTw(5=x*IZOJ(*<|Zc1#2 zbGDnfY3Oa#g@ce2w(yZsIX5}045y#9Lr$ES6gF&s!2>_^nqLUmmw@D)i>)nEuf8vt zq_Xw$ef4Y|^n&z3OIwvuFEx8wd($3cw(}BLuV8m)!SfqpasI|Lw9!Mf@suo!zY_x; znf3%{#HIq@^pWFI^2&s4luF(MD?d%VXjyW0QPM!LWC5!*^PeXc$$;8~hp1BZ^QU0+>TU^SYl&Me-rP;T2--e@=pup~0nj9~ zpW{#>GZEml5>+pp)=Ms;RRa|jYGq`2J ze^f#0D22-Z3yhoyX21^z|JN_i%6^OHq94OQ8AOMts$iXmFmQdy?T-cDJ)wU^p05;% zcYbzR01ZI$zkDxWC5S~Dsi&}CFcc^bo=Fw2T^hNNU)191iGq-6phOvvf2Y_`HW{#+!oVW-eM`5@Zm2DkC$VDkl}rD^lOrDffmW|bOyVsk0oY6b zb1JOi#@&T(1|o1=b2aZ$Xbi^b@tfp*voJSako{th^15gaq(n8oche29#u}6K6 z54%D4dwu%LHa#Fiu{joBSHSjY$kiUU?nxcMzoMxpA)wy=eKv!*)A? z7dQR%Ful+A_PTHm->q{<6BDp_9D|5@kj)3V-Y$D@XrJB3UcL>L(Q7~IWpu<8#-^nP z9hl8N!%>AsxqLnb3>|tx?)qThx*hZ|dd2lklWa}q49f-=vu9{PfH^<&NCz5nN3{;l zMQ9zRTn{@1?KdSQ&>PEKsc)3s3y>S{-Ik=jg=maI5ZUANu{ZbF)NQ}gwR!(OGYc0!XJle2xbADDT{ndTNQF$M*X|~s=T#<+749d`c9L6_&T%7k zPbznpcwO}~+gd%NGVy!7cirxwr$?0ymA{?du$Y|giT;mOz!gThUj*~s(g;N`R>mO0 zWzCpq%FC}vH8V9WSScDu^GpNPduXo8YqG~(_0uB|n!R6r-F-fu=qyG%kuHa8n7N{9 zt_N&h$NDA6lvrq)-!2Jb6>KY}kpEn()x8tU--J*yg(&J2O6t#x7vb~A&2t25s-a|r zcimh;s@>c5`S{-D1GJ3T>m_;GG$)lO8Q>9>f~QqS04J$CL3A+>am88sIeW?4ltb|R z^VCJ`EZICU&x~x!ISr0=4d9rI1=fA28{j06Y{#7+m)f}u-Iqh($Xay50;MwP4 zS&+E!;QMyaIo6i`v&?{7ImQbwaZ>{Iti6#e9FCXWc4v5F551niZ4H8KmOn48 zt=y#*HJicnB6JYE8^nLl-)qmewu^x9wvEvBl!N&~)2k&JO$aQTk^J?PomQ*2)U0e)0kO-wVd`5ix@!dC&5_-7z`*hLnvfi$9#DSd0 zab(Y;#q8-!_U)hs;0VL0wYk#ActpevlTW|qfNpN(1lb_~fkO+os{8PxSZ&!Di7moGjlW}|>|Fey|jnC6Dak_>{00+;2b zfLmd8o`TO|(IxvT$No350?aekj#(?wIQ%9X3`Y^>??B3S+u^T5=6Km5BtmY&bzCqU z94~T*l6__%8=cIn`6>HT%pu(=7JTm0cmYz7>P=tWdS@d*WnBgp;M5GD6c#a_XechF ze)=fM0W??E-zMQbd41p4f%cUE+2-l;N_Pk~#}u@uNRx%4Kj212>vB&w_+yH*fzU0| z9dp1NIJL)Ux^C=SW@bK>h8SnoYu}l&cRIOS?04Ns#IzL= z*8{tIf^?;;t9HCUUhTiu>EfBHIO`qUFmgs7?XPM*>#Jn`Al&?%#fBM}U{$pp-|mLF zA%$hKq?WGZedM(|aHWD_;vap`{hqW0BO|7P!?Xj=t}HOO0dqkiV3D{?=}Tp%Fp$!7 z=LF5mp=h_qirCheIyBuuwSiB%Xc1I{9-7Ij7Io;#CLV=&xitdMp z&#CA>rU#BCTPi%gIl7h;CChD<9eK}sJwM63giEp=>o1kb@&7^Tdq7&VG9<3)i}n>r z_F|3P5z$Yng^uGtsJ*C6Z|`f)juED^W=+HWyjPSSi9Mdg^1%l6Ox5naaY+@GLo!r9 zTlsX|&^6>}4|bY=Px?MJlX1$iZku1sGdey>bVaVP(?%EaTJ`@trf*5;QH14I&a4fs;2b$q(}?Gt4bv z5Y2I3QQZ|XhR4dtuV1!MzJs41gDpIFl|o@aUOmP@7Jmgw>82o)%o~wCj3FX!o&!4+ zf|7!y-pFSsTs_+x%<`3zFNB)nEcaaa0~jeWn^j+8w%YsV^S7w7rkN0#Vish|K8(@l zc+wTdRmM1*iaVEXlGadrV8=0bob36%?*2?}7b<%! zFib+n^e`&@g%3TSF0r673#8ATj|R!_QYlDE z%kO;)YH5=(ukk4sMNCfz+NAjOgFUrVvS2nAg7Z7o% z#{12vaE+4;Nam$fFuv484jqiX*ltD`N~X30OL_gZh4+KMGo8@Wd@CHum1q1cq;tLa zW1)N^@pJt8l24IJPZV1R>Pyvt^Ulx%#fEE=#o}E^^4An_*;@No)|8u1c|J9{yXKc_NM+WNkZ*Z%`O zhG%5L-LRIUUE9i;tNid?lj=(mk#2*C=hgKlp{4kv+J?8tMe+=95xbZ^zV*I-EEdw! zeK`!b#CZs!%`QCQyO%FI>?kLIcz?gW>8mf;vU!h%5(gM&r+B339)xevriv54BpfhI zCNR%6%x{-w-wEbunqT$a6_O;ShR5NM?0*IPzO!dbazYm0Xe@E{`NTuL#(diGw6}J0cOI1wYKxDuR?-yH7m;CO* z@GsS^kaxhu!zz37C6j(=;E6S*jzA+EPd0tfwNM3c`$Dxe+Gzha^ZxdN@LQe2VQ)JO z(TrO{m1YA;cV}9XMSpXPZ%48HjQdDe<@|Gw)9c3FS-2lKk``^f=odww}-E zO{%ftwee1VIngUvSKUW+@_H1r71c(V-=6#0qObVbzO>2{%wj6wjw-5|h{mM3bnoNU zIAbjUMo{9_{`{~vJBUii1h&ledAR62p07~fzveyOre>L<_C}eAh)$^o>54n6RB4L1 zYIWCd&dIIXVcAsr8t|Wc))c_`|6z_~Qn`s3`%n~{@RK78v58;&f)Ew0=T8Xz9v2)b zK|k^JJsgG+x^wDov@u$4*lX~Dg=5=;9`hU=$arh=s zIdtOlr_J?5Tvl`rqZgeKTMD+qK=kr^0|-Abe~V9U_$?``L4MaMs9A)J*HKrJX<$)W z@@)m|Imn&UVe?d@y+TS8>}c@-q}G%KD_XA>`~aSQ2u&7sPvi z!*xyP*1<{;^Z~hS=0n?TarFYv9oI;mTs1oN%q-%(oWlNTry<_TnPLj|r?$tK@cfhv zGbY%t_%VV~l7oZ6d;WafQK8Iq65}cy_co{_{Z$mVb%QqOHX0nPd%GW`r96CGH#zo< z*GEY+O5^F)d-<(b2thY!%eOHbet9BNneJB1E{e4{@lSSG-78*f!p=%f0(d+up-9F5>#-q^7aI|UkF8L3Oji2|;C3ATxeAxy@ zEj|>eI^B7xsJSv4t5a%C84|bBUfJG;ywP00C0~7hsvHF9mk zjPr0tyWRxXaZ*W#6`kh5!1s1y1r7ya2xnXqSReH5;6zVcj91?OT)};25?O%GJMv(w z8v8c_Tx==x+})f7mcH-cpB~s}!DDzJb_YfV%l(iWyZCrI2TLDcaFgIDd$^=S$wB7? zmy=lvcRBaT>u_PWJ-;dAcJ&r}wY?Z<+5HCs*c30(3zLEYeCVT z`BXKlWTzw^9BhHUvxYx>A_||o4ag0{qjkm%=V_3?BAIyB$rQ%(zG)FkjYI@Ri#B28 z#7JY>PewzIjFtju5_F7J?jp06nN^6teNbwLa^>?s`am-uEq9V_oalo0Er|I+kCjp) z?=keTz1t4Wywq{OR%|@$RZR}S<%7%@D(r(JGi6KK(L&o*_x^5-IHnG`1b4^55VO;X z;Ph<`8hcog@-6NU+Fm{ftIryRB-b(IgtCmRvr;PREo^t2PPi!=-&(7cH-2^*%&Qcs zBCnFgpwm+#3buxWTGZUM%Hy4Nc_kINY_=PwZeH55S9!*>(?q3je(e>+Dr~3ukQTo4 z8quSqij8$0gCF!&(WV?ED`o_=Evl))+KI32mHR4bij3WE7kF;(8%*_$q|590YDIx@ zvEIC)WqC8Nj=esiaqicMABbR%#Immg*T6K_U;j~-5~DzC`%jmyAb(xDB9c~gLAk7s zOY%|5eX*SC2B*C>`2(n`O~=;%bsUn4E!scT|8Y?!0r?13wJ$9WB?yOc>z@qIv}Z{s zE2eu0kO5HbV(WD4-4ab7!^VH|&ql&`E>3$4W4u7rSl2F^u?I5}eb;psn%wvA1QJ5O zF4e&WK_JuPuUegOU#p$1WbU;QI9?)ESjso+u@rIe>4Ar><+~25@!zZVzY^NpQbZ(I z-@5XRG3qb#Yx6&{bX7lFu9RS}X+P%#wR*n!3bHzPbXw>Rqp%xJ6}^c_NJY%Q>&qk| zJ7WeQmWL=!XWwfm8+P_Jh}V0{0mY8~ax*@}#W-;O!_%y663D4swrulR=X@~aIbKX4 z=ZVpRBB0E%K*FZh_*kZabadX=`?O{@96vfQ2kM2l#2R@Se+dk56DycT-yt@Y3sbsd zN}&YCBwdQc(xq93%5%}MNq&6tN;UCes5rZt*~xc``<4+JsrPgVy2$~nz)|z`kPfcZ zbcuTLsdTpwjT~y;Go$491SJbO?GNili&V`;o(ZAK&=4PKHG6i(o%h+d=( zr|`jcD&9x07?M>i_3s-F=gL3bU`weh{upyT_fERvf$`l$ke}v1+qVg@*nZi!^_;H$ zI&9>c?zFGBzMDpcm&FoELNe9uR@UmB$geH?qq5eb9q6;A+ML|Ctj(7I3wfMyZTb~! zHvW6~r^DUQ*2p&Yqt(PcNTct4=lvY?E_vKJmI~qx{wBYHysI^)=ixRa2s9@)a@OpE z5aqx`q@JC)eiI2hKe}-qNH1!>(^XiEjErOkW*($DjBdNeJU@SDVWdK#@(ug~a=JRm z3{GI1fqolVm2;p>F^yvc9h0%bAQOU79*lw-b5Lq>tcPJ-g594X5SqHj;bgS=xzIbn z@RbOTcpOv3_Pp;w`~V9+5H2lpDb+#cAq{=(gF8Z3<27#{Pg;Pclk%mscaF>=loF0p z5>4#<%~JGh)>l&Pd$?B9{x#hDxnodZ1ifVpA7BR#zlgnIB8gn&~9x7a@EV3ciagW zJ+9MK<;sO{>!*+d0=aU2*V7f9S=91wV8;{>#IJtJ)9>nhyD?faQENhkDSh$L!Q-7M z&fyy`Q`<&yPU3n(+h0BV_85{Girnyhy61+76gYXXwI?JyKMNr#r=~ZC|laK;u{mNIqKf*vDXuT zFFNUpJg+fXxO*)4v8(;pfH6<|e#{oH=_@+SbaH2lOhIe{PAS6*aZbqqWbvx4`S|WRw8KI8n+U-j$HgJrR25v^6G-5E%d4XVX z#qjIcYmPz>oIe?2V*hFMLW*HKDTD*i`lMa{8AKVGiWNgA&%JaJsQT>aDH?CLKhsaB z&9x?>gqeSz8A_P(oly$m*AKtZqd)@@!T6%;@a z_qL&Wc`&KauK!QjC}2#*-vU+XAMv6{Cb916>Wi{e>roa}r^5{EUBkVFu2Q7l`{$O9 zxo=Y;7IZJ%XTv`FXfda?zTQrlQgJ%p-fV3--WPBO$A=qu{x~cjL8g$LUv<^C8)&+| zSm5h`(J<#J_7XoM4)uJI$m*;vbslI32{i_`i!Y+f%&bJd>g6b?Y`lY05LB_f+`w;;a&NTd`BARCCxkA=#)inrIkjektQa=_=DbIU*96h9XkA2UaaY(-Vmyr z{bSI(1WQ3sv4_bqs%?O(X`+D_K2%K?eJ|=>{Cc$JJr#U-g2t=KXk9wG(SmgP`E{mD z`Q!tRELQ>o+H67mo18ghTmzhn=F_!nxg?qje*j1z_r@RZD6lq*ZHqXuE-Rw@yrJBb z;a-}hRb5h~-i9pUuIQ{dLCGkVu&v@0BAMSN=#_oe+x3{jJ}Bkxvbo(jmr(MgX(_bi z5ok7tXA@(&z#3k^h-nXNE-M-212tB8fGKdlN$h46YR&Cp$Sl3yXak7?_Qi)n0s?Ji z$Qqftk$Y7F#lo$zwl$@g=2k^A{dE6!U?z`xr~%NwK-AKHu;4n-hqu&Sbz#x4BbT>k zDSk&{ih##BnVq%&I7jw3j29nbQnl+}~!lBW~5b`^guod%d?Qwz{a&k-JoU!N+{-)(y~ z={Tq2L4c9;9k|Fip(XIBi14iRT-GAs6m$_^}sRKuTV_xQ5r=H;u@&JL7(q^8Axl zX4`GyU1smOS^LEv!n2X*=SN2!=ft}+tNJTn_#s=kv(7-%lG@YLZInjBBfn($+za87 zj*>?Y<^mfusIs85`n?|@5ScYb;My?&3uT`%Ux2pb9isHmaB!CW`M2+vBTBdXawEIu z*@uG?_9jgHuU-vV2%`*;r;S#YR zyh8l$_~R_3P|1ABuSJU=|ss9Cm#$D8(mVrp^I{hs{YE?4PV6_3}%;qF)p~CRnSPzYo&x)#SEkF1*%|r+|yupuXN| z=xuuz*z9C-b8||{rz3z~o)y)JFANS{Lx=&)5X7RP^=#HaE2SJJ<1{e@!y^CxS<(qnXVmsF(mzWPMye z11bo`N9Ndl@^E5qieKypQX?@1)Qwbt^T!)g{py8~RB`N?O$8Le`-#$O-5RJX9?8Vg zy_v2k{8-KRdx`ZlodahS8w}1quh>x5X#@u_^yoEE3c%DS`!1NULcd&ee4|{cYH_;# zaxL8@!nwk!Rv5+W7Hg+DQ3;O9HMKo~0YTk^Jms-*|GN8fRA05@C}CFm-5DatSF|ZQ z-S4gsrHl@YVPe<+>z_=*M4xCZKrcovbh1|9FgZReRn3c9XX2RIaoGd{_N zdODa<`RY(>1=UDvc9$iHxAMADs(rm%WZ3%=nN_`~4NVs5FG7zk)P0B*weCHg;fl%% zteLn_0L3Z31>*lT!J`#G_TOGYW2@OdrEXhw12bx?64lJYFoqI=^jkrQ{S1WFa06Q4UQ=Qi+Rzs`lM<|0$ zIQ*|EQ~!+Gm?mt13orq~>GH4tL@k5(1EafPAn}Nc-R7w8m>3`-iiyEgF{hf-s|aSm zNYDrWpt!bZBo`-yn2<`jNHtG2R|? zs!ziCg6fl8jryDa-H9s;emjf45ZB~(F^cnF1>UWAK{#e3ImnMe)UEug)*`R{QLDY= z=XVxw+INy@&TQ)K8*Feyd)!u#l}oFj{h%Pjwq z11Bp$gd?W&gF`Dm9ItQgmh;j~>ewj7$*Pb2q2jCK0zJ0(=;qiX`QI1t z(arJyO}G9Dwd`@T#kaqE>y#u|I+n2d=gmB{DG;8yxp})00ADL|{g)W2+psNGxy7?x zOi9|^T{mG~ik2m>SLU&mx?Hxlw?}+-s|nQ zX3u=IYepoT43hV=OJKmWtPk&1b>uuOEC|o1`-5?Z1aJYK{w*ebF7nl@bMA(Cds&@2 zZK5%GyV?Cgk4myA>&4|1Z}1cO4XV-xWX{- z%)E!$J8K4A(mxgGNK;`WdV3=)T?MUAY{DL=xPKlkSx(3P&2;9#&t*q@TPfbN`D4#@D4!=~0 z9h0YpKOQyLEDcQ85m_I@H2i2@;!(HiW9gb$f2P-nn5c+l9{SNN-n}~bcQ-n={`W(= zCng@qVc<~%6eq4=3Y?1w$CicJ+ZvbVV8O~*jk5u`d2Ucp)-tX(h?@y6BX^cv^cZyf z#F#XkTM(M-2d(LqiIV9IT7=46`gNTb;Z^+QI1ND;z=y`Vj}>;wQa&f-F1 zl4MJ|L_>?VX>)c5Bv~Q5SNuDixwP+3)QtXG8~@?h0>v^I&`>ZSNRAOci3KzHxjQfJ zm;;TZ!H&6nk)Uwp{Om+7mi;Y?Wg1R7hM9q6;q&Q_7qzjS0H@x)uJ~iFbD|_HFreg2Fzgk!NG&$5+yaI5+}s z6|-G8^;_NcJKxcFM>(G`cx=8*rcP9;7G`*LV-ON<80_UVSZfvI#FOb({rPQW_MJ}| z8Pf<77&9N0iHr2>!|_&cyR&8|zv>Wbv!fG(2t?5Ss;v0+!;M9EXCFV@`_Y}hAxDct%zEB&mAfi!S=&C7B-YwAWysTl!+ESeZU{kiTBIgKeM|SG$jwLqJv;a1 zL?gi^!;8fHWXpo$l@!~Ce;UF4cUXpJ#%YMRaAyB=;h|~=6r?QVT-D8!iJrVPp_n2j zCdJMg8Sci)Mmm8@iXz;t!c-jI3><&yaGUy6|Ab$vc-xZ-3Jzw1s5<1!!_bFZdLVfMzkm$*o!_I1ulObaN z%UgQ4HKnuCWB4b1eT*Vz9WDGLRLf0M7w?LXAxP?v^WN+eWT!gc*o0Cmcs}U-)Mr-z z^uffuC(Gj_us4WuZB9=%T}&tR{Pb~r2uD8qqN$Kqu$HpK*D(`Saw@A~Z%G1d%}8pd zxNIpH8>T$V1NQaFSZ*RKv#9THV;ByOCsTy+&*KO=zRQJdCVM5b?eeb6Q#wCD7&o3WkPIq#+bMm#UDkJSp5iI$u zrR(nT%UY_KXOdvjbRxkArMQ zh%0K8UG1vp00TOE#?Ra9?5I|#xWf>XjRXnAZ&Z5gR5MKv)bn&1DtjLt8E=wEdgGPm zyh}OA7;m(H#jby1!ZliiyGv|ZDvWCQ1Ue!ZH#fOI1=*fyV!v(L>nL_Q8$L5>Fw=Tt zNg^?`rGp^p_d-q_t53|p-(pW=DqsdXlfSPrIdA3|>z!Z@!{2!ME_Gj{5eql$%%?%D z%cwQ>T>nSjExcBMxIri;(vhGzW?BA z%gwFo!zuFZx*_t<0h9(B^VpP|ntEspCd&>65$ETv+eG1>0L8S21S>FC@&(R=Id8Rf z50o##2m02m7u9lh!lR0f6Ap&L-?`M0c4AU@+joI(Z(Xw{2HRG4lbTFb^6d>gkMrQ5HK02DXWhwU}bm3hh&2Ne$@gI?gLUGA2iwY@r`TYN1) zj=%lwe)VtQ{#_O%{u8)U`(&oDU<)!pE^*9FF>At}2!7!9dGyBV;N^e4O8Bi=aC1lWkMVyJfr3?qi%TG6Lc9SZWZ)u)@#e4JEb zEGoPtT_dB~5`8_jYah#HrhU`(5L>zpZu>F*O3(5j0Ni)-lTwpFXSwXln^+RgRq9<) z2Q$*An~9YJ-fILcqKKz+wH=qkBGRZ}B)*PvrA#mDge2b^owJuUH;a6!E5vAOc=x(- zr?Abh6uF;!lnJp5;4%DRVYJ75$Lr{%wdLf~Qe(zBwCzi?Yizq1=$wW4EU4V)`!JaB zto5wV@J`N;R}QN~#L7B{UQ*%9xPg@?4dj7b|8KtKp>^21EOS8xH@8`0d zm*Q^`m?g;x*wESg!JV1#r*);Sg8Q(c(HI#X{o__mWxhFq?`YX1%^Xm@wm82g;W}~W> z>=Sg(PyV;E;3v3|Sc#8|2?ESYD>4!%xeW~qK0kbL=ue$dEJ!wL<|Y#|Zhe)Q=D{`W z&IZK`GKQ1uIu5*%B4h*e1dj)b@kkv{hDOy-Vh%rP0{NtFH!Wcr?bNTuysju(As?}8 z{x+t!i}8I$Zx?Mr-o+q-a~pupnc&?md6(Zj?I3EA*tm^#J_vN?f7P?C{bb-dRP*kR zQbh%E?7(E-#)jTgkAlB8%72U1t{{-;UH@MuhAF|Hp^et3xUjs7`F&j>TEbHO9@?lM zGj&{P@jequS;H`~pV^ut8OzP9TgCYyUp4;b3%dG}_kQ=hlj{Z!@*p`N-R~`O^`*M$ z=XJZ&Z^@I?LSQ^4qK}sl}^o|)3T=lj4p=q>EAB4xYp8vc|`N3_7S8k{NOVr^rl$4Yas*;HY zO{0{g4q8b|^^k~{7x8`*_O0^LUr z2~=Fxk$7Tv=frvU>MN3o9(YDP3l4rYAp@Qa;cu%_IEqqxWt zX{^4<*_Q>LdA;tEJM~sX;5>`ISy-w{^SuA8@aBGGxyhr=o-d=Uhf_7(ZQQGB zx8pqD?~UN~9BQN^Nw52wNuRncFHg=I;=kSuqCT(7^$;dUtYP#q@*(acJQM&H3Qp11 zO~WJavse3m=gfM0tlc3mtg2UgeCz}c<-AL;0&o4`qjD^k{2E8TV7_1uR$xIzXWWu9 z!#zc`t&aPs^hqr`d%qD^2)<=c1VfeRMi%-9jVmH7E9HlAU_Q{g z;Fs;P*Mf+OlEbBDo66yw(2d$$Nn!9)NwigwB@DKmr)*1Zs2gd~&&p zt$3p=?|dW*t*1kW$Kew7QFYG->bJ&1aGnB1)W&6#jdI#Z!6Oii1W=gUcqt*k0fPV= zGmLoQTQgk(tC%>aAb}1R{MySI%FbEkkVFvp86lv`0u4D;V|wARGsRr=;0M`SX^!Pc zo{2O=@<~u6GtvvA6&EPqnk81jtvKs4F`R(Mf&fK<)ymI}Jk5N;kB%)T+8?!w?Jc}Z zmq1C|^Szt)c21jX*fWkM3wM8zNAwqFMk2IRsgDes6f86DAf|!S1svC@cw+-+WVWc9 z%TDW@Fa6ooHRh)qlOqP6q>4qx?hGltQ-SaYc(7oc7r`vSHR-|sSDO0wV1~cm6OF>* zQG1Hx@KH(7ybRyaD`M?Qt_iFfplC_2XF?-n|=-ujJ+;E((^+< zJTpLqjU$~%XagnDhWoK{V~AyXq@vB~$Ov8wIo1M`->Z40FU#%CUu)|&sh<|PBx|Cr8t83cP< zzzg))Vpjb&9L7?>xzeGM$96FsR*F2wLZ&L*SV0emBkH(nSK2*qqMGG_VUGJb^06zY zhouGWxr(`pXGbWJU&G1i!t zQPn_EZs(nex`OoyKc&o`+teGGpH)wQ)`metuqQno+j_Rrz_}2Udq)NEmP|jf`j#@+X*yZ?1s>mOwGe`%7hGDKnrTooJYE3=47J+Jgx47Q7)LxQ-weM!N0PP z&(x;$l?aN|^n{_?^}h{=|24K5Ob1=8@lLBffPr=et2wGv&Kx6KZ#ew;*tTyaq5eeq z48QEb@re?mVRU4vvXRk;p7hU1H2WtcM$mzbM!Q3Exn+O4CzuqTa2IjqLStFb5v=43 z{60u>^f84F8i@tZ#YuHojG#Ve=mFw`KuJF){eiloc5k@v)hCMe+P{dfz0<6>y7g=J~{9lt`0rFz0)M|pw}BXLa-u+_!lP2P57T? zSe^JZH2f&P*N#+qhc)S->V9?{6swa$xM~WjOGD2iK2ib};B70+}A6-ambB=XN`P!0Y)K_s6>536R3>bj}^5!cBWjNv20bk#_(^K)SzG zo`f!%7ah?ok{YMxjmO(G@bIKVnq~>~8H+UOl0~Wyc;R0EnO@b8yat4mX1MtaS>Ij2 zrzRI2MAD?Wdu!7YCI>)4dXZCu^5kg&0I0FV?hsSYHpP}zkfnE(m8*2do~&~*Q+Hs? zV9%lMU(%~X--&>?W8fgf7!v~VdYkF~H7#zCz zqCR#ZWa99(CTk!=*SK#UEi_5ZMSB^(6tp# znG*2#QADot(JuCmRNTbWl{mn}G+aOmpW)VNVGY{|O6&c*Cd}8qqkdZ2QzxkZ+$flu z|Ic)a`|`^7bo%w(FqWTs5J(`FqWXXdI?9Dgm<(wTB3Savx~oK!_u(F%)XK>S6`CyW zZa8={-$>V?Q@2-Xn;zj!Md5g4WqPm_F5USRq&op%fF0n%yyom5uX#)%Ahf>*dmcf> zt!0wtk*@*=zE~C#&E|9`u=1m~3B`HZZ-KSlD<&Qv0a(+uyj zSM`cPi9RmW!a&nCnQQh$<|10R2;j{Dp7Sp<vaL@AV>tE=8{B+ftcw+m@dt^97N7vXE`fKO} z0}+7Qzfg4@v$qibC|r}B5(~P1t4unVuji^H2-@|w?7Lv8x@_mapNyP?lbAAlmkcGT zCJ6ZKSCtwUYZokeuZ)6C?i!mDd`ovI+f)N5b)`)^S)-%0;|iYqN4L)c=eg`d}?82Mj;Yn%r_w zauKX(@Qs>?+0D725XT#>#JaUs4L2yrwzAxI%Lj4yq4CqDoyGx%<7~^W+sra<=!p#! z_t&j1;8K(SW=90++*sAyNmu^F{5VG5C+}ft>9D}f)AczRw1|g#*w!+lpRkV}4E&Kj zsk7m7hJ|Co*Vu`Wfs0(SMuxqbd~=Y!>3ehXOc(b@b^K zgiFAldsnMM%9#A}#{@+nihK$mwOM5x+J+`dM%~P{>ymXls@YcGN(gP`u`62H@6G|2 z9L0+~r30>wDk#vloVVD#cK4x8L3Xg*nNQ%yE_;y&eQ)AuA=cB)v@&1j=gwSjR&t+L zjcWpZX1{gFcG#)tb`Sj67ldf=ogc9=f!MVcnALwIXD~nRh{*WL_P$%4?;ZZpH*ROu z@euO4#(N=nBoVMMVTXFoA8m?johskk6obTORMc$0v?-#wq6NMOtdb#NT{hSP4*5%2 zLy!yy5-EJ{oHr@0y!5PXA`!n2Wr7S14$nPO^6y~^dDj59T2*P!0E2#P#BE`ziz&VX z5Q}^1bbwY(>5{UY^6v^E5WojG`=1l=7vaCWqQEOB(fpT5e9RAYn#3P!-{s*#f7`9X zq#fBCQ`451S4=+di)LdHUw)mck|h0VKWY~8Kq^llOY1Q|U)EMd!^rxGVl~+3i9j4PfN0x3ZfSEP-e1d75yD({| z#bfT2qr;4`E&xX3HeZ56W-^IF=9_NhqXBz!j!VsN!=G687k%)5Hm4=?Xu8R3uIt=$ zUHsebA_*HS!lUeqBkN^wc*h0+SPWrY(`Z)y%fTKkJ((m@1*H9|c=i_EIUd{pY&KoF zMA+#E1wo8P5%Czu)xJ%MaojuPL@yVV=x1Mc-nk(Dp#X<~4Fg)0c_+>B3C~rsDrjx+V^I0t)}w(*}GbPW9n8qvcok_@ zOO*%Zuj3XV(`1x~edMka4j$%`WdR4)9s3TQpF z3FA(=6LYab3_STK2Wud)CB8v-p#5RKh=>i5bm`)jdw)qfZq|$CS1J2|)J0K%3cvAWhoBSLsqWj8GRg z%wMIP3s-QXvl+lM=w;Z@+!^V{cCN&UVgJim%Bx4Xw6YSXGQC(s|0x=~dd^_d`B zs4kvW!hH($SZR;Ywm%uCMRt2Wf|N03B%k8i{0ri78T$c>)%6z{w2k^Y({&jIZ8r7S z$CvzWxO#@~P1kwZ_1+d-lF8xzu)3W(Fv8%_RW;f=*xa@Bf&cx}dNjxW4!?vEY#5;# zae8U)_h;f`R4!)n8x|yq?1Y7b<%w+7Y4n;5BN#5;vW=v3dpW1HPX1voisH7$ywb^z z32(NB0X{*=OO-egTn}wpfl3X#+vFc!BJZ4;=}@^NG}gop?Cq4CqK0~DOz@IZwd2v6 zXgKmHuTK}8wlcrzY-oY2N#JN?%|Lf0ok%Ql=>dIdjD7mAa1IZE0=9tY-`~(+t}XIU zif?}O+n1ZR7-Yjb!#Z)78%W?X4ns*LK1AA_Hto&i->~huEqTMS&VOr3GYhJE59Lqm zFjzo_35B3-@T&(J1=4kIXp2Y7$uD0?IY(2bqzLzOS6$s42ZvrwX+WKoE_-Bi6PWtjkRhwA^W20G=Ka`Xs(@tjYPx(h(+Uy=}jmdQ%A{t32Ct zvL|$Cx;*xMPw2&S+{b}($9^s?I-5(7RDCdwd|>R_`>WoN_o5y9yNqFkP({THD75aV z4T2a)gHr{>;glFkB9aUX`6f2;b4%w(=(+&d05bpaJ*kn;V}A)+-e2SR_$)J^~2|r}oU&wn?a&8?b z?VM?@S%p-1Kb>q!De^Jw@=e{f?SU$ANC+xan@ad?qG~W|a7~@lIM7#>fPMdKw{_!9 zP&n#Py~?`yA}#9&{<9uK`R?S;rR27X=5L+kn4--Ya4QfboVH)`YJkuN^Qk)`;@YWX z9xoBU%_W|!w{bN&ShL>Den!~pcjL-e;ycG*RCBU3XV)j&o^e#1PmICgrid3!Ezuj) z#gsjln&vd}5THFB4|7Psn)t!D{WF#Q-+dxLGEzwg4;#9SOmLL&dB2Z1@-#sVqqDb| z9Mdl06L=SMDN|x66g(V*$S9B-rc5DE_QMph9x{2+BGT31ScX!w&^VpxeD*uIF-{1a zIWZ6J#J#$`Y|ti-rGdc<`j3T57fsJDrjV3IFDm`s=G6i$|H!agWcK;)Bz0e(!x2%NYscT1S=pxK{IZJ++1X*-&aU?yTKxPQsaR zf@kJ~15hjAqCl3w;0ieG!VM$*C&rj^^z!4Gr&3>MAVpRXYkZIT<*5pyb;Iw`!kH@{ zAGS9BNYNY)c=SqtLR#ri@F@(!uFphVfbZ`srB_3nK#g=Q20nBm(HbdjPj36-5TR*0Z&VJTpeWyjipfekc8Xn3stP4uf_Pcc4&ZaA zjUo!QzYQvZ+TgxpC^SGS+48o1Qbz6ebyIeSxG9*vqFqnugVPx6m3>XH>sEf_u26bl zzSQS`I>GUKEp=WIjpHZP$9NSB*_3PD9`IUcP6UVNF2CopX*nx~!t=^*k0ONYjbL1+ zXu2tH=H8}oS8H>Q6P-|Ni~m!C{*&#+!0;7(aetdlh9uBFrDU$nVATMQIu!xPXUbgO7c~#D*k90 zblv*P$H(>w|BW93%kDt{eGyj(Xt3e`Kwmsf4`+*FaWyrk!@N!iV#HBrHr`jkWv`Q< zPGgH*E@#}_m=SP4d=G);{o*prP$zD1)v6bGRjeLy@m;SJkACQ$N2Kn#y7I|d@+4j@ zTt4bt9{Zu*tZ((4xy^pt%0i_-iG|s=zVwG06APkyPDFE4uGKR(d%b#s7y3gq*D-IFtbI5V^*Ay!pur)(!O~?W>+R!D;Vpr*mE;voo{wT8*4|v3ZrVa# zOqUr)wOsce2a38YDVlC3LfC$Tu9y~Ifxy&OUtS{aY*g&|quFlCOluNo2oLY#bIO7{7avB|2m4SK z7kX(?m2e#fzjQ8D%w(B|RGyGJOnXLwTW5&{qZ^wZmlH$k5FeopVf*T=Evu_t#X{&S zpGaS(<_Q!vpIQrJeSXiPf^?!4aL=1#DEo4QkW57d_Il}SAB{S8tvC0g8MCub3LcBlPFO57cgou3Alk0Q8 zxEK&OnAu=jqlTv^kd8#wpt-Y9`~ET82@7E;0k_TSwB_T_*B~rOTv?k2;nELgwN@g?Pe(}W82h1n#^ZsO5I4`XI&C^vBHzG%?Mpw zumLN=zEN9DSg+lF2=BBG4qSMtSZxb-e70Et0+9`)kF+l~uHNTQkh7q9aE)2Uak>tN z<6Bf!Y}PTm3!9DnDz<8G@-X!O*cQ0HkRau0eMDO7(~Zf+xJSf$(T^PT{k4XoID&B( z@h+-`K9Wks4tHZo-8JZQFWVjY^m%D2jb-R1Qah!DosZd{`)-lenw{8iAe$B=KIT+l z#PfjeFD#I%fWV*MKc^l3^ZQ3+(45L70fT!Cem#O ztbie|e+`4?zL5XBHmu zh8)khJlKw+>`@fZ4w~SU0Cnlez^&yiVf}H9Y4+~bmd<()a+O@BgR1ImPL;%s@$PGj zyM%sq6v3Iw?YT9l_D45yrqw6wZli97*7fG|Vw+(>F`vWp2JkkfZe}oU*fhoTup$og z+onybtc#v$9yjQ!XOp@E(f6T^6G|NZ_#Bvs;HQdbd*N5qEdbyE9svCBA(m>6A>3Y? zJehBU3sVDY7`lE(iTG$iS@Dw&6GuxirgBPYV@M79A=!A3&BE(LTiUtsrJaMS62m+V zrG68q9jD=9$+*KC0v_C>^Oe(g=~n{vy$wnO>4bUH9jK%zM_D~@XL z?Nhu*!|&djappcl(0;|IB6FQ>hT@>)u~&i#nH<{LY!)IXq`zjCem@Uc&plkklKzfO z`b+h8;$n}|Sku-X)VsVdO1Jv@B?(P}OPcz2V?8m{5v1*o+E;)2{Jxy|WY9T{K)~V( zW7E5U=c~B3foQP4wPfFxR$Ko6csu`}LafuQ&g>9lepI%bYDzF~s*g=9ccprA%P=*VEV$4ey;`q7HKwC@j-$UB= zH^T3m_YyG%!&MHRr#lGn4}z@&oa2I_M@CEOMzNYU--nE@AqN*ycb|E5kwThg!~AyB zH#&$TW-#6&)4$ijdPX#$BP3S5Y&qmdOSA@jztJORIhHU>YrgEV{8{VYrzB>;xj%hu zzbdyq4D`dtA_>V)mR~x;+D=`ZZt@&?5lwV9_Cwm{xO@_qdRgKAx^>@uK|2NeE68XN zH<|saJr;52=!AS``zet4-N(d^l`uY*7g}cFK=QiR^vugh!XhpD{dz&mFqYRecm`&| zQ{iU#3b0{sp|^F!*Tv2v_n_gWaB+0s%W9?fn4qate_D1C`YeidI&d|7{uv3up|~Mx zv_GeolxthQ16DI+lYNj&A7Jx+B zEV7mZ$_B^A8B(-dlU3-MaJhspt5~h2FVo_i3xd^?R_?7CenTDSzgEJh5PM&io*->j zY$_nemp)gS9X^{?WQP1#v zRq9Faw)Usa~TA=5%|}N8SXXH3S{Bc^@o@H8bmr&G=W83m5qYSRYo8jKyo7YecQRu>?S&*X|T}H z8lJjON6W>WjhB*v`Ty~^_zHjqAPG~mUyx6l~cuh|~gfrml|D$L99lUUG zATg<;!4VoSsWxF^xzJ36#8Jt#p9~~7+ts#^jx;sBtES**c8{cWhft}FjK9zae`=El zXhnHZ-lax+*o8qEc$ia;4R_6y%ms)Mlr-qgKFSc(24SYV0@FiE#o3i8sI&C`5~csx z3g>0`Y)r8*g&On248-sX?7QDZEFFYHfLoTc*pX@&sYpe4E^%haVaF{Gq&86Ayl z#f2mch5TVy#}gE360=Qiv)wT9~WPd*SSc=b^P5{#BG*_@8V7!h_sDi4bp24m1?t_I{^Sa_b<&gUNV?*@wL+Bjsd5EcD($ zXXwd4=?^^qWa#ViY-z`4RL1u>NAuaC-`t{Sjt%j)_59wNU0DJIJ{xrSzcCeQsmVyu z6v8V8!XarhinIQ*3*OQ*FKrD+4lZORaG)`W_W}j+sUN-OY4bcBfC-6=CO8iu0{MTg z`P^rWC0t4)KBtLYH=8t~1+fo8HQzTk*?X6M8xR4W zR;Msad)h?&w*%&7YBI8ODS2zm(u4Be9Ep6i@v)u7A#FlKP}*W67{59kaWy;S%d?mZ zJalvf|P9StDBth~G7l!_*#Zc$$h%OBO^ zerB;XQZ^G5X=FG&)UJPfZ9h&BtFOx{Gd*;zoMmP{k{B{;h5M}A5rdH-c&IQ|FnT`n z>6g^GM-gpx_tr^8>K=PUG){y1iE2qPlIS|`_YT~cZ- z?~?;QT4C%G8b{Y{PCodufXNT}f#dZFOPvAU&AW3I%qH#vlrBIgG=OI{~gUR8Vc;se66(ps#>)uJ6~$=ir1yZm&+EUETVa6U?E zBllty`G$D(+VF?zDv}-!SubeOOcI<9y_I2*HxMXG`RNn!mKZ`Y1enOre0=JCQZ)Xd zMsSLcL94WJ)XlQJoP=a;IClCmKq&_bmdg6Hpzb-Fleda)M-8kL1su~s-1}BmZ2HsJ z?@WJh(rSpwpfC81e6O764~%IcdKGPAyDu_mxAb))AYinK%%djTSXOr(sY&6a=0?L` zYgsDHUo4RF&pv*M?A|%q&AjhEF6%LZ6t>56pF3IQ142%tWD-wfSNSgPj2mHmT*j2e zo0#)de-2c4hE%I=`9c=(h5T|a#`}sd5EGD`6s(3h*KWro!;FDY1P%?UYq{kkWCsC= z;Zw!i9FgNk_-YND#f^TNj_kduB5zv|GB<^n8KIViTp+<2jq&P*7l~(6>1EH>INY+q zpDGAy-d4>{k;kNi!{QTlg`bQeHUvCB4$AeS{jff62P$CBTpouSR}l!fkIl3N9%o9x z%eHcM*G9ra>c3Penlz~&81&Q}SP9(Z?PQDH0zf8%1{$sgjf3EL7 z!_2%|5pT;NZb_^e#`7LrS?!8!i)TMnE_K+DnS(rfbr?qKHX)E#BC@<1S;b8A4ffaW z#{!=f)tnrF(q2d9W!{@(!z>%9N8n0zU0L)h8gP2Lo0aF4gi^PWLHd zw$)c4`|$ieZCCNVTf)1YrrFGubjy)&)K)lH5hdna$UMCAoHFB(%OHEnkZ*-D^GfWT zF5LA=sW8*9=GDSNYiO+PceqjNb%=o*OagO1;D(1~9x;;Qn{pHuZd9M~j&`knN%FxZ z;s@Fo;>v1lVzK=l+;akV|4qJ-iW)cWZv)NWpvI4lg4$oz&rgDR-QNr7|AqQ_(lg(< z&1W`yB-i=<5=g=cbXbH8g*F(HK)IBn(G~mF_#i^E^DyOen~!+&C?%b;nv`4#8f7WT z-sRRo-vUtDo<~ocHW%pAnK~XX?MYrK6kY#9$$ySIPcDn z%zs!=FWvdKKk*b8z@Eg^mOhy@#^i^g9LYMf;^e@Q&K&!lRD362ZW}IQMPDk#eFi zGl^rtZN-UbZa|&o=h~jksRQ*}p+~docg<)%aHoW*etk}9K}4rWN^Y~+Ke18xJ>Gm?kO z%-75&7#TB(B;P^m2b)Vs6ImP+pbb>lx=0fkx-wINPDpud|%D^J_zFXvy^Tuf)=3F?*YyayQ8{zfGZ8EPb z)yD=NAuUk@Dl9`|o3wV-72tPn_p>ur2uMF9uy0$sk5pc0@aolFOy$yc z5E7#I71((Dzp7mD-k4m`6^U#bj^}1r9I;`?F6XK-!SFzreEcZs2{!ft+C$rt1solY zL1-e2w5d`Op{OTmw6^plBAVQw2bpr(UGr2DfQXr6A`okzf56&m5{d&V?OFP9{V+pt zH9P1lGiNfJ_*?-3RPqGrnupv=ni9fgtC@(g?9G zqE9uf0=wOi-jHq~CG2Ut@SBU}bCdvL4OH`VYdpQF)o4HcLMOw^7uG<_hdxU9p!W3; zc&C+ZF~Fz(nG!hmh=LxVAB#j3s1PE-+XR_aQL0Bq*O3k9e>eTI{~zVt>SDW|v_Eq$ zm|Wn*^e?RAFF|}9#d_{dU|2$QD_S@mY#{u`yr-R;0<7YaO@O6&vi1$750BRVyRGI}O3|!d@^W&`y}zTy*n(+k-vp>uAQE z7u&DzTL#vfMrAk(;uGcA@4Yima(gp5%wfGpjHe@t^CexX`9iI5FWvz|QKd4?HT?$sS)NLPnF?k2O z6wjRHoW^|zZC#bvix@rciHGazvo)kT#^52dr*bZZD&Y>mDhV%N9Oa+ zhym@QaE0=~BBZwe-fl`uo2#y?vGtySaOI8A;5QZaEv9c2FE_rCs}IXx>rZc=_iGJLdyhQ0X=RFkXKglmFTii2#a&5l;ZEP>rUk!kt*OyStE?|G zPgq#%OSOwx|3Z-XuVQjkgzon=3}-vP)AWihFhttpg?u_GcUX?eO=7S(Y?$f$mP-yuhfEw_c3~^s*&4Y^8l*Q!5l}EnC~{eV0~)N@ z^LEwsH5hsAr~x(K4c;f-Ap~x+X)uW%x!q^bM&H_fev9su_lz-_RMl&I9nlLEE9ruemboxi z8I)~Sd_F!-bh42i{s(AdlWhm7)su}VVFGLCach6muXa}Ov=jN**+{sD^lr}8xTAT3 z46D?~Cca7(V|i(MkG^UPx;?Rsgzm-fW*f6l#e5{K{4{wfD9((8?oR6e!I6;vqf2+} ziocs$8UgE=&jfqmke0ow#iJ@;hrUlXucQJWSu87Sk9p0XbGe11CORvmO~(Wva41D_ zmg_K%1OZg|0vVKaQ7+YWnNqMSC=U43#&fFma~K4#Lgw~j*xIZe2W0!|Itb+@w0?4% z*F%UPbHgXdebp)4qo-rg!<0{&-K)KmsHxr@c*0*cmAC*ij$|Xg%cp{cZ&eX#K-E z9^b}$XcQT@R_4o%oELFzoJJU(TmeAI_U_n@6tJ+Pzo{m3_7-zHqxWllR*=~iGS_y@w-OS- z(haz|7wgCXc2Lw*RCu(kTva{XbI~v6!X@v`ycL3*DoWongTLLq=*9 z+a+V-KqCB>us279Od9|#TdJ=%Ei(q|#4JiBo&q&i-?s%V>wY{1Fws!kAaLpfN@0LK(`aYXfNl?m3OSP~74?SLY<8+n_Ff+iH8wh~9c}lA z5>iB8PBsfoHVimpsd6>zqh`bzzuyx+C!qd!I&qzSm%N|*9j7|r3=-Fbq77l)w>e3r zTlTB$dlLD;4=$zGEVkbf)O|vBKwY2?&Be(~Af@!--i5_6KuPV~ zer&RdwCrhw9aE=@lDYVVL#P$ajsxPr)B_3CiI0JSt}*!Wg4}TZNNEX2{ecW3GZ-Sv znIP??4i`hwR6^a0-i2`GBzlyUZngyx%KerN6y613E&W-Hu6gwD+%2JID_~?Zz`1f0=eE^dzm%|NohmdkGZ@lfj&J$NNPf! z8_zl}BL3~5JypyWfOzG+BuCb`jaA@3@OvD%r`4X>I3IE$)6Xq@z>oke+pe%5%H|=UZ(& z1sCoghblq5)FYL2>FBsaq?Wpoccz=|5m0Cu{a@(e(gDwZlarYG(U1V6hxaL=-SOCl zg#TtvPZFNY-Ag?+r(fI#{C7@{%$2Tg6<9d?`Vh&w!vK#=6Vq!WN7C{mD@t#?5dK|mS0Z?C)dL=oeXso&u`h>kT|ur;fBnoVwe9 zK|)YCc~FX}x!hZ}M|em|p48>LQ1wvRnh{5x)f*L8TGGKs(K2;il!H zauGe93M_?_^o}7uUOnsjv;C!V5lo25TW{M4k)xvC7Y9V2v#5F&7Tu#Eq++g~l>B57!2=2H<6}W|te|S>KB21U776 zm3C`{=(I4QmGg(+7ewUvJ!U&tXS|*!q4?!et=G~Sy!=O%lMxx~?w2liFIA@RR#~+* zfM7b9jQ?o${r@OfDBPkkQ&g25;>D+TnA2r(_oLR@vtP7TrR}nxj|PWXaK$n(+ajpR zs%5D)exN$)ocQx!m$?xtC`@`#@oKyb!o#;bIl$*}#w3N#Bonh9BYZqv=S3wA_mxHL zDIqeA(7V|IM3C6iDQq<}jF;~4W%+|VIkE@HFpG39&eWYTR(c@2I${57E0hVK!W7m2 zblYM398L&lZ@Ef|l#lw}3gvD;Q4&dxo@GejGcHphHq1vtk-!VQ1}u3T69ku(pyDcL zCN#Xr+tT35Kmn4I_F>EUeM0bV$Q;P`Tba5)&7*P|4?q3tA1O=G6`(W#r|5 zZH1~befl)h9O8`rgCM)2SM`J-yS#$$gdm%lfSTCp#?m9}&quVvhc_&vkB!QCHM(|8 z-jrWXipaY7)pe?&kLI=81Z4evW%(C^FRWN0o4sL??DLCTOG-HxJ}xs#eOY0R3D7+f zl)jlzKzMlXDLQK*Ck*~15OSP23uXlcmsY*bUU+o4|B*)icp(TKX-DsX4KQ=vaMmN* zSxPeLXqcLv>4yv9wAMSb$|{Y`?GxImxMMf?1kk63JmhG277`RwmF(}T*3TEjYKb#1#PhYz_TDr4owndV69oE? z1aVGwXtpYbUE(BPzku@@(@fiAD@i(;NK~3Q54Ib53h!VbM=J{&ey8qi;?>u;m6HV6 zo!4zfinI%~e4^=~`LIyHp5DTzY`pL*#SCqlRNB~Mn-Q(g=cp2=8a)UfNU{F4V;w?_ z<+R10B=L^>$zvX0E5O`~ZOM?`=Q|mL^b=Rqtqk2Hr^RD|9e||8~@L_KzgxrVz9wjcSK`uE8v5dmkxkQ-L5rj}_%1%hH4va~MVDbJNrg@o zOeOivMV+2%rvl4q>-OM2{b*;ythm4@s0y?X+7ykhQ7O*S6a6fgr1yj`Z)I}7Bxsu= z7E<_)Cn*=5ux!0~+<^awYGe}eQrwDQuy;{S=`G?L+0ih|fj6PJGKS2U#8ILY{d7wappP-(&f}aznS)dh6BxLcX zOt|>k?Pj{Xks^b}So%PEA*DNZBOM?3F;#uNhs`1D%bT3lPdvAI7xVuZiOF6c@AxT{ z@FEI)fafYs3!}|w&=*e<7tItML zErDzJ3RZqL^ZB>q9`#)EpMKf@PXxusf2rz|@}B3kWflb=GWauegFU}Orj9oWslC1t zPf&&SFl!iS5eZBTf{&>`aw@k126+%_+(wQw%m)C1@^$-tdSz~k4|BKhg{6#i;beGR zv4moKabUs}Oj%5c-g*v}LvvEYE%EtBzzQe>`#99*ZlbWFWV4<9*6pj@%ID8Hlg2CR z?esz58_SDvBqaP)f^K!T>ZI}v{tv?iaA(6vu^TR=7jdah2j6&zS4<4qa4v@4$des` z?6m-x>aQbhINzORs+bfZAgdOu!vsTs2r@k)Su>x6p1KHncOevq(xMj$Zgf4+&IPJh zXN7gx;yKH4$fu3Zq~4t|4y~vW_arU2drtD9e>Df})?>2SvP2w0(LR-O!sas9>JbWz z#nh-hF{&px-RGu608mwC4TrdVQCi3@|MKKh0~Ucsjv9d$6mo~Em!vko$ZdAL1zHAJ za1hip+1jd2bK;(7GLBLA=Ng5NlHR&DXRYPjT9Qr^3uPE7%&YmmM}aj^r++ojKWDau zzZobush*BMHUk1i=-7kx@wN;J+tulnyNj=Lg#1a^S4H^-GgOiUg8k@LhKtXIpK9?R zm}v_%sJ%v>Ok_Xh2u)?A(VO_u7?ddd6$$WD^ zNtcs@$Tpx2Wq2j#EJYe;u*A`vcBBe=lMrAWewiO=MVHJSFFLgu1y&v|xpEtyWM@7O z?Y*LJK-OSBq9W&YkZ{$zxjRPLpu^8T-16LH5m=Ea%JjkVoa&%3z{Z(K?K?dm1yOIE zjm&euN<_)fz*NJUx<2ntYQ&_qFqcYs>{{_RmjlGMAcE@DXTHG(2fwC6G_ zBu#FEAjgx#yN|s(f0f`-{kel>$X+!|v7&@<^wlbwqu2^18eET0HN7Mmi?xGhE;09*jy9}8WS(DrZ;Ed$-Y;2biNf}7+L-{X9wF3IuH@P1W+avXJ!{m%H~ z4=RaV?5UTUsq)*SIQeJ3$=oRF4n-Si3GJ`V2GKy3N#?s;a!U<83N=i!+_0QtGAxEUAoLNEpfOqM0*h# zjMpzU-Qb_{$-VJ~ApD7+?R-QW>!gpn@5IQrINJHR!>{rdOYXp0{rHSCU$bp4>jAaK zAk>uI_fk-#;Vm{x5^M*AcqP(A)ed$t-_^)dJ=SSQ8tLdzUC~v1CqWx6l)hsdFXPX& zU?DL~v|z<`$~!&c*JBbqKoL0o7hV9oW3C^5&HWF@iJdS!GagM}LR1&wRi z_-GM*56G>i&J&Hyx8*>?abHXrISmF1uW#Pk-@ZSTCwEmZNWi4jG)}Mf6#Ye-QuACZ zoJ;jDJC&#j)tFq5ZuHIShsv@V@x#X2X^M|Sjy5F*$NIBrS*cvtF4*CNh>of1wzc^+ zUH~D0L4|B(r#apM1A6;Hm34d?C7MZ>MRH(BT8q+K=;m(6Ac5vLCf(|#*W<*#ce3Tb ztxiN$AM2rgK8`2IL0+}U&JjR$`v{35zsd~qd{C50&hZ=S5`S*&QEncxx7YV(ixDzy zL2zEQNolol##QF-wA`1-MVn0jx%>l%QoSGu=l!=v9*j1=+ z(p`=q(~TY!)@7TW!wC!C(E|~%8Jq&S`FoSng2Vh~)%FtixoG)s*3ILn&6bR*?-^j6 zTW7-8P)jB3Lxv0c@Av-ZfEd8>=aX(2Kc0jqx_Z7dj#~K2^h<*|#W<3))v=Y5HL5nIRCWTt>u(HYvJJp;j)ZKZxb~-k*ssvuN~LVb3Wn*qiTxw=;w{`H%q+M_In3Y zoL?4SXSMO|hiu{it5g|waNFvqW1;RBo73N~e(nST8*wd{_n)2A`|L`;=P4>op8A*J zIT2hQwMPk6{UOb1cOjM&LNYynGCWtLAk!4X?iLxE6*@uev6UJL9lvL>YO1^s341_o zQ5zhq(ikglSPccRs(JKz&u$@TaEKUot>08>tJ08ZQBh6QI#GIE&iGkSXC#sSsMPj& zMB%}pGdzNSQJl&-5Q4LhDauSVlkpUorE4=BW|V>B1l5k@wauvc6xp&s;>M=ghu3y7IbvSh4L==SU!^ zVbmrJrB5YcIiEB=&B8vtNUPj^b2o}}*WU66wl8}s3Ty8Yeo;JqsvQpDbG^S(| zgB4L*Q4mMHn*ck70J@$i25;ck3Kc)cUzIQ{i(RZUf2##3!$XPH8mRiU3V{XCMbmZ-`-X1Karo-D5$^Y?APZjxaI{H zYW%d-C)zEv{YrA&b^EpK$BOO#^LwAR2cU$)J3of!XCCeR7@n&O?~a<7yY0TQxJu5d zdC~Xt?l>$ei*a0x@GpwA>Ho$hOhx@m{qRQ^Kf49Sn7zIpos-e57638s&9tOJ^rTAO zSbeo$RXmUd4QJ?d9xT>}D<}2Gr&oWh%F!xNlbSLvGtJuzaD3)lU#=cVPJ0>$i_}kH}D|(7+_4-`> zwXyQcHWBPr(guZiq4?o$tbB8i07XE$zoD;iJ9-|<7*xN!jWPRa!KJ1DHhXOSAm1tc z<>x?QVQv8~`&e=62-ov8Cvg1Pmz0)*ZVo|BM(2kWGfB(^X^T3%4<)qsgMuiA&%p2r zoL+VcQ}{TIC=KXF~33WhoygB#xO?N?n z3?O<&$wO_DDt%fW*{SNAomL#sT0x8eD^71dCvQ6+k@En~bJ$I;IH5fI@axfb6hzVc z@cDxy2NL7L?4lJ%Z*TgYVroD%ZKx>crpvjdwA)wIBXEG(Uc4MKu3gg7tkyYYW_PN< zcg8Q3UkqVt+1U!xiKCyd zSbKyghgv36kvtF=E|p5`wD}zn0#)-!3p;XglL)jsze&S75jX^^#0sJt z08)O={==DB1mh7Q6of~u1XEiPUWR#V68ohDHQ`8QDUst|y&S@-aDNI1mpzwL)tp*o z&UNHqN8Jz(Y^cZan(d81;3!yu;h8Z4yZE!MUBh>9C+BjmUSb->12{Plx$25^$+lO? zd58f}qebUEFlcf!Dz$-b_k;aOk6MexY0+?RqQo(&Pzna5tC(2AlrSa8)hx#clNM?A z9PNl~9tQi5xr&ner(t_d{>TiSmpwyS;kW^l5N*jp6_Ab&C%>BD%%5?#eGN^b;T_uLpdGf&;9Xp2=(WrTnQb^0JT4m7USsLHQ7Nv6He44e4DoEyL7-`XA zLn*XJyw8=nUt#%kBc*kp+3z(1)~NnXarqyETJsP#s9iwI$%_5h6FwN^Km^J-uTIBM z8Jx;N+?)NN>3v2NhDgK}>`h)Wgtg~NXOV2L7?)XSPSpuW{a@_~iw^K05P10ZX~oar z31FF-gQ}Xv9trb{t?_5Kmr}AELI~KgM#XN$U~>Ckc0jR46||Z<|09>*c;B6)n5hw5 z;vFhm`>};$mVuT~m5u4x7It(XFMyAfv*?&=-UDq=O7aby55xy;rcckK6+%E?aHXi; zx!@=*l*R{@d zuCpG6gna7(CC?=S$-Cr+Gn#BcY(NjfCv5M!_VC!fXiIeT^-<(Jj3P?X;{giT&9@2ysUEY5LS_Z8W2-SY?_4vrBsPgw09?R;MbjnYI|i0R zx+&L+ywYy*cI-g>Lpxr_!4$C@U7QO0>zeT)S5g;LE?{#PIgu0V!Q$YTY(g5i{SD1* ze_0A32Nf4~O;z-S`f?WX+-^pJ!jkbyb_r!esrrRI_%eq?qU20i1!b#CC#12-WikVp zu6SE-U@^?c-1tT|=vS@x4LZ&vc1m!in}DKwwMfWcT-0BM#&F!eXqLtLCV6_9YWDhQi|N|h zLmLHcd1e9X6n!J5Hng|^DeF2uAW9Es!?(4MauFLa$SF_5gw#?bK@6rRYZgvUc7^HOjt|6s42pTf)JzhgvIC*MKx|2l|_G1T~P!y7zNzjhj-oPCt<5=zPN|vY=ReNOmw>79&)h~pW$6D`!PXZf zjz=ZMV1ZVe+)bzbp4#H)*>DSick0q5FH>O@-B#*u%P)%+(@%zKFfS_j%C8wrMhyz* zoHHllC*1b@#%)Pgt4kf<$`bQ^=eB5B;R9|V+t?du>w;Vo}YY)m3P6w}_}$$(6lTTIWXXFpkl|olXjv+bXD-w2%zmAkd8NgwVIn z88*CWU02DTckOq4DBbijieslqDZlo)if}~{KvsdS4gT3^h)*OkENdZpGI)*K`N@ITh3sx zB)9Zj%sXY@6CFKFwV~D3e)vs@7=aWe)BoXiI5_5{cA-P4OdREq->D;dlPxSOp9E#( zJ|NpX1dcynpO#-#Oywo)-T3TVU&vd~SwNIPsabaxcbb1{rhJq7MZS4SfPd%K2ind% z<}PP2fm=7H{_H83fRmBC0rdNlL}+<{i4_q@Cf#mMLVC!n-=DCqHH==$%N%@|u!;sd zfX^$A$f7O$woy-x(6dZ6;2FNe($(`r&r=%iyZ)T9&YDaFkDl#gaOWNjD+LeYyu`J?gPU$<(wM3S&W`ZVN&)ySk9#AP8-~AXZTHm@J#PNu<2NBuO zOkY7!Co4L6|3JAl){w*NhICEC9+>~!*QNtp05lI^b)1OxFREX%7-J9horJF1na`IvD$hVSGtjj&LvQ zK1n-@9#CPOILn?GMN|T$Or&YtU8E1X{72E8NiKInwWYU|+V2&ctXFhzU>6xX)rB#D zurf@VwDm-F^;=R~7I33s2Vi+GjZEr+!yJ6*DT;r()~Ed#V`{GQ8+>B2)gGnSmABO& zn(%&_-XDCiFup!G=f)SU{s5E zEyVY_&*ezePKIOZ&u?BIFaR{Zzj>i`mjCP#4cjJFVOm-{nEnOr;2%kk?_rs!!=;k5 z@*?Ni3oa|w;6F}7%Hb66oQ_Aya;i=zysC%@dxPc=?@29OOh1t;^lkOq%dtlmnnEVd9hJTQn&I(EQA9eG<3#%vEX`Ck z&)QqymTIL&M(&NE7wMX+CCz)E2qCA($>qz|o2g*c@++KwJY3lvd=X4miQV*hIyc>a zsm(QK|4|u^bsvnT#qE}QZ-rAU>>s`2i90eh>9j-rD4Q1i`u-$TVdwZjP|D<-YRG|z zn4Bb@RwPNX&AHyVyB-`R62&_1D4G^w;Ud;;tLs#=?AZgu(|izNE{@a{H|ZM8=A5yv z#Br=|;RURcc()VcB<*dY+|Za?K%s9UoGuJ&#y_sJ3@k6jon3mb36uW#{or`cHUQ%O z&e;R#AD2^pwKjXgxeofk@>>G8HOj2Wg>*FG>X-W5$bpn+>2204hrK>7JkOSQ#7Mqyg)QUNeY0z*SE*z_44=tM6j7V9}-o_mrns}TnD)WPsraw zoNbFQ-IAPrlEQHh|E)x2l34Rl!jb|720t(6BIK*T8L$6`ph_pm{&n62pRby%f3*p_;HoWX%e?U1DD%+H%IT&b7Ah%-nXva0Arr5?~Qt3M>2pB=-z&= zh#go&(Lj4#J4|fX@<_>GtCITAaV}DQ!>|P9b`%y^RBZT z`t&0F?6cxo-}HAO76Gt6*>1QPX7p-c6~EeCoLb+)Q2cV-jCtr;fl~Gakh1U(mvWL3 z2YbV;W>RwDdFjb1`b5bW#xA{cu3+LV!#nUR}MA`e7*6)Eevjnt??aLExY%l(h zMl6d9l)ybF239Fxb@Z!l*!|h4Dnch~O^w!Fw74_(bx_riz~VcI4L*&i-p~qtU;ifQ zUM_5v+=kF=tu-D=&00t-oA0fl4NtB9Yz0mFbeNt5Ae$B9N0Q3FPS zM6^<_(f)o!IiuoQr^#o6zQ~}DfLX?iCkV${Ex_DUp0DJ2E*bIov(ZSC_^oRV@B}R9 zdcKHxhfh>}R)rg(EZ}E?XfKLAGOGk_6>n$)psmjDe^uIh9&W0ko_+VS&%x&XUtK+3 z9&+s=DJk&Jay*_HFtW>161MAoXv$`Hw=~lOjM1^puHT<6$$9?1m+K}YhLAAWu9#m> z^AXXG!(CGr@ozo9*F$?C8;Wbp1(%`aQwO?%X9?QYxUy0ZYTsB^Rlm0#4pX4&7pfko!Ng8v75JyZnvS*uS!xoZ#mEJY&hr6!~bbD^XIy7X!soHAs3tW(}79 zPu8v+!P+%iM0AFNp`=MhU#;D&hbbF#DR*InurvSc`ESFH<2}awDS^g9w0ub$a?P5c z4S9Rm$e~&5Twh1~Ych5_Otoli6$Wye+=HoeuH$T6RzG?MJ#M(EkbQLDgQsEad;9TQ zhTcEhkCmn<9z8;BwcLzV*g1g+38tD;RVJd*ZVd5$>T1CV;KpwD%eSE-X8dqQ*JdNS zT$N(=4DTBtox zXfFQ23E_c6{lBo<&zL)D(VLO6*<$qe_SSp^G=F|o_VR?O?$<;@Y=tLW z9-M|BFC1^fLMS zX>SY>niSX!#$1w#3HZaevbE;T?gn1oD6BldmLlJ(B28T}GySy0gT;q|Y5K8g@tNi0 z2>(z1+x;QN{6^>lO%uk>;hLJZk`BOv*e_~k+76mL&EFFe|-1B z5O^B-$St*>vj)L`5H&2Ik1YRkSx2Po|GuolK`RRM5~O`SiT_Q##M-?~$S8ktlkD)y z0O zE#k~T&bnRvntrski|FQREE!PU>R1y5?+2VVxIRc9-*~Hm!E!wz{)!XR?w}q<>OHQw&JNg!R_qW$raM;~t8#?6nz8)G0qI4UD<78M7 zK~(Hx+#z4I)Mik?PT@>w`sI{3$N9LKng4lb(4Xi3C`N*HM+$R;1x`Mhj-=k};ZHKg zF3$V#+N&4A^=lTPNTdU4JM>x4CByaVl$Ti=+9KyB}% zm-1OalqgySZH?u~yViTb=}`itmBE$F~l1vC9_EM!6i?o$Dn~SCYrKsk`dP0N10BCF#Mz zRC-c`H=WR2`tzICxnY*c&bv5LNi^g~ojzMM4mE}p+&(q`p`gZd+hd7o$7S|$HkYB? zsofiM=klkmcK5l5k!>YrRUg*LQM^=*$8iPR1MOg7`%yX07SZAxv)oV1x7qsR3`@|1 zZF53L0s_1*_xP+n? zm0F+iKez^SOrj%OzF{JP6S@33f(r$wWcfEoB!3x12*Tw`Gs*GdE43tsi~hdjs`{>b z{G-J_L8L1gTAwkIoOt2VVWUfMI`ac5Bk9~zwLBJdNd+KJgO`e={f<*kn>TyS^laCr-3SrY*IF|-SqZT-!o) zhgxRz>&9y;YqcS9CqG-I2sN!yi(qT`lOdRszLAA6kA2E?nD($xVhr@>$e{`Z{#oK6 zMmr-E1Ifrlerzq|QTn#WWL7?i`_#dkhhGrDM~+CzT+Lvyt9=tY7tt-IYfk=$cdi$` z$YxyA!4v&bqB%vN$Q8<*MR5VLJa&XjmEfy#1qI66K4_u@Q==|y^Ki&&AnB>cWmk49 z#SBRs;wv*vHvv?w+sC8|?xP|=wGaLqRE$UvzCl&RYfo;*$A9?-$0qBtmTy>$@qI#j z7A@zU*YdIfcL5RAW+#1Su7ZJ>qbbjrq=AV>Z`c)ad@LZN2zwD}FHvzz0u`zwrVb1v`h7*D_Wzggk)Pm#k) z^^-VLHe3}H%NH~w{P2k1n1K+dzqSOM#YJtiZ8L)lJiUm!S~LZBLl8-$Lt0vxyoDhB zq~0;rsz#kK4#&nB(?Iq3;==gtX7pnAC3aS`yMUitCnUUSy)z_)GGr+``lKAc2by}W zm;%=5(v@j$S-e1t+(CCi9UBp09F^zEv)m85x)dhNHhRSEO$=tBZOwbZP^9h7OIYp- zrGrU^M1*K%j=&eP%0MadcjEmY@cE6xn2A{M(8~Drv-K+Eq7{{gM&GA0h(PQwzAdAK z*XLj*i^B4AA1OONGHxURwx>9B{@yXr`2W^{`I@VI3p2v_Fr&Wt>3<9}Fhv9k{N*v5 z(K;eEln5GTM+=`tL|+vkJ;b&!N0`&%^1k`=vp@cz*W~l#=arHp8Jrw%@nPoFs?%-X zT#e#WzNg&M^u4-XKgwn|kI2~BoWVtCgkw(6wB(Lf=~ZR@n=o_7w6vTtV6btp7U9{L{^B5eCgD z*)pBLSem*ye@00Y=pA(iHDLf*Hygx}-9&^c9A+LH6L(D62!g=s6h%DyT#Hw}4eeAl zZAWuHNE@)P?BDp*14Fvl_moNaQNFv0+_zAzMeeLzXX~d_9iknxJmsJ`JbNjD_O+#N zbUesi^m5~viM@nxmIET*Of>g9qTuI;Pn})v6F zM9YpGc(llsrgfpg%=7sSk9Swihs$(p54EV?6*atvh~y%22~ixYl@=zKCbY}d5vcUb z06?Hp4gWtpxanK|s-UiBYeW0_yJ92f+WlZJv`UTFJ19L!Og>cH-Ru+qXbPO*C*NHd zE;9|DKiJ>cNL!smo#Y%mL6$Nx%Q+&uPMR92>A!vIc<@YL?YIXb}BQ54*U84T|Q}_0^*N>>g{6AS8{-5a0pdt&sO)_4<3&^@ZVe#R{>UKk{ z>6XL78ya;A@ZojvZ5Xm~qcc&kaibtiJ!?HE$@=nZgcXwNQ!oV9tW$KC<&YI#8}Hxf zp;eB59hS^QShV1Ut>$&3Q}I~Ll-=jnwq6!4Nzn41T@Hr3Y1rQJjhxzD%4f#jn5RYV zyF$`0PNg6elPK#-pkSGCbb_G-B|L@gLxBiT-zcAKeW7g!1e6AZ*x6^-VHpKS^Q%;<6P6b@9<^Sn7?aKgVtEL^qJ)}fGFJvu*hRBBA9mgz@5=yz}{_Xi^p(Samo z6ga20KHX+c!zPk3b~Nv76f22IbISdh4&*kEM*-{0tS504MEts?3_laxg~Sp@ASdfP z!DY0^i}){BaaSWW0*7ynH?Tca#6SYJ`&v$x8?EO-xht!S>0Gk6+V?4&8^ zd27I*D#UDEx)TRl7960JBr3^90bbqNiiZ>vIf1p^8nhCJ-Il5k*jperj0%LNJ9jAV z+@OkJ$2rT$rc60T6k|6D;zA)kQ@+yAbtyQI`oq3eZFE)8k5&0U4K+V6mmeZaAdb{S ztnpyD^!pDF#(dj@M%4gt_9f_P=Be3?b9mjc8!U!Cbm1Y4VU=GpP?|K+DS2MFU827s zl|(sOd#Jd~>PF7{K+@uvyiCXjCrJZYM6=SH#1D^59qt^K>&&T|?xHQNlW{>sUVaqI z$OWv+)?5o^Cax!OS*`h|ir6hQo`F*n|4fD?)JQE7UK0#8`SGI%RfFwQ-g4T*qyVc0 zB14b_Oq&DtnmD{rB}|=SRjl2zuxz?kX@--Z^j~O0wmr`mG(WSj>x)B|J0o@P!9VZ3 ze&xB#g$wwI#Vm3G!puVG!1CQB#{U;WJo1eW8MFFT)%1*=G1#?cL>R=naSWe1rOeDh z1SC#ma=_fYLAM$jT{|IH^Tq}hkia@wvUBr#Sd{R?8`0v(iH&dy0i)hoTE8tj?WnoQ z^cut@jA+?CF=@D+)A6d^X(c(p=9A< zU)=ih$u9@;cZc$lnqp2M8Sp{IKrGgEHUvwm!sS9w~3SURf!cPb?TKh=I*{rn#dTN$rX(*!VNdmEp8b3cYDBsbii`V>b>hJK^^1q!5i zXa*QP@I35n-DuIxaZg{FTQ`?zJ)8rkzZ1j(t|?4I+%ota%^}!Km&ZnvHcSkV<&3po zrs4V&EzZ-rP8#MQA9dHE+!M zCEF24qO`4kasB}1f;ejOU9|Svd2e2D1LA~$4l(^Qir!g@dm-)2fnY~JBzD@Ovc?N` zfFj#PP1t(rs+&NWkop(#|7lBQ$bVg~r{YKg_z1RHa!s z>{vZpR(L)7MfW^&T|?9thq|=@O(*@4EX}H>&q1b@4}pl#>kluHOnhkB1woRVJ3*H# zYJ2M+RkrlY9(I1jWazN*cwFO0;i3j`13JIjrdG~<6oKLbnH~U>l3(KyLAZpo3lfA& zjzV~cbP`^;Ak+}lrhWT6H-4j|8S07?i=PzztfPl#+o}+M9NIsA=$+wx%4b=eYNXdpIx`H7sUd8bS^3&IkwrtbO$$Kpe4fQ>`R%xrr;u%8X2{y1)2+LG3*!^d zp{Ao?++GK!CwtD;6?c2%<>Vpsn!*!b+^}}j$mHxWt~%h*=}X;h*onP>u4mE5of<8m z26s0;XFw!U)`Qrb4tw5p@NRMGBsjT2aGEIhK?sA%PNyxK$M66In>iDN0w#Dy)dgE# z=CV15cugycP`bS_7Yww`zlEYU8dfL0opjX+vdcR0la5{B^CG-!h#r7T8y5dFP057CzINSG@%mo?=`k?6shRjpq}Z57ST#B-DGqIH2`orlVsK z#92;H_scB(F6d4F@hW{nJL_A>(=~%mUy2)&hnZb1y7dDAN|2hARoNC`_1y1w=3IH} z_N6?d*PO?-m`XNfo#KVPv7}rR+v_n*;sVEG?(VjmBVf<g8#gUwp_n0ctv zq;W>Gm#=%*sIs0@cp`s~2sr;FYwfw~8B9d8#dx?zHzJ{~zR+!2t%G_e9_9FW+W>Wh zzAigZ(tyuGv8|45M(N1pW;6_Nc`_5a6WBfn=g7nUXo6PwdJQdPrnA8$K{OcrMB048^5PFAy*Zb!tW|Q8*Q- z8-V?!a+XZ$jrmIOT&7*hhC1mEVf2$l+nBgxEu#y=1T==_Z)NQ+`N&)_SuRZFGI?O6 zs2ie6!l>`3ajBy6DO)iIz_SUucmB?YmVFZzkeud_hqh5IxJe^+(N6Uv-%3nBd6$}3)tF#^ADFz*I9u7+{>fm zQLvb=#-Qzzv(i_c`F=Hw|!X;{75`L6X7Mnap1=TI-e)DA0$ zHHNMRr<`G%5W&MuEna~`E>puD{WAMp(xu6h%tD5Fj7P!@ya6D%>Huzs?rCRSWsOuKL&ghw16-c#(0IO1ux zYV@(~Rz5^jg$n(muj8hkYQeklaEE#eN63(twp*7c1T*QebSn3K2R$dNPJoA;YDhr9 zj@|q52dW|@hpo(YCre>P3V zhqy+oI3ALiOqk1Mq814h`}z(vfTPitm~*y;Jn%;%V3Mo{Y=zH>3K-V&j_#@L{4sAxPDJL?9f$% z$1Gm*C-CrmuU_a-@Xt^rUS#oBPu-#_FcD~g)v-jCD#_RPgAJB@R z0M1{`T)GrR`H-AHCiec_s=kJosgo#7<#9{A2V#2MQ-1dF8e__X2HuowYAWMbXL+^XG>Uo_34hJyr zHHW7Zy222BE?v~fNifoVT~ch+`JQe zu6ryA9*6m;c~ujn9&4>|GLkP*_kwV$uE7PqQ5REXfq8*Difo1YMu+N*jV-g5U$KH+ zA10>|KRm&c0hFlcT5Ey0i*>H8C=15JrR;1X(=bFRT4RASx_}6tC}jBxYr`19+!bEp(;*QY#i6Ej>&m@hyjP$1dwsn8 zENj_==Gj(~!k`m1aI~7yeY^)uji;jV&pHy|U2NJQDzcv<-%*vBZIjDk#uSM$t@Jyw zdCxDIBRgRKq5eXDt!a2x%NLJ(lbOi(WVGBqyBi9^U^$`clZ0N;i}KCoJ^Nv;fHqGA zSyp(S%ALsLv9zKmiA_+k7e?2@#bZ@gq@hr`97p>Zb zM)L$MY@@ci--*{&iK5oK;k3%*JN+rrCJ)EFmCu|g_$q&{rFruL1TQXAuu&&H;~nOJ zXY5rnZcN!~=6J3c+bV;1DN1reBt&9Gotw}_QSno13n|;@(w2*Zg?~0!>3(u${>@zC zCf9ere;)1&>_5hbE^B3Yy9?pqx(#6*k%HQKuvjO#jgs=Q@TU^JETw#quf}lUP?14x zSfV*51#X?%+W@B_W7P$GU*+2MKVmdOdzf^$9JEu&$`iTFior)HOK4(H~SV#4CEAhyEU9i_6eETQ$h^Htq@ zoiVVGvN8^`KE78Ru#{^1a+t^-pKlbq@4K3(X2lN5KU;eL1Rlok8;g$#5EIrj#2oB| zmEx&R5F+e^t;uT(Y7fdK9wDSq6gF9p@ZizRl=i3OOp7bh$x};Enh>TN(=7aTbn3s! z^Sbs>aNc+D)bAgUjh-NYr_eqLakR;o03I)Xx;v;KK*W!E9qxFrp)mg=U>bMm_I*UJa3+xt}vlLP-nZ$d@s8j zt9{BdER6z-!|SH~mWk;jx_I5B9?bvd)vtHL7J0(KU|AOcfJkYxqX@I`A!Jj$2off) z_DABl(nCi=RgWT$ptiM-<`R(|bLnzSlZBi<~3U<_agF2{DVgINzflL-A}Z z`wHkj`On3o-MewReKzALL&`BWHzR8bbkeCkvLYxkD2!K8|9N<-{TX1ONL)S4ZrVek zkpf61rg)eQ1I7rFhj@=H?F%N?BqwlByBHP0!_Ivra2&mq4D2t1bKJ3c%ck`^4>HREbe37RcDZkz4$huhkxUl(~S zRgMDO%%lL_5k?$Ax8M*@jFIWa-Pv|*Dq*?EY{l-OaQ=959$TWCF&_VPG4l4vFD-(x z_EJB!XoQ|UWM_3_FdpM5<>HIaW3v7ka^Ewo-VZ%fy{+l{9b7WD;x0k!% zSFqbX;bfWK*2#83$#b+VBfbK8ua{lQ8&py(E(A)3Zd;<_)<9+lK zLJ+%UG&o04$trxcgiikBYnXU~3mmG?DvcOKQ1oQ~@2A|7mjg-N8)W4G6fC&e*je%f zeCLCFyOAi`*ZWBh*7UxS1L}K z+k6%@6)qmXqVx7rUnu&+UjX!Nj2T+#ODI|=7`}MUDZKOn+7?`2dQ!7N%}48#rOJQf(B1>r?$Fo=F^G7#8r z$Z-bmf>1Dpkewb{hL!L%81fyHWyBeH4!lD zaZVFLv3Jg_F#YXXPR?(*02=eXcjnN9v`ch1;ZqjuK;XRzym&og_mYcX=$~;D#;4LL z(#|UrCu;amz6vCJbLw?MO1|~+d_W?f8biKzJ`MBNJqdaClTpm8BULJYTq4nwEh0oD zck%vu>yZY5H+}xgGOH8+vSqHAs>oOZpA@BlR;Wg%6kZjnBQZTq5A5@Bl=U8%lPeA* z#Jxic-lw)cOjr4Xr%$^7`^{P9$C1&!f!k#4fbv=#x`lL|_GsgSEX0@n&CkaP>)-(> z<9_aO6-8XfC<;Y9LpB{i_sc4gTpm5uWwlTup!!KjXxrGweuZPRmu4Kd$&1Y#Du&ha zV^%+Rj~jz1Ddu8Mk2lW5u}TSZsq!$qo(@t43S?cwYnQJC2W;dOSDb~1Q1ieop0||=N5o=QSqomP;4pc%cz1LWrm%l)<7|cA<~0L<&UM9m`eVrT7qt0EoC(vj6L z$e2BZH+v^5Tui%dUn;we_{YL94gOE`-y8nF9(y;XQJJfAd`peEjj; zrGb1TMZ^c4FKsm4OyuN7&(d(foFBol57{CmWtMZ}&N<5`d)+pt?tAW>; zJl9v`?U)Bqq@}P%iKuj2>WfP)nD`uv7rboOuyM)k_NhK=)2K;fgEUy%4iQQmO!V-I zu>v>WM0`2L92n5s5*mPBq)aUILQ0<~gd&Yd^1LK>Z_LqN!J>-d-(Zf!rrGZe%)$9X z$pkarZX_k7$+q_{xGRBBMYMLH_5R#qe;b438)bF`&Ejfyo{%#Y!i3V2E`&AV)k=0& z@+(+9YA>i@2{T(-x{R5P;;3q($eqfqyAbu45Tk_WrK^92nAIuq;B88pzZ4q3c7`ej z^Rs>{G)hw8`M-S=9_%YXC^X)jYatcKKVJhvJxAJL!Hh@=udQ9mZDA{`-%=`3s5`7pS+YJkQ|5u_R~61ya^Pyhq)&FU^lF5+pv)|;(Nzl zKgrW8mc8OpXaAz&ajO}psfXRzgKOY*vJ=M*6G6z&TUyPHu5H!c&>zsjd4F@|M|8~p zfDW!om%zsGhHx+cGs-hN{HrgVku&Zc7_Cp590xPR3j1;BAO*gq9^b8l)%V(CIkXCh zk{Q~1xLkI~!P;W#BwNcMu6oaC4xT8l*+`FOn-_kBuf%plsxc=Bd73pRgh|({E1KVf z6@7Mp*>lcGGl~i7M=WD(p1@qFcI~X53NhN}qS^AiaY^acyi)C553A&|xRGlK*D4=c zr8|+138b1Om0D-IqzMlkGJh^BkXnT)&CafyIbo9xga|L(&5~hX%I@41w#~akDY9H} zmuE%$XAABO;11}0|MG+htza_t367t>eE7BqEBwja$m6fONvpX#!BmIh56-fPQ?JR$ z^|t39828eV)q&5m;p+Btw#WY0JF#Yw?aG9v-X;&+GnuG{s;L$@W|riz(Ykr3XfM@1 z34Hy^qm$HUV@|KHTP2GeOrG7;@6S>Yq0wq`vl}SVD>u8xCUZSJ-@HEyGyLHC_}|rO z8|F&$v%RA?18;dN*gR|8JJ<5+qG7k%o6vL?*yV}qgs~smaL+GAW=pf7k?RY{omho| zi&xvqyMkGuIKQ(Uh1a&GDAqqO4#@AwsIXs^NWp){{kz*_XBmr zqf8|KCho-4!gzXv533TAO#Tie36eT-ch&q^n#9eOiCwxPcs79AgP45G*2y`I|NT!U z~!^LTBt5Je83VtlNYW`mOwf=p(7 zvfmS3^xdJbxC8&p^`|1Z5an5OEe;9$x;g2NwF1Ecm9{7-k^yTxYm%{OzI&xV>lt&( z`!~eQuNHhcwm)>XIO4b*XyNOKE(%^E;<@)Lb3xIX;RIeDDyjNcyzCiXX3%yPtGc_L zt+h|#rELu+vP}%6!rXWvPCLcO9!##&aSlu+U471j>9>6&T#p__5;D@;WYR}jBDbYc zZ;7ZTEG=& z+3jLQvn_8}2NQ{Xa&e_1ssubR}T-|=yp-#;zEubr1o0efDdYX|I zN2J|*qfnD@<21z9Ik)ewjLphKuilJSpaN~oJN3GyUZmn}kt_C6w*BeqsT!GSbfv@P z`nC4*JN;aP=D2ojP2&THk#aSPFzNpKMWY@L$uzqhyRGJ&t zXC6mFtI{v;-YDy!06~)WPngc4Zqd$JX4>CGOcJ0q&2sGDg#0Bo6JNF^#vsZ=ZmdBF z86)PP1$8xxP--)w+_TgUe?n`s8%}_hr4r#xP1MnD@tDfsc}Z_>g3=fA~@H8 zMI<$`Hvn#k^@1>}$hzE#xHzR}6?>lV#B97Gdq;Z$@6_2)A~KXQf8-VQFrENCK*GOd zaIWwX&tsIIl4D!~B8t*n@z4r*89Fwb=@*k?^HZZX;o{O%ghhTQ#(37;8KiBe30B!& zxOj(DZUo}gj0l1t_Yo^llHHCRlt9E@5Tpou2_0NPa*rJiY0N@*a^A-xV($T2>p|7k zTN<&nqL_R`6Hw5}ca`_^qRQZ9VRzBUi%n?z``wfw?`lRtMy2Uwt|I7+`H@}2V(fj$ z+eTnvS18l;$8Ll3018qy2*)c`9@)J_jCb2PHbWsYw z%`EZ5&E&pp5#)Pm3jKNQIGN~`(|bC?dhxgq$})TUd^voERhJ%60@|bZ4jWD#W2F)^ z48Hu7Q2xDEUFi59b=enMo}4>MV6jSwL$yV8JA>#Mu>mgNab&;Lp%m7E+1@)xSh}hB zuE(||a7K6EiiQ2_wHg1QE^k3Ea-jy=i#}7zJ7alJY2H1pJ?|DCr?UtI?C*A%qoKGn z7(O~*2ZCVOX=%@f}tC zH`sPML-!un4HM<7;TN}HRBhT0~_gOkXZ%`|&$tS-_hv!dX7`Sz1#)k33H=J&qVaI-C6{OKXIJ`ns-{%3)D z_YZC$)iOgsA3YDkj4LSp{D~UlZsI_ae5$86OrIxE+}OB$$jxTZUJwkbXMbSalqOv4 z?kFqyzBc98Ugl39NJ8gB-R;boqM$c+9=Hb9tVOTQ`4`8S&VOCd=WRe6_x1Mt%!S`| z8e=#IP{GuP0aPfJG~pnjn~=pN$R+d)VO|5R!QtaY4CWT_x)VXh3eSs#V5LX`cC-d~ zg4T!t=xsXCp?_$6zdV1552<8HF{$xUoft?4$E)W+tee-vAfEf{J+um}ZiDBn&60~K zV2!{c?V0rxu~1~l;xy#ox{~7uW9^(byU7c=J|EZ?vjKAcEFasl-lB9DnmGZT+gpZ0 zS#cVBg88{yEJcQ9kLL>AD`=>g7Ru{*42w~WMy zGG%!qXEl&?yQX_WcijwJMaDY8aQ&vJNwvfEsd7ZSXu<h#qDO8{;skSC6<*Vf_|12r#&XE%d=KLHc{`#H z77}gFRE041CKA4jO6;L4`0t`O+m!acGT(yLg``&u^@4~Q# zN${a9GstBrC)EH&T#qkDxM$s>AcwzBv3T`?iWDVo`hbv`_0w6K{^9nIj!{6#4g7Zz z4f zKR#S@EItQa-7dT6?0ttafs~3}4EI{%#(Bb-i4ui@9Fg7lZT1RvDHiID@=-wlJ zy`!L`yxJ$GB6x3|;2b@2UP;9^3x0Jurd<^!0Fd>gok@gk=N`u(^fvr>e}&4 z8n5Tn-6ZA`;%WN4W7Fwd(cKV%M*NEo!Ro|9k89Y1a$t?{CE7D^6YV_z*7Z&biHVIR zTG@>ybe^DelW;M3H-}e`=F;L~AMx_00yl%ks{9)F(XYdEB zS=&NPE{OJy_%~-PpsWEofsSat2Bzjw)D978L+o3o3(bL_GMK68Y#1<22ao#8x8$Oi z)gh?0#kZ?Iz3op07#^L*9fwiIeALv!5Je1JP-82h3py?A#T`sTBi#*`IwFS%p(PES zQ7mC@%vI{~|94I4lp`coAzYS1ui_w)`c~dYZluh@t`T@$g23xWhd%Y7KoYxCFv1Zm z2y7y7o{$U_!984Fqx}gE0R^ntWg(T=Q9U7`p;56o#S^Qoo$lLYXq5qbVmyD>2pd-$#Z+8KKt@VmtWb+4-Fp4nYNuJG9ge0u4R4-gJ`rAfI zvQ+!W|9x~5L_MPSolH_hHMzx!m_OLBnZNCheuyQg0E6^IQE0Pg^v4!+KJomgr`g(* zg0@SJ+O`*TFuash-{vP+7H`PdL&7FM;W?l&bz+wK=2lA31J}hVt;77!f~XUZu})2t zZjugXVreeO|2Eka3<&Aik7gof2ioTX9-He`caIZe4cz)qWxC6n29ml;(pls7l~O~f z&hVMqG0VEm1hMf*zm_j>SECQ-`Cs=;=l`K$P-rE&a!RuhYEvID!%QbypcG2LBKgB7kMx^ABt$=j7&$;0 zF#e7Ta`bc+lANC5J1U$^QmQQ<;Zb2$VXK1|?;IyYh}AcF*`i3-FuJ={1)9mY505i; zYXsEDR9vL0dynykf@rAZj0S6p|BD55=Jde-;qAKvsowwp-!|vq*qdWzB#ugG;E+8^ zA|xC#Q$|@4=h!QI%g)}FILcOpLNpM^AzNm49KUmPSKZJ3eD3{zzQ6B3=l}P4J;&qm z9FHfwtnD#C3k=%36I5v@;%;?U`4iP?JBD$eTwz7k-tNRaJiB5R&`iC~fCC#k2DC)b zPzdFEt|~Pp{Y@>HYYCR;U)y-sZfKpe?EY-L&nzTbKMym6T~j_-4cU!`k{s-vPSIU} z$q1PRBAQ(GLm`viu1qc<2N(3N}*;*Phsi_+j9    Gq`g?+Ji1A9(vsExsA%0tWz#hHcaoL{-T=Ta58YA~VFyg1EaCF%3tB%Jm~SCj?d zfYI-NI-%m~J1X!?-Hmv#b%lqSoF+xeh@ZfiXbFwBHF!V}@8uF#&q{y4IBm!B^f?#m z^RY*Ca`QUBUJpp8w%X(xK5k2rtIqcdAvNf2fSLL%h>j4mXxqNdQjaGlNxCE2ou}(9 zk->4S#31c*q1mWr|5$UFa`A0$@Y);=-|Y+?t4S? z)4dnZ3cS%iQ*D=Apmj%H?&>MU6J@p+(z&14+zLJLt!sP}Ti&8WW>7gDc@=FxAsh)+ z!_Z5&ylz@j`S_-Os>%D~6GNxDW_1tpPX>AiBG_hBz>)5@e_i`Pj*-x8Q(Qs;Tmaxn zEKRt@EJca%r!_#J0<=enEfjl*oIi-`$Wmf2vGTZHB;h%Mun46T_iU!8kp}p~Nk;`N z0w9vB{P!t$g8$*7FZo^T{3yNxiqFPY6{Nr|*2h*t;f}v$NKcma5XZcotq9ITH6;%$ zJ`MuB^Av#l%IOhbMsC5k>E$=`nFfZH}UR1*UgJ1^fFuO6hKxuo)=nPK&VH zz6(ExUABzT8M@pW+1=UHmU821L-1(I*aHO1+0?3TsLKM_`e)=X8g@;c&ZEk zdOtjM4I};l@5>K68K-{BRdJc>h6K(t=SO}vUQ_&Tr9V-jwxiByiHmki{)H^nDb1Se z0aeu1Nv02u-z2GV7QgzkMEY;6PY#XKlPB8h(b)k&{OjXL9X?H0;@Om1%xL}T)pZRC z5@Fo8EeGkv4;6;Pn8>ppnwT1&1)N88t#eAlzrDa0_ZQXJn9iYt@O{kK-aa59_{nFY z4$%&FLOXXrkb5fuicjBM;q(-$*UbUb;;;e1z&3z_0ssXDESQN^3Sj3DobX`ITp5L??8yzdP1e6r z2A3$d9ww3WcK`AT&5k=6ep05RfK-Or3N#|4J<*=Se~~Bd^~&g$uL`VZujJjryV4L> zMBrJHMNv_vgF_Vj1^cfno945AsP!o)6hS=BqyaSdJS`QslwV;hwP>vT@?c$g)> z?AuhDoocAPa^_qYSfY8S4SEN+(gye4TLn8wFCKa~-7Jm5wyp>uPi+OxvTolB;e;vg zDrLB1PRC?6?98=OIXA-X@=1=|=z!`G<&VLf$mf+ZT(5$fp($yCC1c_!(7b(qh0rs* z0!UiIXT``y+v~9{yGCD<*J--PDp!6q94HV2-2Qu2j(@2Cz#+{4Gt{BLJb$`@sq{8B zcNF!E{$69EGWp69PvQ)b0=lnEH-N||PShDBPo?-fe(uWu#A-v*LcPrrC5mpDNb%e_ z>id_HMQ7z&;cluRVj2K3u{%*IYWMa+k?DV)%W16 z2b<5%ddRC~l_tkaXAb?l`fqo^6Fk8C{Fw?XqD8#tJBU!(4%8s294{Dp)t^dYZ?!E1 z2@u{>-y?&Hvk}`GqOjd(sVH<|tM$Gx+LaDp zByYGn|Lrcm87?D^T(o7(dE#tt*N<|PsAO8&JVYRCfhwD_ZP!e`-%Zs-Ar)&{R$IFt zXEXnMW~;uVm9L#FdnZc2SP%b@>~||jHLLB~61KcJ{%+gSq^NnX;n)0rfvml@(yZ8L z9~Z>QHi#(P1ej=pSIl)XLlj!MC@GDmx&UcrrZZBHH4rT*G6tWRJGQ+%PMUkkOeb!{ z-=>Hg`hs?U^O4iwq7d3UyX(V&8E`q115 zBI&{x3@IkZGh2xbi?GKzcqXH_(f*&A40}?9OnA{`yvmtk{`jL^f>SmptxxkOiV?&W zoY2@-d|E)*C?e0fe?m5)Ee>yBg`^Tn-WkM0tAqCQ`{RcaGQ=6K3<(7e+oKbz*-i|* z=M}y9IkK`~d#7zuZEzN|nb*QUPCsz!F)o6;54YMEu9}K6luYTgmpIWif{yYgRI3V= zdi=5QK??Bxn^Re2gq%3h^h?$3GdzMtj^O83=VKUk@P58FuHw(Q9CQ=!=Q=5ji}{3= zB@K8Oig-igu$mb`?w$IRdGPGz|7ISWeA$nGAiCcYMMul7Q4y0=c`T1s`0U^syy#QL z+XC`;#VC&rS~P`a4Nrlob**0kXyvHk?bYcLMIeZFKea>-D7>hae(k^NM`Zm=KbLRy zreH0!HI;+`xXsn}_b9L_DX+)2PDdixrXKq%38h{SfiwG&j>2g0Ocbmz0w0IS1G>4; zgPe6PL)yj_i$DR6z<$3Dmu>$L-Y6pr#S`|kBmQ^@M;^$ar`{LDuBZIaMS6lQH;QCv zwF9Oh5tfd`@mX1;(6>6LlG~5Gy7nVvKMTYHhQCYaLw~$$5NOwjeR1g0=(qepG|%b7 z*xZT~xtuFSXiAipQLEzrTx0 zn9y@Zq9`Ze$i3C>3u5`TF9VC}-sVujMpgo#1Q(2y9`?5kpN%{o?%(pXWgILVI@<4Q^U&=avo89*y;IY zjuSAoE9oh`>;aKv!Q_2RceHylnDG&}Ih06%TJ?c^q}TW_%dEbk`ZzAFiXa!(TnV|)Rw zO}>z$mZ5C6$3L(!S@jJ4T@*KG9(Dc$X{xojeBO{hqmEUwGqKCg=M*ZcLYt7hywfqJ z_BZSb?PA}*A``|Vfz~YK9^EVy*prNbviGwIumb3De}b}q2t+}5o&<2GJM%F2cw9>% z7}@CD2Fc^NQB(jOenrKK%v`a}&ric+xqM;&B1@`k^%i*gH%KQ^9EjsP2Kd^&Dh|Kj zxEc(W83e?NiLK{53ToQ7BZ4UE>aM3V_b1H6Oy7+dRJz$tX_}hn<87rg#q-WFY^FeL zCeeJ2ay0(OVmLwj>n};e|4;3&2WzZD5jHdha{o{o=ikUxo|-|XT96olP24-5lDcS( z5@)+Rbu|VYY(}QsSNzn56l2x48E8T_WojXLe9owk5w<2_6 zvJL&G?2>(sP%xrIMTMm@ zPeda_MRJXmIi`Ez$QMO<;)jCQrx<51WdAbOe`pDy zHJ(FD;NEpuz`t4m=J}~j46Mq+>32Gy@r^(SJRwqWEyJ0H?~xdPP>olpn@{s}UEKEtUv&?P$ zxFA{WnC42>#zIrBR+DGL^UX!(2J-`o=(i=F3zG`I7YUhW;N93?v4DRiyOpChR_-w# z%oUzw5yBfh$YF09w#{QNLYtQpR_i088%CBZgS|@{&MM%ZSRM+EB%YJax;$!aOBZvN zcJiV}<6z4N#i&lhRPpmABKbZC+V2t}l(_K-MM+w_fN)c)RP$sL-0cu?1w6?}LvGJv zl*#tTJ0S9XI{^kdEIZK|4!FMjcx_rMgA5-WB076F*7IVT4-OABEY{T0(i||W0i2HL z2~{!M#llgjW4d!I7=|Vi^%?Usn1DH@z~{l%3FkcRd^m6u;YG0e@zxRmGZ9?+upK0f zYO0-S5n*26Ve8o(|fSF4lmiL|K&n%tro7zx$_ZqgvD^Fv57) zS7H54bqh@VK!b4U3u^y?uqmdJanc9G{ftgT5~=*pHZ;;mNl_A`fs9kbueBZ{BpY<& z0b*ixNo?UJ?#PAunC0=51AZ;X+3iQ2VX^x<^4rVFldqB#a7VY<*M<7?#X=-f)9&1O zUnQlGoQKoSCf|%2>Ns7~@DvSUB8|w8}oeCDqJJLMA_`>r27) z$ee`sYG|@L(b5*c$nCY==PpF(t6boF^-uJ>f38PZ8#Vgav)g2K&JH?M1?^yLGySCL z!dc>mI3n#V+xgAyh*8F2E)0UHw!rrBKFEvwx`NCDJC;-kA*W@)T19oOA8HKq=&7%| zwZZQd7>Sr8>DEK?$@!1OWD_61(E+X58;{65s(E*}8&ZvRKtiJ6B8q&hf0E9l?$XU8;R{p*+A&|<(|!GonJAY#X)b%TaWu+8v~MGjdY88`xP z;H`1(kE&5plvsiW95?M=V$6h+KLw3yg6E^_E%#|;{JA`x^E%=LRnpc|=mcLz(lJ+@ zOYP5rnG-JpRWEPzsVAKE2EEztHE)3ui-WAg1}IPpphO8W?8SL_(dFdf-Ik4BNSU#Q zv^EHv91XSY1y#$eVBj_9DRyPMnddsfj`VshQuiU%Ez zSNGSnVotONF&&`qw(JCQwpo0_Wo%Es01sIEvOne8#8$p3kP}!cLgzjhgz0gbKG$Dv zzT$YTPIRCE`=u;I5ej*8Z+>Ei%J`(RY`iu6=VoWd-zyTTMIHRw>|{A8zILcnOJ;zx zq{4H+Xfb=tNN2p|g+#dkkZIAH8DVnbpE#usF^QyyXleT~f2=lNtn<$=KOY%7>3$+pNM+ySN_y!N? zF44Mhlh6YqA?J~7$@pf4@w*P{s<`#Q)8+ni6`ptJs)+88*A^)>89#5?%w{I-KOlEr zc)s&wxc6W43BCXv*Kwlk^XYW!-3$e((k&?F{jWV%@%bHC__Y`71KN!8>+u)Z8J^|@ za?`pF;X{3jZrGtysvG=QPWi{l3E`X+OZAg;OrsHkAJ_`-3;2k{q%k@U7Pvacp;>jE z!(YVV;U$2AxbB{h&mpA`DfE4NoMQma*l>E>g=CaB*X$(2)jlYhaTjkvD98#AIrl8O zp^S1@&0TXr_Ui863Due5`m>OF@ZAk$+O@swEdy6fl0?4|G*r&MFm6KRNwNLYH#B3NR?vmLzF6)}A$ zWt#1l!JkvYUtM(hKTHW`NTL`q)(TxXureVh+b_`@M*NL|+I@wEkwlf2&4PE(>9rsV z1^B;C_V=SmPJQRbHGom*%>O1X;>MYqZlh4bdP`U*bBlJ}>)#0*dDTV@kMhMJ;`rV12<9n8Q0&o!gry zZy$eUg{2?cxuqWjCkOh>QS8Unxcr&R-640V&8mlCbgF~d{vVA~`h-X=QgVW+horee zsq@t)!bE~MEC{W3eq*slyNQVZ+WLN&>$bfGnjXBCAxEp06Xh{s{MpT`ebDj_`q4Bv zi;T!7Lec*!<*0%>{dxf0d&zS)@2u*XzC3Bl^$vLCRko2F;vvGi@wQ7zWE%0x987-C zkM((godGY4Ic}8Ed*0dG{PNAWpMrBPpqrQKf`Kyy&?6& zzC{`Oe}uW}c$ljLp0@DbyCEZtc$6emKy5ZnF<|edkwrzMDMzjIDj&Q=)C66}9Yn^U}Ld zcpBM(HUF2fmNi7p6sXY#$WnpvJgnWa(e6pNy^jmZ^ zbn6zqkG$mmH-(q0w&@qwsjSBEp}@HVYPkQH`yw%JEhNxmgb)hQWMK)G^A6Se|#Wae2h>59{pKG&*69ReiV}RfunX-X5i1UT27)T;{c6$s?*s zl3=fOwG;n|kv=9RM;$dN!jU%$dK#3#5r{S~Zni6521~4;t3C~pzA0H`LPkt1?Xfge z(t;xEKU*_y8fcuq`(EU0CH{e^g#W7}JU>+_?<628U{T)gVs7ZG2bSyGrFkt2YNk<7 zMl;tcv{tk`@2HKv%ZGjCxD^Ry^Ta{`cXEvP(a{j4ITdR%Ghfn2wF%-1g4-PsB%7?h z53sk|jdw6ziy+oVQ|}P+`e-bKmf@+XDzi`uB6OSx!509!K#9;ZNwv%6lzR5ID?({p zy$|YGkNr&LI2-!Tmw11zoeIWl92}#y-!DLruwjY?WT~dWTA5RzNbzdEF{4_vtVQCh zOZJ1}%DF`esUb8%Q)#;@wLi!h=>mL!GEn&QJ1VnL!`p(v|E;54sW1FT3)=tD(R!_( zeKl-B0>1m7wCSGozR5mKLc_f)JpE=1UdS>#WaIE5?-XP8ITM?OzTyL8oC@d6t-i+x zlzVL^Gjqr~^H1~=-)5qrK(;N*(FR`Y%<3S>SmVGLd=UtQ-7{a(vR(G~_1)VUC%M-% zFm$kuIA~+h?MB{|LoS&WZY}q|@d)w$_;P*t=8JuNrX^^wf*oCh2ks*A^qmTZ9#`_4 z*6|p}Q#GuTuC}qKQYg&XGu|#qan>@BBtz}8!`9wJFsBW=ubsPqN_oifIEO9Wt@>iS zpQb+u1i$=0Cf}n36~vA((ptO22MK~I94eN*O5oxgzN7JEy4#;?x!7vEyP!9VO<_2K zs=DY>9E#zHj#qW@<>W1LRc6_*#qS@sz(HJhbRBw25gFmshYw;~NW2nKb8Eo-CXt<+ zR_%l#U`5)_{mR{Gw27_b{#7x;+ZCB~^hd0?5QP%d zF8j(I8#rZ$<}Kb`Ai$*H`q{mawCXRPz+!B4Z%CybN3PKXlS}T_x&)M|G|3;9|D$HE z-ULREUI&`194c{VpL;Al5+*+2yvTF!b5FB>Z0`k0rxu~1K-2r*32^`QR6|33Nt+Y<$Z%@^r^rJH&~H=4@tMuA~w(otySz5J4xV1o;j zM=zNBPfZ?G%kVobcHiRl<^wLJYt->A!z;8%Z4nvNtskKQT&>8ex?tgu}?QM<8RPV8q_nutSG}pM( zrVyY#{aR>jqB1%C=J->1IzZ>1%3sWK<&Tz+04f1?fE#c5%Redx(~URyK{z!9x4W^; z0o02$>$R6GR05e5_+Jc4PQ38rCu?;6nw$S1f$}scfvWjs1S9NqJrNMktKlzWf=SpZ_Q4T}GPO1@)`7n2>4g%5_#1S^Gu# zm@(}ykHO*{h;)TKm(3SG5Pz&Nee=Q9{!=;)#_`sf+WAk-qLs8uX`8J&Rx#iaQq*0? z=^h^vO3FV%Zvnjb|0x{*$@`_#iw?uFD2`L(P*v`1+%c-y>s`r`1{JQwD~7Kh$chiA z@ATRzB1x8s=#EMizI_PV-@jZ^jZL{k>U8O}^9vL`NG0}0!yF@!?Pm@l&hU&rWM#*O zx;BsK1hk~H5H6*SlF&S^8U1#m>_s+B&#<*~M5RTZl{67neMJRv1U z>D!#QGTF->OT$DHOY}o|Z%k0&ds-DBBiH{et$Hq1Gy{S%X5Od2PB-VOA!$v#*3s`w zO|ocXH8#sXWr!WPA_D8@meq-fE>~vq1rgI?VPW@qgtC>J{fWT(G2P8!0;9x3a!q}P z<`+zNZ2J^QPS|1ngoj7~-3HMY1L|~{?Tl9JH;k~F+`iAXqx7Z7K@4qrUIRG;?o$FU_LR$nEIaT-DC#B&bG*#&{ z=aNK;pn|TWrPi)jEfN{`bLPRN&sOIXl{$9`0B%(*QDHCk3q%3^R+IBqb?(PRJdO~5 zlC~>ok@Vq2cNsiU)p!xDpicR&=Bh;Z7bq!rL|JVY1KD)(Bic#F_%FBhu_;^@vquUU zBgw1{1z3!-p^bav5!~QW6Tyq=!{4;7jDvPqE}BQlUTzognO(M|F!&;QAMn4IlXN7& z;Y7+lgZ^q-Q`G&>-LN0$QwXM8@asVc=fi^#R6p&}-gdx# z2r?&%^HIk#e6-D zU@u7`1kB~Wz{)1OJs0-}ZZ7j%z0S5E)|B$9whnVU|7S#a_MGgMUrl#pVrJrTos^Yj zGyaRIV?Griu!?R%rup5tr-4VhsfMpR9qex{Wp)gV8!qfWP4?L>8w$dFlNnp}{uoI= zb`eYsKfpP>^%(Ru%&Df6Y}qcvDqVD$GB6>{J;fx+bR{&?5U6&u2vjsM;tMGw(#r{D zyG1#nok7H?9@JvwmU}hn2oWV6>eaTTUB;9r-gJYl5Ybhl6#w^$>rn8UVx}$KG3ZC4^(jDEDOC!p#zljYuDK zGAcq`v&!-AFa$j6(#>5nVp)=OS5YeEe97&*6`4u&$LIoW9=}Y60wbSne0wn8I$Bq8 zZ5S`X1x_1Zp{8ONUkTe+YcCxeM;@miD?B#e6XbL1crTnnuCEuM(^>zi+46%)uQ+8l z_>qwbas8xWPZZ;_z5KB$n)Ku^;Fyou_gWp@AvnHb)u~4S$66SH?PZ=f>3DEVCRs49 z7OiGNYud#$0ShF_l3wqqDl~}Ed2+*A{b_02@rN9GM2GR&WI!Sj1khF0IzsihLs?LM z6j})-qY;coRkgl5Af9t{64-EZO0gyI_Togx)lrW2ZQ$`#TyqSOs`N+xfbKt$oCYc} zLEm3Vj=QY4$6O@M11?zr3nv*X=A?%9U|DVt)TSJh!KJTX0J@S-x|sR6uFug_cpN}V zWUg$W#W$L$x2}>*t|RP_d$^GZv+4}jt5ZJWWFRE5YC$o;F{-ZjC?tl zAPD!GfoiP3cMN>lS9Gx&8DkFyW>h%~U>DrjE6;CycxGKW!cI2bbSeK#X@bVx$q*QF z`!*~^>Q{k47 z0gqt6MgKk|hf=Xe*?TTu>F{TVeYf9;P2Sy;gh>-2oq}`%xzhW!CQEHC*e(>7H#AP@G1-zpsJDIxs!Hj-KW z$WO>(&x%nR<=|yAYbrkKqEEFNEFeRLtE_xvBBcBp)(X;ArbBON-+b zve$1*WAOG1WWV~wehqifF|o>r_DdUP^?wwSZ^RqlnYP2kd)NfdrxPX|N*&7v(}Zk* zzWG+La6O_O?AV^jTTwYN98+e;lHgDwR1ZIz=CV(OdV&wh{+t6Jb|Ua0d98yRCg{*k zBYOO7Ol9xxS^z|04_gM8XNZ`F&gOb|K)Va>&OfQmd$9D(6=xZJcV&K|8$MnWHUS~o zk6nZ|v4t9?%S-01JfZm*wi1?qf@xKWYD&EzoRW4jB!l}5=W2Vt&1@B%tai>Jgy+c+ zuRn(k^_|!DAMA26WtQhEiZmzo3JjFm*gy8dg-uP~YRW9Gw_Q{dceAE?@8Ejf=;9jz z;k!TL4=%j1|7(?~zZk6kUdf;$CLy;wQk5Wi-Q9JL5UiQ(E3~%>v0lyvM00pGND3Tx zoDkv*0e{*I4n1MXVz{6DXn10YVbm4Y+f%zwkG=N3bTG>4P$NE=OVYt~Z$81%u-22i zQJc~*d$>vWE2u}-MqlwL5XB~`WKs3`z7LTUIRe`f_dNQf1BOfQC~F|OEZ*3!P6zbo z23TBp1#En7sQ81i*U;D@oHB{U0x@zM!YBmkTZWELPd%35-7+Np7v#LAAC-b>Ycn4& zTt=ECava>ejU}j~qPt(nU&87I&~zo+I%~;ke<7`9tAD91@AYl#QBch7)y@uOA0YGH zfIs!>j5$fyk?l*~&7%5^Gz?c{kM?P2=PA+d0m1yE8(CuyoxL&fM_M;^Km&*!g6+{gm-` zaCIi?XsTS!(|J7pK(v=$!>xo=;q6zI^Hq)B!z8_(m4GKNZPJQUds$3wM%bJwU4*m~ z^_8c*VEIOoRs$f)pUU|xOu^)t^6cHsqBxnV3{Mo<=pxybbD~}sOz9W+ZWE(m?r1h4 z!+x_HLhdW>b-cdB;tNM!*y-I93h*=F$kSiH@nyAxH$?kd8-Mc6+nIvif0WMg1lRyR zAmxw7>UTi<-B<%Q$yM)d)9La@pyG7zQI7DG#RrO)#$f4= z$M$DFLVr;W){O| zN9v%+o$k!lwN;-S=IsE)>I6b={eu_Tz6%!$^0Whb{>>vPvj9-y#zoX*EIG=gV+TM5 z>zrmXGeVTRjq5o!kmZ~|k|}}>Is$sfC@p#YHf+ZfF+nepfA zb-~x!FwdVp!nT;R9tI1m-^faW$)RXi{G_-LMHH80GuGfu7CfM5Pp!(|F9VXY%|s|) z2U)RT&cQd}0r?l7+}PJ}MFf;kcsX)wxeKQ1fbnVB>uK4iTE^#~GTxXHJv2Tuz9^}! z#E^h?cDc(8CCA&Wh&WYW%Q^(K|LOm4qiFK|71^ngmhGNk4c4Y+0}z3V4sYG ztyjKbcWdQ*mmBnd_j=u5wYWT}@EO1hi2B{AvK5i!Kc2jrC>NqQ!DFXemHRydrLLHgL{hMg@E}e zP&p*NszPFSK-`*4Kx!bXJ?)5bb@AmmZL-XFudv1MOC=z{;2?eJ$4c6>nIm2(b=@kv zfnbB~n#JZPj-P!OkJeTgpxwVlolC3IPHeLs0s6l|1swb1J#L)Uy(5r=_(rh7U0a27Ij zlcKrw!KBxMmsk6{*gyq+w09rfNz#^$CRc>ISEQY;TDSM%ZgGZ3m;HcNrN)~wMZGh!LNY+-2Mb@UxCVs1xZ1<*F zKQZTkKCWCH^`E+LZYn|*&(B0=Y=*+gtC3gKWzW{We%MqHbL4D`M-S3?dGzd!d(%xb z`0(Yl#I^f%N;CB<%l@@zZj}bCN(Y%Yea>u*slGYOb8)B0`=NrjLzaTBuTO^+%MFUI zM7g&8nfys}6vKXhJJ^YLyUm}V>#vCPEabUsf2hr$veN1=b1WrBoZ*5xLK&IhhXo5C zE1H0i?~G@72wYi@0JB*R=7dx2taf-D*ISPWqiP(-#vG%aw7e@<<7yGii9-a0ko$RZ zqcqOw$7tza*}3FIY<=1!=_-!ZM(0{o*rg|dc$6Fdz+5`9hv8DZZ zT#f4RxZ1AykDH)$Kq8>>hqA@b)J8o zk+fmb^&=b0%qYDiZsaFNug7=7Iqcdq6!)eS<*}5jHB?UQ=UTIGVmr9~^dte)iTJXZ zlZ94s`cG`5DaCZjR{+#RZ_!Eq$@FJ#Y8~H^@TPjI1I<;PSHm@jmADW#FDVqJt(R7s z?Tgdx-ny(`Hb`n8IgH_$0N!PW*gIdK!}s<+BWDL5C>->TDR}pM1nX3q;fB!bpFee^ zLicgsNXfRLsK+fIVth#$Waw;&iJP-EKzwY20aRCt=z;`|MdJL(UC=Rs95-I@hB97J zvkaxJ!?uHuVOHBBkO19h|ARd}XgJ4lYvKqPdSwD4vzJYGZ(Te{KurR@(uz-7+naa z1W;Kg7G_ePE*Q%~v3N0FCO@NA6dWOK_q%lzU-UNn#X1t!{=8!<_rIh5zZdeP@M6*# zdmNXhl)#&ZD)C2yJa6+>C2Re4Q;4fG#68Gn>gN$Qi~t-EtM1 z0_)bKC@JBeU)ap_1aYy1c62AyO7Pe%o6q#5HsFX?Dy!$-<~&cs%{ZMzi5EYqkl;4i z_%fX95->G)mB9PYKcu}-k2ZQi;|_RCt#SogBDzGv;u(bIB#M*SZB_)G+Kef58|=-+?1 z=+93|djE)4qU3dd(j-bJE%eyd~=gK}$X!Q=OeBboa;GrAj^?Gme!MddJ{UmCOAmB{o85AZ zUV)r}R0ain0IAu#93NvhN={Y^Q_8q72z%&wID+7;PW6Nx?>@?AC@J5S2TuI%`l(bf zV6K%gIwzcXtRt3e5HD2rz7^`l3Bi{eS{c8ZbV zEI*KsfB)`(y#mg^*YP+XpS=}pT-ovDFXE&zVY<`eLWW!bh^()lr!Pm{iCB|Vq_aQo zKh0chW_Q+@8RNRBYK3_NAAoS@uGY?DYJK)`&(&m{-20~ z#%qp|If;LF*6=9Y?fZ&C@18h}Njz3;G2t~hyXY#%H0-7IDf%7hFb%ZGIhR5wsV40uT|tT{L1jFk15polzB1A9QJMS z8gwylI)mbezQB7*c^nAP68mh<(F_q?f5)A$A{AbF{JW>K8}a4GNQdL&bFj6jVz4v| z-Orw{^1tc%=K?_1+g}Tta@6BN>z)0D{e?;qXM+Co)J;jl_crfRCpUCVk@y?Y64>+E$Iv&`Y|f+)}PlOV{-@JokuY z21?tcaP?2{?tNc!mTmu=;!$!gCYP$f`ft%qxcz6f2b(@|2P2GF6w{n_jL%aS+pE4{ z{vus(@agSzM_j%jpA(arXAIb}9y`vh5V%bN2+%6$xOzP$H&GGt-1Z0a5eC*NhBwVs zggv)|!MyEubOA-9Y?mTVx3URDlI-xNM=B|;w{l)w+irti{_x5wo`!zM^1d{$XiQwA z#T;EAqvBv{u%&#UP1;Rb>ugU@d1`?V{xfGX9&)?!X0pB)<#PL3wl)RO{Y%+e`d2Ev z#eS^fs=vG|l_Th|21$b?(iM$4@(V71BBYZn{e7D1V*D0B`gH3_(c2(63ec=xFB&K$ zCLtmrEj#+IRK-uD97J7en0c+vmm54_u83%9jWZQ{QfdMm40frA*8Jxi30wo#*9Z&{ zL(Mxl6>xyt2kE3e>Fgzr2lPs=S}FEdU3wsIzihrZ47#42xv%H3gxs?HN-KJ{x@nXW zIr{wRy`2?IEAuZA;&E2MdO3Hc-_nUE0lUj#cMg{+pc0jha<=?Sgt={nnXTO;9c)(q zO-;x}8Yb6nHAQwo%v}xTeG7yI|z$obZ>Mcx5-*A*voM&B&9ag{EO3*g9iUaoe`k)W_ zY|XAIHr%;*N3`zy06jp$zXo?31c8x^p#cWzt?f1%jQU_U_0Fxai1r;4McvkdhcS-z zbyLYMELI&*cTAm7tm?BW!ITRFJq?Dlfb#}s|) zK(gjZo{tZka4#djRUR97=D9o+H0y-@)Id!nQ>JBDfrzpt5|r@vV~y;myf^I3{g6=q+Mz-3ovo=}h}pau z7K8dj`GgMt6Z%)q14vT|3g}wC^kd=s9uz}N4rM$iTabV*NZlfAPkLN**F2Ix&{Ap{Ou`^s+Z0 zD4zHTN&x@IQN->&1GN`K{kxL2vdq>niVI@Jkb6s#&o-BPKGpksew}z3tuWKexYzb> zVa43>TD|1zL_6f00%-UKGI2ZW#8hH320~iVk>%0Q;?a679dRI|zJd zZK5N@wZZxMZscrn(?yMIy!2aNzry6zuM$e73{t_0+N5?^zn%>WfWDT^72?lwiV|?4 zpv%h+qN8;kg3|yifD84=ZHr)9f~JiKPjhpC;0=Ana0*Emh!yFh!CXbF*Ean3Y3c*$ zyxF2;$5_Zl&l81{bHST$$e68nGigh<0E|5Kja!B-#ekN`+a8~sZ<#K2I1eP;GF#Buj@3~*=T6*EjlRj0k#p-ug92UQC9FDjzEY@O!#g@>MNe)YOvjwF~&wi8sd6q6WTK_{_54hRyE3#q@)`o{ z7u?A{x$vuJdp*M4wqvwGJPnfLFmDhE1y0Omz|2ML? zGv>N4r~=b`NqW}7%%sPiAt++Llx+1Lsuyt0VIB1Wn8G$eAIBAsiw+AJjw6=w!JI0m zQXo(qAaEZEU|VTy>Jj8B(HgHN&Lg)OKF1tyL`;Gm_gWuDE6A^Fg(|mGqNrfHbpqBp zsxPP%nc(VPudYpd_Y7;MTnB@`#HZxNm?ciLid^ZU~YtPvp(=_ z0FLX7R8dcuay+KmR=@ai>OsnU{Z@5JXBsilk$#X!*!xNmB0ui#6KuaA`Xv6U8t2bl zPs81QjFJz1j<89Lx88&ZTq*TjMRZ7xX(9aNNzkSqQ5ra;EP0~u*M8&TL>5B^F37w5@C$LS(VKSc1hgwR0hvJ5uiQt*zDd; zj%zmq1(yqoNCv>R!4;IZJ0KGESbwl1)-c-_7&-4p50-cCfD-Qm@IaY8=QbKbBbRf_ z4sD_GmD?FbLJRa^1B6xCEP`P5E5Y|q>bYKIlxD#;N73z~EhB932`g6R=!+~kh0q}2 zCBJx*6_c&nN{_tyxH+ibLFiwwbi1qA{+o z$JH3_qD}}FsUwAU>2tYzq+-0HER_ZG5>=J$3yz%X5>5`zE- zK={964y*Z+idNEj-JhF6iNJ|J9P(}Y<00SQO7%*ldB3O6$(P~4wTRtV3o$yM$$jFWOYsgAbf4aI#L9 z@#o~`?AN%vS#Av0I4kmy>gHgF(VoYNNhI6HKh>kiaPyurRTeQwh?vc*#3esJZ4ivG z$<)?D5X-#3as1a(z0}J(2;&3t`(96bJm^gdU3H#IYYR~sKfYPtmw78ijn|S#HXX4# z1}&zP(v6V;-gU&zaP57a3G~J@ibrmrX@O;h)`LgeUt3>-8Jz^RpgYT%-_nK-_IE#< zW_zv*^)yKa3~hPE4(&6Hp^)2peLf2H*;ufn^46FdiK@sLl$3A#yzkEH7Ow^a?RJEp zJhdzI?sVN-Z15L1Hr^nTot1!yaSOD$D=vqPp8W-l;Dej~Q-co!XouMzM3GaSW{Flk z6C@bdb4veuob2gP!2~+J&JPJ!<^zZ0nO#!zmD|YJV^hrK1Fci-hTW&nn-v(-r#kBW zoLo`?_}|m)!n-%}SN9gq{nfoe)fCNHuOG<5h>2_N*!E<>8Swz8%r4toF*D6J8>j5!c|VjRF4@p%xlauq>b*Qz`7XW&@2YWZ^QM{LK{$6;m+^o z3%U|zDTW0_oNbiSP7r;ezIxlX4XJ*xUuV723n@fiLwk<5(_9MFz&pg{mGBH(`{TQYcZ%8viCtlz^D6-Mh- z%y=bX+2W?BE*~n05&vZ(Z=q@dTmrsaohwsU^>bm5vC&Xz-x0LM zgi3F!lo)M&-C996Bg_oQ*I1!Iamu(g!gu2>`*MhladqQsZ@TkOkqraR10B(4&zYl{ zH2W<+)jF$ztk2R30MRZ~;jg=Eom8I@ueOp=BPz$*Lg_&^Y=1&{zd7NbRJP7-_hP=Q zY`t3L%~nRm-TUPz5D}3{@$GRr!cm}`l*fiN5iK@^p&9I(2jt!$?i*_}nz`!nHl*Yz z(aZU|5Bu*8zsqXP)_cN0rtBQ?p%TGJ2*6(&9$5^48q|W+NyurEiZZ&73%8l*(9X*B zVULV9i7cNQ^hULaH-~CrES1ns6CIpnoP%1El9SyjC$sWOzIQzZo{H_B{Luf)IIdj< zp|1rN6wKzWvMrUPX*PQ~_H4#-pA4I|`uRlwiM;Q%)0Wfi8L$uu(^K*DsAxd875H`y zL=FTlQeX{p4FYExY^AmxK=_Q1qOF^?nKk6Sh0m<~=eK9i%$_|yd)JF?y(#Vqackl(l)gfes5N0>!>6K>)GS04 z?83?QnhK0T_ZFwsm+AOtqR8*w9ge2iNIjtmXpNq^U43dN>NenD1APb4gnyIHO~~1( zlXBhTk_d7iAVBrN3!SW*zTTPOM5XJX1;Bla5;vfE>zIsa)+2aVHBKRNQKJKqMlTbVml*+)@ zq9^-bJ3kCkLQ^WczF8P9(mBOkMexRpkSEm&(UW!6M0J3SWc^nsCG*X?6iM}%0Tz#3 zA|~x;Fgli?t8WJH1th2wk%3N}j8STo?1l4ztCeB9FyPZrhSTG2>o(|4GS+bugN#`` zQ5JQ7Z_xp>6HJzmQ%S#7zZ`))|3i>n5N7+Q#_uI5NtnFnL?N*JB8Anr_SycpOHX%R z+T56rO*jgqnUK7(@@~;aHFrWW>~8s!MqHZL?uE4nH%X zqD^DS?ObMJi>#x#)l?7YW`DA=G99j1HctOzB8H(>Wen(xmrS1q0TgSO665%Y&KTdl z_}--mY-Hn;4>2W%%|qnJ;iz~?|8ysHV`6TnMCn0`+4wWLXY3KAbgukK&aK=+S_1dY z;mJX-nU;*gXkrs1o!JL36r3eYOe$Sjxr^ysErjtAuz#f}^WW#*Oo4?2j_Wrp_kO;~=a)=M(`d{~x3v~g8 z@rPF)v>_TjM`*s+G_Fl(Y*20kg`sRgr1-TDY~r&e~5BsGs|?>e!KQKLFVfGg zv8^m@cVTpITbnQpMMoZEfJc)Z+?~#CE=%?3;47J$`~lWJ4w|{H&cVl}GY%bUE}I3P z&E=j6@1iT^Ra0CMo~4jXxiA~bw7nfA4m<<^qBx$ePjjdeQ-mdo(Y&3BQ|(rmO}avJ z(LMu`6SJW1CgFb{?XIHXobB^Iz%7vH}Lw~KDSF9Uy_KS1)-e!4f<4|2mQBC1Ax@~lkV{1EeLcP8q<%Ebs= zHFNbq;w{$)JWROV)8C`wSUjwnkc!?#pZ-X^-sK=fjKOYEz_*V=Hm$HPO4IRBmC*43 zxW)93r+3B9aHQtjRi^<2_uSk4ccoKd7DVaL zP!d65b7v6!-;X?|1FrqsLrS7+a)oQg{_&8cS%i#89%^EB07A;gM;<#7q;@TTw*UBd z-3dKxmO(1L`lYUdBeie2AQIwxD86S-0=pF5u zifuozykiF5rXm3tPc-^p#}j96i>cNy%Ma8Bqb0V;jxXZ3Et|Z%;s{bBt9tQ29@0&% zq1@A_X*gM>U0*o1<V8&{@gT?a(t>tQYnJ|kg1 zwNq|e6SuWB9eMAZ-GBQI%1*P{2RVnEMu8>Mv04d~x3-3$#1jAvM0MPF9wNxUr_QD% zO@T>hVAPmTmrU71fvu&N=Mqhg&fLFuId>h!;kv!chgK!mn9D_E?{#wc1O^Oe2I%RQ z-1Bl9D9PuberK9M4zT{G0k`pVJF=bqciOBC*a4h9(5aonAdW%{PPoMw+4>jS@3>?+ztQ+rp7>J0HZ?k^w?(PUv zZg@sZp3j~4>V~-s1h)LTi1U5}s|kGUm%S5J8^h7eZyu<2*Sb$RLATaF^u?V|`M$I8 z4vNS>^4jQ(h?9@YVK^Yd_9m?PS*08Lp9+1Xmkw%2w5uIPeE%fh=yxgLj@<_$I1n%%K$h#j?=)h3qvurVxXY6*XS;BQAL`oZakZL`UnK-8cq)A9 zpX$*>Yz>0?SmHI^9>*$%|AoD|dv9}|i?#B91?|5pY)ZWAzmYbW#`F2`v@8An=kZ|x zinjrEiyE*P011$GDfeX&HInCV@Wz1L>k3R3BK*RN5f;YfusVCC7MUiw;{mdbrkG!` zNy{w(;*Ue}o|;cX4xLG#BLw2$_a&-SXE6LKin<=9&` z)nDgOgAYCp*Sny~zJ(s#tio+KBh%jf5m|6#fe!>>8`W4g>?l4wAP5ep;P@buHXs2qT9r*n{sR1i&JSc>8Up#nwq|Gv6?a< z^7M&zOp2j)1SYHQu6J=tKE+I@=+luOOLuea$;un^GUQ?g7P7o}FMr7N77R$ty0uVV z>g|L3^)8?e9)>RbgVDOn`UnPbY$9}F=6>uC7;?}+0APW54uCHkgphP|>a|Xqu$VB? zHLA11gQMOBdHDhfQ&^C0gh;#!_Yovb>yhk;;1uVYN_{S(aRTy-EnJc5Qva} z>ru`Z$<}EbuP9xiMC$jh+Ed{1X%;ID+A}0vUX1E~v>qgO3YpiKeu(_b=v>TV{!Qd> zBx85yY#=#%l=ANzK_}gW6O(;@-%W%9T>omA3$IUN{vv3!d|CwefgT_=mELiVh_a-A z>h84xNVbIH33~Hg`+?gh9R^bZYEa0yV+Z98^W?O%Xv*t4UwMHk=l%k~HQ1(m1G(kh zp2u4Y$#@)qK7WwNR+Tgy>)V9(c(&R~UHINJ|Cy{P=+5&{Ip1;MSxOVBnXtab2EGjwWrtL$m@i~I=#y@RfL>JosI-~&M zA4D?1>0hnoId;W<=H)Uk@GtGG9e}}ELKhQBK&NIC?u;~rn{LH>FKrETU}%*&wz@)m z`3&jZT>f_Tb*Z}Rhd7GID;MwyMWDm&aitLq*b&I7|5_wW1Bkx+m)VaPZMtfUKS%6y z{DZJf_B*PjHZZP&KXnERst_jZND-)h?(H`&YH)Hsh$HL=%AtEtw$!^OV@xVll+ACs zbv;N>G^BR9b*-}6bgF?;LoA;=$WQ1oS(t+N^nX~(PsH7`QFRRn8u&EXrTTr$W5+}@&jhpH;}$a{JwfwEq*OUF8qTqqaK(GJ+CF=WEpNqSdm+Ba{PTo&ebDNxO*%GLp`mEMuXrLMjM32m=7XpNe{( zqC45tVcmUl%?utKHXzcpu7!>_+s#B|$e+1%RFa;A{kGO_+v>_%9ipTI zZ6XSe!p($lv9w30L0!H-44g!G9MBkrhN`FdDv;&)yuG{w0({-OYOFyax`ai;$1c%c zC%!;Ft;>SvSr)i!3MObS{2RJRf9Ib7uHW4H_Y#vJ;*i6An;UiuwD)WT4>7r1V`zfe5uPEZYsXwX2fCN!v0cV*1*j69WLhRl~Kj69CR z=&ZlxOi%n(&68fcHFJY~yil&3Pd?AzRhp0I{&ddW`r({wN|1IDDkDKWq;|3fsed=X zL#pu+xBbEuw)L>kPZ?$tPWsl`oWbJMelLAOr?r{cNUe9q9q9GCQ&4OR@1`%E5NzP| zM)b=KLsj#4qreiaa1@7zH`%m(|KV3MHA2_%j%qq~2J8gr=ho;yw9{8RBe>U@B-zA0 zmq_!@e-@_ub!nDDvT1`qg)&~jDh6=CZ5-=}do>64bQvhoG@0q2y`!?vI+rPKSTB&{ zzu=l_y>0Exj0$wyoQpy-yOr5&v-t~W;lx*#itdXL2F~V+Tm`9rTf(S~5Yhd8`1?Oc z!e6*b?)@vSQmqlCM;}MR(cyQ2{?Uyz6lwZ7~gwe zFLb)XtSpS4NH1xtVx%Ap6nUt}Ja0;IpPvWfC6=o^Ne;JV>9Cjn>$q zcgkp8G5H-yT{c_vR;%mg$-s1oj4{QTmCFDOD3r^;Mr{E|<#8)AIBsD1c_k!GE-@o! zWwhz0kn$C&`Y%6}OIBrTGyTaNWWCwVUmKrin1*_6iQI%eFK>+hVik0m+meX%h;8+- zQW>!LGL2x@8U)ZdHq3k)vX(@#0F6z?)`(C`y@D|?j0?S0MYi31sa(Y7Vsb(D%x;pN z^eGEFbaP1WWi^fk$BP#_?N}K!)znz01LbZB7u?{TX~=_#R$G|FNeA{=#%mxg@Kkdh zPc>mJSkN8Tt!{2LRe)C)#9)$W4lLLNo71Fa@8(aIZ(8q66*9!gpj{M_ z`O8ikX1vP|30IzObaKd7HydXW3p6_>OQZwl!M-YvZ!~UEYB=QcF~th&NJ7COzA#uL#C zv+W7os=kEy=s@7qGpz=`+V>~4(==r3CQs`p%HCW#8*sDqWd?Mu?cn&N3-c6{BkI4r z_){+p?iE<7nNCL$EUkm`J1jG)ce&tRtCB;lD30;dR$sAgl&Ba+=LcS2G`d0C(iWe# zM_@zb${fLXoJNvmWQ4}5sa5qaQ1sNedInk>Hb z8m?ij>l+sEE+`i)bZ#T{Ms^Og34h zUQC1r>6;8@XW$7YPyr&-q+M6~GHQo??kooT{$-I4#iYHo51#d?qS^Zgx`p2oe*Jg- z$=v9w|I*^HAxMfR1NE~N7~Y|M6cOyVHmTaLeB?U$fUV9TH&Z<7=_fW`*O*L%w~DnQ zk9%z8nMOHfj?b$fs&CF#`zy`VOklctwRUIEZ$Kq%n!Kk4o4Q7QL6d;>*}izFpGUg? zw0+CBDat4YlLtmzdvOesHVU4?Hdx5YoM(1kd#>I0o!)O*h;K87hY^p)xH*GIoEj&5 zj{-<+2)n-Ko+P(OVRkg;_1tQ-=6I%g#i@V#CNxsG(2`Mt@jo=}|Gw4mpXuVol+fBhC9N9*-trjq9j|QmyoNoUW9*uC6cz#0?Pffo@CnKWTi} z8^lO;LTWvX!mcY3404ElxP}}eV?TT6*%+3fdxxuFu;zB4`j46Z+`X7zq(FUI0^7^h{h($PU?c;Nb>3{^wwBn)f!3$IcLx zYL5{%V1ClWQU#LOiLwPGF9AOku@Fa?|! zS+Znnhd?HEh%P@s4glv*a(Ma>|IVcOdDXZ2m zZ1IWzlTD`oOn`rU`$?ej@wcZoC-_-DPX+W7zWv;D1kP*wgdXGU`AjD0zSv-j@bi4N z2=-RLxC%3*NtWA|?i;S>C@zXWe8w9pZcz{#GuK%N7sS30fy72um?&J}QxQ9&tO{@} zQF3b#gLX=$3tXz%@DrCyn8hQfbZY+7O}CnF8a$ys8>hv_4zmViYgc-ZUt3=J)B7v3 znnLC2^B}l6{Atk7Z{L;7OmgS+W(#C{?}1Dzn)C zUd$Pd@SM9>;(x(>kgmQkj`y+7Z2UtfC%!v$J^il}RCpQr6Es$~k_|zJ>rqLnP3tJI zMAIq9yJWID!w|sudRM9+Y!Aie0gw{7Cp{THchAxz0G)hZPgn39^}ZBssvFL$3oMA; zL#2BrIo>zc2Yr}JzHnzuNMGLD-zn9XaH+z%wPUe>(d=^K4RJF^;gW3M%vm+1z}Q!b zr(@k#%H5pqaa4G5)r;U0AgNoi&hOw$hL4O}f0g-Me^XG9W1+t(D0~k{!}xXY3Y%86 zy1dHQ2Y@aOW;Z82>##JS!GFkbHG@nTIaamTnPQO2&TRh);c;snotI zPB(@sMrufM)UQiPiHz29iK|=#Yyr^KF0(fdv>U%@)ZX~!d|;O5(On9_u5g~XTdn#e zoT2EGzPt!aPog|m?gP16r}!|C^V4S;_RnAd6cM@1+`3 z9sZx9?&=%z-UT??Smvhj0wN^U9c{UFY_m_#Z93Zo_rx!4EmMuFN7tWOoYkvitotqt zy%gfMEsPwH{z*d>uj`tvJG5i44#PWHFU~8f?+eMTspEY7s z+BicQ=hc@WNiybq&l7jEnA(vzpFi11J!3JK@4hhYZSs<6;q*kU4;h70)E6_0c`guC znf4>DSMnl7MLh=M|294l|0h-EcUMH`-&C2rmCN^kC)Z?vv|l1<_-_%Evg`i428aHD z>}LdJyCgdPh@gxcLmh~t(CKFZ1Ybd@ALVBXJb8*a=f`C|=wmm8*zx_q_o3eH4L~k8&;jeOdI|x+e_{;!1uaw))+%Ibk3D-ZZV%yUB*8fC<>L zBQ6ml@Nngq+iLDdGGd|5FJp;cO`U9iH3eJwWZIH&sPYh%t z9ko^8W~!f3Yv=dRSJ!PG?gUGY0x%E+ysQAKy1T)BPGPSu87z^$t`26wd7)!fcZX*n zr$Uydj3o(LUuL)6t~+OO`MJ0GsGVjhWQiXZ#z}lx;thp?L{h*W#xCa~FS_YCC7;%c zNTv{A4&YDF#bF$iSJ-rDkpgz*i6Q&c{l#dvUbdv5xi*oj=TvF@7juJOlU~h(%u_7o zhfA1U{V#Ui?O*&of(Hv5AgP z6L8{Opv%+nNCPLK;z|ERNHc1b2tt5!l;7E)9oV*jT-iN;fs9qC2#m z$ikKkE*7Ny=9#5t0`9}Djxk=InfK2>_HIr;JYE7eMqb zOLt4S0T7eaVI+j|ZJs@=^C9eRL9(O;%932n+>Lc~rcbSkCVYscc0Qkp6R4=OPb)x4AslN^!{?~So~!2RL5*vuHUt3?r51u{rQXn?BDvouIPM zoD3X=$fO$`-i-J}mg;u^oyfpe31S-)5Fs^0{-E&@)rMS^rw1KEUZX50*;av-_1bL*nsT96Ac}CB3PaMhppJ@7e>^8yo~ly_*Bn($&6{U0Cuy= ze@34vJ~}y;`>DcfH^7Os;V_90O-W#Z)^!v$;=p(gO1fv4ucgvZoJ4rVg!`83UH6D5 z0AWz(97YDDVXVwNSco=2AeESBY(CAdc8)m1G&evHaHVz>o~8Qq%p9jyVa%Hxf7Iq2 z*u=!OE8~jWfRiq}H>N+dOKxb;G<6NXF~ z!q~Q(CdN|cd9Yiz-`AnqjvGil+`~%B3)(apAEI~EEFT?BSn9W|eP-#mJ+OE&&i!eu zhXIeuSvz0RzBsBp>yOwP^RiowdCrGdv3ZNbPav(I)px+}>k^L-v~A!|{I;b`(M>x)l_KbferF*l4jN7ib&_-*+C zLY#u_2FmACEV5~fgwfHz8?i<&_<$z#Q5#}yN`5YBBQNB8`)DIe@KNaOx5J6XDJP;8 zd1K@@b@Y`my6KBpWss=OB_#dd?5}E>?V{ZewF{l5a`4|QigSTmjUm3NP)|@#CKW0B zqRcc_v0)xXA=&EH1y!Zpi#o@+-Jb%M$Uc6*wDS-hY4V~ZMP;^Pun7)Za&B)+$oP&$ z?p!C>NPQWdW-A>I&roW{Y9-2zcP!k!$mJrCeI{dlHu;(_1%^5>Hb5Xx=q?*y;n``% zd$(_Exp5dpP3$I?=;$5i)WJ7)xv8uxWyQRd%bLH#IUy4A|BQrV@TTECkpKPlax#ba zKb+|ZI)eE=J62Q*pB6n>^mOdLCuT}`z0Td?r-Zn!HNPn#7@^5+j(y|qK~p5}6O7Od z)5*zhOq0JWe^)7O>1NITig0f9VJv4H-7u_D6x zm{nu+*a4XCnqQVK9bc*c)X71<^X-Y&Eh3TUeQ^d?e3sv2Uyz?XN$2VIEJ0#N?4% z5~reHb1hq7>5E$)=ucFIc|uQeeY@x9=U=jb$t{j>^jp5?fgqV3jPLio$GPqle>Xx? zunS`EEMbY_jCA2~_h54@jCL51yBjOKPih~hb~qCYmS|n?gdniH=yM3HBjatkL;ENS z$tJHZYE}LKAV7)IF)l%pIOCP1N2?bqUe|f77^b|yKW}i6{xV_kGEWBx2B`ib;Rc|I*XaJC7bpB`5|%)$c094_5r@#x z95@k@1<LmAvWST%vRhpGWfpcO zcCW)$lL{_(c{ob1lLH^QXq!s`!o-s+U4QoCA}H5ZLj2upY$F4XD8GnVcwsw(DEKCW z{Jkw65;G}Pf6U*QYfh4U`S<~w*hgfBQtA9j#cNBi?R1MK%fE(UfEq?MUjpI=kRQoJ zUxo(O5PbnsIj&V-{@(ayT5GMU=edT$jU+2(TxMsOSTaRLUi(FhEgRp@Sk0ShIIgD$ zCjATXh38*}K59>2|F{;eKa36BV;GoyxzyPm$0A2q?*vM}eH+TqJUsQoERB}lvFrR9 z9$`APWHua#q{YFpsyfS#JQ@;?{qQJG2_I}Uq#4_bAfmCzhanhK7~^Q}E!m>wd=%_L zL7!1fKYxE<#WKh3Osf)bFfw0pf@S z)PEFSkYiyMfB$h}ryx%sO2t+*@tz4r)W>u7K8hj0l5Gk@~>CAAEtWKlCVnKFaI zvuS!c6?Nr95sm35r0g-|=-i84vDW0zO4Va_gpWHn3=bVq`H`zwbuAh7MVds^$7!h}yaYrn&KkHoS$<<-0Fd%?cnE7Yv zo}elEWV^4U;rPUlpZf<=-S1pR#)2L*zf6ne$<*nM6K7w$fBfDU60|)c9gjQWIIpt4 zGdeerSCbPOd=5xFQN8n#YxfnR{liN#+bLY=;Njj*o+A;B0^3F4yR8o04-HLs zLx>g0MlRD!INK1CM1i{>)6TRlN3YCu^cl10LZkL3dqP53?}DCwp)(BKD{q|9Yi}Yd z0uv5hC#o_!>fibDx$HI6qnp(sUR5K2FG*B%&p1I9#8#3*GPa=|XnucFUrZ9PtR64J ze1`T8JTibj)yd5jogz== z2;7py>*u*a?=u8$N#iMdIUt|te7;U{=QUVLhj@u`(H*NXSjD?ax=LbW3arTc8aSC7 zwZ8cZ5rAs&f4fd(cl#}=#&!6g0Z4x2a6qXzKso4*RsyMtve~T6w{qX{Cpv#Pbgse7ajL=`)Qt|D$!LML_ z5kL;m{7?PV$zN`K6?Q~2pLW|M3oGhPpg+QYA}U&`B$DPZs|;FWUU&z*HoJV7|6vwJ z6pzjz1-77{j7*fu<+dJjFHmn-ZupZ!qQ0v6V8b!J>KG#dbWiPj!A+OWQ`@ z9sTF8zXKm_aVG+FK9Qoc4L>3V?odq2Upq(- zEy@?4zc+xp@r`$PeHlw3=-+4K`~4=1q(_))s1xzvkkblv%AM)PQ<@N6EJ=}0Hp{{NoYx3bOgT>kywnmmB* zcwsB<&lw_!4v)rO*tSSoKCMmpM0@oZVubG7rgeBn6q&NU&xM&DFo$30!G)OhJbdP~ zZj##Q>`CBIy)nt^0>tR26T3fEuyb&F0ToBY`kbm?cvVDtPr@lsy5v={s>9QK*EgWI zWqPOVgGWssjKu&o>G@Z`Jvg=uicpfTwtm9Q0V4UmFJS+xp6l1;EJ_5??#>!ZPb}!z zy<@gBj$0Z3RKN1|#|fBLeA3yN9JnXtklA``?b7qx(l-|l7Hfr!a$uJfrDpQU(fK-N zCc@aT<)l2_qC00rA6IDOOMQ#K%{U!x3l1xMua7i%GY-jbLo^IhY8UfEdOhO{7+ zp8j~r8Q2-&z-#o0;9{Q$IpKVw4u$MTi^TEYTsKedPhxR@?_WdY0xHy2^`UC%P}5X}Ovmj}|AXza9D|aw5X(ZNwy6 z7u5si4o?k3m(J{R1kra2<)P%M9P=VOFN_TPK5;6|_Bk0?T&M~xFSEZT{%5e-0K}Wn>VhXO<&JzNZC&UuT?YdR$)I<0ErTc94HDa^9Mr(T!YbUQ< z{KUqahCVW>#A&wp*uQ&dofXb$*`S54KbmmA9_0I{g^Ksf9J;SGM*AZhi^I+PKq-8< zqh4@Iuy{OF3i~{4o9?XOO+e1YBjHlc@$NVRaHb<1RNS^Pv+dEg2plB*!Mp7&@LJCf z8iG4yp$i4JJj$c?+bzvS5zoKR1%DaP&ICetLsQ=Y_g+y)^m9I4{mj~%_Y z369>$v+dbb;@c-e!T-)_cqyHwW&^x?d}vrgY$z4JLEiu-~Yp3<9fVnT#dAF2nh79pgE zuZr&865WhDwo{>&Kfo?gla5IX=h(gQSpo-sU0%9RLQ&yZK*5<=29}f=6v-#n-4`h% zA(+g1G$~^{^X}k?!m1jL2Hnb{MF*GeC(;*JB<45h(v{q*XkM*_mrww%0|y2^;n~tE zwDvu7_;oy6`j{Lu{;NV(5&y^bpZt)FxkUgDfZHGBoKt#`S?mNv`Ftwz0J$}Led(JFgQC|Ol$bKp zatpKCV9Jx}?GNA_I8k!h2V-}1u2v`$R3j<2q$VNyBiPXvJo(YI@WqoK41i`{kv~gf}HDEP&;bt6cs+ z8*8p(0Y!hGcpLx1Ox^Q4e)0My4Sx^|tlv;=JaE_;k1YY4T6ADlh_kk&7Pxal8)2eS({~HC6)k z)4HGTq=_*I+O)dk3DnsDG{WkA@%JG^f0kLq`0fT@f%w+yJkFcXrZr%Hi#o+7^ycH8 zN$@u!a#$E(bb}0e%OOZd*YnPcF*`Z z!CIzShZ0@+6!nkPzrw8yo(?e|)1ibv=mru*fm8emw<>r_I-?xJN-E)#PK)V@;}m!* zjAHEUX1>cjn)2acAQ=E%2V7A@a0rjFC*#>$0B%*J7@VUuzlCdeJ)I7EQ z0~M!R-F=LzXI-xBz^x_fhdTELIhX6WZIeQvh1neV2aMw6V?Gi5!_bVu;AFte6u}LU z*uwZk!`Mo>X9T^g#mW z2kZS4@JU8KgFAcR+w>q129b##t1E*gGNHJ*Zw!^0qE*py{(5F`UN|6IC0CwV*PNuy`!9nNp}pM10Pj_4XnrOL?u0>+2O zXw$dZ^S9=TKDPMdj%jn}ls>Bq-($DO4=B#%y}YDu+LkZ`@3mm#=n|IHj~e3 zfB+eWjtz-CvWa}%#{-l-@1YJuQu({D$uUKm$^3)?%C~7r)Da;}OtH^1VQ9+uJ_Wqmlca`SJ+uehYSD#f}i6bssdHi53iohk3%paUNoT z{K+LW(Xc*CxovCM8ndz}tp~Qt%5UZ3HP%NC_w5>0a$99?I1HvcU3u{=V*!s=F!UFL zD}q1Ks=0nOfy!ox#+86pb^oHM@igIE&%gXEKw)8UBQ!oVbYa?s-}l!;4qxrULOoWRk2b* z#g3zjq)yIKIb;I@?AB?wNB8);U!|7h^9KFbFkr{A65387^8FQGI5;I4}S z15C03*Iv8a{TWKSuk$rbK=P?ShOwEpx&ZO-@Fr8b1iyVnyR zPdoF%!hl3G`U*n%?a^Q{-@%)Y9HKh7A|t*xeTBiLrqv%mJ5%7?mUE7;b!E+V`E{X~ zHqHL|2dmG^_ui-Id|R=8{pcoxj6H@NIDt7H7=)Q)@sxoKiH$OHMcA% zj0kv24T}&u|I$M2OFKfVTGE%o(V;CP)@EIa%1M(RYrfmPPa4|v6kFZ9WY5Uky zPv`8^cVw`5t`fc#YJyGh zvclVnS61%VB~h5xoPEzG_OM&7#4JsgdPed@`oi^eb#+k zc?qeEjNu&GsvIkKKofy|8$Q3Tb?8Ig(q}1cN4dXwKk@yvv(=+u;Nk(;fFQRaWpO5# z{b6Bi)LM6SY#^P^=EXBM*x1XQ(?=YJfj%y;N_6vj&+cDZc^7k<-k^%zcg}XlWB%#L zsGPm10XvCp?U=m1ATU8nimMq#1bCUT>3keQ6n@pMQHM?&XkB8`BML^YxvDP)pAVlj zh8}8vSwg-qzOM4rWPj~*kgCg>EuZkkWfsOz+72PC;gKr{i-cX4OF~kGZswqPNm+SJ zMa4;`BdnF>6C+MEJhqF5gpgf{;1|7Dpw4$c`r>6WavuO4cT3Y-d~D^<5>eN?I2iB9 zJe(z>db`VKqChh?d)HuMcKnV3>%g2oJ(Dacu;bJzqF@H2J$k<<6SGBWj*;(Y(w#D8 zZ3#O)0;(}*oQoJH)cI{(v0R};8$DXoGxMV=?`k`x} zZqYDZh5|&2dG%w%>GK(%h1aFVez+{XEv~nf`Us-*{kRJA2k^TeiWFrZT+X&_OuCn` z{lGHu+}a_B<#yG?Jt+pF4;q<<9Q8Nb7g9&?qj{ z5^s#Ut4HVf zXLf<~&|GG@%^P8+B5tU}0kL_lC6ySvji6M%+MW6cftK?qnNIMSsCg?S$_VT_hQ+R~Q{t zipB5^!5cmhCeT@9J_f1Y(bqT5uCSWMe0RNiF9GbmVSI;jT8m%(J7>vU68YgwEQR3l z5E>pMpoS=8exLL`;tZawTwLzFM-UM~0OL}L;x>0c0_un+yOMiXss|Qe)~T|K>W-Ov zsLViuF?J{pPj9MPz3zjrI5(HVLPE=z)R2mbOAxsSV+&c{DrZW|(zxC_mY3NG6x$cy zCrpqcRhTQat6uyuLE@ww_@@cd$!laPVxJ$Gy;HbNpPpXaH!h+v5Pj@6 zTuSli%7o}m5wmLebwOecIokY(E_TH1LvmG?shF8T!`-ba>u?NefZDfF@>~^QhB$Mz z(d%4uIJH3){<*@!T{kIx_c0SMQ<_+5vO-H0W(mY?jDNx4b|4Xi#P ztmK4*cf9`Sv1TUa6{f)O0#mgx+4iJkZO$%&Ve0uDvrm2^hg@^$4ktkk8g+!bI$Y~t zt@QE=4N2I+UPTOGeg3=6O8fWSRsFV|1boS8IK2Z6GobK?stM64q#(bDxG)CLHY{Sp zN#tCNQH*~oS2DNwhb2v=q&OHD!qQCGv#&p0GXXfEZdwi`t9>(3bq3IL2$n(5fouapm`5G$7!rJndahvN7lbIIQ$Q*+-J z=1vF7F3m|^Z<;0?{mNgYR;BgngB~>n@Mn7ybKhrkYQdD|hHSGfCfL#|@)inW_qqxL zndhd$l4$tv=SYltX=l1JlawKIXf{7eI*$gN%MEho4MbP$(q*`%(@(Y}rnPMAy}e1I z+*etH52*)s7_Okd37-C^8tea8Bzl_TevkVYiEjOYM9x2uXr9LW4d0N*jlH&f?rkJPpg zs28l&fS}!1H!00Z@{v?>OWb;-G&1mTh;;oAbc!UC@ft=ksceqHsJ#dfMXSLB5Q`*# zC2o*D8itT)k$I!cl-{oqdy-4WE}l-AY?cUg0!{+NO?(@YM2vqO_keNvKuGb3HFpOy zaU+V&!)uumb>Rld0t|jQ-3u1VU+zt}(@KE?<~7u*6}t%bEYrH5 zOCGPVWPm%cKanGN;0nqL*OGYyK^$QFAaXwCkKV3aDWgzMbE~K|&vd)itm3ki-Ec772Ys4J z`QV{9HM=Maskp1#gYw$3x1}HG^@o#6Qr-HX3tyJw9;g(@(w2Q@COMJ3jDv$wIL>z` z3qm!X%w~Om#J{`Mg5KmK05hudi-uyb^?j~X;dN3$0^)$%k4!qnMc5L*7VOG>&1a!Uk;0&@kp`l z2I}muUaxxCfW!^K&HYmYmc;tLO6jWM!*tbxO)Wv2QDg0KM%i%cD zZxT3um_G-V(sL}j`%+{x3ZiNpKnBpXKa&__e8-W?r2SyMFX`*8fmcX!L$g;lrlNaj zPJmTkNyf?RWPHi(2OWF{!#uiIwb~D7`d}A{^MwFPFZN3Hb@pIoaXc%HWx8C%*fQ9i zYI#?E(ZY+b^6%^uO(bV2h6}1ugp!B?&=r^6bQ~F#9ZmI|o?fAksCRR%)Ne{%UaZCj z=-3rw-jB`H6f;{6Un56tR@M%iY+tKq>4}90Km_zZH;jA8U2kOmwLMJ^;0Evjq<%&e zX;NRo#I7HCq(XBx;}ul%9w%oD zqXj+2APeY8{~(>-?qwRU9WngOI!lY#;aeOu5a2F4U)HKXh1#m|*ejE+njqg9cEUn(=orYsI4E)rWStOpo9Gkegq)Km?$LRvba005ukG@Ck z&O+W(ARyrr-%77NAPd^>s9G0VnzJzNRM{1p--vwg=T2rEwEQjRLsy^lW1Mm6t4}kK zP%>z!eAw#QsYaIXul$2{MF+k!Dl#KNmN2A$-8W^mME=^EqsIS?Xn@S|&-kDIuPl>4 zUA@IApB)eND1}VM$!m*s6Zp^I)SJE*52XP#Ipg+o;P7Ve&*OS)qKX(dD?tO2AOR8knaR7w$$n4wErxl*7~jWeLHKRF%T$I#jW?pJ9gREr7$wS$GwpD^874!GQi~> zv{?6Iuci;D>P%E1w)sQ7m+R{;A`B3FOt}h^pHYJQxz%sw0BCZ5=zZTR>u>$cS;|o6 zYB>m@>zlsg4alQG0EIQZJR|V+c+d2bi6=nJ1A}8gOgSFJ+wL&|q%;rB0~O_n&F}W= z7AFSpIUKSDHC|ZEy-O{gG8-mc7W)6O+sxy^$3Y_dds{(qnl2*%(SJ{ucvyC{$A8i# zphLx}D(m)R<2(?7DA4f274tLxym}<(sXnAjJA)^>Z(8q!n044S;}_uX~3OZ;%%^Uvka>B!VAk8S&aI0hdm!x z1`4#X8XtA$AQwg5ke)khTUKLLx-8Ec`Bp*+?E&u0kt{m ze%U!@PbhKabRd0)Yl&Aiq~Y$rYdKK+o~77?OI54St_#4VE*STS|T9*|4Xa=WAwer0l;}qVVSBJ zHK%Md7-XgThEEws)fDM)3d_R80(;S_l`C?Y&GCd6-&ZbI1gKi@-B;VQ8!Ww1>S!bL zW^K5FOPAr!m)qQfDTVsN57%MgI76}G7P_2K6%3G zTr-38(YiV8c;W7x@_{mR zoHzV{ev2R^&oY^1i|ObHoV?qE%Ib-`KrA~nsB2o3A+0vgT-s4(QojAjJoYZ{cvxH@ z18tXpqC`^QdIUa;ZZIm4!cu6o;^$G8)wL-SA#ArZERgI03hD4=dl9k_QM6$ZA zr&SswJEx92=7jNCybFzjk{k{bW)r=5n=eNzm|_r(HC&yiY)?p`TIX`JtH6k2>!!?PBZMO1BGDEC`m|f<+ev2h&WCb<4PB& zoEzJi*jYfzW-`U~X2dh~aENK9^wwMuYm6u?4ZRY=_c2`t_IUTO^Fw65=roGFZeXF} z339sss-vx7I~<8|7<&H@6$6FAM|D*nEf1BMz{IVIQQ0vNXDs^dUS#=cayVD_;xNBr zCenCgaov$pcBaO6gU{#jn%TSw-d^4A#6iqZ$So!b^Fj0Qp3G`s^upJJGTCZYSX{$o zV$PRpCX;hM!mM0(gURjdcR>g#1Qg8s;PZXz^_v6WP;p=3tGTGM7w05vy4W_&N4o=; zmQs@M-Ur$sk6fI6%{k+~w0F{GWF7N<&zN}7?yW$W38n@WLb%;LwS&m?x@P5& z=UzeMrR3m~HLE48@mZ!$RjQ$gC( zo}36i^~s2~NIO}4s@BUw$-XaJMRA(;u(UXw(PkQuPOB7Be6CRhcClCr9+5IEdz_K7 zo&64`t=-^Rck{(WDFGIU{cgF!%IMV6G*#OFwF1_JA?6q#`jB4jL%Yl-u+vuti6yu@ zR8s5K3cb#Wt=%Ddwb+OMQG6JELuBO>STQn7eY{r|UVN^c!!t4*gm>nNm?7yPC1MbX zLBB4zT-`cGxS{*bv>8O(*a_lfOp%fP2zwdj{K7ZqyFXzAG;KDdoPEBLl(2#JZTp3* zd;aQE$(^31;RhS6*pu%z`sJ%brEyY_8x)hB&5E|GMJUmxBLD(yQ>^F)4G7kv9)i_~ z2S9jb%G3ykFP$@iVBO+b#d@Qs=3&QN5>w&37HyW1VoBC>uIJSZ=2bY9Bqp9*iGOHy zgPCxZ+FQv_(mMV+vltbvig?LrydF2bO`^AMIIot0li=_bBQK+&2bN6*HmSTaf>gA2 zc{1^MiVI13=~B7_cOIzIsq(9!w;obPc{lC$Wd}7e38M)h~@b9#&IzzIZVFuYKPtT5Cc`%|2{q{e-IRwc9OFC$o(HuS%~LoAqKy z(Tlb_OTF35p$yVHF{z9%VGUau#(PMZvt+B}jr#qmR44`?YjO6qYgN+*1L0<0C*|@f zLIuY%dAZeJ^v8}i`R?ttW2>)VXL`}~Hs4yXDX@D)vfsD*;OOr)K8-dd)9c8+*{mwe z4A!)BM)pfM1E<2yiTF$aAFpS+_BzQW9p_^~0)?9hwn|jUkwe?8C_&PQ1!5n4(=wdJ zO&S&`#L=aLd%UI$<*^yDd?Zp>dm4$84oi+5gUF?6M67u@@QaMHXGptCR<3r zP4wx(ChzHN2a9x~#BFe*w=r5E4O1bpwV?`gvT0W02Y1Eb7F_S~5e&`9Y6oaJJMj~z zJE8VFucyd7zM96m<6-Yh8x+pIlUd`|Dy~_tfQOujI1wM@%ypBQ;_}hOr6@C^-jl0T z<*ZVC()Cn^sg-NdcPE-geddBA2Pz(TC|ntCj~p+ts24-7*h0_5417}?19|kc z0|H+CHLhwLR`ZZQ zJIkY5U?mOFv524!72nj#&DPaH;1No{Y&&;b{Fu4G&Nza(FfG2rp(yei&|H+m-64y+ zDsc2vh^1%DeVpZDENJ+$+;Oe!QAEU=>FaN}tAc9lX4$e{man=*RJYb%aqyUc!Ayk+qQwij;U4PTdS6CP7jFRP;HVk%cVaPdC%J2B9=w=$aE#dwNh<{}LoQ}e}m zF(Uq=Dt8)ImXR`1He$i}9)Av8aO1%4zn3+e3x)BhNpBG&a$AF_7?tix$IYhr-x1TF z6nJ%3F@{pk4NZ*7YmUrUiegr@{Vc0RK;jhfGk4t;=c&Dj8wYj^@SGLn?TJ?K|3#c9 z1*b+6J$)`)O7Kdnpv2vLr1Xx%bpER7y$=;Ye_k+Y66Z^6AfUY}ysCVp{{E-Aq4W`_ znMRWx!X@=lXN%@Af{jF>Sl4$j8}-+jlb<;jdvs1z6R+qz7|TkzLHktgcm&cBtyn5T zEMU9v;`Yau>*vx@YhBeJYF&2<%`9621F>26n1RWF3IkZ-agXB;uJ7>NYEg7te9n%? zN_@ogsQOmx8n)o^+Q;2h+j4m4)kjWtK=%k#@q6dT$cuwh#mRlh!trtB$ub0J2f6e? z7c&p!nZM)?naNSo(9O!ylHSVoe_iEq=T7$p>lBL?K|XGmuoF}#B|K*(LXiC5mLH>r&t}CB5makI*n{^@Lp?} zKZDRr0zgjM4DnetS_wA1eORRD=I0Ta5F&EvtQND)bKd(|;`G_~nL^U+vXIh**2OQ% z^+!rd%&-Tj;IbrAi?Vz#cgL*OS;tR>7u<SeRc94wHRO*$OV* zKnA%HZ8l3EspB3wCt(GoQeQWbzX&}4{E`kF?OILO9G?az+ zKHs{y=QcDXAGJ4p$Ae-AWZ?br`LJd3{r$0}kf=vF))njf6E2J-k0!~;7ItNe`pYIY zB(7nb#es_57asORbWZrd2Kn^Nd%{|D#9SUP@5^dq3W_65;%@I0R5H}YYgzpQ-I8}s za(Ts4Vd;|I7c`01#Wl$>;(E7?KBp_CvrT^Azcn0u`@a4fD}vsnSk^*sb@5kR9eyXe zf!+ybp*-3BY~bu$_z_ldvcH_veR6;b6~$2ted1gQ4C;5NmLj;ukj)rhP=5d6I+AFTg^mw;7-j$Vn*G_3q4Y zS0E=(8sQC<78;uV06vY}dZpqPhPqf^0iRs*Ywaz}L9Z9|_$M7R!lO5&UXtq@^pgkeJ-10+FF#OZ8MPEGO*P5%kF}f_( zf^y^U$g-~2vfbZVP{fJ}e{;XEyEb~UkY1%h1XvhrCBn{J+1sdqN=D4Ej56fzeO+%) z*fZIS5v>=w)8Lr^Y4F_Mrmr&f@-uRu_7a&iZy_TgT;X)Y*WVkF{od#T4y3eQ8AkXX zUDCfrW>%Xc*Vo8zCdT4+E9bVK?qN%$NUiJV2P}gr+z-w%gfReIO=5FV??F;jViTa% z$%rc~_-Sz(ub<8bk)@_&MyYaNirM8q=-cD zx=gIO((%}esV?a3mtuU1%c~{cQJ25u7FAz<_B@;vE2!i1rtXV*R#=N;X=x=0L`;L= z0dB{Y7o;&pvGo8C zn`=bi|1!1}85h=&a2hpAkmW^}F5 z^Nq2nxOA4I>0~=Cf%3Em{|q0~|9;5DiANNMSWALN>Ea?=gMV-l>vAK)tl8onx^NT( zBv^WM=WCplkHvFquRfa(>C)8C>7;jNDw3`|Af>OPU4g>SgYenk-zwD0)`4Rj)v6W- zi&Qqo+}ziOGM}5vn6*nwz8DLJ^<z>j~g1Cp+_$(@3x{n(92$ zCaf`VkHwbtsf%SjK9^|u_(5oP6Wq_{(UsKark#&J)-sz$JXkKwz}7x?4X8>tDe9Yc z`!wE0b`i<*Ju>s=;%8VXjM(Fbhlq3TJ}~+CQ=L2Wr%vzhPK%b6D=Qj-$LPH^hJ#5W z4=ZG6$)kp&9nJQ97`&#zb$@gI?$h(gZ|@eEoiE8WV`cGHF7CJEGH&Po#-F3rvrmW{-SRu{wqI#1L5igCMebK#Rsg-$o>2$U7LJ_q(WpoEzzC ziboQwpG$?aPfdrKma8x{NYZ#t+YBkUNWHGB^(&+&iNL!|{c!=R z+rz2WYpWwTefT^rc5Hnvm}Ew@e@_2s?n5MPduZqG@to%l>iq*e&kNr_l3b)!R+f^U zAm0qIuGVf0dJ)@j7>gl1zIf^~kAns0Ym|-p?;am)V&e!dLlbhbC(`l4$Abo>3~1@GYyyc=Ewu`G z8ZTw!Qh3V=bK~&dk<4cTC@IVTemDOwi~iNEIk89!{b=%qGVEjXl|rr0grI-WLr`Nu zIwdr?CLdaT-rm_9r(XHx-eJ#7Z?^1%-rewitNwfy*4KAPP$PrSuGq;Z9`(&2OK%3M z-G1y++FfRl8Nx0tepj`^7V^B(U{6TD(&2(gFl!B}!ry6XxZP8m%e2_uuKc7kX?;4u zEAuk>c z{pS|&Be2U_bm;^^I2b0Kyu3M2?F-a`P}%QFQWTYaQtCQJxvcxFl|_^Jl@=LB&~+HJO7R(VAF;8BgDK zM#y8^S_=!oLrq?G`8%U_)3P4-m*11|$F3*^8ri6D8a$txQwrxz+v-7_<0hI86J1yd zi6{(To_CcfZ=Yc0i#?QQz6y4y)g+x7Yy;_V^aD{DCACAMWroEzrp*AJN^&#WT^moA zqZx?6_|~8TT>VLwG)$OInr;t!)e7D=@{loF z&NRNXTaahXu3-DrZu%(4sF`Pu?^)oP1?~78H$EmHn8EnD!`KBrs*=()zv>c&SHq6y zZnb=T{MQY*?XKEOSLWdUmu=%A_g^x39mi zL&vmict-HmZLrd!_!bHv{5swx@Lk_~)xPq5Pd1B@t+#q>1E^cCv3UXN0%D(X$T4AY+)*A+w*Au-|X-}~-r zmfOoj>?uFWX|HsVil=9-5os+U7>TZm)GZo)3Et|ZyF<0w=#BZhzPRS{ZSFY&0jPi?r8|0Gh*ioM>5ZlO zy6#=-^sdb=jf3fl`&I*9;~6>(`Yq+kP$aBKM5lK5t2>Mse(UrahV8W~jY-KykGmUu*x5Q`52KS?uA>$Xgh ziYl>5bWjaZu+ARnPKbgH5CYK%Fi?~wZXq)%Uqk)=miiN%2AlxJbE@pv-lxU#HMi>M z3(^=89g1{aNriN>qu)O*F8p@O@wwQaLK3uq!@s=->Y%OxSWX)B(90hwiu4JSYktRR z(Xm|^)y4S(7sli*Zu-n%Fl|Nhrcg8RyuhL41YtW z%Iuh--%|soO~xc{%uHK2=@tl?q!GzswGgdnfzG*dGRe>*kD#eER{UK1rd*&*Dk`3HC;;jsqqhK@V1jrW{%x~bGIn5G; zEQd1z|nUbfxE~8A9}>*{Z!}l+%#zSr^8Le<2V_kaVu7-xKn;^ zb?|80>q4beNv_8{D&l;-8L|`@I%zt4-B?=KG5PyS2c4J;798$sRbLdn)hvvs-tW!! zdiw(a%@r5Bkwo55t*X`joNKSb@xc+^)#avvRvVX1OFhiR;Jyz+?y%Hunq^<0_)hRb z`m9!4IIFr6S?b0@57{}Bjt_62-T0VZMmXOPXQ>?lCl2BBWxjx;3z=vi)8Ds#L*NUj zoV0Q}O=N<0zX|sbL$1K@fnrWRjl?WNF=#V0&-DsupD%iI$L`Wz5j9r2q%Uh+NDNp; zK4>U6t=}4dR8t&dk3P<1buw#koh=l6i3NH?~ z&~62RPv@k;%GPnO*~5A6sflzOCuq45+B{9xR4uU9H!CqrGG^$oN%r+6Su(YZHX4be zO=G_GXQ5{T+yH@plkBPp`B$Y|jYF58S@p@GbAfu4u;#R*Th5X{;elBpy7Lr{r!Z_> zdg+CDC!8tZz2D{;NuryfqAj!-zGX&$Cz;9?by;uqS`I}wxsUXYpg{uE3nS#^TCACG zrzm8~TI1B3WWN3v(}pJaWG~-k7J3h_Q~2w=@_a_b8jBP9tn)&$F!L1*=tz9<1KjWM zXp3lTye&(2>7K?>D6`Z{6S=P$$8$HT;ddIPUrkUr2ASOTXjlx9+7L%l4l!KVLLdzO-r9XSFlY;uj58kaL z?*lViZ;9tiAFZz_6POx4)Ez8{nr4l<>^$6ChE2Q9o_-OL4`8U!cj}i1h6ug9@rO+| zWV`X)ixkW<@+grXyRRFF#MY_K4e%QinXkoULaS4xUQFKA=G^~E4?o%5nkRDKlH>=c zYeRfHpkK$nk4?w<`F^K3nN{~*UnR^|L>zi>Ac$3W$^6z%HrcJ1KnYLal<-Qav^LoaonA7 z4*1D7L34~+h%nhwL-(8Ij}Tcb`tYm^{|yKOZ4`RX^{S0%ei+uP$z!SlS%?|~j;5!UBF=W1v z$qN>He{}TAsfX_M%rMvFJv6x z!2IaG`labXciVywleq#SS{#5DSG zVc2wrCi@(EPq{V*;uit^(DTs~6$RWK@;d2rO?1`;3tpWB}QYZsZzE&Nxi6`y?epXYM% zg~s^PRNHoV*SyCcgWrVz(x%{JI_LT+fW2tVDT_c5tgBYqze^x=2dm+qXL>1nIZ7>4 zXcGdM8@@3vct2#X|Kc6-@K?3rC4(0>pB_&d$Cfp}d6)M*Gu^dN&YWOXC!+YO`7w;$HvPiRDOB@hS7`(XMazl>3KDpxxVeFAdVk-0;#8RAm6THL3uTzC z=$`*jVOkwcbsKglx^-{*&y$4z+x9M3{NMJ>`^}!nANKUrGgnU*BXBcPeZ$w4EcxKe z?qNkyU#4szLz~Pd2OzZq$Re6^2IlkLCR^M6PU-z^Qf z22R)fUS%*|b>D;_h&9Ejv^jnjlX+oZZvH|HSY=T1o7tSkIw)3zy>9P=S0W7NaA@6E zGR7N}9VSIb{VJ8G9ACx4jb3~nd+ROwb zGZpLwq}a!M#aW16J?bFgD>mW-v~AxOWO$nG87lmFk_`c30m+3nvKGvHLn^qG6$eNL)9hwzroYGkGy#GEtDy)6eRGJWGD)n9$040dBd zL2N$`&+smuhssWk6Di>tAKpA4nbsww4`}n|g_?tagtEXyce(*yYab?`IdKVkx0iDx3(hbpw>&MK0nvf3n`?9A(T2DijV!$LAtw zAK}mX+17Gdo@4jhor08~8Y9BV#n4hoeQK^Fg0Dd*BvRC@%_2f{cSR|hb85RI;@mXx zGi7AuT*x`6%B}f0`my|e(Hfm2_NT0~#DDC9`kWez`Ntw*%U{9gkNiOa&Lsv=vZ~Q} zI{*9b5dj8ov%*F_O2n!hU&l31BbnK`sjc|IQhU~Q*77??WXstZ;;NXgw_*aC!E(NU zblp>aM8R#)=&{vuOj!_v9p+eWzjNRT5)-?()!Jv2i-$cFx+`h%F@*lpYa8#izNwdP zSajX&QadZH6X|986tVSesEAogEIME{vWMR+V>Il@%_d5e6e(lDZPnfq$B#!${I>%l z0FKko!sV2jU!Nrh{D;~6za5a2Rv#b#*O23@fZGpvD1J|7EBQmPgfnL1Xv07&?5dYO zdEEnnfr#rKnAb+Js~TDAzPjkuvsSlbp_H~&!B0^!6)@~A^p|<%u~*i}YBszJ_s6{H z;OL`SP&a1T7oDcN*{jKJAP}Sw{E!O2%G79Z{IX~ z-(r&B95fHN>#Ip&GxW^?Eq;pOoxaD;12>^JT3H)=LZ1Vvm%?b z+Lu6^eYR;fo_VU@)(^O~zi&y+tCHE?&X1I;tFnp_1$CW`Wm!xak~O1HU5F3oTIKmg z5M5nhok%utz=1FfsGLtU<=eGMF~g|vY1n3!3?!RB59djrf6GjraX)J{(eL||LP*Ag z_g(x-E|{@?_`TjD_GF=BLHq25)GC3Hi0hZ_pT*mfWJe{6E5Hgf3%XY2NHE-AMByq2 z4)V6Ox*^Yoq?w7O0}#&o(%O}QF= z83c9wdMkk&-~BXlH}i|eQXm~ z2X2x|yp88nPXcO-GgKdNen^vk06y$|Q`DQy;(Vr|W+xk&g+U0N5XUW7WYHw6eeC)) zQ<0A^r8zaedD@)8!rc~ys?Cl3apOaL0oB1*F2OYTw{%1%>+u6AK1bF#w-)<#X0abi zzWj|H`Cm%=7MNA@a;3^(fo7n>HRC3z&gNI5<9IK){CtoYG|gGgpk9EX-Ie(j{N z6O%osU)R_k5kQMUwU9ZR6VErO@~$p3erm z2d%nL;Y#2`3sCIYCkhmHYznl;hoPbH!|kWCq; z9Cdc?f0lc8Koz+4A2Y}PLm|w^)7iW|{M5`1LkoybJPHv85e6BYg+I-0|Raj!|>BoY8kK(fC;GjX7YO%3O> z<|JY2I4c=y^j?hU3>@3ffUj{l)7Tt9#6b8#>eXC3f}Jdg&BzVa9{;Zh?%E#b+K3^% zJE`%LtKvnc_Y04#D1|KnM;I}bD+9IhvK*A+DVI^(ykIdzW36=(jA|zGmMX%q0cK@L z?BSE+txz;XN@TLu)yDIy91QM*-rSk*%^n->*#45_s4P|d9v*NVv4RRq@G2O#$;qu{ww?o_)WgLB2u3ZQxUXay zChIa-89in9$c#<`mBhvvWlHD89E}m1+*XJmUNAU1*?%Vgr-5*y6PEwKcLHGkfWKz` zd2}VGZEGa*Ob5A)3m+g(deS)Ay@OInx)w+(T=JlJAntquhpF)(pL~Hts@nQdh2uM6 zLikjD)n3+ciMgop$=gj1dTt!G4NbMMTn^r>e&*dvvt@il+Z^^TePuIgu4IE{Up}K6 z2_I96)e;<3%-04oPFu#OY|QrroW-+0ziAfSrr~7Kf!>+9R!5fl#OyB>p5?A4PNc*5BgH8l4G6VqJpQA8oEwaK11Y?&Nu!KrYS9M+l)K6bwpSC*MCWBLy zPx#GyZh{_rfCPV=O~m+v|+$D)%MIn52PCH+6l3(JZgFB>h!3O5sol#gQLiS zC`SIEBIbGUC$TUspbO~#8&T#BkuyK=)m$^2;o`4*U_$C&@0zFhO5w1gpeN~^)ob+f z#-c1-Tw9*!@b#kT6PXA551DVR=;SEgWVz?nXY=XV|4otcfuzb67qW@)bf$9Oo=pJY zdq|C&pKkcSy*JNs4|MkOxpx#%QL)c{OaSsJ5-D#>#0K`F@}%scW5#%J6R)B{N~z8#`s>`?VG6M z1J;LNcS1WHC}sC{=2kIVd#!yjQ21o6%X|ktFlys_vNeF|%!GDM4H_G&aN(mUiK{&Z zWbvXe>A4Ya+jT;M|nOhMu%pyLI;-8F=OuJ7Nt$5bUfp$E_rkL#=* z&Gy*bU|$~B6_@Xr54gTxIj?)A!`?b3mZE4OUc)%p`X9{~EszMP|2vcHX*RZ2vHcb7 z_5+aLAU!b!_ios4f+CtB$Zu1FUZTq%v&GgywNetZ*Mt;sjHmd;H=iPC6r)IUxZjy* zp0U5yk68Is!2)N;SE_z7Kxl#sEO)Q@0S77x>`-&fQkx0~=Iu{YlcAq1$#C@zjQ zw-UoOEnYtD2G1o@@`3R2`*&d`dfN>iF_#|3?PG$ml-Zjj$eoo&I8iDFEf=$nb%!}` zH$Znu`qH(vk=AJRR>?@V5*GB5M2?VGn$r+0qLLG69!@YDwC1%;|GPMaEssSQ54Ag2 zIQNMjj@HFIfc0!1QxfZG^JT?+;eFVEsBHs|F(xH0KbqmRD~o})s611I1N{q>!VQ}t zPt`;O<3!=3+db@14Vi*ieYM@z6l+qR;aCgt(@4Hsp1PzU6em+i>8!*y<3jAg!0K**2OWB3&)+wW z&CSB#m6Q-ycYe7+5J9WB8lOvW_(lG8-TRd?!BsWm-=8dGPm99sKh>Z9diX>B`;!IR z>J~?L0-1lwo$%KM?-GRq-)0C>6{@9Oybo?&)#L9=x(@#(JfotBQTti7FNn}fU-}Wl z@h_hLh+$#iePe3n;0Qb9Z!mx*@b@k2MTW~} z31B!AA36ntM%IyGPwd`qm}B6pH4sZr3Kh7g!7DRdshJjDs*%{uZN~5RxGc6{b zLZSbu*7qOA!b9huYJKFm(2K4fc>l{}Dz%+@!1V{)Z0~on z8{r4Bd>X6Td;>+7LA>!o?t;BAXs#nKi^SS+g=BpId+9|Ry*Fmv|1ZAvUwBvADzM{< zziL(ngZSs}1Og|^!%;WQ4+zAis{kT=xi2M<7YEhW@G`ZL8en2#&T*t9(3+#N%S9lO3L7F-J!6TxF0_dP$%^Sn}~{3_&rY`RX!X zqoN}2<8 z8uAcjn`Dn^ozFN=u|S)}VzEe<%|t*Tpe9YBwvrt(Tx6H`L^8rao$xY_@_&%N#PpvD z*Alq;KNBvl7~c4+7m&u;?{#QIHEn}{T&XLP^MxAs zr2xrygyt{b%riD?H%QO+OTlPc6TFoU)pNQQT43j~=pGJWcCHZ|RZE{0@8tJ8PHY+PMbmPxC{KZ7DlUQ15p7&GLrP18~5c zCAEsTc}A&YB8*Guvq^Z|{Ujp~zrz5_d!@OCI`^==D*3jUV3qtQ z4eJ8);r#NyKPu?}On~BlU(#v(#<z1aY}drGH*Dmiy*)beLX&deHf6 zO&JdT^NmRo|D~m+@_Qp9RLP}hPPcLTF?6MEFdMU4SBsXY+i7+LV3VFCo;DoMXsijJ zuy*x5evJ6eTw5fg?bmN@WL4Cd}xu z`!bR32Y=SH;3b*(^$vU>ufB{7g1=d*i62oJd$7OU2ME-FvmdA#+LPeN$Pkzxv7 zyu+HsAOj9=>WoMH6WAsG{ReRw%pdc|;7oU}%o|Cr|I?Rg;4%^_Gzg_w+!lh&76UDj zlDW8aH;yojeM46O)r<(_m2f@%gAa3h@Kh)zh>a^tZiZu(Bvl?G-jx*@zIl8#t?2aMl(IGbO^grJ$%pU4fbb^*!bWiZ?k;s7d(zWoD&5A2(nc(1f2W*p+}3TB zlGu`O@GMIwcFcytPGsu=h3Xk>iy&vn?topktS+xalK1y(RVnfyr{rm8%F|>+3D-Fc zWfiHN4LL`d_nGUo9)R&VxiB=@J^;S7xqK6Pww2Nk?a~S9ob^b!hDC+V%R|mcQ$<1X zN#<`-aD29k_>#G+9t>@;?}YSF7i2>Bj-&_J=?IUA;uYx~Y;KrX2WjhL%nvv{w3c8{|@E2yoz;VGkf&A##dtCWJ zO2>4WD*EI4|I{=~1TOt$d1BPGa3C?spIHnUdQdE8-~8-de8ex<*!n{&r2m&z$P3Sk z0RKb{r;oVU={C{SfBkUvO|fnkBLUy-QCR`JOyzvFOG1_JK9?Gv1y8)MvbdR{^Cp75 z$NlchSN5Oe&)45R&0^p&aG*M4*<_vPtLcw_*RD~4ftE1E5X}#sbF&?>5RqsNquKfQ zDIkqrjU#cpE(J-rmd`WPN-T0xkIEj9?ft}g4YH`PIYJZ7#@H3|aWgrc5vf_Lnz;vQ z2~_uz+9hhP31{r-N?yqyfakPPf4eLis8jglc(It3A(ua z_l`CL!?;;^bAq^U+gs@J=Ly7JE7Vn)qpW}TIWLkl(O{Yq&UP3o%GCStZ1nJ@qF8nV z0%9Oc;qpURg7XEhS-ndgJiO7XCzd_3es?@?$LiP~g;MGaF6M@eCK`>zI@VQj`6^YH zBq0pmlb8oRs%y8+a`f+B&PEWX>1PHgDGKQSlPEVFxbmOab|L$-PAS;C{6r@Y5AXh| zG??_F_)n!l+GD(z7mKtL*aIt!eSm5>8>Bn!^p3>TtE2$ zKiqDG;9-8vP#*&yLeoUyhns;A=;RiWb4}&)fM>ERadR*B_BR-KwQPPZW zbvg({R9c4lwH?Oko>sQbP+*dCE?Uy*nZ=D8gn|Q`(JJT5^u?Vq=IKEL)iYXBw{Dj@ZbVViZW zss(P?<~SJ^NB18qszwtLVJ>&SUhIk#v3gH;v^FiAe&Ed;D{wZ`jfS>6g`@@fwzrasn569L6-#7wF=t zfCc+vN&lA>&tolcR9x?SUErG3<2l{newM!`ph$dfbm-3XwU5R87~CjyL;Rct_;9bH z)^+mj*}tmib;hfHSKppi&x|XfnX$*vxnWjJ3P=p29Leb1-^y^N`O5Ytsqe|Fy|vz$ z-PL>#=RSyED+C}cj+ft{zLf*VKaIoyQf_HI1aOSvRNkk;&kfMl42ydRpf@`q+edl1 z3TdvVoj87~be`S0eSEV>O7?@uu=%44)5=6ng{e=FDgwE}__vAA`mg)iO{&S9c^m7k zrnO+2pmm*uS7EwDn%DocY009Ns?hI%#PJl6oclpxUiaVcP-Jk`r=Dbaiov#%PX|_( z@{|Ka<+F&7UOfwEJgc14a7YHcPZrgetd!d1Ks*(-VbKuSjH!UY5tyGSixDKZyVu!C zi9=brqk2UxCD7N`tByAf`XO5D=&jBwi~b5YRg&;Zk)3fF^j?{bObsfgj+A!0OgU8gpI^PFLe2FulIa}Ht^u``QWO#hB3F%Fkd0x~3}`r5o0fmtED>>U zuSG*sO5Enu)}L&d6DhFlbjE`pak{)IK-#CA9uucnZuizohb}1a5_7?EmqII5g!=gS z;(qTiqqEu5=i_wEF;M7S(@A@EnJfglIahFQuv|7lez9Xu*Pu`b#zQc#_(AN2T%s%% z{Yr7z`CY6Za0r+yqq19)-h}X}(9Kp1CSgl3QeNh^pGyor?d|aI*blzO`d8N*@2mJn z3uk{5rrKYbE_G&AaB2{a4MPL~;O2%}8*l7OE93|7oc=HG$pTm?H`ryoCfoXc(24%F zD~dtGe+N@Pe;4=Pzl#$A$HStH50{ggVffT687kC>bSWqCrdLv%)5sXexd^($miEX5 zQO@h=1F{8`KsMF9t6Os%0GvoJ#79tPt~p-dCdY@8PyE^$#M)_|au@jkZ$+;&yL+vt zZXhXc=5w}+TMK~u9B%F`N4LOgZgr6B>ny4SL+3DupAz|7U>1vWRYo;Ay8iH`UenRN z3fXi&U=&2g>11BxQ&7i#cy_j|%lDi<-^xIKbA>xK3&Ww+L)$z&--l0$St?h{ssrEc%}|lpI$lb~Z@@IGFx%&6My9d~^e&U|I8IwX+?us1}=^6kUM(&sl_W z4E{Cu0dY(9R>`yuDsO&IhRMS~*iRC}b=$?S6I9)$YU;8)hh z@1hWBbKt~`$bjR)poBXN*?Z!%3#%^QH^T-N`5x3_%@-T*9EqMpfT)Q5Dyl*=ovF9< zIsDV)iZ}zQwnH*0fzekvUTiYvK_TR__QO!l!G#_L%4!8U(3zZC1#gbaRi|sf*2fEa3bXb2G+tj zk@ApK-T#11&`*0O|5CfE!RMrv9nLU6I`ZYZI2HwPU{X zy)jVa>KsSb?aeZ1jo;MK{~_%y+@jpO@8O#+X^;}>76uGL1%{Ar1Vp4eBqSt+8A4LJ zyE|1{8YK--P(WfRK|pC3keT_-pgu=E=RD8%df#jQgSq!-t+n@Fd+)`SLvqvSg_hQ; zhhecSonrcNuMrks!XP!4sKZ@?g~P9_V#uo5wj-8S4};R&&HcG{i2FyB9jgCB?$pXC z537NSAYQ!ZggM;@C{94rm_IopDBD{acYyW&wTP545WZB1S^27{R2-HU3Jk{mC1t#6DTsg(OKgGn;&(L9b(E#Y;k1e~su5iF0ploPD=cWy z(!)5Zo8*u5HIeYKV!0K6IyHr+%kxW#G7jmD>~DLJP}_>e z@no;g;H)IOV;{eFq(tzIf^=8M!PF$Y4y*c#Hn?WD5641Ri*_Fl{QWj!0zNLQG9X2T z%UV5a^G3R9_O^hP@x%A?HAQ-OPWC-y!-kLhxQ!fZhyLw0V)H|KieJGN0C@h&UwB>5 zbO!^XcPX1A9)WeN&azvD$I-|{P-BVX)>&Kj#j+A1akf@IpxHAZ>CE88#adITfC=%V z6%$pU6aN#g;YI7AWHFbilYJf=6{Bmxvo}BQl4}(cC=$!Bt&k5TiBzh1O1L^IuqZ&_ z6!2m0u{d+fsXKn6Ynru!->nCmr`&r~4e z65Bn`N$A84cKlz$FHClWWHTw7)7@?3;;t95qgd zXv~e1C^doG#@DVCjut616(Ky)_?>YJYbjf)sdl{L?xk5r0Midbe&siLbg9~;-3iCC z?M0C*PG8!kxhi7#nnI6j`t)11R*xG-z=oloAcH0KySi6jla*3IP!?*)$X(`VO`B0t zoS(80&zct|8#s{haGsX!6>Vls&Ya`aTGIetorkXxe4bIA0n@0X?Xb2lUsvf+t!&7! z=evrUi&c3vUmhGaVL7)2zz+B zWj>U_xp^x%gsLEPK3uGQtB-`IZ3`PhoUpneuhs%xcqG^Xw=nTY7hhypaa+oY2C&U~ z9`Wfb-HZmul+7p9de9gq%y6Bh)IHYwoE4-gTV8y_(D}1 zEENnhbhY3n*h+CN>r*w@Q?f>E4f1&3oY{@a^LXDf8hfG_VyAeCsOeMTL&gjft2Sj7 z3oFOl#7j4`^(y!>;tU_cN;wo%e!q39^vkVNN0+XRwe>2@lwDmr(t+{ub4I@lPhRj> zG6)Y)Fg<$OubqU;3I^^8He}c@3;uHJv=c$~w@m@y9~loT&VR^)=nt}J^|bwH z-5%B(erYy%=)>KG{y4U?&gWWo%quk*jApCe(*e?CD7D$4mU;sg010gE53|_6t1I?Z z$et3Jji$A5?;wpA1&mBRNeEm;E!@JSZ@OHUYHN9JT7@`?|Jv57@QDHfhj-63?Qb$6 zuuv1T2m<)-8mc%7h&h5zt=>9*|c+hY~i9iKHb8VL64gGSut)r zQACGSzEWaMoAqAK+fsJhi%G{+gv`X`-`ApMV3%ziNvx_h;4+IvEMx>)hgGh zC)wAuON{|%)s*qpWT6jj6q^dTZCkw506su(4r_WWSLjbp6{E!FX2EUB?0A+?V&~Hv zk!zL4pJ~9s<#|j|Ic*7U4#nxD$aqB5=ITV1>G~(9^;kim(J6LT{m#YbEco$@cp1-{ z2j2pDa!pv^mfiJ-bODlAjm++Ez@DN|t@p>#oNApR(%pV4WM(-+3 zqk+q{j|B9|d;wGjeE@(`xOCjg+ciQGs&4wJ)Xgzj8h% zJwPE@w=Cs2UXt{2g+o$TTL}iZIzFw*HoZSKl}pnYbF6K}7ub+_eD<1?Y71l1#^V?E z0~jIljmpf2%~DJF?OXY3nJz|og~n6hdL_Dap=Fhd;AQJZi7K@o_T!H=&#qOI47BY{ zac;IA&EjIz&W=l`=wk=r@8>#b1c$ebm?ST3Kc_Kp+I~TATeaPH?*7JhKWo^9odM2N zr=3CG(yE=I3r!n4!@~VlHw)hW{nCu}Y3Q~9K%G6TO9V&kuE=lA#!wQF!viG;OgWGF z>>1^R@4k6-=;n8(2wM(u^0}Taki#@P*h_Q3!9yo6wu(jyg+PpY$|dtUMHh2xk{I1T zD_3Yem2D>mD&N;$kM_k=)#n+7paSyMuQhIH6LnOruRCq95Yv$leWK+WLku(KwFU7)*dW(y=XXOeCOvB@UUtI<1@nKKm7 z?vF%Kb9t*$6YP0|aRjBtG6X<&7|^_Rlf#in6t;+UhLfR4uBo{@p8?d2ZTf(FoEr^` z*C&A(-Ixh2M_r$z5#>Yi&m7b`DU%fGz|{tAhnu0F7t`+Dv*%Iue*M`dR{OwyM3o;~ zm@b&O{94w?~imZ-1SH0nP{8t)Gl=M z$OVJ4H)0}w>^fX~-fWLXSz#B6HUiAX;Y-XJn2p175d3^uPnGeq!AiZSU$a_oRvKgp zx`;@a!Jw4*@|_iulTVks{Ah+!tlxT8&m>xquS$-vewpcHjhJ<;I|LGqbg=|p(|c{0 z;74)zah}{NTN+L)Z+-RBXW`SEY&mHsWk#_5Xn|(oYqjIt^$(c_rlIOa2MaazrXTyV zKJGiefSf@dC4UohH3H)3!~zcc$${?NOwn7ZEw`C~BBR(1%WnpqvvMdD8eFd*Z9!1* zBlDHm*p>0oOIId$Vg*-ThiRtfdRF+Id@-jnx&hPv0cD0e4fSIn$aG+?@3!(~)*Gma zU`jkZ@sW9K+KVfpOxDfwvOe%5syi5DvxuUlJRFw1bj&?RBA$be@LS~yRKj3<3Cf!_ zGrO_z3IznTF+6P}OVXFHwP3L_*wkt_YA6YqAmUdeEDg+ZZgR!lI_NV^RL|H7GL?$C z%q?s!6qdtO{b)5*fbDoJ!}$cUGb|YJ{>pMd;SG;`#ju@~k$9ZtBOT316V6*jNv%n% zUIC;P=EaJ}YgDXRm~^j`O^#nThsqZt4_n`rkaw%BtzqP0rPC7y&vo?}6V8#EuBp>R z@S;;4-}2SC&;zU03r|F?2YKp629>|4zBd_8%P*90ENoDQP&Lw?(@b(*ZR(-vwo7_0hN@2c18&(!3s%y1gad&1*QJk*kGX=&-&p%4ShCg@p-gNS8lnIQDptSz1TA@l8;~4 zr;Ie``us2Nbj`GNBiS59y}zPGw6TWt1>- zogB*YiL`{kmC@o`L_J895CZKspY8|{F_oTqtpcN%>B03Uy=Cx`T+fMEuHAZ>drAAVHJoTa6-C@d<7tu*XL?rPl<^pqyiw#HTio3?)=xw> zXRKfeLRaLO{k)X;iW#=6kH$bXYDYax&L_9B9h#cAKatct=cCNHSFvZU)fE z?(wre{uF6fU2)t>Rr?&xc{w7H^3ikR(@NEd>c3qNdLMG z0aLQ2Wc*7Hif`w>i0Y)#nI6!Fy@+MJ7Ws&9XJH_bj54L3SA=sY6oQY&fo;p-NcOug zaLv|P2U5WVNa3=JW!iu!6_tli{h|h-iFM+@@4BYMa?=#Ky}zjWp$2fa7Q98Zz#=WZ zani*i_i3 zuh5wqwKn^&ajvM`xRfdQWcWd)QXB6G(9T`jcM?bFID?Q@Ut>CneUFh`VF&N4!-!8b3 zNo~J29YMjNmUcau7|-lE8sy~?Ul*AHn0EeGzJK+Nez(eeNR*h_2@x>hooRKqTv(>R z&-VI}pXV%>chq2DF99R27kr(rTUrc@rOFHtNi>RLMrUpW(AoiitZ9cJ@yBB?V@MRH zJ+@4O-<}b+jhRTd=;xVOU6$<`|5pMD57f}EU``%gEzg-f7yi3}o4X7G)K zMyIy1=jsM2vw=}9lKDaJ%)4^UPG{tNtv=7|HgWgG7?&tOr^yyR{5paT)iN>({U)=g z$^_!{4r0LmqHhw2vwt`Tu~`)x38p|U5JN$2^Px`kvf?9x&RXywfwPhH+L2_xrZAeE(oFZREy1gfUGGTiwMl2q_Q5o?z!YFQv71zbk=Hsmlp zsx9+zfsb(zC%tOM`p+|Qhq%amqw&|dovduc2oAchEYe<-jwgCU(dlABZ(k$kRd@R7 z@-*ALsV8^)y7EoK&(e$6-Fn9$Q;GlO3cph{It0AukG&MdYNiSA>rII%XP>_X1qT~+hmYTr)#xon<%W5pK`s-6z4#G9CWvRoeL(P!l~OmuUROQ zh&;KmH&>e@c$YsrV~1C>;+8RHuzOvy#IQHv0wI(*5#KgdiFWvaBQa2szCN4eNe~ur zKOI_(F*KCZBLAsT`oJ5cmzl!%hIx$~Z2Kwz3yVqcc;0vfjK#%BdVzi|B#qhzKRx~O z`UelgxxQ#uuA8~Sp0NSmU^BPpj(EsEzUeC%H-?h*2-ImYNAqabQ5XRP8^xx+Lp_{r z_w_ED`l4ZrZBqB@@WLpI8d>6Sx2Z0Dmi5JMU+ZAVbx&GjJfHDob4LWw#R5!hD{8jQ zU(N)p9Z|3a(P7iF21~p0QUuf3q`i~wIGJY+$C(T&2$w*>ddY>6Fs3M(>b(V05q`xn zFmlwJN>`)@4g*skr@f65?oi5&=SI98BCb^iFNd+lf3a+)eV*(Q@8`7&!bFJ#f~bvu3(G1))8Qru3l|^z0gN4 zda_&TzPUtkml8teKOehx4rJ0KGGa2kM)=sza@MchoG4W{&&O(~dXy)oafSd^^z0rwh|O*W`Mv zxIEkk*^5PdJE7bs0KE1ZO`%No1oW(gq62Xd93+LgaTlOv{xD>6CyWK+;Y}yZ8cLYv zCFN_z!~4u%jS>(5qz$LGff~e|dY?0balu5c#_UeFWAS*vWWm zIg|ARPOR%sr%0gCh%-lzi zGll0*cGsP!>y)S=_2YETlSxo^o(lM4*!w4x2&(Cxktd{GxL**|g1+YR-Fc*p$jsa3 zM)lE?7HaVR53$lsE=^~@xm@X$mbynk0KNTnqNDS!UX2>P&)SF1V1ve8#SQ0GTus!m zJI`V6r|0IDT4zn;&TQ*Azie|0{HEx-_Kd3|4YKee^W`?zPBE%LO19&O?XoFkmzEAD zyFYH*Ljs(ih-0d0X3H1AzZ$M(-bACH|M2^y!d1Ejncs77eDu50{r)FAVT0@b4>Wjj z5d`q4q5*ElL^kz;<b4mLyYWX%+zj@-k_~a2Ww0ipAJn7$QP|S+;yX%XE zyKn>mvNINSwl>lSL6H1c$m1s8ed&BOpd;^ZUb#E**!pQB;_CguB*2wt>D9+3rThVA zs;0Zwmv%x2qQnS}9vrL9cSjhx-1h(W26Ev15w+VvqeTUd7)kuqvw_*~wYxs|rn3{( z+f5(JoMvlg;dR0`W3otVp9Tt}OoJ&1+;xLPhY|GLOY4j6KlOBU}$;n3e;H2AXcu=KV5>I}Tw}{~@Nq49EQLjciW|{F)!kGp9 zQc&{(Uq$>`yuAv`U+pOu0U^NlpYM)zK>$`+kJHGYMcGe>V2ONi3O82o^4s``g$iPq4+0=LffnRW-0Fe$X9R<;KtzcThkZNI6F z=Wpo~Y;NIXHOrNz=IkAL4}KT@1zV+_Q7bP>dWX0{Wb1KvlR3*Aj+xQC4FItko)$c@ z8a!pSnl|{Lv{bjTzBGuFcB%8^Ri)LBUvSkAzt$BUnLqk-f6Ty zys{pMZ#+JPI3n+ACw6`{n=wpRI$)#@oTHe|t2Iw#AQGj&p4L5&k)Sq8R!^2=)P(+J zwaI_xF7HmEQZAdl%^M8)=#HATsK zh*#q2ab38UoTh^!{dsmu3wD$D!2xHvY$4)DR(4<@0hd(0qIFJi{M4-Ag>~!pI(A5PN=Qgu-BAczraga;hPz8f~$(l+fnkoBC z9?U*jBe7|Nsnqey4)e&SP5nxeyZ~mZ0b?P|{r*Z}=*LNMT9J?oL9wemS{;EskDw(% zCN#%eC;*34CVOzou3S@Qu9SM5t`wz5w2-8#Rm+ktQ&G}|m*w02EV)wd$Q`UCI}KCm zgyU(6UGmW)FQnD;h@Cgg#SjhsA30b6lGQ=g;RT%?ItSy>GJg?gZ(3%(`kwW6i=9t} zp5p?ZCMNardO^J`n55=&K8xWGkLoqx?k{nl@;&ag*vi#$*hJBsW6_P7h{+14TTP^+h^CDt?F$~4Bk2Nw!ikiHX z@c{Me!+B%G(TBM^oy|XsYFmPXIL`;;38#cx9)3VVp5oT`5p?aRt60K+?Hif(WSml*OB-JrZr9qM|DuP+ypwch_ zo1lVbebe;g6Whg+y2L~NzE$-zVZ@(G! z7ixOFq<@g}uG$oZdc8Y)u=aXz^Q56jCb6HouCn&%DBHYu*N6Lcr*6jsNGS#?Qv_$W;ymj_T)`HN`47jM=#kMY{m^QB(!czX|I~cg?^p1 zC{RRh4;x!D+pC&d5gqj=oSBRpfk+W)+7?kkV)JlP5>FFAiyMlLGTF4v_f%C)y%mVV45TxF+Tx1pCvHbs)lWP4 zSv7KdJ90j8PFhzQEg|x!&O9l7 zya*1Yf%u#ojN?$mv?D#d$-9(O1r#Ga#4mBbNEMa5M`L)&_C<=2DfT{t8&4vRA{o88 zaMCI`M0itMBySH$G}7Rkz3Q9i7+hjVeXq?R-|<7GIh*qpzdV_#aMFp3UB_n^0t21b z71~IkHM1e=zn5D?RPbXsR5gZip;&+_sc!QFjJynnpvC+#%j-_m*K`g4)|$_6D~W1z zW{w5FOa!aHmbzoKH9VX6=bQg?*T=W*oY~%}=tHt;uaTLDE=>BlIS0q1&HlNu(s1&P z9SsyvcL~CU+$NhWog>GgvKeLxU@c>DAUcBel2ahfCW6J}mC}?bWt&;e$z|brCV$}o zc2Md#2!jJrLQc)*w;|2jJ<&XABDYl0I8YE}QwG-$2Vrm^0O6?`1xniv5|Ay?(@moM z)9a87`2U<77_j~ynE=a=Vj%hp?Rme+>r585^jhM4TB07yLEhLl_|&#L2OKQ*fUmwx zJp&(tMW2p-rOE1F7@Usl9r|;15r}ukwUy(wOnB$VR2l{J>g@Q~nc?fE9~({H4K+-> zt(vKJ-_<#}zo|dxr?O9+@&f|0m|uF&qalD%)e8*)uLskxga`jcjnp!AeMeL$8DD+%p^lVUib2%aPTGDdnCRumtJN{H^uDa>@7cUsr_~e0ln0XQJ~44 zY}3wlbNrajohQD>KQjT;->_gc57KoVnE!q<;O7$B~T#)t>PG`u%LpfGs*Mh#3Fq8=W|Zm4kfWbx^ibTbmXlP zf!aj0bX`Q7S4kcT6)guw>QvZ;Kf379L&DZNAIVhIO2wYj1Y;GO-M=*-Y4ga9|e zUp3YF4&b!D_GQZIUoyiVZL=T-vmfeI^CFfR0K6Ir;Y$*FwUjD728AFJ1;4QvOh8aG zPpiL)I=44E?jpaWHo46`pF7Wz)8^d0_$o^AjrzSmw)3D^G&@diMhm^FG&Mzv-aBBO z>lHgb2R%93&R>*1x`PTpA`z*%Zuop$F!?X_y+$_)yl|DMpqx8q!PG5c6s+Ug7+Kx7F#ng2;oMnVXZKdnNoLn--02d)Sh6FuVzoNPkqz_;|4f>}|CwDO!{&G; zF3l#=KI#X+Sfddp>b%OrsnVzs$|43A^CHno<9Uu*UtDKfer1i#gs(;AGYrC_F_Vv%$eRZTHu6MvY_>`4M)z zCJfwqc|CVj8XV-)I)Q16doU#feE#Z4+R7&vl5q&moHMiVs4^O&SwT-pR9$ z;6s42$OorMo>zFinj0YFbY^W9(dvl65_Ua}JQ@Z%&Y+g3O3!otEid|^4~;Ib%mzZ# z?Yqz*L)3+b;uvbP-DyG>9=~UP51kvSXgEq-s~>V6wozG{!>lj zACL$ZOpU;0Q$S5zp8q{BTKFA_Ed0OF(r^HScf>gWG2*1>a#n5O<}v4-35sIn&wf`W zO9^VCpO+G~4!k(xMzsx=<0xVbW5FV_ttbwuj&@0E`iUcovu7lQ14>t>+h0X2zx2;MN_l8|MYIrI!VtVPRk@QfK#O{|L zAa|OAv!G1k+K>M!)TBO{Kv}DA~>G5Y;v z(s&HwXy9Fte17s`u$>4m=WXzD|IGw>DH>`$^{*d{Z{LxChk+qlMDr;f-h+7{ib)U7 zU}?2xE+9ZDDcOGbfp|-6)^7|e~{W&hk6;0;sjIPfaXn|`gH({@E+*8HV9A}vf|FWdY z%+ogF$Y9H#W_b;xmYD8dw<8A*CJHJZ*Wi-Ccoq9Hvg^ri8);>ov52Y;Aa;Y_ec1d} ztS0Dp#S20Qmia%_&)A@le`v1$a!KQdcl1B=w0~?H(9S=M`R{Ab{(0{MI?GL7b@?~` z3^;Z~hZY(+YR+gsU3yKlH<{hH4CaFK z2o9OaV252Szytd8mxj!i`;71z4-eOUtc;|eps=?3$rn~T4ZqzV3Rnpv1v)5pg~BDj zk$-$ohuT?0`n&BtLppQwrjxge%n-HG zTfZ9%K>xhqY~-(HHaDU=@ug_ewhMk;dUOq2D_t9N4$G4N#52}XJy6VgZQ5|>LoH^V zxZ5CF%S1^99>eF?-;9=Op`|;H7KiMg~tKC8u_5M_6Pbq|5Z~mNHG2 z(Z?qZV{gc>L7Wj!mM_=LrOF#(gV*+| z;$?!5FFufdX(&ZNS95=Hy)PRxhwk=tl9Sb}$t6>P=sxUDgOEk0w|U9IKt>l8mQjw~ z;X6}nv)+<+_ZCmH*MqGQ%+DUVC45MwMolLee*%etUKf?tT)KX+bKRwdL2BA>e{0L9 zKy=d&_HFD4@03w^we9h_z3Ewr|Hm^&GYSoI_YH{uAap@jun}&dFH#}Sp?{^-n?q9~ zpozl;xL&jvD}PVi>Q_clkVuHt(YIetzEzThSqg2>o@|0y3e6C@KND|-(K3_eA>8Rp zkNH-zveK1TvN?$uEaSW#E{{lF8B)fVWqL8do-MocSwF_bS-FoJm#e}y&rz0tq~Os| zxNU4wkReymL;Upe5(hHh^0G$<4EE7b(Y58gEFF=nD`%}9@vmo8-fEN4Oy-G@uHO| z!HVI`Vv#gQB1;2_XHkf6#h9aNv_?ahiJq3or0~1Ft(SPa^b&H&dv?i?dTA&H!b!z! z51M}i2>>RJEyG}MBcW7SDhcB}JuC*7u(1`qHb%m1NssBhMep7Mx`N`PpSjw~7Q5iR z&aR|b>TXcq<)2nP$nWpPM*(P56tR2}M1UQvYG{7@MIo_?Y-Tsd*vD31{KI=_I z0cAx?)=CENI7t+kFh5Z0)e=lb+87@9kdb)rZ&N}^z^&827p?Y`{&y?tL@9K>GoDvB z^m_S|V%4}zlcB7L1THjSVkKQibFpTI2+8p`<@!|+z~gt;I3r=XHXrBitfh)42VIaN z%6`98&vCq#wF$YnMq6bVl|_s^Zt}W2`P7wa^?#dbzTm>awlJ0>? z93&kwYN1M<5q`s_s*IQUu6IB4iPxRmUj8<_Ja;UZ`>g*>r?mPa^7N5LTAhR9saU=9 zClZAxs}o)grwpNTxcT9>_6&`=gQw@dX;>{69=pDfXLG(ajSan?!J&7ug}Sd#IAicF z&xdRgJh#Oy!s+xCI8^T7r5D<w$ zkr;^$rMf&;Xu~~KCU=OKk7DUKaqTDJZbPRFAJo@IScE*L>=_Bm2$4tMNnt&N`b=xB=4YG0@=e9)nIWfM+Y=myls!AVuOr6bJD{1`)tL|& z12|;Nf^PRrB;uQT)B=2c(|rQ+Heg-Z?5YNQSN3rWPDP9oOzpr4#}A->OJAB^mt#}O z$UL#)DST9NqV6e)*kO7boLpQER!7kNT~hhkpUf4Ydpy@=F<|y?!##sKiEy3TOCLrT zLqEUlA||Y}+u-3hL%Q_A_AAQ3XAtA?yrx4YKI1qzV8D3op>tkxMAH%g;}NSs&^rA9 zRv_`Mn*Gw64pBgqJf>>*;p7{&@fzPO-s#-dnlA6Wc2+Kow}xx3=36N%+;_ZhztRG@A1cRYnhnrdnM-8jM4NK@h6>ax-Sh8 z+1$4UmP=~%z38P@KI&-e4{5E0J>}nwHV)Nvu@wxOG&oH$p<|A(FxQ1Q{GrXa1E8av z7yOu4$?4$PR915YG2Z9<}5ifGk%uHiw<-1dnijBT;11iFm|_?xTpKm53^L(*P;+0AH~q|F5@d@>30ngBx$ZkF8%nzU4T_`>cxO zYg%l6bH_1IIGw)w4JMOPE9UxQoo1Q!eZ`zt*y1qFL9#!4xsfkF)br$g#fWa*smg(( z=vT827&w;(#7uc$^+N*z#Q6~|*Iu)-H&zd^3E#P&(?M~w&IRovgEI!tP@Q<}wRr2f z?{lPC1K=vX>Y>2dw)2b-tLSE4iBeJIfCcuq2BsI6Y#m`{Uu-oTLEasm3)*kAe1r!% zbgpA-X;yiNx#`=h<>aP(5hhl!Yo2uboik0QUz&pDOEZfDO4z?Mat?@fHp1EKc{L!= z^_%D0?!#Z-oI+5};a{gpH@0md_y=Y_zQBh>?RDNt-+u=-!3fn+jp+Az-5m;euqq6W zfR-0nB%hSp?u(~-ImnD$)NZbHOu8VMwirgl?wO-tRDpa!t2x=D5dpl8Dlmn*?G z=&MSpRk1^@IL!Ld)4mk> zx$CMhEWPfJ>d8;bv{->3gp~VGt4a+w;=DS;#r^wkIAphxF_JDdk$yIplN%0TzE+Ky z7xkx?>Gr|>TD|>}YLpWnF0(5eU%idM>xoBVQyp~a%Y_+ zX}D_{(-{v`*hL^xexKZPtURBSHP;Z?Jtm{AULGvZqhob$u?M{g0;Lgl<5B?uP=p^b zcn1Z4ia9tkOS+KN)gBM$kn|j=?cLOCGwHdUoyU>mJ z!$ao3>yzJ8AgV?|nTeUt(oPN}g&$Lhx4R@0W>W^|=+u&#RCEOtWQt*bZMK7eXo&3o zP?7FPf&%J<$=H1ThNQ(;@Mu7*?mBSGFif_hN^0G)fjo6Tme{0;ShXy`fUO+$khIv# z6z9B-2;qj35DLq7EJHLy8a}BmUxTfrD*A%8q+doLAx6p>oM*sBq1YXxzS`;fraK(-qCE)FvxAc0 zPI3dns(RvhzbnNin_~__!8HbJz|C#hhuB+PZ^+oPxlX{4d}6NA;$j~JmF9!M{Z?xF z!h~yzqW*B~LHbeKG4aPOUE4nf{EvZra4Nlou6aPvvxx4{a36a&XuKCpRe?3YbZN*x z`c6D+Lk+H>q)!eZ3AA-(K-oyahba1}yw8D2OyL!ftJoGDjm1viNJTEi`rAYlV}?NX zqWWaG?VK=D?FITKxF28>SODkUJ)$AoEs$+-L=3PMkV>#NrfG;qDE=s>E?BT>xO6u! z-VN;1o&Z`4y?Uqzj4`P?Ue_I3KM0g;yFp*ab9M{$p$i@75OR!a2_=t;^i%gW)UKwf z4@sm&7FenU$t`pPD40$@A0R}{tC4vUS_HvYRvWBiurM{vov9JHR(QD5HT|liLr4QQ zR}~J+fFPnSLm3vN1u~Ed8|eCMq)CO71fSy$0Wa^0@g;}8{#xRn_{0gyV$Q%R&Nn)? zRp5dJ^@OWpp(q20TG6mTXrUn6oCko}`&c@l+?WJ0WLHOmX$%VRXwajG9N^1$PEpMZ=?O7A#|?%G}jIoG0?Mv%y?ZOBi^v_T8Kcw=F|#|7)CV7-GY z0_uA9!>r1vl;FkoU|Wyvk}xRo$zB@jtUZhZ+MrUED>LI5HI)CGc9Z}BNwE5wKQW+R z&5fxEoZ;$hI+m6Sjq($nTCz6QF}wW^v0~EWZS4~{u)xyGV;sS3VI~3JjKfO9bqkgF zlEU=ieedb9Pndd6dj$gZ+T(~cpsw^t2}%-vlJA6quLL;F=XtU}O)Yhs+=i?wI{1z` zFk=b%)pWxPieliIFLT%Cl3hv!^gB6@HXD4>j%>$Q=EWjyqQ95b2(j5#Oaf@&nJy$a z37VjA8%*yChpAEQu#+A&F+av|Y!?TtUV^+Qx~4_5SVt77yzCLU!nV*(k<+j#I&jM8 z|1L1Om=bgBY}nm~Bj&H3(baga%~>TuN@Jx^ zyRK|o*a`9Nz^uL^t~tBe@DZ5V!UzhoH!K11TJT(1+ER*Y;eYZbUeYNkN3vfEf)dD> z?hcQ6Kd$Pt+ZhT@{=v%@`slKa=2!SBC!GIT?*unS`jr+kca%-56r;DkgzOUZC|??x zM{wZ!HA$n$-+RzU+Q*?;nWX7pG8lbZ6#uFis#^WLeQPnzWNp`M)~UR1^Q4NdyWxlR zJ;k<>?vVb6oIc1_>xp$pP%bT+TRbAc%E2 zfoq>~X_zako%k$5q-mwJO>xs@rxXBanIq=?gyWJ-!GD&GI*at6(5Y!E4p!Sy?!dMz zIz7(#{qEX-_C@mZo4Tp{>oH0>Ya>3d`6vB~yT*N-*s0k2m_I%_Ed*W7Zu92kccuR0 zDG)$B-ZlSZ(#RV&w}Q*nYF>X=GdOQxx;au`kg=oAGC|OVnMzj&hNe*;H$NC+?wG=V zOOmzgcueze=-F5i1fW>`ErhHj9k&RS()hDRrujKhj_yN-n1ET#Gsuo(=j@sxHda~& zQzREYV-lii?Ln50>HL)8mxzAL*DqrILijz@EggV-swPa_844U$XxHjw0q6dys}e}i z$Vy-Vl?ly&IG<*3urlbr$i<%aX}7D6 z1Nzsr+ZGvJ` zzngSDNaAh*WdzZOb2;RfT*ByDK6hlYwjkw%;}UW#A$RJi&kV<28{FDM!#FnzD5-a2CXs$1hEk|Iy1Op` zjU@9X{g=DL@M2cde4+3KAsmQQ_erjceXBBI-(QW_=>RCYHb|4oGqvc$OYUYf^5A>^ z#G?q@bjp|iFLKvplw=vPsK6(Bgb4ii#Vol(i|2O(L8#CQ$SwOiy3CA?f4iRB#C{4R z&!VMlfo*AW{%!A$7sXMzyvrQb-;B`5+yT==-_ThcuE!jiOb4(-F0uWs@wxLo zOK%T1{fEs1bjVsbr7CV zZ*xs3qXRZwFeH$@D~*@Y;m`p#pQuo6Sj%KmwWMUR>JwH6+we^^XmD-m@xrg}Eg=X} zKyn{1>&~4?MApny{VW3U+uKhVq6SKI+DHicmmF&O&g^y!zAbPwkk42ut~fYi0zLEr zua{2 zPeo`8M-SDPOM6+y79;yH4;?XUt6e5}e_B&dBeRE<;)$FmK0a3frk_7ls^0;GhrKPe zKgWDC22oQ+b+y+M#GOPKvXZz8-q!@AkS^`2buD=^FbHQBLUQi6dV2nO?np48bA|MM z4X0jh%z2AwX}dR5$*?ebkL&4cQA7^c{`w~E^7(@K3A3y$V9lTD9GAFv%Ba+NMQC^| zzSi4!Jk~knGES-o)F7b4_d|XkwQ%*s$YAr;>^iP+%?b*Guj!GUe8NJ9;7DPFtDj9a z1weK{nf^DsnVEZ<&b>%gPO=hnQb_JFx0F!dGJE>(?AwhCFS%+j{K3>34aC7!ON+D} zC3YYBr0T@q@xmgurW{|k!FM^bc}Zoc2nUWowW1CLE+v&|jnb7T=#3{Uuq;xW9kOBB zVFGW@9dLlM(U89rgLkv8X?Svf_awTBXF#%>=CMNCEvg`pZ*Jjz!y$UcWK1aV@9Hh( zN13X*k)2*K={0=W@UgCGlEu7f-qwhy{6U00`+tM(y|782cu_KVuYn~1=hprthP`)y zLo!6iQ;^l0Bl=z=)T%UWOZ;e` zhkbOM)?-@17Uo|dTf0z_=`+KuF7WTsSv+Mdm-;U zzcYXJ?9>k7Uv^vlOfj@lClY=32t~`r9qEZT5*t$n)rCwTHw?i-8tF_X7+nQP#a1yC zUeg6hq8nVODYRNg-OZHGDrO&ZpStJ1zQOzE&U*dUe0J8}I`96){0%24GpbWs#?4tx zlt~)W_s0_nWU#x$Cef{ui4hv>K-KIATvE?D=_MFR4&l^S8#ZQPFyCi+;EoG;Zn)RJ z0@8Mn;>ah~L1$d^PFiWN12YqOjwkpJtGio79RHm72~Z^DCXG0JiLEr! z9sc{$g-Hta3IbZgjbX)ii|;OGSnra^Vx!7R&s$S7P(I8zzS$@9_;-}FyljHSmDf02 zX9g;z*nmk$;8miVO9)9-Ak-22_&^5VBGssmly5y-8nBzEf=Zl$& zK=q?_GOQE74N+FzrloEsVFLUFWcU#X&REV-Ul;i3-vVz) z-3{UA{8Xo4=;eUN!S=7&u%;VNUKM7#{761>&Jn;rQpR?e)jGLR)FXLD9RxA3umfY_ z3Lf{?0TA|$HI}WFr5I=ZqKQ@EK?KHqrm78~@;Bm*WBZlmfvNy#v4 zI(GLgp9o_Hn@OhUJNHkz5U^)5Pwjb|-dwdICq#^6Cm&40cT?7j;=I6Ont%)INqsg9 zr*Q1J0>P-SiTxMw>(9I7JMV;aPTKn!hc+p_h;ogr2=hesZn(@bm!#j73!J!~-Rn_% zQoO7Uu68(q=->tcweLw&`Pozv+0!?TRoQ&}3_f+Lm@bg=*kA7HeoPD3s3Gk1 zHiZ#4fzh4@@hbwZ{5;;)9sF)YPF8fZCj%ypq85*i!g7qDq|@KGe4hBAJoUHvWO6r`!Wt0v1&@pn6^G??PWx}w8`KZqHWH5|vFT1T$AQk< z)opI}9imO`@5q@m&xCghz0cCrazQ-Y4QF<2qrU>Ic~5z!WecGAMh=wssZ0tHO6@}m zbuzSY?oF=;M_#8k#?`(2$mre@`=5GKxLr&2hrO~Vyvf<4@`ZX5x3WJ}Q$R!Cfb7f> zIKutw`V#`!SRTF9k`Dsua+)yp34^EVozbFCAAB)*WOwHK<|;Ty<>|>F%&MqA#va@k ztzoVxq-y);7Y#w@PnyyFS3ZFEep(coP%_ot5Mm^kB*5iaz0*t$YFw^t0k2O#4BXR= zvCFtEC7c0^X7F;5(y9Y`W(ADcstrs7c<#^!Mht`z0i)SkboCEQ%4ZM9*813ByBjh? z)aX4^3$r)eLT~hB9KwPIQR9Z#kl_Z6p|B{ChzkwRjP5AZ?4ryF*UH7I7S3NB*&#fB z2Wp+^h?5iNv~%Aav$fA~lNLREv-m$a%8&H4W+6b4GR@-wug3AU+=s@?;=>Jc<4Z*; z&}i~!IN%9yCXYFLH+2QxdeowUvw7O{d{sw)C-o>dz+KnBqhxD0WWlOW$hiZ@Czpay z3Uv+&N^C;2nq(R8o2C}(*er8t$as=Z`cAG7DnF{lgOY;5;g;Pi0 zrS?<6+Wtp&zjt`IKz>Kr%w-6hM+#h;ZQ-3MaGD8&<5_Dt6Cqpz%avCEKLr)JUEGrg zajy-iInnz3$v=v;v_VWD^5A>%gR4q5x`V#Hu&g?eLVFW0AjCREvU1)#}(%rlEv1qVO)MD3K29j!1eAIWFK3ob%S5*Cgsf!Z@s670A*xZMcls>A7*Snt z{)ieBz&b`OhEU3$>z09*PTuiCgzSNvrqW$;LcU1)uVr{e{rr+s_kKnZdpxTq0iqi6JJ*IrC@{o*3hCnk6IlU zC(_?k{EHjOi^ex@t#gKky8R7-MmnRWNyjE^-?2;^G;#Y;ZWfvXOVi>pb6w+VtM4bE zZ67CXX(88bc^*L{l}Qxyer8B%c~y5=y!s}lXP();B=?ixEea@fLLn34z_QG535u9fYY6zFFNX)pmZ|>!Yz3`QZ7HuTLa2QGx7rR{tco#&3(WHfl?+BTeOmZNc{p9&r2Nb`qLx6=G>Ow|A~ij^Q@H zt9yG|M?fng0whThL24_czpMrZkb@EolOxI67U*joK?2^H11c-nkcp|3jcEB&4fn7w zGfjn@2U=l48iaNUWQ2`6%ew%;%#7L;>&sJqs`^EkrpBI)eRRpA6kRrB3tuYcKBXX) z>DcqPGQ4HByrl|lv15v+$lmkEMMQ<#4@3;?pa(Sl+qF=89P@E)q+s0zm}?Dd;=iN0 z(=*qlw%SdNBd z3XGofA;Pow9zlv)BT!D&Hy|Y_SYgDM+XDy=fYkq$sPjO7IvLi1@5$gvP`!&4w5_44 z`mt)d4E&4BhX`{>x|LAf7^H^XsR-H?_?|r1eN!Y73SlhRA?$B@RO|;>jiqc`5SMC3 ze07`UK-s^M))1jRnkK5*PIJt|$T?L8w1b-5q-;=?qR>5xju!EviVWOs=DHfQ?r@LV z2~Ue9`cat`(_y3swA}%_Y3$64p)F3k4Y4d6t$~ynx&e;p**g3W%T+Yh_WLuk-ZmDn z?1gVWsW!Jeb_jZ%3f~_9QFp${ayKvgFf05BCYKtdjJM0=slC$zT zJ1hBP&Kn~kTegB&bqcwD=P*ZdG=^Etf|0vhw~!_gdfttKTzoBobyl_d>56vJLLzUa z`^~h4{W0qK;al*w_8D-q8;B*IEP`5j!Db69t*fV=JP@Ak0#nSZa?ZpIFCnWJ``ISp zJzG-`+;aWLH%-&`X;sd~oVYLB^21gxNmtM;oIHAMAm9-RtqC~#m;j<2K~so~JI~h< zkhc1?o4?pk#la6TD(#a*I>mUsV~L}-=*`}p9mzR(uemq!a)oZT8DDCYr#k(rlmaN6 zit(gKg!$Ms&1nwPnIM>?WdVO>g1h)5M%aFc(BujBskgdu=V{O-RTR^O>Sqr358I^2 z1My%n)ojqQx_kEw=%`wg7Metha&7*9YU-E{xarC>xmFjv08Gn(Z+r|(GA%B4- zQfP=^%iS6FYKow!?@8A-3dY^;>yVg_+>Wibmx*d}v@&U0k7>Gf)tCy7#fk~O)hx@J zfuh9f8iHf20Nym6M-1c9}@+VIoHOKiFR_#uf_ZCpI1{>s% zA_kFaE+BtU-qNX`2%TewBr*bWY!OTdIvxSxsX732oK*jGYm+KN_3FG)ElBr7W@;Vo z6#d*^)cB>f^$5*H6e|1!rICZJiY7_4Lh;V12Vo(l;(n6(tyS)d>pQ(%kbZ0bt{3S? zCwJ||gq={>LJQI!9Qbd%zF&ln$BjZ47a`SCOA)*AHJp-i4hE9{W|M}-`!yVF*a=Y{ zJem)t*dPYfd8#$}`b%^w7E;g#;o(4%%A|ontLn zH6qfaHYrsEFMCFjK0a!BN_D_XDQJ)i6BiBrk-0meJ*N>AQ---TCBXe<{R>kPJpAjmf`dP0MZZo;Pi((h2r2LT3f%%U&7lyPBnzZ#_dt&g9 zegG_Ch}|kgVKnX%kZM~wu05nqsOYW&@V+yL!Jjtt-yAW*h1Bji zoT7HPX``)rISmX{^Up53CIdWfbnV9rV&2Ma1m4KmkefX; zwOt^4Lgg`wnyV?rn=vVO^=~&{vqs*w>p2GHQKalK{>n~_XcyQSWjyOT z`rR|?S?*@F+HOOnq~v0CrN^IheUWA@*$#8!qqmsdrvGC1AtBD*26ZPmF>pl@BIzHL zV7FsSi}9vW%U`=c?aNxLkO=Ni;I=FMnomXKY5<_#R;)V*qjN=odrCOR5s27Z(BzCT z`@2RMdRd5G$SIV2+*SkpTR7!@#sT-YECo=_bSrzVuw-5Jf^=ctg;{cnCZ5fq^z{_e-@utUo~xa^Q+-1I;* zw-8$CXy;dS*3ft$jCah}(LBL?dCN^tm9M*%=%icIFy13!415c26**M_F09;Ox%FQj-t==Yrd+(J2zslEM+41< zrtYkByo89B0ek=^HX>nmUEM6o9qe0v!sTn6^gIlH`=<9eCj_ z?@4E_d|8&p)r`c!O!2l|$`8kSIG;~(BmlX)JeYF`zxDXb3!?!;va%zxwo2QR^3%sB zu2_W2QT(6hZ`xEV+ltDG<)Nf`!qY}vQS681g%bra66m#t`w{hZ>mRL&A{#wzVkxqM zBhJL2avHO#C)z(xbt&FUB_*X53ardIG&ciUhpp&mVi%*so<5P%ES|DhtubvL=2|=P zdJWZieDNGrOU9YpGjMUYSN~0z42;y=$+L+<$;dx>+L}IRZQ31_+kA}?3uxoT0$GwMgR*Me-*7~*?IOldFh*T(CWsC?2%>H` zOfGKOAmSWB1z{cwON=HO%+LAOd?qpme=tFAZNM0K=0qsvoB!YrXFT|PF`yv#2) z-9#{g2U1<_yO9!kstsJ6y^;F19r1XCRu>=_3T?|G$OVV|9b0BJ7veszyn+Y|1SnWl zH?y)&e{VSu3?>V^CAGQk&1I@ibhm*ms4Y*7LGMryylx36O!}$dB#e~rRa|WSvL@!O zU{Gz#M5hlH<{+27ky9f4j5f#V$Tji(m$*+?c-a%EW!A~gPP-HV4>i2Ib3}hG>Ltww zyNj=7=f`QX#Z-wP;eC*3HGRNH?#1W9-CdlSZZDT=I!@cd-@@qZ#&%=~T{Jpwc;y4T z8lXY-eqpGaU%J|wjLW)20?f=PEnyU!kb9Lp#F%+THLwMv4SBlddWXSvda8M+qz=pX z=${dEph-iUy=QYgU~RCm3PQ8LgRw84L_rXXTx{w|{!b6>$gdP%!e_N7MDsN<-HJG< z`3IR#>@(_N586Et03G3TW|zfLQxG(q7ejEw#Zosu5k*<}i!_W{e0*0(oGdg=mB*;a z5Go7bnf}LHGf;jLlC^Xa_&I_nGr$nNvTBxi^EISu^AcuW4UG@p#IV20x2Vc8V|a0! z1vlk2fmld1y5y*4%lrvpQy7N@0CmxCHooIX*e%6I9mnQ4oN93KT-|TU z!L)YfKm`zszGj}FGC}2_mi7KOJfFjj(b%3c_=jd(K2iM$L|9I;>5C$c?{;Pz-Vi4Q zhy-i>jfkcp4olMxcqL7cn+)3PokA2)=8Z6rAl~3de8cI`qI4e&ssBdU`jRAI5v?Wg zxKt5@!uC}44)}SOW_zDBt{&lj422bizf4MBfsGay8Wj%|nZ2q?_xxVhCx@MY-TT#s zOZulny&LS6g?ujIw*sXvWU)dIJi0og`-6FfVEYsj0(9ce77D;01pt(LT;bu@Q>j0c zajNyeatmp8x*WCUn7|QcfBWwFJa21CgbOd@lFsys5Ho+3s5GkKrhD3@+)66mbd>R2 zg5TFWUfN&f54@h={`D3x>`lu(TdTj>-(v7zLOnDK?9Sn?*Q_wk-hJJbn8DO#A4?Jj zD!}O_d_*mvdo~jT08q&DG&9&2Xp^9+R7ci^$YWi2%H5KQO}sBGKyV+V<}FCrJFR;* zpe#VcUp`PiP&^w7Ry~~M0Su$Djy@&SO5}I&E3Rrpf)qUpCyADN?bkAwI#QqtvazJG z5I8waLYxjt#W2H8_-8Z1TG06S?$-LNDB_G=A|U?2G{JHO*ndhe)6=uDsD#JOO;1rr z#H~@48?&R#5*mKe>JP3U<683l`ku4(s5h(+13;iCK21hq#xm=AzlrhqniiO(t=&c$ zE;5 zrWvQBRDb}J!UO`~i-vdmpA&=z|IzspVL4%8g8>${qG{97)>L zf1m@TQ$YVUH2kM1j*=U<$`9UOr`j3|)C6tBktL=?0R>4I;qpW|*;>*LLzCEs7q){_ zLdFUO2~KguBqb^=S&bIC(T)78`2*NaCS->x0>RL>}B?I+<_|F-1)o&Z2U zKt6yz00KZffV`hP1Q0!lJcK?1K0d#o|K}6b06-`p6fg`J2#h#}yx%(lls-y6dA~e= zzP~Wwix|ovKr|ow>sJ)}I6J~6&Hf1>|q z5@aEOG+;6i1}FqnJWBpA9!dZJ5HJWR1Q-GkJ~2PJKLA7kK1e76iXc4%0e)~W0+b*? zG(SDi5CjwvMl=Gw;2;Dj5r7CjN`L~02qKIK1PXyduppv80YYd$MxY@?C;>=lJcghF z1U^DIzdn!vBnS~gAip?J0R%`qgfJm}L_Tp~0)zk|afCcE5IIO8Lb{+l)Bt&KK0=}( zITRp37yukFHYhuoJNAHs6EOc0|Xg2mwl%0=9?)%m^e3nL@UpgZ}|S=nzJ%A$#ZnNN5s#i*V7^3PDFjwLv+yFlL=kaJ z0<*v(v4j#aEIC%8d3vEjh9v(lS=0H;v4(Riv$Pg-g7!kNt5%99SqW!F-g7VVFI zYU4S;P{Wpfa~heUQLA3xz>#ml`9xfrrj1q2Aq*;bD%V*}Yb@J#+iRWf;b4>IR?N(q zZ_a?;{v%;&10#&YE9&iz(q1*nD~I+;rYcW)bQN~HG26CQW8H}t*^*&IA_JSk-OdgK zsZ-d+ODKD2YqiBcH3POTDlHeY3r`c^R^dWw>G@A+fN!ZA0y9dKUV>eR||_z zPV*+JX;CuD*qEN#0vb8Jpig7 zu8MOOjYVA1B~~umk9eawT3M{Vt`{%YOM%eWVE(`f4Q=MfUmulkAQ}|S@c%&W-QgV~ zq}*mn8~q=(n4IUG`lxKjknzWCSF3Sou4i?$hoeXISUerhzI^s;OJ^-3jwSBb7TW^x zy}{{m7T$%%=-nlA6WeU46d0`r$nRrz(Y2W4o8zJ=_nVmOgNCHp=obQ3RM2}u?}@!$t0Gpu;S(wVdv$wxy3VFt{DSLpJ@xa zmDU%oFU{PMlZi&^n9r%m!x_XBt(S~UXw=STrg8^i)WT)@aSQu)GD&wlLr)>z{Q^^} zq)65FxgYPBBRg*j`Cj+lm-l;hb{Kq_f?@e>uV4cXU2&QdWDkWGYR~y30vz_79D0=) zbiB{i;MuOT@=9k#^_MTsv7LgHj;fQ=Rr)gP#AnXI%{$p#e44CY9ka?CyGKb*rVb&U zrFkZ~-&Kx$M@r67^`za@tc;?gY)THh+9S>Km216Nvn#Mq)wu`iUG!UKvzmZto^$n> zDVRtq##yI`->RnHV|6dt2Ys-!%~?~-5>2(O%Oa)hi3Lyqv7|0 z)!UuttacPMWNvNsESdXJ+TKb|rQHzSin-LguB#N{COz_=tk#T+MW0J%uIGDUXv<*l ztP$D8Y{u3_LeuB0d!QsOL1(kbL+eYcNTG((^@aKk*agH)G~TwiEThYP1d>syE|shv10*jqye~F{kK3vC*>5c@M=GyDRp<&XkD+ZA8DFh77Q4>TA5m}$bgRWHHYh`GP8beMqih}N!V#DBt=+6 zN3>CJ%yEuKh?QmFOWnA0mpZELe3!`z+3(EkQySGw4{9Q!s5G`EbA_f=vBJ-~)Uc%O zcw`B$$ayEypRW^I)ypOpCe3R8s?FG^*t*#qe@g_m03MU@5&s8cS z%~N?7o7dhG>J%hgYBFo^*v^rpU;$ z99!-%^bUAx=a})ZQP>?fhPx#7)*boE4AA^x0Umw%{za z1D)Q6*EZAVN3?UeO8&oQ?aflb4yxx-bc9~p{FaT%Pe*PkxlM^}o7L}p)eEvFMAF_Z zw{y%rPq3x%1Q!~p_*Ucb6K@3G$D!OB%Rri0y9iHjMWZ5zAO|&_iSV?>skt=qGw3H` zjS4(!8Q;3WU6@p*i8-*+$9)?I;FE?I+_2_1vTv+%NCggO>LiVGa6AtY z@Q<8ixU-v%;%@VYf;k$xqvJrTmpH7fs;#Zz8?N?c)mP-A&KHa66aC^#^X&rqodI6H zrYZO*4MrO-3Wpx`cy-Z#X;E}e#vlRyP!h#j4vwN68=^}hp>Oxnve8myK7*klcc2#T zak`ss60|jpe*D)ay48QS1BG=#StD<(#F9$d4fQ?w%{#PI%{39>b>M1 zb4pvK_!m>E=%8~ATv{hR$g0-=beYfBS+O)={GhQ*c@uC5tMk|6Uxp5lYQ%P$WRI)2 z$@J}F&fD9VA*%94*Y%(|;dpGhc|3*%z29D1J#oNwmmUdo42x%_YUY+3780810&lse zm%{()+CJEgfo<0sdv0+ys#&cUU+X_9%VrF#%RXK53lA!b8UpOSKy^@|G3I>v4sI+O zkl!tDz^=HFRkJ+3W7MIjk9!w4Uzhen9HU7XQ^>E}_N!Wce=L?Kl?6 z!MMnEuDxj(G)!j2%>b9stb#A&>Ut&NE-t$-ct{twzCXys<*AI(#c0XyT3@Q0Pg88* z@`H=?KE3=Hujh3z8@{cvWpS)kxy%MIGhDtbSCB@Tk>rjQ23n@Q6YY$wBHJ!i)-|6C z7#+H;^8G5Tviy4^1M^I zeH~pUrdjdgU4#=heV*#l#bK#PBknGBaO>7TxWts_Q61Q8aM`#_)CY)&n?)wi-_@?C zMVy|0H%|>}QQ5}nIDInE_vpTl>)Aw;*}Ru*n$oT?nL^S0=Q-%4PqcryhI(0kG40V5 zn@pvjOvamSl6;BoiA67^p_$foc<1^p8funFwX}b;pf4Z|eq+lN zw*!}iZ-F-4HYxO>7D*KKg8(={$G?uWlQ}g+MjFH~T*w6_ytN!De+X=M)StC1RCTrx zc41ef+LJNFE|jHn(TsT~1(|Vi6d4BSp}MfGhy;g{zW)a-*w3^%((jA#HQ{5^U${#g zwoiq$56yd1mMFir<=^~b4zWfA*RVub+(vq^oH>|F7hKtwxisFI#G7$2&AuHY$oE8e zsWnVp5*e`loJl)G27tI_vh5h3DtN0`a-7LXeDlcd@e`4#4Fyq=byr2-!b%t~3G3WS zzH@mgiuKmr)ZTNUqZnQu@s?yO&N_LgxqKQ_Lm;HTbWha#KD(On zs!U?(o2|&vG+D1RClf!Jn*Le2TfoO7k(>B2&*C~v+b+pzyalvmOu2KVV3;~zpG(JQ2Al&mpZ0%U4c6s64 z-D||9W7n^u@BsD;buE+z(sE%#}Rim_VZ}-sDx~tU4%l zdaqU(%G6P6@z52tkS#FB+#BQ5Zjr`TPB*`SrRAtdpo3VhnnH~k5td*c;>xkXbde)(y64kJdMed97I_-=&NH@A5I-LYwik|gY zo#w8RRW7Ns|IsPP91kg>3yug6kks&E%~d_tBi=hmX zdHe-yK|LnLJoBttDV{RiK=jMI1Y7JW?eoLBf@zKh4l420W@2{eL35(wg3g+^XW!ee zK_y?(-#%`V6&H&iJ99w+jZJ>!{N8-`Krd3@Qic2@6a&MbL{k zUVEKPcv>?Yo0Jd!x~02J7uz`jlnYOze(M zheXHtqZx^Xuoadl#n%G&chB|J-M#Pvqis9tY5UR}TlPuMEBqb4l@K(PD$;vp>uRa4 z*WAdZw=F1ke|5%mdLzfmVU+lH&mM@1nMm~q?7C-7 zg_sWcU0-1v_{<3&?lx`1mTWF2l?^JnAug4Rp5UtEFJN2TT_=ayQ`xnm8U{e?4_w%v zzT_0R%+oy^kQHF!zr~{$ib!CugG(XUXZ*#&kPm*9L6SvZ@c$Bw3Zbn)6g0Nx{-Qb+ z{Yaa$M9CBkiGcW3V(8ufpMp^(1NvVjHgx8JkIO|IRR~KNa7f?(Q!vK7QL>ex{a^Fk zS>G0}Wzg$5yxgp6sp2{v&L!mTg^_{r<;7>kWoBh*C{4+~o@xD3CT)Px95(VNJluMM z6+ui2Bp;oUkfbmcTQ46>CeWI&P_X`{4wkOvHC>VsEAz?w+S&W4Tfz(Mc4MRb5?5X? zyKR^gc#uYETq>`Tg=I7?B^<$m^i}G&L0pzTgV5!KSvfJbyXbH|Aqn&J;xp6IC+1fX zl+vu19D4tZM<+PIW3}fJxXb|*h%GP&-k zkK-I+qn*bx-}3?w?J-Fer>Lccl%+gFh(yz2aJP2BXYK>qLs&`+Ib?3U3jkmPi<`Is z3RL}13HX5h-&5dI-~@G41EYrG5el>QB!NtAa!MMC-4EK7@rEAiDp{rlg*e zGTiacKqKzcj8hEdU^VH{{l_lu!rk2f(R$=~&pzc?3U`e68CbI0-?DIHm7exp^d2VA z`r&)H8@&(!CgR{}ZosMkms}qBD-_98+risR$`u@F>^Z;_rfPKZtduhl|2nbTsDW-` zVFUn$Ci@d<6X`NxFw;{I0bS0uNMx$_-_PA1jv65pqgFmBHRcXePqP; zU)dVLol8x=ON{qK#brBx6Y`n|JEjJoIP`pzMa`?HB~i*0oM1_cxnR* z&`~AjJqOCH_;1iObh!zX#M!l4D!8@I;(q&1b?sAg%S{$urXT?S9cVQ3%!vOS=uPO6 z?LTDe27uoIvHL$i>RJjS1kbX6+ZglIPihUeU8J8Zk@SE9d>DtmzY-offB(8_EctE; z{9qL$Q6-uMkbs3pgh)j(N(tpuNRS+hDqx$sP{$84+tmgf0?oiGyrj62^xS{um%l4I2h!vKOZVYzasz{qQDV= z4zcF|*#F!GB?CAB2ap0RfCI|JH87OeK>-vf3a|%Bi3=1?lHdS}9oIjI#K{>15C=$rqDBUgAW4t{$BG&xLgFF| z2@(TXfuqFw7a&Qo1I0)hEI{F72@8+_=s+XHdm5m~(SpWE8nnRSe-IlW0pS4$hz_{G zks<|-khJiDLqw3Apn`&eP>`HJfuw}AuPmXMGX!$a!`N;iUQihq{IaYH%@SX#f}>oWa8uu1B?MAU{NCjNth%^ z!DELF5;1X+MFfoitl&|j{R^0+*ui4P4HmHQu|x$-0CbQMqdg5+%JSc2hzyJ~>C@?r6LLmSWB0?e{ z0z5(^;6D?A20#i%Do`T9lMxq+vZh3YB0DM&14)7IiBjf3hNC1p5M#>?4@PowCITpc zC5TZaLr5YeNReVmjS@w1lOzO@fmo5E$OjihNwXrvP#P_W;bln+A^~ZTBgp$2#K_X3 z#*rGdNa1A24WaM}5M!T6)u4^V@@Qm#1IcJ_jGCCe5)~S3qvMI}LlHYlj z;;g#8LJ7Za3cmr*F4`b9uyY=iC5YnwlA*f=y z>eklAa5;R_os4V?nV)<^hJv#uJayi$BZp|a<*I{cTtwS$x;AqeqOZde>`>liiM(qmzF=iip4H`;W`L;A2+bm4zA_ zZPEnM?WqI%9o{*V3oE??(Fcz3ozCuz4T(4MY8Ikvd}5Cb@Onp!5;6=trQk2bqk@m@ zi;t+_3c8X~z%~|C?Dx`$z~!AWi*YEb&0Op6)lM(4?f1c4U38BZq$1qk6DZo}KxagE zA`&BWTT z)45ly-nn}4zI9lT8tC!fLEbt{YoI<+utkL7Q`l0H{`c9WFC4zr2qsEaYxCTsmdbPe z)#g%c(mZRiMEzH3bE2|yOsM?yN^vR0+?M?7Ptr_yYiByQH6xK%JyCf^3Acv=0U1w5 zv7KLH0V=(y`~17WxQ`?yWcnG-d>_@b7(4Ir)Mc~2T~T+aoq*sny~a2rB2nt zOk14wKLP!Jb^dv+D%Z0dnM=8HoBG*Ab||{M!B+-Fz?a1rJFO>@ea$)=5KBrWkJ?0L zmR~VS!aFAr{GNxO#vjyCPIj-~2z*@fZSDBI8KORI-+9!bdu3WeXSJEI^PGxa9IPFf z>^^4pL;lEUPQterk!V^QrdYO7nSTVh!}m(TFp`%bQSnuJxS7}X+pcC@+aVU^ST9u^ z154nfF(rD5g+%Oq;UMaqVL!mDvtYvt@6y}N?h^-NB%PmM8CH}%T<0?+sGC=^xH$G= z36RSMnOb9yjud*sU+pG5JA9Cvo>hp-wNQWAy(Z|RI%Dm0nj)#7F8GhQp3qrl{4a7! zc2%z%!YM>ilULO?JnAq#FZSaoj*4=_gjulkL5DOwxH7j(3N)whIq}bUMaIcUDWvhg zo>6WpygvwLG-k>Lw6+D35Eo#FCM$k5OXQM<| zh8bUkIwRZ0o{};cOjhuDUToWT3d-gLz_=;G5M|kM)1rLKl4ah`opj z$R=uJIvK}((UDH1mF+q(H0I z+P2lsQf}`J6pkdwq%L*0-IA#k8`Bl|i%@)aVYPRcb^)o1i6P+pi_RXzGtihnLwyUC>uiD#haM}(2b zbRPBgB4_Ncsi4)4eghpPD^FZcINbM6D}$(4eS3r1F?)aU!5;Mto-JSXd`e?>UW8D_ zrd43mP3Ib*;DGDep%Hnf4o?XsOYu>wOlc9kfHMAJ${(SX;)Ae5wQ{_z1vCD9DTYRQ zu#0>!^CX>VW2XGb(WT;Kr6sAWPEpoCPPR|C0Jis(E!S!NtIz&U`C5XMJi2$>Ji;KB z*5ZwZYg8V@d1Lg>ukNMy5B4;ul###KI^)=7*&yB#{?uso2lMDh@x)f1r6%nhYfT*c zrqir2jIAjyq9LBI;eh(lsgB2XTn}v3R(SB6B|UE2J6Rt10Vrp{e_#Vq0jT5$%)Dm; zyzkZ_Erq(uO{Sf|cMW9}R_2z#ZRwl-Fp z)c)}$)~fEl33~oBXWf!!?sg_@ECS2@i*`D6ZSs1ITd(FX9EKB;k*;H=9n2zMPW`-b zTMe@8kJL$)`r*;%Wb)XH=HJ0B-}nvZnY`juq$f#4V_)-au57cfhjcuylR^9UAtwbNZcs}rz+mR)t(InyzBQQelkR*ClmCa#HsPZ zJlFr?@>e%d`kk3=>8bg_=3Wu4)l_43;T+fEQ&HjVs#m@*OUs@=J9d;q(na!9%Oc-w zzO%ybbti^((<0TvnzN+?K`<0Nej*`ua2;#h`P-)8we_lvjNt;Q;fJO_B$nqHY*8|YLX=YPI~th2&7EKB}8Mb&dSuP#0`eo|`x zmFGV7qdN5Jyk#V3(_!BCnsL2`$0&wg_XtMFh*PI9tgNRBIqsO)s!G%~n?7mmKANP&Q12%JLgmA&knTYdgbnRJmg z=s$&AdO5KA!mF2C=+F#hLojdIqf^efp0d}jZfb`ctL!Qs;RF8YhB^GxbJ=TKKjTtH zkWpg)#z2Fz0|zHcEGUU*E8@1yeppzv!;S#lXvbO5UQCE_&B3)Ma9&F$HE??kCldzY zXlp6sIn$T4g0RyWS`M57uW2b@LgacxQrd960JjN!iY%vTUBn~QFz#uQ*>aL@p<9xP zbfM7wLcU?h3JR@JcFnrrJukJzYtb)u>Xd4b56AmgS%EM`j`D{wabJIE^sPBdB58T} z;MHnyyk#yj{9Ee*ge?%5vVs3Pb-1&+`fev|ZT;SkE}z0|lF7B_8Wc%ciQYGh zs4QJ{{WH~KjIBg8SF|~lSMe@AZk`flF5m10*&mnorwH}`uB?T_YGd8Xs_Kat_uSxW zvpRgHsEv1QeSHXW_M7XO4&ZNh$qIOSSTXeC~p(a!>{B0JeV zp%1NLCnFGX;YSp1aMLH3ywg=WF=zg|{khk(IOIuf5-eL)G2AXgV(Rx1n-R`Y0goQN zkNjFPXd^f5*7@#kUcCsb0b7vt6SbX|ZF}oIF;A(A)q7V)FD+sCe9Lm1$m86m~Fzg$i z#6aTY8`AFSisDM8^_EOo#@1c)2j(B|Qyc5Q|83}BH+CDXMb)`!ErRbu8^V-5XsVtj z4aupF_>O1CV@@uIPYa%C15K%73c9(wtWv2-M`g1lR2kRPN>XdAE&csVc@D~3!j1wM z)f9VH>ji6Fx(@}rPdz5_-svCihTUDozh_F!=%E_A_?Irt1ogji0{*5ghrywYJ;tvh{i9t+!7oQ$$N>j3P zjB;mfwlH<+;Yz{~xz|X@jC9lPn}z#hROWx@pye+G)KO@C4BE8sLLK6X+ElpLG946yVF4k5FEFu2jF?R=&0 zh?)>$5}B1wXaUF0?paH_t@vdUGGnHx=e~HbxNGl(>M^oM<{?^IteB1Sn zZi}`k`DTab9kM_hx@6H%beeK=CYvz@rw)pdK5pxs8KKVzel@yit+FSsYbk0jY`gC4 zJPE#6KdK0Rmu;15nEs)*@QRcCL=jmgl1C~qo-4J_4Ae<^%qT=Q?CC7od|bVb$l781 zWwuioyI&mKSFkp@G-uN0sP)#@HeA6OGQ!$Y{SeYwZSAQ#{`p+8w+OMu`#!@iFvlRn zMV{tg(mZueb?CR6k?u&7i|{QYS^n&A!ZFGwrM^9L$yiZAzm76z^JrrEi4jFYM%Miq zkJ|sG4>8;(SRG9@9)BvfZ?m7Sct>#^O%~ww-$WwSq_K9qS@cHT_^ZocRx`H1XQ~ zw$2IL*4Ix-#<9`jYOA&tDs#{5*&!p=og#ADgG;rg{wdgZ_Ygs+29BQ z9o42B_*Nn$qq>>k5acm&r3(G2XIt}lCWypo>jgJkc9a^imfB;Y)XG;>; z$HDL*h?{o3kydO;ZBwd}0Y=Ub*#qLFEa#M#%g=!cM%LZ_xL!90~p4JOgCbv))5 z7J*G^T72W{yGQf0THr)!GnLGkVW*Gw7h=%>mFta%;Ge~tRT-XCPTvX6VZCOMOIngj zT#Eii2{3X^H#0*`!czW1Gr?X^;JXV~y=3egKOu-XJh}_e- zE$x$ub-8}cCr62IkWQ6AhI9`D>vi34@G)}c6Oq7XTHNsCmfQzk5i0ZI4=Hr`8Ni{_ zk#FCvnhIg(rSPYKE1g=e76)O2qQ$T#CcB47xm{`k<3W zxH_#?N0D@UD&SxXNIXA!U97?*`u2@QLcPM&E5agb_75#p#K|z#{?;flHsqxX6y?>f z-qpkLi=&1l^n~|N)#}+nyuc3pDRPzT^kGBw?yU54wIlRv*8obJpgEl`qaR`rblh&_ zHOIuu<{RU=VbbUefzILgFTvUod5^|9H7m{Y(f(#eH*kjH75?RR7X9<)TU49p6Jh?c%9V$4reIJ9yXESyVw3hNAsZ3jL|mkV0mn*u_EBd3oV!rz}#X zUQoNQ=9OBHyPlextQsGBnZUM2OCw^igcRS`EhOQOAV56W?)5G56DB{7;}7y$*s2+a z=cxA5Qn{VK_dA>YV782fn@VBQT+ddruoT>|D8t6q_}6huEj~;?UMiM^F73K=noJ<7 zkqfRZ*7ty08}O-Sybc!J=S1Ml>6~beax0hD=X}j4&>H6~C9VUULZoxK@j{Yj^IdBl z&tBHTd8K&jqFk&@nO4?Dmih>*wo&JpDpfzXX;wi=4^P%3bhScfUWan`j9DNmSs;6& zQ--GD9=;}=u|@)-8PjFrG$U#5l@;Xcjxxqv3GYwYqT!E97%V|11s^{{7EO=;B0sVJ zg{9EhmibF`!qQxAJGUVk=(lj#(sOn$Z0l!1>KG-jI!^4 z2Ku1VlR%<|IrU$`^tQfn@s6~{5P0yKT1>7N+s_4=@SVI)pdqJJ%AC62&EZzQx}D$+ znHDeQN2+rHQ~JBt`C{^i;V4FO9mdD|!imBkAU;Pc$8GCoV+fJ^-psXqzKl(Ij>b7EzSS- zQmX>FW&95>4Fmw!kRL9_?@9h-7HuZCG1C{KB(ekyt?Avyt5f6G)wi>%wR68|KIp&N z4~@A`rRb9siE{%AIf}X_)O+KqoCo-so!2V!Fy%AZ2vu9Bv453wmgy4yFf7vz60+3} zF-B&ESlUmVI-Vl}{JD5w`_xx)24y|@xR7nu*!?6U5lzv)WnC&W_ZLsEPu6Mdx4Tl| zELS-52*cWUMX-mDPYLzM%$9!-a+0r{ zwpY4;>Xy|Uc--0?!>cL_?zo}tnkiM2BR?sUBmd6&0+kqv{{p$4j}0cvrPb??af%NUCL-trF~a}p|&-g_Qgk{u@}xZ>EsT%iMv3O_7b=U;RiHmc6) z>MzHFUbU*HYNw1S-PG-FdZ|!+S9bX~Kvk2n^$6e{A-INbYTz&L z+$$3W@3O19Px`hp_%f}kWBH)|Ra25H6~C#E3C|BYUsm(Aw2M_yA}-G}=?Cko@*1l0 zBL8}kKOD#dag9rM*`<_lrQm4Xe=Kf=+;O*L9(PnN;oWq*$wgIXSTF zqMsSb28*1f&=}d6q1fc7%a9jVqn)AP=|$3&aKvwEW>PTk9rVaW<0lHt4H-S> zmloi}XBk^@jZ!7>7Xl;`JyXIS=i^f?g{qK6FJ~BthoTwhP=BXn#47F7wLCfaB5pp3 zlMu>m_S5+70!DbLS7Q~S*u(A69%wHY07w>UJ4W`CmrPIbbpJm2A3ODy0>)!i?|A8CKK+{y@-if~3*u!@JxkR( z>7?+=Z6)lF5yzgg4B3SH*8Wt6l_Ar&#dMPRiiW4*JqrZW{o)$l0Ejsc1ebc#s;A!O z#>rYs6X~HJ79}%qS-)jKBxvPPKH-S;I8C+^JhzX`<>bLmWtsyzY6V1Gn+lW>FP7L% z2lq4e2PUoT&6JD!n`Wu+VAlJlf=E+LV9ZmOzMe->6P|Oew&%}?;C^(glic3>IPau` zTqt8tQ16rUv3OzLt?V8q^dV?BSyQ<^QMq#u)oc+4II|p;p&i!F96S7x_5%yO35``P z1Ode%pSz|L^@TX#a{*Xfg6^&yAhyWuN}jaJ^kdq`28{C!jBw4%gUOxStCiy5fJtcL z!6n$Nb5={0igGO*|F3-KoVlHLytQHX6+!F3Nu3$n>8VI=``{}Q9*$&g9_-rkdjXnS^a!^OG!Yd{8uN*`Q=z5_0Na9>MoQ#WEvyvyW0}E4W{5FTT;$a+3zNe46KXW8I~@Mi zm^bn+5yuWm)0A$W;m$%efnYiI0bi|+?7b(Vj<<;=s2T7pF%$?>os}aJLh-nzg#yk# zh*7JcSD2#S^!($t#|Z@ntvq%JtB)WG!ZqhxlwOLue?E+Sa-`Ib!*cX=_*pTulxM0H zOSx09dfkVcw&JVOlbTz_)6vlAelVVs>ZIO4^Ouo7BJd?e7IeG@bau`X|= zU=R65X-t2#A)4#asgOTVnC#N~=R9$=NmnnD;xMxD{_*NOe6gH@vh*C$1cV+R!rNB~ zTsLHh$K0$O;`a@nF+z#?{f1^+EuBs&JVyH6-FIgLzj)R~Fq2ylbl6lYC`Cz|vDQ*l zTIsd#VjcS0)snP7Ap-$dJBB}@QWcU}mO5Ps{4w|mQ;cGr=`#>DTC7e6|M?*4o4NOh z{yFLAlCm8{DYS5jAXpYYwGzUk#y4{~(6)RRUYnsH1?%&>5T0Y}{*G0QTMCnpiqZ4? z_xu*&Lv5R<{)rLOVvqX~T$Q%l_!W=(nr|o7mxq?t^2?8)93$(7R~&Eta9yqKEkeP1 zu|C<~VP6-{dTK6T%B3Q(FX{j#72R6h+`4*s?4i@(C z_F-PX+P1-MFI|Wx%vB8ZoHHmi?4QzMkp#tk9fbolO{_lfH6L(kSEdr1wo zLGoDiDe17qYf@G2JcNi=t{^zBAGaVpfcH)$^}NSWNfK@qAa{Qm<9+1v zZva{C5gB~q;AV0ZSG;iIX#FgNU{JW5mPYhCC3N|G1*mW+W8X2 z=y1r*)bZXbbu|2f3!!BHQlCs4Gli_5*z5}2$SjN+P3Kt_GCiLBqE4t#jr-m_tW@T@ zl8Fl>kp{lr|D;R#Xz4~HgW2qseZJWsD_*$q4eUif^Eje@lMC@owTZ_Nh**x^8-aNb zGe7jXn+tS5jQtKOURvx#d-r&{*oRcxspx{WjS&uOKh}4y8$ghjNWk3&9^GJ(aU6$9L&m8A^P1_qio)&-*4z@ z+4efo|H1lI4>f_ro^#$b5`{$_xz0#s%ahRB#28px=RPdGVVLXMrSL2p8fHRoBq1Fr z06-;;R7Jjghx`9HTl~}*g4f9Q6%5T+waJ5>U;ly0e5u<; zNmW@0m0+yFJFS1Z7b6V+t?#*j>kTW-hM8=PWPC&iH`wMa$^f zK(y^k2g3?^0VZh}w}4)tB6=J7fEUQMh9BODt?8C>0J>km|LpNHDH+szGe!F!m@ft7 z^69XMU{3=!lwuQ^V)^Bnm(%t=v-;_*xDNMIH?k=S#%fO;bkgI8Zu!o1`NX>s^M=CJ zK;5ncvaBjsYxfbGg1y(bt{$^`rO2-xQO?*jEl;V9zJYAP8TTC}t`sg0dm<;10h+Fx zVP#?W&F*%wVFkU9eYCCKZF4qg?|tCnZ|)$x6l+{kpF~O8N4>)jEPXq@v7do8D3SNw zIgnr06d}0f)m*c#W?x<7nAxxp1}~056BZC?4Qn+uXe^ngT$$;N5+0)Vj0r&|H-U+E zftAKcIZ^OkT(?uEXAjD9>y;;xsIfcCa-}R9`7>NAyzDbY`;Bs7tLkbc293k5vLFPA zjgt-{`gK|Un~SQwMKA=Nb5{nA_*NG=_xxkf6s>!Jt#c8 zW>dr-r^OCn-zbWHB~7x5Kq^`Zlxr^S9I@_$QaWD;#mXV`eOE6kCWgLW($YC+nc;fN z8|Z$poKFZYAO5lwSYZKP$ZtE&ANu?%hAk#tt;Su(o5R@>hBQ5B_BA4@kxd4_wvVT3fRxpI^E z>>H$&?GM(q=ZM-ysm7111|r8hkHXo#;$%KCq|o#bR#O@AD~JO1((@f?@a$Xv1{#sU ziIi2)W#lzpV%u;u^kr1<%D+SO+0JLU2}sn$eI4S;aej9v2jbo6Ow1s#|GL+_u4|G` zVSno-LSWuq(pQ=_6&;_LYcPcj&DOafa`XzrxG;)>0V)) z7X#n?QbUN24~vZ|?b252%055zzSH}hnFI)0B~C1X-$Jk@>$Y#)ta2-TFgCZ9>Y)<5 z;^5767n9V@UA{3}hA6MT0eFf1G=fh=Rl|PQ@hgxP0TBsZYm=)Qjm>kGt}}4_|mOVt39?ioN+E`ACiT{wsQy#{H+gIX~uyqtRMSMQO ze2HKi4|3-DW`aQ_pzxSNP4{0PB$*}DmjU~*nOgI^6pWzuuW zwut|Yz`P1NbZ-;?lxo0>5u^m_-FWJe z)O@bAt61JSR_x~21(T(&&Xgpm2@M#H7ZiOz>-|J)u2JY*2n*XuwL%X z?895@U)F!SN+pcb*It0Y&Kxjkt!f&WXIyPg(RsL3`%^Wh*|uHBklhD$?BKmL;3(UF zWl*Ni%LwG%?kejohwBUdzW*E6y7X6+h|9h?xAjk0i!^0R%-ZOe4A(xeA`qsV2hXLM zL45!R-<~sfc2yH(eJ=0bkOzxAS1}Hvco)+V@I-wh*^tx9WbvZ~nwE(m(!rwwRtua~&uR4JA@JRrFcP{=6uZ}tuZ%@YNC~~;BAuX<``4D! zx!B(II)Y^9%b5vy?x86{8PawUfzp?JfITuXzupo%Klk34GOa^^btpl8U#jN2#=l~} zFO;_5izyGfh<8VA4#UVM8LInL`9Y zlEty8uju6rY-y2!3O@Sx(+voYQFMiwDTsP9dOnO(&I}}F{|j!3=8OI1;nL0Pm6jWu zH%(@`E2N&YoLhVhN9jfRo_W3W)t5dU*on8KN^6!z5HlhbCX9avleALtPEg*?1g0%l zCTPz^oFC(n^;5KAIb3LyEgD^54Sf;6QpBxIzXOL4Fa&^^`xPGK`HObC+HZryRwpK7 zbDf|N8Q=)m2KhnT*C3G1&u@n) z2vZs!xv_2&i!dv~3|S@s#-q1_xcgfF&j>OMFQwpICtE@fmHi~xpR-(}3d*KyL%}BXgIadljxuKT$Z4FWe=&n_SN26J~5T3sqGjvR;Ty{EazN2@}89!Lu zLLP>QV+?wsu^|mS{+q&IJzE`8I_RI9H0gOAR<9&!`%1)vo znJGmYJuM%W4eveiv!D;y?mGLYAoOI`rD2jyAff8Nmn(<5$}G39CgR6Or1hDh4Gws6 zy>7^Yp?#!F+!TFGPvoM5G|#rKnXZl`SQ6RDKLceppNAiv^uy==;_DJ-yKg{)e~k;| z8qdZP6@3cFzPSc6(kbL)&!c9aB(8Qpxp$E1FBQy0ria}5-HZc!-~wHWk9||@GFuAE zf>8fubgg({{R@>Lupzn|&}pwKYw3{K6qEW3I;IvT)p?!m_k`M}3>!Lvor!G#os;Rm zpX(zgd!V;LT0+FyNAZ^MM^y%M1%6}NQxWL&MtnQGaEuu9b~=c}@m z|L-!L`IL;H5<+gyhwqhI<6j+(7cpU*(s4rNO9Q=wOeLR8rK?rbj7#QVH|Je)l}Fz4 zbPG*eY8^Y7bd%H+-U=IDU^YEqjYb%8;&V@lqVfmFbBu{Y5@UD+2>Mh|}^v~KnJxIw8 zZUGfwl-fDrwRjLqRNXL9c!?>^Lh!I7e80qsHLoCW*3u!}37qgMsnAe;GEz;?JY*ko zoCN~95RXVHR)eRMG_w(R<)00c9@=yGkG*aN&J=~6%iY}vN%<$pggeOwYja2hlq}=J z@%r4nfI8)#FN`76Op=dln|0q`1ueUB`gn{UFjl5?8BPUsR?{%lv*^nWz-KM?Yn$c! z4Eo}+P&v~KTsPjNpmg_gnRRZ{7PHQcnDKJ?Li156t=#c~x4j!>lYU)Xz@5Tmxs+AH z74~+a$e`h7X{eHXon`^C)s`ooh3sE)p&`q~UaBaWu}PAN^U?tp;JgzpGtu(04D1Gb zw_yB+Lvk>MY2)xsdmjC?ncVj2-cIqy*-k=qAu80%2_d}dX`o69j^dV66g@}I#Kd+P zt(vYAVbPMLqqBJ_YFfuKaUAFt{mlOphlB_K1{&D~i3oxej%NG@E685UDZ?QqYJ#w{ zdP$|)eJ53#N4JSXV&Ape#=m zOy8+_RMn}EGCQmx=5!M!f0r6ZUcdJ=3==|gd7`L?n*S;{B|T#T7K9a({_kvk{3MEC zs|fhVN%_@{Thg%Bg|p3_tkWceF~~P--dH=7tG+lc%{NU{dwOUO%A>D0Wu;%PY~zA;sXqpI^!V zBrpa#1HnY7n>1AHYcG>rdUqzcERIackA9Qt*DpnI0(y7aKYz?ko952HB60z6lwbDZ z3W!|Gb)CVk^qr?49?I=c-^bCB6Z53}F>#JIlHDL-+Yk}mgj`%~&m4DYV zeYS^rAw9m+Ko5~jJx1laCSMW8NtC{m z`ldqG-URi+V^8Umrt^&!ncAh_Elm_%K{B{Zjz;dOSHYLMf)!iaFvl9MXr6N61{pEG z>Ihl2%cu+7@rrgOnkQ;{@TWOC6i}D6d;-J~E5eM|sRJf^qp-O@%`kH{1FFwI>Nzm} z-eJYdEt&#Lv`6$l`tn_@4${W58DtW67SU{({e@2*RZdsDCk)n{o+AI%oKxmW(BPdr zv^M~!K2wZDE;&2T=}dJs?|XlC`FJ`bRh4<866Xgc-&hZ2`d1dTqN6y@=Pm-cy7>}l zZzWfO92%Od#nvog`oV}@OV{RHAdH|2_3*z4G!(mwXQwyoyl=SY_H|l+%jn70GNuVX zWf~QLrpt%bG&svzbu8Z!4Sjm(oqUzz?}_OSeFui% z+5!ngUg@=rW=bo2B_gjkXDfADZu^sH?qz9PD;vwIFtym@{JBh=Pt$9|#3T5P?00kGSI_`SK(@b{c9g~1H%qzhsD4_WulmDX z@y==E$z87i{A>)(t}9Ka{Y^8!W29^0lWxaPgQtDMR(NYy=;tiaCvLO1OiHxQY>YaF zca0=b%5LV;!7U8WWWg@4>6@Pf2ki_9?&g=DL^&?r{Tk8EfquA4TD*E;mO|PjHQUr> zx(A-NV9MRalla!6btfVCggTUCIWhaW2AHYWI2CNxW=MXq{eM0&jiAEn#HOYKqESup#xTbcZFX8cwB4tgx_q0DBe8fA_spw$lZI9Aammwr%Nav!)W&PB8oAtq>L z_#m^66fBNkHF%o_(VFnR#|0F9x5g zD4~gQ80UMX>NU^tWBK}i5w5igg{TjgM^rPq>pvS^X;s|9hIi~$6jwj(Zef^jeE)T{ z%52(+)QE}DUD_)4N2zx4dyD%KWRUE?SoP3vb7-p!s~Vb@;;_fgko`$(@kba5z5c!AJT|n zP>AuGEA5ugpdsPi(Atu4BSO`df#k6KPfax$Be`A%?1{GQ@rIMb3__RJ7y>?lsd=id z@?2Da6F>6z#0~3TJ>jI@sr2?1RF#zsoBrqb1!W z`8HWi+#ELV-8Dq>vyV@PexIP%@tef$%&BmuUWY*@Xz^J9fwDK-U-0NI`1ySe7YQ!O zP<@?2o^2Sh=XvBfe|OO=Fq4hT+HZd?iqCiVV4(Vyu8`LJ%hQ^V{`3;!3Z&RzPZ~xy z&yBRUAkFV3RR-07Ej&IKi*kKYdAE5thC$6~$31R`2&5%n+NZCR-?6!3z}u1G!g@k< z2x?jHb~0x4j)frY5M%%HR_5Zi$x>gGSdE-0(A{Nh?pgr|5s?3q%GT&pDP)FPyngIy zZ`nY*)*SMAjLxfqDb(f$t@~~%TqH;s-iyA=N_{~5wI^rW$}C2&x)sZf*4HVlC{d20 z=B{R{k|ss%{0TajNuXm7Ot+q_aHo_v>3{hYZ_v2P-ZxDV+1McXJUGZYNAntvsUvyCGwl2sKZvGQoFk6X(dFB*$lL4Rk+f8qx+ zv4>5|^!GS)H~>SCu%>n{fwn$CY$gMv%>vlTHk{o;eD2C?hOX*R`ysgcyk34Ga3P$R z*%QD~V1A`A*rPoly0YhMQa_$#zZ&{f5F*;U5C|J=rtUF^*o0%PDM4k=S_xEX47XWz z%8TgTgakG)zsV8cxT++ z26fae?C-T3`=E38@7K=>`=sdpae``x^|%lqkZ`QBmk5XPLl~EZ@#RB{0l@;-X*aHk zdH>!xBoYFq1CAK0kY?;t*ebFVw$<^uA!g41TWR&1gAW~{=iWK{*AmL5V0_5!o#v+m~xM{RFHVJDqYn?gsYHrJ*FJjm;G^#8v zAZdHQ^zcvYU!%2Xnp5jE6u&e?Q>8O=fj0*6W!2xBjvp^NKf>qJWd~C%2 pL>+lA z5*z|5vOc4Sg*#IoQ@UP}(rv716RlLmR=0w?xT;oF#Z_4;xPQyuJ#)_3vWh_69=i;G zYbl#nM1rPn-#>hxBMxd_LX}HH`#}Q!2M0j-zt5h|7Ex$U6{c|eh{X9a@+qN!-PH+0 zzf&m~w;J5?)NOK@lA(g4J5+}-)THD;)km84@{2RddW>}TA0v{A`d zX}W_IUr{_(>H=!x`8G2?)_^R%u@}QuHkIQBgcuW=izk}VBS#~2JS=BA5-uIxdQyG= z@>wv%i#K)cQv<3&Z)D?G51UP28An=L-KmN@f#o(6E z{PX$!B-8f`Y+ExIFev+9`?`=sfA$+~@9zO-)Q|i*?OQUGA#v0eA*T}Y{9)$`+y%lV ztEsbfG=XR}J{73y&?BJ_V4q0E4G}0(rC_Dl+g1(>xupgod49*cKSkV1lM2hR~y+xNGLTwm`FkQKvh&3Gofs<=F6k) zr=}o)5=1=(SDxP+*WmqO*K&cOB@z~)m5Ny>CBthdLQcnv+pkSgWxM#^*r!9~wiNp; z?!twGrgk$~dn{Y6*NpCcxwr>vqymbyAHEvU#^=)mj#O+W;)s(ot+1VvK%MI|!>ELz zU5={i=L_M9r;eo%s-0EeHb~`NUBqFRddiGDO{k4FKaNn{w!9ktQr2g<_~Fc;#Ec-lj;U0dc@9aZlI!B2jQ@b1#$i z5CA^Reo#+5vk&d}{p*|8=_)>Ie>>9yQSzcgh@e9t2cpD4mMtLZ1NdjDA1}dMsGr2h z9aBGDL+zs@E0l9BT&x0j?1B_1hFbC!kky9f{Llx!Kt+1<;F3J&A85a0@9&%r&Rxy=)e{GnVB zU?!HI_;1&=u~8L^aWPrTn7~3#nIpI24yIehRML!vPqw9fRzNzVO}C00mcw!%uHwUv zgPR;YI+^LENyh5P#~XsW!JJ9AeqF#EY?Xt4cLw$fb@&&h{O4JIj%Fz1CCtt18@!$I zlClTs*19V}?!LgLM)^e!StaeRe$CBW9li(ju|EAcIrBnxlTJr}3*9NbTzek*`}65i zu%(^w?<)KfMt8J<@Qr)B z>32Vn1hS&bYvb+*lknhu7ef9PUlG?60HS6?v(42JyP-^gV~Q0I~#0vgHN zEzxTG`Ib?t%hc3S+O%rsj~dC(Tk)L}VhVYp32n+&0IICIL(lzsq)(~PKA(Kt&jX!u zx?&#s_+CwH-eVF5e3(!&$_wG=Xf`i`DbJZTf^-fo3tV}~%eMsbn|Hqij*;^0Og*hm zGR-n)f|qg*Avp#5zfhjGkGNogwnVnX@47}BenOwTsI z;dFwfbG<|-@-a~%+|r+Nnp)rG^sf>_4cG%dzr~U1&p49CPj`i%;*kyG4()9TXMyQN z-)z5}spJc@NRo5P%I%6H0<}kMnFTwNl@0enlGvG6wiAA0bqOqu2VvP$y5Rq@_LTuq zu3gvnFm!i!hlEI~$Pm&E(kYDyC=JRCUDDm12B{!Y28|MmgfMg(gtRbxGZ;rbo~PdL z&-|bJ+I#J_cHXSG5=Q(@EEgC&H$Hkj(nVCh_oC8vY#S+-ug%jsSB^LvnB06m%+WX+ z-Y)& zWw<(x+!bwWRY`OIFrH{_(Rp7_eEngQ_X*>F(4El!Url%^4QrqvNy%|hJl&0m6Kg1R z<+5RX7=EsIpA=KTesma(4n!V|FH()}e>xy2C*Cs^DYv$I^R_PNX zOIdzT@27F)`X~O{=K~@zZC$Nq2+oMt*Tm1zNk(;QAQPD(HhuCwA-GJjmw$~*u zv)HabK09A}Crh8?1}=lsX4+jcF7tFZvgGpDWtiG%jfaxh zoerFH4NaAvhkXlG;wni_W~zHjanj!FP2+ua52h1zkley~IlSMbZCK^F&u88MkY-h- z+;6P0?E=&kD9`NI+Q<8{y*^TFD&z;C05<$cF2AL}#F}rrBy&c8ypigV+fcH>wHMtH z^{D$VpeF$w4U1hwfW@nrFA5sRkG5Aw{4KgPTK48X7C&J-!ICy@_S&jxt5p2&$M(L* z864#uV-G=VL%4mlg;8avyB%!r&BtKPbgS{*@ zV2zy&pdVaIOi{L23RuotIeeE?yRcCKf@B;O*tHmJoN$+bn~6^~%z&%~P; z`4jr5F$tt}3zTM6*q*Ka*7aJx!3Z!(LxQ%uPwTaUD#;ETR6rV!b#L}#Et_s8MrE`O z#k#VAbRp2u{yC48ydW7UuUXWPwS1xc$D;(NYs{LZhIO1Ct6N0ck%|wa815$Jk48%0 zqG&$Tc-4+^v8rPf*j6Lyc6jx^&F?0J1?4oFXMx+Uz?KEVQlUy>K4+opQ zEsDMH5qmV20|bU8s<8e8SJmePDUk-ECv>Om9wpgj8YCLn+BsjvT(oV=-sQCOa;acR z@elebce7Fu@pVf1_ryF;I;RHBFMUU--LL-;MZhP@M|3b`aqg13~i%kX-1TK|2 zO=W78Y25?6QiSwAyf@ASrM)WBw_DBxAOa48RU7V4m0vx{zkJQ@<5Zh4^muDcrQ3cJqgnImC3)rK@F0#Nf0&0@4kDtAW^)ul7 zAF?em-Ru=Ji9bEQ82c>b3&+7oyve5q+f-2%FH1B1L<*am^+O#?@iSBI(b?}i_Sj?h zw4cg@vhqm}6NS zZ_kUwEK+u(N}I36`pZZ*0;^GZfB(#v_{{VUH!g|a_c7HZ?z_fSU2ra-B zSKmOv%GuNp-DgqtJ`%*2y=$C3NS>RS-IG-IDi0#K`vEnC942+VX@fHZA1YJ4wtG)` zgG#>K(2S5n<+WqaJxkqJ$+Sa60F&_ZN(RX{&Jz#aq>a`0%VtI!eK?#9&+-A${v;pC!F^Ih5|i;jOd1rM!QhE|o{K?`PYd z$}8b-o5x=c5-Hwb)RkyZ9OdSR#&JyS%GlOsrEBkWYy0HEQ&G4~*eum4tT1Z< ztPd?UzDRNZOUshb4l$^RNU{)ULdO<5hXz)=Bg>D(mJ6TunPNFdl;&d~xi<3&@|2RK ziDJBtq)E!>$KwOhFY_!W8Q5h|L>vEU7?lwl8*s%O=9FQmNyZ0yfW+?XRRa09QstY< zTAI6V6-)X-gy;1NOE=|Hph2H(xr>rB^^vfmZ?jVDNZZMieugmJF{=7w9g*>lFN%&p&SLahTBU@?Tr#HOa=FS3kJqC1A7(p1( z>-6#AnBCovk0$4oY#U~xjC3AWgkAge*i!gLdlRNeJhv(ZwHvHiZBqDSW`ju0rzV1U ze@@uCNcB1@;Z&7zW~%TL$>-e*_BraJ%z`bCr|vG@j(>#1>~g$G^EaaPXX_07VV%?% z3jIH+u({44gi57p#TL6PYCM)c<0h&a?Y`FRyJhn-2NLXrPFzzySR@<$OujeSW&o;H zB$h)1IR*#kXEx!!-^|l(etH5@F`b+Yz$jY&-HcK5tb0Xj7Z3Bxvs01Qxl(%fQ%Y#ftME9CqAE~R!l#R)ud5n9!nnVTv~4RO+XOd$|LdLZPg^(QbEXm#IjUJ5R7HY zG|-$fvc`NEp~An0&VLNbV>8!r4!PN*WpMTKqt7m4K{PMtcE2&*yT_sBEr6o)3YOzQ zicst85D*bc4DPuZ*^-^E4Kx{68kQos_n4DpPY9DuS--cZ)ZiAYTn#_Re z^Gl(TRDzSqn6dlb=Hgv(eD)d)FusNEG6U(C+f*L&g1u)>rp-sI&BYsDx^Xal8Rg&G z7(W*$VU8s4HeK-Km}&Q61cCq~-*yjMBHv8l*n7r`_h9~Y&}|SJEQDB}5+9&4Uoi>& z*QV0{Se_+D1{BPu(!RYiAIo*H&lb;tgA)-eAHZ%CrySt1kPyf`UVJ%O8&gf#04=fG zM%T|{z}zeV-R|co`$~c`>)Dl&C3&`|+jd#XB%)kU{zc#AY@K7Wm0X+(Jxk3HR#bDH z5E7q4;Bp`k0r5~fso$pq`$8hguD{JVa|x5ImF5d5T#3J1#t6Xf7o7KQe*g$E+ZVb` zJ$v!siP#KJ*`5dl31*YBmDPQ;cuubC+^T254R>(X_*F#)<-uxNZq9|~Q}Yhxzt;3o zeZ1EHZf7Hr%F*6B)z_Vkk_kv@0I;9tB07?|&5$VcnprJL)P2onb(2#jv;!CIFD!R9Ut1z4nlb$FVN~wG4T#}q^w0z=&?=+H%RC#i0V^|Z6DFw( z--wRjsx+{@+;Hpf+e9jTeL20>$>rrU6@d7%bW^naqEleWV@@3#Do~8+8In2?$}CS; zyN#o#XOsJe%um^cTsP(C23#M231B$g^ic8j=`CgaF)dVgi0scTJE9KEnl}%5`-ekGRd?zvlKtI@=@g11G3_vHF#}4MY@|Yk{VtmJ8&ZibG z8qHdMY~@-(0@0lPk3GtC^W~H?)qMCIujF5-7@s7Nb0>bP zGNG3YQlsAE>}-RrTQ#qE&vZ0-;!t8IVVhHLik8qm*?9+jMi zPn^RG5@%1BjUoF*xt=1b)eq?2nKbQD`^_(?TdUL`FQ844xm zG?-O(yN+G(tMWhs5GVOWD1Sgmf&5w7VuNr|UXM4(7}_{^MFiAS7>7ONK@=)4c8X!X znhwgs9z+!Arxfcb$ujxRYj7&wX`7(6qi10OTJ{-k6TXlf{A#-;ezgye?ZG!ClO;I>c4hdaw@?XlD+k>tLp5Qz@TMX z2DR?XBs)Bfl~g`7?bR#jB}45M+6-Qc3-U_jz25uqJ<~baa7xkh)kUu!IAmSAJv$e6 zr50?RtFD^1ENH-I)_CB4x2Qj`_15a7QG;8LMT3~jQLxaQ>2uT;zR3-^+_U6!g6toL zn63IV<<4z;6W6c2F4sNx@PpCAYk{Q}|It);9RKf$a=IVpV9@P{G>7Dh=SXw7bhJwf zM*Eq4qFOorX;nDm@2kQwPmXALwltqXxwV5(_f%y6f@w~?Qnl6EGR=6wpxnmIS z84XTR=$5MaOaM2OpF!fr1FQH3g;?|rK6|4S6Qa|tg!vJ5LnE$xZ!W z`8Dn_NUPDE|vllc!5$&z^V*U$5q*4?>*H*hq-G%(vuhlW;t@$u)dr+|DM zA@JFKpsDI_Q_45%ihjZU+i?k5)9d)oZKjq0=8AO!jDXxTTX`e38{wlL$@xe2*{x^Oyt0;9gmOG!<>M9`B9Rk z9IKI580^?RPDkr}#)5Q^ZAETUdPUf9vTh-HVG1q-2nI4r@QwGhf253_<<#j&l6|e$ zaXH7QP|WVF}+J z>;|D(7zrlz)Aov+a7sBD5aD1i`aQ4A!#=4ehmqq|Ba6SPXl5V)u>JE&i(hJ)5}Q;o zi#CK0`o{8171C60n)c`vWu@6pP99o^mp)D6F<0;t0dJEzrND#T<7N^;%YpHF(A$^26y6{Oi8<}UhlF=F~L=eY7V%<3NmrXkw z{ds2pQCOvHVDQ9+U1=$W0a=vi0B)B7b;;5R&k$Nu$io? z82|tbdgk4ph$TX70al~+?elLCA}L8)Zvk^8sj01+r(n_7*nq2Ryfwj?ZbnOfZD&v& ze)$B~^sgl-#W-j{B>n-{Y+DkSG-`9C6^*#yyQ9aa z?hiN<$Vb+e4CL`!VBfwm-x*gVjgOc+AS!=dMy=i~vH1)`rx-}&w4ETY9Ie6+*Ex0{ zVK%6CcuiWYGU7CT??Ogp4&K`snsfn|gf}$W`GUE(S8=`kR2RBiIq-5BPM4ORS(o~N z3lo;>i9eUcKuuk+7-n#2p3n1-5{8DCH&K{S+0VpTP07G+dDaTZgXkdTnB>Pf+p4p1j>W|mXf zw1Y9RJq9wI-=f_Aw^q&|Pe@)kp+{d1D!EaFB@%BBj%RFdFZ6;B%a;%g(iGbyw#e}m z%lpvU7GY5*@PbPb-8tp1w8W@#r(4%lTrz<#SGm{E847cxk%$94NFx~mK)^m1%1D3! zmeZITh`_v*tu!=S<=+-?tc^W0@XX>wHte3#R(_VSOMi3CPM3x%;DtMZ##3AT(!#}j z<<@3VnPxEEWzu?O(&b1hP9H?$(C2)HOtwh-191j~PD-sv`>G-^t{SKS7+V{||`6hkkX3@IA~h_Wi1v z?l^AUN{{l*v_1={B`1ownUnT>bQzb+(SgiP zxJ+i7lamdSJ;fCgF{NOQs>$I7+O{`J1{~bd1IqtnTn>>?4WehR;AGgU(ZV z3^kGJkSI1(MV5J~`&*)rdLp$TNzcxAMtNPRM?T_l4Kl#7G6L-&f_}q6kTYY*Vqk;N zQ_d^`Ly6f71pU4MiH#q1S+1uGb^^qw*0CLAJwY@3UHj%cgs`-@$HlRt9n3m1{@T>a z^bfPrR?mKfw@6~auM6Zt5SVti_EF3c9}3#vd9SGoKnW3tibWs-jz5Q8< zI=fdzN}`Cfp9@!`nY4;XJ&fFoyNeC$NIl)>LL;ff3lZG=b(L4{I}9Oi&Tm(Y)qpE# zTae*x*u=D@=^-fPLtdp9*!p;}H-?7F#T3mGr=1e8@N%N0P?KD)GC4!@JQ=647G>pF zm-w3MUnKBSB$_)md3MC3S%D2dqUI1EI?lsypB4Rkcyo_y&HF=mn_&_(7Dv9lB&YH~ z>ldZZAAeQAP@R1Q>ayJGlshpD9i@QZvUH;qutatsrX=?t^Kh<1Ipu{AyO%pw(e^#5 zR@!!<*hQiEa0j+K45f{8--=_TE7l20mn29-mBm)sLxeB!Fejm%WmybTCG-FQ{C-MS zGOQ3^2$)Tbc?qO8Njk@G88r^1Ftl2iraLM;X34^$wt|tkhWQ{RG`(dz^06Jt!gx8> zShCU1^3WAUnDeX#sL5tp{n~$yihn+T?qo*$v9_J-@WQ^a?MsUVz z5w(0{2}BTaupmXy?2C+9dOf8%H}nrQa$M(VY*S5sFNa656(s7b%##3?yo*lj0Magy zBp$h}ah7`HXXy|!A* z!Fu7h-aZzjaB}@ao9IkiC=u%7Rl?k9SUlj^7!dd>9J*v;p9rVO9P$7l2_eWQl8H^I>6l(R)$!9=sTsDr(kh7cR4UOMlP2Kw*A>ejh=dFpgrKTKRNjf z45oH8H*i~xK7R&8u-`Q)MR3Lk`FxP@ybuBXDEZC;zsE;l}>}jcpe<#0x-N!SqxA)i>>{82l^w)hVPC zx1Fp%K)qxlhIXNIPn~@q6-0A#|8vnwT-bSrL(^W7B2*81qmUDMyLO;*ir7@ICyQRW zfQ#1U%DTJbb?8Nxa#VKK+V45cd)=Pzz1(jiB8OJZF-QbUYSZau=-*CNN&o_8+23pN z&7wB;+xfZ{&iABCD>(bOxi1aS9fOF~H-uNdJ)r0ED5jLZmpYx*eA`2J!H~OI4;?2B zldgW8nyAx6YyQmFV?g{hjQ$urC3W6_86je}v4 zoQdDbZ&?P`U(U~VTln`ehJ&j{kvx&_u1n!&q*$tGD03v9$ME7v)jgjvCe;A*9SBnX zMh)bX%}+Nd-=sXAp`f99 zb)H{MZ32>Sh|A7X6hRoyUW!Gpy;f?j;Dp2idKhMs42}m2x~~3Rf>fDNpTY{;35JaC`_RZO@B4(J9^T zt}RiGa=Ds%18Fx5S(wlY=l$BUfjnq`GxgzI_)rnLa_O6cx{BAaZIkIfXGc~DiZEP1 z-RgaK_eHs&2EJ63N4`#q`Rht4pT3x}N@-_Hj&h2R6Hl(0bvVy14cwO&RYlvdwdc}o zj(Ch{;?OB>Pm*llJiYW#wufmK=CfbBoEC(`NO=6%L-7tZhM4K~Q^zN85g3#F#qlW; z2{oF1ZoGXMGcT5H>bp1#5T2vLp7MRa`qtfgz{OYW>G-Z**vYa>)8lr?f+tv($>H|l zzsr^98@G#K<1@FPSxE{eR2d@+B{1G64I^SOiI`0DUM$vr# zB4`vVv`r=zJ1N*L6-O(g66=K2+8kgwfiDpThQuHTSzEEfLNZn|k}eqG`f7shP`!7G ziH>$)ZM62%4E%PZucAp2`mKkGJAR45*oMsW)$S{qx{`x=k{Cw$kYZVVUyg(X;X&K9 zA*;0_4Q2Sv0Tubu-GK^~T;BL}8DYk7r-TbRnThGuvAow3s6`jzR2~o0PUSWdlAnqD zwY>6YVZl8Jf9j9_U09%>5FB|XuzfxD$)m-F>b%N5`h1t8Pa5Z>C%FQ(Ru-+V@$7u* z%RCdP^KlKhU0)kGXFq4!`=(G45HM{{_nQwNVZt!t6dkX>I~qb@_z&g9e{8SrL-7d*k(vph=Ugh7I*t7nR36^Jpto`9SW(q3Gv|**;#^ zhaRgWc2A+nDOXVv%JB{{$ITZ1rXBeQsaOxZ)cuRw0vkkYAD~R8uR{8SG2n*yU$LUO z6G<*x=?gvAdQR72Uf6xznTK{lM}$vYoj45z<0&9+9nOC2YU8?|eeKrXXVosw2TIm3 zp6@za;lQrIOgMqd;HfP;T?*D2KP2)j__mi|i3o8Rtc$=LiyQVM7x>j~O+bpYG8cGC zlMA^xDzNBzN#0kOd7VNJgr`#LM^3x>d_{uiUVR0ZVz?}gsaql8j^!S%O|~)o5a)W|#QE)Gv|Lk?lEoB)3bc5&Qw0v1OzM`bcMMPGZ@Tz$f<-C5 z(&UtCTpzD+{dZL3>E!8h>nZVkRHQQcMm3DAHBpgj>rcA$Kda9DRpi)B*|qpp6fgk^ zf4>@b#^En~pL=eBbz=0b%q|^8ue`<9X~7LW*3Q@%ZA)WO?ak)-%m@>6>217mnjM4U zKciQvD<6uXwrKRZG1r;Lqy@!3Ufx<9x@*WZA@WquQLh}W1ee}P6v+!PQoLF|d-7`6 zyVXpX`aOGZPy-Yv#?UmlYF~I@y3)K8E~H=r+&Tbhjqob8#j7HkQnqer1*XxEEz+HS zZLSG&i9?pI z5|$z0%R-iRG34a45D?*+gjvD84PToW@Hjg+;Lc0^wlu~`859Torze87}eWFm`iRILXqjg5<-(&&l1WID>56w9JWuw90??0 zM{zzX&y)gSSVhtQx#o);f#?#Xe=k zmh&vXT@`6Gh+6U8o=CusTDZ#Yi$@kK#tctb7s=!XC$LJcqB`?oCCc#EFi_zoUh@K3 z1QK>#gz?WlEl~)vU%FsVOY=7af&PxR^<~xH4McmkEx1ic|7%T*NHFMhzrmfa{-39s z#%bM~?O1(>?wrfS(c{xDDR3a%A14WkwhZT*+&&ks#vuJ4*gxmC*k{3dj&-W{;jG|k z7LY{LNv81flS;D-Li2c~SJ#t8BlOwdJu!o3;9PigPY-~C@HwIygof`RgY}L1pfZ(w z^xIaR6(iNQnGFGs3ge_77f%{Db zfuwpH2i~0uM;QqBWZ>71ELLo*AO6oGGs;1NO$@uGlq7YMA5j>2Zlt47O`z2$(Ox)} z+&V^L&1)!D^4}!|ew*jyyy1CVWSL~EZ^Y@V_MoQ}X)k2<*^bHw16n?xW?NKr0 zZ#H!Fmkp7yKL5JJCSj)Kbe*ix#%aVZR`W&05JBbMY~`)A?-3hmL%;`O51h}eFN>6T zCT@L?*zB~g^M{HJ>n1L4J_~uxB$i8AFSq0L7Bxe9_VqU38z!B;(wu0oYok=+`G?eB z%F2zJVAq_mP8k{c4N?lz*~+ zVr_Ml=$K*wAC(*^Ri6J3lLP+V-SI82r|<<)5nH+EYh874X8Ie*v<1)nuMrz-$~`)Y zYpMtyxTn>AI=}TI1Xb^d#S+PX7mU<>^QayFbRm+2tMehUTZZhRDmr5evG|igb4VL9 z%69sxA^K|By9@M;AQ*_UwuciaW*DClR230oBO_B=DU4dbw?79nMbjd`LVAdyTM&le z$=8m=oiGm0QCFh7PWby$1kvA8iRdRR7x%BBUYtS}$kye(U#Q%B{4oFkCDx40lKX8nn#6*3I?r3sXdIP%7MUGonoXazkLM4aOB!ydqw$ z6NZKNToiwk2PloA`TYX(zp9e_t8~ygd%x(O!1q|R_=LBsK)YZyml*8_2(vFMpQdRd zsMHz#gF6aatce6$euNe2#V~Xl%Me%@%`6N=1XMFPFm&w1T`79(5vLe?yE9FdM1Vuv zCf%`{#hyXEd@v~U_kyM1uU|Rvgz$U24;2D!g@BG)A2-kr7&p zc438N;hD6ZdH1c&)31ze{-JxQTvVJYZD4S+&7Jh709nl0H>3nE*x(M zpVhhOLw)1=*`hq%vLJsc+H!NEh)-@R{bk()=#0HwsQ<>`-^H7I#{YLcn@o2I)hvM)J+>}7oVs7pR9dG8zVSY zN zuP;5*4#oV@;D(GO4qxMjCHe-}Y!#GQM(31`B<|ri@zf2+AWh#At@_;H+pc(Q&9%v% zx@AGsXn@aVQMq;y?Tua2-uhKpC>V+>it>TG(zp|kd4?-StJF`@an8S4846UE3o`3V zW!I|^EWgTMxkL0S`}*g#EBkou-&&hg=b22RptI7Gj8F*XWNvAq-LreT}$7lef(rq`$NAOecf3N*IKx`~RFaSJjK&IhSU_hhN zSfI;byw9Rbk%|l_QlLa^p~&)xx#;F`FhoW8*?(N3SWNlE5Kd;Gt};53Xs#h{ldf*; z(I;i^QpS#7%uf+%nOe`zuf#Ty{@$GP)A_5fS2KRH zl4Kv(Z@KK1je)8-#uI#U<&8Y$Bzn{ZxS@O+)J`2#E}J2dL$lBZwSy$24QQtJLrGO7 zD6`1H`-OorEy+VVhq;bODuF~!jQsk1jbQEPQ85x%Rz~iM*9#n4N_2Vy0-ewa4<;xLjD0ej+#S6i$AmHWV*HJxc_%9R&_61K3=j#rGqwnq zLtcAh6>`38vNs*NkMhTNtRE(R@yFauls{tWo8}&gqpmMqbcG*^Ia!Vcg3$Ein_MiP z*1DhUwVGV?+A4Bg7L{g{S{=c9_hjJqr7uVSHbYP#H_SnzL=00~L6TIr?t(C>SqxT9 z)>y;LtDL2JSEJ5!+6_kY|L29Q5;@j8jDvuK9Q-vT6}6)|K|knmNrHJ$PkZ|1 z)C^m&Qb0yYy6&GEId(DgKnl#2#UzB(8vtO6+w(?RU6g~};rx`lj;XAE*j&8NM|qp_ z28>8jm4vMb;|j%EV~6MmpU|j@e}!qZDL^X?S=(moI=TPdiaK4pN zYZ~ze#Mq(r@=Vp86K;Kx&xMQ8OwfStBW*X<-k|R~mPy)?vi|cj=hJLKuhwjyx}}~w z2yFOpk9$Q-sa)RG7puMMJX9`-5c-NiO2FcgaIVPydwqF;oS#7`hvjFGManRarK`B|nea&WnPJVz&JOKL`1ml*;YD0W9?IkaX%o7MBcTpWpO53TF?(uyWTyk zp=8DZ52xd?W&L0FmN81cMeQD41Q94}qMHfaVy zG=Bo+6~8PbBP~#-MKRJD{bD8M%TWcfkHSrn3H ztmO+li27{=7G)9{e<3i{uL%6zB=nEEP+5Z$k<&HAka+G3s5Qj4a&c{1oeL|qrRu3X zy70*!GHZ=wsbJjO`(z7)MM^lYjvC)s50}YB2ZWq%MZ6$~`vw=rXK>HbHGn|%t!;Om z?dclgYtL$~&-dm$8Dy4yg0(qU(nrN&+;|(Dm(APAZyBU8Md=bf$~W5)N+w5nv~IrG zn}i2eLFevS>USgxX5<&5K6SN6xDLO%njA*yUvZjweZC`<5v%jUfyPolNYvYf?7YXC z`2BNT=4Kc6GINvAw{VAzH}?{EfnrVQ|80WeKbYgHn0k&h355kVSvkf=gLozVMNK(X zR_t^KTrrv%rZz1{+Z01Y;UetOkyUt_Wq@xqon7Nl7y}w2e<{-$G4l*{2T>vOdaW-w zTQGe?HWT{h@lw`}S2X!V0i=HR*?zeDifLDb55onajWVSfXiko+C3=XP(Iv`nJ9GWO zx1DAZFm@<^LEM|bB|!0;8K7^xhw}&T;Szo~1KeiCP-P5Gf>fo+?i#k|CTfHc7YAo@ zyOOx+;Matq7ML+T#8!`so__gJGsud3aC|!?!z>yk>tt%d4`(G+Lz?14y8~ zg|~YH_!or$7!HtR3FUWe4*@0NSW+RLbnWGjHlHAlan&RQfoJ}BV9tvY)mASe2=KO5 z4&GaYr3tE{=-P{Zu&Dpfh$|}J_`tePf#IA6`v_!JaM<4+234xnKk?A&4uGV z>E1lelQ&G(i=Sqxrtv;2dz#9%9nIpo3s^2I&>fAGQ3nC=t?1B~i($#Gt^6-$uROg> z{tNaBAssF9Umn&1qQ77I3_<`Q&3mUSEX2{!I&a$crU_Y#u~!fq3T462aEf_QD=a9X z1g7JpV*^D>&{|U`5#$%aY4?~YcwY!Uc{0e}Uq z8p#iXh|${HFYin}Lk}SqgIR#7QR(zuqcJORX38F6ICTIuD;XyB5NqTv>M#I`e2{U_ zI>2KW7qJ0N78dS>`2I~B_Ft$D4%uzY1MbBX!(H1F4?VlpMQchU@?xJ2f_htbELMpu zd6pQ-9OsjIuNfv_@Ssc32REhEeR^Cco&gH@p&U{>if*W-lh#*SyHYyhILp%F`ZV!UWNeXZv!X^|N7g z*Be>&(z?u41_ixwx4Za#UmY&m5enE0K&;QdqxY|%r2NT{gph`0`=^snS~FF_L<9=x zLN%wmP>heTU~Gyp%yLov`&Gp~3Ea9arVd*+s0)B9^!qi|eVM}UYiE4OmS5!wf4u;x zodF{SsU&Wf>xNy*Q^7#3OgeQ{KVstF7sscvR6qJ*NG#K99dx>a)eX0_5yyDgZk7qb zSWJu=`|DR<#4v^1WU^}!R|#@h)MLTU*}wc>Gd6!{;3-CFSl?0YfxF>32+zZ)_jRM& z;_YfK!v1 zWu!y6=kU1VS33C%qoIO9T}428-0Q%a5n2(w6k1VDLA(=jejoC2?0e&QA~+^xV9fj zpgqFBLM{9Ly5+M-;?p)=wvJ!~p*}fv?W|H9BVQ#taV@)C%1J=S#eJ;yJPQd?C)52Z z`vHqbUbsjETVX&=qSm-;=W7K@-=;m1NvVvc;xcb;^xRy4#%T2iR4d<@mI=pXL5uG`Js2uD0j}OOM|B7U zpbmyABaKmi+aaCCKYBP&pDMP9I2YzlqD>x3R2Qs1!-NlTpv}szfG7 zMZMd-@=Aqv)whlLMs2@eF~9;n{efx^xf|BuSx5}aJfK=^dmetB@|6z~!@9_+tmW?* z3M=)7^Fxb|@UC8YiO&#)`TxlnYrc1ry^Fc0Rdcv^f9jeYQQi~zNCYEevgY17nG?LB z(Fq5%#k4%zLYB7~GEiWnOOOb`$^rnNLdT^MNV5#{KGF`OV|t`BYmqEvI7q}@xF3Wk zm_5_UkF8N`84NDje@-BsJ~)T>;7+f583S@}4*k}3M@{O`!pZV*Du3v&v{Kv5g`g?3T0YbiCq5ggj zhc5AqM3Sv)!ddsVe#e^9ml+p=abph1R*)o63h|`j^cKZAG!$ohQ*M|?`?B}p`{$PC z-yww)^vBCy#MIx;2cO0!xS`7rCffKpkeb{6l%T4R8xt8> zcXwgvo|{r(@b9{SbkExR{m|$_a_Gy20a?w`(P<^XRC!8ikzyN(VoKZ58N)R%k_Exz zAQ%I*9YqYAkv8Lquo)z}wYB;TX)bQExnd;SVBj{w?KuG#`XQS00L4T5La5kbQ0UQW;bs))bx%43H+`d1-TPDk!cQ5)K zeC&IkHjsbm;LCeatz5|FeZ^_m9~FbJJ&c5o*8!DH1z`Tuw!0INMK|n|9P@0KhN2431%(h62x`GhK_1nKQNV_FU@N`aJ-&9^zz3)TF~1oNXR|jd@s5(2+#7i6L|9s*|`?D@`MX$9htPND${GbeUBwVX%%#<;^A4TnN(|J#F%GTd*P) z;c-SicYh|3iFWh5zPLT+@%6F_#Rbo=M#F&Zd5SQYBRmZPBkX}2m^d~D1ya$lFPuk; z`_om$*D)0N)eO(-beb_3gq(S7`K~)L&gsGbsm%0~(LA6%x3{jqWfv0iUh3@rTSPpe zn&9Y>|BVin1^6s;`@q1^)cd*)Xm;URA88}2L+vUAX!M1OL_6(L=bWh+q z7Ol;Ren1syAFSrWzaC(IwB^0@v~1b*XnFS!`u~Ok;3j`F7w7a?g(U!}0YZ#7Pp1~~ zh0Z>`eKN^5=YE;OTSrlle{WcdhqH@JO@Q=D;SJJ}2-8T$mp(JG7sFELA?J6we};Yv zj2@&cF%Gp(*3&;oPGTHd&P#e~i!&LGd3umE{-7_Tb=)Qo;&n8%!~jj6?aaQvA2ug# zL%IDt*S{5&6KJd{b-5TRNt`D=iS%Yzt7T{Ylep$v@%-bSA%HWA0^nSWTuiGyv}IImaI&PHH**+1{@;wentI%Whd)D+sObN5_M`P zq^$YuFw&eeJ+&pO3GG*VZzd_W6+vq5M!4@!_e z+8&51q%38}j;xR3%rS8jl(*;U-q1o#2|kQypl-_P5mG z#1iX`#X$Xq#9JMPTE2>i(&F1Wl18SEv65bcD8TPaJHt9YCin>71`F^yN?77|sW*zE zGl)HyBJvnU5(A%BVlG?sV*rwaw2o&usI*aXv-WPf3=HAMRp6?IL+=gjZ`hZ3;;NJv zL~!j-5$#(N$)59{ zvE+9#-~=hXKayOHvRvhm*5Ga z$vgU>WwG;nal6a)tYrH!B-nq5&f`S7eIfP6$N4roC|9e$FZ3ytK1*nln=g$qbT_%)q0G0m%$39vBkm*}=D>9}HB`45dCe=AM2EcmbAOOcwAtRH|ey%$3ZD zYlg_VcJ(y-jF!oR5*gPKZs6=&!!))szK9LGZ1+n>%aBVm<4XkMFKD0oLmmbh9(^1n9f|i_aF}=~5pS11MgChJ+7D3pr#$p&6Y~pyy>vaxVav#l z(}@F`P0A!NH<3uABI<~N;JT=iFRX_@UBZ3smP6uFFOrV5-V#4oq7Ic!;*HtA_OA`X z;Do>?@>`gRiSjrNGXf%LGZE(HyVSHkmZ&xq=acit4~s!$m};EdsfX*ii@{D@7moMV zsg?&a#i6ZO=j(?Da#XV@qiA>jA9HUR7WLY-5C353kQ4-j0Yp+lM8JR<6zQ}m=}zf} z8M>vVLpoHDMoB?hBqT+;5dmo&X5Japy>)N*{p|aB{vX~i9DL$nX02DB_67YiBvPH0@H_pj%>&hQq$F z`q%2EYxPcFk9c|AanPu?ON=1$-+CeJ^dO2l%fvbh*fO(?IREM>CN!bHduOCtD)QV_F>fU zX=EZ~>P;9ehZ5pc(k4D|1evgHAv~xrow`Mgb>QcPLw~^N3p&J~CdH+6)stYx5}1U- z4$3|ASzx&pI z3Xa0O%)moJqy^7q3b+yQX(0Znq-3hr4uy z9PyZ3i5;p1qV+(@bR(l;q--5cyk96rrZLOZfPasw89WGIfI=S)*PEdp3JSJa_KR4y z@ATDgHDD!Tx zHAcVvnd8cW0m?|D4-L5vU#D@@%;Z^*XKX@t(7HZ?Q8bk-_`i4x&9Zt`v++$6Fi{vbuJD6vZWUxVfUzp|%f zt=h`!o3PV~Q;nAnZy$eE)6;8KAq16$xpM0_MijCW3SE)O?6|Xz)%TR{|4mV?bboPX zVe*Lw3iaIcy4$P|FY?v?T6)*g+T5$i&rLF3c>P?|wxt=eQlJ0cw)=nuKsg!xDJQ?6 z)9TsG87xxhz7mK5Ggm^!iI4~4pM9bax_%i$N@LX_8P=n}45bhegwEOVPTqXueT-E) zXTM%a%cvkjv@-qY_KfmUnh2Zil49X`GYz7BT_5c|XH^-3jCgEbyFB2!yhn~;E=IUC` zkCf_KN@k^)4@+`F&V-&!g>6%5sN$jns&hIQA3r$tUPI#5=W&%^!?!0830(i~C)BPR z_nSQ$Ka>7t#*f1#VBVD3OnE{VAS*b}%Z`7W$t<%pr#nHw@P_9IbODRK8qtZ_#cS0C zDUTbil%oR)wq;RkNJZhD#G4s%X`!m-a(LAR|5ok1WU5go5)c4HdG%idnlv}d)=}7Xc<%&VI?@;)>TwM=J#~1eO_4z) zcK||t>~_5c3}An2x#PJXk*`V(FX+&I65J4uNvfkkPu#PgEKTvxr0sX}E>zacMYGo5*S@(&Qlb5; zXn;+N7Z!9 zNw@Y}NHKwwkY8x&tjXa;qiHfE*Fl@>dmr{FG5`U<$q+0B%zmfDUDNa$%$DR`TsoEl z8H@>J=zU1NIM4;`H0RtSHskxOXj{xjLVw*3<^dl6Q;&oE4>nO=WDIAN;mNsc>i5ZN z5sCSic}!=U-Ady&+8CTL(wr3vv@vMckEz^CIQp_>XtA-sNph3nVaE}uF#GIhS==JB z`n@rzoFjg`lgnm~CEvy%PY1YWT7vDcwR%Li)Y~E$g91p*Y;-$iS@_#Xs>+sXQ>(G% zg~lz+O2o?3hm(hxfT91*7^F#e-zEUnh(Gz&l3RgWz`V5eI$TTPR0pIF5Z?D7T?Fgd zcVG8B-c8z9^4{KBiuCwxXPe#>2atALN{0TyU12$STqX%p5NK+}AwW933!l{vWK_QF z7G(HnqOyV_QtV3{v!d8%9Jr*6Nx^R~Xmo@%M;0FUtgtOc=?U($ILbb*h?anP;y3Zz zDyiL8wH?QXGrA!4+rd>6{b++%u;k0amL>bhf?1Ey1xu_im;3zJ|-IpVRP%+OBhD!7(?^RQwl)?z_xHq>D6En{5J5=0PvBCO=T4ZZ)z`>npsD=Pel%kXxTfo3S*tX5%MCdBy!oRB+uaY4OG4xc4bxhW{GO>e%OkMj3;M$wZvNL}KwY7@dN>XJ+3J(xzUs zWW1oc69LXuuH%f=tQ>mxOuA8EJx{nlvx=~+m+|VJ)l_pDGRAXFd3!OQU)Sc&Wm)Dm z3G6ZP9+gZAE+mGUxEFnArYyZdf3Ati&y6nShuGziD~ zU#w3R4D#xIf4CSuD>gW|H*CK5WiYPO_|79Omrmtm@TcONR&A%8qbaAc`bKsdgeaX zn$F|8&(c;ewi&~oK6!bgi>b7YJM2>`j@*^uN;`=n>t9m=6wm`?{^m--n_xJMoRJ1` zb2H&q7|BVIr6(DTy$+C@pYX9vB-){KSUWoUc;jen1}UcS=`3HTK&OI$OXrCYSWS5m zU=f={SG(Ca@A6e}Zy-+hPb;QgO_KUJeyLz6MK(ovL9QxWO&9d=Sk-dv5rfF{$;W9a z{g%zTFmJ3rk2ThhKMi3|=4zPb%zw4?+CqbAw}05T=w^3~{5C`#Q`|(C>m8VK{lRoS zUMHDA&Cpr+)^c|-2%erw=(Ijs*?4R2(Zoy0`rpK|-rvAy!qS6!p^$*U$OONQoTa()z`oJyMCV|pVwjj8QG|KyLf@ZB78@qaO5F>A?TchC``BHtyK3Lz^87Wz{+aAGi$wrRua}ohmAj*va z^|hR?DRM9$bA|m{(L62dFwh1;Txvd8pq!$2oYiBQgrxtF?WVQ^*2Q%|sk|L;U7KPz zq^jn5DJ-!7xu4;xk9SpJrTpvo7`DLV2)L^nIZ@W*&FNZ~MHWL-&>#S8@Lycj?fKvB zII5D~)i&f5Wc%pI%euoFevyW4TDd?__8(F%8VDnPE zcQ2vY;|+4|{3jBBxr4>%d|Um4wP6%>_;A*5I^P|3ESLcWmh8aEe40^?nt&j&`!Pl< z%A8VNAfXAnAB}cRZ={NPq045f0!&CYL5Pr_0rKUtL%L|;vwa0O7$aw)>*M0;WLJh{BWg&2?b@DnZv*dmHUyGOy8QK zLYPSD;aXvYs$b_X98JzQ*d&D_caO;6kVbpr2O)xcX5+8Gk?f>R!_aH6sHrl;Hbpv- z>p@|BVrN~O!J{u;xi)_!*%%Pdl}7a<=K@JFVgH0#gjkLM1MJDq!{P`23OR!IEHa0z3HFQerE4^VFO9pA;<{HTK@ z`)9w^LGbF=q~9a#wfDs}FAO8$faD{D`pFDKoA0{1-cq6xDul+XMDSie68X`=B6)ea z|5-)@&9j4T)5LO(_Q|ob{m)ZnJdbJUjyA0yXTw(THiHgZb^9W3#@Vr zbRxM0AXMb22PRm+3e-@q--J=wv_^7d)EzeMu6@jKRh2>_0vEXU+w~5P`zAfMj%n|E z;haA3Biz_p%J3zrA0Fd{>AA|ge5{mUk|XD}AiRBh9ib56n}c8(GQyn`=DH9y2Q7Ui93~ZGxJucGt#KSxIH{Y(1uf~0-a7HV>z_z z)ZK36+=wAi%tF|;V3aE#u=~rZu*=1ZYK6vi#Lo6G2>UL+ze3Iq=n|nuj~q_&>+@AM zb;sEf2pM=Xlv5S`pZseN4f{_X%JYYZE;Ocyny*fTkee!{Lrtrwm$%hAo=C(t?6}=2 zyb~vb`WQRDQ#cfhY4+f-bqBa03?POL5^%Gh&yV-fXn0ChXz~^=?67!vxV@bDD0j3SCXRANp(N4~#I*cM zHCxDfgZ_D(?pSHq>@B*z^>g&~=JlBf@KQax2s~KW??Zd2^c1lPwV9q`+nDbJFHFT% z-ua9;!K5;7h)5sr+sgUOz`w<=<@dIN7!OT29f7!NgGk@>|#iOh3is1&c{ zD5?s+(T~xzp_`ZcdQ8P0tF1{lj}Blli(=>S{s>@+S55+0Zn`@W?ZU4*cbqoRhH}ge z{}HCyq_=`a3M63;J}|@&0dHTp;Yf@r=@aJK%-1O^Kus8(h0miWDxVz=q{uU3 z<8Q(SIiMUgGDq#C3>-SG_m7Kh*T8wI#SL1eyAcfRMvbHE{lm~6G(>qhm*?g%w2ybw zq^@s^x1Q=(0{H*@P+oD^8oUYq_E66Koa9-MAb>#qN0Lu`t0#XWWNS_A^%M9#|OJ5V^h67j)zk}Xpbt9(2J7Wi@ATi=XyZAS~ zkgc%9t7TOSShh@kK#F(_ooRBlS!Zh+DchkE$VD(JKL#IAS&j){y2=^%7ePo9iCY|) zx{FJ1z|LgeG>b--+e|v0$S5+OwNpPx5)TXK--pj5v{(Fqe$$PS%-2qinO^Gx&_JQi zvfS=)ZU|A%L0R701Wwbor>JE0VAo6hsBoUk8wS;7TUGFenMehkT}|8#Q_!&#$(-_M+UR7%7e}p~f#P z4WLV?Hdpsu#n4hUO2nnjcO@UHwOvRo`Y@DCHe;CA#PIQ~8!e(awKj~lPnmhD;K4K* z@D?^vqmQ|B9lY#4o670nwzt&Z(lTM5en`qN)mUs=KREJ9t9hfwZq5RQ2pb~Z39|+= z9f@sF=l|PdCcVo0tX7;v=g);m#ETd{5;5VVc0&PftdShZFTFf6tRb^Mm``U$#Li<_ zDtHV+&I5N}FaN5lv2r&7sLNhVv27z9#e_vP+Ao!z6(gO%+`*;F ze2lVRt-v7uv};~Ol`E7$U+T@+FV~6g6b{DRLTAm;ox*!@n5-GMTQm}R3ri6pS7li0 zW7_?X{q66Sv%Y`RsgugtFsX?Z1d2PbQ|^r6)opvA;h@-?CiV^|<}|r^|K)Wom1p#o zYpYTotm$4vqK=r(d62MNlDvoYwds)QQgE zx$gwlxRLqQd24?~Km9tGF2PP@iSInjXv*fs@lnQASP1PIRL~)<$*9YeM9mmB+~?JK zmu_(Ng{5gYIhGT-6X+xM19|fQwLs{QST(a-B>ck94D94+y9E4(sr1AyI50*8{?AOM zk0}2@GrWN~(EjPJIDr$p%*jEUY-gW#mU94l&2a2pS!(0%jHZ~+7rJ(g2J)Md<#t`E zEd~nJ#WW+5D=ps?QPqbAbnGsT{-?Ra9~})vwdaWTnu`XJ{G@_=O9w9$0m^eC&#kV^ ziTN~&%TIb%Y~Kpuo2awhnpKn>$P)vZ&Ae_U{(!@^`smqKI`J^SPU`f>tz}bMVI=}s zw1=MgEIw_ykte_FCIcT1(&v%gc0H^r;MKx1I!3Y?<^U(n8O;SGsHs9)31DlXgCaQSEsV+`=0?3PcW~Y=SlYKs?xuk=|;mguq z(Z-^TP7z+x(ULUdn=QfTk`r~+vALH0$OLtEk|H4U!Kz-+wx#TwND`r0Pj3>@`5V#3 z!YU#1>^bUj^b)cQ5_=C0`y75NuX(~$+Qzl?yK~dOtM(FcB$)lmCWF}xH~R-Yu0M#W ze@OM7SloW5-Uo@7{y+n>zJYg>9S5g8PciI*GUl2zt2Lw17x_uHf>R<5u&L!LuZZAD zlFEMU@4l&=LaVM^bGZ9ZPpweT5@~OUiG0T1xe9bu>)kI*brWqRhMs+lmV5WuX*-|i z!z(yX*V#{95ru-!y%DNU)fY#aqBthMI>|4mffKzD!97>9AtB)xAAVLidLjnEE>0ah zq2n4DD5bDJo@oyPzj@(s>65#bgSC&dBeC+CFOOcgw_Isd>JY4T@axEkwei0G(y?jS z3#;_aZidUo2#eMxnDfP#{jHgnQ{t{no<|efAF^TZG?1Q0HsJ2|ht+(*F$$Nn&kM1z zbkC=gB3MtIi0^-SO?r@Ncy+HM){AtqV9FZ@SHsX{{jes*OO9`zG3LQ#jWm~iPmYCn zHr}9LIxQHB^U{Z)Ej|$Z{+by^IM9))uPKrAJtp@0bLFR#n$<2`a~tl=&v2l(gJbyw zVrsGrM*x+t{gpVFBI_=#L`v1lO8(WCd6Wr%FSCEITZ+lT*A94ok%l~{MDjfx7fD{B zgx>PwRXdS}e4B!;h5XP-mH#e-z-DVQ*XrMs-4z?UN zXaDe1JqadU$%{`lBAfB@#QyQwGdx{>mm%4#zCl{C+bv*F&filfx;^{0za=s(IT2w{ zJ{yvG>C*c3!&bsXe4mgix7k~3Z?9m9>S)L9EQ-K@dF1vQ*Lq6HwW&i4yA=fX=@{>* zd$L;>l7;GsAWwACgWu(scH0<9gABmVpwylwPVcULP!&eb@gTsfc<;27b?t?>mc{eP zJh^qR9mh6owlM+Xz;uuV)CyO|7d5mr;P>(vIt^XG%A5!wee9^AOldMoeH-`m=nS-r z@a~f!E*}wm*zv{<NJWb(%zdtVJDZn8t-)C1jf?0z5lCczq$|-$X;an*t^#(U**D8?)>nWHrbV|>-U^a zr(^i;@;`3LqSL)VUlV0KchPFm({FmsGTduVD~2uBp4BX_4q^eD;~XXUo9ue?%KPl!#|&IDWfM4*hbKDkoa=oo=0_AD8Z-ayiA*eC-BL{9 zK)zfUHnu>#o@#-*A09S2NfX++cyvhRosAWI_JCFet*U2T(T(08>DNk&KD@%;sO8>9 zBQiC|Euy@G?zbT#97;7I*J@sWkguJ_TpgMP$Z=K;j2Ke!>1D=jh_)`KF_t=Oy-~tzdAjv*#X+{%yfdg#76S9KS{$el zLm7#v2?Ts_4D}uPZrqxMHsgT8BWV{ChisY%#~v%mEE?xL!pEf z^6e{eahg@}#+V3`7iASe_dbC&3FBM_kM;hGobqQ6r;9xy`Hz7MLj=KUXmVC^QBmUa zj+JbYc*C9BdCz0Gxv8oai`Jgk?X9krG~w}lDs82)yl`u|-r-Xj923?0nEIgk8$#sk zSPDA{h2I~LZ$@Z45SW>Q*g~r(TESo@hdLG2?pYChs2JwhW*toDP?8Bhb2saK;k_2Yot5&H`!M`<-Z*!% zr=!*QE;?S@_)Qs^+JuWk@-No=@;zpV_-ZOv-a%ki=NnF@ixK$pQU~2SeXk&bD1y~J zW@aD*qReEyImzFe3>&3Cpf(T~C?!5>`83xK{_w)V;kx##D#p;4evnVAYsz-IfG4nd zuI|ds3W>07B^{M#+iTeSQn@+;5A{CPax|dso@%&$BU|@v3W~RMZ{*2pcUYt*wKrl> z2!-w1SigN-&A+uY1Tt|<mVl}g7eG!(~XJYkI z%aL-ruUq=~P`f@V4OBYGhw7#|>@9YD@|s=ysscsi(SAuzq>szyI4axh_-L)}BQ6qv zx_I1tzBtFR6RX+a6=-ICc(~9$?EHD+BM0-1JZmRPcr|H3rVT#J8D>8stg=kYJc@;B zKPr=8CSQ7I&O|?^z-4kzR+>PqZ?I`G&5S$*km;Dkz$QcK+K-$XHo~~rQ0C@kGaUg5 zJT5i|_7%L@=rk5rJiDUAbp7S7=Th zX@haD+s4xWO;GzWMHcl%2eorWY(;5Fm@xH}sShZMpVS$I7UK(G@#Gc_9C7Dw^o(P(`CW&efDT<^XZoU}Siar(`S{P;7ReA= zWcK~MJN7l^ukOjomEqY0$xsGT){G&BAQh$>yql>^QrEcQ)xk@E&eK3&TUrggOT(^a3@iqYJ z3k001V`b`t;uZT9YmwCu@8dA~A=c7YO#s%N?u41{Yp`i#DX!f)ojNb7b5?8ewu6s? zHn`KRHr1xV;@+cNPcMGa>B_`CIx@WSlnlrxRw8|SHQP8Z{|?A??B!ltt&{ayTcTE` zm2bTZG5?ttE*7;lJlOP;?&TZcn(Rv%N@oG>wf@*U`AoYD4>nMrjZr}_>k&=)W2nya z?>^Lx?>Kzy4odcJ_&U3}{!Ujhs?mF<`IuuGzJM!pG85Hv^8rdHeAVb`R0IGO0)0*7 zi>3Ea37i;7X6SWSee2l}(@6hw2D{YmbGf6MUVJ+7EOs@|s`bNTn}ZE?->i!=hKjdU zXxc*YH10-A?Kx`yn1$3b#@v1_X?zmTB-L#G``YX!l{FfqnJJ}`j_0QKv+Nh7Y$}eM zRoRE_PFC3~_{1t-UPzCdib`_Lv>;Qn+Vt0) zNsaa~|KCb4>i#0V_#>=cSbm7fnS?iRHj5W9^}n;@y*5QfHGbpO3A(L>6xxS6skVbL z`Tf}hUdfoS))ga@O@F+w+DiO{FSMbDt8z^yS3x)N@LV5Qt;Fmm(qf@4oLB^cY>h~% z`aGh=2U|X#nQMo5tb(sll8x&9T7AF!-qdH~H2yF3 zze%uRcGP0+y(bU8U?_-<+@YZOS#*V%%+w^A@1jG`oeysS73GmkU>6i73o9}&7bOYK zSmgYfGquHtFj(!(#I-l<#a3ajC(UsHJaYPHSY!aSD6O>8RHxh)uRYCCWp{-M29wKw zkg;g5WokeC`Palr1=s=szw@x>AIf*lC&9~ZDFJgXa$jgGi%m9|6^+WcHgP+t1T?x* zHSFf3V)kejb2{(jb!R+80^~$Zh7#Fgv7tg5UwShjRhisrSGNC{kp^Gz=VkIuHx%}*v*a89Sl963Xt+Truak*Cj-2Z+)B$OoIvwU zGBA6M!;6!h8Kt8Vho!g4bMAX2^xMn2Ir3UeVw$@a!x`EG})f-bn z9-zi>$N%vbz5KT_%Kw?YirL|fnAu0n9DYXKA}N1Z?5;o`KtXz8W)jnkoy8(GN}LoU zO?3sRxU;0Q!nmsTmBKg#?ZW)W);oTl=%OwM&c_h#(PVoxUO3UGy4dW2dSLd%cBQ}V zyt_`7i>Ad|z{;eWS<%Rcgu06>i@sK6bUMFgqf>zO@2a}KX?~!QRm$pHq@bE~K!*cq zwZ_lJm;;kGT^pS`H@nx0uY=ePyLy5LT8UFkZlJSKY+f-Hk8(}o0r?^nZO1Ql3vGFd zA~4AtJNXqw_kd>;i5F)U+cUs=C}xC9<%PGp)ok9lYjZ)+LX)|e3q-)vk6;;=u8%2O zeCjnuq&YekgP@P@mr3DHYp-nyk2Jir9QCwb9glu-dg=P`we1XLg!=W;^*MW8+I-Ub%VUde z#aivpy|(o}y?Yfl2!`!qi0KvrmH&@A?*H63qDf>3V) zGT8#hIz>GVP`TlDwqo5yLy{Qc$w6p5=7QD#hW!}e|HxY7JVya`nO#2<#wx=Z;G^WK?Lhu>e-IlW0E_{jzuTF7rZMo(b!V<`6_NCNY)`&I+7FrsJQb^?)UdPt@bYajNQv07-RIN+^)eOSRIBCbhkAb zD#3x}TCJWd<^^<`i|!pPjYLA!Lry7#Ia?3KVZ)BOw1$Te^vI4TSq`_8MLztnnkJWEBUR%8 zqFuL4$dqmlb~-9ulLCqCMbz4@R+Shwl4^W) z9-eFrtNM*0Auq3am2k&zBRyb6yGY$*d_A?UL#j#mX&;kBu{UFW(|6%)STd{>;{Wl% zKF3RzYpO6)4Zqk`5aj>>*>jJCS#ic5UEeb!H-djs&xs;J^p!l^<~v?~SkS=$NhLaV zQ7f5A(#NW?-G1E@cFmj%$*NGhIn&MgmB?y&@sS~|$N+ZK1Gh_ygMzwsqIL$mA3Oay zQlu0;vQ2=T? z&jy}1aheU{D!MxbJ?B0;dyl90KW#-(X1UNZE7(($0C%L4$km;iO^qdl@7@3S81MX0 znw1`F&VwZUH;W9(*ke1<+E#pQCg|9^lJyTLE!5N?RPYK%N^fQ1()XPcNzKJ?J15D$ zn4OdHL#6x#K|A8~c%5-*dV%JxB8zO}-*h$_FcVx4`b~J54<@yF>HIcyd@pW308>D$zw-EHUj_X;aPRv@^9kK7GvEV4aXmO7gWmM}9wcu0 zk<#(ioy`Bp2rV$QQnFDL>k-#G%g#&qQ^B5HkQvB*8|h3i9@Dv*ryNV>CDC(GzAO!+ za2OEkx@+(@OgUwe&7G`CED`VqO1GtpM*Q($2#T*QWPq->zqhu0>1rfcd>XmAD{uC` z$O9)1_ zCgd6ct^w)sb^-%co^GQ4)8xyu{TvFfWUXJV@QTzrq#OoRiwx?bF@?Qh+gpjIHZSDM z+#2VdU!uL&k06mzVxjCM4Go!Yad}9Q=%ojT_m{Y$klvxkWBXH*SgRu6YD^dS1&n7c zUe@?2tCI^a>)GMp`sv9C;8MtVLS2_L~3A{u3@ zd%EcLwJ2{P$v+V<7>SNaAYuIkpf+1(HYs+t~x@k?JFd`|C(^`Krcs9_jR&TVO3H>ge$>92H&Hj+($ z;+oQzkWpaDV-dom;XPFzZUNVerM?0t1iJ*i_-s z*L*oMLUxBlX5@(7f>}duTXZ~*{rL__0Kf-W00y*`7QU+#03IL;qB@byjOHLR*kzn) z6}>k;lq?n(mN#eHBB>dW3|A;~WW(6x({Yk+FUa{Za#eXAZjC4;bG;aZ~|0k++mqXTaqXpQ$^=CORw%F52Ap? zOUr%PxISGtMAU_MeD>pJJ&$&jm5T1>uePqx5P@=!XA;qN*%CG~6&#y`EA zQvWhPR!?-$$u%>S;9BK0Q~-c;^HGpfLTu3&1#vo}c7jK;URo#Rfi21wnoWgevs%a)~iClY8?%1W7x4sWXx)xk_$>3lYtWnJ~F3+bzNgc*dADG&XHU?jM zR53P)4_vdRl3zTSQ(gcYgQz}Tt1W*GhIr~dN3pANV0NumxP37H_lqE zi-ysLV;qT3V~y#kY4OSF`Q90pNiXx+9%7LehUZG}Y-b?7DW)$I_>Qk1hn!R!SiNkz zUfah@R-m@@aNPOAfb+7NHl3;oI&zNMiM(!XZ2xm}(d@42 zFd_9lZ~J=%siAyb><*zvvZu4Wgs(2=;yw%d`{kX^f>|p5gK>*uU;;XOf^XNbNI$Ji zT_c@v2S8Drrw=|V%mpMA9}?S zr+G133L*}#YHz6(mB&`v^p)>x(QVt8x2wDV_x*{(ko-yBeHED*e#&GQvhS@4g^pG% zHl{vkpr7o466idpHL>zh&>(~(&xI`Pd;`IZY{~l1k<3lH&J+OTvEi&P7?O@41qHoy z<9&n!Ss?Wu=g>VAJc&tFJx!WjeYn#xFG2c6NP3FI>nm!1RW*qD+YQ($gwqeY$v<&~ z-5dl!-3a`I5xI}WxmAP^jXICH_@jBiHSnY-MO1a*Y!bSrjc9a6o|W5~7R#0}r_t_u zKb;XJOFV8RH! z$+}VR8!z`mK4{mWM7L&N9R{)Y)}ae%l8Nn2EHhJ^Sv(_Pe{O9G6(jVIHLbqVh9bck z0U8^0O?%cvCQcsHqqxn2?K=~EE1-rY^j6aHY{&&9j)FiF8;HdI;$F4NBdtR}aHQlt zf_cN6|FagN(h+{1oNP#Eu8JO0MQoDV%L}^5_59nclECMyZpXv(qpWE5GwPx0VJp~2GPoe8`l6UW8o#-H&r``$B-x!*iQDnk7~rVjd8bPm{fL*;vHI=_!JOfms96RN;;B7}&Y5Sksn* zhRpyG?cxCq{39B?_#^d)N5Jt3LyL@V6cuY=Zq@ws2-OwSk*xKB$+YN=5_CQ~+Psun zy%oCuAl|MV@ebVY?d|y3tIoX{Vyu%oip1ld@_GL(66OE$OF|mexNh7Nd7V0sfaf}O z(No9u$NtqlnIc=S6~2B2us7v-Y!Y9mGGMWb_VJSlx^}~e#-*A3DSmA|7Y5+ZdNhZt zYg798pT=Qk7tCV-ZoT0&2va%~Iu{{57CdKs_466^NbxU{lMxa~0wDU<7k{G|KHkoM zl+FB2C#&{Jto`tnxCqDRA*SgT_HW|TOtApZ;4Ipss(%uh`>`qnW)0^gpDBTz;lMt? zy_hW>!3Z%34_2A)@FVAUuao(bDI1E7^1MqR_(VERfY#pg0F>VX!G$3xG%wARWWzqO z9nau5WkB|sP8~~}HoKK^jZn%52J%XY&WY zK@4F2kLz~d57OwA?jBg410qlONp86{dt4MXZ)~|#|IbrpXxE%=U$CAA1&9v)*#s}Jfj}% z42)t=%k%bBxoRDCc2{Q(WC0CC2rOossTS5oXPi7Z?7XM%F!gRHcr{hHz`)bg2^aB+m5N# zB*idYh$G#g!SS886CVBT2;ZNey?c_h|J5IxDdtk{O;~_H)H1Jv)k||ws4z2T?a7z3 zAYYi=JZ&Qp%xzKR>DyUEfq!P2J8NG^5Bi);V9O#iR3mWNI`mg${vUnj=!efBH8E7N z-+X3(3-$GO>K$I59*cVY|GMN@JsU$KD0cn-Ak+gx5a;SlBiKML8U}jZbGHwe~5+ zoO7y|Y32-SVM4x0|6LNuko#3_Oe~TsRuW zMY~K6LbLG&unN;cZy#}H;nZI;&sLaty;49EB8Q2nKlXnSq+&a#e(e8pFFKrvcorak zu)q}3yO3qbG}MlSr5@cGf!t&rf2^IY^x;D4b(^)k>W9Cc6<|Ed`2QY-FeKF#gtznG zWsn~qQ(2r4^N{`^<{`!d%?lL`W!yVanON(3s?!pRyJ?x(|3dm*mbRr+mR31NF*~HP=x@d}mxsbt3~vgLo~~HX0vcuX7J;4xa+F5*X#vdxq5BrlhtX z@f?8Q3pG#PJPuGm``CM-uQ5AVnLFmAP@jIMw;#v=-G+17ZHiQhTXO^($6vQvq_@Eh z-iL?fNe42K?a>@Yz4aBlkIN`bQ205ork6b`Yp<0YGMYNi;i#dGw!16IAp6DDD5CJ| zZod09s)2AD#8!}lHv&HD{TUaeAb3o%BhA2{syD;n6(CLHca1cMOg%n+@c`L3RxB%doo1M@X?(jt46K zYmX{C?QS2l(=Arrr;&eid6y8L?!o6sOm*svDklhzmwXn4C#nKR`$OR})Q<@>qF}pf zF0n6Ui@EDBGrtk9Gu~7{SU!%R~9izmjjJm5T{$(W7 z?qs-;;z#|@vz;B6(lUpRBJKC}Jb0AH9u;xaaDFEITn%1-V_6%U{)R8**Ruu85xB}9 zP8Kx{0UD89*VRjzTLMXG9!OmSw7CJPY?TYdS zWa19X@)RQ=Ho}NIo9*C?tGGz|qjsf!@I}-nNod~^Stju5B5vVMA-e4MW-%g~p3Kly zh8?KWQ?qsGLSUm%xxGB%OZ+`GHQP(W8BcL58!lkZji^KjR!NVtY-gzOYSRVk4h&VO zOm%8Pik$=oJG_s5=ec8k&%dJYab1jRr}={r#3l8a!~Rknmy-*^YvcZw$K%?`?AMuC z(RlK%?cTL+nAtS$eOSWDwBi-`mWmh8j?XqeJ=}W)I6lxCu$(lQZIE#4UGKldwc+M{ z{8d=2brOs0jBS&5Chs@ab1b;AlhZr9tn>Rx5OV#3JAYSCxLZF!Xz$mM`_oVQ>${L}pBWSjbuwd>A4~hGt9#BJk}Ha7)pW!| zuDT2NIr4rni;7^^G-3mwSyDMrd^9Wt$?|GU@+t_88G)paXS*X*Z0Au7&zx_gOyU;trjTPBJ<2V}u}P*78oVYxVOdYkI6z^a`8xgR9Sk zJqt@l9u@P>+~<5&nOMc3|MJcFR(COemsa#6#ANVBh)>Mlm2LjX^-#aKUgl4(2Z!e> zxM2>`*nB!k*s1c~rifDI#c5nZXAXY1FF0})X3|MKK6Kh$@1OLgEB+(DURGo8Ur!Jp z#uF)oRhLYq!U3F92--C<-)4Y~=fM!GE`VZ{H+g!wtg$ohg%mR2O8LFcL^6-w!!s;q zy&lra@g16%?}a_rRSu7M{4Xip!sQL8K7Sdh*P7aJLoF$(2*Ew>o)Kam&?9v7uTDIh z2+Q<2+y`x3&V<1`y%0cadxjT4xzp>tbJ0%Ths2^n@8`v$*G7}q={v2RcuBD z^p)m$FkGUzTZ@NK+o-4^Yt1ObkrmFh=Qy?WX}72kvEjV11K$FtCX25BkGuE&YjWSV zg`W_5FDjiN(kv7e1w|lqsZtaW1cFpS5Co}0LhntG-g{9&I!LbqilB5Up*N{g0)}uC zbgiw+wf5QP+;iU_-cSC6JhO~3zjKbkm^Q}sMmv|BZ7{!RScPm#Y)~qfqrC3f;%epa z$g2s;F5=G7<=Xwriy_&B7ab~T97;aJ{ z8ohpGmNd}7E}nGr^?gY=J^@?V+t-=RddidxG@Y=_j;NE9R|Gd28$sNa${9(W)U%$6 zL`-k9I_akB2t^DhmKB83Rgx%z&T-{k;`L5f!e*9b^!LoNMuo##U&31WSFhVHXoQgT z1-YYECW-|S>(#P%ZldMkfG-{sD<96E}rrV{8m$CSF|GG z7EV|5V6Hth*ihmj8ugXua&zS)g(K9|fk&Pgv!X-%22m&qL4G0GRMy2b<&99#a< zmq|7+7e@~$ff<7n&;#Ept$@bo0_dG0(Y>|1_D6St4PzJEvGGseE(??wQu%&)o}1e6 z1l_m=T%aElN%a`s+ZXGHIO`rq+tUSzR19C$9WT^4*`4=19fl+7?$&DlHYjPEwZF5($37^g|~!_FOW;2cs z5+0aEuG-!&;U7}nZ_L-YurETyJ&-r>PF}I-Qvii{SO8G#0;OAUHk4IGi=h&Oj>)C; zrBuTX`0xYrMt>PNc>VBAY;oFpYRDa7BHR7z>u{u-x3c;5ICfy5=X4Cwhm^WcBNf*x z9#K8tY@XSsS4~cB;MM(Bs~|ifJ#l4+0X{c(`=l<~j!@nE;F|L(pQKfu&l1z=&)JPm zC@vZ)*8lv5$pY8{tk@qw-14&z?sjhR6x6$cL(btQ4I*CHbj-v==k-)D9mFtcX^}ds zlYZ~!Iftp~HnJFM5X0@W#auV(g*8aagd>b)gZ z{y`H&CsJ1owXmcq3NE~g_Allgl#r;PMehhL_vOPn3qs-r`4p-f3J%n|V@-fCPw+UV_Ys`3*=}W0{j7=5Fu#P zq&LJQ_izA*v^eBnz11l+>3m*okdidAW#$PN-Mh!o`rtutM)oMU*!78J{%u1gG0Wab z=xB+RU(eq(K$f*4w-3g`9|FGRJprn6!)x1GJ!!y%3?@fI^_y(Ic$DNKFsT&IJ@5vV z5W^(f%p1Z@D#7YI4QHaz6x%~I9-FH?NKd_s2=-gFcR9zss17+!`!L{O5|)SiF`tlf z2xew9(`ol+Wy9hkfuyty<_2_`ULb&;ZYMcgFPI@Ipn+04f8`hr0n{d8%~QLoat-~G znIr(ZkglI^Exgshf8yDT zRvNtv(r@v!xm7P*aT_ZUp`2w%JU%TNb6k1(kkjM*TGecf)mrspPT^Y3mx`6O%KoeL zzu+tbY%e(X_r49lnz~UhLcyDh`m&{ob8x3;I1$s>=bNe zsu)48^D!yDq^T!fSOTwiA15(Qc6jY}(6&*NVFR(8HECZ+;w9~I!R+OYHj*NP+-GQE zTa(Tr3qxDq#pbqPdg~!^hi?OIp>9W8qfzI3Rn80B+(RE2&2*$n+^?4MbwzbhNk}~0 zL+GvE1#GJ7n z8bnzs$Py}ZFLGZSptovu`HFe0a+tE?-R~q7%I2Q#u-5&+OL$$}_Qrnld}mFHxJRbW z>*b@@3kDOljT>B_gl|&;5(k($0_RB+%(&Sf4=yz(gV0}zs1P<~mH$JTx^MvFi;sJR z4ET|<-4W+BNl)O2Xvjs=iLuU4Q>{rtZlvW8){%9dnA@Xd*UQd+2vp5fOYt+N!h|0Ce1kYuBer9Y`wu-iI$M#=coMnzJyb;85{cUu^ z+TlSZ>$c3mpuoh@H(`&bUCV&3H`SxQ<18je&n^4L)7y-pbw5m;#%43a2GkXAeA zw@!@^eyGkl>12(YpB-FzOQX`j!kg}eyk>q{TCIG>i}}nbmXrf&B8mPGT24dtz2V2H zjE|BmIv^hCru)I{cpI#M1U_RH740T3^`rG0oU>HoXxpqv#*1e;Bf){K?9ppi2LA3F? zxoKxe2yZ6_5SHt!24vu$jV8Li%q7#j=}boVTx%5Zh&<3|ToIzB?Fk|xe%s1-f-(02 zT{7FnjR#qjb)TRw3bzOgr2J{M$8(5k&0AlTtp-krQ4r@@S}pb^^oECU6s$nsT-Q_k zm3066C<^{d6nz0z5#gRdX+T^W`LsAGIS`OYf^(Q?xW7CWA36UHcEg+0U8HZ}UHVOL zRiUceIel4bQCCA1%a8W-%_n|;I+fCiH zJ1>vH@thCqhY705*Sc^t2d38aF8QiWv(i!78qOkSk*Zdt$>Y#@i)H=FO*pdk;M<18 zywI(q`a?SWofwFW=6fduXr~T_=Mfkv1e2{xCxW`0SE;vCz|pW4DmdD1%$uGufDPO~ z9yvwjfrgObdDw?=z4#@3FK;ldd4%Ypx#hIe1+(&Kj;)F+b_uR zLw;{$0k3I;kJ(xEkQ%04OG%~j(S;W#^;#BLRU9TvMRh6rX2rw=zbUt<{bxj3?C*&3 z9h|)@^}9q7dv5%8O)kM?sCV<15u6VoYv7g)7GO$eOC&2MA6TWgO zSGe>R*!&hQTvK>#$|Ah%fIk&Ip<6hN0|>X~Z?{z6Xy}eRuZnTS*^_GvJGCXMNEu zt@K}!A6i-o4~I;;A0FMW8{cb`sDE*UigZug+Tww`7#_8pa+ugq!c@1_AEiosnsbVJ zx^kGF!U{)i`rpJ0yUq9j)2q!cxqcjx(m3?j*wvv2o<{n5?0DUJ{?Vj7o% zcvK7D3<#PMMA_!p3L+e{3x*Jn+O?4J*G>_JvN8e)5EVWu6~bK$^CB~)-haj_aC^U* zS;MTK;TaK2+K7N-5-gh$Oyhq)W{q(+Thh7~=9g%i=a(awx_@6j>QTjJw>eW!yqYg* zlUzb7X?l!FTC!1qVGKf-izl5W5XGCtoL|A4O~Iwh^nINsNjHW&FDoYKcB~biZGL{G zpyx7(;+>?0rKFh_3<#qVwCjFBHvX}-@by+fuP8-r!Fj8aXO?D#L<~jwHft)N~gg{{k+m+y-$`)si8LjjABEvjc@1Afvx6LVUo> zzebf|^bw^TA8K=b2APs{fnO(Nsk^(T%azUD@-`F!@PU))#9U$ zid99xsD%-bVpz#f@n#p>%7|R$szikqN2ol%v0>5YyNPxz`cyr6?Q3)q*v%2c!fn1h zprWXKNanU1mLu!mn1f0-{_=kOio8IP{*LAD+3_P9%E?+w0FFQUS|7r3=q*7#s}H9l zu;@eJKQ1%y;u5%IT-sSlrQZ!QwZWwXGW2n z&7UJn)(^~h@n)%;g|pIeygs5186M%_6MezF&@XW9$|u6vNX zEW&)FK%I5>uC2_rCpTEyytr1T4Qo;Zw~!)>A0+teH}ou1A?ij9uFYt^L{|^i@!7RyW^=H}OMTH@@1ZvzM~q0`pd5G)<@u}FI~mlw^wNr zPO-YaS$5l!l^++Yq;~$`x;C=+0n$&ht?T_k{kxz@t9#D)ravoyD$)1o#}#_=MM9+i z{G5CQh$`T1qd+8xdWMRqjhNMkq)CK2^xP!}GD&&@xig{Si+;x8Z``*tStZbYvk)=T zRvF2wQfGT1;^>HMzWzHtv(WyLt(G_yxe+xJ9m{=hJ_45qIg<5gqbez-3)VAJQ97IlAZ#Z!gefs1SB$a z&Pi4Zs(%4dMCcZnuh;5cpnK0`VjbScX6y4}Ks>ltjDo=yQKS6SYE3{%NbocrUDdao zSb7?is;6~=6?F-Q@PeRD6XnB=CgK*8qDQ~?5#9golC!vOsO|>!KUVP93~K@iT>rF> zCb9TS9QqPJ{41CZ-WqCpMHsb}J?dzy0}G&Iu6Moodb&;e@SFpAUa`qrA6EMIhvk>1 zyON6V$gVOC-v+#-zukz_-O+>tU#ja}U3NdZ5gC2{r!oF5E9unwfubF^kNLu_D%8SG zt26?WNB18i1H3s-)O-6o7kPSu30N+4)~8INz$hvqHAA?YH<+rnqfr2Tw;v)kJ6CC_ zamm*q4Z1eh(aI+>*FPyw1f-?P(7v2SLbtj0#zmQ?d3yOc)XOx~i#xCCTLJfSG(*^W zh^M^?0=J`N1?%6AD#B57M1b@(!!UHuxk}9w*?BWUfb`mfSy>5Z6V96e`n1|}T}#Fc zTjgndjpyP3tb?ioJ>?%Gk`vHNW!rpr{yDMxLJO7C!~^zZH_8*aNrRh#YAJT4v&m@$ zhmkpa5@U8P%s2$(+rFgb2?DT$P96)ke^jrB978?|bca*zWH0JB+G86d4Q^7?0Z%(j z;x5+PNfnwkb|ua||L3LeQ?8okLC7B==X=-T zFCphnZ1LBSOWc}ldPN|yHNf<+c$m{ur$p%KP6pKN^qI?M%i2>|Kun;OqP$bm@Ghz7 zJ|Sz(@?!TXCFYYXI_268iTX}46jyekpUeK?<;X>ItS8glmjY3??O%~^*ip4V^q~K9 zHT=k*jcxY9`rMf?%$I9ot!|ab$*(;Lr}yVi#0kNAVV&-K_ygtuKqv@}FeWrz7&D`1 z+-O7qXfJlD{zLP7?nHNqlYIfl#YmS)mzz`8Gc>J6E)%}apP>wRA`^kE9((fcg3DTG zd~^4=a3xYh-=FUlaidSdYu=1NS!{=lRPHI8ea4NE(ue}#+2YKNH~eERpVGbyCIW1s zaa@N}EmY>CG$yvn(X+9zN;G?J3=NZfmS>`MUPQbFrtyqjMW}+AJhmqK#|V#YcFS9J zoytoC#-4G?Jxy`xps=YjYG>BOIFiwMvGoFxG)B(eWCBTPmYD_CyOeQWmyAv?NoQeD z9|ZDJB9_To++7=KjKdN>lI7LIzS!;w!5Tr`m(*9}s_S9RWFMAyyJ;yFJ6iDv%x3-0 z$9r^JJ~&BnZI*AnmFN>+Fs6CUZpQM2z;mLHR_k@B`NH+P!ZE*kKAK>8JNKVahZCUx z%cctlA`tJQ?%2(Zp@wia9?RLco0*;AoSH#59y~_Aji$aeA$qMBq8!9#kLh+P&3GSv zS>nKLq%?yDkQX@h(wPe3Nkj+;q6QrD@5ITR)vD`ya=JSipmen6?axy2G2Hxlml$1^spD5Wn8%-jZXw(2>hje07S zEb$Ytx&aBBT6QGaBQ*lVV+u7VfCiI)a*&yaoR>|qj}g`ik`wG@mu~@~yl_B=&=p)K z&32neQ2map@Zq%F)6LLfb)C$ zLFDx$vW}2Y#tHGz zFLJ)wLx7Vw_31XNk)at-YYX!dVA!TaP~r1_c4xP<==DU$tEEf+aOOa!>)TXKOS!#5 z5exsR&+BUrK9<=3@4Hs&jd;J4nE- zhxCHZ1K1r9HQVY}q~)OcA*h&0)M|cIX`Qb>lzDk&;=eLv6I)+hL$<^o&GE6BNH#8Z z^F;GS+;H0=esu8~%*O+~wLvyv1w&$i06E&r&;zg2snL^YOg>HW+i7QjuJVFTgeUR2 zDt4E3AtXDP#%%ku9h4P71OQKY(gQdhd8dLdn3>!U7Hnteldat!>hmY#=F+^`OW|y+ zdI{5-ar6ARcc$~{{OBzC<-^;Ns4-S9 zg#8SYbq0s-NDBAu;J81aKsTgp;MB*|%kgF0fW!%1i{z?&lUXsPkFB32#L?6bvg)pp zn9KXIC9qk=7$ij6L@^b7&PyVFpc(Hnx`jopwNuT^HhFTx!!l6a+nQ$of(B+`^Z9A_ z@R03fZ?VuI&B^m_RJaKi6qJ|=>P-lAY%NQX4RS0I7S&ett(F^8;=eTf6;WCMQpa8j z|CHvhIa<9=U6yMVDNd6QWoLcN!dSxD0_WAt5bE4Y>x$(w#wcF1H+>hwyWUjWxpkc8 z?`KJ#&HRz(^y;CiR5-sX_KsZkLM;MYs^j*~QF{&-S~uR1KIPeH1eMv&kk)eHLd&$1 zBP6J@s5SQNq3$ZNPSIq7z&_LMkU?XkCVoSV`a_2F(Kd`T-tJC^YKK2A7pE(l=HejL zVeVKU2&yEC>vdP$?P%@Iluz_~xRJa`YjjN!M%9Gm~;cS*F3A{cTnqGX-X-<^^)AUu^6*D$ z>pkI{Y(SU~o)i(W9V~xB$DCpr$zI$ML2r*UH`eM3=hHk4bSXA`FGf%3a-FH{jvJts zdG&qVlR-D|A;mokpu#FK_0ciBdgNaX|1Z22-G-49^j=3)>wP@?KH~N%^r)}e2kk48 zIpKyAHx50QI;{NR4Gl+;5+l0^(9ucfNNd6$7dpKlBA^lBO*tWzuDV;j+XJP+vkms( zGR@QTzfmU~Y08PmHEm6k6OgF57Vnn^rIr&p9UwU60||2;Zm@hgx{R?hst{m z0ujnO#2P9W??!W4h~tekhE&}#8NIcB;jVF%;vs;Dt5DlF4%O!0@rO#FU(Y9OY&s9( zOdYslZUk26uu2SX`9t8y_*fYqyhcL}3C@H(D*rC7yfo;@Za7tth9#0Jnt#;7sklk* zL5_rjuHIAH!>QaqL4SMBXLXU;d5Z5N&&HZbD7m#@Qf4t?j`sDZ)Gz5Xj|8savM)+fRtz1nThsEkf zUD1qXd3|AGci8m1`lr;gzi)XhOQrs8Fe&~KOjk%OehH>0--F2w-y5X%@Kf>aVEiM6 zlrM3GCh(&#lCSRF*%IhV%H?<^a<`T zeS&3refr?QTk%4+ggAg~RX3!mX(8 zoE1q+ETJ4jw&y((Wzs; zxOj%yA>$p(cZrlQ%~mZ>0y=&dkA^kA&Gdu{^G{S&&ryF#l@M006`aZl(JaG1s&bJ; zwI!BPyu`TLXB0EGBH$E*D;Wf{e$RH6t?jGVZh4!M{JI{``wu#c-$}#2w26OGP2iM< zgVjlE&9UJhgC(!Wo68Xh+By8_*_;ol44xn`=ZUyRLuk}FSskIc-fXo@(hJx7^ZE;+ z`QmL|137m9!#6?hcO72beP;9lH>I;4#lfT}ckTXwN!sVq>c9IeY#3zu;hZC#&vdt2 zu5hS|FkTvhfx&mLCUUCLroe>_-N_pb-hIk|4|3`3rLGL&Y3CX&k3znT=on76^WCXj zd0uO+#U^sY*#;;H{iIcY&*<^|e;NM&ykC3s18DK{qXSkQc6U;KUPEun=Dlt!I*vF) zZ}6JD?(bqu!#X~DDx7+Df#W5UxJ%<15QT$pI_yc^r^Zks!PCYzr0qI0p%f1p?)cVo zIerKiChZ&u^X(%xd8T==OBCW7xjT#VJe-ZtW?k%t1fOo>y3LQ*!Jwq5n$O`4pkYXL zU<`!7P52gKb~|_dsU)Z^I}+AjI(Acq?2-T(fuuqf=@P?{X;9o;FBQvSgA-6Hii@nm zRa1V9kH&-1RgMil9+el+cPsYpG9;`7f5W;gM=3XbQ(ajJ(;|lbXb*=Yq4JOMM zT*67c36+^H>{pfkX+FIK1_NJ;3yf}tuZ)=t&bfMyFOAo~G zL%I^WDftc6DmD1%?$^HL8>m2uw{A!7Dd`8>5_qML;2Yt5-z3(?oM)y(6Zf}Qxz+52 z<2JlV_@hDW?iIL^6g3uiDu6C+cQfYp zHTmAO8)>5zzxPgcth-`eE1XXZL!-Ajfh<=>vt50#5-$5X4W!mi2F;lEPav^WJwUza zg1i!L4y(IZhNYUeZaFHO;o*=5>rI=y;lc2kDp@M{A~T8#3;qudrik`JI~z*mwTH*r z!$pP10)*hDtVR}|Q-(=uhAC45^vi~CAAQ&$c)BmJDB#mDA>(mKQau$!HeHPelN#X? z8TBN_qo)3AxqJ4H<&NoZ%Uuhu+}(bFEqBivbV_jdWgT!4`%x?_G*(Rp%ywB6qmj3g zWU*oE;*kF-eHSCU!?T?~5Q6|r-n*8sbVCueAkx|L3kt6Qgt}0-mdS6d@R3Hj#}8YX z>N`|8yC1uj*05A9a#w1>xwqPUdhHKS#w~nEN-jZ@UoZREnFRu+Jgnh;0DGh-O0j*n z{9$gfQMkTKqS_#~BTp;=h+W^#6^B;IGb5>YRlgaeF)2m7nkW|vp+)F<|Ja01LtKrM;3CD-VG>B zLMk@)|DbncC6f^N7z|4ps?_9Sc3;o8lSS8W_@#U$*u&=5ZA-$O?ZZn;8T@6y^3x3d z8Vly!`V+rANa6w_?*H5V*H@(9GT48v;(|gU0N?_GbM)T@#LeS?cr6i@bwoc1c(tGk!nsZEf6stHybr@C7nHJ?-=F5 z6g;s&-o86_4=plj)qvJCi$|}X;st>xD4ft<(fM0V=azr;NMz-Cv!;66=90t1X|IyrgT3Zbf!iySr98FR zsKa2ZP4UEmQ}C;X5VG69)@%e|3&{N0A(fC~01?qI9TM3T+(4^!99dzNaUz13o_V(8 z+(tt3HrWWkl%ygs4cqMZi6S7M8;xI~-ci@QNt2z6BtL+a5tRB}4#=JqiUh^vgC2`p z+?7AA;*EFNuHyhxA_Tu$>;af9rMIcs#A^Y?5ACk3g z9LcYhb$y6c`Rwyfx4h#cnbDgRl{bH$kLag%JThEj^>eVqp)(Y_EhkS{v!Qc9NLgrROY4+LPoW?--;)4fRK&-yGIc*%<$x=zjjSpgn(PfihGx zGcO@7(W{21T+i1dy~w!BXfhu)R9&175dRtH8UV!rBfxGXk_Y14Sh-jB?Jl{9_oisLfcMi!l|T5)Qqh z%gMXp8%)phjBOUB4*NiVGkroQV2e{85SW1ax(=kZyB)c@-q>uG8KJ*u%%8ou)d8s# ze%z%`?0CN$BJs59(0(ObHs@)Bs2BYlHtWz1198umIdxren?6-8PY^UQx z>Idf?+B+f%f)k>|@;ovR4_;13_z1+Fu+S}nb`t;UZRrNM11f)BR{UDdPENeJvN<8( z`NOWf$?f|qOkhV8uVzMV+bp6Ro4^G)|D+}-3IjISc@8I#&v#^)vk4JZF-FPr#)gM8JCs_C|+q8{aReq5e-H&Ms zGbDkLUA24bgD;&<9LYK!tfVgVqzDV|D391%^vSCf4u*_!TLFbCm2w69J74;XAN|{# zCGl_cz{+sGRd~So0GY7BL{<8}rQU&nFt7HuRvZKm`gE-!JdlF%<-w0D@)3 zm{~2LkvB&F{MPy3N`=Mf4-!^*mk)B@-y}uR4|K4Jv_M2(uU~CSi@XZ`>cc%^If{y3y?rA*wdAo5DZW3imqO&8bF7 z|Bspwtt0LRcnSd`7p$BVJ8=Su@K)XyMuJYT^&-3uunotTNYIuHPNy}DfHP~oZS@EJ zo%I=8%#0f$wU3d3+*Rc~IYk4X>=-^z9A- z2@`CCv9Gyg?s43N;Ap$xM5lV55lDqe!IBrAweIz0s){V3)NK4iEE1B4PDmAafw$pb zEP>i=kcZf%;q>d@taf{xY>12UH>H zQnKPnge;V<&{Ci2AuR3X?@L&>z5iUos=>cXT{!W*NT9N1{X8U3nRPo|9Bl{KMa8=- zw@kVkFPUSUQCnK?6U1T(lJ5!)CpUU!t6C@I-Sd`A!V4PLeIW)^J{p=Bt)5=XS0FfG z7P$x8n7mkkBQ-$d~tTJui7r_yB^S8384sUZ2Pl ztRTJNi(K>gEHmK|D!pz@$E2#2B%GQ;mx>FP@wwjsY zHJ;OU4OK?%WM)|Wo3gTM(g(4rWnW4SX^y>}&Qlwa&;0z=9xo74gY+XaD?5!;twc|E zL!8i4co_n=PnRcRMGs+xFX0Xc1{xLEYFEVEtiAG50^J0&-2+8L7rnzkIb_@M)E>U7 zX5!>_ZD6o^!Ky{!z~?IYq~N@p_)RhdYhTR}V|0?|IqFJI0}^O8-UwcfqI&xZ-_4tM z0yAWwPx7Op1qoRu%k?xw)}UlVOB*d>?n=KLc|XJk88|()l7B!3sLf@W{=5V|+tP3B$6cOS}Vn*+CezfY8B)6+cTp0P% z^GukN0)a`mD631N}U0b6A-(Rg?-j;`RL^$xTI%q=5`sSez#aZ$9SI(7Sjse}d4vLPN<}tchzQ z)iCN@ghA1+B^7jawDkO~cVbuap>VJ#+4lQH^OeE2|9C=PfA}CmWZaE~_(b(npkaAZ zLT9r68|E8he<~dGPn_A1)fNEAd7a7Vm@J1Iy(oNe@2s4t^gs$~bEGJjUqD6>yWjAA zFWeP;2>&VGzHk)d@Qf|PF-0PjgFREeN;3XT)lXO_dggGki_8XheSek%70k?COKry3 zKx~c_(5_F*Vb<)@{v6Ab7igL|zbZ5g!6BU#>)z@VzJ}qg4p!%&eW_Hl9N}TAkgJLj zsY_gKc~lZ=3H=fCcTZ(OLoXe4cz!=6%ERFg`-6TBUlsh5KAtS z`y6SWHl2!Ao)-s55s*C^`Y74@tpW`XGQ5F@Pxn?i`N~Mx$hh2kG^!`cNq>bpLjUL; z^Nl;wR&5d;CxbOo0`72W`Y zhzW0=AN+TxC1Bs_AHno!nc~>jH(*YemEmOEOa~r45B~wfmdOi_^P;ED?w24v!@aMC zExB0ShBnG%@`s+(De2EoEV=4l{Pjyr$tq~0@|$CKQ`SkmJ@+W#&6#XdI-#)twD&Y# zda(A?P;l|-fy?R)y(PTro8jIVEmEwaPX32PDrRco9|kZL4uE8(jPnl{hps87TR0a7 zXdrb8!{9^X(@zgX7R;d@qAXV9NP**)QRb&Ip5;AZ4xkXcz^CWUBy&; zj9KmGgslK>Aj}g-oo;hKWuuu)SmatS*`pWL^WHTiSve#E$bC~2+`O4Wucjp1y8(-1 z(y=&(EN1)cK})8))NnsUnB~Bm?~5!!QmV)HzQ1=p`F4J~A6~YBcO(e5>@5NG7TnBx zc8+M}CPDdfn)1te*Fsge&vlxdFPpg^#V#xGrm7ORF&V&VE-`=FNUT%!!c{PR`^ zUeMo(JpZKJ8vETA8mGx%zdY7idOQdvWyK9b|L7EdHwgW=mq`ccj>>h3I)=kt0u~)X z!{$>IRB|kv+J+tejPke9+JEg7$wQFiPEqgt2xc|VK-g~JOt!77j}yi9PoP)A0_i`HOvp_jcjmPlZzZuiuUFa3CFlvif_qH+I*ip6>ZMR1Z+#Qx ze+R9cIBW!iJ7e|gce;mCMd4MTWEXz|3s^@ie`34!NF`_kRk0_WK+25ERA0v-E_nGD085F-KslwEIV z$o|R1z#9js_-EHS;ztbqI?c`icXDF5EcL?VOk)daI8AYyVvKIr6L8RlltlgKxriEY z9gzF|!IcH|T_=_Fp-5!9Y#_tsQ(45wp@wkCc}o@rTF#r#cokf?87^fj1#;`vCsl8x zibd?=V=_8FvE$Ug4!d9~PGVMv) z)Gnk?yHyo`quM+J3iZ7H;)37;j$*c6)vQ4xM ze!Dc#_2Oc^!CG@$kwim@=)C)m<744QDiT_8kaHe-ySUa@h480iX!Y>9^ zpSsC5KCG7DR2M(5K-8nKpnoCWAtTAWpmQMcLDShN9g{7E+|;}7(YXrNU8cDyVLhEg za?jQnH8br)?=JZv@R&F=O#RMTs$XV0@e``h0n`A-V+$b3_w5Y=A z)&TtOs*5vnXWBjK^sAgDKeowxQp+bezuuZ{q(5`3))|aUk@OO#!gy6_%qs@)f@be6 zwm)aqXndS=BxJX`7H z#5wM>Q>*BJZ_K zlXK(h)TzVtA9d=C7lEKKu1@i#pa*aQ<2+;iFBA7xadFFeRl2HpyafRwb9|`%vayAS z<8p=Y2HRYfjX{$7m+7I@ID2Mk*2SI6((i%{$M!FOjcfpr zY`nbSJ6l}r=E2|k5WTuRtc~Sgamk{wfD4}4m z1bmt~zck_!8DFkF(+XoL{1VeI;2Vk!B#ZyvXy~}%8NA5D9dsW7ts(m*Iqw6G9xvBE z_yoD-d)*0Bs}RN0ZFoJB^uqP%tPgG~RD|yt2WE>3wF^%juR*=iRfw_s)c$_BK#c%* zn%Oy8e8w=zS#ro%Duy4#+Dhh>8AuO5+9jY45kx?}CA;(=3csF-uwTHae5=}$E43M` zs`@(fA-o6Bxm2#~DljMk7lBvYsd>^WQ7I0`%+xqabpbcJE5L{?i_cj+s%YcFQoB(w z9PD)1@qr2=@daH~UVB~ETbz)ik`jHc{|mA@_=vtrtoq&NtIbiJAdhQShK^5F?o5c= z$*yazSj}~(AfIO6=*ykd>U__GM5t_f&?lrExflLBUj|hjCO`U9Iq9Y> zcpOvy&+Ujq{5h`^%IG2R)3t-bM!<-WOU+oWerC$=zb~R`CPfxjP3>cb_dHtdfugTj zBH`S2x0pR1v3ZjI+}-r}?*pr`KxzU45w%5HH`0Fyq`-^E-od~Fa2i`9{*%y&>9cth zi@b}XSzhd}mKlZxqq!hNjq8G#4`!*6Hk?$RCpmaoIs^?>;ue-_DD{QA)HOb^WSX>?tmL*EWjz2{?EPgEdK&Vz% z;rMF8&SUy2>7u9|Y@XhAAxK=Tb!Y8=!xC>WHaSqQj}Rff5t^s>3+|4fPI zO$*i_oc?9R-lE?b;D=i(%}nG^L37Tb5p>rvkb};ko?$-DL~D2$N-k497efIZX-vSq zl!a9ag3Q6@&#DGQAH6!ulkE5^kSA5f9_=b+*1E)Yrxyb{-*6i~a0kcoEl%LzpkyJUrz?5|vFxFW^;! zRM=(%(y-0qiA8jtYz#imBWJpe8Bj^q&C2h!_*kG&YBs<`!`1bgc)aSdaaUV11dhaj zfahPNp^Y2UDi-~7FAd?-7>0-4uVi4J2m0$*|4*ty7Ll6rLd z2r~mZ{_G4N;otWnbsY`8$;60_q-a%%$J|bF2tm9^*Mv3w*d4Rfetp#w2J`^O-%^YB zx74cI7=nhgeE*L-v;-z1ATT$es2fH?j`Dwme6qcG0b)2Lm;ZGdd87 z4cF)Ai>>veu=(?PrzK3X0e*$Qb>R@|%NM*0HVnHAOZRV3#BAv&?%NpWDM1d|slT`~!b2`cL1B z<0b_5*l%h(S&;Y)dILf_6MV`dFds0XAp4Z_o|+m04B@w9^CgPLQdj1nJykF_PUU|- z(t;ys;kCtdFyUWBjoJ28C6%+x(?N_G3#a`kFEougmYj`4Mh_}De~=)tsm^wJanj$^ zv35t!Al_JGXi%p!J2vOZU+3Q%fcw93YWT^?s>u+{?8;Kz2kDGD0mq#N%r#%Kc~|9m zRxPVLksXUlH>DQdrxG|NIA1)Grw)i2CLu|X6MZqN{`{=V{Y#qR>;ydyGsjuch^b!g z%kR=_!1x?C4W1f_fqT3*72UO#FT@+FD0kZm=PS0~-`&~tE@lOu5SX2OSG6fwOXx0w z8R?xj3At)%u(pT;qp3%CCz7^?#QO3S3}r$ON-cw5+--QpAiDDlCk|}Uorrm#+|JAgrg@`U=3Zp)1lR00Bhku6CwRa zN(tSA^xgWugqH3NcTuOC^SJIh-0Uosn2S(<|@a710O3c>{ ze=3}_Ys!*Usd-}TD*76_+Ch$49e${%GAbnP|B7go!9wYZ74Xt{h5u&E&gyt0-2sNR zR>WqkCGh}a}oVI+% z!9cW+c$y5}9KAn;vA3yq$>kI=&_vCU=0E{*#?~{fHpBm-jV*yPIRp{F8MdISFRM60 zBs=psLZKsz9AQ^MMycU9Mop*yX-dY~$gU;-|GpUXLXtxm@wm6XnSV}-xtYKDbN_wE z2}|<=GhSDtgUkAlEtp{VfSe%0>>l9_7y{bBdw1clHZ8Z#1uYkx#+fw~v+?yp6DV3; z0dQr~Qo&0K3#NIo^7by7S7vKWRz>FZbQfNQT%;7yomn3j(}=DHcKFBIxM1 z=kR|;5ZiVx76=735Mr1Zw=z4Uh>mF|S63{LbdBw1Mh;n=i0A|uinAKM=56*=@@s3b zFyKh4PimKYD^vVAx2fXubY^WEma zO|L)$ly%2y;gw-sy$5P>^bRND89W}tc%w>SxFzt6QrH*c$I}on!D#uremS$A`>VqT zJK@_6@6+tVZoFrDVj1^R<&8&Z$rG#50_)WFRoky$d~2M(w(q;4I;Y$1I8=LXT#yfQ zNp!=<54rv%Eh zRSEw2hu?3)X6mu^vUD&z-?22H!exkbL5Rq+GxFi*jip#znz*HU*T&7K@p@f<|Oz}_q1 zw0i2N0>}%n1J*w*7k>+1!oweHpMb}Mu>c&7r|hnxIZKNrf9zPU*7T782;?>HaoL;$;@_aTar0!8qgXjVmLxicRe{b zTuQNia%}YEOFe)1LrE^BjiJwPh;RmWLzF-%cy*dpxk^w5;zxJp1h3krJ~NnEw0)&x zu%SDSNAo!gYDnx|?3GLc_kDWvzC;Coxx+g>-|2OyEGeB48 zPMAtJ@VdLo_U4??OLZPGoL}t|fE#F$iStQ96?(ynLJ?10{1DW%@CHF@1Z;1~uuRmj z0Uq7hzccxFRy1mjw~5KwLDzHBd{13{Yvjx3G0aZA966jO2BZZ{`%u)F{{`Q%V0%vd zuK`T*v!%Dg~p-Fb-Pn>p)65a;_bM^>!*%|1I3 z`1ZfsbAGc1`{Dk>f9Z5~Q~;K5M}>%=%TJA*VECnG(F2h{$}gWBL-|vac)JnXXvlE( zm@Uf(nvVI9$8Q{)7_&XaGGQfpKoJKs=38R84Q8Vsl;++_6zC~_x;c+!%V+hbpEl9u zf6ApEFZ;obiZPyDX+>?p&H+_ps-gM+VeKmXqFlS}I}DA4fTW~= zga`rx1~N#85&{C!2olmr!wlU5f^>tV3W`Wb3|&fyQW8U#N~gojok7nz);Zt#?!EuP z`+J`qYwx}G+7)DXU%fo5QGVys@VM8|5IKv^Gs9!um20mhVW5?9rQx!fP#PBHvmec# z&m~KaTGzj>X444}#eMrpMK6ct78-S^BvJEXC4f%1-1pjtth*x>&KNkT=l1!1t z_^mw$;7Qc0>rekSX3c+#Sx`BCC1we zzJGfE27Io|zj}Me`c&!HBF@4Di!h5h_PV3OXa(LR>0_6KL;2();2h?mWb|M^>k zJg`Ei&55n`0Q(WB`!=!Re)|y1Ey<6^4D|;x;gVdoz*=L~^Inz&;ex!&YpQy0vd&%H z2JL5B4P@VtkB>}VS?Vn|^SV5)_0l$^RCRK~Kv3-N#48^#%f$A!1EMLK&xkqgUo&B_ z=_=FwC|xzII6~VdQC71jV4!JHH78seofEcJoP`LfP6_c?d$1B+zeItF<=>nl$3A^A zo*V@${GEbX`1yIlWA`t94mFLwl2JGlWtmLK5y{=e8Dqwu!g1=szw-9#4{>&F7t#XQ zv=s_M+!7Q_GERQBnzm_Ox601sLOJ#FhfIJMLNo={v0Ym>?htvv7#^NzNDO=6^c})H z;3-H;DK54k&DtXa2wJZ2_VLGkzVnt}rG>!;Ty)=E6>< z)0O~%>$Ss+IIyM&@U>bBo{duq0R0Wu%NNvg{iJC<^_fI`Bg1VsS ztoi0Mjif_1BHc@~)N%}~GU%<9CO657?ydS59V3%&%0k>*fRf*(fm1Q;i%`0j8n00pno|FF*7Yy=4E zeeD)9owGg~$;XjHE1af+%^9(3C9@cBNTwC*j!eVORE?e*rn#@@1hu7KiQ_$V?|!33 zB{0frpj5#tn$$XvHMKkH}}7LXSUG;-JlO*K(g408UfOPm0w{ z3|J%qzA@$D%CWN%0Fm~Tf zl_o3i%fv?4xj73n#V#zE)FJq5_Ro9<>C{*8c$&3Pqd83po6N?tH zs(Sw-{&lgL-Irs~OM9SAOzReZWq8ycMK{FmmRYlyZQC%W%^UWb-!1hKQnzinq^wPv zcjb?pJH;i|1Si*K_*J$>ZNOJVSwDhQ@r8A**zJcb^l6b=)n*7mu*U!#`;Z z2X}tfW4_&M_r2B|Api!VfV4jd@Q9yR#4hhCBT6Hvj>>nlHX%ER|Cz-jD0QM6=jdddduy@8|N@k zY8TM2-Cl_vWU;jMG3a(Zee=aFf{W(Lm6p0yk{{oVf7yyIWx4X)VorH@ArtPmjq2B} zTJH>ZW>S_LaY00NCfz_8>t86p6P9Xw%p6m`OB?y#J^Zq3bMN**8Qxi%52M@5!&S#e zU#2s<>GVEo><;CiL{;*nkT{y*&61aUNqWZ!@X zVv}a1Xm%(YzB5}m+tZF|{|QbR(@>6@(CJ`tJvy@=M;GJwNkFDw%XgP11G+J56@Kn? zskRD><3Z|r!{$K`4fbdMtGs&ERPM)Bd;qu)+}PoR-|E4^2- zgPoBhl?`q&)bs6Tv+fY=8(Y0KYViUbrjgquvsZ5kS`jlTy_=Q@ImL)t8+fbQm!pQsmRrfkE`IKGI4qP>;~Dm!NX`zo$4 zR^lorl0*&W6MNo2Fj^UO7hjJMF2b8;u76b}^E$-{x3~IW!g{7p)!&Y(u10Qpu>5Ar zlRB@)x~5F8P8L-PC%cWs>RhAV3<-B&^@Du0z)x?ikgSXIUric*$6Wri9V&t<~muIWH9IoHRYt=T zKIdNAkJkn$zb3uKYB<$_S7SMoUtN;bnY+-szd~cX)L%*KE(HmH1qkj$h4BSiq&Ja>I}6QwC)ZRYrmm+KPsF*T52_U zJ&KREL@VbWv`dpmL_j~yx1p_tQT+69lYKQicH?cPU6x2#^Vt86Z2O;EG{3{Vt7Ase zUrIRLK1EGdG@>y-$=GUFQ`6WQy0_8n4@DxH$w^V`{v-lP2z=`XE1CIcZCmTkU=@E7*nYbFvvSmH-!9j=pvlM3xMGf~aEPg&6yTh4?~ zFL%t`!NKRAiz{cUig}p5R4{pW=bj?V4ISSlgCv|-p8Nz;{GE3SPlNDMQ%f6*7_|7t z2L^4vp4d}MX86{GzneUNlAu6+boj0(%HK;F1;UWQuc6cNSk5??c4_mflBLuHEuEy7wjQ ze!mg$yhie~iXG>^Q|V*_w;vJnG500orl2rp{Jay8$rdqklHkjlm=arp-)v{!{T~cv zr>W;93xjRwrQPOs6o9*H+c&E`M&&0B&{`f)l@uZ^Q69p>GcXnBV7d2e9V{kA>j? zd`<$GQ_rDjpTr4Dne3~t%@txC!FXb6Az!_}#-%Tmr@q4ED$`4DGd((D&0Ggz90w9A zf-QFlSgZ?EfLy^AAQzT9a-$(H-DhHCgkOC07%YM3x@WhR{Oz$o_CNJU7VHUx`TvjK2_`i?xYJesn3%LCJ&S&<1 z8HXjU(Y8mEGy0zlTg&W>x zK>qJJm8U=;M!p@4L#ou}Fxj|%y2%QA0QkRpP$B`it@?kXUzSSaQc3)2J~g#Qh&82Cp6w|i2Dh6% zdR#6rS2V|{lz`Ma)I7jjH=F2ffXZ$!*Dd_6%lJ!_CwsZ@$1>YTkGc74tYVXoI3y^s^zzHe;qOdQQ9(sZ^{DoEPBb9WedxR$z z-<9C4lD@bZ&o@{KwkN~Ig`F8L$Hj8RxsQ)x!wsIzy+4-VDCz!KUBoiOsrbI;r5GL@ z961lC+4y+jFLmA;!sb2cJg2@KyO=lJ`A{s8ndl6K;7hxk-N82YD<7lpHBbDJmone* zduQwTw?rhWJVj<{W%iOV6J(@$Y=-usc@AetV9|b(zDV@G<3dmbE!P;b7UjHDDv8tZ zW$zrijht%HC_ee@|0DMg`-_omVpe05^9Pyj_I+4#G`F2j^)%n#w&57wf$>is91VQ|Dk{2z=y>%V_uE?PM4qbZ1q__r zMLCq6#d&UWbOe8up=u>HJIgkPo#{LpKep*L|KY0ZFBdBQld<$O!@MpBxZ=i#a;s-6 z2;U~EgFSMZ9%gBw)h2At(Uwo#Ew}*^!I}=3HJu6l!4YV5m1Tp!a zR*hH32^JWr$-K>1CtHtqF^gAT0H!CwJv&UIw*|xPGR)0twbLI@KMrt zY`w?uGO&U_(7ozA-Ac5()(uvOJ`6rYrytJITW7Qu7%0n4(;xGB}Wd-{slXAUP25i~VEnZI@Y>l~k9r$YZePTih*`1zW} z4C&pzj_Iht)=bYqJQw_2W_K5Wo|V`psjTG(pL}CMeku3W1|W zf>bCyXm3;DIfr}k&Ask4&L_o~uppipZyf$>&$m;Vb};ZBL#YMa{@fOk}5h2yXi?&M%D=cC z12p@`6^w8y(s|sn&3x4d7#K!+P+q7Afyc0)@6zNbD`2u8~$9DA3`a$x}pA!D<{D%)bOr}WE`J$`h7PkW4 zk`xEluTMZ$Ys>Gv!>-KlgGu41?9x56$o(Xz(fd==t;w`p7bk2SP(a+R@YaL@FWk1> zX3{;*;o}~Mk7$kW;?vV*lAKOkYa(K*Yo7E#OAMz#R{bnl{b^Fx{N?A;hv23~J~}ae zoA2GG=$UOT4MW0+6#B>k1@{0_E&o%uBTM;0meq22#0{NMi#|6`01i~sF#<8js4Q-qwCBMBcD^rk}s*thm? z6sl*-fIyeJTu%3AT+40KSJEneUj!r=e#8Ij+NA${1*j*C47AUrB5p!h4?NVK5 zdys54KxZnWO8O|GWd%r(_eSY&+k{N=n4&W-*KEy8))OD>t%n?v&A-EUzz_7L0DpQB|oGy7TM2=!= z?kZh8wnYFlZStq& zWYDjQVQmRuIF9@~~wZ~5pvBKmr7H;U6hl&3JNwTmsPDdD~@TY>@ z89aFaULZ|Wi+d{)s$s64c_;6FUKjNoDWil@*A_Q5AYaZRsZ=W#^9g?6N}?@_yE13I z+H#%!MPX&BVZASyWFqvY*$+U|0l5FCFv~BBk5a#|qkq8;*OVgv;pb6Jj)-yjqp?68QK#1PxS5W;VcgHqTy_am0s4v^{tOhJ@m=flh zn|*QvOp+J{riV~?pGsua0d$7sEVz>d8()Rm3l1WupWSJ+m%oAB7PXUl<&LEz z=5IoV?aXhSF>ptizkj7#D+s1{cbL+bQWkS}MNIbDAGQYfsBhKzzi76~4cgmC-nEV(aa@z;hx`o2y*77;(SyE|nQTP#xil2ayeOK=c9}i@3 zEqn`NHgv&Mu2>2u+tsZj3MZVO=c0&d1nCy-%+=j8OLi*3iK@OxE+OOGXu`SBOb~(K zI!TC(kiu(3M$vX4?W&T%RyKfll`wxWZVeF|Mo<|ge+WbLKeUi_t%BN& z+49 zzB2DwJ!gbKEIw2Lwb_iRG6htU0*}Gv7kVB*$2wFz%iBA$`HbrC9^}GVvN)sLBCEH} z-br%b@@>92yX04PW3tS|bi1)REJ7o1SjX7l6*ZVv?tuC+Vnr_V%IP1E=7azV0RMji zyZoqP@pDH8EBBEbYp|yXu5&0Jont#?hwC7KF<2V(;g{{zXaV2~J_-}c79(=+IPe-2 zs~zDV{O$FCWi?uX>$S#R6?{N}1eEfm(#~R7p5ed*F3H2J_s?}J$*`Jz<6bWBm){WX z^r~dbK*y^6-sAnDB-1(3ej)uOp7)IW9An@dF*ZLgfJv>dGJkTIYm6!cUIK%k-S zB6YBVl{;IAz6Up3D0{m6wArbQ!W@tL)iNmHtdU{XRxeGZK4; z{^ezW8V%`s8_<1@fUU^WQSAIpQCHsR_A?4vIIuz_2Tf!RASauGss)J z4yE^d`_c@CM&CbnEjBJQBS)8QwhxvLpqK?-e%KS%DuCMUS0MTB^`??KD~~_f5OB)H5yq zRuWy}kR!NEG>Dbqm;;e2B)_{vs_y}SD=qh_9vX@YW+jP?56nGO2@K=9e%r7Uo^0Rs z4i3dX374_G*Oh0Osn}1nm=z$TXsPghcN2mMF!xjC zOwt+pjs@mjgkZBj+u&R)FV`0b3dIr!(`qVg+Ddo`EF_Y*%|~OBDVal49jDqWJT85H zGq~f>lMpR^mcQ?r$$Z@E1(v;b8erO50KA5WAa5MEQ8X;D_w9Y*zCK)f^-hPU#I4FJ zz_8Wz2H%m1ODw4UXAP1w8|3>NA#ScnV`D$G;LQxZ&-9B3^1kZ!1ah*`wPyd+^@1r_ zC2~C?&~bmgne-IOIGB<9s|2ZX=nOF>clgesLg9IlT+>x7*W$teMlrMO2Mp0H=~|O@);gVBG9FzA5w{Mi_3hXX0IN0?{}rCw?#zjV>V&>cWdMa0UR zdOV72klAmJztr<`+CySuqn$~z*0_mLpw{t`RN~>3f2w=WJ{2I4)^{hEV3qep8MUK0^+K3%OK~OPub(bPr!oX(^wevwi8-b-m8S!pwxxo8U+FU^wV8u}ehm z%W?50+>*zlkm8J85y%QirM9c+@vG4TtEU8)jp3kUdtFQ4noVOD3$%H(ZI&x=gTg-r z#sAp;Rs&=K<6lF)%@6JG;|JXxu{m{qESJLd3Au@~WH{xEWe6W*S9OVMyy=~18~lVp zpb~m#3*`(!H)1^A4U5KX8F?Hf3>;L-yDs}V@;8zKJI2zX%H73Lbhs205ZfVtd)PvhaXCAB1WEjfhP_4+)MTSf(Td)FR$gBwS}@NXZ%)t#;O~7B6G0V^W*rS`JYms z3`UlKWG8)CZBP8b!Bhic|KVE4Z#+8L0qfDKDY#`qn0!lz?o1=X67htsuy>1%uO~|c zj3Upk-BHd~FgW;9>0jw^(8!YCN?KWaFc^+*Uu$Q?W25`@4gv(LUsDVG3BjVdRs zJpOSg8K-+B!}+wqZ>R*@;<>x?89yu>9i8D;{{HbID26~_jfRRq79p(Dn{gFHu$x?V zdJzSW1rvxsSPUw{?x}+DG1WO0*l8%!pLFnT@xDB_e@VX zXiC7TL$s0HDda77ec;M%-By%XA5Rq~h0kTEDvFgaZ6Kq+nSc3SmHGYZoZ^p3O|HAk zu0OIk)&Aj5{XaD-$X--xV2tM^h!YgAz=a)D;(X{7#uRjf9y_>7)C&!MS3w}YE|G%1 z^lQ2GM~#YDc@piquJasOUr)U_@QswyB zFmC+Ps88o06v*<(gzM#4s>X*CM$^)sz)q-12S+y+FLDKV!D%?gYO*CiIk8)pIxSvAG`__^n| zqLPxwzt(BUu8%&f4u0LAkk8IH9IE@h^j$=sGj!Mea*Tebo8@4CEr1DdAuyKlJDlqF zMM_!dT>Z$;O<@bBaYwN15Q=069q)I;s?DwW6WQ0U;fGSw?}U(?ZtF0W#jkVlj}q-6 zJZ_6WzBog2@w%p2xIiHthb$MueI}OHrz6jRD0zN14)>a(x$FtUB2ES3AC$atfjGeR z&!YHe0eVK7G#$aOk0m1>o7EnLar$?ArFoqXPTeTh_gLdlN)R-AAhO?`hjDt&8gDE+ z0pm${b-UdK>Y18Jn#OO96v`TQQ+CpXf_lVBLep1TJU^=&Lme5^>+AHvut&gFOK4q?l7kBIo9puzbiVy}^BoWQ_tM)#xS@zb^AW5|y*Du6#?E<#ZQ@P3w}~{(?kx5Ri@|2Km>=d*ll$)a$(9j z5{s-PQ#aC@>1h)CD%P1u@)IrZ`R!`=Od~IcZz=OJa_P=0U{HH^+^6r4s(el7f73cX zh9)QG%f8tE}ctisek@-Auz<>5{4mN4O zrxXz3{6;eFgX15PaYHkMPwqWSk-04-@E!^Se|cmr;4sxFE6jyFk#x?q$PZ64n1O!n zb{D|GTkrjj=PRaq(k~1Q7QjA9Z;39wo(%(U zs=Rc*-3>#_T0IY!~* zdl9ba$jV%Ku`3oSq4)PeaB_V9&e$m4>%pAHzDTy1_+(xBlh#*`Z3S3dAmf-f3xBA~ zC1vA^Yi46eqm$1B9e%I?twbUg$nV+hhLA(2d155GFKmYhKymcaF>!tBbhbOYr=If; z(lH`m>?^njrQd;KrNc7n*CvYuUU=fjz0YRiEh%2Eur4WFPKFmh?Rs-z0XlkEGv-W6 znfgP+<)1AEN{Wx?hF8Zbic;zU6S%I?vWuCz1>syq_H~)rvW2HNi+(-X~G? z8MlX%pSrJdo{*IlcIHOLX!+)RzPjXBvQdb-bmqThr8y55XM*7h5;e}t1LexJ)eq)& zmiQV1&;{=OueLtVXQFG|A`T_mJ`@QS!Sfm^S9+oWK~`T@C<+)G_&o_F2P?cz$J_^l zJ(6vu^QB_Kordg+#9-A)H`0Sn+i^3vvYU&UDsnQf!9WCpS<}|&J>eLP`w_unPP)F3 z0bdh=x8TaU(lJ|85G@*dJ?^AVk7;|fGpj5s>cQYC8EmDd%Jha#_uiZ>5lp6`jL67g z(RN9W-Y#iMj?W$yY>-A@z0pJ@P>X20ulrOa4pLbKDkKt!e>MTRy-Z}7LQAw2LJzGs z_Get18@ZK+!$n+3;_pek00}_$zn~YCEwJ zQL5AAp;pp@>N6^-dC73biSlqgqP&Q5<*u{s0rrFxhw>R`Iw?NeRJ`PiW96e0>jW_p zq6@YqJwjYh9dNGPdR!`%K>fpDOXVLl2!D$g>DMP9aU3^_Iri;YS&apZi0@WE&wmpI zqXdz}Z|3)A()bZE5o4lLkaQTcK^HYdBl}t|jSsQP9~IvTc*|d>0eL{#KjXzW+Hbz% z#f4%mMt2)lKM4}~^u5w&X00mfIBN^T%-1^j1e9Hn%EI%x_gSD83^nVPZvj2eE4G_9 z)`M7Ew+QZ^6z@YE zZhMi^l}&*7i6_fG3KLC}M0m)mTgI<1elV9Cy)hG~h~qXQ zBapP-#Aqa!r)Sxqet{y1%Vpm`@GK^-kep6qorjdnX%|53@;#bTXh2pAQcwWl&uB}Q zGA__Cyb;0#9dndU;T+3K=az9D`#Lq5eI?3m0QlvHu`!`I`z>*CFS~sE~U$h=xS8Z}O2*MF^T;?@gNa zBuWGjOC$}jbY&}ep~?vc4Br-9J_>2P@4gz9)=K|hM#bUxt5M-}eKo2%tWoiq7zdJ^ zN)N4aG;LKPxF~nbgY!)wjN#7HG?xub1see1;3$YJW?#RNIUlyMv5XC1Pm%ZjQxOvJ zm>*+QDJbA}ukRtE)TKU7Q*R}X@naZkjT*>A!r_)tVgVsVAO)-HfndE|EsxoYvmK)eP+-OTRe>)rR zf{jk(l1SJiOmuQQ8OtqIyhN^!v4(=d3hJdoN%1?zQU%1+JbAGii)pdYV;b)k3!fz{ zkN~~_#5!;MuwUZ|ZhvxGChqt;Nlf207wR%Ve{k+zmSN(6z{_J>?a96DA#Y!ZjI*(3 z4-pj&_)8n=oUN9;r*PM0#7M9*rkezUBBwjd^59&Kg^60DMmpd7|o35^zLZk zg6IfhadssNni=BC9g%XKX&zhMz~sr(9&d3m$VxUF9{K0+@*##PC-EOb|!Z zw_5iqE{V3l5Cq5I;EB?9=2UY)?se+L)v~(SFu$vEdn*oe9a`k4+D>YfoAjiE9MX>c zH|*`}X(G;nDSS|0yS`1AWxOXw?PoU}Wcb0c8& zmfOS8a8*mlH0lRRJ~x#wv8a2DWRq0KFiew3lTDg{`}CARgnYR07-bDTs)?5g@VJzVL4WNMGibtK4yv5$6xUstkCOw(yr?^B2l?_umTxPA7XcX zi&sMbVz_6S-UHz6F0x#+B6td`G+jniFSaF=$bLT-!G^I1z`g&5h1t2r{QoPky5f)n zhu%G^feSwT#P0c#g?UP#j<5RpT%^OK;l%6e&DPh6Q{ngqx+1_p0=mfX$wd*Hx3Rit zz4U{7D_sqr$KU!yKfnF;#47H~*;BwV7wxAfVT!$&+!+0l9J0MKAa;|s>d@(P7}Q&G zYr?lAl2O>6^$TiyxiySdfZ9!r3^B+=^?61tAKgysy%%FXILdEdyWsB|*?7L6{MZ3J zi^L-B&ag9{x@9vtnc-fyKxJsiu)(q5y@1>K=jqt3V-MHlL#Iv*GF`1lP7wfP;e{g@ z>&uz}k*ZY<8QAd`_Q~vul+9r98av(HQxhiRQsT@bGY@b^aNsPIOq$qJa1H=iiEMU8 zAE#I}mJ~cPoelYR99eMQ&;ljQ2-pP`J#v2aiDuY_%CDfnth{cJBFK5ioKY%_L59UB zqNZ)sswiEK@lKXR4V@+P)8wRl4vl5{!c-5XpKjO-0trCO@8)9-ehp-DBI5l<&0h`S zTyAUB&c&k%XoqCIFH(I!cdkoPK)=Oh({l8nu^CGk=#M`w=ZnJju5u%=}P}lwsNyfGy1`! z?n(g}z-#w_=`TYNUy=6nWBi-8`2WY6$j-mp-fJY1L-@Utz|Zt2yt>;PKyk(0FmP8v z+$4~~FqAD|Mq-a7ctL9~EA*tOrKuJ9BSJ{nY2E_Xh{=)o?4UUb{2cd_cEd~&0uyvI z5Mtj&w#cca(3x;Xi%_O$2$y26o7912MyxFR5=ZRNBlo$uOYA(i18-dP$dfi@Xt^Op zaMIWynRvWOCdEn=p?_Z+Wo9fz3dgc@)e4B6;iW4)PCzGU z8A?h=u>@A3ur`!Qdmz6k4|tDG7v?F+Ef!7x*cZdQuwByq80?SWPj+OuPqRj_MdvEbEbcb9Szq$!f05#A&zyJK1`FZwivWSgD~bs5+iB zLKQ-ni152SvKbvfCX}RMOqnxFu6$g(qhD6qpdu-l_YK~3@TrWJY#Gj!q<>iQ`@IGw zJ*~Zlw$&q#OWSm5(b#pM#9f!7zlhp>3Bp+2BnRd`AXRK)0Fp)^=ooR|Zu&yO{nau0 zf{dF#F|+*@%=6;4m#M7D>H3bYQ()r>d`8mP_X1C(-xCoo<*P+{W(h?ZCk{*>=YF;* zi8@UjU-GDS=ivQQq3>sDxEM=`{hg)sAp`(41En|jx9mYp7n{`(%p`^l8n>H5(Hf{P zW+jMB76Ejv!+KK`V>lNkd*}*s$UH>*9zHv_7Shim=9}dHY>TV6KjJi`>%QL>?#Ze+~3Dy(Rbzd_K6xwh=9yT`U-Av@kxu_1rU)cp@?rN#2Up=DcH zSf0xGNwLetJo3l6%HQ}s1`7l>Fn{ZZqt6l_sb@vAlA;(f?%w7vse#&`nhr{(wYZAK z0#hAc96nhp`!hYHR06QUlS(cYG7mUNH&jM8S|Vs=bCuqQQWt1sh#89CyI4_jGgU#A z68X%@yjxhH$J?M=+qkU^efOiv5>?@i`?j0&C-?+&CceW0=X6zwC>4K z3Qz^3N&b4xjEe0Y%xfk}`Xaq!cKk*fRL|w2&|lUUVswDz(^)c+pp8s_0&QFsK{Vc5 zR@pZokOIZ(wZ#|-z@Z~Mn2@=J30u;b-aeJ_?_TraxnRs|`X-u2s6HngQV`OXeH6)s zM=%qw{yRg4v?ry!R&$z(!ODMCN>O2`cRGn8RzPS7c+xv zxV3ZSP8!2=gs@c5ge!2|Tpb*W*av+6;j}>hXwh=9y5F*O#ILu5;(yP@CuKu9eZz+u zTeRv&_)sHouXMISvK~Ov?=8r8vj~i}xCpD~VeldSRcgzP$x=xy7tPUZk(^3Ezn1%k zaKS)rtf@#I!ISAq7uCCU{HKA2b@ zk0@JE^}o|UYTi{@X5xQ|{JF&IzdT&1qZN;es^#vVHgUkS@b*8!g0WNc1K49A%(}vx zbq$o4pr2pfQC#+ZqT2ph>(XxNwK^~wf-=b4+b{xw)JmD_CX8V?)tFp>f^6(;u4GPP zxoF)W@Lwyztd1Y}lVV{wQj1yAa81jpKx$ogwqQ~#!qC8zym=-;Y(iU8VXO)6F+rWc z+str5&z9rzJ1rQSxNtu7U+EDiOohxPG}soLN;DR-%yYSXV6G88YH-sc#WF=CA>|TQ z7{z%amowy>EbY2^($8WMI9=+I5ljFq52EA=%d6aSC!tC-9ED4WaqV0Or(+8usA-i| zQRa?2c~QC$ODs<`mS~ruq5>30JL5o31JR;g}EeSfYcLr zoc;SS7v*PNSH&6F}?sB@fYAt~1%D7NK3Kb@I)?n%A4u-lXI zSGWx*66JS`>E+Hw2&T)!X3}Yaf7KMqqG5D7bDjS$Q1U>^c(B|CN^1EPO1}DE2|*Ci z<4*-X8<(C6s@LMm3|hl>d08>JJ>=u~03;Oa)LzUv;yGSiIUVruPZe~g;I(wRoN&5d z$aE=zbLWU)#{oLl%y|6E01&qFsO4?K^#}4flDzxzlnNi8y-V=2z4qQP#UNkoy`~Ot zb?Qyv1y6yU>w}+lIRpRZJ^j&ujEwkt_EDCyvfR2Gky%`^=;MWG=(_>Z0*>*ucD}?MqKgUdVd)kRMtVX|)rX`qI75UzYw|8|1m;e)qs8j3+C!Q8f?Z&jH*v3FEs z+{yh%)ZsGVV4lH*SEkjU+t;zCEXr$o|2f&lx#_ zkC1IVM%@7;IZlAfDpeR_e02v&WfDlv^&(#5vsU4d7Q;k~NeEse-gGG0R31|pcm84w zjQ76>ZGS6*)n$?MFEes>a|kY40*UtxbDvx!hqOoIt=b+@mHn{MRXP?Cjft-HnCKdr z8Yk|&tXYW7$T&WL`_jxDCL_lrSXX1B>jMe&7_X(~d<3oim>|X*`C2k04YUt2rMs^1 zt-4|Z8$s*QB!|*^apyQpbOqozc&zb->Unnz&$KH4*s)Pniu|t%^>I%j`?h+h+xB6% zo5Pf$SH4V-_P*o1YP$4IJ|@WM(uEQ;C@~>q_w$GM_$V+ncdefB`~bB6={Ij2BK7^N zI$K#lnFMA)!1%)yf`nX_S<*zjRjksaI2xD`dp?vkh(xh)Dv*vL-dG6sZfGi$&hIuk z=J_ls|I6n)i(Q4#^nifbE8JDV+RSj71m5c-oC#;`MmWJYD!B6>lEOP4Jygk~pZi&Y z1^J8);dpc_fEunK1tJqnNsQn*bc)VveYoE%k!V3&zsi7yM-mD97pFuaqxAesf|0xC`E{cLI0O{28FgB**2QjU^vCX#cOeNS%1 z`I;ZZKmJfj{O0+rr+@ynd@<*y@ou+Aeb8IcTn05SZ6}+}zw+qoH7?x}W;O3-T)tK9QnT1b^a6-f z&CLSrj9s~vZc4tBE?pn6kmhv&o)7e9{#ntm=uxJGF)S845;+p3i+vk)!6Nc(Q^(sF z6mr12q-36+j!9dJw5T{%jJP^iK?Yxuqc~BB;Bn;l0mKVqEerrD##$mU)6La9KW~nCM;gSZVzeI*OM8`0aEjJ+`?;*QUQ6V@x0CI%;kj+J0xpy zX$Ig?C4TOWS1p-v4=Ag^^lnq1)`_SYl%25IDMO^ zFXU>*DZICrvmNI${g~a_(g+bbMA_mIID!lVEbl{%34H(9L-eKY9=r?re%(zFU_1Jv zFs#NO=8rW75a99RsySpq{0UDn)Gn;C%7#%vt39XYgCbajY#Z*YXX+-K0NeeI=soA98<92={;M{Q-ZL?KG=Q?=06QrK0JuyDOCN zQZKAd*35KIw4}f#RO*FtY>7r|?Xup~LP9I)jimE5%VGgUtSE7lG4L!_ASc6p%4v~2 zS@G1B@visN?}{K#4Ed8@SPYe0(%0{eZp^7YRvPF^CxOgpW3;r`=y;7LD3 z7zos?Da;NPF&yEHFdcO#zjmihYwlM~u4;rG7*o67uN0gr$RtG!!?Un3Kgy4N6vz~+ zcV5g2@9NOZ`kE;q3^PY(kPB62-QO08S-(gvICtMFS5gcg^ivV$`*%fH12f9&Yfd1& z{`zRJLt#+ou-ViaEqNM>WZhdBYDi?1>Pkyq$M`*`QDJR1C0xnxrG8VM;)}v&NeqaBH`03fD7g_0&f@XS5%3FyFk1nd`coVZOt37wPGy1N0faew< zWpqHrSXpX>UD5nwutOU~r1M{(y#EzwFZz4X&KOnL0G(|^;c+9RfU~p-YIw}F+B$vw zdkIdFjV`?bJ>$+1LAz@NOy883IK&Y@YI)^oq0SU!LkQlvL%?s(8_$4%F`RLJm5;-4 ziN3nr+^A!FdnaSAQtppQ$Pab|KIl1oE(2x~I_DR60tYq~1hjCchjz7E&PQa2jp%=$ z(M>V0C;W#50!~O`yQ3Pz zu1JGtyuqy%sch2aJ*rj=FETnu=@KgX2fJWsYBbbs<+o!w@oAaBU=Szh0j^q3vl|FB z@Yv&A83gVhv~%pS%bnIoH^8Ur7$7Sx#mZtddnt~enrJ|8etf-%nz&Umd`P7!>jRLK z(cM*}?d6C!ygMP3ucp%)MyHSOgzF$pA{ISTC41^641<1+d`Z7=P64ba&&9H@Q1~(7 z`UE7J5gYk5W&~~ujMX}<&!~1M;G&TS@>HhSnU!Acc3Vpq=1&sVfe`JGTf#QW`du%0 zU`BS=)7;mm+ex#pnsGh4&|frE%9UHc=5fNNKlhfczd^Sv+dJADB{kFs@;28hY&VyE z4lHu<(YZ~U{gnE(#B%-6>=?bBp5F79JrpMCaEAKF=-U-3CW+U8CAfGoCH1Yv-Nu^b#1 zM@10E@jOe`;9BT$1z=A&MuNsp)8t>P5&m5zW7U?Pl%T;!bCWYn&yI)PS?}o{iSzCl zjzt7ICOJRx)J5aK)MHf&ejqYx1VF`d{YS;gm!WLIL0sY3DJfzRMEYXyIyf2&jqbX3 zSUw_U4WE{Xzv9Ug5m`f2P;h!e+ArO8dTCn5tcHBC|1pqNT3*7 zxv<&MWAo+9*cP+X$j}LNdC0)TW(SNi2u+Tz`@80aJn1nA`Vi0?oP?tBs`_DYYRfEH zYR<(he-a_;w*f?QI>*gmK%7Z1jVZ1QD;#&zB!rd|q(E^R=W34ZL`79jAo!eUe+20@ z`TxG!HTNCyIjVM-<`k|@{10v4;YfAg{{K0Sy`zwIWK$u@3`b;VS124KvxHQ~JPxuq z*?UW&C6!r5$cP5AHw`nZ&9CAQ-+*R#_w)0^qA|q-dbh9V7&gSu;?&WlJj&0p2`v3SGlgnj__Yq zX#P}G4b_;CY)Jels_0ckkZpsCXxMRZw#8XvQe+(Og7XeooB$Dab>UOmnIv$wr2=MB zK`$(W#Zv9`B0Bt?eyx)!ujIQQ`UEs{Ul+wPC_HFrc0E(Sp*a z4^JGJGTo0Y5xkDG&9J-#r#lu(NL0lXpUpC{0( z?U|8GCauh66XQ|0<@W}c#=QwNJk6rgBf3MgUunhtzY@d5zpgj9eHSX-%xhUo3kabU zl4A?rl*2pTFE%1gu>$TMQor(zq}Z>Y9j?w9IJ~>4YWcRDYNZGus?k=Avj8B6wbDJ~ z5VYDKg0;RLEsNLlR4>)FiO(d>v(<_x3Hy&`+!2Gtx2OoPP@(%7+j}pjbzxu=tb6uW z620KX-qksOE-x7VMq{$dgWQ&+gDvGkk@NoHFgVFty0|WPD=318bi~V1O_wvJJhJvF ztD{g~O!Uh13m&}!z>n@k;^Vspe(DL&tQ+4C>K9s#H1+i?ShdILHZ=7$)0@^-?ntl= zw+jDF6KvDEaR)~PE?l3se|i-q`2KM3<=Np&;TxyJ7RCPnk1#OJ)RPwxV<$3A(kc&? zEYE&C?h7-iJO!VfQ+tGpz`9{chIrUqfhBrkp+JB}8BH`*&nAZjs57Lg{;`m4Ss!>{ zb3|HP7}*A2R?5xGdD58KG%B2vXG3W0CW9p)fcKpyqEs_c}p5n!lo80o*C_#E`Ge zf6z`;duaO(+1#R=s*9#%v3BNuf6uxzN?}!mUo93d@P$ZWX^|0HvTBW-KZX~OI^WjM z;(mS@9<+04QOYMiJoMZpCL_$ah1A^=OpJO5zrJP;6ut{J9q01tss2* zh=J2!TeL!!YKkN^Uq_PKO&a&L@H-!hkC^DMGw6JJpNHtYW7I`$(~>0mfO2t&_Dj!m zBrUdSCpb2(AX{cN{<{XY!S>dHYx)$p&pl2~JN~(<+6fB9e?9u8^4ax`dY1BKhrzn= z+4uz+7~#Ax4ZK!iiCo= zT9EgUGC~1!v`ou~l5Vooci(>}LJc2A0)V`y_$Wlaf7tX_Mu!j(2zdR5SAD2vTeYn& zXiK#%s{xlL%Qn+ww5%sJ2A3wALs-q5lAnT&CQg&=#Yh@2T)}Bri7PmHTVySbleL0` zaRuiQqf%VKIbCOBA3Iz=v{(R_LK(k!q+0q}?yjoI-@ zmS^fJVs7YFG?X~^kH!4U|593irPpy;S~N|5sqcj^RMFuT;5XYR|Dy<@E6p2%9ARw- z0%rO;;lW_%#r<$;7vOQc^i_4};?ygSPypZ?+6`i7DmdDq(7uj?ZYE#ej>Ufdk{z!* zUqSHRO=I;Mcl<;`W3KDLql%ZVIWmgjvX#c7Wvcr;iMhwlNCw(9^0#*Vd0lk47?&aV z--^$g)on4EBRIguy(-wHO#-{Lk$kk(RA`jv`bQfboL$-7uukT9$VEx+SyH=ivFpoN7ex=RdJF2_i1thTXgx^<4> zsXp>>Tl&NPRN4$J1Mq!uq2X~jtjaS00~r}Wkw|b3qM@8E`Upr29=B z87&T7lpN9>hr@u&J-xV!q0=8QPL5a`zN8Htr^kMOyjbv|STaxJfZ^MGc(&?QZ=>dO z9PelaOcU$)h3?gsA{Q7xiyz3>$~@o3eWvh9$xz@S`ki^Hd(Ox^@bJHgfV5J0Y2>2C!WgZI?*co$1e?a6B+{n%G!h+mBJBw%wn_p5cFLM{HTMLaxi~SGgx} zLuFU4+J|K<(=vVW7xNo$gQ7fcE?qHXOX@tXklxa`Yz2q@o6#sNW=&~<18f|5Z{8kF zll5F7CLOHbZX6ZV$l4n#OAiT8Qh5v8^*A>@E>uqC@vJN}{l+gRg?i*_)LhEmMsI3T zcG-pw{yJOa<8&YHNhX!RQC)0?A1M3pxziq|>ndRzuG{hMCN6*0`ddr&gkSyCUlE%n z-JcHOq7h&&vqZd#W@<~mxH{k55-ID?>ai=HrtPv6Ed&xF2=i=3EPM&a$Kxeo!>$Ed zw8=s#84${bW7pXfv2QH~z3FXE(E9GE@>?KCIH35EB(Rs#H^k$zVo+#{oNE;{Wktm!~od#ecPUF6CO_I?~TnTie zsISUvhV=;3PLbtAUnZHKU$|(>;o8uvRNNO%VGzN6;rodi_qt)ghb8@BLi`ZXE8VH+ zDN4d4)L&bkT^}GUS$i@(SQBccSadYS*}d4)mzHe zVYIBO)R`nSlDv+K>#P0f5LMf$oMqEHQ@JXaY|MTI@W+5qz~`6ac*~TOnA_CVf`4h? z(ZPQ9MSVq9K5ag*&9X{f?5cXX!`#dI@FaW>K-{GR2S(rqH*i)bc$Fjza2}ZS;b{Uv z=7E(@^bMIJk;9ymivw?q5e72uj^E5*-NRR#CvPtOQtNikv)1~U*bX>Q@-?7t$Y0Kt*}Jar+19h-q^jRm%Y#w z>aw5j<8vg{SN!h_N?P$}zq~z@go6k{zB^axE{h-Y@{QT?jr)UKw!g8V-PHR#2b=st z8ILOU zwxY5Xh5O|RXf@B3_=bR8D9Cx?>d~MJ;9fKpq7mQ!$T^ql1R{}=oM7Wju<87I(h@3M z+3A%8^qO%-?!LVyQ+gru%R$7*rm}7}ZCU&|$0y}kqnW(ZvAw~}s|(@zne~%=c|55> zwtYwDp>KGcW$gO%qF)dl>V-owbS zt4a?>Px~;Q0iT~zGYs!4`4O9W&Ki0i{f1q9^(SQCE*SB5J)L zge5cmSz6U-2$A|?lMEaK_vyVcaDvyC)}w3q0|&{fbhx1V6GoN zGR9sUeA}Xb>E2k!{%KZ0%<`9k!{kX~MFY2h=>yce0r(EGETRSlP-9iW&5<2~J}j17 z35;P!0>ht`A51gZFk{CD&wEBx^~MXW;=XR$Q6Ib#0^RYJEam7^uU)pPDBz03Y|iTh z)rYqC8W-y{3g8Bmk90-G-#@e#8+RYtZ`<_545EJ`v9zc!$3EFJg^?cUF1)$tN`U@- znH^v5;Jwk-?JpOers=aptkT{;93eUwSm3Q^_*A+vBhZr}U(#8NH>7m;4qIl>x} znY-qjy;6S@!QTHG!RpVqRLU~Sesh02H#2azcw^)g`cmTum@@GJ@j-O>X}>{FvcGTH zw_2G5^2WY03_23F(k%U?8rc@i)_Gm(&65O1lYq!!6CRmHKcsI^=IL{Qk%$k7t&&K1Tk#~Z zl z*V?H|swH?AqHuW0{^bSx(ytiP5uS@udd)p7>I|lGzTRXz9_qzqA^GN=!Mru(XE0~9 znR=fu*yP8OblI#Y?aDxXpTU#r59PMnbrXp%rkC2_xN)#&$1K41@lvf zzD|w=KjWj%l6~38U=hyGDv=c>KuXx@c&ce~bD97T1=nLBlz`W0`0tM4ae@Bo%-RkQ zrFgp;+j;X+DJ(`a(}CV-s}%A@zLB)+gpyequEKJ>dsc6wwy*jCD+GgS)wr^eQ6&Bs z!7$o`W?KfQ{aK4>;jLuMrm4lTnKva%biO8|?=eD{@_gLolD2UZITTTPCGdSViMPr_V#paSLCiA!2-pYJp?sm5$~Q&GXVW z{BqU0!uwz@1LepC@#`0A{E6tXa*1=sC_|~V{Q_EdyEi`^<^9hZ6R-C|O9{LUDj1FA zYUD_88V^Y?tJfo<4_DQxc(vEpY&x5(#OdNUHm1Mi?Q_^zr4Mn#yP=nk8t1ZpgQ{7X zK&Vh?-~zK@jlbzNCNMD5o+HM+6l}I>AVFXN_)_%uI|>M7n}mp%;U|06P&=E2%7IT! z`n|rE+eVjeww6=}L$`%c{bcxsQ;cuVS!mNOp1XbA<<;3}{okM5=2pE}xm{W4WEQZe zx@+VO#Ui}@a%_!A4acGb`pjMypT9sT7N429E{h=UL5T7tuh0@Pn_el6aD8NZInFs^s!o=y$LzSKi$`Lh8COewCG=6cdOXWWnw2y4(DO2%Z)?h(FIXQF{A)R{s zY`VP92C2+ReIc`2*(a3P)%$Z{Wb{0(Wj4I+!&N3I{2pL-#6NVCpz1h$I>CenGFWc? z3DF#jGA5QjMw59uMurgUyZVx#h2Bz`=Fv)CUQMXJiMWTh-q4#EAZl&pnCSTHM6n#5 zT_$5tmS)g}CZZRx0(bAZh8Qk;FJF>CrH+n^jlbPB6rM^ z(QExnKZDJsa6Lgg&*X$pJ9_?+QduAy^FvUP(EzQ1!%|OI;YJ@Jz_dIaamea>Jq5A; z+;D||7GeEKhIIJDvxn(h$|Xyt?x6_^S=GgtI^QQ;*uV&^vELl0YHWxts&ukKeEt|0 zN0qwW^G_2=$NtjuM_d@ESRC-Co?agC*(dC$e(Q1#VVo6%=r=8WmIseb16UJ`=DDvbS2zUkKNF2KHiBbVwV(geP-9DEy9 zF31xFe*WlWPh3*E5Ia2~N9a(%fAylU!M^EyRpUhsxd&!}ckV=x7WCviO%&gJADZRjA>_SI<}tOb!Jnm51-jd%TUZa<70VgxeYsCgOjH!nE4+4uw(7H@UDWN$ z!#U4&Q{61~7N&uFye(jM0PatWI<_65`F)AKvSdTIBJ@aAH0tsJ%tF=}j2;(I9XalU zg{hc~h|By37!&mBBg`R@+(3Fu+wqG{F;glrssxl*Rx_tb63;xeAx=R9AAQr5pnhh1Ed9b3Yl zC4@h7u^0jXu8yJ#;TxYa3w<{HF-8a>w-L2q&8rd&e?MVDI^w_ft_JYPz12Nx*?g!) zPYih<=wFQVB_s4X1aw}~*P-^}1kbV(*Cq$hd-(PT9~sZq4^cCvXWuGJ@oqQ#OHxq& zJ*Dvk@uGliJ6Ac2@xxW>Wepy9xm<1@)V18qZ7; zPTI4q4qmrs$Z>I5y?1zW$C}jU@t4&87rxn1q7GTn%OXdsry&Wd1=DwnUPx-)B|3=! z?j*2!eOGGy(PE^tDTI?@kg!a>6^S|vtAr=HN&A{ z&g+SsD^R#zhFE@OwQA|z)AyL4mR=@VXbHY=iHoNJvFTLBvhH3dO!t57t@0wtJ;8fy zyZB<59Yn*J)WW_acX&|MZBWtA)n=$8n*%y~ewM@TZO%poHiylb9|dq>YNr43 zO*cYCFigSZ$=cM*B}&G!(FAs^>SE5Vt$}2Eo*s^d)3q#48#9$ICs%84qZ4FNF4AuG zZ3BNlqqM&D?s2-#u+6vhvy>Xv(MVv@nM&!afpw&=B+33{e@4@QLxIXE0cJta>t!Gs zF3SuS{%>hzQ}OZAnUBv#Sqj*2he{R|WGAWETSp~dj@o_xo7eL?rR}8kZI*NkQ}8}q z7Ot4HfJT5lNY(xhg&Yd~^P^HRr(A7v!(?pXd8%ONgot6bI^Un4Fh4O*es3z<>l&{j zaP<)t0hUo*o82pjVjKQZvv_ZgufuS?xpqz`a0K`)fu!<75*hlkemir)wZ8>-`ltSXTs^_P2|BOyovNv-_w zOr>an7PYdH>Hxo=xS5JQY#TSr*xdw!%aa}{cfm(Cx4fbN)tG0CQy=irjf8~6`c_>D z``Rc8kHo`2lu4T>OG}A=?SGDDqM{&&FN{aWrNy!sxmv^Dn~2d9-*sJ@89a7)gUaOK zbNIsOubsvF%Ug^?*+XWF>jcUs)LyvHgvJ}b)Fw+N5I<-nowJBA+}AHQq;bF)dCre9 z_xc4~UCxjx%RYR`^#5SY`E12B?g&U2H}B`*j$GENQGhCllqf`kc26b}5$X?L7J*Md z)O$}^B)j#fj7qt(B7ihc08K!$zwd${{3N+lE7J$n?~M%`VuD&lRN(&Gjj2Prmjf-B_4@96&u|9Cd0Pn}U3hJ6J;`dG>PpWSFm`H}vX=gT$i#U2ZU zisfZrhU@J(i1sIoIt`a59QQ9u78jl_qaqs+`1okXWM_mm#C;Me;N<2r+JI2=QhVW^ z5elLeuvJB|AJK15m-Qj%vh5|FOKU)(E}y_G_>69)=IF?*W`Eem`kMJjdMHf%Dn{95 z_*O!!pbz8X3K^$Xva$)Ubd64aX9moQaS;OW>Cg(gV%?`jEMG;+Vq)AQ=g;UC$(gH4 zPmU;kEakgzfBNvF;r0Od)K>W!dV{L}yBZSlBf`cz_SZ*nh>;(0nXjqFKRlA7Zm`hz zI+d#)!|-fmvi0_1I<6qxGwNVjUwtbo&LXh< zL?y@)-H4D}+^(V&KJ`cI9w!PsUq&(K9zIK1q2q&+Q8!bgo5@v7MJGp_E zypfHE{9(8V`>N|S(eB+(XCU;qo4qLLA%4u*)=D@{4DY_`VmHRk-Os-W(808Vtr5v8 z$445YWe7KGfpFMO`*`GPX6~CT(yNK;-4FBB;&~!eQ%K=94d^qd-ZIEz-gm_!_sj?e z{l`?}t{Rs^nr~$5r|P4++WBLwKR!QR>{`2B#tY=%EE$U2PxXH>==e|67@xg^4kEkj zQ?qMzpXitFr6D(lmXr6Vdn_FIWzeDfg=omv_n@;EoP7UHGFl>MBph0QVi1&_J=N>S zAqsRj%mg64Ae@Qfy?$GqrMtYno1CR^b>`(a5j9i99g)B!Z?bO{!LE(#(BIkxy!#2k zL^1s65*%5f8mwRitKJY$i>@YW!*l)63a&;=l~penpUZr$ zkfa=Kk2yOZe$N~3RvGX5qD4DL|DkYgL1#mCzO)g!hw1no90nyAK1uS;Kkc?)ZDmPf z=~sgyDG2h*6f|EeQ}GKatGYv`!Or%|wU~ctmu35K09a-H2X?vpueO}Zx^R%MV^Eqr z>L%3zbrD=Ct`@Uxmu2NHFMlE-M=}?tC}MV(nGl>^xKi^6e@zovcK++U*=C367QNu= zG0WRvFMherhZ2vTRYP1wzeAcU(*IK2h=K7K4*HKj+JM->LWFS_nH;-{XV_cExZ;MyPz@VvO@yETAqYF&JJx5)azDQM`hnZk8Ys?TxXEAI;5dkhNd0``K4bh~1$0MZ4 zD6`qTc_^IOeD|N@?e?Mj=Zl#fwudsrhXNR4kvNm@cY#CYJXbWJrx_i$KaXXgeTcd!*y~WT`* z-(>ngAu7bS^QD1{gF%0i?d3rmCzUa>i?2_tj6~a3Z*k~W zd^$FC@u$R&UkrgCR^4#y5|F1E_m5MUFE&*z&0dJ%*O%7xx1n+%eH#1E$SH^1vIrR1 zQ5;|6&Ka_&u5@_n91_(F;wJ>hUG2y=)Qa{GR2&G~i>N&*sF83vc%^>_DDOE2q3ER@ zl%}{04j1qPa^j7?PUOH5TO#aQ++IKU`*Jmj>n=9tp_=v$R^Dfuehs98xIp?x^KOev z!!(kuBg|_|2&64H3nrgsx-}`VUmgZv8!^kD25?ZtxD4m6nr&$^8|BQZi=(em%h(~2 z3|g;Hm{!`UZ*%hL6`A@EPu+D{91oS-eg- zVN(9)(=cUw)uk!Y_pWbKJLuJK>ZI_;+V<8&`1d?!+76@_MYtx~^!XYt64U%mCcHV_ zQMtpsRbR9hoobZqNS(3Cv)a?|T8tY~r3($idY@NqV661V&(3*6xO9qHE53iO=?Bd# zGqbYv7_VthCX8ni0)+AhmB|%X(oa|GS(_qhHN+xr(OD|*W2^Qe&V*OHlu--SI9RhQ zr~n;i)Zgz-wx1Pgir?oa`p@^%k}fL8DrZ|Mf>R_5)^;EsUfBsG5HjN`V_8>D%LL2J z)e&jC*wyifYq8z%NSD$^AlY!20wXBs@ZuOE8$#gtl*8doi%7P*3d^^t8{#ruz3Nyb zKW9OXYguG|%oL8~T=eEZ?vXI@ZR{2;n3+Avx7{T6HV zUKeX3JS>y-k8LN}?)j0{^b>Iy=6``+>1ABNBzq=0OlB!pc{*aqyN*@=yYpA=3ZFMb zNBZYh?k{0;G*PJnY{1OGlQc#r1RFeXuVQLc^0fkm%qEBByn5q^+2-GgamWpP;f;_5x@fb*NpqA(Bbh_} zeqT)UPrVw8L+}%ihMWFo$h}MdCGVSfR=OAH8^kozzpyF()cxPIGAX!}_j8xHIK3)# zi=S4PK+glpV$5sG$!?f6ae}(czT8@s4s%&~p|H(G|LU~Y@nX^ZTm=gu!A#O_c!1>W zq+pQRL)*v5Q#_h5?LeB7LC#M<&w9Y}tA3V~ zWdd@6JLyQqA;^ED^3yE@T3PlhZ+S@tx*_-@!RDgS4YW!cCCisZL`Pv-3MDzU5c>9G zo{h0oqw@DNlP!l&9Fb=5nwkCs77ijL28dXRjnNcO8xW^e02)(jQkk*{=?z3c#x3Hr zH_ob`CmCh(KTD!$cFH*=ATy97l_52V7Z0u-EOhSdNGO}jpLO@LqhNFTAC4|-sf-EM z4k4+2$N_+gc| z5;i&XrZh@PIxa7;2*ltjBsno0iU#TJ4KS*8%$kx@=4oGh^~>&#BmQbk1Rg-iaa^Vd zuIf#0UwM(E89}h|meb|I;L#L8z0$=Y8lXKv{^~F;P=u0&u|C9G;n2!&$>crztn5uk zj(P%~P5*b&8;Dxp{XbWOG*w&vv>NpBMW}lBuyGX!nen(?bpBv#(t zO(_EIYxuLe<65_9A!W)_X68!q6sJ2x_|9^v#R*;_^-$He{_D65S{t^r4GR9OqubWQ zVPvxEDl`EBE3{*STXV(c}9&l&oo2SxFmME6Fyt*)^ zi;K#D*5P(K{X(;1Zf9b#9~imtroi7Uc3paTUC<6Ib$?FQ4Jvn>FeX4`s=ivMP%%YP zf4cVCStt}=J1Ohui+a)-VF8}17xc(@oR|!!?2bxQ-yu|qcaYgjw+%&XoxuJ9OlNorrfUl+W zbG^YgZg%VZ_I6GPbF>@%j$*P`R9LPp`JZ%YZpu)C>!{uA1l+R52dRZ7S`)-H@sCia zD|GNdKekFg9~kTPxw3KoYhUrTuhmV0in7KPZc|g6^91adm8H7jw%vK|9G$v19|X_K zhe{Z!!v^4AOYvor%cl2ZIQn6~#*$aPseBbEY84SW+3v@0R%;eaVZg}|B5-4j;XkWT zXW{Z(<4_3uSLH}uE>D7NV}p5O%XlGyiMRe1%DG+2qtEK`qY1lSjW{T?I)E}*=~EC2%%lM{aWjMx2TtWC3oMUFr`tCv6v(9> ze8va8v-au^gest^R*;iIW$I37SXW{AUhU%FwjN%3Z)}SoQF179mN|0NZ_{^!X*cP+ z^Gb3ceRpveefJ9aBBWa5b%KZ^zY)LA1WAmbm+Ezi**5Kjs(9Nz*EdsSsj@kE>4uB8 zIz`r<>~B|bLG$(zG!`NhS7h)Qxj^%_v9m6;L^gotfbIS%gNlGoR*iq&)1wjyJM(<$ zy<6$w5jtXCm6?9Ln-gK=;I#bZVJI2?k@&gM7%*RY%sWq1-}%#GEv_JvY75gu9ZyFu zfBHyH$X107hhyjlQEc4{;b=mdm3`-q^94l89_N}&S70?d5Di~A!<%Hvs1al8PwK)s z42OBY+6QnY{<{%Fwx2&q9EXl>ruC&De#ZCdp@vPIIl%^|;+<#3n(&7b^u58ds>DSz zUQFjlS(T`{bNZBT=CkQ7qHn#zh?pJBLRkA=C}f3-&gO(Hi@>KK>b*``*-3_ySq=$X zW!2zR=+=lqi`u~^r0i+8W#;FWXB?@td*#aaP*g9zq8<_is7Mrj#b;qaFva#t#j?!S z8j?3Ip(2Nd+oQsmQos*z^?>>8Mci+c^BdmEj^KzY@l}<%VndnSyB??xl6kRs^#|ERD$@~77=hFy9 zBg&CUqb+IlBzG&mf8*$@^54+*bX)uOk?482?1%!q8XzWS!$W_7o%$ga>gdq|$@blJb$zW&=e)QuT(XD|2@YwJB!t=|eYv=0`0Z%`^G6SJB9=qcXk z@N$D5b={gBR$Z(Rp92N01X24JK^cxnP)yimE|w`)z!`cR~BiFM}Rgnw_jZUXWE7cKfc^}sW4o*vvfN!*L#1#Do2R=36i`dnN}5^ z@Eei|R?*A7Z>>V?cBO&MQuu{fyA&P^+t`mZ0Pn8O1sS-}@!maNPO6m)Q@NQ@iBq|G zl8i2zP(o%cyTZuni8lDljKojH)uLmclvH-Bwgmo8?k8ff7)?c?cvAd_u1D=kA$u)~ za+8C}Sx;|oT<+jF;omGvu#QPm>D+wciOBn%x0+#Mp2WC3Kn1voSdCw765-LjeFQ~% zfPc14F!6knZQnD;iNM=BMwFbcf_%!jJizT7*>NjaR@Rg>+ji-0&-0sZ-q)vm_-h`h z%#qSv78@Phfw4MW|D8JO?&|oZz+Lc(xAOc(wdEgc75`NoGcaE3KNIuztZL}_^S6eL zh%;`gKuS$-f(Iu8mX(*cRnXig2Y)>AN~>zP!25iTP@>P^6Q6%?9zG7LRV%Z$S4$6e zf75^}Z)WreD~l*qVTpceE$&8z+asU~8lUhnbnIy0N^1Zx+Le;jze+obtBxBP6GsX5 zMC`p<#n!XuU!VvMowC?o91Rgambs->+&CMTM9;HD)Kmjd_-m`*iY=9*#IH@?weIk3 zqQ{xD;#fNoASH68AEeZ3MDyrWQmcL=s*$aVdJhscUtO@*S-Wd{n@uN2t>TVR5=g+g zdhzb%ZXJYrykX~qzUC)+O-ypP7*8af7aVwL%Oa}qOIr6165MUhliE0Qu9@S^sRG3l zb-Qu-O||z}u7}j(WyB-&H$sZ?ve}{3P^eNWbJ^FC*N3g>c$r-nTi!wRt8fv!^GpBB z;4+*!Pnyq;?=q*ys@KwP@lcMWKw)AjCtapT2c8AX)+7okS?oXdtzf z4j2Y#Su)uETfWS$*MC}rXe?&x(n{uSz29jS?`U?YM9Bi}2hI0}^1W}9clP`#f4q%j zfAOP^=x3qUjFtsfKx3W(jrnM$xrL0DAJq#YdbsQlezYzi*jg?=^n&Y6 zJi+aL_ol~o^(IwXsl=`a3gIx7$F4@rH!|euWsL4#UhTMSm~&-e`WwR%g%-W+Wz6g` z@xq>p09v)|GVz@l?J|@3wlS?1d7eG04_EneTBK!-ZA*E!<5{U3yxL);wB-U@38D%O zS?GFx!hYyQ@JGt9i!NV=+EWB!;uoJyO}_`liy6JoCcCD~mmscyMyRH|dQ!gpZ~)v8e`$#($^V@#pj_aY^Z%WuWn`W9((OLyX@~ za>X6(ryUQeig#%{Y8ll%ENlB>$}*UeL7N9Es)O)?5{Osh#~N;h@*tQkz<|Jo@~=6( z@u7;gESXW~Yw9{9FrVZy}R0kn3M|CW;a zpIk;}^+JrE{Y|3~Fe$_?TdWRC|2QxK83D~c4xzZ%JuF3$Z&M9kp=D3Bdw_Eu$07r( z5EoHu^R#Qi2&>y$+5hWuE1XD#e=0lRbPv04jAhWrti&7BJ>WTcZd%%EWSgvE<-@TU zt~ENQVBuA!h48x%Dku31Ir|wo6pBk<9R{jgpQR?11u^WRFl+417x=S<2zfLW5%vP; zFTPj-Z~|@mlHSrd*#VSAiN$>jv!C3# zbqUA+oh1HGRy!WdnEvz^njImg|7&e!G8Av;X9CBE(R&q?uKwg>3|E^FZBV?m-sEHO zIFWDiG3fis9q!MA+IA#hS{+x85x*7->L2`)Rzdv(jy7BlF5gP4Dt2aqxi39Rk8;q9mZFLF|Z-#&%7^As~YuOTaVO(2?KcTT3#%WNM<%tcH*cX z%q5u{h+EwEN*VvhXVn;Quc27bcT4AwCjhq6sswus&t8GT-La8ok!SY294?(w#PDkW zR4n5F{Fa4JEp8D6SSp3)hHruFxY{F=6Whv`x_QC31%9Mc7~nAc8I&9G;lGfIghWqL z`KVr#>1(>p(_A2M*HD^t(D*$L>N@F7jn^f`9kU`WPh7maDT8Hk@V)=pZ2X7kUaOS} zc}3#y(-hh1>!x+FyM4Vm59jMC4njCzKeLIAK5t>8j|;E5tF57*SR~8lhH7PDRH+wT z9Qs}*`Gg6iEX;m-_t36E?T6gR>~Ep2r$xM2=R^}SSHoku?N=?JeCJj< zV^w|*f|6K-B4V7!3#<$YKZ7lg)Gn(uKGNK?0puhW@6QB3i{%X<=sBC2LF27#lR-G_ z!fuJ)T-m;Jt{`CV*7ff4mlj|Jj$sjv7{=u?;j}=(BJlw!0agSM zB7APwOFl>wjCYVtOZ}5$&S*@|vA}^J8?Rbg1G0>klc}z@`{)LwFBK7_85*?2;;=st>LmfWNWJp55^kegv{+l z=Dc%*Vd7T@@dte-UdA4|1fFCa45sB!8}41rGHo`Y4pvR+*|#ZgNBWsWg!f%B;;u(9WS%%o>)L#a7kp`!}kv?j=zB zRjfds`1E!eTUUv36-rll0^ut7!azM!hX(MWgWsRbjp+Gv{Ar6S9F6XU6(5XuL3q_h z)4O6Zu1jCuRr$NHCb*!N$5Hs1Ow_1#p>C)b4klXsh7n{YT-^-@a?lG0h-~ zuy{luxbYl8Br-eVKO-WP4FMYw{@Mg87LiJ>$nYoyNxsY&1yMMAth(^5McB^Zys95= zzd7L5sD%d-o`koDvEiLyyu$oIBuVX1?&%(X_^P~hMv>fVXXbTRqJ00oAa|F+Q3v#_ zL4#B7`^(@8lUKLck&9sQ6V2_3(JqKe2Wjhy%ph$&k_(gocu^moCPughAc%5bCn-ns z-`1bsl%7;wmz!-9%sL)t+o!q^o0O-Sr^BVMXL$}6$jvZg({)VwN6uplg1&))oWAl+ z!Gho04E!taTGYyB;eKRPJtzJMTC6S}`Q;7Ms({jyLT27i>^}9U(6i4pA8)DO&-?x6 z%*z*{P|c$92eZR16a&!`(tFqDehuWN(BP}I@$26gmHHdfWJf)}udgnfcrmu%6AKZs zDi11oapHyeEF9%MWV%G|+~&(9=l?GQIhUd>{9DFNA$5OzztbG!n9!-C7b zI%CpLZ!7-@2kLmgD{@ys%!`=7VO}II^!I|HKV3O5%_Z|yUJ4lc(|HB9R(;T^+{Vb9 z+fJEEe7}(l+aqekxFWcN80){`*{6%#BE~v7n~vWi0|no;4Zzu;kZDk|%0tTr6MN%$ zq1Bf-)8l&p0S7xbC3h;33&XX+F}Q%fa=e{LJ3-~4?fS~6^D21)nRxdk7eU$;mw07c6Dg$Axb{ zG5cB;7cl@H=mk|}Zx@@XErviY6k*CN5NIy~DB&xY(XlSHfq#g{vn51c9lo} zqnS?w1FQ!(65-@T%3d3w8Wj<%K0?|1C~=3hrG)+)JB13)3?&NVd{fidu_qaZuz#7u zIhC{)8Z7oEBBR_X2s5mJqTV=$U-BjsbO3Kq_z%?g$Gv^0^fcKjk*59No=3!! z-sF?3BMqUcfIB_`;_&$9+vi$ocpXfuX|RNmYY%@DulCUP!0IOoy?k9XuE15Y1Pfdl z&eh?TV0zr!H?>{N$G!b)r^Z%a5@}`N1t-?w-hPjt6~&HV&n#?OQ@c|`0BgOeAwZF9 z+~O)Sj&lIDEzPB5G(o2k)3QbLBlJ7M@e2=UM{0w06$LJ?It@LKmTEiV;c7A5b9sN z37u|AR?*E3q!%;7bh(W9<50HOix7i{`ERoJovB2kohCE@nFEv_0gvvyOW=u_c4qvR z_p;#p%b+hSFDN>54{tCet6snR@d@=c;p@{`pWvkq3)2K8LPI?je=05KyGlX0eRI0b(a7%D}2zLqhfo+cuqa}Y^Ga0o1II_gcfD6#C!2lgX zoR@k%EFdo#4A5yD+W|U-ZJg&)jQtIwz4)5*m)tA|R8eA?%4XAFdSh8tV?Z=bPmjX9 z$&Umx@hQ>dfcDV9V8P26eHZg|Dxee|#L2hSk}lV8T)>sX!tx#X4$Ws;lchaq9??3F z)&>doPmY$&wYBBSUr9ab;xo~lqwi1W%>P4m?nF{MTleGtU3HH8pySmG<;#oxW=`o# zXu|tbm3CSmt0CM^hQpGte82Ja=`!*op>bT<7s8vjzgtqfKmL)MSsuPZ$$j#pD_>s! z(ldX2%5yo+SlV?CXtEQ9iLw3qsn~WbY;1-y0^Fwpmy;-&7&(zt>|QNk@>5@7#S!lk1joJrObTwb74-|TX(&4wq?YiX8{^E(f! zB{1z1?rS`S0}A|pjs|M?xGPm!a+Vp;}8n2~v2ol@k2YGcJ?VCgHx8hf?EL%Jm3bms%zPM^J z-g@*m01&eq1z16YQJVnJW+wn-gW!d1E-!5AzIu@w`d;we-*;R~5zjz*=yb~!Ftl`t zn*MDPFjru?6aOY0Kul)JDcs37J*7nIVzF5O>?2}mua4pa8Lir^aQxf6{T;Os1gp@c z;29C>aa<|*Vk}U@ue&&d+7M5T0qzuxJ0nxgt`jEq*$vghO)6n-9!(wIMK@c3?bxV|RC*d0B#PB(4s)D*SlL9ztH ztK&x)&fE7@Ip4RuUh}TuJX&K&kzm(TYP-WcDfahOJ*Ct_-HtZqidA3zh9Xx|uj9aY z>&U6A%m?i2r><|vjoDC1A>b+i*DTGR3n>TCWcpKH|65()X)vzd4oiD}Ua!B>jR|QA_uxl+FPrOk7$~TPA;r+<3TxuLFe6UTb z+W%^$O*(=wfjiJco|GLev@ zTAcb|cXilIuZtJN^22w2?$_*hX*XbfKgwQ(jhK!yA&38v{dKhK{L;Wj4jtglZo=t8#xO2C(gIBamQK zL6*JW4}UrBN3|uuuN1U#D(LmfeXt))7DgU;ET=h6fSwV<6_u+HC3J22a~aJ)=yZctcxy5^<6JFHf$G$Uy~iLeA!v_$yv&2^!2+o?(T~+H@cS= zKHU^}zJc8U{QOLC{-WIFMYzXbGQ7W~+}H_FnbC^u>eboAugZg|yQ0|Mnk~y=<|! zk_tN`$n^iiS>>_+>KEL1GT=jgv58pWady;x#62w3tDOU?C(>I2v z4W8LVF$6)lRg5!LW}-dbkudoSN|35Epm`bWV*l!2?bXpV1kfpiNziu{g?o4tc|kl2 zDHq>L@udRTt2|s+6&iWO+bx%Q;uHac;zYBe{enP%^p{?EK|ljQ{`JDt^_TTy$H_OG zCz1y4km{X2yfabo#9iLZnxd{$KC504h~zw>fEd2zHDz$>ic*_zV`JyyWoEVu4562Eg4I!Q#PKO&sIuTTQ}%gVP-`MGPljcuC88XqWE>8w8BO@>sXbOWDc(Gg5v5h z<6z$p>1qA{oGkHIS4&WTC!%@JE%d|2QU&3RU)j;K$Y0NUB?vpl_usxbKeoHd=G&bO zV*LXwmX@9Bu{x2c*+5uu(uPfY5~pGL15<{D-gI#>Y3%k||BDRStG2HYXG{8X=&Xoo zc!IAuN$_+`E{}!UF**gbO=N9-i4{;c&D4X%@!4Cs0dz^fpwN5WZLka7|EQqeEpT3s2Cs?HT870USpKjkEsw1!m7u zs#g__c8(#*r*vB^{rQY!s@^wvs}f|Mb%S^3lWYll_o;uFKTz>szqz&3%T>HKh4B?F zMYUYm?;727+B7E{EJJB0?HEEw*gStob z9hnE^&%aftW1Qq0n-LR`W+t|0a?OJXHVzF^mN;X66Gp2sA2rlGgicbIsN%g@FGW!Q zO40zIkC(LNO>mw0=kCmZah@i&gur=Wr0z|(s*{LeC45e#h3wF2!KXM?DWOBqW0@|13NHI>}Wg%&E^AXqW# zww!E|L?vyxgT(pP^0KL%8d_Hnkh;1JJW zUJyhRHjf}qac@EYc?{zRR@ugCjW8c(yrAoTLEp1`H%-1w=r3|wl%OX3HaxJcceP3*br5FRdJ{|3%Dkzn2% z_ISPDU6IJC-rf7#shw`KDi{6BWb2Ec;yHaEblrScCo+};-3<=Z9F2U>dCM}x+psJC zQu{e;^)xqd%@u;2u@BxD``VqHe4&+ob)VY?$hctl#nQhZR@%ch{mkTP9(dc6|6{gC z-M{UvF=x9R%tYSd%GNStWf=+&!wrBM*mQ7MLwJbvw#>w(dU0f8;4A%aV;93i*EJRr z6^}!nsYDjJL-M`36A-qM6j`OS0c$iV+(11X!WRCvxgp=7#CSI^zF#hd2hf_}PKs;) z@%}UMW%Celmz<4`H~IOD{X9sGtC817hu;sNH{w4sL;Cz=Hr%^RVQW&R?C~;PnRKbn zRLsBVSy>?dZdlr9q9W_r_+$5tBu~D zv2R>g6D^v|POXb>x;AnhwNyw9s0HXjm zWELdF0k;hmqT13SxzK{Vl0Y44YDaSXt0@m-ff;hznga3+eV57>QWRxq;>?F8w=7#o z+tyF#16l#I+!VUto}!R7qcKuI?Kh;wJyQBJ*%{=sg+tl)RLFjvfLWDA^aD^)vLTJM z>QZRr$5UKpimIhcnMwD$#|jR94d#In9E{grA8+eTS4RG?F|iKMQQ~$TJ_N_Bm?`1D z@*qa?V_%L)Kp&~*mVH0;c&zZT{o^oQ3yL+-?58zC#F*HX#|L)PoDboA=BgP|58g%B z17rK`I}`@_4*fb<;q~JdoY36f$yhpPi~2wkX6!b_8euyxQPE<%(=6((?{rzY?Igx? zrmr=K&lSh9_QqOA6KW)v!FWxn;6UUk)#lD&fu{#>L0n&KEt2U^yyNvUP}0y#HKB@FxHC%W zu%KujTs8qCkcMNXLX5@$v5BWY+1lV6X*^4FDbKV!B8~fKiN{X4;Vx1LqJt4sp%5Xy z^#`x_!Q#PQ)c2J-$`D0&xW@|$P}ff*408@L(+vHH4fPK=24H0_EmA{31;L}JU~ChL zRzPZ12f}$Zb^ZW-)$OAs!aE{l&x3_dVAiI~ON+BjD(q`t3!nPs{7i)aB)7imfO2C z?FCDlm`u6xhpEGzs~BddYZ>&-U-NFvox-FU@v$)7fW@Fj433XZ2Kz&Y42CQt&El=*8b_sZ(TA1`N zE39)s!FRb?yHQWm)Z9!z)m(yjrpD%t1|9Qyw*YgcBvMhFAHe8%-)tbAV%QIsdHr+> zKd>eRU&PdZ$}CZYkH2fnrP_BEU@SPZQmQh^&P7L(XhKz1lo870L`<_3QY8;0rdje2 zTa>m?-YZEd$Tu%9U2BpaC$JK7+~ff%eS1cESTECX<%5rY{aT&9&sH;iKt+tj+H%j` zYO^k;50$pJxVTqR7*jj^eC=lh=m94QKLF$|fTzUnx^kh3Vx8^9dQL|ag!Odzp5Ckq zKr3|8es7Wac^p0Kff6D%pU6qY*pPKyr9@%tE+fQZ)N3fa6RvuBPOT?}LmsyjbAulZ zz2u3lnlmUG$d*NEL{M%_wL_&NaSd}FaTp1YxE2=qx8eE(2owg<5Am2i@iZ{M6m24E zUT|)_;hja-(s0?Yy2pq>oR)vn_cGIpYURMWenUK~W=h5p;z4eKEJW%gRlBRwQ?sC& zdd{xb2y$cg6-H79w@)@rD1@?MH4`s}x^?r23ESpxJ}TyD6@(HB;i68N>Dlr9+L4pP z58#Zr^^F3NfhxqYG`=g+58sCvy~a)NON*K{-kij3xU6<(6MLw=Zk+WTgGy^bg7qUG z**s_+7YUx}+L$ZcC7-%}V>V*8H@#Mq){Wd}skC)M(BxC^*X7YFGslGNkqiGg)2#ks zP@dsu#?D(gYI{LV>ZuDA@~Sx;@&W54hlj~xIR*Ox_3Izz?q`FO$%#{?13tztev>yL zG^c&wVRVjL@FO1wqp&1sQnHq?JXk#)3p zrqW6IV`GRmkk)*9wxJYDM{uY^GK<9`pBM zC(n)P(C>I5+^}(7_zMtOK@}`5qUA_<7DiCwN_E(rQ zEbpB|(O{LYJ6P4-vXjn+NGw_Z9^y9zhA&3Vahk}{Ubd9n=z}H%VIkCm=d51-_D~^` z=BdwfYn0T7J)5XX_12M@Ziw25uz8RcTg4JN+^cA6Dj4|`3g^mbvf>z?iDnWc3)w9Sa))GB&xA|c(t<4 zcR1%k@Xzmk;RLu`00jSXAMlgyMoIx}RklUigSIC=e7Lqh)a@?USs? zLNEL?b_~1N@9@eP@3ynSa~|=C;S!!}vXQ%L=sZ5DWur5}{&uBJiIM2jT6p_$YwA(a z>%wTLvM+iPzP@8OZ&p;+pu5A9K$LgxeaWtN5(ui@Xw zqBj=0yLEMQ3JFe=)HG5_`}-CfFgx_#dhba~P>SIU(ugxbJNuhSkziv8I*9qms$YZH z{&I)HGUX&u{0tt?`o@PRf%EeVy9DpUQhh`3eL;IwRgM^aLmYOLn>ab^7a4v&_Lueu)%9$>fyBVMEl*sRg`2mtlrasp`q z5?fAr(bs)!^C^q`1Pe$reYhFxMIb`C-&&4D?k&k5L%6hoQv7c_8P?nThZkpGyGU1x z;oydamqi?nn3>iDJq*xMU640^>Q9%~ane;p>HgP?PXCOViLk-{Eo=t<7B*<~!YiTm z(bqXJYB0654`QG|Ekwv_oUv>u<4UGN_JKsXLAj~EIa%71vsyXVI&ZUAyy5#ZK;(Uh z@Q#R)*^yY;>JRO}-o-BmyNUf!AXtmzkI|CE zR~TghQr^>o9|xUL*mk#fXO>4xZN~Q{-@jdsYzIh5m;~#3iKnZhSpy5x0WF49$2lhQ zT2cLgwgA4ulQ|F9(Y?v~!|mCTF`S#ipIOl)Bm@a@9%7ow~vp!P~K{j#7J}XDJ+su;~CZ zK+L~A@e6EDGS9_D?A!Ci3!HIEN|3evQx~p}6=;!64xvHH9Btuu)Eu|wKt>#T_9-lx zyaV#bjF#s`&22O3lZ>Ug_|>dTdcp&o>imqXh&~nL(_*%m?W7n(ynl_k2MQ)$rTA_@ z1cBCY4TOE{H6KJllFOD(;js&#Hu{t=khVfn`n-bt_;YEarJ4MVqn83gE(!()oR?6ad5`Y=+JvD;KnnanAcg)%QrmZ=#I_P61qKURLw3lg z&i2&sy-Wh*iQm?W50lX&Sy8*@^%_py)`pa}vo?maFW$<>Uy zHtFwEJ{rQBrzO<{DZ-H<;GH87x9=kib92Z7qiSQ~2`;_(u$eOlg5bg&$;8xu{L z`Ij`F^N{haYPi{p5|7z2QGlGU)I=obGRxE^xUNb9WWBPIeJEdlm0G*vWz|4x zuTsm@BYxE3z!vOCuiI4_7hT--8#^MF_gcrBWH&o##*)E{X`-vbU(CaP_8NhJCgHEW z{4dN;xBW8Bht$i{RteT7Zp*Z%%blyz1`f&^qtEC`J_*WFp{y=|#ho>LB$Zk7^E4k= z7!|u_${8*|jq+(63j|hmhk+i4(Y#=*8iFewwy+wd1DRoIo)?i=aOB7Boa&NIF zku&xT_b_(u#>W(t0s)v5H7bMIl8pSH)>k-Li%=8ZN3R%!tRc)E+(j|W0!ghI(O$@} zGF_>hy36_Tm{kY<{^HljX=4Bf`H{_a!~~&0AixB$5x&YJ0Dt%gpWcyCT%HDe6yAY-)d#nG2r$#qvl>G z9{-G`cVTSPO@PSbOnbMMo)sx%aBRukT^@w8XM-D}#LJBf7(H>-7k0X`nVod9{QTDY zj}8ss)YFNW2Jhvm>pWM_Spmh@ntERDjIuTa;~fuQjx1ez$aBW7P;7ip1Q4oYZ25t~BY*|CZ_2SaoI; zMy0DN)oJ~%nV7c;06Fy;^6VUss8e{Ho6Yl&dqWD-&d*IsEP--DiW1Y(GLsd&O7oISGV}8$FK5!<{EI1yF|-C* zt7(O)Zka(1%y2NfD8IPazbv%~#Nnz@u+jwrD}@>e4W_Js;)>D8=1lHjl@ZWO&rH(< zO2AB;ypct;9xezoHddp!B(WqjSsy5&4;O_f4cJq7a< zL>S#EDCP%Dp2o_D)nWN0J1ihSFWsdyH3wpC4a8!o7^gJSN79$@7`~cytw_D`FLb z%QH)oGd90rY_?OBCR`J&`Ic?lwr$&Xmu=g&ZQHi(s;(~Ew&(ljoHbW-zcVt|lPe;3 zlj)!E1~>|xhB0Z8jYjNIvPiPb$&d3)%MZqPx*B`{a5eVuCAP$WGSht7<|2?Xd>XOB zK0VIjR!(kKN{#;^@#z89Bo*MGY6Klbn3$Jwr;?iujB;4jJioZmE=e z8?{1BCrvXUzXaL%uyD#EJbr&xv}RO0C7BFZA?^f-FlFVK!Tr?$604+&BzqSV>{@CW zLY-uRgld+4gmw;*b?`Kkatb5bN*A(o~pu{JMzPF;wrT z){pVbdFy+<+Ao%HQNR&hL?b^cKR6>vPs5Y8GP|`f^;CSjd%mWPR^Jlj_#**U^{7_P zfpk4`qT}-R=$lF(G~d1RoT2{NN2NQH)!MPFEm?yDQb}QuZb9Rl|9IqFo}TR`Ow7Hv* znyBv93_j%`0SuzB6VDNp{*&9w800cyZ#UPMvQqlvBxR;`F%tSh{C~my2Qak^t*pxw z{DNB-MIcJ{k4fg%`~-Vu_PwyLkdm2xWLOuCWU;W6conduyPtleu%pl)rz&A7KQC&< z7D*x|2C>2d%GM4|BD5M+kwLa?X(I9Mg#~b0LyY1&#k|DKoWg|ctc-N+8pWiH+$zJ0 zxl#bc1+?Y6VDM7l1sd!iI5zbiaYzj7Dusbj5Bob|^2yP z5Y<9uoFRmErS+J6+aG=iKx~cjOab!$x32#TL^~-vKg}R*8j;*>h~(e);q6icyctQ6 zMf9;YPmUw33f9Tp_11N9T*lZVZP_7k{~*`>GuQ(67N}dd5$`cY(_lS9w*jWiU zi^R;xNMR)_BTF+SD?2NFsYZP%F{t4k7E>;qe+;e!rzSOeHu{@c{g)$Z6~$31GSVa z0w^h^=;*1;5GS^7jSJf<-8poA*waqalao;0Z-4>yNzPz5pn*6ZIY9D&#@_X$^e{h?a6Xn<+-9#Bi)IX&};~l%*;h;Wli_OIUq?~A!uM*?+P@@A~ z3qsITMrmn_z}hT69NSf2){}q;rDaGWE5SW-{xA24`1zS>+Q|6z|J=ceI!FBHgWhmx z6$Rf);K%QbqgS)>?+5@ARrzr$3af=>J!{JE$Ibr5WVoYw6y^`BS_*@3$#gU_4774| z6c#;{Qy&z1?uk2D9U;Zs%bX4cl`9;;kB%zaq@Z5YBd7W$e-8lwDhpIq4EF!qzk||Q z!RbOLkDxw_{83+4|7U)6y?zO?sHT9DjjWNhpAvtf0<>~_k|@n{-O|ro(cxW!eZy+* z0&*Q5z_d;XUI+A_?EenO(+n#(!Y=rE*zfx+tnK%7(5hedQ{B4~3?TIl=DPs$|8!?z zQgWD+l$(>Mqh9+jzKt6UKlJdvCE73A_0BeJ=YzGU1;SUcg;&vzF)%*UOVI<-KA>ui z6Q&aTKA+|}ero&w$zgyOx6}YzXEwN4V9(jKeGWX7R+8*;>fg+Y zMF4=E3E}hxh}9#9XZ(LPsgV6@x!p@q0XKK%&WEqP`~t2YJyijbesF+6L%5Y2VCw(9 zd`op4;&K%M0bxNw3Eg{`NjicR>HQzb@*_5#sIqASf1iL!$EMHC1eog$AV7*Taz%dw znnzAax4Wjwb{k(0RIijo6fxOSHe?hba!DxhlJIRWXsJfvLbmi7VXp|lvJe_Mg}j20 zO%y__y>+@6l!61*+TfjP&=>*fcuC}U-*mwK`!VnnO0EAxO5D)+i zzy%}#0>A;ZfW03dXmAikMgRneBx3+Lh$dse=c5`UYJe~hC#s+@5G(3{Fwj>1Zty+; z6@FkJfDB)-4?u_C`}v~6gc3{y;6f2507zJ{`4HyBkzzrA$>e%Ig1_5ED1O$UHk^)3P7-<2+ zK#TxbkYFQ4Mp*bTqCrMh%rLNF2SZFiXkZ~DB`SFMP=X>RHuSKt5ho>V!0>i-CMHbqA0r1^*vQbrMn+onu;C#9E^JJ&kRu}}eAvj~0w*Rq#PI(AsDV&Y1_Xgn zk_HHaP|^kr0Z{^CfP{oLXMQ2@S!7x3!JFv;Y0h1 zfB*;(56XaW5D(G-2@nt3fN>xXKnz6iy&@w%{20lg2P-jEbOYGgRkgA_Sl{Gd^i7d{eV z!GjSSLF{nff)_bfqWE5u6+e6oM3WaYI{eSQ11)~!XmO(lEqeU$0T3>J%&>@~2Pb^| z$o?WHFM9a+{++IeG@Yq0pBR{}DU_0i_j8hWLZF~mT@MldW4oCjgFO(+cX$<_~LGWHxg>D_h4Qw3J z4~PtGU@|zRj4q9 z3yH?`{*rRC&n!foH%f`h*m9*g=D~!NpwrRzVUW|XJh9BZ#e^on;0*~sF zhiB{Tm>1ypRL%dh+89lMHP(JK?cbefB;O|I{zs_AU)+4(-(LPRfZFDPa7@eDs8@k#_u{9AN-Bw9~SfclWhzj<42H-xKoU!L0GF&~*RszLx^q z82Y1!$o9rqv?cM>+n5VGsies1Q`%-7C0JQv_%S8bB{)5EJ(t!OOm+#VysKU)ZcA}` z<;GG%dbw6O@PRPh8DZ$^W{y)b4*a{z_$Y>%oNtT3Wd=tlQS7D2qM3hrXjo~~1if;W z*Z89NyKiYjV|u_e*sLh5q{M`0uj8RHPnK5#>0&fdO?A?mx6-L&YEk=zt%KQCZE~ml zXK1N-4=9duB%2LJ{9*R6D$C$;d_$xE=%GIWL(#o9vepir^&Y_YF)38malvr5?EIJd z>1mjE!^axv``+nKez;YmwU$Vm;m3OOC)+u@w+MYE#e6Q;1I9yHD0DIV_gnY6?gT+h#OuLj4b+~j%6?%ytmhi?kXK=+7c(;^rK&>Np`WTvr9MS`=pM~kS*an>7ALl{wZ@9;Y${T zDaK$ukw1mtkFpkFuVsVBi?@=;vQDXJ?C`3Plj~E5KzJ9}AQ`&*NZD9&(@CIrtR0<5 zyAmzD`0!XeJVKBhdah9p=N(2gd9D?$j>i z$xjeMOqzy5p8~(VQxA?&7zt@F%wPm(h}+*QN3x0ugc(cQ_cn0$gCkxvIcENu4Fjax zJr(D#%-(O2re~FB3I8hfrBa&q+Yl^kh!WHJdIxTo1A!DvsIM;N>j4Vq-$9X8=1~Lz zOEnU23b&%Lm)PCo5lPU+=Wyw{+2ArP`?}f3oitzW50Lr;F)GS9Lh~jsk@vuTwTyT9 zt#3k`PLoNzEbrmDTp*VC#mlRrQ!pW3hr_L~zA`>8qU7?)d@=lMevao8xcz7Y?7!Jb zns^?D;uRql)n_$x)q8l3wrL&xZMbI2S8t z9^$52Dkaw`pUKS6i)(^$o*h$NWY=}eyIREa%eJ$!w~dT%JS64PPq_HCQY(XsLL)Z% z^%)ANbXC^e3e(!S8~8VDj+Q32JJr*&7%b0MY}RB8dC<}wf09ijimwBMN77fPDE@W& zJ)6Fs3U%p*xL#cm(oC0BjWXCA*)C+FN?{(S<43~repbwTF+0cYq;`?#nG<)QvdD?F-Wun5ZniUch_d|&T&IY zl(PnKBun2>dJg@)!Im56HvjLVAvFI!{_O)b9Xh%4<6#Hw@- z5V8~q=Kc3CmZ_AwzKM4Wb$9}Eg+gH=hIX8V-9esJ2X~FvD~#BZ5;B{|Q|2-2D%O7dKb=UBK}40Xj8tuISQJQ6)c zON2KIN;~Kf#K8O~1M+(l^!#Re=O#Oc^(UJunp^}j?H(68jzg9c)!Yc#&EL9aG~ zZAKk!T0pc*aN8rt<+@JGfbSXT+I5C)4Pj23<$gtaziedUeZ?{{rW>>`JXukNePPsS ztdkNl6B!Yxhb-`&pQno|h_^xMq_pV#eG(U80|_1G&nzH(6_9?HjmJ|I8QCl6KgkqPdm0+Er;@ip2uG@Q3TXI*Nl@s3yF===T^q8|HeB;uVe^?Khpc2f5kB#g$cq@A{~u zS4pc%ht0A+PM>5k}Narf$VO zd?Vxzj*{_+=u>`Sn8_=XXy?BH z--`7)Dt|R0d6)7i-jx`qz%JOc`P}3Cz6|ag{EGXT#EKASsMAr&yDwpG$ss!7MWV5W zX8z{T=>>69<9m_1-N7e&&%2k@|EB{)<9CLk?Io*u14$n3N~Ij*BXsl_h$w`KBUT?B zFbXKrjBKGA`QR_26aOY^HZbZ;^TpD<^jpEO6IkoA=#^rpd8IIkF4^`A$~(TO*SW&5 z#r;tI0{iBZc>mHkKT4tKAcYUU#6kVK^KNWfe@+Cs`V&4i@J(pIM{f*@kxj9i{-=&! zu?JlyXSb<*_V_Hx7^51EhKBs)BH3e{OfwqIw4*ms2(@Md68zq*?I%T1AYq|3Z$Xy= z&Hp?^K?_+X1?4eSY=&>9T_G=&e#t~fp#R`GhZCSh zWjgcDE{ut~R%OyUmLhsN(@-uuE^Ft=6(^qZ!i6ugx*Dvj;X&J#H@ zkkp^&cbX9RiT4c$#J%WMtph|Q?E$=I9!l%JseIFGyg10%8JoZxRMJHkq#C1=dZ^5n zqrz4$<->$J`>*F-Y7GI(26okO3^T1mO|@J7#OvOc9xw5reZyap-80$kAk$I~S% zNiZErL6FfSsmyl93TGgy0iK#(7QwL$)^t{MODY<10(N^ZJ{#3@yWSc6&6i~XYV2S8 z_FYF7(kztcP-q;cTc%E@tG1X*N=yaBrrSE$)?9pr>C=tvr{W> zu5@e|oP3Lkv$0a{(b*Ao!Wd| z9QPdj67w6m@CkEDx3g4MKcntrE#Zj_l^Zpl&_IaS%;^nR99PPq!<)E_BzIbqqb>Mp)r;{25QAIZa0q7mW= z**6r{++eFb)zTqoaB5`jX$4o|tp#^ULw>Da?qf!u1w{jH1~BPa^DOc#WusKY5M4lbzshVEVNO=ybA&j!6)ejtsXfDlN9w7Z$jRAG^rh3fCV^?z2wz~S0^lPooEETvqfrfoHw&uYWqD&=1tPmihzpyvd7M{Ebt=^Y6r7&aayC)FhZ8wu}g zP<iZ>}dcfeLW?nV;8 zJo`)UOnx)RrB4J#v1u2u`8{=&eoyulc3!}?%g(3Q4N6vs^|JQZnV)UwMns6Di!Vfg z>NO7xT5s2d-a4x`z_NM>j#bjWW~zRb4E)hIm=%|r(qzBjkL>u5izz|~|G+qaxMsyW zR=p5k`tv3A+OkHPc-ra$Ko}XVfk|ZY*W--p+e0$mC-o8T`r8csUIr^@0^c&9LsB>E z0-^NQ-`lIl`(gVh|FFv)E0h1jhwY7t8il0GhL<3V(^dEatrmFX$ZomKBL>g1Krx7> z=)*5OD)4Ha{teUbzyZ6=)08rq?0i2g=~nrITTg92aNk9j%xvcMqzNFw%Q+MET*Fvc z{DC?j5e1yNOHu%*0fGgHax<==Tj~mUf?|XelVo*z2IB=DmVg<%ci#keI5OY z%ZXoAZo-n>slL3lZ@IWgAXqYlE0w_#Oag8htnkkR6eh#s`PR7Ep!qf##)a zZEPhnbzML5?x~B#)87Zn81nVCd*PnYGg|Qc3wc7ON*yXPZ5@A^GpZLKoccn$PE6~~ zWv)-@L5ax1%Mv#?(*&o?5?^bd+xFA8$q_=fU;9w`5S{e~B4ujemUBWKc`aVon7}MC zSTrVBJL6WxBIK9)xV>m(Bl7hbG5R|nu~s~Qm!>*P8F!68#V&nbUi`wgX`uPb$Jh9c zR8qDIas6idc4b(O?pHlEcxu)r+Uix9hl4?0&ZByr43?Liv|WGq3QtX7l!}_nW;4_V z7?XMt;<7&o4UQDMBvw@suuSV70im*@C zPlkYgO}Ny?qBXOUpxl{T_i`?&K|Fgi89cYlr70(`;K}Rhxka+BDm_-&jfZ=gVVGGL;>G z0t$!{QdOIsrfaiTgJVzz3jO#LRYPQ!C+vKYCKID(@!)9HFyBjR?<+cPk^5~m+oH$i zE0fn+X0;>IQ4yLR#eP03hB6gfwjt!C?yPz+?C`CA&pxA33|FKZ`QAdHQqUKc$`>jb z&fY;o*B{zZLh9pE|2RP^?s)06P_S)}4o7U;OJaVvV~}Y~lGsICx_su-c+cVmAHZ5Q z?i<;_Oa8O_S;adG+3k0od7Zj%1~DZ6{Ic840e4;MaAVKp)3ZUb!b$9YjrC771e{;Z zfs&fo)Zwylb6@oCnYmJmW&D*!Wvq|wLzEp0;bmm*T}~VtPBQWbG2o5^$OHh0c(0au z<^axsTcO*A5NuJcymbAQ4|t=of-}e?Lt#VD=(VQAn?g(%;Zw7w{Y>yBd4p2=O8~6t zSMsr3N<8eGiBOa)fM$hOeF`nDC@Z>E>SkEQBPy4Q<`q3@n^Qy;IU#h>>M%SEpBuT%8J0EP@g5x8<- z_}8eAJ|)OWVqYNT;8plr33_Ezc;i)%q{VJt$y)ciu=iQ;*hOqph$z4Ot=c3f2(1Va z8>H*w!0TAEv+XZ_@n?F_2&2eBbZC+jw*Q!?QHv@cYAI(n-8)euG&reJS20TOOk0B9P#pR1Bn&0bJj>k3pT=%?O{0wcE$n9b4 z8`>Lr->@3md!eDLV(CXh9a$`AJ67si=ao% zWf?S1qyfiPv@}4`U=##Uh2-T~-7a~+_txb>jE-I{VhPFwlb@u1iqi30ujG^(MR(&C z+9gB2fx4XFI4~!=Ry>VT#yfqy3~qUcs%bUn4eD8cTz@Zp{{B7h6X$M$^yZt%7q;YI zyj}{dDuYJxtkcCs0!rbr$@%eGqQ4ycquCJi8uF~H(kS;dmkUzYLaILoe5=iMwGOKt zpDCSuw;deLP{u*s$7?XV4UaoHf7{cmJ1EGMr!s4C!$O2Cxj(Nl?Ks7z92f0Pq&n|q zHFRLGO{kfRJz!0t`T5}!*<@s*-&=B=32LfbykCydY$5ZUG_|&^Csmf_8uuen#b0`K z?ABFu%ws1(u#rw=veSbI8{rY%nw{`a$&>I3lye~{0Zy5G7PCz9mv54ZI>q6M0_~zY z78LGV`A4Uf8ov}v5O4R|F@yaFnikGUMM3&A_({9+5YecT+9_kCc>c2Rz_dMBL9s6@ zSZQ{16T)f#5dwY6(8N70-4?0og~JrJ_z|Ra`o3GpbqNkfu4B)b z55Vx31;j#X-n7*I3kT!~Hl&oyk7aMnvnjD|Rk5n$+7Mo~{!A#XMMJXnpWpeo&&2hk zhL8Fdw^0>R&`AZsyK$i&*UZa>Tu|>Z0zvjnI}0>cPs>!(Ij@-$yVsns^4N+MAk%h@ z79MruZ++hUGqRU{Ztuf`bCAY+C+=nT`0%PwiBRGdc+g3&q-hjy^CldP2z1KT;Vs1) zxU{7I#vQzgok++ui@sO2X8}Qp2Dn|DL% zzVl&Y6cy4#g0~RjkA6O+hjYfD;FI4mqf*|<){7D8urISU4K(tB1DN30q%;93inZyy z_84Z_{<9$;Tb}-F2$K-UEF`+|U{~Hp%z}6n3A{@|kZcO$dE-*2$;U1ZQC@MMA8)?H=+7^sYINtrB?~b;Wv>C7=X}o8m9!Vt zNDRL<_2HMOJ*Si?tMy1$u{fDpUMuaiI4)3$;3$^Mt?X#lR$`u-P2$VI!eWDmBDZM= z9Hu$uYO-aN@j!@b&gZiUZ)9NLn=)RIjK&rcXvq+$LGBkQN9G6<0~PD}vT2sbO?RmT z*=mnD4iNPP8X{ULtf$S;C~>_Ut}e)ySDTf^$Tr(nULO2I(V7q}A7q{P%lhh)9=w%R z*sF;+HE_RyetJG_5w3@xx3#HKsVn(T>Rq&4OVA_q{bUC&sH}uIjs%*Tt^z$8jJNLi zKHta$=?)xY3!P;FY&-$11I~O;B!@EJ*tKtpit>b+gbjzuwhx=XRG8X+mFcErkaFkZ z;3du$%g6pJT_oat`g@aYc;(E-!Z0#$e5P3$*o zhE{BO<*xRJ&{-L8F4^i1`2@(>Z9n-^i1Yv)PU&tbVWGYX6 zF^}>O(%eH|zzBO!F!!AVXyvwA6XXYF*8uWKbJyun`?=+>{wrr7so;r#rv1}lvoP$c z03P$Z8paqmLSEmlD}qFt>KA?HY0=HLB_)Px6F!9q5I^gk=K3UljN7)f6=81w2zpSR*}!MQ zxy56ADD$TEMJk|dn?oY(fIMn2f_WYT zd{=5ua@R@Bw36{QBUsFcx+fgLV@>s0+kcT$EPzP8DejClDp;@t4I z(^~KIULH;7H8ktWvw|#Mo-HuFar0NNO}MDK8~>6m1#N+S5;2Jp(I`%Cx*VH0z6-hW zCiSdAbIo>@xDca>qvKy$?DkGUG(Bp zM)8&U)`l!bel;uKs*093cx5ai!*G^q7m z5^nG^Uk-3F_Ry?8V4#^@#M(oPK~Xf!&3t{^1lI&kA_J2qIlI_jEP2zJpB9XIi1QaH2qVslf7i^f4T~*>8rt;gVPCQKzLZ`lZw$Z-W z6t$)ZSBAvsB@D6yts_o5ZlIvQqcNKRgaz=e9|W_vc!w`K9_CQ(-}tz@cGtd&aQ~X- zmi4mju+LD}vSi?Iwmmtdo`yNFccUCp_L@c+^@b=2|0d&9qc6Usv&%+Enr}vlsix*Z z9U0?&^;Pnd=2?uj4S#{#D8xVIII$J}OibR^lBImi`#I{6EL7qnt{SEsB+bI4L9ug| zkzzndd=*HpmNL)FY@bM}M9{uSjsa-_0MR2^IZzw9){fpkL;hhI5KH!rrR^c1$}H75Mkelex^jw zRBO)OWr`TVT0>4tQ+L}5EJ@)V*JD6DbQCAU3pZSh>BMv2^P9(8gNWFl*w*xQ6xSr7 z3yAx0$2=+Y*`{AQQMvL9h`J0ZMNjUna3A4zG!z>JU?^ax;%-`fXy4@`At{bKApk{s z?Cbh`B;UdbO#&+&A=v#pu->fi?(W~ z12G6dGZr3k_ZLM@n2FVXb2oiH?w&YepAT8~CnORH0V1U2a73@fA0$g75bPiqh?a-^ z{&wYCAH|8m(60-ZP4;6x_h3zXsul?Gpum{UBh=~d0t_0yd2$N<-82qiu2vjjc}Sh^ zl*HF^%GgclfO;Jhx5fjO4vX;A~4pj=;JUMCCp4n4)?8=+;7?0Y)8tcbGJ%(8z z1+@Qwux9>}5C0;~6w~wdFj97735GRX7hrAp;Uju2TQ?gY(~w819xpz51c;tXER@P< zjBM9Vutm2JK%R7=y2PtuQdwvIin^?;>QM0pbxIBvj=DF2UYnzbejG#DHXu(i8cXK$Bj;{`U z?=0NmY6=sHQf_prz(@^*=25)! ze37}{#FO8S7uzmrBFKy@>H0!<|6Z`l28M*NqXYy%ww*KjJrqWlcdtiXqsrgBLt(A~ zx$*XyBswMniRZ}z>~J_*SlEi`Te;C6m5(qs%+{=n>`=Ij`KzX{gPInX&ri^hS%Gr zKAzV5h0%-bmb)_enCX3`^*gd^?$evtm&Dwu)Gkb$d+7_;KS)abdUPQnv{KQnLnS=V zN;wNzA6`S53KY^+i8<*tDg-X0bs4U}U0jicZ(Av*#d>%pQ`}cd3yOS`W|-@dh@|18 z-Qoqnt0ff#VsL39_gB$mr{G8lVp(togg ze=%av7DSyZe?>Bo-)SM7HOVOXO`5xv_|ODX5KbSw6-!1poW(JO>vMPcR-3(0h!To! z(D!14LU+kGOmOe2yfh5o(hiQuo3NKGWIPq3uKb=o*eoJ}D;;$mT1OM?9Q7$P6^Q1a z_hW_@OasYexH%pO#3h1!#D-LiXrw~p8Dg@}*Q*~G zs@hjKPp13pY!x{8^bZ}7c~(I0_DkL_@s_%$?UN*35}k@PN7?spOj4$qyg zQuw6h)f7jj1?JE1G>MU6(6Vn}clc0ieV#MQs-GsIy=ptyRl5s+G4LnHp(qL*bRh@T z{CW@8y}Rm()@@^>;a zKbccIo`%FMh|zmvVN>ozF*=M-m_0e)Rp~c{N;AIuN(>|T(Y!4sbI8Z;7S5d8X3wEC@?Av+u>(m_Ms!)m0ZvU z>ebcVV7_P{pw!=~D?mN$)IE*P7ItsE<=QZ};w5?Z>$}S7Rce#24zd9;mQaqY@@HH4 zmNllvF0%eCwRr{{_JSdfz)oSBLW=(37}sFHi7d8Yq2NW(?J3DA2EWQ08#bInWa-9P z=Xp*c!7EvK!P)|)AZ-d+UdupUz3_>5j{`4RePCwdedtjvUK~es5BE0R@=j9oLW+8E z61$0p@gYGD2g%x|bMveE6Y14>pVH?tj)L)^t^V`1W$8NA5)X`ZO_}HhA&w zpJ;amu`J^FD2qY@;feQ2WZ{aCLA9z<{7KlJOG*|x5)GUQu~SGpnhY`M4PlLc4FL({ z6l1@NxdA7{g-9*5FUt}EH;5vgOrlh6Pjq&SBcUMQgR>6C&or@|De1aWYcaflWZ|m# zeXjr#iJ;`}z(puvY)OEi$NNq|+`FgQm8)Y!-@PdCd8!^`;EoK$VZ}5(U#M5umR2Lf z6Hjh{>MXuF-=*{1uKJRtm$0Y-fr#kc-X6_i@fy!mTf<@J5!{A@9ZBEcAdA6S=ta)f z*{LYz6>{k>MFc8vD>8i4kRLvdZ*lPhf5KoajW>VfM^$iK#NnPVi#64#ha+7Dx=WCLs+guR} zb@h5;OL|`cBdtoTj@fZ{8usO}M1UVjN&L**6jx17ho}r`Clqjsj=N(5kzJ=FG9psq zHQStZ4=@7_ep{sw;h2Ujwv$w}F zq)b6(J4ZLh+-wrgwjZ-)wj*|0&%}EUx%Jf#33}~? zK!#NqOB4~@*7-)Ec&u6BOVQef5)pRlEReYEBu`?Ass96p*?h|UND`uk@{uZbh%=U) zvrb}{aG}brL$t!iO7mJcucFt=dF05((btcd?1dx&?E&T4sWQlQsB-R|7vs4-7I@lDaiKd846R zKkXy6Jf4!D6rfx7>L^s(39@NY8%t$qZWAVhlYg@Dlwk?A5&>V;yPP?yq9y(WTW_Zc z^Gma9V|RFs92iS4-V){>Lg*6$=I{Gs!{EzQ@Ob;24Uhzj`GIpPsHOR6{1qS?*G$1I z%)09rme&QkN0{J>Eu;?`eGyzJ7BJqansZxJ>;mDpWr`+!J#gqSoo&wXNMDf)sp>?Vmz%AvvJ;wXnV#4?tAHU6FsDn-JC!mbR>C$~IhsQs-V+D0z_AO z6>;B%Zzw^IiKDw|(|yheAqe01qo8b|wZRB5Gi%s2&zFB5?K|U<`EXpD$ax3 zO6XAgGYjl=haD4ZT!5+;u0mX_d@3|h;GTCh+2}dbo=Jfv9LCm9E{zZ0koKOI$-N<( zoG^)92wSXPz8pnabC!42*z6pQrA2|}J_62IFE*Lo!YH2PW90&|U|7pCp(c%f7DJI} z+}ix~hi^_?=K45FN|h4_i=10q`cABo{65;XqUX+v{4R~VxIod2->Em`ttYlO=$I#K zTakDpL?kg`4?Gt-1kyegl<{Q>vJ#oTP*;@)43@Il&XJ1@cwF^kHNo$rPzlq>vfKJ; zT&KTn;SjuTlR-dB^*a@l`kKms+!y!gcc-gpCQqYtF0Hfjv+sS=J>#R>rmm~aa(PrE z+*2)n+yCo}%&ea!D>sUqcAau4A^Rn=PdcRt*`K7K&vZ?W=ERF6;KC6?X4md1E&O{` zR4!8gz-IPp?NUt@-1l%+$NN?)Ceuw|qTf5d7Uos-vL_@43g_9kn@tIN^US&RC7j4^2@c&~0;4rn@a~$z+iv{6JLtbm1k0;EO#1lSPC;jWRBpKH`5yK^ zpK)E=LFdLS;$VgE;mhSTR5t!F9|ssMv)>rDl@>%`5e~?I1}cC%T?h(Jx?Wp?-1cmN z=^BIkl0Vb?Ci){e?VqRQe-sgSE@n~I#+nLkEO|S-0cJup!eJRlHBI~R_6~vOtojje z+n9s9#PHD_5+S9|MWR^hTlKwnaP4&6OAP*r}hBY*GRkM1S9%_o-OJc()d!mm#)HSXTg;a zp_*5O3UaF6_6=eTcsSNGqk=f<&A_#UW{_EKCMRosWGrB|ekkX)5kIUxXqw?q!}uls zB6_(J^X`<_2}jHw>A_SB_8%M*$9O@JK0Es*zV9&!d2Cv zvAKn-+OFk(ocSgO;x^}sIsOD@N!UjqfGfGi<8X?(pAAn>NjW5u>@@LqhpL@L&P+}Y zMlXS$#-k)BZ69;8C)4PVcyOj(!C}%9X)w2Y>=v~P{}i>LyHN?A%FiT;%8B5}BotXo ztI_u^2yXt!>sKmmI)pl+nNTWl3$I4+^FkLFxYk2#7H5Z%wb)J&*r|vtEK5HA zPxj1=-gZ{Gf7mL3L+qfxje5GHOFgZx;^XJDX6{u(;3LG18=hRxom)Y@+D-p3uJ76f zsANchQ8ayDplkB2%D<)#`+CY1ONjuIZz|e$aDfpRq z^|_z}U-&W`A1Z-Dinz;pMz3omBXJIu4oZ=6M%O{=N$cux%WEHC?0sSIWU5Dg8P3Y1 zF0}pn;fx@~7Hk6MAaUeP65-QKb;LfJCT+yQ64Kc+UO%_|X}M|a>B#~2ltHYi)-Gn+ zB8E66VxNU!gDV!0!|{o4-c6jAPpymal>CCVQ%BFgxW&0$aL)QtEKZ8c?$~eM*}6}E z>T7@w{T~@5L=EMBe+Os(^|cV(Fk`Ak(g8_f;H7Rd)|y*e{1F#L<~r> zfJZ_xG@dEHzwsswo<2(H;o={?Cfc>3p}?Amd`jJw7ovA0t$i{1m`nd>Up-T6)=uD$ zCq*KqyB4rgVSaZy!^%p^j+>%7AVm4!QYLK6zKO#NK%749=oo5WDS)xInup7>2;*F^JE$lkbnjj z&qQW&OR_RRljdf0q)4NOu0WyVC@K-ArkJaH;uJC@*p#l1!j-Zj+}V?)(qz}|`B zvk{~Z2dnG_e@@yceIFU!RkSPW2k`}JQQL7NL*JZ~{q>|X zO@?6d{{RI*`oGQ1GcGlD(9xZAoVC-bQ2XMl@n+S*^ocb@8#n?s*$aT@moGEg&wD18 zt}FB!77?p)j)X;%s3XffI(cWScCLE8;8g-59oTBZrtc$}ODF(k6511rYJgUWR+xxY ziBp?|)rk{ub@(8o?e*=(ZDh&lr32O`c36g62tut7rI7zl`HHGX$iE4p&ohmG3L#0k z(LaUoy`elYEXMuC--Em;a196mWdGo>n2hDW9SC9yS39A!?v16vTh9$7@sJU`midiC zyZget7dUMD#C(B64J~2p=8|#}8_nv|cV3JAd60;EN&@H#odl2>#UUSJ<5RK{jZMLw z9}Wst;K%1)KZmYOmK$@O&&KUKOqGGQV8q!Hf^X~H->GyANE(YQGG%RXCQ}IHnJ#-j@7q2kKK= zyno((TRVb0+0H7Gk&W0o%KB>RwdgCp>()v$bas<5SFcJZTkp-U2JRQzfOD% zaQ^$Y@~AVC-Cyd8W1&7D{8v~O02CEcKsw34!}41)cJslUI}JtMzLJ!EcUy0kKa6}G zLU9$PU8-ILO<|)FJXlsP6-mrP3(%GimxDs&&ia{9@8dO{ZMHY{ISp!sOXH=45LL6y z=AR}RGp}#;H2TfIW^#J?QZ^^`4`{VQ>UsdSshBv zdv|Tg2oeHn?KL{v24*H3g}zb+w;v^P>VvR|v(64oLC5Y3)tpkRU)kl9&(peI7ykGT zT{El|^SX;K^kpd_V3g2tCQT@f{~PeJrZmua$}MnLIf!Qm?}?x|&7kLQk4 zE7unKK27(`|EB2*Jdk5kaPUsbw)H2 zOHCT|=GxbGJO^n(#bi&blcl=*mGf8v zv@rxJre)V^y|?%G!w5;J^ar-+4_!*TLfNG;i|LSM@7t6=ayz<3Fhae1-?MPto+Qsn z`*758)|}CA30uDC2PM!m@G_K1(+tzgjrJe!Q4ZuxOMXSQr`eei^^-4V1d>&j=Q*cs zuUS0;r}2#h5OG11!kJE&KWiQt8!yM`T#A zWsQZwSNJbD>=0!ePFiy(HQTvQU`@yUxzU>_2DlEEul80I^%t8&A3Ib7f<9Q8oEDAk_9Q%mt-6MJjotgF*uS$}0Mu z7cVqd*5?RKE?KTq%o+0%=Z_YECaGJVyR_dFe29WMn`|4U3U>3G!fa~T7A+puu!mCL zm0@zirfNoyY}Y#<$ffOM)T%ltCe?| zHwifsV2x`r8S=jmgiAVC5bI=L7{Np1DfWF)j+4QGFdh;rTr%}+0Uc>G^PIg5Y0~lhRgp!toRX2g>o$z$=|rF01yK3 z|9qI`;*wkn*Dn1JT~5zPeq;%d!mIrMrLiuFxBfG7d1^u(eWu=AZCu?&^89Q?YVc3XH)^#2MuCt zMiRF}kCw%rUVNNj4?aHL-Fd?nE+yj@l;=enA$AAb9!mB5Vrin#ZXQ+Vl>ZRHINvBKOO(ZpQNanKh3NH7X^}V zEB|TcBzz!(KUJ*cx+^VThmxRmx_m!#g7@#bW0&8@i~0pVw^1(~jDkjTJQl`Z9V)>0 z7EjZ6v{J;?tYT0sb}b$*)$kTTQ_!}WsIqyst=L=hVyc#Q>~#~nd+9)fo4tPD6cP7a z93?N-xq`-XXuH3|iPF3A?b%Lx0*Q3R`bx|47<8-VclA|k3FqLza}lE5mSOrcE#g=8 zTz28g=GPL)AM%?_l&iEq!X?;kg28>6LDCJy_hukKG!(&GdJMpt{qvipUMJEWaJcPQ z`tsrY^oU5{1(-DamM0YSIXeHzwE5S)X;Ax2Z@RnW)aqPLr1Hau-y%a$qh$&EF$X3M zDli49BsYt0==nxdbdBqqJPsW4A&3_oHQjeOP7AOxO&s-Q5+2VIWAZIzP>6h7w0d}VQ_+4s5aeYT?>dt&V%Mcl)jNIVHV)oN?^+8*dK4Lr!^O|fe9-}lMw2f zGdvG?JIH->3to`+kfnyoi|&Om8xfl2#%Km)<~1T|)fcxJ!ZHNJxmq-z`z<@#E02C0+@HNWX;0i0@6tN zGpQIm-wf3%I}%yVO|=BZ+R(nq4<$ON=EGP5Q+w;8xhRR9M#VMvURR0~Jv8~aJNTne)%UZIZHxc%0kV0!9~y+J~-r{Ij@gwt$nDjZDOT4I84h(O**5arEyNYDijIfD;VjqCX1uW96(I)h4< z$y}hUgtRb$JT`R)V1D3Im?1CmkR%3hmoZ9~nSCfmUa&s*4rt=~fC7q~r-inoQIVy% z>OF*7j&Q|XO;n(aV@9fFDDVSaJGN@O^OwN0?nqGLX5t~4PTl4T?_qA6?$nq7sUOiW< zsunR*8wP8*V>2$_Gt?IDfAVPT7pwdZUKrUh)O z(&(Zaxl5^AcVl=l&7;MQ17}|N&47|1!Gr)%h3=xt zi*drGaK2I594Y=)8}kV6$vg&Sx;-Wa1&Mo_7Ll?n6KYZNf*Th4GR}1v=TFY-z_Po z!QqKbgv&n&AaZmYA4vH;?Hrj4A~{*hLvIV*$WL|K+$a#5ky9hiyh4ostBM06NL%l3 zYWNrlgY*9h2IdGDsGU;^Rk8FX3duBI!XN{c5ev7Rs8G%SI@%-%9L&z+BNr(ziouD` zlOKYwH1n{0Lkrdhccc;7vF9s?cg37-YDSB`G;Bi=p*jzu|^82nc8m0s&2%H$>im_>(X$2S>P4?vwON{o% z*0xzd4i_#2syeMN1VIAXkh!qUB2rSB1?t@$R}pJQ-7yRJIX?6s2minGzsrMf#=);x zZB62R;M?(iF`Isfy61YKn&0&cb|a31LJfa~oZD#+c{ntRb?NoFlE2uDnCQH^J=3H0 zTsc)cDB0Dn>f>aF-2LXu#q8L7DYJcF?Y}Q(r>D3h$OOv=V%%8Ey@X6d(l5z`_zZU> zd?KeM_%V(9c45-KO<4A4G!QW-Mfsb2q_@3uMR=Q8hA|r@e)(KW?8nYR3SFM1P9Sc- z$En9oP%l}s^sHh1#%n973`JrlHNt6NwTwlE2RHQVE}2A*Nu=uOd!OuzstoQS(KG)w zw5eL}mKN{Wf$lXr0Ehr1#tckglralO12nia$iOp#VXsobWpaw?M@4cIE7{B{o6YIB_;t%b`%d zKxPAiHSTIH-VQwb(sklmvqbBNOU_Pbo+fV=J9o2SUxq9j@ujtns+c+2>We{qFUfE7d7Z zA9Z7^VihvZV;fw32kDKqlg>`0R%hG83=7UvZ|v^g2mlGy(Vo3ejaqt@q2k@_~UnvF(panz&ZvE}RyBs0#z~2v?cd_Lur8f;RCi1{ZK4<+1o2JFYLf7F< zJ_6>O9kP|M{wR?@!lu}ri`FJ5gxe zB##5e&^5?6f18|^LZHg<>Yfkk4jEaoUa<>z4caOxc}(+@!S`o{FX>HS-EcHU9E{!J zH(|W;_ofcZS8@{juH1dTv%A~{KiE4{jI&scrj2EiLuHCm@-@?YWicJ^@+Rdlb#rxd zD!XgxHu@@U^Pb&qU~5t=O_SP3=Z$g9(&qH0SFJ0e^JGF_;+Y_inj3CyaMhu}PF0ne(~$8Z_sNBh6&!_c#Q7!Jk;$hF)^d2#NHk z%vn+tZgm5Gz);!2J0(r(9Uy6BcUw4s4d03`xo!-KCNz)@#N&xpv$31NNRdUSR$B(5 z^Q#Wqk#lHgMqzyLUk2hq*68!(Xy1bP&BcJ2V?felpSOyV8x02NEBOJ zmSm%Ry#X#oVS$^+LMMN>0jv%3V;C0{=w_IROibOAV9iSws;sCFX0H=A70nx^s9E_!dQCY}F=dtdcx$%Z;0X03Ec9TZ)8Du)#^PjWL8S<{ic*sb zesvN+e32$aE(mX`@t8}uc-R{Z1YN-zk-A#DV_4w^LPY`N;PZGLg0~{bU&28kJWlRo zk}un8-hnHxaWUBl%3*mvoJOL0i#}uJ6w@NqBRkxKIKqLroptW`{U!zlxBx0af>G(E z!2C4Tjg7TJY-!i4r4A&EHrdiMfiP1vyJe=F*WK0Bd9bI)`ALaSY&IV8yFFx* ziYD#MF1%y+uqe9lsl0Vb(GbL9UPwiv&cO%00^&dAEEYbYom{nDb}riCRuPYGLRRz|Y-rKb=}( z3~I!p_ytSQvz2>>5vNm~=H0m2Op6}LjdYw{QX;StC4bgC;p{1|TH`a6rV)p=VJvTL zslJoU;p}{a-gKp@m+*=V`yV~qxS39{a^vynm=Bt~;I%3G{mS+Xg_eeg6@wd>cQ-tC z{PG#S#QlO!CEZgdqM`o=%SoB&ho{?m85d}DO&%qi=;{4;^v*gj;@ZY$;ipd8bUvsx z#5A(s=!NE?*zYwnWv*-&m>3@jdzk2H3CKbokHvq<@FU9>Wbj4PlhqE`O)4}A@B?Dx zgbFL^4u)V-TI8sH#KmBESbY|2`k3lAGvh`<7&p1L+*>fd< z&z6EQ7KOc=S#>|+kN(SD<&jXX`!(!qk_7LQqT;VF*sdW!Gm$9zWVM4<#($H1JuCP< zdur?%1u70C5(Q+F%?TvL;Aw*;OFGA8$Jyq$0;+5ah?Vs5@1daXv**KkSXQAA55@Qk zp)y}Vc7?3~Q@fB{b$-(PR&XpvL6a{r5xyM6SF*&F)F4`P01L2Q`d-`q)5~hLFfPR@ zGokn51(#W z+gfaSs zJazgd-*86Qg4=EJtVNq4)cE9lPvM-+I=sSZY|-cgg^4OBtGuC1Of(i+R`lAFgV`7E z54?91ZsJ!M&-YqOlk&t_yKi?tvud|K;m`3bbSt^`=C<+b(cunbsHl)p^~ty07nOJH z9lg&l0Guwxw|XQJW=a!r8*5r?(E*mm;BH_T?x^QrqQdS(YLuCE;ru%dx7`>??8C7^GMJ zrK`uTijM+_@lOU@R#TOxYU!!v70~ua?hAH+p6JiFMY6L#=kRP3Ir)` zTYl*!r=M?iy)2dGfctNPRyY4d>AIUR+_0VPR zeLA;3OJ72bLrdh+!^SuMVzFSQC-?KQ>di(G91}fluU|jaW&1e$QnRQG_6Mu5)1J%G2uC9e3@@98@KgbWeGN>`u*x-j;y|k;bi)lnko;1D&&tl@R zS_^6(fiK<(28p5=Ga(Lgy)={S2Bn-fLkIwb1M#`5YZRwmDT^PYpl@r5kKsJZtk=p( z!2&Vh@K{xbg$l2VN;Ma0sM@Yl;G}?^Y}b_Jm4qz_24Ue}Z_g^Kz5sZk;M@}8IRO&&6ZlzO z>HX?7(B&dWm!+!q#m-mHK;p}DUbh)^&F!r8I!dd1Gv=?44A&z4F}fhz931^qP?R@{+&V-V)!Eh5AYG4*KrS<{b|IKVRMO0ce$Ra|^eHb} zR2W(t8)Mk3L7SxrG&dpDP})T{nsC%Uo2aPUmn^~;XJ-A9xjt6(5Fl(Kj(M9cFP7D( zX4y^2hnXRPbA|{9(u#`mF2PX$Lz)S!7*&x0&cw%pFK0?R^NgZ~oq*>NiUbQGF0YND zRf~{oUOELPK#{K%wVf40Gb6na7i@;wmG}fjf>p|oRHsOel8d7_!?k!kDmk1cZmj_O zj7X9&1Q0>cWi95c&z8bO(_<&4f0#&__bPpvvk=;%VG}PbYT=@T>p2WS-F7z%F-QbqnKNDSjoT8uQN{xJJpBow{CEYu- z&T3pvCdxE6@WP3S*Ik z;paeWQRW*?Wok=jiU{vh@RlESzj$f(2}p&5kqE71shxeECi(hHrpyW-Dlf>I8V0oYm5CBWl;6!(ip>yNwU4?!(|71^mUwWagpL zc*}y1{oC|Ac#ZIs)rD8?PEx6Cfw2---_IraO<+!Aa(~scgt(k{j|3&PZ`` zded`suLmi*0L$M*PjbOUTSh_bD9(4((tp&?c7}CY>yxmSK>+y^mrOs>PfVJw zpR`!alyFMI$3T8NTO zUsjD##5y*M#lXb0416gkoENTbjEHk`_b?crDOzD(#yKk`t@Q3}8w_C4sw!S{y_pTS z4;!n5OQ>gowkt)!TDs#9rj3{Hl~4xH@W-|`icdmGKV*clfc10zJLNXA%w|1`?o#u; zWKT_ag5bjh{UY`vw)zkRAJ)i6KSNKuwYez`pw`@a&O8`MgOuD?by?3jK9+Zo+JCI* zVk!Ap`A%@_W7UB4)lZjG;4jCmN;)6a4`Us8YPjuH47R> zf^HVqb-7kRGKtMI1nc&>K*`noT6PJj*9WzHxd_VBT8!(iVtt3q;iT59))kVuf;fzd zHB9^B0O9A~R$}ZFO77r|TJ&hGYZt*{-DniF&2`Hyhkd6XmRqRgT2B;feZyXX7As#> zj<+S*o-NTbdQYRsledtMV>x7a+J9f|brc%y?dLqjD#eMQ{cs$;d#}A>KC`>AH(1&x z76=o~rVdoDHkBol^Ee`?b4b^OP^+)nx~=Tir?abzt*szNyo6ICrTuE%SzEmuoHxJ0 zT|qdKB4fBsTf>RLFPkrJaZL+3Y$+Pve4?yQU%oA9Vm3n~;64wMd2V0=r`mXQY*DdF zp&+_ZKu46v+erhTwbas6tK~5bAi>Q&8W%rgijic%xppw4lKw(7&aL*WlEwq{bvR3% z!EKom1~dIK7pkC^KMargx;*~y%23x-7|(>YiNg8u%95(!?G;VK>tA%oqb88JSAr z%TU=etxNgsdB!HXB_ypUg|mbfW<^PF#{W#@H+}}#;)2Mh$=N>wY|Z9oNqMI*$JX=e z(ZA<#{|TsQJyojT%0lRXXn@AwD+0*IGR6gN^$k^qj^FBV)oe(5qLs@=-Z$6gb$Q?1 z^_7;C7S7`52sE{)K<|UcH>wKNe8l`{jT-pZMYAN69Hh)7_&TXlp;b9O&tb97O}1JL z{pB`~XGoDXmh^zZaLrh5VouB1A^o3+%l+j9De2X^=X=JtB(&lJn8hr zuC8*r!|RpyEk{(IdGLL%u!OoVb4?u}35A1?NnK5KWNS~QLn-bnH52K1Ym2_BeLY!M zUwU{t{_^uu3$K0Sn_M6fSRqP5M)&6RPOl^5GiL=&$mfkFOuq{{?qPPLkn-+tg5LN? zlCg^)dabWIadq$SW4|SLmh^qlNQeV;s{>WL)qx{E6)V8T#k|Dxxi?t z3@hr@Ad>+4OFP(oF%3jl(~b&bw<4x9y1LROd!#g*A7n4TGeqhXQe*H8_0 z90^esW9?DIM^q)=P)wx~6?I#s9F43ONIfu{z?X9uqzq5z3h*qHva(9hNJC+cmW0SL zy`IaZi|w;rq||!J)lti zWdd&Rs2K#>3P7uRdr$qb^zqBsSZ|aBBe7y4Pv+AQZs$?28Ug*max~>YY%)x7KCnH_PZM|TSH|~b6YgbqHBNTTpIOWg)8ziEX)b;-jcoL<%@EwHVAXCyWJu>KmK1sb z>r4EzrEOY6>`|4~k}0cp&Y`P0ukOc@QOe{IX){y;Nx6NCW55_trrZATRHo+nXn?|J zAQqStzr2j8Ka^{u|G8d@0SF+E8t#w@eE^c+{}%R9GeGGK%^dqVFMhc%BWtz2Znj}D z@af?U4vwl-cV?D)^IcWD%r_A}z#_^90kxhW^XmryDD@3b4Ue}Khm_G+FbUZ( z=V!XOKsxI$IV_e3cCSS$rxkq(>o2j-C~Xos+MaI@Gi(v!KRoh%{xvhEf*I6ut&TmDgoP%X2>7Te_5OV(f<2|gq>wg5T+qFBjwoqy%};H z5%pLR2OjeG1IH(r4Ew>hgr|A`Jwp&gbpQX$Zus{_-4|1#%_s*~$=$aNqJFQ^|L4=4 z|9raC!vB0ar6B76%hMTzo}+XyBc5(5^7qai1W@vCV8tgp5B=B9T|%nOpPf5o%6;{` zx4X)^*GB}c0SDiuus`RzAE8lbHT(Oti)?vI{h^HE5^$D$TIkCXSwsROeU>*;G7+U3 z5~}1?^Fk)iKBpn=oAwLcOqYFQoRi4shE4G6`CONZffbYS$NUB3o%lcJFY2$q)kbl6 zNvQO&dSj?iCFAhjJ~Kj2(@!YxEyj+ogr22z|+{eGy3ZK4eoS{Z9kE+2Q9RyPrTwj4x3ohwjQ4) z!685leE6w*tOEuZn(l+suxj!rShds)$u)kb0TAM3wg35KL7#9>Z&du$V&;I*9}73) zp^!Vpib+Gcf7qQxCV|ueVRs_)UcmY37i#%~d6jVS04&X$Jhp=cB9U&VByf(A5&^Yh z;$RJWjjSUSf10qDobSACF3!YXz0A^zFH5H8fgu!mxz+wt#olLJ9p&O-LP(7cWbydI z(L6C$)(MiiB71vaes=0jBE6t;3X$nzW-Kd_TxxywR3D}HjEI@;9sxL4saLvAZ&N&9 znxS9xk$uhm$;YCz=0llsAeJWXi*p*ot7B7W?ai;e$kz9^m^GSTMi{w`g~4B7=bcW| zh{$eJ$3N@;IBm1$v;4L!V=kfj?A!7Fmi6Z+mo$LX0JCw!?r8r;11wo@|0gX*oUSwV z^wz^*1Vgi|oBzqsF^YmeN}>UBsQ4Xm6mJ4o60Y&S`B4%jvJ%S||J`G+0&)P{2+O7X zlZyTyZf*X*-!<{wW4Sb$kcMe_h(xa7f@(FI^~`#X3%b<8}EzLo9v-%tP`7eGX1b=xG7*xl4hAPp!t5 zmv9#)w#y!0fGsac4~0)(f~FWVFQFT`>|+ynlqGp8D$JDj#0=F4&drb{O9%%5ySNW? zld|ICwK*oGnG@Jn3gp$g+5afuHyDr>Xa6pi{X(6)w1aKu*_$XXB&4erI=q1JSr-gVOL&4X0@B6%T0|ge zYT`QANUQdEasj)uBYWoIGMI>Ag z$d>J*J}-YCv(CgOtb3DAdC}+|!+iDe_0>b6Dd-Z7Qh8#r;G74^=`+`h9kS5AHZS$O zWd1ZKT2Op1SWReW>1B~P=avD?ba^1_C_?#+;msCG&>eTgQvm1Z1+CuNUE@M0JaQYJ5%bX(o;3Ex*dEvBS!y|J z4IArS#talG@rzoS%om>K)+eWP?awrMspp;F9_BTK`j@7 zvlG_i-l%T=#(z&^6Q_MHYl&b}bb4j#|E?vj7Mk0JeJMtboBrlzA({~N z@8&DYQ<1rM+0vIYeP)(myXvo(`c?JU?Bl;l();YY^~lgp_W!URvC!^fOE`|O^N6nM zck6N55U#xa;e6I_V`y~1^M5$sC(pgyj(j{`xDNdieL~3 zJ)9c$`b)`FEmnIdt|vk)y+Ws)j>W3IZ&I&{KFK-m#%SCqIeB9A@vz7+s%1Ox&3EUE zN4dSBRv3-$K6vSTR;?Ohv)PCg!*5Vnsdu(M>3Vm7INJ8(TGy?CiPqT3Ko7Oe`T4Fk zWwgzxkN0Ol4_GtKcaLS8N2*x)+3UVJ_ywRFUyLf#E+MVgvRvAXw4O$)GLg9E+(>r8zdT4wRDP%xe~GZCy!4YuGhX9{0rFDRf}ta8eT0ipnqO z{IFs8qKApt?l}+e4fTubNuedquYUFwFfS|D6k(hd*gtoBf9Qi3|1?ziLmxam{tSB) z{7?FTwy^-I57@X#<1qhBCRpW?Z~I{x38~GuGj|msYA!%C*S1V8f}cYe)UY`=zgq0ydKA>VzPDTcXIW&Qf`z|=zaQidseha z-J626PK16eyWG2<_R4;~OtE&zRc|>yoBM&%dh#a9XN;yvuS@J%Jt}ICeEJLC#;`Or zowjcY_LY40jNUszkZa1td#c7i$<_CO#A?p_Ur)Jmf#1}P{D0A?`S(=nr-o4`mx{;u z)lA4f&WnHRkJ2*rJ#${~}DM%ruA%kJ_^&vD+}jACoTqInSaGdN>}AXT$a4_O1XlhCIcSHnTb$ z+B=O@N!9j#Pp*qtZN^<&A1pI!F*EbQ^lL4qAVupFsI~V3Zl675)!ByjALj648@mb z!Y~T2r?-q0<(hrVAa>`a?2^nmWGylhV?la0k_Qcu{3tZ&Xxp=&pfW;2REwA(gxW4t zG2C33II*n|HPtxOEL2K0!x~gYKKmyjD{dXpNvl$AY?%3pj`xJuWLr^(Sz`6XJ4u|b zYIJ$Ygfl(-i%rm0 zI1q(2bF-&NPs@3xftWE<($t3;C9bAass&;=_E~~B_WIoS3L3eW__Nq+G}HYKA25N zm-yCqO2*&ckLP9e5oa%BYYeEX@JQ*~irnPIlWye*uP?o`!u(@4yUe2rz1^AV;WYOJraR!gGIxTo1YXwwlTbiFfm z{DHkrA!FM^QFa2$u__xf?OVyDMWZ#2OE*Om<2ViKJ*ckqmEafaHp6qAjpQ6ZoA{k$ zX-`jmF^0fe0YL2X8s<#NCu~FRQ^bz_)<|&2VW|u48P1MA^RC~oJ_bkpa0H0{@G+d7`%%|V!N7%H5C4UDCPMpo`Ko7Vl3OY9CJBWuuOTIl@y;0u7ENX)i)UMka z?0yn`k5YC1_^T>^Ijk0BX{@^k(5sZ8EL`akE?q0elnUEUVS4^yIIfQpR%vj3t)9`k z7qye-XdQsZJiC}J?h_LKfu@Z6_BBmdCh~#=yM9OeQ3Igy!kWCQsPkb3TGR(P$qOku z`Lqf*YE4pRG<7*eTXY16%Xh#e4nt5)H7<;_(ySV_*cNrDpCUu8ua23=R^Bm@8qby7 z%rGLkb?p$rAw;NU-doQx0=&lLpCovUMfi~RkUtCT zOhPK*&Gnq9?2%BeSZ(=@yaezs?w^UER#gBk!tA<$3$(x9GxdL2EbjKcyswoT^5>f& zG3M~&J4g^1gZ+ZBmftZ}xJ!?GGw`tha^8dC=bH(ihsj-MlJ{e=e&1L(ywIIY=(UMM z2`d~7J3_go$T7gEnkXYY%lVZ*e;`tk$m&IdE37}{;&mv^)g61un1XO>U9V;Bv0|yB zr??^?O2+EdxG8yI^H38O`33O~d%-N_*OYR!LX~d3v9-PdiVD!-Y4Rk%9r#TFUrL%IJ90 z^eNisvxjS=pP&Pcy`40aywBYdT5g@cTkT6*_Xt;}EAD^aeAcrU_$kySlm|L`qSY_) z++ecFJ9Fa<*^L(@hG~qE$fiGXdRzdJ?pAYw!SJ+g=>5s860@FwqdT(#5ps^fhv4aW zdJxMGIR^{Ee!P+tMrTS!6LD>Z2sy`rf+HqWj4I1;3ndlF^ zrDZxUDW`);q4#GX={nIp@z;3Z>-AFjPVE8VK4?TVUDpj-wa|!Y4ciXdeZwgVln6=K zVYNax%tj5X7|Z8!sL&nl9#pP`))w->5GnWP2yPG>W*if~yq^YZ(9HV&Y?*tilv@QJ zk1^wDHqF^5&f`=@wIbmMUeoong=gPr=gi4Y67C%Nmhuuqo{-c_&f8DX6wdp6;#wz` z!h4TC;mfBbr0REv=7>NV^BPjGKh@0800}_$zkm@SQX_K85#vx+q+W;EVKDlkS9~v{ zGe&TXc8$7L=4M18E4`ECo!iuN}PC)(hh0h3$|HlhJ)IYM}e|h2Z zH(CMw|NN&~0c3xB;Sxfv2nccg_{tdsgxGfIz7rDp{1`L{g`V#xM_AZ6=7@*TP(8%K zAA)uHv&cPtm$l2Apa(7 z!rha&&DKIpbMX&fhl{0hZ@s!k(h$$Cm8>e`rpa-gwL4EyD_=6?aAi7+!(?;DOMqMI zxyKFPWreyO<#rNR6lp~Q_NBK`dWQ~MH!4TMLb>c`-{G@bzhk=-Mw>RcFyCG{bEUfb z!vkq=^iODxk9Xo@LFlyP^-@$zKXfHj@HNU6M=Fo<4qj4( zJ$Di`0!fG*YN8h4^Ys3e`;6VB8j6+Og7aeK0kZl4x=pP`6JI43LW%bgU5U17yl#uq z)26@nJX#T+ix0p=96cl8?2lN0iKW=Z(~4Y_6g*|EFl*DU;jw$y1{r6}9REXcZ z{aXwXQiZe!VbX|SZZ-KGUbCA%E<<`f^SytFA)dR&(BDx&aEXGqmnayZZSwov<0ibX0pT@ zpt^#~ZL53-{BAwgcJTYrkYOSg6@QLhhsE``O~&1m3tK zY*9IH0!i+C54j<<65t(=JxDKr#9X?=CTbNLE{NNsDYs|&*8d1KFivmZsy$Oy?A+kQ zo?RsaO#@u_VHxgI0^GLs->kOpX2HeLb!`P~ww4;p=!>=<0{k*D2X8VL4&~2O0k7P$ z@q%}~$UHRcSz%XO@?m_0-pb7B!XsA^Qq>u6us0S1R^(B5iv3`_syuhwi>9G9;VB8P7EQ7&q8I9s6%tGf8IttL=YNzytx zUS&0$ZB$a@FkXu(MkvrlQaakeyC(iIWbEF!7w8OamFx7hXdq7!bhi-$Ek@2>J2J@g~D$2^PviFNIe0LQ4Q0ZXorsQ?7I@6WOf7 z1~0TboRRxHMQ(VdH$^zv$e1-FGA2jBrQk9~$`ER<|^*p=r z;wFjB-bUwh#=Pxi@`RP~pYu(=>Ti&jF64DJcnF+?v1t~UHGw}H9mo?n!kfHwN+qA` z%yY4#Um@q&k#_{V#-+e^8DgsRLDw`!V}7W*?ZX(&c%wXgBFl872#=?pS(EL}aD~o; zR16UTqRf|6oy2t^C`25ZV9HV<-gVKe-s?5nLW743TOt?6knu>%XYOX2c2c{jC;H_|q>^We2lbLdn_Bdw+_D)*@+~}f zi~%Wv2Cu2>cC}+cVBc9z3JvQv1o#22t(k>teo#DF(N+j+R}zOOSZD{f31q0fJLx4~ zzU;$jjJ`ZAn~*tFePimi;SgMIrp}>aG2A}>u3`=6L{DGx5Jz8IFcB&3_>G@)6MSX` zD7*%}3D@+Uk%$0Qwg6k#TxH%IsgjVsm+NRg*i6T?&@|3w0gZL$%n zEQCnAE6T6e!5u%K8&$%T0VF>%zylIyM>2ws1|D3H7tBkw z9?P1>2r3^e-rq)xl4XEvZuCZA=7Dp(%v~oe{P6VRjjb^R+3^zyr1XC4q_hIq5Wxd^ zC!ffFClZF=g8PZiG*X++ znO)wxRVZ4aGc zr>0TOgN=kb8d5<;5{GT%L2B zKCR>Q2Nk$FBTt4XWN3nDzL$R4RIk=#7Mg`}mfBpFesd%%ED=%?JTppdmf99*s2t68 z^z44yS0ak&mZUdJb{)VZ@)$XA(zLB%TTqTl&#&d9W`x=K0Hg>rNh8d3%KBHcKO(oe zG&|No!O#5BnV~J_FZm-&yEig_{K^ga{ZHdvA}Zy(5;TFEpkkgIB=Ri04#J}X@1RF# zNd~ttTzN;HnJpbeiz|r3z6*Qvlv#w2*ppxdfdIoE#UjUL#bhD0S*k9+Uq>(m92mGL zG+@IqM{qQd^h?M)031aw4=Zfn)2QX|BO^U$oA)6KFF8MFKeo(;5}->x*=T!t@!aKS;M zlgGP22EVtrOR2v;w*;YMO^f#;zgbahXcf1d@{?$4(wj*I-P*f4sb*WHrG2DF6XOq% zh^4l=Y5=@TgZ%EL{k8WwWEtrLH@Qn z=a4xT2mde=-B{z4nH*}=`uwf2mpH|Ew5SinCpes5Phc^EODR`J9(z;(;>MQ5L@)Jg zTfP+$YF2(!~0L6@q>;+F>wFbw5K{rsdF+Hj|JU0mlqzI3TAzT+kDRkw(r(#;0m!*nSO{Eb`l|wKTRbJjZ8R)R4&_`q?wd4JI8jQ0 zuy*+lO2~k2rWi{3heU05y#lL(?M9&+axqXjLd;YH)%@2<#HJj#OP)a%Uph0EF)2D) zj7H(HUkxuzT3+Z&gHN6v61O7-~>O?mFl+P{FC`s0_>xkJz%4OR=+)TTKqcK)XdMX z*~hRYab_AlQ8x0JqVJCJj?8)C;XA23#-|`PmQdt}N`e$+3yd8;lDPB-);2y+c=(n# z(EyIVD{16}_Q8cnhLz>l)0OF9*F&GZO3D0uFZ>DLf`-Y>Fi3} zp# zHj(#)Lu`$n7np@DGfjb!aU_v=ZHG#sLd(f^al=G9*r0+`s&5gZ3+E9hUe#eT`u*IF zvJ(wiE*@IO9hE1k6{nO8E^Rh-`{o86sXy~r1H9RxjtdfneOj$JoFB>y6{%TFc@mO} zMm#)Pc7KALs^`!}3oIwKhDg8;u9>MgQ#1JGBWG&-jHqd%RqS=2eBHVV(V+F2+Ca~- zRuy)$Z+uP2G_5S}X`> zWm?Rg!wy=g}LDKw@&=6v-2Cp=y zwTNQ_0D@msEILxe1Oy5OtF#&<*=L8bDIih%ZXYGEg4;`v4Hi0bRn(qI4}>(d3u+t z{=?SFE(-t$hye^3+vfOv+aP4oqpjVJIT!v9G1X1RRy&*tEnWWgJ(~*2alG>ihTaKM7zrI zrb4GCEfX<5l3gu7dB=Idz2-Le?SMEQl$2qXI9Xa4x!ODStOL?F9QRvIZhMSE6?F`y@!)E|Tv2<%pICO2Q zfjV@~61kzi9?%!GV-{rT9*?8lq{?bMkOLJ+0fA|6t~lsj$J!e#puHlZz%->{L#CX3 z^C&JYorp-(@m542M6n0J#ipP*;E;XH@7d)IMm)!tcAD=ug<0e_9@GRV0(cp4U;*oE zwYxXQ*#f!$T~G(D!nU7*ogobw)86CwbY-f99+T`+b)|tpb5wFhb3))QVUgOb6K+qzZfI#)_c3TrEg=~@KT|-ynOQJ~liegC-otBRey0j&d0I$Nt{aYg`_WBk|7%|lHbtq=RbTWb>KP_FFeFerqcgAy5Q ziUsPhK0KnB-D+#XE@b}k-EhlQ5eZe2*JvmiO2EtAZu`Ijx~c&Q8*fX*5U$s=a+eGM zNZr{S(cCqYJULIu76QuI6mfTAEO2&6%*bNAkC98UFBS!yd{w1U+&H7Y5lsTt(J#Kf z&|P|iemE!nx=CXH;IR(my|Jnb^9yTBYC2;V$-6>29}*QC>(Ibr$l01aT$j6yX92;& zue;Y*1(Adn*+#Ka3})U)oHf3Gw%{K<%mu#e{ui#;MgM%-lYyqZ4D&WUmFK&j8UooO z%~AkL|DeErli2orn=taX#J16eb=A{RkIcdMT~v?S0weE7oRMo zw8!qWNi()dki=XCzk_WC}MB#=Mu9lcE=klwi62r-dU%)gVcO z*Zq(yA@e*B$v4O7jvYv(Oh9VdEEb2Re}2q6{hR9nE)15?yTRyyO!&vrSz^R2bNz|d zRgo@$+!(i3M)$aNAwoj0<$Zh^r6WaAYvQM=L@pi4g+Y}kW0gDbLYf+slHlz9VsQkk zQTcGm5w;sP-o-8vRT-%GQ{$%%uAaS#^d^pCRwP$^ps_PgQ%1=jrUYrl=eT5t$dhab z)km5i%tqSYhw1Nhs3SdjJ99epEVB?L!rxV8_mR*lV{(Qh)93e&EMM6EU(oo&HHoRPr_G4P>pzmVNZBoQn-*!Ql=P~<~c57^UI<#QW@ zL2GNeG2D4NDF|5m3-#xYPi#%$sGg61-e^tXA+YkcbLJx$nTUA4Plt$cj9%_3q5gSF z0thNB7j>)dSu4j z-7~VL@Kj}Y=MT=-x3C~u?b59Mu~ajc^Abk`i~6FJJhZHeRh6vL3b7#ptG*|6D@>0> znU;80=p|+#(x`O-MWm|AaM%@=v(?_OmXS#9^Y^Zm9InXF%M@MZBitiE6tmVDDJx}Y|{uPRPJrLW7%0` zm(DqJsM2e5)aV_XkO00xf$A#xk%ysIF1{787Hcl|gvl%i2HtbIeu?A3IpdnLzTo7V z;o7(3p=1yaJd2L*^WoZ^c||DIqO|-i6zcrsl$X5Ev0xxOW!MDl3+TIp9s!|0H*hgm zDDhdH9BpuuTu~TfqxX-f1iiL`286<<5TQchiP z3^xSN9i2aQi7Yb#_maQaIso*lr`ryWnq@<;%}xvoPR(68Bq76do;K#01X99f4xgRq z{&OP_u)msznWlMaO>+FNCK;c;i?SRBEUzQDRY|XnMnzdNFi-8IvV4T)SaLi>Tl$MD+3a#_(`r&ojAzB(q>W~*NZo&(EffDE`qR^4(Km< zlCys3Q2vcX(|v6bmw@qkf_{f)t&av6F?)pqO$3OKHE{+LmY41GalJr}EO!dX_`R~k zHK&%t!?MO!-TI(IUwm@YvB(z_UMn=;4*9ze#o`U^PZi`k#-2AExv`DQ?HSl7{F{xL z69kCf`eESW);Id?Ty-*xT*en2-LufAlfqRN^)ifJwy&<;U)(C!Cfih2ojTr7WEWHkLq?x0O%hpeyY^XTidwEl zmS94{t+Vv`Cs|uw@uoUGqJ)8w2UF%=HnpxYsY;i=A0QL?)#Re7^LYIz&~&87+?|FR ze`xS2R=D!M%4npL{mhGvP~e5DEBxr{HmEinbG0f@387AFf1 zRF7*S+rvg7BEGYb--olnj{@@TGHRCfUZP4SeA1kJ2Et_ofk7*`r@OsxDN2r$UC@}x zKTK%O;^_)4%C9%)gF_rI>?w%AGs-{NdTWE8^x(=J<3)n=e>iLnANGuFQ3-%x9Er*w zmw9ZJWfi5OO;PNKf2&1e)=dSzTvN!$UtE!*r+@S)FiAnJ?|{#>#{hRh;k1i33i&+% z1JADUY9*ZnCPs!GNC8V=^3@WKyKdOm??RxlSo!7O*s`n0;?Br$?);vGiV2cSJt6Tax&>q)2y2 zszZu|fQV8e>d2h$`<`=ip0h9av)8-U#d@FpyLfeUwH1||wt;Qn*KOcG2L=EL06+l% z8UU~WfCB*h|88M0009F~FaQk$urL4z1MvTQ3PS)01b{*SXas;o05}AI|6e@}1t3rW z3I(8102T${PyqgaqA)aoKm#Z=fJOsYG=M_``2Tf+VF3gdKw$wi7QkWw92UU;uRRP0 zAaDQ*2cU5P76;&P0RDe7!SDbA51{Y>8V_Lc01gk}|2rrW1|Sgt5(Oa901^uzaR3qz zAORQ>216oXNE8f-h9R*qBo2ne!;k<12}2+e2qX%DL?e(`1QLfp;t@yyg@mDy2ow^9 zLZVSfEDDK3A@L|AfJVa5NCX;*LL<>=Bo>Xtp^t3t(^n1`l8W7zPHzAYd3241MS!6Q2owQ@BA`(OEQ)|b5%4GifF{7u1O%FZLKDzv0v1icp$T|2 z0l*SqSONk|Kw$}JECGup;IITdmH^-gFdPAaBcN~uG>(A95pXyH9!CK11Q?!xz!Ok- z0vb=i;t4oB0gor}5GakbEX?n#n`lW(NRkl!XQ6n0gFqy}Z9we5=J>Bo00;pff^Z>P z%ZFk}qM-JZ=yHu1T1ssc!jyLeL@#K$F_{U4I^I?z<|NPBL!@$x-q=mG^XrEL_S5#i zMZXy2UomV1dWufXyts;_aM0D8ewXK(dds25ZN{wD$d1%O>E}x(Gh>48;2&AtMvTmh zj%;n|f~nU-k*@#TFFe<)UY4_*s+fP-t= zU-R<_0&f`MPF*s>^9~%ckKWIV{kTr8+!{jYhK^a+YjrSuOuAa2Bw&enC31M?0s^Nmqr&S zTI`!1^j*77MS_#5*l~N8-&Z>)j8&h!DD-zq_-MA;aJ6Zw`o^o76W14}z$%XPD} zpQHwHXd1v+-K8oCxJF`aKL%qZ!%HS z%-_a_WzSf_JD%M`x-i7TbG>H#W0YbWb}Cv~u6rq-;Hb2o5?@GW=OgocmBIJ!pKMK4 zf@?Pmx`0TCG3mW-4de5vRR;S2aT^6_*&Z`u%V4A9V^*+FFKmiUpE=0dTD)T7Fs$RO z;X&LZA|g!_G>>COwKREQU4rot_K{^XRY=aub@fp5=qA%p^O%t_LeYMcy71?HBdIdh z+Tyd%iuSzh&Y?NA_wRvsJa{`@1siI=SLb^f1s!}fkmw9KHtZEg?iwnj$Ht};U^{%e zqrGFEb%mc-P;U1->~gb|r3#&~C%sJdZ^SyrIUL)sL{};ucsm>Y($fNu)UvSIUl5b2 zOv|cbPQqen2Ax@kfYe*Z8znD#yHF)xhu(ZrtM5&4e;D|s@oOEyL%P|(ww1SaO2&KL zziSX!OjlJ|-`&+2^G~ib6mH4+rmwS{BWrT}&Q*WethBS;ph@p|t?6dEXtYCSKa2_J zRC!8UxI^sRS!CtB>4{{ndz+%Ep^z{Vav z)vRc$J?0RyO`uzA8#A-^pBMeH5s>kv#P>NZNiJUe&lUkDRj^zCLp*9&e(i~I-0ea+ z30*Ul{sy(s-FsX>|AxU9oE*3lp$CBOpwGZOE%?(ZnG6+;`P*3h85ciXlhM#ZU6W+p zcV3g^zU1I2@Gjr@-|w+HwP$}IC8Y+EXNCKmlGJ3`{cwonOSM0LKheT(9!D*YAMy?s zC!z$J3yR({Dc0xr>@krgmV*Qp)k*uD9~tWc$FZFZu-IlP6oZ9+UhPX0aYGjcVP=Q& z&cP;ZRMhW)?WXTU>J&x`0E7J?*lDz5Bf`YaBTILTfKV)S=J)cf3`_0{s?_*a?ja}= zM#28zNsishLcIJ64{66(Zw>G{;o-4K-Mjs)8;>t)j=IiC*7Z$-Td?E?id3qECmAvu zCeex|v)xN;*iG%Jr;hX1E~UJ^hLZpLi9DENg|y8VGYI)&%3Cvu z#XvsdPzDv zV}2PKZI*B?Y*u-GPFePMLKK~Ba_^DbD?jeNBFyyk#B^L)x(dZZhb%aWgSe@XYNVdI z$j~CqV6$|ViHqK^ksRiw3ZJ$()OS**HG&U@TfMm1CrY|>?R%bY^xtOhA0H^@udlk* zg(>{s;_{PgN{%=rhzgo5TX)$Bj-*x_({AY#9#*CEwAuHC+K zlBIXDlxn_Q`O`aB7|*r%)-!M=eVWN0kIS&Aw59m01ar^1H~REu8A6k1 z08zIiwkM0GKBuNHRQ5tBLM6WXQ4or(m=`E2a+%95b>pQh>^|Jk$JMs)6->us^btz$ z+028+=SRKL4s=C0yY?ED3!N>ogEdt^n7~1*SPocq_{C?JYqVX zo#{jIfT2uzjwyR7tUqRrQ1ivN{RKpPV-`G%e|Dc(YM?tT&XU4q?>0E4N7x- z$AOC?GqVoIIX1X2l$gu_eV0${AI2{)!W8Eiiid(QJ^evO`8!tms80Ujcm9^+3ttOO zI5c){U1Z(X_o)h9du2)8@hbS-=P0KLQFqmZsp1tYv=(+`sj}>kLhq@|usd)Q&pb|I zgeUXP)tof6YyXOMV+y^@b0I7c=Mz4=CB1^p$~|2X(G4~Ah_P2feNk*_@IK*tRJ?f^ zdx_~ybJxF{mymqaRamN975Fz|Q$IjfKWtjFfzfSU?6byOCri)TeqNmZ z?nuIf_T(>3tbFyYp8)RdF(n~nmE^&*Nx@b^&OClkdEobgaevgy9}!=* z)QE@*;EWt_fHd$Wu<`32;TLClEEu2h=cCTRpXK@IKkZ)pTk^g6yR3HlWPS20p_|f8 z`3;k3@)f}vYgYfB+aZ`SpHv{(-$ULV@u=8^cjs6QzQcxB+zk3 z+$@qI;f}M0|6fb*K)2{rCb!>GZjCE;3Uz}0GPi69$&N_tFP>7%PxyvI9#4dpoRT1#e@Lx%h(Cfa@g*;5m$Q!s@_@PmByAS?gUyhv&Y z)UQ@kWYT7CSoq!h7jk<5*o4T?kEjf5sdOE6ef9KChbJx4in^ZNgEgvJJ4!)Y((^Tg zZ0vpTHsJGBz`|HNvb9=TtWMLte?eM!egw5Gcm`*_wv1$6n&S#Cs(kgA3!Uq zj=r|(8fSPfog{6_oIj}*fir;EF#qEb7AXkkT8`d>#3);{$s97=B*fSr(j@*QZ5L9< z9ws>irQ9PN@|Us2W@tXiFpE{L5SC~6lhevkWlbk((7bDk%n_7tiM&r8Y5h_mjZqW$ zFbOq^O8XrY8S^Ok&*T$X=?uhVB&|ohXE3kv=NDCO05VbQ*du|hhjBJ9;h-Xrh|XxQ zO%cVDSL#gJyq({Cnja-x@Z9r36nQ~(Xh9UYd`wb7x^8F^dtst3Y3fly+VA}O#o*YU z!cIR5Qn$ED=1h865j`lW!-TRqvM7(ezzhn0-cuCoc{jhIAR&~sO`*7i9H@pCme`2& zd@OGHolgXWb)l7B`B6?3mUL~?kPlMSHxxA#7VDycWt-BK(9&fal@C3oIC5a)cPWUp z>;s{JZYQafO}&)fr1U7MOt&0R9|TJMoE*?jNRyZ6HsnmLAbJ1wa&FK84wsz3Nvn*_cZ1RMA%b)8c1*;c}e zlq?3p73Y3BXI_%g8M^}UDp!;vu!{1e`|B|ku8^p2@h=z<($^m=Y=*$*3%oxkbJK{c z^y?B_11SmcB96ougHu;yh_TC|Pt-N63@oJ~D_UG$`8vz+2^m^EGtHMc-rPSJ!fHgv zUPFJW8pl?9!jSbsIxFBjnp99vNXZ<}s-rcDevY^VU)IA#^xG-Z8;R2}I z2ap~hH>54KWKD&2ys?gpUW1>$lY5I<#?}`vdm>XrnAF2=rq7@ZEkCfaANZ}T`g&c} z`=|!f;JiA9YPF815^Z6@)+wS=`KCW}3EL^+i}iG#086V9!H=k#zlM#1Lz~!3x6nTg z8)V85*PvaipqnXPAMgFrU2 zgW59BzM9MK`sw+bPL2yr$BXOTjg{)I|GrpXg#;`I!A~ziuMwcen%l?@guF)TliC_H z%|N@5!0O(#x@i|ex!GGy;OGx69c3dmkj^?&)j^S%Hr^26&Z!fY!B05U+N*7)aRhET zf+rxF0`m(sRht(Bh+Cal?c8@sqY%x+r>d5^L22kfFQvH+z7S&T+woXyKKw73>7$w|W z9^PAV+!L=<&@|KQ3ioeTE_lOUP+YF=z0gzpx2L%k{NSvoA*J`t8@9o}Z_3t-MkxE5 zbUAy%`cxZP3 z{B4jNUV5WE^ovvFPxug?bC4l*kbQQ5dSiggeqc>^2&^(NWpDP|!sC zl@-hYlALl}XB&Ykx6QCyZd{WdyFZ*SP7k<7SRq1lc0C%A38O8QquQ)KA#6?fJmz3> zb8T}hTVu1QvLR@1Jc?gu>X1RKmyuOBiM=O7fxv0}jLW|yG5YN_B9zsT{by%QBHWmg z|C&eXg-Az)S-oy@-Le0pfK}U6USx-2@*!{fM*BoNXwqW0+wwnOA*4OUe#|wvCRC&G zEc%5-FynlsH#g-iN0bp*v}x2<+l_3B-BYBNl&R_w$4@3V=ESF|F`eroN%2FYZ|?d( z5Vi@QG+;%x{EfmYWZEpU=$6RI2`(!#@GAu|w^shC8{iNq8BXEtF?F>NW?=cY+kVtr;3H=e5edNegKWF|B0FfHWc!ouw2OQ*KysdI-V zsV2%@rde~F>RLYnCpT>;3MLn2$7UPt^Hd4RV>&8}nU-DHRL_YfljO{9K26eUI7M_D z`F&L_O;xR4;oke3E7i~yo#tX=?-i|QE81_t-T5n8X)Do-6`SiTCjV9(=LXH6tvGtG zaDQb+I(%?&V6v-tZ&|vcrOxS*wt_0rb^F4Vh*vb22 z^1t`V4j(ib^xi@Hs zwO9XEOG?+uya(D;*9*jlI;q#p934wl*UQt^d;hIAlY=m&t7Q)BZ&laNk5`8_izdA{ zCN?*wPBs!}7QcLxjYyel{l;K+?B_>7G3$M;e9G=+QBvXRX|=a88%RHA&*FQs(ej;W zK4WQfPpVaAGwuFSN(6DLvp{Oi=u9)$V&R$W`kXiTpSQ9&W0tZW&zl9o!)>B>(?&9* zzjcgTakKdDSvm>h2&z1uIiD4kMCYF#+Um_h4oPv}ymEs&dg?m@ zH;g#aX0o!jvc;w=yKVhuYy=dPZoii*X?j$)F?Y*q{$5!y;>**si{k~EIkV$~-?X!$n`6#5#Wv%_nZGDIuC zyn|b2ZB}CLmHlH%r4mSb%$Q|{Q6Iy@t!>@E+by>~%f5v%-Bh9Is-NH59n(H+`rfQs zx|sanKvEp7r_YLjyDL6p%rCp+b89Q!sAr^KKPJ7D#Nl-8Qj z$>y7r?VFl)BZ@3K$VMymdh7SSLF(@}-}hRnr?$VZOVX@M{`hw9#~zegxAnyK*U8?4 zH?^-%-aq)BND@w_cKTxR<7c1Kj}l-i+B1!lQ@Y#VaYkp1KTcR*ob5`SLLaU~zYT-^ z;1)1G&v|w(^6)%*lZl^jz7_RWl;C@A-U`lwf}u|GcLy$1K3&|Sg%J9~vLr7Wy9PCV z&p|IPVT9hsx=X{)r;@Z6B`25G+&^X1&MjV^Th5=uHzxs&&sQe5-RW6=yS z)%U#1|0d0hYB|<**@N<^iiqmmoF&AUC+y+vNIFaV=}l=_W?HM7oz11Tpw~#R$6@YV zGIQ!~uOuoIm&A5uoacY#sxjOqVp|T{i{qW=5jy&#J>RFuIQ7i}1Ny`->y7NMs_JKa z|Md6ID4s`z&3dB@=A6C;Q?_h8CZbHLv-?f0ySZk1Z2Rc96(s} zTq%%(NxJQ2X`VwEtj7&m95^<3$9ObTiaRZ;g-s|LlJ8_=!)&C~m6K)2*Ek9;9Ern8 zb7?^3mq%)VD(zaMmz<|%D z`r|a2Yjg8foqY->J?h1_tSzKs-7SCmx;z^++1c&4gb9nZmCbkEqisAdFTQ4{Tml1!kZEi!i^0c zyx%cwrA7Yb+r13+O|P?f5_0^R>|4dZX>ulTdqOU0(7G18qAdd-QVd0hJw{F{%k;3c z?VG1F()R{^3WZdp2<+zN!EAd52)XxEAhr)!-d)2+`B3WVOM^hvOxew`x1#D1Nh+p* zG&kBrgGfl0V0xc1mpD@EHqtrC%>HPUNniBsN>0FCQ}b#$H3xDuOx_f`qMnU#yko9! z2u;#41jD0|tlmG7_vw-EErl5hm@OS>Pkve!$2&AzJ4YqTynCNhvs^0FPT zLXQhJfduuZKmEhBXmWY;*=8l>!|vhD)P=a8ozSz0dCtsUW~nmC)?um@d;4{fI*`M~Pk?7&6CeaUpd$%k> zkN@L8a5%AAbB~m}PGBPG6Vl-tKlBURuUVFOoiWC%J!Wf_Nbo&%KoGxkYyGZ97aMLe z#)Mk(ZtU{xy;yZ&r*o(-7v~pjbmfMMzqg^RtuRrIMvUVRBE#-2(eF<2r>lMuN`yZ= z63yq6{ns?bWbBoGUXL_+DL@5Q3}0w;D2FSA@* zfI>{2!t~J7(7ji*)aZS!x>l}5Oz*O#!&8iDE-_um&-cws*hX9a$JU{7%k3YSfSX<~ zyC)pT-tU)np1EF4d%Z{iJ={`!`nUvt7iS1_Y;9Kj=5z!CM*noQRl`vPVW@r_xa2eS zOT{t4ciXBzJU@gJOO6O3h;Rc8ERW;4n;ms4Y^5Df1gqV|J&Ag=Culmc*Lu+|jE#1Y zK2EI|?q=`MpSZPKA50thz5T3|aiBr@p4K=qAo8#M`c-cCn>3&DT8~@htHWB;_h}sU zM9g$Ci&?4_j7on4j4G;M3x)<7OziztQzss0o(@bDKUG#ok#UN=J*aVc z?2hb{*Zcn9(Hl#0--jq+GxA9>)r00lflrM~4AbNW;Bz6KqEgV^-B$Mxsily~oSu9d>C_$v3m zxk&sySd^%}g!6iuvCVhqgXL$_)TWfU=Uv}DQ|c&95l551?jHUzB3l&dnX5*SURcU= zF0K>X@tQ&O8@pnYc~0nbmz7s_@BMf=+~aECBRprOqQKk`Jxh4E-?FKH^eN`2i-Bd# z?=3>Bqr!yh_D|B@(%7xGI+p-5K+L})kg(szF872m2ETt6@w9L6Mq_2t&?q1y7`cB( z;c=67|9NiL*F6Q_&}o08CqQ?^0iSW?tmn;n$y&shJHLJ}>Tf-%d5v#Y8|Zmgymi(^ z^Xwp1OU!v4uzA&0$?Q7V`*CdR>8k{XZ|fxJX z{p92&rQeLgTrdVk=y3UZSlAV;x1U1$4i()LzJ zwShVCYpXX-)mQes4M?dD%SqTDeNHxSFbgk6(s@PKi(O7&e&Le$V-q{GV4swrW#A zktH4cONDR5+*jIp`{hja zbg7?TQ{M5(+l#yRqKL~rRScW)(O*Xj9?DQ09aB@ub`v<|z28uAM09ha-z zG)fayh5im*rriHS*JH^u?!+Y8#xd?HHr^#N{)kJv_Q|+&4ghu3cC*lSO&xb{8b_g+ zfkExZFl{f;gzKfYPlWap`*Ghr?WbH5-dq#q_jNK!06PydI}dR9=tTI6&iM4i^OcFk zi-}04g0Pi|NX@3uS!L|&xM7jW_}Pg!GLz4%g`-m^gUW>oFD~zf&mtyM5YJGr(nocU zye3n{5PMKCx=*L+zV7otM1m`)iACb~yPXnDErp;F)i>;aXJqAM)#Ej~R^Mni%1RmM zXuV95WVTIRc>+-#;(vruPk`_ulxbzCs7f_u9p@vY?%V^I7|Fy(CXMnCiV z%P{3xsc)|vm=tAEv$uH@<7}za?WUtd;Z`e(m7b}DEy)=(&Os~_M*{CyzNA5sHJ|!u zS04hhCaca6RYK`G_7^P#i{Wi)n>G4IIVUE_BRY$SP$ZIRF$|YKPnqRGaZX=pi-qm8yzqHnBjrR#uCIDg8jK?qhL@Q0mJ7 z+xr{eB%@+vCgEI`pi!=h`PZWcQ6@9V2&0<-{oNS-J#!g@cDeF;eT@~ljqMI`-BGoG zvH^BlPNG?$Y08f2HJ7Z+C9G=#Kh`qZrqFaNiY_u7EnpR}6dtlocK<8v`9U%LazB8^sOYq?_*d@Bt!RKFHMLf>-B#ewEOfvtTDdETxD{>YNFt9SLsh`8 zmCP9Y-k2E~zCs-adLM?jumDe5P`|fiE-;OIZ^BPfn)Kd!|BbCG(l{+?(eu5vc#f$S z(wrGd4lFy4A)h%R?Hv-cEi6sI(efUareMJ31Zh$_;qffaYI~*j=(gENr??zsglp{F z)!<@7_g`NYufsB7HYGfy#+#qLa#X%xw*MCY9eg$Wc;Bef&U{Sw($nLC{=T^2P0Kzd zAk@&X`!{2ag|()0 zxrP05u?#+|D`GWXJ@+hbcGPiRtT{0liX)+0jd~kb^O+^qP0g>_ik#8T%Fe_=Q)@zV z=((W+tcJFQ+}0;FuAjWnD-5tHnJ03n$3{fxNR}Q;XetBPPLuw&u-B*fHx(DnR z3ViU^%raXGKS`-oH9mdP$rC3(fi@s`EmH*V=aGd(Q{XB^(DqI##{JA=!;2f|8)JhD zR&uGtopKJ_aqwm~>-1y?l8}mUoRD{q-TwY!yeedNu%hv7-7gm>Km744yZyRkv6|8b zUrpyGLP{#sA%we-hP%}GQjRw6-HTJ{4j!<6bO+ykIeIDzVYWDW)AWt6vLU@oiIt08 z9UHNUb~o=?`oQU}O&nZ?x2`vVi7n}tewkSM6BSebWMO%fYcfKU?%{!BmQMa z+-OJi##w+gUK2qoZ4@t^=S&cyiRFK^TPCw3MzbsSVCByD9i`e8UIFL3H=kiNk+-!0 z=;1C=cbrvsT(HpIeStlcY8;VAvUZSl!p~PaL9RM^L%KoM4?4)WhtdsSr5n*O>$dKh zJm@rezBj|M?At#E3FWXo&ZeOX` zq%t_sOSDZJ*fyV1gde7VRZIz$r+l55enV^~Eo1goiNS=N91dUZoYG8xI=_4L?n2z| z1&+$tCi@Tpsf@CIoA`k6tb#%*RIuIfgD`E+ z@GVj)XDM97r{4vn#yvDV73j!UT7ReysZPlEL#AlnCenhXJ$^O0#bk;jMk zI|R(RqwZ2u96@vu#KokCTgb}N+NJ0qim*BQ!ppWyp}q}MuQWxntc>d@#fv`foA;{nfv%|Q#yGq@kpVHDp1Cl3Om!=4{|TCZ(@?LQa9!&{GEEW|<@ z_TN6ZGb*|9a;1bY>hN(jysBNL!u`=!$`xlDj?~!m>(bbl&BA+RbCj4{E{<>@=$g!z zhS^_YGflhbo2&Y(n*uIwN>1TpwvA4MS{f)K+WQnDQszD$&)7pc@UQ19+Jp|M2Zs-S zFaC1e_S=C6@68KZ*Cw$_yiSxQA~g{pd;dK(;rr}h>8McF_azUCZ|{6&J2mEviY4fM z6PS-G0yVjm9(`>+NXc`D7sS&L?B<`MP|4nD@Yt$a=U)>3C1syz&L301?7U+0JLeyDqR$J=raPv&-}0{9g0!oh}zg z55_Qo3kEWP0IJ`W_^~5#uq-as@kgVhM-QO?y(5srQ6#3~gcAL-O6CDFb)hC1$Q73IRmmPf;FK#peIWw{n|mOVWEq-;(7A%UH>mLO#D&x37~qg^^c6b*(9CwB0(Mw(s$L{ znS5VF<=$xC`g=sXBG=8U?7`vbj?#}TftOq)l{+OU{x&-QndHb~xqfE!@h6p^Acw~5 z@%9a&r&ZNYjBs|Du&&b{6U4p5MLY8#X5wzjAYN zV{gU&`pEeS7dM|Q$k1^E`lUtG3vg_eSNM=977h;kQNZ_6N1q(ab7n__Vxdz8nSAAPPwU;9&}$~_GkIk zwpP`cotFkyH`L|ftY^Y7?(r3oL%#>D0M&bq0%D`Z)hoPNHeeK5F!qS=X$q>kYv!h@TK==P< zM|u8lb`-)WVn5khF_J*XCjcUb(6y$pNIlqq5Ya_uK<_7mpl#Jt*+S+Rd*Q{hx2~K~ z;1qU76KFUg;Oj<1Hf-i4Jcfdsp)P?2SqeF*X$=gy9 z=l^`y6GWg66noQsmhepc57>gX(QPx5;?uL_>+YY&6iN3<1=xwriwYGu38dS!Z+>5Z zv_49dPV%=GsZ+!--y__%jv8J(R(Qo-upeaA1G)w6SCO+adg8B5=l@skn%$L|hX~de~R=WTCA|*v)?P4dDfX;bo)J)#>IYC7a ztVt)3rI_s|!_1E0h4KJq#JbN{cC$@r0{PUHwLRVZIW2n@9$N(+?dLnj3x6qa&GYR9YIE__fD_`&m_)X^*Hpp5eEl%OpU-B?Wq3DMa>MWXh;du3|v$HOYL zw#ZRom_+Ykb#PwXQ4zx>`4CI-PN;FcH$&GUOL^9s3rNgU#3TWeN1z~LZwpiO5(W=Z z95=ND6g+Oe-G83WY5;lsO>}~SGK$q4?x?TYiDNv$F0H+eVl6 zosIyU+i*cPj-YupJ6X!1dJ31|_z^OdQ_v`Z#@=s?>?ZrXgWE-TA5NXGA~{8>0++J> zsTAim(|+{Zf66$-bc_T_IF2N8W9wUHGdqpo9wE)akEwQ*oCAZjh=>b^pYOYLF(ChM zSoeXAiwh!Fsu>Wd1(GUjtj$x;IOD>p@Kw8$7dU>q>BEe zpodJ)1_yOTOBvY>4<3g&o8(8te=>G`_H#Fi_WR05k3Y_4d6q>EkwoAKqA1Z?{CV(} zz>D0fcZS8DE&HItdjcamYL4LcuD$@hIsWz3&nI^s?7Is)756X&?%7;z^n&_E+QYrO zm+k2k8gzr=VD_gC2{qkIA?KlLDw51ySTlXm&IR8W|2h#oTB6N4OVh_r!~o$mc;Ii3 z22pKyoU+eZx?T5h6h&X0>f{6;7^F_5Pgk@cVgjysT0!AZugP3r4)DXEsybysu-^48JEwTT zz3q|o_u~Q0LZpEGJ4K{y_0WSer6kqQ6s4ktTm2!^Wo}t%UB&u>g!h1G_ZbDfsGGPq zIe-i@6#UV>MgoZxqv+dVoq8Mzj%P0}@yL=0n3C*)5UKBs07lsb^O3g`eCEhyuWz3<8#0uOrVRpo!*r6j6UhTU-%$9B#AIqe!F|r4 zU}hVw;ekL~m*^Sh1=R)LCB+bBQU$>H80c`4vD76G7mR2n*AUXiegbZT&}go2AyW@k zRCbdztDNPH+O)DH(}Pw>2*MmuZK>||pVM90Gw%);#_PkV12mm`_pfI%UVxn)VI=f6 zEZoK`B~;I(0(ydu$T|iYjd1O~ar=bmNJo*BKHI zYsDpR7SyuP0_JeQN|-$rhTYzLpzlPPwE!}-Zz|R-SX0lnSSEtDev4v@Br0V?_|R6y z!KhfBMmD$no^fiKp`DnjPlLKgsK3Y9f4qCppG|biM!%vCO{G=};vnSQ`P`SJ$jM9d zNKlM!x<*0s4vDL}Y)##YJOb@u+r`8g{XpNc;j@%{@Sx6LAgW42t%v008#|tHm0Oxy zgnJ0~9H?LHH)S3=f}l6EfJEn88ch`BlgUGB=Tq}L9y~{#J@10wU6lP*d8x*#Hy!hw z0nPaD(3yj2DgWJNt@6JklZ@V#I-{Q$yMJG;=fjZEO+R0S{}T-Q*NdAly2^=t`_04n z&DzIzSDkPEo%m(E**Gu?)-LV)4sXL+(Dkzh-_|^O5e(UGV9Tti+v_N=paR{kKEs@< znr8em2xS=j+DdV+Lbv|6*KLhPsF5yD5!t_|+dn0O=O`ft9DTY4fn4FbdJi6#SSq{v zrmW6Ce`VC7r3&-Vn(DbZ+Hc@(d{Gfr`KDg`uG6m3H&c9Rium>9d~IIbC-kS6wmOUX zTo=Xt@5X$^{_64F5LTjgIm=Klk9xh*n0MuGC0Kv@vNw$Wa;%({)uXka(*A0V zqm5y?>rBZBi7U=ytTu^`ud#pIMu2PTn^#A({)IMK0Vk zjbSBhC+~MLwkH~6t>bBT0Sogaq9=*OXK~53j7+( zW0Ng)Xp;?h6#CpP9_1Vv#e0X?DVgjncJbnc*k{_G<@8!+yoOV*DGKhA`vFebti%EI zl2VK|o*vRfknAs3c`-SKI3BwpCpI2x=blKap%}AFOZ}u=BldhH%K)P~dKUkDE_r7& zPm3arTogOAXD+87dG200mqQPAYgQ&&`pMW?9t#o3gvIT=hSNj%sgF-yLg_|}!v zQLyj#&__zZ*_Y+|10xk_o^#hcDub&3D4s^;>)Zd6c~a2nJJ!!(+!uKK;LOU32DSh9 zP3T6uIJqqPCaVTmefL`MpPXNvaUM&W3PF*#2l}OW@b21jzYP$TB*Al!j<7crMMlT6 zLggp9I&^tMp96#!d$*6NK?Ld@qs`FmY{8%#j2a~9GN|s!@q>Se4oZL<#R=>_qmon<8jFW$q=#rrs4dX@27g_IReN~5l z7EtHC=FPUaCA*tRc{g-Fjn?w0DqBojsV3j`Zc6o3oU<%H=8 zgSf~oL>l~`L4*wnsclKwtVQ30B%`|9g)%OI)fx7JaX$KTo}1scD# z&&nTvd%d77dV*cCe0#H2R(=tl7xCGJgu5XIDgZ*gb zb1_~4=;eKEerv=wFq`kt1>OrLF(Xsz)yP+Nj3l!Z0*1sfw_7h(#cLxZj_5Y5fSmz9 z;vqqV8W5V&JW#fv)4x^DW7?73_@`iyWZN zUV}B$<2;)%))W1uW_u4rB55+X&KkMS+z)gSKNzB*`VTi|LE~Uzr%_aa+c#dsSn_s# z4}3H;h&Xa6>T)I;RAH4)XO^QEB?)7m-LBpaEPPW#Cv2@;@lt?)gk$YAUUAk5dAEKf z7E9vBovBGh2!}+$urXZUpgN+HLlgqMrb$^qC3SNUS{O#mW~o3ZaKuR3=IaTrU)|&K0i+oupLbaH3$)=IQV&B` z_&62|QoT_z=_n9PeWyUntd)fLXF(NDLcvJ5kXhO{4>AFsP=v0=Ah$hem|MnE`+m`q zy_ZFm=+8tgS>lo0mTDMF6qFp&8iRJoAr+*So^70($!??476R`e)VJCS$Zh6-czj5g zAdcZ$9~6);ZP1_&yWnd%caSDhf2WByN3RX?u|r$m8KQ_G2V5|v-Ow3VKfdC1HnMx% z4Fp3^*buyUn;1SNxnP)@3bjKl-gVD5i2{g=f2|1IYwMx*oV}=ID?L!FX$9-jeXZi! zuLmW`j<2rp5mPR0sjU&ss%|l`RioL}Q!X>iy}!V2>cIfYiLWj*2Hm;A2Q9BI4|kk| zZ?@-?ftS)jFc1Ac7f*4AKW0MeKO4w-EfgMlG?xfeXc=7`v1F^imtc@XFzXeYS6R$C z2*fb$Es#ofg;;WW^OW5F#_|Xi1)^e?3m{wzvt&EbzE~N#wdBE{bCa9}>Y!;5ek@|y zQ4M{8U}jr_>M-yUMRnRPV~lGUqp^PseM%2*zdCH(TVnb!$RN$WTd$)EtFSI4)X2Ft z!{K=>d)S9G=glYZCb=UoyS}_2TR0!b!{TxK#pI{0-{6{eY{;oxHv-y>8Ba2)hTnV6 z9p*eXL_e_#GxEfav=6`EF(LRPLEcGN$>`cxP(VtH^w<7umz5{kL=as2NM##V0xICN z9v_tA0pj7%TU}dHB_(Ei>b+>tfuw*0?m){FzzL+hdX{ppP+%tQAF;{8S zQf)o4XZ8AGzCNX;#?5g5)`yFQ_IE9>f(#ju&s>VT%8Sai;&YWBGS7riw^A7nd)O}a zF%_zvmbL}$o`~OCdN*s?s4U@)?5z5b&cWQys=BF`{9N~ARtXJV=z+oOGwT|{sfQj3 z0O9U3D+GiGN>6Y;4`GvoW}RJ)&VPaDKi9ij{n0~`@(MlsY<=sqc`uKeQtYRc4TeqH zL9ti|DdXkOnJFiEJTTb5%K=J&I2GUqAPZ3XFFC;cU*tf0kv?-oKB=|~rlDjYkwxxa zY4G>=Z^Aj%6E55S!wVc~EYnZAvLDUUHUC_ftK`48GSXBr`7bX}uetKC9LNy8J=$C~ zTVe8#7uZrg|I()E-pA3Fm+$NT$^lOO)|#az_u1;(W39Cx+I&Bc=Igh;`uIBdWbdQr zq2Q+q!WD(&jwUx;4kO5vn@fo)OkbSsfg$mf3GkpyKcJ}n*DnJMwU z$@!D476o?Fwn&EFJ*>3d36N*a=9y1}I1X`Tc_0Tkg%^s4+6L)F=CYrsR%?i5B}5X_Lm=eJ`QojU-}NRlDnSyR89 zXU>%XI#d>`UwjlQy?-h^wHYR|H%x%t3H2KgxIaj~EpSJPhXAL?@CeirHkdDgj?CiTZs}7sUo>O%FD7%xI(gsS$}F}+BqL?N#fo7a z+)<<~Go5&kpgB<0Nf0T?dV~eO$kmHZp=`e80j|h98fCrXaI1>0dBK|H)l-EAf~vjU zQFPVQJCoGIp; zXORNIli?1P14(tjGQ@!LWz$+x$DV~lm*^gR3OYCj;KKm0QKvbCxpn39f`|Hd92 zDa#tNdtkS!A5LX~KOTC$kmpxyU;-QPDaB9AiP#Ar4T`*iAB?isbvY)?xMi0hMXq=9 z50APrljAao4-J&3R}z7qYOU?0+){YI!`9P}F-MXZ3SM_)YTJ zYUmm9;Ovv+5;zzYMC)_DPD<{>jY{~PY_lGxBl&AH)7t0PR<_%#UtjWqzWv%RdTf8@ z<2`x6EJFU2bS2)b<`o|D^NA1)(EOeT2|@z!uMe7Of8>9CG`9Map>c!&k({s@Nq;PjC@dw#2 zH~xXb73^n!iK$!gcBdXj4f5JUg09gUQyW2r?Jz%!oRhzJk`B#BVzLKLZmWKH9HY@DP4z_?Xm_ealB z;|gzw(NdtOB=?$vyV*?ukM^`=L-Ef@J-2K*Nrz;kzhDaq8Bzo}MHecwGv`C0e&R)3 zk@`+t6YV+_j4sKMUy}u~_&BLvQ%tDAIMe>QHbO~OaHO4&_3?`r*F@7eP8T66mpr-x z>2zrYrkkL0I;caE3>|VOS?o!Np2@9rgy+}0Dy2X`AYL~00ack(ySu)Q_DsB^JinN6 z^Pse#Q&MnO*?qe{{W~B&RwuZ#%pYMq7Vv-33_&)OXeB}cNC0UZUwtaKN+cu;5ZC`7 z7hr!lF}*obw@J|O@eGobeMBx>_u0QLK>53ofePMNkvT~fIu+yz|6G9gM#Ev)#YhjS zUG;G3r@H|ZUPASS;6FG2xd2USK&icU*PjV~bTI_03OpyQO0{9Zlxp$UdaUXco`veckc-l0ae!)h?sLVRdach7q>exdo`16pR* zax~!RCgRjJ6}=Wzba!PqLq$9^`^&x8#9YxZDxNgw;n5~fV|P}$@$H8%!ingOj!t)c zf7{DD#;}L}htfS){?wO+rGdR^PvF6+Z^D0eHne_lb!%~+j5v*3p!-XiXP=_lSYvDq z&)P{byjYiDE;kZKp{WP37>s2&QT!=DB=^@z;vIo3_es@+-Ife{BVAz69jdF6tULEY zgf~*nfOtABatjOoG!0ORwQP@k%WMW=hnWg*Vdb&))cPWBE6eVGYKFA#e)%uWkl-&> z+XZ%!!`p>VfucJ_IHpd46p~-8mBm4#qm|hTqMJJ}Mhw|9%QC|}a!Q=t({@wo+c){j z3nSwhnakAT7)9c*E5XFLxrn7qR<@-Vg%YQD9+FpG%d=vyV7}`BNFDsQQ9~Vvrx&w6 zz;izkz`=UZto2ux0T|?`0YEBPy*C&gx?f9AiNq3`_FH%daPlAhMu9qn$AqpQ3$F}zA`}TOo$_q|TNe%71jWc?C5t-|C6=+M}NH`5+&^Y2C^H- z?|+ZM0}CwuGQOZrAw=>?s(jp*z}RMk0z%wE4eo;rFotY#FJF1iPD*>PGyrLXZ4ZTL z=JXqvMvA={LIu=_UPy4o%JInf&*db!RV(j3he`QAmnVZKy?;ie3T=+CGd~pm$ZuS4 zv;G-G-A3sPHhS$psOqK!Y?#P5zC7A%KSfka73J=;`=*!Y(KvJxC#` zpy_Bv%V0-c=AYIY5%hF`2tvmids1hfCIQ(INomkNq#8C|NTL7({V4`jgOpgKl}oLt z)_dW?i^@?-36?N}L@1eTPu~MyQB!kh3B^Vg!@XE)$)sW`H_X5DP%%x=a39yjVJwoW z1V)U-^z*VE#bFFe=${Su3tm5pH%TgiS6~K2ZyhDrjF&LA4G%~@I!bh)dciV;8I;XE zN^&=N!S;T5P@(fE**ED0$2Ml@`tngq@c0X^+sCQjTl554iZ>{Djx*p+4LYsRQZgQ= zM1FD6#KxOYJB|^&l|=(eQ9qHJQ3$_Qbifnyl}Cs>{wV;C)&bq*5l96ybF5b+c0)BZ z_vl2W7G!xCY(PM@qiB7*o@a#1yHYQuvE-QbYJTD)<37b8g)=INSyvbX zC(`y_RnPU|CxS8sM|zbvv;<8STX;onyM`ptF)9%x@*-!L1PMGr3)eeF{VKRmnQL1zqkTmyU zE?++5%MA*F+TDBS3m9=;iekBa0TKl(K-}$SE#2Zoy_-z~ZK`V&kRlMoD8lfO_ryIm ziAtmIKq;~tJ-6%Hhe#qHO_r#cCH&dlMxz23Vyhl=$mhNv-9A@6` zog{E1gCKC}h}{sTAoOD-eJLJ7p48f0am_o;z*b9NqX=H+&?n4fB5qAwJf-H*S`f2a zCxx~*F>Z+qo!~s*ZFQU`TpdOOG={mMvc73FybhM1l!jG#UCI{bU+{jk`S{g-@b%*r z$>1UV)#Kc=t{ua=NAJg0Pdm@v946n7)eeRo6*TDNB=H4iJ*Fg+J?y6G8t_|E!iutF zQ;>cGMee&f2ueF7gIEaa#18FJLmLNAi?wK(n!zid_7XCTGdP?_nrz%F_=e%f zKLEdzJgp)c}h~6C0f)_y!PTsnN%=Ze2|6XxXP= z)EvJ^yAyOg$3EgHmd{Wa|8NjyM}^QRVXocr73n1RDu%++r*Z^=V5#gueQ8iy5|u|O zuPY^1dz=JlO1s8X*FK#MwVOKr4ci+Uo%H}Bc4q!@( z^H1+v3!X814PI)B<;+Egc!jCb{K?iIm0Ybc>8(=KajyO_!Ty+!QrgrkcU8(hgzrkO z5vGYehx!=pwNk+o+&@(h;Af(cBZoRF^bZbG zOm2M6)BG^l-FH%o?2_O%9-MI1N=Xyu*z#=|=>zh4*$O$&EMp1d-V=0rWcBLpXvXvJ@sW^1otI9XWaVK1JpoZ@_K#?VLL4OtNHhNYvXoJ(mglAm zqA&)#pwIBsIO;$tP9SP-m=#2ks^3Gi=UE`6Jhv`r>~~a6bL(7!L1ZG+@ZaI?{tN0 zk@>S~?~jNN!fH1MQ?p}CkE@s0g-aVBnf0)>78!drDG-qqX&UYt4`@HRF@5>Mu=!q! z9v^n0q4%NK_-iBlRFQH1X*kz3E9(Xe2!_PY?>A9w^7 z%YBOFRz^L%eD5C`6*e8jK7Y;X;StZ&Q3Of08;B=z=!`bEC>&`*13`yNABF##(c!RR z2<^jSff09Q!c@{3mQSNxwiubS$Z6KWHEZ9Uot7}JA~!qY`8xyD39#;{~d*$ztzJ6l*qM0t^gwUPoheT_xpB&DG$ zEDS8q*8;evS&TuDZ-AsGD9?X2;{Shyr=Bf|4({{MbHmoBoHqwc4Au>dJ|tT={>= z+d4zEp%YU7^W|Uiru=dv0au|9wX!YG*MH_SQvl0OsU2(kK36TZ{uUy7`0P_(oGGtm zb&$7lV?3WeO$c0CzbBQ9tzf+4r1$kRO|qE%j=#WdS!aK@s#72`^11PtQJ&x3&x4Q2 zvnEtj_P(ux-wWOg`d*;=c{ou_=xc`j;r4A6K_DDz0^3_?xBNE>Y0E4b_tIU|7h;|w(H69c@&WR+gtVb6C1#c)euPa_NRzD(~9;XlnaWhVUV8K zHJl<)$^H=%5Q&K(HQRAC1%f~rRoEgQR}}5?$$G-Io5Bt%F#Zy!2=N8?j7Tg$2tqHN z8#^AQK&qRb7!LsVCy~5eD8L(N!;7%|g68}Q+IowdPpLbgfcUh^^Uoo5jv`+)==Hz` zvO3jq{L**WT(+`(Dd0}hyYn9QkI2uPT=GI^f<8Tg2p01e;wUm5>EjXA?j=cr=@<|a z;)^ehR>;92&yaYL@-FXNyPCFy7(AT71Dq~^6Eo7&Zn}B1^AESD-TVuOQ5UTQTotLJ(lsYj6S%OZc!5r8DZ_Lp&l9@dg*YOo^-gt$6ILbfvn=oho6$r#4!)fY1&Atkia zjaU+vHIQ{XfuSTQ`mgY$FmM1t_@%Yp_9r7c(_t&4!^}Kf5H(*M0p3m+ye_!uS@_7j zk+ehyqMLk6pMk!tKKiBX;t@X1pp=(+WLW$9aT-3UlwU}9MDNyddiHp!pyJ4gkzQG9 zfp|i-hVH0Y?r~S?%Lxl0hS5_7}&`2vp_LvE<`c z^bB|LD-@0?BL*>q#+<37a?KI}!x(&|cS=yqBW|>tLcj)f zl|VU31oicU67O{N0ZTGgz=2gmk1w4<>nb(KxUDjO7GzkDbkv z>HPF}V%M8&0dFnTq%3C zg_mP>D9zrhtuftl(V*^O&bpF;=LxFxN7wu?k+|!bc1Q9m9q%ZZM7e-CetMZ(I!_ho z(h%My;*A8r;)|E7i4xXeMNA2Cu7vB#eFzH)s8|JQu#NMjA@?362I053Bv7yXh>mlQ z=URoY>UF}(9~Zg+oY~Sy%*b!9wY9#EiN+ImEj6|{A71sNnh)T0Z@BZiOXD@ zYs@pFP=;6hSWodDYSwSL#z5|0?{#*67%Vs?BDyxawm&{YKJp&3{d^`+i|NJcvJBr; z-{Nk?x7uWvd=0xk{y_mXZ@T=zQ)WnDc}-0~-^zv~BN_038`m)2)`g+=!4rB%sphCZ*3w8yV&*DO2pLtt3Lv~icW zm^WV-u^~JT_JyZh6gfJS5y(Du}ZE0`4TRmtULFkKiqKshi zVk2My2%5$K`9|;Ms@d<@3L2my#eg3DMobLZj7zN7r8!VCuIrRYa%ao9|>iIva~dFr||Rh;Z@ITRW?3UCs2Y(5hg< zTQ#`FM@|rzdR!ZmQa~q%!0x6JMPP=}g{ubi_1!9!WyYM~N~rxV3=sL|get`K>NXo&dBYW0f3w=Din;B|R{UUU0jQsHVnF$i`zoZc4 zuM}GOAEnTG4**C3SMuG&Y9*4&x{9x%|33Fu3c+{l|Fa1pfO`yk#nK-9pge+1F;U+4 zXI;f)p;{T{-0DB@Z+l zYdu&fGZa&JXo9X@3h%C_;XL=;S8v0D+rCnm3!3LC(PHO4&R*H~;Q91j7ti$5l9+Py zI)VnEhLy?R^%kGEe&3v}B%{B&ellCz>qL$JL+)7pW(a)XefxRm*`^c&t-P@Bm;LR^ zGk0^X{ns5QB`@og>K6$;RoSziY*6p4Unk$!EW?OL;$88qP(o_#E7gYq=Vyo1SBxzA z9vP(NQ?>wGht}h!T0psND5OCCj<;u7P(SncuD~%BYr9QC;K;g`rTuD-)mqFXqry6n zMhQvqU2?QQ-c0$q$r&qNaQe&`wi0RYYv_Tdvv6HGNmLCvSmaA3CM}qwq+8rw8~{)_ z767^73ve=C{m~EBxv_==YlLPZF}9I&0vKA(e|c~>*CJg1Paa(8>F#zRmh@j9+`XOR zI6-n004#vrDR5sn-6_qqy0=>aZ^lO{!{^ ziSYnPcL3YV!rNOE5wAQas+VgCS00|ZT~F?bqmoGXzj%VWP7<2ly*;jJo?PX{Hu{hj z0T`%~FlaDe%m+f7z6lBjTg?46rW&t2Ki9t77&0bBtF$X=Xkc;l?A1~F$SXTC{5HJn zPkEKI&9x=Z=_hGSp_`r*iVo4tR*PcRU_<$@zKPOWAy~$?F26B`gxmdS@NXY+!G1cU|`JLpn&}S2|Cs$MT#IK`N5{q%mu#RetjtjD8@lD@1Lgv>S})* z`_oHJL6oI(iI4dHIT)!Tl~seD%lPR&QyZ`0hgZ+`!(@@W0Q{J7L-`^Kl=v+$Pvw0d{f`NW@jRg7Fm|=A=5sLK;jV)acGBO+ABjdOe7qD z0Mg}3BpJr7>JPE$f$8AeP&4}zs5@1h(o|}R%`QJ1piT2 zpf5d3>zs>p#4lad(R4)1F%KdpWXzTn#FBBSYyU{ru<(2)W8$leU~-7T#n*aHBtlS5 zB}d=>;#fW|Hz>VI)A}=g{V}BC?4X~ zGb_Z#@j(_NpnhTL&wlOeURTbc@BAso@a09N5T-p_6zm@vPZ}yOc4qd;zX)dtVzIE)}+wiz+EF<-?s(Ui?#YP(S;0|GFddJ0gKF(#_ zwdI)Fj?n_a*6ORSYAYZe)BxoC%Kpi!hGM+#0+Hrg%bm;(q{8~L)xv2877%HBts&af z%aLKo@S9+B5xL=<)zjW&AMa>=U8q-aj6uye6!tHgc}20$uEl5$>8E`6ab`y@;^3|I ztEWCFO53S^Vfr-av0oI&*_%4ekz#TL{%F7f$p>OD$yr1(*jd(BA+;F!ozr%3vbw%fmk0fd_ZL~d;9jK zOvuC%dngW^m_)qYeN%In8DPZoULM!sEBQQi+OyIjo-n6xk--Sk4>bZI9waHZ=Ph9R z%0aDAXWOgZ&E>S7A$=i=#U&r6wT}>^M+oQ^KxH+*C-GkQM0e zQlR#z41Oi;i*iZAce^i%^Inr>IQTjt(dhs?K*YcNPK-w`<-4D&@TxNa6cJzfe(Pzr z$ht69id^|LnIlx#kpd(gCFJ5AGd{7*P%b?nQeK4p*;)FA+_$b?w1-sy3u#JL=;?9l*rQz{w{)LySXfMkqxQ zGM?Q3@Z*>WX|P6mel{$*wlhrTb&z?|+P50qeZ*s)fZcB>x;=T+vfP-W8T`psf^lO! zux%KuhH95X3yYH&CnVDv-Dg?;z^g;tNo4Nwjac0IAWYe;q;W;vo%9K3RR~C#qJkY~ zjk_h0NyYV&xP4E!;duLb!?;!o;)su1z*Qjx$vjSgK;bEZBs(3mx6OWphDQJx+lR#5 zOS7NdegtBLTDq6h70#DnM9H zu3EzY@Y;7+_V_!YT77q8yJFYn7r3Bve}n8N-_bii8_)U`<5eSx7vWUuzeq1gu0Ven zf)LynRKTl~g48FmO5e3yJ%)%sUrYU9km~+5zp3_%y*Knw<>BT2v&mnc6<6nCHfFEE zYZm?YiP4t=z{wLzr-g8kqXVfh5CGE{*$hYW@8H(@1U*^A9zY}fF|iM`qj=kKFK1$$f@6bc?nU*+8l1;uJF(?r zqj=Am%?awV@1>N9{c+DgkS1O`{59(y7*&Kxd_10N1V*Kch?h8Le(Ot{hcN1Qi6<@7 zMVio%pgis#Ge236edxqiIB&EBfw)-&owhOU_foem8uS1lV%d>TP>J1}wy#|D+cCJe z=ZPEqNk^rEj=>_$l?kx|u-Petm)Qmy{gEy0NiP74|f=G-=3lzGAwf`xMC!q2D!Z7j> zZRrN;%jP7_#iR}4lpg7n-qM5vSb)y4=5Kc`2`9GMdV@~1!EF*MB!NbDSQr4a+r${O z!|j9;ly1CqW=ltsrmQ>C7ikLx`wnd}bo;Y!2N;0$q^Mb7nCocI18oyxpm zSR4`n2Aix5H~dFrxY4C2T<~A={H~6Or1QTymCxx-Hv*~{G$!+r>Q@}@Qo_bBFsKrU ziF}o5w$Y2Qe&Lqt`CQGqdml%eJWzFZDz4O?jGwz3T=X>z*%`}h3w<97BI2@YKfVqg zps9I&<~mtmA8vM_pIsyI7_iJ~U$eOCG!#Xtk|&lUh=n|pb2b9F_KcaJEXaYrgZ`Cef|epBJ(gWa>HLrd;6oo@BMCa7=F z@{@^QZflzQ#X(k}OU@k!4L1_s$)GOFbK@X@{P|i8q{T`9Ka;T=@q(TIBx8|NB}gg1 zolgKThXWy5VZ$AptRPFkfn-U^d!)z?+drpib0N`SjnQCA2rXP>E7M$o(ErddMR6Ty zYPKuDKwVT?9D}-`A)VWzP%L@rM%rvEdlr{6f zL?kKUO$Me&gch$ZpycG;jqe<2dXUHTkp@`W1X;*@9x}ijrtl=1kxG~ch4MUbp+)on zUi@6J184l?*Hm}mu)jQdPlK?p`1o+S2ScxbcHxC1S9EW`IbDtbH0gHT7bEiXEWzaC z0vCcRIO9G%A(GXg0&KTvmI+14IQjU1{GfTVKl=DD@tQ1|OHtsa2_+aF+_4@Pl(O>q zmXMPS#LYjsRWS7B)rYm~Iv}7*vFDR>Wa5Vu=50`Ze#<+bp|{tTv!l>mhjPG|UGL%v zG!W?#@xluo4$b>>pKl5^P6)#m`I6XEG$TsGWm)XIf?ELK@6oa$orS z`I{>Co8n|i2O_1eRJ#R4i$=w$uigx$+1fDD%4@(;Nji!fP?1|a(GYwj9EJn1s$nQ( zk$AW*7AeNnPRPNaD3FK&1qGQtwi%O{=XeMx(iLB&TNH&1UZ*9-p7-+V@sI<89VtC$ zQuu|3xsc9KGDdX5nZyHrVs9!%+FCyFb|vTDQ8xUMf1#+ojNp9_Ezs2G)LY|alEPIb zk;4uHa@GAzMrc}=^^JxbX0dSdCO-YJ#=-FO&+^+cn>W4IB_>Bq6=H8Bv!?P5C*Ps<|f$Q9Def^)p$G znPp-m1l=mvCJA9)H`TGKcj@)yS$a-o0zZePsLJqwNpdA3cV`!d6lJtQ}(OC(sam_emQ zp5zk{Uk4WO9C;Bty1jUY0KtZBt#+BhA|X|}6CeCyJE9>0z3_Z#%}C~?HB!wYN~wn) z`fae;9G`yggXEWHz#emG=~QoIGl-!Jy#cl$FhC7uG_CWTlBmvkoMFSO=|%6^GWrG| z-Hl43qljR>bHqh0Jd>hlS1MLK#1*&&Rpu!9oWAPRk1jVSN!R7H?zL)kePYjf02N2* z+#Dv8mVMZ~lr|U=Z9qZEp>}^7!oyxn1ZYVkft4;;QIn2>VEdQ;Dv`*Mix7kc<&H`u zRfTzI@045iOiKTF-cdEKjgoRNBXJH({&0h`@Ng#cml!)Stn}t|3#(ehwH?l1`&{Mc zARwI6)o3hcoSvLZfL-H`iqUdE{-R&%7HXRU-Rtw`}xm4@LAnM0+}SYqg6uJwTeKDIGcWMGdqy(%5e{m zpJe67eYo3sFY7n03md8qP*Eaw{fk!a<(_k9jrWtLDL@`v2H`dWCO zd#lU^D~JEV_^H+O1)TeLa=<-3<6D5zq8nwa_gzl4=!eCa9#hVo*z;6u4=cZ@eI%k> zL~j08tNc!QhRJ)sAgD)*s2D1#n1rcWDyb6iR4`wE z>r#NOG@uS9%0tgvoF2(%4UnIvsU^sDl*!2mQ%}Fwc!hwKtx=St5UUM=Vx~&g4{yxC zOzIsC`XgNSthqprw6+iu^+X|*=&wJpV=K3qS$i_or(92`OT>G} zxdimXl?;L{8C@(F`J&O;>=eh8YEYYj7ZwB)&_-i;jf!GFAOf1et)>Ofu^4gA8 zRN1Twwx0SEF^C>3tbb(6@Kf5>yN_n1FYXTXe&r@p+RnW=i`ZM0F&-IIOO~pezN$b- zY`DAO94c19G5$mVRzd;|gT`DrM~^$wI=rW+_KUlp4QZaGi7Qe_K937HiXZ1m5G=kq zG8;>bOq_a=s56|z=Sb_Btq_l;g0o!jAtVe#xi9)TZo1P0;Z!@>iWUAh%x%nak5b5C zsrf*_f)c$hAUZhPCLRlM57v4n!<WgPn>twM%^%`KDys9lOiF!hrD7_!`wv%C<1c?7{Z$#p9pU}_Ks zk;H(z95-)N_eYh=8(;qw;(0Blz(#KkolY~&@Ht!GO~x40;jb2<)6kM=NFetFq0KIJzyt5aD=04J(IIoxw%h2 z;;x(v%sBm8)>FDn2~$IgUKF3qGg6*lvYr4pG;4ku5@PPJAC3weTYbBHB5Jz~#SglA>+4#pAbPo z;#(n1)XPsl7v-O{P?L_9Trjc|S@Vy~VUSxTs@fE%$QLUECGQprr0#_h{R-SO6p{5V zAEIM9M|Cd(OV8(0`B5den*P2&phjnj`BFaMB4jaML0=GdyVDV!%l#?sg}^*UcZW`u zEjk`w0&jMFUW9n3o3$OEvc0GRo}&n6EAQ*i;gw-#>tRJ9qKwaRwm3QUY~56-NlqF? z)ek7pKZ4FGH-NCg0-1)tge#o`8N1hL&+gt)bq<77!S$pKy zKFyr!tmQeXS*eCGwAA3}V1WM`Ap!pdv;TdT^FP7N=8zfn7%q43zk=ES8r~RtChe?u zsfaHhf3DfkKuvesl9%Ff8QN`1t4;IG3?xIk9z4*^Jfz8hynE%z7sD;Sd21nfGLn?5sw3^T32#{n;?4Q&Mn9V-37v-!Up zR2QSbWLH!?o(G@Bb)Vb6<7o6;tBdC=ETHF;o~aJ1kUV&Ncz?0GER|Sh#I@<}fSm&C z;l%&=Z4>@n{SSaQW&S1qO*9Plr>~4diyC8 z^pTh3woT^o8*T0!w*_UqUsv>SKJUTh@pIL%EAUPs5#W%RZ_WdVD1pfA(E&VwbQt=0 zlY1Uz*@D1+r)&vQ55RK`Hc=FiDictmbS#9ouW=HA0en6Zg`ehBF5G(=ara~O{;RJ4 z?^(`;n>qT=z1H!q^I32)gXX~3X7p0tywUBxtV6I`7o}$MPJ!F{=?P1{?6YlLx&YJm z(_FNqrb9^c2>j0~TwT#G=kiYBu*y+!_Oq}`?vHNsT?fM8mVe!Xv&Kqn>U&? z%!M5$C0?y;ypyO~IZj!ydIpS;sF|Bdkf_ZxY+W+TqeLxT*MV!n)4hgW9|cVQDdnfH z^u~igaH2^+5X;7jRv+*FrPzTz+&Y!YkevqJK5*P~ z`E-~RQIOhr6f{n{TX9-tRs{grQ`WB?F!-?tD* z>?9C)>B}EWdO@#x$>RLi?M zF#xTgPO>}@$cm*1MD6>AAdk0ou){!dFp`%Ok_oZOaaxt`07_gvoJn|{qeF>(heaxt z?3dM5YXdcL3I{9@Pk4T-D3b%z~DCfNgL4){E_bUo7#e7ZcR~n zyib^w2H1@rkV+Wc_f7IDf?Mk1105t;*!5f=)khT=^702@HE#piu$`u1__#p3d~m!n ztRO9ts&oy_Ys06e&_JrOeUF))Vulo&*#ozBj|3(iVt=x}^VRl6(kVF*DHQ4>snz(j z*g?xNR;3djYUDOtKS?2aqRwoy?WJguj$}gsiouNFjDpkyK=Nx?>+@d5SH4Jm${dg) z8%fWLhmiB5C?uZj>dS$FZ>SlPS}~CnBoC`oeM=0kHIP$wbm%?j$4d8#l&IfBy~1;% z?@%2rqsBEdEa)F1ls|fi8U2)2K=<&N{ISuvg)wMCu2qUQlh6~EVFh=;B9tHJzvFVH zM8J&I4t91e?>o)CH}H^dncBC%pFU6D|JKG8X;1*8#FZ6s2X9c6X$|UYjm92E-K3>_wog!kZC3S^ux%+AiUW@thth*J&P~_aSH+;&co)=%1hW3mY{gSdVWoy*?k1 zPH9MTGx*H*;e1f(-O*rXJ(l%(4_roaJn7TxjqiKsecDViX_k4LeE#htMk(T$1<|F) zNaaz+x3Y23fX`oZHyCZSWem-jzA$944(e+k<%}6@Tzjo#KjCULY&Aai1^QZf$fGJf z7F-0_<*3oeUeyo1wce)jWbGwsrOl(9*wI$Gn8qvU)_mjGO%-46dHVmF zoC9s?KP7iGFT{7OJX6U~j|aTYJn_%GGRCC#I3OA&@59%INtFvLiLoLeCuPWH=}+*Xxg~pkaAg4@py+Es^gB~MU%?)yY`cne?VN~zDZemWybe-`Hp?| zK7?&2Tq$-c-p>#_+u(XPzb*B6kjJJfpswS@_N++%z6QI-{f)&IK^PPPOcVSRFHI!> zghhdQx5Ny_->rUS29s@!6>@(<(a9a0c)lT9SN%gIV{t@^EbO|inYH~OEVMa;;&(g-BnL zCt6fC4oOJA&5Ureya$tbZTWKtyn(Cyixf)wDFDt|NxBe71H(AR83tCt(rDOjZp8m& zJlxP}2^m?817|x%VM#tRkOcyHfpy0KYdw6laMeX90{N_F-0*j zE{>wXmKIZDuMChd9P-1Py}OK-93!6^85JHIGeX$JyfxAcJta`DXfk&UpmzVPPlyrw z*1nnL`K@E8{N=aKudVyvULOvMeee3Q;3JpK$rMY%bTYK42s*tLcxv31ujYUqIk+DaNFL@R1!SpoHbPwXOq zM}t0otZACK@##j|JNd$=gj<43>F6mRR>L3a2f;}JR_wHf&C|xu?^4198MDkB4Lxmt zn)5F4{k+8)Zn!WnqMBtU_CeQ6!6ZJ#n_2JSAOD{pY>N-F7wlAs$!BJcW4GZ(Uc~&T z5}49_^K~C~F7uUyLW>|VjFKAM;ZfbZ5hQC!K*{WG&=rFitlsHDFfLQ`-Cj+T{om5> zN!8o!1 zkY^45^Yb4&*4W5WyzvLjB=z!ao~%UmOha;$<|n}<6(=7r&HWaBI0-MgAS@0lA&@i| zE#IH~$r!mze+FGd6>*Z*a7)Xh-a-b^PhKZNAd5i4X2)bCr_3jCs=^XB3`i5lFdNkWiJ&!c8*#yOL52p?WI;bo<()#tk!z#-2 zVv=HSY??|%5}@3mvN-$^YkwjvTN}ooeM3x^78Z{f5XAIoB^`cGhW*QA9lAban(~jy z+FTLM*hUq)`J;OUG-yiwGVCy)Wmz+l9vo*B zQI!en{CjG%e@@VRWYntj_-Xxrq&EF-Pj~+J)TZvZ6WdAlutB-(vypMv>nAx=XzN^Y z)(KNtedl>`0s8!>h8Chgkj7NmuHJTnSsv<{-Ll|y@)Dz4uKSz^Uj;=zi|AWV%VjOy z%etZLM>Ou+YAUq<9rueSW&*uIUBtm7thp@M89IZz`JO-S(-a68Eh~$s?ug}Q<^^hB zQ9|7fw0Y5bW?KB3^fTEwSpXhNjwyDN{>chaNRT$rDDwLhs;iP)`5Fo6txz-UH=d3g z1b)2fe5c zwEUpEjGb2xguxc7hFBSRIy%a5HN2kNj&db^7Xp1;%d<635ye?=YN_iiKa1%s%k()A z5uYP;nn)ZT@ZQXToqE0)7zRb~P*b$@b8o?TT|E2jGEG{Gq9G2lFd4PSd2}@%3}4le zzh$Q1l{QU)1Awg;Rk?)@J@WX*t4ErL~OOmtL*E4VSbB*@P$73 z$b-1Wp7m0;(G~`XG&v!gqqnC^Mq)IA1e2O5%(ET5(J9DCn8M@c%|7uZ5#`p;l+eWz zE*|}c_=6sXEJQ7rd{5;~6dGpy$lLkBWho*fBQJu3YG@H2uO*jb#im)rq`*iMXWOf7 zWW~gXtul0+7rV~R+;2(B6c^^G?1Gz<%%qcx)2QlDm6SBYJ1T5if<|@4gi58y+Hg(T zS~*{E_?+za4?^-$>-vLxzb3H-EYcvu9S!!2sW_u%gUXs6?QM^-1B(B)fRyqtx6c0$ z-0FyWO!Jpp|6}RE|Cw9=y`lJjwRC`z0n4f1`gKCab@tJv^%OUx%?FiW`JYP%#@gyW zbz{MlTn6p+>wR$yV)oSrrg08$n>07O5> z3V~4v6ouu}1y4nI{jqk?H~-GBs zCe9qdjIi=0!&Xzx!~3-3>;d&BBKGYYZZ{+XJE(!ugZdRU#ZL{@SzJ3uw7>;uTD1yX zGV*SLpg7@WUfZ4Sf6|U&&g@^P0cj*qWM^-9`=R(FR_K;Itz) zlXdB^*OP=?E)pr$-eL2~QG6FxsiypT=wmi=(pR(~_0!nC0o}gKSkr1A1#HAt8AzmXt%d<&X7fI(5T|JZ*xb&ZBTAgG6&=Zl-OoO zO5xWY;$0@Iym=Rz)S@GF@^0h`gXPy|dWYy_mcxgygIQPt3th(iQFz%zu+=Z|jth^Y z6G1vJewL@`;mrU;8A-5}F*sMaVwoT(hp+O3g3AK<@zMWVKdpTN~d3h0sW z$^D&P|@iu;`_)a)+6Z^ z__@vsA*DdPy8a5zD}qz-_qOe@pKS{zt0H@4Vb$>dP6#iOT-y_~D3zDz1G$5F)A<5bic?i%%5A%8q-mpMUJjrU>z) zatQSoKj=+siLEGr2{-_=OBs9g<9Y5*--zn@_`!&Fs=uU&8f@dHE<)XYn2+=kKXuwv z-nmzBH_{0FnL=D3kE^+<8#NO6;qt=Wm10Iddd&TteMMWf$lyvX|GQ?%0#vmKqePtT zYaZu&R7gG}9}Qx#c_mmA@#2GthFa5%$3iuam=t#B!2M zQ&cz+Fl{=mn_qs+nW&Y(;$Bd}k)utT>*7gsYcw$4%w*ij3TtSYWu@@;w23Zfh8AA1 zjqa_uiDPKhz;cU_(-j}Rr{k6>o8fn`jb?$#&lxJt$gx*^7qg%Wd66gC(DAsDB*HoI za;BT=4qv7eUY*tsM{y_+_6Sb(GGK;b4p@;|^k;&JT(B>LEW!Z3EwpfA=_3;)HaEmb zsb!rZ#b)Hq{3Sabm>_7r$()q&z_p_eV z9$zz%LhiCYe{z*pA}`xV;5jci?0c|}qrrK5^fL36SF&qg&U!f_h<4ns9!Vm01NmZ$ zi==uYsX%v_u3EmILg^3JzQ`t+b*Ie~hv@QkGOnG*=<(l8Rp&1lignJYW@@%^m5Z<- zSx@poKj78X=fIb5JAIOgLynN0C44|?8-^wCoWribF%j) zr^=IX^M313XZh&0q=y|BO=7sziVIrrsa-210RIR3qBK7#8Tns(xce}S_fHSUnt}h- z!`Gg~{X-PAawkFh@;`gHpfKoVilrM2~v6Lk3GN}ThYz{BKYrT3Ws6To8Coh{gC6rQ1 zxAJOd$(-rtWYjqjrp2V2L3WG;9Jl~BWR49loxdNOGg>c4F7%8{Z6NG62 z3_w-|#Q;0+d*#jqs{pW_Z;#){@qsZI0tB^80HCgKO?qQP$tW7Rg$fXZe54lWHpZJj zv{a7O0=(^zS{8I9Mnxw*n1QcK6V56ETxq7lr?E&KcK1XYWN>l{TVTh$!{ZfigTT`h z=8vR%Ns(@8cwW6#&Gfrs&!y^Tjxm|c)~SM_<~|!tDK-Ly8EG!bmjoh=?+w4YVLrvq zk8%bUYqOq?G%^|LokV()D5;8W>7qORAHzIVIfN6Sax+>~#*!P$e}3BHni@w>IJ%Ja zXF+N07y!eAQN&G|mq*~aFS$W6E*hJZ0+gotI0a`a28Oe}-wV!=dF8ukt`s(J?> zY_i?dtCgM?Zpv-~6m4p8X>f?A#30ApZ#s!oEx-K6S&IzA?qx9`+|2E5uD8#%nWx8~8!F#4aY@+Vl!zr(Qop5l?!mZT*rq7*CkyGC+K-If z90CsVq?U-xsDKp9cEatXg%}4apbDdd4AHv@7`jTBE`ErqWH(XF(6xA}a)|ZbZj!Xz ziWwdIW1N+=pa;yD{<(Qe-a$}`z$8DCq7?afkS;sHqkR&kaQs+UB*_vYL#L@&`8WwS zo7N*Z0_woaiir#hLH5R3`csRn30lT58M+5_8rrgMPgvtS98~x`%PBzmXp8gR)ql2UZUs4x zNA(^XWT870vO3mME<%JKDOJGk&@G|Zm#M4ex`adg5fD5-wN?3qx^dqS^AvQ#W5?6w z{UlzIWwk6B|IN%jW`=nRpb+f`RzZIIo)_IkNKwkk=*MLoq3KZsu)_+gR|OcNx&qMj z8x-$ut>9?E+6h&A#dinRs+;s0pwg9A&cQ_qeTy}jZICQlB3!A^tI|p*6@I5rcX7d2 z{_4>4TGqV{_02{SsZHv(ZZMg72>_k5Tmgfu*iF9(Ba2wCl)>+sZiKo&PZs~iT23Hi z#Wyjx{SFGfE2nW<&}ViSkie5m?yg$VSigsExio!49L6EZCN&4HZ_C6@z4RyVgS#sk z#Bgq|V?YMN8e@W`Rsd$R#_bL}^F5!6VaN#UZX)`20JV?a-cutX!cCee1Ya*HQ7tE; zyXyBAhq5aHOkrP!RZa#-Bv~iP=f%_CVu49GaB^dJn=M9-&V&+4sbYA3_;fxHo{M?A z*~ZBK!UzzmHh8wl`2~-K=!g!B^ z<1)m$AxU}(C2#6+4!1TQ``Mad{g}#9(ONdt&UJ0|1 zNDQ+smh0JaUknV6VJ56ZjG6o3kpnNP_$ks??-BRsqh2MxrK)t@PJCQgUAFL!r;=BF z02X|^?Cvtmq;H6AYB%}9$qMpj->|6hUb_9sD*JTb6NRb0teBGz7{>llUE}?n(vy#3 z6qEjO8&y+I-jgNCoBfjpSMt6*DN8=Mulw_)bFzFob~%7GYa`+`ECIG8qb-`MC-Ud{ zZGL^83yhwl?!06Usd&{dk~H6ub=lSAtvHT+7Qj1(#$^0%kT>~xcIs5DE7$R(^jOf! zY4&096aSY|%`bWLG7r!}TKi<6hL2}+%f4z>eADvf0I8#(f>(u^W=h;l! zQj9vB>AX@bW6rQFX^7Tv*;BM@kcm)Kj&dUe zue%^a7|}l0+719s+fV?f_tr`;7o9+D3M88Ykmc)cV!S)LAPtvL1O@^u>&6`<*H|H1 zm*L2+xQNlXNS_q&E<(f`v1y_B2p0||Mv1L+mG(wccHNbajL$$lg(-#1we{|-p^g_j zBl&2UOY5%|h=paRnv{-)DukrkX{TN6buIKkryX6b>2vK64Uu+Dy;+~ex{!9`Tbcx0 z`aPNGPTO?iZ2CfepzUm0q)2qQDC&R)WN4n4u5L-4lBThrp`{_@wT?>U_ca*wpgwX3 z?=p6-aJ}1*T#lp#FQfc>kd^hpR5R386Z!Iho1C6H{&X!}6i)7&kC>*E*Z`v${3?*yVha2Y4 z1ca&E<2>%h@(y|?CAj|b@|XJ@uZ`itiwSsekkJ3kQH7+9XXE|NVEmP_$TdRf* z0uZ80nXb2gOap%Yv&q8egnDU!BZRrfl<`p7V~IiqI~kH$OLHKSV2oEea=y{N^Frym zG7>Oue1d2?-BcR0BDPrulIG>Z#%m!0w@C$lH>v0}R=AF$kaR_;6JL*;&#@KYGr&=3 zC6;>!bEd>x6}ju1S&hXWhS5j6Uo~T_sDiOZR^H5|`;I|>cX2NXt%$G34ie~2XBg1~(afGqi2@rz?NMUu{)wLM!v~|t$j%uM>IWRm4G8V@4Qz=Jh7u&l#Kc*B%8nLik@b6`QHXD{V@ScCE}6806~T;*=GC*=#Ez&vM-Pc|OUxE1icc zyuvJAYpluY*WFqhm?NXAqk%&6%S4^+m$gwxZq7wVkRC(uO8Ls;B$Oup1!Ct>+$l!w z`hbn_$T$oHfWZ8Jsd_FZ98Pk8Nf5xa+4vS`*O!tp!dL;1e?-7y8n9S0dH+Zah;t6Q zc;iw~`|DAnu*K`Lp9fqSD1U0s)H9{h^>`g#6w9OVVA? zp9e=Kf)?cDF)jo^SR+F1!qNI8L)PVRqY;JoMCj*yx(LF8CU3;%`B5&&J5b7U63Ze@ z#s#F7x>E#hL!Kka^n3D38sjxvr|W;~dNx*Qv|^Zl>hPzpCcPv@4JJ0M`s<}0+Y*Ej z4bRP$S0gJ!x}L#^LTif!e3`;Oy()8TInnm1-@}gurPU*B+_D9|eCu$ApLFx|4sl|F z`3n1=mWnUUE7)U>*NSg>yFoX&vK8jigd-+K8p;K>PgcuR(oY|6R6SZ05X)NrTgc=(_#O8_)4dAw)FM`*U8;gLFx#7Dow6kGPBL- z_TA;2x0y0+Ww*nV-#`_2$Ic!tnUB0;WH;5xoL+l-PrqCt_<*H1e3lv#WEzNLYt+>F zd%Gks>W-w+`ZP&;C6$4}Qp9saMmc3b7k-~x6GbnT!*UtY+C}z~j)$2X; zv-?)EF!|3Im@e=1Uv-LX@na!>#lRTWE5G`W7?}0qNe8={+=EGP>_p>|iLVZDtHp&H zGD$GLe)z2dX0*W*&^VLa6Y92=F`-1iIg>ul zD}KII*DBRwMCs-2)B{(YegiA*P{#6;gvxR`I&&dym0duX1Bq+4bLIf7TZMASWQzcB zhrYAzex0D%uxb#KULuUCv{NRP|JR0j2E_eVO2DYq-uS`;;fPy4=S3a43>Fdbot;p7 z|1V(he~@2{04*^4zZQJv(;NRA`6Xp6=W}*-IRBCPpMrl*5D6pr{fqqa6bUQZ5F}LK zbDgvxm9=yArsv-!NO!x=5-qCz-u^_g50+6E3pkA3ZCS)YbEAaC=3CzMLf^n!bXnRQ z?=v?;xfVJ%$<0@(N`(wi?9xJUj}cBMFM^EYG6dB!f871NbdM@X|6NO1lk-$XzW8H3 zPX5<34Hi2rsjSTK6P&|!Nz#cedWjOwQw~BN#A&tH zfyNK+egCrX7y0$z{?XU}Aip+dnN6l2{!Mi-#K@ZsMTWk0h+Qh-s*n$QX05zWnEpxu!bUwYS#|EC%P9K)s8UqXSB6`#V{DA)PRLBT-OlBw!%CR^;Eu_m_~cA zE1%9b6)cKt(tD78_|@4>{iJ2H=0-~tv_1(pDC%MIfT-VhTHItoO1IaBbcuxHI;`a{ z&oo*a3$<1(cyU!qy_Q$9|CfS9LM-=F-NXuN^FJ#{f_mYA3{Np-^d&>@5mZvzrvz@BZ&vX~|)44{A6N@;5|JCzm_z9OtmQ=3H+M^t-=xyU`D}gTCfv?@KXCIKC z)ivLsOqX?k+Ds>q&?AanA}gyOKCtN+&$o>H+5#<#vpYSEyAOjww3x`2ETu$t9z~A7 z@@mL9;6&8bF;RGCN{!vC!8KXWaD(5=LD+>h`Sir1Dp^#d0vSu&c)W5$_)_0W20OTe zFXJ-*5x`Yiz-Y;A$T>SyorOznP_!Yd^|)+;3# z@^BL2p+&4!B!TV*70SXl+S&rQ=+9NlYY_FgY>^ij{pxokLO??mLaV8zt%+1vc}iKV z#(jljP4hbrg{VXN|56r<>j7Ks5Pn@m$0$RIf2Zr2KU2wL`8Sf?*5b2B?0pU3CF#XO zv6dcMZBSsZkKLP@-sd^L_zh~&c`xpN(7*)V+q}M!#b!9LpY`x$=gU_qE5l`dsYk_1yk&Z!`XrIf%7J`t zi~;S^^+H<2XrI}6DQp7ESY+7s%q&65=_QTB9>YWAKvmC-*8ShNC#cyeyEqVoMwQKmxno!u^m@tGzC>FCm1SXp({EUU2$V_YHxeiddyThL8X6+^rBpK zODC>6*XfNY!;Hg-#GR%6agN*7v61GIW4wGzM$@sKAe|P;lQ^5*lm7QIT&@61Z+GWO zBwVPH%*68N?-#X-bk;j|%rF@Ne_D;c7FuT*JgxS%^mo=y?_}IfHiH&5UK@YRtXOjO zO#JEUV+6KBOjC5o6!gt27gSfr0-c;G@k*4kQ<_n97@Vh^swX>Qcj60uwVCAc=|RL@ zd5%1C1}v!(aeX)O>Pk9HuSkl0UL*mv^gKgZv4QP1NlKP`g+kFke$1s&k*HQ>nJgaJ zEfDXCKbNHiI<`=6!~Y=Y>2ol3ryF4NU3qH0w=bw-O? zC8w0=naV>yb!O8yu1X5SQISw)C-^1NSw9wvAt-k9Hu&=ksh(~V{@%Hp618&nt2wH!r`j2eHkY}hw% zQmKw-W8b}TRcsri_EP-p%hi`Tw}>!00$HjgSXOY^Qqw0>91{JhcihAi(7kzKoP>YJ+nvy(`_2$@X0Vczjo2_>TIFR%72(N!I*8Mn zFVd!4V66L)$$CG0{cPhUETZdp(hVU44Ye!iK$9_z}e!K|hI5D!USIT0tP93zb^@Fii9K>(s8-$#Ace(oXd=kRJ7FeK35lqufBTUa3z z8PW1nppy4Hr3ZtzObDITu7p)zqW>KCFD9-Hu$1I@|jwOWiYR zO?=n1(#aoZ*~c?Gyu0C-e??y-6yPsh?4n<2d@VlriJ4X=Z_AKVfA+{%L1Exx*0R55 z%D0|_v+wCsF%WewY5%Y|1JqkR8d_ryLLd^cUD8u^6vP7P;c>m}0`$sWzgx`jT zd)}s5n(%U|iQYTT4c^V~CZqFr=QADehZ>shZoPX%5;So7)7I^81+QJN`Evg2$LXBg z@lIljqL3bYBG9zXs3A8(CTEq?_ls>mr;>VqvGsX?S6hoKYodi;m~gt zg1^6Ayj0KP^i>@L7r>1rzYpl)Ie2l2ULJ3H8EkgEcxk;hB{wLf!^RT59b3;KjkirL z#od%rd=n;Vd>*PHqEV3+PL*GfpA0@_X0zEQd*4(UNxE8OaL$MyC1@<^p*#j!MhNj_AA!!e9ZI zmY-Rp3$=iabT~D$p-JmF#fnq~nH~3ub91VD&IaXP{&I)>a}$U1l1znYA>Qe8MFeCt zezB+Qa&C41tra&Gxa!>l4Y=&w3s$L-bavI>-BRn-dLwa{tZYiaet1&KT`2Y`x;Tj9 z(6MumX7@|;pU$7N-LuacTlrx9`d_}CzBf2##^VbB1wi`0{v1#ny4UWm$`^#Ye;#iB zEh5)$*LIj_9T79J-s!)j7enSak~~U6Z9gpCf95$GW7l@sG3n03hfFiTfQ=&10%CCa zB{XGtGrz-F@!iMH`_dAV?d{a2O9%5FzcrJesD}9;fT$Ih(?)gFmM<^rhi%Kd)D->l zzvqme4Wcqc|I%jp?MB%ie?#c1`SyJJ9diH(dU02~>@RuwC`E+l_@fv9Ny|$LC$4K4 z;$!qzwA3Izj!-E1-a7YU=Fh+fK)d_`U-wQMYK$19ZY9mYO zi_G+vX2kuGb0Bpjcv0N6leSI^i3BaRnLqi=^Z6Q!)v*8jVg^+k(A zr|ACc9x|I3DZh``@5I>#(H{i-_(fZxwfw3UjMyche%uD5Y>{xBhYnpvg+`CIbeohZ zhGOB&gjpqPb7-keRXQA=d=&cewvA;~h96?rF<5LFj5Yu}=6WegFG_9)aCzXdXo5ge z?6wC}_!~otX#?9=qgodI-3oPieoorPc3-v(SLzTiM%LF=JU#7CS-M8GV+N|2=_8*? z42wOn$(YT!)2GW;FoPgQt_S!E0UXmlPh_+9SQis;Qtvk8&59A`gadbQ{3~oFOD9Of zz=Mjlo$#6Z3;@P}sp!)1##Wsc`V|@sjZe?p=j}K&Kj*$rup*Gmc8sdHYwt8*Amk8u zlG5j5yWr59Hpe&Y(gewD;^;p66L7>ecQ)px<)L4;K{3>>uEuR$lNP`ikX$t@XlkCl z_w5Ie?y;3z{Yc|9y$mQRW`gVZ%0wIwu5h5NYQ!y~+gX^w*2ss<93($%qWXnx>L4E_ zxl2i+wFSA5OFpfy*2p`bCnKENhwgMn5mKIoIqcb2E+Go1hp$zJ+)>1-Gb<1r%MEW) zyXPvk2Gc%O;V+`WImt|fP0sMtlury#T?t&7^mkfewe}&aOtn7~M3R|gV!712e1UUk zaK?1!#Zb{3<1FzOHp$JNcL1G4@v>Fc+wAstW|Z1{7gHJ-F$9*r-5VyrH(;;(maI`2 zOUX}hHXRfjB=K(l^(xwFX6M4&MTL&jR-U`S64sd5?UXNd|JoOpef?R4`9W`Hz(-C@ zE(rj0^^vKq2}t&tv@@22wBuCHiOr_)GcaOGt_6XKozC@a)#DF_7?Xd%}lB zNzK=;ZGw(ar&EddXP+G1xQZjtS^{s?AmqLVjftlhtS>XLG4=kAWCnvsYh9tb{niz> zT@XD!`sZ+vxzAuJ1t5P87BCQ|syKg`Yop~0&Rxe5)QA)M0392qN4l7Dkv~55AigiB z6aexfoOUAZqDjE>AHq?|C8f_Uwn6uPOfy}&QLslxs7mZtgZW}_AY1cV;aE{rY`!xo| z4y*LMg*2%DDSK6<=cm`IuW6f|WPmL{7Qnru{CxGOW%N)4W6BWPUCq6{E3fqAZB}TI zApE0ggl5%S=S@;P+5@A_wfs9Ux)%)FyRHCxx{8AXIItd|NUkM8;<+xO^nk>_h~Xp} z*U?Qc{2;|~H=S!~RS&=o-qL8RFQ4EgwBNAVk%hA~Usyb%{Lt=t{HY*<-A_w3+aO#( zH9Jb4(-7S8bo>f+D2bdb*!23a-S;}FR3b!faz5YQ?ld&}J@v*p(P+)Ua4wC6+~R=K zlnWZFd_q3%x_o9VCC=jC3+Z@F#3={Krz@e){N1^bKEIwp>avXS{07r(p8oOmu6Mn6mZb^Hcz zDBx7JUIRSBTA)s!_CMrVXz=?Ezoe?Ud>Y%Zq5B@dj8H^I+u?h;?@!il>mZTm_aEd^ zB}+R9a3EE8e)2glt-L$jrs>??IZF+gV?ZD=YKRyC5=p?}5O3Pyw0&^8V*-6DfRu|c z>7dTRz!3_3sg*>Bj>w!D)zeq#X=~&*5CV5)e58ZiCPc>^p+?&|5U=9+kE0#jgla|n z74a94l~LQhDA`fVa~n)vtY!+ZKXpToMx8-zorSQQpuWN6EfZhPh6+t#=hz4%FDII$ z8p33l=~snW=VGx%Qu@cJySfZ_P)tUV;r0trc_`*kHu{!XhEpKo8-I9!tyivDGMPm8 zkpP=DWAtlCifWLwN~Nor5!|ngNwiJXqRer|!eC?44Bt%|OP1puW>d59LT7qWb2tV) zBs?S~DY7pu1DwW1q#HA3JTy(qsYuR6TGoxG@$#pA@Jb;g!@CsE#4ZpK_2O4MlJAd- z&4OS}mBCHf8MO7BC!qMpc;2<%*f%(a=eC(krkO9q{%p=dEXNr~E0?A<5K|kO)#~b2 zP=X=0`Jh0Ts!F-vPYH zwhNP22xU23$bxgQoHHX8?rEZwcmrHaR*qxXl(<4HF5B#Bw$5kgbdU-M4Gpz&NGLP( z7;DXAJ8>+>uT*akNAB%CG=vgCu&>CyxR@!=!SNFnq|X$jr5B_P*Eq2DRO8_I^%U9K zo2Tfl;WU(OnsyoLs`iT}+v+nDlFjR+FF!~9tO>lpRD;taEzBXZz^}i+f3YC&M?o-0 zVW?7J_@%;#sKUtN!s!0O*u}!@KMDyPMe#~SiI<9!ql!|Ci_-gxG8T)neiV^7igT2T z^DY$^L=_hm7nk%Gmn{~T|0vFajX#clXv8UsDdP!t;HXccWN3 zJruvr0BecnKVQa#>+%0eV*$zDem(>o6x$v37ZxOYQT({vnKR%KFrIE$L=s zX;v!A0i%~~uq6~EAY|#UyU_}ErRDxvr(nit07DZo^pNa2^@WngOR6B8 zrc8XpOAp4pbylU0@WYwwehqDPilzFM4fizIw*fSl?X}a_H{A;)o#tXw^`jJscLG9~ ze{F_u(eUPwb>T|>)AC8+7@p7S?C}jTebW|tYi??_`iqsy>50@G1npxGDT)gK*i9br{_oW`nBegr^T4x!Y8w*8ck`k?H+y8WsXrUe# ziTzE>uYSeA{Q`HWCm5tM!lh{}UxbYu$XKi|xs+J(qX>(yn2UL^L|gPgqU1Lf`@d5Z zsQ^jLpS}3eKTS_G{~c0C5CJfx<O6h>0tK|7wCwz6FLwS5QpZj>Q`Lg1ZNIH**{gT?v26YYsk_ksL~){ zhW%|X(s7%ane*((`4(+z8(yQIYqrE2#JgOinsSv!pZZTroXX7~=la4FhMR4qFhbuw zoDSZIZ=9{Fx+3A6C4Kh;XjDw(+i&*HnHs&~%c7Y!;@Znp0>YF{Iz^v=RFA6229 zb&2hg`7iomzC=uqN1l_{td?0C%&T@;O`=YLkih?>0gcoHm<^?$S9)Tkl^95PJRHE* z+yL_jld@$+;MPnGsK$@EXZcNU*up(^N%_3ILazWg?tkEYp1|DyEAQj2ZoKm!A?Hn8 z1<7?kysA6|#>L9a6}!i(y_7e_YICt6o*LN#-T}K=4YQEnyy10;Osrrj9 zxFMc6?tNqD%XibJg>-+vP}ks#C9{0Jc!7*XC^$epByuF07SkRCe50aPqyXD0_5RP= zr!LyFb(syUKAW4N;dUi3TgO$(JvP2^v12vr_A2n$dmt*-ZGNzqbg`qv+YGkS$T~*SB zdx<=z+@qYo8<<3xm?RCm-Fawm>3Z3#B;+DgomYu)zRiqQFVVxGs79~$1e|fpQeO(f zI&S_cIib#SbjSQ3M8C2F>)()iTm8+Re?jV@{LLq||3{+#ACUUL?!EjsqF--1`t`T+ z<4Ms^9%J?s!;7*>KL=b%i5z91q6vq)ThHr`8x7)0WJ$Cq95Otw$G0#$MQBf9t<2(C zODD@>p7XT`!SqRzEOjO5L9T|y_*{WsSk0NtMniA?=OwQ#>z?*E-j4tL@IM<=82{AU z+5Z8D^g+V@qA)oj5Ey%VI_o(f!E4CA+MVe>?KLoVBtA1l5@gPkQw(-Y|Nc(r z@nzJY(=DwI%O(-tr^7||~@n>3utsj&dc``EY&AqKXYg4M9SyP8^^7i&Q z3@?_=iUdna!`fT&I zO!>Rbr@U=`BpFO+iPNHj$nAPp#L3Q(nP>q50WXp-#02Prm|)T!)XbB%MGt**t8d67 z2zs|v6M=Hl+x;<+E_X#!6SSpi6Ga_MnX54_O@v{XYISQ20v+03fL8)ilDE}5QhtwTelBs*@*AU2REsV%0i|&K0q&QUAJw#(eelx^xdEj8oNKC!ff(} z@Q|ggHgFijtTdx?r`?YXh6-+NQVo^A33)s$!?iyttY5nMZ*$8gimoi4C9~VFdG4{C zZtOHy8v`kH$k^G_wd1Kszz7+zBu4w-TmJz+|H)DQh{C5B$@I(h4x3sxZ}KAE|MOw^tQX%$xV9eQ_`rSoz^(CLp2K}fs7!n9;Hjw8BARsG$x zn(5JdTFbcCU*4QGOZn3;EKYO>_SGxU)SsQzS9QtMVEFal8k}>7g>@HDNOPX*iY&+~ z!jmTvSBtX%nB(gX7pKL<&~woq^QE6M?hwxjnj%$?aVUfq5sY1#ZF!3G@ElFd?$^dr zV7pm1=sgx4EjK2U5a?g__?~uMbiJ51Hc$ghNv%4|$*v7_C#3fryRp=7w1{?O&>%TVyvj1SM;r zycXN5j2z)L1rfgz8znJQRA@q(k*uh)NyAjWQ|&Ei;>4+O?xbcxXWDVN6_vG~`9sO9 zY)T|kCUkR*-W8kA7CBPDyV)eL+u3L zRaUYs4M`G{%Mw;^>20aohuG|0Z&~Mq~@g3+tej6I8M8^iIbd&T}jtf z*g`?jdKK^4bI}v~d`}pVPL{HjAWitD4MT}q-tMdG4$TrcDXEz@!-bpb6+OuTs-Nznc{a+EO{*R-VraA@Mj+Nx|k z%43lXnv25P5d_dh7iwx3_=M;?H$|B{G4|J|k`r4vQ5SU?hj6#o8d5>DRD# zcb7HNe|^eBE9Ku-On7#E9)Ol54h6~?V^PxEE>v)Aq$J;yv%!j6Px9+kS< z_)?R-+jyAoUf@{s$x`#^46d-(-q=aRKTB(4Dj&dV5QD4gWUJ_I|4e*l>CG7bnmz+J z{D#*sNXe(Jgt^E-a_h`PrB@SLtk+{4MNLf*u`5nx#O@31WeZOTJ(ZbPn}2)S5P!Tt zF@-#=*V-Zb=qinZ{32dF_%aN>U!wIfG9M!HJ{Y0mRH7=;drk{Sw-~c-1$LHy;rE`+ z28zYYAY}dH(WULbw`?9z12xM0Ed?52FOw6iucwDKkvChc9{o%ea^);deRxoR_S#=& zpoK!DbdR-S1bGR9O3zbVxR?F_-Jq(>!yUxGQgjPlgY6JJST?+ z1Ji0C!Bl0P5}rzuLC3>lB34pKFpi+g)VxSkL1Y?9CXEdQtHCk6CUdaNrj?7a7pJ1e zFudPg7_Wq|4Wr^4@d-C+Kow&2A4!mG!i}#`y5VeM`G)72j`TH|zfsM0RAy0nEq|(L z3SG?%LwgZ-{zgVkWm?`4Ln@BrI|;@4DnlqWGn79K4WgT?5O~3UqcS!9V=B74!5vwV zhRUJ)KxPns6~-n5=iJXCDyIH_tiAPHlx@2$g_qC*cU4N9j-Nl7b1 zHw+C5(k{KXMAzji2BB^iyAEd+ITZbk2K`DI(qiIW$x055NyBh>e zR>?<^K!jgm&AA9^%T(NjrC@g_k~7QL_*GLDu62?n+b9nUQ+~Ckpy*TOrBW5|rz-oU zqO(%f+EO*1h?KA>}1v#4xB0OXP(`BVaQTnI`3#>1|0kVb}{o;1DdM*Ec;U zE34>HvcPP5;r-_oZO;W~vuF~;>A-=kEi{P>^VF(M&1vZz9zD9KWT^;9cm5I3YGgq*h?p-fxzWLNWOWFk-87hzTn2vkzKJ%^Z6gm2W ziG!#OEYvpG#wWv>H4aa%I5;<1XIfnMy7$=K_|#q#1yfHEV@w|u0YJoL*(f)w z9w0;jjZHq3>-4{VX0~veyE&XGm@>!Sc5|fGh3_FtXZZ0wh$wSFLaSlV){KdG)l0$p zo!Q`2ixhE}7tN2cZBIO#I{G#~ej2!W$}()|^Lb*hFu-|_$2QT{7!DKRqYKKf9x>{_JvL)le-kWAH37`1$$Sr;t_` z(O{DdQUW$3)65d;(Ag9FF*>zN&oQAhUuZ!To8c^T#I8;w1^{Mq!vKXD4%XsCS#UU5 zu;2jU9y+=&XvzF?+x*_V%{wdB*n5<2%4z3< zKag?CxBXCHy^u$o>6#3}Md?a+P&iu9k3xHI>n;ZrLCIy$28AYz9O^mka`a-F+_S?~ z&|}!wf{ciSGvXJC57tVvNQ&soy^N#>!)S@Kt+8Qq*%x+J6J!Xfnk?Xq^wb zq1>+ydPpR84_wGq1P*)QM$UgJiVypxsVi9DA)9nR;1CNC#X+8%8ubahLYyAM>NzCG zBS|+(3sB7z)MIsS zCkFyhg^E3yy74OoUo0mU6Hw4%^nq};W`+vg`dXdB27A9{VFelXkct^miWKvbwNwV(~3S(L_ zFY=m&ju5<*B*GYs(6jT6&_oT0PTXjx3NG;aspuXUd{-EU07r@Kc5XLoG@;2_HE75Q|kILoj^ zre}u)AXz9B7z5+*u3!0C^o5+7>)@GCZ zW#mm$}FNDEfpyS zPL?&;DCevGNiv&Wy~``SCwIbqlinbqd)##d4Re@jaa(V8zSz&ucr9{xrDu4utBi$Y z3^8l8)~aV}+s%uaOxxUhi{+rA(eUAdl<7)<%Ku2)-eHijEK4W+YsJ@L{Yb_X5#e0b z?YWfa-i}+jV7!#&(Ev8b60MVhYW|e{2PpKm9Qq@L+y0kvjxG(2$rBq}hOFZb=FZn= zI9&P8by|gpR2Vw?w)FM;t2Wn3Q6ch&Pd%%~6i+XoR*#CeT%Q8elxfXjTep_=YY6QU zWk;U&F4+N|u@z9wz1gtC=I|m;7I(J{IWj zf8Qn-gTK9z8$#n)=Fo0+3&x8_;lCB{%>aK?NWy&J@Hk{H871HThsxK{<%8uCFlqM{ zk2Qr%Nx_(c`AWc?*X~$X*QZ1Ua)Dja&kJCG!Z=T7DV=3{=H;Dhzxe{Y`S*o?4mFI* z{~Y3OmzSUfNZZHmH4{SqQ}7Pc{?_)H2KNO#efm)gG&Sm_Mp69H>2W6+{Dac#B0D08 zzIC7NBro2g(*bbp+i;D8x7@)uGwr!2%H72*?gzj>n)oo{d%pMi5AZdXcu7+oaeZ`* z@+9i))OvgB~%)p36<=gJ{gZR6Qe5+D#K#X7&;$FwmAa{_` z2kv*SwG{Z;?&>u?K1!4TRpj`uv*pO#Fp%EtVmP;f{;Cq^SD)@@e^~sdrl@OtYDkFy zVCw44E_t^s;CsmB=QEmIu`ufT3T7aOUh*es?=w&mssLY9%0SRf%=&engKzPFPeI8i zG8*_t3QAl)$-k2N2aza}0@!urA5#A?3awEFTy!|lVEv}I13;Lb({9%IcNFmGFz`Hvahlv=H zA_RfzfIP#|KTF)vbVC)9FCIkQh5PmfUG5l0%LB?FqUEJ>dDQ5r{6b~I{7(QS2|@ME zN#oVGtFz>MARes1PZjkdxfQs=>}slaQI)y16Tm!Ao_0z_<3+JhCK#7r5b~tMAI@*w z?O|X0z#R~4uxrjO)3fjIKn~~?5bAtw{QbuGTMOTt_P;8*h62yR7`4V(4x|_Fe*Do& z(CZpm_JdY>KmS@Q@YX0)!-<>x^^ehj%L4QIn>cjm^&=X+C-Is%jqYz)i|-FycPq2*W~jpc8|dXe(#IU-lXwAJu?v;{=JbC>7|^ z#;5>=Sit?qco+ny&aPtfLy9DXB|93eD;ky>_mQihsn$SI z7%|@Ief`T`4kw>u?9G@8pJUIGBi)v96ht z+yG%*eQc4Yn}=1%hc95c##G+X6GWn7l;uJC$E?P;Lh&DAF@(fwX5xZ;j zdRRhy2436TtSn-2#`d#^bqYTbG8p0@g{IQU@z}EwOb1Bk4@L^uwuuNkI+J@#a=jDK z7m0OOVJ32dK<6DIxe&O7h7HKz4X;#3XPq}ewOpQZC2H|4hcn&)Fg^FPd}~?{DQNNZ z#vpDs0##cdPi!L<&csxVNt9jzzu=^ZvP{il6ibsS)gBb2fu=;>T#Z9OVYj*t(_GLz zH;7Mhk5pro9l%BM`_z%K5kmpN`Y*ZwJ>z28P>7m((xfg@Ea!cr78U|j4DyapuS3Ls zm&dkqLFiz!J*cgp@=8^H9Wq=4&UndjApn~p7pb*S7Dbq21N;qH9JDdb*a%gYDYH0u zN=hMX9yWC@s$?JKVTE=MU+jNf0fLA|854(V>(&rKys1yr5)h*2Z1yKjG$+5_UKc3# zG4i%54C!FU1p;D6X`tOg)Uf~cx{{XY1I{wIT+IK&%U(EFcvuHeBte~&PoSihS_ z_~8F0!qoDQ%KfxFsor6-!fd#?YT@_6Ka}&ok1$<-hvoRs2-C@v(zbt%Fdc1g+L@`f z8Gfnx`kxV|qpw@`KjA`&S+zP^57!4${)#X?{xV+pC&Kh!T%2kDba75rTKv<+`S%FZ ze{pegm92!3h;OYp$qAp0sgM3Tf&sec-ohft|I@{ZOsAWrCT+4Vuzk3sz~+r_s}d&1 zWPYaR^JXeDCt`7PAV(JHRk|Ld=-OwKgm%5P@#It>Yr{iJ*C)|O<<9SeAK$&-=@1kMe)qjw3D!*i0=(nNI$2tuv)kQ~q0S zXTVsYg3lHCnd|24j4q_-&!kKnVy`V;%h1>Uo>bvIh(kv}Q)wEWJS&Ouub-#DrWV4B z&Y*~iGVT?>i?-8G!3xwFB#7;4pO>K*Y&=$4uX#gXKHC+^G!FxtNvuXP=QJaNPg}*; z!ZD5DwHPtH0ZUn!Vt#=iPrU<&3Xx+OD~uE0cs(BFYReWH1pslxlhM#`B$Gy$70ObjDXXcE`J4qT7?~869|cNQ%#XqMe;9*L?ejo2p$mzHr`h{+gYcM3dA0BZ8H!2^ z(j6U3268I4D_%8AoBQUAT*E6W39R`h<@%Tb!9a3txt*X|zVAek_NxmP)uP+_uj=;a z6qG67%&y1EHfuY%G(Gn1eia0;-Ee7HdG-PWAVInHHSaeiE4uL{S@XBgvGD_6U)LVp z=$I?7)O{%H5>g&{e)Rc9r>!M}9Hm(2JI@Hw5)3bLW$vz%^}V&saUAk%j3-TWgZ*-oVRb^$grJ%D`nDRgD}7Si+# z12)q5rA9@-q~o{SLM1f&g;f8lm)}6&eT=`65$Cr(5!vTYQO($m<829)4*=SQ1IUkMZf~ajQOFJ|G)NH`0C& zpq1?77caXrytLvYTq)8 z;~wwVhR0m%>l4u_a{`FyPAghsydRE4vO{7_4vp@_AyX@1y_#*i4yGK*$f zmk1BKA?6Tr>)+ts=^4>?K&>=ZA`{+8ITu3qmU(DZjJBUhI$@BgH`xFdk=Qv`@q4UTH^ z0b>WnLV45uhApi1GB@Q^p^ONn$opeWs{s^<$>rlrBh*(%aA%s42uuZqsslZ5TGG;i`U5Xbdb zCcwCp-($R&r>=Yh&+iXatII{-2{v5-gZV-s~0g6Oor8g21T;+m6U1u8#Jj^ixhl-siEt3Et8Cn*}m=e$(M~M zfb6F@{*Hr0lUl4%OcNa!vlIEiw-N4?s%$)S_t^Bg9)PWEHZ510&I@@=Q=r%0UF84( zw}zk?vn|t?JA)?POCfg#0!QHhT)VBn+hiVz+%OFLj2Gfpc~qQW2!P&)0M?C;NG;8d;#HR- z#NfUpv3ftx0;Ef;*QqoLe5M9!pN-h4-<%WY0o-DMT#sC1Gj&&-1uYsw;irSgfu{o8 zmJkr^#X+P{qj))q$q*tKPC7%Ts|#(kcF49OlG+l>JBC;@igOGW=>}G27P%u#l%Wb= z=~9wJ?0Bzmp2gE(sH=krj|v5vQM^C(JyrpgY7c`6@6`+WpCZl$(9)G_MtZD*oXW=p zG9``M#4hW7;_6~@ZYlEuQqn}^H2YOQNAggQ6=Yw3EHJ$3iPcrOXj4>Bl{pGzd34*3 z(F+8T>Tk8ij+`n0M?#Ret6nJrsH_!&!FPr-G_x<3PKX@@0iZz$ivI`{ z;o6r&)9@4yTf!Ge^&gSTk3Q%53yN|pKEm;}_o69E$NoT3&aaOxMe}Jxm735=c|*m& zp{UY@Q(k?(wxBEW{Bh-6gM&&%ptP9HNaJrPY5<*L$NFm)w&98WhN2Ccc=dQ(H&tb! zsABw08Bfc(H&~9HOEqd(vj26m6%&73!^51=PwbUt^jwInT)j%ySp&p)Ovbg6T58hq z`gE(O0<~M>|hnd?eANP}MFdWn;Sb ziR-*=8Df#&{Hp9_`;Rks3MeEQVq9xk9IQI?m{l8b4(NYk0@}N@`3*%`q7*SCB?dI0 zCc3S$yHslE{9wX^RI5;?8;X`82nJ}FAai=ul#A*tC3_U(^Q|>4Fiqx)5F$izIaV@V z-w?#d&chO`yvbv$%F1xQo(P?n+(?Gy&#uQ1G!JZ~l2fbNJRva8f~1*>Hm;{=Uvb!F z*fL0HCmFN*}zfk?izqw(r@9{)^UXspqfscNbToDhfvLWC8m@WlSetgue5 zZFsvExMtFqx{5Ui{lqn3C9iH*3}22ersI+YR^54q@7v$i$?{?U58o8w3w-ON2JU2V z6YjzHs$b|M<)4|xQQTcsqNar{R)UEhI-7UFIQ~H}p*ryOsiG=kyp@12>it`fDmzK1_;g;73+GQ*!Gip z%=G=a@*sGaI{$;ZaHMlBbrHadQ64S-Rn;=I2!kTh%k0j9jw4jDC=iM}oNitlq*}&C zMg;(vu-!x4b!IcXE{bH@5qL~29~S$o;;k@Ogiw^~B?-K+HH~*UDh8W{_&?&H=v#5J zf8(GEbQmJSa0Zlv1IpHRBnff<063s1|MU)n{|a#aQAuNK0O0a_p;XVC7NkGOE~2XX z0N*32K8%yY>Qt_TD6uCVv0mOvKKo&qZ-|u`MZP??6Af$WhLG#908tDf z&3P|gwp%biNn;+kGlUu+0T6Pf0uh5=syfPz=?Ui%B8SU9h^q+^2(QxYtuTSXc>*D| zAq70e-1p4o0AYi}mWR?kHT|=rk&6)$HLdcS@OOG|M)jjsHsV**tjiWULZAegN4o3y zlLU16L@VE|HPw$|Qk7mgM0)4%=2^xFZUsXGH;+gGLQ00G~HrRoW>Aw)eN?bSUyWf*^(wl7m z7!KH@wHe2u8ly(<0xNy^pH>2y#xd2r)zvaqp$g(a)mg4h8o{*8?&jC}-0(c8{#F%w zfQ@hqrGk*Ci$-BHy8%2H0L?7zrYT>Y7rhH1X!fBh7J-LAvr}A?3S$TaqLIjar4+Mi zS&k3}sMsts%uR(6f1qXQb)lRU*6%vQF~*ma@AdWEG}Jfm0wptl4Vr_DK?V zGa3-$e3r<+`i=U+E=F_Sonmz39lI_eK5T$Z(F9VAjkm4!`IGlHAos8%BPDqXTud`HuHw6 z2qqF-U`pd~q}VxGRf-}wRJ_pPd!Y*R?Yt%XiTB{7B=QOCklE08a%2#Mt7bxaHr?FI z!eT){o;|ouihHv)vExe>T`a_ER;e>r_8Up*6CV?{GdBX^F3Jr!?QOc?+1@uf_?Td? zUY%G+c<)b6>$!OMGR^P%Mk4P; zT>bY%LFW)e1L=Hh__AG29QXD`=p;?^!6-5!hnr?YI_l#M#Q~R15nYLs#mXB4plw+> zb0OPB(pI1D(wrISe2BxZ%(Gv>m;28~+J7?Tp4SLn#)n2yPN!X|c+~z1!OqyQpC?Bl z#|Z#XjPJJIei>vxN#uxxB;EmO0);H2Lp0i3UX(eWFib>&%P$u8qXJ4YinlXZId+MA zGHQrzD}v8B+O2XcHa;Ehe zjIF+iwO(tBC60q{2Ae}Q0uP%5)p0m~Y$rYp6$T8(S7gZ6>TK_Z@yW@DcH(tolEAH= zJN@wwPIoQHNmgjOWF(Tf;6E;w{r9_*7F%Ljw>mp^IxX6xS%nW!J}LO{%E#B8)3<%V z{N}GVp50+?^Zk4l!CLFFGhM~K;CU0sdj8Ac=)FtE9O!JZhvl&%6#g>EK=72w2V&ga zCzK1xSnwgS$6KN37I7nTAz$n_2WtpqvKgH9(k_Ax11T0{Xx5rpm587Fvg%tz;}HWSgBqbS4L@@; zQBkL^sqt2!R)`=j9_6VZHR17SS_9wJ=$Cs`~D94c%SJd@dKYyhQ2LkvZP4*W8y067e%9~~wICmST*X9Lv zhL0R88`o*5tC|-|O!WDSkf@;!TkHfr@5zL^8b4|ywnk4k8@XkyiUwjJc-#WvJ%nKG z0~GM}PpWoGy{2Nhe#6`F76yf4M};;2z;U5{_r=3Oo}{>=ApvqP_hI=Y;7vX7n1KEO zc-8q>BU-F+^u2_Od<0SrJDKpoEqO9#LOG>@5K;vDE;cTJClHQI27sW#CO)JiGiKoA zFr+8Uk$MQ8!ag`YNhnU92(jXNm=Y1&{ zzhUFO^rNOJ5#fU04-=oI_OOD}yKj>fVZljrs0+?UP3QciL}chVB&4QHX%tJ+hE#w@ zt456Tu+a`%b5nbW*y`xY6?zRYnf0rMtcav(IdgJb%~9w~OvDq@vr|2$$Gf@wZ=S^Gw)kWX9i46?_jCC}4`PZI;1fTW@fNXa>X4(jVqV zDQf4`px0OpJ5AVt1D*$_@#qwq^UFEVqrvE#Eq4Q5ow`l6C1B}}LkpN*(1p)iTb}`` z)ZW%-XHK!OXiNjDsRhXq0haYFn~6O*$lY0LGx>~H3Fi;LDfHrLZiX2lMxXTX!*w4- z|M)K5&UXJP(eITvVu)rR&Q@<(`qb&Pg z!Wn!MFBuQvrN_pAjv>Pl63KS)jco$S%|^NQn&AqA{9178tLz zyGriC1A)N+{a6|W=l<47WVHO|UDN!v*sw@uFhz4?A+z)K?gZy8N4&C`I7(Mo*n-a> zuiT{^UGfH87*xj}b4%34inw~9MmBp3bo z^i!)9b>Sjyd+Rj^T5B(N!Phx!c7dJN$hW%wIh?@pPf>v%{fyvWMFm_OW0FHg?~l@7 zcVtyOZ5w&`UetH;^0VYv+k2n)$1f{6>DK>QQuxncZVEsLp!>h$1~)|3zZM+*OG%-| zx!WJyz^xXo+F-911a@b*q)jRv%|+2(hz&d<3Jbau#3tr1R;KqtniD!j7+mpQ6y!(7 z+1m7VvPR^-l5IP7n5)*oVwpcG#ZgWdG}G@=LETd#_~^LWxlpnEQ$SFf{7O+@<14U01y~A@!BZ%Sv6xf)D4o5elVZd904)a7$Fs)j7dSU$8P7 zt2^EH1w&=O=ovxvlX^Tia5z{cJq-D>bOmU%=rJsmTa@ttc<^fyp{p{QK^HA4?-mh@ z=UZ0zr2Oy;!V~bB$I=L|n!tt{MpF7bghbS+Fiuh4VFNRr?`=gZExP(LeUX&V+{#ET zYT19!Z=xtzs9&$uN<-jC<=U8SLD&KLM}C=?hEalsjiyj~8;lN%>1+Mqx&1zgCjU;xB`Yb@qf<9kOZXH?#U zPyD3xmSOaycoqD9xsu~f+y;W*%S_*j@9bhmbj6k%rVO3-nr8nB=Jx*2re8j!@xN*M z-S0gAT)E$MdAhUz20$QifP>Hq95hmV{isceFR^>@7WPL~Ny)DvqShvAo7hJ*7t-5@ z-;F0nxz{MKC9{zkS3k5H2LkesyU44J6k+B2xvHGy9*n}(Ik7r_Ap-?k&F>!+xRape zMQUt#llShD0?HALaENd>Acf@P1MvFE$0)_4FZ~9LUWw!8ta~TpPGF8`Pqr&6AAXVy zGi`)vX;sT|^yT#&(L1W2x5nr0Gmd)CE2tlq3aTOk52x=3bjFKCmM-~to(0Y%xF)D; zND47^8kXqW2DrCjb#`gs@rUs&te_61PTJWwot6w(=tk_cTf<0@+qfeg4N z1A4?->$)E~zYSf;%!UTn5(+xY5pip<5g9=Nn%QoF3_nNZ&k5d!o%hm+E=F5bcvrud zTQOAwG8tV%kLnm$if&xjhEv>S@8$fg$OBeg)5mA$$Lnv3wh9)PCAPtD<`xQa% zh!DR>5KI4HEYuQYa3D|HfgLhvPfjz_?tw7S}HmOKaN+c zZWQd*5LYvB&ynsRVUP(RQ_faOR~3Cdsuu_ZwX9H8F01xGRz7BK*&3D&y!GCpI85VA z`V06u0&DSVld3BaAeSLoEWxd_P&hvab*sq8OJmbK#Mx!;|fqrWiEdxSB&q(Efe z>n02V~qH#>pig9 z{#Ah&N>9u(Vt$3!`t+(;RRP(ZbtQ*=(|pXmzn+pnZGxTor^)0t7l8ZoM$^Ycs45K~Zre z^CQk552c*UTT!vS)h*e%)_)GfS;(A}ONYk53CkLkoke1hffaf=j*+r=I2{6<^Igv# z*qRZo^(CB#x{mmz_@LbjLo|Ikt`SO*E8D4^HYc9=8&UQ&ok@Hbq0~B2FN3~WD>0_@ z1St5gCgvE^8>{nb*&)Mi5T>^egeBv$nw!2~>3xm-O3Nl2w^rGySJ*muF)KV3V^;)M zvXGGj}{2cZ@$^S^m74ll-g|OEJ`1A&zI9)x6J(V z{;>(GkIW^SYCY?UoH3~~52$^-d-R3&?*3Vqc@P+w$*T914Q1|fX(nN5u-8eDiy>B? zHUgZbW-NwZdl3M_D8#2@1c4n#LxTH@YcbJ0FSoQnucju{1YxrvU>q~cEgXP=F3%=G zh5>7nKq8!vir4w&flIu@6}~PP-ure`*7I5L=X!T6F?eT+e%(n-xipg?LL&PKQV~$pE`m# zr48}j=LEM!?4&f4rB>S1NcTITz?!K#YW&5G`8Tu_g!8^o(H{8q(FW24hEtR9uyvz| z_XG62$nF_@mIG=q9rnHb1>#qrQBO6IM`=UKZEoRX4hPBN|Ii67yGO&ql_U-7I%0}L zBP4YNRLVt->`I^_$kUtj@+ozy<6=#sMDN)I1-i#jXtusdeRY>6xGDf~Ym)cJE`nFv zeW?jPlta)w%Nx9hptmYy*5y|XT&i7Q0T!P=zMA#+F7tYLy>Tl9GvnPn^3=qu#pQIa z-+Ps{iK4}?D@jyhoqkTRd&Wht`opKEoD!LH+7#7(3f?Kdme&d&UrBldtG{eljx_N( z-J$)O3YA*G76yD<_?{pmHGvd?(Kf0h`)FcNfPteCiBBB|+N!#jihBUi$jR)(pw1JC zdv}w}H;jXx3eP_zA9$bV{|2T#ong^+vQ_yBhFN`m{p|8cV(O0YaEg3__}T)C@Il6# z@F{Y-!P>~X+g9X{9QFfYv&KNGo;!!Cp-ie#5Xih$7qH}E2vNQTAVnm>e@cOh9EpIE z>&or`LG(oqDquY58)`mC;F>zV42-SQ-;$jEax!}Bn#)*jN2!qc3Y!nEYq?E5uZ@sY zL?F-uWRHx6=y3pc1lOTr`0m)V8R9U4*AB3+_bK8uAP`74v_}MnB)p2{klNU~D-N}# zG8rdit|(1~ zvXck-LiJ$QT4UDZf%dnYR7c8SfcONQG6x{(df27}qJ+*JR#{x;5Zv!Y45duN5?lwFgd0+{@+4h7w z5IE`zYJ6b^h?+vE0_52#+2U={>m8q!YQ$3Wqim6Eu!VRbpB7?3iRe23QMpnqL=8nO7pbb8sdTkei#TwV-bUjUmli7VfPu3F*D$`^IFEoF8{Pxp=$1VnDp)Zg7cH@%X***_S#&AFa z&m9Da+l6p~V*Q!JNq20>p%b%~r@c{j#B-&$IYpvUKHwu5Rb$x(wtK=A}N?qofh9hspEoC~n?+Z9}BG zM>9|3*oZ9JMjmP%KVSGKosdwUd3^Z=MSd|Px?Ay|p)_EC011EvFa@yvmk=--NZvo& zzWz6)!vEo8gu%v}ivRi;@gL~HL7g)LVlvH)d-l7Ze?Bu%Z z-2&^r5zVVgpI-ZRk$iBd^O=V%OfRPqD7Cy zq{&-2#dK8gyCTnfL@GGOv(&@){I{&5?e9t1w^}Ev+}6g?*JO((m8R|6HTrm)Nvfqd9 zTT%)7$LoiE$w-e{r4pCAa5#5qoaj)p29N|H{7%4g1jQrbp;U!T+PeGa_#L}*EzmfQ za#r|;qdyEV;HUg<|I+iU5v(DsVD!sV6hrs)Sn<_XcG;AWiD2WH60!d7M?y2D7`V(G z(&kwOo2py$u+0w7F)I!cGN}p3HM95JIy-6dyz|`*t0_G&m8LAXcAfmA{IDR*+>683 zm_Pgr*Gmh&tG8t1gS9DWB)H?(&6Pf)JEE&DUrxmlQP{T$ve;`TjgM&xQxP+!YkQuQ z>y3JeVm4r~@3$ec+SBJQ|y_V_qg|P#GE*dwauZ&VfgIo>4VFMm5!l6iK1Pbw% z3%Mmsq~=Vb8U>O<*L9P59E5H}VXi4Baw0e4?sKFLydj&&uvT3`JXB{RCf5}Vc8(;` zp#yC4_z2RoZ^(ivjO9ekjbG7SXot(e2MC8Huq-S(2p4n#DfyB7V<{bVaJiKnj|!Im zP93|lSo(ICZ_?kl2=p(-SIEDQ5`=M(0Z@Oo2;eW*VWA2Hg&!d%srygY;n#yt7-ZGe zU^egMr@JrCQvDX&MCf1vpx|Xj0DKDt5%{x3ydH&kixpn|lS`fH0s+`#Ke zeROYoH0Z1dNbKFTelJAf&F?!1t=%k;&|2%3`bn-qJc9)O=P1+;?zPVjpZSO{W*E{u z^0KeI*`~hoF{9J+s~?Z_-{}g0!Jx92Z-5{0ACC;RH91xLyc>d$UN}GKF5#$SL_D{# z4|cckj9=64!C3TgET44voQ`x20upbG{XwHgX%L=%kI#ev^z-D_n7+-8W45)m-P*FcCp6*y@czHAk?qhke*XnVrvI1z) zVN>1ImAvC>REi2f0ZF|?aLw5)0XwfRMUfnG?^ysW3)qhJo^2cOd7EvVJ!v}3?!fYZ(f$A!sOdP;y4Y^GCZpCM*?50qLgpAWd-jPH2T@x28k7V0{)v(X07x2ttUfTz zIO6}Y92B^Q3G#mlVK)-S_lSX#E7CyZyxJ&bzetrHOi|o6r~?ZlLLr1p_!h^vJF4Qj zKFJO9^C>HYQ9J2D&U=V5Hj>3HwcOr7tp=~Fy@C9%w!;_}*@$f*t%T1xBdD`y!0!PK zMu9PUQu^%PpFeg_oNThTKvyY94#G@)3y8py;6XAs2q;fG7U-C7o9CbuCXx>-Iq#!p z8s`ON0~2nlA)?igiZ(hc@RoB3qfB`3Q{coc%ItNG0tXDi#vzrVZ%AG3`8yRpsNXvg zbvlV@3_<%E$r*IE1iLi|IQ$mv;0GpqBV3$mmLDd$Ys){)f$oBeScuzhF#7oA2;MHH zuBycoO=;6G$?hSh5n}+p8Yc>#3dtNKG@+!d9d-3cc|oKN0!x1g=JU-b+PcjxU8PM( z!~=XmT!&}#t+{ZSw4UTpZvpUK>N?ej1Velp^5sbdp-i!)ZeCTljxnsLo6#>kuSbVj zq-k7;n*IANypA$u2}j9TkMD&D(sHy4AB8$5WL< zN)Jhr_)NS*NGcvoS6bv77#4#ej~w1|FhoNhSa3|cIB@VYd?gjGkT&YyxKXL!#H$lG zHQhC#RPd7|#1cS38Gx|j4!pH%>hXmV2U6g;cM|C+CeJkJs9b^bb(F;%lqD#BY>257 zU~d*7VtiC6%ET++_&VhOVeUQOqWZG!-(3lcTx1Y|BIh6?B8XOzvw#RFIfDcNm8?{e zGer;q0m(T_2BA>0BtZcsN0AJYMe1sIpL@Q2y8HZ|d+(e3C+zh+bM4QVbFMKa{RsS? zTw6#L*T)<2V?)WJ4?;G~0ybV~kz3IH2@J1ypDLhvdcAXobgn=a5CAM`C%d+#q^f7O zU9D&UoZ^%337Fo#<@}`;Qjg~#j97ESjiV}k-*&y(-fFKl%`ep3fQ6Y~7Gt{*Nv!Vr zhD^M?^`3N}>SA;|iV$8I*s?f3`?gAvBvV{t(paYW#PGdwv0E_uv1P3dQzs?u z6%XYn7NZ}^^u3aeN-*o!qa|8{HWoD(ZnIHdSm z^zB*k>Li!?eH|M#h!{$WAUGhN`=5ZwYf1x7)-gYicjsgss@g8z9qs*KKXGj*bQr3% zGG_TfntM>b7d-qU0%sleco4aMR#xB%>qiu=*%Nmxh^p0VA(*6XA0I^F?>;QFW3rd! zr#V7w6#z6(-~}F$@Ak{F_D^<3iN|^a)D-#)U9#qvan=yg{l-}#st3KDEIZRVix_sn zro1?E{{<3=F2iKb3k@QN1%U={j}Lsi_egj>Ih(I%3RLgFn^&w?cOs;gK#HJN=JJd~wR8j(j$59{i7w;cmm=SDyXf3tv_cze06|G-uw1 zWJi`1561MDJS$}&&~21{X;R-PZ(J?esA&BYZ-*y(#mWqcsRo$!xW)S!UPV`vys^1C zwaIHh@?nC32Kdd%Q=T8T8jQK)i|YL2IYT@MiW;s46$O5-MuphJ&1TK)CkTNDstl92 zrxMgF4Wu}twi%4iH zal5?_NLrb+tp=m^wU0<^`ReSYxj4q3;%L$`fI^Xz#!2q3EFQ{LKjqdiBZXWp(n>=u zhnFAE+C3szd#P8Py&sx&{lQ9(d`@_qFysRXf=^Eoq%c$NsO}v2|jZReLB&+&yG7Z0@2!~W1OL}J#(z|Q%2HYL?!X~8ZCMuz|T{tdXOg75y8jgNehkKB= z*OTzPG2UQV#|1!gzE81=3)+LY_6Wc#(m9v4r>e zUX=B+D$V1?67%pYF*G357&k8X#<;GDR}$xC7I{)^Z6F=WF$Af`(JRl8K>lcYN#S8m zy62j1JVmjF))9U3voDduO@v`G zlMb*_Z@y9lh#wwVr(T0cVxFB@Zn3Zrw_R&eg5( z_hv!DbuT1uk$ZPBN6sUYZhho*(JA&YlBv-mSzgQ?O4h(CV?$tdgwD zhx$Cs;@>3|!@VH6r}bTjPgyOaEz#sv;CG0bdlU%wo5m&ZFo$ML{#i}AxsTIF@AXW+ ztYrqi4l2#^e(LP)d=V>whabT#MZ`KCg4t8cq-Sx| zm)+uwbq>pJe}=2yl-2IN2rAXk*nD0wAABR@u#UM{V3eEdOFk>=_Xk&a#?52qA1{GQmu!EC}q~7B}eDu3BJQf)4Zi8@gy<&l*T3lhB7 zmeJARLl}e~pMpH3vUqY>WXlHXD93RagRE?E`?AdCst+Xm|Lw{ z^oPN1&#SPRZBd?dGVqRfLkq=`tLp0?gF7kwmiB+$_=N*-U=ry1mk}MDgG2$Vhdh^+ z-pl?sO}1#4%Le-V3weKVi%zz474YQdR&m9WrgZj$3 zmjIZIA>Be8fN7>7bf+;W2vH1+xMh}N$J~b$wfcmF`ZwQ(8njuNw3;QqbCKV&H5-&! zQq)Vo{h{r+zt&|G+IH?d*m8`{t+6GPJe%@-IM*BBo{TuZ>M}*-`fWC&)psucItTD* z*5?GvmCGVcaMYRib0o?vD>bKh@5wu>sjC0k|rYW;65gk;sirY=Pi*A(MCu=2)eC&W7K0iA2tVqnOHJH z8(R0fV%gVuR#I3`?Q`5_s12tbbrQZ8C?B2EX3MpAsugBDz4)OXn8!S_;6i=?$BI2- z@YS&Zg8(!b1&lc40^;lZxo`l1L6A%SO)^$H!qR-dLSdvx3@MSrg^;tG>fx89I2wQv5ANtEx^0r_JhK-qOvQ0jZ76HzUeITebgobLW3F+5WP* zW1K8sSXD(GR4a()79>h|iX%$e59{}TP`zX>^W%vqdS{ru^xbW!Dv$?^4vlXmy+52-#zElFvf7+V=Gsh;(67#w&q${#;bjrH=h0kxC zo6c=Pw;xnA%VEgH{5RjsCARJ-oIck9XWrvCPRBQ}weI!;xzT%wUazU60pt7mrNy~x zsd8n8R8VAs!{fJ`7dbQ3;mBgQTa>X^6qVkjQJk1+QKJ*U{6QBU5dqATG+-PR-{`|*;Ti#I+R zQOb&iyOA1VaABlSD|8-TTLz-~s#<5}?V+MhzSRweOXrF7Z3v8ctF>QjMB;kZx307z zoNxJ5C-he}g=2bU+?Ndm@f?7?I=w`8^C0^;P`a}gw;SG0W+qXu9BY8fWvztF53kh; zot`)XkEBD=RB3PD=+{IZmeO2^7I%(!Arddny5z_$perF^B)QX2Xgw-aiyK2{!vqsh z)*W=Qd-H0%Wp=5)SWGP45|bYhE93JeSj#(+HV7WkpA|4pl?DV#IS>=OTItbzf{~ zb#$<#jYh=7VJlMZpaVKb<(qI_^UJJ(E$xpS#{?FidA+M-6*Pvy=A${K&YOd2th*Wbq3fm|xj&?9~ z)+E}(Z`9(xM62d-AX4}Qru!)LoULp3xW6;xRec?=IjSSy+C=5K4_FZlc2YhkzT>UO zPtA;uH5iBM(DWjW;o*b+D`c4-)D0vvd-aXo<>iMZ6DFTs&}kVfeeUla2i6T78%B6k zHfXXHyWYk&yHe}e8weS184L%G!dYB~H%yOjHZ)q5W^!*X#Cc9~UFso+il?!3uj;(l6j|a6W``^LPol@;qxYf`)hWS{lkP?Qj2VjtHFD9Mj-!)$kh+tBY`^rHv z88X>ay;T-k%1GkD%}Im6*ls{g%(MFn(iUPYB{nVNd?$0ey8YcrZ6UKmc9SW6^F-GR z+-DgXrKOKy%$9L1tz8@qt#Yo~$ybF))Bvl1=&9Ff!{p)Izy#s-*|G!pg6LZxuRQfM zoaM&;K#Yg8^|0k#MH*!dqKD0QGxsE4(UmaV>635Z>U>tK)Qg~oYNTjo`npQ~c)%Ivi^9!#B$HxAWRc>QuP;94$@xW`oD zFLGBc>fXe4<2qDe)=FN}-bWt#VkKA5=0HyWqsK#@G_`WyK!^oW!$v%9#V*u6Xlh#9 z2m{D`s5AiZ4CR=HR^x?8@3ARba()a%yz>#|_7n&v4$zzFl79J`8$alyM4hgdCWvS$ zzw%0ThhNijxHSh<@MYHFTt8&7qRPuqTZ73QjhWhXSfVFPZrD(7UzW3+mfV;=0m;M~ zw6e9TtLb_>c8>AfEQ`x{;@VP-FMx~!dI=$ov$^A7Eu*4 zOp@gBMLivx=|20B%Hq&V9HSxY(X2=EbMF#0i6q%=49v6Z&DA{CHSYV)-EE>zquN_R z3|WE9HmiK8cP+Q>HK`5`e?D4`Ze7m5A#YmmiV-bJ=j!2w>laEVSNm^v|dg(n$ z7RQ8gDC%NC;JrfjVr3qAnu+0CM-7#5HGqR;h**Ts(l1jDoqW%*Rn>Jm_WJP0R4xUH z4zF4Ab4JnBcM^3)#PTOk84iE8vt4|9zFkEz{Ztc$q^Qpduw4=Dx4U=WP!bx1sHnOU zP0w;K$9fue6N92GGjvX=M-Ud+7GGfTTsoya-a)qaRe2DqID`Q;$HEx1JLj#cCu*HG zw=j5$V=z{}n(mb@w_p+j*QERkk104r(r)Rh+BHNSM$q{162lr>ImQHoq&TXgsDr_m z08{xpF)&nOtrIs8_f~miuedocGde=~RXns?65T}~h<;&407API zD5xm=b81RIBVpd+mPG0(b)#EDs+IC$(pd{Hz02-mbJF(ba7W zs*4r#Z2*)|MG~iEbc_E+`gP%4{Cz$nD<9b1gsVn2ZbpS(jxKK0zF}7m*9_>XaqQ@a zxDeLewM(tzILbciC@)9WA9wGD7%kBt{^DxH27rKeVD3M>YNNi0`d&l8)9t%M9!w0Oy39s#3H$|nA;qTz zV6reAMA4UV&z4bJQ}x80%kgNDn3=Lp)?` zg?KJP<l0O+1d1~sx?#;}}zpS&uIDvYc0g0p&v_89whk{Fzhq8BJy6)*pp-n{-IK5V$R4C7TR(UP2 zL87=Xj{O3W3X{~qK!5~hKZv3vG6bro8vk>hP_%5``ysS* z9OiOAw#5!8_YkB*a&5b;2dRSgudK`ov&7gF)(%jicNdMBdz>(HkKa$@tVQH*EZNOo z(7oMvLI#qokcaDj1$tRTwBXx|H?rXN-d!TX%dEd#*6lza=m-X?xKHOhX!A8NgkTn* z`eS9pRCYNV^}yj3QTpPI%9K0D&W%FWJh~CL4(36ZMlg32spXWX=MIxp5hn}7{?pl< zG=1I((WC{^2a@Jq$vUCBI4o`cpJ0a$6$$6SxJ31F{3IMi3eP6X6ju-!7$H%X7$u81 z2!`fm!2~qXq<>tE+IeB}l!|~N9ehlg1f&NB?6pUkp%uN#^~~I2fspjXo2Lk`0sd=I zciHEk6cluS+<(#HJOvJ#CynN=cuBQqJ2~EszkjjM+wcGH(xg!CJm~I5oCBM-*5dmp z@eTomp>s?Il^UNo2VUxNELAGl$6kI@jevN*uaQX>Z#=`)Ff>1-<#?uPE52|8=NW#N zYb33JyB>h*o$uu^46d=urdGzfcI2|ss}4_23Sy(>Y4NH8;y|`uazRo)R<*kvr!=O8 z{19h@lLxDUuv*bcxFbMWTwg~21r0r7N{7A6e@Y?tJ^5q*8BvXAG#ZOHEdbq~@G`L6Yd$v(=CrXu;;&AP& zMGCA!H+mNgdqiH#cgE!T;5?4CZ-pSz1636OiZ7nXauYo1D zpC>&i${L(IO|VnJ)i(k}5)oenhRj4yUoC5f_cEo`q})HXm5{w6rK=JN#W-7AP^PA+ zfiN*7oebnW@Q9%Lu9pfTqQ;^>eTOVm^IiX=Ym+n=?&DPaN7oEec zGoO-zr0+h_I0H@GyNsu7hPV=U%Cm8IL}Fu}h?>qIr)~*GqdS zKEiMMA?$I#khtq*F3EY;j!d6$omqF?xJ&mL4C9A49!H8@{{D@ZV5~eqU#?n6_C4j^ z#}8EV0JA6vTrZAtX}tS5NoB)a+UfB2t>(_K!)a-4_>oAcmckk08JDt`8hud6QQ%DA zjGk{GtO9ZP`c4Ly{*1xw*tOoo46BV90SqiuJSae1yk zBG19!w!a2)r{5Qk=i)n{JUN9>R>n&uWmpNb8-)Mpli-G1kkmS%sPk^OuhC*biYXXG zTRHVnw+H|dghVZ`nk z;9lra73oMW@f2mVu%V5W4g*G?>LcKH$b0W&vPh1QhRRO^cWbfpbz)08ff#X04(Y%& z6eUENlBYLlvlcsQK~;tzK_G%9;8N|?G}ojl&f+i0I0RmWdn_#lV(>=ssNiKBTm=?* z|8a;QN1!bX>k22=glmg>n_$I5ZFx+8V>H;^ZK9I`iy#l=06{>$zYoWb==EF+o24Un zI5F+H7btHAgDTZu`|hQM1we^9BG+?;8<%MkbleEa*l8&FRU<@kwHPrh)W6>Pxq?)k ziWUX~7YvQeRKyTcGn-T8Oq9G!0m5ol`9dTMMn0tlPGAZ|4IgNfHP z{^m2)o2KU6ba~waBlr3ba{7v|sOKm;(hT&~mV7Ry2$GL^xHEyO1b0}mY8VJ;zc3C{ z7x_?Meh(=hC4X(bzQQz9V2WDl)T6-Yl^2A}04`ZEo!tKLf_0#=8vo|bSPislvU&x7 z4aC$<@_A`x*eg!PMc@awX0n9i8B5~>3#XB)O3o|kDYlDKSIP8Z23Z8OlN@;N9Id3) zxzDz+8Ys_7#LosC=V-isUOB&!+vVrPmx@9nkQgwCKnjN6^!c$u#>v!ZX7O{2G!Cx< zmtwB^dGh1i9n7<~fL6#fSIl$hEF(FhoJVCMfLu!791Clkm=8vl^CF)gq2c6#&_fjt z^gcx=5Mm(!fs6BT25NHhuQy}{vhzOa4r3&Oz>$Y1OT(Dn^Q@N)9wP!UlKUeb)$~o-37YkdY^elh@sZnRw{DEc z^#iozkq$LQ`5CvhC-YIdl=CELV)9d@IGjHfo+e89eHGQ9e_bd0inb3s;_&iQI_xu| zleyPPY{-X2A(4F+5qz^gizO#BUszv#eM z0p!I;6k+%DQLs1k!atp9H(eynt1;a1;q}w58!ns?Pp*FqCZxy%Pu6A=MW2}IBCKgt zNV$vC@5Xa>>C{owbunDwMDOS~$Gca$5jKDBbUaGn1a-h31-eDwP!bk@PXsyI#Stc7 z8MGN|Qjd2nUy{7iK_4Vs_|A{IF62G$Rd{ZwE zbJ8kKWP5V4E&!ydqX>b{mwV0nUnf)rqh=^9Hm7%=-dJA3>EEJ*g9H5TA z-j-q1o*sBY9nOLjguF1GoqEX7ws`uT*juy&RL_ROGv@q!*;GWygx`c2a8ovqR`)3T zVsPv_c!>QI=GOI!_p?7e8yubogYHZ(s?kHzEjx=;5IQ%aKM)`NTqRp2Vm>i?d{RPU9;f5+9kuj>8(t0&FBdqno{+@nQzIR48$`sxeLzq>~}%?|u$_vkIp9Mlvp z;G;IG#mi*6?XT|9lq}N!!97|%?f?To{J-nNe|C><@wlF4dj5C!=s8Kug$)Qa7GQG< zp4Yi(U7p7NagPo{DDy7_(=CRmhA`O`ErdP>(HfxssX0OkwERcQAKJ73KO8S&6NDNt zs(Kr8Or)(Z-Lxl(_9IhuFu1^$GA@7#Z3+%YGMG=lOF?0fO z8)Pk-YQf)kRD%KGpo0ft!LUmM!?y!Sf{=RL)LBq91k*gWpPF0A6)b0g2mLu-kjjYH zhugo8*D@3ke`#z(rOD43eA2WZcblbPk|T(C#v?9>OS|Z`h`=};?;S?}a`PmGUHtmeJ}&BgSx6757khAHD**cfL;wHN>l0L5|Y- zxx?7HlWmg5wC4Lo3{|CVUpQG6*~Wsm;OIEwn>u)U6O@~JIQ95T-q0x5l-0=2l@s0& z3V#nWsQH06G}z0|JO4%I#T!0{YM;}2dUJmF4?>KpDdxG%7le9_JbzNj?;!i>#0?>g zhq`U%<;C8d*KaAsXM~7uh8N{9OOciqxvbh1#1TvD0GU3#h%Kj#6m)qhrJHUy?`pi| zY|{$ZLKM18TC3(vgF@_p^>5mTKA!Ph3~9?u$_|F=0j(i)P)3toD>6}jomOE|ab9S= z(oG#31i#yq9TTCzy}X#ZcB&gmwaoI8n5U5^NR7@%NA8ln|49AimzN(a2^nZi}~F}(7oY+ie%3JU(VbB)`_8fCP7w(&w}-#I4I zdL~sa<^8tIO2YY%?`bZLgYMhS=CiV){G@86@9AKaOxqO@5PT26%T|Ibhu;>4ZY^gH zNN+hP_()W+QJ$38e!b3>3Z+NJ6}+p8)Qalx1HYbzwr>Nr!+CW&vu zKXE;$(u^;Yq-)LAbq#JQCWd3Lyc>9dP@?T%0ud-u-eHTV5Uk5Yr`|<3ER5U&X`(HT z^0%3H3vtSgh&?0q1pK{S>kLUy{TmnURD-QJGZ_>00M!t6Z ze4*~AMBPSD$QI;Jw61i(2P1emAUCr}ZTw&$Wx9Mga~AlM{sSAZU=H!M12lYQ`J;aB z9oZX}_vkzv7}FTkBRNYzFPmY(!lVUn1!8PSO^a#GK9ayEIrD@aM<4?StmGKH zi0D_n7|tz*pEohE?LD#r`FV0iR)J6az8E5-EHlJO2-?sW5NxdFQC8A$w`kTOVeL?o z4$))e^i}t#%@4Eh>SCjjy^qNQ?;H@n$fR8e#l?s5`_NOuc74(#Zk`eVR1S`YP&K5QD{oDNko6u(g>zeBmaM_K6Payr^eJy(fHlvhxk6`(UJzl=jn7ueAR=RjGn$Qx>rhR`24Jhz9%90@?`WQ z-%1w0S_|pmH8+yKJpN|_*nkG$-M{e!%1Q{Opho@I^F8gnoPP!jxeJy?30tKA(f{XQ zu`y_F_RnCk)syV{53oon-(H`QJunZtphBQk5;kKFCD9ySnf4Gf3 zr1*a6%23u!=Gi2X{A_EsmY$ltj=B;_v91jN{*9wyPKr%5%DKfp-uP|AI)n5V&YgL# zI4<*{R-c~5As@lG%o_ONhG({Q+!xG|=WAP{1j6a9$E2r;Ut#Iyw>_nMzi)CtiTdta z{KEZM+S!J)|FGh|F;h7AKT<#b!EFzMAA~Sehc1Ll9QG`P(SHGRDy+sdj_`ujtT+|6 z%NliLMZ3ePqb0^2mSQ+R2j!wfKs8HoC?bJnoM^Q?LO}MR>T(RDsAEo;w(^Z^PjFK* z5#UtjqBYHF~d3*;?X%ZZj+U(3ZQ3pxe?;8f@f z5}$DHKrARn6&b;upjec{o1@B^Z~{kx0py4vP}NKz6)H8#B%V-jpR~Rsivh+qD)G>% z(yGyu--7XI`Rv2(fz`8N-im9b?Dj_hZFF6S9m{52o9lbwHy8dw+qZ3u`bNFnt`%%G zZ-6-m8+Ro%BAbW|{I=dsdI*&@5Q14MTMsVr3yVhpG;ncZr;Y9Qr3CPlx{j2JNd4$q z0n3f4?Vlphh3)Ywr%;-cjg~C{)LZpMbU+yTp<8z6`JtZhNJX3AfLJG6oUv=*7QB=k zx>Xcj{cfki`N{h#`pJa1cy5RyjePEOdE+ta{K^1P>_F+r7Q>8irwAk+IW{_AQvL}u z-D=({tK!dE2`5fe6@R;~A^d((Ci)sybBvrdykoN@`uMikT?WKo%{JM9`F~d+f5;n3 zi~861t-}14{~+(CUG@)=7iRx2$@}TL$A2X6pW_Vwdn^t%dEjrtMkx-%Cb!S{N7$&; z6`mD;AOBk{j`35>*FUj1GYJhlbEpsC34zNC8=baT{Pist79n*ylIs7fK;(JMLZEeF3!#_) z5{u)o7{Omqv=}Mu;g#)9Bw)A@MHNbn)sUVIXOE#*J3sl#1nl~O5uc8j1p2Lw> z5HzIRAe2y^j-%}hBRwCM=VhpoSRurx(MTcS-1>Kchp zF9?iaAgQNzX)Xm#tIvk&PVLhizoxQ3_CAYP%WWrrn3e4?dorBvAQ;J>ixB@dnw@-0 z(9vJYZf!L+IaK{y(Sf@lE;kNL=zsV;DeZe)YRS7BLcabUX-lt*iAwAOFHY_{m6i5~ z0cihi$6|mmy+15=tTRY2CUV;p zefOi=KJ9(M4CKnY%2toaek$(P3V&k>GSKUt^nnUTdVTj0YO&Y zsz>L0)myQxe*GJb=RZbN#C{!on)}ph_Rprb!0$TzcbNbj3P@s*^k3?5Z@wV*?>{OV zRS9FwzboJfcJn`VxZ0g0^?%plHIE49KXnKuP;jg33XB%&Xe~znu0z!TMAd_t{?DsS zTKAc?x76!PM*R?*6i-O8MdMZAx*TDRm+O^SFW4$i_+fd6uJs1#ks1+)F*cXzh@!

    G;L-emdkcY#2g?&8v?u_91Y7_(u%?~$ zsvaY#mC!PY%8P=?PIDxJh82+cz)$K2bUM z^SBRld`wKvv<7|I`YiqAsz zES~;y^a}Q;l;kk?qNLO54(7HxpkjveHqHuAT-wFcBmDPKUWjqS#Tc>`FOH|o-Nzx0 z7w(LU7-MylJe1rsyqDNe(&1OR;inA5paCZ67-!+b;YTnZ9;SOMw&*+&_r4w|ab_ z4}QnY0d?Sz;Yf(w?qPn4mhr|wlv9d}~gZN8>dPqw?A9d+E8uf?H2V2Gw}zfRCEjXW~a_ zxwH@?q2<>QpYE)%#G947XC$|k3kSdWr0kSW2fjYLSW`L0n|E)De1=Sa62Og;3b;&` zd(CAr&BN;6;VR+f@EKno^;D1m4mo42#^EZl0m0wV|; z)t!go&E`V22l7aHoTfeQ6CbYV*1cp!r#YHl*vo^i0O3BiyV2Z8q)yy1kG0fygh=y~J-S(r zJT3Ajbuz&OnHc9~t_9+7&#GM3Weh(<9sl*qF;!7N_g49+MleMFi}ZzV_*adI-9U}5 z5~j%JsHzuj@~{MZdU=j|&>q9R-uX!Jv{ywHO$1EGkmn-1_O~HdfD`DZD z{@ho5IAH=n6|C5GvB*$Fkg?w_Bqp+7PJ>^;N?Ml`ds5D`C6fUqtugqlh9;%)7BcxJ zI8m=@MeE(QAthBk@m3uSSYYuFetZ&b=)YFsks$;N9)aCS$)>YgQ?%jQfc5xVmB*#N zL*?`&*3mDEBh*w23K%Q7CpjG#X6QBMgUw=w2F82p zE)fWn@n?Q`Md0{15u@xh`RCcjiXB@xm^-nxZ+=QyjAD~^sVXd0ZRX74*onx|H8T*H z*%7R6HlcphWAYH;mz@8EtT?1K^VQ7R*WD z1%0{eNDQiuiPW29D5Wu%P41UNsV?*^u!fQj}O%wjxN6*8&&g2b~d2jT(JKlv@T@tp7D!J~Mp~{!b zI*R#{@vZNhv{!Sg^C~i%K!j!S5dVl*uCDq{{{G$+bHc$C(T${d+xri{I`bFQyr|6h zcGC1JyNJQPWn2*Se&|}oNwUxp2HE%gn|1Ui4fO#2{*<=2|(o#xS(Lk9)2&%T-Q`gmusJ0FMh6gdxnx5e~U@>u}t zGy7%c>SZs=`NyBF*4bv52Zd($B0K%Q?HqfLK}PgoLN_h* zri=aEKaOU!OfS;IcF&R4vr~c6vvP-Jfl1CVpXr_H{jhJburi$rW61@%tZzdQ?T-Mj zMG1(wC~3yrwaz!+=_nx(mEZj4BwL9804Mg0@ZB`;{Ckg&OT2D>*=26t6|aDg{OqdJ z-{yYy(y|(n!ZEV0mT|_L=8}NXrFxW^Kp_K)OQK$fs(wVmxrS{c8>#We0kAOZda9Q> z`m{sY101ZA4%HaqHi@#s5LYuXe$x2<@$s_)kD@y4=P?Nm74u{jR_#3Pn}JyOBSm8m zv|2U0z5%8#pJALY$m~FU+%kxaAwXW$`HD?2tibq6Z=nCX;2T$DSaL0ePXt|TSmlgp zdPB_2988$6V&`9os)K@cRD(Zb17&qN5cAiK!>%wivg`5Flml0TUP|N>f0JAb<;!JDjk@RZ&>#mUGH6*+Dy(7>pvY?oFbv1dPx-&T74`s z?yN|Er(SNG2!zFnpuLy_#e(vyHLpBDGNG}AE{299h1HJx^501B z9bq(U*hjHX_?p#Q>2!&+*AbW_`SA|sS*s^?2s>>{)|ll)CByhT z2T2njiKHAK!^4vlCKK=WS#{eaeAm5Gp`WBh;vk#Crg{I#)l@b;pC|PgC`3P*ixvU= z_fiD%H)vY|9{)PYa|3XHbBTt#ll=j0z3CdYKZ2Loi2OHbD+=)bZPc`a7fCSbKbQ8% zd7CN?%NgiuMdjupmxA7poAzD7^k@o>xN8RM6!tY5i^y#MSPe zTF6?vxjlhiDeCAoU75-ISPEtofPn)0i=SUz7efG#?FD?l?ZzVFR|Is>#P={@fUp<(+z((@Yi8#otZw@PZtsstr=m2^ zg?vJ^n>2N&q(qGCmgY1}#BTh0YkmPbi3>16jBC4`ffz6-%oPVbH5HphCy3~>MK*{@ zj~`8QrgabzLU+pz523$=;tV0ijV>sf(*cW7^nq&_PzAZo*GQ6IPn4p?Bf~jE^BehC zW8rmh>j*I!9aM}W!$BhI_ID=hXg)L5YzSJa4;QZ@J>SU;@z-6TW*BnlM_;dyUIuPQ z++Sd9IQI&PW4wgRPbPh)8yrGG2U=yM?Qi7Hgvj-+X20w-LO}}2-C*H7U+?p06EXYz zM!i-EJTvq%Y87-f4gDRh_rM4Pv(H?_7t&pFTP-5KPdMw6t^9P>&xpH(6f85Lw=D*# z6Mf*Ss@zu}p@-eH=!~YxIVpxHCoK#ChdLULph+1NP)jA9RTfTl8B{?_#b1?G4Y{YA z9J4vA+W$CLv|1>Tq&%YH?OI9jqyGfjZpw}f2P5z445uG;!nHO(b%X1cszgW-U)9m& z(bJ^NK+-hn9~UYKX3>^bHE?HTc z2B}RzjXuO~?tIiIjA{2Cc&hG<-=BJC*=`gzK0bb-&N^$UXCCv!4Ocg_oOapk=C$ed zdva3o2Qma(GNAgi*U{*MBi6&25E8?b>GR|aX&VPGii`!%(if(7DnEY?PpM6fWhJvd zmFx%dsDQuh0 z=Y!XN5~sRXOy(XN%4G#OZG>HU{UyKf`S{0;{k9j(6anTWo@#3OW*0Ihc6*YX#orYv-!VJ;3pH z1?v<({N=rq-HsK8J)coGupcc0+ZKdm)t61BTh8^v93Q#RcpPu8%#=1%;td=LBy7Sb z-%+)XX(7P7#%k+sOa)j?+(!cFClTeFb-uz}HJqgoqz&z5SQrZi)vgSjS(w^lQN$3b z7AY4Lb~Br;Y-;ZIXHvt4$?i8qo9CgwiDZ^yGt7Q&%U|T6Qz{BV z088DXuG!0`e!+Gg_3ZHiWFcVU!AlvRfWwagL{JCW#~m|+*VDc6pO&a-guf7LdbE*u zCupo}Usg`AOCrq(?mE(Txi^V6NFK0xVx2E{lXKBJF$gBTl&Ba| z$mwGfs{A#9@XLTH_aFc`d@JVwsonS5n<<4JO{B(F&NmN2BugMP@_vQR0 z)Dh0zp?6YT$ZSQ5(5eS_ zR~Wi)?LH1|^{`rd#t4`Z4m@MJkRw0N2XfV)8s%3ViF^B+1@;rbMOalf-}p`aa;+oj2l zGDf>_bni~s`cBOArp&u4HsD=0|C-&A|_2ELU^Mnaz-=AcO@Ixi=gkYx;;i=qU&!X9Sg$Ca|%Qi4ZC2 zW_HSQ5YDTOcWv{|GtYrAvBnO3qXJYgBi?F^r&|L#AZL!sLCcs%_4`IP$C&EDUn9+e z{G;TA-JN*2gu;Fu*hi!b3vWh|^>dMFovo6po6&}`xx9tsHrb1tF&4d;?HCI_+)+o* z%N(=UX>(eN>iMzGrfdRl1rv=b5+q!M@fbl-8B@)KXJP-s zu&54yFf0nxD%BB&KK`%o{{CTrR-d2CLAU4M9k>7f&&K)naM3?8tZU9m!@k%_(IuTw zUmvI5-g!-k%s(@%s-&!dLS9Sh|6*7Vx8=_n-}ZS3Fng?|BNQ9aT9q`+UPJ_r#o&*A z$=RV6$(%g`ykO6hTRxb=0|x!i0PXwF0gC$90QLAYKp+1RpdtS*Kqbdf3v$vBo9=+2DY1)|i znYMel&mXkTCVpmgky+(Eivh#~nD&6X=;QwK2tyP>698;35%QUNjpV7`)~i>xY3iaC zU+&BF164e-!LV2=(Q!?`))M^+Sp zbc^tfUCZh&N5nA%lu-_y)tx*b zZB&nvWQd2Nc#|0_>*V)I^Sh!8{eJhb<+gnSfivk=egv={y$~F`UBDvW#`%+Lr?T6&8ijHhgW~2G1U}Iic*{pnZ%56LY!cKp( zqg<<9v=epUVti@m$6NQ(+Oh9<@W$ShBc3f?s^f-0tTKR`RWN>~l2!f+vx)=pMi4~6 z0w919Vrpp(#OV|g0n#lft(&=)O2RA@+*zxS#wHRp+t|OXx$HLDGr_y!G>^*t4{A)I zpw#(Ay1dc`9Nk<@D^i&UomuV zk-RNFUrChpXf$s@iA>X451CLaY%UJTR9Ssh4=Mw4kybVUe4pDFm>NlK8!encTRj4Y zKd+$tqf{o%JJ2?p-N$OPHWh(@_1Mbct3Qc8l^i#=?#P%)y*TWT&%>oeY97;5f!J#u z#8$hVNmSV=_kZ}sjo8$fST$bAOWSVcTQ@Tfcx-3AHwlti+dvjfct>`2|GF>YLErb= znZq~NAo^IWy?WHHH*T@K<2t_4)qgM)S8s46!P_h=lc$dWhA0z!I}gRjsl9EOCEfNb z(s{W2Wv6kbtD{$4pDFcr*6e)j z4>`^`ojQgy6cNbSo-!t*BR-@%`;OfN_Y+x^BDtsFvZ55-u>p-c#if=fpAz^Ej@m{% z37*k94WU4CAT5o2Q$QE8^#(Pn1uM%9LPM;lbs+2I*Q z_ZwyIu<KSND-dEcp_myD@Ey29XM zp?M)h!vrKPDw5xQV4tc9pO>)K`opn)$rbGT)3N@!o<|u>%>0rsaz$v}ke%*cIV2&* z(We(UWBiQjS%5Zj%k|7erfr!@sh%SvZUrQ0F@|K|Lgs|%<@g+>PX$o7vxZ)(>@ZV5 z6crV*jsk@HC+CUTaz3>fZ;r#p=kn#FS|rI8Y&GsF9QmW6Wu|O5?|o%PCf54ZDoa4TsvsD0(qFm{lImI`cf8KG$Exx%%HPK57 z3A!#ai-=PG1)1FK_BmZMutJAp-uuVTw$7m5bL#=)B5BhqelPDM|G^#{ z0>A(-0VDrO{r=xj^=Ti>|IZ$j0#J}W%HME~PIdABqJICh2M(<};(xaXZe#HAQJeWY z*}#uUlqYhCxF&TE$EWpQO@`qTYND$F=#|hy)&iZ*=Zy7_u08^dmlTvC?>p!xsV*X+ zp0JXu1#6`fsL**tkwe+h^1nIYgWnMN+t&Q+zd_*Ls~>ujPC2QGQ|oY~uuSAsKmjb9Ek(b}v0??}aSe@DOq z&Z7h#b(L?2qBN1%IS*~`n*_e5Mo=@~Y5Vj~8o#2J^3NGE%$?F^|3JX5drs;8sk-O{ zgy2mk7;4=Ng%lR0qiK|1Esn7YgQpC$GgiCozz-)t%>p{&^yx=Rp@NtUBC=J+$?fvN zjWz+w#8x*{xZa{TY9OJ=EiPN!vmM{^GIz1qcI>pnQzklK1u*-6Y$IvcNb6@V>nvqqdN>G;6$~z5!y*0>NGk66plgG*sIw*k%V#p-^YyU$>$Gl z7?n#oNu%k!@bl=l7KD~)DlfuN-ms~gZAuc7=hwU00YaaA(>ajRTdLWcQL;sN{lIb{ zydId=s@cxXE7T>G8o1Rj*e4(Fb>c8-y!(tswFpvM3U4_;`K^8p&*Db>j4zx^?KgKR z{W^?G*+L9j?^6?!k3O7;(PB_v+#f}H;RfEb3w5aUfh-%VUR&sOnZKtTKfoD|b;a># zOdjD>(f}9e8r# zI*Uk6slE55s>Ycg*~b$!U?fjpp;)6bo$AJ1xQKgw(W~f75Dl9{Dd75z%3vn$aWKJiy zpRDb3MrldFM)S=nCbF$boL&)fE*S|qrN&n>^XJBW|6MWCkMfc1S1>_km5}AE@=k@_ zaF)*tvI52Ih|c`T69>_l%4k5{1GBfN13|c!bq*kVRS~F5D?Br+dmf6-F}iT$6*N2k z;C1ljY4-hU01q!Bd^tHuy|i4LDDcEEVmO!g8DJnnQWXamdIVrWh4NXW^{)M#(Y62H zhCh}3419{oc7oXNOf)T~yjv1)T1>o+jmUwY=yv9iG*z062S4+yj2;cM z_78PCr}zoy$DLxuQvpRJ&zYEH48iz*@B2MEnrs|}JNqst>aws&0cerG>6A7&>!Ji> zQ90j1c*(5lA`7fB6o5Uk#Mk~_I1l4TQnMb!Ne0j_Oc2@PFMiI?4p1uYmM%B*WAYgL zHrnl6F_#!P_>=T;B&8F=A`F)ZTejM&@mr{J~C7|_9|8lX8q zh&VdaN2lia%$$MQEFg#TKY$2&IHI~(`OtkyDfF+;X2fd4hKTs-i0j`vOjG<0fd%RL z=Fs&*@1A&bWh3;sYo`kBz5t+Z_W^*mt_DTXFurcN?yrskMG-WANv!-PXL%6Rx`IT0 z#e}Jshw>zZXIxr1?Wt=iN3(?oj^>K-z4W~OIcESekym;6mjZ;a!rof2@*F|Vz^R~t z@P<)-QU4133Oa6Ki7QNok#i|lRWy9D8vS@BV~W%Wh{*UZ{e@x4QTYN@Hq{8)sZS|!k&7)^ zH#)gCkC(OTV!WdadNL8P(_dI)jW~2(e+m~G=#8D16905K-*)>M=!)BnZkTBP2CEu7 zXGa^XxQ8S{PSJ!s7;;o36{u(9EpnnmO6mt5X?^*y1^V3N*5MI>!g| z=7RBYJF+f!-}AW>G1u~gp&p^+G5;l0VgkgQFg2Fbf|5eMC=P{V<*{%}&3+E2(AdC{ z`L&uXI)N*sB=ZK{^Ny+qKHglP>;Xb+mj=qm4E3?N?X8F^NIALROk^)lki=!Z(TjxB>o3OuDJLSn2Yr(2Rf#9hyli+2;5LHkn zS2z|x475F{A8mQLS-+R-xe6dE5h68lwdZ^201^tTID4{Xam{1Lhd8{Wu{iZ}4H3`M z5j%D_&~Rg8*Rw-1{ftlPfG%ncKCaoa-$F+(W5(q^ z8z{1vow~W}l^Vd`W8#~Hb6ZxzDxE3Q$OefO+HGl=W&%22V5Z`H&d>BKrd+D>Vg`7Z%BSF_VnP^ z*lbb5E;Nd za=(`Dp{UV};0P_G_iKfsjg^$VX);DZ?56~}rF6OwFt0+V=XnJGSETQ0qT|;EwoM$! zx3wsbv^9)%rN9tPW?;!ekV`tprAWWE2mloZGzCY>a7X%hNghvf8EED@;&P~I>QEC( zq?ZLa@|C2$C~_jtpcO4*jdg1ov1-E8sm4m)LC3i*EhHWmSiw&opDfVh=s-|um!vBZ z#?HY*g(sF5R1`bBI)_*ub?W83iD~hT=}uNIsS^YO0cw8%F-`L6xTFeroQ!&PP+Q%* zZU_`f0tD9+l|+F7uJ8eWUqAo=4FCWG00aO4KL8*I0K5PIoB#k90Kg3Z zSOfqr0Dx-%-~oUF0H6Q?sJ8&*-(OK4Zxj?XB=D~#_JDwU;NyjbLCNudJ<&;h^D0oNFSuNZ&_OmqxPOj9gy4;FYG1b6_U--58(A%If|Mic}y z27=`eA&kSu7{W#K$HlbAg=FI4MB?Ga;o*nlL*wuXq6nbT1bCka0Urr5kc4NdS zoDL$~Bx0}=F`*I(P67!|A_>%wgd9l<*dhb9Qh*T@xIUDaNJ^@fr-c5`Ah%Qmvs8rP z)I_J$q{cMVnKYaxw74_$fO$rMISUDrlL*O+AH>V6&nIXl%$O-7)Mx|%n7`n+77?+2 zF5oI5`w>7G^jtbzm@bBvF_T|1n_M`Xg*k_TE?-o>fQ`G5imr%-y_lJ~gqfj~iKUE* zv7C{mf`+V?PM}^$qYIC)pOJ8c06K{dn4)B#`iYz&A(|OQ{``qt1LCX$FgJmqEdXe1 z(~P08vhUBTo*ICx4J9 zr^u7@qf-Fj^bC3Sct1RxBkmsScd>)KI6z)J{JDI1KmY)U>r2G_0|JS>ez>?nT;E*Z z+}zyW++W{3+}%7p-2QpEJ-xhr{OSJa_Tlah0JuZJxI+WnAs+5-FYj)z?r!hy{(5kI zclU62e|Goq=kDR*9)Nm}g>}Dyygz%mzj(O6y1KtY-d{i5-(KI}-`qc5$h&(a^5HKe z;XdF|Jj{+i{Qmi{)BA9+{cv*dfH-})Mn2q}Jg)fopNEI5$0Z*gkPpZ|56E*Q0D#12 zMP_FsbIOo)t;m)(WLqN=-i;jYN6t(m=N6E2yU6)jJJ2g4N};wPd7Va=e_}sE?=y&|ITE0Riv@^z7KIZ7yI*d_N(LBQ~>P}(I_O9#OK!Y zGk{IDghq54m6X@UV3SN&=l$*Z?&$X%Ve8Ar z;@47VF^-dl8F_74&Z>lEcvO>w)5kKIdHgIY&3{L-)~n{uJut6=`Ju{YGm-#4GaM^= zxw9E3+1LYSl_ib<@swcZtG`ec+}%oq@nUJbIBxIVdU0%lrNQ5n1PjE;3@3;cTi8mB z6SFBU`<>yixEUjKv!kWV!epesEPB4SNh|Txa1c9(`m_EKU`K?T+UivTyLjb{`E7`E+f?>QmHuI z@%Q5IOxJM(YiP!DI+J)ZD=TwhJUPgwxPN7yxp*j$dAD|q_5E(ayyMMoiQ*!UU0FqF zMR`TdXTvYHZ?0kA#Bb9KINn_?YiWoT`=1h3_v~GNi5!%Fm0z25l(|tCccZdWj{+Y3 z+2F$SVdwqQ`{&~Uk)Xv@e+*8`lFlu zs;1YQ9;8k8Y90uvJ9yqK(Kl|v4x>v@!m;b=)w0K+n@~xS&_vT}hyqu~TDtSf^+mM8 z%IMaZ$sgaP8LfwotDG?L){XIP-y4>)iSg$L$0M=%9t~ISFC<$2_yO??cSdjrW)@XW zE>R%H!T|4gy%;M$2I1-#V^mX7Yv(<49Vez5AC~LZE|^QZ_mlsS1Dsl0y^BCQ2{a;B%AOi{V}ya1Jf0xx5Z$AlJt}q$QMH*kjS|3WC~h*wySPv z!k4gYUK-GeLS;@(JQZiKDY9I+*K8*K^%GI*pZLv^%qd4cvHfYCGs1S3URO^A?c`sv<5~(Q;ONHqo;}g~j zHW|9MamUwB5@NOrs3P9UNe#qDg%;A#I+hOc3`lEOI&7U$V}4VVcS%T_{(Ql`5J!VP z;O2Wmxg(i2=ogI7y(w%;Ri-#B+36C`t6D^&hb2!L%R^MXmB*fasi@fL8kn*5YlC*M zLrwEXs=ok5K)S!T5mV5&KGWxTz8_?Zc@=E?Z_^Xvk%`5{^i#-IyT}@+c^}eo9Cuo4 z(&^mb%uy+lz3ATSO_C^zQWMtd)HlLGTycq|GBk_05zmT&7!5mee9tE>)Oj*vmA;D| zT#srIYraU_-Nr_eMQJaO76h>DNE5JdZ2W3k-mrO3xEn8ZMTN%M#P=~fr{S1^UcZ_C3~4<^676#-TGS14tB>E84F|N{N)WZ z`_@$%q+N~N)vw0JEg(zE{ZuCQXs($#;&JMS2 z@lyJ#;hm8gS`80$4+cKgKd-NUCOV^S1+Ulfxj zL5J(qN6ua+r1MJ}+{QH*dYb>!=)2-3Vwwf$nZ2v5z4ZG7 z9+{tO3T-JTdU2Mq>a%ULeZAl(F5~KDz9Z)ezg5LBZ^gzUIOxR6Hn3V01ds7uJxxCUj6&zh?w6m zjO{M;2N-C({c)G63CeiA7n|bT6LUa>x&HD^n7?Mavnf+jADcmYNi-^ml9|l9_j!CK>WX~sArD|lqL%L zlWNyZ94Y;TD1aGMnig5!99g<=F zs(mPm)&+nj2~eQ{0HJ_!M6@Vg3;>2gjt8W7jhRo2S!|A3_KG2bq5L3?-4KplCj~y8 zjM+(x{fUU#QjB#wjy)!gI~9&Q*N;0%10d4k5T$^-`M3u}9AF{trZnz}3E*2A;b>Ys z)^!YIAs**C9+wPQ8SfZfjB+p^OPc;^1A$AxLifyogx>WN`8NV{SOTMW0&{u-YfA#t zH$u+q1a7iKUa~~o?r2nQ01(LyK%oXe7ZSy<;{lSNem;q%Gf0wW!6PzxdG||K);sC> zdNdT8tSOSLZIG<%o%~WH(Xb`iFd|+gBJugRLkR&~4A8ge$OSu_ zDeJ&b8Lc9C?_D!O8Z+TofNGH})t7kV!kIu==9Wk10WoPeR#wMuCeLNobb7`ok&Ku; zrWav?u`jdB$g+Q=>#iV0J_mbe$E9Z@s( z3`5*s=%kd9pU6Ntbno*EBR+fa=AHF$(iC758seUX=5OlfV>IxPU9nc1fbdBmZ2%l5 zp=|!=g%0W2=hgXCd--1h1!uekvg9#>b}-@A0+Gdn?lmxzr~S!wq0#e%nqQC;C~h+~ zcC0_94;-h?A0nF%9+@oMnJapMS2Qlk8VLYVRs$sg0m&vHBn;eR;--yJ5~v)TfQM`6 zg7w1%GX#L?j|AY(0dTV5#f~6|^YcUHoE6Wl98_G5lGBO;3_y7nk5&l9alXQp+RkcQ zti-1w`{57v+X>63#&%r8={?46iiEW6k&NF!J3wF#?k^MoGzvI?V+}1GjwM}Ed01Yt zdQ+8vLgX!s(*#F9@>g71teQHhT3-Yb!vV^00OXiRmK$_qBC~jbd&2rP_LFKpJZ#hd z>rgK4tzp?X`4_Zq0N!yK82~6m4eoZqb*-(&-mgrc!SsWN{h-FSrUqxZ;C9A?RNuRt zAS26aa)DZFC{Ip+T=BrVNl5k&T$=ZfUY~528q*Jxh~ovg4vu?e;&yFY_fTGBE618F ziHcW^VzOSBag6nJhDFZ4zI>6?SrgO^M|YS+e-A?+Fu}c@6kOE99nGlG-PdXaKqg=y zbr_n03z|GPXfhsjtJD~@NP2(|SuueeO*Rj>V0S>9?zx*hmMYyq*p1cRwI-0P50K$_ zo)G(H7elkVN$^?|)*1=8;qndb-dBSr5;IM(A34MzBRBs;b0UggSw*XgRa#YMjlm!Y z3B}off!8NZ!(H2&qe%OJIBRRHlWXtK6q(5dfORw>Ao_V@*E-yR_$vHZ}r)cLdR+3|<2w(>x=OOi|QS43u)IZ;=N zeTOg~Ttit8eYp#>odu){SDk^E?{)DozQhwPgp76(-NE-ibssJ15@wYXS6ZPy>0xf| z2KvCyGU3dk`K$+3zwMn#?|LZ7d+%Ah_^z3f{IDiCz^xp}c0h3a3a!LHn ziWke4Ty)clCMK_Xyt z=p}iWeb$B8fYrkAeK~UL4TM&FF*L0h@elPigU_s+gFsw)l^ao9<$XQaLJ}z=-<>D@6C)+$2ww&f z$Xf&^zr3H`eWN?Z-<$Pfpt+Yxurk+dn5}b??{}IN<9jkId`SXj)>;#{QRbYX(@2T8 zD_E9%S@ghh_QBD>xZFHM{zxjk5EPwXiF2MQxgLz%OyIV(NN(qINJEr@0COIMv4-r`<+*TM&*9v{uc>vC;efziuPL5GGjp>=T0cDwuTtc zdvgp9V{JEeI{DHN-Mc1FdGZ*wqiam6{Wpw1=BE8n;i8EWJ?d0fhLtF>f%%JTcqiJq zq=c^6&nIO`9M8WiGF6@bP!&2npMoiHTuf{0IbO`@TURy6A>STe%)Pq0yO>7?Kjm4l z$b5~z=+Jc7xMY`je!2X%?rD>P$9DGB&!@ZXhgVBJ#^+aSHW=dq>)}g|!YUC$IF}pP z0-OPxuVtRKtR%*9;Qda)-SU$4+XRcO19fnoFB?a73hzGV?RUT5R5%I$vXm%nItZA; z`F<$ymGgSn19j^5;A8Kb7n|>2OnLA6QGLEU9hf`3+n=QS>V5uZD*{{wBnxv+={?V8O)QRZ$aX((X~$N8g(Ztwz6x_QQ#fmJhd7MnGoCCTO+4j0iRG zWj_2#MqvOE7Uq98v^0S5pAB7Y&}D^5fNwMe>*)%EQVAyIYOK~-<1^V2+HmLf#DMAK zvq~p2B|ee~MF!9RS?dn&$_RLih|3wv6s$5xv?_a9Aimq^NWq9gDQq*T&Z@(jw=f*W z@uDY7iC|j|Yj#>Sadzlaj`&6>u~^1i7OfJEsm}bH$-Ob7DYtA|!7ute!bfJVkx#G* znd?+-i(j|AvwGEZ$NBB)T$%nCBZuX5&5&uXwL~C>1PA1_dP(bhq}+24`{icABVlZ< ze#4n+Q-3~tVxRvj8kIz4tH#_H0OZ>=J(bybMMw>+( zzrM7-Lv(G0rgH5q&pBu3nl%@yxHC=Yxacf2y>OpBZ63RBc+%+5qG8-@=K`-8Ah&=e zYKUX>6Yh_whP7V)QGLc@786~8>78tpA+p>~X5MtFWCU(GzFRC%8nh`EyH7j6Rf zO7m7QwTQu%WaYJvr6$KGecE6=*IsOH_*$PuDnF<%4((!^4r(v|#%S0mfAJb@`CRf(G_HBO$~#t*U}%Tkn}{(e_7#pyXr^56K}4uB2N1rYtK{C`he zSs%-v`&fPoWus zRH~aUeM$9p{=oZ;F`h~ynuN=$o~8SDR}_DCYZhW6Bg6IwxyZ&828F2aQwCCjbj|XG zIvS?m87cdOHLm@U-L9Tv6l(9!hXmzDu73BaI1YNSrm&Gvm~Zr9RqAOtuMV>EeU8*+ z|55g8bD)aH>W?8iR`gRDzbqcZC9@`VOD8^iiu(ST3*xP|IR|+uzrx*>*H)PaAq7S1 z=S}J`9;3FR?j@@eq3$&shP^Z5FXWWM_OnT*B%@2!bK@h@z9cIyHdQWj)!#dG=Q68I z@3Im@v+xL+^|SnIBKwi&R2qgleykTZN#+d(Jn~uvw=pIiDu-F;(qWPy+SAS0u(y?l0oK+p|vjU zbZ{R5x{$o>$kx|YmFU%QW!XxdKmmcNLIJm}l8Meb)Tvs_+hbPaPZP6?jzfCfs|#u< z5|ouq)iUFeTI_4bC_|5^NuebQH*l97H*{B{@Qj>qHtSz&j_PM2=7;6#HX(hz+S=WN ze@|UKEnPBQEJSSp=@vzzpklC7@|#Eof{HaRMniDL$EwhtN~CVN4GU)zCaQ2cG@D1V z+6FK^Q`#yUT=o_(!$_0UStlYjJyxR5d2?O^2u5Y{2LRF8r@12~7BLmqB+|(+f3kch zB>2YC6wRkjapb}a@jb`pi}r$gW738ZiLjU@xyI<{N8F8;fS}m*n9}Gt{;*!2bZP>1 z;b%npKLWlp>yL7PV++Dwz$JqUS(SOyP2W=pviwT&(`liIIL=NZ_B9oplthmT1O)Lc zGP_%42uq86QVc}Uj1^}UOO>FlV7<*Vtxw08$_Yeo{rn-AQDChCF^ZJ!#bn}MGsEK2 zmja0)cM3Fw8xFK~lW))KVyXMrddm2ZhJJwg^OU*m?4M}{+m#=nEKYu4=y{&&nJ^iT z0&Yt2k6^!8tBVtAn9iT5e(`!{LcB{C6@${g@+JBUcM()OcGSS&L`9zr((er)-(bBI zg9wdQ>vpnBhhXc?n}i5W-X$Js_LsLLwD&i@!WKMg4vG=If1s$A7xt!O&$**yr|oHC zxr^_`@hpyDuVt53W=0eWP!~O`6aA{#e~G%n{j*_)73MrcnKK%Ad6AfeBT~@Nr6i&{ zF?T-N3t_@*kZ;tX z7mF3uwFh@;x%oA}oLiJ^ylSyOJ~*UietBiue?s-FJebJ>F}G-w2;n*e5vX%_ANQy% zIrVZJe^0@crSi&@*>7nC-glJj6ueSK?3RwaQRe8uL~T2WpL4pCJ~b17_Toct&ec)v zX6^%&E@JF6hFgrRNQ1GZlA{cOts~H-+Ds$#~aK9FD2ln91 z+HL<*ZRTpK?tpvzXMd4|Ouz~~vyY-{9>*&%HfU?`&uUjBHOtKz$texL4N#}dh-Gq9 z!jD?uS(bk7@9uDZN5S^yczUV0=PEapWan$&lGt<_-+E6LO_it<0FwbTwrcjXg$8!h zrMS<%&n^!$4U-L~zJDoDBL)|g(Y0iHUKwn9msMp<1gW8AL zs<8%ak{8dKCn!2wB(CgMAY>L_Y+KavDD*1lpLu4TD*4ODt!&NtUiXBKwKV?DNR^J; z*Jjt_6ddT0OM*>Y9BhIa=(z0LuBS3e@`<9}?#(v4+Ak|Lhc+Tk|GevdfqbX_=3T!{ z-lN!ibi}ZmGDlF5NW2p9{oT;J@B2nGp2`-0?j_H`{nL6a!1hy<_G_B!zq`%c|7|4OXN?1SZ;8B8LGBqvIWiKBL2= zzxiJxacb#iKCnw1WcBWnT>K&}AC%xM$grJbmyJbeth0BHr+PlU`H7`iTVTPGf(X+F z)`7Vdh1Pz28ZQCS{HC2&uLOT7c9GS`oE0m@T3(n~oXo;E%Fn(v74P^|VA~N!d{bgE z#!m<3#&Fe2^uCNyNU2K4j@s9(k{>Yp+J(OKAG#m^2%slj6;CIVdQ-W?I6+eFdy6LiVYb>) zBB{~g9=v6W5=6|8>6-wnNNEZbmeS3(0~=QMhL?5rh2idZI0N(K)#ja#*B&DtsJ*cm zmZuVrI5r4TBWVt`TWIwK^E&0SJrm`gmkdI$_nXlrgWP>T)$EHfwz51RJ>4qltTI$f zC)+oDnO*p+CxlIpUuZ_Zp?}y*z&@#lY!o>#DXo+yH<5y}Ny149+xk`^=)7_we7E&U zgV))Qbgsze2VqbYe~HOpBpSABSez#x2Q>vcw!@kU;ki#jgu$Wj+$*2cu~Z%3OAf*u zW%{F31NQQ__+!E?e@W8!we@jM%uBurQAm|wA}Wpp^&2!|Vd^0I2O~_?P!`wuys$fe zQZs4g-$GM9VZYo)gxS@FVxJx7ZfM8|{oa{XB?jfMn}zRLpd5azRZC<|sUU@BF}acsIwGv_%=x$?^& zh1xPUc{6c*bKJf1lny(e&;{X&@H8pae6$#}GG@0BY`J;B0Bu%b?F83&XA{)*?yaFi z=^#$UfyAibk_O-0qaWrOFrw2CSao^@YvLnH|3*7xbNY>~MsZT57$**4}4@}jwxanZDd6S!9O_5i+(3%XatGtN|yQedQV}ko5yk!v7NY>TUcgIsK zGcs{Shck7mTtXPh0%7SLx!8LH=+}Z|3zmQAJlQAH%s_HZXAmJ}x2dPndvq;VE?-W> zylkr9jcA!rHgp8PsxzS!x-Q{|%Vsqiep;Xy zK+QmJVKbJ&uM$IkA(g7O!it4Bp6S)ADGp92HSZ4u=}^ynRL_&Dvjo#fS549je@OfB z4`07+DByFO!0BjsRz6+9pQAa}+EAfa^}dF4RDffS#cVT@C}?y1mAHO!y{)?)(br!i zf!SuCS&kFdU9%&9kq-{M7RM?#Q7WeNs*@kBd~KiP!fIHfSmG2z(KJBXXqQW?QOurc zwNl;}1&+3Ap0Co{46kw}IY^(L`4&Z|@Wr47?a}f_AgPSeGn@RB5(j3FGQR45y=UJv zVk#y5a`I!|^wlp8vCfNwZe0TXNjOuE_==)GHlp9qk{Vfb(8XJhxOCC_Jeyc?(BQodh|`r3IYUhn{LP~k=t_XcKP=>*^`GmS3~mjpxjCjppDlsu6_X$4O4za`F8^1z(JYm(^H)$l;fq4VBv zLLOI%1pB}Xm33;y`0*aB#>CTzKTm1SUmW*znMJZF2?z9HvqYPL`8)POcaq;Lc%#`1 zqs>OyYZc`xe@>p+dvhE|(T$6DlH7`IoRjk-#qr8xdoVEO|xP=Rg_kzlYDC;cn1 zd73oQkK-Ude!YL!_0UHq2>JKIuW$t%56Ee;gGL zCmM_9T9jlQAa<%jU|<~2NCtSpI=b!V-LZC(k^F{FWN$$o38pd>0YTiGd*=_2%twm;MOl@&X+cU!#7 zf=|||I%0kIGW%?@IAfjp(T&VTj@Y1qP-3kVKPopeRB1I6)c_Qulj{oF%YO5@e6PH+ zac{4pdZ3YYzp{STZojJeclrL8_RGEfuW)p>gX%uwznNg)?zI5xEo0r82nA#PsuGx? zVbma#uCc{dg|^B6{eBi#C%~Peh*61^(|IccY$}uUWrTvedxy@sqWi7;BNJTi|LFqE zVHm<;II7&vcEt(9-L|aPs_acX(dF8I(CJS4UP=|lWsFj`wOGK*8v0<`U(aWDo!8LqlFnZH$&{1pXt@Of=XbE?qa}aap@6^@+$K zS;;25V<5uNkeXNoBwcro>{e(4&o;U9cEn z>QOhEU$i!#6O9PO#G}3)g+B#ld9c71T$RXK?cc-gd zH)lzEb+NT~r^Yvh-maOT?qd>D4#jVA-Cnx<^u>6i`~6OLlmm`e7y6=Glt6`j5ze$4 zjA{VrGpqb={gdf#dLY~X*{{A7@xSH{5zHw8F+%rio3!GDnut-pzfI$Lz`smm-OF(5 z)9FO<+e*4{{^nd}>=*Nu8)H%21I(+@LD6d{SR#Sk5N^wwG`#R}Z}SO>d3YW=g-cd| z8uv>3X{4a|C{nlStclSglvkzOB2-;6W&yx7A>~zvQfg(B@3RS#I+>O8qi+J&s zV7h1_iQOOZgN_DCUwt_7Wz&XvXhADIg`X@dd@c=bH`@gl9oudj6| zpgw#K9gFLm)Wb3ktOp#;3+oH2SwF)Tx1ufm+w&WyEQemGpm-As-tdX9O zGhJ7%o{VsrsNbZsE(e`;|CF2jio$qjKblj@<4e(h8 zWknX0LbM=!ZGM)ij+U#H)m|C!=w~n2$ys)NAEZ|g5EUI=kN!lfc;SupDt}9uID1#K zYmio`$+1fhN+^lRPCo7%d|^JQ4;>l(yu>*QlT^^cLQ{rRuSLENNp=@iNG|bUZ%ao`e>TU-01jQ3)W0hblt|#$je*_!kb7myu>1rb7dz_O7OCc&M{5NcWr{`+#;!0el0hQz0GL$zICX2((8Xzcjk#pm5Rl!DnClz}?F2Xa0BvMXn z&eXP-F+w-1#wc<%M96pprw|AA zT@XyFK1HqPocV=3V4?fvUWVg>u2CH7KB)N%&XghZXdxXbD&LV23j)Yf=k`wu-r(GmmL<=BH|;a%UN$wqBr zikIi$4_MfNRiB8SP?QQ-M95HH4O5}A*+``*!Gov?l~xt?rLBj+2JD6DBX7LoNT1VN z2`1&uxa|mLoxt1-Z)u=dp3Hr~45;P5XV1Ub!3gmDzXwLumMXbYSYI=+gBYPyT|rPf zWzqSs`Eppdqz|`D)f;_r)VMbUiri*Brj!iR;BN`}a&cq~g?HOEMPuQ)Kn}8#jVFXk zWvT%;=s5afJ_y~F=;j*RsmRIPlg;gVyWt|LF1&;q3#hp*eJftM zb~*1t`4ckJpb!WSFSiMFAh8^lYr;|?lnK6gsWWehgJpB<~0S6tv#hjUq zl+4T~DIzvyZonrLR48|pWJshT+p=Qsh)5*TSY@$A@?$`;Lz<7YGL|&HYWfBShf%X% zY%kX{=NZu+y*y^=CqQi@lu-Jj%2Nd;erA__%ZQk$$|@OEv|7N!YGc0>A)N& z{CVbX&*zu=i@}n>7Fy@~Q93EMyRF_h&WAGQMMGWCuCdsVB0NTXNwHcrZA-^XC;pen z6;%AQl1ZHS{Q>H>6j6enRd&0Y&P3j~ow|~ok9)t9a$}EvFTcb&?2UF@4Ot*HkEwlY zQxtuZSXO*1wz&J((fofJOZWfXShgQAAH>2lgbu3v|8Eg9+iWu`>PoiXQMQi%8WOw_ zOYu_T!6E5OD^RTOOBtPYSTC%3Eb{>vKG0Ii`KXLf{pMy)U zOVvjlj@?|}dj7pVX4*|_Q%xlkMo7VUZVjb+#36%KlF11TzYS8e+~+s0rE-L)BwUL1YmxNGJ3*GHU+9$0_1%fVj3HpMb09e>Cc$CFVDuriqqWU zB@~s{er1~QGBdhN`n%<(+msuaXI>zFYY3^06GpLAQ%>W0^IdySL$_d=1=WYx;;3_xXs<|xDq zBX^g}v}pNl?hw04$1(BI3@VE4+}(wms-bHsO4Sj3OcUAS0ib&(?jCe78Ks2@oI09G zCr`3l^IaWlUxiD{8=@h2Ko9|rWR?w{96ob>b-k_oEzg(rY@iTxqlszIYt~wNX4Z>x zJ+2;STngfyHTzA>ILmbM#&QELr!*?N6?UoqNPLK-1ss4jN(;;Ln+0uidGyZ6w2+^Z z*bJk$GZiLL(|0Ga_)YeNQ0%^&S41dd-K=UWdgAN-q>>NZblfZ$?#0&BMel z=nLXLDV+{YJO%-eNf^d@5U#{6cMyZ(yhy}U>g@sDa$0#67M<+>*VNJvfDLd1$owC6 zx@n|8;ygLcRKz3a*C*vOb6_-G6*swixQ|LZUf>IQZ1*6=+9FBco6pyo%rJWCNFpw~ zQRYX#8ZY>om_I`Em!yyM)Av!kaTG$<%V}me8`Il8Q(qy zdG2ax^O$XhueUP9#MJ0~ovwTB8FbB0<%BJqi-rY}H3>jfispSOys=X^@!@m7P=Z_I zYk5`vW}_bw%-0R;<3hK63B!?W@2yS3P}hG;g7F-xrO?PEne#Z;gE#R)$?XJlRQgm2 zHo_nM$A(Z_^-z9P1DVAPA-~reiZgYVe~8P|-Koux)N8^*MS_^5Y1zO#({P%S>c}fv`EnriHd*d-Wikn787=>Ddz9p7eEi7Q9cVxthWB*2P zuAE(_AHoz3TmY)DCkKB=CDJC%Q(~W-+?o)-H$qw^m`Ev~KjVKG4IF*ucu+G)Hjqm@ zE;?G;oX^qdV>ik66_GGZJ7_{X!qloSPzd9_m{P4m`tK!htl4PAy%>O)L2|pjdLex3 zv=t6g8-1C$5?PAXQD~R?!8B5|Mg66rsp)1Qrep+f6mMc_ARc#h^N*3<4$MnVsg*u9 z^~%IPRw(~1j>V@g3Mb4QZRxUF7XGqnHWa2}>LuF}WNPkH^TUag-3fygXAEG5WuY>7gsWE-8A&sdYU0tWY6uVtE7Zl>& zt#Eh!NMzkupquHPt^+>@;s`ez1XrTLM39N+qh^~H?;LF1+XXfI*UuW4wajgFmJnOm zV=KN$wKQpYY>UR}Co=>I5;Akg&Q=UQt?>5#Qq1}f^lFXos6%?) zUBXZ1m-9z)nO))ueRmNTjC-(v^>Rp(ATU0JNP8V@ zst89+!h-qQO~Kfuy@c<+hT-4QKyl2O`)FpX!{wuwK#BPO#{aJ$DPjsB@_(QR({8_e zN@XA6n*9DKu(9oO_;>2gIO^+@eJ{NH?r`p;*z0-G6Um3bkTP*vD&3!nj`cnZ~ zk^~LV5tDaoDEQU^pI_-^U{|s-Zl^X&g4J{dy6c|yx3gr9SCc)FYGrj~FXJzM%xdlK z=5oH)XK>-bOm;8|SV%aT7L2E8YV*{Id|Gj`zgBJGo?)r4eYlv77N)I9?8({P>V{lL zt!&CQh9@JRhRzbX_F7v%Ez#MPltvAtmJ}LgX=`*0B3I*+9V5-z8+`IgG2EAe=5S+3 zYMasbd4t^usuWbi?>dKXXu59B=qUcGC$5Ww4H?1h(|WnNIF-272`xEWMdh&obO$~$ z*%_!&5Tj!wtJ5xPrbXdyEG;4Y0BV_Aa=vh2gktq(n;4tikOdnXofYOsN*a@uYKLI> zqM1nr^_!Wo7b{9G5{aVGpb#fIcoDtFD(jop`ME5ED^>=AysED+_)T=hAY2%gpIdM_?Fgv@q>)#eMW7Rz0w+iG}2I_8${ z+xIMZX&_&It8}w$Qk9GfLgoO^mm5kyv);b5-OeU?PtTYRPWRr>YtS}~LUZCQ*NzNl zv@0#X?q-Sq9M4K$7E#@9X3NnXj#{pk<+ilDTk_S)BT34Zs9{?=Fa~{$Z$g-^l5fR} zz%biCT|UFsXpGgaitnc(>?k+JURxnyhMDJ;^D>5SWY2{(s^fW>OdU&%HJqadu$t@%l>znq-eTCus6FqE?(wO#sBw!1*GTiMju%~3{ z+A4{!G@q1IvSA$0d=*`+lJMKwkYR6HkDwNLgjt9N1(WGxno3#XRcklCLD?U1Kkn); zY5O2yx13O~(!b^QjV|SYMn!PmK`V+2PMwnI2R*%dDpziRri{fPJQ;|JlVpv`GTC>k zDRzGSj>`3VcW~BLPRb06K2`EZ#77s>#4*A7U{uwh8c|ZyU;<6j6`7|209jNB_ap#J zJNL?MFeJAu-9843VxZ37CVv zg&vct7`VzuuMAtYV^W^~og%kRJ0{#hi26Hi-63(}15T}ojjnT#NTJNr?`uG)hliak zbB7`~e1EwyTE!G% zzn9W#WiZWFp{<2Y=?F(rAa4*YS7E&%G}8Do=J)EQZLr|5pCfIBXO9)B`2!vilO6Y+gMUhAe&Csg|5v6xUno3m=L7qk*nKd3T&%)MO@5r@rtPj z_VJ6bs7;k)vGNHcwU%SJ$gBQZ2-g8+`-_jXM*H)MN0%SejV_qE?oU|wP--{Pi2F&! zDyH8o_uE61^}*0L)S`|t6Pa=66dw3Kj5%oATkURQ`T)ag5#-EHkssBNwV0}Qbf`d* zBllGOI!+Fr)d13}v`fSJzF&gn}%ZD)h2k%El5 z^jA^s5Ii+X!Z4TNdQUrN1NDgJoaK#rxf}-#;iO>1O2#@%X0|swVKD>jp~x_BVfg^EYi~g{>*kWotpdG z>&d%QT%D`+BYl-{JO!w3pJHoGU{mmC9@+YF!u z60%$S^z+rUM#7ayPRibeuLdu~&B_kDA>3w(7eK?fz>yL%fisK&h*1*hsj`oFT*19- zS8jbw-)v3PKyhjogzCBfSWku9okJLl-gvV)Q5nvvPq~&r(0cfcP2`QM5kB7|odkwq@iYsj#f~_Jz-c>$ZV!DD19SyVsLY!I>Mie@}PD_yAxAU z`f3NoD}sFNEU6G{vyGh2l;lmx(I1NyW#q0VlB2-MPO?9Ojz$jm^e$ zF|nDvR~b_k&E8INgOf4NKR&F7?UDn^-q#Sp=j}}aEN66SvVaFGwT3~h8P}5nXcm?~ zNAtBk<-V}Zszp*Rp3BRYFS6d9_2)7E$`VnAN~8h9LQMhBXF*Vbpyl3ZfoK^OCyuAu zYD+@u@32YzFZ3Y!v#`<(ItB?c|9hM8P*%$swwu+a0NV2zL`~Kj!DS=4D^@FuPjc$^ zt|j{Lg0g{7d7cG_@dx^L=Yyejs7@LZodQUW=Kwm70UrcyP_IrsSCkb-svl5ABnv&! zYYD+J-kuaB^`4W7O4ui1;nFVtk$&~EN^4xY)%Cw#rHuImDV<vOyLJ*r1Os6l)d6BIOE=?OqLP!ve)g?(PEuu8B$7reE6 z`di^kSEV@ACR^nnDg2L@@qz!157!N+P7&$1Ogd(DaA@xwCLaU*^Z=re_cB6~= zvX-?y^+P^It*WU;Un(O95UU{B8j@}&vnWD_`uzkbC=QGk`rmBZB}u9aHK(zKXbgfs zscy?D^h!N!cQnZ1uEpje(=sJ*xgLcdt4dBd)JpyswJe@VaM7kMW>W1@j5Bn#*mwL9 z?kxZ_{k%Y_Fj5aXV%qJb9e!Ro&meS`J2Xu)pL8MKi|yx<8eiyN$ODSPHSKe({Cq3- zri~h=o(*Z4=n={x*h^)7{0~2V9sY|S=upx53X#)N!dli}X3a*Ip_IUN%8VNRWFmpxzG z^2n~&Kb(G;Wfbz&15>8ZQq+hDu$ruk1V2#byl0lwOxyVR93Q@-xg+0TGw`C)*3v|J zRpvlgpzqVey`y8ka=wr)OeRP4=D@AJjp+GqyGPAUm-bsgAK%Vyrds1|CYP?bF+E(v^cd}*P%!RCvN@)?^JzW0^Atq5NKI+Zz*}r zMvo;);~UTg$82O%m|~sck)zP!AZ8Pm^MRwl&mGTm^9j?nj|qMZiQ49>yn0=nx3g*R zMH#h2HVFzuC<8V#2(q@4pP(zXT7a|iTs@Iz*=kY7{k7HNoM@@Fl7bY^wSRiPQYBvr z)h)+UYkv7fY;H1zOT>T@8TlyBIm#8$Cd7v(I}2@o$Q0fhSZ zMlkXgw72Vu$fNoF;%3T79VT6t_B(JuB6MOhZ3+-8TiTQ8hlR;|IVH>(FEEwrXV@pR zwMM2|*3ZXv;pmlHP}E*nsk@%?(#wU#et6R}HbyC1(CPJ=j2q}d`JA`0eehS_S^EX^ zLB{uMH@d#Ge~sxxr7nL{sn}uhKMzK1_o_~21`bg?n$A;1 z0+e&B8L~o}YzbrGF}Af~$?wNPkiHo_TXxbQQIi_>iKlJH>z=_bYcY{yk2)V$U}Xpf#Sgd$8M z{s}dQk!>EGFwkR0vpdAlw*};BCjmPK6G~R{Vx^T*md1k+yMKvj?~vIB?#z?&jigax zF(OH8Fo}{(!rT7z<0wg$PrsaG&-QhQU|stQD;hxtpJj@27t+=T6r-EMFQ=EW z72kM^p6$y_w;;?kaw$L>wiY_i2r9adu3z8aP_DbJH(U|2!y)k~mTNK7aMGcx1%A?A zQlkvGbvkfAw>V@ojGmsWu#$Dx*K}?e%&N#zMTk7*66MC7@OHHBV2I5sb)G8s$1=tqB=kVW@#?yT)*Xi_j?C(1pkYI zR8Di4+JL;zJdn7uZm4mpD}J>o^WnNab3uV_pXifYm2cP7D;h9j#-E>w7%R79 z3UZ?p0JLUTU58H(KXTq(lg6D?cDZtwKsu#{tcaMdO$4#8VFI#kVYzn{|}SL`njG$&&64a?|_Ow z$$yadm{GAyllhMCkgf$bU>E}bp!nWcT}G+xh?KV0?r3~7Y3!Zn@1T!Yh4%zgJkyUr zX9Q{X;|ahK{{X(Pm728DWO$ssyLD8QZND~r!O)F_(ls;!Dgp))A`GFRFoJ?~GlYnw z(j_5-NJ)o)L(34714ttwH6SsBq@=V1^A6tL`RwQ0d%y4AAAi7FEMSIp9@kO7->Ism zB74O&9_Rq}xd8kJ7yuvu00RIx03ZPX4FK4m-vxsK2pE8Y0XP^yf&nxb!2bMEFa&@= z02l;-LjWWMKtllR&))~b00<0#!2mc6K*9ht48Z>UqhL4yfdeo&0EYueIDm!&*q^Ti zMgkBd07C+BB!EN$Xe5CB`SM^i06_yVGyq2fNHl;(1K6MM35*3GSOA6v;8*~O1<+Uk zd-_J9U;qjMpfCUm2cSp*iUy!q01AMiU@#N{hQh#5I2eiqL(yO;77PU-P%s1vfk0sp zC>#PsLZD~}6bpd@Fen%Xg}|UN7!(eJB4JQ842p$80XP&4heF^`7#s?RLy>SO8V<$6 zp#TyJMnWM-C=3aOBcVtn6pe&pkx&2)1*4%5G!%x0!qHG98j40kv1llOg@UnA2o?&% zLg82_5(`CRp;#;w01#jR0Ra#&009RONC1Hb5Lf^KfDvFY0s=#oW7>0nr5HJ`54nrVe2s8|Vg&_bq0t`n$;0PET z0f!@ya0D8Tz`_v#5&=dcAV>rZiGU*!NF)M{L|~B!0F3~n5fC&2hDN~A2qYSTMkBCj z1b{_=u?Pqj0mCBTSOgM_Kw}YDECK*#=*ci zI2eZn;Ac&Ut3^AUE!InS1^#Xz zS0h^~^$3aRFVHWyo9wD6(0h5~arm)%rC5(Q8y|0+T?OB8gwtf@BesNf!xuIK;>xu; zC1Z_#8WeTIA5pQSDy$RT=c>yd_>xXvkn}qsh&AwOxjNMY#h8$gzGRops1};2y`xnS zUpww-AM@@(Hzyg(b`b@MioA)lz4YR=!SS)##j8@FMpUd*J^3Lgan}dWoA>W<*f$$| zycMgm{?KiIqjanR_mbMk_Mm-5$@+|EM2xrZu}^`=p?1&?iUN<(aSvFh~5t6 z@Fk(hXK|z(?@}BL9!s=V-6fmPmKWG*=aAR6TYf8;ve`^7F3AjzW9NUPr}&}$foYvJ31hW-`h5|V}GJU*py=9`rV z1Gy58_kKjMp=&<3#eGs?El_i5W)V!D@yZ>mT5^jk%+lnvNUz;o87c^V=wcudFFN_f zp+>Bzd!sR`Xw|9a1@raXY-j*tvw?Eeo)tx>Dz^3U;g!hJ`m-`A7g`wIziqX1$68z{ z=et&Lp>yUu+jf^&Rit?*`4g&3Z9=uM@1JCoYqoj>lf}OG3NMQ7^ef}qzV}`J@yylz zmdx7Dpx$|E*UuL;95{#ct&5&&+!vLr7}b@p-9=kCO+Fnq&gPH~5J_0u`y!rx`x4r< z;1TzP$KvGs3Ar!0wf%{ymGcJ}j_*;flYW9PC1xUJPTtPI^NKvm9j-YZe*JN)VQ8{G4*7nO=jczdKV?v@j#!Rlic!Wr?h7<(~DiTiDpMMWRi2FXEe7^M6o_cGdD|) zcQ%dB(@E}(ZXBOug2KR`#?~aBU<&$&rOu6vT z;w)-<<9wk_dWlHAl(V(YiEzhuwO=8Vd_IXOHKEk2RW4IOR2 zHh$MFP!qgL(b*w>Q7Z;MWWa~B=?K2H{VE}Oo{e@nU)5|o)@qfP^D@g>?Kk&cWzB>6 zYZp3oYF^po=zbO8IOuw4cAonuZ4ndU84FKa1V9{+_mj2=$k-wMoK1mb3?hadrI{_E zbbRV5WSFz95$7)1O`O}BLiutLoZAtCgWFGOi|9)KT)lw=?R#Cd^4!~@;sg`f3ON+9 zBpQws9O{gjtp=Jj?03g8X{}k>*)A!bCI0@TdMv2dc-s1CEEdCg&O2jSr^IHkiwJm+ zAb79oNS^YpvRE%(t+ThHyQ-iQVLpA)gzbIFOfvTtp=<_xeWqQIVHJk*xDZL;sHpK7Z%rcr^ETzA-K=>d}P z+B<26o-4*0tWq&InhwolVa5%Co#R)HOK~UuINA-lnX%%T!93jg2O09f#`IauWc_2W ziQT(gSdbqs-54MY8`0dG3{8TqOn#4+0MHlV9KH%35o}gZ zvanD;E$EPrH%e(r|BRl>#n+ZL)>}epfuQ4=ZeMrgOo*r>^1Y>}^jHF&FB6oJI~h^r zX`v+69G?RT=6Id`$YA_%Tr;3e&HB+~r~6XY8J-tk(kYXWg1Y`^^n|h#tq{w(DQ+vv zc^M%hEBQI`b}R4la}g_u!pfDEg3`7Hk<~)vh}~*Y^#WqGxNdJ{wFE^X`mOXMv;DWS z4uOJi?|Yt zmQYa+mDO@H$UJV~sj7Kg3TM5>Ajx0F&feYWgWY;7KG=D&XU8$|TNi3);?nyY`&4CR za-i8uW!&|PmoB2H9*NQUantX%Gkv#(EArvqSdP?-tUT3dBI&wZWiJT)k{R2r^Cw|EM4t+>j)+zLr>PCSX0)Z;2aCWs*#U`TW)eZZp*upN? zZ7@Kf0TuB=WM_F^UpTScd#oB#=x)fBYrI6Z$agx4*9h3*i0|))3k2luw{}&mw&c?_ zd2w8G6B#SF8ALqlt}OaeJ2`Kdc;vcVouSiy=lE!Cd8qpF=*#45 z`;L{@oPM|+vlY!3wO)92$8q?=XaA20m*m4)WtKvH0SjDfs<%OVK+HoN&eA?3RerHA z*;A+OcEQ}>BlU#i4cce^^4p8Pj{_Q$e{3BNeUTWW&Pn$<-qj=VBWME~+3j6%#n568 z+f4!m*T7RsL?@DP+9Ur&45gK!Y`m-Y*S!dNAF7>DiM{%zw#edYf7845(; zqYN|2EHtk93&z9{j$O*p*j#hXgllXEft5ervbD6zx&0zVFGul8VuD_X@Is#A%_wkDTpFuwd5B;FyxxQ%sv|LdD%1S{d*+_DxCksWS11qbg5lON@Nv(<%KU-gc zmA442^py-|-Ooo3F?%oFK`FC3R)>FHpiA_)I@(`NWh*T70YGW2)`ewy3nF*IUzB9* z5mIn*ACkb%U5sj`bZR6VC6PppY_HhFo*+jGGU@dqozu$4yp|m5W|_CLM2ujIPpBSa z*YpJW@^~^Ih1cbElb<~cRRI18+^pY$yIGRa6ae}wXG6+nm3#s0@u0popGpMxO+0X4 zSu`2rw0XsTT200)C$-j_d_BpsPr_3?SEoeZ;9T*XtvjPCa+{UTMv~uVAW!?TG>Zoq zF2hx0<1i+#z_YS@;vU$174Zffab*rm}4_Pu6O&VJ7X%9GXGTK(U2aoeME7 zVGN>1J$S3x{E|;2?fRylQJ44eeP)}pm&{OppPr2 zNNF#5P+!HSw`xm-2%rel3WGE_eWoms3mJUe=om;2{b8X^1qd0m&OqCc9#1Ixxn6{bSX6@ zQIeFxDkElWnqEl{DXF+ot|IFd3xwK^6=}uE{mGHJ!VllBMIC8790K!8_|+mdpv`{cQN#o*!bob!FiI4|UY z`Wr%OoHNobm^r?M)Q@NlE{P z=d4zL9AMES``k~-^b99Q?N0=I@zKd_sH0p+ zH4_q>D`ltqZZMD0lC?tbwZKq;K5lYV-*vG!TeU3wR?X9b0dHJB#l+RhqR)Z1sjhpP z5M_R;RXV1A){X)IKH5aUB%-SHQvfLoAZu2=rB&}mpu0(l@iz;2d8uG0!FkK_Q^u{D zNX^WPUkzez78IjDn4*XO@E`yU=64TLMRTXFdTXvDmU6hr;Dg;l@6Bm5my9ESk^#z4 z2a{tw_)E%Faa4`2m>rNRLZ({AP!>`uv+=m9cmBTXc8LzL1)U2aVcHEnPkD^VijYX5 z%ckTHdHwJK>6Z0`%A@0%76pk0q3lSjA;_zk4wg8R2*n1>Jh>zH1A0?c`Y_XI!ehWM z;szPZ9OW&9a+(M67CQF4k5EpA1bGclmZ(Em9;43M`DM%t7nM9dzAi4dUe?D4AC~^F@%-QMASI~Rl$zWqvVyacO{Lx9IaO8`#UDQQ{#aM5 z6OfYNUY^xhuWRB?UQhfGGF+n8cINte^B6&FOb*p)bYj$lZMB{=puAAx$Ni;Dp_9x& zKN(*CY5Ft?LDz76n;i#&q()Iex||W zIp7zbv4W);fA`cD9!^=mdFqa`?Dk)H{+BAC9FJ$&Co2=D#sZILJU=7z2eKitB9Cg2 zyn}hnzD!`U%htC7bO{a$SQ%FvOYpBo_hT{PLAjyR)Q=;lB3%RGHEfJ`cT|#$-$t_V^9R|R(QM*IDb6!WcE*jqxIo5OaPVi5+;yNMQA37@xIMW@HwmenGkljrI}Ff5TV&HaJJJS zrP&C{HlevlsS%sGDA|SlxtG`XmiQp^$3pXPY2wa|7;eHqaLf)t%e=3~4K?G}WcS80 zwra+5Jj(p74arx-av zGwFVzY}`^~3C;(~aX_jtW^6OyR`KBSYNNNQ?-;AiqSj=M)zs_9gA^;{DaNfQTcVqx zGc6K#z;1kd5_d{XyJ@aJ(=VI77{oCegk2m_?2qB+KifvC|(`%gUu%djxJu{R#NMqbftrViD2=D1d`*6T&AW zPY6Jqar##8zQ!9;kZC5I+$LXmX8h82;h}NvShj*CA)RHACIR1Kik?U!6Y!*$d{Mz~ zek?c0hvepD5CphqlS%JL3_Oi|XqhQ*IGmJ`6`r{&Mat@ksZ3Cfz9t_9sVUPcS7J5OZei2c*sCJkf-cWc8;a zLj@w9M>%#SA|oX{$uy}~M?}!&>>EjiS1StCkj{fsH8T76Cu&*f@0s2URtpry4b5`& zYzfV@h>>zC;62n_JDg9W*Vg_0*KRoRC8@|O;sboCHzPQ2*C3Y%^K}Z0qV8X^&yPrL zxYNPHH!x<(AiYcKSEVq`Y#@H4AK1V7&WiekpcLCzlHDi48Rk&OGk#cp9M5*i zaeM;z&I6x(_y*ak-JX4sZt}HcI!TnDZhgTZVW=tQQPQ-ttU2*H?J*lk`_`roAy@XlsVO~!#g+N6WVg{cf%N2XmlRd9iWVID;z73e{#ikzI-Ikq+)ShQ;vgwb zHpNYGL*5Z9xfS4(yiFWRYrJvc$CCs3Fw!)(uSH*@?Md&o8!KO~&|?GS6%GPvrv}a% zv3#}7O%I^7S{Rr3!laxmQU1XcZ)dI}dE7Wukj9?nVwu zMJdB=#f?5@>rxP?!C$HFscPHYR%!!Bzamf4j#CtPZTgyiru}Pz$^S2j^@&J+xFTXO!btu{8t^xfuKn}#(33Lk9Ab`W(wSXYki1Y@3H}lt+N@vTq zQyf0>S7tJ*?@pL`80a*-$MFXauggf_$&(UyJ5l=L42VL|V5;Xsxpn?a`Pk?73OkdZ z-`*ZPuE;ah)HQnfKfvcJ+GG_lb%b>b|_b+YyD3E=xg!}#Eb z%#*0z=X_X@-;Z=-;wD_iaVmQ?W|i^z^pC~Gj8ypzrwZC6&-&~)E3H1jw8m;%Bw zjrcDveUW(cbg|q0SNnIy9K!A>sO^E-+UI+4nUnmUjk=eGg_4%^n4!`KHQTSIYF3_j z(R5q>YK^^h_*VJXq5YwF@j%YM)LobVacC<4sJjR$y?)hQqQBK$@C$gSbLAetQv7wi z)1km{HN7NLJoBgviBDPDsuv^GD-a*~sG}a94yZ;;Oj^P@G%Dx)-~U#3omVNMcba^n zF*-X|^H=id^$cplAfIrt${LRP(}PrmB&{?u!vYCu;!&xZd((}^lxH0muJ29D4W_Ba zk|ftutPC5;{dhuMi?r{nFsGGL|M7IaSjX4t`KQ`^=d{)}YTfk>kD-x9UEC?bq#L05 zj@)j%a$hlSgi+&=-$SuQ*BmEaKW<62SzV^_?Z7>rYmHzK9H$L7dX9N^B%kO!O?gDV znAUdMq$W_RH_+VyQp_UB?ccQq)~qJ%x7=_h(^<& zV*K-FBLka=d4MOvXL+ME97YTySA{F|&9uz9{yno@4>IY~$et5$BV7)TTSz}SYyos&w zoociFMxB!vGl!#FoBc*=ziq31LtAjNLsYFCnZiHIt9T&e5AIk^#~Uvk^*^|SmcdRp z?3dvtD%~<}`l(@8%5vmhis6md*w$#h@4n;LAhq&5t@M51_sN6DiQ#WByg3!FULn#7 zqV#?67Z}Db#QcJxLq`UkR!yqGme}7k?5145!sf%bgk`J^-WeBWS{>PBZ=fthEGnuJ zH64L{mxrCCRO}`uV^79DOeMNHHS!+qN)_yVL%Jec3rbOo9 z$2MC#hw7A!mf8N;OHfE^&Q9@=hMliI_Ui35OpG^9?~V&!^V(c$h-$je(NPp-?WNb# zoECfg(Ats#7F>$7g))$Krj8S9u?3M^TTF#@D-@Gpf zgs|2K-SipX1Z6eOpi`8d-8ov6P>%29i+av^6v*L=zK6(@<~x1Vq0$%6^V`y91EUY&5M&i9roaUzepQeJ3(t$)35v-0|Ib)i0uM|GXu z>%t-`ZT}esf0tA?fgnw!wvGB_zFQo{WilTWVC)9-J8s(Rjst4SAXGnxmk3> zAC8TG$qkQo{ttu#BKn&@tJ+FY>2kA%mh+U{lmQl=TjS}qXWLI$c+HZNpCUmt% zbroHn{=B%pc7Fr^KTiBJUN-4P;~!nuZ4PR8Yol||+GQN;OhVX7mJ%i9S5{;F<%SoFC3)EpU6j(+Ul>Ef2H$xPBwcA;pj6U4W@^hcq!iNi2Mqa1O zC!3+PH*)g(qkIlYP+Y#)%|Kdt^frfH<2hT4DDL90M=`1=n9yj#$9i8RE-Ssr@}*fP z^P?3Rkx7D|)?IJ-2y9chBux(%5`lvm!xW8X;8_)aE>R~M-cB!J|IL;XT3q-MVxURWalIeN>Q0u~xSZ(e*HjcG8yFJQS)Qwy31J~5E`1^pYoL~GBqmZJ z(vnN!rIcz@A?n9x)N%$XTv_;_%z}Ya&VWfwukx2gf0Svvu`*ffENYyhkZ^KpjZ9Na zsM#CH?v~BAGW(vl?GM@}>R^>?_;Jw3wGcp$d%;=nx@J9aa<$;35H4jq;-r?xd|P{S zd7#t)1n;<3<&>Zxhr40pRlPNbiV-#Lt^4qOu|HcQ{Z9Rn@t2YEAHCPuYxg#MJ1GW8 zC4H)!Q~d064eS^ z+9zX|rz9N64g%PG(esn0*e^=|T zA|?sQ0G5FHA70AoZw8q7B;wEZVCyUp&3-fdqkbu-BUXs(q{hAd6XhlGshkuw9TOHA z0i#6cyXAS1Xc@yakEX%Afhch!s@rxO`GXPShRv_ZF?|wn(lWE>E8i2?QY0oOK74N< zl`y6hba`I0I~vV(;~Dlr_0zQhZ%*8S-;b7Fg$lP1glRUrYtiygn7oMC69vqxLXLq= zL1)Xk?#9B09Ek@m=7nL$M~6I~yykuRq;|JI4XZvJapTuGIVjn)MPIs22>(`tSevf4 zA1Q+U7?q!EdQL;)jCEg|ms=f?reXK$%OyO%;_EoMyth78HC;=h)o}ge)`I;@((Bis zs|@zc)uwCVFxyAJwMM>r|FzZ#VyZ)Is>j2Pf}bOkWF~Ym_+&3YmM;be+?T zJVPju9eL)K?7+(lS!a2|m2QObhI0zMn}4M#y*wW~KS1`&OKA}j`o%vqIv1vJ{$6$B zlQS#1scw6WFEAz+cWlPp8JpEi6*qVDGhb6pG5>?6x)LuqtW0!XGVSq}+}LGTcg^jL6^N!$et_=kJ>fPrMBWSakejaC=)W}lI5-S-^B&vgGJ|sY9C(^0EX(>f#d)GLcyA4Vo?OnyC%>KN#T_oib`Jdv<;C+$ z84h!o4=^g^(m~K*2A3c%HIQ0jrs~RbNUUetu~}9quexiBgj=bpDw_k+>|Xf!Kt5ILvhen58Q? zf9ch{Z~^Hdp`V1>*Pw!%zF{91bgNEkY(pi7AAT#5n6ZE<%XaP*Dy3e%cD{nN{x#m2 zEVMgReet1Oe?R*dQ?j~VyV)kXQMWTJD~j2QTH5aPI3kx_C>WvbX$H)vF`lervVtwrdL!-GcC%R3x9x7pcCqh0Qlqf% zpTh3;h)4pPJ>|9XpmLWQ5(4v`0cBcOsn1IQd$K#E(3v%!?9fFy;kx_16x9%V$Xrmd3+G+)1* z=Gjb0L6jQ86Y@M2on z*2X{A$_UT4%k#1Ic+a(`%0K3@w}N~h72d0l% z8GdfweloMzZYPfmARYZQUhXtZ6ig||`6I#QYmb^JmCuIf&PX5bN@MYuS<|yQ(hqU^Jkhgk6>{$ zSyF#O_XcUgCXqsy#lr4_<_advJU1ZW9wx;5g7ER%d!c8JXVt?b>M%-y1fS6Z5)Rd; zP9{<}F}ml&YzUsiwpzI454Lg?1b98~2I`+|h0+?HaPtJ*zo-0}*5U+Z=|>(i26wc4 zb9mF$&Shslm_J1#&RZ1@OtED7gQ=6^Wwq*zU4 zz&&J-Wh6IV-GUN$cH}YwO*PWB2+WBW$<=arw?d~@dZu#-QNUcWOPht1wy=NmAz;kB zgtqlawDcQwkZ4&AuR+24-3zDVRq-}^6`r5A77`Qz8m>Rl&A3Fu(=5aGbAShKJu7aML1g>0)*{H4+G_qt?G1X|HqlwW z*z{8B!O0=hK;A$$c(OWUc{U_fzS3um(3$q#aKYIE3hZw7YP38t&XhyrUZG+UL&Zti z%u|E{(qir*)3u@1bS|b(U6X9z`=YpfH>iTTt0c7ye4p`?__=53*N5;$7d~V}J*<<5 z0e)g1o*2J~lZ$A`A8o(co2_HK144z;%Hz!1@89<3wI0m1TrK|eWzTA?(PE!dy#E_) zvi60jd*rjd?J2M1Yre&GSCm&izRhmy{gtr%|F;KtzxN5Oi60dpCQG!y9+tu4rShAz z4@2Zn{%CoK{+v`q{BbJUnEHUfjld6_FeL$J1%Gmm>Pu45lONEnjCy}k`JZD|d<;5W z#mG@qPmq1GlCnf=2?Ehlv75LQnDwPXVnsN}zWWV=q%Vo5sz3eOWI&z9$?Z~r7%kyQ zzF`Se7LElokzNEuUN@CwThDcIc$oY#sjwUmc>04vg;gIj-Slfx@v-IQ?}q9s52aY- zFV0!_j?GvYNY%4|3Mqzf6cCdk}gShpj|FHYL zSN%9O_k+Bi2!_!~Dfjj5+#PQmN()D-&6>^e4gCqWn5I+Xy}5|BvExA1UBNk6cecCIhaQc>=YHS6dP@EYpL^5y|jjdH&rv{wL$tJn`;6ym5Wf%Uxq-ut1+< z>KY{2FHV3u9{c4@Md4VWYVYrho_epKHr&ONX}11TzbeOTz0|*tD~hk6O(t7^y8A;6 zC4cYOBZztqrjzsq>1s)%|0%+^|3QXcjw~tI6=!CJMHJ!>t#%CpPaKuCm_|%=h6Xsb}7B( zCpKF+Xg|O1m(Y+x5rH%{Gy0#AcrJFkgN{um6~sb12JY|A>pxQdPC$mDy}=Czhe$Ig zCE6)3Nb>=T+X*=_^m5B|5k3cWIb2s1$?pXj`Y>9+uc*D_*QyL=8BAolB}*S7z+)ls zM&FdtO554WGB@$*$c|Oo9e$<-C6DbF_o6OSoPx27(dNH9HVgZYX1$a%g`W7ZvxmY$gSS{Cah2a z_+0Y-y*R(KBKeym>q7Cv?|t{)b5gC!J)n4Jml^T$VQzg!<-3(I>c8q208rw=>GL}{ zA^%(!0t-1~N`6~F88E#S#C7fGSf<%mNGD4Q&y*TrT+m!1y6v~0lEoxwbPw$Xdeaqd zAN`PFJauQWcCQah@;>KjvmxsC`;0#2>&G35{*0bX@57(h#I53K6_{YtcPl)d>Ap0q z-4yYwbom}Gh4RCcc>DT1s(V}I+(tnHWLRc=Kkab1Yd_moZ@0-?8m08ZLf(Jvqe+?| z-t6vO_>`(x@baKLZsDwip@-4!TJ9Kc1E-2B$IYM8$59NGZgXA1$y~lZ2a24APiIq^ zi78LMx3|--8f;1TU1dZ02XXm^;Y!v)M|%siMFv&gJ{8;hjbCalwdhS~ zk}SQ>yJB-=u zh;XC&)kWW-hzK?DL>C|`i6Z>5THO4S6;_0zVk-OyOj zzpsyFgXTkXPyi`e`_soQy<7&3u(-72t@u#`c^BNl&03k8b@>E|w~pFF9LE}Bt+8_& z`&?69+43KkR!5__MmHeSC!3s4762bc98vG2%U2Mn@1?l-iJ52jJj+WTEn{=+QP=Z> z&F>|?zSx~Zz|Z%D;2fHsf&M_l-+$8DM2YCkLB=C8&L%*a{-Ug@4wS$I1&X?`r4dSJ zwS)<|{>KiCpb;ZaxasltUOCZL*^C!bX2a6~@+)X_{WyA$=pYbO8f>P}y99@8G+Xe# zz8erc5C^(Xp01>8XFL4N=znM|$a>%Ek)7Q;P4)W}5MDp<1-rNQ1cW^ZgcXIRWC@`u zL&w{5MRuZ+#l;k&#fh^k1$hm{E6b(CTk5CFmj-k%x_&KRI;5rwP;?+;j0(<|H35>M zK&A?X@;krCuZCMTM4z0Y`@1B+`c;$TQ6~h) zDEw2Bd0Tx+ea{cs{5}PA*5YY@wMP@r{&Ve)*6G59&3XSA5>eBYiOxYh#z;8N$)e`t zo6_+9FwQe$nIMJzteLKQg-C}1b;_B`HIkXbB_@0q9qjxR+q@e;pUkr2CE5Q$@C_5x zR(#P#$DX>!Pq&Mv)9%u80@~8q1q^~lb&~W4CFW_cK*IbV0*}ouMRD0%B^{Cx7yC}s zm3|ZWzA(t6MX9b8ygfeXYpY zCRBseGw8!Emq4DqyKLM%iWC$pfwNA#XWE%ES*3-$^;Rj6gxO63_kA_}`R>8lcsa;r$l5J*LEI|JdXA$u7GD+L*v)9wEa!jR$2Z)bi*pQWZK@p-eB|gZCL+)NW`@ z=$GT|_F$Hi3)-Rknfa3vB@@o=QSQpbbAcBr^w>d6lKm(m(zTB4K)KIuLCUmgiSsZL z&7qlCCgLH(dJ2m$2T!4^tmj8k#23C zaCzeGWz#g|vOeaGAEJFGCfITGQR)l3HX-+@|FR@s`me71ez8%HR=0O3Ya$~0XSea} z->T>D>5mut_9u?Wr*vfBn{=L9JMiJ^)cFKa^R5)nXu=qg+pJ@CBVz&~-Bmb&qQBs$rRSz?tqR_LeC>wMW`oIJn(*5@+6-YjVhD}J zq_?v0IHGyMjNK{Lt;Ph7e72eN z;zQz&{%3?)0z)8ba%B^Lmq(X}sGSZ>(iAB#3MD91B|b%g1Sd?5o-lK^#{?=t`$r6T zrWT?hx8+Pw1Vq4+Uih2Cr_lQ(uv^*TM3eWLB1o>Ig50mEFAOtMOUBF-@k90J4fl~d zpEd5>I2@JNw|en5uFZ70Pu|$4^Q*%{yDPa?CUHq$leVFqR{9i#bc@NZBT4t(xc$an zoR@@C>BDgqX6YytSEPt-R@+s2FmD z^M_b*8zJ5Kf9W=+uGm+U5)%5=ZFh+{;f45Oe@RX}IZL7GiFsdf7J{W9-mGgG5r2@` zQP@mWB`Ta|_t#SOx8*3aKMM(NKVK3C3g!GiA!(-U3rIfv%SQ1bnDqGcK31|qU!p9e9B_=*Hb|K$=G`QuI2&ox*2h9~EpSlpp_%7K1 zS`uE-NypP{;FQaInE`L;;&3O&2bbK}*<>8Yt3A8*-q!oy@FA+)A{FcLe$L=@op393 zuKfX%#-7KSE%PRMUacv^3oeh^eaBrmQ#buk1|ZxaA%;}Jq}`w14t%>7<42~jk4?k%km3=%Mz={sBT zA6vNw*KtH9$HCiY_m*6Y)3+QS1}sijOJ35J@?P-dy6>jtdi8L%YqEdv`Km+wJ&%E1 z)tKi*(qsU`%`I!pO^YiK%oCz5>zNQnITMADnaS3fFeY-Oz9~4Dj6O)@L*8l6q(2aP z(_cS3^4GD-n)!g_FnlljOl5Ndb(yZf-BleuC>cnVJrzK=r9E^&cUheQP?e^>Nzssm+zoa&vP8f1c_=&O;g4q; zEenF0Urh64Ozl5snm7Mmv0!<@XEg46wVkwx0r5u^r;F_ui^}|fScl3s!Z`;?6cHcn z-jr9jL)q8(FZfJZ>Rnj7lJ#T#=31|PNp1Ux_eKMDrGWG2h^WMf2DVu=F#d*-o9Mzb zC*C$fg0PQ6toG66=h)ad%v!_ITi#p+(9#mV4GN=9(m5}kzrvLc|LGRMkB94zQ@9?K zU>c2aBqBuI&c9gbpg-eQFu5?^*3)O*3goyMK7UZg?H4Jiiue|G##Awgfvq@2cwkl{ ziRSL-i3b<>Z>A~e9g(TO&Mpa+jbLVybGrgl9a|^BH?T)XG zG$*#=liQ=eNEL=Bx^!H#GD`9P=D8pT#9wF>DnE3#ec8DFA!)5`#D1{Rak3}s`;J2l zAwMn`Jtm_(*8-)``pX_d&F(5%mAB95*jVB3tbF10%YS4PK)#A8VVh}65yThnO?YgK z)*EmUKn#U8ryGq0WUO)&a@1o-`GZN^hiu?eP9!k$^-<^bFC% z*o?jqSz7jiv2q^uH);fAn)nEGAJ`9Hw{R)o8H^=OzbG}Xu+Uj$NbdGjeUk%7C3zYD zHkmn}Hkl+^Go@h9zUc!C!CCCJ&5~Oc;l5{fK$v#L7UB!W3lW>E#%%!Xxa@oTqlpf3 z;*+DMqjs*qu)AE?TeZ8ipN%Z96;2+}I!sa@zrcRYzU)|};oq{i(w?xvC2myrct)Ud zd!o?@M&eGLvuP^Eza#Oym)^!BtViI`sE(6wIwqBE;KO58 zU#d9%7*!&2>wfQjITB`TVsfh>Gwi2aZbcFRJ`?z!d@)0*QEUKU;Jc%cp?D+Cz5Ck! zgwE89gj8bl$P=uNdDNv>f^(6^sg~&v^54GZ^X!O|nvbE~4LfCJ&&lY0;(u{*9{*fT zW&4MVNqZU9yFLM278eov&^f18yAUFuubb1M=HGo6!&Cqgr{@n<6 zdY|jR?F44u2UD1s-b|r+cW+3fVznE@|I~#EQc*ZsuGM;xJ)*McZUt_(i$8g=Xfjl{ z=!RcMy#0jt6VmZojYM8_gsyMrBP^xUOuP8mJ^%|q^uJ5W58IG37vJNfOQmIB`<~@l zomi6kaRHUrbx1zB?%o|OG%q3EY;Q*mmRqjIAGDuU?yIfmq!ERe-^2h|;%KjrZyV)N zI4u06ePeqSt0zkz)Nq1vklujI&ML#u4|x?{}DEZ)Y+*^VtJJAe8ug9>Um zum68$P$NO>|CCz;{^VAQ!M8`o3Ur)>~!Dhz%`R^C?6q3n)7dM5cJzsh_>c8 z=6i8aUO?3QKHNxU)Tmthlw}%0xH(X{K9ucBZf9vk%xjOR*G}OE|sCrB?c0`nIOY z?HMojx=-LuKywsqisqPFmuA};*5qqgE&07Ul5L*;aGS$(BP8PVy2W18jr*(nn*v5~ z3HfIpTf2`cramZ6wo)H#>2S7lNu4)40Z7gbniBrI(Nd1-f9D>bICA=B4}pnRG=RO7 zqd>GkcB-TAEu!&$-tx5^p}8eqM`O*^?>|)I2jrX4mI+WoATmig%X1;=`un6@MaMHN zr{Mw*OCsVIr{RKS&drBp-0anbT%rzz(2*oav;y#1C{`|wUpPamZC6X5@Ggb9uAEC( zcHSEbuUUgzBP@uQzh4t1`VSat^Hfer{&7Y5Rw&7(liFWUI5QNc{qvfj$`GDz+x+)+ z@lc4JE%BCdnylS@pNoeA@jdQ)Z`A_%1Tb`FBZ7@HeL<|?uMw@*6eQQ~cMF*n` zkhUSH5f!Gx6@T2k5t$|#j;NAojZ!K$?joPPJmCDr#bG<%c8Izw&7$W=j(l(_ z?KxSLpS1c6;d82wkJxiIK&mw7$c>(_J@(&c6~h_#5?$?2NUxR0R^$*c>@)P_VQDZU z(enEnd3v-rR(jp{KZ2)K<@k5IY2^>#7+TQw{;)im%0XZt6Nj~_Hk(=)cd#ALYc3!8 zQUTsjBg;gt0Eg47`9*Lvlaj>Y_g}0fj$AO)33hOeV3U5c8NhaR-Da4>YRDs-t(RqN zI9hs2UnSd8*>n#CG~2I8p zklkfU!XyhTZ7{fSx0HCM7exWj*xywwd-T4dBvXrbv*h`nz!W7PNQ1kiHXbv3t1i*8 zuQYcS)54)T#Dg%bJU*nDw=NEo#gVUlh(Pk$Ac=>0O&YVO+Mff5JizzwP0_6!qZ6d; zsDeTtmTq4fY7ctF5zAL+=cJjHIem%5Qp{BN)vLEc@%X3CoT&jw9Nw^_YTK6}x!ALk zdzpzDDDKot#W{8sDCR-UrR&qic$V^@R;>Kug4m*K$z(lwB%f{3L2g?h4$iEKXvODS zEI48P+Q;QPZKnR77ck7sHilk^6W=lJEkC z%F=|UBc`By0a_6+K1qKDIiNXE+Uz#DAL3XF8Tg=L=OSXt6p8!oR6bQja%y$yrH#=_(`!ciUOU&LvmL>UE%xzu_(y-f!4{51b@X^2^ zV-CF>gSRB<*qE6`C^CC~m_f*By^}#)I!8mV4bWo+(girxfB!TfKoPk5voYY!1OK9^ zp>n(bmuZ;wRRk_^@(jiqvs= z1&#^1$I>;!J^Y&lKEBIu^FF1;)42?k*lhtjp!RbURg$|`--%Lv1cN~zzwjJ)CwA7! zCYCt}>#b=*L$MJ@SB}({vLw9WLnqnUqXo#Tp$KPvj*)?9LWGwY;;J^mb{H=slHDJy z3O?#@u(Ph+VnEYFw*?&MMW6aB1ZrXydLJ#32>+^UMGO4F%RH@sRbU7&dl%SS*wDv#dC! zJlmCcz2pW9^dcnJ*VF%eeSk`6vC>2_RzCHXPO1}7^O2&A9~=tCKC-iG2XSLMd!k?V)2w4e3K57pOmJJV`07i4h&^aFUj*PD6O4a*^urE#GTJT|3Z(^? zaeo(>t?`sJy1Z&nmkDJ&-H`;1#9W79m7c3zVZl{yeoZed2s6ag9u#~Ms`zPwn0Z^U z<*SL`1P%#byO@#lHmD+;k8Jj&Vt4t?v*mlnFSg4$iIScoGbCR$&wi1^nEqD$;bi8( zowM0>aLDHNFctXCp_#8dZf!}&O_UiD202vHm^%2DE!=*TFLFX zHko7ng2DYts)axYDR>{G4w%OU>hKgS)bQ-#i`w^37eF$dllp=UWr>q{zc>lJwdGjq!UMH9BgD>7ydEQ-WTHkupRvbetY3$fo}2#*yih z;LMqK4u9o!fnyB@mx`ReM^uBO; zI>EWTQx1Uc-Dd)@K$6UFFpJd#wpWjbB@W<*+;^B6a~%WD%_U!Kx$>HD4)QyjMWY-q>7&dq|6If)b&nO?L|tT*E# zbe~@79qUj$ODLCdQDyFEymGdZSJKxxv0@&X*_7T|ZWucd1_*0*6i?{HEhozCOfZhAR1O#e&Ms&yf zunvReJ?-UuMuAcn5NQ}a(u_G-bnc7SaEjn3r{<%>9kwb7dQ(oD{(ZDi49bE)C^Ni_ zxdq_3y@O)Ec4hL#Z>`BxMNxVU9qwAziyHFAE3^J`7_ZDVBu}45sUAGRi?tVt&o@!z z{=5bPAg#8G>fWfn1VDp07Q;9ks9B^Qosj2nm?MoC(IzuLe8`YKnU|xD>mZ?Hi<57L z;i}!Rjw|9QU%}UVSvmu9IoVe6-iTc%ab(P1pI9P}!AE&Z$EIg_>8K%7CM}{!KRPe( zctnS-$ex@jxZQ=IxR?sVk{d0_B`YeN_24e0JhJDmpujKYt|amy;Hv5uBpj`LD~F}< z^Ps(tw!B8Q7%fTK1UGcpif%P7CK#3_4JY7Yp`e%IY5fF+9i+a9F}GAj!0zFe+BqA~ z`8J}3yId+*w5eQ1zqdtn9oqhS;QqHd>@o%T&Yj%(K1z7p6y8biLAS*{w0bu2@ES#2 zBRcJfjC^*fGY7*5=QhP6J*q8s8mcg!zwVUdqX^K4E2SlWF$RH1N$AZUaC427n~Ba^ zbZQlkR@jb;FT0id=#(1adT8I$<)H*{cLWRiXtqXu9B~=VoQliMYu!n?%jh1Tc@4^~ zZDCB^jWPV$IfdV)aT2eI=6%*jiLXQA3NL(5Tue~#5ZoBALnkhZmbA}SM*@7Ek{2E4 zy|{I8c4)3Io8CUp7cJh=xC*_pZCCCvv(vVX^!7n+lN>d-LDvRj!Njen9@_~YBB7?~ zh{ZmV#t*1Eyox}W*FK{EU9iaXwD=k_J&j((7sK6Yc z?sI82n4|9SCvJxIFyB0fa9&cPN8u(Q@ws6H+?vClqaaeY8ZRdQXpc?`dO~I$BPe(@ zgdSH`NC)&Mhb8m^j{FjtoO9R`3?3=k$V$L3vL!$eyNk)xcC>5>XPcH>EyX(_>N5+k z(N+duLT`lP5RW63eJ7zKYY_)!-Y7nBR~jot6=R`neckb0*^Xn8Pd8sNeu%E@{VVP0^0?Bk2FMT$OkSYJ z^{z$T>Of%GlMl%S+-WBM1-y0e7oufVX$Cl-f5Scn;2NOvmtmit<4=D}s_jmOY$zqa zeVkimc1Jjq4C5Yr`@MWLhfvJRq$)PqNOt31SH0n-7p8}*m_yfH@?>K#lYjZdIhvgj z4GN^#`skNAlFOU%IhCN63zi2$>dg?$cT0*w+oxiCwU^3E+Pn)+pV!_SG<6y~hMRZu z3^&|Grq-r7L4Y>W-6HJ1_iJEN4`_qiZh-TGFC)u^ld61sx!~EN-bhJxqOw9d>zYO?|J{{HaN|4 ze{-Vx?%WTNtP4LjyZ6PTFCd>nT%7eAX4{+ZvvD(9UmPE8M$gETVyRO|o(|k01wbUi zb3V8NflM;Lr3I-2!T>FB0reicetVC4WH2c|A{s=>`74VrL&>-qUn*!U$we?qsJ0|t zVTGB}Xe84I?Y@T?0w+Ml{;j31cmiU?uJca%8;Mq`@e+OKOVU}4r)gMDgzXt%HRSx0 z<1YCzusAkXDSSKdeK|L=!M^?2nogv|iRAu9m4j@CNg={^>2+9&=FKhqhqgm=;O~Up zmt!OdMc4&*I|t0eANON1ep(%0^7o)5WHeg+()qk>I_S|T&e{vp1Y-ln^S_9Iv{26_ z^*;wXke=)RvEC3oxm2|*8G}TeD%p)#_9j9MA3VjsZ6lW~;9y{c?~lD$Hr6sf5PZ4DV{Y7zE zV-gZW-qf34x=F$Jp%Ye~Uot*4OO%Z4E0C2LHVAforatOYGpKK4WM`Zv=QXHz_>|;) zTM1$rM!wzPjv7D+X%Zm1Eh~A|!8DH2p^k3d{h>nXGt&kG(XMv}ZL~%ot@o;I3pzeB zZ8_Z3SzK)LhiLzNVPGO#5m~EoN=%G6tf7NX*juM`ln}-i$%IvmY5waoMRL#9s)GT;S&t>IphNM^kmQ#6tZ?se_SLwu;_b&`(1JQc$=e?eK zX-cKN@m$onazy6~hidO>=75TYoz>?tC<_n!dr#f1%4RaRsVIUy>+koodxMtxMRx#z z-Ivm3njn)`>UGF>J-BfzHnHJsKiLDB!Bo_UP#T5taTFlRtm~iHl^iNYNBoI$tWM<+ zgKaa(1RcwoaA%Lx;OAmNJ)}wOyl`XCGHO#P#F{f;3Sb7*y0N(-p4^Vy~Hd3Ta~`;Ip6CY z>x*I6#`Va&wH9ny0&pj>sN@JQtcnbQj8;EZ?4tRB zl`NvxeHZg7?76M#}IJ6;Y&UD^9ehGsPY*@c4D`Bq`R+m+p+#cfm{G zwiB~;QulH9RdB%Zjvo&&QN5JeN*}@zV><%x14mr2m(OaWQplEgN^$H*2WeRFrE?~z z;Y{b9ur%u=tgcEblvaH)--(avodH=`r&sTmzfj^T$~!FC*d(->R=CCtnrux>zzF`g?hs5{v$GYxFaJ26rsBcG1Q}ECPEU3O0R{^zieiJTq-03kS)~*S zXrJ2=6}`)eCYusckGopY3u_(;|=KI-0J*ix$xn?&~tglUL+SOsjY$ zWz7lZmt}gM1Pw#V8Wu|F9GtMwEdT1~K)#s=MNM>*Un848PvOH7dnNe~z3!Km3*djC#%&JAK=@zPn)jX)xg5Q4jk+sF$bwedNUMMQ^}*Ger0g z=#{Q9{}?$dR?0o=KQ~$_)!M#=9#)Bu>DO+mbCHCd=R=L0`bmnon$C-v?I3oe!Co)- zoa~`SH|o~g3d<(5Z7>l3zNpi7I|xX^Gx@~FE7KkJ9rYAnCa}pk-q>?It3Gerq|C+P zc%e3}aCEr3i+&eL7+ew*RsCX|(T)3X60Mf2-}?5c8K;Ys@1yY(PThIj2)z4}4cEs! zsfp-^TAx@MY~J*VL^AYcionzDwy*H*NM5EKTzl3TDRE6q=Ih>9zxs01!cS+mVnsOd zKO7*Dz)gRtzAMyt{}?Xl^>x6OCCF&SZBdQK|K_3wjU%2_6fH@xMI?9aHFY@wd$%~P z!)tFAG`JY|aUDVAwC3R=$?G5~aS!}3m3P zkws2YvQt;IpBxVDN*LmI;Ue~5z=UFwa8k*?WjOy_j%qMV1BW=rn`fm#3MIdWrsn9K zSqXf2t}O*${w=CB)&@|FrRbAljjdv)py>tU-o3286g5*42c_B~g#?9ogBkmMREykp zoN>7xUp$#5BBg0J0#n51G8p`qC^z#qYf|pt9Ou@}!5H)kNwn6}eH3)}W-sqB**4>9 zf^D5VKHnNf2je&uKb%3Q{wvNfWAwL+;;3q7Fww1qNGRNUq!EH@6vj{|~k`{dTeMz+^5>Zh|h2&)TMe zOe0T?f`~BuDT4`Qa~VR;C-+!{O1=;t5|bF__(dfqpZCU1J%AxuAtyj?F=;vc7QxYI z$Ym1zu|S1m+g#;#b$Fnhz7o8f)YMUt@n(A*J`t1gRd^Svc*POOmGqfip5NW^BTzi% z8~r>#hq6fK`liR9mP2l3u$0IsrEtRYh9!V!%Oi z2v%GWmv$l8#||3@@4QZ=VZHa(d7VwY2J{d>Pf3yA za^oP2O#$2Z?8@p_Ed8FaVPaOemsq^U}?gkwo zR^HACx{L&d%p8mBc|2ck&&*|JM1a~eBnYF^y-x&OT-F~WPL3_Pj~8ZvCwIM(QV86= zp@+ASQ>wGwq@vI)f;}O1kJIvZNI2JhaemXeuHo7=&*hGK6Sugdth7gu8+Lu^cpIc9 zphKq%zJxaD@R?Uqc~N4d>8KBMU;R-OJ|s+dkq2XAzfX)jrocSZ`+CyF5Qb!7vT)2@ z4&4Y5oFg)?J^W{64{?9hvN!=|z(s%_^-12Hi3XzPC4e>pE+gA0)F&Y%C5XeTT$Bts zN2{sqwv*8fZDY0QFWAXej3E(C#tN=lei_e*5%x5x!dxNpqG8^(qs>Gfz}r>u69-k6 zf{!>JqUmv^g-XoLP!eU4$|5bFRxAYkYxdotVm#~z`@3C(kqK&qh!4F7qa#6o}KhLuV zWlt7mX@^9-n$CZE{q}=To8xICHC?UN2hB|`BE0`Rs2OSm5&%ShS4WT|ZlR-%0uF$i zVepEqSH3Wqb-!DsrEEBi0)ue5GN;27HG03p>(9C6$F19%%E_0fl+5o?_ww&4$3KsE z_VH|swEeWJRf%z6R@w!~&JgN`3OhY0a!8I^^jwKvtva-l=bP(u<1E*5k(`wG`NnmC zIssK`6yMuMhk>{8A9hXD>TD??@u6-`X1VvaPe)PnwqB#-cjbe5c5;_JiNm5lA&pmJ zg!k}qxb!=~sAa-o5eHVGOq1Y!YA_G($PX-Qef|L(5QUzcR8sH1kKB|%a6J43|B|sO z?yVrNHgza`@F`J)9dgXj%+R+PI`z$^?__^_3L1-i1(t*(CDf_tfl2eVuGnDF-mEp~PYrQA0R6>2+xD9&C(h03$A5jklKJevzLqqU-Hi@O z3xnUCS3HxopZ(W-&KuK{LsqKkl70tNN?HYKF|x4~UWGNB3VA@LNZJL;BCR4l(S-h9 zf#tSxGg6Z+C7tJQDop!|UYtLmnMk<1Byy8br_4CO@hif`v%b9DleZ1gr2D?y+(!kV zzhaityU>gvxB!^VV*u@u{CDfuM<* zN$NEYY4ej$2ehcUX)zP##|EFgHUYzUi#6ecz36KCKbV?d9my@KTcq`n8FMB3>x4>F z;#o>veY;L!evo6mBne|3v5XY`Ck?S6mU5h^s-1PTt|})hpAjY~D1p6b!a7mr!O==0 zb3lT6Vuur_MG`g!h5`kw=X#D9Fduv=p87MsflS8Hn+3>IOi7MZJcfrC&Dth3Lg=g; zh6%D*5sSTqNv-oZxm!GV!w|>R9bf$92O?Q?&ysg=j5wL8nG14aJnV5y1r0WQ>FN>a z5Tx}Dy8usek3L8X)`I3g~D2CR4>#_wrL2l!$hV^k!8^6gZ#`9f{IE zL18qFkHxFlX279cSE_SiOu`$E34=J|>kL;@M{H`g`fur3hH~~!eD=#5c~}u(@j^IT zZ5Na5w*x*hR5?2R>q#!w{EtBMcPJjy9f^hQ_+6hzP5kIgpBIL-Kzw8Snmdppr0Wq! z`!!uPRnjjZOfl9*`ME-@(z`dYXpbW0272Tg^OViDpHYe;h{Yk|Qa*Z9i$2M9ZsqS` z{cpLm=3RFQ>%=bWT6KW1-&{>srsY}&5p%W06Y0Acv*Suz9$c>XSx3!h(cYefHS0Ks zI?Y~p3|#FetsmBjJ1xU7CL)|&Uc7IIZR6Jar<@A-cj&THT5v~lF-YDk(aleZFFC#U zYI~+D6#3sg_H;Lv5y(3}SWI#%&O&A~asyWLK7`ag^O`DR=cCoccsTxfic&0%8wr}% z)|v^ze8oY|2Mpmcxq-B;24w_EtS&x*q@q{>Bx!*~rkxeVM8r1k$VEevb~-}|I15=} z5UXb^bT!W(y%33*#~x*<@qmsC>9`{;@aVSYE~1K~YJhoqS^NOOM{CW}8hT_DVkeSLH>0;;Z5$lxrV{9FO) zcFZm%0jSJ(2S#-050#`|?P1WyIm+A!k}zqADX5Y=f7wqQh$``8reKTJU@X%zQyKQC z8LeS@Ex1+L)cf&Wpw;zPJRzVmh0oG$>BXM|lEtb1rNaKn$AiW61MvFcgy;(6KC7@d zK8GYfDy%-4Uf7~tAwCY`veExBa{dJl4#|19cC59x*$OJN77sE%`rydLDy1D)thGNT zPGvNE5L)!U9A6hxuZX%xp~805=3%eES%vwA6Tp0?|68*0Uja;}rRSc&CbubYRWDFI zAJT{TGm{Vq>piI3`v66H&%~Y<*KXDvqKFLfw?_t*6@G+=rv`3Li?6~z$VF)7NG|oy zgoN`RUYFhYWkLN96((xbYSijg$hkGshR30SM0``k>W8%hn94A6as?(|GEQM8Kho=N zOiyXX(TOU?|PVn%HkFRB63S| z=A<}ZUmAjxrJ`8RrKF)@eb%YNP&)bK9bej7V_fEFh?QL@9k+lx85002G*MDh;$(}* z%(Y6B>7k;^al6?xACM|>vg-$aD1J61SKaXx9JkFUl^{Eg5<@bD6g-i*wy`eL+nA`S z6mjmjD^mR8%M?eGNo7o%R>vqA3j;l*-R4p4MD04K8K# z%xJvoo#LpsS+~jCMAfU`P(bX0c@hPzeWCQWFjIk6?$FgUu0d! zl@plKGSke4%vyOZ$bEc9!)}t~zN?if{oK&hN{rCVUZ##%tpM8-^jV$8vAxM{7guub z84SP@Z>ZmiJmVQFdzr>QFy>1JdS3?1!AZ~)p7rthwn3z~*f~Ag>PYY-L>$k=-iG8* zCJmx7ezMypqC$fI#O2u4zM9S@gW(iXx#Ht>>L$){;Zc`FJvLW)LB+8OSW*|%ogT&f zVVKGb&IzBU2R^(_rLB4uVcejxP#N5byX}JTge$qEY+tW5+n;b6?xysh`jy{Wq2At= zpPTQWv=1Mlgrw zfnk!NYloE~GQvsXEG5 z-P2?y)y;g-|CuT40Kr4bge|&U(UrF+D|rqd>YbatA1(y!L`q1U(0?YFpM7Uq?|pi- zv?Oo+8jA%n(XY??KJ*Z9syhMk*Z_prSrFzs)iP#)84ZyG-=6LGBLL22oA5}=tlgSs zjGIrGPqI@ZIgMktes}p&Fc>R>{%AI$BN~HdU9?cHp_CN!GI|H@mT+5J{AQnbp3+B! zIhh;l&EM4b@pHT6Igopjs|ac<0~mhexW@0|F*MYf{r|X`{s-#(tuW}_ZX}*Xf~tFe z0KQ_d*9&I9u%>nK^UNA--3+63wJ#M>1PecCZ?}ssubsVXmU#er-&l`G?wse2#V_#V zRv`fd#*2IFh_<|`b=)NQ#GjJ))wv(7CEh&odLhF1dA@t-+ zU??PFy_59uS@WB>&5*<#gLg$sOmnbGff3F*sJ>UpDh-AE=8!6>x-$S!N&H*}A$GSc z1*Dcf3##SKcgF$qHY|W$5$Ce27VR>~k-aJDjt{r>$8I0bWZdVZ09g}L3f8{-sFcO$ z_ogaABzv}*I2a&U?nkdZ4!I|8~|#rbvb zIb`xy4Y;$~v95A41=mcm2gPOGj1p6B(k;pF6bxOj8W5oR&96@exC7e%(kV#^!hRgp zAuuZ{SvrEzUWvw?bUC!W@TObw8y^;&Fs&<%Re35w2t3s1nuqz|D%;M|D`_pD7YRoHR$KgOP4%BKdbhVZbeUtu6wW4!Lw5z)2?QQIyGqx+V-8i`!l zco|G{DJe6|u)7i&uHH*G<;4^L=q?i6= zH+;Q|l)FiOh?MVjQo17htDZ&DgQr49VJwJwi*P|qGeW7>u=@1S!Hdt9rSz^}d+v&f zzp^A*ch7bC5#`J(;oauk!*0pli)%Bm&f(QrXNx9|)R?SC4xug^9dYQ> z?mN~-%7o9v3d6z(^1#ZqVS?gt?){;XM#op!{Jaulom8^J@6SD|Y%n(D#|8|GIHVu% zYy-wJYxqXf483C4WOa`upiEIt0>~6J^9D0Gp=>R$mSMvn;?Jj`-Wt1JPvQs^U+gO! zD`p5*QVBudg?9^ATEKoD@GCMMEmA4?YU@|)R34z>2Ea4`+a>hx0O@Hoqr^P8_0Pod zw{Sb+5w?%5E>Ir3_NN%J_4HLdzngw1WNj_$6MqndD5j{fa};cR+OuG9CFg`@c- z+XGrhnjkPP1JhB{LAVnskig$s}vfWcg@#8NEG1ou0Wr3WFYBIFQpA- z;0pn^F@$?Beza5{Fi%egp~Uo;KrnOy(_9lcc{h(L7GN=eE_^jJNc_YMH<&J=%>g*a)7%@S~_L}lqx35Iw_NTgv%fHhW9@fuKdu-tzlkUQ+JCr7c zK@gUSdu7EsfttMc_uK%b3rxw6HaKOy(Pcy9+{;-AU}dX&$77;*H4Q9WMPkSyrQ(8H zg{)k+>TQ%(veL6A?Fy1VUn`0%U>D`8{1vpX0oMR?;Kpy{4TSvD5(dIXOf`_B7w#Ah zTuT8~oIb!Of1@`K79hH4DUn-fcLDGVa{UcgEE*zMK7@tQ@U8T;3L-ub9aZT>q>6evUP!KF23$j)5cx2a6u{$zLa71%wuFQRopWjjX#@9~Nn}U2+{C-f1V@jq4-^_j!%cvGq;s!#8 z5E(ZVt}ng<148B-@ARwJzE4o`J9BTd>24)Z!PnkL=0lSSu7%VT-3)I-+}PjkQdKQB zV?*rjQ`rw!*kZtMHJrPoTlNSzNx3E9FxG^>6&~OFu52#A6a#*AYvOrir*}A`O%RQ< z0a$z$k-T|+YoW{aqmEq4^{=ZHX#&tXk?VW5IdU?$9iL3)+K-o-7Db#s;N0@BFrS+p zpo7VIwt9A9#RUoP#?~o%*AbUAx-3A_BmyHeJzO@%N-h6-t+Ib`;tdRpC{l?e>F|#} zfD)zx)YB6Q;F)>G#^)KjKO+uGcgU-JZ{~qprmOpOnGJW&e_P|vAVz!?(B1htuDSgW z;jbh}pPwg0-Ywl#Y`5b1EDiSLQ3C{*P$Iyga2wSh@u9@1{@5N3!o4s}iC+@Sq6QA4_n6rwAiB-p7vY$LVhg+!X!TiIddzcn;@86Y3L2biQrRmCv?6 z7TqT}nH!nKxlR~))GYjY6}(`@yQ3fIKI=J^LjVQKd*L14N1?ap+ZBZ+Pl*SeGtmy9lt=D zt3pu<5^=T`K@!wfi-8iEcyYm$b?X+&1Z5Lh4E()&CJ`b8J=tpRr*0I+NBEqv;5xiL zd6za?yAklHhHh#shq?TUf~&r}wc>zvJzK(CFGWpHi@(w+XpV##P;`-PHo{pv$2yH{ zc+5K1P1E&G2Kg!ex=c5%{95L}1*F)u14D&*8BOw>d^wmPg@Wt^rH$O8GJCt?0&zyA z_k}gDxj6ZJq6ie2t?{}|8^+xoSzaE(#{n_f-MK|1bOh9>( z+vM@RwZ+{zdUS3Kx5mw7({!KVl)BZGe5l7FZBP4_P%8c0^{hAAd}#eIyEqi3V*mEo z#S)ZG1IFMJaW&1_2KK0Wu-b2rK7pRb@_aMBGvPuy)p{6N?=bD7_zPsvAJAE&HP{8sO)&I&>WRTErSx=6H(w`S!^6YF0rtACGG>_;+5$!g_ z%fY*CEurA=Q{^Yz_p&q#hh4rB^LH{0%U9|%QBx&GyBq0){XhuOab||(c^kYjFqNil zH~3MoLIMw+2*(g90Dqokctp->8N@&GZXB|JF)!nC9j9?F@Cy4SRxO94`6YijMt!Y2 zZw3OgV>x9N>Eku@kLVKAuF|q43SiiMa540?i;K7UGr%Nx#pBWHv-ipe4SD(UFjflN zOsHH9IbRUF0f&4rS5v{waOPuViZ)4gyP9-PG9Ez22s zlv~iXdu->Tf@%0E|2DIzdR5Ye#Et6bsfqz!XsZ(IVlpQB1I35l7lp`?bm| zO-zi5*zbq~ZS-(nd_^Zj%kOLSrpmh0)KR%5%1$n~FGX0i^GV0(%apk5egwvNx7i0I zKwFdaz3*1ab$F+5p6;#%f{RtEXD=&M6h(qfk#-E8UjrwrtwudsaA8~e6%r&>+P8!j zCySgnH~is(+p}#%uP_;>)%g~>h#fAPd>Z1je9?-BNcjYJp7M?3He^hB_Qj^wLi8>o zRr{LVP@3A$v#X!C`8J^S?I|DEXWP3m=~Qnt++2JW&aV|rPgddA7S94MqhH)x8)GmT z{E6kFzhJqz#ooaZMxngT?4H}V#ofU7gT`#g?dKRfCXXr${&FFr6apxb~A8J3x}- zU{+V!TgEaGf(Fb)uX`Pq{-bIwoT7YF~Gt(??5dkvzWMm$pwE$@el$C9vRr9&_3_@4aQnq~e{`yiR6aID9Xhz!L9AzcmM$K5&)-lUCA_m1@^Mn)?anX4~&l2zd z?RWeq>v2eKd;oh9Z8GfdAsGLD$o8{SQ=H(!vR|Xmd@rvLn<>VypzV@78MhL@UyQYK zD_PjZzQ*o${$K`=(3hw9H3b@eZ3s4%4bA!Cd`X8{OM=>s!5_AyqW4J%yjBnjqcNVE zx1FWbc2wAr&W3K*{o6>Qb9v|^BtLWu-B=%R_C!x29_%lw+!1=VIs3qJQPz&X?X!QL zn^|7U+q^kp6XME;cY%5>1hd02_D1)Ctq^*$-USisAnaFOhZ71(AWn8Djp3b-i6g*f z4>nnQ3M0X{kcYc|E%uvq_!RGF=WpDKkXY=K&p`DGP>E_Iq&I4c^cWz)KsZZV5(zYD*WT}OaSOru7Uq`a~um}aDd>&TdXAS7K3%p zrcu6lsxZk@3`sY!1-*Cb;vO7Yky6Jm->JwRV_C&W=k@$8H%k5bo1eK+m}Ad_t>u=j zlgY8)N2lDavLj;(9>Q0kc%X#UpZ;W(mOR7jxRJ~X_L*~C#vjw=A3AkYEE;wT6|pdU z$&xb!ld+g20g9&u2~%Gqi_4nW6h6kpkBD%qB~E#87Umv^aObe?ZfyP%iFfxp?ce;D z|9*J(pUT1hSvhW`=5NqyfmF9`oxWIZ%icKm$nQ(yXK$+S^n4G&1M<)zc*iX8G9Psi zZql@pQf-lRrcCfGrEP8Tn5Sx>(!Nq17bSo@B*@YCh+P+83#>c4b0`FJ?m|oMVcD;h z!|kDrUI_+3-nO`X?d-RFZNt^Zx4ZaYktS2BZ8r>l)0R?50sJWX$Ul?@>V2vq1{biU zk#IcS{n~xbp`HHvYsBsv?(VC$&CWQ7>26X{;Fvwb?dB@j^feToB*wSx8jRwK$iU(+ zkuGbG5N=44UG$Y~^AJ6|O^2$a6G4H7*m4b(}(2)1YuF#ML^4hwY-__Hr znWRKqLF#^plUVZr?$SHvH%Y}OVtu@y!kY0p9d%Ufx2BzrWy2~%9~W0L&XDr`@{{l@UB3wlA!-DlXl+_0N{|6#~AmW zYad77SB1^JlW5%R^3}d&r6`Q9Eq8W<;qoquvR^W3ajw?pTj+M2Npia~8D`lN1#Q$& zK2ovlPZCfoJLxWa(Ev+8w7-|R@(ZHyi|#KqKeGM?ul(?61KjTSjj_ooQ@@_$6hrTP z>S5%@9#yr$IcIDw_P%Ay!CP0 zvM3P(W<&VYEk9<2p3+)}N+R*~$bGOWR{Fe&xon4(u`r{UlppV2T2Lt1f63%QXLIwp zzw(vy%tq|#9HSwcX|CoWT=vvw87Gzo=Z)R|NUO`y5HWbS5PhaJ5#z!CqjuPAHN836 zZyyltsXb$ja3^O9u0Uapy**Zcs_W}#CCv0U-r2X;41zD5)&kx=tGsbCo^dY=P*BIX|w zwei0CD$;h${?0~Qz3oy_4NVUVs~3l8metyWvXN_rh4q(87ljYB?7|(fY)>)<5N|ZlH6vnXao0SVS^O3*wly{1Df~|~&UuDV!aun})PL5O@PCS< zc@+u}*;DC)kc6jiB;)$FfL4;@#+-dtIP01`epwTd+ya$L$6 z_zL)PFTFG@>urcO=q(s4=0{QV-GE}c60_1{7hB1NiJwB97r+A0fU*_#U^&{~Pwh#Nil5}~T!k549xfi7+NttPSRNfzvjg*=#DCe6 zALUV|H=D;GRu7DkJjP$wP&>S4mU$NZ?MD(LH;*BIK8ZzyAXz~2qQ5eSLY8CO zy#P)+t^cCW)N34EW+MCY5Y&AVDyi;(LVv)2NQa7 zB(*!(^`8aL`E3D7H`p7VNJ89Wc;$nY_Bf`qkSZIg7y>D|oStfQGSMA`CC3*uq^mkU zy&knloUZJrl%^D)DfRTaDB{RkBTF|_yDQ}MI>mFvh&J!ZW0%dyaIo5(9K)@8-m%gf zmmIN~hEUBP2oS%q6V3f99*V$2Sd&&Pb<=&rRQROLEYlIie+LmC^L07+G89g5LR!5# z8nFNkK=Qxc=u|!BpH^xJjZY765;Pm;2^DP?{kmk+{u!;9zoVGcp+3 z290=MdpHNX5YKMtg2fJxLOWbCZ4j6DW1UFFA7KGFT5z)fT1>vgfP**e%q~nM=cmQ0 zW%Zb!#4z+2d0$nmV>UY-C@>8d9H(J&pjt%%c%(7Sc@%&v>PCuP_`k=E{?}xp`1TuI zcBvi8$@w9F-oX8{ZU0nZAc2$P`W@3wzhBz+IV0hr=iPy%0^i$q2)=~K_ryF7?SQA~ z9ZeV^-P;xvA8%R<+lObPjrdRtS3~byLA_ni< z&;|6miv*7E7aD>e$K)B`s?U!EyKPfkF0L&zjdh#xH>Uk+I!XvE z5hJbH?d5v{!xwspPGaY!A^fXXPBtI!(Aap-Y%-vjp^rQ(68ZJ8FP%_meP-G-qPcl) z3;1-}w7)ifvi;d5Fhkktg(Upwl$zp63r2f81VL3Z4DX@qPeoo9 z)tMrX+3=FwQ_~;Ll1H@`{}?iRG@$S=A+ukstP+guS#Kxe6=LN{nR*E1X<4LC4vQft zzLLW1F=K*KAEe_yPyWZRjtLNDh8EFM@I_&Q6tW-7(c0yBm=M_IMKM=%KGqNTZbK?c zkH}44MiRm2Vydt7y;AHsNv^J!4R{0auHo{9dO@lw4D||* znJ)Cg;KqneH=Pjw?OFONj4p{(8PQYg&B%}b-jeU%U2b~V1xlx$I^GSevj37?lOQKW9Mn+%E4f$yGhg zl)L}88ztg8-`L3<0A%(B89G>aeEI1HT(Sk2F?f7FC59s8d^K4FL{e3PT6aqu7?`lSDG|D2>03CVV^xTg+m4NW5yd4<3kOAxRqlM!&J zH5_j}aRa|lt&bLrCbY%zp6u7y_oiu<+6&;v*lXn)cG=uu+8OT!o+JcaODM_V&pxCLu8wpZZpIGx%LnaUk-=!?l4l8X+B-!>=+=b|l)uib;${#pXs@5t8-0 zU4!7P`N?`}Z76)i`0>s_9gPDKf$cl+@hM>JjUiGnb`MFFWTtkG_n;3#wt8sp&tOns z7A!k^hhP1_243q^Mc8a$-43&3$JR+9SA1oY@_^tedzyYn z;V2rP1|N7_#ZqAQBe_r@<-+)n^~M?34h-4l$NNs1Wd>tda`C}}j@JwGm3oT`%aR#F zh)NcC&X&dxigONtp}XZNXCp+6`yZkWbj3LSYJT(iePN-o(D9Up`v+&iLs|!b?ke?M9k~Nsro*pJ) zHXPR!t8kHo2|H=}Ja40MG)a=`(<**n&(?^UY1S@{-zk&NAPNY+8lCR z74aNA*6OQ6Z6f}S6xfdLy|0$R{AOYQD@&?zMVt4QGCk8g2Bvnsp(c|0u^I>@?gV;* zKdw8(gF({M&eI`VFe6LIxC4U`opg>lUl12E_s0}8i!|{EhD+v5s>a%> z_d%9wq(@RmB;k<65BIziKq*!CeuSS zA5Zlh@3*atf~(YJRfshEt+NZ6@-|wzw_Ac{|7|&ThE8T3rXVCOR;JUxTI6WIFrv{( z`lc<}0wnRWc5f=@*MzjIj{Oas)Uhaj%m{6Cm)i3WO_z-2%601t52jJ;V=`8Hn9~VG zdJDwC@2t}mI2MJd1lNA*47dz!bi43yW1{{-U+;{F`evlWW9Wp_Qm+cFs1xYUG}j^H zxUVh-{7o}F*Rs@?#q^y>BCm{(#MPz!i;dNnimC`LTl9&a{0tZBNO zL{kp#D|Vr4k2ZJGEXTF{8_)ximgtlx4U!kS(!%`NgpeJBaie-{m11U}L=a zK2kt2>J1N+AS2Ke;v}lmGdtM_wrwprK5D8YbN?7&2DaFywVPVr4Pp%GJm1 z;ZtMlw8q5Kw(S^1)6zr1BJ2+dDTVFjO`_BHgRN8fOL0*QX=fV>Fy5c7HLEpq!B5VkS+<4PGR%zjo$a; zUC%hrIcL1@_s2IJga6&v`mHtRiaB*SMl@gF7=bt{s#N49Ii)-uO0CF-<74t-tir%2 zh-!G55%POA+=v15U73HehN@-zar~B@y!qj=W-`e%ES4G>ye#aFui>L5C*_fu-odd<=dY>cleaL$FLN>1Bpnmv$W?@Ub ze8vV5Lx)(a8h>7Y?WR;ZD)`m|YaBqi!eC4jZD(_qyN1}?x{K3^MYq;NfA;fd&79WU z2eo(6Jck@+-!eI>IvaIDGw~^ zHF9|Dl06m{(bhgwINrSDwA=y)@UG$BB4aa1M+$GeBR-O|eQRec4NEVxlZ`Gt z<9QuuWDoeV-(*O!Gi_QVI6NfV-L^qkT$fsEwaIZi1#S)r80X+8m z+)>3vHEvVe#)^09`I1DkP5Kd-_!|X~z)Yf^Q3tXUC~t$zvL1dR6dDO+418@Qj%zWo z7*291vlIa|kuyZzR~9sm0>LoN!eVe!vW%r!??l^NH*3(wt^B`I!#_kH7=Ur;S?wgX|B~={d8UPj4X>sF z)ThJQk|9(Sv2ZxX(&vS5O!$?Z`v&!+yl* zaDF)O2~;XqZPSJ5Ih12i_NJ#0L(0pn#-tC}cl7wU8(` zG^(T7IX?<~41=b1{!478sVcIH&N!Uf_Euoo3u(6l5sx-`X_<7b(xtj=X-* z1l)giS>rwBMkQqbwPvH&+kF>0%O!I|dG1OpY1w_h#p3%V%=~E>)!F)Zx7;EUW|joq zjBz)P<=4V_2VrOrNc)-&>KFvAMFrjWt5T8nbAZHs`uTBvb`F(R$eM}t@wp3*27e32 zY*<)KeXN|YgV1;g(MQ)MkmOO3HkgUQdfDV1_qy5Bhtk?30cw<3W+57PJx$}Z7f%L= za0RO>-B;bhKGQ(xj>QSI4?8~@>>)vExFg5N*j>{c= zYj|%F8%h-#VEvou( zDsCXf23J9tpBa(87Bp6yQ5j<2zG90C?~X^k6@!6j zt{o>Wy%qI2r_oBPoaza_6t1&-y;I&B#Ktt(_oieR?DO0jwy@YmMdmP^XrQBY6Ck^- z*krYkvuF3}@*6^LsuPE-3cUaFn&xZgnU;jd0p8SZF6#>0(^gxnct@{aze}xl{oZV= z3oQ2~C3(^qe?Po=k)4VEA35_{EqGD>xEjPXddiq6JO+B`?2)a^y87TzSbQ_PiS~^S zEvE248{w=LnZS90wUK%oI|n15igbu+yjr9tV}gouNPZ$jebOYp*X#{bh;rg-)-y_J zse+UrovFazSFC^McM*W;KcjW_T_cIGpALaCewPv^=T}k{*it1A5Dt~0oR3@5<>K`LXqkL*6RitI9_MaXrXs@*p%~}(pG8i^Vv@YxC=z$D%MvjGF1-(Sq>fnH+qCP8r5BNK-;u;qy{cCYv}ANSP)1>+ZR_Obo!KAYjy)t_l4kxi0rBK@KF*aq z9`;y_kZF--e5`j9WEMB8I-jVk=r3z-x}_vH>ZjunLIt=daIB_SeT&!r#|~qnF^uDX z;IV~`kavYykNinj^$+y_1m|zaTywzhWDawChA4&bivEi>$ppxupkFU2Ld+t1emJXS z{@Nvx+o4n5w(dmAt@?>ocB7@@boU(lVeG>6yn@x1q)~7P2?`?x=OW$VOw(9KH1cY}Q ze7?f8^*;-5_*bJ%otb}$68ZFA_YzL{zDKQ0MqFM}&dneu)ACa|QE5%bTRM|LuQpol zJv(Iwb6d~H_MxU?AT-H8U(9#lIMh4Q`B1B}jNUs|(IEDoe}fQ_sWC?Jqu~ggv{Yg^ z>A=EtksTLG4$3~3 zWQsUYaF!88C4VAc5IJVt2h^rAo^l06gB0b6EKW;!5+V`#Hg0#^&q+*phM}jJ~ zIaTvh<>fxQs<@R9(1%3JmlWZ|g!~>J68}>Fl`kyWUEkOQwXGq%OG$99(ZAhmD5~rp z?YFCZPa1Zu=3V=bHUW3prt9}jM7?=cbo^ImKl}e@gK4v|-%OjA@buV!+&Rr=^o=iQ zG`kdzqV145O~b?Ln8u^FVxcmWP^_Ekbna81G)C^rCUS_yI|>@An?`0|$~#+qE+#pL zZ?46l({>(Rd*p0$4~$;m<4v@?KKDw2=ctQn#Rwd-&2l8O&peX?elB=ER`1b0mf!_r zz^km)*d)1wap_3meHSt?&l@^-=i@A$E(g;2=}j^BXW=o{WEZ<{WiIKCE4$_(0J@c` z;nLPJ9TdGpYR*qAkJZ?Zh=rW^$@yT9A(cb&f;7jPs;qi2?&<0f-A@%7P)_herNbd^ zhnH3zIOa{5rt0p=*^4-=^_5J)5`EG(hWp*~+hJ{i?m5a?Lp>hwPW*ICg*-pn?DUA-;mT^JCi??>r$oy8{t4H`pt4 zBtl7XVV_U7$GgL^?u2$?s?EzrvfO-eE<8*t8^tJ+dVh@oH0a0WSU?@3@})E47De_ZCV9Vb>h?yr**~xv);R+7Ftf1Cu76cYS&pMdy}9 zdwt^QD_xxoPL;RqCoYLfLK@iq(&jgOGTvF{#4|4LD-;oZe|-MRipfLOGs|o+9tjZL zK;91<%|PQ!v|Ejw7-4aoiM!yk`BHQZ+>G~&_zE9D1(*YBe=q!lgL^j~^T!n+WnMBO z=@MUtP$N2MkyMv6OrG%95yT_|qm+$V8c3rzjjai(w(LtrmDDRRr^>j)|V2l=u=JN4yqu={kA@a|$j z*hrq2{JHx8t}QWde_N#eL1YS z{JCW5^N)%QA*{=(t>CxSaw4m9APBRX5fp_M^_}lER#^f++`H@q{y6f&3<0EnuWOg% z-fM@6$WVJGfky~CD~gRyI-k3&!lWuh9F~SrT%(WUf~bFmyLicMj{--Uh zPB6IutQrB6pj22kA5JRs{@e)TMBKaBXrW1)m=ZGv_aRkaOplpBB}@J1VD&2H))&Bt~xPUU|-4KYf3?|OQ& zFYMPXW$47>6(pnm3XJc<>>c=Ro9iGcrWJO{iiYy{aCZnPm!6N@BG@R9lD3y2kjh8a zNFHLQ#VCvB!%iP?skHjgFZ9>xw;(_V82?;0{&%Zye-^-ELd(%h+07gPx9U};TdT|u zBndttZQ^*urjP)5ARpclC>Th_028uU@2`{_`e06}HPqw{LR)~7BKb%2nPvB|W|!{n zJ=FA9$HHT2+~pF7Cv^jJs!>`%7xnUdm@4U2AD!59R9NMw?H;ff5(-c9C#Qido$I`A z_xUSl=^hF^T4g8LS)sF^PIi7&vTG5pzJ1}m-_aE&#G{z`{$w+eiWr_0*qOX;+kE2n zZQLDn{#p)y2gj_Us_VSEJQ9tSPr~l>yqhSq6%%p$@uNHbdQjgK+e zRWHu=$&fz}iM2|=E>438Am6cs>7nSX9)x^yRT47?f72|7jUdf_MeYv6<6#aaW9`}qw{jp}CG9uOwZwvNTWq|@)flhTfVzWM9Y|Bs6*3~)UfDZ}!irmxOo~+v>vt0e^Iqy@% zNMN2PRmB=XqOWW+$OTKHiXXk!(Qn|h{p9b4kO5P8Igx}6zS!AadZ_i*WHBjRY!lKA>z!4*_qS3TCFd$H(ux)qm28>FHof41Pmm^=sZ8_iW58&s?J-xy z_oEn=)-Y4S6V>NKd#e%86o*C(rH;QY$>6d61J5%-lBDOmcAMSE4-Xxa!4F?~%}@g( zm~1RDz9q!qjfsw#B8Fi>2G2#^g^z+b>L%!%&nlDu{uU-+|Df^5okqD~oNg}cx6_`t z-i5y{hQCwcTdDUzqMmr3vSac7_1*pEuH}6wx|vjK#M!HM3lt1;%)z3&JnIO*DUklX z?Pl}}V*$CD?_1?8xXH|!pr(`0S;slVPnDa`2J+ys{|`a}Uj1B4W29E6A_}L6TS?)D zkHi!cXiwQ<%>h;t4GUyxbt7MHns_2V6wc)=4LA095`S!yA_ikIyW}xHT1qW#i8{b3 z3?F5QVcsWZRopU7mx3$lrY*mKTB0tF$EDLZH5lMk~%4pDfPRx{Pv`7XrAj??zLgy#& zW$nX8Be=+O=12-LT(P_9lkfRgK-j?yc&wz_zc6VnxCvfS@AD z7U#L8{b>~xp_S|PTP_t=$pZNTc#MDYIKw7{HT=k>yqrxk!&f_fC0|d<^9RBaqGNp{ z^8;XJ>y7EYC)tB(7(xiO(tMGuOGQ#R;$A8TKLpR%Vu7s3O0_6b{ij?uR7V~z%Hu>b%f0VsbL-N0C{m+6hdzK0{ju@k}Ewx|>uABaeGeJFuR9dP5+` zyd}q;9Voi&OU2tX?@Qq-Av2opl?t`Uz2Wi`ar@nkv?(zYPq1NEDX6+zQkv7DO=+V5 zo6jpLgrtWO&_mb$#*=+2kb-Z}1^Zpihg5m{+Dy);-1Nh#m>acYsr_5^GojeMZ$fq` zw~|;VQ6P<*h)Kyoj@ut*zz8XQ(kq?BhQAzGm!?vK3z3L7F&4nP-#q0_glutKLoc#tqJ$SzaMty{2BS@r`m_2Sqq2 z%~3SW^f+#D+TSr4L9Bv`TJ0rqk=f5oyh*5hX_=OVahH|+4`W6oz~-kc$2R=&4ATLA zs7RQKzjQ$sd4n#0Nd0yOU|?AsAev;6t>5yST(?QzhOwFUf3?~%NJ3ocWbs!Q4n>e5 zhAsZ=egRBiPJQ_ueLp6Ay!%Ejyc}btVga-h&$6JPQmwa%?>x$p#!(a$DFg6MX!Q8G7v55?FdAH=Ha`My}*2q;UV^ zvde!dN&XNHkcqhI}ojWUSI-36{GbOU6rD(KVfoIw5hKhGS~YccR6SFwm~u=N@?F$3R_#{f?*x<4{4WE4ov=n>|-+KK)N z#=6PX-mO{@WaRMrMIA4D#`%IG0s;s#5@zF6ffYWm{~pIa@X_SGKjFN8g&aInuKoFh zTkpj7C&yXaq5Fm7kc#B|^wE1vN>aoR$9XhJ9_NLzZ0|Ec)}uG-A6X9P3c2s;Z4eia zCEiNm5z*Q&)`(REE754}mOPGrkiVCHT5H#9cMhF60aAk9vwO7>Kc{;p5F7r;cXgWlW9Z6zDD2H zH2yK>XRQXTvf2Tn-e)$e6JK!H&OJvAeX%JGJ%Agj)&)^+I+-{Qh>3zl{6?|o({G9^ zUvey_7i_Xj{<6XFwelD(#L!U+A8>NLr(;*i4Mvc34mGH=8 zCXLJn(uAMbJiwYq*iK!eHgZuY9Lbd>{vp1YDwiv7uf1b{QYh(oJ-xqnvpfIl;G1;u zBZ>sY`IJuGe@&6>XMpV)GXZ?Mi39sIKo)XJ^z?oAvUPBgTn}zY!=V`1PNUUzOUF%b zs@6VgaP67B&5Q%iD7tb}J*dvkp7NiLIsR6BaUj!CnCap>vFXxWIf0PG_dDw{)DFnq zU*>#X&FoXDv5J5BEY=EmjEUi@Vd^`!o*0wRhd5!%5|I&+T<2p(nB9$CKj=|qNtB(T zYpKv$4Eb1Qr5?tt`8b9f`E;iTUu>$LD0B+JFl1oEYLOEcD0wl`l+4D@I*=gD@$%n} zm!rfUtmQ)i^1c=PADK)s<%Re0`9e2B<1-M)rtWxv+aY?3C=8U#$z)n0>64}4$ZsDX_UW8DVhO_z|G%1YJb;&rmW}tr5Zwy z(Jk&z?BqNEwSjyOWYy>qnu-}W%cn*(;W5Ty*!fC+&TPzsH;ax{~Ady5&X2i`SHe@SdvY+b0BE zHkk!xzuB+8?*sw9WxRHHZ)~Z^+nSb7&X4J91Y&_;G2wSW6%>GM~qec(^yN ziCw*i5x;K){_NiF7nzk#42>%bpw)0uAAdaVky){^0v6c!_J&%U+AQ5T9`d6_i3wqR zo!N_}t<2ZBGyMK(X|Cp$6tOkPheMDrbB>k;e|lq&%r}Dv*IqfGo4}I z!p3;@GAS62$aoc>KI}DM>N0eAmzG(62#4Q&9ACzz94}nLpx1HpW+6HsyXXDt#1V)> zf`nG~uWpY|exFMIrD$pmBSb9Ne&V!aB2Co!?c6I7i!0-oBu>ONb>*@M=L1SUs|tU_ zSw~RMDwu?uK4ZBLvbYrj} zD30U(xcB&(^`O`AbTS4W+fV%zv2dDmN7+`3IDxa&(5N3d&8#y~K)~+hzi`?WnZUT^X+dAmRdH` zX1sB0L#qZ~woYlCpZwk&>E8agMe~n1ZS}T_^mU5~6%QbJoZgLgREQ`LFQRwa1vy5T zA()cLUq1}_`f=1_STW`w4U=lDu%J~e~NNti7`lx)cJAG84~<_WNXAesmo zQ8@w$E$o`*XXD(*e)uS-?F4~sHS4U98xG5>X($%{tKtTDf^780*lYP3Oc8>XX{Z)? z1t7QcU4qPTWX6I_s~l?#AOyA>#7)DBsz~34iXzJxVDW4LKM$70F5$U1uBFy`zK2mP zdTtnPm4%YP94&$}!#Ypb5^u8+;DH=wBvqc6YE^oi*O0{vuHPu3;;m=k_SZE3@Rg2d z!YB}4nZ>zcv1j%>sMyl^90C~v%sOyZQ@wI=hFjXW{Y#N|x^XsBoqHsIb#_EZ|I=Pk z+JPUGmh88fV}OA1Y2^LMV{$r!(YRor>j|^F!l>^UsAJw-miD7#!xo$*G?ny2Wq>W7 z;IN9P5lem|S+g{I2c`2qIjC358ioo}u+rQ;U>^?RFtn&an@@}q6RUtB74M8k(TTRf z>g!21#;Z+JG*Jmi%clIamtaM5?(CiATCH`?rrk=#RXEPG-&a-x!pH)V6wX}{E30e< z@v2Sre2Zib>Woboke#L8xH-!6mHsXRAvcF+Df!pmHpZ{rA^uhI|3||S`l};369%4O zz0jYU?b1!-ZEZ+VisUvw7yWTq73&GE+$J>7g`$_B7_=1AIUES#Tvm-&cJt9}!@06i zdv+CjK%B1J4?|~-WDLN}JZog_zOahzy2x5|1Yb+gf@IsWFs)w}2Zm~t z2(rrZvo~lIgfVT@#RjpiBOjLIl?yOdV%0T6XzgnC*9hnz`A*IzxFQk1i-*z|Q^z+T zMWfO?w%lP&lgIwBkgTx(6`VRA$ZkE4VF$o`scX6WDZ7Cb4)MD| zG)Wd4J#=+~oO3u?FLpX;#Ndynct3=*1g9`{<1K#qds&Q)m*AQyo9EX;Y&vjWw%FJf z@69YpbMe$HNEc?XP8IWpK0BS|4i;T2&J9xD%=LiW`j8ERm~+iVK3{yiR{J$VH_m%r zg5)oI+R%U7&3@~L!Tp<%bQD3s(*=xtjO8SE26Eznn~NegCtYEkZ%y|2H7?#8f8$gNV#0 zO>4KT%f|7h_4BNDtzP{Y?C0b3l?a6+>+W-nJ@|vTro~EgWZ*ZHnWASgb2uK0uIThI zS*g+Xz1IzyV_IcBrguwSaa@j=>c;ykHu#eNjo|uY!lu{lI2BS;Kx!aikS4b1oUuoF10Ul@1e__yr5r4wU5^TYSy+vt&H@m}x7LI~I6hMobCg4C(K zfxzPjA+x*=BsN7kAr2dU_+U^d!SmOO3OeigVFjGg1wQA-3tp`?i-@=~?%{T=+X&(X zP8K)(0NI`w$tKul1ro198%*5>_1!1}3F<8EC2skI89W&b@#6Ma9sOb_^=oqW03HD} zS8}Vzzrif%F&?Mh$_s~|`eF*I%EyxXTeqg~R#@7qb^ z*|(G2^cA9SOhTE!T^A^jSK_ZdnvvO;3LMk!xxpsRo?t*|fEHlcm1}1%j5x}1`c(WB z0~m1XOBKPSs0!;(z~RwU+zD7G6M{in!j7{-#xgm+XFY#oYwPDQb>J#+k~A-7^p;5legJR$PR6xhOCOtf-5RVMSReu)80Tc>A z6dPG@6svOas)O|?B-UP`#KQDk>JcHQuVPqXuY~Wi~)J;ia~kH#&w=f zAWmO}_5!8aBD8fHv8f&yHG*P`uROE!n5)HM`ur5fJ&SiRy|OBUcHC7eTtc7@-D$f- zI@lye8uz7jjfq>tioVP3hXVBftEJ9jIeD1LPxsOcF$*zMhH^f>y@r^r1&L9;yx+F` zX#`m6X4?`S3_pflcAhOZtS7m8` zLy4U~*CR)e^DUl~Lot$&cFN9A=6Hv)?o^j;ZfCB06`JB-K=n4BreS`lX8{JOSgJ&c1SzdyhMfBdUx5yY=PLz@}1Fq*3 zEcif}M~#E2M}@)xG}%p8M}8Uv@&Sx+TpCqTuSJt6yE6g2O}iafLg%}0Q^f$Me30n#L^H`- zKBp}7MixEMSe7mG$Z576SP*=@QS_J$Hv=|ifTw=`3ofMYv;ZW>iADG#c{OC_1@F#j zA->%#A3-+xZ8J;0HlT{e&!J{fg365qq`i{d(reA}00hLM(!aqbVS4(2h+{bKY2sa# z$ljKEy(pAZ%yG~?1T+Sfio&xytdyb(W4TFm(51%J zRZY}Id1F&0`?KFMX1BL(1-E_({yOb}%c3QD$bX|6OLSTdXCm6%%ICLs zuBvhWRW(lf+iL6$hkEd8%%MDZwSC9GjrE*98fl`M-C9&VtZr*S*I}tYpgu$8rO?K8Jz$1kO$(9OGzc%ww!4B-hdmSO%Jj7V7YyZTqL#6 zV#44Dee(ZY-@MwycJveueD5^qy=Z##Bp7sE8vH-4#(!*cYt5RFZUlV@M6LReEzX4H z&j6;y)5?JylgHx&$c!Q4pBawI{8hcp>le%@+$Z2n zM;h4(1Mu;WCzNSp7SwnBVPrYI`!+@FhM<}_gWCrdcazX{y#s z5^QQ8RpgIrM+~C12P354XoyQGXEhgkkM@H#QnDteIN zCT~=d>1#P8PnXZu&*6TTwUbFEa}_(-bL8WO)eo|TJlimngpbz8C^gc2_iy$RLjM2U z`9Fvq6Gs-AOxuce_s>358$p!zM;cae!tfPN7}$_wJN@BEF2_HOus#(mz-YPjymflN zhx;CEcYuz-*ox8%vFl0U0Ec9}VD`yq?mHGIYJ5W%i&swd9r$`~lhIN1*iWaEKWV3! zC+fSb6T}|zO`r=#T9iRYi=Oe40tyjTPKav*(x_v+qxOKs} z&I<_YaHzSruoLhNe_Y{^Hq&{q0=lIsQ=EL$xrDRu!$+;$0M;_${tfRtNT!=*V>Dz- z5tXwxn@=i;(1OdW7iGQ{ow6Wd5^y5BaC$Y~R$r_+Z$5lku&N)%?dkmMf|nLx35Wok zf9#GC)1DV1f(aW;XvQ9$^B|a9fb#r!n^q>2QZn%TL^ZE3ocRuRlY5nUM?AfL|Bbb6 z!@*R6Cq$^I5VoO!yARP(_?{MZB$N7?G}RM9#0phRk&TK;7RyUpVY1T=F=iV6HwNkn zKd(k*!0`xXjQ~NxhjJprCQs*%u1q=)Fv&st?zl{W;bR!T_;Em3lLtHr0|HLsAv}6b ze9tkQ$6Fj*tC$r}D)0x8$%NeWhLyO4?7vQ*nJU{^pLAg1QARTL&V5bCvnbCgU8CbP`jHOowk4mcR@#NTHq0AW+X|5oi=&o z5N%{&=+%W#`?q!g&3DLfY3%X>ja<#;5teYY^ObQlJK?YVN*(?~ivQ(@)%sHbuJ{DlW5nk8fI2~Y!3p_^Sh2cH(nSnrCV1KRZF1IdIio0S5Mu-n;lfvj2+4qspDgK@9vsvK-DwepHG z(|oVmX;6rfN#|^CaP{}Z;%|LV6Fu~9S_oYf2mdA(Jv{@G2wVB^tl06l(kik0!+R5_ zINVg-QNysKjimKsIxy;8!4W7LD|?&3lN0t^5V{3p%(?X(#D zbK?^P2EHPJJRIpIf&ib3vubB1h#1yo2+4Xycv_r2Y(nky3n_39cz#hO6uaC3=2dKh z!`7gOnP_(+&t09f-xJ>l!>HF#+L}|PuD3WO75?#&Wj}Q-Krv?s9u!> zUpg(;^o=pl!(-uyI+c$c$dx8O^Gl8?>d#j}5=_4vUGONP_1neSVUTTy;f?ahr>5OX z6hd`ykZ+0u$$-|i1wO}$`S#J!!64mZcQMGY6|J$~o^!RV{3!i{y?b2wW! zUt&M6Z!f?sBQ>m7H+oV$kV^(u9(HhE8TCY_Vkdca6#KdLi7={k%YC<9oPi|w5Fg`5 zpp}K!<3+Suqo)K5oA^K`bIj2;eMyqxgWoQNeNH zFVi!_n$n-sGZHRO9@9V>JnAAnt$mC45XYOl&$FPK!%w|hY_lfYkXT={=Q?nmfh_xu~LMgT8B(7!Vd z_bfl=0@+zz$Lq=QBG~PiKqw3;HaG&pTuIM&A2D0Iw?}?0%7|jZG)kQI4HE!D#?lY7 zztv}ET%|ENVxDv%^f6Ju26fy=3&yxg*LqGkBFDZW^oYWt0>LKEox@UaucJOnm8wf( zffEVm4lEuxk;oEbH!HxmTC|>1WJO#_@DSW(mlS>1tLp95FyDa_JxgCK88W&?o69Rm zS&W~Fr_Lc0I%=6jRaS2{97to{osv%nB!A%aF*(8pGH?SpOG0P+PF55AM@`0Ca#Kag3=XiP;&PljG)L<9k1m|kN~?MJ;?eNi-$k6qP})rhR~W(3qv>XA}+ zh?Tcp^IdAdNbRqOz7hYta9%V__lMnQV7F)rsMlBB}s|p4a1)j`K z_Yp4-B;PbVqLz<0QO;ls-QfM8iT z8HifNXgT4Y(%ZGxoMHfulm{fS__Ish8E|OdgNQ>6zzvwkh>=Y=jn4jr=$%~wwPfq!O#7iV)v|2nic(n=(d2*)W8fhOnAD0UbBj0==l}Y`g zOEN9}Le4#&H>N(p-x_BPCiLsu0V2#Ay3suDK+7sAdw;)>dedLItT3F*>R)l$-m9!< zdo}qb{cYckQg33I4hTB}FiC`?b35U|=W~9z`LHh*fbpI8#OC)V@fjvMOI8ok-2>aD zpc-L7cq6#CJ;UW{*nv>q*cPYVT}6oG~EL@#17=02?zP!Gl>jr7ayIxB6aKP1|M6voui>|~La(FEru}$XVh9O^ z<=!x4)j-6?jY7cni;Sk@hZa%^yiYc!@9gIF#Q^+!@RzmL1rPsE*K}%%A7vbWn+c^n zEdE%^h=s`Zo3UW3%21j{0=BC`2cj3P8+b$+=&dkbGTGqa{(caV(woo$zLVcgg{U9T zH~M>#!CzL2@p6A~1TY*vS)me;bT5onNmy{b1poy=`oFPpGc*1b`q6K@s=zC(Z*v@1 zamd1J*TC;HM)KP1VBOIc)0>ceZv+qfqFBJ`Ub=_qSCjLl@IAjf6``K*bvYE~bbn@e zX7I=GuyAeeKYLlk2|@pbm-XB5ur7Z=4aZtG#m=#-^Mh2g{>gwy>-k<#ocP&;W#3;| z&4gMsx5A4D!|_($tOR`Czf&8h4(;JMQMn?EH2ZE2HgTg*6bp0~73&2rys#e>CKJxC zui}WVX~SQUY>E|^L(M`UmA>dCi#SVyzg3PB;{k@GYwV<{_mljoX!(L?A8aHokgr1u ztjZR!=SeqO4Hs8IlP9xqQG>ce7$J}Gzsfdtyf1EO%M7<&4LsmM^m~iG-*@VHhK)oI9efHwpT$0&CLzsfm$<|O0{3mYqPGZ2A-Ihj@ zlMJ1Pd~K*z^DhjyB8JiHnX3vZVj`nqjj zG`Js?FZdo~5*eIFRmg#Z_OWT!+Cos6NC3siuR=EgY&2^Bz=1bc^L)1Hmsfk2wy6hp z6^>u)Upl%s=imE=1hje?9(ra81L%1HT!yb=h4#M;3yem#2nxB@y)PHTeWky?J=bgR zeksOrKpMB$MHS9A1zaGpPL6*O+-_-qdU{i-#TMMw_Wk5!#LU1PkEFfL8>#;(KAKeU zY8w&OH^B*i!i{)x`$dhDgsUeh9qoPfB(>Co(X*}o;Yt4IB4aMrGzM?5TvrvVkpK!a zBNlg^Kmn|S0bX#^0fvNY^Q}T%sW%K(6b{pSLEN51%Rvg(G!F_jB5xg8$%B)3JFs}0 z5ObraXOiM>2Y0Nsb0XzmP$j4Cbl|@5xf_iSOuX8~$SQbaz99>)ceXAlbJJ0I5s}y( zt83eTfK$R-G{ptH>b_f=Sup9BSo7dwG?c2WW?uS55UYGN_tTlF>-5jjUB6_YZh1x* z6{Dd*8lmwVu_6Q^-xqR~uPZQx9`$c-w*nY9cG^(4Cmh>pk^S;It{n>(cetEkzUyoQ zkSk+I9yK|2laci~_dL+K!QU&1OpM9b4#Wn>SbqyP3Pl~cH#{J*5wcXLz zd+so5d6LCuD9o2tM7ySV%pYD=VkvKZ87ozaLeA>Q!-~UJfLV9R*S(yq=1s}@*S!wc zXj@z5(|=sEYxh9n^LT#rxE&=KUodOBPJ9(b(ZN|7UPHd1$H*9Zf4L)oGEy>-DC&eN zurXo2eT&ZtA{6(bJ`A>KmLGbZOCN&KgSEHTY^a{}I1$kot7}Uz{NL%RUg5R+7!iCO zw+VT6iU;vJY#C37LoP0I7J0!H^(rbM&Q=faZINgMQVA#Sh9PE?J#@eX5`-6+M#(jX zFvpt3N;NzA#vxj#@nV3AxfDN?ot(byX`i_&tMI8Wj=I{NX#Bfa=Hz_sPx~z5c!jOs z(wQ@o#3zDjmJ8B6rbt%f%sy!kX9R6cnyX%QQhBCFO`_0F`H3IevVJI3;3|hw?hfZViiM;ZO)bU{`Ur8>&K&Oc6$R zwGv#74(9_kTlJ>&*=VdQF*V~BlVz#Mi+60jDoRL9pi4(3-x{22DSv55paJ|q^iQRZ zBtD~2NFOTjU%7_-O^NZ>x!%SfoQ+3vWq)? z+p6r?UE+1C@3%(5?h>}-mm_c2Q|q|b)7R*HradI+fUm-H-nByRG2nLuX>0~F&pL)1 zYM%*VJ+BWV6D{G*^WOH<1Q+gBJFXN&PzlurXr@WLxnud!{f76^q}7+3@h={?-*Af) zA>0a_Hz3no9S(r#qeFtI<&%?x7=OntEP=g?Vc1Ada>4~k-55eJKb~Xe^RAVfffhU7b*33~BZX|GsLrp?DH(VS@h7!jW7aYg&ZU zl5(C(uIRbYj=o<$Iv!vBp?b-9Ic5UCJpf{zs5UA_uJ4cUlDw_XC^Gbma?r6AP1&!kcevH?ozvru%75TAvX$)K(59Mx6RE^Ajf9)l-HSKCW z24kTXwqEFs7?OIS=c!lH3ciOgis+O>otL`*_c8*U!97jM*SE89t(#G0(>5b~3WQtD zi6ii17|u=iIXUxsRZ}W>GFYU;99jLW&lyR^_%G5QD@L-6@nwk0X1`NN|uGwZ055sq{*n~zvAg-*C~MIs#<*a=d|BkR=ka`3`Y zN*!qEulpDnF3BpY{PSk_JqkX69;X-dVhZD0)$nayI*lRCu&+M#*{LsU$K`P{wIKkp zLwyhtCSnOLj>osGE)5e>qsdvGhvjMewfXrZ<$~KExy_$r?*D!j+T17mnIp(bk$`I( zzvYP2792%#^B+~{Wi%F~GW=)EMR@Y|h_s!8PzIf-o{F|DldjdiP7ZefP6a33bM7TM zY@JzepvVYUzp^K-gQB#i46klI#qwjt4OSsnPG3_k-Y&TCyjM z%pri(BhHM+GfFlYW;q(j4Tl4vt>+^q5&~HHb0Q8Ov;wJo&W#7?+yTVw4wEET27Tc; zh~jE?CnmVgEbh}OJm%9{pPd&gDtrpZJm(CiJ{qiV@)DJZmKT#wG>Y@E6m78D5|Lfk zM?|G~4zA^juS9_k{u|hTT<1asV*V9o ze-xUO8ER0X+(DQ1^(vmhyOV;&ZUs8Y zt1wH%0MeV^{c-=+SzZ}&CzX%$|qI_osBubx>`&_YMHxC!=5fbgDWz%OKZ$sNT{q#onaU49gjRfMnen-wMA&ea~ z$LmjIlM4zWVy4MpX|WiBwq|(b`10KTt7sGpS}r|301y%Uxy*(Be=_C#0!s=u-LA^q zy+N)!-w=-X>VZEU@5`IJI-mGe<_sVl?*%tLJVrR)p9p)2mY*~ieX4qEF_rlg0z9&>h2UM)wXrg2eqdTKxDRUc4bnrz&I$Zl z(oo^BHwLi!ClL_-ro(S9I{)9p-bl`9qb0+z3#8~9+F9lRDVij#>vWuM2>UXG^kUqE zLa6ZbdFB3>_nA;P>L}J3vc0g+Y2kmkf$zq15Nq4@YlDqh`0uKz_5;3l zblH=+C)1^KPnS!P2}SZ|A7^{d@2`!{$h)l-B&6MTyZ`y@xyp&vo3=d^6jVE-GiZh@ zdYkRxAbK0vZavA6FN`A3$mg2*DuJ&}8AhAmg;F9xKqtiqV?A;!n#OH^L`A?b`L2Y!e_i0NKk?^8 zLSIi%NjI}G8t0SY8N2Es8OFWME(Yn^R?!K=A`$bkkmoG<#&B{48p_+8C-V(M-W3qR ztdbkJ=VZ^;aI;06|E{C6M+*fgl+8^=g5A`G0KB=BFc=UQm=H!}l`C-s4=afLP*bE6 zm)5p;)dDWsm1%3>)|+K&%RFoud9{m}J|g?a6EuRD_-_0M;pgXFn@H4UC>5Z}srt>5 zVDuCBGvB?xpd;Yfo&U!y$u^zNgUR(-`1F>|kugG6A!s0w{LqsH3NN~vq($_=RRG>5 zu4ORsq~>OD&c|N%<$AY|ey5HOPNE56n=g;zyg5HS{oE>Kms_`IwiI11#eu&cSAy}4 zAs`i0NVU8kltRmIaG=vNg@Q@7By@+|n(YkR{_qaG-PUwzp4KkgO4dgTx-@O}d9ujJ zhk%zbezpWPZ*dvTk?y8{li^;L7&>5?#v}L6=`^QsT3ALKcb*&%Io{M) zp-;kx)~wLhO>T2&+1pSGXxO}XuAMmN(krPQRQ*_}rjS+vY*K3*Z+p9|w-qBhy^U=3ou$aGfPfPSV0!656r zrn-S$K9h>}d2mC-`fWb_6E`OBnOmbZPjQno#%ETiS_AUt5A^@{9k0Jl%{~U@eetUh zorx;+I513r^v(NRmt@-lBtz{oDCyw0$fr2rfz_S0$Xm2p;v*2tMu#p^u6jR+Po_ps zwn0n1NjvG8RB2X+78eiPP}V{`N?q|S#gTJ%&&Bg1G)3N|+7Yy!$_wF2pJ|664rcSC z@;ABXh}{nyx*>IqN~v0ehscE_PD(CXr4|pH4zT0_#k`qa&WdGxZvjtHDyUpnHy(l7 z3V=u*3|htOX)Wf*`*@`kp5uK7GLiQ5*$aX2(px9XwNS4XKoEFEH}Z^SNGxtK-bgJ> z3MI+sHZ%3IoGxYOFt@UcdxTGtmGN_(_@ifh2iibjnPblYj)YVidmhfoAt(idx0N2T zw?{D?9S&JPvgi-s(S4GxD^#eOD$X*+xXf>PHReLVYNLC^a#IFl5U+ea6>y4EE&006 zR^`eN=21)HmDbp{85RU1(fG+u`N6p?gi#;4JU*p5WOm`<+8W~1LggqKs)uFFS zEmpOvGxPLWApNz9vzpL(kRrj=#YXJSP(thM-AsZhoOI}Q>=I^!&tSWdkn`w4U0Yyki!@3Kt6M@x-SPAJHtks%3EXA`c&fh!jZ$N*tCgRm~;?PxG?Jtf^pG<7ds z$T{4@nFuxQtweTd*$V1wX?OjUi{#+@_*5CKIdaQNez$kHvW&-;tYusYXqWT0zALgW`8NU6RwYmzr z(IY=ULb0%z1n$R0oe7w5;IZXjjj|-2678nkeTCCvfi`$cpo}t{s2bMFS%9hi`YvM>h^NZ=kq=ORHRoVrRa&rtzjDWct1_s;Lf!W}BY zp3TKdtOk#VG#||LgWc$!(Fdd)#>R$-0`Pu^W*{bCD^P^UG}`CD#00+Kx-tXSswf@^D9*Uf+&At|+C^&=)FilSeaH4k)k15) z0$~68%mI{qRW{78h5XOleNZ$oB^?29<8sdhvnhrhpT{_&J4#uJ z(Kl`>3g~nvE2E`9EGaHM&sLs2eB1CiYM@w}^)RT~Hz6e4Vl7!`|KhGm|7!yi+omP&^Yw_bE z{N3Z^ND_1(Eh@oU9cE9Xu)CX5F9xcWUD&H&t^Q04y<5-vlkq}mGV~~u-f)M}uuD>F2Mrvb%ctwj{5DS?XB z8Em|7F%G(Yu(3H;X4=9-EJU-+zuD58{Z6t`)r&>0uNXm5ZSdqXEEM5$xcRvw`u%t) zeiw7`Shwm#tLqosYzE(t(_sNEYCL*1w?lgqLkcb!xPCM~dSsP!Wk1^fZEcmU<&qEm zzV*jBoNbPEn!8>*;W;xJ)(ZGj*0#BQHhCb>!Rq>Ph7^%=IFQhJgIY=o6xqBKU%x=+rk^7!L(x@N7ll$ z)H7bl7ilP~JZGtX%flwb(Ik~ILhr63(s+t~H%oe(Dfx!$AlWkX4#5@cnIOW?!9Q^P zSphwO7NEg=*#_V!+0${clUS?;;KuSE_{BJlx5L2Q#Q!YkHI<@EcKKfj$2qT(K<+@2vG;bJsW0S48vq0|91JAdb z^L3e8hS0%W#C!0iGlE73RGNikKy9_qHa@g$tehlmfA897QzhV*1seLL(%vu@ai?1T zOaHoY1cd6E$Q?SEN>DTtzvzMJW=s3sX)p5!o{y%pZ=AbE{=;H3DW-sN{9U|DiNg1% z1{K{Vq|Nx7!hLrjE+O+*g?sS&q-O6)Y$?id56>Z*B%VZ-;cN~)wtxX)mFBi#0Wly! zT*JPExh4X6ut3Ax>e9DibU_LdUNvRff#;{CCE!Q^REl@b$8cPPcf3yPTq!<<>uPI* z12*lsvl#`d?vq&VyMHTSPHZy$We)R6FF#W@V#*eP*<|KXGU8NS6fp&XuRIlP%$JTQ zQ00Eke6T(#aX-lY@!{?|?dDwjG^+%9o!QAhJKk92WqSQ~ol}{9OQtrTFLHvo=!i*5 zzzg+%Yfu>l5AxIT-e3VN>IY3i7|FN&N%_;&O=TD^TFZo;9iy5hwW~t42d(0Z(RwhI zTjz!?mbec@OEhSNGgj0)bb}GlJwk6^;I(o&r@50AeQ^)V-GE&+aLCFsv|;Iy20L=^qCpz=)jpw?=f}t@S%4NoYz7# zi%imN_mEbh`nVf}sbDnBD9C42XaLe|txZuFDnjC0V$2hiQ+(=X9aEt5h0EQ%LC-Z5 zOZ_XR7+3?u>{2r;9B}&YmlUtv{3`0C;47?DB;(8Oq+RyBfsWM?eP9+{Mf7qkJL;`Q zu}S$&$IIF7i+oR!Q9sW4NCCcYH(HO5{j2q>e^IpZ`x7#%q}L0W%7xQ0qQ`qSO}ip# zF4+mA9-6n^YY|U1uu;#KiayIhGLbB_(34EHPT9B{ZKasb3852&<8jL)02gxM4PPG0 z0;MRG3=(Lm48e6>$zv_^V3ii!AfDd~$T*aF^{FS@2@34o*Y~4OrhnP@NnxHH({dUuS!cX+DT$u9w>0nxz8}XdQOb1)qef( zvo_BX*UNg;r=lk})-AKM>OalB%+jB5?3@JS(IivoLTZ93{6w*M_=CnmYykDJqNWxJ z`8G0GikjL)3vsX&!Y_r|9M;0aMRve&d_-6QCbS7_gsHrRNeJi5`L{03twebc(aAT7 znurIZIp?HVlW)i=^R6T-*@n%tU!hcamnkD%NX?PZ_fXg(3C=K*YX!#tsW#U5V+Zty z3P?7X5P}_H%(jJ|I)5`ibQ4QCVstxDe8k=r!!6<^tUP4=ES|^g3j1C}!Sl3}>tv1E zaaPLdFz2Z!VsFd_@<>YrY})hNPbMqnofL-Fl)@08X&iwK7?cXQSdM$h@zrXa`F*gT z+8Vc&+$x`gd_ZY$NdN)gT7NEPophaf7|0hJU6NJtJw%@_9eh`W8lkGpjuqPsH z0y6FHNiVka>XJJa^X(EOr8yj#HgApbkF`ii*B))|eQ9#Yvi-B3Aw_FsSS5zJ_w%1X zZ(#5(YG?w|Vql#E!PnKMRY(uBVhMIT9YPr*g&Ld%KcnZM2;v_D0E?b5owQI%05H2# zlPDGwAhnvD6bqBIPKVSIZHMrmWud1FX+dqr+X~1lYu6+o;Z#zP;2ldmm12Eg zJ|o5Es{(z=NUGDfBg;%_Uf&-!u(r7K$6Z4PGLL1czm7(O{?#}hDUB5}KHab}1&ilY zE;Z=gnpFh8l|cas=bTq>*-%=5Kw#gK$~KNggLy1w%sO63&Y>c8E*kue3Y^!V6LvF^ zeKqf8IJFr853HbWnQF>}P=D=5zgmYp^d(|=35)<(?3(P?=`Y#WTbt*l+@tTnt^5VXE^Pwecud`no`j|ywYMm*Ty!24N$zIN?1Mwx+u?a z<(lh-m<0G0Wex=dN^=qHM(r}5OmVoW8HK5hNYiNdaD9_qj9lgtd=565JdL&#V=_e0 zNK5C%L$g<~i16(YuzO#azRMy`93C_bK6cV&2?sZ_s1|{G$D6NXz!N zc3WZ6$C?SZua}efJi0XZFg_RgR2{lX;=)65pL4d{XNUV{W&Orsu>Moeccm32Lz$N- zTIb?wZr8otEZb*F&Ts|SMhboL)MyP+Xxw|dx`JI zJzMvDwC~=J-B;2-y!QX!-~|@X%i&omJqS{@JC~v2%BqlaHCk)3@6^n5_bXsJ@mAQ) zk;m3Jzp3@RmB>ecvJ{wz>dLWU>dJ$(n6Rl9*dFe;?(3=M%fBL#e4Df$Iw%Klk|nD{ z@9s`N^*)RLcyWJyx^k@E%2_`A%XHJklUGw~hL^-|1JC>z*Y|1moJw8usO0t=|E`61 zDZe5!_m)4xr}HT2;oWXg?B&L}E+0i&KCGDD;T|Tc;G}N&`x9sd4*2q?L*aLcnns{k zNmPoVnjZ}7V@OdAgY6Kh7*sceIb3c)6?f1*;tm3}5fE9M-$M~Zb2!r{3Ffs4Gj~Ep)aGb`$RfV<^?Q7iK=GdLgZpAxatlWz;#w3uS zRZ409TK)-l!9Co(T*anCdh3GV`_wjRuWsDrNhSAOADw?CgC8GHk%bkaH+g%`q93T^O#rY+@s~fvbE!5nqLnb zV$c0Hp%%;5n-6xrF(a=oO#OInS{&nlI)4Q>e@yxvkxk*p^{vJnKLELeo#U`Si=e?V zkh8(gaj3@|JwCMurWq5OtWNZPN@AoEF-{=veOx>kCwVp9!*OYSYGzJx zy5qM^8a@M>hi0(|QW68zQ4Ff(JUZB)dmjRMp3Sv*T0;H2bbrpZXu0wyA|2@S#(8(RQfwPKc zK1=9B?#uV-(zuQcC3&-BC&&A0eIOdF%6-!0a-x8lLL=AXLbV81#0uWV7N+jS>#rA_Q{W!RQ#4NZ*lZ7q z$ixS;W~exogwbjrk|n6r*eVymCn*9~YtD9;U4GG+y|+FhU12$AdUmxdG2)UGdj62D z&R!wo;%79 zD$PBJt8d5e$7`s!WdzD2Or9`RsaaLXRbbSa?7J}|<-eZEVBzrI78CWG3XfM*3M{Al zf-cqZ8UBx%o4*X?Rn+B;(D@buzO zy{{3mo&Hkybmw+MO#tE1p;o=)tTh{m;>x(_)??#nDfjnxE{?3eEgyMN@ki8$WM10) z6!MTwINFHq%Z@7F2{IPCQY} zcA#WUh6kNKB|hU%$DE%YfaAuEd4J(LF~R-HR;CKg1sFS@Tq86K=6^{Sdxv^#G+UZw z_0LzgLMfo25@;U6!B=e_r>@Ye6%Ui-!SdlfX#(}!Ru-gR;ydTYtls^56I&%2tgYm! z`|k7Ps`F!s8d&h0i#WSl3|2s!b}@1_BHQzl_cvr?15+KXbMFQBS!W@<^i~RTrhHdm z6)&vQNQ0k<0lazlL|40E`7gOw82NNkSHP(^Q_Zvc@Wabul|`OeJo3k$Q0|oe^U7mC zYryxo7X~;0^vAk&`9I@c+2CDf^4~%?vYthdgKwU;9%Stbq9lHTZTN&i5I)5RvKy}| z5l2Q>T%2cv!e~XWqc?DTi(r=!@b+!4jaNOXT&|_J_B4yT^@Y`%)jiF4q_YJ+dLHa{ zKaUXOH^_d|>y&*tSqNXIX=PUa2ILay*iDkhQ2Bdg5+o;xVU-=;K2jv#`BHL~M9`0b z{m)16tLrb+sP&{@J%S_6QdQ4m{_HxX!AkUke6qN6b7R90>*WksK&GnaRSUIaF>d#D zt=EMrc`7%s4~18iSHPw!aOF)UoEA-TJa~q6=Nd9r^N!bDeB1u_gJ-J70#vzyKOw{W zsMGX-S+5INz3cf7r!PS&p-_Ei{Ns1}s=~H*(AV3#e!H+TOQ(0Ab>EbNt7@ZmGH$)7 z*8i00^=UYcK#f5l2YSatCuMz2!u8&c;OEL0;N|qpVB(ydDJIVKjxoOGm@9v0`GCzE&aOR`T+>7r|hOq45O} zTsf+qE9XZoi<{l^Fkd->{c0)RM*mwwy>ewvkgo+=hk{$}Nq)tmU137wjo+lx{DZs>{k7YE*JN17O?d(IG) zsio_?OJ<;G)&^+#j1nYDs~xV@-p%p3b;&`n=Nasjq$M{!%%bN9_>2|q_fPs=xc~HI zduG!G5sR^U=l4@Do;s_%Bl1f-@l&@m4)gc7fNc~jT=~a6Pg<$t^VA&(?rWyl^VFAv z>24I_?L8`dk;;oGY3zNxq=+Cmnv7q;0Jbn(jny|{Tg9)7h}I3MRcd(Nl<5#M4%OH1O_P~G9b?&(kZn-Fn+%(Yss#6%)e)ZrDk5NwknFVOpPhd#$MJVx?9R6U7CAL$0Ovml`ChZq4>EI9{!(; zNpi&G)0#&dSluU;)5J*TQ=F%gg>^KqVZ{sxzmv^f9h_|Z>^; z69M|Y6K?d{2w#5#0P1eeRd6iVas$;)pPX)I*Hx<4-o{b=d^wp1+5UGM3E#bjSH4NY z)kF(jFtv5Ec{HAV}OAg^!b(8LjzbLncwdG;9IZoLrI>#WUlv z=sjv9%oFviTdxU*dOAPS0SN!tgZ>4`R5;++m^$Zo@|!pKox%zH>J2CC1s&mm9Qcvk zW_{@^8;~~d>m=d>y~iTTq@};Xwqc-T>3H5lk3U)qB69<|WTiJJy#^Lr-LFI_EU5fj zVi%YmiuDGwu(P_nKDTHVB}fR|F2c7WoW_igxQz~$-hO(A*i|nRy_;HXJM`d1^OcVm z-QKhRlHtq9{pr6; zqaSJXl1E}~+DZ8y*c5>Qq)j*EFbb0`F);`?)s5j9m+xFGk9CGyg>no8phGcu^3`uH zrPy4&8IzX1K$Ds7GPMWF1noI4=P&5KE08vhBcw{t#K-yRT!02}?-!_RrVW96ba?W3 zpiRq$I}nPM)~1jR#(2ein>U{PetqeQD1s&B(a9(*^=z+}w2_3x({|H~_q zzIw%UR<1ocWIFm)h~9!-dsnDYW(nP$UGGX1kkTsDu`xQ&;cUPRiK` z%?3-M0N)^*u{fl^`Qc0#DL4~~43(m*&(UiUC&~?EXE5)Eh`$IN=C^!62ld0duQcGt zmHQSRNO?skI|yXM9jCv2;n|>Goa1QfogY_m-CBO+_x%~qf0^rvHvi1^-TrifK0hJk zld=`&JlGp?qS|Q9u*XGLjwvgCCU$-3d`<6M9B9P4kmtwh6> z+6CmoKx-=qFl{Ss>Vub$r`eFkS!91MH_UTg|L0tvd_MYl|Ge)Kh)j6;_QoC%ub(?P zwR+GXJ@Mv>;1t_otFO(QsXR{F-YP2TOLS2fifLJ>QA+I+SvqHm0<7ZSQXG3D8c;7jz$D#x~U&qKxH^d<% zNz6GyRkw;QF6)wy7MNeKv(a=WF>07iT5LjMl?)uZ!47D@7_f@kUgmYq#+V{v2kf;G zVoD3EJN(pM{-pLhDHO4r$`;F{eA~oJWO#ef>j?x6$B*34I?D<7Vhkj*FUcw`wtcUt z0Hmu|8il9JF`xG!CHGxsc}jT6Uv7tB5b^`>`sP#L;bIyLE`|VP7|)6Rii?r0-UL*< zcpGPCT7r&IVpUFS2$<_8dniS28m+xiJ!*3)q*n@yf_LpLWCF2Ltb(8J^+F_pd1~t= zaDNtwS+n@&Z0CzK9&Owk9{ps4DPlfyhZ+O-hP>&i(7Yf`_TDsZ!cy~t^5_OEd7hclF?IYxRQYye#H0z|6T!S&Sahy zAQa+a3$AVYpyGWtM#%Tkao=gB6t=$V@Tm9~ zm+5b&rbpxdf06A!Mz#|r7~5xxeAXUO#X&0~5yVWPP$gc<8mQ7vvH7b2ysyJBrI}zb zzO0Gg=|5}_SqPd9|AfChBfou$h&4iZm@X@v8yECEa&KjsO@QbSqagLBQTT5GHX3|pJQ`5^S+TLEoagR5wRGUHG^ECS4;ht|M4!;R}aEJ z+$6oG)#M(*=+N%j0hXK~_66g;YhOxU#AZaUXSz!n!6);VR6ZTfD&)tKdo7U>z1vjO zu=3ekrsuC0EXnZDd%~@RXCrmh8Ld#eZP2QJsn?Gv*_a7b) zU~2!%H%?x{HozEDYV3aOrcwe?K;Z8zNG=EXPuq&cF;MBRGYrAmPJf!?8HRbMe+#-q zwPYwCdu(3sy7KGk?qqT2b_36*i5-bPul(?pz*0>b6<*Vm$=6_vGvs3vG z8w^X$R30#BK&v^JTjAhyo@uA5o#2jFMDHAkY)$Rn{-$9-f2@f8eZ4)W#XO%*0t*mfb)4eY8K4WjNZQv$I*+foq*UW5l*2n@ z5>gX-&prH4iz#HXk`z67PKIm*kvhSbhk8I+aus$8U)HDjUl%Kuad=j4w+t5$mm0li za9w$K%XXwoO@H_H6#VEEbp>O+{!ACfteLbI{$2g@!Z%F4Kn?lF4VG&{7D z^5ZdZX9Ua88>L3| z9FTYIe7Da2h%?p8@(9aa9N)&)cjFhDbr7?iE}NnHMbkvhO{AM*4`}72r1B!f29y;u ze7@v}>LJJ{6(~F`1FbT^QptqAke1#Mq!qXkx#@84Z@0;Q zwK*ktrTy=7W3)p4$00cQDGh}7z^_rbG^)xx_S~5d8G$A@cei?J&zyu3E}z? z%%@KjW~2+<_YE5;gI^YvfbF!tsj^mXwuQN!3DCEhb2+`TsOK_+u%Jp??OVk9+ULj* zi+UPlVUTxi;<87Q(np|cq%077VB|;Z<(+-#8cOqwY!e z73cY$C9(EP#O;_Zh$48j>>JCe#{&%`qUP}+SyRqYt>9Ilo_gRp+{@%tN1`-U-e@;w zT-m@t^dJl*X*KEE%`AR4y!{bFCTxL6|NZZit+J1`zq4@G>gSu6%M5;=7dZkqe|IgU zJ+jH~7p}!ZL+ufh*A|I!tw4{mUK}_n+&9 z)Dgp&n-2Ths9+Cmo@v+bvJ3^LE>Ox z0cA`@5V+1T|66O|{Hr{v+vJmB)ho{t96^g`c_~U?mqAhh@v2j0+V`{wXQ_LTTBDWL zOqpu^JF|Kfw=pq0;`*7Ze`{TAMR17L8$xo#6USOjec{@gYCbea&}w{2V4Ga4e#k{<$yp`7hS><~sL-b-7c_DV$9P>(p-% zu+|M5-k>C8y57Mfr{~|e2aecMD23SXp%i;SJnI$v}(i~@fQkAP_ORo=;v9l1UX?u%8(`g*7= zN*Y4`-0Q|otqMa4e~V@QTQ80A2TMiD9-j4mc*{}dV0V3_Mauf{#KGnqc$(JADBZ#N z3||)ly*zvWn7vasRoVEbl+-(_AHQ1lCU>Z*lE{0;Lun5s9)R; z@@tN;cOlNV$PIdx5kVM}ncWn(5nnlCP3xw<`bjW;F%vQCgFd|NI-vcffEuX)FfR`n zkYM`)cs0S0Kx(b}sX&^)RmL^K)Al}x<-dH$31x{AUraXfANcEQwEx2fzZ8H8CZIfi zjf!C{=rHQ9uTkQjK?UYNlLnjmn(JL^Fh`$Bm+8%BM_Gea7mpm-#iur!4>00%($ zzmWKL!s3;`k}T~0ybcU+`&H7!JSLq@sB{4I0=71uZ%1%n%f*h1`@?om6mA!76}(6j zl_)((Jc)tA9HhC8Vz1dSwedOh6ZZ(7*O=N^p;0kN8K0BvK3(G1+L*rU#6NHMA2o8I zKJW(-l60u0(*Y)(H^BnIw}!1jy}0xfL(23Y&T+*Fbz`IqjqXT?yr2Ml)LZ|{W z4(@Y*5_0YVe+q&*C_o`2ZW2gpgsG4iw8UpZPB?NS{l2x3M~Ag=ACAUE@V!*b4L>1G zo23qs7S8h#%(P&WVQucDQbELEQmWFol-hL<_K}Fg;L_ntHw@zh7WY-LdG%Uuk!CD0 zi_K!(Wyuv9==DO3nTb+tIOA3Q_cGOaV#5Yvm2+3jlQlYf7Q=5~ie|83bGD_vFS{u? z@dB;@M|NS!n>@W}-g(nP3r{&Jm7UYqvvm`K*;n*rcjmJTlBRByYCmW`wJ=zmA4T$B z8B*hwRUy2yTIQ`z@E0cXz5jbljzPOL5UYeA!;5(FBe5Wo$92`*v%2K9|1KX%M&EYH zjXWAkW*qo@=}4i*pTq)DEU`cpODyPFt^LK5H+~ZQmzJCz<@pO(X|d<7U!}#YYwASZ z?SS2FjmTX&DFN76wQXeym`)HL`1y93h8bhZp|8^W=_}`W70K^I^{Qbh|ISkoJjYMI za^%&k7Q~A~G@0Cp%0>L-7*$|DB-@YZTn|!yr`W~7*OwhfiS?9Zv58!pbuv<(F?dwR z<*WSQmB3^fh)^hH*d{R^)UdytZaPr?Lej7g6G`nGDSYzzZkQDHUN@MpaifdWY7_;D zh-48rmzF$GUtkhDz7PFu*X#z~DIUu>gG}VSb}LcskUD6OqA*$PR>K0Z(PDL0GGZcT zo{_J9^KFU^ZZHx;6t*XuZt%nbbk7I}q#EJTTRafn->|S;;D^RflXb|jp$<#Fc#$IC z-Sgs2qh#E;lI)Xhqgus&a+(UI+BCQtCaiuhE)F=i@|c5x7WBoGirxb24-Yv`MjW@~ zZu#^{{>e%E4!Kt-58ks~Yq*oxpY>$9_Pw22A?}YH1#DNx`tO@{+_Aqm>mIVn_@jUO z%3p~9d=B+XoCmLoBh$YIq*Fq~e+DGcmGauZ%`7JQ>v+r*JxGCZPz+O+JTbjcd|h1sl&R((EWkUZ zCHKxq_9u&hoHo0)gA1;s-UVlj-N<&nUW$aNUE;MLP@NX}sMaXn5z~VcDEyzM(%O}e zki&bpchS!f>PL4E(3?qdclU|$5`A$p^zazibFyu(0EQ!E5J(^kMcUS z&?@IHJIOOfPyM}Z!s|ZN+n8do5rv&7Aqj$(_Av1Ti~XV{GdKK+`4jxuJa~qyS!MV| zboNG%+%*FB{V0%yDz^mwUwgCL?`GX(evEM#URaWqr-0Nyl^2BrQDbbF&=@_eBw6H zut)Bu(hOHflvsUatgi;LlrzkfRq^6&TmRVZqry0b&HrGcZ+_Dm>v7EKJl@|~*$ESD#ht_Rd#_0a>Zwh&&9yZ3q*NCb5ze6$ z&7*VWeQ&KWAf(lesqA$t2n1S>{rBxY5A2QwChYVmNQq|OLIMH$$r&n=4uMXfnw+R9 z+kuAjRBuYcR7BzPzOk<#-eqM%-TV9wRB5slY_mxV%ypEMp2}?WT%LUSE5}&;Ri?^1 zv$c<6Bgs%fAkg_TWG==C9JCdMmL5G301Xr9$@VX)5vFiiCN|HJ0}WeII-eSTc3m3V zsSoAD-odacUV>mtxmhN@bdN4lUK%0{9&Cos5{r<#&V^!%g4r+$A0_or@%tYBq;@Wy z(}rizs17F1W@J0rT1vvqUHTm>3ZU3^W$1bvHD{sXM+xGYR8UC@jr8e9@7|_tp)%1^kd!loY<~ls2p+V<6uG$39)2^dC1Y88N1} z_Q}S;CB$KH1 zsdIS>A-slt-;om(3Uw0EeVCZ=%-iUn8~C;B=phLx>YPbCV0TK(8@Vz2NE=63;Q`8V zv>aqFm~tl@N2lL>T?F|&Ul>IH8}iJi%15COS>z>Bf>UwNM!4Q|?1yZ(^RYK`@B zMLkjfg{h({$QF5`MN%VRhqgVxMH?*}i5_&JLezW~jk$K|qjs`9ii3Sx3uibV?uTb$ z4!z2ASPfM{jQS5iqykahSY7>KM#?96MAB#MJO=p@)bkzS^DlLBF+m{KRIRNZ^$LO$ z%*`Igo?ASnCU}y1Hj?GOPVT+c85fNZ2+0N0Fs?Oj<~V#}p7|(#x)kOBypwn)KB8pT z-(0gcEmlv7ZX*W#oonC4+@J5&RkwG^4JpFZem80FN;D^#hPM)xF3X0E6yvp#63l;o!B-Q>zhy;91Vol#3?T@*I@c)67DWInI-kH8h{6JNVMhghHZx%)4^ zN$YD&$n&j{h)ZZGj5_ZA^Gw}zYE|kdV7rF#iTJsv1O&`<;;r*W&r`9oR2~t%etCp& za>G8C#U=tJ X#H8;C+HLw)<#2G1nvFJ5(fts>+6~?bh4H+Mv{kO7ISUBSJ7Lj~( z*mrC1_Tnp>OJ(>kbhoXCa@5k~%lAKM50snrbD|hvtMC3BPlZEo@fO{F?ap|PtXAdW zz%1(h;_WjB^uj)KFS4l6qen|25w!M=4hrh1L>hVZsUE8Wv%M@=*SjOP=ru3FMJjU)%*M|p&V@l8-r zW99#dViO;_$I4!G(>9V;lcX)s##0N+xI1$yu)yW2sGRYA+uOv8$;c8#3V|gacFvTq z6#0vRmgysE7dQl*AB8QYj?AoBWzHOiae3q%68EIv(d4p*nz7;#QOI%nPbT5N!-RwT z;nbOTzc;-eU;vG+l!w@*6O`^YpWsbk8#ERpT$J0-v$}d-HBG@q zX&ZbkpBI60nNfHTE2L6XctHG_`)w(|YNK`H+p~i)`W$+A6x9{F{_b4l93C6|$fw>c zBs8}?&T07FWZ|P0Z*XjDy?-7woBGhfGNcU%M60ISV|&O&mS-KRSHCS>3%?)x!}O02 zPg1{Rq;v3VjXP6xml0;M&`*gDChtKxF84-*vhA4ato+|l*$@*N)HkdJO@nZM43;jy zjlT<)GSsN`?5}!>)-l=g$9kNe4|A2{*{_uvT)>_}7BDmYZI86zd73z_R^zTv(bcS@ zwaJsb+eI45Xd-Mg4L6?PGK@K+ZE=6G;xh%BB-{8BSop`0z}k zFD3) zESs3`c(1HWkfKqa@|q%Oxs1}K5kfi9klpoqfz9^kr2C>JBbnYC84kFXhcL=aUipVj0b`hJ?g}?3GiIttIS)ZKP=ahK)Z_=~Yq&dZ<7dU80f7M@5= z$d>RsD)e`4aDl&$6*E4)aNOa#4aN=ScHq*kIP14E89|-XTB)>4@cI&k{i}8*oDU0^ z#2%V;M)*mHrtjuVgquRZ31@X5F9rEXVTT*SKoIGrw6ARQDv4|N{`mBhk8*BK0sbor zxI{|hau3eap0-{${9^bdEM+5C#L>B5cE+K$jB189Vda_ByVs@bp?=P5B z-XU+YPQN3#x%@V&jW}!|L_*S<-A;)!Av20qXoOwP_q5o8l9V*{2|zz}>GuO4f9|Y> z+n6Vs<9^~0a2XEExU=%srs(#}3UO$<(W>%Nx+}WZ@IHV_=}A5^7v^xo?fhPu#}0r4 zT)$tV1W+8<4lRH_-nx(uB4;DHP9(_M7IE@C%>$jyOohNB8T6OA_sD1nLcQ7H(`;l~ zG^0Fv-)x{8oF?Xz?BO^oPwy`vfA6f#M*bkJ($#vLeU0VkxeCrtkbCVXu0^pRq8e`s zFEHi*X4p7kJuoE~O=U53)_Pm|1e^%_AY6O3%=@ZyTgwRjh?&y|`Q@s8Ny*7M z=`7o$1Bj(K%La`*dZL&v5p)gM%T=Ge;5@AQb;$Q^>*7yEj2qrG;$vxId<0&=hrDNi8t0EB|bNE1ck4SpN0`=zyqii?rCm=X;|z|CTkf zpBinu#mCAKaS>kyQQ4(A2{oCmvj?yw6f^Q;rT+v>sBGZCE^^b@=UneSSv;5rvzdBY z;>0ynq;4XF36YrVj`r&z01yrra3~@OAd{1bxD&rD@ zhB5V%^+17G*3aMMU%!AaHi%xEaMwE2`zK5HXO3*Nnl`$lrPdJj7#nxe&lk|{1IRhW zqf2~m+`C9oLBr~I)?O%16XS=A4uAouM+e#Gw{;X=tqUVl-kUGnF?EQr^;iI9mpkEI zOn?UY0Fg;_z(V3~u4|GSXG~kS=PlR)z z(z5~S2*GEPbWswVk?|wtmyzS9Y;E#`;7@%~+?s~ygYE*TPaT+lnBSmL+)Tl&7tw;I za9P=U38vz|(#UQ0cpegki{QSo&$>)3Ca=6mU94Ghxd0Y4d>Ylyt>S++5$$Nw9+alZ zWPe5N-B!reY(9e+6#$ROH=K#y5)ZI8lk+>0IVrbTx6#@NkTa-0*LetQ4;BhY!jz8QB1<4N>@(2qO7VAkWMzQ7h{@rM1EWVMibTr|&FIxu7H;hkp?MPa`c3an4;UGt@>%BjF(<>qu=m%u07bxw@-5Di+G=XMi#kDy%*r2C9MHb)P1KDxB?)OG$ z0q-{7?>YuwKN41(ZR5|)$KOBF5i@xt_a;jx<|FY%qAK&cLv}^IJ|t_f_{O&IT|B`P zGuPtHCc=PugwV@?nDfxBlDHhOA41c|+#^040yJMCPxK&deU$~Ml5f{l6<^Coh+kp% z5U?JAAS!4+5c{sS7}ugv8ej6@Iw6#Zo(%oL(rAeLg5K@ojVXf$`}b%Pp53*{G;^{i zUKmW8#)zrf)h-H9bm-eXZ^VF;AZ@@8Urzzm{UYKu|HMHY?`BdO6+b-5WYX*5ikO|b zLQ{wCr*mVcz$7C~H7NLO5Am8B_Q{npif~rYQT!Np5&@?_Z@nKMF#WN?FCRj1-sl_Q z0^jlt-SM_K{XS#zrcrx>fLY%}XjNf9;=J?f1ToyAE$zr)>$D2B1r?wWLu1&to##hL zfH!6PALV=3$G^hCR{l|UGCW4{R^)w?N?$^%%>Cr|P0s>S#48#_wvYq~7p&yZ>N;Bw z{vsCt%eWKj{H5|Eh*QCD8{fJERx2D+QX>7j}u~ znSmB)LaYQRqc&ZA?cJ85!g0rzQgf*zXbC58Cb+wteMnaFY_*91g9@!_sMLvimJmFg z(AWGdJ7NRj9K5^T{89(8b6_8e9tr^fNJ$UDL(hPsFOZ;SBBe!b4AepI9+}0>l8sJW zPoUrHh^;*#H}8x;g$lXnGw)&MT2XJ(#mG02Js0bCpfO3x*$7`E79!{|$lz)mu!`6x z-Y#%y`k}5HzSd2S0!jaofx7`$+)E0xgLzBkc8Te43XHR0 zN-1AFoE~QqK6+wbiK79P_imd!x>HK^TfCEZ8}5-z-TArG5(ZfQk0zV?pPd%j-#abs zzjsH1(TY3D&k^2kf(G=`TGXSXrv`Sg5h+--|tW zGx)V3JhU5cB6u+9J~GBA@-NG|8eb`|?tf&XLW)1_+RZbz|ZHOZ|lJ z%L1*Kb1f6t3k@iZNwDNWHcP;eW8zH6=__?J!AvLfF}3!gX+N=L<3<e;0?jlNlA)j8GC7;^4r9Lh%!tS~`M?xJkH1AA{`2UZ+ z_wH(P@47}W5+L-@q(ee)Vo(vW(2=g7B2}t%6cCgS3BC7@^p1)kBA`+tMLKTbrKnD{QQN;HE%35Y3*Pc@wLBIp$BVXMw$T4wbma=9rLfZX7S z&cOMSSAi#ZNq52@l*(mDgyT~Woie#%?YxuJUI=@g=(Dkhg10p0TMLN;1*+$hXJNs{ z(^VV z7i7tgYY{WrVknsr=ait72_NNV-xM8nJoOr;<1KF*rkIKg7+VaGD5vbkH7j`ntP^@V zP|P{j;~PfN_b4%a6-Kx$y*QW11Ifu|^}Zzi=HIsEH+t;-IF-yy)6V}Yb;5(kPn&Pw zm2smY_bC$m>wwfkA&3~tRQ*jY)Pi_fW8zsHf&N_W^`siUoPN$Y=R%-AXMOCG{@>A` zyG{GoTqLxVv2x(L^X%`W(UVLjB`#RWI&AISsfL?hnXEy(d=ne;4T)Y zUGK3<(-t}tlg;f7EPX&so-=Lqja^w<7<~mRY}))9{caQmrtsi&B^F?bSDTHV%vGM{ z7Rr<7Cl4|U1m5w%6i!_yR&`#t+;}BTD>l_qDCq$K~K;`Q>BF^FlHjM9}2}w z?=E*tIFW5(=mH;Psiyf;?7Ys31ap2VaO~c}#0AmbFXeTt*|unh`EA4bRF%f3r=XVV zWiWHgE$!)WMags{TdS1LH^CM6)z|>L7~xUGy-eq4l!zf`f+Fu&q)#n@yF2pn{rm)E zKkG1o**l=fQK!S)yQ%K_T#A%K++3pzpDbtn*kjBt96dwDsgDVuB|4hvLcznxtw4hl zcyp893ep{d1}=;RJt(WbaE8N$EHjKfxo#|6-&DL+11~K0JQ}hIN%xClo%a0}gXOGmxwW1U zQYfXy29P6ecmJ{ zQQJMc;F~M*GmjE69!cVXtNSpr&91r(`;)!oqXWQV`wfs77V^F6d1n~+nQTH}Z3}49 zd7P>CV7e_%@;|F6P`n4e+sneM?;{L@0?12(y@++fFjDI%GkY8lBUd7s zRZI>~qtm)gEuu{51lPiG#is71LtOISkWi;xr@e}Ns{M{2x!+Ff)Z#=}4(x!;&ex6e z(|Zdu+2ftoqe!f9Xb(z|yWdcCoSFr$0OcaqbnHzg?hdNG)c-n|!$S3zys+=Pv0xZI z_Od+tfj(|Hi%1xozVIWd*J1u>CT;*JmHJvm6#x;#<$-`(h?CFt*_1H+;c<4Q2WBW0 zWHfJFQB-yzEl|3|gAMf2pzZe~IASIwJ)z%}5;zO@WJ zIkr4f;SJFLY%Pb!r71Mo#BD1J^Y^x-bqKAnLHf=|QT=PUTBj~q$YW}a<8l8}QIZ+*@gLpISc z9P+7KvB-?(l=WjZSed+)fOw+W#VTj`;-@X?f0wjF7GtqCLwyde_cEn#u7%)eLK}~7 z8@o-khM1a;nmk%Zd`Um+ZaND=mCH_w#b(1^tIQ0@idQE%*zJJO@*~1JBMQT;lYxqT z)GQhPR6BV0XU1udB9I#iw86Pz8$AdaTaU5O2Rr6t(1Wq9Xa6f4JvGacg9X0kkLEQ=I>&a5v0f%x2-6aS=-_wutiOp`x^M1Z)m zGPlJP%zbjd|3kHF_b%-I__Jt=Q%98QR`e9(d1V8X!h0rFQh<-?-5d1NXxT$21{rNd zXVj8p+;q2FE{sF-P9q04CqhemU^0HH()ORZfU@&Ni$PL1?N#moe1>fl$@6Eznl5Ka zeI9R%K83uxvl6_T!p?wLevDRoXV)%kZIVg*(ORI})R7b_$-Ck#%IrT{_vqp06wZ1N z<6|M@z3?}$a?iLe{#;gS1D3!UK=t<}OwA3aU-rhxCA<#+3=YaF!btmYwlJbQ#-N zQA_oBm|nJz8be`shQ!X!+cJh34LJY>TH8EA3luyJ6EV&>B8<0C!IRP7EL7%Z*JWOn%xW1C0=Wofz56ZaC_rw_;DOslXoATY5I9lU_z zJzG+}&c~)!BuQ_1OEAI65>iXU7E%3-I_Z{0u;~}h0bl*- z>*LI=L29LWK^!0ycM!r>GS147U{?34U#r_8SKKI_nC@M=5(Ylj!R;pXofl5=ef+QC zMF@!crsqMRXK3{&qQLGkcK`ewj7#V7k(ieu>YLmW8Y~2rP}Sd)TM}$^K)oTzxme)C z%hAcsBx)~wcuC@Zk)-Lu^k0bd(2}ziqEZ*0#n1rT3%Axrn@q8^yKI^p^9{&Y-lL9} zZ{8p4N)*;=i!yM%^fFcYB4@Wjy zfU%!p2LDeP=8ufw=>EQ47&rA^#_-G(>)2M0?JB|-9tSR%y+O$b^6$F*7rIM#Amd!N z%f0sCM-X7>_Ng@{hI;D7mE^6(MItbSdgvtDCL4B07kL$CC#H?g(L4 z*5eT;adRJ!WIv>58VCtbqZ0RzgAl{j0Qw;0@BkPUr=VNPVVirvj?Efio8!AaUqm(> zJdx#o-t&TtpozMHeWz9G3*YqNlvyjRhWtnK8A5jYl;2%uLz^EsdvXR0xWC-sb8(j; zo_229UJlKby}Q{3YusiYM#wOVl;-#E(9J@VaPlLtXi0mBLhcUh<+BO6!=;zSS;HwWnOjzK~SBk?bVratr?HFll{5@e()Y5Zf1c}fRPd<+hsw>Kb1aEc-oA( z2xtnOaOb?OPeS}FK3oFA{vIEsR`#mKBYAt}3M~m2i)XI90-=i4*|_&qm7uR_h)+M2 z;`!M7hdLAUG!ite*Z*m`q8LCboVMd}xBv|7_PP*u=B1xBjXy8nSTY%H@)n#f($IG@ z3k#-%#iXSzTeLbKVE6mmv#WoN4_og|q4$j{2GU%yFM>aN1?^_1`)sym?ot(Z4fJkx zrurUS$F9UGYii|#t)fc>l%C+w2mx9|+<1l6l*T?jF zzgLq$t{8H|VqlNQC&IDR#uS;fWB!a%uxB>JBW-6~Ir>;eo^echo%jU1BU$i(m|JOS zzNoAl<8QznkN^FD&db?u?0I=__r90!5be?;xK0Cd=$?Y((1Y6EW>2e}Ai!NeFSC!I zT3%J;H2JQRjk))@za6v5_560U6xQQIEpH<{l>ak9A&N0U9$|CL{rJUgZwi%tA`FTDrL4;I=u=i;qggij&FZ z&Klw5X9}*L>x*EwsvB#j4p_8nhv}Xq6q7+-pn?#N(}(#;Bv5O3+e)?f8u+d&`-p=#ljBH;<&~GMU-h`pWu5kAvffxQnP{WB zuTCn*@_)*az04o=J~-a?GS1s1`qtAc#galM*Q|o{yCN?gqy@a=GM38v<`aDtWd+ZN zlyA@E!T>i3-8iK$$0oZJ*Q2qi^2qHsISRW5)?F|^e4!WU2ADEs*-WnW z`1%EJ#m7YaNmfbyg`$UMH}Ex{<$5HiGTU543f-?5OlScl|A(TE@N^(yaXWPGQ~LcJ z&1F(-Zvu>}+?xQa`}E085eX1m5k!+G)ova1#y_3TSVM-$b$pJdBhXWAHRpL^x(>$SW% zAo%X1)DNMl!BlHx>VyFkfs92c@vSB~U&NSM11%a~31Wn1g$;6+S zl`>{}f(_i}rgSg67-YHKXC=!^8!0Lt@J%ce&dIQ~fXw(d-OwHMt-KG>vM|x-%V3>hZ=?sSb+@$;-sYtYwmPtAH!AevaGC z505_C#-uisO73jfE*S0FqnR5ucv!+?6DH7IRtaxK)YpYM6;rpI=b(VCNS4RAua|KC z#90-Sl7s+z{fyT;n&T0ad&4C6?Xk#vg@*2t><=@R11LRtwp8+-6IfQd8!^ui`4_N} z?1J^$*H~YxpSr_(sBSF8y=jR1YWcNom$aDvwTzs{_kU(ffdiEQ7r=ryu(t1&{SVDJ zBmh^xu@MJXih%IrKGd!BorwZ1QBSI_@~TGiAJ66!I#TpJigK5Bdn$fX9dO$wt6meF zIt$2}np*pmKpwrX*Kym`<)LE-Q7uF1q z`{}D4x5!Pyn05qKwVs!nk{zsfH~(n~4Gwq!Hv#RxqMyz9KEfJ`_3h3jMr*iTg7(#SFS;<+;Wru^dQ zOKJ1A#KFWs11JJF>9#!cnw@X-$++~|i=B-;q4+K6siu*Ft4P|l^2MA@0EG$>#zFuH z)1=Dv96HM`nEST5NrLR*^0N+`!>M}Js6g?5eWVOATneLCTZ65ONYrjs={03A!+x!HeGVI$r$U?QG z*1$o1WKE;4QqC-lfqhZR1@o*mg1IS(ip;^RQ-GWI*~h9Xo32d$kyBDz;_sBur63{B z!Re>Zzy`IXc2K!bgAZKYlg}FdEl@eqi4%b8RJC8`J^7`<7KbyTzUgrHNxI_!?S6in zXVc$~wjdgD7zA=DFZSFn6sa}e6!s0kbz7p9r!<;$}zkRzD!I3hWXeJwG)eo6g-ibj@&0`hBcH!KG-ghaU&3 zuZ?C?qs8ZjEs%(9F^>H&+obCAcjpib|9}|ncMgr*Gfh;CdA^NnHKvYj??dza%3c?b zhHREuzp~pRN(c)VkSQIHc+yrwD-JYwjDOQ)I3!_?uNl(%Vq`2J6LCvcwqtT=fPo4) z!bWbIpa$aM_0v`;Gfh@A3Fi+66A`J>qdbJ;5|;+RsB~*m*5UKul}U>Xpi&r?ZvWQ; z0JC{vY_+`M0C2+e`VA)XabI@jL9T0MzEZ-$x9rny-T3wNfd`-l1n}=4_{T_^oa3^ya** zwdw+qk^P>Ug#Y_hBkmZ@G{XO4M-D$Z_Sr-}P0Dp%N6Md^Bu4}Hi9=nYhzJZ&i3>fi z7H;>ZqBPLO6Y9`=|JW|gRMOgN z6MR^?`DmR!iIrkh7tO6><(Z#D$N+Z7Qga@F4@4FQNhviB5GXDkzk;R;8jlcvw!sv% zPuQNaVc|;)n&1^jRqMa^mc?nR^~65qQ|s}TR~F{;;RW$pz)zeM%#>RPFi^l) zq_Zc+AZ4I8WhdTlI^(Kive~cDa`2}o#GBTi&cR>jJ&io@?X)6*nR}enV00!WvdMOQ7%RAi&m;Qt!U$@a? zvjq0=yxa+Uf#FmWyX21HJ!`Ns4U0MEv1ETfV8QNTmRo&x^m?-289ycur|F)8 zi^+d_!YBU)wEV%j$gNRoT$<-qYBf8mC-GvE8zpi;i02$QMKxb@oyAhie{ZYoFc;o? zjyvvi?)QhukP2IxbW!U}#_eU7CqXAL@#I8cRFW1sSyYnVg%f5e2I~;MRK4ajez>6_ zxwZch5Z#p+?1>XSk8r=`<3a6P;*}SB~<*jy)xeH&5)+uQ`fpbNTl+N3?~9Z zsKn)@_Mn1$LQB}YXVtw%TYJw~!0mIy&0Dk;BY$iib5p~=@7*g- zi~i{Qc`nld5vLHk99yNc=%y42vnsFVxn?YmQ$pzLLyYOJ%f!oTM$U_)P2Lbvn#*;@ z<>)Jrhlybz4R*{|KH~{x~G4{W2%0! zAIA+^?e3ea!9#jgXrrVh0vGCMf70}&Mh6UmtWr_2YxweGfAja(m*Urb*w%+$kVbKc zH)9@tsysu~A-Rl3+1+ZI64}Y%xaK_iqCjJm%=b5^Y53oMoiEVr@$>zmvE{~1A@~~6rK1@DP0`64p@Fy zN|#zTmH%Z8lFxP-(-g@lxKA+?0UU~8^z13XM>4?}j`XS`ja0Y_&VnGK5KW30VR-bp z_*D+=?(mbf#a`_=xx_7E&sYk7Qcbpl^0%x1vQnTa%l{V1?y!aJzvv1@HvtA8KxhD| z9?a6&+y0Ovbw$0-T}GN9h@&{8*{^P&MlHR;b!YJzR!y}X%=D1%-%$&;{q2j6NftLUqtLkVomH48 za$~gx2BBcVz}&XS+F><->aZWMz0w6`5vVf@kv-ipjBL{1fCTUvhGn_dh?nw&%PMtb z$#Zap#RUn1FfGvIBeNWCV2ai}sGmpeAR^g3Z34Pp+L|efs9y(gi&+7I%{<|=GIJrwG2?Wr6sxo{m~{kEN^Ge$754Y^66LG>cr}i z;O?(Yyo#+jd|f1XZbHRC*07y^+jT)y{~;-IN7#E^>rJVGFfRV*jGO>?bvXt&d|SuXjVXfsP6MjcY&Z24c>TX!DXhST$fM%0|k)3 zgChn{2p*2;OLc;Yd1^TVnuu(2LmmnDn+LRd+U6qe4ls=S`)!YpN04k0p~UztRPkBW zHW8pepI?#{BsfQghAOP5z}f zIi^jWe{V4Ga&xx8T3ZBzxc6=|fYp&BtS866QJbPDZw$g$(0`zg&WzIj=$3gVWFSm0 zzh?Tj=#9uj#F}@Wkc!^*sH1WCc)#+%SBIRAk=lbnubA|&K6t0=FyJ`yb0bI&2m&qv zhkq{M{dZ~pL=7Y?(A{x=;;AWYZx_6`s;pn6OW6(9r#}!x zWu5jtL_Dy<)sv(q42qnsz(50zm-imxQUZfrK>asa;MI=M*NhWlCUj&*NFUC`7z!(vU^8(o29)+|FPw*9a+YZAmm zL#!GuMDdV`);=pOU(vcvbv)82>kO}=+;CVy^vUOKS7L$cP;Nq}(NayL?bmH&f^Yl& zZD7h&jn8`;xO;@YUVooH^W0?Geap^39zy9ypRblEM$qhtx9l_(x0@g8bmc`wVbw`L zsa2PDBe9MRPm+@gUYBU|yVR=$%Yg-`Q>3E*uEe_O82#P1ckIo+r5Ydg;CrdYv?m5Y zfvK5th6g?1fkv+UkqvEy*C$A^61R?dRQS(;H66ocyzg|{_1t;Uc=XeWo3s5_oW$}J zf}0mV)HQ5@X&-MiOrS#{rY|p>y8tZ($8*kQaXVw)KRDWc?;olA{C%p?p=(XQmuh6# z=m)%_wtQvJ9`cYIj*e612|yErG{sST0X*X3%|eFOjBE$4L`5{{@anLScvx7RTsHXwM_J(B&wOZ#VtbkdDT;q2S< z$_g0~BxFF>_)YuUWhn`FkYPMtsWEQGwY2UIBqTajcGt)b>5q>{Ec*6yrQP$7sj)M= z_j5ig0iXbLg020jbH|PPonVh(WDGbqsUv#=lP>ioRYhmy!U2B&xvFACIkd|-BiZX> zt-uq3q+4o9@YY=UNcvCYfZe$w*>sSD`T>!5WoqU4?B_jiYl88fk@38(663yo=FaND z4{^KW&!fLGO%6S}DLb-*?YcG+?FP;sk1d{!>x?3oXx?kwDKu z=T)nwG-$wx=E*Ng4|%=qd%F`#=0a-xwm0S3(K*k|K7-=PZEIU)W&DV-;M_@A-%UYL znD6o{vB7FeUReF{aRjCrxv_NxD(5iovUFIN#JFxb_EeSchA@>R_QNAz<_yoHqis;x z0p45p?kxxJG9c2H*^~j0Z8K%t%KHJi;hke#iU>xQu^WrjX>maiKbt%y#~ZU_ss|iP zd&HWORP)0qe#Te{AnBVpU-SN_hNDbLp#uBB6^ zn=j!$A0%64ZA@{vR1PMK4mjWuFZ4E_B3WeZ{F-@>cM~Rp`m1ZMUZkT*{p+tYF)&l4 zqT$Ue

    4*W%t(JG_4%$$)rS_vlPb5qn46i?ra4t`(D`jvhFks~cy^UI5!-zlx9% zU@GK~va?qjfc(}8Y74_2!Kr(EWBILbYemsPlv$h^_c~MzOXm+`a#3Ibal*E9!lZu= zgl*_++k?{A1yoDN054RhSfO^o{16f=AECA7q_T@3KYius2+nn5`+;9lp4;R&DsE?93Bg2gJiq zy$@atGF>}~?XN%+4RM8is9l+$xDJ8F)Ia(h@;pbw-Nff#iIDvaQ*cvvPCQKQoM|^h zz29^<3i|q~*D&A!%?#{3!+LwXY}xx>(*P8pZ4nM41cG5mdYYLqNR07Q(EO)3DwHmDLf-bl#mz7gV&U!nTXaZlAZ!LjucIzBh5^ z;B48A!*Pi4MT1ejKn25DW110cC{z|@-wuPk7amPD#W}0w$!y#&^|P)&Y97LVNQrG8 zS|g`2N?A`Jm?GMmzeA27?pUB~CXM`9Mbs#N2Olu$G5A>dH8Qs3SDHz2;LJA}h`q`h zj{52NhMY-aZ`*-N&|ZlkMij~}jIXROOdz89F1j7`aI(-y78k=kabFk~%iy>&f7Rn# z)s_C}!Ef6!(OwSwb=W0av%PS~w8LL>*Ps;ghRfDD1%vjce#0h(#+ZG;C&{J8-b;?n zb@u5xYm^w>#0KgudOwI%xmlaLlbAU==*WQ%_pgmlR2Ytqy`uaYL5gAh@3tL&vZDpd zrq6{h2P9|Il5!Ojptl(Eww`=C?x*a58&zxO7)9bPniiTG?L1-{F~G;3P~!!=V;DIk z*%!q?_kopAG33UIN+2ZSAsmps(a#-}gHJ-kcxuk{hI8N!j|X0hc6){3OxRGgGbM}+ z2{L8De)_XrA!d6U?X&~Wg5lZxlLXI~VJW6LQd@P6nIKr2?4uEMWq|6km zrc|RL7Wgl}l-hL*x^yhg6YuVsa;(isYDY`Y2Q9fh|FXGM`s=8z&CYf2;Xu;xJpedL z5U?!@sKw1#U|v4%1ZZush74DYe|xfW+S8+)h1gR*hs(p`n7Wj@G0x zNDf~<@xNSC!9Uz1m`d%{0dee=f3W^49bQ*DS3D)-TS7u)l=SWB9!0u%Z<4Y z#~l{hb6@~yITQihxYP@i*^bMFkz~75qJVt&I3LrUfZPDtjTMNG=e1xiFWI5794H&k z0UjYSEExfN1q-rWB|?HW4LI-_uKZeASjWR${&*g+aW?^g|ln( z$f;!>hbVK>R{B3IbAVcY$aVWveH65`aZY;s>B8*I0z^f9!<^~juM&~6w6mz~&|A{G zw5o8&UA|Iy0L2Ke$u$c4FJ>Ne=lWsR;|3mMA@}V{*?lnw9gfvVxq)ukoeK z8p_kTm5fI`+*zTFA|J7)j5!q2-%ltN(aO>;x`lDTHp()-XwO0W>G%M@deU)uSP{n_`j%modJZg3dG&t8Glac)U(;b={Gsom5g|7-b{(C zS#HnuF`Rq0@1V7s$37q&`7%mbgP)zS$+E7y&R&RoysRLc>?dXL^K^FK2GV}aYP9~; zvphT|A3(fo{JoI4@J=;hQZ8n%khoypnLuhpXcAn(7ZMHxUH8c*Z$5v*Lg>^hqb0}T zC&5*C2W2RuwM7>`O~}#yI~!;>a{O)shSZpEHZXys*t3D7P$8Kf;QoL&H6yze8YsAO zzc%flwa?O#fsV*@^cu&yZsH5-lZm8XX{s$=-+4M>*o;!UCUX0V`={;2kvBt+11W9# zA4vh2mVYb$_ZJb`IFLHb^y24+AhLarzijwV5BH|V#AfP)2ACkL$Pp^_$#>oz$0ho%Wj)GF(!_BJ!$6FCArFLS>aP&_QeC!o@R%>ga*>wv&$9(`nh{xOw&nRDYEso z=N0^wh2o%~9qrOw7VfYiFXYf_3oH?|+zxB(wkvVR34IBrmfctmgaO<1f`KXOw`U7U z93LB(JR;J-hL%Qbt*H0}rf;=#CZ0Pv&2n79b2?rvA0WMqm>uE^sdZ{G#(< z=bSG`V4)h?Sjf>x6}|HO#>VPY_bn;P$5+XZmb^+o=8fZEqo0!XMtq_(K1om5zQ{b1 z#YRsH+|ha9MOFN>eZ>g9`b&&SMTi5)uOr0ksm^9k4^y)%AG~GLOn`QyO$DRal3%?;Q@C$wH~xes;)-OOa~OvY3JSn16pnW3bPk1U2<3bNfSmTD+%+s?yFAc zYlh%rw)dALjjWe9tMy;gM6n1pqpKb6EQv5)H$xLKe5fsFI~F0i@xG)Z5`6sc{k8v} zX4Lxcl?{Xoll-YN@U8`&hmgVjfungr;Xr~p&lb@{UNHK`thqSXVCwr&I7Jf1M<}Jv zG>m(l`AQ(cU)@DVO8?js=JUj7-h3gz)zTa-%3d2SHLXFtL@kXA>99nN#&?r(m)+)m zCQzEZ7C!B&nGgh1!|Fe}?2Y9M=GR~0?i27s4p|oYx^2@5W}bcdXaq{ZRydpSvg}b# z-m72JUI`%Kr^j}}RQ&0&+k&yA-yd6kgfLp@*uObn;W_}KPLt3Fneh^ii_e`FT4qWF z%Rwb7%Lr-jF?M3Y_4u0-ZN06_CnfON6C(r2qeMa|(8%yU-g%t4F9D}JZ}WdRAOw^+ z{Hp+c>JUEvJ(}EobGPBi@l_N39?RV_lXAA3m)}Uip&%D)&K070GJrn^yEUZYkO(g!g30t!|DuUrXWd z$;|z^=641H0scL9tf*fUR!f(e_g>dh$a%aK$#T>%YD4r&yFblwk?zv-d}k64oE51~ zjW2p0i~>kNw4aL<e{V@8_#PeLl^^P2o2OFJiREBA-S+a8hO zLjv;BNdM|S3N4i{UM|@;$yQe1mQ&jk6Z4ne=3JRwzv(mr}$efxbn zKXmdm8a4I4!c6o?U3y|?MfTqf5D;yJKL!ZqKL!Zki~9ov8BRQo1QA(3St@1h`G_J1 zDso0~TkZ=c3&`CIZ+F~dYpxBrm{;_2vL7EBtn|mOe`1n!>Tex=CK+=YZ+Tw zNnOYASmTiuDYJi*u(x8`$s!JjpMZ;ND$H4;E~sx>CSP2I=qIBrH(@CzTg<$);^Qpv zNL!wt8^rr>=llPHyl!QHG=iGL4E6eczmX=>(MFr}gOhEV7&n?o16T0Nf~SpN?@HZx z`_5pms+K(PuJVbCpSzRr;t)FU=%gkG+G*@RLr3EO{B~Y9FYk3;`~o>kF`nyG?{a-n z06m#FokAKJoLRsECeCqztUH8*v?b4n9MqE+3eOH34TVkjjR)ZKI$l@)mxLXh>HRoy z#U2IF76t>Bwxb#q4K`3UjpwzmHB=M~Ug9!j!8B=;y~fEx=O45-NgBuac&z@*{QLTf>GsM@=&tOR*kmA5mP$z5kceIt z?&orBYLuGzB*au6xqX2fNeuEBMKlqnZAdDqT$n5d8;T&orPAuI8%X6M;UYn1-t4UW zJ+K|Cj(i`x!=VM##Di56K8QU%I*e@|La|XWl^h{QUKGsX5;jUHXtvo*X+r+4(EF@a zI)4&`lnk9>^cWM8ij^+oPwyBAWz=@fu$%F@=~sO|dt+eoO_ub^z__gBz>OJIX_C4U zY5oCMp*`O&45c6@15$NGw8b;_q}VU;+3riRmlofizkRoc0`K5xK`$Zk$~Pi+vT6oT zI^3&ul+wP3`oadUXr5;lrj4_E=XvtxPH&A|6wegdBoz2=J4;GsKWry`%D)T3rW4{_ zL1g{3R`NkZ+vEf5*AAne&${^}=Bo<{0t~rC1AVrqUS)MnN;*EMP1)1TB^hmcnqc)j z>EP;QPl4ghlN(FXUI7n6cz|8;kP9KS^hy?Ld_)-Y2#~>SJ58so;VGK-w@b-7w?xVxf}C%==`cMM zt*9diAI6NSeHTYnka|<}LRqvtl~P{xsyN`oW?xD9#y(blG$I3|(d;IA=+RI@>Agx1 zQsUnY%6N)n&EY6kptA~-h6imb) z4IR5a>>BnKQ?jpDj69iH=Q?-F!*CoO-Evc{z8nuu!bjY>{#|nbDe@WFbfC}FsDr%2 z)Sq-@hA99n!U#uo9+T<~BwB~`!$8txMuC0=qA*UTQd2ds?*_Lb`Lp9$a)RnYv>_}Y z+AASo+-js4vVSF%8rX1-_eE`fZbuL<%%Log&y1V$ilvV`vaEgSf+=&AO~`4i9-oX? zd0F}{@!^_8Z=@vSW=nGUPBlF7p)fg`(y8A086|S)dRN@JD)PjH7t%I_wvbsf<&ih4 z2V8m@a&2>+(t=AJIH?(D3lU0eb_!;G5}C5;_R|6Ecs#UI%nJDuE8Sph$gYf}F>I!` zCbA?b2zO8nCm5uVkO5Cb`r?&AWzR2(*Urz4;|*-VusX;|4TvJ6h+0JU@ABS^$DoswYUC+?Eok(cH+DTkm!s z3kISHkS8?B7_mYp)DyujPzqBQr>LWg0c`^42<`+ASd{F>N+47ex55qhiMv_&Df5IE zh8*R=8yPi?Qg5;($7YKvZAu+OP>h;tLa+&LMKO|cO7(5bDaYv$R@cVsQ_0FM!3_3I zb*s%eh`MD;bxZn7^r1n96Ppv9ad21{8T6%M?U zU1!|tN0ir%-M(W*qzyvdt@0A-kq(?zf8%OI0+NHH_!L-!ZeK{(6@T`#?DI7RwC0}1 zOeB~1kekER+K~(7hiXS-j1JYJbVLu;HHy^5{X(t-0%Y+ivgdbp;^URMzqh}yk2OU$ z9n$^QnnFap6N-cHrCv9rTqX&+@?IQsuLuO`u3W7fOMTMuCj%+Y%sFA2aOW8;CzM()QRQqPsx^Z9!m7`eE z>2;>bHtNOelHNKiEA7us_m^5>>aVuDW{FV8n-bmP+1|>o=|;~wEhb`FI6WZ1QypJ z(GnqHS9DdF1SVmUwB(lhC*3;wDEBrUkU?Zu5)4+pz{~{dX|2;JDO#;!wn^5T$sHP( zK?WMpNw3YWR;d;1SsCkG)CznwS4`=cmOgVvR;^4g&8?9f?Px)iKn7Qnx;>bFmXo|g zK9%CRi8$c=TsR~*L_Iz)|EJVze*RA7v>0i3*@Jv{KJhvbShajS`}y{UcUesk08DU$RL3!+d`T^?;RWfE4uoA!GA zSs>>znfl|j=Bn|q!*1KEJt^w3^hlx&B8*mB=sLw44xtByEs-FrK}%1%yzVUEvcl>4 znb6p*8se=v;W)mQ60K`+tGO@jIWQyc1itWx3$39?kYzjX*A}#sD2LPHzmgv8FSg;t zxvb?cQ8L>@W$LG)TRk#Tps+_;M?+7wH?gN7&=Y=^3-l&T>^E18!5R42|Fnb9 zaw4tpeN><{?siA8g7}P`-A@F$L)hpr8NntxwD(WuMGzIb);q3c3Shga>=D3$j=y50C4 zdA+lm@yIKqxViqhk9ow;-u~Toy1Nl-l6p4f;)ZlhivJLQT^tesHZmNF!`t9S#B`Ri zS1j;|3In~EW+E&Wjhhj3;gOeXq?bwDA5Esp6IULlYqUXBa3vgl@Y!I5tX%)f@UQ&NwU&IFYz4Ldgy*a057h4Z)jjGEMSfF*epu&T${DFKyJUD zN>#05E}=HXp#=aNYb-(4X}KBZw_RnGHz&@ZrXYM%cT=5@H5yBEY7ql3zFIv7c_B=bNtN~(PeBZF$^w;6h|Mdvk>+JYfww@8N{*L(`A zc3K$OSr+|=Z2o^2rg1j|G(O$hC4n``KNXu&VbbCwVS*b%C{p6jT8JIHpd83=+k58N znnZtqTo(TuEzp(~We8^k$zC6G+;f|pxI)g6oE3q}S`XVr3=fc)fCu*?|+MR#L|inz)q~7=QFk2&-x1SYeD7TiGz;UVZK~ z<)Q6W3(wlR>&=t`^_Y-k;}~oxt3x*C8Pd7&@Jyz4y@8deBVWd>d`QI1bb-Qjicxfu zuKipQw1KR^?em@x_ArII)P7A3`<4E(&ecF&J%hZy zz4h3I(zhiCcZup|5{tD<^&fa;l4F3tzzYu}IB!)FHJcK*B>eM1MR=^-h=ZBSh^Ypf z`0LY8j|_Y7iks*xY<{XhmR?LeG_3p3zSr5fBxFZ-{pw(ytK#MIC$&q#^~nE)ZTv{# z+?R2~$C7)H!E{o3`lMU@$qwH98>!c5?(wr*lxD4GT(&pxFbeXJ_vi}>^|WOU3lg*a z5K2YL)C&W);c@n$>>0&Kl6c$f$j`%a1zz|7%mW-84;%T`3h9>OjolBxl|q-I#)tXh zVzm^)`2Y?Jl_{9E!T@jbc;4jp z?V5Ix+jkn-Zf6^!fQt_0i}k&G7%E2fueS z_{S|Hp^dEhBN%(gcx81a;+k>JNZqFCQZad8g3W1}6{xJN z&o+xK<$bR;$zFzoyT^B8Wo0Ls3u9|qvnu-pisB<0#mwQ2L&pU|yGva)5%@mMs~(^-UBCrJ@y40h%RY38DKu1W}=r z+h^hklK?7-I;>VfXCm9urF}_tdZGxMN>3&;NkWqx&HeDX|Zv6L1 zzC}*fk3ibi<#Cqntf+*rOLn9y2@F<_L2I1X_u_Xzc+K`oBqB3?IUrVG2m`k}#B>0YMQUOr{5Y*QREPElLJm_<8PO zE|i1Lcx!FYo0=-h@tV=z9XWB1*<-B11C3<1|MgnOkI@ZWj(K<~_0EDfo_ zXoLcjUEvKS3gV7<&6n>d6F-hk<|OST6Eg&#Zk)_XJ|SE)ncy`5LZ#I>5Ad3I0IF7| z$kR0f!G`PDX+5F+Adg~ArWzkHyY|qxbO` zEs!E+^Wyt%7|kbi7!mJvmA;)hGz$Fa_6OY8nfq>I{a3Ef<3bs*g98tcMA&ksx|I*D zwOp|5%Xgnny!bz0K!0~Ix-YJt@7@u;H+uFMb$#}}&iMEJ;Ffsn3k zae_8+?fPcREm@9q*m>iv<-ja!99ARK-K8upZKjN`2PQeR8OZtd5#|}CW8E?&6}bVj z_KSYJtU!)nc{J%KxOYTXho;S)_na>yPsd@KS?@evCcX9bKFfR=s}q@}7P2y*-|P(0 zYQ}=!fd(n#|Fo3i^n380h03)>Fnl|F^(P#d99SisN&Y{~y=7RG{nkEwF+;Z^Qj$ta z3J39R{Cn+Mjyv z&bYC_OuL`4HdLtA-gZoL7n2XC3s+&7;zcLYPUDurLn{WnBpKa3?)eGm7`UEj{P zII?G9m##2+!GIY3fCCJ&u)tQ#Sv^lGP!1~EBo`V40+~SuUfju@Op@?{{aK&&31NsP zE{4x<%j@h_3`3<@igFEQs6Wm}58T6j=_fk*sMlA0b}1#kZK6IeL1*ukq1WBEN7_lB zXoK_JUY*2@Sqrq(FC<%fU4x`lN;OO-dIY%kB`%L#iQ=+(EzRl%4Rcv`6$AR%G`BI5 zO>-ha*Q4_07X*uP09=kK5-${}%t=4G0x8TH?PJRf4Pjj+sT93iOqxHZD!5uErt(Tx zvnhMGhorbIxzNKp7QM#uc+0%-;^&Zfnr}0k$C4%P@CL@R8tY5l79}$$7@ulb8{S*y zj!J*^p?Rip?hiJ`ivozh2^BBkPV4?*;;FN5x*3A~?Q)z(eTuJo zj;_OpDk1gOTGy|xjp!Rf@?)zzZpJOaE3E)5; z_IHt?NC5MNBkGLH{^c8c|5uZak8HU5c_o%py_0;PL1;vr3#sRo#)lWfa=I>ML~NZ@eTWd- z>zs41JLvonais63{j2Bkp`v@j56n-}+6>vl{sT1&L^~XMYS@S38#R-Kf~N%ePBH(! zc=9PV3#KBuaLsml4xWOhW*VEi;B;<%ViYwqYfrsN!W*@MdX*F+b(RzV?c$*dW)c{! z!r(lYw^+|NYUXb+g?<(8z6Pg%pl0+r>qk-c^G(u#QHy25_6>j&>xzbH(&I4*QaJ~a zFzBE&xQ}>c@@bTNZLHn`9(Rq#*l_(*a zg&50P{)c!Sj9v|gG3Qljnf5b+*t}dI3;^&U8^(zZ!VN_}>+ItJVH@guUf+Aj$a~|v z4Cs=S@D-!ECyG}!)T2oxR*YO)Rm_-SYPyGzcvXY0yaX6PmY>kd(7yoVxKp&4>?+r_ zm}2|-`!10C){|Wmb`2->j8>xgCQMLh{xwLJ9|*SW^m8#g2!-~D05;S!yP-!)Wh)r~ zZ)`G-3kNp(F9HZ@xh-YwX5pa}D!c6`&Grl$EN z`yVy^MUsaeSMdSiNZ&lDn1w^g2sCSgYz<#UoZnEoVAh*0;s-mMr+8Rt+#belNPNuv zy$Ycy`v4y63xY!-h& zQJA(o`mch&t)-iTyVr}iJja@bbf8NhHM0J)=rP?=w4c2^<<=jjv)^WjTcH4UKiUO7 zuywi}F1`cHIP6)<*A~BYCkk6NlKg=3GkiyoFi=$y)jgCL<1wB=JsvX?@l_7Be!QW5W zqMz#zpjVVLC@TBgedWZhA$TMM#__Byx?+VdQG9y&Czu1h*Y0sXJdqKtu)CMn5qZ-W zKm1^867CLP*SOGKvmybO25vjbs}{n&Hu({rDMs2 zligATt*ySb@bY))usBld^_PI;xQf%b4Tl@fm2!4s-fijWk%BMwwrG8@hR5ofYNBq6R@-`*`w=CwL0KRW1foZ_VZsVA&((bqRnMa z-c!$h0lm794)&NIweuW$wVbrH2Tl!{(zd=2`TCwBQ*I#mi-PwF(Sfp7eZ!0%m38cs zOrNvpJ=r{|wmvkO_k!;EvKk%uP?5PcC_|wQTRsGojeS|cM8^9`G@QwsK7PE{zqCg{ zk8OQ@xr7sQZiyb-c0`C>aLJUtb;a*ev=XsOhPc#c({_?8J`~I`(Blux(ISX5qw(0Z zrs%=zd-HKp(ZPN6PB%x4y>Gb*^{MLWKFgsUWXxkuy6B0Q@0i_HkeP5dePuDY-r*W+ z4D;gf`E;CjW}%Kp&&-yyQ=L{% zekunWlV#5J9wFm$HEfC;m>j%jmAM5dLM=fTEMIFb0|ss2eR*_R;DnpM<=DhfV>uPm zo@X7b`=jJ+)ED3E>4N)k633Z{_SMKf zW@{AC9(@)t9jF?a)Rgqc$C}K+L;CcnmTb9qay@?gwrOp%p&P?CZL+CxUfm4MwDi(iD8^c@k;)(PWMVd=3knN>tCBo<;(LHXt9<8g>w~X zj*}Oa($}hc?iVWk0`{M*G3>vG>Cn4jqB&hEFzzPs z`lHM3`ghWCC_7QPX7w#L=M#ja8Oj=?aauxXZ(QL-MIYpoQ%qN9j@3feNe1}Vqph*z zBj-OdcK&ET!n0#Y`!6+n^f=G zu@|1Ozcv$Nc=1t5-8EnaoN#N9Fq7Y|!8=ZDio1DR#b@Xh6B5b1NpU#xUS`VvL4e5D zotdsSOwX~2E8V-SPB^s-m=>Qtah=q0%XUl_juEi(Llg`W>yS*f|_Jzt&KJ2`~V4BfQaM(4?JcB@} zHf8Sp`4J2l7x3kvJs2Ou)kqGDVO*%UOydR!N7pB z#?Q}|^GiO*t<|lSo3}Py-e?Lq-F!K0Q6hre?5Qa88l}FFSlRYSTRWkgnzuR8ykO1t zXSJ6yAPYzVj=#@{F~EM7QGDMLfZ#SCXCr0&cIE280v#GD;?Etly&@3eQOunWs$>BY zSPWVcuT?FCl8L$cyjecWrkq8C1P*q3mXat;qBM{9d*0EP2w5i{K6}7%I}+r1LiyEq zMZSV;1LI(~183xhJhVvaE9p!(#HgnnQz0Qssm`@*ueNY#y|vnx=pB~cT;+YB4%pRlA~jXZ_zFKT&o-VlJA?0 zzhcv*)@%g4up_ujl)C?EuJ0Ciq_r*o4}#$wF$fQJq`xa_bg=y|2}ZfIuJ^?e0Ud`f z*Zt7PlY+x(e&e;TM}2UA20jL|urO05ht}M}S!a&&Wt~v=D%?3IJ&f$^@Ln^Vi@dD5 zHNysTjOR&zg`huCy+1)vZI>Sq^yyqHbl=+F!(W@AD zWPL-0QB)6qS_hntK39$tds@ibo(J3JW;tC4b$8i$C#x=V2ru-E>>(6oSG+ ziKA$snoP6a=fP^m< zhtcfh#cNrKipiGt zShqR=U> zcQh!NlQ^#7uq5kNt)QviVO)&;-)@v;DPGPFySloZ6D7^Dl6%_F{}*8)Tf1mFS2aDU zFe!SLdd0c!Y;mSY-^fLFhG0?!%hFRtaq3lELbUGgQ_V*;i%+$I#rvh=V<2(XIx^73 z&kfNL3f8%>uoUZOi%wrn&%#jV)D(?xv5_l{Ap6DO(5qsTe&1r>?tG61wJdU86WJBg2=Qn)@!EYcx0;GywSPdOHe7x41M)PJM zCX#~|R)%gYraMf2a{fqLENmbXNUQI7S*8I5_l_@X*3 z2ZcHzqgYA0A=4{sg&yR}uFt^8C8d6>UMO zf5-9V{SGR!f6IqC;-TNppy#$j7`PBuPB}i-16>a^Opd%lFgf1!*yvq|$iJmZ(K=Fi zuKMV1z2#iDV8i`O+E^jyv_2#GV;3w7+I6wtQUm@zNPY$rUw7f8c8Ynaj zXDC?3ogKbL@J)9(-w+eHaQ%xhT$-42xHMT`2Y}jw4?GtAlOimzCN1>D_rNTv7?>3& zPo{ZpKr^QrBt{&&;)auij_1|__vxS6I3pE_Vb!jmNRS5yUaU7=zeqwoaqik?x`b`a zWE}I;H+tnvpKN#Vj7;p)E6k7RL9Bt{D}DAjpXb3kXNU#$&B?3I^-Fo-ez_GO)D=7C zM@#L;7(|dn-pf3%W)Luq{|n?#n!DB%Ta=E z?5uu+$?8fWD41+=$pU3TC8VdPWu97&?nG8~SFw|&+yI#^T3ud!$&uKLPmrP27?1bx z=Tf>Uzz=YJU&;B~Dn$f(m15sCJmA|z?Fea)~i*T@3@8s35d$m(w&>}+QNB8;f*3BcHf0vG({iv>dT3NX|&K6}PcRE6Our|q+vhVd4g&~$C zzg^Vz%j^BcNsS91pE~JhkC|Ej6bhD_khnt-=MaTqe}2>Z-It+c=5ryIU(c6~KPN{5 zeeQd22>4yV`~q%$gY-9`pT~K(+A5|yh^$T$k8jzi`b6;)9N#+ZpzYZg2t^ENR4n(y zg^`PX+rjK6QwC5fF!;0qkOMfJ+qV_J-1c7bYTz_(Ngq zJWlkKdDqQLa?LM_GLTKLF1hbvcRYVM6cqJ*-+O~PQLI((`BCIMow^#j=wt7TOL@)^ z87#&x7l2yTN7`WQ3w;b^AYzv%3Y^!RKkz6B^!Wvl-DcC5I(l_3S!rY3Xgm-zAjm)s zul!rxG{=_f-M?Ko{WqP|G*OTk<)n4wA*QjHibhP6BeSVil91y&ks@)$c(P0>)~hVM zp??wPB|%IV*GkRb$y+4Xb!%Fbn}g6S^q@Fj1Rz!5EWPE1EB>=DIslhZ(LMZ!FDg$r zpEetXHz{CZaHcD%iG6BRLKHp1xw2acB^2DcskYTmawMUE7n%U%NylIz^DZ285Oq>8 zD70bPR?d{gaY35K+52W9#blu4E_1KLXoPO#bbs0MU5%u#XF{x+-Did6K4g>+m+xC8 zp%EjtXTH1{n2`$|epkaE#W7HCaf2x=v@iW(wkY-oa<&*teF(|I6++Iw$XRF<@O4=G#6;=kjIV&dl^O|=esC=0w( zF)8RS-a1$vx#88~u~&ofWo)BcZXH~=a1?E zumI;z>q=YS0^D>f1e;(|p|eu|#kWFVTh@yp7Pgkk1K!zS3RYZ%!m>@KOcc#E&;b~j z4dv%~qke3>boVWd(3?z$gN~ud3^8>khX&g`l1yj6^CGx<^Aed7G}p-wHiqa61*xu| z@bMN7KyHz7*V;uf16ul2_Ts9Ogtkd zqJLL1Y-7;QQFJvzF|JDOe|MJd@mtvSi=iTd3_nLC1|a?itGVz8G3yxw%|S2ZTWg^6 zOxCLiw3u~ebEp}@CWr1_<Y@u8OcX%w7u3l#MY+P271I zoKBB1euGlAXdvYQ)};N;mzCSza5i1q8`~BV)uJB92Rm~CG!MB)Rg{DiN_C>l2M`B# z1jg#V=a&YX6)5K>+eLAu)2JJ!r`w`%Oyb;lH%Mwh^r3<@&mK#@1CvyrfX|DoasZS- zq4-G6(h&X$P>H3-Gu9kKaMk;*`wYa7m2C0Or^foFE)ytW&Be`(k)q6S1su|ep)YF; zHb5oD^c{=!X`<^$XGD+N%P#CTnRMaw<5sc7fgX~v$!jl`w_gl(WQg57Iils;)Pg?z zv_uNUJo{Sr^e~I;vdC|p962IZhJbV_XG{sO{j!GJiYF_R5o7wns8Kt{W~d2AMC}*` zAGOTI@Y^4UM;0r2%(LO_!q#_oGrZ=6_GqDR)fqKau9&&Y%%gUdhKD<(Y3Yg zxrv>-p@As4@$~f_sr5`4A?6$LQqa*oUN9E_e3j4DG0wbBQa`KG>43}DvST3S&kgA6 zTH|_vtMzkpQH7$W;AE^f%?I4(XIeUeTUwYW8C)G5H14tlOGNE;;kmv_1TH|qY)LM3?apQ5E962y(RNL}1 z(JJ(TjP=mY~BkTJ65*NUD(PKe>77$_9E8yc*3PpzD5FO-HS zHENm!Wt*#IB>qv`3uL#ZGdQ+VQ&X#yUP14dV@Jwc{mn)JeK z-#Q?OO0)9heh7_q!}M$V`eo-yuet*eb$|m=fRTwT!d}4@oi2HHCR{F6J9!BIGrOK5|zGz|W#r{YUrZZ>)v3n;M``=(i*?&{|4V#o(e+pSN zn<;vojXzLbgbgBOXvI8BSkj6aV^!OeEziX}4Z290oEbpFT~nA$P+kr84<&4PYcgum zl<0M-T)g#Wd~w5m+o+98CpPD1Yto6r<_lbz#LYHBoS}*rMEv-&FQb^GHuAFbYB_9k zRzv4r#S%N;mzukBd8Dg&mB6;hF`aNbcb@oHMKdc3GTE@0i>^zN#`mcP(qU9c4fnk0s|UKVE|S&e8u`r1N%H;YZ!b6-!nnw$8ZM}-dNJbqe;?(2H(Rxqpu)ruZfX|SfdRZz4djZq zJqQ$*pFD`C$eZEC!Ooi-eO)={QM_hx0RZ$Dmy|Xt;D?Ee=i^-lV4HF!IS@6M%b`ub zbu2RBbSN_KJkVVlBLRS3=05vi)+;uot5sm_V1J{a-ri?y(cd@gX#Lr;Zd{+(uOsKI zKUNi=U{IoqMcyEiKkPbyVxt4xmbkCa`R@ZL?nKw=^z3_1ud=FrEZojP&$5NQl;vGO z$F-V|=t`u&z>TZIC>C)i|zc+&q5ug}xIhfP|!#A3lc@AJjcF++mz!@!gzS+r`F{+aw&dI1`= zh0}rhR?7pv4F4VV#UJcT(p1?W$1ag2+ENnlEU+_T*q3@&jS`pF{qr^D)O(<+N4y^A z{R(1RFBLCmsBtv10fE@7Y`kG=*VuVW1@VErSo5xxeEmiMSU{)0D&u^cZ(&hWtpJs? zi+u>ISdUOte-SH4qJEIW$m8J=OOlPMuJDNCu|`+P(>a}_qQX0%Zpg0+Iac8Qsch)4 zViOhO%BgJV{KeC@i9{#^ncD@q4~Cs#=-9LzIV<<%vSInLcQWJEFgb5IFg8(q z>*=i9zb+G5wqL{{7678ES2G~CQwTfUXnimk2bgbmBN3+(G8wNqG+c{Jdxrf9XrQp4 zeyOGSA>XR6e)T1VwQi~lDgZ6f6Wpb18WP{ez_u#b6U;jF5A99I8RhM4D1V`)r1;>D z$Uvow zvc0my&${i}J!DKBuon zfNlEGk4#cfPm-MD=fE!jhy%318PvzkKo6N>P4`y#f{?h{Y@0)3fk*@&rXESFh>!_y zduH0XOxg0pDnz$BinkPrct|a6Qtzg!i`w5ndEkGKe+^$5rx0 z^!QDqfI%xR7%)yi9PKVq0M7vg?{P=MMjO5~9YD%i8%D!#-F=MuV)45E`$tFq1p8Aw zJ)Su&wp+7)p*O@Nk?L%&+u~hMl=!(!vv(HhMQ_J&dsf!W7wqdUi*A8x*2`=3MolVi@%_`*!Cr>Z@H^9w`gM7=x}K#cl|!2opP>+afJ3DgH= z&^}npXb&L3Xkh1mA6_Pg0&5|-t!lRdiO(SHQCumMLhOEVm?`PJSb()+vWwlV1(FCS z(EvsD#d*0Dc8ygJJXmH*62C=xSmJx;_mHx2S;w6vHC6vJfuA|$D=qFtoYC>2RZzR6 z4kZYVM}_21>b2%QuZ@@0tS5qgtquR4F#i7sNR$qj5@y<<7o5IyCY;8D+twPd!>NR< zP|((p^{HfPxF@LL)%?a3_P+XWuWo=;$Gu+j!ktKwUXNEeFL(eHKT}O@JbZjz1nZ2?uvDI&FKMxHza&|{ioPGYNKllbFJGa) zDFyYpZ25}@rn4H%@#^=3>C^vv3FA8;F){Tu>zqzes9QFYbTLsemJH=HS&9+H`?8dF z;V{YA%J=GFuLuoB17rN@X}%}rIP3sSZ`ByRK4K;AaY+0^!N_~FQyg8al2neu(RH;v z_$VA*hZTlS7?*gP_PUKLT6B|0D`E4=Y%#6d-NvLa;7n1e28hinf%>=h=Z`oc{MT{f zmM@YJeVA?58ykxr&Ba=qvCot6J{x$ND;Io@%iw0$2mRYV3?!7otuUjW6e89DOIS5) zdm_2a(S^V0FYGQW_&+Q`+mN|hkWPEZf;pN50WHbsK9W`)Iy~%FSnK$chus7!oeOFB zp%cTWiNoLF6LFY&g0(@d#YB_v*6gduoOmB zgyrs*hh3|F@itNi-a_;Ii{(dA?ux}QN9SHh9txe#EcSA>K|L^{h+7BO{LK36URz(8 ztI)EUX}n}9QrC2t62+J@|4zACVhd-czKH|+v}aF#B4a9XlGXhy$**&$Xy*(0f9!m$ z_xx?fesjLLqnF>*bnoe|*=Aw@%>)^sgPqW=8L?0*^Yubu-dEVz#6X|r79H$&S`bA( zTcdbpj*G<)PiWI}%RiKYReSvK#aEfxgwo5C7NlPR--bAHw-(PfC!0x8!XsT;kJ&Po zSC1|Rl=NNwO!3?ui44#XncbYdZF*N5G4bm0B;Y}YsjE91jV+(TseJdne+rU&d^-?U z&*nhjxhrQ9^+>$=apa>t@!a%B^~d)arN`dN==xA}#cJQ8vOm?z8ud%EYQ`E!(#J0= z+jFtJRu1C69y==`G+3AubdC7qT*QC7=F7}dqZ0;Gle?I(nnyV!e(Zgy#?UH`zak#! zgom5^;&wMfa;w8CdtR#BD0_}6h>3$YGu@Sg*Uu|>wZy7TI;1$suWodq6qEcLi7ioV zKya*AdSNN7KkbZ1!MGllua_LnCaa&+JZf>gFH`g(gdw@lKD4ji!unxR^?W?dz=L1o z`sJVF`kyAzk7b@_i~rX1dliDd4lZXuD-(h?t(C2LDRg~A;2?Ci8M3eZZoKQk0z6e5 zvUfpVB?o*_hU%P7x5`Yj&pCYIsP6fP{Uz#D_;3Fyu2ZZ(Z*;+-V{6!Dcn5(a0G5Nm z#f~Q4{m)po7Bl@2>+ai5XfGA3BOEQpJImMALdpm~_0s-A0{psVjZ3Cru#GpjOAnD5 z8ISexrrqeY@nD(471x99$fOT>_X=3vJTtLPRHVLv1yBiDPxRm+J|Ewl=VHNWqeq%f zZ$4xfT1FEAI(|nWg~XZUu-y9qOqfdVAx~NfEm32}9QUAX{xac7v{~P0iTTbU(sNd7 zXj*-WO7A6xG;ij#A7rB64}*kq3n1IK>|}h1(b91*#3OAv1|8R*h+*;_kcQ)ou>D`O z6OvRzGM_F#O1$oL_#sO$`S9$cH1gFv1wh!m&?DO3^@Dhv_2C1lJ-uR<(=CmKEa`Z- zKtcMd2`^CD&5`VLq)@w90*<@B{H;EsUDUl+(fWDW0!1`KDR?m%EiB_&1=cK6-id}T zD!d0}Tjpy@k6EUmcqe9Id;Z8ugSAhfr+)i6GKxZ2D6g|i$*kZx31NcI&p?k`k#&?r%$8Z>TJz$B^Qpk~CaB)C| zE*vDerFko>N~vBu6DG$$63w^CUSH|hWe2A-AVDNZ#)gJzc!wsKRSw2#=XJGw}wC4Lh#NKCiP%QMz9t zIx`!v=?gcdXO+ivezEYvzlToAt??~XfjSClStZW+=q^sbvc0|c>3_ekO>ZHEs&tmHPaSqfm+8LU4 z>ffh|KoL!%0kv5lw*Mm0{F8kZ#6FR#r%3?09jZIH9Tn-t({UW=$02m;j2Qr4YFS>r z>!YU9gF1M)60LrCE8;1Wu10y--m5Exx$*dF-uWop07*dO23KR|M%GeMkl|VxWHO{A zp>Z_Xq5=-8vPfv|)?Cc0VJTTx@jp5HHzlzDrzHw~C|@kWZTEL! zr{op)&KwtmGm#wg?@v`g6VK2KZUf?v8WEOpcKk=Dcc?gfP)>LK~CV(0XKD1ld;? zpk>qng1FEmaif$!hJNyByA_4zzkv*7EbTmj(@xhpEk_2adDRgsSFV|E4)9z)_I)vou2-ql5 zE4KpX*`+mu_z#Qg7K=+X3alF~B2Z!Hc4o7=S+7_JT~D5C&)Ej%>fY-X%)TWj=AUX+ zy(xb&Br%jB+5gn~g#S8ciQ6js1iv0}45~cSn&F3(49GnlU18ohb5nUfIDy}IJoeCV zDUvH0ODg2{>9Q37(QIpnB40Peq7mYSk1zG1U?+#bR$E~xy=W{q^YE9dLf2RLabilB-w7p1V*rn$Epe3Js&OvYRjFub6qxfmPhJTVmJYuS4=p z=IM<9jNW;icIluj#39cdmiv|IFRO_3U0xl!BxDhQ?$Y(-=sbSs`ftb(to|2smP{R^ zKIL}1nIhP#|9a9WqEPc!%>o1J1;T%T3{qID&tI;)d4m=vMDAa=KGi+?BXB2k+Hx_T zU@P3|PL&;Wnx&H+ZR(OX6UMN4&9XDN3 z{z`<)YVf)12p?c+Ow@W19K$z~%4y`*;PHr*j=TaQT3ycs#%zpYH+jR|`4c zc*Xq6lYv$LBnq4H@zD!=Q_Vy4$UMxxZWN@4XgogYcrLWzoeRMgn~@e$pK9hqRvBu! zB#pn0k=#$)efU|-S~s^pogp%-?s9%hdKbh?ax_WGz zepig?TC_MO)49A@$`REl7fXHiv;<=vy#KE5adNr;2{@RFBPI99Y{)X9=n|WQH(*U> zQQAs{FI)PeGbA?N2Q;@3WhX91;Gc8+P zlRL2gE<>9q_9kq5*k)DbYfF>(8$tg)8E4Rly{=n%nInmot!o_F9nV_kQRS>MDHZ=U zV99JqAz7f%n_>I>?N{=H<@_V~0MAEzr4dG0i^J%~Xs;+b_v>7fY4+KZdq3%%5yK&7Ipo=I2c1DkW7a=6R~Dw(UW z^$1`R&^p64uY-*#o~afmg8Jd-`xX9PrtWmyO${6E6-gTb>rF1!E5k_9`a8UCP6GqZ zAxlVZk(={t^95rF=SX+A?a8VC#Z3FZn(BPJg-Y52+&H-*ps+afDrTMM9=N%0b%BKRYv06|x(Vu_< zN+wfdkzF}M;YqG={@SKLO^3C$(&YNyka1hMj#d2e(om#Wflh~~L7Jj|=YctdPG7nrtk--Zu) zO&lTaJ1%wWNY;g)_n=^(I##UpcD3X#FjxWoDq4JSxP*7^?874r1|8@fc5-@G>ZS?yT~T4`dmmRkXv@ z%VH6|0F=B$|#25E^E>FTz#&vJrGV>2DH0WWjw{A(~A%PmFs4K)*#FW}B-`3$HQn zrc3y9AFD3<_mhdoNneuPd3djYr~8spL+uPb06{jm+WC~GXx)D{dStlNn?_?C-ncsY zo(1MEw!QkPGdEAS*@;Lu*PV1@ir%}QI9G0m*WeYHAF++qiUVtcB909Ro@G`*;;mq` ziHCO$0v+(~*R5YPUG)=Stil$*R`)TiffbQ8{$a1Rg%`u`SoOpC)fd!K*>^`(Q|409n9Bb_bOuhH-B5&OF)%7d7vK(1*D}nf!HHvL8 zLlcLX_A6_lcsa-}qGm3xFs%~HuGoJKEUuIInB88gXP0EW8n%WwkvTxIEU3px?Zd>% zSMWd@GCIy@yyM+5JcbAbQZBTpt>X)4V~y#DOfcxz*c1*xf0Y3dAwZiw0|fj$lMNb) zKMu3qo8uB$~nZfyB+8(w)9G=VYpybTA8newh2MyV9M)E|71!ULzsN_i(l&yLURw zPHPkGkfS$;*DDWp@KS^V#wnY>ZW;E#Gw9F5JdO@xLSOf0SXv=|oaU8A(*7YY9ZW)T zF6d=FrUGq#4Dg(2yB$Dvb@P0X&i+)OB;Fwt$Y+fg_@ z#NmUE-qK;TQYa9)Jy&ss#~|IdA3c`|I)mV#+L20%752sVJa{xsFT``_lH+3w6o@3Z zI(fRY^o}TxMF+#bTX4CVUa#GWwnwm3y~?yX87rmEO3BXx+K~4A?0SY6JmkX!O_k~7 zbB}#%)nx)0NxLuj@_E@at?<_Y7b&D05_ybXM`_b-tcNC-~-qW4?)yF#bIb!Z6{*BpxChQ;o4_@a2UEpj8CCE|Rfb(jB zf%`Kel^l;Kj07sf{S_z8$mCC*$)1fX`@JGbBkeh!r-d7(`%d%>fU?|z*9#!ZI3DK& z$1qG!-o^Nbv$8m~ma<>qwS8R#Nwpuup-O|*Tr=`?Xe)w{!hrYliTmtk%U&ICmlv~- ztjK?A2yy^~fCW(Y0Bfu<>nZFHJxx9X{wXEa+9DJ3b~$PE&aKiD2L$Xx z)r2_+^&4?)qxFPHyxi_O_VMTT6d%&4RrMyH!*E8k6b9~0x8q%7(p?9O8>U%Pz~jDl zWEwRf=ew;usCZy+lAtGF=SukX`Z9%*Ux!kEF@IC00~e*av13_Ni=2Ed!n^(Xe_EMo z{_i}rX{HU`w}RK`-k?W%JJC!W&3LceF za{YDJLTc}` z1*S9?>H@7_wL!S3JR9^!xh>|V&F`JcHuv-2Myg*x%5@W6w_d;HL62021>MtA0P4pR zN!hpOyHkX1t_mEKya~-<9iP7-e{P8<3C;fut5L2?N`pAu#c8D%5{qekhK1q)eolz- zGw!!m6dx+Viya%uT1n`|W)bQN&Ro*Z?8gQtEY7ud#kkC^E+`1k>G|w?VIk}qD(AxB z9%qiX9<9>fh~$49YY7%#Dh6E^8dN;WGNK1Oa=G3%GMRtGJgH4j+sya`eNwia`!;3T zY7B09grQebZ#M;phKs&?KC>|fO5kVxbYXvEWjJH(wP^E^W@77MJ@XMCBa_$Z*cNii0JjKf<2i_^BI&TM=~`fq9rUqAaT>s&k(xj{D}t z;mVnzyH(1}iJ&f)z8gf8n=lNe{5+Oq%##I%G`zHxv*{j^U!qa*cXa7Lm)oBCMQ%Gh z@pFJD1uy~6-}iBRw-Od~bN-h9Z!=6maLs4zbNR6;q3C`#-uBC23O1!E^tLNmGB!wS z(-JzsV?t70?Y0STrHcE8jJuxA=gN|ep*!AL>!2^-9+!2TlP^*&A^vpS=N4Gh4iTsb zDLb^8M$xzr*H73h*N5x5+C^2}&y`Q4+}T0syVY-WY5{WpDukyKyLln?;Zca?-O`D{}0k{-^tIC3<1^35p=q(nDat}wOx(NalbYM?lKH1##NV@Dgkw+X_J4Q?J5%teYpfx?e+ zHOumuq9+%Q4KEkIFOZ^<$J(sUA))8YGJkuhkkwnJ$7P};SIa6>px>`~{OEC}Vuqki zaT?9mkl1r1g6os{l0OZF^>`mqkD8u7;7T?nc+e3LN)g!;MoUPZtJ4(BpX~O6YTKw= zd4%Jg`v&B#mi3(`WvfMg)9f5W3dF+d%cKDx$l6Us<+_!rf~iDd{aJsb2`%zDDQ${t z`Y&^Eww=n=92`~~jSXcYu5IkC55Yr^=^<(V?}3oAv}=aN{T7FQu|3LohSN@Mf;}TH zF0Wxk7X$B$%wQH@L(m)xFBS8_VoG)w9*}B%HE1c#kqNgz94344DDF3ZWee$-__&-+ zhJ}Gks*T;4DWY!Dt!J4Mkibwt;A1{*6fa?J?y8yZwpy(J^jG0I0L1-d5Z?p1;z7j> z972rb`hko&Dfwn!f4g8^nR6=|y%PhEVYWF2%v^xEL7QmG%-+#NN0S7eux}2jp(?&m%b_RVjyO!)2E6we`X7{LYFNPp{wH! z>w%oS3p|^v${L~=C@+1GkAU=D`OsnjVM^B6B({}cY zoa22DfQdL~81B~d+O-DY2mmAgoD=;(AP(;fx*%$zIL2VQvkpull;NF>Wc97GWT98z zUCP10;Z-(_xY>;RH(j>~W0<&(YEoaErhVOQ@YyST5($J{h0|i#_phnk7B((2zR!|2 zznEed9c--VII3}9!D~gXFV*9SolS<17IQiCKMCM}THI6;8;GEWJS5d;j)_uUyk>ff zlsu(SE<}>1FoPv{H!IB>h_b2+wJs^E^#w7lH_XRdSU;Qnwzz3*CAhQk@&(>6gC(ZT zdn_S5&HJCnco+_u8hLnqK%=Ua+bnO`)jc`ZHw!g>o!f$k z(6n79o>Oj97kzSt+6ZrY3J$qyMVHH7UCm2d2 z-3h^Ty)$uCy>iR_Y^^=5dsT6JD%ARrbQ$OJNWDlI3nvwMxlXe2!1Z&U@Rx$k0L}tE&v&G@^XJ5NAxn&J%>CO7|IZBd&1;juo-}FavK{4 zabo{t4_LP=<_tFF`vob)IU@kTz6pqj8oqPrPkC6Iq^u$_;TxMuF!M1M=3=s-nnB>Y z%aE>$6-V(OS>uy`UL%%LaH|{JY%t%quB#xg_qusZ`unq^AG1Qr9b&b@tuLG1ajFV_ zqyoDHl0CaJT3-l*eZ)3n0Am?iS08j^Dz+T&t!=<-UK4D#*5WHx`hTV4WaK3jx@p!< ztbotgPIC7qJLww8hp!>bXW?lZ=_Ems=P<_YDI^!nu0;~{$;8P9c?A7K(sQ%ulDp0QAqJKPy8OOnw#UZ&#Hyn%|M zW((4_-DIq~%UAWsvf;jVkC(1)Wjl8Svc7BT)EReqH3iB7D~qCVO}KS;1DoONH;z-n z(20m6sq|0IkxTA~M(5Cz=G_$@bOTr7gn5ext?op6W2Y{#z84`na0bXQ@NwUJVS@^% zZ(~w--Zm#u>+I(`|2NB;wPm8%(!{X|U%m{6bL!KB6%~17f)ys_(p_n#1yzGEG;39Y z*vOd&?lOHax)gH#)9}26xb?3Z00LAXDgUb{g8%kJQlW5c(eH*Wj)MnK>vW%kiKPMSB1Rh3MXD*6k>lHTwHEK= zsGhmeYJGgyN>AKX%=PR6C!j@zqlU%nT&5FbM}#=sl9|qEat6y79a?O(ia8^OS1&i% zbN;e{900W4GfQTq4(hW*0 zr3~FENOy-4f|N8$DUu*}bIFNi*blnvgI{rt1^H#KGmXaH`#Ig0QlC#?HM`|}-eLP`Ss}HC zL<4;BfriScS%dWGE#a#}1<%ptcH8NWAG+%l8g>?5knCA>(Y&&I?&Hx=s*&n0d;g<5 zm_Q_R`~g8LKH$lf%Bp`CSAxgz-|3u(v^PTq?pgU`Y{kGyjSQR(ats+5wS>!r#A&O_ zmUgl0p~u-{p>s(Wy}^}BYHs3qox$-DVWIZPaL3IF)!onMK> z8ABu%2g<7nbjD|R9=S)V6#Dq!gTl+(sLy_^@Bw-)4ZMNb+m)bYn}W)>Ak*8=-h=ok zt6S9O7d1!i&REq>Oxk?i0QYqtAz2`aeTaA8xs%#}vL1Jn-xhyivk!ifb`?5wwWN`X z`LW`fU|300V2eD^CZB4`#`E;cw?bO7Vtt`AOecD zBP0*i1rdu%!?NQ1ZMP7e9T;2eLT*-Hozo&)ksOa>US8gx4B)bP8!ZAZ zLgj+jZ|S!PS|4A^czC#EnweJEyf*RBbA5Fz3g(48hFDC{Pti^kqiDeWq#y2yH@5G* z8ujEw5WA_setTzBuPiHld|zGGe7`L6E;CciQ}aGU^h1M)<5lLRw{&`C&TT}MHOYQ# zaI5_A>f;n(vVQov^JNH4XKx|C(3gZFnt{u;H(uZWw4i4f$s(!Hcwn5{+}D7X@*A(& zdsmJ#?qNFPA41d9e>!MHA2l(u`(02nz9Yam3}2TPAF|_w@1eJt+`N6SKe$UmMb10` zcv0}(y$n7fdgTO=Vy-*nPtNR^6$}mG8VJN^9LNr0%hnyD^gGq^0(b8P&+Mi035DTE zbs|DDRfKUDJo=xWWc-VlSClGmM{Hh@&bR+gw&Kt<3VOEqXwLDejfXSH7!>~a=_HYw zNX?{2!Q+-~;&5nY@^VQTXI}C`S=U06fXK@?`z6r+K)!EF)BVDMYy1!xu-=3mJMk0WGyJ&P?+4Twx5B_|#BodC zgku9?%r_geADyj)wJ}IUSzd2Ezxcs9snhwdf`|;8QfW8yTMcFWY(5r#PPmYXf0D_b2T9Q+VH-OcWq+tEnzr|)XxYUx+fQkikw^+RZ2Hoz z=Hcqg)kJq>DEpkUXd~e!MSx84d28VT*U50-k^vW&F6v|$^O(rl-<5(hv_urn$U{e` zLm4dM2339dBcaq@BK;phC)E1Ste_O%yxsFN%3LEa5pyQ=FA>kGtP&@WCH z;Rp~2Y_#Cg+}>y<CJbaM}$JD&-Ml7y6UMHREDXiSWm3dJBrYv&)Pj1#rtJD z!S{uLzjojIHWL!xqTS5;N{)$+CxIwN1+1=`Ys4zYw-}gSisonXXiy4J0jFG&Q4Q~| zTx>}YgdJ45g?0Pn5YVSG_K4+n8bS%K-%i$qH{vTpMCz2D@%KhS@j`b(%dM{GnGN;e zGTgm=Gg>!B)Tg+5t*!hlw}*eX^GZwQJuu7h)vYMuR)Se(JO`2r(`vNW;U>5xNFHX8 zbbc~?BTDeaX(D={1x#Rv zxEpm&cyi8_y%~0KvZ2nXjRZG-IsrzQElCSBC(5(>2r(wY{c%-Xdl|SJqO<5IwU)B9 zOH&;J1CbA`PU_s!w`wB-(_77iQMB@~TPpP_v$=sIBnBKMO8wbyGz&OVVT7d7%_yQK z*C>^@Y`g}SG#lq~?{5U`OvbIK?`XxTn-rxy()ZXmH!#Uh7)vt$Ka?fme}HM8F)Gxm zJOw31aX;B+c}#XUju=a&TLMorK%nqv@8gNpL_XN93W&gcOIR_QY?T`0lVjJ4R;oSM z3RcLEmz8h7w_R^UER?K&^*6TnR0LX_@Noz!`axPYuB4FLyau`Q=ZjxtYpIE3k{dqW z>|k^~&#nBkwPE~WW96Up&VGyD7T6ijPB7V;^-&Xc;5z;N^DT4lTVq8`d+sA_Svw!G zSz<4%!KELu=ehm117aQ^-~kJPML6XsPz77#%QH+XK`aN^{25K z>JRDj$<#u6b$36l=m%GS%t-bGf4*Z&;w{`7OvQYEpZ(NKxixwe7|_j>?M%cwh)$Gr z9_|59IMpBNHZxcX&`7~Xnhh2P+51Qy0Zoq8XX5DbgI2wrJGI8Jw8+EuewGi-Q>e{= z`kD>HoY=4vYe9QQ*(B@Pe2k`_I(G?Oqx<eIADpYC!qZ2~TB?-FK2+dG&Q`E^+w2PY#q&r?pldK1^!DVu(kt&RFqcg!vGGWtj zF_OcL|A*(q{~bUJ!_$0d%^idvomhBa=*yj)rPK4Q&@;=OH$j=#f{qYKvE#ESNHtmE zWCh2ZUHS>2yko^dc!PYaVJ&mTwu$v?#1>_BTf~pcy%7hTU%2`0teV>Z>o0x3OAZ+AM{>?L98s3rZOG1&EOtXAuk8`1zNkOlkdaah?G;La z{Z+%u!Dq$C{CQ8$b?YhWRo|kEo0bljqCt(EDKF|Q06ieBirt>hV1ePH_Ra6Izb*%T zLe}kcQje7i!Vn0#J%XM>(Z!X+X^7o;-Ce)ZapDmzYY6ZV zH>mc6I4$cnCYmm2tJ%mD7CqS)TN-FtLXa+j;G`M@;_f(`K5?X*oK1E&A;S>Cx7kEvTO-5B3JB z^GqYY02!l^Q%Qb*22`o|_Oht)K#p=Od4ozEky_r@p+cJ%1*EVW|G^E?31ZS)SK&r& zchF2S=RXfW#8+p%3>DC-Fz-EG0YX@AOvrV_@fuveZ(rUVkg0@v+CcBT6cUJjcJGzQ z-TA>#!?4Nl4-!3-O?M?<94hhK#3p;%Qp=?7nd<@JvbVfWNjZ;@!rn#-4OQ=~lH38% z1LUZWEWR9#EW-WK2l87fuu(lX58f}yZy0tNG5!fH`*Ona(e=Yaqjtyuk<~kv8YDau zMywj}P8jdNn-$PCr)BdZIW!J~kyGp3@?tnJ({%x4tb)kwY6n9ZIJNG!5edJwk`I?1 z#ixo8OWpuQYFgiIqZ1xLut&?h8>WbnW)3ma!cc`gokim2++|hD?OwrV7x(!qQxEsigri;?*FCLmn9bqt-^UD!~gZD8Yy;gYc4;r!cCVdR}WO-~%r3ll7OuXF}%3AvUi{9s7qumDok+BXo! zdxcr5xc9n@-MJs=#z19}gIRuDzGe?0Cs6~(LIkzMrtZs*Y_>57*G^xIpg+{gBC*&; z$kT1(EDzYS9GY|xaW^>lGN|zJ4hq^`pyPaa-DEqA|Kb55^pU-SQf1U=w;xOj79@4+YVP{Ovkv+VCfIz!r*{>GgqTAHzOaR^JN$uV;S4x(04yF$9|2ev*0!zP|T+cr?k(ah2MI?Je>gOxS>SGWxR*8OGSK0dckKJ5hbfZI(s zTks14VBB_mhnlL3jO(g|D&mw0myjgQD;v5>i$TqGcbUF9>XpLgcrhzc&$i{^3*5pF zI*&^Z7x+IQVbQp?xcmV(h6YA^|BOb*>$?(B=~VMBE25P)o#}5lFRzUT_@c{ry&6=r z6x5SD2^kN2BXE7nEGGviWnmJF>*I-r=+}{Z+hWm9D3xszFSMI0lMCKMvuIhaLnb%jmghC0?fRs$GWbFk44rHGWG1%%(awUM}%7tV%>O5jdM zNG6x(L`utS`_c)gPRmC!296j9DT<2G1xnU$_=T~WxN}Bb5qgm+2YN7?Cm&1XHj!u* z=yEe0DDSpN{^xZ!`ENMv>C*kVlLZ?drM^Q1@n?HhUxty9*}V8|&uZe=J*yWvZfIdI za-BM9wl9o5j-~Xp@7sy)i}!~&;s^fziSFwqE!7UQd>DT4o&b~0oqbYN?xPPTprw8H zC2mc>VQ-Ss!lQp(pKFi$ocR43IkRr%eR7xE2}JUJmrnRadL$Vy4;CtATwl68z4G?^ zZpWwDObbm9D(D(~i@xpaut{c3OH9+mrH1h5=^6EkWnr8E5O*G5t2l*L_&9#ngBx`uO zlk)eF6@xw}?(DTO&~$K5vXC4)vILx{STyjHMX*jmw?`$}f{znbr6{yi&}w;!zAt?1 zoP2Fq^T(+%x1)t96qrAd3ZVDF&~j{d*fC}`^N(77(1 z>F#E41#%84sp*ao!f#E+zdyDdQwd=1A8>tX>1qq)q2| zYBz@s(W0iWZ;guTyv{u5IZ2=V+3HEuB_)}|q`Nl;!N$+i)phTgk5+wO2YQ*`Qa@PF z#1RfTWXm|)Ft<5>okk{ba%Z=FvFnoxx5mVxu#WI#gR{ft?B#tH@E{mbIqKpXRx_SR zi=GL3@P*_7Q2x1>d1>!B(RLUecj|`xgT3h*3ePM@0@e%7wZ`!KMA-#XP~_nrtKw9E zX!IF~a2o6Ha|MRQG_6G<0wJaf=c03QE)!ieS&Jqj3@lNF3`4bm@W%rnG?EWuQ z1~00e>g-PPQFXLo8Uf{UcIMh=%*dsC!tOy+2|$n|rc@gpW%-try9jxuB z%>4+NwwnLJcIerAIXv5q>2KCU&o$jZ_G~P0pWQ7LIq$LAK}UVLnPR?WZ=1%$WqRQy z0}TvG0f!KD2kLp@Ihu7aa5W6HK`7*PqnThxlS~Ifau)~;d%o72l6p?zHUatZDOXPM zr7V(Tfl|?svdws7IibYq3=&sD9#%OR0PhSyTsfX=%vm=r?MaxsEGZ zNt=?xJa+!olBOHC&9b>>-018rF9nnR`Tw(GkBvDR7w4z(BvK5aKl}Ho4D4>QPESU0 z?CUgZ_td5pGkRkb(w!h;s)`ifppFGgLBwjgv9A-nF5cztO>{X^IX1FN#iaoAI{C?^ zv74;)u2kCG895HTdScY7@zPrC=ymc*t8SEx{d?_+z4mg*d>mn`;>VT1s}`BMJM~;I zbN$Z3?cJ$n!~DxpTx1eEvV-|IG88|JRq2i1wC(I*+BLc}c|+$5emVV7vng0u#ZguIyryZ+gt0p13I zQ6LoLU<2{=F!4I$HWDsUM7ux1fk?CJNO>$gIS<|=llv_FjHoX9^<=jE#H<$S<4>C!L(h45c~{F_xokt7Uc*8>KN-QxaO9yMQ8naVu7NcW>_h zq!8hxYPZ6xdAov>i?t=69&)G)Z6|7(chlGz)gwOpEwBKx zq^J+wL+0{SLTU|Gn>kijd?e(z#+RLq~h1eM~q@O1fxIzwH-qt5RlhOPTZLTY?*a#cHu4q2t_WD=35dA_iOXF~E5bb;iM> z+5c@3^w&g(!5g8SE&wXOOG>ACa(xodPIHfxFwy;UUVp~rGG&Rp?^D&an>j@IVx@&J4FP}4Jf0d0!)8gK z`%pi>$Qp^l``$z!`m{Kmn@Df2G`=DrWi;BTtJ3P$ej;kMa&uX_3}Y~jv=~+ta}+(n zFVAacI!^yZHc9_rBxve6JK8_>TEqTlSO}GNhRFGYin&&*aMWGQS4@28MJ0z`kSP54?Bjfon+p@US}rxZ z+U!~*KQ%j^w9*V~*2v+~A1hjcanRAHw}B41OpaR>MG9gr-pt z?hUDQ4XI+w0Z+iCb2dtGKR7y?`q3iu|4+4PdRoz+$%OEy?58bU8X)+Cy(5R%TnPR~ zCX8-jAz*fHIu6F8o-K;?D?D&-E-LJ>GE#Tz=+~dXE-}60Pv*sJ^mMCmsDRxQbsNCg zyC@ZyD4oYwtsdCG`EN4-m|rhd-qIFKRUi(62hY^tGm|!qZaa<|qD4$moui^RU+b@?#l0yFc4U5firVeaA*`xAyCFm+klm5LuyUE7Wps$Qg0^`??{HxS74cc7D zi(`0eX(;8H)A#sI-bbjriD-LhPXMH6Qr zGEk{#|Km-;hH~3y;!@;Q0&Cd4G#4&R+U??On8p1wzkWoVsK`-SLpALITAPlGHN=5f znbHizzEWsZD#llo_kgIYsKvj>rVyw)H)fNR``TvtE1ATC+PrrCB6YYQj|O!arO}W|-*i*zr<5h2Stzkz_EOq)0IQfR^8(sPTyl zi-~}Hbl!A}AQ8JkU*dFUfTfNEGoAa-=dAx51rWv}d!39>cW31>4kw6O`JAm0yd5S5 zfq0#F|2&gb^Z0gjVYh>b&&V5!l@P{tOoTQ>TpJ zn7Gd6!>3Qb$qWWVGDWAmum(;Whfum@2kMur({6BQUA`hjs{$ z9icZy2W*8&zj(EGWWw7>c=L5NgP6y-vm-dt3umDM^|U$R$C~8W0(t1)PRDG8C})#! z<#(CL33a^Ci^Jy~iS|46S!Q{F|4@4(ksvgl#mDM@FmSH!-9>J@iD$T3LoiZ3ttGb@ z6Y2ANY)Txs~m{r z!NIT5?nxzj$kB!m;Jz8}{>;|v$LLASpIizzH+ZQ)ey|4unq}Bc@l+xXy1e#>G zil&!QdmD5IJY@7=CTS3t8+g!@S&*KLkm__Vp%$7c=;wE?VFDF5DN!Cr9k|DYziBOI65?>$O@sDu)~hn9g~#%G5=iwytVERyFAYl#Qr)L# zI$1I>UTOdHYQ~i$3FAp8)0u#fblGhDol^tX2Xj;^QM{kCS4lFp0u02H#g~O8xl@~J z*~AN>#rIH*iZ`h_UenmSfex3cy;-tNSKr?ntFqQ+xk8}z?5t}pJdWxHbMB?j6-NYl zk;iEP+LI#v=9<*m|HdNyp%$!7UG{uyv(7B0Uoe^Rfi-+BET_RW4&?*|Hn<7m2%)T~ zIlNycjSA^SC_rFB5s-?+E&pR&L2Ul?cC{HnU3ZkT{PL6HeoC84VLr6}Z;|NNJO>Uk z6#Jn=QQubZM6sMEK28LmgvJs==yAKBn#(@|j5J(SV%d6Gz-erQ}}UsA56 z8`xIoEbA6+t1L%q16oEkI*^HxTD!S!mLvAWf)@#O!Dwi&JM|w%rhY4G9#slZe^N;2 z8b=6Ew|IWN>dMjW+f2$Q_O<>05GOiG(_AhI?1;VU%H3hOXNT9gN8%05P>$tG>pl+P zagouA={jjIzA4ewljz3Kib^ziO8oZ8wxzB_)oP= z;Xly1e$T1ybw+RAsP-?&{uj~esKXW3Hh4&L#T!KJ7ITe+t%kw}aqW7H4f)zIwIBL$ zriaAxDD8xV8!Nza%{0AOu_24T$-ys(g?tVS+zXa6@oL7{3nJl$mXl$m0X(x_Trz-r>9JubZHVr+AD^S9g8oPTydH_%dNEAHO9h@5 ztR-9}0t&=Js~;=9K;WZ7x?tfO^6iCZkvOxTYVLZG2&K(EjFtxCsHWM*Bu?xe&2p3TOcsOi zt$ja;g5C1lND5y`ibov9+tcayDFZU}*mrsVhpJ`RPu271Gu;?jOWfBGWFofS zxJh}At!bMtRG4*>tH-`28009(DhE6*lm!{@-C<8w{@TRH{ibA-`43 z*zj)7dk9SCz?Y^fp8eL9ES=d(z7o4+N0GSUqa;qWohJ+nnsv?_NFKGZM9)GT^ zL+Qd7-74tou}K`M;+i?qik0n;qlsoa*}<0tk+$YWU65&$@A$ND68I(Zd=?O3*gDqo z%y~6%ADfPE=KdJZzcbmKv~Kl~d2MH-d2t})bn}<-DcRNc1-%i>`sYSk#L?33NVL2+<$aY%`Ok-m&(H<4DQqygOlr`#QO^W& zc=*Sd(}|0jgwzskvxV?$A;S2DQ#XA{WRWqm5Lt?CCG#15H*A_ulOswQ1aF5{1HAg~ zCs0>9NVJ>0%^rdeMX?I_y8JK%8}f4NxbK5*^2 zwcCg5mx>&{WSDT?!B~oYa|}K{p$Hgq`%W$f z;Nfkb-}EL+E2;At{*f>n{~uxgJ;FTU0s7PFb}33H0I*_H!PR&-9mpzUb@>vgE?_W( z!AmMSl&$^+MHqDxyo1EPcB2y_i2}SM&!RTt0xTMg+ekQ(M!Ar)AUGKCLLR%Wbt2ga z1IS6^+!%txtVimg)|0M}v zo;^RZ$)*&89kfUEz6(1oY=A_Z{!Gv&3-dbs;O>;ypFs&hVQo+tjieXPPQ^f^M!K|* zYTDV2gd{3%6TS%>XCsY5?e@UNc`n;4r4mWRF2vU#f5;FtY&doE>T*WPW!I*OFfE=z z8pYTHM60$SP<>~guhi|Bv1u#=yOO>d-O28N*kcx@SZuyb5SJWB)`?x}$p4@Hkk`Lf zjb_+E`;nV+c)b41tSzj;tR5}6L#!e`(R;Ka$FFx81spdB&I+Wyh|g@jjKdE9dGmrv zsCR8pxO#Axl1)9tfInSWe41{1&N^iaMk&SpK22y9oz_h-~E)4;rN4~?0Inh zHR~D0A&hfU6?8jZ^rqyPp_TNeiPc+QnQ+w8&}Rbe`M30y56=#CwiXM%>Jdb=^@W`l zS|dDhb^`*)I?TRi`naGt#v+qj&uR6Rd9>Bfk~pq`BCF3oj?M7IV=P^qloWlj9DRD} zAF!y3|J^#GHR?Xs{72k3{b}pBFaJ!$IKNq>Kf5d+jv#@)45j+I7U&`jmtc~CP|HXT zyB6ry=*FaA0=v@DR*BAUf*=ID0Gbb5aT#PMH6zyViJO0pe-Yb;ylysBqI;TB+Tn#2 zIPXS#sJl@9pXjT^PNt(LP z(|s>xl$akpz6R^65H2%78qN9m;yI%YZNx>lHzOYc-HiH+ zH6yx-Fv?=wG`ABcT?_&lvvc3O9uMST3#7QiGwsTBa{D(uaBLQWf4CXs^g-@O4;X>M zWpPz@A$rS-lw?p2SSW zlsp>Q%CUCHohaU+1l1f>^Zj*$-U9Xln~AdngaA8fZK%4T4_dNmMP*eeiK7`9@sJ{+Jv74yDsZ{UH!qBpwDzN!X^Z2brzb_3QNIGA(oO zZA&?>_S?1W?!NbYdSiH3`g%vH*z)6%0>~YMt@_>dF;8}M z9LdomJx4>Gbo5AB%4noy|KgFpS2XbDbMM4J>Et^aS5OGS)qBT8FHowGQ(74Z5;blF z1`8oK{X$^ChQci-9mHA z)PJUR)1ls+Xw~)ZO>O>+qyFH!10q6^7SO6?c`Ko9e0noJNnd%7(1*pR10eb*m6g}^ z1QXsL&6D2ae^P_JHf(`CUBapi%oIqY%X>>RF1C`)r*H1w5Kg!2IlG1(nJUX5$N#A+ zfkocq|3_l~3yBe@akBx`28GjsYR?FO05(S+1BL4mgjrg2cE6jN&GvANKpekE9=VMr zF$_(e#sCr9yTBevj{4TS5G_YH&)y03+^L?F=wYd3Lw(&LQ2KbT z6r(2HKtOQj#(REh(NJQEGUyjkdQzbE!t*7U@G-yC)-lTMTSg(2=QrBGQ|=|f#6MOv za50L75un6CGV`b<=BGj#$sta(t|q(%H=>#cwbM?Iji!Brbze%sD;Y?bVm z2)UBoC%Ya;LQoQ8{fL> zdk%jTWB*C_bLpCO@!o|10?XN9GLidjNf%ENo%DWiEo44Q+Tw;;%q9%vjt3ENQ|8BW&5|Igo1@Q<#{>iO;p|J`Q%FSOb0)E|~2)QIM?vM;K8TzeQJ zvC0#?TbrAed&Kmk8pMQob9koFA#L%t*msn&TC;+2Wpa}^u>()n-(QMrh<@lV417`L z6BYq36K}K}%BVWD43S?DVMUPh&9vg3qsL{10=~G4Df3?Q3BTtZ6wO~i9=vrzY(y99gYJT+JRyhbe()Jpy$BBf&e-*|x9e`Y1A?&PIP zcPZoGlNWW5t8!c<)D5L5P0R`txj{G^fI?o)krt34Bm~Z_6&kYwxL_cf_CR+g<^+Av zZ!JL~=`mW@t>^Ebrz7fnL<*XnPoVQ6;tKq_lGKB!E8fhDi13lH6YpZ`O5@zspS22H$gH-_ft+4e?tA9)H9Uxa>-K8H<6u{+?Ug!enQ)$l}2h= z%8sa&x^2rh1yWZi1&N=n^cn8inK$nB77xcLtzR*I&b{BfF!@%vwv; zz26%0r$OH;GsDCh-ZsQt8wP$u1c`+P zX_aCh-&Wmaez14o?K5)IdG`DPHZy6?OGsn0HSuJzGGXYWs{Ywmqa#oF8%wZ1^W-Km zI-LCHpUE5geVLt3x-448{1701H>HU~i975)yS&FDr{C%lD~}dbPV`SByO1R}_mbf1 zA5n{Ntz*^$nZvCh2slD2n@l#)t>|Fg?BJQ`4JMUnZ585eFd);VyAUBMPh{jrq^KL6f+w$S zmWm6KvXm$HQ29s3d;))G8e*2zptEk=8CTipKV~YC+45XQY2FpMyeDPO!>R%$Q9elo z@?c{P#Row}0IQN!j8@b$2%L$xwaad)a$l>jH0dpMSr#Q$Yw%?jUvJz*(X%#vJ{c;^ zx=r0<-{KEqd(s*_2KH_vZnmYXBUcjGd~y6JVea|N5P0#IV)8pQ&kFEFvjb7| z->#csf}j^iA+~5JV%qb9@S%BUk`(qH%~jJ^=}?^L?Tz%7?ifrE41fEzN=tobA6_yQqeBp8F-!Lawy3VzcB=ga^^Tg!TuNH}$R?y6q>i|JcvX5v8C z&g4W0n~lypvq|Kikk_Vm4dyAO%d9=1V~|H=oM2^QHIN#rwVjm};q=}Z3r%aC zSc88JO_#sDpyW=OzXrw;6CLJ1DN4F&0|8)m$z8=QaTvKcA0PG!yJu0aX03yNyG`5%a04=YW6p1 zutl4D;TTtgEn3ev1P$f)XAmCo-x6h}vTglwY?n;|0ez;L>WPI$d`*Z*-L+@@z4jx;R3N`P|uQd2J97Dy#~;c)`14iJ9g zM>Te1`ka5%xnJ7__=(Wi?yD=rF~&p4y|wst&$_$K$UXb=O?cZS4NeV4sm(bF0;+^a z#(I^7krF*_>*{lJp9F}g(CI|1B8a~Tco{d(q0yZC1aSEo&I^g|V0xYrU)*<(;Ly4= zU9`UYcRR;fnh@|#s4Q|LC`dS!BAbDHD`-L(hG;Pk5(vfkX%4Pnra-)k5#wOCdWtbG zYK?@MP>E2UY&y>CgrfmMho=UD)SBRPM5oBzW}-lMm)apZO$ZY@MXr%NsGrY1+?C8- zGT+;jnR-xrF7F>6mdPKxqO|+*GR8nxzAok#@d2nvOFdo__m#Db6MrmgBi9os2W54$ zXSG(Z{JJUU!d>&l_X+kMB?Yza55pEidSsUL>py5v|CF<##fYj0fcsxHN0K8?Y-YUp zIS#9AWhn;Y;pGcp;;=%P2t-RvY6v5$2rq*Q&kW3{ra~1_3Av+vp6=K)IE8OVDJ=$K)X~ScD`EvSGq3SR ziUn^PT?uBLz89ybIn(O>Lrs|#fC6^`x8GNY(ENFvv>8v_oPtJ?t~+Ar>10YWp7=F`yq*5hU8w<<uCC;Gdet&_#OuGD!Ie9}YbX;M;zz0)uLftRBy*_v#YBki>Zp#)ptV~uGjY2PTQmC8P!oExv2)-F~N*inrV4bl?1LA();{an&JQyfZKrm zPxm*#pXh%3=C~VW@UlP}yz8k)&y`A&nUM&qP0(7XFS8}^3>IN@gVbi5e_A-V^NN4r z)>N6S3_c1mZMplt*b1$>eH-^s<<|=G>~+9V%($WnRS1L9NtYq)!2NxeElJ> zkKI~4-dU#%yRL;^f1)2^h^An?S!MMm*K(bh+}hE2qV?-H-y91e#Jh+O?{}PSkN-dT z=0bn@TtAtWz$wsJKBhz`xHK^y;|24lRmIl;d5D$tfYHLc zGO4SRw&gnjRj$!cdSCrBpRpR@ZaZ^P%6~gq@>^8pQhB>%k*CZ{8@=MawpiSZ8f@4r z+gZ*->-L)+-n3qQ`$T`Z!}@g1dj50c_jqsySBFeq7|Z44RD@HWV8-bL>?e$9Bmy=6 zRZm!QP>%oQW*MhXh;;xVxX|NIJtfyv?RX;e^oQS?MU9?lZ=;k zOQd{^e=UO@G>A^lmqr_f-AQUW#D9`FnNKH_JbTf0t!~-4?x!BlAK~>c52}8H7dta4 z7G9w!!pH_!jC+Vjo1ngkZ(#x|6JCB?NM*t0Fj2qvNR=ucARZjhA=XpH3nZV)61uDU zZb2hYp~Oafe|1u@Y|rZs)9ORwSbWO_GtR#=;rta|=a>`kXUffGi!-Q6IC!@{=~c5o zm38ZkfK^U1?`b)Yg958JWV$t8b)q|FTqc$u7>_z_taz7D-55VR`!s0wc<(Lu!MWF- z3z1lOg|lBJ_xJj%mcy5qF~-(wD8}{^cVC%96Do8(i(69@hVUI=k!3}6VpGSy45%XR zA?eeHSE3=ZW>)PGvg8dTf6AWiAQ)YI7o3P2ib&UN)>YOFq_r3^^0V-EoAf?825Nzv z17#ZdlAkTi_7te^Vh>bC!Lx$I+ky;XXq>#M*ef2n`j5yrgZ`Ppr6MQoKz9oDc;ZLa znkidm(Fv*n%l)!BV>$CMZy?&H*yPBd)}ANkDv5n>QzBn5dE_35mtCq3_#T%^kn_`u zba63QJneBh!&KQNMDnD{TkMPZ)2<7)AFIcS3*X59JPt+h(|+W?8jiNXb#VCI1$6tM zG>UI^zcIpYyEt+IFJV&7GgaM{i_o+)4~PTKTwN^7REVwK-izYZ$XCrNe&JrVqE@6U zIr$~&v%%{)P>luC^G!kkEYw~6?r6^&T}n$@*d|k(p2_leKy~_(eFK!`t8*I(Ghg{C zX$I$|4*wD5Th@m^I!lc;8bbv1{;1oPI4fW^yNcZ#2hUbU#VjYCA7GpJpE`aJ_|y`N z117yrV(%U$k3-O(Bv7-A7?xhmDX8pwzT+|3H+BT5^wWqfuKpYhpdQH0uVE>FiVJ!0 zbyL^oK3dm&c*+}>Pg`4l?}W0Z1&BIsLXC{CHrkHTDQ^fWOSs+6!hj_*%i4{LcX63< z-(I6bZTLZC$2Wb6glioXeD~9WqeG><+}cQl;o@-~o}+}&K>8WCc9L^#%IsbjUtQpc zWb~rVAvv}i6svQN>hL^r@Hzm77RkXJG>S^OH&Z-6?J|3A2Z zVil*-Vy2MNhZe4nr11w2xLH-;q5PTNz>sa3H;DbVNL%1L%d$4=FP2Zc@ks!Yeq!QJ zRj)g2K0Tv*7W!>556vrFL2T<*W?d@w>_k1treAiwsn_xf6o)T2V{bP}ICCs}b-mDn z{O5}kSm@>dhXz!|7mbCU7Q8H8#MI;?Pi4M(So7M%8+YNwR~b-(qqXCMoN}=p_j$1+ z`~g4LoktqaG!9j%%~btvKxm~};O}n$_Psr?+yJOPptiV@XX{#K z7@>wD?ofUNFRS(YL#X7QV*42SO>J0M8GTn9W78iDFVTygQpIt!O+rsjj6LYz{S=Qz zF)^=waqVuy=929i7N+6-VUm(b4F@jc4o?hq_9e`nvVOZb5#ZimVkm2H_DF3=p8IMJ zgf*s)%4!nVb-hU%11F(t$8;%7z$l#wmL_1M{m6C(2&YD8F`%Y(Ft(FW%s4(Kt}Z0UsX3rbRm&ayog?_`xLS9s!JodcQ4(`*EV7YqyV7R= zq`Xb}V5RtJc9PEHir2~fDJ{T&ZJBi_(PJo)h0WJYNJ8(|Oh^h3>l=CCdd=hp9i9*WU3UNJw$dqoMdYP{w@A0AqC6l0B=p?=KnPnRY|!ADyT5F?MX4&EtZb zW?pZC(5Ym9mCAhOco0sanKdR*hSVf7ia5cQ{ZeIP`8cgM^ub^VDWj1SgttFW*Lu>% zDRb2)=X7Ogz&@m<5#InpA$PX=Q*Wjzp!PBz-Zp8*^N!KApl1E3LZyd38M215E(PAbYpKgDGft>GcB)xH1y{%6Y0@3xX>i zFRFq2G1d{VT^KG3dK&!mLaIn;N`IP9znFxHQ{dO`*1YvIUVGJ!Na5g{jWSDR* z{9^14FM@f91Pa~`ajgaB9uZzw9+NlovRX_d8oR-b1sAuaCkop#ZFAh}Z}VCPbLekr zHie(kY>&=J6&5!frP)?Im_%Ow)gIul3&&*eZ`aOs(EUJK@)%os6}a<5nMe*m|4=4M z9T8zWA}vNY3~2>7Q!tYjMbCA{G6-hh+TY0Ng~9HVA#vdSkQA6ft=n`*gi1UQRTFY= zH7Y7e_TePLct-JR1~jK0iT9b?J6G4xN*YI;uQyAP$mFo1YB_|){ISQ5l$eEFs-*r$ zm%+QAR1)@rQOA^rIw8%MTqW-0ykC16%}2x~q&rgn6<4G;=1l?gJjYA2bT8}hxV5_qG`P>8MM%&(9gRxmM9+ks`Z5-eYk7)ot z0iNFlFS~pwvqI=w<;}^iSRPprko0t3ITS@&>Y`FP+nge0|9;$hkfSyAj63lY1c~L9 zRE1cHaH8WyZwg4T_osIzg$s30Xc@opUZsq`@o!cuxju%L6QFmWVZz3AE5!|FsXoB* zrJfqKu$|GnQ|%AQI9QBkhMp>PCG&b31m8Fvo1H_Yg?yGP7h8H<>zvsBTKuQiPw5OM z(iqs*{F*&i$ptXqJFPY9kKbFVxw!G1VW@XlrNu3#nCE+S>JStyBIhuJMaUr4aTDM1 zD$5E^)!ydpYceW~#_)Rp<$XKpBb16Q(@@ZAdl8@<+DS-2OPN;>D)uHlH}RG(8p2K_ zvUz<`u>yN<`i#W5FFp&VHkv-uF{5yr`8dU&dyT)M&=)^bC=ulmg@%5mP!R$XF}lk9 zS1|&3Ad&8^{Qe9o`%`C(=2~L#&-(D`+eFVl$H!;J1mQUYt$f`_M-+;aD^?l$Z-&&baO!5*l)3f!#F9zfutOaA9Mr?@>IHfyBX^WuRCusz};+@6#KyKr|jU zKF3Y<|BK|EZpOXkB=cx#5V?Pv8s&*ZBL$FP!IT3Gk&W>mMo0d`$d_qWoo8GaMv5`8 zvPdI^P+1o%jBWK#G`xdG8i&aX(Q&o2AEFOi)rgR)kZN(fn!HHIuhW84`n zpX#%HzxQ|VeLU`?M}L5MpVv9h=Q+>wJiHbOQxJ`|pbwt%ys=zS=bF*eQTrzg`eUhZ zTtkJ1H}18S>HiToC60Tff(}^^$gb&&g6#q*<|W%BcMl-b>>ay|?OSK+~nkqg+Q)uV4*5X-y} ze{{b)Wdzk3Atw#5b93x?#*)Bn`Er8*Q6LYMf{)#Eoc0f9kUvhgx2a{Mm3%r`G2C6V zKrI{a^1I~jTllKPrcJMpzIf=TDeSk2rzLGg+XGZ!|fma%KKCM?3%APyz-4JwSE86^XTTv-z_P^Ew;)xKQEQt6Z_(o z3Si$3x~`CJ2Es`9)_i@w4ZFP!?(Y9F{1p!u8S-yNJ98>p? z7`R0SFxj9deJvt&jWpoHTGJ26MMyaPghC2vuZmy+`2ZCxYAR4ti!mx3q$Dt_jAiqC zbzeg~b~g5Z-Z3y?Ds<7mJd*y9XT&KoFXr#(jgVc#xQZd0Vq@-K#f|^JX8FSUO3REC zx#N-`M3q%ZN;ak09rxujc$wjT@;I~A;K(+1Jmx0L((?NjkNw8Ickgf4v>6T(zC2w& zSb^+4Wk|+`*jNxN<9q#JufDZ>93LO)(Wn1zM3c^lk^hW2#Gj!La7j-pVm)0s&c9`{ zGQ9hCPZWoTHNN9vj6y=BpBx4cXB&;$AR=S^oOZ1Obkm`XX{a`c5J);x0fsNh;%U03 zp@0gbSlo2#I%ZZ zZnurk2b1vG+zpB{c?##)N1sB^DR;}kF6Z1erRMB{w_vF;YwbiYpn{hgjV_>O<)Vmh z52_^&J}XUwnZBe8i77~o2YG^c+y*~{7W26^5EkC$xCE6$F22_LsM8rvn3;96S1P+B zgz#6BP7p58YkmkXAo8s~;9mpXq#X`krP#3i7U+Zyc+aO6i3bPA+5)%vMXiw;tS#w9U(wu@U+1b}64 z5sy$j_h=Qlef6y;H7lCDE}s{l6(0n;93>Cqk3e^`z3zXP<$Iu?Z`OAc`rOua(HqKc z)om`ku6)@fa)xt$*^a4-K!9 zu|K1QZmdVT^>CWv6=AxpKw9V7Oy_RljdrNex9(Sf8Req$ILr$()sCZuliz8fQ9_C{ zfQgEXP;2JTfiT{worzYPaPW;hr?k@-b!!hmL>#L?PeVlNGAA8|#y>|u!S@#0scuh9 zfui)%tfow*L%6BTkm=O<$#!myK{dyyal2)(Dc@<(I^*!Pjl>Ft&G)ye#Rc5bIXewz3H zk!@Eef-5-8cb`FLS{mQq+3(>Ct}<}^wO0)Q!GOjOH9=^%_fcX>i8|?Dk)qIqZKNnC zZNOK*1+y;wOQ!@0tq-JA#8?*e15;(9T8oac?#ldUK7uR&dbYAaGh0fg;!ICf;RW8y z*{Z_zS0`bqSN__WIey_uVD0F4DC5f9^{rGL-;^$2)M)c*F=4YJL{2=l+-7zIkfI+R z?PG6_H=--F;}ci+Xj4a=hKm!gxJsqQW!k(cInpA+c3NmCu}tVK(G&M=?J>ap3;X4i zEY!4GFxNkfC%vKS*s}HYpGy_Pz2vM}RBD9u8Mr~Cm-oK2^t-ci7|ls45xj%T?J((# zvtP^7*g;vEG_t2wV$6gbm;kY~4H^4MH*!q-^W;hOQ-Rp5jC>dyybT>#+xYwaXeW%? z{YfWuHQ|(pPri>r9b(k{^R8|A5B&aDe%a>(iH>8ReD}*fqe63Fu2W>^cH5}KgJ`cb z}HRIc}$*0cF6rsjr`Yt=a*eR+m>ec9bjrvO@0Fq z&jb(2J5QCidbhS+E6}-ZQXRfI;cmA@9|KOHl5n`VQ>^B_&1@!%G`chQIG&2NwQ=X} zaCjyKYvc>R%B3K-FSP|6-3``l+ALiZH_Yt{6cp+SfE)$>A~vB`Ci2%T5ePRC#Gvn{XkJF zSwZ|+CvsvSJOz{SH}cFT<%wPXO8}8}o&B%vRh`la`mGb~_u41p5!{+Kg%%)CSW!_D z>GR^&MkTV*gra6a+r*mDML__G3b80dt~4y*0K{wCuY$IAP}_247cbc-1H7k7p~dLu zh?>z5y$mG_8*X9vbO7X${I6FdKtB(s&{Tg>vd_Z7Vr1v6zJ^u+MtPm#!UHb+eO0aL z?uWvrJt>J5A1=m9I4zDPuI0$3pYbKT);*9j5OX$0a5E_CgG}$JLjLn(ZeN?Yc;Aadcb+EZlO)m=KUkW0EuuzL-y>2Yw|Cc~1OY_eZmv2g z?P!Ql%LjV9auq^o395};dRbHjdFI&KQ%I@mnmN*}#Ic0=_OV50Hp}OZT}Q#wYsJh* zk55GXNVP^G9>osMYWXmm=R+pbOH-IOjvNZ;jbrnm*^tA}zZSbXZ;`oZRJxi67h3|Q~^@b-h@EX)jb;nGf^@EQ!o)8R^uT7&OUy(A_YIsk!9}sUQWTKLh_jQr{7Z{&9AA@_y1QU zKKk9AVK#M@exl(U2#T5IA6<{~yL%BV1t6*y+rJ&*ux;_(kGpRvzWJ$f-)H{|4`%!Q zBo@n~Za-FrnrlwJoeGh0Yik`G0a=}v{BqZyvD5=sG1yHu>SzB&6$m@ez~n;6TQCVH z5&0_Z$~2j}CCz#M$aS}=^KY2Yo^0S%1BSkA93NI2nO$IW(6OeIL>+v7+VXJ z=_ma#<;RqDVwws7NI!%eqaj8@m5J(Sf@BF7;^iqQgAO~kVAgTT0Zo8E-bjLgwV+YH z2`VmyrmQJUBHq?dG}&m$zp=w*(}+Tg2PE=L+%3B zHUG_x#y_4l-Y&%Ciu0UxNR75m@snUJV$-57J=NO|;vLMK92Za!6)p|>ZfcJDg9pxd ze+Bof8=<9TG|Q-`baI4h&lr>6#Xn^#a}_Itz9oC!#r=d#pqr;9WW9^4?1m6Ew(Ul- z4A8IU)H!jU>6gwp@M9JEZ~O!gT9&enZ-FTS-dc%o8h(Jx{c6ww5aq<0cCBULFW|ih z!vRtCaKXb;00%06AX|mlQpWtjq~LCth&GOPazM8%{{{~Ge_P2fcn`IPWq z0$Ay7@@;@pB?C^>KtCy8Cl+MnNW7@S>_N0(r z4xIvEMv@`ZB6%(g3b&GeA0iS`mjQ#L__KnTMyKCX!7)bM!H`{toak+g_>|}Qai^Rp zS;D&hSk*Qw)B9UvQ$3K|jguX0t`1#OAv=510Zgz24vPR0$|`dY(B4!Bgl}>|Nz-c4 zHI(JvN9zD>0J}b$YEefpsaQMK%A0IX7dP$mAFY=-|CNT{R^WCVZ#-jIEDjr0^k|W( zaUcmlFT;fdAK>KpklKwX3F|Y|0V` z<#mx?G)kXVLL2%*Xjn5^57Fm4G0wjU(-_G5#WAgug(dpYZ$+rvi2bV`G z@yBZL+a#6weF^07G>jzbeT$)TIMJ=bss*jsi>ZP)yCbJ$3YE^&9MP?wgyvkX&EmF! zAWG$QlJn?Jqo{NM^+4w3Q85vP$0|)^W>1EQffmqetIW=7L>UzB#qXkjcNQiW0v{F= zkXoGRf-_dqVU~2vx|2FV^O!kGlc$C|@pf^M?1( z+{CSEf1}pyi_?b?$+Zv$nE#TOChJLHiN%|t(K0`rgqd28Kvsj8^7hQP1Jk|X$w?cVfv9w-?Wa*@EA^wVeD=Lq zu?qEtH-HEk%1s5PH7XA^nA7s;hQZuRVf8Ea$08-WPCqhebex*Ltk)cI+_iDG<2i&D zHa1)xx*n0tz!&*sN8md5>Ic5d2?QAoh?bbn_l=@6N)A2xF89ICuGszLii7LB+sq3pk}7(%`?Fx?f! zBa}v$sAbX-cT8Ui^Vv36EvN<<-ZfqQa0y`XF2dYHDYtqM($Yxk(jwDyHBJ+~NI<>C zhdo-1IWDuzP}a-Ob2M)14b-YTo-OF=WLb#`{KJjYk$Klj-1hr(g!;pi#qDoouq}s{ zAsP6nTii5{Aheo;X_MWIb&dL;!Ct1EFs`pac7@2$-hI{MZz!EbJWJ1QrBi}dt@>s+PpZk!zd}Q*aXds&Md3=dP%}&1YVWw}~2pAE_&O6+HpYZ32y#}BQ z$m9P1^ecPge&s*U^TNESBFLcb39OZFr5URSxdMem;qA8kP^fb-_~)R~%ExGVjdpN|wmWfd@b+ zy{(LNxn^v}CIRA6X$_8Dqd#Y_QZ*jNLCUFzsOui!Lty#@Tt;j0Vu88vGm0BO#qu&> zW*46&GlU&el)aKLlAwUE0%wDwp$7_^YIz!I}ygDA@a%i0qM4(Rr`hr8;pMQ1Nak+!gr4P-naM??Ne<0 z*3Uam)=11@+zQ{8cX_RHkQvpz&UA?%WJZgPc}NL+sZHBug~t$vG?6PtO=KSc_WSLZJ)*WxFP0$5$-k6XH| zD?|A}d@9g^EU=WwViwy6ZlNp_1+a)um=Kp##1GiU{xuIs){fc& z-f|@z7k{eJa2yA`8rNTvsBvtAXh=fsx*^XSk;5#{TabjNu-1L{8xCykqbc$+tq+I#{8HR602V?+>JNJse{Ymkc3AR&f+9XZ0>M)N|>(7@W*jJds>t82x)s z`j6vM2R-QnDuBM7@oToD8%7y`{fHCBY=#}O532h)J?fmcIDB<49GFo66EluYtKg4I zF<+#S2tx7TuLUYr1xI!(0*Y~}sYT+81b|Mu)_Bim&QP(MSJbj<&2mS9#jCu%0$dDN z+0VH6dt7^Tv(RC6@fid;bGFn2M+J`Q&OZrMBl*$L@0H;A9>XaEhUfcINZTbrh&%HD z2|)J0U9l8r<~nyLAJaa~Odc@F4w_C=8vd_Y{ zwuvKTu*7WVY5=2NsFbYk2Oqf;q}%}9jgi|^bNqnN9-j4e#dFzfvglqaEBpEk?G}fV zAE_<~p7;Q>ykk8VN%M>(g9U%C8)BWc(CY(B(U^z8o)*me1TfpQL8!TQ+nD%LkhH$x z>|}Tbi_B-Co=}`Cpd-t*^v!rtAn301=<8Hzbf~bRh9ORdb!wLb$31(5{15=GUHK_W z!sm}$Ss}}hpDT9RRZ@Y#h^j#)mC}jV8a> zroCT&n<`lc`9bq!_jJmUed&6(0eklzFL%}^unv1>5X`=VAC;c;M0nqt40D2>sCXxR zr-N)&Qq%!)D&~EkKFqq`?+fAb$c*a#N8OE^*Pl+kDqig_-}|T0=6@T6|8}fB$LkSW z{iK50qmRZh;!$oIYL{I|%;T5^&zZ+^3f?sj6go(ncBEMe;1QF&=sjZrt|YJ6`A%pv z=CEb3+=oKTVA=VRxj-RO=ZG|qkgmg-aCf{hQsbHrL^^WkVj`l!xGg&(B$zh=z~C7y zV$nzAf((|CHKJY630g=7gSf9 z&_%JU_mLY*w}Y|RizdiDkXm^wg5*qAOBY?(boCdd`(}kV_Go?r+ax&55{`??e>>&z zpEJ{Ryw8DZ75*mXe|kpsKi*XOW74TkPqM?H@G|L(57DKByT*N>{X*=vZS~Zf@=L@I zKAl1}^;KPCz`i4N>sD*30R&M$+~@b&ij{?mM)?2;CGBz}WF|1>yM|=J zKQO920Ov{%hn-w^(~VT%A|eNr*%Ni6RhWabO;n}sP2Z?EUpV&t&W^zFJ%a#&JWZnq zOxn~4fZ(Xf7Uq>ja@R9@K0BwOt)!IDD65ZY_p_?TtbQ+eQ9$&!TtkOepDgnM6L6^g zLZU(sF<7wHp$=|ogf!M^8h0Pqm+ojDZGUhK_gzmu#tU-sCQhF@U#627Z?USICk>o-IJd!)ZpMx6pi_)h*EDj_VxV*C$r1)NDu*-uxkqH zQxm^smn1G&ZiJw5WLpyDE3dTbE^-c%8Xoc2AwcVeI&q9vk?djCHZ4WQ^IYF^VbUfXr=bxJ zwr@V+*|0i2lSjlTK?)ouP7RcsEY#jDY{_m4+uzKmg-ejC99Ac~?q}pf3@WGfJ@SlL z{#UgD2=D~5{$X~cIVl4$Fzgn8b@o1dz0o;>`$8l>{Rp7w8Mr7IB{J?@oK%X(l@$q+ zN%|PDG9{yRq+|X-Hvg)as8I$I%qtnCntE!0uOCbtgu~m;Q%yv=k74 z!NRbj3#mu2T!Q&w@nn|YrcHN4tW%FPi*PKFR+!wK&jb$^EkRg%cH2mi_H{y~qzgoN zoQ}&pkL=uf3$nO&6CWr4x>AB4KNkO!s$ja^_nYEX*$=cI%t|_VK-pBeFNo*#;M)#o zy0mhh0$Vax3s7qK~U|`q^+X+LHl5OBNHse2?jj3$^IUhToe&NAqfDu!kwY8?a!8g z-35sKzwN8a7U7W zX9nI(*owNGq{a;1Pe-%PJ@XedJ z5SnsKG>rH0uGf2LQuKz!jDz$|bg*Q2!bc4UE^S`ZTkJ}Mnh%04`c$^Mx(Ce=+=s0) zydE(zaR3}4R+&fj=+zXzJi8W|6s}rhWg@|}I0Yr$TO-bOQVmL_!ojsgzfAXFXnqQI zvNO0D^Wg9jRL-w!gsKXgv4n%)yJyReM!QCoHOB|hKYrr#qu#mg_U0$%5|N*^1Z{@0 zSR4p@%TG;1?He6@>g~5zmeg*G;yh0+sL57 z;}*7>Wq3B-5DqCmD7#NXZh{_!trO2CTkYSN7;DwbtI^$bMD$YY%+*(I=sk&*Xs+_9N)&bK49s zu8{M!B8SC+)rm`u9JdG9Z*GHITTVDl`x$P{ekQ~CV^4e4?@GQ6fxN6=gW3>)93YP5 z?7HlLm9q=|IZE-@g9>8T>NJ0^A3sZq{z7{&-R=R1UjKF^>o8fPYiQ2b`k}Y>y?%t| z;%JEsedDta$SAN9v(LZh`-xs>EW+8NjD-0CMTB}kUNuce?7XWFgBKYqF9D_wXE;Xep4(`R|_ z8v*jBZwh!=2>M4I?YfVUmp5P9c^gfnhEShm+nJDyUR9OleW3H<;8ck+=HOJx86P4fTpwi#Bkly3hQp$Ny@m#e|{5O`$SIY4*v0T_Wr+r{N`R?@Q;7Zal|m5r_GWf z$YXDMJa{@>_oJLeQt-z_t|CjSyIUzFDm1PW)_ed0rN@}icqoS2m}6@p_W{C1_x-)V zhKG5<5dqyF-6Gox^AzIU3zA$CUC@ET;JvlL#MpRV8c}lLo;Og^5yhgr;g^Gh7xha( zlf?JdUo6H!;pp|w($AS>&&w_LjFum-P!yLuy0PO>QrRuiTcSxa?sTl;lQ5*KB)G_L zpt|2e=+o7fy7e{|{K{QHOW2|TjzLB7)!QdT+BcL8EgGa^v^MeGOzLt7NgD95E8$T&r&DjsMl$&$-jyT*Ai^WA}hT&%ld~_CFGx{|; zhwSLTMJFd>|95irPtgg8sNkaWvYA=*Yhfm1b71&5bpXLjViHljxiiOM6H@%n`XG;} zS@g+|4=uz`7$%yhjqv&*0cy(>y21K5Ul;L0KG(~5jk@Pn$80*e<00c9@Jq{g){O+x z1@{a-4O#g{bGfJ6#fb-oX@)3m5NC%L)qKdlIzoUD&c_ny+jSiR0SP9jik-)x;DLse z+>pJF*w`3zOLe;z8uNK;svUXljMRk?JGB-xiaPfBg6eB%GX~X3HEhE7sY=%9i*MnO zK%lK^3ud~s`b2K;ScRDI)s=de0-MRlF?;KsY(v+-So@S7;3$mar|M=kfl1hQBi??e z5d~^n9;)qEgh<=9LAIYWINmoCST2W%oKX>Lhut{F97ix=s?36I7*O?_?PwH(9|s4C z_e}ikDaZ(90}Qx-07>(A0UVr&l#32On}Jh9c~khQv(5+6QGx2L#7v)surl!4bvYbG z1hUFdl8=bZ;3-YU(-Uh33Q)lA#W@x6DQJictq6t`$iE!LY>vfDDVr-8!ZWZtYhd|n z{!j&Mg6d|GvLA`(-6I!Y<+KLuuQ-2NjGeq*#I={|yg3~XVURY#MlY;AWH}B~YjpS= zQ_v}AZ;+y~%v;c@Z%?0o!uaFE7bSGGc$HhYUR>=mK!2mdu>et^|LlAD6&r_kekmr3Y$q>RU^=*Fw=OueCarmsW0H1>pXak?{-V#}ZLr+h+* z!x4di{9odTKoFb)Bdw7oA42_e6HX7s4#aTds$C(q2F*hR&Eu2{S9my#@Y~bfl}F2s zm(V3$GPktsH6$xpvTmJGCtMa}zEWG4XpoArt=Q zgKO-Kyc#xK*aVM&O55+2Rm^s)3GaUnJnZ!1Ws8LC*7GYHR)YluuCl2~`IAHCroAiO z4H=W8l}_1`rTV$nV@)TuDLt9o9D*7nd~;MV73&dnF^2c;HOeRDYKX8quf|Sn`oT#? z-4?M;jheBfO%LVR*b(WAAK7+ykMDcle;|j&4DZiim7o}cdFWXhy}ML!3{y_pJzh>+ z@$;_#n&aDDS{#UtjqDRNjQyvoYdPKU$S+t#cVjf@(D-ymFkjMP=Fr2<8d*&C{dN8L z4y;e0c}c-hg`d^TBme>k2UPz0V^Hrad`s@c6JR4MWJehOV-$)~i^}tPN8RYIm^~-& z4e4$5HP?Gu^ZW1KD)JH0N>2T`~83s%l)3meHF+-Nd4?q=SyA zj=>=gu@b#d^S&P)QSnPczjs9KFS_DBh7bXPMhW9%J$Z4?(=U^@(r0^PT1LOr-gB}j zdng9aV{Jgq=Vij<_MArxpS~?M?KQCaD>Foep^li&(vrsdCGR*`am_O8E$hWM0tTIF zsm2>i<9yhtM1xZpMSY+jjCLw(+zEsewt1D|RTNzd3_1G4#7mgnldm-qM0i$6JW7ld zCS=|ifXQMwGTA_n=?xJyv)jEe8dW!AUs6Z(SwENiBwPT?bvI=nc$jOJGEg_A#3Wvm zkm1f(v4#758|V)BmsG0`2ruw_r5i&0toGTJ3}T`)YOJBmQaH$N!)4iJ_$)eU)pJ3GD(tSx|-={ zW5_nU=@JI>PbPEut8V7JD!x`3AG>%rcc_eEpJJH|acsDf=|$G>eQOw~=i67}yC(31 z9VR$l;*i*ux@@y*INK8wgMIg0k7Me&*hyC+wyPT!gSnToCwA^tUw&ME^Iz7H9hN@! zXI~^P*4ltQK9mlV_H=JhSQ&1j)ggo$oIxxVz3@0x<~0f37|##fCP30-*1LJ8Z_AbE z&iQNwZxaKnhr80{;TU6OA2|$w@NIVE)Apr1HJ|P#yrVHiwY@CO<`0*^p+q?={%QH6 zfT7 zmW+}RW&4{7Rjq5yP%-@LgoV4+hm;}-KVi8Tv`yrqwzG<-GUq4Wb>poD<@X4JND*W) zyvh$>6LQcMe4Vf`M-u=ZSAU+ccsNgn^H^-!tzJPGc}{yNcm1MXOl)W%Ug#%(B5s!i8CO}W(n|FR0^vtVd4+} zS%q4MSD`-QPY0((NjZi{QDN!J;34bE;fQhRUEUvIt2&ybKp~f$%fUyqzrznbPm0g-uqBQ zrzeH!G1v2h+aQG~-pgt7zLz(4R&)BZkB8DXIm37d^M&H1Zjn{ns1hrciH#$OolHhl6U6)2tI_?imDbBZ69HhHfCtc)p@x8E2N6VnnQhEdm_-MxJ>O>Jjk zs3q=zmpi?!0zeS@x;X9&f%!6qIlVg7{bW82@I#-je-!2=3i~i|-6K!an}Ym|r~~4{ z%!J^D30Vg-2jBNy<`WPv#0lE@fol`?qBSJo3eK%Ub%T?o<@4nw!gnZWB!#NQGSsjJa%V z^%~{{MCJ-#Wftg{I}+`tWg`cmlrJz!ZS^KvP(J+_Mt#S+6Y8a6{V_4THE|Mu!$Vzba1HGz4*mOA4ONPZ20 zRZwoTgHN%2m_U+oVOcY?Z5BONt0OP9!PulQKMAGCQjym^{P>wm*ZhO$QWrTMG5p%n zB?c@2o4d%aA9ktA3+Rhtz?7R^^+%5S~_4Bmt+U|oI-p3RpFP}EQjXb{Ir8O`|m0o zn`<$KlAc!$REYut2Yp^T@*VaSIH?k&Os=(Y)01|J3?V00qG zr*5UZsjp{DaIu z#aG0%r;)@d??p0T!#tKqo8Om64;jK-xXYuSc0MKx(^(~OC09dX@$N32_3L7mRP6dT zUGdxS8|_kis`RF>Jofu3g?qI#TfD7i@-Drrv2UxU4Es=ORJ+<*yG=yzsDUIvAc2Y9 zn?L!yRW$X4`ii^`W3u%H6W-SQ#JY)C08^+am+nv^tpW0@+q1{?3%&cRnT zAynI5lc(2^)#Oq86Dsc{6<)k}_-@1VqWd6kUaG#%Zc;1k={r-K{pM#MBo;je+Mi(V zCtYyfl4*cu_w1l#C(33ggai2?b;bYlx3AbQA^@B5i>gSopXecf8$7oqx z`!S9dmObeVVK;+5Mj*Jd^CY?$VBwG4Qvb*%ENb1dMPwcb=O zGSW0CfCdmSKSe$MSYiB?5ctMb>L^J9)k68WJ0ZDH_}DseRved2Xf7NCumN&aqc-{1 zh{or9iz~CF3yMw&SPxLas}_B;inN2LN}ggJ^o*jgBRvp%(cYy3AF{ig3GfEvHptr= zYO5-cH^W%AICgNE6qYins4?KJpI!GMdWk4f1zTZPY0;?vF?u@5`xC@mg4m$q>?3Bj zM#YI!rHN*sFqGP&NJ(D1zD1Jl)2ohp`SA>viDL0qBcCHWZxDe1j-#%mDjW#zJ9lce zY?LbGVNOisPbJ>keSUEwhD7bmy7YDIth z?J?IU#-UPPE*2=T5+x(Y%f^)^BD;~TN0+x5_=K*N==ikfk9{S(xU-GY9oJV?(hq!};@je`>kOe>S^MP>)j5}?{6PjL!NC%`eHA0q|rr9LJv zJhtWyY{o=>go3Yi52rBZZ3a@=U?+)zmJ{7agOmvX^9+{`9TOd|BGUOdW2p=CFu5_~ z{0z5v?kuJQd7=+ibU|MARL>Lymf`>^K7BaqZb1-l)vjXc`8&Z}#Wl^;rln;=iwnW_ zVb@DoyiRtH2>VNUWE=I!3S~#^+gYfGeZ8V(0(oAG0BDbjwnl#WRQEnF)V^W5jeq5d zc!h^ZvcSvNBEWWo&?o#I#MKsY5D0}NY+=0BO2SuyYWrvnIadQ}x9XFlRcFnPCr&TO zjXHUAOe~D)37QA^RVKuMi&CGz&C7o1vgsB>dcJRY88{ZgYDImJmwD+Q^eu?i`lB-!%u_2#cC_#$^V7?%NOI1 z0)Td(H|)B0Xl;|q+5}PA{>jI~sjJLt7L&v4!(A~fV@H>Fid#LBm@I{b>8rfK%I7Lq zLh8M}iOY2Qu&POOD>!QR9Mr#9$^X4BmTK!|UyIy+`jIy}v4~@DbYd>8@A)0} zlsz&$0d3kL$n;(0oCO0lfG)eE%AVMJZT%f0Bo%;JBX<0-Nt7ALL;2Li9m~o+B@9ZFEh0aV0Ti%gb({&qi@JC^H(L##P$LPvP1hV&#U9YeZX6kRsw8s@Hs2 zU;czr)Ru2~rB%%m_<}J)Vf%+pvH~~E!Z(DfK>|WjI$D8R?7kz7Y@P9?=$7)kRD;yPO>&3V@SEvV;xXnEl@On<08>{Qig1M zab}%pj=^UIrP$5yJDwx6D&KcJ-Df{mc62k40QYyhIsDhzt_^4saxuO~ymg19D}2Bz z|5ou?aFONp_Z6Q6$??778y!@9HFQ}{cl1$Q#b;W}xfm~M+ID8A7eBsNXsAl8%{pUvteu*rZkMF=!Ptw)*Gca4*OKY#vHTGb^s-3;dz^4A;+ zr{547oj<~+GT{k-y%FtAyaPfA18#L!+~vrxa`{6{T$vPjMzf**%c&%YF!s#Z^c!&W zfjUE!I*6(wcS???VKzb@j_#kK@&c&M!mY!HhoS&8bbFjEW-VTULbaYg?(%vachJL| z@za`GX~C~yc36kb2QoIK#`iTxwWCAr;-&)9L_|-1ym63_ylXbwGWl4UIl>$XK!7N4 zVlgs>gS=M!7;v5IFRz02;16H958eMNly)ZVH+X`y4yZu&nnPYE3BWCNIR@znFQqK*wlA>Qs%i4_aY zkve)rM8)l>kn>+=)JMt*E2S{bt8d@cn$2os-hA{=W)NA#j@UTO!lZc(&eE zT{d&ceEXQ?Mbh|PsSCp>$nmT>UcX~I8&+>vbclX|3;5^a`^R%(|NXg4eh0{o8BGUp zArYlb3s%yQ5>>(LaH-!*icKSUV?2%-FOKc3)5n12sY*UvsIuzvpzy(aY8Sl9r}GrN z1ymPZ3s{XNGVk9gdRqpf0;nN1C9mM%XVH=e&qZ{g?ca5{XPE&9Zt3BiA7X0k?j^)I=xy}k5Wt#gtf-)cg}%kti5-eNd9DfH z8)6(~j?+nWn1?*5u$<3^mCspaMLn7$%1CZ#oe*{k5S%Q?sd;Albyr!PU}mv+xru{p z$-SaM)x^CT)XvUxcocNTA@&o(&7%Y#BqT zR8`5mp6qLtHat6YOQmrApQ|?nnI~UJH2g=7WkdIVn)u$B)o5{`_!czqFN*Py_^GIF z_{_x98{6+`h}7iOC!cFCHo>>On9f{sUnzr^o9fs%Y1i0|+TH5xxT(KwH(?0}|Br*_ zFI5r^+Yn1CsX7Bc_ZLL*4(l(rMsN9O293&ZeaKBVIY@Xs6V6jX{sf@sp&}opBK>Ow zh}1--n+v!A7VPJwa2U}@;_W9#=yM|P>?veYsSYs?+aby6a>F1y076*CgkLvS#b$&q zz(AZRH`Z2wJ5F8>Ky^Zm?F;GsRk56rDb_tFM`F!)cp~vcDG&X9)-U+kCm1b4giAE6^Y6FB zy5q02JYMvut#e!|Z4jXo^k|7yC@O5_suj9YowQcO)T_!b>>PgCR`?n(NRx{LjblNz z6Vsh0RZEYp((5-fFGA~?Kn_#Cot~h<1zF$^DNg9`&ZI#o_HUVKZAzGzGK^7l5NY!D z9I#uN_hDipR%)^O3LJH(lVcygm%;rKZya_1En<#3{o+H8s>)zqZN{OUPZ zD{Va80QtA2M%7rBMQ1*M$oMY{xvf(P(iPmY&h{-cxi-$~e539q6=Fu-J_2uLKu83# zBk8^SmKyinF?hk1Y^go%YTF@eh}Y#kA<5;MN4MTvB(rST7ln5=V=w*R2(kkrBf`xB z-$D_Sh_>Rz3EKp;BmoF+t@paf*aA*bw)h0j5P<)gfhj{$mx-r%{P2|g@zFq&2hKb6 z#;#2}XoJY5O9>kG)#1bX9*S7acMc1r0+BHI2aESq=>#0P_Y_U^rsVbW0(tIZMJ0GW zzk`5C;k7tYB-CyfLR2gQ!uWydir|G zqf7WY-_{T%6nkkrdLGupY#HT%Vfxu+k^}I=Rc(CTF8bMQhY%Gp)fNcG*KJLF-KH>z zP0*a_2w-FLA{-f*P>JT;$Luq=F_9$Wn`vb2OZ{HB#o4IP;<+ zqH(;()6u64l6G-?PoD-2^Jq*`@Yg~prqI4;FTPnUTk5QS=b__|E$eJF^1fQj zCqL)+XR91qp+i>|4&-P=o$_R^F48E{uAInexLW)+Ot*)^dtJIsSl#xv)2Vz zmu(3`t1c#(80Pw%C+bBnrO$K*(()04_(?f>j<_lY92?1Of~U(Fj$!JY&JI$YojA%i za<{eb!FiTt<{kAJc&+{u_8?|tcVG>6Vl z`kNDvh9ooY9PB`e>nbn*wKG<62QKt<+k;rVpPAH+CGjXUJ{=);J=T;W_c=7GIl$bh zf;Z4WmB2r4f70Io?8o;c=B^^kvz>+jwgQv1fOEleb^!jD^U!$#}es5nr2Z^9-TgXzhQcpOMGWc}D>MQSR;CSLn;U6$Hd&CXQv z5d-?4X+2J`4cEj1Clp^s8)DDQrW=s{J!t80XHGP={cCk7*xs`O; z6ZcJUBButArZeGRHCF2lp1tC#@H@&otwDa(e79IPTZ)$h$bE(v?>n>--fi@Miy!%9 zMKiX_mO`6nk*YZMmBi(M*Z60ce#X|gzp*vs7i=y5Dw4X!ir`F_Z!I281RlHMFO8`U z26A71fmzB|AH#Hie#mrfgg1FxW{plEl9_`2MvVGR=2A54BM1UQ@yq6~!y(Hc{-czT z#=oDwLP7u%$c`a3@7+e&iTK)AlZYt@VW|T9r^5OvOk-QQNcfr?OVrsA2InF-w^pTd zMMTd%LF9jt8+5R9YcJI5O4$GJtnY7Zy&hPb9WRBGq{##*X#~%> zNvOpAXW532u*Z%Y2OJyU&`IRJikF?75n6w|oS?=T1t4WeUOxVmk!UB47K~#qWXB`Ff(P9Eter%x87ZMMA9FG6P zEZ-IQ`Af*id8D?C<(Uh}O$m=wo)70|oe-ejv|G)B{aPC%{e?>XWXk5F$gnb*SF{dl zLWNgz)&B9Q-9K$DrKeqy0m@Cq4R-s+kwXdB@u!qm#;v>?>TfAA^f;)Flf;rAtmm_C zMD*R)vcwx2BhLD8xbLCo=#sW14s>Q`_tQ_JUD|aRv@$#!A}+pm{cz3_26q#i+`f#s z*o+;i4#w|SbLPdShPgtwHWW|sRjxJ=Y#)pf*d9yT2spuBLvoj3et=!mlh(zd;VAp; zLsnhS9ALD64W=Tgd~=Mzs1n4Y%IHmfH;6EG=Y2@^Zr|~6;_p!l7cE`A0GU7Zs<=u1 zznih-LP7ZLEry=6!4Ljw?K`dGg?9}&i3SnKyp;m zkk!NC=dN(B{g+i6jp8p{_jjLGmDafxf%aEl?)fBpgs+Me77W_x_lmjp)9yGH&+68> z4qd$;Uut?5Gc~!&;;=FXT{>`C9dE?@_<|sgVHS1wk%RkCbDglb)xQ1uBB(M}?1SPtu5zW~M_27q69r61ypJtDBbNlD)5dtyj`aZ?As5v1_g>vWP8Uzw6!SNfAYiB)hvc&*&<{V{uHvxw%l_mwxuf~krf&e}4Z1jkjMBld_A8sI}Kpf{O>@*;_nXq&wc1Li0AxZOL&V>PN7|SXkz-PN_Cp>-~>|&ky#qe{f3R=J@ZN@?1uH*8ln7 z^QC0w?!XIcd_Xdxch;h1ifBtZW^zQS4A~W?2_Y!o(=&usYD1nU#BW(Nt3F(q=H1`E z4?GXjRtZknmY7OJTUb))S+s@GKw8>sFB&b3pe?M3-~@7ODFy<rUH!B-}9%c(vv zIy9{w9=y{+S-b8mA8hNWVCnRJaKI(9V>G~Fpr%nIVCPKLAyWB=`iz$96<=MqsAj5B z_-V4&G(kYa><9R#8{Yas9xj4|e4p@gV(w!Nt_AP{obnTTsyA?v#JKW!{q7RnxVLOn zht~-ngLeU>X~@Cq6r=e5$xwRmF7p7RN!r{S^N85pmx-oC25fgtFej;;X^p0h;X`3Y zab_~TmGDkLJT>ZoIS6rAWrY2;228@J+~6rr`(Kog~7mz$Hud?EHkK=rY4 z?^B7)GLr>m%#L?l=cDyY=CaJydoq0xhCx+grhbJv)zK z`c3@OquYzsZt<4@MAZYoGgJchsG91{pgK?8&k%&;Ww~%3ID()eW7V(R%vcrGwC@RJ z2t9O!FV`K`?NbfAfGLGXA3^cuMkVl2eTp)SoXdnAyN$k8AkC zB=H(5t}79h`K2ZhWND3h5sGl?BJkF7jUPFV_`y8DiE}zB>#KMYZmwk1v&o2Zc|Eud zty2yHWF*I#WFyu;=ka~sgCG6=Cjkod`)B_hg0M-yg`j&TR_OO9I(2slcQi>l2ype< z-;~Xe1K`I|+Er_MG(vo>`_Ia(u{<0~o@P^%PVNM7KD>)a!U1x?)u zEYP8rDFs?M(qRDN=2){wh@;ps-0#5+W4ot+US_v%zY^W>v+Yh?@usAA`T}kgBk{Nb zhF5&*To1x7tf3>aAP3)pT3)`Z%8upGu7b+IDZapN%fh zsN9_76}qNeQ>nL!1C)T5IPc)681{KsDx%JipPvsAH+))%%@uCml6@^RyjCGJMK@6}*J4t7s-&>}9-J(5VaI(m`0``G?IBr6N zTZxib|miUPX#2%ruI*3GUVM3 zSp_crNZ}X{_Xl&=RU60I6^kq%!QzZQqhIi8D{iy6+O%C)(@;Y;2 zym1cJJ&nBu*vpQDt#-^>&rUx}n-U?JFN(asE<)$r^Gd5gwQrvOQD>0eS@a97W|pMz z)qAoK*OaaI@wx>Opxp8_p3MZE5)WW_`_5ebUCu(B%3E6GaF4)hdArHp!o99C!w*{t z+S?W#vaTh9074ky-0&wpV&71=@3MBpM+RPi=!5}8y<=2G%oE)Y$W7-6!f8gB`@UH7 z&ffXXWH@te#D_;pw`EdfpMe}2K~UVv6uIqlFu)Us7s{f3Z@~&Q?sgS@kUi72Es-P zCIQdpevS!`fqVbM9?z2kf$|X(d;r{a?)c8kXc5{Bdg=DM&~9Kj)W5605JK-E?m?EA zw;CN!#i3KPQ(tzfK0(s>8vUtN1|VOnWMp^9^7}aMk3}jhUus*t z8`)Io)KE`#CI7@@*gc!94SzI+1ydZ=YxF-GFY2r~qmm%hL00od6E@VQuNp7&Rbg%A zSHqSDK6gS3y(#wL?R~NIN$7UlD^o^}q=c;g;*2o!p`rZ=v=bIbj0o2_$>J;J@gL?E zdsc0)&t9P=acF(?kH<-LQ%P+AfPg{xojZYmb$|!sa6==AnxPZ0J#Mj~0-<3*B0cFJ z7>t2rVU?-QY+^R55U4>DmJhGaMl0Dd{;&zGbGdEHYWJ#xN8_Wk_KGg zOJ|e9jDYd$xpGnR*Qdx&M`H!y1;!wYFNDTfzUJZpYeldyb%wbLiT-Kag(>Ib}DSWj&!S4hPFQD#FHJ#gLBdmC@wXN5W@& z-(Gr%8RPk^zs=jSaJUaXRe$c2*KzuK$3s4>Tw>z6_=q`dRN^6fV}$-k`b#@jbN&Z} zw;5_Kc(l-~>aX<4)jD|uHHC%L^{4ECW*5-b;W%2kwfm2mavTN_0z?ueebP%>TJxlA znqNig6?)g%uB$e=9L!fvCRJYb=QhgdH01rBG3W;|w3ne7!-y>-Knu<77H=FjxEp#@ zz)^x zAYmLAb-*>7BEeiSlURSuhBHwG+BKQ>aBw{&Bm8wNxt~m8DQl?Bz<_CDGMWFh>5q}( zINAvA{$K=$MBMwCe>4If2g{?`P}JV0&lFM|8-5Uf;cVvHcq(`mSIPUg1>$z7e)kKV zQA%;b5!|*p>#zK%_|QsRaucUM>Kys$;g-#)_-uBap&YZvDT(RltMZih-UMFHzSc}u z^H6B2=r$GGb!s|&IEuWJx zHkd0PnJ;_>OU;9~bNF+S8TH|Pm1jrpJ{WCsgAgiAGl;F~czmdf#pP%p8nc*0MhhkD ztPERED!J~aw-1lH09Zh$zkM97h3xNq6#jEc@54bs$H6cb>s`^F?lWQtyJZc%;)baXr5 zkXLJ;Uhk9By$|STA&CB37ji0_mocja;x8PCYOs;f+OJ{=N9K0}1b}Aj!=dMa!XKHxci=cT#qlf{^6qsN6D8$w;2WEBww) zmJTN2P8N;{4L2PCQhcNTq$4a|EzM3yFCNILgQtbw?$(dVGq4L6Cw=KV`Gv79z=w1 z0i7o4_eNrpBqH#CtjlAb-1^7ex+Dz$yi|?#r6=K4N_Eq3Y|*FEDjpF zmM8Rd>OB^G7)*Oy!o(xR#sSTD|}XK=;2xAhqWPexM0(s~-bF%6%N5bJU&nd_NO; zy5vq5uM%C#7>IC4e@u4yT=5v_nEK91S!xC=PjC9y_T$oMLW%TNa@ZM{spF{!$S_i_J8I_k`&MalQ{PQ?(7ZLmE?H40`u5J|+#e0Ya_G9gAY%P&&YLL`@r z3BUpa$qKe}#{4d{7uZt_@TE;Y-`XJrLFO_drTjS)A0I8AXW*R64pwN%R6s`JA7e_p zE0UR;72a%?Tlj*ydMYWqcixN%i3O2kPSi2&kU&{E8vQs@gE1D{p997}8p{K{hVY-V z!>|VNIxFxF+X`_5ybhK4M7~u2gpgq<3N1l<87~TYDex*#uyjzJ)-kDFPhTJgbe*9T zw554k{@Z}TdAH3sghhIwet^k#e&9KAk&)=Cdw!q0*nHG0MF+iG<&W+lmqa>ohN~|V zD4y;g&pAO=VKqPLsK}(18gZFe>C-}A!Mlsjw4QWYrGMWVyB~Y$WciCPsm_ScJ2-G? zGbm2)>OdekDQC6X6Zowsn;vV5FId_@COc_twnl=tMYf=m1n@UvqG){JB*xwEq5sGF zdQ%rEKs}lwN2VZ$WlL_>NNYl4Q95QpuW^pjpvudk%ZJ^_^R7RKm=V=~%rguy`6{&c zWrQ*b3{0q9+Y=s#-I%V_3wuH{M{FR2dKeOOmlp3xf@H-4z`LyQdVs_@+(PWv!%VDY z&|`oPgCi$@AZ_z(x*r*yX}V+-Zc}=+li*~A8?~PiiE*!dhA0rKU>1Y_YtOjT56c*F z;D08N;)4X3P~iq9&N9eZ$@orXVlT_n`IQ*Ud%<0?d3SIS4e5vBD21wb+3$nvTnF~_ zXF9s`uD`A98l3N|%}`GEAEw^$87MK)O;B4f;_82__ogL?^LW7>z`QTc+MMs>P#vK* zcxR%uaG2cjFK*@BPQKAM*bB6R0h&tz<+>44Bl|1Sd<~L^L-+@+LHPw1BLY)xgD~~f zt2hF%)Bc z{q@`0EZ7ApintXYt&-C!2@b0=hW!F;$)WigROA0&^0%RTNSyw2=4Wj(7y(>2e0&j{Zpe?s(kBT*n&z@a6^(Ib z#b|+nz&O#uIiql^*M7{_4}Q3o=Y#-y^_eb8(KJT5&6{+H2b^|du_|7o&JB*3zR49- zDSyjI^4C#Iw~J*1|?AC2xxkdsB- z_U}Z;W4J}tnX{Wgbs0C$_F@%B&eb!*>Z={B$6p0JF*%Pat>}K!S1e=^$?JT!;EsCaS$y_%0noY8tv-^+K6WWR;v~P9a>Tue{DV zr&|H_(YX`)_4cRsnnq38mudIXJ1%Pn^m{8-80d+W2QT>QBW z84clo^}h@_G&YqO)Fgl-VI+=mZRoVKO_U3`nFq#e>OF zkgSC8Nhf*@j5$a*VR^BA&g_s*VDXfjFaz64_NqaG|H^4I1MqUUCU1ZTSf*5trXdW= zDKlfy=D;4J3})eoL#kiMj<>ybfg{oGW9&?%Ei^7OX!qw{u#7P0L;t~P4dvh z?8Y@2Hf{0U%V`nyJ(Fw^X(|JGkzB8!W>a|p)_0@)Bxuv{`eGX2@ch47mV6l%zq{Ng z=?(`b+jjG_kuegy@1E>FOhJqMP}G?TZ`1rV$#fCAJsfTX{b@ zanwK2QoZM*H*MZup??3Q^e2x64WWvWd0x4#kdP9ao;=$LJ4^Lv4_rujTKj|Yvn-CC zcp>}nOuL(*<-xlrTF?H}V3F9#_-_{>d_x7_KgH+k6eGv$;D_4>f`sQo;Uk1M1{|Pd zxH|4r2bTvtra&0gB}aV{K)_;j_yH?l9q%2lA5r0*EE7L~G+_utcHh(q5^fRJ2^I{; z!3`EC;$#SMixko$A#)c_5(W614Lk_GCr_Ur$LV{&b^A{|nhYW7ho|_>Z>6hNM%v;} z9nLAqV;T+Nu6PFz39_#SUbH7G#YVO>YD*7CTI){-)u!L%VDHIY@zE?gI)F57&|5-S zu|~Ip=Fv~B?x!VclZ>_Nnq!wAklHO0R3q^6Hm!@7{(7oE1Y zMXP&|eVkNPs82N3@O+k;{od>LD=N%79RBe~^+4#jAKuT9-9*bzVB60SvI=6E3|%9X zH*hur-DM43;ccFDR*>1vaJ~$MDETdk3MY#LR*q^v${tLf?9tCxe*Phqj)lZyjNGuKMarv}{K^kG^}Zi$bP8bTc9z2uZxX`=A#U_vtU;1IL!A!;c5^W!@a$N^Nvgk^GmnUVG4FNfhje`V)CK!#n3=GId-jjA4@4Rv7 zy7VBBteq}VR;MQEKmohbRLGUcH$d*QT++WZQDTjarVF8lKRfvcZuH|l z+G%yY1bsoRX+^ucyJRp&y8(`v<^r0WIFGEVYc}mPgB14W;Ben`1nVk06Xfl#AwK}~ z=SuwfI9t+&ah?Vkg2*qd=D}Ntl?UlB6ey)Ro+>m0g9D3R6gryomlxtmXCUaFXQ7?F zl_y^tdh>??GbXgZ+oFy0UxP8_an5gGJcV>Vt#5W|r)XNz6OJ{%|0*J*EgA3rppwZ# zNTN8JN?+ZB=Dc^8GH2hHe4H16sl_$ni}#0c?IJXnZzqUZ9$dTqzuf5lSN{HhQB_Rv!&-q0`?f5}v}N)P zULIAum?IMmffyDi;<*Vw*axw|TtWDpFgGZ(L&^gr$#F(2LK)wY?+0w}3hDY}gTM&i zHtg$kx>A|vJAGd4u7?C%qp9e#K1&-NrFM=sgT*&ZR9NlT^o7%pB{4jd$PVA1{P@Z2 zGVEy4d|Narjg+Y~OF@YX|DVPf|E+J-Jn8H27lp;u$nuvQ-^2Y^~sbRrwIq zecpK+TkmBk+LE@9&<$8P-1dyJrxpz3@Db6$v<=Xg&qqvfNnyHO;2^^B|G;1ni`;{v!^4htb*?-)-5`qE7-n*cHQ_E^<;i}DP$`*>;1EuMle*!g!%bLdr6h?Ru12eV z&FePUsv{b*6lo}RH`tLRoGsHd>xyAvE*oHmcIH3amCj>|Uci%{h&d{v1g+uW@<*t& zaPY@KJ#uYqr_XU^#ovwQ$Vt(`2l%u1{O0q#Kd-kD=3aR6I<zrR6853lHJFkdQ zgcgfQ4tSm|dWiPpqzi6uvP+$>vw6VIpEXi$2K0}C#5$Sp5N6x1s zJj}f|BN_|IMOChXPiYjIh`xSAW1H2_c)G<%Sy6MMBU%jKpUqsue7KH$nS0~>dS9|O ze2~?40$Qx^>rQyP=3-$3O`prLU69@mcbGG$8A;^VXXU0BZTcr@-LkwpI<0) z_9$$V3!J+t@LftHloxOBd2`8hUGkyp)bSnTlL+w&jjh{zq_Z85%K=O&yhHu0MRC-% zmePeQJ}P09`;t+M0p!NRR+}wRtUz|1KM{{MCGV}Wi+{~~;R?>yiJQ-%93*!d<8{rq zxsCA7otCT@UwonVyi!1_@+$Tom}Hw0CY18@7Q>xRt{D%+g+m4Zm~($M7JfzD484ip zztBVXQ_G-GR898A!u$O7npcf(q&O$xpRKs=i@L0}!{PTW zC)Ah^5v(Wam;#n4j{!IgawG0sq&G=ECm}z8&UQK#xbrqwV?^?K=+^A`*mX_6*E z*KRcKR|^6M2C(fJ-Q_>C1(il1hGyeMft4nGpK!mZAA+ke`YrANo_}tG{_++I8`?#}eqKNT-FyArQ4(O0ZK7UlA^>|;~W zJ-7Kzfy%Jng`9Sh_9EYQq)q(GS;W(~8Xd34&8@D1sF>&F} z7{yK1{>g@fv0oD_?bc+?*(wwfk85YItrw=v=#@LaZKPt~MA2iNOFMCD)7Q13wy zgkN*2932y~w4d4fbTzJiAU9N2KM46Whb6cUM2^c}Uqe0Q)=-9p2>T%ifB`@aK>6pk zP^F!p*7=#0?pUG$5ORWc5%&x2yd?i1Q_<`2ZaE#A-MkgAdw{CiTuWDq#kvnFUVqc8 zB4jueiCTrLp{PD8ILDvV#_Zj5ji@t)Ja=;V4=ibd_ z=S{nT{pRjrQlXCn#`@5T#6g=j=fhBzvQVS%70|@tP*1 zl8(Uk>wI{O&ggSD9Dtvz-?jGO6IuMI&rLWcMA3R`yNn<0bz#_OwJcQTP-rDzieo8CnbQCLi zx+KWOuOZT0-VN(#JUryRIqaHseplCgq-vT~^4K-1yY5H2CH{`%)V*K(F%ZA@d3TQ1 z{kG5hR*ijVMibhFFgkPUB}Qk)p?#nTtut@kZeE|3!061X_3XeCXDt+VyPHe7q>>KJ z(BbLvd_X?O5RYw34_CMD861rR>j|7*s$0+V?&EKEXaM?e@1iZ=qNPX=JDAse{1slQ zjU+Vf6s4dZlYpPa?@o(QJUPuO&1!q0 zGg$j~I`jL0#c@S54sR?TR){CZ>;Vq+za^#Ky4wR9*O6fiddY_5eW6bMiyZC>%Z3uH zp7GkT@uiz}z64HPyS!a(c_?n0h#!HkM9V|-0kd2}hKSlE zCGnscvpJx%L;9lSC&ns!Bm$=sXHqwS7OCN3@)1#PhS>Mwbvx@Jk;)8Q zQ;+Q~?DlEhGBxc~xoL)-7=3{-gFF^Dap3x6lmEbW;zGk7JR8VLQ#n1&Zfe_ggFIbW zWY0*!_~c-g=p}i14kjlk=8(9rSx$6D5@vTN6o*@_;Ptt$7vzP(5JaKl+8!4_60B%k zTu-Fhr%=fx7*ipOzlm)S6_typYY3&w=$ZdP(>3LR9-FAQ2-0N~Org*~B`#y+7 zaif>9vTmrdVSf6lv9AiTju@@r9B)nXza&+KB-!h3!m%BJrw^vqeY^83_V;c)SC|=k z>U?N#D!j3?G^RCjF@$VZJXvxed*7*}=FwwWfVxWc+KdDXG4%=n=RPCJq<|{W?Bqky z|619l7{h32Txl8~)~KBrH+HAUJN~kyXguP_;L)jPa%zw>Xgu;2WUy%sI3t(>3xTlt zPJp1Z4gqk&t(^&w3^mL%RGQ4dARJ7V3Jba+<{1}s*;{KWhL%tsjY5LhSAT;--wesb z`qYC9IbL4XyF>9)cm|vSV`uh~T<+~nL}r3x2Uye(`zIpZb+ck}c&L`2ixFuxm;t6R;=|@gH&=*(c)05vRxw&DCu1A8 zvzp)_62)aLtIkYh1)hkKNZ&io{eCngU-M1^ny|l>$%Wm(ua%8p>SYwtu{yt=X53%F z3V$2v{C-i3%IvwTSE~?f4_pjCNx8MBF~HgR6L- zeF>M~!!0#cHd62KTI2pUDZzr@?J56`8(vTSn;U-WK?$%U28R^g4-Xko*+Z|Vl%o!` z&$}pWO$uy_2$=(xZqhPP)`hw6e?NO|-26NWc-Ch&utggzrfkSk}|T;>j;yLqL9Mtsm)o zg-m+vA|b0IKwe?c&l(KSIU-4MsEa2bV_(_m6eXnYoPsFv;~7aR@*|}oN=hlyk6E5Y zez>?-(A=pb^2{VV3E`wZBl!?&Hv3DRO2h1PFS07mLFJr(jw61SC`8MVlL?vOpBLf_ z#_-fijcSv;Qt(;xyr+ZyY%zwZx6FeJdh^WFUkcn>%>KN(OK@Saq*Lb9QfaqF04Wmy z3*gQ7#o%0kg{Mv$s!jc4s$rr~WZDGt;OXkg&E2#d@UML-zeBIFKcbfa^U;>LM<$&| zr?|fKrLf!~+wdtde(Q7Sj{1~&9W6{`gGTPom6*zs&2sT6|{TCi7T^);c||h@lgHfRux5FN<5?SQWX2;Xp4Tv0Qh(j(-5*GFa)lF)j7! zy5sP#wz?j3wSl=4s>@x2kO)Xdd9$EHqc4-U95cj~P+HrMXZ*P4*DzQaqquxb`mW^h z!EYtefl-N%G#Hm(>`M;=)94{c!3DwwAqr4AtjK86sm#ZhtL;P7^pjq*nccW+G#yL& z7>7dLIBv@*flFhKD#4~*ki<&jMa)bxMO7(t41)4crglrnBum=jR!)jKRV!P?A zt|a|@h+00rG%QT%KtrE67QA|al}R25$1My1KQrM^z-)p8&~?acndKqou=B2S^s9#%4H;f-TVLteQlkT#{yfC z15loCfwEr>lV1XWV*pdKH<%uFSC7?bJJs(X>}DsgQjE!62(eNB=taUN2Q+S-^VPHF zXRvAnK|5#1;atP-5g=(Xq;$|}a4{A{=^U~VOHF47raV=vuM>%TN08a;w!3MTwDzFQ z@96q>fl_{s&}=m3DrL5bee>(I0- zw^)jH1_AW18>JfW$^DQ>By+C$z@7+iYDrGp*Uz}^~lPjG~pyb%^Hm1owW zY;QcoA57bJ}!-a_Zvex@~(a-5Lm^zq$gVtMq3;}kt(W1#9LZVYwl zvdJ!Rqjww;n9o{J0Hi^5Y$oc9e4%t~WQ{rhc{aN6V)>Uq$M8{WC z{278ZlOT#K{i&*Lyd0CB?}yhz!j(?%j$gAmV>_;3BEQReS5>OAI{;eHJrqjvfQ$XI zdY;YQUrUIQzkj$oB9!y*I-t|*Fdfh~!q(rF5QBILl4_dCr;Feca7-PD^D0mmF0$IMVm9x=&cU$?=g-sPI{}V_OwdzsKV+{~Jyr_h;UFUH$Je z&Wte+s}+lT3Y^<6JiUlm9F#;IHWGR z&m=Y@4D8j90BMOyutr09>j82?{xhHm127Y_>+)nDyVO! zL4SU7J%(PKUg8eTY5L4$ofDpkDG<=}qnT_$${@I;APM1c$|7Yvn82iJBs}0_!Ud4q zDW;G3>hpeEfSP4(uh<&{&z;{4X~h2Ns?G1VvS_UwNpk$i*&j&lvE{L$3-2fE;B~jj zzBK0^4vjZQPIP2T;SBDc-HeJCuxP6>ZJ}W4PKLsLDHe_>M|EE2&iOp#Q*n!!#OLLO zFH_WBHt+@r77W+n{WVW|EkV0i~^*ogbyw#`SllsJUOk#VY|d^dFJo}tPv3g9jSNjc08i64E*v%Bk4Pk z9YX#T-Qyc3~J;zLM)v+tAfth9Jvx1FWwa8*rb7}e|u+nl<<87XupOiH2 z{VBKePgbY!x|z$D@uQzsqM8=hyyl}|g=}%vfryy^ET}RBU_v~}F#g;NnC#AJ{(1HB zH&W~B3T-RtT*JFx>a6mKCs9iBL|d7KD-?5YaJ`^(Y`*1GNz1_5(CU7HpC`Yf{@%y* z$4g?rvyQ`mY1hqcm~gO$e1SCRaGK9ml8Z81gw) z82e)F=x{gPp$Q`5qP*WWEwylQlDy+OYi$6PDuZj&#Ow)Q%jKRI&AX@|tnNoRSe5S_ zps46Su7DpUBQ$t}h0P$a(Q$EG0!Q5sdd7=Nm4|F&7keTIkef^ozPRR~ z6(WsaH7+6X_U5>;l;B*(HPt`2fB!l4jJ7Z`tnTDX;K%x^V0uTE(Cf_HnQ2#1`W@ou zUu=ICLYbQ2Yli%Q6|JvC_!@4TMBJ9?eD0UaKTg`Yu~mZH=GW5294B+iRQyiySBW(* z6xE90@f4{rU^f-Lp`178NAeRaL_CH{&XuSPLJ(!W>~TLZ*vG&d0EGSlxL5!Pt(PwT ze|et6;kCF9BIi~>tXaEQrAeaUUr3Et5GSATp zWIzA_2oMYd0AK(b7=RJRF$)vPr30bq0RVbB6dVA61Ho{{3pg7U0RSPzzc3&%3=$=b z+!vSFFF2IG^3wtY=)i)vp6VdE?mJIEdr5sRFYBVL>8D~CAts)J=1=Ww�tk@_;k1 z%6-9VejT1E7Gy0RxG0|eQ^ElFwsZ7-^RQxby<+pIVr!>jYyZdQLFLkE<;HI1#&Olc zTGje~)%tPO=62QQUe)G7_1a4WPl}*y$t>)e1_P_tySGPO1E;_ciJ9Z8`|E_dy9CR&j zb*-#-ZJ+imt@bW1^==>bEid$~Z})9&^}|8^f=vC3YyHbx1IvF0{;rcLf5^+L_9l5bCs%eR|ISZt@BR`M_B;fw$;s&%^^{6I>l-*b`geAGadvim_8%^&XH@F>-?8(<&GVzZ z^OM8#GwS(&pq`ymsTV--#nHvZ@#)10_2Ts6;_Uq5{N&>N;(~g7LA{`UOr*BwQAf7_ zn^OOG43&=x*S}-*KuyKq&P{PF_&@gs@Ez08(gV-{zKiQ>uGW@r*3Wr8?A&a5t-W8^ z*@!=T<;3f5_uTpwuiGnL4XwMp&tAQBw{~^2b_Iz2FP8onR{+?5fr3B60fL%?VMr^} zE@D<|I4dt8HNrBxJ(^Y0eo9UP7shl2hB3(E&-$4pWJbh@+?;P?z9y8Ip;a!_#~`5V zn!yl|Gw_*R;%raG$-K&LCh^+i?uvp3xtL$)4jMlSM~h6$?VX@_Gexh#?4L=86=Dxu zrbi-un6H~vyqtk0r?bS_*VlROt&ZxH-ca({9hy*AT#>#*uoSGHZ9 zOr^SFZFQ#zk8Ekp-txfvwT`==PNZFr#N+Z0i)ZURmiq&#V3x-En!VBCqP%`xj`D*g zpzi5mZ~l4CQEcs#$IJn;g8;}8tK<0@c-8|ZlurqwzbP~0kpOg_Qk<1Ho(!eEU5S5o z0l^V00;mCxGQTn{xjzIY=v9}{yYwNLgB_NEqp_`Yyw91bF~7_Vx~ z7$dd(JV4hc^m1Ysr_x=KrR3Y(j9=L8+a$X9A|t*FN^p#>uA8N#@|x3hB7aQ$RSJ^b&8dsUe39=1il=!Gg(`iDqfB&>^6Henphn!Y{g0 z)_rqLtOA&!QNlUqy0%j}8zd@T^T)n5sCK;;URu9DmcG?cvU+4Mt%3gB^!CUJiip~!59Mt;P1Cke<`>bl|F8gzP*NUWu_0^o- zjhGvq?~Izgn*KL#8Ta^L(ynwm^!F>MXw{Tkm8A#PbM5WxS&MU|X9O%o(z7gd{N?FJ zuQ=C%KXd8>O5L0`%5`eX+u_xpbyF-_>b!z$i|unT>TrcGE-g_*F;=IzZ>Ws=Wz->SB!mS-$oq677N1Htb)_DAlW|M=@jno+E%;@Pxoe=`d< zSZ|Xp-j#Q@t+8m`ycn#Teaz>%)%sy7Dlc>)^14@#12=o72jqOG)PD?)$9}&`9A(?a z&*%_9tt?`{l*oY$8F<=B50yjfZTL?5({rk7M3d)71Gn?Nr_)AVEm_sup$}CE2$gaY z`@5~Mr5?)@;p)Z%muR!zmPFRkZQ7!@(ZSqj&|BT^O$>kTd28xg(bJkrzEWg}Qiz_r zT%g$+A&rS%9L_!wi%@Y2EJHqPGh|Z>yLIe43utAp<_*J}&7sY)ts;`+Xh5$( z;++`o6+^qQujkj2(kq^|RP6pNZqrXD@GVjFf9md*7zSBPWwBI!4ZHOv3rNcvfHWWm z(0W=TtvrKolx#yRqXJVq)N?tq@~FyE*x`>Z0e=j>{MBLz{4O=IA0wj|+5Pg@X1c<< z6}slmH@9DS0P7M7U>FaPm>FKukM7~DFI3(FqiM~D?{gUmplSQ7 zFW9q#!yXn-3V+q-FgLdT^gcRd-?DN{+bH`c;F~Q4#Iou=p7%$(@amGo!<1u_50H_s z=1bR%>*>`(xMj>Er-hzzoSM2&^^-H~ZLdPb3kQJ)yhvL>u+7I`X1*V!Dt1+YP2Um@?%A!n}qJf7S z4^17Ih~~1D2&ZDRX_9tVZDQTi2f=)mWB{!&0eBfp!$p8X;%kAW-gm_JCw|KuLW~=Y zvtC^;@|jp$9WOhhy=wl-LGz}%3WJ?yEB_D{uMpAGjDc(ziLBrI(>Puu!By-Aw|Fq_ z&e{}xPGqa#FnjqCi1a^dz3~+vDxb_}IycWg9t9s==$|MJpb@iQC|VfRuliAH9TTps zyGUg}M*5RNn8qAKHO}U!^fvL|lMBwwM5CE|Oh<0|zUF*Z=q)4Qm6hHl|F%)3Lv4SwcIA%0NR1Ih(9QL{atZV#{c(Cf`*YuAs`6uf^d!R;aJ|Fvl1>Ln*U*Cy_q+CJ$7tr4~QyIUcI(BE8+n zbe0RfchtghJ1p|n^k1eX&kos{ow)FNPr=S#b()RRH+`EO#AJheEXDV{$Y)$KdEov* z!t2bvsbw6?iBR-QH>H_<#qdOkoZ-gxZ+)j*0q9%hel$^!HPtwHRysXB3qm7I_~3D! z?2A{a&OT;qCQaU++`HR!&okgMLK_U++gLFj^P8rxJX-~Qfhb4Br%m}=wZc1@$cDU= zIX`s$X6uD$3kUa36g1rC3it6(7V3@}QIP}71MPb~v-b;deoOtHw`}U|$b1u}wa3aH z6Q_NfK^J;d<=Z=S1`+d@b2Xbz#s@AAf?F;VB9sOF7-C+X34kdPx@QYFqAvb>-BDBM zvq|&?B>EzQ0C2TGVH-}_Mr{HBixlKJ#(yn^?jwbcIUjrCVDJsrU@5s^8BGpxRzzJqybF&yOz}UD4_1{65r+jo-Q^}E z1~-zqH@5>ACeV!(xHfC3k!h&0XQ)v%yb6Qp#iI`lG_IOrzcfqIgJ^4!qQ-jKHx>|Qh3xD!Ym}hEDz8&tmvr#WQa>teri-Y<%FB4o<=sl^yVZzyq5#G#@v;1t7`bMUG!86=2Z`cA z*YO}hQryKv9N;((qy>^7gLpAOcpz8~0N5b&qmJXX=6I%!MK2ey3&bOK10oMd;hw_@ z+{X#LbcsfD32e@Y&K<5w&**D4iDF*SJR0bP!7wojdNrQSE0j-;YR1-=byZIoeT6kD zO(l#@8brGr&dbKe$CjwcmdHkA-QVW+zR4|k^SzP$`-QBl-ma)$@mzp-)I6Sl%lZ8; zEcbJ>4>n#Oyob24a@-6BNn97U%A|W4% zph6g!d$5S8kEt>8h}h%QcWfV{U}-flXHEu*C%O({r7Z}>w-q+aH+ z`D=M(wpJS1E-hak{Rosk=Jn}}KfQvEr@JoMco8DW*&~^j?>(mt&^v5yds?6WIj51aXisP8$xmtOWHvtw!Bmn!>rJ4u z=ZGf3PzqU5LRlDlS|)lImcmT-S`AL`2Rzuz+3MLsX-K<0MH`%|fwtkUPp;t=m zc%UQc>1se%w=HGnjY!7999hI|uR&TF#SQK6#nu+ao9iYj+;NK1(@4uu%K37R%X>?P zZn;8Y{}niG3+G&m3bHRuAW9cT<(AMxO1-tqPV$>E1&-PUmLG~zRuleB7QOieHM&x4 zeW$oXyYQ!Xp`!-`Pb?f@FR{Xw*l8Dy)900bEpdq|xUyRE)*13ave*Gom~H=Uh)_IN z`>lT!@@@4C(LVRPdC6{f=7f1^ZdnP}WJx1^@%DTvz^C}DHeJszX-H8%9mfq1YpAgS z#QGO;jJ|AtHS>~B=~P-N3&%Hhi{dP5ciD1O;Wp#<;wU1ykZ47KJ;Bq8=#+?Z*i>E9 zK2W$*zgl$ms{nQ9o2)32rmlQ_6;h6&y%Y}sJA>&kv~~oTt_`iDvo=Dy$f5iD`;+gw z(WPvJQdZG#nJ1+(9KKR8K1S_6Qr0f3(ugmt#E2*@)gLK`lZ+C^6|tbMPo z#U|I4o_p7D==|WeAeuM;2s*X0cpU^5ZxLU+C3NrGS~VR_%>%(2I-@!li}H`Z%Xoel zN2T2l{B424&@xxyWB)*6sB6U1Q}bR?wf<8jvO~@5^g0in{EWJ)m(E~*Ty29i;1(Ws zA4|Wp+c4)-vB*)!)m@nsU3!IFG`3yzNTBMzPF2%9T_fOnb-~wIB8<)#(wS}XS6!#^ z2dyzLt+_MAdi#0I8oWWKG5ld8om*q6S~e{L39T=EOdw6G7bDL|*}`Z>U6|l=u~!&c z5VnbxskXh&=4p!k3!N4;yzV*&N#{(-RDxNQj2^?FV{tGK9JJ}CM8`aC(Tbh~Xr3n^ zdvLI@ZACBhTW^W9^6?;%0K6N)OkEeE|K9$6K${S!vGAGFhgyw$VrWqu`awM`BtF}$ zxZOnk_T#hmY$|KV5FQo{Q1UFy>;6vry@EC@9%6x|Hzv_KlW464#fI1+ZyvRDpDKBs zDflSfX%UVIX$JknbVcL3T<~4t^-sH1p&pcWd@(Y%xK+ZpBlHKoiz={6de?I4^MTw4Gv5c_m%7RM4QX`@AuKZB{v{`(kqD-ut3mG zwEk4}!S4EA7Nr4gL|=BuK*?G64i4tGEsFZ#s4fY)hU}P;hlP-&nKfVr$Wb>O%mMF0 z|KUn+r*H2~v5BO?s|eEFUxPOnhu0|G52m2hMhs1ydWInEdBf~|!-r>sYXK9(cw`g-im0+rn0}dH)8T-h zOm!kth?C*uNn^yXEIn9W!&GkU=fXa*l8sSM#R<-aL8ah<;}~G1({E~h;S6m+a>g42eaA1@`$BMoVd9h4w6?{>Z1D6NVfx40-@neLK{Gnf z44Ut2&op_={B4+VrOX`B&F&xt_qY(=R+XQ0?q_{-+?!~R`yWXFgu3UF>1I$J&L0Wmn21N-FkLcTf!7)QvGMz3+X$D31#eo; zaYW7a70m?%Paimc>D=L7NH7ViUJlZXLIYz?#nlrjJ--~dc`;G`*#jTl>+ z31l+>j`?j&WNt7SfUXcRkU>Tv42Ta47N>v~xxp)$o2!kRtKY!l81Uw8@Vxk@5E%_h zVPcwO$857@+A{t7R{xStnCtfTqBFacfxsUOcl6*#DhAm`NENSVVN~AnvP3fbBUt?r zLWISD)|JW=1Sb!#iziz`GiWs);mN=jlYkccj8556(mRHaQ@6RzwO1zt{>u69DSY0O zdxiXhLUsc7-O`W}AYqIh^A}*U^l-e!R{YTvrvRD7Qv!8xh}N43x#6Fb2+TH*ONy}s zP4IGRb|FfSmKt0R>VW&5@u?I7=Wbc%;;lL#Bgt&_Z7x3+3=u^&Fx(@d&T&jr&i~Am z*`!@p5*R?{-3U>?zL?pgchsZU0m4UTI>CPWUu4c#Ji8%5SZ(y*gxh_SIfoCX4%Y!k zQ_d$(G@1NZ*?8$A0wQF2P&=Iu4*p~a`Q&_wKY?^uwkSKXv=TD2|R!=pUQ$^6Q+zx>;1XV>a zF`cxgG6)3NUW~+^j+K6vPrRa+ss3?jF~*VHu`{>#2g#l$?KIOCnXCUv(U+b{J2Zwv z769ThS1%-58s^epmHCl>UqB;nZz9s@2Zpm7I_Q5{=?U=m3+~3Oc zPUd2uLJ7oWQpJ6&aNyQd&LC>DZhGkcvEKgz^-V^;b$)??U$*gVQ*Nx})bR7m26y+( zsRkC%z-0Zm-!dU zx1zqQD+pbPlxSJ+C4AQwEpbGk6;bkwmrzO>kTyQ}34`XPb8tYqK$4=Zmdv#4O>PH5 zi1S^x(zfz$v8wf59q!JvJYB|tDKsD`lkFz8Up}XzyiActIvAw@lUKj2+`29yuW~Lh zqNqkyxvY8T@0TqJ4ZijVhT7Ms86Mzdr%Q)-ZX@5f-Dmqch%|cesbWZ{o%53M-gCWo zW9p|LRrC$5d*6(UKK%FNAwK-FyP{RBq_NTS_wA!bk3W?d_B{0O8>zNWs&lV)beeV_ zR7j^y(RRt9eQnM=AT9oQXzbtCm`6jb$rHtn_mxv~+g$pyFprz5?fAu~R|h9T zu0O9doo+ZSL|$!G*q*RM40K`~iWQE?n0!*{;JuY}^owgQ%xyawnVnbjl)lKOM_qc| zCtteonyl7jcXbGENh9er#H|0OFj}ZgoOvWml;ib8!KdBd9~nhL9I?xWI@jMqm3_Bw zC<-#%)L>MdoG$%TqvSih%lMRi@^Zj+3_G z#wFl2zty8~B|;g(_W3hcp#sj$_$z3aHZb)v0ec6gHy5cJLktm=vZTLXt@EfLkJ?PBryZGz&1k$NHgj4lb$c8}B| zq@*6q%?AY@Zd`Q{Fbzj}L?j}tN_nJB!x0~wV_w(i@Saz9Dp72cg_c^z495Rgwi8nzzJhsVvH$maCUPeQSDY}n@q(A=D&Mr|~67FP@ z003bCQgWWU^|gmJ+Q!+5$)zFIV%G9H=2jH2P(T2kTaC_>O-67DP3|=l0{BudA+;kz z!-{34e>DR>QQ!w)pe+ula2Y`nVEY>prqEUP&LKWHeKBA~fmz{x zq;Yb#T62eV7^oX9c!$8lw4$nDkBR><^5v>ixO_tH%~!(6bTHqt+@PIA3)$Gcs^W?fWW3L(tTND0V)0@RHSteUc>lq{N%`Z15e|rolVqZ57nZ z_|an9f63;ZG{DMtA_cB``c%DAOR@uApwwvE;px>*RM5%@4~mOk7G+VM^W*)4SS~}u zS9b$yg>txN^=+n8!x`fH*;KS=E^|ZnoD9 zycJhrTtrR3yadt`*mzZ*-`8Gy*?`A3TzURioZFaqt?+d;u5@jh@4a#Ul+z8XzTx{j ziI;21ckI;1(x>N8aKd@}hxm$#&GneBnfxvUgSr!AQl*Ji2XEvax=7Cckw!h zdg)a%)N6q^94}=Jlfk zi$5woH3n`oai1=2e7LjbRnoGw7x0na4kvo@lXcUz!(eUUwrEX<2P*%=btPNf9J67q zIkzOGjXxD(wN5#?t_97$t6HUBhYTuz%sD@{eaw?K{wa4q?+vHp9rWp7*gL80gbfGl zrjL`!?v;7GN>A;26ecV5*MGeIv|;AtqgclMn!tv9*!+5WD?*r)LbW}5%hR|X5uGmb zC#16fq|DhHp25XY<=!b-ay=3Hh;mx!UZVAT6A(TnN42T`X^Gl2YxXjhG8wOHklBrX z!C8A61duAJaWnP{_dyV^} zjqa{N85^%&)G5y0KFqFtY^`Y(PAl!;>QfgcNSmK~!ThF6sk3N+utp#s+e<5X7e7JG3 zo<;TfEBg6_xva69xaGT1p=^fpJ4pwI9@RuhmX1}gd! zg|?o}UJ=b9gyt>_R#0;N)Ttbmqd9^EwHJnXl2q-TPDAyx3YIFe9p41|@sCwg(fP|o$WO1`wG+bI&OlDn8Hl`J)3-U9C*u;O4cpf2{ zQ7Prt5{@NhCbn=DN63b#$+f9or9f?0pla&sUk%hbp0{D@Z_^m1Ft)<`#sh)x(s&RwP#%|m$lQTS}S$C-Wl~aD)yRr_1Ikg(zdAc zO>bJS?$4>-A28VeM~x6u0dNDt0m1*rbl`z;Y7%*YFKNo3yDZqmw1#5@?uu!aQQnI( zN!m|!mF0BCfzd*bCJeH*B6)cp>HhrwwI>B!>h^b+@})0a%+4yLz~xb;pX)r#E&pr& zV7joE;W~f0&`=VZY_R{~e%?^b?Z9rc^055D?0bAqxYiPNn7_08)5i$x2ewo>FLXbQ zsQmUg)t-mQBv-XHRTlt$T`#mbdsP#y?zKx}Z@AF;PCzYFQ6wo(H%Rh@(Ag_ODYVf& zl3TB~;_u_M2Sum@+^?!vCX>kIyjq!iE(J^NU^-xs({CUMm%4+4z?s>l)J^*pYxb7O zlFzXTmZ6bMqDJ|Y^MKW@5$20QYGoDbvywBVx8P}9W5em;D)lFIxwKJv=2KJAZ6VBO zf8VWo$=LuJW?SCClPg^bHy7P<-YNY3k`r@t%)quUl+`AGE_C?OM64X|`~mKk;0N3J zNcvm|PD<>17eWTxA-oVH)pC^PdobxS|4t}3e=R720Nt0ku%FN47b6K0t2Uf=2?%mGIOj|8<|8Zhp{C&uL?Q7M1cVXVtHv3v}6XTGu&M&FFK82lW(cZ|VS^KhO`@ zAG6r1ov%FoX1LSuu+;>bzeTA&t-OqrnN0-~8$jsdSIsxLGPYX|WnxxaF&r_(mdh#j z63uM>f)brv7QPZ$oA#9;rLh>bF7S(tK}I#M*r~EkWN^l0_d?7bspn?7x}=88tfQ+R zw&GBO?WM1ZEntB-JP-iq>H9sTr#u}uqB`pGX3XOC z`J0Jn116O}d4UF~;erCH^T10qBD)jpH!_&OYK53)0JtR~iIY5k`>JgG&{7e&P#*CP zmY3=?)zfhcT*oto$H=3l5RS~X<=yG?ot2~)#x=P77mc+O@2Kz^jl`731F(13VAyZd z@lo_db?~>kmHMQaL*=^T9l;%Uu7MNfZ_9`}dxI)OJ?<4C%y+sZ{K9*$3_kW&Wo}m5 zZxbQeRJ~WI2ToGsk^uLgc6a(;oc$GGxsc0NzIgx2i@Kn>b?`p&foF@r5J1Br(E~@d zL%-$Qi;xi%M$%gi__7q6$$Z}l@JJOf5Z2A22TR9!5rer(CpjRSYH=TWgRk5v!Z#QO z<^IdSTC>Zu(oyxQTlksFLT|)a%G=~${_G_f&U2Q+V6viS^F|O1>;k~J0U)FziBWY|W{xM&+NdwE4cp#j#4|)UC!in^!$C-O0+@pcpd&sAWZEx_Lz}AbLL&ob_B%kf zjw(niQV?@8fu=dD3(=Av9B5?7o8pTTy1~`~tq723h(~Ek0Z?}JY8)Ii8gP@J-Nv7>fJEqmKekq1kt7WExod+t#%PwWkJ9Pe)aQn>nT6e? z1yPX>`>)ji|CV7-gS(Lsm>|&fu5=s_VMX(d4HGM@?&A{k%+<#80U4u}X|;7|a)%_FXd;V4%~p1zk!d(%xLjZ;vz!Ko1*AUu{k`}N)C!?<`G)z^Ne|V*X>Qxe+}Cl{P624dgw9;ccNwp z(DV{}@O1o0PJSjff9F-6V%#G!*70ihYE{>iqNzXPhyNoHaPu<2th!Zd5)YT3RL--(D8j%2@=yQ%!kz-eITZ9 zSPJOU5y`|8?3U+1JMcud>6{qGa$pdqNDREr+D{CV{W9MtLnsm(#_iuE~QDjssKplTzjGfr=9Dw!_Z z_g5b*-{k{)lQ5g5>$tzbYJa54LUOj0Yn5;HAgdKWa}N{U&tfPcxzdgqsWlUZ+O0oG zr1g-XdOjx;j+w9W?{Faul}RJ{#?nO#lvUYQ^D%G28&FYOa`FAF-?hvhHr*QuuQg1j*WC}x-ZwaU zDW8rUWV$#0i=8vw^C3?i$W*^a+;(*U3$n0wVqcwwNpqz4U7~uFbMF&g+BE&=?)g;} zKkmLxdz#wn7~je{F|;L0&O)=JWz&CoumpJsF7B{Gc`oqb*liWWCw=*G<8PaaiJAKqLa@c30DM* z_U37S%Q=QSXJh?t7e?eJvzU3U{{ZgXV7-kA&0EjJbz!lG+sKlT?X3wdrAO~{LLMMs+K98{4A;KTz~WC53PP;(M( zs)?^}Q7savB83bpqErX^n}rCICo~ za2N+Hgsa*=LNmlpGbpwzECLjn$m)d0LBc&e-Ehzc6R`Uf0Ez^F;ed8TOxsu5XV^WF)C*9xz*3qOhH;uq-VSPl|wg65y&@N2r2;AUcHECBC34LOUCZj%Irh`X$ z<@2M;)y*KBr69!<19ilvJ;y-|Z>9xlrG}Y#_>zT~Fz87D@*N?`R_4)#^wY7_^jfdz z`4?~WnBZMx^vr)Z1J={=jr`BMVK5%=1~85*ep^CU>8crdWsr+tXEeWXH1!lp)z zsg@M5_)sM<;4kc@ACW^8m=8YP);Qgj1g&HH{7DPeXa=+Thi;ujQJoP!@zB@t8TuU= z{l^((FX#vzWXy~9A@3zTi7AErsTTvBB2hw>Q*+o(3cL&JZcodE=AUZ|9LjtXB|j#k>l&kTD4Bsm@)%my5a*8#ae!=0f{ z3B1gNEKYj5jh~qt$B<{ZY$I=oDD+``09_OQ@e%+adIC>m%Kd2u(KF7qG0&Ax%X!d} zWgL?KQrYL83XrZho=$pd}{Mju<#TNJcl`eHnsOhQpAb@sODKz#K2hIx|#$Yu;8 z+XY_d46pPqs!l7axyIg%y@%aJJ?=(|rNL_g;5AryXB50D0N$kl@9r+{G%s#bC@FV_ zx2?iS0OSNOhKbj_3dhiU0+>_yR?hovN&6d!;hQ*v>m&(9`(9gCP2w`&wN)U1=MYgq zgWBiE{LAJ(U=$j?%$u!>nLa_=GphXM6Q&o%OzZ)Wr^GD^5$%jpZ2mUMdmpo;%U)N; z`CC_DzibvD@Gr`}lNY(O|4lgk`>2uL&Sibthr*qbXh_|)3PDUZH5rHwC|5`?9~vyb zeOj)>0bs&#c1U5^uwY3Hs0|3x{avBIR)Hn}Upxn@S%B1ZDu4l8(gt8v2g6MP&M+EA zo^5oymw|hk8rnd+Zyzv5GJJVjd41_#i&WKXxIR|3YEA;VxD6O2qR=?hII)VBubNW@ zB^V?W08;0e<9Ze-1iE!Cn-MTXKy2Z^_mB{6@il36HR(Pzm{I*QdVL8YU8VrZ=Po?k zhN$*qz%UtEa#}mhiQjW!+C$w@WA`78>vfIq zwqa=$*pn+wyU~rsB>G$$Br5=Q=z^ReAz}A1cb6F2&G^@|8?(Cvt?Cv}x#fHs z3jlQG$BkP>W^oi)LVPpFqvmq0X5L<-0sV~NswqBUC$|R5X`=;v`<>e z;+BK!@0=R>@PHD@AI&3BdXl!-r;hr z3tqp*Jl|`Fz9YJM0qFCANV#Y9R|aC(iTWGaUOM{0vL#|`M89!HcR`i3z6=meT1c(z1Gdvb}9dnSkY8> z8_tBd9##G*?T2qqO$IQhzK3hDr!WRxoKe%F2q}Bi+Zof7RovTq#@K(>Gjs-P?G0oOcn^&g-0 zfA3{dI-z~)+|Ldm1NAWA^S-&>kcAj>K_$6nii}1wGQkSFT4vF+Jh*BSuufHYp_hmR53^|AmOO`NzV-egX1M`o1Z!`=`Y>-8q z#QtTy)3P=(ER?>pF(i*1Rw^0c=o`79jC27;pYe`fb{csYJ7RJ^EME3yh=}mied^f! z=H>aA`LjX#KC-FP7`W`q1Ys=D7v_W<_v0FuR~lz~JnnuDU?F7nVAI1%(1FPef8@Ap z>|oaVaOhjN9Re~^Dc$>6C=NLpczu$Ln|$BE5FR_3_V&MQGWls^EJ1HP=bYi|r^!Ih zw4(E|5~YdK4TiFkNncV*+VO8vY_tPGI@6#l`_u2sWxxFiFds}B>3qudCsUi#REX0! z3jo=LK}2ms*T3R_3~!8eOhFvy=o+_OpM*?tx=c@&B(490K0Tk7W|>)dJk#Smvr@va z%7rLrd;BUuj{bZWz)b-@xlDf@8XiwqM?#SBAQ9hLri9tkKDTrM%4Km1%>^waldh43 zXe0oUn-p3XV5Sd!9-cmzLiags{#xU_*loIKOZsdgeGwi$=!^<{f#SWOr1T)l*b!Ox zi)z^yYrBgYF^E9bh<}@F z66#VIJY{;~qj|is{L%J&XTl!!VgJyvQ5kJWG8yU$M^D3#s} zJE!{Y&~i9dzo2Mr|J3+weaMh9tLMM7%o6}&^rVE+m0)HS;M@-f!g+7oJ_zwCQ$xwK z?eF1~&23d6Y^+xKS^8YAV5=-g+NKyiKn-?XH4(CQ`j__j}-Yf zF(0kQSYsKhE&g-yozh)8!S@eqOSc z!a4b4&1|e+T>JPZ?Y1`C7Y$Fp=V7XD+%_*VZyYq!vf<;)mNhkYP0c>=pjHLr9NAB@ zKYjUNU--#9#9}%$ZD*b%y;%fvhLB$Lr+< zjjLfCRuL|yT%68$QVC>&)QY;V_??!^ZQZh+i|?z&r1Cm{-BMmply*mb|JbdLhGXtO zcR?4-04!95HL-))qPVP++eNBaysa}^zU+IJpO-{Ot~vBBi@Uqy1RI#<`aXb@7js7d zT9k$5bPHJAu9bi*>?MW%ih14tXnm@@F?^a=703ze1A>KQRl~b|NOQZb3_WM!?b1Cy z(gkB*Yo)E=a#$Ql<&Y}c%<~8;wv7$QFZ;_Zz+@Ohf_8q$Rf7eoz@zcK}Wx3 zq4xnV@9;UdF?7zY2yy?5{b|Vcsyse@r?BD3l`DnKLMkx`L@&jwa@oCDqIxAeX>)KG zTQ&T0m7vhCuvq$vvo0scATf;ZKC5b{a8&$YI6EoMZ*x1BXo;{ZK#psnqDN&)5l#YgMbKC9`pS%O#up z5Rs`Yv?oc8dUmJ;2;1_cDNY?F=QWCB$!{a%P5J7#p`xr-@VC6`TvGXA>=$wTJe;2+ zDAtYX31l<4E`EtR|D9G%xESLj%+Dp&)x=&n7e?o0{+N`Vb}jdvoPXpHn8uA%69_~& zm@oVWFbUF!7Zx>P|KRZ*odOu%SE>B1!XxrW=4a|P2~2Xw-PbW7KbhSapAJ- zS||s{1&%pW1)rm5GHzKp_CxAP4ZXop^b7{kja@(Wy?5`pRfGBNYH43Hnsclmbtuj= z1bD?PaCwu1apZt7(UHUt)xN>v|3Y;heGq*7POz8rucp?uC%ZgcgI|~xf9mQ7876<> z{3^m=47V%({=PKkN{g+Xj>UmtO6~erxj|k={QQDMi?4=+=+BgfM0tMP%TXlqLE&+N_hwtxYH~`HOLjBsvhLafe!+k7vAT>WG0#d}WF- zV6-ypqyBO)PhNLa1z5auH43fx5*zLuF8U}C#v)&;l536-MEwBZc##Yd>bbWbN0=V+ zk>7s9VKEG=-iLJ*H2wcR}YVV;g~%Hv#~`27ozn zCcGP%2wyieA1hAf!D#r;|gnLEtixOMSU^VvSN~zpi zeQWyu3M1aEZ%i12dNXDIR@$i&36I~dXM8!CB;PGl=B6N>G=mwUyHuIJ$J54J1^%W$ zWN)|jIo%U8&O{$|O2w22(OtFZ4iSg z;MaO(k?YDXzybS=Ey1*nsRDm>)!^#W!)v1r|0!-_sb}hI|GQ>}Z_W%7Q*+%^&yY5e zmyqai;KEC?XQw>MKqBB#+y+}sy-w!mB%KA;WM1{_O{RM>0~=3DPvi<*#CK`@DLhSc zUIQlfv7aCCGn%Y`27&hbMSO1&+Gu%q1nOUi&biQSXf3(a8G^m6^hd)(ZOEx{Kj>D& z`ipSAU!^m&m35!ioDXZKn)N-(w!+r$51g7Dyn9nZ<@ytU>OJxC_Cej%Cp)%zwWH%R z2UQw=+euZofLJtaBt@6B?<(%a7-6!_;YzeC^6gjb%^`!o9=(T$@V{yq|JZ(r8l^P~2s#Jx?I zd&(}de74ZQU-uLKR~JHd_RwvvArX|2!|ssJdtOXfi0=gwt15q6`Tr=Z zDWu1At!h`OkT`VI4ADvwvM}f8R1M*BLyg^q#}lEq@t1{q0z~-2#NnakUQbWBp;3)t zi2X3Z@lapIa0ddS1s~RTA;2#)Ttf;KMGu$1EVIBB*(4F6?-sET9Z?$nG_5YeWF^cL z9&y7uLd8u$H6~2GBFuOtLK_}ccOuM@kMztdY$-ErOa&f6)q*b68fZIL*oenxWq-(#5Sz8AzxsJ zb#TDP#1Yc}QTOihvDeVqvhCUO35X1OGIJ&QXazx;MK~@Yp3~V|Nl8`}NeMCj>r&ip zBqR+$run4s*;q$@biSE^eZXzT}G* zwB&li1l{WP;$1a3u^NgJK0vMV#G4d4iXQ-y{;5awRO5-%bIrVpv8eyR?=v9O1pd|* z!PdAPy@2P=-{Sv!3E-0lW9(T@5uV-6es+os5+UBkIAF16ERtjp!VAPp0Pzq&+*lxR zA~Wb9GuR!=hXwcuEZk(QGAWIw%ja-Iu`t&@`1sEa}EGeDo6m} zJVnP~cX8|Vn-$qP6*k;eHt2S*^ew=;Tv_E?|XC>NwAE za1LYb%Uo{;nt`_LQ5fJqhvyjqP!nc+!wjIG%H!$aJYkZ)m;Lis zvh&q_#7>_Qs^P-4N@FLoPoBa*AHfy$lM5^msL#DT;-K`s3He3$fNd|a)8@Ajq6lYh z#0CiU!3L2^6vfI@vVOR>7h+S%h1m=M8FvzW2w)F^1bMiGZ4pKuUZzku{=-!yEFA@d zV{ou&y5dF(DwYN>s@Hy?4|QmOmKGNER^~}aV@ZJeBYg4rr{^kh#d6nS>!6aX3l0DQ zbIw*Z$Q;+=1;XID15?iprxtNl`I#G)jsbAXL%_O9kcW&~W1w!!K)&d_g}7RO)!|rtcx(S5*#SwwITEGP;bY_)i+y$368ia(t3g z9vF9$td2?TQw~4Wic~LmeNu@bS0==j&y9P}3PoGXK4oPmyR=Bif&1$l!{U?J{w7>6^=l@7ZmLwROY zTM?kUIthJ1tzG{K6$wRdslSv^fSOUklZt=}89?CSi8x3-3Ca|pE~rso#Z@jJ0*~n- zgGx}HMGkISmH4M~V&33JF&23&B#udvN7TcX^sCHWs~~}|K7ZAe{l+2J!Ubh0H$DP0 zbycM5A+A(K4cS+ZGb-Zmy^@k`IJLuZ=_Y5cIE#Z4fVQu{n_h1w_~QE2R~GCI-O;ac zb+0Anl)2SV%D?gdj`3`pk!^^bSw!dD9rCbmq06O7KTvQ0h4%C$n z#l@f1?Dy8X`Rd-UU|m7Z2eO=*^?+Sk6PTnpA_lVunq34-Sv_il{a(LR5gv$;@^tjtc`#BmZSl#z1k$=0E;8O`|?@MRi4cX54}l{E){IR6 zYW@(|Z1wrA2ee+&)wdmBC3Qo;6apo8l>JEx^QyER?5m<+i^ zg;+4&+1t08)VHh#zG=-ZUyE=4cJFuP2M70_)(^WDtBUK`LlG`#AWuJdxGfKTUroZ&}} z6l5fT7~zCk&wh+!dKy_^yPR%y+2IZ1OG^Wmy24@IkBqw!g|F7qLR=}`N)fQ9LX`Qq zr(ypvy#V44fM5f|EwNDHwMS=RFc$zGOM*8gz`d#cE(FySDmz;pZx91`fQ9Xa^?K(G z^t~JCt?u@tn--tBP0D*x{tqDA_f%EO|CZ=Ky9fNY7xu2-o8IqD>WY`dc67mtya&7n z1~#e(hLD5fi1ZL+lCr3qwfw~ zXP*45c*NewV5ia9J87Tali5{|fL3C`G6T(Vgt?*4Ue5rG21i-yMz5XxLb&+F^l0=T zmcl1Ln7+tivJET(&X0BYApH7Y%MO$!AxikOP2q#UJjAHecUX<{kmo zm!N2y=G(SY^`8yU6++fyJzS#CFfq?Dt~FSJT?{}#v$_h#x0RvHC!S4}&Cqxn=lI)0 zK^P)&*yWjv+N7(^7+P*G_4~lo_o4Uq)^Q@dW$2$ou{Il?J5R;RO4DdQx7wN6Sn_>= zJ`gZ1k2>$buja$`^ZoR%Bl1%a@b(Jt_IvXv9bO{@iBI7KOJP2XSq|GC2y=J%hBM3g zPblLq=T1EG6A8&a#B=?Ch}FvAJSKXkpz^(D0$4%6SR6zh_?gd<4QkOx_{>Q>G69E> z_z0J2ogXioKOCEvKc4@33qhu_jZpbVUeC23ahN|Liq{!FJ5yJ~h*!3w7@MoYj(9e!2}aTU7=RVV$Z)uI&kc-PggI^_El3>XaEfsw`44|Gpv&$54L z;x<=|fW)H#tIu`>S9y0hO4Ern${Sv$`nX*1#xBfABdK!(+s$C12ja~GQJ4RO2Q*Fw>k38*hL_Ag}CS{!m!9y3Me^^w@&w+lWe^7}A< z=iktd@f?OhKr!$r26dM)%eqBI|74(oPGa~!{@JeG+PSlHa%RmTL|M%A_u4hikK|nj zh26sxb&ZDc|HIdAv8SxCGxycIAvJk#<7gA~?>)4qKOS?AM=;4_=okHs82 z_`x&Z8RzR~D<%nEaO(2g{(_k^&Vg=Cb#@Q)Yv*sT&DCF^+M9&4 z@vE0)-Eh~B=Noqa>|N2%{Y>o2PqXYl5?3?jT{tC;2!i$YJOtaO6hZm&*t)! z;Hyo$_EM0bLG`$zZoqG9s8iP(sO1ZcIU_cems@-Lo!8Fx`+Jk;e|Elh9m<7m+**U# zdCUx*e7fg4Q&^MAbtk9k-U5_;H*GaYd1ob84odCcNXkEZd$O}xOHspNo~7k8YX}o` zu>EMb-0tQ3z4^~ht!+Pl8)`$Q+L4XN&sbiD?Vz*4NX1D%3ZK#PHo( z1#Bsf*v@utfT*XbKJc>x8!zsqb9~L1O><}Z;BNB8s*3ppB^vhCW%(q1#a@YTpV{?0 z2NcQ;AP+me#7;8jJ{jU7C0_N2X0ZOgeCjn(kx#00G`UaawC>9Se;nvin(^{LXPVmj zjy|X9_V??iTFjm6sTR&JkR}VGx0IMgDytJqJcfc(lIB-Vmo}IiWq*oN)N29b1kX%> zOy5DIZMlSqD+px^3oDBZVJowY$~+&=?>lOyGa&AF`V;@kMFQ@4XM73%Df-6RDe=J?DpC~S?L>FkCz{LfSjVIK$ts674DXoh!lv$d-$a2ZJ*B<_rANf)OQJa zOiztb{Dph&s`x(QeltTrg~i+n^(YUEgGjWrz5x zxye2~b?96ydq>%wlV2D9bMo0T-wXHhwpa0G(dg6w(|aWb+IMCbuU@z_|HBJ*C!n@6S!;eZUPJ-+fA|Xm004gfFMlC) znM!F3XGLHyU#93nP#{*8tjl#pALDsY&=@yI(f0r0FKnpa>S4V_k}bQqj&~tzYc=(* zFK@RzRQ3KjdhoJ*xIp{wAEnUY+gc%)(`23k4VBj^62$@yr-@TE?k(1?c*Ep;g13D~ z9Jfb(^+f%jpBJf_wa?8<+!x0Fe%-9RU19!PAcCAR{mx%EvQ^Nyi7}fG%|-v)qvSRP zKb-k_)XP;joiaP*{`nL+TLn2%;;!*zgE3XGc}?$#XC2h^m5RvJD!2XPGBB(1PY1A z^V5y*|L%<%+I9~m0a_c=Z7^Q#)_?xqtAFaJ16kX7XABP(!DHaH^y^7%yWcZ&f@e5R za{)^g3(4L`ls%!RxF{*d7+6&LY#2(5PgLuKd6V6J-n%iFXkn3eIx&hb&kg){;6DIml{3R{%Hs5f{U-~kLU6{*)D#d?cS1t5E;G| zN~CbyN=BS-nsaf|GThyTbt0k-e0j@N<~D|vZmOXLS^4?m^Hb)%Yvi zJLtZtIgRn(2VSO~LDQ+pFIj+TyBe?hgkC)>MxM z$kyH-&kq!6ylC;F)^V=&FINk*N%o=F)^t2?Nq^3ZbB+hc8ZQct-Y8Ym5+G;f-|zm+ zJ=+fMwqdlf%lB_rq0%d`{ep{|oiW1)#RVt?>z+yA`nZ1+Ct zU$<2gm-U}5IpMKyUfq&}S=;Y@t zimkwfBL|%rZTkz__U&w^G#Q}>D=&tF4@I(wt^c;l3xafx9%qc%?e=_TGLPA3kH_@z zEF+zUOl&mMZKV0(lqv3)_n=4m=Ug@|{Dr<|3M9lE{b;{K_*g1g^O|oRESRJ~EyS%_ zHu3Y=5{Y{KBd{Zfuz&$Q;a}Kjbh!e`W1kWIdBp(BjF##YS##O!V;#b(ZBYVsb!iXtmJ*3U#hg|&eNTBu zPO||qo;|-yXCH0F+oyNRmY8+D?h#M)@N^R5MRcp}>IsZjIPy>lk!Nmfr#>5yI3*h! zdCgpros!ofWE+~Gc(*uY+{{VPtG&na*~b)e`r_#emA$r)M$#-QSa@FH`mU#^DwU@j z%1ksFS-ss3KJTD&F-tU2Poh25ju?Kv{L0`x+j`Xf7(F4hLiBZ3aY2?|1GQs?XOD6^ z9@<`AR?IyZkkS8~!-zPo^19dLdXQ4y!ucQ8^jPM3AI0C;XljYz0G=KuutS00?hv3# z!_@rRc_;P_fmdy)j+L|mrQfbfVB5%0v7KTO3n{%NLhNn*$3@(71DCY-R1#TkP;iNr zI+rG+qMq%PaJ{ZjQ#h2(d^B3Dm?&*>|DJj5@{Ljr7kBmb!hw+E^A(CGWv;}>nLmG) zR$j0zZCuD^UO7ElAuTyzlpASWR{Oj3vQM3dsLFh#v^n znEKhZ@Ybhi!?%25%00~L-JeM-dQy0GlfI|T|lD0k07V*qVaZ(;E9l@)nPEtOPEJ)dO_dt*7DZ;NY;}b zd=d#xM`8z(vOG0WkhMj>eI1PX(1VPZc}sGp7e4xM%#|;k&!`VcWStc(%YR&@JaErSM0Kp;ns501 zL=XL6IbU0$z9ps?U3vPGgcj|6G`keyEFb^nnngD@eS98!-EaQ;ib|xkV%-D6KNh}X zo7+^glcy}4ZhqT>N1Y>q672W?&H$Ju8l*Mej#XS~JO2$;PuJ$Z&Tc+;|A1F?x|SDT z%JL2SS+QBuY8XC36q#=Ofi2V;JQDYwc^Ic=Dx47&y)-uy;|KqJ66tm({H4(Ffn8VI z@BHMBg$Y2-QU4g7ntMwfGn=6m$@w0pY8vDmcgUti4MSQV{j?qqfUpq4et13fFaDVW zIfEnDH7>64<25F0w5q7QB`golT}G;4Q~6eAK?;5(^tKk6pM+!KkiU#t5ltSA#4a6G zbPMnf`zf<6p1V0gRUcJ214{WMOXJ{l&r0L#;#%ZWXE`$LoLfPhg?Ucm78JAU4#(<` z{}h@nc#MT&d}hJm&zHNvQK1Rb9&!TxQYp%;EJ@*1yzot%F2yK?ge8p+2OTRP&v}YI zy=na6@GJAux%@(;r(4Hv+G!G#Bur-yMe4N_5rFbhv2BT{Bhmx5UODahX9;L0Be##eY07P0vizUdnB zyAhs9A!-9S zv~vSIg5>s?5MeM5yA%`g7!SWX9&xEBB7zQc?SlE^NhUkMIs+KChhPX$e}*iMf!F4X z(DQ>D#Y7mnMclKFVEY)gRSFBH!GiFl5kip7K5&N$vy_K@;PlhlM;VkcM}*kp z7&Di z3l{ox6$&*2F3>=>YCtUl>~0Tp#X=p(ZjwcoY#*PEk+Ksh@JFlh&#zdAW8p7xIVA^} zve+DSSq{VusO<%XujZi0(3rchMt^8?uSJV{_Im{C1Akr~(>*=E(vq@zwfF!i_0M@H zWny@c?`98FX9L?d0GP~j6OV8uK_;4?6TUnb*m^#9KwhxPEO5{3sm$-|%@1CNGGd_* zXdp3M!FDeQLU?XY$UXAMy={iRTg|0ip}bSgE8x#rZO(U~=c5lP=g7dwYL4v{&`kow zd^W+J@?5CT0z-$2u;q)s%@#Mx?~J9KeoK*_q+ljt?(!greZk9SNFoVxivDVxU!|fXEPxU7lAMX5O7<}#QIaNKbBeMLq1`et^SLasLKloYl&N0E`lUD-i-;h> z+GA>haL{)#wF;%RYd>o@7(kyq%wPUxh;QXYtXm)zhQY$j+Fpvzy#$L=?~<@QSj-53 z$|fVC3GkDG_5W2~6JY^NJX|%mUOm1Z;Jzg9mypMy%efEm2%0zrcsyW0%nyLWxMy~= zDHtNWB^tS<0F5}sE#2};?(ZwbZ*}KA>NN!4M9}M#u;JB||CF~N-M>I2p~15XLS%0g z^moZ>YA}h9B?Q8=vB(=Q^m!?-=l~l9d+B6L&E#9yo9w2zmZp<};aLFU4Hh*{2#PdB zdJ10h#=)|hn)6zkCnby`0r-s@P!|9m$pB)B2-+;a00SS=@+#~moNNyXq}05*+7h4J z@@AqblF{^rh)TODjInR_8-J5I-rVydjUxLt7XP;FFF^XQr0{L620F`~{SBVIn)*KM z+xuyDwn`=iF{jZw{{rz%kiCw<{sE8dt>a19LKNWGihvIo;KOd zcCmkegl4;#Q2To_`{pjY=oxG?9!VpkW@wN0yg?JQZn|^atA)JbAU*4f$Ypu#rLT%V zaEgYWoegvTte;dqZUOFdjcVV$Z<>9y@I-A<^laaKyj6|Y8nrEPingub&HW+ZQO}P^ z#og>s?iJT!F9gZqStRU1-nY7)tbAnUS@_7^fz{pq##$m+;ASHSs{@M?mPHg`abQ+} z1js&?upU0TFDte;rJ5yOs8<|^%?{E&ogS(3kK?&GKMsT*AR+r;sBVTHzEkfr16lnK zIZWXkCL?=s=-2;{^H|g^*~@b$Z*4%j;E(R7A%08q~&$_CyV;637a z66&|g0rcSw1A|xmZ=0{^|!(sVJD6vo&h6uyl7-#hwuT=c6 zzLu>A<}y1z80BFoZSq0N#EK5NL4dsIVY?BCettAjcyV&Y9P2uO+8~OsGLS$1!IS8) z`xKXZAKh;gp+RLZFUq%Z20E-GO>>=tgiS_Atj_)SbDwFu{zvZTr@zeV!ZyY|K{xUVQ zhMENht8dPtA;(j5w`UN+Gs73%Rt9H|r=~fFM8X~Bpy=tH`!h?=XM|g4Fv9bk*OVL* z1-OIfV6}5>({rH)Zf9GknHT5h&E{2-=HdAZ%Tx2{jRmod=`yj!)A@5Ut+R?>W^uxE z3fJb9j~As|=VXUwFW*_xM$cM5nCELKc?15t~6I=3nHbJ%F|!uXG0Uzfs@} z$z6Lf2%9`#<%1^_3RJOiSB;(>&vSb=*}85yww?x(-{({i*=n4mA{RDBmyTetxX8G2t$nfaA-3L7k&I-+LpZGjAN`1p<(|xwj@mu$u8`stAw`BSz*lI z=}j++CDCJxj|us5=gtQ$ClRLUW;$9%;v6UDCFvmzb}i9FYxDOT;VRjJjb!1f7A?#U zo-?b)#9(UYn(1$3$ZyGz?GSAAMdojO$S-I4okH6kXF8&dv}2SEGv3^}i}|zi1VkrK z#xYO@3g+9$m^IY|arA^wDaI!Fv$C zvrRyM1WVuE0sw{K=?}2~$C0c(98{8>1~r-2kK>k3>fcSAGVuMq{1JTRWkF9cQiQIT zJyLQddDpTvB32$xLSpVZ(z37r7|Ih5I*!0{>FNSN6x5=yXvMtDKD?E67tVJ-*BkE|Zyw$8H zY{?EL?HO#cXEg59wyv?u)#6@V7EgZ1f|QSNHfNWqun$L>hq-qP@+=l2F`D@$#r;}u zS(S_xI!5)9HJ7DWIn9U>R}G{5BnAwLXf^%I5}Rktdrw6MDfUUfc&FGeR|)*?SHMVq zW*VPocaY3#{O;T%a^-YJ10@yOEPh?Fhw$pUsrJ$ zRk~YhamDuGN982-_E}{lA*5nKO1t+b{U&6Ypf_vGzw~i3a}4&SzOg!aVthC>@GFhg&bw*CEw}j z>EtDLTy*@ptp8SbG3R8&i`z^neeV|neC+0th52|9o31`&mX65b)Wel zs1qo>KBir_dyo1$T+Zz~;of+Z{arBj{S$@xrVcVxU*3Kn z%<1OLuAc5Hd@}EeT51ie2%`e3SpmX8+dB|N-*U9^Vp{5zNj@tY1&4R7TKshun0(E@ zI5&mLqd-J^;IyU#$ebLdL`I>~Is|vvno=F5$lj>OVe%twuVH?RAl<(=A^p*9$PrOr zK~FnzfjW(Y=KBCdwGeTiK!Q>k$fsF(f{qo9HD?$Bqt-ah99r^YKCs}dJREE0s`Hzp z&PTwx0MZIV?a8H}#TGM96u9Y1p#hNl%E)LtAxei+-s7CZEK6GtO(8BxaB@Fapu5}P zLhGlPUkkHBzk2k$KZ-`0>WhMbD=;0-RI;MxqS383gr8?|#<%e(4qVJD!z<=7ywwu5 zvsqWvWm3}o6FOLg2)bNo5=+#0m{^(0>tOCP8L|`6vbG0(_amI1VVD(5P38u@w|;J? zUQTy9w;p3Svnq;Wms&=Enj_`b;PU#3ZBcu1 zyYFgR>$~EAKMkS{N=P6~jTrCcFOlbh94vhVJNPu`*nfGtR?=_OYn5Cp@DQ5E zf8HBZ6@HxeenIzAN;Wh{csRskyq&*wHj3BrMzPRToHj!z!ELs^v108 zu9TbVg=|9b`!-?Jkc2tr8FQ@oI$)nI0B)aXT?R@@itn*lC^2q z6iZDLtFDGfM_p)Y!-No7c?M=-XPv*Wg@M5~gX+)*EpNlFPXkPiA&$zo(<7BKgD+I~ zWS|!k3NiiOdX`N4NEN@hY8T}Wo6j;?!T4wVY66xnxraBxUZrPVmptt{g?mA_b!d=V z8+GUXHcnCWRNIk)D~9RVIC8U|jlHa&&8kj6o57v8!@x;OeTvpxsD`V?PWTJ{}dBH;2yU>2$&>#F{YAB)Z(1=eW@mc5-Fek-s~u z%j%6nQs`Yy;c)PYEGmmckcPbC-aWfZmgcrAxakHydJZR8tTBee!G98>zB&#DlAj+yj1i4f~V7VD{mHow%-Z!FBPs zzqe0qMm4rfoJ0nE{WiS=jiScZBzmac0;K*|3iVByRo z+3}Y%;ZHWPMtT23!15`v-RPl<>L>n}fUT^X0!NDftBhC~dfuKX@6;FJP8RRYRprRh zuN)D%nyC~Ht!b7oH%n3B&f%&k$~OwdMj9nv*{QHdymDX5?QmzVrP4m0b2Q*}^~cN) ziwFq@@Vz)Ro>MPfx{be&cAqSZ&mt$U5#b1dJS*(1JLGhU%OHRAiP)E|T5Uv$t}>6@rq29c`#SBT{bZ^!opDON^0s=^|6vPF&+LT-RtIv)C@Cd533$mKxN;MUpIrEJkso5dW@ zXIyvVweBEX^W28iPGvutcTdTEcqm<-a(H~C)4LvoGd0FO;NR! z`Ba+KUQu~1FPK5g4+^%$mOCWbx@DGeM*J*Ef5TL#zj&N&*MzBt?a z;w5cT=2`W}$$+xp0hRtURg$Xbhc%X0_DJ>PcYjsuze-83xjgB}Uehod9}E8?Kj`=w!heH{kPoz5Fx2)ulsa=@-H5Molao1!+6&i?_r_THZX@K zwf!Te0Y~nV>!j6n^9LI1BKh6yU-crweTS5f(ia}7%_NBb9+)9I{T@_ZI79D0_v9~D zb%(w{uYq&B?pP(6u$rhWAh}EpwmOxrsX)x+=t^yt@JwwK2O|M*W;jJqan8K|W@6^~ z^T7)<@J^Kp{?Lxy?~lKthV(_S`=;Sbe$0iv84~QQ>g*!fX?Sk8=b5ASLd3~o&9#b} zH_Bu0p6(7PY6DLU=$Ahq{K$_+4}y!e0T-Co;y3lycD#2s zF4x~5`lNWtc}y(r@^-I|KXXNWKKjYk7G+%}_UWBby&cq_nfo3|m*SBlN8oVMx#0ap zc8Okamq-LCSoQ4iKM%~Udiweiwd2~<{twf}qkq5cIQUs@!U5Q28mLPYeIgY2>)0;c zeBRmPmUqm*{H> zWje<;fcV{>LiCfS4lqvy|9j3D7G}m*40kOx@|=(%(p9QMVDcr4>S7_6_ezHz*E~+o z(~hIjw+U7)F49ZR_RJdBKYdpp>1JFk@I?xywx}a!6Jf?S_ehmx7M$Ri2jf>*(f7^s zN+jt!3C`@3WO*c#PUkq8zTN*QpX8JfIBF-t26h`JeS%4`J}cFstAktAP;0ZBX{c7PvEbiwSmvE6}yN<>y4o0qx5A|qNn zldu0MJLQ>p@6y>RdcPgiI}1aXC|H}*eKYY;+ZKucEX;u3oZbeJ(l4W7BT*{YB20uI zNlKW4^1S!?+O8i>>0LHcgDuiKZ%MxxKN3s!VJmsQneaoi&0*MavR7i5XcJy5aV>!S zIS+h$;SyHBRQGX_97>tT5GV8Lgm6U>FSnGs@;#0uEm|>7vg|Uit~qaZy0#Q*MGaxg zk=dSJsK~wga>vFzE~%qPeXv5{!q1_kV6KXZ$8MST)IUiraTUwnmN9_Ojb<76l<6f_ zm?XyCsNNnGyY+PSx}MC9s_C7JRu-!^vZ-SQufAA2$W>V+=ZrkB{e--^E^}j3&7vsD z=cP-=FFR1YW!toc%>zjfyNsN!gi)UwosS;Q92yhVmMYYHe`FlKu$o=EqEeY}v3lvO ztZikgWkD-@GGSZi+a&X68H-6SOvEzhs&bM8(s)V6O1mmCG-1~S`#k@~J&$i6W_Lkv zm%svnxm`0*tENunS`V+$4484T#w-vWc-;2=S=QH+@2+lYoRPm4Ix3289t{h6%>d3n zo@?EE9~Qh982Oh?KO#akLW-!%e&aG&q=$qscMdbpJsd6V0zOy{-h0X+0anjy z1K-8{;?JUuovHv;U9I_b=j-g(57Wv}En3(iKI;A{3jGQ#2umeMXit-%=c^SGT{ZQD zzP9=*j;16$e*E{_ydO!us|M{QvG9Gx;7Q#Drw_0M@ffv~`j#3UIZv|5WCy8B{#o0@ z;1dforUqS#wS`D&C%s`-|5y1Fud<_9NhdGy3&rg03Wqw+@=xq{YYaJ{J%tt)a2Pd( zq^*Z8QVV>im={#ztcH+|ZcfARy;M8D6*ih!{88mG?b?*zA{lAUa%NH@=+a)XhI~gH zNq(JRvALM5`tzJI_N-!@W^mb4VjOjN19>tB=8%L?;vLIr8YhY?~r&$8`h@LLU;bm#jGqV>BgD$rgJV{+JT)F$?p>!%nh7M@Biy-6AU8qN-yswc}@~F|iRnv4(T8oGI~%)cD6# zs4oTX3M9D7CkXAw@P?vSsHkrg_O~>|91gwBP=1+pK64)!c8mXH&G7^ab0Ilp=EM}F0o&nH~3#NSh@%8%=LfI#TZLl~s z;375oZx0AeW^t#(Z(?EIWSBSoYG(`?W`Yi#&hTt;5G@uv+6JP`l)QHI7q(UCk z;m#DOH#OrCb2j&UGbRt0r@o5ZCb;!nNoV3ht{EUP^enm!c0?^-Eff|_hDyl;6V1sM zm67*={P|d{k~GJ$J%+T*YC@+N&LU&m@?oF2r%rP|B0vqjU{P3zJPDAd16=m-8EO3b z=rzEZQmBS)szd*qMgC(H%@Z*D_IX=oSQ*eI?kW!W`R5lt!OY~@qKL`vSV$BFN~A*X z;h~ZNMKCYQrDEYpVn9C$sjwi%tHT2J3gZvOPrGpiQ{Z>#@N_ynjt(^?KwT)%|2jk2 z7je#&z(5>UO+Ev7`!PC>AC%46R;reDA{a}?&{E^5kT?c}2*BLmmffF(8Z+-4w!Uc`z4Hdtvul@uGNQ3c1xNrCRS zkSHqT&vmkqSXnrzfu)_RpO1F7@$2oUlgbQqWgB5H|XDNxRlG0UaV$xRoOV6gQ%x`MOJ(ahh3Xf;tYIko%`Ch<@K#^ukZYQZ7=x7 zQR9vC%{O-g-&_cFb<^Oy*Ybu_lv}n0`}yY^H@*`xqKzF!jTzRi0qh|PuRIgR8Uxw+ zFxaNB7SEwwOq6Vs1E?ulqe-t28~?2_Q5Ku_qH#zCo7wUvRggE{5SttS=K0kVSHN$R z*_#LMHkY?>;1vs#W%;j&yA}q%q2_YK6{Y~kgEzAYsjv5?SGE{ ziwH7os2DC53?C7)O_v%e=XD>+HF~%TCR(?tMfPPB-l(#yzxVX(YtMEH>u}{oL$1<^ zM(`)}t^VETH=7)|PK?)JwDor&mkQ|dv-+wWx? z$IrArHCDAKh9)Dp9LsiEVLoB)4N$VHA^?MTr&<6-Oy{|6@rcuvGSl8DTZ)q6xTcry=lgP%%nrK*}4P zq~+MOpndTeUSfEC@WgP&?|~CYrSp}3b6^dFgZXqp3xCOHH+@$nv(Cg_DKpf6?lg2^ zh7^ucaq^!)=L)<%IAy3GE6HOPuw9v25E2$)L7C+eC`)!a)%U4r(|*YYvqly5`z@1Ouq8^KT zXg5suErMHAG$iU_vq$yUpZ01}r!*B6I>;0*S08%KnNPh# zU2Dk2Vem9k!n9i+?L7b%{LoO>3QqY=J8yB2n=gSOrS{^nEu9M8av#p55A27AeI62Y zJ;0woK5x7%ijglm(W?!z5EI|X$!iCRaN=ai2 zQielQi*bTN%C-+b#KMGG9hChDNk7EZSL4DLPBf~1oJw7-<&pTi_WGB^==z!F{J-Y! z(iQ{PY5nKlz4+MuIH;<(qd~KB@Wal(@=q=2-E?X+W{_2z zqcJufohY*KN2^jB`xn)Gj&q$*=|VO-*x3xZ>I%09+VC`%Fi}$RuY_Z^c5UV#42eZE z9K;@jA45Q8*|}oCQ*CW%UHLE^m;K3~_Bu`W3?bGDdzjjO8#as>AsG~Zz9f`8gGfF1 z9oMbg6EIB|b_XTo{@nRSHfb{TkIRECp0V65Q5V@L?92`U{7ZrtBT)R>nRiyG9 z5r)WSz7#=EXeTjFU>_H6EB4>YUid-pv?8JEWZ$FX3?nREeTyd9W zNk!B~x77 zBQQeN!*Kif2=xH(HxpofZ%#ZTqZ=MzvBq(}#iHfO-l!6;A1=iL3u)9Sw9Ffj7KAV# z$|;5#6~czF1mKGhRM16UN(7WW!p3?L*M2g^n5NWEYxK(e0`h^f^<^d+qT1S9g_WIoC#+vO2vl|?y=)Ey5jDxj@ z<(wZHcDqE-PDL-qXfTD(yAToLLlEUB0Z=TJ4!ULC_5AlAphKp6U<=sI?Y?~%^H?;W zQNN2Ztz}lFXK~mLxqA|8Sagxv(Hbxk#!myFYSCJ(g}=^e(<&Z>^IEq+PFaYY(4gmA zL~HL*8F<>Lk)scr24d#xipvch|4!ti0p&f$+oHW*ik z_rE}zFG$PmjMcfJe!-ULGjp^M^Zrt^>d?=ySua+YNh5bX@kuC83kGVU)X6jN=B=qF zdS%e(LxG-xm7wkd&y!yMs~S7bJx!lQ?=1A$ortYG&zY6v(lZGZ#wTUZFrMRL8+@a? zltBC~=zv3I>4XdK{EJZQnKX@M4DS`G9%lE7^1jL1oYFC9uxts*1U0o~;f6=A^ZY z4~<4%hspBr&38qk-p+jZ^7W#RF}3W)rSmRJ0~wbdzxvER|Mu~fPB!t+i9*INGc7Vc z_2_@O{&A-9q^xf^M|q;`=}~=Dl8AJp{SH^pyjq>-cR&M6vnj7c ziyO;dU=v6Bos@nNdCG7l6RkYQduSNC)ct%u7;Ug7sx|AwF%;dct8?wr=t$SF+7>99 z4?rF6QZ(u4z2O9<36!KA^|ydsz|x-xwB(d>6_((BqsYQQO8@;RdRp=w?l zgAJ3yk<6p{(WjV469S696)O?i#GUfPWgf2Q$6xfDD!AS6U2t)hpi*+NF#X~8(rJgX z(sX%lUU9FJ5qhI|S$%rh_cEWyr;}6$E5p)MJx`WR)w{VLZPxsvHkHOVt9&h2{{h@8 zzS?VLSWz1TCOrlKCRzx5AqA(+x*Tg53g-(r>S)@~ptv^6#$(Pi?P(J#E#Ka!Q4}Vb zHohevv^|(%oOIYV$uBf#Hm)p5EE^)Nw(JtRL7MNSFFJ_TvC8$cDh$dCpJ%M?k;u4R z+dH5qKGa3kz2^R29q|WHn8q@~CE~H>(c%#+tTE#85-Gc)@n=$c#p2PAFO0>{#=}pM zG2#k?a$&oa{)@q{>cssoAUub)rEK>`@1EaMkjTWb|C+cbD)4p2?aT8S<<;BjYeNAm zFAoHQ!|B@}1Y8?KK2XDXG<@bG{w`jSU$r(;H1V84ASEI=J@n@7(>kIS?3r8-SDi3H z3e)kb8dl^E=I}o$O<;;~LexqKR5%Rk+AQF>Qg!sB8JZX+61-^-mp>+3eExo+EV1cn z)tAg3(PtxxAGShvU0Z*IEV%cGJ`K=r*>(_ZU-{mo%%r7`j9`QC;yp*-|DxHMK!qIY z&GfNpLP;$mX@vc*1+~2| znkS>{AXLjWrXw~S^G#al-T;ZgqzAYCyg;yw&mIo-(LS`th6voo({IDNCVJgMFI43J zqPv!8A!Cz5u1DdZT(SzFGhB_9-PjpGd^3gF9`&C2}l308y40t3BQ6%M!KECAxW+?<<`1T3TPf<2L^|V)Nd8>}_T4mzHvwCr|saPkgUgo=VLa z*psxJUhXqFbwz6YjH6;D9CBa2E(^Ktc)|HMa4Ys{PEm{Fc^h77n^QFm$0RfAWDmVf z>2Npiu~U~O)W9^jf1JlWAhtr{zKPtDMPBd911-jzBe5SAN$}&ZCn>yyppmmIHY5hF zVI1^&y5)0|!Z6X{kq84!TB29Yy$h;;%+5yFvz(GVZ4gmnp5An!AZp6#p$iA_bYu%1 z)!519^$;?f6%tq$UuIdr3`t8IV?TdA!BkGivLKhS^d7Iy57!LU660SLUs}q}e=fMb z^JZ9(K?{8Pj8?3{Bo})+A>4LLCB{Invzl#^s@R%`L+Br*7ql|Ttg49$UJus!>vSL4+ht=I;vHPF)bD17_do=g5nN6 z(Y5kL|NJ%u*X|LA3flZrn zzdGzCOy6Z%C+}^~o9{W%+G|=U?~$&}BBL$mqkL3Q<6E&mB-n*XtwKw(&KZx8gpBJe zENfmmO<#^9D1BN;5%cZXmz|s1y^Cn>H%>zPLqFvvEen^1BQDJ6hs4DzJlRe^Nbes_ za(DYVcfwlUEp^WQ&^DhLXb6+?5=D$(U#Y-ocJY5Ex<%*^7Ljz#?&c{pG`9J6#TSc* z+*Q;l6`xKXp&q^7k19JUT{3FuwO*yaq^J<#r=-zDucI{n5pyOTez#UV#PRymP2i;< zy%#c?;Cii4=N8JaU3)xF_*vXFb>wwul6q+FuS)vv8vUBeWHV*&qLR>eUSiPXw z?X4T`bic`-MDJ1#BI~T9laoQ+1U;f77E&Rr!dMAlMykBH^qJ$>(KcjSb&ii4_aF2hFkH==%|ZzP<0Gt{V7CrKZKXwd^unqe!nr5HReT~jw_RDtfR$bl zCmNi9Wqv{fKO{k>#iQ($!t8s)@6jhijCn)NoX~GGAW&KqBQ?@<1p4b~bWE=dw_H?$ zPgIh1WXfci-v}&W1YoAbauOjA0Ehz#Y>SC=0Hoy2*V;vT|X&oGYm{`c0RFs+EBn`b; z0=Y#6ThYMJC>jQ@EjM*MzGud9<_GA@dn#HdynULmI~nFe21f!=7cA5Y7r<$g%mYtg zgv+vkK>8_&Sdxbaoum$Wld3kAda~#GliSqB6#UbRx%q+OGL8)>jo7V=u);x(of1!7 zhVtQKS)}nJ1OBvwm5uy}$3E_(H2TN*Q(^fGs~tcw zEW>X-Bd1iIa5^*GIn#rX`IMB|OUX>MVcS&r=do;CG|}TXGGYlijTX0wX8 z(9qW3FWS~$=v~ij{rt$+6n5Vf;%l1SJYti;$L>VU{!u4$N&dx~byRVm;`1=e4tUPI za!!nCPBh&ur|(@(mK)1(R`%74Y@hU{J=c-4Pyfw|-gRc)_su)>&HLrdTtUg(hclCi z@EvEC0lXR?p0LrETf3fX#gI2mWoag|uxGP0;_|s~=JTrLHv+7J{rT*EEDb~!!D585 zI|=@f&Fii}{0?doKz1xi7%y2Vzq%>q$9DQ5Qg-?k|645UJ((~MS)_t=+5Pj1O&p>a zG=^As3ScmtF1+@q@D%zrgFOQZ_u5hgtw?-*h6cz0;+Of^zu^%KX7?R#7WLDBaSZY! z4&6sWuG4_;K=F5Mu?+3CBpVw;GGfZ~R)8OZX-W1w93<}RKpQ4o8hp@IYS47BH zblt4rORo4>bcR0mZ-%^B2{x%o{86{`6dHjdIhS!-+4CvE?9JF~E0Lp>3kIhb%fy#u z1SL?zm8{Bnbe?;_h_Aec`3TS)CyN-mv#Zv+n%Ay-)_T>~$k1nM&Z@ef7Z5FB ztDC<`zHqDVbaSmROAY6NE`uyuLpGF0p#G6V9qR`9imJQbK;2+XJ#M4klIcw%rPe|f zQh1GFF7YDA%?1pN3`V{S7C(}<3VSss!OwzeO&E~>Vz#IfC`fY* zZH=|5PwKI~Ub_inUETp0a0+R;&Q8rSW0|ItZ%byrS?_w{AT(xMn=B*53}9mM$jH zEAun&4PDSQzxUQlRz}zK+1<&!i)fiEx5ozF2gbd>QSn|$uuG+aEywbvh8AE}a~4G6 z-}B%-&nbEE2lx$eO%c-`Ll9GV#NAk4Nz>cWAp7W?UPgLAIjyEAo>e3xn{-(3W2e&R zarpWy<~A|DFMCr^#oiS@Z5Fk`NovV`b>>~6OR*%@)qOD0Eg-vU5tpK*GTK=4cWf7=s}K8O$Q2tN|2}L8cP688+$Lm;V3d6vB7OjehnPHi4P{*`|#%r!A>e-GP1dcZ=prwxiFlGD}diZ=R^gacA z^Z2#xf<4ZFP{~dW6@x_UT7?qqu*p;!F`r)SiFR_WRYXo&xa zp5!Me3avDw9;Y!oCto=|Nt;KZ7x^7y72WyeFQgK0w*S9V1^`H)7kK;saLVkvg3;pIp8qLZJ)G^`Ch5iq{ad!G zUHRZAtyy;4`M;bp#Q;-4tiR8<^cC(7zEFDlA;-qz^>7|8tbeo5ykxXM54t1qP@#0Z zSUE%f82YB{Q@Vxbow1vUgV#}c{4(rJ?lO1OTdIG0^gxviQDU=tnxl{Eq^5=Y3j&88-@}dvk+p^?ZC*F!9(}F}YtA*AUD9>^@+agLa`@%$>aZeDUK`EF z7#Jl?qN>3#%62lB`t~@C3^r~-A{!wn0Bi5~R-4N@lflX8s+uuuc_f6dMy zH6hC}(YYl1K0os{ECpDOJKx0=_Kk6OQ2e9Hl7q_7h0mT36D}>gSr}Vw&Hj2|_+t9* zfNZs?yn%Qug)XyEQS)KCxAKk5-6*k^#4&@j^|uK|>iYq^Mw#~fc?~cn3vVJ+%*tSyD$c=c+;OP+gLko|D!6levJ8ShxWZBwGU!(3AG>O%MbJMaHEKe^thLHwnz!Wpq#lHi zjr|(Y`3R-wO!Z2h&$Iva$@)RU6%ntT(Pv-SWwoQF8Uckt#<8?Rchi@sp3bef4*10 zIZ?@|88E)Im4&MN&GmBi$L}9bdx;R?cBrP>Mv{n#_YTWYknirttLN3%nr|hk(@M&R zTbJjL;6aB2Nl#RNtQLqg{h4YoX?gTvcm8i$&xLPuD8Jw1d20bZ8jJ6$edai*z}Zmt z5W$fEc%d+zKeNfDzUuMz3k-O}UTgLAk|TWc1>~xC>xuG6n24?s?6O`mvj#Qv^vVn~ zUx~Wbd(%kv!u%b&0Zf+jn_JY`6%s0VtqH@N9HiSK#E_J!!Ch4yh5ft8`s~%krAgDL zQUTV;Awsw8+w*a9vv-Yio@e}&ED=82*pu1xEJAiKA7MG0$5q$E>Q; zotgP|%r9JtiT39ZaKRRt??#IlUFg#Pr`q#}K<%5`(hx;iCz+90W{HCdua9YmjprlSP zoFiB}q5$co$9E;bsQ=jL`(u%vUH zEf_Z`QsMXMh9ehB_bm8E>CW}&@K65t9)H&ntpmB_d}0ee3=^8iMt1V=lcO)YP%V9< zFGnxsP6D#^7t6-em7Z4zAwKBQNzs>BFV>KZHt%kXSexRkYG1!PV-r&sq9`iPT?~Mw z<^M%Nr}y7BzTP^0%{|tFEXIxf1OfH`PHJ}bd{gIqRw0v3%5L*6PlHdDXR+bj_a9=s zPnjWdoY2I@SLwAoPo<3w&`Caf%R=9*k=nTvBH_9OJ_~-d&d=ne1Arv>3Ix3=B#*n6ec4VwXju>ZrZqVr5?*h`KW5Ew;Z#n7dwfAB_rEhUT^4uzl!PV1djoX} z>`wM|DAL^6kYYv%X2HIw{$m>{F* zZi!ye)`X;c_cDm4E);(CEaFOajM0G6n=*^QCWXL5P*0<$SD1UMqPm=@684E^v{t;` z>=JJ)@0IfeqoO-|OFUi17s(mwoAy)&9*5W{#r{WGLGj`|?|ddKNZWv`fAb9WWM7`i z!}-7gClHi)`OMpQL+0sYY(cycVoh9CewSXQWi&E9b{=JU6fX|gML;fzbU=$)wJ~wL z-D14kK!k$WA_{+qlV=lU)#c^O?mZNlJcib@=dibk-RQk~60N5s#(qTq0lsh|L{r;b zL`S*xi}AO)MQGyLVko}{e@C!xxhjxD-iHdE5W%XPt&eOB{iug-u&dZtTyL!-MZ}v~ z&sdAH*eIPc@1;U65kpwV>bN1%A+{9AXXao^=yVM+LI(G9>j6P^_8>h<4ZE~nL;Ns7 zH+v>aeu>b}&!vu%pz6}4BF>3D^3jVhy}upI#NaN0C3Mt9Vm=FUKy9dr6w#!>z+Ops zE7HR@pWZtmiq~uXexv=*#rj3sF(zaw!``d@o)#*tIzXTaeW3E^TkjD(58Y67)+KLc9Gi9Ge2TW&{6F6~EqxU;PNm?3jNs1)&?h6o^wmI+69F>`UiO9&DX?Nf(D&6KIku1mYzUWL2#=mveqnGT z$*b{8$oy3(18*o;DfC+uOFuCbg-67a(RL)XI}M0(gFurJ!(T!X+o8J$U|XKBIy1fv zF8{h!iz_#vkEtQTjS$t|Feon+>+RjnWx<5!*Ue;ecZ0fNpoWwFOvyp)ljy39D8J08 zcNyVO5NMtd;VK#pkqf&gZ*ILBfnN)}s~5>c*Ykbg6k0<-{3J)}l)!=rY;FXo8y4z; zSA1rPjsr!o)XFi`Ycd#(>;)O%Cb8B@dyG;#-xn!+GI-)Rr`Zu38?ga(4`;WS(UT9q-Oabw+nE3b7 zkv*C5EC;uysqsI7gjj<5L}QdQ3HV23Fj7EJB49=aCs4sxl?cytA&)UPXwMTjqS5+m ziSHPb`jnEG%#yAzqUz{+2xa9sV+v{kgBmb%W?GCJ!JZ;}Kr#=o5TE$@tCsE05W_yG zkAN)qZ=lEtP@{mrSO9Daa1a@s$&iOw=u>>k4MK`cOrlJ5I0{Hz)q(c;Kq2<2rpf=j zDZ~sh;}eOciNN$1i(CRBwUyDj&S~(hG%E|jHGP;dAB(9zfWkZ{i<$70MnOiN3lf2w z&Hw_Po?r@jN=TE+O6eu1JnBnyr>2iFX7nj#?AUY70I2sPNj2{f>c0_v#5-Z9(Gkwt zRyyF8hO~PVX>qhPGv(*8xGbx8Q9`E~GWtN%u=E5P*x_{c1O4oxKJ?Za1m2zPp`TQ3 z!&OhWc|pU!_=~@zHI>q`PPn}O0?m;VtDF;WlauJ1ld7CBP#Ps<16(8}Oiu!47_hhv z#4#rQb!qx^R`zU`=KOkgoPNZeVIO!ouj?ct-fhLW74*>bh@2$h^Mpq zV)BZj8K9;B1b8{E2Qs6Atw@lEZs1Jsm(Dhz;`J2fY+U8F)en9C9cgv9 z0QoRUdl9uuW!mdw0eY|8w~Tqe)* zVV+cMNm>;278(4HoSn;rcw$W*d`s{9fgkvhRI*8Vf0XlYrfOw#mL(%QNyv2~@;eT- zPSdacUibr^Q>#L-Fvz?9@YPOuxreDwEEQ&sw)Q7O54?lZqDnuMmGPC$JpBR5EuLH(7&-z!7uW#sUFa-?5*^i60G4jN1F z$^?8L;d}*VYWA{=AX>hbnK+4~B2zPAY#i$BQLUUnKFea<7_IQqpGsSFm4pG(yiayA ztUQgUB9p|V^9byKDY|;ALb$n9j~8n6LeoUH%#EM3EJScE1l22A2hIYGQKVb$rdM6? z3A<{ULWNi0Sw>A`Rw&4ifJXqq-?15To8U#fRB*Q}JPiu51)BPtm{*-Z-y6{ZIuJZn zIp$fz@J$)hnuaeU4Z(45W@H+}%VBmj@O2V6Dz0&&pfP@ zPq93P3lFgShR_4myw@Y}GBPun#ukodWDu-;aY4~ocr*bP zPk`zEdY=G%z;D7+NX(9`ES)r#6{=J(6}FNAy>AL>R_*ar>q(O;3M2M}ky`xJU>N`_ z?H6F+|A)AAL;4HW zo$U#z=*b|mygt<%o`a640Zma@yRa;ESmqix_;-(i@14v|Edwos1EmD!2dpf8MAlhC z7Zcm9JmSBz1P0GAzs4|^6Pe|62IU8tt4&#!v8ZEHhMDI6Og2z;3u^;rpbpJaM_{R; zxvUGaw31xgsl&9v;dUBJ50UjF`ShsiAf)Es^#2ATOC6ErL~w-h>!l zNl(!bd*q?H%^~9g&@u%XCeKQt5L~rSIoJLhb>EF16F)m9aaK}9lphvy=NmO()s63b zEAP&AKLxt2&HNcOSKuh$+|z+KP8!GKxK^jorPC3g*%)gD8RaqNJN!vr<1EP_G>+je zH^>6Pa2dkDPXqYy4CgyP@lrqCITkxFCXTRYU?YN@UVe5O64wj`F%cvH0Ce|QT;>&n z?9v1zA4Dhk=)atRcSsNd+jtjGerp%r6P!1h%dC$^CO5Qw^|?d zHtp?e=(i7ZTOZ;Oc4cn`=~P12OE&-e6ED?Ig^x`OzdZHW+@;Hlb%chPO>ZdPLSMZn z&%BMyB*2NLp^0v?4P$3N7{Sx1u1O=-JGXp_bm7T37S>4?0kTp%+Wk_-3~qa-!!}5m z&I=Dj!+l7wqWak=5-bjQx*j;cS_w-=&$Z%_U9>xcSoC`3oW!Dt*F}W6JpAVEf9C}C z^&0f8>ER?3Cd1Untf8<1SRjp=f;nZAuy8(q0YwANQ=GC$?_*4T7~2fs!b@Vg2CT;5 zySvO?7)cJ}MI{g$FK}v*3X67wg%RV64waoq27A!h8wB-AG;8VR0~7#f*nZ=RLtp4y zy0eWsKkOuLynL}O>fJa*bmp1k>4YL+B~E-r@({R^xDr&d!nY{Ms;cGbHK|h@gJGqXF6LxpcRR&B67XlLqW^Jk-Yl;z~bJcAE!#2tz%M zV>Y5#8MdL9hgWx3SA!GTLV~^};nv*oYZ1?Jar0|Q=fY7msICSf9hsU{wf@8OTh5Pl zk_O~eqP4p*_-P_5F&B{%^!@q#s^iFao|^TrJ8O^8-+wE7Z~U>=Y#Ld-$&3A1ZttXqKLF*kFTbyQFxj+0SxjZ#HYJm!53dW~?l=p;zv7e>2|b zbKQt<+k7p&o_eye%kz^8re`gmF^?w=ofIX^ziI^RXxE=VmAN=0^;ey{c&I32!f_ISYm}sa!TuS zF0g@?39MaIm_Hg~Pilcrl-lxi-#vWxLWK1vkeOh@cESCh^?$qK=btp5BT>VnNBAQ< z%5Ua3M=kB_ZK%JMB9N{;?VBX7|HI#&e&QxgHt>n_(%$loOZuwDQW&%|7D^l^L3f)m5=Az0*_J) z3KE}mr@eo{bAI5A%}Q?w6RV(7V+2B^?e|kY+p7_q-y>RX7%sFELqobEt#=)Vna<-F zp$VU5#&S=+yz=tquZJeNGOZ^}ag$*lFX@jLT#xv6vGW~XTTNWI58W2qy?<-}ke&4S z)}5(Ni@f#6GuooE?w{RhPSb5w#$dDO+{~fN_P3!8@7&xL50}#RC_CC8%h0KXz*}!! zaul78kLggi9G5oxVw~4-oRrqyTK;(rsEB7tC}2P&cEV*V(AtPg-Ro*7O{Q%%*E!U&vv|^?s6x`yTti!GEK-A@)s<|qudaqPp zD$l`L|8#L!ZU3p_?yU2l1j@-*WOQdF)POlHCP~7 zu^C~~%<|V+1KksDjB&yqQbW#`k;PbL$v!x{OM+WBOyrs69aywsWlM3;#zt%VNP!=m zi08tG+L8)aE{5I1Rq}o%$qk}y98!#icNd*m#JCbtL@>YaXYS61PKORRDz7Gc7ny}c z>jhF`H4l}RLyF)xeOBZ?U{n-MG+1Nbl1hd$n;zm8jV4J|tzL3rq33)tanMa-v|X)# zKrzNhn7W*Rr35t(H^TdwstEswlK=p{|5vL2|L(boLW^@V0l5F|B-AXo2P>%j>m(HR zC8LeYZvUs}reAHUdf{Iu!L_2auJ~gf>=6X*#8EO@AQ!=YxxfDOfR}NBfe`#z*{5>r zx`&cBp5-^HZQEm8`K=s3)j08`2ToO0&NMt2FS4Dfy*(4`vCIEv23bAdt~{`Ptk;Oe zlRa;}6}UY~v#(TpaIS48lXkB+CQrU<^FV%TP*&>T>+IC-M6cte8yAm%7def+wo$z- zw5#YmRjgHBKg8EyIgRzL-dDeMpud<9!{c44^3J1(@+j%Yv7qx%nr>Xsg~O`mgP+ue zv-B0!K%Zv?$=gLq$CwuXpQx$4k88s(4g8mc4|i|Woob{5cg3j$CicUjOS^NH3xU74 zzSwC3|7po^-}fv= zT)K2pJd~)mzgBSZTwAnN{L$8dnUChFzI#ONTBe!QbVU8M%6HB;!zw3@7S=|;JlDTw z;6IN`Tr<_yH@%NGbsm4dKhbbs)BeNy*GxGr{OU`{RLscB0BtnxzhTs{e__-peuGdr zAM2lv0^*fFYs})7{ZcQ*7Q>Qo_q`0PjrenRSuZ}(IK&jG1g@@^(P`b+F4A@BjRmm= zCnW_J=`Y1|eM%hDlDsQwnHdzm1%wUCcS|b_Vh7GNU#lCqB3oJi@tk6XJiN#@4ezqOUcb+?yNeLYc zp1bKeq^g=*_xPPjZd`m5F<_@k+#hzky7}Nw`T2HXMz!-2nYg>yPPnQA3iR7&{9!Qq z#gQ>E__ybrh|==k4?U-A)3feQP;=!m@q*G$nv zYB&e-f^g8^nkdQMy)NX;i%N4-=f__ z9~J!BV&NqGs@j3?Pk&JO0KnqZ&X{{XgoWu2eg2>|to|})uYh{>hj+)`ePYP$G9D@p z)n+g!h98&Y!cD9j5gKk0gOBoI3l43(-%LU-aONQB^r$Xq01YI{Dav}kq#eUt6LKPZ z7n!`)$#8Bv%6LMU^SbnV@l)Fo+&H0KCuqibslL$%-T>tFHw|NzGe{5#!6!u$Uf0v=lgC&z$D?{r*%H~AX5iZf{BpSGmB}{ z?qnyfJ+hNBgX?qdlOq++*h_(84G2p--O@`KY^-)Z`*kElrBmRj z&9Fy0-{2bija%#l#t?bA3T@czx9#a{di=)Q#y5{$k9G-05J1rT^5cj$IZ;=)W&U2H zDv#s=gg557^CUs))M?d}&r3!RVKG};j<;rlsQb;|Mz$V{aK48AwYr@h5$RW#KfkXn z&QA-xav_Bc@q6q`+n4AMisa-37t6IQKM(+tzRM7TrkDg9(DI}^sSwpkV=(LGX> z>YA36UGH0g`k}4!z?K@?)a#zaYH4=;)X+%!TR|jSRIVGxgYFK)M%g-N>8zvKBM?qG zzAm{ap-c23;*jq%InsxIZ+_szK@f)g6DoLD@{VgX<&%M1n0QXWf$MHO2A3H0_m<>X z$tT9~A7m@Xm?35X1giQWG+kAsAzCSjV~> z^exBVtw(r9fxmFx@-J!|=ncbLO%k53{Ew&E?ZbpW2uIcCy~M2gyqo?)@oMO@MKI*6 z>a0u;7~kjFcxKVqd1-#8-6GJ_T#zAH_fBrEn0mMrI zSTc{f-;y9ogZ9WM^1k-0n4Us%szqjT_P3*XU(L0}-3XRr>Xq|e<6#|}M-K1HZ=;UF zdlx~GPB{7&3nml{174?1cK!@G|2MQG({}<~==Cw}FM~W?BscFjgHWHUOrZvHI_eUL z=E=An_~rrQn+yUL4Zekj`IDg!(XenF46+^Mr0ctLAaUcWvA2}=K}Ik`O>q8rFr9}$ z#}ZgYz2MOV_3`xI%R&@$iSMnS>2>8>Wsbd`u_=HPkhNH;gjHZAwEFuyc zfwqoNgauRTLmEbJWjuJgvKxX^iquvL)wK?FiVkHF^;}Q(AU8+ALwb-4J`rA-9O?NXaQ8Tql(_q~ zaqc5=`7k93r|?dnXb{=2(G*VMKk|9g| zK5snq9eQG);6?>EUt*R4L)i6`@%ZG4=;S(jWAfBw(#&B}(_zGExs;Ykh?pL+z&iS& zw3qq+;I*pP-dncRL4GNr4}8Lul4Jh%M*Sn}=U{33G}JOJN-{=W%9;3jE%1M6TN1VZ0tFEB#(EcEot6^g(vl%-pfC zVj3W34q6@wgqbCqE2&vHfgLcZ95EJRm!Vx5$!A70R<7SsU4LGS7J%1mkQV4V}MY zHNXC{E%llmoobS*6yrbxc9mi>u&nDlz%u2DC0D{~7Weqn^Bf=LFU}$iArB%jUQV?dZGfdt&8w} z9J=X69~TiW-}!DG7J0-97kS8d);@%ET!LxA0&(bzQ!vyBa~I{+vwd`K*CoFubSa+g z8U10gacT)f3l=*9O__#XR934bcrkXsSlvsa`_Zr5EK#C1MrKbdPP3mW3x_S5RD3si zhCx3=mshKxpW&d@G~i4&Vw47?Qeh}J2zpWPZU$RnKW9-k$AUMya=MJn53T7}b&P_B zsK_zom!qh#T1W*-w4%?C?R_j~&mSIvKOoo+K%|wU+`u5lsM!Mq&9wK%D@{Zn9IK zj)1rTM+isF*N0@8M>P(T-gP$s$VfPx2wDFE4aDgCD^BEFAY9)h^(xIh$*b{9>8$jE_};m|j9G35K?5@Sb7!Ek|z=P!klE z2>>;LLwqu2HKw4}F-Yjrn^=E0&0FjVvi0|0)TRhPy|KWVOoAYh!I_xoj)g`5uY8(o zU}lI3Q`X>emhu^<&luz(UcYvx?u{XPgDks_z&j^Z$Pv9h)VLQYK!fY>|3;0>O=tH% zLLcJaya!O*kKPzjn>qpH7}*JL*sN01D&$`4*xc++jRc!gTzf$f0prz`-6Bu}e__gc z{Z_{n3hG7Yg~eBeH@Cl7FPBn`FoRQ;Q<)ctqSOWbKRA z)exvZ6&yfTFYS+WCV{tY=<0pF$p^QYLaSJcFPcId$% z=shX1+N8i)NU9z}P8q^o7 z*7X#6{D+I7(&^5lk%h34c&P|hp%IJ#-*Aq+WKVP00p*r^zraf@pMTF&KQ7ACv75^80xIIbATB|0;{hi+0m=IBdRQ&TNQScM_>rM@Y08q7^x zvjv$7eKI|k;;{N8iJssNaddJ-e5N9tJ|pVH(Si)>Q=kd!IkdGl@+(c)~+hAf~<};;lXz&04Ge_;Wx4I_Ru$2;z45<1fB}lZ!c% zDDot0mw4jwq)%&|pV1U=%_JQ?nT?pr*&5GHU>hZ-nwO{5~ zd+asmzvegpOD3AKZ#6CyES-6~1WHuHZ7zxbMX9MPYMyq5?Jp|_>s_@|^k!L+e!C>| zr`+W2ve{d8eUoKPyVcpL$&))zUYCp`>-A4g`lLUveq{WH5c!r`U1#%GheP}#5B4+C z{#zmAn%55$=i(~WJIS{I<)654eQpi5yLPJ~H7a=_jKyRf#hw-YmkkK~zw#Ip008I# zkp6%27#KO@M*7*QtvudoKJvd z->$J;9?EYgu7*GCr5L36xS|EMqCpVBYb|8^o@llUUMu>k<#q!n_9aHGuWZ^s$GCk` zSx4D-g|kHpUT9O;pvH2`POijoz4E2X3>SP(J&S)j{B+jO{vw$vhK<|db__|7%bfk^e1Khw{!4IcK8C0-lj&0L3>Ek#IOdOLCDgbHjC1HHfwt#V0W4y8JI6ZYLNhGqVsb1NP64!|W|Wc%PFzS5h2S z)=T*U*1JlQq*&a_+^_ZdIi?s%&#t5@>METjVC~$#*>LokuVffDZS-g2<|Lhy9IO1k z6V!nAT3(q6+<_yvVHO41XUbVGv{1g8zw0jyP5xLnBfEU!B#!X zV%1i1$IrR1b)JkW?)A;{T!4@7(0QyZ0zL?{N93eY?tI6RsQ{wzdOqMLhEu) zyHxY%ochP@)7?h39W|4mzQ36wRPWTE{@AQ*djuGMOxZygdsJmVi2Kp?2DDw<^MR}5 z(ThW?yi0vZhn(u(jXQZBjjZ=zuU_(+j_h`x3;(@4^rFe;Rka|iaYGSDVvkAuW#wbJ z?xHc-sv(uTbNk~#jQI|4#+z;&Oq5BCrdwIO;G9sU=CNCF-ahbk2*Gii!L zyLSFOPBKvD@WiANTT9( z1%Z82HBLyDz#;}~$CtmH_<(nEPmAKk!@KBN=X%^on6kQ+hfDQMR znE9%h)-?nHET`8p_F-@aMz~(Nj0v%`&|MeAvmmh)w0)?W98a7D<{R69ak*Tkf9#F; zV)S^Bvsn2o)bIer`xUPXcohnFyS2nl^uxv@Za}-T_JQA28I_sDP`bt-wp2+b`5zGD z7^pYzhnRO#+&%R1;FDC@rp8Np89yK%854%&Xiyf5XNF zH+MYdH|8yO=pLi$YK9E5>qbALXmQ;cn#^cZV12gEAmD5W(L#l>+#+d0m zK)B1gwiWXPB=1T$s*&alH=D^o zsmRmtv@}w0)He1%KHPyujNW??t&HmCD4hnY2@xT%v*P3m_%?2h#@s9DSRY>&?I2>r ztW3|-e3Gz~<~%VZqHy|-gF@83)7VCw=Dmuz;q^!hS8b7OgU*q2in!UgdqR9i5GGE{_R@`E#Y!-F0-P#mPh9G(hhu@zhwZHQpBes3&V)2>L;-p|$R5}*ncx2EpWIW5}wZQJ3 zotI;0>JFo3?!fHzpIc23{YS&izTcp$u0N{vd!89A;xI$yyD1a}&n`NIpijjvb>cPSF z_lWQyhM@WWN`X^MVAT#pHIBP?e{DC3<(i|m3ndu37`(6#4Z)j-U_%aFynnp_hm)a2 z7})7r2xx$v4-aQrhlbD~8CSuAN1?*pVWNs)PYO6f^WFjxSbRHF%rlhd2tuHN1r);+ zxWmrsa~ChzO0|T^aI^7IV8eLW$t~4vvj|q-G9Ys}=8A}<3rB~R7k;tjG9)1h0 z{+7(z?!js`6KQu8iECkfyTH1z&pJSYFJK~XbF=Z{qfm`(TpZC}iqRJ}kuJLkMsk#U zmii1JB2--eFcp3%9}{5_6As1hVUhQfqXUl6jC33b2>_mvMXUN;)4LU$kp<*K1GxW5 ztf(=<$ zp}+}T@>%~BzU&mP#uS-jz!4plBF2--Ymn;Ynkrw_qf5yW6LMPdudx-(Kn9)$5@2Xjf_7Y=@tgq4htJk>IKtTp$>4y zUt+X_WrmSwhO58kgTa3rrynzmdP+qj%@E<1h*bd5f0oVC0QG#s&dkl#TQLhsMMTAB z#cr?}4rV0sWO+(vr}<~~`ekQrpu}HgmvCiA9A{@_XT>R@$~NrsvI&)z*+u@@FQrg* zZnpK7Tm=TX{aKmK$AARp+!s<=4a_;6u^C;J)@_z~`Tlv;%z4?BDEa$2M7O-ujojka zyvO8hUyuCp?EC?GYyMO=>OFJrS2yPhse*~hywzCLnwzZ;vB03Ipvp40=C}ZYE;N%b z)YLlPAtjU?>rBTdcIij6He#REXTTQo>6RCOyh03Cn&6x~r)1Li29i;10sboJaDfe+ z#&$iF>&F7(bvRNor&zkJ`1}wX7XbYm2u7kx4Bj(=Ln)sjIMFJOutZWNyzD%Kh* zzQ~J~t3vUll>DTynTx07>Agn;wALtQb1xsxRX}9V|8NRn@(7v7oM)h5%Geui-c#X`F=E z6h=^$Urw^T+@QfWNQgo5%Rv-;kPK=|tpTIex&W|MUbUU3nn@IVl3X)v2%|ly?Zv}x zaKi_rYZ)h4d+@9jD(jyBSpR>mhy65IKMB@}C9|9-chXoXG*%i5VmiDIjJ71u0B5S9 zX*56@(HO+ZTtYS`(4a_~R}#g11Owy6nD){DDR`45UbE*Us64CpC3-(J8Q)NaHdE$f zmRyA*Dcta4bzVG-apKI!nQR7GQ&2LuNtq>R@z$Bkw{P*C4G=~Jks#$56dHAgWaeNP z2b}U~jHcy3VHOmZdmW8?9Y7KU?Ls`sETDQAn$oDr%JK4DII{@5RfI13J{u#hvO!an zF|?_&Ofcp=#2EvYKKswhuaP8m#x5vmF_ckO%(JW6a;146*b;W0FG2+s6=xM&E%>MY z=B7PM7pfI%-^vi$YFni9yGMs@k|R{*Vi%cxU!;8>+m629UdS5MvDLm0?2u9HI6&q3 z*mrPBB43?nPuoN~4kAE|Y~_;W$x5h^^Z2o99qL15A^ikCJxmhxd{Z%_|IMR)Mlbsd z4))s&{>cpfj>5c1L>yvag9{pOexs#g-);^EAK_reB-VFicpHV;43F5rAUjE&&n({Z zZ*+F$`X8CW-T{y_GuSk#%Y=f&OUL{vDTGaQ&gH&k&-0(gcX1oTrQ#V+Nh;eIga|(y zN~%i+ij>OhIs-!r)TsRP+YwdmzHHM?9}eb;@8XK@5|Bau#BgSIBQ#}tH(h%;2bgmK z$WQopmO+e@(#X$^8fL~lYt@aQ`ZqIr$VD>qCvu;a4B`}r+#;&u_}{UhyRHRw8?7Oa zh%Zg3jKM*SW0b0&!g@v%2vOPI0PfP$O!7dSd*_jvO2h~vN~X^z=FJ5lVS%XrIUF`j zy@!-Rf!yA_&Puwui_GQc`GXlu zHI-x^cCNhkA;9cV?P+q)DM0Xem>WI9u{AQAOt^1YPZ}9&mbv&buyWFURBdAPi!s{* z|Cp%k*b0p8ThN$h=;)>l>i^X$5C8z6?SHrOr~!)Wg7El31?P%yD2UpXY@$zA9Ej@JD(YCb*cdPl%r zK;Q0Jp*N~+zjme=tm#6XoRnslf%J^OV}b#Gl$GUuMyjdJ+m*~uTP}aU)SS{KCR~Jq zGrqsL7vco0AnAj?ys!%fzltWAv)*w{#_ynJY$LwB(anz(R=;-d>PGqpWb{I0MX|{r zTYuv{bfF2$}C zc%IBo=i+N;mgCa$4P1*vekyXt6C^goN?u9bF;znT^iH#o`o6mHqCf(gm!9^_x$C5Kw=CbIe-MMVlJ;w)0YDWrjNyfL? zSPSd(M6B6g&=+`L*1o$L$NdT^6DQNKH)Ha35o^reHniA&3>`yox?6KA`0H2_Ch&kgu z(l)w(-J5WJ%ummni9_Q~!LAzSt;3uJ72`+H!=tvUl0J-~gTH zGrM>$WXIy^K2@czL;IPHW(=0%+7pj0k5o5RUD(Rl+GRVOh~{$0^Lc~)qyZSS`5bsD zIBUh!r;owbrfBK};eFMViz7(Vc4?*%6vXl8!MEetgE#-2Ni5q0s) z{~NS~J%rK%*%cOUqWrp|B$GnKd}KOho(L(dBAiP8Q-n!-deyNO!;99Xq)d{dkK2|QIO;WD z!qEI{6eFdfF8J8)u27YnuyyI@{Q)4m34>P(uZ*IhT=82h( zN$P8Tn;-_Rh~3TdMEsidB>GL4d=zixG_-n^oE@ zkKjJ65Es^Dk|u|-{ijt(--wk)K~E2WKyD~NkUU)zz*&v^vw5p)bDFCW08zy1YjJ@< zsB6L?8`~--8DS!eD&FDQyDi(+wlJ9M!b^nIm)ljTKxddXUt20z!EtD&Pmas)^E&tz zhL#8+i;=#r_bZ)DN}E06x|BH4r@D@Ai{^I_nK2tsy~rk?XTQNc;D`U2g*I!U&vHE; z)~|W&n>&M_Gp0V?MRi$!W~MLNqF3aI6r-VfzvtwSu9E}Q{<9m1-Tt@sEdTDZgv#+? zvPDE9IyB}cV^10qL%J*sWWI*&?#mp=&FV4!#&;y2Rb@C$QuJ@$Jc-S%6YHiaa4#h@mid`b=}@-HRTf_ZC?! zTf&4kG20QpiHl!;?zhm1KMH=?AiXS}iBj66+t-wM^%xqEMeQ`Oh2N(QVU**u%_wYFpO#{5~G@ zKO?8+Z#yUTP~$G{`I6N(jFK4Um&dj%Oy6w!BBMUE{-DKVh9X0R0v8~4&w85GcH=r* z=7pa4))=Ypob8|}F3Bc&c8b7)NPY)eb)L@NU)BrsPTyNUj!M%Aob?}I5P>ePPrBHW zI9pG&YJ(3i*JvI)-h8gw*X8ZiqyD+E^QdpU;Mu&8-A4P9R{BmMJX)3G{?WT;qYxP{qQa*CKVijeF4u*1l%Bk2_)!ak5HbLP=TXRk@et%ev||^L~;fR#KCuQZ~_hz zi~(~&)VO|y{e8h?C5&f0dVVz~oSy<@zyPkA0r-VMN;s$w7UGYAxOhbHTd-_H!qgUn z1kOYPQSx{6UC;DKUR;j^L89(kggeFH-4y}ubUy|#1|UTQdr+Z46v)#fFGq54d1{RhUrfqCMaqw? znDL_&S*g?(?tk4qB1q5%OoZBTs^)QWXG`=$YV>MLn#63HRAq|HEP9#7O4!n2i(<%XIlxrq_Kgo~{o9NRSKz4=v^tN2L^nn<>j!*9XUG?#GE2 zjx*wBnXJi73qK*1mC)+h9KpfNdgd(gS>#z^h zxL_g9G^WRN)B``9dfo@mY(FLe7nYq;I2Sb&w&=>;;&r|`o23X|d0EVBB2Pg+TSh-- z(%*C1ANoP+`b=g1JdcN`KiCBicE*FhnM1y{Ci@Z#wg!WDl+b&#Oa~iJeoCQ^c-Xr3 zfZMnm;%FJ!BJR(4_~ZgKk`{omdRDDeqF_}LK!ik-A%SF0Ul1hX>s!BA9Wj83U1{>izIFC6hta9Hr;kAB9RrRP2}-C%*0Un^s! z10&&{WwyMPR&DI}9~6t-D3{#@ERo;?0Om+C1nu%1BVARhTvhiAqH+Ss0x7m z*mA1WY`Ch*a$c4{cpj8dl|^AuKdGYAOd>JWuLG*va%$hqRrm1L^aL=^JgedCsyV2t z8Dn|btW}&L%{Fzy2B5*_Q1E&x%L@+{_Ge^hE}2Ee9-PZk31F?q!gdMmcimrC=KxVrmjfI~p#pA$BeFk+qpC!EC% z1K<+?c*Mh&$iNmtZcFUL7I$syMkc@x5k5naE?GR28HmVM;k{mOd8LTkGMQVc8}Va` zJEN@GH@Eq;t@$std7cLM(6(;#wr;d;ZwgfW3TUlUZk-V}T|ik;_&N$Bb#tF05_YB1 zLOIpCuITQ=)sk718ME({qvnX3%AVEZ?hBu<%LlqWFV6Ic1MLGWJVGCr-~7ly;FvPbby>+ zg8&rJ?aLVDiQDG4Il0W0A2O;^#Jm>F5%YMN#CH}0qR1^=w`2$6y$<3dwL9cDkS?tC z$AkBsbtPBVdFdTUqWM?6{I^vrO9%KA4KRg)k9+j$1^r8nhWu_e19Dn3aPA5t zFVK*CDCE+=U^l|}wAVBUeKyBjX;rK&8Tbuzv+-%Wdpta9tDa&tcw>0V-6OCRcwhqZQXSrg;V{J|nQF+I0$al`DzUv{(4 zqL827!On6H6{`+KRl`02aJ?j$PYvuy%GA>V{!cDK#WytuHBnG5EqTLjHTQam)=jG& z8-Z=sSMLpb*z^UB49K+1I*oX8^(wG4c!?l4F{7I(`1N@>k_x}jF*-j#dT|8)uyIgz zu_G13oZ>N-LWXX1jBU=3ZSgx~1ICkSP#hO?9Z7A{%tFSBIUVctAj3J11Wl$w)3DUE ze@?7?s=zmB4#i1r^T8P!U>1kSz%i%dq45jQ6fq^HKLGGc`|BAF51Miu{)1v$$gnrU z)<;Cx3LZIGrb1M;R-pf}zG7?bCo-m@Irc4n?B>rYtibd}9`jm|bG*m2-XE4pqVo^wjc7+r@0 zD82*|O#=Y(oe3ULyn{m{mDT#L#SszlaRE_pJthBqD(dr)H&|Na1}{h!fuulVsh$FQ z(1%!vFMc*Be>VF%dQFlu1b&2Y{w8v7B6hm4;9f=Q^#bl?niQm9An z(fsEo`ejLh#Tg`EHs$)PFK#wB|4W{L`}PIMePM8=ybs~&TqPWSW^}IZ``p0wPjOgi z91hy^@oSGSw9OW3|NQ|EVBS=Io@K(b@aMwVqmM|`BT&~rF|%~h0LSt0Eh1um0XhHu z^Jn-MFW^n|;*5#I$9K;mW^SH4f4*>?`5H}TDNBH#@!(x1I{Vl{{IDexx{yYJ6)lrb zZg-$DfZDw@;O8rdBpO8gB?@#kOw2Eg)gGM45cz`k&v9sm{)mchC zy>Sk4L#n`8;N#L`+R}a0vL(@30Rg{~xa{@~5=~iFw}W^pt~A0|u9+;vQ&_sp3=HY& zNaMedU=rlEpvTRzpj%&1!x^h3_Lh6Xh%cyb6Pn)~5nlvvqb2_;)Le%yW)h~{RxQan zWh8j{^W}vS$O=!iZO+%G+ON;4o^DYqbRzHzK%GDXuRD8!L_E%VBHhnlKI?vMi&4AC z1YJ_Q)Z@Dum9<&fvswEWP8LL#eE28kEt`O?=EV87U}(pWEy~-F-oJo86Sl9ESEL)= zdz9_fv}@z)=n1>c=^v3F-fjj^H|r2upZ|XMG22=y__`7dX-|YLP`5e-k!n$ohru2y z@SQ#N9rJ0^K%X79+ULg4Pf zV>3N34kQ-I-o-D7+!HPI1o(cZ7kvK|yvHQ;12XOwVgg+aX1SE~16B7!zK(sT0Qfca z9g~8N%T1(O&pzAY{sWUup(mRnb@5{N4$h8$ZzwpB>O5f72r?BqL<;Rb!R+d}9lj|z z6zM&**FtQ0pvBT}KD7J!`)`;Vk~Vjd_ULO_0ujvY0RZCwtYpRpGL!M8b?>jVo2G;( zG{D6f_N#xDcbWMq&2R9uGTrKeezQ?$y0l0hY)hmMDHv^WF@_CVfeRW z9s}{hLOk@L?+t!Y-i8$los872%k~76g`AX5&eYsHs!Q6eH$4E{Fyw1|a^~B+xV3&$ zAyuDsQYdsHAr#2T>QMnLnMl?3I-ts4i>4N7zaWPG&g|mT(s{Ps;lI!qpAEVE=wB%K zMVF`A@+brEE!WOqdc2@c&aDlN{OgZ?24ZhXPd}LzB#K|FxSpGB`zH7)`D-tH48*g#r8i;{OT3NqAcC4Q$W>jF@DrYx*^-14A>4s1FRQ`TI;%}*2aua3g znR1U=o=Q(ZZr2G8wHs4IN;K*s$DbukAhO25cOR-tr>Y9}1gI1MJn z4bx@l9r3Lep7%$t_@AVywpKW8>h3NEE^Goj9?=Bd1BoU!*WJIqb^F4PE4ZEFi+I0i z7~nIbc)09-J>i+~mf!Mx$3{tFvvV#klv%RM6~F@X9~DUs`^}mMA@GfqO>vvo!W44? zALp%?RwuSWX!Vd6KisRLRrjg6hH5_%o=ElMZ{9s`f6om(dHuRiss;gcid#>RWK%?y z5Q<6Hkm~$FW-X}_95rWEF51*ez?FPzULd1$F7^)7EcNoA=6HxQpNsb(2vm4f2@ARk z;uur3S(qQfRCao&#qJjpzZ*~($AXr zV>#t-3$;v2c}qmxzGMC9-TPNLV-rz4j$!uS3IlU`20>T9hBBt8{X}WUJ0xEYA6Gmr z6^*;Pf-0kGcXmnexxIEhn252t>6)bRS28^D0tKFlQq#ITnD2{0_19-mJxXlDn4>mt zKMB70A}_%2Y|eVn`38gB{Cqj?9B+jiS@zLFee!fmKB*>dhXnLxNN$pL)bLpy$@=?9 z^&{66n8)J7WkjaEuC8;=qo$ttyiXQ63!%@!R_n!a>L+r=>qX~w@^Zla$rXI%ZJROa zZ%UTp+Kr*F4?0uu_auZWF{7b`{8Nz%G#wTHz)c_;Tz_F^-h1d zO(kUAa{n+lCpciYlCSgUZnfBW&R)W$anQ<$$Xwo_`Z`VQV8Y;PzWDTQmp_lc@{~kB znfvqhbHRLS+eZG*LoFN6^WBowtjB9F?ge8gS+T2^Kah$EC&m*qOfy}HPjEn1EADV@ z{?!Wg15FW^+yd(?@syfdR1VW(qUbAJRaVV7U5W?Sza91=aW8u(NWVVz;DA_XLf37xefC5?f)z$Qs3AIyEG3o%{kHIKs`)=7}3AT4x8 zc*LrT2us<3x*H<{PsmI-Cr%Q4A#LJc-OAWlI|KWs%W|dypYv^3U9F`{&tvxnnkNjR zb_Ip>lgwwcmW45My6|6=jpu7p)OqGp!nY(0Q8IMxmbsV`2T;l8gdIhlCrzX}-CB0b zTGdT>S1;rZ(-zp_PFCgGO?p1>Alz_2Mw9DR?6Vo1g-k3vN|`MsMaG^h;gu1qPcjcd zWtOkkjLZ4TR5m>Pn_RY3g(;&0&m49I_=QYIIXI>qoT&RL92#+z@X7J2kdP-Q zk3M>~pR3Nc#y#L~l2~k>|2(Tq#7-9G-s@cuuEZy_?LVHg;$8ShykDc9+sw!c&ofNb zm9g_A&!yJ)C+TIJab2PF^qh$ zdmJCg(Q4ZlIod7ySg!ESwdm3nFf@SwfHrxK3U=TOoauV@p&?(Vuaj5=Z>+j3NWsV$ z2>}p;bLoYC`IVRpJotvHpHh3MXw_R}gnx}TnF@I95uOWob#&nG)A`|*U7!D!nRSMh`AZuci;#9k=Ps-CyRPA%Z>7a`#R? ze0+OwXrS6#f6sHf=@os6`wF8B&VwxWM9;rpg#tuusADljZ#L96(CrWuiT~rwIi*_7 z#NB5t0_#k>&W!_K0FJ_dtISx!?T-6g%`@*V^JvD4U_j88taqFg&%~!4%PckA?~oM$ z#IwlaJ)a5d+GAT z1IE9oXk}Ef4W@Z^G|3;}QG!lzyv6`pCh4hi*`zZ5Iu>x0kT3jdpT_>HL|{v}Ul}D} zpljqcg1<6!D}D(XGV@^)8B1>-e7)LVHRRL@pKJ!>ped-NzT zoNt8!VO&Yx-l|~u@XCZX=JNWE4PFwwwR6dkX~D8^O~ zeZC*~^$+-{lRgTi8(@mR)L^P3hJc*DGTi*k8Lvlf&E$E0%mM5G*p=>IU&SHY1>Ob! zy11;i6Yy2#E9jT+5JRn37+(+@G&5?F>u-{`BLC7tfughYE(@|zkfqN znTY!FxOKcqXxWeQzLV@f=KE#8@R#bD7E~0*QJ^5S5TZ6_|8T2Ad7BPs^!Jtn1x7Hx z)vAwPM4C()dMe$Ki^?5Jf+h$ z_O&&ptu=~0G>Stz=aV}(tfNnIAq*WFUxqc1Ecid?!+$J??J!W_GM$quod-s5f^nS; zE9s1|Jf>0&HvkJ%L9;lQa#)EyNM(RcV-YJ@AY&KET_eO#gRCOL;f{5-=n`=6+V|IF z=}-+=U{G4`+~s>aD#jp3#qx_|XB;#?2fmrFWQH9 z&t;x+TR8%oFd+aX|7@m5cKe?Y)vz6^nzy91V4b5X*Xp3GIzHOE#%eRwv3iZA`gy&E z9oG&9uU%JAHOV`tan#Ez-h1Pe#e76rwK{e>j>WOzB{GHp)atWX>9OpM)d<1^NRWqE zDOrFHja|oXo8_KrXSYAVeqPf^UKf2$*XWk+9mhWB`Cga!KDQcO?cTm@sxEH3ulu;~ zzG|PRjGniuUap6(Ynrakz&ow^cfJbYek;09#C6}Z^m~f;d(HRft@l4=)PFGj?qQG~ zJshSRV62~Er61%o5PhxR>Q;XUY{2Xk6egp8#iBQ!zt6Bi&(pXsiP0d~STDq}pP*_G z);r+uK4=sO@x~;P(g+rLgSj#KDW&?Mjr#8M1`$36pH2ozHt3<2-r`b&LHycrs& zbws->LQH?xNWZ>GJvn2_Kqy>)b2gvKnV`a&Fd1B-G$P7UB%Ei^ktu!ox_Kq3?&8bX zq#kCcpqrAf3BDCd%u028R#Sny(O#^5AMobh4CWOY3A_!;$RQ-QY7oMg-+Ky0< zFSq=IUhe643s&C!T$0qw72!UK8Qk2bnE$y)DNrZt*ou_c|d5^CpdX zsrYu+`8}r<*l|}!)V%(%oj&F3Yp=P+>`#vp56wLm1YNydA2)f1zpi^T{QL*M?pOU# z)^oufCk%m%d;Y)IX5|lFj0y$O1BG{Tk9V17)sEahEB*{xZSkciDyRm8euGy%DQwB7 z)BoBpAno@?^$LyKW1G!XLH>P0RsfKsaOh`$Uz{;>;>U5-IylF8G_1D&OE^yF=qrSd z@)Nn)&V&(-eV@H9L|El$S~9)x7Fyy-AJ8vJxH^`$ocQ6;bty^nL_s{cKi%9RLSG`5 zqy*&XFZDDLJT^_ZX=1sWjyySs%CzmCE>6c;o%_b%m~lLw?bh^9UFOupZIdUPsOysX zY$SU{`8jFdOjhD4tGXizDj@uq+y!>{uS=(#q zv|U1xK3`KgdgZVtw*XB}t|H%2sji~_EO|u!956in+Ubk5`OBDWBaXHvQ5E*~lvZnw zj+<3{>~DTbM?Oh8>F#~f`712Ytzt_;ht>|(Akf~jTj4lxzS;XQgAXXFK?@y-;Sg%mQysEe^hO?!g!Bpmrc*x)DUn9;uZdsI3r6j6yd$036A?jT`e$037h@!wCB~Kk- z0RA9e{(w0ixpwo!p>3mJk%-XP@>t&+CEv{=*$+d-NAJ9JJJbx)%~&4Yp+GL3(z;T$ z6V(Pj^@zZKo{Eh{*PR=jxX!9623LSVF`>O{1Z)$|NASM>#xFjrRNV@wBLJk(cc{a4qe z;HyT&%NaL1BZRetfAt87vUSCvGl&sLZXD1zBJ?|>TM)7l(q}Q)dT8^}1!<9xmZBIvg{3nDNum~t8FbMmGgf&jX)1qAe!W2igGa+0T^ab*BJ#DV zSX~9oHDrJvR%B0)g-z4b?(1$m=S#_OpXueOc3R{mKW^Q`(MKXUs5m#bSF{Rg7Z`}2 z;Jat>PAMl5{H((s@eA6go*@HR#>EIjT8UbMUV3pI$^2_Es`|AK0_`(87)iIm6s2A` z*{4VQh}J7v>zZ~)#!1;Ed+7FFefr3KeHo_>lvUVd#+mNb4WB`y+ppNNBpwc6Tr7vd z$|>0uKHn}C8N7@%ZpeD?SSog1Pdl(iJKA<$;_AJPD8=4TqHoJh$tSb=dKvq9%GaWl ziw5Z?zTM^p13}Z!523QoL1@JV{hM)ev&1VfSrJD+yhgUdxDmXO1Yt_ls_Zq3QcM$J zhp)Z(3j}_N+|jzwHm;@KhEncLEe1X{H#qM#$xt>eB6B;&2&gG%e zIm*?}RvHLBor+Sw_2Lre#*1Jp^Am;lC8p`Obp)DjUA28$W~ui6j?>(2P3Clwr!;M$ z5&fF_jMysUvPm-}+hkMU!Aqy>)0Ss$-i@lUhu>7Quq!*b9no8aaDOy#k74+BPT$iy zmW?efuE3Gf?)TMROdBdfT9!IrUK^n*Ai7>9aOd@>wP4wB9*)V#56?u1k&ZRSI_-J$ z^sEeWwr$J<%Pp%JT~2Y0F0B2v;tWX875#EPO|_f`LxKAYeB5#WbEf=)z(rBHIPwuAh%c(%%1U2H`WdUlf?|cg$+nPCYB>Gujv_?y ziwMf2!a_s?xF09P#v>Q^PRnMNTW4lB!3n0#Gep+CGBc03y0+2^3Oz6c&;SJ+tCLxO z?opbWo4?xq+K=f}rU;U&i0r1iJrinz{{`9n8;>%9nt)z$!2L4HlJ~`A& z)CgMK2lmhBdy58mTG8XToYgNid{;dtD_Y=pS96wAm}ux(+E+jlDq0YexGNkf$SLQ2 zx3nmJwPpOdqCYzT7nTy1x3a{O?48r;n&@5AYrw&LmuY4q$hu%k75P!Lm!?qLs*KDZvT)6WB52`<6 zKEEPH7gdZ-^23UA>_R`BUqU4-_A+pZ{MI=V5#e?Hdg&w>I=BCMr{hcq?f2cIo^!#w z3BO`BIie0d(sA1R!v4ni{dsrX+tG>i2@TZHpVtm1yg6nLbP-3>7WJU@XV$0ei8AH$QG^V zGq~sT6!=WEEc5hp=w%cY6PMq;5pm@rUL@T)cSem|^p0h;ImK6(+-{V_zYB_x^* zal+sWad?~hpdpA?{~i!^BqTd=b)E*_#)%>Z5SEjOf?fDGihmE9p8$r-RQLu6L#l4; zy5j6?xI1DGeL=& zK!sHy?R9Y-zX;Ho=Y~h(N&69e=LG)AJfrEvsP=TWNp#eCMP}!is0Ra4Pu8P6{h;0! z(TpwHwl18iP)>F3i+{kve?K5raS?bZJ7+OE78;4fN;2$%**GD8E`voTJb^^`eJJyV z^)M@d9eFy^Hb!@-BHoP~Y)@o1z-8-6NX#T?9Yxoy6GGYIcx2=F?#6-bA+0TO zBBAgdT3nZ6{9Uv7uTSF#Td)>hAm%9WTX#__il}c;w)Kj{PT_dxnZ&+z@UCRievIxRG+|Po5a3Gq!+i#JJIOU|>lmvxz`AYal5BL%^neE-ZKTi`kT2i?U;`wHip8`?>7EFF{r2O&|GdK>v z-byf&Nk5V7E>ss znXlSf_;uBUnF0h#gI-wz5JYekiivfOgbYYRw{futq!_ssZO0ZB3>FP46$%bzKKCzN zxXdQ5oWbi~?27>~P=E|L1~X;AC0-`xHYV66Nyn|Edyr?tvS_O{MZGEy+g4Q2l_z9~ zLi6P7Nwaa@$v4`x{o%tPY6iHe3}C_lFjQ!ea*5SkuAYGqz)Z|m+R#{vXfjBsHAiaD zhq4c-6@SdQwKwySG_d=Lt^UbBo17erE7sTmvn~LnH%pn!0P1AM5FC>=_Fu;acMABv zFqrCBa9*yWR69GyDsLYzOEi2$l&&l`C~fwu{BoJ?b~k*Gtm^^*FFD0O`OgpyjQt;U z{pDK}-WT=_&j1s^BrRPcB@T!*I)EY_VgP~;ShR#F-7qwRBB6+b2+|JSI)s39NypF) zLrDnC{rle6^Wu8(+<(A6_HpbN>p1s1*SS9VU=%2Y3JE8IADU!*=PoF-%oWu|0R+Wd z(+UjJ;%s-476thzLBK%_q7Ta$jlsWtR!Ac%o-T)w%2VwYjN(WjOdo(wANY_AayJ1# zqnHrvz|)J2KLF`JY*6zmcQP8Y#=mi_brf$gec#pl{_z(J(DnOxU*KWY_YQN23n|~8 zR2o47#StUii6rxO(h~{^MVnxE?q_Ft=mfb_Vv}EN{DmV-%e!%>co@sb;t4sBWD%MW z*%EyuHu2q=47pE)I`%q1F>KS!6bKpzKv=nB|ib8QM2e% z6ZxUWxTNNmeiaf|MbtO@vJO7gtM#NlpjQfVB!V%_Wh7N_f-35wPz|$rodeJ9*n!-xE1>m%7wDWpK<=e`u=F$G)_^Ao1gK;Qo} z#kK^kDUcHioFl?|T^ZvkCCZKD4OgfK0Q{6<;zo*c9Q|zX-RQ7lZ9xF_^RWG^F4Mi( zRK3n~(XUCOzG<<5EHT@Jk#3fEYF6=URNH7SWo~&UTQj0BaPzdXFwtrb1#d$!r}e#A zyimVQh0~jZ?%^VExCmD@19$1R>5Grlr7M#M8#M}lX`PncI{kHyp|d$I#tgyE3K-M7pw8O#&ieY!3<~303S)uEjjYTzi}c_Ala-u~E$6tV zK{A{?dtIf4#b>Z%pTvxa=*}zuZ+v8WqZ?cy)*4$G%_=2@X8kmPJaFZ&cjI5qtR&2{ z#Mv|r1HSiBl%+I~(zlT!WWId>FBBpr+$_Xk-3NAdPQQx$OMa1RfErM|vs&LX-`}%! z3Rn9F*x5kD0fBQk_zd~g9!g0NEew~FJDWuA2f-;OSFd^W$(#)^I3tgUENZkpU_+~8 zLGqxf2xqXHJOj=?KWEUo?3%pEkbDrm62XxR0MIi7HvJv?ZxhTEoh;VjYb%{Q;b5z98pm&7-C?xzPrdwzSgWMs~tzwY=d`d}tFJt`2GWxmxpy4A}z!Bj;3dA>2 zsBLl=xB(;ZR)J7kTB zR#}F*Xhckp;%x4g z+iM@x2FZxrL_RgCi9o1y^}WUL3J_ujH+ zz{@bGYwzOqI|@h|c?3m5)XYLNyr3^B5EmK_5)cHAnO=!~q%E*&eHRNdr-F$97>xwu zpu)8{p?KFonS1qi3snJx>g<5Es2J@Cd1xq(sSyBwTPi_AEqk~iUb`%N@w;x4mQ^@m zslVNW&u@iY$G;>(2x@Kn|3Frl)%!u9FDMvN5dLv`^>YA74EmO-+s8j{o}Pm&fZ>$91QcsGvJ^Ow*tgemWdcLB!=8NG=JrgS~{lKcNm4&T<%iH@d z+pSw*w7|~vpY3}=AOLcYo)(Az1yaWdROT6K!Z}s)=k7E49fu1Lj)a{k;PL@gXn+jk zo0$>#zBNkTYQ!K8scJ~f_A+MU^hN#ivGD!QRwtPLXQJ^ajyn88#}Nx%IsiM+z(-qq zmeBpZ^|ftlKJA2ky`THN@3(|#`yw;Q06xI|gaf!P08;no{&Ml5PtW0|%odF0DE#_SSZ{a)?Z5@~mxB4L(0jCUc;q$krjznR(bM(zUMdM^s@W*Q%Sy@AgOJno83PDO9|wd5-n*_N7lHlS8W|{0>)e4sV!h9{1E2?cMkTqp!3U7 zmVqZ4+eQyuK>*Wiy>z9oga-QCwvQC-VxY5`gWnQ(qS?ExT2j$4u4w0Rl8wmic!|gJ zC+gU=JBxbBm#u$jOZ{$%3O&4;Waae{D=Dhr{?-qLoU7f5m$C6J}RhMRXEH4O1D5WhR0YBi)IXS_D*!+7$cr)9XxbZ z$ozqVb^-%U{q^ddWP^|8$P)`M0DN3T)3u=|TS=-Tj2?8;K&R|kC%44YR+H1+aAr>9 zx^pJ5z@v*c&5=l{0P<+b)8Dg7&I9+HJ{B9EHbt%D@5@K$h-8C&$RTgJ9?5u^}E8C**69nF>Ova(er)j>oE+Y z(quyfH`ij+Wl}pUs}yOpxc3GKq=3b7zkbgwUq zI;j=O#haAyzSjpgFqj$0b(fRAD1dQq0Ep2=Y_K?6$$Xqz5vj@O`Wb_~d7w47LaSDs z`rlTE{LGvW&))O7^Z8Jtz`7U9icBB6Xuv&(E^|6jJ7>BKJ^#qRL}doOF?J!MCD^>1 znXGr%pUga_Wm{wiJ(C`84HJQVSv?5V*!e970V)+e?v zGZnO+xT(r;-n-X(b1P1q0mu3X@n!T2^OahqC$OvK*oR92W9K8A>C9q~&jEe9bc{l8 z#p(FmhSHWldpmr=&t1Iiax(Zclf%yeas_a_Cf{t72?TKyvBb;NDLav_!HhJIK=s@E z=I-sfw^K8lbRnO-ThPn=BNt*C%(K&6x?hv&Q)=%fcq>{h)NGjCOcaMQK1{U8@9T-a z{Pyu9k7d^UZru{%ifv9nQlu&vN!{2!!;_%B|{QEv_zqOo(}8WO}RJbMz{ zC})py98EOJIH%&T_t`7ly*t4UdjIQF$@~pvAcU(qS4+;e({@~ZX0R{yhfuMx0b2N` z01BV3G`EU$9Mu%qs{SUqW_s}{!=kQbQ`@JrHqPPQTkjOuJLH}_+IRIuh;TLHwW?=d zbLl?XMcqV&qyu5kYMS}brH(17V+ZQ}z+=~-ql zBZ`rY#czt%d78)&(TGfh9oY(Fh=w@}h5{zNAc{a}}dq|D&d7oVVt%%e5-ULM63J9n17Ai!zjlM~GQsS-@rXyT8F=aGd zP)Eso78*YG$^P%Hbaev&sbu&uxvAze03?c)Wn8;IQagCYAX@s-+Sn_k_7nA{B<=#7 z&aPSDFZrg=<((+wyG^X`U9b4f@H{^0dGmw%{^lj8V3>=mR&}%g!{>3Ts%4TRxu2(Q zDZeSfffcKhn+_P}LMLffFqW#8)T=q8#7nj_2Vj45! z+>^2{U2EM{TPt}cGnmT1PR7S~7Cf;%73Th(+A3dI@?4>6totQ@o6=}v(RscW-~Sy+ zR{`(?;{QK9#Za2?x$pmvQ!SUIWcvSlio)e1MYDBpgw67;Y6xF-lcb+?7+AREbbnNI zoSXc8^=nTCHuTP^M0sw1#1a3f*$Mgc&n%N*G3(kP`=J5_S)3OD81w*us-+|A-M@I< zZ(%rby*1hDPpPXc3p%)V#MQ={>?l6xGtb zcIsLaE;@?kZfdB%Cbk9hkAB||4fv?Yx}xSfSv^Ez{_vc$?ytsRE-g(#srU85)Q`~5 z+LzUosB;wwBKRV|Qtj3sR~77 z>+Fk5vv|P(Xip?%I2D+|3OToF(M2r+Y)dhR5b;$T~DUYBGTnnI=3Ndfb^uT1bvlbCuv9z1+ zzkhMPNOu$E_3&#K##1xf1rpTUb+3?d^zqo?)79NhOj*`u{$xk>STkDxOPOe_`NrPG z8~?g?gkyPa_l3NvC3umGhXXMR@#=kYU3UZNKQx((Ogi5x$qDmNi(+HJ^ZKj=w~g6- z^RE=}Cl%=kyqpLIv%i`^>&pxTUDYbusi&o<&MYOhQYkDpLlruY@t51?+*@Xk<_TJl z+7^-$GC@;!M?W!NrKDnv)b>*M44)ON%*Py?D6-I(Wm|-Cus*LWzO6vb^Cdm))*-GiXH&Uh0t-Mf|;HYU+{~g8SjunkTsP3 z_L4?Kt45&E#talY?DZncz{v~UR{3lCh|2hTE%a2j&~$n@{0-g5`e52(Vi&dQ{t-ILNRF3(dNlO z@f@{oU010{6V2-ko!5l9XBY4f8R*qsCjw$yQ9t z{3fkRlPoC>^fp)YquOn=-QVX)=E!!27|347rFHv>$@jPav_BvC&?K6p@PsYJMeMuP_^5vY|3HgrIFXLM;1ZvMmXtR~wv!2Z#8=o@gcBl1$Yi=#-OOj; z)(0UyiNpir5-qJPq=aK)>B~nydstLp8 zNWo+Wi_VDg;mX$)Ao|DlhoGNKf7xpOw7U56T&6IkRJSfldyEuVE&tHEv(?vXf9oDp zC&>nJRP-s}u;6i9BM=b^r3#Lc!9e633%So(eoZLBPA7|RF?YxYvG;Ah^suR$Z_`(o zZKfoB0PWjOAJ9pjVF{f5NxOy#hWkZzN4;r0z;fh^rpGlOEgEyRl6+^^;~~{r+~Z2W zB(T6}W3J0KINZ0cw=yh9(u@26`h`BYivkpj4AS}_FDPxZs>|sX z2RXON)4wPAK8ZWB93tcF!}e8%hQVJW0^yG4p~C&7c9mi@+_$}1F#n5=ZCKcs*r&~! zbzuDWgj?=U($dXaX>_}huQyZO)OdMW;~BwwHYxhnSktkfh}!u67#r%T)k;W8A6{nv ziMR-NNXT(VMj~7$gPv!A6r~7e`3@14TkFJWn@W0mMEHJT7xN7h(O!zYA?}Q{Gp-vS z`2jIhqGKUM-jXK?5uWrR=_uhZg|EB*29)&I?2E*ITo3GL7GKlP-2Mw`?p!{|GMN^L*7o@= zE7E{IC>`2S+vCdJ^z#2h6c z)~?9d(dO67>FOg&$QM+O&-aWwNq2El3&nzT>%PwHC4dQ14_9n1!V36|! zxOQ{!OBC4LHN=Jrwk6$Ay9aGLgcMQ(kra5lBI<<}^c69*_s~yuG}PY}!X6BRmso3a zL24i&!a5;B&5%n+VN#G2~jx;Rp*N6yA;MD%h;KMx?OFp zE6&1>k=Y?aoI)Y5un@Z-u)7&P=^6NqtNb@?ScMqMgDWz)JycZ*A4XQ?*m(^tQBArC zwxtnRoONUZPd~zh%n+qH23cb4G1Qm)Tnn#93MhZ7Zsx3wmRwCYwLfm!mDUnf0 z*-`0{u&BRLndXsk)KCwla0#1ey6|YH+~}m0=#39GUU6GT(Nv|F3@?-?SM-188Ua9$qi03^mlBuO!3rhfhoCIOk#XJmzee2N< zD7?qF;@JHoj9=#)DBwx9sV`CIFo= z?bj%<%rV+q{QXr%!ZnNcUXkwv(%#$ky))E>&|?6+eS|m405}00?Fu$He#;GfKmRrP z&q{>rX!15R#oV6YQS<(lE4^?Z!P`5<-vXp>pKP>>wwI8amqWNv!Tu%?1^f505()9T zDeuYv%&4R{765uI@Sz~+4i)r}3a~T*SrF0Q=HPY3ck@*6ON?^%aq{F!VqDF;gw?m6 z6o{!HKp60mUNGrlAM`#2B!h{g1UOSZ2xf_Y-3NKn_X#Cw z&XNyj6+#H-LXkn>b#s(075tP63e1Dv69fg4S=_M7MkimBvc6F45u?;EvJQB8hpcM0 z%$?k<^MNcRRRcU5sa*^5CxKqzKprNlQvDE@lN?v3Z|^*aS!UT5fGipBC>aU8m8dN1 z+AN%84g~$s76*y12H~gxS{{oDZ9Mm#NVae=@vcL*w?mXK6WBjL<55S}JS8i{0cEKC zwdRuEQ!>jXpL_y03$z0<69h6Rf}UXTuaxoi^bae23SJR1dT*Ocy#uo4LoEQGp{9_h zScqf|=(!;4YE5C6q&c|u%l*Op@7x8iFa@tn3aUO7)kvDx>BS%^1v`CtjkTYkriE=E z3S~3FYqiB29mT!*q8(R})!!g480DB6_1ZG9BcY&ogsGWeM^Qn%#zC6%P!-QuSC0YR zwT0Wv(ci1HzMnck&zVS@5~M~;h-nFld4ePa1a4xHyCzUy6DS^|Y}r>z zQUT9jD3enygY5f7&`8j7*IU~J;20(2IX=K7u}D)NJNb=8)Ui~&pj3LjwCZY^5l`7Q zeP{#@`W*usC6%c^%;bfZ-R2DtDf$ zSE`H|eT+Fcrt&_fN)*id9k2r{lhwx*z08>A%9bI>m_ewHDX5P9SN)a>h*B-x6h!W# zV2y&(b)=f~b>@tJHAJ4;A3-%=A6^*;AcFK|$e7yCov1ojJ|kJW!BWIdlWV1K-57bss7ym_d{4nTM~y{BO?CI!NldI@>?tuSGTEvcn>;G^S65OST0LZnm3*os(;6 zvd16=F&Fg9@)VHb>hqQgO~kc%`L+4PwvCjyZDSqIOs?H;M%b21cM5B3u{jpfe7Nf5 z;7fvtTsODu+t6qXG=K_NtPu!+wTHmC!=2hAt&q{DaD1XvFSR{U8kIox^TB)rxmz~g zIjV(7|E6X$o(j91sC(K8k~WZ~YMru$Y=3d>F2S88m7U`R*deaw z=P9z$57fLNkiyZ@%G)lJC7EN}Zr9kQFN=x-a&NwaZD6~LrJN+;79@x~#L;WwtkX5Sg9~qxW~) z?24vJ_wLvC{oUxxW9j*)cKJ*iwMp&ea7L_k^?;}P88>BjeT%N zq&uTQIC`Wj;8h>=4vKlSp5^Qdt0bMxm?q*E%>))^4I46n|DrI=22ISNkc~mG++g9H z9dAyJvAAY%d(q^_sWCP;dL%9hg92S2PoirCS)y3R3^|LwT;9S?2VWljLj~xf5FI^R0;6K# zL)cNHKDIT?tnwr0S35{`BlL}o*_6u*{6^^8*8#de=p#^|EXrGbZm=a_Q9x*MB92mJ z4XXtp!{96bP_PywWzz8K z35l`R1YTi1u|h;d7s2rXuF(Ords8b~h*cQ+6P?DV6c^Baa)>Qy)q8r??0nVo6Jn=t z(TKuuTM)q>iyXn)+;3icc8-cIp8b2a!j?$SPy(`~f^5i;in1H^TWBAnMG?cyQ*jZ!NJBRr*!o(TW zmm5k&O`JKsXjsX~`8%{!7E9&!{*xbu%o_GNjo~n%bZwz`?~+e3uSP0{PtoqKL%qhG z7X!~qKc9hC^>}8_DBsVbicgs|?0FzFV?Y0`d`7JyFKqne{l7b>7yv&Y_5Ul4!9>$d z*D)1t`9ECV9o(-3+J9VLchM?GBsuu%i=NLn-e8_|r0Wh^|Nf}t!LOF){;V_QTIj%# zZh7u+$FsZnY#pz1`oEb7WfS6+_og4$ePS7Qu234yJ~Jm zp@~{C97UR-xIuaI+~??;gJw9gFmC;J-AujrynAnqVsA#hCpnb)@r&)P&54SCHIIH^ zu$5!Ddk!CstpD>pOC4v5Kc!W=FK+G9augG4jzh{7alRNNcloo<2zV#|jejMNeny^k z(AK$Vr)}bUUGh`yBF1X_?HeT6xWi@JjAR3MYAq+eW4+ zePNU;BnulTGRk@Jo3mqBXT#VXpuYtaId*5`g`=ypL=x>n7^U;b!p1Q&au5+Ew4U5= zA&;9iZArdKCAT>&tTGpqtsK&R@dQ#Lqa?=!)4n?DD@ZU3nD%Hqd+&`KTTFl8cR@__ zN%Do>v`$?HjTpq}&&9f7jW=du zJs2~eC`q>F*Zr*ey=GSRrX{JCZ};1?Td{ z+)Gt0_Z^>BE!9e0{6QzR!PUaw=};Q`ytI>kOpV#1y!MXfCi_pYs#D3hc1ID(q1}!n=E;I+44Z@^|`ez zzF~7|osImMsW^Yr`{YgJy9e>!OCB$KYNn#T!F-IqJZAN#Bt49;St=;os+sfsDRXDT zUg}QmX3BiVzltKxO;z2J+cIj4FYZ1%9vU=uK07cA1!^yj`|fULbfyf|3-=D98}^qT zB&h!#_XvzQL&)MH>3{NvmQOoBF}va~MUI>~O{yNg%+ZVzcH6<_a1x5sr zLMnjiNC}xWuvvN=AumYZ;HufGD=#OM4NZDNLcnq zKPYeFXkTeVMMa`TPMQ@42w==`YK%gyEN5M07Vol7n>E7|u0|*rlPc6Y(jOrd=m*d@ z7v?B|F1WEUGFxL=-r{w_RJzD<0`JQRSi}HtxYL4A?u7i0tze|zk4A*t zYf<`1Kk(JBCA3(1JuL}6#w096#Uv;sRD}YFQCAgKwYR#R(W!Gu{;RBMpymD-`&e=L zg{v?p($mZf$E~UiaxnY5nk+Fod3p~%MBc)D$-d$#Ven9eQCmlog$Vd1nW|w-uocZj zMr!LCuMKE${mG5tY7v%ff!-ATl4tBGdH14?S$aqa`@qU=b$=DqjeQOD<6EJ8u#mVn zL4rJMeh$2I#0XWxF1?ZE?wuWg(Vg{=Q~<$h$$jEH;X+Qfzwo@L%5CjmS=atWjQ5>x-RRNz8|k5 zd}8T;?EWg|;ssSH+>C~ed$99UNh8K9DP^vaE`hXv5rDxea>81(RWASxx_X2{oNood zh2bvAQB54l+n2v!g6UyoFk_@9H=7VhDhUX4YzXCtxmBA6Z?R>wxu z!Nf4ZbmSwzE4wf@TQ`vWf>)dRxh#RPX_}Q&`Pc9w00J`!;^D$F2nT7fv1v3&{FV{#G@9V$oUW{^LT@5@ADwG?N~p%c?cA`hrHhCO(X1 zS00un@PMD_YoPBRg|ctp@J=OC79H1*>KiX_S;q*O-cJp0eqn=Gl>a;DOlz;MT0OJ7 za>Z}pv0!;i^AP^>{wyp`J+0#kd1{J~E#1Ewn%gE+vPPXXwtK`KwSBrwbAw(ld~ce4qhag8%n45ViTcAftg;z?R3SM zcI+k_V?nXy8WRw)qGN0yhqe$H`oA8TizT!fKzNq{EWc%g7`%`hM7W;pomqwp`ww*g zR_QS8AeIG#Nqb0>?Ls%6BZzFo7WpPQ)qr{iU zqN6yIu!V3MDuz3iZy5kT4noCJ0sn@fVS=axidi`tAehQCyxuQT5nDvKo!LuBrX)l}T5(n!;pXc?NOoN1LhK~F{l)mB7=_ zru?Hzy(p9l*GP@NX3k8d38uu1ArjIEiFqmF8fkPkX`3NwqM2#1#x&3@OU7|p zLO<$@?uV?^REpbs8sWpcC{%IYWs)*WX$`7e7gd?Zxj;=1=ufXii@)g)s{5Uq?frow zo?f7v9=Dq^!t(Jsje`0eg6h`g=+#B^=OwR``G+i!!^cS1v5!HipMKh>c;$WYvq$~% zmi83MSQrB@TJI_{{6`{;OUNC^5ING&gTNw?O41VJ)ZF z@@+ba@eitiip$e>72HP&rRT#ld@}dbGRaKo;T>=*i?5KY51s-3%XuKNU<(tFNy+`9 zj`WN6$XPO@sbx{}8Vs@v7_Tju9549HRGi6Oh*2(HbjV*m0j<^+KP9s7t`#z&xmNoa z3@HqTt_(fYv_0az4azq#Fqdc<)l{3=>zNC-k#?*oMgky@$;HgVZ`a63`A+02jf~{| z7X%e5Wza8$IF@dNmx3}&9S;ETb>V$1vXX*aIAT5vgA4yDU~VjX%*4-7_sNBePg5Ty z`4FN!Td{DA(k-~xg}HNW)wD*b;^#?GaC8MFvjWjy(fL7CsQTOggfx>z1=mU3>$D2* z+6t{sNhA^ECPYLu8d>Tg-_%)n*H!wrR}ogLTvAZ2n$;EVh(<5O2h|$Y^=iM=a`I}# zn>-yfkXtjkkMj9fM4YI4sb8DzTkAw671?}Y&NQLVtf^l|egB8&vqTeD zQ6AZKy}NZtO5LYXRO33nSs(pt-LALuW4~&{tNRVks`aCF&&KN-$h5JB930cXg9c(p zLybMky9L#4tHI%>eIRHOdh>ruIOwlFPYMIaRA+Vi!H+Z4PjWlUekbD85p|@hL)YI> z(}?QBO3^a_QLsj1E&!WVBZprjrvZvf4Q1@b|E`f&1_(SSA?B$3ES^o`=}nTn@P0u= zKLGG(){X)M&`fv--S~#xnpM)9)i(Il>zl_>@JVun#u9;Y5e>jJ(FFOrTuuJ$w3zVz z`s0Qeryx!)p)UM+B;bkY1Ry#N@A_Rcle^X`d#@Gq%q*7ZSBQdrgaQhcP@!_-l!{;0 zr@vxRtxu&}pBuE=6}EnOgK$xM`Bt!vlFb4HAlifAQ8Wz1JP5SDCQ0D^{Z{RFBJV{a ziJ7*qxvc|LwrHogwnx>s$DOt(S#@MP{jMXyCK;|rlc3Dtr$AC94Rx{YVP~aZXZ0z_ ziVTW!|2;-U0I0@GL}(bbLxiJq9@FVT=Q1{UyghTUo<21mq+nN~N!KvUN;!ma7v2p#q6x*{_-@6gp9ixNv6NFIJ`dGRholVsro*;`#o~ z++>%Br#ORi-x;QR3B`C)oq%ct$AqH_XceD%gTdAR0JkR4M?F2`3L6-e#|%g7&Gko$A#Uw z2NFDN{|U@Q0Q_kMFk=}XQXtV3@T)1%O2|l%C&ITI^_*Km>1@O@@lUDg=s5lGjP$UW z*>Kz^6d~@qFwQyO_mvyI zvR4OG#o3Qc0UiP%FZ&>5+UD2;z_`yQDtk(%-I=Y!dX#r*oC=$YEgE=XIJNSGzGljS zo(Qm~f<6{O9t%&lsE;$7PLI1!-^!fq`lP-A?~CG_*mmv*odPN@t2c^0FB%5elflkJ z$mto_LU{UD_Xw{RxJxpg1bzLC8f5(tVoqq}X0U z=w%*=%iQ+{%9ANdyfX@I1o0x<*7)1=62K4t+qHO6Az3BD9}M!9VSiN@O5qgEbb zn8c!Ib&r>LX|s^~dn1XO%U1D}$WP0gcOKY0LZSH|+^0bAlOcDOKpyNi7fm}`cLDQJ z9N>WENaw-=%~fCinUylI0}fkt*Ut-*|~heeQ=A`{#E+q($xeuntp zbIS47yMKtqS0u2Ts~zCw*4IbZek)T;VDKdhG!}L7Dd*PHOW-y$#k)Zu8xn+79{E5T zWlklOWyfn?_}8hW56pf)JqbIMfJwREAS-~ zKQlQGo0lN=*ZY>FZgx#(Cjpk1#xEYej1hOgM-*f#biMZl$3&zswkaMbU19u$V=ACB zm1CI&&H14~rq8b3Kv%?=D`Er~ko);0|L4gV8aCp_J9G18@Y84(CBE{+a@`nSnlESl z%)d9Xl@%r5mSHYsH1@lv&cA`DKYpI3-$V&1o%*r5V+P5V0S>0DCj!D2MT~s5E{@UTc z%i9OrGjzH=0giPi*vWQ*jjc0Pubu7Y1XWr9F}-GQx&NOb#JNEw=#VPF5JaW}#GsFE zCyAS8{-n{)cB<%tXdpsxFpz=D&mck~(}7W%|9e1`j?!#6*K;V=7yo{K7C+M5F)Rh~1`rcJWq#I+3xEh<> z2y5+P_;cHsxFY$ABDkjwjhZfwr` z{hXgmJL&rv$Ov8OGTz>y9r40PSsT9%(A^62g;#}Mq|3JUJ!SUaA9E9Jke6rIfm95c zl}Ymr-^Ihu7YH{Ji0RybAZWBlhG=Fh#}tBO_g@AJK0$zpv}I~#Q+ zqjuiqkab|)?P0*RrPqVgCezjH=Wn*NXxFu=xdlD1ciMWjyf;c;Y}Mbq`I=vSWhll) z{ckMm%>6^t(B9Ldg+ROW<+^X%|F$=fKTmdPL61*m@Yi|e0X zp~t~~qXtMa1*-|(bXu;N%uERWIzK~y)~3Md5EjCIH)G@VfEt`%G?>3j0NPE#&f0&c zTLQExtIY|8{dyT*<}nMVb-qyr?HUDM>&lpSuhw$iC-AW|2s4K(-TcQc8^O$A!y;s^ z$+t`5y&n64?W)eL5VDESZG#yOv*j45Pk$6HYz`T$(<;ZrVED5Upp+c^>uMMEh1U9G zv2mq%0A?q~=II#ML%Z8uY>Xh50UwwTIY=(o=%K@@)P1>qD7N z&$J%)vp~uT!NOc;sdyo?U8Ht}ky6aYSIz4eI!r&5!gEDGO3st@6suX1OXY1IJ1M7I zt?epM0NLW~I}s`^La8^+p2;TYCU|wdmF#L+6#L9*XdhqtVWj1GUqp?eY`vV#%#hAy z&w(ztw}XQGvX`zE<EDWC^;=@*F%QOQ86oUHsgVlMKF+FuWR=XK-`%)HvFBWz7ycU`ea{1XN?EOGBt(4l z>ijeF+|bePck%10f~nD?W^H6UMD>S|UPszt{Zd7%(opBLn?aFsObGdg{9Iyyn&BfA>ZV+>UrihMrHWQ6><_1Dw2nf4vMiaOCY5T>%PWOahH4K%{GG zGk-N1B#oV8`Iy?ms7GMBD@Z`+1_8tf$Y^7OC)}*LH!j-&wKPe=a4RC6awV48mI~nG zscF|yaA%N51&b+;sa&vQQPPx0zGoEWLAGnFwYufX_>P$f$qfy)2q8RkLj_y?0qSD5 z@gb+HG(mw;UAcI-*kD|F zCSQHK15gzjQR#=KYPvStPpV1N53E21qTFDr2ew?<3J0=2C2RM~PX4@#uaGNEMEHHx z;$w(hr1t^F-op>8aO3cAZ8I4Sm735WD0jsaJfrkYn#TG9D$B2+qn^*Rv;04VoryQp z{~zx^mf4uizB86&Z%86yP)OU@mC{%eLPPd-#=aZ7C``vraz4zR6?mzH;pYuMi^FFWT`FcK{;>@&bjVWgYUHOZDKljyQAr1-6O!WqH z%a2RJJgjDHSaY_{hc#jLSU2%m(d7T)noMS!s_D8#2^r@vram8EZ$}-nTlxy+gEE`K!q_)n+nS_(w(@_ zhuOM)ri&b&RZo``=S@WurNJ6Rmj}w{UykLL>c&WYLLYsmS5R)HtwnKJHu$xtYtf!m z>qK5TZS+#O;6gA~`@k9hI!6b<{?Mm&tT~#W8)+F#GGZAcZx?lm{2cz({^#R2@x?CZ z2C0mX0TSX-SX0Zz=%SXauFVjKz4^mNdvuiZOyW1sPa-3`&ARQIa`^_I&0mMSebmO{ zZch9ngi0Kaa9*br#;u&W81_!Ua3|TEYx!ynI$(FvmuEFB|3}cTw9)AAM`C_|wRsly zvE(4I;E=_fUVP#c-us^RvJk_$rYCXG7)jRH6njg#Z2#SjzoMzSrj~UgQDmmeZxcDw z^*T1%ZKG23B^7*(N<#(RU*uRsACn!8;`HZG$wgoiAl>8Om|x}|OkxrZ`iKtKE(~8L zMSBvXsv0yaDPRF20Kx;j|1l0BGN67e!1RXDcpUf|Mpt|ay-P!#qw}@`C>9okryy6E zs3yGL4wm#4hX(guhc%NByXI&mvY=<^7ZO>#Pt%_OMbNombPhS%KP3`G5eo64#=6jB z+iaq+yTJDs=)ZV`cXAvHk6ra%D@oDIB(eoG4WFC{L9X0@NSGc8#{=1-DLU8`1OQnhL6g&>c9SrI$#Ky1>ZIgU z0wmotJzI$rhK|jy0*0$n?yjVufK(xu42*Uvst!>NMSX;#&NJa3^AHT&m1$yTXjSIR zsgTxWa62?>5|h5;n4a0k0m9_;=>mnj>4zs%#y$Bai!wY{P$e`NPfFZx2kvbQa*NGW zuyM^a)kUuqB}Fu6zS=e4v`F?A$x5-v!V5t{7r)>Dk zQ6)HpPgTqUDR&D%4AOE1KIC$4Mj$tibk%Gki#UTK@su?8f9aBr?9%eA7A_Iz|gZJe`qS481J3y>WNhNy$*$wdSGLXA8K2{08ASuX9> zMXE0o9ZbGr`H4?-3;2phiasvC*?0HbRj6pYjszf%Fs~pV=#gv)Ck}*74pLNrBtIHHEA?JVJS9k=qQK2Mp zm^fap9Ll*65Jadz;T~6HlOHrxp(tdrsi(CAXa)~*3RgqZ)0w&qnVznied(7= z;2p%&E`=jS3u`@A0{m&s=_Wr8}BGNumJs;hSrKY$tMWuw?||n>dn&X4?n5j2gB6= z)E{837Un>8-bNVsi=i+z&1l|AEN>0gs-s}4CL=e*Q0fNc+r=oSr zCmzqoz-~Le2#k0UybjBu%i6zvc8L?-LgRgbZH{+pPK;=NMdQ8H`M~Q{6B7-mGt5x` zVI{p_0r?Rvcm4p)ETDzNO9x)I(Ibv=aGc*tU8XnH42XQ-6BD(Kl+C@n1!Wh>BP4rO~Ad5wdQ#szN^6AEE ztkaNaF%svN{4S_<9N>|N;@%Qg$fiq2(sw9w zYuO^#tT;YU;V6RVAg)7TO9G&muF(-6&~mcdve&jcqz9352Ax_$96?Q{i9MY{J}7`7 zzsuqs0T785t%kp@Uy$_pK!ZOuL)Cri(ojVwux^=bpfv_d*D3HbPx%^MkyC(&c4Uv| z73=|8j)A)|oQ4o5!pCTxjs`Fk<9=pabJO3Hw2-d6X)>4n{8{I-JVl01= z5Z6xJT;_o#VfzctL7UG(i_i7HRO5SjQ&DsreS(9-(0V}C+L*tf?`CADMtPH?4tmri zM!oluxT7|K2)4t4!fE&{{y})TfIME{C+WJJPhZm00cC!_Re&Ia?)x#Tb@ z5NxayRwa-D)@*}ZqcUD!f5)i_&a%kQ@l3Izj|EU55fq5(jd9tGaTF#cTQ?`W52(a| z8t5Qp2t>r~y{N5d#%Y~22as|okB|$H=Lvv-NiKR;AdA}vfLzB-1P~^O?;T?xAs_m| z5Al4Cy7>WZS%;>e*Rc?Z#YyM)jy|je6bs}{K$oK7^Q;F*O|X3*N1#*+KMfQR4yn4` zpgm+DUW3n7M;By#C?5KNkOg~@IOXo7bJ0MFw6cd$npGL23UCE8b-sWfSa1Be{V3Fl z!l}#v5QKbhIxz@excYs_os;pVhVgx8+K><8!={1a31f*A?sUsa4G<S5R;r*r2v5aN>wAG^=%D=UeN%0V=0p!%-@A}RkeflwOg zzlgCimygKNIaFm}l5A-N3v!(X@uq=$949(k=h#sTtgBMu8sb+cgJg#_#Z70QAZ8I( zE?NyC&;2zOUy`vJi}dRY$A+&h#V%HG*!?*(Z7{5<)ZURueh(*8q2W~Q->E~xOQM@g zP+0?;+9gNlI#0Cjfx9mG{pa$9d!+dDuB9N2yiaG2!KlnH#J^txq?dy?V36)F?izdw z9j-qxsE>z-N9vRm>!eqW`OVT%qHMY^ig86WX&KppIEcI$F((u~mrMF95i30_)Gn4( z8V(}+pWO2JkXzal-oU3v1({t&I_l@>Lx0iNJqb4Sga{ z!1)+wo`)2JEc#y!`WgnQl!ImKf7|gxZL=gt`1pUv*Z0Ewc0*-G%o;vswH{&==&;A@))aZlcx*C^Pc6*Dhj`J_gzyrNBkxCJW%s5>6$v zV9!{6Tc|+|kzh7IHo#RXNFLQ~B^qli`RjDgB&Afp92wB+Z(G zI)|QVMMtxMi;rsG`~M2N#veKVYlHPmhi`WS-(#~4Pn<^`*+<-Xrja(kn<=x`_)RwV z+8(b<`pY9nN=9S>CTg7(aR8`xU)-0i;ES#MHkXavFNV)jk>Yoe8!11%%9JKX5YMmu zk^Q8NWRt;n*as-!@BBA#=-MCpk^MK<_IugL{%8BQ8cbhPw{~&Ba}U&b7GknY>BGf8 zGb2b~3&+I8f@GSFi}13YkRZ5WyeQwNS0TWD&CcYr`Qw@M15wHI9>uu?YNba$ku18-zhVPwhIq0^u)v&0vDE}Qsa=7K(9=^^@5f62JlO}dN zSmAMyf>2tqTG^|0r!D9wl-}6u zQ={Ws+_Lp$Xq5A0Y`h%Lwb1?0?avd>ObRt}ey^{8_;yHjzPCa)%K>OA=|HCDfnP; zFG1(B>+J{jmN$iho{z@PGOeAU$CnzMZu#ERnU(GRF+NL1TRL1l@*zm%vR6lb;py3-al!p2@w_pY>J{G2p3{@)Vulp42$!6XO*nW-pap zaw-!8*B+K_imjVJ*C}2vYV&JajUfqp`qaP3XRS0{lnOGfd-hWo^M6XH^Zy?bY9-q2 ztANk;@O|?wnXdmzsOuI_Cw-ecBj`9<(VMGe^Z2`{mU~x>#$o#_EsVQ*#k#k(eD*f4 zpUXKKB;NS1`9!5fuFBPc#%h~5gEGr|)0!&ppEwF!9vaABpQt5trAnP`ditTk^WEdC zZ<}gnUJw@t?w$Rwq5ay&w=bT}zb0|Pi*ErALoUgqq-$pj*dchWiBJVcsOiM>z<3U`uu7HgkJ$o(hgO&#+ z#`2Ho8{z0e&_c}qn2cQ?c6B~W)i5#Es}&G=iZ`-@8H(t(S(g`b9+zi6f=hn z+Pj&+i*=)JnYQHeL!x-gJs{u4MelvVl>nj7dA8Y_J)*WP3lAl(R)~Em_H7qiF7z7H zdtX8+FXPK`#gST*)X6_Dtz}qa6Z}1^2LjDi)~FNDLcN+IP|W-E;AuZ z-EJwl*UnrDYK)<1qa-_ft@sRBm)**eE>Kph<{#Q{-+y~S;A_)6LHc4#L}SD^6&L*Q z8a;x$gY~9T+hlw2;t$Uq#_{W%QKL8VIIjHNT`Yu6-JBQsb>{k!zq?B}vaEiJh8|&r zd>`x@*0u8cc+7rW-tOt@zl=B2;n=x}9up zbk5WyIWj=|SA)5(uCbYo^qAPE8-B#KwQ{N}2-`;l>8>1&WIX^&q z<_O4FxFt*P6WNvd`_E)H2xb=>`9{KA&tram=QBJ~KcdArm2C2T(SSmRzeqk>EY`nl zf=d!eZW_(_POUnP`VB;QpQ^m5m!wDc&nzhDRd-KaEm#50XNcZV(YWq)c7(&`PEX#$ zQw`O&-``muR9!!BabCrGk_Ik*m{7Hh^}d{E`Cb$Iz8XFlYyV=vuEseP9giLm8>aH9 zd&*+3p1&#FE^dD>{rS@ypmYTl%@j)mdTnUilfwosJT{W2`1@xPj)2cjb%GaZea#>M zML@d0PwU|9tP?-$CSBY;Mc<;s$32c6;zL-9S(Asue>U8^isaUrX}xu7wGsWp(_l#! z1vaEh(=uvcH}st**e{zR8?7^7r%hYa^0IWS*3=>WkSoEFi%s1@8U9&s>GLXu@#QYP zX3=M71UPM9Qak+2s?IvS?>_SC-66lAwkIEXQ(wGl^4bi$@m72(^O?-z`Dfg33afSp zSaL$>%`=S(Nzt93z1klfsoLNr46lXA2k&-@?5%e|#tb^Xk@ zSs(e`o8c|$r=#Cyf5>s467A#b;ICNh-T&#{m1AHr!haQ94u zqGhh`?EnKn{J+G!>YN8>iw^tpNIh<-8cosC|7){FIHXHdPqGx6el|b)#L;cSeNAxf zRW;#;{g9>3jRETGrAtv%#H?Pd%i^5c*^s3VF~6i|{~V@$ogoK&o{(zY^0RA4VNcym z8!20)A6t-9Jh%7hW6S|s)nj8y#=SU4HLV?5-niP$Z7qNudii{PlIE0EdXHB6b!O)a zpM@E+?ACl)M=kaIEc5H)59t2e-kR&6G)cylnROg_C{#AlEl5{98Ix*Jk#JbU(GE~a?fH}>v%e#k zcx=eg?fP+(JU2bdEv4KnBmSR^3{zpR`{ovkmnL)6boTB$9x#*+xt#lJx|DD3(Q>HJ zE_`(<*DCXl9p$SdOfk>(@fqO~zxF#1@0yQ#eJLS&r!3orNB1w^i&+y~xnstES?NO3 zb_%sDOHaKkP9p@lkQLTtZ4n|jejHtdZJon*W&&3QVa5}eEmrA{ilsewRY+kChZ@u zG$ou~_r1%^8CQRF(094EP$v>|qI^TH2zrt!1NV4?)tWe`WaS^x;h^V#k{To_!B6OANwz)>Vg z9OzIye0%8VGCTk*_e%}UtLEt5!KKiM4~*wOy@S&`SGbd5Jg=?(W8Bu}JX(Ka>7kGn zs`baXjUZxY`RG@rbe9{Mc6P=F69&ML77EMt%W7mlh2yxiv2Et@$Y1QQzvfAQ&Lj1f zvJ0Jno$ejFW|wv;>$D7k_mbTb!Z6ez6{T=S@!u3_Q7DS$)s!Go`y97rQSnj%bt)ib zo%UMe&ob4G$ZR~wyHM(jIwx{bE@7n{vLa+m{BzD&V-_{-b@mtiTSA{KWD7o2ggBpr zvJ=|5r+z@!O&a7gmmELhezI;>A8f2QFB8zWiz>1*G+AC8J{k5;YSJbX-n@_~wp}t` zS08xtk0vT+Y1dq5ci9LKzJ4O~@wKMS8(I6x6f&EaZ~>!q7&hP;26;;06q$MrH)LIu zj!(a_v5+-pg^fXBWcM}X+ER!_d3WkL9%h4meCK$s$h| ztu^r=h0xJ!ScK2;rn^a$OvUSUcOYV(R;ImjG#5Qmm9&-3mqF2v6JMm8ky|x0<@?KV z8x|sqpRH0^RJmjKSGc3Q<)ua~uHTv^>uh6|va{pwvprT5&l*w@T-iO>h$dziJl^CI zloMr4FCnh3RKOkznyF0kAtGt@M{5nBYHnd>mw%5GGz%tc1dYA*`8$?k3cIdZrFkwi zPfNNJcSx^!E609%G5`8~k7G3{LsvG(6h5l)Vjr2X-zay+iI*EZGOG3e=d_(WhLMl7 z7nLspmm3O*tq)GNd^j5wYokQwzOJ-j6@R09QRHWpIX2eVI8;d`w>hI?bzL2N6!%jWtboqMby2M0; zhC;)3d5yJG=!~a=y!blJ+%LT3vZsn*y=mT438kA?48%rGF@JnDGV^A5Qo2Oit01=| zRGl055`oX5N!o&2fP8R6$<4lR;);6RyLXIvFN<1WX`@#YN$)YyZw5lTc%8E-g%V$W zfHVi{oO^@6w5fN1qrOJW>x?gO?6qB&@K8bnoJ9{n&V6Lw9Oe^db_aZ??`-s9mS|YF z-f$QU*yli!_l;5ScD@-5krL&Mm|Z{eohclWNycb*&+kg(X)cnLg^dk8JUOsTrwTV| zBN8eC*yftTC2E!I@rrRX;rVJwS~13K-ld&xT;k8eS?BHyygx6qb9wt1?Q4-(Unovi zngXfxq~e1O(WNs_mQTK}961T58&NvaO>lCk1dW06^?j81X?V|TU+wAL{pW=Sbg3~$ zkMe~r03RuLc*6qzhvmMUt9D|f;wPlK%+=fWW|zpTsW)7HTcj(4aw1)Z!S&f0K3A41 zMCf!7mVZ0@(%XJ{85X$OPs-bhSK;)A9%_)&X1O4P*2>`3!p-ujXQ z(e9O(S`{vto}@*IZw%bWzvX^D7L`g9MaCcGrm}bj-PSA6Zxpl_pceZbe97hphs=yq zdi4KpWyG9PfM&rgx<83$2UAP>B~$ev;XOS~mXuQJdAa z#QP0;j=(8A*mn-hpO#@uEagLa#>1mW@wZN3rRw4=T83rNDDqgX&HF{b9-iJ#_t z(=6c6FJ*sYzF-ezfB-C;1j%MXf>>Zz7Wfv^C$0L2qh2b_aJ4BiRw+(<#mN2sHCNq4cLjqKI<$ra zCw@M|AjY53H9THL`e_cGpNzLslCiho;|7p@IJnwMR1^{Hi3OjwcXtNiPB%g0D2Z>W z2w}FuDNf_FdKO7ME4YR1B!Rl5J|a914I?Qf$2%q`S-5-RplS5vOkI=xqoz51qzrsg zAR*;FIR#$+??^llogC+w?1@cHv4B?XCO^?VRntex6HOC8CDquBZkkFdYfh^OO+M|B z+EJC7aVouGDfPT#O24C|s4IFXIqjurYNAqxCk;xs2(0e2SbUf1k4`%83BA1=bpr=> zL4zAaMLmv+aPwu3SY)<&W{y{7c7|qbLbGmr#yqot-uREb)81|d*FshcG_`dlZMi5j z!;!LSVRl?J!HWV0Qvfb3z)JyV&_I3|Fo6M@>ytm|OM0=JC8A7uWI>r=Pi0HnL$(XS zJPa-{okNlc@X|nu|22BTipCH?J_NKGe{RW2x)3bm^NQVT$BYe8h?Qsp$}{-xsI1okGEv&_2*yR%a6wcZhC;R9spth zUPN#d0pdc<3*smYsLqny1_Z+jM~ezqs`5?7^Fx3lFuDLn0Lt_L1R9iq0lA{VJiCA` z(vl15@#>0 z?BOcNnhtitg5x;ujrWLfXx(9p!6(9^gf2n!KR}}BX&P%4I=lChaM6=YBsUciumfa~ zNK!rd85A&<4k%LrK^iC&fY>uZC>E$VFy)kS#Wz}#;%}&uc(O`(oZ9b-%c)?^VrmDA z|9Ej_Bm)x4sPw0S0?|3%CwQ&_5ES+P*~|QibafMBzKdjzX+lLR{=p3rgunQK6Hlcd zG48#2^i#p(ijOOgd!moBEx^~P&>L7DI}9iSFghE5MS>jfM>{p_B?9lv;`M(N#siP| zof7?Beq49r34{ZD0|O4mob+bYcvA#a*3QQ;t8=T-72@#wFQ3Lt zSBzYCe|n;__9Z{f5-#ak1MVm|stXCAfdd&e*4DLyP--^;qBKx&LYHcJhsIMvBM{?m zp}|Xs(H~qYxMv_iXjf2q-jg$sKcCMk$evX2P|vpFvpzusMbMzHXeu)_%{~a<**lxK zy2+5bn?$q^y9IOGnQFYC|2%jG zHu%B>5G5g!afH0!-n3;ec%^ig;y5ZMlU7N75ms zX#wR<$IVMXZnUdaPhQRsyhITY6C~bBEKdysT8PmvWWusBuv`E({i<#DNn0ta?GBs4 zQ$ygLB8k;^wl=eX8Y1i=<_7OpTPXwfBer>-g2>>4w@_aHF5&GE_F4Jy8jz%~9J5}= zAgF&(KmGtKJn)q_XgaC3gQhVM;C9VLV9Y;2)F;5+xI?4kxYIh{?*%?9GrkWL1l&qC zE$7G&An=d@a*?EmmF&_o=-R>}85D#`q2)&=!bx8$ATB`ZIG-KgO4@^M1>h;TZ_Is~yyL)OWmP71@~iBMrDNHm)SJ{I*(=ia;1iUtNX5I5=zTjKNo z58qZGz5L^E4!1^#j04-#@N!!8Nk4-zb4sbk)Le?Bi*PWjkjEZZ@nDL{|w0@mA z)ZvhlA*5_PJ+_DHC-%N!N6q#lW&f>9#_w~tP#=kiZ`{BfeL9FWjdGlCl!CNQXFhjoSv2M8V|m(( z@c$L4r)|$O%bRT5d=Id^VhM9|Bne~W;Ak0FZ6KK(fXOQYAmdQ8KyjC2_j zJ=zG9L(gK+ikY*K)D}N9;J$ct!H&Kp{1^F!rgS1xRL1$_^fqc1y_9p8 zdCG6eqSVatg4sD_y$|r&I(oq_T3KI%Nqvkq(MQ{H08b8pX#CP8;*xXF7sHY-N7}#G zZYtj@{e0SBd4}z={Iq5{5@{QaT<{(S+su8rmGSwCJU^jy*>!H&hx8@ea5dg2NlpT-(c z!q?WmTtA0ztX;FcaHGXntXfL z&NoT%&Ll$b9e@LHXLlMG(6!%gom%k6^K?6}u1GJ7kFus&ERy<`V3u`*wqaxG7Pfs$ z>b26T^qx!XSj7DeXaZ$o{1^VTBG_Plr|G~95Kef)aGU1bxuhwtJpxZo+KHs@crtbj zFHYc|flQ2IHAg~FENw@ZlYJ!kw0;!GBF@$5r}oqi>Mi7`C^$mud&s}Y9Aue;l3V!? zEg9eg{vL${BfN&(B0!?pzoKP!-H-fK4+nWNANf5)d{_Bp@8YXxxDzGHO9ObT0iJ4l zWKr2(5sD`lA6|-r&A*Mj$AneW;iIY8h4-4m6Ba&UGe0ssMnp_TvJ6ZNny|5V-u5#-{o!n>6BL*&jpRtu|taJF#0 zJH_S~?q0JWci-p5pY*xbGF%*w6i=S?$*}Fd?8Tm;QUig_#(RIve_VV0#PN^V$ARbT z_OZ*uWUgQJPd~(2e1_TnA#|T08-1vYD?k)~d=oyd%Ub0|}bx$);PYtYo`D1-j`E>aAWK@}HBy^9hr(KGw zCfLfj%Y*&K?whg}du2|&Zjbtt#IGK(WRd&Ox$k>ihCjE4ty92Tmt3-T$C>)%xWZF% z3-N;X(D{)01>-*1aMCYW_{*9}=R>y@a{4594rwK)s@1NzrYUN5CM#w#Vh8mZf>n#g zrHFhE{aE!*sB9W~H7V3Wy83=-XtQRbFy7lAB!1(rvAN6x?xPIdf&I8Xg@8Ujb4l;K zONtWSl$0;U-n4$OWPqc~8yQL?|2ZSQ&8kPZlRWvx{{(2gMRt;pMY6PncW81&sP>7T zT&sUE+n4yq56KhOG`~pQopbr}h#D>a@NSG?=KxlBv*nS|-z%p_3`$+tB$Unks>>7U zpryy_wXa?-tUvR_rzAeB@Kna@OzgXTk_vAeO%!98olO*Tx4rrG`TP0~GwCov>poTb ze(s~kC+gQ!p8n{wR84xZy;fWMnp0wf<|n9%2R@v8cTtYDkiFizW2VZYGcxteU(6@! zYqXqjJ(hKHoxngH-`wXj6OD}D>gkaT+1Tm@mVYm_1&!`pmCDc&`!=A~E8!s>5z1TJ z3}0WxcJtf#X!eN5Y&49>PHJjdF5-@kghVq(&1Isgqm43djkEPN*tbuKL;7=)Tirs} z?rx0`LL|R4I0U1#!z_<}ZG2~SwoU83=HYWkCqUl!jvm*KkqVx2lVCgPXe(B>eShck z>Yz#5?rj_P9LM z@4Er^g3jIL_NHf^I3rp^Q}m5x{Y9uzq|uv*;QeEx`*r&g!)Zro-^ZO~AU~&XXT4fJ z&sBC%aoX@?(Gk%Vn`@VFJG#I27EO>!hd%;f7m~)`y9kTEIgq$h+ltV zv-^o-3(q_&T6F?qf{(6wat(CEvCZ@9#k1e12)_V}XJ{lbK2&J|#IETAz$QR#_t zOUFBSOwtm?y33L50jXe~*Mpl`F0vfr&FJ3%UR24`}z2%qMp z-OqQm>Xc&QXtbM}qnHe(>!@`$(R*K3B5tKew#O>->V0|ms=4vuFQf3|ON)Z~Pz$uP zKjo6bCs_fl-f&4&?Abk3uW$4pYud^vo?S9NvDd##uFF2EcKvT( zktH%%Bcn!eNfn`dHlCrCQZ9L2Ssvh2LR%_g&%s%4SzxDTN;2 zAN(>@I=^KlP%wG-y4=Z>bNT@SS#t)56n;&a`)^|@@4DTOx!qx3AiKXSmZO8iT$d~wPHXkay`oA|rL`b7>~ zuMPuRIpRFd#B5IUfqnJyhqb})FWFU@6seRyh~_3pGeXVfL_4ZA6IN@88Bea5CnOZL znVd~qzY#7?%ztDS^q5Igk>}5TU*rA$i8JVl<7b?Gd+D*7o4z+CzgT`~xy-?vxhw3P z)@Lg$cOSGxNYcY`q>Np*g^Qf6@jjw+vO74R=aLeaTL|A>fQ@w%^FG~;nY)}g_xtAy zI!NP1tKGCCRpnXwx*C^}#B_AzCjp8Q_Y3^1>Gw?&&;3a?KB^08Oao?9%SYd$*t5>t z$QS50ZF~VY?UAvKi*2loBay!>(Trhv=_cA~HAnwco89Htz~5&vf6fxv;9Ze!3iv1n zlOdVkDTvxUrEExs@fiK;6dh*WJ}sN-&?+a~B-*3ZD4S6z9N*5xp?>l=hlD$JQLi0p zmVQHRQ5>?^quWz=Bv{RjTeYY|FZ@F0&DKRZiM_tl(pp*t9dLe;Y&(2recq)6SMpSd z^k*hm=YBTA{nFxF0!pV#j!H* zZSVTJLjCm>PLWVLNR7TG>_|8ygzi6UF~-SHE9MpfaK@+Pa*}s{PttW@H2GK#1ck&c zrPN`;N!{SreZ{R1*3?*bp5MOm_jI)e1C+^^6}o?aCdE+OCeLIuAlC`=ar5Uzcx*{@ zUxL&E^4E)Z=f5>7B+QDEs_mUY%l_$cN&t#y%X25w=GtcogmaHpe0usa_i|4+)U5lk z##SoCd|`}krZc6P`N)tdIZ4#kZF7r)XyxQ;1=tU7a>E(E(bWjS)w433{I(fp7;z{>V z)4kX-^=7!R4qhd7pp((LR}*xpzCo z{7NTZ!o}7Fgbd95RIHhO2I#0(COHJh;x8_WPY3HyJ?~lBVY%2p+N32OBk~x^cfEn5thE2c9Ei6VSYUxF|7<#sFR_lMemX8~xQmDF}#W z9#>vSg2;5j`$!yP6oXBFh(8eRje~?x$(4sM)m~S%C4uJZc<8v8@T3?91@^}Qs`)N% z&e|&7T^AM#U3tUPM}<$ZI70c!g&27uPjW&mS=<65Ii(~c8aswYOk!a;|Ym6iijuO??om>0%O}-J^1gli>X;h&ig10}*tU0UF*_>fldw*;Tqk0X=X`z0A-l zCGw&O2=2~|FGP4R4wlD&-o$1ud1iXxDLxp;`c&pOC}M<}z2m9+qmR66aiqOI>joo> z%@(C_oq>VEC@?)RZ$ITGlY$n@cEo14=fGZ3LVC)x14UsHq2MLQ3ta9wO8uFe(5yBr zPaZ2L9G&~CioCb0s*{onz>u7)*(XydS)6&Em~`YG*enHbpg1)fM>5DQ?&5*9{4h$@Z2TI3t{jKA}m@j zUqQLZaW8X%z?*a?OEOp0#sVrn1qlN5fA(eNu4LV~Q(z!=$4E?#X%1~a18vy`rxGvS z@6&u;kyr8|HM}1gxtA@wR|p&}b|B`&c@-y49+1*QJUCcYw7IH;@+4!YmmS?igw5;y5tRxm7@;s6waA7In zE3|*DptCK?v+k4_rj&=$IV1^yv^Stb=hCNuv+(I`_OjY`OU2;oonhsiQLl%8@8q{x zz&HmUriMbL#UBwb1IiSzH3L%H@W|qFIc4urR*tH*WmUoehcc6D1ww$s z0|gSFvMNw z#x~r-Hyk84oOs)ijkY(EP!IhNRw2&2@hlC*e$jX^3HE0-9=@M(LsPgv6i%m0dB=f$ zQ?KSoKpy?7d#Q^&L=UlZA~o|Lam7F~Xdq<@CxQWBBf!B7$emTN))~k>lJqQn=v-vz%i|&b*DZ3U@atpz! z{&0zEsVbN$ULh;qV9gdQPuEL$_6Jp`|LS?fD?30v;5jIf+2)0AaHDA+JJtHYy)~(% zl}X~5W(Lk`X0^x4O4cilr8y32L1o^;Wt}*7U*x9}Ai+#kUp=t6JopL&as$&~O$AB3 zgLnW~n-$oK6$&7sMk(-2ETnk`O#CbrfCh^`?RY2(e(3&MWJmTJ!?{OOcwZuDRBT%9{IdljT_CL)h#e8^NC&+;nlEe6c?fPCiHCYksj1(ExL_Kz zBESKim&e_@Tu5Cfu1FfRwq~eig`>Nlo6t|UDmy24FVg9~O!~PE1px!_RpLL42`zQsUPNlmLO;51ezI~Gkn1BWgyMeCZK~4{Pe4f7E=M7@(_xfSzQYPIY z=UT!Bd3aiTE%)hs@IC=JIR5Ts2LPPf+84;^%YqwaV)|`?x|dHNMNj(!V*ASsjE|?| zd$MNv@1;w5iNX>wAM(^c2!5Zr`}G4Eqf~Jcc|h-VNlO-fF#UwxHad1ftxgKH zr-*;$-vyb?@O%tR-uY841lB}^Z(vcq z4aju@Z!?3Zk_fA$z}~wo-l4*%)Wu>J?Bw4?!_7r1k>?rV-(f``3-B?|F8XO zLGlM5J`EkKOI zAw>JCripG9-uz^^*{X<^p?~Eh`|uGtntHr9|8BDUI6RE`z*hXV6P%iONg0|vz(UQj zP|?N#myWZ<3lUodoi zXj|dhwqnfo|6uDc!=mioxNUfdDPm?wX@+j4Bm`mT5D`pD7z8Cm1*8!ea%hHb970+g zL>gh}PL-BaI))HQ=lNgv`{mu9`}uyZPv=_OTEFc$j(rz>e<S*Ok0WZ1;rKuY3_@KGsj|=5LxBIlopckS%Q)B;!6D2G_Ta@IsN2?u>rPTK z!mR~}F@A(QK*XH{Pz^$`AnA`xxSroY1SBB-q3Mk9v~zgc9{^wu4t9!XYCeFy<#iOX z{p-EY=(i7gw|@)=JfSf?#eNcq2|}Ej&~*dg|9K4n@a{3{O9cHX0JgsL^uF$2@bcN) zx*cJLbH9*tmCxs34f-E|3xfKEF_KPr2iE-l*VoarT=?Z9zDrt-OBM37%kq2Y5Qd9k z@;M28vAN`FbM4r4?V>H=>S6vBP5afYl`CM#B}nFiU_*EM=Hf2~4#>ZYRPx@t)Sa7hto;u z&J>(aGhBr8F}3)(NM{^3Hjm5vP_!>a;vjsE`hl=&;O68{>h!nk?Ri?EpxYMTSMTEt z`U72}$9Y>lHgk|pAzhFr&n=e_mG%Ah$uQb`G*>j=OQvghhgl;sSXVUaHcpi*%8g~! z-ue7F+;#r^aw$4gIbP?IMAB3DVlI3XC9_@?yqjcR6P}Rc67i1Qc>ZIk@yMhjUng0< zKip`m<7)bBfMr)>#0lPayZ6bQ*BTDp7IC$6Q9pqz>HlnZ?Z$$veS3(EmN4$K$52DX zc5&p-(vyfnxxoDQZx>pg4-LuL>3-cf$r`zhTZ&1j-`TM1Qv*HP#`5@eQNfAc3yqF9 zD7|N6#M?ic_&eg z+Eh)^NP8ld`O)Qu&YojRqPFO(AyE;D3kq;fArg7m^qyMF#XzRlJ}FP2h&TAbntbA^CBho-=&icEPSO!K_-GcR^T*>l2~kYQ%Dtc0f0?g=1NFns=)&664a8WuIhx%NL(y$YZAT zEK0qsLNp!f^O!qY# z9?=_YIi5Yrj`>(M_H^(I^R@O+=?vD=oL}5q9o{^+EbF~@7q1CZAHa)-Kltu*Q%>1nZ=Xz4;lG~H zyzUJMV|Z+qA5Md0%l{c-%bUuBlZkdvrXD;f_Fez#OU!m3aB}e5{?yBJ@uqIQ^|7nJOkv^XYRxz1kpQvFVkdvmQ4I4 zqLYHdgMU+e)xERF{vfL8`-j&)n+f*wtK3BLe4g~;6QgJRhARD)ddge2gQ(A|oO7|l z@`dbd*8c?9>n`;u3eODhM!e0jl8aX!FT3MebriJgDx&(k?2g~zLEdlG5;?u;dqIzj za`zwhq28FA_$SG|hla09vCoc0sXoM0=ZFLl#f)P@z=fMbLVCd~a$f1_w?P z0nYLt#6LjptZq(3MSGNd-=5QXX)zGD^00(a>bFD?%~Weysj+C7{@!>3WP7}bLRXq|m-CO$n$(^oe9U*-ev}zK*6`V_ z>R)9A*_wkR{Fi5mJ-5AX)WDCbe*%PEr6n>NdI9g;f>(EHZ*p)-Fwg-!I2dKnw*@P5 zy#Xm@%;WEG?A<4F}+(k5t)+FN!G8ndh+hsQJt2FN}@^p zTdN|cvV(_i{$;meK8PPYzc66?>%RWv30wHwDX-GxLHrL+8|3~AYRxKe;@j;6fdr<& zQ>y|}PDRteZQ?^Xnj`Z^%L|2Tbo{lUZvkutDy*eMPiiD)2Xk$mam;u$i1jFx21?=o zrIS;B$hu%KUkx{8=a>1V_tuUrQ+6d4uUG#onM}v*kJMRe(F-yBjlM!ao-iP7Bh7Pr zInFz)0(@C()g`qxxYLgcE!(k?G(4^>{;IuG>BU18&FnAD@1;%teJaJN_ewupDN%NP zeuZ^EP0LK6bh;;=7m)N-`_#Q><`m=gWxuD}`Ofxi`2YT71K$?^RcBOMnyvzY0_HRq zH{V%z(m0y5=P6+swcn%HvM%+CteLy2`8y)3^{Zwg<48mP>q>ddo~UkfqX2KLTkY$p zkFM{T4=G@O*EImB_zL|{cFNue1@XXxT6@(YE_E8TYom2!i59m@(%+M+VEAj-LS(sD zBcmc7FJWH;P`uxK&@_mmJ>x@Euhi}|%)knq6tD797_21uWy)3L} zT}kS{rGW9Jud@l|a(#I7yW;he*uEB#$3YG1NE4^zpiZtH+gsn1Q=T^QMsCrz`}h4B z`Kx%9%N@Lbwq#@N6Wb*shkPRnifp;PqmXjw=`HnIdjH<=MJG=VST06=@gO`lFqh4- zZ2O}5{O`NVZo*>O4)oqd@&@5wTEn7ExbNb{T(9+MCp>EZ4)on%65E-uX)Avw?^UBZ zFaLL&mE(jr0kLSCs9A5dzj7;A3?XgRD`d!-hW)>)f}qRp(W{f6qliBOkyrA0ZCYFK z_ZJmvWcnHPpQMjVv~6d0p8zee0ohhck2Wf+Pmydy@TEHVmoy+!6gs3%V|9ZzR)kI- z#im6hE)fyb$#m51*ElucE<~ml!L25g@Lc_ACW>{Qc+KS_=Km^Un|O#d61yEcj;&yPH+_ zBF7C?L05)6FqVWc{Ay9l6Q3ew!;|C7vp%O8Gs1O=&f8d|nJYUIY51(%xsxPeGotD{ z2-bgoctYQDO9Rkxz>!(9CA>lDXY1UURj5AN6SOyF9YiDBi`EMTlJKZEVeMGmz}R+N@IErq zDtyUYh$rI8@f67$q>6~peciwp!S*5}7!mlBu)g3gD(=mv*Aia*&`^lMKH)aoaVNc^ zj2fvYj`=I!YAns4qqHb>FXapedNs8Def{%7!DM51yIx*y<1$k=GvThJbPq<5a16EU z>cdP2g){qQ=V?--B z8ndJ-=gQ|^l|43=sA_qbI_F&S_nbq5P%~*EN&GdDo%l6&t+}ecvo-r+3*Gjbv<~7W zr+9s$F6a73@(a~p#Pyz(>qT%YtqRn0r1o!=oJ!bY8WZDlcf)Abp`CW5IwCGhdQcLi zaZ@qOMme0P?x+WT0OHDrr;sr6LrU%tUU#JY&XmpeL6&AKgc_hYbtzzJu=i#6WX`pB zcclN%{ausXFh2WQ`9VkWaP1Ttd)rI%xw@;@?8_LzzbkIvav5$PPkXnk4%y%C=kfj3 z-rJ|dy#PnYVqt%CxiqdT|DHusLf7CfFxJ`k*+S<#Z?{?T0fIxx^zzkR?$N?=sjI_O zjvu}Vs`ZI{x;z}Lzk5z>7(I$p?n5#eNw>q>?mwLO-oK~Ds3yW>0ms=r1xy49#L4&3 zealciybQh68&cW&`jDz!{-4V4_cv~Vm1#JZN+dr5gVKhe0HB^O%E`VSH^`5KA&d?coY130 z&0m|50+IZsaxa<8gQs_aP;>d34=L>kGrfXPvk#5YGM?3|U!Vb8ePa>wxakb0ccDB~ z(NXd|s90Hk0S?{c8Qp>QFGuf|cn5JkqOyrmJ!Tm8Y3UUzL~x|$sT(|yt_TSh>r+=+ zp}=-gOGaLc%Os4ey@_2Fl3k9xHSs&%ml?uK1&vbom|zG8d=S+#7%=r>WX$jZNJcIv z=nni!q_*ghrDlY-aIM^Icn9F|>K}YMxlG{*5$0far*M6@%9(`g;#(sQ-Bx0Hn;BCq zUy_Mczig*@Oc0jzT262p5{#Ve2XHx@VJr%S(?4xKzY!!hWE)~j|1wgG3vz$R&XDrv zP)?`ZUuiN{_1DpRz^k>JA5juU2m#&&Iup=Ha87FbIYyP4A5N9IxjRxw;Y!yTkh0Nv!;UgS#xa=I3L>sT=?cUC+NgBA z`!ZfzUR0Q}r2GkYH`VzPIMd@#h3%E7`EdDUYn7+P>nnPzpNdmW?msHfEZf$cD^J?; z#0drLm8cEpCtYigN@Jeu+g;oO)h*C`j!Sg0b(omp6sxO@{pn)y6JT}`nf6&$>?51; z;FBNZywuuPo*y4UY3DNVwsgfmB%E))oSUmlD{l@!J*MBbTpD@YP%Cx)wG?o+Mc=E@ z4>_#6fx=!_O0T0U({dX3eHOJU6p`##h$sE(z`VuUh7w zo#(f5HGbkeY{ng|L0TK1JG!THGTl;LqJx)eajR0~N{ds&L6ZzY!O6kD;vWlGARx@oGn{*GVrb5al2>xq@vmgURH% z#ibr#?#Mr5Xc~%)NM2b*Qu+4jEa=B&h_Akk#ix!oXvagVS6Q3e>7E$NrPbslN_l@I z58A}*%6D-=Wsl7GtcaS{<&|r^Oqs*9Zbwv0^8@zsnTW*F&RnJfU~CA5W@NV-WE-#NiC%ww7ECMB)+ zC~@2RB5R5yU#n&EaSKhD{qZ(1tkP*1Cx_>ovRd8?<-Odt23t2uWta4D_bwwaMWczxPxvf*aacH$Bon!LQ9%J7p$=t1=f~V)#ANLS2){}`Xv>=7=-G(> zzau>S>xSOF^NH+#$M6Z>t*6uRp{0R;I6E@SsVOO@Ie)R5W)LYr=um2HK6(#A2=nO} z(08Cr>Ou9>-B78d^RYfbg&r^zNHj>dA-R??WC|KrDym2nwRB_% zzt3H;gIC-|#`)A=+QV?_$Tk4KXd49wI#TMCOy`dVx|_T_dmweBhj=9opdwPVC<1`e z02l%|uLA}&fcMw^Kt32MH!6Ni5M3!oWej2}P1C}Pp>jk33WKSt0WfI*2#*1p{5QJL z;Q)L?e8{GBU0hVi6baylgaVg>rtL%Lupxc?Y8=xbOFTq*3_vpmkShh+Eb-ZpfG)&m zx@?ncE*aN-L;0-T#7GO|Chury* z>d;@HwKQ#q2(ShXeluoF>4sH~6nzQtEnN?qBHP~$AOHiFTs9&i%}XQS^}Vx1ffG)_ zKB(XTq!An)<)nW-YJ%~$e2_H@XoeZoqKobLWgf*Q9g`8gIA!iCEBR^K_%kGk+BODC z5u5EmF@b{b5W(+Ez-}f2r4i0gq!Aiki1sgcTFo%tNABr*vHQWX1AVc>r?Hd_aUhvE z>V{BguvUv3#KBC!+!eSh>PBKqI23iGtbVjMjo3Ja7xCTO*@!_ z$~hH@Pq)Iry#_h&aLOac&{6VcvLvG)2aKfC*N>cb^!pEM43In}ffF_GBL*W-h>TZ+&1jg9}=q3(! zfTw>sTl}cW*VVc3ExEkN&P0IyMTAvW{;3JwcT&j@L~-{L_2m-$6vu_lD$S}db&vj# z3oPpvx`)15_V9a25DokX0ot@gy@G_H7|Toa$`uVuN}NmTK9pM#%P9{^(zZ&12Fgy5 zbQWKVUvK#~&6c7f%G)@=70v}u&jAGSVv=haiCA`u(w@bXHlEYRI+rJGRg?oOJI(>w zM5yg~ImwlJ3kio~(r1y^iqk0;UFo^AAZy|u%ezg{o&wjz(=Uj$nKYj+QKkRTpH40O zDA2HZ99{U>do;#*Ji01dh+u^xw4!uCDKKLUwec9WE(TU-%D6ABHN55f(v7}8tk6zpi2A6r7(iRz&mbX0tFF9(W6Bm5AE zI=c$s^KG!zDmZWq9HP|(F_5PT~xrUbQX5yV1t`DI9273m-vOEeOoaCe% zo`WAFZ!XCPvEZ8rsX_L*hg%*kB+C|-ALiyb;A@GdGN0DP^wzlM*4>Ndea<$LN}HWY z+uy;oS7-2j68(s_TMG!(mL?Fd(9S$>UiyOAj%nh^Y^I$jJ|YnrTtLjK9Z!EhvcR-+ zsNxzewQv9$O|VBEN>647n!E(fLqZi|I~8*}6;z>kOlJi`=VN-Rgb`d&EJ?7D*080^ zXuivMzAI@4rj?UAZp!-GpB>kkVqwLJv%7VKhC_@RwW(oGt$NU3x@%4AN1Iblmf-*Y z9%0!3qJ9&g;QxyH5Dbxuj)C8Yguq~QVgMZ0^L+=;dSWT5s4)QBE@A?Jwfvzz%08S% z;#xE_drk3hHmK3fpJbHv*@f!;?HGmQncDv#Sa+RwXTN~IrD$+gkMJtHPqhZv^}-&Ena%;nBV1}1Ad zDa$&yRkO&Lzr<=gr?EYr9X*VdY|xTsOq#@0&gQq$2idf_Z1c@;v+SmJh3TzqT&6p= ztSjTp%@)4~(WXbx&(8-v=Pk`pO^|r)GGWiza4#?zUhPzQUCIMmu;;L|xh+MXhBm=Mu`**f5%8CM?G=`kc*W*B=X06IVFb9qGbEIqK0I!T-ZLDUZjn%Y-EOgyweMQ zN6kjORA&oQZi*`<`mTJJb_XQfd&8|a&l{oH7Zg7 z_2MO8)C{RaBKfggE`;%(lHAYvx?IkY&dz0}pDD0hPpmYSC0@NCgG6#>kW6v8NryA* z@54_FH~8Kn%S8+0HM1mmBtfL%51FrLH%su*4{W$b4@QAsj0^c8KH>~2=!vJQtcjO}x0oPne?XSL9JET=m zuddZgfe?GLt7-n%n_g9Jzih^co{zNr33+?tGrf58n>I2?l|X#)W#@44#uUQ_w|Ux% zv;rdCA**Yppn7u@HJIN!rhVOl_s_cIjy=^*IaEube z5*7X%_%pqf7F#(LkF>5k62^@$@5DlO*Lve600Hu5= zXla*`SE?DR8Rp~9tt+&+FwX_&@nhG;?el46w0pGr2oa7ho%C-TI)mqpVNJ6SIGBl4 z40ISOuo;nE*A^@Vp++BMzdta-wuoDgN1M;Da5edK+?cd?Sj5EhZlUDavfLtGDHXBM zPc{adI3WK;+i`-%s9?y0c&`8KuBJc)~v?Fza+CjsS z-BdiSCXcmSROMzCBL4pUJoa0{Iy#036^J06ccd2AJ|={kdFg#g5s|$J9sW@$te9Ky zm$qBypw(T3O#dp@Ef})Z62C zSQUz}&(~0~=YS${uY2ESHJJ<0k@958XHmol94~IFJaC(3PPe+du0s}|4aZF;Y3fK3 z5fX@6HO%`@UqiW8q_L`8E6{kH7LOoXQ!vi9rcO}ARekQqr@b!?RLXm*_xvZnugFz5 zam+3ip`t{Vuy8McwRhXxT$a~K@h_Gt<8c*>0UgFXb*WeGbIn#iXu@i%xM!CBp_Xaj zFdypW8ybXxnXmErpMG<{c7$5tyz6k4VLHO0{ls~&(zPbHE-Kzb5?J4if{|tNO+>qd z>8S9x0^NxoU#O&!lW2iv=$*r*&%M@n0|#vM<$$tT#abO3M-=DZL)$Kez-k*B@868A z^BgW2(fKRZ0HLTK(cTb6ge*C1P~zIeAr5RRxph}i0U`*MyDlV=*)Q(p&WW!Ci~SAd ze$*4ePo4Bz%Kwq0A3n5bQ+ic)|JpFMY0c8?0PzojPBZVTt@(ZSsS}EHPqUDfd z14!@_{%Dp4AGR|1Gv|jDSshh@f6|}pZ@YhMpVSJWmdHqqdvlz{4xgqXgg-@4G5(qc z8F;kypH!RN>(06WX)Z!DVDr#x7Yo7>dY4olZT>&A7q0c)YduW2xttNOV<%F4MIj9m0yBP}Pmi{q z)RPl}OE3_MMeFnqnXp692){zPc>|DAwh`=16>6y%eDMaDdGJ+>D!4w;8W74n*9Q~nU z5PS>)A7hjYW<(wU zs*aeKW;H#HxTh3;qKG}ISFoS}EF$1!1c$BLHBz16EJ1TZ?*0*$VxC0Tn9u1_%5(>U zbtQ{pJr6D6cK!&c<^%}S!Dqx%~RU<4SGSaCba(*l_UNn3M5%sn~LEqhH zcMQgO2()sCz)^hlrF=~dG0comED@PN0%W0RkJqIp93lsZ^dA}`OEaR~8lo#p1(lDY zrwL(3F+*4w56xks)qE$%Qf> zEED_&O#r=Ix?tmvD3cT)E07?494!CUy^ClZ#h1bJiv96ywN zg+xR$E4VFzh~J)*4q?9`ersc|H4Ip_%vp7!QuNDGN*xhFhbi}bEdTlsjf(#>aOckJ20>hem(CfqNfey^Qin*E7;OM$c3Vsd6+R>277AF>jmw|pKU zuB7Sv38@cJK(H1_>PUl-sr1?;0&JR@-k+G28Ia2!^WGd0Z351B0u@FneK0^=Az@`_ zkcyeCPwd$Q1C9TL!qD`L%^y@=s)Ouy3Mrs^1bE~QMa zU2tZmQ|^o4+^+_RdG@RYWzdosVr7Q+my`6GZ{Fv!ya{MZnn7*@^gS8*e%1g=#gJ1S zm9!j{bJ?HuJL-izGdLOmw!vJdG(@~Iz}L^jL$dJA%J1PsJlq8Ah{%FRJN1a=4;bYC zbINB6&9^22xMuUum0>&R6oG*N8Vibb0-Z=!fw;xJb!h%MwD8JY`u+c}0bUxQfCHEk zwC^tgoe7|%6R}&faMi6s^`=ar0hp#?Q3?^M=WNV-3?SCB8;KVh&w@+=R7}zoK$8z( zG$ign)P2HJjx=tX;@erQc80~d{VWtD7>TS4H9Z745(@fLimr3yuZoobl4ZC^4}oJ; zAS5L}fy&es8evk-R#2#7QR?Vc+GJiFGh3R#QRcFhV|Wg@U@!c5CicS~07O!XJqOUC zw4G4k2uw_=Vd=Ah@`KItSigd5@v=tkGFQK{##@Ho+8=;70WgC02aC!&pGsTj(t)PR zr~%lgtx8J1ie$rz)TW~E(bBJ6srYdv#bMeiXc|*#+T!z%Hn(7U%2mGxGOZ{8u}yUG zvm$|jk6B$$wwys|Lc&XuSS}K=0|5LKfl+GYZLHj&T>?;{On{F7%s5au5*UF3d%Hrc zZzteG98FX%S<0xF zq0AD9_woo&>Hi(kr=L^*5655H+?oV*l?H!XaoozRE$OSx+N?zn()CK$VNno==DL*Z zx}~C;P4NemeD#`l>Zvt=-rJ5^0+!kjEhD69t?}Z|0pP;V&#@OUYC>4_cHjy6zfN#C zjt-85hLQl$n7Y)t8m?<)o9s*pI56dM2DOa*V?4wf2`pR!R@npn(ZIJRAYaT^!=oku zuAycypqA|4+}Paw^`f~wyABr(>TYI?mj=Tz!0Q%@&q=^&Jn%*;*kAg^b2Km<{Z@@R zcYXQmVnx7mOf%dS+TzbkY_5Z&8tG%1qVetz@hwz|_Rn!`&q?6iXSduQT%VU9&exX@ zaI{jlH@7wuo#%*j^EEda=$j-EBuj*WF!(IH%>|{;EeYansimq6_+tVGpy6NEAyLxu za7>4oB*Fnv=32pMxB`PC=%08Y)VFJDdJ@cr5=uFUs`H)Ge>=rAME+?g=yA1(w{&eb zGv2FYeB|;vxwyO3BINUS_w;%#HK3ca{nO@P*Sq;%t`f#8yr8`!)4Bu#9#amG?)4zM zuzJyTn{PnikIkQq^`^vvf?5#4R@@RAU@EuPC6B&aS$$7i`W)TjsTg8|5E|#T1C=eU zG}YlZHGof$gI`n!Ya|n&kO2M@`2{U~Q%?p$RTHC0&nm5ks&WtnE~eU+p?V|WQ&(WK zB>wBpU?0N}IzKO!2dXG2NB8W;L?mGLnfzl#l?$V>DaJ!>vZw zXej7T-i@B-j1mcqD^@&xQ~59QC1)^P#}2SZX5qye=`^{lRty~G5?>Y(bPrZo4ot>b zcf%6H{IBNuSFIu_uIQH~yH03lK+3m{awFGL9Ohy6bNE5uWAN0bmh8QD+E`kO={~P3Ujd zzR(!8JY}YIqg#$*-2t%6P%z(=5>mhHJkbO6e(X2VIQhHwvf6a9$r;AVKd*4#m+qn)p29uZ=*g}Nn*ki#!wXK!3vz6W8#fo#F9BPwiyLtqmtV&& zt`-5;Ehw$wa1)l(ryPeFOCF7j6w)yIy`_d%5nD^}EzClGI}2;z@{C9XC(P_$wi1pvDNgUS5+s><^iZVWgbgKZNyrgyvb zMjjYgvm7Eg-avnwE*R5X0e+9c4>3!^Bjt{X%dT!o`I*0^@1|QV!2bZ&<~~;1sllGN zBHE+YWb(%b+#Ln0R;`WJBp7l+oY!fB*B9~Y;Z1#kcyA*iJ|(erJ@Wc`(8xw+)n*TLGyMAIyM@lwDuk!;7DQtsLTz)gc?;sU`RCUb zO@2Ub;D(3MHrRH1Eo>W_zg@Do`4zU6wg+xJ21pC+Ks0u$^R`mrcRZ>LhFUX50uf{J zTkqA@p%gn$!ggn@5p%D10*rTSM|Rq+D@LsMS_1bL@(}eSdmXL27PcHQR>*@q1fX04 z`)y;UYWwy8?C(GNv-mw)3jFI^AV9+QXBfpF66`mMtzkLRbvSb}6Z5Z$u~D_3hYa9` zAFeJ;P5gr+aDlHixu($vUQq{)Wd}0j2aaUY0pI&U0h_H^>{MktU0C~};pSmh6e}x! zS#gyA^L2#E;*qP*L3PH#mGrUBo8z**TKE#&&X&!P95Fp}Yz!C4bA^(2c65yYJRU72 zG9;`PW^e!{bs}x{B#LEe+4*Of)DS* zsbq@hG%jmaYHwoTr>KiVBh`}@a@e0Hi^Ch80gu%lsJuhfqI%ZNR|_r5OzY~A%mF7%><>3g3UoB`v@K+R z8~RJxyol-FTYp%tLN+MUd$Ta<%Mus(Q=e(a>@(++%NmjEbMB> zrSjOty47xUc6%-EVgNHWfY}%2H0hYm*5>b6maSFW{Mcf%hP&u z`kEqC^EWlyIBdwC-wN7!yYor`vx z)ywhOIso4&>`A`S4&HdX*P6>i>SYryFDuS)#xDtSJ75@hD=wH*5;P9D-XVCMeS$@F zSaO#{xZe`mh4XxxZXq-fsZ8T)jJ)sgY%HGUGCaOJ6e2zBcO(XSIK2m}bGdzw8Sxvk^($A@;vy`*(XT@7GIre}VEpQ3`g*Ry~qm z$@Jmz!E!4ks_qVFucW*s7Z1&rjT+n%n(IxyT|fRU)8~@ndydBKcd+r_%zq-rLEY~# zIqo~{y-~uCZ+Ec2>pSS>eLMTCgxgr<&vA2Ru~=D-9z2eP_uVO9nwi^9@Q?TH&T({8 zR_ocD>{)DAwZ(RwI1T6?WFSTXQ2Aoe}&u*sX)M{>~{( zYHA)l_Q?BZr)?siJsYg@-73Qth0)fE+XomF``k!rvyFY8uJV}8I8!}N>1Wd2k;sA~)dh*x z_ZJ-ID_P329AXKLt4`LN*ip}{mBOOBrk4u^!xk?S;!kE@4*ttO+pKQ9KDSl#t?DGW zT!p`R_roGA+cAEt&hkY|c)aCi%<10Ta@)m^X3|ee2!Owb7pr=RLt;>FgW=C4-VQ6m zo#f6aDgWOk2g3hVa!}F)f*=CA)DS@a|G|b^2oe)9sCWPXZKnUZsQ*9jhE=ACm16qO zO7lA<*SUv1R=$5d?x`ucmv`fb&>ck&M|sE=>2_q#v4u*SNugn`;Ugrq+pnqv$@RNA zs-LWXe@qtnlrvM~I2`@7_wmnL?=~mPOdRfUDLDC5Z2cVcz7gj6@9^qk>J#rw=i$~5 ziJ^vSyY7>nTA|VpxUNV|}FUYU?;`C_ojS<;KHb|halTOa}7ay2+OQ>|oV5bp#S7D}4 zZ}Hv9rUPHl(HR(hn~GHw%jF{}x)KR2Lyz7%(Z2RGV3Dx)%=UVZs6fP#V@9z2kfRou z+aX6|B|!ibSrENQP_k2(6`tX+ySnnWi&r^;{7Qkw6DMyi+S{=PhI;mAIirhvX3;y^ zCCgOaPKDiB7ABi7+#Y{U{b2rd@@!OPzpr>b&u_(`IFp>;&z=3M)v46M?OEA#WiPKN zk@vW~)phmffp1@Ah4GY&*+h->I~IDcv2W%jKlT%|cC-rC&CfcQ0crS^lC@#a|C)-~ zN9)v4IhXpJY;AwCPb;$pEdB}yaeU(W)$@7i`nEZj+jqX*FK(xNIJRfFha(y_#%a8` z`l?ci(zY~O$qyaTtMKXxh=+RZ)`R9Zq7X{q!n47 zmQT4I^+;sJ?T_FTT!s9laK8K6ev^dVw?A?(0XwS5yWFpj9eeskODaFJ54*Pm+3mGi zz?@D+GQkv7%u7T@OSTozHQTU*Wh{8cj*h9@rI7YUjw`>B6(RQd&rUL-#LAY%F1ES{ zuh0|Udy0{i?HB&!a?&sLR^}1c>!NWKC6pOoL*+EK+{>f^m+I9xH6cHsyGbIhddu8Z zDm1=7j?-CjTlI3Po*Z!dSvrN*GJz|=TMl$*96seJ481^gD}a^Y=%#tGB3mqXkXf}I zClq*HMYWswjaq%vs3&ek0Cu8Ln4opNG1t0A3#OcA_jZ^Q?z-SR@IIO0)}666mnQn+ zw__-dfa3!_wokpV(;bo$!w6N=k@25>FVLqyB+n)V^TWY6fjviga@(>S1-)*5@JUY8 z*3Q!Hsc{M~2*vzq50-j>ch(LAjxlwgGE#7-+osD zgk4a~%=thGHbgLZxl6c?BH`7vE2l21mnw;Yy<0Lqgcxe|i!7_p);qNC{^tVuuRx?m zi3?33{DlX|Yzw?j$fq|GoYm;4j^daci%>AtQ}Z)R;C(#irYT9!pg65@5sdKTx3c7# zaK(rYyrQ+lXtAL|Qc{dXvYRL@6K;%J@E1WgnLWn*T?2kKW&45lP8Nf5q&9qEKhBYf zk8HLAQeb;VOYexHVM7IrR%*~+2vD)E5*7dVb?T=t(B}pDeeTltHR3N>1O*IQ4C+jy zMq?I^AAhctlJ7v=j0xdWiyC04Iz;{>_TI>7Pw^KN;OIw!6?YQTc%LmLwxrnG#h0qy zc_y1YX!u+D(Pv4$oqPn=u}c)90a1EK@xmU^DM!CThPz*`TX1-u8+BEGAA?swrA(jz-j_k;5gAvK2@6giAi6OOxbYRwCejW^jAMPL~Uw1%nB)} zF=ylsipjJQ1W?Qci2WQY(<7(vq1rBL2a$RnkDc=^?n+Z#^^N#19npW$#@^H(cxnA~ z73NKi=Aj(V`H~ji-NA7}7~3l3A}9?RtKy|u0APb56i9O{-@y$-Wj2G4xG|qh7r@mg z0&<`9E%N->?zsIx_>{tH!@0_%w!+%sj$k}F;NzztmQ1w4T14l09U&^N6U(N8=wRICJ|Dg|7hZ2t}_n|SBhuP@iB^pp$Y@-kc2)&82xO#{G+m9f7= z1Vb2XjgQlNUDo94uUWcBhGu6n0RD`aYaHWyX(l)l+*DV;?MSDWdYZ#tX9&r5ko4SP z6VZ=tI(`sdcUM^6-L{x(VrgLOp~^VGxLEmVWG=`DY?!jyr236bE%n??8{THLiJ@h`(?@ToEZ0LU9TJxa(X@m6*t1H$Ru>w-}?5F0o(FBpuumh zd0`uY^?V$FpxP-tW6q!WZMn>3@;a7PIdA z@iosE|DhtCFLxY%boJf*O!oEZ5~|9JQbhuUM1xV1sH3c2D*=uueeGw)2nb<`;#Pvb zPKdEb?vq~v+y+e(w+{gj0@v+vD3J}Q)>?9+XVeP8v6|I=N_YscPu`dVR#kNa4;^_F?Ue+UpdH)3^86m;GP7qhvzG1sx>VMCz$(GNhPvT^+RrUCk{r zbb_G-BHb7v;HLv@8W6ZF;+1+oa4kJ>bDciPm1hgfNP585ilv+U3l1WH{g*T!lfW-^ zfnF%!6W0*YjF2ZwV4o#pfDcWd35M;z(B@6+6M7^a_~irtH$KDy36MwvyheiDq#?8l z;5|=-Y+aZi(hh49OdTAYqez#9hwlFheoS~_9tk9S0)E1LFKJM(dsK#xacw z>nthuo5myjA%IMbrlN>>gwvU41FwHE{^|%zouq9JB<6(U zrIa#+#>*%t-(*j{57i`@rKtumXnGQBQ)nnYAOd`vkpO4gqpFpErHgJhV9WjrWLexuCj)|eXAm=WijZa0(k zyb(vgnSQSocZqNwZy z-z=!FY(!uB-LrI}Q(^}++toM2-8|dVH>J@aZTKuFEhmC*L6ZZD!7V zV@}gd4vm<%QGeDi_MEk|%0I8RL=VMTP?fMop5& z4|2{Q%(60csFP=wK13rC9wb1m8F=hjv1wGXGdcR<6|#to5pHv(i`lT|?ZCdpD|*i2 z-AL9^7|16gopS@Fd;^TB^YvC#PNFkx0#6r>f`nNh#*<2;EfS|YfHNzlI77g+bXjUS zIDG({)l@dNQic~VpGK9txRonxm2gC3s{FuVW~I~`Qnk(%&oC82e@b7S!&}bp5vQ%& z%h8<{CEexV|A(x%42t@D-+(``Y{3>t=h7kQ64Kz(t)M6DXB}_%-F}FyC z`vMs2|4Z$rORB!?RQJ`jHevaJ{7l_5u1|KjPq$4!c^kd+-7b3Z^;4t-Cb%gl)T%yj z4h#yZUycUtkq}=iqu*&qidi#-?i?L&ro;K{A}FRuamw*Z{V)e5M#zBPrv#KMzY>0c8V zTFKl%%tu9)w5=a#r9GgQ0%T+1$rxfs5gbi|X8`R2-0ebVKn@mDNTx4b_HAOrvh;Yj zVi2E~8JbKxWQJDp&`-8P0_OA7^N*}FIM(n zB-4IA>u){lZ{en41o{jVn9w=2H=qCw(1#|`408{XRR@Pv87C1Q!sI?hB8Vl)6HTPS z;y}`-AP#^DATaVL4Q(|K@w*Lfoqb)#@s%Yxy$E6dZ{-ljZZAiBLFzL3oF~@h(^J9`mL@sBK4hxIx zy92U4ZI3rZ&-GG^O+e2md3lsfb#(=1&^Or7Tz%cMG^jGY?hsZHO(e>6l37n0(}Ck{ z!y>(pZMY(l{;wYSMzYPjs(#0^voE7mJbUiD30t@eFg_WbA-fnya$ZB)W{iwqL-GKB z(84%W;skeq#aKUX%uC8IHgl>cXDZeo(dX(XwbnI)MbTlr(4^>AGDCh5B5q+ifoCiT z>cYs0eEWmxC#GbY8R^~3>Y~9W#x!bvizWCLdz?IM#CvuWnpD&h(C-aTBQtD1LUEFi zJ_7Jq4AdJ33&Fwu*XXgWv&U@n?N5Va+2Le4#EX@7C5s`mzwEMcK847YZ#5#|6@}}EXNp@BETKtdf(7HwXj-2ggb-q#C z?W>tF&s2UL{;Y%E-iaGR09fQd0O@WD$re@{c99oN2Vb;=W-l{LQ&7Qfs`IV8L*I5E zq`doK!m@JC1%6+^`DITlAG&c3T8=?bTM?`P^5=c%-Up?O8cDW4{12Y*>@p#LQ32#L zETk&>nla}+V;Cef2yAIi--bg`U0DpwP#8WQY&ko}=zgR6XpCcjy?35Vo8q`?EU5#LXf zU>FDbDCqK$1^!6z!Tv$gkriJ|#=(&z=J?VB814)6qh|+qhzC-o2f@7uq9@0$|Ja^s z{PukQ+xIdH+wkw82dD19N$9hqh+ojV6>HJIj$YNDoa9^DCn6qJAiS`DQh(*eT&ezh zrT+C21pnoaa|k*+@$gR3p<>@*V&Csq=V#=%mf`+j6{dYE78!mOWxjaoMI}Olmj5{} zLkLSsF7FjS2LB~(+-drfqs#~W2|iQ5eAY%nt^)vH{KecMRL++(`aNb~c81)_8^x$Y&jz1&CYd+XmT0h#66ZfXo35;~dvv|5 z#qa3+jZWt6-Mh8W1$7FWh0FY>f2fY>Klwts$&MRJhL3pLg zF*8smb^*nVk>($KB^8_juB{?^Xwvtkd{8wN)FZ+!bBkJV<0Jhtm&%8h z{`2#}+b5Jy(Dk8me{Y+PmZ$fwDqQfv<=$#L<-hH0lJrK|5v!+bCqf?wzkrC=$Ayft zUd-6ZVl>GpsvRqds*XHoIyT9uE4uyiN9ls!D!t-zq?T#<3adT&1C;XEG5*TFzY~Jt zvSbXs#LxPC=ItCB0_T%#v+zXq0b<`}lp+q01-aDQK(*?`ViMGj)9p}yjAJ2L_RJg| z@zP?R;^GRYO~_UcUMyJw)NhFE3uZ4Em4fnTX=w^rHDWqs<-U_Ug7C~q+84}r@-8bd z5vl+-B}YQ4Sf1D>`XYUsLD8RUAU!b^ir}ce5Ov$>s|Z8DC{D&Fwus71P;Kghm?$t>A=ElQYv{ zV;LHe;2$(HGQ8L#XKd2H!))@!1(Qp>QDo9aGQZOzJ3LC*^-#f!H(#+3e!X+Wj1Y67 zPGjxGo77&z9)0;SX@x6eV{(7*wuaC!;a-B^%Kj^8sUORNJr;jbUrtJOm{gm(XC5BS z6UMAb)3CXBlm(wIp_DP%tIwU{#OMfBOJd43?JpFZWh}4rHvJomzsn)wITZzuvt0Q| zpBHNzCi$xNcKnylR6D;+@XWhAb*4EqMdBpQN?e(K4Ap_h`iH#DYqo2TO|8>XA(@pe z?LSOEcrJ2W3Jy5;U5Jg)edXw|?r^;7Z{27cduf9`0%iC?Z8NrX&C%CN{ok4V)r>De z*qy2tnZI1M6HHlkQfExlJEHy0DOpV#$DFJe0Ag_-{d#)`JbmIzPlc5r^%Wk%IL zxyMyE<5t}-@4sK`bN}!Dfb`I_jiLAtbBk#={PN>OnUic^NtR)?s;5>+Ag>O8?SSps zq@GcC}x-OjbjJ&h0E zKQft4!#}AHI>dSCgz;@8PcjWzwgq|iE}d5vwWzrbQ?0aCI8R(GWGiEydar3=Dam$h ztd}|sFV*(X&91*&rCH@8!|?iLdocBaPEB2jujA*K$mDc7akLZ6i4=l(hyl1}G4yunu};lbWjBGTyyU%yeGwKUnf-JuA6}6Kpsckv<|JGPMp1SG6#AEb**t# ziq_54%@BufD$vGpYC1^VD#?w-gC)}F3&e>w2{u9xxhDy9f?KXSl8YMXzu!Q7Tc%Jo z=9_4F16zJS~T=Oi?@YBJOG$*gNK9_oAcZYTH;m0`j@nC42C* z-`Jvd1KS)p&s#4rfr%j2AgxIIy7sunJo3Xie; z-<7if{%U#%qWp&t+E*+*VLan@UFlKTvq3HCz=miY=E6)ilCJ8Hy@`}O**CkGenWI8 zYg!|?(Cf96iIVn(ZuwVLp7mkW>ufgBUzupDJuAz^U1<^=Bw4tpT_w!lDbK!tC7*64 z6|UWIcO*&1?6tob-Q6kWxw>xF4{j43_t-}0MIH;%&ZPIK*0#lcE@ZB>XOv8ZSy_fz zE_1axbX|1Zdtssdsd4x`J zGRFRyO#A%%Us-^pfwCwkf3ROCN>1o5-nIXEA5SA^>}%@D1Qce;^%?6Zp9E#5uHs0UOD(v?rOm9CI=h3YEyXZHCM z6<*KFto}Ks5x5~=pwb`f*gE+rHjumgKIoq&hPWP~S$)~IgT6>s7O>%N+7QD19sh?rMV|#OKX8u_SKnJUIL8#(ET+uXeTXLw+a3-CQPiu=+ z&a`nrKPAsA7XN(&``el?t{a%XwGKf3+T!NSu7w1rrzJ34KDU(f{e#OHDQ3${OolY5d zpdKvcRaaLaY)z`w2>hOI!a^DM$*Cmib~f^x)Frpwz!-`gR=A_;5us zD7+CKC4&d;sF%BkTaY0<{zPZft4AS9dTv*PPl)0M5C?ytnxh4ErndYfm2UJ4dG*U5 z5S~V;l~%QtW#@BQX!vbMysJ+utTq;o&=8M_Wl&W{e#fKFT6OM_#5gqSN}w0C=`{6= ztwW=2+zX=hW0EcDG#$Ye?U6{Ks8E7Z@m~NV8ST75@9ZP@mQmxA5sClswLmHa(AE^W zr%4rdZddft9C#5fZXYSp8R^s4UL+ghztIjoYsVF&ORK{8<+N-#XynX0)N#q3Zt*F^ z7jer{A}NtB{zO&o283lYqfn=&bE=k{rlx79ju4aHJXicid3Ub15wnspce}}VZHc{3 zIn{Pkxnwi#Fg-aP17n?IS%ur%sY zvbt<*b?(J!KfBv)_+3ZYUw0!`*GU_uU8ZYRfUzIYc{;Dd*4Sx0{v`xpb(4cVrqg+% z+G8cz<*cpqa7S0+udb_PhkS~jUqM$veRqI)_w(?^V1!zX4b(RP2(9fo!|MgkcT*qk z=y;!X>G<}B2=(~M>AmReu^-n7vhTSyuJ0&>`naL*mfA}SPk3RhpK`A^l}tvL zNNoZgsjSPtv-|b;?i5emOlN&vMt$Eh=)>u+*@&)ab`dMd7=z+{0zxX zqAFkXy}md1=v1+EFCDc?R6(V+hR^!?mN)g7?DU%iqdY=-p#aefA8Df)zGKuLjW3Q8 z>lS<*7KaVj@&P}rhj0i%18~a0K1M%Af&5yP#_--I=4O3Gx^UZ$s1qYa#6YSKtXq~!4k5G}U)9%qkY}O_3=DgMhC0Jq?7>f?A=eF+s4owqs_}#U zW+t*($x9k(jdIPiV*|=LgY(rU!)L||-%Xb2u64&TtOyw@s0`hwqv05XSiHyXNH6!`BzA2iGrZvqVQZC_+@0LCBIMHZ>4VEL-}-R80|V z-)hEWZF=FQF@^4iEb|S4k02fVDE`I`kq8un@2J?D8$42GNe3fTLYa!2eTWY+($2Ew z=4tY!oY32ip%T5JYHMTa3pW|Wi?tNN+0l_>_%R6{rQWb!`Ip@a@niZv;TJ1G62)UG zAEQ-|#*7DP@xmfvM__R^0L_BE$3yn=$ zQ@{iOlD!PK#f_Wkm?y1vOd3yEgcJRIi2fVR8rxC+8@&+(vV>ACSA67K+-;53DZ4-- z3oN_Q*y7qf3!B)n2OTv&l@KeCs9SVYPS|wxyDN^Bky)HZ(zd-*j<`XdU5lW1?8J#- z((_of4>7_@QTe7KU^)|qs|PIaxY0FuUbs_i7v)B5bOhDb)kL|#TC$DORuvOov&`m7 z)t(8?=IhLA$jufw%;pY`7Dn843z#i_H_P{Qvb5{2Q`T(R+H4Wm^at?V*XG$uVXG>W zzG@w-q`$MZ4s(Zh=ISEmJbmZt-_0HW?Q86s!#7*KfW!V@!j1?40D1raOsjj*n9z2*)ovE%U-3Q7J-boZU6;9`oDZ~5{9Nb zQpFDP1@BZval6;^uSBnhz-)b=*5*dAV!fCKRE{2RRUt7Gv14rHBrbH&m@BgKYeA}1Wx_b$!)wpbyaq-qKia>)7mqrjzI z{1Jz{`3gg}pL$Vy>Q-5bvCL+=9r>y$OX=3tsFnPo%Uhn*J2V<)s;)0YK0VIc{?NSg zVq3hGcJ95%7tBFxXtb~Ln({I64#s3AKyddzjc~5=rjYV`rFCZJ0Z?Wn!_WFmb<9A+ z24uCrahcU03w%{qc&DbtGm)e)X`lb=^Vv_8AqD2s4fGSV(8jvVq)-{99quoepn$&G zOt~^ep0lzhOEs0LK?a$he#8cMpj7mDj@C>K)CdVGP<5h782`<=301i%<8t|B#Pg~+ zLH&YIdcioLm8^XMj|cQ0VThxKg1d5~x_UfU0CrXE6&gd6!+XOk(}L)0+oX=cX}hM? z1f^dEk!!Cm3hEzE)2Vh0s>}``?c(4uA5GK7OQ_`GtHOLb7fjxStK7x2 zxxZ@^S6%;lm?#b44ytQaY*u}m2N}gsN;MB|{(V{hMBHz+h01$ISz=CP-yg6c49l%G zpUDia*vWFtTK!nOv%ZEQU9W`(r406}FeZBcUP_alIXOrT4m#cPhyI<1UVYk|7<8$M z&l5*MJqZYtS%%TmiC%blN)(mwcj|%inkIOqYuTd-JxjNB+L>4ll@Xp;I$^~&`n#?Y z?DPE)l@pCms)TTnfp4V+DKsA13<__gGcnYKT&-IU6cBY|`RX#$%M<&8sH@r!Fj4xQ z0!&-c2B{1e%+@+%@go{H0-!AK9lO}O!7+yOwrnjMTDXT!aSq(ftYf`wmp&v3JfO>C zFJae{Z3oBSEnDTenB0SFuS)bl6x`cU)DNe77qyD9F*~(v*S=&%iZ|{R9Aq?Lzu|uC zmARb=&kS1MP)6)6-8)2$&sUv<T(myj8IQBftMyV*%E^E!n!Jbewh zkE>q>(}{>gD3MHpR#kdZSMSr=_KV2Wync6vmCKJA_+WZ7X|nkTUWS}#%bAiHmJ!W# zeIwabJ?cNiSUTRIuesa#W+-G`eocug22>`$Sz(4LUG5zYIv>lXVG#?Ye@uQZAeReT zc%<=fldK_Zo=fK-rnBud602+e>Vlf1-i6bU2$@4ZgAy_IiaVom*HXi{t;J=nB+P;e z{E>_(%3HCEXkRRce8KYTJlv8t!!yjtoLe=iG z-+Dn80sO+~yUgBEY{D;>9F#brO77l*oq^u}BTHCxtWE?0*fP%1ulN;y^yhAn9a~bQ zTx+g{D$O!T+o;TzEz>euQC@(_7l2Dv(lQZ124;pJmRUcEkns@6%5}kq zl!oWEdEj$pR+mF@&K1ij#n$f6KS#qKjuIogb#bWyS#m{a04ksJKahvldJvxUK>plt z>2y}1`DaA8%&`%gixwyXzkCcYbkl<0DlXN^Ex7Gem;iZihh0gBUiQ}h)?HjG%h9VR zw7v4li_Pf*X= zP8#XA@KR9KGv`|!9J0s#-_IHqnnUBOQ;T*M9T0ZxJ+0jYil;5J!X6(xDRSsX8>+-v zwnX{+luTvr<&-l!D(8(jj$X)x@4DR7Go|8_6-2bb?GR zjykf+jLC#>At9O<#$}-#uyA}@3pX=DJD>MBT}Wf(UVDzd(Sy*Z2g>JraqU@ty=Q(O zNi#(i2(d7Z*^>jTkTeX>PKzO0hUYL!!*sFj(=qtEYDVc8gD_g=GLr5yp=u}QB~$Rq z#;EwZb3&4a!X-hj9>s#QYN&Px-&c0TF+uISz^~Xq-yytw`C-$Gf)t4$uu{HclQr<-2G{Q9-u{X3HD zPD%G3`cpYxH#dCTI2)%y;u4hx;SRVKEu*?Dcadp_Y9fD%m?9`Cg#{;f) z2LZ5*?W#vZ!B(>dhu_GI+abyA&95cGsX|r)nzq<(E2D%GrW{rWohYI`05x%U-}kD{ zSKEC$d^U$>{N0ebMRWDb-~UyK=wvLuX;NhHmHOFJCM8Z+)F4V_+(xlWA#x8>dL8K2 zNc62WKnS>}WX75QSxF02Ra{pUAxXb(1o3ngD72_uE9s#RdxVqYx)Uo+Lb*h1am7Lj z)K~4WOdbfn{WH3QvVaX(PV7DD*8AMfP<8erZ1JbYWwn}OKHcqL=-&@hk{{Xw7mJY2 zUhbCb-oZVn^(pk`r=VTE0E&h0eo?>x6g4F3C&cWg!yH&-!91a7GxZ2H^V{z_+w&6= z;+HEiaLSvZm$>m8vb7$>coNhd1=-T@V$ly~OAk@*3{hqdd?3qGXTdIL8M?cS5*E&&LQ!&F=WC9%{EbD_VD2A+Rz@o79G0_sSKKB(4fmJM8t4m_^ zrtWfxOxgxILrn9GnK}0)s*NrB9W1yg{l)te)`!uyO$w+QcI7&KK9R`C4LaZeACU)( zE>Mi?N)rgj!``^Yl^4gYOvZgwyhSd)v91$835(kijhnUfo!^LGPLEsv823G0aTvxs zvf=R0JU&SA#n0aflxfJqNj!7si$k9q$NGj-;_*os?KRmtNt|XUvHB*_i6=s1;y8;*tG%Q(_GGkD@`!oz1?o%!i#Um0De>}G zQ_M1QBGUG6Z=#rR(vD@Acu6vNGewspRqJ&Ero>DiP7m<)VshXQF zpKdxlb9(7Qd>QN;_gwtCDPw9HJIjgV%{5o9%#%I6Sr3B$g;LsnXizci!3-AitaHpj&OA?TC^{A~yfBteL!>koFrJni&kU z;yMJJ0k**BTe#*cX5=e#W;>cfZG2#Ffn27?DJ`kq+KQZav?Y9r0u``~EQtvx4wws9I)9jDwuE zU55g<7it!ZTwApcla-g=6ONH74u3(jhG8ZF@Zj6j4b_H!ELb3TS4KL?(M|B^_7 zCAfa6Y7H33p&0;BJ*}`fS6D2`qgx*S^%#~)h9$PPqUKvM-0&n4JdRZSJ_iZ&WNERH zoo}~cx@b9_|4B|ubBXKo+Z;H_wf$F9dm5=ks=2+01pk!@f1PtN0V|}i1FND4^kQhB z`wUF?TInA-N4q+Qdcndm@Z$7Nqq5E@3_J=4XF0`;owpkk5gWyl^3JdO+vCk4=Z!@^vl z;Xtn^9=bp+?p=WO5>Gh3dqp!OBk4v{OYJ|9He53t?!g>VJQ*2vp^h4&oe zh<&uGtqJ%Z?os+SQV%0eIq<9}M75XRqSu?y8&2w7n2v91?vKcVIfTNBiB9qbtfe>6 z0zO=m6ySmEY(*qu;6YSV zDC->$fp%VjJl8UcR4;L87U+y6`dYX`tX;<+PmLU83H(;&rOZQe@S{)MMvq9N>CG`T zyQ8Miv0gkp(R559eoT26qPjb_fk15qz&vxj^y9}L<&4J+j2kdPuk%ctXR(++op{(i zTE(Sz*#u$zZ$e&*q__*Qvz|m##!`|dpFAD6UY<5UPPqpPcp`_rtVexX2K-t^1Exl6 z4yQtpli^a6c9mmMgptbJo?%XVF{+-EW`OJqKc9Z1wCL}b&%GW0cJlArsnp664{)`4=n)wz#=Gp) zL&I$ffLuY~RvG~SWQK=Z;h|QoPzO@4J^ovI`6~O+Drdq%ko8h7&-V`4%HQ2F-JeSa zpTA3>AdprX5N=I@0DxNor%JFBWzwCz=1Ex_dpfG}Q*(-1J^{8{np2xy;8}@|M?7;` zG0a}k3-kiv052ENYxV!!UgL4YAwSm$>Jw3)r{1)DyG{3f-o#^Z_lKVP;>)4$1Oi9^ z2M7?>Uw;NgV!-*pM(DK-r34luPlMx-KpJdIT)^aiQd1qXCyoM zIeyoT{J*;7%$@`D=fFA%QQYfgobaEW!>5FwK`vl_(%LWHUmY$igm*fn;nyVU5VOQD ziLKBpgM1>v`);IR+N&Nerj%8()(;9|fm-jS!;|qaXAK8{#K{0N9{Qjae2)Tt3_$Fo z_E>^g2+uS4Dv)W*umB27l<(+FgZE-0L||nfM+KUiz4Ul894o$qMy z`2I8K3j*{O8F~W?wj_WpfkRr0S^KC}y4=)C)PQz=4a45u74hoz#0{QBsPyt_WpQWKJ|Wv@~ePNW}OBt zom!y(YDoW$sW|y`72>k;_ZyXa$c_jBU4^{9{wj;_M-LU{_V%nt?i8|nScLv} zbZphLe;?oebFqzZ+J{=U@&Y68`qM8z=&pCO>o(4m&j}6udyHSve!(JjztfQs-w^zN z8sQt?dlT4Jq-1~hGVgHr;G{vYg8x2&^*tUFK;5_spNM^)=%<>8FbW4K?e-Ia;y2EGMbr zJXzP?p1*tCEphup1oAo^`O?lx7h4Wkt8fF&BH?v5-j;Po_F?236$bPAXiT{SI!6tF zUb~Zpr*+HFVCNd(D6OJCT}t@()uO>YD$x7ZfMxx3Pm<`(KOVjZ?nBvguz3SVMyIjj zTRjb{8OKj?1lQm%<(a>q?X$I@^r(BGa)jQ2^FMa$gAY!R7#7nqgk!U{ibCIT{h)bR z%O9DfGoBNMknJQ?YC!qIE8B4G#7?%SdE%F~Iu#8oI*w*uX;KEH`?U{@uM8p%(yVVBNNZukw10&ikNcU{+$iYBeTZa(C2N)KV>vRsN#ShV18hj2MfPDJAD z*64DUf~5Bq2qFALVqP#G3~NIxW~kUfFF}R@p>Xw+Z{Z-HdRhj_-g>)41(6UswCYAq z7*c+EN&!HvmOG?9xKxzwHJb#RNZ>8D;Cw-xP2+YKYF~a0J^NOeCAF@>xoN*?(CAe3 zR`+xwwdlN@%KN%R1PzMWNrG@xv^?}~rAg!WPv;+&pL4FBiqy?;-i(1hX;?4h3=7`+ zrp#$e;?Uhy`Dj!4%MK#iOtBmP*);2R8k6bL!*}^I(W5R^C`C^o<~x&~Tw*w*t}aIe z6Illx=t;iZV4rFv*HFOv#y3*cYeqIXYu^}`*vY8t`8$hoT)ctK=UbR)cJjdGui7bh zuZ0DAt-nD7S&n&1-!ZBbN(WlKZoLDo#_XicsL9E(gxTkkev8o&_}Y%DxlQvU#5ubt z_;{nJ;{0^y)2A@uG#Cwm5739vKuDPB4o%j98!v1QJ2;QRmRBzk68ttb(6A~H=VI%Z zU57-fQszh)-}!PhqgOlIKSdL|ThUh-0#@(4E7EBu??X7<(Xwu>HmK*(S`6(`;f2u; zoNg<22Hslcn$76oUm-#nB5nM+zGrh6DC$TUN}~g&L3{%;F;+xDyf$5aVqCQLw#{{P zfKLZsPOqALT~fTiWsIQ2NY~}ak%TWCcJj$SEh6ek@wey-xId)p)C`QodTW2TktCy^ z@`R#KSGCytFki1il3#?$#R}H@^lE$ExX41VmoF8?UtMmCe}xd@;`tV1-K-Uz9qyom ze~o;?y}}qS_D*~*+7RzP%oss z=m##IBmR-;$}2h{I} zUvQ>oAAMg`RN!P+@Ep(Oi{H?PZe5F&x6KxN{zJ{8&nz+2yg=roodK7dY1Y2RJC%rd zBRJvMdXyk5+DRIF0IWfaIcS(fm; zSaeO|)=T=EmDMk!ueUhe8&RJ6IQu;MZp3?Y7Ky2f^YIEyIPq$Z8^TcRz;Ez7FE#9% z`CG~6kgMk<+cvj~ExT`M226PfB1Mhf_T4`&%?RWzlEfuC+5ANA{?%f)ecydG`rF5f z^0FY;zjqaybJ5Q&I|c4D<}l~UfBa{;TYHHSu0r5fAJ$<&c95*Le#$ezNw&Hf}G>xY0L{jfRWpTi!n}O=_XO5kbBK zJ@_+}`1xqJ2mI5m{4Uv+1$|#f+?e0yX1eBuoZe@VSF#?Ma&NL!9vH&_}#L-FpTWVp`R!&K1R9LuRc~~lIR-V70WLDO3WLhE+2GR}eSWS2!zI!WXZ;U>DSSUr;hS(LF+`Z{V z1P#j?`IOauoMl>#g=j1`2YCw`Wg9cKDt|6CmbyFc$jCpOB(IRC-B`OhnxuKzwo{B* z=3y_b{RN+mwmhvQUbDr)%u96cqqVR1YiGzmhTpP9{A|9b^k_Yk4VfG+W{#U?d3d$h z+TZNgujM80n5S8K1l3pS#7y;3*V-#X@P#$2Y> zE1|R8`^?t!ycd*EX?~%9)(h+)Talto%%=~t+-0^l?+H69UE}l7RxV^OE-c+DErFbZ z3p0kODJJqkl-#Jiw}pp^8YhUu;SU-pICF{5a#c}!`Q6f;YS!+y?wp8Mt{V#ei;c;A zHjN>X=KYYN%FLLA zrx7qFL3o^|`46}PbIfY^f)I3f3|O8xOLKCH+Pk)E7cmHQpdRJJO1!5HtGLwlTHJaL zb-DI?>$YsF-`QzbY0-!w$kjAvVgfpRfwN{Fe9ym2l81!gl2^zaqUMJS3<`0V<%~S2 zn}{gd&w2XYvF(5@n`;MByuhrB!{#h^k;8Ghtj55dA$uEz1PKhJG{zw-Q_CqTQj7`1+oc! zeGar2zaKhW`k=t~cZY8-;rZ9Niwu8GNFJn6{y@yLu08RfA=#awtIs}d!cvT|qmzFZ zPPg~023o#F|6N>p^*pR+&jsLK#)I;3FC>_sF?epYgJyrrliP-U$+FRh#olr)@nO%V z=H4D_|Nf%sd)=*pVVsure?TBL5yWg6#$J{q5(^Ix?W-YB2XZi{Zj>Niy4-VWc3lV4 zW%( zQBq<8P*ySa%G#pFF@m?LJ@z0xKSA@fa5SS_6uQn zW85ps!qgXf*1-i;VOwl>?2}0_b#=T~a8c?Aa&e%Y{)SJx@Hqjjx@E@Sq^&QZW6@}q z?2bR-vV);HzS!5f6JM2TF_#z&irucEceT`)M^~G)Js-=QxUh~fyF=}~m0|jNqn4RY zI>z9NK4V(L{Hnp0_(!#4&|98byv50}Fm(rg>z`y6d^;o_I$mn|;O->HJRw5O^nfA5 ze2lHW6&w~I5G5+D)tVYH7?mLNjA;Zb@9Vm$mzkgvN@9WhLsh@5s{c+lI zY-r!;4}L7}A!yAG);gaq384|Vfw3l2a&LIh6&iXI%I`1#kT@}X8@<{89#+>HrOb-~ zK3rO*$#?7$yOtYsbhjS6`({Gs@IsvR&I&8`b3~M$@d}gwG}~vGE{+ruXRk`&K|Iv6 zH9sQ~?wxdls)%yDt0vEmAM>05fI_-Je3`utr|?LECdCHL9jw>0CpMzE!o>E32o{^9 zM85ho;BaYptA88jQ4xtiJ%xo|7N$WM?2OopMm?em(!z;Z>yqnH(|?A1P+vhJ?T-LD zEQqcb58yCF`rd}&Xx&=8K>LQSQTVXGm+Z42(uscKt6b?rwMqkf3DEl0u^iO%Vzt3| zC~0;4dvB|B0Zqc~XBnff6F!!So2=dat1Vi5`$a+6VZ!~tjZ=`#qoM|l9b3C)o>_gP z>c7>mY8C&^kQ3@^(5q?gsUWu|){%-}o*i#Q({yJyQT~1h+h-x&*^O(zZF~+B+|zS- zXIRbNubcex3(KEcRzIprs){pGY-cR9(H5XncdcN$w;ZL|$#aMomT-$Rgj2O~$tFoY zQN4@o&#ruqE&t_BnOg!LnY8I#a}VdsvlOHDg!#Q-*^80wSP_`59e8xj!vhvRsQ5__ zvY7PJ=tSk4DzcF`6m^)AR*Jf+)xptGyuMxIwc6BMx5v zQateNeD}?RfGYfDztNPoL-}j}w(eXPomy?)t$f=xZ+7i3d5-sYv4z^N?eG5bAikgZ zrU62;#Hf|Tpil4C+c%%{2?Y4?@!?T5RJHGImBk8Q2D1`B@4Wxj{4=4=ws`jQzQ*^! z=S04zco)1@d3M2))Z05{wM!t8^m8z#9SqEq7weC6}^>%YR83)V35}K zK^edOBcc9V3V}PKia(32z&^yKZ~sM@Oqe23_3%(v^d?}(8Xa<;6H);U%vSKvlMVg# z2(p*vf6&8qZi<)y&`AC8iDkNP81yDN{K7`S*{2ZGUSB|rD4-ZB3=4y6g+Y+ zhhdIHcSb@DEr)Lb5v)_eJken2D+prz3Bo!GzYF92TM3`QMxrc{d!~`? z}mXlwE78yxQwdX8Z`b0tB=Q3oX7b3DrVN8GOs z^jk~(ms)IPGYWl3umBo(Y5xF}W1cUgkH6s1Un}_~8j>IXNXBs>WyIlOgp@#H%4xBc zEd`>!nZhrgN~`Hq2|=lVI1o2q{q6C)&hhf5Z_4eGL_({Q6+9X5WTyq^P1Qvok{>^8 zNONcia8gQna_8mK*NM+K(%egUHSVM}+r09bc_PuB_OKz~)p2q!9+5zTJ*XB4gQv~a zCvMqTD2$>H0NywQbOHy0G?Uh>@cJBs?4-bADn&Dt(ydC8-o`+3mC!U8-iL7KKG5q+ ze}LC|z|R~+AKv>hi%4Wcmf@$Q4KvvW@z{bpuRg!dXj97S^36)fx$Wcu9=MY^WB_|a zHHD7MIJJEhLwGpfk@gOzK)qU_smW5ad7jp!L{DL{>@p2|(kyT{+asvGuGW=y_F86FMZ)0w(S1*TxjGxS+4jA&Z37VUe*d>Iq(^+NWcb`_>(h`YMXyb3_WZIFx<>@l_nf`P z-Qn6+L7lF{0vVZcY@*edsbW{)?Bz_#b$)ov{Alx{q~ANb_D2{Zjo==DF?DpGWQ_b> z$N?0&&w52F&=QhwN9;knsADWn|&OU z#D0KJAZ!-l+i^!&^rmSHnpC(4Amse)$8VR8M{S7^I;F>K_3a z0l6Q`P`qua|dF@_fsWL>X{?8u_N`dyvPJ5!&Hg=o#Ik~8!FT}(m^3+C6z3X{wpNySFADwv%T+O^3Vgx)K&SOQ`dMr3FvY3LRW`Q zL&BAi(PHee=hg8_H9nhF&z2#dm8wI$sl!sMqd05zrE1?tRRNMUk$7lcRE>Pk71~_) z^z)ihu{v?JIyXH#kYxD?uI{FD4K}LWti_P9S)0KLpL9fuEbt@Tp)vx3c9JHUWZU|C z0rlNI^>*|1V8&`EzJ?<2Y9b%PEe|zU*YI0Ua89f)^{6^8AJL^(UbD&1!l23{0Ck+_ z#h@Wi&<)>6jm`xqQ&NQu23BVXqAsB{7=(rLQ9Gib&ZH}Joj*^{tMzY{50ILiPz1KN zX6KeDnytp+K>{tmlnojT(FZ{=Ab1JLMqiG4ra6VKWd+|PxYgtZBD_PX_O!In_7cFT zXLS5h4f`5$gLfeM3UAR+PaM=0$2ah#g5R!6-Lys1r$uW^SqBcezEyK?t4)dvTJM5> z@EC-^L0(AG`s3SI4X#{2Z+GMDxEkG3>e69TPq;DDCbYnhQ*T}(kAOlYY3*_C&$mp` z;%(A;ZFi&Fy0h9O{vkvEwK)wc-K;0VH6XaO&JW@(AH_S8>^eS2cj%cC?b1*@Wp|b+ z6_^HLRT_|YB@Maoj_?0EUvm+Ql!>-JL=PN9{Bh57Ne&|nfBjawT6A}nZP%;lu1{HA zZ{a;*T+kRCMHBCuWa=) z`Tgr}<{D^K9str&b|C0z1@tR=NKXc0UQ)@0fbfvt4G!OHA=vc=1oj2r8wk5ca4muQ z3}(D0Ksm0Da2ItVdOG>G`wPT-;rDx|wz@W?pq?bCHxByRaa6`^#G9+{z3n?ngBBl0 zXc$3ACSgp$mOtVTFAIa$XsVQPWk?_Tms^3|Kfm4!}YjP!MA*WR`uVtF;<`-GfqUYThv|8t0h5&Hok; zDcYXmVVEvAn%ia(0;^Ex%=*RLsxQo0vT_$O=fyV{MxyNa=$FigOiRO0YOJbr5CI!m->wzdG}_Q9$YR(PuVsFCMC)(Y5Fa>(bGE~ zfw)R3Ud&S5M6KK$q!8B!+oo7sXIRq*EWf8`&7}CW&j)zu0Mw=YC(f_URuB&rzgcvw zxMzZ2>)0q**epIBsiWEP!3uo|D*~@Mc(iv9r=RvEC>q@L?_Ta%VoaVb5=pl3Vqd2Yt4)vy1!= zkalUL1~Gi3$=F@Ofml@kZsX3bP%{a+J3qyi(0O z07PI_jz`CA;N*n+luPyW-;2{LPN%gWPtBT66Pk{`kb#N^3|XoWilF5*@_u6RnK1uZ zvHwyja8%xOMjP<)CE@v;^x4Y94SUowO&#$Tn_SXHeG%@ z7+^RuKLqXV1XMz)5j+~n`U`hL5!|=`EAej&0d;$xCzby0i2h$!;l&^qru!KhpC2uE z2eU}MT@rb;((AuF`x^7vaP>h3Ql(b5x|CZJ2UEIzGH$s3DCesGi!G7FmB;a@gxldq zpwF^_B9o=I98c}kVJHxq077q@fJ7*%w2dn9k@VERKDE>Fqa&((FD`wbXSPh2QNf}$ zXJjm!`~W#GrIUAWYMiP0g}gCwX)Aq4TSetz#e9~--|m0ux0cy^JEpYx%h15_x1!Gy z+3rJYSe)FG{32iS-R~`PcKua0t1BwZC(bMIcMQa&b513fU(IGs_B&>OaoVZ=P_nXO zfv_h z8+^xYmD~(TJtvfQ@%l3;N2&pzolziXs}G@{Pen%|b?{sFa7}>uRs_gr*Z|7@YVCX_ z)t0mn$_Z7N0q*j3w+hI80Zp zn17$cxq31|<;J)V><+`{(~dO134=#nvXgBi{I|xhN%bpiZte9eL+`^V5!+p*KPqII zg(`-i@I&a3!`W$Asivy)r~w@Fc6Inr`km{DfA(Kj)thRs4Ie+!<6eLKxZ&~5t4}{T zAN~@t+;)9x@~)lnnVI{t+kJC4H>ru|3Ku_(EW^0oKYg7PQfmD4OURMBjo!CP6G-u) z?DKaQi${+fnzjMsm$p}`nm{6EZYHiH?;L&`_`hSG1(y46PP%WuLz;U9ERU=DlYigB z%tJ26F#(gy3AudCu}Kzzl>Z->UeJFQ@&AGW`H6W_S$rT6GlQrpvBZL!j)lSR1dV^$ z4rS*to$$!SN5=7~`-#KAEkPedRI^NJ=9`;8Nm%?y({ZtiNWG$)0Y`zQvTN`cmt2#FpDXh#RL@y8@EhlwtROQ~GRH zH^^w!R{S>@u*v%tAGvt%vbyJx6@r-bArq51Xu5_H`OI&yF^eYMcB= z%3?4N6}O3%y!Mczp!q;pEZ+lVfa`0oB-2y*UTIc|uM44E@ALR&IPMC*l%_Hr;gT1T ztD1@wuRK8-7Dz#ny0h9}`zU)c`yd-JUl(ql-Je=q@+B%bBp5)4v%V;+SzG$>p!@1C zO4>`8`M@XAdCRIuU6*K~1aWDiK$v8Uwm?+!=c|IYmETuVqqP%gt`V^=i#MCyhG~+W zu0G27^5)Wbg+GF5v6dOSk=e-~<`W3_2A3QYu4SZY!1$x)_Xzxe2!oCwI)f@l=tCM6 z7u9da#Ru;y)Ydl&ZUmAU`|QLsEDItF?YrvGbvB`x1Xc5VL-mQ1Bt8TceFbSlnkUin35z;-zgqTok&JyI4l7Y%@PaL{_L3|6ig| zlIdS&kzD178-w+R&9~$wUc&fY3oZ8hTxX;h`#ueWC3M4x@i4wmR9_L&;=~VM-ktj{ z9*pQ~T|XS%Dxqvv{j)V5XZC(bec&UNptYn74K$0o70Ii(H)Z?G$EgF18QJUf3by@-%$DA<$mYEa8K^W z8Igxa?yG?!$Dq2yiH>PFJ%j~!LJ=!_#V0gw%p^(h@?IIs=WTDGQra-JmVlo7_u3{?>)WA zbGtG3XMXx_5TB6w8E7!B=jX1kMU?s43MNm!Ei;h0_Y3@uw)*7ISN|=TIvcHZ-uWrY z4Wf+UBr7`A`zwRQCi!O9U$`?U-E+8N6fSV|elOO}8pWHf7C&-_sCn>{5_E~`q_m^H z&BSmoBn*?uWeQW58YIQsoS8$sOKRl~%nVaKut3pkx7o4LN9ejzZl!4xg>~uaK1^tO z|47!fvn!3lF`1w7p*o~8EM2Td8906s@3OV*eJCmCN4gQ!6_yO*AWN3?aUSVt={@j! zR4d~{XX2ylf5A(9VNN{kT@nw<5=2BTuim6Vybq2A+%{jq)lPK`rgk_TV6E&s*4>6v z4sq(@FR5`SSVi$YM0uLdbqA*&Q;yE}4Z%w{+V=paCx{QNTc%QR-}4klEkegM%r4Vx z$aUk?K-_niOf+>QTCR=XV`6Jj_3g2s_bg*tx%h7#(Y#h&5e5!150JUflB{daho}#2 zQ>xng502XnGZyi=JX;Y1KD5$_&abTH{&_;MS@w8JO*~z*WZz?}kvMors6g%Mps?Qp zn86pM*@_w3f93jhU};%Pm2KE7zSLwv9m_2cnz+QR`W^v5XQzC0zMEe654%R1Gxff3w(8z@V*10&7l zTg?Yh6U{6~B{<_kYR>Zs;%t9*;sq2`7(Cf4Fdl8D0^x6|9WHHDtN6NXW^;H9c;VmY zIKXCiIXVKsjs%Dt8yfJ!1(C#HVsYo@x(Iz*RcksU ze_SIHZ7;+GHkq)vRcAG^V=~P#D-(T;q(tv}j@VLCWd~TjFA4Gii-es?95kMMi|{+` zc}%{;LKxs8>Cny0^xD5y8U7}r$6ap%v%OU`b0`tgzF8NF;A`cQ-}N-qkGig%2?!v$ zBJSIPnR1Ov?I#@k2lnScJ@Of{Rw*tHX)%`K)y=QAOXCFLOEkHauAv5^Cdfi1 zw`b)i_(z)yg4lI^{}_H<8puahI5jEvN~%hX38&voDW0~5t(p)Wu1ecd(o?ISV8IF` zFm;w>xLTqb(%Mm-T2JTo?PK<-w%0#L^3rt}j)HVk?n~HLG+6V@rS#?ea?AFA^H7~l z0D&r|Zd;WMbEiw1^_DYz84&Ei`f&Q!uUUX_s&X%IJY^b6CLjG+*iw>ytVa8&I*=)! zL?!k(fX;j{vwxeoB^tJom$`A0uPWgrbUwe8@3YXjXv55;Mtw~`b3<<6 zwBPf^pJ=6X@SP;N=;L%(86LA_%`bh!?CF2y)D>qEcG5XUQSxHi3xK~PCAWCBTXgE# zrp2yv!H7g^IW&z#p5Kzn-4yK5vVDu*hho3gOG~MDP8Cyw-FnvZe zf8gE=752PZ07uBE1_C9k>awTv9g@nkh11N`(uGXdsuyLE9t8FIGAy-{z8bwNN~rzX zi%{jnAEUd~1YLPI1_c5MEFqNUQ#Ej*lk| zfJ(~;#5h8i%V0$W_;^W>%1Mx#2vnVH6s+ZOt@&1j$9%9cDC7|v{JLQX*DB=sNrden z#2(9FN(#9E$-JCEzN!u_oQSlWKyKoIHiMfF1;c0`g~6P|rX=C9Ae!sdfnvn4;vmS^ zr_pU5(QmL&cLFR7En0|2jH8iMD8!rrY|IE2F)til7Be#u^V>tkh!E&ANZVY(zPjfI zNJ0b!p)P~8D_|L5UUo!{>%n5=QC_HUCh}6!FVQlpI-JuzIjjK|{ge_pKO;%V@BniMPzS|L&0lLl>*n1W5b>q(nJqAkSG72@Cm z2}m|{OdE=>RFbx+l`o>2K%mel!eEWky zF9AMc5O-Xi^e0GfV-Rq(V4aanRy0D^+zC{OKs00F&4bB}q+}fX=LCh%FHj6rnakORgaM~TU;_=rI?KU+dNYgsyHAiNEe zLL6k9CF>)zN;#Nuh=akD+6BN*io(|b5D82TUIi}algNCx9i0$E9*9j;%na@}I>9(+ z!DUjP+g%~9D;`f9`FT+1qkyyQCmNp71%7U6Zg(FdfKszy=zq$f)6959 ziUK7`AqoQNzRlL>`1VLQ8wdUdptCvfU?~#l;oEF#3;A2kjEGW(F$`k2na5g@b@B9yEZBEgwR@3t&Cs0|aqM-v!o#2#-gB3=t`M zf5^s24-5GhUpdS7G`x);eGUPi-~fPhV-=IhcJ>{r$epE_|5XvWg$DLfpduT-EevbG z8R)b`%`bqJTzy5r{cHbSNn9p!yyUyspYM3D%pQKORwpi)qVa(~1@D<#;YNPo2C#tz z%1wZu`B$nU3nn;3sRq9Q`ZS;TXg)g9B%x^*b$=vceyn)W#A6fHKL@7F!|vJ8tztB~ zBV*Uc^X=*L00%($zeM?xHnC5hlqY$w%L4ky$&{jTUB>#UqB#?qO%9qhlOM|)645xC zSVw{R-?!a4Hq!|T&J40S+fSK%~Yy= z`m**)NbL%~wy>^#2m~iO!b0XD<|XE+Q1qB>19gSDfD0|#N!`XNU`_U5$fj$YPrV&p z#oJw1A2wY-z}YaO1o`FNFhK{IN~NBut4Tj^SU3kQKIgOYXuQBRZfurq(KQU~HB4S< z+JD}3yxCAt+jQ}~k@}9uT3sz=S`qbDBTU(Z)~*gd({OrDcq7tSw$QYq*9^96X5|vC zeZ_kewo(tii`8{J;ocRZPGq%2hUp_X4-SFHiM?xFg`7ATTQFn_NK6tK!DqDb=xX; z^V)TPG$qn=b)?vJrpLOYpb|mVdlJ=SM&FEY4#OG-;w2ao-KjTqlAMTVdkX9^CtsGT&;e+gveJ8 z6QZ2Yg7T`GiPFto89oeK%H^?~1N(*4$1{W9J`7;%2f{W7K-`1fPx{7s6*m7l%nnL_ zz{0SSFp0K7#Jd3?y*)dP`ZM1UM(hf6I%jCq5PNza9l2}>&=)+$!ajna0^Cp`vtbx~ zNT7UJOLpk)!SDx;k)~cgVW-hzyFsq&hztCPvdS>pY(V&;@7n#*f6)aJF{4`7M_MZw zw#0uu{yn0K7&UGn-0b~zS7ij$N0k0Ox)IGFAPn?ikXR7R1q6j@K&m(=)R>e-3>BA7 zkc)i266e1#MWesAy4)M!yN;E<7vq7xlfk~!p)pbszlR_Eo}g?P1tZ2IOc*_UkuLMp z2H`F?zP}LL&z|U0TmGl2Jv2gpPIUj8c$D>>?V|f;A5-o-YAl-iDURBFaO%gqfs*Uf z72NR3?a8sD;S$Ah-_UU{GuTsAL{(R z7CJr!f=v|tKG>SNy8nAYWwy}~xe_x={oJGW0@&J~4QubeUj8u(Lz_`D|FL8qKTm_$ zp#gXn@RBqv_6yE07dX%9dP&G3EV%GrLk(i=T#91*{`|vN^ShXo%*j7r6;jgK7o{$_ z%P9V=|DL$w_eaEDe(N7%*D;ABklEcU*D{{@ZY=YnKI@(&a<+tGy{qt%bHM@4-3p4^ zXJx#3YGF^xc%PLoUh5yaM$6S>aPza_iaq^*njgp$1@2uM`EV?gS&}Z%%lDlN&YT(Y z)W)17oMX3_LqE`~S;(uEqEt-4%1#W_nqn#MU2^89D-~4zUYzM96$F4|v7o3+;?DGM~^RixQ+|m3Qvlw}MCyhlu zpTa?weio?W_@2A7Y{gMOVwlo*d%H=T<-QXjGep;h-kzD=y0!-U z&9n2Ozan<9WW#G`MPg?|qTK<1ZS!&`toVBwz}U3jtL39y0BnYVMH&0LhIqPK__w3pb&uMJ(owXR&+t z!CxqXKpjQE2tWm-hidzkb~C%G0y|ZI0h7MQ_p|*zJKSX<6%2#f7Knl-)z)$q~d>gp8)(WL5kctws1Jk4L$ZOZX}f;M)eV+gGavx zsrvEA7MxUY@oIJCNyI?yJMu|X6JnZloZ*CY;RW1ykJj=|Qvu#k2l>Mqn$*F4y#L`{ z9P(vi_4}AJ^W3w7Ii@0B(Mw0*Y@W8(@enrvRPLH4`+u#E+Y{puZ32W2yMUc}DxG<% z2aZ&ZUyws&9B_m8qR!!{!TgKw3&weX{lMmYfdFjA{R_?)ohv^7eB(0J9QoUkyflBN zc&QO^wh5xYFMk^aEe&aLEuE zeCqh}UdqjPR<21SW_NK?d-06I60f^m|18DZCdsbfkCXX)J4DxT^+B|X#h-Fv3A3`q zrpMizIj%ONQ4hK@H0kJHJqRbyG#0HFztC}0{}r(rs&8i!!E|~VW8DA8Bp*4~@{NXp zx6yi?L+iN~^2xU2U1tf>JPjY z`_6gXUtH`(OqSVbP3V0O*^B>ZX-O8TkW`P694*Nldpm+_MY5jhYtY=^0yKehu#`IUO}nCWR@Stk7B*WUlopKx zmq91Yb8Aow8UQahv#trwa#`bbTpj*FxxdCKV0 zd)_T;cw2VlvBQ12K9~Lbdw&wkl@vD%cOEF`j<^gg58K9Yzc&%A&{qdEM+I-Hex$76 zRbvd1eWb~n4p8XvV~CG*lgfb{=iAJ7nGXZ#VE=#I{C}zMV9W^`6xbXDLr@qIQ3N$O zN>DfW1hWwIKW16P(yyIyd^aI5{mj7jNEFP`j+pSeB}nwKEL^Aj>RmWce?N8XT}suL z|G~|f)_&Ro-IJ!>73=LDK!;9WrOYHH#nMcC)DdX`|Zp1-#t~ub4^$>c>ZZs$wG9`Cn?LTF~+iPY>PUn_iDmfomaAL zDSL$_Q6wMjf(J%a49Vh2RStFIB|RTGz4|z?gVpN~ta9#8w_Brldc8qNE~8%JHw&@; zdXFZ*hJk=hsG_?zEIYSrw%<(7uBoIoiVQe3)c0vdu}Reu;l^0IW*qm+;7k2%^{B1I z@h_?gS=w=2S)KeoTv;$4O5_QRG>~YOp*3;w)y#dSfn~ycvEgVwlqWMRT<8;3YcC}; zL73O~l)LOpG=2$ zolG+hHSp4zm>I_7{_6YkC(~J}4mOkE@9oUiX-*Js-$aI97bO8tda7A%*sBm0q|TCu z{a+6)&UbA?OCJm#blv;-lHyt(pntZQc4ta~CpUw_Ha#yp zB{Dz1u#wibXtX2JHmi1HYO@$mYrFofh`sKbQkQ5{DjQI}*IxVJNTxr3E?Ir8{$T9+ zGUv%k>U9pc(SSwO3uP23+X+ecs#KjvZzp$UCA2`eQG>d;{^+T)L*tp%^lno;GCI3z zPQWL#Wm%4Es+G0j99_>+B@XR4*{O4ix$?(*7PY~-oCJ2riJ zlg&f5L}@of6WmJ02A!@KLgpRvKt<%5XrQEWa5`36_bW3`B`1uoUu9 zOQkIcP`7;pgx5RtQZkcV{w)jN{Mczx8yd&9_mvs^xwp2SpOyl%EJKIBD`u1lCkf*B z-zSsxba(#7%-{RN(8;b#c-6@mkSD#HTw=Phcr@=8KK!mmGs=c9Sis8Ss?5TJFSw;ltqYn* z*vQeO^a?>p7du<9y?*BUt2Iq74q^+spYD-kE{2*?Yg~TRK2+8q1^(ty zfTxfJ=g|{fXyV@jv;-hbJ*`c24)(VXL2;u+ARfrPo zBRe16%lIv9tbRTqmhdz!@)sW_V}I1?3D&}gNI6U!RVA^vE;vXC=arN$7MI#ZB>T2% z65i&lP#+3R;5sNygH?_XBc47Z)m**Yuz-Wz4sSPgzB_JBE@3ns`} zbC=7)*)=IXOj)A_jz_wE%&%|Qj#xmLu9PK(@gp3gYkJ>2*iR|`wPMs@Oy`ub-(8)s zs&Za4fUVqJhxpz^Mru+2T|-=z+=JfR%0vOlt(w{Qx2-KTR1%1lLXQtX;;*RK1@Cl7 z;|yTd&-sDZB8Nv=ToCc<^WUWR?jPSAf;y^ed7=yezx?bSza+2O(X2ZxIi?Y2;wxYx zs#CR1U{QTHq`j5L)u}URYZgtx7%dT8QDnCLRuwcowNiPP5JPfO#iO2*JwUbReLEg? zTy&tgCt_Q_)o0h9qn$}rO_mSD0vP}42zp`VYZ*R;Zr+-BKwr^z%*g{wq~sjJwo_uj zQAS1=8%6_O;&#o4KU`E;I2FPm|LIsuWju4Pq^2;>e1xt?3nN=s@XYE1_uAawOgSR$ zl(XP=hF(F=!EjA6h<}8k^&e9B!+#41|Huxd7keou)KXp}AESsJ*PWfBo^M$I*|#*; zZ)Q|`q#EJ1D`gHm$&0gPV&C1@7sr`M^d`Nwscglwnx zmNg2R(&yU61u3q19#CUngFPPYG3`kPtBWu%;t*QN!P?cqLm?Od1z3YYH1d+|V3t&; z5F?L};s6%wfRJ0nH|$cbto*E414hZk0hdC zQVba<5qM%K`;fnwQKZ5iWBdEiZ){AtC!qfvzEt90CLcA9idu4ysuJP67G&{IJGhY; z@|Hfjg)REcLX^IIFbW4hBmoN6*3Ns;gD25`E67o!m@yB1?qoRE#2fyJm?eUzGd;r_ z10lV56EX?xsb;<+Z%Vn3T=>P-E6Fl3$QdC8Kk|q>v5t#az_LM@)wRuMK$36yS&y*5 zCKjOf#L?);!zXdZjInk@fS-kRsN6loFw@F0{Y6!5zz`oh3DHSH)X~RT4so7(B%D;k z|H&gS*#I^yV7kXeX$;UAe}Jsxx8CfPSW=wa7eBO00)c+01rA78S6PGJB1 z`jn=U5An9^)9n1GMsSi-SBT3g!p%6TI5+9_P|^xtJeOy@PCREm>{)0RtJkOGN0Zi1 zZ5~$JJT$^k-(Yp{M?pgHJU9~V6b4~?3M}gb?M^JLd!L=Vkj8B3&#K++oLJ`YpXTr> z-fHnPI`MwZDPwynrQ}K0@(4t03H*a3O-KoJAT;&OEA@u;RKjQ2x=vb0*Qc(Uv{rtu zewviQYw1He>3=cl6J1E;cseXU9k_*EIAzV1&nUA&vhyK2CE@V|SQzfVRO!qP>S}3- zUXYX}PUJ+5_rKE#~IRjL^;GJliC^R$} z4YdV97_tpc9~%V^LQk4$dj}CSC0U!2-&QgOFG|>@j}UjOzGYEmc>ejO!jaQC2upO# zu|-4Ql10^2mvbDavRzDaVb{J*%x49j0{X4FE~mLGB6;?A@=zQ(4QQHpd|q^IPTaFrbo&nJHNprg|ZYgvUO<^F)-_t{Mf0SLXPjwy16A@-ycVEdUbyv zT>pM6r9i8!pwuR>k+v{uD$g$EUchL6LS*4bjsk>YfiG=Q_0+c#g?t9xOiKD3kKd6$i0mp?iL(Q4BU8)?;dgDm!}n zU5LDya<)93U1?QZSKjZ$g7@-DCP-jwoxwdG`vJkdIS@t(wnvoda_ zLM_`uxKddU2S?vj#mZE~-ZxB2>9@Bfno&P{Ln;sj6)$=!=%dPp^2J-8cCAPq{jqE zt(K{*)=w1{3r(}?xjJ70537ZJ(1Y?6D9s$!S`FWz+e9RGzftfIx^Y*ibiEvqu1%KC z$cd^|Kl@RrRQQ9gZstmziC$fRZJld}pp{Z>)^m=U>8}G|P_7dE6$dQ(as9MtaXq>$@zjP^@0?k>#w$XNJ$j9J=fRC6}erxv43Td>9y2Y7fVzJ)a1@@BX} zx4tcU6k~;HmH!7)gf}QFS9)?ZV<-8!- z`G77QJY^ncjHNc#r^eKGQ6y16oA1K^>&i-_ww8okNb}{QcoouwqfFbo?{ungbbT=G zav)F}splXp;{)=m=*xfup$t z)D=cGUeIsw7=^)+8P+HV3UF*5 z1)EfOm;nPzi6|RA*E1&ud$w^WZgEk5dUhQ6lM%{!JM`-Qs9f_fc~G5wFD}!7<@X?B za}d#-dF_pb_2Y{P9QWk@uQ346JcNTM6(Ne`f|3$~SpRZK}B^k$FbmD65MMKIqLCk$Ow3-6wmx_s;!SnJ$Y&qKZ3#n$AeSq z)`&$Cf>HiAy%EdEeaBjOr z$!7>Mh6^)PY|#_@+Gw5!$uP%?xWvI_j^>B2Ef|Fhe|gDrq0i-YgnZ&_z1YY4aGJ>b zV)lc9r@(_hzHk5BZe$nD0838+!wZuuj~D%X7YWjq#q@zF9CDWrxlKSmQkyRGT{c4E z!FaB3x2AeUrWEBDrxj*&cNXr=BKc218}rC6kR`2~ObiPD@a0pNO5thDH-v={NklB5L=SB%Ks2}le79f9K zbIPbr9a=L?Pl=yh`0!vo!+t&B$A5NL?2#K+8`rY^D!uv@7t=S|&DOE@05k&Jg$2?t zSIS~TDl!n?{5C%)Y=#-gdk#m_Mhey61`e?+Cuqkn8z-nMkkWH?2e%W$WH29ZIwOHC_}zi^Oz={YVja5e+@FG=bj zXyk??0Qe*6$*Ne!T_)xmmJ25!_AguA61u4Qp(D`WrpgqaSb#(VkN^YXIoNAWMb!*$wvZ?;V9R9vhRB7EkxdnB$bdV+-=a zn=c30xwKvVlU2tPLtci+hZ2+6Ll^Ri+wP?K?in^4~s}G zM!X$5TBbjXHa{z+IZgO-MtFLPm^!<^0GwkF$_M^Elt%^^pC1s;#}>}N-Z(aKIE#-v zOC+oQPZ1YCB$PqL7zZfWeq+zKNfeJH6@4MYI2zJ8(4HH2mP=0{M;& z6yI;G=m!jce;&3wr|P)MBH@EF_+PWEt??UwZpCs_zVbfC&#Ahs@cfth{!JG~x82}< zT%4|&a`V}2y=Q4wmGk!HIu%F7pB7B>#AMlO+^lx^o%GuulbtNJVtLhL#Ul(0#y*i> zbfg)7Sb18ento|V`=m%Cix~T}y!~Loq?YaGZM+F{wsykRo=dZd!-9LImf%%aTBPYz z?L*e1ufq2^WnO)lQ&66){_(ipv*>T1S9KvKIhZc$#pz=y%RV)k;jf+dIHh0pl9SEu zPI_Hmt)IJXc^H*a&@*h&PWy1(Lqc;5u$J?iahO$3)+D$07M7TalB2#9uGOO{};Z4Q^#jnHJ~wA}mRH6_NM;n$IQBh}7} zGcNtTACtn(FcAa~QW>M%&eT-p$N1U9s=cQVMwwDnvAy%YDLSel|vE zOF5cEIdw$qqHMU@U6=OMc!RQCv>hX}rMs@A21|D<6jMF!8+{>Ep|7sF5k}6*mfu|E zd{;5F*EO4+ZAeU&7C>`)tqk2wQCVGQY(W=OjWnr0)O=DD&vY0zXw;xU+^yzuoF+?q%9h1g2D7>EV@{p@tkK~jYYvNaw^9C;q&FGy zxjf!uI#)ltwyo!+T10%LYkoE7`>ZfK*;mfw9YG3u=V*!={`}ou^SFxUc6^wU_gH&^ zx#hxZwqGw+J!5#{SK`^4A|%^Tc_F0cFl_MXg3)ugQ3V+PyU&_N`X4#0PDjH8Gd!;S zmCR9S-Fon6!Vq@)K8xRa#&a)ygEQOZzxMK}&BC)#su)xen;K=|T%`-W$_HN&53SN~uRMN*?##%jPm z{Q4@l%-+paunULRu-vru*O^tK_9H_N&PjitP}r|AV--FFyaA zI%mT%`Rq+<&s1-I1)2Cq`eb4hp*c9YmX&}R--olFm}K4+w(~IH@^W#V*XnZhn+@;% zw#*_&CmVs8&5XVo>c(Fx>y8g=o~idsHM?kJ`z8~pgt$u8G+p({^obr+S2R2#=p!AO z68ZRP)Y7$q7RG0XZ)*Lpk3#+Vvd+>ErLJO0L1Q znFQ#)5$B(50Eai)M9*_ zwm2aoaJ{C~0to)fwNCu@dd46q&LfDo-;r8LKNAQFe$7vJsw+Fnfbh+}Bgo6%rIXRh z+f^c@CH=kYe&x`I;DpEx5ovDq>*^fO3n_UR6o}7w{=$EfgO}NRF?#k>nQlc594DCzop883m_K*5|Mv)C|0H9{J>}OvV4S8Sb32&%0Lt z3XtmPfA6=R9@cH7Q7r|x($AuQgQw&;sWFH>V?+dPeHG+49^6!J<#IMz4dY;T=J5Rb&g-E& z==|v*`@99`H?K4Lo9(B*_7hP<==x~vnM}Un;WhozjfPAHQAJhar9R&_90I!O#{^dO>7;&Gyw;m$4fy$^@7F@ibzMW$VB!DafU(p$26gr){#D^G~@txn0;xT$7@ z+ZvU$y-|Sw4D8O&zg?KPEuHMfR=GsBHf|DWNu=s?cTRiRr*F;6ka%dX3AalbhaKzk z@5l%s0B%>yenV0e9E|9;%{kATu>Ii9PrpAZzpFOXM#mr=K|%eKy8bpT{CA z-M+guY(73V5r(9dB>z+6)FcLsq;)t}{%m;HfOHUWzG{H9trg`xQ!nT4@AoLnT6n|thz}yovhsgti^ydnRFFFv{vlzdvY0NC&KCbc3VtilaJtJm7qmB8)8XKE&FTvia zOt@#udKIFLGf2!j3EYbd^I7zj^6Wl;$y!(4e*HZ?9HSHWNBoAeCP+W45%ev2IFn7q zx}`L`qg0lRH{E)z>fLQD$hmydkpLorM5)O-<}ykbW{I3+=XNf;C z=eq~Ft7#C=fN61vGp}62jy70YV5HK`l1{Q!x3N@LLq-s6d%+k z9AZ5|u){Y1*$wNYTd_2tJz}U+LgVJw#t4t(2rMG;dl=LM62C?xnt40In9iHMi7q;P zD--^^E0d3@g`cZMxH;2y9>PEsmynE?+4vyrlfa#YpY%+0cMKzh!{Qdx=k~aNFWm8@ zXqBK+pMr(9t3$ZEBzV%oZEjm!5SE< zyV}8;lA4c|+aJGbe{ z2i^WL*4nj50yfzl;v69c)>L+*9d=81b>;cZqp3{^)OPY8?j?rbAB7nG zYtw|a8;G|*3D$a&sP)XJ-AE(UxkA%)R8#x^q3b@Jnt4x4R6zLrVq=qVjkbR!_{dQ(&clJ*>bLPz4 z_nevQx7ndZq70$z3Ax^*&eCNU(P=M6sWpW&K~uf()^Lc4((Q@RjG?@TZy;;QQEK+^ z(Yr(>9QP#d^w4j0n#6!HJyG7$-4Jv)nuYSN1qj81yN&NnE9gxx)xLwl$;IH*QgJ3w zDr^9R`WaNqIZstCB9t$pz_jm^0Hq>NgjyX=fe89Cu3i3>LbQj{CNvlOhRV*6@}5?~ z{XUA!ub_LIAa%L^n*06V9``Hb8+3^g*yQ-+IOubZhz~)%sWrXLHQJfu+S$jwQuI1l zVI2oO$}EluIjw$>OkLjXKjqBbg2Te5O+wqtN^Tt6$hPhF1Eq$rN;Bi#Cl2vT;&$>2fq z&bVI=L+j5CHpV~Sb_Pixaiz(hX*Z+24naz2ka$tJ2%v#9K=F}QU^L>>En!{dfc}Y~ z|I*(xGwJyfm);?3E+nKCO5$lkrNY98ME|)2gcI>c?(o(S@0ST0oef1tMKvO%mLN>A=QkZ z9A!Ug2xhLEAjQhhH;>H9*>SRG1OC=ERhg2qOSl^{6PhaKj^aO`%8bb2b)4UuyJSJs zD?xE=APM4x)US!&V;qhVgf+9!o(DUlAY|{sC?|CPb55LZCq>lJRASb{z_Lj#s3xam z_uBvfM%&!j(L8Ou>{pzQ@pbSsrFWik9Ij+ zG!`e>1LDYvc+F_w6G|0lMwz}vVdFbJA|kE4WTgyT^6?K5A0S|$Ag zDFDDjC0Tm+Evp?ZGsaO;76xgsD957d3~`Gy0bCX=0=Nmx{T15uEA)Cy_?v3^IEFcE zJe3a=;A0roz6(mfi+eIS)0I9#mOaC1lfZffb5)!_MLHkDX6L@`&J738KI|LJY6j(g zdcuyhnlOtHuhXG>W9=sV*ikMjp`%D;#k%Xnn$=o5kN=5G>D;RI#G3k}p_9i=b(Wi* z`oGoZDTM(|oB*&EP9@O#?&iY2ew0R!wS|by{Q11>8D*uT#ER?Ud-a9%fw^B<(|>&* zxyCQlXFY)n1K=pYm3yd0Ot{F@6JUa=1k2<7viVsV%)I_mmw3%FEb8P}EA-Z*Gfr_@ zk0 z`s@1y*0NY1j}x{C9sjgI22N4@;Eb$tv)c+kSHi#fs(2O1Bif)>rgA#JITDxZx|x=(!t_ z5&6fUz^Eyp$1?eGww*`0o#(nk=#54w)u8RVQ?=x#58O_5ljqq4&pnF>(XUUf3LP9% z?OEW%H@6){sn%ECBfZyfiUFXlwcptZW^6yTjm09P28J`tA|!4I+?DH3e6Dk^@^i2i zPBVMRii+BsF*~6^P?}_x!w!fJ}R(Mh>o^1E(C|5*)krD`M5UXzLY{i~eqb)y$Uc<`&!KmYA7gHGEd@ zaSZ4iOcN8K)PqA+hTl7L69pK71$Rg9xR?LBt05eboj3o2m^8iy`sD_inX?S|>h3<) zYp~Q5QWr{o!SY#b1+*0WpZ70626xR`l~&!vsq;h_;~)Dp?Y!*S8C(EmfJW{?K|6Q$ zc5fLdFWuFOaSeCb74nKuC4#04ck7#XPlk5q_jVUwMRYgI^Id|_Z*24dS|#Ts@)RG-ShQ^<-KvRKWS$H7h$S#P#zk9kGmm)Qx?=8TO%F37fcta z(XIne2Y5R3m}h*Iw=u&eI3B3K?0fJS?0~W)w}b+?NfZxAL@IYaM-rDmcIQ^aJ)XXY z7%I#clIAl$QDIOTB~Cm?U-ghAyn#jfgXo347Ne+M-}eFw)iFAGt-Qi9+eEM+%w!(N zkZA$YUe(}70bp}EuwCSlT*{IBWP}M%@52D_qa5MhqR83}h zJkSbeh!0+=(O#7uLEV46Dx+Tg`4a*OVCNHZBjomO9$i#2`#e00)ckX9_$OdTTcXCumYECcGOa?JR2FXG(2bIG7>lvo7%Yt5dr=f<^GIK=S5)7IygG$9*-=baK@5 zx%89AZ>nm%MMhTl{p9IO^M-f69{b71rRd&QtkiqTZBga$Q?-ddMRc}tcdo&Dx^OUE z#+U!qUxt923$=*T;+>1{BsJLx&TY4gifYHsn}g1G`0JNbhu%GS(!${*9t z`fN7ki>55Z-96bVFH18{;2eYP&VbEI26wN0avm+$&HKygwKGsf%=-Fz?0dz`nv(s) zThW)R-h)h8{p2`!Ga_?qqTlb2_ROq4iF%*-P5zWgh5Z^;U{YIjLcr3UXVKS!puPTM zAr#Xnq##9o6;(JuBZe&kfvsF(LJ*U>xk9x!ZJm%J!iH>!0+r|}4!^VISZ`kEb^xUe zA2K$+%Gv(mkfk#|j&-zXB_4fComb_hA^^fkMfIfs@4<85E(wbq?F!O0fBnu!k>S23 zj~6npi%r|==DJ?6!(RfQWOl#`o0hTsdWK+CRSvUXKyFNW03Y>6HjmLp!o1b!SjB9W@9q>Z9gD8vF*?!At zat508k%daofjzd{96psvMP(lqi4ThI_!5THg5w=Uii^vqMef_TvXzt`{S0(1Z|ob{ zt|0V{x60iTZ0i9zLa%q#*~aijvFj@vma}fwzS&cc?sz%TbUgGjhAW!nrNtFFkkrj# zYC*5T^+_Pgy;j71R@^MDShu`>AHG)JNa}mE-}z^~(euZPHF_+|%j^G!TcIH}ad!}6 z{PEH;dWQrqibr9QjOH;C$a=XAw-LVAV%1tr_qZ93i3fp?dPZLF0FSseQy2ymHWUeZ zy*Khpy>(-*%a;#G&5C&pVSd=s^DBegYl%+mgFPdh{&Uij8~62PwDuqKAoXgRs$+%T zYmp^i3%p$$?^}vs_K;hq=sZM=R?%Jci8gD2PC&l~%97-L@1YFt^y^&`ZZ*#FH3Wlp z5C8ry{QQbulA>zw98}Q_zu4&+f1)7OQ^Xw~AM%AI4p*rNx;n&Z)fPN9&0eWUls zf~0>1@DCp!j*ktWG>05tC?3U0!L<<U{xex4wE&OM1jbS{7^;^;LGO%?!smd2P&AkVjfA3k0D>_&RI)uP6ajbu zV+4T0l0ZS3fT8^Nj_wa-8{DrH%X8slbGfwvcVEJ>%XLI6`N=hN?qR_l|3b?+7S(6C)R0Rr}j<|314OD=jlRGYWJwXjoemic`>}!y}trXf_@TK`|?^#ZUvYgfEG*Tg*`3J;hq%M6cmwot!)_2 z5a09Zmj9Q2EknEO$pIaLT)9v?5x&$KmM(=HAmFBlf+6bG{ir##-QC=k<5d_8WMpVMLE$bL<*L( z(op%czsL|}9xOu6N~Mhi@DV&Nw2zsTp$EP;XE}I3QrC4~P$CAM*@JvOrE2`3z*Ght zj#P(r@Qmj~C3V>%3+sfWP5X?aDFGaM9XiIx$P~!^LX~T}B~`~}G3JzNjtCNjg}*f> zwo3)(0Kh%@qZ8F~{+=!*SH$e77%WcNhAaU@$wwK51^fNvv4M&5ubCCavWPx$4|dgE zOOMmMuUdR45$~)%d&Ba#nH0DF4v1xYs+)llfU)ipicja2Bs(~W1rkBU2D#G^Z2Oz+ zGG`kN-N#Zpz7rajwrXx= z*IT~*N%lc>H=QHr?Yiy&bD!9<3xMkJiyt8g2T|`i)^A)puQO7XdLI(>i_vMi%@Z!2 z3en$C{rYF<+4^x_#r7u6h5WeW9`;k#KL_utswADN8ztp?hGjXPCAq^XLa5MZk>r#021i#;~ zcGL64fa}5yPb99)sx{8*IG^euv`eD0K3cIu1@2Y!vTej{O6+m#*+aQ2#)TY6yc9nR z0zJKL&u}r#-+n3)6D^r-+ZXav8}F!YE1b-R}%7OSA+4 z-3)Vn1tU52+w~PF33XHYN!C90$xvbLK-3_OppR4 z3uGIAU0fHNQin{ih!HRZNnjYGjovi!@_^|9ix?Q05cO0`(;W(%=85ZYj_L}CYfn|w z5P(f$VH6UG0|H$(KT^$^)$5wfVY||d9D)+4I^q+;*&4NoN^q)^?+Hlgih*rG?_pCT z&Fz5Q02mo?ZzKm+CJ;YR8O9!y_-`ulUtOZTv<8SC(T=1W$Wfo~P^9TqoT|hrq$Zv_ zCyo+eYlcbtBn0IW(>_t@00r;-isxZ{OY>T5fS6$K4J2aWFlx*XfPXWV?)J^7W)iK` z3p(lPx3c7C?{2@KI}^q;Xd#GG|G7Z_4kE4h^JKjUty+5pn`64KQ zW%$zG5Aa9nIlQ%d3mgF4!-rB&Md;^W*U1Oyx%}02qt^TJoH~0=_6w436kz=e5H+&~ zO<#g$QKr25n}U7u!MPohCxl7iNnRwuDe-iQe`RqfXhbD6YDwzdUxN=X()RD)Nx{;5 zCeqXZ;_Tl7`*}X#3D5*X{Wzpf@{;cNj`Z^B^z3A6i&mQFtTO)3U^A4|BONK(;~81U z8QBij?-*}iwEdTlrI?^?2Eg}2fy4O3SNzi4k!jqRagJ(rCpr~U*Q6slkR-ba@Lu=JWvQ9XOT&1Z0h zIb zs?Rmtg2d)P6G*uUOHXb}-4GxqO=G3?oJ^<%U)bF(blCDJ%Dep;AiDtoj1mzOD2O8g zY(sn~_ujD67i@+y_QOE3D`}?yIY7q0@Cc?#`9%_yZ4R3s)*s^Q@9zY$Cx?PVNRR{~ zbqoHJUG=Ao)lZgyLa2ISrbyw_SdnOa;RXQi7W=^4kP4^r**DP8zc}cXAtZZ=c8UaN zdkb-N>&>s=N-gjASLI4Rs_P@(Fi_7=U%eS_u>_}8i=S778 zx_Q>p_Pg-Tz*2j|Y^rai5HrIZnzB)yvI!BYslHoG-^=EUA&YF~?x^x#eM-B<93oWpbIK6d)b)Cd+Nr@85+*Ht+FL@Ab5q1ZjB9a zfT?0@EZ=FUQk|``K!X!SXwF3Pbofe%DER&53P4G<@N6}N_G=#o;>q(BM#~w1_3-C; z!Ju2Exczl6y-Kye>gcZ2X?hJ?u9|7hJ*if9YFl7E1eSz_u5EZ2XBci8SC;UU&HBN; zoD6O7Wsek?4wD{X-h93GhHjdKU?hIs5TSBpuW`C4-#nyx{imo&u$s9FeoCs9FJdTI zMn3P4$~6UB=6rwo(lGd#BXf#Z*&;8k^ZOfclEQZB610IDgCjmx@-D~a5VN) z=mnrgp^j0O)(8cpMgh*MDA%7u@`*Ohvvt!X!1HY-Fqjo*TN}CU;DMsbxS*=}T@PMH zTmZmLhIF0i=I|W&HXc6uy_t2Sd4X2#63{}dOz+if@pEa}di zKoAcFH6HK`3C_=f#=iX8GW&z900+iZeb9C7c~q0x-$cEPAXRqg>cM6WVFpO}-_X_> zB7OcNdJaW4Dla7g7uYaf*9i+2h%;PwrAlPGUF8ZF-D#oh26CYOr@zQU!GQ;9sY4yW??XtuC~TedW5N(3F4n=U+X0OOi;)`%g9=DJ#b(_mnEn#ok&ChysgQ$M#_7#dXx@Pw6 zoPx3cy7&Oym6=reM94ReE^B;$*j&;Cp}$3ug^(HWhAy}>4mn#1>lJI&RqXv8icoOw z_D5&S^6a;<7F(j^smBkEs(skcNg0V#CZd`u|$M7m@xY@1ehZ~`F zPPk(ZnNdmInL|6K3mZ42+$3WVm6d92LG;kUUap;9I`O_@qb_=-k)p>VJGB5>0uUTJ z1VRI!8B%$pp#Gs4+gW`+{n3-h1lg#jGsS`PIphTo6aYhXqG8LS@GCt06r(iu52=vF zF}2fqZGsqC!C#Hw>LxO3sLwfpHYsmI^)O*e%hXJKgIBLD=8 zf<*Ih)A!eX!u*WM`uV4{=gBSubQy6;3j2qj`^Pgq*$D5=QOmxh=g^CvH=6N@pE0;I z9x&MDdzaRUj0DG4g5z_bUrFHaN{|Nksphi&L%rb}?!$571ERsmc{F^Ohi(m{^aF_> z3H`5=0!Tmz@|M3H%sEV~m#iN`m-dMCL@M z*!O04v!>9rE03~9+na~MxiC3O4ci*{2x(CqzI5AQ3C_C;eKPswa?oXN2L8&VS6%Ab zD?Wchz*$@4mirq2?y~l)QQd`~{8wvpSeWsi`WbQYVZyJu*=9uRFSHVsYV*>p_>eET z*fa#S9D;^KVVULvZ>tY2Ea1LTaBPs2HV!YRsFtT^urOr@#98RdL#}y*BM> z@5zs&W4~2iQN`Zfc1Dhr^1zmYU?HZywQU}*OH?{XMO&=-fAS#dMiz&y%lF)V<9RE3 zX(VKvZfRsfvbNVKmw(S8x8zN?K8nv5acy$D!QAT7oeXop=Tub&+p#X&Z}YnvUIjF_ zcm$6~I97f03xzU4k%Pf`_hlRGn^R%=%>BVpi0jr#Aw+v*=hnGGu(R z4ED`^PAFjhc(5)1Kqp_nl0bJ%fCEUd87y5Q9ukd$Y}6laOy6O}-98VWwI{bepF96P zO@cEKl~xImHOpIe$U908id4CeS0w{A8X*;-9br|Zd>B1Eu^|b$ow!xRSdq;Za>DMJ z8S-D}o_2Hz#`oaREBdf==umu7;&z3EBsWC50&*wu=m?;$_@HQA^;j+Bj~_RcRDP58 zFhR!-*p><#?SOfzCm{5nAm3B=3&EEfe zHF~GM^Z7v=%>ef32mc?%{i7zGiU^uJl{Uw3hNLS9cjJTjLqEdSdGyc!R$WUt^UFC) z+&@d;{u_e&TZx3F?7R06=(MD+f8@UGU_IxQI=4jLt{~8TWh)XVM~JNQ?UpSbRUs}y zDj|_e&~IeGWBL3i6;I!n#mo%AY3#HS= zW~C3y0<~WX?Y0``smF7vr&@HT9&W<=+!9hXo4rBdA1DL8#`pT~|Eky#OoQxw=<> z_R?PZKRjB0r+fLg@j{+k&V8;%DIf(liwoJ8^ylicyyI4IOx~Q29`)O=2K|pN z(AvtG5+y!szdo3znXTv{BE2hY4XW>%9ih?XhNa|yo(sQ zD!a2GQ`&vCzS$9yeZsp*pEJE%ScTZ06$WBjb1CSM>=NAqCVESUq9Q;3QDk}v`$p+F z2?vqnT?`cVJI1nC2P8I{`N(d>3|~dvw;XB+b!oDZV82%!t7h7ZB_nO#$n`Ksu`<(} z6br50+EubTnbb@Pqnff0 zckJT~S}+yreX+^vJ8rzJAY=9{!|Hvb&s&}SFxT2&wX+^o_3L3{>EUZYEuNcebI(W2 zm>D!VUav89@dXe}+^*ABS{gG;{mgtuHghpktA7!F&;8r$r$w#=#{k=C);r>l3i}+H z-`*Q)p0bPMwH0{Vwb?1e+4;dibMbR_@NKKmhqwB&g?ha<3PNV8s-I?fc1`QNOC2ph zYH}^{Z>ojgy6>EiI8aMc{)b_MNf-C1Cgpy}Z6Fu<#Z-&C1`^%khLbNH<=q5q64GSS zzlna%4)lK7nAylA8t7dAH?`5@9@IAdnZi!b3h;|~S$a`P^yti;a@M06rT3jrGY@pn zc?uHb)^eJ>`)?-lC*0i<-lCC?;)}qW@w~r*^ZBP;$es82qSW^RuaxCt?&Gf&UbE~E z*AnjB_|f!dKz4r)E)pZDUNrjPLGX$HWAaYRR7O_T>Fj4jVovW}ULxek=3Ge}^^?eq0?-dUF5#)6O08>U}Ro)`=lTl&Md`Gdv}Pj@45u+5gRLh!el-b0{ z8*_`$exo+#32c~z!!o6Nof=>T8zsJGiyYoiW1|^~A-7?SVxS%Tcj(Qucd}gz>03A? zhhIOWyw3J7RQ*==zK=Df{b~z~hUABlIH&veY(MKX?$nONdq&uE{F2kWJ28^r-(k;n zQm1JOE5G?QE&eE*A${_%tMXXiF!VgBAXh-!J+V~t6+pfh?LCsHm%H|q@^{8Mtkb3k zy{34*S4&_T^w#B_!ws`(b!cdjFc1m=INa>F7#jWXof42JVw!BJ$^-fu;fTWh&84Bl zfCbS{!O4FCj<>$5w$T9+Op&>eCCL;Z`3#U)Q?KuF2>md>lczLe#I=3DJM-N$_8~rM z)`@?sQUF8af)Bif`5o=qe>yfmsV7ZF>YH3AJeqGirr}|MI`1=C1)Lh7>9{+l?$Yh7 z04w{1x64khE^WffNAC=EUTJ`fUot2oWj^~Z<>!{Hs91_SWV^k3?TKCEF%C=0#5dzVA7uqzzrOQ&BZXgNFcs%hfZeoo?JkZ5rm z<7p1Sx%(YYwS}8=yV8$ASLXRUb9!vDwMRhBH0gf$&_eB}HABt2_%OPkl$&g4OH{IR z^k2JOQX1#og}hs7#eGO$a{=gzz@*kkJ{__xdp+F_yDt~D1HUUdvRzAUX`Oz1DNES? zwJW>u5_A4E<;|=kR@e3d1GQIcyM86+k@yKtXW!1jw$3Ckux%?AF4qy%yrUq{w0f{*GoM+f=XeNi~Rb z>nZ!iw5I9lM<@MVJ%yK}t*6gxo>}kcFQyI?Vc+k6w%%L5I-ezKeEEF>-cV=?Ujv2#aDbW1VK(w`S6FDUKC}AAZ+3U#-SR z%K-on&j_pwUQ{6*tNVU@tz-dq-|u6Cf_3p9AC_y!_Gc`k?!9>X|K;0z{(ty(9Se)9 zl`wS*u=xMY(5S_(X^rqvLT_kd(DI|JttlvAO{TU_ejMbmwLDmtn1yXosmrBv8WWwtySqScIVW%oe42n5+yH9DEVq->zbChVBT)MH+B-v4`E23i zv?GFCKclf0i|xN6KHe}oNiCsP=dW>gbII+KvTiHRQyJKX0t4s`RVqH7sMXPJW}PfX&?L z6Wjl>$h$cMa>!v(>@^L`5Vr5@vD@Vh=V8L!K{4S2_bwGN@8w_a7)5auxumEwJr0t1 zeS1ku4K00E${#Q77iO~c{rS~csd{nsmPl#!go|j!&S~s+MJxQ#PF3$Mg*~s~SG?li zChSVYzaNM}_6VMBG9LA1L#+;+zD>)dcBCeUL@t5m0-@BUj7ea8iGAVeGOVo7d#me7 zXq<|)kZ$`xpSX$m_kMHP5)p;6 z!;yj^YajBZLuY%G`v`Xk-rCPxTi086#`XK9-?)$FURk1_8txBoAnjIk$ClrMe7ioy z!mwi3t0VNXXM~EVxl_%;xmR}M0fE`R=wgS(Ml9%fN(n;Z0Fa^rBbGJ*H46d2gaoU(<7IYP`|(FHs~lk#niPF1VU@?L05*J- zGI1H^C%kqYVcDr(UX^_OQh<8whPIB?TdvfBf@dyW+ERI5FI*^}Ud{=N$+jM9-!vv=6| z{94Y`^(BoNVwB(^{RdDEnvHAez5;R&c5PQsL3hjGmdkkR(N2-(EQRrd*z4&6?{8ch zr1!bfn0{h=rpBFD34k}5hQSNexOXsP_hNWIzn)tHCA&n$G5F-GaBgYtKQj&1F!`cq z=6GWMonBI!j|PPe1+k#R#Pup+zMDAq1w+X5t?@j?&Y}myG~>F(KgHDtH?$yW7GhmL zzu0PVTFZn-iM#RAyg_vcZxUnfR!Ubo%c*hGu}^&Hp@EG`i%PHbj&ze5-j}^q7jf(j zoDO|z?u+QjF5Ak9(a_4Jf9{Oq)?<%uEBE_`F>}&wAps;qk*U?6+^yUjKlE^!B{xQH z-z=FnU39rqSAK-!yq*h{iCdv|*u?R%bfJFE$Tl#7d)2E_t!wUnXhgK_`v2*jJ-GVX z}OeZhgHRh0w zqQq=;hQY{S3-{rIfV%YMlk9Zqeb;Lxy7d7CnnJ=@I%8J7ezTB2DqVTu!{pG_Ty~v7 zn9h6e_24%^Nr(p@4%q-D_h^if2*Vl4lpj#zmzP zW5;}fjEIq0M|DS!+YVjtv7nNYHq3Y2Tt+$pK)iH=yB;qaXCIi4-V8s0JEI9)I~dLQ ziQ{;`|1vZQ-VaK^LkHcWb- z`4$V^DnzrePxqZlRAvrBoOLs11}ba0s+)IZh^=djWP|$LcJXKv9uUX0oPWdQmQfBE#E&#)Zr*WM#fh^6S`Ttu1kW1WUk&3LF5+K!=;s3D!r0CV6`)3s8)&&L3S zmU~`q+kEet__h{v|6drQSL8ll4rko6N1UwDTK7LDcSJV$EcXTsH!d8t6v!(AScAd< z@@K~R?o~H8Vm|&^{25kkkg>X<5>45C!}EFG&CA2bcIRV4wr{FYb6b(29bDrrC*>In z+n*r{{4~okigJ9rt|4)=WmkZDR~2p5xyHSdg+CVe*MZ?S3HEfJM;+`xIGJpNKHWvZ z&8JuXR>Xis?UHD=Tvg8}ozzdiCtg}~v@tF2cam=xaK*dw;#jJ3o{3HjU!69RFLy)K zMV${qpI}yj8$F@w2_v+BU$$M(!gId|&`_QMJ_BeF02)YYC~Z+F)H(Ejz3w#<$0^}P zbu=B32ev>A<2noDVF~A(3PV)F2it=FgenpZ*;|)r8%g0P7M!FoPMn1fzs%xD4|Dx0 ze#aTdMo-5Cj6m>2+;{fHAF!A|{N6{?j-^HtdbOd6!#3rqIK%_-#WD7A|&6&Oz z0AItioEmb@*EPy?oJ?%Q3y9JPAoKN6b)ZP!a8#3`cSa+y4RUgud~lyXQtXu6A`Mp z*9)CvC-B#93cyRIVh=H}op$9>JZ}9VM|~Ib?AtdDb#I!q7*|8DA1p<^fr{)Ly^*6w z_$I~m)W!8p#r0@0GVvnvj1X^QSdee|PaNo7edTz)7-Z-Hz#Mu?G=L3AkE91M0ur|B z61LW>)m?zDNnrM-*@O|^xC%ct3-H*Eq!ei;%o`->=l*J6VGiM^qVPq zJnX-+_vtB|rdoV-;yv%5_oL?S3G^R$zl*nMAzEYD$QXoM?+2ly4~owa$D+&R zz_OahTSJG|ZsnZO(It#t$Ke+Wx2n?^FDso`cNZv(5_z567 z`8Y@TJZBvP4@&X?>A+7b^W5HgK+}El=m90uInqgaoea54=v<}-cNQ0CHa1s|bTHhW zW)Yd^eLwGpPMTt0{!V3{cpv;WTmCvKf5$Li&OcwiIA7sB2Z4mO=fDz{phS?}pRfL* z>R>H4h)(){vxL7VRL5FBVtX=gIdEp>ecC$(k`3XzD9#-cOuFxrTy#DrmQk2|x3HZ+ zm$C#Uw9u6LS#h>E6i?WehQ*flZ6yzGm3jlqzPdoZxs)8+mo>6Mrgc8I_LW^~moDD* z>`rIuZ7BWaQf?VmK3ZHpKBKi4TQ)-~IczBV*igLS!nPz*bbhzoX{zEXkP)Czwlh;+ ze_nnt^QFzI0``avAzJ?L{1fA&Ys{jRtgcjG+6qWXS@ulX%dN^YwsLy%qiP{`=9`V> zJ2Tas3Kg{5s=T^v{4c+*xm3%~GAJlix~+Y^olz-4TP-D8Ekj#HuB_0^sIjQ35#6qV zZdVgE=!{dyqC$wL=MQhbAnABg^$~D$XkJKtPR#9DiLyLZe z8GpA>sB(GpoeNkpfG<~cE%BuN=0#iVv&|I#rWT-46?H*BSxGmO15F66(&;o7i!3oC zl^2G>UX@gaJ}NcSt+sqw&!1F3LxKijAj^XKIkQxGx($W2-)(8@O>}EYTx%k>8`yY- z*t}q?B)A`{(UV8D;-rZvDr#MLF95AF`UsLup!;D3G!m%|d-QdW?DuUy%AU+~4U=9f zW81_O*3_3_H>f~0!v4b<)f}k-napUOoo&uJZ*~-ISz&L*n0hbs|Ti) zrOB2pPMd>+(9uXw?&W}TAO!wFZp5_>m$Y4b3}Gy7X8zac!q`q0HBCGLJkV~E{nyG@ z+AcUpP|j?N(`XhI>k!xLxJ`%AQf!~=?2w^j2)&8;G2>R(fl_PedY*IdY!+cXU~Qw= z0(5J!AE0*(>Vn%tTn4&Wwz_6$yEPO|edq|bVuTLSXg0D@7mY z%prKWJ&G*tsCDd4Dd}Nq?MaC1s$uATsMpM(*Bg!LO;zmS`qpdkyEm#oKX#<&7f>QG zxxb;QpYJ)?nJA>cLam+Jn+WcAa`t?knfL7<;`={T)4zd+oqj?oq%Dqa06m~d*CMIc zE$8PM(4Q~I`+3)&x(rYK8B4PsH?)yCw9_>78B6_pC-?_ua3VbZN$H?AN56hk?UES6 z``j;a1V^Dnq!cGUn!-gs9A=Fh#?WB^mBUv3`K<>f-@Mp6H?Of=Cb2e;uvsUeESbw% znPP0?*;Gc6^YnSk9&R_;Se6j%Vr;X;KfD5>-N*;Tm}N;(8X@~}F$VW8vLSy!?kC^7 z2ER_rt)BqusL`n<=0uA^t3P+@1 zt(=ORpGvr#iff+aCQb&s)3adltVqCZ-pLToYuT665U<-vLos<8k;34a!sfhUcep8Q z-z5OJ%>zF(WZE`-?PiZqe9e;7$h4>NdXfm!^O79o?pr)!I18cnQ`N1SBT4JOzRXLP zWu{yCOa`7TB)YMj8>5wBV<3@F4YTL-2pzF5;TC#a)r!0}{OSlUdbboN$tm>;eua*| za|>>0wS1j*-A*}a#CrV~38?afJ2>f+TD-Pa`RDt-qxOm{M0(Dg>Y0G6^@O~cZ7O)j zAs{`8F2brv_wIU>!TMA3!s6;ulx#SE>2BkMM2+6U8j}PIG6!zpwXySbCOX!~-3}gL z8?|D$8g7st@oKXu8}@qPrXc$5We!iF@q@RwHe@nAD2emic#o*qjpDKm3lI1bfq4VK zyiG!E5fSSIm{RgqbopviJ8TcRoxigEL*aL_$97%$wn^W1BG>Ppp}e5w4KiT1YXPxO zNWPoB$K)}-0{~0{c1@pfj%6=mhnOZ7)@qe~H@jE}SCUl=JjpdTc92WNWyBU*Oz+F~ zRtst8312YlwtiMRTyi_PoZciB*ozl`7RLW#>LBvq=L9|Jqa^9nkOWyoP?LA(S`J@a zO!NKgOh zr@?l|%L9{W#E*p|9#1y(gF}9dD`Xh1k$9~AV^8DqP$Ds&9&zIJievoCc>lwx zKxo`zdug+?v-BqdepV!Y7C}ZtCFZ#&oqfp-nhKVF76wcfJ}CII@kxSn??%xc6_6hm zo&NL@0sz2yA`EUX2rZn6w+1YsS`5>eF2_;^o_*%}R!n+&(O_L+tEULOx_BjT`sWsN z+oISr@&VAiq~gzx2co&uUFEXnBR}rm(bSL2fZ5B%kZ;S2@YM%bt}2SI&qRKckGDOK zbapQRBYTU45DM!5DX_Le$N;hb7Fh2=LA%W_9}*We60Ql|2|^mLX}uNvKLyq>?mkuy z=-BE&6vAmC!s1}DGj3PTuj31HO(&0kTjt+G(=VeRZX^2WXsVu?7K3~b$Ir%W%%ZeY zBu~!GN_z_>bG>(6uI&z%uzbCGoFVNt5`aiAsEK~Q^W?4L$1Z(2#>(zmrEebA3dy(o z!rm}n2nO6KUutEb{Q-_OuhQ|JRIRn&u&B}Lkr2IB5EQuV_|ZyeSU+t)tajuZ+2q5m zBCB$%d1_1USJva!$}_d`e=avhOZUc`5m$a$EO&JL_hPxb6CPDRUl^iRjTUWow)YU2 zw5^C=wyF=gock)}w3qYp{A72sJI09T*V$ec1k=s^asU|3>9?M?k#RQgrJ{JyBLmJI zL}b@*PC1vrXL+_f`>aNss=s~z{O~2@2H>2vC)1PexKf?jCW2lA02UykDH(SXl_8%W z72l;%=jTcH-I*q3xW(RQ=mw`aNVfP40U;UulnNTb_~ zFca_#VM`K{jxqkqzoIp(QoUoSo8QsL5+%^}ZreazYF=Q>+BY`t*F zyh4I8->H(70T8rvpvnFe`inTh;`Pmmlkr1vXsS*23yo}xN$ne~z701&a}Dm4r2xF7 z=O-;gfJ|qYtZNz;>?W<9Q5gx`sdX!Fz71HcGRvO^SOyVn@hze^YrDz!>}-{p0A3AM zPT4jv2brBiE;&6>JZjExfq0b`%|++)cn3wzZ3y|;aA0Z9={q{|QY)6;2kC0>xPESZ zXPWwACaf=EsJSN6x^ef;Luu3BpBR;xLB6VO8$vGyDEC}i#UlX}bSUBFFa{kcPbi~R z#ao6C!U_=CllDmhFMl%25Q~)5DE}G|p(}k?ZPOYo_&o-;FUeG|0 zo*wx7dXwv85Nj`*N|H-1&YsSmrGHZ$o|x4BWXCXCX=fQG-V31Wz&T;*!Y=~ukIg8cEk z<94L{fURUWJs#5`8t9k+Atv#_j0Yi2hp?nn7w!LxtG^73!i(Cz;R%=_XNK+uK}xzo znxT{uK|)GEI;0T>28MWqpGI0rO^u7KtM^#lzD=qq-(%`jkvX0U^#baVLb7K`ox&}yLC#GT1r!ZSLTMP@G(9lU&%Xh`{ za1g|*{2$aKDr}9b%$fwvTZSKfjDA%UNm*@qi8pqxnTh-t>~0+<`M7xc6)NccKH654?{k9l;wG$7(VkGk;V+yG>lEFA8YDq+nd=Mp)~m=&R7Z zdC&RuE#FsVH#yM>+uI9SO%!Q2lK2XH*_T^=GH?F)*;C}n%;G{XeT&3^=bopHF8_AP zY}*9!2lsbj^ZV@5jZJLt19j$Xme1%qC+S23?z`J=lrXdfJo@lyK7UcV^Knfco43b^ z(Bm8Dxw7By=OR8IJe36GG;9{$i_%wr{MYN34%v}sBopfrfaZ1AoXCmkjF22fM^MlD z%UTpq=RAT(E?YE0RbH6b^w}yMr7Hl;G>^yJtB9iXwDUn9_{8hHcwxs$ymo(=pav)9 z;r|3}7K?TDaAZBmp;Wm8#ypzRa1(i!k*`EemTB$-?gmK4bA8ik?WS~|5NGWI z#IfD9Pn zx`9+(C3zH>KjYrAEw5tkDNRp5JMn%17mricDIGtkfa3Z_9Uk!i;&}e=MTL%yTbu&z z%<&`X>ZjNH7A-EbjqjyL?F_c9!kDI05z>{0hAuL1pOjR4N!LC5>hy%a!*+x|;oTcH zPuu&AWB9J-+RS24M}L-uW%i0z24#%6^%|KIUY5yk>5;<4_Z7m`FE(zyNrDFrVD>w^ zWGCfaFP*>LFxOgHrmEW5dq2Y{0RzFGi9dXc=IH!E`nZ*A=Xy`;{?fADygFSM8KmaJ z(v~Q^I-hjRg(SF>V?f1w5UOzYH0v*!O)H4;4*QGjhgVxVX>Bad>?!Vyo%6CmL1O_C zeNhPf$PmtnW8WuM(5U;s_Ml{=zM(yzyRXT4s4a^FQo8q+5@rE-f@vm@7Y{O{0VrV{v7MJZ(2L@( zWz^?u7E%x%1-&jL@5NJ0t5PTxl7|;j1*4`yil8J#vkOIJwJ5p@0G(4e-3vAHS`;iE zMbZk`=K|1Uf%E`1Y9heVX*WkFJ4=>orV%vK0*Yjchw9iw?2r$G(Q za8r|P3gT43`M3k)Y6q3d26aU$urWJ);K9j#y@EuR=2*H|b3>LvUt2pgjnk5;rG(y)j@(IWM5 zF)uU+knu5EV&xVmz5EWZbWuc2riucC-oW$r45(T`_H4N4bctI3s-I(_mtYZ{&-=gtxFw|3EDEy~V)QwIQ?I03I6sS3< zj~p)3M8-PBXvPy8y-f-5QI6Ln(hMF> zd1K^*NF$>A@gr?nBkjrhZDvG@gCm`d;Ey9C=2(c^7gVsotN63_h=ZzLYqZKPSj%;E z;Kk^m)2IqgHZd{Q&Iv+Dhd5hEhaZ4&4J{M2hGF2ZlZ~<;-N8z(v3A%mfa?lx>yfY} z9mlh~jrCu@_Zl{?qkn9oWq0*IHh>8a0nlz<10OL1K~;l3YjA_Z|6pj8@aW+z^azpG z*DQ!i4tSgh{B3jWs1x+@rf#NJYfeh{Dh#DTaz{l~NY-)u%IV>b(*xX1M>X{W^1QN% zR z=SoB=p@gGg?wCR=;x`Iyip&bDRY{QA%S3N#yr_fu9?Bo+m|?VRFgVnIia)?aR#Ip;W=xXgYqg&74J|$W3|g8qMv(@5 zhpkCgqXpPBijp_-+C&qGJ;4#@laW>r6fDr=n+kWn#@KN}oCL9uqIQz^7Z0rd(c@Gj@1(|*QrYt?q>p62jc1D2BSiQ{TN!de>iKz+H19gir1(cbs z7FsU_t^W*ZuVthnfM(O5b?$rY5^Mgv%-mfI0Ll%BN{cXt(8o7?j)3 z5F!6qVdBx9XLE#=KZ1$Ta|Sb*K(RRKueIUCmhvynjc)}-CZZ#r89H+Uyrgu^`R1eP zro9ViVh&{=a?Yg-Vm?87=Q|iBE=DDT!Rg3 z!3q`?gnMF3@)l1hn8&>BmS$2WU*A-JERi+Uni;v({`g~QjPLs|vdNEU3Q^?5{H$?+ z-8jnUL|QoXy=OM2bYR(y|Dncwl18z#MzNh(MSceh<$z^!{<}hZ&UG^qf(Y6>nlYV5P54aQoTy3EsjIjdo z(JDBwSk>Ik*Ol9*fq#!;K2(7LuvMVwDp*E$kJxJLncYFaI5gGn=+gFt?x_MEOsTRW z!-(Q#2f&aHFjbU*QH-`6SRrTS&m;TWtrZWH6=U`qXX%=$v>g$$_qk!4f{aN3kb0p<; zWV^Cv4_uY9-H?Gfkh8CIXFJMT0icX1B3B10EK1}W&3_G+9YjmuDukTZ)_9#nRh)oV z)y%e=*X~aHo_1Fk_A-!lvaVH%u}#@{5YTAl&NfN_@$Bw3xKM9P)%F=f;MU^@n_^w| zU%%R0VVtb9H#v${^gv?aXkpn=)8fH<>R&nYz`y|!Afan+A%yg5z}~%i@gQgCU{R-}qMh1|t(1R@ z;UAV?rfy=dcG8cwvgz8qL$v%{=V&i& z+XYwdgc98|2t-%cgI@hZtIUHWea2(++|-=65|11pJ^Pl;Dy;uX%=;Rw0voJh2Yo5u zy_fA|=c8$~52WWox}FbPvUhXKck_ze303Ze+x-|Y5Al|}>~&Rbl7POC`tu5aKY*(B z5Uhl&$W+-AlSYfzMk`;JRFfI^i@r0r!|>?h1VUo&BU-BsmkErXZ3SIBSi3uVAfCkG zpI?iddWk(JXDzi*W8bt7yG7Ar0$>$Yv}C+zW%jOoEwNLrC-E1E-WSQ`qeH^q!_a1T z(q4#>^>e{%v>2!9$85*YhL*oIF*_XqPapVy0a*z-HS2XaO%I+H1&hCO6h|Ft3J~Fy z>_~Q@ra=Bz5om?rZ?e+Wrm9O9sh=;MF&IU6VOJCOqT|pG&`rPhZ5LF9j}lZ!tEE z5a*NSGgd#R2_i`Xw}|SQxRKvG@zG zPao4f6}Sjy2#$hdqM-dT53el;iLC(`j}zvC(-0p|ylM2@ucU$NGKyqkiWFv3RsY11)HzEBh8sn7no=-c{y5 z@-n7Ksvb7Az<$rkGU$$xUGKQc$seRk>N z%Iuc_2ClBDcj6j@E47`iDx z>UM~l{D5yF2^XvX$-vL(dpj246n;s=ElC8Ic&oG*^3%snF9fucS!6frFfG>~x^8cN zi4W{p{0skcNjY@dA9f=my<&CEXYnS2_{sFj3wXsZqc9?w_~|IxyA_;hdh-|Tp;Vwj zMC?fbBQ1vJc@AgXlazB5d)&3}blh|u%|j>3gf(hLJK<>_^OIL{{a8n{X4)jDJuXK> zUxODv;uWj3ozm%>vS~9cQv64G5)EQ7309Y+$r}_RSn#gQWTP61b{}lby@%nwQthvoqV(4sU#{8*@u%ac6&i z@T5&8rpn2EIts(|S(_XZLp~WgnWOh3F{@dh?_i=n<1z~Jaq0|t2%hyC@H9HOv*oF= z3fmTL+UH;ovj%fOvzm zi5CUQkA`)DxEbON)dr(Gl($rv>98z&`;gX8-2>{~*B2-$PL7f|7N?q1JhN-Qp@8Bi z5oMSjJl6hhg}H3}CHRT|q4YdOIw9q10rm06)su1*^(8oxbIgblm$84Xu`cYg$DXa? z)uxeN9Y`irSKz0&lc$8x-+i5__h7GLASiUdrse3*LGAafiJy%FULA&c0duYrHw@dqUd0wl5x}>uIOv2-QxMguJ zbk|dnJ>F20gez!QoJ51A^gf*Eput<_S(ZF4lG8VbgB2nU-LvbCxDI1uzc7lE$r9pr zGt$r)w8O;i)~I@-Qy{X}dGmWwQLrp}BECTcS#*LY8vIm`M3Zl}X(LuJoRr+=*O$QF z2|h^=8%pn1_3K8l07VaW6#^zcEr8aS?TUn5%Wqemk&KFCUtpaoGnncQXfolWWWP$H zak8gJ58WCvPGOoWSzMck`QM3-R!w|6XZsk(jh(u56)??>#DV4%IiT>zUR$clH$mks z7roU^n#06ol5Gu!=1aPcIK6m^JH=ZVbU{JGt2FidM675YKJA4Z58EC-PMMQ*5c;<^ zCClK4;%d6CrfofGvgUw7iQMr02QevS{-pXLSI4vPOUfF=iG%>CRWRCV9U!VfI-*TQqq@rbw_A@Orj(EsQ#W8=eCJe z_cl(vu#%I1I3>$zb4;Re%aTo+ry<&_^`4oV$nD^;v(Pzr*rJ1 z4&Tu2cXlnC2b&EevfFwqZDZZ`z9`HVC zp~vVykJy`~>+XsF1S7sr_d#as+(UAlS8&84PHOJqh(2u*mdAq*GJN?jEnM!8#7(4L zeYBz5e_oeCYD7ReP`AV~Pc!>zfA)gZmKBOvhMxk~kJh;cTeYRatC?1Z&08sfC|UaG zii+eU!LN^JqoB|g#bT5z`y!lJMX93bVKVwcZ2g1 zhOUE`qAx^B5V69%Tu@6SilII_w&EDai1Z7KdjMT zU}3F2dxqW`Vwu{v1)>~ttr*#1t`zsIQg*RyZ!HUnG!Lgz7P%u#2xUBPP+<`Us3|ot4y}o4G zzOHF$v=eP`I_qGPmW?NTbqYN9SzWS_PhuJiJfbN)w-!^#&mLbmJAVUwnZ+EzG#+F7 zweW9&s}_0tWEIr~j?0rqdmxTQMrs;)zrY!j)5H1`l^OrJTSg=Q&vd!h;Qgjb7tJ)i zezxT^Z56hC%-}WO#jVRb$zv`e5(RVMuq~vh-vHy1?ui$^& zN37#m=C?6*=`TH=SwUF8UL{Zcw9j zI=}aVWO6^IJE0?y$m2k_7?asEi{~Z6nji*$*%X7L|NtQ<%h7s*}GX3^5*ZHHz+rLb~8f6 zFzHuKEro|7GgWRf*4y@CZyAEd6_-y{DBASlu$1+ znD(G|L^jobs`*ADt+%P({q7a{3B)p{Zrk!g!karf`Y7_Z{-=o%i2ZR)?_haXsJq z72M%@+|A-gPJ&J?JfbZ4^WSxOf|f1j-`k26?oIE{e;o(^{r-68-_`i%ll^;4l%pJk zh_%SGIle!SQIKF1Ne+sv6-6n22BYP6qjw#EO7jRoJctLULh4J^XsQU~ zvIqBr1OaFRIXga8fDvP?iCip^m|b`V<$Nv8lX)ilCInvq505yE3`{wtigAjMV}*N zE->|~E-E$Saj*}>6-#Vs2X#cG4WIJX&;X8Wy{2%Kr!o$EyLbKuLtGHpCt}!901WPe ztwduFwrTR6v42_xLT3<}XMh(I&>z<{b!gx}G5t{-rN?>vJ4wx&a#gn(gbiVWm_IjT z>YIY_UWQn%t)`g)U`Bx_AyXkChHdZuECnL)0?{XLq+ zi6j&Gz|D!c-`^104UEAEvQdD{3m2L`DxtgfGJf(@ks4H)vYKK+{I_;qPE3mCA=v@g zZ1=mAo_o}E7ukmERQ{DRRK~BV+YpD7h@j_i^J0#$q1UTDJdb>c-6V5jpVK(1LN5KO zKAGA!V>r(q3Exyiq!;aD?iV|lnLq$EU5w>aPF1aN$s1MAKsU(4g+ zGrf64oUnix365f)ERa@{_Xj|qi9u<6CB8&OOgyIvQT^|+Qq3QbfW;-1#xE`H;X#kfBUxWA-t^rDJ^Pl^{>DCQA>a*~2$UPvS=AJGk&?9Qi( zf(9W`bv!S4X)~y5oyK*X%U`m@Z`CtAbZIf ze|a9=Kx}*$GeuV|dDklWAc}$~pnAmDI;p;z1}JgqOKW3sr-YW8Io)Ygkcyl%k={5} zpplxZ*~{YbdmVZN;Qtn8mH#hPDHcae4hLlk6qo!b%!JI9uogaRaZrE+NZ|i$(m~Z3 zb?cpm09>Y3)iPA(RncU03SI@{rASM+%sV3a-`*J& zssMKCG?k(T*~J{or7P#67T7h}Zp?6v(0SFXo8j3FiRX*>Px<;SM@#;zwS`;30i1GG zo@o(bVmTgWb@pJPtGwNAPVEGkTK3X4kViX1B0!2?cZyphgIc`AAXK6h`ikxzH*Mu4 zw^q7E2Sy}}stlTB28w8Dm312^=h2`KmOtMf!_aThk9A%gXQLmnbH6S4bFrUPPHxRR z)Do2D$393%-{?2YKHvTJIuZwU@>6hG41cQo?w_i{FXDB6&0Uy&;Lp*yZ^0W0Iz=IS ze66-&#KpXbNS9uIoZi)%%Nm$rs{5%v-~kCWyJyUGpQ^JGVx8GNr+7V4xSvXgEx<<; zAivZpIly|^x66eTzMAJz<0*Gal~)XmpJDurH6v7j0t}eMs0yY;(I+~E#PWY& zMWNXQ#{KXpjDlO`yK)7WtdtCYO4;yEs%mdo)#gs!;GF}09PWrw)gfj#YJe@E(YUJ8 zi-fi$X(@2zJ$vX$cbowHLNcv%Zr)xeMkQt72<#Fr-Tyy^?%v2Hk{C*v^Ch$G1~}_E z343s-tH_6iym?GZzzd-=9ZB2FqxMsnzyOm){j=kK`$STcF4xlo?`$G#9wVMmggP{F zt{th3{N4}zpfhJs8sF~KOaxV<3*&!|Byx2e2g{RQjNj+Y|C`J$__0%a#CPzI@u+0b zi?361A9rZ_m2a*jk!o<83$RXRCroBsYHDCuNJB&#DQa$mko{(N_-sj<2L#&=1+~r;z1veq!Kiv6>NRu0tm4n(T_(-pHPKl|mbgyIA2Ui!nL+I?hQYTNd~=7Id-!|z?KI&b}Z!)8B& zk26nD?(B-+-tcq$IlL=PxZach71RF)Z~ErtmV9N9C}m2&^g;Cs-G{p{L_s=LFVlT9@DeI^Pj62Fy^hw zi2F!JXY^+%J!#-g&+;GY$8qF-yy!AC2UMAaI$4=B=9LiF=~Riz#BEv_S5)yi!D(Ji ztR@CC6^4R6W$4&2K8cePjz@fEXyay|jMt?dq~>Z(?KRkeiGHCe>fb2&Pf5iZEA)b4 z>10vU8^9$LMuPS3yu8u&41^2G|osOG4_oV7KWBjt!V%I{*`RFYpU{2Y{dw@xtZvRs4P<`VH* zoC18Yj@oc6gjX0Hug+QW24R>;*Hv%yyoQGvzM3FIzi*_eBvrzBs4HwiMQ+IZ>>emV z+ngPdbbNB>tr#)$Bt9xR@52l6?cnDIL3)y&VvbCzTa`vhZz)TtQiL=N-K-=e5aoiG zl~GK3M(nEw9;$t^8z&I6Pa%yT0t0-G@Z@s+%{ksguGL~}tlES#hLsDA*RI5O0qs%A zL9W}tAR7bgb+T{nnqb!434@@sKj}*~n#?~og;qNEXP~Z+SnfP3^G>h3mBp4_OZ@JRr%q@`$MSLWFP@EO`r$j@=l`hx>LMOGetcymbNL7J39ucH z`lN1mboC2=W$1%iNy>sm{_6guRvV+MxmId1k7r6f!4TKmU8?{-5trYKA#|)!Kfdlg zGZe=#l&(>A%=d_Gd%N9|FVhFj^K+Q{##;)nv4G_VB$iJT!>l(s!b~^|*O6G4njGev zd48s3KQ#8I9UnY*hzEva#xp|23hed8sF4m|C}!kNCJ9>3+6}3*qt3+4!^Cv-9;LeLt7okyN$F)iX6%Q6#F!rL6bX^1So zJFre`Qd(uZMK)DE{AA_Calj$i?r zr-@dqs)JrO%F8)}k*g-0xBHLl4G+`JI7DUu1c3q7_Q{k&rx&>zjzj%)q*(zq8lm%c z*Qw>=QaM78+Y)m}w0n(r{BT#h^H=fU??9mU>Ytsf+TRZr+eT9!r}6pw))b`_z36id z%LQ@$cG!*vzOQEjSwAM7N*hZzQ>6d7#f)r9mv~z~$8?A{PyYU?4i?rKiX2watKynS zQQ(m{j$%?;xjhQRcQD@-c!ODU6vmlB?O3k_0ye%<1;3#n1!TCKuXT*kk2u<%yZioK z3OoBf5f^gt>R1QxmJ1$P!*?lPMs9_P0Xoe8W#_@%h6|^qgsz5i22MUVQEf`j|6Fv| z^ZMm!8pmP#Pg~Bfzrs$6kFEDu?*D`{W)Loin*Tzb08B3is3GCh z5>W=X7tjP!#r1zseB=ykdL+AjB<3;fLqh~ddL#o^;OPzvhY?I%_AZ4S7o#0Y8;xQs zM)~R>yyjTotpU*ZD5NhHj-HV|*nUMVns1PvfA?8ZakS7borplR4iGIKPb(>aQn5$N z?$XHTL@NuRxhl}g-DtHz>ifHF+QBh`=`monXQYEp`hziCYcaZNPNv9M4*OVSRP2*^ z8tY(M+v`{%<0y?mdW+!bb6=_dJ}Iy`98mAeGIvi!oZ*a^+I=9MfX&PY%~zy+pHEpCq*P3gs4`6jt;60p`(N9IvF);rxrDV^z&2}P!wmPY@BRJ&$dKS|%pz>+UuuTHRtJy}2-&15ZhiUyD@ z#aoAjU(|-O5(EN?nIc)Op`WU~mppw> zax(W6>lTpiR`5Ibw$@kp1iohh*SQOa41t>B;Z>DM22_qV(thuo4bEy||0!{qcc7Xf zS*lt(l_t9TO%xM81%r1fCvXL$q?D?c1>R8wT2%0JJ5c~9%>FlEgfGFi&KJ%J->myRbb3Dk1WwUp!eYhQMMZ?2 zWrG%5peC`I5%lE*SzR_IJs!3>K{;28T&c~BG6oirRh78G%(rMupA?0WaVGXsA_jx; z8WottibO$CB6ZI5>FPL6!_0Q(Z^h8;eP{r#(qXdljbL@MW|gOX)kt1dbspt=xBMzd zwSE=0(M_rW1eixq#O;%;t#c=?i!6W93n$|7QiZ%MblO5PyR!6$9iq-?@2ck@Bf9i~!UXG_&+t>37m z--!HP7Xhvn5ONF$HQi%rkaTbINotbsXtFslcLyffST#B&HR|Oz!gWpZzMN8gOSXI zb?5i$7Oc=`S1bGjMUHF5PV){tQAaoAUy2UqgSh(PvT9myGYT?G)N(?)MVwe-_^V+&VjuaM17b7w%Vsd|^GH^V%;cxO?TaXZQqW z`BQ&N3KI1bLfN59Wi{wvL>k8mFsL3p${!TJ8eEcN$WR&z@fcc;8p``QQ22MK*kBOj zFno)ES1ff@dGw55_6uwer?ZaOY7Gf=4&-(YC1XdTI!8*EMmkPLx@-D>{IW?&9<8kz z9Z?!F8>Mfi_&W8oZ(6v&Ec`3JVA!3av2W??vDm;0#h6adXs9*s8IEEH|D{k%jGT1> z=&5?X&Rqte_?+DT?m*eXmpKVX4 zQ|+LUtNe3kvpyma(rEULQ_?Ow*Z^SG=9Vq=<+QH;?6V(Ju7$L2Da3&Yh&N&`QDn|P zWya5LZdHuprO0gf*O>^<=0>J@mru=`c9h2m^7j*Dxjtm&K4k9zWbXBZCcponTzXdcB;V!^f)nn3%CO@DGkczSbh~`4?04bw z-)95c2LQWL9HiP-a3!`L#Ehr!-4OW3msMH1KjKmtJyTsWKULcR+P-V~Ez4RK3;-x+ z5pag0HO8(rCRi~$toR-mjQazO2fHHPzakY9b$hu2qFP12?k856ZPcv~$#J{+u}pac zhadf5F98YHf?zJTbSGY#wqEqOY&tt2gAc-^4MA?4%Osw=B7=Bvw0YG}<;(P%Ty2wr z-)rq^%PuwhxETQe0GSe~?2ERaclrD)2eFESPF3M&MzkPf+CZn=)JiYkT7BF(XYy)K&6qZ{%1wq8aCMSm z)`z7atu2q(uPEAoceKq`44Tn}%Tv9nQ#nx4I><3USVU3WzcigZq1al018{DUb@y5$ z4tjJ8M(s@N=&4OY?Dqt)=dHf8N)41=j=HtE#&`e4Y7!+%GI_hpOR=iFxG;TM4Y=G|jG z(j%RXJa&8dJOEZ$XSv|Zl(4GL0B2(Se98=GO29tA8j;j+Ml9>m6|FhHDP|S)W@!;e zyjHbL$5Mw8_$c>+rSvQr>cU~l!{(L)#i`*iipUq5wpFbwf=0A&};aLKS8N7?1 zZx;#oA$}q^j;c4SoSDEcI+s9oTL;yBIAo^mzw`y>M1bFlF%Xn_Q7JDfAux|OVLftK z!0FLaM-YMHe>dsW(f|IYocvNcVh3P?0WKEn*w%rnY7@xq*_ z9%m{Z9R=yV9JNeIe#Ro@?~TQbX1id(;mMi?60v0BvC`G#m{a@ov;R&`9?vzNT_qAk zs!hx?%hg*f!NHoxqw}CZ&SAOpC6Bf!LwmsClwoc|xxVKp$S%Lqg|hgK*Glw7&iv?j zaSDg(MBv4W*)Z&O!Tq&-O?M?^;pLIb2~hTfExDI=%0w|)__KROhC=Tiu(kRSc9#1J z8NS*&za&7|l{%W=K;`Kq1BGeporXOt2F)SB{D7ir0I+M z03BNK7vrZF*MErruBO_lZ?90x7BPnSC61h4-u%q3{OL;~`L(=fOyOta`GBDR=;@f^ ziFHrB(t@kvvCQTEREq4%dJh<`K3jWC`FBF@1MDBKgElt_UY<-fu)tw}orN+>RCu43 zHl^_c-d9k8D~`8*kjvAFs}t9s^uD)e$&N?w?uK1yvc#;#28GBtNga(7K7G3-LcMn* zbwLw!_kP-9j-l3XnEVD%#GN{^4dC7>t}tB@cG6zSemzheG+La_#sk-CB(SkOa!uB4 znwOwZ#=iYOJ9hOW_j;su6twmw8(>mDjQs0l_|^r+hoH(1Iw{+mPC}U}OK$$)-g?`nOWdtz1zcIfNb1mF2V~ z#jP9POzBy+cGd3Pfbbl= z&)xfT5YpEyd>AwO59&Eo`h-QM$?JpAP8$s)=Mj)mH9J-LYL&a}?S@7B-n-HwW`Umf z>|pl*2Ja<~Os=5aGqga z^n2UYlc3s%sn)@(aM}p2_3sg3(Zo5g@}s3la-XFs?PeT!&k@IcI$1D1i(kj@wAQA4 z4cZwo{zF~g^|kH2zhC?pA~HKb`zeLxOQRTem^g#@?mdD}(f*sO=3f9no*F8deC3ED zSe{7G=w-Kq71);^s-ek6OyPQ54Rd?1cJv{m}5yKrd@B97#{gOa@dxf;Uj zV9i!#S3%=E--0bx_viYPQu~r{q$c;a7*r^>`wbx(&3b}>`aF*@0H>gJvWDpI74on= zJmBS!Yvyh?lY$*m^sNJY?pZ2K=5=Q}d_Kw@e|dHO=AToN;jW#LM4eV~^?8OoyI^bl zgRJ}X&-4|!NKrv@R7LEc3O;=Z8N$lAv9HQOc}rTLLSi$O|FT9NMnr#06vPw{;vq7E zfYeGk2YC1(`pKwKlW~Y9H*2Q0fgCg4K2j8b;&x+Od6Hck4W`D*6`JO|#Y)6*Rhg7k z>C^Pk>)So5w*Ca=U^T|n4%^oV2_6s}Nk>GQK##aQ4Zm9n={$@XeI!6vabWY%CH6jc zw(%upPlc+9uw$&INHz&X25%4aoItb32CLfs!Vzo3h*{_on5@wH&48FYAGFxWTwilP zL0WV-mj6KCe{=wuP26nivKS_I`<|vQeEH(j;hYacwBjqN~1LAqEZx)S@X%1tx zgqN!mObbA?xni;WC)H0MhL=#|uD$t{xJZEwASQSfUU{7=&9YIJ{*u_cv^Q*lbb#C_ zU-|gHOrCY19S=)4Nxx{IFFEy!dN2QS>^-v@2Sm$!pU@St1a>Wp=$=I%{G?B`d564* z?zQ>rjuT&-*0pil9pIvBy$rYUR6zu9+|a}RIOR>QH0sfwVXH8m7&e-;lz;Pz4B8|_ zqxtsY{@0Sa&i!T{9nPUj3tzpKhqJ0X?VLF`fVAKIjOdr1u|1Mgcj!L}84=Dx{a;;1 zAfOU3@P8X3(g<1)`rukI|G)9iz9gzXuAcwXWfTb{*R6j$lqv8(UB(Ri=4>&?;R4-; z|20J1mm-BNjlEIjx}wl)bnPzI&V*=HH9mh({-AU!Mv%_AK+Xei*?YBNGZ;9b#EBQHpdGIS6x72obub_=Z>Ug=!2lys@15!8Nn#!YJ zR6{GVH9|*lSQa*x+!@DdaMWz<0GTH%ZnJ-%5jW;eM_l5I3{M3Qr<=Wxg@380dwe1C z*GSHfgIA`U6KRKC;Zr%3Jr{SiE04aTb%NOe90HwOzA9x{A%ge6cMl79zZp`u-Z^~y z78<%)An#xH?^TW4G!ZlHE-;dCdm1#u;_Z~)cemRP{S`dPiKgHH-Mz|<7JL6LCG2EK z=grIOycr4*Ge)KM-m0Q_ zr?&IG`WG0Ih2ICk{NbxpdP479sYm$o5VT2mYLpPZc2_xbQ$(c(Ty(Ch4 zr8tqEQ3*QR(iEwl7_w2(jwOeE5xU>hf>H~kl9h?wM1*rxO}$s(nEY|)w>`yei@EXp zcnFJ!+CQa}hjjzY7vr|Kt%(y71rb*UQer@#uZjz*;IDGCJ_qlm;mPMHC5k?phplj@ z^RKdFD)^L#a3s%1sl7@l6#x)!2DG{?>a|i^RzjTMf%;iryZ6X6@xiw;KmUi>~O!PDOx_sscDwRY5Q!kPs)J zi~hezD>2~*qQ$8axvI|Fp}1rZ25n;ka{LT7fIoJ%<^R2UO`UfVLuR>f>g=DAy}1t@ z3MNr;7C?mW!o&HVFOU=e7osa1+15|u?h|G)8kw79;AM*eg~4u0FWSjF+o%`i{JFEycnJmNqq?X8DFBN$~1hPGL==XS*G@9UN8BSFWZEdM5LPkb1b7H zN1p&hX|+4)Wv>SN+;ty3YAwl({w-@+%`NjQFCXyuTHI5FdPGw&i*|Se7(Uu zCw5m)WOyo>*@>nAxvR5g9OKfv!Ekn>Q$t;E;@SR!^JY~`+Ck7XqE>^o$|BChz0dTW zXm8dSK8_}Dm585Jjq!RH@}Qf0$ldlunhIMyg7`~X`2?^@+}YBy3qhQCW@k8q51vM#Q`($24FJW^X6Yc zaT~Mym#kRx`lr%c&r8LPk%SMkb(n=B)x6=U`TpUzlbmt)!j02H3Qp?CqkcYp$wIM8 z_L^}^?1 z6(RXA*HUuga9kZ^yv9R)@J_6acEliKU;;s{Avfh@_;FD#dVDt3FL-0b?n)c7U54vy z9kzCT(2)15WG(3D7Gd=N;p#nuntr~p?=J+B00{}bLntB^lrEhhMMTt~sDPq0DS{Y^ z^b!aq^w6siqzWQIkzRsGM+il#NJqK>6hV+YzyEzdZ=UDf&g{(2&YqdG*ExIbe0C@M zeG1;@AgDuu9QFAsW;Kq^7xe^V;YuWkh?A4R-=XR0an(inM2BL?|K8+Y{cw28TGk^R zI1aAi{IY(_#f3lWF}Sol32sQYyVKZC(P@@nue&!rG>;2^E51Oup}II;Klr=y@-E@V zBU|p)l5ldw_U2>sl9R2h3Ts0*h-<7`FG&0xV%pe~~N)$j{Cawxi8Yh-p zd~qH4CkafU!K%Ux(N|TkBJ5)7npY|zS{E4+;pR-@2REVTF$L^FPMr#;NpLn3$8Ret zR~nNF;vXu$M(!v0HX{n60)GZ@{N1_2bAo|{a($a`-2JY^DXJpQ2ne;)oerCRV|lLB zpRCCF9NpVN?`MyGgE?S90+tK0A#rxtd=9nV1_zH+#ijAzZhiw9+MYQR!av;jK|?`# zq3h6~HCe?YWkP%0HPx^sRYcQd_bu<-DjPUPOi$WhJE1)4L`<@~vAbsQWfi_<%lGW@ zRKZT<$Gfu|*o@14~x>3H0L`9i5}T8Ij!*9j$4hA-|qc!@UQidqq|uN0S*Mzd!!bTswE! zza!)OA6v-jQiM+6cQLUehx~u58Et?zY5n`JpAv;vmkHIJ#OpxApHo9~6~he(!Pm?Y z?3v*g%tK8k!-WpR75F3Q=y1ax+&JeQx1A#;#`mpS^hVH)9 z2ylH=U?xur7o-?~iK~X?GDKCLjq-wqCg7s{d!iEP{35}bsyHYNfx(a35a>~aJ`((~ zBMLl%C#`{19Z~Ka1RK?e#?WYwv-+O((TOyO_k465M+~oOj0GkpOeQj8Eu^9>I`2p14*l zJ5?sWt~BPWk?;TQsRnqk(x9xnzN~xZiNA~zvjEnk`ot_^;+_-R@I3nv8aN{Rr=&*j zWx(mk|GDiNXv8AUBu;^(I?VZEkMsQ3&$l7r6GRx3U-T?VF|JV2)GS%rFZsQ`lDcFF zC5$_Ofmco-MedD!vX|m76k?Gs$9YwsU9rQVWyGK%EcvQMGLyKzrvW=&iR-_qpDz{l z)KY(*3E83{a$d`h{|`W!X4#PTXe!Op`O+RPWn7L?V=Cnz<-%sNP4ix=j76&95n$v; zRFG%914eUwMzgltahNeMnlW5J>mtc$1PXly!|1PuWPJt48 zydnbP&vR-B@xprFY6P1I7T-}X_QgT$k|6;|HukGUX}{oGNWh4d-PoAf`D5hgdP5ny) z>CK$v7q!;XfxAz5?8;!`a0VRQs~qBjh7OB^nG|Ac);x&n&~Ujoc-B0bQ5z2Y z8@7Tsx!GlT#!#$!`HOjo9j4q3TON)qt9XW7JR?soN4}qalkZ|SVqB8sT6%|lmzZe0U1ahVv zdE^7RCkWAh{{rOB%i<^O_?G*S$~D8~?IiOO#UUGg@0{k}(avY@ z{{@&_st%*7{OYU1sZ|V2?3=Hlj1S56eHA5^5Tz?nA7okO9m|*^?&u#L7WC?Os@FV! zmP;_8-)~vJ_oYHEMX3@%rIAE13l?C;0BmHYcrx=H62ymy)MR6aj6!435YjVp;f6h2%Gc<#l{!;_D@1^q!oWxd;;o@E z^UU@nFoMkNh!q5xetb{?4MIbFDIb{@KbAmhZ7bmRF12)Ysv$=JVaY?iLSjPi{T%LN z5DoBtSo2a2P>}=P^*4fujLvAN3kK?m1wX`r?d0Ca_ZKDQynk)p(C!Jb(WspSfPZrO zIkC0+3k^*h;DQPX1iI0@0_s9+v;#_+X^kKX!@C(EmQMXIu#mc(rly&!`#Es)8K@T( zoU9T1M5DPgwxK(xnewT@*~Pz!t?|W$#?hurUjtf9ed;P^T3FB3v{ayMg&KanIp&$ z8=(D1h30xws|&S~cA;6esr~n6`+yMlV1FC-xUJab(2^CsSXuHd+DeP>N-?A}8{=9vp=Oc7XUAX3buufO%=OIV!GqUONW+#a2U=Kig= zt+2^^^X8fgztv+KL_xXwci4KWBzKsE#TZZrr=WuwM=>)c!PprC8R|JC?STirWRg|4=;C__GHmj4>14b(NBkG|M62TD?an*1 zcy-P*wIuiyQ7U(P>SpLC3lok6RKZyWcI#?_KNcz(R*|!GW?=T44Gv1c&V1%l*Q(aL zzAzIY%wfZ!I1TXLrgK7~C}4MN=W@%}!JzN=f!R8(vI0*XNqw$K0Nz58`86w3jb!_0 zE}2CGbXZwaFf1uF7P=Kni6_^p9AXz4T1sTeL$T=mEd|^YN2p;g-6vMi>+uRk3~{ZGWV+Wr3UQz2Dzjn+r_iNORA!<4@9rJx!0vG$;U-_n=*_iBW_))Q(L;d&+1BHD7u|{4wHE-DVr#VwY_H5z>U!pqqUgzXceC z>OFTDk2|7<@@)bl`_po|>#LW#3F1cSf)I(~NA3k=hQoUa{ zdAs`|dxOT3ZyfzAy$M3{^qOYMYgkg*nQ7`>Jq_zP_#(V_d^p<{axf%W4@uJ`LnV3P z9(hJcUwo$B78x{%pI0)^_29m5WaFOFvY()w{M+YCXf_A(i~p9_;jKVqww+k;MPA2* zNILC@7%o1rb%r@5i zzQCM@F(WnJAWGOHGDR{CIfV6iy8k5`uHs~G2Y)7cR;BmX-PVS6vU)Uli_w~XQd&!DJI>^bC3DAIy&*n9W zM#b~0#jH8!J>pn;<0^hpo+Z4knf~kiZgXpTvudmMcr&|8!E-<>S-I(!>u`c{r2#*3 zG*NZZ$?lOB6Ib)Ldt7C7yxP#`ec;OP=v}Y~0w*sE7`{n~SDKeEne1MHUGN}Z4J-Uo zy+2s^!hA#%O=~^*%SNVj3Ixj4l>~plb|9{4x_9z9Hp|=fIXnG|zAs&=6!W(4II&N3 zfeWs9x#QcX$nMW9!szbrRsrAd zHv_pwBoF6zt4FcGGlzY&7Cp#@3xEqkGwsERx~}LR7)OM9S{GSVNz3?QoiV_{teai+0QD5y^Vk@+H0NJy(Fau?t>K>yAF7=-kllsEx{Q{3f9spw{O%fJm}?TL9Pk^u_l5ZcZ=kheG!{UE48x zG-br>tV_9bw@V);Y5H4<ut~aZe-Biya zhG1y++OMcgR#yFHoNZ5r5PV7Zwwj4BJNe=nWoJ&)X|Xhq0bhmpCFR0z zZhMs9%{v&&rJmTK{fgnI$vXwSN+lu&YuA!2(Jfv) zub=!rcXm5e+cOYvJo#u^|JNLL!e_^w>swOTRZW7crU#b|2|ZG z_Nk%<;|A@G*v_+*NJhPHi<}wvrqN)9BKJglG(>TL1`?ymRV1w~F!=QBi0DgL5t#~~ z@}t3OwBhr`uk;LlsCEyJ_kW-9c2bD+Lu-sn#i#F9*BuxclEP+43xa>zdJGKZUhg8@ zH76tjGu>W>yq_g{{gG^+&+wfv{a*BHLU{doRpjhu>*3%Zq2DV-tYZs^fA8&#WTcaA zwR;!Qf)^SF2$4Fpk}2p}Thl0+x!Imjn1fN_y?F|4BvNy^{8X|vZ;wDUQeET(@d>5G zye|T-C%Bj$_1zy-Po9S4g>@8@FQAS2avDRC>M3(#bxn0^N1dau+b3+YrXwiBdW7 zk8{+N6d(l~UMbN{;LQ;v(iCA#2L}VvGayY?w#;0ImNm_#6*OV$RxwaGi+>zt`nsCk1GhCVb|KS!APD}l zs%iEx(^co&msB$LVfw|3uRP}2UJ-lJOnAsRW=6C}a=BO)@1fL)!o4-6n`t6QDFRm-sw#+_;bm{PT+qOn?E~Cf1h;- z>-XWFxL$AK8?S5ayBV06B_G$Nak`t>-IN$HX1C(NyERGd(pYx$X_^b-?+04e^ltl8 zm;QA}-Wdh~m^t=6d0@^Ltdhj};S%re3Z zEZ0;Q5ko$$k}|r{re^fN<3rlN&lN=9aAM>vLWAilDl+j;&6l~8d?_-QY?5wZKK-MG z=$w(zWr;0nJ9~KqP)V(Z-4A@iHNuBWkYi;s$AQgpKYdt8G?q*pOEuDN7{}0IcdIKZ zu&+BX1Op8rLxjzt2_$GD5GKJNCRGMarTsU?k$~WK9Q+4WZS@bZ<`@#M5X?&s87BR= zNGK5q(t_g7)(2fx4NoGnq)~90)@(g;PMpwtnBk_@OmIGL9wLD0%^WZw}{^fn*8CLSJGaFL0_Ms7NPN zFoOD9CoZWb#3_U0?)|ST=0;1{RZWykO1u=E zcznopdI$&Juw>yNQFQA>h(H2-Duvww5{`t%;h>z?oj50X_>SNL4VNdVzGD**7dfXbXp(KmjKxF)gmsrqGIe;ZBtc{u!j)v@gHhT8vo+A_5sR4>dvT^yP9#}u% zU-uvNfOiA1+X88tN@)&eX?R?kB@5>xzvRbRiPqP7ZLg=#8bKYi(x076Q(j5;l7qUd zWki@}w7*aFV}W&0Gg4VHRi!e+uO~)MrGbyqo+H!kEYiR8XSwtm6=oTyvt)QUr{*|k z7)NK+*}?$A*DNAVt(0N5I&Vhw<)KqN*lz!>syz7)smZK+;1cD+cAsYJ8ym3?2T z!*m&HTIbobw3O^3H-9Kd4iF~424MgY1#p&QeSiYnVvCDdIm3+eRX581d6wDsWmT!m z>YmU0@C^LXKTXQ`O~W)u1p}y1E1ISOXAJX0BskDmKBW9@Y2VwF^|$nna>7P=OG5!E zDlCW!o|xA0CqbgHMVHsCFZ)5*=KniNpve+oMl(2DGe0IX^D0V9h^My!IYDTs3$a>I zD53oC+vnJHo5uVR%PJGUszfT}CAsQkw93Yob&sEnT7X=B12IJbIrMBYGnqM_#{7g> z&8bnnM1=9iRtx4-3$s-@0BF=qxt;NQKNk2F2^vEKyHSaD*pHeyA1#|c#+`xuD~G(o z!QO#l#oNtKm%#2+=tBSmBQmn086J|Y{w#9=L^#i8U=R*6U2MfxUX^xK9WDor#iWW| ztvWq=MWrHQD3C|aAI+OS68k?`9e>QD!0a??#cXQ-ErIRG%w9w=DF>|X!r?i?_3~kL z#ys?(zHaJkT_kHg9toX30wRwK{Wt~xA;C-L#dc)kgX1QOP;+~~`O+Fo4-L*M&o0i8 zPq`p(NnsAcf@3hucSubW2D|_k?z0KKmx=ZvLFVWI5VpUKQ)lYKZ#3NusE_9qM9{z? zIMoL;DW5dj+HN$zux{=-2Kv}IEos1)3*bj2=tEy-M{DN$NQei9cdDsnRtU1d*1B|D zw{oG0E4o#(qP1>QZilU{WwYtmar;4RTdxr3zYEP)Sg@Tnq>Ht>@(pGBm~C)^g1Fe~ zJ73qqgkI68=Xum5*U>68&;&bbM{R*S1DYiV+D?U<472if$YzvGyb2SI@}2_-mLba|2ToSL)$m^Um@#n?fdfLV>e1x^<--<^{{g&Pnb z>#o3W-9dp6C^y6^x3%_$wDrX|1lZ5avuMO20ny>1jLT9O3FI#QFuBbMSkFoQL9(@~A<2xQ zd0e?;2*hiD6r(uu^;~c#`yKal7r*nz*znz}Wfuz*;97%Q*c&kKbA&DO2Y-dn&hzr4 zbTcMoq}>+zu(0R=g|nkmy=UMtf0jq6Ph%hW43oq0hS*`8g71QpKJv*P_uU_I@JJA_?Cj=htCm>HDDAHQQrH^83ft3^( zL3;h|JHAwQc0MFxR1O}8fkt4VTTWn>hD|5R=5R&2vYFoSOn(dx!kHBfkX}@_Zb|NL z@S-+HDS@6u2urvbNdJ{14%`B3?fxz`aJnttE@ABkr_MubwIM9a!xP#$osS0oda|~o z0_U}Vii+=0(L1$|p<|+ujh)Y1ZUaoNVA`dhzqEf|=X8+=e zgN`gO_l=93>QoYJHq#1FN{ciATU!s{Uk(jevkMU0g6QWiN-cA79|jv7CRrb3=2wAlNcfOkh!+`L)C;b8at!>U+FWIoJ!DTx{9BiNy<+Zf zl^9F4&hf|F$8}ZAHk_>ezk!yY$MK9OvY-CGLmaJjWeJ(J3dg`zO@f(j!9> zV>&0DiT|eZPx|Lh=EY7m6iyFKPsd-JbXT3u4xN_%JcTgQ|A^B8j_5%_MP|6DzEk9v ztC7#AorXQuFf#TdqV|Ue=q1Ah5v^b*f&W9XxpI2^V@S+6UDm>5i|?*9gWGVPDW{R# z!eAIoNIOooda*N3I$qQuyGF`5?b54L{5eJGn>nhGz2$Y!0zCrfIm^#wW?Sa(rLx`y zPAim~m0$6+oqdxApCCPnPVdagxHt-VmjCXR#pw3Edg#i?Z*`@&kGtZ(l<4R1J~fHI zQ_5Of{(;=yf#$-?ZkYdaig!Hoq3DF-p;JDp@IrgCp#noKqCeMA+4gw3@l~+UAeQ^w zE#XP<5==d~IX-*p6?fEQ|Gz8GzY$;U>^zuy|I?*Citlo!@DXX{OTJi|_N&I{K0DuA zT=?2VjJGQ;d`?OK!Xc?0MZv8v#$5tSktvF6sBoeIqbV&xql~l|qpr!~Ab?dM_sATn zEE7e%2jqV+xV9dZ;^b)|$B5wCJC=pi6; z;rlnQf-Q<}X+YW~V=qP)5_-j1%xs<@P}mVPRs=#iFD=qkdkYpuzWiQExgqWU5~sP z>@(EU`nDX^|Gr*wS{gCG(=Lm}XBs*tNH&dbU9+DFD|Q+^bBe!8BkslzaeObO$;k17b`IS)kv zz@VYV_6KpCHB;SOq>x)J=n#d2DUnHGtc~x)BU|k#;@Eb`*=BIZLnj-ic+@1}IaCfO z45}2^ED_>7Er=4$c|Klte`tk4#4BJ~pvB8isc)V^U%5`edBUnr&1=f>#q7%$&AR^K zF9+gpQ+owsS%4x_MfW_b#A~OAnaPf)hliqM!|XrUh-lGxQoEYtcHP`#LFZ?w;9S>M z3x>WB@8AFabUlumJzSlOnmh7l?CA<{NZ9qsgDx)_eB}zgnwG)mxmN3yzfXRXFkG-?DJn!>C%&7p*EL|sY zUQVbG=Y+x{%3BPjwOPGA` zx7Af~p1beC+TZz~q${M0J{$QV#z3cl#o^8ntvDN)FP13#B05qeZ8Ud97|;L7n)$+3 zC=btYB!Tp++U`IyE5~lt$Tp1OFcq(lin*>_F~#TlY?LFI)*;kIlXmwXy{5e<&G!k{ z<6|tDHGVZUPuO)6Mq0uPeJ*`Q9;YWz65RG=kt2bO7kh^!(NjODeh=>v z-owUNo15hs#u;Dxu{*rA_CZWiIqLeC(2>wnyJFcOXA?*N$n2f961$#t;aj0Fd+ViA z^w$JT^t4IQ`Ls79;NG1pMPUzq+o|AtnD*9hyq z+FKvX9qXUnubci_6{ak`H+g}smG>F$Bg0iB5tL;C3dnAMcdy(nV$)&Sc%t*Wav?g% z)jr=VzHjkE+4Sz_m0OJ#yz>Ejs!14 zq0o@^tiqFu>_Ufws+!8myI=huc5LugH{W|}CB$|=kRQ*|IT?A$x0>Zvk)FsNEvjf}4T zA{g+ei_W~`+};;`(-kB-gWXq&BPRF}3X6<#ei}Y{Yath<2Fzg$#gCAq4A1#9zUzCv z=Gp#oIj{isbpFllh5`ptq#f6bV8z_8;_U9f@cO>19zhpi4q=Sh{DD{|>85~4z01>I zx7n31;t|1KO*896h&B_mt2gxFkd<_;8Y8QN{SWu4FTu4nD>t|0oL-ksmel-3 zo`Ksl6<%%1?oOTeb)y%R1k8314r&s4Bd<6yz=y_D@|L%s`OgTYP7PjwWbNXDJ-<3I zUb*7>!~tv9%|FHJSkF6Oc~;imW-<2NVT*F|&pCIqyOY*n#mVY!kGevGy>&wD_3uAY zx+Pa0IC+1zI^EN6YKMQVRc(zB;wn;8{UOR)JJiqc%ZC1YUi!9bpU{r0hRmz=n<4%! zPZQi1vh`}`Fsttt_xD3%zimIh+z?@)p79k#^CE z?p592>K^-S@CyG7x_=Wm-0`90Gps-04_tshLA{U;-gSL((qN}hoIqLz(bpgvtqkqW z-G^Es2YI5ukcY(q^wITB_a4DX2sbgAa7|2Dq3D5MU5 z&tyGtLn{&0mME{YR-wiAur1q3#}KD;RH1Xs1^QQrK3Qs0L1{nfX^;^@$IGCL)zH6g z==`-tn^3}yGnu&GhV$6B8xzBC`L^?iw+p1|-d+p8br|uV*i>l*@47T2ytF1Y+SVkM z!6!OTXmwI%pU&8SlKyqY0aprHYwt=aTr@2?!o~o|t4vHum$p7tjY5&S-1;`!ADti` z^QsOQZi6?U=qium{ehG|Tui1PaJi;KW9upfomIZuo>+FZ;M3I%>Zc!tTAF5c$+0Z* z!g_D8tWebsm$kaw{XAEazLsFrjnAJ=#y;OuLjf8<2KN0=1ij!*a!d|3M8ls z-8aj55N=@psPmp{Cj^rdFBAG`%%G(g_gEk%U4d^_-&FgQ!i19!`NPILY%`>o-cSHGnkxh)ZUm8-kZUFKF2vJ`)zM4 zG3!EmY<$1kf0~|P!>}F`sXPW&zG!S9-&fRZ6erwKypHS|+LIw@|8l+7+2 zT{s4goz~?cM8#Pczm~f3UO$!n>Wv!Beot(6JSwV(*y@8$jpdf8lw#4a3BPXI0m+0I ztfyYZh3lik^-%+#lmQr!>^+v!U6$t{~PuCZ@yx58b0$V)buBn zas7#nxIRp19RrJ&NK7X5#3IE~O8h zKp!Q8)O>N)sGGTVk$Rc;!7q>R*7Kn`=fB%GW;yeYU40C)6^ z1u$SC#v%%0q8YWQo<>yWSNH`i23m~2zn0J1LJjGmE^blTUm_Qd6Q!8(oM|9dwP9{6 zQ@RNo|8t1IHUxkD29s<$qzA7lGbo~rzb|mp8Mxz(99c}oFA!4b%0XOFss#F4n<|aT z-WRg@CQ9aBcIhOslsXVX8M(-elieP5lflE|@ra18=4tn?%R-G;Qf>s{ zG@hDH=9HXUxO+RTQnB60G$`s6HP*F{mtTYG{Wdj#-_dsZDjamrqIeupd{2>l@5X<3 zXph-r=7uL9VN=Io8UH9-zi^dv_mZ9P>I6K*={9znpoofn{0Z;RQ>seCw@Hqj2H@_m zm3xwbm&Xt-qr0kQkmC8gXrV_Zk1*=h$)-ubE6Ig3k}N@3+!;A0y)uw$CP?WeQUo6PqF#Dhh~4Gp z%h=fZeW|;jZIq{7(3aX5%M-h4W7$FQ3P*L>gpw1ENe;(Io@Sn8)T80g zu1&YJ;1}c~-U-fR$~;syn(4duFeiRA_vOR9`0DyI(>}!R3=cd;oPCmne(@L4)(K)K zS+euv82G<&8-cE3nDoPR)O~?&_`@>ua!5Uhjvy#KiPKf-sgV3!N95{1!dKN$PWmZ)#!n!A2Jv?TPM|*)8QLFX%;qoIk zE7h-|{B2!8BdjDL_7XpLb@1Sf2SH>7}OBHDIk=k~X)e#5`3@IM~toT`nO zTi*V*Qatw~{`<%*?BB{<5p8Zu8DL1CHN=7Rr^A@Wt-!aQ%s-u5#6Iyyf7fiWUNxZp z4LYYonOpzTHlhgns)b`l&QBBrdbBVFETOV&jCpmI_wf^G!n~#IYV`MSylSjodn<4d1N6zwkj zTjg*iEU}$jr{PDHZ)ZQMwuV|{OS{o2mDvP>z- zd$VA$^0oFet~bC#Vco|c`+@<$by6~Z60eDj#zS4-?Q9fU-@YFW(jw^)u_|G*PuBea zc6G@4p0M*}VV7v{JS8?nOz#XzBmY(~E!#&mf^ki!jq~l1uRO!A&fh7ux{v2~O;>SM zk{#?kC3L0ZyLlygBa!s}OB-{y?hLZKmH%~>hi;!2-)_1ey>b9a`UK8tkbr!J_G^dh zuZ3F^z$#b!uRSCf3Jw1Rzu?$*72Pd)hG%*rV;nED&0+7BhIJcUms+pg($B=-AvSDZ z0;xLH{9FBbddj#f>IR(qDX04LSFOu|+ly3cb7BwE!a%r*YQ)_`X768VjrI|2rOfvM zW{fJc=^7332aA9(-vuIsly+|>ops8Bv(@nWJD9KYzKrs4>UJPcpBM6D;Z@9+6U= z%4n1Obx+MBt;#YauIx$vsY|%KcjhgWzfQV6CP9GjO+s1!uWz~ry3yRGBJncRBC=qB zIYf?c|7AVcZH-ftYWiZu?dQ1(sJ$=lDbelqfJa~%fR*F0Cq%jtqny{kE^GTbfA-No zBWKpZ9?*R~K2I?ph#Ptz0{#C{Rx*H3|DSKRj0{Ww1!IHbI4@epZ%QucjOB+bj1kC% z|JP6$jm_JM}FyeIH#WLg)au$c@^!-N8=BS&C7$k6R#n-y1z4+5;{53WDn~QGbKiLI__Ek!f2JZ`Z=gZZw9^Ho8oA~6-Z|uG z@5t((+lL<$WW+lW=q0b(_8t4d*)L6h+{T03ugdT>9d5q+G5DM({$btEA8nx_|E_$i zf7$W!tPbkh*GraP<}c^d$Aeomr?wujRc-G44Dh8boyp0+&8?*!aJa2&o_M*EP6rzK z@^ub%QS3p`82cx|e|P!j@${!x9*07tooueLt=5U%LwZjphL@W9E)c*^%}IDsZRl&~ z!+0`#v>a=ymA#LyHJ5f~@!DKW=@h?YoMQU@rFgXsnR(pl;h`ik`y&Xgc5t}=J=Ki& zi1P48GBht`W%Axi#?OM6Ptz?`j-)c3BL4tj^JM3RbjOl&Kc2eOu1kku(-nW@z5ISu zlI6d1&M`lDk)@1;e{WG%kmVljlpTKQ+2g##*XNwWQt$s=DUP#UEi1X_Vf>_obnDzk zU{QMc>YJ6~@^Xi_>GZmzg~!EVY{Ik%kOQD4q(l5BiZ-sINwvrFT$#)`=AyK38JuDhnwnv^*7j%&`KHA(zr zje;mEgsV(tei*BBmR;HFm%Cgn*EEgsN>v&WzW3NzX;{hK7+OuM))EilcS~l{A;l|E z%LWpLokh<49y4R}8f%=dIi_4$ozEQ`oCudaQ2w6!y!N^}Us0I)VHW<2=?#mEf4+H5 zJl~DpD8eJumNe-pCJO7C#9Im*gbuE+gbSUw)P7oPWw8FD^urWPGOozq_PU&=qU%!k zuXfQp>XN^PZ-pM}y|=~68Ok}gm3j@L2G3m=dtGY0&ho%MWfa77pWk}KRvOt%n;snj zU9I9%Yj#iEY5pcF9iM0gO1kk5lkr<`A9}IU^QO|c`94L_U~1>~6;Qk+Qf*Vea8P;K z=ZpA;`Ly@^x}-2RTR^rShg%XmJf1+az;k(Co*JY!_TVu%ant;I4)md%40jtG%8QrNSADI)YL=vaSzwbzbt{jeQgdFeQt# z2f$v4ZWn&hWP+WxW0j5xJ z*0v~k03^Fl>NZ_24hx+BY3wb?3QTjzXgRDh+&2Mh*Dm&caRq}8j3TwLuX~gPhh*hr z1)8Z>5&lJS$&3vM>E8Ffx6=0%m*c4k+SO3#wMBVFjpK% z=R7S6J$|){#bYG%(+xukFg!3T zF(^2vDn@|^eDhLNmxbI8GN%4v`Yog<`|NIlC2z6d{sX1T%(Kj&*sJ|L-1L=QIvf9* zZ7B6tWJ^?7{3eWtl<}C`hq|1DPsz0Kdii(;-Lv}HH(t8;^=`01>Lae$vi858m444R zZ7y?7-jS_MO%RcK?09~vE`f&}$7ANO-8fa`U0ane>*DlB$+E{|oSgS{>FLnN#Y?*Y$)Y$!6v&znGj910E@5$qJpKsp9vIX2-I#qc`>U4dcc*An_bhM$DdDL}X z-RhGeZ*5uHb1!EX`I%?qjq^(O<{z5g?0VESkG*_;XG-I&{B+v;p|56upuj{3#>CgJ z_$O&B_6fsX=QhKQpCRkq^EuOXDZDc-4{tVD=YkIzgq5Y-m0ae}3H+%zEp~nonr1_{ z&8!g#9*PQyUSPd}Z9RKh9(h}M!Shyqr$j3@`bym5ot?T$RWQxudh?io=VZ5`CzcQa zvqLtebt~6yIoyBLhf?(Ikr%DNngrSlf&6+ONXHm{IJqJDXR`nH$M+G6jpM5xNAv;z zqW9@mnzjlz`1_73Q*(c3UQ^#ZY6F%Lpt=3`Lk9Pfyl*vvP1gXDlN#AcK~XT!@r`zaD4-7B%o5?0KY=nTY&yvC;<~Mkzf3Iz?bm-7W<#-7|V) z@s9?S1jA~g1v`9iDpAj}2aHDfTCgTM(G*#(>q~HE%nvA7o2`p2REROqabUGA`RMBx zZ8?Tm$9qB6@HQtepTlD*H@1pPEQ1G&O)m&+HTujm=orA%&zGz*FDMn7qODf~03C`S?) zA#^EoFGMx76~mv)^$3nxv&P#jUA_0CYgOLfaUfj!*D$JO9senEjL-fzqm5zLMebvr zd(N`_gvh;@Pnb2og-W?v<8>ENL~EICWtU54jRQ$1D;cdvLt@wIzYv^Jx6S7k@_Jj# zKH*QCyi$*xD9x4sc8-?Y=*PR)PEPDT2k$iL{5d=ua*M;q{71;2441Nam=pXXC&6B5$a5s*B?jU}rvawE&~Q{VlPR-5iiJX!7X;u0 z@h760$K$aOJPr729`7y(Ha!b|0(crLKF_@37H=2F zE5|+!ut%T>el)-n#fitpkNZO0iOd%B;O9t>IsSx2r?`0{!xWMWVa<|`;x3>8%oaq3 zETZ5NmI)L2*POePpnNI=7XZN!^K4CIHVFopuo_IbHwj^#)Id`D0T^TK-TxAJjci|Z zs`44xvyba?u#7y@?h!a2EyYM5PTs-kKjGpYC&5PL;C*tbeJJfkMfUGBc!+oklrhG@ zlz|VK)_I@7%p&cc05ELGE$|BvBQkV?-RAFuG?mirrx*o@440O#YJ`JSt)I&>N!A;3 z^I#FTLD$^9GJMy02{joREdvLIR@_+HqrULLKG@{u<+|4vf zulmUppt;nOSalBmmL<0zl}m|M96-Y5ZDiTI`DBvdBUl)fn%$4g8MAmbo|QeR1{+a( z_5ON}9066AbzO?C%lL~D&B($+B*pdBL)I;z3A8+#K>pr3H2eQ4>n*&Re&fIY5qod2 z1cQbVB1kwwNnr>mD(J94sUa!lq@+ZoC5K5f1p$>9f^>{-7~MKR8e|C4xclSxzR&Nz z&-wiW*SW59p66Vzd_03Vw)bFs1E>=WPzMSAip-IW&WOimZ1{4-GoY_YWR87k6dB4H z1?yxW8{c8NXA^e_>2HJ5S2#HCf;cQGF-vVwQS}dDm=B9D(s%d}>kPyJ2cnhCL83zy z_95(7$?=TGs^eq?KSwl;vkOh+x%pntn&pW4g8x4BE-CA3W!61BcxyN7`qlV5Xy6(C zVW0TrsC z{Yj7j2FQ<*^UMz#Mf*35>W$Bhi~<6KfTv76M0{FUUnlEQa=Kt;*0m_`ouE%fm7lIJ zXIbL`5(BdM5Y|RVjnO_@oyqJehxkWjJAD3!3Vt*W@xepfqW~|?EH=^1H#?vJP;LYr z5+=M})fYhyCUcb0vdq7=S&Evo=T%F!p`K0E!IIlj;<9Gm1S1?N1Qw z9M9`jhj;{m%^kqjm?CpJ;6|YEIs;CSf~;2 zAr~I*gPIsHR%^&pCMSg`h2~2svL*__CQ#*n)1zT;+y3JK#(c5CluL5>-j|SI%}Zw(D36+ zd>#V8CFe4I{}SJ86n}qcDH=Wx`i4n?Bkq4QN%{7ORBmzbjZL6bZUsp2FMaJ#(dx{@ zbyj%!myrc3C)F#B4$3ex<)Iot_(~fF{=9FTg;FLes0ZK%4NB%j zSqsx2{`m{^q*?9UM0JrF#EMvCkFFtIt9ksTv@*J;`dXRZLB-@mjXkb9JG%0@M!CFm z?aAoco(aebtS0WD_N73b-QyZgv$|UUy0kBK2UR6mD%W z1Q3$K9*%>Wk)e*{B6oaMnOTj~Vw7?~tNFOB=1bvdwdeitB@r}RZ)q!=Cq z{-0H3HYy980F~+Tdg0g%i)p@Z*r)=px6-V)cByY(X$*liIaC7=10ZZn4C`aTR+lc& zLn2^C%6cMOqtHbaDrya=uCj2cONX^hbv90$wM3-0cv4y1;H|EjAOx<}fzaw1)LL$K z{N7}1nqYlKOxg3UlA|xIt!8cZlYnDt3(*CXTirnl`0uA_FLG%&G3-3&(s@LyxK9QT z`*rV4)L3_x@UC_)e1`fF0SkPGCpG&l21Ly3_%|a5RfJq zUK9%R!$O&pjS{X#&noL56Ke0RfI}+5wxAxTCqwsYhAuA*&4PySVTNsu8h`r@XHC?} z;ZQ^U!#anQHaPNFZYL`P0ad*x`>F{gDV{ zRBRaQD&Rl_9-sjaG(f=j^d*j^=^}8vUYMj%c*rnKJbhpwQ5NU5<&JpI0UaBW!==8JGl}gr~zM4 zV*QIgzqDBDClDG#f|dr& zO_Aq*%tHOuAy)G9p_U7FT9A3>6J$`^d|m@<+67cTkux8=luzPFCUWGEIFj5r)A1bX zbPh6YDMNTU{TIha3}+sdvlMf6C5Xj;0Fi@U$suqS2Xc}LE9HSJl}OHNEeKfPHuezud~&rH-nnB%@EjX~Jv$T5E&IwJ&OFeN@El>E%`NwW^>5KE7WwTEBh@ z|Ej0M+Lm5kC}ZuqimJ!L2a^yq%=#GxWSH-|y4=~3i{k7=k)NQo@jcOR$_#dG2K=?H zi8t!oAfok{$Eq-D3sjaNe$F5aP5;U_^pfAZZ=IcWEnj+lgQ_QX_P0+`V$;CL5f1gF zP5RbDV&^4v*U3#qruOEf^#kv&n_gXYG;yX@%hH~W( z9g+WO(`*xQ`rMycqBh3*82*{iVfmll*MF|P6eNgjKhT!76cLZXa-gWtNbfpJ{lcFou!KJ)NckM+_PaoYJFti2n&PJNbvd46Lz>d|h*4UrpvcSro9 zV;_m?xYfpJ@6A1)c^|#U@nO%ma}W9;e_VY(6SW^FvX^yZe{|=hn?g)%#{MU(?l19I zQf>E3p$C~~4{WUA2Y7fh1u;b7I6zOzJ zIQ-9wP*VW{i+rtfs6llpb$Fk>b#2;(+-@Me=ulz zbKWrdoU8ZGQ@N%Jk*8xBjpf%Zu~t87r}UCYf8NG%Aar3emEbpzuAfqQ5bKsnbf~E zb)(E}q(-Ijt;NLo(#b0D!6i4IQVqA2)J`@?Px;2rf{7BAR~84}zs4NoMb$!%MyXvKu{dwrKevZ=}-#<>n`GjlULKElm?j_g`#Hn?b=1-In zXFJlal@&}?`D`vvb(R(Ws0(6)2^*GErW+%V$yj#%TVyARUrRTv_&V2_sO0>;tD^WH z-Z5I(s8YH0S?;YbmPfr?`ZfnXOJ^=2I%++a2J<)Hk-j||Dv+6(XbnB)>SRbJ>%N0$ zH2k1_17|E2@}FB_(C>sSNC>*QjxYB8_6JSvuq_RU?#qPNjF!4Gl5?*`paSXw=bmQX zKyiI9b!xSH$F9Cj4v9IxlkhIX>0o04e{R%R7ZG5<2tlcC|H-5W=c@50$fVf5OAFP{ zYMk&8%$}dt{kwYXL*%g!A-yr2d#@fu$eh338*p8DI+LqhtrhPqJ->)oUsM&3S30C$ zveH#4wa>c9U%PotbJ%nfuGVhKX7~Em&9PqWdW=#$mcw@zooIbVxmU9jE55+C)s#6+ zQYl@MOm$ZrxRJIs`6LH7w=_l~A2ns>rp=rx$G;Dk4_*AQ5jm2Z8B$PjKMRFf{*-yV zsAVZ9UH&-w|7}0uKy9GN|3BsPpA3M5KL+)oKu}xPxE`-x)*CV$tQ!`r3qNC(ve>OgIze)Pr=Q}Ys3Z16zV`WAPO=^OL zV0UE4ih&R@tTn7v2N4IRI`y8iyHp} zSGeI=A)Yl+fqhgE#L5mAx%f{8o)6*d!Z-r2dLtX0V$&HOi*AScHF`X|LRqt&s4lr+-Qv}q) zh_jvXXBiVJ!BdqiKIr~avHt}%o7A&ShrwOgE{XCO<^EBX-VpSqUApQjAtcsp!Gm0oU^6tyZM4s;t%T6Xct4Lg7N1oT*`)nBd z=59x;aHwFvyy#P!=z8^5{{%!hLXew7JfE%zR~^n)S}FZ8I-c*xt%HeE7cXjzGn=!2 zRj*i&o`n`Ydu<$E^=pRZtdK*Htb3cpz~e`?ZKd4N%HyNQzo^5nvc-l0A~BYhDAPt6 z9n_^D3G7mMJw(Lk_?m+7@AUUu{f06%mxQD?H9E0!8cPmcN;bEOyXU9#im&XfZPCHn zHZEWhyUF(v<4-0NJO_x{Lf<+%u{N(Vu#jX#xA{9J^zoIz?nbovBvH3N!4a~zhy{|o zrjVo_7UaQbgiw(2ji{Tic}6~lilIAOHut$|LZFK*vkDwDEnFHJs$#HvG+jaHV+bg2 zJ7s^A%GFn~=dfXtqP}-(1WEru=R^+>H_fE^1|yIM(no2J0O~Zxr~7 zjyMr~BgoVxW!ydy$|m0!E^iG}TQ!b2YpIJ|-sQn3TWH~z)PEoy8NJ59A<5c@qfr0 z|Jx>h`jU=|n9!j?xOS~|oW@L(rk{Dc&$cT1)R~R==UZ(Un6H7p{LPot*Ak~t#!&=w zM@iq36Z~t&u@8>32@jGlDWy?S_nwRE`W#-Cy`k9o97h$V!*jT=DH^!Q&hkcRHl7%M z817xUnZ}68;rogYYdx`JyqPT?HJW{;QkwKA>R76z z^n%(zMUPC|=7--|3$q5+U52%K8FSM)0vUWpD(BxPHxMxAt@yaZ-rByMeHfuD!f))s zPEB3B9jZR|qA$JlR;C|aN=y5*k==*K0z+g#c?uf;Dr#2b%mt}yGBf`@$5T`3goDu| zb>p7SGBQc&&Eh;=>A>_iOR&dx(X*kFOf12mkr- zU`y|jyEB@p3CPDyLz*6MK3Z(-<0D@VWC#5DY{lg7r6RxOg~}b*W=x0v zkFPypRsBN(3bwCba7U>9&IC(CBD2M_&|>P7zFJD3A8?1V2VZ+}C>rO(m@RK>hG-RR zAV_{$b#_RM(a60AM0M|~(UN+Y*G2`?jiGiV(vL>k9sk3|GHsuIj$4nth5> z@pASPle}RY&7Gnf@+Vlwno}<`#ZJ&aUhf5k%WHm(a%(G2-|2EtY12W*7&MtWu)gE& z3en8YR7BtUIp?SsA{f8@`(1vPu)?FzE}sh?8vTc{=T2vJyZd@*&%m3tQs2Y8YCQz^ zyPPgC)o59u76%n*BEPMpu-la_jq|IhD>wY==j(%1Fuz;1#n>N=mw&-s#l>%!wq4YW zDJ{uN<0;3pM;1g=Rck}qv`U^MTz8AhTaA9XpBCdH1%;l{O5kLBo-q8Z$V-!mKw4~$ z3+;MGyblsuwbsrWwO03*d~*b{*EnFIrI$ANnyy)xsOcB1P)pzSK{R{>Cg*y5T6)ud z-Jtl}t;dPj-6yWFD=c1o##HEyXV$MfAF#2}{Xi?nSsB_rPy8TU=EV;MiX9FP7qlG8 z)$)#r&&qq)Ht60WztE6#a8{(dYN+tGX%3^*D~;XTYf3 z0gvP{4^h+uosm{&#Thh2HHb^cKXwCza`#gQZadhZ7Rz#0i zgig`w$31I9!|fp;UOFEL9R?z^-Zl*r$Fd&OqtojETR^10S!Ho-%5(q|KQ`Ja- zE3ES_>q&gL9G*pnN#d$A=C3PaHC1J0oPaS>X#0%m>gM1Pr*g+uKR`)f$ zs@2yhkRtlMgPAoq>NiRBHwHCDK`PKuTn~6}HEX`V#K!toYX%$Bhl3A=pu!pvClcL1 z*rU@-SU1n33>8DpL5Z45sEi$7YeiPi*LHn4)Fwfcl)&R=95?S1w!()KW4YOn=D&-)IiGk;5=7Z`c=$1VM;NIwGMq+%qxllJ zITGL>Bu?^a@P3k@_n~I-ETSwtT?Fqnm6xjA@LCXD^2N*3NiR1e;C(^xNo?pGAJIQ4Mk@7?| zNtugB1wOAhRx$ zUxRNx`vZ~cCyFi*L>`*RXrB!A@P$=j(5nm-3&j<5J_u(6NfN@v$_Xkd#Ca2ZhOa_Z^cc ze>1W9kx$$;})B8)2ZnGgz2(v6H-_NImRPMbdz{{0|KL7Y(&R zb=kxsf0-m7wj-GpN!#yP)s)F++u=I7JV+k9%|1;!HE@^j{U+}HCJ6C^$dOD_^~UCS z69F&39Pi|u6PI&5q6`i_6h33&Od_0zi0q=lqOs5rG~|$J1Ag{zIX$yjNu%VJ^x%~= zZwBPGlk_tXtdD^j#$<7F#G&1@P@x~M-bc=`BHJ;jEi~!`Rp)j`U?V{^G3jGr`9}-C zXKkSDaZL8SPN;oUb`=5Eg?-OL%N`?g#1SB-IPh(f0g+ARi68idE#x2se4aly#EB)G z6aJzyH%axGj&jr39j=Sh|3Jy-LWDS#kj+(*tUwXYfoj!m^qF%Cu z?IGG{aUV>i4mXVcxgAmw3m7|;fCz+*XI=_1c=`Pg5Tb(t?Y^30LcWeUC=#1c_KE_$ zgCHj^r};8CYH6rp^nWQJvXcgVPJ}o!X$FcCy8ICiZXiGl2Ds7ze*#!JnX{3kbSTQV zv&`c+jXO*h{`Js=qL&4A{&#+LlMK+}U%fU6wg9PEF~A;JNFEw)piyzntipG? zq{dV%S>u8vzbI|YzUpC+e_)1dyWh%gjJdjlC92d!Un0n)$ zDpzXoG4uM9vQB1u;{09p!dSP+t~QAZHv0uM1mrbb-E6%t+i0lNX2WP^i)oAq zsCnnyvS=K-4r+?+YKaec@bOxGalk3}nA&BtR$i~xWX)#()CQBuhO++wn1C^?QY0W8 z2f52&FR>|aF>A{;ud&VRfOd5(Om|S=^<^=_39{`M5}V$n)~>)BZX9<0?rd*ZZIu%Q zYhYQ95kO)j&@mi`Cj*j9fcSiY48U8$UAiO9r6#-Dz6Uf-%l7PC?csOnsSYr%O>L!i z_1s;~h0whA{-E0(Ja9W!iuX&? zq8WJEY~Ytn-->zj)5G@tXb2l|n2pNfAjisy21gN`o)V-@v4Esejgc#IbVbW}E!}Rr zAC0Z$w0I7$8M27&NrDec3WC&J`@So^t`PqKp^ROnsG`SY{q3&jwAXU#Rv*gIP&^ z&fP5fN06Hgh+@=aN6p}N<=F7VWZL%R!8~H$0{JqPHnXbUzAqF#>8^Tg*6W8KCK!QQQ2wD9S%zEaHEFv9~dqcd1}ETnX(y;7;W1haon*Cy_tL_VRz2 z#0o#VF!dZk?@*^JB-cU$>1(eY;NJnHJn#lnbL@N12}_784lu((9+CiK!oSuX0Si2$ zf^F%59U8bzC^(%y6F+(dRn~7iKa-pLL*Z$YQsDecm)U6rX~$!k8nshezpf+P0MzxL zGu@+pbjTeyC-pEu)DqxKcW8m-D@4t#Gs!b3(EKrp`2pSeS?hU)K;eg4ON;gMV?imt zq&aK!LctN(jd9&8Z~;X{`Ud_~YywB&!JZg!6b7(F&ycS#J|HYw2Q5bREK1KT9>)H# zL(XzqEDgpkx%Diie;?maUcM+_lYM=<(x{q11gxpO)*!$P0|eY%6bW3ZCoNhbeX5a5 zYnY{a`EgF8RT5)9EpXK(WfkYP&}q4x^LMpTU`-);PD^9OFMBQG``Vc}vBSQ_s6C|3 zuayMk>Wu8_x$CHQeqoLK^m{=-hsMV7!s@HD#qknnV{hwcP*?99LIZ*T8+@6mx&D@W zal&nVTCzh9dYbtQ#ZtGz#do4?A20w0}V#)&G@iCrTm=O{Gl<;KR*Ir@C#Y zM^%I%HBS&=Puo!3-MA?M6i$qLr8s=Ps~@<&LtC!T2f>pf{?N5H&8;@;0yo`Uh6XQe z8yd(uWNf=w?Uf2BCW`NPYNNcFfKQ!4Ud@ghV<(Hb;|j{ZO$LW#_--fvng0IgXF6kE zzQ@g&`QkP@4e`u!6i9xLBnN@eTh^mB>+C-2xh+u@CX5Y9AKQNl6?H zKRX&m<^TH4X~~Qei>UikQ?Lu^-IY$q0d;AXOqj9rG`B9f-}+Z^}?>t%KQg z9`>w5mDP_x&%v-0viI2kc?lBUoQZ(eaBhRfqac6(9&Ntcn9+!v)3p58erO{}%9lDX z7};sOl(TplN-IcJd3)A%oP7raQhM0`pxMvs^N{oh{n(S|^p!v5-pWo5R{VHnpfK`C zSaX-Bd${-R&2wt|%Xd~yi*l5|U7T1RDStA1mr$M9H=`Z5uKBjgcR;W1nbe*BE&)~B zuB3Bd5dh+rqwLxhd?9f4+Hc#AS`0)toNsKtt4+e;m&*FGHZe)$$Hlb!f1HNPyw_gB z;k+d-tIQVbxoBhYsW=hTtHYo^%Z@ju`rEqc0b6xnx?=?(zFv}5lY-qAh0aI&-oGZt zie}SQap#GGDNOr7;jHtn>$T0VAcUIpTW7nU2|PKla;p&VpY86fNWdDzt5Pprm9L*< zWv4%1`t0vewJNq+(6CYqyHVsqVY!@a=O}PG#pjYlQ4{y3u0r$UL~BU_uGwg3B}e!F^Eh__$-XV0$idvH{-7Wr<<{3Lr!zdc+vl+a#jBG4mhR=s1X)<6 z>Rc3qK&b;aPb)R+DH)v`Dj^%epcsOoL|^r11C8~jTZZuMCUP+7YA){TsUpM8VA!Ox zGSgUv$4N7#)qRyKBGpZEKV0ZF{&UV96N@(v z>7q(^Ha>%FYY@rvS-wFd={|Fww_H@a9^SI7t6wiAyz}bOU%)uVoxXSZ^_{zZi$%1C z^@2BEW`3-8=gl7%ih)Et*6ul!B`ErxsBo(bDZB40?OHJUc)eNWhuvkub>RmeM1dqb zK(ajNdD1<=I_#MJu8E^x$9xKH>*J`J$A+;9_9yx_x*|#&Q2W8PpU`Uc;JUDVqTOxD z-9+oxqN?^uknnk3M~Q5epZ=Nkszptf%c_p=^Y=$44lLKGttDC1EZB$f0xzAteIu|9c@hhrTPcqq1@r7-IF48*d zQC(HGIex*=l9d%&28oKPKRO_LEk5K-qB6`B4*zKCLOtp>Z zaP`Gjc)N9kxabvHVC}57t)09FOQ^L6pen8}DCPy1hO!ltE9JLn5!$QPf-}VW&2M77 zM|8|=o|l|tN3VH5;nD*M&xW7Uj$yK%qScx_ji8qVgJFErpDt|pS&Sb10p5atsQi*z z+w;gU8oA*0qLySZK2-MQjM~?b267Sh&bjmHx;{Ed;v>rm+a7cH*F!tF$H{Z2|q&|)xHcjfqp627`X_je^>q5d9J<(tngcCtm;HO&ed1VM<`l6(A!|ny;#P>-|Pf6KE zG@umo82&>>SGdchQ_bxc%cN#H$8)~_QctC56lS5ER?#zGqGY=sbNa&E!y5rTvjY2D zmO}d+G|1Ru%KUv$b=owX09xd!#p5gc z-@v^-uIwO-pyJzXDRzyYzIb*FtQ{WDeS6HQyw>4%<(W61uAksx-p)}U`+Nq&!of=g z@lt|}U36f!XqkIMFS3fWz4Be3I-4&=<69n1agIFe_)}J=KP4tCX4W#~ zeW$EB-^f$r_V;0(^dBFciAm=sKlRs_Cw0q06emEsKA2M#RAjveO*(&Q}axUe*KXPn5&ZAKWS38-&|-vK)XA9&l+V%3V97pwJ3En`Nr9@ z!b`(^hcHh(;AmNLdFRR&73qek{JX_Xg3S(?w(%10Y_PTBkM zw0MTuhd2vbZ+gR8#M306$TvEDig$Y8ReaC918X{)869!5-c17kq_Du!B zPVvVyQdzQalDIl|RwPHH>V#5wyNkVY#$KASk4|KKb(ma?7uoY9;|^nThRsB+-ekM@ zEt^o84iCR$g#Fn^?dG({?E}vX|GEe)qQiheewO0~At#4HK>1silGO37*G%bf>>-QS z)4!WOqUM$y$HoVFIEoWZEbJdw@DFF43K5f<8d7$#K5j)|D@@s2F#=CCcT0HlwosPw z8a{(|GJlHeW0oHqRSmfk2q(Vc!-5T_QKp0tu`Em&Q!d~+98giMcyCU)zLL~SzP|l7 zY0A#GSD>$+d&Yvt!ZqM5{>NcU@QHK+*!9tSPRrb22a%w8LQ>U$|6k>p^&p7xSupSBA zgXQpv3VR^KN~-##aFEZ{SJoS`4#R<%!kFqI=KJ68TC3-e0_S=%#CXXI?yhM^F7Y3e z?-^d0<%_o`nyN&M5X!GM6mk#GQ!3Tgsh1%0)1sHLtRXyl%om%q3mebl*8>@igVsAQ zEK=D7+H=-^8%-TP*3sU-tWmsN#ku+=7y=3jF{^;kYrimRzkswi#I-kNLbs~OigY$D zgHPN3b-RrrdtbEuE7Uh%qWX(dArwxV`)|+sfGQMmkCQ?peLyGzB(y#BifV>7nF{M* zgB=f9TdE)4(;0c51`7?fAfOJ--f&4=_z4X~7UWCTF$fA3G^|sUV(S`ckb{Su zcghsgYR5sKa@6zCiU+kzk$kL!-BH${W^ zjFn5cMky1i|p}RprDRp2z^nK6y&N*MO*6;^wc;;St1{6a+va02Bs5;Q$l~Kv4h`4L|`96a<1oKu{P63I{=vASen1MT4LK1PX#c zArL4G0)<1MNC*@KfubQ$00srYpb!`o27|(3P$UeBfjL?&`=N> z3PD3*Xeb;FMWUf7G!%`70ssO8ARqt&1|Z-70tp~c00Ipl01yHMLO?(W7zhCeA&?*h z3WPv|5C8-Lf*>Fe1Pp?JLl8&^0tG>!AqW750KpIt7y<@Ez+ng^41t0n&@co5M}Xi6 z2pj=}Bj9iZ5{^K@5okC9Kq5d$1O$nIArWvS0*OSRkO(vq0iX~d6as=mz)%P{3V}o+ zP$&c%g#ge95E=nNBVcF*9F0Jt5hyeQjYa?f1_WRr00stN-~a{*U{C-C4PXEe1_Z)D zKo}SZ0|#M{APfqGL4z;=1OtL#AP@`;f`LOYNC*Z6!Jr`+0EPjt+2&=?RJ13_b8Xbc>UL837zGzN{vaA8PvR16;8SJYLJ5|hBe zx)v1I4_sUvfCCWuT^zs51R%r!iiEV$Sh$Roctqq{(H^UPq=dgqL76)^mPXPdH5$w_ z05FTjvZ^-}52Xk^L_8X5C>hC+cszW^V~7(zu@=Iun4N0wo(W=qZD8*a>{ zbGxkA!))?+X)CE#kCeV}8Xi^Hr?T3>F7##Sfp|mra4L`571ZQOuSge~@5Qgglg?s# zoE7sA?M*H*h4snJ=VSIEW?wy)+sm?j-Im`x+i#8;cZF5Cs&l;m+1~bjTQ9?5(IutV zsd!lB;XQo`4aZ$g$#4uHs_;gCXXjI;_>V-Hi&B}ag+7b+tBGG{-#PGZPj{TGU3H$1 z1>n-}x zMm6&~^m#9ymF4qe`*jQH;nUTZckbe4*yq*{eh!@`%c4zyu$#r${((QElnEYEfw zEN<~_YSLTBMB$m_nA<%JV-Yja#n%zr=8P;(GVC#wr#tSUemK2t|-0`F&I zz08?4tZyZmsj;;KnAp#V8w1e>$yy(VJ}byu6Qgd2aP#if6-m*}g0Zc8De*5EO>!d$ zmk`Cs$;r>QMT}oG@O1F+1gB7ukG9j+Z}>27IWptYDd}<>WYT6Q^f9e?wu`+}_Qr+z zexfo;+y5G%{ev2hFiPW$!h{6x4XB7kiEyr?(0Yf{)xga=F%M(4cZj7={Ojxcv7}ca ze#t7w5AN~s(1vF7r|$@ja1V$S$9d{F?B8pej{4k5;K!1sqqZ{4HK^P|FHyh=FHy0B z#BCm6sfh6+`$#O_aHXE)`aGbM@z%^9FsncCoUmQ?7Sq41VzG{7Dz0oO)NK*W??{+_ zW1yoJZr@|5W`H@;>9!-#F+ys*bfC92rkxQ%#wiKszDsr1L|*UWTYyGRp6=)I zT^l+T+}zx>^3R&8WIRpdMk64?3Y)kW#TQN=@Y0R9noA-FGbldIGQVVcH^`(Lx%S5M z{j=c3uOEo&^N621U0qoX`&9P+xUaij+QK@h2ot<2pnb$jAa`|Z9CTHb7-YE7@U*(z zhr8YULx61c-YTiDYGK}&fwLZJQ;-Z6dKn;+RU%g!ygfH!QKb90j~SaYRIZq!LpAC( z`&N`&V8I*#MwU^DVs9*xJf}e4P@sA(rY?*!^z**4%scd*VFvP7nUUoJl?7aqF`HD$ z&yF&>=_(R#X zj60T_Bs5~sbtjJZk&e$%7_A%Kw`7+gTT__|bs|WABAj&dDKmJOkJMc<_7^xT>^AoXpH3EWphyc zCHNV{`#JCPhyxfC!Z=_43f_2~o$RRpQjv@k^J`lvv9$RA^_#yCdhD3v&knsK!zB*CXA9R;y5ycueN>tH@;pKg zqj>Y`Ns$iCa`|-Hpfq-H^znf$CuV94^06w64z@8&F4tp&ccNkDXC<_+Riw3>nO!1c zFOo7l=}z95vB@mJ&pr4_UH?gRS^T1QM$4pWmlvWjSw+&O>LZTxu@=Ai*6ie57p(NS z+DMdBUR)(^4Z&82IS!@iSRs3N*QX)L6=pB|W_CoZnZdT^{5qZ^pqQj;gGDukccNEXiSO zeyR1CVwtBEs8`{EQB8A#D}LPjcc=sUY^5@t(?{ZUf^+!&MZPb6tR!CCcXwz|7}7b_ zp*auR5AY&d1`%6(OIpLphVi;v_-BV{iJ)}Nq5F|pFcEX?@Xd|ak?v0QHZ_~hfY zk;o0cVOFs@{66WYBGp}(7x%3meBOU)lL!L(PK_`>wr;tfnMtF`?ELOCzq|5l{=Rs~ z@9}o2QOdv6>-KyIA--#?{I5c~DKc>ZpLR@5T}QTpOex1*LwrjWuh1JpX1UBKN+Vf^=j4{`akONo`4vi42$n*o~i4 zR-?dGQ6eBT?L|UzfI%uQ?o1GUCMi85VR8~17NOM6HvT~dY{CIX08v+;cxHxVO*RVB zejlptX?+H`LEyA8D=+Rl++o`hXFVYp_r<=ugC8&v_RjEyq3?@AtdR0;-*6+Z7_bLY zGVp9XeBMXVA~mc(^lnQqY)Iy*8{F;XOc1;!v`;NS-_HM(K|nIorzaR!%{pYZ^oGBE z#O9=8kS_lFY5Z;={IJ=>Z^qL&<2EERoG3ZqqBoo_lN4zkM&J+%0!0;KLL)dYu*Hpt z#b@x>d~i&jhFa}BQzCfYHxv38<)a5?Ne0|{IEhYI9=wEzDV<`ANjxRwb|o&4sc+#5 z-C=n7_Jw(|>w^o5!}0Lmd><$>)=2IyQqs+C1^??gg==v9i(~n34spG@&)=%?T+G~7 zrxfPMARL26^rQGCzme3eVB2%X+}XjefqNHh2fvD-ZMk5ljs{yN2x>6Cpg5?6JHfkBW>J}VSIE7U!S zCtoTp$kwOY@b)-ke57ys)%_qmLRXI;uqV4P0>4PbrsO3^`c|PMYjI?tQ%I0gY-pS7 z;W;*!oLh7jEXpZ!o8ISENrtk&&m-FuoHExHqpT%*%0HZrhYMf5NZ=)!AG;$#Ih411RZ#XXu))#gI9EaZjgVX8JvT*y}NU5iL3qy@% zW3RzsTS!4<`9#qlPB&P4U#eRqbg9;a02g8hsV@%G@qI{?u8buMZmOJZl|)vSm?buo9wMQ z)v(>F-20_k=%?q>tJe?Rd4E!Pn$6@wM_w}fG}_nu3XUylKZBVq^(0PjxyZvTCVNv@ zc!}TmZ#N7VZC9Di%USHAe3ke!<=*&a6^M3WE_JRz_EGaK8J1)9V!zH+rry5p)+_F5 zabF~;d%)p{{cih|-%8^rnIorq5L@{QWL`5I&MAxD&4bj zyM^$xUE+DCKVKp~BNsgR&fczhnAEo)J+Cdl`IVc8^0k;JDmMIB$Fe{Unioy@4}q*z zPX?7^n8{;y9a^;N*Mvn(tyN+vW8@Kb-ixjev8c!-#(Na1`W$J3YE2ENQ>b+y#p;{*MBS-Y0V2|IegIY1x~j-dOq}tebsHB>K4Yw*raTG?3-}xvJL+WS-#o zRHo~6mHqSUOm~P86}yuJ4DQV}3!LNs{6xuhKb6!SZ$M?F z#a*c>xI$@kSD1v5l#}}LU@%tI@_RU*1sSW{Do&^!j>rHdw&eCZy%kmVgmfvpm?%Xn z^;|~f$UuT;$$1LWb={wcvIfak9>i_G&3*W^RQtKp=6o8Jl)!~BF}{djuf0Tkf&Ane zDr{~`d&*)yAaq4XaH?$mv}w+p{{0s`0XnPo z%co<{$xkC0>94}^a&5NotP+E1+oiLdLSMXxJO7MGd-v22Dqy3@@X9A!(S!|*Yls8g zjuWTCU5UfbvE!O{!;de2tj?%YqWc1ib2E{U`P2^<=&@2;CEyf-=3asVx^a z?$!?_T|X_KT|GmlE4-w85cBQLEY_wHt_%@kMgq-JI5`jq)39}3j6E; z{8w8_%-;le3N*CoOOm5yOeZb;SL5L|N_GN}TfKAe+BTV2Syk~ibK`Y5tHWjuqs+yyI=bd5 z7dTBtdy22QD6R++Nj0)oqc5fjZy3`UJj}u@Y~?EC3H2`_uZwTw_VM({UYmowT_SBT zXj0sMb?+t~4Vi?V3Op-K7~Xe*??)j)sYa^B*{kI!e*f0uWIq<`Q&Oyk0;=IhA zd1x>$w`czl$i_-pTWs6$E$?6qF10lE9zvuZSxxLkLZf|K zHyC{5Ddxas$mO2>1MVKsQx;0Yec$r{u&wB-)^@D;_rJKyCUub@2 zd{H}W)OT6d<_r646B-5UVJZTeyn@xvPd&MK8*={deZSnP**o;i>t9tQ$i;1y6MPYl zrMj^Jd{UFBV+igc*rAo6!{CIE-N&KkZI!3!M`E?7d_AFV3IFPPn%-XZ<*j<_BdBtV z0hRmD-5Wjq$b&e!)qCoIGC;G+rlP29ah{J&!w3$?M-l5>(&*}t`iV=@X+kUFI3sGB z%iWnOJx2;ta{5bp$&zhfkO^`5Tp3i2R1ZI*1aUP?m-VSK5&Pd9Eg<(Dy#K1Tj@xO! zfbuYEC#8aR91MAsG3EhLR+FrcktZ%x?Bh`pcp2?RSddSbbt_ zgUZd@4{)DQhvM%-O2&QIWi= zR}}TU8hyqh5^UkGP$FCP;Y2h&5j2_G2evY7)TA$QasS%^#Q{!W4Uqr89q{64Mk$W= zpAJa&Q&UdrF9(c|mIRNau^s+)K>OfiS%NiW#WC--1M*yvX#DQMw}_TVIB6=b9gxj5 z=cmga__bc6Gg<6+3l>?o(N@p4rYe)dprsDE`sPEDDu?x9vfZtSm}c15=ATVoHOt*h zqBROh=kKzbp<^<5qsD91GBA>5JD079o_8jzGE0v<%6smmP31BEs{F>(hiw+HHTv_b zsBDVm!p~FrZ*jTRPcfpM&Dy|nx9RHCa}P+i8mU5?Y;g zWhst9tdsyl_<5ZEbbm*?LG=5;k80Wdceh~~U4es^H5QpfnD-x8j5wNXmV;SDp;OO( zrL7LYaM^t6?qgF66sU=sG$r}d*zb(%lSeAQj^xw_ErVYO(ppCg&>O79f&#p3VkM@- zt>dNpPOKB;bPU*(#J6_V6Tx%gCU9kXrwy&!>=hfy%6&JvQ*`m+wy8J2$nmB@ePqT{ zjflfZ(*rJUE4;sL$0ZWaevk|a-M67F&s=jpZ(me)DyVW z{(*9mr6PWHr9ifv``+lSPX&FigIvm~Kt8&(6xZ$6{nCQQ*I^FT=$?9&DDMU*`PzZ;_Y6go ze7{#Al_qpkk8^8y{dFMwgp4ohWz+t$C8o=35jRcR!hbXCC^V#|+$ZVD$j$2>A2YPE z>`JrwZaD%8m0k6vQVeJE=MOht2@#j}T67?kh++*J<>j8Tcu?K68f5L6#t|u}FUNE&QT5)AP5JpP){F)#hyYwxi0CorYGF8kD*dNZ?m_9%n+T#3EzL^ zImY4EvC`*{t9X(HoPm;ZJhj;~Msx0lDi=iFao#NpynVI{H{e}diwwbE`ZmG_V)mor z{ECMiO^eOTH0o)sF_fEeXX#?Q%;xfaSw79>#DKpW;dmqvlv=Pw?z#OgoV* zc2--eB;O0^rxap(oIE8I0d77Gz!%}x%gNEir8e3;9!ZiddN3^N$Yw$ruYoIN%K61u zg!09T=g$Z6FCq?Zn9G+5+bPG2wb!w@3&=0Dru&;(j=5cmcxxHnTHN}i^^1_M)7baL z=4Y#4F5W8r)-rS{j+z^*4USDWeNY}j3(3x2Zx#w5Dj_#L7M^b#E}YD;)|YWUbF#ef zS&rua(zF&Y>o&3aQZ0{n1Fq@zV>8o%-9)(A0c+51n&#`#@p9M`4B^QY7Z{>8BcjyV23HECVA6ugZj zxJOT`4!4Db^AQvqFx9UQ{Ml~oB(D!;2@urS?ogCLqQEp?Sp6ujy5(O7gepVxcNuH3g__Y)otLsOLe z$%FAgg1TnkTBK5kB4&4~^Vvz?%O>|l8uGTL-tBE5~L(K^+vi1uS4&|+PzJ5plD4Q4G(DtH` zAmk&w2CcfY5e2qeGAmy3G4{HTYH2-sB3YTOlYB`ZxwtbIUNLu!v^{uf^y$f6Mo)A(}Pw zGF4(oQXs^}j?W~-$06lD6bU!Wkhzix_8?$Q+Q7nx?eyH%+49#@2}tFGRRLfXjo*%mDN1?)c26as7$_qO{T`U-v&Rc^fuZw#Dc*+hw z>>;}L0)+U&x47Q&F3Q)A^`^u|A)jUqC^_xFy|M_s9Xi7C?p*4Gn8t^hOScEN>=9)P zU$bQJNHHMTVuScRQJ1vS&*7oPDjR?@{Nq!oBUL-i6h6@Z`{GKNmX{uj#OgTnTDDjufW%zQta! zPIjwJX@(ao`ZT6P-sOZYyG2*;T&o8cM@#zX<#yk;N$Qgl4$DXu;`0OV!M3;=R#nmF z)cq65yIw&SyCf@DJrV7;OvS-TA=;OWWt1l&xH4O&R?0p6J*mBFd=WAEkJ$&i*AGB2 zI#O%HyAL^AFI4&5J@a9Gylm0t8`j*Z4~S!w?#C-12vNtIUhQnwGybfLB19KZIz>VH zxtWin`yN^c!w37L&?Pat@!To^Txt>ZSoZuvjlQv*g;?V_j=(8Gv>!4c`KmtNu_}Tw zBbij9{WyVAu@6|`RSOwRdWn-%#GcsK7;-`tToYQvdFvrkk@*`uT5E&r*9}Ume|4X~ z60_1@?n6q*>0zZ9ij774uZj8ZMjw3)yidf#khRKE|Hp1MNBui!C{mUw7PeFSCo#(` z{P>n}zewq?-Rh^mcB@T$4-182sGE!=93F&NRL8Q7eA^GMrEs2mqxLNiQfxmooICO` zf5Ni+iqpKcCI5W}cN;n6+8V5T9cVt8-Sn$2$;OMpXg&uTv2-KmyA#L)yvK7TRNLUNCQq;JY% z`@>ih|B8NhX8jzC+lrml^P0BQl-oEc9t(*_SX8S)8z>cxL#zDO?V8L0HR|x?L@8qR>g2Y z{9O4*U!HV{Tyg|XwATX(^+;}%+wy;=cdIpnQ>l&|C=v)|>GX|H(yeNMuYlh68 zPz| z?I6GTu&)82C5C`}t?6+R^y#CGVHHtp-=0(9{?WRcr3Uc4&Kn5gCLkp*7VpZ`INMuHE{tj}6b9UcP zHJs+^#%AQ>E`O+*xXP2wn308x9V*2SpLHZggi{)6>)0FiZO3ry=UqqA8V`qK*lU#g zDnV}!zVO~PUaPL+j2rdbA zeDJ@GVmSc3zoFt^W1!YuVs;eP*0?)NB` zrm#EY9-cfcqjb-BtT+MR;m~zir^u2lAzP5#@8Rl2cNd47_TRl$iLkHQYpR-UB)Em> z^xa>qedaD|bE($i1~fQ+--fL)A=*EBHwf@pSzXJ$SR=$E8lp_~N0x$&0cdPb`&w9TUt2krtlL-6ktG493DOJOhjg*yvasiDY)T zOx+bc!m(66HiF&MB;d0lgj{^OQ5*xUP`!30ur(m_tr^!I%W4oFN4&$$h~p=O z`ABp1b;MgA{_9qa-}k-6k(S!!?qCXRDy>XT)sQP}VS7Kn`t|ogEG&G4zmNRir{@HS zusHus;8=4^=qNZ>Ulb+YRqVg)8j9X0G&7UiLK@aP6aR7gSYkH$=YO9*lvb{Bd+;^u zCEcBhKLJ%zh*Tmae>8@jNhf?*y5jouagle7CwJ8fO|DP>=1GN7GG-Ft8tb{%`K)DS zDO;oY`t*1$0h?{jl>;f~ zp3=u_wF{{f;IZ0H!@xv4Gj(PIUK;JvW?|jvcM2Z1r&u_dY#njh9Tz&Ev|Q8zqsQS(QIsG5{W#H?fN1mHA2^ncg-$(PZcz<;|ee zrWM?@8RlYG1K-4!QMA4$sgLl7ky*4*1C!|DWE^k&(oC2sH2J|Ds5h_jN4z@wY`<8# zwJiU$tV{*42~)Wx^VH6EfxLJ6mdX;>-cAu*RcN=^`onBeNvi#|!0VfA;>z*^g`CTC z(w+Az>bNA~2<&?1;7Y^*QF}&y`ShMX_&+vL|DYNiTmS%DDTWbH5&kb!^IIJeXJDNZ zhG5&l#Q)T`b}EK_%L_8O`Z4>rwuw*p?-wcmhqj@K3Yuc*A#ZELVxwzyRDQSo7}Sk@ zfUDVDZB_xAd6oMHdN}{aFQ)YLR*SfK2<$sC`>52uLT%uOrC$t2huv?(c{HzDWx<>M z_CX)2@3b{KyaZi-FwgOmTp25WcwfuqbM52CLY&IH!ygWh2JD;q*jGbDrtb|L`K~^Kgc>S0o zS$%d2Y%Qlw4GlR0)7e$%yHeNg^XD0_`N~vBQ+YCXUC@WW{&BxB;qlN8mHu4~2&;M& zTQg50-CjSp@?E!nW?wO9TIwW-4DYIn_LEdz<$@T4&4=gVCKKt5xlqe=>r#`A49y=Y zxSZ~MPl_K%DyhxHM(Xt{(UKqOIfBJniHm5X3qpXD@Ox*e#H7#O8Sc!zyK;^f>`gZf@EClJ?=eEM1C>X7R!8h+6i44l!yBHr-Crnl5?tBHC)rFtM zNynZsK6g^rW?AUDO=NU9(Tt5p6|69PGVD44K|sF0!R9LKqw+y1soIIAwv{bq!5aH= z>#^&bJ2k4^9=*Hk=Mp`*BcUx`nY1AuY=e(Ayqo+~&(Ys>=(ME4xHu|vrv9q1%hXnH zZhf@ge&t&vAJoCKEGdbEtO9x{(n*f4M)0ffC3I4BQy*%WgDk z3d27;kb2BH{u~KhYIEpMl?pyBQVW-L`rU!HbutJp$!LWu0@C#H^ z<)s1r5CaN(@82CLlI0#BS-!#U%ykFSK);_ZR4Z2sLF8)sip{@q6|`t#z=KeOV-2av zOk3(!GpvJFJ1bFE#u2iYKX6prr0T4CKt>9qZBGY7^BKh51ZJE8+h8%XuK4h(-QK*% zBlypBk%#YzT-i(FZ#8WypDkJBUVC`YB<$;AmrI?r(lhm7-iHdgviaapOO|Cf^p?%L z!rmIWl{97k$r30n`dwQh^Sf3pRIQ8^crGw#X-F5ib^~A^I;?J_?PeR;bCi0 zo}-GG?IL?+R>BJs=q>Jt%@$cn%#`421!ZF4tVBCx-CXrC3E`)F?7F(8%h?O!`s|J1 zwKjOhzZGq0LAP~t!tO?L+Mu0!$5~vsJfFW|#P`c08)aejwK#M#9kMEBO*R-Av+$TNY|w0S54N*cKCwBSXbCq-YMOe##G~AEwK3 zce7vijFcfx$)=2mDKEN)l--g{`{CTDG}?&YkRs1%cD7KG^Q5G7ucExFc@JT#2x6iY zR%(>Or7f%7LhRQxT)yA0WhLXQQv6r7SOHYPTR`&v+HA;l@c*l)|G&I)>uThG;T35? zm9Vo{8m>Qt*LA{-%ACoLq;03G%-$l<1cp`mL3nSJ(DEK-^MKt4_C=E>L{BUI@><&d z$@=`8=Ou=nzF!SjpLvTXo14#0PR{9*NvFNx z&1NMy7tUn#NL!bQhOFaZ;r$zvMn&4?1(A;xRZD$zlkcw~7m`{F)OZ5OWt=wCZY*XR znLV-*d}yc@NL!j=o9?ug?ech(P z|KGa4P?_(4K~OFk@Q(Pu{z+vptA7wQ^bC6kt&S|$OD4r9Cp#Kh?hfkC+hc7%DZhp| zA~Dxl{k?l-4{;Vq#-%D}8&%5!%zx2dBk0W-))V5h<<4eYGv?rt7Wct4{Gz*bq*Im4 zZ=)6X=*rVY=j&s5G}8%HjC0l`jf-uJ9h~7?>7~M&*zwT-KKZ) zL#q@7R@IMq4OU&+w2f2E2^R}jqd2(ya@{%8QF_Ms1&(X6BF%o#CueN}kF5ymjp&1K zYzhNJn<~Q1 z1mJU9QDUb%hQd|EMc+A2XIRbA7T;sE!V1l`f)Qaea-FQ(xZdEdh{y`JcM_myx)YP0 zEr#V>EN)^255a$A2UkgFvXZpOk@|VV@pZTIk)}A4Rg4Ou>yth1Vh3lqN+En{Hmd@` z$M=3J-bb}evGZ{-A;9$z!3pa@dLI?xh`X4JcGrDAF-B+D3z^4QQ z5%(%yf~uNwu$b^lx;`g6;hWMr={IdO4@@O9XU;xIViIcC6PBZ6albF;77OoY?fv!D zR-IT2Ud@^CtS%8Pg!ZtMAUCK7|8(9s>QBK#i@jHHBmEYL%Wr!$XcAIdK8#Zz-w+&3 zO53Phjne=AL1sJt`$#R;Q}`G@)`!MtHwc220Wf0%a#D>X75Ry%qU`f{Hl6lqzmjG3 z>(6iT*(m=#vXEl^Kdae4feS~H7+p3|_z$&js0Cr^GAmb17XN$Tl1A09fBbD=t~)n) z6cbdUt_{pfaG+TG8cLnNIRP(?3-o_y^4>zjMw?TO=BgP>NG(2C_>5z|-hO=;kNC`a zzS((uZ}HjpUGt9TI3zWuM_(6vqiGWRpBY!Ky@%5=+q1Sc*oCJwS}Z+kD>JWuee<^c zMck_6$8t-b=h>&OkV?ykBOkM}9$RFTIqV0r{roz*@Fwip$vjHHdUY_7M}&>W@7p%{ z+oikC&m~XxLq6FoUyXks*!##EiP_f4lRx~n)AsVg*wWU|v+~d8xpZG~^(z2z=`c-9XCp8EQo=AAv$GW~FzJ z6!~A4bE}JJZ?D8k@bz)UONV|g1m9Sb8;`nqR$idkeX$+~mt{w-hl$`-sL(Sj7raYQ zQx(*W#k!zY#5V~Jjz~8q7u?LajqkyW&twIJW*J+l^JemB$d!Pva+t93y`0s{h&!u> zN~3=fAtSOl2i5uUpW4?sB#h2}kE7)b4^h0QXdnD8pOJNL{rQ!nwY_+k@~Y-NyWydt z62kVZ3|eFR{0RIjbTV(23EZXKCh!W)gpGTJe5I20G?G)e5GzMeu(Hcm(EEEMu7af=5xGlEj&w8lra?eWL+cTpStr%g#yh)otNG>t=aG`_Vx5BTa!1 z8ATlzThGknboYe0Ij`9*95foJ@f z+4;fxzw{?E)KEH*^*iwYpf1tQzVo|*p5&{>-7};A9f}5vnJ??Teutt+nI!Da;U*Vo zy70}Vd4Z`Fg<8i?Ted%h!Sle)so`9fPaKcRJ$LsONu<^dx@CV5vy)DUjdz682+Nd1 zh1cG}x%fp{+8Q?F({J#*#ZcFnf5^Z8hV8--=~Sxmpwi+)d(-|*9aZ{E?QG}0xz?xq zXFuCkOu|}xNi5~wSb2uLj}pFz^0)oE*`6{IsiVL(ywr_jTl?{H<>>%Lm@g)A@p*@T zU=DDdkp}s5Kn1Bbmg?bDz7m9w|kLL>dpU@a{ER0zs|E*wJD=7g0<& zdB9`EwRZx{VJ)p#g+yWtHu)^HU7WUV?+yhDn9f^Q4TW*9Eyfmeq`8Nv!zC{^~v|SMKdE8i_YgbVXhT3`?F1xp9xu{-t<%|lx{qpfS^jv*dHiMN?N^Mspy_juXLy%n9kRNYiv`#rSw^hM z)f1K%_s8yWtV~x+xuseYJj1af=P}0N*1FP1!k6RGN8%3h%eUE~>bGF>z2n$N zLKCO4Y@m4A@%Vj5aArYVL&}=G4csXm= zAoRS$3P121@2HSOa+rwenLBD%mE|V86MqbT)(*dUH*!*mMGW1S&%z@%%#ecJ7WJBq ztbJ8Au)EbkFyZH-L!xQnOXrM$J(Da8paQ|%zNb__tie!Ow-qZ~)11urxK_t{ZmEs} z?_X{kom|U>@%jZ2-u*2Xfg_IIkSp^4k>LIpCYOiV#g&YtOIQvgr#?c#e?JoYuTXdZ zaQ%Jrgm&eB4}}y#)~pxqMFR={c;rj$+X(5uF#rCNW)`L0=hvl9!)sQ&lShv3 zzt`am2^A=0dfC&z0odOUF- zT~IZhR=wB+_yf$D#czn_xzDuBb{4{P)b?NA`ka?Kl1@NEPCsa6R)Tb?ts;5W^1Vnv z6zB2&*z(>8+2yQ(h!N*`E7j9`KRCWl=A8s!OJ~22Ka%Sk%Y74OD3{XlA}8buA3MGO zwZdqRAA#PJ*NYDM1}ck>v!sG?^=UbKiTIM9;7vQsfIqO6bE)cVW`u*)K}ri=*r}1a zP9B2xERVHBZPy|tY?O1ril`DT=MZB?Gi{H~Df8!&_(TsAEEO`R@5$dyO-s~_;onGM z`~XtnM&0ry?;-q$|ilXoUdkWX=0H4H^%*ghn zXN>JkV5PFySiOG^@^JofrQGcln>tCuvShOZ*>0YWMo%>shoR0>**M5hI>O@XgmCAQ7Yiwq z)RW^eF`QfZ_bfYuihNQ=8e;6MBj)@g; zdhv|&1>N4{X;iRh|9^!j_F;=7T73@$tIguk10)) zi%fY--8V9mU(NglQ;`Rov!J~ehIY>Y62WK6ud1}TUS8DEwvm|1-@Y-)(QRy+To@x0 zAy8Iiell|#Dxvp|5*jbfQsPc-8z<_D!c4@9Hnz=DKFy*3Nn#xM3TavRCNy?JSclc> zfNDVAsz==N_~Jc_IQUl|a^mWVW-@2QtI9gH7wxXKIC6Ty^&9L~Jvf>N>ef2$9iDw8 zU%OoTNNV718!Zd+xL=f))_(CeQZQOoVX3CN^av_QX&xdSdPr)CY^RNO6s|(;wz{ZJ z<7!pTwK$J@D1x86C5SJ5iX;&qt!#K}*P)oCn}eVIwOVQyvEzstgiVNQv3e-ukhL1r zjP2dMh++|!J_sEN%9K1L5;MX@49H9sOHR=(ehRK{>EIsk*+}kK6Lph@g_5)w??-ls zdl($6F{Y$e2HjBWZyY0+^BhswV#aXRQK>MXq&w^v*72>Xby=Dc0R&l~kuZ$P z2aMb{G>|Obac?~y5LL@QR=B9gtS9bg2c&i%n6QB7zIY1p(e_|{l~@Y1PiZz=C?FM+ zc*{0vPj!o{*6eB|U$y*%&Tw}~ho(INyQwKn-daQ}WKAD_rB7$?4N)dz{v4)y!8GIQ z1d;6@Ch^2ADNZHiC0|=6{b=|5a)czea)I@B)succ`D4S22DL(g^gIf(o>;?;B!bs+ zAYS}>un8-Ku?;dnCs-fvs8wX;#-n^w$U&XiYupmK#wKgJ9W`f@%Vz0yliJNQv7mlr z9#H3$BTv4GTfxIs-6)!c<&kvZpIu;&JM~EO>lm$B)km-1g(uACihJUWB2J%V>N>GWEkl&7 z#GJG6=7NV%F#L}mN;AhIr(f8O z#jhFpkGYn`C|DwPGy>!H*B&mDpXy9_XT)wi^cD&B#qU*Z*}WjL)SptjK&9^QJ)HMy zXU*?D+z1q5(%|?Bfyc5rV-%wHV3^I%z^yXH=*WIW0{lyG(nxhlCIo% zMZX8mbNgLr!?(czoC9^NR6=-hz}22TTCE+~lT*%HX~8yaE(Pul9>Px-U~-H}I8R^F z4iMj8?(zSqkR>pFfa;awZ;&-wGfVAIU;gDgx%Yhq)!2fJx^t_}s@|yu^Hu@sdtzDI zRB72wrLmc+-i4!yuQlsA_u;rpTtVUI$)^K6ffbdIe!h2pXP@z-pBEkaK%K6NWmm#! z_v{`}_M&8Z?w^S;W=s8`N@Z}O+A?7CMvD#@MTj^Put<>bB{BK+GODo$8x?G?+>@_C zttHOR-XQW@rW%;_9TM@?^D0-|;eqR3ZISz_$J0VWR+s!hIcE zt~}u#8R*1Nn2E3(vBF;bK9Y`uj!6m3N^hU`!_{NnL(s6L%^CT(VPYwBOQ(Nkl*=_Z z@7L^A4DZ+K)kF)|^%wFwDRm8-95gs9+}ddzRgFupoiubgY(jA(V9lD-w_qN^XVVjP z9~=BoZSzk?^qdXrE5%eQu$*_^u)iajY&qEyPI_~g>@`uJ+t7CK^z)wCSC!Wd{>E+K zCl#62_`W0VYCL4;7EN5Lq+gvZ?mFr{j$8Uh-{q6H8F!EWb=C0z>-W+%C11LJJSr{9 z31vh#d6^=u=~2#?;jOgkqSij*~gR8|; z#2%ca72+@+1D`ZlJOH|uo`?!Zerrx_d@f^GSwicRC;GiVLNTs_ssya{NGdIcBjQ^9 z+;}W{P6=|=X_KgNlA*nym$FKQi*F<5oQlkQa7z2C@fM3MUF8=oKE26)N%3EdeRWh6 z`u6<{14F~m-QB22DKeyVcSv^#NaxTk-H3p6cXy|PGy>8s-3t7`d*6HazWA=+->e0! zS?BD}KKtx_Cbq9h?2ja=>rMyIGwt1=L}oI<=XH3qAz^k)TvuKa~x}?zKW< z@@cYt8%5pG3MI8WL3}^v0V&S&28^2lc46zkHfh4u7J4 zgPKhHuKsDT7p>0BjXpII5rdTNNn9pct{7)wWwO&(n}V|(MjYf2-x`Mm)T}sG9^{CC zNo=b6fQ{O-WC*RbIiL+Ugu`DmIs_W{7A|`V8n4OlMr%ExDI!nQ7@iGFRv~rvEHV@+ zcG&uBvD6Y;KDQT5xlZUa!#;T2C>fg0cZQXcm+$gXUgN%UYpZtht~TrOgW7y{%IFum z;nL9UVrYY1pnINknfEM9Dg?F-C#$T;M!Bz#c{YY+elFRrB1GYiB**=Pa9&XJl_}dN z%rQ!$(@a9XsFJ?&EVI#XOWI6I8RdNV&GXShIVSjGN>j)D_FZ~~T)cx<2M0wEK3)HZm%wD$A5`5Ce`WQDFgG&Gk=5=e zsQw;44AS}ek!0TTI&Kf8!ey~Pj2xCq`D%{{d~P}j)MySPlXf=6*ONue_{DrgD_M)Y zh#yZ^!n~+0{*^uy_|m#f?s{bN}}Q~Z?^05%C$by2S&=p7t`@E8PjU1<+ReQ z|1Az22)@K|2imzSwK39k-q))y=tEAaZlDYGg&|ColJLX=o%Wi@hK>Cu4?jYxX_YWG)BQ-`vN5G>T2ZDO!CxD=p;B==;aAIm`OY}UQt1KIQPwRdN4n;g%Sp}3X@-)X{zdnP}9Pld|N+e zOZxz1mED2@OOr(n-#p8S+_IBk+?n35j*FQz(c(JEBpy)eODZLdYv63}@rvw9m zUsgOFLR6?Eu7$s)Jw<+|ZqtRDyKY@EhI?m?H`5DN#hH*UwuXTYbySq6RO)qWr-}SO zELJj&+iaKW->1$A=#t)NI?hQF*okFYv7LpJ*y{bL8zI;bIMlT>!gW)v4wbZhCzRsZ zfrSov;ylE@!7-d9 ztbOe8$d)IgeqUwEcAPDdR6hVD7twl#)rxucgTtidNSp+evvfN;)ULsI%9p#b81f49 z=skfX$x6VXl%ReO@cOQ2K|WY-@h!5|^rlo&?rFuLpaU0p zlo~pRygl_V=FXLttp)l$m4VsodE7KFSC9}Z`Z@Ak0%%xQ@Ex-TrWg}zJtz+l@;cO7 zBzQ^T-&eeT+cm&8l05c;u&V-?2n!eU9uFgkKfw3l^e53dCeG5$$AhcXj!Sfjac6@e zqv;FdEdjfOWu@8L8v=0_^@Dg8;Fz@>Gx5r=M55tGH0y2x1v4EmOC`m8t5v z@Ga%HvXN?t&%)hz6hlFwMw8J~Oam|9v%V_%B=u{fJVNF`Z?7}^hjaXZ{a~+cXQc%5 zv(va`WTp2#$14n{paD$X6V4!7l??IFwnXEqBv-0gCkNyaL*moq0pl3jxIr25+HEQX z-_7hqq8NieWg7TDlwV4~)SsHlf2exx;JX*R$65bHrir*$QhPaAMJJYo`opA!Cpu3k z4Z(yseE=sJ6!3VMqNmJy^e$}!dmCf& zhX8WTMY?-2n*t~@1)LVbN0wbR=Bm58Lk|=)39N_52W{Y7DaTOWv?)@r>}ijtAz`)a ztndXT-ot$JgI0SA69^`%e%wVztrJUC6yL&zw8lVb?s`K3WLL=`a#JQ&G_Z0hce{ z2vjEcMi|%#sW2R+^Hs5WbLTHCMsA7-ifGjNgq~bW z=4+9djR@)>G%n6WRba=!NB>(YH`-VT`dNaY4t;QZ@cYP+jRXyDM?4J$H98X^&Fif^ zd(XkwrYWfQ1Bz+y-X}7AK$*i^*9BA&6(o4#6MxSLqQAk(4&xre{`CIQEcM5tU8>IX z(G6%}eE-)};G?X&2l!7r=s<-*mj@yh|6?v%>`$+sQNIt!hYWXToN^$Rkst%D;^h0# zA9m1ZbmHNJZ}CSoA7vEyTjVL^M;V1&PtMv+p%9u6q&yK#Rmh);mnpQIx;mj%2~d!% zqHR*mnk@%qIeTf<*@#EmY1;N^aCE5D!wq&tgx8k~H?V^|0$dLl1-iVE!*H#h*y?wN z(&U?5yOs*KDi%IeOU|HyA9jpZ0=0>BESsAuzM~FN>H5bAk-EFf;oOCwJJBUehmH>4 z@9j~H?B7sP!Dy7xESFPJHp+FBR`rP(*G_YD2}E+YsvW+*h#ejj_nFw*R*7gsaSWKi z-jA&TQa~eu{dxD%>p6xnT#CDTAmU*S-B)`;!u$n$BK-?e@9Vis9Az{$^#ol-CwqYF zt(STS-XSusmg1ecuMiSVVgorok>jjF05(5TyhN$M6s0H659ddSto-=4vvxxof6!zs zisKG9dBGmHm-m|uI~*N_*OR9f(Q=0>$KM_x@=Vv2A}kN+(E4i|m1Da!i8kEx`Hr!- zG{0jw9U|ch4CiN{i)bQMRr?rVSbs80sTix1Jq>FGi+cTtq+Sok)I96$`3VS~hSKzrD`K_e)In z<@nsm@Xe7e?}jp7ivKJeOPuHly($GwsPIV*A@JnSc{L+@EZ=|Y8tH%pXrxeVYh0{|dZ3BZeV#%JR;C)pHw&9vJ zOtQI{GtV!*m^armQJ4Z1BJq54Y(^S^y3CYbTG?(fv|fHUV8DVx{4r^HSHM&lSF5jf z)_K;Vr4#MfEltV#B3^3xU%8lJJN1~{_aHLei z>Oe6t+VJmBx$j6K&pUVD16OlNmSy2nlb6o0N&aE>(D)OcPh9$0Y_SU&~E}zd9jF zxSG{G%a^bGYf5tw8r7hnClj9!SQ?$xl0Hg{wDk7a`8NZ}Yt&j*R*l*uLrK)HTwX|| z9j00zdpofd`XYuz<#`r()m&bxrjT&!i|QZJ4AKZEMz-?)x@TCN=y2Pby9B&NT*t%m zxQ(2~gF!MJld&)gHx~r1pHIe1MIH0t8es*Qe+43vM^8G+EIj?Hxqv?NU5mQ!YMd=& z@ia{HjmndFKh(n7kI=+9K#ctZW#$^SAVm>*F* zKA|XQjQ{RF3gCTQjIm3uE;K4IeEA3l!TmLdg9tzdzWCP+qA%nD{!cF))gASf{IYly z8KNgG66L2qfB#dH0pPvi5bkWH>F^(J?9;4~kAfB#q*65nik~>*;1Ezw3m%2;CY>S2 zWr;$KYD>1Fc+06I?d~u#4{X-a%r}EYy2JRZ!ycUX*WrAC6{Tj`Z@Y`%xYq^sWQtKX zq#NvJKVEFpV^}s+8kL$Sy~>c$%-8gN_u?izjqmNomrU(Dl*T8U+nrg;#fc3mYp{_B z_2CSG<0`ZHhIb4#jhd&>h9-aiHs+RwWY=2nI9?c!)BDoadU|mbTofh${l~%2^V4e4f_9(r%R2<~SV2jEl0NkcazUmz zXfEynXaI0%0T6ccEDRm3?Q8BS~|q>-==iLUR$$nOU-G%qB^ zse*Ck2xTZg_lBfITnF}0aY|kp<%E%CZ_z{_%{?;~00nI41jC&Vto!*w36Wy%X4j}V z)x$!@1W1`d5|l_lzqJ4ZxD*ozoJ|7_XJ`%b`g^!ei>RlJC+7JPc1!Q%QxWb!%OBxT z_+hxdv{j$7LoOI}g*a@uEOLNdPK8w@ zQZjAHq9Ta~U`U7CXxf9e?MjV>cQ0A)S7M9?vlgU1|2Yr6`+JHD?xA|1|B)g6SM|g! zm-`d2{O2Cdc|;>Rcj#}{l62R%kKQnFvh@K2wt8~%Ns6s6R7m|7GKsX`fxg@Hl|0im>svk5Yi9`vAbfQ7NYs1q6T+Xb z`z~D(KVg`|T1A~twcZ!EwvBlACxjg&_`bBsRZ57*qRa(?0+jNBMloJ!L+jGGLH6{W zDz$C|6SnP42uPkzR1?d=S{7p|hD4tD+V42T5(AV%G!WsPDEkV&iiJ>t$z8C%GQ?WK z3-x7q93f0(ynl+%6l-?-k{95ULB~BID9P$1&nZ;H19KBMEgV;9H^&lEU8M>#kpNr>Hn;$;wN^VI@DQ@K}f;vOCG9l-a3mFk!D)dDcX&&$9{Rl!$%aJZy=o>lX$zTIlNS?}lNc|e>!zI3 zEUI6{u#_dX?Vf#U+E4Xh_s^{Jh;OF6Wj<_bANzSAal_2~w(Un-{6XuDhF_<4$2Y7v9TF?Ur{foks7C|p7cDB?3RbVrriEQp zsuktTN}amJ`$a2T?i|^id-R`0#5S_#Hgns9Kuw&o4qwo@KEEuM7C6gyr?I(Q_9c)T ztMVo(yIPIlIk;NOm%?LPkM|j>5ZE`=xn6?nBplgvSD#iffeU>F*vZd-&;J=Di*d78 zkwo&ddHR(KPr={MApiiQe_Ce{{?T6u0Exe&GyWm+SjeCK#TwGP+8*5s`8t?C9GrLqC$;`Xw$#F~#$ovfACR>7 z4)$ZCae}1j2}0VNg%P=z_TR@qFr#1%>bvh4B>?L=k8=Ko z?3(Bj^`n3m7r_lr4IA2%5TL#<1h45L%3Qb(Q`K@)2-ONoQAnI0f32q{)r z^1XE*(^h(`uOx#jA!+=Ep_s7Pw+}Ff83p6FU}hquAo}@lT2@T_nt~u9=I9)bxt#)` z7|qo2+?De)(=0uQ-{?Falb4_tuU?S+K6F()C;NVWQZ;%7533*#&XhjA@#{`TAl{iry(bI{Oh~2B zz80%23K@XYIjz>=6Kb9e{R#g*Nu`Tv630xJZ1N z_y{K0-hs-FL)u7tF)K=u{z;@>!N9umQbWA6r_zx{_001dHWzMf`+^7%2Msa~U;7ly z&TAQgtedkDOXIX*(fkxQVW9%WF{w3W*tqaRPP^8PQnB!vn^2}dJLL60ZQTF>D+^Hb zfnNb1TI+#l{dH!g1YiJAf$$GMQ2n3g;9fvGJAF(3pFNk3B%IkEb%8>D#q$T_5@(g< z?1rLYXi`rW9_Nhw_InRJgChBJ`TUdHJ-nU}05X6CNc-@YjvimntN)Qhct&q+c9_)n z&m2NbsNA32j?y0)4P5X+&~x&X8@c`|-i-?cRlj zOa0;Z4u4b%xqoo4_q;r(Tc{o)dL)p)i{K3Z2!j9P<33P@%TF+iu9ip;)%wu zz&{DD@%A2_-*0!I1@?M>EGq(dL-O%MAoMaV4Ek2el8%ij=__)|CNZ0h6X0(SQH7&lVuP{h=!C}cmLW5NFM}ZJd>s_{2 zhLc}ssE*97O$P)35#aD6xGcscz>tT&tlbN98J$nKjpFVq14+MdMkCV5^hnf0r8ual zKK7Ts#;8f^aWtOD1r^gj5YWzEiWf9!uPCF}XfISw@3}`8o33ZVg&aPkx_7j0c6Nb? z^*ONC2PZT?e)rLMVbd&37Vr%K+XtgTY|p`HK4_+SZwz%x5wYHlH6ny~MLn>GAdJ=A zMbTWk6@2P7AQpKrV^FS|v-vfMN(<~I4zV~HH`C)h#iBO~dnu<%qF$5ItKfD|+nM0e z3I9YdN=6t@Oo+1#(MVLA4j7=HRW zb1DoR%Cz;tb+d3w+4*26UxEm!VcZggHEo*aL+`Hg5d{+OHg--@`4S z#>?EV9usHzQ8W2_4QX2Zqi(@@???S|5X(WsdZN`q<2JvEF$7e>GS>711HQa@XWHtp zbpRt#FYB+pj|+SSzymzCI29lU0Dyx&x_hw5aR-hv|ExLtRY3s3GyY!$kbgbKkDj2) zKhI1)JjRRwz<;#sYKI3Lr`Dfjm+bD3ft-qT41ta=kAGbm%EuNokuk^N>xijXx>J!OdM!+#Xbw>icqUIOO9&x zyOklPRtR`Mrz9&-TeIhHh!Bye>w2;G4bqMMDzf>qYqrv>dDBbv`L|UcE5PD(g?Du2W|Bg~M8Q7IxgmXX&` zUUQBnFqlk<*MQTv!8Q6lM;)$ma0l(mmuTiCbG<0uIgqgcTu8xsFx+v-y4=H*BoTc# z`;I0=T9!}6{nD`jjnX#V) zQ{XaLb6F4ucE`zrlfaUo=6zaJAm>E=T01xH0A4?jfy|9mIN5M$r(oUw^`Dhm;x`$_ zR^Eq+vR6`7XSwHko8Ve3{r6}YC14T2^H0&{NDtr-|F;@Bnt=u&KM}5xP|WTNC;Km9 zp`sRsxXjj6b_|R=?6Xld1fRS; z50`vt5`5L2k9I9>CtPNlT3txlRg-uWPbZ~wYfYvxD8&V5TR7D!NRd76@ynlI82HyG zE7$^fJ3uS#JoM87ZAS<5>9RiGBjYxk=}TK3qyg=OM=%pO{45UJ@g_aRe=;Em^{BFM&iSSU4gCi2|qoW$m<|{MWh^^Q}O_)mX(KCsgap zmh40VRXf}@kx^eh>d?C*jXW9m<44t}U_%w4jZs4uivEg`AV56QDPKq2VDrl-XZofB zYiBvBNp-W^In~VfUORr8 z22qy;V86w#YqB~N=~z$WpzMIhM$4REV&N2m!NbXL-cXaGK`yVl6FvBBj^L#1q+LDgBNYtCZSSPCj}wITSRa36a*-(&2wnHUf!xjdhqMhM zf$c}fsA-?JOT0`#YyWlU&osU|{|Sk6P~jH07uf~DsTqlZ3`$K6iW7aIU}JMCtA+>= z1@Ka!A&uIljs6gnL=#n}gqD0rUO1qq1d6u!$_&?cL}J(;js<-uv0ELj8m3*_A>N5z zi>@Yy{5_Ak^Iw$C-%W#dFZ7!N1pHI!#Crz5a*ECoiJ=$z_krm z*J^P40 zwhNv@%0?2{Ep$yt!$vG79+)|2lB!>atgO@4_%P)--VvBR zQUJ5GqdAdj0pshGgpUEjjSPW?IiJuE-XU#X4kFTk`&+fs^)#={%FYWF3_W&^CG7KKyeAyR?zJOBnpfP9M z6xA6_C5a<$z7n{r$~y}sHXKL`8g<&v^2EF|S`DQ^aa|3gzdO?pyQyTP2@~wp_|EZc zwIMt5e019={H3V`B%B?a$uK${s7f6s=n!J;Zde&&6hkkkxfZuGDnS>g(O3T|QQg^M zFclw=8Ao&>aPr z828ajA4iQo$OV-MmHov3!7`4nHwsnSnQKR>@b`%i8%g}^;w(_;#hLn7Gofsxl`vV& zY5l1}Bwv&TJ_1(tdK?2&vDu-6j`vnxC~6H29HJ>Sht%=EzS?&+>5>esvz$B;(}ur> zyG2zgHOWF zI~OWZ1;Qb5b^?s_eFfNN$V4K~>p%Lt>N;xx$_ec|0tw&Dl1LT{Adv?L5jpmgaHK1& zqB0!xC0PEfB`^@mB#W1!grC;X=aj_FUGgKU-8K~Q!CQs~fk_qC;o2eRGdi!ldEW?O z$Czh_(zht&ve;Nm=XfCXswwz_Fw8T3eaHHY#I*V6qU{k$V+?{YI{I^zE%06BLmAdB zvWe8%#;|1e7Mz*Li{|6hQ{yP{hQh?yyEX{IqhiFo?AF6E)g@m9XaR-}h z%^Va&(&`4NtRwrW`EDNrYa~F`PtI&25;2hWl4n>vWN3bP@;%q~}ybinrnf zIY^YYQs2(5ij0u!2LQ#9CQXBA7&fTeAh+Y-FkARBaez-e@BrB?0(EKdt3A(Tll}!f zQY25J)QL#{*OdJ@xYh~L7-1CJqd@ob=S9Lkl4q#hIQL&s9GnYoggYzUw?E3%#~RCX z^172EE%G|xpH3i*-|wb>N3HGI!~&CIl?J}%?5G=AN%=)p%lP34?9j2e>__4uEs)-b zn57Nzg|je|rv}qMVfv2SytNjNENPqt#H>)w36$B7ub5GKdmSxGmwGHtMAA6kwV`c+ zPi5QZjGQffNh>@l;LC^|V%L#{QF~Cr;fRXYe(fBEC!>OyD0{87KgLnkpcsTY*th~t z4DKX0^f)0=A=C#W00XFDyl+)ar%`ndOcMp%iH>+LGXan1km>getgk5soO>nj$9ju)0Z>)=tf+}sH|h6mBwn5$)>_7>0cTcD=9(?W^c-3X1*^uz|JcqZG?KTUE&6u6d9 zI}=4Iq}H9qNvKs7koum&<*|K|L{ZBi?Bn;|V7GahwY5Lc|LUeP)EjMG13m>u`1fhK z9BP8W0r8U`j_)IKXxZXEQNnk`!WFL5A!bX;$eqNI!R0c(mK;(_{pxN{8wrK`+v*Dh zKm!W?AIpmel=49NuKCa91={tO<(QcNeTYa8bdOBNT(c^e3G!y53z^KH$X7s&r{(7= z5YC}PNNm~a8zFgSi)0!pRaYWgWZwVk*V<-p8iAdBgNo)a5Mn_kbkYdNPV z$6_%Bnb^HaM4opyPc1an)W;XyNVPjKhNe3%ox_9>Ckq)gWzizDi? zwV~13hR8i8NuhWeA4HzMNd6k=7#U+@neE7xTi5k{IVqhNQa;IB3695N5_`hMBug?pMkfw&B zwYuw)OY@qaT1}LExd|9gt1Fu?t>^eWJ63W@3zVRS(H>hhej75^s@q8Om!BX}!B)a+K?3}h#sE)1 zu)kue;y(Z|%BqSkU&>>tK7H>pE8K3(?iHr(g{j45454qpqtK#vBgPayak_Odu5S z`MT0DOP6Dn{X7=t53eDMw*)H2sh$1J`b@@6}LXMb%vAZvgo}6$HmXOzv&8s;zi%1PyVN#~O%@FO` zz}Ivwrg8Y;B~4^JE_f0H5?OOas5~tA0wcs3*ajRojw1N`w*CB2Raf7h#7Yl}{3(VD z9t-~v2!IoMre@hMFVg2v>}1yOx* zNNO&|9v4XUVz(0WzyO-?V)tK*fE4Ytk+jTkvBe5uI@(0}yHoLlleVEUqLKuBt1bSe z@OmN>Y8#p2W$`0J6pmSB_^9hS@er0EgO*Ty8pHZTjeQESvp1At4xduy%&D)WQiy*b z^}P&w;`^SIoR%iEK~m`?Hd#f9#d=3ZzYeI$R zM$4wtP(E5!V#7!-JEGca9R&ElXVm_3LK7fLDddlU@ZUp_H2ps0|DiR|AM(W`|Hd!9 zrLe1^6^@U5kGX2xHTfXKuzww!A_G)q2?VR>KL$7qRg2($EXKJV?g3JDCyH)&hre_9 z*nD?5Q$PQ>7-&joDhtP`sOL*LUPQJq^&Q}&m2D1-?yP|JW4e&5na`H#HlZR&pCK|5 zReXWL0`HQ}8CNEhJPd%9#2_RzqN>KbU}WG+ zyRJ9^u`33i~y7uhpO=CyZk+*HzK$6zIH` zi0M?&<2^`tVPe;r#Z(}5XP%kP0cX;&;D#a2wPI4I7@rU%aTJ0?NetJ$l9`@*rfgkV z>Pnp!DRma-b%d@@;r$W=qVocTd%tdA2aTB8DWzeXiAn`=HKtqFXa>y*&a5zf627Fx z)!>wbrzR*Hb3O~l$7*n*#INin91p}9;#n2DpoB;1xDEOc?IEKebuCVps~t|`tZU3xyPQ!QF9UYJVCtzqP$^KQmG|LnOw_PA&4 zoJCxXvomfddbOA)i;2qu9nk~JPC7I=lc>VPA+7&H)=WB ziQ^77kDcmpYLV?&xPI25 z^}M!SKM0k^!-J&gD-C(u!s%Rr;D_(2yHdw*IlBRO%RvMUFlW?3t(;vaj-$o5OgKe~ zm)xJCwHtsQ91Xo32S?yk@3nR$P!jh1KD?qZD$k%QMH;W%TI18FZaxFzsrt%kyjl>I>_%WeB7LJ0R{>njT%Uo5=rmvggvkMpU%^)Sa3Rg<*KxcPNCt6rW31^COf?-jS zPB^e{$H>zVFJuxeLB-L7Fw9V%MAAk*ALz^+4nsc<)&^)wIuY556Nb)mjK!80IAP1a z^^|D8a~+Z+dQVj-qt7%tGo&?t@*BTg28U@74yC3;O*n!jWfEx(07B3_;rA&c#wY

    8 z&>DQY)7ylr%AMwHyw}D}5K9g;N`3zj0g|Kca?q?ezOKfPo$tyU%m)HQ)Hs1nPXZJ$ z()dCcLHkq`6(<$KU+pWu{6KAKo*pms!o1h_^6IdLXqbP*93Q@c(dc*gv&qag-+qck z-P^Jn|CqR0$@co2X&X^)s`>e1ex`hEXD+drnn{drfVr#$2^e`^Ey>;@g#85YqnZ#Oy-hBbIQgd%=4yr*8XBEQG#5rE>R^ii8`X+zZ6PXO!Rl2AJ0Q7 z%Y!JknRn*s56f<=`!l-l?usP^-ZPKeXJ#QJ5E4CLE(e(@u3AYscyG@ZFD8G+jJOSt z&UO`fkF_Kj*t}YRxd@sUt~CtW7Y4sP+OPl2jmqAz`sVpTW7kvZA12#%&8pQKO)TX~ z$DgeZTd$g=Ra%u7lMf|J?|Kf~iKmA1#SvW`Sv%3bxFu!}nzGne>2O=y%QTpWI77>GDtgGQD-LH8gePTX*T@yg69O)ar{4f~GOaEi)?Ji=RpK2(O5XbV#CK0bKhb5sw=au4dlR%_ponL-9Y85_u0~{zChI!5d*=tY zF5f$^*c^66fL{*T4towd_9F<69rD&f($n>vH-*z#6L8WgVoO`Uy3~_Pbt^9R`#ruHCB$ATR!$)!>1P zwuOO1fS4Uj72l;Nd|?$mp}Chs>j_Mo^@$6^Y!=xM^Lk9?99PX6L`q}EZrX)z(>IeT zWPJ8E%evmlqVrt};yS&9cZ;yDW! zUXG&3x-L6OWp%X@{1SGswd*cgMotP{2(i+U4!Bt$s&dMu*E)8TD7F>yrQ-Aj;xzw>wIxPbbRz4}Js+BB{d`^y3)I&~d?6O1YV z$Jv*33B{H6&~|du62fKbKSf}N%>LNl6gGBkFcC}d{YV4W?-88*<>0%O*5C`ctDhTh z^%aAkIx*`?J+(A7$4-y4MqKN)z-!s;=}~u=$I- zaMa4Jh30p%H(dND50dy$(hvu-4a9HiSErf61lvpEY{MvA3-*^wcr26h(+FKX$7Q>{^Go2JNFJrRU+tvBlRtC8kL zxth+?u1NNm24mTGFxMs0fp=R_F`uye?>jA?%DeeQZ%l1&F1u3iqiGs#UnYC$gFggu zfB8uDX2L;rX6)@)W4(f+x7Vh;bHJdQw(H2w^!EVDEQUqLNwIhd7vEXaD1pJrUt*TQO3Y;+%!6Xelx_t?rvXQ!Q;L$d0uFdib^3eZION~nxDJz z(`)cQw~1fgpoEH#Iw4Ah8I7QMGBb1L6-P7tEWu=*ic4|N~rIBT*h zw*oGl1QETBKkjRZUEH(zWSf4}g$(|v??+UCvj4+Du+csk8Btxvo<-FDg+eoiy2lsP zw<_s&y@1R4<7%Ank}4t&R>b|m+Pf}?E*toxU=ogXPrpG<%;p6)TLtT*mUdj7<2sVd zl6~U`Bz;FzSQmSWO3z4E)&`!?Mrm~-wC+Ne(DvGorks$TVB>4XDxY2U)(s*dL{tFx znpnr<_FI$ulxAU%7dPnU@M-xW{>kly_7sOp{IVa3^JJR+%HmKMv&y8fH>>jq))~$w zB7%`y46WU;80Wv0Atz7a)#uV}8?JZ4&j5KRI4 zuV$AAPy*)vvF7`Q3jN!E1ojcaTz0G_luv{W-91P&wn41)@Oj>K@Oo_9Uaz zJ;g<-UKY$m41GMiYB7)`@j>cCs#123_GN|ERKe-CVv&A##-Ad*?b1l{M6r69;o6YR zmC`B;4dTh}Xm|ppVlMLHox@MOqlNBwR$#_u{c>MW6v3}5b%iG)UCCHhOQg#)t)SNz z2Tm3E)SDBEe8m^nC*%nr%#&A3!^u?}9dL>`Z*(DqhZ!7ip92>&eMQF+(Fou@`7f8t zhkc~qx}E9AI#!jA)R`%dVBDa%;qT<`l z!mSL?i=BT48d4cdFVYjY74KY~z9a{U^DAKk;T0`};^3Upw&uYzpEWM&>ZqF~qIo?Dh-T#*5~c=y?>a}kB%cO4emgOi(JD}GFR0Yt zMx_>M?+h$A4P^QGt`n5|?*0tX%>k*cfUejiq~WT4tSJt3{@W&Q{%jkeQoUHF+rCnJ zV`Q@Fg!*GoX7mD4MiaAiNb^Esjc@mPU~aJp zc?uqxM0tyX3uN^vk<1rY;~ZL|N4Re_0GnjO0m$oQel=O!YjGswc!^-+Td`~*wAtj@ z$dFcqjU-Zw=c*!-ClX_cAe%<=WNUWXTy49ez=Z^YX;;(8cjy5+X%Dr)V5F{wyZ!Oc zaEla{Z5a0-XVeSDeh}8ILyg|>7v|JK94GVbgDl4s)A~d$U;8V~D-`ASIT|a3Io>v`mMSbHYRA`z5EVz)<9sMS5IKQ6-;><`oc6I9G3JJ~;igLEP<{s+ z{tzGmzh9*Oi+O*dBd}NHsIn72E}%QA$-|P95QJCY&ME6zLQ6(bRNU-1)oeESirkH3 zIF0+)WF;pJzkS>DpKct5)44ihLwq7`<#|FBmZ)W!UvnMDJ@Z)DKFbteoj^y0NzbP= z%u3IvN$AnwyFhN}bx`D_(dUZm)W$#NUOC{2rXmWA)`hJXS(whhYp)?+LY}*yhD^GZ z*YGT$sD_}47m4XAGNihV zrTU~FcOi51c>AH+LTs9r13w@ot-_PZokO#`*?*y$HY^V%*MCzXjWZ|4O-%;1u(u*} zv``6X4}Q=2d5}ht6Svi^EHAJ>X;$8{NqiwY#$nrY*F*yPWhj1LnfCqb(zj2@zPr(uSzwRzXwuA3TBpufK{{SC{m3APVWSI3Iiywo&g_Pgmqr1zxQ*QhyCQP$l zTjk5-K0J-NykE%&vo#{+9>0NNAerieJDD_hp_&gU?0t(Fw#{6ZAmk?T_S#?IB*r+wM%`$2;D(qZp$y=R+FmZb0YbM6U_iz7r5 zpAE9!CCPERqzwOoC!!AwlYUlLi!eJ-%l;Lc22lX9YVwCbMu8-XAVCDAO#k<{N$E}? zi)2yiP(b#85pDL_2LWx-EHP|tYY8&hsVo-)Wn3WKNNfi@C?b|GZ~gb-7C3xL<4|7? z(xpjCSV+nlr3j4n>w1bD7)LaoJPS`eflA*ZH$gEo1joamtUg*$nz@iRPX7G>z4&n; zObDzxgbWA9aMSnw8R6`-+ zRH6vM3(6-@fk93pnF^xoPe3pgZCapUYjRkbx3nU|P7(?b8;SFh-XKa~Wfc{zF+kiG zNJwoQBP7^ntY`w(v`bMgIc;*C7qX=rr==X6U6W5SmqIB-RgbFo`k3J-w z+?|qYU}2=FwQ&Ls@?s@DBVq)NM6rbMwJOpa2${Z z1cMqcHLy(*Euy?^1=kbtUkS&E39{fc){l5S7rWR%rV4Rg4fEKtTbJm6S?X{vi&zA0 z*D%JiOnX-r?OpNB?JNw_mit<`JSeAm54W@J=K&@O+yCS4s{*3lyLM-Yp>gQ$24Mg} zFwhygVd!Q+x{(%vp<9Md0qF*jmX-#kQ$RpkML;B!IG{V+``!Ed-g7R`H8+d-|JGX1 zs^?jce1*OUe(bI}@cjeTO~{DHtJ-_tLeF@OGb6`SEzTSjB3peNr3Ow{_Ui8$pC4@& zUa|Cl$^4RLi+gYm*t^E>clPNOo6mW}>>Su{QT*w#L+|}LPj2^ho^0)h9^i|oJ^tt# zjJnX?WLqkg0QkuQ_Fli>*jY|H9YZO+^%Q*8@FdDtlK$eoE2lJwd0_=MfZ4JSA z1K6=)ji{JgTq^```2iv``DE&k{c&iC$#XUK`N9(9iQKUJ)Bx3lwdTz z%0WF3^PYZsTzKnKDjm$1kNuH99;bAOgo+>cx-lqmr~$uv1Smwx^lP0Cy^NS!fII(M zr$<@+LC^a9eYYRF&&l^S6;`?5bxmM?z|)_=D+SHD>8hAeNxHhvi}-=JB+m`5iisfp z7iOFhOp=o&@XRaCo_yI{#B&5E?j^dauo^Fk-gGRP1WR^kO*Yn5EY!AEBieJGnIZAT zVfbfm^;OH|XY;s2b3B1d&-r2WnToUdm7LOr^4Lp}9~TF*CA@>SRM0hz(QfEqDrnby zm~ju1{6Dd{EY@eYXaPE@fX(YWD0f)?<#ci^SwIc|YRzE|N+e{dI0buI;eW z@#^8{oWl5vf(n;M%DbE6rN(VScfS8z)@r-gfqwYgvQ}W+8OM?&ko$nvZk=F8!yny$ zH9G)&vkMVKD&Dme2sG7M4xypzGx4PWWiJOZNQ*uTLo*LE4aW`HT)NNS(_jiCgLyDV z5_eO+humJ&(UIC&E{cs5I~2`DeKXWqjpJhrTQ!mA(_M`vuvuM>52Y(&iBS{mV@cYR zn$3@54WP1sJy5o0_c1jU&5yAS(Pd2*1UZ>!d?3=zyMM78Zs}*;8E%PMk|Sry^2Pne z;msd$Y@Uiv$GMf`l_Q#ypAZmcol9*ouwHP(T|U3?&nwZgx@{YESqvJ0B;YwN>4&n~ zu@c*gN}GJE%F#32kMYR`l)u3HpBeVf|Hm2jUAsUNK_-#A?7QAz-w-8m-Iplxa6CzS zaJk#tbb*k0hx$9atMyrU7p`~Dmz}mSi?21`Yi`-DV#j2+Y;`(6=#AOR4K4D6_>5Ca zg)%s8<8;yc7Lm)}lSh0LNRGyI<=^oQ_z5L7{S#R?7~!)rG;NdaTcb z!$8)elyV;3+RwzTE)!|ACBc`5SdDoiT8D$rcsL~cHmL0wojrvzNRAam#M(Gx zV%4*Ox{s}z`uORQ>`1ERS<9&oUfE>UFyl~LTHP~KA=W@cla21w>R+*E zyNM@GUi~dz3S?|b)ajOf=$Zi+uy{Li|8Nr<5QJ2ulI4fZ-`H``Gb4hL>fyI@H#yvr zOL?#=CCO$Jezzc3F#NinemJ_-v*7#Z39~Vjm1uX--;?Oj5_|kQL zQYvVyttz{^@VGgVm?)IkpA{%OguWB1B?gyd0+H$#*{-W-QcP#KF9;Ow`GWK|_&Obv zk2X9DYHgyydPL$@>pk^!R>yO3XZ3mS2%I-WABEgk7JZN*s667dZNhgmVd^1Y0~Bq{ zaAY+n*#R|FP+0NJcZIjuW2&&*_ua?RX@~DWWPDYaKmy3mXkX92Lp#yi?wCJgXuG{N zAmTT=4EmKcKnTEa>3)CPza*Md_~Cwpzb#Y;{#>YL=`}{b@#s4C!_}TRprp8l^;LCE ztJ|3&_adh&UD##2ChUc!Oag}g&X1%1AVCtopiKyLJRg`%tY~wv2rWf>e(!)wm31-~ zArf(YR(-RpzzUW|Q02}cpI~fKN1))5GaZdLdb@m3wPYp%*|<;l7@j{I<@_B~Rg%d6 zwzu_EDwAgP`Pvhs3+t)*T9;P^XcTh_XqiTjtlDm6Kcu&Bi^9E-j0s_%I3Ryo2e5 zCw2L4&|%3kP7)<1lT$WWX8TiJ=Ulc}R&rzTP4ho6bl1lbqUD#%TOD@l29iv`hx`En zptutfH1ZMuC9pxdc`}0>xI%W^fm4vvNF$t<1NG=oBb?s&+-x@k($+T}88)2inS)Ti zH?!nH3LG=`5(WX5G6vQrSU^+=CK#ZV)@nV~+Ngjbor}I^M2v#E*93sH@-hgO(4ABv zABHWfs55mQ6$S;@O=#Wmr_Wn;$ep?g_mdL}mh^kYHmVa3F}Xlml2ge_b7i8=*Y%70 z>C6?1WSv255tjX_Ly2&<+-0LiLb4n)`0-dwGNaA_Sb_NDeY(Y*jRHkcdx;L`^D08t zel;Z^NsthnXg!9#RywJpS@*b8U0#ixZL)INJ#;3BoOeQ*G~eE8SnSZ?w(1N1{E#VI zqmU65vrhx-6BY|AlBBBT2OmB^AbO&)6BxWOp^pHy2$Qzad$^}qlHVHoI3z7RIn(La z5@i8jsxv^nAu0Z<^VE$^>R^}?FH%tF`jI!q5X+7HaioYxkB*64D2L7j#n+E6;s$=a z?h;CobcZumcUIlz?CL+kXWQ?UyLmWF+aE1>Zj>J_K0abA7a;}~VKqRJc=y0dVc*M- z6)hn<4T0ctvW@o%6y3WD5D&6nP7$s=XW<_Si=Mp-JSRN&{tG3aKMBFP9;IM=UktiR z&d2S3`?TNGtX{>*l8i-hqO1XhlDYDU+c%1#iInDo-M4E? zwKMNgeGB?lTQ-|PGJdX4Q-^q4YSuR#(pI{ z`XJ|967$boOz=1;(+;hI6l(K&z-TUC@GC+UvgF5)=Xz|ply^Q2C*Q}{;nTiI5^H%p zF_&Xca!}~s2r-8H?%DX>l63uY9G>G*=DXw$rpj1Xf#|ME8!1cEB{8^p<|k*WVSZ&n zbEG^Nqa!UUK74E1$Gd5wXlK%EEVX0~4t8uspe_@EGiWxHU#n(HYzj_F75pd<}JDfcj4!j1eH3R5Ljv8UJGu1p(lvdt) zrfFg$M5JvlZ&r|ME%ucplyX!P1*eHSD3G#upR$zaJ!{;sWt*icllVw=sZe#dI@&66)Sr%<;`V&g}^$OFQe`s zUPjUGyAE%B7J13dO9&kCvtzm3V@yKFjlrdA_8X)TWakniUkFmni8LmXbZUB^H z2$jx1(O8f!h2)>$4sVlE-x_E6ZEb%P68b3!98>y%`HWF z!8e|)sxxk~nb3tXpRgFfSobZEL}(oz$hd3bLW`bubZ+l$Wyxj??4B^qU^q(Og%dr# zhCDG1h8YHPz_hE^W9;N+9$aG{!lA_UChzVw2k@Zv0<7GLlMACrz6GOFfK?~g7|&-& z`s`xE%&f)dshpuw(s$cwh)~ks3`6;!JWB)HhOe@*Glv0IbYSQ6uYwTmi|RlU?WWrX z@RJVeScylofs%}{rIP6wkB(qQOhwgpDCXyR)9WeS$`~ooM<+ox+yKM zsBt|MM_$o~Uf=IyO}5+b=PWGSAK+aW+#huQm4WZ-)i(ZM;7j=T)2d$>_=JBWk>Z5Q z|F~I=IWii^kW*hCNc(jP)XeqA6lj8Zv_RE*(VM!~MxjLOzZY9I&eNd*{c?3*c4?^j z4@Cz@z{S6etGc{N&jd(C;qRd_V3^*&sjQ*`mgAnTwQ*Umm>ru{6R=ECGJ^0&$g&iPwIz zZP>&Yt0zITrcle^LAS~iRY?uWfIi1X4$=5M6@s}J;W&K5Hnub#2Szl*iKfqnk**5~@ z3dryo+!nGz#vR5112$DkoT|5_<9LYx4y#0%8D^{|PlvOKl;SuT5j~x1I{A zj`RD<00!QBLZT0^n78hLRQuHsvSHUA5MJK`=f6%TFs_)OLKaA-UYM=W%9AoFCDNkL z=z9@a4dTqvg0ioZd0Q37P zknFn_-~3b=YwHYhD5YSkl5CF~HO#;d9}BbkUgC2*rD`ou^L-CN&5T2iI|*B4t0J}J ztI9|b+b8L=h{XU)?+AL{DnaWpBnl`oHo9XI#>8=vczRH`m(22ssg(v0Hn!$!%oByL z-rJ`uw(O_}i?hs>Hv05X8hk@2| ze8Ne+LxTbCtht&YHXqEcZ$E{}IpMEHa2PB0hQRe2qg)MnJH$YMR9}yjUI(se%oZo< zt)RKOqD5q6sUL3Vvl{P7zAT&T&fDTSshEJ-$YQ;+C2EMeK`5uAV6G{w5huS$3jWg* z5dbtmKj2>_0_ji5kU7vu;=c{6<-#ciGXMPgCvdi_@i}pG;Z;1s^*JHoCcTVDd}o-u zg_sH<*O?P=6C!F#CZk)bkN|Wv3=qCbq!FFM6jKP8vbjX8)ZBWu-U$nLVf$;TUgg=! zHt(QN`MX*piJ7+w4JDP=DXXkd-zFgB>A@@;ACs5M+Bw@iH3=0pt2fd^TkcJQ8HG~1 zvee%la4_>?7^B^=yZ2VN);_82RBSm)N!%S=ADK zXY#(`7k+saY`81n3H05IQx2n-Co$q&ruT>GR}lH17jRdtQNG{5AAfmaXN2PR+Q-`D zSVCevkxcYCB)d}bL1UDrhf}~?DM*>NEYUmZoEkqHabdNY!!jAnLP=2h71#!Vl zgf%7+!L`^d{uh0e@1yX7b>2Tgn+;=nLfaxb2Dw8nXBG=3JTi_|+)v4~SCY1VA0I16 zWiGC+rkfb6b-t046y(;=nvxYB&YBuW=B1L1Ng$oX>x| zP53-)Lti%YR*|LOj#W%fXn}NWG6g}NWv+~QICH+Y(j%5kb|X(zddi=?Xgy?#B_&t7 zZBJ^|kjQs^)u)Ygd4*eF*G!$PjTx}Hw1BiPz; ziq@#x;DqdoxG{~nN=doV5{q8@RJOL-{B_aiIfbzk0V_nun30hE?b=VzkILjJ>u>ugUJz=z?>o*^S>m-mZhW}^K@nYQJxbEnxP6RQ|Dspbm9E>%t+N(aU|b63#_NSd!kB^udaC; z)Ci77e*)tlsi-jElFT&RbIg-9gH<_y94@BzHBu40U!0oda&I-^8RnCMxC|WsWR#-V z;U;-*#ECBTQ&G?|Sl;iH*+xd$^wCTKMrSRG#-eJwM=WsX!fIPbq)s+n+An3VlM0eC z@&V6rNYqy63Av0u5?9U!Mdf(B{SnWhd1(NWPsl@F5~!Oz4vi-Z->8H%Bg7k)o2m+H zb?A5)jLPrc*?9@s)Z=Z^E~Z&VVVn_{SB`*CJPik?nq@70($wmJ`Uc=<-kesTW4lSW zFhyPOQ@I65McDTe8X0ri9UAbo6#vs}IIQZO?>qHEif|*pi}xu^axbi zM>U@!QI=-b%iT8}PS(I&3aGp^j$btvuhP}fZ+cb#n64=PFiZ6u4&69j87gV30P{Iw_XDVzW`G|Czp0X^6a@khC|}S@~7N{mp~|)v*v}L$@)qf=K76@pn0xf5)kl3FPap!);O**j zWP~IC&d8^*{gduVXcm0UUtv%-jZtbddMdw4K} zhT1g;dh3vL?FpnIX9XpC@?Iql(upk|MRi>k?AJE`iX{=!u+*9Q)-c>$@!ZOv6~xr1 zV+m3scU=*yPl4uV@ii|M;_z)Au`BTgY{fv%5lE>h$juZQ@wNu>d>V`S{ZJY!T(3>} zb1P^eq*Zbc58~ECDV_;}A4SrLrCscG^0v)C>`kD>{9-K=9)R+XJwJ%UOm%p@~ zAwtq1b(`K(y8O%|oO*QkPgkvp~`i{-XmpdNMzKmp+~4YIaH(~It)$R>5{Q|U(d1N9x(Ehhr`$Q zXxkqP_THpMD83D?X<{k15=N+33NbX5-7DXK$@65m>32W5B3k}lC;Zwjvc+{y11N}^ zcXG)ug?eD2juz3kTGKoh?pKngz5X9#dMET^yw7GY zkL-WZ9rwTZgZY<2n)I0Gr%SUoJV*5Zex9rlb;O++Bzt3JC|T4c?qFlr0vUI4(|zL7 zrF&9u`f-1?uCeYjK5~$15z$y5%Z5F*(e#SJjRh%Ech!jS7l>!ye{Q-kFtswIuSD4Q zlt-JpH`KcS%=76vh0%syb?uh$vaE56DT?S*=ZC5#GWl)jk`x@8oB5B7p z>_JSsk+S}plu4VNbloLho`*C-M4YD>GKCxhc?x_6sCldT=0t3 z*(MXBh9kqAY;uAC5$LRG`5E^VmVy# zeGXC^@)g4fD^5JO>qS})aW->06fL?`kdf8NQiK*WzBQ2Ox&)h!xi z&BRUH@|qEgbTBcns<*s;0km7*u==p{Q{!fVH!(cBu3+b7;W;yj3>VRDaNXw}J;&Bl zwS+jO5Vz^dDwqLjJ3|}m^fd z5>7T2G4GByV{0lmrIl1Z;n^;2;`{zleA&CQmhjhF<_ZBl2i*O?E7(`N{VA3HsAVD) zxh~7F_#=ny`Qq{9`&swAaOu8{8Uy_Fs?7SBJBy%KlN+TOJUncEMku1aN@#PU5`k%o z?GG!ReXjQ;#PH6@d>+1KfQq!2uYl+Qia?(i%)K zim~KnRhkM+6D2bDqLEPgEed2fv!In)1nqMjlrc9SX^tCu<8iDMr+g8Lmt-_KE|6}! z!8BUno*7FRWo8j8ioUyWIhHs0csM~Fw~IA;3lkATio7C5P>r3L3D(dmUJaB>$}>)F zOW=ABI|7j4^CJNdEE(G~@(j|=mHRmo1)t4WW?9}eBV~-}E8fiLXCzuo3na!h%?lJ? zjd_}4A40rih1v2NlZhywCkak+&1ZE66s=bKbXsR@{TX&xor zcX)fOJZm8oYNcz*k1R5GRAM0Mhjg=CkO<1L>shlmp~a9+DY?O<@osWoGj@}`$9MpO0zz&|vVKRTM6Bq&I zmsLLid+=`+zX{Tq@PDZId2ki~Y-2?{^Qs7k-6(x#dR@N_k#2-7`K;X-rTyNBW{z=K-(jek>kG(z zD>#>E!|0wq??#7%QHJQA_&2;8$Se_qH+0oDwJlit7p^Y7L0`Al_0QxTQcaGIvW%N= zsa#C852b!dYdxc{_o!RO1kk!+j9K>J0hy2JAKQu@g;I?xZAb(BBxRP|OWckqJ?t?> zv)$T)HUqpS2)Z)i!S)S|E)HymGE|ebL`W*&>DU0CWuHvC#R3qN>Q|b-woJ1gtRo+0 zziQP<(96`7V!~5B!zex-oMH%=Y&6O8e&>w6<`?l=JA`KYY%Ijf!OFwbh1zCD+r6S; zBm}Rmc~d%#>?;MU!LwkiP#QiIg=w@cAE|jn#M#5)#0;_~R7@6~29?TPoUU*> z-{oNbc-c7TES&t$Bti)Yq#ys&Q(^56|M9Zp^28#J^nSeT?qr}U;l!5HGA!%n?=PF> zLC?@F;v$($+1#lb3Jg`^QuH$ujHcbJ=}LqA2(qmSvpWbR5)-{A$sSM+9N3XO#``$k zgueDf>LSWsrxpcxOCg zXN}fr{F(p*K>WW+?P5!KItHmSvw`%qOa`ATh%lV+p5nB_>MF22BC`yh+=IK$d6U;q3e4WLgl~yR#Tg8#l)P# zT2oJ(wg5X_oE1n$6v5q-GV1gySST$@G_y-J^46g&+)UvF6laA0&FXLE0GnU=jM`eF zmW7u&ufEpo6CU7<;93e@qezPPXG|w{*iNn7x;;#}*8n*{#=jKM(QLi()bpk$O_cMm z3@$&13`+mDCP{xP^4|{`uA;hnc@`s##Qf*ukK4u=O+LqhM0SM`vcc4CLoUn8B2|#I>3h?N->b$>?T>^N*MijN%~K z!#Jr_SsN~GM)bS0tS%(q9{aqVPTyBHKUkA@G(eqxiXW zETru?4v3K)(oPTKdaa|CSmef@TbMnbhI|*G);|jg}Nqt(MlWCee&} zZ7^9lNW@lrn6lgH;rZO`I{H%c<>7ki=hjo35XUwE*7e;sKFnW8HJ)77r0GAvz#q>K zctOZ*;K|SQ&)LW5l}ysE4yjgre^}%b<$O>(ZyO=R3uWFZhvt2;}#t#Bymb zxD4l9ipHa_V-7UeBd|%Bm9U!(IMuhI1qwvIKJqtB)&1*nX>|qf* zq75f4c%$-%&?x3Agm)4CuZt``&8-c3uK2UqF})B5i##{2ioU_4;T40sjD8Itn_Uqs z082SPUwB!jSGhi|D0u_Hthx(~P|&_pM&pirGo5xP(AQ02&9bgRpXYTXRl^rxeM*@R zBJbwvhcNwIpb#f|9Cnrn;(OaAg5*EkE--(+xkw(`5t^Y5puz8@jjG4~6vs(;x0 z8OTBH(@JFmFi~96h*c|9`l@2nLV(ain)3pa;q}eq(11j)1eCG~uk^BtK8$sohlGC= zN@{1Gtts#-q!5#8hZ^m1fEARe zy!7Igw}M&!=((?7br{xD6`2J9VDus`RGMAW?0Uw-oXe2HcoxQyDXo{fnu!E)7sWjy z#GOwH`Kd7Vs3`X{6Yf@C0=0%!zJ%OujbD39zL$Zg^xvkNxPF)@P84(3Uz17rXpU6e zAGm`@rSmtKB%^n*7Q>+uDCmMqS2FTPvBWuAs7i6ziH3%c4W{J3enwSmtBudZkD$ZT z8iZF`1@D(d2!GHiI~yBl3D+nytk#j3xpSx9QNdW3A2GV*T5Nb7cE(LV3qjZMhx_h5 z%$VL<8Ds%IFf3t0TJQ%5M3Qf*1nFS6VpwFKrhML_hj247axY zzR%x!_Wk$^kHP)8w(|>N0jwn?AC4@-ZHsP`8U|ucSW+Y)Oi7U(k}5(NlYvksArz*E z!=ViBqG|UT15!o;nHM!yLfgtxnDxjiHgkd@xNPsEX{S=a(Ok=(xlhD(HB2K#ZTeVH z;BTAEaS~j*%<+HR;DKQ%OG@J zgFUo~(7HO$-G=6+*_1$wGv|u@Fgpn6!2CJlIy`p0PlJu1fWHpFhX7jt%R~X3D*Jb6 z%0bu_Rs|IIjT`7=Bbr z1GPNj5lXLaC&-iL-Y1|ElpL)lTNkTzdLQ!itws*y^7SCw+eIK#{$}rS3FSTb;6WH# zgk30LrBN?cHGXE4(8tPx_M@Iqijf=}2d9aFgYi$aK$rg#RzX&gu!Vxx}26x)u`d>?mytqiA zy4}66Od<%NT}nz|Y8=KmRdXLo<}X|#%Z?sz$7?jy z6@EzaHw}95ySWHPV*`DQAupTtq6Jxw@J!&UL{(_ zoraY+10PiUn)f}E~eonvEgU4Po=+JK8Ub@k|@WgEx zD6@2@0{t@v+hh|5{%I%GhctxK)3!HvCSm#uH!xeTZSqvKfP`t{X+G-F^1rlbepZro z{%IZDuyS)G2ugU3W&`S%vKP?Ap-Lmv&2hJuBj~2cfin8a05=MF4V6AV)YiPWk5_++ zW8Mc$K963|q*?H*3E3f$dVq-)Mf0f1vu)577 zFbGlAjF=94Q7ZJ*-z3OLN`#%3>D7v|y&yXGE4ua(KUc94JV=A{=Hk6@x;P!Mv{@JM zTc>1ggW|l)mU3H_eh*Eg1Wgr{nM+xEFCNp;brQALw|UU1orCMC_)Z^&(oA6aaoA!* zT)oSWRNE!V`lsYoHv~HlBBvc2meCG2)VtD5gN)W+d@Z_gU92dN&`k-1kVsz#;iKI^M*L^kGGRfNz7Aj` zmBmp>5hrb`3TC;UC~w*zM7^$JbWxl|!1R!WZpVqgo@MHqzx9N~a~;R9z^#-ulqh|W ze|Y!+d}a1Ac6Zu{x&FOeQ?dsuUE21eLOx$sD7ePE4-Qx$}EU#fB1k{ zSEf;P0M4&SG#W_nA=C2O{NcS1xr-smG%{fiZ7*pFokqUB0|hCAe4KbNuy;uLQ=n-2oO z4>~irUa-GTz3#F-K`F=~mrUxtLo4sPD(;Q-nLvXZWi6A6e@2jCqgpbVqZ59PCa=ob z=U?B-y}G@ZtsQ7d6Zd+c&K{7bUCg(NTg$3e?|JL|?Vv|_XN;KR0u6pc&g&-M*Nfk4 zkEgS0yjbpkd`VGg9(vfoZ`~Obq1H$!_bvBvYPLaPA^LpNWRDWfouW~B_JJD9AwPYt z_dO}1PVwk6gxC}glKfH&jXW`8o*S>jNg&y6VT36*E3=YrwvL{){=rMHu2EiywzJwP zeDWc?&1S{k#PW><-=G1V+bN(2y#N z23SMX&S0+S(Md&?-&ZJ==hJT`@c>u50GdM>x?=EnA<#HPotLZ~;@*DYVae0hj?3Vh z=5Tw$Ub%H{)eWg;jwHH937)FDMmY+tz~=onUWg{eiNbTZt2IL~Az%7Ix#S5bwxmQn zazqmo|9e-d@?8NVuMI=aST(-+r5+lK$E(tKv$$sO>o{O$`sR1EYK*;(u-Pdw6|L@2 zJt8|>k;kcMSkkerU~Pc3G3RTgx<6!dpS)8!uij0JEwZbugUixy)^ITszpePt5MQg@ zqh>vX&Lb`4xz-&rtrRH+rYR?w5%A>~X}WhcQ+nE4-}A)3qkWATqV9G(n8Z0b<(|A2 zeV{IV@4E!U)b(&JJwqg4m$0Nk;nTVi&HC+{NwZ%r8%RH0b@}*?m&{2Vj+etY_$=VR zuB%sFI*RwN%u0xkzv+7JzJvbQByAYIqJP18L`#Lt>yP79d4$hJ#4?zSYuI^rMs>YA z9nJpDz8j@tf`rCp&$=S%li z{k<5sBsIS^a%2%2#M>W0kWNxKTw(T!xP^y*tDGGR4Uh0~t>f!RW50f01>t9=jWI8& zpXjuAEVY9OuNMz|*;#3J{mOiPnrJ$SF2jfCzRSmK-S`ZT+Q%@>(J)Zrvr^+Jqs!6& zXnCMV;d_xrdbh2@DNO^6#J|MxZN+iphX<*84`@PKd07aNLGad1#d%DEP`7p@*s2w~>e6hH#QWQR)1e2jF-jzy^6le+2k?7P- z&;*e3!@Ko#aguXj&=ZL{*SqTnc~+rx8&nB!aqD;=&77c-WITfIJp0@IDQiwHPj#mw zY@W!{ACPR);x# z11D#7O2$DO30BEV+>>sr7vT!8e;s6)_1Y}))vR6asc@L@q$;iNBrWmrYuF3Gfn;Qt z;k9T{+v!*|AV1>tw}L%*+&{beuEbha;&bA zJPr*joxI(3#JeMSAg5irm)oa86E|jU_@)@SW~V0&LU|8o%w)E{O8hZu{foL29UvZn zjh2mo`D5VF3l)w0RCkKnbj*?R`_)m}9&MdBV{eVJ}6#dyxVFd)3F%Bv%2M&$frxx5%S0BRV}vPF7Z zzZ`h__V@V`=Y?So+NrvmoQB>zS!59+e00VOT*GCifTBDq+62?-?`q);ro+JJTjg&Je z)q6FhQax0w58~k#Ct!OU^P?$o#f-1{FsBZUf+VXP4f8a+v{mzTr>jx>pIG->dH;V( zE${RouF9d`b?PljxMX7-en>6jC5de%i8&O)0wnvSutN}%` zOShT^5d~tqbKA-B%N3}4&hsj$s`%0;VA#`e@sw%s+I&~5S(0btYZxQMSgNT%jpJ*? z0{Yu)tQtzk)^~QgMbcO_!+U79xN%|wbdvN(AP&t=oj4H6(*PFL0F zd6_aXhnvB5UxDkaEu8~pyAaivNP$i(GK{FMy`3SUa=(OFCmLW19L@vDfLDt*SYXaH zAKEKjx+?cO)m?Az2Ov#Ry&; z?1_K5o#|~L(lNK}WYx*+*4iZJUTl%4+QL3-dNcZ$%x}2GFll zJg)RW-ghucx$tz(X~x^aXwl~gr}aJzc#D!RLwvO7|ns#-@grtds0nvvN)j`Mp<@;lZ?nc^9`Qy{O6mOVpH?ABc-c_LGa zC0GZ(`GgpfjGJfI?+g-@%GICXZFG%^B< z-C6S%%FwE*vtWgZVK!yeZ$x?Wn$gU8%DWb+KR!>v@q||p{CeiO-AOcxXcl;4{X^!=i~rH9vpDox7VV zgDb<8&x0tH^jQod>#5{l2LKg7e%dq-0D(ta+jD*KdeYH))-J*I7)^Fc$u^|P)a1j2 zRqjiHt4rt_7w&w{&1%8;#}Ub6jeCy9x=x=>y%`RlC6Rx-Twaw<*q;C zcOwaIX(`$$HL|>Hy@V&rp4}A={aYc~`zJo;9AWs60K?pAOmrdD#i^Rx{o|d?qF8=T zUUlKQy#UDlG_xN8hUgQUkWQ8Ps}#{dT%W!#h|wr?OZM7Lq9SO%M6Pu3qY1`H~==ZguD;Eex6^Fqz04Orve=Y8U)_6FDrN> zkz7W!oRs6uo_<@kbM@5&_brT?MV(E*SVY}$8^9!ws@Z}<^~8;orm84yk(G7*QnjUM5C3nLfw z7tCKqz5R(faF&MOIgY<&5qZsX^71TH-Y@*KU1zkO$ZI+?73RFHggIn8qr{vsUO9+F zv3G%k$#-S)FhH+fV`f%u)BH+VSpYO7IWHx;9FCECpkM)Ma8@yCOIV(z%El2|u)d8` zpYq_E_3E%Qdk&{rgyRfIi|S(ZkQ^2bbRNDRj!`}Hh;CHpa$F_NtUNn9C5MD<}`k|_?B+s^b=t`(>MEV0Ne@yE;m=WE?_DXj24r$op zmFy_4Ki;}}v(_4W;~07E*8PcUTPs7w{y{t2YZbR`D-Q{Af(*CCDM_SH58v8^cTmsU z7($scsyh2`&$D;+-3(r{p`cKFj=Cpbw%0@KzDA593{WgBp!c(T_^_C$i@M{&pjohR zrYW7xrE0hmhk{cp{DS&pm*y1#oq@+| zJNO<5B*C;>!zzCrDE$0I-U2^xzC4tB4qE05Amj;q6TB)vd zg|%J3S5_3ujPzr?MNAsb&)W9{;G2qHJjU4X&!;`mEV`dNl*q5Jm~pa~J(MIWvV))Y zVreK{%qOiTT!U36n@>F?LDbbo?WqK>R&BVTiFAonr8^x>()+<^v+ah=+~O3QA}M?e!g54Tzi=`B98Qhtg@oZ3?dN4W>4o!XA^nM zbxkNf^@a88iUamrDbH_gkXRa_7EkM|#Pl;=k`EQrvL*c61~S~ym#$e4LkP8$w|ZV! zbhpi0P2C0K*e5ZOhNf~`r%W3i~OO;NWvlBoZ^%V^#l8z(a51# zJM|fWilm~|`Er0GfT!7JwZ&i@dN{rT*9*43*h8bSpktL)*<2|{t(txfD;a>Sy|gB! zx`j$I%=~;_^b0i5UoxKFylIzwv{cDZPfAqXPA~X2rk2#pubE>BAi!fO%G&(CFpG|y z+J2`>#1o9v?4VSB_O3>3g1f5!hqAYfii1nKg&P_gceen+Ee#|x!TeveSp36zhP_ z57;&(@p~`w(!rfxo{P@)h3DjmM57>xeQ%+YyjSZ`oG>C$`?pBjyPBO44J@QI0{u5x{P@>it0=)mY%ye37s9wuv(O9G$k4#d%el~U4pU-dPgLC8 zXi`cT049Q3;TA+gU3@D(!R1eS2;>nk@BH&VLp?Evc6j~HZq~I9|39-k#>5)>Eu7?^ z*^N~6>Gi43Z@J%hbHD8rKPcN5Mi>LqUp{5G(i*=jubT2F#F~P~?ABdoo_qB~d}5I4 zMw~Q!B6jsR?|~s|uHt~)_tX_7?XGVk~Wk!Ac)GBjiAERMoxM700}dmF~(znmN7<&w&N>{#Lq8@jHWKxzJZ-^4)9Gc|BsV1jJVK+HziG)hHQ`3Bovpo}voR#OjdSpyg}vz74D zZHhY4B(HEg$)fHbxJR^3FC^7@xzIT6)AF7Kpl+1m5%&l*MkHAB<4R?IK*pa{kP`Vw zo6Y>(82^1sqj2Fz2}7cZoMgMY*t{kOgTh=~c^7^64F4k4DA3Zw={IoN3x?FFGUfUy zjkZNQNg}Hl-ESL5*(Muq%&9;b%2r^(NWH_D3!`9tTn(j*I;CeSK#&1x&4l&Q!x zc=YpJUj31Gz0j}DCTpc%$0+p>gY`zZTE~Mgot9I@H2~RzV{4_U0=-3kb{S4@+1?mMC;%R0qSmJ(gn@JBF# zn{~aBFr^(|xQrhi*cp?rP_k5(p6~l3tS>{41q>+@jz-H_n2t+o@XyOgE_lhtDu2pL zlWBgB`yoOC@Aud`x4k=Za1o;It(-+cfQk}LQDNi$K6!;TNFsE zyAWE2ij`~EFW!*gadRc zT_oOY3E9?oIzT1gj!d$Sc{oX7s7On4xAM5Zx`xhuiamuNHvJW-JZ-vr`;81 zx$ak74l}+zEd9kvYpPxG>#alh+4jpS!{qkIR@i#9Bl}7iwc&FJL2Wf1_)J9i!%mBn z7w{c5n8_cIFe1;2AdFLtIfN=fnmj{ zSlQ`fj>**Rr+Pu`S8XByYZGnpVHGEvvuRqWa0qkt;KyJ_0dZzH`r6`qM%HfPErAUA_Fge!sj8QBkkN=3FY3HbJ_5S6i_yM5TS1b2dU{tcQAocU{(vI7$(Q%T zaV7@o?$8$ZIRP0=yNGcre7~P4-Q{et-&+u(AOh;ufSP?OcPF^NfnqN3=fTJTmI>Pp zIM#S{WRLNTwq%`WKtqlPruh(M9%eEN2A{&<2l%-0ZO;c)g}d7 zfT5=^=ie(jDeBeeVG8D<%xQGZGU^E{%T!@u_^|Eq8zA{OGm^q)czc>kS{+vy6_ z{qkS>j+Gb{kVRl(d@4?h&M=hG}HAFX9FI!q0%tFZJyZi zOqOM(J6;f@rIxdNU8jDTnp0J+#ymAm*2a-)6i>4rmcMQJ^VyFOjkf*_$7}*C^W|P^ zi6Kdc_EQ3@BHHvYoIGtVE6q&Di)qy9s>!eA54Stt7NU!yS-iyLP`u=4`!}QnAAa`q*r)L$iKTT&gfpC=Z#8Pkj;Yn23IeTD@aKR3QqEzYoim0L^7GR_2*0WfZV4Fd`J{NQ5I;`Ys_?R6@Kk!UKiF6h&AltuQJH z%Y|Vq$__hs10Zs@L6{hy=(rB`T{5 zG20F9XGP+(<<#nbY)iZ-LenaM0TE3ScYyBJ8QeibX6~U*${|#^)_=E5VVyv03v6G9Kq(xQVeX zv-WLrY`z{N+O6~XFkG{lLmpT!ppi73G@SLe!2aNuCAzC~`AFJCn&e#BuVZ>5ysS^2 zM1-i{y9mm$j`K;#pJ9U;M+hl2MYKLv4O?H>U98K;y}Oi!(WQXu26()_(%^sCYl%9& zp}UbA8`|oKv>M@6I{Ehf=Jq8#R!iZCe+%!u+z-*aLxlDW>JB$X5-3b6dn3%fwHI$C zqM-y$E8h}^H*;hpkUX#>l^VvHLWJ8C~7T6-z+lG=BLAxudbFRIciQ5?A z&&>Y}eb4SI+0T;z2s1K2?4MS#Pi5gdK;VBX3xI$1&Ywa5C<{2rsB#PuNPn4W>{+k! z1zzE{CO_%lrlTP)9icVMT@=k#lZP+g ze!DD$)FaQ`V_z4)m~V1ss9QR)Iac+t8f;+L5Gj~>MAvOg1BAz&jUJ252~s!RByGVj zcPt05SCzm1wv$~K+8YCF-W91SS05CZKirwUO}75p_(3+E+b!z~rK#LU($KuTly6vz z4}$`Bw&ue!SjZet#?5qNDuuF+`!37I;5eZ;-0guwP^`czgV_*a9+wK$Q`w_Cf?;C#W+Qv7;Dqt6d~@%01H8uh>{OQ z8190)hr{#?AlovKB$%!NOFNL&4ma8NDuF(Z|3wQ2rm0bL<{E%3i7iYMW6(W0W%Fm1~o|aEh`+&u*ECdfHrrqhWV*j3U=^gQ~}=YW0u2XCC?P+ z7#UbpHdxgj%64+*6!VKFBD0hsRu<;h`CH)@J7C^O9O;MiEQ}Rnjn`t2sVi~H$^7RtG(bFjDe~sDpfNwv7D=F6@grh+jx(68MtjljCaxe_X zm$!)Q!vC1}v$Xr5Xm6HD2C?WA27~ZqfCna=iNIL!n0w;yb@lVXm!Q^j9J|IDtU>yy zlj)df9P|N|JR`};@T5TG zkb!LL^;Zs%>ft9Y5E0_TLEPd!TQOt9K7*7WjAi|uCl)%U!IAv8DOHEh8=afs_!|0{ z0yCHfm~}&UFs&;qw7Sd_JW`1B07<5NG>=Z zIx{-S;&``qE`v)&`}g^xKA2ueTHR#9jI>^3cY~Wvh8z&nNiKiwdJ}w<-1mOQrgE^) zN3C+6Ny76=qTVNhfBI?5eMWPlHDOmzF}fObg1$)xu;swQT~Eua@fR94p*|A_3$?#w zKJO%N_4J3H$SRWP*?H37XWY)RM`hHnL{PU(m)`&dQ>Pss@`bGVg>vg0sb*vR0;_2@;#*} z8MC{E7dbi0PVAHk1v2rjCf_Gp9N!M6kCdosdj%WeSrAz*>>n#iLJaI|!-WCCW)!$S zQ$oS|S@Kk|XcD{EowDZFiQkD8Y8OzQYwElw2*!2e-9!^6b1ur zslJ5+J-mMJH29KG7QOp>rvY1Y??<*+;v<5ZOtgcJ1YsSKopVL(4rhx+r{!FK-}X#G z=X&%0X#*=nj*qydzqVKq+!rzC(LD}!2KY=23z%5HAjc*s@xgjp^@}&KFMo=cA%1opfbzN zjZ~0?q@g()$#;ssq7!WXWs@<>q67R!%ziqd(}SWwOVVo-1#^E}IF0Joa8aH`N$1f| zmf0ed;O97akB&;TeX!dyyu5su(d+8<2`DJQ$yRZG9hpbU^K;G8wb2kiP_Fbx+=>^s z^c23`d}TM(6_o+h#q&Uoy`nopS)eOpK6V~&iGt-fuew1LH}6m1(={nXW^|qmxthfh z!v^P>=<6UDsw4{<(#Wf5xh&BeNO>{Lj4Dy06&k2BMZSu&krd&Mf?`pMN5m=^HI71k zO!sJ{RP}DQ{vwtCIfyXDa!Vbf!*p{-kI7H9OKN=o^Z--fzQ}Q-^iN_iEP&be@qo4e zrr^Q<+X1h(JWZNNudP?heF*5x77HqVd$!q~(?1??Ahqa=PxP!?;yy(33cih_xg8$& z+7dNt6A3uFk_x zg>b*UxD?mdQgKZF7GdP%dEc5DTQ>k_jz0#+P?W)U<-NFkyAYlx2cDJ}B8_vLIq{aL z?t90f7;d=M9!<9{EHzI^+ z;UJM>wA_=CK7URmOdiiA_4m2B0`C5Qb_w+XUjJza{r|lB`0u-f&tO6?2>xv%NeKu5 zpgw*ciT|9Cp#P`z0-B8C1$1rUPBrtlws z{>5ak!k=Fkl|#Af36h0fV*mUl{`hrMi7k=NX9m*2`7Mq=zc|+vy-&YS2oB1C=}gLi zeb6G4#Lp8BX{~}jDOlp8>YAZl%v6;uHIb(pVO8S;5!+Nd#fa49n3ul>;?WU7u%Y#IjZETqAP(&D*)A?6wLYS`>8!y(XmE45;05m=m0Lo?cp3|5N zqnY-j%ZaG&IzE=xpON_T^J#n)FscaYGRls=^u!fc2kP=~lOZDy`*Mdc*HPMhpZqYpA>$YM6f%l8?7<*=_>#} zc?<*jQL-BaZ^`cSggh}vcVR_7`Y}_K%t73 zR+=I1JWs;haSuo2vO+me&ruz3Cx-Nn-B{Y&f67?e!S2C{9Fv*Z+9M%1XOF1IkmbEqb$R-kP@V_t` zYv-NE<=Uy&51R);K%vkonPRJ^7|^OOaN!vgQjft5+2{ZD{MUF~HSM6aI58SLA}fw; z$zfNJ=e3@IKQ}6)%oV9kR~?nLdHY)gs@)%78BOch2dFFa_M7`@CrXCAAOuYIXs)D+ zC0BB~o_(}@P+uV2-N)krsU$|MuTo$n#A>lYL)&Ul?2T#d&$3^Y6VI2%ee2L(QOdPH z7wjt2HUIufvA~>3${4%f$f2BP{1r*)Fs0pth2_Y%<@4cLgl4+|VH0wId_}H03eF_> zYm68LE)eL;cSWDgywz&Ecrjx)VUL1Ct>Jr##lUs%vsd;PN7tNpEQ3-XU! zm(tJ)>S-Q4legz`ohY~MauOtyuHR7zK>IxC+0>~#gS3>k6pBC`-@x@Mah5Z#piiDy z_(V|Ob~MjlawA2_t`pNMZ;!E?T@h^Rn>*mf>T`d}5PgxRdYcXj<9r26Xzo`0i(Fz_ zs5rCWfQB#yh+wZ23UWd&LE7D{Z}%xc#tbTe zf{4Z|ai`&gu4dok1xK{HdLGEaBVhz4f8u0LMEyZN>XWwdnh@KB0jwk2g{&K~TNjv9 zMrTFbizn*+_0CO5m5*P0#h`}D-xom<8oDkH>P~u@RxA=cP9bmqiZ=WNO_|VlBH+ebIh#KO0m}=id-ke58gi?!p(u)T!7Dpuv=FoDP##6&VJkoFj z=`Qm5JF~tf{EspxmE?ga3KGLq8+6(9b-BCUlAnkgNrHRYDxy0WDJ$s@K4Cr9Xf zljnis`=4RPnRs;bNq5~Fk5ZD zGThR1^?mj0*#qaBMySn?rIBnM@HzD6^l+v)TsFOt|N7#LD$POz>Tx#@AAsi$_{fzG zxxZCebrg&>{=4Ntv5TMmyR)(HrF`sc^mY7t(eRY;i9R8A22euQLM*5yxMINfC4n&b zB=dp%Ah@2_nm+0Uf&7Tf0ueBwyxQ?=!SBYMv%`=b9igEw&Dt~nlF(u(BZQ3-zgHcN zqA>8x;%hOeTo0D_FI#+=K=ft`kh{D4USnr2I$ZP(fSxWt5W;oz9e$-IJEH2(9In2HxY zJMZ1sKAM*{Uj|R;@WbcI8@A$`A6E804U7jl4@#27-~d8^F=X7MTA;4RQ8n*T|55#_ z0GF!ppFFwK$4md&A2jb1@(!?{c7 zGdm)m5$W9qoa**@;oViHB#v(mCQ?`liFC0olDt^PlgnQWTD%tXpY@R*4*GomZOb8LDV4)5F zJFygf1ncV8z$ZP2S$!sra4Nwx(njU2-ZI!(vs=tEfe|t~P*Y$e>qs*chnV(rB&C_3 z1er3Nv`wv`M=R`gr`AFIQE4rZ`Y;wo0tBb;uEEo=5TAfam?b#knH$$O(~0r(EM6ED zgt?NfUxS$0{l&4F9N`!QY#V(6L7_Z&(| zjz7yVB?}_V@zlU=9nI=-QZ?-t6aoTj*Wvq<@UwfrPSWUC4cs!%@g<&&SX6vq}6|OLXOC6^LZe~zNL9WODI%& zO4l}z<$;g`|B%p=y7#d>1WTz^t+sf9)%r^s>^29JU-RD|h{;1of@wb>bEqn=e$5b! z?q@a|U6IX|e^R#UV;CEyO>Dx4l3G73-@~zEmX=wf!>VA^56qKcy`a!a5CW+3CZNJ0d+zWPZJ+K0Ji z17;q_5RA7xiC4{+hduulO!|Ca<)~889Pxw;h|i`s`i=MJK)OFYG40hEO-1Ra0lvGM zX^+=DT$tjPWWuZ7s8&6G7zugzcVW zbq!>S%=5Sa))(5i8^LC1_WGZ-)_Su)qY2k+gu8J};)e=33S$M4(-JB03zYS4M#x|* zO~j~js?i9l7;)3YXkc&Y1KOZ!v_af?M0sIyANw|wkRA5&0&QricT&vEaHUc$-JSDd z+&RU!!kxy0XwotHU+rW#;{D8(%0 zy^VFCRP~cH*1qVkD{ZfVvGVsta-tCqW}5l3bTFB&lT7!p=eG++qWBuLxh|b>mb>~* z65%X$vBFBgLp{r2`nk#r<(zPM6ium!R~@C7QSd+z;_UuZ{?{lp7UyEs{stziBfLl8 zj7j*LkXR-&uWpL+&J~k!QYc7M5<`S6aJZ;fkRT?d>8)Uat^nKvImSC7+>!KZDi!DH zG@kuZ4Pf+(zVnWreaeu_VZKeU2O-o8#-&|166PbyRz9iFr5L4_eR(6k6h+8feX937 zORlcRlA1^ap9Yj7&AB}`UKoVAu#+-1Sr5T)$Of<=+*jnZP_ZDj5H0oaI2o)8mb?+k|e zXZoMkx7Q%8fk9r0cYp+1#DJo}l)^)@b0ch6n#f*XHqHk)}PKpRW}_@_`Y4 zp!*nxWiQL%n3Pf?kypCKFtgcGtYPYbh3>VTG3HPUVSanO@AQ2yF(QJ(vaZ zHgb?bD6;XP!1>ref%aCv`&oUUQaY;6ua!Q;ItIL6^rhG+S(=xJrvW{bs*y_5Z@SnA z&@)NSMj#VRHA?&+RM#N*^PzJ3@O>N#TJNm6@cTf*M<2#ijY}PzB&?4vW}pEqjw1+^ z7MK8`OrFbmtXjR01)H|JIwsp{%X&GLl~Sj0PHUtuS1Yq3BL4zYumEP*$L!eruQP~$ z^txE%AKzU?bMo;>& zNFJxhUN>fa#7h=__R!gek-r#N$+^`3-c-Ry9^N;0x3YMyTE@6vQ}S%1yUJvv-|z;{ z44Xp2lbT5Vd}hr50TyK$(Wz|hE1nOG#-MYhs+S*Q5xvDU2*T~DiXk8qhd1BV2iQR z(MF~ifZS%5_+Gfv>npYo(n2s{7dH~wy`*qw9SbZkXF^TxE(!f0fN^CZk~AaYNV%fe zY*mHb_mmopJo%Z?KEiuj($BHtNCj)9FE-;%l46!1I({m|?xsDB<4WfEo7-{D{l8hh zeUXArKN-b`kFhyCMbNJjOY^Y@y&)zpVXw@T zN*3x-pEzYR>6+0Gr+J2m@5{B*JYN2kgblQLo&~$msf9Y79!F1{Wk6A+$~Kqme)h&M z;bvwfT#cWT)CBx^?|OQs@>)hz9wbP?Uw6X70t4p>1)!P9a?fR2((aNzCp2E$1nRWW zO5}WTsn~&NOMXXuUkS--v@|a#d)^Pu77DU@5Y~A2vir1W1HZ=jyUYG$dy-@TM?kp0 z6>rPEcq5^SelPUb<-I#mX1q3k8xp;y4C)+}@Q1S?T-UWu@q;iW1Wai$qZ>%xPe>bx18=8J!Sqvq#AoC}WGfkW z%2^*DYp^@rQnf~8F#$(i$siKj%>m=XV?FFu@q9H)n-VIsITA~FQkNbNFFi+@jzwml z5s`#vzb;@zODTsP1ZmLA4%K#)bAMyTrs6@YI+Mg;1o9U#P~WzMTPuKls)GLGT&bZ2 z=t3IcR+G7bFbb2#NEhI-%cevswuane5P!hP@-@^rtk0^9IW({{jPPrmblY8s^gMna zDX#z*1?M~OElE@9$c^uZFBkDPw?qdvXxSyp^ui)NYt9Uu^5Rb8ec4Cs*NX~dm9;0` zyxMdSXVONL+3-($lkpVUng|xtLT^(b^KwQ7h#T_$Ryv{#EBNU?~4LoAj)1@B0_% zw@wjMLSfH19-GFf&GM*6RM?#V_YCb*LgiXt^vmBfJvzZS*dsJt#)@q~wwO5VMPE(f zM4C_lb{99r=A@5OIq#FUe0MG=8`NSl^;qXjgpswFSc`=sOgH+aQ5;Qv*Exod=kf9Z zk97`R&GE6$@&2xJKLA;N#a5)(<`t{`a8y*wgWQHyLhZOA=$|)j1smcNo#Hf{IHVw< zx?CWG5zB4w!Vl(-LrqtmxtCGCMk%3{>ky*sy5$0m+V-W^Y{`1Mx8l-sU#(@_Wo~X( z`wJ~4cgRKuE9_K)t4sAOdmwW<%hn`ha%)v@KG16pz9#4W)*ndkuAjoMYm& zxGzh6gtJn|)>fjHDHo&*1T1 z+Az+3+GcQ%402(}^!QVq=jrU}U|#<(XOETjRObY49U`)Oo(2Xv*VZ{*QM8KQLWO^y z{j1(n!@i`y&fZuo4uSx}dp(&ymQ7@f?dB{Go$$SixMx<4&J`_Gt>$fJR47#T(~ieR z{Z(5s7fCRIF6ZjF<(aIv&_DIFylm+U(cR7X!S|VoMue8-?q8QIy6u6k=sg;fnS3nr z6z12-&TFeWjwl$b>W$UC#as`%+M(Pp1j0yNbz!``I1n)@E3UH-)ppmLlFSdL17Dt7 zF52K}TSzZ*_+!eM9NhCb*PoRQ+P>WmghBCNFR3)xtzUFJAN-si@_=w~*R)Yf92oGr zj8_u$!%32)?Cl$_ogddYVXcdwJ_zc_YsSOqOtpYu3DYj>W9d^d>!^)-%o&MQ}!dF zLB$N<(B+y`wEf6w7IL+fpz7-8NJ%G^Pu{k7`p2nK4RcsRshQENeH1zgRTx z8olp<6qapSHSP8M8dp3TcTMtu&7}R_sy4R}UnGNOfYAmg`96y0(nkHHlckj8L>Rab z4M!0K7bT>$!(Q@3%zZ9ejFrL%6_uiWb^<+@BtJ!kKL`LL#R>+V*$n_uOHPMq@ANSS zslPJ69O2wxekF|F^cDikMh6TX=L6#Oj>?%E$d1ft$B41>0C6s+@sVzASqX#{K0KEk z7`&KMLPkX#G@x4>nWT;my<7z0xsCCN_coumdq&7JpH0{!q{&gVH}C+St64KOta{u4 zL)n2L2S3I=;aaZO>1pi3hpqsyaI5hFFXXy#WJv?dvhsPq>=$Wh%XO6?gjHo#&RD&j zd2I_iINJTwDl_d7QiT2Yv7Dg46s!>L>}KX)uM`PC;yP7+`%Wliw0S+?-m^PrnHN%)1CoSNE@Pd`8pQW0;%$+EfDKz`c&F0CI3lk{N5)fO^ zUZ63dXF$QRRbHh!)xXqRm!Vi?ZgVIQWq2OwACvvNLs z@NZq~)iZJ*jt$KgPgEvp5x!WpYH)R|Pga|ASuFj$JQ`>Vc_1&J2Dr>?HMwJWyxJ4| zzOymVJ&aR(S8@G)C7utD_u&B$*OkT(1exIIO%Tz~NPK=0&cA1Pt>~RHH||ejIcuBh z%faIGjf5`cM<)_E#u?T}60cWR1Q&GD6G#^!mW2dPI7M+|8lF@QqB`9Lg^LR(Y=|H& z+o61>TZCkxfD@pd;jht=CctH{_PqZRx_`lX1IsTU5j1Aec6;3!UVm#BfbwdbK2KVs z^~C>gwFrBn^<=Q_RS4OuyuDx_&&%28BD%y4i;0+=gzac5wEg*FufD8_5Vb9?i=ae` zAZ+JQ-zi9-STUZ7>r)__@`t>oCWLukW&Yn!Ne~W5`MB5#kM~=|-BSpvc z6^6&U7zjjwFAxhLTRt6NscWze2)VjlPxYqfZODW;HQRQQfX}ZxNy-;9^*#$B8~BHT z6HXq3kCZ#d69}V_q66Vr000n&@b*?d>>w5?zi+Gg8u62>o4n`tyJ=F@oa z@76m}$`Mz4g01LXHj|JzicCKeG_HC_z}>6a?U5A|BO0S{lB)FwTJ7%kkOxmEr4_;5 z=?0KkHk^c{l-5v{d@dkz6(w=383W~yOd}fAofrm|0#@wIiyP=U3)UJp^NNq1i9k)q zO(*yMp@8<$h5v{);{P$)NbWVRnf`oUC8P5uO&lff^P`phH|f|lp-7 z2T8+q_Izr7Guce(RO~o=_ROd2yOsC@AK(oe|Kaq+%yR3J?uTws#*e%Lm{t5b`)5~% z9*GY8-x1yck|0}=PM#2FB+oQ=)JvLKm3|(?IKeV9%2nr}rKwf_ZnU~ZK^`*xWgsnl zJRwf5VeIn&W@4qe5FdG=DacTeFuocp>)3i&UdL=2fay76yxNrWb_a)kROc9fG`%~H0rlsBLF zqaf1Nn96vH7t!nTo0c!F4v4of`34Mxer|k!co-Lq1rT6k{)=e-gzkx8xc|4#ZoSJDpXL5-1KU5I1F;Rh zMFCm=HX6n_>hQfbhYUT@q!x*O81bjWd8sRLL-HWmz*{t>{_3a8B9S(g@jathj;N2o z@6Ik-0L?pAY?5L+OM9Y7EBT~^Kh#F2KZS{vd7;UqOPFX12gVe?l_-K|wJw^e{^mow)ntAom|9}CcH1C{4M8E;wejlv!~FRJr{#Lh z@iF!hmTYHC!$r%=hATqxuR#=?xQE>rZdqAS*HV~`a6SmC9EM{eyngH_RCmW|u+_Pv zq>lBuf0~e&L7Xn;)ogUA4UxpmYM|3y)&f$T+GPk6xZ$J|g-Cve*r*Q*CuTm&@R}s( z-5XG*>BH^m0*h6QN*WX~HD*-A6AlO`5(I}G8Y50oXb`Sibg@p3gMOYx@|+hg)D{RA zsbJm@7qykdn`GoqQhF<3;9_*^y%`16;#SpTpL^^N9V!{`op&6-FIo0K zAOv6}QefAwSKW;8-IolMH64|ad~Z5fuEuKE+YYhF{;6DSUakl8ZVJ$DGO5_MoF;+Y z23%Y?`HKEF^rQ>EgRxQ^qbQi zV^f(Zs_0_H5Lf-Xvr_!;gC9f*JUGtBkRMW5#yMT>X(o6GCe!;QfGn>jU}Vn+#x%HC zY^KKShCT>$w;^8)L8*S7&zX3WA-%MRT7joD&@3aI@ z!$rD}N>l0#eW`V&3v&@`oN;%Zjdx{f68ice;}j)C5N;ghL{eTf48zp+Nr3$=pN?qs zgHp{!;=!TYS*!Q6mseRnvb+r^SHeMWuV6Rr2kOhab;>lG zU39mHO8LsPQ&`t%%U8ZdkS*KkulGPB>xZ%3H(pomjKD9{m`rSIlXwM9Bw-fd)|j`+ zYcybr_iT_AalL6<9=e)o86&lJ__WnIWU<_P^Tk3xJ2qJXY)!;c&Cz&K_5wJ|u(3Tu zuHmR|`_LkjW~v=-(>H6X_+fMT^h7btez4Nv+x4~l_7b1XyMe14_`q>zSbURJFGSiW z|3}W?jjR(gOM37GK0%nUdxHo zcLq`fKWrvYg>WJlWQXw(_l|{g-=FG7%mj&TMv2*}js{6$&ryH65+d4)>1`~?j+Db^ zpo(*UZ?Y8&wR1Lzhs2p0g}r|I5`-;pV z%^3Si#;J*n2c@iE)FVs(oX!Cre;8o?+zNnTg8ui7dIol22R-%Q4oy+_*q2PZ2f4oi zKk1D*bR_i&f)$-U`^~sBoU7g&&zH-d4?P_yXirc<=;|tZUo=w=F6FJIJ4xfIQbV|f z8>lP&<^jC#`}~&cSb|wgKdO|gJq;HU1Q0LU2`^sw8pzTaJJ7(kSQf7JMG{0$STWF! zO+%H^P`winCOBi*SZkgINnk+$ru1s>ja2A|;HTE*@=O(b!B?j=Y8|dr5j|iZJJZB8 zyG8GU;!qQ+CKGjl>C4JxJU#b9-$Do3BMcc@k4zsxH7Q?Q>N7nIF^&|hBcRC$fIZ3O zD)zmdGca1J+g;lQ*Lx7Ei}-o125Z4yK2+&&tUCd+SZFXp-L&V35B)+jG4DhK4|&Rr@wa&||I&nAub;jbct z;^8Q48m~8^S%x??KfaE17~we*gXxM2#zoqSmw!XQu8-Af5Ta&yH=d;qtrN#i#GDtF zOJuweiBRRO+}lpGo_kNN67qv#H(Un&ziN%_<-Wix+VQs6V@%HX|CorGhVeN;AtM91 zXs?J$-0Dwxv_eP1N>!@@~tP) z_b&(EQ4DX*EL$+ve`Q7UT%de!g7w*C?D*VgS`q$G^P{|pwt1kuH6I&+(soCtB%}N` z?T0tg+I@$06CI&FSh%rDG`SZg_-6>o!}9YZk_Wp*wl46*yCv{#8m3i4Tf?b03b+|_ z`x|>d^LfYXwv!Ps%*wh*4^i~~2T>m3PaL1&INmu}NIztrL8O~>FOx5WiO%QR@7BwR ztZGg}riqfn4EDd?*kx7On^VAN3N}Zi4V2Hi#FjA)n{vlE;JF`3(5A)y=L-vinWb1o6MYEC7J+ z&lIKbe-gbQMTGvB6cz9QLgvrJ{wARcIom+vV%wDeGG|S?!}g|bIMW?T2YM}&{wBpD zKA(#lVSs<%72q$rE{TAI#San_&RFU%Ose}tLg!2e*UNu^(9M9&N%FM zS4^etCCGQ^qsOwWjOZB#2Bo=b*l{xHUd#N%9O4mN(6>dTFOy)1+xdnG@aNxI|yJ-de}UM?-tgu z>hOf?0$gGiEExUdFj=zrChs2$Dd0vx#fJ5(Px2wHsv~p4rYT6aqJr_6x9YPm(6c$P z6piF-8sdU{%A=`V41UeHovv_yOD8W&Jsb$N8pi8d1)yye7^w6RiI93CT3(NMSuB0P zVC4r?ZN|AA?dl6bJ{yb}#=|R+2#DGhZX09Y$CC;?4+T$s>!8% z0)5m<^8P(QpB9?cgk_O1$%=4<-&KtKTiI78-Pe?o=lk5B;y z<`1Z_+V1gRt^ul_IDxGHa_0CN_7Be)Ur*$$PrpZqTwXLCzvn&_IKQ>Zq#u5htRQlXCy-~+w_ST%-0um+Sq@)3yqo+dh6uXhG@#q`Q|`$pwXy%y z{P>%YPPTkY_%A{_leh;Zf)w6co03rU^)?$acyBlu1e<^-_qPt0#*1y9 z0WxIf)}AfYp4c?4eXrNYJtC;0(WDwr$C5q0VJ1q{Dj_YHR3ZMT^HqBjB-Nx+s3o;V z&>);hvXF=)liquZ?ByX127cEhnA&I0nwPX{)2YsAScqM|?~J0L<0t!GZy$8v%UU`A zay;H${W|l2p2OV->F)3nZ$64@xxUhrl&pz2`LO19w2uImYATr@Q(B`XdW1`p1k#>x zu%X`DchN)^`!{|OS`WV6f=or!*IZ*fQ?|;4=1K75}NFeRIHGO3rx@KHF*6Hh|$)V(K8nNaPPUB06n!YE!3+aAb8Bp7!?MAQ65iftln@4 zQJ0F{pgaSu=CiqI0H!z~647HEEk`~6HTT{!zh;;zoq`LJlB%}rvHebZ!JfmqqZEwU z?gv6rYI1&uXZyQ~x_gB4*}tzxQhsqcbsNs$y?lBaGr6oR_j>Arf^pnVNO9ep_M*Zb z=u_?u8=3#ky|AYLo7^jYl2|%8s$YGQSlWzR$jaHyD?a{mSa)#Tbn@hGaXxMSz2$ma z@;D3gpZaeA05cPe+CSBWVBp~XJsL6oGaAMIk7$I4?du)OXoDd?a`-zMU+NJi(B!u3 zy!dO-FRTNEV&A@wr0ezK^_l-2jnNP6`-9vzdt;Aa?O8WBMd9SX)+5l})Jor+EVox0 z5ryF~MzE+=JVqnw+q#_h`N{{+awO$8TReD^`^ltUQ%WUxq_j$j#eQ%577;BGM)(Se zVjzfjpJgzyx>v`szrl2a_6QJ(O(Pn{Y*Q@=Y1bR@Kl)N_+2FC*?EKh^eOFmfP zbYWJNhZP1NgQ&_nH{wOpD9v*$x#Ygk+2+Ne3)gZ}}ayw%aX1J)4eLq?{Wn->Cj~Zz|0~qG~ntL!JsI~+=Qk@$D#dt1jiEW znspnsi-BW7s>7P?CVyRmIxIZ)Lcm@j9eJc?)_FiMT@0Hi32ftOGLqsOIOzKx;mzz1 z!$EMNu_H7STdC|k30hH0cc&u0ef+QVld*5kNCSb+^b7W;>w6}mtT6GLx>tam9u^J8 zZ`%g<6^_pYdAk#LP5EFI0okp5IuM+X80{Biv{`?3O1*_Fh)uE2*yB@(TAfPt|I+>rS&w z9A_8lSGg43IUur`Dy5E6n|H}NUDeD^dQ`Rq|6UL@O*G|?;S?P1HMhA8;` zRL9|@VHN+v-mbC3**nMkxBV+2_hpw}IdWZ7UD@n!&5PHn?|vRBn|j~$%o6;e9V!G+ z0fYf0zk<~raQL6R$}Y#L1|qQT^_PZm3I=|=tB@QU0utzyb*KLJDkE%gQ=pYj<8fpt zbf_!-nDfL|4j;9F=5s#I7YzPb&K|`=L=0k>8Nfm~f}uofzr|=zG*P#&%jB_}B z|K)Qt-?A#*Om|K-ILT)xzj|x5j7!0FkSSIJ9S$V_b%FQkDvuNEn$H8Px z68Oa_irK+X_t%Y^BDa~0p<4Q zs>+7o;&I^MHzGzE+s9&zLUt=VTREI%1k|~kYa%px6)54q>O9gSvjXfh$zR2unnjsZ zqV8WH(p)^Psg)@qIqsTSGCq4WL6EW>)j~*wlS$vVqQQgIqKfVVK5ljD57N>aQC`63 zsvbhi%Hn=X>EhDS{6wqzf2HvJdj%5uKiTf3asJUHi2Jiifd0=Wfz;nk0@c5p1Z{sb z2^!bCgI{+`QAj6O3iw@o&-%c(Y6`0ZE#F;V+g3xHjZb!`!-_25*Pnss5)hbhs3p&C z?1LZ|IbVJ5$X%DETOAZ!M*twiK@yF8qp;48Z*Jy(p`gdjF0%NgC!6pfym{p)yM3do zm`5*XbnG~Hzx1##cw&%g)PBfb_D>*2PDW4Vn$0$UaqlJ9 z<@BopApjuSI=wu($G}BD1yK7BG^cT`b2elcnJy#F`y)xFdkq1&MnSP=hA}LYRAEbYzEpoAAq8rMQzxmN#GLvA~V+g2^ zrF-Hmu>!>$WZKX$1wDEnhS+jK5zLC&uPt+i!Ax&_A%QsSDGB&N5Q8~Ct(K@9EV%)= zSelH+hZ>cITAu*Tscgwk|wvlrVB~!mOY^ePbe$>s{)Dog@T02r5-8}54?)D2NXn0!p&6M|3oDP zDu!QCsc4|c*@HOxfPy}whyWW$y$6t*x`iYentPtlK=D>rcG+N}iF*|W1?6Ps?WiM~ zx$Vu@{sbMTZG5&}t=VITzju7~AQ4#|y@N1T9`)|35%&WKP&bYMSkf}b1L&re$Ajc7 zh3Rg0|Bj6QLyi3JClm)^MB)hWhcXER3-i7I-;>0|VI!z)$Sv+QON--QYrvu|QP;cu z#6up0qvr3x1U}3%yA8i^x@;NhcV0N-Y@~Rf!Go36IBQp_Qs)b3CA0~@o1yUW?5`SG z;>4g~R?TR_r+W%@icjIOP@~u3?VcoY&DSA>BB@)$RQR_t<$Lqnm()^vb^6>9gaUCD z-E_VRj0^L`Z!PP+eCCZj5f%bS zE&OWfIr>?jI;~N?NL>?hTDIY)9`3v?J~4>EDAg&jlG?By6A=#7ACmaVf2uFp4$+$Y zAxd%P)e;wJ@mb?(e+ImRreD?9e}Ws^rrX52$xxb#X0utQ%Ikd5Y@s6oh}Men*G>(k zDwO-Si!W!Y!B1BMzi_*)g)&xOu7oI0U9PJOJh@br7Ph$(?a+L4wVCO9u<+H0QRTTj z;`~0g4Ugc#ReQ>rN~>fIQIp47+Rw;pIYaf+2JQO2tbMi%z^6F&J8bzI@sabp=h_zv zJxByH&+JQP_{?#Fz_S^;+1vfk!#jQiV@Zv(h?3j&1GjU0e2vf3yh=zQjB+1Nx;0(-s^*{T)4eOV|^Zb%_`p8 zyhA!wg?b>m0l?FQIC@5Yfx075uvjQCX->|iOi96BjzGk%?i+)gnrsflpXizc6IpOD$hxBOXduF>QRJe2d&!$97<50 zCisCz5%C@sMo8NeA69A3|JlpvB)P{IT#D<~4cR0(p-o=D>hsX*5kh&x>Cx$)A14EN zsm^EJNPO=GGX3gc0y`Rt%f(34dR?~QyAeim#4r2g@zc52y?yN=Zx3?Fz7n57cNKdF z>YZ>eSKO21)<=AYvb}zjY=gbB$T2^ zr;7Q`$`rU44`Tm08UF2HP7N>wVE#(SBL9Xq*WUeJam0jeAR7KZ>gJ1O(+O=q{=FOd zH>mdg1*%~H++Pod{rO(En=Lu==Y=&0`|jh(8sGzQ1md%D=6V{{MqAWQ5TF z_*BV|L%D~it_cQ+YyFq)fQ*|hTt52F1=aZK?e>_zxDM{kfYZ)+331e22A?xrArJ}m z?&e}lmgI%W67da6W2JDp*=MmEg4LJ|B0KqxSWhu(?RqyNDcK6Oy`D-Wy>xedgX*<+ z)d#vC#(3;}%eqU&4#r~HwEG;seN6IfuG{ic-d8*uy7*%77IN4+BGc0N`l&5a{#|6s zA^U_mNms`r_#mRzY7Aa1o560rR9u0Z+1X;b*4vy~{fZ{#a%)62X4BiN_8oMr3KDT| zjklV@(pOjYLd=t&_h-oc68?dvSld!BjHn`{t-1S0a7xy(xN=I?s+j4|G-G^+m@{K6 zp=M$g4UM8&`Y&k6ACp%VtYB*ChFDn_GwePcYrz_1)#{|?2m}Pa!N{Vn(w4QH-wc$P z4-%)M*v_{VLcW2*u%mf5H`W6XK;PCi5OMN|%tUP@j9|#aE_z?{yNcoOQJc<2Pak?z4N=Y2azD9i7iz42w9NTb|?c6}IvXYo3#RlRW zZvqcxBhzA1^_~>?F;*X*@!7kt3Pu7{~Mmr&f{-G+xS|Xpz!5PW-yEugy z9%dOX0Py%j?iy!-PzU-33D%8z;KbbJrJD)Cu6DoI5C;N8>jrHSzq?>WkpjRP@FS}l z4Wn`SM(TZy^pY{!#2oO)U~KDlO(W=m#GJAEp<+Y%<5+e@w+z41B?L4C;u8nD&DLy> zvM22{r+qc-oh?5t-_g)STDKuz#e=|bqgYU7rx6VYf1taW zci;ZK(o3YT=8e^E^n-5+HsryIMzOHIye5)!QHK=MA6f+le9boUU@`8DT-S=MomG#+ z+rmauGMj|trQ_1=G_YdDx^1hn`G8L% z9TT$z`llw;`&YnL{B5Q4U!euQEO`hm6cV9d0c+5whs>!($HeJ7=uh?HuYmPrP%aPL zw*3{bcUoN!4jQtah>EC*ys;$%(G{@Sad{oz72Wv zQ!dAESN4wx@wzccA?WCr5&)lI6oV<&Sl+`eEbWzY=#`q;=Ylt4uR~khFRE7aB0dng zRiGTNc{7vEQXaRTeEd4pZDF?xH1?}K5j1mm{^_}O4B{DQyNERbc!T)uL=a0<3+Hr9 z1yYlJ@z+hi_yIRup?EbnNO=}Vs>7sWH8QYOwnOY;0G^X<;Gzz#+U3|wRgQN69N7*$ zLeD1q5N$Wk0}~Aie%)tz3iotGbe8;_s;Y5|+%IC)XrAz9%S%1(PfH^BF-IpUt~T;5 zwKVSEtaO-7ResC2ol#Y`vJsEdwlaUS|Bn?>Rs{b{o)4Za^#HQpfB5hpo-H`2GQX22 z=$_^j`FHZ9xEp=2n+zaBmbxMS@tb8(?j%w7b@)Nc#IN~#TZJR;Yhquk#9?lzdO5t03w@`?Y94^^XL2*tw69!e7gvm- zU@L>qrqDknEUmyM(ro<|HTilhhaEi>23J5w2xRX@Z*RK=rS7key;=RjY2=3`&gbDi3aiG;WzI1 zL_8rW&2)6!W!}!c0F^8M$hfqCco>vj|kz~j_i8dijg0lWDy zZ!|dbhXd*4%a>9o^1G8b1ezj8;!XO3z>jvx&3>#6<$z2V3Lz?R#TaBgRF(AF5`Z?M z4=06Y-M8)viID6}Cj7$DY6QZr%+WZ@nR>042m+{TJsQn+BhNdYl-R!7px7oRr_xZ2tOehiSR5I6bx&fUdz|w-rS%EZ9=xWWRt2=JC}s2Wi0L zyiVSQUmNJS`qoX5}L#TCGI*VkjybSuyNgsDe6{rnp8~}jXXrf$*gEdl73wg z+bQqhm$zU9V!Nz~K%jcZk1>I6rXg}OTv3xmCrlHtz8)8Xnm*)qd|R5G==ynVNFfQl zCv=p;Zry95u5>!#u<@HT;iYlxfN9$fq&J!B$%QYpnB>UU{0XzBj-JValYHPzR#2fJ zfkCVv{K9x%TLpm_-j65-*#J0-KRb>VW!IfG&o$56Hc-`VW$w-JGqzug0?F>uONbHo zxIyvu1ie-vvKI9hU~Lh_0a(HgnW1uv}Emh!SwdKh{f>Wti6uXhvC$ zAbcSuTA2~qHT_+H0R0^jaCdo~KpI^wBSG&o&DB95iO4ZalD{a6Lc>ims0_y3_7k4Y zA*o8NX|&%!@Mg|$Vci*1o_Mw&;c2Gk09QHzr);BA2~Hz1vM8SNo?nywFq86({V1y* zo$VOM-t^>q{sF6#4?J5|3IOEe$_KF;M$RV$Z0m+`2{ima5bCR6kng{>GP%Tt{4boH zHb;FB?HB$gx@+&EbEp`8&Xq_|x*_o0p(+ODJvS;p$x-zePqGlTs42_hvRPaNB)6MyX)b~x^zR?x9)W1v72uUrxE?tcojNs*QbWlEvk211#=&|1+8@# zjU~8`*r!SzS>)X-xL5SXvltm7xUJ2{lB!k^J32f^1Uuvt?|E6r3AXVND9}WH&g~v| ziQIP%eed)Zn+eN%(RDV+eHOe3e4b`pG;f+KCi4wMq1?T6(u-zOTB4x$Lq~K?71<;e zM-$6`dQLwqiG~Jt|E)pISOmi?HJDW@aFV^crDu`dp>+)v_K4PL!Jh40;jqTfnyCn;#5tD;W!W?vS3Hv4p%)G&d7dFG!)l}kp!Y7@6xO2j zQ%JHLbpNcN4~F|1Ic&CSs(75I#cG?zZ*JxCY>I5>IntIhNf|$$+7>o{@_0uH*_?ML zB`#kxxi}3K-%R*aM~h0i>r%PAD1YA}w5EC4y)=C%|DJ;s zT)wX+v%KC91V5C&r<9yp9&u9JvtIMn6(pK>H|v3jV16ij=doN~Me)!Csh)Mub%tgD z@~n0p;-N+vUH`4rjannHLs)cxPEf%YaU|W7W@y*reS%6e2h0``NkQhV;dS+Zt=CB_ zf2#h43iOa=e8K$a-H=I}gLQ)6?ui`dn58dHTf7&v%&RD4X;QLEO0+|iheqKYGQWad z-KMwHYTpvgZiMT-!cbzz5tk|y_=5{dG1E|&NijDQ(p0IcPK{=S)QJV%@vt>OSUKLP#wnn=c@0gYNZ5eEbMu5gaZIhnYJ5*Pd|8oN?UjOc`!Drx6 zR8?N{z^m(4XRSN8zBezw|J1Ihk}s{>N|rk~sWPE!`N7}HeVfs#beQm#QKv>gy43rS z{sKQLt?**Kq~?`0fPCui`e+Di@+WZIw)=+tU(F-{fS&;VKLQr!2L&_m|Ie#m%IVB( zSGxn))&HY4T%zH7fH=uD%V>l7ry`IKj@Q+px=fi|y+snrbkUd(yw8S0`nmBeK%=i5 z6wz3@*!<|W3gn5!y43E*e8raDqztI`;5&qr*O~F2BYTtwPKDYGauz8};)0 z-LBVs&Bot88Q#NWUy_t<6y_=uar3``uJ@E1);!^9E#Ge{HQqaSS9z*59Hw(fd62-q zP@!LitM@UXUj2PznsoQhC%1#0xiZiFpW|-lhxCj7_f+zyd=JW$t&qTPvU0?>(*C;K} z@-8#2I1xRlH7?Zt`HNTqb%aSJp>w8rIuW*aVG|~PyGRlo76<`>Q0d3>@Iod(MFm3H zkrpE^HlTxX0tDf)0s%6A53{b`ojH*NhwBR7aAhLH4-J2W#vc|g7L?`rT!{$lv250N zd6)a^WPGL0y>ibl5W-d{uG(ETUKMfl6gWZ>_;K7T4z%#JAkv5J;zAmol%tAoeR zP#j3f;_&zBd!+Rm`6CGQC1iOj0(-_w8dKuCLFCxgFvAQiTfTITH!6OfMBZlo*e z+MX3H?G#hUq9=!Pw!o5IhZ8+8bU<>Lo3qCs__c1x7k~%hShUGxr+$QdQa9=c;7l38 z%qVggrt##j7r^cFsuEQ4%0%Y~13Y~?ez#G1QjCg#E5iXD4}ws0CwMiqI2b3s9`70c z_Er#9L5|+1oKo>@yoZ`bo$DE5g3a8sY>%@c-(k?%D7-$ZvY7HwtRgK<#mh|l5tkvY z=5dl%EBRU94)T`Pdpb@HT4)RT=K41VVBkQQhV)f?ht0v&*3)ye>rSy4-NJ3ZG&!R` z4OkI>#nUF>ztI`;j8hE+-9P!2&F#O|L{kehyf=$29R3eRpO}XfQ|7UzumlDi@W6wU zoTo5|A?osT>12uMzU+GmRj6vIMt$|o_L$zR-d&H)jd;^{r7E+@=MBSZ()bvP4Q`E0 zu?NES#?!mYEKbyOx*895QpCa#RvplD3iD};c|Et0A6YW<`D&HH^^ekWGCaVQ80tz18AJFpz>bJiN(iQ*MdHa4d?WqyYS7#1lbPWT;*EP$X z_em~i2?KD2OT#qVG!#Z(c~g=Ub)$RH)Obm%5yl+I9Uxc>>`0$HZeU^@k`hSgvat-+ zG+LE*>^e1|geZoJzSfJM<}nE4aWIN!*Rq>S@}e<1j`K#HcF~EJcp{Q1iayE!Wz#B5 z)e`K4td5zY;)o{;=F{(dQ}L!iT35mHHs6FXq83i7%Dv@Q@W4=|-%4|uF)z`zDZkpx zCIvbx1h|5=UQ7BDN|ELU|4ln)61U}lTqedY^Dt2?$In2)uS+ZQvDs|JF3I8JGfEZV zQrpNWY8y({(C51^n}QZTZ`!S_+Jq#U-oIx8l!Xe3VduB5o5Tr70^%Xml5Q^8B_L!Necqr$cs)XMR_wm_DQ)8>Ml zJW?ghDWbaT9sFt!>KJXdVf~28V!yM8?sL`PRa?c1;Zs9(NAnA&I2$e{<_dNW00BV$ zzX>80%lFvf{LhSM)L4Bdt`XK)T3UK8UF2$Ae#nH#6!Pyo$^ z73&qf4lt4B)BDOR`mK-Iu*DnJwqI%DpC9y)C0731EOA8n#|I((Wsx)zcPID%{6S~y z-PQlJ1ODd*T@B^l|6-*sHYKrj`&u6WH$ce0Q^>eg4fI-ya^aT|AAhQe1$jAVK7R`x z!ytP!#UK0BI5YNMwo67Ucga5lFZ~P^mD93kv%gPLfF*Qd(E67$i1kxYXAmCTi*pbLIqt_x4Pm1tqba zD z&ZAKuBlZQfu_m^~Lef^$2Hx3U2u!iWLPndg=1aGb&y%Qj{MxlXuGs_Ck>*2z{@6zW z8`oGi-!h!(Emv3-4+oPng(F_Ia6KBE%oVR7wD#EFovj@h%>Ka3y}#JvxVQ45z4`D< zPXHQ~+|r(5z!1yzqc(E)<1HF%cB}B%YxC{)qAe$ix4h?vI8_1nUp{p1Y8{Rw#Hh1T zmS2QSG}h$!)bg8OUkWyv^5yR2++I->>#A&FBxbTTHRy}8G|h>}dI7aSxypdM<3Ti{ zu3Vbn?zwo)Kx#sxFAlaH_PuC~*=D_H)iT7iq5HQEpMpu#>()>hTDFq{PB9?UTDy&F zy=W|ik{f8o9eyd%Utvs%F+MHFy(H`{xf@i*W@i5JAl@rlN4YrQ1|q;U;*5yTnlZx|6eVY#RK`@~>_U zkPDu-aQWs)wsA5S39Q(o737O)ibdr5hZnQCB(90Yo4vB&iA^bcm6Dy&$Z$PZjd(d{ zTuno1x?IvTDpFzJ$*{NN+eNs?YB%)ifr$(DjCDO!tzZ*Ywq_xzK@dpZ2lTjQSl}We zaRS;9@+zv^_Avn>Z4gn*CvVp<1FfacSN0COt@hiOdal9$=4FWXx0B7k_KWm;Js&2T z!~oiX2$Xw@snEKDemzRz@IuH0d?b!ZmXbfsfqEzkQL`ztb)4!$2AF1-+^mlFV-B%3 zmsgll!DoPY0@3X;G3rv;W6JtQ{^pakn`6)D zL^706Y_r|wskeJex%*U@Je)p^Rx`IFz2-aUTx@wuIaO}!i8qx>}`2z{0q`ZDnUW*pTct> zniU7tL0PhwrXynG0zLQ|aE>187BWMb&*#z}qbN>dKx7p+ksyKtRkFKmh)iB;)t$tzB8zHvYqlWEdz}aS0Ipf ztB=!}KVi<0<^Dnn5zMdP3q%kYIR)lgoJ|NAoZQA+bbSwpfqR+Dxx#xFM%_FjS^B5h zT{do{Sw03;e+`C!aNjpvAm#IY3gcjU+bByhBWgV7;;D(iEWb+BIn zifx2wA*YDe9L5i_%^L;?#G;t+43SH`XjvURLP~411DrGUJ*7rlu;3T84bG5bad3XDA9|t;&uTr<*TK+d zEp#HDu66+z2MSMBL^tU5J5X<1gyVScglz?xv#r)k+1E3RqeFU=yNtw#=h^yM3HhhI z_KjqBtad z3!@HkdOEvKZnVRDw$_rzAt%o zF_Q9Wqtl+oW~T(YGxAm)^E7wZOQ!2_*m$sieK=+m6Re_NAxvv3t&(e7^u=TG=-fTs zsjsCHczoPZ(Wiw5=JVGIqe>@c4%#f%Wip0%1z{wP#XOx7%B12wnb(O7`U}yYw=XC2 zck#uxamp1_q&oC(4f8~d%T*qgRhMa*edm`T(DOtb%J{^I5m%O%~@l(z#iaSyc z{F2|Wk$XWYx6@DF5wnA-(PB6n5%!$T&9Q7YM7ukD;Bjq5f2jfrAqBjNburL#Xf#Wj z$Z)C29yypTbjExz!+H;uQigbIvJc_p6%t8H?bjjad&%tYb~Tf9&_sYF?erJ+atvNM zN?F3UJCiKHdI~mPmVS7T!C15awZ&JzZVl;=QEknKN#S9;cPZKpL#s#$^a$W~UPrpd zN0^Ma?QPDH98vNm-?3w@4(oyQ_rlS3FVFV2C#7F}y4_UsluR9yV<2DHvY6+ z^{#6dWTxeE(>NF_uu2ajaQlJLI3p_>r$%nMfnrAE0SYf9&>Q=Qb-|lDP=KBjm9c2L z%WbR6T0|$O+N!W$znbB1>jGD9?n zF%}>_&{e0wXUa9ooUC=y!nem`19+yd6KZUntb)ldI4X{Zac|cK(QG?%f1t3p)Flto z4atnFnRMv)qOsO{)w6V4i4t&dA?qlF;$ZonjGAV~a*c7I?Vk!%Z+JZ$dz^9HD<@|v z@U*BqG4BVONveIs2xIMcgd(kys!B4xZo9DoD`i=1GmD|f0$3K;!`sE~h96^WwhaO;Gvl|5j`KbpANFRO`E_5IvCh6mUCR0y$~~U`a#?DO{l>t7 zEIj2Ek2uj&Z^t3xy>A}NwI>Uw`y`WDYGoa0S+hs)uns(rhZ%Dky|s6U{urx0Y)<<8 zHyPkhn8Y;xyHCn+!^lZH5DvQ4f2i1cBS0`IeBRhX@n8nLJ3ij{Unm{Ot)5ZJ|6b|(w^lV7!5pE=%@#gc0LlqwZB^KQm-VtpXxT!^z~I}Ix!lEJX2D$X5-zunk!@-6INdueFG?UBR^ZNM_3hL%8Ug` zCNZc&ehA#~aP?-LH%ZLr8%a+DWRn!g+_2c|1k^;5r@q;M?jtrIL@m2d9eR+v^KiU52O=V;E=|mxV>P2A&w(3#JJsh&M)#J=pBsVRPfEc(7d?)_Cs ztA13Svzj6o&!EdMTS3@)&cN?Os=>Hd`7rmcOWb$tx3TKf@hH)S8=lz9y+|%#;yR_M zrm6;V9g^MQM{%^*j$7`=Z;_Vc(Yc+aK~|Kxicu5x;-+@vu&Ibt4c09*d`&oj7V;)L z(an=+eZ}r1a706BP7UZgjWO_ZqRAML zj`DBy@pJk?0_gTS&w^gZi7C-|9gVE5ZOrK~roH|#B%Y35FTPv;vR>Lw@qN~NrW}c+ zICg6nM^Q@cz=d%M#?Em+BF56E(*N#c`hUDvP!IoW8xZ-?9$Y+t3NLs`;w_y-41qYN ziutb+G>M(30XYj<{|ppH3I9 z$DDc`YbjK&UUjg9q6VK2*6l+$D^FaXkCKYOG}f$rq|^9JbAb%IajQRvW^w%u2{DIM zfIsMwN@fGKVTdjAY$7qicipt8G^t`Q}L__-^MoXO>odlz?oW@{dTg4=F+(jfr;B6$ZKUNk^{$a$t6cD-nr zT}L>o2-hJLR#-sXP*3tEB6Ms-gN;DMGZNpNFqNCPdP0P{q57fd6w~W|SZtT8QS7Nl z{w&@2uH%uDFryfcFaYAP$F8I!NhFqYpx&F)tV_8_Z6FCpXAA< znTpM(r_FPuxW8<+?gRRnyEx{xMm&$a{S zsDdl3g)N#=-s}3}&sUbua=A+*BJ;z4q3|>_+BfD;=nDGOZP30t`CSujTdxQVE0T53 zrEiA=?oVUrhDx6webk{fZa+zqx_3J(RIlb+ z0PxOO%>sUbZSb_#qi^dXwhVh2TnE7TMj`p&8|WL^4k5)+%TM0?~6SGDw*S+BPWu??b! zc(2rANSpQMv3SV&Q2G-W6DJkYCg5HYjglt}o_(w)6PnwQ=OjoXb^4Mk$2S9#M8v?D&!@duZ=%dv|Bj&wyKNYHy#9 zFYDjWtQDv0Nvd~~>&S3&wHXfYmncG7NApQMjvN`IbwbAC~aSquuzb$+uV%~aO z*lOwr4$J`}O=?p@+SM4{Bc)hu1qk-JQ%YX69Y)$d`T8Sn=m=}#XDUAA!v_)aFGO(~K-_=SJO0Ym2)BQ1ZYw>a8j8WE7bAB&giFT$ zssXIF6<66k5wxo_B^`^00l50vm4CO6pLCepMkKOHHTC0N0-Q` z(`b^3BsvL)FShh^k>m`W$&c?JH?qyw(Tta8p-tovzEu)V9Eb-p^kz9BzC2AbwgUTf zBfxvO)XIDN{U1F&ErM0D>*08xp`oc)Z+r?slm50Bb2L{+)$C8W5b-p*vMbl4eac`Xnhmrg&( zTt@@jS6ZKE(u3>v$EQ%byGQvU&m^}M!|3xh5-eB;hw~$t-x2#pvMoqFlA;})dldHK z0D>FIGsv_N3mS1Nh!b}ZF^YqNH5E8y)sSez1e4tilE?&5ih^GY6|Se~S3g@95G6%2 zNkfM{Uxy05P0NpC84W8;v|K2Qlkt)&W0XC<g)o|WU7L@$iVnv+!2hHT*?_8z98&*8wnDQm z`^{Lht#-6M7N#+eW6hVo(9Yo7p_b#G9$+rG-PR|1A8tyvwzbt3 zUjR_MlaIJI&2rj5Tb68ndohyaBX=q*Hpgs_595ebn&bZ0;f(dcEW29RY`JPZm~hzc z$KEgzd$;WP&g-w|@i+G-Y;U~IMvz^#hCWi3DaRLleiFkreq{))_PabWWj{NOLHIno z;wxj0q~W!|s}?KPo=3D|dwv;!lRQlCgCCAOHm0SHT5M+js^AKUsup14kA~@mKkl z3XGS~-*z|7%z8qa^bP7+-8dtdDI2z3$O;61wy85tr76nvV4%$sr64FT%u41{gvaiO z6~YSBcXyR?y6_)Ym}-Bt*we3DDYw*aJlJIs`4hMObx{5dw}DpvIw;_MSSHia7gz-T z!x)KkttOa?(EwdDmTu6W@nt!AtOI=PuWbvH^sopX&E5B!pG{=r_&iZ>B^yfnQp92& zTZcW{=}Wv%=gX36ED($)mbWk_nKm7-K_huW-fq7)UavaY`|Rf#!$1t3U-Bu+S-y^6 z6Zo!e;n-NN%mT6E8Ltw(UbyqUw=r5vZhu8c-NEVOmLH=DxbD^bu;vV+wpUy>bB2Ky52(l2&nl$yQoy;L`P;t;p7_<0=0UPx>ywlkW_eW7+c{E=g83Sk;QyPp9e zA_>d7o30uuNzy3-F*8$EfM9ZHkf%6_8~hFNrpcNk{`^xN+8%(=veSe_pn@Nn!*;CT zep*O2qFAA;2qHd1pmKP-u?AK2iNL8A?tPfXvhzD6Aq6AN)_y|^i7cmS0hPiFUm#|w zt1gg`y2;3p5A5l6?DQxO@ z_$cVO2)FlJIWA!C9UajV(6toaCJ7vxjYSFFwIRx0NH3y5FxTP21tR1ZF3|8MmJK{- zj4j?KIr4?hzw%wF^;sz|uM^ohFY?o~!I$7LS-~4_o1q6jHV$IGl9#y5JIw})@xue5 zm?CZ>c!wSI2N}Ha?ddt_6+DV@(U|ZUu34baYM9RQHiyXVHD3?An2I|Tj~iEGw+fa0 z`;zLkQsOYQ5ELI6M^q7<>qw={umx~xx)>GMcUBzmCa@`eJ*s>=U=;K9G2QX|U!X~1 z#A(NJSIAH3xt8Z|TJ5uy_@@fY4nYmZ8ZqaeQnsF37wAfr%}nGy7ZjOTW?8GsS~MNI z?@#R_a~3*CTWo(O>H`T98GYgJ$o|L%Fn=s0AcSiiE3E7&7>7Aju))`l&>w(?g$~0ieoBon zxyBKce+MbpHA2xNtDl!m@bxTHnXdoP4^ z^ErZ23>(3ya2?DI=w+SzM1`HH4_SBXzcr#p^LY;-%i|G#^aCeE9|BCLU5$8hj*Zrp zIz^)zio1a9CMk!v$SOGYZ35!xEL3+fk9W-UT(A0jy{n@DK(G``1Y9Mf9( z5!`Ai@ux4->95p94xs&&y3`-;IHn`2|MNoJKSuiDVCFkMW_x<+KxVtbop0MwwX6S$(kcnCO`Id*5lPk=FhgScl^ zrAVV*DRQ)NeL7SUbk}e(%{cT{YqGqGkK_Bh7@IhT$i-UAsbZR{w>o1ls+5C>tgDwL zE?25NyRp$;U7hEA?TY=0i+NPx@GZ%SM9Vqy9IjGwQOr@+Q;2HpuSmqW@!rFBV-{9$ z_ttPwmgGgrBk!dpkM%)q5GI7kX4N;Y&tR*`c|@Y!E|Bl-~W4Q2QCPN(fQ)(1I#B)aoHf@`Tt?2987#W~czAH&oa z@la-F-1l^uQD0QjxzMiq9MYFUcmgarcgLznz^1r&S{Sur*el{?($<^!L- znhW-sON(9hz8VovitJ=&9A}duw+loBK@zj=Ag$A{EmvXcUlTFkoDO-Lz7APTVmB() zG(?&aClDg)3|IlyEYXg8^a?0sTX1?ve9J`n<{CkEx2!Zzc6W9;qX`ufDfGZ3&NK25 zzroEB!U>n$DsbF)MX+hPU6*oUm%uj(fiX-?yhKmb@C8~v-e!=0Qv8_8FH_1Lukd{N zK5SXA8^C1gf~|u@ABc7vFfSlR+8PIZNAa#l>W=l=)Hb?zw-UV%M39nYE zsvbM2PoCK8WuIPNLlxu77G+S`TKR!VAoGla)F7Q}XfG1^%ZnAZ(=Nrq-moYXoSMf^ z+g>M(dTT0-(LB20lE|{3RkGO1Z%cr3e)|4sn0vOA^plX__SYC{K76Oo3bU6qaN6@F zg7H^d1ZOkl&iwVTwrYkZaT-v|**up()cJx{^Zxmw6ypb)C8t^Ii)A5TZ0w5L#r^)p z7kraI5daj4qXink_$qci1o5CnL|yPntHe`WSr>-{Q_QJI#XS7TO>6B}@!BQj#rc61&7l0{T_KnZv1V+W zXoSRYfd3h>Q7r5&|NP?3!vDwGTfarUuWiFK3^R0h3P>X(JfZAt?w$N-CWa zQqm!fv~);!cc-+(E4ti!t-bH(K92V}o*(8fnCpAR`8h)g+_W`}ROv4cA@+yAL7tY3 zbe>qSo7;>EqXUGk9CW<8}5-fA@3v_j^>b{{{_IJc-XY?GF3fk>yB*@VY9pFAa05(~*(NjI=<0)}m08+0YV zJ}XkHJC3!V$zo1adp-`vj;raTn}bJU>t5l~{9TK7zv&(Rr{nFPP47>;^t)r@>dPHk zDtY(2g&D)&|F*GP6Mu(dQ1?rB9LI&Uj$iZG@MY{7CZyr#rYD;q#bSD`fomfCNg`7w zWiuwPIUld+#as_h{&>3Pn{8#3%7UqK%y`H}XBF!}q18;?89Ax!Bm0g z)Gcaqb*Vwf0}xhfGEeF3*NBIryITo%D?MR!R`c~$i;8{L44q(#r1Gtib15^gr`T&N z?Vfk3sc)(*XWBFGFkTdVKicT5@Y^{ptUEF8t>woh{8pc{wopL!^hIO*MLI|_74$TXm+zS#$s^pD@Kl%2-LQ$Lm=}Yt82nx zYe8|+okM3jn$9Mj%vnqlI!DHH)F46m44HX$hWbd0GWMFQZ89$$rQTTi&a-J}7>Ahg^ z1!-|z67Y2>*+HE>D5bi-Y_Gwb9=T80=Om=9X$avJMN2@innaH2?1%)Qx;`ooja~}z zFtcTnz_J4L3O%s}vkA&T@@m|=6$6`J#S-_9X4pJ07=10LuTs$YWzK^)%8of^o|X|2 zhEdFec`QgEOnNVge3_kek;okJtD<8vGjAqD6vWg=kPhn!0E7D3VziSU0gUV zY}gf+-$OC`!8PV6kSr(zzkHG6cKPLqL#TkM7#g^peq~i0Wo%9HY=U&X?-F-SNX7O8 z?Pf)a1xsS6mJ|23x$w8AQi?2y0%jt`Ui2dW`H>y{54O_ggOvaN$b4?uEV(CTR!YKu zePoKOO;N};IDH92ADg4mbu#WHy;6zD2UDJtw)c0pjTiTm3jLvUpteODKa5@}3zqq= zk1@g$PyiJef5!at?hfDp+4B&`=}$#5Gi)1QofOPeV-RhRPF0$V;uZOj0`K0Fxa8?-KvP}{`P%8}#lmgW8(*$9z2b;71jWELvvjTf z@UsE-YC0)W#d8WqYUQ7^6GnDk&zIFhv@bXJG?Q+G(= z!I|m~(ac@(iDOi`&Ys}OHO)pXlbK91PD0saVpgHgt~P01Kqhaeqm|z&Q%Hbps*I&L zRJZ87Y8CS~M~-Xxn_V*hq2noeW|1IP2a>^Masmrnwhi z4hWr-hva62O=so`h0v2WCkCj>M8;snYQWyasZJTZO;QAhDWt@qDySzU*t%R2Nl3)%Ci;+gHMzQv*2 z7dGc>Gu4-*et`XpbpBW2?)_&T?SBDA#3XDVYB-_47TmUr9eruP2a+_;@*MShe9rZs zdrp1M=})G%z#R=%GWbO&&6cj%kW57pV2<5V%2oPsmRk2=S~=@Q7IJ7tOpgj(Z!A(J zp>go(4z!%AEbvqvW;0POcms}LA)*=2wN+W4>#ky+ZPM{Z!G!ZJw|>3DWjM2X#XeRm z2Wmr&$A|O#ch)8?nRpS)q?)cX2On9ffPFZE+a9 zJa5n*v(wK1xSg|9&0NiXb+?v(LB{8QDyi++o3a*VSmW+GBRlIJx8SEZw+!)MpQV}f zZ?_mFeDiGyY3VKWW*K4)KwvbqV)#HEX=ClYNgc$YYn&=fLrJ7WLR43sVfdYgZ{Zau z#9{=E9$i8m@U!wI^a5EgrVJ(LPX^3IR1=3Bl4C{YSUM9O(Ztg)MfKcN;}VD1i)sK1 ztgbanfh>By-AgdjagrhP>bc{2vQ`qNG!el%bu~MiBJ+)p7?I-GvBa$H2{AzuR}tD_ zT$b?`a+vw?Ss&hwZDpp7^E2v12}AOvb5&_|a|_q2EelGo*v>2qD-d|CifVijp~dyL z;~}EKQ<({+ZS&JRWnJ`WYP#$FPa?|GnOXM4@BZa8h4A}E`ts`@`^89RJf!^Z$crYQ zCS21$EYuhEQjUw_k&H_DC+2o^ZPC9RIA3LTCnFfhSO*WL4is}F|nf5cPN- z=S3v6yVw1+PJR#)Bfi7uxDB2?O=B1(IZ35kU+5u}^;gaZMCL>5ei?yvl#koA2{6Rb*2~A`Vc^K|RHCE+qjZ5S8d2mv1s*geu9; zE45(DaXV;fCh~hEjYxg3RxA3+Aa9BnD%Oa)6&@l2uGzO7{2}NBjfd(*g2Bkph#5Z*CU(IPa>HzbhX&b zoqo^3t$mvIRzv{ee_Juwn1qmpA#4(6@XTM|xx%Wrx3-k=NvL+qNvES(v+7hJdpc5#Aj4qA#=Pm@0vbu_x#NI52ShFwEjtJHg1-`2yFM+t8Ldw(1Y9k352D#!M@~Sg>xLrTSi1Q(0B4MN zeQ^q$0UuL?ZdL}oaNA9Pp0BGaD`=+IG)ymq(d`EWbr4LMvkMy$SHu*3@3vdni|3Yl zl!>Hqq;QW@rngH_IGvevzH(jSx>v)mBuL8lPh-?25eh!34R~dZRMr$T?^PA@2QZ49 zVnEd^dBl5*Di%_;U^Uw`?u!~*ifwF$##O(?U_@a$!($enhYkr@g(qSh&C z&GH=CG%fRRv7lu&NP9Jh9f}&_tevQQ2(P-*D$fwlHdOnWlnA2n&Vv+!O|M2;VsOD~<>{t+rfdf*~<}aPevC(o`R0 zMk4=F!c?H{#JDMdF4lspGeh+qi(t zOHrF7G@z(05^vp}wQB+?1t2oU=aDi8=;(3kx+Nd>n&5y>Xmq+|n`sWpIukmzJ`B;b zOmN(wg~)hoBr4&gEH!4`qW`f$D2yh66&(=))&(at-Sut6p)Bxm9-+>@E(@U4j4+vY z97{CJ5%Y3J^u{(wda*4+(5@1xVnH%dZ!J*_k~R+2TP4i!5{-*DSaq`4{icP#%vvT5 zqq7e)>W1fZ5)OURLNW*;`61OM630r6B%DPOb{F0@=IgMWnrZr+MFC{EC#@Og04JEqDq5 zuov}xsgao7&P$hj6GQ{UiHSM%TbdE(ek9gbY=zO1bb{l;cuK{ZsgROG&nxz3o_3=m zBJy{`y>H2p0}No!9VxO}jJUOWjF^6>Tp-=VrRc&?A4 zf9y21A;dPAT3=zgHJhR|rbWQPS*I{AxW0SMNvNec&_YiMN6xP#dD(d5G9cK3h$H2T z?-HOZ+CvBQ*UnJcG*Cg)^;dRy5VwhM6xPxn9sGz_MJJS_gK;H{d)!SWgm*o6<=wNh z&6W312otUBle&u;i9K`;ykJpIrdU>){H$47_P99X7*&-|*}5=8rr{W(O9n+%bd^io z=oc@Xvg4_;nXsd^B%BfwL4DHrDY(wv>yrA4{%Vn&8pedF>J;unNgkA^y+KrwtLEwv z+p73kF>(Bnslkq;W@2_zih*QlUNHpO?+CBx^9*i=)x?5KnTHL$>+gY?c^t)nkS@JmEAU%mB~8l zAl=zLkzh|K-7NmOD+Yea9U{u8w5_X1A2zHO)0`s^`C-Ad&=`kM#8kbCo+mW+qV8zP z?S1(D#@KxDdt>xL(|w{R`LL))Mr+8mx6$sA2d|{?P3CDS0&p7l-cv3rr8`F~r$3Nk z|Fxt=FcZBcZvS)mw&TJ(rP=i6t5xNQvVlpud1B;>r;K`h%Wt^ zwkz^aq(V6RFYy|^cnt5oVClQ@KpL+D0!0su@t%d;$emx2C*$$Ll`H)IQw>lmvE)f` zX6Y~1{Ti}{;!-$OE<2R3BT)WGE6aA_Dn_=txUX8l>Z&6FyB}ESENEX zZ!nw+K@N>Oob+^iq!cW|^CbD_r`9M)lg&AA%`^8^j|U&g z$l0Jz9XH;%-_3lxSzw>Ly}r3S-U#(PT3Rl{Q=M!OXkSLSw&+f$4uM=8j)80knr=k4 z5HEef5Mnrt3PNIxt?Rk6bx!1Q;}Ab63NVZWLa51!MSK<7HmKp4>!%dn4`rSZzy~iu zts%sO$P$6nCrUa&SQ1@&qNviJz@d<{Nke3+;Tkn>l(`@nHq~tu7+IfDT^Axt$q0KT zGHsISjmL!;%TB5Qf zoB39oFhSh?7JZTi+2d-=ELr#}+5n>PgoF;lMc*^waz2Ty5Q2bBto`;@8ySJ{HpYTU ztJj7P7K~Y|8o{pdM0H9QSwo5n9Q$cqoYn*Ha#>c0hBinXxe-QUTx{u(Wi46X5?q)8 z3U4bAyTDM^vKhDvV3FvzNiHhZc!4-qwh(7!t2k_GydAbtwNtF#jErqpkM(zNjC5m(QvRdV4F46geGhvA-ZCORwKo5p-Ef~$(YN_Nz3Dz?yYDh;waHJ&PgUF9 zk?KXJK=iGg%&?K{r@J^!r>BP|@)>fniv2x2S+G1pit%UD%dQwAR&>Z-nYS z49(eW^DzM>DkxpHUp^MDC^o7lkpE#LHlD~z)@`$xVja%l#qQrw#cLCMPJj6AO?nV$ zX;9vlm6G7|>UPw_c}>GwAke4pz1a4HyP2}1U2HoA=$jovxAnjPJcjq@7v_8DnCZO7 zx_7{%6Y%ELn{(llKO&ZoRSQ6#{_{kCgd~%BMew@GfOi~taR8J{G1M$DqGE2S=FuOQ zAy&(ipn+HNfiW|HnbLIz+GJlg_3oUb*&j;xjWJ95;~3#M`+;(c7)Xo*u4pAHffB8( zY^t0MwU*AyehEe)hky!9F4#exDG7NR0!)1~M{7WJF-%=Vb={_GjKsihu1Q{SRE@{_ijPPh8{AJ%WBi zQIGVmr6BSb>V1nQjr*}J>i1Hx()IK2*NkBtniq>@H`$)89n^i{D}=R<&U~0DocS!8 zxsoY@rZW;2E0gk?ySYTqyV%zMik*3^gewJ7V^$EJ%;mCwgNqhvHBqd;(yHyz*%yk( ze!To9x^nYt^_>?mFjTqNARKqB(cQ25XpYFni_>P)g=6t+0)MgK5q-(VH#2Q~Q}bFU zSVp1}UKZ2fSRxVqGXa;C4g-@Bd!PLIqQjTT^=xvZw=3)78+Zr`WRED)3l?X+++~Tn zuHup>>1}7PoHsb{KTOkz36Z^_&c7I2A2BLdkrU5y5n5dgf1 z-FXJXVL1{eu^KKK*j|dM34^fy*x>B=dRBM3Bx}gjdNM&C%EkxGrE0ogF{0}+p~x@wA>-m7220sPkub3EYjSBWQnOhD=#DG~tPBlD0eNLxpUR_fQ4F*aOKAa?v6;jAQx} z{(4Qga{10X5<#j(xvwy*1|L^b!vBp{@8 za~7!NqMDt2reQuJ(p1Aq9swOknME-R%ZrRE;AFzKlKU(%<*96Gwkkz6_t?M%_vSdz zO*j^~{?^NsHug6TvX?jBCJ?E6_VV3R=F?bbyPkvvD0{+={p7p*t;jS{rj0fUfxl%D zDAv^p%TQ&m)QZCkCemcOYM@E|xfwl;RIcX0nPE9^lK>Ks&b^&dEqMJU*P^Hfd-?Z{ zG$a#Bts~N|dXJjmJO~cbjNew}HF3=q5iUCEWuVpa5 zFy8Fd>y4o^>hU+1T@+8Iya5lA9A^%s;@p9;nVsm{RdF;hPqIIMRPtfEBj9=ZIY%;_ z!udFhIbsu5jQ+v59pLphnri#=v%*2$13D-gHN04Y2gRomAwEW`RUO?nHY_@P|Q zMf6DMzhJk>3h<{KW2uV0k|rB4QH9K5Ml=G1lAH34e)qSE*Ee46ebYtyt*`?+%6_*B zpHX40opHH{#}beb>Gxi+HXn&}pxE@tT+%r-3v9=!p`DH*x+M9n?u_NsaeY5A1!?Cn zB~p0BwFEKH#dDFOy{GT1z}q6nZbOiS*X7rDXM#H)6N33twa)>PAMmz$&HzS4Hrd+7 zKm<1|{qN2ylZL^Zt8ko9X>2@C@f(IzmChT(Haw(fiv#SCC(Q}?9N}Ugg9uRD@RWuV zPkRLgGz|hY0sHj&fWfV-3Bl_65ikh-kwQC5(+ORq+7U&?oC~(FYf`vC%cy=B7bZ7B zoUcY@tm-F=3KeNvcb-*fX%TK9eJ&U>%eCzg%ShviD~OYW@RpLj(e_Ya$H31|t7&=3l05 z8Gr2kx9pmmd>c)F8$rZ1@y_Ev8C`GA5V}8m@eTTc=SlxIf}q%9&u{qzYNkm+j-#6V ztTmFaW90%D%zEiZ8e^qQHRsLO`Oc)bntdP%K1UCX39{F(zztn)-#JK|tn>Q0OS@xX zr|UPv4cM&JR+b0c4qi~oYZ}M-@}OAhVuaael$PRQS))&vt?0R*$IbUA(p+#Wl?`87LLul0G#`6U6au9m`+(gSwunbZ*UZ2&nGDFbtXl z9)Clax=HAbqpKX_TI)`<8g11b+XXwH1R-O?pU}tXbnzgs#Y+qOBx%{w0lM^quisLL zyl&V?eW%}XsT2r81aC;Xcjs*=pW{5qNwYTPn{l<;xA|x-*&0oX@#!^UOF*>0cFH-owtcztElpWH z0h3kX3rFC3sL-|yJzHfZ#dR?hE3F!gBhR5Au6%8)vIQtkxR1khLoD(TN+F0NR5&aQ z_-0kRP4jktk1nZLxIwWEC`rC}0>mPBknOe%KYe@ukbOFCXMaL6M5!2ttJB~QM1Yv3 zQ6NAXPad9?tzS(p6=URUz(MjtZ{R>A*qv~to~d(;NKtUfxDdh=x8@?e6E$#i})`@_hq^ zS+*$xf|o)+=Om0z2M&(5u|iwYA}V%Xd<1&Lx9%NK*}ghiQuGEwvbfDM$>-BVEq*)O z+BT2Djjg=#;z0}9xZlLJ#~ z`b-pvTQiH7!+X8WaJw&;0c4PKgy6YljBg&oC))u+b3B;5w*V^StEEu%3tyt?WbG7R zQ?|Lv7c|x82qWnF@YAh507=_4+TAnKCe z=8KznTn4V|7M$JI5vH81Kt^q|z;V*-EpCpeShH`hB7qFXW z6D~s%Vjh!Ocw|7!2;pc2g;7Nea2BB)B*-bqod*g@SqGjLe5J$E;C|3Q$e@TXOvp8e zrWeMAN_w%{zSzM+W=ZS&3_l1j$Q3{n@j?p3q)j#p3(>1kh}XJao_(ztjqEf)ggu!6 zsreMpGpDHjkY$yE9T)$*{2sx10x3ARG+DZ?6>MtX!dkozQ5)Q{DO3t!uYig>TfkEk zCKJw&18U8#EX0%wnz0iG@BD|rk6H0v`H;=)2l7c3DrSks`EYIEHJekR;SjF@fVDA} zYwNnGTIzMYGNi1+>%lmYW3(f&3JI{nET1gGx#Iww3S%oi^QpK><82a=k0-Qmnee;w z=|9(Kgc!<@GnHpmD_Rr{KRg>nh1YzNn{q>kx^0+A`B)IFqJXfL zNxgHrD9O@4Ux2PabD~GJ2yv0h&Zan zX*w1pNgrlu zR$-Z5=%$N{LvT4;IV5kcOtngL?sUJawKv})^1Oa?Y{W<^Vg9VHny-Zz>bas72?zZ9 zeFyk+n$GO za2kqVASNM)s}BWy9mtR$km;+ z_eYZ@>-LjotPd6$^}5tg6*Tpfzv^t{AKPBc3vfpBCJ)%39MpNs(j>=wdAyMo&Rc@| zw=m4ZT4Xn{VSzZK{W=n6{t93ZvXy%fs5@b%i+5lZAcxV>X{r=@};hK%Zo z9rdd%`r^(RzZAw=?p*Xoznp;iXj267a_BJz z0GXp1QPe2BdD4$MjNKPek;Z4yd?kW;zw2Qb?|d#T2i(n#HpuW zn((;>E)_2Vb(-K~=1XcD-C?3Nz!|N=Ln-)$C_P7ay#_4a0vA~-!UsST0M&;)Hu7@E zO;AhbK^V>o3q}scP7N47e=q2R7~Sj#MNqQ9)|n@=@bJCu%HeW?95V%-%|;_97ea6m zsUD@;J^dIbgJ@_-wO<+%6Zv_&jYZqoot7f2K`|)C+n?~u0^E!=rttGN4%(i-m!QC2 z<>c%0!pbR^og#a7eEeeQQrwTd+RZB7eSj_Ts)+DGPC;^rFnW3h^aK;e-gtEu3adPZ z+{YayuNsA+KoI{~gUtV~L7M+l4O*V?{XG5j{#Jv$*V%v7;G=Gb=7xXNpvu+(eQOdZ z;x+lN8pP(G)?zNDm%XpS?2Fzs_7Z)tu1;6kF2F+?K&NO$xHVpFUpD_iYcMSkoixMz za6;u#n+F^K20^S3HJcppSOT|M(2&uoXpFid=<-^7ApH5PeIyCfVsUe9KR2EYV{~e& z%+yEiIadD41GtpW+^RNQvS00}PEoxY?xHHyhhgxG3VntQIlkyfK!CbEWM%99|%nT)+pFMsmP(<-8`Weh#timW@63`Gw=XWaY6-x z?IFnUjM`7>wg4{WI7qHLL4X=bE}@7oDiBwSV>|X7%b@w9dNF9ET@_~Mb%VQLtT8yD z?x21WFe@m{jkIVQXdt!xu8u%IMQAc00xZ=CRWPtTX;@zjl%CyuU?9EOwGxx9qe2KV zfJx}FxAwb!cqauUUJf@S=%S7_- z^c{*u9{q_vk+F%+*!W1&&Z-FTtXO2x)i{W)O06PQ7WQFFsmT9{*HTG2v2~UCFe+@57vPkFjFSX5T9c&!RP$Cv0_ ze|f1D#y<4)F6zYg(Hfb{1pb_2Gtl}$3-3&|f1$Ima_~*XoSfYohY9fqck)TCo;z}p zc$lR&%`K~tr#GV(quEsu;KbK~Ik6*Wl zE8pzCj)cF|yaDiI<*y@;PpO!bd@p`1-ZbJ_wOCt@Im9YkItO(IHG3iM0ZN`<&A&nN zx-vA4ryOF%C-*{2?*EE;RQ+xt^>`#}qloKj({7*{_?CZLo0>Ab(CRN^#Ie%}UHuAj7*?W|CEN(Bm=M`xtBfKXwy{u(>K$x!n zkb01x@Cw<%_P8+WdWm4hO1dald~-F=62ZA>7l_BaHX7&sYqBs*tS;700=BljpmTiy z9(J^7%+^cutyk2e_)5=7p083ixk(Zv3<794^x48@`Y((>i&lioX%{n(ttOab(Ft^i{y4~O;e`QORUUG1+-);ggi{=wj;q%s=$OBDUiiWBt zZ!DKkv_ZpTExe|A4bvy^*M& z{?**5W$;TK9{$3sl{b&lIwMIG2rFCjRd5#Y_Rfz;iVs;50E_Siz z8~5M-$W|-T^OFSQ+he=#-=j~9&bifP-AxhTi(^-!Vr?XebxMm;w~t1hkrSAw?(chM zks#!Q;abfI?E~SW4=eaTZ?Mw8CF6!$YV1c&s9vISvuIaorX3^^yB%&%H0TB5fcONS zRXJ2vCpGiv!>gA-BLkXgO{u)@vug-zZ1H5boSl;u-3$Jf&0U~}2%KMSj^EMfZF_`E zMtRb(cBVjIvyD&YFg}b8|Ljiv`KL7$4fI|Nyz1;lwa~5w~`oB+bV^KYj7lE?*pMi8~ z_uGqcG{_O!pYk^^#^)!e<%`a2(h`pIkb%L`I1UisEdH~tD z@+5buP|NeLsu*&02V(kgc%CW+viiuE2&0Fmv0xA}c8#!cGnXriMT@~>WsDmSJB>;d zBm@}>Yv=D!=q*nm>zAj4pK!SJ0Y21iXZs53A$PhL{3Q(yYe;wXU(VHjoa3u}deK0t zp6;eQ*GuB98ku#m0Vs1_GnQ0-9+N874G2w;clA+!w-J91Ka6Gu$T$8;qClZlSKp%YXD&d$j6}`R}pr_l_uAUU- z&S@MgqFEK^FLahpC`Z-e$+Rh30ZtGmB<+d?BkP;Snctsj)>P*vql&e>2zyO1B{F}@hKGJ(MTd{@TN1QPJbR_u;n#!Y(PTT{#6ZV(RS-9`%&g=7_E|*_l z31cLjP3t+YKN|0tcu_N~7RtK!eH8rh)F9ppg8>__dK8~W*t}cjbTR@%{%p#84ymjshH_I+szEBCFOeba7`TS0PIKd0ac4W?%NcV|K5SiED#Dk#}oP`j{;-PF# z3FE5O1hZoI?s&WX+8_<33x$!icrS&GxCj1qTd6{8ucac&*}g!}XB`LFo359}wqqaj z|2RMtbna!fsTm5Su;jF5@J$CE)I+`N9a>2kJ{-29(bt#05-pYgu=BmV>8O=txJm!b zS&rM1V+U*9jj@|7AdyP*6%9oM_!ygBdrlW3C1-fUST?HT408IHjU_YY9oBG5H z3D#e=koDQW7pnRA`t%3V(kd;nEYIPqTXXTVUDr^(*;ep|M+;=C8k+cq!?6iYbqN^^ zjEOPuqK79$H=c_$;S^vp)Ci-LX!(3WGkh6JuJ$gGm23u@L4xkMnk|YWL`-+dRaP<3}>FarrcEiEmY~- ziL~-YQ>GO7EX?gFUg?u~CbLZt_m*-HJ5n9jN=_JAfNZwvwF&{#R66cmjk|+y(F~tq za9qj^=dSQ$+nFk!ql~_i1P57!Y%P_%IUl0$%(ltVX8J**Xt-=M90sbubL&0l-fGtQ zL^!8Vj+?{&)Mw*|#bZBCvF3@aF(@fh?2ol`Y)=~PU+>6o5T#rAWj(9mu${C)^t@Ty zwlP|wH<_tQ;y7XD^8{ZXd%P)c9vpm_8ucvY#K|cXLAU-A#B&t-6!sO1;YFf>d85s2 z%#}06-K$GH`3++ApxR|$61LflII?2yvj=e0%f}=tc!F!bWOfNhflBe>5Nf4ZoQfg8 zXF%Rmm&!L!(u~Y05K{pk`|&?ODf2fKWV2YsGQyt*xQ)F>Zjf4mid=QAL^9H#7(~kO z{+c^XYca~I(k1}#zl8mNy>^xT=il^qo_5fEayH@&)3ch0CHCJ!#Hb8bUqaUD|IcY_qaE4SUuQn=g;Y>9*j#s_&`+dnm>=wthtEz*Ul=B|k2EG940BV&kMT&&D z#raNz>8q4kb=v;CS7_vueV5HRKp&)Kf^;6o)rKa@gA{GWR^r(zQ9dtlm5q)qUcvnAgQE>sn4I_7rANlCfmh{JBqy;tm-8^FW9g#A470jBw}!|hz|i4kSF$SKsmVSc*t3{)M_jV!P|LuXXkEc3?9=OMVv9Av96|$ zVlyI0jN2YdcbNh?E7oK^W+4J8kzqo(PJw$Y%^6asXN|ywq#TUCWulqx!#fH45GJ9i z`%y?Ge=_EoZkwTcgv`^3Ot4J|PHO5qv3%+5oDjU+{3;n-Yd#>!Rcb-SxTTdkfIMcW zsQzp^rZ5z;V^#7`HZCUMJAmtdVdIu9w|W!)c^kmv^!VVCW&_2nZ#onY;HFf9qz)l{Vi|<*LLr!a`P*K{M&}=0F+od z729~hAIYce^F?|S7-45BK0h~D^95Tich!M{kD+&1+SMm;1k9=j9HqC-QTy{n`>_J4 zd~5#ZWd&*n`qtM6fqSYCoDweIPTAv^eoFD3`FQ>S4i=P5_aDOGATbBcw)(qZa7U)S zmV6dO;I`6EM`@bbFamd}o~6O|n+|px^F&NiHxeZ+HiH8Z_FUm%$6#}#06J20DlMZ1 zrIvS)Cxb>FNY54{eW2rFnXihE&Sph#uN|>mFrQElf+X}M!LceL;8gg@)F&;G91`d{ zsOfq*%SARzddyQal~}=A&TT8(8lB_gMykl?txr&MX6_k5e2x|Xh+uYQ?IXXzMTBK~ zUf`f{dQ#w)5^cuhtR)6d$0ooi?`ghLDlQ+sl~_h2YC*)uwH`W)!eh;XBH291gVOn% z5yrCD-$Hi1d)Rh%%LnhnsksDQlAz<7vNpPZe`kdLtfeIY%l|nQN@{4S`xT<0Ejh8n(s+@c2c6C${AdN~9byeq?P5=8fyN!eH_a!ijbL&rKHJ4+U%iAm z!n14ZggW;fb#~i@eR%D6|Dk>)Np^zof~Z|z9u81`;oDd^PLb48spc@y24Un@r`RE6 z2TmF{$MWeogJjxg1u%$RWi5c4t{<%LsA<#T1Trd%Ku8wj;A>1uZW?F0=WDtEVi&ca zW#n#79t6KUGxX==J)ITAA~%5J5DDswIimNOkH7y8p29bkc&6VB0$v-!L6YRnZH_jq zh?;PGnM-^NxogAeBy>*_1y9M&;12-zrxcI?QVI6d z|Eatmns4zS{{z6`l%@Ykn!O@BEJ-g(09h^nDuu8nuU|~to!QzzDP$K0=sx148YQ~%AYapF$^}a%p+a4V-Cg`BA39q4*Vz;tr?rV_z zJP47Im_z^F4^(5a1dgTZ;Kx>R3E&+-V}vMZB^lXH)i(gFNx6Hl&{y+{uQ&B(Y~HBz zo!xs|v!mlhy(TlX2W*EG^MlFXdB+y&Y; z)LVP_PnrtxO%@H2_UU9ks|w4`klX&5EO$+v_W>aI#t7XB6{~T06;N$;&Q0ZaZY5aM_ ze$mG0*xc}&rZxO9A?{am!CUW6`u`cv@b5GO`TQZ8ZhxDJp8mWt+qRll-?!(`PS*k6 zcq^!s`f7{%UqLt1D_;l?I;+7mPbI9exr}PzkCEv6rWTicqS<%a@0-go`OdADj(WvT z6Zk;EVAx-pAjkz=t}D;)xIAIGFwfk*fv--(2%=+-a91kyJBs>(ErZNC$y>9zRxt+%v5$g3 z8LGN|NPSsQoF2m52*7a;2+4Fn$8ghgr@v86d1v78b9jNCP3eM;1S_-x7(_Z(b)l40 z*beQS4xyqF%UvYCT7lxj_@5HjjqrF{10IH0Y&#GCHHzwAz4WA`TqX8#J5v9o9&wpD z3w4NMVoZ7)LckbYD$l%`m~N7!mDNH~VV)`Om^9&;rI~19mnO`-t&;9Ix}A@bPl3v! zS@P|7MeP*VpIN5pHS2D(yspIb(D*0j0s+wduR^2y9)#>|@Ne3K#^8hBt4-qZ0wDg* zu*a!80w^2}dV{e!)*COB_UOK1oEZBOIgy6@f?mP9^_o0`o3Uv^1A$Gu{H(t!1}ic| z=k=gzx2^^UyagUm1mVB;Xf>-JqNvnRek zdcG@=P_$M>EOugfPazI)>tq&t1}$jfB!=pvj1ZzNVq=RX)SuT@*>o7{t-mmjwl*ke zz_x;O5La~+8{4r|3x1zh_m;LGn@x5@8%xE>utWyZhz7Rxehs}dBXOr!-Q?2a8oiK3 zg`ks^8gFjM+Mhql>fBina-?_i8RU=G73oAVe(wz5T9qXvfdg9g%8wY9{D}kNCPct- zal*v&7KeC#$ETu8L9{YX6R+0k_y$=a4I82C*cUNCPtG{=m2gyHh?2XsU{)l%rj&gu zGR!iOC1g#BJ*n#gz0+E+_(Z5$fXms|O`Tkft%-Zk9jt0r>6p6U-o?;TnDRYVuR9F<) zIDUDv!zNnI?a*8=*(Y$o-fieSsQ7H#ZS)HO>qHiWIr($BMtT&fr%p%@wrNiPY<>uo zAl_{jz~-^0OvaY=z#>3;#!&*KnbCrSG}peAbYL>jnsr7Jx}o7Akt;xfr1G9{Xggs= zvU%nb>i3{(WsU>ZM+ln7_0u#b-vh9)V(L;u<&4j4J`g7!bWV?QT{SJ`ZXz!1}hr==|&8fR{rM?2RpReMOV?4jo<3t(e0+O zu~?NCCEK#>ryWlZZ8Z~lNA`5s@$c^P(;r-a$8O??I6;uM6FiwTE4ew#ecTlBVe!>_ zfurRW670V#iKzhn|G`K5+lNVcO;?uD{I8oMA_oRDo_)+ohF4cj@^fmv?cMZ zC~29@Ryg4C{Jc;h4Z1htgaz$FcDf%1e~28AmDq!tiezhcn?Ip^-`DAzz}Ihyh}UG5 zBU~==c=K}C`N;SCd?$Uf1m?47M}sU>$@w}HUpuP-_TvtG!KYkH#m)r#*K2_#p@dp< z=KES-+!RNu&1iVtWG_O;wUQ5PZ0etm^hCcBxE}ctkPNW(x@llqnW=M+n`9!ApF5S` z$t|vGoVs4FzZv8%+xR5=V{xPC9OmSQwlrr(C7De|m_#w)I;}y~O!L+f1dFBC6E8mgE!BGC zgOg)7T}lTqLh+xi764%QBdq$vFarT_a{n+4|H&{Tx}(5H`Ex~1;eh^Z{`A84fcHuK z(Y$aP@2}y^C>%xZOsZ`8YdB*plf$|Hw)y+1U|pfIU^9;Sm@?2*zCA0LseU=-f`9pb zYw8QW^z@gF$9sS#dk|XHvN_!-5|R@me~LxSH&SDLZ~yq-+uCGP#mF=y{~ zoCH8XVQ`-qx&U$;xGs5b*X6gaTOj?*Wd;sTBH0}X1AARm&zL zooWV&1&x2pDWQQMo=I+-clTk|s$;Rd1-+~!p%a?02rl0TKgsNbBymTLptjH;fFU&p z@oTsBG~r9PwGa9jS<4#p+5CW)2qsUO;d|(%5rH5CpNHuQnr-IdaA63og?egyqM~!G z8jj-mvn&k`X?%07oYHT1w8L&%!;1ARpP;y`{TPfA`)#KkHw;&$Gr{W6m|^7%D9QV%u+h6I(3cJ_u9nx7T0^ z->G(frv3PaAJSD$vT?&5M{hZmWI2au(`}vYX)Juck38hxV;tou8gP8a&0BT&1~C>l zj7Zs%)8##ky~*(bBrq^&9BP=@$bNJ2-DQm1BsK^Y3iUe~H(I&oo@8LzjAuVJO+W@| zB-^^7sVd zVD*^r2-v7sVR}G0$*vsA!AC&qncM?_+J6tnF}M&Y#!r%8nq82>mJ zoz*f8n+wt}Fr{^}`z0TJagp5sb;a?4o8@2SwGkN(> z7=NsvMl93dJnJHf>De{k%44sW|F*sH?BcTw@7>MCcJ1Bm_4(o59RM>Qz@8KeL^1*( z6G#GaJYmSB-+~zHjU4pLn*J~J+u(`}Rn>Cx9^u~Cbl>u3fa=xO6f){8 z^0k)8@=Z+5_cdF3d_?R=CF??mr$uv4%=B_klNTaJ>2MRv;%j{j9Q}8mbOvqyH|}<_ zZQm7wUHZP~*xr+=7pjlT%zkKm-TuTbL8&I&RI~f8+~h5Nj@PR%pDENN1b>_!X)W|3 zY9O;@wQBSjgs|jUroUX<(8YhHC*O6nzBgMtmsBxH-w@iMn-fEW5$X5>?QzX5 z@jSuSjARjPT9yUch_O2bp(sn~Nns?pk}I;**oF#tfnyn=QEW>*dJc7MrV1RrNKCZQ z-k}T{dy6)n^rT_G$raTgTI!9IK7bQ#(%|LUd^8MC$~0jZfG-)Np*{E_O~<>KEJO)_ zNSoRV&P2^LQx_-8l1mH4hl<=CDswb+94cBbcTz&WGmz#oqfZ)% z8k{jH=Ci7izSK{VXGjDR$+;=u*|$URpfT;2FD+=o?y?vp>~Nz(VZP{@qP}3gD{J+S zf~6bMDsB9PP%Z>g%m#$w4l(5q1EOJyIl$sy+21$TFxEEK#A z7V-2Nov+4LR_bFVx$>bV@DJTSGtcF|q9%7-VQ)EfG4G~1^u|U>>q;Qed~vkIk$8AC zK%ik;uTrC1bu{>F(R#BQVUhVej3&euFK5HK>iY;g`~o_<{)X64lj9D>@qK-!Bty>m zTC6r0v3_u+X~uZ#kqsdnK^|1Y-fCeX-j+p3t#U@O+}W~rGFH=49nn@iyOtsJZ1vVO5E5jRz0DB=75|7s-sx4-FC;P-aXesmx})e{sp1V^ zWxoxC@_Z_`hyyOQt(TH*+*}m8ulxJxQGb5(h`t@JY}v~DvT4fx#G zdCd0A@u^V!kIP`WTKlmtWDkTUj(t2F?-9;!2^TvvQ2JxN>9?yH0q_9$>^I;(!X)8? zqW=$4DBB%}g5Ujyegkem!SBG8Sc>`UP);FE$^Sh$pDXNbkVM?A?oZ%`wR{ze%TnC@ zbPwG2Y_E9qL=Yyk<$d4$a9BqSn}n}7ts1w|e0?ORqYf6*WQRk}Ul*D$*dB;z5tpUm zppSBh#G?PM(MBH;XSRm*TVVa>D4zqT(dYaq>DRXLhLm#JYeJEZXiGM|XSq}TD8jjH za%5Y!;VtZ2l;v6tK|x^wGnDeau)WB0W=a3CBd0HGV{gAc=ese|U-7}=cBJU)@Cbc0 zQ*%cVc@9X{k?g6AuumNU%qVL-(o$F!&_I`A^!3wFHA=ezXuGGVJ)$FA2`X|MnDUXP zFogz##Ti#at(#5MHX#M7=`7W;iidzp(ZS79w0#B+Bt zQx%mJRZx@`+^U&+cQU_U*)duTll3O=Q3+o&Hrl8)~$Ifv#jyM69P-6 zXY({-&}abla3T7%c3$Q$iMA;>S#u&6@*(>y{l*^b}p^|1;$d4M*ZS+`fB@rCRHodDT`0sRf3mFa=#QDhG19EEhPKjj;fND}q^%#-4IC z8baoAUFoVZPwtTRL`3Qf92teYu2)zCqd=OK0js{B|J>K&W6$#u=1E=#f_*p)^C1T{ zA_}4;bdx^R(kX8KyYm2t$UEntYM$yx>ZIsF=QNM498a%Wgh&8lkuq@%?U+Xd+(h!-a9OI z*GX^$Cx!<3+8Iji;CzY|zYe{J9Qb73{SiE8U?+j1TXfr+mmjFG`v3n2_Whm4*fHq+82tfyR!*E@2q& zR)Tf7-(l5!o4j5M=iEDx(U|>|b`3U;rv*^!2DQFbG;er4pKVS*>ikP8&1fUr`UJJu zcM}IkUrf3~iA2}_)!T?h1TPe)?C{$}YJR^V%`~Uvw2@ZMO2S>Z@tUHtyi5CBL5>GY6{y9y|a$-^_RQ+R` zSY&?GMyvj=T;=mZm6xGea4iaH#trl&2m!nGsbF<` zi$i?m{s^HwF}rZmxoyNV=atA1kDekKln>s=18-EHh1^!%z;ICzOcd@OC@ewufzvFx zW(e)Tpi-1p-T>|HTlFh1&QgW=Z&gx0$z~qC1_5@&qz-3vJBgmioSumx;1UUP3*3Gl zfDa0vVIZs-;)vvr9-finM3M7$8E&Jb{>B~0kx@IwHj0r(5-hG-Bp$iwMlJUK@!JAm z(__c=SkZPjO=E_p@FjRE&{N$*VcI>A(2vln#Eh0gi#&{w)om1!XxndxBl zl2%x$l_x>!e4q`0PryO7!D_iBrJ0tLs0j`Rty-1lhT6N$CPm0DU`UK`O|`+5^1M=Od{m86_43a7qqu0@6S^wQ9w|`>c=2(c_sc>5Uumc zY#(MNN8c2kJ7b5cLK6{)+A`-cD#SGZjY?_(W|EK?!(C3}_&sq%4P7nubNoD;wjm=3 z2@UF7*R}{00|}jmDiib)$nkJUF|0wQ0EK3)^E8REd>;@eUxPppLvZlEjs>o@l{X>- z7%Ih3ZwfVG@zd?gZf>SaI#KfRR;ITC_TO`wH3d{nLk=&wro;F3v%MLO34*g^f=xU9 z$Gl&Uv7X$--{r{8x};ND5e8S-Bf zL*JQU!vc5oIlf2ajw?mNZTAi2-$iuSU#t@+za}LT2GpmKpn&^e{Rg+8SlZAUSGiBi zD%>i>Vpd~{h{FH^E@;Qr0T>aZ_pzc~Q)#0kF%G1l;<5;#2-iFg_-y@NN_fZCm4mGtCel5Oc= zkz)M=<49p7kY;bG(jcQ1lD-CLc1cE@Cs`V@rvmgz`;KLyqf{NUL`efIX=RwiY)(?cdKI{?L_S8{zEbauNslvz zWVe)Di9>-q^3ZqWutx<<@2!bI0F=H`u9jiBBGd22i?-@$G~uaxVJ=2pPs>WWy_$m- zlyY7oq^aeG(~9*cYicT~2p;Z+JameCp7(xaLkOP-9$31ks7NON9CMzfP!pVnxtJo@ zaH1F~uBu{FSuVg{!DLA1>N&8wQzcvY@g{0g#iPpdiGnx1EOv@6vYA!xyV9QWYSmIN zYAa;oXqA*Lvpm65f>LhCq{Mh*teD!AT=Es83)f>68NUkEGScJ>Jkwy!YRSJ3k|^)% z^3%W7WfUOvU)k#9g~J~@N!tGB*a28DFQt;<-`Y(61iHQe3a2eytydO(K8V*haN00 zf2)#$c1-+cFq3aG>MzB=y+0DS9gIBK!CMFG+e>>MbI$X@K;44Hz#n&0Xk)NI z!2ja2$DyI*7p^MXYo6o76`jv_Uuz!me(Xf_Lu|-pta6Xo=!(cw5c2R{$87p6V=Or5 zW;hJx0a&dsMyQAsPL<`WBQNo4!IdswQF2I<@KABw*9Wk^V`>?mKPnsJnh91CFNG50 zd@`zoFi=SwI@9GbBVCuuw-(osO0w+gE7xiaHj+bK7ln>U;Cb`WN2LX!d@os!zNMo9 zLs`BP#lnl-b|@mHo85%d5hsRBEkIMLMg(V)txMi8l5lVmq9g;oD_Ss`evXuw7-bF^ zjZl^Yjvk8Y5IIC*?1jZ;zFnvR3`-Ug$VH`6N<@0B$R!D-24|!4Fwqy~H*Xq>!qC)k z@XCS8mM+?9#Cw(6i+8Vz%D{r~;ODN~|zBJL%rL#0IxR0vVErtIWZ(jRL_@F=`6$*w^hVp)UQ{R2e0=7ZXQ57$y%JVD=4L@kCmd z31UP<@n4*!RWF0ZP&ES$AKe$z)uYj;LbNnQ_w^6~#4C~jC-9mpG5RR$K)MZp8YZ}m zK?8A{&~KhDG>+2Az#GR5mmj*0>csoT%TQPZ0mmL#+;@)#HhMgRUy)1tW`FA*RpWi* z))v@oFg4~K734f8H39nWiwyW!L2r8`vD{HjS+F^cdM*$V#+b_N9)YxUx=%G6&(|>~ zZcn-%*&se-eeyb(j5Z8pXF(N2P>Z;vq@-;Mb_KQZV%cwHq8PZpAt$?Y{?hZTy7Sw| zZGGpH+7OwW(^;}FPvE5PIRDf>0suJw^pKAP1d#k&q9hS}hP^~~;m|vj_~6&vM}r``COEF_!*52_b0j=iEmrt>-B^UF*ru;P4`BQLV<8lYczB-nW=aB_k?^b3hBB zw1-lsm7iO592S#VW(+5yrS5hYJKO{HTLTdV1f{|x5kZ7J=$k)?tC!yf)5z3#RSp3W zQMfF4oQY{n1{25>DCOEsfYHVhlsd88fNT$*st%3q9k&F2sh$T(Hy}u1?N^j&`L<~q6*Lr=NkcqKX{jy%;1d^ zDkK!@U!omUMKK2R%e`zVh^A^Dnv%>gqF1y=;c$x4LSZe1Mtt_#S&P?L6bq9w9-ac= z6T6+w1P?KxY$Pk?i1y(TketcufXv_%Y4iMOo9XhYZZncrW|~R{hNV^n3JjSWByclt zhV2O1msVwlXxPOYIlj-rN?-wJOqP*tSk~p@BpFe9@H(t9MtD3Wvz2s|dWvPqt!vtL zake*7Kb#5^%nUEWR<$lGZ=X_xSC;NEi`{6t4t%O|=(gV1c(L^J*>#V=qPEu`VpTU# zkz(^N5BbthY@0q(#vI%OQb)01DIB8^?H-Vc&0D_!x&O7}+aSwf=gIURZhgXY(X4kM z%x}F2=QiK^xb=Kk`*r^L{r{IxZSwy_s0QX_{_~09|6Qo|xA*@rDd%dPRJ-MooS)*N z4o{JP7(xjRV#yv+&i94VQOo~k2<1gA@^1{GCnGP{(YzM& z%8vc-`McPF2f+WR$h$`z`d^=|`B#2_wjTaoFqrn=e5kDq)cjM0p1}A6odkB`tk}`9 zVt9uGWTLCJrAn$3;pmn%)1ySEC5#iBzBQvxG|ms7<>^B$>Xo2iP8*&`PuR9kDw#6n zBIoh!XaGBLY?b$^c4s0aB9>godcQYIj^WFK7IbqA*#`%3aMU~(5Xb5dVgr8uw+nrlNcM4FyTg3}+Y9HeE*(knlD41aGnwJ2I)vA50W6ybt~)>2uPyAp+xM56 zN14ONnizH00Gd=fcH3mPIssLl^aSFwA1GMm> zfcE#>S_HkAmed?5^jyUnCy+_X7|HIPa11vio)VAebme(jr1P740ib$%FcgIzt_h{1 zW}IWM?z{+xoL?IaVz6~{q>Di9FLG?u%*4&4py7O!QN68UX@VJense-SvYtDvJqzE% zWAKF=>0dk%M>mecxxC9LgZa^ew8+>qj0OSUgRc5R7|KVOOX}uPw~JEoQkLQ8w25YE z`@jh+j;QK$849+f`Dj=B{Hy+S=S@i(PksI;7WEz!NRk|q3%fKNQV&@&(dZYb7F`fq zSAu;j@M6e~oSI3Q>Yo>MB`%F1oiw-Ec1dG8!qMoY_sf4_1vAka|))ORdlgE-oJ~(VwUZ@ zkUtxH$8}{n7Ge|F|E1K{;Li$n$ok(`FdGSg{{3I``Sw67iUg~awUx(6Ajte9w&iEe znuUmFzWEsI>92axepKCEj6B~A!rbvy-o0?Fg#}IX-p;IC_jGAqA?72+@9&*|)K1`H*>rf4Oj8Z;Oy z&{50od?NB0@v(wDLZlCM;yk~~-G`g@_T?>PmX2Q%|JI21TQf?2zxjf6iw9s#aNQjM z;aI^JsX!gy3OQ5Ezz<_B9c=mAG!b9@Vo)_>$|6^I77l`NL;>()BWXd11E$DU5vm5H z+Qf+vLS@N8^MTRJ&Ay{oFGP=J_2kqKbI_; z07t+>bwZ3|00*%JWg4c8uoC35K2n#Y3F8PxQPZ9_Qq9wm!gj@h(onBbL2rTO%epB1 znfM{3BfANqT;1m=vFx&uMqbLoOq(f4aC3#o-Y+F(f)go`FEY{+qfCSBhb|CYTRQ8v zFziWNsEW zELt-={rjo)hDb{*d|j8v{l%7v9<`fx<}hpPXT$ud=s=inQU;RLFj0xUA(n9Q23de) z*#?H2$w;2MJ|g>ys$FqlwH|ckHR4i zFowAVd8$?}VF5&K!(haz>q#p;_7j}vfuwxQi8!R89BnV$!XCMS%7%15+O?q9J7Aw) zfXC%>p1mhBvj6C(I6a@}=kZz9YCCHg$;te}2yvH{aXsOMI=_2r-pPfLx5Za4>5Krj ztDWL>`>WmZvKnt$AUEM(#(1dr>=g2Mb{Y%}M3rR(2la>i`!U|Xft1QhSI8}B>-U^W zy4xpw+`mDpVd3SUkfIrW&P&H-rhR``7I$A`-E1B%3pF4N6Lm3upXt(ZIIN~fIIUdl z;Ci$5aP)k638;&+bZwK25CoDXT_Es!#kTY|K>_=*Oj`g2K>EMKTwF^u`Ru#7Al&vR zG%=L#Im_mJQk3#2cQV{8gOJG`w`>Q1ge`$m`%7|(Rh!Ei3`w259T!0?hn_43_H_@x z1yqEnbV``L3PPNzbzoZyY&+lVX8Lp|=Xbd#NdJP28{)l!x=rKKGwQ)iVSF*U&Z5&O z(K52-A!c^oYaaC8VnyndFa;`Z!LPBzo|frBX#qkF)$>6i!Cej_SKiF`>`_YjBzJfb zv51IZ=CNi-(abmHM+;=2h95w_%>h7`8gLu7v)L4@HK4eb7G6Ey8eqC0lEy+de{(%q z2pfuL22NujfD0@(AnEEZy$0ad!kOq5OQozJ)Zp}U$L_M{8xHv&-F(>koChu91^bX(4!e3? zF=J^b3w;>taZ#i=OV1<=9Q!ZV$Yi0K&#DZE?`CQ0hYpO}Ul!8RVam4MUc9dF$Bhy- z)yB;gGwVS#xnfH>ztH^%K%aZf)%||+X^B6sf^S1%&?Ffuht!41$Nt`8I%h*J@e8+- zI3|Qxu0W*z2T%?fE??esXJ{JX|Ww8%?+v=dkR2&rh6pWX4~ z1zg@$us*CA8pOExo`@#&RK6`U=F##yp@7{lavC;%*)JN}k8_F?9(+GDA<7fH{45f2 zd-b9Au&}hPy~pPDca+c3Eaxr8GPli*A$PaU<#_;JBT@4ySR*c@t=I!KA|< zne5T>HEfS&(W@+}Mscjei||yaDfnD1WJ_>g-fGj zQ>*i-BJ^T^sn#HXi?oDvYD=?$>~L>U^OHUvECAoHI*JD-Uu~`7-Dp-B`+~>bQ@!&g zZxyTnF+h#7vd!@I3z`rhf_OW3T|i_!eYd46~kYQw$+797yPx4xQ61@COVj%KuZcG>YbX5YiUf<=~@yG{P~POJ5>qSx6rwLE9@ z7M2|e0^^1ktJsIjr(ebny*?EpJF5%7e94e08mo3r>2BYeK%jD!7^Y>OzFRV9{#nm_ zQZSGW4&)bGYllCtWtrq^nq>@&x(7-{zB*sYPah{klrbrD!kj0}(P18*hV7&~6m?MERGwX8{ z&7Svpy-j$&X1A3D6JjPFk_;qI*X2N5;|C|1o)P!5yXl^rvN*JHZer{jNtNaFDru&K zdgy8HTA+AE1;VJ9(+32*2}cs|N$Jsa0kCyN*i>zB-6~HnYr18uxCC1j-`SUR%Zvl= zsh0YZWzKw%EEcp9jltVGq3#C`C*Jp;v_~4sCzr;H6dsT_i$=p*a*iN#=d*cPy?B(7 zSM@@CBT<3!a*6?pfbL^iyy(B7o?_nW^rfN)HdExHWv!(uT&>nfAd250Jr;ZEKgl63 z_g;E3gvdY@-w3x%ZQ>R8&4ATVhFP{vhc$emUW(a)e<_>IMJBNqiO~J~2nR-JZI|&) zDQ!O=ir&t9VPZSzxHv=H@qje19XCK;f?`qgY(eeulz?>g>9no~>v5^94=azDB0d55 z&z;JI`!k;nc=As>mDty1e}*9dM8XMwyTDA|#{5QCzf%n*Q{P~9YBZF{g!K^d#A@N0 zz+bSt>~TkaqLxY%2tn`KJ?=V#z;!kI8R@6|+PP6m zWvjMA;$yJt?D5YKgOw3Kn&ukG3d?b2ZW>;DU3~VDYh|kzN6`{5d_-Q zdmQy^;i1S2uTC}CR$6_v%t6d;+zO4>N(=$ zh^&#jww@~VrrAqd6bH;2)NF~f$&(%06@wkxfDc#F4G(gV&F@O(pFWms!YAvwL~Q!G zQ#qXcHtw^LdGF7i$`Xn@*UKiovJ~f<)}I8Wbu*mh(+x_#-tB@puKH7}=^T|`jD$}t zGYsdMW?G0w9D||?ni5H&9*I`dtTu|-;Fu-RMC)2Kb3gZ|xW-{m#ph^==G5_`&7#JM zlFZzuNwR<0sk}7ALqMVeq#+o+EJ@QdXWEJehVtd7vfzPCO=U`3Nwe5qPHu^0vhJ?O zIy$=+-+vC0X)cyuYX+=XX%b(ZAB4CgZhN&BUlells0;ubN=b3^``T765=>eQB!Gw( zgk>YRS|>zhoV76(A-~y>A%Gg*fmu+Yrh#GXsaT9Fd6o30h0?f|(EYmlaPXVrm|?~h zc=XxGEbRG2bYG*UEQ=jG12zFfahuLO_Z5NLUVPmfV=zMf)_luYO-1mJhw$3LtiGaK zQ`ke2=B6S0j;|o?vDSvO;|lh!Spgg2-cCNM4$$MzwIbc7RIEhDS4lurL)jPA@YLLl z>9?nX*9ooEnENt9w}yswZDE6v7TL-5cB zsk<>2Y%IQLqV-xxh&BxA7BdY_L@FATP&xxCwwk;%uNFh1p7osVpADjzn4vGIb{7-I zJ}H=>soh9^>5yl4{sJ%O!JOU_Uj1al3Mm!F#j3)f9HT^Zdi2G*l7OsYKiXd>K|e_! z)!)UY!S_F8TNPLP{!OYU;sHe`%D*FhJt4ey)4%#C0v93z^xyg@Ej+nD`Y7!r)3+s{ zFyoT@K5D|?&VH$#*S+d}9Z3Kd>L}4cZ?*pVNE{?>Ly%dNl%w47Rjp``_UFRWmga8@ zuOoG$S)F@(gWkr+mA_}ytXmz8dSui6Bp-(OHsPoIg_ib1iv!EuaSJKy!uvWC5%T-C z*%IZDS)K@!$M$fhAq0cileaMlC7y!V(^2WM&U96TOdi`eqU5(&SyfFp7m~*#Hhn60 z5AGiCy!OGOTt*bgKMRWXV~wzQpVoPktWLGH+?tnvx&cB})d~D|sczu;eX%TG_gpn$ z0WlPEzGM&) zOUakS*=kE(c2n9OLzo!1*5V% zPy0?$f;jU_ktB6%D_w1U?WEkSUnw&8q)(hqdhaE@Q}pk|D--bD-DLRkZ(}Z}CisB9PbS(dUJ!eRkvO18iR3dHHfTuC2XAX&j@^wU!jL7Qt#1V* z%I=)Z2Vs$LO^!ap5ycd!xe<&}#BVtllLq9YQE1LQ0!0H`*ZQ$)t|tnuh5Gr_Lq%?k zqtUh(e=-4X{tQ{1y~i7J@l2NCWa4WVF*>Vgv_uv-BYazOp2L}n4!3O)iUBSSDQU@w zUCs=*;$KEY z@9n7rRxk;SazMc23Y8c1f86fCU#>3?fd6zraQ`JT?GgF_j3Z|7598IO8}OCD@0BW7 z_tz@Ei*+$%CHKQ$4v4_j8k39J&tcRP*Q@=!sS1Puu}Nmf)mopH-lcWsv9ifmmT@(@ zmQxiy0W3K@++?t?g6K%399AnX1q$s}h#1W;o2rdo1BZk=)iwgx-(ts}kTZJJZI05< z+;+^IReuhWMVUy|>Fry8Tk_JkI;O2@KS2#=6gwwjtLsxM4xi_Fd&^;rI)8^9OE%AI zA60TPzB=Ew^@*M$84nV&dfTY;3_klY(fuHr_;|*0Bw&oGeS>&1oS)}xR!Z*`~Z z=Qb{HahTJW9q_)v?+AdJIci!IMB{>Ug64s+cYf!KUe{ublRlX5iF)0zsy99f9CMxr z!pLER7`r&r*zq!9{Z^o-dPs|d)xe4mgM5t-z^JJxe zR;uDOYi6mG+&{M~hOL&#I6PlW>s9QS3$s*YYb`)LSI zCsSj&QVx~Kybfw(i9)6#qA20cBV$0JXuP%E+{s>F4cQ~gr){ZKy#|jv(vSqkzD&2U zrFWZQ9Pjp)t3?nMW3P&)sF=W(t*wLV)QP-Rwb}pQh z>k~Y#PRwFrDU_RKx7<8mJ^C<|yqxyXid~DtFfeejTW;)NF8$8Gjq;%`m2INOLMPvG z;bj!D!x$;1lzQQ})tRorC40x+3DGFmn7d`H4Wsqd+;gI->#g_4=j^%ZNguA5uRbgs zCpTHJYt}8a0coHdNkgd#=Fr862fE~MW)?zlmR3H?BrvtEI9*zq#S^TBwY;$JLlOoD zqiOMF2!>dd<;tJ8I`?CJ>)Q^;K!8|gM==Gb6_}ESp~#xwUf}0@x?YOp*l6exXN%+S zokTc0v>FDJ^*p_V#%;Bgnp%(>wVEiMUuY@-Hs{t0>_LgBdz=H z8KAQ{hq~HMxykv)#gx2b`<-KmRQ0uUj1POpH(@`&_Pt#tD+&TeJm3Z5Qa~tWsiGTy zA#s-a-2&nwg&Vr7jIVvC@OYZO*||+P6}qUc-CdMsUWL9JeX_WJ#mOLse&=d*wZo1y zo!D!qz5N#P119@=V2X97>Wi1{?7A(wq-Lp)+MO^=i1vmKO_g;)lzmuM;i=u-%`cM7 zTP82XA#54Aix18{Hx3K!+V_qIsFli0lI_EJbfP3Kg{k!ssU%$-Oduc^+blx8x@h+8 zjgRaytEVyJEMyu5nxR;BO;h%Zoyrp$cTGERrJ3tJ*L^HgXjE2MPU(QHho7p3F2l+3 zIN0pqY|%5o*8QW`IrqW&a-f1~dbmHt@?z;RXvxhhnwQd(r3SVX&6@OiLVhEqZ0lkx z+vDryREP=rx9wsD@_*Vp`|$cKazf9c~Ul2 z$>}1jT0TWJSo4_7J%=RC_{ov8+ROAJUyk1z z?>4QB-2-WN3G0i2ZYb~Wtdzj@<|Zh)1^5c9&y}M8#qYsUPNa zNxmryVXA0A=PHq;p))S`B&&UKi@vvcXV6NBG!Rbi32A_@+M;hm4kpWmF4!O1CG^o@F4H4PJJ zNtsIT5qqHzfBbXiFuOaLBjJ&89!_@N>ZA!7$^mS-PQ~q7NPnYZdLWjXO%ZSQy#+qSBhuDSLbj4!Gr4FZJVpt7(V)GO8EG?VcX0{FU&e9}-S+K-BBy z&}fbCACxw`Tk~Kyi?$4?*sNEW`Vt}wmK$y#_cm`GUlOb#w+OFB4IAEX+R)-w%P;aq zLWQ4oGD<=DE+rROUw0ji1Tis19(0rmzTMUB3%AHD4AjlAt(0B-KFaD4N>bPpsEzTk z6Dvt9p`Uz{cFabXpL={@B)oXPxt1cfE-^Yx6k@@KwP{(KDX-QrNLCj6O-rUCmUYy^ zo;z3gPLKcCO+K!wI@7)23zwhC?#8h^$DQrTI5PYx_k=9(0k;jf7WvD>yQZp>PdNyI zr_(cqW(o!dK7x!hrs;NPbDDZ~^z*hYM1LV9wD*kl5Ww*t96IXP`u==G$NXP=M7O8A zv@zC7p%u)<3UsQTZL!i=!dQv_+SlfwGC1|Z=}@A@%W4`~Dyg_DA_AF+awpx$xWnxT zl<<(hZloa)jptdb`zx^qa|q5CTdl2?jNG}3GfgjNt3E^>jq3{W*R?+RM&G9vB7i-ii#^O{aF0>Cx z1&Zf(YL)BI(qvIwK;^Z!D`Hi%Yffg;-oKoc; z4rf8k+;I?#xJ*U@OCco<33}dWDeB=i3luwJtUh6HYd$*@hd8XfwM-(m5Y1{LzE(wC0_Lh-CfMnR1;3Hra zI}7`TP0R6CWXBcg`H@D0=s*Mp4hX5uYI~(j<@PRT(en%zkCHY>@$CBIygJzwgG~I1!m4Ux)p3=y9d#(<%nGS?X3E%&u`*+2O`r%aTCIsAP#2A8Z$qm z?g#va?FLO;Z={IItVubebdpWkP}cMOj#?QCW4X2&B_epPC<9BStappj%`0~Q(Z4-W zE-p;7-qVZ8WnMEau83Cqdl=MD3nRAwXkpazr-jk^?-oX}#0vjafy4}mlZC(@YSmdv z>FfG@v(>IPmTULhe_Raf0Djx^lT5Q)4m#Tjjw3U)7ikE^#Sq^=EYf)$^C~*$+QwO| z=WbSfI$9&2Rl{r{^y@nh!>v!{ye4!%-oB5SVbJe!6qzm1n!NL0RhMgWTJJU2AE3PL zABa(Ig4DcupM9=YUuW`~=yr{0VDp`dK;HaC8rj-l)Qes(7eSxn17-GGC$UAO1A{Lm z=^v}eeC&sIGQH#NI=|N*PQJIub-#qWq4+qoI>nQjrBZ!HLogOM3{vp2&<#m#!(GC_ zHTS^rCtyD9hoqt*EuuNlwkY}`$&(bS3}&u^oX=bFUOL(KGhH*6Da8RXjCM=d zzp2DL$?CUs_AM>t9B90w@PRs@re-KfzM?fc@s#)`pfYhv?(g^T*d zfi#NXXMCha1|N-SaI$^1^R46TcGKKFy*c)(v-u^+yINbmgw#@emO+U0Ec{gMpn~qN&;U*95X1cA*Z!4FoGEZfFPtuFp2MJ zPW46m-2pBT50zt`P|}c=^JpelqOXJ)D7H#9+i( z6@5>bf2OKqHvyV3d$-}L3_CS#X0f#&f_hOn7>Fm!fk<62le=lRwXG5!#$=EMxmb@O zk|4FtDRS<=B#x?~t2g2D1l}oI1l^XmSSh_j%X*V}E2=~rn*2B=F<=gUHC>OAYOxE5 z!sOGZ-ldxSP%U?f$V@~Ui4G6tk}uY?C1qxu10d~%l?i@8C`@3T-aS&v;=)uZ z!gPBIekZ}MMkd?2N$xl)iRPEWs%>)MHf}IlmHdh5fc$LEk|B&ip?Y>An#Jzzfjg^n znP19+XGgWil2b<(SKypGN)6szMdrg+1ydYx9=E>Gl+UXw<7PH939kutHZv@4A8%&) zfa>G2J>+Kt(qnGvTH~{p^c~ABJaH^#E3v1|AVigmaRs9vriz^Rx}k>&S2enEwJ7Y@ z7Uc->hevKeho{H!{5XQv)ju?R;q?lKQ*O<9Rer~Jtan}8>&s_e7Mnu!=^MMe26TaK zE$nyK4|A3k|32^d)9LAdBB}iWSOI@HJqZt#Czy&QQ~wGGmq=mz1y`{QY^VAEw`=*Q zgi9xc6nW2-@0@j_9rb8_QAd?wut(`=a8p3dVbli&OQDf>Y#H_@RObm1AHNg4H$e+I?JFkesgR;LZ{(TAWwR!m2*?SPo&kbR-1q;IVQO7{9DkC38t^N zNOIRe;Os{sq_kzUOSfFFrGnB4FVtrn874o?kvVY0Nie~{vGfG@GdL0)l_YR%HG%}T zI_kba<*9GWl9>UGN@?3R_QiUhZY1$jL{38Kyi*WCa`M{PmN~Z^k^m` z+TbaYxV1c@ics~>=!-um&|0y?$t2JSLPFmi6AXk!4g?3bqU7nm72JhbpI;;)gGemF zFb9ZfJq^iTYG`g05>d*$C<8XbRwCgOR7?|#tD>C{qIRuu7vMfOrX5rD?RDripX-<86T<<^f{3FONe}T zyJ9h-U|C;8k5Rr!L>pc@UGPk__b0d!9%M(Cwl9Sl0igt9gtsbn&MbASD^cm-ZLDY` z8h}T$(yAE-h7~Z;etMr==ChxJal+DDptFqqf zsUT_1pjPTFi)YM{6SRW(HL7)7-Q&Fy>}}WmPQ^FP{BU(+P~=q?C2sH=*Hch5FI5wI zxUJu$dwgwE0!)f#3%*Hw_M|6I#>~!1*~j8a=(dhcXRmRTB73O`#a$pB2`F-S&{j$D zw~+|SJ=UK518a=;M>+)I4)|P`{bxmTmhkR>a{1Dwl5N zoXbuViJ>3LFy75Hl&Cp)+TOIE5p2f1VLSD8a=6xI-+Wi8!67Bz1LMJxhWlQ#k^`Q> zhihv2_wS2j6g(TAxK@37C4p1Z)p35f@;c^MSj$ga2ZN8@T|dshvAE>LKk4>MS3o3r z>eo{#_zikwh8u80I7DHbL0G;beZo1g7!1ZHq0+qz^7zOKTyUle!grrdPDQ3THILCyKJhe0h%6+hdO9a^~ERH&jQN^Z1@P zpZ1qm(noav;6K`}p%b`cz|udnJMzN0Vp&e_lze~p^wqX+=9%u-W3O!?zf1pv8vuxL zB@l_oawQ0rzI-JZ)4o|hI5z=Mfh3yd4U>j!Re@A zOPyzYG(&I`HL@so9TMW%b+A#a&d>M{7^PDn5Jch>0~7eZqZ)}wI2H(mVKdR;KVfE4 zi>3pCOA>Suc6$Zzwb7{IssOZ2Tm=}K84%l+(OlJXu^%ach&yN+C`|dS7w8~Nx2Soj z3EW_ju;$;$MKK&Q%@a=z8)uaRTn6F-FG4I*K&nIK3MNJ*<(W8D7Fz``!H= z1Ahvsp=7rhzck)*#(P$syjur;1L2jdm0dp4mDWZuSPEq&=4PY}ih{6^{52>$TSNDO>O@t4{M$ z2BUU<mIQl7fsMY@myNpMIn2RRa|K?(AM(AcYV`^nDTv^U`3CvNM5G#3py zkd~G0m7=$IVw6us)4-Z_Awy&F`5Qqi>psLEr zPl^ms4Gja4#beP17OXzR73+k#bV)!}hcAnOAXM{vNxsh+Ie2o)8uU6+$%gP&v(w@^k%DtxAa!vno^r zZnRUgMY9({%ayJF{H*q(vT{(I*WmY}g*^&ute}8ZTjS`;q&LmG?;prxVGbO$Y^2%P zHw@*Hr8cdT=6}_k>t;FZqAUj(vVC-? zmZ`EpD}9sU!g^ocR=+=GX))9bw@g`ySvx!{;q5P|NQrSo^zjb z-_JR>CXehJA25W|W`JkN6KJ9IqgRSTMyaCk+qUAW9peROp` z@$kp{99xF|#%z?&T;zkxt)~I9U?{ma!`FUq6C>T4iE1`;c_YV_>2Hpdih2w5TuhiO ztkvUl$xN5O${rctpR(8Wei+4D@g&QeW@e!SZrL!u07EtG2Im}PknDO*w5%n%KR zF|xLCnUZMvKly4vB#fVrcqkBk)et2bbR(2%^RR=NmGcx7>~J>xPW}V=Hv{hs z@S2e%iZt7^Y9kVmvc8o{IrBX4waf2hP?S* zf|Zgo+Wf_BqB~ zI=x}`$^B*>4a7u?i(OzAOqy7~dVEF6<3CQjknv+xXAm=p-LmF0<3PB>tEUy|h|N5r z&VM7!SQq8Kr$z=eigPsE>)e;71L^*Ai=XnY^mA9~Wrku3EN{!74(@s#<`6+n0|%dZ zKz<$cdYIAJ7^S2=6^(xD7qFc=p)4tvhp`@#->DTees({qQ)n_<%8QQP*pX_Hp*cl} z=Hl_bYN3IumTMxRTwaF3crH_ds2N&hi6ni+a`c1YaiTxerd~wGlu?U5F$%?YeIt{^ zSw$2WvE$FBu%Lxje?(Wm6TlyyMP@1Uxdc-lD2%y|<8O_jBJ2dMGfj~vVP)DX=yrp@ z3dlIgyuq;d1Zrg1WYbz!RdVax5y(xQqqCGrX|bmcDp@lw^*ENg@;=W*hc}0eCH}5J z(VakiU2MzGV`9i#4%NXm5?`nCCr&{u72%*56e=m`D@m7Bv<6IETehoDAj8gav6wtkYONl@qG6tY-SzyxpkV*UCZj)K z6+H?6f!e@?Lxf6jE>7&kTQ4j*4Dwpn(DL%IU!+NuO!ROpS(t(`iCs6K&N(*p87#VN&(Pf4-y9-- zm#C=Cy7%US9Uuo{@7@mbABVq-h{`P%aHw8@TJ(}1gqaeT9Y6E?(n9#=M3(Ww$TkfK zBH?F*f)!ZtfLa6p=WlBVuUj5GBnyj@jK7}VlSD=-V?PLarPz|4{;$4T)m!7YIRL|*`dNiZAhw)u))9Bon865Jte z)z5=(OqesC(ocHX{fy4?8N$FOi7g7NAhM#D%x%`2%w38l-SHQH{qME?fdoAxlRb|n94uCj1*)^X_zJoup}d8bKx#ZccM*kA0(BP8oxja;%$ieojk zFLQBT;bK*s`g&QCOQ4r&y7>%DL4Y}eh%EB4pZt>O+2K(bbN|KS5?=Pb)~(^^7}T0nesuVChlvS|@Pz=o%0^+<_T%Fem#&eOj!6rFOuR1{ct5cZaIh3c zw@hVVqa=BqT?q>(xIRb#K_DW&Sika1!ZWigx=%-hm-=YWJE|SqS(-cvoQEtGZ;;ZW z%p1%tq~@O-dLr_~6GrJc*Fkk#54X~e%Z^8(@fY=*xV6K{{Q`T!1E&bp59ddQkLadzNnV3&y>@yCn*8lR(acKb%Vde+G|o=$iP zd6lt+n27%suq{er#;RpkmEtws>gQ9H1TpLsB&Es%${YELdQ-QMXA|g&gndaPod94) zJV`BA2JE_O2}lqRxNVn3k&JF-dfah8u$v9u9aJDe+#u=uSdGl6AXDP%wFQ`Pn=$aJ z(>wMQqMNlwex%4jNcdIX0Ade3uq25A)cBe7_A4ne@F3>SHKV)zH%--K5Q(~i;sKQW zepXFMlHjY)B9y4RE&3-qDg`S7C;{wd4Z;}I72_~l+IKfTAPSn7`J;jOEYi~QdW0h4 zHZ5%z^1V_aeD?b_b)(i>$~R-Yl&~-bRtcI}=bo`r0bUQbZjhsF{7d!Q4h>}TCZN@B zBZK9~#ANU8B&znAKj26{rp~>akN6J)69yFg-wI4tY7;eKf&VKLIPaL3#fUqvwo0jP zz=cbVeTe(k)F+SqAOA3c!!|+J6tqg_TV*J38uWSC&U8fGl>f^FUKxtAo$dQ(Oz8h( z0=FN&wl-3rT5M3ddv|@j+_X1i<7Lg}6z=)@?D^50m{*ezG4ZumIOlo1Cy$}Soo@6B zC>;!=TJrQ=6Ne;*+m;{60*G7g!mdBN$UNRt0E7OX)j$Eb^*_xy{?$32ymS5k(jreU z2pfK_^T`Hk* zq*Z&ntW@QodByIY%^ik0xWCH-_+?=o$DIdKh6Ga7GK^bWQUes0_B6EcuI_mG3^+B!Z)!^JOroT1l0y z&Wm5MUh;$AOWFCef@~i2`r)3i9s+$)YslJ#^f!x<^Nag#Y?i>yE?$dMZY8HT(aK@s zGvgi+_4N;iSo`j*ytDParj{8>8Y4vQ4n=AecwRXqTJUE`G4f+RMu!^bxK}PEh+YT9 zPGnrTk}ih$z4yont&fj;pA%7M+)Dmo_VPE+yr;vc_n!BZiG*Jkw`>?pRcccpLmXk9 zE7^=Q$znRLHf9M>&%mzBO>JH=VRZoR83(uchukQFhqC7y5R!kI{Qk~(!MeZ3dVvWw z{Qy}(v>ED)4&GrfL3BsIn9c6&s9rTE7w}J%cE?E=zHXj2dBbc!u{Tk^FKxyY{4K@G zJ_OGuM-@{R@NlFE9wOw+;~uvV$OE3<>qsT-)x9IUM+a_{_dXb_RZwk`$tQIg|ce-*c@%C{Ta3Fm-Ft!!EX;W<@}L_U9%}GS_(PIuU== zHV8QIYsZMlIcgWmHZTNBTJZiLKES|!cF>#?@Pgofr||Vl49y1(CM+f3rH zml)m3Myw;1(?!x;^u~|p<}TAj-5*X~CXRogAF-?NM9waxoG|fPy?GFSc?4=YCZE4X zYa~HOKS9z5}@3 z0Fnxkqk+?XuBHwg1V{_}0x!KiBG7jz`CU!ncm!iPOJE*$>C+RsKt_q1zDmKA9<4x` z!-~l0aBs!j9%Pac04#_ntzq;fd(u_k>$|RJ%pPVwtBhr90{eqMZ~=b` zVX1Nj^Rf}F0FU1%VsgszPGH;}e<^_Cb=w3tm!+jA-+Noq!*2CPi2Q;0}iXeY8j{tsSB=`S>k-)!+`Tss2U?itGG-6!QIGSSP>BNxLyLi+=~Arc~O9x+B-<8$!zYXcuV~ zkZ8Y0hx4phd6!A-zIBz!vSS4Ny~vv}h1DDzNZ0%LjOmtsC!_qj*)LSp3@%y!Ge>YL z+wwTQY}i)swA12bkK2?Uvf=0F{x0JY?`vM@<0RMZkV}TSgf-iQ#V#Im!WGqbWWI6DEes0>Q;B}s~nR9e-iae@N3OfQC*n|DTJqs zRR}>(UY#tUQtO0RW&P8~rm!2zu0c$w2Ip27s_&!&_060Br)M+*faqBvFh5jq@+!5V zBJEBV@PttO*Zcqw`<{k684uQ&-Bx@Jn-j21?OQ1W~F0;Pjb z2oIIV{3qjhC0u*MQ3tX)@O$<1%jOrs+?-JK&k0lCn{V#}W1}9kS^_B;@9U)BtP8Tt zP`}bPj&e7vTTT$H-}RF~{LB%LWDI8V%VtGiMq!^5OLYz*6|OpAz)u@E^2?%_LJFn* z7rL@5pcFe(=&18W=sU+afFaYX40M#i1)3mFZ)Pz_87=BJc#|IE&=u`+EQnu1ejqtm z2Ub(qB3x3hS(Q2U595}Rzj4m)Z9XUMHlV5GiXW;mc;Ol9LnyOf%8iAeSzyI!mw{CI zD=i-y?u@2xc-gnD1~xagcu2oZqB{tqU$mei)M_8By3}#L?Z5{~J;EoZ5;JB!7q4*% zH>vcPO--qEz&JnEjrDKb6 z9YrOBVthE@^$6kF`BrA;s{7KkOjeg!0b;Lxqr@4R&%BOi3F+Pp~;)P{1@ zkbW@)K=AQNTv&)&m!W_QC1Tmar_K1cm8nZ89|V~ifGCSc)@#$6x>G^AsL z{@mhJKPh4)CeQqly|#TMm(E;u0Vf?i;q^4siCt}lSMwFmQ;liV1^8)o%7Rq;7JU`? z0eqxwVc>EN^J-xPfUGSXB%Y%Qx(^VF+I>p|Y!zR!qI%$`eDj6>cZshuRF5-uBdsCq zI#m8?j6b4HE#{3K;<%b6Db_q)3VjUIC*hLPz{o;Ue zI%-4PO0Xvl9%OGyzJAw@P#5_G>5HoP`j(5~R8ET^5bv2ijLO%_j8uk&eO8FnkD|j^ z+_;-O{EY*&DJ7>WK@-&2z)7Po`{gLfF=$e6ac@lHi<5=3D&di(jYn+1+O@aQn_qA6 zeV4_^4DV;0TwS6wPJ|2qiT(u`fxm%K?7s&_|BL?$7;}@9je3EX8h;1I=S#hR14iQM zhbPaC6tMEdS7rVHqkl8hETBleSj^Dy+`dt+rbIhHnfPP`zyXGkKON{v_ZrotCu2mf zwnqWMTJ^qQnlDLNj=>Eccl-whcFgr!qHd-_yiWKHsDkev(444Mt_*OD-*_CJ>>5aw zsaiZMe7{(~!KuZf%TEMMA81PFZtz97a+T`hOp0l6duxI0=<9v@6M)I&>>&7;pC6NC z0?EZoc={B9TBlKu%|4o0_v2EN&u2giY zH>8XZMvs6H8yahqY0bZRXTB6kuWSZvdEFyQyb$Q;IiB=_!i8&5?~cg(>0mBsjczEf zbf94ve+Sq1aKW)ZInjiI?-7qGaVDW+_i4?dL^-9{mG5xCMx+q*ve~LiF7!juHWe}5td`*ytz4ade`F85)j-Xgcy~g?h#e7K;b1t?d)W#L}-APrPT94Sl&Ad zYpZ;96F(8Oi^CXXMJ$|i%?fioEr3TxlRmIgmgMZQkPEo~qc97Su01F(KO;#*L(Z?% zlBZV3^gciQ_N7}?9$U-L0aQ6@oVi{ttrMP_7y{p*Zl1nNpHWg4SD26Uo}|m?|CX{Y z$wW7gE}t;z_0PS{Mb@zo=cqy9aIfjX~*3GTjj<&Io+rN$Xb{{Jtgn5Mm2?fxyDFMP1K+1C`RKS(^1AS$4ZIbSGGAUoxG|;><^c0zv(dL zi64QdRNk6Z^9B4;(h)`Z*H%%VS7J2cRC)k3e%H;*hs2(1lw;x*eQjv$=c?3Vk)3Z5 zV{*~9v5B}J52Azgc1y;@LNvXF3KDa^=4Yimz}PhPj5>VUYG9y@TZ6;I&Nh8PHY)k; z#~F`yp5AUg+wU;^(5RTA(R_Y9Pf7XuWGT^S4F=9cMg0UP8F}M3Qg<)j%8;owqYqac zF9D}jV$JjaI-j@>_yrL8Z$z>3LSrb*hvlnP)IavYUoG63dlJcV*MVJjFjw{;fnocZ zq6ru5^?!w+q&#HF))RJJ>IFRU;Z!L4O=;2tMtHUr+K>@Z8`4Si!;5rUGa9*LeIx5 z^VvvU`;F>Kx95BKcEg7fWxwQi(Vrt!dm-z}&emNR$_wZH#V=t~wR30nbd#*BgE>!* zHJ<7@I!(YtS6ufAZSa*rL!1Q5uJGRC#FS2RQ_MWdi_anGQfU9I4)z0{<;duKYYQKZaWnsH+oY7>x@hH`(I(@+pets@K`}~~Fl@qR^tPWvor~F0;wPEOWR)17FY6y1tp`1#^YZ?(Iw*P{cV`00FtZoPY`;U$ zlo1DUA*l50t%gN+@2QQV+$R>H;!y`VXV~&N_R#}Z6%z$;7rjNp5URY)=^tI) z=XBRORoz@~bSHQm!~&Thnp6}Cd>5o}OJ~Vh&toj;K~|2n^k3K}MQ29gvdajA?l3S# zUnO?O4=TNSdUE@Pf*xZ5p;Sw?QT4C>7i~}wpsy6Mdjo^s<3WER!cX0sj!<$9lRks1 zm-g~J7W#LKBZ-URjwP`%Z=LmkXXP%8#h&wd_s{J7#m5G@)S36Nj*Wxox_r;kA64FV-8P#oV@R<&JRA9QKetqq-g}`5`j#!;2FX zV}tfL00@7?Mv-B-MA=|tcTPZXrz2Ajz0c4HO=}+4r21v^c8X2{G0~I-SM346Oi$1r z_uo0$u4C6S?q6rQUd6F?qKcLH%hrvZZ%L^4m+x8LEBiS6kG|bkazUP`!gTRXGj}n} z*wQ&5))HSq)`M+rHYjR){lFJz?|}J*o0$Xz#>}hB=))X~z@!)Ehr}A`FxVJPPsX>p z2CKn2Pfh?Uc6S%eEI~~GO+d20jP)(0Ku~dnlu_f!FkZKQjiq?Ku`iX*oM8{Ke3GuH zUUkwzJ2NKLrH-uTmWf1RoSO0^Pq*ix4FFZf;rPWFWEjnx$87( znY?L`15KcNNEzH3(+)&`R!0o26Y7>C#|YUZI)x=l1WANF!V(uNbWeE-AxZeMt!LVb z>X?2s;8>`NYzlM|j-B_=*cn+mx8x$(HGVni_WB*F_JKz`4S#)?y>9YO`f>MNQKTgS zpKs$7s?$C?i{dz(?KgsXr;3}+{iiES40U3t}y!qhqZE#W72&q_fHAmAql&0({bwIufH&hOKefQmM|9-%}1v;b4W>>)nfY- zRK6+S5b=1d%c0^vLv#B|>`j!6cP&$}jMdXn2|UOZo&SKP3A#3u;*s|IS<8Ry^RG;~ zNS@4<8?QNE#JTmx?|do+x}$&2R@XM!t` z^cb~^=`1z5|PY(cqKYUnV@!}92k)kiCkgmc_0pPAFZOtcrMe`x}Ew?j93{krhSobwISHA0@ zaqi|5Puqwb-M#RH_2>MiJzYul7Jnh)T$cME&^fW&bYVPNxZO@E)>}U+U=Z;vCJSL$*1N)@HW8;l z0R^q7cYMxZx5yY1Ym18N9#%wfR2|W^}q-0S_m##jrE0&WrToGuH0UN@<6=hS8jq7Kr4cmw!LU zMHfUw0nG(5KDy^;OtHD^n`N+v5-%?w4HA$4BIOiwlC0yyi~bl{mFm~c%bSU6Lp3~Q z%u|?LY9Mj#n0+Gv^Uaq)vtl2gj$=Z2oP@Z3CDCsopgY7x zGM(>MZ{ddO%kdh@N~r*B-7!-+-wTPbqv9qBWVqnhor>Y#<|yMQE)a5>Z=cxOqi)Ph z6Ap`WyG5>UP9V)~4Q3_17dB>i?|SYBsT6o{k@Zryb63f-?@!>ZbIb zRD01KyYz47k~2box*#fy_+sDS&v8%m|Ix1Q9LXlI816Vdc1ukEBTzwx6BhB}Ct?@G zhV@0Rn#t?=0K3&vI=wDx_@>#(>R;Dd1@#a(ChyJ}#XH4Q+sOmn8ptY%kz9Fle&~&i zc6tomEGo40Rc^`Q>f2JdUd>pr?8WZvdgelHb@QpEFnre^u^D|=XeCPs+uq7sFdUpN zK1_wI%~sCULm+Uud-Pg@ifci?7n^!S-RVS#`_o?u-3p)#IEVQ1Dt6m_R(hSn`3U1r z+xcAAAvYiRayIU!yP{wup$Nz1AXM}ma~HlSIeY&GHqROIa(s9_dOFXgwe))S|3%CC zN8X9wYe+~NVIu`2i8OoR^?1L2F(<+BLU*sd$EsxWW*(1TOv@bwck`sL&CrBE9{!?? zqZui>iiN6&Rg|MNlc(U6qcKLVb&_xO4)K)Y6^{%l#L%TS^NrYVhl{ zs1~2G?QN7Dk$a0LcZEOG!k;G!qdYr`aGsK=S9DX>m!v>ys*>apn>W@yiPA?l3H7kl z`JKVfhDn{@PW@H(_2uqG@JCl9d_|Y}t=!#$J5s%h>VcL0l;=svAnv+-G zy>ZEQMnfk}KNf)~K_0Noi%H67%l5M$ zO`@MUf3AG>%WsSvavO(<%c-!^pZ_2@2=~C{*tV9taT`iNOpeF}M`Hfsw7d2ta`mFr zI9-&&I9a7t*w=zwVYWg*c0^xTljh7E*Y{+CIxatzv>p0uKr2-?mx;Pkap`**Ew!nK zdlJ3+Q69cOiOx|_4p8VYSD&=HJAr%rJp$1Vw=>2>$>NoI694mP5K&Bbz0PuoncyFP3^ zOs<*?SWR|Y01SIfGet>E-mr=rt`66X&15>w2Eoc6QL+DkN?Xh#O-BhH;fk*kohB3( z7P=sIPUtsBWE-00yYR!x7}!iA$wb}x>lEYqq5%CW5x0yE5<*bWmK@tgYAxOv5=H2%A@|q5kwiRm#4=Ega zn_F%^B@(-W_G}y8gkL>rA?PGvWcc|TwTfLU)h6WEQ8+y8zme-Okrp_o9G{lttjF2Y z21`dy&&9WLF|ca5>|E7ZYx)X2o3p@0bXkh!l#i^RxY?m3%mLTd9jAyTJ{j!q*~#8= zOM1Vg(-2-W9pl; zHv`p`MEO?DYl_EkYZw2puGTJ0k#{%V>VYmbne^SJHDMmV*7H-l6Fs_>9M`MM&8(VC zyT4{`hCTtFA?Jo@{f7Q5AeV~!?YO6RL-7;%+v|#qVj)NfE=kpx6Ya=?VNQGUJ^mQZ zlQID67isERP4-n#wezw;QQO-I5X}Mni#J#(R1$A+0w!T=BLJnz}r)#?rK%1hS({!x<|- ztz87?UE8Y;w*=KC06=|^X1f^Xzxdc$RsK6)yply3V>YS2+wFX=g&_UWDt6IWIi6@z zG8ScVkr~ynMe-qKs_EfTK$4a}t}w=z;fBM|Wdgy`ZCO|w({r3h&<7ikUxpNQ^j~G|lm4Op{pL-h)IU0@T>c`m z5+0|^YnpDs0Icx7rBy@xXHHfCWXSyl2W(xC22^(Pdm(ecAk@@&{oF0~TR?rOFS{8Ty zHJl}O;x0XwbL>b01eu8$PVXU<r_S=OV^10zcKiAEe?H@ zAj1zVbG-0IFC0t1n*#oWVUd7i|34D)`|)h#_*}uO8`PGreawp2Iq_N7$#1li+k%0> z!1)5>P`SV?hd_yv7rhAjk%;=vyG0pUJt%PPFoyLiR#5 z>h_%qd+My-XZV-VVaxgdbs}pQ&9DFA0{4hRaKMXk8FAA_{yknUT>}=hNG%MyL0ki?w;1zv~jO%W1>k z$b|6H1;PeyOOG0eFKrP302=V0Km!1L1E2u_xBl%SWB_Pr5io!T4e$*TP2@i*sQ@wn z95fgVP(}lwK$3|7_P;KnfB-aTKmZ^Q4L}4bA_5rs5=01w6j261BY;IA9BBX$Fl1l2 z@NnP2+WfBLEUbqDcd8 zyCVVv04#8TP(Y6|01pK~0WbsskN|-JA}nxlp@3dxGCUM$3a}vv4G9nsK*jtLL&#d z2?6G@fUsb}{a{BtIFz6sBV`^WlrZ2n3y26-qz4R#2L};EW~dB}6blBTv49|9Mf$+f zcyRDRWJby$q}X6!JPQZ|7LW%9V8MX{0m{k%1QZbL52q@w} zFbxO|29OUX!-5MA1S>0}A)tr?MQA{9FyK5e8Wvn&pop?E90JOBAQ=q^8U|bdR>p#h z3=~mOra?fF0OrwvuwcN2U`H&t*uWkYWgY|+Dd08@hzLd`4-AI|7Z6CMtPG71D*{BL z0YSou6o93%;35LaRFpvovEsmZ8W6@{aNm;p{8D1-g696w&Y;Mf(sx!tLt|rrKGf!W zGwDBO|NjGqmrwTN$pH-r)^l!XesCU1KyR+S6=!^+9k`ma7x2^J~Axh0knXu&9_9E*cJly;Hvnq?@mSbfGQ0?9wZ(Vx#aSw$POrDpz z4P%EIDpk|04?jNsV(cMZb8Bj`cw zGc86Jb|XXO#|x;+^D3B-V;!4+myQ1)w6JHr{m2yh*%dxE>qf%cV>>Ck4|6k_EXE`; z@4d{X%b@)*%WB=BHykSfA9GM9+8uM?g}13lBWSI8cFhb~{-@8RE&gmmACdg`VoOwA zCcJ_%s+|1md(@jar4x6rw{02r62^KS?`!HaoikH|+RpFWycs!Nl|&f~asx4?a0=cH zPKOArD=8%eWj8X>$sy-+evzi`d%exdqKDO_Ds>rN<_n)U$J##)$uWLb^T`RR zKW4=iMk1&b?;7P?gem?B!wIcxE_KVI(0NXy_Bs1){)^V@P%}sKz+0Oxm5EUSeL@Gh zGUfv$o7{nrX$V1SpZeulae3__jB~W}J(m?S%7NPLX=SQQ;h;7aSmF{gz9LtV&~Ha` zNZ8D38O&*3Va3yRM$DWg@IANHL@6BoQ-U_$q;FKF&n$uOuqcdKY;4^~m^;KOD6VuE zKW?EW5&H+lN%#$0#^~eb&2T$$goB_caPJXh#PbD*4{;1hYBNu99a=6kc>KlG9lLEE z$H6fD#vf*<`or3?)6WzAquHbgE1D5l9|EzM5;4YJ5fADm>&ay~=!7Xwx7BSe^ajqo z(!Ff09Si=}D%;7VYPKT$(!4~+Gw0}T^ys4ev=)H-Dqi`f2#NY0Cs^!a8w$~o!lO-gg-OyFf&18fZZ zgUaKp=&0{$jXn1AOgzn{FL0DQ?`ApH*O9?MBD9yn;0P^`m(e)Q!hE-D$GKnOHCD(* zMexIQ<@}ad2>%kY82nRx$P9Gl41d|ey+>forgKMU-Gc@sdVlhmxJO$idQj*lK_H!h`XoU|o1-z8mLy-wMSj zF(s_nF{Lnp1RC?mFckVBM7-4sqka97G483d0upFA{LUkhzkwCyh(c+hkmRx*tCu%> zfM(Felv21!Hn)mOHmiDnX|_o7Mt7zRkBCG%QH>8|L^ULsm^V*DchGm_uNl=lw$BXZ zAw6F}X$CoJ@2h1DsKR9su6oXml$y6|IZ5b7CYhBnI;-MiHVwk_M?g@Qba8T`n~#^X zUySL(7a181ZVC2AVbk4hpYF%aqh+g{b!{kAQc!Ky zR*svkT%oXGem$xJpJ7g4+;VbQfqs#|#TX-IR;OgROk9JGc{;HHVmhKhPW*E&-_N_F zM(c4sb3Lo*O($aH@``e;D+@JgfpfT+Ll3r$c)9d=g!ugUCG4aytFwTL+!*hj{%e*&OoZan3WEi?cn{ z6Fvp%DG(3<)sxc8jzMzWL>cowZ=Yck-c?kV{qu>#Ei=jt=^<-{MgCVTI*VRi|ElV+ z=oJHwhD9(LOMER+x1+4nCV}+8{WyXa_tZLv9W^k{5xdfZLi?167zaMiNN3QsmA~I`48ER6ztwxkt5AtXqA}vNxqgW*5-=VwRnS2RkHGP zm!gyO$C|lumaJ)qP%mornIE)NkWf2jttDUOS>hH9A@fF}WOBAtI;I`&8KK}g9a2Vk z8WBdrgx3tn1tFs3ce=)pmia6P1DP!5lQT%qA-jxLl=E$Awftcg?oW&x4!8sb7p(I~ zmkS4y#OdffdH*202UA^9Z*n_RC*qPkiM~hq#pb5W*zEU`3IPOJido@~;^42Gy^S4q zBBcBu-Ha_$oLW9He4^4>_n#Q5Xq};xg8YtQA}82z5dJLU+IskoBFZ`qqT;YOVe;Qx zCR^4S377FUKfwHDeuk_2&8E+jF2H6)Zi}~iM6DOh$UVv+#WG{Yls!IdJmGAPrI5*}d}jAZe~D#X z4P-=4$Zm+#GnYMoAN|J5<8*_8LVFK#KN@$CDL3{`3}y#zfgJr-C&R(-q97<_w?>0c zNh@DrC|zi>uI`7$g-k&byJ`>S1BGK)xErNBm z=T5_B(^sQf5ww>O*EgluISGh)dE*tBCibSQwPIUk-HzN}OGz94n(w0T&lWQqf%?-$ ztz;V!``g_qz0=$wT?v_kNCDx&SIsa%xs*f-)iGT{PkQar`sHxi_tyAZsYU8a5x02Z z?_M773tGlARL_>q5!Yv8RJRoY(^GJE(|coQLi0*WRY#nUq}*y_KwN{UT!&1CoEGxs zrg=XYe^O#`6s{K`Ju5xo!!HqJy0bar-)DbY_vqe?`JrRBV1zeAI{I#iayj0HNjfy` zq|_dG*SHxBmi;@~ZR^NpvZ6z)p`(rgRZ&(Sh4P`2(ACGQ`*?qo(jg_aldGen8$E`)u&R?6(cQUMpI@qk z>ZycssuVw|gjJN+#;K4>h$2;96EfVy)E5BjW~#$2Q>%}a$XV@&XELHzH+flwh*Vy$ zE!AZvN^4ei@Qe6qs8(61(#H1LL)^edNK}!xb0k0Yyy--msK|4ER6D0pp?-IZR0VjwInL)Q{!tstp*tvY^ILyyBG^Zyc zm}Bf(eCL|jpiug8+N-Y@OK)dmIiGGH{)=XBYfGxw1)7c*G?UDmO;26b@LK6H$h|l~ zx*cqnLbcE^z%)pexRN?~^78&XsfR8QwHPD^rLI+g&Pogdu^IH<<>Ue~~2n zJwy&$I{lW9N#+}O;uNcIJzMe?%N$CQjCIm`zHWss&Afzx5`N z2OM+4VFM6ZNOWOX1pP_*)zCIjrF}Bq)Z{l|j_$~Tq8qGgEj8bDo036T)+f z93eZPBTP3Yk`Z2ujlCGhMQcT!cNz0>C!JAmlv4U(e8x5^Zq&tWt?r!R@8^&?s;uc# z69X}FUL&B}N%U(aJzWSAPYBknEZW0Q1@;&GfRp3qf=~H8Z{_cC` z2>dbduoxzoSwHB)ULHxR5yP)3Tf}Trg=14B_Bc;@=qJ}Ofla^c~-|x0n zU52fu;W4}7G3J_b)yg(kDsjK!RhtleKz0AbAKR!uCnwe5!PH%so^8ou;a$rNqC$1s z@M5mz5}$DAPU$<;TO4zB+yR0VAL**NGXtF&q|s~bk9?cP1Qk8iZ^OxNI!8~_;Y5*x ze}`MzSo~MD^1Srp#g@<~@90r_U8Y+-DsY2{C!d>UY5X=z>bv|lKqI2N>c@T9{7KoI zLpjwljD_QxD+81sKq)tk&&3~&nt$7)7iu+gDR^yj4cg3Z#yq=5|SLq@DGv6u(dgp+k25#WrxixY_V?jvqApV%C77$;N;u zEykCctS2>+4;)>yAvFtbcRQ99(`?YoXufsEi#Tyb6ofqNC?5mb)qbnzr)%nFM9tUF zd~rX@*YYWJXg}0ynDlLiffgvWp$8+~DRLc{?Qs}3%+9hunHbQ1tRP}?lKjiHOGrz$ z0G{cm5HGXtsHPErTChRtV?Az!y?x>@*Qab??vtuNcFN&Z7a{8uxLIeJS!>t$yGiW2 z0uXF8Rn_5JJ*+|Lq>MZut34ImrGtBt_gGM{JACx-u+O56O)zvQ6QG}V=3+zA21hR*I5|peC z4+#d@&s(I!S5>7d9L^4z(yF8FF>+1Gta9S2Wkp{Rb}AZ{DuFFBIy548@K`c-?=y_! zkFAlUzgD>wh?UWy=^LfZgmmkOB`8J<#yfaj^b1kQozMu1EVB_^C>P5=>pz{AO=g40Q)Y9_`?ooXYb7dV=xqn6X*N4dKIcyF#(!S(y3g zMzC8gzlJw&{ku8IPgd`+IJ28M=jub6`Ss5l8?Sj`o8aBTH>4lm=l4U0KkrY_b$tmb zU!0x`+rDU9-+bkxx7>*|`1f#pKpDzQ&CAEMku&0Bg`s!`{iM}-P3NDSL9pd0C2e_s z9{$W$;g80a?zj;|JltXiNh49tAMp`~M1C)UAZIgNp(@Jhx7OPpa(uiTpBw@+T2gJk zUf?iVeb*kV5Gq}EUOVSfV&RT^BQg2ZUstO!+>`$tETCxOkm!rm@1Mf_po0=c$0YAU*S$ zO@3eZhG-Y3v{x6qm5Mj1@%RqJ4RrJ?V;kuKGK}0_T%wm|%`HAq%kAaBE5UkftMs_D zHuJCCD;zC83LwYkSC@ z28^!UOQ7L{*=$}7^g-N4pR6(OS{HnyI^S9#1xla6tm@IxZx?~9;>*ZtDCN0th|ui0 z?wr>}hog{W#pH&W@lzWZ73mdjlb?1j{T=xQH% zCxFj3@~RiJJ$f~(W-~qdvlTnL7nv3hYwkiVzKEqX%R*#ZO`*m^x;e@1^HB@0M9b&@avc2XtSYr^{rnIy zbo5!iD>t?=%g6uM_S2hfKR@{+K=UJD9CC#@<~~&LhrIB$!Mw(M2JBC(ixFt!yiWCz znN8y8o{|#y0_zoGB&nlCnBD8-n4y`30`9N*6=JzO5ZAXurj#;CcKX6F)idJqBlxyo z{UCJ1TzY6|fhUzSQkzs-Dps1qliXk9$rDAr88iGku||oPCude0FY%U8ksoE2Q0fh;{SqW*&b4Eog!Hwrup9hN=h&@FtL#R{*(Io{thJbm$1J;rm zbMtuo1uCO|eCQ|5`(t^VzHUfQ21v@g@5BDHfhAg(w`0g(n1EnGL4$Tod) zO1ip<0tUC`upZpU?0y&5>k4FSjuSI;575A5!s})GPrF_+%;U7{y;yBY8YV0`z8*Q! zvLJqai@Ok8#xJ^M#;6%zD=DHt?91>RbARo33(v_e#-s=~GBI|NSn^K26@AZ5_X|djA@~fCf2tn#_r1cCH^}r8(fs_|G z+(OS)@8;k}t}x-k2ETaLW~P7;{x?AR&)I)mWVrvnWiAJ%hQvDGx69ERym zE3^;iC~=;L(L}b@+JqPcDa<2ra&-C6c`Kxs)sW>TRkN3!mK}HH_d@F1z8LbQHi|56 z3Okf5UoLzOmSlD5ZA^Fl6Vc?pgRjJXzOy)Vzm6;66f>wCI?qO5>b9}@Z4_CNqGH+W zcgB-cVWT-y*`|H9xZEUBX@U~a+;{-v<+)%LDO=E#4(q!_*0Iee)0(}V&9~KH$Pv# z+VVmi9x`X|X=$0cWs;p!C%(Qr@+~Kwwz=iIig!dAY24p>HNyoBL?5~KP(h!Bzgw_Z z*IJ`GAs;R`#+d>_Yg}D&-QTx-Z$lQR?AqGhXDPY|OTvf@yt-lKtQ5DqS>Sm7pmg(U zZqMhvT|_=UOdYLh?OZkYlfaf1G0!gIG^<6`Pznn#tA{}gw~4<5=Otx#34eAGCVlGL z@%&7=15>ai@jXs??$$Y8WMW7R3=dp)znJT~XxHZ63(!RPTC`PfEkhkSG)`j7i1~X| zYUDPGYHCSL z)Lf4op*5D7!3S)MZ_6mPNu$<&W(0>wYyze?Klu7aI#2tGULT#me0IoRcUrspmyu?n zH6dVfYP?k0+oUQ#?CF->Tl@iSDp)f(&95v=r1V8bG?~6}w^B=P%= zoHSLNJHJNB`UYAw%z zw7oF6`s)rj7t5{}=A3eAMA|}7O%VTJ#XRN{rqJn$%KHA{ht4(92)tkB+_kzV6qDy1 zD~K~25VfXJC9#?=Po1+52pN?&HG889w)P~N&=JK?LOXtp3yca73~1W=&b#qUoHwGs z2=Qemr#a0tcFj#N2?>tzfQtbyiT%;B`)mmFH}9{cYpQR4an1~kglJzu<$jEpGQ?Ik zz9hLf8Q-+Zc#RScF6*B%`gnP@^Vp!&a(dZ^ID;81*v0*(a!XAeQxbA;N8ZI0q_ya0 z*SwKC0@JH>RKgcDwB`nVOp9~6S4xAwIv7H8)X*IQiz6i)indMFU0XUj;_}&)?EjF| z|D?<%y;c6|@Eg_|{4JA<0x-#x<@tL&~79;6eB*YeHexU8x>|86=+3hejF|sFz5^X1*$p?j z={42k!@h4XYV+8P=!1AvE;ZUtB>4J< zx52Ze?u4bH?rGVpriTIuiN|T!nV?r6feQUeBttp%jVm^j3D%`3Ex|ohWa#2ETP~T? z#~OPC6@wBA!B?e9zMd$DuI=W^>Sn%E89!t1Bcr(&=L*d6iugk?q0;xrMf#niE)GFw z$1@$A3*R@yze3{3pEAJ{`cXf!d*B{2%e6?u4rAVAUt{&Px5 zo#;|EO-Yi4F1E80&En&~6R$?^8jY!RTNwt*!9U96Q;(Ts#GVtz5dD^|e2PT?hG}=` zn9-ePS;!XLwqpy&($emmd&M3M&?c!q{M$r=RXQqTjKfGC=aO{ogC05tQmisd70d3Q z9Ivimk2wt_JVWkJE1ogfz3y+{W{HGo;j0(BA~_hucJ-U`At##a+wa1nOnqY$6l(e! zjBl1Ly31%3URUs!JuuK z*o=twDiHb06}5i?4I~kQsm-<2hP#!7Cl}fwXQEeOF=C;*gokb=mY3{Eb)HQ7VXV?5 zs%UvkpW#XQoT!k$d~-t)C~o{fU06c6N0%g=AS0ZmP%Md}OVXi+hwt^|_apdG4_t;o zDK4>GJeeZzr3Kg*O(ES|P>qLsHKoJwWwHB%NT>3x6Ka z*8MfX51X!&cWmo+A4ih<$`+K)N+~6-VH-r%9m9r|qm|T|ZrZFWmy$-yghq>;aZsEN zAFhtrpY`||o-0z^gY<``)T)7jURXi7qF`JmK)(D}R%Uo~l%dhrz}DC-*!101b6SeLBRm zr^$EpII=3qPRBj-#hfjY#GBJ`V)pd6N?}{4YH4=`pMB2G^Ge=$egjPX&Cqk3*oELF z4;}_<^HO@P$1>uFUpLCx23vpjuQTG9@H(mF&TS8PgrzCs-n}cXomrgk&jNnYiTtDQ z9zx5_j@cWZZiVe1frzaOuyWn2;Fe}XwCOnwc7 z$qMb&pG#M1!$HL$qVXvGMRpFmlDkxupUAqF%Hxl{qPp>_I>(#TYMS#}MNU!SXxACL z;%tN=)waXT{^;xl01V6CJ)4=ceTn>t3$Y+xZ~}?(xQ_DdDA_ z=VXhtkv=}Nt%`e9z@ec5d*0;TunPrXi^9`H$bF=1RD6Xn8_51Mc`M=ar9$v5$TBsT zCDlyfp?x#8_H?I~Cv1;3)a1z0>z`#!2a)MKUYy+=4F6aH+*k?8@u~NjM8q$*XEkzp zCjY1!6H;g{#~F-;AF&WsEn&>o9h59o23%K5o%H|MPa)dJPjtbl>4c`~232%vOap;k z#5YmWRh828e&LEWNFTXQFo>AoC;w=A`LCrG(O9u>)=c=ZJ-`FDY-=Xs7@#qFe@AG_ zD3t#741-HUlG5jYib`(nJXrQ0U`n{C?~mm1P&LybOQ;KBQ)*+>b5(2}mWh|)jVnuBUVJ#{1?f4>Ez1gvydQr^gYss;K40%$8ZSbbp z-cZxR5@Ggl$XTEMYQ>OaKA6hcSt8jxFOOLh1KL4iG4apm2ctwP_06mtO5q=WzM-*D9V)s1-R`~V%i*6ZR?#t)54^B_2wd87~&M`g7V&Ns8bEs{#_KB@5 z>)}=3RtT4J>?{F*?jqWg^u+d2`cj8&N+?s6u5YX0CEW^N6b<5zAXC`*iW>xHR(myq zgsu#}@-*%xwH{mPGmf@iuX-p5C5NHP`Kq}hLIhy#A9ClUWiIBMhdom4QCxF>9B3u7 zYv!{Xj&O;~85#%$w}WX&rf!V{`{KFioXw}Z`Q5u}NhHHXU{_UX7m3XbW6R|LlLFgM z@>owFEho13@-{USwY@ijzv#(c%?HL;@%#i}^j})IdO73zFV2|FW#fKdI6CM9Q+I74 zrn4_Q*UmmV90p5Ha`P0@*4ug(ux3Nukb5~-WfmIMji^$CNdow|LqY}QZ?QGGTR)M^ zgmD$iF~15rf|Wc3PZhB}l1d5HA_>=G;5*H)Bvs^cFChpyS|75C1|b{Ne1m`b`Arm` zk89igHN1Thw3*^JJA6LTd)eSR==qN4Co6Yb6tT$bt+*t}WN!K4HzNs=c6$T8WSg1E zXKkK9FozO{cvo~u(JiJSC->_InA!w{T}mq+z+LI!kE|nRvS|@dS~NJ{bN#oVvuX=D zDq$=`=$`ioKzbb`T#*w=dDkE(WBb@DB2jz!$9}TSCu#uhGl$kxd@iQ;J4YYdUwJ5S zBYYQupOFW(g_s8DA#s@4%p0UQZGiVSB2^~#eEBF?89c8shg-9D78Yzylay9IC*&}>rPtKpmI zx{{0lhq&KPtr9g4Y;{tuU^&x?*z2sEoT9B}DZ{n18BnFn-o-XO42McnOzp!y=0wZ#oSPtzl8%u-m9o`OCYSTh^y=ap|mvmvs)L&x(?6AQz$p zE8KX0RMaL|8-YUSlg_V~$fF#uWZXGxl# zU$xD}lfz{g)nyBV^^`-f7hW0P!D+`hfIYb0g*x(U8pKd1q!hQZallN(JWyjS`4e<7 zB!kVIvT*0XPiqj}(o$MJT>ncm9ikVy+<}bDWweu-mVQ-1kK4jTgpe^Rr^49S`&+U# z+vm)9h3c;SH@Q(oae)&1&Ut}g`PFAJVgh_QAp`ifz)_!qW7sNWPN^DY5eAB^Kxr82 z*%irkQ({hX@n9%G9=AvO_0KK$t{y8_3fHKCyo8gCUp^;_EdY_b_EY`mb2_;JnMu+d zf~eZTClXvR)AMtm2S-aX%Ghl6x1m>*6XuO260xt@N;f@eNa@LlCBy_~Xr7U(h{L+u zGSWpz(z?YB{nD#E_|%@&9R+Cr>KpuO|212}_RzYybT)^4{%wS!v z5#XF%28F$9K#ElUK7Zfz+QmjL;&yxf1e9&3xv;XQ!l|=$`M_P7zRe~GY@yQ9lRGzoL1|MLZ`PzN#XSaH(5KvtQOT_1`6G zDDi+3O+ML=jnmGmOIw@$6Ro#oi!%SZDenvtLP=6EWJ;HDcRb_D5lMT!DI#hj#)`SC z=eS<`%3p!+4obBYK$1}IZk2{m~`m&LJVv^g)!x$Kxy7JFP zYU$6_!&=W;dShpk6w?`+Au)%AWfoZ3XbUDGps@R;lKJE9uuvP{92z2*TeP@nKxl+g z(1Ej+7KFnyvY$5$tjTMM^6Qh94H7|c+9|Z9_sbk!S23o_L67@){NQ7ZFf9vR$>A0o zy1R!Im_cO13V`GDcl607bjVP2OUlSAgS4s- z*FPn0l*Kw9hTp`*I7K9rgeyBv&-g2%-u?a%H^VDrt327l#-t{X=fU2KicCU@@I3Nj z))uB!Swtt?Mo*<6&q>1$GZ4!Rz{h1XbDRFtut_4dh7L0yLB|(M8ymyiBXiy9+)-cb zq!#(4&k-JtZN+KY5T^hJVn|w>){GZTvDg-1@BRqgN?@W!+LewOx@8&|(%zJMRHZDU zFzwE#3{|77`;gDdFWA|0N#8|VDI!mgQi%Nl&)hDz&o58ssx(#!uUlqROV_Bn(RlB3 z!|I-T^Ms{xr4(1I>=11Q!?2!;CmHD(4Jj#ibBN;PA6u4elFxUQAFE$x>QvUxHZYaA zt9Bw22HK&UoX<}q>(~K}uPtrS_cZwtRhW%={UFHihVdyLsdkd~D{#yTUK+hB8$qb= zgn_0VC^KRIfeeXypEGKDel>?f{Q;Gd%*+b*kCRzYa&@H>@=$&S_i^m=t`tfn)Y3j| z62l0_!?eZxE|n_b#4`u$B%x}$Mwhb3(2XKcH+B!1wwW7l1~t<9W=FLDF8~oh?!Pme zK>nXF^=YFplPfaK_%fPooay#0{t)MoEc!=Y%LS5UoW}~Pb(3jD;Y9i8csPFcatzjH z%?eOZ^+ihyHhLG5>#B63Wp|L9jhpj0EZ&RAjEc?aEYH<2(sccpYtAJ0%1U-zM6P5# zFH<-jx=wmR1!$pG3e+p;=1e=V5-w9Tt`f|j!i@Eb&BxB>in7%L`5SJbj7aR2OfjToh;InpwDy;9+WKmw~Gd|zt}d&oPz zFr7B*SH^l$kz})f0r<+8EM1>W{gOY|=6y$y-uaQq^MeYbtdvy!Je4^tbS%ehtzRuT z{W>fY8GkMM5F;H=RV!bi+Y>k)6KkepjV5OGr@c&-IDVg*GNR&>eWs6vs^$1fMa->> zedW9OT2^Bx@Tah%EjBBXRkvkfd4V3XI!bqVSK;&tv3!q!a(X#03Wp3+2~EI*kN{Rd zslTAvNoCbZP%-QZo7np_hqY2h&_9gxAydx1x;G}M#ixdu&wg;hxyjA(o~?RpmCVkV zry@XKXQ@V7$1fab81+9&rS=qWA14(54BWjA~MyDVAf(#mQ~Nui&1F0!`2N&x!5@tZ8mv4kv? z+8YkJYOu23jERLzuQi;4?J&>k(3L;0p^KpKnjd>Az!R9{p-$L4fFLeYn}5}@wOP1^ zZuM3T%xti~Sq)qc^_(=OTCiY_UQH3l^sa;2h!um^;zcZ_ji!0QT&PWo%XT-)I)~Pc zrlHp6?{|%?v!=gfs`7E*3#zRrj@F;?>Qpuhqyl0~+K8eycst5!qK+sSVw_ke+h6Up zn?1I3WP4nkL|}Lj{eF<3O(lllll;aleuwA?*)_2}Q*>nl*!mEDtm+I@YYxiqL<=p6 zuWG5;pa7eYam11AD^mEbbx9-2A7Xt}1ALTP8!Htf`Rihc{2TsYa5x-wWzg!5l#v*U z2<2dQR~$x%V6|G|lHuqVehAF7NLNjqrV<5`gEv}XvqUg2EJMD2|B_8pY&ONH+!U&q_&B9;pmq1O2Y>D&lY{oJFAsY;ItIN&5>+Qd}Px*{9G;o|2pF-kli9=U| zb@gsvcMu;^Lx`lBq^PLHwOV)_{0(eR;_N^AXG=oXPYg+v5F7!TO3*>7eyPc4U6Wc* zspdw28*^gsWCx_2N3QBqnv^;6KqQA`{6RMY9K zpy?b538vkNGD@QVBN<@)S292c005l-Kgj@Hq>J}|l7VUgXwhmkoRZsVYif}Rn#?G# ziL4H-m`>xs+8~{7wVh2zSDElqX{(yg7kM=^_upBUVwniMw_P4Ji*br^42oS#_TMVC zai8P=dBSAtuv%0rRhv1%4KMmV=+!>FS#NV#f| z5)Z3s3VO&3q2rwhBBh){UyKh9g<-$w8F!;UI^NGpzffm~M$=_+ki}t3lnkMvpbBMi zybRt$Z-1%#%*MWB7ST)Wr+1Sfef6G68;j;2!q88G;lqKQ|ww7jQzMHXTMf0L$-GxdpDPqN&m-xNv$-_PRYgxb%ec zxLEcomp&qr4R&1iKFlC1oQ8VlSl3!tgtPQaC4e)7%JtMyE6b?%#J@0Dbt}3&FYhA^ z){(~usy9cm+tnwEp2D;$os`cS`fg9p+aA_JIsC_!%}ZTv+XAz8KJPlna=z@Qk4tb3I&D0=?EO7~f7Wrm>inz!K7P%0;Lr8! z?!Ytlo8Lq61ZupAGeT-SLob9Sct_i*C5jb*QKxRMB+BrjP%HBw!wCjyT#qS76_ceY zU{e&oCQnySb_IA!RbAx}L^%I=$3*!4+Tm8=SxzR#3-go3UOOPM5?H4S*80^Rdn?=KIUP1VTR%`kB7wjffw^r=f$d+(=M%<}o? z(e7u8e?MzpOXMArwV2n`6$*S51iV@-3eduB^ZItV5a5iM!CG>*Ki_9lM0OA5H_&Lt zwtD^ZYe18nY>c*v341q*Bx7$!_!7(@FN#hLCvQri?4MGX9rzDv zl)cYn|^s%n>(JgrW^B4|9Tova_GKq0+eUHhsy%Yh0C=csmYD9 zkbf=4PBkVl{f50XZrf@|^{mtaG8Sos8Wo%&o+Pt|rOh848?YK6@@n0RsH)p{nf)16 z4B{pQIUSu5({G)^Modz_`!EU=?Cz~vw__JxiQ#l{5&Ki&Fn`GIS^dvHttelhihND8 z)CiXswmEw0y4Yu`-sJ!F9<;L!tTlRxuKF7f5i`)1@(Xj!J&=TZmm#@)Z_A#7JNG^6 zk}MfiSmE?tZ6Q-kBtgZnIppOWo%Hw47$$nwLz+&D_pWN0AWL>ST*!Mo1I*;zc{f>B zI(||;VA;-CwfK*m$RCF;V8DPiZ*?4IluwrekLH{vGuNmLn3M`=$E%IEp~>(lGF78j zDNAmtV?cs8<0}AD1#iv2vT&UuY-EjVQ5U^$%$ds+_(L9HX~XJ#fEwvQ@?BDSh;En( z;utlY7XB#!OZ?D5v+@3OLsAW3M2@D#O2NxW3y*j4HrNW5fiftjlM>yYcu{fM2|xwA zW=Bm<{PTN+%xB_Zo>wvsw<>0j%0(dq{8jZnUXBfvG@1k+75LrfjjEdJx`2F597y=q`PmvE__zW$ zj*9&0!9W3t!CQjv=n;w$WN~{nS$x+L#bapT_whxD_Z>uCIl#jp8OfY_q z_wtKC;x50JAfIuAunFu?^ihEN0>F_ZoP{Ht%`jYD1`Uk}Ao}8ioUAu53*gk3u(~r? z2o!PN8T`wQ(5=;*N5OP)H_V=oBP`q$uSDcEx`8%j=#fFBreCDCVPs%yB>SSEsuh{s zJ-hin5tZ;K1)L~?VFOjau#Z$!)9r6f45OX8q9HjPhpX-`Q<37GKJJDwJ~^+A{9*#S zVuBDc!S^wtRI%YAv5|(cDBD=PnArHP*hEBZvdn$#vo*!9+ZekRC#YXs4i0&GP8_r= zuHZgS8YezegvbmhzREBDKSd6^7w~tGZ0L$_rb=jy^=&guu&CBUL;Js^lIZ_CK+W|3 zlFI*wudi%~W82#7hQ=k0yE_DT0>RzggS!QHcXxMpcXzko5+DSZ;7$Tea?ZK;zA`g^ zpuW_qy&qi#4=VV}hr)vZ3e*=2M`HO~fS}C_M`H-tA%N)hMdOLc@5*%Cw_zuf8Pr+8 z03cu@lgsgBV_cco&-XW|B%oz8u8YHs4Iy7^q?t`sD6BAcn9NzI)M&H}zO3DrFA(tc z$B1mQ)+EfoSWpbahpDn)cgRoSX7DCA^Ku(=MK?+~jyHQCu#j^R-ZM?KkmpmIkStcSTBD zd_6BZC!azc_-3%m1w3DW&|sYrQY}Ku!Y(GLRNR?uo76!Kt6Db#Tkwb_V3)U*Ita{B zRW3je(%OBRNBLSKG-(>%c+PBb{QlHfL!lJi#(j|x4aK-#@0Z7v0|ZD$)@*S3VEy^R zFVhwcIA~P}jo-nheNs8aINr1JqK92J!=*rh_4V_*$MKaXMNdYtN5Wo;(oNAsje$Rx zz>6aXa+Bzj*P9&(h>KCixf^Yf9%gy%+`aSWIfu;-4MMutt9uiyv`5`eW_Y z(}?VbPO~Rj*~tP0zVW}Bw{AFy+`DfjbSgQo^#Xt4eMSY7vY?{+z29I8 zdPkNe(CWp~2lB(@>=PNm$~|HBg9j-j3XsmaTZp7+hStQqcim?(e0ZU5RArle&nopYbV5m(hQ_OcUeFUvpF zi`^F1Ie%$IB$6eNNk*6w_U$X>E`0=2jaB^cB0@2S0+xGiey@+07O zE|f>GqW0sb!wzw__9V~Et)G`eKS7w6`f#(H7wkTtux{EpIoc<^Xx=&B4x3%|5Tf#KnS}b`k$(`QCvLkDkF`RJg$Lpy8W=4f#C7{C1W`}H!yT{m z1cZtE(>NjjfOQ+dUa#X}@5)DZ9TR@2XY%4hEC9cSN(_kO~4?g(+#T!WE(l z&@0J!u;s2IknBk@8{38%Jgy@3yCN_K!#=QNTt(S_7<(rkI>OOm@8*eDD7=R~gx?cy zuhmsZ_%Js5@%bv&53z_CMsiFD_c|_2ric`4d`yhzIzBF{h@47tTuSRY0k0~OV6<71 ze;y_A-jRSBk9ANn<2tF#X_u;1X+lM;Cb1@}_`TV2k?J*NY*QCJgFD2e&NFIa_pg1r zcIqjqZtLj&_a)H@T#>a0JF`z_Usjp5@D*ccN9tv#S*lfE7#N*wiXGb?9R ztwkM&G|1%>*O45G97+zcwi`hWyIdtv1t6NgC3>ACGVCwr^1sofjyJPXpDvb2LZ6U_ zh<%P~m{dGFAMYXHNfPbkl{~*TZ#zdR&IMt_&|3}6;(uZ*S|gXrn>XRiP6HQHhZp9C zkz&S9)Dq(a(^!<4EiBZdkWwv`?@4utB*9^@0->Z_ki!y_x}AK)MI}Yvi7(~n)g;Tp zRWK%6F6D~RSIm=H%Jgfcwvn7xIV)RA+cgYeeq>J|E1s-oV_6XdlEL~ZlnGO_Fk2}x zC2UsBWLhd?7RP~tbes?QcchkWP+G|7cq9$!q?Qu^2Mrq*3QXJk6*H(qdHRxNYQe`i z+WwS^z8_$>OL)*dx0)JKoZp*rcdFOF4RDY)pN!>OCWFe??_1u{M0Tz=YZl@%`dol) z1#N?astGKm3;p%^P#)Grqwhl`>fwbqL{>K>?pwV(YLMdR=9fm((-Iv65 zvHVP`j*_0ggRqzTNk@ib6$}z#TT$3p*F4vV#9wUZjQ+$|_ac;lc=k#yar08+1z)vY zt)dYqAsEDe9k_KX<>ozZQ$}PK2`fMhp>p|BcYrw6k-)1Et@oqOLKJmnT&KFk{V~5m{0Lzam9jt^JzUzCHAJquuTJ)xhHLdm6tsLmPoj(x)2RjRt`|a-Nax@-O(f zHy^njzx3ea@j@C@g<&X(N}Eu`()qh+V_!!6+=(q&e1R!S9j1t^SZ1GHS2a$%gttgt z@>?l6B`@9qsEChlJ;V^XpGmF6rEw<2G7I$PW1M>c@#zX0ulTQRtb!W%;4r8&fKith z->XszjwmPc6Y2~AD-k0JZ@jS;dRGY0ToXWra74(8sKxGEryY|1Sr>Wy?-I}+6v)(l z5X7oMF`eQ&?NB4W(n-dWnCrm!00E%>)+HpbAIoxn=`>U`dS%)w>rW1OVSZS-vQ zce#^J!Z9}iZ=BN71~Dx!k3V4Qyx=|-dblGq-Ml&_d1OC@t@sW(5(z7P32^D7BQ^1Z zP|>C6mh#jBtS-cY$oMJLXZlH=en6scLx9W$DwJ)UJ&~w(9zVekN+B8y@}sWO9VdVO z_(?Utt>w&~vF+S{?p*v?d!Lr*rQ)a0?V?_0OJoq=9>qJ-T_U{ZkLqxP>MY*JM%fXN zf*w8bwGz-^L8C3Zfw$o_VxQ>bUf3OVKL7-HEfnNJ7d0{_w8U@tZt%1ShucpY z@5~GU%Lak%2LwCU-^+~N%6v6pWnb`PY4nY1!m~Jd_eq($d|&gdUkEPEPqf*JaMdc0 z+#i!Vz%dj9tRg&-0^ST^XIC;kSf=0(R%Hwm<{=5}2(&692|x?;pI|2}6m;Y}3LH=( z*HR7e1_iMj2HIf+$*cx}ZiQ}5bQhA8xzfPwD#3DT!CI?CV$>gX+r%Z>f(^Msl7}?( z+_=r!L_Nepa+e6%3*{{&!1iMyQLLd{m`VnZHWy-{qK2Vfzd>ahpZg<^2}@X4ORx|2 zzd+^NV^~}oW(aopYhqZETlm)yrOjKPwAJtr@c}s!yeUWFiDkjXN9Y>Z5fu_xg>4aX z_7OFY;A-s1CXL7zbN+I-$d0ziBEev$k;uL>A%2%Ilzj81qkvCPJP`R&&Dg%!_4eN& zycl*c+3dp=sO<+1?aEp#Uq?V0B@w7t(Ve88+vWmCf{ZJr@_DOX%`gtaC>W*vn0v4~ zb&s0Rimex;98mgg z8z4s<4_B@9q&XxS`;qrS6V0;}$ZbfdL~&=wNE@U1keU zIMD`7JP0t2h;W+dU{bIo8{a~TVVVCgBb+fq|4GX9j#Uh`c)wIQS+cfto;02ekyv#{ zf4zOW=A*Rsh4jSmr2e)HP7=Z>5*`bV^oH?_Ce6$iZU(WofX=X>9?Xnaqrg_2tUmV0 z0nMxt_pGt>tcmulskN+`r>rkH+4J1li|H8Lg&9klS?y`r+1Q!c!P)g`MnP{HnJbbE z8{-*cZD>P>PmC3RJio&_Z$u&R!=3UuVI6pft7-F8mTj zG4gwi5D)``gbhE`rG}c0J|ltA9t)A!19u!%1(-Kc6vs7z+TrpJ=?j2YGCay4Z*&MN zm=Y{aoWR$CS~40KKc)qqNM`hBjvD}prHbECt3Y@F#CRMp{EFP?F}BA;AuIthYnULZ za8S8}@{>2iapBLl5|+eZx|0GQcywKyQgx5miHAah@%%o@vZ%HqSj1#{c*{8C)B;>| z@V|n3zX=jPAR2%KOa%N*kb*92^ZI=u(P(wJ>c;6kVR59gB>oU2Y*x#SvA=_QYEje; zzX=kxa)r+L^-gsP6HG$XAJsjW)#O0s>9pjJ>K?E`(^NM1cTlgXe4$jKP%8V+hN(XU zX>-Dvwys8hFoHn3rRpEmy=8Z$<@4@ug4FsasCRSnC#d(WS1leuDAV>wb;m(9k#w@( z8RS#t@sVk7*q=;ewc46$Z#bM;pD^afeG~LmBI-RIDnIqfTVpCiXJnWh7 zY`yr_7idI5-^I1EnHP@t5arQ+bLbpTm5ki|m+Efy%}0sbb!fUba)D?ziRbC&XrWSX zwx|1#>OPq(-`n%@Q)7CUzvqwYesv79?FYg%-}Z;4EZYu1H2E{s+h58GAZIZsoG{>ZXq z6erd#Y^*Fy=#B}-VJ_cGRDeFzO@ebzQ*(nbklatvaXH>k)eoULNHb2cI7l}yDL=@t zZaF^4{5(Q)nB}x=ahQF=9&(uDb$@)A>jzDHloy0)d6XYUsg;t<1r7S|MEGCs3JL%W zn@1cM06@ll9|nLW)(4YGf$l$(GKGR!OhI}74rH>B`$eEBRzw|v^Cwf85K^*!zh~J* zX2%ZRzQ)p7NO7NQ-T{3-abi;7uf~U2y@gV_LZKHkX2ZG)6rjZ~82{7hTK!&#PRsZo zBF13gi}qwo_16~s!HH~b<>RlNw!tS$lY1$Aun$tCj}Hj7zu^EGY5rszbC4b(8kLw~ zn}g118qU{hFW1_`9HcGh{gWkfvo4f5AvP;DQkLp;quKYHbCyHKt;(fpp3i6dy*U2D zV1~}Nk!hI~O1+8YW360h4>A5rea3k0S3v59j+>3^fOT)-TarvXh5hc4sG+U}%GmX# zYcHLwJU`8^)^A@=kZ6Jb zl7bftCOO)*>56v42@y5LG%}fSre1_xa2FHtQ}AJQBSUaxT=N#+;L(p)v)Zoe>^lVT2)C7VLYi2JXeEUTq_ zmpak)!M#rej+r)9L_?x}r4}*IPnHl1gF-8+OFIiY{v7@TmRx(B*>crnmFNCMTU97@ zA96yibj5j%VjacR8iTH+=3Vv^*b@r~qbc^5@^dLo5>!A*{0kzU%*}aV;*#G3zG}_e zDmvuQaTGUx4)XN{l$SgGdh*#(B|TV;DW0??Zz#$hH;T{1VbVg)EgCV^@BOTEP5r!q zyMUM`JXArL9T4wSLPvo%RQa1X!>HARw%D*^)eKO*?eaC4H0j?WwR?HxXgkebpyiy@Q_}#T*h*1)wXh zZOT^U4f0tjkHFfeUrB>+Xu2GJH(1f747ezk<7Q71DG3|O7J0S3(JK*#lJJtywDYy6Q-%Nn z1q8?iNNy*`0oqvoX#jtq3@kb{USALr?Rj9NRR9D;URhKe6HSy9iG_-J>+Q>aFhdkc zN_}EJrQ^1kDY{(7Oll!x{kDW_ zx?Ik+f;VAeq?8w_LO~UNF(3D?jDVh2DRpA8gy*hYA^Jq37=9r_>8?U!xmqjRPKa zN8JHyw&G5z1CEuB*?nUkxDScj4NIM5aOa|s>oL6x^;LbC^S*<;}8!^_6xTyWh@bSd?jdy(N_DNggN#!$t?Hhso@)+Z2Zh*iHunz zA<1-a?it`D#TSlAw0yE_3*pHs+6W)IbHZ$lq<+PT6dyue4%zYmA!=m@1opZM3rlsy zA0Jj*8jY-xCYfE^EOac7)9u8j-j~>?3&b3^oxRspd;u3&)R08mK^?l#iqcxeOC?v_ z7Fka?v=yn@c+9`V`|&rfbAC@9m`nsQ+KdjMMsb?HJ|(x=milAEhydAV1Y)GSiog{U zs$$5$&%go(E`&M#&)sa+(FX+k&YUw?gJ%x;xmvB~$qlB5ZS<8cd;a|#=7 z60btJV=sQ-&7I}!aX_Y*_#)QerM`It6^q(vmtb5ry7JZ0a%*N+_&%pH;nU-I{lc{5 zW#%QmA%^h&SAGU1O~npcq)~^3*^fW6v_&Lqju*=x2tWRs;VY@h(aQZ<{P0Cl#>q>* zbMHH{GVII#m(2{9-Pu+b?4|b_$tz16*D+~bodiqqu1bgFG?llKop+hN876Aqh{Xc8 zE(pYgUT;kb>UYhIuk*dVzu59Lhi=iPzOnz1mv?K|<5^({I8Foij}B*UQ9AOshW$KO z(j*bG%@k64(}J#jUbNDDo>E4;A>jAC;m?h3SpWU=Azi{x-r6#iDd(AdCWxmHp(YR; zL-U>slogd2G(qEr#Vwc%Kq|0Q!ba6SREJhIQClUJeLWT@~*Y;El z;=ck&j>&76D*4TeAb$Z>pau60*&Obx3Nds1A&A?6;Ok)1Hg=)!08u6&_%ni285TQB znEM+UIutQS?-u~l_D6|T+5vSU0SbJ(Ri)7ZK|n!R1)8m>P%wpHa7mhStRq5i9moeZ zY}5fzwG$Tfh{jIJ0j&=zOp9ou#*5<$O@`&=p9^_h45!{Fd5w?CH&8PWC6Hgm@?DOk zW(hH&j)q_j8CsRgT2;kw;=(E7`nDQPkRo$OFf|j749IBVq;YUm+a;*y=9po?i68p`5Pk_bZ`U8@ z>JwRyS$WG72hha1B#9Oa1ERT-?hL}q(YOcZ6Rd;%2_2H!lr-HWv~u&4c$wKGB$Hkg zlcHebwBax<+=5u)xT{H-mJX8SDbUfuQjQuN2I+*cXrkVap`JieA?#FxD>Oj^5bh(q z22HXD4xXqv=p7ha;P^uvWJ++W=IDG%YaIxB8&udPLq($+I~b^PiEu$2#U4%&cpS-3 z!cl~i7KM|6WgZm0Ac(vTuyg~g_-E7ulXFSZ8pg4W#%SH$Gg$?rOUE;buej9P(SGkpMHQUjYxT2N-Luze0JVM=rNHb{VG{DUMmux}Ct(`B z0ed^;e7jRg_w9046Ll}s(zsMf-|2^(QfQvD&|Cw1(@7NH3gawp4D%HiKd5p|seXr2 zVBW8G-`<8?qz<2V87AhK3ApXl9B5uB5hfQvZx_>ibU3G9t@+3kIB#$TjIEkt(7x0e zgm-YNa6AR%6YLM%g)(N?@FzJO&z}il(Wnjmb3jGB$=+5}dC_;m#D(4;EYUu$7Ri=# z!-lFdTFR)xsY>7mN@o=B4+L%O%PCeA>*JOnc?5rYj@J+^Ia4W)W=k+!FKJjRu*EHP z;3;*|Ds}NFWi=^6Ei83gFZFpY^}{U-;3*5zDyzHA4ACmn!BsMN4Qec*kjFIm?^(Guh#@G{(_#EK(+b z3~eWPaZ=jFz{+=tHXReW?d}L_;>3s(nK}K5dN`E^YZT{&6$5D$gTnbX@r4pDncdH5 zI3ygkd6ARwgaEYMQIAZacoZBJKpi!Dl@tl#x?_!$wh$1``olwUOuB;^l@da0X&(jvmUU(AGAk>G9DB% zt&~Z2X){GvK^Ov{WQ~*v_weC$2bX|q-0zirw7ER+IfFRs4G10RkH?i+Fxk&1BtYq^ zo%BX%ZTWxOrjY-#O=pSr`^84f`pK^}2guFE#4mjj?ItX9o3i^oNBN_5{mR`k1*YoG6Gjs%*~lZRS}ha zKOoovMuUahW_%%Bc({Bi;*5sYGRRzx1I0UTtKgd zSI~I^XkEd1f=E0^S!@oID8$zv#RycFn=6;{fJGAV$%eodw9=pMO;)vEmrqr^t+zZ? zKOHC?S17pzl+_64;ylf|!@K_YTKG_aZsGV66I9%xi(7)xjOS=fQ3Zw;j-NuK=u&JT z7@R_>AR+1#X&>&#Lk0Vj2Z4xZDcQE?B#8&uh~bz-78a45y;Q)EARz2AB)|a;6>=jf zXjGB84-{<5*(b7Y=6lK!3=g_g04Oo;#Vx`D;sF3yg||glan3UeE{~vhAGtiGD9&ql zr2M7scZ85^n%!W@2`=brB`vANGb@4PE*`yqYS<9XPneam~kN_N7K2~wKee$AXM*hk--!LM=9b%hJfB5x^y4aD)ga<#`=*xE}2n(<-v+e44`I!X-K@};N>4v zFmf*ZV19%lXi5y?u3v_5O&KVgYTyERf z^v09@Wt;BA#~60y5&4b{M*x&U7$Xcwu(-#H@f+gfhSbQZjyuF#6ylP22`O2}C8Lk( z6Y~&@sZYcDWMk$Nqkinrdh_Fq(y=A0d$ zap}0p?h87kPIv2Zo^T8Ub{&=l9lw3_M3svkX)54f;Fj@~a)#2sfDpWF{O)SVnE5<2H61jQZn~rWpiADya|L52G;#YF(HL0Hq`_|whH-DsU?Dx1Zxr59HkHKgym|g zikdGKYWv5k$P8j-LBI4hRmauH`R)>~gbKAPvOK`2-@-n&T}IMl$-)^C=$I29yw-n-^w{#&jdG-PoKyXzj+b{vS{00qwt7_<>2)1J8qAL+l>!V zc*>hbZzIpEC@$(WZyV6`Amy}YAv{hFVj{Dc&#STNQsF#N%H@h+!xN~W|NY5i%;%NK z?mq)OQ8X-g($JP0YFwvJZ6?;TR~Ek2ddgalbhUR6b1YPR;QC?wA(MR4hhuH){h4(! z4+@gCOY>Xmsf}kRr`fBEleHR(!)#UVrjHrvh*&&lr}l=W*mU)GSvym^I>%RBv>i$QNA5o2q@1uU>0nM=KHj_{fw4PmmbX!YH}x*thh<}sgur!<@&|$^`-!i zuaLk-JS4QoN*kZ=qtr^}c=qe*fZSaL5k9&4!?PCk*Z7*qZW^x|JP~BS zik8x$GnlaPHD;Crix)CRO=dsi)f~!h`H>6nhc$kbEwzM|ZffV_bg8P{o4_+H@mlw3 zud;#0B#n|~Y|0kTqAa}O(x9tWUD}^R-fjhZ!#C6<1N3Ph=jO4X?qC2f8+1w!V+c^^ zi;8?Y_L-6;m2Pya_Ha_v0)>Az`Yc^N&gsf0mFt)Rd)>b*opT0k9SwtFpQmNDcWfbgLv@f3o==pNPzV|JHO09qLVkoAem3OtufwVss% zdMMEA#Qk|p)TcPneM{XCFMY_$bdbY5XsMMyC%8*DX$>>GJD>zS4thKfnX&bAh`DiC z1qN9ba-e|$6p89}Y$z5eR(FX)^kJlf0o6mn>B34_o zXd^S;kT31OY%C(**ROx}P1JA>hL|`=@AAjPzP>z9w9qZI zg5y^QrT((9q%CbG)vNT?AjKP5vpKzyWRLb1Rq;q7__qGPKLv6Dy1WXSh z-wmPCSUFs*SC=k1$jYLeD-k=AWL3*_ww!gAtd*SOE+5E#W5(I|;Ev;HS{$LP?&Q_k zesf3@iTCZ-3L|hnsMJ$+1|9bHVtdf;E$L%-$2aH)pX37%`Of)>asT1i)yH!0tH&E< z%5Q2ny>;)r4zN(p#!b_0)0*!UNMrc+;sP*{O1}l6?~GuI=@q*uL&}z^Cu$6or*5h@ z)Uzjfp-e+z=uD{|?L;cYs~hUEbt^41`Z9l(wuTNFA~Kx>-bC_$h63)!OVo((L|M-n z_a$QQr{yG&rgJ#GQ>D;O{{? zgP>$TbbYt%HMv=O-rJ)1b4kIx1?~c9oDHQ~M*ug4=~@eXZ{JrDFnrXHp0{^duP*dH zw*)OTYGp*kM)af~7SeJ}`aLU1q`{9EIM%?zx(c3VB@6I#C4Gp}o}`ED2cY32ZvxQ7 zj4GX4WU0Xd^4L(sO7do^HhD`LW^60V;Lm5yE8xo?H>(=)yehF58;z$m-7V9ne8)&s z&0@5HL!ZUfj8qsUZWG?lICb44o07d_qnkd*h9%w>XPGTUqr~b77w?G0yxDL|YO)d@ ziUkLrw)cKUc72fWG*ju=Of4ncI;vlVyIHf^m-~76*_VgyP{!BCy;Qs3fVH&t_hVWw zUVmPWGQRz~U9o$6c|5Frd;Rrr@%9El%mctk^g-ZW`oYTl16VKp!BKf2Dv5r0t;+!H zsXS=TzW{3>F=GC2z$$b_TpKj?fdpQCYyh+4GMFhUAJI%=Fs`gFB(;+S*>!A?@cA-S z;IEt?Ha{WCpPXL-Dzo_z1<&6(zfy@|TCJ-Hjj006*0H~Heu#zGqY@*3a(?DAg}5t6 zq67_B(Vr2w?|9~iF58)765D?R*09k5=NY}^!L5hSoP0dCH69*1ggL|wV;}j*VdYT1 zKHv_Fg=@^mr-Go#{h^zf!|jw%zpn@<>?U3N&l~vzq=w5lKE~#cku;L#(vhKpjLJ`_ z8u0fGw*w0p{NE`w8XFI{Ks2jy?5EWwDUf=WbVjkUB+8T*=d%o9>L6cuHj-03sgXgl z2>?3hkMk+#*HV^MPveguBZjSd@y4TFVy6aLYhBhO&Ta?xo;@@#K^`g)BJI4+{Wyul zDk>X$_?H@yh_rLBO@?e1EYq`wf|hU^pA^)v3^^xm$_8bHu=3{1cm!+-0THPegD-x| zVzC1^pG+Q~Xa#-q8af?~y68T5|%#Mu& zRS$9!a99HfU_3#U^Uqw~Rz#wWwN{#dzxe6x=I7xVU_YH+sPN(??fyH$YeADzBP!1fK!kHy-?9^0CBZ%fo zbQz)dejDrJvp_4)px;-Qcz!D+GpQarG#GS!h;G_|EU0nO4>uUI)ZFQiZczRZupMK} zIE^@gGSeiPMcVjU0<{@T7UY%9OltSAWwuw56oBUiD!Qi1O3*1URy&ZRBNP|~I7ypo zkJ3?Nlj=|fAfV>~Ei4;k(gcoWlh=HMl+4*zqD+j`wG~3n^%{&eQTU`kh=1SOX=~p+ zGfK4Ih}>fJ6S+Bi-;ipOL+&aTI@!KKIwtlXKdZaB|Zwm&3)h^0^cWKx)VX)95RRNfqYO}qSSB%a0tw|2%} zqJRBZUdMXzkWHqS(B9SPaBR{@WxYi`I}AduQ>ux<34YEy)KbJh*)lII7v?rTT8PjT+u`{#W2dPgJAuc*hq=gvtIonRIY;JK&P z`SdcJBi<8SJo!9rR=r_luBD|z%`Ui#wrha>_2;W9M&alvMZV(=}0Z5m0_$FFJ ztsa{RgueIix%Ya6p*ZP>Zfpiu%MXaH83 zhg<^lA9)*WW>6pCj7n2V;aRdmk1Gb{01H6$zh}x6UEt>l6(Hd6-@d~^LmhOE5)igR z|22ROc^klljaFFi$5!OqGKY6iAE1?NDs>Pbw`VN-=&v{Gw`mrgRsF7Uv0(s96#55#cI(6vxFfyb%uRr@}b9jBqnRXmG=-*NAi$jFgAs^<0E+ON;E4 zz-q;gQu`h`WR5ZD7NwjYIl+Z7z8Xc_AC=r6B554e1XML8kN%Q|I@=aal^;DrLgin? zm!wX&&qcjehANBxo;@W7AQHU;8GFtZd#e$9?-u)*7W>>5`)f7!^)VKJ69>T^2hxm# zc8`NikArWILtKkPeu@L*#G`V@qie=vy2tZ`#y3&LR=NsX{iiJb8}HKn#=F3O@b15R zEhGSqvATTWP#99FEQxs1x}g7tcezpfdjG_>WHjq3L?K~_$;tjMOFN!SnO~<$<$hGp zagtQonfWWWB{g64rIea#po%M*^sipa@bRslLajn5CPhc{pV$_Z3sR?pwR9Rdp!iR0 zYr5H5v%|*XPi%`<@fnTB%|x{vwNedgs;yqHi>kNe*Ga-ax3;Vf07*c$zt7gx+WWn( zH=V6#=UU)WTDEef+%xxuKN%W5Yul8_&B@m`D;Ijie&-#?2V$mn^>vr?-7#*Lm-P=n z{0ECjHppJOc82#>8?mf6f0g^gsHQW$X^>`5xDKr^SlLZ47yG-`MqsPU6LJ2mQznvd zR_NoKf{)ircp;-3aD6cu#AP6snYekqm zVN3uGDQ`>!lCEV;AQB0^zaWe)$0T0?V#8sLzO`$xFh-!qF*bH+$$T$fg3^6~zngu0 zFHw$zCXOGhAi1BUYHA^$BIQz!o{0V85SR8Tr5rcS^iOQdvfy|w1tR~LGSkUcloI5+ zrFobmGd=z%w&jtTON1g-mKXM?EFHyha#R>6wmz4ipk{eooMu{aT$1H-a$K4hLVHqH z#87NjURF|ZQc=}%a#C40@rnwHQZf3S|r*3AYn@?{9gdwvJM#UPVn#QH6&TbbJ z#Od#PWfOJo7B$SO@0N93y_2#y0?+PNjZyfj9OGI3SUlvZy&xE}_Wfi?qjnttXU+UxUMR!! zQSztm9EI3W_Kt%*?VjfzR|YB-e=6X|_Or@w$R2$;rAZ%7dvCVBALi}Wc%Bcui=98I zfeeJb&YpeUQ_e>~Mnu)dCGxnYbEZ-`S6NSlz4f5$^}1U585qR>%KPSVE9BlgfAj+q z=+7q+7i;$eI>24YuIVStIe?RbPSwu!@-*i}1O`@Q^_}l?q5K5zFyYfvjKBX^rOR8B zx|$+63Zky<696#ms#{*L1QPv?bb%jp=ACyh8o@&n)*$xGXE*i@{Uy`<~c1k8)kVR~C;11B;G>`{@CVS*rA z^A0?rh!keFeG4TM&?+`_!AL`XiJ^5BS;{Ale%5CcTR%kOdL`;HA>0JjAl@QUtS6eE zlG5%`ztl+OIV59)ixLI~jPd+S#tV;_wX?f4sOdQ-B9c+5tcR9qFj}OT9Ili3ttbZd z$7#gHu9eO4C2(P`$F)z$ll$Rg7(*qe#Ks%`5ol!+fQ{*2qDt6GC8sU5{uXFk$ER&Q z{uAdpu#9m=lN12_$DR6p5CZW3$uh#gf&c&s!2nn|T$L~|>>si(l>`g=0w9L?_u&V{ z0*Jk*xW`QFFT%4RqfD3@aT zLqtm^@B0H~Yza0emZ+yH{$y5970p)~#at_Nt&|Hjo6K@2g2Re7{mPAROszj(U~Sas z-N6&eXdHZNv_|_j^z;65uH7L}v|HQFTCL{c6}o1sy>Y@fB$5)t#_9G;lO++hMrVhO z`KV8%5H$1{4Y(_(l4a}{)R zl-P-`g4yCKid@YuenF@^i(3L?I7_;U;~1*iB4Na=i^9#xuqfbrYFJb+-23HZyMc0Q zBN;njt3l{n*^&U-*?y9qE0QhLp5{ChT0Gw?fPKZ7)zk?CQj4IOM2CrabuXD{jg!Lg zL$@P>=TtZ5H<9LMPAoAjLvBE%sD(&z#^;nJRY&tPVm_zr4XBb!Nm#hicSBg~c}m%gvpAzSfQF-5ft!isKrw_;gsXBp@M7Ian#A*u|F+ zqt0ueMg&oHb(s4i_}7IZw(%@su(}UVbR9 zAWAW`i_uu>b{{0a8S0*^e?&q#&i1tH5QTV&;q{+*Abb%E!F^U1n+Z%R5x>dh=QhU3 zHSf{5`zo*ch6l`hLr3qK3$OKqGM)$Uw{$|GhRXCeV3+&#@xf>*kX<>BKXc7Dzf4m* z<&Y&Gib?b&Yu5RH6a20*KQs^nQ1Gi!R)+PzN+g%uQ$|OeL@+BFKzL@NfE5voc)Nu{ z6d?wyQ2xqCO+88o!-n&DiX6L2eFy9NV~{kh9{a@iDYV$T#WB9M@wZu@l+(?<}KknN+{y zlJ7`Um4R%fRw(OoTC;QY=jckEL#Y*tp5_|2>B>*39_leo&9xg-ho4}iR|jyHt3!2( zwa_M4M>62+Vxp^zsiZYc1Y1&5O^yw9kH)6e?i=&El>~*Q*XJ`_stdiVxQmY!aZkSF zf8edQHv3sW^@beR+>~5yd7;YRNsTnd@Sv~r3AoG)G1o&4 z<;;sdG@27rJ&z{8UKKBB%}P2_E_~d&jS;lz6}wqjba%gn;;U9sI@MU@ak(>4d}%d> zPG9!^L|vJSvw-YUR2|h+$JL*HYN)*IWiixTpE~^`leTm1OH51C&-POb?aUG`9~W;M z?=xFZ&$6Z0mbOvZbBD~I8%J`j9V^r4E}cI&k7iuNUU@e^5B}VGG%f9R&a|_$>ipIP z-PR}Cb&>Y!xh-mICGam{|F63qi6f{?^)c(~^xi^|__`9D{{k8S; z?nJknOU-5t;QZuHtJO@;-}eFjPd{b2F9eN9q>77VcO;HX8k6VdYHu=aP;2I&eo7>v z>~r4yeA%CVik9Q)N}bVP{gkt>ZH@~Y)qnR>5QyZu+OBrT66y42y4r6JzZ6R6%KfXK zGSl7paIrlY`9>t))AfJrr~JJY%J+Vn9(eTazw}dV;Ku%Jg`)l*gsx`(Js8`x?0bj< zx5U~ffA=vG-NM-`kUxv;xx>- zxXjW&s7O~gUu>ZDF{-z>T-9?CAk z=+RNay(oI~n6T@NTgIao$*|?Q?60GjA@xCKeQ)oIBLAD{)T+wlk6^nZ5{wdtzIP}l zOSN_=KV(BbLw))-CxIYfo|0rVp9`uf;d`1`Srz|DyO6s-{#+|E^H+ly& z018L{JBS6=?j(H*VzDL|{k73=Gu-h_rVWBg-FDYPz`^wEq0~vgHadNN;d*$l@vn`} z-bcR?$u(oO5yiV*NaezMyt@$%foIr^k-)cpV)T2T;)d)7Hse*K7`75Lw5+!hbu5bh z3SueA8l;#F?U?<05R2?Xs@~rj3iua8K?NfIHMQt5{Y1g=zZSF50RRa2&&4Dl3diNc z@~e3Fw3w;qYa%g!E~a7(Kxf~K56izJ0|MP5Jw4C=y_k8a&v(DZiDXo}G}S&WrX9X( z%A|pIhJmVBBEReFnWt18+39PesdpV}eUf|x_Cs;d>^MnT*Ro%W8P&ebq(up63%rN_ zwU``zU4c|THo~zThrIdm1Xf0fPPGQS&#FsHs*lIN7OLOFy>hvH70RaDfm*CvceE(0 zi-X3PMrW7Z_#UO9$oodUCz*{z`71$V+uC@J=|ib{ruwGK$`0Cehdq(``V<){IDpLgIE}J2QlyKS})+j#kLaU?Q!g*+>!9tC!)t`P{1_ z@Ye&R-XI{10=N@VHaDgdw1k8BS`L- z$8ScJ#(=N?VmYg> z5ni^ z(jjQv-unm?D$HN`yBFm|##i}UFN*TRAa0PQs{YW)aak@X2yM!lu0!8D0F&axr)<1{ z=;sX)cgN3NeOLXn+ua`Dw@-3pyV`okUvgxzb@X3yB#OYx8yBt}?q4|*gH#yRhx>}gK`7FvIGKX|W zrk{V|?1RV1KWazZZ8s!fkz^t;tN%;wNXkeqEzjVflGpif|D3e!klIu2h(aFY4PBdN z^Lb+BQ|$;*M#RKWKPSkw%5ZMtE;)}ohbHCG8k@;M2I6JNNF4EY~ zX#V|qChg**MFJ)N>>vKQ(|@L|Zd;>Xd?y5A{vFFv|H5+c{{_ntJx2U*TyY&A{=DMQ z;>G@lD~`m4Qa4nsAYbS&ESKRu#r(Id@!zkw<0iL%yW-qz-u&-eaXZVpS<8EWV!7eI zl_xA0J#P4Ae{)LzMMb;s>ubC+WwMF{}m7E(x2^bwk3-V|Dyh65UJ zO+g9zSAMk|ksAHuTe%+PCgSVNH zs%}LlAB=*DOk$n|59FoNzb%T5yNp6oz<%7c-g$4Xm74w8(I9nOoWc#h7jQktmmAJf zuao=ny#*h2piBzSUVu60kHXAYPFRx=e;0TeN|Tk{-KZGkUh4u7UzlaKw-i)ouC*byQ^=1K3X4L%^HQU zT>trKeW7vw+9m(!g@j$hXwuS~*|Kq8x1F*EF@YGy(7wLp0!AE zv{gVf)#a_9+Owz2iZrunJK3_mX)EP97}A@oU3j@wDB{Pz3t#y42T&nqmn|Yp`~L2z z@%sv(_IxGW<+$^^66^_WQepO)^o(%a8a~G}J-u{q^48-T7_aOrt;t$msm0I1_gfzk zlzUkoX)9b#CwJPt@7kaJ??83mowlcJtXUQZ2@53{9H2`}Vn=t=sAwEmikdqj$pzoz0y9;>_Ou=w4mXX`XE`}_ZKD*pFJ>t9oG zSD}DkmwUwZ*HkpHru#J&m5T>(SpS}~fc*ve>4D&XnTH+d|NlV#e|9fC&BG3IcFf>Q9C18Vu#T{{s2pznOdxyxui+5O~xfVHbOX{1G&P zn{f*0Qf9v(f7pVR+Aqj2I&Pkt9jEU?K@YpW;kM0E$N&>?eC62U$9A?qt z%-m3^fxN7Q>jZ|Bbglh8w+t^L#^{_Ck3C2MvBrKuY5jh&Q_Gio{_*09a4RyY+R2oC zvFaCyCgpYEO;-OQ;yuFK)*>k_=GJkVloW@y@aNizeph2UZWMD%U&~l)PAP{RELgIN zAJxkrw_f7Ib&B2)C%h58=VT~qy{eu)0EnWavbO@seb~ERtPoeG_XxlI+~d+7vC&Oy z97&p7qE{N%_fwVmY=AoXrA2SCzb&UCqkdp@1JlKflMJOg*TvfwFVcxoj++szF(MG6 zOSUjU)c0{Ix}Q74niRgr^*VfIm(w^Lqe;^m&x7hJmd(+~lV0iCeV@taHmjVqB9L}l zutjLD4}H_Xa`k;{Fm5r^WDRV=b2mc?a*^RJ$z_d0lL?>{Yev zmw|zLh`WemqXA*|_t?_ABKUP4;0*9~r;El4S&I68;e(UWX2p|ryHGPf{f=Mf!2;Q1 zr@!Pqi4sk7WA#?AfpnZurRT*!iWLE+*!7s9SZyI5%%E3O-9Z~a30GEx+KzNeX8=p> z4#tggTg7`6<>HHFmL9iJR*3_Add!}l9Qg|CNk7+S|3tQ%Q76ukx+%J9DEc#@8X1uB zvvJ?dyS@H#=ie87AO*dJjXSF zPbEmX4;D7UPjJ%&)6aH+fl|gN+|L}9ok9f$0?6DiF3B%0E)-h8~$`y;3mN%I*ew;gtE222)0wCgF5}FefBZnLZVOP45vn?Ct z)hEFjdc00|7^6sDA{d*4$zT~jWzab&Ev}N<8F`-Nb2&CA{K9XV;ex4p+}pT}eHaDw zWha0@yv=!r;d7K*H)V@k#e~&;3+Ys?+imSUr)S%XS-d@H+x1LDQ~^)-d_czHnsamp zCMV{aZ^C!0lAkAaRQIJ<5O?h_&)!L(z+UdmG19BNAw^G(>G?4A(xpDJN3ls-+f&GC z3mNn__3$+nIx44b8zL&DMp6d9bc`j-U3+bf(EZWCaH6X?=q|oiyc?qH?|!yLaIY`_exckKJ`E7-Sb>v^{JxlCm80RRS+x4rL?QC7cO z^z+p3mYK|prF>F)A4&|f0g1Ezh@_d)g;Q?tr}vuTB{bBcX^^Puos0%r7H5Nac;=~} z@RCtIZet4F`D=GeenlM5+je@nGd;W4QFu3rlJ~X~5r&&4LjbzoGLIMs6r9;L+>kGU z3fqz7L{tZdUy6l)#A<3gD`I|%FBW^OipE$=pnMJ&s$v};%PNLPF5omES3YE8YnDYT z`=W&%URp1mOG9<#TcMi8Rzf~AO%ttW8yhw|tq_+YfYTUb*{RUB2_ zv>J(km^^;_2SKrqs#kBrup~!WnZ8EmD+Uvh(_;6gjUz9YuOq|Dx-lF^QBg&Qzye{n1e|G*jPv^@pw#KI1)k zRlfF;*5Kwxrjdk1g%9@(U=nw(-D9gVs4NT}CxWn(vz{u@*H$w|6tkDOwsaun|$Nzji>ICKI!KUaEw}%EODNn(k}=N0v@?;dE*=`$8xQbmy0j@p)g@T z<|i}nlOcI8r+=@p*+#-kNd~2L#{a-~v=Q~O4K#3!fGZsab|fbIG1<;6hxY>EQvAZg z$6tZoD~6+>+Xj(>D}cD2z=;g@4&u-<(u+EwYF)oD-c}rp6Q-^UQ%7AYmk2{||nFawF_w!mmm}G4hAuGJ+}yV4V|mMb5%R^bV?6q$hk~q~!qR z4_jN5W^^2Ev7PXh-rKPoZANa0GbPsJQ6%Mte4&!N36)l){{{ooX}zY!S=#Uef2tGD z<_=W?aPBqBwgOk-9y{>uM2#SJ)=EAvh9KV$kWA{DbHEqAP{HGPsWF^_$%1w`HxVzx z!GoYx?S{aHXY5wZ68G6OfLp{lVnoV@t1^tVD9QXASuE#btZn)S8a&_*U|JE=dUm07 zObHP#t3P(EzV01*4u;P_8P*d0aRR#8z%bRzg>0h|Exan>ei7P#()ivlBW{OQ;(toZu)xp5pt&2kxEG z<`YALhh;2`rk9UB?Bv10AZjuR>s{&!ws0rmy_+Ci_NO z-;J)#l@1LI@cv?(N(~hAtfU+C9b{RzH@p!tiynkVV*67Jj z|6U1b#Etk~@$98y{d=k_@9tR7?I7~7dK?KAwqiW3J0Rd__4(He<2e%&5SXn243^Y+ zUtcNxItPcl8}KIIoIf@oOfqQTc`$92{LMH3&B4Nvh6{ihy!8AvhZP0L!MwsR_&}0g z)k+I6j(h`W0>varwV+ghggCtF)^?{iFx#a%;0zG}8fkNcntur{*C2c8fuS@J-~#9A zaVvS$^^|dcgKtbJ`u>zgTU{qeDll|D?tOM+a2~xRUtw_2m$3YS04Xaix0DEy&G$cD zBNf1*C;x1}`$x_l{6D-%8H)ih{UN^)C#7V_zc8>StWJ@4P_b42&38K)S$`yp~7KYR{tSNe1T1R;F*DAM_<&FkhUS_P9GFaWvV zx)9j5TNQ;uAK`Aldp?qN_5!X^%{1J8aLdfGfokb~F|g{{P9aSGLSxe-q4BklL(e&K zoDHPj)nNLsc!ohIs&kzJFdWJ2k(Qcse`QW+9E7{%ERXX7JB%pdr4kK=*Z%VPumIX-qiM3$3+^nR8x2nsEM1e7?T=!< zo_;C(E9!8ehkSgHFhV%-olHX2Tjb!@Wt<(UPR#v6aW_ttF6@pa@t`PFnY-P5Dpe~# zAo@oqIV?w;p}it94qZP`iAkDqm|r%Ou|!l&DU>Sa$eLlJv=u+FBm>)m_;p1u`6W{f zL7LuSHHggnpez-ci@By19ut70z%>C2QsC-@ZCDAow`)+KW(P>@G|eo)@B3V3`lNf3ItKG+RIrnRN)M5VHIs2CDv<`75c89D7Gv@ER> zN7eec+V>M5^|y`ke2~7N^bZ^5Fnd4qf2xcBS6-yrss{}I`6Bg)jq>Pv(V_o01GD4N zH2Pk4v*L3DJ-YeshroKf8W{cHxE4xR-WVK0lFB0*&33H587G3Lx#c9DxY`aPGkDpL zY`B!RtxqZ8*@vI^i67iq`?dL?40O?aScL$Tu`DiN#gderE~s{52I43j_7Etyo(>>5 z3mii+ZIiX~3Q!;9sy(sz50r!@BGTOJSqT@rDyuu4r_HWT^KmDxl7fMF7^vBac zx{mV~GShhD`O$dw{#Lts7_!S)V@igNe+fAR`Ul=lg*~ca6OYD>yknZ5RD#iIhn?>y z4+9(hfa+z7=XJi}XAYZC5|(LMS$_X^0mWENEeD6>!y)JWBoc=In&>rVn9eas z1k{z}rgdb-gzfFzCTyXnRilffmVY)o2Vrey#qxu|vh&jSVW9W=1&R(4;cX1m2U_@n zKCS|CUHB`f&4Xn&so=HGMUFq_##ajF!r-ZW)cdrZOs{KB%+u6o6sL^j6rj?JZZcw8 zJ`oN^`OcOR)&_Yf0#`E1=oWk8NIp(Rw!>mPI3){!0Th^`h!WVFNXn4jDYu(IOE9t# zG!R6%rrvLY0ouI~u4Tt^L(F@1U^^=fJKsazArm9fP|KGHSQ8)|jgopkq)H_Sgn+v7 zfL~qk3lUFqD=+hhdHdDKny0CC!}eUvh|@|#=`*oyI+<3xM-K*_@-k5N^=b>T%^yM1 zaOWC94-!ggq!k5V+x@=d!v3Vyb1B|BOvWa$L8Hf7OH{M5IGAv9k-WTu8F6YIzSM6> z)_bLM2R29bdNq}5^hNc=Sh8qd?MB6kmJJD6a?aE%p7urW%ZbNF8|I=!m9gRK`D7zmQzum{=BElfJ|~lMLzSWHX$Q>qEC3?s0@%x^GOx}iwTI`q~S>& zlnbstDa_kgat-u{nTu^ZbdC0Q)ip3Yu8y^Qz{z<~@yAhoo>ri+Jgek=&Vfe5Eu^RUG5)cuoJ`W6z6j#f=m|MApg@nv`#waVHt?wg-!4997+ zl#W-=D$Wb#Dwrmvv+6!`j9M9mNX*u|Uf1}N*@;!@nReGl)D|eN+4mFA<*r;O@-f(l zkTw_YN)p1m(P> zPwJh2>)@3BEH{c)53|qrLyS13$&L^mzbHCG3+x0N^>+&rq&@fL zByEg+5Xbe3RoF!&g8CkO(OFy9$`xJ@Y=3X7zAo*j(_G8kp+i_w;T&55x+$R{S@Ik1 z9QTaEv4N4-S~aV7J+ihlEGD6J51L-qqV1ZIq9GB2(ySTQ)wvDm9wsdMeX77yyn`j4pv+9=@KdyViJ;JpxiO_KYH^(N&@M=liz;Bu5|E z+Qv@7MFVMRB?+2Ee25&V^jeBJe|FrD-*Sv z6{R+Q$0Vh_q@PZY54pby4Bkr#1nNsXIzk!VrYrqT8VUr7Uf2j-+h&|w^KSleMG2N{bHZKJZJD{NxHZ)Ji?8dm& z5q#pHjuY^npPskA+FVPGwzMwD|Gmaf3I@mw-~}&*M5n)Sf&b}M0DYmqo{q2mt(Fl_ z0Phx^W4)sYE$jL^%K}WOA~+Bx*8D1f?Nw4}xRw7l9s38j#t@PnYy9V4F>`n(gOE1YDV z$bpo|ArI2>#>lU`c%D6Msk_hfuOnv$UXRY<_9JrtT?~K$U;>DMg#SexMX1NVeQoq5 z^zTp<*Xs3OVt|@yOVS@ubh|h0Nem#u0se-fl6PQfuwG`6mD$3 z9vZTg#CRrQ9Eo|DVf2%^CZarbh`14)=Po8u>}@hCk-S5&5K)oGxf~?|^!dpssYwYc z1rEUHr(c@+YP98l*u*>U^`j$hIhax7(yl!YZ|RRT}n9}cHC z{{H@uw>jdZ{TBfJo#@%CvC(2W?Lwp*!D)*leo5OCYrk9BjjL_TA%!WlTX=UR#nekkDbS)jM7od~M|u z{jRFnB4yv;(cLP1hum@0+TuPQ3MU5jt!y;E14Wb*-3|3a9;qLxTd0P7#S%;x+Z!N) zuCA|UvWhqyquDbt*@r1zEnml-wJ6@;-Z-FO_)e0GC{n_@^xASDCMOeeEuNb-6&oE2qnu8N5b;@ro`bq^X(1Vk4Dd0O~Xi}!hLx?&zAESEIQCv0?z5tHi0#;>#l z)-=CB#Rfpekr5^8;ej$kOrU_?q0kt?sS#XNhGGa$1b=K_Gq4MQ1A1Hn;E3U%)lP^r zfdK$f6o*-kUv5lk;I`|m^s-<o)Hf9jp^8iLK$GDr8Pg*T*foTA~;_57H71^*!SL-W5mIIDhUc z+_;m8V|E5f>R=+=NH6$GvJbyj_Tx)MKdZNM=~%hr5jh#Fc|rsgUgS~526^1+BNoQP z*#KR|Q#He9+~sH1N}j8e@*NWm9Rd|?=Q6e@z8CY+k3XNP8H5wtK3B9K_Er3@W_YG^ zeLjoP^79r%F7x;Rwel;}V zgFJ(bT9dIWy4187yb`-VD}2&WDW)wfKC6}QCSa+hUN`oMRl_3ib%w?;o5~LBF2vbn znJKXf$t3yx>@h?}>fUxM679ZJqmS)EE~k`|7gx2j&5g+C5*JrEA&WQO`GLfagAKdO zu^;!1nHKiLe7PN`S&9htP6@GEp?Wq0((Yvz|2ypJt0g#S8AS>=VP=%CEfYN5(3pfALlqk>q9sNw;x$nR1?Q)et@DsRVOr6#L&7K4qHyTr zJPUQqg6i~%h9sjOjA@vN#{IAJGD9ISK@!NYK`ZpW>}I>NSV7q(5*id;DOTT@!^fz9 zYJ`)NEozPjh7N?eLd_CuOF(#ps_OpjZ3JJ;aH&biMOnf-JAut{l+MF#u`obd&(daJ2H zV&DMGP-+*q8o$-jAqkQN2yQ1LVhaHX)+j}FvNWgy7e&-wy?GK!CN0$XQ`+9WWbqy0 z8j_BlLVw^eHcV(CyKf$fs!v+aLI8`YnJ(JrT&87zCkA5%>rN)U19{sFS3e!njE^4% zlg~}^1Y`IiUI*V&Kc4hmmrpuo>`zS}Y`^6`kV?%Ly(cf|14m z9}v~|P0YDud%jb-hv8R-7EIx2?cs9eB?d^JciQmK=e~bRnaMP|kW((~TqIr%EL0(E zdaYJs*uFRf!?X>oxcve%y$t}>xN{(*b42Ct6EIrn(6cx~fV7##l#RC?)?rMs@0^V3 z?OlgN_EJDH0qv-Uo ztW~7N>xM^ zja;J7e0sPmm>I+jid_kOM4+$zHO$mv{*__!>MQmfq@LY;yScwge|$I<{!Tp)eT43l2@{@)x1BCWZ}(nBJ)XQ{$sxqyzf!c3 zG95FkZ~0sm#23z5Z^Zn-J+0CR@8?qcdIis^4nf7C-cLiy4aA*!q0_Z53H@6N!~l?o53-eda)2U!p%l)oMe=ieqFX7dTc771KEQ2%v;lt_ z`Tzl$fct)2st{R73_v^uHl&ZRpo&m7lwGb7Pzt98jPb0rk z?>=mkVC^dR&^C+LL}np(@gYqE-f9UUUp)k*sOiS3jN3(F6K}~l2E+!gZKc(@#HV0~ zgFg_1UNecm!}521b5kEoR_zOS<9G+pyJqzkkwIc1 z1uLp97~frZN?Obb+aE0?$`%+Um>Y?-_ZIFB3#%y>OcY01MDmO;?tP*-p>-T%m>2?| zFilfj;AI@Ut~d=t{D^71r631CZ)yxva@_N~c)Hs7vki4IPbro=Ey#Rq0Gyw+r-VXk z!XQE-+(ZI05?EtTMte|3w@HQnCV}~%q6H#=86AKD=mGrMb@p#eF&IiN_mTIM{12PV zE&F=68_YXL!ClASOwp!RQ_W;J?M<&z3MhpnOW@MDK z;`oK6RTzlzc=3RkfG4KdjEBiyVc_qJCxFkaORbqx{HmnVOyNBi@FD%_%bw;=nAqjv zWLtKluE=(C`NmW6CU>00qhXIj1F$Tb{HI9Fs`cThpEi&8u11?V3F0{L@7xQ9Dnxpf+f*$a-EAy3-*uiWZ@h_Q751>J8i<3DgGE&}U0xid%amFh zl}xbq{>dF9(kt|0^*tK*&I;s`h!88FAq+2G_4+jAv*G-a6mB3jtp+yWn%+&r(=B86 z8Mquc31%-(wAwI+-a5~!ro6Jmpltqm`#F>++yWLxI>jv2#YGb#QbdI_MHdqX-#P_| zXjDWU9})J22pk%Sq`+xL0a_2$%9kCFbE(oHh+y58(()E9vZjLpG!9Z5PMgonXSjv; zZhv}fk%XUVjn4SWj%eM&!muk)CPqE94*`IQ^zzHMd>tTA2Gbw*j3LqwELvRji54i6 zpI975XhqNsWh{}t(MgytLtHH`(Uhn!wkUOlL0Je#3^lW@I-yA~sTi5rQ}LsH!36e+ z7nUrlr57DmO~f?*v-?46mj7V+sr%u$`LO<}`=O8NvC3cqkv{jFz~N;+g%2XrY~U zQ7QVw6ve0%*T>~}jxL?V-(7P_(JR|seN!-e5-mV!S7TW3mY=#G&`1Cle~1=}j!j*S zEH_KoaAG&h9y?_>4JIc~qQ&bQG~RAi4Ex(B(c*SJlIQq#BSxC_ZZly@D4|ISE&nb6 z!EuqJAk&h2W7jN9Z#;{1IFT|L%D(TF6<1wa=o|Dx%6vdgJ!Nj+?W|n)$e@o^<(Nc2 zf9GXC6o>A#flj0DOU)R3t0Sj$bf#m_3uyIKim0Adu2pfV)j9KZqC~=QL|)sWfWh2P z16jwE_S1F4z8nYV4?z+NcPDj^Lp3eW9=UK3e|m%}U&^PR4aBEC97T(=s*OVTEeUkm zl$qf+HA8}NpG)dG_!%Mvi^O${EPAqR`}`n(!f72nS9A#Q+aL z@V~(}XWNp-n-IanEGYCXq6CFjfmarYtWYD7QH=u&Q&|JAVm!aPab!QdigB>KzOnT7 zYTsP9H@j3Upwu+ADpI&*sd_|{<;dY{6gt++t!)`;0W4)}ac={%n}HgEb4OdJha9js z&VcCeKTS6W;I|XRhDu{m95-7Z3I}?c6yj96%xHL0d1V--;}6cM>7;epRpQPQt~kxR zCBlY)s?~pn3y3wZN0MqswkBN$hjSbc0?)h6HFWscjIY(~dTX_3;sz8-NmqZ{X!H00 zz}yjE=e18J`PKj{w~KTH_D)dqAgquZGb`kwpKG>B2_Ep=lupi)XMRx8iqk*TP1fMm z?jUpw)L8R=?~qPt7)b?;oRER%VYv^pKN>9T{AsdvXK1wsLQHFshTGWBa=Zqhs=WBj zKD8~GXB|uPuug_nl)X+w}q?uUd276W+{ zt5s=5X4|D1>tZ0!Y!W&fnl014qVhrsXmmC4B-=O0L7IHMj~>dK$AlA?Q#Q>V&xB?q z21@G$4C=k8PO%fSpo1hH<*5T2JfzEB@GrPKXyy=e+X1}3>7>wXQ>Nq0%Xbs@twHgq zX&doW?cIl)+}qVm#@N8%%I!s3?75~9 z_m69+mAqZ^itv|sg#x6uTs_2PyvM`lKYvP_AZx;j#CIWDWQ2US zHcAgq$yj|!7mCzWj=15{$;W;0Jk`NyY{p!WRI7jsZt`1&98Pb~=v?L}{z)rjiE3Lr z%cyznu2ltfY7#pL)?AYbZN$!d$>4kq!!isaqK8S~&`tPLABRD%R~e zB0yALYmi2>8KC|$t6{N?#)XtMwEQQRAsn1b~P`E@2fz@sCQ8&i}At%SI!s7b|#X<_&|S z4TfA8#-DY)dwkqmx*IbX`z-9MM9M#%t(Z>t*Xt`0@bsfx;?78kHk(}8kSpCp1!iom zx*u-w@7Gt!`8t-F1}+49JAJ&vStF^z^5w+|^6wpf(80U|bxWlT?Pr9I&5e0l&4TYd z>rkxdW2*h0qdep_T`V-k?sW$-sZ>sT$GC6T-gC2Aj8&Qr(<1S6uTSVaTihN@qsNB( zXo>PnNpRo3q_GA)U(dtlVg=`HpGB$7%c2YBb%W5uiTMU73Xcgo)~;Ty1h-Y9^&u3I z=xa#ghB)l{Z1c~1@0~P}3TKh?WHc@cOa<02xXIQpv$Vh2bV-;+i@fkpBzTLfHJ9KP z5Jo#C&;XELaMEBH_$HC)2#=51sav=nUaP68r_D(O-tc>4>>jT)+BQJuE-5LK;8IaV zsOBs7)q*OR?YuX`JzR)=6}0k2(%{{ms`Zx0fv5(Nu5LGvVJs~i!nT7Xe`xZ%E1&6L40 zqb?{RH%m^+>FVpX>dpd6`IdheV#R?(dqQz~?wj`VL(pl7r+>qN1#f<6Z)xamdz!@9 z%oA=yOq{4~ODgxY6U!TD=%%vj%dT?*r^?bMss;>hENYki5%OvfAT1;=I21r;JpjGV zf=p>ZuP$KDBAqBNn+Vaq5&Qx{w-}RKC&6GR+kB%F1Y9hKL7?+dZ9Ruap)^~;?wvpnJo+`S0X0p4*%pjt;8}jP{qiL8dOao!|K3JRFDA>b}3#p1;=uo!G)RTa*m@ zw#X(q!zK5!aZIco_f83NXTSLZ>m!f$@T#>nA-=a#dgAjX<2N0 zbwbX2j6-eJ7;79IZs^bZvg@}NnXoHQ9uOlM!H0(nhIdw+d;veE-6JS!$2@;-OkHwj zvv=$6y!&J`GjvGT>a8LAu17u-o@Kf@BeUPB| zMvD+kgu775kNKdSiMr~i{xLwY+LRmy`bMUAe$#<2jKL3-P}p&XcNS*dw4yN*BK(7< z-&bWM+u{R#mp2E%SPyk$P-BiKnQJV}L?K75v9Y^>IXc)TlIqU3J`Mm2uvD}P7`sN~ z`b^^o-iyltmq}wj7ljw%5GM&4@2jt~9*RvBi@kxDl@(*PwdzH;&g9{AYjm!m5&~Xa{*7}vb-J~UaM52vO3H|leNwn%fn90 zA`PWa5+nz_+Qr3J+79bC+lezmG8e=Px?+5#v|oMmQzD z1A0?U>sJ*hK?jOgUK$-R*OkoO)Df`hxDETyV}c|Ob@hxnhurj7^`uM`H|1$+?sP$;qiUTeRA z$q^k0`Y?HF^U_I@8nKV7$W+{ESZVJ--8BsT^?eIjQDNF(MQ z+bah1fg+h)QUBB7qR{SFfRI_&^|HvfXiMJ_rEf}s?PhKiaceL+`-aqfUd zY{JChfg8;RcQ7Vqp>wAaF%xS@i#G%OYwUqPy-jMa=b9w9IeaEiBFS!?n{m^IT^Gg7 zZ#R?r+Z4n46}UOo9c@jYiPscm_bp9?eT62(MqEs#(8_kU&yU(yFE6Xr?g-OyrR$)l z%Y>Tc*d}d__xB=Z*$wYX2`n(@+LdNJf4(TK|6?-cYi&LcOA!Q=RB2MoRUoanua^E} z8ttmqQX^_#Yk1~c<%habOT`1d#UC?`U+ci$LdrdBl=!k(lJZ8;}37ZttMp`Qd zp#Rt_G0NN)2_zhHm~MgdJx7-uC!l98&!%_o4p$xFL?F>;s^!(Ul~)}MNT;0Wf_($S z4L|br?D5{x2@azeWUD@`4YvjQ-9Q}x>E+dLBE2isIE}wyJ`d-Xmz-<;$6RXmuvbVq8U-nNw>RUvLSNp^(ihyjg)=l^?3)fK3B&qAE!EJImxl+ z%QHt7o!*5#%Q++Er1R2nW-_y^fgv8VN~{?{umZyPg>Zu!Rec5NX%dxbz~I=Go?o;TTd}l16R0q)dhn>G!PchUkFQ8Ds zuGNvd&6%YQ$jSz;k#ba9ZB3i{(Onfu#s)#h28-7QN5%$kz?y*9hDe42g_rgpYWFX^ zMg9wKq5lJK;UXG={oW%!Mc-5?z($zQqyiBYOeDi005n`gv%m0GoLOH4zIt{mVyTt)-%=&oBV!1030pio%H;F@|4w3s;9diZ@5DY!62#GhgRD-_fpf@l*7z zTTrh4P_@*G0ltZ@zHGKujztXCWUOqt#iKt7PcPwQJl7#zcvQYISG`t`g7yQVn&n!5 zj5zKl2gksFg=9zZpfk}vkKyM)rwsbLr=_!4CA?q0ujMEKuY3w3QnmMe&(u{ z_(r%i8wn6=2Kt;-Y9LXhh+&_Vv3k|SAWaClN&R%d&0M^pI?WFT^_QKi?(|yvjB?-2 zP+^Ga{T*pwWWavluAEfbnIcUSKd=cSyFF*(W_;Wagu(=AP3YaCTI}RzwxW4Gw@Ff@ z)69N^m4OXzdX3ohMy|n;j;*A?PcKT9O5~POy_*qfb-=l+MVUg=+XUk9DVX?=HETg~ zT`yLiwFQ?c z9Ht}dwCtUdukp*ky=(DJS2s%*$mp@wk1KS$And2&A?5vln9V&>!;on)YF z8fG|~aw#qe6KuPi7a+n;U;(cr7MIKyh^vFo=6wiWL!;y|UMKKP0;LzDwq(DVJ#`W% z70HbiO9KmrJ3;z0T@z{i1sWylO4C=mGmX+DFq6^t>6B!9Zhdh1s!>rRn^eyj8aRVb z9Fi2MRyp?u6j|SRsxuE!kMj1)e~3$XM1cTOIs<9B{DBHq033%Rd4prlWCa<(fP@tL zN5uFPiuYS+^OOL%)(!HL#xKvm^Ni}Zp7474Z`L! zx>=%A?6xNG6TZy;nMEh{CPmr5wB~alN`w?oxN1pHZI@oqoCh zG3p~X+q`h~yTa{eXx9Gvs=^Pg)aZAG+jV8AVyOwi@LV7Nbydvo3b*+oz3b|f(NYVZ z;rXv#*EN}FWmeMi3zI((2=j%|3zmkF zb?xqP#l``}a;N0srEUJ3rtzq9mje0aeZ8CJxzS^(H;W7YXz>)^HA9{JtA~d1wCg<| zHa&b1gcX8a8p3j^-Feath|9GTF1|6pS$Ms2kFFMLGxElQNJnDq?j|JBnA)oX&!AM~ONKL?&YEWIRatSYh=a;Zh;EnIY2 zZ)NhUi(xi0I$iH%Fz3%rKAjVMkM(4bb(nT!#(u^l-lwM{8lXJXx=oxI+uvIxFMX)G z%_M;G6?cLt2>*!f|6uJbxZ*(bFI`o*y95#_+}$m~-6goY1qlv;!rh(V?(XgyBsc^J z9wfMvgxaRNXQpR*|GQ`Q?00z2`?GtW%jRCn<7p6Hj3*qwlbNm;B^v;8`4Gzv)gA|8 zyClPqkx?ABd1y#hfjFHpsJ>S$96jcSk_cfb)L6~>RKF?^g)5^oEbdo5qRR|FsU>_N z!0B9klMf_Ak{Oi7ek58*efXiEh0;Op&W~9|CZQ`~9>UZO--vp$K360dY;MT!u4hAA zHulnD0&gp+ZV7FFXQM+?JKx2U@fTs})#1)L-C5CK)v0y*qvm_)RQ>ov@}6fR%i<6+ zsQ#mmOj--dNN^f z4wlwKMJXT0&rCMal!KKU;*<)ZxeSI-^5KiY>Js|HI0b5r0_+L76Y6D3cDT#bLO6qi z9XKSsmZW_SxI5?6{Fk|%t^Dm^dAPzt>?wGHpoBsA?`0i1!mt-0-{yG5tTpK1&hRVaa4gDj5F6~>IdT&sPo_rzH`fb?Ik=TZ zSoq$Ec+OCNm>B!_uxO5QX9KWA*OB-XkyDM)4r)QfBGIYK+!*J`Xte+(GvUJJa6#8# z{DO!TtC(mj^6C=tcfVCvKs+7{&!<*3Id!1;4B)dzENofqud86R89+y?(3*5y$hDQ_&bPT{9u4eZ>)QK>$$&{adOi*YqQA?W7>s$Qr zRj?#qxNw_F=R9mXXL2uRvTC}V0xR$(875+_L_H^-XxKH(A(V*^1PI{zFdn|NE zHDP&DGyJ?y(5M#7w3t94;xRPvp__sVV(oJ{YZZp{;a)@i_*%u;QZU&B-({B|B|Kf< z++#P7`dssa6rxrmpE(IRZk~)*2cKi*wN@XU{js}m+k@uN;fKk)kAY2@FDXCj!DUhJ zYt3s*kql`KX-k^Bq_4hs5qjt|hPk4-`y)h^=J|=DR=AL#2@ag8_zQ{ z#Y}m4d^lvg(q~C5zF@@rl3qX6-zBjW3rX@?Zm;B;zc1F%y5HiqHe0AQIPPRic=z^- zQOikFlWtj0B)HLLK#CSDGTn;&lGbj35GRmMPeN;!7NNR>0Q5O+dhxH!MyK=Zk(UA z1gXL-jKmp0sSq^)$eg|6Fbcx7#z9!vTVNP`b~1lSBDw=(6VTe_l+a8x;<^wm(H({v z6!sj&L(BheD@yk!2oC_=EiH&)*}so`^ZF^UAc_-T8$U)Elx~^;Wi2a=W5>2Oi?^qn znNk*en%jxdI}I!FQVa2nW72Z7wn#OMC@U7x1i9{{JMQ!DX1E}W7b|U=d+rNa6l>=s zxzaCND0^KJ(j*zt1FUkP>oNyWnYJ5_elwvLNs7^F5F#r0&qL?;i+@7y|6ko%NpMR46EEm* z?yNr!o&V;}B6vP@3IxkVXp+(`*}#^!RRy_Kd<{} zjHhGKN2c!1#vISO&engB$X_tO@^jangAwgI6G1h5y}?ZNRq$#qwx+i`xTeP!dsG(V z>8=;i5a4Io6Z-0%m$YN}W5!Z<1unDiILe0f_jtsIa2C&@0n#l<$?KNRvDja#pxiD;|tmTYRYHmhGH2&}4RU^Y*fUxXwhAMIByheKO+$rGLGRB2LS#}(ht zM!HK7rHT^RFo)pIYBafMh@{J?crRE|!gFSJ;d9rAIlm`^%7%&TEeTlYl zt0w<8B;Z_??jIrjV@UX5Mo*l5#Hm$PE1?`pKcA1SLHS03UfZX9zBEdmA&W@ANRnWI zi>=A5N?Kpz4Ufi*{la_w$yK$6%Y~`n8)tZWTV9&fg@BnvXPe12ZGmqQ#m_^6(2R(U z=R)V>O%44q!@q_EyDzW;d>N0?bvfP#!j!taqx5g>eAbVJ86$Z|bVeFd(N3n@r8#Dh z_p0*$vXqUkva^)?vP!dh+uZoEVt+e=Z`~rMCDqpU5Ms;s)g-1hnd5XXPL6L+C#H=A z1@T`CCQbklumNEBTkUubNP&A6$^4D2#7X#vNTys+%lUs|E8(p;|6!-d(sn7E`*;`j zIt%#<0wUK)K`az)vR$HpL!$*8H!-T#(HFc%wfLQ_gqIsF_d8qZsvI>RECcR>n6KIr z{b{H0mG-DJjfy6f%O>=!HxEf>v9)MxDFIxs^}Kd+u^iFmL?I5&CB|x4BIyoA{++r1(Ce_R`b#1 z!=*d1JQswkdRf5Q6&TQqbAXQYc#~OTWD_Mg&O^JX7x>nIPdO#;LHh4(r49J7(B8(e zY#mrKj<_u64Fb#Tz0aQCguUr|7lRy-VS_+V=Z7R9B6Rv0!&^!bnyXX)X`S?lMNA9js%Yu9CSoy%_A? zkgWl6{dOHJW zC%dK{`*7E6lz}*@)HO?^0K<8)&}-8U7H@cg5AW;So*{XlVMfOyEz&55DF_&dNG@a= zi8@06PIk2t0*I^hs<>nD`?WxysUdhJBVPk>fOjPUpO7GKGUzf^~ z5UA+m&g%)3l-pt4)&xw8xuJx8NFrxHujd7*x;!QXD9vuxez<+O+YBY0yQL(OS=#<8 zMfc=NP2S@xdH93C&+n>31FC-~m-0sW$JeDdAE0xF1|V8e(C`T4m<}0))*zh8=}5cM zF$!bP-UsVe8W7yK`NL7H8|`xNLm-YUf>?<@dT zndSP`afCMN4seiG625W5V}C4TrFvA7o4r@*btO9Df=4a)Tee6|GtuZUOYu{e@5W+s zDX$PbAv`4Ujs`yw$WBT}n5muP?KyXaw?|*`KcU2$4$dYZjs@wWV#H&q;SW|z`+Ki# zlc2MTQaWj67cYE{Q29JA@Dg)OC;rt*l%xTU@WqNyL>!K+`oc6W>K}Dd(%Lvo`_=L^6-7a*;Ao}dfta39` zm`y}^J(CN7d?^5nNVx&F(=wk5y==Y3ft;1pMH#+-f>ea69RP=D1~$3frHCoGOzvYXKny8JWzSfL97+b! z?c5t@pZik&8!Wat8w{!9&;0YSkw^;Y z$Z!An?l_-VtideFf5m_QJ0tJk{_(BgHHA=P5hC{atq^d?@?0pn32g5lSZqL&2PR;8 zh<4`UcA_Q zZ!L*?5THFiYMzDs(>bmHKih0SmPz!Rd0=9U=V4F+x~x?i^2>;oQcQRx0Jyg>qP)Be z)KFeg1rG%xMmClslCkz}mfKdh6^|cRLp;WJ--NO+AN}@^f3(5wpKYUqRNY2YHjcpK z*fr=QtJ?*PyG1f$-m>zPv`TFpowW6B%rO?A=RexipOO;?wI9Fs&Jl&Dle4cFed%rA zh1nfJj}3abVy-L;kUQ;%I<}Q|JyF@d!px(dJ?ZJM|B~C&3k#CRMqb-tYFM_hvn&7A zPE-zpXnn8f;dsw~2^Hd&uLRS=U8X_2}wC)`R<;Mvjkjk<5Vq8|( zkVC*AdG^o!dhCP0vb;|_bw3_A)%QKlWv<9&bZ#!M(~CF^>s_VZOc)?s#IEigGJEbNlJ-IXG?$?c;<#UNF30mz zU$-Ah3L5vqR4ARc?^cey<^#42PUaM}@|1g@w!JZWIP2Bj=w$%-zWe-b7$d3etdD1U zkO?3!nFs9UJ60g>xxWj^$=!n&ROu6H zS(}64OqMvj-pBA5V!=O;bmxWmrc&fQLcT#j@&G@HgRi&WDSEDQE#;sak~r8{7DD$`D}BodDdElE;0=2#|z85ZYGed~$3Agc8-Z zpRvU{ll>&SC0`mq;y?5uZOlYVMwHlmI1kr=HNcjmiic4hk{-HHKg(M(}p@B`ZC= z6L}Oo*5OY4c~{78f;{)(pD_dp8coU?dMagmDaCZ2l$iKj%6US1@CQZ0W<*IA;>gK| z`yXdtAbtBQlk^eJA_t=CVey}N8!l+bhY7ZdHr6=_2LB!`m=d8df-R=pZ0sd^b zt9gDqx~hzSzMWb17Y#J>Swx=HS~pdq>$kTGU zulu`3m2Ys}E#LIM?iOHBqxbk(eXmU)N%QJEyvtCLx-ZF903hEIxHGTACkAP~as?>< z{0#dkK#I#MS<|Fx2iK4}o>nd7Wif4r4psBPY9uXru%Qz6u)4A4onw=+q(FJ=oCWh_ za*B=cQWCgPB3xLQB5Q+bWTIYNREeK-_#8?(Zkb`UV`p7D%;B+|RHDBU{0pw0tIkSeK*+JX)5z zh?Lux$v}=zCF_aEnClw0d6k=wGLWrVA++bm9RN~XySAJZ?{~$+2}FqCv1BjC2B4Ik zawR(F`GgRH{Kt!mr5K`;0v+Q>hMsj<1#ExC5JD;?|R2glUXDz8SrpMCVDKoH&<&4{xx ziZy#8vORsv#=Ge22@FI|$4a)uqcfq~`yGsGZUN}=ZHvKI=jrFjgvh2hUmh*UA&i-b z)#A`!ZsQoZMnD`i))~OIUS9F{_6fH#E@fZ2Zal3kn@YB>SApnlhiB-k9N(PV-qDv^ zC!>285l<_+Yi1*Ts5ogqojR`Rp4-5~6b2-SgGG2Tq_1S;S%hwW9a#(3;9-6i#`zi= z%ZGy+W#|hd?}6u&v|+R7J9gGn^Xs+>8Qj_v@Kx>psil*`1Im<$V&hx_0pg!SCW*pM z4?;`eh)(>EIKmQ9+4DzFJOX*hnp)iWzSP0Y zQN{|gwEuMaT3+?N`is-If2Q_tPG773W~{*obyy7ihX2{=dlCKei0prpdH{KL`uW!lcc}-OWB=lNTZ{4i&iCGX@g+~LZ8<;HV2g^i?Q$aFg<0ePY&dId+NWsThIAP$ zYR>q}Pu4^d8H=#8Tn*_scFs67LuUItBV%~(h=q8~!X^B3V^gjP4dfe{`)~}i6dv9G z5K*2?vq@I}Q$$IQl@6~!D{@r!hlr9&|Ib?e2fy=?q!w+y03Y~q8q7Z%6o5-cwO1

    20Dgdc*>mO$N3!12R@|B$dDf50yJ)U>oUNwS zNWt!-w3U#sD492>=glm9sv(8A7xC?F~;{F)op>h~0BXw05NP~WQ1lSj= zaihFG3mWk)IzqW?w9o1KUyOR6f^vxb$^B=N!+IF?IP?(@cH3riC)yI;;YjYi#0bhl zQG#k}nkZ#_i>>!#0pHvjgO}Tiq7eMj5KHi1fFyg6hRIyT09i2`CB*cH=M^>pwHw5t zcNrov=auvK=IYt3gHi2y{&3ha1AA$o-L}>-{Tb?Vms5`rC^PlJHuKaeG6CO<)dV+l zh_`)&e}HPpFT>y+C3`TO;f3uoV*0lRi$DmjNOb3sK#m!h!n}9@`+O0Dj?y)#H-h;0_6JlBI8g|QM`0r`Z*{Hzu({` z!_e+~Ef!4OZX+akxIC}+PhV9%Ppt696(O&&RiJPDdXVL}M&%~L>^b>qG$-qUPk zt08tfkL{2lH5ygO%Q)BaZ7iX6K7yXAg+g|D;$#=NpC^vav|^G~<4`)22!3koxlg{M- zpl!{Zg9)sYyd)$9>(KIxLp}wA_n|TC%^`1;%@hqW7@?c_ve4N1QgY=g8zDl*&sdw5 z_uOkD=Z5@tECn9^xd2%*3XOBrcJt*_Uh`okM|=$6U7Vr=#MliFEuucg7ZbKyk>Jdt zq8=#I=@MkgBT5V%)e@-dd{3CY#JdR1sUJDTyO^cv60VCRR!kR;O5t9mAb*DG`NbQG zK{g947=}0u*{qF6`+l4wkUtgZ=JG9K?bsn&Z*&={x+Ru|f?Z@%6u#Q@N2_&pTaujEAw!1IStN2#4kLhtsJHZB|6eld6tS85aLyuD#!%B$IpAs78vXlU7~G_ zws%6=FI4CZRznOjv#^ke()qnb;{q(Rc*bv-2s?csYJu>CLzn?%{6mm%6N!wS2lWGu z1EJ!rY8XsI$f2+ya=1}MeT!{I6oS}w>{7kk&eZL0Tm7fKr_X{3?BJmfNh+FkHPo*} zp3%4*JO8rvpuEVbttGv7i^eaN>e#W{vOePxn|TUua+FU7F1Uj?>M@uhvi3@OzNQISoSy2o z@bz>tX~F72Oqg-8{`l24m=1fg?8fYjZn6F@_7r$ZsT_X+21Bou4=i1Tb)=?mIAk0D zVkssKeN+li#BS3Svv3r|ah_o}Zk+i#ytUD1JyV}Z(fg;kW-M{;{Etrpo699v^#PeX zg!_`XrhOLao>$L#ML(4eM0CAJ@*!eJQk!lIy z!M}QZaqIi{e}HuG2<}0yW)SLmB`&uvzS%0 z@+*diA=1^W$IA_}I(fG{E+3`O;&sH%l-UCpNK4Dg2WkB+m*prcgQvFKiRqll6Tjxp zo=t0|Dml)WjaczJ29?j^K`O~1*^WeWMneS);I;GYRYQFXV9U4gV_nDXY0W0v-#dyd z>};~^1Tr&SSHlxF`hVKJm8?AEfUI;HyW22mJBZayVtjp?JZfPWdWKFPg0Kv6eFhT; zQ3obFPN4)AYj?P$;y9r>_qwHlyE52qgH}FC+TE%bA&7_^UO4S;sflyX3_}eK)uAo? zRLcvt84rsAKmJN&$tm^U>x%{@RPt7q^k6YqT!tgdEWCvNqABlUH7W8PKkv%vT@keP zEc;RVfK9n+TH6O`=hMnwIx0OvYFL4r1aA#UOz|qwFUQ0L*cx{$9SZIgF0lDiehy%lI;1Fycwhhy;~{A*lY{>h-8WIEx7vl~^#! zC7$S|r`8TW>E-^ye1`I*SvD*V3C<}wEis>WaWS6N-yopy zXHF7lL_%SyA_{}~$PN)4athHT`*K6BtR1=Dxr5uO6$8}T6j5|)zi3Z$5BR5k^hEU@ zgDx#w-WJVZf_1Qd5%s%E$e*!&)ye^3p-rhone$Le)g9R=W52AABs!`{v4wlEU@*b< zBZB`l7CtV#52eRo$PdfnnZVN;DS;(Y5ck=FE^^M{ou}LFx6vl4XqMHkSB#{kmY1z# zr>OM~r@V^KiYJD}#9jby0qr~KE4Lbowr(F?{rsHGdgr)G-Bo}d)Y(Y7!HdBppgZ%D zli$7jq4_fJWG()5<5h`kpVzFQ)R$yOEIQfUzi=opNh0I!PFj$3?r>|T5d+0~!6sUX zi!w1SLVQ`h9`H-jT#>@Z>C?Wf)StU+7x_`8dNbI5M)J7&DYX51o#e^VqXu<9b-8{I zoIv3ufUij1mY3_lS-cD||0+;Zt4ec-PGjrIA$95gC`FdEzD{>$Bbt++th(T6`7O2W z0pG&shqv&VYy%{ZyxN7leDB83Wl=KXO|2r+FLUf>x|Us4&1;_^U8)wgd`A`g4=sXE zcK@_+o`szB^3y$`OdOyG3R0=s;E83i2c3FRsfD|(`x{T;!ZqRHGE#Ik#p}PN@4-Lvn{m+`#|8bePgGx!mV9=Df ztBj6E@EHAVHgrPRmEW;Q|ED76bwj#+bnhCXYut5kSKUbZ%wcd9ByFH`EL=K%6ev2mZ#gxmdkgduHAvb6nLpb zFSBnNC6ppm?Ouv{gc-`(Y3`Pm12Yog!?ve&iJzoZ%!jO-jB+|YG(f(|=BJDxj z(lqwgV0iOp?bFkh1hbi2R1*v)qC(sb23UoAK4$9Pi-IL&vczD1mu{s;C0*u0xCugj zOaCm1b$NffEDk*i25Nz?&7DCm?wM`<^)s}$9Y-o_EL<{hT&nLB(B_lQ%Zy} zj&sX{qz2&L)#aqe=qBx;B2<~BvGc5^ zew;9%kQd!kM9+~NUO3ZbUeE*mapo=`22-2BkU^upX_~xD9l(%%-&a&wyc)|V!$Wa; z(2$>tU|iAc;O3zkErap*U!!HPG`mY}G-jby?2FHk!G-KU-TFS56gMSLc|BaZU>9F@ z5yHb6ttxcl5oXq#d~Q#l%4c7&hC$a9*nD8(H#|%*ohRMy1ygM9hty=PnjIsF<~Mj3 zOZ%M2nP=lTh1emt|M6OAEtHpwd#JL!3$J1CEZnhMvugh^+KPFE-e;ZV?jlJ9e{bf_ zIY4&TEY-zpYJ$(rPaKnbEP#$v^6%3ybYXlgJy~8$vO=IS0Qdc$|3Cs4hd9U3G~HAb zjq!H;4hfHiAYP`5@*l1~+w)M}6#vyKWsD@c6o}kl09g+dU*~4&pC%G-m#m)5|I%E& zOWtX@#mWR0UVotEj(p{C1AaCp?N%LJ#Lh$v(!zsKn^f1QN@_o9Pb->Da^RuOanNu- z5Su-F{aH4Edt4+MJ6{mJTuw`2=xSo%))2Pt9EV%%vXSlmSq4zcb?6h1l|k>7r&EJyc!5uYqDw1eQ^VL9ZJyN%4=MPn*3PlcMi0? z^XTa;XR7!8TJNH;eDeTv>+hIOA09yLy`v(lB!oaU8>Fv*h&| zIw8=~b61@AiAT3V$5jMyzITd?yp!WI2;czH@>?3_>Qj?Gvs14?robeLN}+(--@44Q z0(9)O=zT2;f>Ax25*rJJmVppm@?go{bHePia2Bs7VIh!&%}SSHAmpVpA)&SiX?eyl z3kBF(J>;@3E{~Mg08cMXw0?xTn-wD=B#)~iceymc=YBtf1CYJX97ZqzQvD+)njLJQ z8Z6MKDXan3X)~q~#{C1bPDcM%=VjpCM?>#+IK0E?@~sO1 zAG+s`!LTPED863}y+>FqmKFc~S`jtBz`}t$&Va!#Aj)#-e0f7ugLO~qpg;F%g_5|x z9d8?hK$&s#|7&xQ7M-aKq(=ZZ-Qtg}$PO<~qHBqzK8Sn`7k_*43zPneT_`nyr464c zmD2f7i*OmZTRK}T*Duf!aHnJ(Gn5buND>6}JtGhDaBhsK$#Lb3H4|Kfh|ZI)Sth8r z9G7^DCQ3}CMWJ8Z+?l83`8GtZ|BZGNoSx8nry(SUK^~RSHTxD7)j^>J_n)&j2nT;o z@BM5K?Ln~NC+ZYYlND(JY_N_?(=ks0-$QvNT<8Z^102hy&*A$XIn+^qjDoz43eN#cUhUzv-s8NMVF+>!3 z-}ZEg<=$mf`i?FxJ6|54__3Foma|&xU-#?o2ORxBiQ1!N8g>vsp)a9%x%|y@?za{O zxO^AZ#L<~4lA`f7ToiMWs3lI;w;Ri>yVafpFp2#ePSD#SK`8x4O1M z|J1)V92NR$QKGEM!j`&SF#OmVXFef*@EM7Uaz2YT{Kt!;_>UJQ@GsC8q5n_w@n3By zinMD#`oQrYH*B=|Go5g%j;r|Va9EVm#%(&v2p~R|rqN6g4OOO|CBYf}A*hVzC9EcR z`=73(#5H^8%Y*}qhmct06%ri!*pEZ922Y z#cG*p7&m|%VE@a?7bjAcnU)9(I< zCQw{mb{%rQPAL2ZjDEPf?G`?HY^oupIi4P-5XL41`c#$q5xV^+-I@Hxiuv0I=;A63 z5+wu{e{)f^^eqG&uxOQvPtCHkc&x^MwZ4xmNdLcKkNnRELsQEaz z!3!h(X>}+!$xVMGQYHCh;OYFiFI&%s1Pss`++UZ(F_=$=q22k+x;TAcUk!MA1%h6% zi$acE z&+;6?Po8c00a$m-Qd*lHb+jiW07B;kL`M$78Sn;KGf8<@e3)I{zuSu!p2n&?1=erA z&@yB@);4OsehJor`r5zM@@H$oBdX)Gi=b7%U3g%?@zT`cC;rFEc&2WKt}JQvR_)g7 z!+|bfhygMAfG&2P%Sz)lpYB%Jh6Iwjc4Rjig?tL+!3YQ!OKi1pgNZ_se6QxuZp*+FK|^*?a^x3)oAeEWN9wJeu- zNt1|qRWvzOrN8R#9Tq}s=RGt(T}j|W)}4#&r2$pm8Fc!ejwN?>+;P-um9KKVi|gOy zpl4A1+diTkiKF%8gT!k%+r5N=veXCF3{JbmUCatox1L-WY0#i=oL9Tmpf~Sc-}2mZ z=b@LFc?hcC_y1+Pbq?5l_1_q+9n?rz$5alR3B4kF@l#cBLUXM=Y`J_aPbAS-6cH5u zBvL7HbUQ`A4i(s@7~c5P>M~*!g|thMjGb4o=Umz`(|{#fBwgJ38Rv56rYOQynN98O z8KIz(myq}Gs@O|Ut0cFi6%}RSNeN8g?2jK&hD{D?{;Hyo_Rsjxm$CD4tye8iqw*Y5 zx`6an0V>aM|Acz~UI>2#EqUy9sa-}W@;*0$cjz?kvxy5qehz3>h#{U8_HhqxCKVOk zzOcmrCTJbURh!ViMz!|V=3Qn_ZXfEUr_=R+asd1H-c#nL=7Qme?LH8=vyR6rR76{< zs-P(@>deS|s!K9Gd3$diQvEbd4B4J-&+WgqTq#q~F2`jJJ zu)4qVxQk0*H9ns_Qd3K=-Neo=&=z?>9RDsoz*+oGb?4$`1TFqgPn!@3qW{i+y5%w* zmIedbzNOcrqVRJv7|hp<{=z4395pL#%}Xw%*Ji56s-CQMRiZ z;P`J*Uc(Jz?8$iBmEtdw5FV2YkW5tb`rGUH+vBEH8L1C?6S!glZI@fBrHJ6m2zAn@ zO66U3u2Qp+;b7Vv07yW$zkxr$02BHvv9RM81j4MBv@aPf1Ui0xJ~=b>fhn$shV9%! zpejb>e)ck;_Fp%l7D@y|>YIWgzdeNYJ)=N-gh`-w32y_3QND`A0$FM=qm*xBFvids z%T5|r{3rfW&K7joa{I{}RR$a^%%@=Wty}V5ZeozQuz65@tkMjwcRgO-Ff%& z1CRY;Fsmj^xj$Rz6hG~o;En>dbVqKC1`W$syn7V^qsbJ!B zf+!bFGS(Z>gq{88l(m7E1kpFVOxUO z)G2VnIy%F8L!SVptH8VG4JY?r8eq3Q(-;3148{Ok6Pdq{<@%?tx{K1dS_g4SUld`d zn=n<4Yq-0Un5#6OC74a5y1+XGAG(kij9JsWgvhLtT9*-6^lmBxMSXJb9c(J<6vf7~ z)IiA07~WBcZLk$`6eBGq2P<)jpP59_SWmtQ?$2ADAIEjDtX!a=b!zKTmJ%w@gFt-q z*E*BE^a`u_OuHvge2d6trc;oHsS4`>Q<}ONrM?wa>k-qH$>b)F=dms;Z z8x=R|7eyB8h!u)7;an%IIufy!(rG*inN|KoTFOU!@FoY0AePJXd3k&ZhoXowD+c7i z8;K`9GgzG%7hMKT;B-@|6N5xY`)c)ijhRhssxZsH#^XD$zdiE?-@S33^9 zYjxX|7{HNCs8*`jKM;c`eic4NepBgs>W4Fqt?le=HPJkAGC5OIYRUl6c~fle47Skq zZ$j4T0m4ydA7uAHAH@AjjrIL$mheaD?Y8~{kl~>O_2=9@T;>oj1W`XS1tyv=q#d*J zZwvlo*6-gZ<8=XtN8~!*MJ#$8FTRl0@67x^cz5rhCZqRz5GkQc6$o8=?Q3om9U{u?0{>X7WfVsd(W=@5ebloony2k&(%TO#9(&lmsJIs2dB7gZTLRE7>Bu6KY!dfbzl39KK8+cD~?{k`KXd2N1T0d~MzwY??@e z^LlSWL)AVAsbb;K)zrO`U@u`?8Op?IOn(?<%$B4Uo>vFykQE?1yjVuc-OaCie;=Zg zf69YhiY;F~Gd{ueG*JLUCP8fJ32LQGWrmg|Yvg%j2fenVC`@JMTe5XsiQ#KHHbRXM zlN^=u3%Cu6*{v*`yrba$r+>KGzVpDR9e%Cv)Op?SSWpAwUFb7(0zWj+>os{8{A z`n{n*bZhX`#a8AA)V6&hjPe|=-3SHqld4nE7L1Ap&}s|4GKr5inOHo$z>rq^*jF#} z180!rWF0KpvFE#tHh5?8T9PmGX>zROM^f(om;D%Z%X`lEk^pb z@5YEo{#9|`IJ)>TVkz@i^U-W ze$)WYDHroqES8nAR_AKOrF$4JjgU5cg06uGK5~>XD1o^GK++_H;DnaO9LXq%-~vd0 zY9uaZ`H1Dx!TuqZu)hE4@mwkX6V@@Fj_sA5v$gk`6!ouxFu8+aKOWz!xBil3bRQ-8 zB`184C_`oDvLk^}C58^w(nqqFnr*L{4Ih>0gEM9*TWJ9XBF>IKmw&3{E$)~OiZ)% z7<->2lhkCvq0f{5y^;($qC)Q>$W3o{e%{?SN=TN4{!97CZGNwg7&XQC!?`#Y=HA{EL~0q*`UB)-Dlf{#=6io&Mg$6GJ?+A8y(p zoM7Z-q};bP%(1Nn29A+iz)RHJNT~s%Mmq^fATq#4&yvr#r5q6|sdzBsoy|drU`vME zqvl`+-!@OLA2IP$^>Ayn58~1&(m<>V&kVdD`DEV(boe>@aVP430t&43c(b)OhYzZw zp*24UUbbkPa1tHCPp5aI-*r&7{3^=gtEjL(mgRr*%uQ(+LTxXStN|WY9NRB8+zPH* zfck;P)QEMwND*w@Y2Qy?-}#rKOuq~c{Vpn9cQZT?CiRG>r+K(9q_WO^a%JN8Q)vDDeEjC)xA zB_pa5to|`k(>4kn-78xSU_b5gQcTzu{8Ets4G!o966@73@fD0zlKmFT9+0EX=8I!q z^J=H%EI=8!`cy^#^~H=QLkqk=ib;)IWTp0h=%S!3>ogw7H5(@`3!ftN`TR}F@<*`p z+$%LPeQAvE^H_!*1HUWwLx(aam=RxSDwPo{Z6au6=A-3*~tq)s=&=3uLsjQPdQ<=)G&ruO%;ZZR+94iMrGTM(6O&*bIAd-YONN9#qAf zXu}uG^7u5(HaQPtsTCir&^&N@g_ZZ&@4=1cDkb|5kF`sIc)$txS&gojWkZ*8r#Yfs z(8VsLfDqz)`2rs?Nnm4z5g()I#RERt+TWfy%SL>87gQ&+&-lc@g;of!ngka52Dhca1Ag3IanV-K?9X1PLn1PdT zJ(Ba#+o$Su=Ww|JV><0sqS120bF_EJ9@ebP^yc-7c9Z}}oj3K$jDT*-Gk5f!HPA_W zAH7#t(wnT8qg~=^UfO~J(GVAmJ9b;|^p{O+X?W~2zT3Uj&K>!QC7}A@g?Y8naQXZ-FCKct z-0VJbI@ly6IK?PMqK3Bc1vbG-#)XxU#YthuxQj;|MnctqcKH5E45YlFwoTwDRBLCz zi@nr)mwQ~weF$r3E``VMFW^E4w2A)~834zMT=hRDou?;ywz;_)m=R&V>+b!hBIvhQ zyl$Bf^+FkdSIygcl}2$SBx~&RZ^gLVE}SZ7$(L0WX6jX$r<3D5?Rl1y!cW!vxxaCXUw5Sen=3v$qWrYKuM#=4iEgLU_1y>v@)K8a(Jpt3`|H?Z^nS9LYi(A!Vf3ki zBlep(O8>C)$vnnCh3-I4Q@i$Np}Q4Sv^GLP<9fiudZa&JX2Y!Rt}Qj~F}OoErP#if z7j-p*S;Wi#Y`E1tTbNC-tSKgEs@!+{D31aVgmPo`3u8oUJk6#4ydS76pDl zdoHXd?wV=TNbaqmDQ5IJj&Ev=rg>C{kf3({U`WMi7-H0l4{Y-#sTe7;(I)<7#9P9q zwVOs-n&!MZe2nar!RhhigUO}JqMW6EB!lv1lq4Bl zHSmCwcOXz^VS5XGX;gl$QKgmt_8q;vHj=48hZ|8>eGSESyjne*680auUgYtn(A2{+Wh$USK} z%-w_r-pOmRk0u%Hb`|>j4@Xnu4@>3{_1OMJ&%Jy4QUfnPxG=?+$0fd0N-=8*Yj@rF zv%Y_a8(t7&e)Hz}HjKEAvHDV!o*X50yu=X&N%fDFw!;{3s#6r^q=2-W)?+G#`{ib8 zD0n~%ekUa;iJ9KJMPUz4?;9N{B(8t+W-kOg_*G`cDMe{i!rz6q|GDM==Q$vuHc~#H zhkBlRsp0kPAx6m(6lk=&!lfD(2_`7i53flv%OfM{tmD8>*W%4NpFbb$qM&j3F7_!d ztmNmOZ4s!S8r0dmn>)bSvbKas9liU+0&e(A8& z)DFxUnEu~JEs}0_IJ2+ZP#TQ?Q^`b($m!L{-=gzhR%x##Ts*9_C)EmzbG+Kd_t=YB9=5-SrX&6MRS&$ zob#!^|J4}63G?u}a=*ye%xZ!Bx>A0Qw?xC(<2>s1!2GCen{8FPXm(fFqEhlAX={74 zd1_zq=8^*YhHlDMZe+#=n=~r}BG$X6OAjA)<~rmAs~!*D)b=H;kpdx{-f?{={Q45r zl)R4GG^oCmDf=;%gt)k*c;Oy%Dv{4hqs$o5vU#t~YI;w?eDX{3se!P303)zUUd&J_ zt;o*713T^SC}>l!O79mgYBArlUT8je3A@p17^I%0rTlgwuYs`+K~F>&rgf{znd-KDH5IFFJM4NT5u)oA8FPc+lmZa3xb| z)NjHkp?q!@oA7=P@^x7QW2eqFM7Q}(xR|XdMru8#pP|0>N#wA7=j1XUNM||ZVSkU$~}- zU1y=&vXpAL)0%0TW2nzjmJ&nq+f8FEu^T&m=DOj%axJ>YTyH+QYbk{!a9b*K-MCfR z`8{$Zj?o%7jhL0#V2kLQ2<}pea8R+L{8NZsaFpXnaF_QNF$6Bg?H0}c_Aa<9e#v3I zmEyCm)7{j4Co;zx_9> zI}GVeZ(aMDcAN}#ucRWMT+E-wyPZyad~<=hDf-0iEP?stK3KS12m_K%4eA<9&nPfF zh}btFU6UR%M+0tbmmM1O;Z8YCVPYeaOXIB>DT( z{=3JD-yF&_<0?HH))U3~_`{ZVQ{qA6!MS}DLQqK-(k4jQ%aXP;!e;^B?_o}J*oy^H zu1ncvMgF2@{uPNzhkw^c=mZ;IkE!M)2)tg(XnNsL`e{1=@r)QOdiDrGm#Gns%zzx! zjh!G5eZoo5E()CcPH^-A)4agrrnD)|H~hCt-{(Nje?e+9IlIYw*t~5-Vo&?sHsC?( zoFGhlv~qM$_YwVkTx*y^UuDF$P|63}xYJbTJ#Fv?{Oj;hm!MZ>hZL&P&hSaZ5ZFBR zb}aZsa!#(%)I=SYKgR;CbQth&e{>rSJk zI+efi+$BiK$FKkAzP0xkNGC%0dhe!*eq9xH1$gjVe16n^%=; zFmZ@e=!L*Gsg-~O#ZfT1ys%Vvk{Vvw*0hBd0uFgao9w#Loe<8fhDZNML|2(Fg@M`n z$$so<=m4I|Tk`V3qf#J7aqleTeGT9k1O02WZ3no@A*cWM2g?z$g5q38Grdum%r4?# z(miKatcCyj>$kXz%$UWz8FE^@GzFVJ*w*f4Nv~zR5QPI7}{=G0V8<2`JhhF6YKx~uNwdlkC8qU59*im8d(o%(1Nwza7RH`QemzoLtnzn3 z203lF{_v(H_tXHGT#$xuZol(vWZx|=K34P9A!R{w$!v)tzcZ>LAVIgf^i{z*Ddzt=fuJ4e|) zV!bt>6G*-La!lF>85xs(W^9QtO(?DT`p0D-f__Cy?h$@1xxd;lEY42Nn7$7QFZt$mDxD!p`xvUHYZP9OITf5 zlw>3{cn&Ab9JA;$)f2~Zy{BtC3~@VxWMbpM2kv3%=3K*UWR0ln0U_Y^;Vywx9w<_8 zO~{u4pn4IA3vmXx}?7(vaA)r#yZ24;8!Ct|OK1gT6-(5U@57Z*gEaJihF&0C* zQs^I-D)gs~`^Tmb`GeSBew>%6>>px#a}(EZ?;R%mi-smz?4IRexy;lNS|auO)xnft z@3WkMj`92-V@XrPVhtKdd>3-kL7-vq{n$!!|()YkR#w|m)U|*S)8lWemTYu@d{57Y{OZU1oTy|#kUT@k8H0)xt6KO zncr~q{oM@S%pyynB$rn(Pvn+o6b(;IEh~*$!HfQlOl1`9bkZr_L&Va6srrD>`ld9K z)L~MCT}0LF4`665?5(=5#vi~XFp9fa$msYh7h~g<$TyxwF2bcVJSgFF199)NXdpJ`a45HyZj7t{4`~gjZ_s3=pB;xrQ6d6tYCZy9RN(#q|+|!yr zmQr;SK)H=4Vsfo>Mo8^=-&S+pL0p1EH~?OzJBXJyGT%z|fh(Vts*B?S#Ec#h3w=`8 zi%NCv^f_A}&9D@Y87B4K|xp9BR@4KAUqjjS}cBLl5|7GmK}G11?HJADFabe+%{ zl`#}*O#qFF2fk;6Vxh)GJv`T%`l6n*&1NG>&129XY8dt}c=MNOU;jlt*!HVz#qDAX z_;DNB>I~`wkAEUuEB%rI=MJ)By=9zxWY<*oCzzP_m2xfBU4;ETPAg92LMN$}$6%Ku zL4WpX@a9}qSsUg-Pn5l~SIV_SA>5}RZ!$ogGh8BRgcL5(waO^>HqZw9npEhr6F3WY zrze{F)-YN`JYAI%JSjW!`JhS6{>q>L>C~W&*yv0 zs60k5>JyejRFyA8^CUZEw`OgFNb6))z~szvyicWPgEwcPf@x(IZwIUw0+*BMH;zHp zi!;G5yxgA3xz}p>$33=r(x_|6N@TVRy{~tlAnOJ`X!;G9dl$FYZ2^4KcQu%tN+0O; z!kvZ0fLNtrp7g2o7)hdf6kDW0NmTLCHdJSvXPJFlzh{-0U&{~2#dBN@#~=A zzQdQD$kp?=?_84wf+SBNeC2J@uStp+uw2oEi1+JtVN5$E+8IT^YEUw|#yDc(#ZCVr zYt@s$%Hi&nRYv#b1{^AkL0|)Ra^}`yB4eEf%z-j$jr5}aNvoF$?$66#pDY!s44!+F zSD@?TK)A1i#BF8DFwIHf_v@evD}nH#Y$d!?KP>vekDL|q1I;FCWfud`jGV_+LI0?? zZ@%pGSrjO4)(Dpy%^o`KXF8Q>5AH+`Mz&T#vnXg$((8s_l zwwL447=8pxQ1H<$tBJ&7$Jmi!Le$0d9_L&^$Jj`Sf8PC&P?n%(<4+^M`~wFrtFpHi z^81E=Z$sbdQG^EM{fjN@@OWpMl2&WqC1x#5fFkQ|Z0y19w5}?FEHQwOgWCMA^UtTh zPY~w)Vzm6EKlc%AIV@?LLWYrUH(bdLPV#+xbI@I{3HlzK)a3{!bxb>-77b29FN8b& zUGjh6<9i$`_4tSlC2zhAEURynDS=#Q@8dFY!cv^Omh}FztUhIj4I@btN~#AWA}$$D zOgp^5SLGvNaZacg8GK)odo+LhN!~nv@D#X?-7EI`H(X`C1RONxxWXAOq4&u`P4oU$ zSAXApZQ?<2F|Dy$9=LX8oUrEhW@@Xpq~bu0$WhF9B)%iq@e-=Lh6^s(u>EG|s{tbI z>X(Rkp10x2p}z#Qcc!8aYRZZK_J<`4BVN2ehLD{qTlCzn2gyK{W)08`xz2qv1$M@6~`*vyMJ!U{eb6L`CFCT~dC?1WS6MfOqkp=2^u&-`o|1mnf_{rCpHrB7`A=UJt<{rY*HWqf1`05 zUuG{zQGJV)o5n5yKb_tC*k95_(^*H~$ zIA1MU_qgo6>=21_`9UPL4#r?T80QI_<~|!sosdHQ+-p7~R)WJ!c5e;P3$t9-DPfvx z2g8X${g&({D(cbVcam-xiik$W$XnjwLhW7CF+e$Gko#M!XMFr#lM z$LAGtq`o4&;YQ}x3Ox)v44xPE7kq36+_Ks+xAurvRY46~Ypiv-&ti{7rUZUF<>Xfs zggWZ)+>4VDxh}?x{fz~htJc0!urx+jK?Ph}q;4ReH!0p?eBYX_(Md-%MV%Xbd@su& zkH=&Zi29k`oO9WK=gb05o*3pnv$~I|9UH&@gx@Spdw03E2KwG_rNz8|Ao*+xZ$TKa z;Xtcm44#oWyj%IZE(3dFKMUbUzLhP!ugC2#(0>xHLc=qd5_!8<1vX$^>=E#All!Cc z3JOiG?C#s5yo46$>;}Ju%5^<;fGL4=1mDV<`FLeLH0L*KNN-*+e`nEmz}-=cq#50y z7F`L5Twn^@NmMW;gPa$Qwf6=#U4S^%N-*s|v*mD+k^Fyx$c89bbTAZIlc$l}0400; ztxGjRvJB%pNTRQTHfHrXWBsX6lJaL5{PO89`r`NP@5!0xxS|8gWW}Lml z4_<`5#Pz%(%ODBjS}|)#o7UKhxIjj;Ytw`mud4kVyd60jQXT-`63X-I5f)Evzmn2> z$@ta2_3e?}M9+pBH6HWc&I)tEM>4Uwq2f)f6vzQ%(1>?)nWu4ETd^OZ=N0puyg8Hi z{5lw<3XFh$gzFrak7dUgAvSD0Z==p{7bnfj`9Jqw)bEwy)HS8jj=Rapeao+@IM^0p z+`T73UF&3>k3^LAeEtTBl~Hr9Xl!%cJNq{{Hz7o1n^EUSp10>&S>#@h7;D& zgjdhUj=L1bymL>Awzqo{f0KCS5iUVPSj3n76^)C3e(dvp9QmO|x7J8gVd?>a z#}h=tyV2*kJ`;|kcb9naz7z9oY*uYfr8g>6H;?=IU7JJp(_5L97e3z#-Wqp{UD zp%s{@>{eE#Rp$}>>d|Lo@?62b3IvLG6}LXJkrwyaTgL7U`r$M(D+v)&PW-?ANKKyU zTC9f0&B!;SlmOC}GAxb4LT=F#dWIi~$EB2WdHCk=o$`uxY5pg;0X)mIkHhfqc0sD8 zkKl~}$A9t$YPSm-aOZy5qer8L&Y+Jyy9ya+2$$8uXCM; zeD1sS6!FH0uM6mG!J?Ho0nR@j57) zqMZ0W05eZ(euLs+&z^exsGGVZYPb_nb&H;7e9`LbQB|o%BKL zA>`DxRBa&77mrdert#j8&tkp33|~x@bg9qW%{7Z5ka?y*-1NB=liTl+_7S>2HcCkC zCU?tGe!wPgm^N`sU-YGGE*M8VLVMR=&26>MGwc_OTXe1Ce?PbXfd32j@V`Konjo00 z-bgR^nfFspH*WvJ_#d;c4d;qAwxo8R!#+FivnBn2bX$!vhYN25lb;_Z7YLOfR26(g ze8)_4u4N~@Z|}jYLUN-e%kUM;eEuZl(#tJyvP0?=OeElo&Z7b9ZkpWw`AJeN%3GbE z9KDL)hw!;#D2gP6^>R)48&)LW&As;F1{R3uwC8R-O5yyRdYa|Leq)_N#{7@cKVpRp zLDlZ4*v6~GLNkw$>WlLaXaA7iC+&$FKJdx0u4*}nJS7q9X*Sr_^GIbL!uo2U$#-Sd zHRcY*!{*bLHl%3D!Kf^poXG>{p~^)DPEHtczB z{=xoeaez>j3eW&JM2m^!=HrFS>llwQYX@B=VZ`51RqL{{ea@Pc#OZUZfy%Qpa$~c6 z_4oIM=cy{Uyg3y&+N!UwY>F=xGq4LQq<>OzaIVwKODw>V&hvp4$0|?1@USmipmg};;i{HL*M-^{){84oqx9zqU zzGnh#{l2>#?O?T^a4<3gKFBkFieb4wEdujCRXl`;3;bQ1xYWn00IdW5cfiu!&m&+v zhs31+ag@^?k95&m)GWT8_T0>0>1t>p-RMSxi3X@ys!(0oe&_v73BNS#r;EYtzB;M< zv$53)QK4i-3fhx#MZ!b)K~XSM8RlT8_&3jkR|*a65$~L1}#BvC0OG_YCmhu_PH8GJzzvJ~}b7IK%lC32BUaFHs( z&o}xgQG0i-+XleJgcb4lO7cG~Ke(2eiq8I6rt|@_B&!#)5mNcSJyvJes|u>bzw*5S zl~jN-w;Ho?*^a^7;1(0RGu1(xEmMER5mKJyGi-S90-6bb!>S0r!svB&e?9){k4_p0 zx4DIYE}=1)gWmAE9}v}xpYdq;vo_mxBYS7rBA)c^%#Q=yXHS2eQxx6JnE)hjNEp@h z=eX4J?lbT!&9}~Jt_Sa_=&8(c`O&%UpO9wqRFP?*VLdn)L#w=Z-GS%wT`z*CSj8g^ z^xF!Xcn{dLUvN7j_Y!|sgRF_bQff~Yd#W7kAoPz7^XxlnNeW(*f|Rzojj?{z%|Ug`HK#O(s87t!N2(EHQc-!iKdsr4*TW5Xs|0%DNH}qHWQTX!>8zX4`!oJx+NQJV zs(S&*)o`-|&v7XosE%&HXd{j4YoQlRDeGzfsVjB~AS_Te*sr+r>BfrY0LLHs8aQGSGJ?|S($50Ca(kP<$~9m< zJ(**Rm3#=R4Eh?81#nvNJlZmr4V`M3S2O%>n0I6c0TD!vU(|m?ctko8E%xxtJs5Nw zKZ7gIpHI6|XL}iYi1dCR#tS!e-iL>N4Wma*hqZ-40ZI4VDg`PwF5f$BK{Y1wFUsv< z3K`$SZ00;u?d1~+vuY^$w}>_p*Ngg6Sdt3#!Z#Lxpt}T)=0RkasR*jn(=_8XY{g^9!G?ZoxiQVam$+>Pv_t;;Oom2q~3h_l6rt%_RKMG-c2ca zn~-$7I4ZS8(~h|i%eqn9g`(s=FW24j)bI>e){g-puYej!tkPW}+^CI{BX8fP^-=KC zBV{;YVJyNl{5jLvcnypXtxF#9W-y7?c|;@qj{(#n-isKrmbInaU!f#S86+ONZc`F->3 z>%ohrE{>G)1MQfAVZ7ru1}t{+y{$Y^YDb!$7FjISH(LmlAJAcAArU*+80n&ARy-wD z2Q8Fw#~0R5ojx9=UuD(O3gXzGup`r2>4dudd)v zaWZu6ib~fAZJe-|xSzyv|ox(@nN> zk5-n>eKTL9mpnFjog1wMFjNie(GXw_+xmqntP`)V7tYq ze#7hqj$Kd21YwnMcz;Lip=0c6zV%?>h5ySl5&C4KN9iSu4D zWRUXFjl^O?$PrhKK|WzNf3w!}!ZI-JQSd#!JIJY}18Vio23c|w1r@~Bm}FnOR_QPr z!>MP!|IGhsUvi~BCJG=ufFG#mJ13la*v{LQFZ<0x;F}*65_yLm(@$4Ck)yf!;GUDU zB|4G+>=xS21?4DIoJ^3}7%4$#IE*`Pp@iLAhkH){H{gFYkbty=E6>D-R^1Mbd(G-U z?SA;2JKUZX(Z9rIlp&5z0Xi2ue!)XFSpLH{(^tVf6ATxiBRI*g+^zRFr*+rCl`Tkw z`QfxmS7Td%epZVvHdDNt3i3+oM-$~tGP!>qi;GAm+j^%o2T{3bS& z*hq{(_Joc3{~Yo?^)w3l#v-8t|GY}q+>77$r_~e3J?UR?qvJa!$O2^j$G3bh|5v$j z>!$X;oC}c0NF^=tKVYyFC`fnmt5ZMvrKVRBbJyQ~=pkBjUB{l_~w+8RVHIY(ojM!`uo^x3CWe=Yw+ybQR*4c05gXiZM- zna)9Sk`41#`ZF9p2@AZ)52f+;w7YBiqvW|6YHgyq5ZmTtWl!h1Y0~xu;VZ8NyuRJ{ zm}jbib|d^|Af%gbaf1t1PPWc&VFoU8-nmGbHmAJ1`$e|1kbYwqG$M$cVCy)s)2UxC zs40ar&stwI9TieA5pB)G&E%SW>K>S31jy+B82Ea3x`lt!h`Bj>LQ}5UfXOu@G9|&L ziw#gKyrtzFe^r8@^!No@QYy;8a9T1SxA(oDZ9;fjwxJR;(Zw_}%Dn2r+G1D-DRYl= z&WoXiOy}a>JOr9wd~UG-sa#5SMmJfW%}ZJiAF?Hr@n}Zg7TPuPL{vd>P@hxU_OJ(< zIurM+KZljhplZWO+!4I4=0Hb$Ublv^<={3+t=VCA`Pssy# zhfrrMW@gLIRv?#xB-(xf;;;P^-_&CG)9UW$1Gd|^$S-d%rg_u(!YqL8H&>dQi_Q5yLho7x8ZmCaI5nb-rk5o!Kih?EzKaSW*4ut@3pvzQ%jR56e_XrfSb(V3 z-N-b@T6;A)*`%i!ewUyly*62ZlnmhZ_20TE4;k*Y_=x0A3q~e2DS-snKPm>OfCz2X zsj8?Yg3CX2aaM85;ZT< zb+e*xMNKhee@Sqf*>Wa>%)&^~2&P-Cg4{aKTdB&>?OEF-w>d_Hvi|K8ImXOpR%$$< z9eD+we)F0TV(|P+@`>F*{n>jb-Al1Uxb@#K=PX>(L%4~F8rXAlrtkWEybm)yvi@HE zuGC3>imfKeH|w?CnwQ7@#G!+IJh}x4Ok@ix?0`n-JAQH{WlQ!jSmJOe6%Eyn3h1Kl z6xs&bRM&3vy@tD3BxHUTuJmK2`Ybvtz3=3Win5bq!+?Bg00Vf{_B0z8?~`M^~ zk+Sr!Pk_xY$oC}y=Mhev1;~@nYa`Qh`RCIOw5X5u4au4i@zr+UP1m$>VH&TBKC`2v zAiu2RZv^!f#H)0Q8>NS(!kuD=*-b~;y~h&Q_M}f5X^Y3xXrT>_03zeFz|saf59p$1GH{24-wnY*2Uq34V+XKR1Y^s{@tZWGH%vR9U3c#-fwNW zh_A>qynCc{JA(3?P}6m7Bb~6=hZAsnrW#FjTxbc53DfYE%Y6a-$tW^MG1GA6V7V$` zd1<$+CHixNMasJ{xFLzM2~ja-%sYB}8(Jlda(E61aX(8->q8j9?}zpQV-F&zfhn=3 znhvyWy7R^{z~w)owzuwk)z>KFve2wa@TO>`|A+$bOEP))^@r9KGS=ZVMT#-@iRX1} z2W+L*gW*x_|x|I_Qt# z$@8GMMW3eOB}$#zqxLC#1`-cp54S&cxaTrVj&7u$l;~16%6xyrJi4I_v?8HWj$*5s zT7Z~DaHy7a1NIFP!GPVavMsNp(T#69ZGP_Ar_((1kXPVE4Q{L_Ig-8>+k1bLTTyb~ z5jpxazX_1+pl;RWLijR(I|CNZzW`0|56x5n!;jW+03xmWK|-{fa0em;3`ZrWT?IE5!&1*`k-5Bc~Q#r2u!d}3?3(H@oFx9$u7DVBIX%>{Z83&1?sPu7|(m{}4n zf2W^QaEs^JwCEOR@}MPSsZ2%QraA+fnqf6qqxUhAe2AJ8mJ^6@7UHDa6XXmNz;bg8 zB+ys{(wvCF$i@J3)!kKU&D5j4k>y`Y<;aE!id4)&EJSSJ96kjomx55^C6+&a687ucpbOHJUC^+RvfCVbTkQ!Z z(VIzy-nsN?f^YN!oS-B!VDLP z+eP=n^d`s)G|WAaRuL^j_m=BiP-nN=3}CDkO`Mjcql9D4NN*|{?!6|ijDH7_CG~Co ze1BwXn_vP1r6PUmT|Kmg^L6L;LC&PhVe1S05#0L{D(8ad6b#H4q6B%TS)AItrp&2) zwU}78N92)*c4egUy$TxJu&*uz9Gu6Kyzc@3ct;xgeDw*nJWTxU-fWa*RxrXnhJ0Wa zZzMVNJe9bV7#x1N-JQz{6S#xW7}on7iuz2biK9o7x3g#1j2ud&wz6kD-xQ=Pha1;A z(K_7|6&zUxW+wN|p75GkgoK(h8Plcr1#62tI5zm@CRwV17B}wOIXU{6ucu=#rt>%? z57;iD|31J7B-%iW#X_90$Dcft()-~4He(zTGNe~j?<{&#!iC)~L?gYfIWD2MMFCM{ zYo(r@@dT3-xlUGSov6U;A1?rxS5_A}>3zH=dsK~)Xljd<_*;5Kx6}Ui@mtn|^20EZaj;Qv(Knt-4YU z_l$L~A(WWg?3wiKmHo!iHp4$V7D@uEj1LPO?m6s!QvyBC(ByiV4&9jTpi3eXTUJL4 z;(>K5ub*rJ$KjJkamRq+zdoH?%~bWkLEM%;wCT4Bk;R^WUGowpLma%fu<_fGfv##! zIvpO^Br!_ctn4Rx%ujd87MC1j< zTKcNQ4C4G99QyVL*^gi`~bBpZ|HIhRVMYrHXm>NmlNyy7j9iDuY62U&U=oi7!SPx`F^5AP!THn4t zbi--=Q0xq%{4P7;8Md#1+&87%;@0FxWN+FrKQE&W<>_$cJ!8nWPPonu3W&GbFiz`# z1n5HKD>@o6vT~0KX<~sE^E$kZJ;^IT`h}mx7c28%BKtiTnu1-6d ztO&?$Wynx`1^v55AMC5K>9GXJ%FSV$1qzGD54Hh$#>~N3g)Qh)U@)P5|UX7eZi$t zw%@bpbyF5e>m!&xW%_t9YX3KSm5Ln&pTd#-gi%KsM3G?I0tfS)r6p8VL5qmkvIDrv z&o6Ogwy3!et?%JKAajXdU$euc-^g?G@hK}4lV}?nMn_ZAW)>IY6L9lg1Kx&)&ZOh7 zlyimonh2a)iYMtZm!l=D|388MYxi~je}yn7oUpz}vk;I-?pdcwtz#VxRar3esX>*S zA?I_fZ1`l6p&h7hN6ZCtVfC>nc!+dNdWL?5*8=3v=~7i-Fte{`QOX1&{;a$TYnSuU zaLrC(8gBnLaW$d`a4jP*&fIh7cjw%>f9=`p-6eay>s{-4@~pjky+yM*D_-$XR!z9Z z=!uF=n#6UJVeLaI_DJ*qoe%+kOa^h)br4ww&?9vI<6-HX4S7jM{yuwj0p8{-G zMpjR2`n6MQ^6Q&f&@K>SiHUy1`dE^wj3?LXs^Omp+X*IKj1SVP)kq%I$f2LHvsNz1LUQxHwHk~Zwf1CsmpwWAq`@BCOGPUt2Es{-LjOQ z9ZwEFNqJL`79viU*wN0yHzexhpQN zTyk!2Nd-R40rwT7ZFw%iIQf=3-2X*+;#v)=CvKu#)CBKP+*3Nvwqks}$tbu7C3R8) z$XVIUfoD7~%m;2ZSj1`?{5%J3$0ej=J-5<@&tAu)T?#q9%5RN{FAsHf(6&Y7KVtWE zJdbGPFOi#`8bC;$5>nrQ4%Ep}&#W8o&@+X74<`u8KRqLLrz}9To~$$(tsn_(d3cOG z&i;5E{&OqAs11%M9dSVpuRa+>G(WcS+F?^4^^Vi*v zzx`I?)y3%w;O|oSvp%2vFDjG+$AC(yb1gc>**0{6W%?Bs&dOp4B~J=`4TUf3(w2vA zc&4DsKgR^*TGk>vqfvKr3UNa(MmB+^wy;kaXpV5xn(`K@{4P)EnTpLg^lt!`i7{m7 z5z-%MPZ9eT z=JEr%y^zPz0}P?X(Z9#=)1i2AManTyPGgAm3tsJDGve~#fAxp{$POs_(1NpDl8qf; zknov|!@04#5gL+ddbw1B$G*xxboro!b&939gjB<_5GAsg>DkMr$Csed`C6T zKcYEDa^Hc7ed!UhMoXlCbq_iwwu2z11a{~cEuFD`K`i-vGa!jvG9B3i===MM2j(e) z5je#IWX-E?i$NfJxp`zovf)3%lVpSiuwVyN!gTL(8s%|I*3^J$d(SHy9+t=ziT5+W zzqDU9;9!c=Bg&WH-{FQ~Ndo;0m;{DOLC;$+!3aQF=;eGa6mcXTZV4mKx#i_*{SDWVU#*?%Vqx z;b}WG*|+y=;U6o% zg?_3&4z>-Uu6REPB({ah5qvuYf6sxpziIX~(JE|hkJuzHyE?DZq;RFYD+ih0IM*1t+m4Pzc)U~yR!rgqaE zhGh)UL1azB*HIC90KVf^58OK06rb;$if;{^1|{RkSxW2G8CIH6UbT;+0!+b4)@p!9@g9DK*>Lz?+B5w;+y-;djz>N@=@s? zOYK*$uo*EY;WMCiGm7v|^gzvnW(9eQcpDxg4eO{y6bFw{o*)4$hauSwPp`4tUBMWI z>Z_H0ooUh!5Z*RK_{KGJMrz?k=Droc0tJ-rY%GMbLs&K^v1}dE75Y=>c zKeC3=D3yL}(9upY>HF^#np?8?RP(hH4*6G$`VSsm3Q?G2XJ5Jsr=P1*ke~aoZPThx z7Jh1?E+-G6TF^z8_r*VHRt3xohW~`p&DwW++jtA#fTmIl8XjiK=_u|)_PwY67l~^juK(-0G{`sSfO2{=l^un;-1n|?9*5Kf|Kz;w(qVc-IXJ2 z0N8EVqL@3$$%`7=icC9OX!LnQ=z^{H))?lQ0lAc=G_*KlqQg47J=eF?Zel(zc8fSy zZMaec*+s1O7vQvUm1vQsIGIX#QU5X_tlJm|Gqe1o3uHbtTO0Sk1P84963qkG@fDzD zpzg74P|J#31q6b)(W)h>R_#UktIBU^5hz47S9RC_lNa>rW_y~t_2EBF>uFMe-GXWA z7``gJWDV!q>;@<~VpbEd(C3BCp&daGB%RkvhRiNHLAtPJb;#<6`;z4{JRFb-tJgY2 z-0o%>S4y?OlH?T$9Rp$4}DPJ-zwN^WB|wPI^JJl?0g6$K3(O)_sd zCs@V}oG*h!(8PTkOr9;(?Ah)0EmC2cS}XHFODhfU$?Eo+O`bsz*CqH4g(v9ViPl_syBx-c->h#M z*AL{X6IgG2X6xWAXPT4GWbbg*ctq-q5$%b0Uy6}lZ}BsSlbOSL(2BR1H+~BAqLvNf zY_b=-&G|#xMp4`bbqXo$9aYU7l-Q{kOwz}rsH32gUyc}@vd$i!z2fch^5anLJ%=Uo`EJI`cDe@&+ZOv?B0v7@GYKRa*UR;gg=^-qahX8gKDns(45hh zaWp-CdJgXNq#oGFn|(@V0<2l3f`u+%iF_0*-v>74n1pQCMC!xmfvTh@~SN6efr8Bc>%My^)+b3 zM%#3?(Jg$O@NYA+)ihxmq zBNy5zs>b~2-Ru4nMoV?1%g1-C3hl!`=?IbO=_0euI>lHeze6gY)yA=bioKCfK0N|+6gX0^y!4S=gAw}|)eKCK z{!RH!xCuyFv2+xGm0KLNzHkG(-ClubV4lhfu+NCq^4lwMn}cn!<@~2Vt6iEE1J|z* z!u1f+UU&1b9nJbTGt+q+yQEG7S}s`PC+5H#GxllFK<@Ivbrdq(?CjI|h7Fg&wQEss z8`ep z+I6f{jGB86(YY3ukhW0fz>|Ga3}mH6>O0Ck_T3tzA~FV@QJ;vmrX$Cx@SmPDY;o5` z?$cs4O2Ep?YDQ=m7u$~;P8#RR9_0I_*s0O{MsBOY+!~dUfnT1G+dGfqiT8O|fZOnd zgB;tc;#R}-cVTE3Q=XNgU2b#W?CR(G@fW12hsaVptB0yeZ$O3Qd*dcgbISI}Z^oWY0j^LWVzlzkI`8yOto+~>9Yn^=r5-q1Any|M~0`@_3b zAqz(`7pj(CG@&9}76KLjwtpRPhjzg_|GZR^i)ef7icA0f*cDCgC|p?<%0-QON-*~o zx5eXIB}%Xl?FiFP<$P4*?l5XkgoFt-rr*v+Ol7Dvpjn5T@BP_#utgY^aa@fV9b_=Y zy;-{MG6%Y=jCY}L)C5H_yb5#Xd`TQvGp$sTp2I1tw8i`aN?T8rs==gnRs8!7#s2!I zcNHQ~TzXjIh^=fyD7n&*o->B(I>^urn~a3j8NX9j!WM%V=76EZj@L?zOA)9E+W;l1eMDq_k_B>_(fbZzlm-wR zE=@*XwHr{-A#G5!4UqSgBWD8cxXwdv5eu9~S>LLqQ;KE>aZAkXJ5)IPjtIRZlu02B z>czq0cEop)GAbur7`DlabQEpl!!@+E5GHs8??w?z^+Ukp04~6C`s()YyfF~wyI8f31>y6Z#)Sm|iJ#mQCmLa3g0qmQzVD^PlKh#kL3fl8!)5L6S`;&PitsLi!Fo zKwkD%(oy>${ zV=+rMa0bY|8BX%mjMvQ^lh6S0<@QsUFwZ55B*e6}DZ$*oV+9qq?!7Y}I z$M#;zs=jn=qD`j-LDd(h_dUsel8&Mg$taT&a6V1nxtij~eFyj33VYBH)Wu7{$HzPhDhj~Gie$Vc*R?#(w_=lI6&kd6oXlY3UAl3%GRgpYljU1OuAS_aM` zlx27<;9SbSYI&iW73BFyt^bRGI5OywTosIP5SVZ0!?t`_2ZGv)1{i^S3^Y(bz2~vY zqIQAJW{e+bi|Mb%r*yzVqDDh9t_^WxqR+O zqN|BRvYMsu_zPaN(d%jEC$iT?8d1=`yHItl8?2aB!ij8_+PvBZpY7H~Dj_mkw!x>t zFL!tno;jek*{*`dQL+RcOYr$q@&#DyCb;NS=@950t+>R`?NdgL&{q%`id4Zbu7%8hX@W!L_i`AS1Nqif*m4}b;yIz#0$NLqm0 zSM~-3`t3sWeW0)fb`YWg;`DG6!0>`Z`n|sv-D@`O4!?bz;RB7mxw9m)u0?ru|GzMi zFOkz+naIp!w(DPle^~j;(sva#9}#!Cc~hPCu=+)s*tVl<#@ZBgxA55@JMgQM_SVK_ z=EaW8KJU`v4j^D0R2b@V@vfCuTUl4)X$G)Gs5 zG;0%Gwl?kN#-;bT0vOS!c-|!4QoRt!f*zn@S^D*Qm-jpwJpj|h_)r``%~l+^9kIm& zQ?K2&m*LJnZw&l|FoQ`{9^)2~4v6}LcQ}3OT0*v@kQgK^OGJhY6_xw?En>!Q$!o~^ zpEB?}`*dOt+(nd6TeNDY^X7IYD-mXakaUhstTP?_3tQFF{2gF>N>lrkr(Xd~Yasyy z$tf(Ahx4uj?}$aW1UlngR=90lvy#vQ=}C)+llXjL)KOsu=r=_3^!JT5(Mb{Jrq;TU z-GKDWGCXz&I$X5OLIU<44F}cBJ*1P}ZtwpluX(vqN@K)2jUFbflk(D8;pA*B9Ri{# z>ZeB4KN5RT&2D9P$Q?sBCq7*K%10D#C=bw&^Tu)B*c7{5U?zSlF|hLOLyb`tuYrNx zJt*t+D)S(|@~!^Lx;AFZ_D7WCeFzlX0p(v;VJYBkL`jo(7DJ|zl&^klY;ZL~h&+-u zHrlGKyLG5)<+@`d-$+f@A0T2H)TW7Cx4BPXw;PSh9utO*kupIR(&hMF>ePBK-UIL# zLz}=qxa%<6Yj-Ner}IUzAJ6udE@3j}kTFGYhTo!D6O%2GINfCyo2mK%F~C0pDw*Qy1Dg` z4w+7rfjKwFR@L4vg|+3T{o8W2^}U%=59@SFOSV{N5Wi>7=lxHfiRhjmz-x!k0j~V72De1p#ML z|KT8^GLXgpgie+P;T{qRz2u|<%BeeVq={1KiUiRABroo?yb|4a(Dc_xbHP=2o7}ju zDx8EA-B~AqgUPS<2$lCVvaOk#{J$M3AMmyk{T6*IUJvw$L}8)v%~u;}>)~;ci;b@u z=U4f-FAW4vkpsf_9*v!X?a0|;a-RT`to2WFMLV};#d%oKFr$j2G63TKWH=gtVX%TjTUr;V|_oeD4b~NzwbfT6jGwt;ON=iqdWA1yd|0v^z2dT zfgKN=6q`+RuUkpddvLHB+<%JT6El-Bfc=y71>H4B$>=x16)81#s$9Mz^tT{TzC&Pk zw-a`LYo(WK&H8ef5EI;#5|VIX`jpY5kf0;@ubAIIL>R4WEh9WciFEk;+e&8(Jv*+f zn3aj)tx4*;8~$VpB$`;M^J`cE9C*otudnOB6TUXj7G`&#F3j`RwU}fNfs|e&uHzNb zC2PXK6HnF76X!gJ4b@JW7ZOpSY1?Ga45}e$){0rXQOvRnnbj!fxt`m;+EN1O1TIq6 zrV9kMnDj~9+l2B-6{AiO6uP(z2NPWnsLI2TiAM<*9BW- z2hf`(R42OJz+oDM5~v2#wrsccX31SROZUIkEc$GOLdHSp76Bpbvwi>BI=$OX)Xj!DEJ&t_(jE#Cbf7RMA*nM-g$V_!KQ z<@hx%J;6`*w9_VFHn`gMp6+qyKuTrDV-_Fzv4g^I>QLa;E3E0P;XP?{8p(IV+cDI)CbO5VHo~WdpKCu!H_#c~5xI zlsqsc1)8%z0l0*onI+tuK7NVYc?gb_aj*rF3a!M5kBRk|&Ow%c1D^sR+ZEu{qUBJs zCGz~bL*P%?KPs~wnA1!1j_vLE-HyHIJwG?ohJQk`&PNs%|x>*l-we%47 z2PI$s0P350-f&)mb0oVx8E!zip`@x9sNK7DJRosmtkyXMG;F>UXfJTb-vqQ~UEkyf z2jdG8KK|)^P=ya_?`+&*5J^ zqESlid_j&~reJ~auh*cEja)ro&04>KvD;Us5Oq+5yPkK~o!B<8;&G!^oNWBwn$xcB zrP^A3!@p9O*v!hUTPM%(`IxZleZl0trg-y@efLz-9nZit^gN6oO?%tdIx;%~mb86k zuJ4ZZ;(`BNZV3ionV|IJ@E0YE^rT3b7w;LklUTzzbza9;#_!>bT%jQU)ID1SZHvG| zTslLy53K-DK7P?An%4od7?${iq$Bx@{Prpl^b0OcFX_!=4OWDCzyCSJ3-#%r{UYD|6pr9x=%3q1b1`nYt)f z_MBNmzOS96;)>a3HWzOHI7OCUv%;AiIq@9*$4I?E_*Xb3U3=)L>(=^nHN|~i;#U%L zNm4EyAD%a%lHND^X*O7RrK`^vsBhKqZ7g}`gUNxxXU0QMDpH|;KrJs-nD8M}sGj-` zexS^PgmDPWFXX4aIU20O*Lf0m?1K_8PHxUupswl2fBwl^bNesfMa zf1pmqSAjRm$q$e+Cu@hw(;#E^PmIj#rG%S+$6NY@Z{YEA`1@OHi4QbP1(zcX8Zu5% zIRBn_9BY_I$dWN~eYn9w!lW({W~jCHGM8=7@QQ7Z*N1v<$XCX2_xdHa>N6AzHLwH3 zMa8>K+C%K!yU`-?HS*?w_>RrWA-~3?l=>V3F)p}#T%)>`1Lba0I6om~R zvCtm;GoLC@11@pDO9IQ^H!_>wr&Wr`8suNYi5W;8RO3inYdpS5AE9@VNI`x{u9O^4G>y^~}@ z@b2-k*5|ocobRYp;2BsTo|x+eup_xbm#ifO?BJA0pLse1B58t%Ie@+Tdo??$hVFuJRo?~AP{!Y`~Uck5E$`fjc9ia_8sr&`mHLVU^eoa(V=^Ft8k<4g zgz&BaCpNDpsW5w%cy^oz34vY#F0deH(^w$JIJgXL{YvcsnvAly$bzUFob=&LF&qKI z_W%lkW9}Fs3#@T3_ID^H@8Tw)DAvpiemDk*^}jLdT`p7{wghsR2KAS23t9q(yHH`K z2eWyJ&4n7(Pj;b61i5(!4mr5sP2%6*vmhh);cwIx7_JmQP`~90OyJDf_dWv?!~%@L zxzuk3j>qr8gv!yZ1IwhOS_b|m=yJVXCrmukRWf=<8r<`aoWZ3bl6;sXt@T`N4bBwE ziX*%MT~XLD;7p%xw|g7{n4(BcZiimqz<3v{yp7}O(uciUMZO=Z@V9JA`rBtFyrI#V z6e=?V>Ju+j9az>>p=X%g3v46VH}J()5u|4@;pkTUn=hwPNr}x)2@+OGJ) z!`{;K^Avc*kbPR$tk2T&-bd&1=2jnbpi)1yRIM{AzutE%61}rI<91(;st@}C_tZ5udMq2_t}UI) zx1n|#rzhyS>>K6Q{(Hb&FH38(k-%$V9++Q|Xa3pT>e^bw=nFc{y|tN>m>|=)JZAm7 zKh$8lW51biqf*oC3;ehICPfhmke558Rk1kEa zO}U{hf}L`cGe6Q2o3C-D-1>I!b>7VvFhRpWFuHRxt;XV9#TcK?L zS+m7WBzo-i@yDtXpy%7sX9oOPQ4~DBrQyjG=mJb--uh&H3JUa6#qi2JnETd;gR!va zk`4haC|2#@SxkB=61E`_OM?$_Mol($n(gMrur6((JR$fOeYJqM8NQ z(q-Hlj}Fv{vVh><@b^UJi2x-xva_OZS(91N^C$<0*m~a>HLoEQCnBbCfBim{>Eoae z-5E^zC8u}(D-#Y{c&d$%4SkQhM%fD52*;T*FPyBJ2l2L&gK^2c=0Sd^M%>-#9lXiV z!TnXXG!;pN#H|+=pJR-)OuxD{Y2`=w&(bDxhmS0VeoB)W8!CfWKjW+t))yo?plV)K6zuN;rZzxa+{T$8ELv* z=nlknt6NMyc-if#8)7DA=X)#+PuRBmwaAKd5!KL}jq!4<3)f;LE4Paf$@AL_qiV>` z$E=EJ^i5gQe6dOj1ev}@^HRUZv7>m!+Qefros9Q$yX|P6hq_ap5Z7#Ja{6JbR+C#K z|6h6czc8%;CRC=*>qC+n*>NKEr>|K4zr6T=Ojn?fan{a-YI05L=0Ts$bY|H;r(&SI zfEA1V`6^A7Iq+0t7%YB-uB{s-~vncG-#{C?30r6 z#`f!b-s^zs0TjrO3luYbXViwh4{UyCCEK^ozl)IV2CjEOF6$rZ(gJUC+Sd-MJkW;Xb*y+Cy18ds3$GOidzp`il#lGbFag)uGbP*2mBO-@!_2)2s4E8Xw z8+8+wiXLceljA81Qju9~C>!Y_{#F7mdRFtgsr>c3@>SX(?P-uE_qU};36#G_6(a!> zCg%N}>WIOEp99234jkz2!+_;<6XW+W%;?V;nFA#c!NSF}a zQQIp)s+0A|5h^Al&vRF^w<)19uW5-r^t|$)yT{MPiqJi#2`wYZ_N)MZIDKV*?+*4{ z#S^E5?V-*r*wM#IaL$?YZHq3w@`7NoYE+h7K?+_36(wLy_lYE#Hg<{`oZRybto1tW zh3Fo%yH4$oA(Q>;!IsYR<02|-IbAFs+&GgiYUu)&ll)qh;Ox@}jm6rpfSvVcOc34= zfW@UaRWHRQ_~rLQGz2{x!aMM#C9mp_WpbZ;A{P3X)^}}k~J`d#2%iHW*Jkdw@sK&H-m@Q${S;Dsr4xIjLxsaV8b_2GIjPe<$t zlHEMW_|_wpZkDps2x<3dsfX)Uyz#sVVr?gWB-*kTK~}9+v=NEjRdeNceXvz8DZ{4o z;ON=?bK4a90j8Dl|M}^Hc|2fbhS;i9gOB!O=%fn4Jn-xdwDn)A=1}kYF9#03L0m-)M;}3+ywS14PIihG)LmgPFTwGKJvF^-#Gu)(&Bc)nu`@W$X~I1&_LD z+o8Reg}Ev<_FQ`SR@o^>VAnvXf6=M*2{Z{*7Ker2{vND^$0%C0eWJ2S`wl}dI*OJY zsl467GdXi`PvhnPbTh4F-HZnPz`0GZBI$lZ4jlBVuRASgK(I-o&Tb}qv!Q*IlYHC` zH2pzud_d?t&=si1-;0AqCpB-q1a=l(P>wGFmlei^mnpY#u+@v+{9&dQ4MtR2KR4yy zj%{)cLPx5ao{9u2E@7&Zd3Y#_)~f9E*a5CIsIyKw%R#PHg9P(13=H95okHY$$GM$M z_6C9_CxO-AuMgM1i>BLBmz|!T8F?~}Gx6;4p3=HddA|st2s8(NMN)Bu%>ma`GEpgL zR(qmnEV31HhGyB_>c80RAnlR0q^XwLR~hCpk-i#njD#~uJ!jtc)Anya&ZEYY@GFXA+JZ--MJrgAkY-i zHb@7h1&tGx_XC2~q2*g+^12?4j>hT5)NeS#&5!VbLn_8mcB;F$!4f0Zb~o4J)Y{jH z>SWSEvM!$PK{wsO(~uR&%`E={%>*83YF)-j$4+-#jmc*m!ShII@;|vAoWM`>8Wwm{ z(*u{ zHLSn~>o7}&ZB{q+YGV0eA9|MVPHdymO8dSSCz-fK5Ul`Wua9ZvlRWAo1S;C*zkF*$ zY-Etg@I}xOi59;t_tyu=n#=zA<~CU%qxsF9icvt33rQa-*2bMa$XA8Gc7Gbr*=$U%TrBh%_>S+7)pF6qmz}RU=55w?pNA*n@J30tAScr|{R`t$ItKa!BLL z8rW=dP0A2snrqSSbM?EB_XgyifkqDUI2lSDcas};(X{n(@AS~tA)~y?m1g6U# zf`na&XQOy!Q9posHt#WlTGqQQZ9_kV&riqIxc@qzciNICL_WZoCax%WJ8ytH74lF} z20D$-G;8`j=!fve=0Biiy7p0kCPD}({(;+5x#Isl|8N>)TD4@Xs}qkdf3q;Aig~FE zU=qng?6HXU1D7XcQJNQfJ?i&God5ppi0;+5S4rJ+oiV#x+Oq~VibAbuCjpq2_dJnr?UH#hf< zJH%gT)ens~ZSSi!*#n3F{O01C?y zs>)8Cs~odB1Jn4DtKiof28zy2f*bCkk}eaJzbCE6zKy{`kMtg9BTdUew2GglWcGNc z&1oBNL#&5sl$BqCa8wR@IMb<*{G!KS77wsvnV|-VFJAZ9KjVlIh zfIEF_zH^XN+n50czx(}V7Dqb__@M3oW7pWx?(&()Uy<#2X|R~N2%)}ssPVaky;_sk zl#jkfzFPAec!SLB<00cTXh4Q>nEo$Po6#0KukZZRi%Pa>m>xQN4KherQ^J##uPZK0 za|(A0vW}eH!#E86CHiscl;1pIf9yVhPgP66zu&18e&EMuR9WhYlNI9Je*RpB1)gR< zEet%KtKT{u0m}=i{IOHUXdD8)-@D+oK)pXfrSK*Ax7x8T6!-{wZkN?@unM}Ul6}M}tVP=#+@+m2>*<=^Gq3ONoNwq1xIFQlby(N0sur!6k zt=tb@>4#ul!_w>wj`&SjG5|jx*@}^1>Bas7f>*uv9UZq#|Exe=rEPPOB=gITSU<{i zOu?&-0KJ7TcG_{fGUkHF7l4Qe6+Ib(YL<>fJtipT(NyFIn|$LRw6lC<2rtc*AU#4) zY1=jiF9nE8(|QzJuR5yuS_;D}_xaT*qkOKt+~-uCwm=!|7Ev#C7w0-{+%Bt5+sAUB zQnpdPm4!#bqGNHpE9GHp%B)2-eI#u(8%mJggGZ4;!6$aQI);2J1;+^pZITV7#%0ys zYx)y2zG?|O$RlGm%1*S)Xz)^@0%Qk3Ra4W$(pGoIaHjTz8^MLb2uVG^d8%~|rcoRW zlG{v%r{GxXu#`IBe0*6#<^U=yk&hVu2Id&mofq8xp`>FVNHx(^8O?zw@7_y&-gr>; z5G_Rx89?vU5g_8lg56~yDhwbqWxN}59m>JC4d+%!HVn|Qw;@e5>3VUy=ZiPwa&O*V z<5S86feJ#&c?NVJm6Ov71%;wFt0}FyF2P!fuNDV7`e%bV9hQm(DRhBMgVYm^&n80j zC!eVtP@R@U2wkIcxq+sCkfy<3+~B$wmAB@*&Sr3Q%})n``I{Jq>3v5*`Wm=>KCiOO zyKhF?5p4|=y}AE>xvdOGv`|)vo2a8UhZyoJQ8uKC>Rao2o@Zg9V@jRMnL9s9>lsRz zi;yRHe-KSi!cr9v?~j&_rT#0r8TH!K;_hm^Ai%beZR{0{jDFxY2QrC~pH=pG!<6=* zf)e7Y zbQ#9{2g~by`MIIx=dB>w9Hgn46rE@)2l&}~zx@ka@UJuW%>AFUwBG(RAQ7-3E{;c-J|{j+8gT_E&wcTrUFz)Ub(Y zBl5_yGrMQv`o#kKzVCkD%vOMhdlFx#j?rVORRj)ik@$bJ zH~Ea7&xur{Ub=)yb>o?Y?=+;gQ+|eA(hkVE)N)WA>1GLz6#8snfSp>_Su~-j0M)D7 z6IFnpY(Ig2#VHkM`TK_w<=#$0*fq}n?f4b6My`jo7^!&vz~o7Q%|~7k|2{N({=0V} zKIK$^130I;W`DsamhX_G0&XS?5zb-tB6KG!vh)3upq6V?TmgEfOi%tLk=PD zQvCcqS?2~4VNc1npGpG*Xoo;UH16qf? zXUm>9X6W+~IpfN$E)Z`7NT|!U)joQWGH;nFLd#2=6~zEDZLNQ;Bc~0m>hs*s|61{xt%`T%L8kpFn~5i_MHbBg zVv!Hxb=n?#UQpZ@#+_`NmGemAg#0_H+bB1Ok*j(}0= zT8iN_B_a+T<;3Sq9*N6Ja+jh{YNP9{Ef1+47BMiUzVu+D&Yg4Wi5LBpDlDwrnMP_* z=cp&pUyT{RtI+Cb!*GsU*$H2{Y5`1qoC5>4^ACQH@7N{WMu`7(h!Y3YBE3(rLYVC$ zT%qnyE1)f%bARg_SZ%pzWFiv@Q#;tAK>IHkiv+u_=w7@ouMp?y&!24W$m0umG5Fe; z|JPN7ROV54-0sQz5&Peh!LX&QK544}lzv1Y7VO_yCciIp!pivQhOTupT<-3A3Ao@+ z)2M2VhIKbH8UnYF(Udu;8P~q{sZ2$)?(u4jPtL$A*In@QwN@CR?-b^7 zf_+zAjL5wqP!W(^5a4|(8dVkSGDfJ49+-suX?(#0kf$r`kC)PxA_inleJ^Y8kdg@> z=My}5%(n+snAB#{rA$9@zjXJ@JjnI6^-THcI6Py*qz|6e#Y!6jHau28$p5C^NO-FepRjs%B2J0*Qv(}f;L znifOW?DN47G?TEy&erCvQnntoUFcGV|PZ@|6Q40T*`ody}1 zBj)~qw)Ugyh!1!JuI`lL{0%j4{&bFo-;xrw?(o6EtZqZ@3O3`En(}xIGpUZi9=^-; zcBMvi8o%4Ny3a)4;)j%(9h~d{Dr3GVRUD%DbLl{O=WHk?8KRn3w+@dBhd%!M217vn znkIfn_oTlPP>?p70Br(_`IbTSCxq2H_}bu^dXP|}8y(c+(`T;T<~4m#92uk~_O8zg z8}p$;YjcqGDL})XfKK3K$X^<0PxIR|R5-x+lbSve$>U^*a3sXX{T)knI#P&;b{lZa z#}lQON8#a0q{fXJsK+7R5_a6R1NOHle@sGTj|{#z;L>(T^(gm=@E_tO=+ zFljit;TY&G{a^oln8%#&C=|rg($~=9VwWH-C6=PN&^NsrKO1y!KR^5My^)s0Rc@C5 z^t92IyR)McJ-dT6UpaOJT&=6qa|%x4i+UBtQ#$R>U#eA&}kl0$1KhdQrDr#0zJ$89&Om-IC|>7&9GsaefQbKk)2Kc_o{r;fu7)Gxuu zJ`-tJ-u?6NP;3>6mX^cSpTiFXN}fy1UQ*L^GNaM{)M zC=l=NER7Y0w-ncxnzcHsL?Q5q-ZvV;sGrBTLBb#cQPt_C{I5({2jkvE_PUpb-?A7a zoO_el>jZKh=}#rPs^Q1d?`-HReu0!RPQ3e%m6F~xGKi>`kuRSBL4uDn4Z@B88xxj} z9FcQ9D0owgLx@eM*yOxdrujI!;d^DF^yATq0BeZh;eKRz`{|ZHFwCn)yvIvxr`}rQ zTjpqT6c?Fop8%VNlN?;Up#$pD-LtBMXP$bkOCp3NOup zJ7I4fqHDeYKS030XjW$5`!Z)mQ%Uyra%@v~3?c-GY%ntyd52`h1R3UJ@c)78Xt3TB z#-2mpj%jp_ejn;de{9<7iLEdBaNW-ugl?S`mxCnyDqR#i18=YSK3Wz4b}XU~)#`zI zYG%c;dlIwZgjz3S9swJl!pg*@-^|`#?#!F_Yof&#m-B!B240YICBVab713DX>vC8g zt)_3w=b3;2VXPlH$n^FNi1k!eWe%7(uC@14OwM*GX|RRl*2)FK=IPoGbs6E+_Y@>Z zKFaN+I|pB%(b@#5Ax8evs_Jp0` z@4qTNg`eiDms~%`9@##Rp!X(V4#}L^lsWxU8X0sIfTiu^TRCqIRn~3;ojxv=gEakW zP$`jvu!gW(3z8wi7UOov5oUZW;yDJR_{FXvIK!6#6n`(?hoUI*8E&-65%s}_hfbO{ z80fRMzBS!fr;#~ds@Q}^y?tmO*r>lj5 z73@NvUiz`?VVvuC#3~9Da0n3WL9K(x#n1zcP4)wyKH#!ZX30E{Zh6H zXs#Y#8SJON;a3wE6|wsZYQ9k21$opv_nO`aiJprYO@)tma2ZW01(@OP?m-2*Hi376 zgi1NbVYezvsSbg*Z`3|{)Dx?93_nw-CO(?hGi*NKvzNIqy}d;Gs|a_^P-<>PhUh*G zEj{K41@`OMYsMg?sfAKTQ(~V}km;8&Wa+EBzF+UOGEt9HZ+OxOU>Oq@$Jh4`ffD)0 z<-5Gd5evmH#y;(Gli4QYj#R~5YWP|Js{0P^RI(A%mza=z*HH@)$2;Re6NaWBI*4oR zcG?@G#4`6JiPUEh=Lh%1^hLz1b%X&U@x_)O8Wn^F2Tz=*caozgh$zG&C<;(x8&J#-{u0Bg$UlHZo22 zUiX@&Jz#BT-7q%!?r_$N*M^ecT$^fhT6k=l*7eaSj>Q7QpBmV83zNC z!^0k=%o&)cWaDKopM8Te-17$HSU%H{N&a90bjyv&gf&_B0U<&c`Ojq@me<8noy){p zp@QML0%>nESc`eJk+x0qc`P9Mei)|-WKJvuL^!P zr<{W?=LdWNL@EL)>yueUU?w^}J)G%ln`E?$khJD=wqEsgs+^*|H>PSaR5=*v&T%#? z8#=d_FMnT0g3tcJlNU%+qs;kE^v+>reB2Z${kY8uYy1~v@XlXY0$rXo3DHKo>5TtjJE=8hL5$Po$(wlTeAoO0P_uhp7 z0s*Ol^co-#0?Gd3@1A$u_nvX?{qA?h`2K-0X0mqnT64|0o@cFR?u}(=Q+fAulwhl3_SYiSs2FHGdEVkxD6** za_LIhJ~@U$*!rzI36v{=;wMM)laRu~`vd^N%x+5v`@nRO5FAKgz_|kE zr#{Db3X9PMsqh%E&=UQ+V+wdH=~&-bi8lB7_*QlT(*B`S)|&uNelz5?;Hj@uAHhyq z$gcc2!aF1*J_#W26;AWG;69~F#2U%BtdZ^zQg331bH%aC-9p&me(t21EVA8yRf+Tn zcE?1%g=oGyNEevU!_Gspy&7qMf0Q6Kxr7%k-?y;Oa^Q){=T|k0nSLWRAfk)opTEh# zI0@Mk$5(8Dj$oR5yJ7V-=MeOIS(Ztm1C$Kaw#-Mbe0*B#krYbmu+6Q}+njL;UoCmk zpZ>yw45hlN7Kwv@(%K*P%Gv4=dE{*a57OtduPju8yetf|Az*g#$qnb88_tYr5~(-E z-PxnV>ON=lkvg1+uS{6mjYfZ|uN^ykC}EYq{SaNjg(+QoWVKRF0PhnJNy$vN6NlgA zM|1-E`WO%c3zmn~nXi`CRiM4KftFYwXWd2FgGbaw)XBAs{@HuEV}X%&X7ijN{^8PH zNNZWqN*nCjh1s1eQtn&mM<{Waw}2!#PqN=E2GQ{%ZF#^}*duJVxYbHh8@GI@TWd@f zUNY!x^SjWkm3_0YpYqowtWvuVrN0%@x;iDfUmmHNFG}U_vHRZCJ9H+jud@XR;?>24 zTv&3>y-#AcqrWU@QuR=Z$8KBHORem*QX$)?J?`$tY+JO#;C(xk z&yG)<4@Kl!?!=Vu`ai1P*mse7ZVhbkt5D;2fd)Ut9m)$x3A6TTr9S^BW)YN)2Kp-4 z%IV7|t)^#ya4W5E_zA*A;0ql9^%)pxeUR0Nr?2_JJTiAE>EE5hI&!fh0#lfV^!kzY zG5;yL`p8N%=k;Jdo+%QRpn>I?7^Ixw%g9vvl+myv99EyjezoKdp-mGIgjH6)k1&`f z5T%xqyOWWgDy=Elx?r8uOi| zq-Nh~DC^cK{xr%+`hhfFF>&x)QhnvbW@biye*&Vm9rJ;A04P$=e}&u&T8P&&ygTqJ znMK<-^3L>Gv)tbBT~Ahb4W7&3eoN$}d8lfLnch+zUDn}7T5Otztm^L+3;j8!c*N_~ zVKt^Vk2deID>H3YQT(k67>z4TC6EiH$fa}~$Oq10^+z;Dtq|67ND^o7&FglnZ4vg( z?C$17B*m;}CNb;Mdk}R``PR=1(!DCY3!f#?48&lyoasy4lu6e5060CcX-%+Rh~;|b zx{}oEc_|}#{~&u~8Zbk>d<^KdR{2}oP7>L_zH0u=c9G{1n9C1IKOfnf0E$AkDK$)9y;;H2*WP z9nj8?j;Dsvju^k~ygIW#=BC%fNRkmD5qF$t=XKQSQM?{6mFUKE_qeI8r_-p7` zaOv}KXZgF}+qKcRmX0XwzCj7x=?$rEL`CG-S|x36U;WE=u?vVuVV=d?qTF7r^1(mV zBn64;Hq~~eeF_4FcRpbY?hioP41i^e5)v~3*}mirqzNzBT+?+f6mN%hR>E+9S` zS?tonCeEKy5bU!F_;63|z$w7R7WVZLZhNe1%z5^3GK~ZiJtRH$qy9{ueG|*oTTj5;FHq--giOXHX||6c6CZ4u`>({t$I{^gGyYQ z3ymiGx@yAd)-a|%&}P}rDMww=k8G7}dnTEEc@Y(#OeNWSc1Pqld(e)B$q-xI@^JSt z-`5w!lX0wiH{EP-t5=lDVI#QD&S_KlVYZG<+k3d5OYu4q3nl0)pq8uJK&fkrP>VUB zdhN5&LAOPo%{RHv)ao!H7_KBXNEhssZQz`DbI04!69lnzfOAnsd|k}4mt%TUr}d4b zJbyt;Mf3sWzCkqFQ?!_Q=LhXPII{=or-813;B)|*Smv1wwALavg}d$Mja(S^&1CFj zsbl77oqV>othu7Vy{x>O9}%v|4lTiD7}gzHwDW=rpCe5?c5!BO+TgYtSdP2p`Dn4W zCkZV}7@DHdZJgC+)~Al`k3#uTyCkmK@efGMN_JF$8$?w8VHpcQf?hdpXgIp;^ct)h zP&F2cVewQ1JeFI8y*?%5I>{4Q)VRv8qyAI%2Nt^J>f4KZVsc>bj}m9m#EG?C755OB zG(~++KCdHbv@l8S=s_im92RR!B=>YqE?U)EO!=Fb zfvaH>75non%8fK;Dg>7C-L=H9>r#+|eFxrZ&OxoUkT=-mBGnnRva?9hW8dC1oZ-~) z$a&kZV+e|@UKnlpDwdagM% z4Wn9Fn{IyuL}AGxR!68c)}^C%qLs3PrdEh5!K+oNOKsKUIgLAy9@kdET@Tx@qT*p# znOoqAZux?aQQCNfb0}TJK^e9}PG%C409Pelaom2LadJ;hiePF1`I0#{_HmfUs=i(O z^?VHI!IuB`#{Per@UT>_Z>#GJY{S?B$1J&)dCBiC3VCc?KiRA(34Ixjn`(;3$8khQ z&2)YAY{bXnJ|L2FhT?%7UZZO?LddW(q!Gaf3GY39Eah5EaX&8tLe|3)@8R?qX@opt zb~QX)yLcLyefM@dT@|liC0tOad4+Y4@;&1l_W4d20!!Y0JSX4sU0m5waD-NJvU6T4 zRoA-7o%E%yhq|SljHp#;TGDd8k-5PSkLFur$cpP1kYHleP_4<7^*D(ybeer7!?*mt z|9O7(0%1NprlCe776*x#`VFPN{mM@LJnD$dH&ZPWVz|}Azb_`tKk_ipCpt^!NuB4_ z0&vV*5PRAdN>FLHAg*_TufSlp^5Z4EJ4fD@#`*Ac?tCNc!(0S^uUrO~C?8wwWy7yL z(vci;C%NFw3T-Ztu(Ji92Xc%(C1XuD{@HIJ6Z-mtW3ke#fsB`9v3k$r-X<)5Bg}N{ zX|kwtPZYTo&A}lnP8tIlVy!*guz9qAn1cMv2}@T|ikunGdv1q&a?)zu`)~O1-%S9y zTuc29oZFKJiyOf5pwDdBLJm1(8q#k~0Brbku3N9Eci`34gK@Q2SG<}7;-Y6#ExbR5N6#EYtSm-G z&#*@-eLcQ_s0`OEZ}#G~#AI&nfB;6qUwB#TA>W^4SYP(X7bjY^`V{Q*wg6O!z2L?L zWQisfiiW9^u1vAUhed?RwJ^OdUWriPFt4z_4UoN}q4qHFq!$%rQ=4%Cj?CE-a}~dw zK%hIF>Kx`rM~D`}y1QueyS+Co&*TjsOO=i%C2bgk=VkX2%0{6g?=!F(o%%_x#VhGo zv)obmPFYULne--wt!fWTg4gx>*SRp$Ap>^YO{_B5>TI=E?92w>AwJ+BEWYEO)6oEP}-W0F)T-%(lNT&<}WTY<;ShZG-1%0*z3QXA24K_>C9b#{rz3dkvsKAu|>4dAo1zRp|z(_-aF1ktK#@I zwVn$-auP8~T4zc2O>UkQDpW`fHZ5!;P|c-HQD#WrH_=g}hu?%eDuT9zgOreQ+W6*( zP}*TpoX5xsm}$r1rXvSCUL??S zr)fuEKFBz1`e}L}hu&SOyDR>W2Ta2<(vP>{Sn{(v)w@kiNx5pyO4x!2j5-sVi}#i&-kwvV4|s;TksItSmuDQl;a~^4a}3)OX3y#e;SMC zKV`$T=}o6T@@&C%Bw{w@?&Dl9PK^JS$LpR&t>P|&cKWSS6LyOIq0fNL&M*@R10di} zrUxhPcvf715&D8P99uxD8UW^kJjLn3&kRFf0*T)_-jE3@hCcoiz)Sb%bu=e?l5o`M zVW@32#38I*a0{;0?!ioQhbDH!tMO%0)sv07m$c%&KB?YG)Ggta~|R_pD7PD zwOM3T!pEZfwb9fa+%kqeC4&<=7h5XWSCbIuL)hW>c)*pf%{ykzbs!h^tKw-Af@9I2 z40T8E?yq9lVm}r7{WK)qX?*R?5Ehz_eou;%+RICj5qG|e{jR4^Pi?LBcHTmhrdjyU z_dJ>fpY}UL?!TX&ob8@QZF|@}ar$A!5WJ{fGk?Dj_Tl8`Z;2at>p;V|tFo%3nx%XP z+Vzp5`$XD6W}PKa;AO)l9AgPUnWV$kpTB3|2}o>_dZovNPC1LDD8Pl99kqQ)dp7~m zzNMajUr)T3Kc&O@u5y#AK(aR4itGZy^)paT29azs^d8N5IF|gr3^fci=B~{jZP$vG zX-2$Nyt?OT|4S$oREP%`=KN_lB_aDMl+HaP`glqS7kNU>4t=}hGIdy;jaNC;hGdI8 zsXOzRLXKlb`~p~L?>NTAKFp13Ne7b?ZGqIdp3a4MU4mpr3#C&YCH8^3WqN13pZQIa z-1gAQbRAdDds-t-6Rh2}dXb9z#kz*IWkr%@QS|k># zPLH@&Z!GZb4KDxboh(x&l6enFq@vyUa>})t?`QKKoEj zLtf-+7Qqb2nJf-;c7Y!yVY*L?VCYWe3>@60ezxk73vO6dr6hRM>a%SQ?eEJMnxyvu zDCvcSX_)^jv$-iHQh^&i&rX?g+#@$J53VJutB}WHfeb#HO03?=> z_uVXo%`7A&Vv$VfT?G<}Gs&7vQ(V_M#88D}J!%e3+o3W?a|61>z3kZ+lipvZz%k=S#z~(4h zSxWGT2G(c5XW*x&?r-@P@&+;;uC97;IVk=*#R^q7y)rSXNv9(A%F21}7`-8yS~VN{ znajckCuYOC4RTWSd=EO48qH%FDLqEg%NV7~)d9&s$EHwXMt)-rg|t2=DSdh?v`xCo;GaUFaSA*A0J~ww_0;zcUgkAP1MM;6&VwV^b7g__lgGDz9BcxaGMFv7 zTuR=rBM{gV61y29)URc&f#@%4s=+Jm1xh~@O!)7Oa|_Md&B>{fx?huN*?U2LC{&3@ zd-)~SOV0pIDT8wGE?{|(vJ~bv($~N7lgSj2>I6#%1LJ{J9azInDqmmxY*+`LdcR@O z7+{VIfePqE?^wkBm=gx{=7{474cRVe5vc^Yo>1XwWQ8wSx_Bix1E_Qv4F14tt@xJH zi2|;VNS8~l>{x`?@QH*WkCkn2Pb#UKK699EH#gq4NO*hJmd$nn@o0F{kEceTx3%I8 zu7E&aG4L#=D#OiEybP`7{cSg`zq@VZTBdIZx9$BVu^0I26SR9LkCqrDLf>N_$Q+J? z-yGKAssgI5D^q#_mmHQ*99-JvucGkS)4KCE?US(zreX|46&J}O(q3Ut^Z)r&_CHJ) z?a5&!`0C5_W?5R7t-{QyH%*ecLI!}c!B}S1Cg?Jv&|MqwwROg<^xsnK5$FTCsjrRN zgIU@y7kFP5U6U52MFQ6jW(0fU0KH#bGhd$rpEe!`Sjr4s6l~R&=cA@Bo@=jH8wnk@ zXW}~n(X4f%OF2DLbInQ&1dY!}SlR*`E%bsp5$|dTF;VQ3@{!RA9=hUYe(85h7a7N% zzMJUFIDCAp`*C_)qNURwOU_D_3u9j}q_5GXC2$x@d;^)^oh_>+Iy53wu_x>K*Mwso z*YHa?r((}r%V!1Yxv=8zBsrV#>|_=tH0)3g-7(|hX5z!S!q?ptFyfhcZgAz^5iITX z-`b;5id=gxl)&Vv-);6=t-p7Z>wAix{ZG#e( z2=O8s$#qePH@Fe&J1g_7Krhx*7G@KPj8Gv zFrC`KW(UyqQVy4}pJ|?wP46v2%M(a#_pKm{Dg4y&MLk+Hw$aC_nmt|*4jG-id^Bg` ztb}vDEp%V`81EMRmhKsFcqNa>K^0f>AlEGG&EicpzI?wK^^dX23mUZlk}Uj%NVDgo@)3$)>=DcPY}doF=5|yMTD%MS#tbXG#d^3QB3_!IA z6g9Fy`vjrlEWGQ3)}4abo;qgZRDcHn>vLe{+23>d67iu=x@-f$+pQ})reXU`p;lXD zJ8(w_r4l@K$>V;)O7IeBSm;B;aqv-v(dU)lF5%6A-)bvA;ouE-9WHFX<8>TBTUWP0 zV)WUv+H;p^Zu;#0M?_+Qp7P{xS>!k}f{+)xkpge)YFM^GcWlU}A;zaL-P*&a;}G%> zuWV=zW;pdVZmc+7TXyul&yY=zA01^3!jfwPo(K$s^xOEUsorU)+B2|SEHeCF&{zRH;j`w+82}zL~F$)*pryX<_-Xl6&%D{ zTNihM(%iRDNKB(^NL_CE(TpXZjWRiQXqNtuQ~*2_+GD#k_9d+op%x+y-BST%l;ck9 zbCp+L3%!esBcSW0@0BXC55=sAzt_Z6iO@<`cq)~`(kh+}cmf|YYeJ~K|rDXgdg;ocMDu0g7ACmP*T88KKP7FDDW_FCE zCU&Q<5%o<)xfEuSXPB@KFEBwg&US`kQ$J zN^~ra*#lo#<#+N*ol#mng1T3_uZBzUV?{b@D^3UaQFg0_w%)uDX53#xmGd?uH8ACN z_aT^XPZ{^~OX!B(H0f{JO#oL!{&P`yD_nNrwdfA$P)pOElirKE;^X-cG#9qJxC{i? z`YpMt^POs#jvvWBIa;uP_M;>rlfu8!Dcxq8;1p9Jp4F_=x4C-KAe*VmgUB0rQet{0 zrhrg+XJ_3`SqN(w-G)69qhGePt@=}4xPvhlP5`*hPlW8m3xGguKQZ;p!6v80Mopx& z{X|sF^S_U;dgLZQ+l_@9C*Lqf&s>~Z6 zroEy$IDGTDh^E!@@Dyp=^D%uJC_JDgOtQ|alkF45$5At{Tha|}bA$PWuH@mbn$_KG zuc!R`Vu}ryGlw0}JA)mI>q>&X!HbBKD-#eg?Q1S~z3pkt83Yxb?HGu^8;9M;bqhx8 zhP|jX566M5s}IzWY%!5EQ|-or4D0{HW2fN9@0us#0nH z0HxzkdCFmqo5ev7Cm`Q0E@O%bUHk0H^7WPWB`8V%hxfk76?AMXyHXjgv_Wh2gY=j?$eM=Gt4O`K49 zk2sgRqntWd;+r$UUDh*<0iWR0tdj7d!8Y-yY)Na&v4SZTG0soUl^0xmVU;+c9%K2|K7 z-8KSxejUNebDoM<0>``Ta}z^RFMYp5*QX%om+&Jhl+7+c=QAqPczs3XkQ4}!Z<%O| zKDvMuAr8Oqa{|z_golsn=xC@962c@TI1BLzZRU+Zyw(Y9Vk2Z(|O<;eySLu69qL#WMaqx9)Jb|^gl-^iz>#s+$U$XQ9=P#)F!p_1a7>xi@ z$8726aUkguJa2IOYj!}HkhycPR}%Z%jfT+``wtu$ zdnhI0s}!B|4S2MK%BCp|wRqnFe*!*^V}`PjmKaTK;Lg-@VMVszN?;!m%K4YfVHD2@a`Y`9>bgoq4v-=VcL`kDdHH1h{mA$!5F&`8ce&#iFG@K)_jGx`9;CH8Y`DrO4I7w^ex>xb z7mq-_Cghj_KoX2^+IChqiql}WEb$wBAG9CiJH`(!X%J@QyfD!)G{M+!a&$NGJ;~OnQoi)GiSd=M2 z$D&?dLGh(0E^oXbU2($)6G>g|nd^5tX)GhDrSbe@eclyATa>zqZLn5lg!@UmrBl%y z2z`HPLKEhJ8@NKYS@#`Sy^^(neOzSOdCvvT#|ckZviJR&80%U-s+P98C6|U7C#)`ARtQ9MLPK<1HD9M+En$V z4sy^rKU_l@Csv{9j}zNwUS#|GNYz&K(`f;oS->U0riH*bw;l;`wF9oHg4%>s*9u!h z=k3(V7f77)Q>|0pGF3C3#W~m_NLkQAvQMtrmcwdIIx5bd)hald+US!4`8*epJ@h$& z;WapHpkZ>WV#^b~1i*N!M&|_4xOW%IvQ_~LZtrpSjNZ1W=*n=lj6I1}GMtBIw##Q2 zX98j)KZOucFIxitO0LjC0aX}@S^9Va>h+R0el`t}b;%+`nu2WVyNnXpO}uu}3F3u7 zk0@{(as|NF;=-`)GKV<@o(3Qo|f<4?_J)C8DYlW1(Jy(bMFVg<5218HA=qeC$36I%H)mbAXH-?0FTjA8LTTYr7Qp2E%blDC%z))&iWEvuI z8})d;E$ur+Y<1{d3fKDqF25l>BLCfI5Y)^i=JcxM4RPvy&q9;{CVk|sMp)|3CwuDV zeEx=9MJ&(jd#}~RpW$3rn6DC*9rzRmHM)tfF}dwstIXYJO!;^TAF{dr9dPv?Frx7P zwr0@lThHG{g*+?i!pGiJ|C73(amc_PdV(-KmC_f@h0XiU>3+J>m|x=k&^zGZ$1jEX z1jN)mF8?}d!)VlTC6Ms_KM61vR)F2~8=qZ#^QwykiYd0{-5r1BO=2;<{{LBlp;*2Z zS#UO3V{j6S=y+JH-hgD>ev`=C*84Wv`d279)VoFP;^^N`(IgC_>5(;aFF03bz-&n| z%KoN%x*ylaNhC5c$rajMRIh;zSvn5&5ra2oWD{RfDP@CF)bHDs|N47;CIa?2m_3!g zbpjbI-?}Pb!&_d&MBQVs+UbO`RlYsKyEoTFpgDJnb(nceZu-&a{zmy_@@bGNOT%YN3I z0hm5EWA+GXeWVc@UZGdqr|P@#c<>I`FaWQT{>IiCBLar57rQW2$e;w) z2HlRWUZZOfTu<1C$J`|6g^=E|%ZW2$AF<(RsB6h9Ze#lz@)BWOA{{ z{zKLiP@R$3N&)Ux;`s;L-Z&5*FTgVteHnHV8+3$F0q7|Ft*0=r%>q_U!T6Zy%-8Vu zQ!fLp8tfhp_NFz<8BRhV+2X}RC^u5jzqHA#a*He(IWDN8gl8!~#*&L9*W#=2KVVLy7EU$jJ`O9HO=Rg9M1Wk!t|H9uc2e7YvnU(Ee^En z)B>>^mSl#>Mu3B^Ul9~18IC+T9ZgkC(QWS z>QG}{`mPSG#s@>&X+Wv_0W577DQ^w9OZjkG-|5w*74CX9y4JiOth<3Cv-ee7MWU&v z+Bwa~!NUz{bmK3}nERPATU44Z+)C8D>fD3C_~iR+a@Oo8o+e15{}3o&nS>ss3x%tgh{)ctlM%fv_(J-`qkj5mD*_dfIUGz*9`p@lj2$a8bCU4`5CqC#j8V*AZN zc62HrC-Po@O%9y}`OT{a(FozLE`!xb$=)WS z{70i4UD5qbev4V#SA+`LJ^*-qj)sj=RyteN|H0t2>KlajLbJLh*4k&jU=uWcfUa0dJh(w;!4s{7 zbKRq+NW?U~IIgL56!)OgN6PP#?lUuX=ZQQ69jJ7|Qesdv*m--UA#qHh4?ay-?LZs;mC#JPiHtq#Ui^|dfc-QZR@5_U;*XjfYAk7Y z@fA9UM3$l4h#Q;@n#oRnpHTfRV7mq=g&(voQ+2+u6Kohn8b6o|ZzhY-Ra=2HlXdOA z3zWzG)LaEp@s@T@Q0123FR9<^L!7d{tB;i7KbGhh!tQO_Bg{xzbBjEYV&M91*ZH?1 zV2O6|wi*KF!2D@*2(Q>_g|k=9Y{vQB)Qv9f%x1>D%DNffh^POl4R5y>P1EMdUqAf- zQ?GzZ$0{kBJlCPW^2nRnWlIFSQ9d|lyg58MkJC@j!@H5fYv;>Ve5VWX&#=l_>&BQs znP-&r^!R=P2chWyrn;AEI_xYIb8yuF_;Prq3aBp2F2gkaXHoKlHjlreWHCM0|Di~p zG;IO(pPfoEw>Ki>O)cKzK*EJ)37&b*Dr13#je8!W%d$h0^TChVa)rUro81QvR(9p8 z(*9l|;Cga+H{B&m{`$n}mTSkrmpKl}4}zDkUcC>1F?A6>#F zw}=S9ROj@wBA!~ub-a!BMZqTM^3*es(BhdN&oKdwmvFT$&b`Cg`!QTEv9{Ykd)v|` z4EFeK9pEXzN|*#^gnMwo?fexRn*#4I!loJW*I$6Y$hE5qWUqh!M{Mk-@ju1JKPKm0 z1irkZwX)k^H%}Zh1+E*1RCtub%D}eva)aa4FzAwqm zLt2UNL^jK0p9MPa4!fg%7gyD@U;ce#&r8{B07GNG1}9so0;G+HJ;?ND@zIRWhUF)) z{eENfQoIyb5wgLHOZfeirzI2JXas)i?TGO#?$c&-fS=6CZiJZj^pM`94)n}x_)qFD z{?nP`;69=UK+346@`To`+9k*#Jw6dc2SkAseLN_USkjExj8#OBnxAr zI~%wwyHWFrr}BD(+Yfijjn)-?zR_w!_Z zy*ZmD>B-@)Vd%fUr(FIv0;lcGBR5Siqdms$u7UKrXJPjo^k9(WM@82=S8rr7%-48% zF7op4iQCYa7&BJh(wA%BwkGtBj^`l%O+L-?B7M3E_C$NGsm%gn1>^I&L@;!NJQv0p zY3LX%5hHryFegCBMtF90ZNh>t!Mx*sNM8TSqnH^4#r-4^cdxX4PIQu8+W7P8i)dxD z6oSLD%ibR$;P(7Ced_j-dyM#fnmQK}M|6++EGaOjx~4zb&bYO4@DG!EGACSI6fm!$ z)aIOUO7wDRBmaYer{;SZ91Bl&ZoI8v)=hjaY8Y)HGd`QUZdEbhg+>tI@Vaj%2_%2L z_(Pq@$*$GHa{W2vys$BWPNaWw{%aDtlMt~>41M%3!EN%^I^P@YS}t8>qn3+ZkQ%SA z7c(E^7d;q=PKS0me96u7M>aI^IwHcW@Bn{F9o5!kA6Moqyiu7wy zqqCN_G_2lC=H#I-dJO*scxk_QW+Cc5IXL5M zl6B8$0C@YT;vOTc{-&(`6E#XbwMsm10m*2p$>vfT|GkzABWe6z_$H65Ry^16+t*0TZ1KGY%cJDg@L+6qnRN^rd)INP!p8HSH zux%Q!GP(lnQ06}qQT?K{ozWnAFA7{}H*bC$6H?kZUo0|nm3{F=5to1T9W|4YicQ72nJT}v6c*21 zV&qy1o5|Bs!rd;qA$xJ|I&3{JvmMkYUYfLRG2yKxECx<_Tp{?S(azU#!`_HL6Sd{x zZ_eE^#{a9_VJ&~rbg)3DmsgPJbr~y?6bUoIQSSaD)mEq*DQ1_pe;G1nOZ_}wn^$O? z`K3S^%x78{i$fgvT?Gq8JqWu8xsy6s9MayeUSAmD0Qi8(o{Jy z-49IdV|yoS7GE#6wPKN^r;YY<YTqB5WbT`{MM?AX5&x>%pHRZ+KXkp?ydR-13#XzMU?ZRjT?KnoqAVXBwF zvu(A!P>4zgm$Sypejo%QC9zWuZlujt43I~Jmq_^)BH2YiIv z6-o?zH4XXIKY6oY8+5uJT`g&^`SBIu188W_W!R3zV#Ociw#71^9&QJ+hI&y0J^{z# zdDXr;ya31#SP40eg2tL~ZC!@LZv+s!Fz{z^kC}+q%V48-A7h|U&U(3ML`eQgg^l0M zJU4gLyD+7A>I`w-d|?zvM#^RhFte=RRnz4v`qI|>X$zSK zA&VBeA&Y?ZW=Vam{dbp`hilXAtd9qy3niIJn$I^7P&ukOGZF$rd6vk%123b0fS)eE zw+%m4yf~AQsfcg+MHpf~!BuXb zA7xyEn7t9)3I+g?uRUSxi1ejrWlpAoAE*ue+Jj@cHUf$-V@W+6sr(F@@6To7vpB{a z?@ELNjfgjKT|nlMfA^T7ug_Je2%wiS_Lv_CYTxe9d2G*+`8B$Hkh=k`YJ-RJET}eK zQmfv|0FTL2+B@9v%d4^7W^OosGjo+!2++f1or&1|(Z<>t?)8 zaBsdyfar;v1wN~y+~N^#+Ps8TI3oZ>8pZQUhIo~NQLDrAO5g4O%Crc(8)Yabf<<3z z_MZE_o6q5>-`yK~*ebqqe+S<6Y(^!f5ayn^SJv0Nxz%(T zWI?X&Ax7US}@>0$l8W87owKPR)2$lAK3mO zA3CY7-VlJqoz>Fg&_`gvnenpeDDsP~T61)x2JE4a8|)LdU)5CI78mr?EuqD@XUL_MR>Q_K0CA;xZ$}3!shw|T7-2NPNTcr7&y~0UZd^HPSL~#`BJO#lJ z{1t=Kor~x`KU}W1qAA5dEDb)Epltvs%AVGpbhYenVOav~i40~6yOz`m)jyop`!xLd zXR0}QK0PHKvv}?(^$kl7A=`q_PB9xpLbd_9L(0rfsFcFH?u}USTv&?*u}qT|9?@p# z=^hEJ?#46p=}|mT`;< zvjwl(k?a~}M+xkIY7q0?@~3dszcSlSQc!fuSgP~JbOFJ zFfw9%x)XY#4US&xgpz;V`O|r()@KqfJqd}d)ptbRec`{OFH3rY&rX8O&A6XSq*w z^tAHARfT&RBK170^Gu}g8iVkAoc(^N?`r9(#Xgae1lJZEf65pC_$it~!8SYCTyd_+D_+ri*P00cXB6M%@l_Oqzs=ueIAB#1ZU)N3 z*G^$si)r1B2RfipIM7mE!&@l8<9ep2&9&=}8!_@!75UXvoSua>R4|TtXJ7?vpxB(> z*U=8*C#s$~$=2zUP`Nmvj{%k#2f`=pD-wxJeL2@G zE<;qsu582aTLh^E;t@aNFsGIPbaJTsU?NHwLsuTt+26pF6*?M%_bS6Dk@tOjiuy2( zWmvqj3zSw+zRw1CIR#TO9z5JbF!udy7H$>U;bOnMUipLB61hMY#XuXgBwUXVK7dXE z@^7uXfkBd097=X7oB(I+D8OAeqx$b@bNbE+R1>cXyw;2*e)Ot^zx|eCzwn}ikb43>PH!>*%adOmo=flV#$skFjrVOx zY0lC_IBY58rM?E8rN=q)8jEnolP*&NwR+rD&$~8C@Tq8`{Q(mLb?}Q0(6&8Kg*Ug2 z39Wmh@FM&BWV{Z5Wfx+4HnUJEF~@kBO1X-bR;OX>_RQ~l_{u50N5W>g%3B}UE{YaZ z7bC3DVp6qCfn2_wpAH_b!%iRh5;9wyt{79F+Wv^MkO8nvcI6B@LYT}*vM59Rm&AIe z+olFU0pk}GEf-KQx7Atqss&GHN9Y(j=yF0A1M)?DtJSWNLAG-e9OzfbpJ3T?L?tP% z_ZwD7K?Q{1;35)^ot`W_s}eShLT;Sxiw1aQ$`FZ7xG5uIaSe=Dt`oPu(gFx_-FR1` z>vq^3!SiA{XX~n+SeGtgdVOrJKtYxVf)03|xCe;01T`?ZzG0+E?xV?PgE4&!CsPuoUK*SmTi1%y88$I=#0MfWc|0!v!%U#)$ zVMfcjT=u73C-zC%y$7%2Mcissz?I0T{W)Tn#Z=)X*;mt{$zgG5#O&;!fnQfpEoL19 zN)97JvYkCjyB^PWjBUdoW_z~4u>W-_TOm0Xjz_%I?B&Ag2^eJS%N)T@S6m*`uXjj% zS~tRhZU=gTL+@pC>e`0u_tJ;TwV3x4*7%NU8yCZ1!@H}u5VvMag5pV6@&J{_`}$#; z=J>DmR6zf?hnMw?RgGrj|EghAxbDZ|Ef9POI;bDf6{}y>|1y-bBo7oS?fMVd4W#GU z+5$Yx2bszh9E zctrj+VYOwsY8$GH0=ma)f28g2Hr+n2Fb!j`p|?2L1#Bw%kQX&84n2bXf1l#p2Z|-Y zlKPQ9Xg)mk-n}FvyMHe|3-tVUh~7VB5A?KP&-AJWNKqk)fiCASfce_ufK{A@m{$2na%?NmEeiO7BST z9Ri4S2t7c^y*&SY_S5dQyYH9xy7J-9T$wp%=FFM-P3}3T^MgEFT+mi6?y((1_g(gm zdtQ3COpy?!UN{d8??amPB^}&#ha%OlYS&KEg6gptc1y|xmJ(1njtSRO;4CL8Q2`G5 zjrb$0+${7Mm1qrYN&n9oMD4OT&N-mzE889B+R+izQ_3ls@h38V{MirP4J@;$FG@Gs(kCus%OPcv)lRCnOWHvZ`CbI4uk z{_nqs%`9|&F}?=0jN>j^GJ-O5L_p-x8S4N<5a*z<93+yCWq7B$ExL!WX(Y4*O6Q1| zqJ99ApAT`;1_a|u$&QPR6Gfy9>OPgk@sbX26~eKDSOl>1nDIPsmpd}K}ZZOpTr*WB)iI_bJpfOVU0 z2Ap`Ttf=c@B)4K+eyyoA_Up&X-(9G|=aZGVZ=2nOcS)zf?9p4!OB~t6F@OYo6P8EA>H#466Rauk9&L7|27w@ zO0Rr6sW$S5NY=0or`35GMh^T)lPL!+mp0j>^>FYe6EZ3N(ItIMFOHSr*G~RVE971t zHdW#R!XlZk0=sY?dd?Ix9}44p^XI>pRKok82H2vp#md;__F4&T@P6Ezl2Wl|fKk`i|LhT?RW+c}c$s~O+N^OP30**0!X6Y)m$Tn^&Ud9O@*(7cOSRXHBzXY>{v}XKL>EmQ*cTjh@AvH z|J|6b(%c`Z+&o2`as>Bfe*Q2s3~&FyhAetRD7UDeZX^mFcaJwbuOf|eiK0AMiO$<#ZPV#_%6~i7sn7YR zIZROH&pV$-Wzc-2Pd`8iy`oMjYCwZ&H2YuO-Z}N6uw)8AHO#((`NOYG-(4xw{?H!p zm{&=eqnmx@X4AQ{HSK(Kl^;X{swQfU3=$}Z{HVtu_NLB1OLH#SnT1sNxlDnl+>EWc zRD)e3ZSNwd-*d|sDvE^p$W`$;lZ6Ju2!QWHWLDs_b>rs$cCAu1`_Omi=AG`6Vn@dw zcLW>3RHQx{9Y>^E6JkmC5G_S}|1P1dD;Wa9cfRmJvCQ(nU8`@C%>m!O9@B@%Bf6iz ztoIl#yGpLo#v(Wa0L!6LW5TqoYm*l zgHj8je^A_KJf9LS33zg*fq7Wyca%BMH&OxBS@ZhX2fT-!kCwIG+dTG~@x857&|$(m zRvp=bsEU`ayI^?<^N81By=bM;v=8ia#F3TWw~2zkN!G(9a4s!BC7ATUV?bN?kC{pL z0T8zY>tDE{ZvFq&qRIiAh!fwIx12Etk<;~rD(ccA7H%k`+=4UrX_&5^L>}ojvFDTM z+zOFYGhj(?z$u4F-JxAT{e5Jhh<=Sdc^I58xuWa3RPZ+QQSZN+7N!^emaCb5gStYB zDvrPwa_ZL=6NRfG!lnwVA|e4rxrAN0GTUvDKl)8E#Q$v$c+@&Xd`>=^cKfBZ2lyyK zwo0=Ps5_O6>m0TziTO?T0&5Sd=D6=aW2~qsBx6gfB9VB#6_Yo=5ci1j(o&jc4OOqh$AfO7fkE=1 z9B)T*5@OT-EzQ8*5GbBVl&qJy7Vd4li+38Rs<{K5PiftVpZ6&Ie0%?~XE%3NS}Et4 z@ccp$cTU0`rW!T7MJQ!_>cfsMrO@NOIwN-=aJzp_E@Nvsq4InnP*GQQAcv?q#TnUb zL?Qij@I=^@Fbws+tus0MbocKpxQPE}?W9IN&+pU#VmaG)K5zfdzJLKcz^5<78NFEj zBP)w3W8@^$%?G^GhlEu;C&Kgp(`y=_dgm_74x7Cz7`*PIS z+EqREb8;bc#Do0NB0uz22vhP&j)tB&UChjJ;5yOt=l{Q}uI@~|aCc)?Ao&r^paiz6Is*|P;iS)x|(0z{oTp?WzgETQws5KZBL;rm@EZk;0 zPa|!KXs!9U5F{OQ=oI$1`Cw49DctE&K7;#i|M2`E?ug#T=2y6zPJa=q6Pum+VxgFI zD|-uPvz(7Z!hE+UM^8yB9C^7iiN+Hsn0!wu1KO$L==@zyj4;Z%jiB4?9$LCW;7fp6 zyqAEP#4t)=Wj{hq=qFY{ytM?MSbZUmT6<6bn$M`<;z_c<5Ah-H36bK+_dj>lN^gUE z&&GF(o4%k%O&hL2UHw+L#qLL9@6uXT)!Gu$j$_WijPH114I0%5OO*Sg;%P;uO;k#A);sX* z040F46t^_BQ3LA}Fm(Iec0B2YlIO}t`-LSECs}h{Bp(d2lPcCEGa?zfg$mL6hjl_~ zC1j8Sw`8Blj=Y&joNnJ6-^&Bh)SiRESjV;A4~tnF*T)Z4Z|$0t$$^N7S53+h>AJ7@ zHdxP7$fw?_`3cqa2EF8O5L5(jisSJV6UkR7?ajKX znufb-cX%t@_*QKdWGSTHHXr995v%l!r}e2tnhUU?JiQ4VX23s_wWTS0idQE{XMnk1w-q z$Wf!&d#duPafhWlSen{FetCc|;Ww_=)+xUF`u={l2O2E`~0 zx=YmXhMp^*3OjEt@JgP`(vC-6H|#3^6@<$YtNxXV%Q6@*&d<6#^y;Uj7T3x@_6e|N zSg=u7UprVM?8*Z0jdrsNy2OLB#*2{;HV;|x1>O8&`hvc47;hDwR;B86BE&ZJ#yQXM zD_{xh`m(Hm8~{v5v1tEkr|w8ci9>y#n0EIb#;7Kj%@hX3%O@?KyJ`npuFrmgqfgkM zguWSZ*`*Q0q{N5hEzDPmYbZKl>;w%<+{_ z{hk%w7tpM@4>&*JDCB$q?6de}5ulxn;CjiKP+l~j;L9laEwt~D_cJsbv;&;7+Uj{_ zo?{%()^v;?flskl7;b^HqT)xPPS35-|0wQz`f-W|kp18dYdqKW8ET#VCoFDRp~+_D zB0z1@-wAtXh_DB|JpSAig}2Lf)n?M4vPh0MqxK7S0x?~qagL{z=^m)*4bLmYaUtKG z?DemxzCLj9;_L^V$7oKlWt-?4jZclOo9}q*2n^avR&k|u^Vmq4e|#tTi`f@lfrgv; z#o4^u-0IUuNf0XZOpCZ@x zOaESZuI#ppQ%jj`TkItRJEN*JPB)qd2|awhowgh|t6LUkq7tPdzJ80YjdtyfGbzcl zh+APm(_>O>x)VXGxkcf`Nf6BlFRfSf$CCOdWG__Den|hO@Va`mGT>p;y?9EzK}&!9 z4@H}Bw+tJU5!qdFLH?FS%v>w1^QvT6CszxC@x$X$)nj}(U^$QsE8FjvNLsGq{0v|WfCF3iTKensg^ zBp{14(QLioQwENA(lm6KE&>T^%xJS?Y90;tM=p7I1qlVmf}Ypu?4$S#kygWr*2C+0 zCE`6xzDNqM?HB%h5(~5Q-*~4>!gUYq3wzCJ;z3H)J+;;eAdX*)t6W#zh6iR5_s-pG zy?*@m^G$+W(?~ttH)Ebe1-d@46X^l=ZNJL=G@SW1)lG_0T7pxM^Wec=MbMy2nN6>@ z3288=vRoq#yDASVzrnaOq}5N8u6^9C4r-idT3zz??_~r@GgJy6(oOXhgc(wWKEn)@hJ# zzN45JINeBmHrB0C=f&XI@b69yy!}GIn_tifJnY(nQbZySprw~#n zYWNn(7NG^f?-o^P@CbXS?df_j`c9mBf+r5?5is6rvc^{H^(c9iO9l5zF;q^6T|1mr zV*TR>P+nPhI*Y1p_iul18)DJU;_$&LQyC44BuM;?hC?E}O3OJG9cS_qT zXGjkF3>{~Rn4&xgT^zQXgI~0S@=5I9$3AK!i>axm2Qt?8B(M^`F~-~%@U9kZ86Ijb zOF0jBP$WSTbV25Brtiz=i1SKMBAt_-Mn-K_RMRF@N?MW=VFngRxG~1eGeDmuNm}lef1e%$LOV8}3 zzA;B#w1}BHyGSowk~yjy-&y0I_cy{kGCOaEbWJVi=9)k^J-}7 z52pe+f!OutwiQB(NZ1(EArN=5%|N`4iuBDF_}N4@GPY) zXIFg|5km9fB4F^zwI(FD6PJQnWhi}J$`6;KUy~dMyVjH9L)XpB`$S3;1S1!mZZ%1Q zSQ$v*2{^AZ8OOdKGR2iJk>m^^R*4#X075{$csCHNsE-WjSepk3m0^l*YM&PmDRhpz z_F5Psp@M7^chaA?GZ6vQe0wsS9blRTX#VH-onQ;Ok2|d<*0Hs5--;Lno-2bm%0!jP zA_*q^4=za3{w!J@Co?JI%{m z>J0>y3bi4ZVI)BVqi=8njS4Q3Eqk8f|o?vEK-b zAj6Cb)OHwQQlf2}@T5@WUd(Gy+CKNjl5xjvtMQ!)JTF4jsCI2f=sipq{>#8ac2(Fl z-N%3RA1+$JejHeN47v8ruZzShCqifW5?9S#mx4^DcHq-VQO;P~g#P03s2WG)Tc~Vc z`-2t|Y{Qa$yeB`oh50jK|QX-+%%R$qs{f(?zZo-^ZZo{Mn&+zH;e z^|xf-wvze@fw|j_wI0TpMiLZ4Ek1M*AfXdfDjJm+0k&Qoh_nxj*WD+yM3DNOvxY3C z(`t=6bs(oEnPXpWtc~}Hm@p|CacBmsiUtG<&#J}rHW^b}*)XFBxK<>=Gbd@5uxJ96 zp_c-h=&S2zW4|jm-c_S<*`3A*WDtp@-oxvnOxrOCgDI&j4ZVfI$xO$?19rTTX9!sRal+g3IW**ixxc}d&vY!&-Za$(k|E>$S*9>&jg z!9xKs60=BpRuR7ZBus6(^<9<(+pU`%YV_7@okp6`(D>j39hQdJR>gK7{EMg?s%&#LB!8IoPZ4f6cj& z*+TJfb(fHU)0qA7fo$AW?bs91^XiBzJP9tU&0%vSo1DtfNAfcTBA`)v{o^y<6#UVR zj#CIqkyY`3%i|G{iSN}L@+)LXSALH45I?PW1Licv?oxsG9>!hmCYdTy*bd`X@JSFN zxAZu#)FD-6T!oj6r~Z7pCcXU>u`Fv&qTS%8?p!3o15e~RR?=Dt#!m~Q-5=`MEtcL6 zg--*T_RG7K@%Yt^B=$Ar-18%Heu9S`k46}!p&^?cZeO2-SMQa|t+Cc>sd>uD^f%CJ z{RF;Zka+l&5#DZ-10Pymk%g#g_|9IWF>*I~lyNjFrJ0o^;u3`DV`N>BriWfWHF)wIteVo>I%r=bt;r`lXhO20 z-Zr!LP)gG4lMe73pB@ z$7mRn8oRS8Nt@cV`yP$U;Lao629q|**Rh=Cby%&5lea3{t*yft0o!l_BtWPW5OuYJPhvuw`SYU^AbD4cd z-va|Tj(nj&?e$ud`_?YEGJ0lO--9_K4y?Siwja8yXo&e=%Jg@w zEevNA(Jr^(>-&7y&rjb=x`%(0?qjc|c35Z*Z-)m(64QErC-?G|ctCbBY?Ea1uK-UV z{3+gxk4cQ(Xy_EX7unVpSbmPkp-DC!=03S_xv{l$8(XG=jYNKP1#R@JkWTm@pC)%| z6yVK!ecACUz%zB`uwx%6{XR;MGd~q?^qMp$Yo$zuq=iX_K>qa@nK@h?TmQ1ix?Cmy z>e394?FHPOiH<<>rE4l}4e8IKF`Ri;7A;%v@4wM9Wh#`jemgM6zx%vTk`3T!nsf<< z90Un>@Xn{T&B0ap>{~XVOeLneH9=#+l|6nf1IhSzo;I=T(AR@cMr-Igmr>-9jO-%k zqyJ`yg4aN~%DX}8#@;fOdQ=(%W9<~XW6Y_WAspAo6vgG5T1B!vt(;bUG@ID2oEGdc zTlcT;xA*sRKFcKwWwk@X7jR)|ytLX_{G*zQS8LW=b-DlQIWA+(hR>F6ocont?LN4h zSGk$^4o>J{?)fhLQJ+hHo)f#79D~*avYE)_^kz2ZuM6_;7}eP}|90GT>fzzH!lrp3h_**7g4iKDC7>`Y#e^I4z;%0p-^eyQ<&`; zAs$u&6HuiH`B4pJ!?4mtGS*qn1(0tbxU9_$;Uvj;rmkPLws zW2;T;|4*?pQyNIdnK{G3)I ziG6y^5<{6zntPFqUypJJcvbL(m3e>Nu^HapVM)U;ETgp=DkSDMf7Xgbm{hSobvpX` zIIO0#SB&v2%k}5CN*H&g1G`6;LbEcJ>)D;GtabspUr;^^yhTauzzc)!b7e{sMroW> z)@OVsX&~iz{XqSffMqt!T_AQ7srTFDah1|tFybzcnOxXZrtIlx1!>HrV59N!-fsM| zy_tsf&V8!A(O=$=U*6P!k0_-k>xpR!SPm=&4)+@Y;r>JK8hXipjG*NbJF5z8hnMf1@H6Vn_jN=Ad@v7Jk5zl`Of#alRz8PG5*T+Aky~29 z-TIUlTpkoiq}Hq1r|pqEc=&p)gULZ0GwgUDV3wWit%gBgyaE zBE3!0`flCYE7zf7!wPBLwpO&|=KWplcpOfm3Os6tC&ApoIRdYOQ5CP<|G=G`$HX|{ zs2eJZNN6v3GOnQ-iI|Pxf1K@$A{R?tx88ujE)wvo-;}<@xP=)#WOgUwD&+V6eo0Re zk3R~`_lBuivmZ3thFz4wxb+=SzwV+5J$tWgpK_W!akvpFj`ow&($`3Yp=k&8eX@)B zU!9*fM>|%YZSCE$eX9&(fuKAsXxXU`+rdQ)p0y9L0-gng=Wn4ESB4|X_3x9S)Zv~o zVqc#C1Y@7~8rHf@@qT{G&2Bc~T4q8!$y#ZV+J~ly6{AK~wR!k48`G&o(NIamD7S&8v>@b zQw7+u{hWnZx{JpfY-Vy0R44jlG>K~Gm_PP%>&?(!7)B$-BB@C9@US1PNu4Chb`L-s zVF;ba7U{Wc*ayzJl6_J7=M?i{)HndQx|6eXL(kV;Sq`;?GwrJ9^<}kEB>lhc1N&RV zc`^82W(4!8DvhmTijltZb8~Idf<}8A3EKK?9l(lY6(`CTi)HT7S!eW29q1l(DSB#R}+Ns)*au2dCkjMPRxmH<27Y}gju zlNe$FMYHP`B__a!|2VN9Oh|=rC_&!A+zEXJY5u^s=(R#q^-DT~QOr0lVt@KGRqvkA zPDOg|0vY(DFsNgEV$B$9v+k{SEZ*$_?DRj(b&nZjoz$^eaf<8E3DBc$VYfT z&mZ-=nedaZY)KRgk+19sEA_!kB?;QW18?q_Ist2TkG-vWw>!A=2sKLcvHEZbNb2u{ zZ?%K%C8eEz>ks6sWbZG)orcx%YzX>a*@GTG-JC0SkA$ik{NCK?Al?zgjFNrV5QKx$ zHb@57pb3WEcBkNIw*HGP4sS{N{+a2+KtuEv_qYokZ;(OD?RGX_ z#aVJh&+0=Y{!wr*<=AI8e;9*4X>ncdu`4J1sz^>`vvf8Lf?BY+gA^WrD3aZ+vNB0J zeaj{U;ns2+3^*<_f}QM|q{ZQhiQf-V$cGwK64;$!dpk{a5>nEI*1@;Xw`IcADJ1_g z=`|Gkt+Di-&iD*Nymr>TMvQ0PzZuAJ=B?ew_+6M3A(un`*s?BoA7AoB8*MxLS(5Wu z4NwQzCP6i5f6Pn->3yC1GX=956(zB+?Ea3AcIs1(2)Pe@{mSp~9)c3ATdVFsJNY_n zlqOn0{Nq)s60tIigR)WnDDaVngfcEr^F^ajgVzcLHhWKKk%))q$kq#13fv51?FHL4 zPb=5c{v%6H?bvU1bF5qIxBl{JMG8|d1>lfdh7a^fVLy(uZ=H%`ZRB$fy3l-&;bdBP zHb$8@B)nJMeTd-jB*uQ(o8eE3jRR4oUC*fR-RdR|-Z_W#wl=+sd0k(;_6JD8mMqZX zvhykad#uqr#wob-*v&JFD+r4DnvTX0$7Nv3-Eu3S&UXmLptK>w$i&Qz*KrmH78ZB<!TnQAmn7J z4e4P_PIU7-=(PxI^RjvihM$tP*;!)@m)!E>7E@}S3>rbDH6(=b>pt8x!y~t3DSYYp zhLC*a7nXLsEeaU0w|_~CNdbo$B>+uUqn`ha81$Z8H`czzuMYd?ko@>-hw_4*Eiz1{Q?3gNqLduA0%!Em1<6+B(ycq9cCNBy0#Zbr(TMBN0PZ;#L;NtB3 zS=gsxn|4p=^N#OWLK~kn{eDHx)5NoridE>7wsOM|e>9&6A z1ri3_k0d?_?O=E42Xv1eE(C~JqzKXMIqB;5F(}ib3 zT~>9gAAtzeG@B-{F1IYV_fmBIkJ%0q^UAL|>(^MDXQ^wkSrniR=2A?{18i!g_mEP+2D`F*g(49& zVux_CcgOld1CsvpkIK`YtoA4`h(y?1klwm|u`k}grW2b_7AmFvgA88YH~aGX>z&qg zfhQaL(s&aiNS>NbVA7GyT5^Mp*i?>$0+YV&;dckDB4RXZc;HPDD4Ml6^IfZuh&%Mn zFr-z!VzFx_VnI?&1Mc~3Jl=K@ukxn^>)zmyn)P!fb>dbON6tz54dJGBQRiunoMNF>4P_8vpBf z7E3=|`m*>JL*;4EzufnU@0+rVEDEq74!h~!nvqEdjoDVG1Bc~bEHx8Srx(Iv;yqWI zbM@JC5`~=;?mEGytqLuO33DqmKqYi!QfqL zQH|$uUq*Vgas<;2lD_URh{Cp+EMsTTr$NxvO$X^H4zG$>WrH6-M@VfT;UygT^oI=q zdP%_jxysm5TgpTRGLs|?Z zA)$BK<*VDTF_PqC$?Wr++l8hF{_r zsq9Ey6lquwl0vojdF@}1xY=*HqC>4Vy>>&`USmSst2dvzog)e>l~Yx+HaYFUE?-(E z8p=|CggRyKJYnn{ipb+gbd9?t#~^}96eJPiTX)f6A2;66_*_LTyTKN+QKsx4ZT7Q?5(}sIVD?H#Vi_o{ zjT^U%gg1VPrC)YHizFMXH5crBw)|blk?+&2q@hNSeMu^@@V%#8`HbL6F{#A$BCHsm z^bI4q`i5Tv@7CvDlE-ETC4pp^k28T$$;fmhW2eDe4h^QXW|iJUaU*hAh^umnwlyolN4D>(AO;O4twk@y(PQUH%3Gz(0g@ILlZriva! z5F#m`>W>?_a$Uh{N=~3|14MAWg9%K2Gor@sobVB=qSow!QI6MCYevzxrgGX1 z;)~=JGp;`i%s!SBYI2XuI%9hLQMndWkT+0*;k&Q_pVa zCsJ>rlz+<2x(INaya=(lC5ha zQrTd(v4EtlJ84$wKEyY6TlbaaovEE<|*hGV}U^J`XyQN(_K;jG57f&i1=JgaKN zp}}Sr*xy} zptqdCbzenNI6!BdSv2Asb=jSKWJ3@-IgKI?k`08__*ZE%^j5addLy@8UXL>pL2Yd6 zOmJHLqv~P$k2^)PGV|XDc`(Grh{xohoY|cpfm^J52ebP|(>ZK=LnR_;m%Q7-V5Y%q zXLhZ$eJbVNs{Ni+^@oL{Fos7hKCgf6ym%;}(Q3e|NT@D_U1KR`d{Di?f-8A(NVKA& z!G3Km`w4n!*_5-AglAqXZ;qW|@I90QR=$nt0Vwg-2VmqI}ciuJA8n&UBIfU>`* z6f2DcD&-h73+Z{-ZtpL;qGaVWg$S=L&p3RHU$gps z+geV$==I4&c+$a$*U^(CBy}9El;ulNirPq&K5qY<-^3qU`>%l7P7|(VVD~R3#^Ep) z`+bK7Qkxl+Dm(o%q{dz~z@!+6rB-4tt0_8?F8y31HC$ACFLX2B z4d7t~4)t|ydP5&8X8Wi9Py+1OaiuIjvg3`2Z+N(-UeS}n0RL=#Kao7HDa_@Yv>SODB(Ix1{wtx0uMojwsjHR9~_sDExgDG(GF=+f&J8V}yn z|L~i-ZWn*6;VdnteL}*i3Ad1r)ncDHZp5xvLFCMRopyPUuW^1X$F^7MwC^5pdMN)S zN%SocpE%;yYxmaq%YhVAz(h14!M0el7Rb>>kc+9gQj*nh$|e)uuNQkg1c~kh4=eY+ z6XE)Zy!Fd*)^BsMb@s!uAQ9L0aC#*ZT<^H?K0aP5m82(AZfzRr6}s%#NW~T6A8cQ* z)7SN;`h=!3IZ2Oul~%_dfk?5XlcOXcQcQZ8A;lVWV#IwDnxggkcTzpT-`ln^ND)_` z{=KpkbeIh9GiiQy2ConFpYLMH<3IWmDDdXQC~E6IAce%puITEH zgS-}muo-xm(-E|F5Dtax(u~C^SQiF;UbVh=57?(?OX~vssYu_*;64x2`U@%w*TsFo z-@T0&zWbg`BE(K?MgERP_G#nN^;!ru%1vm;nPf1i>c*iF`?QI`?gB)f?@Ev!9IX{U z{<4W|P5|vR9{(YX#i7?s?yVTE)JmM5W%Gkh0`UzQ$VaxLXFW*`6_*?|y+f0rxo7mJ zQ%XwuYR#b^<8ubZBk%Ub{3K}ff9dnT^_A2A+UqHOR>fYh#Zec0HezrH|B{ILkF3nZ zzf5>QH1v6*1}=um_+t=Ael8CV!H2;uUS54CCbuMC^IhvuX!uY-p$IX#kdDG1-Dffn z*yd7magsF)QVUF#$guxX;*HTq5tywMaDOyGKTd!vc@*sXiil?fBEtC>X&K_83f$u* z>t~;2c+{FD0#$9KUSoaw6{C_?yR%_p5xAt=vFU21a6*{FSm+flC# zViPGNyhgX)PM1O^qLPdnLaB@h~^7+0x_t~KbOfUs4WuVsg%3RaTvbQ|3UW&z1uHQ*h(d-pZ)heekXxECc;k>;4rxa;cfn28Mgs06(iF;f!7on}&3$1& zybiPJXZ$0(C-DIgR4WBQpa`kUUp$k)yR(Tg2@!xPIBQ37C_8(sQK_X?9%`FS9!VjC3zyO{NI!`r}LQ6 ze$+l4kI6-}qv)(6OGU|EN%m`cgmui(CjAPlyKOv|E zufwn`67rRnTt<-|gM~6<$b+ZwOo1IU#{}GgN>ig=SE9t<(o0CFPIMYp@MD>+6Q1f= z+ZU64{l{p;!R_cV`So>B5;T7icE4I7^AagDo}d@*)WdhQmQYWlWQJ0`p)Q?!_^q~~ z#U6hff0^7NH`!GWl|mQ2n*^OG!L8jjJ9DRc*sT$vwVH8uz^D7D1UqA5~u7ZvAL76 zl3PD-@SseQ4>$99Wyga zV;mlnJ22hY$}#-O9HJ+@Ho0a<>VMZHrZ$hXL#q9-t*#xL<%W+rg>vuQjp%0jgjMlG z;!MoW^eGU+sV%)Nc1zUH-cRycCgT-Lf_5ZSn&D49I>U_g+Egpx58UaK4pkH|P9Be) z{*b~<;$VhZ@t(Nbi8y^+0#t(o=xKLF+wDxM__9YwYR~)|B z>B<_l6ZstkMzeKEZ1KWT$ANhEl`v?;kH08zHSya4Fx)$hj)1kf{rAozH5V^lV)j8W z{PUSu2Y6XylTl}H(`nnP6X3tygD1P;oRE~Xbi_)ZRWFF4CMz!qNlNlOzbGJ~5uq;~ z+n+qn2&CZRl*zdGV=akBn3)JT@q1JZohy7njf#WB5`g0(1C5bRfV$FZuUP{Rs9Drg zW21HZp62#2xCXB>3NXEYu6hhM@6P*4Ub4Ka3b=BdPm9=tUL1dtL$ei9T@=*fGZ1+E z*kH@m=!uL5=C5h0t95S&d++_2*W0#AAFe$Xf}mW@`Y(vl2!6acG60c4Zoh17iuh*7 z-tZ+pA`^6eGgOPqysXAL1iy#+A`pP4SR_qgz+I>It^hm}BkBJ<+C> z`@ktNmy6Avk?BDv-VQ9!WOjeOkMXgdJ$8JMQPj7nFn z%)SN6b1Sw`;AL5y^nQak9o$T%ty6iur}+e31`Dp(T4U7cR};;L=`bjx?#?E?j_#LJ zP1yij{;HkKpVK(s!p`)pF;Yq)$fr^*91Kw%0+LVeIL^CCIe*N@PMp-!-3}t!!e$Q6 z{4z^C&&2j5(a>K+4R)6R>UTZ1=9G;Gtt#{T=+l|{{?o0yJf4Ub4>;9-6bzgfz#=Y% zN3Q9B5nWE~g(GHC69blNac?@SE?_Wue(=VK%aAMG%e$sd7n~7i6T6H8qLiy%Fbpa8 zf=8}rBdbF1{LTFmdA@nXOiZ3%4YiAgMgbHr`%QtL9Pq2lEHZM-bZ=3`wGKClT@Ku) zZe~DVfQ>2Zm|lkoSX&)YbGi8h=EdlL7k_j?@<8nSI@FcwcrEN)@vPcAanfJ`vIc!~ zi_MWgT<8F5SW|+cq3@_z)@QBx52As{Pcaij*y9suK}Avrw!Rh(z3ytJSY*JEuY$wq zVL))fxcHs22|&4DvfRl_6jU`6%#*6k%y2ZZx8V zf|ifmk`=r4eGb|IoSE)L@PEq#f_Fz>VOy#Mu$qdH1VCG4MSN-_^R#i3wNHf4_1Fq) zpX8lCs_(ueaud-z;FjcFUmozRlVIzOxz}S2`8_RRt2e!y)Q6c&euiF*?&Sptfrh07 z`z?wQBeZ9e=?-W%SH!2GJK!xZal?|nMi?qC$-64^dv9=5O=Y*F7jp?M6Vn#uVdi}G zO(n(1{M9?}e|dJ?naN%ReXdO9b!)F%TIFI$8kYXusC$2XvXsHI>J9V#`bVy~^Jj@y zp{(=cCLJcDzeZ`Vt4j7=E8fWGr*@LOt$XBtQN6kEePKZ{ zTW6Ch+L+mwkw|AgC4RI8TZrX2DDbXsy`AoEM`av}og9eKW&4CXKQ zZetY6dcYEok4WkoA9EcAeWtyH*F7)qiNPN#qs=iZwo61Jc#`7UGqFlJ6clq4;8i$2 z$#f@U*WLl|52FwI%+b)ohyB^DDs|pt@;(89*SCziZpCrDib364A z^(XNxe=-5hx-pLdEn{M{KW`dPq!2zX2P^rVz=kC&EBOt!53eJ3Ka7&LJ+JD1PH0gP zU6?jEe_FZc3R5M;*a@2=h`k2`HfH@FQ)=GXJk1qG-+;L@R079Xri*NeU z6;k3dLAHF0`EwjbnDth^Xm7SoG=7xA!+XG|RZ$mMRKRr!&cvsG9U+{(V2J=<+=J~g<3r5I4bjAM^?U5lCLB9=LzhxZ$vHrwDnxZEraj1wH{g^n#Zux z*7p~6(u%oMi}~Lf|Ku#t&CIN$YJqAB{`vR&(*D_$#`uR!MQG`Wj@D0TTK%ZgVPx%v z#BvudIFS~3Z~_Z-Xmt!~7q0RzocgSLaij zp%#}9s}~8kT&+^{(rmcpR&wd(iY;Zhq10mI=HELbZ1}zVmj{wAYH_3QZZbd=wYcaj zl@Br8P>E##M5=?tvT?dO;h!r?E#Am@*BP|fR0c`1YKA%*_mVSKnWCH(WKag*D4ERTi zFcKS+;GZ|92Am-h{8O&gNB?M~@XuR6m}vw$w*2%gK*|N+ALCT&!}zC7a`SxB0s-x= zAT97dB;m!tKZcFaeOka4u{B36CM_^r@XvVzrMOQEh*~_OQ&g%(BeK1NJKsr5lJ}a~ zS@6#VT?DR7Qgj9 zgCww}vl;~D#XnhtZ3U$T^5CDk!8M71fA)^$#~AQWwFV5Ffq#ary5a)=Y!F6OC6+5N z>;e9HeNj;yu%(Qb(XnN1OWr~LPfN#^v%G_9 zeu9oI+a%NN=-9GU63aC{-26C+<&V*^Wl0ytBn{EAWe;CRsl}UvNJnD%;--d=fqw#! zSf2l`(4CT4{&3sZT!TqsIr+#8uECI4Zrqu{G1TJecW?0LLC2QN%aSNiZ0RO{stNqF ze^y9=X#wGr4gB+3AQg)LxwxB`{1Q60%xY{Y9RE{ch^?TGEqVB#aNa>~I@Z0CqR~gv zv1Ro&28fr?v4sI*F_MlgjDMaDq1bg9f`7*Ii39#8aWU^8#s9Pz$y*5g&*lfRg|^Q^ zTHuz26#oPKGvRUAQkNZDRAPA&AI_2%IB}@b!N*7o*ua)vSQuq<&Hp&#pX1_|d)3~t zWsdYnaFbXbnw}f~fd4rn+^9+mj0qtd_@5>1x%LA8Ge4Ynj$%t^h1>DZYhON^mK3(M zH^Uy_pPvF5_5lC%*}+?BD$)WIFY3-|fq-gAOTNu;L-?Nst>^G{g#Yc%hivKAs zl&IjJzA3A(k&%`xKrPNcv7G(?c4Aqhiz>04Wu`QH$Cj7n5JJb6JMnx7m8|rBF12|3 zc!3g7i>tKrD_mM&rJO+y{%24oI@Wz?$(5EAwfOR0e!w9uaO;$uK`zsh7288C-pv=f z1-1A$zEBL(0&PS#c9j;$zhld-l*iJ(^uDevKgs z(gMwX<`P47Y-t@tp{ircH9mvTv87R`Ax2si+%7sHM-pSYT1OEBlznqbll(YaBXl9zKI@VpWICX5%(XlQo z!vgCU@K4JR4IC9}L&uhX%8~8Af;!gmi&e*#@;MF2CIJ7;;Ah_80{I{EG4PMQ16d)} z;_~3drb8`0Vs8Ov;2)w*k`x91^Q}Kk0REX#mtha^Pq#Htxs0@c{W}j+_=iuU$9Bt! zRb&V7&r`ft8Ez>2BjS=P{KKz?TkZ$@n#B&TBKXIfLKXk>Yw6e_i@^W<4+VjAtouHS zbg0Eu>!rz;4{eT)b@neW_$N*phs{J`rYZFwjiTYQYPr2Qv+ z10CzW*XUy-Enu>KUFxY4q84)%W%Ejf<85?JLY@sAX#s$)xJsL}l@R&}gH9-8sb zM@4K5p~r{5TC`jx(%&Id>px4OV=u#m>EM6rFiZgcv5o=}#Q4Wd9g2oZ$KKye##de5 z`_%fr$d*+I3SAzalN;lm2qiv9~O^6vjUx;)FYJe#gX|4rGjfyk!IPCmO`~r=(4#1uEMX zwBoH$wMS9gkIzf1hX z>tUn??B|NTTZI0V@IRN9{-%FZ!iyI4NKFU5t{0jR}0 z2a%4Bb-h&OoTe zl+&lqz(4Y2G*i4%i`QD{5<1px{nDQzfPa3jf{4H9UrRoBG^$hlk6dR z{>gcJ`T^Y{UgDq7wOcL{9qXb~;XqLOm!nZ#Z@A@(6Y;|fZn@G|d4aZ1L0WR9nO34> z-MapewiN!cw;nb4A2Ve~hLQE-9TO|HU)`Bk2p#M0AJ5=IPde6>&%6w6>p^+JKgN>h z@0J^M$wgZ7Z~t-yy5%f>q${^vErvbHEjL$e2C8H44L&8o|I~VZr$!&F2mdqGOvm7s zo6*2RcYuH1_@AGFf58739qZPYDoFpz!~c9?c?|tab&I%_BCfm5NDHio>rQp-^{3Fg zuJx~>s$=gNc`tiffOnas1-6WrLnwd!%Vo!2#x||N|1h>0=vx088~o<4Z7&~~E3Uip zivRfoAnAR@|Bx*o{mUjD>o&*S7TaAK_@6om+#)T|aRAX|*ZP+gX@M=Di|OeE@K1%? zz){}7Ka8XD(Z5oGe;#J(&S^=;W9Zm>jh}UH^sfkX?EPq^>>#6KFNnif#s6qz%SZn* z(vqd5w3sPr$;yZSaD{&gr+@M3DE!as zk82b!`d8mhtIcF|>>X?&1OH^nyGEj8?_1bG-lUI*H4F#=AdP*Ym+YF8Lo`=-0jv4kG>XL9O+|t~1K(uK1tLe4Xat ze|`}EoSpyS6<28i>6qXuE%|o{Nm}4eM#?s6uvwqGR2UF46+K zrpPn`qy_5t$NAC%{7-qZ*(VmXneT%C*}}WS(Xp4`9BIj;pA$tp>tA`gMT`sMkBV*) z3BN!f9Qy(~)@c-%Psd)>En-CWom~1?p*!|IxX0fx$=)rZm+JZK=~!1_$GV392V1k4 z*5Ll@0gA{7c%1CLXIv9$*9UyAnZSgWM5#&wu_G3+(;0W|71!RecLlMafEjPywO7=! zi)-)LyT-Lwup<^?1r;gM1>rr{72VZ)z4v*Z5AXYXKipqOOeT|=^DpOrPByOz{+#C@ ze~RzNr9HiV)Jr-m@iwXeke;8PZC|V3?nqI(ytqnROY!)D*Vk7FkaXkt!PJDfK(!0R z9eR6rd*@vp0`=1Shw+2>mF-U+`H;**+OwM{k7lNv5fm;RP8|z}E3viu@0E3PES(M9 z>@xR`X9A?0w7a`H27^>wkoEHV1}}ps6&L2b7mR1m8YEDtv_4>LI@8Lr?%?fVMu2kF zI&AK7d*DSv8jgLHRL_jd8s8{4j|H-wApdG}5GH9VxhdUQ9n zV_cA63nBN?I&E&fyt+XQTt% z+1y6<)Pbqyz3snj`(?RbC`h||9UcF{(^RVmflEtu8dL4$LnBH2nD_a69us(TXV0jD zx^17Gr!OA6tQTrSXBYjDWY{0?AWQ>6?mH>kw|@R^ibc>5(lmDVd7PAUKT zDj_|NOeiWZwp-O;#4Rnrunz0|r{q8J4F%Zv^Bku}p0EP|6cF}lf(Sc0<5^XNTHT=c z@kx2f69M)Qy1JrK)Pr;hMm^eV0ebPtF-UuGIO<^rKvYqFo}t(|K85h$&^X#9d?4YV zVVR{K#KBj2uFDZy5b?FS2`YdI!*6eDu7h3lf}L9Yv=|XSJU+^U2BHC8UX`2lp9YZk{?XCNhO%d$ zCCGW3AmF)`mOx=afm$OOr}F~%%=X^+$FH`Icq{0Czc9z4mh~|Yd_4kwQ?(lX<|~Ww zwSA_Pm#!_Z-TQzR^jmMNYB(x+x36Bhwa$-#5*$3Fjm_p&tm6mLoh|Lw_{Aho!0)un zu3o3-c9psjz&`;)W4ttjuUi7>rR2kqd?LoCpzH&F-uKC;=QZ(qM#!g6A8bIB&dne$ zLGyj;y&zE0f10iDfx%aJOE@;)_u!kGoSzNgrIK!M78E`UDuP5mcBBrZ^hbwA-E{-~ z(Tu03qKY%Heq=rW15ZuT`bw{FtnnWgR0E{kxA%7SKHHBzDNjyDm8tsSe}8p-jo_`9 zKIK93Q8i%wkSoO~9NQ_xl^WwuzY#4bxpaEbYz_~7F0AbIvcLgzQ_pwXCT7p6o2tPhMSRaA6Hlc4zmx8~NOyuYz^g+Wg}#oBOS$W@hc zakme>j=+;Mzq&f(y)p5_vx{rGkHgPjQfi@fF^2x}tDUnOh}i=#Ezkvz4Z!dC+VVR6 zFRiVRBf+{rgQcCAJN(X;)@oSXD{th3J$;Y5X~g`|$%n}kdE)`j)m^FS`)R^E*xmo~ zdR;NV)6)jQyb(GP@)~$!d-$eHgarM| zwq|*_y0m)7)mjLxc6~+N{(nrsXFyurc-)J!8#@}LN)Tk=?5smQZvWn5o(ke7)V#a3 z!v;T{Ce-u(&Tb@BUUFSpndJ882?`YLW>st&CIEj*K@EqN>tiw+paZ?Ao&_?ec?Cq@veN^KQ$w$ULcGHA5$$WN( zfVytMy-WWbpjYDcf25`Z@en>y~BcKQ<` z&(1IMatFD66GbJt-4vVh^19)qj_lo3Ey#gW?X2Z%dd`~v>=!C~HMY{q-t#PMD*Y-z z$f=d#gDx*}RWA-$HGar-x!$k`r&dJ_OW4qKmeZID^7Q6PSJ|%;S4DQ+B9x2unrKb> z%R4Jt)RhBb1S_Q|#d>t!sW8g(;uO_mOBPzGCFpT*sZYWHax$qO*-4 zxAC*K6G-`5{cgBOazZIzBUUxoO2tvwYa*o>XV?fqp)z@*(8>wvmbRKVf}%&lnbT1~ zq0Zf4lm{3r9EcFx-^0;GV|a>PE|zB&&*U&z=@fCeJV0K*%$^{BQ5?8Ds8cc=tUR_yQqph39RR7J5 z)g&yoPBWeC{Y+u$VUgtov{d{tOAVnVJ|ACQlWVzaOuo6PSUtA)(?LGs$1rzqS(9n$ zTB|?)*Te*&L2#$3MGRrKN@q@1}#5B6R9GhHFs*8@y;$c}6gImm0^4|QG(M3Gm zJ8Ez@O#56UD6-Vvk~p0f{78+2px|0p>kSeGgp!(5*3{f$#UvYxYQNyM zs@?m90{?_z!l@AvqO>l66qZ!bn=zppK2yY2;Uud+*jeam0~vgk^O@&_XEypVLO_$U zx=#pshY;W_$>qjaRj$11fOHf5qB4*#rrqA$c1_TjsiJg$ckh#y;D03j(SE!H{Uems z%YFVNnX9<|RAKh3i)xn@9c^k^H`Z#z7tQ86l0NdeF;-q`y)oqX<_kD+b(gnyHuIM9 z^H9DP)l^lWLrp@p4D>uPtkORW1@z3euHSUD3` zO^uT>o}P|y_XcC+$JaMEv^2!_KjIu)+|yR23B_uvBI10H@PQYIe69y?FK6XhSj(En z#8Dpinpel3Uhd$atTn0jqXR>)W7V8g=kbxTDMxhP@_IX!liX*cTIVRK>P&6;_6{w7 zd#2Lz*Z+KVyVyD_Tw}vZtmQ=6szU6j73Jio8`bLA4M(jFt|}LeG_d(RXEONSu_a|M zFA)%bbMRd{xb;$=+uY^&(rO@k4Y)MZS1-0-(0U^fU9Q~cdo(G5BUa!+gV|pJQM=mV)mouIa$a-`C zq&zO=G;gp}XaZJ-4S-qYYIIv032=E)rIsrKdVZ9ER!UomR?0xN(tGuU5qBYETes7_ z@&NV?xV$28&5?A%bFUtpUeg3p&Q2GEAP~b}US3k8`}$}RQWG||6uj3>o%sH~ph!Hk zE^0EQoSWtBj#MTm_4c;zXM;XVgqUJ$w{bOa9(}IFSCVFFs#3Qo-P~og^cl_fk{G7v|-8kMrN{IV;o+=O|rf~Z0!Mhq2 z&UH%>70%-$f+7hO&N#qQ?_M_=q?c6T-0D+=pY!*&8#?Huy!S~52f0Js$%0JN`FUmP zV7*j?lE=VPASlk~xhoqwo(*D}CjnUu(>zssB$}E|$MX+ON__q^(>y-m0P`y}Ld~<} zr396lFRW_CG*2g$GLgar48FCwXDnVK--#KhdGHePoNcvU_dEbKPu;G5!0@BP7Qcxi z?CyDZtid7?HBWqhL80&cAiXDyxW9ME$w>gm!e8Io(i^W4A-Nf^A3GZnSS$n-Kv`4? zd38epM3oR9*XPm4A=i0s^qVJ-8bIR4=8K#H!8qdHj;{NL?wM+oVE>SWt#(Wag{vVE z@y2xSSHWFlL?!gI0 z!nR-oKuJ+vW@^fYU?U-x<+Vy=oHMP6_vo~&jpqre{>xf=WPSw_mY$Zq=`w*96rHNJ ziN(t}9MOp}&g~TdZmsp3{7!2CQ;o#9pJtWU`RP#@!3s`kx>r-8!e|{Hd9WLf;F0^a z7;x$Q0ZZUOQ*WV8fM)>tA3wZ5iYy6rQQgsn@y`iyYS0fgfpldF3E4Zk z=)h}28b>^%M-n%Z#dXJgI`o#5LAQfr4JeF8Kg4s}KIUX&go;U-mZst+StcXOps@_< zVo?T_97kHpg=|xq{1EC6;*rlMC*zUDHPj}vR01ZIKBiuF_tryc38m~N_fVJ)N~eaA zi6#J>3XLZeYF^+a^60i1FA*%LNG9YK8p^%8Oh~I0Rju&Qa!+yy71W#S1cpC2G{t_{ ztu=UNSw;5dtL9x1#9y|wT~${HFtW=IM8z$Bda>J{p2*X)@@g>!Rn1>7_BuR*9?7o* z$0Oc8*y2x7kt(R`lkuX>u;RT(`-sR{U7D0lDlsJCqZV7U`;q7hIe3wLCR=7zm;kq- zmE1Tc8SDA3o_yG6KM@twh=)h|Mj;)hSo!&60p6IiotsS4isMm1AvJ`l6W$_LZVS?j zlRi8?&S#milA~A|$Q0DDV2YJFZyz4*XuuOGE}{ZTcA0Bi&Y))i$sL+Y=EX%j-1y9jKa!PmFaV&wGUI zMAhUdJ=oLdmnjEHFgP0IE@IX+&1WAmEqdv-dyK&1qOPrT_I zSi1(zFU*o)zALIG{ef3km(@^%L{-yT|8Z`paVVVfJ`4N~zq`l7v1xV`P7er*V+;I` zjs;Xrr5~b1>G_3uu7VVB{Nw9u0z|3;PoQYpqIS{fPv(DkLL$plK>>vXuB{1P6m>v$NR7*m3}TM=0Pc%#D7@`Dp(xy{HNdSboI5~dLyE89nMHa)B+x& z6&0w6sv;=FRlmG$19keInOQVmgw-gb6pK5jc3AH(095RfhVVTLVNQJe%hvWJ;Ma;> zN8ZCbe|Nu&tGmAMwuk4UEnK^Gpu+0r`bPeGW3MA)o^q0F_l!Qrd3b`t$_(&IO^>Y5 zKfu<##ab7*`yO(qi=3;x!=5p_ zzBwofskE7TD2+e|hV7&7Q`u67Nt;n9TcS|fz%*W6H=t00ON**>RE7HHyt%ao<~i3Js)%1wHmKzyE$UcrWD2o0I<41gq#@=WQJ~a*O?zkG#JFPe1X*@=@LQ^*^mrdaoGW^B^Q{ z?Rv7?2MNYKzqH&&8r%2e znhL0<(oGfN?zT?bEaO*C9QAOI0E+|HOd9j}=oqy!e$C`@DIf{Xc>hTR7T4&0X6=-S zH$2x&4qE32?;q|Tb{U`D$(bqVcuny7=qVq0*xdc>`WTz2f)9dXaqxy|F(8MFHjV~G zk1O#Te+Fa>Ydyk62^qskg%KIUA~ykC7-R70sQQL6bbfRo`;?H6?+j0`J4?tIc-~DD zhm2u1@#uHsSiP47DpSUA;Ptg{8G~gYN@KB_{L~ot)&H0uw^hdHPnHht-*EErtqz1u=jG-VyP^|2D_Dl>QV@TO&paRLv7|#8R z=V}j5N~{F4kul8o3H?7YhG9smW~!?7axhYtw6f0o-x-62;hz|TjI3%(UdC%1BbB0M zDwppW!y*KRkmL2?#;Qi+9^jE-i5uvV6AghfhHgk5sCSyMH4w$n6RZFxo$=^bg<6K+ z!`0;gAk$`afmZVUB?-Pzqb0;ILf zNi`fymEfmYDvJGg09Mo=^KRK&yu-HsvS(zT=Yw557xv+rBYqtNFLEBE@@pp_qKcpn4uPbCcIWhUY8<>zW{4vYkJxC8oSSj|G_)2Mawsn?+@b07L zYG%giVhLVco+sd0PtAn~2ghZe`7FW5)a&%fv=YeA$;_mR87#q@DP|i-l`>0Hnfi#J z46BD#*VT8@6Q5AL)s`^N_0B5^YW3v8it3B)puva-hk8bcu+E?4=2id?=DAk$Gc$(9 z6O)C990@kHmK)mv^<2BEm01-hjeZ2fA0CN$oDaYlgKJ^t zw~V0{K3F1S3`hGh&-M6ZAqyEp_H$J2R9yr0TvJf7Fveg=DgDYAq}Z$fDu;ENDr$v< zgkeENK~bozzqu74VF(y89wbH>ta4s|OBg0nbCwZ?igmE;P@9#Dv~INX-gchrae8i} zS`re5sUQ46VuT@Z)THkSLpa_Q5+e-7ACYhvkTA^rYO@k*vsRFhFx2Yo26{#qP=9|z z7)IX1dLPLM!=YhnrR0A!<0(unF10HukN^oouu6}F;f9RQ2qO%;I=g>E7)lDT+^=MW zA@7}_h*SrkF+>9rhO^`FIv`;v7R-df>QBP(myEx#vQw2y?uRTg?Vv$j$#pXg}gDqnkQWzB!1`0X7}QP;G@tn)YY`#d%O!_#?! z>=|OCceWK^wnup7i*uc7p^y)+a(SU!@FcBw?D`{4^58ovtzTQb*!qdo4c*sE2cfkd z?(KK>U!PG9p!QGTTJ1vZf0*Zt3u;ZSE1j5>`Qo&{9#YOsE8~>0t0G)IG@%VrZ>=Yi znK2N`7%BpF15UbWLMdY)rdxZWCTGZ%o+zEFC>cY5y@L|=2I3?7?CcD+Iy9i@3NoLnmEH*D#gLWy@NEAw3Ce(1YFjPVOEZ-Ap3M-g zm>00NRXCLh4Msk>vcC!{l%M8mTbGv***pKP)x;sPdy^|`TOX``lMssR_EN$IaYm^& zHfy1y;aivXEXw*LyB{5dJ3IScpM>m&BDqc6-lr!cKj0I|r2juPHx;zGISdfU-Rhv? zr$BCLJaPCvk{eWWRQex_cq%JJztQwx+FJ8fV9d2dQQuMRB5$mx zbnp?ubZ$D2?_72)oPqM0Xc5wGZ={D#Oa$7{JcDI5>-l+lB-(5x@0(_nvlm!YK!y6p zOeNNJQ(`5F0`OWNJzSdaLXX_8A}vM$*81w<+Ap zxyubuLKorpiW0iIov9&(*3Nk7QzJ9lh5Cbp$eQyYP+p=}IE^M-I}4$or%LH?-C;oXoTCCS(Ludn1z0v)p z07TN?MP&qak} z6^TKj9cx4Ws#b<+r;L9ME1OI%eNrSb!&wD#(|m<%b&;W;5cZ9?Y@DvJ;8??PCeQgq zE9+I3~w> z`RLH7$CeDwj^hIf0mrgX0Nc9*!!z18)U&Q3e{5mk$^3Mb_EZ2%*{`mz@d~D0!}P~I z#k1%_-cpWftE?e+Y@llD*kz|f*F9#$+1vN@G>SMqcFEa!+=l-~BDB5!--|@Vw(6=i zkbC=vrMMZR3NtNk2q7ucJ`>2zOif7%7J(DWYkl;LzH`IVvMODe#Y4z$7KCud_c3L& zAjDj)##Hs%GEWje+De3WGzg*aHXD8))2~K|67~BmLUTRtY;GmM@OQa12(gjoC_EqR z?&WcG{8NPbr1*iCk-m3>Uc)cG`2 zLTQN{EUp^}W~P2{>`lBslMW2KYZ)q+Ubk@_qDN_RXB)qLY+uMv=O8RI2%s?Q#pPJ9 zj#xl)v62YP#mf?4`?#{8(vnZvc>e(HVp(A&i8o2i#Z$t-X2iaxHH4LW39OP7elhVH zPAMYczgUj#}dLCigz}{%EpdaU0AoN%VFf>w3N`qg9!TvU8D`JI=ui($;8K~ z;izACIO=Ey)~>>Q*2OaNoI~R;Fwn3}1z1`Lz7S15486(21S~BRhTqn%>h2n~DLmy(|5J9M^9&HFoFNlOCr$^qW{*qELxso7!{%b3URLsond40`yY5 zRWLI7)SMjhM$Y0CPTmTn*)$EYvzLX`XqU8y1N_kX-AM$ltqz>mcu(>no;w!v-l+`F zZo*scn5=f^T@4!jfx+~efkqzzu-4zEur`ZSv{{1``|vy3&>D-6bo!$MG^OC$L^9_V zK%xc6HMk$`-C+STr~>_R3EA0OkaSVf2Lh30bw!{Bh(|qeniimwh3BXP=LtK@y6&(xT8n=X>K+FMAv0go?L#jZ^YqNG#Rj|^POxZ0 z0lgb&w4p~<=sBL#wpiPv9?)pRqyR7uX3>V|DI(mW(FRSg+~IdN2_UZP^`!*|EZQ(8 zO1l!KP%MV~xE1JjAu9Wd6u#K`|T2d|yB^e{vi&&81o}jq0H^zT5 z$l)4`sjX3HQsb?~@`@7ycCi*+*Z079xBrTCybY=Gwqi{0BRtR+-Ppb-CkT{=Gv7^c zq8&P<;WaXB%QrV6;)E%6pG5$QeT(zIH^2S*T}w38j`&1EiiuCtIY$i|q__tcb3>J(AS$U^Oe_Z72@EcHet zZA+^0E;3o9bnpb0wUuo2{rZ#vypqKeR!+k-tcyhE*XaK& zpor!uhNV3FjG!Vw9QA}sxZF% zW#wwPAq>%wL+8EirhtUF5s^{bbdc3Dk1_J}s+>c;BOeHvl=I9EYn|yF(Rht5iOj~g z_a2u~;+vb6h|n~gY4C|C(o$r#k+@oR5?dgVjXmAI zVr}l;p`NsZ<3E|pDzYD*xPi9>@2L(We%aD?xeBrzXe-B}iz6$}vC~dD+v>$W$HTFz zg%#8VIcuO*1?wK-U={)ks~|7&vvsoN4o!MW!&?W3@$qt}6oremMMHCSzq>R%cN@^d$}u8oBfpx*Dzuw?Z(HQQ{Q05wjoEUDU3rIhMdla|-!k13$R z)*GwpQNN=BC6Q4dwO@=AP9r(gl(w`Zwz61Y>tLs6Rte{WHN4j$nJ9P%>ul>n${vk8 z$V*z-+2gCy{t5U+mD(ofNU>RmoaEX8R`&9f<~BC`ryL@Odf4gYeYX_0<-YHISkQFO z5F5RGE7!uZ+>mR8^f@!Lg1xcb>PBPkXuahDk!r2u4_X<>ktA25lh>=%)=somW)HkH z4{MyZ=f}Z7Yvt%Ea=E1zo_0zN*1;Dr-@WDflETnQOivq>HYkV-_*`-aagA=Ep-2R zVeKjAT-L4KkLt+-I#|2PA?+28vG?jWHdrdI>@x%#pp;%PP zpCtpEGgDz@c$o-`+?)QUV5UMf0YZ*1rA>z3G_x?Y{hB7^)xG1fskaD7mMo6&JI__^ zdSY2k<}v(=nx!k&I*xe^dwH0NwQh9JgM$q;vw+3!D3~rS_wSwoFgcs%5i*|<0$A)^ znCIZ`rjclHi%`#jI-1@rxRrX7w^V2XS5FuL+d6IQwr(oGLl#HqoiD-B@YDx(%!-In zE4|l7hQDM@hCW3A2^(9=LF-QB5}q>*et2e`KV+tyn|0fGD-C@NObd-m?$wIsAt2zp~Z--vfosjwFsjJbPLI86LVYyewpipXGARlo1%f0}}VtMTKC6WQ& z-rpvG-dJ3~dI}p#B`C^ErOvq#INppOGq6gJrlMrF8gKH@>@Uu_sUCc>GyqVPbU!fq z>UdFFS)U{-G^RK0;V)4_G!`0DS|?~D*B^6j0ojDO9X4gz1C4tQ{2 z%=`R0Pmc0j;-0q7SbL}DacQ)tfR{*9x19?fnd`_8PcN+Z>N2XNy`yPQLB=!tyYT@g zn$A8j?7ExA)HvfY!cp2&SD-Bg8+VP#=i^HQ9vc2k%k)cIcVH=mqY)4Bsv!XJ$dOF} zU!4$;3O}aa-8b$mUZPJ0?*)KRn9dQJ!h3#>5SPY-Y4~7HdD7)jWVx3Y35#);8+>IE zK3d*WzT`8)D!_L!Zo`!PSH5~NWMfkbOSv9EdkRCa#&p@yi5|)G1gdIYqrB%;^41Z4~M7U}6hth6uCM$y$_%(V*NUxLA>|FmX$NeTS^v8*W zZEB?ycryP3oQNt@l@?1h#vQC?dg!4VAoxs4QSSQ$ckeHtZ1`Iu;kyRIpi6`VOjA@S zZ)~-+TKg4R5K#}+>3e3hZ|MJ`!Ql8#f--9NP(8G_?{O7qs2&<}bE(ezOCX{8zadN>H@gH!UyPPRP_l}h-6I&v*Pt$57njauE zX-I{7=%r=8u|wOh<~a)OHDE*Vbj#0le1hf&x}NCrRst$xtfi3%?R!+pc$yyonl$W| zj?d|BDq}t*6GD;QS?qgqa(1!~#ZP|&(=l4SOjO5M#?wsp0wfzWKfscPYw>Agne4z` zc!?;q+es5gJ>1tD^`k`d1L3Kc@L@w}pU!K75ZYfL*XVv`tuIR&&ccUnBFzsxTpuHv zk==WJ^pu=eNFn3b$3ztvGM}Gy*GLHM1?1Jkvm44viMiUvIjad9C*8XD`Vy^(qPyDwy zK_42iSW4prv(AzihdP z2bxoV)p}VCgms!z2Za93H6x7Q#t966<2Q_ zMhKzC1_)5)xdJ9QcpLO%@w^mc}4r=}LX!6yw&bTF%KyGT*GFxQ?(RL&km z2eS;QHbqpM4d|hqVeos)*t+eff@F3nKb**u;J`3}8jTrGU{X%ThcyLw`}}6_MFWU0 zdCq$r;;;mFw(v}QzUR5H%v(A<^5P=TX0>v7h#!|}wk{Rh$GMw9Jk9-!b!pJj49c%8 zftdJQ0{fa=d}KZG+b#%M*02IpgXOh)9vSRw#{R$J1nCuEH~@_RwLnV0<+q_@NiH5yLbi81Jk^mInqzeu7>&?RmYL%Oz~TgB`yi~{J8Xz@_5zwfd`fSub2F@*8%HPM zZ8Wa}V!thGmQ+(K!zx4!y$L=nP5>-U5K|EaEcpxiucLhbF04PtSzx`BtAZk|R6BJ7 znBxS%;sgbyPm9gaZ!ly3U25hs?0>io)iHHB`7AT`+i)%#Jssg9tfO*)fM>5$e+wg4 zrDHrl$`(vkDOng1XrCoF4S+%apJ7CU8T-fLE%M`V!FWMI(}?|8_vJk*4u<{z7DlXw zr~jjD!Ml4rN5hDXSubWUDh>wy--Hnj6#K8z_RJ5$1p$SG&@`eOYsMVJdxl~E-@=F} z$U$V#A2h+|`!FI$vjtTFp?`ftR2>ZZzs(lNOa36PviQ5;WHsdG`sQo8)k!Auk`FdI z`HUOqz?l_!L&wF`kG#1qcoy{|-%&qu{b^sh-x{vE{70O+g1o@RBB+WUXjYJ$?K#o& z{tliCy}5H@E$RDsfE>O|=~`10N<}%*ci0`d{{LWNB z{vjl&jHg~DQ2Fjd9aBw>Zsy3q|0*M>U>QLs-xoyT<9&9zU``Tzml3S~pF)Bh4G9J^ z`ECdfl}T4ym&*ST5LD2BU~7=QN7!jA|3egzqfx+GlAL&-*VZH--~Fe|AEBAQ3QT|Z zck`Cxj@mfX{D+jE#M*o+Cb-wB<$B8mr`qOJ-w#85BG7C=)m9qT=c`P8r)Rbf^=ORm zHw7Z#)mZ4VckrX7SPkk8xkQiLTmusjcVkl}X!@%qgfukX-UQ2H#SgCa6?!VJf!y0^ zk@x_`^*BKh*W20H7f9|qi$pp)W)KwmJ}&0!9jXPR95O}W>LUTlXo|P*AHv{-t$o+z zOK^N*Dm{Xqi}vu=mf;7xy1W)u$XGjRmOti`Boj*u$0mx<6#ROCyt*!gP}{WtK^U=9 zL|oHURhrDUtJ0K(-xqKpo0TT>P_vx9g?DU1b@Ct@!uw7q&!9*C)X5_1WF|p9y8)dt zJ2BBHXKfQ4>;G?bFⅅOLHqs1tz)9Dy6oOC_OnrvNCk0CemM_I#|&_ZTc#JH737F zkbGE3b2B&aX3?O_V&0*e#G3!=FSw`0=kO zX1%(HnKWDP+}IYv2VR_Kua?JZh?A$tRIDW%M&5xa^||7tuTgrDMd>ZGTqDDt8wFe0 z*wezn^B=IQvIPW}9Dq9{>hvXC0@Nj^M+Vnvf8M5YPv@1SC3 zu{TtudZtQwpZPxfUtHihpJ}$f^`)!JYW3cWfJBjrFSPYLG(1r*ZKH6iqG2MG6qedy zSt4YvYw$v?L{k9`5Pv8TksfmY0!5hfmrfZ|h6~J{vaP{9Cx?F~>;BvC|L+Qf|E@sz z?+S$fu0Z(j3WWczK=}XV3IuDpbfdybEdmV>ShCdzzagKg^w!!@UKGvS$nEXbO6Au{ z)Yv7f+~oC5RhF?=_uHw+1w*PNS>IjadF`1swsR zang(rCuLo+g_~oNph(NmK_rk%i-b^G>QnKI7tbWla{uOPOSv>tuq*%lpk}$vmr*J% zwW*=1=BF2b+0vS})^lruT*^!@ z4gOtlmQT)8X{`Gk=Q(P`d*P|*tnAqYfM%1RV5uY6)A#IeAEsiB2qQIQuQT{y1gWjn zu>+4d#1{8-P)zU%MeI!{4cwHTf13d8%RlzoLX&BMS z`rmD>4HMrHo-;126V+J4XZqzC6gT)x8+f%UT9jU(Y!~99QMTU@6!e*%2xjjdn0K%C zax0MtxrL9k8x-efq?~kz(4f!MyFVMOhcYJPJv3rdl<{sT!v~;@C(X>(Jy7d@Ntt2L ze2hxzO}7NVV&>fg^X~nY@4Sbn zKluKM<+Wg6|8u`t_ygM#u(j)&Ng^CYS#RM@BS;fbveO6>n5}zY)_u2AKG_C5e|^gp z0To1AO!~WlWG*I_fqHQS>UuupH0t^XJa=J5;F`u@_U?gs_ky!3+9&}4L_oX0-4dXZ zrbMJBtWyKq67X<;e|L=(g|eQ^bPir$f`vYDTlceWVpIXO?txkN*{NsgJh-9E?A?d_ zvv>ce(x+|t9pe!by$5|61^khB@8oE7K>gnV2;YVG0pk4XLhNEFmRW2Il-4mt_ zQV7tt{4sAJv&->4Zn)rQq{{)?y#koqp|Dmo`;9(!1%s<{D6xK?fmhp zq2B#hYXRcupus6FH^UK(_`QCoAv8=wkak=9JR1L+Ktt-?=cL}k^U(M81Gq=J>@ugk z_TlR=>zwkT)%brAIYInLL&9V$(i3jg7GzLa~VT)Pd!S4Ul zC?Ji-gcimRpgn-0wkEm_r;7T^D4<1{c>~VxEe&dA=%+&?pbWV9p>`(Pw;zOuijR!J z|CjZ~j*SA&_;XKA4Z05?H;oN7eD-4^Bjs!C*g3Qxiv`Dz!`f3q9XDlYN|03YsURmM zn6~YC3z`qE^b?C6Cxrpbv9D?9b|`ZH3w(-vCMoEL1k%b9zDzIxn_yVQ4Yy`r);fPd zkyUiJ2L8?L@)ltXx`v;xJ#|p;uf>94G!(8FZ zC?K#=z#?l6Kv?KGpwli{OHVViy?>-G*=jb$@KlAQWgFvJKWw}YmY4|i#?V)DY{cE)W*fd0O zrj6OV=a9vt_V@{GUHzid!oHsi{2Fg-jvr0G1+@n`7H!p_`VGPo$_=*NTFY~6Mnd}0 zTp-Z7z+DgU64@%D@$RJkG_Y)#m!JNF-8)|g1HYXM{6>2ysP-5NYyG22)2C4F(RNMH zf=|d(=^TZf{s(h`z~%xM1{7uEo$feRuLAL^o0b~ZZ7DR~>Wd?#i+9&(P1p!!2`0s{3p={r48eVFi zV_N|vWCrIA_)t=2z|*{5*=HWidA%8eo$8w*zKxAEgU|s#ws}c6TnH_vg8){}| z;1(?c-5(^xuSI5XQ&21pLS_Kwu0F7?zH@NDl5ZyoAJcSznHgLT){AChV9@$#P<|`z zD{D1z8<821R%=`BuY8d}eL6}=*9viTtx#1aY?@J3Xl4cxNS^3gA*;A;$P5fBrFaOL z0W9_Gdu;;VaUq-9nAZvsx>m?hLiKZg8f~aOI%r-iB(k+a76$PsGJ{EiBI)|kef>e= zZQN={B1NID;f=`bZ<#?IY8F!0kggTdv$aBekDDhBzkl#UX5d~I`6XQ|Brs-hlZF#c zpD?c#(y+BcWOkXmq2rl=B0}!$>Yw1on8A5>h&%JyyjDmrv01~a9)I=|fH4EQR!C$^ z0u+h2QF^%}#?r1n)a<%<@NwjqG|)$v1W2r_Z-vYtuBm}0wwaGux2xY*P1ehB|6SKz zJU9Av^8H#snSrA}utXnS5+Jg!zC{RS29u~y$oAD)2uKLB2Aro&2P99_U2KU*J`TB# zM@%;=7>6ZpY(C2kj4aT%uUnwrI4X(24OPg=#_;6$#LQc$uaFu1N;?t3=2iqBs?r;g zKG9ahI;2mJ4)Z2FQjiH3SfcOy-FAW*$Z`cplm}c_(70py8f_@Nea8jn(j(t;0Y(M3 z-Xvtw$5&P$N^t}>$wP@KncL_wh1K$YbC?Ku~zoeR^Lekwi9$ zWD(|UK=hx241q1wn_;0=GGS{D?H$X;6Y-#%gCd6B;5iEZ4OzRRI!Y00Jn{_ulX$py zB*t~}*r&&lj(m2F41@D?oSSr@*Mb_4rP@z9glwl<_E_^{oVy0-bP}P>55wmo0!c?F zm$^-gF6247W-l0iJf0M-EXTG8{i4;db_iiI`U+v~kosG$E#o-`{w%Y%nHHXP0a2ey zta4K&mf5prTZCj()U-PSv~i#J4(|lo9o>{gn~+tjXXZP=+7ahuQN=MVIt7R zy+YIJMYFl(@xL#oKBO*il2#)!ACk^6eIK%apan6tgCjK}bA5lb1yGI~r&ieiU`9!f zZ=}%4nn>PS&K*Y@yBXK_EuLf^EBd++AVFuO3jxl~Dw`&YvlM!HA^q$4x0dSIBgGo| z(MFY%>^#m&E%!NMrSue8gWp0=cmE9*6mPo4OV2b#A>W&adO2W{BH|D3mvGh$t!TLBTiRRYkw z{l$4MUaBfertIOTe+b;i+NHZz{s#^JC0Q>Mk!UM(tcNfCLxx^n5^1fiOla?3EBzoW zk8EI6$}JuKT?HF$ZHaa5)p6!-T~!~v>uI-6!6y3EC{laee$kvw^luNB!}EEkCftq9 zzEnc`MsOzDo!dZ{_4zsAk=C&si^*@}xprGBg7H4|Jjj`n@4tVz7r#CI7T%%0#+cC; zrVxec`q%Az6h>hV`8OE-qJ2@<$8IdMm-n<(SQ)tVH>X|kZO+Z$^?$YG&d`|G?M-}K zzo(xjzDE>KWihYkJX6y}A9+g%WkXci&(*-X`9*oD?iypT5jZLpO{S(`?OcEK6_i1< z%U7XC@B>$DGCuuExF{{I{4+g*w8GP}`+?EN#~bByLo6JoAx04l#2d&n>RwN(J!9hum;kpTxWRX%HPLWf=a%1^(I_ODRLzlM1Z&HV=<9 z97%;Tg}{jbh>m*$ql|K;;Y!E4ocB1D$u}47kdQy&P%wRtlCt1Ztk<|RNHyO*VhE>#DJXB3tCd+OZ{Gq%xYM5`dGTtcV>Lw> zMKqwRZTcHThG+tqqi8Da*C?7Cldo_cpp+22mP4<9PMzIL!>4{VO`UzKTzd(MLhHu6 zd%!=1&g64iD|{GxoSE<2K>CMxne6_XwYDZ%?Qc-&r2ipDw$I&Lw(e+CNp?wARgnXi z+cCPpJtW+5gpH?Ng)GU zNDXYSXPr?;(rIluFviA&e0=;%6@)TA{`Xj|a|831uEogN*+h+bl}LlML}(FPcgzV? zy5>|Z7EpZK#JyvR8)rN@CMeQA6gLTjKW5o`b%+M()io9J@r+YN66514Q|$W(u1b;3 z&IqCM#_1OJZABKDRb-hLhX7j~vcR8kybf3oJ>L zpZh36gpEJ{O8vWDTvPL%MCjB8h_>v$3mLQ5M%FxjQ1T%=jd-R`+M1W;u zf3RvRmbQ@I0nlu0avm^$EylIr10j(GZ|v|rp8r7{noQFcWfQhnL_0M%l{Z*I{4`w@ za_qxclXTRG;g)&*crFHOF5NaFnkn}1Fpz$w{+jakbLsnRtP4ejs%<$RGsOl zRUeDXoSf(UxNQU#1Xw^W1V$Jp~hH;w+tXgD-VIA+J%_LJ3LSCFXb;=Uh|=VSp2Rg64eRx6ov z2FWK4>ax$koU=v2WuNiBDg5G`*|}+}`RI)__WJK#d{;pMbfp$lYrnqP6gkk9U_~iW zRP5pbG{)4(4_NCg=vSKc*hvRmEEdx;*INK!1OG!V(@53f&ZH_T?$|h*d7hivz6RDP-dRZ!&>vIuJ?jMa_?Un4f4h}3j0{q zuXYAKT%7CV#a-DRUgh!v9wavKzqBf{@g3uP_ z_$1oli}8a$JAB{MdSGk4j*NYT7bUdr)9=@M)%J+=etr&V5L*RsX}+MCU3*IXAwWlL z=TG_C_pg{M5!rN?pC|VlW>e1jgw{Kj4$6kVG5c=8(=8qSa;wYx!qx8bl};6~ z0J7z|JPc=M*FS8;l$(yQtDzaF)38oHnXPMkBt419C*~5p{Q0TMO0AdZbSb~G-SGoM zt|902=xn2vKP6k+szq}Jqh*wUNH&RvD%Cuql)V!?^e_JD!0q1_;^6%{Dh~wzNfN`H3wC*dmJ|7LHZ_eW~+0 zYW%XL&b1((y}Wm{ym6bYn|^pA&n;`*m*$#wb!QGt`a^t!c;>)x3Xm7AQB)ixV+R~uL?C)|Ys1fi{<|$DY~$cGL@O3Ga-*yH zzeJ5X`SOkij;;r&1at@SOdq}2e4cYXF7Zpd0cNsaD~PR1r~2u705(I|`3-=%>ZRTy zgsum05a|wLH=RZr(Ro`((#+g#0NQOZ`w#lXTbkPhX~=1xw^KK zkfzdhgH7C?$4A4cdrP|wkvY$F-g@(Rf=jz&@z4L@fm$2;2LV^PC5n~$_!;YudUm6N zM7s?$+%!zAknR0$o@j){ap3j!C|1TKj~E#G*Nr3ck9jCx z+}N)eY;uRV(@eh1qMHMLEMFWFQ7fW+`F?%C_bZ8ktt4icpl6uC8NL`JK!3#`trKaz z+!_+^y_MMh0BhsBuDg5gS1H*_Vxap2R-}Jjcv$B#UfCO4dOxfSr>5j(-1i5gdB03Z z5|*qvj#&1SP}F>xA^h?BfK6@&?|*MMJV1&=_Xnt*L+H90;Gp&}Br*|c33Qe5X+NN0 z=7)G>GFx67awY(B-`?NXJs6nSN!aE^*UuN_(S$2jTViPmp>+L>5`_#~AAnVIJC+)0 zenpfGhO~0f3P_8EU)2u#6@zv_hTc*zwc{gvogtjAa?zMPe%VZ2k%|B`+T2EGUTSvK zI?a>CVZTJ157NJR?8J7OUDMGh+ezx2Ja%Sc$HwgoA~on1V8HM5%F-%ybnxR4pc`v$ zAmO3Shq`r4hfwWc?H~SuZ8{r7|0sV9{MCGzL5~!Mv*yDsyrI%q^I;ugEIAUPri&W0 zUVLpnu$9E6r=>zOv;~zClAn{JHvB*Ay?0zx=hrR1pL5PILz@|eL8OkoV8Pfx1;$ik zEEr?d7H&wsz?NgCdx2`sEDanaR$Ye zuB7d2X1xMM0g-Q(mht0TzleJ7MIiK#%MyN2O+HTbHO(x;*AG~{nu0Bu@~?w|tg z5W0n1Z6E&)Roffm%G!DeoHQWB3`2SeoTLvgup|buB&HogD`}u*q9W2|K%5jm2&5Tx z^;toU)aR)P$5Sri85QyPo_JMp5o2tDC$SA=@Na@TgwLb^X@nvB>`OK?K-dA8X~;VO@eajB2NH@7s(?m2fK%HD`DBQ*pw3@=553cjkf2SBhQ0Tk z=zx(Xny}Q7V6ffnvUyDIlk2FCLeYU+g60PVTJ4IWHQ_Q@2X%g^k`lp4y}>4vf&2vkHHuvXZoheUyJ!ubh+Bl7#-UGJgIM%{)trO2 zeIZK@QX)`qusg!4Zl7X(O(Jmitqco4{2P0NT6$&1+bid{q~73r+@ja4gR1Ul43!AJ z)Em4iK-;f}7k6N^qj;22Yp^Eyn7Bm?JMUUklavVKMTeg695h_?y{hc#2=SuTMU%j;wXtW8MwGfcCP^Aw9-~#lT^Pth@Yy^wg8m#VrxRKR~5=F`Cz_@~yvXqRy z0chSr;OvaO!Nm>3pOcRR^#&y?5zNlpy$_;VVNZZl8+ZZ@Jh3&fiq?REmnn8*UIgS@ z3jyspEht~257WS^i+d@#xp$39>auMrOz{@dSNemmOMF7ODVt1Dcr-Oz`uL^DP!> z(~t(S?x)(#5+bq777z6p&%_w>js8B_P=$WF%l+F zH_!)FPQA!YsI!11hZlET>qgzca`CAV7K7#}0OTmpaBTR=`zk3DjFf5gGh+Es!!%i< zm1K!BB^2D2rB$LEhz4G68S`~WSpUn>ob+RSTqyBxLiRZoF!9e|eQLA!jLiXJb42|Y zFm2?+$L_vV|BWbXDhW_O$L0XBIkW<$!VpKs@lBpo|L@=qRtotC)oNl@&;2FncO7u7Ar#i08u|1Mis0UIG*kn0O}AQKe&C_ zWKd(aJjsQJ`8q^duh?e9o_&?num{Cf+QHbi@sU>&|hoYK&&=AoZpcFxJ zq7HGU{L<;F+(nnbne5e{pPJ<<%JtWBAV2dZ^C*)TnxMrGZ21yrnh)<>KeK)ImMC85 zr02Nu?DX%G!pZ=S8;=K@?X}Nr9?PJ}mJw%9+yg7T#5hPL;~=9wF|}4;g_l?a*~lVD zpTuXkRD!~aWC;c9i|iBFXrDl#>1939dQ83I?%S5TY02uExIm*rxn1^2j zHm!Dh@ioAeRMuiams2C}Ft%#IQM{6sPR^~PMIkt}3=09=7Z`j&*E7+%jHiCY$~OEr z8olQWW8!5svFQd&!+)d6Qx^7Vh>-d2Ev*hZ#GA17iP5*-zY^h*mk!ZPAnL*w+*sQ? z&_PRI_p{hv1-3a-$^cV()ML0+sK)6rw^sCzZV-nlJO&(%VvX~u)+z`9DZt2K7b2`w z@!1Z^;qmFSqqMDtQ;r3m`wr98M z6|%;oVs?78jjZwXaYz9k(<|gL0QnkwAdP)q$dgA5t{kN>+sCqW`(8Cl9FUp+%Vl;K zwlm)5Yon!tfxPfw=># zr7A;PyAbss6mmE9e9{f2Mq=DDy;plIJD0H9oLQH z8zSw#@aEYq++9)ymqUTep}^%(;BqK%ITW}Y3S15aE{6h_LxIbo;D3liLA1`Gv*Glg z_7GPsIyhgJn&BRzin#%M8d+%%QS#TO|A5nfbSUVj_cD+&0B&8=S(66f9-=Jz{uIIn zt(p{)b`=dPHbwzW{}*?eInw#m#;_k-H`o)L8}0!q16qlD%ow^02dO!iQ-6N#IbnbU zNEv_`hbWB5KSDicv=$%Z(S3nL_H8MUg(al_SAzv~yXRFywb)a%5TjSDC=8(K|9a6h zE_Af{rIjxB@U?cy?e%~w6YW#FZ83@?8)alHxqF!NdMAuR3UJyU~%x2LL-j>b9f@ur_QK%E1IV9EhlwniXC@rtA8UqRtk z*!xo?p73=+O?p*l*5X~7bMcQT?rcGG0U)X$S&XdzaSNrLeO-bw0H}oAyLA}z>z$v< z$meHugRjeY_R`|8Dz^;*q5>KYK23-sNqqkX`|*03yKt8%D+ow|&H^VIP=z@yP9q8{ zt{si9+D=Z!!?&-%{I+H6o#~Di3r0JLzczOiV6Z@I0kIyVn<^)#&5Q6#MxS2{`umFO<~=0x#1{yDk$1@6Mv4ziMyiyi;YT{taE{TIo_FLDZg}d+qlqvmla`AsFSm53NO0s|}Silu5;0hLS1q-->1zf=bu3!OIuz)LA zz!fau3Knn$3%G&>T)_hW(`kE`!+^_Sz~wODau{$q47eNyTn+>O!eJofMGgZn-vjJ- zcqtRJmQxlM-QRGCx@o8V4V{5xJ$&4H?+R92*kq}0PZu^fAwo_ET;2>d7kB~ zKDUK27Eg|>n>RzL50%Vuu|(L^8^XJF_wY&F+1a}y>#*iyckye!e}x?q{pW%`;{UjP z8|?+yYXZSp!S?PiJo^v#;EI!j(e8N_#yw~Fk=^r7oa$n&xN9Ls3=F*X4+g))W^Z_x zW6gMFJogX(%2o}_4OoML+m9p+)a4mZ+CZ7LVutwuUODGKSk| zh~cJ+R`JiT#Iw2oAPeFnSGjwXdUQt>C-sTV+DSdshR#vmRid3hUO;7#S_ZKZ`o z2@l#G3B2o7eI;+vDvlHJAWAB=_wx+fYA`oD*qqc9yGqekWL6BCBdT*ut#K9wE^QdC>q5 zZsqS0sv4bOPc*%2aWHe`_tu$dk`NtaR&j5TeT>S1^M5h0eSww&e@~U&lvvyw1QZUG z9Sooyb8Q9w#AD8bCJB>_h7?>IgrqmWYt&A(3+kG@f1qP&n{LW3Y81 zOM$@QHrU`8o_$G2lZ7#Q#f*D{fXacNMD5v5mI9HY_OS4tM>dwVp>V(v0_qU56nHbr zYU^`!SvQ#`3&&;Y!jVlsHUvCW_0eD(NR|SVk=DeuL8FSofh8tLnMsxcj!?5YIzH#J zp5bI6_hDkS=7El;gh)+BcX&8iq^L2%|f?Qf_KR%8V6 z-FCJuUU@YoJ*b16dVcNE5e+F$^&%?L=fUUqBQx57;{`GT%Jl znW^FjhpK{RRSYD~YV}L?q2l8^I698p#LexZr`|g zqQGMjoLPuhYPV`)x5197G9(2#S9B+yU}HQ$>YO*GRPOx4?SD1f0C1p0Kme8;@jbR! zVAOi&zT!@5JORW~h{53np%h|}&6K3pq0Yj#i_EMPvR%}+1(SPCMT7nZgDxOpkia0J zM(exyyAk2@?<6JmiRsyVEMYdhmnu5C|^mklL5#);OP9QqUHe-LZC>JC-)Qw5K08vuWdC(JlMz+ z7|F-@PZN;GPR6li;)amN4%!$C5Ww!=1$B}rj+qfao;xTE zun{|;z*FE;cR?#a_u{LfBPPoLGJs@g0BNDJ`ED>+7a!4t0ahXhq=;`CEIGbPmevst zQqjD4Nn=n?Z*m|zaALT#z@0}F)`?{V&WT*12?NZ;4zLkVyc*TJ!dUYkqbCKZ8ae`! z^7T0}KT14U$PTy>I{IA?5tEx}!T3ZCE2PXJ0P#) z>@!cF?nMESgn+05dH>vw0MY`?dcF4`@qv*LaHqYE-bNrTK-8GKL1r6Hd|1#MKfKrb z0AnNtaH4U*lJ|qbrdds1Z)E3r8$JE_q~pd?UY`l@f+*;(No7pjSxFA#34bYW%8ip~ z_Bv5C5#G(bjUnV+Jp0LGcy61pYu?x=S&V0|kgD*QS&E0;eJA4sBV=EJgk;e-_%L6` z$f2UY6KKRcY*=y{V|(YeT#u4bUZF0@VcY%TrmW*0^gE1r=g2;NPOlY_R zs84upigAohMcIKqshHoul(9g`2UBXt2pED4gvlP0?>ArvVj$jI7kgGE)q#QfhMJ-Q zAO?c^hIcOj2@Xkhz)*GIV?%g!JV^Y0n}&}9L}!Jv106LEE%gd*G*aX6!%!f_A<7O^ zk{x)e{Y2w%LiC5kNMv97v$55N8;vsc4Zq70ZZy`RzTt}|VAIsgKgHWLE28ipMBgwE z)dD{_PfUwH!4e~pb9s-+ zK(Yf!-_B!-Bee)|{C`GBEl%7c>?RCJ=$u?reEcK_t~|c>wO6E7McIMW0sRK1%M!^B z)E`AVu{#n-cHj|QeTa?P37vD*qJ!cVd4X*;`VFisg5_!)V+>UXe(DL)PW&DWq&mpC zwx@9+DcSwiQ?w55 ztVS_!fgYk!I8WXJz7EkTVDBKI#z5Q;H=wEJPmu$ShSY;STZ_Wt73@JV2UEqBL%rLI zD=lUg_znU24t4wU3(+xT9s@Y!X5Z!KL&L|>OD9;lb)3l%dopi?s9I8uu12STmHX$2OQ%K-#n$1Bv82{P z+#*ntLV;R`^C-%8 zu4o?a>y;}6awXZ<3!vMp<2`+FDoSgZ@6U^{+Heh1%6giC5MR1vjY0ffB@3PPY~m%N z03@^;Fq6EHp%-+x+y}Me_hVctgkL$7;zL z2$xN9Hg%lh8K9`R+T@{!5r#Ur+Em48pa3L5=e`4?OD{14L0fHVGD>pXcdxfFHAx2l zsITiSI`B*L0v$i@Hhq9~tE3Es6~SgJY7B<3G;OsB9zM+Vj7Uj$$`Y+MseeukY}^A2 zS4kNNZDQ18uTxb@ELspfB ztS0N0C%KRBn7HWnYVhz2vRoJHodZbzwYiv%7~cZ6bZ4 z`2yhtv}4Yt?R=}z6w3czrj(LvN85S1gOZi!&QwY{F5`$(RQ=psS+T}hw~mv++F~JY z8+qxK2~`J{e}lG-CJeNxlS5)YRjkEknGXhu`>M?{2QS$BqFJVuW|`8CUx$VrT_Q`W z=pJgQD|+bm&8^?q&X?Wr6lfm%1%EjE`xmCb5VO6``M)MJ+vFOen`W$*=oiYu<-N0F zhU1HU71Iz}O%m_yk)>U!Ur?!E@TR4sQ6>WgCDGDRlLsnLyI{wqqebG+wKu+3bgVFC ztWC5F)y{0wTOHB44{z%gOiFS9^$Q91rXUxxE*u?UPh9%9Ur!b*+Jn-Ri?lh&tm5>P zT`1;M7=VK)^gDR-y8v;lE~c?Uy9^buJMg4jq&=eZm@c{l%0(il!ly=skBt62y{DUR zFB@P#;6_16Gxdr&O7sIz+?sF|b_nQruw`6c2&f}yj_{V~0{CQp=$o1*X+cd0XIkucW(O< zD)OzD4S|u09Y?*uKHjOsV*PHYSJIF9I9mhz+(SpKilB^05&AUEmfNX|6^dGZg|#doFpbQ--cFEqwg^=q;84PYIL(W+$r`! zX~Nmp0O5Tz@m5a^AP`-l8I*tP+ZvIB`0|O&&mm=H{!NjDK9!{)emnK$$=Am+1qrEI3 zn(rfke%)R$_YHkjQMRrf%KNox`vv%#XHrFOIqK<3**%pe%f@t!Buq3J(y=1vqj{Ck zyfIx=Cqt9H8JccB`g6p}Ut3D?B!h0YOWoj7H@MUdE_H)T-QZF;xYP~*R^4DI zP&e@1fx7BEc7Va%Xo$rv7*gX7x~TGjr^V0r4)D}BeN&VnoI;UBSJ+YU&|~q*?6|6@ z9_%Mjhdc1B5h!aj{?4UfaNgA1uSEgEk|f*tz5#~X1{--=yQT7p0iNX&_paBQwcS+B znQ~cAlat4uQY`%9efMB*<;s=pA}wciufhW#Zd>mi9>m;y-0c41u>XpM_s+(-V(#4{f)uc~sz8z9#q5 zN_hN%isHX^SWytO#m~&vSKzIGm)(3+==58EG=)h^0kpLeMTy6x8T~?^j!u9^Cv-NO zfoK?RG6mdYI@PY0&rj>zw5{qWr$wcv3PjxuH=g1H3HP% zzc8UIN;coR(%@HFG1MGcuQzA|2^$S0+-Plbl^%91!HtQ{@h1Pov)z0xB@!0={b*wd zpA|2=w<#3I+~%B00gVRrxwVKN(2evR{_n<>Xkws%#*e1IW5`ctd*uN^sTDi)<9V zO+c%TQ_3|C3FrGi()k7yUT*AcGSFCoT#Y;NfuG4fj+_fplAI9TSzu}0XWTsV(Ot%z zUzUjpp-IB*P%}j;vp+N=V?ZP0*^a#2JizIHO48QP49w7B9gY^j?O7z_OA;A@5*Z@~ zpT@|JnnWyEY^xRmG)puBO&W?D88b98_EULQl-|E;Skkht42=~3L6pE}=NSD5bW*RM zkOWwGL0a-&y)vWKru9R~qj>2U@sCh~5Zby6&NeYDdGjXqUX~7Kv~~cYhco*edD@Od z2W>~9a*&x!3Rnw74}AY8%=Z=PJ;`1T1YVF~_q>iz=MIUOcwc1hFbxy$=QY>G@~Hz1 z{tgcsQsd0wIS&F*WJBz_a6o~<`{JO+bOLeZw^sl?2#z!Xav+{kx#mc5`YdXm#p`lPjVUDzIpk~8Epa)6oJw-10L0>Xh%`WY&U1FeJvx&gaT|sLgI^Cbwr0>Nypqz5 zCDiP5H~AQYBP;r4+^zZHTGYW&G_piiitutfVuBJYd*W!LvUfTU^HIt={H)|Rz`Vfz z7{?DhEYzGHIYht){X%1Uqq~{MSBn2Dx3QXaPEOFAZmtMKL`#Ra&Z56Di3wzLdT0_V z()<+hBELHRnV8@OVUR$4z>wD;thk8Yh&Y_6(vmf1w*|Dw_c|MS=!!~9hXsBE2J#yy zE+*guhA~T(B3{GKEg1qW8Isp9XV(>67(q6RllKc^kJsmLbAry# z+d%RFn(&{_<+(d`{?W~_x@oF_7wrtzh_sSbnKS|cgLnflNMOO0cRqs}xet z9Z44dBhK$>G*PoL(ENV=m=nYDoZL}Swpa=c?E8zi9BF@f9S@&v3gXH8W%tTfTK$vn zdy86^E=%MGobNbR$Z(8r?WMEw)Nc)*++M+;muf`traf2~w+_B!Q=JB|MFS_?%bv)cC2l8;H^OfseVTn)X`{DFGkgxYvj1IQZp$`plVuUac zHQTVhv5lg>jTHxGrbff($`W{q`quPQFqUtz(2Q^k?Y~5Q`%0GVyG8p=1Nq0|4Z`{+ zVf(%7I~kl4s|~Ka=ZAWFC2z-~!Btw5{ntZWch>BG79C)EHtW?%)h)oy?L90GB=AMU zy;_!>lfs7Aq8#GY==r{xT)-m4697Fx!oPy`&Wi@}fhL1B{*{Y6p%xMNKm>lxk*lsS z*6Qb0O9CjKcrDKE&;{)L-btfulpUm}f&3o=C7=tKHjo#GrznTq_NPwgP_^(N3N8K{ zXY^5M8MMt2V+FJfl5rrPmn(ytHjqaGe?}b8K)zV5Y>X^B)yny=3t*15Qc5Uh8=z|h zG@-BD?uQ{Z2ZbB!C@|nX9HOIdK-aBfgUvLeXY-2r2AmEC8qqVoTN5iq9m{AxcK%Pa zE!OPEYMKl*um@W)+knm{pvt;-;F%D!I(&fR%oc`$0qe3vi(8R)3NP=`D>Si(;=TbF z^*~ud_immNB^slTsm{jasjMJNXzOQg`j;#jJ_kk|Ptgbi^bJ_KVBakHCA)`!b@q8Z z;E%(^Lq>YQwY5&Mmuw_bUO*8!pd;zEZ`{y2jP!teM&3a^;5Ij~s-pWClyQ2`sCz0n zKa=(1%E0sF8!-9-5T{>|o@|1oQKDiq8s$B{&x+axgsNkJ-MU<8lxG#i7?gMkWdbt62S)37Syl}O4 zI56I)NV%SZSBjPEsF5}}w5Sah;MIM|<*g>Ft+8CwgpYl0Dr1B^8(i4IaKaa#!`Qsm z9OZxlQ}(@l`qZd(r2*WtED-jW@LZl2T6a_5G{f{MWx1)&B6!Vc>J5VC&(BCgDBZ+`-B292HO)u z#aZ+4u<>PmXR&Su{z3nR#j`lFqnbo86EgtRn){ZS^ zr`&a{ji?4aO$4*`KuJNlkM5WZfNsi)yM$h(uYQhJ`6Uqcx6Z5N2jotg`1VF1*M{NB5R}%LD`^6eShVrhj=k-n0n|<>~#sJ>4~w`lYDp zDFKFZXu225_gNabyIGfSA_#3B#RK&(U{R{g>l`s(P2 zLNm6vCh@7unO{${T@>{OEIprPXE;;y1@!a_2sYbMch78-st^6-khqz&@_%&chk7uC z;Y_W^E~6m{XKH{U0S#}cAWHlmPYMZmWHeYugrEH#jL#r#`IkUi1tIMK;LAtpgUS~X z>060Sm(la*(H+rDOsVaNRuhx1g;q3Bqz{?;waF9DL}=45eBG%+w=@={{qd!}SYUaD z!h1lY%WXuL>!=oY*5?QN@$8$D53lit;$BU{te$Xk?0HkX4hUmqD!n@k$M`*IV&~&p z87$=)qS!#2Mvj|_9A{W+AKWU)?c-Zpdj$SRr5GL)neki~C49EyX+mF7VmulF>7OLB zUpNp{RK*oXuLeS|R8$9FE}+SI*ZzjUhBkGXeTxLk4SBl8aov z_uVQzq@p)Y5gJ^8P*ZAXzEmxOXmA$Z*FU{52!#gyz0&Idfl%K_JMR|mX0bCQuzVmj z>D4biL@~pv#*n~TbWqfgcb9x(0gS}!f!tIY@mKE*19cVhR~c)rt1#&DfhC?~e3v6C zAMQ|rn@_F2>qJY9!ZfdRe)@e=U`3nebK{m?#5Ma__}!O2dOI)G)+4Zjm6rJnnf{AP zMO4*nQM?FUUg1+tdH97Pnp`zvzbM{VnGD2Jy&+&>X;GN~jk6D2Q~}jx?!b?IZwRbz zqXau0;l_@l2s{*JiEv}5VK$1a(VZ*k(bjyop#Re ze36(Y-e6elYwr$(S z7aRA#w`OW;rfyBm!@R8BwO;m1_v&4{SL5qYzg_r_7-@N^Qt#H*pW(%K3iOmO+9$VJ zp}Jd8u%JxN+MKMrkzuV~9c1WeH%3~yI!rUvtPitywweC_O0GWt(AHe9V70fPfhXvG z!PsR`gs3vD0qL*lVtT3~b7rSG_=bBcf;2j_iW?o1nUv*q&ul`zXTvV4l^D$hS?^oS z4`lOI5YFW}qEQ7|T>>J?Lab!~i!v^j$|2lmArKBPhwk?;p>vuG6(s1S>aC?osK7-h z#!#3AHp3V;VfeYPV}f}%{zr1KfK5Xe%=tr~x(?2huH9-$)mBz!nAj@YLE;c!?^kmH zvjDa5X0-}BS*tm2q0hkKYZJD)aXuPfVYv1MNw`~TK^hF&@P}Slh(|7|pbNi59>_N~!7WZY@QC#piflok{Mpac0&yrR>{b}($LZRF&=GD-( zB{3lVw@~Sdp10^@jjs3GO9Ig@n+%6%BnR+&8tha}olD7wo4!|5Q%pZI$oCNWkRlFyLe+FKiq zk!WXi(tKa1hocQ|{gASQ*st{KPZvx)OSL0^%3Ivi*~KSfRsYd zt%)Bl!LEE4YWiU*DHZIP^qN+n(I7JpnoN9oBT@LM?@g`pezPz__Dd>t1DY2TE06nr z&MmjV8^dP4Uc_xVxcm$j92H8RA4wT-`2l_0End(h-n07<9`$$#y5~KD=uC9C{3A>c zkN)7C;W5u!TS9V2tfpURP%m9vxuJ%Rg)Si?lK5P@E2W!TG?@J zFj8LYA_qxkBRL*Ddh4!Y8)Mphn4Q7qm_qrwAi$qccDA>#n)3hpJYwjm$R*WKJz36z z{J{PwaArbnde~v`pZ90dd_0MqKdUKjwHtdtDPD!RsmbmU(=LC{7rb5J&Pg78So@&u z43ANg)taeW*Zu7|XTV*0C-tssIx?vEn71fa`^DurX5y@IbF+_UD`&@oT0^k>!gA~t zfp_IGe{mS*s7k`x04C)sk(Ia#`%t$Q`Rb=R{*x#-rz(gcs)zpc15u!l)%_#=o7N;s z3ZhtP=uO=(Nl$j}N-*F`Ft*CqtSu23n@F$1B==eWL+c;?;tiS?ADC=Wlu2Xn03L-V z*o7S162!9$Qi&MW=J>jn_bQBK3~SaF4YJ*jb=z2&3)CJshe#0Q@qZZx9;|( zheV$Qs?;~!{~n;*4xA@$2S=9$J1*%DodD%v!FEzo(0>ei4Krtgdf0hm?jYK2)n-uB z5FWe{zHJzJl~4S6T*)!mK@99Gk+n?xeYLeB>P*3iz4WfmB5&2dk(Mf)ed!KFATSdI zh%npc?Woxj{~A8tjT~h@4sC_*u=e7zjX`0A<+h2em|YvE*9#5hQBZYeTasxYg<^~U zQpTJn83#lKhzDVqi2b`P+=iFV-?w-yU5b!PE-u0>KqJ?fKl@92g_2z++H~2I&I006 z;G~CjUBNZ72}XDr6P6|xdHPx_f<}0;&S}JN@&uSKO~rNftEKV#)41V$Y2x>IfyW8vcL(xfWN?rXr~*{WgTGR7TA4W z8bZ${7+bfo^I9T9AX$l;IJ_UDKx9X$x_||Ew|65(sb(J)ljRl(oVEnEz-u39zKhmQ^mw-CYt1%Pw-wuwjcc;jDT+ zTAFdiIq(o|Ud%9dF<~TdUK3L~df;_wX_l6^!7fo-aIMx^!BAdjEgFHbFJoKW;3w4z z35Eks-AidYIIKoxtTi9XT(Zmmq!1CqSE|(|J>^5hWfWQ~Pi$pv$iRP&4||QGCkBuy z_`?!s*N5oNODMM^UdYauo}`ml9DRN!?u@3hvrlJb%AV4iN8TvE<-pc~~PMPmMiBKUO*Id|pg&0DjGgQw{a7NG* ziB3GfnFE&399V4jJVt0*a$?k6XZ3xWi*C(I@)R9o`9aA$+@*8905LpBdjq@J=-*0Y ze28#v#)QkX=cV%ziC=foIo4f_k%B){=UWOdjk6gD^=2b6DEvMrIEk#gq-=k>?sJX6 z{w_OLm+GxW(*3zCusC~b{VVeEOq|J6?Wxt}ct?OL+f5(ykjwP>crh_a)Ae4A_a_Ma z0Y6y2L`UTT9$0JH07Wyffc_U+jmzHf54rfYH>==`HQiP%ki7wAj zngsk^ul1(EK_n{*3^!;6GLC|Gt~=8m*frMulG)1>=E%sTb|?VMMlI0>2T@Uf29bmZ zF;RccsbGj==v{;Q{D%<_oHB4fUY^AxFp=;0i3d3pAx8hW)M!1Rcv0)n<{*^8|4jJ@ z(`BmCp?K@TAd}$v0Tq?=HGSa`r>RjsFZ%UlLbTt|aHRO^hlP)&fqUwmUG`^&BACIJ~|x7y3;vNds!tMe{}N>5;wv zF{vjmvVo=~L#Qvn%!YbM4|m(Qfl>tK?y$Y*%!6?0L&q9M?5W5I`3%TMLx-hVw2YXF z%3G4z#v`P`^iVvk!Of@1Sa!7tqRT=^Xxz*3M=>c9zk($OXu_xq(#*QQ$G@hyIqr%z z%qnfGQrk9Q%kn+LAM?FicB;+fG8$Bb4jObPbY#DDs=Yhfs#!cMZb28d0E|d?0EgMd8G8I_NW|g4`+@krph> zgQx*9Y$D~cj`rw#*VWYI8!_ZtUQXQO1}6#oaB75iT&7ZAU!nBByDm^PCa7Gb05YlB zg{gWg1IH>Dy^?x$0?l5lA^mYBWdbTMMmpb(1FDyqA=E6!IQ`vQuJV>7UBA8Y#* zrF>;b>hi&FCp+)o(mssXSs2PxFB!(y?t2595d~Zda=BdDZe9Ustc+0lB;8TV3+5N>xpB(BV9qtcJSI^tZLF5Zp-dt=V z=naqE{WJR-WtO+kabao2XpIS~Y@cI@dNL#9>o>n`b%#3>!=`x%& z-5~dkndoYJM*?l0JsU;{&gx*XtG^hMMDK#*PA5P=MlkF6LpT&aCmSn(wYUlZL&|sP^oTP9 z7+%Kj=y(+o!lahFbwMu(RlkTaHS&g)L@U^uIua?GdSa;wsitb@+*a-58mW1t*S2Ttl*l?4lop186|uDr7%eP=OH%6}t;zlX zyJ?T)eTI$3qvi5b(1~v5cEeF{#sX``PY!Gl-f<~Hj??d415J7zuGto<9u;W{yBFV8 zEBWYqHwi9tKShPb7x>r)+N zQO=1{ttW&seR>;W8SRC#YX|}Cu?sAF#eNP>?n78=v1bJOO=H~B%Q`gA29C7!zKgC& ziY6n4)1|8g7@4wNi2M$ZE)WgvYav`Z{AlyXFAK!c2|CL?Az}+ zD`;7AE?-@6N?luPD$kr}LqG%o5kwdDLxVf!f!o^(VM1@Jb*TFCh^Sgw zuv2F4uhwxXsT18@PE=*(P@-JbX-zTbpkqYeny8xsoNZ9=6`kzN!*+~{0=!`%rGQh8 zGJlS;8DJ(U&kE(x?T7}o?ZXoc?Xa^nEQ*q#r%0AK`r=n^YY9XI1GC?zyDDwk@|TZ z#P=@{ynzk)B`dg88db$u;BN|%8Y?p(hRLU4Ar8XzJMXPC_~Au7&cWGZ_8Kf$f2GCx z7A?7Gy4PW zoh#BoFy~Vx|7MrQ|B-|gSj26fN}g_`G+DX(%36y$o#Qd2tn?OlAsM;BlD7Fd>^sws zVZXCo5F`YFs(gKKNP9Y`yY?EHs0jUC{ar~o6MHk^uXH_awDyHBZ*TiXiF6ug*|FVI zR2q(N$yP*owXTV8q1#U=clBBdpbhgpBv~~q2}_K%F@tl>qdUO@J=8Be3t5myXbo_} ze{f{wRqE(hOWEM(h^gz3DX$7?YbfCL*7NHtj=lpq0$cYeDs#HsX&}VK4%;?%>ht}5 z7w+UQU! zJnTK6%VPCIDl8E+3&<#v&8{RA3Oh-~W0M{qZ96w3Z_Ajul6bkfWce8YaJMmg7%tbC z407a5e*a+LndW3uyanT0x9kc+|3FE#EkD1j2`rOcT#E_v9zB04Vgw+CCk9knebzmMGo+YN5z}K`2QB=@Z zS)NcG=`pT2I#6u7?4m)%b)DwtXIa&wI&59(deb39yFk|*oM_Ia5n=2dpvbOUI|qSV zlH_I8q1M&5-6c2s8*j3r&JOcqYj)E%D5$``y_opCQ?!;$G-3dmq0B@AuB8&-gL)KcrhR(6xG#&f0FScHaHF&G+jAtby|t}`h1 zHqvk^CNQzXb*&j=GV3HB8IyODad#m3q{u%;qzxCFH ztrdW?ZrfJBe{w7t|K(@)22s8FSBT~lh@YYG?^C~OyD@quOoTO-p3?#|kvzJ!H3T4- z3uR%V*m+PU5m3=RX1@T9v(;D|P*y%ZEv4@vT%6(QGAyRDqnHxdYzL0 zwl_N_-Bm8F6cd$0YV>)ugg+hm65(x8gy+Y`e6r4P{%@&U6Au!3o73}4Dn1?7-|bS1 z%^QRoXR7JMKkyE!O5Lae3#~7mY&^RW7kEW35}GWcg#m?5k%$)Cx?7$-4*4j1BljLDLL>$5IMXYQ4 zz0mE(ihUU(!kJNN7T%bCw!?zh8Pzhct&p7{q@-?d7gs#O_CCgVK`=eGTht)O7?y^<+ zWYmWtIq`p~G)+9hf?nP`iFtv5(tBND?($9yn6sGWjpKXQ8C1C`u-`O}XUUhXzi-m# ze213m`lb!Gbyv7C>9wd)XS!E~GBaPwA`$Wr6rJ`zVCWegZyyW`zKy=vFt%natIryp z(5g(BA-K4z&-=*FKkpq$jipA}Bm|UpRacVjnj4%^#2F zPNVAs%yCRF(Ew3B0Eh7iBUcC&Gb&_S+3tQMVaAdKAXLc1TKQCKhZKMbDKG;sijgd_ z7WWHneYa=Znheyb&{RL;JH+_j@Gcr~d8kD)nRCyj=fp`cCblW?x$HGp90kj?Z}=WqF%p zbf|sRss71iuQPmAMI3q>)9F+^M*>(U%87Qhnj+^H)r@}C*2sLsdh*wVG!fA@Ut_)L zHM$Y`8MV@PzmCSAGaSS&l?20T(>j9;Drp;sE)UA&(xorztC3ia!>+Oa z0qg1HJva(Ot{=SdMxL}oS6$0KSdLAdlZy9!sRt9+JJ5H#SFrY^qfPX|v7PxZR#Z$U zZYq!ZE+^V0+fnIAkR#gLXti6Nl9J#X`2S*Ys})}M2YMiN2;`N0pX0UG|7QB9VbsCp zc!B5KN!9Kr!`ms1_3ajvXWRi+5X%Iah3XwUPU(DEx>=&liBf#l>q+ zQp*Cq_TCu0qzV;%ni^6%2kE|t=ut`gQX}0P&-4fKHHWs3)ut5I<4ofE$0j3j+;DS- z5SU?UV%^e<7u$80e9ck)wSgpG4oBCt{`H`lLqzW0*L#`b0K_#@7UZGe-(AxZH~AO< zMRB-(sRn&HJah1(s3fi(nK_ETyJV~UvYe9(u9$SBV3Q}Ug`Dt`l?SkRSHG&hwsWItdH=NnTjCyu?r zA?eiKBpN@s9u#N%w{TlIZT}I`m?L-DT^=!OwPPKQY>#L0g4~F`=y{<4`x|AfpRQ#d zLtRI})tmm;n09O@4vl0qf3^JeCuP^%OmDz$$N1mPtSu%Q`9d4lFYHFTi3#YFhIEFM zwPDmwbwe!aE5isKy81|RS9*|6On)8tz8XmVRWK%ckPo%%Gv+~li|-sFQ7`%Q(8fYB zuU+ru0ggGkde2k6u(c&fu&+yNzLwBdKlbsfEL|lXywNUy%qsDtEVNMu%OMHlmaa}Nhy>N&No(vH`f$p@SO7iL+g$h4mHE7rJYsKG7zS!m zU{`CSC$IzpObh030PkRACj*|95{lNDjTXnLKu4vT`+Fu}n1m)2;V)aRS5R^_5VZV5 z&tBk1Y*R^4j0FM>jC}BPM2ZN%qrtDn{mp`=?qA001lWufxqgI7Hxi@GDt2N9lx|W& z)3xDPm;1rlbXv_!eYn&@1H{nW@|^gw5^Xv$=IO=mDv;Vf*dMz>Be>3@MIZ5cn3H;P3VXM9ph`!#NS~hDz|@@U$<=FAv*f!cw427(ZDz% zc_4N3yKP#YHMYT}c;ZODRt{kfydrGq{3}BUx%H%*UTz>Mc6%|`5b#ESC)2Z+GXu@W z`pQf9?7Ai@vh^fBdysB@Q6c~9z<=M95@6;jkwt+gak7`3sUr-IECf>EcM@goAgX5D z*#r#2xUjXk(S3pY4U#WATAcLVMA=f{_5BOIM(A4D;1hUgqMA(k#Qw%U#a9zaE5dob z<%%}fw6`%ooPY9IHu_?Z(_uwL&S?@JitCnR{K5!WwdKVF^=1DGLFUxZ;|r1%X%)F2B&uO@{F@Hpx~YAZb# zjSmnQ#mnf=u>=r(JW#J95XMOJ7Q=8$>%$?Z1*X=arO^V`C9V&#C5;Yy12vbnm~6$Ok>vl74aG|2K=Gl*N z@P^t`P2%ZHdf1LR&smLn5YDUE&o^J7$Xhm0M~rADJ3RC@-z=(_ zh#&-&1T`eT$PA3OEt+3vD1s3gTXrA$bP(C{b96hE!qZcd(k-jF9mg;n{*gCp-aBuP zE7SoW`ScLk3VNDeBoYHsk_^uMp6L-s#A>H=(*f0j7=@M1{ZD3|c19$}%Gpi_exp&s zHj*0eTbv7bHeyr}wW08um3m~wdsbv|RK*~vNjMDDTGr&nAdBvqFTgHch9;&bh?i7m z!0lHwL4!(kjVb~85C}M8H|LlADK2f(Ul*{q!UX?eC+a>SrCSze9YQSKN}hlujuQ-1 zy&XcMW|FfwL}3)uunOwBDN(&fQ!tmLpTv28?Yv(Ou@n&9-30|1>dn`+wS_Xj&M$L! z2)!kX$xU903?HB3Duo{IPbw>kOc7F(=qvomH%Fl0v3;zM8_`8&NM?0n`W-;j{ntq%l+rH5bDLb_;79oZA0z0jrE7M(m|soBW=fYZ9&ur{8UmpK7F!uwt-#kea>DwE7*D2h1*-P*{@2oZk6X@JkH7w0 z{m>*08vzii{)r8H>FX8Q`1Q2;{WxfFb%PsoQgJtuaE2)W&?2c?RI4_j=;$VICW`@; z!>DQ2$wS~d{7$m4%4LQGmNDE*L5Kh*S@zQ)e34ax{oPr?`M*r9CGi}8wRR9|tIj2; z`(TjfjtZqxbMJ2|f4zQmGU3jMz_v+ds6%p|?_UW2mj=LbavpyW+1H7#dxKz%?=*c( zkp74~plp)X2uh0U&dH+E-7g8rJhHIRdT6 zjy4V98M-$7hb&I2!yg43?kY@UA?zNX?Lf;)BS7cvV%(R9efiM8Bp+lz-)K;$_ zR1hi9oD1sgpC80$%?) z&)I~E1VbHa_k9fN^S(tlAD_^@Ce)131VLrowbq zEl~95n~8Oy!jL4Dl!?+khi?9!<@lS<0-~vlkL6|KA0G9Zv#9CIiWj#KJRD(y1M!;m z@Ue-wi9j~9w(fH&i=Nt&qqgp^%O+)XUv1r#@gp*oq=9frHeqW!77zwTa8qdhj>?7m zhE@#9aJ`J2@sep(ZCz7>7MU1NX9>QU-z(1@rAa2A+nR&|?A3D0tdnd^Uc}Ort+O|{ z;BS^E&Xh)E5{3N7P{R8Wl4?G~`3mh%TbMUP%x)~WRb_(U2Wgi8CFmzJLgk%amiLD> z_M4vNNVy4gnls!{x(w~Cr1P@8SXA>zsWLNrA@A);FJo9*H~xc^!on&twnWNX9;a?F z+O%u+LY#%MOFLF^_6+kQg?!V5qzn=o=T%8}h-{eQS57wU1qAFP?ko6fP| z`o{sDJ4+S~%B>srQxpo<*|gFao7K6h(M5)bcp5g>n5}vY`>yx%x`YzXGJ#-MMJ542gBgQa4 z7DIgx5Db}o$Ut(aAHapDtp0AHIjG}<^0*iAFZ1ee3F(mqxQR*;F|VFvjHKB6%U?0Tjbzr_m-qkNGC%2Ig{(EkH#t&tiLKx59cL9@Fb9=b z|6|+I8OAY?r8jO{$J6Qqi=j1r$@G*3XL2chegJr-)y3j^(7IDowhAK)q59<_8oPS8 zM;=hI?fwii+)A>b43B)p-pVzfS}0MsP8~-MHsJP|MF825#aN>)7_cHQ8L)wWt8MRo zu>NnCr$Q*ft~~#ZsG}zDVPz79U_EQ zQpc2IplMgYGr=$$CAsmkisE1*fB0RQZL6(ZEnx_rqGqe7T>g05l{!DZsy{Cg%7GJg zqCOb6+in_d@9hhNs33Ydiib``iMi3B^&b4X*M8CQYIgu}!`E~7T@{pczt*n%dW_l6 z@o`n($#WVBGTd29o$C-w$r4_&%3ZaqpL0rFWqkS@87m44aPRtM@YPd5MWDT*xKxl> zK-I3G{DpbfIf3}}T&-Vh?}^jRa+#y+3naJY=ag=+41kgS?G?pvd{0avm zpcL|&%>9g><+k;9*oCB)rP&l2t_hConU;*ld4Q zkf5X75Gm!VFwQ`;E=Jxt;HR_STXUe}{d78eYqs@PZ-76y+j{@Ux9`ml?~?67i17DuB>i&SsFQZP&7sM7 zDB-N7;g~*XM(3wMeFoB9^XNlhpDrY2r=o2U#`CO?6F}7anz0SKGJ7mCI*u*L)kFd+ z^SS`#x<5gafw1L>D`>H|u=$~U*O;2%T$BCo(o{Gffj?_f?y22cLE=U5fD_-2huU=W5|Ih}z1i{=sNHId$Oy{MIW4XZxnl z!oLN}o~cKtRsn3^e!C<)^W-$eIc>vGXJM~(>9az@*8zZbS@oY;LK0kvgiAiFZk~U) z|8CY!oQm7I{I$yi+oMtiHHD-4gFr`r_8DM0zorGu3@Tla-pUBTbatQWz3N=H`*ewG zDclmud)o8qjNJ7cF$QmK z4vL@4T=AN&rjzRP3$5y@j;@)?JX1*3891~*GgX)-7A~iPB;+?8&)lSh?WL;I{?3($ zQA%cWRp72^Qz6e`ng^3)z*XLLQFI%!RjBkgSVUH9+iPwsV;st+7e-0oKxYywg2t z?Qx-KFz8pS&2zT%^~)3@7v&0E4Qbexjsh{gJ}$kloI}-^1T?%kvdN%v+T9845_}}d zyRtXQ{A$@{4h0ylOwT8+!s@3bDn`^k|I71UIK6@;pFW@JvEK>5GrV2&dz`3>z?fw3 z7Uh9u4R|o6Ava&VTm7~_?&9F7zIM6;-x_)!oxXV>wOG6@$cBGqDm!})>$E#i2aYS; znDo6$R5PQ`7RXSS)Amgz2wS?_@SVL{nOTVw(R@Mmm+79RRY zEH)(U#nILlRu-$atG{*!#;H|JFg1TC?t>g#9SldkN58SmaekUKZ@PdyTNMvcqd-^9&=oRs^68H`FK>wYCaAkuilv%T>6I;Dn$q?rAa-h?P(F@J0^E#TNB%*%LGD` zrefYgEH2$uMgzB%mM|b0k+|Z3RH)=(h8;wk^2z+)?`URx!8eq}b~di8#{HdsZGHy4 zH_Y|C7Y0MvfP9{0w8R~RzuLR~+Acx}ar=W2^g09=iCd$wgy+BSPsM6y!}%i*3(jW0 zpfx?pbqlIw-&I>~tr}&;by?^AabO#?3cj4sX4CSVfEE-*{ee55<%4}rQD2PDJG@;H zm6w>iW~N0r7}6{)M-&JK zYb5kni-yJk$r8F*nMULM6O_8D#hG?_LR4aAiMk4&@`}WiImu4*zTzg`@zAJH9W{S^ z&&PPz%e^f|CtRdiCH)~A`qxU@>=sZ7BaeGYl6(fqOSSBu^srE;%+K;faVmb0?&o;b zowk$M!QDBW!oS5xc?`VGGL$;fzM&eYakvD8h*H9n51n0do>7odi8RrW$GEI=xF3H( zJEs8^+!9awkElj(Xu0H zuIQfjuDOdVnk6Q=w(+tfb@(cd{%UtPiq{)|LDV*c?%wU3!W6R&VekI6pk(dAn>Wwj z96DL&VGQbHY(j}$%A8uu|0k4yNa=^ruiuNQS^L2b)j}V#+4lA0aSoG%|B(;~Pe*OP z%4m~MmDs42zMHG&<|#6c@jew3NoN;{O8XHcNyW+@Q*lz5jd27}3w}rPPSuuV_l?__xof$2?a>dLnfk z@bYDiEJ8#HK3V%wF+Lh(;z|L-(8Z#C$U!SK6F>}(#_g8K*-$|WN_QR>70RS~FHyx4 z_V4jBmr@+$>DxvYaAMK89`Y3y#S;(iRLndM$03%sO<|yT9JN$0(>OJD>yHz=K65B*vXjRwGQ3C#T&>J zM+N7=3KiVs9O4dSAjR!pxH{TLY#h%7_HA}Djhb<``X)L4nfKPtc_g3&&D?3PuV|UA z`cFciaE|>Yrzgv%rZT^aB1Vz+V_=-J64qlDm3l<13yJWd${XU z?XER!kvDzQ&&aVzAnhiFNu)hDra4D{46_f%5qS79vW&VWOWZd&tI|Us@pRE;4@G1yl60csYSn-m0!O z)NYEz%Y!`1nWFphQItr3fRx4$D`6RgdyjC561uicZ(-l94gC>*$A6hk4uc^;wt04ugYOZ7FR`o*-;dajewV=_jo-=wf)<`Fv|6Ks`-K_5Ojhp*AG!PHyY!+qUR?9rCx$YfGlNR4!3V4$w6=1F zG^%0B;r)i1p~Za6>=4roVk2#CY`#x9XK?eX0Th@;u5sI$_{#FSPe;vl*4~b+?N*D9 z)l!G)B<=Hf8xg>8uK=I?U%9#rSfMX%cQPzw>=`CtB7wRJu}P*YFMxSn4BhI3ZC7FN z-oza}*m>WS*$x~QJg%^?OjWD^o8oX_?$P}RjLLovGA5AVN zghqvLRvekbFmf6|PhI!$iR@^N;(?ERO~~^%erGDd8+IRaD8m^)M*qpR@6V#voKe-( zyu`3{`hS8r366;-x5wuK#?+X0yE~N-GGK*?ru(Yg&)j>vZ;}Ii;$FF=ZL(jf*$?C| zvUg0h-_Y(Sw7Z+b5g?PVVR9a?g{eMku#3R>=3rg`x&@EQ{#k)bh;+Ua^p8kdIs$_{ z6YrfDxn4q>O0Y9tp8Ozb7R4gMW&DTF57J_KYl z+|8vmvr3)x8`~;j%b5*=#o>`&Nx4fu-Swt+r#1-292}Ur- z6IJ7|y+(p_whGtYj%3xSb}&-&)*ARtHG7?wbAMbZR4UL?;%+Y^U%eddjVMrHINDqo z{$2-l+x!=P^;-JON--^$Pn6276er1<_{Sa8Ou`bMovx?$LWB!7rWQ%nRnDPXV8`}u zp2?Q1Y5kwnx(M7Y(h*7uoN!%QtUskk;10@L57FwnO`WMdzdU_DY}543LBh01lsbT1DAI0wdhyXr@mdO0c;Dj3#$oa=?T#n`4Ylp^E7EXvo4ADu?X% zZrr$cAAL_a0+B7IBt!=UE)fTh(ZrB(djmJsNFuLBx9Q1~!e16O+9*KS#m@V|Twi#NJyfp(kwxxj z9lZh``TEk97}#gZv31olf;@S^AF3w@=>aj>W}Cwh3)dD14fAAl$pYiw8wP*l_@JyI zo!tDtgKuaP`fn{IuAQIs@fnrb2O4_oJA-)$vn;b64{=3#-~130jp)t35^35PO(#f_ zi)<`db-5MZu4*KWN4K9nZlvvryd~09GK@&2w(i-zzC^vlpx^R|rWu%oqN)2eN5o@v zyOxJdx?x!NZE`syp{!GZ>NURfsar8kliH@ltO#f3XZ~`VBXj(nIN;#a6t7o=0*}|H z+j5cK7pg6V`?`Z`DMPc=vwsHsRq9v1HYY{=Z6(#$QbZj-5R`K=cu({ockHcX%Gq*R zghd^$$z!M5q=iSmXJuW1%}*(N?z;siFe8_e#RWrTfI;~&!F}=vG9S84TdvyN`8IZ{ zOz|$R@k-f%z#!TT5t=FT<~r)^>!Te~SvCtll#2#4+r@+D)-`YmnJ0Z`#k$R;%ywKV z1WFq)k+TB7%m3K{Z@W`oQkH$*Cyh|bIGcGmq zp;F1=2h5a-9{jy~zB|%9dfGWCsH)Ugqp-Y}y^d$O%1#$$ZB%^$*>a z?mW5{>VzK6Au|Xgb$ltVyXYi#_s~g!^=ys6652vR6e?s;y$N~h>R(+!&53XCYh_ft z;Lq{vxOOZ2o)+OC!%?#KEpip1J=SV&s{*R%mkp3wG{kwUfX`ogwmI2-MTVOCvwf}H z5V_MUoJihB$uR|UjR6mWXXqe!ta*t7I57{j2y0up&-VcDS$LBYlN zch-X7Uu@5fZ#lXRBXA!4d!-z9Td-#LB;eflD&NcM;)6g-)UN7W3`NJt{Vl)bfNAS2 zu7U&555d@54l#KzJ-YLuls~p`F`pjkHnIn|3%} zf*Yi0#?NRCoN1dqBlN)vNqYmUf`2-Qu~-R?zaDrn1bnW(USlDg#;ZzOirigt%!yN{ zM{&VjEj)D8VWk|fZPEeGo?WD~dvj<#d7?%sEvp3@dAB4rGv}~d=nFFNG<%68v7{5BjFJMRY0o0c9@o};p<=j zZpzzsA8Hk-W9u6<;b}EYx2TWg%m@E|4FIfZasFIl&d%GbA~dM78n$FGs$s>_WXr( zYT8qV_vK36ZTyvcmQr6Z^0>Vz1UFSP41c}YZki7klFDL!tx3S>aMhJ1F2?GByVLf( z37Ittz_*aga&|rF>*op-6w+<*|B%eNA)p;o-$Jt}ywVhbN>8@rk zo#d72jXb#K-_~7@>SZt-ku8hQrK58g6Rf2p!VJKcknX3t_CDhhey&An*ZY+$VBih) zrfCcW-PmK?c~(-O$9}D&G> zSNRn}k3D13K#6JL9?O+<>i{jWIcn$%)?jIHw1YPlBd@Z`(^NiOc37JCRUZyZMHq$) zW#D7gcXz>z%Yu27&$G-xJor$I6!-^C-V?Pd5_Sd9(Kv9#Ng@k4602R$d2nw zt=&GJaU%-t7qfm`#z7<^2=3G2O2p>kg(>SSN01e3##!ihr#hwG=ro{Ord9lHX};Ggb(6iBNs77UQM=+ z3+Y+4J`^5Q*pa0^_~1r|dPS6rRUK-g0qZ23NXEaHcDt1 zav(GhyaJ2VP4L$u0M|%LZ*vt*0R{eOvGO@M&DslJ_;GT6o%Qm`d)v0vz3$d%5&$+y0=NGMPs} zrG{&9zKNrQAX@MAdIT{Yp_V^aX27oqICDQgV>Gob22@iQJ?3n(*0v|okLI{(n(M-D zqTQdaWS6ln(yuns!MJL&sEm|T&b$k=7>_LdXzN$IR#Cb8@;cXZ!Kj)F*Zq-0fxUX- z7NKoRW0^VrcItaCk3RZxVJL--Ze_a1QlR%{(hBc(W*ON1ZRYzGTR`u75fPM@;)u9E z!_mK?CS`<;QKE9=;Q8E(o_7#2F8>mwI*+&}G58PwM|JKT6`9aSw}DaSfA36po(g#Z z{MSU31bjCDn{EHl9cSb|LD$GOivC2uhx!shI6^#0*%j4K2bc^iItw!5JhJ{pj zHT0M4NWg5-v#B2zZ+T&)*B;ypT;8K&yZo#~`>Xx3OHp^H?QfyeA;5BvT4DL4-5XZ> z2t{=4@)DhZ;pu3>KYQ1)#y-v0cT$2<5rFNyv89A#Y>JWEL0Jd~SCN1|;s>H}p$X`ULNnhB?w&JK?|B zdp`KxAZu?k&_v2{&f5uci50nt$f!Ge_7+y-Zb!d^5E7mZa@;-c!QhX0ph+ z3m7xVTa>_%04!r3C<&D97QukzE#R0%i$pvRBpaRB=vJU-nty$Kv&8B8HrJ_iw~{EH z1poXd`c`FnzJRVuynNq>jVZk@x0Dk9y-AIrB=gVJd*f+613FQe4QdckY#dJTqG?O_ zw$BfpBZsU`N#h}FaooK2!ep2~o3>%K6HPRrx8(f4igX*ID|KQ!7VugXaZoO^Wwr$(CZQI`Owr$(C zZQHhO_xqcsX`4UioQt+MPiDsDTFJ_oD`SAAx*ee=DS$2>W$$JVMA6|GAL6XP#TG|& z>P!gJUcROWB$#`evvJbWkieWAdgjIFHD-%wftyd@sc3+eawtLyD?&4)h;rWtK*(N5 z$o5o25wvMd3ag*8wKWE5u$V_f+c^<(H1rw_@ORfiG7@&4A`V20h>IsdpLiRqT(}jx zI1&Vx08L#-+VZ9mFiHhP*kx|KP`ie~GK-CbcAVLQ1`bWf^VVU6u}-_$ikG7d7@~#P zos=ZjE&U20qrRmsrgvOSR?!1$Z?E@oy#fdwy$J4KRnS=cEB8Og4?~E}nA)JWAi3W{ z`4>xd?*@-kVK7&7 zvcR}!<695;y+_$xjJ)AOH|e_18+H{2VW;1Tw{&<_GA$sLEkU#XM3L_Zh8ULaXdm)3 zUP}FOt~XSwpxG1_JYdixkh*G?8s_T4X)2LxEfA|xF*kp#V1xC3L;Dkg$Gud~`g4ma z6jwZs6(oZLTDHuH_m8`U_kAB8413l$i+#V8n1w4zfVup$NjI9d=!fGoVKks=50Xg= zDQhyMs9h9&hx#CVR1CDK@&!_t74}}X5ApsZYi(*Rcqad=?GH2m0}*wLYW9@#uhlh2k@IctGF6rbL?Ls8-gqKqTO7 zPxKX}*4WT>TlDY%q0e14wFUfaqs_!35NSQY zuL9t~mlg=d(%Dif^*E|8S*N>64UwrL4l_UCtYt$V4&T8Hn)0~9%@ZMi znkit^BagzQaYb9n*Qro&bW2;B&{aCqJr{98`ae5LEgw5!ipIxyG*%*C$Fn?y>ctPW zuL)G3_dQJNnais<#b$vS zO+&o%J@2@QTed#JJK|pC({=6P9`L;D-PC)8z~vgf-VC-vch=a#4i9_0SkEJOv2ocp<6Tjgr#D44?}xT( ztFx`t148nLZB@#S{7ffweI+)mCcYbR!JEAv7wQRqeAf@~f+eKHZn)DJu|2Vf?iVh| z`|s(bY~vQuxjze6tCI*sN=GK zQ#KL@luo_B+swgs3cuah>GTmNSU^f0_5oFm)nR8`OY7qK89mi&!@!-j7rypkJwqtQ z3C*3e{|II9?M|m7EsVnm1+ptdma4BrZo&v!uU#q+yr#_och%S4x<{JUE506L@|V@_ zL`=}M=75-WceM^@b9u?Rr_$5z%TZy!64b%AoW1F$qe48*>Eg|07Ex>Qli~43#a=Ur zKQt+igdAsNgsZ?xbSjEv>wM@DFm{N;0maJ{-ZSHc9sAm{s95csx%N^?C@);KQ8O8Z zhs2w2lBzp{>WhLZ{C|VEf0qdlc~yAq)KnA2k+N+3&{-Z4mx8tN&9=_Kqg(yaZFj_! z<|WEzsI|dUX(MTZw!M+qYUM&dhG;DB**pVzq;MJ$&em=-vEW`NLn%pJ>hxR;^eOpw zV%BJF|3Y`6a8tYCp_;-hmm^Odeb&B~>G$Y)TLZzhfF*e^JvHEeTiE}RqOt3#M?AB& z)#h#@mloH*7h2q#Ho5J?21aUdGqg>NYxi$%^w_3MfpeszxIY2U5RKSONvvPO`e@R1z%sKq#|n z^nnPxv0fO>Hu|!DOU)Coal8G6db}#uXN{+iH~i|oda9rX+yDm`jYw1f8K|XAq~>K6`+4 zMmB=uNPvf4s>7{|$=PQ@8ioqEAYEML#>qK}j7AL#a0QIHWVZ?fV5I)7T}~ah!!fRp z6ILxvkUHVVBArjFN!d`##VcQ`w<b9v=??SWsR#23Z3~ev9Ne6;q9D6O`CW3R!Np{M0q=C-j5O)H_X83 zEg)|+bk%AOQx*lfpvkTgqY9ML^0Iyr(f&-%>ae(RlQeL%O1C5hD z#c4kUjNxqWk$EziY?X6uNHFI^hNoF36$>$)+XAym=VaL^nRl9Q*+}92dx57cF=FGy zokluMQ|KT?R;|r8S#UVEgGnqKp=^&C=H6a5NYEVUH_C)vgLABxJ_0&fCbI%nUQUxL z6>x+l8vEV|ad9q9l0*NkT_n%PQJgA!5vE&0MN=vrHI55uIt=KUA-kq7Sqrm71Anx7 zoXaBh)V|bs3|5atWZ?RveFR;8JP$O^A*v^Nv)tYSv@{n>q9y(lBcfIJi5;`=d7=wF<+?r5DsWV z>uKqRD9`-Og+mCkHkIlmh+Tk_@#D)3Eb9!08oB-R_`#0*7_N6hMLMx z0PyFTk;!!$trEr$on(otNf?TO77BhGz-52h&gqI9+@2mG>1+jWJC{Zf8{*Z$)alY! zV94c=hC;>LK5kiJ2j6gf1(VA+(^v>tbhs5q$waVNmp*5g#EIp!4Cows$);mMVg}Y4t-ZK={_MhOlg<^b@h^IHnT1vy+04Bn~cPPQUMa?NxTc!FH622HHQw6Fp^FAPeY2lF> z|DmHEbLLmxW;}--N)Q&sx|of$ae2A5hxUjAoLFBtl=_T)JWIcAQ$I?KRgHF9HemIJ zLob_RDcM_6!8|fw3q^BRe=|hzv3ESqvqlAMox(gk*ub>i*>0bk{-IO#-KK$gx32pN zMd8}b*o&e~lws=cSvk$Cw6j=B)JZ4oZOp5UyOjAVyQ!&0VY1RGrq=O3 z#qvcpf zSVg*nKT(;{a;zR~jFw()N9p|;<1U%9&e*}>CObLG&YdU@qTG;Blgw+3(2vr)dnPuL z^{=RM(!qvwv!I2d?ls8#!%6~yR_c+!qcjyAHQXAa^H7-x|0VJ`SG<>epAN@kwT~P= zIR!E?!-^zFAKc0hL)`9Lekkdj_zrM?mRhoi>5i#xX1_L@WG@xyu^lNQOIm1x*z&0Z zGp1MH$prEb{ivqZrKuKH_E3lYe)@^)Goxmip~{WJxOGZI+ox*wPe!)I3SGQH@rX7R zV~NF}`^48?OoJBm8`;AujrE+zCt=?5FWsLqzG>g}6GG5s@8_a%1Rad!j=DIHr`~ny z-1|6R0R%-6F+>f?$%$)-gktOZicMMaC8B)v1W*2v-C$#!6%3gIJ1(^)?%tgk-|tI3 z6ye)~wl3-nsq5$0numFLvhrh*Homllr`;JiU^|wy;wp_UKqg<5c>!TlJpfkNUJvSV z35$33`;j)@Um~KzLqO*1O`BjTk(a{^#+)|h9LtC@AL*O^6`d|%pCt_iwGJSpK6HC{ zZ|#}S#_Qq}oatz`Be5lHnxQK7VS%r~;-~2T?h5~E9CiW?-T!@jAEfmuy2_2n9e2m` z6#dQrKh}yh*I08x>bE0D&m0$h-jXzTb33*pt_^=#w45oLYUqCf-}ux+$-i$*FnITr z_9-+r3OFARr#^&ed6GvX`(Uj_YW_LGtAwR<*0soJFmKU@A3^FF`WnqHuF z#fr2LBHA|&Xo|R~H>16Kpxm=JWS2qu7Ay8q&#e6w&SJ1VlfbmIMPYk9q)=!|+2TNU z4f54UMi0I($O5Rg4V3;|fUUUICquw8`F!(AY%oA(TTL zfddV+uy9vrlyoPf5iQ57FaDaklpQ}|@H%T)vS}g(GXW}ILwQ@olkIyxga`5lnyupx zy-)yhwger*!P|Q^exr#(SRsi!knd7e#SKRtKb}2)VgZ9x84Gc5`m4iW%C|9$DB(;f zuR{EE?;{9-H*ULuv}>}0f;g@kkR%NF)B=t;VCSppTakCJwXGLlb{2(V^U9MZLE$1Xsm^w+;a zK47GzV59{LX%$KMyMP+2cFJA~HR^T5BJ7rh8YWA!PAA@<7eZ6iHYuaWL_Kd4t{}mU zyI-C|b!Bl9A5>#uoPV-qKV>Y&R4;G4Sj4*+s<-?WMRjrgn0XSdQ2wnmq@AZ7s3(`j z)_9hw8f!3loY>R8-~C&M@Y7xP6pLM+NLU&$ay&|pb)RpT=975FyW_AP&d=BzK4?GXfNo2PzJn?Z2#6( z?^k#L6)f?~$Uy*aqT`e9?Qh_P+w~?-a{{J1GpQG)QvrQDhM)P&q{dZ_E&_*@#0R~; zwE)lF>FZtb)h{PW8LyrZi$Le&+Uald*QFx!6a(PtTCTSg0#kUF{lOsftj}NsQp=-1 zOKM3wWkzj3rY-^vbqW&Ivu4P1vZJa(2cSqOby&qa?TcFKgC3R_TD6;GdJal%+>J5g z55rplZ6#gUxdv=waim+2BDCLJ0#K9W0gV4}R;j)4IEm*a3o-E_gf!3CCXQcq>a8{Y ztE546Vw=NRJOUsG=MjacC*p51RbhQ(#zWK?+{bc?Z*0l1J%TFADz0%~VGNjN zs55gU$eR1fl}uD`0v(|K%a$Rfmfor^rjpNF-T!z%79V6#uwYUR1KD#bWuCL((NSZI zpllYvZ$f>U{@)rVoqvi+ByCe3ugN&_eU4^Yq}r{44dD#*7{2u zN?|@`Z>Z})NFiZN#&&QlUfxX^lIF$@S4;k4U#Rsoo6~>4oi~Bc#i!Ol))R4w}N5~J=J}83&&@-C(zt-N}jLnM>CCG2JyEh%Ui=DV~!C^66Or~O$PVp zEE8DkbLY!5eA&9e+t1O*2ux!~aH(@nUap1f*y3jhD-kO2N08;YXmE|5Mv@E2Flc^=#qdw~N#OL#fhGv~&>8gzEU$Cu-a%K?z)+TiEb zL^(H;!8&TP?!6mxT_dVW!uB}2+yNQY=e7dWj`^tF%Qb9l^t(S1#)8+|Rfvpq9SG6Y zAAY><-WVF{@4lL&WWiUi^qQjJpUpLKI>@$<9`J!L!`8O5x>-WUBp!5wH6aXSsVAxb z^$jc4_ALbSD@Tf;O8EZ=kVz6oFR0d?$->ZCJ`*wbaz7if;B3s{9zzU#<0?!5e%aw; zG3>)xDgT-ew7h@XSa%?5v6D-v@pZ27s1R@YLiX3bEkhgEUH`M~Ao=&mSEF+BTlz~r z)s4{ZlDF$&+AoHEWzEQ9eyMnHh?w=%g&OvW9m}`3Fn-8=V#yME2saNbMwN}+Hl>iprrM^Eu7$bfR??SN~fLDrY2`Pyg&pHWZ>d0(7CoR z5%J?WUJ(`UGn^0(O3KlZ9O&~5c-7t{Za}X1SY??@APc)4A2(_dzp03?zp~ zJkIpZxuoDYwPtLE34pm(Ixdd2H3Tw4(CBzl<7OR$Dki7o9p`0xaTd>WCgwVTo z^-zowy_?1qm2KpjM|dVms-7tNS%1zkh!Y$GqUZP~gGZ*c!14Slq$svE7A15qJ0E3~ z|5$H)QS=~~wLC+~1x-Yd(8xuJ+bIF!BxfcP7^stFO>=7{`r9*ewU5C@S^Q*55l~Zm z6N%Cm{rF6Xi!eF9#|vLs-vh370j}vTfT~zgh3MR>|9tez%|h96IiiNjAovLFh4E7a;-FjN${wgx54i6`^#-mc; z_x=44Z-h%f3(Y!&)I-OJY)A9e_`xcc)r$sE%**Ywc`uV!S zL@lCRddt&FlMboWs{aOm{PL+)o_kiiZJBR1r@nu4r@q#XIapXdd@m6|p5}hpMz03O zsVXxjiWbG}pPx6d`ZV952ZBAE*t6S1QM(Dc(Prq_5GPJ-W-pwr*#=49WyiwpM66;! z;1WHND`rPG`cWwp;)Tu9@j_=@;#osIIX-Om9$w+$PG~K=Xtdrr4Ryg9KG1H7YKqu; zMae=AN>d%jS?Ts7CqwZ5L(Z^gD2@ZXLH__tXrH6LGu!0sEfE}xGCv83sZ#FYPJ-Nt zEW*=dSk>7&Q5Nr#zGQN9s`qD&rwQK&Xq=m6(*-kqoL=`nx6X@U)xG5LanSyA(|eKS zxcVUPS+li_&p4j!QhbdEjaSixgvI>q?`c0^Jsw~Qa@C}HzF*wMSf#P7S1&E%lLA^# zfFLlo`t}Y<^77`mvVN6{GsODm-3iAja3-0}98d7R@t`{$o*&UkK0K2J2m}|RSvDLm z*vjD~CyRbqokxy2t|_F_c7qSsbQH*A-?df$PL8_8sPL4BmDjL`9Q=6D1$)|y}MzWc*SV%pu$^pD+M9`4yQ-3cva zYobqYWWDYaS?Y+lkI4J(FEJuR$Rb9C#^Kk^*vEqW*hWOaGZ$Rh=kuEH;_#J4SP}^e zVeq~}u?ijLXq+kfZYmzhkE+I~8^>*5r&Lh_?DgQ0il|mWM!!~ZykC4ZKVEN2?yS;R zmr}QxtIQIk9H$s|QJ(K34hCdF$*>BQV(55YItOe&zht^_=|2;onH-rfWh7-4Z;@D* zieka3vtSdFLEs42uFqnVT?P(+)QfT5F%97oBAe#p3d;t`Gd? z;PF916flp0^y$98^ei|xzYmfWnW8#Z8!)r8kILbUi%VJiJ zz}uCQ5zx`jD3<^PFFm^c^GJ%E&2%kf(PqGtqW)VmCIz{m`IJP}3TQmWRa|c|u@JKE z2gP?r9pX^N#P&#>O_M3NligkZKjP<^_pZlkYyVUtp8L-Ha+SoMV=5fK7V#cu!B&y4 z-*;4!_%@qTVi@kUG?RH@?b^D1c{C>bdUk!)27tjcjO(3L%3Kxc(f~}#`Nm01y^VpX zBEC@nfx?!i%2!+rw3ub;>MeQMVqhWct8v>Lc0h}#h9-Pb27yEC<&u($Z8h$hmL0@i zP||hmMchs3X7YA&DU2(n9}cYD-Pc}pf!dP0aNd&?-92r?T^ExK{Q&WqZ*@lxl7|qa z&fgD+lYTSB@JWN6l*tcOz4CQ1OQ&nSypjfi?GuIDpnQ5^>wL z&Ju{ZJAyz>#d$%`Z(Gbm*}kqPyIsw5(vG*=W%KRowai%-RX=k8)cIDrcBRA2EfL=G ziBJ68@!2+ZWtIsEgyQQ>s4I84`IK}V!?=%)d3;&)HbO462N?MO%n?<(61tW+%PVm{ znkMm8e(>ZLNzOUmzhr5qWUf8~Kx1mW+pVhlJdra3oyiGnDc-jb?_lhw;N^5RCyQC~ zlcmIUN_XDWvQ&QP^&#^pY~KC;+qTurNSrY-OaLsuJ6S5T+7WNRr6dG7Gt|TP-GKb% z3fNp?=3k3LNExS95P31wmtTAmqak*Og#`&$^{Mk!nUv~1{Wv1!K5>Lye`BjVn@e=+ z-m<2P;~j(jyw3%HCsO@VSN<1VB^FXn|1eOQ$n{sP)JauR_5T+B))YD`OG2zH3$(P# zQ_=n})^LiD>}d}p`QqN!xzdbTO$jdLk;jV6asP&=pqF(-FN;=yHv0TEm@ALo zoL1JGqPF&UUX2Cx`UfX?q`>dz1@Xk)rCEBAP;hARnhoU4DJtSY&3IROnpwR!J`PK& z(!Wd<4Bp$>bKxdHA6uyV*A=A@1Zd|HfsU{LI0krv9BaeoE$?naPwoKm?79x*eZ=FO zLvqzy5aW%4>noM={n#Q2#qW1=S%qSl-E{y%7>t2f%ZKCGZzPcM541V=rvnckZ>Leo z+lvU>iI`7U1d5+cROG!fw!|sMFu>ku^3pzlUJ0morSh3!>K+3{Xkjf;w|g%?5>-=G zaa~XD3QV0f@ltEbJhkW*81*HVF zgi{pOZ0qC{JMVlce67>fgx}^1ghy0|#HNi?laudS%1}14y--4d+LS<0voQA$6ypkX zZ5eP8R$Sy2I9(eVy2LsOQjg};wIjirX+lAz*#v8q=)Z2AF2l73rx}!>vu0A!@j(kk z5#7Qn9g*Dt{f+ywjM9B2ab+F9K!3h{U0nHHKLv-_?w}3h1YGeoX620s!GRHABdl0N zNyxptXrYS4Qf5fU9=jA<5K>k*p0Y5LBnwu}wVi`kHf^z7qv`>xFNxwBOC;1)Eif(R zK-ERZ0h1mn-Kyo5{~l`}`Wa%oo4E$rz)#o)NkRrC$M>CuHHO_L_USiDWbu36-(GMKr5Q`#q%#x2MyZ~>w zQ8>4UoOp;jFbw+Zyel*(**KiR%VlIwGB)+*DIFXz=7Gq0Es}`q`fC&+;UHh9Uzdsa zF77E#lty$x+?OAV>)&{bnDh18Lbo?~k??%OSyGU0It`(WAYbNLHJVKcm}{#Ufkkp_ zZh$YFM@Ny8o7epub5aR7Lxo8y!_**Ek-6loce?y^gDKtQEqKRyTXv&W)8)|=?cy3- zePHeVwl3h3z`|ZES->vq_xGmP5v4dp&)#oed={{)VNj+IhjFr_5NX>v|D^c-9EZPn zojuZDFW1Kldg5qGkNm<|7m6E`pwx7y$7Y(QQyy|k!JyrIjZhrbQajjBL2|c8f>?~+ z?c0)l0ZWrq)997Sl^xjN53;S5kce;%q~2XzAGq2Or}m4Xo-5CAf)Qc^BO zNMNQlho=9akh+?fTK?g~+un3PxoD=%B7Vb34OBtKA6McuUU~qU7HebWi%b9a)G4OPDG4#@U?PkLhpIw=8*1S3ig`+L%m4ou(ho{`8grBrl+%)7o7TY^E z1gB%J_o*>F0J;F`Bu3c!oHi(fwx|9ATw2ik2H)p7urpD%y$Wp3B;v0K+Hd<*Upy{~Hoef!;;n$!|p?xy^>8C-!_t&MKxr zwx9H68}IwMlCqE~ZCw@0#auRepI`m2DSQ$n@^fzK_xEWYYv26Gm*OH$(myVmqfY^* z=J>Jg9jjCoV$ze8fuAE|l^AQ?_2I4B8i1v~Dn~Yem3N`h4h;TwTcAFbc3~}za4a_O zrq-vsLK&WIX$K&|*J)P7fMm-~uJFnf$sT|?2>cg1j`Mh45; zq+PfU?&di% zx2y$!w?pPq^i%{TlzWBU`e7-{NC6?T2vM79$Kc#{4Ad++pPtASTZ!YR#2`hgv7PXH zQh7X!g%8WV$tsW{i#n7Qm-LMfXRUFpZF11cp6d$(`2=CAp4GByjcmF_y9 zzOIi?NzDRqiR#@3A?w^PwfUWQfcNOu;+mr`k8s^2p+X|aN*=~uP_*dRuy~O2h|NNl{W*usEV7m+EDn$J=5f`2fLm#$ffNUu}#CHux)9R;jXB46d@`s zpAb2bKk6yDHZEUDT=R6)dg1K718he)~5;hEbT02Akki>#ag`JD2{FIyhVU7*xi087^2xV5R=DP zA}z*ehT6Neeqv8@7KWd1YZoK-E0zhI+l{14f;3AhS+pIS;0D_Xd?MyT;L-fuMY-VF z`=!y!?8UC7Ifdfb21n}F|9CCX`;$py#?GpQXa&UGMA4?%vizMW^(|wgUi!9`JPM8@ zu39QS(X24W6q3#=_6qHC{Og64LD?dv_WDPQs_d(WdNyUm*$77PKooc?k7%)P zVg7(_;GOfUGn~rJx0EY}{XPl^{ zD&C;OwBXxPs!HbE9rCN$B{1-baDJ!a)y8lD$ z{Hcp;&IBl`fd@p@+X3VSouNXUICgPKn5P`Po)j9ke}y;HYU300aE>A%KG~qa{-e{s z(^54f7pn6W5GgoK*q92_%x`y^QSETCxK$dZ|u zTNIUbN9*9yN?zP*5KnwQD7vHG&+7*d-F8bDvi>`T6{br`=VzYh#?s|HeKrMrrIC&%7CW>fc@JM+;UF?!;UBe;; z;Nz7pj3#k4pYKZj{=C_Op3rRbZJ3C)u35;CIX@U84|O1Q{}Ktx&@q~|#^R2(qq%rd z_dn@)xe#p8Q>}jz?%|fmw_SwDC&*XkqQkH+*N_X9uSVsD%4q(HAQ?mE{+(jYu>=1w zD!M@Brh#k;j_QvQ@nx4>%H;EmiQYA=^FT(jfB%)WQ_|2@>#Rp+x`UG z`1^WuaH<09`jS``0p){m>NQN392(lNwxqSS%uABM%581yx;rw2vG$lBECn0)LDyy@ z(gu3dz~14*oc$?oU@r&x)kdFSFBEal|Cwn!Ep%a3Ts6YwB6V8uX}As z2S7Lg7#=L`AJZY8vQzQyq2Ah*{EU@u z`nW>((ton!rMTDvDj=Co$30|LYo^d;k3Gs@H5(#6Ywif;^9}%dmtalswONlAoi#7- z5}<^gHE-ce#UL=^PnhT2ov_$z*#VOcwr?Onv|<~QDSjYv{C7u46*fb$ap{H_(BgM% zLB!PcpS|n==+XrJURUgV{}kD$m3b>6VRd3-?vM9oBqNBBvPlxx*#R!d25+owP7|aZ z*!Vsp$2!qEQ*gGe++&1f9a!QuSfB@sW5Kn(MFe3?=^eG1w4fAD*7G09GU)Od7-ke( z)$Z2-T(AABZwAlt!w$SYj$T3ucAf>nx53+_2bgdUj(*& zYPKAE;=X&2SF7cJR^u^+vG11^o>syHRdeOOmqjqf0FQ6GEsK)JPaje`sLl(Sep$bK zscX!bhg$>@V;-vkEr#a$_YAjvT`-#86h3&E3tF-Vz~i5=zL)8LwD~g$7+RMO?S~aI zW(jNMl06Ang;W3k8Sei`z;RxT;ND$(s*@Vbk?*Mf=BDeU^;W?;Z;ran>o}&mvu|}_ zzs@WGaEj`;v)c})^FVc{rz-U9G|L#`@ejT^1FOFHnYsbzg6Kf77@`?|9UuJ@4t`Rb z8L-k-0NMW{oJo4Z$4T#`q)2Ke;jW(_sM8p^UCH9D%`!8C5%i2ZZgQySolv*>r)Ls{B)IG;~PRif$@0;1&wNF8oUML2( zu*47j`)A$a!@#ng8Xf^VVCvYJuBd*+yJ=+oh<}jxJ>dp4fAKHY)cSqbbdUb{0P9F* zlrPept5eJah^m;=*nP(pl zpR-H7u1(`Svf=@!S^#9cJ%V7N)Ly(xl}mCZz|UR@&CMJEY&6^V zXv zs*4`F8!av!b@S6==Xv)n2v;0uEVX&ndBrNY;23@pZS{7+|5+C+S!Yn1S^sJF4@jE* z0M14u=gj%XJeHJ1R8*!JJ+Cc5DeXrTlYK7LHX0CF;6jdS+?wUC@|m<7k{zSzt5tXW z9Z<2o!Ye{(yHp=}JmK83U;2FeRp{syO^Bj@Q^30UcNoJU6*?4^naEV2as}Mp3a;wn zuwk={VV8>t3f&R{klZ9OO-6e2>x$?O#YWE6L3KTnObPNA+@~4YI6%-G0TAMaYtwQR zbP$TrO}bc}$2U?8Pl|uowOjU#r&&;w#X7l`<`W)!13}li%nHGXA31M!_o;g39yssN zx$vgTzRw93n+)8$#=lbP;V_`GSSOP^eT7>s1Isn#JJ7blA1ct24H|Qn;ZrPwt1nmE z@lHI7v8}Uk8N^|{!Ny6K&tsBga#!?(Dg#u6Q7}5+4ZR2VaC8b=ygq(YyjR4F^W9mv z6=BFOanoC$SCA*$K4Md}Wb&|g+3HSq>3wl*4^nX;WeozZ7vb4MNVD>i4!fAxYp0W~ zfOkgF6N^G11r^EJ{)ATmLKcwdE1NGBCp={puZ~+kfDk47CF^#*fq{bNd@GK( z4M^u6sj=KVQ^`ki$7OonYqh^q+1N$ZRrNo_SKm^f0L0z(nT*en=>(`hp{wJrL{pDT zjW>Wbc=lc&kCdXEy`TY`=daM-VZh6{`j=g$WZhG7tZ2MDpTN2rH1y1`5uw1zZ0T8t z*7XDtd+}pxZG~AOE8uZ!)rbG7x%J%moTx#BulZB7;N8!(aL+}HIoxfiLn2_K`7YhW z7MWOET3G#o>RWageZGyO9kH|LwfwS$(c<|jL!yA5>wo=5>KG_*jxZlTo}z5Z9bG;w(8)p6^B}tx z?sM8fBGC6b+^_x76l$SA-niXbHXiwG6h~=+6mPMnEP0zKp(q%IN%+(MLxawut!6+u z`ymlC+DkjlH$bS(V?y8n<}s()`)$WA?IY%Lv_rv1@wtImOL=pBf6N5}SvUh>jrheV3XG~-#J@3U!{Nc~ly!k~pLuHc8YjM3!@`j6-%zF`x zA4tW#y?cBrBcgb1ZGIfiB5b<3x&8W^?Rc#J5Ek$9_Vw4N4fL1m`udR%x_`Hf|281l zPbcTHzJ4O13G8{o$nKE#OBld(c?$((;x>FP#XvNFP`D1{@kUS5#!9iM`PZuEUFwYM z4*@Fw;k#s>ZrdOvyeQcMvM4F*_s%C6-<+hy5*&jN>P;Y#g8-YWp01W6+RV_Bb>kjjiQ>UGUio-04ow3TW zRVD)wF8rjw-QN>_?HV)~Z7c65?X~Oq2+~s&D3gQQjkDV;-;DJ)`5C%a1tTHo zSfWVfeD_)WbbNv}M|!g}qW>%O8z@vU9s4D=13K}4h7 zqiRck!n!Xqsy!X}-hp>Iwz2N?8C%dXHHUnV6`=PRT0Ec0b)7Nj4V^6OWA~8=b3;de zQu~UtSF%nuhET2}D;+@d#GfaJfvm!!L1@YCf6R7HrRHB8a{pMIfcnMFwmAoJaP{j( z`4c*#3~r(y$$hVjzCc59*z~3|nF1|Zl`WS19{STdJfMpO67c<{vB#;a zBzlb0mb#g&;<~&Pots~cD=*c9b2w;t!E@M7bG$h z8c5kE5-Hi%N6|opm|G=;EC6L7NDDNQ0SYz*&JExt^^`2wfbf9+qZR}kAwl4Z5Ih?J z7l6eJpU;#A0p_)cg~mVLI?wjj-hIyQdjC#%KWwIo6=}|O-tBo!>U2R*l&v8F7Q3Fp zUI&V88xHh_yC}VVm3}1?QvnGf8dPi#*TG{JikYdl6Uzw<$A(SXorvxi5hC zReSt(86HdHxC9{W1UQ&g(xJ>gBUV5b`}QiKa!EIH&uTs2uHsV7z4# zRs=4IBNANkyT?Mvh9ZxD#bpg9srICz_P>>g%`txb)NR)i4OM3NY(E3Rf-AbisJ&O> zDc%DMgDAWBA&RM$s0+{O+Q@A64eZD%yhGJB>ZaUO+Nhg$QEjPi)K0powpRV$q}#tS zTo>Gi9Bv}6f@;BJ&48|TXNQe(LUhV}!i4-)@Ni33`84)QHW1kPb>oE54Uy+%>_81M zOcgSH-qyx3EhvfLh4=LVBFP4l$ z3zGD!o|t&{)h;VXhH1&u=Z3(JUv3Cmi$Tjp3nU$bz#@g*kiGasphwZ;N?U8}8X|zmDQDIv9C=R<5D%22ju8 z?5!*Ag#-Ic;H9wiCwDZ&gow9Cv~+h}76qdrXeAHxX=;*28eHPsj0^EH{zErCRw2FswFVd{}uegQ8=Xz6VbS*3it0 z9OQBAQ@BIh{k#E0Rq5}?Gd0`ODU8lTXBb|X0@fe7UykPIby!T;4kw!!K*cSeFq@+w zt9=EDoRO<;S?s9zJ>;wW9}(RW1kryhn0~1?!RoOz7xiq`PsW%|W)E&xpoK>`8=4|v zJ5B^m*Vy8&8sg)9cAq%2OdOFzFOnQl@jX98*ZY4l===Gqi&^QN8%9H)ktsUKK%(C5 z91%QK%VbvdH!&HO3;co;K6J6cVAFOEOnTJLg|-@Y5@&&E#Tr^k+M4HCT!l z%fpyE^_)ms1soX=W3}+CWN`8>EE(2Mf(rzrFIZ1kwVoFhH_7=7bTsl=}7<tyCF^2bQE%9tycbASw_3* z1vozdQBZ8YoX3lRj|5rRO0G@iq&62sTDJa=CP^0JV2#D7tjw$6wqnw5?}h1%TJPVU z8{u#~OTnzH)CTO4B0pl@q}bV9;j*e3JNg&E$0v;mY)Khsc+%hnv}&b$dgepFJ%a&aST=||k0pVi^led+ z2$bzN<$>C$+m-ujplwv`sslApcB&3l|4-?US+axciCkm`!FNG809#({eP2`ECIs71 z$w5?a!ciD%*+b%NTHUJ0NT!DyU88t^(_1}GXcT01aLF&}%|21Sr$Pf6`fO2y&f+wm zi9Pesf<^BrbF(R~Au;@=$Oe;M1 z+?rl2itq;$`zR+#luh{~?QU}kWb5L=pJd%B^&z|T=(jW3QH_RuXK7+-p)}9{7=8}g zun2#A{LS%W`fJDJiL2f6Cq7D~I+-T^diO**8mzPM;p8Z`+XjS{={=K3cZ%8CAVe9b zf4&sVSoK-rggs(7-^*mCU;mQZ7x%I2XIjC`@zEPeEcXggVq135tO^7ez1cE2OKEPE z3sieLpNS$Uu7V*;XQ9oh8XM}#?WkTMQ_kB&bZWuwV-0rb?);l2$Bo)ni);B9U=B<+-u;sU672PWcz>I> z2m-n5O24I$)HbbIpLuez?&neNDs0Ew=3>uL)f1c@18qNMJ-O{zO2#W|p%d}dw`-((6ND&Lf-24G$zK6BzS7V^#`8?4bEwRUL%V+@% zR_#g@k>Bro8G-Dp&DSV?ocOf8v_((_S`VQX&oskTvlOguavy`fT?Wf4NsNFBXt62( zm{|Blg8g$)si%r8Y%>p{k(YAIKV=QHZ5K8-rn=~IEKu7)-_x6LisX;FP*!cHvD!Bq);_^1GE^se^VB(fW@fB#*lPVG8X`|8}SdM>-GSHF!~%bbZ+ z{!4(HjEekMU3BbQ2EVENS@$H*#Hid1}B0f(Y^jRjW)p<+Nan)oVtJiCd^QJsV8FLy8d9nazvw1z= zMTs3uyU_!XY%7!dxTtkXF3ARcOLQ)}SAQL3p#58G>(_hT%CKF%%o3N7-k!dKI7ZHT z*J&35?KmHdb%6!m%%NLo_el;?mBJRfIiQ!95T5#Bfq%$nOxAEXk2pu{-rPUq468wl zKX5k`B%|%mfN~K3c&i+LTA3sy0>5Q+0UGRu)zp<6q#bQAo}BoAR5 z6TGTT@5Qg>KkIReR^x^_mH!c}-jE*=K%z~Ad|GEG@0@rwIrh`Xp67KeM*N4~qwA9p z+$dOepyMFBDAIoctM%Ito;{P4t1O!(H5d~B*8{~U`|yAhLu&26ttpJoA3Tg0tizXY zSgVqhr0&&><|yJHW)0jdbw-W%F(fQ*A5rjNiSw8#D+6QAOK-D75>q}Zm=UZsd)4@4 z`o)E&8MO%W>Hmb}Wk@_&PH8OG8>MiFF+_ujSEA&4Fg4*MA(@(GBWYdtH^=mDs9SuK z2~)L-GP29oTv4b_UmmH2J4Dzsr=rZj3f##K=G1w-qI`wCX2O}R590oMV?t|Sa|84I zEW_=TS%`AYM2FkEbO01CX4WF`fTCV#L|P=DCbURMXou}F6FBvKND@pl+}l$v4~={@ zKWWyezHTk-NAl+i5pz0x5V|;M@WSM6Mv~1j-YsPzSUJIShGy!x7mbc1-aIHV0(Wg6 zqgaX)P+U+@Iq`$I-i+em2Z%g8ha|%|Av2waNJ--wx1ARK(~yU`QP*ZnWizF5FmJ>V z9tah%!#WJvG4raiG$(SbNzKmJk-WZkPOd}e;l+%@G?T*VMgc7!vfu;PUIqiw)C)FB zzcL1ES2-M8XPa{qWH;5v zw#4F0<9-Xm_Fe0=}vFarMtqM_Db&sgbOwbOlEX3vjgc4Uhqb2`0JYCRauAViloy zDk|YU89KN=$4p30>!KR};i{3*o3Oa?oooPQi96A#ne|(=#gXaX0)+GkBc%Si@FANh zW)C9sZNC;*9@lPU%LI_=*h0P$?K{FI|&S&s?w% zt2o!pbtoJxuQ;AM6*>bh>mNf#z(Lg|%UK2nsP2D!>3pItDm%AKXSUeVKbZuZL^!!L z?$eJ+bu?rRXAyyKWK^nlPw;mE-ydY!GJl=VefJ3dJZT(Le!tJ_Ofu{hedAc7bV3d5 zlR|Y)aFNXL)0i;vbxHCH?|#3BEREUcY)xBel(}!;p>xHc#3Nh}UPx;`l7wKvYE~I@r;p8yg^>t{r?t;< z^dubNBJDn4NZ?wzJs-Q9+ijVaY98>i%*WJrQp03iP8_cLr|=cP>iM?ru!~mMJTr2B z5Co~j1rJ{V+d+qmS%@ttbD&+W)~B4ci!Zdx;rVz(x&JApEP1?8H6>TAWV6p)*m_#W zqL!jZTpK&ey6reM{9pS1)21qecb8mIWqdU_#_?Hm5X>_7OJCDy-g!EhzB?(WQ z^S4$X3CwA37jtQ0abVnN%bo3fX#ugG3EJ;M6NtU?yMRVn2q%5H2*JfXnbjxc_%Com zN}v=36OM9elo#+1#69FH!WsX_s6$g045|!eZV4~3JQ{elNP4oF1ZGsb2bf?+z zcxbN|YrH@0&iGO3vaf~rNgzegtbUfXQU(hpi@yal>TVD~QzFn(+`0>2$EOFSkaC94Fj8=N4)< z*!#h^kvHrSL=W7%?-fV^0Rxx%4Tr!S*Pg$VG@j(ar%Q;tT#t6N_dH{L6Q7sL-|ZZG ze%I_njh7p16zJJBc>a&&U)=n#N1yZj=%mUL)36gpbr^{z+h> zcsp_fw`3=!zLoj=T>7x$+M~8gtkm@+mJTN@img`?ik0l zLp=jRK(f6k!}7ohna>|bO`Wj24@d{iyP=5#COPx;szEt)x5|3ukQ|z86~i(Z9@X`V z{{wvz{^qIrX1&&V+kIf!3&1EtN8b3wzMD$7@}1zJl<~J*roA51l9wkYSKW)$qDiY+ zD9Sq3VxOHiSJ;t@WkV|O`s zCN~L!8`9>i*=slIqk%e=$^LOsqXd+#a@9)n}q zVmj>p9@x)-g{~hwNLpx06oRxM>%x%f$Q`qKq6FZsQv^hCvei_@5Z81jdR}axKgmm8 zsVv9)vR#(2F$Po7?68z2vn0t2E1%hs`0Mp&Ch)eRNjVie6biQb>M*Rfk*Xl8<(Sfy z-t}3ARzZMF@aK#6V36MF?UcboL#WD9A=-I4q#HOw77iUU=;yIxw|TSihaiFXUXjS71I(9k;q<5}g5D~OB2 z+>_0qnlsfg!yUOVV_+loe(1_4vxdum#3&;YejV`ak%fVUt8#_Li!{Z_$q^Plgpm@; z`kRN}J+>$=?s3H1TQTR*P_+B@`G|LSu7b}taFXe&v5dm;)5O~)s6Gl25sMO$myo?$ zFxebB1sH_O<%!=s-a5Epd1D44>I9@A zxu7YCHFi0aqvUM<({Dz?HMFd~Cq`hlD~}?Zt@m-7TmzxS-8IcZ>Ef1eMC=FFIMmhQ zf>^#DDV)9?=^ERU)-lLppMv@72Nx~Jb!{z_ZEU91WEzR$BqAnOqa$E@h*CqM=2MFN z>3br^F<)f}<>mWp;=&A=R(mimYP@ftp0j38q@zZrJV{V*VVpgRS1^ixY6X+h;H?Ai zqQ{A&>gC@orhmKLHF^jvVK8t$Bg~o81@^D^J%hIYHcd!e_+PH5zN{t2rP7axjo(`q z4J^m9TZihB5*AqGEMNc_(F@Gw2?W}mui}T|kgD2NaSU_&;Da7D2q7LyolwxO=U>7O zIYU8TBT@|d6@dPA_lC_MI%2KP-FV@?JWjc909^OIj+=mg|I7o9N7~Kbwm=vi7X#&#Y-!)K*4^) zAn}z68xn&m(lqbMH}(1caE%}d6?y$`us*+uWU2QDY&`qr_DOWA%Ld;+$R~sobu2SY z!bdJ`(U4=I)Moa+I(c~SYaPLrl$~NXf=a=Z{~N__osVq>kW!xbev?u*_24xe(AO)e4yA&ap+x)^)%Y) zTlriQW@&@!J62=6daz6k7B9hIVRB@1QG}H2NwR%*Vbv3Kbj&VQ;kUik0i{N zdvR0pbu+w_z}uDxT?@uxeK=1Str_L_+PBCU`;^Z(+23?`6I6YW;xPj;OBVlHnjpCq z8NL$+1bw_G?ES3~9Ad(&ogq2weH^an!X^i`>^C0IXHSGt?vz{oQ98Dl7g0(3Xjee1 zP)6b@(gp2(mzcw=NB(UNcVa%Thb-VGrNmb0rE0D!-)Xlat{5P<+(Mcq)GXC$yoenQ zJ6*+k&g~)*(tSw6R2e-ZS3LgKd;Y#KUXee-gwGFi9`}FeOpr2e7SqMy#Kfxh+GmT@ z7TVKaPrIfeU|Wb0EJtI4g=^(wM?PL1Rez^80Mh4624F#Bh}y6( zBJ(7)yRE4h->#1;W5WYpV*%kftk0KNFv>;^Fv%ObrUYIvGxWMn9f%xh&_j1AV$>{o z*GCb#eBKXMr8G@#>ws5=6usgj8NUK>qr3n75ZAcv8Im2aW zwsz{s`*NHn1w4?(26?iJ4oo8_iIY-A5+OgDM8IX4qV&Zg z9Z7tV_`Oaif5wAd+rEH(z>yl0Res#}T$ODgD_y4kD21sS{@m8XAjqKI2)f!$hI3Ye zLY%~3BEQj4${pFp_POO@CYV#*pRIFryq_DcF(>OF83nbLaxMgA{ho^@Aa+5T-f3Y- zTH5glOgX>%#4&{m=xg0Yua?~UC&$jscGZHZhl+j_z6KEs$basQh|I{>=Ce$JABywj%2x_Ca zr;y+wcNi20I1{jIyv9KXqT~>*-$S0hvC&g)4Q?7p)taE~u)qs)?{%u4xyv3BoyA(j zY4Aq|H+&L&oVMcm>Wf{4>hM@Ay^qDE^mF|lbAJ8jei2q^QqH7WNeu)Kc^nTNjTsPm zIGsHh93U(?ut+lN!6Nd8S&#Yv7cOEJp~=gcR<4%XLA|>6DQrKdtV}MIVphx&?^uXX zy(3&KS#VAxq#mRdaCaoy3xnYei39-fQnlFw)%}%=AD8i-Pr;s72G$s9hd?5dXw1ut zqK^?DX}DA;Dx~C8ohlQRQ?e_MRmjW!547~-b=*AFM_RTNGG}qQjDkUJT>9;P{Mhe4 z#5$W;^S89Lvy)iN4c zYD_J~BZxn_8~@jOKr^OPkrorM5ezp=%*u-O#VaaW`-sGmGsJZOe$Y)UPo&J(g;_z9UW zvD=bzs3ms#R^|UAijDM4S;}5NiXpduO+e^SXJpWsPxk@uU6GEZlUf46%NSq2f+lAl z;HT^WN0HmY%_pWL#D_h_H^k!AOJ=?O#XG`)mA6mv9{^k}Uq^M*%qbW-)+STXcS{KU zrML2hU753BWz$A&ZT9+fSd2ee#x@p-pW!$+k13jVZ-S1Vx!TlXX><6STB+nq$*Zw! zBgqRya}j|S*q!${w0IJ1mwY}02Q(I?Q9JUrA4XGkm!|V*il){0#&X)a=)OyEs4pM( zOIKBFSf^^EQkJmg)z}g>?IhN<(0-r^?AitOaTXje6ulXz$wpQc*17`@REs8Uis1!d z6&I5D;W`w8R>WCrPt!-<>XyucwZx^ue;s4CSJMzZMEl^V(;nKY{^4W*Ins{f&x zpS=dN@x4)hKiQ{F0wvK4U^oM~rn-L9? z<1gn3$nUw%=6Q*~bjb>uou&Tvf(a7B4y;a*QTR_5HkW>}*9&G5GEuZKN+>u5O6FTAm=C9*#UD^i=|a}Vh{W073}uy&->e}Om37P9 zuN3fGu**1wexXBIy|Jj^0bbq5dnVWBlcdYDB3k_!1{!iX^|%|{JTz4%Eqoveb!k1n zwL-1DI^b5oW39DtMOya3?=KZm+!|%z{kr{}&=agKgkQ0&9>Loa{(h`bBqTA2A0F2c ziKH2eboGimVr#8FooSkIF_wX)z8sA@u zwfI5Qaxz)@`VQZZ4;EF@-TGKmW@4DFGAb5Fa2hliA;zO)E72YNFM_HbfBcM(R#NUC zJ^+Wx8#X|O@lD=}Zpza%@7KG$Yf1_TRlIV20J$69ss52sQPH);cRu}$Hjebpi+cd} z+!|6LpnD+tiqc?%H%6DHgj5oZ4%21^(b=1Wy^0`sep4_l);2bI0%bT;VF7gQu5lLI zJ6tTAEd%S<4T6NwZ37eYCXyod#AN$RX^@{V)pbg5->sgjALUzl-bN;MNR}=5@dtq# z*Sf)Yo{&x=mzF%)T}01Kd^ZY??wY8gDlJ%_P2t>~k-}2P1QLFgqt{3-H;O)aP`^*{ zvCPDoyYRg4Dhw@YOX#{&7jIl6Db2oCS(5lOF~z8#wdUI^l2MvKXrf{?{%$4>w2;KW zmK4R}w|&2RMvHfwnYK_meqds3%X^SOIX~I$D#}Ni0TfI3mA8Jw6FBAFpDYW&YM}}* z_-E}vQh9ZQz1GlnLZ$*|GU+AJStOB$^0}VV*Wv?og_(#bdZ1WC=j1BF-iyAxwX5}W z>8W~ZUu9TS-IXCQO^CC-wc%hBRERK$urMlEN})bZq$(O|LL$cw3F`${FC-7^dlePf_Z!;fxqePr(pInD2`cntZP}#Q!JM)EvWt&YMCNL$A^S7L+CG9|6 zTszS-m?kUvz_IpQOAG@_b{toYyZ_NY>>)Vrb=KIX`X|n#=s8cwV!CCleFM-QhR;z+ zonHOASJ$twH96Z{1@n`Q8RN)wY~2u=_@!78X^rvSK2G(s$+CF3a4Fm;R)L84@e5w()4Jmh0fato`j}}=wq)o zQ_feeSa#aCK2m>Z9C3ih@O5tK+{bhObtkBW z6=9D1XM;iI9eu`i{ZU>olcLy@FkyH56#rr8xNMnnta|naXLOLDJy3CYQg2bTl0K?Xd@CGXY)hm~ZC=Et%=E?1! z?h34I8FJYQu*yp~qYv)OGL`mQpO`^nv!-QUbXL?3r$9#D{>%1^(S9mi$g>EfQ^_*L z9(5N&cQzt4ADJ9WUi#8^wUnuLZQ=dp77o*`wn=|A3Vz<-708vW6S;aV=}8kRyAe+*z-gjRlZMllTgz zJ+P2tqul+Q%H9r;t`Ya@j{f%vqlKVK9O8wa6EvpX|3%RuubCGkWSKzpESKX*;q)lJ z_m0l8D?S}wy6<_4r-smiGE%!Du$6)rdg{uUPrQ5ht1Q6| zYc0*$aoVbr<3ClHXcf|RvOP&}W|_RyYIQ6rssdcQbYF+qk$O@zY3Y90@y-H28McMT zFS9eR{#V_CHfTu9Vrqo7k&S~jA)ULgJxac7&9LG_z453tNg2Ep8EQ*ORZJn5K^5C|G15v?j zQ-5*iPzfKROlO62&;e{!n)`eeZj3BX)#ZNdGRCrf8y6|~q)CP17M*&gM5grPTxnbr z;kWltB9dRx#GDX_Nu%k>yuLq_R6|;v5GSmk%IC1+uac`pItaj<45~(9o**lV3Nu?u z-d%WE&^%8ifk?s+jY)P}c1`Wbk||(3%hbY7pd#%Mk(K)bB42v^}0gPqnRi4&|g{aDk4a=ic=3qsxjw5BhgZ~R>fe;<$4rJ<*R^{pirg! zsQ{BASD*<{f=>RwqVgY2ut^5G(=YTr5wdxGgyWhX1|Up;&`sZ+|H%jJiPGh48zfUH z7!4NsanJJwEah*pMzlSHm>*1_kZ>vB)(mT{_${BeO#FsvE>#*yL`O#Z2XfN9bmS510IOla?LHh zztrXJ&!1$;Dxxxsu+LV`?oJ149^AaBgI!B7+fps~UfVGWXNjF_W# zHCZF=(C?fw56Ak3lhT{LE7dG5WtSbzDqIcDo~BR!>h1DIx&^B0~Bi2c{@_y2p1$%#Nd7MOfBR1Er|tgA_Mws zYn#r8Am^@)#gfq+anu3y$^Ddm`)F>eLdlA)q&!#D5A%Oak)VUU3RSZo{0pxu{dLQ_ zj~-yK?wu6KJripN%CE1;Q0_{3G=Y#oJHW?agkf|ZZ}9hFKPIpVXN$abtJ#srvKpR9 zF>kF8qUBhg$os_?v>9c+m1GAHTDq&1lQOlY>CT3=>SZq-?d?pWG%i1qEspN{ zHH2qwq?=5*jI2woI)m+S6Z7$6c<(NEhhfkMmOa)?K{+#vsegsC5)S_;%z@Mn@R)ci zd!2@>jbB*)pD+W@-}ZXslVbE9z;5wLLdT3f&%Zumo#U(Gi^;rTNd08Ro8RW{=R0Xy zGq#*cB4#D2Fb}tk7CjJeCQc$eC8<^sGTE>3QU2z7_5QU3r&R90 zf+8sEj$V1TGh^WsHq1)sVAINkn((YD+d-I>j&qjL^M`LgFq3jnkk{<00qtjYPK3Z+ zEtdAlUTo;>()aHrXu69M5*jQ*l%+sw=y0frBdGi(wpcTHW9oja+Sf3c1U z$@GA=ky;HM0@ksco!c9euX6;in>4T-{mo{mMvlb1qV3-eQS!BS`XG`R$&CM5gK0kP z|3GAWFz_Km2*wnjd8PF-<$j=D-K(`bh=Jw02Mg{P3Z|ULs?umQ=k~k9KNv^RhiRLW z28U#95vRaaim*gVk;)#O-UTr2bEUfpE>K!i&?)jE&~{0O2!qCA8yNW8#&!I9;2bOn z`vjS9jEMAD|4`V#u^5e-pm*MgVu`ih)=w2NQaqmAsjFsyMp1vWqUUbsmmxRAD-o{N zNIBfvI&kyO}26}RB_cs*B;c|w>rSSOw#NewWJ1a#IG82J7*2C!^uWA>VSVd$1Xaw zIW{|=5uy2S9um0Vic5;kN#n1TLUS!bJ!&EjN@cD^t-OO@b1u}o-utA;S{e~W`pmj8 zEGzloLv?iA60N{3z`B5LWQE5U=WPdLy=t8&v@tqF>*b6wLAXOv0RJe@sqw^zNy{(j ziFb)ocbNSx9UbtQPU;Hnrun2rrIWwma#qxtGh1dh3$TOZ19qliWDAGc5v~y|myKM~ z)(()dUfrT;MZ1hZ>RJ~lDGC#{Q@+Q7Zg+Tqq15< zS@2-$>Y_j@a4-!Q@DBr%o0Cq>s)Y{jw_mAyM06g&DQRMLkl*4wK&8n5G%t4LPoO4A z+I(`YjYt*JRHKi=@&*>fm$;-L`lN1o5wb9(!Sw_=hp`Zg_)juspqcM$TB@yo&7tFp zE9TduzxoJk=I8OA`h75%%!LX`1b1!SpVjiJF$`R^&;%Ya0G{C}nqvKTW3uUcc0_&| zj{nUE!IyfPy52_{BJB{(td!%z4?NIkj)N88z;-|yAxtja2eJDz(x`NsLn zFebNayoFD|JIu4oQ6uDv^847?ZrbnHLQ}BA-HI_vcDbDgDGc>=BzJ;9x+)yMREBC5 z*Ar3L(q_-L_`$zjChB2e8vx?o)TPEph`|oXE?ghM^?tNI&V4>Tnt(V92<}ET(!pu~ zQ8X4W@ZTn-m1H5Y`hmVHwuQloGMt|&ZR?-q*+fE~BXQa#A=CxVy4PGYq!SX04E1wC zG!j(kxlc4Vb5U%7YRd@Oyei4D4le)Uvghw9nX&)^RARCcuLg^gMyV`ohDp6@qn~Hx za-5!_;jBhRh4f&GQnrWL)b1)L3NpC^eavfnvcdJTEb`%^2xOt{sBEe!yBVoM=X6Fm!Unsf^L}H07IhR74vh|XzzwE z!<9jucm4G>q_}V+FU=t7^2W;bsQ8)EZn#MgIh9f~`ZwZXu{&NCD@eo8e^j}KjNf;wIg^av)fz0AD;!xHCoa&VqD$#SolQW2 zh1}r=5nF~*Oqx*aiVLbRcJc|mu9@JTbLVhG$eg$d5eXi`qiTxL4v}Lr+hotOlc3s; zm3L^X`)!i%%a7JZP0~E4)V}pA*v@^(!jt#+t_-@gOd5ZW1IY~z%dST#|k9=I)oYdvA8 zvWcP~QKV)~i)t)SLq>dQAx*`BPcf7{F@ax&uDDIN$f^vY0}xrO!&$y@VR`})A3;8A z7Aq;c#(w4x{pg0x_j0CctW8+|%#AGQ7LQ++KdQPzG*s$kkeOV^ZC~~x;OhyR2EFQG z8(B51WyH6A)Bf@{vq@#>`}_aG_b$xOY046F$edauAoD~bQzc^jWaw3<`3XU5DmsEX|D2~}LL0t6>Z&fWA?N>5{iu(=arm~kg-F3Ey8iuaq@$ZAEO zF2N%eAR?f`{#?$F5&On5kZ>^T8?xlAhWj*-6ZgZi=!-(N0*-`Hk0v@dLfOXI&A^1n zELN#;pH1*{F!b2xkL?Afd^~=QD^f;se&z!kGo0P{^;`0P! zd11qm-8w~(x=W;!!Vaj&)LeDYOguU0V1~NvqSLWnaM=3cYUFn*7@rBWH`9e-OdDju z(9^1-fT^r8q%%sbEUuHim;82wLLI~j$X+X^OdneoTX$X_(_am9Fq{=~eqYCzel}Q2 zk(**^dN$|Okd#SzyvfM)R;Uw^9aZq?8cldB3cc<*>oDexSOb>ZtV@x4q_D^6%xtG0 z;Nh)&wW^5JLFH{eP9Ye03Lrq!tzAfFf(IE?n3vW1=~91D zHwh638WrsYr3ZH@H4A!~x}YYm7$sO?k~i(l{~|Mn4A`BqN#6|Dn$pNVF`J)6(>vmA zup}p}F`d*2MGgIIA5(l+dEa*E?u~@S~DN9=(9YGw+CTOW>ZEH#p)61tUc61?`oSZe35c*u*o1m^#pC^ZDvC(o zs$57MK4KvvV;ba_8m{D*T}BoD8nUkU;)>(R#CJ)-dgDr!h*)=w^@SJHb{^GA7~0W8 zhj!-Bn)8^>{)UdeiTN4Yk<8l7c>M@G|J*k@Zx~n`WX}aYJz!xjDum$D8PY<8`~+dZ zL9Yeq3mfol5eIe6nHSxu&lF{U-YC{MZBvKI8s{ymaU?t7W*occ2*5?`$O34pie%cq zIM#}u@f=uAyo4+?xD|-;@qqPtifPylM@?)qZ1@FqnE-j)5gk}|M0z6^?OLwx#x2y& zGSW;KPjhvSRl|x=@L|_ z;ZZ)jD=6EDuMQFj1A_5-^41Y}!bL~H^Ef@RwbsgR1jHxU-95iN6*JpLwptz~P_cYt ziO!sD+Hyz|hjV&UIAN+{MEJp#$j5`c432FMcht&CaqaQV|J2qIkQ67A`aPYyd3Nn6 z3fDE3pWV|tiy`(s(bk}4@R*S;AU@zpP`n?gB@k1x;}JSpSXLlPg3IRI-HFKqzlZj&Di|rp<>b->LT#^whkx%73Nj6My=zt!`L|#NbIxLyVVjCv(jd@s4EbRg-Bcp(c5~#aL!{~?IWg65EhhBuQyUSXd zg-q|pO8wfdMu^MW*+ka|VDXMAQ$Gf2*Jo9xmS&}(}w%WoG9!gpX9qwiNr8nHv@$PSvH`kIUT z*B)g|~MkD%@fYVMlSjawhVX3Oc#x z;R4`;7S>pVTRJWSj3D2Ofy|hrio&dkKgrfm>s0kj7rQo;)eHU<)e-0XRPm=*wD zTdux*^5dW8M4AA%`?3_3zFa&I_*9zTVT z&eZbEXf9(3Xa%!Z2i``_BSPVc)!6gl@Z zHqL%$A}|dC%Nl>aCC~Dqj(N9CPbW5JRM@xo05#aSkn1OYYC< z5XABGZ5&0q|Fw$S0$nTOB zo8?-BIF0?ZBmc`a2UtrzyN0c$x?w}p;s5s3ay&t!Grx?xW_l53mvK9y^^#fA=d%{8 z_l<4y`y~sR(5!_z3aE9q&c1~DOQ#7^<= zcUp7kJV!_Hi##|CwKhFm`3gH|m{>n&j^VApXc@u%c7;B-x7j0Tw3|$XETtKa{=k%h zDXHy@^N_#IKBP`h*D;jXUT;tLT~7(H z@um43sA}<=iG{Cuy9qjD6TdiZW)RpXCLe9|>%&o^#UF6XAvxWJ5%MWD&8mXGjyvJl$c_z3mwtR?E1Qb3E+QL(6Lh{)E|K@xvM8}WJ&_EIpT zOk6zbr!_}tvEqLQGU^((NA^KTGi~41cHL}gC#NSvR&L8Ncom8D4WN)PKu``uaOZ7TMU(#4SrceQB%_0oV+BX(yNqR zQxsRP@)UWh)RlIRO}UXFU&%WFMnJj0M~0K6Ue#Shb#LJ3D9V+b!0`y#v>)7>Vz_Q| zcN~lwhHo%s3zJz1&!Yg+>4~qpypo9r<3A-evi973Cdrui+ZzHZh99_W8Cx==M*p;h zLSWjfX$|emGT8^CoxY|EOOdAMBnesmV0it~;`;DV2M&jf-KT25U&{*7$l zAwx?3$g5_yD^UhlGzrN&j5Nfq7%rS=7VSFanlyQQq_%c3d*87bW$P_$(PYpRl}#BA z7?u+$fREr}f`t#hbeEWx{(*B&ZFQlFL}_n}ZOlqB#?FSUQi9v6iM*QlYI>+0>K`^? zP~uW$o_R;2E~8nxTI;15qBuM@T+{`Cv202uww)-mzTP%sC{uF#iewD@nz&{EM9K}IM_4zwQbXOhl!2c!tmd zO%JYTT`xFAoOphm!aU|egAKIR@;_4=2wcObvIngZxI?;Zn%WiHC}70vjun?VIn8Vy zjAW|CZ$P3AW7%}OaA}3a2QojNE_jWLhxvr3HWx)9TLHcVss-W*#V|(x_q!mE4Zq8n z-@o<;<;Ef|2`WUtm*V%51vf#23oqU$*s;@4d`pHbkw6O( zgo#_S^#1dMP}Z&w)Z8#wIw~0W1$}(099&wmn$BIk>lWU_xIgA-kN9w!rZp7DM~ayC z;E9K+<3;#9LBFU)*mn2#2|iYvIIhikT%UR>610c{)?}D!rTYu;KvdS^U%@ZO@tPpx zgldrjvL6a^hsO@^s^(B8kHsJjAh|w}IA*Ioiw2gaX%IXZ1d@nYI!tFZ}pu|{|*Vo@N8H(Lg5hG08N4YBa5y2e= z7*eniisxD0H+knGBFT<}Qj1lVP#~F|#B)#sq`RnQ{o-);6r>U`>XB^i3)~1qEygX) zuSYDep}DhjT#wCubZB|JkAZLq9RQrpf}giWMnHE{w1)so@-Ml@fs{rFC6$bivIG191=VwVj#O-xk`7l%tX~_H0`9?EV(!%ri1G) zxd7GXN00u;mQ2UXg#~-9aHzWEM`7h8$<$guCvG47W%&qU)snvgRBZY95?T(4`$Q}L z$~HimaCd%MpCkxVTbDMy%jdgi>m$YkYKchU^hjbnp*8pj%>|+vxwAr|*h@bkt~Zd4 zg7*m4?yjgug3b9dfZ4$xmt1fSzs%yp+VY0tI9$obAh^X4k8bVi5EehWq<^a*t&T zYxPexs1C{Bu*7OyNp(by5)($5;cQ?2 zCz3isO#W1;+28W5y;8owiC#+p-~AWoUP>#L@6zSZ{OaDd(DTw2rW|2kO6-^=Jk<@D z$Qe!`J6rniT?S2&yhLcz@vBAX0{u1N1wCQ%`VEbV-4zZXLs06IoTya7MuBL>h6s?i z)BS3(n?0yB0F)dKM;6>xhihA8EBV260wc1{?tEu=4S=|T4zyp zX-#q+WBM=_EWBVqoZjgLITvcSD0t$XzlDldjpMzt{$G-2f39p$kLj_xRyD4}`S^Jr z)$=jDZgaVZr;@M0!!DC%hhMsd|lVO-Pg zPhnfDU=I~x_7t|)>vu9+D2x?Xr_f^$#r+A-T2u=DRwbB)WlMEDjm#xZtf@4Klqnr{ z#^ylc1sK)oUH$#$d;BIR6AF-vsacAbekIhVz`{U07m<$Obm2m};7^SgM_xdZ56mVh z+!Bg}oR}Qgn3*YIDZJGKQpOFc$Kl5RU^idbS3*totwkBO-!hS)col9(dT7;HSri;@ zuoA+Dsw#77uMEMC=Xft{X3RAYL!w+oS(7$bLORxDm{<#a$qf=^5fh`S-_!?YK7bxq zb>#h3qpdr^f%SZHvN@AkkU_F3H0XBJ%XeKB}zu-!bFyg z)E|l@#0m)B{k-90x%qSJe%o>9x#xYmBJlQQq$0-ZUH8Q6UGv0CRT?B}SCANgfLLIi zz5iGD?sAw3;Mh1q_%}3+SZ3{l@hfF2C5NaE{VsTy5}WiJoA4rIcp(DF6+D<|lYW)A z#~KyrxWn|nIJ<}LN&rP!z}>NJ+qUiO*tTsao$T1`*tTuk?AW%Qd(K*K@@9A4L4C9O z0X3@O$8LMHOchc>K92RH{l;+ka`KKI>~O5)G-0myVp6fZ2Z+1S z_JAlT0A|e7*o(dXTS}6O@`$pS&Yu@kpa+VH73L6eE)B(QnK(|;Ea!WX$RhZ%P3Mxj zw>E;E+g^1$98a?g!``DbUJsiZz%Vg<$#TopUybEwN!!i;&FBybQCmEeMvlpl-T)(V z-;)%=cW8O|PtvS(PgQe#7W`ZKV4V#bI}R%(ZwXOqlYQ(q3o!{jc3>xs1?UY_BoH)) zjJS%v-c;F+0#>2kt(F615NHob}J-MnB&C|)2N{yi>4D)1-grTpwH7J6@p;KqO|4u!i9v1(*F$!1IsGiA+;_yp-QT&u zx?2j&X|_YS33LNAw74P>qYGXf1Of^Zj1(SmQmmN6Q3kvErmGdNRt8!N8*x_YG#zCh z@~J3$r>ur2U2GsZvcXGOg^VAMY3x9~;f>Oi4$r00r0_ame5fan#m;EmAhlmtYoOez zLg^*=ygYzJQ}P}+bH%<4O)V3KG$%+fjb=)-Mkk)L%Xi! zzkMD~ZGH6723vf=iER(AU2Fdv!OYiNX{g(V1Z&|xXCoR%XD&dxLYFH00$Xq1|502I z0f7c{Gk6B>4Z4&)6)gGE&cDd53KjlP#0^{fKs>=o(JDn zp;UYDUv64lb7#o{0nSZV+|!B-0KfXT56KlJ(`0Sd6kI;?T@ zS=*M3?>tZ`XQRK-m*L$%2%^|ti6k{OWL8_6Lf_?E)9mmp7jd=Jm``^@CImeH1R4S@TktqGy$>hTC|90V-0AqK^@eTd?&&XWb>}1m#>A{ygLoklJEHnoHC7IE4xRBhC-T`vX7o`)th!UQlpHyRXwO0$B6U=o;j4S3B?^z<;9RNyYm**dP$PmKlC6CWBB$R> zea;)c{gVhSuLEHmsG#DRYLiv_kOc}MaT7T(%JWkw6yxLWuMC_;Oc$)5pnh~7B~6Kh z`>Ap1L>eXheVH)>4yF^TH~dauBFPHEHlr0N1=-hqZ)S;xT@ynDF--Gl&1nfXDK*Oz zPACKZTLCu``Cr;m)r7xFr*A?Is8+MlG$!X|rLwq5-G(-8n2+BFC0Uue@hLLBeed&KRZlk+3$G4#I#h+<7QKVMTP`^H`>>L|)oQ`qd%p_d z2f9@iDr)oZo~*+BepDPJD4u0NgFWbJ5k;b5w$)igqsP;^KDyy6!G<@zxW0EM+_i2sxix_%Cr(}O6xq$$Qf71BE;O8ZO-)O&9AG- z-1OBM(PM5I208RnYpN~=&0K_^Z8xKa7x}$cOlA1ctjQh6qt8- z`SO2K(f4B*-i9*lKTe0J*)`lXA<-s>zBzgAU`a~t(7dTK&U-}YNYheCNj_=UF#~-~ zWUr!p=bq5;={b)@8s`7jyXl?2 zbd!~u#+8+s_amp=)Ti~TEMSx835dD2a%m4@g6(-}$ul|a4fw(Z8{NszewC+&}E)}xde1QLmt-6$FJvdLNYd-AlGLGzaTJBkOW&Je3GB~ zEu)YLWrHqp|ChG_Xg7q_SUf;>AMw3XMoo)h;=CB@(~9vnZ1*e{Nu}w>;K48CBjlKs zcO20WtV5gnG~r^4IL7I=u!njqRHf5n_SZ|je7Y>JTu$!tBfCd2Uv!7AUZiH$vfsAd zJ-Yx47LUsVR5(8djFU0cWn-7$wt{1tDW}=_=%0(XSV1jh#n(MI0MX>gc%D3gZ8JXU4EIBB$#Y~qbu#aMl~iY8VZlQElbsCITgb!EyM*9HvA%v0 zp7qdJI;)Obw*li=WQ>DT&(7I#Gc?niA#C2445eQtq=k7|gjf|apTRvucR3+}`O1UceN3QFE1yiBbfBm^5j&%IMrSA$AEJ<` z6Po__7q~_swcK@r;vOz*agFUk(*ykl5e8T&lXI&7$@Mm;<_$YK7Pt4MVdkzgGEU zO|egdKK{|i^Brm_*hS2rla!^f=-vw=qp;TV(oqqNxs4R>O zJRv@&{NT!nbYCm0iCGGmI}T%T7Y&|G_A^doxI1>)Hu9R4NpAnKY8Vk$m4ih7#tU)) z)ej14f6i?%m|N1pcRAuLj<5QisOEJME?$Ju#Zee}kn)O6nT$xSqV;U{iMKyGTP(hh z7Tt9|b?$^^Kz%u8Snb#+p2_RavjpULo*k`C_#kGxd=Dv93EksT7$QEOp=QX7AFnlh zCk5h|FK6@KcKbDQC%2ki=Us-j~M( zsbJW)M=1$0>#6^18K+u^5jgldSL27H!exO^?u2`ijvnExEbU4b`B9E3PYlpDp%H2T zCFA`sA3yx#v3?1rxnD*g`UZ%Sn_aGti@SWrsEWb$j@knC&}|4BElNg5r%UMkHI=HF zy`BTn?iM8U6>BQXO}-?krBx0Gk*?)ffAv=?;dxquB+f$IWZ3Piw>N0RIq+Ye%+vVML06+-`2Yl5U~O zl$^#OU#Gv7soj1?b*_~v{C^I-3vZC70nMwQ-?=CTNPFT3SlN9gna*>g-taQLnXyO- zp*kAK-5)$5C}AS^x)dxx9DiI%AOThBZ?ke_;NK7E*ROu41ctdM<}pSNRo@?K7ydxHNIGF1lvkR!$;j*?SLVVFB5C1$Z)kOQgf7WpW`MM-X9>0Fd(>{-G zPw*&?jYPfiL!?{biLnV>pFg0vwhEA;!UX&}c1rxTI-!D>j8l8oep0t?i@^O%UeF$! zsu52NnyY7b^t4CposWE9koPt`3Hre}zoN&P6c{!{qpgojg8`7ADq$J-fo zszphY_~48q4_&!x!*oVv%oO0{_YbG}TV@TdP=E9|0pU49eZWX&CH~{ zj;#8}l|NLJYkdMz_4$D0DnCY0E=S#>+jr?F6`n|8QJV=GJw}OQ^|=4^E~UDu{!_(K13@X43@7Ao|H~iC(^-1b!YVH7Nl6~Y z%UhP3k*bR1I3`gCo9U+a^_uf=t1g<^qh;TcaWa{=)1{WU!418ONBEL=LErV7vPll( zHbZLzKvjC~u+!SQE+L4I!UdQO@4{QG6br^A{E`3pT987uk{m;#p#-{iqNzvp!xWV)ojC z+D^ieey&P~(U`BxL2F~WiH%rLpJvfUC`D7Ry;&QvMYk|e zZx?G;2C?it+v7MGw3(QT&zrUQA3<2*AB>&oFBFT^K|C$}s9bxD9D}lEL#V7~u#Or9 zFvweh*pailm+WsHNUi}$;@bwq@db=d`e;SQlcd!WL|FOYdXV* zMmaY+;Bm5ZgCTfbEBDh?7`m)aAVm&^BKdul2ZjqMK=I`uP!o+$(h$q)kT<-5OJW&E zp~ep!{-S$`KzXFScN7W?fIaZ9-5d9|SV-ZM``put4pyJ6fHOaC@Ls8D@RAM$!hLUl zs>9&M*{xJQNt-J~;CMeS(Mb^ljZL{QNG4)HfH&>$pV;ZwlpB-*f0giwcWuf!N_Ej~ z5<4%)fa;_)InREJ&uJ|9JHHr^D-Hh|g4FNiOp8|qTN&Ij*%YbXk8t4#8jM`kP~b(B z7Vlzp3noK0(;D$L>&isc=niLJp!Cndabiee@a`qBs;GD_oHcJag{buBwHow zr!48K5olUTLnntOCP?Wm+_PfoKKN#*^q%}HVy65x6(?$mglobwsnH~|+VWwAJBW0? zg+H3P3vND8MPkgt5}5Bs4fHwN-L^vQoF)BJ?5eimk}1wBrl{N7V2pi~UHFYi^F}pp zauT9ZGx&2p2M?_=ek^3j-vjYr;-O>%OOekKr8?ySWHN1T5&7L=>M_!ZDI%Ss>DP-W zSPT$%SiYoW%ZAD}SF;ER5p?DlICD9TG}rhPiUQM6vr< z5DX^jE>5KAc?o-Ak{*VP)0J0Rt1Ov}j{V>Yfm0KE1g}pN2M7g%FzO#;KU`Bq_6AJH zzw6V&H^m`MX7Fvz6fPO6jlzOSyyBGWmqEIAUEoxcuSyxQI_7()W%58mw7nNGgnylthSS8f<{i4BYs8#7C@!x2!;#ERf za;4P71#XxmhRXIH?;Q~N7zd}`(u6U%v_6%5w6afI*Oc*FjAM=01vC1=oOx=KbbHu# zxZn@YT?<|>Jq}+%C`0P1N%_OaxNTIRH@1ZUggFgY#mTI;-=Nr}``Gh|k>QAYWi>5P zKmJh3zwNRHW}pGoSfMqtjwrm#YP3|b=;mB(DJbSX<7m=4h006MTY1Ntv*u z%UUdX8P{=BnqUHint59^HZ#x9`3g1EC?a|pLP84C<(2+o1gbB2yu2Jj#mL<v!TbXh79ZVr37pp>55CVDe7u9BGdc8o5>2jJ$UM@bvEGjY!w>VKwWJL-j6x$Xxmr zp$vK2KnMLR&7!qUwC9DJAG|d3IslH4*ofLeNf{5s*DVPa#cD;{LS0$PSvSxz^EY9Po#U$u+XctDp0nZfv&ZKK0$%&f~<0gsHX&9IXVtfrSR}a^<9tm>Ge}1ij11{YqcPgAR~JCl~R^m`=%Gr3K-(fCKh1e!eZBm9M2-%i5#-ptL!lEG%=!&krY?Ybi*vbqmeC-V^cN2nHHISmq{Q(dNgITL#bl!|0S}N(+~{oFX_>mm!;haQy@f7+yP4 z=U>W+EZXa+jJK#6v|PKznW}m){)=%-`@Zq}35AA&!?)MU{LR4S2$Rc4-k~z&CPv#& zAqDrcUM3uoL+f%#d%^iOo zy;R>lp);rt@gkP54^CtflDm=EhF&3R$}M52nd`;iQNz$)EAMWyx7wJ;+po^g3-_rv z`st?xe$CXAht-5a0(@iTsK+G>_>&f22gVQ{yAP@+HM9cj#}Ypm;B5@?qLfj!&!JNI zBX>)WtLYtUW0$HNrJQ-@_9&GM`{V<1$Tc0La_NRbfE7EqiHLe=Lw|)xH?ji{%WBQ^ zqs=&}n(X&jA}4wB86%m!s{r@!K24NSZ?~Woe+>!KHND5aKO`-5;k}Us5gjt-o4I%g z8A~R*VeDCXqk%^eMK64vS9J4b}=wfC%#eoga%Cd!-z2 z-P;TAlDx;ZKSf0j`GUw0Khg)!ovY$pccA74Lra#K zaYfVfRN^J_9$Eusi}np~u_h)2&wjGABcx5s{s@~ofyA}z*ia;o^U)3tLKnaS!YqX~ z@*oy*1y@7QTtXBbZw$#6PH3?VPN_gt0%D?Eze0tb$a&x}*qD}B_aPtW2~|63=T*ev(aJE zKtG1k8rIB)4SvGgGU*b8U|WuF;K2I4Rz>F@?mY762_tAAuuQ+kz64}0Hr(8drZu%e z^u)w*Dq$Qi>0#IG90K#sJca*EdE9|&`o*IQi4v*5O3%Z^GHM34JLUX>@MBu~34*s! zH!f=MjsxRJmdGFX7n`siXa5V(rO=G7^I6&_w`n%Y60Ip@U&+ksGAt++B1jPq&c zIix?GdMjZNhq^sbtea*JUGjv}RCPi$_7>XAtBtzYVAgGq{Qcb=4r4!@Gv+6mq{WSq zpLDVN4z~?K{ub&eP?LB5fmXo{1=BOHaRjOC^V{V$%#lAcmK8HT@eTh=JfeynN2@|~ z227wW$U4VdzYCVMNz310Tn0}@qnv%ZB4)q1C(+5fnQnW<#TlRHH}YaRR4?*kZU#}H z{NqM0{x{yp76n9zUrf*gT>EdbnI#;ba|UG}KLU??!}G~|s;&GfTMg_)r_f#c^L0t1 zc>eg-fkEQ;a5!b06}8&%#`D8E@1#n`(MJUGSY)Y9tBCURLmUHqoqtyR-0`ZFiw`uE zM*3oWmQHYtXRFZgJtBf0nh{Gg1Gk3)g9(7wOU*&Xp&9y;gO+lvpi7In+tE;-3vavv zT-FGh+>ksqLB8*IEO-X&0 z&aH#kjDZm5h`{jpLyvz11Y0B8Snb78Wfg+(0(c8U??$7W^$LJQPRVE^;Qon*-=`N% z34z-85SXe!Zj{Grno1J53{UmdJ&?_UD#z~G0{Il1(Vts4JhwN-D@R%wlj|}w_q_eQv-I{ot z4qkBHfOb=xo~oHYP-S&5kHa>b0Rj4eT_R(8m9rl)acukNQfI?u^9sl#;HaKs7d>3U zKsqm1LarqmctiFo>#T_hZB|zJzeUsy71}C>g(xOBW;69llU=JKST%KC9uaH5Lg$EtYz>|Ih(B9;0i3B8jbibNDNg_w#zP)U?01-NM(`&V_CjSO zG@g$}Zra5#89Y!bK~fsG<~zvytHW-+wpF?{0?7^s&B!?+MLGd>MDm4-iwzX03a@53XU3)~u za1*VqwT&L=Yn@(t>Isw8T^VH<#Qh^iI+yxtXm_N~>4R57WYowEH3a1P)A^Vv>`a^( zeri@*Mf*Szr>(i-z5IM#{QBVQ;;6l{Gt%WT$*)D<9}YgG*hC;1+{8nt$>z_)jeUu2G^K)i(Ab$~bR@n;;-3yY_>R z4|T<2i!y$YfOax;`i@cfjn>Oi3FDxbtpClIhfHIoNWsP8{;oiMF& zQIP;s#p!0R!P9x-fbewEawq@FM{0Sa5U(bItT~-&X59!wR-FSc-yJQK8G=G;hgEhN8b1Bw2QgtBpXpes<)uYcc0-ffQe*B0yCX$Uh zH1|8yP&wdeFCkpNj8D@uZe6LyO_XDFZPJ?wY?y3?8z+B9)nP&26Sl1S$Xfi9*7q)! z`1!PH6l$yed-U!KR+d0T1$l4-_o!oOWhAs{G(jkE76+aSGJ!~pHgvSW*~B{}zIByF zvgfiZZfSTiVB>=D9_RZ+n^ZSI_Jz?n;aQ`8%nIM{Yd(nqLxiAOTP$~Hw3ue;27PZ` zLPF<9HLs(4X}<&x@nHM& z>8CwTF^!6OXwcfWA3yWPlQu3K)bD1UJ6gVn<;ALWo1B}U==UXpm&($pGz{i zpd$ZxVls2_Ny-m^NhC}ZKD@HRcI?jFmJ&cAQ8mZK%V-O8cY45rUW)}gptu=fiuqVZ_%{fX7 zVjl3uqWVR8KW{?g*~~_hXTGrQ>h5PX_r`GijZ$WekhuRe1fAvqcCU%g(hcrHZ#w88 zB1O^^)1HA63Qa0Z$PW3byE0`GiuNDgebE@v>g}qNFh>7uxv{*WSe>b1M+G{}78{L*84X_6NSjtIYCqiu zT^k@GO3aF(!b`||M#GiOesC)NUi;eS=Jn;89QW^x`=3y&5BNJ9N zIkmyQ)A?lT2LI|}9mh2dj3lSX`)x@MQ*VJX=*CITtO|z*TfivXV>PBb7+M z2-Lw9e<~Od_7QQ~LQ`N@Dn+3t-(yK8Ko8SiH+Z4m;vzWpMhTUw1%N6{jrx1yJI8XOIF3ga#d5%IPxsG`7V}3jo$f*@^CF%HMvruuQKK@xj z^&eFu*?d3$kZK1pE?&*>f*TCsLvRj^$W<0iCtWCOSZrmOa^4W~cG<~jf<53TQ5Sl~p7<1a65P`XxTaQF{a~45OuE3bquZl!Ri?!+ByiTw^8)x- zGG}!u9*kamm?e>#&Cq9KDzk_9G@VM*Q_+C_Kbn+q9k--QJCYkIO`5;oXAgX$9v$~H z8B8KMYJKe((ePLd;arp*4%hZXVih;SO1iuBRA8CidDyZ)cKl1r^J}t#80A; zg;*C%V~;bqUNrQ2LG0Oe!%X4KfZD~$&)`RM1zKSh&r1k>I6nC z%O#`4#bVE8qt-i=Ab;*_HncDgS65Eq9^VT0z!1z8)B0+&rbkU0e01#cgIC1#i)S2yFY>ypaKvsiSQmUvH=lSMNIr%cPT~wh z)h*MVF%40GJHGG8c3V@cVV4jN2@)q^a*O*+!$G1_(=VM(1rHJv))I3BgM`D20ny84 z;L}2o1<}jaU+$-6sxgIUP|oy!Z};7U*`8#HNe&E|K&xyLyI%cX2)wXK;R z7!XTT@?Cl8iX8t~w}^)kf|x6{{@YhZ6!x60a5BqgikHx1Q7IX{Xo_R*mnzpY%{h?} z8Cpm{%7_kL5_a@SJY^Wfs3Z$to{K=;;)WL(c=HV4^Z zCGvu<9K;RhiaOQ98hGTU38Dsyj5Ikb;1F8L=v2Md*R>K!Rko_6Lp?us4Yj zyJV%4YHibsGv0cLtFWxE4A1$)_W9`@`U~t}B*h(-Ya)ThMNZH=!V8%(mw?-&LRlv` z+_2Wa3nsvN@TYj@Hf1O0A~DL$Y=XBhPyVw49L2wmnF#KWk9hdR3mVO*)zY8|S!(20 zo#iPNdhtrr-|Hw1qX_guCRQd>s6}Xevqtyo*P83C8;IrfpNv$+nn`5a>1mq${D4AM*k#SWOXQZlqk4jy;+qsl%^*9&99Q|$mOgS9T}AOjDv{1l%Oe79XZ{OrLAy>ir`-!aH#1&mUZlw0Kzd%1sgR10j2zh2>k!&s}rRt z0O&vTiusN_`7`}Whk{~P&i!7DVM!?h+!nmGvh#Dw z@9Q#*v? zc?9!LHwekhN1V&HVWy0!0UaMFDW9TN&Y^(DG}t5I)iYlvPs;NFFXE3Ybtpe|+{my4 zO6S6qZHmO1!B{&#huXsV$d^Q-hTsNXgSqoVMfXwz`>xj9n(I<=FD4F^DiBtNdmKN? zsbB?XgIRiN!P@8vFC-dH8p`#!-s1!r4a9%7I9~?hT5a)#+oDJ|JZ36Zee}-uv3vsYp|!UWhgbHBwUul^M?MeadU~Cw zJuh&mGf^9a9-GAZA{NyPw=Ru{G*g#mjn3g`(C^qy>ExtjO*6rkS)Ppiz}l8sg2Ck(5o_`9A? zm$1ZZNd$-yVBaO08k%$`l)f+;^MNW^y2Rc4yD2y@*@kP)LW&Nu-A40wRwH(o9^|eZ z(sl6w{5Y8^p~04$hI)5wJd!kT|HbeoYcl~Q4SHBw3Y7&DKrUY-?;4B2w2)G+H85}k z=te=GPp#HT6Q-cESMAq?jhd{8#T!@q3;%pRU#3y;k-`m!HvdvSGV-j>tUwFmglT+w zp7@DEqn1qci&Ph6G*M}n0m^;N%|3ecysFn1&FzoVh3j3{xcFgf`JT*>HK%;@K}@V^ zw9Nt~GuTFACxhp=el{jhWMK>C@}kL0Y7ZH18a9ZXK}B~xF1P#ZQKL)EA)Yc=PprSv|zMl(~Zk zd6w#JqR4!leHy7_X+|V8tMn6<(7N+HKUMEqRf*dUPG$KYorYSn+8f-4AF@2?YV}%Z#W3cr7A@(%e6tUR9z~bT*lW z(ny$sH5*UEsLQNtRXTMHb6i%a+n8qOB1$gJ!UP?})GoU`wu(v7#2E7M<82AarH z0XWL1sPRqcr59~9p?OGgpNkFZ23CGtL<9!X<}Fe}n{20?WeP%|C3rm1jnMB%jYcDO zGvM0{SqyEenR}w4u@~+nj3{z!8kQXF^RRx;((Yr|CH<}~f27EEi|18tFy6XMLhw1u zr&k%FT+=&ve8Z6T8XHhi1d^O{0gF{)lLR)Bmd944&U#ZkdL3RHAe#1&Nz)zcIR%{gTJ z36QF9R`_7Id~sP=H8WUtwo}%rp)UA!`*eeqOE93s*Hb(Hhma+`)Od{h0={6V+YxHU zp5GkR3{MO}i(3u!g8|Z_7Rrf4&Uc@7tnq-y-d|sMybR$yU>QR=AZW3Yqe?a zJsFHbX};51!_*UEMIJ{;lf*C#F-}#%?Sv}y(V&P4kO^WK({d>?bZbBfb`Mf5(L2kz z1kTAmAR2R-6O&d<_*k&&Pl*tnK=PD*%B~0>{vrk@R77v90+bdf*$vDj6&wBx*VpXM zokyc#lVG|mi|XZ2Anl~gR8@EVEWib$O13Rq@s8MR#tSCLa2WkX-g1}yxM0aZev|ks ztZv+&KZBgw?h;dA%E0wD{5H<=C@-az7ENGiUC$3L`IuPARD%%v@zv-w#%Z^d1Oz2% z5^LRQbeB$|`z$jD8M)G#PBc_kK*O1h!`@Hmwf=Eqj6r8B`O?(nPt>@_upaKEKLq99>{%osSb48HoDU^N9G!!W# z88Jb%LkL!7eapC67A9B*`1>6YV>KW#e|@Y}j#9H1ti8Kyj^??E60E(u6+pa=OJELv z1%@Cb=H*5>L!~72UQvD_FWt<7&u2#NPWfev&cQuj1b((ans1u~{i&t-iyAD4uEwQl zK}Pmy3_LY}5IIEa?lzK*q(D_x|NN@%J+E_fj#)Di>wQR2bQyLb&k&fM_2RPv7Ryl$ z(cip=2RRX*0+Sii#-p==dlHeudGiQYa@B?MeHmX}?22t~kT9HwYqKIP)R>z+Sfk@C z!5vjz^2;Q!1v+O3hTj)esGEyk2wfXqof!e+Ub+7$_J<94=3ff{N=#_SR#k@Kii zyzw`YJTPW*B5CpQ&&{=CfTce><5a4GTwf6!X=EmtZ19O-1%It5JH(E1Xv|=y9!Mwt zQu>mj4c#|ISek_nOrUg!pz+^X@QrA_MDUnnRS}t}^pMO{MBCGz>(33j_T9>6ILkEB zd*gHObb8%f2r^Z`Eo}7Zm9YXn&?Zsb0Q6hj(C4|R`{6VcFGI>AvSoiFYM)-LmIiDC z=mg1sfg}tpeJ`s0ZIEug8&Moh3&mvh$&F<$cV;tHV?65n44s!i_$T+D=?)hnklKcH zN+Kh1w?dvyBTW10A5Ys)X)DWv5lR0c5Lv4PvXYgJ5?*tttdi2xRI~7Qx zJJui1Xv_+|cbMN9Mc!H0 z%|*&Q5r*Q0q*HLr6y;d*9yGWtzqIZkEU1d`3; z&@YQ8urTy{3Riv_jHYr!8)73tZwJUvHYgE#JEN4vf5p*_V9tB1ab{zl3l2SIQVmLU zmBB?RFMH`nGOpP?HBX9bJJK6fiDAaW@(Q_T`5v$-tCn&PL&-Ia^aju%gYu~o0B)>A z+$J}w{C^(DaWt6XK6P(`lSsiD7b@hismMfucXs6R^PAM7)+x0-2#6t{{y4B)KT9t0 zdcRd#WDv^X9vh(~WU>(fe?ncnqBweX2KQE}kCvE61vyq7e2Js_va=ImGZGSd80Jwh z3q1Y~%wf_fk)rfH08PInC4RnJ5wyPuIS)K5piV1*-rp8zzk3N4hPVur!RSOASC~i` zZb#sFo>EFdS8k1h6;ZtG+XFHx;4HDOf$61Gm~@&{j{S1v0v`JCy=G~+gJi~y^)Z@t zk8|rZroF$%A0!dlr}FGaTc-)xt4`TQGnBJ(d^pSa5WeElmtd0nK*;Napy?DrVC$BD z>A?R#en@ou2!9-4d2vA-6Yq?z&If0QT}9b&bD`z@B9sgdPYOVlCvlinJBY3O&_g{Y zV0Y2uMJc<^a+zJiH+1?DNx#gsaPuM(YFN?97j!2E8d24aPd@Gn4j0pvnyckMI#(c{f z#e2(Ir1HODzn`r(1`sapaMd#Lt)=q+ux#}I!N{t{t)^kD(C(}}H+J@S^zVkNCcUH( z@n4iTs5N0*XXzohfAbR?iT2jAdD-rRh5W;HkHpl*VrRl~58v(XZ6i@vK8X?RFZk=& zFRQuWqC5eoBn%Egfrbl>^xP0o79?u)(tshAgFuncJe<*`1bMt%1f#EPye|_$(%rX% z^kb%zW>p%*qWb5RwukF#=+@BH`(duYL@W_#=`vSf#FeOK7t8e-OX!rCB{ib)Vm{pK z<2bz2DVn@a>DDini1q0!SL2E{A3Eu|E4Dp!apIqq?{L@()V|dQyI&!Gw!0BIpv@u< zv=k>GS|Bh%>iGNf6^)`y-K+Pwk-v<+L1^+zC=-RXn^z!D&VOSS8pRbp#TJ{QeIqkB zUSj;nY^hWQ~ z(SIvjd-HgsDrM-+wbX#h6G|ge+Un3D;ME@_OgX~+s|FhZx;^EvyUQ{W((Pr?20I`y z@dV1M0r^KP9TuZ|7Kug~+4}zT155tI{lxjsnW{jsT!9RT(Yhh;X+s>w;&hDr@T>j= zQQBDo0z^enV9jzu|8X!GDe|yy$3HkL5)er&$W5)UzZz#Ht}iaS8H)X;ct^U2>0 z+W>Y+JK^r@T=VZUSn`{?!r1^ZfZeN^9`n5mdu7xjR=rHN{p&2M3jF;^KOys*rYv~i z2?C?3l*vyV+H@eyoM!pjHIy&f7+EZ|fz`4pgt_olXbkUp*=R z+;|n@S;2LB8+^u(#t}h>Ihw*Id{e`jhv(G_VH>#u^je&z8j+of?njnKLPI#>gTW3M z-|J25;;lZnTRq)%D zCnUAxS<@1+*W}>$gjmcIW$I>WPrFtLRdH*r(xbYD;%y^=uJHVEUu*JZn9BnmeM@BX zXb$)zO%naTB{+ZnIc1fzKtcYBlRVtbOh+k^KToFdh73)*TYiOuhcs%%UT^?x4pOW2 zUGC{gU6n*2Do+1ehWlUg=>T8{HLm*2yb?glyi%h$Ct4eFSe=$Rq{bXfOI}T8_#*vk zmKlw4P-EJK8>KoQdTe5(D_TNlx;mQMQrAExY(Q|*0QxF7jbS%g;R`~ zpQKlK4MX2%eolXR<8$X;&_l^XAq%%h$KMaRPoZGdlmu z(TgZiJp7a{#h|U;s}TPD<;n*M?aHb&$05!@Ma&5@8+S zc?<)B6s4m`H2LlRG*=U6fSiAL+Z>7Eh}8q$%WL-O8GpNJebjK2%3zj}839Y+GBc-1 zDKAxx37O7U1pWS5h?IK7j^PNwBveiqD|liLnl3pqu*Z~=+V8c`inO-I&y$Q0rWoib z^N_>1WQ7_X8V)RHZWQ2W7J!!hVxuR1 z!Gst|J!RCK`=>Y?k)G$(riQvNe7JXc5!3wiCJMAuJGO|iS$b~}0yCyXOy04j2xs^?U-G95Ym;pEM8 zL&8DClc;>=#>ZTZc6NU3G9&|-z$WD=KPkqa{~nlBOTQc$SO`#TWst1lHVUspaFLf% zS7-rS)fiKE(~fsP`2}ekJT^lHBLTfXp(0{3F2%eTPpXIbzvK?Zid$T9YTJZ}hOciS z#0k#TSTOAq@pHbg?ug&fqo+2U6le?^Jp`K8V9~-lM$(FfXF*|nk?{zl`~;y86Bc@| z=S%W;@{#hy&n}YGr10c{B=A{d%Z!Y!O3%2SkVUyreA<}jq zq&(r-wD}tb`)a81#CF)ARJfOJaLZhT{0-*h_7?X{?}<}OJZPw5t2v9GBYmFd`koZ;X4;G&*vO^-^X z23B#MbMSi_s(3R=T(u`3=f7+I_N=a+wRZ z6vB}N#erPmDC?7_97%~3c5?(^iCkkIbE$`_?IQA?sAc^;Wl z4~^hzU5cmAvE?Po3r|JTeKzG1;4eh?AZSHR(wu@YMJoPx+UBGf`QB`qq#kQ^`XS_% zRKI#ZNIEcy2Fyml8p!mDm&1D&MCGhYZ{EQV-F(K0+Bed;M-LPCka<}*6wr%t7H9_zDAm6uunxWk zNl7s-UnL3ZTst9pg3H1VVbyN_phDn))KQvSiwT@`VC8IT z#bOeGm%sMzq_^q3PCKd~vhv<%Fd-sTLj2WmWIxRwi8f`(-&l-tVf7>A#IDyN_rG?Z z()~sWMQ6*~_4rDc==ZVeH($~5y;`#0%XKZ>`zx37#6=_jj^{snP@lbLBe287qk>3` z9y36*vfH*ThF=)QzuP7Oh4#~CZ}@OE!T78oVv=c|mqT|uvfukpxLKjZV#*mmC&Jgd zZ?Ag`Lg;HgMuu%CB@fFUd*h>v*y0XejeoBV%V86?!xmzT9n>xx#0iTq{pVx+$o~q{ zTg1DO7EUbGhV<2Lp8WWCSo-<{ldxRpP}E+=u*2;Dnk=C;g!SOQn>#n8k1L$G#h)Z_ zY0<#ow`9+OwOrhtoihoSRr}RoFzDpMmPt&(jomSU0~Cb7Q(IZ-xc zoMabMAH#EJ6B2qmUJh>`^gIcG)P&z}Sj~OLBrbeCyCfCw@F^DMXoDfsVMctRrQ?SX zs_nOKIlDn7*K{kA5B;iS$0idb^&N?8ZKb>3yHW70zeuzPyE+`?ovbwEUP=yemS~SM zy;h76NXJsj82vpQ7$*sMF|1m#nJT;Jzsd{Ce_dw%t161|Hh_nHr56ExSq$m^)05QC zI25kJNRwutuc7!O(!rn9_K;P1L_@Tj-MlwvL4}E*vM0L6sx-zGsr}J}E|WzUdFNZ1 z<^m&su%#fW85$>2ALJtz!&;h0dEMoV9Jb&|gjoxy`Nbm)M??(?Gjwqq4d{obE1BCX zf+_dkj`*X>ctz96vpF*=gimH65~Q-I3g-J1j~bLF71%4$XblO$0sc-|JvZcU_Vta# zA?Wb@fVeaGmiNYzighkB$r9y)KL zC1ved!S3ZBghc8N0XmF{kbwmC}Me`Pd>m*fQjt)X_eD;rnt=1U~^p`_Ad#Dh4 zL8n3@^u=m7PIjgF^;ThJJ{M8r4;vvG`L84|6x%zhMLM(%Bu~Tr7xBf`#^eQmZZ!sl zo*Cpn^961tC~e0)td&=yP(&g6XWF%Vd6d6qjo2Yp5wRgeb~q#xcv4qs(EcHpa4cU4 z0^Yb%x?Fb>cE&W^(@<~7z2_e1`;-H}0C_~Ei^}(7oNj+@X$@o2$Io3~;nPZ)vbDhu z<)hKpyi?jf{r|@oSt~EyFEJ%pu~m&^9Be8!|5DNajkZZb6&efQ1ffDLrv!km;R_>k zV}j|qh)N1WqIhADJ?>;j!iDJBcrm?R^ZgcfO3DFl&HC=1@SA^6oZxCrOBVO8zWc6f zd(&-#dlkLT*_<#A0By+WLYuA*6Ybw-X%SH3tZU2Ksv|Knl>KKGw&14uc7Y*uU`-fE z*_dd>Pb^wzD|x#vO5vck!`BP7xj6Wpja-8*A(#YIP5pmWaDmYjcz?k{W8Vy@J3a~7 zBL%%of9?8{$!-7DP#a=SL? zc|e1tprvuuI&t6BCAfsittB-6t_0E&Msvqil)S2O$QQ6P zx|DLKad5NqcKPC$cV0~$M1X#Zo68Yfk+_oLSR*PKy~~8|^`zZjeGE+m&qD7H84Ot* zgx|f*=!ZjiZPxyLJ}X({0>P>AMl0fY%v<_?&Op+zikjLgV+sm5$bS*zp^GxM6z?H) z+VL;FrjTt3fq?N~Pb)mt?%s<-ax-4#?#SL;e6(&#dbEgPJ6uCDv^;BwC zp}-x&!tMvkLtzbCDI9!w5Y@BbKt)1?=J!!3VP03qj`ypdf)U%)K{;yYw5w#1H5wB3 zT>{SVtsOW8c*J8doH!?>-G#nQ_*Zr*;qfE?u8r`IKWjFq`htZYnE9NuRs7VLh?DuimfsO&i9N( zyZ5G9d5G$19{1N)Hf_`#`*&@)!j4@x`g|$ZZ5$K^dDS#4#IgCM5sTN2I~OlxJhZ)k z+r6uBT_HCOIUyqqBqP(A;%jdoEqzFKwj)->X5MmN;IG=V)ZAXzFTY`Hv7_Mdz4Nz2CEv3qs8BA#`4*p4~$i8^=2TcA}xQlKLBY=|}bax9avnD9RUB}d5yoDu9^6IXSMNpN>4gHfclLY$QiKDa3t=e76 z!@U!bC@2yoT_ig8HfmHPCEa!Hc|2(K4#qH!G!Gw0<3t$QCeB@DEmIjr9zstyWG{KJ zpEQ#Lv}pT*IS=AqZ*SD!n5z~`!19el`pp!}CkEu{n^uJPEh>7e)}rhM7`}SMyZ6B` zw(A~PloabK=umTWa3Y6egT&u`qt__<*W+=j?+>FiHApG@FkI%1YYO(NE~^!`(4vqU7U@K> z31|S>enNXh01+rIDryA8a8LF760lTM4+akZJv6+4Cib;bAoTu`2<9LBq4Ugh(}+LS z)M>a>P74{ecFnocaO~@wl#AiM$zFCCvbxKu53zy^@vUj$@=HA{j)pV&;_vmpS}PK? zhqUZXK!Y1Vfpeo?iRIRYqD)A2?Yt%}2~sb#DG8+CWG6?WvkUe9B<|Q^(mGcDDOw3P z0==g*h#f6`LicO2HhQ4!Sl-%e54$KK2k5BZG6`(G&lT=>aSRFhGth(M45?rK2=2)2 z+1OE1RlYLsknb+%;u>>bMi;Wb^5c7EGuZW3tukErmmk*tE*6BZMJdE18PHPExu*l) zUD;ImlaJ5;(gpebcEDP+H4Itv0H2IZ3Pkv2{=PM0ul^ z`K3bbw*KKQr0*rw9u6}ylJt8znm2YH5IMh1AeQ_0j>m9TW^m_H(4c1uqOsnjXF2Y+ zt$w|!m)0axQNy*|V*6|N;!W`=O)wpAU;rn}z{6A3Qd7l%Edn-$yA~?6#~lr1OXp~1 z-vuROsbA{@oqZoEEQPQ%nEq?we3$k)z}-v#^D1zkWOwEFmNH}JcWcHwM;S&){p0V= z1$0OZSO`ducq=x>i$vyM%G*GJ2_)JR^xoeU;nmVZvs5O#QBs;i>OHiBIoZ0R88cDC zv!nz0@*l!KdbhjWaM93s4YQF%lNZ8E!?`pyNca5vK}5g~{(svu#Y&4Th~?}Q{1`+} zWaeRw;4xT__(Z=CATC_4^|S|Ud(dDH0m#*RwliS=_Jy!PH9dDuF?Y+86c7 zHRjue3~r11{)IOUHdK_eh0ah{VK_D8CsD9byQ#)-LcvF|O{l&N=lS>NmHHJ9<+IZo z+^{t)nV@}kc5PoFeR+Pw^@C`h!fi_F&7!r`~{bFTg6jws>)kHL9bog^0_}(dA}Tpj)W7) z+)Ol^fvnmeE6c$k&j~EpUP#2^U82E)Kv9HpmH8oPxW0hTK4H&11}Hh183~d-NpRm9 zKm>CbZ*P2j{X~nsC&*NI~HhkY^O^PUT$pV=lLiOCGBx8bvhxnhfe;4EJ~Qe>x6uAV(B6je;j zN|+L=DQf9BLV$O}zk7!I9-$xJ*0UV%`?D>eNn1FdJm>9DfcMZI&NE={1n4ZvfL;UbCSh6K)*|K@AphADm{WgH zocvtfp>K@%q;RRP@%2L^j|D2#GfI3rLjGzh?9>P+%h2jeSsX-X90>Yi`KiJR(8~@R zw%x^)W-C&P$Q7MHLU@VK1~K7pC(>vgjOH1$J|C7eb6@dx14j9K1mB`9bx&B*e0mxO zOw%!xj{aJ1($(oViRm*=8-kmoS~CE!MiH#$N`je91|cUo=qR9PHy5VRhSMLaT1)rCNDzJPBvoj2cx(=I6{2OBA0nYv1bgTgR$**UPt&4ZKE(dT?2f&MEJLf z@~;!5?)#HSFYa@9M#Oc^#kDlwTiFrmafhp-P$`i9IzX0bT+~YJpgq9PxEb+N@(c1Z zV^ku#s*W9(POp%hRE1KYw`$L0)f7t8>?%)h|0}TW41i%%>M96Zf@UJ1XC|4Y_UKtf zEX62Y#yH(>3CyDllb3k?<*|~1QR6S0wpJe#pd9xv2j;!veWyB>kF6ga_lWF_iE1>e zXEl2PI0qC6kv_Ek(HdqM2b+?bfc9psi+Dl}h)JC)8mQ%zP_x8aDLiuR1aHxW zeUBWCFEVq59Uha*?DmaeY`&ClOc5<_Ul?@G^$;Vbsz}sXn6srQZ%0X?QYahBj5euC zsj}~nfJUB5nN7SR+T$+;b3eTMm)6qGt?R459e1bvW+ z_BWo6)l-8(i1G>f@r?G>`jIUmqu)2g-z0pb!4+X2_AWspG6-9E4cb=|@kYk>M|2Y) zDWjxi?gooYF!*@q#e*A#AuAjQ>07E!yn3eUK3@v8<|M=bHL265t2a}>-F$V->>ubS zV_}vc33-#tLANRl?ktT%h0*z{l3{%ERrodQ1y{B{{3}oE39nS;oh7PXtKKJ%!D{;n0##lOEa(AprGB_Rqa3KTbEt{o#S*2-*cyps z?duhznZR{(PS;YMz4wEDa75kAxjGfAU}!BCiiK^r3O4RSv6w6C&b&SGzrk19HN>D1 z9ek)#=Y|1hax}QZiaOM~92@RZGzM53Ap&S$eQbdFgXh54!V(mAG_~*eeovOZBd#Hn z_WhYT1;P{!+u*e;F16@IjpE)CqO@S$Bx0|MyC&?gr2q zhE!p+)Ic#}2TDuA6UP+-B;hf5M`1NYmsOz(z{_3C4ctWuGF(v=l8e`-`nL~+MO;>- zSi3Kpt-tbqI+blQv8cM-x$*@~JpD>~h(BX06%6rG4@MJ~Pf$DH42{Gp6Lj_a`uE{( zXaDQMxfISx`xMNq8YB(J0d}i$#!2mBKB#!ZSfm=k!=Vd77)5${rtWiOehNygt?d4F zl0T;F9e4+e&gXIa=-xM6q3;z(8+cd^0r`C0a0UZU9={~f@G+r41enX%Wd%VGQ9V%f zTjyX)@~HWYZv-91SA1pa0|lGkqXlyQe|%y9-saq8%0`I`W~g=E~JT8hU?N%30#XHP3L*e#6q_<2V=C;c<-#9 zAsx)VF5TXK{pD6f2LFeH8rNZ;A{NwcSrLX1f8N@_EMjncG|@CsMa&S~7#ru`4!-D$ z7Ij>}+6-7g5^T`Gwx?^rhv&sFF@K|pf?IU?y(e;oUrPCKd-7ohVWB4d6`*2{&fMZ& zwLI1GgfV=@P)8{{j`L94VbIR7>=7!=+@Q!eb9TLrb=dj_E)MiSfkDF-DW3ps&Ym{t zJIcc=Md!+{-@7ppyb#M-u+DW*hPTeHrPOsg(Y2 zaP;F}aZhv&$>&BTygHpV>>n-ELj7oi+}U>oJ^!8ungbnGoZ^Mqxw;1#YyAN7zWInd zcb%z>lK_JISUD3OT+VO&>lm^Q)|w)o)`>NAYNLnXF|#-#AY}*Wk42rZy>-ne0u|xA zJm6r(+9tANt6myCHP(+tseZfW85oOy8OkpQ{kLTL@OOrx+c?LA$3sbT4p^^IK}DoN7wAx}+NU^&3EK<%{7dzA2*S z$8BLCU2+Kk@}cCgL^bcR+xukX;k)7K@1PkZH8R>G`46Lol=0^Ie>8Hnh}MAru~ES{JjrUXG63+AP`C&=hGt!>;QR^pfaEPiyOx0H6~n@a>)h{98rUOO*$tTeTuMhSBuZ0c z!k;*PX4OP-^trJJ`T{ro1O!$R!&{lF7PJE-RfITHC$cLKRf~JhXpeZPWKAQbVaU~Z z2CFG9++fIp2b>6E{C7c1Dz#7L6cg!$#Y0i)+*VrWwebf}U}dRor64w<>+zUEYTgSQ{WxZzj2l}QwNDEQ(Fhbw zCsAQln&=hL#-e$~`hK<3EvvNX=`Nd+iry$((H=u%%kA^9~4njsldTc5$qb@y3L zYkA;oAs{?&Ql{!IL{S8NUX-QkWhC5z?jY!i9P_DFZ>;xx?pG{t)c8*i38*c%u(d`D zJ~zvRG?;{@oLt8)Pdn4K_rqYIVx@pcW2}L2+yJh0>)e1GbUf3BxOtiF&g&!o)bx#q zg>w;|$D;R>exkLhXkhd6O$S2COGLIHZ<$WD$on1-3)*&1Mdz!t}#M zpe2~2HnMH1Q|xMxd#yM$I9AR=?j2n{zmNKY_@*u6BRmG${NZ`%)(I#Ky@N3^w;tvD z!N3#BRM+Pv&p!vs_1m(nP>5jZbV}M*o@x`ha~zqQT-Cm#_^T>&eF}CO^IYL_3-)+k zlAd&qflHlX<=Kk5W9V?j`ig09-0YJXQa!5+nO;84lH3|2uR4E1V*Lqey;PlY|4Y%l z<;8F>m=g8_7q6ZyOd1+5We3Mq2P6CNYsGM(d5Mf?jY_tKe%$2bj<)60cZ1Jpc-ZTw9FoZ_RhrKQ^5 z1~r#b@~kQz5%~ZY`|#!!FzK1b#_@V`c4+3SzFD^WFFUs6#P%?;cX$y6I56?Uw~Dz} zy(S9D^vfa#oXQQdaP|1C)d-GqgcJKu-Gzh_33@D{Q|sIBAf|2%<;^9-;2ra9LYZ;| zVwliTKh?OFj!)a6@i7FP$q<-31Tj)AqI7SKTu#0n-f2YZWFXbr;Z_*n=qo8=`SkPq zRFGCHwUdVl);GO%a2h+uKy{QShxPXZ#n76t@6g(xdP?Vt3_M|*Bm0I)blM_Mx#QNk z$QBj{ouZu)5WX|Pw%kZWTeXB;+VZK&8FdO9+ldbezP{{!m?Zxp)paP9vU?l6vIEpw zx07j41f{Y+=b$>}mL$!P&A1N>2lSX2mf5?0yXRylD~~`u%ea;g4yTv-E$lxc zYFTp3R1+nn$w{)({{@ebC91~=QdE8SY%%9@3fMS5Z{{9MdN307_d-YC37Kt=(h>(( z4>%AQ7r92-L>oN0(ZNAdSpgLS=H&n$#;EfKW9j5hiyXB`MF;(Hf|<+?4NP#qKVk`8 zJrCk6mV?Y4EnbUb{xR|l_*#k0|5~rCb4R&~(S&)%oFrRX`d;pKCXQe!tDn9U5CF2D zEG69CK_WYuwvj^yVW`=O574X=Dw(_3Y&fJg;Q3MM8&VqRTzOZ$`i~ceKJ1S#gcNpU)x8-g}ze<{0N1L6(sS$1@+( znTraRsHDj(l(R>5ZjTkptH(#G#@y=aC zD?N&4;2|oN^77njA+@_!Uo;_)?nRvh z2E!_T+2KmH6L-lN_~oB&Mqj{dy7bnQ1SX72hKMjN#(rbJ2 zSoPUO+?D8H$Q_Vyoa{%RoF&520LmJm5z_H1G`YL990+dyJRq96?rU{ldz?o1!p%tO zLX5o#(kn;nZoSfRzJfZaduV;H3GlrdPiN6tCeo$yo3rj8H)$M`y;>OcpMXu8?9MwKn`p9q`gMEB`!t`+6bq78SKg*HW7V7O!sF1W44S?%+wN4Ur1 z2gG6~YvJUe*~7$O=@+*Gz2OQ1Gtrn4<@MC&Rs0igRa$a88g!wQvHapU4^dJ+#*3_S zQLC|mXV8kFvAjKgwxZhfHs|Xb8nxHGdknl>oMX*J^n42#4{dgk+|kJ3)JGXbyuTJH zv1Ltfa9rz8=%w#iS~^2ip0%P@d9aZ9=MST~a$ql^qI`DEN+O9r8j zz#_Gt?@*$YJe8KZg|k=MXF z!5(duqv@}{LQ!m=l^ik(8LbMsw z8@*pT#S^P8uFBMV00$wp=i%RJj(8rXv0}NiP&b7uoLSh00};ARofnCJc$-nvZey^A zPp`?;^=>NTM6utzRD-vGrwFkp>cn*7jW<)-R63PPtxPpj`3F+bYA;{0>f?G+FrunZ9c4h( z*j9+YD8%bK5-JoSBHR8oR&Z-jLVKWN&@QC}p$t4_gn{w#j6OtgK(Z*$*>9mDhJQlE z+6tP%*8IB|Jd4zd_$g*-5SPxjpX$)92PTRm$(QB1W@nZ$Mg?($_A+KvJGs9z6(Y0v z_1=p=I*Alpxuyw?4_&A~sC965sBvZY)aMSi@FB;d!8@GS+v5XMi``J-sdWFJJRF9m z>P+CX>?50m#K#;#JF`Z+uJq18I58MLOx%Nc&SA@dD7mRbb2o;+Qsoqgw&ewD$i

    2aqUpT`wtOit~$MX zTey6g;ti+fWI36A`GU4{(?5^;+n>rUiuMEhowbilj%ZNKZ{GW}#IPaD7pl>SqK0qXNMGyNn*`uUoenFyy&=h(g?j zJXZho+jL;t*VqD7)ESfl#S?!mkH=S@PXTmd_DLC>qMugg&6 zd0M;K?eW({vVd!EGM`O-N?>2G+9bzG=qvxNdq~+`tb+YscJqiqH(O(x%5VQ$_h6Xl zSce>>lYFY#>1oa0_t5fN_W<6*e=X0)rce^}ra1Uh8^?Gr43!#}JYTVSwEU^3!K4?H z-i#2(T$gS;b+C6!+Wgv=_7;b-TK@#*UkTs+b$d=VO51q9oqe0ZiOA;91YAT(QDYiG zbno^`$atinq|-W0o2B}F7G1uddtv;`&NTY87ENsY=f;jhNd*1ow=@@0J~OonxlX^^ zH*fwLfI?4;2l;M1D2^wG>4<^%iETe#5!DSb-A_b>c2h!1%dA(RB}v0XGR^bA#1qjU zUe}Q{oOVuVsL5Rgh~7Gq>A8Os^eIxq=^#CT8IO}f zhrZ9!kN`>a!7UALYP>}WjCm=w4;ae?yD5&Zn>VDv)Z)}mY*?sw>Vl7#{l+~i9-N5Hl@8`I^|_W3CO6J*ZoSKk69cr4_)N!{g?5{Q^C>DN%>>7sNXR0#mF!@v6@ydCkj(942~Yn zD5E!Zz5GNmqa*ftvt2_bOxH`bF~L-?%Z5*&`nc3EYW4SeJTz695IhRJt1^GY6Q%N# zcR%+;Uy+KtEXua{SjxkZ4s$d*mvK6>iq@>%tR`yMo@l;dK#3Bjqv^X{|LA+;rfMn% zwQugV@@p&?Jn$4c*1nsj=y0V`$nxrj^jAJJI2J8e_)Uq91g(d!Uk2KFH0v~kzjY> zAJQws@xGD0xM41%Z|LB%?C!kUpefuuuJNAIIe2>N_|P#Pi`#vrT4uKGJq-t)ebfj< zfId;drFlFui@bP3yzPqXjP;{?5e6KFsepe?i3pK?#7o zG8Xu&92Bg#oB_OtHp*V)Nz$#kwXAmG@Q230x_s^+u#BywV3-wrDyA1#Yy_zjVOMjZ zL`!a+GmS{*2y@E-4}WpA+iw-%gGnvg$!`_FPWg`tz^>>axch|tz$!rZuL=N?h)MZd z1yD(-`&$K&k>hl$`y=7uw`uKB^-a{zqUYQjosH*MVvEp#vQHNLbLpRcQc>qT%jh|1 zKX?8!;h`;)ns-D-vD18x;~S$T&ca`&HNGjX**_{khXr){!>LqeFl_0YzsTr6(9Hz@ zRslM`TMiI%sf0bknW|5emSw<+L{=9$P{Ytk1Gi!niyO>gTr>p)A91{zj?GxnbY#dp zRXdC3wcl2h&AN;?f0T_WvI)=Z4f3Mz04CG3n5O{bs2gRoXJZJ(0$hLQ38_~drAcui zb?oNTdQ+N|3Qp1kIolZv-ZqsQPad7eNeULpzEjl3t&qu&K&0F}tt9Stp6Ki9J zKQ5G%JtBl146%kjG=WRuLa?Xqb^4&clJ$elHNj7L8Q=#u@VRppFk0dhRy{qYh;$Gv z6=EM4On^CM{|=076eRR2SCxN!6cmyyCJ>?=J9|VBGd%I-GJ!x!(j&lo(R-bJ#=}&o80zH>)lgk&dDHQo(5Y)HfLL9E8+%^yk>0Om&)t zTpNFwW^BXi6@hqyo1L2sLz#*S_UG}*$luB;)J~9NzEST3ykzso*fiK#mJ*|39vxg9 z*aB26iz}51`x9-#!tULJ7m#L*53YWv3HEq|!pOU<({lhuq?!#>%`7vbcyj?by$V=S zndSkM=@Xc@?EZ+o6r5#a&_VM>=j_@$Qyh(4APOc&`l07mLS`HLiDxg-1e_BjhS;&IWXs?0JL)zZ>pd@#xsv8q1*G&U)poHk*nlkDUD@HSw*}x zI`fWiiy}3hhZ$z?DADbVYTWcI46cYYn`}fu-MXLj{A^0ubxHD!M!yFlL4?1O$2In7 z(CxgNDeq)$xwsztnd*K-uq=63 z(?1(m`|hNz$Cvrjo9QWSL@;EXka$kDssj?|6^=4-oDlD@R(87RwQQ7s`3YXn^;SAS zdT1^#ko0Jd)g5W>>%nafquz0P*~4TO97{Heo-aLt_6`hcN zz*$k)%*)=Wv7)-z5w+bm(f&jNroB5`{GQMFT!lPOdoCek`|dJVXlIP3wJncdDV|*A zu)8@4Sdx2w{5oH;?M#0)(n?r{v*TB3km`cPpZ&<)tMH=3f7ItmTnla;RrUOFaoTr3 zyur>a;m~O3))1=6WHxa5wU(_Zmg%cv^_Rs`0_dJ0GAXpcRaCAd90Vk9Popk;cOegA@)z?vwq&gaP9& z{(O=Foma%)Alz1HxQr*+fY`zz1gJ2XZ2^3Gzxq-Fe0r8dRSYVtF4zb`Ww==;){w>P8s28R+NU8tT)k;Jh4)v>58w8tjD_RxzmglOrso zH7x2tB_bkCId(ZL!9zKoH+*zZCq*kHQ7hbWS25En9Cgc6A=E5aE8^{aI96c9-J*QC zRYbM4qMdZ0I8J1ZR;0XESo1Sa&#w_NxT`g?_g<%mFv^!KLXpsQuoYr!75M31A@;zp zH7#nEH~L^cWZolsXg1;@KRR(Rdig$j)8q9bVa)D5`9@kaDHi5mmRtYs^8818iVU0k z=W>fN4Dj5ZvVq1%9rOQaPai(#^y*Nmm7_i=0U>@*Ph-!^E%}h!*Ufr=F1H?GG3Sa^ zGFXEDTyDV_jyGB+9nGh|!u!tH>dG)*W8_Nz6AzzJCA}wxNTIxmY^C0Dhx(eH;B+G2 z#-6=Ux25K5pbzwQa)x`==0F%6ykcb%t3k$FTMO6Lv^w)R2Npje0W~9?k2%n~1^LIa zs##H8>{C507v1^V;O59zF0MbWOwO@=Q}vm-@Fsn+ zDKrD}h*vt}empj!>;TVwuqpDZiF+dumtR`TjBrX^LmtO@FG~&eEp5JtFaY!^oI2lw zoRzwHl$=#2ik8Bc8i?~Lf@j}qCx#!MV<)zd?QSOGIb$D8b1`?(vk?XiYAIyV6L`D_D7=(~=HOUyFqf)y|PY4L(`y*aF^{Q!tkU0ovFxm0Q zYnoyB^$=BM7_PNtHnOnpOl8uZrM3~gpfUtb9&ui{2^NmJP%53fqO1j>>p0J3n|(jI zS_*ZZr9)v%m%2r`S^X3^07Aw4C^7t_vL7=x$TSN_?FClr$X-Qy3zF1#6#!Xs+HZqA zfZZ9JIx*9Ej4JS`-Z?J1amO(9^v1&D?4Vpas*C zQ?v!~dTQ!cayJKLW*&XBs1bPh^+OyH=eN9um@Mo-D;WEx;liff+bPqkpf#y<*Ju{E z)I95!I-yYLtArJct8~sa*AoSg9RE}o4B7QU5}UO%4HE>(ML2hMd1sK>#$Zk##3ml9kXASJE47?)a;y%ONS z5b7OMGOW5@B-Z_KTWT`>tL%&4blKgu+@sG!M?!3O-d4@mlE>2#7U)#>v;*|l{g+ff zkr1_I=+iTbOh&Z__>Wjd+Eggpxj*J7l1*P#VHu5DoxaH2caOuZ0jP?BV0rzounFAQ zA?JK`Ix5H-X89>Y--A+R)XTg$sRZ>g$#{bDlwg~wz>Nd`{85pe(9s}9n7vv?sa>_s z3@U~mxOg@}&P1G?v^0+=6*L8u7r@WK2?5uA# zCslKNjLK~PY&`d7AWJf$w0XX2y3utBqf1}|iE*$EU0YcL z4t0n!$v7aXLIt4st-F2CwkJ#WOPHWi^2d4XqIO778FcEjlAclq(%yp5ZuB|G-Ci*4TacZg9*fF z25C-fqiJlmck?|)NVV$}cy*1y70ZLozlcJk<<(!R;qmZuvje=oe<9`I%T=s)PEtcP zeKAK^MuCoG`kvm5)WLH^&7v;3=4Twc?E|zzX)U5w+tss^UobtrikFnncG#R!lGk@$9;S(;>F_Dj8W+)y^91twF=0T9~*+IT2E*Wef@! zO(%8VEd{L}TYR2Lvvc^U#}>gNj?upz9LdMe19;Y+Fy~L0(N@!r zCpS8C8I=gCA_H?I-{CU~nJDn?`VWimVCx8KKi1>5N=@B%Dt1RvhkcrOiS+Cc(&W&-P(0a~S@L>z;!=2^R z9S`NxElb$lm0=#btCZV!JOPju=5fiZ6DkM*kpLkL@)JQBZW$tZ&x6&J@VY1l9yUhXa!IcGSj35AoW?%YcUZl^X|BL!xGZWF9qOS1aPMXmKO)Y zv-xKR1PT%ciCUQo^9BKTec1zoa1WTVD4kC)q;4&39B_=bDeS@*eY6lYwP~qK_IWo?{89)-wY(b;C8Tj`EJxCKq3SYN2B_V zz>^jSX$D&{jW6&XqgO(eFqzpAN*&f6OyCX_YJ#))z^ga`7Zw&DX8Tqk{2-L-20!#3 zC-WYqAdD{uG2-XEp8o+srj^lnVgw?LdPAB@s%wOhzk7mJXuO)6r6vxlxt79^!V5eN z(o##Mw}e@x{4pBN>`ofUaT>bj+C9DQid$OaqmeNbkx1r%{?;%*X_i%K9rYqY<>pZN zRDxa&X3ct?`TMYHtr#h?Z~>K|J&W-35Z&EI^_E&#c;P5C^Jv-l7qcm56KVXDRvIX- zRC2k1IjhK>uqeKe7;=1_t$EIKxVS*7n6!~lgmC^fL`76#4z$gf=4BNsb%G!#o#Q_I z5ozA;v{BWVIo*D)lL z9zW^4RKTQE9~EHmmC9hVDS0b6dH;qz01?n9p^D}h0^{UqJQ@me#5i-r2)#*ZIoEdd z)XGUV9DYjxucr8zq*UXIF|kb|s*Nx607pQ$znR1ZiyWjXiuK@z+?$BLhoWrFSJ`5q z=8^gmmhQv?TUlBj1{lKyr+bnjBs=;hjyt32njp$EsT(UTWgwkVLO&;Z$rvMchHvc~l>7VYEHs5#?|Lu^@_zNvb29xsB?}zl2^CjUNc>l|nvW=xGQJg5$ zLjM#lt-8?tL%4K27h^CG@e0MQa-~6h0$V+@xk@+FJQ&~XM{~7)n#(D$msMoPC8i%rUcKExYgsz^CZHQ*X5kEJr9t&uD47JQe@~=c->V zyPbXQvV~&6qatDgz5#PJr@Onnz=#QNlA)QK?_ea+U-B6Vzwv`2z?Oc6kz{=a?3NNh zyivAmMF+dhCHzK6knl|cPf4X%jpNbA$tW0JEMQ)ZV zG}y{FjNK8Vuv#%%;BnUg%GY zN+)$I-y?|pc(avN*-XCxlOnziL7vi@zh=qM+0pmBg57SrBxOaS&B!-$%^pFcm#<$Q zwy-Yf8nY7?ubF~LJ;R*hp!^bvbphPgR&Tmr_A-_tuwpe2nC{$f#+0s`PRHr!KpD#_ zWEdusCY3s|_O@)fohm9yFTd-je4zr3HicWQQ0=@zX7Etqe$y)qC#U=w0%Mgw<~UU{ zJ64gh4$tSD=Hk@cCO7#yF9*ohf=HE9)8l)0v}%(5@ckRV=O9jN%~;SzspX@Vp;Uu*~f==baHixSie z?e;8peJYH=dy4P-CH`0~^ zzlq3YmmQCTY##F0>&Sui?N@gI+%e}OoHapeo>Zhs#G0Lt81Qj)^(g3~(6pTaB61m~ zLbA7QzN6SalZ6~gxMA|e9+<-r@gSjHS{u%KRA*{EPzG3zds z1ZTHm!JYTU;W})k09=<`&Cisg zv8u6w%uadwePHp)gz4}M@jU#l@hEm8l_+O*dc0qyVjw3h7}UW+^5}7(>c1S$=Cc0m zQ%Q6vt!0XBrPWz~aX6FD6-x>sW8wA&`Ew?Nl@#bDRxN8aJJ0d8K8!Hs7iv_K$d@xC z9V%2HyiIhJtW=e-Rb8-Auk~oGq%gKs*^^aoM_;HG9kNxvL00d6ny+~c_gxXLLajfB zD;J$f5jPKUDS&mi4jIDeM8tp{0!}8knqV~NnOHG)`@`Y<(R%sw?ZO{KnI9vw<4Kj3 zU5IvPmv1egic;Vb6*)5#l&A<;5g-*TQ#yPAcffUu35cxB*4RsX75_c^ko~Ar>W}H2 z*Sjrs9}dJzE2cMeG{F$7(b$dtbzz;mPKioX-dMTBa~+;e&6&(tpcK$ptLFaDsR`zscLLnhM^&9k6Ql0u+aDNW;68h zVG`N&?4drm)RFUJzhLX!h#Adk-_MV~7^)X2>dc@N5TC$-;}9QWoTm5Ij<2ef$XKe} z=WbZ}$))vW1%qM?M>hK*Tkq{?%*axV@2wLiHSOzGpRX`OV=A|?fE1u31+NhDzX7B-hoo%97aHB6O~x9dFyONIynGdndz0iVPZ+l*{=-LiosRSO6KocRg#Imzgloh6dRtEb;Q_Ja@1@~0l~7* z$~u#iX8F~WDuZ+-AcMgttsc?Pv{3-ba zhi^-@;j*0P!iQb(?pNcLo@uTl*>!;^2=qKDT-N^jc5$u&}bwo6g*sC_eh7{+wG!Mk7b zL6L9?5<~BNFK31#DZXs3C}O!8Oo$DuVup^l`CIX~qkUnXw61Ovyls9@^cFH13J&OD zyk1ekqW91-J{@Z$3ZQ?BS0LikRZ18um@$uIhTus18Ag_$GhN=`qBs>Iq>5K)`5BLH z9?10?UnfS*M!QH^Ao*}D-6G?~qKPfGi+Kbs*TXWl{v@Tjnk_c(zLJ*-5{=J?TXCZt zP@UC+B)})=KoYrfrC0I+1O3B zL6chfyZUr}Mq1AMDw`Lr@3%h-o+lL`0Y%}KCW`cCKuyP^l3(_;h0Sf zKtUGm`O^vYo^$^S`o=}>iu2x`74H|_5ScaNK|n-TMjtgX@59DIPSN35t=vo0FR0+QJPiOv^KAUCocRfQ%XKd&)<4lON!foQ` zfmb^@pjoOK`;D`9Hs8EsR zH!o8VWMF+Da#mu*1fV(=G1#rUC|;6+xM|u^q%4FpccGvD%0ZNc+m6Vkl<0I$GKw$B zjOwL?gzpDP7X!hV*|ft`2bgpU%hM|Li2S|Z9lW%#whL(T#Q%vO);f(YLgRy~O!T=Bw7VZ|0>HA)Ibt5EotKLwz18AJhV=QBH z)b{m0Ev5~Ub+)tiEKaK#0MA01j{0MyFoj2XJXsQ(l=Xo=zLi3cJnl!h&)X7P*c7Ou zyP4#{6E|@epN%}!z-Wa#(iNiMRdF?O9#_Ofl5M44SRkehBJ5oPAS@3buk=j3{4T{N z?ySqDSRF|i0N)!#trejvZv?<%mME)Gv;iw6%c1QBC)YdASKhWtX*D2ewD+=Tw&mmL zy$>d7pN-&P8@DCBdF1YDc4eF2K6X(pm#i65r)>LaE49?5 zE&Lnu!#H!wAO}*X>6hu6b_J}*M&);zQ#$S<2Vef@MaAD(T0IT`H%aHz%Swprf%^o(Gz@j-Zc<}c@`d=I8&;lKJNwk&y zy_$Ohr^`f2_VCf#O%0C_?Q8=~le;~Xn(G;B60;lyM*YDHs-Pt?Q6)h99-m<`t2=iB zC}PH zN`s5;qn5I2hxXi~sWmI6e~{Y4&LWoHRjI%Eft3SJP~a3%WzI+K1~7kmk|J!iv6T4% z`zmE@$klDTyk+)-cE9Iao?CqVJ%tHk!j;qal?a#ioH;L+X4)O&ffs70FETd~F+{<* zB=h2A=@fX?iW1YZ7uU5-{I;?<8Qi|_9&cU20Zi{~Nf_sdrQW{ghc=&!2wXZ5xqlo+ zs$=|Q*_{w$c)!Ov-6nnDp_TO`lad$u@SaJhDDLy;{g26Q7NXR}#VEfHmEG06C-H8| zbzgN#=hV&y4!8^|{yH)$xfoX=xvl^B>-+v>yLNNTZF9%36UW$&wQmMh+;c*MHkwWw zA3E+zul4sVp+9B!rZU>YMk8?C1c9o#+lW&s#=|A{C zDF}aGn_z$du$ANzk$CnxIUZ58CB>%&I5X2g2(V2!7 z)|Ky@pT#3i188Zop$?rHQQBGw-_ql2t$9q zc&q8kvW2(lA98|VpsbR`gYypY_3jG@mNVS*n2QY?LgbtV8%p4tCLBx|(oBLGRb&+# z!C2jD63!X`-}|O##>zCBclACt8qi3v6YD+MTJT2TNo+Sxf*J`BCx1z0mY__ny%WR1 z-4++{mbz&pN{Z;kUaH(|>Ry_;(8FFjlu>3sLoZ$1EY-4nY(G;$NSgw9yvIkCq%nK= zO-WZWoF+qk`{7%jP@7C?LKye|qU|jo>dc!h?G0>Pf;$9vcX!!naCi6M9@t25cemgg z+zIY3!AXJym*CEH(*1PzBj=o%ci#CA?k{)!R#mOKmMm%t6|D_TzB6A2Rv{c+IeC)L z^$33PT!4*Xo*@c;Ns(^&Xl`-wU>RMxQ0I$IZgFHWZ(OBsKzT(ltnx5s_5SkRPL3?( zD!w);scu}I4jus=L<5sfnG*t^Pg>)gq#SM2qT+$Q?ZeJ~Rm&zi$2b2y{9?w8IbzLY z@w^CH3k|^hPq_VtkFjs(GH0T+Iv$)Rq{y}{OV>2j+DgJTw(rLtSP6z75YqC#UV^$VFXYw#_>()-6pLT3TFh4xHSUMbEReP_~dS0ettY+2E%JWa`f0UGEWJA3;7&sEK zwbb3t>F?4n?8joa=N_6q&njGQYJ8DsG_W=mtk`f_wZ9$IvL~Ya0^M7lw5dew*Rtdt z`^X$buBv~o-3}vn8N@Tvq53`9&QfQ2cxe{|xSoyKU#l98_87|4G-z{-6 zsnDyyy46knVaH>kmRjeaz-Z`klo9wi+t|;lYEbBH^_(1k0BN22_@$9H-Z& zWmDU&ZPR)Kw)YqMPXZY-nr}p&(1eYiVwU8zx2?;`5v2`qT5iiER7+RYo<=={$*1j2 z9=2ZRC$T3RwY+knV9!$|NiE!8ers(kZkRCdK{c?pMZ7o?qZC-&1Lb$}daHdv(7@>x z-iAVOFyb3B02-VKNe^>|Z9d|Q@M0q@W;cX_6#$CPv2XU6CICJ-cHW4(KY7~0PI*@A z)Z4NJ?sJ5b{jwewd9P@ipolZ6s|gC8G#Yk{69v;&`ap}P5T`~p8C$##%)qTwu+{N< zE4$58VyR)|cQg`UkwP+7OcuzN+E^6QLJEj!02v2Kc$ysmgJT4&z?Pqwu_i-r$2zK# zaGnIlMDp!D6fc)xqsbBDkh|)AeZUYf&Ur|Ed2&=dy9g%QIOUjY*4Rt zR}~z$lRgvqoypCaO`NvS9+FqnT7EFjIeJIASuEAkmqQ^`Z7y?xBbBG3NkQnkkmx&u zrC!zmTk#Sk*;UeR#@Di!DL<6$oaakRddPz}V05*dM*%Bdy+Z;E7D;q zm(duVR#>p*GeQ;(AMhJ!K%BQT@xMU4{{t9K!aFs{`90*0e9#>mwF znPqX7Is(f#F!$ z=**}e$Vbro>yqBV8KNalmEgm%nee5GAO&kTu=JEaY=&JogrcPaAt++?WSN}2Mk`)2 z&~eLsq8%7DMWNE!>Zu@1B?C#+zKTi6C$mmD{78%MDI8-c*xXaY!epa`H0A%YCMT15 z&;=82$B4Qa3CfjCd90*_4~u6~iyzbL@ulq4eqv&YbHs!6Dr|9DW>Rg>j7R7~M>`c9 z_sqvnqPT^Ecc?valZ(W@5=!pujthr|pU$EH@}&7XeN-(&Omx+5<^l53YhJdiJ+Hu^ zfj)-6=M74O;BM^5stpzV9H`%C)l!B!`Qb~jIKfaiPDO8L&DWG1HGep zlEoGlZVQ{3Ox^VWmZcWN*69$YesdAKy)k@*`izzL=DvJvk~dZ}!n5PwIad*0Y_lgA zk{udoUj$hdoS5oGIH0Ln-m6ZL({5#jB3XIvD7=**bIB(qvl4jQRO{r>P=NDpB0x_$ z(2+i&WJzkDo=8W@Qr5LZ9DV3eSvKRt^<_>Zuu(>ocH+IPo45sab8YaLt0z@McE7N~3tzRz1c`F^8O(WuDr%9Z@nXHR~AW!b=WGtY*x!LT76ApP<-h^=~9ema6kVh70#~;}I4$(E)fk2|)`h}Bc;AIfQ zDe#O=_!rhi8_9PxLG^j+F?$=$dvVI#5N;k7|ETU3^ zgm`j*dfa742A+(Z=h(TH;4Vx*SyaBKLUY)fU{-CE_deqn_TPbOaZbF`CHj_nxiFo@ z0k4#E2(`>oR~D>JPn9$J-1qur4;66?LRCQke8binOmCIkcn+uXBUH3Bv8o}n|C2Jf<^Z!!<|1Q zRB-0!VARc_LYu!-3PHTp9*n2Zz*f&jn8GFKq= zcY64M2qmz#K-!@C-mU4!}=C#ajwM=m{X<38dgrMetN6Eb^yp31nCbw3zW? zf6ZmJb{{^U_DQj?78=7tihbHA>5@57Pt!EZ~*dH4&U6hHLgP-1N?bOnH|H7 zbGdzQ&_Q!904NfGj1vH*LGo2UziS`B52O|yBFtSMhTjKZf7wIj?jhoEVV_WC$7ctE zILYB9XuQ)nkVrzwYx%Q10mv-Ov1uydG`u7ntn!W#sI?KHE>d^mz)R8)B4}^dMbQs5 zthu+Lq2ki$bAVGvXMyC1c27XnUgXd)aP-m{>>33-9SQ2;KmkT#t436}s6@kv$&vH9 z_eDWE1)`9$^0KqDkw?mi11I*JNJ1r9%vdo=V;mzGZ zclHDfcuKG(Ka>n0>D z3DmIgJz1qNn@8dI0kMo^q%WgGsZl>J0c5x1_IMIUDPriZoaSM~yThbFcNo*42#0+Y zN8SjAeT?uCfWf}1kyq4gSelF~>wCPY$Yqu2ru2k9VD=sAN4Shi&}(pbgr^mAhO11~ zzD#wij0{C4B8!e78UTU?kSPJoyOhb9my!RP*+UC4S7*<@Uk3-&8BYTOK0=HRpV_sqSB6fM6{%hie*c)fH^;EV4%$(zSmv5~YIkij@=(!nG zqDqwXmY^sY<49---t2)Iw!A6uj_m&sz zV3J_KmPZ69KvkBF8Rk&z?rtQOi%*kwvP3-!neVk=+gOlKDVy^M75D{Pj({(}6_ihI zmA4x1$3Vx(s`DXts{rn=_3m$d%5QB7fCPf(SDylbiVDj6r#|KH>)pM#f8I$`NCi^; zW4(*?()0VBG?w}1|0$Q^A92lG`HCOjuKwGd^xPmak`v2+x|7yGHr-fLW&d?2-E`#D zus2JqocW7O(Nu0&B1zkB^|7+S)<337x7lK&bNAU9A7dLW42Fo%=(VplR-{2zDH z{J`SB-${EN1Wb}Y4Rv+;Tmx598*XBsMei@yxs$4e4t_iXU~qH6-an*g{DEBNMuA3* z!+Dx0U6|j3Fcmaj`iqIxu%F&=rRjQ1n&2Z?d;Q|$=u9lpCx|819 ziWPyO-H!WdC5;D>xy6coqFhcF?c>BON@O9mic5At*3yeob1zeqHbOu9k_I<&`8gfN zDV>6(yJ4g_(;o7oILj3V0uCSo$lJ~Fy>8tVF)*1&W9jL)w#pA147VaVN>{@$L3G5D|VS#-$N@QAi}bpSMel%YcQQXHb8)v1Rx1Cu2VPKPo(y@5ON{U zZ=viY;M_Rz6UHt%4Mp%y%4&Pz*IcVC=g?$4ZQ8??Y&051y_1 zMTdtYq%{oA`YdOKOoK9 zr(>kG0?w+jEt-d$0q!{8gPmQT$A>Skg}V;SUUTdmW3IW338L^Q`<@&nB!Ae`{1iMt z>DZ>*&Ajmq{1X0FZ4F=d!OWE(7L(MY*-GsGGfofu#;^~pV=p6|VfiMdTLe4ePU683x>ZM*ren|U_N-uf<$ zeG?PW0>Gj|gw2=+BD}+5_E`=yz7c=@15jFqzj-F9>J1MU2`fBjZyPxg_$Jauf|DYG zHQezrm9Qu*j(>yceW4f;&NtKkjoWGnO}k!1f;^xJQg0Wn^Dmz zQgh8D{2pqDTDr9#z(5~|HjeX=3xix`S(^O56+Q)Vj&|?Dib1_lpeQ=+aD=IYlc!ZB4*Pxn87du&%p|hVV5wGDF;P8(GQRBTpW_g z&1%^_-b0wFGL#jQWJccDD}`UD>BoXYH)5IxJnVNO~#=V z$ak5~WKHw8TUSiVHreP13HI~LQc!5`471SN82A8-vp|_9f3g0A6eP_y5v~%FuVY07 zM8mhBA~t>fNE4$>tJA@o#y1c>d0Vc-O3ef8*JkCV7Er-THM=?ZqYb6ifqP0mgLCzw zoVY6M<|xe>4JM+E3Luabmq){Vq4t5D;{diC`KSiO>_@?GvVm!jKkACjlk?a}=~?9@ z%gjD97)lW=>CMnC`3chLji4^(31I~nN)=Ycyu&Q2R%eC;9aX6?hG`vg&ui0GstwKB z_Q$BJ7M|0n2PQB3wBb|ty=>4MVyISXn5HaCsmmrOHmafOJt(|KYAk#7j-3v>UjEP- zUAtIe+@9gS@VOu_I2lC`EVfV1EG&{22cgzSu8eb?uVN zuzlvn!9DRnp~6~E9i=I)|0v9Z)uR;>Bh4A(sU{HXoNtl0rHAwitxt+yAwlA{5UKUd zO?vbjiZpBrPPCj%EQbksatnx>Dj|rYat8C&$beX&Q=AzR1^&zMAo}&k7>{w&jy+&% zcdt`ysJvO2P0cVi8e2l@_%>B+@PJ}WT~dks4xR4ZsMcg%isj`R<6F-$S(uE}PnA0? zhr{E%V(c;VzT{l7cP|sk>6Gs!@m09{(xsG$Y2yNXQG@Q6Nw%wG@??NthVj5IegfS9 z;747W1PAxC3%zd=L%{-x$Q@laCIv2wdq1jJz57gi)lOv({3+Vpv!0nS72X{TYI|F! zF{xJ&l<=#I{{Y8LULDIXTJQ5A_}_2kejVi5tL2hyif&s^@Ez_Zc`jIjj%qA+umnK$ zpUuI78pV5jbfg1VT;<9O>iPB6+w0og^+~>=qdW;)iI0GRJOD9VaN$zI(IUSd7et0a zhis@>@RdPh@vD!Yo;ppIm|VH7p>u3u)V@Y+o^W?JD=b9RTYmLWd)=DIbqGZ`^?9qW z1-G#i&5K%QCDxWqucEPZLF^$MBF1i;j7a7}de9dVb4ImW>4XMiZHwpz&b>#H_EJTI zh7{uzo%JRtj+gsM7R#$K0z&?(NSr85EA(FcNH6odXz)P!xBJLbhMLV%_2e%$!^825vU-~5+cECNX=O?wvO!GOM>oECwPXWv^?f9Vf5qtErW zmbRY%!fC0OB9Q;dX=x;nee+3ZuJih0Xt3sghtuL2?~4DgoR-qf2!^J;%}AD?oR*%M zU-|XWG#!b9r<$Tiq?0e$akUu#saZ+^qg{sOn`*OA>OIq?--tgTeBEASB zr3EQAHcqM$z;5B|$E|~(tc`<%Jhf`N!>Zcqd$re zxNMd3WbRigD@U&fY(2Z~*%>RR?%&gR_O;=w)HZAAR#Z|z@T(-x4)Qs)26mMnSDg0B zp*9GeFFD9Gli4~Xjkk?4SNF&+Gs#?ik2noP%*E0Xe{je=5?$n5a+JH5m9LWdLWcKo zum+pZynppsm!f9M+~-j6+5jGjgc*vQeVl!akbPVP5U7;gO=0?zjr*=9nPrafa)u4o zDb|{ZUaR&E7kN;_M2}LHOEAuA|A!e6fqla)kkg(5)UC#NH7`0f#6GR}h6o5DpG&B` z+?|E&x*!Vc!a2j$_mTyKc)kq+jaze5Q7)@%W=B~;EwkTFg&+D-B?f++<^QZK^lo}P7Jan{2GmW2^~Wj8aCbl4#{kr*JRmiw2SMj75O+Ke zhI_aN#aoH@c4S2Y0Qvyy$0A9#rcx3z*qdKCEm{(hk0g0bk)SYUMh=D7Y=BEaz(;RT zErwnB5{%163^@|o&9)gE2V!xml?#>ZQ0WiD?WfVPHiV~7#h8K~Fr%H3Mr+rs)Z(gO zzNXQGdVt01q>*$jax$eW%Z*Gc9pGwC7g`vwVwZ8j+FaGP1VlpCL#73Q{5d3Q;TorhVNVtQ)M7l)b%8 zO^-JwRa_TeRcOvg&MQa1z?NKQ)=QR&DTA=Kg?-~Fg26NbHkOh|WT-15k8K4TcJx7{ zju$a2$jCzB=BGE)^iuk07pdcag@AShsLXoFNh7GH#&CRROq&Lq>(r$)MkR2_c=1OX zs$FC(cR&lMdqsLYZl*yS6|*M;$6-0TA>@oi`087yDUQ!`uq0@gFz3umQY z-zhEvfq$I_{=J@J0hB~g;Y6YTR!`{!M1Iy&96UHc_fK}wYf3gCu#!S1kuo4&FC85o zHIBtN;*WZYwWRtrwqzG-QVh|YBAuMHi-^R>tm8ueQ^f@(y*}q zn?5cgP;%*S{T$`9?3nxBOfwe2+QUFn^-`^zz0GIbHss}}cpW}=qNN)G>+L$7{t9zMWyUMxTUz)?{ zED@@5kCGpMiDYaTJl`81s#q5ux*os1{xTf_9_&8!!cARJ4y*(OO7?GWTYz z2!h)T8U{_bNFbsEx28Vv^=)>r_LEwsp)n^dZj>kj4Pgqjj#)PpaKPgYmH{8tYolCV z_zihB11$MBFv~$7qSEdblR`B@cC=Hpozp3$nkje|1(9rq)pFbd;0P2L;)CJH{AmHh zNtEk@`KrW1xTLvR?tC+n@@logkT*)TCaU}}P1X{EGuFvj;Jp_POaB?;^74|F?|c-J zFdZ&(rqV|J`jJfcc&4%x%Ue@*_10Ly`MXsawgE>9x>SH-L4R5Sz7|oHb zuW87-De4zA+?!f<8_u4s)ILUX-sTl|vPK*kOqjr(bVdx)JEmd^PNylu1ld@XYX*jz zS4@RVAE}(8*k#nV5R07pz!HwE1ig`b&nk*`v9r*RrCV9&2J7P6jV>QRkK1Z1)Zj2e z^(1XLN`GUA{|mG+E%D0Ubw*4MRYYFiB^U!0Oo0YBF9Q&3Tu?UEcY|c+oNx_;zZ)+B z+;pg3Gj{zNH-$T^v9W zs%qy}G*(&>H3R$-y_-RfQGKTX1>qLAo5#G}+?Sdv#H*P8MzVfisHDrw5l1uh_OM}* zSVtST-QnS+rZwOdOZ1mVi$D~Poe)ucF{|zbg#M` z1+Lr%)!gg|w;vmIk*gUUbIWMJJ7-T;2BH&q~En$GD%0FtpTD8Bh+%a zNW*@brm`km)h5=x1if6X1eyOmdMybLyH1tFhp74Y>Gs7yyf1b*X5N@GS^0%3@eo?W z=kJF#6zVe@(s6@FWirPE5Y1sOzj5Dazo0vw%jNVjP-w6P0r9U8vs+{^3-l5v*xO{& z-NYA#D_rS8Hk9zsB9L!TbP_1Spq1yWOYv^*sATpj;L+=`KoI?W;)=uKqCFAzv%GRk zdoqQluGB2uct{d0!PGR_d#V};Sl%{L+Sf_x8N?%RD&)(hXrDq=u9BvRGv=5VJLYXL z&BaCAMzc13izRC^5`55Y^*GkhJ2DQYvZa#QH}46_0|1RAM#wg9(82masBm4f$be5s z#nKdb)E#&!f?<@&BE#yfgBmK{wef`eW-Wa3gXnTR$VwLE!yVNz=cF?X8RvC@3C2%D7PK_<@64bZ$gI#J|IMn z=>e#kgE;M(jC1JM54>aBez*|1J2mPbdf&7Y9ywSGB5s_+s3h@-p6*+qkA{>sM{P%g z%t#mG_M|jPFDOBN6JuYLyB0~C8jpQm2xMfHFS?r`f{n~)hc%oo%BlG)lt=J^RXRKl zH^*cbHa>C1=alR{aJKKdaKB`NY8VKy-e~V!)Jjk$nH&ieDz56(!L{4=?)oaecVpi~ zx?uIny`&P*hc-durmVQQG_f!xS2dyt5Zl|d3Nf&otLj~f=VO#%b4RBRw<*>jAfp@5 z@gAl!EXqUpxU1h!s3b;=@xsm@P_7|Hn5`DzYMEkcmY8c7E2CdZVIhbaYj{M)P_nF0 zBNkS%jZ>CiTUX_DzPTlOp^8pMB6|CAr<%*z_# zO`~xcu~mN*A+~@x8(cU@!tkoiw(CT_rW>a6BaT3v2|K!?hg@YHR;E_q$D!I7EIo2v z(pH-}>=m!BZ5VCc<>w5^jt!LN7GB;H=gfz-J;mm>3FH&E_J{Se9I}8dg;VdzFhZVzQafYfRo@MmFxxUY*1?5%?~HI z7bljWTgrwDex5wSql(>?^s^%ryl|K9$+*ucBW)pkF_x$o*>&3CuMM%0hn z{zXAx-C#60bQ+Xtp!1E*t%?_8A_>wN;`>D}$fW?R0B5PMo|4mngi9dUAUBw=02efm zT55KaVs~ESK#Sl&ggg?lt)LRF_ch9nx9ZXwi@4MUUZ+N$wbB7sKfIJ((O^(%Xq*wD zQN6g~f}t-ck-q{Ey8%*guU~trCO6YZI#EcJ_>_hOHRFIAJVKd1gbq>r`fG(Aa$w3Y z3I08z^5=vf>kmbUMuS@VUrzX4t^S(un~DE9;ZJ!s_|;)6gne@OKPFHAKH)FIclkFG zD!27e*WXE~C(91Crg2Tga-Wwx>&&73$)&mk?q2^+LY?4b-!I9@{4VgT!)8#Wk5kY! z)`f0%>2pWS`$<9#fP$3rZE-Rh3dQ(&q(uTP!UGkD_%-+EgkSgOCbs3~^h+rF%j=iN zYlqr}#)>JFu1=rtGgGMwx$7JEhbI@zu-KIX4!Ol6-01HPp9*sP?t5_ZOe+R<@B(g( zhlm1Cf9&RhjYV-u-7ybvUWed0v=IAKoGcQ%jNMu0*ptW>eG`VU!PycX^@4>XdA?Q6 z@)22^E6{vp!@X*E6CY+G&8oii#b4O{%0v{cU{9Z<8#CAlRl)D7z!D9g*F52lvDG9+ zbB3Edm8D#B%TW#6RnJE-p-esQ0FAdGY2*x+%5$5QmRbXKWtq|wVa+90koWt|w_I?9 zCyD28BJ^i4<&T_*D4na#^u-A+)6g%B8T)Fke50hL$ND^U>beeLbXgTCFV=KE{iGM= zX?}wTYK|S(_&?26n|kPABwdD&$&i_FuHy&z>TXIXBSj7KCR~N0MCP_tO+S_ks_ILX z`Hvgmcg(6GcXJR=G(PiE5F{b9Vb0lqsixQLfca*7nuR&he)JOb{z5mlZ40?06c;w(5}>*v=p1 z)p+9s&s&RM7QC|yP5-hS1}Wy5(YV*Sfoi(`HC zAoj_4(a>1W0uPmy3^IL3g3n)zeGNT0(Ro)A;hYmb+4D-`%SN_%jJRTE3x=&d@g3d3 z%i4N&=T(FKWH38@wwCXnOXH{9F!zgw#x>{K7-Wo=Sl)Xi3Pul6?jqaHkqch z`*`}FDt#YYwVHbH4?TD~8Jv@D{!k{H@o-m0$K|g#Czlu{pUSfd$He;nL(JBhs>n8` z>^o%RT=z}bawxGl(#zJNO8mXL6Z6Z4+=e3?DAlhN!w061SfV z&kR$6KW8eVn0o?#Z=k^#XI3BuVu`?IT5{xtmf)Wabi*3?JXyHhQPvEEBxs8<11v4S za-QO*@`^}!8K|}K0iwHu94+{69AIK8RnB1wmG*VioQpj%gW{-L zuZ(JTEyO9<9BJzAMgB~C*Dp$KiW>1NfUMgkPM!vT+TCMx7I~U4W!q91D^Kyz7d=)@ z08IQgXmW3XJhC2n~H7FWfP$ugP2mJh+vNvq(&7XXk? zDtOr|41;luR9)M|Y98KYOhAw;?qbUoa@)7oF-9!cX)M<&I?%+XlL>OfB1P6QwJq<* znz|jXTo4PJey;xRUy7tu9>{nSVL+{%xVbdqU{;&55N5=Vs6FXdSeMfXXMshw9ILQvBCaY&D_FPE9CFQ4 zBsit1FNf3mVlv!R^M_yi9-E=!$zJch+}dU@Wea|Il~d~2+BjKrD+aGUMQ$YOikL^; zJaYI*ozMD-cQe(R}hYuzFdO zpt##|47?xetICKd1gW3!+-svO<+t!xZwGk{w4z+bw}=x5dxcQy;zH%O$q4R-L5g+1 zCj9(&BT9cv_%F60G$!j(n#axKE4iCRPU_O|uIgDc@5ZeZ>oXR|zjL)~vyU75)n|Qq zjQECrLxfz$F2iLfC(x%kwF}+g1^uKCbDQ2G=Lwhpsom<7tF1khxWNh%@}J*PhCkj> zh<|xULjYty-%+xj(7zMKFRAOyqyaA~zrCZ*l(LDx6U8lhBDq}nB$n%_<#J$N_J1Bi z&!mgRMr`lb?++Bp{`QXYVbU!m$opT-ubl5J{Cr1~M;ol))Ie9o{3MF&hL}%0%ilHq z_Kwk1v=25Gy&)!ug)KwAI8@cfo%)CWPPT=SzrtItV5Zl83RO2t#}?P)8X6<7gCy8GciM0Eopk)pUZTe5>tu;;hLMQKpYLR5&1w&Q8>t;rH-h^I4@n96Z? zl2nKR@@%XI@bWA=&im%ky3ZgA1425ZZYZV#?VWV%f?NCq$3b|@6xYuqMbRFR7au%K zHz+L=orzms zNri9iZAy#F4a&(_A_iX&4pZDd@2|d(=S!k57jLAauIMt*p^wgsDBrE_Xv1R2>MqCM zsf^%i+xy5Z8)jQ)H*yCks>WYFipN}YLB#O?i~z^>d~52EQs^x~(?m6}F9P*QXd0S@R3u5He!0v>P_2$iY{Z_mmb}u3d=5MIngzZ`(q#i3kg%J^K2Mp)INSCbr?Zs zg(K)5KW{0&yR8g>akHj6@JVA3myrbQI==IY(l|W2`cNI9|o~~Csrru%50w8`QaOSUcizIbq#;1Jy1A-KD{OK_LqZo%E%Ex1E)m*7rtcY@1DcJI@t&-e5` zeQ)-2f557$HOCxtj5({!-c8VDv|dvGkevw6DsBOn1?plE@{jpIeYRyfRpqB=rwi>0 z+j~%XFXs7Z?yyaR`ynb)b#$){KHvZ)LlnOTofVGZ(W=;} zLE+=RZw8-W-95U}c;A^XhOBZag=c`aq%p_9sAh#2ILT2K{gZVrN&U4GB)he&J+%aqkUuj883jJ98#$BNj!Fm7j5T; zOH5%)4{QzCRn{<>X+#e|G&kttz^~|uOhp}Lu_vd!jJV7Z)C-&S%Gmg|Og4ZgS$zx$ zg|GPnQRBU?x?TkQw4mZ6s1O7ddA7>)+#|X8A)rN zd0{-rRJZV^W||PBU-Vv$-!|Vp-+hMi3{#Zkn}ZKM#qGao7op;zKnXqXSesvKX$o|5 zPmDo7u*&F5W+VO*Id<|u9CbQIg90yeRu0GOLMU-P+$zgSB$k=$v*<>TisU@|hWy6j@!1i-@WM-$a0!f@pohmL9}n-S>#Vl6nMidm+tcIiV#k+Tw$fym*E50|GoZk%qgYtADlN4}H#-Lf98?iNwf&UkdU@Fs(Q{REbp7%XuFu zxuiB)*)Jqd7E?kvlRiyuEaHILI3_wTe5{3WFEbtRtiR|Ol&9Z#GN|E_;OUz@bgbN6 zT_J1o^Ya&}vB_b3tF`Ato#7Bef!*A^3AioV8{@0x?YM=+;KR(em_+BZ>41@&h>#-e zExH-r3*W7gaC>vT_-RxCupD)->Hkt4n zKl!cKMViZG77fZ@?YlqQ|g zM!oC3q#9>_n_+5BpiOWZE>>lO7){nn#+ zf4%Bmi*yPuA}#xTbniR1H8#Ja(-jw86EL1<&le!otHX}qZEzq0O!kvQ;LB}7lqE8b z*3w5XRxYDHdMY;oKz~+|h+3}(@!skM1@+QxscCf|3RZe4qy^)?2K9F${)vyYqNOES za2+3X+J9;#u0X0wXi3=@Kp*=QjAp@k3s6 zgI31TJ`bavt3kN131(v7q8NiexpWlHlw-U9;F_uD!1Fm7k-OIRI3G(n- znBxz)FowH*Ev^Y~HdA|FYQ$?Gjh*_%XeW6tiP3SGdYxlLv0VSGe54#X0ZB_&2)uiZ}bx1O+|_2Lyh zLJ-)SdAXmj?dgNU>_ESAJ%U|Dbbo)ocp+*)P*z*(zz^gYhJgNz6jK0djS+(OCE_mQ z0maf>MS25l579Ox%wx>HiPS5iB8|js+bF&GjNBr9FDTo(u6bwD4*EsWxIOS&giQwc zyIKzSJ%p0d^^O}Sw9NOK&r5D}G@fjFQf=y0`WwJ?8PkA?x=ZsX2H*YMVB)OrcJg?G zgPd`zh0&+vSLX(MZGN{OZJAzIl$;EO)^2-wPDMJsT2MS8VY!hD*XxOO+h zszUkH(d+lXcJod6ZfGusS-DWBXTwUZ>KkL`;D}zAyo6X>Q4xYCpWWL>`d2M0!wjEQ zut!l(bF6iPdy9N9AP27Gd%Fgt6*xZLRwYbC=1XV?xpL zfrhCv(Dp#n<{7u&Vs^izO4~IXV3TC6nf6R}@px4WccOic6h&{ndgtEN+lQ>et|**y zN7)5-$S;orMIbVa-!FqGJi-oB7&Gx&e`%KlWD?#XR`+}PQg+vvAWPFj_Kh+8d7-KE`^hx1q3%1zU{nS0%@msc~^`38w{SZ*FUOLhA91`SGW zUX%@;C>qE!UfsjfqWc*PVx0u)rGau%oLrMeUP*KtnwHA0nl3T8gOWd~G|UTlf94y` zGnH8y44>f<17|U$?~L6RFrFDF@H^vef>byIjv4XmtJTuAoP#jpDLU>O^FD}L5s-+A>u za-zYf&yF!zPfVi{pjgH61l@6BQ!yqpwm4v}I$oF$pJET8CXW;*YlEIRalT3lPfu+s zOdj$-7amprEX`*5AzV(iymfef#1)>T)#sAgh>a|fsg_jjxaF+G56@S{?0%q zi%A!+sw(?d$z!5j!J*3*9$8l!VN9_Tq8;zY9Q+iC-{Z~8i}|_JUJ10HA zaD52M&H0sC9F2syCnHA!nq%X{%-mefg{(6c|G@M^TPq72t2K0yR1M3lQJ%<`hNPk}0t@&ZcP9(1{ZbU`l#0yh7K~G$P66d;On$={^dbRCs<`nW z*`Fgxa;SI$E^;eR`%e(VYi9*d*-!1cP=Hc$$gEO50|7F&6EyRU%Iua5IAsWWOf|XM z;xiIe=V@gAJ;oPH!O)y>dcg=~$wJLp%rD-oLuRF0R43Hv1Vy>;9|j7 z$Z~D%I+3ZjHiqJH5e0=>(iZn_|A`QetUU==Nbn36f}pne+lHYkReMnck#LO`a!>Gw zVE;&6(2{5Q4H5xFowXi9@aFz@v^Q=A_mHVC!x6)i@2ji`7EqdY%_=dOlm(2?3&0U7 zPmc)9;G>&rrbNbqL(Bn(AT=_)ghq5?6BJIu9IiP7{S%|%RJdWtrA-8$cci__&Obde zof!_D52=qf=-kIycDQgciiBiA_*N$5#;J(wlH}$<(pdGu5LhXKg9-JG#pOBCp~)Rj zolB(ll7V5_#OmgQOY-0JpEH-YLkkFE43&k?Kq$nBg>F1>&joTSe%s0Ic94vOM`k0l z+5*QMsD!-jj08|-d!69hbc9L5#@>&nLPL`A6N(UmF7)QCpD^Wde@GN(6bX{+-(0u7 zZsdO=V9~r~#ABWan{)a^`!SU5B%X2W?3_@Rk+N=D#*}p(q(nELA;Ws%fm;NuHYAEe zX;FHBf+;9NVCLCcE`E#xTi?Wm35X~%_03mfM~T!_jJIf6MXNX6QhKwLsd1D6iwyKT zoh2zlR@Ay0r8q?uttKB`@3I=iv|#&=R+je5ifUzbHf(VWc)mNU3MNlNIE@#E)VLig zf#!1b^|e_kO(BUOn%st0LRe_>c>`x%2M9Sw{!GdxLx}Z2bSEJQ#t+-w!#VczkaL1b z-zX5MSRuFfYiREmih%=Pk%5kCfw}CKOJ9Aye<_!~?}-}92+3b)7|Y(%XYno5<{PXy%M+^u zL&oWQ2X-Nw+jfa`$Z>%g%Y+GM^AfnChO^w{%zMBd>cyo!IDim{t>>tnv>luzI)vky zi(i)a?xhQ^EFZuw=(5W_FH&KpQ-G_-pOYI7E5K2 zJJv^`#4(uGgwU&Hs#QkL4^;GE8;)qP*6e2y#4_16Chw?Hu~s6;938WkNS&O+%N_Bo z>piy6@B0YRYtWmzp~9g_D+*Netb96h@-%eij%ZTuBbnDhFbiK*vIHc0LS}BUSK&9q z(>zB#YH(!ivgd2^+6c%+Z4(o{UA#wWE{R}BtH_~;28D=ZU?^!OB*JtDgx1Irp=v)Z zl-R7oeewY%nTvdn;boL4Dy2&K9BFlwZl@qEC0SY*;R@Y#K~qCP3n8nFV|=%0if`0? z8LRhT7G$gc@+}qo)#X-JmUzGcH) zaeVcRGWy2RdE=>tjH6SfOn=cO72cv!Yi{!#C8znSQ+fxajzC7SBuyejf1iF1Mca2| zKog<}^`s)>ZeoAsa&}vehc|*S4qx(XInG110JEI6#aNvcyk0Y1`jEUq!BY9BQ@2{S z9hap81Z8Udm=OL7z3+nayw zFos#&mW6u!lWHY4KZ7IasIT}U7M6qJM}SAw;A4zLqwjUvWB;|aky%K)uM~NWopW##Z(N{hxeAK#(|^gp<6J*wBg({XvH_D3;dRZbL{j|b zva%F+$g9Qu4!ne#*Q%@^*~`nS_{HVo>b99{$=fBTp@CWYNflj{`FxD87vaEj`)o7# z(e3(mD^O#)IzL2k(rBiW%YM1doLW~ zsFzEKWZJ;gmvE1s1jro=BL*xk z?oqpBD=dsLnFDKcCrPH$X3vqC_6jY+2=Rpd7OB;ms&r;EpLecjaQorNp5?#OrXhn;}TC$m*eJm z)WZzbjU9!2j0_jMLo>WC)I4B@8wYEG%HqS`BUUd+OPgX#7;7P%7mVd#gg#xI(<~)e z8RJuL-7S7BTyT9NXgJ`znMW;gcIvz1x91bVOpO{J*d|4!>*qi}p=DfveJ4W(mdd1+ zdSHV|Tg0@o^NPqf2i+!h7O9Sj9ggaNFSA=u9VK1p^n%EzrJzuKbGp>s{L24Syjgpl za%TU8%6DRc_Zo3GWNBkPUdP?c1a)b;xjYoq$N#(~>IX{i3IYa%^HKa(g@3Rf&AcrT zHB}EDi?sim6w~xN9d-Sx_~F!>xBgBxrfEeM&$J{dSl{?3{c6S&AU1+G3d0;c_Y{bq z-W<%)hqnHRjWhoFfM?TG9G^O{FR2X2`>g0F)5Ksf6C%CM1&|OF;nQjaC)jD~DWyQi z?n}?i2IWYFtjiwDLo6=j-)=)l*BHJI(&A~S8=Ov*lzPX|TbCUxL|dEpU`tqx6+{@* zXO-$5G?4FuI*&yfgDN6VT)jfr@`zkwmdlzE-Im9K>z|O78fTA$;NJ^!6G$h+bTeIO z@a~%wXwI@iidhlwdZBQ^^9^xA$d4ekAa8`N z7^(slEUcci-Tmz-AH&VBXx_uniRm5swy zl%V{!+GylPkcckKF+!QJQazhaPijO5af-ltVEVsm5_oezk@goKMy7d)5^sf-k1&46 zrbe}rXjx0B_oVh+W0`RFH;dM@5~9m_VgL)QzZ7)V?i-iK-km=&?r~kkG=$p#e~*l# zW)zLii6?{(2ms_jcObG#iH`;{%zSTS1(_IT!fzcPs^tWrmy-cBTjZ(;)km1({?20{7Je_%7o*oQk93h{6V75U>w zB7wkg=pcR&z&`o8_o|+7bh(_*XAjIm;mN~~^rNJ`oE!cGIIcTiFPeAXeN{kgXh5ds zGl3Avp4CKervTH}l28gco#zhBk2S@&w-v!%_tF!N@@`g*7_i_(5iuaHHGrb^&3*2o zrOSq$BR9t4gN9qarTjXxg_T|VF0?I^p_+T2D)%c^*frA&5$zPwdqQB?utCIgI(U5l z-4nxTDdP9NrCMsvNaZ;iiYXKk6HNIpAU*&OL1@jJP7ncL`wMZq4-&60H`Kcra$?c7 z7^0XDAx=2^T>;zQX`?)3NnwzE+t9^8?&!jJ3>$ z__bw%lF)sP#XN?F!|JMqazE?}q34c%g<=pHQW9sH1(D*GsbwIGH()mPlW3@8du#9VYtsV#AC|nY}=;l8%lN>ze@MX-!nL8D`-sZi^^g!1o38toLtP2 z&I+N}U&s%d44|wtUZGkMMEJ<#$IG0Xl|nAuy2x82pY3{U`w>ND;QMG@xpkl6H@da1 z)*59cH!|f^q!5>IqUd;L`_^UDn5TDXApqa9!@Gg6XRa zAwv_u1pgU2cs9O_rgJj{yWTt4^MjS-Xr+;-G0)ydQ-0lt*3iP)Txb%iPaq%wE`;Q` z*i*2wY7*CUpbVh;H5()wB$^HNJVp$KhqHsP1xY+(@h$+t(zSsxMv@+t$)zUF@j*TL zUE~dlYD{&6;HIQ-Oq4dNMVJ0o56St5Pl}(^igbi7huYRpV}S%*t;E+Q5U%M=%=GkZ zEXpBK6Qd*4p;7Zx#YSq{aTQu7?i&?$Hcuz>+NnWs?^?B_V0}IkvQ(8PclpYim;$Oo z#skv|(4({-y`2kO@A&ga!jWO16~e=?Xy4g)yhG2}bs?!FARE6K*CeBGf&*Mpk%^Uq z{bM^O2Kj&8?rvCrH9ptFt>m?A7@rSY)ttL?+(ZvbJ~2HjEnW>o!H9zMU?J=t3-;*x zk#>;z6Hff*c@zL3X^iu;68N?v0{|rOdXyPrE`3!s%jrCQW$D$}Ny~3*%NwnF1yC6U z$9M|#6>&}1{8s1lNuWLi`H0yrXK5^oc3f*sUzavRnZr(hy!Ry(P!9`x_ZK1g`ZGF< z>S3{8gEh&0D&^;BS0inz%^y3@wAB~{iIOGG>Y!(4UQ;Zu+lFcsQQZa{)gT^R=Jxve z&Ey(DSnbVn-UhPE;+(q;pue_e_LZKb0suqyWFAH6uE2k@pDXL__hV^{Dh_uSi%P1f z7Q20M*>fQjPy#uzaWUy}85&9xHMh+y*fTPkzANp4@txHs_LT3HN6bp%Ue*c*@BHg; zE~aZSAaK(A%2MsUe;UjM7(Ft{feYXF_1znn2*Y}=bOox`zAwAA*Il=sQ&Vs zWcgG=GK2-|r`%%{xUXk>*5SHO65`>-Z?0oND%d){msk*l0fH>s_7s>x$Dzg=~C=%@rBq%AIQprK)6@1gQZ$6nmT3(r3&eYqF0hjmrx z`*i3uhWD7!L#q!kx32C3Q&F;hY+klXnSr|DOT;y76pW=;ADpubkeVlUXcyF1<&8D0 zfT7ntkUfxSchr@SCu~P^!bHk+ko?oFdh#X>jkPgR@~CE}JhBXZHp1R7vSWnCbxqJs zPYf~F4z(g!dt0EwZ)-3CM*p$K57hkm(mg94_8B^rT-;L9D_44~8nBpaUDU7E)lM?} z{YdRwGB+$hRu*oslpSvr43ASFziYbt_;vaxfu_NMgWTgMaQr(2Po)$GWf7T*VM01v zgByH6e=^*=El|pT-AWbJkxSGp%fMJgoQi6mH?Hf8^OYfPjKw}Gxgk>YbG;_~Zlo=& zl4F1#$|$Z70~nf5<-ey^wAv>ZcTa_EU@gdzmIIA{I}a~~E*#--ER#Z9tzvi>|FHFf zcHKJl>caJrq~XrNqK&d;DYyp^u@_hOH`!Wub`hnPSFjuBX{zI?hwH9yobuSV4zS`T zsH_6}x7><`YW{4jIf9gjCXS$<^un`+(=uX|N+_gmFYje%=cR6rB&MyAq|&s1QAYPN z;uWzb-^ulnw&aKsFyv0iR|N8J7Y&&P`o38&IA9qi#Cxv$l4!jX5v?+l3wXZ>AXzL&R0z#=9XuZvyfDO=p@G*ECws_ zJk%z==@=QXi&PEGocG)*Ft3eQ1p>I;2(Ajj|2?joea`*){p>`5Tf{8$bY-G z8LUN(*|5$==TrPt6dFp>dy7l#`1b9nPNbKQ&WPc+DXs!EqBFy!aQa3r3qA0pZ2eC%d3ur2Zu3pHg;4D^yL1 z3Zfyhv zaScW2(hNma+pC7|6YRRgH?ZGlff&4 ziQQ}Vt~YLU(bayy2yt|iH{j*?-TYRia`Lnm$*l#KtJ6-{f%GqplF6pLo`#_AyeXxsmuwy zyc>(R`i0P08K7q!?(hfD5BMt%Jb~^#UWH>5$+Qd%4t92L)kX@u@DL2wH9f7{GsV3} zNTn#%=M9lxfm~(ue_#0?T$?}+z=)csk6it}k0g{NW$57`<V#c4ks?f_g}5VuVh@I3%uB56QdX!QC+zWI2>9mGEJ1f)rao3A~xJ<*wy$r;vX1 zl}AyK__bo8z)4}a%P%77*;;zZ$N&HaU4j)fR=^6?e*{X6wdXG6*pLNpqatm_Yr&RGX z>v(s2n2AFI|MAzb(2xb!>s1_A9wbzIhnLDDqYYO{g^zPf|5NszMqp>iFz6{QO$vh}onc!-;TmMO|Vl@v2=oX;t{|lA=_r6##q~@)9T;IjNC2KV2sbEgka!*6= zE1mi5i=*dWTA)g}BM@bR=;f`DjyeBXNDW9k8-H?c`4%pfj|IxWps%<-28MzH;-FzJ z_CW$_1%6f1*C11YyZ&Qj5$B)z>Al;QKZoeBFnSi!iWsc)SGav=gWdtS3g8p}M;GFs zSa)-H*43`xEWe^7=`G?|(1->KNMVDC{)-TjFWT&GKgFLM)rtz^-LqosLb7l>Gyk*B zQ1N%Z>^pl>9*Bi{u9jc&+qqX4kqWi6kymFE)E6i7FT2%`dlHVLKzgiI_q|q{R-3ny z0sz@Flo`Kow^la;1z?IQlZkIXzWPLu-a~e-Rh;aA07gOxt$!bwmQEBSZ*P>iwvswN z&u8mAaBR+i_muzu>I+fC7KqP8z%RGFUl-;EO1x#{C4yi=LQ*hv|K;T=RYeP0PA*za z3l@+|2&Huj3^w|29J~JoiSB=69HO;VZZ%e$l;DnY1eMH z2dlMKu3Z=FoXD)#QX0{Tv~}Xuxoj?T!@hfA4* zM|ip&gI6w6h>zKFa`CCLYO%Vg2u4bb$B`OUZWRvR9fg|KieIN*h+e%idxsdm>=Jnt ztq6Ok%v3mHghx7TYI(oBf|PLrye1f^kHCE1vJof=sC(@J=>dsmMJ*Y8!GXJ^B%BEA zAGcok9tY$N8;OxfMU9i;3kLt0zyFzhdY(Brf=WPQ2QlXE%^C0`Q3et^``Rp+nu$$9 z$T;}$uJSq(neTyUym4amt()m~TKq3igKDZaiC%W#PT%z-JV?@#i?2QWc(&UNGyuq$ zBfb2EH!qG2>)7B~wcoVFp1YK0ihWd@Fu(;$c&bw%m20|%p1EGC9F2Vh@p8j7*YJwcR$pLaFkkAz z{>Y=$F&!A7W(QX9_p)GJ`0^C~qMHhfB;afzx{roNg7Gyg|W!F%v*0OownbO2Iv8S`~2TOp+}pWn0VfNNMNztAU@DAk;Gx- zm1|%=By?5*=nR8fI0nXW{u1uho@RZ%>0MJP{x5OZ>epD!q&7zKz}V+ri8Nj^-(nqfE1Cp9fXNAq9&H4=C+E)gbx zSB>zaMI@i`5kmx`rce*&PN*?0#t4Ip6;>Bl&)UA(sIAXYL_XZ_c>?ODjzxFh-pkwP&w;dI z;E#>R%nR&&zjjohLUw?$R|vfK|pG2^BQEB7>O?|#|Oz-DJwaWpjXJ;ge zfta1Rfw&1{k8yJxigTM5R>R@iD*H=x%Ja~)Oe!dVQ%ME&4dFN41>ylVUT1(az&9`K zJ^l*va3l+aYv`3-{Ek+1T2WLPBN@8&f{=O)8}I0vjuy@Z-%XDQE(~}VnCd0W-&a)_ zk^K=wVNwohV38d?G-6JMNJ8f0Mz}4~_%~n^5UnuvJ9hioIg4wizqZg)vYpH7Y?JBWJ2~D3EiFhcf>>3AlbTdRbme{(|CnKMumQe!XVmOm%3qROt4X9V_ zR+rmtyk~JT$J@0{$W=lS@(oZQkvIO2*aaE(n3IZB4p}6F1ZYoJpxUEsWuSyhfLuB# zy!nbIQW(Yopz&48PoHh>-JgIca|EO#aIRl!dJ$kK$Q`HaC|vI6@Fq3Y-fvQqR4Gs= zQ0}K`E^ZGI0v9{-kJ(m`J!qJMBC_OVxQ`==QoX|;Bdvg1H_B;e5r7pguE(1{TC@1m zA8pP&jPQ;aHQM;z6m4;x$gYSVv(bYCB2!@^PQc36klyI<<8FkN-#+(*$ZIe6Y181f zW`<~$5!4B{G8A8!l%YoI)YK?#bS!zuGT=ij$6|7c$0a{$%1~BHJXw{&ZpLUIxACR( z?nl7I5mwXz#2fIBBhJB|e);wq@?Mn!)&?G*jVpEjtP9lwCf593@}YhqnQ9F{ z0eGm$g15llezCRZCBw7Z^SjSKre{P1+MhJ^OEpchD|=Wj^7glNM8?OszqDdCD_jj< zhbdesD1?)q=})DNj;=7=8C$%d+6Ly46YcJT26Oz`asJ7&)Uy2b_a4Vi z8%-cyu+)LzH(N3uqpWl!ebMCxiqJ;a>-#h(9@i&9Juf*Q8 zEI{HMv!{G~A-1Z0&MZv91o#S=>!>pbe$(FSU6ioLfc6hitmL`KP$}kH{?87ePlb%~ zFD7w+>5sO*%xoXeM%_3x%-@8Ycfgaq+nfSO1BaH34=L{bD48459L1s1|~;tb^sd7^!7I#fc`hfl>2K|JP&Sop-ormfV+S) zdsHb=>eI>fGbV~OW)2?p%04O-6nJnBhXM~vn7uPsZjyW+b0-P%iv48=LhQ}C^u9Si zfqF4~7u${5^8Um2mHEar7L{48wYrU;yk2mb<*!y{> zZWPdJLxG6osHg>n9}>4Q`arP~V6hs1f?JJ4<88AT^48nf2;R@l&zF_;eiz zjNeE3HyM@B1QAoBW2urt$Q6r&0kAmXJ^rl>&Zr+>cI0inoOS4YU7`Gboz|wYd-ee! zflUl{2I~LTyR~+^2HFY4mmoPD?2zm=hR-IW?RqVJyK2aNkxKp*z(2MBc;t$=;$Xyk z=A+N7^rf$|@XyYddXz_Ppd{COma_kj9J$1(~v3 zS(;6nhn@ZF^)mzAQE$~o4Y@qqApwGRL=b17!KnX4=YX-1DoeK#uA}j?uds$R*gQiJ zpFQl48$>n+-gd^M4YlFUcq#paQa%wZfh@eanGWOOiHwCaXuxniuKE8054&E;v%r4V z6=Y1@RGFtoZXe$2O9u)-LL8~<_u%=g{cUMtubXF-c)k=ZS>`GMW0)J}KhAl-Gm$am zBLD8qU9x8NW<<*CWlxH+qQTu=n6_%|=%zpA(1f%Ewf+K0CgXf;5##gx?x|*QMMpaq zX&f1ZvRu5bR=RGYMz3uK%e`UwON+rGcDJf-)s@EV*YyV~A!&_Y7Rp~w`RfXp)AWNs zhO7`PC+12k9p2}6M7salRc>$JLeLK-rc=W>mrq_Yj@wwWKf~u%=94qf&@gfEmNC;c ze_-N^%6Zi4+sgl@>yfs0gZ7fSkcZ3m=!GZr>65sL3K^4xh&B7>4n{8fz0i2?u?Y$sM`I{q zkX1`)ZV^fm-An4wwF}Vy7k}dujdoF)&NJ*Wo{be6r?ja#$3q>S?_Xic4*}L$gkq;a zy+56I?yqXCipoVbFp`!d*OdxKVzgeC2d-zxHyUHn*j)o>c?zWd^C;c=G2KarbV?4? zR?4Wlv}xuoE*1Dmn^U3l?ZBQ0*!#ali?L>hCfnispS73-XHEFQT5 z)S|?A^nElXM&>l>`f}YEr{`r@l~)}1B$@BGN9=^Vs7W6ZjYyfa61h0^2SyU~%T8A# zwh&Slv$#+4IRK$)1{80-k;9(|-TiV0?x+JPyPtY|K2}Fhz(sbTR5AGniLS@g3Llpt zu>Jr^!TDtazK^gH#`D#FBxI^;`oKUY%Fm|EZ?jeE{Ur{Asmfp*1n{j2CgHaj?(aRo z7AVr_{)oaGfdAS4Qunm3-XNG@YDYK}9K1xh4|Zw1%;^-yE|dMoRV2-e3s1Q6G7Au( zLr)p@x02%7{CVE!HV_{{l{TidqZfT=Q1RlP7ohcxIPDbZkM`!pyMEQmda)G60`B&@o$0(8m5o&zhZcVrUs5y+; z6s*5_H{p1uC?GVrf)zRxDLzePnwgci$#ld5&^N;WEB~$5|0B1Zd*L|bq3d1Q(u2{9 zR;stj9}XygHvx_5TKCAYzWF(ZE7!F7W975xPC@SB=l}Sif?AN=;V=H+ul@JaN`5c0 zRv4fgi^%^ku3P`|MfS`#$%@>BA15;TL<|9MCf=8Ial3#EQ3cF1P#^jKbAQ3FabtXZ z+{`<~Uz_fYxeZy}rco_xV9-egXkx{&J_Sns&*9C%QOdyG5~5%!3(Y<|O-{AxN2JFX z$*{m-R9|Z2J)v`<)!N?M(j1&rs2Tz9|qM%LC)0Z?>qxb^(hugcOT@ zPOrQYkvy|3uxcOH-Sf$Zk!B1ObS!R8wAA0pl2F1-91=O@nQLIPv%Y7gDJ@d9Wk(l2 zFa3h1Oj2%%x_sj7t}j#0eyvbp$?uGqOkfLn(S|P)--L--HVbXE%DVJ<=Uch(sXYfm zDGarC`urwYzl*Mu;pnd4XZw^j5}tsv^nC5+Mlukm%G=@`4gm~(1W+~9%FC6y+>gOg z*|Ghf%`gcm!!~Cw7M*FNbG{pJ%PK&989jhsdJnh6esBa;m)6ejc_HmkB5-gLKcSd> z8SIuQV8b2zQ@2$P|a}l zE8Vhhblrd7XH)^k1|~+^(ykzQ;I9Qx z$(FiVj~QD>jo%;l?;Yqs`lpks_UZkCB)~m@`3mt&`%(TQ4j&t2z$KI}zz9Hm%}J71 zK2a#fGd~O3wqz8lgQ#=1b}&k+e29LWlB4Lqx?kCnglI%RjjAC=D~zqv_8%WU*+Dez z0w*X*{CFBuqML75Y1P0KyysG56bj!$n9JIl%EU}%40Fm;&`pciLUumQHh#r7k^JI4 zP)9Ds8x%F{n4^XweW=>i6$6AD7nwenkliEVUTnmpmCy}h({$*Kl+ZCOk*kYG3gpE5 zl)5lD^Qc%$=#5AqtH32;4s(Jswqy_!1lI zXF%ood_5ZbW$Huw_LJ80@zYd#Z`)LEujBKI(CYQD*0X)Kf;3O%P!}~4Ah`i2@b}cn z?D;w~jWT0*y)kBW_iml5(!-)^Ebu>FV@i2bBRzql>?qe=KRLmQ`)q%jy2TVXpn z%TpG8Wv>YPuBCYGbYC}R5R&Tm=Q}!Ag1p!+@s{q}FgK~iP&On7(^GB-sd`7__d?qL zn@Lw|=X3l0)-vOGKFc75EbX<57m>DxPh~RT@t|WJF#eNJ`)}DL7O4h{&+E2RgkOvJ zbO&tI8@?o(Lju^^2;}~kZqO`sPyN=wAxdz6Xx>Rbf_qg_McoP$=wBjgmr6eT5KuAwQh-aln<$5nRd@C?yx8oF-w2TEYyXbQkm(2Bnu!la4X z#lAZ-iK<6swH12+pRtzDZAsjrn2@BTkA#59>scFUpH{UC=(Hc4z(IEU#$>sA(|)j#Y|d>@5srAOKg>n3J_cuIW}Wd^Y~{h89iCQ#zEiFnr%!89MYS{I4`KtYr1q37A<`HHA_9{u`xN2kaQK7pRd|pgb46#* zG~MbO7ms(27VzoOE=MKVQ*sCR64Z5ayX%`5X zTT%xNrB38Cfj+3KAwu-ciU5Nh-_*WqiKw~A!IATf@Uyvxni zdA@p@Un9nx06ijJ2tKi)Jf6z70G23>D7Hcy!q?qC5lJ4EI>(O@O_87}FRD}Swg-ho zK3`E6-V1-ENE}H>#BYa(j%(f7hO7lGcPae$wD=FzA^m~c{s*h7#E<$k{^ghT4h3^+ zvSC<2r7|x5-=P0_N=w3#$mGB$eg(~~FnDe*jC)NLu-w9bVU1CfqY3VE?=4e480R{a z3hrmd>HFSra73p9k`#_U*p|n`R?L(Rp@Y%bB=u{L+GM@BVN}4?Gtd3sc!>GKs5AVE zMLKCs9=S4ZUNsAY3Yh``s8zuo`djq>NfW==7)et$@$f^YX9PFrw&=5K0tkT9p3tZC z|MI~>d~UCObnFw|)I?Pi7xbM+P3pJzp5pYUI{wT6V;8*gJec1r1(seqw`6u%m)(;= z@7_;Hm+#=RJxS63ODS*~uE9dNk}b%$S+4dR7QDyiQ1>{>E=2v`y>BUnI1Me3;=arD zpmtm7`;g)C+L@8V?T?k`hg);N&YYOqTSLv@n!Y^Rlwv0k5MuUau=&{Wbh&W&mR6@K&T5g$=}Fm#I17)o)(7l(KX&JE3e0S=PcRO7;t&Te;rPFOmcnaKF}KG+RakO zm*t+nb(G&54J@Xj85*T9nOs#}?;E@(kuXhc8_nHI>jO(uDg*#}%emM7BBSMa9`vA( zo}I9sZKyJB)c2R{DGYW{e|3GiPn(-hH8VufoIuG;aq3fbvbG)@Bg*Pk*4g!`Nf5v! zEIwr8+oF>^kxp?1#g1=Z-Rp|6ts#O;%5JN zk8m{{hE{$pw0dp3%t->wh{=<>xmE&Ban(T*rqQ%C@U4KHFjR=Zy7IuHMB$j^ZFX6>EY;W$kYbH; zq37l?>2qZFBR1J`hm4a+(v1TsAl;q0`YVTP`dIWI9d*!|7c^1QLf;Qas7XOzgCZdA zS3`NIFVbAyMhe;6uw7qW=qf%~&pF>75GbJD$M8L)3NwLWiB1^qOvq*yxq?W2tA4LU zYL=%Er9PA5;3v6sD>->wQi={iv7)PgB6hDCIy=rFv#q75tc*cl&?(V5*H~qu>YFiU zqIeAY;RqcnBuXccY!g;JTan=lPK6;aqR=R0bk3bY5=f*Z=}bm|kGuTnix}%S5YHvD zsXXC;4bk36J;*3tCyA7C?N;wsZL&1-Jn4wSyIYk|_bKPt%qeHciRB({LL1$ebnF{u zABRj&=SmsWpzz_He2%5CROe@#HFJeRwadhJ|>8u%Qv2mQeC4|5A(mW_UT|Vjt`EYp;2w z>vs-OTwn5C@m52ntvob7$-s*VZtdpNkVz)9c{;iN&`enH948uJMJfy0jP1=&SBg#~ zj>9$acH#8k397!I`TuBl>!3KFc3Tv8x8SaU;2LDG;Fb^|kl+^FT?Qw(TaY0kgdjme zaCevB&LF{E2N>o({Jy)-{?4g7yY8)h>Q+s?^G{b-cR#Dwv!3qNH4+`BwG;lI6O?gR zo+r$N=#5Fam1oQ}v~PIJ+y+xm?Egh#?6oLw#r>a|gZn?+4zZYm&&dTna4+K2$WRKy zWSdjD@pX;=H3HE-=XZL~vRK*wezO2UK)%1wWr$an)>02!ZAAM0LWTPnHInC_a0SQv z$>Um{_3O(ip`vsC>k2lGk5lbu{KOcaI{L4Rv{-US@~`-wWY9tWOk5->3}T@})JWTZ zMLI4htW+b=P1WL!R@r};Q{-h|+pAfq1`{V&;uY-Ky&yXJ9?D$rHU(lw3QMIh+eGg` z|3`4tCIs}5e;>N%aZYooRnwN~92N2tjqBRB9eq>@%pD_q*Grj08kfoeSZ8Fscq(X~ z*N9BMRvA;T5fMcdoD%eKca)1@e}=TTPaAfKs{7Pqe_K+3f*@|RmY5)vw`0G#Quv}mP(T?SsrCgf-+w{hgJ8a+Kec(%`>+f-3bd&M; z&41hdyad=k;|0qRK*A&lj~i;|Nb&6?v5=Bas9*g5%ic`YJYP%w%x!LWEy1yoRzC1X zy$oV2teRi{JaMD6f`r9S3?v#XMn4wg@HzPiC0VatxPg$^VKT}8Rj$m+S_7%d)j6#H z)pqbtWL0k6p5PIP;QU{aC}+REdRMn_v9-|#k&&{smGRrM@y(0CLvr_J4QxR9e+4fI z5>F!`=ma@K?LT9pQopfee`k0bABTdLMwY|d?_kWU3CS)(K`Niaee++$1m@#9e{N{- z{?@-h)xd2fdqpo8=u-Vp`}kgizR`%opOqo${jpy|x()j4+IX61NboRv+5d~6R8yR< z&|ch>`!izajIb#?=inQjrL*(|-}b|Jq9ZZ`hE06Wv!D-V)A1fZBrQ%hz9g#BzU>5W__( z`o_k6f}CyfpK%%N#TAtNrcF!wYr@WDB!+wtMrnbhMgQgbxk>wo!Ab36Z^mWvTw zCP`||3zU9dqaI>S-%lUEUO8JN;vm(W;PL+VBG`CUTIp3%##=>tN4@G}JdMUTX@cwBg|ndBssT z8{5~?lA_H|^0FXBSoa9o(>JfYG{@}41?OMjbKRk0L7#i1Rm*w#Y0XP68Mm(;yh!b3 zc)~}>od40zy;^FPqoWe~X-AN?Sf16we{;alM{NY>5UJa=Dzg%Ki)xBohR9;9_E8i}EE3TL+k_hZLp#|buV3IIVFsem9b$w`7HU6LXbwo}02Bp^k{n9b ze>Mybj%@Ucpf>D8!l}o8W8f#3lEF`wH@DuZOm95a?V-Ps7oTzg5qJYmS=1HAM&-}; zLK>g*)FEFb-qv!Z{lee)G%}ePT0($3W;t0=_x7s_{|X!toY+5U zhYP*zan-6?Cl*^#v7BMA-X4(KH`#fllb4*{JNFG(QJ zK0jojCwXW8uP1eW)_r&hye9#hrZ0oRUlv=w_-14{O#RWq{kKP;hfQu=tn2GEyt0>k zyLWNT?~t{)5*f+fiOc%S|M`*}YRKpQr~Xv+27!*OJPR`5Z>0Zh^AYQHrtat5bfff) z>^x(s+(dksrygaF~T*bsrF)6O=Fx_|3A)K z&g8J%kdbgD@J}|eL!j-bzIpHU8XBF$2weU{(Z04YQ42m%bmH<}?+ptZZM}*a?oimTf;P#?YW4Q(5e^{Hwleo5U*RpVPv;UudaFh=pi~gD@@nVYmTp08Z4&h3$ahzJm zCL17yjgoOkj<%BU616aPEU0x!fI2wYg!pP$y*vawN>{8bt!su{G8#D^xJsRG-e%h7 z(RTKTHoZ{|@Q57_=0bW^^uqHH_o=9(i~mB<%}*K8<68JS8yAac1P|K+TdvR*3(eVK z4>ZSrFXr^@SEoYqkt!C-t25uV=!CR&AD{J+L;v?{eYLtt;j0vybJUJ(hQ9qNgPpr* z2Gl)sES7Pae?8`EKPC#vy$iERn&JwQeIEJjL0(*$w=c%pP{$bEdCCF8X01X}oS~gs zTEvX|#R{)D2&;#;zbu##vuI9H&En4jP&vjRS{y|keANPhVN!6*1pRO3^7)}IqldPJ znqAm<$GzX>W+@>Ev5k!M!w$b=19kk5<5SWZze=~0|K0rWJz3+N6KdjbL#?2$p%NRA zPBx(&ncuNX1@|mN`0Kk`3|b;BWReyBUb=gN%2Auohqv?dH*Mp1{EBul_!*fWXWrGY zjKxq}RxGrTaCw@&6mHz2S56}WO0 z{~+KPH{+E$wd|OBGIF zIM(A*H$``n4f~xE*lv;HrUJ*@(~w65MV_ax4? zMCa2W(1GgZzG>WY&(Q_GGN=RC6>5Gw8kTmE0jyKdB#&fozmENC4zGG)kJ9X2{#irA zpI)NC8MJutVA8PV{Au~n#Aj{F&GMnI((|?eQ>nw(*A}g_irV!c=vH&$vB&wH+{|k4 zTJ%7>xGiB}q-EMvnu(NsJUAZzcxE?vxXRWh&RJVI^`GLT#tz#vFK#5ddRWAEmOzr=K<}HlfLG6K5`u!=YB0W5M{YVh z9#uE3AJB$iV)nQC(~_pEo*vBrZfoCu&ksx5wgtHbc>s0jSK}@wef^@dp=y_!)*fy2 zg^Er7L1p_LPE2wcT9W?p=ziXj{ab6j#~T=*O+6M`hO??eJLjmN!@v9$Tuxw-%YCP- zx$~uZMM&R`LVb+tv<~&u5Ab=v02?76S z=S)TVSLUYv3PHyM11Li4hG8;PP#;Y$A&$1oX`$z?^p7@FH-+eWlBd);ZCL?2E4fGa zZm+>y(tF+;^(PNX9PTALua2oRmN~hM!pJZu#kBsEf6pwNUL#nO?qD?xk*QKOS{WO2 zqCIyQbfKrGhU9HCwJX&Lpr1(Coj2?`$ePTx7&LfqCE>6Rss!m_c5jGX8?0x;mf-$s zD9cKuzZ@}sXKOi~K3sQZdOQ~uKs}p7etU1N8m_wD+NhGWm-{%DTt9NS#Vv~L^e|ib z-Og^YxU8#rN2sE+;i_7TW|^mIEZ@Rm_vb1BnkZRPN~*4nRgyI7#2aMo5&Q8!#n?X$pSCUDF*5}~Ek`g7DizSfI%uqG?u-m-_TN@p5+R?oi+?<_`gE?np z#PUs8iaQIU@?#NFto)C4il!zsdHMJEYhMuNSAS~-g?{E+zd-cQ1vQT~T#$)0qS^Wl z+1i*r&=#x_Z%cSP!(^^SKfr3;bJ4wLuHqbSWIm#3U1a>lk1B)FRC*H*4?R3TCSM(y zv$Qc1X}!RiPtKdVaKf&?eyOu+7a$gx7Xm+ETwfPbLFKKbJ`VIX+m^O3m6v$w-p%d! zyks>g3GP!qN@(6m@ zU-xY2s5gYW&>t0Ny1z^(BjpN}Xk^gx>sDChx>$|l=3edhozm%OYrA#k4ov9XD?(6X z^}v0pq^v& zIL>n@enu-IoPH{d1%2?gS^Asb?t)(@%$Rlh0iBgx5ppN@o_x+XqO}&)DR15sH0Jhk zP)Ms~wji@`)td&V<>bhx5KS9zKwgvEJ7C(9&};l|l_Y?UmfR9U6oI77&c4PY)bVf< zOq6Pxe1h@vR9BaW?mdPF>Y_-#{ru&shd{ZVZuE!u$HbW%TDJ=e&Q2qdZUx7NlyHuJoOGjyTw1)Wb zRoa^cbIK}Nj1Ji)CW>Ew@C20 zG5RHi3-a0Cxb%_WgVaQ+oMgk!H-A3f7gL>)bZG}2&Gr2Wwz2bKVqWjo*Gx2Qnw#pv zsM%yzw8u79#qZ8z`0GS`ZISbG4N7||p@Tjo6?QUJ-k%&^YWD7IFgfR_CTZLEN_Moi zxA&Y(Zb@8?Paj73?d)3p*g&ath(t$E9qzYozfA%|%MQGJ5mjR)hRc^|=hGGLUePP{ zuU>v%*5+Nc>MAs)2-a@M!fY@!+MV~HV6?;)p?|pW-tIP-XH6a(Zsfh#S0OF&>-}n; zFXr>EMA+|ETH%D&FwfNhWpzf-SzYO7r92AXIEj~P_F<9}?(VFvd@Fqb`G`)eK5zx1 zV*k6e#dCJJZ)a+7T#pkldFQu|Xzgv=l?*BNie8hhm%1)STe7p3*4Lo)Cas!2hQ-FP zoE%TS3%OM5c%4L$kl=;12-Y|nS{J6)o=gn#Mog5g&O@ax5AK^GS^hVD5_@X}Uv9N3 zWIi+XA*6qPm=@_udt9$U$+zv*VRY8OgPy{NQR_{J z&OOEkvGyl$IysF-W?jO{!{&vc$jWXC`C4G6LJP}+dc#=~A?r#Ur7u;09Y*KIId)PN zy;GhB*m@bUV7|>@{g<~J6i|vf90sPPlNFaD0OPwi#PIbp2!sP8kxvFU` z2fy61vW_1-yS_aZyJ{3bDwLY;tA*@>AErOxPX1vnToTyp(%fdiu4GJYwL&zlMj#Fb zxOzb{7S)Um{D5NxQnRF|&ts1y_9{!TUN~t|g}IM6ez2zxS?4?ah}-t8ncIbL4KprY zk6s-)Mg}#cMHr(+Cmn2vu5sy@HJvlU-~@_zr5&jF`^yj8b zz#dQ_-&3NaOla(}x3zb4?nd&v_%GcWKLEWb=uYm1dtg#=U0uS90y!a#=`Ir*fD-=u z4|^in;57m?SRxntU5Wo~+aQU`WG|@ajhL}DI}-xi#RjgUevlBW{jee0$=*-D4N=%4 z;YM01VKAUcd40Ko*swqA%8ZqfI+-2IcY^O%39hlRTh0B6%b}E<&&m0FkQ5y5L3EAe zaeCy>S5@(`0$IA@?MP&u@?YUk1Lx~w)K?Pt>nrP$SHDmn4X=oSnQw|>$mFm|-a(N1 z;a1YG{rooXq;a1yUzSk5aGl}lOWZ@~FnbBp&v;edv%wLy4TEc*!I2@2zrp8XE-bp!WrW}A-*Vg67Fq-O;QMvtm+UHacytb1;*DCZM`i1IfI%T?aF-!y~^pRe*K z!!YG;6%dQ6AT4u5(?^WDRFUuTW99S*<)0>zx`K|&PN^KK)uRXbcDo(_M?0) znMr?{7ia;q22)@3T>Dpu@ck1YqZ`N7r3voX^d~pDA66tO-~0i43a?jK$g1&Z%!@}a ziBOez+8syCH&jzq5&6mjh){V?TpjWYUzTF{`~bSNx*w%yycSi(n?tOLpt{ z&A(+2w$88t#(la^(fL8z_lD0&Bb>cXNB<-xZ9O*mtY74ZxFOzw;HzhGQ(ul?#Rd1HuI+? z#zrG%?&0K>PaMO+`3mit_hI8HAD;k0gpy?e34qrZZGQHnA8v`gR<%EGv;%P>@nyN; z2jB%0XP@BH&MoVlJsJc8%J5W(NuQG;k94`;7}73B$w%_P`q_sIw0D@G$yD$rL*U0v zNVuhIGvQY;JiYhx1m0cwLmv=s2O|y+xf4CkQIc`Ss8y%(2yT+dgo`KeM+#@l@$ZlKi}qo?beyIM+dZkr8^>$(+sLUWrTq{bMzAK^jaYA zM6i9>H|C(NFI_FV+oT`;$klBS=VYW!Sx^Wi9RG4+`Q3-x^-FEodPnJ(&y(N^;p+Ii zfqDw_3I91T4^coi_}=kVgbm`(S}IXs2NI-LrG6%PK0$$|D75DmbIovq7zLpoF$GE< zf-;>ML1vG(EexFSX=zm%>SSs@n6iGB=f?obI}Vkyxzphd>51`)zFlIt?f|uC1t1Lc zAa~z(us@cxVGfRh9FD?0U)}BArY81+w9nh_O(iLE?hmLfY(4xQVK zRY@wMmd3|2hiihj>X@j%*7L)W*A$qQw0vskrL<)?;~bcXG});Dp?E(*{tQ0e3(5%K z?G(6Cm^U5-h4W4C;-C7(!Ibgu3Ga5F;roJGHSZ^k98HY_=jo&gmx{VI1egogV!K?G*sDB0^!bw^;v(e=;6iT=k= zh(_Ww8$?sX82wdUHdtRs5W|K9h-Z3?M(VxeI!_|sOSqqCd?x+}#$JOUtqisA1vTI{ zcQAu#&&D>>nw6;o-Uyxb~5cqo5JC z#sndXOKl}jQpahaSUf^sm&e!aRW_RZpnYuk;lrCN zQu3(nP1ee{F|+_)02L#EW^LQ*iu&_*eGrsz&wss;im;Rd@w1y`FnjdH+PcjHlwgv zGqZYeIfj0M2_p zWVTC@ZF3Sq_^xRT=a%Z)?VNMilG;cv2Iysk1v#i3q z5?=4UxE3Reb;t20nT|%J8fE8SE`xfmeVOj73_g|$xqbxVK12X5z9t^pCPbNIXWM2{Z1j2c zr~?m-V@4}jnc;coaF16Qlp#Qj%AhJlo_oU5&;??al_{OH`>n(8_XKgC*0yyCo7B=+ znWu?!WxxJm#C(nGM}bu(pV&&4WpCO^l&b07(O=COW|h=on!`T+YxZ3}zWI0gCcjPI zl%LAM!}yjfKZ+lm2N(ED$>SvZY<-26ND$Pief0D4fxJQP@_TP?;03AWo+|I*Z6!Jc z1z=7_Mj$*U>3^#RC2D8{28Im0Qwa_&H6~>BWdTNOTw>R0d`{yWu0aiDwvSn`ps=x{ zfa1HA4^_y=2PXSHrzb~4Ej1~Ggz>S`(Jcb>7M=GKh9V!O>;?fv3FXnG)1uzqYqfomEs2kf!H#1ukfgaSyh@2qjL02gcLC4A7 z^r~msSCFM37Pt5h3*-0NmE__Az&isgPMHCCc-$`AoKrUV^JZZeq8naYEB}%pA-Kd4 z@C$M1tIL;5NMm5Ir1$W9xLRpdn}d$M=X-GoIv=!5gpS=$Fys`##!ue?=c5Bn+&97( z6ERX09p2F*Y>xwYd42=dy2T?KcO3=&V$~?SAk1XMG_LKY{PDykEjZ7pS&D@ETsvz} zOA0&RuouqK<60dIH?O)y*?Y20J?* zG?Qf2C|DLS0-o9^XW)m6@jt~a43?+-T!-zWxAW5r;be#!Wp+cEL|91&>flziiRr_Z zun^%SK}%iJ){e6WN{uc*P_gZ_Ti!B|0;b>c8uqKip?o;vF~PjEg_j?eC%~td1uVRh z!57i@+MzU7>>!Db-*(_5u^boERmrCMYFWXH2?iVLJ*G|5#D$_9@WE{l#l7}<)uF?T zne)VX6+{@VPq44UsHzg_f)4wJAgRgI>_M8=A{HSO;gy@8^a;$vBo*VPaF#?oE`ji( zH28DZcS-2YOALISr1s){(bqwcB=)OX;&MU=yG%k}{6c%?Ui8wnBoSWgqIgtEbU;>O zL;liR+zqc~UURK%R}c`V&i+txru~=xGT%(l;%Q6eUG)U4Fd|AvT)XeJ|6UYs?L~?m zg1i#2q?q>a!rxa_qH5HmR4u#FiNSw0#y;+DnPM@L0nN1|Q?i?}h;Yjue6q&@I)f@9 zq2U=SlD7EpXHH_*Oo$Y-?PxfQ_gFtHi(uDg2cnD^VmKp|Cpn);NLa{eG{4b>bBj%@ zK<%ndkLao6)M+bC3`B2D15oY;$gXN?H0ugelU0+!>{*-!`979|pnPxG47#5vDe=Vw z{4`k9o|2fwwBrFWEQtN}NY?c4Qd$3ZE*JXrupehOU|0CzPSl)wIk+3-OScI@-d#6U z?(H~xi6vjhz<`>T-Wzekco;nYkydsC65NWZ3&S)+FCy;i2sYvDeYFJKv;IO9MZKUG zk5*(5&{S2ET6#AJ`TV^e{S~l3G13BMSSbT-<2G5RAS|6=Ix-bNYtJF{KraM8JvHx$ zlAxpX+Y~;^$3L8g^miNw*wMLYfC_);?z1^C`>ar)Kp||txT=8+QgUec3nITNP_EVv|*S%aa#v6Ab zpPqHFirJP!3+8ReUz%qY(Nto4o81h}E}P%w5q^%yq|p9*Q;N@b7-5rO3s>5ncJo_r zs`_)TJ91RKnxCp$$!gTlLTrsVT_-N=_8Y;AB(85{3%khtfcSQ8??D{g62`sBz%@n; zfT%Ro^ktXs^fBUMD@pO*Y;Np(Vb%-pfi58S?{;KL(*o@T+6_7l^5sm{%p%C2B$C`| z(GnJ?0Fzyvd{ zORIz~pEFFuw@j&J5sQ~sM@Q~bd{P~aKgx{uK-|Sh1p&!miqxfW`1Ew)4IbF@tIMm_ z!p^KYBl1xiPhrIpgEuTW-QEDFMl(=Y%Qb=Y&$*0Zc&a*MNMU z!HGU^*9~qKU;$;>j|{<8guMDI@zA`IE?6U1E5p3&_V{gmp*2E6Aaveun<4yDrtOi z@GsMQ6-71$#`w_1E(st5oa?0w=h_+NNJT$!nDXv8cm^stfk9cgZqTM{h;ux%JRmSW z^t1k9*5A`rnD^sgq&$%Gm+&HHQa(dZGmFgMzvM;|OOG35?!`u%OZ(?RkxrBb60 z>Kxd$UmbH{76RL4h+cRt-MF=|T#;o9{x$=g2!FXi<=M7h3x$ z8v|xL`(8Qu1Ar5VO34Y^J`T+B%PCePwqPvsQo8{nIaCAsFAb~-{CriK~|?7nBE+JQtV9CZ=_Utioa z&w(lUdlctD6TsJpVF66E2|%X7t~e}@wxjCJIcP6-o?2y6d*j70J=%&9u*8C2DJRy{ zqtgv^RE+q&fj{HrtrWU)yJ8KYffWS{d;^YZqJCt50Ma>ZOC-V)w@h0klaDU>5Reo% zu0yR_U|{#yXV<2*E9)9^m-uZN>(Q2`O0NRID<`~oF7mA!;P_A;YF`2TRucVF^5{x) zS5hpUNK6bg_vY4k;M%{2;{wWXO^icO78-zN(;nBw^i>(brME1N6pb;^@#{va=yj_# z>p;g0QS^w^rS`FRC<6?v0#IcfW_^;zyoXst$Bcdr0gne^{}A-13=kEH-6$8LGy=j415nBqSvV<4_V=jA3c#ww zO#Lnuk(Qh9xDENdvtXsm^ITLgw1#nl)3|89X5Yn)#^p(x+nQ8epko`QA!=)`nU8Le zp9)e+BS(1L6{McqSb3vAX=-^d zSB>#Ytyz%>co#kj*R5$sO-9uIP&(CIu(_DkbCyG=kSN7_Q{ZvN?&JM~Fxp7Qo^$Ku z@u}Kyj!v~mgnID|0UkEgCIk78ATO1VS^+@;0jFt2esONBpRO4~)aUvRK<;uf`=Fg| z;B^Y?)~j>A2c^*wp+!LVS?nGtL%7iRNlt2WSMQJBPiU@m94u9xV{!2hnOEh10*gyCm&jG);FPCsL>EZt%Em1CfgR zoGnvq>SVfyXJ}7FLw(fUBd!8&GFT(;onhU5<%{!)bnqu|3SfKKn2^Ov1fO=6@Gcg9 z1z=k^k{}>BNDZtyaM1hNe-v${3^jk%hB5*Y$4WsA^cLWl8^^PX+iD&}VWq&_oblz)2Lt&fO43|rzAR*W08|UuxDt^a=D_Bbl zrU`@w1p_g}eE|4r+$Y(p)W6(c^m+x>8wgD?SkaBprU4%_@kZEI{38lT1wc2kDuVXg3*cPY_}?$T4DQds2sq#+-9f zt~T9o5d#e+&98}q(F~)#HNkN+E7NOh(KfzwWfKj>Zm~i9jefH^)ajy$*FjjI{r8Rc zYohGraV4|NDYLg#s4`Q*>Ll_6aS3SxM6B`r3(pRQc_QNLk@EVVAtoQqT<_i<$@3zp z{Qh#l6l(}ex_fyo!M5Dnn|!68~Gt^T-Tx| zMv4L%RR@H2Bc|7uOxrZCune{22RRs%Qwp(|e8`!;SxJ~jv`&N;ZiVt_F=D}aYlv++ z4S$3LeZdZwwPg(p;=@KuZS-;aSUaVbLPdn;PIECRHOlBZ3Sq-XLl$Wq^~xefqbNs} ziVH!a3MEFv@$KoTC?)zh8{JcZE8Xk0fR&bIhrjK8w>mTRrB9F+Np}`1vsk}1O(CRS zLv&H`_rn7_Pk+79Sx}3}i{ZF6e@Ayel+gQDk`4=^LRlgmMl3nA1x;*yKdsUb#_GD{ zx|B*hlwl$^qIZi^^UYc8AbM1WTP3AP0c><5_U3H4`*Ue<;Qo1^TsfhS+c|% z$~;glr4ozN1?Vt1RPX1@g%$=76tb~ssF+6$U@B4GX9Fxmojx{1j4b>om!;Z%X+)a~ zJlP5{%tXROoyhZVnIeAIgt-cK&zyiG9!7A?BTD($^)3%cRj_jU6d6(ERM31_bx`Ca zXBpoMV&PfXDTahFC$JJ@&BREL(=wyTHI{Z;Yha^o5Lq%=O|@rLjLzV&GASDNpZp@) z>?c-4M>|taEJO<%9AS#**VjmnC043nS#yjqTqiTAS;50WGmH++OG}Aoocx_UMNGz0 ztz?N#jEAz`%uM%O32JS5E=G%GKZX4|CNdeF@9e!lWJ{LFIj>;dhN(BdTP7~>WhHQ@ zjTxX5&#(JIj#yqF%A0^9XT8$g*0nm4plO`MyLhLwDLeUwDPl!pSP5$IoQ6XFyP=T+ z8fI9#(*cqoW>{2#2j&aRu;C?4=7iUbdyU+$hFiIaw#5#ZWC$_CERq~Lqfq5`ccEV5 zM87LbSq5pbsQstX(D~>v!(36H^_ibA5}-+hiJFFy;h;rEk~0D!KcD2qn!!`0Dya7A zr6`J+80w4Mv6oE1(W)<%8PeC|c1;6==6J9@=0g%0$LTfE-QMDlM;?PUc}1}O7eqxmYi&P3HQF7eA6mes)>EzjV^#2Gi>jMa8*T>zR}k(Z9A&FrUc}Mnqr!ylcHk$lqMHBhLHJ&dZTc@3+>IKL4nrz zwjggerW|XOY0?wQO*Xe%g6n}&oMdq7PbaH2!$u7sL_L+sBHIG|@N}C>97Kz|#|g;| zHy9T@MZ17xxLq?N3_sM~nw-du(8M>|F-Ih;V&EQ*;?Vb~Z)Gf3dgj0Dj#<#0-GYr? z9ek5S+u2P|IZY#)s*ai-H~AJEF@vNPG1Y0l1A%H3pj?BsMT>Rtu$iLF`oB`Q`7kLY z6K%R+&@wu-O?i-^4|-q89N6whl+PoGlZJ=MYxv4gJjOa7tNDI()$e~Gk>?dKq*D#} zDJ*YQv}q@)tn@>P@$mk(h|Y7jSdVOZ9-_b-hS?h(IIoxOS^q5&<*W9 z`}UFfttbl~OIgL{zlXdbSCW1+R`~VcLNFlWesZLAhvVlTU0eUQyp*WZ= zKkw<9K6F$GTJz224db4e(6lzHoM}9NAB}nJG`RX$1+epQWeu6&P@*KpQu~ha24?&G zA7BmU@-dnIPpof><)p&P*{lRwq%nreoGbJJoUy&a>aEhKS=`gt{SP! z|K#tUk&CcFG|~6hv1;Ur6CZGkklYQ#=-fYueFA?@Exwhw)P9D>GzS(HHeW1x+U`xS z(88XFoKS2kJtVZ9^sX;PU52G%T{n<(h8Ai{VaCW3jmtT{iEqx*P@;ryDzRvEzR$Uf zj_1#s?FA9OiB|e)y~vS3`lUj40A-O}D$)m=)<$@}==lr0tqtzzTGe8hk3%)&Xn)des|R2#aD=W^{IC{0Qj8wvau;O64MM zy&2>|`;`;VItaoo^Zwp~Qt>e!o+gbTzxj1nhYmRY39*Qv8e~j=3aRfh*?2Q`zVefZ zfI->gNxCFYZ}7UD`(vPY%0g`74as*L8KtUdVXnrOEf!H&XfcYeMa1lz*uerRShMc+ z)a>7^gQS1VLe_t22sU1CGN`cM*$ie_LwkhuA7-6Nxr&zu9E^oZg=L;`gozy-#^ScC zL^OvV2A-;PT;&P6_{MGU#CN8~xHvB^1(`{i=8%4wBFan#2WP2mbw|Jz?zKTU*v3|m zyU6uKRH2u{CFhhL1f|I`@yxoxcM|v;$lUVjSZwk9yzQ<=sN-L})B4*QTZtY}(6oFt zmCUpY6@RDzqXSY4ygW`CmrKeqI#y+GNV;ky)ESq>J)U;VXSZ=zMyxC!H8jFeA@Bgn zq2Mjm>GLhq>z5>Xj|ro4_dP-f22R@T%dJ&v25Iw5&c36L)z>@yri1bGt~C3g{G%>A zNe+3o3tHj5i&!JJ8f`ktyS+j#m~IXw7h>a_`0v?zn1{!d`pbC*+UTzFxs!8E71Jdy-@ZUPy6NgP_KHIyGjNQT0A7UE({AhZn{iqr<3o zyX+>>3H*PeVJ{6`Af_g&GzMB*_|?v8PqCv#l*lIE*x}-!9q$f(x{D_5GCU?$qWnh3 zx))VM4>4S8IuL0MC{T^$bu$5r3{WNCm8`lK#ETQnobrq-<@KXDqS04i^kId=(`1Ri z#Ka^`R36Ff)S&rUMW~+dXw=+EFJQ&fxakMjzhUGgcR6%^qAr5*!Y$d!x86x^j6axh zqt%!j(Y)h_9MQZx_2Uf`7_`u|{Z+m&rC^N*cnhLHu6Bo1RM9X8=b;Lc$d}shk3t*m zycrR9CnZ$dW8^VVXV_~rpu%k7HA$D8NvO-~kCl|>$yV%)z!L$yw~MWzc!niEN)!Xl zgpNr>y%aY=7z@x*My=`C+rME*{#N)}Pk+JNSZkW= z9AQ9Z)bJBOmWJN(I-*sEI%M*|Y|B(hC*7IfKCAy$P5j+MAO_8wy5&lBJTSZy2du^W;PhIBphhJFIH$%g#+29>0YLgH!^D7EV zv>Q$tMT$XVF4~|-MBSxCIl`Ea7v109>@XAqWm1u>0N|>nqfcL{gw;ehBs8yQ$YQ`j zR%jS$58YTjOQD%trr(>x2SG(5)$yCA%BZPs`4s1TBGH_R@VIUl#|v^L%1lp1=0-LS zd9P}WRbnkP&@s}6BbruYf4=%H<~gBzcO+pv{|Zho=q3*fq5Opb+oCX1QM>LFBaE;y_I2|hz%5%lA4e6Wpnl)FRkk%*agL`jU+YTEdYh)w!bDHW16AwN zGzC%XL+0EsuF_$RHqeni{^~v{$45UR(B?ZAtQ8yS4uvL^B>`TWWZ)By2zTtEQxfR@ zbXqjQmSO6oJF9josMTzxCiV)^g-H?vy*&>d#7euk64uq#GJ@mRC9&*-T#7lviP=N% z%b^S>4QwN%L(n$0QYPj?makr~hG^fk2b%~Zu3t05L==#Q9gHuzp4*!Aq=0#l%SJ-r zwQ1$qD$$Q5>d6R}uCGxdW*k6@3-2MQ>XaU&SabPhdT}IKUHFz#I1HKhM6_volZg>Z01+QJRe ztFPyA+FcrYD?bsxdjf~;JJ&)**KG)A$YQ+oAgElr0F3DrED`arbO_fL8td?oQ+-&+OdKyQ~SyDcarq- zO4&#Pb317W)beETkq4qAMhD4zJPiEy=TrHX340QmQ=ggP;g9-c-CZ$JyF2?8fbbgy zrU|NVK=#Ym@bX9uqZ-6?Fq;MGA*dz5E!zCCM2RK&HnttuqP(9Tq`4T%1p8FU} z9qMk-F`MK|YVo4`N13or%8|fN&4O3kNE?+QT`TQ&;8rK#|;m&vp9hIFT zv#(!r4K6<8V~H;!M`3@f0{C6fGhN4zlGIrl);uQx|fxsB0TC4wbpaD-uC|JNba3$QGTuAE2Eyf{{40UT7hPlEM00Kzt{ zo_r!ww|S6CMCv6Q{~2NSyTwov;9wJ_d1qP`C>5o67JLJ&P!8GzaNGd!kd{coD~20D z>fI-4l{4)+nuTzCf*r*(f_PzO;uR84hTe{+p`R+)H~vZ?{Q|FUp}u0s%!ouNcz(fP zBepNaxEci=2LHz3+ksRb+7YtD23YpCU|C&&?kurq9@eaRDaW2TF?^PHNe>7GB#nf1cwr=m2Kr??AiIM#Ic%Bw{>I0y7P!kdn~TLtgpG zbTvub4TAi!)shjlww+Y&bhU`uAJ4UkjsULCSrnQdA{R-Ap|IswxX8~DkeCHoEi1GF zy&3|s$&3tVbN#nRJ2#m3glu7;j@C)yyeF;i`B68p$8m?jWL3|QIublA2x!#(Rz}?2 z+x?LbIr#=ib?`GVT(=QZGAW;F`%S*KC%H_Te<{LwUL~sHu?;L{u6YS105?F$zc$}0 znFZ?i0jT^qU{*M&>Zh3uP3Wc2kz8)K5_D-Ma*ZZbZZ`UgK`X)?>BS+)fyOVe4A{0A z?4t~I8yzG#0v7G@FI9mUiT^>$R|drud|w7j7!raE4uN1ZcnB6GFhH=t;0}WX87v7J z5+n=`1HpB$L4!NNA!u;-AOV8A2A^4e|J|+K5Bp(jx2mh(t$wfW>-+Axr%%5ReQ-!< zG>w>^&SD5$CBoS#RE?MQ6@5#6j1845@U3(*rSS{GoKjPFpulg<(klMn8upd=yW&Jv zFrnDH89G*GaDCx0DRyn;vT10to89@9wlfTXk9A_p6-vlSgJr`K#;@bSLo2C8=^W5O zwdhyWQqCQ#%hc@7&P|@GxXx^`Ge+VjwxWE8TDy7Y{STfYCs$k*S&o+`LsU86^hh2O;>H4RcRAp{1)JQcrV@LrT)ug%ca#mZB7UoOocTd>}3;x|uQ+;{x z;*bK+$68S>65k~(E^~0RqlDeoGb%F%V0(otgcM~9*cP6wqIbTL>yRhOrCg`Zmj}05 zsH}=t0&8a!^c}|$EGkwB94a@l1MI#m%;c}vV(Xi-Ndz4=-ojS!kI#eqyh+y3D7OdF zaV-ISvY#AW{G%~~!CWi97H7>HOJdl2Ptu=_uV@T}iEp5Rt_yUo-TEu|IFu(*CcrC# z8M9;2SSrO?;ThIml5FAVdk~<}c@=ZV>N%y(*6QhY z%!9o^;eadH2PFoi8`E-FLKfSBsh7>DCR=XcN9)fI*3b-=Y`J9RI}CtrWO#lh+k$-* zKMe9sEjsv^O-a=n+$+TgP`njH+5rFL=VF)Lx`YD~ltSs;6fkeU+wO%IcR`rN$mO)( z*87_u$OOl7gi3nu)T+VAt&8xDB#t&>Iq?2Xl<2_T#%HTVKYl-9i9s#njgD{O8>)R$ zp9Kgki zI{71pZ;Bx+OutP4m@y%O(q*Q92!5;(Z!{zyyCrv9jC$rxj^%l)Glk0YM4Ey#bQIxY`@>M7qo?x>a;MNl|Vv9I_)I7`V^BYn0D8BwcNV!I5nv}_Y* zHN?_dvDxIYrP0dWIi^({jm0P(s&out?xooqT60m%ry+8;Jq52f*XoG$W8q}P!QXxC zq_n7nOybOsU=DZxTcL=@mN zo`JU+-;&#|N8IR#Hq)Q?^1{gLGBg#hx0ICf*vc|gIzAZRL>Zo z$F~9_1QP=t=pXdQD1{j2URZz$BQ{A^%^n2Fijb?37709RQ)F3Zlnu-pWz> zuB5b3?b^h_zVgINONmtsqBKf|jh+`$%W=$-GJ#5E?h?n@~P1dH!{(^_*`cc zF217VzG}BwtTD>UeD2F<)Hb-sex7LniXQw$&!8KA@#B2>AuF)eWfh1xU`mZKBEWL; zgI_eokB7Phl%|Vld@GUb{9v-olwQ#UO<-jP>V$~#)1;-wDH+Nl)cFjDGyOF|(c@uV z5dAO)K*gwERN%tc&DPd^9NA=7N)N68kZ3Qq^kgMPT+-$+F&@^JM#P2$jPU?B)BSOZ z86NizPzco~{uMC~mw^Uol%m3tR2CxxHPF+aaVh(18E`q}t{T3cZn{jB6LrrdCK4KF zXrYXN_)BVOiSLx{kh^bqg~`z80@_R~&`cMe;Z0}IC?{a=i$9qfsb$uT#broZ#aA@} ztP|gy_Xe7jc($TSJayGyv_Ttgl{tpmDHkv?zHYtJfg0BFkHTl(n5Exqjh5 zDFafdpNOTN?Z9yjv3jYWfQHF%3;1z87j?beReDoYdx?$=bFCC_Vv%loOVLCjCk@0#T~(H;1Oia&=ln^iE~FwZ(G^F(MX z<|ZcEZLix@irj^mQ}!1tbL9)pYm$#m$FKIVhL`DI$-pHY5N^<_4cWE2>eIus`av~= zh3N*rLABZBeT;m1#j28ToqDB8tx=~UF3*+vad~}@J}2f=n*IDqO!Q-JXM{($nt@jM zCpA^;&((rP7&&yWE?niA!B zd3z1-4;;QLzs zl}(Q4W4?tFESuS5oOPvMpvQwj$SjYRCI*K-29WC_@_Ae&RjVqkpW24?1nV~cCx-?* zq9A^($cZ#g!ZKXt188lOujGO{&qyM>7?*+svR9+if>=LA+Dwx_`nGD)|B_jHSpkiFF$&6h0;6~1ydSGR0tjV1cR z>K$pG0Z=Fo8EKFW?)Lqr93@SyWK{JRy7iRdU-l(+Pz&yoltj7vFI59`&yIjMn9A9N zOQpRE1K5&lwV)s;<~;T7TkaISZz$<67dtJHbL3d1GCz-tpNSFd&M$bmHk5La^}2|| zK+5CtXma9^&dtjnWO{|-!zK^Ja^O8A3)#x>YkNgt?Bzj@w7hSz?gJbJSwAZuM z4EP_pbQ~~cwSNT1Q>@H6W zTe@Kc-`TM79qRXD<1U>bQoCMY9yOH?>>E2AasyG9+uag63han1@zy0(W zBOr4NBI3<9p5_vn$$*p$hSK{Z_8_mx&2`71wn2yvr;k+Z&L{-5^gZNLhZ^=Gw!U;J zscRjYnRv=I8}RB>y_$4Sr34F0M?Pm0@au1lxd>xr4#Fy0$=(#`nhpN=KQ=7n)eMpA zTgbv?MS~SF)0-i?T4u2?>0XBu?;AP^F;}{7-Ri^hR|_5knOfuT<*`X`VFgQ{>G>>S z@Lb{I>NL19UR`iL{KO+5t)Mh_jw|4R>HhKM=5>~KO2!>y(Rvq0YE1CeCjT=Ep1~jX zii9@KSqhKTJ^}4&{ZH z1a*_n8>5))6;R@f3x1b9l~!NSy`Im9ji;TRtW6C`Mk^bv2!B=yZHc2)$SJS;8FC&0 z!;9{&30>c^lfiJ}YEy5{*9BPdju9)T?i&ejGB!`$gEQ9t1_sqfm%~Db*tzJ%9gj$q ztA%@xU*Dw`vDaQU-|d16M!%#XU1J1T1;QVfU%15P#=?s$RCO++z6T4is%$r>&!n@H zQ!7U|On+Ibs(VP`yh1giEeq4&apXisHM%;^bJ;_qm;8zgZYL zBD|3E#qJ{ToJvvhL|uelkh4CUa|K3RGbC+2Um*10ge1LgZ$fjP-1&`l+{yXF4W_iX zM|GyKoXvI_Rq26E;)BD!WrpY&Jw`c7;?1`4atW75YMdr5;gcT4M2`H~srVsxZMU zI76*RG6&!=!Qd-!R!Iz^zwfz>ni_3|wJR^WPVj*ah}3*cn-Sl8=_!X@ts{t%SFq!@ zv!}QC5S{!6hZm+=&9hkjOUVYFFyZ7BG*Hql*I(}FsLPbGyOPN4=dz?0Mi+{R+|2C} z{E9qvr|@!*Jn%*=m+4gqD)?|NL9+>KX=d%vVa`{b}xOJsl?Lm~WfH zD+oAMUp2Xn;Gg;4M2T2ZixDz@$V^3T6ui8^I{1SU;=1%SRDe}qpWE4Qwg21g$kisg zJ1DCEn{PWTd@*UDM0M8l)NG*S_=0Scw@0o>X5`kDrT+5NT}sMPV}(f^tD3A5Den6U z<k)1)>&rfm00#(vlD9}h zK{nRr@)0d^=!a01#~V$@F3WMmgA~Jn#GxPQGqDq@$qo(#E0l9E`Yx}D>`h*3CKz(e zfM)2&E?M9`gLyD=%ML=oodtycx|*)YgBtvVKG@9}&(u53!548Th8U%3(?V!$ItC@&-fk9rxNb;1Ss_ zha<+IyZn&=XP#(IYXzsFl{?B*6-E}L{d?LCP<9yu==F+dpytxteD7Yjhb|LU#^Bc# z4!-yK_^`EF3;wTESgyPEg`mnv5xzd#(dtzA&=2QNGQSpYyhUdCij-p>Yr_Aq6r9I#Lvw4nQ zxagDr&Sk@>$Ug9j&WbNKTpZOrgZTHlvvo9V=tn8_or)8|BSe102dZK@7fo>CzPeTqpXY!X`#2B_|N=qCz4$J$6Wow}mxag6V0Pi&Z6B1mDX~-6Y7N7$OI!D{sl(x?>2l{h2 zd5H9@eJ#xrVC{v>)o0K7matK4qGH0}I2)hPA|c~+qOa2jz6s^FAvpxKfEVUT^@F_* zPe+Imz19lLEe-B^1SDM5pmrjnviG}HLAmiGDJrwD$1b_;Ft695)N_Qd_7SeAxf{+) zE6RoX!99Ys3Ny|nTXJW<7bvq z4%%S0o01}Fz0KeeBlIO2*Kunsw=~L5dCYBm_?+4zNl!9UYCB+FqZz}-AlE(|y{Gq! zF3$Ignl5WN=Mkpi^x=a)zI*hMTVA|$Ld{j%k9zt?25?Wr`a?P)=oi(A z`L%fL4Fx!ccKI$YnkEaWM0YlKVu083s2Io;>9Dmz;^d@n2c)xvh%B{!u!IZEvkrs1 zgASGqv4u;dn~Wk!&;=iTf1nj#;hZZ;x0RezP(P8jQ}>IjtjfXnIJNQoNrNIgko7;D zx1HATKA6gFHH^sdXWXStNaf%jAsgyBUv=W{Q6WZKn)u8aHMduXHG=cPz*@*ajOD}O zS!V;R*h1wI?Soy&V~$5aXfq4-69r7WN8+Yc5aLN0EpMtyquzskUOJGkqr1kpqulj_Dfc(FAmquS{yc;1 z5HDFWSAY^33{FKW-BWQ-PYQ3=HZL~=P7|HQ-(ipnEPuHyYGLr&lrgfxdv`Q81dU(Q zV5Lde`qUu;jfY(1GMn zGN3Viv%<6%K@wX`faS6B+aqSD{zTaG_FFzfJB0X2`Ej}RblXDsIoeGybX87jz1cd4 zDdMl$*9w!eTgU#vaZ2hjkAb!Tp?sIdHpEQ&dwJz(di)NL;^x~dU+f*T*!v~2n7s3Ju>0nu z*6pXd5sRJJ_`OoM?RUa0L}!lFy29Mr)F+b?>b&pC&5O5GNRVmW?ii3Q}#Q%=swuV7XC+HXAhD=-ptd)b2Lc* z&(hL7U7F|~W!v?L@!#<_YxrDW9w!}EYml8dIdcv+?)*TS_91B)LenKifR^(*if#>r zwZv#HRHngNev#K80O zU0WZDP^F79QDUS1fQ&Ui_FT!|tN9t`janLCa7i2Q(aM%6Ph$lMPgFM5JX2JLC7U=0q!(<0^QIg%T< z7Pnt(EZ%R(uqoe3L-w{ox7GW@|31=b^zThH{AFpdCH|Lr(QRE&B)JV~vUCnL3P_9# zSJrj$b(%)}9{lu(y}^@me~A+_b)bY-mOdWv*Nm)JlQA@{1iRP8NHJjjukZamP@=4V zvc~b9s&|<9$OK>?b=WDxu~CX;`06jp*)e^ZZ_@7RcilezsL%Rej?em}0bezwshRGK zFZ1uJKtk&CfBe|j21mHQSDShX#b5M;5GewpUCu7TEzbPnqUx{IY zE%6prU_+GAa(}?C`*oqbX|PGB=adQSm->_>OVyv}B-ca@*19;>dyRj;O%CQv1Bt#o3md6=Nx%eoH$9F(QFJcqzOM>ADq+4?KjGCJI5#Z~iQxl* z%vf`qs_$13_j`o}uS@^I4WrXK^(UBnfplz$H;X;p{0t?9!8Sq(G+Z;u)d<&m!P=fWi(q#;@KdC)x$&ojEN%_Cb@(| zahETmCYA=HpnD^^m9w_f*@1nt&GbSE`JH*ss6ljtczr66mwo&~9_tZwVS(vW zom#%_xSe2WGl2i&eWs&Z&6Mf@0b;M&8n#q{|2b5y1AA)3S^ekJCGk}_`0T!63Dmc) zTudv$UsT(8nWR#z6kg)wYu7)@lQsJ@o-%qJkfZ-awLold@p+}+=+B$heXXhJRQlHO z@bSM)lnj>Yw!f#1$=~mEWsL6c@ltwOSThAW_!j~b81-c)hsjjp;w4{}0?22VyP_OA z23^i??o*{WvTr}2$<8I3XaZ65#*?IE(EijZ=}f>S3UV$p7|YpQ0;f^!0@gQP*Hzd* zlk{N84Sm@cEbK?l_dvjD@eA6i2n%f|{x}WczJqn3EO0e)Hf7*?|Ky-WBo#))c6M_$ zV2mc-uwR-}WpV&snv=71UtDT5;~(95OA@51%%U$*4B!3C zf7)i;{O9bZ+U9@krGU3ow-%*|(`5vY9@MIAv|hDTsPFxK7EuMRT??GB!#0Acj9s`S z`9xH+JPeEKy+pfyWL;mH6rXiK$(0iVXI z;9hjS2~x56ag$5#c1EPc3r~2^fJ0Ctwo>Cst?5((CH4T9@iv2(1f}u#nnB;16^Xk5 z`Q;pcC}#qnyH_r!9EB8?<`eXu%*}n*@l1PzkyINT{{DNfOoqT?9is2aQ0~gX35xA^ z8%=3DgesO7=m|}07@auGPT%BG9^OUbQ(C7 z+V=IZc3#?R70=s~gyFHU>Q>Qi<>%;{o%{rA!!b#bAjvFpIm9pE@mNMOxte~;!_+AY zbvgPBwA{9{xTdmy%o?&cx6>;=XH)HT+LR5(fv3EOyd(F(^8V|pj|1?wo?U_g`G=1q z{g72)y#S9m7C?S*&Za^6B(>>E;A8n@{$Kc81;7>Lga4cm*bJ7UwIvAycLbb^#yGt+ z8nj=*FaUesJ=BcMz`IIwR&OS}n6^;ggWt)tzoctX?%4LJH3`kpC$$Wp1IB@I2Rl8~ zy|u{t%}mYyy!*W6JiM)Efm?6W*>H6?=9>HE=Y`xd2UUiKu{_Z8Cs;v&elUY@57W&c zshB~?A4RG-!l^n8$PWh|V0r>4U`4~nSCY<^fP4wtp96q=OUq}{RDgWCG*7vX?fYI7 zv^+mO=zh`{*#AtJG2094kro*KFk#HYf1w z?%mW!rRPK+-If9$p9?GjFEX?B?}NURASXLS*8On38O%st3B2eM{{2@k`J&6gUJaKW z#Lsyy=w*@d_>UJ$*O)7xWy}xf;w-ztdeWL(uayLOW$|!5j}d2J6{KnNn6>QlqSAYa3^}(0pdESODeI3!b-n@#6H@(bS0Vkf6dSAXgx!I&DMBcJ_I6@^~%9B z=3&VCxZx3xCu!3qdtn)IoX=E(SCToh1&zUU4UaW?$N7G_Syntri5)hIbZ6Y{VbzpyS7R7MCU%ffnSx<88<$Cq94GCKgMmQ?-RQ6Fk8RkK?8`< zVA`jM*?x5l^+ewq1h=mqzJ2>a#2f_YqKFxXTpyf@16jvhlZ<%Y4XAU);lT=ae88mE4~razxO`61(PhVegGwgGz2O99Q&K~dk*J4JhD;yo-{md(-qox;+y&RI?9BMb7on0N26UXA zQ|9rx{Z6b29E|K~{A8OX591^kJDsLxtYlR|lm9>689?}}tSVY#0{p+3XMa|JDJH6E zVu8fi*V{XD=s)hYo7g_#<9Iojpo(-KTS8upN4`n*>3n)iRN240OG`z2&r;;)W6Ala z!pEynD1O=5Z5jCYTf+IbyWjqScO(RYM*E3kQ|gt=c{Ps!)_2u3Er$gIfNVsK@0>_w znj_CH+J-^9$jy2Y_YUa#GWrbq#1lCRnIA=r3%=aZiz*)#m_E6=RF(`MU$K^ajS5wq zfV>i?pC~d0a{bcZhmWcNWveBCy@c;r(N$j$hz}I!OG*&<3(Fr|KB6rh7?Of$*P@)* zG6a6<9;6Zj`-LmH1NZHyfidb3fpe5^CSKnRBKHOQB;mu1ByfTNx;1EIg?M)Ts80X} ztj5SV)cwwk&vo(kGr-nx#qL9&>@vNR{hau36yiSEHEA0VR$f2YMWJtl`*?~2`ekq` zOsA9{ae);6ok#)lf9x^Yr6Td@%XAlE7O&pJy6Jy~!t*z$)Nz71l)MP5PteM)@f|sq z636JtE(Iy_JJ*YiAd~it>mI+9FE!cr9U+@OYD`wWs5YB0^7Rx07NchbHe#(+N`!q& z)ne0Knf7df)0v3_D#5qj<}%y_*1vcUYQ!8bld}^7kp@~RvKl}$@#f5@V!0r-w@nm( zJwTbVvj*6${fKjQ)t_tLrOQC^Il(|xG~jHQLe`(uNDMfNT}Z`fLgiYZ%xt&ox~hXC zA8)y6p@FaNv@>aCl|Wf@;iR$T*I5T?L?Ac9pWnsoSF`i_%IsHVe)JW}&Wex>G028* zQ}L70Kb(>w+)wzM8q)nd@L#K=xpOPuD1`SW-Cw~UbESM3xdY9#5@CL?_!8*q%^QF3 z+g~^y(t?w$QC9ui9JH}R=0hy2H2jhHp8iS#g=7i)J`+#Anq+3fVaH?I04h@r$!!hY zvxK+6GTB#I<7)g-8{7%yxM>J$eYOU3jpdgWUqC6%Ff}1`YuSf9YyATVB$p2GekNx) zjIZ!yLMqE#b_yk3igSz=Lu=$;kQ%i{C1CpGD`(Rvft6^E>3yKM&W6b(s%>U0Ln_N^ z3|b%~wUxgDHGhWHLKqaTuw4eK#ao6K9s!&mNoN->M_@xM2fO&2)@KwTQ*XL zTZll-czF+l<5qx;=;ABlBY+7SiLA)?=hd~r)%(5ZyLTYyM~l7nozZ#-jLAY6fOD{? z%9Jip-;aj4AR+3xMZuSj8Jh&RKBoLxkcyA@A2e-d%v1jkeJh0bW5#w1od*Laey6>! z7OA362t{(}{>z!m0U}@}*B;yT-6-TQ8^X8@H<@rdZ?Ho1c&06|l}aNG&hrQ;K79)H zMPyvuwFQz#gxEI%K=I|fya9 zkMh~T<=0WH;gtRo*!XTDGOyWk;8H-1d`s)r=M+*= zTBX@R8Cdak*+SN1g*vo6A^s?d%x}v1>Zm`1d>}FHi+%Nv_jog?wqG&BA=~IdC`WRr zsF;w)PQqKtQ&Py3A@)hLvC}_yjMqP2JfP-)0x}ukv%-1^Eii#wToba7xj6SgcxY{K z=htBZUT8O;CzH?%b)F^cAd?Q2v3wF_J#@pA7lrg_r`v$bV&fC*>PSU=SOM?mr4q0T zCC`Tf$xHhWplt*9gn;g7c%#;MHbRmr^=^?|iP@YbID>3bpEqs$~ChNP{g5F@C+dVTd_1?Rka7T>Y8=r+*3= zhg|jfkcr`Vf_`PLltfSfHQ&O@rl8Q5KAy~h|MpvH=XrZ>Q1oGrHAs9 z%-2bMDE`>)+-0uK)2uw)m>;89fK%m*6l8_g$0g5Etlc%ijQJK;=cBI#w>BZ~&#b`F zDvUGPYg)l}_p+sZEH3cnlCofG0I~odl?AsriNR~JVH(88ZYg|W3rsvGA=7`|)E-!C zHZpEcejiXJFXx?hoGtH`nlNzhyTlLmMj7JjZEQ3dv*61B12H)ZuUORb=_BQ)Z8b_+ zZc(pk)bixKOCC`H%A8w%S<`2q?HY1W{QSD7=xoGGXB7L=#^S7;ECdsk9$x-?=>}r8 z?fwKBD3jwXTR8mFzFs<)_Hlb)l|p|mf=EvqyZL!E{XUj;uP0HJn?cAn0Mh@C9~i)C z`I&J4;j~gRa_&T)>aG(Z2)9`Mr1s!vP6hxZPATo4hFE<43@Ruv{L`n;4I_*h2hxQ5EfOMOkdZpB-kD= zfpO0hDTb*?pG7^OC_#9aT6;JKgO$&<4ucK-jjVgKe12D`*p7j<`Z+n6&jB+ z^tBQ%;+Io`TB0VMjx$OSB6OahWDtQya~fhFb!yMQeOeCO@=m+9l&yT=++==d8UFbt znQFilByXX?BZJ8~<$~wvpu8-H?HvnG<%$&ouxaLe*9DXx{`oIc*+fL6`4z=L-?fc6 zaqu459cxv-SG8e;31lP{@}qYocH0*&`@$=D^j9AgiWVs@2?I!~#e1{<Bq-jboB~V0{T*sIcX99Hd|)R*JU4+znm(eJoT7rjqGRPgq$YDs#-^DzsqBre2L~a4 zR!GD1o~1pw?D;1T1Hun#_X10Si|=}K5+2YIz=6MMR!Z%;(RGwCp1~nkiu)R+(IVCf zw4R{Rlk4om0d+w3`N)z9*SezB1(G$!6o>_0W?gd?plDD4pQsa$9VQ95#c$qD#z53E zy!ln#W9hjw_`j2sCn&P|IgR+zr&Dzd0=+-2?%@ZOZ z$gS-P;sKy|02JR8ey?LL3$7I19OvG7J^a()=E@f!*&>#TGRFk}1_qUj{Z#PCW_bKu z66oq*x24H=hrSRvM=&NMJxtF=21|KJH~j(W4E8yk=CYk$;U+}Mqs>>EEdcb!kC(AgA?P%n8~V80-k<9_Tj8`fwSExD!W2iUh|y{Slht@$tw7)8aOx7F79fAb4A+1)b>g`ECdXzMrzd*qn5iL0laVbeO^Jlr!C+y>Z&Qwgb zJwccBC53w2H%tQqB&D-Y5#P=u&N>3&U$W&su5p`EWpJRWWU(8AREBje_I=PJ*J=Ev zfVRNk){P+``*m1b-jN*gv+VnJU>?`w%%)Z3$ws!6or8gFIgM{xvx6R;KCPI?ruFQc zB(M`vJlF+`Lz!3g^S*!b_+wk)dKW@Vt$gzuGM_XVdCwVj<{gdQT%3E{E%|0v? zAJ~WmUtH-c(o+Arnn`yhl{J^eyUMCdH^T$D9rEa@jzWrW8YMOT>E$*u5kkP<``D*I zaS?i8at#BARh;O+Na8hb>rN}_#-ZaF|L7oRI#qF?6G zogHT*v>a)}0J<3*MOasloS_~$Ayv7Fnth=U5~}lR6Se=Fd5TE?u(=~M^K>%c!=@sR z^$v-?=0QReBYOCn_Ih(xgTbPU2?DAmzNTmCzw4@NR@$KMHmPpTTk(Hk0;NK2zh z%VeNJs!FRl*Pl8X&t=jeXOhxa8|UqYEJ8pzHL1@G4r)uUub#h`=&rRdq}L4fXT!;M zdSf;X3o@U-q8~_gY2RWciP^(FaDUf_j9d9{%4em&NSrBsD{lQ zD;~}7S}rqj2Xf(B(NuQLB#8Cg&THhaaX! zX9X*7=R~0Whvs+HP`kL0=P9Ni=qp48vlp!2zeJ?d-F`WA#;;9ZRU()Vj{7M~R{Co; zd&zd=Ewb_7KWi$hB00%!e`rF%qbqhA^BQS1QAwEZ5oxw#btVy1bfiJlKaBII zA$XU<75fa%Gw7NRc>Xo-j46rrWajjUHH zA6p^E;X2{1Y~c(wTWF^Cd0eA2gpYVAgYB3;hbIaGop7Jod?>WsBppk0B--xYjfZ1El*?B z{@zgIJwOf@6%8NG7YR>=I{6fRpm)(M@si66hGi~RXCd?q-p=b0`&Vi4dLmrd-~k-T zwdshN+Uk}+@iRtmuxS^2*dwO$_w<*4a=xIMd9U+UBSHFjwYu~N?DQ3-_H|3nA?Qo( zl#eANQ5W<{3$?_T>wdPsoMZUKQ>ngU1#bslnay%WEne-!Z+hE3d{1w?th^r0X2hX1O(5GM;x#B^!oqm!pOPDVz9{-S8w0cNx}wCyd(Cylkr8hCtpIlToOrrmz}ay$ z(fM|9z}->7!g&4g-F#nD%<~ir$!f1(r%ZsSEYtnp&#=1d%)pwB25k;=}0Q6 zg3dvd66K_;=FLAhH1_8LV8-9(NF+6%1JSsJCKnc83Fgu%>xRA)hP==%{;AGlFr-j| zAUZ)<+)Z$kq`PjT#Vl@P;@wtMy}B-{R-Sx_q-ue%Qynnhp=VlHQf{Rzj1PLScIUp2 zy}Wbky9`L?gA-<~tm+LS!p>liOo^t(b54<%l|LHu?s`2>$M(ya%xbEP=q4?(uPtda zPu{3B9rov4LtajId0O>^Di6(cD?8mxrig4Sz-D*n7pMs4hj~~VAM%C+zQ=SPg6M84 zK8|Z0TIkpO1hY2sJ{I*6F0lF_BSasW$T>AOlI}A`SOVOC6BWr}HLLfjY>4EZp{Sj8 zu&A!u+eD--=8wEkpT-@NPtVG%9Dla|@BOS$T!hUOmW)xn8iQQ>t@FNkL!srS*v>d1 z-~;8KW+QC$*J_z}M4Deb-aox#s@ftmd3A?(Cs-+=sZAeIOng_}CYdcj-V~!i5o+E@ z^IgGh@hD1N^V=5Pna}{^x2@LO7kn+5h^NpVSJQI@AE~^REQQc5@aD%3R%6zQ#F5$`=*`j(+jGl!Iv`ns**99WB;XRe8gAOdMFetCoexPwvW@)+v)u8kp#OWY;Af zR=)}2m@3~wX*eV)%8(Zv8kq+b^Ff71hL{3%6)G9kZ;dE4z%?)auEukOYVwytpSi3A zvhg91OY?jD$Im5bN^X|jG|=D!VmQ>rT4?X!Lo}ecR7!U#(+1# z{kU%79Gm#20dNOcE^jNa4ZLgC*tr~86}`su2_${i2nLcB<}^>Qq!gyqr~XC@Boms# zkNQ&d(-|J$dlVeTzg*JL{GP(djF2<*CT!HcvZ}NirdTh_X&!YS*EDTh=M_Hof5)B0 zRijR&FOZ>m@buD(wJ#T7nE>P>A1dC%Bb5LS9-MI2ph}I@!(R zyX48C5&IL}VDx2ili%>BaftH# zmkCG;a@H*)NnYoW*DJr(*KxY17ZY_Vk)mDiRNE#DQ(@1hQH}(9|qrO;eIQ6YxqFVr;+Abz`Ar$D$-giyLnZ}o&Y`S zx*I~)4z5`|JF{}RvRV{UdPx?*hqLCZz6Jkkvm@xI6EM6d8?Vnree)4L8WvHX?Wb-l zTf(4cz|t1z^ZgKd8K~!-2l1||2bp$o6G~;hvNb_uszdwWZg^mFG)r`D{ z9G8Kh9mlOQAZNgsZmVdR$O>7)Q9jjR(O!;`-v#vZ1A1Z%yr2 z4a3MyD8H)MuNrWxmPnP@9_kG=x!$FRaa>fxjH%zm-{>mYO+$iQ3OUMI^eC-EIKo## zKCT%6Up&lh7q0x~&6Y^VL-Xy-Pd8qF=GuxWR1oKNQDjMEgDaNiaC{Xg&c(4`wy>hP zEkhr?%@O{Pt7mB)X!L49MfFzaJI9Ou{Q?1pR4OYJ+#=XnHLYvJgVNe)ct~z9)gdqO z2w*Zw#4fzc6!2Z=1tp8{%NhXOg72B>AHiUTqrMZB4HG&W+?A2^6DyX>zq+a=hoNaG z4HF>ZiF~g8>hm9>#vS0Z{&zkRVO5s_Q%}8Rqj)~uSn#E*%aSnDl zsV>C406{>$zlgSAq5Eo=w|xXi*TIR!7e03W3gf?(@~&#Nnaj3c)i+e;K#$h>oa8&- z&x$EsxPQP8l`KPjDSQM_Z8AyO69p2MwHrT84n-^m_?K}&yM|ni+FskoZwJ`m9s%xZ zgz=*1wtNJ~I^@6thPlAHH;p8K5V-shZi&wb86YbMFe z-dTIEz1H`WnF;41d8%ZTWL*t+erwg?V!`PM%cPV2<0?+_-1|>yr#=#!7JiL?ccT$q z&M7kR;&jCMX`cCrtU|$DK+5)qtqB-;gaB|*{H{Q^I`e6sH6Be@f=u8K+Jg5_y?Nou zi`Jo)u{JffJub6}t%l*sxcxU_S0z$+2Fz2^)gz;Wj3mW{%P*G!YJ~Xzy;7Y;yIyrR z66Ag)i2c<75_G>8!28g+cvY31t8s+*_+;d(27V2u_ZDY0o|!Hf6Q>?R)!6p|rAYUs zX7(__cO)Zz`GxW}++e0(!(=~j3fPa5Tf z>``E<9g~ciIIZctQXbLof>#9@ciVRlv!|5u)Dc#m0=$kXKl94*rk8tHYh8Ln*;hXy zY1(+c)SzQ^Z&*sR z3m2*^19li1Uc}g@pZCA>4bsXKgpcwjFDX@Syz#qS@gBgiGtF54jhski)V*dG(oUz^ zdF*xvaK*kXmPVK?H%CGH_2(&NPZ}57!-JauCL{M(c;Gw(Q@4oujSs>Unrc0DY3+ z5jlfQes}bwkVt&#fSgy(X~G|v=s}2kym$V!p8dLcoLfp*n&fT9eW5!Z%IUxoW}>mR z%iUg!mk(S7CB=c8I_&Q=mtaDg6>WQj;1IK+Q`)wEl(_T-_e?CqY*@W7^NkAKBlBKu zM3WC2KDtXzO~8lvb!QKYlxwTk)ZKbR8MEo}Qgp2EiDu26e0c4Poaxzg@Tc5YLmS8B z5t=gkTW|-9-5>OXBy4V3+jImUZgK)r zm}GAyy2JfJd<7DhF*CuhW|!%}Pz)m+XFs{P{|K^HMUW+;5O;iI89 zFhu5(Nke=t6x{1`rli0jiQ=>RBmh6%dq!|B`_>39iOzUmENQ@%SB85)v6@}$@xW>A z9JVhCH!`W8;k$ex!qEcJ2)w!*Vfo9UtbrDlBX-s78#C7#-^#fCr@mG6wzSx4z1juVeredG!w3F)L!VTUXs|;J1sngre)8+m6XVaw%K#b*MD;bHqh~QYCPG_WCoe>wVp$R!=eqF z^+2osd~GtPhSAr8?!3{1IN?pZY0b>!F{3M>TYYms|43#ACGX=xhSQPsKczl60OE=4 z>-@@V;66}IJWuQ0?_kZQ-OP;q`@b{dRj6gUlC#MF1!0*hpydH2@QRZbc5Bm|6+)Iq z;zA5X$y^=VYcxsw^$xuX$o+pjrF{mVJNYWK5(eMm0vO2^VB3o%OAvh20P%qq(|XVgjcW$yb`p|0+Bp}WOv zU8A?_A`0pPQFN-Ar*=&M;Y3=R@zTyiVdF1e#(dio<4!Youi# z^A=9;3-GbTXGh_VCh;mM*9P;4M#lRN_hqXMD#v82$AQB(#9KeD1#alA=TLIruWE=G zJepF|9`U&b02j6`(`MX_f$O-_p*pzwDbOY zd%g&hL+}0dNMb)XFy4pfO>9Ai(fLM&TXaay5jKW@BKNu?mw`_Fn3MCup&9>0p7^WZ zh0oJGFC&xnuET|!y2Q@qFSf=wlkS*(RF(-A4q3b4EH7gvG_p`}4HTZSyz9X03g%hh zqNfWLVAVU-$RxM*&*mqI&2oG^_f=8_Z?;4x1#G*&{!NT$LhtG2)`^%3$i~49D>1m6WKHvkt?wI>|DRursYvOX0dG$Ah6e0^ENZ1( z50I|0XVR|kI~8ck;E%-~W$w$KLzm(u?4R9J=U87jW!vqe#1iqQoF#R5_dOgUZV=S} ze5f7*O!Ni@=eIuN&0@D#6+)*7$T_GwU2%^1<1p?@`E$BQNpi%-`;}I-0=%b6i~a+2 zE-^o)SR_$v1%8efN9J&UyxU(N-%^}m8~v2tP|fUHH+)>|Cr4GQyZ!ywfH5Fjw-hh; zz(G0Oh88I+oHU6NF&SXVdAhPU;KL8 z%PU|j>J2`mh(eY(INtYQ_gj3^$P|X459?2i+VG>ydi-O_?25*m8RV`U7~=oj7(sZI z{NTv~$<^kP>ehFLEqqP*z%-pphg8@k41f8MBmau`f=je>Uja?G&ehT1gwcDN*Zokz z07kyvFf>h9)P5_y9DgGGk7*Jp_WEo;+!H6SD{^F;KgV}Ou@eN!pHm|VL1W>@7o%N4 z(ky1;OumjiEc!o=)TiSzm+~5d@BOX2ceQ!htW+5>(SGDL%+u& z8{#}IUV;q@_h!|e`h7qFk`;Gpi6R$Si^Wg#6@<#*ciM>~KK`*~N>liS|8LiA^K-s! z+K~|EzV>rsGm5Tg`h0xc_#1!k;o8Ar@5OlJ4NPSBwNIXRxRkHsKjE#W(t7+qlIicX}dl-KraNA3Z=ijW)Wjj<`&$Ucw>J!o;bnUxoDoE@}O+1M-MZ?9ECD zU2^6U!o&06E*{2z{uT7TThHehpkh5CPUk0J`7M$dSMsw(J<>K~R4M!eiq6HvzVi7< zw!#m>vVtYIKFy1}WF*2%`d@bIZQ502z3-&sVd4pOP^B|`$mbe-6j`!%r(dt;Y2HY# zu&Hdw3fBhqMktV&kU1oz6_^PFF-k&&r|Y}tRn5eE`r+>t(94jXz(;vdGyJBu?acc6 z&2H}^&`&ZCzcje>2ItX6+%j&~9h1DuCbXTIv1!E}MgB@+da*mET{MO73h2Qk&R+Y9 zkSlOhmjFP?V?J*hPwuCA0v0=s8Kci&Io~A3*R7p+gURqJ1X|rC!#zC&o)oPY5puOQ z9rz4dMR}q`m#u-%uJhnE8}2GSsV+v5;1-JkbjpX`>>k{0)?_|SrR?mseR3#E_PWlw z;(afyw@P79`Zp&_nFxhdz71bdXPs7(_)4>T4W!bcfZJgItIqO76qe zxy%6#w!cWWCP&KY)UqbC!xL@FZ26i((;#P|y+9*Hp=djJVKhia;V@cd09C$;Vn<-A z1%ohJW%?U+*RE6+h<1ADO~y;jukC~!oa6UAId&sR@%y6^Gt;;#&?1HmKLro>NGQ_8 zOEwu-&Xm49XUnk_ALwWs)naGoweCjk6zBhE7=ZZO6h@fJDH$>gga&)^PO1ejv$Y^N zuT(A269@IOfN<}9C45Vojq}@;m+cH;wL6o$S=^2e)j`e;RS^Bg@P=}tGTT&T&)sfyB%d4a(EyyIjr=cYb zjfbW!NX}btVdhC2t(srO#0#Jds>gpm;CaKi0xJ?j*X$Igam_HWR)5 z|D1P3rc^VF-PU^wa(^pw=~oyUD^nkSc{=!B|Dt?FK7?#jpldN7Jkq*WDn~{PHEH_P zMEm%3CzX%vVDcGjf>*?etZ@ADQcXSZ%4ub33Bk)uWAzlZYpHmsx?t|~{z#tMaCmS% zZwS6FVqW0b=h&}!kIRad8uV9e~y@;_<+M!ehj)4$oPduo}q-6T)w#N%g~ zJum(=BQLHD5R!sY&Rfc|#dBL?WwsTsTgNDF-fm3}Hb3Gfsf#w93gd(0nJ(8hAZKJ( zW_NL31E>L$hcVfSJ=R&4e@(wCe(6W;@SC)?SHiAR5$2+zyQwK=)fVP@<)ajS$8sg+ z)ynTdeX5>}w<>+s-;}`k@LA1Z^M=sdWgroTS*BRnra!J`ypl9f(zR`aj~c{uZmm6$ zrTc1)o@``wH_ln-tW%_%qv9guKm;>*P0Vel-T+gZKa305?JuEaXIBN>-}J4K!`F~& zR*{$^%e_0a^Z^Vs&qgV(@=}+VSn_$9J2tK`gT&nUW9~|y8k3Ij!h8nJqmN(w%FE(l zeC-ZB#0A6uo`R#Q5xzA$6_J<(uQsAOS5Z6A`vH^#YxSyZE?}{F%lGi+@{Jy6C(6&m z4fj+g_@81ASFuUk$8_EtNYLT-QK6^RF>PE@u&<6eN=lW8}lfKlQqH^=X(0N2= zH%juXUd4d$?yo`ORi4|ATP+YcoA^#tZLkYS!sjr28LONQb5lb-TVA>xxYYnN&wN4J z24||`w&lcSBtfl_bG^2*2@X;(&!PzI6(>c!x&$E;Ba_B|k8F9hnb3qI}sKciw z6!9?#tH!8^Dg`8Gfy(4#c*Naf%cR?Oxh9|c)UX!9*K+|$#j6S|)YL@3)$q@ep`)5| z7;L&;-JSMj>d~z8{k74`kuj8H3%RDe=abYOrQ51^sX_PZ?!UiN$->rOeY`G?%y3f*OXJErtto~~Cj`ST84;4|Ci8R`jBGFJn)MJvbtXs^fn>-PX`%55 z<2^9BvH+$$*>Zth!MuE1#a4xD?K9-;N?T3da(;_H?SatS4=N40wC6hSV3-g2&_Zkc z@5k+r3dSnoFOs@YQD-CKb1w@%V8RuR2qDqM~+BQDf{=Gc$7LIoh{uQ@MYv z2S5%MiP;AEJKj~A{|Zz4M-AfXL#^M8ri3w3IAm^-_w+2B8EQ6s!@KGWx#K+)QNb`f z`VSC5ov5tAD?&v$w)cIRV;P0Ow8*4RVamb0q0vVa)jTpX;nWAW{da%SDVpjFNJoJ# ztQI37(cis(sLZx&BD0)}%*(-TpZif!)%3i~B{5}fqsHv>SH3u@eJcOm0;2wpW)t2B z%wv6_xo(p2y67}RFEYP_`K$U}t*bQ1*gf@6SWg+t?OHWGMEHvYi;sadj!_w6%oBoN ztr~KzJ6_$1)b-!h+U8cnR33az#0lIYLyUZI5c!D`e9m>URsqf4n#7A!n6jK#%%J8B z%bX@qg4Se+lCNs`cW0#Tv5u)rOWtuf0h{Q*)?EHIKZM9e47*nD)OB_dGrk)@Ee}sm zl9|r;8&Pk>y$ifH6+Rv52dr2uFir^Co$%JtP6&$B7B-(eP%bYUqTU-Qwmx`+lf{QI zk{MRzy?p~(Qtg~T&3ysAGnx?WZIuZ&@oL$@9}awY9Cr#WmJVH|7h4ik{P-SrX(^OQ z9RtJETCpqiC%^GXJ-B!^M-39AU?KtM9X+&F$z@2h=v!Ipw^P$c9CCJHTE2m_PtN!4 zac^LJ7B-{6m2a>6bE#MCk@@EiwqIZ8Lk(oKNInJDgbIFu@vTqJhg!=`&tu_%9v@$_ zvoAqj@`IP9^GhCu*>=@=z4je2EOn1Px;gqh3LI(MHKZK{!hS~uIbc`udI*TR6 z@O`!~V~*O@Y9tx_DyK7}74FcrXdH8@T2t8}YUj$hU^Nuz7u7?6n9* z^5Yd1sYSQ{GSPf#oWB=ZMY)dmzwSxp?&eqe>J`ybM^Q~O33Tcj&}lyNwrW}K6y9#O z$PQ!In(DF{WB?|xj{+l#`A$Y8$^?r8tR89#N5j;zhrhxYI#%19S@uO9V&N^C`EfWF zrjY|vn^VKMw*{4G_N&d42lt=mutr?csL5LD7pB=$Wa(ATOy{#Jc(Sp~UyVCX#Ct^y zwBNJhV1Fm)HO%m=Vku2pt;cU3#eN$O5st~AR#aF z&f$*9y!2y7)dG47Z7-AmOPZyzi11+WZ7d0Z zLU5Az-l`OGuv7Pm6t|p3GnSg2W3Z@urCt${6sQdw8+l1&*N0EkN}BP@SMql! z1QAVlvtAbI51?9`69a|FPH1s7a>1blm2q&G5A7PK!ovN&?ka38^`E!v1O#yhhJs_I zjYR97R*tWH&8u*C1{c|SHr@-B3qD?~xB>>ec%B&e)a_B1g&I%e-0B0hF^XSa|LG7= z`S_}r>bVO3`9VXYJWAu8vrKUNB~c(N24AN0u{w{3zH*7Q&>!EMtGO`JaiO!98Hb}8 z6Zq=~0e1rqa;)VoUn74~uwIq)-xqE3!6BOkE#(Q^Cr2Uw(f2SR@WgoM<^$JU(x;`z z)F9_E>TmmqPyHwb$Dmf33?SR{8I<2+>hLAdF_Xmz{*8|Drk$6v&N(z#-_L52{;H&Y z78U+DBUG`{(@5;_UM>T8$_JYvQ90iG=QD6+3FG7`&<_vfBChoCGep$gg797O!D|Tx z{99AtDx>E`ckRl$F!jdXaSM)953v-Z%048`dKvAP1tK4R>H7S7^;g^|N%*fi2MGl? z7|iWQK6k1$Ii=tUBVwF4=tleT>zbWFELmhIdJ8JMd}V6gJ~|YLx+U$b7hI zw;~x(o3Q3xx~I%u5pIrmOLM6z!M*;8DlefV{lrs$Uc`{v#qK;3J=uNN&z?MeUl=0x z#>Bn4@oD-n?=DbDLTcvzqKK~-ftVO4aU#WV^;$Y{YUHj&0DT^ z5i$*iS-aU<+wj*Y;)Z0H?yU5~G2rI8DCGUh>Azu5lnnksbnefV5$N-7ETHl-!JTaq z$n)aewrREAO6ny*m$G1!McGdq~l!6XKl}{ zy5bL${;K$d#zC%Xc7+e${~AvSjtPh5LL)aLRON$<=fV?kNd`5$*1+Sjl7O61phr-} z!s&<|UKT6Kx@k9-KKdOr(cJ1|5c>LX=y?r-$4YnrRjV{fSaB}O@9YZPhaFn!bD{10 z41eYD@0ahb%~v|V-CpvL3&zE9D9HCcg;Yzv{s7N^4I1}))}wED zu=&+_C_DnR`q;pm`rC`kh*A7@P_(z+Jbs#nL}Wtn@?z5kYx-zvAGJ^tz@~%e+wSDt z)+)n|w8-@BMEC~)5#zw$f6r*HfW2CUazF9!tY`J5DnFLQi?!<^UiO86HFBVBQktZ8 zbUwJVJI~&|q8scG_{B5wH{&?_yS!=iz2X2N0Oxb+Lm(5MPS-HZ>y>&((g;^bi=&Vp zB3KdMS?A^cLe`2Ax9NMlGVw{IZ4|$32WUAF>&C0w(Fz$I0*?%yOC18uHKap;_&fM7 z$^N$a9&wRiX_>Na-T_laBVqokIXG9%eJx*|q_(Vs|Ku7UHoYj%aRy8YA{+XsKajdXvj z$Z2!3H<&ot@PpUF^`rR93g*RUI`~$WlmKwUo>2s@=#++C!NOM2o|zZok)KARB~qJ9 zO8c~D4#>yskT*55pGCQ#{gua)TA0(d-+U$}EQ4parlR|%2Go-w`8un+s5r*Oqu5EV zQ}uOZbWKGQbX|(Muc){fL@m56(qlB8oA!Y5k%N_`4CJYi7`uuHk8rfzmwcUJc+*xz zq}9(0?+_y6o(yQSj{fb{GifvXo1p!giX85~DAdn(whM>En$hcB)9=;iB#?)TpDt(% ze_nj)2=+$u@p22FYCscNTRq|hRa!A145#apTrd5M~!lDo9|fNpK=?eJy5R5~v(PaF;H#e*SzgPaO(VHjGDZcv zB!pmx0wX3u@97!_B4=QCekoe-IrG=9Edn%`r~t~AU1-(J1o8gfs-fh7FtiY)W_6Z{ z1Yr5QD2GCq%l?F&vTJXX3gFYpy!;hxqgfv<-sznENXRKhuGx9G9i32iFb&~Ti*@1I z_1>Za^f6NoxtyFoeY+B7o1V3!))pHZ%Iegdn0H|cfRrg=tSw84U{S!SCAT^XUAsP^ zt#V3IkXMJT5+y7Of9yn+g!;K8r$QeO3e)l+7eYV88(CPoYiEKX=WLzhG$G!|5u+ZX zfan(vc|yBdlg18tspgh3bu$x=t!&)xXo-#vt+MS~kZG3}!G-BHM%wk}JsH*Cr4$P~ zRb6IGV`D!$>{Tb`DV6I7#>DQ!dy>Q{Nhdu}4UD8jd@p1Z0$V9`r*=h%0G~|!=2Lb{ zC@G@E%{E_wc8JlwQ_`hRQ^iyE!!&T>+6lDOCyNE(z1_1M0c6>mqqLm z8>1G2^h>b`lALuzzmi>&X(yu*Ee;_2k$X;8r1%EIvXNsrNL>Sw}48?39kN@of> zY|oSZWIlm7tOyDBWTZlz&}+Lh+@jP^$U1*0b-!$PdD{7daaUyZ@Yt5@g!*3?TLEu6 zbt&p+UVFa>!&BxAzMlx(vK>R^zZtoEbA&89se1`1N84DAwBR1;Uhp%kZ6^g4kj^mg z;1mz8;_NEVlDx9WbSx*!-oh=*s5&3_q^*Q;mZBu98-J;$B3s9Saujof(C6sMxHluR z3DHXs*SadcvuS*_S;NS=Ut`2~nslIOtJ~p-F1!eMEZ##IUINUXK;p)dniZ9)bQV@H z0#_vgr!KJ2oLdl|A)Rx3%77hdexV|0-9)>D=DT*sqUF2y+Fs9S{yVTw@OfuWmw>8f z?OkfEnhAGt+r6Bss7G?Zp@8)8p3PX*y*Y{(mmsYpj8X*nn zoY39oiVPJ&jt@0Fg^cUYEZS=7UnrSem2eG;m2xno?Hf`@w*F#<%t&Nj!>8{Q@ZNh_ zn3xv|vo|T=_5DMnV8+62>0uM`g9soa1B%Y(Vgm59J(o1YZ>nXwYFTcjMP`Ec)+QcY z@eS#ij&2$i@V>abp0&ZkM`Bae@g+x_t!!mAry-5Cr={eJ>_+-C%u(#UlqQw`{;?G{ zeDTBge!nLZO`*G44l9z<)VK8%DUMi3K$ZQZ+~N{q>Vqe{p>%jNMFUI}B=v z;b@}6->^zCuPX;>J_6BG56_)e@TaQ8L6c znsG&6zA)5Dybay2TP}n8TV4UjEDxxih$DM-SplDn>8iwcKGQjDyQs#M$ool% zRl-LU(?xhc4Bs~{CwJs|M#fTK8J?~zArCBa;!2rnIpdOGWmYPaFk@@m{Z;A;7?ghN zi8Gda&^$q$#=`HfZd_e*jsN!SsX^QP*U3LNJsAfi^nCLW?~!O0EL@{n?O_$O&nq_% zLMyxvIcs!YE_#{-x?}x-U&LK=PbVA_GFz{o#EW12KpAE)Q2r3VjF%J>avEYYr2Fxd z&IJwXG<{1icp`^+%?5-13}Qqv*J`?nwwXQE5t_3?NFf5#uHwEyiGH|k@K(IBBG4saH0p_P4n<5Z zUQ&uH*5_4ur|O8;I;NeXC&Lu+$DJWTe7DD$T$hha5z<*5cn+neS?MN7*F>H0DC%lL zu`DP1rFtp2CPIQvc7e5E7R>Z$ILGnvzcd5%@<=P7a}IgdMd#I;`U{lB_nDojThe%h zb2@4zG?z-aEH+qrGA>WDN{iyb|7T_$N{n>I(iw4;@}y_oi~AB7!1|d5piEl1x<4EP z#De#)>c6d`C=IkUNiV~`UBw{)xoHvpVuzAGU5Z9Xu?Q z9hRzKuF^j5Qi)RV$!bJz#bmMy!>?pw7Rmz_e%3L_LH>1=zN>Ty8Os_-g^IHnYF2hR zs3rVV;05j~K6N&(6Vq>xT@?j%X!FQ5S?LTf2VK%6Q~HVm9@TeYj`u~T8~Qzrs-g|k z6KO(hvufK0RPle@MU7y|=1mXHs?y-T4wt9T`X71|MOhZ@ev1!TY2ioT8r-`~!hTSH zwrTQZu_P!Y86@99b83hv!maFxTy2_)3U_0Umy+u(Rj?xm{bGZ*kKt7|U@Mlv_LtR- zmc7M>RE+xlAEHw%N<%kdz|SzH z>zC}tgJcrS-+sq{lgIXc{;;4px_*c2hnYWXc0W5dt75U+g20YVk~V^k0N`^kbZpjS<>zuo_QU`QqlYAStFkvXtAcB%b$~8B5o=O$;U51+i1{&cgkHp zb-p5rr@va7ojo0BnhWVpkS9S!h-! zM*j$nCHg_fM}+)a7|WraRm) zhi{4gh!-Dacwpi4HDB*9=-{{fjJ#R0AB_U-EH)A~U7p(fq8Pu@YJc?oZxqf$MrOx> zWi7?JAT1RC78g>x`#a9{(@9{yk#58vS&l0=4Z~SAiN25k546dvB{`TN@79<^-!D2U zF_vr~$*4>txGKrPgVGTIj;EEyaA!Y>ndBRcU`>uF>AfLAYGp^j%GjZ7$KH5tlD2-Aw#!*b{q1j&Gbc zqHTz+6`sMeCI{Ow)lKD${fuIHS+L@R6rQDrkX5Bk=zdx?tkYhCcx);b)C<8NS%m=_ zH-9b)jwLtR&Ll-{L$;Oin;3U?dQS^)SbeSv4cb0T-}hym$^OGrs8TX7uDqM0|MK*5tb2=oG9Ujd;2j~aXbO|}TDWrwmaZEVj0IJvZvZK)7It?PKB1-` zyr9D1mtXM9nv$BQe2R%xn@u5s*h^&!e8l-66Rsb$e(^%GNdcdJr8G!Y0i08*8NphA z-}83OP4ia!atq1+6`tLwJcCI_d>^j@UBiDAH-i7&^Y~TNbfirSoZ$$<)~OgulR%m{ zu!m~WE&C=c*@Mj!hQXWGimQw)K4!c(!56x;71(> z&x46r%y3Uw)+(SIO-6&Sl;~CEOQtxTwck8WraW$?0Ng!G-0WqxX$uSIq16?-f{Sq3 zMmi{;3wfFV^Mua!soCp!Rod{Ot-GN-qwh~HVStl~n2(GRe2H{CBMoeQ#Qnbji@^RCOun;_16$Q0ZNeIppe-z>i` z8=vnDPRW**JGl5-T67BSno(G;=EAzhB>UaBN;?zL{Eg1&>PP}N+lbIzyx>^O%uhdw_5>LWDXnb4Z81TI=7U3){5(||dhcI>Ks{JygU>*Lhr;a}~ zuc^g_Oh4F|1X^Fqb173D|KU&<3s067u5JWBQ*C&AMzcG+&J%xTsO$9bW1Kmn&E4Ar zIdY<(5O@X!*#_x z=RS5KqapU?DGiX!3?IiO95?!y48qdgau>fq0l1KR%as0WhY9U${zn@|0dj9}!mLbm z5Gh%cewbc%rImh$r$(k?<98lL`{Uxy926*7?#WczZ=zWMm*d zDwmXIi))&u4);hfA?HphYG`F`?xhhxg~|Xgo+cJ@7reK=a9A?$jts(a489DlRIwOg zFV~5IIVOUSX}XZ41cjiY`XJO3DTL0gewZwZegk}Qnshoqw>U2~Q_8IVdzYgKqf{>{ z*1+M8G#oC3_VFG6abVp1K&3(jIxWVCVt)a$D{q1KtP|d&h_RB0Z6@D^O?~_!TaJuU z^i-}(Z}z`%qLtj>hEu3;z0J5L{_Iz)hVVSWwZq68r+mC0Rtq#Lrj6g+enmGb)p0cR zv45O*JMv`C(~PYxN20~;56Ru<9YvfV6X@nm1T61p{&bT9QNX_Ke!l`p2<;()42fO0!u zDIdD>GUZ|Azb?~`+p&&Vry{SZoqG_UaXBU`iAmsv_kmSKEQ0DIRbEns85I_8NO8!u zM)0e{q$=&jyYhUV0`OkN4?1F`FgU_}8ID^47R=pfn1u-W zMW?{J8E>vtp#Z|7FKFy7@6N^ij%Je|PUS{8?ayrrh0p8Id47C={ql2BR{Sm&{^BOf zcnhCRqb$ko-QQgBs~w-pL>Wji2CSSs;_OR0pNVE2dMWkM`+KiCRdCx4F}Bvu>pMVg zLLI-PXJ8|ee*UJ8)P%BoF32MgYJs-yU_35_9P4gAZCzxz~=b4ig1o<0)l zI9#T85z11x93wgmX5-ncZV^0MD^BVf@&y>d+HDe17+)u@I_XJDBFIm~J=ZiHe2o`@ z4Kx?%Y=2?!Wz72=NZ}WTr-mgWqySp<(%)y_vHmH+rbe)RVQXS!h4f0`!|5n6srmU^ z*12S%IV|SMg4M~$G<+3;)Pr;#(~>6n7+3&7!58)_tDN@_YFjDLVMflj}&uB z)uBrfqxcZnpI9Xw{Nc6KL-}3!UL-A>)?@+MHSh<%#aLJHT^re^-Nc6$8E!tKK_=hN zFLIz(6f)^Up_p9i+2V9&W$H1YRs*$+*x-{Ok8CK58RMWcwWfY~OIklAK{E^+N?vSZ zULCYPHkbqot=$YJoihO-4`7PVld=x4mFb1hKhmxvmUcu0a~MwNq&Mkfae36U9cSuP z7rnvovyo==siv!5oG%>~{_y-9OoRTJW)PSS#ja2JPQ5+*wRW>hdzKVgFrN=Cbr78d zO4{Z5MzacMYM6eLELHAWm{zZV-sz7&5yL;uZLzx)JVmzt-p_}!UTgk*pLsTt_4o$9 zFt1Qw2ggxyC$Vo;W-6&rIVKDi^@r?rX?{rJ(x49-EjA1nICf`;s^pT!5eIX_FA99k z@^R)S=MjDhms={mBKt23-kAi-HHVHC*<`5>S`^PWpZYj(2?ovI-*P5btlbXE`%RDg z5+0gE!{es6qRkdt*Lyi-wfK}(Uy2A z#*O5&5nm?di|Y%WfcM9#JFk7Ty*q+M7p~!pR!@!%O7#Ng_*Eq!olrY>F{0P;*(X1B+r}2V+ ztcde9slNGbbHS2D({sIO7x5GPl%!sh2va{}$7RlyonfZuFOH%bObYQ%qv{y^_D0Lg z(W3moNuZl#URvN4$(7c7XWNeLCc5<-PgU8kAN1MpqL;)X3my`JVy#8E`N8jwW>O-T>@_8-%*i( z^KGNSg@9nZ-xe+@8>;f4`wr6opdWT}8&ES|9tVfIa015U( zZHbx)+rr=28|7}6#@9BD1J%Bh3t;JbrTP9ffhdq>*V(cV$NaSax@z_RWiC6&kbfJo zvpT=tfWL|E*t8^9&yYoSa}pt)v9<4FM{~L+*gi~4A5&fYDJ@FLqN|YVn8=H!ejP9% zRI~w~;ABCrnMl?vb`+mQUDj1x8X`Q(e!b3-!($j-59wXF4Xq7CNdhu` zb}`mM^!o!Yq^;7Az{;R8U{V4Y^f!(9Sg1O;#`ZNW0T+1uTH0=7Pb^Da-r}}xlyH3c zuA2KziVKf>Z$Ls^Ovo6VKo^PQ@!7|o3keWKeD!#DE-O)$5+_99YIBH@EubaTAIa^I zlwn#_`N$D_5?sooGpu0qnTIdZ5Kg~x9v`}*RX4*$gJcIchSg>&Q)SJb52v2|%G2n= z{Omh#EGEF8j0LwZLZ(fhvP? zbcIRAx9i(Q8xo>zW^Y{Y=z8R)!RM4ks$S%r4p)Qsm`FcSkLo!_-6cTzWrVP_8G_@E zal_2n{ze|O4uNWZsWW_^>~w@QqLzvjaydnF!y3!lI$UjBv|-Drt1Mz#ozG}dx_%$o z7UCD+dkgt;G`)c{m-O~_X*ZV-;1)!-=vMlYvdH3AP5;9psO{Ap0c^Gf-qGi#v7J_- zcV%pH+P^1zAv^53E1|_#rnh6kqV7>d%(37dB7b=!*h1WehB8$ZJHI{{KUc11kV`5? z6^A!}Xati47ZI2-phZ+R3fwLgZSiRwSmZ@_8~KvM50kqb4d&(R8FagiyzvthdYqef zZ^2cA810MZ0I4gB*l~6TF&Ygn`5%$Agx{$QjBNdK4=D`!pT;0w9+J5XY5MN|XG-V| z>)#u;(set(1>8plsa-g-yR#hujK)2KWWL6&(>=cGRVCMIXup&;8mynSFq4Zusiu%* z^xp#29zdLXrO4q-UC>{taxQ_0z zxe0Yyd`zL>u6sOIrVq!-8|en&Hy0pUopxJzbfAq=>p1vkF$T=b1AP6K#IZmdV?Y05 zIj7Y)WbH=iZ^v73IW8$Tt>mPKx`j>$%7h$}!SDQKlFtYH?(q8mE%f!!LAf%)YDR;)H~Cq{M~o^WMDYz^V*b03ODE+t z6F=4U_A`v#;vQ@L%*0LB=`}ayG=!)SMGm^euS}p6 zr5dT9&S5G%Pkr%e9(N_psH4;Tl?u|IdOi9W*r8EbX#-g(&o#iuF)*O*f!@0sYO~Bl z;R2H*UfvDg7B@kHVm#)M9-aq@0#TrADuo2cxO5+Yz&qXH1BPvKw8YgPqsO8EtfQ9C zwnGKSidK9PDIcp8tMx0qBxthtvU>=w9c$ni%E0KI>C(#j9v4_B`}8L})u)qGC%V76eh|dESgT7?C7~3p z-J(Ii!C$5p@Hf1buYyl2q0$(TFBcbH14S^SrA4WCn$QZDHyt+$y83AbCXkP(0(rw3 zg_wn46OylWGc8i$<>$YJD*xq7=aSxx1eBG9lq|a=#kSqU?h>FZCzo`>`k(1;{Tm2! zwM$wH{YJPMjh-c(2(#<)Xd-NM{6s#2!e$+8J_wB^G7EkgX$*DtRjZg%Ck!xP!X3j$y=rQ9T(W7zIESND0Wj^EsN8 zSfS`EgNj%#dJa<5m%|SzndNV&n>infObuwyCA~LWJsiWzuOtanchMT8~L^4HD}k!^vNu+dUbge=?c^trhPVE8U;TjF2~@+ zM3v;d(@p&3cI#g(Voz4{&%ivi+HcTT9u1<+MuS3e#N4+aO|R8&#S*H7giMA%0=s(U z%teeow$65yZ9?A!Uc5LPlz#)p9f=>PxFfl4yy!pkK|Bfwm__U~@&+ecZaq*C1vUsy zvm-)FR46hfDvOY1QxvlB15P%y_*jZIe(di|)(`H8HyPb(i0G>oUGt>p8DnI^`mQj`Ik zzDgNO!`E5ViH zdAmks2JA!x{w1TXMJ}|x^CJ<~uco@Wg7$TMG&t>b0qK|E-fc>>Jx6u!{3vi@lr)RCFzRj9p_0&FyB%anFFfzX*!JZV|X9v=DP zJ_(c_{EIVCl6r9E_M^RGc;k7QY?9{@1Gd`68T$))(Ws;7T}vR4e;YDQC3v z8wNJpn`*9FoYkf#73hQ{3(IRn)UIsI51?B${^(oQ50-35$qGW-yidrQYYNG~eVUYz z^>+h2R~?g(m7Q!_$jA)1q~uwl%VABV(;a<80{2LOOB?G9m(K@nSVH!{)avM-tgEHh zs!M@17Ig%o0f6svr-ngQIC{vZCJZM7K`>ew8T?<&yJb{VZx}BsEg~R-ARwSLQcCv_ zQZjTm3WyA)NY@Y|(gM;kAT_kaP)bWn&Hw_^APn6x%*;Og?>YB=IqR&CVA+oOo9&gEXt%b7tGHq(QNA8U$bx% zj?Fw%Ylmn^&sh3@o-FxieCRlJY6mp)*dsh@|u|x-^frm3j-Uv9dNX2ga8p zdesd^W&V!59xlA%h90r_B$Gft_5k1|uocCt7g6hs%>lKVSnPSHP|lruM+k4%=xWwY zKAi?J?=Q54c{Cq|%(bD6vKK?!(6Bg@z&<#+tqG&}D_o&ZjJgVCO#b;KM&08<`BWa0 z_Akul<-eNP_j|pVe#!5SMn5;<~VQvaezGQy$z~gEJB=dku*A5;@ zXt$?*|KAW@+;fTX8@KA;+oAskbsw{-*A=HvOQsj$A5L&f{Gpf!Oo>AFmbjT6mpcB= z!~^PD#oy|p-}qUsK)Y?eibk8$=WXYApYh&d4UFPsdM@$0cPK|Jp8Msl_5Hy9MQ@!1 zXUMnczWd@&`pjM*PTGy&31CtKpc^x2+{Emvd(;(9iF-113-ycY@7 z@+*Br{p$y4DsYH5YyHPFPvJ@%?=4d@OoD700?g>)UDF0uF9^h?qweXpLkRpLw}kjrBK3^ z3KQ}FJ`|V7Wnt?Mn73eE6mCc&c~!s*&}!!kM*RX$H1E(sW@F(rE}Z#k1t-C=3TuGt zn03S8@0e9UpwqeNR9CRhEh1UJklQNc@B#4+JnhMcFxS-KXY0=A2t?0X{k~_6_;9Ye zvDLX-FUi*9YgaW`1u#yuOBFCoQtH`QDh`ReqDSgS46SMCX~DL7=64bvPk`vf!cwP3 zzda|A_^C{J$&{>9l)x19v{4)jgZ}J(y0I1Ri+C+gYIW9V>nueXa~tl+PY-`cr&qQZ z_4yM<2FpfkOVNs-y@NVV|4Z`?3Sam{@RqO!?>8ptD0{f_la-Lz?#({QIqTe=YTG$ouu`Yf4_7pWWnhyB{vd|6&@HWr~k+JE4KeNA^OX z8@a-l*`C^G?UQRE%uVOlB%!=U3{nxv&fGnaTKPMbgI*|$C z_V0m2j4N3I^dIM(2pTp;B-Fs;TpJ*aUJ7&FY)a^?{d$k1NA&v~fC|@Z=x4S3`>Pw^ zliuGY{U89-eJEabpIx(uv9NI_v8Md~$=4dLMB)WtkOZt*=|j6g3-EMI-^VE^3CdrB z#4+cHPbH5x;e)VifR+p>YV+d^8fG%a%v4_=*M z%bK?pl)Vp%D$l!dy%1nT^i{Jx$_Htk-26h z)>YH)d$4KmW=?PNLmYctJr)$+=gZl8K9d94+Fy`sZx|^#3Q}BorxS|5N&#?wQB& zUK+r_!yMyl4IJvqs=JL1<%CXxQkx~@WA#%~Hzf;aYfh;tJm4$~z>}x@4RhJ#!#D$Y z*q#VQ-MuXsNN`_W`%-imK4>J?e3)qta4snH&uGgeZLzT+ehcDceBoTf^x}|j6LMCD z4-tYU@xAStqxPuvei`ii$$zgbc4L@61LQblxv)M@T={oAI%KQ=r$d_xF)Cksl1Ck>P7Y1cJPU6IFTK;vwk|jDb(j#%?A|cox;Yvv`L33 zz_3`H)mQ|Q+^=AEs%u-r`}g$$IM_OmxS?UHeAJ)ipcL7%5)h%H7m;gTNLp0I=BFa_ z`gIJkSBsPn-NG-PAQ82p=M;LwbdL2>XNlT+5ebXWt=fj+nl*tG=WmS+)wFHq6P>HS zUpq35>FINHNEDPm)3E``37f^G{PHZINm#RcwPx<|Ia#{4=&GiK%Nt*D!`wJ@nPw>>Hf% zoOtRTm{27JiMUjIIpgP+^OI) z9hn2F$Wo{kABXoh$~Wcxl-f;kuF&VJJoS^DPN}=3kgL=a2_v|pJ0oHh zAH}EV$3swQybnrgurN|CmeQnZsDOk0EDIm{A9?CXZ6riwjoA6?@ zy`b+l3qY>Z)=-FLdCZi%P?jM|L`jbbYV!1=29IVGJKSHJ8xtD;O@$zayqjk$BYiXW zR|-NRKEMaLG5%$d2(GER1SbOnA^a?C>AWa^g2bc!?tiC7dMd+6ZQ{g=*GoRKOJHzI zd7h}nKxb64*2w!!vjaEgR5Pi97vDctJm6FQRs8Q}i3w%xJkHd?_p!=uChrqYNwVmf zq7CZpr0IeKJA}-&?rADOYz;pb{V*C1uurBiI8zJ(rNl6NyQKgbp_8LB3HBpyyE~qoBaa$-ga=i&W$ZQO_obLV>nvjjRIt8{iMv#IzR1m@OM6x;%7Q(2=h~R zRThLZ_PdF?o8WX;VGHW(zk&Thg|2pj?{n7lZ;wH`?H}^S&Pva?rkr5`xp79NMe`z4 z497-=Mve%VJpb50Fhv`7+UZTXBn@1N0F_dXLhWd#-V~g_Y#>@g{L8%zX~6)AsqMsJ zL6bnE5|S+Jm~pwGlRR9cw2V~X@fUSjE z_$+rD6kC*>3>F#)@@V`x0DXlpww_6&j!l&KUydkqnf)t3{pDuoE<}NwsZgF#nR0iC z5IMYv16Q-XC&hyj58)d&)yjc{2!Y-iipul7y{pmiDFt8qobzu+S__b+DDJenx&HEw zASwgFg|w_@@j6|EmnCF%dO%@nGPsO9H95Gvc;0|E>h{YK6@z;PsJOeMVN(gZey)s_ z=ML2c`j8T?55cF z1G%RR^o9#7E9oHL8F&&VL9GW5#%bUiAS~}wCb(nI4RYfd{mNO%A)S~~1)WF5Q1^rK(}DAC(X&p2K6VA6K|c+7yzb+ z*NCNn@4F3s#oT(s-ft}8jSfau;AAbRM+IE#jyMjd1neXjQ)(5P$|j{xl+qw` z(I34Pv3p_M-{(e95+BMhG}4$4p{}|?3p;H1kcvByP?Cqd-`rVkCtqhaNV~fvZ=bF- z#^c70%jgB+8(i1KJdi7GmN@1X>_cm_CjXh z-?(5!7HZCHmnrhb25n@lNqx4227=I_T^R^rg|!{8;oo>I*#hVbDPrvN3CsMjkEV-P z$u_tFH}mop$Og;jtK1x!aDH*$?ual8HMAN54}v?nSH z9t89$z6^W-$13SZTAUBia*&09I;+N(e zn%>(BR;Ri=?Y*2+WoMrzc&_MozN?<!sb z5YtkA+E|H2>m|roP`D6^GD(Sn`0?UyCLK=U4vcQ8_Ry{Ma(3K zjrRQ!zIY3!8OeWX@No+^`K0ibScTC+xr55C@s04cK~r%?9M!pv0`B_ z3Si6n*ngd%TL3kS&G+ENY$n#^23dWKn@iB$+(yOYX@Dq1yfUTP{4!%g6?b8j^yja4 z0Un#eqc@kJAQBR>2;jp#m7fsQbYiziA*oDyA(o9?v1uG{tb^;B8SW4S{Mt7D>5a=B z{0_yh+|TVu_o~4ATdMh7Hz5~;wNs?w#6$;B6= z1LV5m&3=z~AGhdUAAGOz75$`s1W#xo9pv%%yI-_tA;&WUJax`EM;?{D{fBhRfXmpk>T(5vXE6kZh2-lGmPo;NA8q;;4SJl z!(BDEiw^Xq4ST2szOPRGtV52Fb87xYCDcO7u#G6OJLNxDTu0z!f+?*@kuxZ4^}xMlOA0d%Ft4{_S}-P`?ir-3O~U;^}X%2SmLS2!oD znjkX#VIqv_kGT@m7_PhmZ>$s=Sc_17@EpQ#sXfX$1Sc0WIH1*H0U+8J9HJ+41Ur4! z>RJjMQwHyI$3fT~M^xJMwLU7+?Y?-oWIpbQ*HxVSiELx5AQgdT1|4+mb4P-OL^qt= z|1?JX^2$vd5_oD%w#%!}kwt`X6~xuJXg8G0J6^7zl{`dvt5RO3I+^T272oJ&3A^K6 z_8)TYd|GPhwY_H`C`&i3o0GBrl0}kSZT#ZZ#;BM@V{5LBV?Fd~g}Q+WkvOk*xJ4cQ zipF%D8u2M7d4(1Y^^c5JN~=cLNG|KWkF*f-)GMp? zv*#JLc^^mXP%u?^=QALBS6yvu%p;o*uItUe13S=OX|6W|*15I=U!IOd{-si$R0TB3 zH&Xvr^vhL(fnZeKkFo9dETIU+b#yb;Tg6 zyH4XV4}7DA607`~X;$XNqbJ2sgIA*4m?+o4h5WSp@9qGlvjXOR->|Ts&y2{dy5Qs! zev^oxk^NpV@6OXA3M^ZYb=ipXx@i2sS@tP$Yi`^>=F7^NZk z&C^{^tJeEQUFyTcM~Q@b zTw#}?i+C%HN&NSzqB{v(y#O6Ywk)cyB=Xw|VAdb)QHX_6znAI(X2+x8Rm><)CB7K5 zF#L>@cfICW2o=I=w0MT*v}2CvXl5?R+zAcM?G7E1I0OZGG%E?Z)GJzzzrfh`xJRph zta5)Y8-B1uU}+qj=VVQIi->Ni}b zehSlz;lXtjXsjSD{mNU12!XZa_I5=5I`FpJwdi^(?{F3omxayf-zP=v2LA;Ipax#H ztEh|>0NzAC_P-BcNAd-iF!aA?l5od9lazo&u^G6ydU?}_3QhOK?_hn7v4kwKtf#Jt zqL6s4+=%)LAMrB%lsOA!&EzWXO6bd@$Pahn8({?r_?e5k_fI$=AA40jd+T46PX9B; zY_oFSf`+UEPkD=4-=T=J)ml8#j&KuX)SunwAshv=>+Y$fZ zn=8T*M+gE5b$=7U_;}_jGW4S z8D0rZnKWrVEYt8T3SbcatWUcE{2an;Qlg=_I^P&tL{Mah3A38EOr@%S$t4WpknK@I zciIsW5?lQ7u!VJLsYSn zHNMF=x<8QB$|`bqjTQ7Vh(F}# zS(Xffjs2c1*UjyaiEne1?yvToh`?G&(&s{Oifg|5Ym)5MGWsNkWwMf#;*wPqBD?Hu1fQE5Y;aP?$Td#O**gT;G-TrcTHoO_l>cKGEPllG8!v0`a~gRX`4Qp4+h=% zG5;!xNU>&eWuaONA_(^7Is;kTNngI4RWe?yi&9O*AZzRLLXD>-^sVMEV<(nQaKIG$ zj~~%?{xz&!jM8{BGQRSV=AHk073(^V<;j!hGL|timzFH$V2(obSRkCi(e zqgLK~rlO5Y_qiS~ANMCGe!ll^*4ZmrZJ29W#7!!F!~SLv;LdG*gqD@Bh-bSN@^-Tr zq0r#}6XQd*ZpO|RqTK#}!~Z<;zfKMVd#GZXch1CCOa$pvLIeNXo&Ud^dRjz?|5F$S zV1;S!XNh~?M3)28vDL0;iGiqPF=*A-Fs1>dHMJCPuQDloBRdyH?w93W5;+d4e;VV4 zbL|XsV!4rcY|9yHia-vh#hzA0N7R&pMdy$c5!Kw`nzi1rTm9<9y{WUrOl9+PMd0^u z4$|X#0y5?MLay+R+oQ}8|1@{?8It890G zirz4nx}(=mWLTP(TdWFHV#NLDN_@(9gRfJgWM2(|->Ehq?=)8iDRC@782TDd$yD9K zEij(}q_^;LgrG|=Wb>M*A~XVb|DC`3{qLO129-Yby-Y3s_mnuI@zI#x?#aZ{&{Cb( zoIn?*bT(l?{X%pH2J`ttjA&2<6!&;hp}kI0N&kHUKy{zqm_$xICuSvEo1CNs6ja`}{^(P`mV4I}#38b;G)NoNJ;-DHf zKNT#@FGOV{%eyi0WwNAKnW+9yn~UEQ^LOld@`O~!n*|R`r|?HY?cCltL)7YPMASkD zru|%v`IXSHXV~J}AJ0cKuMrd<7>}kI+VftSo9_uHTC`G1!KH}_($&q>jrW)GFgY5V z)YCgU@5!5AMAq!Uf>r@+^~TA6KdN%LDG*yx{lVY#3GCgMCY|fk<5;E6Fkbd)5EIw0 zYgro@2dA_2utB=0omA=C_rvtb(s zf*J*+&`3}6VM?<-{<+$}8iI5)YC9MwcO}6w!s#g-8-CBK-b1~!i#yEQxU-Ar$A}RN zT&W4vf%uYO5k8Hm!}~)`w5#zg=a3(ogjB0K^>*V#RG+T&2?>Qdj!*i-Q#$^kw#iM2 z;7WYcYMVJ2r*zu{EX;{8Cs$}z(u%ZpDVw)aHWDUUiH3~@EJ})JSq)!u@j9&^z zGBpcG4GQ1yjK|kXl0vTDK8wXM;e_||43|l12=gpR)tsv-bXK{_H`&ZO z%X)&tET-W*05;kin;xtnDP(Dze+Uv^Lo zR)CMXZ$X_8u=la2a|%KW6<+;G0{2QUC9$<@9Og-e2eFg622qPLqct4(u?323?5$gi zWRO8qQ%`yV-JlbGROs-(F(rxeP#zw)*Ac);t{M&34dY{8 z-|^d}sDfsU|>7+P^EIryfdzQb!z?t@&wGK))yj(4S@wY_YAS5Ghh#Uahq*a}I~{ z3~e4lAF&ly<)>wLw^Z7z7U>N;att*7H}UgW+CPj0=}h7Q?Le<^Gb=%QJKa&J7pm?b zOZ=&RO01qX1oiQNUaZ@w6G_y+G)d1^@1oq=JEusPmyk7o9GP;ihp{#0zIQrS%t7mb z`Y1;@r$&NJ@yz7+iM7Gq2}U16xnmon6);*V>UTG;MTcbnzMlq(dd{|3Rm$(~Pl@&i z)+y#}SvKOL|0IS?5Va@nBt*1L9s9U|pI;6sES_lF)R{KuAEfZhUP;{>q$)M_Nwynu z*?eRCMx+H4Y=ww)M4uGf)v z1EkA$?(7D4%wLCVod59esH9B1`2K5#dhEfA<2S4EZxT|J1*N*u{{+21xw_r#)wFzr zVw)_DBfSF?(IsY5j`t5aX>UaDR>b*J{mxiD?sp6lBLnd?Kt0#DB|Dt^0EFm-8eD5~I-081i+T=HAqi#UxXH=5zy?2AF zGy&#k9gt$#1X^ogz4(?wU_LZ_3447-;t}<4YGWyJK`EXD@%2m=^Zr{QX;9XN25^{7 z3h_5?`c!9r$c+=Gmrj2U5fRJSJ+MO+%{&flBv^uUWnY2{gIEf=uN?mhW`}x^H_Kge z5VlJmQUrerJ?Kh1$vmYdY!}9?^Dc|nqd0{uzKrA$TfK?Pl1qIm-UB>22>J$Py)Dvz z2jkoqE*YCm;;|EP<$@!HI~ps>(I^T!K^(?Hx-6B-+CqXTC>GQHX#j8%59>y)rE+%q zpI6IbEp?_?w)WX-#~;g8xt-}Rf!$8QM*DE$pcBDH$$1hF>f)z{xEj(>zuhfpF@)~q zFMRMrVdDc%v`8Djnwz|y#?5d+iN-5z%0Pt))Yc*%Khu$y#>tGat(i&BQJw-Ik(L$n znp)66D^2vF;oFhWTAy-lg4PUA>8`a6DL8MwM>yO z0hd2}DOiL~1;`W&!}E@r0oY-}KcFm)dS$?wiHALY<3t*j~+`!c$1;$$%3C z_)j?CjP?E(>QTLTM#7_h$w!F$ut9C2&0hdL*Fc}e;Tf;E*~Drju4-+bGutX*!c%(Q zVD95;)?&@-`A~JVP}E&~6toPt6hxv%JWq^pU1NU2w^_IB5SUBGRJ9kVNhricZF~ia zp85Fj0l>2|B=pvK8e}Sf&3phDe%3ZeW$fq{wALz##DiNLa=)5IU?be3mDnl{ z02-7CnzZ8{q0oW#KZfq`c`{S%bI7+#*!QKgP>&Vp?}m?T7Tv(zgn7_s*DdIo&2ZE` zj4jsWYL*%C)Aw6kR;RE&LApQRr;f*$AWf~rA4kwujQ|G``W)=1{y%RHL5Kl*=aP@I zO_4yYIq!Gdm_rbW8KUU%@{a;ev`s&JPwDsS7EX8?sN$+KIMp5Ncxia|6809cU!1keqS7HHnsS(Jv%$2n4;>SC!BKITIwuh63&pU)B(0Rw)|n3up#%EF5XI~D1RSjtVP(Ka_OyBl+Fq|*AK@EO@N#e zSL@XUMKQMVC7EZT2L}%TB3%9B#HKDWCU9YG!@Sv?o4}q#nh&$qkNDz+zDaj`=cfPf zaE4lhv)F$D+mY*RbssW+7O9!h=K=itp_pb;?Ju|qwm(X1EbI_N5z4Z3_GzD<+J-xy z*3Jp-HTcH>DVV^TQCe+eMBJKBW4n(xyx6nz47z2E_=;p07lxLNsz$A+yX5O|GI!36OFf5dz*f7 zc^i4jI{_XJTm{FDNh2>89SldhOxub}**Z7D2iJ}^G0v`<+zQCW$1;eYI3_%F9enMu zRFoqcU(!8A`bglZd~m#Ox4Q8Lc*jzv6m0Johdv=Rz8q+Qp52-W_GO#ViIG3o(#h;A zU28x^MIVK1q3f6Lj$IsC!mj8latuq3mR${N3|Soz61g7Vii*^Abgy6R4nK=em4&Ft zIU^FM=KN(lMv0sa^GleSa}OuM_PT5e$jeW25<}{v4aBqZ)-FMV`VlqlbV$FY6Jk?e ziGbvPn&gKKgR$YoCD#JFG&5NV9k$hI(!lQxe*m}43*#l*T<7Q@QsaZb^86|b>5$<< zLkfe=`%343&rYvkFpG)z5#N)nklu!_oEd%Zo1#in+=j>JTO<={zP@+O;_Ky+`kYw& z*8o{{m92}jLwP|%qF#Ekq(%D=%gVNXAN1OI*omyR6s5|O(CZs`uE6I{)Le2cZ>UXH z^4^@nO3HWCW6e}Re%j|?+ZHM4$Aa>DLdojlwexpW$l6sGR^-fAUh-iH>2!9IHz!}^ zzUU7}zcE>c@lAj0C%fURmf}*VUwbBNAYFRaz8i)4iZfJ-46u40ALw;pa)9>TU?W8S z0q&-U+c6#FJXJL@M#^_ulaDD(p-=z^kKM(oy!2v8nDftkHV-?dAK>M07w<`tZ+G(~ zd~}(AlmcYCd5e>O7abZkeROMvN0IcnLg2#|EG1airQtfe?t#O_YhgG4iw~j)_oDIW zcLALw@yQMsUP0-(mwS+l27KlO5+s@tySPnMvsrrwxut1^G#*n1-XY*KUO@x{f-)*{ zd+I9gihx%gb*yZC-)~H*5Ky{csE>4<1Yr;Kkw=ohnxvqYD1xaZDrj;YJHc-a4_*}3Tndi9e&5tg*UO=u_&3IvaOHapM z;=_{V4(rt0+p7U%D*`0dbz!kT z(%|U1dLPTMtbI;8u^F$@%Qcq?=BNdsxQh?Ua(kom>gK#p_Ew(W*!K+z;a0U9L8-9p z++_YMN{`xaZd3nEBFp%sXEg{Eiyw>y8Py<}i%VoZN{ZHbSSrb16O+UX^NF)i{3f0N z^`}d$9Ymf{&v;3WQ#T)%%3;E^J+*my>w=#7%{fzJHKx4s79A)sv*(NtYD6WnVt>@= zAs5z!;SSbi=%HOQR=C`rj}6BM<3Joyj&LI+mU^s4@?VJN77Q9dUj=H-_nWJ~yd8lz zpZ-Zl66*XYn5;kLHPiLZr=PExJlG87aT9rbx=#<;@E6f7X8E=FS<&Jbh)KFmjFVa< z75~J$98=%lNICCM%Wu+ApNdAYx{LqFZc{u_x-!{velfyMGQln}Q$V z8=2EJ!K;L4@xJ{@Tdf zYhOj;Tbc`mW>ZUX8C(rn=r6VTHCmBDHgyZj6~`A7ftY|LT8vZ35m}IT3?BU~hvZup z`r*4Ul1x+QqXyqsjW7J?WUSk*PIc!FWh?vhq5HP4+$3a?)Kb0}ye&zP?ETPa+TwCUDjC19i!4w@Dm7(a;+#w9#A>427>`F#o z9{0Cr;M{V@81@KLMY?1$4(hp9Jz;s!g(;nEiN^_?7H`3-dOWTThyyU_vSN)H5anfn ze%|?7Qh+WSX0!j^>1iNaQAtOBDKvR$Tr#x}_49*ITjY7oLqmT2YQl3XX(a$G*i}b{ z+1#U9wTs0|*WJIxu>*U$#$hNnnb}8%l^H_ntj?H=D_t}Y^;~5%VAsylvP-S;DutmJ zZ$6jG4}uqpN|z8TD1*1TzJbtymPGm^@sVQH)I`S;`w69y5pe%Zd(jSzG=DD@U#9dc z>g1QQx6`>xQ+4*e2+`}<(zZgn*`iYD0Vc22VXy8(_!~Z}`(NI*dscBcQLlpZZt-VI zZ=rp8S-viS)jh4~<#8Ur{8V(IA&_LXiokEB(DbFmEtr1v%ppN4euep(uR?tqBnzFp zGScFqcwmfAgbB~40w?rZOr=22#m`I7LCwZf@$oXKO!Rw7+=m@Iw>l-@Yk)m3T9ieF zAV)GQF_TI`Iv3M#80WG9-0fHkxe)WWTmk65e_1Gl&Mh=K?l}8~6@>vGk~%y~@r8iWdzGN+c6ZG>Laso|eGOlGAjc4>>c2z9>&~r_}{0`UMXhCOR zBDWy})7544@)icBd8a|O>QJ{W-E*U95Ho3>yW%+wdHMxdkb%(s@>h8tcB$`+&)LB0 z8~piiQ5KToz)zV8f2Oc~+S~c^sqnhO_ibJt`^X#O3h42$i5u=QFPEVG>N?M1|cD|B}eM$ z<9a;f)&tR)7XY~CFA&AaYztKN0%bJ)cb-EWp+``OdiRzTZqX{>Q%~>B%z6ox&;&>U zo{k77y55kZHU!KbncWZ=^FzQZFiUMFQ3WU>kX6)OLehze+=z8u0z;`8n?vZdWO9P* zC%m&*ee3C|Q--sAMxV4t_e*_I7VP2L935V_t62SNwVbNtJzAaK;)YD4C@Yih6qEv# zLPDpfNXUJ+_uH@}tdkCJcqSFdL_R`cmIG5tHq5#Oi)VQf7_QC<5|a1|wdB|Y_lroU zJww*kA|JB<_L7sn_`)V>8x5Rnv10It(1dJfaW<`S5H)q;n(hFmbo%3@E~_(L8+o43 zosK6{Na!!vmvArLYzi@j9CN_0qh&c)_^PJ$YFP3H-C|SkX7q6{)eh+A`GCwSGn}iY zQ{Z`#r+Ij%a1kL@-3};AU6}LV#VFWoJEjjjDvw9s;ObX*N#iBzxAifbyEUEynz}%= z?#m#Z0L&`z+wJhooS^S=eB}bS7X{=1QSf15HH(PV80#eCKP#wc_E9A+k?s7Jj9>*>scnfX# z5H?aHNsBw#A&8f$$V`y7v+%;LsS;!uU8-SW$NjlQpQ@n)jg!*Ii4o=ad?;xS(XHsr z1Qo8&LWA-77pJ+W+55mftPY5 zH%x#XUkfM1k0p7SN%a6_lgrQTNVZ@O@w@OLgOs(JM=M&NJ4guoIq+jUf>%ScsiHU) z?gij|xB3&{_~|c96ucnvG;{i@ovsKA83iPN`h4NbiQY#z_z{Ehr0nrf@ufH$9 z30Bv9B(Efbe7;WFLCJ{SKY|pn5YBe(~f^f&JAROkEf8n_~+Tcr%aP?x;DGOM?yy< zoG{N2{Ni*3HSJw8MOtzs{Az1p-I@s}98aiU!H-J`6RH03M3kkn>FTJxDDt-#YPtWK zr(4YZ=Lh%mqH1)uP1+8^IDh% zeSg)Ci36|MlJ8Rl6n6o$qO%08b@-!6>sv?A862Y{nGuM`J#+Ig51v)7Y~o1yTBw?;{rNVfE&V;MyJ1KS`zDD4jaRizO(5S z6Z&8mOvd1yWoh_4KI-Kk62CJ;%w|mH({3PIKD>z+Crk-m>+zeTwDL4czBgdUL@S0% z@H>5s{|p-yGdUg2KwoA8o*3|sgGOjzjd)ghMxRJL`o_7_7KJ&x%8VU3y~3?s-gm0Q z^J(z>Jw5F?9OKWkK6tA#0FH8c==JItCkuU`4_@2xKOL?zwYrl>nZ|`wyn4S$ATS&fb;e zwTCd5XC}}F5nUQcosOjuRD(I76yu;svU`+a3Eg0qx)X`J$$5BI{B=>3(?M49(_ej3c}BV&8Ebkd99xJ=WjZv($@2)4l5F;a9GUT> zkC|t^&^_;Crio~A^AVKrQ6?+bfRBFR4J&n3*zdl18iSw7(|jAq2lS^C9=62=WvRcB z$w?%g0#-{IOb@aPpzP|0GlV$NmRqbE_I$5&zUJX6pU{r9ip@QfGh=peR=*Gw*T0|dy+EBorrF0oPm zv`Z{_r;0ZDIxKUVvJLl^>BfdqYdNK(k2t5&Q~~~_j|`3hsP)9{c3=u6IS)=W%GCFo z2j*;fet;7RHW30C+gBHLBd3?D|BP*)A-9E)h4Ht4G5l`G8;)`Q&9PQ4oaJ1#{>d4< zad4(8TD7+DEM9?QS2)L65dF#7imjsUU5Rm?+Sv1Wo?pQMUErOF%8b5Y=%I%jCwMJy zF!G6A2m#Oo_lr$$E?3Q?N``g++PH|)UyD|Mfj)2!R z$o^(s9Q()dnKpD_`&}A3h%+^<(9s8l*ZnCCQ5KazEEd-NI!Ee?4PWB))dMjQai*Ly z>)|7=isHVGIw=AWy_85ZeDkuzz^U0HX{(Be{0sBk7}@#+o#rmJH&*{TACe({V}(fSK)na{kHwCP}M=%vtm!P)i*>MK57#)$)}^ zv`Uf_$+1X@^c$?$!^F5vyM_$M~3q=Rq2#E}kOc0@VI{i9i21!$7>r~2(#Gw)BvaNb+tWHXc=7TpoD*70eL=>d2L*H5-tVQ-{xhAt|u4aAF6%l>0#Bt2$Sw|c#D#e4K=43cDKF~`9+H7IF`ALg_tP!;Lhz<4CbDZHNjqCP~# zrpa@^;eP60IzD5U!eo~)6%FaxO9kdJxA6|C{j8C9bz_yPT&A(T4~scX@B9%!$8ilE zX%^<}{;;=3MmMvAFE#+5gBX9G0aWRkk=6al=wt^^DTmF={t}d!R=#*>b29(#8yU); zrL6Mx^PON{uVdf4jx=1UR_@n1^e?Oii{B*ayg<)+maV(PCIMVM^2)rJHYsY~VP{`? zI@Ri*7yFQtDpz{-hmcwDrpHlX(Bxo;&s%z8X;V+xjzenwi>zUJL^;E|^UX+Ok$ntdMN2dpt{F;ar<8 zpr0%~K`y)9Hd~LZ!2metZ z$Ms9xT7NNQU-nAmAt6uT2VNKA4%1(Q_`oJLrHs*o)&m#3%*mh^{?wDrQ@f<(q_1m=inUj(T>!ZwIuBO`v?$a?_3}x!`+ylzwqamBvd7bvG(R43%9D0hL)4rPv z|HSKAWv<7387_8}GJrJ8@^r+q_Wte27Wnl5-VjPsci*ZWBBU(8aOJB)AgN4&L zHJly=^_s#)JDg$!X?if5wnHwa=oe(_$&~0gT;pFJWv+|23zYGv=!P})v-y@}06Rd$ zza$JBAd_Sj245$AQPQkaDwyomCo_m8iR8$A{)DvHvCZfe>ZiRKAZN7lMZ*dFlWrC4 z6c_V+k}=*(jU#tL!|7&CK!1s73~+Kqk-O_ccG9C`RL zg~dFZ%_cx*LNElBF`C%%w!jPjF==c41hI7rHG7i9cb+VoYr%Q?qR3EpP4bghUm@(4 z_ju{a2wf)^0~Y44)muAqQ;-0L+@O%3n_Zm!-JcGtJQ~d=!9hDIO*%q_XYF((xcW^) zO==okeciQ0d-yWomC#b^cxR0jG8ugyn4p>PYX__GITZNtCY}wA*`$Eh zfmpfFD7VA$R^Pn=fwa%3mlF~5X>jstf z#Os~U)P4z9kQLh}pk6iW?W)6triG_bVWvo~E&kW*cGdAg+k~Kshwtme!JZAax}`bj zxWFs^8;f&@&+SnAAFRD+RFhHECTc?v1Qa6((tAgONE0Hx6N=KLs00vEn)DU{k=_!D zNQv|!9R&*z=>()p?+~Q-P!f{7lkc1RW9H6XcgmVSJ7=Bpp7ZYV?ES2ql_d9*%ZeVY zhGyN0)&M$+D#FIlg_k!Yc#H4b@Wo4q|}-*fMB_BNAju*gNDlXJxL!%ITv3(LOkn5rq|U%H9TIoo)j zcJ__L#h_YiU8qlcf7?3Hs)pCh4DT$@B>?uwTm14D_Yk@*OBZtEfHNaO2^-v!~~9Y2i?rbj9-eP*mjb2 z69W;5FUWY$Gv?3mXKMOc-TDgyONH-1nD0@5P+z+H6UAYFW`$T-Ftu~iC+xFpwsdXk z8(czH3z6|QafdtMwqXRf>@)xj7MJ3adp$<0|xmo+#vnLwg)(BqIlVdOHILWRvj-2^S zk~YW+H`rSX`xRzHmbVf24Y$6UcM(2vW(m2uj9YpL@yt+=u(!2+&)p z)78bzDnqd6+QS#Wjxr++^!Ty(?mh+>)VRsh&&?cT-wW^|EA)U0^5o5HQ$9c=S&cQ8 zJWNRQt%mDt`NB*Dt~r7Lp=yqcr2!qlHS-)|JA<(lPHpot>1nEN;zIvL9-R`EALVOC zk`GL4a#n;s=X%OXMn(s``^F5$lv6F;NHJ=OW*>*h8W4_8=KXHt+gu*1yEZp0pzs@? zwUiTia$Z(tGA#lcmONgUk|eN7c|ivs4~5d-zsmD}DdCRB#J%9eb^3y#qv%Yma z;k1tY0m;QBGEB#P>98g-dt)sGHA?j(tl3FW45eBy4Cry~<--2-6@}>|q7sGxw!dkm zUq=C#W?V=T z#S{n^Jw5$VYzEN=pQO=fQ=WppO{!*Xj`?&jSW2FtmE(5^5%~coqSy_*gb}sA`Tb)6 z&)(QNJmhsAH>PRf{(4v04##YAe9Q!YtF##D*0=8o>nLo6PS?<#tS>;j^^5*+W#nIe z?THVdXP7GB+taGg*GcNdkEV>Ap2q8r?=74`k?|HysbsuGI)Qpdp`k>;m31S2;<=A5 zf|~Bm)&R%K$a?4lT~5F~C7#VtxFm5Yds-xa(yZpZu&y*W zYu4b~Vi~ve7{fxDdHb`HVnM4&K-%mBa3mx~;UyvCN~5eneJr>N;0V*QcW8PYRTyH_ z$G;W(md%!hTyog?X284a8e%cpyNma>he;_pqtGl%geWPfz)R5_Gi5O3(mh2Imn?>d z5cN_JxPFG}G|VOs3OFk|L|=%vf_lN`fN3SG*G!t=mN*C9A{k~rF}e=BefXyV89IQE zJP47f%Xxwu^p*|>Va~+*a>w`K19hn%$gFQ;(cH2a&kQuTa!z0tvFg4-T_C|kUQi+S zNm-^QBB~u2#&Z?HA0xUWZ#yBwnAn8#{*OeZ_ho_m#%lk)XanvZC;fjfAZHk>)gs`h zw^BVG=WleB^UCWnc_lLSnv6s-W=xzwk504AYi&aVMN;=W5dY%tzRm2=zLvr5=k|>^ z)l@J5B3nV#RC424=^9;oPf#WLlmh0dTnQvLri zux0M8G-8D@B6{n>MHPl?69)!*y#o4|=TA}#u>bl#zpp_s)*IaQmmWz)WHH4~eotMV zU={0IvhzaMu?wtyAD?|7TI-<}^UU*pEQp6qfO}GZ+rp?RAkZx1T1CVtCxQPga=N0< zH{XuC0#+s16vmXU-pqgQbLFDVcVK1$bgA@k)&21Kd(+J?9a}F1gjfnU!-%1Rx24{b z6m29EBGP8rTaN0?#HJ67?DN+l5S5CIDM!HDNzFQ%GTFt^2zAB^b zyO+i~IBh45Z?e2BS~CoK^xiyh&bmTNt3BIR9{Y(+1s@v#@e%51_U^*7c(J$lfndD-l7vfYLI4LDxc0t_c>qz(3 zqgHB2ekx(!Amx^c-jn*|5Tm$U!*7HYFSC~P+BWO-V3zZ=tVPt;xeu<6_`|eZDnnP) zo-8;46SY@il}($LDz#V{Ofy{<{hV5Eh1w3Ov1cVm6rXI(j@5Qn9IeVUj$qUmstfMk zm|E+bSORZ6!S-n{Gn<`L9X}QS6^#E==?%6*Yqd1FX*%afp%0t1`~a;MqVJud$?BC7 zuFffJpA?qqC6)CCheztXs@PnuvEWuPp20?NOjArb2b!xrS^ZEI`)5wS&OV^qArxX! zV~>x?#F_sTkeW~uF0ykmM!bGmbQ*n~`IptVkt}-w(bJS|ix%HU1@|m5T;xS3G=*T! z>G^72bHDwyAuP@2h5SwHjT!4BmiMNw#%;;rOjQM`)6WHy+FlkxQ=Lv;Z(A@dEz_P; z1?^U5%!Rflc-HOhgHGt&)=oz+oTs5r3dW9V>=imtuM2AJ0~U_XeazeSTAGk>r-vtE zdn@fP!?!{mX&)21SsWurAcV@}asF+Kf7WlWooxM~RD)A5i>Mi4-y+NjXu}%P-_X|0 z&GDg1>=UBQ-5UFSGjolXMZihkRwze{ERGcZQ?E6&mDX@8Wry};>oL3LPcmz#+SN*2 zWu3~$hm>)}@(5CRRF_Ak2%*FUwy&JvO?o?LkT9f&4x!~5s;hSvCoX_CqbU0 zpVdX?=TzbvMlERy(m0O~B(Xol{_AfVFol+!&^3s=qsOnD`^>McUDGR*8Fjyrm#R3t zCH#81ezKZ^iw5Ri^R&L$KJ@1_N#Q*5;Hd-pklmd4_SJP9?G3Rhe*9s0SW}k$eg&ND z!XSQ|ujsGo>V835%VE1TObPAt5&Bvn-vb0CtS_HE;dzuvh}c3-*e%evSA)od>!mHyIjh5mA$*qHiH6;wt$i=E9ng!xwo))l3@Zih%n5JC9Ufid zn(Y{`K1d{m14NMhm$U3my3ieC~=;& zQhnL{Fi*F==PJo8&m((BvjFsyEMoGhZC*nx#J{cEwhq^3KA>jRNyZ|&~Cj0X{T>&WR%;z&ax7g(cRn&9zqEhg3G&#Us z%S6TVnSeW4n4@uA!xA%!zxaEarT`SwJ^nAqPD{k-wM#2h>YM$&2YLlAHHfG0WQR`a z0*7BYi!~PqtD3|O=`Fl0o~wXg^`3t4N6n-LA@nfJu~17y50R!{eZi%OvE@Vvkj*bl z3&oZYLJJ#5;4fURhy^UH30piD7<6o{87+ego*xLkC%c?!UanIEvdNz8H!HBo1D3zP z-!MN^Pw-Vb9vrwqepf~4R(q2MtnBH(5U3~PuLvj-QK$Ir%-z3`v@5jU_4Y%)Ai|nl zO*$Q$0~$PRT|#D+>XODmOB}_E^5CE17fVvK$#M%(TdsC( z2_}?-;RSi>ZqYBpq_meFm)d`2n6`?9JwEb(I)s6QWYDM|XfjD2jGBH}rVgo_F&G;& zCZZPdXIGHh)I2xh=yz@xZA!>bWd;5nOhM)-YP3Z@NPwGaN76Yef!sN&MHSxj>IK}w z9X-8;_jwNs*6w#{#Z)PMjC<&Vcv7*G3*$HIU9o9n&<$|^(t?{*82xbU?ORq391qBOxW$BFR<^Ss(2(jA3-yw=8(NW{gq(w>i38 ziq%sT`;qE_FD*x+Eut>vAm;i8fa1J3;BBqXXn$C*h`>NMf^{ZMJ8H) zBj1>DQogL>8K3A6dZ!UFg^w2lh3wHJuv*w=2m``XQA~%jc;ciWct;gH5Ki@(vXE-g zO?W=mD71%0a^pij;x5EUC9fQ7ii1Y$xvfRtsB;hwUAQ8-0sgSuEMCRWw{R3FoKGm4 zqK|asWkvba)w9yR)du@cNumuDrc^>f(o8wTpGCTR*u7HiLPdN)j`~3-*QD1ZW{`$z z`tKtlR%sC2-scb%e+Es7w{aFm9Mp?{Sf-o?^eOXUaUK$)hMwmK)p|5HT+Ovs@TD_D82busQd!;^+1XwLe7O3-h@42+ltM{e$%~F6dZ~dCjhwNi zGkY|eHb>JfER;#o!YB$pb#bMH&tI@c`=_b>Ee#pDWcTdouH z;0PX0X7%q~g2j=C$0nxeLu9W}^}0Hf9E!hh4MCJ7Ktl@P5SbWqtws(U7jE3)sF`1g zKpJbPyf?#+v@zB{^LIVr}n#ddZhJ9bu!uG?{6x|-!4jMk!eBID9oSP4~`R+R3@K{`+F*T8uGX~%Ug`6*YGe=P6GCRWj3 zx|yG9=fe*LKwl)&vhh}=m4FwHb}qGUqX+~lwFa_)5em7X@nJbF(1BDCu!huS*lB18 z)%Y2#3dTscLUQ!uLF-)U5*JKd@4-^Yr0M}R`id~e5E_9MDw zLBiMhSH2uu?Yx5#73Juhnvw9Co3#Cxh7E3Vv&x;+4`pz(M$xGhe^@rtPuYd>)nCnf zl%zfKmfzucp1Q0oIbPNcg_XNP`j7jXIOEUKp%jfL-SDN!A*R(-=D~T*?vLbbQ}3$h zT`oEbGhTPe0D7#1>?1EpD3<;hhS=99nBV!wjn?=T!(PAW}Yos7QRFwwLP|)P!+zh~h;_ zP~eN3>nPW=cSDfy{)YlDNN)qzph~D$R*`DXXOQ5*2$;=Gg7A>qn>={9VlAyV3vDIT z>w2fSF<)|g-~G+bI%YH<9BGCJHcnCl;Gh=yy0X*Z6}ma$HE^=kJ&g z0N=sP0b7VD|H?mAxc+>7mhsCqe(FEh&mc^_(^!9a@-Io$K#HiIRR2S?#~my+wr8zAD*(lf#N@$=X+-Rbe0 zDj=k}m?~cgVx_mw2`pK~H2SRR&d|Y=$4$x>&nn<5Mo)q)5v;{iL-0tA-&aJisBZr-R04Ab&V*>cC0Lvfu1N{wb5?k?I&4 z`Gg!GRfgs4I)Xqm;3y1c-m2p#(*1X8$BedGPr3<))b|ANSP~$VbY0-M7r1seIexVN zYtu|7@xdpO)9EZIkoYt8C##X~qaaH8rRnr(1VXSV_B;HwKjxTc< z?ox%-`mQ)d9PGqYji^JT&YziR4@2tDoCn;kqKsNB)~5=^ag68y7`X0tp?AQk4JHN| zqK<=(8u@Nuvory$Jo085yS=I^DPP?R*iRDQIKGKvobWxhL+ZIPP53%8{8`ki;=A%< zEVtVU8@s_nSgTU|YjX@em3xuyMJX1fHb!G#_szI3PXt^b$!y>aHk-Vrj_hTZ26hae zJZ_K`{h>=m{EqlF@#MTJJM+?u@{^&`f62dsuks*&9!3RU%`wjW`;_b5BbzB8+n>~K z1qc^PbKwTZNH#lj%IHq!AcKP7=g;--;y%z@0LuVVVMRxIfPVa_maR(f04`wQ)&$E<{sB zd~=qRh7h{wxPxiqV}c~?O>=%kdiOV^ROVmj2fp1b9fa-!Dgk93rOFoWIgBJW!{}NC zya%iLrMsd2=7TuJivtmCPkRi?UzaQ z>gXv->|F$|Q*fhk38gCNu&^SYo{ahE%OYXDjtVUBo?*}Y+h=`db|HIP`XE?^2Pv&# zIT>&83cO+T0Io*+ujIoz%Fk)-7ct*Ji&=!5)o;B7v5A$-0on|-$n~F*_W8wxddnr) zDP-lGGG#=M788w*D@}D)!cRGa1v`570vnJEA8wmW@BdJJNl$lEZG~yM&l>1ph<7Iydq;%!5A`0-_GB4G5|*@VGg@Oz#mum~?3h5|mL_Fv-d_IfjA!BCzLb0E~p|b>KPcK7_nNowUm|(V$XoD^V zn{ncMqtHIfAM$hZ->{eOfyG~MaeiA3 zKOEB*T#0(Q!H4t|v=8&UPpd8HK%^RepfFI_@;uKN2JS*3FKBz#DOVrrt`9-Nf=`42 zq3k{B4Xm8?(W`$(p(a($Bum_q%>g|9aOyH?^v)HKB7U{@-lGY~ ze_qfK_;3VqCatjEs9l~FytRW`C^RfK&_XgQ;^B{yM1ZJ03deWJc&RSvP(;JS`b{8* z@%9=Eh;+2n@QFx8-x6{#)L?kPqN<_g#!PBYi!lSd5uC%*EhSX|vqOttf*g3|pFjD*8z5Ecd zGe`?afUvfrjLP6u(&D37G?hykoT88gJ=e1~x}pOcs0)+|wU>N4xN} zKiTzb+0E#!Un;JZZA%)gn1-p7@AHjOcCs%W(SP+Dw_wH7MTiw6Y(hND^0AM*BO3@U zFJ3mE^K>HNzN=u=8Ds@%a0YqGresgX-Yl3pZ2bj26J!;+w{!Ckmge)F`M^a0CZY5N z2iu?l4UZbUe>`52lvwIQf7*-o;>_oAtRPK5)sJtRE|IGa!?vZDk?|I@sO(J z1XpO+r9{tzg#2LFw*H8b7EVc$E`l>%?1PmWsBEEl!7m|FPF>%f0A7C{?hNSE^#w*W~yoCZ#oj>^fTH2Mm3PCE)9AW2obyYvG;0BpMy9 z_?ZisGrC9bWVuQU!Bj7r1HpSM0`hvL5pDMctvY z0`8NyGKLYLZiB>pDD~)B<2_`=ZEsF8l)y?}jampJW2BMo;zhVY8}jFW>$=LxNgq?w z)xhQzD3in==)fBj)S+PXIq^ZB(@JhiURhg?y2Dh7w<=OMUJtM-hP5O z89H+YCV%=7Y{crd2qU%c`ecNOy3BMp1u%ZSCrEY!zO3d9esi`_nT`!gG%f z^Rnd8yk~FYI>C?3NT8`a+q+}mWF6&eUzS=l> zeKdz|-mHMv(Dqqq)f2C#wdWV^ zssgFJYi@hFf2F=;viCoq;6?(TJYX}Jt#iK6u_)tCJ{TVotySA#l1WtKv6E3ayJM0W z?78w=qCEF6gB%V1E)m6dHHTKfOjLoq>ftY{@JNiU))XV1dik5cWZ%_1A)k*~_*x>D zxqU&jTRx(IWSa;mlanBaC!)ZU7~;wkJ<6S$!qKGe6VyVF8H!Im6f5iK9MK!>>5xIGLJf9m@V24d5# zV$CET@7@ugZ3^Uho~O+77zfQ?=Zdvu1WHUmx`L=<=&FeqhlM1+5u;4=F(Y16r~`e@ zw4l|VDUqS#)XZm!S^97*1(USaOf92e9TIXit)hfsXQz&?{=W6zmR|Yvpih;H4POC| zn$$StBFO%)E`@R&H=4gvUE>oIJ9NyRR`_D%IUyYZh6!rFIydwcJp(y;#fNqLob~MU zS6@XpnD|(#SHSIV>?R*Wv$K?7BpKbwt^(JqHZ;*(U5Xb8&YN=vf3WLb#Pwuv?u$E# z^c>tezo%C2ZW^=Aa%SoZ%Ax#&+wZyFE<(sYY2l;gcerT?euXeW#(#qSc!Z7vJiwcZj3a%7z=cl;u$SU2nQu>#L5P}pWpM|_{g7?9Pd~tXb;A(>*7{OMj0(K zZQnhWm1VCOHMF04cQzI4yQ5xg1^!OT9*{#fnrLa6OY%E}juweEJLo|3xrO#%L4T4QX;gGrrTibNTdy zf3t7oH#TQmHaSy}_ea!70Au<5%A=e0)(S1yJh@n7|Lp5f$*_&sdON?NX9gCpA5b4~ zX?uym^uECk>HVPje6HFycS-}9Zd)oFT=ix^yWB#gG!|Vdx``nz3-=nTM|q-#Q&)d_l7K?#dnd6dO`XlRh7UP9lh+Saky6zkR$+eoIDXN`-4r#{SrNLIk&nO) zPZu|I?_*>4Hex|9H@LROetU2D$SNzxNTT*BJGc*xK+-3;qBce*qyhq&4yMw!yfyz~ zDUfW11J2f1G;^Iu?ST#^C6zzqrwRm+DB^fhX=AtdhSfN#o;-CBztz^-iJM`8w%tN* zjA#pbeh%Llp$=N}sKd`ZF>LV}yp{VJbVE{^^-V`g_V6{$17#Y%Fu zpu7bhox3W$KLXC_t*2O0W?xuhM3L^5whQ%2aa#7(9gIWY#56_9>L!!T`Q_3b>5BE? z$Nte7ON+t9Cuzc*b|D1tjR)8!KO3CK@D4Fx)uC20=IhWV!HF9>k*g2PJJMvUw zZ#?xl$-m9YFb^KA;Z^?%d5aRMoLc#TeT8c?ATuN1YHry-_}^{LWl*I|(|%eEE9odI z5si$0iVppGQ|dbJ4^`)(@i4w0#=0TVWRo>wof~yrr%o2%%;MI?3iyj2BhAlnqqwJ) zflU3&lMI5Q=}^J$K5o& zZH*5JAC5vwWD>_3j-ul7g;wrSJBGv9cW4f*FnK}X&D{8(LIt~+(*OVF{|LJOldc2w zCs=U)Sgz4iRj?j3AJZa(Z(4l>csKZFZ|g{2%Ov{28BQ`WYmSj*g-LA**9PQ{njm_( zP-h`CxK837S+~ZQJqBw!_`0W35K_G{GJ^E%FjOzrk0bwZDFHzR(QRL(=P6TVKOmI{ zyv`r`J*pvClseu&s zGPi*dUZ@8J0L^wnOn2B_cp_M3kkGV$cH zt7lzxiI5g0I6H-IQl+Kp?7ruBvF1kHrEbsQ`*E-=$rymeOdK3pE64cOT`|rintkh; zxhoG zO>Rz1zFyr|fAzc zTtg*o$U$rti@mDf5Nn!WmS*mZ;hw{?x-3_Fi;zh_7 z*Da%ypDp_tkD;%=94{_X)TJ|9`{|=(NRfCX9p=vCpQikoYT$9!%uD;5nGfz338hTy zw`UXxnFu6)eGjvdJ6txNA5TA(uRA!(n$Q4do6Lm4y!2;9Bam^PKuVf$J5rG1ckpQL zF^C|8go4@=t}9i3K<4;R`Mo^muY2&!p3rRp<+U>xLst9J_F=S8BJxv%Z`a4J7@Z`L z6#`9>{gfw}=LIs&vFj9Ozh*Vw{X)^bHKBW2guDy$TW~Kq=4!{-{rlc1bX@TK$*xab z-#ovc;DQA-UyPxp_M}|s<`!3I{UtR{*E=jn?NZ`$cI*~pQg<<&TECayX_bTk-OPw zq^;_hHi_ytY9Xq%c<3{Fy$a~an>GZV*XK#nApS6Y|NrnX0i0oz9 z>0X17y-<)3o7C1VkGjEKkK{Cr{T)O#GKa^lu@`$zoh}0XH^_BWg@{UyA3PW0vOOwI zjxPvr&fc~>gH*#AH2->8ojSm))?2Hix-s)S@Rt1^(L-czJjIRH&}z`l&x<-zq;@*e zNBJpv87o9-u10~H6<^NG-8HTWx9*`e!*R$76P^H(dC2b&6sl+m@Duc*T?r}6z`k}I zSyu_~dM5^18ED$~&p~0TNrc+0=?uaM!RPaLkrk@m!w1c-{o=}?@U?X1eYjHw&&0-6 zO4)t$SIQHNDlWW4=r|+zsOoLd+N4v<2Gq#ni0T0kD|IrCpRkO1fIJh|66n6Tp}IPN zMM|vVQl6w?RyTBUwP)+s^p)6d5k(q4kFHi>d*WdH)(&R^97=WNjGnLAs&u+4?PC>k z(n7}{)Zgq5t-2^eRCVr*x`LSuotAJXwR8PdF;RRsWeL0vC+8h5!;&QTYwpDW{s9#` zWrCr^**Bcd1s_2h!q-WoJ0)h6h^k*mwi^*yJ4u(p<3w$4lA$4)!^MU8x;>^l*<7@JpancS6y z_2se1{-l@;;H}+p;k@+YT;%DAWMIL4;Dsx75lSSrWA6@*k^1w<>$<@ruxZe}UoHph zbzr1}3yr|s2PDbrP7GNA?qL;_y=gIUSl>a=4m9Ng0Z|W?%PvO>h|exHN;e!zFkcy2 zaQ5=YLyXh};X1M^(CuX)8o7goum}YeztKy`_o&Ny`M;P8{j2n?id+UOc5AoMu!N%=XdX& z+4K~1_sp@;8N`)8=EKGatLHDrpWf9+tkAuz_G%h1p3Yj$F&g>ivyji$s`onxD`9Qh z$qqaWnW(mzeGqM6kg@wik$$u&s01b%`RnGyt;>F{iU_khf=?dn6y==DvieLyzxw7~ zd~|VxpKMXu;CRZa#AvrI3<)`Er4@Is5B^|41ih;69kj>_gn$wqIjN6&dNIIG++jNL|_)=CF2CXZtCWErlVi$H4hdE9X}+4c~}`G_L$TJ!T9zObrJSV z;z-(7iYAydV@}5~qv~z{=Wp%zY!wqA)P%Lo05bbJD!!hmU@`HbxX)~)a72j@}CjC}(Ad!v>edm*@&p2>}Emv5XciDnh zaSXqXC}g8*V+dm9`&x|a5wOp5AbnM(o{ax$>3GFNFwhhNk|30*x@86<7XjR!9**{U zcF8c(+EePoIx4>h%jo#6ov_wyE3;WQeUo(5x27upRWj!tpzcjQe20jV8`F7~Wk%#O z3p91dmZ~2YN_GTV>2%4ljJJrMzCJE|Abd@YhMx}-mOr8jJ}Jl!~#w+wsq8LXxuZB85f}hX8(hJ6?nZ8 zdD)1cu=Yn~RdSjbJDzKq8(QIhX)h9Z{KjFS1R3P;2 z*i{F*8SeDwSJ%g=h?f`;b96qO7rk^R7KE8}bWYqD(SMZrl$;b;e9v}zU(61}PHcf4 z9-YFM)thv+W(Ue5vWm4ufRgC9x)rd)y)TG}%=O#Y!}Sqt0z&P z;l8|Xj-*@0!VPGHjNDs`ZI^r|A5`W)JNt5OD{oB0a&oa-Z*jS0xzUlt_JZUfC61T! z69j%flr+V(64tIudbam(xdoj&IT3WDTvLOyDi?1_;NNo?HqhoUr4ug?!o-X9?~Yv- z4n0k}NPqRfWKWv=9Ww^LDbr^kTd6AfRf%j{FgB5YWeCl z%69l0lce@Y?ZXoNiH2f*uD&-vy`QaUFpt)m9#EW`sy8FUz2wD{K4T^@-mv&EBElfF zm-*vRb?}osem5HBGUq%YS^ZBXzD5Z*R`{E3RxBUR=&oiBZ2=7H*0D*k@>yJ+wPfj# za|m&*0)9Of^+d63Glv!j zLy*SndUUr?DT@Fol;vW;GwUS=DflJ8`SjX?a+%1`A^eiC7TC0F(bW`lv)V3novYwy z8W8JRt0te-8ID=NHQ5C0L8C}Q+N3<<^oYa=`t|~$sGJE?!M(!xIy#uAq4P( zbo8(xwrU7+>AXNs9XFWgxPLFZ6#LtNL3i(Qt3l!$fFbYxW3NAN-Y3|zY-QkrXU zN$PLHD&XdkL!~K=?soJkHAaqyPyh8`k(H}~kG<~Pzax~LhOrQ8qma3Nw`jdYH&PFg zlOPlkc|)gb6UrCxe1pqe!K26oV^UB;pf@PPer|WwzG92+3o1{JM~zivN5Bdnxe8L- zW-YQBQ6c4-9Te-&#bTzeB#05Kjs$cHDs@i1OW(WrnIPoOw^-N!o}vMhmk zEQ^@tp}pVU?6RZJ8d`6)6nhVHyad3uP-$vPPHX71H4YI%3DM2Yn5<38gTJbuP9aV# zGu(a1ztVqEa*xP@w~|rV{<^A41e9$fwcg@JKWI|^!B zh1zuh3N?9#Ze+%VCF6Cj&~8iaa=pZ-;x|^-nwnjvqn}S4;@~}ly>L&Bw&9Px@Xlta z;2COy@b4c`Jo8G-odXR*+4is7rDly&cbqKOljeuh(n#69&UJl8x(8X3{f zumt}nZ=)cp03Ma%-<<}@_I@=}#lV~IDL(qH`2<6t<&5uI(FWm6n8h;mq^o*>treqin`a!IVSpf<$(?#4&VO>Uz^XF+P9++2Wq zxYy>4>i~WtUcpDlPacfiimQFJstP{joJ>nw7Qv#~7#dikcP41328O!}&7F0dtu2U9S`ePH@3p=3sq!HAj%@cGav z@Yw6GMkSA^!nkuva=a)&MdBk5hW;{q8}&&k_{Oer@ES^;#xvv1!t(Na+Pd>KhRP7_ z@9p=1gWJ~1YK$EmyK%1IHJ^FMT!or4Bao)A8g^82F9)F$i zjSQmWg2E~HEKl%S4tvsxLd$%*)w_|{EZk8P8BBL@c=>IVsjBb$q$(fvk+`C?e_M?S z5_Z|pj+PRxXrKTC$~q{l;aQu1&6-gn=Y zDi!S+IB>$zi>{9jM(+hz(mYukS~zhj%P;-hXx2f~VQcuzbP!TgbK9PA%rd#=ai<|& zE5}K>`Yvr9XN0k;J@--Zh3JBa?VBsSxiWi|pBwkDQ(lokJNRh!$+}>lbBC?ZD?9K@ z;cnYY+~gB0V-1h9Phwzku+r`_WbJ#OQKsMKCsWmG6>3k59^3hd8-ALaa|a!^AK-bW zp(}TR<7I8%*beayC|Bh3xTbTRIF*<3^eso|ov(JCwqxvALa|GRT*6O2TH_@r%q6FE zywLadt7mV;cI!52_YQJ`@4mFxTM7y5}$`j1*g9O%mT zb~biLA8J4UySpC`bEL{D<&$XZvIG1M>qXZqX?vT*dR}GZy*wM>7+<+fr8AjnsO>cz6cD+E$zo`TKq!t1~H112~ z_x_Sws>z>yx~%lpQgt}1l_PY^fqB`v)tstX&d!AkAfUxEYF^bH7`nJk)c2hLs6e(7T?jQ$eu6}`@xGIQB0{5m;KHHXe2Ue7ZgO} zE52_f1&OU^7+m)I9s~62ctpYXo$mK$DHR^-0ay4^O$ptB-|UE>Ssv}8pyi6p96WO`M&AANv=t39Y0p%#A7W=Reu8& zB7^vYTx1Zdd1X$}7B}v&b^U@;uPCJq+b}xu?B40hx55o(ZSh&&e@a-3&#}{-C$(Li z{T#%c|Fd2BsFlMD_R~@Ed%G)x-adBjX(E(YZC;r?Xa9Uql;qHyyh72>YG&l5!XRzn zIAlCg=E%?^>PWf%#*C6MJ0xi06T(`!wKtl_AJbUCPtn)9)SNy)GL8~biSbZ2q)vaoA-}F z*01!3dOrvGzfkwyQB8bbySNQeP>NEdgH-8N=@BgSfYPNm=^(umY0`T~AV>>M5m0)O zPCy6}1VK8XcL)$dlDYYO-`~CKy?4F$zVG_|ao4(k&6%BPiK;a=6~Cq_&@wM%Hp)I^AGw?{<|Ca ze@b)+TBVQ}+k@V}xi9(~a+yDjV)PIYz+Y|d-EF@#A!#lAds2E~IcK$_ACg>mwj45e zOK@>@_S(Y*zr00|Urxut4Q`mN(9Xk4@YBr8bqel-PA}! zz6%{r)Y@}Ci*r^59Mj?yb<()yGBFbW#bw^s?k6OCpH2a4-_)*kB&CVV|>|M_U_^)1)MRTwG%9nem1K%a(aAWRMOX1w3 zx+c7gB-|eq8zQ1SA_kUCIjbHj|%SZW_&&JIx3i!#f;e)UGg7EzN(b+iP+R#z6+E`=Gt_nh&-}Wjaa7=b=+wIUg^bzFV@|*Qz41!R zV9e?lx^YGR{ntl)GRty?F0`6a`g1<-OQbMke&SBueuUEZCx71_7C84l{C2#LsP5uL z&C9v3EolaCm=79wzT=r&?q3x!BDwA7KM2qz7_7-bzur=&_4}N_H60Yw=H^wwKesHf zn7#5dX4BtSJrFnFH=mLhvfZWnR4tV zzp)r_&&9jX%=N+_M09n?TNLH`3koqMSMJYQ;ge;qT=(<0Jg zvb`9K_BUJUW^rSrRwfPtd~0OUFUw-#h973|!9v6j7S{Z+pp++mLr1X_AS=s4{W&`; z5gQ`r<7ohDC61@C1U;PnW8ReBGN zc(8eQ6S{GIvMF3{4*7_FQRepCGC2@Fe+Q3j$>4rM9HIY5;BIrTAB3WOV2R=!@wSer z&W$dQF%ysMH&aAWeAem`*_>60TKnmHIr)Y{DEqr;*RB_qtyi$T-S_mCC=f0~qke@m zx3C-6iUo-VH$^+g5p8|5;NJe>SjyD5>mnr3+*^Hec6nCN7B06NbT9YI>Q1zIBX>5K z7`}`2-&F?C3uD?;Xl;a)A=kl}hv8IxgG{hJndVA!JdKV>z)3aU*B%C!lcmAz7aSooJ|gK}}k z7H=bxS_?>UM&^%$MrKct-B{jVYys!0#89~81yEXmlS;5AZY}=ww@G#bR`?SM_u+20 z#ful+{{L<*Clpa*a`pqrHTz%+V~RPDr*2CON%8duA!#z1>FG(F4H&0Qv4yS9gR8JMKkeX?q*TLNch zisDs3Uc;(ffFQy8byvyli7emDD1h2=KT5$se`WDvHZxu&?*75()ntS1T;sCdT5EWz zt*}&+rl>UE$g$Xga&vD}_UXb^@70cu_ysC&o4q2mL{xpqMt3$z#jaOnf?6>oP9?j{ z`B8NC2mgz)V#pYq=@csro7sR9tRrwtfk>a-z<<@HhPQQLUa>D4<-Lpk*J{t}K|@6m zxH!lCP6?2bIUzZ{Gwi@L5piG6&i1{W_c3&kW^2m!Ukg9a=Rd-wqq)=kEdIQGAzG%k z(6KrtLiI;uA>@aa;)cxQ@5lQQYk%XP5rcH!%TUszf`iV%k2XJ;_E9;cl)C+FZFBMMfCIv+<7xgPBPQC>fuK`9^Yt)9?w0G6@pivol zen$%AkJf;1y;&tVwgt}{1#|e0$9Zur*7~tVbsotQ-{hE%&p?x-(_?o-Of|L5$KX zaE7;B*ic>N$8M4J;gLcbW&1rRHji_o1L&W#9G^oNIk{_(Eu!W?11FdEQ@nA6+w3GV zk4+{WM>)U9Z$a!{zt zdg?SkYz3@)b)eQkpaA-u@Rye zkR#|&#D265b7y^cZO{V$+H-mSi~Nx+N1%3Edz+q4Dd==v?;PQn;q7$CJi|%TMK1~a-7$H3o#aq`)_rfr>LK^y0AO`>fGv__5lV?dl=qSvJkcFykRf*zAyh@@1n5-~@n;;;g@9`c*2 zeB7R}Gdi~|oB?<}i|Yzr?`oVJb6zk!D0wM-QM)-3Fm{Q3)m{*cW3xEvT%ZffNd~;Q z5{gUe@leWL^m7Reh@^iOjsQ{_88@MM9pR3*k1c@K?wvd+AW<)vtFR2<8YpD`{s3;@ zi7CRA#DVMxz-s_3qWys|!@B@hBp~!j``w4vDf*T5 za0rX82NzQXitxdpXb>?77zx6?nk<;g%$o?_2LBDfLCv~RZ(>`KiFVa!gZs&Kb|lr8P$ttA!LglvY*vW{&w zx>G>6h9E*A+4@N8pO>=m<@WP-5pf9x(!}so}`^y)@FG&heOQh90XWA9HQ|XB8@(8M{;367STui zKY*0oxzFy@B^Sz13yCL=jyJmMGizeo^QHi@wN;0$R+r83!MjQL4{xm2swV zo=g@D6z#8k9BxMvH+wRyfC_Wp$RXtMf_TU!$dO~eH9jUM=){5-@??1O6?U+yjSAr9 z|DpZQyDl8S!%!BkI2?!KNF#a_Z1{H02mszk9z?)7co-`r+-uByx@AdFjOoRVBJ)XtD zcfl_`JNY;Dxp--9r&y-a$m+|H#?(dOiaz#)RAzmSsasXyyS31hIXEHqN|IlFd1FqR zvbcdh-=`;Vj9T4B^uESf;at9%(m^bmm?)^Nh>TAm}i{9XNn-yI^nmoGsa*9;Qd%}l7T2E08k;A~KP8Tmv% z)ZgQqUTAE3V$(omx*1G_G{A|sT4&@O^|t;tT(zp=SB;_{LNn71O3E9=;jTcsWD8|` zYQO`$A&c%%7N2T2+2ZTzKscRoFc#9LFwk&R^H|jK4^>5_(ZYq|xHuSg?zE+g_wCs( zUJ+_w1fOw`E2oiRKl%w%aO(1JmUK2}s+B%$nPug@WK=c_PwyBFbQEimsag<<7-Li7 z@{2D)?2?IVb7jXa7twO1$vNI79UVipnM1k2k7r*Lh3~!!ZP-p~UNx*Tp7w9i<>h(O zWtDr)JU7^*UZZI<*K+*KF$QJu2OfvAyeX~EW#17xlZzWl7Hhe1Hb~JGYiTjKMeCLb zA5FKqYU$}^JaVm>H{C)$p@s~r#5{4T!u8}Hb6y-u{Ii@M*J=@;=Uo*p-qU5jdQO>W zqJly!^pAhtQ$%#ZzB;%1DMI&A>mohdk*~dfBU|*x zvs;I&XZ8*C#@SY4sAz2mr{#_xG8q1D;CWI_CZ3YO=}w7-=FR=|`0$VcRHX%eAs#w1 zm?M&wiW`_#BI!77cO~j>D9iD=4Xg09H9lj6_YPS$8MdLNH-3O+^tw%Pk2p@xKo?DT?$a6z7|mmqY75Ehg_hpTv$YI@bFT_oWAPV?mDSS+EA zjuC4QNshRQKO?$&aXF#Q;O7rp6l=cEGMjQ?HxeMY?#Z(o-%IjcoLBP?ZA)gD;T{zx z9f;hL=!v%7VFvhTvoNyyD@iEf;+@n9P&3Aq=Sh%?XX0#ZddH%0Dzl(RN4iCkf!i6*P>9pOrcc%^uR|F-vkjnbjB z9l%SNvOzOkkg&c&^;r^ zQzfZaN+NFGhf#{c*n;$MJW+K=4G!SKVBQG9dXb#kfjF{KD(1QMG2M z<7tBt*q4%^G7MLw*Idm^4@rbxNlqWtu9Tj!dvmJKgz?{mwmR@cX@(}YqP8jUULT?C zl|_rV5XXglqHjPK+*%$Vo>bw1z9?dd(?hEa4;Mh2OL7(3VFl#u4w z`iYoXL&VIZ!0BS@8c0%)lp$8-4o^KAq3Q_9*5)P!JPk-fREkBX?!8t5d|GZ6yUISv zF};`{Epbl$t+dHR;jBjzBH>7G*TBDdH?T!8w<^sh;*VN@p{GufzS2Pve}s&8I3c## z1xt(tN$^a6#B0a`J&(6&y$sOW+hV^SE25M2l>n77O@cA0)B@2_KOa^h-b@(yWP{RG zk5C{q!LB2IQ7arGXwTDi!(G&583^Y5U`qjPd@Qel6es+suZ?)GOw>_FMUo+Q5PX{m zkWkDCk@HH=5Hh~HW*XrQNc$~{1~Y(Fc9;tAP5nuSfXvG5o`Vn;))6wywBwYsK#Qi? z%pk)G-OIhhdm@;>+nB~V{7Z?iiUpnAvGye~00RmHiI*tn;|7Uu8Sk2(8UI_4u>l&k~Z)nDMwp5Oi~XvFq?l(Z-?< zzf3shfeqek0TfE*-hBwU8BdwJ2T116f6c6x*kKD=#ylHY0{ru``=I80 zvy@c?wW4q*NRQZ zXV&zWX+7jO7Wr%+)@%>fRcTKDnbAuSODg^AbF<+ovqcM{V2FtU{}j5#q$ zNeE^>>8|CM5#G)4o+UaE`kSd25~>55Nw3S&N-RHD@tv9}P;hZVn4;|}P!-b#PZcuZ zwW-!R=bY>e$;Y6U>uE>yLog|JhG%@Z4{Z@Eqt(ID5x(Vrz!EbRs`Sd10fvxG> zfIYVgD*-W`CNW|L+yP5Cm_9%w90|BBfye21rFM}$>AE7MYvtWT~qV|&`j_buqWj> zZa`DIH~+~u24a9!t1GYIufEa9$eb|(G-!m2Tt^MR@yFgTN<6@-a{H6v)-hW=vU>D? zDfLvx1nu6;qp7 zh+i1SW?71)YZ7u+tlZWIk0C}Oi4T>zcgoM0arT!W!qDGpoS{;j!c{1W51qX6SdRya zNabH3B$Kn0g1#-^#!_QpvbkPGPxZK~qc@mM=yHQ2R`#nSB%@3S+6?EeMC+@G`H<$V zlfURj>ef<{k|1?>gM0T1oKGy@v|*cFp$~78XDGrY_>DJm;7eO!p2JrHM|FfZCq<82 z4^S!i`AMPx{XJN8?gBO%*9Qc|x|b|#dh7Jbt;iZf+L^%5c|Z#)q@RS_#xG!}onK1x zxGN}>?}q&lcKH7A{Yt!SS=y>D#JP;lTd}_Wp~XODLBlqUABzOvGk#;;zwXC+|4SsO^?j&Mu&!8CqO{l z(esr2=sf0#i*Oy@5h$FY6d(cnO1kT?;!g_L6RXlnCd4AX3YmV+il>pDxYkhgDR1@*o2T(j-txu`d?))qU)3CTA4b;+RiH2*lKp;x zR47kZSMh5i&jeaAy6cQ#zXspf&{3t~Y9P{&`Dlc}XVGVefv7VG9lU5(o?iwPe(x`} zow*{5wcW9Zz5OxyYhNC~YA5Ynn%h)1x{}UZ)YW8vwQ*-Q;tO+# zs@-t}zC;0h*5_4zpc#6>N=lbA(nii1;L4Hj!tAmWZ4MGqwMpB#4hgzYZ6X+p6oSy& zQ`)e*RUBB@n`2?=na`D%oo&H3W+GYl(x|D4RG2ItIB9q#osjl$31}BmEzzqN-tP$0 z7OAxD@D7^yiFgb#lj(C*=K)nX$D|}w>O6O%v*vdxJgxVDfc4j10mo!*AIM3 z!c1n(+G^q@Hv_bGbqYj(v(@wMaT%}8MC;`j+kU~##*cv`PwOuqGMpc8Skj^E^f|65 zP6|G+WCOI{Dvcbt^Z*ybMSsLQ@zAbOWt7>Kh9Z#2+Rc~N#(Ie&@oS?k3(-TUgGSxmkK``^!S~Kwnam;5dGP}e3^V|Lf9hNk zGsixi;_DQ%L@%wTFo(e}gn~rp`NArtC6o@?aT6S=hdeW7u zaGzKi&di!NdCXljVs`tvSG(E(=KDB^=?cdqyl!5;%Km3t?g-|7BS5t~@(JAGEDC5~ zaE6Ir)YPp%J-hRiTjeX0ay^c1p~NdCVFl3qGEEniS3u!LgSQ#q%*PJCA<>6ZE3YrA zy@V2yb9DiuCM&z+3)~S8gH0+NvVQl0(h#b46FezCDCn^B_>DE4_nl~;1+J1u3YE{8 z;#|Rv$9s($gpne{gHgy4sXC#03uhTOswBQ{0Pmibela|Tho@0$MgauDm*e(*h3yKUZMOAVhQKnjBtOr*EErhI?>8t!yjTDUu(9(c!D2o7=0 zJ)~4k8~fW;$&<#_p@H+ZVcu*A@%o5J^-WZ`m*xA-q{|eTl<09)<5n^HD&P+A>brH~ zk4~R$GFx7ULvFjI_-~wPeUGzEO=! zy5iVX1zCJ$!J=Bvr73pi<$i%R;^*{y{2yCz`T;pt0k2bsIu3e4YU~zF0+70Dr+?9x z-)?kGs%6VdQm05s_(XCMXawl_J}Gq(ERg@XxA@0KxU8l!%bCYwiw)jCm8waIK0B8F zZ45-J@={i!a+qE4g78B~pE=t0*e{Vw;(l&$2FL8g>-k(9@3q^VIr{MJncwmkwSygM zFGYYsM}1E{8~pVbrQ#Qb9rF{Qo2TbI3v-5w3MU7R+pL3LH&D$x9M1oc z)bo_zE6QQ!;uOCq`%);;Zm%Swq@OYjx$Aqd2i|1M5D83A!2cHP>q?>4#YWRV*dbz~ z9qldGM`Rgs%qDBm7MQSQjbOJADG6FesG{`2aT=~3J)SO!S5ZNDw&RAt*MA>{|E36h zoh*%b&8Scb>Ih`Lvf`>&U+*7jZf@ms{LY;>i}2|Avg0{2Jx>ICdZ0tRA8-c1*)|ih z#n$QYJAJnPgASb0^ZYUOTP2kj^Kf@M@9Fak>ZgQiHJH#%sr;L1M~i~@Cj}jD)BhIS zmJ|gt5P?JFm3TZ{)&S> zyceW(AK1vndmU&1m|awN8(dA#^Kis>mT1ICrCMboi1fckOn{h<;?%gE%a(C*u4hW( zZw)<7vnw>$y0$-!yeGe?WxC0|_EMZ~r7ZqEa8i@;SzUYkE4|{`Gxkri0f%x1Go)~T z*qz)Vh=RW15=SQBlw??q4Bcc?TnSD|D3dvRLkhsk`^?pW_IPD24Pa=>PXwNm>Y`%< zhum%c3eEN*`ZI%5W&aQANh|HEro;j4+bi6}r#FLNyG+1s&hxSF^2>OACef+Sr@c5C zfTL=qWHs)c1W4%cqLY68OC{<`^{F&lA|CI^%;69l;_jT#T;>_BRgs!JIwb%n{Aqd- zoxHR!p3Yfcx)pDmKbq%CdTinPtli%8FuPSTmhOpsLwk2>3j%I`Px&*RUP#C`vccs|2HPcAo( zlszLT$=Dw9yegvf$llqhbdC(-6S4Y9bQ}I?! zQsagqw!0qb38zt7^4HTmKg{5;if&SKKsR_*9rRx)Lm-J)E zz({lQE|?V@Z;P7t>ng~ij3gLuH%|k%S5$sXfCgm-bQ+41J*djxDlN9Ujnn1A#jG7} zLMT)RY9ud&e}SGfyzU*+68AkN+V4h}Ig2}FUQ4|3Re`(_@RRRBMefT{NvHBx5{Y?^ zlbR)O8Ksc_p=bcu(X=tOIhhuFY{kf%f4C10foj@RzAZ=d3of1^2)G-S`Go7DiArDz zkMu1Of;MB9b+UYXJ^?S0==w8W#o)Y-_bWlH;}j7hp7?Q=1LkgzPj8Sr#fv4@N~Pw9 zdYMS>uz6mB&POF#g9dz-T?tLAg~3^U$Skl-#Tc*?mNUTGzhY8HSVja?J#1;+8(DS) zQ6Y1T;67d{raCRdT=SJ33VWnBj701cnvX4&TfHup82wSV=N>_5_}n2(2HBl)1MhW) zARZ#XRpv)MRsQ-D7ko?ihk^|L(uVO@x!7JL;ID!%82Ys3`(}capgntCB2fcUdOLZ# zsg5AiK|5dBz#8K?2TG=EJfbhJyEZZ>lwJgp{Erp(zY`q+qR1S#zfH~3IRlcN0h*>i zTX9!3ZE^q%*=JVj@;XORZmSh^glHnOf$29Mijbec3!i+~AV9MOH0bSIMTT?`Q4;rX zc!{rlwzKdA+5DUBK7Bv(Y+pKPQL_XR#xL&i$`c!mU`8rnz2kF<>YVx~6(a`NqZGN& z&HVIYEI>QI6HD4ghL7@Y*+dPkzatF}Xj%A#-?_VCR9DWP% zVvdZBZ)1ybY-i1&oA~1$XGdeqdY_(!>po1TT!56eU(8HPan5c9z1qwQy$|DzHWN_( zJ#umy{HBU83NbGcZ_`FSwDm;kiZt+}#~@gNORQiRE+Q`|W>zbU+HYR8p~_`O(hjx5 z{cIB`>Q&xp+AlC6jakDrnX78YDc2kq+to|0Tl2E-Fb1B0IUE-b|Jo~2cM~;G-Rr#a z5V-grFWR}2$Be{VAw=`=-`N@jk4>6Km`Bu>7O_g~+*thpfkSL7zTvjwyqDm4A-> za<}nypZ9Tnk1M=mHFPZYnbQiI&(RkTlJsajnhnYsIC+tuBdzxh{q&*#?9Q9RYaB4p z#XS)nMjndqiw?Y?KJTuiteRFz?Y}c5nl{D^7B|c5q`Hj7J?i%T1auQ{JmAa1X1*?5 zh-SWUEm~Vt(4h=cV<>PCi&W&LSpGc}wDK)B_SqtcT=mbku5D#TJ1>|M*3~GBdn{1i zknqWr2OLjWE;`rupQWc*x;gy_Z}X}rXg$KQ{Vhxg*)3`Al`;XcWE}JwztnjZyG1C8 zMHWMfm&gS0Hha?i<<4Uq8rtYbD=ASupCGM!XZy{yJi_(=mYPs_#!DHkBQL+lyW^E} z6OY_@$_4M^m;(9Kt@7c$l!U6u#mu(DT4iMSFY~0eLbC{U`=Zx1S0xwnMWA%$ukOGT z^eB=4mX#psrnGCV#r%3Puc17`ruB1hpw?9+-^OjUHkLWTwjP~)YBLb#Xqp-i>nNqq z8F&yGnMq`9wxzVzc+Q}tU`Jso7QAp-49#17uh{&t*MQmW=b41mbPMa&-*K^M?>U5w zhlj_(U*~`@gR|?%95QDD7&U(9B7ahDO10e7S@z%0Zy-~%WFJ{^DY>nf*sA@X%=G`C zB7&b-qJ}X*bR8IK0wq8e+};aZ#F9K5x#2&%_}en~iuPX_{+#?nNN4CBxlv{m8p@lV zeEwzdhh*cLyH5{tK`tVSN`4OnIv?9QshF~M&^XvjJNMS8$q3^Z_>HM0UDU_Q)35sV0HpYZH zdr{Am3Dhc{l4yi~r>JORn;g!qBEJ&@@+GT%v59Q1K$?1RHC#%bb=|;A%!5f%kx*kc zTM!5qG2(7vOKh;jSksRXjlasd1zHjfbLz@5Wzg16= z-yad~3uy->EQLi7eIlD?F@#5+kY`I+0%~?6Z8QUB z&0@B{X6O{2MRx}?K7?8PQ*L2&$+79}UL#fES!lj>7D3EGrt8BnlF$kM_R&#(2$q9F zM`t(V=qeglym6(GBOmAoe- zHlZ=M>T?DnA5350gdR57iY3C^{;6rRPL{<)hr}qlYXt1DdF}Ms8w0jx9C>f-{}hCO zml!+F)=~mP0C|FVF4DsxAe&0u4m^mp+WJHz>?NxxVAjH#TLbB=me_9IeK`*D+40XA z_$m2M@rCM*3$G%Gcw|#tI6=+lxsgmr8H6nJ^OM#-6A5#6_?C7+HK&M9DGIIKA=?es zT?RA*v!sX`-Do3eJ4IxQ>x;rxPsSk>d&x2^r;7ag8Jy&0ki2v0rCwlb|4LIYf*O6#| zp?uQz;6-cB7l7ufs&xM8ap6p2Xt0;nUiI_%SpfT@-%3Di$abX1N=i4Ge3p{Z<|I4$ z$s4Yk$QSZTz^?Uo@A;xcwhwx&43As_1ks4^G&R@Il$s7|CpU};$8@W$) zqwGd*WFpz%4`E1W3(yv*VFrnE5pD4dS^^bA#&!kNz&Bu7!0H7%8`i#bc+2k?Iujf! z2H#@q-T3wb+$AY<-)ocYoa8r@=HhOedW}iv7B2x&p(B3j_g5pd2J+@Uiz!a{#=}^G z%$2f7)sCZEhw?lI;Kd@am8~+6;PE`wOBHJjnsLz}qo#oFguIZKECo%%=rSOgyI5>pvve)gq+FkCOE>Ya zab{j~_pLjrc}eMLTILOOgZS;@uc$5x#wjMN3s4MXb@qVKc)Gn13FLOhYJP+Y$&YUZ zT7A((ykP(J%~k9s^;m|YDqeTSy)YP-PEmW?kbsjukpH`T?T&z7-&B) zDqlV9^LmXu3j8&2DHGTvbYPZYS_4Ue%cgibFhD%fsnG%j2 z=(F&}ZRUl#Nv5A7tiy;_^kB99G}+ncZ60tY)2`4!ID(b-_3_aMGfja+Vl=7nHrKEj zxa_A;I6Pas^^^7yWUjp!T^Y*vC>l69;-rPDsfTLRq?nVW6)w=LT>Gd-7@&kpt71?6|Mb=i?8pS2K`P9p^pB`oQY}ZQ+Kr*$k|Uv{u@6Qon(t=Q-5w5 z6%#30d^907kKzcT84l=T%qXwN;HHo7p-@PHYZtYK7yeMAkfFj{^npG85zh`aL+^`F zjSofTyOjUrUN-YXNMNIBnhF@|!eHo5mwYw&Q=ygs(BE_De2|wU zi2>Dy#0rsK)c*Y*OQFO`ByW8UYc!QI^*>WOVRx&>nq>UC4+(08r1TC|a0rUs`n|R%Li`2># zi(%mC-Fgsh_^Zh=uy zibd*Ewg~HANuC@fa?64a9z&$RZk(%LfGAHn zbuU0()kkDjU}NA*z4EgA&m920Fmv}4cM(LErr3c(NEGHmoJ5hn!};}bDn`m}#C=zg zhocVC6!J z5<9(%+Qo;eMIlnKn)|P02r7pV{-(Q~(?E+^Jmi@^^4Fb(27P3AV8lPrI!N%@KZQ7% zi+7w@)R+`OSl5N@MX7PnAp=9Ad~9#sutW?8?RTy7`vJ)O&~~g+D#w5#7D3CsK&}X1 zi3)Idw+Bf5Wmk!FUjj_K9;C$+M>?3+^co8je@s~hQTL?k3$$T9wnz`gfH-g1{rp_s zTje-U6I=_Nq!uNNgZ|#M%BHyEDxJedzM5+P_p^wVCwdGETh4=m0Y}&uOGZB}ujUKr zlN)w7fQCKteEaH4vs>kV@+=wSNeADqZvo6+H0>H=iHDHTzD6$1lA?qr_qXPayw6j& zh`KxJ3O)p7k`T{5c|pzmst1Uf%{?~V0O1)&d}H&HYAiA)>0Bi(leIb;$YAObXKq5r%x$VOjZeD=4OVg771 zMBq7EJ2|qf0J!C|(d9e*79|BsaG`Bw-Y`=PXzEzb_vLibjVlN&E|(HY1>MqIZJEa@ zSu$z^{7K(M6D#eX!~)M6dwPieZnlWuj!z8(_0%fIS~7OY_S4TR=M3;RY5PulYkpSX zd!Oe2BrXbQm?K60u~w|r6VDl-=qg6t!D-h{JXg6Z%Zbx2Tipp3O{9$fe!aYIOETZu z3CsWpniI5j*)&(Lnt#IKpwEddNrtQ(;xiB@?|VB02{E5ga;$I+!Eqn?9v?vz$#|;E zOYR-i&wqk+(*KZU1MKMz)N_!5r?)^#6zRN zK3qr|1Y#dW)+$%BKUUEuvAzFqO^ihfe0rY2$M5BQ!7D%e$^SxA3W@u-*8~7Ud7gw;Sk1{_)efLf%nQ8%CLOxq$=2o-~S=mKUtkbW}bdjPA*$ zJtL9b{PRcpkyceOsAxDuBC#H}$1GfUR>MAnycq|>U(vMj1$Q~6O2%&WAv@j6Us<<4 zJN!0CQz8EC7Um{<-UyF z;%a64SI+N5HqeSKQb+AV+zzgbB`y>;{!alJc?&#o}1bg`F-KFbf4Mmi6&W!_||0O0e|$87e(ze zFen`qZPTjd=eGDRMy$-bm5h1VL(}m8%lx%zrHjFIJ=HYSS9lNB+c0$Lr)I7?31V4N zrINwRF?`o40MxE-G>z&MSTI$j&>|JwUo;MF`7JPNu0}cT`ncE`dzurYjPTQpxGlG< z-LDJJZpVZ5eqRX4GzU?N&AU$B4WdG3wyN)M2h%`J#0u8#k|B*CZy2(4E5_XABzucU z_5dpF%i@9$ay_&2jKOPM7!51NWY_vzH5q*%;xl!nKPt}|ux2i=+{P{Ame|YxDjLY! zl`N1Re*0a9MSs)k^G4{|D@5gd0Zbi*mY&dQ+K>yZ-$Y_u5@|wZxXYQEQ`nAHyNYDB zjw3A@pHM{0-7P=z7>T+Ya{ITV2^%{snT?Yev8*GMkKtP;VrCUIAD97JF*7;~%A5hY zrUc8XYvo7FI!W*hn1XD0z9r*%VhAAI^{PsCZ;~$Q6BSY_#}>BuLo#%bR9!yjUsawh z8UIz~nYG+w?>FqWrcD7|`%|I}&YHrdX>-sR+z!2lY<7%{sL2maxnUZ!wGClb1N+aC zmLEqDvBYiXP7+g;BkpWFAfhbv24oDp1_n(Rtj zLeG`9I!JDDHC~Uqc5Olf{oebQl=r&7&MO`26dtXu3Fpl_>ZBZK3MF-A@~g@sEzbBt zd!G9fAI|3v@C!i)1p2pUyB#OH_Q33$k>aD|z>~kJjxKEuS9(R@lA)4^6x4@^R`t4@ z9TH^u`&78VCzr#ou4pGE#nD9QA)Do|wI(W!oO6xjN}|hdToXh#X3-Z8 za@sP|OEt83w0;wvxL9<{mhryz57lR-ve4vS%tlz8-ZfxOfo;17^y_W;ub+~6 zxi2M* zGvZ&yv{N0uoo2No7!s2r%I9OAYSoontMQh zol{h#STsWuh=NwMWv+3~ChbkB!<*fM6sl#*$}fQu?y1S|uKA4;_s166GJXQQQLFjp4LMV99H4$F zoSlhEu}A-nb~g`)@{QZZE5v`Jj{CZ=`@TNc=RD8P=agfg z_t5kDDr`(=Oghdotv$1$L$gb1shK_05$#joXufRi4^(rY#PaevA}sg+csA9VZ{SF@rB2 zzuj#8?2#0DtN6rB2QU6h_-1RJpV;PY`E=x%hSXyhWWkJ&xLd@^;Je=o9d6J!ItnVm z1DTB4l5vzbHeKuWUK%~PzlZr-=J~0}G3TF2dt4_T-^U<%5B*o3W}K&7f8uGfva&g+ znWam8VzTf=?I7>lq3PU;?EFHJ=E(Ma4s0sdBtS;n>%GLgS^Y}QiM`g8CvQ%zQGDwoCFv@uMA%{72g?XzctBm?b zI3_T-3Yz}StyH(#Lm zn@coDUPOZoxW*!0beWoSeD(gDXzOT#eQtmf5-?J`v0Vy+X6 z%9lluyKefo&CzTBrRn6DpQfdkt!nuOwo!H+MBxkkeDS0Vxz52BN=wbqmA7j!;UEu_ z9;myQ=nut)ed|N$`-ia^cTtnMhSMZ^D&+3Bm~s&a@8U{}u8s5ZAW55m3Dk8Jz0u_VxU)Bp)S)hO-{eHAi;JY91BL;$K#35;$;86dDV? zK8KH}F-h4Nd>V%t?STGn_!}H@WVPK9i8?Y5wst@Td0hEG>J|M};9r%6S(W8S&dsNd zAY-M(#`?lVC#}8AXR_8%zxepniBrbG9)fGr1&RL0+;03Jzl*MqOcu^tXUhF57!id# zkN)Bt370c4?LC`oSXZB8a&bXqBGm2AO8aBaOZaUB)n6QI_`6cQ*e)fDEGDVHzd_t> zpMMVz@*sm4vK1Td1c%s^=w{#`9=gSTIb?F95=gYOzGpx%{ zy3JdjZv!Rx$+Iu6rkW6}9Tw5z!FsV}?FH)1Q1#iB_G5Qgp1MFG1wPB$PjyK<4gXd_ zeYc$)Ze{PQ&svQ|N6D;wD`>86DKm$CfM#C@b&aJf&3_o;r5cUASccZx{%TfSy^|=h zUtKl+6`M{2S^M|6!#o{NPA3bB(8w*D4;;XHG6&xkY z<6)dyC-0#O>|qS#H-YcSFU7q_{(L4qHXN*5pSTmO=4}f_1xD*2>g~f2ue~U1a0VSE2Q9hyYrt_Ct&Bw z(z86fp|AbSsutWm30e1U>qL^RG<)wqvooz)9lUj`1UrxB`}Jx>pL#@5<#b<*aD%xn znfjoe?p^SJe|l4PF?`fY$INbJFvRn~9P2x%waOKI-%Kdh$uIx*dTYv5)Rpu6%OSB` z8t8k=PU^*>>-vvwPGIxSIo1-F?~kAncQQOLNgaA-n^|uTRKU3}i1)3!^+r3{~>dH^#4yz)-}Y!Od4D!v)J>H|uKSympIT29`b4j{oubm1(u@`4Tun#U1UY&##Q-4aaoFqIz1os5Dg>@Tp^eO!b z;o8#O+C7%|Q~Jt2eiThleJ9*$wrp09zkBMUSjf@L5zm9aW?!ISbEvg-i50QH^2Nv= zuh&U0;xp|KxcPqzeqrKT(I54MxjSi zcqwgdjf){Qdm{0m?^c#CP{uF6s76X0nH%>Q^(nob`NYfoi_g3k(1#79Yw)@kirx6} z1!;lL3p{U zJY7I{+JGABx9vS(N;^}(-%mC9W_ae`DJ^q26dxnWh%hP(iI}?y&#kSz=J&qt!Jni- zOc)*`9ABr%@9w@!$`1JeRV^KBT3xx|>)co3t-<2;MOJ8cmp2N}!4VclrZZq7wihy` zT5D^E<%8a$j@a5gL+8pj#^Z#!U{p`VTehea%Fb>v(o0NLM|@we$(!v10+-R{w~$cHXHEI16oUK+BctEqC$muQ9?Y2imvH z{Ps-KCLwq%b!j7kt0cr}q zPpMh4`D?vQ2EQGRmlgEt#F@DIPLIjnxizM7i(N;31zpMV_(D<5(Dd(1G|_)wl&WYNc*&ZQfWAqVg*rf_ zHH8ZQt{csvi#1&>)eVQeG_hR;q(ddjl{MFMNH1SP2ysV@ZWXhN)+v25R;m$vEr#^t` z5I^&rElRt>v%p^HS-=d*_^@xp>4p49>X!_4p_&bB>Qd<(+V_diC33B})HG&Aus1h` z_4m*zD)}>$(K`A>w7Z(vK4`+w&3kg57KsJ(Lq(bk(aa4N$Bud#!jlDK>PKB4zjp?n z>w%7tWz65bI_Y&;e*raJM{9>)ftln`zh2kB`>P;q$|#2a{Bz#l=rBC9**ZG)y${^9 z%8eg=`sce86W8;I8c+Km*|D^^W+E~)=@Ap$+t~?j)aV5E4^%?lq*MaCO=mT#?DP1q7Q z5`+vmFp$b*M3(r8LsxVL-+SZvEyx?vt}GYMR?W*sZbB$F4^*^E0$rLSCvHUv#Y1hQ zi$qoOmSC*z3K7>5NMh;pUXoTlD*rPeGV4X*ii}RONHSEv=x|kR%>HOee~U+HVnSn$ zPxj0vVWMXIMxv22HEGFOD0cp*UmKFFJe0G=Mh&S7!X`cl>csK)KvnJfe9EKDH*z|x zntkLjm;+vpMHo`<(WY+gP-6;;0@~DG4^H2^S9e(&+KN7zziO!NGq`m#A^GB+qC+;B zE7kJxl2hO9{;0l1-MU46R2WccKgaS_^S)nd<%?jN#7bFC5Xt|lN~0NdtSXkbLxrlE@pc~44wN3P}6eQ z^QgF&4^qolESSMET6c!;@GabE?tTzevMQsGB-TSg24X-+C3~DJHhocjfoB$fL^xb8 zjh&AOUb{;n^pIjJ|3DKi3w-*5Hq7R`0_&*3K6Jvdp*fA_sJg9WV&Ut9|w2Qqzal_6v`@cSB zt@D5(iK;roWlZ3x0Y(dw$Cj}&G?2TU$F~`U!4M4S3$aB`n$u8c96Gb~d|XgGO2?PI zhgX5(*&VLf!%N}hmc=OckZ%9HM~G)nW;1csr!Ni(ZQ<$V70#bngRi7jC(n%JT-|>M zX@};>$4>%mM2Un^^t9wM3;13pg?SjQv%Jo~K>qU?(mqAgL?`b>eiT#3{|=?ouuIw9 zOmk0!A<<@;o|vwk^n`mCFKBSa^YSh}3>y1fcp=|;r*j+z^NhN+DKK(kxzkzPx_64I z^j&Zp82dBCA-~U(GLsoM-vqke!)vwt3SGY(xstxMbq&}zqBgzL>p5K^KgfN?D0!wE zV7mkt>rqp!HaN%B9pJsy!l=9Tqod?`IupP^$amZw#h@KaPt@CzH;fVb*` z`V)Fk19AW`Ag22jW_V(^Y^3jTC(mpSOt0(oNHmeOH>6_c(L4T@#a(s^xra=7P;}o} z4~{(Q%Qmy!Lo-jxo8{-QUaO+R0Eh^qYlMp7q}B-9LJ7SQ{?({%v`&~kIp@dSsNiL2 zG^UJhQm)j%coD1LMPPh0Z`5 zD2g|&9(z#&o?1WVcMaKdYW~7z08nl~Tf>G&{2?AO<;RqP!k`I$BUlF!MK$!NAWuA?kuj?=yOTYJ$VgLwtVaQ>HJ|o$^zH z!(s84(y!r~+ioLY-TLvgfn|yXZ?;t}FaDSQIteCEmUfw|K|)vw68Ad-j(BJDhb&}+AJowrZ*fjgp*8N*#X zp%5!PJ@Iiw-gf(|uWbedL2pzX%hau@O<(`TzBLj{fe$9p&PpLZ=~2uU*|DCuLMM#2=D$ffo%_y(1}G(l2-)X zaew2ug6VX(TFAM*2*9ClN|&3xbfv)vyWQ8ag+5)qJQ98ct|F8~FvLJs%Ey+zQHr0( zV4j6a+|>6R<7(2iN6AW3S=b zvJRj&mjEX2$`#i7?uj}qZeLrf2ZdOZ-Jcio{AAvsG zXX4r7k+&ruW2~+L4xGB2keBMD!_;UuC{r;>o*s1-@)v8al{jh|f0gGNu2AQ9FiBPy z=XHsHH(Z*%RA4qX8CXfkP8uaX8lE&~ZGOnFyLCfayIwtR9wQm+qJU`prLvNlq8Vs0 zD{)^t31X=91b7H7x z$@5wy2y{{JI#=mjXHq>%M*q_vbH$6-a9*NlmJ+3Q%n`~LH?c4T5F)hSNcrOYen)n= z<92E(|HO>*Uez2Y z9v(X{0-j*GbQR)v=W&U~s+)lRzFvU9KxMmHJA2<-F^Kj1&!)jw0=bx!T5RH z{AZJ}&R|hMts9Z!gc~Y%C#=c!|HX2mZ-YSXF00Qmm>VSJY%ApB)*d6|eZA)nes=K! zZ+~9+!j;$-+Hh+uHS-?T7l|xjzK|In>~#4@VUh(tjqyda-kt$msNn7HIIy(HP<(5i zVLN9?MtAGi_jZ!v7zgC%N2Tk`-I}3n{2Rg$h;A`4{k-QT;GK8qlLChKJ3<(YYqe6l zG6v)BtF%H#4+__P7LB}Z9``~v%rAb7!rqb1cC;N%n5Q!#d)uGlRB5hvg~0J29f?f3{^TdCy0~eT2#l?8uDKWl{gRkIr}$=ZWM%|~ zl>apJ))?M#??|ne9rE)@KQMG;_4oQ>Qc{MpY+^>~-I%5ehnIjvrAER<490eM9M(~a z{qYF>Z4X?uJciO{y?~&c<47(g3{t+){#K8o;P>t~-kKFj?HopxEo3`OYOdZM;`*Zh z>D71g`{#tTF~EcDs&spt`SQ)19LXtbqfBQE50k%ry}`eD60uP8XLhJ=a~MX8_<3ZjVl_esczr9aa~b((Fv`_h9<)PTC* zH8=s?*Sw3>cEdajHyiGxht_in1)WNzr>ZG$$)1xve0OTYMwe{Lku0PlB|};oE%ZMP z+1W*jEU@d<6+PuYamMX%hCRLC5pL3Ss;&k5pecC7IJ|I)s`L~Paw>_P2QhU=H9b8K z17uM6;1%=+uP^h)rh&6_5V~%d)NkQW45+x%#*yx@;Ag$MCM^Uy_tD2PXE?o&d7s8GTQCHcEJ~=IenvZ;YzQt3|J~)vE{e`jX#*f6Qt{eCSAzC?z=2 zj>v`39P}SZ@=YDN_h}m0f22$QIyGjFuaNS5<~~59m{oK0`O#6{SD`PRf?9f@F-8+M ziKv?4sZ&hd>mhSS#rkfe>TNJA+6U(Bf(w_Jw1mSP8Zl_3`Rq>8i8eEM@ZoucTWQgL zOBWz}l3ZzQGzFx4JGH7xHL4QRvoHxq!uC#%!*0>;>%lh>LYa!=y03eFt#Bi5``t?n8)@COA(B^d${L8~)$nBP8m5B(Y|(Z&Zs3e3{s7 z3@<24zt^r0@Ax?Fif9Zm)c~v9>5nOz{Tv8!8%FO;mbrP9>;_#pNPfUZV|c2!M}6%L zXffF8T2vBo+v#e$CU|3CW=4_XxmOS03>`)dPw^Nw`9LDR$n`%P+eVHd^1>Ml&Y#Ub9QX%|&1Z-Q%kpxU47Dpm6Q z7#yi2q`Ng_HK}$+*I!&f{j%!Dt(kJ=ack`MvSYO+F_OJBjMRJ?k!Ry47@wx`x%zqP zc8yBiMd15pB{bEOp(Icwm5Cb#w)#AMu4FeL>C~uU=q1I=^dJvQcb`#^{z~q1fW-he zAGB}I-2K(d;~OfWRS7l9ewRH*P>M8pfo#FMy`Ltx(&=hlARSPXur+@aVvpD2*6++Q0t$Xt3=Q}Bnf`7JkSM* zbZTAGSj|tQ3`j!GQbyQ*^U{NQx4|ytG0N}abK$IBL;4l37ZxhcH~X*r?uQtvhsNPQ zE<2y>xseGm7ztL5y62rAUtrN<*vJ4|)89gR!4NH3#|honlp`6eaN&TZfc}ab{#MNN z@m95uvlt}&2I9O!f~P3zsdu73Ky|-%mJ;sR3V~X>3%xH21--S6nJhRvClKf+ zX~+gW_5BQ`Nn%DK0cP_EO_}9>o@DV*;LBzpQ5Av+>HaBG_qhm-?OvYqFERjA?@F?h zmTCs>P>z&o&+hJBiK=AAKKM+QaS8Ycv^TP61}URP(f8R)Yfup}w7op(uxNwXJZQbC z@q)Zm@_rv+C3h~xq-orcawa&FciuLH&;ZP)=(ndh1U+zDAdXX2mB@Lt@a&P+Ts7@% zMu%m`p#~Yxpfnmd88OJ0{A&s*uQL9H7e7!E|5MQX_edqm5F7Yhnv#sqL1xh`p~cFU zotl4A!xvM?=!nq6igq@-=utgu6B?c;lJd zN}x&J`Z)ehaaR|7B@i-j1-SLLVWjt2_F~0uGSm=YI6~=0 zpJw(v+nvh*+6)m%0<-7rB&mvD172K#XmJuewu$?-i53@Mo?8Kyy8-73a|0;S3cz4$ ze=cg5KbUY;U(ZV3l4J|Z&&2#OJ||N>5>DQ8O|h%Z{!q*=&aI}8yuZ9h!O}kk4r>gC ziZ7U78AUyA6nDIU_)ML?Q|iQC`pfl+np?>w`?HG!=LTz3WzGt#(1X0K7IWY^W6z$j z{Z#wD>8!$$+&F2JKK-nZ?JRaXF7@^Odxg<9T3v(sp7~dU;2jUw*A#kN>2F}HHQZGG{21V7 z-}V_XY?_`Hg-q~eUqJES7dYRl(7(4IU*yy>m&+;sZwieJb>6+-%g{6Bcxcgi(Vx-d zLS;hHtl&w|ES_MWE>OSDIE&9>QPs%LPQz&P-K*Ddc$+FuoL1hvOUt9gzce-AZHOeg zZGL9BnO!vdD~<8x^Q2KN0Sv}X4%MGJ67Kz*IY<=fJC&L7FLm=cP6n>hJEtuG*l+>| zjL-LhYC z!LX?|5E6TZr6ln7jAk40g*l47|yegP6Z z^PxovgE21&VO~k==Q-rk9%^{e#_*-#4jcNduwfGyRZV=))dY9uHLipM5^+RR#x+H= zr|3a(fPte^iDe0DFwr`NSjo15&n$ocJztgs0wD(^+e~A6b|Uzw#PmknCF4=lp*seH z_v0b&{JT0h;egXo1_%k={QV$8lb3qykLsUBkN;T#N0UA?f@p6V4vI@cZD@Z;ks7Qy zu7h>(>qTzLf4x`tYB4_pcJ`Ufi#=H$h#PrFu4C=LrRp*JPl?C#Z!Qp;f|AhyxJn<6 zqK0>hW<56fYq_g7#SISuX#=O->Cg0_PqV3FJl!}og?8AazGP>S)4Q$Gt8gbPj%3M{ z00O`{>xm>r1?yqAedL1v}yJiJ}=rvz4bAi@)s} zzzgu>B3f&mc;m3zRp}Et3G3|_?sV?=eR>2yZ}_HxJb}nJNvJjx4%|kWm5mw6Lie7g zA7t7EH?w_z_tjv>6Q*t{#>Q9O{XdxxtF|j1Y7{=ZkpAJPKD+p*jw}%d8uFeJPeew9 z!03?c&&nqDoW*1a2XCb-qc)HH+Y>K2-0AsM73G54VGUd6Oy1#PF@$%VQI7AD{~C|f z+TNeZ1Aa+uN{SoT(|i`}(6Lc-zk4e$QixKFvJl_w_) z{iM%qeO7Oz_|9v*ZZbhCUqCb}K$7v!&86((bu(>u3m``eJM2OEkA|0GQoH{4`tD=l zY|_P~uPG1mFR&-~u3lZ}WOlbDP3|c7EYICAc_9jcn!hwV$JD*aHa|h&U36R=3Nn3E zQo6S&MA6;%=1P;j^whp%bjwn zZXwyaSGhEW>NCzK^Gx>17#>!9E$9dg%b!wGl1!QQ2Y<_*K0<9uhaW0N`wAbte;5cm zl}ej+?M2^52t#p*q4Hf0)uCjhmWg(y>!djA7$Jd=;Km%XKse-Rr#K^LH3=yPVXnuG z1IGJbHgCEuv4LQLZoBtmW&)MU)

    (2D0w%%O@ZA4& zp;@1YUiWJXef0-E?T~GE-;9{`Vgv6TdmoH9K!H$}cKEN)?@9CRZt4vmozw14-xMAb zbZ?dtchU=VfDHOfeuDeyqk_7~t#4d+By{M>`{aLxO+JqX|KYy<3&7ym4nZTc-wwrM zqaY0oM@Rd2!^G@ZwX%N zYiIxDn(<)lW~N2|F9$N}9Nk1D637>@M)diT$SFNrDU^5U%B&pi-(`C7zMZWTl_ zlMcoVTnoRbHFGiUI9rf?b&iIy_Xkx zhgX$XS$8%AGZFSHv33_F#q%P(KWy zzPhZkD2i}RSi7uQMsGnjv2DtvjJYri)NKRS=ACi5D0ZXu_z@*Rz~)t}+`pvkc2LW| z9EZgiOjM>w$G>v>(#~X@jAi6J^68o3OgCluY`akN%A%f5bN3GbN0aKTH{$T(*L1O^ z+1m-vMw_J-bk$r?Pc~8W{NB^46B)p7Kbo3$9(F)Nr{0KT$xO$3o%AuIp|e{uPnu(k zryrMj)kY9bNsJXI^t@@efU8Q@$+8w)8#Um_(Q1<*a^VI{E`F=f?f!T%uyr^6Rc|7} z_`H8A<5ykDrDCT}&Pyr>mZBxD5tbozm!8JJVyghRz#TGZ-C$Y9`+AuZq8Xl#9K+ma zeG}O7IHR;bCK?;@zdtt048=-b#`8+I)wRkzQ(PqfRLs-;40EUz?4(H|{jq?O8|N@7 zRhU=R%Q@~uQ@^krcaze-HgnTq)#}p<7(Us&Z8jIU@%ZQ(C)PGRN|k_&bDVeILaBHyF_IY^gN#ryAl!EA$CVE6R(tD17X-U|*$RG+dN4sYyF zpz9BJ&OA+xKhyIfpJVPeH9nTy0!OloAArc_7{F@jyVsA?`VlWSB*)F?o`C2aAvSj_ z-~Rsn%{{^n8Gyk6OG~0f$95sC@MK`*$lzbT{a}#i%rgT;3N~mYOh}AQ!}AL+ip5+w zdAtfl-XQs5a5Mq{3_$b0i_sE|Sr(GK*bvD}NvUD6n#4uf5Ji1b6g0y|5B6>f=Z=_o zem-rE)wQ=GDe*19VC9FuMxUdROAqp-4AcZ~}25$&2h}4UhN2M+j(@S8id}!`P z0FVjKMW4WhK{jqsCW;It8Q}&-d#6IfxSTcm5jRLpK|5_$xFsw81(2Cyu&NX%+(Sfq z+1Xu+1cr&pF?{eVoDS&9Op{6owQy}ghWHj#pv}cbmd(UCslvFHMj@e^aDeLw6B$0I z3QL!k8kaHf{0R;}Qs$+Q;uO4-A4yP%IXJtz4l>b9hXNdKK8KMmNK=3hMri5yO zCUht^xM`0*eDe*{HdBZa|3Y%#7yS=tvPv9Z;SBlmW1@;EXjgrNiHNM#5ygjRRh=)k z3^&9RT+K^Up>uMH`eiIX4KOqr zp}6S7!HN(#V-b)K)s*&YV;co1Azc`DV@t21+2XSmQd&Mn17UDx-JDryNmKrJ9b%Jp zK9CFps^ne>3F9Q3(i?fjQEmd0_8Xax#uGF|Pq2{^M8W!xnfMK)B4f8<`3wTC1e`yR z@$`k5M=6&NsX9W$j|sd5b&IjlF?5?#dNCYtVyw$H!@w#K#~PiS=_gxT#o`3`j0a8K z74&5u_KI1dqW4f+Cb$Acc;heoO9;jkG6E=~-WM{$o8;QF=C2u8=eA1(tT=xtnVq&~Y zW}L7xed{~%s`=ujNDC#FRalEg;}>D3hIWMijukJh+Ux`?hL$lIrg7+cbTe4~rTj=2 zHD5|<2NO>Is$-q{9c+A><4!57e3fo&;KsYNwm2ufW<4wo_?#Q7b-rY=v{%T>&-fDK zbY`k=y2x>f;R|K7AjV647g#tg)1##VyM$Sk zPC`Y{_{Cj6fPC^D1sp)O#;S;ZLCAV5j{TEF8ZO~21Q~ECW*6NlZRzgOZ>t}*pgYBL z_R~0*OND`@gmy#o4M%>=s>Iv1`8PCou0CI^^DWq$-JZWc_;)*V)$a#tN&z1W!(%}A%kX%o&{qZ4738$HHba~E5hoQ>po zeEAoN!B7TaeBUN_8!{&^!H*{sQWG;)j4rZ_xxkX8w{AyO7cB^8#8#7+C0<8h@RlTh z`DI(NEiN*bGX6p-4oJsXL|zqpQjE>O;ThL_B=9Vk*2Qg!YkZaGSi8X(CZSOLXVjGu zQ(|HsRBKfb`cm*}QhAesNc)+M8LEy0B-erx(p0`-mUY2Ame4AaCnBTQvC*+t@a1#rB4n(F?`izk^kYfDi>VL5*V-bRA{MngwCC0T-tdu4M*i zm%EH-cr0B9Yob7;7v=0)P3FAbu93Jhp8i1uwGO`do zx4+ zAsTl#@-i&dJS>wUcG((Hr_L5K=ymK2N?Z#?TpNe^Yik1>#aKDvxIS1MBL_p3wW$08 z+v>P@Ju7uvu_b3pe2s@VEg6vm7V)uVbe2H8F%XeoY|{#17e~;l^4rX?B5Ig~TII+l zM4cNBMIKP7I~byPhlN^E?afz%kv6pc2uXL{#FrjWKUvRYGS>6?#6RJruSK{YVPvuQ zp|53Kzx#4W{N!_#OB`cjd1)W$KJF?*4#5;l1}0XYFYYw2YqsYxwqBVF&_N2FftsylYG zzK-E)TFy}Tm2u0M`81s6v=B+R$O9M4xqFB+CzkN^OFz*OX29f%T+1%q&8}d|m|O9~ z<4cf|%a)nUknc_*|0R;Q_PWK8TZ<(JJJkcbCDSE5r+QMDgYsQHr%xkiUh_$2;y?eB zSOy~}Sh*-U6`MJIGFim2ERorrU5kkcwuVF#$ZPF(M=xQ%TEo3sBe)hTyk4WL=+0}g z&hOHpz9zK13yDV&F9LE=i&QY7ED6v5kO)4G9=48hej|t7A<--644liaE7sTf8Nbj@ z+TTvE6q!rHC%(H%*rG@BKm1$N|2{2)i3ebEwHOLVBlw^2Bl<6o?C;9|#1Gk-Nf;@ti48nd<|PS{iPGwKAE zXJFfQzj=Y>O2KB4wA{Ir6?6qF@N4ih_z19P) zCxp*zcn0k|A7wqr)3l5`k2F9aN#y0oP}tcTxBs_7-$U1_X%zc48&TFlIkkZ?TX-o6?8ALQYM0LQsAxWirJAvn)Srl4w1w29Kfp zZYE6dWv)r!(hV_$gGk^Q?*>OXtj&d!?aZR~vhQ~QV6L#kD9|Tt56&-tm6uaTcX>`xn zU;Wg+C|5-{@{t@!)xPE(;jG~|h(Jj>C1^E3v;T zv(+b?lc7!I_UENO&6X7ov8v!a8oEvw@a+h}OYVm9kLB;(>M1+8bx4S|JU&Zkz5lJ% z1^eUuQ#J?ate-ID=TB)!E}q7&X55)x)$j{{(yAyaKH4l9(+}t=wM1gyedQ{l zBu2a0e2O^6eso!nR(O;`?+?38n4{4%eODwF;`qB@TbyecLvTMa>7kp}z&uZ7)J zzscM56pNY7HzYe*zANl|KlZ}3XMMV#NYMnSQE~sU!CGr0#{97Nai!402I9KdLeNZd z90PV*RwiGDJ6CVEEk{#gwakTL!IkzUS(}=aKtaG{01n|o68>GO3{ke8;xmJ{G$EB6 z*ZlAVsC$=WVlkXnlLeDxZGPa!%@r`vm8)E$71!@~ZK%pQN%SgWAifER1p}ZlEg}$P zde`m@swzjTWE7drmhmox>Ds(s-b1W&`pc=CZM_=?{8=OQA9HB?0 zgel1)tyjCtXyjIfyF^DgeueKVEb}MIunngLsj`ytwUYADjnLVg?7IDi4`l^%rInN^ z5nvW6S}+V}R^i!D<16N(d~u6Vnqk$KPfu2Q54c2uQ5^4m^s?(==#n~XxL=P{fVGlX70%LC%0QU*heo{Ypq zcrXwdXsOu{!)TcRVH@s!G!}`_qL^SLGhv#%{FIKCYHwLwsu#&bFO7IATT4WW%<7o* z7~iK5P+MS<v1Hg70nLb_sDKBXzGjfL+AP^hQU?V_%_0ugD8{a(^@ z6;$DZQOSZqmKta1aLvLniHGY*+N=2hqSCpTSQq(X0)QrQ%9rSGk4hMn7Nu#oFz-i| z8Uf_gGt91eo%kq69E|sZ7g>}LFL<3Z08aniRZTh;RR2O1<;cvoWqNIqneq!PO1O+M z;bX_UaNL+m_=R$TO^SUB9lT6t*(yneZa-z^g_5en8iC>S8}wrHctdi^Sff@w!T`fa zebY6C_1rG!vVQzYlUWhc=lYEBm3U<MJvoGF9?{iGl~16GFayq)(gWt%#kc;L0G~{VLR-` z!7N=vqR#fA3glA7J!PyE1FniJl9t)Ljl;IA!)!#nA$rJNuNKHPK(!03?rTGW8gdP> z!P`ZOCP`7y$RyelsV6%_#)UvFNLtz4zYbBdho|Z(-L3nK!DZ>7?l+{QAL%j-v!&Gq z>Qlqr4|t62r|GU6d=3;DZ7;@WzFtFp9Pimvc0Hfdw$rQq(`WX@GpA=;99#Lbv&XN< zK27VZzs08v8z8{stIgED1bAW{^W1mnX(zUo5N6i#8;gJLfD~IZQ0Rq;b%v+jA^bA~ zj$SiTI;iLxX@K}&f&^@46W(O=Cwx3nclk%sm&*$&X`KvH)119+TMzwRSYkMnI1Zf( zf3ew59}Ng{(CQ$%VWy1Q+RoKu`*X4;w1jJ_@vBa=2lf*qj_Q6?f!}f4p(XvMuH;WE zK*o!XB|Qfpk0;rQAeTz?vdneR%~zu0jzJCD?|DHtdRof@`Zdn)boP>$`GqUza_wDJ=4joSgYQ)~;HtT5L7J5DW)x7%oz8I0j zc0$$NeV>vn{0+BJz48_Gf8-o7Ugi!JZ7Q4WZ4Q{he|AyR*+`L%9Plf3v{|d$Io$^^ z-qKRhl{@d>vHm&9gFKpj-6Si%JUEub6tF@DwF1aTzLSf+$52W2ZpZw65R&;`;roNj zd+Vu4i&cw8)8y#q4t0!)5CWJe6#TmDrhQY0uz-%=Z_*y-I9klkCikv}8g z5n!`#;fA4+6l5|vSH=)Meh>UU z^1&I7{a%d$LHsjP5eh^J?7R2S7tt(Y)4mo4{Fc#4nD^p8vbPqL;Iy=q4_AOM%i0h- z6kGDzXa+84k^r)Fe;>(QfbrQl@aGxuH#BX&wS#$%)HWi#^L;yaZ@=9KR;5b*A~*Sa#CQwFe~Au5G=-^W@}nArAkY+T`2j_` z5`6rwmMszE%iv--!3046P+rMV5QpSJ^DnWpn=hEwAW(uNOkNalex;I*rm53~KW!T{ zyX+=xD~)bxf2tpTkqNmJ0qj4|sRKr%Lqc@dGwlzA&Gz)ciUyn`6%i7P5yy6h=e$ZX z04%sBLDEFj7o@_uHrTMs-Gm}8s~M905Yl09ORxf7&5S*`VS46`b*{pfPNSz7@vMRC zL;xe2cS6xErBt(Gt~*FeAD}S|_@zko3pA{GqyhImSO_!xD@%9`g%8jzJRXa%x+*+5 zBw-MY7se7#4U5oxV+bXU1vbZ=u{#f-#zO23`yQaAOP~uhwAu%flTwPjCcmc+r15p+ z23eGekK<3hD0V@>g{E>48di)2Tt-tpBmyrB?g_Sx-xG>Phe!X)g7_YTU$EleS-eTK z*PY}gWn)E%=zGXK-*aMyA_QTHs7(m$MZ}9{d`VX9ln}%7MMYvl(#AI)slvf|t4{VE z@~TRNxeQ7M*k}_nxL{HU?z}F7>?I*25M+?yixHHKYJzi=;Lo9;wWzmH^7e`-x^36w z;$z@Um)snM?S+SdJ7*9m&6H}%g-x@FVSmW4m=(iu=L=q6A$05x_Hh`p1DJuBK*~QpxV+VxGR;t93s#9h6#Z)58D$y6zI`k09w*{O^RjAwL z69%REWmU}NrE(b6RzhV0!xgW5s>|P0yD1lX^i=;}Y`t@P9bDJ;o2+QXE4IzXPGj3? zY}+PeL8=&g-96rJRm1*TU1AI2 zo=UT&s&O;XIwF@Y?kRpnBUb_UiIFo|ixVUcrMRdh0gh}=ixi>@r?j_-NAVGF>p>QJ z94jjhevzSNlU7WLaz_3fwwE!YNA_;(#t4rg13EZT^ncsT%>UPBlKwj;1Q$$7_}$64 zKLUwF78BUG_wUDn|8}|d2YiXyLy-Q|8_5LIz@_Tp{ZF@$w@H2+eP2F}M`f?)+x+*` zOyN+3IBQ4Y+0c(uDt()TO0`mj!fZm>R{GHZmPWq&<9fq-pZA^+to8b>mHJy6JthUEWjcTbwq4%g6pzBXTJSb<=NBi zANym43^LtLfW<6`vS~(!Vl@rZ#+OJtJH&^TN}29Sf=5sAJoUIB@Q@0y1W^ZnuWbo+ zP?lptdHG=AXS&FiE377dQWStQ!jm5~hpb#+2|NOB$0N6uhQ(}5n-_YE#cLU0{_$@& z5TOw*)6!9z)51#>Ub-s?92>JTlg0Hf+sS~Bh%&NKcUdNjgu;cF$JAR*0VeewND_#; zumSa=Xg@}IVo3e)ytAslm2U}loT{{n3#zuieMq6KqLX1Z0_P|>@7o! zPk~Ob7CsYIDq_7?nffwts1C?Wrqk1e{e+6GnJ5|AE~{ofC@twRkt!3;Ji4#?u3?=9 z46^-D`^OzsX2IAI<5QYwJ=`A|INF4_9!p+A)8lHsQ{5#93DfEr1kPHeF}NUIR#9Y{ z5f4i?oc+L8li68~ILdjv=NINbYOFk~C_frEZL=xWaTI}BlhYQtphPDnJ+oPgsz~*Ick%f z-@3jNpQ1?*B8X34xc((-eW;mIz%+s9K*eFnF&HuN^b+*$Qm+FKfx-3VPCw z`+4V-Fq_+Zf9eVM;wMLyc~Esds-$NitD3<%yDd`ip|`+Ls_YP?f$+O^qrAR>^!n@J zo!si}Q{QixnJ;B)(=S6>B5#S8p@W4+a4;#g{8&>A^3fZ>nC4`bMJ1^(y1O5uxP6b- zmyDdMF-V>4d_D4xkwbK3D4+HgAS_-GIGY!KZs=Z&vd+Qq8SA_-VEKL^L zOJe%3QZQnc9G1k2L}mk}gBc@*-X6xFl}f!Y!GVeV+&?dA>#PczNj^S|DV{ELsblqe z@eXVYPZ)5*QbNW|1hY~~gzq*fF_tn-x<3k1y@@on#Y~Qi6+)HpAURoY+9Rwe95R(H zwPIt+2M`IWbCS@gKbK;({DW#Z)|B3^o<_=KIx30{EfWY&{T_xlRf6OS;Xg=B*?mhq zYUJRWR2j*Xnk5T`)f{9_K}pPz$@wi}mbzS4cG{sMT@Pk1YNzn%{&Z9k=dvmKx}+Pk`0$CPE?du5J7`B#E7208X$(gUJ==ZixpSr_CPA|K*!4+O`llhNQf;8 z%RhFL0G^(ZUNfQ>4S^r%m_Q3As)FN%AY%}R5JiuJ5pVv55f^9GQq0l3u0PT>kVd*B zUGzR`K_Ztw_xUkRDxGW&nIM^0p?NL&IVb6z2m#Gy?@^?J3&-y6lZAGp7M(pAql{eL zU35y12bLZ(!5T)fzK6rx*QV>vrKhVf@bFc`TKrroL>B+IL+ABHA?dr`fzzTSO;3)^ zZyT12syC`tkr1uk@v zwqZ-xC4DLQf=p$P0Q0J^O zBz+^SpKu*(R0jn zRUTJ^?)M9_FRzXp>4e`lJ&U)jNx+hrSYC!@76;Qq^Xpm8Nbl;-bXLcZIu zZQiXS3BRSCCyh0`>`07i6*O8pxPW@+>W?d!4hY0g^3F_OQr+_pJeULE#&*1aE zR-an;Z1X~Ebq@~htM34J``_0#zdeS!ZOm@!^*+b1-6F(qCzfAig5c=lXue&yd2qJ- zdvdc}H{%ebsvxokWR4QGY1-L%206J3%Pz@As5=_N*#+PFk{tUPSla&y51us%fpa#T zZS#}E2_A+F{`16EJ*3=W5!7P&g>lRyCEOWySJ?th7`N1q_Bhb{$w`vO{@600-xI|D zs4zPiAdBk(;%3Y;wx6^JCqk3f(FW<#hE2ecmmjc%G}#5-2aUN!m}N#-R@jD;(JFJ> z!uDG^WeT*Le%T}kd3u4CU|4}krV?q95?EkAX`IPs=?L6_NGr@JFVGd2XzWc1IAH}l z-Nef&)1O;glK2FD?8qsnBGfP=D*3>V^VpftSyAsfyezxD!ueJ`bJU(h_gH#Xz{N^H#tiTTuN{NhDr#ifLf$>vPabi5N3*h)= ztAw>Ny9+BHSjdEU@r2(MF+I}pPh#<&1@vSlaXaL&m%)>vlRBL6p>3;_`fk@Wu|x1~W4|G%?dWUq*LH(SCf z02KsfnI@`19}4}hzckHd5UK#{wT*F%;;?+-(8~qyq(hlPX*~4z(mN8Bn5O`&zgTKE zsu?^m0*g8`bP7crrm`*BE~TqUCiSrIRg2Z@WqLHSz3x}R!QF5u96S=5weo2uQhX1` zX$|waINRS?or0VV$dPqk(&(tj)XFnH%`I$_F2j4ii?MrgX3?y-S0HNn9a6Of!xd(+ zB6Y!ZMeQQatF7=)8hfL-Pac|ioaU>?W97?y@i(zF#!rSK=xN;ZtD)t zhaLs~Xb(r)P!hQr`OxQ7(kgyX8Y7yD&%9$BUzEg`#i1Z1GnRv3Y;aY+tdcFH>gM*k zXXyd_RK!jHgENEwgEJS#ybCNS#R3vl@J!189BE4Yn|E!hz$2XP>Wy~;;mp(pBym8E z9{VoJ>Mt^c_tbM6pM#3|5}!$ceNgCO_$MM$d??F|3KCiPQ3nzQB(1FDMh;D3%zeSD zk+tFPWg*Inb^?Qe%%K8PV{vliQ4HoaMPVy!~CQbPBD~Ek69$Ici@J+-z z2EiMbfXyI?;Gcbnrp0*K%N!?8H~_F?glHm(tZ^R2mQ$}2?logj6_R_MPY_~x2+f=P z(ZlFEwX&GYK0*}=J^z~p4*7mY6j9K1h@K?1B}69Q4>ZL%i0dw>rlr8Vs8wz!*MwZK zdMPpVL%?lPJ)`$v)uMdbwMY&_++n?`jY-p-MNS_z)+-~gebdYE(|&R-%Ef+W8G*MK zG%1u)>$Mjvrtcwtr_ZP&cpf|W>o7B6K~6q`B&KyiOBea|#PjLSb12A^p^pa)F~^5q z*DMCwg)*n|n6)mh^}R%WVhoxWih%;M3__U-up}jhy0p0l`aRwGiPg8?%MF~!onp=J zCvZs<79g>V1)+*lHnE3w?4JnLee5wIV)FiQdktm1UdKe8dU_-vk>OYR{qb_I;cEaP z;>ov76h;W<$-f!E`LPcvN-)YcHtm%0V^||m=Bg@uCt$RHy`k7E2tu5^g$x-69TksG z47dqBCfZ#%9X{Wk4~fbaFTR0_H;X$eG?pYJEM7z^)4jR!iQoQpq*V#xiURzDnnLK; zlq>|vf+0U}KK?HmIsUut?@!nc3CM_FZYwdkXR&ZVO3FPF6*p37YZ}nHN+KBhKz|6K zf|ND{gq!>h`?nlAV!sGighn575C&iIG8}Fg);9g50HTHu#z2@1} zLY!d_W)`lo7O%whu!xW*lDkJWp4^O107wc4FD7#N%g5MGds=ZMUR-)VXHI&A=OB`q zR8~3JN&Xlu#cLdz^G04VRD}taP08Y2QS*@Ro6h&g&tc)VWH(0N(qdW77#Dp814C2O z+AuJ`EFX$JTFUNjg@;lqJf#Dndg0w8Y+s-US`OB_(Jn&KObk9{Cb9wz%$+$#qnjex zvaD(y*YjK$k3EC`r1c-S?4}=wQ^X@JNHRl3%IW~t9DuftJk|j7mSBrrmHPJ7t2#jx zF}+zc)Wky_&Cj}zrfR+Ss=sZoi_v866ohisV@ngP+R0`JSPbDJxl}=HNJDkdN!5~L z4N*GQ)%9*z5HtvEEBy#14Ixx^Ch#UJ!A_#}(!3=Qly4XlijPf#GJV2~44_{ok7T-_ zzLX6ev@y*oNmbbkW5$q*Qf%nDts7Z8*7>~l&-(-7@E0kU60?Mbdu$!F^NiCHc)G}oOAI+tMEBo9IJ&dG}?HK()G?b9WOrP52Gmg>!P5JR`Yu7lTTOu&S&nsPA5g{#bTMd#x^1^{t|V}01c zL4{)Lk0Sh6Y(u z{iPseNXN!sS7&FDOUiNzoudNvQV1}3KFMTUtS-H3m_RU{Bjdey8OLcmG@EH8122Hokra$PdWDPGtAmnQ#@n6hTn2PkI$w$SP(3By@HS+RasOe&%0cXUO@_)Ny*MIJ04) z^*NMEJM1+wh;Phaf``Qp=f5*|(Ed?w{t$8hOSxe~vw)-rBVmcC&44O} z|50vMGuy%m01*Vxc;mlQ`WUae|3f=4NPSRbHI*sEf1Q)LF8CS>IP`9=n2mXK zK|j|>sGKi*coj+0VK`j~kbiq5^!z&QlU}!4 zqx1H1iqKxKP%R3D`e40OZ=BLAiG7cSWoz1#NFujo9dofeiRX_@i6J2@RA>_!fqnj) z?Pwkz!!sA5HN&)0(XicPN`vln9k_`AF~#3vWzD1x-dzkGEjmP_bkB z9v<&?^U}0MJ3UdmO|nbY&rjldLi+p5O<%+IUk1HS_l9p|>iQsF?<1{0bAqTOrGAZY z?=(RcQ0|XezChR~wLrSg^rX<8L(4RMj3o&ILnMRr9Up=pS~G%pHk70y6y#=x-n4t8 zhG4EVD-Z!}Tlr3GPoovO=5T3QN?3!&%6`I;v6ZO`hr5-T67p7qtaMH>eV9^WxmKVJBDACDrtZ| z?aiE1vdTnzSL*akzX${)t4vE-E-l-{krDNLpE7mII42AIG7(u#oAq+QI~2WS7>at8 zM6Q|JsRa5%+I)2x!Y9O;y}7zgu{E%hBB-wC^3=Az3n8q!)(|GKPZ*MvAxLZ-A(gRt z$UsuR#ns2G7Q#;dOrJN&#lhnlK@`gV@8sB*eHHzn@6lUZE6$6EFcb=xDVJil*e7;6sBL3cqy1pjs5v zTgvsk3Nx*8wN&_vc^8N0P&{>z+a@Ra8p~^~h)Rb<=7_b8tSSQ&Hn(PW_25XB00jd ze(5M9umA0?(A~bqa2V*n)O{sPNg;C`1Qe1LX>-Hu$o`tV*u2YI?XJ@QsD~=gGQMJp z%6qksJoW_qmjC0z=0d*PlD#s&=qnt1S0X&!{s}Q@D%sCK%9(&ZTnRzGc7nVOcn>6u z4Dkl_L&}b($miLjy%4Qe z4i!)XJf9!^oWQ*+(s_U$is#r6LgAMP!lZ<^5J;fQcNX_243Nw{&LE1jIOivUkr%K$ zZH7z{;z0Nxu|*c;aUcLkK)An-KXyWl4u1N%#Do;;m8ygMJ~F!YZoz%ZuJ(aBfFdki zOe9ea$J=((;S3B5X+$5p`IX-D$aqCgVPhbY`N&S1CZQr!j5@735HIef;H7dx5Kl-M zbiHWlhk1LDs&0XCaBeHfBmkZe5RJF862o7+E~S8h7!P%7OeVTD3=(b65Cp`3jeJc| zhj5b&wzIZ^php`&=jm{e91$a6L{Je=E{(!t&w#!GTUa$wnf?G}@#?OJTK#ED!U~Io zAV`!%;r#rr(AX3LB}b0ei$_;6R%4bLqb)PlDbs{!7nr7pRf#036z=9A=Z9b_LffHG zu{o;}|6(aaS*1LoA0y4D0*2ddYwSBeXZwcH_eEV)FyiYtiEnU}qI0yG@AE+UmsX(6 zI%P?e_gTeeH06(Un-q1tp~_bLF*KMWf>8BVCQYn9YG_ed)fpBIMJ!?MQ9e*pvIc~) z%s3`=19NaoTJ33cwJuDoO|m=l|F4}E)=R(Nbjy5c{6Gv!LU{${@B9Ko2C>vkeS!wL0kVYZfh=~ zR`Th?KxrV8ULhW{w#DW_RjMO{QK|YvY88soGaMDHcayYm>CyNUROd$Is~QRqTf})P zt3HFu5nHRLsla~8<_EB&=LEUYbo^~a zt(PmI-t@}Vgq{enjWcMi9~BX|kk8w>G7O7dckhvl&DHv6r1$Vbu%1B72WTUM62%xL zv9WSjXAyftM4M)ZbU=!8(#O8Fom8uS_;!TrG>puim^n8^Hmlqi%W0n>mv+ePu=*(} zLtLMH?6*cEmG!+*mo;sfaFt;)b8PG$l<}f>{DU~N6j8AyKTzIk6veyLVQD{cZ^MjF zE4$G{vT@*gXzzq$Z5A*y9SpIl8h7zBpZRenovtms%v(0q{!OFuSL~^}F3ubQwp&Ec z+wMH`w_gib7d}F=cDEn~fPd^t4)7h?5tkTE z928F{&mk;uyZc$2O6(VG)d>z-+jgP^_w@07e(P@Cd|3ZDX1e%L&-uMxrDP2oun~gs zcL&x`rvIn=;)n9I=D7JxBMKskKfEwJ^iA~$6PvRJ0p(r<^%8BoRHCYKYgI|<$|<#~ zqDES?U$w(*9A>f+CYDg)V2P8 z>GivN>*?3ikGJ=$Ak>~6IHl(Sf{*I}YrZ|4-_IjjHkPuUJtAgIFB7^Sw~TCYM{p+D zMlgb;|D{}@|4}Yg5HA0_<|Ga@BZY_rA%E5LYA_aoMrJ}7q7^RgizTA_KYStFC$c(S zVktxh;w0`pO;Fi%HjiV`zkH!m_-Dw!0S#6&8Pce)y8q!c1OMeTzbpGnHlsYyq#C6L zKl}|xwpFV6*W4=PO}{ zvrtDO1qtP%YV~1}-qkzSy3)0F+jpTDE{4vQKN%TnIPd%&%SHnY`jd4uZx1Gh2})-H z)vs(?b&s(0c+&Ok#{s!g3~&mr?XD?p$#v&)y(e3N0q_89B;SU5pC5l$Kbwvx4R&;* ztV8SJ?mFMXr(qJ4L%O&vin8G*4)l7oc1!zL@8q2B{VX@EuBbW^^gmHnB`pVMs^4V zM`#4%FJuZY1tSd%mD&V$F0C>>X$uEINL_mi7pNd$hWiBjKOGf5G+Hf0>*8VM3J=1o z7Yki(?2AJ|6;@h^s4TFhJKvC<5+)&QN-emH6|Uu{yzjgOLGY@6HiQYI;zCvnqqT^U znOsCw>H3+^a37n0TjDrkXy5QjOa)NxTDugs?VnjkOz_s2*C1;qR0AGlphZ<%_Kk%i z6#qF)-vsG}O91lGH4r)U2X@*iRkn53B}>E0Nq_UZPCNGX3u_?i2}@x3mGU(bd!J+_ zBN|TdVahVu*IXv~p^a@4F>0RGOohx3mz}Jg(J`Gh-EH`Z!!Vf^alMcd81V+DjOZ2RK*sp+uMy@*Uv!p8jc$C8Ak zj%CLYSuk(5G8^AJZ-dF6U@6|at9l*kVyZ(4l}6`j$OJK+UDy6gIp)hG!vrG@G6_Lrvyr-bZk53LRRD}LDD`Wb5c)g(uH)k;0OC6Yl(g_i zKQh~+@1!_pi*N{sav+RMx>?!GkL8ytEoqe~SoX1gWS_)PK%z_%G^mFgEs9!LjtpL_ zLiBs#3=?(@Qt`|oDmA>BM;-U;11zjaViIfN*jE#h{%*Da3?!}IXkIU$|$lsqxzz5zY z7DKv_HDc`P0nufXz;j6+_@(s7s)e$|N6B)2s~DBHxaaRj3F_(!(Xeb|+KFhs0J3l2#ykhS4ZAbZ8Qz zNbR*T`*VD*Vfj-!sKqa&VOh8)RD>Z+F_ZN6BRbeA?7z*GV3^AbT3G)ss1T0^WbGoU z4bmDg&n3YKQp;vTAtkt^7LP#yq*06Ng7NP&c{TJz(<&3C$it|n&vEc)GbO#yhjL|l z_GO?4QQ|x>aq%DiNM%J6q(cVBA5NL-7*-OetD5J9ARdr2c`cN<98u1K&LoGUW8(n9 zL`>#YDsqtvGCoOmiAX1?o^vwmBF&@;Sruk9o#N$yd6KykIu=kcSlsNfyiXV~;8B4o zpxV?Vy|EaH{)VCZA?Cd-ny&ztBHjJD^g#JB#=sJi1XO6*Cn-2QG0d}3FCkuP8&4(q zP%KKW@L_X7Rmj%Z9Mj~lT53wd6t)nQMJa&9X#pCgoZIk}z<=|sJ+?1`&@Pi6j)$&m z=uOu{#1ji4eyXF68Zmf((XB^#%2|VqNjUeNdHaZMWg4cIAbiyWE7Db0rPYLBp>Oy{ zP<0SsG2Mt~=W2^;50F^UdC1(((FG{9g>+SVocb817_}7#8I1}h7iv)gEsOz@b@9Av zqJQI7%pTZHlSv~fQyW=p8`$l+9LM7QB}MU#$s)Np5W>VEGW|(JY<(K5q#j`JM-hY1 zPN0QhvBc*3@>pGE&dHF)V+TKp&y8Ss&qem7pLzO<|DsXB+iLl?CV63KN}CeJ^b09* zReuGro7u9V2=lzo3JXJ{VZS6?$ffeJsKs{n&u~pZ~3ztWX3SFlVsyogcax z(E+yWr@146tPd^o@CCA#Ds;Tm*Hp81XTd|Wy}>B#g{&2Oqp2=Q%J`3pzmSLeQt?-8 z{x(EX)QJBEd(qQ7hetX8*%5Dh+SwrCs8S8LUH?=&`P!+gl(%3B z-vpxq_x*%YzCe9#KN*#;Hl;|2E0+dr&7(u(y1%lSA#_kmSa5HaYwD)6oEqces{0Lj2A_ropOpstLyHRSx@;^5 z%t#qgb`Y}f+Mm&o6F0h*&be;3J1&~*gclM+sRTT^1i+r4^J9=O7jnvT2MPlNMV@se zEdpK9oTbMk_`&F_PuL7zLAaRV`(nNbW43t5l;m*6teGtD%E9}0;fC!Zx4WFqoS;&c zkd|~;<3djUWuFD?5LOsDxNx1nk^v55dg6y6)97kGk`7Fg>Ymc}YA3cDD?xrDy!)Q| zEGNbhDxrIy7~l|EkQtAKZDd4%mGbv~7JFb646FCq2!57kqA8&6Px3%9Zkn$u2Kb53JS;=qf*Ar`ymG365u!vu1{^ehqlx{qH6)%s+zs`~OO){!e^2X<{fU0Xj5k;+}8}Hlju9 z{|Wr?_)8O1GM-GU6#bvz6hxVU{|Wra`fxbi$Wo@34gMD zRY*=X(U3l04R>ZNu~MJEP_O1|jWuIiy&B3ZvL?phQZrlWhyV#d_P3Ff(IQrj34lJ^R$hoRG@|2uLo`NKqKwPn-=J~tCX<5jg1 zTs$F2rG2^R?Z7fGuhHzo`MKfp8;=ry zr`QP6BC9AJl9_rYM0U=6!jtD?Z`>2zU_8k#7bc6?9%|84F0*HI{=f#6@j3!R-A7A2 zfwr!~Bqjv=5IfWDmU5oIrnD`H=4WiwHX1U=>_md7#RPQ@AokF_NKu1^&=!@Y;h12X z(|JM)&~#!GlNaz=nWh9661_+Vt4Dswbk*Y`10~UMyKffMpfg)- zwLeU~sP;lrukRi7cvCSvM=-~1YDD*)$%X#eLLlnN)R7*jK7L59i&kGkDh7{;c`JJ% zA)zVqLvMps&U80NSrO|ercpQmRh>Zz10&T{Gsb_d*~DiUrr2I`vCY~TDvk3}gEf#B z)WPQ4veO3`rkQE_HeC7ekBxqxH;41KzYbAm7k`}BdZAXCxB9|yyS~e8^$^2SZ}+r` zWFPjl<@fY?kvw6^fAO?~*e~q zk08Bf&0M(^%!=;mnXPWU8&sEFyhF_NIjDI=^@;{#GJUDPGtRkA( zMuXpLA8~S?ZTG`X=M{f+*;|HvXl`A|cD-)Tl*0h1Ch=W=UT(t&ev%Y)_UK$B!#z@T zbZ@=fChQM>`T2U@ztszW^rd6_zVEleHyR_WLgb+|uZGq#qY#96OIT*2 zC)S@FgU7w>ka~EAbC*at*{`;b%FeeWQ@OHZH=M17e&XHp{Tz_(DrA%VvqPN1O_tTl%aiX@ zRmyNCWI7pJz1q&U6J-VUK2hFIip{`0gV`zNw&in$j08B36(|>p^k2ya^C$4?3LdC( zR3)BR{C(&y=Mbr%LQVA#f<$Gx8Sb`y^Lw{F{rUAh!NZC^)w?H>vJcC~K9VmS>&QG@w`UKh+zH|u=+x6^ zMT4->kCxbGL5$U?NIEMl*)nE~>iO5yr*FoX?K5Zw0SU5(K?*&sa?7H){b3+t=8o_; zVN1cHh?%UWPeCuC5yWd=zvm+#0E#h~H`vO-qZ~a0Mi58%axv6O3QC+mK)b~V8Vd5_ zdAi}0lHNDdPSq+kuR~LbK`ak_Ul0E z5Xr2x8r1YpS z$c5Ho5PB$as-=m8`!kER+=j!@TLh2dHOjPg6YZJ1o0&heZOCR>C;V|!8gftQP%_@J zbstKy1_8~zLB=r)359TfJrV!9OA?|CTXv{!DvuS>U&BdNw9AS%@AXH~bGRB3+6rO~ zX$L~-+J@tGffA12n>m$`wLBOC`T+tGC|+YK1+l*%uAjvOUQ0_?YD$c-v-t`~3~dWE zf{*s=H_#KEJ^UvI-n%GjT58L@lLxe;apT_8E$pI1w~!wRQOJ zRe;$yadvLSmA_p#tNkez2$PdAY-{Gy^@8BP^ zO3hOTBiCT7qCOePh3lszj~+R+p8JQnW?V@#6S*2O^L(j;`ezedYF5`FvukH1-f(Ql zi)&-}w?~`+y;IN87JPL7tk2;##OP-;zeD_T=?Jf->`YdwJ;z62^gEl4enQPLT;!Cv zR`s}igCvfm*6?gJ^%yq7-bqT;(zd_~4da`!bZ!aLV!~*l zgoz?8lhf2jrY)Yy_C@XI)Gu(&Fz7ZT{gMBx-thu6{5jkB*jlP7qZiwpJ7TBfI%ym z(`l~PA)tJnRDf=;7cSm0B=VI)?zCCh&e(A9esFh?jNX-)+4(_~SCR!mxv1_h76TdO z#A=nYr_GCAB6L5t`)Au-Z6q6p`6TUcmdFHuxHH;-!}g<C-PlEzb4-_{Y!z~MhRsxTU5&?+}Qmoq|3SLR|rkw z95-*S37E+kP7>nLIQ)35lTq_AYuGjuN(>q*+bq$8mf0oSAdEGu!F}4KwJkfRd%Mq0D1+}fb29l zL@}9?cBr|-n7Yh;bW?mn!hLIL0{M1Ki~N(D|{&X&E5{7mdx0p3uq{X|^#b^t8hBpTZ{xI7^Gg=8Cw3K=9%o zp$tHc`$BRj8f|rL_Xtst`FAHfFA7OE73Y;eQk94$Oyu=MH3|Td zM1uIAgG114&I-lQ22m~`pjlKS!o2ig3;mQ=g5LM3RG-PWw9JjbvD#oVrA(9UJy$>w zbmcJl>Y}y=nhC%t0{k8MQy2v-90F#EJ}I}X%fw}*@@(P(Rj*i{ach#Wviair`&UE> z_r)FTM21;`!JbU9PeC6iG+4C3O)J&_qu_n^uODFb6?h3q{C7Nle9KVXB|) z#=k0PhH)boGh^pRb!q|O{7|N6;V$vG9%I_Zzzn7}5d8iQYKJt+LJHEHGOm?Ww7^mH z^qt2QccL3DO{}RF)vg`#OW?>d!2<|cxKGpJTS9m+JwOGKE7dECHa2fwA0gCv^cnob z%E*O99&E-4RcKJ=k$%`j>_~~MNrya7iVPMpdU|s4)Xu8#jCWlQTy6YXJ)}6bB4Zwv zoY29@e=GmU>Z3zraO#Sg6rLgOO^CaUBQ)-ogQHW3J`O_4G_h;T@cJb4_R~&; z<29q+x11ghj#^6GhILHsl%eXxg+U8+0r~>ILsRgYo$}_<;3cGZ#!d2A0D0IfLO2yY zafh4e!SmmDMWy#>aomsGFYcVbQ3#v1$!XmBa36;oW<+F`GmEhr*?RI{ti&hWczE7A z88-Ty0FybjY}|1J`1UCLEj>9sb^E zlPFX(OKwjRhaOE{M=)=W9AHEcN~y76BOQ;^Jvx@q8D^unRLYKx%WUw=?AFR0*2?r` zVRJ|!N+2+u`O4j8%YpsyfWGp8welc(P@oM+&j++F0@C9v2lSUlee!|AtI7e5@CiN@ z03%lH!O{$!N*|lb?1@V3=5nZ;azFYCH(3NPgi3%@6|m8@h8`m6wyF-ly1EJjHVOnA zW#6LdaA;w#m&Ds;laEdc^OLWpFB-k}G--^!dg8TgYOUrM0-6LR z%WKb#*r-lW8j^z^hUgDb<#Hfvi?BvSg_!?=5ghl1IlAKSgxC&@#%ZVK(d89OF~QBr z!6xW#Kk{fv%FKvfv!HX*540W{V;^bGS z7RWW%cFXYSwvQ8F+BbAzj5S!71$-i?J+StUAha5Neha;lX}{&w>}BXkIKxT$=Phq6 zR$}e3B)6J4G%RZb(HQHwCeXL#23^Usg!rG_;E?qBo$ywA0NX%bB3Y|jw#gAhBQ zVg^Xj2y;B6TnJeBrdS0C&EN(dDhX^}Wr$ch8iwhN2m009NX_N!@D=P(l?aDhpoTbV z2CzMT>dckd?qa3x;_CO{W*f)r>J4)?qB~n+B>?+17(0}@M;xZg&Ynl$RH|L{MrR~p z3$m*!>C1ft%5(Zgxh+PiZYtOwD}b}G(1jHt(miADJp?9ajj{4LT;jtD)wM}}D!VnPM@tDm}i`%K0+o@V&ne^;L`2nrbHc zKAe>h(ah!%;dE{1E5RitI~!G4w(GSTz2teiGjto>UT7b;mWu}4KQ&%p+vvT%3x<50 zQW@AKG@1^_kwrqsKGhjjy0H@k=ytm5)qPfIViT>=ThBqIwA?S(YxM@y8jfCS{pPe= zZ#{ddvXpD=YjITajgWJ95K5^NG2l|@;azVX{M0{^%d_tej&7dc`sm_+x>_|&{tO`| z`m*gSGirYIfLoCoa$oC?sz-owROlAwE|KUFJqu+ZMD6pW6W#LSiYt=;(z@tII?<7@#Fhn( zAK?H$ydMG}ot?A)izRcA)I1zP0oH!mL)942hoN#MQ=y&lsO1Hh3U@opB--dmeACe| zkb* zLDbf)66H|3j8@ktqm>++69mi&m%|Ad1wqQ^IxokNX3>IT7Uqr&;$>T*8Pwnh=o1sQ zPH7xG6BX%onf@QT&ay2IcFWR5VTD)W?he6%69R>MaCdiicMlZq1b252?ry;o2pS~8 zAtBVf=XB3Z_sCy(KI~`hb*;Vb0hXRMHj{sr=H5b5QyCAjoPe@2J>ZWc)f#Tuiq972 z()!x(IUbV>JFs;Y8!$W1T$`X2?6spNf~J7JpT_S7NqAdUPln&fOgj`KQS>+&LO$)- zbOYRD@@BgAm+2xDHQ&<$qa9) zo<~l{<(d}L5bR9pMu_Jq5(<0%sf~!}$hhnnvExuS7%t#NVjtv-vxkcRYmBpV$76gQ z1+@2F_A+XsX)}(bi&!PvJFlXL>AL_;H!C_ZOIh!f(p&9{!I?+S!ol9wAI@Q(jXMbA zD7G$76w2LJ@A!=#9%vM?>h)1VUat@_`mPtweiA~SzTBto_IsP(B9&eeH&8s=&Ic*w zrMTUG@aX}b=9|jhvZ%Zn#QV&=HYf00dI5i3Eag`2Omos)1?agmG9gruimUfuqm`{2PPX>KIy_iOJXJ5hil{ zk0Q*2k*vI7saotBNMnS>GzvJ0_puAkq(td*$1d)`r?J{TuL3dis|=@HQZj8lzPO`i z9(Oy^0KqBQ!V7D5k@V!wVZqr(kXnK+YgSp$DSa5Gi=f7^Rke&1%T8dVKeUbV$~wIt zB0U%M%@Pm2S!cY4y`J+v-ozUaCFyup6=O|r>8B(w=-OJ@D!V~K28 z%&70EW81r!t7^uR69UoBv~U@l>*!9dVhdZkJx#ZVGXdw5anJXH?Yn z)_aaeX+6~(IL>^nROcbz#ou0Xq*}q5uxo zS-+jxob&f?DJsA)R#V>q1dg@=q4faGTK&|>hxVqAmmgMXx3}PwN9<+@7t%!xdXukZ z_?ALS&SVaR2ZanRY(wj=QzZt)qcIH0A35C0*^ExD#})8MLEcfZV4$>ZR`hlW;YhEM z#AOU$!r27pnxf`C>v;dr3YMu|4Z}k=x@|8e8rXgx2Y6YL+gGZo;9-FguyN2nL6Z|k z%?n2PdN&j-ToY+z(_w-hslw>y8Y~8^Z&rS);`fXZjda+!8D^}T7nv@aP9TPNKe@tL ze)%_n(&>K*aO2H#ANWJG%50w|F3|F3k82E{yKfaJgT9UoN`kfVJ5(h|XxJ-j!PQs#$)$7);UKVa{1KSecLkY9k> zt4wZi2^S8>ejifhJqq%x#cWMoB25LerT>tLPoSjKn1`}lkkKZ054|p;fB|73b3|I) zuJ@dzTSY6}F-5sBtRWMFaKG0Qfc)k>?RWm)H64g))kct5Y`@NZm2M-f1# zZDhp1P@)+(AT&A|By%>=>7f=6xLf>BGa>e}JO>m%y_0EIh*Bn0K#1bhQ5Eqq$UsSOym%nl0|XeEzvjB zli@BM)5dXX`*XrdCD;cUO8Ug!7rU_C?5OQ@8r&p#`pfxo@GDkg#KhskpCEeV>oFd= zGd&^vAQS7~+}Ki%hDVw{o&2vx%#Y!^$Mm$5(z_%d+iJ=8C{R(r@y;&JIR3}FuOx-L zhczY7@_-9ppMNE-ogC>dL&ark1{-FxNFdfq$xXbPNgyeD^IJAqVE>TgY^LGs-JkKf zFM6nab^Mq#VLn=$Vcf3XdaSNYt#qPY{6*6ebO2K(S%XSj|+2@5)L{2;*DPm3oyN9Y}H#$iG z?Omlw7+(YXmJPB%(oL<^-x<7mNxFJceD zFh}^&xx*q2OFZSXV*d^XbNfIq;Hr)()L8LQ>@6-=aTus6ni%3{VSp-HxDhElBu#%u z0nbzcipRx)T@MGb)lQ<=xh+9UCVhnAU+H~>n(-c}v5ahSE(Xzcs-ja&V=?H|+H}?E zyA-#Fq^H{$wiptzQk{V?5b8!a3KJfG7p{`6-=t-@1B!0hDb9DUk1HUg*g=!2Y};mE z|Ltu5-DZC=Iy`zCC^94fHG}iwm)|xu)-nw9`{0L=apC?)pI)A5At~nXMgMAKTPnhe zr6yB(3Ut|!T^67Xm#`p{@qhGzT!$Iq(<+G5y*0PCj~kaem6fWI<%y6`4GoWEJyZGp z#moKF{58w4>9;(CzL>kDT=n4YEfzNCcu_m~h zzMw}JRzYiXl#FkD8E-PJ-(l1YKEc;G!d5Xfd=FB>5W29*zgls*uN?pBwjx3Ff3N~|`+xQ&z7u}NSyS-ytA zec$B9nh55YCvIk8w7Rkab4#-+2pIvVGyqFe%bX?a9}i|qQ^fm(@wu^dS!XtK`}8fX zq{)Z$z0Y!gkMjDTlV6V=#G%xN?ummr`P1F`FP1oP8-pe->oGnsATNZ#=x*Vjj{`Ui3nK|58C=k~9r5lV#kY730%26RG>W z-c95BVNy|Mum()mKciS*r)&VNOfug10=Yai%3N0%tj7vAU}dKn)-*uS#G{j-6HudM z$lW!|UzA5lMvBHmru>Xgxh$_uBVDOb&hxa7^9SYa?`W5yTm5VrT@JebvQv*zsd8X=-GbUgUI0?bc} z3)!Gex{Av#wJ)t|x4Y`tlc^0@ZA3r>$;jFrm&q0e=i1}ulV&xp)J{d^O#IfCX;uBz zQ%b8|6Bk)ig`V>lu`p7h{v|%R4yqd;Rf{fBk-n9_7;e_2h3=GV=>ih*Al0ABmD{RH zim4$3;FlKj6+hm&_TP&1t?;iu^KG$eG(6#8VjoC^jpy*@X3tpphRns(| zz=>*?Zwn&rt^fXxgU8G1qc^)ERG^oN(%&h|1|&ANHx=l+H$_<0^-w~#D{x9PV{2N! znYBNSt5dTX?BHvn-`2l5Z&U5nxpq;#yd4uF7q*^*I) znGN~Fkeko8yWRfC^+Tz-9Zh?nBRqy9Z|k!6&TzO5tE@ZftTT!G_i2&BTa3vxI)~o? zcZiiPXjL#tRLDux*qJzoP(Tj4B2cEXiaZVUe4}?`N-zCF_PS86@nC`y>-=$GF`)S< zY%%_mJ$p&8*O;5(gtZd-GC!qn-9%Q;1gfx)_dC5o25 zfLH!>%Z1Lh!&GPAOyADTz{|`a`Rwq!*-^vU@xWPYr5QKJnaMh$PqVXgsD5hZ95wO|I+zlgcJh~2+1{j!KFx|CM87)0=&P6yKe z+v$LXgYeKXxnO1hz<)X&&;<0i_L_3d7Nf~{P-k-amZHy*aHKLN7FhXo7>5ySp8(8S zFE}k%kzg&B z4)%=(qrYn!Gv4*v9j;#o+R<)5^_O@L?fvZ$;QZQb8>K+9kyW)hVvYFtZmZV#bMQwh z+31Vxmc!{ZR26B%2O`}C6DS<9o|xHu$xJQeJ=-r+yN@!v4<1SZ9hX~OE{}Mp&eTg^ zeX;%!Z0ppC^QamB!DP#C^~kADXS=gmAi5gwXaADWSHS-`o;&p(PPL#fJY)izsulRm z?QPiJx<6CMDOSUV3=9i#Y8ZI;Cp_W^l`!Nl!ow$tht&fLYfR(gB5!sE8so5hitHud znkZ{Rf5P0mLPM%?Suz6_%p62f4rFDTb$@+{kWSQ6Mk1&Xux4#lLF;>uBF8)A{ZXJ=CQF9!rmRbRhIq~$eS6t|Zvz%96eY9mOSKcCBgC$jD zr}>Eq)l?@(5!M=HZL(~gS62Beoad)YD1MC<$=#sJ%*Fnbo3Zga0PiOHXk>GyjT>zDsaR?y4)J;0JlnulM$QwNg z#1G}4kKmkOSQYYTq}+xC^N?31f1`}w^@gc=`&g_$!&_>l+(SqO%jAe&Y2tvYm=3Fd zh2E%+^LL8iZq8@TXVNqHeq1IkA)}U$WX9H!@p^y{ZV*e#PV~ss&%fkrtY@t~0FP(yAct7DTRob=~V<{?#k4A{iUo@=s%JKNGG3GkibkEXTGK;INt@-!TX&9U(AwvvTSQsG$0wC_KL zy72d0Y%JJ%(scJ(o2*K%E{1k_-j(C*A(7y%Ofq^ zDWi@n>637OF9*pb+X71|F8~0wA`&zt)~iCa&AIoV@#L_pw&$|LSv3{9cV$!1|#ib*XXdS4k0)ZGxP-U>t)6pdWIq?$4}9?KslR zg)AeDc93A-!jsjDPtbykt8@crxeiCpyQG|u83^RV&s=tOq)(;!P~DT+N>{!v(}q^o zU`^kxqZO+w9XD$l)$hI2yNFnj8jk*i!q``vPs(>|sMJZlE# z-be zi(#~#GBX80$z%|c(Vi8?^HJ8K1+2d222Ovmz`F1Fyz@i$UO*;{T9R}L7M%oj{E!NB zuP9J<_4WJ1bdqFQcIfCDwx|dmqo|`)rL31mOO>n>6>#&aHxuR4JIGs&dTaV!UxMpN z8sqnEpCylF>+s7^E4enjcAsEhBT^|9vCvYWY}EwH!?wjz*qO$l#iJnGCvJ%?71&Gb zK_hE!9)`cQM4oY-*chlm!FbymR8doSTPJ*cQ&n3`c98QZxF4?MlQY6X5 zPMnlfx#=WGrZ$8x$xVwSfsq3952=%dW^J&g5b6TyQlGjAd0KArvR+Q;I@K3L2ChoT zMB_hpmva7g!>?ds82R8M+`z|AT*bM|tRoP=(mH=%`h>;r_b~B=fc*q%ri|I#r)RBm zz^gH!KSnVn-3wcmscZE@$x~iM^w{`Oek$rG(%z2eG5LuQqJNO*hZlpDdXa$d zN4uW*xHQL=B9XZyheVcD{u>hUt}dK)Y(aZr^kL@fEY?u*n}n{PS6mO?-$YIauE~uF zEXYdlIv*Cp+i{YuE=AznCO4@Dq~C=zCxi7#rC$x`gGzRPRzC$zXccVnfEluH5~S5F zO6h}dd^rEeGW9R|77+P))A^l5==@Rl$`jz?7^n7m>wTFK=1}ade*8i8khI!{slAv& z;)swhgQ`(Bpqw5Mu1;9cV)>XA!fMMlctUnDh+f)+E+mLu9HTaFL!Wk_|4Pm3ixhOU zOnNY6ea~QBsL3S*3R=JRBZYF?o8fH3Fc*Np)5x26I zt7D!#wt*<+E@*BNuUyzE0Oh4NeX?fxx1dO;lu%?d90q0sd|z{WVjv0 zzVsg}H7q6-)0rHit3Kw%BzKPq5P5Mmd{1m^L(4kEnP*m9m6JL_4o3hxK*YZgVq_2` z`QWw9k;p!uV@9P4nG@#C7c zQkCf-{3+q{X`E&|nP(fU)l&Yo2x}1uaJTB9d77Yb7TC!}AFq@AT{D&^o&4{h4`EKc zGsG7Jl}#naoR`9egR!`fk$Tc0{=`YC2Qf$pvH4jBcscLEwwm-d!SF5(vA5pem;i^= znUw!5C_`C;p`bx(loUpAj&7FAb7sG8N--#dm?TTwJf6ZYFexWm(%R1R`cQ17fkv95ZQ@SXM}_W0Ny`tm$e>F>?Cr(8H~^@W zD{$c2Yf~;uc=Kc7NqTTh=67bkqmAeiW?Ij2wbw=LqBA9O6wnq|K@m@K04OBME%`TG zE}YhvOubvlLnsV_1|!anc^yRm&V_zW@+rW^HXss783g++6rNC>pV1xuX7a5rT80{w z!QMknCWWfqRl++u0g=_LN?!CNCO<2sNIk_^A(p6aJcwMFb<->lv?SSyVu6MP1xrQE z%93n8M%H~1(BJ1$au2-^CnK!kw7DUh3BPw6~r^KOf2h%_>vUZ z(24w@Kt+P=L>5Mvap5fcYwIg*KuIo1g6jzJraJ$JKxz*g{FrPl(pFbSW0yp_UYep>pCUT9Ciizu&W~U> zqAD3kRfhsamws)RLhb6l5Yw8(bi~^tU;Oq7!YjABUIo$qcxY! zEEeAgx^_XXL@xea3J?J!V8r}hcfn@@gR@-FFkFGZT(A`zz+}z1=k?EOqSGt2gbwvH zKLqg{gngnM(T4;+$mP-IRc5Dw?NeJQ9NvX@`q(_RcuzF5{b51iiG;zk1T^DYlZpIl z(vs3VsJ9qr79J};~f#|Z+Ti-Q-$z)a^+lG;j_~E=ayc(GCu}n8fK)# zuEmZ)?3cmg6iHn6U|@;}`VtD5{T(4&J^4jH$;%M6q31E#umsOAJ$o~H;4o|7F#FCh z$ICDm`3TRu5njU){=gBz{1M^4k#{>I?_Wm5$w%=VhD%0NBmzg96-VU~M-^t-Wt4`M z`$pB@jggR#f&N=$fdYU72nFQ-_X{kX-(YkGVV5ZD&)7NIAj< z(|D{}m12$dmpPvQMprA${#dQi&9`eLBeBwO*>1sQ#5SmgwhHzsLU{ zU7h{bVQ;p(^<<$)GJ#ULr|oQ|N~88a(A7;AGlj~%9hci(9_M>=y`4Y54r;`(pNTm1 zYa!`1W6T>mY#lF_Dl$C<_S~QM*IIm?59)Zh`ZkhaM)hH!@A>w0rPgF&p#RU~ukT`x z#1u_0f1Ym7e=1Y;{Qdi1(bZMo!ZG=MA#SaIQH@2gwQcvKh^?#kqshHb_hUrmL24n? ztc+T5262M_D2kF(S|}?eKKTC(iZuL}SBD~qV>6P2LW=+ zy+AwbDoj2ZU8%agcHI~$t?55<9aak!(hiyRTOEHLg|MdXzd(CAt_fGq+c{^w5ev|- zd^))HN22lYDp+CZs-Ka&B?@;wh+BncVOY;k;a-`wxk&bUo!1sNPE~18?O?VV1H&Q; zmL&syJIuEeS}7e#o{J7sK5OAv8Fd?fn^xw=I9*HT+-zUmS0LJLT;JIP-o5_z;Srop z^>@02&vEP=*@gd$6JTRV?Xp*gJX%rT_1=6ODCb1*E6>5_RqEP18SwbR^G%^A;}N3U zw1$Yu6)FzGq|JPzQ!D|@88bAvcRe*TqssN#lihW*If#?obzeY~CI5DSfZ132O)lek z7#nwx{+60+5|9c~yxjLxe-p;nu4^VUo_Uz6jzzz!0Pmjo8mBpIxgu&z{A(})6+9%) z_BI?zTl7Mz>w6M}(LEE;Za8kJAlJmJ?@1lMZf4u(*GAzJpgyG|&uTn)RD{`w%P1#M zSbZ8HHq1B}i#*4~5|qx*vY7TYL++?zr~fXVC<1#r64{h8HzfzRGLMOzKTvvCLB(~;;!FA1G^I-l8tKjlBz(T36@ z+Yb&Aa9xi#9`;m)iU9u}xcITHl{oS!(Of5s)&OabCQiAh=!4|-3=Jz5#PK0 zLaf}mo#y25s2+Wzu2T0c2g2CgT;gUuUL=n8Y~f%|PWxbH2Qx3bsTcsBnrq$CSc=pNo((^y@Mr9y?n<3lgUU4`~N5V!cpow+9e*3+k;jFpZL zn@*=B@(N+TIsim)5rRG~1IqI3BBb6xu4bQ3E0gN|3}zgzmZHV>M3j>GEgX=tBi0T39@M<0NG!y`0DP zOjZ#l2RkFUv{gtWsK25w0sGS{NX;I|UR8>1dNor`x6ujrXQQMQF4djBApS z6SZOgrmMu9k#9BmJuY|hg$2EQm+oXGXgs~DtW-XcwkWJjEujj50rAlzE%thx+n+tz z=Z0BPS7khY0P|Rh{h-PmeT7;cL5T;sY1WETgAw`*9bk+|V*kGN9Y)_Vl)gw=qrE=n z+sm;4vMS12ex%dB(urWNJnf3x&w|&DB-cIpg{$Tj>T>fqk?!yCWdFjdPgqFDvYvEN z>i20Pk6daJkQ!ALiq#kx&GOwwHbn4w;Zur1j*#D4Wl|E@^nrYVdshn<^M3Uc8{G&B z4Jj^5+kOf25!0#?a%OJkm;{|&>?iYF5y#?+LXyjBK*^b6LBIMZM!qHaPu`1RZ3+VH z{j^Zco`sr1+hR=xR(-U-zG?bRPbiK(3&2pGu zwaeGzcLM0%Z`In22cr2!3s}9ofNpg<?Z zrEhx#Hw#~OoC-3?Yvb-abh+4XxJbowNYx!2mA-~SCK zob?2ybqw3pk<+e^t+zHOE4o^0wC21hJM;oi>!T5fH|Ehz`_@EV_GtU|R@d1WmLeO2 z9mq^lwtN}e`Z+)HZJA6L#k{{M{z%X1i3@Sq7#s?!b3p}Wse{z zABaj?4mFe8m~socJ&$+PAU77C3%0?i)QGRbE<^3Kr`_7igWm|pbrP)u3-|-DvP?m_ zLkxix=8odFah}-Ww!7l+Gk0T#*v_*~F2`W7kiXe3m8y zzH+YEV3Y$#_vJr(a}Tk>{4}9zcOqhz-yR1LIDUn@9#EqaO%CC`LZYdp&|e=q#h-Sw z!&Kji5fX_HaW?!6o9(5hVw#+g46Y87dps_cv$|#dW;y&f4y)uUH7mE*S{^4WF{*oug?g7=bIUY8mc1socaHisGrNMHI ziTq%EChV(XW?V(H#Nj&W@b_wl=*$qOAAa{|S-LTUNbw!-jJ4~g$Fa3`t54NRDR2mhA5qyD!<$=CT+EJZ>#@A#LOTP6oTyLOyws?MU zck5TEZGgAQ5x1={VK^WSLeUcPBEe=i1TDDnAy^MPkoCy}1~giQpvS7O5VCBR(@*k( zUj@uln*D-?l+%wXJhVkMB1HwY_@}-EaGwU|OGhx#hTX4zu+8BABfvxa6!E?*jFvdq zST5LSKirsDXw#a;!&B5DHR>l*v?^*u9luW+6y>oqN)F|XuxAX(QiLs&?|Z=yVFIBt z*@n?-;FZ?M00}_$zi-xlD!yhoRUYb~7<~A~d*P~}-lOQx6H$)HA(YJI)Xc#a#KL#( z(TfmbUvFIYQwO7Lwe;-R4}v5fzLpG;QfWLy_+w=LPXX1jR>b^~w8e%_G-gp2ghXl) zwHuLMRuRJpK|3+A%FN*JQMSRnHXl9g^-tpi1+~pLhzd1>Yi-O-I=rt07_ukhZaS@b zo_yU-1J=s%pMCt7vu%~yNZRCrI!+bDZSnnSBJCNXnbjS7vmxRz|5EM5Bf*GjH6x}d zZF(rqeF-eHm`qt)OaESPZ$gO~lJpZPn#4D5`$=D2+}%tYmr6eFxbkCQI8VTnJ<^iS z@Ds0R2E>Da#m`n`)z-6ECQX*c`6qlFonMH=#_L0Aj0bnI&U@taU~d0VU9iNN+XBA3 zC<~yD3tjj*UF|GglLZ{am!Zw#1}sh+-gkL<$RH)kH1%^ewF8qsIx9arj}t0=TO&|P z!FQ2Qab?N!M9WINPL+m8x>RQQ>v;x3(QO!rG-$m(fzN^%s+asq0g^rbnu<(v3vcAg=_yy|K>gjG2S zb{N{0KTXoV;o(0Rq&xJ~OW@4b3Yv{PpVht&rm(xw+!JYyaLwe9Ievu>6$5Rh=Q`E4r^Pgucf`Xx15ToYDm5!zG{mS_nYP#nifl0TWW zIf6zUX}Tbxw8)+&R=h`=gH7xd9R&E+rf6wj$;wpYU#+2;gY=_j&5!;@pP0HC?_v=9 za1r}r5u0zjh!4FEfnTIlujPwW9_W?CjNil&NPk&v zxL$1NXy9qM-yz0U70gyZI|P$Vvzs7Fh)$>o|DzNckfU&jzoJO^A%&2vq_I5OsL?#S zLnKZTPS8CjJdlT^qqHI)-kdp%Uj^72NHw>u-A~H2>Ogd3M<6ZIRxBFW7oLq4Fj&IX zU&pOKE_fbsGT?VXu&{2b zKqU(9VO~G6??Sx{#boS;pIH)z-A;yqOT%y3NACSc79588ltwCO=1$s>v`QpJd;467ARuGg&{46xgy7(BJ3l^+oV_j@6Ae zWz^(o(vZ|RjO*QvT_q@7+m1O0;wM80I(hPYF?0~2nqKUmDq}pFB6{4lM4VnGf|nF! z;H^!Kwb?bEdd4x6li?*}A|&~_blu|T zNaFtU0t(Xdo{~4s(o5X?iy5;arEyaNr%SSCIQG302Yb()B3Wz7>fcbZ7sF)Ww>2Fu8!N`e<4{~FoOKo*a|g(7LWu; z{J;B#nuhtgHcXGgvt-9IcwnuP3k9es zw#;VDXN3d5pKr7Lz0TKa^d^FpMVdoHy4>K?a{&TBzkbF*pVME^5e=D0qcdsfGyZ%$ zo`rFkxcyZuV6f4q(^yo%r)4sL0Zs((>$~+om@1Y1`k{|=d-MAY4hNU~#_ep1*q2X% zVz2NHG}nc|8#mdnx=Zy$dwz%QC;1V71(^%q{2Dvib|HGm+%~ePUMKf>&0#tSiJg-A z7B`sDZt8#X@yRfPy+?;8o);T`H0b%hk}l*OP;gfqahccz2H%m)bmQcCD)Q%IeI}1W zF3#SM-gINu7DGEGK1iZkmjshYM5FEb%eM%Y_#aOZTV*q0TT-ERB7bO`=<0gQGi3<{$1%m5yLHg!q`&4^rv9vdwhcr?2rJDm;$STR z6hXRKT&!r$PErH(^*Xoi{TwY?DgHr}8T**M--^t<-a^6BMjjZ5vSso--QY zbDMSUVtdgSVdgZOmF2a@zkehS&izjAByz8+)9Lg+b1aw>5_XC=lkX|a$q*)KY;JDA z$ktvs_UC9T!;UQIm#Sy@1+C#HDINj~XOq?$*vXxbsBd|4%DlZ~0DOAsmaFsYokuzr zTPN`?nY@@zU$!-f8pXHCVkF<6;rg6kIJ~3YUHy>{s}rJ47L)7JvqZ~S7KZY={Y6?O zC%*+M|EX==y>9dSmRp?~h0)s2P4#@yuuR7Kxv$|tB^gL5qOeUW<=I6I!bZmS?ONy? zfwfmTWwqVnIsBD2J_I-n*R~Sb8UGti7^5^kBF$ck4I1VlpKLOllaw9V(*Q0IH?c*2SPHIg`~(6x1AmR; z;Jk>-ft*#ron4w6_IMf0ONJ}5AA;ggXef)!5GjuPWe`0$4JDMnRi6s%Z z{ZmIdTjSPY80l|jjIEjQ`hwB+84mn4CV|xTxqAc1-N?H1nn?FPf_{kbL4rVHJT0G{ zrABdp7~4Unr9X9X-u9d)w(T|DTan&3HK;y|W2GNIr&L+eyt6-CJt~4BuATTvC zbO*nT0`Zz87Cu`}j^*x3^VPu=6$>~@$rB|ha5Ldy5BcnCcm4r!Ka#LX2Zxx@HA4M` z3Tm4Txd$CeBh!2z1G$UG3Smk{;f7}HK%|n4z&cbtv1Qu6U>4f(o)OxcWG?sQ7~gS;|Jm zdq@}@$=lL0{kG4H9$7+!iFG+LQR%eT+yhMdJAz_kwUXZpE!NUgJzIOW3c~|5LQ(NH z60gmIJtSIyum_fW0jrWqK>#URT9(n^GDBf#P2udVlykG0DsatgyAD+VGC8Vr7y>|k z!-9N!^@uss7v_59nIEfG$~vUQENFgI5SeJ>Vd_sVs;&W$8>`^m3L`61!3S?;V`yOA zv!dVimmZcx*wf)!SCOv9+a_aCy z+4j+9{N~d#k!dFQ!^D$9Z27A>_}C@cYpHUu=^FwArg{D>%{sLIXcyf*_)xBkQw+Dq z18qOWI5ndoiL(>X?t9h?>TTdm1oJxX*4R6vhijSdr_?fB76MLa2mR$Fv!=e)*|g{E zPE--lI_*L%u+$f|=C!CB6dL<n$~7KI+V8hE=jI-Tyc1K^rJ%f;nkN$)!e{o~@^YSWT72OEHGpVbTh1L9xAwH$RE*fv|41VAv&u zdbCoQFVk%`#+u!NXw^qD&$+p(z|PM{X6)h&aj-)QqUpD5+@A9Okg<>4aGJo;SliZI z@)7*h(Jllt=ezuqix)Qkg51~_6s!Xxf!O#MggKA!?9W0c3p3;IE=U%t#(@W(AC;at zKMq^9M&7YoEa2MdSat3bGy0{1NHB5#)JjS6l=s1$YMAzIjvl8HMcc=DB#rOn)o-$* zQ^si$rj)>=B?8knS8F(Hv&_p$!ZYq>q>aBZ$=^FL35Kd|L)+>J+xT3m4K#CcvJ|h} z8eO&(wv~JGDz5Gd2wA^M$HtsEz4vE&!ziQh*5t>WOTgow9k!f&IY-@8L5(pDR=fNg zl&N)UN_ys3_bv%jIkz}GvI9aP5E$>pIVs>Jn*#M&k{MStHIIBT4!-R${c`#9=lJLx zQ?lQb8iy00<>l*+H=51fh{yV%OuV?x$HnsO+!CTnuH?6kIsHMn1!0FtsuLE|WGLjR zUNPbrEGAE@S+|LGV?j}jZLujDpY1gGx_7sHwB?FLF|gG#UyCcCLMjJklY-4Q3XWf@ zuuUUBSiK=K^eDg{^XabHW%VIS#xG`>XTp+wZ+Np2GpYOh5t7_fhZo zqbt7;1LmC!;6}GSjebsrUs_^MyI*F0ozyBEB;T%%fBtbDA<+vzv=|PU{xEj4F!u5b zxyA36P?N2wvPfdh!2b&!da{d-=$*dH-U!{1T{Dw+OBTH)YOnyIU;loH-agHlj_Fy6 z-&hQRichd$lnL$}n3-!gfM$g(oIF^U#aC$j+=#Zd(G% z67C~4J#Amgjg{c$OgAHLzPnbwaY!@?Z}ga3w9ZPPOqJs|HQKND(ayv%_Q;|{#q5XC z9=@k87HaR&ZG+K=2=>N3Zy3Shr-auJ3~T~XhLy2`zPJeg3&{PCDx?HZ0z3e)|E3C2 zf@uCj75*DVcwN*V|36h>%B+P*dvA zOmQ50I~InjrD!6~XeHLrf1PdT;3lg7`npsLeA&FU`oN|W^4EXub}gA*ui5(C?w*K; ze!UZU&09pe_*SM^Vy8ax8WF~Qe$>+uOE}2Wj?n%W_dWxWu5mhFl^hJHq~nhQ111(^-uOU z>mSdO`xuuaZ^x_5H#2Xq)U~31Fe0!2ys>Ja4PUSA{B?O7ztn1U%KtL{b-b4?^Jkaf zO_cVo1^e6AhUYuL>%^w#?+ndVy<4jZJ=bg@{&gQ@ch3dr>rr%JsQuFISAR|E(*=(8X5QY7#f6ToR6SvM>CtQ$Euo{S<}*S_|?D%UkC1BEgPznNIak8s+T z@}pnWDovOaPdY91my>LEEdBb4t{4*~uT?rq^XY-Ryt%J|^?6uzN|N?XELnsO3R-b4 zghXC7iYPN@KU!s34-%`Rze$(Ls1RN92{Ue!%Fn+>*1~LMH zo*BtduH_3QzE7}olxDkZTMf2x^6kFrfi#xLpEz}XQ$VGtAs~>xlCV;8*I)s<@xXeo z(0QlDZ`Tz$CF(aEIQxLE>L^U*08AGWq%i{&mikDW1^|Y@JJ<;DP2Y+prhFiH6i3ey zHhxl*gJiO3Oyp!oo8_3>T#qf3ZiCcQ4uf7oNhtKb6f50M@#QJG&jQOPTg5&8dRH4s zoNKii8e-iYLOC%OJ#9fpKCA{l6v}N@67ccXVx%$4E!{YpQ$`;&_aCa9z@j(ct-qi} zZ^Mf?zU<_1M&?W-b1Zy3iAyk?o;5K=Nk~Frq_trF1?`el~RhYl#G$&NlJ%GrmD#5 zqg8SZYClD9TawdW$8|?CMNb7o(ua$%piqXGa0Yho+Nnn6Om(9O zuun;Z-!FuO!jRY&AKJpx_r0Uv;(ZJD20vO=IJ=f?M=T$*gtcL5uceJ6KX-nH+l5o~ zQx|_Fj3lEf;DWA5%5}ia>4#~5SiADziSythB%1QNMTY`Lp9pW7>mtEH*5R>e+DIu|dQD#r#-d;CuMPtK6AaYA*^c^hr)S>_|t^socScDLykMmn@ zd}g)}rzr!_q{>8f_0-|e%()v7sR_!rODR&n88y`C7!nhwAK`@mvVboOaz?g+*bJ}G2H%XFBeH-VWY+kXXrXg+T z(iY<&66hKHS+#U78~kGuCZf)pTmS(*9!Ow2NnO{7$3`BLrcS1tpH@Ev%F&0Mz6;_Q z*LJ>6m&iPk7~Sy1-wr)z9ET?8yur7&`uHhU{T4>^KiE18t~kSNOGDvaxD%YhT?3(T zcXuru65QRra3{FCOCY$1;0_7yu1N?r-M81=>ArLS!nfXUt-bd-&ps7P*=OCAV1k7< z$m2x=5~+gxM5AuA)`h{J(G zJiD#9qE2N+$ZgM1YespSQLYi;K?@akmblY7VVS|5V5X(9FP2hYJ^!Kb&1GD`Dv#MO zya+6C_gn%3wIH39cyC z8DJ`A1i=>znZAi8H29g?5fBkLP75=Od4SdYBCZ+ETG=Xeki?ufYJy9A)M8!!s8kH& zt}#OSfaYTz16HAWTxMN;;i2sW23v6&>^44F|JRiMY)fXn@gdH;(W$pCx9sYKLvB5T z8M~yG%Z`~Nfp}pUNLK^U+j5u8wb?!E)8ATnpKsf&y8bO>#Z4!!QXz)#n{lWrUuN<& zu?&LWj9iL)={4pN-@Wg5ugp~Nha)?-&%)CtM5rH8@iIf3>X)ic^FI=!pRosv%;7Yf zYKdSuR)3(+p&043OKDu#cyWKwqWE;|f{Sfqel?>wtTOE=@NI3B09>LcR#9_#&%v8)LafIjA z(Rv$;T%+j+7$fgg{fvuBp@`NEE8p05{$2VwSvP^2PqfI%Ep18sDoW)GI3f_3d+=p% zwDzLK;J%}x{pLh3sUxeVvo_!CPf$=iN2?Aw_eUdy7;I?T-+c9UZsrhGs zOF$FxLYgiL70raK*U_Dgg94c~`bCX<;~z0-A$3HRhYjy!v*=$w@j8EMaT7e1j8`$% z+Ja}_WDu#`g%sysciKu@DomS^IESRtO-kAp!mB77zlnyrt+V(`n*4=4eaU+ zm|?<=)N@nC7mYaN!<5(0kpU9w+V`vi=a@nzlFxqXVXZDRBMh5WWvLVAuMzYK8+$|I1b0hwaX>;_(6ob zWf`R}5&p?GsLU2eLdK@V>(Q6Nz6W$h!a~wmCXFvNAgV#H)!oE{r&)3JnPZAp^A64oa(j?E32J-lt9oK8;^%XVYRWBiiHoo*F zzGLm<-bzAIXM((3;%C%E!rjD8%eYD(Ty@#VBNES_Ld=}<-bAYjV2#MM|5(@iU!|)5 ze@fM_|4*vIK73&NFRA*(QXx9YLkkDbX+@N-a8oXk2f(K%At@^V9TFXdiRO-VDMks`We@g!@>N(y~AAH zTir1Oe9kfo!~L!A9;9y$1t~LjYyJj{ZzrVQ7u)2X7}A$UwITEa;&9$2Xm_S>FnGA2*NZS6A2Ge|`H4 z%#hpiJ%yK_chi^g&ItQ9$&~MpOy%*}ch`5>Bm(#G*i2r-qANGT+FlOmMn$W=V~vTY ztN)$|%R?gsVEjYFyrZcs6abvJZF3G&F?}{4pte5B>nr8n>_1tpZ z=^N=-)d!bG>mc#S_1ZAa&MA5=VLkWzLXzkQE4UE7uhvnF_A4vj+@F3Wx1g^a+FS$g z#i#D3G31>2Bq`4>)MuA1^V>aGIe#~QrwQM}>9q(|s&8{cL%jI z{@xF1*RCN(1t0dnyGDty!?hgIQSaTvSK;-k3U%wa*4Y=Ywu0Qv^7^lGhz=^KDW1$&;;$;RC1Mb)P z4v}{IMQfhp?@{9)g%_(O(NrGCeT7(mSC;Av`mO^7gI*_i$+cU*;hZK0EfW87e4V=? z{`O~nNU+W0GV#0U;}X_kH;W5a=-iGuK6C6qQqYV-k@W`;Tp~i%_b0Tm;1(UQFC~!X zm^{99uVrvCqml^$k-GfI7VnaMj;7%gZY{MBFCL3rWVr9L4vAGG(DX=~<02-H2f2^H zSx)j$L@^SET+)#FQ-+szZ%Oi#rVtIJ%rP{M*14z)S3@S0K_exEpsEX6>@**5Oo5tL zG=dJFZmE<{jYNjjG$zn;n`k~SnLZ)S2;U?D2Vt}FzRDigfW~5aT+BvNghkLtrm%3{ zjmv1Z$Zv9M!X*ckNiAFp#bCf~fZjk2vZn-s)FCD80azM_5rObav z-qqB}v0K_;7>+q*mJpB*Yx@p{k0@t70vU#*NO>*Ec(WDxCS}S#6`K>?=Zq)VaNo)T zDF)rr*AhMp#2#Vz*ClCJp5aP6%~z7d&iGjl9|<=zPelvO`80;xEXAPl2h-ovYuHur zJu}0;TkN<~FOjB;@9-` z^KEp2FUC^aXs(Bua2%Kl(B)D|n+s67*a+8XZS1&Iq94`M1M7Ycj(Hb-L2!v~$g!xL z!l}7@-PV@cNUm*o8f)BlkeJZWr-tso6OFp|wC&K{l=10xMcOlFQk)Nu(res3to^}g znCu$k*#ur26R*R>90b4v@32lS?>E+X&UsoC-$03fF?f3{o?wtX!ih2to@zgNsk+G< zZYWybF&o22>bj67BZdSWoGeyz{G}#Y{yHT!7DUshu2#M&%kRzsIrT{J4e=4ZC#tP! z{fIgGkQ6ox_)bUD%O-aoS{gJ3CrPqNxClN=X`Bmm`p>0|d`S?S&#DsZ{R6Yt3Vowh zes-mGmVaA_j!+x~P3~9ccFw)ykAZJ3IZGHulhj{@Tm7J=ln3U~T?(vlCmaz3afmk! z=LR$A`*r>KREJjkeX1^6XWuvPBCFj-Ic#`-P0ExOUV#OmlC%c}XDglxzS5KaN=X_= znOhR6bCi+EoVl8c9T;E?KDU4Tk47opEtO#kLTPf+SUFNH<2%yA z`D)oDRjKjaHMsEYoH!pv$NtbWK%>hJ=-&B;@qKBRniAFIsLrGHEh}+}jpLM5_&tyF z@bF$m?CkN|Z}X&b7AL`&RbIg2hEgjHZ%t7+hE}1m-|MqR0}5;T8c?%IC;Kk_mq2`{ z|DM!YZy{Azyz~4kUdC~G4SV}N<=)*G;B)ivj}I`Z1+j&d3^AJy{I3!v3x7>m$U3M? zgC>^WE!-L9w*&0D2>jI^Gu#8rcIAQ*ECO&N1J4-UPnkkot@Owl0y!B2mX8CFeS$~J zLWKB~-kt-m@j{4DnO_fV@S}ooNkA8Z&Ns*QC-lL)r6KP#0*v)EiHn0Zw5fW_LaF5V zFo#1a*TSB>UB9P?NOSwM&j)ho8aJ1RQFejEvIFn=!ZelwZ8JmRD#M8;!a4LE#7`n* zHN!ALMmH!rwBZp7LZNFmcIeq2gI?k1(vfT@?m04EV@j1ihXOz=D<`~5)b4ztNm(;S%xa5(u$?H-}`N>C6+ell}`C> z7j}b+y6u)eL|sCtAg9DvBzR(Dq0Xk=dT;!!dc)-(6644;am$oz}~{Bnxx}8i><}uY@EZjE+OR{4L}U^MsN*tdiZkk8}!R zaOn?no4<|adCLc_|FFj@G$8qW#m{7FCU~b~6^y2FTv`lIn@btHjYVLSii;BEjUvzrbHmhw#@{jF@it5GVV?q?2g=o6Toaad^9KARF@7d#f zOKM4Cm2%taa%?xwuN{bE4e@eIUmdyRSUvrk60X`La3#Kf=)D<1gqOG-C0^IeBqDJ? z&9=G?WdjG$Cp)=K(gbhHjwAfW%i(;|&iU*-J%3^H?RsNBeu&NXSpR)4*=fO?G)|zA zwOA#6rQM*5M~1kotn-hhDgizB1$+R~&Wf(vrKdNS)weV=OBR+RFwq}6h`m$erGWda zLBWq%%itq7$#k?f&x{8Tm^0*u>7RQxSJBma4Y9|j(wDu96O|(gHMu~#qms)J1}CJ~BVyubwMR=tw%?Aa?z~R%DN%8^$oYcjk4y-#w#i;X)O6TwH} z(=WdiuG|4lP>Rwk{Uhq&IpUvnN8p1yxrE?1mo_H8ahq&+qZ%9&lN@H3)(J*((Fz+% z9FfmXY5h4H5Gxg;4)Z`sz@|hq285cSo}yGvwtp0y*egbJkH!SUN#Mh%QLcY?zCR}% zaTJ)M!Cqb>rD8mY0nAeZ+isIx)#n1lM6k9IX+?ARD;<@-%x}5yO7fn`YV~OC3Y9<; zq}92j4KiLGuXz^onQdDGJ*!D2Q4C5 z(e%iNv;h|lnRG*bOdyp8%PQ&A2+=ix7I&$~*z4MVscs_fs7M_uFvaXaWUT_}g$~9@a)Mj_! zRIMyFD*uRpPDEx%X(0cS*8eO;oZ?=fyL(A;)*_4jqM1E&qNW1HUAbNreJ+{o8~6gx zP(AUfC)DSlC|P21HnT@Fn9I|aHPAuJ`w1iXr#6dl=!&c$?R4S7(P#6&XCG1&bzPc^ zsnIT!0h3(7js{PB05y^kt0cPWNCpAgms;=?RPW0be&RXABJIYzGrxM#hNMq;=SCtxl!lu8a=0u1PE8kcmrfiX zCi-&fW|_OS!;3HUt^A8p0@hCzuH>#YmK2XvARQ)k@q(!<0_{EYw`T%JwHz##!a zDdpvI!A$#vT;QbF^m!p(C%X&L!hYE~w^odF)>P#BZPKXZ#k(~8PkFOXb8jRW%F)u= zg*!>?E0APY434TpM!_!&SiJU%hcyV_{&gPKDv!_OfVXQo zRbrM$^f`@SF{dN6k1P}03L|PQ_0i5m0f~pPcqOU$XY$l7=!kW)cbhAzI1vN|FbgSw zoCI=Lk1-r)vhWPf0!>o>GPNM*1kqitd*@jD0op??7=6sGYok`~w2eP&iH{&}>EINL z7|vBw*4P5!;*5JrIi%P#Lo#^XkV&peg2fm&GERDD4Sn|Oi)Y0)i710ccJ0TPgoXzH zP2D@;$!#nAizJG&Yruf1`WhRwGx?m}DHvJ>FJ9}Q^5^`55S2N`z4PzSst@IwIy;^+ zC%y==M(taznaFrZHC?sSU6QD*=0iui*)y6$rS9f;Z%2@0jFyKMw;Mklr3*6jB{OOH zbgj0v93kd0RPNV^8VIHRXT3WL1gfL<*O_5=s=KDAxY|_$zaidJzHi@lVIB;!=qp*% z2BmmA`*Pw>xAdfwO83dnNJSo`lo&5B%jEhdY{zjCXQFb))U%aVO|-t2-VDJ4!lW&} zDhs|OvLWlKZ0+<8kF>z@RU8h4p!`f~v}V6C9*2g?04Z93WEyqaO(1cnn%5CR^~1{F}-Kz76}@>FFx~-qQuf`>qYwUy!B!A?j0hF!O}QH4;bUqP1!Vqk*N}g)J^4tXxnritIH77-+|A zeHZ1F6>9)BK+3;ViR(}y+yw#cGSY}uMU?4BlSf4}fG7#Q1PmZ?>6*a-fEaOF3Y08! zn1_FtT-cLlY%2;|1zxP8tk)rTaA^yt;j#ZoXu`28f!IihoDZ;t&aX8F@4Y+`eNo)F^g}Z)Bp2X^U`0sFaEs#v)`H{9txZik;hh>1z*+3#D z!awqo@&g<^Rf6yeY=~&gw=_PHwz$w1C^as(K{k}Fm7S77iSAZSUC!u1#sP6RO7O&k zN|^B64&O?{ougHo#3!{sB=kHil_e*YSy-!u$Mp-1j>MogVX?3lj9(`pJyjiFyWlXD zty*Um&*=KlRKx>kIUBDN|JhJtjAGegFMFhVt?<|eo$)1=B zb7n&+<1rvIcu5B(i~Iw;rshgI@;)OEov^N?;!SE?A1Yc~!ExLouLskL+PV&0lV0pRMGj5k-(n6Um7Q){t|UV6-^^8}du798FTD>|fcN zA^FGmQkx;NMWh5Jv4w`nnOl;XUSoFkR+%I7j0iLz2$sYTq&^&mqm~wBZXRYz>ZZHq zI1Ql5&bM*+{8GB~{Q!n4EFhqwIpC@zWV;(^^pnQES!REnG>%#o1boV$LqOd$$LTB; zKP*MNdCBG@$+54}IZ_DB{KeRt5_%60cG3ReL{Gr|TZ)2(^uAg^lqG}NK6lqUSFlW} z)_{;>xMVT}G0UUu5CJu}s4Qt+p;w?B+tB+V$7$vvAD=90gt^Rt4;%KWSR4_PlEp0b zDF3>X0uIejg09d$vqUGfJOqqwkpBJv0ks6I`NOwF2u2o1b60$-U~weiVM(m|_)~Ts z0d8|n_B0&R5s^R0x8g~-27N*<7DU$@YWvZiyip3hJPWvo|M}xC4si|Ev^@~=)3B3X z6EbDlpA+&m_Z{zYcDqDv2ozO^@52Bb;{F^(r4;Z^Uag-nyWLh}(wg%bw@^@|@Vlnp z2(jwLT;Z^ES-3t>Q60TUu|Vat3J*|SItiJH5-tS>){;Vhut2?jQyv@C;MNpg)KFIE z<_1mGI1C{-iqJ-TM~KP*Pd1FLri^8NH3EjpMx_+(&+>+IfyMAe0fWuN7EY#B73Or! z*Ug$ClT}-aehflX*Fp>jUYAGOZxy`eT1gb3$ptymfXMylzIF`Dh(;0G}(6u0Jb}fIo!1n+Uej8DP{EoX1bEQ`!U~ z#FaE981~t`b@Gb_`j%%x{kxm~zdH$K|7KPH@12C0|DIwr|Fd)e+2+fCpM8~UDVt0M zsHQ60^OR5g6Mv`Ao;RKJXMyd`XkK{lBpsG2MgEv<1BQ@x@VM}SP zQoCI$zJYy)gmLR@*r-6^E4Q2el+iYSWB^Z-`Hab5cCp$Q&`CWxZbgF=>G|oZsgYx( z%4V@*Q@!K4?l=5f@gIub7+LH(FQ4{A6R+QLjk>SO%8T>7u*rSh&f4nLM&EvScwR00 z%fZy*@kH$|H@{iQavB8gcNLmoG(iN)?4CJehpgkFU*TYH3Ix>503;!3=nH#!js@Bk zBpD8VYvE@E07=3k43C37b5RV(O_6(ET(>H{n`3RmH zm##J%hBRy9; zf`-WcrKUl#vfg9L=(NKS{pV#P8U{7T7r4B0?wShPC61g~o~^ldyg%}d9X$<7T1^8P zHarrwcy(-p9KHD(t=AtMTKXh(S#R^7(MOym0Z14PW8xUn58!wB?1qDMoQ8KZNPm6m zp{y&l&akUlMqC>*#L7+q zsKN zJ)NM~Pk--%ZY8PRy(zk<={w+)Brpzu`vSd2LZSvU16(D?82i$8d+%n%f4|DT%}G zqv8J0fP;abLXbp@;ITj;1f;olH~~KaG(H2Ux}Q9HQph(Eo4&@8oY_z#&{PC8Gs2fu zQ-#r{ql`Fe(L*AQF-tL!N`$Cy3S6J-cvu2PK6!ANJR9B zYOG{A`4f}H2Ipy@guOxaBWXGGL|xSfq<0dFLK+E@*__(cQF`VSStsxmV5nN8YA|_qKt6)6y3$6;`XEa!htG>OqqqYHapTeA030q`L zOX*xT3AfyoLdFm-6W$bQLLeBS+ZoCD|>aWa1rQJUvD7EuF> zK6kN>k^fKeHxq2cJj6rBkuTj^pH`0fC0K2FLL`!$oe)SSxR_`)pP`=EGbz;m$Ha$| z6UJnkf=b5A+-Z|>!A(Q&Hudd!jL78Qt=FVseLJ0VLVq7AFrOw6Pa^$C2v|ntPEiU6 zPylMET2otksI(kviz#YK-?h|4h?Px%wJNFWwR6nGYJ?XNCrbtil6bz=)jP2O^Uoey zrS7YpxwJ+CQ}^HHHQ6aKq^?zd(5xvgbFkElSq_Hb(br8@Y6V|_b;^wMl{?PWX=u=g zl5rCqd8BoZ+X!>p2@`2yrOR(E-2vj_93`C`Uad;o|MSGWchdxF**UNTofHZm7aVLi@qGdQ@S?Sp6!;FV6o%+7qEu>MT zD|yX_kn;r`2EL|ldR^iaH@2` zLq6g*jwE?5)NzPMyH&~P^Bf7t$gyHkQ|{e|%%((ua~cF zMQ-V;no=QaJjQNkDuWg^r`mFp*}2g?t23~w_nB$EP9 zC{UDX{m&+n#prFJT-@>F)sHu9A#&l8j}Zi4s0{}S;l7L`!=L>5i51*6MBIN9v})7- zGz|@$I2e_p{G#>B2(rsRJ;@}J>(w(bG)f1+k$zs7$JR>}NYbO4i#bKdB`LOJM zH`Ps>J@V)G+q(7W+0F=j9ED_&7$6@;O~pO?5)`dmMPm832y%(qiY7XNlZEaY+2beb z&#cCp=d(2w&%VCeFjNsgLWOWiFQuU1ZOX2Ah4k2CEco|eDFd6{?x@1>*2UwqlSx7= zVWK&1^4gd;g*vrG?*Z0({4TA<<4=4)_lYu8{B8-_aLG}bhA#P7H3Bt&2LWW1Za;g6 ziSI|9^Xr)0e2A2~%K$fXbc1Vq5*g4fYxD&;&pt(g7@NB^oJB%gGu&xgj1hO(L6ts} z@gS?Gkg-16Cg;(2vn@J`IKmb+oM$C$V9B%LQ1v3*k9{>piUI2q zjF+gd%2JLs>K$+w5oIO_oPPFN)07@*SI?2OP&sj5eP)#EVnv^El6;E z>jRbQHi%iY!mo-BBZLEJ#TU zdZ_ECgZOaa80cYXQXlVtcBj7Jl>|8i#-N<^<}M%Q>cmkmTxCJr`KnwzdypUs-t1~* z)3`w&UXB-O%smO?iY^X{tOZWW`_z}b1w>D=%2Xm(AAI4Ilf5wWPP>(y*pSKaz)p@o zi#b@uLJ2vsEU-1nC}QgiId9ma0*ua5w$;Y|$Of*>#v!v1Wi)3AkI$RVhUUcTdi}F{ zgm;#_T~B%dU?3M^%9GtAZ|lXoN8l5HrfAW(s4E(Xj_y<72?8QXki-eiXY)n7Te5q< zQ(|<){EdOR;TBkeOU_Ng)>FbNLKJ-L%2UI^%K{8L%;)AV6%oN}mwF!$;Teb~ZT>}u zIS6E<#og>6IPM@1W6($3rBO2kDwt#NG68jZ+;n>!H;6Sas>%(fa0&4%Hc+T)EreO0 zsqOqK9C9m6o1rdpz|%}Jdyz`VILdic+K+TDX)Rzx0=@ur+`T=vtngBbTRWB&TYYqf z#0Ahd*dT!`ItjD}FOrU78=mb)i{L4@ONgsf zZ5E%N6#nsTje4tJsJq!1ZAAx7?d~m!5^qIVw?u7(P}pyZG<;sBd0?A^)UtGH8gfl4 ze$BXIoftyNEIq^#-;I;EzS6a_G538?O!E0&sx>n1@NcrXJxbF3THY9+5_`#pn)-5w zy15$qJalL{N}4t|L>gL~;!E-_ zVR8*UiloPWcm(mpDxS?9S;1EKG|mep`R|vD82sBxdPvTT8^xrM50I@u zhCplTst845e`j;F_nQ z3-oQgUl;+5sDs@pi=|WNrBl0{vqokg=ig!Na-#jgr?LEk)y%c36hY+|Ch;Gh7zl?A zR{&S^zoE%e5l>YA^2G5Z#Bb9R+`D5*)KZ48NE-i!CjZkD=l=(q{FzG}K>2^sWRt13 z>cvX+QpLv@trP7sT@9PQTxc1%(fGg75hM*$@^0U{_d>Xy*~G6+wXa4e}tOD z(EryWKAvIvfXfXPG=ZkW>^yP3bn2cq`>ImJPk!=>)H#cY3K;3z&?l$0E}6IAm zvA#gJg9h>5&t_X5{#S#yTAU_@U+&&cFZD=C$L9qba>46KkOL6i`DMxVUsZH*@i37pife2G0~uVP==C`rB~IZMS}Tt=)urJEp&w_%*v zENDG~4cV?ZVfeD5*qRdA;IlQ_Syxe-WTq@61EGBan(3f}O6PhO_{&elN-ID?d8SG*$Q-o)a zO7c)(>4d6Y4K7C=kaaADHpjx1W-6XlS0p(=OYk8M&{`yC*nDm!Lf^_pT@?B!)`sfH z8W#0zEMz;CpuhO@vJ*pD7cDVF9{FZ$ky(drqG97kaneZKRrwFh%)kax6?yu2Kib%R z)xXwA)=+PB#CmxRir+tbSjZk4R5l7@D>cre%$4T_aGirZwLqoMttFHMq1};2yUw zxChXh!<-5L?X&|!T+6&b;0i!xt_zmAr8KMl9;+RUh5xWU0H7&Yl;tvFB&QMyf9vaT`;;FzBDLOZLZ_l>XX^2`xxBN~D_Sa(~I4_@9LdVwLJC)?~>Ya)S zPCef;6vJ9z1M)fWY48e9dg{9#OI_N^R;mwlK2XWI=cnfN*3rJ%a(hJ@Zm@^lIzZI6 z_t1X-Lx-%G+ql-UNBrsGjcgDr^eZ65*zdH8^E+=ht6Z|hnixOT`wWETEu6bly~01w zJI^!1PZBuJi&qAg&gzpFrBv;I)M{0W7Yy|nLIzh%)hsWNihqc`q?hI&a0#;G^<7d+ zTg7f6;(d9dm=zpb+?;(I6GcZ(3j1!fxOWch+Fi&N^|l!3YYY zpJV7<%*}wW&zjbZ)G{c6?)Ju<#9%B zLr^s=(XZ)cubY}8(JAC#51!j}_EJ3c^2;Vu1v_50xf8{dAqNMdibEuOUNihmY!LUXTM#6lAD&5Q z8fB^obY3u75_2i5@ZDFEmqRNgjy(UR!E+6?4EsCA%V@qs?0_$s@FrHD>M17f2~m2YDztK zWw=71JyL972t@D;Pyw#rRtr{~*!YD|B?qz7lCEV`X% zIR0Dvk+j_edgC?AEjw)>Ip;zo_l_^^fU8Wa01$1pK!Yh1nkEoXYs#wM(Vz1Tl*#Kz z>aM5z6*yt#PRPu}y{bge_dP<6pibf_SAOttIq9#HdqMaeNMt^QMC5V!eP1>Wy7UDDt zi|5`ACpLUyeQx_o1gVYKcK*|{xg0j3OML;BrX7h#rBB%JuEAF{+(cO)N6zd7&l9m5 z*&9&UaJSp`C614@^OhzekicPcaRIrp7q}67Xj-o!&UONNx@u0Bi8pA|L`Fkc5tJeF z5h3uHsM2K>Dge|>@lWTb+btb2QVQoERfvs=cu64Jn`2SJ6y{*3;hy*YE~dX~^xAW8 zd$vX!yt*CM_7UqWU`Wy-;qVF8d^y8ZjO7A2x_dMjM59uJ*QWqm!)oW1-q*+IgwSGZ zYBO+OoN#nBBxVQV;irVQ#&)MsgcxN&j2w)79t;_`~`#D$nqq<1Fu2SHM_ z(wFfHVN6sN`EG;rf*UV2vF>6!j2_qi#-`6tkfGT41pxAp&>Oc6nZF@TnaWKjVStq7 zFd_r67YguvRZFfB<;VH)DX=c_p(cwTfY#FJRtC< zVb4Q{fVx3_$suvMCrkcX<>H2h&g5hp^<#;%tja+0&3|9n)^BEmi%W6}dTyem-y#h! z5GZ4E$#&M(F$=?WkNW#+4cUCmLYcJ9!xz1Kc3tn6W>=iRu0NOd&ax3=Ef-&-+2=Jf zB)^l2!Bb3e^g{7)I_?`f0Q{CfS*>sn7G;<9(t#q&VD?ctu$6Y2wLM5 zv;kYj-<6Pmav$fB((uRftQ|C*3h5-TLFl8(dYe zh=k5FJzS186Ox^gB6l#2g&IvWY3p7HCJ2b1&B*&78*)@|aB%Mbb3+cFg#gw1-^LE{ z|AHszMuMrN;x1ksspZ3&BJCBIPbugQwP7Nk6$Z*9d(^r0;vKty+u| zhbbT_E?e{#qbu!*c>bJ#M?g-CLKrqm&#s-! zs)O-sO#E+FvKrj*r}?=uE$7;xm3XEk+Te9gY)n!0FE0Zjy#ADONlxx=gDn+8!HIcm z61cqh?~LW2v^ z1B}fL4&%SF!Ngj#8^Se*Vu>4-n{=N2tMNH+=zpeJqF632Qf>E{YnCY%;APoJv>ksjr%4R?p9|ag!Q+*_sS=#AQBo0Yi5Q4i(GCCYP~$lfdO*Z z>w5)OyeoF45Cp>ZrC4m>Tu7R^@TV~-M$LEuIGmDs zfU`kcJ(5sGeE=~U5pp_bMPpihD5N=d3XWP`AR^)q~omLh_v7A+vWjdTyS5?%U)zo#I zoz*st!C21g+EyIS>${F>&Ktfwov}(PzKEPQ;>z^Y5{xkBu`&Umc(|?KI^ycUiyySv z+o$@^l*Hd&TAe$8cum)L?ngN`G?%XID~Y`|oxAkj^c$&(-!))x_V{{jxNu(GDP4cP zUm0%v3c%RH`Pxt6SziS|zRA&t(~IFcOz@77Eri6~;p5;R-vXiPw#=hN`V8 z^6L?}PG#;?07%GT90BEmr|;cr!p&rH4cqN^TojY$f=3E%59#y03)cz6<1c)Dj7D)+ zlNzW+4c{eOG243oK3YLn2{Iy`#^F{kcm^T0)4WysT6LZ)*8+YX5^|kJ(-I2Abq_nV z=df+>rJy9Cc2^{w!1iIIuO_!!T8GPQ^xZv^8_D zjJ<#>F|@yLBOf~dnrM7h;rfXxCbIOC>A97AFOGG`Wz|fNz5|lZn)q~R@0jGU*%v?C zuqEr?DYBC_?h-iAmPg;E5rJ?0a&;#l?sry^q821MoMs|=WXoCqTD|+8^X(qVm#RZH ziF|2V;txJzH%u9W`T`Ne19fKgCuug3yXyqnq7{AogrGk{Kf#YAY3+=kvx>lP#~c(TR_U99wV2#v5^9Y*`x(=-xQ;1=`I8{t+P z$~TnD!nZXiM!ONgAr8c8G|}1sb|VmSM#sg_i;@XQktDD5f$yoLF~Slu!4&#HS-c2> z*)aMxsxU;;?I9Yu(@eOqE(I00yR;P6c*64W;iMBTg{rbtVsF$GMXy>nz(d~Q3M5{X zVIdq~ULf}x+O$kqN2G_q_%)=P!`#csh?L^pUS>7V_X18j%}+mfH^ycll4Ues*w)U? zyE|vr_k+<+%~;q|RSN9g?T_l#dx$jCyC<&;Zrm6diTAn&)$a?Z%Stp(kYF?Lcb9&O zP#nw1DdSuS_K3u;Ls-qG1GVHjwNArDcZq{K> z3j*m@RHPtA49#xM8$V~99U~M?{$2WC#DvE`iHSXPb9^QV`k$2hkL=F&ty|#KPtY>x zq{H7LIoZR??~DIVOi0!^l0#l?K>ZtpW##W6RnqV*O21t#ohk=+1ybP?dNw?32lSCR zF&qdgf3-$5AFRr?(i3#uYk>X}li&sd@gmAT|H)Yao%G5>s>|s_ znUrgzRdcZs^L1Otg-*~7dL0OXlP~rt&Qe3punnRDJ*ikp4w2>g$fHO}!U2%srlB>Q zBtDKSxoAfGU|Lbe=g`CkapBByn#*gKH+AXl+F7=y>`Q7om@EN*m!FHTI!x=YICJwq7Pd;+Zvaz|i>qRq9(+*uuk!f#k=#Z%OjNHP z_nWIxGN8WX;=1&4uE8XVW10fu>rQkuM_-}rSvgS zRV^bCI{kw0GbH}oUyOAbt4l6l-2R=7V3WczmTgW2?Q1Y)JujV5qLQ*Y@;iIz z)G9FAH5Vp0K-Uu;v8(Sgy#Lx1{nhp4Zzua#T}@ugKm3+M%K29^tz}ehyCxJo z<1l56l4Gnc?mz`j3=WY2RNJdYu|VLsugdx>vaaNMqWWS;TnHnGt26zPR2ipp8PdJY_o6aWJS8L3k%>V~Q zX;VRDq4XsswC(kn-8XJ2g4_Q_cntICtpItc*pFEK*poJDUAQm8rv-;F%A z_nJ`qvgC>qE_ik6jT9&&KgmYirE$94T?gU3{>W?Cj3(lk4O4sl$+>5-u0N^xqKewd z_}O$S%@|IaePKVAH>H1VpY8LHFDd#9WAm>J?3KrAYGtVvt!;&mtrGH|p1pmsS?Jjt zVRBdQ@5txHgHm)KL~2Q0D1S)xzlG|eo*{rWrOtKW6n*K{+bEx9f6-??-JZ+5Y-4*{=xn|^h6jzwYR?=chK&0(EM>osswsyg%llf z5Vd?DwjFSB-soC3Y-ue%W|U4hiJwG$i0Nu9D`FN zKQ7HR=(?s=PGZ)F*m*NGKLh5qCa1 zV!>D|Wd)tM}B!!&L!xn`ng!4b3U;?;gxZM9WIt>4R z(c%ASHB^YiRr`x=hdr3SpDFQZJ%L6r&*AQa%K_((nLhtiDh zB#6(XLX9wpju;PyTZ03Hp@RJdE;k|}NPwyhP4w+vV!AWJrcTZuUvWdPMjaOm+5P*0 ztWP@M^~A@a{#|3#9Q9PytY>A}U{vVb`PMkS-RgEIerQwx;r`>w$47E7s-eR*B> z%vT@h9f|^ur6{>8$noI!*I_vG%T`K+GD>TX z#8+b0a{DDy;HWPcV(U>rK9093%+e-h$t2w-YjHY2fj7o49Tv-?=|hfL!lamTDwSo9 zunuyb@zHm(>E=z;voz>99Nh4HC^OULswpeuD3Tl<6xncE?cjc*iRRg!9tX^}4|PD6 zl;GD2`>9rIwIyZtm^0u)R~ya)12vN1%5u_A!0J%{YGoU!x89mg4!7u|ZSF*H(tK%N zw3DMBo|+YxQmG`8tKJDTXy$C&_E7PfpQ$^ZzToB{SEmL_nw;T+$;b_BXBp@e=g6L-B|n3WWN!AM%cS+rqtX-t;!1P|Iz#puRZ6AWtUTDoZcrnUg`uw ztGt+CL(e!Q7ei?0+^v8=R2n)#dMQk>517zm#3&u7oA(IYs$#|j(A&+}0@phQF-e`l zgP4A(rpbXmfevuxubXEZn$yD9(?5x)b9?j_<0$aF8pNKvn)4y}-45ZOEMAP+keU46 zp(Fd$>4-VLVl|73LF>097x)Tcbo=E$oyJ&~W~7Pod7J*152;eIGHn@cYUe$u9gk}1$J z0R)rk&cG!cxs)F$%>;z=F?1kHfqvkFBZ)EUY?i&9TQaUP4zIZ z4WRn^t<1%MLx5Ko@Vrm7l!S%njQBu@teq6~!^T!ogB{yR9==3QWCQf_J2k%4G;FK?Qb*W$~)k5HT%CfP!bo}3&MM=yelZ(-`6)Q@j z%4PUwux5J&zv8S>h0m)N3;N3D#%jIa($_EB;T2ckGZ2N&Rkj^o**w9BgS*RVOS%heM zhM`(8Sx`fbm|5>L{I>uSuD4gOM-v~t$dJ6gYPm*A5g2?{^9&*CR$Gu9-8m~4(v{=1 zV-ZScDUHxMoYKYJX6UDY8QBAD7@`NW5k$*$qi9cReGKrEIJWOtw$mkZ@X()o7Bn= zRyQ7Rd(hgFT{$zenS#4}hBSXVN*ia+aozRx8>&dMJ836C?)tgGNmA{Bo7=Z!EV^dy zw!Z|ouGn%1suh39ir!_E7ASWvH}N_v7p;B6f%VJnr_Px%_eUM`a$HL{g=J9_>jdNt ze>B84BI5xIiQf)mW8&$CWA0)TDntZU9OP33f>I4BdTluqb>a(6&?18dXJ5fzwi$!o zy}Tc*k_Lj8sVq@Gh;dfV$<1R6eHLAcpM+8hxVyWpt+(6Nvc+H85~51!zDme@%{ZQ% zGjK4JufQW`Cnh&C7hBQc>&H&d9GS)TVII0tAwp8cWNlshtZJVr|}eY$Qa>1V;HD?u_GqxBU| z!`+ADr^5Zcz7ifbgilPaAavvJ2G3o==lJ(6{J7*y(pVCoQJ#gZm4z0y_4^pa&I%2R zX_WPiJ?4i3wp1?Ia#aHBCu`q%LT!?rg&gzr?Pa`cfqHK956(l?Z<2z9rH{mBDVN3Q z;AsV6J+I?F_xH)%y{&kcS;mw%kIJFfdRJT>m@~D?#%N=DfNvW&kmk+XL6B9K=8X3^o=GirITk^lJDLSg_|;@rdso)_64)TZU-R64B=ucy~RAHxCbLNw(feZ@c#r&KoNW{@Rz*4%im> zha`5<-82>5u8}bEe8Oly!{-%im(|Z*!)m`R=(J&#^5L=j&Ue>O$jvOQ$fZt~SlD+w z$SThYcSRn}JX2MQi>roJTp#95s8gu=AVcLXFuYpLwNIi z{>luLzOS>qxE_i)sInYNoJd`^s>^T32RSSv1p>^q?{kK(Hc){{7tM;3-XZ^PZ=~SD|nXZQAILi-7-&RkV zKgBYVR_7FoVvlrrPI77Z=$?H;c=@8XR&~}xhfp{0(KQfWY7ERMcm_N=ZMG=s%<^v< z1xAJeFCS)jB@wK%^`#iGbiDn3hF`EQadbX~be%2LgOA1&SKji64xZjLd~bQfzcn0+ zOGxmyn?DAb$!+<0pznHBT9%Mbt0d^=%n{>J|NT?nUF0Tm6=`nXy|Z?WRu-s{tNM{gPU^l*E)+%djBaBYbUuPK((+O*5_!2Vgy;D!5hX+3&ylMTd)3UspeMehRV zX+IwTS-4k-Q4BN_AUovnpN=P(+CmDnpxbGrqrUIg-F2u;-f5J0Z_r%TZ zWeqdibEMHEf>Bn(c@h;!AGE{}BmL&p_c5eqY5#@S5*V*TW;Yij7I)$^CC~y)j?&i{WBPr`6-fadLmDEItEg~$E`fJ`s zU`m40b`RGo0IE92W^rQp!3W_2LK1rwPZ};n)hK$2TvabG)wqhxGjSSIKdd?b@ImK% zr{7Q#_0GI8ZQ_r|KolJ(sQJ4MM|RKq+lPj{!=ncNS07IezCe^jB|R$_7aY$a1<%Bn zW0ROSCy>Da376VY9FnSGw3&E)Ud+KYwy>+w{;DRlU0k2!KV;dL{}8r2`!-bEY`2lw znS8PJ+8YcOMehh9;;MDF^WoZvR=s#6gIg$7i&<2E9U@ws&leb8H1Z9^EUp zEdahDw;OH49CpB`r-{ur06lxcbx9nVNTpn;FVZi-F)DO~k=T)bo$uvy9(^Ib1s5IfkCs zVo;L1wcLCTb*sw_tNq!{AGhy-ZEEzUzXE$M>>6F#67>5s@gh!K5q9a%ThqWViGXeL z#RrmA#}?4df_@9)frNHjw}fQmE#j}g%@N`+`uw)g)n6nJ0*MOPIR}mM?U8CTTBxU1 zwT~CW?GJpzsJBoUgJ?EX5v)KGy3OIvl1mZxm+QdYZZ4)_`0#gkiv6__I-gt-RsqdH z5Cji*vS#X z#*vIz0hGJea!O^1x)eSn4f$5`pBLn!?q?VRUT%TDIe=NISmIf16GdPs*jgc(q$ncH z=!~55Y`~^8?t?;5)fmK&Gl}8H9J+vm>Z~bKG%c3K6pnp>1Xc-{(GK?N2U~P?LaS|7 zSFt@iN(SEPY8zwpSE8l}%HgNYITHz%L^ePM+-`{o)bo*5VNaA#=14F9^n_%GN{`UT zFEe)*T}fz%GF6i3B1MpAqe9CWsm06H6op5!sx%P=xBG|Q{W7Q)q0<}QHPJEv0@)-> z?lpwk_e(M;l%=2N(_(#usOU3Q4%JM>G3qc79NslAJ*U$mThXwu3~I%}$8Tb>s&7kZD=5fIJ+)fDur`?p|4CePV!9(+WhR&jB6*Xs!=dqgdOA*& zr7k3fx>buyBZ+P-B~5`t0E7bPxZf|oOc@oI!s=fJ) zM;dBR!=jdHAQ5A46(?s+zS<5#RQl+0M=ShFA;ZA0>$s}E$|!SeeCabo(_k9TDNdY>_0eNPvWe$5Yb z)cg*-dcsU&d0zua&Vtshw&z37oYUANcl!Volhv2sb) z;~bDxi3`rNP!$=W`;?buY_8S}_B#rk;^0F)L^NLIlu&M0@gcWww45CWeJ)0O91ku` zrLCz9>${1L)cQqT#f|D=Qv_xamRuT(7xDSa@|mGTUM3P>&d<&D&n5 zO5<>tCYQ}+43cmkm9Ie$;em_qjncHRc<0`fIkG5o&4r%o;wno?>HW1{^JKiuOk!fO zD2k{(#6vqR4$o%}C{uyGqIGq)`QzriUaV8UmqfB4zuw=yTKJ*js|D@BIv2RqeCvb4 zSEHYu7tW&DWbM61Hm)3rX5ZHP4QPY(6V_p4VpormMJNBLD&yS;z(q0qR_oF#VK~`RLJbU6{%cM#7x*UAJ zlRrqxtwPx+gZI2A*TOxVwY3J-WvJ`3>^%Q^P|Ra)`^Oj;cI6udeLjEhe!6#YBBN^6 zVC`o%ivP2d!{;S|Z9r+dFx@%Ud(iaF&zb1^SFG^1$vGog-aptcttgr^e+)%j$er1F z(#$?h(AedRt`)Ia)2Jn?pqD=ka~h2kW_XIEYktleJRuwpJ&vUPag_3U1XA!$zEt1W z1=sH9nyiX^axl@)Y(IJ?L+|Ye+mFvR!xX{q@D}(`Z1Wxo$DXMd@Srn<*J^rS5BwiYSmjz>+KIkvEWY(-0N9=iWN5hZ(haV`|k)`?$Gn#fHR^_F%s?9Q$P8) zqxlN0CbL)d2Ez-<24y&5f$3b|vW>qo9n^Qao3~IrL2m-jw(?N1uIQ{Ue&(8WiIw%! zZoFs))ZPt{cvhxw~vRvZwueNdqf~ga_eP+lH-@gS6n@l4c=NrdYvvwp%_y@ib>?AYxF(u9x!); z@oNLWV2sfM-bV3oPLla~M6nJNmx70fCGNabEQm-R74jVlQnA3QmLwU*Xo`)2QN(4W zZHiK??Z3apQ8?Hc+bWt|%^t_H#_~ z^cZ){3={v&Ls&vGZ}%ZY%ZGCzk;}*^gWm?|^@USILy-Gb&FEKV2#wWME1hGX)^gda)mQt{1{jnpb_Jg{aOWUV z{*ON^ne9w{irMVfh1rk2BUdtrn|f282*qev9jTH1T0t0;#)xqQP_)I{)G+YXA1c5&Esf3jnK*S+F5+o-$fxh@qbr~2LC)!jB~(PqF* z5SmACuQ#E={ctGr;@vJnIW5B#2QA)zEwje(VLLk=|LSu-VRpdrt;bJ??|Kz)9!?vV z*LtBFnUCL$q9by|yNYzz`oliHmvMvqhhY3~pk@633C3~&G)Bq)OE4xuM*W{PI5hXk zhjJO*|IV6fCJ1=`8))$oNFXhC$KjU>HbqR@dXl=YSsHO_C6d4E4AXDX;lJ5)8gyEW zKlm7^b-5X~IdK=hD(XI0Z*tx5FG&_?w&)K^EPUz`b9~)zS2xPr(E81?A?jRBIz_C_ zW+HX>idQPQ&1lMlRC}DpRQP+JfOKTQtnsg*8As(B{E)sn$Pf^}A%N@t%B{22l6Y`a zwC6nbeK0P)jIg+On`Y{y!Fv$4N(DyQ7e$|e{+qLv4%h9^gVilj+Zgf4Us~@6yiPC* zih548gN9yXZr|N1M?QVJJ;&H1*O3h!dh@jD^M(7@fX$9n^M)oIYdl=3_XBA70htXE5jE}Pui^7`bG=_qcziTD`&;+07)*inl8bO41&N>=WW5=;fbPzZ-P4pG!L%IPC9`j{aZoQFY0Ipiv<5MnZTkU}Ey zKujfsi>)prT3Z}YMQS;4rJl}e6nyvw!~29y-iiw4fD&!XcO->HtMt%QpzGngqL)%I z01bB~kv-Ay=l3cA#pljC7v8wwVK4#7MG}CJL9)DDEBtXalQ0Eo0zs>JEm{7?3QQS4NcKxHxcG&0cJ%rT=7AKyyIHjk+u}SNQ~+!9<4ypFp)qsm_0h{@is@Vz2U5) z(gCF4>C7W4K5122CBL<}93%a0RMzCV6(Y3|v;v(hZ0>Gb=hG|+LEo0!CFrR(1g3X8 zCleA@{u&m2nBW;V5&Z{fG=V=ZCm9lWuBKS>+=R>q|4v?xvhR<3&bo|zoDV#H&`O9}osR8+7$WmDh-k&w(06##$zpe-3+$^0ZY|HYU$JH;VS&xkKv6!&1EA;c`fufJ= znj3|+!bkNt%d(P#YWsvN!O-u-+2I<4k&6+pW^?m`epGhLBjYc}$-pXw_mr5CoIIFuB0n=ETla-^ z>MzPM#?=1vT4PEfOwQpbUs4s85qY7_72iC6Wl^+?X| z(4%?hddtEY^7@3ZJ&yWxTuLrjYtQW=M}<=e8(HO7+B3IyTmpaRw$8(ulf3a4STwam zHfOuO*oP|ambF}eHXRrHa8|Oul)qbQ#9Wp1uw0Uys0i7^%wx}g5mA>{`f*U<$AB!W z3roeL#4SCvCRns?6w6Evl!I>TaNZmlvS!~?oY`+TZk~hteQI-jyt$y~=s7hwx zw7%za$Vp^6UDsN;`uYoXnePPV7L^a}KC%Ny8NZ)Cvjlgmoyv}1uFTe7y66hZ+-;oG z-{>K3&@*zMZf*|37mQJzhCuUB?wRr^{!ycCMgP2W-IEWuapCt=xrT1Np?I_TV>5uCUE@cXF@JgfYxY#2ZZ%seHXIwwa`lfo*1V_NvYSjkQ$vQS`C zm0RWyPRPb%!x6qp@}=3ypL$u1Zc@{84jfKv7h$EH{@Nwqu0R43x98Lb#>#nUr(KvH zbXDiJd{?-l8&zPZ5EE+EU=uD-MaXeb>fxdw&~k=V?0NKqkz*{cV`*WF4l?!ya6=-G zFjr1`vNjd6ij~pA5fWzB>3r~Y=Vb|I(07!2ucROe1l?dMvjBo|*1eDjtkPKxohbX_ z^i!Yu_n8?}(cS3x8*h)%k0$|45;ZX!dnlCB)hKkOx@-nzg#uJsT&hWMCj~jGI=l(a z3V_UnfL+AFoFf-;8Gh|gVX8f9J#6-gOr^3D>t@qAe*Z%Kc#UR;o}g%*)L{U>#SS(W zHWykx0vbJ6L5HRoFY-J>5Xw?fm0)-RoruD=VnL%^qnMO!r&3oK{>_xy={sa4{e5eT|LQRV!tUaD}f()zm zRAgsN4;+DMRohz`&@>}Oz*{mq!J(0;XYi`|s!97W1Rblm{7p;HDB48`Y;egvsYU^l z0altg&14)!#i(q14UAO)kzwDFDFd zph?pB5DhjFKpiGy9cNT&YGkWH?!A1up>z*uSB)5yDZuA5vX)!+X@Ubsv7v9+Cl?g* zrDF6i4=a>t84Y2KiW-CwoZh$mo{S}zmQI<4z}$4a*`NUiN4Kuv`ZQ}6YfJG5H11>Y z0kwxT&FaY9HH$LNk|JxgHK$2YO`p)cvK`5)i%l)UYp#87HAepfj?O^Di0|{af^Gq! zBm{%Obu}^=Uq_M6tdMSA8T9ny_FY2lt}&QJhH)VMaqP`r=vqD~e3=?s2m&dBjinfx z#3@G+FFjdS_6oXlK)Ey^5$+D)uf{H@!<{TBzByq5#}0Ar456dDT{DromB=;{pPEF;BTqyhAgjA()Fmyp1>%z#$>=DJJjm=h8@c+_Ni)GEriO!|^F5c|FG5H=(f5mgxl*X; zqO||pISkmVXeaV5%#6;;2eRhk6xH2j;ujl@liT(7rs61{Ki2%ntH@4PPg_zCL~~~p zI}#E*1*31n@X9OAxDug0WA@U~4%rIt)aT_w2@x-ZG*Wql_MH?F8Z>%)}IxfICz$?sJEaAUvyjAG!L^O6JnN+0^OYYDo+-ML*&2akhV z!SkE#48({?RBXwp1n$74AoN^VRxBtokh^GEjF5?g<$%!ony_3ksyru(tT&2kB`Rmk zZ&3(J;0KBeB;tydw$|0f>5WEj3$)aS{T>DTqAHb>npst9Az?A%CNb3Fg4=;sFMR{t z*kc)V4Ow)ftUIG66m?baqV2!ESjeO#tfk3}jm?gYL05_r^NWP(Lsz2$z8Xn09Z_Hz z#v>|Yfh*B4W)=ui4;}lNg^!xfCh|;Tk zanENlKDQu$B0EVnKkpnBZYzbThF+NNrBM8ojwOZP(75uKDlRUun0j&DbTZwm1-O4X z5t$=RV<#9M2uGbG&dX26osugOCo&kL7dc|F#s;I0Ff8e(WeD1C#3v2Kr&^7rTnnY* zp-5bmnVQR`X2mA=ahPG3rP}HF#@7m2VnZXb;V$KD&dLE3`LUDYs);LUw9B!i`|QDT z@s}oGlRaKTJJ6C-+>c}L{n5nDV=C4?dW1w`*f4Q2OhMW1J*kw-bPVU%w4+S5I>gizi~-G3e}vABJuy-*nF^?J zzmoC7l&4LcT(g}lG#?yy2MN66h=qgddb5a0B{qYg5#VHT=k%W5to9Ge0;9@at6_Ay z(KY#g1bx}~Cz+qg!zEFoDccnEA7HX7a>tALD-^P!Um-Ca;1E+V2W9rx00o8K?1uXn zzr7qgnSFC`njIl4%Px5JFKGz|e=4tO1GmoIu;0C0NKd z8oRlO^_qxHc0eNoV7Bee`ShDpQxHT?q6&gjH3zgG23C1RKC0r_x6)KXFabDkQK%w+ zV`XiLm0l!93Xi?P=XygqA{yC-0XlUZ?o1ZV`viiM5X_x4mmu*5Ye^Z6h1TvvtM zs;brD33EV6&C|+Dy_&bo=`y-CM`fU}ep(XY8jZ&q)K=awwK&L1xdu+nG!mYY^P7yj zwmRy;S3#`G5MC2^TCLd+N=gb(P_v13sSC}B3392;e=IF8tWL$L0NeZN%+#l?Ws|m5 zEKt^0_t*X5s4sl1KwQh5oF znXc;(shbF?$)|cVs#ZT@kf1ij=8{;sQdp19R)=5%yx6S();6MFMnLYHWRtXZ^_nFN zFc+Qc-l|DZn^g>PwXB|2-x^Y1^_8Y0#T8$J?A3~SU0<%{bn*))X{?i{*@AS{1=jVtFs!?@L%Ym(?J^d-sPnq4 z%vlTJ;46wE>(Fi|>R{(H^8ZwD(f%1TtN&BMwSodL|M!07|KW^t%x^=|QK3Hn5|-e% znAIe`dOJf|0=6?BS3IpWqr^f#ZA^DKj)lm?;P3|PTT7QqR%u=IyBOMA#7Ppx0yDHCX7khFb1(OvF1x2k> zv~oB!g!paa3F`Xq!N>aQ2U5m)Z>gsf91xoFXQYJwPGWnGDv;4 zo8!UL5dyFOy!9%>^?>8~@2Yw>l>kvxTQGMzHXdcG-R)n-2ghF`RtmEnQY_}e?G`(? z00}_$zbx`d@UY>#hd5_eb=>EdJ{~c~UuKbio-CylmJS_8Ot0Jj9y(X~Z~Klg%9U7r zR{Tn+)BiMCBt3W_0`eYRwbhoYQ$)#$p&?k*VuCO*QZP)vT-MgapIEkbz{i?2WEMso}B9=_}7+OavaxLHO&7#Nd&t1MS=TL2oU`{?<7<0smbn3ugI zbs;r^<@#ADM$80(SQ3B4uWFmNFwSg-f%(cQQOo` zB8hmAt2V{LgLVyoo};9*=6)Qti3aq`kAwX?g< zvq(`>=f-n~dnHtn6V|(DX3&(Z-nL5EOMFxDH5y0tk9y~IfO@TbBHI{W*dvi*jr`#R z3r6!T+20?{dY$vSiuxy~L@nSE^D2N^hW%wcKI38YJ2b`POF7JhjYh>S%)E@Ghou5C zMU>g!O;caC(3&7aWSxg*dh<9}%?T1_y0PrHdS9Y{y~6YyU;Fw$@)n6&ER?iq4IulKYiV6Xp{Ag{PN(=IMKM^N7#}nT0OM=yCcSl8=Q`K7EO83lBqKafuf-8OZ~j_H-lWqQPW=&7I^-2T zj#66_0Rwv!yS1#L8=8x4d2Qprfaa6?-TfH&;98kg7U=qA5Zym9pBS??tsjiVVKuR$ ztwwwO4R`Qe@YfYtcRzDypP%8k_Y2$+BLst23a82xh?l2lc_@R>2T51|4rjvb_KmDx z%|m)2{2~(P02-=0oY9^gZ`=S3=tokT)&Vq%9hB+h9h7V@+z*``R^eYaHLsNV2z&&b ze9a0E;JHH;i<+#fMA*l8LBnwd7pP!22r-XWKXa`I>Ms>aXf%j_i1L7Ku%QHmysXZx zJm-gXn^WUfd;^WW96=%M93_8!nS5QB7Og^E`rgV&+Ry)I=o)U~=l3ddKdKrcov1Cp znW-Sc2IsI(rO=jtdZQdwd7%Cb%BKs?M?4IUr91v9(~O>iuJxU;yGz;pT8VB$<-4>( zV~m3qV~?N-SfLnn+MIF(2}kn~$XKM%8o&djiF3gcKySlIC8rr9p>qy{#Lp0;dzI4> zZGgbbVK9vj%ouwIB|w3e?n$I$66nJqkH-XNFW0coY|8yir9u=|J&aY_48XsrV$)4f z{0W`mD$|&Hrx85pQ-8u}$5VAQip2}*C(guqh?DiWB)}pW!FjW?M%fHNLNN_cH1%BJ z`=^Qma^UQs#I-rBxO_2YS|XpMMSA8tBORxov&?tBljJlPi7lIZBuBykVlI>X=mT%J z@}k^iE%dIWU>#TwbGkTEu}w&;xmqCgv!Rpod^iyD22}uqA72M{APB=D+-2?{7@vp* zfD6F(*i9D;8mydcHJWy~HZtCE>=wHXh78UX6Y&)CFG!=*oT3AM{Decy+8miQe&m^0PBM5Z4lzEF;yVW%OJxHNRj-DEqe zdZ+}{eP%%u#VroJY2QKeHU{l4l&k1afrkmk zG+Il5ufBF=_uL5B!8+}=vCVV2f(hD0QZ+xd8=8|5Rg(1j!`-CNiRgjOtlpzst%c=9 zuc03Yj%iy$mS+*VI z36(%1jHITZspz8;Z;4k<0}NJ-KTLkRA7!T8uemXxC0`Yp_nlE^QLx|}wR{#4q|m{_ z1ed1~muBR{YwW_7QB9R4=g+EU@;w~N z1otB%;2$fRb(m9N0EXVF-FVSdOm$DpFJ6Q(J(!!IReYZ`jtUSqe9M?^j) zu|5R1LHNCsaQA<%(tQ&Vq7@Y5 z{J_wo+~&S7Q<{R#DNYJg3<>$xXFZ`yhV5YoXP&>4DP{xLEUX!V&c+k&9>#`^Pb(RvZX$AixBKG6Ujp>q=8mzz^ zuQ%@rifC%pVOKBYJaI=uU=cE=K78 zXgb$FV{-a?ZIG#pC}VwkCWp-n!e*5twRXKQ^I^hZ6onc&y$cx81q{XvCRPEF*23@^ zRk25KDu!_?b#N*fR4ZG7mBThwLY~%JvEvC@^}KDczfiCvg=Cxdol;p3abS1*J)D;s1Q?mc|!kMl~Jg$DTxnZ zfD_(`y!fb{-chKP0q);rB@2oK%naozXky-^&P=RCo=PNQ#Xl47Rqqhq4yWjZ>%Yyn zVTg)p=g>ioO2XYs+Qw8mIY=rZOKvwwNrbyt2ZZ(arVz|0yWb`clBICjr5@xaN`FXP z4NQp%NT|S$sO|-Qu1&HZ3%xx~>Ichi)q+->l4%pdlowO~^wPRvr{q`LFnC8%z$NVS zN!E|O&wD|Y6bWCmlN!b0G1x?zr0MM|90ikZgr+Ki;_+y!Am?L3$$00l%5vP!a_;gx znTDB{OaQ1!@|9A8cW?($WohIfX@av8A%$~#V{OJar$C@d)_2m=C;$+k2>|;Kn-?0TioQs!qYfL2 z!}?D)?gW}rh@|=VyP#<}5>H9uU+;og_4Uw!|CLkaL#Z>E`bz%!EhNT8gAG8Of&y&hrI```Q8COAUJJJFz9TZSlQ;C-&c)JctH1k(MUK4DCn#YUQ0W|1)Qrr3!o ziIreXO|Q^iR42uV_4jUS%hT01|0AiL6~9>(g`LSDwkb=7(h^!FuZSADxhwg&N*Uyx zbmF2j)n2WAu1H2(W-L!Nd@s>1jnkokoyb5*$;btr$O#U znXmjN*uf+k)$|~n`f!y^Ia%hWP&&lpK|7B-(X_lOzdoNc8I=>71SxeUqpRJ25dzZB zI_WN`&;Jyhm71}ItUZp9VM{z{l0OQulkM*1nsY_vKAZByU0Ct&#eHWrQbFlRUR z#S9o-3?ZH^TJ>8bl1aH~7m{P+$zFrrNm5xh+hpu;QOh_)hL;m>UE}~`v z-4c|H;@l;tVWA-Ax)3TFCeZ%#uqkUGUiWo{&r+&^9xv2LZgm<@jf`u`rP(SL)5KbH z=wg%x#pFbI8G6{?HkEC>(TZ#)pmqYL z>L-P?`!3m;vz&QfDpXpsp)@bl% zK1bn?o7s;CRHc9`j+g`11BQU9s90bUrK=s3%XA!P>B{%h0p823V6LPwC$70-puW{1 zvP`87Bub345KX{3$G}>vf+h`WS(&ab1EK>zecTl@7pftWeSZ?ViVE^vfz)6GSz8zBNOh;dL6;l3(+QYX4#5HKRVOnP0Ot41V$us!DvQH>B*U zRe4I{b_G)5FHExO2f1h}NFubo%Slj8h>d24fl*E`8eV-mRZ!AJXwr-|zGYG|URFcs zwZ3RX4RW>kKC_gB$Z13!aW%c;T5diyWN^U15yKEz8NE7bm5o|gBH;#4jGjx9TJu0@ zl9lSrx}|ul*pwFA8MxY5-`Zxw!*kn7u;)9g#iZR`9Mu{#4^(9dlig7rsT&fI)??*l zt22&R1~Xfa_1>tAiW(SIgY|LKy-Y&b9^*Ntc}D1piv8{iEA6D?*pQyE)!Od|al(H{ z6CJnXJiHk~vb-ff%Uifuhm@Fumr2n($-P@b?K2U(T0bl9?VU6WXScp9DKdsO3R!H@_os>j)DfKyJ#WrXr{$8?q_pwLhiK1_ zb?>EwlB|q9UZPy^p4(Uje-hzlYHkB$vgcf^wP@^oTYifvw{_g5Q>^S!-08qAG})(` zK_CAs!Lcz>7j8phNT>cXP~S2YcJ?7W!|Pbh2n6DnG#kiT!I2lG<37?!~i$v z5vN*4=lc!u&`r@KqS^h$AM3}woR))iA_7$87c$d2OQ>yL(fJ0ZZk4R3A1NI`P>*Hf z7Bdy+>X?+wStt_$N&?yVcoN!X#QaI@;j}^@`OV$?$lt8u$O8vf`z|_`V+8{C-=(U2 zp#&Yrzvx#+0UKXZ<`CvGmZE4LXHfKy!)4fe^z6V>DHW@qFk?l!FHcp`Y-rvu_z5I# z>bgy`ETWziY1fCoviM`FLD@Db`ne>9{Aqt16M2@FQ{)p+I%|>F$@BDB|ke&jc#Sl_3E^4t$w(vT5c)nY|=%BRls^Mmgs-nYB#70orqFmVOe+Cz~;lM26I<^X()$DSghsL>+wrK2c$YQqR-Do0vb`kK-0N4)XnK zdHQhWWf;3F`3A1NNaSdx5<_$2v>_@=fUU40(fRdfXqVV0ri8462VY5po>7e5(I+y~ z2UW+ebV*xXDP;IaG&T@KLuOACr6;IIVcSW~y~{3y<-lqa$+W9o?&tJV8H6+j>58$V zIB1D-fCZ01wt-R`ckDOa(SNL?D7(Q4P^-CYiHtIRr7CM!psXyLA9s8tU%)Gjj39m! zH4tOeCq^RN`&Ytt92=7iX2-5_K}yq?#OF#x-fcvJiiCE1@tmj$Hka|*DiJDU%pJ^O zN;M%>3dFa9tdT+g3<9b}!XVvAP`o(9i&~2WJ;H7zKrj6m_x1w*ERcwRGo??l$Wjn_ z*+KuJ@%+8kIuf(JvI7;c({mZpoO;L6nCRca^mC4r+U}Dc4T*_g5LE=27OQ~zYLl70 zlioWK#W;atNQpLBO!*jr{JJT5ht_X1jZ}`U=fy#80-!1g&RaRmz#IuXd!vfsRMef+ zf(|2-ag${ZK&4me7ZbxD6c)~LsmB!H&p{x%0Hb~mb%7Dul`>kx5uKb$N^NGYQgMU@ zvG8}sq#wG8XJv@aUc^Zmf&ff^3{DX@$QVUa<9z`t28d0tS`#u_v!iC}h*@*&Wu(gzmYvG6;lRw`3?^F0Hq`?={YBe3)fyz$QtMVYL?`1=iZtoa zyJa)%w)V4&_fs>La=L~6?U7PH%%Zjq8meRvSk1X*PBQ5Ohx>YZ5{t$9XufgeS8P|a zUe%Jl_Q5s?xj+@79(R^gX7akXbU(=@N0CCJ=V(a?#W<3FInlw zy*Aqu6D4xZoHfIhas(H-GZS-M$n!qkybb^=22Up?27}sn^LVlIy+uGV&iSjq3*pmw zRnCg}zTkBw2yVW;4{-qR! zvPfs}*DO0ay%2`A%;CiV&|~0GcDRmE;ET_FSSWJx*Y=~XS0 zN3iB;mH}xs1Z-%hrsUY4Hw6W`K4!%R5qI5_aA5=(o2g6JfxQc>Ly~ed{XtX*2C@$| z5vN)IIS2v%`3n9=upAl%A^4XqF%nC_j00f!_aG!WAU3xRO-F^2k@V?SOlM&r@eMc; ztZG@$HZ#A)^{ZipV!oMS)5NxeQL-4A_=)pdkP6fc$=}B5bziG>-m0apb=G`Kk?+9y z5!ki_X+a&q0Z4V0wuB5m-qbI*yyoq8+I{c-uKTT7_mkWCz)TO5bhdpI(T1tZ+2RKS z@E`rPJWu}k0DPardhq%1B_<7Fe z>pGsxTTOqgek@Xad+nhqakVs=O>RT>x$BK@S<`anP5-sqT7TfV&2}33yVw5!Q4L9V zEE)X4EFYttTCJmEl%wEjp8J2X^_E?6b?dfhQMecG!3uYR6M{owL4&)yJHg!xcXxM( z;4T4zTW|{yAh_Q3uD$m?YoB(0z-(=P7_E_y_y*6I~qL5GddMu!P@bFD8>PaNI^)}1G(XFb`d0g z_lrlkwnXYf8zCE38c;7(bfk#DJ~#$&!zJWH-1c|gHXdW%ZKbuR9vbHq>2<*o=#gSS3NOjC`O}PE>BE7 zi7<=b2fQOkaJ|3a*g$=xU>6eLa$3M4y%Xd6Nhp$$m0}#J#K~S|DC`)}@fAjR)g0Nq zt#q@k8>+82b~a{)qpZmrCJw1Sc>n-yIGm_Eibx!^|3$_|Fy~Z2w}7{wVJ)P{7onhN zI!tF)G11Q^D}4Gh>K~p~AdGxByta?EFM{IJxg8e@vwilv+j6EEhmj{pFoF;XZg3w! z`8#jz?5R@ZI|_+y91gZ1w1F5^%CU|0`=P$LcIJbkU?SSI85a{ioPqpZ!HHe}_92>@ zj%#5TV+eu}<7_ASqqsBIU5HR~V1&e8ycat08b`fGU1iGxwwCT!=VUM3+dg4n0Vi~x z{@v{~L+PV-LxrQor}|2M>})N_TST`L7la}ajw%5xmgChOHU^%~-ZOLA+5IlnwyQT; z`tt`RMsiK)c)yMbNB1+ht1%M^m?S!U z^)Hhwd|B{;NCFyGX&6N68H3sTQ&~+<{MTgqM2BJ=cET%=@;_#@MRS72afG{ou^d1Nq zqQoAGSj-xGO{AxJ<(uP`z%|Aydl+QArU3x^3PqaJC6h!#{7{htfQ@64tW>5H1}{{E z$c+-)mPWA}{RO1(?4ttnS8=R!J0~#%qnM8k@c}o~GHVnGbhHN)vW|*AG4mhd*{@w= zxnyc+`Nv1N;}X-SOlxL6aST@WK2k}<0M9u@Lmx~6_(jbaNiOlltl^Ny)DJ%3?B%Dh%+Sr)u$abLQTo z>C1CC?kpz^!Lfb}_ZSn>8JQEqsZ}*3ED$vljkR1L13>87RY6e>S7l-w1+!3~NCT7ruk?em-gm#Me0e?*3n5l4 zR+(>0S}1Kr@%a+Bb5Ql$Bvy#{dF611(GU}zygq`2;0nn!lvmZtQZh>>F#Xo5%x%sA zGLHd-IcWer&+##vEkD55<||{3BbfY+3U=hpg>@>wJTq0<^|}q%y+#bgfD18&3@C~N zhv?N!w5CTadmY7UC9UDtUn)hCR(Z%I~_0!}WxT{^pwjdB=hI);F=+QtEAe{(IE|eOndySDC`%!Z$0Qc_H_w^ojLz1J- z`L$7vKhwXB;Re)0LG=3LfPo^T^ z@Sy|bV>>9~2OQvgD?nzYvBYJrw1up+mU7$Oh2!>#)F8*Y2eh8k* zVgwc98L!Jd-%!I=IApsZZxDI#dhjke?OM_rson*XjX~fzi)S%X_?AQn%sS)ut)v1cB^h7U06Vit`k61|=mbuwh| z+hIc?WRWX6fu@sjO^fbx!#F@Osh04o$$`tc0S0*rxQ4wUaO~nw`fdiVOo-DOJ4oe!Y;Nr4~3zvBkS9dv$BUBwr!ilkSZmNT_y$k=cmr3A$y1hVG?UwsTh2r zk;NTp3tC`7I ztBV(o^^@*yYX#LhS#5&p}O^)=T~+ z*?2e`1fc8G1##Mn0gCHCGc@|Yo-x7A%@IJ}Zw3uO67k*?5&Bn;&z!zNy8bWo@xm`h z_u?JtC%L&`&A9R;@>&2fWuRjkDPAa~6LnjR7(+ZT1+(*TMh7@2vp zj;R<+${_S>pcMqFH?ZdLx)L4;)`_z2VYx~k2Lp}7fhM@dR}M^)q3Nz}{mn8!B|>cF z;IB=es?|b2wFe7DxDiu^00YACL_=s-!U)3E9MeOC!-e?VsCAS{cT5$C+QKCzg%vav zbn&scR|IK1l>LuG2nxd3`oqG@L-pQ~h}Cg|l-wh-LxSL8LP#k&MJ##u-HkN)Wb!OS z{o#Y(V~?Zs z3HTB{g9jrhv&uNeBPc%v@~KmQia8YZE56;#I>^xTX#UTLCtA!XCJP!FhO%FnWgxi>K z|6E~eTZl*@WR>E;#m9GyBMh#-itkL1a=pR)MW@Hc64*j0J*;Qdw5r{X8RHS13#Iuz*&vG_FaAk*rg5 z9H}1)BCM3^A7>@zzHX(Vf#M1n74=~hnRgl0|{TB(wjG9N35mD&=_(me$uGMS0O?Fqwc zA_X!zKz7YU8Hdts#py|rDgE^9bZYjM$k-`}WJi0v{7&h4$h<9m+}Q2pic(lVy>eC< zxGRTocy6Uuy|9_^bMM~ia`&zL=LmAKc|m3({-)SUyyXHr7QR47Uqt1=KUSb|FVOG% zoSVIzK9212LR_FXE}S>S$U}N3JwK~BpSeLsWJSJeG_U3%b0bs%I|^HZKgBHFEzPUI z&^75pXPj?=Qm9tp)^TjPJGKa6p(q5jBOP6hM8oXOeEbmpy@TAFt4Q86Q(%oYhSDii z%U&pok$6nDnLiaG5F{SO$f{hV;8`Sx!hnz-10XK(K#t1sNLTj`gEuPHV8}?c0)6c) zQMHP|h{O$%D2*5|vR)$)&cwA{EB&TiY(-B zI4BE7Q8$PxGt$p2uh=P;L@J3BNVes#*y&H@2`x{Zh>lsS_|{(WtzR~5LQgX>FW(!R z!n;yPv{2JJRD+ZSbZy99#_Nfy8;E zRJL5RL0!++D(^Vb3_Y-6L_h@s96i=96!?eZ6RNg-uXk%+;&&~~T4DFfU{u7^au3mN zlxOd)t_?Pq%`TfaDNbF!#-&K3rzlI3oAZk{O6m7D885ek?POKy207Q#g-WBW$@jBw zzaNSQRW%Spj;s5cQ5I>pJz6$BaC8SGcjATbz|OhWGKOZP>ge1A+P(zNq1h6d%7j<} zxG=$J?BgqYn4tF+)m-a{zA$z1c637Oz}>I=dC|(~DM^mWk_o$x%1D6auRs{lSvy}Z zkAlQJyM-NJd33sI;WsF~u76#iB{XK!Z9FxQyJTrk_`E*s1G{JtZ2;Tr%RI;ze?W#@ z!G?cfCketatR`4N(?mhVppuFf*~V~}C$8swvXVx)$mEYD+)gqxQEI=&R+)Y<;u+et8kk$r!z^&Vg&H1{V=gem#4M9A)mvQq*1-Xx!X3 z=R$08S^SP;=tAt~bp6c!bIO5otF-nrv1<^=nnTa1+iuCwi6ZGtIWcK;gtG> zJ1%X*1lCCh{Vpstwm5zw`QP917A>76uJOSna*-Y0^wRI`$$$N=ZeOSdaZ=2Z`bCzD z?PUpTikn5bsP%3?t^vS!UzP0Vy#JAyni*M z(+M9npfMA?OgvsL36Jd5`T;f~03P8Tf2JAib_We$HO55lLfNUX_x0dmkD}jW!GR_^ zhKRyreD6IVFIyem;_eVgKR+3WffX(jV{vrg5259^;D&G(b&J3I}e<#BbpBB`mt27QcaxU8XSWZCE#;n5E>!_EFD}Q zvS?}{G?*Fufd&X*V3z>G1{Q;_%*NH`cSEo_Lxa2>F#N}4Y(22`TfV6FrOXN!y zl`1BR_*ktVK4#cDRusH**QxDO2cnuMTzXH}X?77x4 z+L{KY!U~gSX<3PW2hx+=EHklgbd{f6xY;b89|l6v%`d7u?!;(dPbjx?+S|Sk$23$f z65ey_PJN4}Zmn@V)0&0Ue7qp!_SJMGJD-E#eP1`jOLAP{gfjMXviHuf z@IBZg^o^@N^&zIC%}F9IyT7~W``JQk0ANfYLY$bk`ERo3e4Uc3HiK^vVtC%X z08(Ij5>EsmI3@y-s{dawc2b<>K-{QFuG+v&r{(Ugz(=5c;6<+#b887Z6)%FofD(rw z(?IQBp;rbr-tfA`!Ons&TMXNMyveg8e1bKGg5M(0KzlF6vp!e-Z7b2S!w{$?@GCu~ zv2#jWc3hC%;6b1+5Ev1JE@VmI;IM>bA@qI0r9eud2FhdazleW+(P*%9-l&`A0~rl0$x?~6h3{4(=|Jl*!Ih1D^ywKa zMI8+Ep_dnK;xiEjV}ZQFM|U#^7a0t1>TLeg z^y9ca!UyV(T5B{kY@Z-E8Y{W5&di#P$ALt}TY;7B(k%6FCE>|-Cfmg|jj-~Z6OGh| zqNd;59S2oDZ{%k`SqakjHu~zwDe~C-jTcgTZ$by1TrXIJ@O<~V&bOcTsSk$huTGzm zk8ee%@&*RAI`ILPP40HVz!m~8T7u(ll9KT;=ofyQUt?Mp?ua`ThhCKy{%Dy z9dtJ6Pvl*ngA+Pq(X_7847-kbw}{R66DugDp`fp)4n(p42O8xKZ`aIhiVQt;eXtKIv40p^;$3XGw99F)XYP^3H2}`Q{?PjPw}vK*pFI z@eL;sF>!GqL>X1ioj$<5_#mA5ssQahd<+BtivW;C`a}RqAX=26NMHaJRvBcN(H$(c z%M_6SXPm_PeLq?stq2d+h4d;NgZ@1MKoCDX!qa&b>*%^eIJsNp`&E)c=S+t5{z1xw z%!V3rA`g4C!WBGpL2dT}PlD&gEcG^+05#o{MzTu&nAxBM$1o`OFeWqJ(ExsB0OR#o zj5{EQ_yLvnXayf3>ZzvCWQi;?^dOSJjo{e3fIXJyF)87RD@re;aQ5+j3}bC4wN~#f zu{*+XDOT(_P9sDLkrAG@264aHk0jx>#WL#Gb7_g-F*e0;+#D}7z0au8oE`8JgME^p zhxy4w)lR0p4~HnccjqH+hTe0poIt+4Mi3T-ff2p8!F)RjOSW}L>l_l3V*K^$Pb6b8 zK!LLqZ4;KYWg>)6ps~n-%Lsc2JI(C;Oj$%ML|6_lw%d6+w>*yg8(1XueFFBX49}U3uwNfBhn**lb6mchxs$dQ+c*ZmnpO6NG?k7-q z6*pB=PqQ&>hkrp`Lq>3Gv@tl_U+&wAH1`vPnlE__`F*GpiBQ0?jEv(AwMNtUWLnHq zm{E~Ah}9KWHbZmd6OuDNh*d^Nu+9+tp!jS!6|!0@=K5N6jcWK~TX??QXGjS*X! zHYffTx(a0Hu}I01%>+oTSuiFI#Gm1Qp1;X=yrCAQ?=jC}0MX_iM3Ha9X~r?O9eQSVzceg2;_({-y)+{Fwv?x}Tmw6HNok&%KxsKAmfiO;xaviiM!xRl$+n)2I`w6yOQ&5stq z7bK6#GMv2i76Yx&o)Y^xG0zqMc(2mOJ+V;hM4>{vcvP>@x4S&yg3Ax-%p}k6y9e%8 zb9lu2c%KSH({^!?#=cFb`EH5%HRA_jYf9FQc;X;g((PIS`%z%~14&i_x5N;N*@LLc zv5MRS=`AHSlbvA;#91DKfc^ez#TLHm!6m>z2BgnxrM6Ki!JjQ9>5M?4=@#SaAv_Vm z!bpUGQMXx~|J|tw0E7b~{#VWoEDWidC6sa?9F9yXgkFSdI1-4b-96S&BpvcbAI8OZ zO`Q=B_l~YMew{Cy%x*G*CgWnL5clqUh&$Uz_pg~vXM$9A?!#R8Ee5S5y;;6cj#80q zPUqk9uW>!8`ZOyJhRcb?o?@>hNw!~_-e=<$J8sL>gB;{qCCgP%h$=n*Hs(avFN^dd zz%2h=uOc^!4@Y(9Zr3#23-YFxJ7Bu}pxnX3Aa2WeSz$V!Ui108rgOtm8wf}EZv98S zNgb}R;#VDY!^D~bcteEh=k4hXpu7j!hwgUOAdmM)RaQZ^es^^*VJ?;cQg)H$rst}o zXm9KG{-(S4D&Frco;5@0JZaUd*XX_YDF^5E9Dy;?c@XJsmu1!z9 zfhbbL2Po18cE_KQF31Rg{9Kn_Z}BS0vj5DLAE^Ijm>>ps5r;f9f}(I-_#7%jnY?+v zY~rYyNkBen)|I#kSP2So|xSsma50%eXOO&UDP981#-2pEZAw?%#3nnXR`JTWD?c&VPfnLF!y_7(etUrp;GBj^v|H*# z`(*i9DMVuA3b?lQiw|0sG5D`q>3r7bhDLTCT?8X2uw-x}C9q{!U^!TzKQ-YA*ow%z znwO&PYiR{$YS731x<9Oj|m@@ykbs`1z&34Qa_ z`Mw{ftx}EHy5MONzT93F++){e2yE$2+-hjctO2~uC|v#6p8iZGvEg2|dR-ute*c$V zs<&G%Rng&9b|vfE>m9ITP)KykM-kL$tnUXp{8i@v6#(1dj|V&*#r%nXz3d=-D+*KM z|0aeE8h!n##iJ3(&FWCSm5_}C zW#NQf`f~Dr1m8`{^{fOBpvGO2vwbT-qCx6UbF2xyFPcC&A08}+!iVF_XUV?PyJ1t5 zDn$TyVz7P_3=$togn379V^`b@k(tK^`(*55VIJjj7RRvli|-7DJYdsUbNDK$DPJcN z;{VNY^3h1jCf$Q%@h)LTCR-Pg9qm$XhF-dwZxoqFrH=@FRRYL<+5yJGk22k{$64cu z0O4hbz@23g^F{kgBM}l3*4N&lwll<76{Cvk*wKPFl#E0SP?^kY##CG+S_6IpPTGXj z%s^RW;lBgoj2w>bT&q1Q$TI2(O)SXsldKg%*(Owf9Y7Cu0_@Lb;g#ZpRtSA37sldIKP%Ok^orCSgfQBJ;#t+yppNy?|a)NOnGip|qG@ zp+Z)*U8RaS&XpzBN?4TfXC|W~*Pv6jf`_$q&h<^12zsSb_=HL!@ohOop;9GnVxe^2 zIrU3yrCOQnVx{(NrKTQ(Les=zt@mw}A$pZ&P1|Ba=54j5LY4ON#8PYLZH;5>Z=zTt z(Er&V!T?|a)d0Ky1>L~2SMmRJl1hg3u>L!uOgaoz_~{w)Phb&H9RF;&eoHis8TE;n z5T|51iw`~ySMz6KPqs)f3XNvnl1SQ{F3NF>YQTI(FGk}OUvuTx*q-An(iOF`fLf-y zC^9A|FomJT4_fP^Mh(SA$GtD6|2B|%!obK5wKzzm8XlnVXx%Ugc1PgA%Bg9JOl4q= zqvwA6NiGPE!l7RGQ3IH!ApLH@0Yo%fC0YC}l>ByQk6R}9Eb13=)_fkvf%j_~4M}<1 z2QzTvgI|N=?!PUi&BPAaWwDVia4fzw?52H*tA7#q^LiFMvAo{NAU&O4v9G5zpGk(h zc9*4oOwv_+ZD{s?Yoxn7CwO>y$KM>@5?smyF@2#WP_K?E%$R4|I?^tcEr#duPjtj<`W ztY&9+BGX&0rmtpd)=?pvOS`Y6dceO=nUE33NS0K)_1-)yCOpj;AWOZ{wC}vtEwm~C-KT`7-6V*a{7LMD#szZtuJ3E0>EJsndV6fXG!@I zs>?oxa?SKtw8?tva^$@9Nd5aQdyd0!J-Wip_@Y!Z&MaaPiR*w$mV8SemhURpcT+_M zt{fEtwdRJlFO%*wXty^_KkR5(T-V@ya~j3PYTuciW^tQv$+s9*H!i0zBVU}BqJmc) zD(rs9VI^B3knkzQCotPq@&`vfOf6S3bk42{u)1_xmtC-Ha@^h=sKEAo;NH2kUuf?_ z&icT*Uy-y172^1%@chTgs{HZt!}QYE#8Tp+PuJx9XsYoS)Ti^hFf^Tqmd#3ke5#HeRxkzYudvS?>d->BV-0-$X4tS>PiMErWe#|0Y z-1eaL-?7j}p;QNw)Aw5`I0g-_&u8!uV!@x**a-ZM=%$6RU~AXwUOq>n2p;z*_5Yy4 zBQ->0fDEDr#X2H3OETy-43b%c^kHm_7T=M0lXN!rP!ZYtU{Wmsu$n*Eim{DHkUWYF zy7zG9b&avf^Wk`3tgKa%b=L}whu8>r0E(Ug*e%BaROv8$T2yJsI-cFi+y@`(*n^1E z9sn#K@{uq!gBYIn)BGuZL$3!iC>fRDRWHlZ4iW1kf3M7cE%uf{9-a0gCGW3bxZRd_KoM`pEX~CanvlI~OppC)nw7R11PAwj=5;s#9H0Ua`JbR4 z^!CTql%k?Icxe8sa=j-hp<}=e>ZGV3C1*4LG>2AN;%Qsk4 ze|%`^-?@}s_ZL5VUEa~yKAV_bAE29GcIe)-iP+N}aMTax+ zI(q$!8h|w8@yi{a>X;%R{p2Xu2UDGJJBSoLLfahQ-EuX2i|ipk0)uCSG8o5QvmgYw z){-)U*2*)}7qeB<#0Seegwl!@NM9HL3_$b03%*F#3udI^GxCB(W1;|Rl1w(FABQI6+f#Zj(` z)B>#&F9Wzp%xZBl=NoY2dYnLmYweiLN=DOCG>uOwpd{THE64BerpqJ^CMm^fIU=s( zaOTM-9AibXeq}|-TgDmGvvPFYz*CQ21f4!6Js81i7);#R0Fi`L-lfJuVn=wQ$Te9zoF0t*b65mEiea|C zQXc>;I^l$#NgJPX6$}Ez5VLcHvroc&SzEds@Ta@b0UPvF@Dx;swExGWj&P@sq0aMZ zf~b=dCdcmp1(yQ`UvnoEPn}&*WCU28(%1ppGcN%nd8vCo?u7k>GA0rJVm|1CAOM|r zq^dj=`LgWwG0!tqkb6ackiEU3Y{;qCF` z^cM2R@K>V`=39)Ib=mr$=ChS#x2Mf^&vyU`DN|qdZbLG!G0#f!$9{dJ*V{~{R{M>g zrB~W^&kyC}=^4l)_AE-{BHn34OkZr>c=u{ltUO3x*QhK5M?e1f!q@6z5vb50ApKjZ zdGmMq$$X93VzJZ?el+TPcFoK7VS0YrybW;~=Iq*Em_5l5nd_L4o7zTM5GFZ#wFM9_#;BfFM(l^#^+sk*rB|KNcT zk5eQD0DwE-@P-f%Q{ZMW&cAnU=)N+5FO?!W3T^6u0vU86s^bJFRkp3O;a-cg3@t)H{bXj7=uJGXbHA%~u#0Ku(or>cu4d0Vt+Nwa*PQ{9V)D<*tw33h z*frMnI-%XRIT)ypm+=NRr{rQLvFg!^GXL-mw*T-A@1;69p$5Fx9=di6FD^2RR_`zD z3jS0o^c0Q+vT>Qt&bGIs<%}h_TeTBW%OV%f3mbgO6p2Wj5DMVVs!2HQ|;-FZDXJl{4x!g-ae)Tx$CBZo}((14GLq+_%_wRix#{I{o=ZdR0 zBxSu{yKP-45BmfSb1QWeSqvZEK?5JGY1lbIxB;;6iQQST2rBS8Ti2&56p&RazBwE9 zySO79N^~Xbdy>RKz_Aca5PZnT$6--ZdJr*H37_0!XtwK^FiE0h2(VrC<{fSugL(Q3 zo%o9u$R4J&DFVzVi3`DvOiu=XVMN-7=@V&=NQTpF;Ffl$-6|BKjHo?Fcq;(a*wDJ05;jCC7_4zw=CPr}LU|7{wFy(5 zJLn;sG98Fh>410BWJkqzFzXH{&Dnh)RDU%U{sJ`R+e4lx zff{`}vew;m9+cws#$lcT?|(v;RSZ0C|jLa=ecBQCdh^XFTH9 z8k?^QP-D9cq2Z#LkPTk>Xwl@Hr+M?J(g;|fqq8%M3zL!Eh>eJ~qlggnnMjD3;3~(v z&FUwVep0WG`daBxwXP@FnOx17sscPdWrX>? zB0O-EJ*;Mlx;&mZd1psqk8a9!z^LK8q-sGMU8Czcq4D!QPC}U2<%}X+cB-+o=`-YO zr5*Wz=He-jJfaRRXOc}@m9~33#tx$l;@awHVoSaGE4p^UWWvJCO+_qxakf37Hng!x$w~rr;@Z~5ywg^iC2iD7$Zkn{ zqgGQ?j|;r}Iu#rhRtf}Oi}fFB%S>J+rGGvxwa>KFdQAV;ubNu!!)UE%v-xdA^1L$k zrL{3_`pjJLd39F*ciH{*sin^I+R998Yuog>y^l|X?L%w(sKSMF*7L^km$uH5>Pfe* z=grHww(cW^Ki-1Hn|Cv9-|wgY_&p0Q4PV?hzbjl?h_wzCeA#)!=MFuv z{39@Pwfhjp*Dc*{5d7amFWmnK3#$KP41iKVgOL88nOS4XeZ6u2w0A}}Y>6a+|26b> z%zYTuXEvFTKXLe{O^RkfcZJR9N47|?*b75`>1+{1`fi;8VBVYes)Q1d`Onb%rbL{h z4Lx40gZN0ru2pTVRH(pv!%w=Gu6k<~jgUcH?bf?wb69fIoN1(u?xZ&a0Bho8_7m|XrABhjgdpN2u z_ruyu*5uur)2{X2k%e^~6RFX34b|1C;>p6jOI=U~qO!v?GL2#2m9k{M*c0W-if@Wj z%VG#T_X^YxU;B7lENx zreIyh%y&g8u8gdq<)W6%I`Oi0ijb6MI>=jPKBv$0>%WhvcFuTBC(%=7qW5!zLeUv`7t0K zt50oBf&`!MP|-O1@@=p~?OL6&8%fDg(;CqQb&bAbYv^+J!Ymigwybi7U-qY`4l~B( z#i_Hhk7qM;P?=NhUzlHE@akKrV}M!prmboEdf$Oqv9xoL7YE#k0eUm#;)Gi@57Dl( z$=Z*+Osqn;@@Vqx^WSb1X9P`jlSTqF1e!+|A62=kF{+rIgw6r)J3oZq?{rSY+*_x( zry?eqo!h_5phLj#&|t`V3=vD;k5YW*8%LilJECdlLQ|FPK0pG`(Cb^)kgd&lWyjJ6 zO?B4U$~E{}(Hu63hd&b{F2EVl&H~u@Gh_GZ$RpUVZsN`I?;qhe$hF6Igyqk_BI=FV zd;?7`f=y-2yQ6W!i9ZicSCI)~l z{VT4uv6!StFifM;)w*+*E5yn9XMuEGc-YDjovKvu@Pmrn*i^qxGoGy><1)Kfi7k|{ zVp5a=%1|*t<_nmJhZ3q3l$e~4+ee2wAoB%nC$1 zPNaLZLy6TU8B4{_#PXPEiW_Z25@PWFF?jOVth9h0vQbg)<@Pg<3O5lmJ#y>uhD`IKB=I=g<5*8GVtQDv9gn zmRdWO9LohJaGthm2NFJ&{WJrlYq4)o|2SXgm4T1!H^^lzD$mSS%d~J_bAoe>{Dx4@ za*V(t0;t=XXE(>+R{j7l2V8@SX*Z{?Es&~-MNf`o)#}nxl!~Us|mP@{W8_$;cYcQ3`hwD`G4J%9OEbzP7HbqTOcDsfNtcmsvYrs43x|cX%F4`rZxX7huAJj*Z9g!2QK!bFeA~ zgR2YxB4|*hvT*9%ptVq~@jD7u{$Q*-w7Bv!ZeH*|;f8g6Z>J!|8b`jQz~%;Nsak^J zWyb2+JJtAV_?=2C&q}X}YRO*Nb(Z?AH@f`T+U?Yz4oD~=r>)qxL`>Cn-$|<4QRcjH z%kF`QSV=fnK3jFaWPh_X|BCZbYWQs@z3(IJ&Q)sncbM+7dG-#Qr9142X}0BYzBQrQ zFSm`UAFd6NwH<(xtlYu)9$cNdfBt6Gi5F2o_bIO%}<) zo3$v8c*tQ8%blQMJXM>yW|0SVsCWeEIQ7>N!|IGYxL$UAYy4NKUTZ&77OJM%$!8?j z0yPucwkiZ0z1f+Wg!eEO1>x{%`v3w0vU7jzDrS*+*VIbT;73l9{2?9P^%0CHtByog#zuG9*9w&iX& zPtl!Nd~>m_7yPbN$a)hWyHtOji1M8pNG~L%0Yj1H!C3h~0!@$#k|8d9prJSqB><5) zBq$psf((FLVQI5{B+HLkqP1&zmkrADiAZMcMN4wGm18abm3c9SkZMO#RM!e z1zfRf3g7J1y?-;Jgj_iU9#_Q&7w%(5YfX?dyW-W?9}Z1uRcp{)8)EP$HqRkrMOfod zg+dZ4S!OV)7@3i9e9RgQTCgnCgQ>f1DTPUM>-}Dz3P1FZNuyO%#k?j#t4c>$7*(dA z8dd33J~Rw0x%D54G$p~*L`<5piI_n#kcu*ST8B(fEEZG*C)NLD z4#O_lIe8h5mTIh@drb%_fnacL2+d0p#KK>DDeD`K>BB zvx*)NnO+*u{W3YRaZi(Yt>9)-=dYo+nibSXw3R_O#_2_*qx1pwNe0F^Sm;R>LfjMM z(q1Yee;zFaSB>KB(w);`^-4t4qh>>SZtNRf%aLiMqO%ZON#18eROFD6zz_VyB|fs{7}G<_5{@4n3QIimYw2Qpt2S*dyjd8nP@yWW5(_?H=Y@1E+V&F#dBfmRiM>89iC5o-Zj z)>`0x-{$Od?EOL`j7Ae0*`l^zr^&>!R7;?JF>R!v(-Wl$RZj2hHmQGiyTp)A|k-%`^DZ;TPg1mm28jaorm#ZB(;{pL1MoF?L*-u25!!A{7sH5=99h8o=#MQ@?>|&Umvk`YGk?${{<5J!N(=aaR<{1<)@YwbJ zyr%Ahwn%bui5yBlXpYLR3qv>MW|3>SM^hW&DK7?!tR+7(fU#|*o~BKqr)DvYB|WvZG39Q z_S8~Rs4y%GK?W>kBx;G+x^bs*^c_bR;|}( z`FBEr&LAXE`p*n_U-8Jb(DwhHegD)56jOX;GDc4wT>_0eNXyFl7=qKGXli_ z($^nA>sj!CPc+Apd+GCwHy`Qj{S}M25AH#6|L-41KSZdXx`1ke$$e48GHLk|eVy$a}~gSo#%SzrDp-3NG;4#I#5 zc)b7Cdl`sm2zSmKR3aI~V?@G%w+cN|Y7D8}nad&rjcXxMp zO>if;7I!FCoR;Da1&TvkC@se7f{`uLOr@hr ztDUNJI!uM5B5N&NS87z?ICzr*TbBoOasNaXcN^QCr}Jg#U+l#e|TIQrx2uhk}N!aC}?o@#LvZ~7ELk-rs^C;$)g_d7jp z-#)@FL0DiimA$bfB6gw1;EwzW1ZlyWv>rakS>SKInvmXifKL_DsYB|%HYL*yMDmvZ zE1uIsr?@+svuX9M=c$Yvqh$3Ro{1a4mo0b_wcW4Dfz--tSp5> z39Nw3z!#JKkQUwnv(O{fle&o9QPak7{n^!($Yeis^D<*643+{>h*?trs_wYHd41sQ zY7r6-8CNN-K3ep(#)2!fgqz;UY$X9r^aqm?>zJoMkpQa-#`?AWWpP0Yh>Q52Nd zjZ7kMvK%>)}7Uzzj-PXwu3KxQ)4jS$f&5fHRcRhv3V(%YMmPQ-|;1$$69i2 zdc0dMW0rYOhicU6f4{Q*GvWDAjJ0iI;=N@D=k}b>fOYGx*EnYUslWA_8OPV^PTw8h zoc-@9Co9ug95*vDhh44j!17OA8dA0kH}j-86;A1|f?D=#ymLg_Rsu@iT91ZsalkrR zdXc~Xz9V!eHJv*!ir~y@x20ez4VDw*gnt? z-}ueOdh**)2bX$Io$U6FNgCg-E&f5ti5aW#Pw06nkUwa7g{*IQ?VU^=Rf-%Dz9s#; z&)18CrO&@sjipZu6G6ktG9Fk8voh5%ZkhZ2aPDCy%=k|;f46fw|2;BAT1TlKQyN4+ zH>>j?UeIojKPDsgt*u{_C4Q4%zyhzV41Hoq(!S(UCCTKV zVbSvBtXO3miBaS&2{Yqn42p>e4d-ISNSY?C@ zh2F5lC{QrDxTVFG;CJ*5=wph4z4wK-Aj)V6^crZT=9Si)CkVdX&~_FP z(&EFXEo)G4Wegx>g$|hKs2n-2cY%JFrP@D|ZPPAX)!y^2LBe(zOw|gu)X;^!23N;kOUqcaT(v15vV~ysG@!*phk9z3-mDqRoGnLhD z7$pzR$&N1k6zq^6yJ53-(5bv0Eb^r(-dmuQ2=8~oSf={=vQIDw?-$7H74Kj(+^g?RH0usOj2&b0 z09MeW1G@QPD8Af~#XtXi+J6o9m7V^riS_zmEU|x##{vWH`lig$e?U=m30vrwT6~}L zdC#eO`dg+I%y{?bpBUj`RL~Lt9#`tG{W@u58TD>XD?+5Uqwg@m?xC7;4-_*Lzb*YO zuJ7m-7wyR4Dtx*i_d3$7JsO5WMH6q`B6WdjBZEYZaP$ZLEB-O~O|)Hri0@fFSxDS4Nowd}Of8|bV{Qqq_DUPY2@B^ASf6w8)K znis6iKf( z_CLh3+AGW;lp`(}aX&R-PM7oBf@B(e&q#ikao0d$ROR=ynUPQEflQ*q7syPjk3&s< ztxd>=ZOYh}$b{F3VD)9XAj?V_NlbUnGQCD*W{Xf1y@ozyLHV=(nPkyzk+iuWQa2Kr zlgXMBNcgga9yhZUQL+SP#6C$DW={D#3MZ4<2&GDrlMVDi0Lr;*B)LuSTpHKhKPto$ zjk!_08829Q_LXS{p?Pd~Ip31<5-amsNeKVnlmr3*008^nw(I{=)NlYklf<+S4FUi} zk?~OJpp#1ladV9)br4TwBGKZWRaR)oV_q=oELdTT>Tp_Rk^tsCtC>_D#g`P;|23cz z{?$X&;K#>mnH&s5HXfCQsveA2-HEHOSDH=Uy;O^JLTAD-NBDB9^qSYIosqS*Vcx$# z_S}KCseTzWe|ckiNyjOco2xqrLnR0hEAXCcawDo1bPuk8j1ku&urnqYzBsB-ih z$ilL)<5vWFh199Uz3zsCVrsYMYLMjeJ9Ivzv`GKJ&to9rvtYDg-^kDJ6L(Kq(L?{v zFd1u{zL7`1?Q}&?+O$JsB6$^jFmXf>iL)4NBX8PGOlxmC;Yb#s(vp~KMdfu&Yj%$J zBQ<0q@^4Y{hYA(DsHe-`0*wgI;F3_I2_vN-`WR*%NxZ9Rnphx%iB9P00#>!{ul~)i zS?n05-*WzbG0w_1u4EpL?lCp4fSFrOAxtC_9u6Y(2PVqmH016|jfNvPm!hw0lS)gO z7hUa&3|`*l7%Y>YSfr!hUhAn#iQON>;OXT$=K;HkzEoMgf-*0V|^5D7`PYY(`<&KiREQUnN<4)Zc|z zt5rE}+m#t{HUE$uUwgM(v#O@vsE9B-Y0U47R|B`A7Ew4M_A{L7LM`4t9(xa3tuYO{ z2ANqPx|$k1WJ^9Y*EdHr^)~{dDiu~e4u~zd;+pV(G<-zLsaHkVs5bA)OtNnCv;#R> zXa|0g-0byY5aqg!F7!OnwwXo-{3*ahsrxg>edlh{1I$8qQl!mE3}q`_C>x$f zzj}c)*Ta{WG?t|#)=;kDq^j4e56&^5+3H^}?O!vwZ->jipK5x4z4hoLH4vMYqT7D@ zGlcni(5r_?t5jC=NuNEko9EASjedIpKtprnxufXQG<80FFKytmocNk1IGW?y`7AwhDgfNaVtr{zwxrFxx#2CXzPXR}Xb5OBMhuIvk zoI8(U9$rMQr&Lh0>QUxy%4ZI%DDj=z68N3^CMA(6%4dUl7{l}|G%bD{orhE?V?e;O z6+fgW38zUg{Gi?Xs;FDe)7+hHujq@)S;Iwk6_N zo5D~nN9{j6y3(+Il~2gvoZ@7L*j9P#XGTn{C`PgVXY0SSQ_o&f=nWo@w1$39dwrwT zMa8@RDF6_GcM)~;;oFFpY`Z#Q@>op7dUNdE&~d5c^X1lZN`6UOJmQ zZtl+N=2K6;RW!W|Zvg>wJ6|4qdV&3(o;yk;w^$CKh{hA2Go}8YT)p-PN|k^&;o@4N z`2&cebiau3W=YCRTu_GgR{@~Tq{ge!TCn8bf2OZ_Z zw2D?|cES7)Ch>0qd%+t-b)@-B?QUK;PY_!RD zXN!gXDgRL8JXa_9r0r?mOYhf<9O*;JL2M7(+)4m`7(cw}4(f#tH zwJI%tM(NEryJ!T?<8Q+@Yw5)@+30f#v4EqnirmdD0C20;?l7F_Z4zW3-II&CXqkaij_~^+cZ3 zmC2~(UQ$Nf_?na_j$*IxC2{i_?Eqz#x|1JFB8Q%HuwrPxg_}gWC=CT zOjMEK9ja70wugNc$qr&ig-)w84PRVe)lC+0-)@o}RiI`0LiKtLNouN=)+Q-sRh7)j zYrIBPUF4GYjgO*HKmoOIGbdHEf?E@zrN{!e?KO{*L!tfZGA5+yiH|26`KHWN?KZ=k zw;)%>H0~>MkF+)HCiaYmlr8t_ADm=ll1h3_4qf@L)a@K!Y;pE_LLm2cft4@VxfU~K zC8Yr^T&WG!#WQRzUFV&qZ~N~ro4isGVe3a_{CDCM&XX%8ud@;#O^%uuA5qV`aL2^k zhWBoJJ-uE6XRCUnuAa`yYqN2DqfWm5b{YKxqI^4}{{7K6mQMJwxjpQz*|I~pHR9|8 zmVDF2jD+xW`$C&xA*UnMoSZ9P`HAnmQARn9e^9J?*Z)1MuqC%ACp)@p*)3p}x56fv zQ)mOWt=GCXYZ&y(%U!s@U$zUv+-n8xCAC?}YD{&)<0Zxwb5 zyk|h1m1b^&m5I)@bkDKARPBr$5ecgSxwfa&SluD4oLD3cQTe9y16#CZ$tT+;*G(~=4b(=9ZM0@8@i1D z!-%f2heB@*9-cpgdt|@zdz_S<0oTIh8%?><5cOGP+OF6!+E}rIqiG}c9<8Ryc`RKm zbGUI_K$~x}i^^aM3Ll0e{R%t+x`GAWpDM#3x>@A=8lnvvfO}4Vfr_8YFm-X97;vis z%G)=R#j@ZY{(U>mLZF($@tjerW}(W=undrm7K@``bAL<=7PtcI;Rlee?HOi<0cw-t zmnr{t-POc;fDJZi)tonFRs6tyPy^p;Z#0|t!&G<(*u0gkU;H?4&um)RyVV*oeE>~B zvcIq*9h%yeJQ^KUePp#veLxJW(4K_svRC@hHRqf!0&e@;b~|BAHxlr84tj!BX%n}TntWZZ<$nA znsjB}8MRu%a>uy#kEoE=vWdF?ILCM9jMoEo#-7Q(GF0ung|aEcoFWb5V(0z z^IYVk3dBFk&|Hyy+r<6&6VR|%fV&Y>9DstQ>O#->V{?+5kPy!-=}&pcT2a2xB)2YO z7Y}@4TLvp!9`%(W%L^GMUDq;xkW25YVNE6R*+kjQR!EfrWxb!+oDL8_)%pOHT5WRV zOQat^q^qIL(tJlhosixBo#%oar|O*RD;={k^$wkv4m?i?_?h-OrasAdJAK>CM3 zaR%O`u4{bJ*s*^*JjTqh#}`I7knhx_ywi->_dn$mq?Dt4E10O)*Y00eBuDKEw1^w6 zC_R(vHeB~q;i%nhy<4VWQX>|WKUC2sm=UqMZh7uJ$QID{0b_Xb%UpwPUs$3q^$zzD2v*UVSL`|yP+Bg(GBqkKB&m{HU~u&GeA2d{A+2+x&R}zW zIYx~~Wzh;}U1KB^Ed7`8dPKq0xdla|vk&Moe}52p6l6VsTHQ6-_q2OBS+0R`Wp%O{ zzR5!_be%uty+j+Npx1%8-ogNdW1*8NkNZ|hAn^4WV zj8m`xye0)O>I>%U`NSxbAYUhi$lzb`Nj(4Lj25AD<^mMspye^}&iY|g`J>!e(CA%o7dNBXhktCTlKM6Ie{2o*#mqRo40 zb8NlVaXU)9SNQ4Y`F|?x-Ow%F%%HnMc{U4TUwEzTuIp!v__k0VeYC-I>6-KK{GDPMXwcO00of?MKBy(P6B0)*Q2HtruMUI z#ALl4HfTV{%4!ES*MTV^!u6zdnZb~q_0Ts^vq(O(Kwa@M!>o6?d|h;TNBROR zQOpMXRS8u6gbe7&7Zq6?X>8+!>H*6P~&>hj~Dw%rxemR?0+V}q*<%Dy%R{ll7&UoI+vlqLpB`BXi8H1v<@1?H&O4a?^tn0-gBJ<{V7A$?E5z8BM<5$Z zT6OskOr2S6ajpe}hJ-@bQCP5J6occvj)M|2RVYvG?WxfUq7ksS6so>%*+5;HUJYdxCW*|6%1xK zWfBS_ z#9r~kP~A+Nz>-3=L|1Q^`P^09jQiH<`tP(R__zn1IenV=%IuirW9Yb$4B!CY(BQ0y zs9x8Gl^PlU28<3#j;_))H^%SGE*3Y9sH649#b7Hn@UNnVkbwqfKW4NMQYKf@|4O5n zZnJTd@=0!C1j8#BUt$eQLk~ifwF9C7@u0J9+0B1l8dkt9AeFOSd^Mz-!1U=La|bBxq1VHN zp-mh%-R+v4%6_t>V1>eBbYcus@ZSs2zmyL0CxFHtdP!cv9da^|0Ro|%4*15#k3Xnq z$zP12+o}Y3hv|c+T+*eCs2=GlPhY3$S+ltbGwle2gE}DEdjPU0fKqCQT7?g?!23ci zOC+Xnxt5YmnDT>}kq0Fk;UJjQ2kqZ10LKTT?E*j&-l0r`@>f<=_O91&RFY9#D%-5h zqJhGU_bmrIKpWOeu!r7DBcW~6rG{XQVFqtBdE-#FVLnmguG@U2Ft40Ngv&L>(gO%! zU6?v3Ru`bZJ}9B#9YiuMz_1kCl%CQ#8eQ=cefPE*MUS-94m24+8x6pCxZ4V+2M6xr zz(*&U-=16);I5uXz)sG3Uk+m8AB(`7Q9}QGPEtPws|UZilP) z0v}HY&H4g=^cwW%0)XBQsN;b;i1SQN4mO_*u8xZVFLCO9&^0g6HTTf9FVK~20Mso+ zwyEAUzo|Z4b|}D`OPkt1jrj300 zLPyFFs*XGFq@q~K?u*~{KT*bHD~&gMf&#`4!;XF8yA!wZ{)s0!7oU`t>mUwZlsbcs z#$aqwQ;Y5r0q!o!kbIXW3lvUet`JHIz({Npl zX(d(hj_msIw25=HDf8WlNHy2~_(oeX7UjNai^Iv#JHe2wh?&342fv5|Jotm>rDZ!y z8P?~hu)^aTBf{Zr7QdpCo*{o}Gx7N9nIi@wli|Vd#sbWmR@ISc)JGSqUJByyST%`(M@Wh%+#KWoc0Z;*Is{fe_ZUnUce@zeuEC(6{03AgDD$-%2m;glK*q!#ZpUA`~(2Cpa_#gDi z3-wJwC!<=-U$pG76ThFsVL%g#wEmpBPvy+DSU1XDEOXVO$%hsx&_>x>F*r%uc5dTZ zlUA*pV56Q4EBx+xjYvJHWvkQWQ$}}hH-mb&>%r(-=S_1(;ONAvNX*!PU&ZL^<&>a2uI+O5C)9TGuF zeEK%h>!^PYzmj$51G-~lDM_>W8*)P0nf8u#d5EvO~^wl}a zH^{`fN5_~ts3$r8vo}PG5dZ<+=55B(WYp7y-rnm~7_oinwsJ?AJhC%IRn^^*-MW?2 zmVDXjEztz-qrDFJguX=pXFlweG#8cw={rR|59pLrg-g z@Dr@E%uWm_qZ6-^&SP(iGx6Jth20%Y%o@9^M+Tgng7P~?MHzB9W1d>(l81A0rIv+9Fux~PVE=a$V zS{TFVtlVQ>?T^+NLe*dD!PJc=&T=Gkn*pGGsAlauy<~v6N>SgcjAfJ?XR_VPz*^Bj zRlie(#}KPb=|@%jquhKU95hZXid^=HZ(%!3m7ZsTbWLSm>!m}_a)q{;Kpl_@Uov22 zQ5qTH-XC(F52NE}??ZvY8E z_P>qWi-c|!j;h!S+wjc^pK_=nlq_LKsKVun2_}TfoUJmO42Z=o!(m3d{&Bgu>wX}$ z|IFAS8lHUm3HlMuPbz}9NLOw#-y?7`0 zywM&S+ilHHNbIVbrt4nyZkt@u9ev0QZFc&1faPyQ`U+ymBX-+f5=53z(z|rP-O@M_ z=yg!D=jafUHc|nSHR#6jV}O16y!`;`BzH^Z@B|#Zy0*I)A}v;k_(}Qg9x1R%$l>VC z$N@Lm{~p{o_A`|??b=ouxH&vyn!zR{`t7*sSn{uoP6~#uWrzU(Bq;wF!@P3#=huJdk3=r}k=l`Td?gawWF$nR zIcnRnrI^&ta!AonzES=~c%^NJF6lQV$Cy^ZOr_K|c}G2dQ& zn=e?V&C8(Quu16oRj~G$s+k4;X#5_M7dIK>Y>R4cJP7P**9&X}S}RTjdpUH#Z&^)w z!?ohEc#~SlZ6yb?mouyUTdDHcEmht07FYD;l{8aJ-rr8ck%fFMsh%9Rmh9`X2p{JZ zdtBK8eel=2o{mY$a|iTICONaUj#lDR7R9jbUv;0~DJ!y@dGed3$ZT{q;c-g}hR^;` z%Vz9w^a_VoY3`C3^@<4oSCRGKkNG6VYH51ee;nML1_5$c_xDY&ime~AgL50u+eC>Z zHqu$x-V%JxaG@>QMmOIW#pG&>vUP8gmI~(X{oYWVFzQLp@pvsK2*lInY|MnU(CcOImY zC0!X?^DoV37{>7O%Wj-S2i>&zs&6WQzD_k7^A5ihCXl&Yt(S5tu?z5}{ytevALk>? zt|z{tJ}n?^Jk|C(-O>G$Z)7XLPH(B~Sz*!Ml0%9)lIC#}%((I$;x0q=?O1L*p3$gY zt47c*8ZjVWyg2h)&hG14HdL@AEm_Us!*Tx7c4K_ZJK@0@_mgwdw6JGE9U)qPopQn^ zXn$NbmlMVZR?tS54eG*^!$3jAK-WnAQ=pxmcJ@0v6FoxYNt)+A1?Ds>Tsk7+on9zi z<}~`#>hC{apTpy>zn(7&4_$I1f;P-IYgK86BmJCv54Q%~ncDTBmf$H|2PROIdpx7m z+V%iqIpol@`p;yRhC0pykXByWSKT#2EpJ{7NunZK+plV}F`kX?Ca9e427YBN_U%n; zyULfRZKTv^Vhmz`od5`OBx*#(@&(W0@m4v*LaUx zD})~hD%+wE;w4`e_YOFhKLA^pHhQ|RBp#%7xU@@#wkMRQ=pXx@FG@dM+nT&qF^?uE zGET{vx6V7>R$sBm=*mB8^-*MH_yY>A$Uni{!hs7!OLWZTjARRBJs-PI3I8IdJuF?^ z^}hU{Uq82P%)`Wha=#d8!~i7|YK==-(=FYHxtHg0bX-9AchU-gIaY)|CEayYIJGdV z#flGwm-n&MDM=gb-|tMo6TQeVcF$4=vbg)_T+C)?w#KKkYDnwXss!F6y~ud0pb5i*zu{z_5uJly57fxVBj8`{-yFcN+i3I zmR=zGZ9UBR0v$I5FkXQ6P#+zf7<}erKo>(j?r-`t00(IU!?-}RvH|t_5D@4@w=tt6 zlS1N{0SQxvShZ0^A;M=<4k0}O$YD6f1;EhezczwEdoQeAIR>=mZ{#0NH4<%G?~lZ^ zVmefMc*jyyDMWrxAY2&R797#$LqKYSw#|%!X&<8*A3RX+@+HP33h3Z&=%j}gj=lt_ z_rcK*0T{5LGbY7<8jo1`>%D};q|+m!T0M$*bI0SnhX$Gk-~eJtd|=u(C}>L{%vya^ z-l&gyQEfHq=G{8U%my2|5?8*49fh<$&RTzbFf@x8lkw0* z7pr5paOE(EX>KZ8hqfg&;|pwSne_5K0QxYxQeFHl0~~q}ZK9(hlJh?a@T(XQ{6fT4s4d zLi9i7TBIIjMdX5okOjVmByiTow9rb3S^?r^JZI-P)RtLHxPz72mad2i1>BoO7iy-k zr4v;^WJc=8Eo2G_0PE5C4Ib(gq35qFfcZpyr1z8NY6w3;QD$`WvuV-2T=Q8*@)eK0 z^n40J`VoXR=zgk12X%zMeLZ7X3)h3d{Cb7h`yMBin${zQ9vg+34~4Un*fql*z7Oi8 zy4n$21yY0sCLJWogd`P3MDGPS6Isa-s3b2JWX+<`mWSe2Rgwl4H9~&bg^&`507~H^ zs;qs2+awX$Rhqy+b!j_}iIw8tQ^BY{OmC8iQGyp^NZOPzK&d8Tfek0T_EBhD)U#bhGEg=WA#_ zkl2!Ks3#CYX%^P{xzgj1;B>F%T_+#v7>IxTO>{?9REn~Rm|6m6HMSUuX*X#8swSxd z8yh)N4XmqpVyr<^t0`M16BT%rRMX!PH5%D5}Z#~Ah15+8csSo+yf<98sTix{A(0ItK;{LisnGIYbDoyLg z8bpW9LN6)I1`2;H!4RV}HhU|qTe%rVHN(m&rbht0e*1$4@4NyxuL`d+Cy(E^HQJ!k z1Elm3#C9v#GOABKP=qUU@z%DZB?O3TiXeh-^N>2c;dHaU?edLlYC2Aid(f@QMpj_2 z*i_E|utdY8GPWg-XKVRIrB72z%4?yCthA>UcuMq=N@EQ}yk(4n(A%H&WxS;*zpHEB zRdm9MKg!~t0(wrD@Nsf-(0My@{oevpNtTO+;_=}n*S(74=uPH$+so13u0r4sAaMkJ(wFY zCma#4qMhk1-ZVEoE-D7DjdHY%UZlEzVV7{aRLBh{i`@4SnJ2tsC$9mB<}toQGiN}| zjcB0Nle1CD?vBfslA63rTu~=6u$VaUtyOeyTJ-HFH{kz0H^GucjI6H9a3{HOpRDPm zs@&-v-X?P$PxmUN8i0>OOPWcVB!knae*KuJG4Hx}C6Rlcinn8OMMiuu+nzQ{n>Lb% z`r;(7h>_{z&S;cMgj+O3rp?6DO()Ctm(flpmX4LBy{4?5N_%E@YbNoGn5{*b%CYE9 zB%}?(oohJdD&crHm@>!W2mZ`UCMD4rxy|za*PMXwcox68+w*I$&WUMU4qpw~XS-Kx z9BfN1AE#Q%1iI#Gh(0KhF3h0Jj<|L`185zdrxeqQ=PXJG>B`@5EQZj|eHN=JH=Hee zzaX_-ZGpD9TKVbfS#$4!e9v?q!*WSB_7ln*(l=T3e@>Rb|CS)0WMWDIpZHD@OQ1QP#0?#Tbf zrM&(ByCW7@fMJ(m?(qfPNE(naWt9c|WAcA70`I+R<)VQzeFB@E zwK65ZUF^PKFtcnqYUm=Bnm}z_2{N0FSSQC-uTG~#h|>&uteN+*!EH^=twJwa?GGz8 zdyt1-IC9wROViRTb90XaBf*lJ0=>b&KP&&7jk!jwfSn|#^~CMmACrXhv%Uwk9ZcDT zZs#T=+GXcJzU%T>`ZmrerL}v-DgoH=zv!XgEkwJ zb+P4K$BTqm3n(ewKx^e_x+W*KoRG0b9_;TtTw%xRjA zEjY_;w^88MPetSGh^<_T&~M44KjZc^O{)My9vG|sJb{}B1vYd*EOEtI<)RcscOfE} z??ss2pZBT&oX+!rs=*f=b3#ui<56c-3_^GtXwZ$RH-0jC4^lbti0fwA*tMUTk9nxHPO#sax7s-{0%2k3I+ z63{e3xL}w9O1?l_OcEMh!=xFp0g!jy)z!TW#GK}3_f*`7@$h-E35EsBmkb$tq@`QT z5IXM4pjOzd8enCstYXq)L2v?qC>x%mF@*q=FstHoBlq-pl0h4|*soFVN8^3jhNu!| z9;|1%*5-En7Q>OStzR91Tw-E>T2r->njmtN}XdRofcz0?0u%}dRfgBkiM3W2v((xtrDVl8hcryr3zqSu% z|MuUVIsPGtpF6JfqR{C9Nm7XM`W~fP(!D+|$-*4{=?xdF?^5>={c%#d)?<`cL1IZb zzPQ+{A+ob4o<*Pw*5*Zf&U%g>rndQCRX%(u#Toa26j(Bgq|J3I&G}EC`abIqoLRO( zaUDsD&W)B^|MXU)sP8z(Hb1dXj)!jAs-q$6ok$Oz#d@dELZnHPa{eSL*|*Sm>HPiI zC^$flr-e9<0Lmmt9VR8f)F%RrgO~A`0HI#~1ZNdX!nFR2Ljb~{l?7CS^dSP&Yp%-N z5nn;hD!KS?6_`Q1$yz~$Ij^oR5~WV}osc0a@@TX$1|n>1h~!F19`R)|26hZerN|fy zl@6Qk3>`_wCMQ`KZ-P|-j>b3(7h4rev^x2;$CCeBAt!dFx5)=_Ob$5Y#l8Zik14Nm zMysUi2t~c1#LGkSnL?sx8CNr}vfb7g{)Hxprlr)(s0j6k4P$0iWyz7d*-^AB)G0;? z;Q0A5wCR2IvaLUzTINyZL8)O9iIUKi#b{{GUc=Gtlc-Iun^x;bR5ChV=Dw2LqO+U4 zdM&pPW<@XRm7R3S`;ex7=jEGe{{DM$n@A%2>}ZYGgM9XTV0~-N_?(N^&jP0h2f>nI z0Fb8wfUX0hbZDISAkVhF>tYsIP(|h%+fBXx*f+&q<^zkNxB*}kIuAOLGg{{J={Bb6 zi9Ym0N-bI$XHw_6(`;ZdL95l6CNOd#;{2GsoFW$LH7QoL;tn1E3Jl2@6e$XQ{#|)+JG`KrnE(KW_Mj`9@HLcNspP#5)~C^5fj{L#Ty^wg7_b9 z+!dS^3>WYx%EMew38{2o#z>z0I@-unlXX;zE~6E9mxUI0S3PiA!q3;n_opbJCBRO0 zPX)o!Z>vYpW=>f^?U-Td!TBXVCUgW;)_z8s?G6JO*m|f#k{mvP7F$i%>jUP`HI`IY z!gmjd@Uhw~wSA3Ic}BVrkxhYy?A)ib8#L|I>a#DlpzTFyB~Aq}hf376OH7~&e4FV* zAVD45t-?{>H%g=s5zlsdTPQ>B6PT=|gzp`cr52_kyk@@tZAiG3X{S9k+K!|HjXyI0 zg%3mD{EL3)`mlJ=PeW1x+hzysKAj-%ssM|cm|G`KqMLahSyK&%Akp0DuJx3u-R4M(fY1R{*Z97Q z$4MoAv6;`JRxkLSLPBPbD}_S=vwBKnq;x;#CqB@rE2QV>y}n?xM5}K4IPEsFh~P#1 z_wGl@BTR`Lh)~HDWkm>cS}!OzdPxAS@JO&8!WP;B2DAi9o?76T0|rQvTI>ILz`;8- z?+?3Wx;Oz|*V`LFKyc4sx|9j~AkCpv%0OOPF*B$o>IM2+^qdMGZ&Lehzcb-NH_I3Y_Xn)c>ep#N3D4y`VxBW6{ztw|DpdW#iIM--w!xCuu56`!Y{u9P(*a&<(vn<@81;lVSXM9P%Rr_ zOF34?MjND`kC>GvV^L9j=scL$+bH5Fo%ex+$JRG|H#VP@zamSoKJVB6Q}!3v|7>(q z?$dAEa>QrSrUFh?33y^Tm6t5Sa`X1zpXy|^C7ICD3ew0uO(JJA-J=xmpj|IhHHXHk z&Z0@uXJiBU_)6g?_OXv^Pir~+k(YmmoNp8yE~ZFT5C@-ctwg!5=&6a{azdGPoTWka zQF+9X!SZ0Mddbaq)a8>ZJ7bY)hmlI_Vf$0zf0U!3@xft<;Z(;=XU=fy!|2b3Av6Nf zcKLeaiP6TlA#pJd^djNG4nzyWFd>YvG`h&aBeFGT_6_48aS@X*&Xz;26ZQCLOY;+z?inJmD900H zQjvObZ${~?`;xvo5Zqr$7Puq>Ymy85l8an$l&d2_WcCCsq?EzR?l~y~1u3aQO3n8v z%?u$iO36+&N#?#O0=aRz=cHlF~OqAiD{Wk69kqb>LbV_m9hzV}ZvT|TqBG4|CDl1xWEhL})}N*K z+8;QKaUm>*3a1ZoARsS_Qa4dWr47XDAU^}+t;iJRV-w-UvG;bslfLI{JHm}vNSy=4 z;~k<{}l)VkMwoK;uV0i;h~$o?nNBj!?ALqaJv(vmZTa+Amm&(cIz zaxl$bS1z;Ht~0JiFxOY!*FwLhRN(Ck>89szJ+?BUIC5@9r8E0>Pk`fk)l~MaHJART^ z-yV4IgJJWGA-T|9cp(E;5;5#c*b$v`+mxfZl6DZD)GP+P{J|h#iT?n>Lw8`CpOatK z0dH#**-f%?&Zn22rc+`i)>&E`86q- zv1f)3LOhrnG?dD@3yGLhb0BnV%y~0ndgeEd%|HTZbj&N}Y@fKkP)4fV6#EPCPw_U_z zvG5T@23Ie;0NHD6V~VcQDy+JIE$pKEF*QYdZeLHuj|u7e{E9J?%+{YOnZ#<7m8)t= z+1cmuE+Ke%*$u`9(%LN`R;Fq+Frh#+UCerTe6&gjY7IMIs+$#1QiRMHJGEjM+-WI( zJs(WR+HfFNHIWUv4`zR%OM__E7rhgAh?mtEz91>Z3{%LsjJ1-r~`ZG8Nk1Lsl39G`q0f2Awtup zK{LQ$Z|aarpjk_U!$5KH+HRACqo>_PU()+t@=FJh>AvKf5}}cHjc#3($ z;hblSIsaX}zU(IFSUZ8_Q&nBvnn`eYQEl&~>|Gw9Y*X6021;iyHT3oT+XLrkb^T)B zZ`V0cloWjSTNIPb@j53JDJID6#N(2~^ihJLt)dcja`4JKpS)Hdpc*v(OZQKJ^A?9& zdWO|OK#F~5B_ql(%FQ6xMsc3wusp|LJ<9;SY8R!G3e8rCAXVS%r-5b(PvYDFRHH`p zS`$p8F#l;*!CJSalS;q&Fm6?2BHV!?(G&(rBWP}E!?f)m$5F*Uqe#7@pHqi3bO)_F zEP3LJ`KU$$j>m=n)Qjehko@j{m>%VG8!}gkwq+eX$%w=;bP8=TE1RDJ0|d)&p!K3TelJnU%un znO_Pkg$sRqXSEvJG3^Z(91eRORB(DbbG{C3eln!;?xva$XmSgs912t=@<#n)fgz|) z1pFT!Spq-~An$+mFaEu>7xscG7YB3QANeoG!M~UGWdC!;C54%dL_4~hjk zqx5sOsK@!%RLhAn9L41qm9mzdd@(GA-FUjSYOzAS#6sI%yLvgUpK7INTBsZlWIPgc zpYhS6|D1D|2hsv0y^kTK4+SfA|c4w zxkbD9l$ppQblQ4bNdMDuP-oV$hyZqyf1N$;nQn$G5RpHWxK8DOMF8+0^7k%S*FXjtV4ODqC$bD+%i8h-+;9k zjbzF(J!0A4KK1kKv)Y)>wvw}moZAh18Ka%@>f)+w{d0)Z^snQ%rUXalc4U}nMsxbc zMt%J_jY>(wNWL#~@gqOGOG6^}@7jUy!>o|<=@+dl>AdOKE0a?9p7$NYq2vuU`wp<{ z+TbRYs?x#si_QK!ibuyrrtIvKzMIg`=FQ%JH)|*6u<70pVX<7EwqX`gW=L%^V>eAn z{aXCwx|eTQpG&40cLO<}F5_!S@&0T(&CVm`5W;Agce=t#sQnPfIe`0PO=DldH77T; zm)ArxfU;@CIiso^AiwH+zfC6cCGqQ@AMR&Na00i9lEHjZS;kZHNtQ|rEe~3bH1cy zwm|aI(uRJ(W&FwO-^jf@zyM^Y;rz#g*am6k`JjhWL@$!r5URR(|LGS-x^B>RZOpo$q(XxCPHug<4Gk4gtbf{U+c5UvowS`*xXO-V5D zy=BQ;=O-X>nmWZGjFAq66({c?>xG6?0(H<+DD3(Yi*}Z9ot!QkdjpKfHY;T%#)7}6 za^dCvQ%su|G4-qP%5jizR&#+h(Jt;bJ(g&s3Y}l%NUt=`h+%~`mNi(Y`6d{quOStj z6lb^t!d{IolhLuSBK_$0QEikCTvrS;qbs7mGMk{6j}8y?1Hx`239~O5 zS-dibDo)Et2O$ddUXd7~94S^QW%j zOT=&vYpVMKBdtb)C=j$_Nga!-A;p>vV{9ymBD@)U!>I|SEo*day`1G3h*T|QEvMB= z3ukW3#Fb-or-GVey?wPXn)%g}PN;0lDg~rByA($4S<3(fgrL@QYw%Wa!o@ib$PJS5 zUGBpj?nY(syb(+7?H9Z-4^14c;J*82euGn+40^uU+-V4@JgX_s0z;;-ZX1R@UBIdfv@ve`Dr!`=1!hwmyAFaA6 z*$SQg;lme+zF`|vKU^PIfx=CCbeOzK0CZ#k1#Rg#?$C{C>T?-SlPlebHD(U8L>4_| zTVDLot7tC?g+|_rDRmkiZ&%qPJ)N#;Y^58rwvi(~o6;GxM}DPAnPaphF?n1W0ezQ{c`N^{NJfck@&1`Mt_!xYyQl|EqVk{2NVVYdl6uT;<3k z+cNU#rO0(2j*ruXo#-647gtQ0^yfzP$lh#QyDp7%_o)s2o_3|X{jU%_nOs`<+;UC^>zt=0bQc6`wn$xB-9)me~Kj~SVATZzt#Lx=I^x2E)p zj;`%-!wRNVMwGR>KYNZj$UO_6aBP~wmYh*Uve)luI$F?N9VnG3*J}Uz)cpQX{eJdW zm0B5ENBzZiairR>z0*h|UIr27w52cQKl)>P&+6O)qTYUs-+!8n`r8t-=)bEd3m+N_ zb#)WT*(=h@YM}dgW=^_!JemG$O!kXY^H0eUns20iu)i%)Sw_2W>&QC&zFwq#XFWH- z@V<1i#R*vJyY?;EzC2HQ>ZRP?dqIv(w)EQvQRa_wU%5LBe}3i~VKy9Ob6R8`0ODQb zX40#|;`pUOz(ppy|3)q#0O3F667#>wWsdfL*_r-3x%@kMnfjMpcrB-Ml?Y2m{)=4x zwKF;HNQeqn&X>Saxc^kNvMB_-Em6)>ZmV9d(Q9$oo^GpIZ7>_b4rgz#U2m~ltTyhs z&CqCeYTKa_75Rr;NVE1M^j%!Oc7H^{p}}qC*~<*Z=Xk8~X;K;qpk(m=)3geiLFU7m zGXnCznZh8%0dOVW%T|_SmA}U9FdGPKP^|xYs>5!J)MPB1D}@POH}6`}6K}vBq?^zwh_ckAs=jj{*HJ_XB0O z`TB^zt1tUVr{986^OsCRAXu1-A($3ryFU2BorU4wj+pZ!P|RU_kyPHNdr@?etov}r zbYG*7Z;I^b-EosEieoA2Sr6hAe-bjpkGF~L#7gY=8bkii01V|n0kYt~021T>4Um7) z6$C|ZivC9#g#WPo?=rOf>kIh5We7;dRc`%{GO)i{b2QTmQNxhD>L5Cq3)Q7nC^cx~ zm<*CdynrxQYSzOU7xC0*9%~IF9`VLMsQ8!#hxuHb|0h5~TWorjeGAP7IVad@oI3gT z(})>J>Q9@l_9kdvUSdug?xuz_-?`K9W_T*LJ*`ytQyCl_8$Oxj`9EAQbF5kNfzW4q zMAlcbFnamr3!!f4#FAw+5WmCoojELF3}=EzQU;7>N>Q`Hp$bFf3Jyg0+nuM8PNORP zUZ3(o@9Q5pqO4vjy~9gKGc~iP-*@_O*3o}9DdJ~e& z1+$TswbA*JM%$VigU}~~iA@Fh%)H0DlJ^WK6*3aJ5YNL#P`uZo5T{M@9}wd9Jg^x7 zo|&D75mz^C&Yh_0NlmWe@OhN7;g5xK2wG>Tm8@k*8J!EK^AwW|6jr)pM{xWc46wo7 z9S^|J)0>Ww&;BRbb%bY9BUQkZ%E)#3OU*rhvRTal`q%EjX2ASgxnR^>zClVSkXc9! z$;+8xp#b($MP{b)V+`Rd!M!nwj$V%;LNqeykOW0iMp+Uj6M|3Oe(9aIA)(Hi%```W z5%`}A9Xp5s{)NWW(A5_{abwy)fm0`jX9pS3RHt2)!k$BSj7#R!~WA{QlO>Az9&H` zccD*+gDJw_ii+qSkc+F`3Xyl=4_=^hlZk=8B^12E%^fd8Ty37FmJW!F&`-#%A=`pjtzUel5r6NULgXk!eMr^fse+{lrw1~5nE8^^Pg5rbMar`pb$*mV`_;nXhhfCCT5;ejda2d;> zns?5HM^wI- zPXm4z^I25h>GWta{3G-i8e&wsqPV&l4Xm|t4EUJe!{5I8>DHUkHWlPG|kOaAQE;B zQ%1LKy&b9{(CK#wpu)Tv1nwi7kKIOF5S!qRDFDr$AwtQgd<%oXs+a+ z8arawETkf1mj^%133xd>aAGhGio z*_&fU8}r>^`S_@FSzUw>jL0Z<1&Y`p2?bfAsF+M8RULQlBI|Kwg(>^Am@;czcWn(M)YQkO8k7km7xj*&`o=>PBTbuK^Q;wWz}akO{h$O^nsFGoIbg|p*F2rEr+ zOoaQ{f(}Ke=9$YeuFP4eQ^_hBH6S5LBX$Z@O;^!=HK5DYAH8qY^Et%jk4btW+zuJI z01syd6H*=oKV(_#4xSj$G8=$a?QqVaWi!!gVwIP16OH^ z3ze^RzV4jzFz%Jt| zDa-`D@k6MmG#HXeh=9G{!!sNm+d{^2`GsHnH*Bl%M_DwY_SmJZoh+PD_sf8NJ$A;p z9<#-tTW_qEAk@krvZ5eN(O_2STNI?gh+APGOk|%hgh)>u#A@&1$#7(@!MLPi%C8#f zZXF#ankW#;EUJVW5<*@S${C_m-wgQ-HZB~sevD8TkbT?S>K^RnmpB@lUit=w;A38s zzjV6;5Y-b>syU4sekx+oo?)%&CH^^8N&A^8m(kdG zROHuUvQXe@4AH52G0CWM@AcIC_#*ttX!tSTc$PyO$)ZEyGEeD>NW(B@VNz0=ryw75dwAQQcKlh!a>486%zNE1=GdMh0I zbLf(3o|SaH9??L*25$ALn>d$GEPui;(2C5c1j#kT8s?q zYd1no!cq0d#H8bnB=Rct#~;DJ$z-vff*7%a=S3)3#TeMSl9P2mSfKIX3Ydb&>+13zY(RS{7pR2<^Qm87&f)?j~!?UI)b`Z zvGoiV)Q@`XsqDotTJ^5Q!L#B&BT5XlO1smWTALYm;~bqD%@E}ZnZV8V^MM}Z>xDX; zbS~p=*iZCjmChd~Lo%xrD>$S#b7PT|Ggjw%;KMOOd>=VO^6PEW3>;hZDSd$TXpC%Z z{rYz^$E9kliSV@=v#aSU%Y(Gatb>cMD9{ro+zr>89jqkFR*c+2)6a=HcT0W0toyz# zc+4bl8y`1>O+@P=qWrplo1Slby*{ZoVtPCdijtj`5bvA*j()S+p_W|d@hka;g-(88 zn>jcHecw|(faQV2G?*rBWiH@5mNr=a*GopO9J$UC7>28yS_mLD!32MBdnNeJ1++q3 z{DBsabub2Fs?0*^@6_oM(OH@skPu_ z&8m`6QQVY(pEM)SQvRcxS~zjKE$tL%f}^mW9dwTn2WluTLb(Y6ejRDH}6Q2of^_JkqfoLY>M-OGGzxl%|`Bj zz_(0PnZUz*J*=uU!-3+uGo@emy;DK*`DN9mg34(nvfI(Rlz$ucMH{kxk2=Sn4Vf6O z&uwekb^F&FpN23msvRSJgq}an#R!eZW2E7VK3KKURHCwpIpk1&4NZ_ zd(IFNn>z6!>8v4>l+MXG{wV0mcD%L5eiIlm!%b-xOiJnV;MVh)6L!x#hg4mE{6LZftVa!NEeGo~}C@>re zEfWETOhBpoTC1+{zOH{|A@xiHJoQ=(j-eU3juC&dbb_#%#J{TVwDaM%&q4!#&D%%b z_P`#0+iAXZM!I>`p%XxoaK9Tt$@JY-Ycs$(9+}rxe4I4%R{0=}Rvm(r8T;OG9XsW{ z_r}*RY%iPR(B2%V8^7~cluxun!K4F?32xsMq&fKaf+$RyPgkW!MGq28B?rl$GWXn( z@H~@$9{>6Fr>nfk@|$mhI}!`oASMp)7|*o2B)XoAO=Cd%CG%-*xzrn6%Onk_w$*pxy(Cej=Hh z6SPK#ep11v-%F^8^>SntmM%mH+dEH5#kMZ@XHu4?zO_e=TQuxl$`hDp&}=ZWVEWS0 zpJ#vjsjhXb;oO`fXpKR=f2oWo#2(-L@2&2ZOqp-FVl9OOx@%wvTuMghi@>`UA6F+r zVNzX{Uf$!Joza;ZWGpJFGnmR?^dQUtTpnMULRX8bB)Pdx09Ha(LI>ast`Owf@#b0fs}yxa;!v%G95u<;ay|ajY%nHElNP;zIG-sF8WAt;fPC%|K}A z2JC`MQwY{e5nU>i7&b3kr3Yq_`(p(zH?pmePGg5wjxOvFFEQrk&OtKiN)&4wU6;`UyiV^3q!GHSnsG9;5#TqxL>RvzEc{Yui_{eF z^vDPHDy4RuV5ddO_9~Pi*AVoa*_w?(@<$eBGBWVc26=JE1WU9<5trQ!q7dk zaL_2JMZhUsTiPy%YR#|tvJ7rGc@yUnyc-W%y5xr;4wkwRMEr3BO6P{6dj=KLB<*bx zQ`-JWxK^IRJju@(Vm~(Z<)E^d%5EkEF@_CKLc;y_YL?m-ZnQ7B;MOj2>j}IO!w_Jqg)EN2 zOEzBsGUl+snet={TUr$|+mCw-mE%aEb%xDnCw_mbe^65v(LbpdC0&2szey}N;&ERE z-TFOxz`xX^v|=x{a-%P`%rKTGk1SCyV{EDo?{2SJ(hgus>?0W+SG#mpV_p0ts3Jq? z{05y~TmrA#B5JZ8>4FL^%_@{rKe!Ta8t$Nl@6D*w1OcFe4*x7i^?D9`Em?{sw3?-) z_x+5uf+e=rbZOmiAQgMTBb&_3Rwm;Q6-jtdcNThgPJ>I|_H5N*cm*`aVbt;;6HXae zxR5{*p}$IxQF@B)o6$ilOZ(UaHMzTxK%lp=2x&N(2AHj%%PrpPz@tYgz~OQCBQGn_ z%Bwl;jsxRkY;EySRu9Tr)OJq3QJMjLeoZNz8s=OC=bN`X7P?kn-~z)2lMghht5n#B zNk4xo8EBQe9lX{#LjAE~5)4CHj<=@Mir9eu)lPjADi7l=a-GRc>Ya%6YSmE%K(2bP36SAas zBAc%%qrfrcOJf|Jg^~KSiZFISHy&e1i-Lj5H6-D`uF| z4f=e*@strfuPF6vf=u{?Y(-aKjg)_5)pnB9awp1|6CQlXB3^}|!r$pIH1;v3n3SlK zm84Vh{>gO4hqvq`ggP_mk76jkU8tq@$1TO+QcvS=y4((gdM-`D%Pjl@KH(9vR)o(V zCp#p`o+C7Ton$73n+ZbUe4!@yL3J77yqTdKzQKWM0_M;DXY;|UNgws>6q8ybm@PwI zR|W=?`9m{hfzC3VI@!~f~ zmOAs^-ld(Q2EIND=04TFK4KQ0GAy3tM=%?*P)t%1=_oBsu)HWg-Ty?(|L?tn$;Ley z03M-?plA~~F^rVqKhRP!(h2hyrKo$UR5G#x{hucd@J#wYXc@-}-5tmgL2KNqYWeq6 zGk!V|oBGk`f4kqQGN20ST=fE(K0kcha)P z1X;y-Dx|iGh(s5~F3LR+ucquVz=das-e>BE=NPYK zy&+DQPvpWNpP+HWseY#Or4!fPH~J;EMjQ1P?b$r8id^(S8_jFQ*oWKwDIXGXbD9Z9 zVIoNPV&Y%gOQrB8%Sl%G>$Re7D6?#>Z-teJ8iTr5MjvLIPfnGd9S__81legLNu%iLm{E)MQJW|e3XinRf;Voqg_ zT}M((G)?2~VktO5>nyKEO4qM;c%6@+D6R34sVsyl^>;6qSzI%7cqi_Y&nI+W|2ldz zk^N3L)AHPQqa`}pptcMnFUA2{gUzv9RQt9^w>DsWC7tI0dzYo=d?+KogQ9CaMdmHG z++O|@uC}2LYuPkc*(C)|D!_aN01QC$zoW4SeT)NGg$CaG9w-U+`aYb--kl_YtghMw zy@GtU1mM1tOPrvHiqi(2#84H1SYMSj2Xl1vCXS*GIW^n-eO~zV<@#vr(^v5n$F12{ z6rR_xA>umrCIJStbb!`h-J?ZOg}QfLy3Bdm(gpREUa=FeY^k6#Sgq-9<{T{E>z%ib zZHl#VRtKB1!ts_e3d?y7U1&+|kE+)~2(qVdc-_x!3h$HBIg$friR3p`Wv|Rj-G4LO zZPU`{J!|N-WD2bsh_DtwU`xemPBT+F_+po(3E~njb9Xu z0HN^j!2o{ci}Vp1T@46+R(|{psmN)c_3voCua^MJmxmRGp|?`^MBbe@R9vqYyuH#9 zzG;Y`B3&+*jP4)!*HA)6#djIfey-%;)OY#3B*u(F9)F}1Y|P_5BklWRwC*5<3?}@V zvhv0&rp|LteO;=A_i<%p_jhd+rrg$-kO`IOI|Z`NjxM;7Sq=HlVFbhs00k`!((on% z;3HM9!qKBcgXvC0C#z;nD%t3G5y?oH_Hz5QkHNI)NH@|{?S~(8WA`g;P6h%WUXf!v zMVw=ci4^1KJoTVByAN?~JSD&dZ*$rc?! zPuDsm(|Mg)!t<4#B1x_CHRFd+P=RJjRg}z{QO*}dqO?C{o2W;#`h!Y!tqDY*d-GYRP`xad>0GZ;ce*$VxOD9#vWcBhI+#Q6=CihWpGxOQEv)qoz_$ zHdHyU4ui5DPL$-4JaTwH<9+PmB8V5z4i4Gys zp^MlK2M}$7-g3s%B`{%HqVdM`sTMWnvZAmkhv*}DKDOk);?+m<}?HEE0;a)j?nTQV^R=FH7Hbu^JEIyqaFNYm#}s0-Ku zQg}6V{l?P0=OjKqiHk7xXXMp>P5w;K8{7lIfSUB?>#eq;*Ii8sKUecV86Ec!K-YL; zSa^|AztY^R-u~4IH+Sqht5zR(&_$Dp{LPPAOASspxt|fNXzL8LOwCGZK^Bxab{evK zEzq?hUQq5Vhv@hHW0kA4W;Fv-6u*<9e|wv2ix(g|9o&E7N}g=nleh-HZj4|vF~`;| z6|PTYhQkM(7D!`HAyen70!O)vXfGXL3G?fSM#<`st$iYV(ctQL#=#uVXfwqrMisVw zi#}pMr@)W5@%zkTox@~YykZf=xQz}v5@tBB)_F293^e%+V?V!QPtIjig_cH$^HSc- zRB0JIy^|gU)b&Tm$Ydnj=WHg}I`C`X#@Vi57LHid2$tov_xJlp@>7eA5#tWNx2VZp zLt_Q76tchLp4}Yt5|7qMmCv$ze^x=C>mes~l}Fttw^UruA?xh+3FrAYK!G=yu=_(W zT5ri@7b4{$ndL8=ms*7kx;bBP##N{P@6m0Fs#QXrHi<{zx(h`Ww55{k#a_NN6YXNdB6Jc!>0dBMveZ8eA}VKCtkR@<0v|gNDLfgi`npIm@zs&;C@D8PNhJvaKz`L7L(%^Vqx7G}#LTf04AD*#=0 z(Vb0QEij;-pGkaSETOK*@7JG1NZO?l#9m5@40?M6A8)YB3cUc!zcVLwd|EVJV)lAQV27g5JcXs;2g4m0ZXAyPa=nRA@}L(Hg3d* z{IOpmINL<&TP*S0)lGZW-t|?+_L9Z*i;9&h&=H`=9lC2xb!try#Z9lp9a`WVPOwU5 z5_-GEW4LK-bgFE?l((ManKdGInn`5cG`e6q$E;$foe3-&JnT0*(Mtx?sQ}|4?8#-~ zTTwehx9F!zqVgl)Ct+arm_8~Spv){0(=U5Iocf}!Z&B$b(6I5f=rpRVopaX*b%|FQ0+l;R-0n( zasn>wB9Zq{L5nFt7I)xn1T!+3JTDu+BR^&GqnJ%q8X3Keyg8X1p_yEo><>41JBsLc z5XhM?nIB4F!~Z_EiL5Otl8qVg@6rKXR;n1t{LWBVA#j%H{YX;kiv`)e8RfobhRQoW zoOeJ6R9tvrc6AjN!d3R&k#*^_D02yoeo55taGtoBK62$wa z#jl!_wd?oc6}`?FoV89MBAA|nl?`x^2sw?0jF%F}$Y8IsYyl>PAHxa11}RH z!NItXuHt{HvgZ@2G1{yfqZeAK@0r zMY7l=7LVj+?9faT^Gv=WGHVgW!)Qjs2vOhpLMp~r(ZR*Yg9X9BS5(%(D>3sQ`l&bM zDNe|gdl_K02TkOr0?cqs7$cT!DdfjJxXG{VPFTRmFUP)FjmMm~LIYp!TxO6?Y^Vgn ztzC3jm82F^INs&^GFL>eRXD}RLC!#QHCQa`3Cael8Y8+%Fs$=>MC@>%tV$efbwd=*@8)-T-E#q ziBouoK-Jw}bbdlHI`2vx_v-F{3V93*vV&N!HtH#hYLi$6W#&z5#RK!j1fW=KSVblG zKX~i3gp$$WE$f19empwWF<+_(@*RY&L8V9OmQxyaO7J@EZeqxBDh`zM%{l?3DuqYX zgYj1u8@+JcO-aOE6HtbuRfaj6q(E-H#I3!-5wX~?4$k<*SMuJr0DQpB17+7UCt5&GH;7}d+Aa~;7kWLm=D6&2_@+K5YQ2H(Gk@{ zAU;hHozqF>=8rrlXKXEyj$>@$)s-C}kh$3@x}<9tMf;JNM*zdNJcp`+BO(8yvjIeA z;~TkuluX(w)>GNmf$rNa++5SqO_$u1TjYuCN&pOZ);OmV@lyK*fAedaE$Jie>UF&6wJs$Mt?6Bq z80a8lk>ATN=@2^K{M13>(S;)B0^ZJmoXEgT1ar?m3OyA39KZcy zyFA1w?;#nea_}^ak{&5EnN@JB{xu~Gt$6swkh$KjV83#-06K({6iMnd!W}x|jU^+# zCAu3MHQUoa$kA)PQZmNY5{f=HNclGOJjOQg(}LlEYap1wcC3`t*bKK<9!j;{LLEqz z$;>#MAvuin=|^|q>>2AjMdWWy z=}z_)T>`awh!b#M|99IH5`Y#^@c;6sqFoW+i>Mb31)zYjQ{j#D&=^v#|C&x=Anb|k zZ%AQb%o_RsZ93&qHk>1bwlXu-QlY~5`rqjkQE7THEJep^Qr%n&Nv&W%PB}4dp+-3b z@?hPdJJ4X-7m3EvZn|#r(D}=P6@tCe@mgM-$ZXcM@8iSe{&LQf^S8c_SHC=9%5;XE zaE}d~fYn;l);I=<@e7Xcd!w1A$wa%(W~M_0h+ktW{+=6Mm?E|&U-+=JCm+6!=U=XI zSWTp@Ht$?5q}$F7zDX(9rD&9HkHE1w{1Ql~v^w#7*?;S~{o#D=6NBn{@9#nH0a*+L z5BHxtqmG?LTHM=>*7ujm4JZA$J{A6-Jmf#E(smAbBTZR*(z;atB-D5M@g{ON#O@^{ z>jR4Sp{YR;l1-6sC0M~UoEREK>SUpzNaK|HP-19ERV|_`9oV+^ErMG-(;^(5opm$n z2N!Febu<)~Dm6&=&cwQK=-I-AoV^XKF-3GV8$i3kx*v-9l?CW8`;elS04ZJT}Y~P?xaw9L1d>D{TUI(Jq75^;j`C-g-46c_r=!a8EpMB?0j16y}^<2l_&J_)nPSYM)S>FJH^kI zb&WRyT+Uq~%{})EW-fNTItVIV8b42T45NxVywWnr=CR}VZB%(MAY!Nuv7Pa4gI*~k zCMes1%c+W{^l0B^?8>2_gupY)jPoNJr&r0o=y|kIOL_Zd8wp!Jq`p(IqaW=J%rSFf z$9C2ex1QWSEKP??olAW0bsnUmxqZYYjj@`P&qjN!Qg?)QDR(r1e{bZlUcmI}3 z$My6S7;{>n~ zv`sXR+N9#EuhhJoQa(e*S=y|x7|93Wg{UDz2|9TV`U3*M=lT-n-ZhJWhUxs`#)qtg zMp;5ocp)mm5FW`*?Ca7>n%z`es`*Q_@LK)2C*^~FR<0l33pE%Lh<}Zy00uHbP7D4R z5|OsUT7&N#_|c6F17W=~vcL7W%ky;;36W(Yl;Y^<64i&4jb}gp0>3q_zD@k>$rF+s zbAXQqc^%R?0+&Y(r@84Iw%e^U(^Z*6+QXzgsABOep0V&lO%Zb6B>+aRp^KFcO;`bB zD1K{h0j`!Lsp^m8HRFo+Mio0cuMQIkR8jxKOloHt3WrDPlI_doLg!tJC6}X}$En)f zZZrJ$=%M2widK!XAPrGGUTsV`3=+CJIznjo6140LlCIJQMfjZ=` z^ejN^WTlqEt678ofj<)IRf(|3s|eibEVAUCqNIz55EJy`i>4^i;ANXi`F?;6M4SK7RFM%>{i3*<~9ZqjSc|bS82queg%l!VSx+=~J9ZP$F z45W6mlnQ-G5t`L=|K#g=E-=9e$#uAi#EhV_FHP;55z?m9(8{g(^RJcmdwrYT)=vGS zT+&DUoVfL)UhM3JK}e|paaQ&Us%D1j#uA+YS8WoW#R9FpK&6BD*Y$FnB%2Rq2`EYD zjX+UO!iR(E9VL@j#P*EsbdQXF;;Qboq|C>NFIr&R-;9=_M>-uVHo`P#)GC5-YIA~% z&Qf<<>xTWAo=d@I@7rkCr${cL$C2)}K6PsfYaDDZAKq_mioVf|CX7iU9YRyDzD+x3 zT)U6;T}!_4k$!$FIWN!aCAPYo;K5z-u$scJ^j52NBz36tV(?EN;oGIKooi2SJK#sX zaH(Rn4wyMM)gBX6(1=J0Gkk=Q21c%K)2xwbPE^Xv7h{ZGNJ4sS%}t_g-YAEzxu!P^ zI$n!4j%y;sF(cHmmxo4@?9*FDG9n=E$vEYQL0gtnYH!_BW9fI99N$c#jPR$la0aK< zGf$b}wPa4H9`k;8VD9f)8?65m%irUN`261HN6v}SaVev|qi^|L&i9g&*J0}{UK;TQ zFPta*ruyY(?TW>uA5QxEtz^luJyORc?Uhp2XKKk?^SKs!y6o*8E(9iJaTaVuQ@WNs zgkfPx}SbSk@0 zN++g2oG0Y`+CkG7(!|C)-Z)eIV2a50`3l413Q^C2a4Jq#EMpAZa|y0sfJLWpjDIjb zB^{1LV_Dy!Ym~D|Y=le@PQcZIW2@J$Kf#|5v88`%TLQ@kD5c z|8mBqbF!|M#=&vGqz|6iXy!J9%eLWh8AAN4Y;*nIzcyIvy>vTKr>CW1Qn9UQ>&?Xw}V-e=o-WkX$7F zvgR@BY}qzhboKXl;ls4pE(;vts37y@iYi>%R*pe`!k*lt{rpGYDtBf%Wrjz~o?+jJGJ6WIsa*fuf=~Is{x3AGv%&u2zHVnrtj9 zLX31}%87=loq{A!sZz(C`cp!+SOFYsp$e70-5n87 zaD>GvNaNW@OfT{o43NU~(Tt36AS0ALjWn4IHM0vXC5fV$4|l7G;&Y4qa2h2%6shwZ z*2fI*9_H}sgkS1-!R=WIt)AhOzF`oCXc7W@UCi+GQv#1pU+?D#x(ZOvnnDa2yyTQ1 zcrB!&lfVOsE6W#NGf7~)23MSfSCYii?FDwO5mec^^!P@DF<|sa1Y~+~!@h9?dIZXu zQN@#S)0yE@WTa|j|9$Zk@DF_!0^I*EeX17@gklo1b3XH!%ZA{i$r*{d(*Nh%D7_yv zatSCwOhigahcdCW(Es)}N^H?YCI{1vyOt%GMJ*e78;P2N5GGTC1Y#buORfBO@$~B! zeOvW(m~abrW3*R|Zk<^ly%S8Q_}{bQ)zoX_)rRqZ_s1WsXEUUGe1A{XvFN%O_50vs ztNvteYOqqzhOcJ1#Cmf)JPF=yjc~=JTObrE#AP04pXB zzaGN4a=fKBS}eaMQ+O?^yB?|;R47X=%HzR8I~X&baPvDDFjz|eLscc_K`9k(WYCP>|#B65cwru z-pW<47?Z|NeM7k1x%rjVQpO6_y*6RbgWLj5`zh-@kJ}@rf@r9NSr)f~{&5kVMRpay#UaMLEcHC{JP%o^qN4Ic zHew?$Gz2ugU-+V`pIH)vcXCmJ{KCwkn(XddQ9a2l#!x#VlV%aiS?zo5Q&^96QCTKT znUXt2lXLzF`4DR>KhxoYs4!x|;j|9-YWlS2I>F&ub*6(HWkc0&HAPJrX=6%1JQ#IezoT_uG zp1=F~v8(p3b*=ke_j=9CzX~rXZNX_o-;Yn9_4tJ}ofkJ>8`AbaW(3R+8LrAu1$PUs z;q!6bks9s-LE=%tecsPU4)U3JTQ z6R_%qrc6ZWuo*rfc}6NrBzL(Dgoan`R1;>q z?|PWD1TPnseh|}pi%%+^xym9vd*ChfK~VlGU(PS|HSt~lL6@uG{k(c?<8K6}NDJ=_ zoxQscFD8r^eSx`JsVk%G*mF4-cF6KKfCT*aVD}GavV_c5OLh2}X$fn@s0@WWRpiwo z3G*$bu>$&{vp=Bz+9CeZ{)8QJ%$DN(;$8TB)5xS|NKD2Tc@JZ^Ory>fwxUIrA6l5C z0P{t^=RXeQ;jbjtz+853DpPI9Dk~B z8(xG_Ma`nD4iq2d)f%UBXr8LEtul|)aX=y}e5(mDFmMc3Y^ke3asueVVxQ}6TOhm5 znTLFn>=hA%mK;9Mx_-I2Mn>c=vpwO|X1Y2+CK*pYrsK9?I4(Nzg~AE#6PQtp5Tl*M zyjlf7a7fDvy+vOLsHMS(PHA=KE8czy4=VSo6?WL?Pc(xgCioR+%@*iVKv-fAfirQp zos|IhT7>-7RB*pw{?1B*-~kT4!Kgw6)|3X9f8wagSSG3qU*B1QbKXp0tePSkc0njZ zm;eOgRMWWr0yaUlGTIf{OKME6uGEArKoQAT6YM?wN#1TU)K$=#@wlM56G0Y+z^f+r z{9u`vj~%+YXKGEZCqrb>JJ)f)@hRltN@i;=Ix?7l=YMkA@ZA5<0WfoHjSCh*pBMyo zer9;u)>WfSVjv(6S~YjTcr{@B%>&_5OsjAi`1NM26|O$h81ihZpnzT267u{**z$XE zPAuoBBNCUTjisj14WDNzlud{ijToPhwJI-#}DZfaDITAT`qX0G`p2R8=Ig zfZ`TDQ4N8mYCAw_XiEn@Jm&a&cGcmj*^{znd{ZSAsu|XvL71HzCWHokQ41rdM#0OU zDrMd|*u=XvPwB6+X?O*6aqQzenpXIH#L}_E0JYX>eJ1=E27BXVkFVnRjld1d=9S#g z$HxySlU)Th8OJ1hc6&3QG{k5;?NQ3!r}&1ClYi6}?iuo@W7JqY>bsq=qLcW*nCkfj zd2RaE*O_4#N_m(BPEv{W3yi}bWpW3e$Q9sg9TV#*dRK8RIOQqWdQVnN<(Jvfsm{yr zD^;`oT_7!})8+cTh{$EMgcgY^{x$kI%qPjFezPBMNlep<9n+%U6e3OS>K_nEHYTWH zq9bA-06*K#FdG;zSa?WjZKrMXeG!sXED3MTXm(Rg4t?>FT*v2*`G;SX394=MLxHcp zs-f?IHjE*@cO>P+`}^|Pi~{#_?J z^05Fuh5uy{2d|Nwdu~W#uxa9|p#-Y`ErVV?RnB6biWKr)TXfc~{Zxh^N+a!`rT-Xt zlg2vMe_8gXN-H%N|KBOr=$^EP%zM=)!(#KHh&L)dFEy*29&5~-$Q!>@Ic!hwIvv-p z3_K3MU6=50`QG!!=KzR}i@jGOP`ctO(s7;ClQ*Up;9vq*x6fL`zX%5_Lk2=&dyK7Y|b zg2vwp6^u%}FS#E@0bzj&%!*7-~NqVE1Bp0cJt!u z5(u}>vl!mgr*nB=VV|}6EmT_`NpN~)<-ZKewv202@1u^w%Gj@nwsoeo45f5nw+Uk@ z4B863>!+v)I;6Z?{OgBZokq|%4*Fcs z?qOZ2P>DYjO~|Rlu5@*&tI3=ESy#)Fr6%4wEG)MJo<1pdZ|5vb99R{%afJo1pt)v5 zNY&#mrZ?1eh->lw#d+LGu1%*K+X!f3iLaElmrGhn#vfBbJ4K@$ILijDL&|~@=}4Q} z>J*wdr3;6T-eE5_4opYRYizh- zU8}F`X68J3c?`U>YIR?~|0eUPw6Rs{itmK48CU4I3^c`f=+^b=sHwg(#izI)0VT34 z82Y-kQyRX^BAwL7l6&knf;TbaCEG%_(vSoRY3%d`GD}0>0{vc}H9iEScnnUnQhM}c zv37aSBni-5P3#4*d&Q9YuK7hVlS#}j(8a#->^5+fICaOppYid<(iDR(@@dj`+cAFn zeYoyX<@D=YEy3fCuXOtp?x48`9yzCP-{sqXZeEW!w{KHNrn=AH*mJi7jKR5HE4Fw) z?sh`E*L%Y_T3?&RC3Ay%0bWd|P0nEyBAXwDhL<~~qq&|3t-#X*@4{6d2}LJ;->;t| z3RS=CFOe9UZ6ScUc0?lx$ETMISp6owt7`9Zd(>G5KKC58tNiFY9`Q)Oi57Pa@=X;k zy1nu7bS3Ke`%*KxYYf7}^CxCnB5#N8=esMTwBY;^4S@4*J=R*fZFy+&kQ_E|EPf5k zrvl4cjjDVZn0?SQG4&HWL#Q_YAzVFEIjiSa%gw3^_X9fV=o>5a4~OUT1PE>5lP!Tl z2v#spMoi6P+k)hH+b$oG*b~`XW_~zjQosiWC2rDswCh}I1hCDf%!hS)r{)B0byKj@6$=@-AkC6g&ZC{n29Ep2|PcGGU_It=Sg4{8)pB4#LnS@dhh4uWDJ zr45m-&8N)Izp7buwWDI#C*zic6Kx=5o8K1Yu@k0;SFRuy!ciZ0*2X*ItWQWVq6QyqlPE- zKb0|GK+_iq#%i8Sv2V{!DUAGbm`g~r33Vi5YS`N_B%;EfdLQ`xiIk|$5%J`7*R~wm3Qqyvj$e+>%hZ7qzpG@gKcb4 zT)i0A2X4QIHWkogMh8dFUPVbBg7KK1&TUY@?sj?MeaUFNYu&(#aChTl3G~8Shd5CQ zYq7u^lUIB1ZKc460I~%9AFP${?&N`RU@^2ndZl$j|NG}@W9a@kbQoQ-X3!qjGRQ1D z5VbNcTsmVZ2Y{(+Owo^@Kau)i)`H*3km&{*Gf=U^TyaW(Qq z?u+>}RbDC*eeISACq@Q31)OAX-D}Zg@i&(?mPsSNxv9P6J#CK>JyymdgmAZ58l8w> z0YG_bf#j&`E|>EKU(c8yBOmJr`;m;ia1V#~YM-$iCWN7L-)l|3rmQF^1#NShtb+-@ zl4AtNzSP-Es4O^g=5di+Q~T~QPUZAKIl;d8f{-7%Rk!Yjtj$01%*WC*VqrK@OEg2L zL{<-tu)Olid3heKMHvM=KZ;3+Q7cf0>v+x+@ZmO7tHj5yyTzfqiAv#1Yr)m8Kwc%I ztQE(2H?4jb81i`oael(wNYDqK`?()uzh*rpOqOo7&N&J1$4Orz7@@Teyr({q`y}JR zxZCotZKBk`_p*npI*UOg7^*dVeY>w3Dcp2Jk|ayo-b1q_D9Ile73v?gz{(tp!BIJc zaL~pBGScILH9_9)BVEypFbtq~G$uxjG>F9ApN5fLO^QJ=JR4TchwH5A~5#u8MNXm4m)riUBgj#WF+rDRnZkV-ecXG?Yw}_EEO;QMe-R zMnUll+kwjS@eM=qT8Xr#D^a$6@xSyE%tdvan4v%)JJ!13v1OveI%k$kX#F0E5}!Q& z0$9sB}i52yc|YY8bCF= zTu84h5?9zy6e4K76B)Ge7UZ;xNeoGLwbFX!-i@<0i&#y6_WbC8a$#roqlATf(=KD6 zdMHy+?EZFOQ4=}ptMWvnd5eN&7qid2Sk_6;yya{26KAgyKK3OY)wx(+sT(tg+}0uD zkjNp*(2@Kn7P1k8$nXTEAQOAkJ95YzoZuHFLE(lF7$dbdo|bMT>0sRcRxlt|JI28q z59oz!=!IeE1=jS!J)4D+IlU+s%{c8&)pGP*CI)LP`+bRe!31MnW=oOvR&v-Bx;e+y z*Z_d708hn07E!E}+0=ugK@6)zOmnhVk7#y$CW1dpS=)TZ!(_65FLc}eeRy(Hs@%t%=+_*%>RC(zyrg72Gash4Z2#FrdRo~KzIkxMHS!zvOtd&IozLftS8tjP(eWH!a+3t$vIKRInfRpXD!DW`V8rb zS$uCC6yZ&*IhaXw4k6?KJfaG)%RyLEc|_!)JN0f>`Fzv;1oHrZXocge9FS)X0@a^B zvtmzJ;SJUe29aBVv4SyHo%1#epv9~10{7`{nWck0rM{mn&kFEEdQ6c} z$`B<|_+23*2ebq71ZY*MJp~}}!Ud`11vG$K`P|;XlQr7BTEm3}UO z?r7Q0&k|a?5bZodaaQB;w+Gpn+H>(;YW689*rL(e#sGCrfJRh-h87?<=EXCJNY0T_ zw6Vco1knW+zUUm`0v7%y1m|Q>-6+Gm^dNujfS=Jt)d5EJ?i`l@o!81&(|eBltshXk z0nN^L_;_C+EKZqL$#ana(njNCqW^go|D@aYheS2i3-`rNsVnQ-hGa^CB`pfWQa< z{vZgKPfn~mXBO2)K9vfW{Wo~@$40x9GEtVly^&5D&V)yqH}TmyZk^W4-VIz+hQMgH z(4EIyBhTQ_vj9$nNlGh_yM#TgOv#X5MH};q%Cx-=h=w&Ng$)@v9uUI$)XQH1zmAsdv|&C&``b4iCp1zQ>`4RH*&w#ee&Ce z+khg_$O6tx$@aS-hB}G!nBu@zCUq2JcAf8EZ|w~2Zl z$1sw;zgbCJ=Qr_&OKV2ZD+1XhxmxX8$}Guj@8!6T7{jQ;s?z4mCJ0hdsy#H6hbUb3vBU+4&v5xAbe* zf!iB{h9b=dWUrU2AmD^bw)dT!N;m-x#kH3Mx!X{Qao42CP~^p6Mqk;7_d^J5F1bVA z!`$8swqfbbq2EnqoF^1yVM79cx_({_M_uIEHR%gKrEmyvV0}H<+&#jAdeto&B=hyn zvy`E=`k}9(qX@;(j^j}~YX_E-5zqH7I&Nf7{|qJNjv2s4lut%pnP^+Q9u{>AZ#5o& ziOm2xK*qmyS2>6qHpcOCdsv?M{u5EvPBO;qkKo-h}7Vm1fi4CsV0vL~qVR z8fT`P?x$O5-nWHK*S~}|OgebxysyfCZ`2n*@VcV=4{XV=>QHHBoMw0^Om7b7wN>^^ z!B^s^720zrul9Flps@emWC8_L14{p^6M-4Y@V`0{|KEqz|MwuLf9*j3Gsp?+OkGgQFVFbIsI{ znW2GOkxX?zY?5Y*u~!kvaO#wfth!FzEsLoYKZtSpo5fn$iuK5@6@%Bo!BOokp4*h% zH(apkB}FAnQ98yiNK0FC+u>nN*|_dXx_@? z+G!o!vs;moUSS-}BEeunnAOHib4L%Vm~OwF^=R1YmbkcDQEb7Qw0-$_CjxEex^8S5 z>3#8(zX`E>#av3gu3Fg(eU|kx z?&9eH)zTNsdd~f0&v$>0j)?}vAn-P_O!+ED`rpRx|X2K;KTUrI5x2_Zs9loo!ti!f|x(Z&C~ok zajWFg`-tuEN3$Uex5YS@j>7!*{E=t0NUi?UQjj92l(`nVz%*5?r6>Q}n&-+fs~+l& zA<8X>E!OO`Zzs#Vme`@)11>!~!Q)(fKEV6>H6}2C&=~bSw3f z#e<5;SgUr1Eh;co4`u0jN_)z}oCd|z@Cd!mXhNea&R~EDY?IFg z9-voFJH-qjS@&W;CN_d&cRcT&i-$9wXpH#eRG17-X7I$=NirW!;yz@sPm>^6MH-Y_ zB+t@M)SPr*?VzuQ{_(x&bE=yICwS7U71`l}Uk zmqkgIfQ-0yHRV8~w?M*1=F%Bt8!n^r_}E1n$2@TrE9!$Au3TJDbz%Pzt0c?cOa`$< z3&D%}saVnuh5|iwPgISdElNikSH(dD&-k&){z;k}@0Om#!jev9X*cU&eL@wx_eQQC>5F9XeU{3p1Oa$wy<5d{j}$1~rVK zwBP=T%9bV^7YOlLx*P{M9nIe5q!beHT;cLPL+)<&@@HJKZEePOl0Vsic+Th@+QtiwqEqL1>Ho zF|FfUr+RPZpl8~#gGcmNm?8jf+kyrsn3IK|U#9(P%m2adnz^D&N*7m6r~Pf~L99#S z1(whb+STsx35wK1G0T>z)P!i5+T$^v%sti~*%mOil;|v=f%6AXLJtpVw^&8@1P2cTHJ^KruyaZ_K6X2sZHREv6_Gd)`Ac9UYDQdbM3RxS}8FR7b?jy&m>m{ zeiQ7mEp3)|p)PzzkX-_w^XtS{(KCbjnCEV)`4r69$-!+>Bj*I2yHu=|E&J(d)jsqT z*0y!RfkMV{z^;yYgi__;(0;6)wphmY4-`CmrGv%8E{uX<9d8E$b zJn609><5-cL=cEuXagTt%1cI9ErD;8Y>0kG z+c{st^(`aKoQc%$DsS*Z(6yY*f3{qYD2l|h0Wvw;drT^D=@ZFvy?l@7^nAVQVp%(0 zGjGYnd|W6~zdbJLoD#=`%*P&o{(Cx-cY2oA8pN|IA;XylC3#f;8`P-(Y)rW~-{0$p zTy(&SBF$0V6UKljMS&<9>B#pKf!p$YU+T<@6Z4RP82qGH<>GOem+laxUP@s|UUsh+gdFi%xIc9!4 zpckH@7eJ*i4$BH8Mj0T%4tSLi)o$=iqKBHLf=bOBOy~t5+K5*%0ui=>7}%lslg}}U zqd{BG*B!-jY@)aKV;C=R%i7`;trArZ02(VkElbb9`MfpTQq`D|yhHISP5>e+fNFgl zcmp8%I0-fONdP29y7xuSFGWhPgz|1iQ}bO=>;ea0;rtFy&j6HUwP6d5dBFvT8}AWF)RAH2(j1-3d;)M=|$uT2N+uY za}NM55(An`ZiM0%|2EDaE#)<8}8T`zN6-vr!TCq4w7i`73UZd10It z(2^j4;syY8jtlfkS5-jgOm#imlV$E&V@{z28v+j!OdG$0kRt7x?I zWOy$qkvvG+hmzQDC5~Vg^d|@mZc8CRfVAo(IO_e+428LVhPI9opCOX5yl@#Fjj}BJ z(xDaT{{6I1o!XvCQbe-Z@9H>sDxkRDc%@zdg>!(WGk{}bSNyksJPh@h(==po3Vv?>B?^J#2{p%r5ma1hJE(BJ4F zfTJR5Q_y){HXG)feJM%~q6$VVB?Sg>;~K$&m?8}zVpvSk`3Z{G2<9_csq$d1a%b~v z0GS;riSIrP6wG~XMFhAk1cQn?sEQbVWf)24e_srJ6)dROU-W9V=rts)bjUckBeXs+ zWHPw;J)?KTa&cufRm7r!c#N-=KGe5B0IpXuxlcaP?-NuPZMB>gjn9ErYgdeb1X;lT zdsP4q;3c5;|1MQxKT7}q00c3WnW@hGkEQV-DlvtBMFUS;V*}?n{{aN1g11ylxGhGX8k=gA z09;rN_WbIRKzLIQjq#w}`+QzYy2){GHF!q&>hV&mpYwd3dev#~D?`b-w{SePfPT+N z1kkHByWXH&GVyi9S zr*eLKW{ypf=k?#8%Wp(JUcKaNm$F>bUuJrpK!()`pXq+CSS-*MOY$i3nQ^Uf1 z0Zw+B03}>X(+sW)r>{k4)_31xalBV&P2E0I6vQ`h8*dk^$M-Jf$!XsehWOYtu=wTd zLUi(-eu_9$Qve%lOVSc{9EuGCrfF*P?K$bCKl4u&W>P<_re_7A8l968Ss#xMD>0*w zIOWkA;;iMIIw6P6%4SH2LT3Us=YmU5Hj46KGa`B#_nd*ZGz0MhV^{%&p zN%8?|!Kc}@5`T3X+B@Sj)H6om zY09|#REG-N-(Z#<^Wx+e;*)zNl%p6&50|Y5?b9`<#AV(EP-PnJL_Ehczgz$XK>EJ` znw4}*i_)88u{GoH#V?uq$`>IVKjlK*o$H!mK$j|WtuL5!+`V7?zl6cK2JatbpN#(n zoO(~hnI>xS0^4&%rQgyl1~p3wkrz!we>d#s&2xOww+T`&p&n{$G-4THXN|kv0t!WR z9I;a{Z&pCIC?rNYh{(ZIux&fd)=i!1;|knO*4JA|ig!%`)tM%Z-nYQD*S%khwY8$v z27h{#64rNo1^G3rkFm-4hM|2VBPp|JW{8siCBY1f-vh`d2c*c|;M%;W*_suW3z;Sk-j%y{ zXAA#}+`>^`ROb$X;XI@$(7~cVPU=1LA30FTtV3m0Da0FKiU|`p+`v(PGNG za?4}Wv=Bk!$8&71p0zQ%*Zz&HnHv_#S?@^<;`83paQ==`hH{-MQ6pEzMJE|FxD~8v zjN%9rLGv@^m5@TmuW7OcJ|D#VrzKais!%h|Gq#UF%r0b2qc}$JQW6hysQ^_fmHF*} zfb=MRA*DPt4{!J$tm|PvYRH}^IQ^vgbLiFkxvq#1VwBG1fXm#s&GOEy?_FsOEb>Ri zqQmCd2;su>4}|+bb~LHZ=vyLx=fK7ROML5=`j`4A1n&|Vc~(*O*S0wB2KAL^9~)Zb zR%rttnARx_C@mR>c>OelsvHVxHA&ub=`*jTj}zw)a*CCPM90xL-7-DDg=qTPxe-12 z(r3+=&?et_$1xYArqp(rJ2B}`?*pTW>AlNTJGlFlyYeSpH-#;DK*)h@~ zZvXP&$9t<5+;1;_R@9-lW{1PU zvRL-wszLpOQ@PpwQd?U0%fM|SnPrpNUc#5n_S{Y}*+9`w~9lK+<-u%MLCp-QW3>R*QWAU z>nCq-n+yp)MT`=O7?2)yZ95im;2@UTA1!*vjhjzvw4SeE!M^R&q)?l_dA{naaQnfU zV20ErqYt5oGZgsX(qCY=3UwBq+x_7cm+vPA`SzAR;a zNTW^%pPZtS<0K7!rFtHZ==W#pp6R;wGZO(-qE9&dA5o;fm%ng-Y3yN^i7%zL_{Q^^ z!_}oG7m~BksLu5*$4@~!yKfNb`y$!2#k*igchfJrU!RuDPDfp=xLq1L+h~&%2>U1Y#3W(40Qq^P_+x=Emr6EgHbsFf1rrG; zn5Za;iZs8>t&^-0|CM1}dV&?_GiPaJ%yZE=&4ai-0bV9+pU41Wz%0}_DBf5eO7x8P z0+v8Fq_QrOfGy2s-$yOim!Op&qaX^6g(Vh5Q_1EFdm`hM^8MWU5`%)sb}0nRl$^2n zpRveT(Dixg)W@oHL7gc1UOe6g-}UaWtD7e-f{? z8;vMgxix8+wG7WT@iR=npI!=!srH}-t89&aK@ zKN@_FTXzm%2nMLl0zlDhuP5CGDNz&bDNf47(c|9e{1lBol2i6<>{&po7ey;PPiYF& zTnqa=87e-X1%U~xc;i8g0fY!@+SMFO=g7MJH1_s1-uhImi!87=o+e*9i5F2TUuJNB zc34kz-d^_aTxh5<487`#FGA6CL8#ad0(ucD-sRX*@sV;wkx`L+C(j8nioNl|(6GXg zhYGwK1YvuDz*-+o#QTq=r;fPvOm7IK2yNTBb>*hz{#ms7sj~8c^sK)oHx)(R*$=x?Hc>~F zc}=3KInemtR@|}WYrWLnsDN)+41&@r3&AmL1zF<3snJTX{o8WsLV-KcIJdb(6_qe~ z7Z_;>YG{&7N;G*uEJw}AnW~WO{|rxi326Q=lEH@IfCSK+!wJ~5#By-h|F*Lt`3KfL z310{ER2HjlrP*Eke@G@kjd8Z@KT^kxu~+uYzfuQ_lO{qt`ufk3?zG^cPTfPf3b$CM z)l!8~r^_3q1&4Pgs>jj#;s$Oj9kze(R6_ilzVv#5ofddRr~?@RvRx zi}dkrj>LyjX_$F7xAsIbsg4_+Rj&`LKpY91sD3TPpxNmEu!(ayV<;(c>VIvecU_LP z?v3~x@}sqFT!Ihn(ky=-p<>E*o5{0m545Hm{tx z;nOyEvl0b;5X=nR&|mo;|M-E-B#vsA+&a2xjO`2ZRLFo{y8rL+rZlIIQGsXC-GpUI zaI;gDB6wb8Jv6SpCqI2j$RLLl+#R%?(QUMs9-QpHSM4Zl+_jlO-aItW%=D=Es6E$z^Ag z@dGA(PxhIsuX7ChT+g2GdYW(V6rN^fCOF$vS%O~{-_L8CYFiolkvITrawU^z-lhonh4@h0gGxurSOR zcM?TT8u0v7(_xQiU|M)xxH97ErshE2%?{xC#tR#?8k4H!SNJ0zE9U&W0hY`LqIq zAmn5?wkN9}UWz1bf?K#JF!2Z!IaEwZeI+H#b|u9y<%|>cHSGH@0s!9pz!>u#KZjHM|l_Yg5d^7Zrt5Hp-iDxK&H$U8p7+qFij8 zJjeLqaOk;7oc9lj<9sj+Qr&#_o@92Nz|-0jy}}bDzBIeLr(QIBNjqjAwEyIJU+5^Z7i}ugY?$=!3PEweSD1VoZgQM(pNbeM1Yk?G?t4@9! zKvav|)1mHysY zbpXe%@cD_k>wA-jZ2uZIN%t3cJH}8y8ejdfuXoO&wM!O0eUJR7tcLxcK9>5dOd5Yq zhH|W{dkDSST(?7pnSQAs%a-nf*qJX3e2Wqg%qsffHtpueJf{GA{h`=`FGFI{nJ%yf zJH~w$C@B0bm)-+@YKJlv`%H7+n-zH7`>y}Xcf7}mpNN?omJCx<0AcXgN1ElgTs-=F zf5y|^Tef7x(JO8Y*WP%uDH_G?2hW{9{PVICY7>)xZ)@h>b|)4*kM<~%*E;Jf{6u?R zV*0As>JydBx4)jj%MD+_o|FLZQe7afT!ffn|K!{UryxhWs&6d@(*z$l>VCgj(2l(8 zL?oXz{!;6ICKULYjQ%!wAZ%T1p?){)t$M%2@5QX2A&PDml(7dVoZ`!2%jw*g24>Yh z^_r&&`6Yhn#ISu#$nY0^UeZBV^N(qdKk4UQA4R^E-`t#=w`7@Tg6=TYZ5vFZ5thd)WJ&!yeGndSj+zMZl` za<_TUXi60hFCtMFzT5X0ESysSc2Znu6o?%_QV!DP17nMV1sEXU6_@}EK$}?&Mk!Cz z4hC9B7neuVu&R->Ltzd!lGad|5|q`zn(a1<|ALQmlB2~6`Ug8=D3^O6&|!I=eU;Q7oTZ+bE95nr|rAM~|JVdMHAv zFHR^Z{(?N5?!t{`E>d|f+=)HTBrksABKB}RA!`U~QRgF$m1rdWO#$4=MW?ohZESU5sH%QFAMZUj5 znYJS<_feGx#OQo(bp@$ez*79+5*35pyu{!za z+}lV`hNl-4v3HKhZpw60;|$t%UP|W-&-PTUTXg(EhVelLBNYXHLq_<}Kfh-*35v!L zhEDwj%MBpIF-GEwSrYYRay>#QKn*cam(P8q@W{o{l5OMnaD^Ype<^Dqic91Lq zc83>K=}bekh8qCvb3piEuKWf-l`jW?f-#~(ogj&==Mc|eQ0ieWAv_OLJF}`C^<1%_ zwJ*=Z`Clo@^Dq~@0kDN+#VBNG1!I&30|=q{^5+E}Vi@wy`MDFWqz!ouRE7UjbZQ6& zsPN@sP!+upbB@x_s#D6XcLrdcW8Iu%0lmNq=LNVMJo$%No2&pGiuAATMWuH+t><_H zcLncelna^aJV+ZD>x2IvT?t+4DOR#0=1A6%=ehc?mIOQpMcn0X7i4xcB<(tvlJ@6& zuL9IJN)5FD)2jeEE4<+;Mt`y7J;kggj-<2lA}lXl-GYjU)lxqWCzx28>mdZ&3PcoD zi8TvQ8;q|OEAqRh*qSu|LX}52Sb(`v3V0Plf0qF&%uky{QzZG18&`Y7DzZ9qoUW4L zt5we+)lyXDulTE!7qtB#v678wEzS&nhHBy`HC9h*ainW(Olr9|YiRxtFW0|-Q~lq7 zgMp*h=LMI=Ac82EiT*FZd41c)Bm>97bfa-@TN;c6YBj1&`P9762&IlogZ@y*Y{{Vbc!rm&UY&4 zUeh>zYBG%#AE z+D(pgNLX0(TdaI$)sK1jO5Jb}(-bZ8!Tghu`O0_4?~imu58fW5=3r)sS$O?sCuqSn zGKpB3yXk|_v1Zz;x9$E@e+Oc{AKh$PF25wnmQ2>C9KOU7jvH7=?ROh&!~>Nx$}JH% z%SI0ES3pMJ?4_7hUz9z?6aR(OaYkncs_gZIXZq}PL6->1Qul|Q6$8C@EcPqtQ!?ce zfZCkVf{6?y3Z(bv$38uBS1f_itJd7;i4!oRqFNU}B(N$jB;(FJ9>p+Nn)@~!wGxQF zVqf zcANNd1}8Fx?p2lundGBq8TC(;p0xxv*smmrIu;lDi_^c%&^Xq!NMLQ`Ods{>W9_S+ z=k!o`>s7>Au)2q9+{)2H?G`=qmikC~BDd*ekQBK8RC|u@s=f>R?_uo-$XvoPBH!aE zzxQQyRhe_K50?*EVZG6-jx9Dp*+*JZzW!p==dL4Tmpo2n-Ywcr6}Z&4^UQ zCL8 zHOqM~L|4)gI-Y$vJv5DvbHJPyC}B$KQcvV33k1iHGC={ni4bd03d#XCq4##zzdBxy z`^uw#d?2Dz*t*@$Jdw1B_XwZ^ne=Iq`Ax>?YiiI6a;m(lavPy2p&T73vF;a{eBCUa z0VW_zbjKD=GyxLcMh|&bjeUcNi1z)~D$FcFE1ybV9nj;E^T*}`#;3b^uN@bILZcm6 zjWElyM-A2Y6&C2Pew~mXgy>i447_-lGKY@8#EA>&f-`4Z5p4$rpd@hP72i_)Cz;Bp zNoV<+1aw)6>*PxYHi{Fy+JMqbZjjcQri>l8MBzvEcG_ha>=3W0Bc^r~5Xn$gu3-s9 z0?X+h@iBkWHO6OXh{t!0Nf&`5WNF#sQ<^r#SR;yo^kfm~@(j}zT1~*nnl5P-MFJsh z$a$P~HU^WJ!TS8Bw6puaIud8IAhO2S#f>a~*;tKTPnzI6!J_}!*tM)=ilZ}`+2;rW zG6u(*2*(WDSuumd26DCqAaW64G)2GlQm<~7rAc*0BSaR$jB}CWRlot zXSjG=C{M%Pga==@x$wbO#>*u!^#_>?TJGME*yZMSV4tln-9;yAmJ8Q8Fw^&@KE3gR zpy(^eY4yE5WL-YNx}n-61S+^=D94e@&FJn{uDJf*RSU2)>%!M%N$jRN(Hb#3PChcI zA0@C_7pGl`JmX-A5t#)2AgE(o8$@yuC~*EMK;MDIs0M z)R?OXG>K19fCOrjuK%4guB;$K@j}rwE~KiAO8_cMr@i|ty#)W*fYMZ=eo%AZZEZ9# zP}M_(u7)D#9Jx`r5^;PJ6nZ7=dAO5crTW7wCE4*z+IRJEj@u)1woy&?7g4-N$XGs1`etwL${qH1i17<{FLW-Xrv=Hh1R;?dHLI>G zO`uzX-`kjmz_yi)U@4~VfWT01dR>f`zVz>pQ>8nB!A%otuMrHN zZ^OuPw7S$y8wAvGf%_NC3RG|KQ|uWV?_NrN?yiHy1!V(H$lqT_2@e%+KNF0}{Jo** zXE@ft9#xqd`XyA`e=3~)-2Rd1wroW7)aFZDOr^)iSNo;*{=R~YT|1d(DO1l|zThE3 zfBztdA3i=Kbl8P49^`QMPJU}V^x?-RPr)m>Kj@0QG&)XoUS~*;jzZ`%!NU<+l5=SG zdzs>PhX8`W*Z1G0Ik5S1KkKmjI9yp(tXJeq@lhq*5Oh&y<8wM*pI~0Q-4*G~3gx(kj z3KJ1Pmt+IY0xg?eB+l7c2t0KwzoV`C!lZRl;RzbUZQ{Xf;=!#PjUut4$6zsn+WMrP zd47kBw0lEN`0%eBv{{5qx^qLn2|BlE;qCgGje>Pk56m#T$X_e1tsS)ibxscL+Stq@ zNO{)$q|%zSVM0$qBuETbWf;b!*e}uO@9$jG7!h;hN}{wV<{eTvju3?y@t+&|)bTK2 zMkws{7Vn6-dN)5b74;_L@LBb3W=ZfC}mF}T2I}W1i`uihF2>(;xP^*L^$SkaIaGj zMqtbaIvFQvL`|&+8=YeZiV87;rZHXwm&yGf3|A5bJE0Bd7>X00u2tX*3GEA=MWybi zjc3VXjB#|S>T`_FRULogn{@~Y@kxGMr<#k`NEpR^9mV|z#W{;dUtO^{u1R2uPb`_W z?^%r;?9*S%ReGq6blptC^GWhF--sJ_Bo zUA2{P9K)YblT)X~|06pS{RhT53ozqTGP`+#-#Ui6S)~m*#{YieAMwMVnhC1i zPMBtS8?BG0g^I0ImR{M05rTsKofPA!P5uFvQ5>IfSHWHu5X#Mxf50mnK1T_pRkU_W zQW2sMF~q~|H~Ke*`6cb(54XeijP)Gq?2I2VjwrwXN2rM^oUT2Gm9ltCnftIOuc zx8WSYa4+k$w0Lm)Huydu#p+>D$UGx{d#2bw8E+Rjmol+DC?xmRBp$L9n!n1XM#dTi zZ0amN7e3gF9Mmh_DHhp!q$y;iaR@LCou++9$$Y^uIIP5|+>Zi~M_R~ds8kB!Ni&{Q zWo*rH_1n-4J;6re-&EWel>+eOQp#sjp+cqN?NTtLcg?KUzG!dl;o zco|tP%jL_?34c29URRObWaiX41UayPx#u~mpDH;g6}lKy_v>tBSk$x*#0dSvC(sbF z{*mSaVN3-+%#V6EmE~7arYKq=d2scwLZOxyifWKbCKmJUHQ4Nd)q*D(=d6DPDO6R~ zz8gtK&8MQ;-f$*2N#R9~;y_J`mRpF7`hb|1GD?xGJY)`86;8(6^STn0l1P!}jF%x+ z58kXhe$dlp&4d`LcxxFXYZ;V?Wt0VZXPnoIpfTX5GwjrrH0u(Acyn$QJSMjklaZ|Y ztW4my^{oU%1LOip&L*Zp1o}7iy<$y;3Qfs3;##!vN?Z6-`dVDXY={Z*V*bEpqyY0V zksSgdh-LGa0e@|9%i#rKY96T$OmN#dKn3CN!0g8W@l%qAB- zD*fna){kU9b}-b2ch3A7eC&W8fO-Oi2&S94SKC3FR8J0nM|9E|Y0Yn73s)4O;k?vV zrl8S`3Qq3^amx~ZfX)TdXbEJ zIJWCU&zcc~ZkNmwl;i`3$P5WsDTsd}eBVKCO80qZl|-qB(tQ#$kgn^lF~;W-?%L z+I@Aj6`LhN75qj5%v3%xwm8f2+h??Rarw_8Nd~pX8p(hJ%@p?1cJWf`-r~;Q()Qxg zmqke?&!q#2zzF#A6ma=V^709#Q1ZlbHSoiUYx%FeFkw^5C%A`MT1e_PeWjwZr4cugjOe>2tW{=ZhS+-qx>?|4@4d z$#OWC$73Gf|6O=;-)Ar+I{H}j$ZCi*LheEx%-1%m@5`9{qsLFLF_2J;R zF=L!MuhbU5f*)_OI$iRT-w#jRUwvMXs-u4YW#_iGh$NoD=jqj-Gxux6`%nHf2oQMy zH1dOyf@UZt=4_eEZz|Bf?MQ}jc?ej3_H*o6;4FFkH(yfPFos-xE5vxl7M&BTpF>3? zZv}HjG}ajEMjXp@b#zQS%K5Hz$dV&eLE_i!N+K=L=qL+vzHQ(IWOO)z%MfKVO5dZe z*yT~UIepBrVso&Ew`7a16#_Nu!C)YW7EtWv2VzE$2<9@!o4}s6ED`Ojbp`Y&%HMn) zh^y8s#0BTGU~ZEPjPPXr3&+YgYRc(S#V^(x+?mlm=$)I_q5$_m%K zPUXPxb%`$H3fVT&0JWtJ-)hWXbkf?Y@x?hBrwwR{zSoSu$Y`|B zu$f$p&Cu&h$Ptk0$oBafkedmr!RqA}duK<+g_pLEp-Lw^!3_S0>q zV3!`Wro4rb1{b(Xy=Rzo8k>Dkz5alK$Ihg_GTvJZvsHxX;@fJv>RMllDyA<|tVXTW z)}<~q%$fK%b5(QWRS8r>#|F3N+0)Ry>+WJNd?inJE`NSR>QwglJii(0Z&}vx0Mbeit2Yz`S#xMnyAwIY)fN z-~CEW_dUy3DgCf3UYiEw^QF>)Q*_UKFr}rB%cjnFN7_m6_6?Tzub5Y_QLoMf7r>AO2(z<}aRyt9!lQ zw+$zQBSAD7%bs5pN)+ z0epn-PB13T&wiN=ggCs3($oll)iE6FBI&U}T+?c$Q~mL=6T6~Hw#R_KdwJz~5`T2! z3ehRBz(I+~`MsKKE2EmTY`ram5oFJU!d4dAhlamBr!I=AjJ-%lD!~1@g_by)?k;{i zZ9gzQniHsju8=RW>Y5(Nd`vW~&my`CnKFbmmV%PxCfulan7&qiC0!C}l4N4HS$3g) z+4hH%wS3U~xl5~;?Wl-!r@*Ftwa2Jp2M$HTj?lmR>Y!gcHBR-5krGttM_I=ZTz(h?;ZQ}rLTCQVjCFf;Kr4zk7-EZ8vN;>H2}b zR+3$@gNTq7`&!9MZFU=>VZ?#CzB^@ z3Yq4hj|iv^?-mZ0rTWy$w}i`?39sF~h?5Pw7Y2Ks7+W8j!v;d`tzkM$7JpUKui0xP zbows!s+b&l^gr1gdLoueu9Wx*mpRJOu4*LzOsOdMB)%?vb>;ZYOa$?_L=Ok3r<@HqOG*fC5<{Q|BYKy9v?X&_6%ML)eAoS#1qg%pf6cXOsa4DnKZn5GH`6nyN{KWN^a6^*QWTek~9g}#xzdP5)0 zMM4lE1io>yX11ohLdkuaC6$R`;heu=S0dbwSeUH-6~y6&E*kyWRRh_oS>C=wDE9lg zdK5Y72i!A1G(P)OBkk}zSBAHO1?_}AWKln_THX%`q*jNC4(z5b2lS!vkOq&{>@U7@3D+SZKt3+4v}RsUgGlgO4e>um zUQsv=zMxh8GxALV+$A1|&pA}ThtB@J_4W6UI+dK?TTiG9?92e5$Iw(!FQHBfO+&71 zw#t%^aOGj-- z+}t1*O`!Msf~}MWN-Sc0y@iX#&Aw`rUh;-e-+{d0D_U=8aul{42I?RJe*Y>2y1b&E z5dmvuuF?bXJ>T)OG52{^c4dmHYl++Ms!y#a%pC)rX7-SHaX)>UxHMNlh5q_c)!2FT zL%=~{AMG7RY5{pZTw&lrV^$~%`>*sNZ_sOt|FDY}M$-9f3CURz4j_Dwo)oQi;4Zxx z+@v6}@sK`Qdxo3PcB`4b;iIzZ<42k225j?N2Pt+ss39@=U(JY^pS_LnU^dnd2et)G zo#Iut;Y13%Mu2ggVtC_(ng2{kJ^Hyb*1300xi44nupQgiS4hh+fqiH_n1peu(Xr7r z36F(b70YnG6FHKgqtERV9$n*oAAS2j0vzV!8GGahH+00c2}Zv091;!&+Nc?6d%=5g za20SXR|vn4;*_AG7lLI?*BBA)81b6^$D00g0MOhhrk#SAzwVnLfiQ)c(5DJ|SrX`s zuFLV6cSyG2@geIs67?Pd05w3$zg$=wj{EDk*f{`T2g*S+hFb*l&R6t4))2&P=mzdSrBI<||QIPvKT0{*cPV+qx^++}>2J3N?2Wl4=IJ|Yu|cAF}d z>#E)Ofy;-QDQ7nBu}T6H+<^g9gQO|eP)zKPW-~};GPI1RomK&k@(|k~S!+jmdZL&( zjbSvI5eGe^Q;tvQN@u>d-o(uINJ7deVLX|z-oUeQZ;2HmQOiLF_PEuB9ph=DCh zn44`bQ@1i5gO}%yE~}oDpU+4&=yi;+a@EryFT$`+O20Q1lwnu@ErF8P_DnEJ1S4M6 zNfkBU$VE`A#}bd=gj>%D&%h{0*9cEPA!nV=4H@C8OO}g$V2h_^E~`)6JVCOvX&8sk zA}**aEvM^=;lF=C=88d~#$-y1(B+4N3Ad>`r}8wk13+@%9$$(PJBlz?z?dGm($WY7 zF+e`Z1-9$wSA+eg^nWoH6dM(^@Pm_l!IOI6k96SNz`|em__LQiu5g$GqtuEH9TwsoD;AVViXNHX%A{RLlq&6k4l;^{M!KTcrrt9~PQHa$A ziD=bY2ajH1dLq|WOTr@a-J=oPzsNG{Upfpr9LG+9%(rYmQE224(C!tT91<(M#Qaf5By<_82 zl@_B&@^#FQ=dOW|kcaW03X_=R@%ld`QrSyA8n{8$62<H`%!QDWRIAs(O4@kdU_= zbt7LdgUmq_G$hu%?hLTqhn~__xzDQ%5jJt_X1(&{KsPhK z^sZFaxZUNr8H~}06fXiSFAgE63iHJ8cLv~HW3+nUjqr2~1Ob%6t%cInDTNW!Y?x(e-R zJJ=ThMA!fQ?cA0YQ%XaEUi zzYWh4Bv0W(5KOg`m)4kA1xHS_^%w5E&@QoUv<*TjR|cSrcB{9wplQ}sKX*)!cW&C0 z7)~^ky>MI18++bhwXOiv%4%$auv&S#u=i20c&&EH2U5-qa}*$t0q8h z6rfrLs2;6BhPP)7ybt=;MTQmeYNMM%(E+^1ZS}xxO&@82w$sOGCp2o5UPe}e%IWE2 zx{ZmywwoD`jbLOzNR=~92D%KL`rfb(-wMI#yaYn6sB}^Ee!ZyPNjLHO^zao5qVe}# z2D5$Z9Tp`o;X$fG3IY?Y(R{7$>C^9PF06X}{oZ%GNAa$fOE!r(pij0~8R@A1@<^m| z&5SaU-(`0aaxo!QOyjz#i9^Jvf2CPzJy;hxh-xf+yQv{-6O4K?&Vxijc?H3m=)cor zp**Ds+vJrPX8@8kkP>a+p;kP3WWN9!SaxNCmUjppjSx#KU=qQ;Ve20cgzw0 zML8@w1pfEdEw13sZEt6()H_%>CXTW$%W|b+mc8pamSbzs-xtXgF+DcjM%~u`4t-z3 zQsgH7U7=++8P#bHH<+1Z)k-yL99cIT__CRGX?(i)oRq<+)nT)MiL095dN6^jvy&2> z=lZ!Zt&*-(m8^Qd(;*K3pGtqHRjE<%QVzxcY~dg3J{b$C;PLD%wb=YtRoiVGUhAX1 z%rzYp)?PsHk5F?3H~l;k4ononB5Azco2@J|mCO&gKV0if8yepA(VzL;o%A|_9`dwu zxQUczp#k}BH2CxL(bR6*@ZY!%D~qpEx9?x0_CmMv{Xzd6ELG0tYQIKyQa!y3*_Yei zj!yb`wjcUzGz0Y``9iYqeT!je9QTkD_az}uQ1LJbtWHW}&sExU7(iChSDwi4{CHqJ zh%-y?E^6;k5lK$`$s&F2^XX@^WWhNGGh#X=ZLsObS&NK|rMguE9Jl8c1?Rs6HQ5$? z3^fI|U;Iy{&8h3M0B-sWmB#MY$Eg7SFQVWa;C6*W9$q*7s8gh~E{lO8B|1zEh_mrc zE{U4(9AR7~`Y8i-xC)|DG>)-eRO06hrVwhLOIPKsPP}NKBGXL*SWmxVUN&`XDql9& z?C4r#qdXraWol^=U3qEah`UOTyzpfa>b^;N0lAYS4m3-cba#o)o^sjJ1srW+mooc=*| zPQi1$i{v{|CnxWws5dv=0?%~CZgD0zhy#tiFn5$9hl0f z27SqHR}t11G=rAfNmEom5e+I-bfi6_QB`%UXN7qhhd$|kDYo7QgBczIN7eeHtO37` z(Q6EUy*Z|(BhN^m{;m1=UJ*TlxJ7LCLlY4Z%YBQ#25Vu*YhUDGuV$UxwNu7CDiE+= z&SNpbAOA~}Bqh9Uf@al{>LQ&NjwFl0^kKL3PAHxxeMkjh?)_=ICA%Miev$A}sJF+# z0wAZWq`m@-uG_~C$ojTPNmWwVUZag7ajFWU{ob|FDk) za{O7OpQIx&=F9dH*iO9z0P$&7U}TfxL3Zn^Fi}`vn+9WMMSpw818Z+bWG`YfI6SMC z=*~G~{T6JHi5QtaA!dSHqdX(hr+MW91+iN)cxsxUYPw1c9SPOZ%%;pEj-uf3XLjSl zfOo10Bv6WwtR?cNaywN_RmRreTwh z)Y-;&L>b^6ZLRX5hS^F`}SFP-Z{t>j#jfJFCHInR7 z(==7Z$jcx90Cv;PbTWR&v|2zZ(73^txvlzo?(OC>Sg7AmvOH@%>5{YC9GHk-YdD>0 zG841&i3J0?x7O}J_qhp>83FyV_+vn(aoenQrx(GBZv<+{7kfRxi}P9-s!WP^r^RP3 z3)rZ9ShM`pvq_|{t8Sk=;XauBJMEbKZ9g9Wm14Hsel`nMIr~ixVqsw)p;*R_rp}|R z-Druy{bUW$uXlD`Xu$oDL1zy93>)n-$18fu#)4$Y>vGHxi!Va^jemO9Sh6N(M-o(v zDBeqtW5lIb%mDj5Q{5%Nv}k0r)pgYANfPPQ+@I!95x+7iv}4usHL;4o5!oaiC#+dW z=D?>8Ku41e4<5sUBXuW0zWg*Hp6t~h7{SkW3cApsNG1dBuPiPmgl0D=C%spG^_DvuCp4U7cqNP?a(clN z$t1!0TWipUlT1Y+EGMN_{#@*v64yd&ZgYZ|E%!tq+0I9!x0|)wbl)u%HA1%aCkhAg zSWA)vYC6}G4XRlZZn|24)YRucl)_?qbVL`H!bLxJe!#5vL5f{(maUf>iN-j|EHwS~ z_t8m}dV?(ni@9uMgSE+>c4buVf*Mc^%0)^(p8Xs@SZ8_rw9KGL@rpn5-qvThmEhur*flCp|mZ_-S69@uVbNqgNnA?4JR8Qkgd-&$KPy0G$aY@ zS9>X{p|xRd!H}A{+Ug}iu{DPUj_hq}`POdbCN%T3Af7jo0#)?_YrxBxA!jz$N2lHd z|BNnfkLJvtUXBmcU(-r;=<3oyufOzq`!)?2_G#>%mcP_15{0>*Xvw%U(2EAVBm5VS zPw^h*H*Lzt-!$l#E8cGfMR2qf`&LONQ3zi3ksBG!H0QAgJ-0#H6M-B!+*CCDatV3CJcyp{T_ER@884-MTrY~e|vW*tLgQlQ3>;C2@lwcKiz7yvd zta7q~`-2`wbQa%~G|)v?zE#Kd`waMRl+Z6`wQk$++HQl-o{7YhxbbB;$89*fJh(?t zj0ivy+KsmNI5Xod_dzugohRP~B!!159+xx-!Xj-;=vJ_XTmK}2ih;fZ!{dU(^_hb= zGraNPNm7pRkEemc=CPL)fH0$M5VFAAcAH!Y9f?x0SVb=N;_8G$!L@uOMOV7>MR;ARQR?*F#`~QF~E1 zhj`7?Cr*GZ+U?j5th_Hb;45k=N?3}IQpHDrC6Z;fjps0F6)zXukeR7W3&!)$mX_n4 zvd$Zi%=2dr`A#PyR?82}b9skib81iA9gtrbZ{pxYP`2*1y5U8RMsLN4;K-0ykj8tB zEHJMs_&ZuqIHxUP6mi8DsHTL!BVrhOPVl~i05+l6H0RRngoky@-qR-Skr`N{i+6X+ zrZxsn_Ioo>pLsP!WNj^t>?!gG$o15t(R5Bskc+=WR`Rx;z;`^CRfxc)A68ytZjxze z{;MS3k234EgBsI=hKRHj?CB_kaLDu^{I2A{SYFDgQHc<~Z$pP3;Zb zh32dk6b!TcVpW37HDw7xkSm!Egzl0*+=S{RO{9+F1m9J7V3JK)a9w+3of~(_CED_W>8zCZl?X;ZhGsn*w)xkl+oIrT0&q0}s4xJ| zC`K<>p=PZrQ@8QiIV(7;k=UxC&KTddrh&pq*#CyWp`u!eu{qBJfcv}czoxY%3b~Cr zmtUh*Eme+_#VS1oCJXl-*9Cy43Bnmg(Qs~3r~@m(-&O3uzoOQ0y*@Xk z5;nhHW8x%rFk!W0?_-$Fx86N;ko~r1IC!_K#kHk_$YDz+$gltW3s=JffCp%Q6AWNk z01=K_%sEBNG*(#;>bgSO2v?<$7h?72>473)XN~M=!R@sUg$N9E{p~jn(SWjkf^O+wYb&sgsDOtXA&q=uCKb zsp_5C8F`PX=K{dJ2HCcC>iupD-tAo^s`|_(hPX_irEO8&C1KHT(b#R&D$-3o@56)j z;$LI^f*U{RLlIvRQ=DFnZ#-X zxkbRju@rm)X@(kT?i-^~we+98@LD}EE&gD(dSIQz#9B7iK597z$&r`sihn3%iNPGR z6`q;(6?whU8xC&q+Zy-3$WJX0dMxDbdKpqpB((oc@wK~w=+I^oC$CCX*@^X-2XCwP zSd8m#LD`nSR;jS=i`W$1A`#zj3W;d5t&Y5%`=;+$eb)wl$tF%f%VNlCI%flM;GS|J zJUi@ad;ptELUk?)b3~CH5;sIVan(AV3@J%+#?e}YVOH5Evd?f-%n`HbDAWC7{V&QP z$AX_TOMMg+&#^}hcg(@NSt77v|5i-c5kkt5QR4E8t)s)d?2z@giMYCc$;B!N8)>=VyncVtJg`Z-MlO0p`?F8 z2>zeP$^Q@ABmMu*dqCk(WC*$GtY*Y`EC`7~7Vw|#fkyIXFTTBUCi3+SzN_%cyV)2f zwNgSQ`LcP$BPcqVmFILZN5s2^u&tB%L$btQjQ?oFS`|{>e%aC5>Qk@7QSRTJ?(cKv`Wc|^tw!^au+Udj)T_L< z{qdfQoxf8agddE+V!hdDJ?>I|DB|u>wK+FDe3z>lk!@luUQiH zZrT^8^&XvC>bbyjZln>k2mW{iOV0!3&|z7Y4eImJO!~(XF3F&;lk>sc*@xRS_h%o+ zl3shM%?6)inJ#gnDae{G9p*C~&bjeiCC<-}N7fZAj3L2$)@vwYwng3F(tVxLl#! z*L-v~jLqV7dCbLZSB!|Q*kB{#ikeU>QvJ|moDC{lmmtTQf}+T|(*!qs*)*vu*ek75 z)f|98Ey;y8f&N|aJQ|dum^P3mMY#0D#k3h8WFx61eSTFu3m|YKyyr=1gc-&5+6B}$ zIHl#QmXNggALj(x_Z~C4ta#k3Vmfto(nTp&7ji1nS3EtCRRSW+1;4?hHn$t!d#uLU zMszPdR}sE0=EYY9RihKv<~U)MaK6LuvZ@pDG}Vr&3(CK+vb1V=i%4hUr@?>ix@D+} z51H04^*R|ma|Yg<(3DO6y66ny?$>b-O+2V>>r{F=!ku)@5Yoh#D6#)kuWitK`u+g- zxLx~5{gLG3ULorCR(eTO6O0wPGFg9g1M^nJ^`q~%J>(=gccn;NftI(vx38iMGH6@O zXT@4w$+HvPZwg`npn-4gnzRPk74;-vLWyR~RH&biiCWL0jX*alOL4D;KX?ld5=M>l zu0oTKA3z_N!D&bxz(6;liMNli7HqfDE7VOsQz^VHn-4c!Qe-hw==hWNlX~ItKLVU% ztF<@FT@>DOTNGV{+PV}DZ+d)ZK%RX*V+PJPcMCH3f9*$3DK#_N%{BSsgIs#$&Lnpj zPWR8G^csC5PrljCI+Nnl=>?)aluXu#W~^9+Wx5{^Qe%5)g7u{*vve!OUh+~j_XmD8 zeo0B3L)aY7pKtcV2y{2CVx#P!YrcsX${7(5qEM=$w|Teh(m6M3U-^}9X81<7RltT` z?Gv9PxX+4#=_iHN56bPyAq-phH>R{1x|JjP!wexVR#{1f@i45!WG_xxtbhaE?N;!=BcXIm`T{=ZL;37T_l_;S+b~+8sE}M7iqz}l)&Y1S=yF; zd-za)dUL)GKw9N2zd@TO5gJVg#2AGglYIyji>3vn$NSx}z5>F6qD!knPW5^;X;d(? z6L_!_r0vF6V#K=2!M>VgLv$)x#kUn5lXY?yx3gbd;}ygY_BDhq2r^>3^nU$`Q#P1i z%N@9@#u?Yme=oNV{d-c=a6VILkI5AkfgvJ=^R6Khx!4__g;Dw(jVot8w>S7_}qpO?nJ-A%=qc5r0h@DV!OS;>Nis_Nw=cg3|yAxFdciV+R zxf#h&vnoe3VQ^yC9Z7nOD?B8lVY1Z8WCKDVbL~aB^J?r26>gd>eA`gOWu#DBoP41- z*}&Uaquz$Dwzj#2ZtgfT#*?Cf3T!uvcUM>`Gva5WwT!Xl&_tq1`%?Chi7V z&40Wrgw0>+<^?7)iEqPGpZ(67AA&SFW2MJ*^x}D{t;^(c^_nvHqW$=cSPhmg*OaHY zw7~j}p?~6KR%Pk&J;i9ilv%q7*9R%1CJsmTUz93J9TTV-vnpB(mbpSVF7_LH=9Tm& z6k^R(*NNH1lrmsK0>|E5oQ|64L9Sz6)^HrBFZw2XmN{&B9g-%DA?VsH$AoBP+*yRs zh&GJmHiyKxNGms1Ti)}oSy(vhPbtGvH%;g22dsiu#HstcnKux~*~U(Oiu=nxtUPTl_I7yp z?s?zPqu6SUmFpsQ=nUV0n$ znD~xrnp7c^Pc*Xi#u+ExGQHCHU)_DnCoNWHelBUdQhvL5>@AnjpjD{ngG#yDTf`Kq zAr?d=JEHBotu8q!%n;%*%naL;$tVoUqgpT9m zDBXyOpwe+@8mQl$-$xxdc(YgKaMI&ZJf2jhdUbuZ$%v3)Q@?tOo1rKA)U18CIT+_L<6oj=n&!5lgui;dAkuZiz#tn%oX*}723ekl&xka{ z#x_zP@$1XWe4bK6_{5m#@RWXFn~~C2zszV(%N>(vP~0~80dfW4)33HK!-ni7J|P!TR0J5g#JuC`JGyh zH=OepR|O3zDC^!JAP(V!dER)-iKB`M$pSshB1=& z$5>Az6tK|)xo(u?9N2!MK;8=pacaxP#VQon&@(FfnR@w9BO)B~!qv9IvHb8-a!u#y ze0LbFT7U*42MB7Uur(5AHCYq%Z*(lg$|Cue{{Zek{f%!?9W2*m_~K{<^lT(gLq!6V zS&$;W9U5bG8QSQXaUb|D%g9NLMdTdf8PHps%aMR+5lV+fUhQ$da0n#Ce|u_(xkBFW3sb+qm@YE z!&eip+*+Q9@Y0kyPHbZV*upToSOCSANio`X*q%wqT9{7=8W1DEjR$PIVppY^++~BcebDL?ISBI0ZlOc|wWtOHgZ^IP_V)E&nVXMo~s^+2D})obkk z)KOvSnKsX;1Y{)=x05ks5E-I`Vr>Uqc*Rfa+A?53;1K~-=NZer+7c|Do!*ewwZ!)m zU@ZMCj(SVs*LrQL1UmBP*bi}8T=!W^HMF#|;8)W0Pq(na?eOL^KPsyzFob}aHIoJ- zb8anYuAA1{4nIH_0c_7myAO~h%XrkK>5$I>(~?mdQa$JRAbVyyI^?M5>2ZG2mVWkO zhlmi9<@wHOhZbbrJ!FS!XY!wC!JP8A8MAFz^JcohfVmLtZ`uAn@@oSBGc(xS3t;`ceO#`5E$R-YcjRv>`ugnNtS2bU&wWzH{(l< zfs&6MtCwkCIAB8|Y-sBL$?@-OdZAGa@l8GyuIh7=ejOKy(JP9YNk!{f^kb_;A3Yo~ zm)-D61}%RMcw-o7BdGY@I|`Ip+|ZNXZQ+U+R#YbZ9_#jZO^Uiy!XqYT#ouP#7 zqte@0|eD;q|)(e5=OHLl-6@=jIg8jMTP1XpHIAPq5^o{3|ub&DIm2!_0 zik5S8nyW(EkaM^{m8qh>(KVD2SP(Wv>yiQi`6bFgy2?Ldv07pxl&PvpGs_?}sB-yy$>_k~O7u?vGY0Zkb z8&=vFr@Kj4>a(U@SXB$pd#zXnyU$hLVAM!qR^hbgQLol63l+~A)HUDvtn^kj$fx7J zu4&y{l|#i03`7l*8~k`0GTnehw6rLk2T-iUaJ^%^A|&nmPc=V9>#GyV)~XnuQDVO^ z6S|V;24hwaI+>53+n$_D(DyaQPdDbu;Ah&I`_Jih99xZy7SZff)1PbU)io`%xv#QX zRxm?BeMD5w6*4%^J#$P}an$9YQmRFtQfO*o&i1(B{!ABFK z(ijn;(p4eeADmQQi|p9bEja+WPz5hhV+{lTJ^8;rAxP+k?YXXfQkM6p z!S9>3yMaXi&vXFne`NAofam`ulUV@hV+d3fCXV>F19&Xnf9U`%KtwbE)jZkJqdWot zXQLis{#7Q0#H{qMD4>IJK99>FvFTBJf4&IWbG}$nyMhq$FY>>mJ>N7fxK!}>>U9D5 z`I<2m*kav`diBigJGTaDjU&xw6G23~L>vYd4}}dmiaPuHogTx}oa_cLHyNWXE4w+tjr{AnKWHB~H!r-9Vsv3o)tqhWY4|CV-D+Rm1uFQ4&8b)k(5e`izG zmIS+9KAIu4@5lxHv=arE#;F%5{8}gR`@M)w)4V;0XHqB>272i17stMn>6Z=#T^ua1 z)45aS-aq{O66o;irfx3$c)LI7k4EhoG-5mpx4@I^<3B`H;nonhJu#>)GF$1V*@EEA zIyUe*|2<3WFybQBBOHdEx981majS_eho*N|kFE_Q$u6Ms3cc|0FFwT_;KEdhX!^ zLTAdUE?E#?ULzU;Q8>+Pl@v=Sv{l?l&3$uG@4z=35l9ASUdyI)wU&|bAG!lzwHhz=EPt9)R=ui>rxI=Hg(FAT7_0H^Yxw< zU;~<9GY9Z0CKK9}+V>lVPO-dwP*2S1z<090m2PbqA~tT*27BnrTuc=I@~yo4CnEB$ z|Lv)qt8A6*YPnTopb3)^NN+$#fAt9+UG#s1a2j6^P)SAYtx|r&I8;W7K6r$O$i_jr zhrHG$Vv2k<0@zo?43605|JnTD#kk`5Vm!%R+|sgmT>e0noj>-KD}=hxn%9LOm*r-b z5gK8b++G7K_A8qC%Tm-eUY9wo+yAc=i^KC$xvkYE*a=y+G(k7?c4PF|v_MLd{^s^m z$oebV_9$MCm4_6%|D|tL4hC$xjHNrKVAqt%`bMZr9fxklFHqv*Kt+0v`pA%J>vCj- zE~t0`FTwtk5W%i%+LFW$yC&x@{vH4MXs)@1RD?gr%07no55Y>7(yn~7rkND*^Fg!i9VU2j1 zkmG8nAMw*Lk>Labe0FOIYzb|24s3pg(=pueC%Ly!C^lp18xTIO zzi#Nj6N5cVruZWG>;_i2NMu{j&XgU5@`ob6dhePBxj&gx{-pIt$ub3((t}(2O`Va_ z&%eFcJY<&kg3P5IN|K%81KF(%-erC{7YNp(G|Ri{m)XTniCJ?r-@JYb6FcDb$Bk@~ z{u)s|*x}-k6C_I3I!!F(3gFjuM+20M(o~VV$n8EN9?t$k-Fj{zcYk#im48qp50`g; zUTN@>D`{F~8=5mKpu)XJv?pJvUhmLA zZyY}2xrAY*{DnJ?5@Dj{6?LjEMZh?PU~DOlYRcheHB#je;}*Jk&nL|u3XQis&*xZ5 zR2I9gj=CykjsILJuN|X@EKElOr_9~MBlRNh2g-%z_Ouz=rW^{vd~`i-+<`T-@v^ZR z3de4mk^QrA#J|}dq1<&Sq-SB@2lDZylKxg)b+d~Pm#doYva=#hj2tfH85FfHaQLJt5NK8dw<;#L9{fn6rbN{&#LJD`=l55dOE>yK=l7~Xc zGr4m63s!wqSLd%%OD~L63V6g^;r~YKkMO>v$UQrhmyqF`xqnce@?p=OsJ0*)X2?<3 zFzz8FleUDTG*dP?D&DuGp#UuQx3-|besmTLvH^r2n+M@4qIr6=xtB4d$5y2?S*Z2Q zmX5{2Jr{%we|1H@iD9XR7ftT!aez6XJ&El^r>zOgb zo-eHNZ@ZQQ50zBh>N1MgyLB#5KC}6>r!k?2*qq)K8c7bbG4sx!c{FSe2%4!7$UAbT zXm-YNlG@Cq?>&F2F6rm#)bMi^ z$|T!($xf;G3eoVX4|^4o(h6xYqGg;jlSWa0=8hg|hLL+8m~>eJ>f~EA&`>LjmFtDP z)~Y0bom3T`_Al+WB%03?kU0w-<nqvd<1#~maeC{ zvnzc)n=OM!Uvbtx!X++i0Y6X$cAKF)7 zaDN%N?y1R@tj9EM)0~MsQ48ToTPFV8X|efcVam^Clj8l^t6CHfS>N%2e~3)?kagcl z$Y@^EOoFdx#*QL;ZN4JFNW!%k<`OxYFf#(b_FS_w(NoVy);6Xv#;zL|2O;x{BQW*z^=>& z`ETcu>X*ZRpGGKec_YxHNIw09|MxV~(a1WH_{ryBHFUaSBJw_j_+jT$w<*L0={nzw z(fPeKJ0#J;1840_q6&;{msjWBYOT2nmtuBDkX0nZAr7ekB9t@^cNi8;cSDx*9fqbcgS< z<;(@1^I3hdKncbNh||s8-$R({&YLUTU(G@|kc(y2a_Yb+6;WSy%e3^TVr_13s)!Ai zl{m$>$z>>jI3Sp#iMTRJ;arjbz+8MDwC^M>g0WKQrAC*8PO~92xZ;IZ7D^qDJ83ku zRn0R?@E>a18eejWZK7kf4pW1e(=4qls&sc26~QXnPz%2aRCar*A=*>-mEm@w5Usnm zug$!gM~)(lX531p)9l9A4B?PR+o7H|mfFvCEhyB%X2YEY_#`@?`9xgaD&+|$aMb@j z5Ru;mU~PUKFixu3n3|$(^k}J-mH}n1;&36$(W}_OE!wACD{VBTnz$S|Lvz;@1k)oq z)}K=$vG-HvBGbi8n>l$+&6C7#i%o~2CN5Orl@BisI9oX*R(vFgyjQ1qc3PEBc6!o2 zdv@X8O|1j@TGB?}(Aj3u#>58AZ1RguN^XJncWe#P?KJq4>PrB1ZOXb4C>19#(8|qr~ZFZv9PsSw_ zq+zIg4|~KWo>hv_Yu*%!z0SWUZ}xw^85P z_LaTUQQvV+8H-PDph6^D zGTT3=Vi;iLQyY8qf3U2;Lt|NrfOqfK16dEgNc3L2PleY7l&G?tE6|@nERDkSVsQrC zvg|B7A<`c-_Bg=OJmHh^$ARMnfmB^&TE3zWy)?k-&na0?^&$wEEeeN5N|=>W5=s7S zk`n;Y3^hJalNlzGSqA%-lE{)c{|(E{@pTrnRmQ0>7eXgSL}gDP36w#$3;0t*{7sVy zjJGtGzpGT_lOgYaMzOL`6$Ai6H_ z3p{47HJifATqfkbq={)UVvF70p{iWcuvZevOFkIp`nnHw57)~nbAS`a^5JBa4tDY{ zt-1OXhq5CrKNk1PLY-PQ^<#ej4wO4DA|IyI!(V8MLj};3hGz}k$@zI4Y(F$eaGgRH zipXyKiItsj(_zH4aFWm4;I4+{zwb;*<<9Xc5P`d0TSWkSExU@5sUmX>7^4pkM}Z7_rfj=|^vGPg{bi1Vk!FL-~e^Bb8w4pg?OKkdJ_p0I)P9f^O25 zE`*aErd5T#WHT&}I|gwy&k)`i$F;tf?UQ!`mn?lrl#$;Hb$wU!i45CF z#A?Z+KN#PE+205}UdN+rZ{>>?+!nO^6^`!UZ1IM<2~yrh!qWyj2jxEbMF5!=0dO{7 zn<>GFSNZ(_(BlR?mSTJfXi^{Xnr6f=e-K!G5lhAw$snBXP_TDNLv{H(mShSb&>q=# zB=kp}HZ+lw{t67?>&G*B9RU)>dpC`Gs=M|l0!+6-V$Rouh@S$mQ*mR(KN|kI?`?w7NyBD;YsSYX6QPTj^`8ZM>TNp9o-n2{kZe zm_hOw|3MI*yZP9;wvT@Tf66sU(N#YME#Tl3ZayzKJzmzk0RU1L10cc?cv=W|>%ZT+ zfs)Y-vJNpuEhP=iPR)%^`>m2vuL5S1H5;2Jj(uL;tZYEq+J|*d0`hKF_w?F2zfGLH z>{`J9Cm7$Z*L5Gvym~cAv)|*=W%uTnRWIz=W$%9C*9FCDNy+?Kwd`xZCqFBYLsxeg5Jb^S(h!=bZ9@A;RS0epp`T!uS03 z1bqDee^ZhCzxo0X3}y2xhDCx%{=2Cl7s$wN*i?+B5>i_o`C}guiANgl#plNn@g9UL zr!F>f8UByH=zP6|EC;+LH(hJ)Fy_@yzVw2uYR=hD7eHMf-wwA_&*mCxQM;Gu5OHooxOiY5{R`NggxnDAi8Zjg<5g%E(b|M<>__>({jsC7 z6nOj<7+&XGlU?iEkt5MmebXgr+~vi3S;pHbllxWTUYDXZ#!lhl{qR&u9T;n#jHMYtN0L94%#Bess%BBw4=86+A-{BVL78_B zK>=LwYMS6^_EuIMQ;5aa)Q3VHtd0*Z+?+hidzRoRa)K9A5%8DM^JawMF&uE~Qn+PW z;^k~r8rgJaaXgQ$q=69ds?;IY{^dbcRx)H;+s6(S%2!^w{CD?--l)Zzn$HlflNWtN zKwJhhOx>I+yc$g?iVG&*NPgW~v*X<8DRj#YYuLr5U-972=1r;>Ppx=Y--;Y>!|5*cXD-O&A ziv%;0XMBc11g^bMtP)+R3ZS#R@&L%`5&zeU#+YhtMFX8<%frcT;gKCPZ@(UH8B_Xn zUow-^@o7UL-wuaoqV&FlKRr(eqt`BN-0+4c+_0ewG@>iKJbM^e?J;h(p*fJ8^rJRR z^7!RSoU3Z!1VzVE^Oh>CZn`<6+B43k&IR@!y>^FlZ;gB`~0KSQC7YR2!hi_Rk! z5r4M5A5tfM_G|*%!k)db-Rzp;R$*EzVdCmGTuL6%L2wR}v9>k5u zkdnariQxt@{6&tGXIXW8PESI!S(PE5a|TyO8bD6sc53MN^~Dup50$j zYy9!*LHHiAyWBx34Pg$KD$j;0gQs1|hNnbnLmC6;MA zft@p{T275oPXPIHsn1C12qdEczDz1FcDgDZ(f@e zlkHDCi&d0YKXQ5eZQ+sV_J|w97 z$GUT^>+w;I{r7r>>@6Lnh*o`7nP;nQ*qp%+^uuj7LK%?k#`Qc&g)i{PZl}U{p0PFZ z0j;o%@OD2b13nSP?JzQh=bNfwE6cYxKP|RiEgYGl%7Kd@HUI)l`+$S;=qfFwN6X31 z)byR9Mwa6iU6MvSJya2!8mPd~OTl!E*)FJ)E6uaItDrVA=j0Yr!2{br|2Kw>@SilG zTJ1T$XcQnj%R*Yo?O+}-JZ}}@h(MG;pW@?;r^J@s%b6tAqpx^=qkOwRdkE_)Zj6m$ zMWFVNDjt7x?+(=qN}5_3oMkH;a|eM^CD;pXa;|E)Gli6om}G7(Q(nKJc|x#KvMFlL zFoV^=mw_0#u^4qV?~=&Rj}N@fy#7i+#R$8);Dt zFK4k%=6;C3Ta#J3Z6iCCeX+7k^avbJtj#h7QATafvXnL3HnotrTfIcgWhw|wK@tD* zR+*8aR|LA)mz#wQUUL6lr5rzTE^R!TM&bZjK&HP$h%Sd6sN9x4tWr}EJph)}lI=ST z#(XdR&3Q;V&$)|)fTRxoR75+4ZYq>k!$epv&2ip_Y_ZI)>T9*|o2_l6?Nh0HSi)=L zS0-F7t!wFZHi17vB+2}{(1?vQ=MW}(WmdYY(hO!Z%H zs4`CGsB~Qtg-rZ?(rU985n?RLHJQAz#xSXu+%s@G8Je5U%n?C2Bn~bjbpdeFn}P7Z zypTNGY$i2Bx;~Y;cL;@1^#n3>w%{&cX6qG+HeU>U!S1(x5LR`d!d81H3IT{Sq|c57 zf4XUKsH<$EYh+oOazsMOR7D`m1}+*K2qJ zehR{(0jUZ0+$vTjo#woUy1t4H>e*CJ{RX`N_^)8YFBBn{zuiWEzY%_CJ_diCd=~nU zH%wuXglk*Di;n#Q3yE)V>;QpB6djbe!zegk8S%L`2m2|{sowo2--_@QjKNP;B(V>) zSEC82Jd1+6*|3X+3xx1LIoFtIBO7BQn}J5X=3$|&iu862WNIc`P9>cwCi_0bcaGKm z(H7+)$))vI=O?*9fCIXOv8{MEz8<9e%sJE8NIqs~Y=ke9OwkpTEB%9-STk zxh+ICTNO*zP$b8kp0`UXdd^I6ANhnx zvGyNE{$zkq5tY^E5dzch71D=_rnH?pS9l@WS%8>1Xg>t9o z^o!BOjiQcPa+KzbOQdpb_Mrv;P*B6%swI8tA$_UV+}e0evJz^=vvdws&sRIOb5M47o2DfCUOJEvcX8p>=?Y2d!S2DWdf^U|nPzvV!tj`ki0@-{;^L7bpLBP}}Le5o<*dtSOp(R#J?WrTV-${uRrmEBxKq*X+6{ftQ*LB)q$BhsR^EZK7w2A}7% zYC6*F^E$?Bo2NTvafc^;vpCx?Io*eDc9=oIh+k%YGm^P?+h)Lyq;0CAhs1fLE)E>6 z1qnkX{{2Oh@q(S3r5b-Fn=?#7o%iZVB8N^nujkk;5Ji(L%|JtLx=PuRfK@EIdL zjeFeA4sTX4rja19H zZNAtM5v>Fkg-9oB!eld)MT#*xfsw*9E#6YA;^jg)Xj*H0+pKsjj`SyaTxYe%Sn%L( zMNdmX<6AO z4b?=}Gd0$}>HPTbqSBJPs?V)Tp^CpY;UjrpmFs!c^2>(83!OQJr)L7<*}1VGZJI)X zRQki>bz*~x>uMH3VYEpDO&0~Rpj`uTB_RbtGjU43Yc2J747RiS+On#AFSQqU{qY4X z<-xv>6b;LkXcJ_*${-hALs0k7fj4is{v~k>C%)(#v6lu+R{Bg>qIiAl=CX9KYf_B zZB8EdI#$kh@21?ju_#ZbivRdr=yE&xb-Pn=q}ejmXbK)8`aA^9pF`qUhU$1<%kRgH zj_-mlQ&XFo7V4uqb;}bhqg#HR){09PHZgUHOLvMg8VOqzw03m`QQ$ERbYYpg`$}^A z4_V)8fu?Q-a+ZNd5- z8I;75f^A37?#ic=n@_))dQW9~&#ihdU-e$)^H9dPqg`cjNN_At9>lDeQaO)sF|SvfXx2@jQ8*%y7l7$ zc$`f8byU>P_rQ-ofu)gdP-^K;Q9wds>0G*{SrAZ=F6mrK5Ttu)SW3D}kPuh|k!}!> z5*PRT`n-SVdw$MK>yDsFdBfM0T>#9qX8ruK%oKj{}~De15gM6g#l1F07U{&6aYm7Pyh@C zgP{;G6b6RE!B8X^iULE?U?>2Af+0`{1PX&d;SeYi0!2ZfXb2R5LBTL61O|n{pl}!z z34@|wP&5n*z@cC`6at6B;7~XmiiAT^a3~rM1&~lM5(+^=VMr((2}L5IC?ph(gaRli z7zKr(pfD5^j)EdlP!tM^MnM5I6pV&K&`=l}3P(ecXebH|MWdkrfB*vs2!MbA2snU1 z0tggi~xfX5HJD;M!>-cBp87LBhX+306~Bu2nYlLgCO7#1QLQkK@eyN0)QdF zFa!jKfWZ)O7y=1HpkN3z3<1CqU^oH-N5J3+I2?h5BT#Sz8jb*v2rv=>K_XyC1RRM# zA`vJg0*yofC<~8b8+X-R4`9ZYd-zxkuyB>jWe$8}aLLXWjO@yt(VyT#nljzU;{9Ybv ztgEX|?o1hsTNZpO`r$O85PZF;`Mu$AzO3^5^=(ao{BonsREl;`n|x64DQ<^ZSG&j0 z`duY$_g4SKMuYay3JbqZi#*tFLe_c~pKmn*3>!1u)20t-pI&1S%MCy8F88N1Oj1K1 z?yMg?zhm3y^gLYM{W#uV_()s+`)#~D<#bMQLQ1b%O0Y&4RNfEA9%yp z(6wi7V(3ll?iT3#?@eWS&#Yl;%WMiyRbe+p0&ijVlh1p!mLBf3A5-i9=wd4}GVl~Y zE{$^JcIBQGRE7goue;+=+q@a9Dcm-8P_fSPzF*!9<-!kfH{~5o=>hy7T@zNFD);Sp zy@!9PlQ_#U68TlOp0Hw(i1kM$k+8;F9tGhczif-)TO5?R zVQgC(tf&*PAJus5+|0DUDmt&0nl#?oS#HYvIS2=1C7J9z&s;HWchkYu$@|;G=j#JX zB4hSsk?QsMZ}!Kp!3Fl}R|}Y{pp$(Zx?m~?f}jwtl^0UoPI)HQC!a?e4U?KG;i?O#{$J@8c3w!!h=&D>_PKg;tqs=1!LZnlq)e^R6Fys@>=K**^U_g%x9+tDnq+?GQGJA%1`DbI<;vU8Vxvm(KUU7Of2QruH>cVL#6N7Ax@mIz_e} z2hSePS~hVy!*X@-c=G0L|FPq**IQf0czwH7Vf1HOaYIZup%p!)NKJb-&6sZZfI^U# zkhNZwn&2f(*}!?CW^>SwIQAR1NQR`h?i!{!7`=9m#r#7%oG)zg{?als3^BUxOe^|m zMX|MuGXp9khHX_y_a^gewdvKl-)_`$V*0}Vv~TqZM_LNV*s#ejl(5<1v(%*{{6rq# z*8WuNeI74HxFNpVU=Cty1pOa{#WC;5$K^mipEra?!OHUEIQyCJz2&@uN(zRv00lt$ zze$E?D<6|;sqVotMmCdY&A$^_w$qO&iiN=KWPh9Cs#jH)S{|xw)hxU7;FDbT=-9k3 z*4?koe8D+66I+U7m=krt!M-&f=U|`TPXHORnHR}uO{56;&h|*5J1wK1yu6anbW*UQ zDRJ~KF};adi}6`iQvChbGRkb#H}#_OcL^YN$_Pvxn3bJq*W0kG06Eq+QlDcKq`^v4 zEpIwOmQL<*z%EIpukLf2=9UOH2Y>6-FwV$j8Qc?ae$|kkYamsQPL_h^3X*sx=D&Z# zeI-l4Ys%$WPJJ?6wB$dSm`3yh^n014uCpP}e)cv7DH zYujl*?K)GQ?{^83N@5EoQ36y(+%IiR*j%&$uSE=asAv(6$3OOL+EtHBeC*CqB4s6B z5HJxxN|^gBt)bLB4hr|*sv2r*j1@ldCZ^2m80L|>5H|Xw4QPNyGsQcpRQ44B z?wchU9tBoUtC^@=in+iNml;BOnVx)DD(#lUDMEs6U^JC&`sbQ%#3$ELw}NOd3g{ zwaU@H@=3Qeil=U?oHovhjM(QaRG8SW!!nX3x{~^4xy*OH!a;^_9v|+RYB*kku2!(4 z+Z14PwuxU9D?vjyfjNho$RV2klwLH*R%- zzVTVX+{JU6E5!xn?*msezqZ+GUyUvqKGAq?lPT2oqbH+)_+USX4DEX}iQen{2GjvRO~Z0rV~K9a-{&t8-Nw*( z=R${9!e*%x7XNO+Nxt^4&st3sF@ zg$ufTJg@h!UB%3Wn^Ae6_ry!$0$WU>pVZ{qFJMc%pPZzFN^Gap;|;!SjGE?%I86YZ z_^A}l>^}3g)az_T|Cg0e;S7J8VbTq-Smt|zhb3c?cT9((FY)8T57hz)_9Y?)yi##R z3x`-}_q`ct0gDno^sjuFcfG`Sc=<9zRa?SoBQ2ZK6iguzYn#xJ z-_ZYJqyuH-6kEtl8mJzw&F!MIqQ;b&C1XQCqM-01S0$8thnFCS6cKb(N|k?}QjS44vq?T%;L%ieskiBER|y@>V@=nffCclYbEy6L_afwf&@y>-j3n$0z z-BR$mdi$2rtYwavfPf(Foa_Wnp!P;Z)F)!}HYVC22GbwBC6<_Hmsn^Q*}x-v^PgBs zujb9fOb)_-0BF!Z`kjmCj;MRU607mr1S|&lVf}s+OEWrSg9TU@%nYQH28rv&CfK*n*Bp~c{EC5q9kl#k2^2z+F0X*ehlgdxF(VG5UnDPcPuT+>bId2u^LuGN~SEjj;!gzdhe1 zTi=)aJ+I!SD#|V9`wr^supjoXGS0G{;}18h$RV>BZ~XjW*bgV4r}1$t*m+qTFZ{tC z-q$|f3vA7p2$DHC$-HgB+^p5JEMDEi1Ry@I>3{=Iovt-5OQA9=`nG^@7jE;>?8d7A zNFbTl<2kB$yB`1P28-L*H#=`XApTcB)rn&y>K4h~mN#3Ak0L0}-mvpp0w?=qP< zGC|)y*|OvHPg(yCdHK!X35aLS>@)v<#z4@? z&_Phq9OeJeS->`5PPb!xvysE1R7qx1NsiARBFfibR~edH=|9Or1gE9V^`{}iT=C+J`lj{bSi~u}4s*5IJhP;KMM{mTRhuj62XAsC( zPL@*`T0hGbKHLl9gDS!L=-Yj zlv6}h?oG6-M2uZc3|&p-jjym~iRv4hz!t>dt|qRa<~n#Yw}s`MZM8jar6dN_;ea|hmSbLw zo+aGBFuZWlZjGv_iAucpJ`Jy;3iK&RtWWf=Pb{W0=X*1Jo9S$y`}8-NBOimac_dQ| zM{tuB&RFTp-s4o~vOd)?2_qwS>^iHlvitqAuTiFR)b+o!d#2p|(nhlQhir{7e^@?i zc5YmnFe|&Z({~?H+c`(r%82UXIQ!pTMX}Ry7in@8?Q@VA6f;bkR84&*Z*HW!rwW-c zz0UKxi2G7(wG)UT7tdA zcqFb8`zgc+ANXlf@%lZnaw?|0-!qID%rV5jCI00F;CTV|!N8+)!ZJEGtQ}MK?CL#( z@K}@JPMJ;$%g#Bq0m79b8E+y|krj?c#i2m zEDQGx#QZevUPQsc6Kef{Q!VW>gNXRYakj-C5Tc^@eb;g;m9(tYt|qNtQiXW{ZJ@xY zoxs$qFCda|H`j>aniT_|zBpopa5e=cz)8WK==CuTT#l=^~o?G{Ae)!W4K;ex?Z>~vZERzwoEXM zW&FgRU|%+vrCvA3$d|bJ+(EnLUNSEbsVzY3MfT_~?vm^%F?`@DR=prh&wz2r6EzUy zJOo(HVAvi%jqJ_VFO)^AhB_0A8VIZqJGE;7bl6fW}6zcaGU$#@86wr#Q4e%T{;i>bT_P_zcBFp6G{k>$FNg?x&>qg5WX~A-nQKiA(0jFiK~5%TX;@Xd5%lq z0itc2c1>C^s_Vy%d$Xza#ygnHO1r_m#Ew#%%2^M<#V7!-ApmaC_lVo2gmw>;|6sBM zk@tL6l5%$LJ{>e2#{Y@mX_nA>gV1^wH$#e!y3Tyte=b9AlIhXd-SONE*8KU6PJaH6 zhT3=Y#Z;?x@6d#4b&o3>dOm!vROWwI*-Y~VS@%Uw`x^!Bw=sclqo1}^wYUDf@#zh& zeyuj%ruhTk^GA2rN@3bedeaQ+RhDs`H2a`o=|n%}`{lcuW!%B#smzm%F>m#l9g^F_X3;8@r89xO9{>CfPi|$h+yY&9!t*Rr^Vjc!v+I;!mrs5X zikzO1o?#dM!YVui8gUO^oR*%Stj(N{_MR`)pWqAQ#u!~thh5MXUeNbl(2@bnmoy^l->jHl7XmiF1siam2>*VbdI7KLlk4N)JQ;U7 z>q3&|xBaujL9^s?J=S8O)lNOc)0xR4y*6U;FFd0q_&P)+b23%zs;+D6_ku3!5B1Bc zz}lLf!nsxMnX0A4Rh#$6Hhm28Hmu<#`&-1)mh$2bp$x?@=t{Gm2MFJ8VIMnCUIm2= z!Y!}9@j+iG%>g$NuNeP)Amw^Yzo&n3CFyhRHh25_;#P(FE_?GjXk9Pr*UhlyO>e$D z*Ngi$v2WGCY}9_;Oj34N{5smOzy0ufJGYSb(=XZrj8>66!y6k8!^`fDK4zCbhHxQ< zfnAP{e|LfSd+GYrV;BZ|ox(3Iw{H~ZNin_Golo@FbEF~_@7&4$;aU7Ej^+O|_PMTM z#F_2jpl_OC`yYkOlCBVI9cAag9&pq_63fjOuBEu?M8yt+L{APSKCS zW-nrFy^o7^>+RSULYxq)M#WF>@%gS?bBgpvaFV5HtEew+hN##uL;imR9NOTT)}zFM zwptNo9y7Ho9U}S#AS$NN33{LO?qm$-Q%^)wJu1xP@VkemsgliBaJj}0bkJ+Bs{BO! z{X{RJw@2N#G3yU&_diCfgd053pPNobQ{~#Fv;rC($3L=ho19o)IFIwH5G&i5S4@3N z7!cl7mmL{5{35%0i6hW{YTQgzb2N_aIncmr={bPE*&_0i=&z4Z*FjZ==Cf0&?i#5_ zawIx({h;?JeKbM$gkso_#JygCvJJd>Zy4V+h?3xEd=RqwZC z)>0U9VoqwC@-);^EMt@qQlji(JR73gyr*bn9jg_oKb&Pw<2&hAo@+SEg^6;HmoX=4 z^t(}Cs+RdEyQyLB#cRKm&wyYU$jC_^3H!nml}na|?wa3XvRDgmc=^BLBqO&sPA1#& zx7W_I3Dr4$TXfqkT8-X6cMvi*?j)l4CV?TWn~S#Jx*oT_KF%^&JYg1AMU%d~F?F=k zFfP(c+j=7+kQ{YFrx5SPt;Hg@r3L=FWcgNrxg8$4#ygp{W$Cl5l5O>FHA^W0420iW z1s#uMTk--O?LUJzliwwRDesnT-p^^X#=obV$zG3Sd68os!|h227?`p}C+Uxm`){W_ z<9bDv{1VS>Ct2A9Yu_)%EdMLw4Z(p_<=!kA{$&wb#o(ZUQ@)7mXtz zb5heroix8SC8*?GYs`~xDVN#xqmM79|28rIoop@a2|DY%?LO!JDkb1iR=~;kU0EGJ zf&Ew4?c<=6?O066z+dI_QGt^RQMkM9!$fV zSVg9GvOuVE5+U=`yAkt_W#{ot_*4Fa$7M7luL!O|C48H{hb&(>>{LEo%I+@CzI@HM z8CUDv59<9D^lO1Rbhw?G2WvD~Ic-DUbB@sP?EpzW4-Ov2PO+?8Cmd5%hz>SQC3Q+m zyzcnhyphTb2msu(x@txj`cqAjV0hM(f{Nhd4!)<5qaOs7fW$Mss_loLst86 z9QqBcD2Xy^(_(lKvof0zIDuyWAYCZtJdthf$64qy6MCgFdRWtr82^%k#-}RGpmRiS z#tSfRONXo3?$YR7urYF_8qK?BocsB!J#nK*-CQlj5f(4aS=*+GhB>ma0Gh_M8$@TU z7K*Hd)aw7#SuHSSKhce@@;V9nVc8Lj3PNdG__cEc_~E-0jCxTfBA^t)^is)V<1yNmMDXL@qWB^OWgIu2LRQaaNf|lXq->K2}sE6WCk3k4t6z~pN-Hw z4v6nR_Vx68hx*45}i;*!Zkq@T`uGt-7XGb=xwz+SfmMPhB3*Szf})tVqT;H}L~ z75McZ$s4ZuwXMfw{)05b{MH;IMO1yJluG}`r02f0)VJ*NdNM$iUofUsxxJx%&5asY zEr?9Uy{pBV`CC(6zwAx%ctV-!FADcqs9DsZj*!rCVI!QnO-Sog<;g{YVfPkI+nxEj z3uF^_Q)2LWrl2m=a+P&!LH>GoKSWeA(|XaL6)gw}d{nZQCbGA@fB6jAri_Oq+jV!O=O{0Hu3n<;kEUR0P=24@q9VsG z=c;a!JD=T#G{JS;ZNzr0eg3ReXZ^miLx|SL_f7ndd)f@BF5IMw>wJGNF_WKaceZ@$ zbqf<43^ydH!94IKRtvP#Z&h6Ai!t2nTsob#R zA%#8NKb$Y(le)ULt9sjs1^4tTT>Wv1*Wb1K!!r;kV*U5%)6NL;UqQDP>oVK#dU0pY zFur7K8@qT4{HFTE#>f(ydM|%|7=?G7Oz;+IL&q?OC$RxClA1H#f9$c4PipMi4u^c_ z&_p<0db{kdkXpbO3ku<6i{S02?Ey3Hy}$jfnD=FF17^csT*Sle4%C)AXFWVc3)O=6 zbza|pE&fLb!E*Ha7q9SWgwI)=*9d;imd2E>M)?=2=oGXZEVOx9) z6|tW|yNOR8jvU!bvimNBlqZ)m4<`n~yI*?BmJ~X#HO8s#N^+`Br$0Y5o8tgivzULZ zsa<8ST)Bf2pWoGdaPYbah&?jq*~lyIUaaY!;aW+jB*?cVk2BjIb^ z^AQdT>}n^Lr^##neRiu^9 za(>Je)?YAQQzpUxV&0|q*>(#}x;W)jt>NbvkyYigL2x-LC8-tO^QKq`5W@)qaec!J zOfSSI>0x1%XBY1Zla{wjr?^?j#HmG)!3h4mFPyB7bub+ihYfNX4g4w2$FLtiN|5gU zw<@nqDP}`XFvU+@FOzwwNj=krc(m3VfDq)gR1&vTaE)r{h)uW1(iSX|sGLkRq z?@zmHyE^FMpGTM<;45nliGxNslOjhGYU!^^7?f2eBb2ErDddP>}rhgCkX2QbWAv6EBg}M!NCjS!$X#4_e=fwuGD@pIE?)=7M^)u&8!E)R;Y>% zBv7?qE0FrqX9yarMGVQ1^k>ocXYeR+<2+KxtfJ<4X{#{oQ1t0Dp^Ae-6il`VKS=?v zdk}&oBxa7+_(t#uluYW7KT=Zm#9ql3j^_%`jJ8w)vyRXyw-_1qV%C}lo)Ys;eIH<{ zZDB{JvnHuL8px_EQn7wJwt}Ui00Bi1j0V+mgn;DX4PI<0xde>vCa6~Qld=CD z=E}@00^!AjWk-TuM#K%>XAIpRMu#@O1cG}hi>e~K$;Vwtg%I9W46*S9c=Mw zGt|}4oOxHx$qMA*Xo85fI8E8czbIC6L@Zagk_}7=Y#U2EoNsUvTT0TqquG-5jR)nH zCC1)l*8Spm6+$U03TGyk*JSkrq8#F)l;c1S5Rkt^tj!7Y$2GO@fx}_vV_CLS$zGbt zrV{A~6F>h>{3_K_=oocZj`W)?5xQ5sr&9CE%*PD99EwoEh;O|dS(yM$X^7|X&vC^1 z+2Zdx=mfu2c&<2l7$pM%7cR+aCqt&=SRp!dS)JXG6+%TVw*dnZEy+^Fq$uU&HI)?D zmmhcIX(Z~6+XLyaug_$Z=2`^LC!WlrdofqMc@CEX%L31gk3g38f8o&#ES&|+~X z$hgy5f(|V>&Tb0%Ct1Ujxj^^;UbGM$HI1Rf$Rw85;LG_l{dtX+f!f5l6X`D-T3Rn; z)knK^pH7Te4EJU-_uw)2%>SJi#Wi#+>ks>+tUNXGpM^TA3v)t`ix~5muI}Z#;tjhm z@UIdCWV5K{7E{3tj}~0;d47BVccn``v~Hs%ZM<7mAJM9LeCs;GH+^r!)MDh+G$1ZH zEzwHI^6L>}pTng$j?1q@iVGE%{p6Ozsy_NAF9+e8xDPCcNG^Y<`b>Cl7*2z_ zN|!7#O`n$~WUPx@8S05ylyg-wN_Ii|X^mqz>pnqROu2Clrs5?g?8Jv`qkUuvhie5! zU|+@?M{b9Ex4|?M;Dv1H2;a*Dt)2-iGs7ZN?f4MG%(wBIOwG4xLTF8m#$1hm&%XJB z=6GL;&de1kGLb|*CX7xd(ABPwYmJLv)KG64zly4DNY^DR7RZu}o1$F38&)JQ*8Rdh z*iiRG*zx5mxI8_O;HM|Mcn?_UC1q#XpdhG%INei^&^$@vYwzyy^thby|@8 zAM;nwH;h)g%<{+0KA_B=sgIv4f9<{cKKM=Lbf{h_y7xw6<*)Jz^YFm%HBY{e#XoB8 z@97rzh?SD&RlSkbf?bQ9<5e>AZnR=E#`D?t2|c3$3(H*CFsfWqTN^dr(?9W}b@5QL zbIm;Ax$e=?S_X4m zGC{41P9IzI!M)k>!zwr7S}MI|GS6B~xlNp&70OIuG+*!rm#FH2p!yNfAG!ERm9?ql zx-Yxmd&;e)7G7LF`VrGnZsM>?#B23U&T1T)IL5sB_(A--budQ|Rq6W^Kjvp!rmd#H zMoi;Rm64xo(?9B!CjW7kV1C2_ZE;xjar%FU%F1=N9XF11H{1Ql|(5#oQ?VpN@G`bR)iZ$li) z95nvZ3uNv|X>ZP9pDePKps;m6Z5K;#{efqF)^b>>jo|rG9IsOxpI?=V6VVGNf$3`p z_R})b54!Ut{p3pnG_)C%TsDE)E0V^WA}LyDLEChpn?;j?<3H-Y0_1+1o&?_=UZQsP2&K7_4{I;iRqlM?)2nHVkz!2Z87u_wR< zF!;YRF>Bd3@8*Fl-#;0($myzTg8q~AyMT1eaqB?8iO{HV0tS=Lo)qU*cu(&l5 zmMAYyj+@0ldtIj5;OGAp9T&zo&DO0m-@uT_2*D~0LC{);?paGS1{ ztRuSPy!ffmfIUD&u6a{4EULVKu$uBK)h%@2Ttdf{)&1!5|DE1l^{z}Xo&ZO)TN2)= zFcdfA&5+JgK*6ulrG((XyK-%c;10gH~zSPyKv-glcflo_P?|-Q8GN z!kjVXm4us$rLavqXNFMLAI$mxYkXCfcG@VO2dC}X-$}h%RALFehOAhG$vQpT(Zr|R{VEj) zuyav{vnr(m&b=V&Aj0{aqGXTUD$ejnj1KTvw)OorH9sO&7Qs@0qea&Wmg zE+C=GCFpeG*}k5l8EvVb_)oArznM?wwo=Tt+hQS@q zQXFd_w@S{(!`gpteo%#ds!kr%f_+%rWH@@8S62IB*ulERiijIQA;%BM^d4!rh=qfP zs4V*iT(&##oW7q}W$|&bWL|&U7s@;z(AZQCvG3v+Ne8FL)&FSoxAXPqT6O7{7YeHP z2bI_4Aiw?j=#~UJM5gr=dj9?@f2#b_rHb577T!UYp;_9ApDl3Q^=m%&n`)c&u{-;X z?cY@STR)kMMr~UD##CM$Gh)m74@o6?Nd}r0vlJ_uvPkpNm9bX|+>udXhvON6U124ds>e z5Ip14J@B@YE`-GR-|x6Or-hP;J9l5-#G+a;?txIW0DBJ;tGn9*ccfP%E5jGajfCXj z_wZvkz`-;wr&1Ab`l}EbuP#<8wAWT;3Wg0Rpzn7YNC@LSeZ&jNF&yKWv}ffZkM$Gq z-?wr;vJ9)s48<8W4 zjNVi4o*PSINvjGJPn5dcn2xou{nBPM+Y*HMWfWxUxYNi2diuAfuGfm z9>%4-$&A;^5vQr}7kl1~RsQy^_~efRt|WKmOw~K_4;)*nA{|Yd1g9lq^Qd&rc9KVw z@s~=<6$X435fHOA$QNVU-`V3JbzV@TR)oOsk2q#ryr8X|RpntC%7Hye!a*Xew{KYM zd2pqM)OEabTRt5`RXn3x(E4inhV>Urg?F7HeHA%Kb3+|2cQJvXz=aGK0yGK=___4c zWzj|dw74K|t7a#>ONyEQj0p>I=_i=KEurn9R~5gf+%VcLHNkd;s2t@*Am4LTfgu|B z%?5Fvd-nhCV&xn(7J|NAu}Dqx>byfN7z_|Z6V4}V3W`sqh>Qs`Xd?`=1Dhg%P1ZBU zt%;Kg9vjkYWs&j&t(L&1x_M-cd3lv?Ye%=HdS%#?rXwiBc53ox$qLHoV+%<>;|LHI zBKM6IZW?EXpR`jQ{`S!qJf?*~%A3X*k0YAIrK(XppX~5OO3RNLX633;Neb9$S4-sr z)5Z;PX@VbZxi!rbJ{UK?f?5-8W(qx| zs@uw40uK?W909k;faKRhAbY*vN_5@B+I6q8h!ys}n`qwUpE{PBBbPTH*<+*5!Yo=B z0Aa~iN6uenbh6xA_1SHs$%H>mh}n|3PlU<+fE)mg{x?K2H?)GQOH*fYvp?? z5y-YN=0UrJM%`5Vh+h#j7`IZ*h$iF&Zn7^g_TcY{ymrO5XXD2o6{YJ^DF$nIo)Q&%{gNy8^qgE82+y4Uo%x@EQQpPAKy7WG7TVc zETj~ap|rlp%@Y6ENgQ(rK5W9$N0tbtF?O=ZHv4rx#qBLDw_hYLdSj`@4Q;on$%x=+~>5oz_339sT$^=aEdz!dEZ+mrv@>(?q&A$x}Oq zxpgiQEYMr@Upq!e(C5iP=xtWA&T%!J%k-($9UjcS{g~m5kGF}?=JZRyj;2dZiuPaw zvA)ik*!0qTcfz%o`CPpt8dus4bNm+Pf68EKvP! ze4C;4i|A(jK!|wv`%0tKp zXr|r)YUjbu3Ss$Uz~b@?BEyJ^*dS?d+!_1;+znjpjqv=Zp#7)8KF%&9Bd zTJ0OSWL~&*8z2)eTv`Q?)vPZC4eg(i=H z{f>BTEJR6EAC}_=s(SmIJAyMJIwi%0Rxu+^Altz0fKqioCMP>Imq$L?)Hlm8zGK-h z=Z?NeI-vrk6!J2Pg)P1kl~8||@ZZ0nM5o(Akl5y**pZX?xg)U~mDqci*iWA{D4jHH znl$R4G@g?**^%@Gl{9mgG)td6FP;3&G`X8d={QnS@6lenW{$E6eG!_lKC!pd}Hho<@6#E~deu6ZWj3m;FyZwI< zwYh8}gGca2W1_iyDqG~$$r(z5D1jv`Et>I^8 z8y+65^j(8pEFmd7d87Zu<`|rVOhPlD8R*XDe0dt^Ro&lUBI+JP5?Hj>Qkv#A{k>sH zv8VbYWqwbq)M>=EGVui3Z!tT1A{>)>rEWT33^?QcVcM}FPQ)AwY zMDv>~`)u17UM>g8l&5ga^an$dwIE84Er`dgRcAe{(IHj+P-ImPKBX(dJ{lDY@65E=Raa3od8Mvpo4`1+_)= z`fKpWF`Lo+xYfRvB?UxYqg41;Zs84hmKUUoo{qWua_zF4;SFRW7Q-|RkibKJ<14q* zRyAf@1-iBy#}j?Gkoo=wdp4bg=w-b>6>i#X}rBQ~RtqEXN5S&jmFOfERJ7N0< z9Sq@;oP<+Y_pe5^stG9Y_e*mpyl%ji-iRK@y$b~B+L2xRU9{99-XjgcK8jc%BMN$i zUC7ioURr7I1YYOhi%}X}qYFO%s3p}@!_=k*0@9CgwEk`8l{RY4aewirMW@_(e<}rdXp{_V3|T zo-0V@dNmb;dC1F92M82~U^8!gp0lVE+>zZlevQDANI0Gq>l>p;3?%6;#KJP}jZ;2n zV=yZGH&lA`ICz7o&SH zbb=)sN3>~2nF#6Spe1(}rY6MkM)fn!PXvmA*&b~=E3yf}ruvrFSK^qHbtbCX-#LcSDd zDZu&mcV7CNXgGlZk$GpyC!EfYwxsh3-k##@G{`tmy(??mPD-eAcY(s%1m<(_QAVNH zpw=zgwThWhB8OnHGJW;~*WBGg0+3Ac#`G0=2}*?(7zZnIOoHnyT?EM%%UOt(Gp)as zvRsR}zR@u!C&#^vC`zifNT)7c07XE$ze9avd8)#&@;0FwewOjl^I}p%4>6xva8=de zs%sI<2c$fAk(O3*HS*wFTzD9I0(v~>`q{^zzW%D_llQpdOTqD zyR{-@Wqn4kg2<{oDlU>{nDJ?)#)H7CmlEG21h|`L7iW|)2>sRPP{o{&kK9Tbi_dN{ z>kCf}Ts8Jrv(1-^-re)KcaD8K{_a}T<|quB2wabtH461NZwOijp`TVjQ~_sMK_Dp;iUc9Y+nkkL0tV*3v!-5#ZF=P@$WtMwkbTB6H-e;W~? zpbby9gCzzUb<$Ry3b<)Me`RBftCqu=qBPpf5@f_M;gaHEPhn9)HRgnoOuyE$j33gR zL=8PL9p(L?O>BJdgk}`S4i{d%LzWtOAg0u&mY!DYD7gK0RO=|(4|f(GXqr1eQjT=S zr`(mvLssSCUb||`Y}*8{sG1fN%dw{sD6nEv#K$QYD;hbeh?`H3`5wpEOVDn$aWM1) zc&w3vCk{OK{KLDuu2(J2rI#x!vr@b|Sc2q=!V@eL5ReF5@x*rQd8cvpb;(%%o|>Az zQhl`=I8yG!lS+mG>81M;N$JQW-_=l>B9-QNvWJ>2rE6H15(z&&vsJV2cGP)Ozg)X# zl?aje@kELopdhj5z@`^k5-$QPdl?cx&(d{s)Zq5M$klq%~VJ|Bt&p^WnC8xuejxPSjF+s0MC zW9Bm~>kBw+_Wk}q#pkYAYvQnum>0}s*d;3Ut1sGnSe(ypT}CfGZ6O@N0xWm^6#82+ zu7QUVWaw$N;XDlvdlWm{9N7Fu`!TC>8D1#k_88Ks3MYv7fNQ+ZL+@0ojJGveR3A^G@zns{*SHA+M;z| z`cJH{=hIF%b+snC$wXGJ8Velj>3(0slB~(g67N@v*F6*V3YAkMGHw164%{y-7Y4>Y z?JNf|2-Cf83i@>OWalnb1`+Ui0exPG`NHi#gC7#b?N~HGRX#^!dBOIH&2`0uTxilv`B{}f_#k2o^qv`+fj8zf9p&&Bi2`?UqC5uqdf_QHZ|G#*q zmW&0Kk~M8A8cGMt{iAC1DewOvz%uxWA){hCm*vgT+C+=r1+rqyUN zy?VY}FK>`qqZI)yQ*I%eo%F3Ssl)Z?84hTHY1QBzQz)~}uFJRUZcpYj(FuL;@|U^) zZIasXvo{<=M9NHG6%Ke{egyq}+hov@P^jd0GGD$Jig)O?F8i(s8>{a!;Ym+x;i2`l zPUTswd8_+U_3?^<%sK5rWuMovacmY!;l=W)f)NvvLqNwIAtdh7M``4M~7S@i>ocZ3xOTWQZEpIZ5n8@S}Zg}9pRc>iQD zdJ>1YwzqL(y?jdUB(GCl1e~+Ea;ct%LrUGzi}Vi3Pf5W=IeAdRxjY)535tB5XaBjU zb3ye^M$|5EaWuALd%nY4ia$9{4Atd@d2H6|>8Uw0t)-3>prdNVuiA{;RS(0wvx5B> zJY4Xot!igQJW|XKPhsH_(UGkdt(#Rw&;%b%p*Qpo-v`e&LMwMxKUd12emu>Jy4i%! zyP@uFhb)S3q>crYRi6bMG?Ke^2_job?@v2**vokwMyDnD-_Zq3IJNFFK)%JDc;q4~ zJ|ZBqsXLQuygfLp9R;RbM>UP9??^Mb1C}Dx4Qg)=z8m?IH0%7vuV1K#u|5M z+}$CBrg3+d;7)?Oy9Rf6m*ASQj?s>+gG66d=`O{vtYj}s26U~tVPF@Th)xqH#+xk zI9yRQj50>)ye@J^XY1Tx*^BB=$TPvQ%Be~g9NFDOzk`oo29vcaB#Pt6Y;u1BYU0qY zqU8%m#%nuVfi?|YJjDV-hGiJbWIZBVxfnmNc+8w2aH&oM zvcOqexkn|}9>3##kky=Y0C$p>pi^g5#5GRApi8f&Fm@69*s8#b+$|Y?IEv=Ek&_B4 zQJv1sCp7dXCR%DKrK0lgO z!7A3(GX&%KEAYpNDTZqRvg}0Nu7t%m0;$_+3)L8E&M z;jeTVveV677M7Qy0%Q~M2WOQ^YBLl5jn=7yYfK{|h~qpi>Sdz!f;Kn0?9r-d>KtPY zH?;74TCK}y9a(o@-)IY;w1;AD=X{werTW;YWGtwY%*Cp&X!4E-#gaAJ4W)_mNNGmZTv z+<&f!7I7bFVqt0!mBxf2Bof4*HmC?2;^m2-O^jOcuxUF=FSU&?$Vs}3jKA~NcBsqEi|8s$Kc%QKGtG zqMyW(gYKzpcAt7gg=}oH1q12WQ{zx0sI*7CS{Dl(JnzrOA0kZ|-L^inK37n$$RPy# z0yg&Mj9|1EgMGhL@_G4Fkj71EGzCR$jb_Ua_N zN9PAfJ1t%W&fuR5%NAw`pEeq8e_~YYJ*EEXUwucr1Z!#Cz_)UBuE_&ycjz-zF@<6e zj`e^?mR_zjxljFez(_ysSB3@)QaGQf?mDwV8k12|sPU;~PER5x`_79WqxQDqq;`w3 z)l?YB7Zv4$56{!?n#lX-C`3kL>Hjcy8LT2PLVl-7A57jePU6lh&N#MMUlha)H<;8T zk1Z)quipgMTbNeD-zM*vp%T?>e2OD+j3-uqH&Ap9`E%y4t*6*v0t-+GGS|}!U2~XW z@q_JpCLH?yVh)oH*F~cHQ|$uJdG@ZV2!podxTXO@l1b6gH^Lpu!+JYYDs9!{>VW$i%x1D0^rKujgT(?_ zL%JAaDZ%GWwx~^%5xHrgSS;=B4t73T)A$l&Ah2rIsfmYrfRC$Z)V7GP{SYO?7d=M! zihGzuBSv;!$5PKy9lP5A2|)J0v|E5$fDu}H8^fpwC9C02kkwWIlI-uRde-72h?vol zzNDmG)N_LI|KRc2tg2bL@pB!dwz-5|jIxGlf>6u2m^j8nKbYGMbvB(t{V_2I9n)J~ zVw7CtR^7xl8)b(wg`Q)=jMhXu536|~?a-8h9}4Z*I;7Desi`$-YAWeVZqgSF+T@O; zHsa>(mC~skBXYH6dg+KcuNcx|)OmYfDO3>(F{|?VX9OX4XqUnh zBh%Hxv{Y>#GJ-AD5o1b~Ya~-3Q>AIiw<6)s_8Hq%5xgKumR>~Rz4gj4#ZAr(WAy&*^gAgS8=?Tcn!U0Kma9!K1@QY9PNIdTDVrioQK1Zf<4 z6(Y??wL~T-`bM$NIe5eA?G!1Ah#6BgiAJ~apN*0Qx?EBa!y~>b$P$1$4A>9PvTUaj z<7a}51`(FWsw}-$$|Amj6IB2(mb_hBNokb3w4xNqfUx|I<&4cbIUI{p&@c#_7t-&` zTFb`t>P|zMEvwe|OzST6E)|*`W8edU8Hu&uaT5C}#J3BxD4#oP3>DZ-i z<(|nSCdprezXvm))HiEoNu#X{qhuuOXU$+^hvHd4RayZRfC~3Vj>-(!JI~yI@e;#Y zDFMrPim@flEVk3I0vbDb<%sFPFMi%7=6e7UxU>=-4pl>Ll2|{KXe)b9;pYX&2$qnS zDG(a3`FJdq;>^IU78jxvP|C?WI0y+6E7-^`U;wm*dhmx@P#UORi;ge9ke7v{Q7j9! zyP=TTWF-q@Kb1IbsAeWpGyYdjOP!k>JN)Ic_64BG?gWOqgCG`o7OS+UQRy>cd2_vC z67l$FaIT6c^8~-Hgj0wx8c^)L-EhO7Im?8+@Y+wTp9AILB$KqKiC|gza zl>I|liar}dBTMn0g~_H{kRwWU5h2+tF*2YV1o)v6%@Vx1)9woC(EHo^K-a1fZ3rB~ z-2^mR$>CyBb|xyq0g1E`eTg=S5})NjXo~>3Qk1|l)3y&EsGUPtfn|J@ZD+=1qdIu> z6NX8|*4h$b+lcxmtNP{e`aB&h<`WIzCGHzuC=Lt?vH(j@b4W>4^lq9V0`f#EnWpqy zH%l>>CHfsFOS_5_ZYfWKSX2}dW8wEcgM{QUv7@9| zuVP4oM2$!KUu|Rw5y*gV1VxrhtNS$uP|B*~`}r`vTv@}qGqZW;q<_{G(@Pk2YtTf7mb{=&A~17J38?c%9J(8wLyuQu?dL+?7rS z&rOB|wVXN2p0ty?zv@dO&kR7m>IozwPlr{@oQ0_6E zMMaszG)ivb6fteZFMu0>7w3TCb5lvA8M5yBA7-Y5n6+CG>xG}czoICKdwp;NEh#iE z_xrMy_T@O~i^cQjivfI2g1%^unp?gG*+pDb%z3=B#NW?s4T^fdIEDDIM-R5=(bV)* z@@OZ)xifNEHtS!|i#lW0Mu^Nl`PqpWZx@-U&Vca4X!b?|2^On$Q$F*}BNL|Z!7xT! zqHZ_5vMVg8EBQ0$QVDX&^?Sfu@QX0gKA7=|rB7vffna{Je4&~}2JeNF9zk?}2`(90 zvA@J@DV5a)u5d?wBFb;tTOnaM0fQIn(KG>IIJmoYOb4YzT6R?du{8M<;v7A^jzZzR zKlFR36J2x7=LMhvU8O}KOnc<9i~2fixqNmquAG|y=vi%YTJv!Q;K9C%OT)rknTP2J z#5WddCmRA(iOxl#_iiX?rR$}^0C4IWyVlTKstvC6CjFwQt|+=MW9wxv8{XeKXSeY& zcB!M{H+{>#O3yfr?|vn_!u37}$nyf!+#uPd>v$FbvxQ|2tC2cw1-GWHO1gFbxKA6( zeD`ixC7;)8ke85qH%P+EMwaF#n^jb=+BfMd1{v30IkifK!3`kYws$bd4!YV5*qn3y zUSt_{lFZmp#J$n$SAW;XOgC;I#D^N;f3ExlRqpOk?w7%*;kytZHv z4sXk2@ETYX?fKh>4xe3u5Yn&IpKj$n;a4CZHNbGg&%|JGKoh~s`46^_RP;OFwFU;| zB{i=l%exoAXg@*r&=!?oOf(CS8MH;r{rk({-b~ph=X<@6TZ{Xk|Jo#$h8s$a8>Hwb zq{I!JM7?^Rv?Zl>ppZ#ky}XsB5<}&VRRs8|rVfzR1pnq;Hqi~CA30p&qN3NCh~jP! zQWYyi24j`DL2@oJ)q-)$owm9l$GA&-MBDj-sN-Mhk6xiHOQ(I${AiaTx(G^ZDp$$Q zQ(ZI;SvOYero5VOAUZiYIxMnVLSxjv&L#iCb;`sJXy_AB%^o4(pCUyenow__3^_}u zIgOe-t+4zBqdJ#=bw20JMRoPz%LN-qvESM5QL;DtF3$b{u`>+-eA7jdVEp(_BUGSCGZ0nwd@k}(vT>^5N+}wW5 zSf`wL|Clj$-f`6U1%7w$L)p$ya9?+JD`NXlrtshrx@=JOP)l{|YqazJO{(F$;|Qc> z#_vS!=?1pk$B%yZN%OwNNF@l$BW|EUsMM1s)x&)vM}4R~q-=Ql={6quw2t}|<@%RY z_o<8fwzlx8%=}S)RMxiTFE_zow%i8UEvt!ll|Q*Zedd7&{`l)l{1V+Ld_OzdWcva{ z?$fqky;NZR+ok=tKmT&_$MVo-YB2p@^9sk`xBo6_Sw(4)H!W0{si6B3&<$>|3|NvB}ZzZ<*h4ZAHyPkG56>SR{CcZoWjm6<=WPV{Zc)oq~B+gTxfO zUYNHzMNsA>@P_`RN^1}%bDrXeLJtVvBM5&#ot4%;?YIpA-n`;TNNUz;l|QGVIupi( zN_?2CJj+w54zXb5hchxpXvpCoVY!d}#cdl`VC!$wtR@Wy>Zcsz+0&S(Cj z69qL)wF9B*m?@*qFBU-A7-|i=3RT^wfGxIi>%eG|ih!jpHXYcqOhgeQ+l)#*Wj`wC zb~UFio3SSs0fC4QK#4hBmR_1UO>_fOTSGI;jGsk&{=k4v7l+8g4Bk3M5bsm#sXQJx zq6%d{e{0)Hv+i@=y~}8O9b4$LzTvkw&wD{6GIe>LF_v)GBx=hj^hBcVZboBZo2L+< zKRp8^PCWCnORC|wXaZMcjoV0((e1!nJ)L}P-pa>9m!w!_PWcdv1wwO zq<{z!Us*R<&C1fA_e$d24wJs$HDTn8|7}ue@A>+6pKfhY>$k)HvJgmviA^dxsVN}e z=|aMNoKy-D1_CJp0I4)Jh3;*$!H$h%oJYQ%D6e?fZsrw82fdaQq!hir_$Oz%#+MV7 z7JnFdexz`iR*H% zqJs`IicZy!U*GAudo8Q%x@gYVJd3Hb@x{>>;>NsHF?1107;?;u-hd0Q`{{V*9t7f>Xp5HZA9G$*#ZW%FWc61dY^TAJ3Ug(n4 z2uCB$7-3zu`0(00#sIYhaY45KhJpT@SEmPD0G9q&7|3+0wQMvN!&UnK2m|TVoBppb z(0|;f{|g3!P%CF1ekxbYU^n}JU?Anr`qc)*cBgMMoegU(7NZH&DqW2m?GFFp)w`Ou zy506>id4FrzcmnG-u3sat(nDzfbRlC*Q+Jd{;|>s^*sDm4-Eu^%nf^Xeog}^MT$+{ zS=)z~ym{(B5_PZ&e?f%4<_ha{&dXKF`4*)82mZA!ap}wFKpT$m&VVg7x6~f#y^#zK zgI<%J(*$I>N{RYl-|gY7%C23{izNZd+GrxOq}a^fc7<@Vq5Gdu+wuj&OH<0*cftO} z7sT^5|2Uq5GcTERHJm_QXp13#!jJUALX zPGvEaf#&B9NmrCt$c`j=K@h~FpC2$q!xG4>a?vdBzIk7pEx|L;1pVQJ<{j6G@%xtD zt8u`y@dLH2Z~Uo|QVAH0;Ssok>$W|5Yd_LmY23z8Q^d&+GS9pnmMp<2;^n!85xGYt zNk+cWsBeWI!wT?tccMA6>Z91y12&NlI#voV<<`X)s{*yDb?4#^4u%8@Ou^M{zbXq9 z$TVfi?{yuI6peSi%#$StAtyQ3Zz&GDO7#zC@>VyroeB(0wVY6Jy|k%~FCN>DR6oid z)>y=l)H#<8LX26(WW5_h#cDW5LtQ(Ko@AKqmsK;6c9T3~-|$dRJX*=Y#L z=U+Kdb`2eI$3^Je!N_%MjRIGK%WQ25 zJDP(yQti38ytSP*n;jlMdgcEkjD%5y!%FWtq2hqN;jH`xj-oSrE z*qiOwm}-TgfCM$+D5~&(^Jn_iS-Jh(S`C^!hBSSDTO4#8D&MTMyPAc5>e-8CG51=M zV`=p8(;wwB{dS6m{|;(?`}1LG<393xx>Bb3&%AR7o?M zp>7_#Wtsluru8*EiAD#h7&l5!fgf7|!ypBX#50jF#-oMrs>t)9g}Ye6bUWxt8d$?P zAT~rMtQh3jyW`!8X+-GK74VKQB0$oG!{E)^M07L+5_=SX-Tn_?O@5pQzHsi^xBN%bEELoeiV-WuS2R zWrNd!qqrwaB$o3GTEQOn@ehoSf`l}jlv~gDoa1cjMvyQ;g)>{;_(y*HDjdbc#YH|2 zP(G%kH6`sgG{tUI4REJ*UVp^l5yrS=a8+Ae163`&Ta4cKj)6REiVX(+uy0Se}4g*GEhyK|JYc9v$_yr>${A*;tdqfi1pVr;d2g#xux~&QWQ?TRTbFm;$xlLG)B$)FRvN@^tp-Hh#^rIH+KDmIc3h`X>#E61obi2g4HKqF z`=xHiP_P!ymeX4lUW@-CzL;p-QC?T3X2B|JQuke-Fiv#F#FeteP>8gxY>wsiGPwzi z+90cb_0&Up#i)LS7+SPA7F1!m8CmI5z2Rc(eA};X_i;|TzwF#HjgMaGj09uR@SQ%m zO%GBaQupbJNA-7r`2bo+FL+~5w0;ve)WwQ&-;aZSJzTSIbG84-td>iXSmF9G@e`5I zyF7TtP=EsvnkIT6T27)T3yUPK**BHGX}|y-gK;nulN{r@nNWZ&H3dg%G|p6o;c45Q zTm#EHrMGk?1id}n&bEI#ZBW!O+)7YNY_>;j&*VJHv%T1Ihl7kJ8FOerK)+NvRZ4TI zeX-l+>XBL3ONO_Hf}YohmYRKW#ISh7}lUP^g+32#Py9${hzYtKS#EBX#rl$;SME- zHF2Vjo2BbxEw?sS>2FoPOMw1hBu6n<@&v0*IcW%-zq;`K{e0@$DsY!Nfz&1DNS?)y zQ<%c#WaHvUUH3)Ob%;zED_|L%(yk2m#PWLfBL77~wKR5X3`?UiHZ2rLt%6#s9Ud!# z&SPDbwfx_p1i!#=;^3k0vas{q{2mnSH(8jU*Fv%Q-VDJU&`9*QY!OE_+HGG0P|*Wn z7@^gActfe!@5XV@0mf5d7{xw;Po$2t4-x%e$@kW9g%0udSoF5MDPJyMeRPX}touoi zgjK;%KBGsWUGctL>gW0>S5!sbD?oq6IH$<-NjXGQ3L4J~gfKs_RoO#7GOEqrn0@;b zP_q+a7v~2;!FKJ8H6_D-n-$pxrg)rmZTE^LydqzSHk_}}i3c-aBT=GLr<5BlV`HYE z3t1TFns|%SxFW22`)-;QADlhNVghdcvM}&dKalPLFg8nZmM#-Cv#@!WV&rydy02AQ z6GL`cfl0BIZ4qEi8xn*sM1(x|1mpoPvKafQ4OXjN^Ms!Z%bQZ-18Nn_CVYE6`Rd(W|cg%DO zZh8{#hireN%$D@zj}cNH{sqfgGK>PyJf*D0G8RdU9NNtMQu4G3dmj79!m%+M5MOQ` zxn#qyWReQ5h#ocUt3t$nR`F;mpEXnnk@swww}TYRiV=;mk#}@|ER3xyL~E%L7S zk}cMy^~%#xk>*Ge6};?bbvPD41&aZj*%10n;&$UVxy563MYllJZ-0t0b_&D?0Tr!^1C$_2(h6Dmy%uAOx0>i zuP6?L;PZ(TmOR${TC7>!k&02K_CnM?Rqxa!J>|zM=cYh>-iz0f5!5|7)~!9lVUXhd z1gW?aL`h#=mUw+xL~Yp2{SlN(Jib>JX_Eb-fXec8PTP$lp4mAwqBYM~y zn#FSf^<@QxuI>#`VJ#rCrL6txt`rGmR1AwO{)hzb2IZd@Pf5YAmU5+AwaK zV{hK0XzGb?-g2rwQ?9-+u0Ew`VWO*X(P{a#N;0iX{Qu2y{09{5{@+Wt|3rA*i12z%~wl7(l6h4Kbtjw$KZEavlPuX??_^(X(0|$ZTSpWQZ?tHSNoIFDU21nQc{JcITt3gb24;*T-iR zz4dn%ZxRs)kdG?s&EL#lT&m3G$gFg@G|r9AMY7s-XI%xXCxS?%B~i3~Y0)(G3jSJl z?ap_d>+?o5#hlK_Qx$NTn5cnMH#tgkhCYREh1gC>@K|l$qOEa13x7C0YWBnfOZK+i z&aTyXn6U<1$uDn+#3EDrDEc0cXY2)NlL|zBe*!vbHQ$rp*oS_5b8kK$+G{d?1nya} zPSaD!H{7*kS@jXPm79Nu7jZ}`QxYCa;e(#vwJHhp@ypi#SgSy8I!&y<>)QLcui+NF z*i8%@8p2>8*rV#!G>^#=-Hp%NsJ4zzc-h@a<@Byu(!;c?E=}LOih)Z0>uAT})ukt7 zAfRU=f=4U)h!yM0#*i-?zr?eeaNTx@ESTTo*fVAEYAZ1T&p00_^8V%J(S%vuIpT){ z6|;*BG3?efC6=b@Y761A!|Da-?wH@x&CR>YN^;Zp6H8})8Lp+g-WAql6aK&o7xb7R zt0>f`A6Az>AV=P_#dq8}oW2jTtLN}VC5~e>%o6TREu7|jV$M?B&Pj)xKwz6Ws`|#? zn^*Xw-qfsL*Hi;q4sY}%<0~Oa21qKbqy`yjc1;UgJ4%!0)d}a+7BYE>+A=uZhcoyT zjNTq5^{B&H;d;ozX1A(E@x9|(xUoR3FhXJ1p&gh<$?={3&gcU#vy}#@^I71s@Eb1C zX##x&ySwA1I495G0$P7*209%>g#=R8B2jXO;OyLsNU8pE;V$Br#{i<*1}|UYFt^y$ zp`ILt1HOToFFX{r^k2bbAqZl**-50H-otFbn_$K>=r+ zn1yi*NS#&y0>|reQ1;yf@0Jy_8PtVcT;nQt2ZM`mVQN35-tIoyWa?b=P;hcI)5U`V+;$(l7|7 zeNQPj^zp{t7IzB2IL;1O=RX-n?K*3|z4ZSpvS5p6U|jZ1C&2G6*NJI$tlLSSF9B7J zCPezVMT3!CVrd8VG0Qj(SR;U4XXS<- z>Z7Y>hIy6zbE=!OLyN}`T=_F~+pY+E_#7txl!rQ9lRlI;AN?s{>q%FU@!27o>vsXf z9I3pY7(jFtC$G9Z4vZ!m64PSU3@XPjs?9BmZWj40rtXoiGe$?hnCQ4`sp8|1&QO&S zXt)9S?4lqeJfkUeV7)|s)FZ)%I~@_~QQ<|Pnc9pwmI4G+!V2i8B7bVZ)bq=*kq!z{ zO^e0~(imY02&IZlzH~W9o02a0M4AiOnimre5z3PUV1%V`^D(g*KW>->NUeKAj}P*XgY1M+qvqZSE*H&39q}J1rG{?fiGomYG4Sb;jGpE? z#WNrWe!Si$KGX(` z(kc2AW#~H>6E}$v*C42`yP_Ao8mAeZ`QL~e+J4XCvr(*tmr*SE<`bvA%}d=(H5Q9p z(5r{Ly1c^cCu{gBRO5P^c)@KAZD)ihGrcs5ixrTpMv$b*3b8*^iBR++acwHq)i|6@ z$W_-Ugy}dbToJmzwtifYFe_ystUMbw2Akt+i9IRIh&+sHXa4lP4?c6Z-QCy~kJosi zY%;(cDQipZ=-E=M^EH@-b6w$3QW@vWL8YLx#?nQW0IflQ0LT%H2m@jpBuyd52bYmXfh_J zt9Pj@_QoQ7@QQUoL+RioFR(z!biz!C=w5Toe!@^7iFuvmRaxY?Q&+wp@*9OC(Y0|} z3@ge7vC$E-Bsb!^32J?WAalJ~5kdlRh?^U(F9M>;zeX&5_PY&;ELGXI*o(NqahR85&_1i#! zf$!MwazSlTY_F9qV6AhRKuFuwdn}8+wj_~G=*fAUSIq(`QvT6Nj+-aKDsT(8&Y>#D zF*OIBK&SY->y-L-W|9upI&W7+6=k}ou>n>d@K7L)R-Dm5eJoE{?MaB~dTLbcg^?!D zlY^c&onK($ASOuBm}|~FmvHRR9xu?@KWLjT&`cFZC7|>@L|mg`$#U(Pd2QPg%NKA% zvuAqgoIiawF<{hu_cLG8=Pt3Kg!FPM>RE~rZ@I2y3Fajgx2H8y;~X-KfzCUGA|D!M z0QgR;@?OmIx8ENFQ@Wdq&@hy~OWai+ufqx30{&azL*PI;0FUhDSZ+SbMAm#&_S3C1 z*ek{4)iBCP>r$wxQ@jSFU|u1rdD&uDuD*0S%64bgK#wgkQ}J>k>TL7iSR81hc{gP5T0uAYg zrREk+{s3CvkHjJSuT$_C`~W|ELq1(Xl;UF86{!OCLaxvwr*8r)3_vZH5a)G3`C6<2 z%z_icA_g3_NEanr8~NrEqW2RM7#0v zV2fj&@CU-4oWN*S06keSc;UY$;^F}U91KAmtnq&`&AQ_fR+iLe86y|hKqaD)Bj`b4 z4&pROHkWA;IJoXcqJU$ zTYEl@lIj{0j~7sr>PGl04t*%}?~;jNciPNa8g54XtQS_1Md~{@h`k%+%|bkSd}<1# z%{j>Aj1l_#C&0ih-To4j6v>b&ij5KZC5Kk4XdZEh}WwP$DZSK zwl!Y-uE3ryvu3wo3xAV)zhpH(txHm#m!CA4Oy1e+KH;nR0a*rphFnGlmOVW-o1JF4 zwr0KLdI+|-zlS-KB{t&mo#eO zMJf?c+81;==1J#C{*CuXzbcW@_TzRhah)jPw)7%u6SYLc|67?Rjb5fJLV&$VMBGP^ zu3%_eRYnDI2&E(H45PgTIPBd@*4{(4duRzfpalql8pYC3rE(@ZBDEcvejlPoOEK?k zJAPY1vXh)Xp%U?WNBe4my;7V#Xw?=k?m4eZua$CUhFw~f)%_mj!*HTYcjYvTPJ^b1 zow3N?CF!HPceIXh>~c4Pp?qC0kVU~%no$}V#9fH z_git?=~?0y-YkrkK3gM-cS(g3Y7@QjMDi^rX1q{Qu-_qYk-WG-7*1~j#+{oK_NW!< zZUgfkWARMRLv5uvq~!(9G>*XC2LS$%vhWgsr3Bv6>FK`ZgO8N8GVl|95N}i_ZN@wy z#X~f^Ej1$HB8?FkA)88~cMYNLqV>I+QA{6t?Z-?i zZm%wkad1tr!#v4%S8jwTmkZK4aGv08T2wbIHL{Lsk&fZtb?B`lDs^ls5+Rvl6y4-( zBDbmC8WdH3E2~2(wcqf!=lWVbP?TUm9R=7$URvopB>G?A!gd_J?qBt#5EKO3w0uup zZ71z-P5fS6P7lS25)_0g=c^Tm=|4 zd<0lF0vsBle?_2UF(S*4rlB=@pfq|QMzG5Ul~6^)SS`Y58MUwS3t#d>B#{u*TZh^- zok~!Lfl3tSWK^lfm85ACJ|IQ@>0RX`D|L?hh9#pCcNDarM{O{Tb zAemLilx77yg$6T%O7wbfBJm&Ecb#eNpUOF?=a`CYe=0v0;2?MY;}mN|iept%iMO*Z zR>)`kGwsu=(LvFuKoqfeT15EBBQCnREz!e!ohg zJH%X%ygvCh+s*R^#r-H7{QUd7RA2zkk=$AL)k-h$LnTf{zR?L(aUP<8hnlA9p*z-&fUpo&)>~2&iOw#XNM%vX}>?7*@lJ(^5FCxG@l#= zXJIMAGWfiXtMI-u`1x0SGB4vuL>|5R`0DWTC8%nQJ^14G-g~QxXkFu{4@xWBa?X@- zDc8fAlHd>Zkr%sCcXSeT1B7>{uU4cV%CVsb`l>IXW0X={zTB|jE<}@Yg|ioPi;=DW+%oNcja5Ky&SWF41aKMSrX9jfetn_gvmlr_oSHW42?ji zabxW=ge4RtA0@)iTpu&?DG4_1LJ%kS5IW~m0}Z* zkti#D`^!s3_Ve3W&P1LGhd+^fCw}{D4Qx*W!E?LzLK|rAoeb;d9C?~jY^WIo&{*ni zwtnsm@%B$|&ym3vpG^!h+dsB54;Q_w?#lbwxi*pcZZR{P170)D*w`WIgZjs74=RR&eL>OoC~m15hq-K(?ie z3+dTtcKd4YxJ0T;(dmfw0vV{KsN9x=JqFvMy=XPzKau4)8XOP*g1Y^eDE$CJK)t_} zxt^k`faCXfyi@gI8(&0j{fhznHYu5~Ge?|wjIoZ}+SK@F{*ye|Gj2KG_u6J_I)Zd8 zYtsIRb%bmu!*)Fqql#NL2x5q}{r>8vS~{=?7EGW+^)}vz3wiNuW}Xs39(7{vcbp1 zjq{|t-l}^d7QmeN0MnISx$Y2U=hsSiN4qM81`DMA?UC0=o=uUxA!=I)ejm#l#JHx- z61FNf2fJNNDzd|*gndm_vqt%TS(r=0nxLHJBTc$KbQND+O!QvSO*0_*W6wxW zNm(;3L)7#Ret~)j-&L4=tUG*vU!k~X=$rZvhak**drexAC5M&feSG2y6Hh$moFf09 zh}hBnkh@1@J1#`W|K0L#DH=N<0uoh(hLXa=no%6{WaV}1S1Ze)tVPzRQ?S*z12^sc z^n@|3P%@TM5p=VV{<0i_k(bn8*|C(vcFmP^mc#SSoAtghGtY9#UV-Nqor*Uf&uX9s z!iqIDKlKZ(f?k}<0|Qg&b3_%IR@hhODF##HN^V0YMd5w|BFccy28(kD4&F$`l8s&? zj{An3Mmb?+Zagc6F)&Uy!kK3`#+-cE(rz0aFzHtp%d^;c`J24%PS#om_CM4QVrknI z5N#qy^k9<|vo>2sTi=(u>;gunwY`_{rWNfRr=nBt8&8bhtdI6O`PZ7zIMI+*v%H%y zX%KA)eBGcI!(qLcem{}8eSokZ0H<+S@yX8-jNA3r^H?d7NjJLKuS7CZ}n=bagUn(K2hnGG(32xC~u2-Lxqb92to9AIf{D~A9 z(O;-Y8D9D3A7toLntMG=tDO>vH8zy8kDmf{vHeX!F{)Z@#*$)7?$cU9^WNuXeux3y zZgx0=_1Y_;>sh^h$bgRai)W*Jq4|7uPfwo>U;dj{vXSKY_;COrY<<`ya%KH+_to`YBQ&pF;R zW}gY^Zq@0+s0m`GN~8XXSJ~(7&v_Q_`qe>81Muh8ELEkCdxiWy1?3dg-Rwt2V+5|t zPz<;EBQ~GmJ+l&@D9kU_Uoy0wHWp)+^FEtK(qqO7`0eL~h^yIK;nktik#|n1tfrOz zrY&e%NHw@jP%b|6uOoMoUYs(Ic%td-PCsz<=>0)Bc9Nwm2}QyYTo=Xn1ttUA4lx!M ztT(LVOh&4b-x5xQ61+fVJePAm0U2)u8fJnae8yo3!V%GcfJ|&zU2X^D&k~K0nm-)h zD}a#AH9Xoi+NksJ4AlX99^X^c2r9?6{*|XE735Z;) zc`nccYe|Ao)zXh&}f zMTj^=qTl*c*o754#7vBdiLc^uA4g^G2Kd_=bFsv^83ZEeNcEW65v%~&bzjdZXOMKz z>s{uuKm7ULUY~Pp576C-MEx_WPW`PNk;dJ}nsRvo982+ihG06H4UH5Hch5b4$3dASNSTa3V61XyP&xnZ{ho=YJ*xD~OmxhoWuA$p zov?F0L4qIh>JUa_Kp3V6jn_+M_Dp5Lup+sOD*;$p*{3>3rrIt;t?qnjzG1gGr`1ej z`%$LZbENqzLt})~tNRRVZPTng)56=+!XcCl2xGAV00L&L{l*+ws$P)dU0k0bu$wGZ zkeWLTG2@Vf2I(#;B#o5NmA7XqilfMH;-)-^N8SWB8fCSNXA1l1rwSk(8BE`AW%b>? z{vw>IhW567KWm01+qr#1XI}5Wli)dNG5aVQ+ldo>c`*$YiBE z*kas-kXU&bj|%CLS>))PCrrE`Aj3k9qEd!9V80+<(IEw@GtBfmj`=57G9p= zG@gsx1zJgzoOToz$ZQK786UE_k~c1dgtOZORejh}QG}gy_@ChT)$o5EvB^b(&lOz8 zMp$p0acUCMYM>=(oFC8J$*!HPqC?zR7f6V`C#iEs-u@TL3My;JuFU0<$aZLM|+kNKXVGzR7`!&Uw5kK(;N=4dEqu;sk`Tl!5C( zL)W6-N5swMCB_cX&9k9luG}Yec#Evuk#KxFC1pa3q8t|-E(8E;2B5HseS@UI)_Y~C zQy~^;O*qD8b!$pDLHt;Uryl%_OOTu6TH@^%e%Ko{EG-b256meNd}Lq6j)vUzc7?7G zr?_wzaI$5MYM{I3g_d&4L(8x%0N`m54w%C^#7S_!3RGDJJanHGBbN)5$YsR4<>bEh zG+CJ;E9(OgApkG)i>ElMMO`Y+uCxb;X`Zc6*A#>nC6MJu@wsnnEwTjVY?bYB0`LLE zU@gE)KFY?GJPbc#lEy0ZMJK+wDoKml@oD@M0BFoY^hJ|C=94S-rxhp|>ZgkB!BSN~ zV;pSR!ZJF*VcZ6J>cfATguvX>1s0Ur_LpAqgMiK;Op9j4X~4^AoPA2F#rsB5U~T+j z*~J8j7Erpc4lljoS$~i$V`2GQ27tID0D=JEo9Due>exXoB!P9Am30*Z6|-B8?1d0a zL5xQ}>_iwO6dSkfg*LitC22FQt|s6G3;>3sfh+*~DrhD7PIw9pAN&a0GNn7Pow_)A zzB54jpt$QykZ_mxg95}<uL#h9;&Y}Avt}~|bcPMA2bAzp(mfqEqoa%mxnRFJ6Daq~uD1#D?m7PT z?vPX1qD@(Vxs4y++z2k{JRNRe`Pn*m-`acE%-X4w$N-t;!=8mgR-pB&Axh=gG=Jgt zUNisTM$>7CD0a_0N5>wQZMz7_mlF=glcjHiOQEpJfRa$0eSMXnjSC$m1p`s#>VW0j^G)PWDT*wJr$b)>I12sPZ;n*Ec z>D_-P`{gHz(56PzrpB_TB+t?(R;SeJlc#aXzDQ19YwAA^imWwJS6}9hI}JssLD!Ez zQ?L-j3VDk~J`XL|5cfk{-OCPE6Mlq|@9R0)hkO5yagW&NS9If{YD4;Az;C9Usm3|S zZ`9rnsAq*hf6+{zoGI^Ee<%sjU^)-M&4~r4g@G02NTl*eO`l&Un6O#e-w)FM{h$L` zL>lW+>CMml*6GZm(x?lWMm0`zKfM)9PGxjF)=FX}Dvi(vC_a30b|X>ZFeT?&Te#k4Kvt1OGP4?eR;wSM{}4+otflIl^DVRO z1DDs5z?=>h$r8TG9I9v41qb~2awAYBrvAngDVK<>ZBf$OW2D6}FV z#I6qXmlqX*Z7;p~yi($3vX*Z8W;9IU?NUoM>1blmz?yje$=o3c{N&>MXz=$2Z`SAh zNRsMR3gRSD`|zc6rrz}DHQZ~z1*42tljkp-gjOGv#3AJ@x|)cC&*(Zlg7VZm4{_ur zS%SOYfBQ@J_Dx;ImVM!b{^0>Ubc4|uz;_Rg*09W&H=EYZbA|NB|hmLq1QVs zbYDes3qsR(B-eMOA9vt%%LwUR`PaLRF*`4EcU5|J)z)`kKJIGL?P*Kz>Av3657;xz z-81gld$qpz`f<;kZr`$I?~Ta5&ASoifPIIa{kzx=ryA)0q!Dz0Ex^KmMfuYq0=q%X zl4ErYvcXqG*HL1cg#*!KwEu%fSdRywYi0iDevRT|hiF16(hT^tMRHoZ*PAQu^GJi0 zY!Q>4vAh!1MUP%CaV+f!;fum@+Pm?zzvyblJC`Y_A^v zj{MrZOZtvXgBZdBcOkpqE&m8d=t`wGGcib9)eU}8z$Nth<^pUZ&vAYD5MdN?>0?OX zpeSC>s5UTo{WNj#kj0J$ixQSylks$~;P-!#LcE6m}M*6&TFY6tVZZzPSxuQ5n@wIoZ z%NWY8(Ep?7HCv?NfPMLkzL3bW9HqZrOx_8FKAXjfFXr|O(w~DY5|PZr$}utbOnSU0 zkV{h6n@Ea^h+@*r!-C7Xr=$ba_~ovKz&LYcgv_tE#a5YpFIQ~zYi%-%rA}ZL@iMB6 z8K7W__2k67AZ?t&$`E{alKNq)6`V4p{5gR{Jnc+PpE>FEv)!H z$^)?ko^FXlWuAnhd5tf(n{>hhj3pRb7+qKEqaLS&DFh<10E7||)4p^a-gh*I9Vso2 zho9G8%xR3NzNoeqC=@B4t)MM6F7%pN_Tu}NzTf30G_IuMDb$r)#WAbW`k7VV9EU^C zhC`j2XBgB?bI{B(@#Mr9r8E4%>JatNb~bV-?cg#?(aWXT6;!3PBTHT&-!9*6jCU%h z{Z5L{zI2(!C$-)Ml(J1;_NDC|Rit!Bs!)LB+VSW4$82PAca4PHhm#1VOh;_?E_cSqvPGZ5OeaT7G6ZMWNa0{%H-cKZg#R82a=*Hd_ z44a0AiaMItKLO`RJ}FJpJ=Kl!u3F%*VxQ7sg_xAGrn6AtMZM2xe6@gd&BhHtz%Wh= zWbi5aK6R<78ZhhtOugIsoU3$E=d`#pvj{*TZWPcv6hPMpWyJzV(UVxT%Ca*@djFCT z-7bK0i}eG=hR|MmgGtFFMo6$Z;`$tvXi>LGeg`vX`20cWh9LK;SwAVD3JK22&nEH&#?))Gb0|@P6^oq$p|2nS!opC;;dwM5E>CBZZB@ z3+uEMQr%X5EF?ky+&ZTN`$_=$lwv_*n~*sY0E-^;06lhvwv&klr`zmQUe^GM24V#x z7JC}tu1?H6bk`KM5^v1t|D^G%3r0D}fkYP?D(IB|!MA(>%l||e=&*}*)0UV~)bA!T zcE~0O7JC^eHpc+=?vn=JaSLhhW#R{}T61fc1ooG@V*3y=;jkk)DK@cyd~h(9D-s>+ zisbNBnG8>Rp;hI(cM_}2fHN_u2JS8d;orw$GxqgOw?UhAwe++SeExGf@%ti&1?ADq^&MY^e=okwK1^sP&q=1+w!L?#mh_K zS{MqrPCP06h@j4wfTOW)7!IO*0>wEtc`y;96^HzYqkXK{aHa-VZJ~s(;%7tv%s2y#62)E4-!tmvxqs zJ|R|Ye@M;528pxPk=CO=!gDkZ-|n3JDAqp55eFAc2aAC1aOzXeOG7=3f&pk_^^fjT6%DX3mxF_ zGiyvz?O2EQfkN9Tfs zY+OlmW(wi_zh0Md5w$1fj5wR&BL~SN2gb<4tf5MGQ?r|kCQDazJ`XQ!WD?FxlY)b2#Q6>S2a6oaYsMBfo}@Ov9!v4`-^2%EU+Bs0lPm$@U#L-4nP zjC4X083Qod46R>;veqM!vYIvW{DijzePAU4D>JEY%mh=M3quw!o93Zw`Aq;9M zClt9237n^-snKriYBSk=mw*#fWKr9wXT)F{==Z#j`MHLNF~2M=!Ax zvCBp7vzq6TMKxT-#lFrKY>nMIU%hxlE`@ zAs4_D#x72%CzzPZkK{ejN+(IOI~V&E5l-NeWdFnhliK|tJ}<7T&8P2*;cz|0Q|5q&?u!jOTlnOfmbq$3m&2l62@t!F|88{-M ziD9P#u~ZQh7$~mnAZFc);GS|}V`a`ymQ=t%)CN0^r81YfC2!^_p+jV7|Aqg%C6nf9 z0*eOTXdv!`Y~m~`d|p=Yjs(#^ivWibV}DkSYT~NONdG zdt%05(svL${4QzuN7_;ZRRa<0IFjPgoNWw=C9%t_BWdQt0{0D#9Ro8 z=pqbDA5ZxxnE24g5qV1NEt`$WAUF18o#{}Ro!-W)sx5E;!HrE!(1)P>DDy56ob43} zyU5(aZ5HzZ{P*_R6v63b%E`wEjD7jkW!40$L7-#J#I0q5Z>Wf+3?84sIQA`q<7p05 z0v4)IFmeDt_mE2fl0RUT`;PtNgL46LGqWZVbdo^%TR978mEp&c%ydYwJB@#gEE2nC z9@=LHU?(yzLzx}I)iN0P-7`!w5>3UEfht5hcf~G3#hwRr6Y>E#ZP1gmLN%>1U|76RF&hL~ey8NulDD+8M3EWhor6CMH-*LIyXp zx!MFFz^HuCtsHPjT@e{O{}f0(ujmwahEv|pP^euI$x!iPxtNOsZ~`q?wWp(7C?a2E z&})=hajWcPcpKb7+R^=Xs6FtMCyC3sFfM0iQuS9ORF}#tVMII zv)z(#ovY~^tlRIeWwp!tG7&=L_YCwWk8%>Kmr)GCDZr_UuT}X(5FIM1T~$>{{jk!| z$`M1`$xr=9mtPf1Z-rAwh^;D&lkG?uPF~ayJkdb}P(t<7?s&aKmEqY+6Vo=7 z`DZ2T9|?hs2GuHR-^dzy$0p(ECa6iBsv|)QKx(M5%JPq-jYP|sRf~aNW$As3Ds}5C z{;KWU0#)k&zG#64001ujdjRzRaviuZeny2T^gwILaxHU3Gq%>pD6-xW4^=yN(Ui|8yPRrSSi!>zEXI{n@xcf6&8;=Wmm^PuEU`HS&TGM(c#aG`sgQu_t))r z3yx`QY3jGvzt%p2=L%o_7CiX+*w(pU-P7yZ8kNJgsojQPMOrVQSoJ-2?86ge{AYHu zLh|p1r)@V1!g7L2-+FvS!URfEc#pL($X!DV>`dIH5x06~S8u^se1cyhhQvmAlG7~M zGs`V)ny(@XEqg=km{sr9MJc^;!$ebpv1WEOB)$WzSOkO^&rq^x=1aS2YKb1Z>HH-( zW|8&t+VOtqCvWnc4UNT%QuG=7;&f{=BKLiNf|d63NJ%M{6D>)Pm~C;JnecLAW+Cz< z$6Y=NX&bSa`)}xgl^n`FOFK;Q;}FW-x9BOB#LoT-a<;KTvym)B$0Sd>xq#t$7S4pW zfaklVT_PlF`E!9Dhe@3b^0or8Blavrw}c@kX}}1w(vojqDl3Y+_bEB^9_l&ZpTzz3 zG#IOZJM%IVK=hxQOS2Oj8LI>Pwk?;8m%4Q{4iwzEaE2{z#28HWAKVfAsp$QYfrUK| zt9QB0!n65j@@MmYn$hj4p4)aUaa6QiTa4F~4TAp`9&XM#oA^RM<5jwI%0|Snb~Oz9i3j`q`(1 zqi9xirI|h$(cwCVPuayuD&nO?t+7i@)y1(7PR%0llEwGm)?+#NJN5TD0+OZqy2)3DA?os}~Yc2-;Fn5-wcYo1*%Pqq%FFH20d^s zvQ|TaXNf8>Sklx?i0lim6qo2<+EdfUqtyEpnBr4sS_!*tT&Y7pBDLis6X!-pE1ib@ zVa(ez2W3!Nult307a z$DE`|R1D)CC(Z9FOVMoDRR&}v7QvX|3gabOFAFDgDCA?&&yx7SfGJBNoz#yj*khJN zNZOv_q+x2T7YXAeb{!zV+!U*EvKNu#GXmE=NSTPiHnG9LZ1#a<+4Eoy;#v~soKs!f z=a7aJ&n>M0FltMTzTs&s;Qk^HfM+LzN2G98Jw)$ASS|}TMoH-Z%Ec~#V%Xb5kt)R{lgzTL_m>75XpEE&YKDuj};Bw zUyZ_9NNc5V&RP9i?2vzu$kdCugb7fgjH!}N*1h&RZb;a*iyb;CqgX+JdnHj zukSh*Ki;O)x(jP)_8-j!Qjjt%xMhc)mg{D1lGG%VR;cDvEk}iU7NCo-A?}Hoe?Y=> zL0VJxq{hZgPfN`xm-z!t+QP=rwyVYm&OAwV|(}NNl(Tm{aw4Ynex3L?TgW% z*-tb4`848adJ;!D^daK728ZSDH z>D>1)WGw;d6qC!0948~M^m?p4X1BBEL-=!eW^Hi3o;hY^w>tbFBEOlxF>O+dQFD|R z70DB4Y$}$!S@YZfvDo+63@F)9MhpSzG}wo=jGFNaHX8gV*@;#nsMd<|iJ5bz@o zqFyzOj!>M!G59rQV?{5)aQL@Ka`7`sOuAblUfd5!EBkjzN`>pU`bu2SxDqC*t~P*x zwm!;%_jwduq|zMV0h5ALq(t@xx~*6**t{~Tb3hm9912(4OF~k=0?;aoja`@~zEm-X ztv}8E!#&PZam@Fa&D1rmUb=z>bRm|AEjr|YXz&{0`yKm9rXK;SKqb6+0lVZJT z8z!cyyB+2V0N7x4@k|8V@vm*u)q2bRF!<^{+G)x#_Dt#u-lOILH@a3fPcFNaGf`SCTNSzVs|lc-vyg^ z383ffYqkkl_X6Q+g1=m9U;PM*b@2zD{#z)xC{Q3V;o!(n-M0DQkBlFvM1=j$@bR6& zw$9l9NCadslhS)nTgT=RH)$+d#S4bA|LCEcVhd7jgBFe2Td~9b%NEoHV-otgc#}C@ z1p-ffjYq@%4S5G!%prh}rN?D(B;BSdyIG)KK`o3#?fB1y^KER+~zh()f5 zn7qD?bSE`0_KqqagJ$zV1=#=^sHi%M=mxRqCJM$T_h>)E|EQ2_i;fcWOlS7&ZsYi! z8a;>=rEOs1q~|t#U}d48KRU)mbsIhNuhY^snlI<5PmQ&1h}BYvZD0+hR`OXDV_j6z z+cb=8Q7}R^#~{4o5T$YZVm`X8@vxwmI>J%cV)_M3um9N54Y!#b-o~p1D$gknC&llr|+Dqa@4F=pB&3WqO3^X!7W_RpCL(yOv~2dm?~d%AH{<51RgEd8#@jweLqN z7vj|cOPbl8`?_5kGrx&;X_^fr&F3(U2$8nNlJ0k>VV8jnGa__qH;*1qM~=tEi>F(Q zXUvADCE8~MQ-nn<>10y+gxqFe5oQ)vB*f`wYI{UvL}nH#XMQ@2DZb7uRY-T?$nvAm zHp#GPSq`ah&+2M_+cO?aPZ}lb@iJyEre7Q?^TT6&Ir|HTYE9ky_!mUmH`y*W%sJL6 z^$`rBXHG4;B&9}a_)m+8nHcu-MS30>M0JsG<|~rCLvn9~UUhS%66@&Pdzd!cr#m&r zcrWDQ`Y84i<|oP7ogv-{?dPdM^75~Tw$^ z5hJcCcPrq&+=n{6pg#Jchio3e+BL)O{u#@q!8>K~LYH6~j6j|pu~;(wR8=7!Z~ER%pb~ewkjPmth%*tasgi3jOmERq=FpOBo?1wG zSUj~*aKVwN{8Zs6sEZVN?h^nBZCx&V9?5vF!{3iaqi$TG{5v*M1@XV}@6 zcOs6N#7=UQo8O*gEoC%?6^UGBJ{YA=W4%8GDN(1qHaxUSO(}q9MsvEDwB8cX14+iQ zRXdJQQw}-LEXt>!S4Rh;+9t{|w^Ua;RFjmoT_-~Bv9OtV2rKZw0(IS@Ppdsol?kYM`f{CTyx4O3AHq1`cBb>lcfUR#+tJsmHwTkstt2xtA*yyRM z?Gb=$7{ooEPd?_@6Nw~YNbc}5p$)Gh526%NAXSeE^XOg5jD5}|rES12pjM-$xa32^ zpg<^e@@E%Qhy=-0g4)Wn%!Owd*|Mr41IL6t-jxpQQ`JLOlE1kIRpkZ;u}F+ShB-&% zjWaL0w+2C1Ox44+hI0za?X7VL=!eUpbHCA15XCa5*Yz}qoZOfxvzK*!O&H}TQpb?d zl4aEm%JDe0>f|ok|wi*Q)P>yjyD4YQfruUf}Pm&dz)7sW8XwxL9F}-miK7VNV;cJ}@kt)SH-? z#gq6SPXUuhZEODe@$MF)n9RIax{a=m!G=W0+B-}`FUa8!4Jn*wx3oUk{GNKcac|i+ zL-{D#SK(PR7{(v&SiJHgZaLgl-}k4m&_-!l9BeBs(Uw|7FRSxTwKrJEtPGF~v7%j%D7IttvgXUSrFR_p|y5r$-hSfg0ToK(Qg zn#qkuhtuu@3T!`2PD96yA{ekYi<7*lNamc~MwD}`>4;Ncw=nyG>lsUZcRy|$e+ zaYhHtc-#B3gHi~pfEI;LtNh(7XSpn|LcM_lko6EaaaaZ7f2!O{g=in;7t4@7x3SWC zv-0yzQpLnFB62__KYxr7gIuHCmx3|r_Jp!JhxUPTHEKykc&VU-2!~6qfhr=QaeMiw zqy&0hUShm8b5$)559p{tDb-BSG?ng8S|eKOp10fVNhtC6hB*(Vnw zPd7D(U;M5}p!fcL`IWC@Uhas08z-Xr3w2WpdZ(hJyyQ<#Xa-3XS4Q2sl>hO_c*mYg zBR=W#BgDc$2G6KMDQ z7A_1RNS^euWY?$eJXDG~f@yTi4iVQGk5RA}9dHYjeJ+88rI?t9cMa5-Wh9Jj_>pLJ zhi)g014ZZc=%-$;3lqz>Q0vo?b3;w2j?GxCKQU`l#nc@~aC55mjlgnAwtpINOnt?K zG{D$7cxmD6gVftcx^QN_ek60=ZNijx4!Sj+sUVB5KUeVA`>jM~$k?AHplo{s#oN_k!lRZ16xq2UYKwvHO~ecaITzZx#3PvY`0*n&{$=Ftts0T5hPV_KSs!J=nfDB|?sd z^u!+A2FSkqJtyKN|A4m<|b-mnKAe0>SHo<;0k<=-2E9#V}@7gXJ@~p1bAwCKDN7 zPZ)VSSq#0XNz%br!4KGMo6!QECl*j`&8`IEngh_GtWq2>LzkE4&(70nN_HVS!ry$e z1#B+nSVSlfn}V=h1jgOoJbMFC4WC1&j8Ll3zPr;1T_Oq>XK|2)xbpJL*4tibnwl(> z4tY_!Wt4;z3S=k`)9!tu9fwzFr$lL|2E8yeS!FeBx1VQO6VjHS|60e#I`zK&BDNlv zTS|eVA*tr7t`87YCnBgR7U1mgy=SnIuK{Y7{lLm_rYa# z2pfj~^tspEdd!+ixN(h2EwuIHmR3DGD(zJft#zk~pmo`gzrxt2S1Jd)<#aA35R%jl zKO}xWE@*2F(2r3vS&m2!o@5ptFxObV_i2hB;Yo8Y*t(j`RI!~PS@VXA#b>)c*@x0=Vl&jRJ>T%f5-!in z#MyM?HMWs=B*KOR0q!nQy04&UzugpGVn-j<#Ob@rH(r%6T_khJC%g!+5cJ_TS z)9^b=-@W%92c!4t2D06K-a6nHH`N30h<~`MGUGTh8UIUChzy*2dc3B`eFSTNNYBT5^3%WjBZ8-v*_(jG8TNVY=LywsYnTzS0cN!o&R-DI7a^@Tm? zO|6wh^!VV{jEGjJN2FgDsoXI`5FyGsJynN=osjWOpsLpMH~Z?ZF~hE)p2hXg(FQ_c z2hhi9k1-)vbz3v#^f0sqKVi{ug8H!Y7FOFbuMfAhoJvF>3Nd{wcb^nKjW+7>`moO~ zFCw?iyq{z_shT}#Ss1mG!{OfijA41~GXT>ZfT0;nc8Wb&L@k8kaTp6*XEv~~)4A5CWIH3y7s7`pXgh$hBKz=R^3mdf|ACHZRLyodyB$!2E9JTcdn03MB{GMGNFjHDFlb|Z4rEjLL!FaBBo#5EI6Q^ zTb6N39QRJWb^aD)njc$xiXt!v*d(OVrcw1RIF__)QSAjQ9uhqw)B2FO-#{Q< z64a2G>_vpLgby6R45&4VB|pXXJj0H$2bi9wH`|li)F)qhzY%9k;HeKYd6JI_8_qa3 zH)IV3ZKD()HK``jQ>G{sUU5J#Y@wCPv3SnlZ2%rNRb({L$;hm8c79UW?aSk9r_OACZBwD!dlXadZf3FbJG7rc>OXC$cw(3-yFec1v$U@qqj zPp!2TYYF%fj*ZHPmKRjsB@fQE-mM+XVM+V*m99;hBSnTdCP2 zBa506bsdzumx=ObUX7p1W}ce-6tkBSEttqCBsYeZ<;Rn6f^D72HO$j7;h-`=Nkb8j zj3wy?nMYho*@ig9jut^~JKioF|F{m%9*ir{A+v*^GpulTexi)CVFA33{3jUXr87pE zbA?%c{L8u!hZ(9DnE6PHM4Ii?vkEm#PKw8G_}`&c@VYqmX1va}wDAmk z^G{G(VJO~b3b0=F^cAgjvgz{BV!1IbYx zDnl(vgB7Az-XwZT1!AoU_FA^Cwyr;RpZz){c08!I+xOV=jeNUY1X_tH%OnFX^Xvkh z5vkFtWVIAOp*-mc-xR9C4_HIbr62p;g%m=BBoyeM+!}EC>zxckxDFlTh`GsHat({! zc4izZyv@=5BAX* z`^_S=&ASI;4+AY$SS|KhO{5l{O8mhZ-mRWo!FY^xUVcp2wk^P;7dDCwA$}2Gs9*)< zM7u7HNNS46?``OT?g@XM6>V|=;n=Ar?ZAR|pm|bGRy)^LJBXnp={xkms~z2?qolLr z)Ax>=ERtOYAMK?E6WPwCRI=Wi4p~bAjnUvzd!q+<=y$^w&-6~DL{~*p7hQD1l%FaG zL+fc_XVid2&q7hpPleg6?i-Lsaz&YLL{i(6ueE!6H^nx=cV>OkPTO4^LNktVw8o~- zlU+akUfnK9&sAAc+MDd9cRRb_lsRMM0s72NAwz$9rhM%#CTneuK%h+kXc_>X225be z1}+(``G9vtAQC9di?P?0|Hcc(ydYNYlQY&%^#rv#u?`Se?M^tXyiaq+kp1Hg6@ zgy)RiyEzi*1W=d;lYVe`uB`hhMv!g=WHt?cKRp`m10D_@(<2&QWsZxDV%sq37&^mv zX%2W{9*d_5vIcY`1cvMk;_bXfNVsD&p7+`IeL3k0%W1O|%;M}hh9Y9L>V_Ir!oDIM zBQa?I^%$04=MGy>9vGtjva$W8*|>nrbogmi8fPuii(7>*c9b%9_3ty_ILOE;;7~#o3JFQe**-ipC9bjOtP;nYDW73PiOt0+0ID8_|Q# z`q}g3+BE})i!U0E&whG+w|DoU_ct2&j3_<%o~<_nsj`U2%WxtVIVcs};9t=__Sl{I z7@D(^TmR`A>#L@OfBaMIce;gODWwoAL$(~#<yXw?)JwNBa>^L{->xNliDzt)1&Z=!d(Ga4_N!R6v?44Nb2({NO*a$Z+Uz`=k8G zSxVf?j{~X$Tg1h&{`0dAqX0rcy}zL7I(fh0U68gd6T6++!)YYO*7p0i^Q8QU7(pWK?qjnv^euuhV}Nf&iMuW zT-od}kDUh27zyeZPwH2`6n8v-$zEe=%E$Ho z^hQ8(?lR+E{)&whZ*o25gWv8{`h|@t=W4f!ei2kw#(w;iU>AOcgyD&1yk)1lCI1Xt z`!auKcii=XQ)y6hv`0wiy*BF9WU}OT##vS^tpBcak~VkOC}))n9acGjKLd}6JiW`Q z)s6XhX;N?<+V;&sD$cKv8OG$0%YQ|1cXyqCI|gmqe!jMG%pCA>0V(6cC?OU4UVHYG z;G{0D3XJ9bwBhXcGD4;H%+~adB$_uhPhnl)?yBQK}4@MBPyl_f)DFA$={*St-U8O`8^;}8& z_f|4!-OSf+%LJuE!Q6NgxXV9nx?PaSI}o-SPA7^`>_v>Yj*WIHZ_>SXr1p_+rC#-G z)3I5b&(S0doU5Fh8)I>j{;cS!SQG;K%qZt6L#=1>(e9m|-Qjw_U-$Tr2U%AT7n-Aa zf?pkzUKcG403EI_%5H7O!uWq0wK$0iZU^){zX#JA3)y87h*N!PZMBl|6Pq1bHiofQ2Qgj>(ln~M2 z@?%n8a!W(lxX0Tnda5EjgIy<2@)tY4JGZPXQ1f@kJ}>4Qz9mvolCCE?#~ef%zls=% zb*jfyq;+e;Z93U5G_{=_j3I;s5oV!rTTk> z@s9pOO3JhTqVA)spJ`t2&3b&%{@`V|8pxQuC+2#srdeav{^}>>R;4y?eKJ+hYW-8Y z*-CSaT=dBDQN8^qbpJa+r=|0a#Sp7MQj#4lZ!^o>E=UqwTz}N1_G9#N)7%lTQDR`+Z;ay7v9U zpKF{AFH?du-fO@!%nd7aFl^l!i^^a%Xo!J)Gq?!vmcU(qnPttP!6J>pqX}_PU`I6a zBjx=rp{-@mK*`NkR8(kr%7;&;sd+I%=c9R^-1}px(Y%(sq)}{~6vZ<5M|RdPJy`Bm z^}HZ`TS?RyVLK?@l_9J6B-AZ+u!*a_b&7eWJ4;dpWIDtAPf10Ff2um=eil=#GV6O8 zY{+iTXn%!mcAS=&L87{>eNH~@FZ*&eS7Xw3>mMw3uOoA5IVpZ!7PXl~Wb(>IG=;+2(8EKN;_MyemDjFE6yQ|55YG^17rtF^qHLeP*T6 z&YNbSAA1ctgEOqQg!8tl!B=|WCD_CF7qEqsCSOFhsw5Dg@ z@Q_meHs!jh#E+#Fo+CowdGMlBK80!W7&e}?sAqMhu_Jy_4z w!gYYd|+a}-4vUz znqIbHQor5@GNkcuRx;L0464?_tK2cScaVPnht28rRqVIHp~ z)YtMAdctFP{QKY-XqZ@7a9AMS*bY6qx@=hO=U6WnMAljqTh;?ssS~MWgv`O2B)Tgf zr3f<&I;Iuv3;4p4M9rY^B;vwD&hQIX`W%G_m?wZ@!w4x?^GN~&BM69?BV=imf9{5e zfHTJpvnFs6*UgE(N;625(g+3$5Mdni+1IEzBDfmPglTgKS?B&Ru= z?|82Jp7VM(%Aj~(`6$e{=;(-`{q>XD42TR_EN!DQeRd9iYfm=?)lwU`khbqqmA(}xZ`jJriVwQAzMWK?ASJ$F& z$viZkK=V+dtU;~!a+RIB4Am3WfX-R* zid$#sG$kf6nn#!43t}*hZ7exI2Obh=r4*xRhNopk4J|_ZCE=epwmc`P%5>v!J~Tzp z(;g@_2E^d4S1ZDLv`DW2qGu3N7Ann4a-`eO{!y^`k}PA!UnC>_XRxg%h0fqU|4aJp zh2>+GNYU9VycA27QysF5#c@006ujhFY+A!K0|DEHA2w9VZ#IFw*q)UyoYl2HIj=@O zMC5LPXraHqt+(B&*U*xj{dvGn2eZg~1T~#1Qxy=$KrvyO$*kBHb}V1>+?v!9_w=dT z4F?p_61gOq71Gxx!o>(X^%!Xwa3`IwPNW<}nX9N)x4b|}T?^~kfyQsiXkE~vXHwtcQW=5B0nK-<8w ze$H)8o(tfRZZ!KimXjaSXiZ_Bis)dl%?{S(v_J2fie}vbrQA9d(<1h1UTj{UbRs*} zgD4^l@Kf_YCfPdi%g=}Godt_o_hLER)kVHfuJT7g9h|V2ju`Xq0Y(hJ^T1jdnQSut z{0&>6z*sp^<7*$VoL0)a+8O3@9XoC2u)ZkvpJ*)rWK-b^gvEb)+MDgXpNRUFLPqnJvALoUAs;5Z$fVHjSyo9E5V`|>} zKl#bA*{qcD^)uttgW-VO_gV<%$A*lflTrBQHkGPDU3?Sy0d>$SZ}WNsImksf=tO5> zUk_||FmKjCaWtqH$QPD;b&WaTXd$F;q5Fp>J{EI(qf)G*CgfmO-NiMHB}u<4+4R_l zvKS;CFv}z5)rpAT)`;la?!%Q{YYuvoCwI9NaNZqoHn`(BbU*eJ@5XwHTDOQ@`4wQu zf9S~UWT-ZE)NT8Q5*p%Uu~cXors^9(u+0yP#_iY}d^S9#Ydymt6@9C)TOhaPB!f-3Rk5w`S)bwAj zRmi;bW6a*9UkbP_RpTPwMXnsxx)xkCm3-X&=Hz!bFszcEI>?0mmjkDy>LEUF@K-!sbl%;Ui+6ZKRZ03guK4sMsZD~QHNFi9HAZ9qfnD`4eih8Rc3IW>N z9-=#M-Hyqyvr!S#o+ua6$R1DW@qI!!0~gqxi-8#Ux+YSi7c7-)winCrri{#LLq3l% zMy)$~lv&RC0T{EVXc?|L^OK0j6NlG|+=xXb>&WX*dVnJvMZh{xmCbF1h17K|mN{L# ziIDiOL3S~23-MPWE8g8ICM6&JD90s~5gpSDZdimCZ$S6?L?p!GRP@|X_6!PO#DQ%@ zWWI_seE0>dz+qnnKQL$c6aWQ4`oFEEb{I=ef)*j90MB-|i9NZ|g9V zsP6Y-6PToX6aJAN&^==MhRtU~j4^*ENqY%8TC0dEuTkNfRBf12Go37qq(oJzvJb)E zN|QqYslHa$_BM*`>9s4*P7IDqkys+oM#Vk2BG{(`ECW)m9SII>H15W{VIWxaOllp5 zA4;X-0R$g#MQ~>+hc2XrqMMApPp}>bW35WQOvjT%iMol#eu?*GsVEhGqQ*p<0EOAb%)jPW%e0wN ztRiOWgJW+qC64)xtXN0=d`pNPVHyIW@K^W>F>3$G^Cvr***L=lGn)WOmhL8{U<5nq zAq5V{+#CsbB^LY*lSxP=hy9vlb2okj6Lp_2hY}-73pqL_p2)H-CvhVV)*+@E_j*#4 zCQLYwrArQaZHX{CJoa}CEesr1I&0hVCZ<9lS8&#}jhs+=Mf6>bHSy?c zDpvo_@vJ^94amLGoj_=zF;$6ip+K8Z=g(LdW?7QT{0|ts8z($Z^m*^=fXZOQiK;Yg zL}gMdx(fk$0gn|K@t7Fm%Jg}+__8i-&IdT$N_xw#jqF)clv#50sVHzA1n38m?U%&2 z37}%L>hb|i`G6)KKs;w4L11a(DAo-#fLsin=Qx5J7D2%QU^p(rTMB0fM=)+i5Iq&L zsxnh_mvcE=v5-r$*Au5Z5O|7H330?ZvxvJ@n7*K>NJh!)ekhX07BY`#J!c`6J0;cV zsmy(Nt-?X_Z6$5xuF}O^ibSM8!(Ip9@k1FaDstsI@INvx?6Nv{@o)Pe^RyQ|Vd5`v;tiOmrd+!hJ8 zi6+EvFmk#{>*P7AT%7CJJZn8S>%8mhGHhya#q0U7>Oo~PS(`N)&h?D+wcRSUjjBY2 z`L*kW4NV;NudrOYC5T=+#7}3)3`h_*C)7!kG!l8%-eX2hnbbV$@c8Gz<`Z;(^u+!& zY+4CyT61n%dIElbB;j0A|AyM^Okbh`Z5Hlo!rg=D#WtUBHg9&kX!ZbJI|I$XHEZ+R z2@~>JB{(~nu=Avg?FEv)6Ka*$X+>3QJyNdnx+6MYZWS!2&D?0QVQ}1^jZ&+%t&3~X zTMK1p4SJ!?Hx{01cp4mJm4Ti3GG0!1!r${-TT3RnM=VNf{b@b{We6N$)vk#RH`8lt zIel|n+ZOt*&Eeh+RfH54SJgn?X00j&pbU_@ul>^5$u{3Q;upVB*J;YZ3c`_lCsF%r zBbETYGppR#HZgR-vD1*lM40d%bRztxb8o>1e4V0J~RGt+y2#eK|%8JZtY$@wlZ64Na^}DG0SI;Cv=&?{uaO zET|kjRPY=M^-1Ie|J1V2OzVqU{{v2>EWo25CrR~XDoPlu#sJr`L<*fR;y&FFl*MzFb} z(F|GuUrig(MwO68ah6KHM3|lCkD8+JQ!P}v%m4{E4DTh!P;5qP%V|7E+rl|tyktlm zJQ*808T;i7GbSIm5CKZv#Jp`77q}x4rOX*xB={}KD{&VDAkv#otiYNVc)J&{@N+D< zD#&$)%!wp6qa#0Z%%_8pkw3#}+i(J9%=ipyoXa0$Zx&PMGFHc6)h0?T)=nnVS-_b- zzGSca6Ks%d2Xn_H1bby|k7bMBPa4+vlQGh~UK4pb1F9iLMTLFGrTqs~s;6p^Uwd>J zOxXA{2@r0nRb+TpP>_v7S9>xOi|&}zTUlp_B1F3M8z-t}kVmjX;xXg36pYvnNIFZm za=B!>=L#hg!k_wSm zlp?ZljZ`L67JB{0=1(Gg5SB3iZ;jTV@iWRy`;3J}aOx&}I(0R_O%nXHH`zxB9~&oz zM%Ja;13Q5V2JFglTJr{GtWpE>Mmb{iNHlo6M5HMOQdL>r3E@#M# zrS3Rj2?dwN3l6PBe|S6)Sr4-8L2&4yn9i5Lj*#N{Xg?&&7<}}5;v#~@-_$U>4WEh$ zZ*zV5EN(HDi~6-wS6+B5UQ>u-&+-1miYL>atjLv=u-Y{d?gJ+E~7?_2>FBje)=vQ*a+?KCdUDiN-h*{ zo))4hotfVHv0f+?egcyAzC<3y-v=b|tqFe=dh>2i`(x4tr>)z24>SpA59B;6P^E&? z*jlvA<`Y-j%CKsHX3?g+2&gMs=tjNQyDJC{Do48@Vkx!S_jYUk>%85&u~9rfY@gY5 z;?1>%0M72^#pKTJ-1U|AbuHGSX`4P_;nMh*riq`KlS6h=2Y^EA64h=qFv0K~br&rz zxr#s8t-p9`-iEAh@1!tKu4m3xJxu$X9+{$4-Ku4>ZuFwD`8P6!^QFjt-seOE008HJ>;HSy)yoF}Cj5IL zvr;?)H42yAqjDzpUix46In`31Q6~!7J6PTo2wO}A|J7{2dH%ghtm;#{RIZ##^YXFd z??Ps*r!%E4!?*Q@zlQBsB9krc`l;H z(@{`jk|c>H7RSFJ-qjDhcznN$cUQq$`s(e))jpZ#msf*-YU*iOS3me2`$o?AkXZ!J z$TO^k-80u_2DrFQZUx}25$DS4z7;L?1iQzg(plfizKNha)zkC+(s7Z%N$f0 z9M@WNkWIVyEj7|b0c@XaqnNu7iK+TT6D;$L5-0VHJ6i^z37k1B0;?D@*an$Qr)5b$ z?`wkYFtKRql(|HApnP9XmXq-B*0%CayIAQMJkU=FQfK}+RvIfyoPsh#vYpF$G25p# z3b~QP<;$*^5s?#S@8g|a7K-s%+DBxoKSFlHSMtj$ah@zMmXR6z9~sD=+SJu?D&Cw|(wymp`-a zHT6I-R}kFyGH?u@hutW$XJISjNVS@^B#U!VpBN>cfL#=N2<@o4!*g#)3__>(C@NAT z0IHZjW%e<-d*Cr7v*(rANG@xN)_H^Jx2KB=oJzrO<+nV!YsliSemTXEsrFt+G7`WJ za~$xmuF^1go^F;63B+zu=-mU?Ot_V7nrc4czqRh&e=Zp)YeLU~^2f_On;(AB04^s4 z4fDNzMRM}|>q@gZco%_UR)(Jm49s%UL=Jc*Uo9}$(RSbs#QnSrVqo`9XLb7hF%Cta zNam4Rp3T_w-Bk9sbiCzvQM4z!2W-l$w-=|O?emuMTD2KWaq`g_jF>aHUwOOFaamF& zDe(}J)^A7SH9V9Wi=nv0nSP&Z9gUn)eu3D4)voSwl%sptGHk>%K1;$VXPAEsVkfzr z+dR-VIxjf(N@KeAiHVj8` ziaeL{QLkJr{)m;&H0ftBB!Lcy-h7+*^+v^sU*1-ZC!SE;UdKbiR@sFmAe)XTDvPe~ z6)Fn*=t~j8$g%nyN4h(DZ0JNnv(GHMu(1-6H4ql5H}kxVK2MJC9chDVcb}grq0KRs z^?}P?4zpX%KE5yMOC_wNkf+T8>7P@G z5BB{(OP;JJp{NOM2Dr9v$o$8fr1@Gg40-?p-#94iX+k_UyqwfmZ(BqntrzL0f)0;; z$|OhLI7XvVHRGM&pLm@)I`^vbr%mymwRaIPJr5|dnFgA^bVnL@G>TQ-!pxm6(s5%j z^Pg$LpX_s}aNKGRD3xr&=JFeeelCW1wp3`DeltF5E)M^m2D_+k&u1Zzdx*;(vT|?A z?;{uMfoU{c$U4g6DxgPs)__(=4}$IM0_dH@lg!nt|AxhD5u$)Z;{l)AyhTz6o*OEW zuZ(_63)ARwWfA|H_~-BzR{=IcNVT=W2p+)-n5_K${M6ClejxFAFxHwHVls`89>GE{ zjj(NirL{b+MShp$++CJXNKf zygO{&K452LWDS-hbqkM?6^|HAlkQ^U_nB()|I&jwqdH*sp%GU9iWkHLbXE#OeD^8! zw&|CXmc0jV<4c)kW>0 z^0h!dJu7R#jkv}lF;1S2BWe_b4USR|+(QSKzLl|$Q^CgFzP2MB1)Vg?_WR*zLe`Pc z`K~#0Ee!_@c0t5>Ex75yM$y~*q?%$EWP?B}Ub8(Ubg>b3+^ww$+g?NyIp~rddY)#Q z^%bcEHp%|pBjHvAFi+-Jts@ir`QFzM5RJ#1?54g`6e7NW$_A&WHwXuo_g{&H`7!o90?7f+=mpmi)JVk_z~T| zwAxPkM@p>;;5AIv=FHg;#cYqdrLhren{u8%aIoam{Np{XRum@$%tRmCu_UwHWoT5^ zT}j=9)DKu|4TYUjtIU!G$YqP0Te8NLEyNbR3Z(lAyG z%jV-Cum#Pak12-{RgBMp5*pjb9X$ZTA-XO9?XSeiU2p`+pyc>z#g7Ndwm{;cMJ(N5 zhHH%0G0L^)aS?=H^g6+BsYVEDKWQl@&LY-;un(z-2rw$NMc)P41HZ%ri}D_;nh&lg z@3RpD#{!c)lEK8%35A7$)NH`dS>cSLWadh~(e~k3jxN|n5o~NUwCfSJRtj7XK%<** zzFY-SOzKXNNb!xxI~Ftyq$ufH@a>(u94t!tI8wwhQe`7b;~`259IYc3t!ET%0E;%t zjW+3xHrt4P^$=|dj4Y7+maS_JUV z&o$ut|8=+&I(}U~nfUMF^1qape|uEj|9UKBGgtWU9`&@)$g}1{2_zDmnxd0-sp7BF za?Z!()f;s<6kgm+i)V8&)1vJcQ?({xbM#;s~T9EBl#bh zs5kVqSkwk1uxZqD-S?s#P-tH*^tQe1O<`0kZq(to>q_Hw*;(l8IQmd3mq_!nzw_i{ zt!~4scl}*w>#a8P#V-fCFSf9Lt&YDN@VAQ`4#we4ly5oQGeSYgHH+~6Jkn2W_-uY9 zc2k|#(t_9gR#NxGt0@Sfly5%r<65jv(C3ww)YCn5l$Y5QM_S*4a&jvtK5TgV$1@t+ zE}V*)FAEt8#C*evO?vYolHeJ+EQ(OpaWDFaxeFC#CSIa2L$LI?Q#cH~UIr#O)Rx_~vnE`0 z<{f4O`!^D}FGdi~^NES|Y+Csa^>r@IgHPl3Ro~vi=Gif*Ea6B~6dtP43--CYF}k4~ zO(n%opGwlA+G`KlN$6ZgP`-qo;WsoSI922wE7pIj95$Hy$~!^I`F#+-&`Yh5j3WLZ zqePLTDpQA}j;X%>RWCQn&U?-lB>XI|<8gxgd6)hH9d(-lKlFgIC=4q879=#bU;x^^ zGA66P?~d_B31`m>?&L_02cGuH&|ecI7%%v~6E*g;wPl`ZG`cr~EoQB!Cqu6%Bd2)O z+Y8CvH=Q~GCFHHE`pCI%Wr&5KIZmEXNnYZEO%6aaY0X=fR%Ewxg5@cS>km>MWTxSX zkJvYjbG!ox^@}>!;at9U&xD1jLK9Up{(Oj9g3S_zDcc^|OYjziFqX#jC$W2LZA;Pb z=S1O3Ljp7HdPR$eLn({kpvh&#;HySLhsDj>&BX7goLi&qj%b7}gWgl@uRAa0Ir*0R zzJFQh7-Y4z|25rJ))EYbAgz-Dye>8s)q(rFHCtkNX`&C`WM09n6-#l<9WajGxN}St9ZjO8!cG5sIfa(04oaaQAx7PpHSllUPWKQH%ARSG(_khgcrkK zv@4|`LTwWh@s?;5ZdQRWL1R(!yn#J=J2D{QN%=O%F#C?!7LIlWNvCUT;is>pqhdgG zyK{EU0EX-ZZ%Qg@dpe$jBC3XVtcWxyT_t#yjc5#u2eUE-nS2tAEUt`%DUgQpX`d=q zAYQ?*k|x5-7H!=qjU}{_0Wx<0S?rzFm|INlI!RzlU!m84Rx>CkQN!Erf4;&AL^Yajy>OcJ7~s zYZB$6LJwKpq;(~4wCu%}1T^Arq9g=2b0kQXIj5p^b-u3@$(Ghqn0{l=^X;>Ty7D8^ z25W0*Z=p7p%g8wzw-=>hAFJd9CW_L9^pdU0>;+l#=OS;SG6}8^`Bp`wDagHxU4s(= z?#CqUY0Rn%jV77E@(;FmKZ`FGr^LFUAKe^L7&sbA_T`nTUh~+BOAji0#Q4T&f^Kn2 z9NA3YN3-PQ>(VU&qV#I8G5u^iV~kb_l|OVqRa0*0rlJmjje#d0{;>R>5 zMXRcOGRiD*NsL4bcJ9vP3>*~Yc!nk@r@kjDX?UN-R-%R0`TF^dyG#FPB*U)$S;d4BttP z5wc6qM8iP_2W=5RBt)Eaa66&q#yQrTxQ28R3*69bjC5(MvCwDt< zrG>6eX`$yJq3@z`O70?SyN{G161xiT;`7c=j_$!!wvg7Zv>A>a1>CMic#L{&*BA6L zr7XZ%jtSbHR{81H_2{5Ru#=M5HTT5w`^h5iH)gieA`c0eUJo|7LPbIrnO3XOo&JNy zRZYa#g$^~g=QJ%>xmkOEq6J?db)e=7nAQ=Q4V-6XuRa#4C4VIdJQpS+u$~4*7UCM@ zF0y;ke8xUABf!UrpS_V%_V~Bb(tzfsEn)FI&P3O36;U|jzEb!t_el&4r;ByJ3rX{i z%pSZlKQ5#QIu98i_?PJNnd*=;N|^I5uA1%WsV-W5OXGdKy;K&bKApNRiN+c*In?cg z``!3g4YSgFp`o;xp}!`~u{Iv17;~M&!o-GJ$nlOirM)6_LBRx97VzP=8yb{bk;eLxCPm~_bW#3)s8h<P!MDBLaxUkG*SXTAg$rV)E914Cx0hOG%G z+2Znq;$n>AMvo}?Kyn4;5sZi*ag3PptY7`Z%Q@n_;y*{y=NQ>wiT(3^kT;G60lP?P zL|kb_Tyxi6#CKVl=n*z*@g;8PqiDNz=qDbGOSWz3D=AynkePGbGMVZTo~K1IWI_8J zManarX!Rhny{S}v1^DF>E=iPcB~tW(DLlb0M5YI{{0W_*`3LODDeoA42 z{kviOy|rmp3_ztB`m>FcVI$k7Udtc74Cs+5?)rqiIjJIghLXLsi)^mXwHc~t-0QGp zbap@^JZ&mlhDbae2_Yl|B@R7!2>Q&HPGKE(4mLN`Wt3V2HP8c;i{yZa~UTnWEx;9c-Z~jM?Wz=@w2v?H>Y<_rcBqGTf+%4(yEA2U(7p zaaW$w0P$o;m9+KVH{n0uSO##~if2cuu!oW7Ah)us=e+rFlkcHpPk zB225IB`o5clsiK7F*P72@ex_^=_9dUUorkt;3av-1n9I4glYM zDxoMV0Z5dhnv|jkmSX0YV)vKgZkFOdl>#Zsh$PBDCS|07W#IfWivBXH%`%#&GCGQK z28nVelX8~8a<=?(j{b74&2pZnaz2U*0f`DBlM0c*ilA#L45TozK#G%qlK-^sBLjed zLBQVsiYP)xum7EY|7*?Ue<4cJAL;)SqKNZXLJE*QmjAAq{0C9EMQ~L+YyUwMsZ;g2 z&cFHhUqtDu*ROd-wHNNatzS}!?TyZEIQpr> zZp1BHoTXH^Pc=9Y@j$#Ay?Co@5%M&-ZWZ|>q<7C^QL3{L0In8GLq#4u(v<0%$+46X zT-%^a#!ysdNFb?-vkIczVP%S?A7iDrm3}@p3�+yqk$3F|+F*_DPpombv=wgOm!k za$1aRD4TscR5jPmr1nn@bx;#|=XU;vHkpH;7d{A@D7v|knyLLmydnjbhi2)HxK`H6 zi3~ll&b5s=-U|t+gH?yIG4$`iLr$?6vYy11tIQ*w+1ScVyqxNMNbQvmDjFqD>*M^R z5gbRp9j^jvA*>o0tW7@F4y?sNrzZ!k`JsWFX42n8PRnZSF)BlBq&J-^ZWQLKbI`Rn zT?2DP>YO69V_Ztxe4^)yJEE%_4!hT+wwya#@cT^6-o6)a0lYqyI2syk_d5!D3oJMq z8I5>2O80t9;7(m^oBsy6gqj|7F87;o*S_hVcazyjdaew$N1eeY8#YrWoj<`nMK~B!Q|(bkVSB#sibM8$&mCqr%*r2A%eAa z0|n*PRWaOPO=MM$h2|nJ{&8Jo-6VRet^`#^?DA8Bo14=D>brCCL6GiRhcD#l9q&QN zcZbB&#$VPSPM-%u+t0tVzxDo%vbU2nrEBQG>^F?CsftDsXN8>4Y(@SKyR?vF0lH$d&qNq6wS5ck(=JtgU<6{Tey7Jc&@>QY1K^p{m!%MpU#5mtqcLE}^wrts1 z%Kn?)#BB$t-ivu)@}?N!C!d%IH*2$7wv3`n*Atu^v2oU+t9ad&A5LiL)*AgY?9v*Q znO(>DlHFy=UK!pbyp`z^!(vkzZ`^WiZB3_I`v}iHW@Ya^6p*kX2zUsoC1y=;mYnO? zzPNK8y6S5TNkyc%7zrVw^IR)gDw3Jr5d19kJ*tie?N)ds0)l!T^FUniTJZA=KDXBnep3fDy z7<>Sg8j4jEp*_%O5SW*jeIdaqI)O-{mg?6M{>z)J8c79o28Jh~6%p|(O4(|Nvu-U; z0F$I)nzP=ms(xm-tV9=8ZSRhXsxA#G>#b$1(EEh7B4h^gb1gdXyz+*>Mnf!Co1G%S zl$wXcvK}1^Diyum>Ar!<2lVkIR|9PrIxfWy4rD zz3m#FNJ42773-sJ$Ba{l|LEpe;t>PijI3vP`&|c1M`f=2qa{JOm9;Mgu4S4Wh4ur6 zI28`=%htOIv6!PyP`PonwNiZvHyowXd%Io$X}H!7^LXY!A*#~=fU@Q!08hj8qca-7 zdn0cu8MVbcjYm&WRAc^mpc)yx*Q->D@0IET;K_AH7@#lBsznQ%>n!QXeERG}w0WH> z|DabB2Z&-}a(Pu=E%4S@mDu}}8mEB85jKI?B}y>5nMlD`(zQLa!XA~%9@TccqB`PJq|j2oV|td&nO@MpPvZMP9tJyrJbk=BV_ zRliSWu*x)WE&W9tIVg|x~R!M$XH{`iUFAw?^wFb5U!Y#LP$|D$Q6g$meWI??C=l*lIm-P-bnzbkl=my7 zWj1CHWBd8WJ1MXZ3rSZEA`Jzde^^s!xOvmKpEd5H?hQ9Jc0gB#O{R)d>(GUAGp9VP z#!mzSicj@DN~zOiJV;inW0VEs_dJRXp1-)e@YPvWcDPJf!g-XCPqC#aKwa4Is+KVH zdH*zr&Xp}=F}uqxA2hZutslG}cYMu#OWMcx-c!hW!s+QH#P zM9_Q=!?2C%#dPT19)p^4#4zH>nS3u6?K(p8+Oxzz9?A%4@?fPYr?nGw2;ZmqE~*=$ zoB$?NYdZ*W5~Yi0^Ru1cP{}4Lr>B?^O6=YPRWJ(Y$Hm@Vsf0KfqwmvXz=)D*eSZW{8<2y7$1Tm^74KMC;;DI&zGs>vJ3^k1sS5KhjUxU<;exCpdwjV_;e9 z*`eWLsa-szeXkX3L11dOJ`fpu60|@}(iL^ms|`^t(HP}HuuQ)ucr02w>(-IK9aECP zHBJs)mWCxjq19e@6@W^oi@KD?6|P=Zo>api`0ATDNSuz9p2GMV|GOVR)(@@a3X2!Z z7EcQs=mgvZ6AvsW2mMTDu2W@Y7x>mo)Z>B1qyV_K$Np>1^svX+1t1+((1$_#&id)WUrrbOl3;t#`*?yfSN@Bqc1>V7to{v%4N%-H)hRc z$UR^a=tiVv*zp5Uu#`wu5R0bKP7AREoaLX9=6jI_%ak;Zd-@GLXczm$1O2VADo^(`J9u_GZ)0Q`0U*^H+)H1C!=&fz8MH%>(&r12xT;PtB!mpbPAl z8F!AZLjRND;ME2l&1Uso4eKq8f~1q!-L6W_V(;pi zA6@<*7_PT%fBJti9AU;gE+4^v)l9|b0YdMVt0@tZy%PVbnTESWTe|>T-F9b}Gxg4# zNW+mRv^>E**Zam>3>NPP`}D$4%5buH;DOeMdVf|6hTaDHUnDuTAi6FG2jPC*i;29w zJFK_&OITY&zNaHdDCn1zVs^lR$ArZq9cU=}3oXm=%L&oXVI+TaNQ+ay~^wb7u{Qz ztD63dFB?8O50?(O4sPHdG+AZ5z`p4FF&6}EveZ{I97e^takV^TLR;@x8zGT~NBetNFKV zDWSIB)u6^tu3uOhH=ih6wgL@O-R?ibj+*XwLKG@T?!TZ1xawp?DXj?aMt1G@K)S#< zts4P2DPnCGz?I&eG(%?2Z4fixgkv-3_~q+ZiU zdXJiaUFJly!8wNp#(rH*kiGtWJ;U1a`^SRl_uoGkm3aT$tQfxjbGzZt^5<^b|NEc& z-B{k|hl8Bg&yUA7EzeKq^GpH1vJP<(zi-x4H-6o5RtG)@Q@^)_{6{~G2_OW#1ML6z zgG=zFo7?)|!WY`Rx&MA}SMb1B1|Gu+z{){cizTRS46hW=l-T40zzWn!tEBL?l!-CJ04`-GxkuAFO z#V>yyI@hAOY|cx0_+NZUHKqnVr>{WwM>07Pzt2j39xYa=m4r(642M3qyL@^7R?;6y z`=u^|R+A=hH1O@|ZNpJRvCRGH-f-CP(4XLY!LD5%^C84dcm{l3P3D9I{3Y!7lYV~?w-!-sjCaC6tA#Iwn@q=s7#Qos$3-2Yq|2 zZRphJ{9LNi46Eo}kWeSD!{GOs05L$$zo&k(3s($D zvNr>b*7FsvO_*6;z6$t`g5kyOdEVh{6uGrfLCqI@-J|ZfjsbKo5)Rzwt2SyMHG9aq zK57|WBrYYylPoz`)l*)pjy>6EZrBVMc5gV2cfcE=2a%*w>s%r;h!th=z%+d zUl80QELffbK-Q2W?nRd9d+McB60FN!`6M!s_%3R)fZ_o%T`87>Dbs=75oq0A^6R|Io7>?MW$D$DdMo9`laj~x?zC%Ux|SDdGrTu? z+%`~{Q@g$*U)ngbxa%7LY6NPC-WLmkPKXWsC^h6yY0JgbNt;L7CfW7LyRX67|SePnc!fG0}2maW$~QvLCn%g#MZQ&mON*gelXYb z$-Z_yPebSPn^l7V-QOffmDneu?LIi``L*3s}{CDh9qaJn?&dTXecFwyyf9}(|fJBL$S zi!;T1>*K?f$lzQ=ElcBI2XdzX)9WUP6i&=Eq_M0Uc?1<}E;sG%r6WKFl5u#&T()DqE38dU z=;n+<_$w<$Y7?q&e&aX-Mz^#JB3)X?a_%vfOHW zhE6*6f@&-R%DQUV=+qQ_bbTdyyoO6bqH|?s@tjS4{5ru#Ogwj0YB8mq>Ntk)4H%Nv zYAIn_x3mXW%nap76ZbEH5ind9*$?{AyZFoAaayhAP z7=Eu#m!7RLD=pd-BtPc~Et$;6qVAPBk+G@OHpWt?dvMXg`MxRxYv?GK&fMS3mOeQv zv@W}{38PGLB*;$&9DtQ*e~ID0@kpzXb0HU*MF#wSK8qpuTj|QERlP8tnBCcHc2TWl z5y?n2NNjv9*-rGvQI*L0W4w{ikYBoFF)QOAphGOl@~#M@AA&`BFhNjetK?+TD~jjk z9zJLJuvm*Ri`-|AKCOeV$JEBXIC~cO)A58&EKy;Bew4n#rfuz@fXY+ZI+O3UO2nJq zO)tcMhBcg57v)hz-Zaf-OQ})w4?KxmlkSx`{7&1|KIlZ9O#zIWwB6Tg0YZyaUF0R! zKF_+wO!?ursp{5N9MuFv` z{(zoK9p64bA(a_$*A#F>WMZeFY|D`+1h|B!Nr$M88^#dukjYD-bH{m*JFfm7nRe*| zl7N{exQZs*Pxul9P(g(*MZ=BEymD`0Gm;u!NQJP2bVSFNqBai@74zFRQTWyt_W0`z zSREWS-r{$Ehxhj@sxP5+_zUyq4ABCrvMJc5%l^qzX1bI+z)J?%3xs#4~`e&!w%ApH-?!L6Iywr%N0xI<@dw;&FJ@f~~^Z`0O+N8nuXV_Ot+p6&5-%h4- z!4j!|_)3P?K7G^(X=9FZ5sHwo)=Tvkn!NvoMRbaQU(w7`xJk2gVa#1L{A3-|eTojS z2nk4fnKp^2x-MPqf2QSqZ2lFN+d4{-cH}yo6Y5JCLP`{(bC4gQQlgJUh2#1xun%hM$iHf<00TndW3WumSPqK*bu-NToVROmOrzj7iT zr^A}oBM{RJ#ARX2cVL-ZV;nId+n*8b`x-|zQN>D;GzMDoY*9V3MDAcoDmEb+W&T)0 z_@H-aqPCTBZ!~9Zu+|MO7eNmFq;({Z->EWn8=)%R?GMke3~aPp#2GF6*7(9B?zh9;3xxym5LM0eU)$qXSl#^^J2i&p|B0nZr zSR$eA0?@Di<_UUN*mhCC6{9rbe_1Oj`u=JpQ5R8LfbpbGgg8*il3f7RY&seQpbto+ zvx{C9%OFZmrEmhU&!*!e7XJrnXTcQ*+pXten89TRcLsMSSRl9tC%6Q62<|et!{F}j z?t$PA!7U`X1`5a{=wo}PZIU90U{{T0u?_jNzl26ZAJAeL09EAIQIe>{x>*(taz z$+Y^(Z?YksF6d;F0MHKj?h-?FKAG|qAlsD6+m#l3m7-gej0{Qn;DXV42_o-CD6v5L zQ!`-2A~0Tvie0+=JOIlD1?U2ncLAWl(_c@flOUy0=x5%!pvd|H*d~>>Pcmg!&~^0z zCNRKz2>=QrQ+@!TY#JO?ttjD!Yh{=9-UZX-GTS;k)AQ7#(I=5%H=`w)vNj|cbcyk4 zAX~x?ggpF$6~=0P?t2-VJ9d{;~ziGM7Ivhx5L zbrd9Y0(>2E#5GQ*L`LN$M&nKjt&^hMwGiGVGr8ep9z^oP*+zkuel}C8)B4ZRb%(dXCd0H8OTP-&|CK;0R7yHH6adb(sxfanX)8lae`Cx~ zdpBL25_8mUf$%2d|A zNBOm%Ehf`Nt)BI?U2JxJ@V`a%^LiKDq+OZ#*TCo3#HRpV&xCug)1$6p?;rJfl^fxu zx}Da6J-xTP4XL^>QD^JEKYsUCC#mI723+2a8`Hgi|8V7gbpP`^*Gu;E;`vXfyBzqp z^rJ$pg~RbekGyK{9kqeoj7=weGU7ZR+_?i&NetdD>ac95tX);W*m!=xcCj8=B;9kH zX%x5W8m&6R&~ZUbO|uZ48h221q4npOu6<<&BB3IGAkJxtvsgcKX$pOgE=|B-NF`kw z4LXZ@DwW6vb%p?8HEjsXkT*jXXWvO_l3Ix(y{kZtZC<)FZJCvv(OB0`KIaWDq7Ybg zdt4Nv>2Z_|q49(2`>&YOWHae(7;ybH(nXXrA>FQ3G$V(@2n0gG4>9KZq0pS>5WY>=9Jj1gyBD8=$m zX$g3lge0<>!(}-x3+qBL{;A*hhVDHw*O3lZP^~i1$f!zcEg4KPf}_UXGJ@KzglLty z8P2k1d%enwcu&iKD8zD-`8|fb!Va-DbnGUvk$5@JHUl|*S@}FND9jW)vawe~Ktg@M zR76aAbh2_|THkK>rTEBm6g~U!Y~3flUQ4|9(D%I!^3BYGhA3Cs?Uv;0+jE%=-kIbf zbe5OZq|#`@^&7jyj3W>An}eq~{S==qGM8k@p!NN*;K~+pX5wYt+T5#fudfa@zFCuF zjJ@6Y!0x}oU+r?ENF;v%M?0H`BxC7(OfdmB8K9LEzDcQwC~?rL_>ACB z$0MMyLlVdB6K~*6&6~DI>*BEJ>?=^S^fYT-lN9Fb82Jl?U)$=}o=k7|t*5;?a!?;& z?csNvc<0vd2q+5o5E7OPobLLwPVAYkJ50M6q=u&!}{!*cVI2J}_P0RTey<}517 zKnw1C97Jqi4a~#B2U$Qdm1c#1nu4O0-jho%hpB#cUg6XJc!c3#zZCOAe@s;1BZ0(v z7UKWWtTK~B0;~$1+Pl<{vQv)N7|z7PVl1Y7o)4#2iXCKb8rJsR+NJ6wXHxpw?#RnP z?_?J{w1|E{)!8+KJa7sw3X<#tyTDnV-0;g1Q&gZdEU<}RxSDNy)Cb(=yba_^ZxivX zD@5^$PBZ6Yw@L_bg+|#_T_d%^-Md5<7dVmMt)Z}XcAZOv^G^*XdLf3}H1y;jTUJ?{ zC^H(Owb^l7NwmfV>fAr~jN)TSz`y>u1ypp6x+Rf~a{f(qBKi_1w&^O#z{o>3HM3#= zkPY($5_q90S~+oq79Tv7;2SfKk`6e=ZtNavktV!zn&6*Jc?f$UNSO~8IH&^ywJx-8 z)a~BxFMsv>$nVboMqhShrhs-hrTL1Sg?C_i@0}8O{!USTMTqLE7tzuR#xdr0M zQS++xNGAD#Uzgx(*>OBQLs_YIRn5mDySw2{FLE^H)Ft?c85hoyAcqUMb zJJT>?Y{q_9#XE6UqE`mj z-#=Vs8?-3Dt3N!h+3a~ViwClO{?fC7{g{%(_xbw^ymxD5?z)cI=R!#>x`;%owGR_+ z58uXHpQ7?SB7CNfde5LZFbD0}5&MaXRGvm%|2fKWcmLK$^vgoYLk}GT)A+0TZPX#w zx9NSqTRG78=~BFwKXf~??BdC6Jtxid2Hm{X_uFe^)=9YTTQcl%H$@W{|3GG7T#FZf2|zo_R5O{=+P>7peD|wL#_thh=XzP*|3xw__;1Rg^5UTe+J)0kKk2bV zHq19~4#=b~?l2D*fPX0Y)}KBKUT*3rdIH zls`KOU#$nfjlv6#(rXiN02T)e+?yf!Qm*sTpiaVjf?3ZXbzu`DR61n zy}Jk(r64YXQEGzTVUaEb8sTd^T!Wh75Hm837dk%f3B4q5ADiaLS-NO*)Hkhiw#};o z;8)=WW>N7~5P941QYL61GXy7-aE?!lZ!~BQInEF@@{G>0Dl6i8kMKY{aBVOStZjlu zqw^t3{1d|6RX%=!jw-9g6}aFZc@RHZB1k7?!7l)pl~0h1N~k$$&)nB{tB3t_cMdRR~5xJyFj!9eDL&j|UW z7rUW#yDj-ZR+;g>F(*>42SI5PoE9`hGNH@60_UB{v%|@39-e1Hkd0P8T4BgyiZD8i zD0?+wv@@9V!4HCtS4xsfogB+=2hDof6Tl(y5S+AnT1&C9^%=cF2neQ%WkcnB)1b{X zEecUelTZ_qG|kgswaJtsz4WN8G(GhA^Qw^iOe@1|W^tA@F7ymD7I9tQ4F9fFdts=` zL%QfjMkZdm2YT8&eLFYO2-U1i+wRPOjZD0UOp2ulFZ8s~m{{yS=FdH9o2d7I`d4ct=_+!xR2=Pg~tS{+{WZbSB6A*D>0WAje;4JP*{{OPvOy#WolIvv;2EQbt5eCr-1Aur(L0|+RGU6$_neA7sVk1#5Fap9hCDpIVeWC-r z?3W{3g0L=esC&w7Oabu+<+h_BEbfY)7(=k7KDGqNc0~b49b`CBsQw5Bh3izhU^cpd zL6*{S?cuCmHQ7(@=*=D&pXR|fuyQcut!$Ar{uI0EFP!(L(!XTEm=Z?1kjh_&IE;Rn zg{IOO+BKb!8Uc@gGT=N*)n}=sox>P4dRLdjk z*fz2nT2$V10~#S^k~LKznd+9kToSiB98=IPSp!}Z$3;d-<0Y84tpTNpOFgH=Br<2A zCU;S@k~70`FNnNFs1fk#9i)Fh@6LjN5CtA)M_!+`vU$l?{a<8ktz zBCqn>NBn5ls=naoXXC$>H`{#;?;9g;FjkKS%G8%?jB8uJNpu2tsp_)RYyV+Mn>U%- z1C5#(Pc4+nU6(^asLCk9H$7QFtFYZx{2_i4hA+eNr)i{bWld3V;m?& zaq|*xh2P5BZhJ!JM)xerRy+*iC4TUi1W^;&l}P$JMZlcFIMEprYA7t_asy(vR^dZX z)WgJy!<9n`bY+$bc|kH^dZU$mlAXk{gi}T-1}ovx=$j1^^9ZtL7K4Bfx4!e)FrbpT zTq$(oNC%t_Dlat0VA?G`R3Ra@i36?}XZo1g6KBPHv3`yI6CrXC3|8r^C`j_$w8_)X z@Ue(^<@^#3ttrl~U@cvOR=Uff0pvE>9^FCKh>hqsrm$Iq?^E2@Zv=^J5NV8VF^8T9!GU2PL_j;YGg6NR^UFWswI8Cq4VfJ}^V+8+0S)OFax;s_N ztHl>RCcpmW{plf#@p`fmvKyn0<;_hx`yxaxX}7p0hqI7J_gDm{#k6eOdyc>tW9pNV zHRQ)2-|bl?P-dLI&(sGMld>wAH=OrP^>--5U-8}x2#z9un_YB%$m+D>54GH61=W z9zDk2!j90*AL3O-n#=cXoqZ=elN{#?C@}b-1PUjDl8PKAYues9PxCYMSkd?vtt#i? z)nynQ5Nh*o!%ZWUW0+hl_xsQ5+L$Ofyl4GjUObFDNyH94vqY8LoTPK9b=N7O2+>NJs5U|pTLaTRfg^@&)lGXO&M3%;vC@Vhar*A<-Ju`W3Pd+K0-{i z8~1HWn(B8v6Cb4KyU&#zb9+@gohqGzX_)1FBEXh45#LC{Ke=%1HjFAhu+H+WnuV?#T`8=y9y4~Z=3~D(0V!9jI2b(1g_P0sPGiV>Ou61<0XcH+!^g4~% zyJ@ZU3&@p0BwTz`D7!dD3Cn$;Zt1TP#NAchHIK;zXray9t85OTGufVI4x95=hrK)4 zjXt0LG&CuOVhN^g%(eBB=Rg1K0|ab&;f8iShK^JFmL3E(5zt;gw7i+|zjzm9!yZG5 z&~(a=_urp3y*FtTj0jZC!RnfxZUKJ^a;%KBTkbEo*P$a?Wz4DLGe2g}@&aDG4?o7F z@-J)nKsaNjCav6>m*t?uui9j$I`b|asMMLUh{-6{viD8dXJ{__d+GYabRlh-Rd(oV zC$HO5M4R?_Ho=z{9oCY`n^(;KWLNS6)8?7+$Dz9MJUafdht3h*FEhjf-<4>h5E)Hq zb7301sj}V-ICoQaPml6NzdmMw-8aUIuVys3Wo+Bk=R`>pJmRwhlCv++=7yEyRFC6N za9K8Ioy;*4aw~1>IlJ@UXXx1dD5-95%T97JI>ov;|iuf?8>i!?n{Ru|@in?UyZZKp_3>Z>l%hew&@zPTIJ8fgAxZQ|FP>E}72`4^UaL{o-eI{oK zi(+caM>(OQq@wbNbGlUemp$BuIVG@pxe!Fny(3e0LydYmV17~Sp3i(_noUl+xdQtpWgj(JCuq^e=JlHe>NKk~N z`{}Xik<@dz?wlx72);hf>4lmHJTta=s~Kb#fr7|C2m45PE}3MN`kSB2`}D43o8Fc! z=5sWf!oS^^08mFV^xg<=EnvF9>Gb-LD(|K6C!FC-Pkp?VF7Wkymq=YS*zdGYL%Ups z8M?~P0OI=*a?m9P@Z;AjFjJqZ$?h#OqzIsG+P5=q@~}SL|M9K+UnrI-fNg8|8kgiX zQV!yN`&i~g;2{(VM;(BC@JxW6^?aRs8N2XWoSugc?PalF0*)H6M|jsY%EBoTz$A3{ z=Spspt^Z$-NZuY+fzU7^m3nZ9GUc=KEQf%Nws*k%A;=z@X1PBTj! z@|dC&!52bs0)^}Fl4OP0ABR*ggurY+G=?~v3xujShcI`Avag2*3u2%mg>edn@lTl3 z>V*kqg^6^9iLQsexet>d36~NIm(dHC^9fhT3Rmh1SN{JOS*$~s+9R}+|4n4U?EeE1 z0lGwIbq|wV2LYK)F?4+ZK%@{I<`^L4L&X0svbZ;}?F{|*;Suz&;qfQn>i=-nmKzR; zL*g}Au4<-%C*l29gAli;hY4Oj9mY=iKZO6T&wm*HOZd-M`H!fA>hIm>Z=d{?3*{oF z((IRO_0s>r0Bl25&(9H7B0B7=`8ErC*V>T1VHP6x;Fh_Fwt5o2aQfJ(n!G~WknN=__27L6z5mUb&m`KyAb&kH%HCVGT zllQz||8=~H^0(MzwpUQH+w>o+#WKywemUlko7p^XwxX68eBE_6SMl#RHFq`Mfvz`; zUAb{DndddEI)jUi0wWE*@B40lLVr-HZPx;ePMhVD&T{;|1by)wr-UO%b{Yku^@ELq zmUu?KhW(I5qp^#j%-V@C&t2MCZ-4 zIKbDf#B4 zI6O|#_ArZn_mChXFIf*^^HcKlD3vgU#U@+0q`Wi{q(RCUvgx}~>hF%i3eC|cTGliC z(1CvJ{0`&nL^r(RrX*E72+h(`>P?NER5DHmq(qgK)xID%!%sb5_X2f2kz#YT5W}hTQPw0dfWCbCmTml?YC;`#E)dl zB+gYKItOj;0_5in1bcq!rO09(hEY4_%+9W84BzKFL9!@S4epz*ke)}fdb&NS;FkPD@o0| z^A;6DEc)dcnHBalop!XcpJ~9$bN9~@FY!O7X%*{SKcl@p{%maYFid1ftAqMI>D`BM zq6G$Qv3a8`mOxTt1CFm1&U~C0h2fT%0i1pPN;9Jb5*#X7H3gA@#^p$R7susm8!tn* z2(&%iY;C=vzRSd=CS(IGpZ{m<+LwRb1`5mIdTK%DW4)*dXsrt>cR)c+h7| zeX;X*@B6w5?Kv;bHN7ZV_7kQ1{~Rb6U>{09H{qkhqML9*c;9WmwBAnpJ#*^%4OiXO zQkxj7JB5}~{ipCB?R-Kz+>e37!?n>XpEd4>jHiI#Wuo`nF9uS}sPUEnWKv2B7z_d6 zsGE#;uAaixeSX*8ci2e1P^$NN-An=sM23PWV0d^wLih#fdOQ9U;1_@S5fTn2B-Cm> z`ZszKrpqqyXxOgF-gS$&Cl<@I>9^AXnf_%`8J}WTHGCu_G({f-Tmc}%(Qzp4sxt5G zLQi|SHJr%j1%XVuDB)}{!Sj6MeTYhPVc$mx{rQt3W@_JIwMXRh(bAA=n2s-s+Dg#F zqyVKccnd*#wSqXmLJI>vL!pa_wp$u_&k|C4dCWs^OXO17TFsZ*6sG%betuY~mgg0X zhBw)o+ignhGsx44?zG3H=3mh>@88o~JeH(U%Y*{xC)2$0l9Ph?i}O3Or5vhl*diTi z>ez3FWUrXL56f@2rWjdxKMg;u^Gjnbo> zQBqre+JC_G0tLl)`z7%s%p{P3NDz&aag!FiI01pT*$*Q&n*?q}9m@P<6ACE2c#=5@ zKov9i(=4oSvTy_b8zIC{88t&H^r5(d10qK8cR=>@e1u8>E#c{#&)V~uBI%~uGF6%_ zXtai`ZNce$X!b*vxj4`t#V9x013)=M$zFu4#siH)HmRmB?@CbuS&e0)9Hb{P982ui z=weFbP~Vi3t*%VWQcZ`bOiL3{4jXib0*|5dakN2xf%$6g_p4s!Lt#g|Byel1{*o_| zl0Nq46Ym;ZLAp-cweY+<3Q(WETf?KjX?)TAwc*B#$ZSVjU%iP%t!{<&W%(ztvrGAb z9_TMylO65%5!*4NT)x=tKs3KTHo)zzU?0Hb!WWt1TXb~ir+T3#AN;3n37Q2M z?%ZS`V;!h`Yn&M>6z?T(-j#l{XaOx+br^hHjRI!CQcorw>C4Nq+d3FK^XuQ+DDRW~ z873OlPO!=wdPOD3P%qx`H7VVJJmlTEjXaTrIiGv33<@>L1<#>+*^!8r4)gt_(o702Z-mOVvh>FW%nGfR zJb7K=#e+ZReF1t4P?pQu}0RRdA==5zsGZs9Mk$tNM!<~1oQv=YmUAU5(?n- za2eW53!jLSvGTvc#Wvm$$7Mi@YnUwk+08tettvNdje}Z! z?pjNRrU*u1blmTK%F*_QWGAgdy(WH|m%sK>1GBJt*i-;26ZH)Kw>*2Ol0dyV^LAT; z(}`01?5pC5CtCVBj?G0en$LWNvpuB5qip;vx&+cQsZ`S9f-JF{?1sAvzrSr&=6<>M zsG-5Ri7ri)8!~7e)l$*Qm1lBnd+r)f{oyRd#3#svyRw?^SMz6gjhQQ{V+M6Qso~F9 z?80bbMql_vBF*zpd7k&tSKha87sMp564RET$T#(cNm_UNSo0rJZmeRLFKhLM#>B91 zU2#W}M#(BSkVl+i>ic#DQ2jkV49VtAQTi=+_=~9o<;b4?RlzXx8|t8~`8@RTVM|i% zgIn;O!|RGT9jA|@L5Z?xgvQ5mpG03mdAXhj^{jrc-hG@3YRI2YTowN;_`XXN%a^MpvYDOc1mPr$GJ?uK7Xj&LLFCC{X8}?IFw0 zC_FUyhbwQtL8h=SCKhHlSCn;_!h%M&Hm>(kxYM3sPl#R*O@z9T;o!J~q9$=(XE3xl zA~)G_I}Lx9$Vb65^3EetxWb}4!4*t(_u|^$nC1sI=HzDL5F@_q-LE6TH zwh*4*L>987xVL<{=8mX5D5bEdMP@)LH)L(Xcr6QJWGX~CW-@<5OeqIh=}J)KNkFEF zLMcu>B1t^5O=K(qYrvAO74WnMz{naVeLP9Pd2adLBr1F}!DRFC6SfZR*Mo=gY77RJ zsILKmoe7u)uRGmPhUpb(50_FA)<2J%qZd`7AdvK1h=U-d)+d=EIZk0k z*)A*cbt=C8VG0jv4DU;KWDc#zp$=}15bW;&t~Xzr7%lAuBRYl$e)Fp|<&Cs84I5gP zlz|gAkqwBTounR%UHTDYQ`##Wvs6mGH%2Um$WWKyYjQCfzC1JCkd^Rv3ZYb?dYOC4 zblDJLEnL-!;P_=Wl3^>VyG+h^l%G7!I9rKj2gy)rR0JMUjC@o1y167RC{M1UzA(oq z62=Tp#IUbIw9q0vI^1_`UP;au&aeJ&MzqqHIwCtREuK*AMp+K!Z*sGeKP8)U> z+L3L{-f3Pqlq<>n#y}ePrU+YYxeyEje9@BvYgqMVyUoVDS~#_{eUPQ5&E&``@IV)J z;=!o{p)Jni9PFqGgFs(QX(Cx5ZYP$$n&GWJOP{VykR~NvPLY8j{!e07k;FvJoAYY>73N1)xB;-(Z;i3t`>`Ihpv zyly#kptj(bJ+VKw&ez_glV&Z?6U4FfkeK4ek3mRJ0swn3p?lwg&duPj%>}@P0tS47 z%-BjWO`(v!ixqw%pgrhC&ke2FLN6Iv=lfKfdF6I5*Ah#Z7GF4-=5mC8- z6N5lIXu>XQ!v2zjhK(V*W?L%hBbm@rYNo4uD;2ZM(6QgOU)W4TfW0a+|) zi;8Wh#e?QXnF=-=geL;RMxII26Y7K3>Rco7g{Mm?k}QS0T+%OJ+ta3wq)i_aI?v>^ z5ZRfy{_Z7{Eg;uoO|P>iPV$X#YQj=CM57}pOGGEtxy?{po`w2=OhYkErao6+Lot(# zA?1zlTQ0-Q^&76KTFJ@d(a zT0&L2Lkq9}te0=J%}op>;2&XNuF=2sCt}s})82Q(``%A$vaAoa2zu>LLC_B16E7eJ zf~xGukURif;J(otK!+xRXa=!)5{crIfZU0|-9#W#)Y^UGD#;Kv`w*CC2t#u4&_3Z% z4{>{msim)AV1}!Sh=&+4oJIEwWDJLur)WPEykduHpE~qg)3{W{txsp!#eF;}L3oOA zD5|?-B9-YFq*3cfiyP7LkY<=}R_n4e2R)9aKEiNTMlly`uV44rD&sG)0c0KDz#UR7 z10ZTfU3Clq(Or8bsGb(P>xSK^t7zv>Q9~?!@O8%<M!eWdf;=PfQvG44XuSOs3Rs1iQKv@msYHZ`0FX_NpZKLJV| z0XU|hj&p!2qS_E?ESao2n|F5O3Ioe!8l`Aj4L&<%1i;yu!3wc+V$Z;HkXfRb*1f|3 zxd7x%ks2YqPma^y;|UrB{3i`YueQ3`e_>$5Cb2KU#xMYm(=^xx?2lBdH?D(%Q87I@ zuw*n_5&$A}!9^CE1kV!72sXyq%mMfym5`~DEfB#={#;4G0_NXz)WIpQ+$G{{qY4** z+$8|4jzr4`cy9j-BlO>r>S?z63pY9nz`b)p|NexbOgUfZ0svnElrNTOk!E$? z#&QNi==RZ3T+nqdLAnF0auP_&#(h(Ky+O<{x5%|hzA?^OJkpEkOmzUbXc@&6VCvLm z-q-m$@bkR?DyA#u9blvL5(mM-6J&%h__X{Pf2CxE6y!L!c7w;|Y7j&Dy4SAln&l1G z`FcP|n$>hN8VAhrS9k#SCfNN;wAUA4@j|HQSnU@byNS(4FIHpMSSsOB?e(r83(Q)M ztx+dtA=0mQdYg0k_-4JjU5C0G#)iENhUG64u`4R14j*6nTa6sf#*y-;u9JQ1HxqST z;ycUjzY27GGxKe~c<`13nimI?L~MT?!~kkw0yv?_oI4N9fFBT-jWj;2XTQMf9i$)I zcyD$IOm>Nac1a3$$@+IGzU@*y@4~3|Xy5G7o9r?_=Du-n zXE{>8-;|LNg+6judx+x?H=fhmBigXW*)R>_@#Q(Hp21UA2Y*lnsldQ}j{AAc=6{NF zN%1#yO{Y5@LG*n4e62+^gLtYtpw1WeDQT5sK@*Bg7HH@Eaq9W8{0>0YWs(u6o|PVY zaeh3-J}J8ca?#rdO{hT4R@wsq*z+LpJgAF(5^vxOYeP}^w@D7k=`v-2K_F&K>jnJCkR*f@(i~K6vBC8_NMKS6Vho4 z4ox^Wf^4Y4Cb0$pFVA~T9j8N_5Tpwb&cHJJFPtAA0k+P^brBr*yO0bHh&F5)7k+|* z`0-hBzkBjX`6xQFel}9d7>rxE&o7Pd?Km#-^HYMI4vW1)eCO|cn)cykpIgaQDh_-K^1~#FcJupqDQkd9gX-O z#}z_|31dOrZuH{-8fr$P(ZlLdq|~rv0kJ7?3A|Y>l)MHSi&0jTKxt?bd+v}rfS1Vm+dLaw@p+=NJuD; zwVj<@+n<7vk&no)SyX$X@wu*(A?MJ)ZLxza(ZZ=%;>;t=r4}R;LZhiQCrSGoSM0KeOMs z3@)pIpPgm{z3zUt-gUg($}e~N{*E86rT!WC*Q-AYpWycVXLsP|kGPzIeT8ko%Y9oN zoA4H>ZIJ`;b{|vqef}}bcz{NCD|`gXl*{!8$#go9VzV>ZbJZy=A8-+KR}jNoBmXUo zUFw9`pS$41*#T|lV#FAsoiT`mG5Z}tMpNj(WeU1X;jyUw zu)#%nBUK`CeU3`21G2i=`}5`j*dmUq*9wqPQu0mq zNm;EvQ%K!{a` z4?U7C@1;bjQY+VMz1qt!rRTV3An)wl0hQFA#b;21-Bb({V;=VDkLGJbM?91D!Y=T3 zrAF0Q>=ZXG4edxxZ9TB!sFgxoiETz`+BQGyv(u5jRUwiSshk%}-bmv!eJDE~iSdDx zlN!FS^sM(Y6=S29aOL{#%)9X%lP^4&qL6$B zaozdEa>5r(WnN5br=*Pm?m#jr)KFqW@3{9Twp%R*6GAy{?VV&@Ktx`rI}>t$rS(ge zANw>}U%>a%z}Y8;1NLw6eqXzI!tajmyB`0m9csHcAHUt&W?IWB6 z?RBjzCaGg#Ml44ikbuLr49|S1Y8O)&Dc(z$g%_`(6x;l!MNW`)-|pbok16h-RJ|QY z3qOcmwBwH4bpU_a;i_U-7|3zbip6R;yVXvvkA178SCXuE0~3%+@<)lEZH=&9NXb;k z%f9JtQx^Z8S~&L(l?&x|%+nd)?ET(S2FB3`8iN*0JP6M)J&YEE19MIy@|gd`Y}Q_j z&n)Go;kKEUh|2-AK^D~`SI*{ytNr73p5#1gZ(jOjpp=LD6O1pb1A8WdYYWc-G*1>{ zTdf{No<=~Y1DksFBc$CfZLB4Wep1`sEw&c*W`IO)JBu>{PCw(Rq%huG)D$c7cH-Qq zEBO#>ieFZ^WThlS)PYblF7LnaO(TRs&o(OwAKVfJ(yrVWP);jcdV(NT4~X+Lnt0dbk+1M^!Bx!FFn71nRw4c z2X$WGcVB^Gac!#0j}GUwugwGyu&&$C%=}mucaqWWF;9N0j~}cb=nxv;+B3{ap{@~{ zvlPoFA1gMBZ7_?PT zotw(PE}6lhOc)!BrYl8oJIZl5fMn9*nuApMXp2KMU>2`oKXK`cxAd~u_JOsdWK-;c~?`t`>A zzLE2~M6r7#nGpiJPQMXJHHtqf&^6BeTH!au4dR6Q2l6W@eDT-058fxS6se9~hv-199$ByiEF5E`x3nGfwd1227+tw`=A?ECQjm&j=Y>drj@ zmdYRn30e`VwZ?E~kMxDW;S!BKTvhxg$L`Z^1XJAVACd3M>2m;Lvmh`6L?(}a6myS? zXa5VgJb{%|UtVG4Zp?F*@o}h>hnsXb@#|*h*KNCJmHi!)E1TTtI~N<_obprq-g8Pw z)`C=Ehilz{zs%I!Yv07bzATlhBKnv@Dc&!{i-{7) zFXw*X`Kd@fiauND*T^HZ+W&du?t^4Vc-jRujb)VXhsZVRJstP~qu7Cbz`T?@rWW)C z?imC70F(?FmVyiEkW~@UesAXB=JiRIH^)u!lgGV=!8hG>34JlMkj~1jCnX_N$m?NEYU+25LlBi>`4fA z4kRM36QX$%@S2Btloi&G z7W-IgaaR%hRTfIe7<$s|x)5cDed90F7Qci~IWb&m0#~fEVd?2CPk@t1hqEbgS?>_PDU5D;<$?1i=3m-?7|H6cjxb+VjN%81s zF)I!bz);O_Wybs`wZzZDnW%~x4&;@{?gL6gIH$Y_Qj!76yB2g+O0(7yanb-bK*_&c zbXVeVE0ZHal6j!@Q@hS}G$aAfKU7|h@6_$tES38u+fgPt*^PkKO@`o+zzuUG5a|RL zbBIKKl0+#x?VuT^YM1tVCGCYTU6UnU%{N_L7}DLDYEUkKTTPski~tVk-j=4~o?==m zLTsx^RGMQ=Z!*koXqs-=eO7e56qCGtGhG;n^cK}+x-;REw3jF!G@HpA4h`xNS?1+g z7Asi>z69a=kYr&Uu5zxjc?_iT^yCLkpaur=VD@85RxwLXi6SJ=H;XSLrz$2Vku=9v z2U1YY)yx8DKE)_^$?DXH@E>7xR}y?!!T9))YfhNgN16o;p+^bHi)|*KIn_We&hvn3 zTOASklUq&;$@&X|5xd!7M8ko!<`31 zc^>s+AC5rT(%@!9heoNNM_G+PVm1T%o{n>*JpK7;yoNgkAs@wLYWeN2e1J~5?q=!D zZl!LHQ%rK1^jRW-C!u>55ot5n{Y9>-+Ec`4FS6~v%-fo_RHmHDH-6KbiW5V5p4sVK zKc3L%eQT9;=`Opm0)*qy^rJVkx;bS^s2$1?4{=TPYqF{?0}oF;1I^tkRV6{$9v_+^ zoVF&6x&e$fbt|N}IZytT3)Csd5l)fqJW))}O4_Gu2Qg4d5=3Rj}H@|G+;f~p^fKDX;&@OEV!30g{wiObcUk$6b0p3q3U%p?2No^kLygjIs%3Eq6nQu zniUQ~->jnkWZ=t7TkSmN*=m{CVecW^<&M+QOf01_XO4Tm1D|MdfkUtje8@;yD&-a2 zroa)+J#I|)@8mF0eG<@MD07!N&psmVp!iR0 zbCw75d$a%}a-QECo$yq&i%YhPOK*uLY^(3;YlAgaP0 zEs$|fwc0$-cb3v(-y+a*FLBRJ*#`k=`$azY6>YfR>$3OT79J%v|KKjLu|M0lfAyo< z&gsY@pxE!RF!}x({sI79n0i`=fi?%&Kk`FtYdetj=rM;nZm0Suz@5}Jl`04Mt!&*V zG)+#cpZt1_k&Cgd_q&0n5#vBmsYI#FQWtASTM$h%^Tt5KVqvyOcS~+F>v6a&K~2&a z&IJ%lv57qghFa`HF{?+%+uCvyx>1>|VZ4CxbOYPJ3hVtNkrx_P4WRB(>O0=`QDRN%@Q%eFtqV&YnXCPXJjh4*v3x*uJT!-d#BDTtxsU1H z)aNecMcWyTFkvm)@jJE|u6N_(?`D!>dmBRIh|@gCYKon+I*lsvmt*5sXR68Km}Buu z(Pbbh^W>IAlzA0Ogjy7tbP16qb7!JF+=jTf+1rB%L7MD&9%~ZZ z>1iMo(7RVkFiTpPT(9{wa%ad%F-xQOMfbLd;(%t6Mumhva1m~W*=D5xR9C#&S>*mp z%+p6a^}>@JZ-$Hgun13I66RR+yC!DMhpbri8Eh`Noo57Mh-B^ZkvJ`oVmF0IriV*X zy%Sr}Ou#qGZr*k@*yLBKULs-CUwId}5>f1x%r+TZozECip%|ih|BJ{zpFXpbF&T|T z47}zV2(j{nSPr27+FAp`3J$df!u73pxZ3VUwbud8RNQH^>y<|l2Sqo-wNBoMPqeiaPw=o#_0h* z*~%9dhKz=K`s1g~Hg_+qbhX!MThGK>d-{^WBK&q@8f(d<3$Ob&r?#kEp?zkv*>+|$ zFJ$xUf1zuNg!yJ)dzA>Qh}2egXqi8GGVyWHE&MeKz|S8qe0+oSovojUgN8-_U{x^sX=pT~NrM(cfR92eQx- z1-<_x;Ey9mo3zYKbKgkZ58ZbHgVh1puzg+=s#P9!I+G^8#e##w_A%E?z-=h2);s1g9MI@q(}ME=d0AkL{`t|&Sv(~fsR?EX3R;%kYZ*K+Tox75r^9~+O z$Ki0eVw#u&P*efP0hi@-0Q^rkO4s0le;z;js)uLB=nxHDo#SlC57&(#jyE9SH6R~; z4SsS#FLyzBy#$J0UOx1Nmb_y>6}mlfy6T|EslNt+T)^cn$ayZvI@bU>6$kWLnbd*l z0909R;Vb;^J4q7=h3O?b`JN@@~@{iCM{EtQYkowDO{k(E~;EmBNsm(RWb9T>W z7YlJc>erwWkVFYp%(#B)!abgFv-gqvnapC@3(zSV8V`Asv=!(4Luy`I81|TeYodRp z$IqS^|9!K7f*APelsVJcf+Gsgi&%Z`FuE|yB7QOb+qJu8rpC_(PRm5oxI z3StBGr--^Q_UE$rf&X+S7tDr!K!1cCEYQlADdmFS2p$*~ziD9p+h|%k)2=h>Ey`gi zu-9y|n$D9FXsik_G1=-1t#)x5=mh@8TC$C=*{<-1KR*~Oo2&(Ye!pF7uiM|-5nGPv z`mV9ou<84pzHs!~#o<`4&^O^o?R(q70;PQE>4f{UmP(^;EkzgSt-xk@{yKR@;=^Tc z{TG2y!30jt#$gdmg0Ff$#}nUk{m3@fxjvoE*6!3XuXos-b=QAmwafl^Q|oSV@>`hv z?Immg_<8K?Oe}~y20Zjla%rjwtO8@Hoo|a)7{p^0zZv@X=API`I8dA**f_JU)>TI+ zP$NzL2^Z2gQPT9O<|S^3b^H&8c%1CH=Ha|VdkCre1h7~_briqxgOb8u2vgi7W$8M2 zY`6{)M~U6aa$%aj7wj-}#NF-Ql(_w-JdQhA^M39Y+df%B7&CSXUM#*P=`kS+ABvT_ z0IDz6iIb$d&G321-;GFr44G z4mu{I99TaL!kOim%;RG(#4XdUj-VO(>}`eWz#uPZynbV8fj06w_l8p?UcZ5Jx~C)W zHz2Q0fd{p|4(}SHd8*+fea&dT?^Wvxh2Q;A`}P<3^omIre(IC-*xPAop7D!Z2Sw=l zX}DPgU0lar_?ODjboY(K)ZL&@Q{Lu&WEdTm{wmAu1p|Wrcn1uY9zypbtm+8OBW>C` z%_7mF3N$qgFyrh+R{0VceRD1oa0vhXHN&Ru@@;8`Z3_cpHK0UmxLdA=Y$#HN z*Y~{F00%($zn$*&W>yEfB|aMhkqZ0{kWPucoMfP#wjav9Hur{gc69gp;d-e>g^0#b4zB)Ihu)cnU5fcbx;JA z8OsC{;>Lz?F=#j1y;_*#kQ5J|uSGi$RLh*R(T=c%)p(xsL+!9LXg=HBFiYS7fJe5h zCOMyiobO}q=UQbvLucZ@Xr_}htl(bn-7uZ#>rlbRWTfRkIYrg{?0#2NA}&9n5YJ)x z0HX{bMVJ@)-gdw0XSc$2hrW6jS0qRIm=@XBl-<+6LC2>@(ii%y#?6nGnM5R9UoN4pMHgK*oxfg^){dXa0t`}vZeR`Nb zHyf?af9UQl@pU-Xg;VRL5CG+%N z(_oqLBj!tD$RvDlXS;J{T;4&nub;W~a*s&o)%go;1o5)~P< zoJ_k9O9s_p?(tg&!FgxPh#IcLN4!(fI6@f@r<5Gv+KCh=W55^t`wDcupJnR4s$_94 zpE1o)upwBHwa!}EcT;3CJV|7j3pAxx9{C6Q9onI&SVk8BX~QGp67sG4MCGa1-1_?_ht9wdvN5iYfCSI^saS7t>+FSLWg7fs`2_U;r# z0MMwru*U5KiU_z6kKr4pSD){I zmA8iAq|;{}vqcvm;S2FxKvbw2mIcMRb9fc>Q*m$GEG} zV2 zJuy|GpY5pA6f>(&sG){Mlvm3DAPmL=7u1oIw8%lmVxBYAUqdvIAh zoEsn2+2(EE3t;v=E>>E^)=m)y1%?a6utrj3EdZ-M>u0`@ieKx;pPBQ39|x~(iy>=R zlR6cIc%~6ZzQ#G?xWn2?H)88j@xF&VL~z0>({(HzJKmDqgv)E#i6RSBg>TFw`jR|8 zYK3>gySpn?R2383n3Ni?*gKrWvVG zbEJBvRZaB74eFYI>JxT$3+c{=d!2EQ6DixbFH@>m={iituDEr#&EFFdwA!kq zVDrha>@3aG4bzbAc*1c&CPuf2J>xu>%YT;=)5T11YAK|@rPzzq$B_5#d}PRPn=b8M zGKN;Wfc)_wl;yOWbuHDMCYkB;%_!@oIi`W?)|pnlFIjb6eT_kHobu7 zh__Gc^Trot4O>(xC0d&U2|FaIn0&~CK88Sf;v;RN15G-f;|4|2mm7?~5n#+{ZP?3*&{?kLHaC`%k$!i{o z#>;|WMpZ2gizQOU@=ky1tNw?eHg_Vm;(MUU`}IZn$1|a7n`$qk?zW4cCeDr6hhuGZ zs^47ij9Vu^S8RW`M-Th0@TX`f#2aDs<7$0_MT*s)b*f?7ZpLVa3ge7`%cQ1cfT&Ga zXXf32aj^!0V2q2ST~mzKq*p}htwBEu$?SISwdYY_u{XNRuZ`c=2WUnFJJJ4avIw*B z>uvnK_MV+ctUNM(f(|FU8{wbfJN6LIKWh`CqrxJu6i4SPX7x7pzJl(7yQhJk6Lu{Fm(9$(JD8`M%4qW*bM4(O5_E?0fS;%5+xm8xRUrGAiz_sLQ2q?4#f z)mIUDNMQt2SE4%XSv4tKQ1}EriYHknXb&^sG>`))v z>_ujZh?!e@FiJ@BfQ65iLXR!U@1Xqw=ykMi3&H22vwwfsx))QG*wt5D64%$`zdvT~ z7zY%yvWdz@K(OPTSL(W}Z=xQOVN6wo*+#;Bos(;R+_71y?=@@$V+0w&-*_>mbi$#{wUFCwgQT4ce6Y@L4q)Jtwc(a|J$Q2tci{^?3y;g4oRs0GoGL! zP;w}&n@A7t$47F7<$$S1UOhwNxA+nPy2%N=gN<@-`xY=6G;s+r6r+qiL`8iw`7rNe z8YVj{u{S9Q8yN$2Bd#dRXTAhO>>&j(g0)!4^JF0EVmd+sjZRs?lr@#5;yyzJHH{s0 zFwE6LXx&o7tz_C!93*aDn8>T3M=kTURQklS1Yqb&j##J;y$qpqN(q&+LK`7poH&6$ zR>C>+#1=$62FE_8YcScbTj-$v))FVEnhU#}$h=E267$~^%=vg_%B!#{b#8_(DvYcW zd}AxMDM_ECU#hJoH(7EaHXodMuF3kaB#od$t;|6$%c~`e52_@PLfhr&jw)OaITN(` zi&KQ^Lz4+(4A|xJC?oc?l}MaOxD6b3l$^HL6p;TJ`bWBy5f6QLu3c>7rzTXJ>!<|7 zva{_>1uDwSD8fglX;0Q#J;V?vKly7ZB<-P|LITei&78I=RI%EAgTAVRUze^AxyCBy zP?ye6*eBOBG1jQc^NfKCGVn3*Wl7J-!h2%@501{&kgE0ec0_5VG$TC}fdorx*GLQDK}7}Y$#yj<5s^TB@LKxY}bU4Q;XS~8E$t2IF@Jqb=+-RoNM zrbX*L8?^ZC!Dgi&>Lb++&}i*{3f0#c%R@i{%~CJ!Oz^t1&EwnVKGFvnn(nL}L-v%y zH=%sfW{01+a$3gF$4m38YDIB|lM!t-Wx09rVVv(7t+;6sp=%1;_RH5}?=DSc1#D-2 zG2(G3=d9X!SPlytSBh~@ zMN0zicRkN|&ot*fO`Qn8KF9r3KHw>vk^ z_Fd-LYpoBDwK1w+UoFSyZcMMWGcATe{~vAf|M#})|3Wq~5s7%!Or;PAIMp}lA)-TJ zJX%w#EBX|GAwFb#j&iazsA=<3QDbP8B02Fe**pG7Lm>7t7}o zHWzQAG(Y{!3-CjNR~l6>Q~c+SSI<>wXOWXb`upEXYG=Fe+uvKf zciz{=`G~kMqu1z0i;h@If>wpjrlT6*^zGJWeDldnLAMvV!kpITV#SfM@pM)lt6wO> z6OCfWw#i2J7d@x+E}paH{^0T!#)J#&y5e{ym6@Jxy}+`S^=N@Do{d1X5QiNMwMiwv zz#k-Vatecn%W+3)xJJG(%j@{&!K9Af9IxB_pQ|0r>C{TKb3on4od6*AeHLp+=<=Eq z`f}TfaT}yvH=qyCw$KZEJ$=WGgjPmht*t?8ON3gzVmXw)jT=|YB2p(Wy5*<(tdE_L z^_HLg$F%*pG`-N&FmmUPoiI=)Dl}f!_>njm&DSg28*<5Am{?9@cce)aR7w$U;A^X` z)*2yeWcZYTwv%eR%r&C2o0wh@e3*H%^X=1KN1orO(n=d8k)WT3>3?ocNPRUQ+fE8; zYcmL(6P|r*RYB*s4Rb$Js_Mhr2BtiP>o2{&OEKeWazSv&hGp}8^f^x zm!-4!TK1)_>M-|G%_nc9JP zTec0D=de#?_E}#x@x+HW45BZ_(E6g)0?&pFoS8)1oqW#k3^^%)*9s_BHNRMymBJtl?PQQ1Ck*?&vfDs>Nqx7e%) zoa8Rx1`!w&(D_(_miD7~W;(R8f7z#rCCh|JYTEIXKC9<^>Z{vBUJr)q6wj_|ot6fN z7Koh@FGU0Dzt>%!L##G2FA(o(5-wYxmKH^vGwHpKh;|&VZbm3P+T@(6Jzg(6zggld z_n&C?8VpdNb;TRs!|fAosjgQANT?|_jnlNZ3q3zS z!q?ze8pdM!1ij&~K+1OXU2%QvfYU@&Z_FI!$d>l+*@qWFG5k8K3n(J1y{uj!HE8q< zPr`L%0~DdcATG$J98g*7hDxMgFKCr;6kB4)Vh8Po@{+a3*aq1nP({KSGYANYZY&U~ zQHg-+m>3u(kr%e%xNB01t6jp_wS2h{Oa4Ryzv^7i@<0c{UHvHj}Y-uZJr4o==J{Xb;GH^5g;}$ zkBEf2zH5q@Ynu`*x}b#FMMj5)2*5MH2t`k%Y`aC0Ngcx?Onw(eb0ynWLFp3aKc6b* zvHAlA*_C+>$2<~}IcDt>AKj`(g(L7QM-Fi?X%0BVZNHa}RW=++K6xzOZ(IC$ILkid zu?9CTliruSgsuma(Ikq=@i!gNBJUlDxttb;z;+ej^vMg5Q#j^3?3mb&wOyqWb_G+I zIPt6_9aJ~4`wte-TYr9n&ef|f1kmrqL0u+!It73GT17iaOqI$g{6@tk|2jMCLrBM67ewhj@UAr@v?9uKj<%AG2C2%73FUO)d<)4w zXp9hV&M4ovgNhVIMu~h~$hIN|*VmF?O}Ihx8ez$`30o85qF@ElgkoL`gb#9BFNsHQ zAqx0M>-yIGixSwbbw@!O+{t_iG%vWHzYw92v7(_S!?Eu()#y!_Vi6_&P~Nek+g<$a zi4;i|(oQ|_(9@wllM0L}ZVDg_7fSlye|}gH8IxN2gnr;kdeCz1p9><9zn!w(%Lj?4 zIIG$tfiX5+1Q6MNXE9xS-0DMhrl{UeE_?cnu~a-wb@pj7^ zpw*f1^&PM4`(@W}q!$*fXKDaSA}l*re|@6THn`s|I*s_(DuZM3HK=x5^uyS%Y<)YT zf@9EoA>7rCXW_$-Qexs66C=D4J*f%Kdd+q!**HyqBzXFMv?zQ0I91qCrAKDv7c1Gs z>cvAIR|xODQ$}4lL(6*HN6Q9y-}_VvllGxA6D_Ra>0|BHi~$6 zKVh&xRX6$fPQ00TQs9tb&fm0CSr`8v55K@0L8HTMp@~*{&>ihaOBr)lj{Hn0s%cSg z)*^>kB68wi?t4K32a>|8Mt2nh%aCJJp&${Sr8Y6w{? zKtRA2;gQv#nTBsADxWJaKXtNtkBzp@^F{Yar`3KKAd)OGxq|7}@*`7RDugRfpBrIn zCWAlLUL*_UUJYq&N<6D`?dxEiBr0BIHX<;<<7eUm5(p$YU5{`J591<>~e`!$t78+CJJ+xK5c$gPlR!e z-RjhkglN*EpO#~%t>*qVGs{V+Y<_SduTS2ql;p30l<%!;cIg{YyZmhXC)RlzTsKJdi3f|g z^=5id00hmDQgas91b-zEd0*yJ*w^5~uNyzU0vqK7SF}&5U3k>w1^}oFrkD?N#to8k zqAr1Y13uF>L4}{gzVHO>AMt-UhRJuxVtI-qw7efPmC5SIOWX~j2BBv4W9%&oww=KE z9z+!PL&6ciV5reku3FUqIVs@t7)>BlIV?1oxT;>od-DjQsgrM974i`oFqad*MY_*r z=3mH`#Hv4P!O|g0Fu*~=yF*co-w-Lk3ebhBm7?U=@b-{OLb|Y3SK#VbT9|TiqXlLl z&Tml+KVE}?gm*OsEI=3~2Vnr<&$oP8&e9XP&X53dZ4WAXwGIpXtOB&`g7;Yk?;X_L z(LMagmStsiKM>)>tKr1NpIAIedEf6rn9b~X_Arm#xoykcXGUpOA~+7D#n|^G>)I6a zAL4P^qK?wN0}+T@#1gTZM3{M)UNx;*tgR+ic zHfRVq97zF?@hAfzH7=Ylk)oKyjfPD4SwZwvL43`CM@CqSPLv|h)r8%4Xf-9MNNvGJ zD~y8^sF{kd@l~dsC3#WsahKazkCXx-w#)+EU_RAKF3ln^%{nuUb0JlGm5qq#v-H07 z370im1hxb)&038Q!~j^N$A84iuwu{ft5gWJh=R?#Sc-bWuAp@W|lDCWxlEa3qtEPa)E}Y2#Dc;Ru;uNmY%cR3f-`peH zW`{)BWe?m)e=L;h(}Vi+L0-x}<*msn@9{AW*dw_n##`8%hS`e+r#$oK*qJF5=}K?% z8NMqvBZ5&!!9$>p>cV9jd5qz?gY($x5Wdp%z!G=m3~+t@EVO)LM)MtVnmIE zaW;b}8H@k*r{waMTJd?h5fxgme@b4!6!k5V=qkde5)>dNRg5ZG*r%w4lp#Ph(?7gXKD)tqh7-5n8P^FJiF@a_82TEc3hPXCj<}X4cxYd3| zxMfw95rdUXrr>J!F-KXFUVwHd(4L-Pch%wwjkGL#e}2x+V*}i8wo4j%b-2AF&nX< zeUBqESr7Wn4^qzn2(E^V%XP4*Wy_gkeup-@y><+B72LWuYGAdeZZyzR#!DvVq9y{+ zGwQO?9qg!7xF~Zx5xAAmJFB|8Y9_gspY045yCTH9Y}7?eg7G_|o0<$hJpIQaA{*^? zTLaR%&smwwF)>a5;}EANdp1kEhy5a^tO4WMJtz%5>THhlevB;w5cfNP41BI*sp>_6C@w%umtN^zfOLP)s{ueo^?L_)cl{jT z#jh8`1zDxx`>P)S_?c%2O{2c%JOR*ae2k4fqA&qfGU@zXp{- z0d$)^QZ67I#lgG4foTH-j~~EkKLF*29~s!)S91V9L(uoR!6^d(+lOA5AW^$1SlJYa ztO~$|0zNeWkavebst8vBeY@tJVebFIgVNCN^`<=S*JzM8fW_30&ZZ$`7c`S=ppFZg zjwuiq(rs@z*cF8TA18P{fDjYwO+kdFooTz)7_~+J2Mcvuj+`0ISl|S{v2Cu{?3J1o$S- z}Wmu$D9n6vRCiEK;*GT)okQ7*s|P$gc~|+0ztPCc6ru=o4-g zl&k3Fsv{H@Yo1jUrhDgwtOr)N9M%js(8Uf%GnIF_oj2-s_xeT#a%28Ou@J{eWje@>dA z(Vbt_sgtWsldW7uLLUt+IFwdj{c-*YMqIA0+WpOQSgAI|TY9HN$NQz`1!vw=ta5Ta zBrbY9-bh(SC^&?9<*%^{=&2tR@$s#rIzl#$IwGw1L(ogtYo&tA80?hBJ2>Wp5Fqr->#OF3CZ>HpfvCx3j0n zE^;APxeL&fpSSxwFAUga*$|p&cfaOQygD-Pt}Y?F6}@P#2%nMx`HTx&1OfC zgVm(VVwWQ!^0jO(>F}5;SKLE(d#`hC)){5~P_8!u5jZty> znW55C%b2X}B>pmKuTBOmd7BPzL4(hMQx6q0`#tl>t3K=~34gCYaPqz1Hq1w}{MUf- z!gBkYO~K!ryYb%@zo+qK-pyS|(wo>lqG%V8@-BewJ-p$*x#Y6a@&WF7Ly{dUN?XEn zaPcnEh9&K476^o7T&0Ep=Uv==8v|8iNs#Cz9ks7>3}GGE2(%=^bd9*xDX+yxR*-$g(*y;PLwOe9pVJh+&7CxjpK3z_Ij9;SyeW{-@i zX+r+0j;0~;R z=Um_$X=nXwg2)%AnS@OEY$Dk83heP_^LVc)dtUT`wfs}AjTd)~h4sdTj97v@ zl%JzS9s&NfvuB-yY0`w$=Ic--`-QX#^7Qf2pVk^_gU7cuhjaA0jZU|XKBw%@ETiRW zfB%7|+RseOgUgAfH+fvf8&yeGLKIcqEGLqgSw|E(@2qEXg}xz^>2t{qgI@Evh4foZ zljJ~M_yE5L$9{zWd(Y=T3-rzZBJ%tHvp|aozmsz->13*2hKAI+a#e*F65iG-K(%hYeXBg1tY!sWIk?|;XlL{{yY*h%7j#YNV!ihr`2M8c_D~@CMTc%5=ccJ_>Q+ite;ncT{H287#%G9Y3L~XF$OH#YK)*pKNWvku9 z^kCS8nojxcvkO8i%43vrL%K(ZjpyBVAO4+E`Roq`xaXGd$LG81^UOz@S3b`((p;0= z?q#Phi6?}x){Ilx7vintP zy&Xr>{_tsewi-eta z>iTuKlE6+)+{Y>sonJShWW$FJ;GMRx3M)Jvh7+Ip{ze$JiyL*)%+!B5I;qb1y!O*R z%yB!5zV2J-XPZI_>kzCexT0=m3sNL^O<2QB6|-Hq@aKck%At?823SS&g60h^=Ghs2vI1)T zKDb|gE!{0Z@>|WQ95^1&ZAnS|px281luK znm^E8BU-#paxFc3+_O<75dFkBc|QUESi7H@ajof1tf^O0kq6dT^u^oY91OCG(Q%p# z(oOJ{0;jZuq-A~}YDLsTe{|fC?sXE`dW3gaEnjZev{-!(FTb41GLFev6Gn6T zXPme5U!PlAI}?4#*kvKvnjNcA^_MF*NItX)|8fH@qX#JcD8vxGA>F@2z5$7lp6;=0 zy4gYlAe(3iShv|V7gkJ~15h3Nzz|SS><8kmVNpqX2sS=2Q0xZ8WPdF76=47$)EFvc zN-seuJ{*8TBq!&h%M7B80^>K%BCWXUn5V;+TC+s5Uz6_FPSU ziVC;W9CDs30I`)r273RvX+%a>np6;~?Zu#D(Sm@` z1VDu6hcS(3*hUY%T!BK;1b>v73itjm1~m(%Qa8wO3lfpC zCnHtBKolSIVa}YIjs0@usFS)KHSPOq4bLh$laFpI}< zr>$pRHr%BzZ6q?OpTwG2|tEDoDmp$`bcYjCG=AOsc6 zOIy$Q(}<0y|Z_{(`dp2GYkSAgqmt;h`42KjUCJ<%}>>k|IgH zut=svvy*MFWTKH}o}vQa=%?nMU##+kcCXA=$hx&!F+!6PQFA8n0BTaG`isw6n#Fr>082z$*k9;H5}029kLuMI1io=L)3>G z)`Q6ax|5#nesiYp%UiRy3>*ao)aG6+bG}j4MxKe8nTqme!YkO>AO$<0iECNg-`&zb z;~NBCf38bP7!r{O$Cz??_J7!9&f%Ei-(~5bUb>Nq^~?fFF7p!MOGmg>I_?%JD@$pm zut4j~<0W^w;a|L8480#{jceNuCP?f+b`<3q#7LA_Jds15J~@dejp2nJ&U9kFIVRuk zKMzI%f+9eQyVBE(Uww_xl&+_+z$i#qc8Nf?lD#F`R z6seOi!=8{&TX`jh*iHpOmKy7@Tk=!PK*WmwtW+!02$lt9wP<$l*Vdzf4Iifp2VjHiXB(Vb#9 z8D&wgJO%S0f#2J&xT>4PUkX1ciGrj98)~#j>pRCVW534t81tzN1qCoXh6=r2NTqr0 zVEX>*d+HY%9~;OE`HdH%S@I&Qy2jZsg!nyE=QoN=H5EQhQcq;yy+vr-Y&w4xdu+s8 z8T<~F@1MykQfM5C@}KW;)oZ_e)`MEeTX;Q=g*{DT3# zSJlM5=mlJZBdujLofg9LSFwhlB6Mh=C^sZ%F4&>A5u^)>B&88rH1hQo1a#)oPM%aw zC!*sKdRZMT#2mr3u(0Sx`}EH-RPIs7h&XjL9!{$f^Mu%qzX%&XlhY$gvq^@ya5ASC z`@F$E)h@;fW5pCz=tj{*cm~I$k6HipcEvl<|Mv&yGrkB3lCZ=Hal9IAts;J-ga7?0 z&MrND4hIW=6Xeh~?q&XCbp}tfEjh_E4&(rOL&zlA?8wGNB&nO+SrH!%m(lF&NJ2zN z7)ALqGo;RF9)sK&TD?jMurSdXOcq9vURvct3Rb#Gj!|^w5Axw7&j1-Zu<{wX?*-N&e{WKozDT$qi5-Iz$ zBBGX_$$K1!#fR<$@1w3wWS~##m|F%sP>x)V%ZM*RKSStOM3Q3$aT1YNkt`QU^2Y<1 zI2T$vO)+bDa^iR(X?`Wn*eLz}LIwySLz*E2y;57AFS7m_i#*l8WeuxeDphB|pSUgh zX)&8dEDLr6vF7EkjEG9Z5_yEkU%1L=raH6}5Uxf_=Tzh=PB zd^|oCDn=)viOeczNS8S) z4h=5Wd?{|P;39A>Ww1aW8ZR})DXnBEy+A87dnvV^U={RDuw!J`UN3udi?`q}*8rEV z+?IKD73*bj8f2B*?U&k~m2(glo9kCtF;>(Dl}A*;9N_jT(OI-nXJUM3@ox4NT>9n2 z`~~eemGSGb=4j%%jN}2niE_kM&1RJx^5x<7Rh)yBH7}&1>&5kqRiE^$ocJqitBRVl zs%h4%8KQ8ziEB*H3dtC2{v{R)7FJdug^sM(kOQksUvl|h@6?1|PHfR}G?Evk;&{QF})@Ith8wZ9}BL^*YBP^=nbfR$PR3~Ozfg;fKE~IJjwzmJQiK|<72Od?#Yta0zx>>N=?fY`GsDdI-bQ2>< zi?m-0f<8J^NGW?ri}GuWYIp4dW2OsXtF~Wj98V=Utd(MqsNNKWc?~r4L(pjuF+9im zd_%&p2Yhb|z}#&E?E(y4+5}wD|4Z?@0DQaKeH{SlqU{u}gpQN#ey;#&Q$P-&gYu@; z{}m9j(Vni_qLYnXKL-eP0QjDF_!$6o3Tm}q!Ab?~??nN4F6cIpE07X1)fe~F*NppToANP5zsDx$h*L&1X-lb?!Ojo1qT1Md!_mT zN&z?qnSHG6c%V0MIdri78cZvhU}Mm-N<2Vm_Wi@#pyE1KBFc~Du0b|M8J?UUB8{zU z)jxzu@z`yveQqQ~42Ps+LUBBYtFs5(g<%moKi(>s@bZW^UT`4UeTXS z+x1Ye=_Z>EJuq$h0eJmDJ41v^(_!0cT!rFh=m&9v>kbrAK)xzLqeO+fM6-4xwzcDE z;|8ECU;s5h%D=FyseFpD=m}oQM|BU9hK`|_0^hlSCDa;8Uq@|B0T|G+jLHA~V&nIY zn%@sS&!y+n20(JI&;J}CCF*>ZTvc(_1bqW#Z1k0@4(bGS`WgT+R2hhZW8sXG?Jj6v z{QhSh3qWw)jjRf+FA&bO8^_GX*0{kiDQI_90HKSH>okm`K-!)B$Jj}xsQu9M6FQL_ zrrTYn?=}ERa{~%doZOjXfV08oaGaKRJ;3c&YlI7F1qhwqv^-ugzSKsGLz zla75OZ&y+L>I zAb61|mw5%7C|>t+I{_%Y#e@u_se~4!&xu5Z2E<&8xUx;N+FChdgcy8VCmq^)(5JP*UZ+<|C-qX{2xj)G?f3l(u}?6 zM?ATNIH_G@u^a+|s`>|p>moopo1SwI&$Y>PHsN2q7P*8nrF_JIb({kCU*pBH$+wD8 z$^QkD6VGzwTNr;8NV>)~vv-k@A~`pK~x=#6)dPhVPgxRdmmz+*DYOXOfIn#p8q7GGlssKna3q0{QQ zy3+ikZbq-;=`=1d8cv|l@Y5Oi`FUY+hP>PFJeOemy-+^XZUOZ7p(wTs?f?gHteA-O z{TfRBj2c%C^+C*RCvhz&8{G*((xx{RnmOWvcn}Akkbj0c(-qkV_=K4n?J`#s0Hpco zH~qdQ@|wqR;V-KRo$m3fenHVXoOPW<;4?@V|4FL~q+?t*(El#Fd}yJPhIhnDB1|vp z+B%hHWm56bdgNgIl4>U1i^@hJ^gt<%C)Sb4Q0%p7L~!CoOS`5O#TYfZA9)&Dv6i3t zUJ6H?cNqzlEu1_mBCY~UvrO#ux6PKTM4!#(w}%dz{y;Crv=eX8Ub}XnIEJo$VxsmL zX)5aTkN-4haHOf406>y#>L+Ge)oh9x!u{-TKi5t*yuT6~Mt6e}$Q#^V&Z--VGx^S2 zmJoww1^BnL6ttevPMs2WSi&=aKg6|a8WYtbNy3Yb_3S#Xf;KkAdwcYmTgL%GsCB@# z#be*zK{V=!u{4DX(bz%~!=%Z<)$N>5VBdtoMOt50wsz;qb+IIK89OfuzYCMVY~q-r z!Nm8Lc@@Tq;5t8qiQ}CNlwtnb8UpIUIx&QmUo|n@(l#9GCq;_J15pg;w6W8OuMVGa z&_$WpXJO(B-0El~wP||}A2;)M6ExatjGAeK>3{lM37s#0Ai(9EV(X&K zF3TZzR-cbzZEbRP&IR+7ag08$Z-Njk_=FiBJ^S~j!Oo@@BTc-q{T+il(EvlT?z9}!vTgk$ew6!0tyAdV#?68; z&BOTW9VRxsqu7VTdLmvF(B4w^FcmN_VItixkuXFPAJ~1m*pK|gG~O9We|VK!gP?F7o~shU$@puf65~5MZ{gB-F9lAx~e4FJqac} z9C={4#6^Y4(iq_zD&L;#hX7eoN!;Xs2zr-ayhvtRE5?t59Yk^f3(i8h<)?W1*x?v+ zF$>*17~ooOJo@f+`CVwlkg8cofTDk14k6u03@@9Hr~f=Q#fT(&nKYmZVK0^Hu);PG z&8uKB?w7Gdm#{O?GfIPqN-ta%JaFt@Fa-JGzbYw`DXT;)NG`66HZK zD&(OW7Xov~!XgYt_!6wdqeKY63H?Clw2d^e#%z#v`W_x^(kh(xP`{J22o`P6xYk#ugeoH|x~8|ydMS{C^b zGygyhV9H1CYLz2EP~$L8l1VgH@G-4uK;yBYGsONqMk~rWbMd*UrQ}zg>%GcZQwQT?1Ht_`PCHL>W4lv zH~WI(s`*f{*V!iX?_>&cr`z6aj|+cN^qu)Bo6MRTYxCP^$lF+YZ&8%9U8g{TEOXO$ z z)lk=O;##A#z%t*4gOQ5@6amHJt*sY#37nYaFa4YwH!oVesauGgr${{}_EF?;QyEwS z7TOmC1hnd3AfeTB3#qiozKVhQ6>i1#S!|slx_#D?-Pa>o@&)+h8D1RolLdj7t+TzT zX?VhpfFyWX2TCCAuEXL8`HRpr__#+K@m~B_$p_vX^*VzZKwu2hr)N$l{GHN1MT%>L z@}aTm2-!$4>4P&nD8pA)TaLdhM$R=1L# z{P_HH<-Ria(IJE_1t5gj%Kj+mtU}pINQ|(wjMe|IUZVIzZw^^?p52`5miC?To^Y`c6(J5qL^avu;SYAWcO~il*kl=k+NfsPo*PLS!fpz zCkQR>|(Tc6azzV}#>zUksE4S8pbhSl3PNUBf^uxvL60k;V2wZSTlh z@53li^*i00*UX~c)GxRfs}COkNMVuH8uq%3zUdzI*^22U1OW$)mprjM%}_tveY?0% z{AEp^T`Z`f>gL3yN337?uBP+kadCEYdRWE?h6&WY%mQ#pPx}`-Q@B$IxiyLrdI4E~qGlAIr zD$xRQiJXi`LU<|yY>in?#+g#~gSCk8ClP_HyB=N))IDtYlTiGL0gufn$h%`}g5QqY zSDru0f|R}(i9{2*Vd%{vgCw!2WW-+5oCIrAx>aTqai@o9ikW$h@%+7D(wC?Bs6u+? zshl6_^zG&)SctpI*r(6O?BX!m#?ByYLa3m_U7GzRw};a=geg0Kc3Ems`=-4}dvM!JN5DaQigDHLxHDK*%(m97UyqUjx7(LW{FcbzOFX z8^yQ}>On22)k|s^XyfiBJW9(~^AM_J3>&moP%QSJMv>Ak-*ktsH zlF|ysD3Z|$vC|VA>fB#t37t?Kp(WN*PzqeD1Fj_JnUY;cVQ5`MY%Ezto+8=`X|Wze z7K~70&lm53%;VYQdt|xR61mn4xtZ!&fAz?ka&!uEpLJbPW_4wi`in#{KujxWvWY0k zu%t6mgVg^7X?Q@E4>GWSWwSBo_dMF$AAn~t|>-e6>9N`$B7%9LLQ6a$L9>s8l~0GwJNtnVTQkB>)US^S|A(Mih(Vb%(lsMl{4Z?Ek(rmu;fnO@1s)o{9q|i}JlRSslfIYLuj^D0ayuZmzol#CcT5B|wAxvu*?x=hBd+wU6foI&p8KXHGZg}-;jOfD8~mq3j79T=9}e+(Ae6BXeul&J5TS%=2; zqm^~(=kc4CIF#mLkLQW#CabNLao18}Xe7)VCk%xa!+{0kp{1*T%6xyp=1xgZJj%$u zVOvUW6cgp2?qE=}{N3DQqE0^zFxAH7TNBQdiwNFowX(=#GTqvuGF3L3`@Eru3X(s% zRPrhUU&<6+GS9M#)$+8AI6dsw%O|hOE$vJ4LQCQT3Jp;b)vrn_0QaPfl5D<{pE<*=(OiVZcr*{Ag3IbXTL=X32{8J|Z z(1QS~#RC+NpgWx&h_rCy1qiQqfG}kM%?;>?`1mXV+6V)pxeXFn0R=n}%SS zI#qK3EUQ7x9f0Enkeqt}1V`gc7}jAJM85XhhA;8vOz+9R$F? z0NvM(Qdj_ttwxP4D0sLBRCUnq;A7}oqxPGsI5!05v;&40xQzn0mRfe9oWw+K2xF))$%_#C-|LjcDy-T@i| zrc|0H=zAx-z>uplo$gz$g*JYy>m=?A1wph-Sweat0@TnE4s z1W?`q=)-Xhnma+-iN#!SGdvwPzhF_Gm3|$}A~>!n-9YTt%(PoxYN9?oQ3d<}eh&jw zUG$L#fmBl#^9qxFz^iY$)&@O3!^y@QbwFe~Aj=&9MarN}{gm|U4MAl7LhWaxxhdnj zSpbF(=EC<*Q4v2lzN?nT49J5pI0(1#Vw*=sS4Mt=H?gDLbu+zg^St*f>ylHVQFu>W zUrrA9Xzf?>8n0z7mfzaj8g)yLFF#^huBv(z5n2v7ia;nV|6NLWPYGE4pQVKE_o91R zREGZcd#8$P7&Uv80x6T)tGu!X4uiFVvin&qtI`y^gqHiViTe`n+pXY(I_7c(jRV|@ z_tM*+q<%E3Jx`fDKX~!{Eq~&aQ;vf|(q_8Mp?LOT#_XY}$g0KrcdsKvRfLZQM)t&H zkFa8o?D}-S^&Y)={<)fB(W&pCN;6x%IoACLlKs~#fn~@4?!uq|0ATU|t)MD~!!T7y z#dTq$@x(%4;Kc<4BAM>LtD%uk&%s(XX}kSd{I*jCll&F^`QopSdN~FhJ{R<3sz83H zRU@+1k~y`f3^-=1bjp0}IIIj_pxr#$Z`J;zpgx^>x3lfMA235L)M|f!Q@??PgBVjZ zki0Md{|f4RbW&Qwhxq1wf=d(2J zV=DvU$2{yltt<0m=}fOI8$ZAnUCswHw*%`hzOQ1rm^$+{`3~>c;-Xe${9hl09M2AG z%SrS+Oyz(`<#@Yg-`$tAOTM)D7_xo`=(zo_eTwz3eL8adDg;f1-7x5Ct*Rh$BGo8^ z$5r<(Ihyc^Z8t^;q^x7|H8z{^b^DUyHtgHYxTRTQsuPozfvKT&6#1T`wLF7@-+l`J z$0)1N_glAg;RKku)TuggN+rg3FHP*D>|U?#dcn6A?E`hABn*9b$*QgV_Rq&(CAiPV zyf?QQJ55Ql6!5iw_eaF|eR}WYsiQpWk@8V07OOvNI!1ddYZ3hE^hADi&H1qWuN?;@ zI^q3oby=uqm0d0u=E`2Y(&f=njg|X1wnEB|JT_051FE*B*4l^Dsic$fkF#|QG4prx zwpwcEI$xujUDN(r^0Oj#_l~*-ryqy$dEX7Wa=oV;IG<%&d4Fg65Xc(;xIn(i1-JE!s32 z{yVo6L)K){B6*fCXRSrL*LKtM2dw-M)eQMr;Te6c*CN^lc*lz?#vbQ4hUl-+JAzUy z>%Sq)iK;W#c`LVgLNT>nb@esyxjNqn)vwZy$6QjyLXo7DL^PDHV_7|CcOw$ z`*A@k^|bXu1NWY;?^WM!f0<(omIirb7>Goq^zJjX;&sZ%S16n#;_TH(rN(<`SbDSn}bGVJXcGTRuS6REjP148LVN)zmar3LRzCZRmQ zV@`QxUE86wprSJKS7QP&LUfz-05=;_I|;>CHSjFsge_UJmQl6s$p8$ZjR&wGK`6Qn z?X8-MARranr`*nLRjP`ErE4K#mTmXes%C04LUY5vnUv1&UU{xx(h>hva{fJ~~n!}p5(`=_6-kRN&B*%dO;JN84h zc~u$~^Wbrh)nPhvM6|L37%ii7q=LU0pfMjuj$1s+%&eC1s=x*hsf3Q94j>*eAfm}W z$d&!{iH?gKA;ozxBsbZZmL2r<+Kf*M5Pm+$IY|*)Rb$Bd;~7DQdns)1>47l-SN&ODlZ0=H;@XO}HPqeR`v5&2x?&<#wrPbxy+j2kId)Y5ox z-WkPbF9^CK5!JePLM;Y%W6rkkAwjWECU1(}1mvsCSkgYd_59YB8M;ZZn3tqp5_;@x zT*|*xfo|HySLUWJup#jvsa;>#_B#C{FXkFU)P`*KB(kPzbK-VhknNqb({K;-_1bsF z)QH!9li#+DRn(C^qXj)JZksIx!~?M%DQ{bU%|*^@S8W#Zi`ob5(y9AUAalgY)@k+` zaz*J>-MhSR9%iAa?1h2E_G3-!fH@8d#PFi_+t}9pJ>goT86n=EiM|$c7@-7HzS}85 z$md^YWF%^08@4iEXvhTxODwvE{m8gqSMC}7liy(IR&7C>N$q&F*h5xV5%{rh_u<-i zSvDWr=3)QEpKd1om<^41Um?Gsq@v|t$0u-(DPaK-&T|3I&+YXW9paBhX62b9eDyLo zQppEQ$=^GNAA`5^UPKbJkw;-j>f7#bH6vvXf3v%Mt1h}Hz|pgbu;X#VFn#yU@x{Y> z|M4$_4jG>?XM5;PpVPpXzZ!HrK@vQxV@mHmYMsX(_i*iSKPC5FSMg;|@4vs-ZH5x7 zCjb2oL)~u?hM>p?`GUJ){g2Fya`#{O#BW;VIDdrnKOPgp^~Z~zn1}n}KN?X_@il** zC=jaYr06CHnc)!390a%%fwzj#L$gD%`4WhS>MJ^GG5OUh1ht_fa7tWBe}{sdLOw=A zusefg9JsVfWazQO@Wi;dDT#N{x%pO!)aKcQOdxFRZX-rvAWBBDP6*+zaQpPvp`zgu z><}S6JCDKezsBM4JsJhIuqTgjMzx4#fnZq@t>k9bW1>i$-3U|(wwP>xBKCUo_$*5RW1z!i0Db86BAu6W2*i zEowc5GFHR7j6r9JrE!X>CHIsJCF${nkhg{D42dhM(pn3|+33cVi^s(E0K>5`$~dlF3qn=?j~KKoSTB5v_?jS5fluS~Ks ziK6o$d<6#(@@G0o01P3}&|gWC{#p8y>2{)SxWn{9*REvteqDQ^QWKO+9w45YboH7H z8XdbA9Dv3-fP{0ZK~1_c9ON~Y$^%lAiw?Cm&9HMI=ddP!fyi?B&*O`ftV^&qBK^O- zg-)J?DZprl_%J7t{?E&L?0kJn;{T*?b?1LD0b!hMaev^ia5@7IaGXCt8;*;-z^&2& zNSXp}{1O~gr7dFx4YB#2R=m56^&C<&c>KZWIv^zK0z}c2P6!2+s#D1CCVle%U%Uo? z&d<}(s`+dQEI$%^?~|2;F|~+yH981JNEcNO!444opZq(QTj&qK-C_CBoQrN%;9#1% zw^nfE44kilq~)^Bbi^CjxJoGmXw~lxg;oN)N3vmA&js{1;&kK5kOM*r#fP@PGZ=e7V zQh)`0pK6-)cG<;>AuQW8iiO0M)P=kVo<#_#FuVX)0kU~cOLaJCC9l8+a5OwlfUrOK zKe_Wce?qB2se5M;9?5fIRJPv7R!ixOmF?4LhkQUynRw{CRUu_;{D`D`4>>b zOGKQ-DHv=M8Z*WA(lpURxKv4@Iasi~xr=v7oQAQ7mywI6MfXF`a8zF=1lu=$W0%ku zXqsIHcOV9jqOPH^-yRe`G3iPm3NJZgp~DL0SY3`VoU*wkPQeLKP23|6WyvaCa^P(3lfgoImukVhNRZS=t}S5(-+`r zOYg$iZ9+4St?`O^5Xg2hPs?C}m|2n82L@f1bgvCjzC(I9N%;pElZuQfXkJk-vWPLW z0J=4aQ#2t9-8~szCim@R(V;X2W6nTEo_z+S*e~LRdGdt~!QOe_*8)u*p4}Tm?dBWZ ze?9r%qRJ#2?Zl7miP29Xjq|P+pqAkIKGx%2QN8d+8b)5Gj*AVV(e!>U21&21`20Hs z!3VOEJ2u+E)N6YUj~fcMR57tssWuXWLvNC4KV|&De$j=N;yV-S-oZ3Q6E-vrxm~gG zXvlgGQ@VQHBvGO!vN5krH+)HKUy8Vq*{adS5#@L0xk+QCLPsEA zLZM*6fM&2ir>T{IjVEC<{0SBm z?iiN5*f~Fe*2uVa8~CCEa=PK{d2QkT$ljI(PY9C|nWiWS8W$2KNF5k|9syiygP_a{ zrHQ$9(y8xbdFuH|wL}KX6SPF^RdUMppI^(vs_LD&2ZKN!Va=1lt9+}06L<#BoXbQT z7pfS{_$Yq-7uVy?5mZL<&gse%+g!Y?jImS7uMbFxXPY&(+wg5J2@ft*X@J0uCTXHg z3k7Fr8xi%yBU@qbXB1?lx_Kt~MuB0T;1)7F0Kt=P$NvcgN(d8da5^_^az++ZH_@~y zG>f_QG6rD0=uiTmoqeLu(BaY@WNT?k5{T z*y|SzmglNE!UKW(UJXYdTQ}X6b*Gn8Yluj{AI0G}K=z|6hm7x!} z9Mc=33k$rHJPgybqo2-AaYnWlWki$O#SIGlFOCz)3O86?AQ?SfAva4OT^+OHAq92B z9ae-Gr8{_iJB_J3gr9b7>N&M_cRihUoqGz`(nC@l(ui(%=Rd|c42Kp6F2}#!V?s{= z1{-s~M=BEk@3~1J008*;|Lmg}5K#~!bkpBD|DHr6SaR~%8Hs^l1og&vluahFYgPjP zy;h3-7iUNB;_E;lIvhMQDRd0WM-N?zP}gDCjgm~}d^y!tqrihRRS$SaQuJ&9h~94VGC%X&FihSZM5mekVXND3QgUnhi@i)p$UpL1C;#r?Ev9=d zO-J+oI65(|)8Wsi-RUgC?|4m%buyoepGj42AiYjsVT_lb8#Wrtt^Bq85|3{#dcIa~ zen-Q8SatHX6jXXcvC(n4KH<30;8;g5-!(A&HlFj|>a}d02?MmoooZ83z%hL_gBBblJ{nt<+QTFC*Jo;!F*FJH_ zRGvdR?1Es*AE2*M+a57%uhRZF>_( z)q}HSZn~_lx|Q;>B+JM{jMcf!b5_RKI^~u8X=aA+kFhJhlV&lsascZ%J7q11;f&ZTb!wro z(TNfjJ+jNJa5gQYsRhAoZoHRI<%)s>EKAG5%405Sve-`Uo4co)a=(>P8ZQkH_ceBI z;b5g*Tlrg!a+7jQ*OUs&hTj#i_(>O|u-&xFeom4#f?6=B&jgXdaNl2SBR9(8EqX}xs>4XCi-lkzE9`rILh4RuQg~l zjQZN(IWM69T|3{8t_$ET`PLxim!ReNV{0+gt5+De3gQvym_s8F*v}7A^Ggg{zsujcL}Gli<`r3oUFrf|FB6bLV~8Dwbpj} zyU^V7F}qjxhxT`m@8@lQcz*rF$J@B%9sYE-ypJ^PbGFd8zvzSvfse)%)_r~If179i z<_7j4X8Gji1Q)lWqQpi=__TtqQqccmil5+rqIf(1qQebS!uW61vp&C!`!=?QjheXU zM?*OqYJRY=m=um5(-~H0sj}S>Q(R)wgb#VyT~#u3$~;QAgQ#FnNC{r5w3^)%5d$OI z*>3`dm6jC}N72kj*rMZt0t*97!YRI2odx_ zW=(pk$#Ab0)zqMaN1%lCQLbJxPFx&wVc5iTdUza>-rPEY56zb)^wf6?#PSv#E!jGP z4F2ecI+VYrLrwIuAsk11{%e{BqP$rZ39P(V`x@<9S5SE?HL|;jG*@~E|9Tv4Xahx` zYd33pW7h!;u$1-!9pARc!(PC=bjG{8JZHzBP2q?N_0dT1jlW>fSMqd5Zl!?MB5AY2 zFD~AiF*{Lokk?pJC-Yh{<*m@9p6WeDJ=`BlOr;QeRT+oYsqiNDv^E2_IqLD5lhlQb z1=0jIB4T_fV6PGZ2xB*!f@GM*F|29DXnhOChLMs%)Bq!r2-XRAjfBunb$iMUk}iV0 zIr7dAmBNU@j0flCEEu0A*aQgTKt>>zFO@0ST00tLKkONoeF4(lk(RPz5S1*a1RT^ zy7lBJ)`*PtnD55@ai&&EVRSdlF)ltepT<%4q@v2|I&7*_Nh7r@bzb(9e+}C8-g)Xa zMPyd$fLicZd_*V*)Kn}&`3VfoNWBszTaxpFT&fx#kmfX0OdXFcFd*K+aQ1&7UR^}eS4tN_pAx_ zKgj7n>3-4{^9qm6_|rBWbSP-FOI^>dfvk}-3mJ@@ zl8*wdmf5SM!!^ZO=9HG|t#?CNb#hF99e%q_(_p*y#+NwUn(Av~yPIM@eyapTz@y^y z01Ti^=LXS0XKn2KUN_<+Eq?Gm78Mjad=WLz7WV3sS&4mNr=x^j_{iJxKG2QacA&nQ@N#29!Ryz25}F=t zshqvn(ZWafR6I-3W3l26QsX?kj{&-0s6=xW_$RD8Sur3RPSL5Ug9P!Fpt7F}m-4|5 zkotd2I;%Rlb$K_Xh-|1B%H8#H4tSDY^n4*jzj2ORW)bjUOV^J%W=s^CU?`kfWINk9 z^DIIaRhbv4wJi4Bn*a~W97Y~uf1$-O`MKobIy+K(DKfwP`v-k{npiaz;gKY4&w?%n zMQvQ^-1GBL@xTo;A}PUghTEpSjdFDLKQ$~FjPkAi?lJ^oDR1fuJSje&07vf*e zX+FZN^#Z8}({B(2^QetW-ayx}@4{3;p)J3IUMd9JyB~-JnpGEzr8QqLq(-w%>*2q}lt)x*GYi{K9zkrfw_>GQ-d zehWE@N7nCvEBH;_aw0L%q8ly(Ao5YJPSNQC(e@`=U^wKR8c{_mfI=WRTaj4PPqhpl z;~MeImXt^zm0=#lq{eO%mSGah$ga+y$YUlim9-lCo-#b(7Y)dc+_YR32-4a`(~c;K z3)MwjgoYQ}&_4;rqhRq8YD^z;EFxGG$jf8Q?Cj5d750>P(-oOo^r#kh0}Ps(GP1-E zNJahih(30zsOu(*5fKS^$gDJ zl8nKS?3C^B1B=0$36cgO8(-xB>AnHjG1U;$NVO1AD#7gcm8+^r2NzkA?KPTC7;gkp zn`6Rpy!a^95oXOv30G2XnRFt=&lx!qMsuK3dxVpnGBrX8!-i>0%;MJN;g}@|&TKgI zmBLq4PWoAlqBKsDpf$;XQUWwTN$Y!spKuh$NbBUN4WKBpy zl47c!LWx1wdnHOcoUtk{Z`Z}X=ae=s{u(1yTj4b2$6Im}TS=Q5?OAcJ{S2BueTm=A zX*-u}fELT&Cq!~UrMHuiH7AbViNvTOzH)iKXm!I5FfYRR#TL{uD4 zzl+zzGE)DfdFGbhhL@B%>-BmWw6!_`Z6a!VsIpfxsC=96WV%gJW<)T8s+1XMJmj5N21bXm@dX_X>Jm&!I21yxB& zPD*WtY8MUnUV&nbwOt0eOl~^f}vKDn6f;ewi29Z6j0iSPJ1Je9^Ju2 z#8_7$SvR*ziwLcoxn$=Wk*nvF8n};%q>}D7(GUhI2P&izV>9N;E6z-P5I*>@hfONV zDO6JvqhCz=1x);+t-d&br{JP~kCT6Qij+YS5*q_CXNP>}1e|x1{>pAZhc=c&L(a__ z&meQsukn0?QtDI(Z zk7oRhX8MO_Xgtr#Ap)~GrL?1o7@J5CCy<+q36IN#p+`iZM;UV&A%qHS70+*#>}i$W zXqA0vh5aYbOSLJP|0BPt<+o|{v}tYpE5FgS>q)g6nExZcndG;d^|V`Tv|B#3ThVmb zY&8Cq>~Q#(&3bm#{?GxZL3;iVn-z%k%SQ(EAcHoL!P6Pr_sB4*|I23OcgFQ} zCTw&jJ#?nfbfrmky)o!OpO450q~l@_yd!V6vWC40YvHGnVXaeQRSpwU3Y)L({=iia z_A7J7eLf_f_F3&MA2{7T9zNRr&KyvYnX5!RO~F+c635=XnN@XbLR);&Gw`isnwSY< z4spY+4QnsuWhyE8o$t~UxHN3M)zj;VlB_x-?u3^jD$JXRnw6FVYXq0e_~RfrQrZ1* zLgNb{p;iA2IOu1S#=Itdif%LIhKHPJ!Cnnbhm-; zav8m78Hh=vTv!9C{CkuQss(t}^!hkkK0$!mBP~9D>itdb$Dh+qF8`861phSUK!PTq zXKDcDV^^8Agi#oEwKYHm4p0jMswIpVSq*Z_7i!jMOVwI>dk*YAB;kgVG+y9nZjNH5 z3^xKupaCq>x|pJ8qlAbN?Hv$BHfapA!ZHU3>n0nZqS$D^uT!c;{1S&8GW3#m2&4lr z)EVTpem3+Apt?C4i|QRwg@Jel=&ru^&U5+6tC*qNjFzSj%=SQ#aIhNQ5F_`HbHWG# zY%r3$@&sqV$Ob?-FvV~-ZrKZbVL^Te$KBIp0<`y(`}XK^hTdI(G&iT9(o=e8gIKWOBbajlARM=0XA}ztnn?J7ydW^ybh>XZGn|oV;`;a=F@nbru+*70nwu%# z1j^tMcdCn_?@STG!7^I2D>GeuQt=%@3^Weo$0tP}gCKW1AT?=#^&{r>BT&z(f4!0o>KiTZ-$Q&*#%iXy;4F;yDl; z6KREM3CA^V8sVd3lm)8zhC&bOnx4M~kXlTCNSr5y6JF6YEAfcw*+BXhm7Ij~NW&-d z67o!V0F661Ksfj+fA*CXP{IxK4$z0jZB+Bf-$BPD7)nyL1JJet8O(v6!$Aza{o<(k zcgj2*0oHmAIZK3HOT7uw#MR_?7ht0~z?Y9(s#bvEzgq-}gI(>jJI{&(zRsiLT zb>=$AKHau`sLwO67vQRk-Dlgw`W~~U z?^Z}?NVM0z0e^RDgtss*36K}yLg_uUz^{b(RAY6|s{U?Z{veK!hu@phbifuq%5d=r ze5~{seAgV7|9zh(Y6*=@dwYJbKfi_BdXZ;-KRRdS2aff!ZBEK!`c&M z?Lq+ZQL}t?_sfqNNa6&ARMJr--XS=2gNdbd_>fwe7xF85G2+LD+;*2As;?QV&V5im&$vSgyyw9__b59*Yqq`-P+mq2K%lziE0M zYg!*Ud}{wVc&HRl?o{H7o2sJh4C#G-%o0jY$Z%Fha>i+WrZRnCgU1wyRIA3a$FU|P z-3wWI&i*#3`p7t#DgS)h@SAJ_U0O3}vQ!YWPLQ3pSI>-N4ygTMP&IV^$6QXGuU0gp zt@@7L3pFHyHpkz7MSMOxjnDtU^D+9Hgh6+2yfyO=(GFhj~4CJ`}TNL$E%C8mdh=X-GZSUZAMNpyv(C2MQG{o*g85kbZ4-Yn@BfAU9H z(R=oFGllWuJx;BD@WoH(p%=En4nCwAwJ#)}AuvlWvK!uSF3cKypozOq{o9}!yGzSk zZ4|8QsvE6`FJFHBBmVbbABW&l?Fd*U5*sTD=yP{4d8;pN|7(YGc-$YC>zSQJ0F96w ziCVr?_z934#HD@fHYs5@JL_Mr1}LiMxBf zi+AJ8Snb~RssYX3fQmlN?NK za@o&x2gbfiqU(aa>B-J^Eb2h6uPShOeIa)3cYi<^m-|<+SNA-q&A+^Hz8OCa|G6r+ z^AZ8U!T$YwH2q~^x!8Jq@o5E()hA=aR}P!a1UW4D&6H2|_P+m}n|#1IF>EYSh{B9rfWjxnwGmTfZS_@F+t;fkkCu(3HvBdhOhm)kX0~i?)PVCBU+ZH2t0YVuP5t2u0hvEn`>zIyD1Za zg+?q;0C}f6On-qIw@9UM-TLEljB$To-kkA7Cpu~sjT2XCv42B6RSDy@C&SsUF`@^ssQZPtx}Rw~U8Z0seps#)0*+rlDEVx6YDy&2`QNs3~-2 z;*k448q&2hT;HH`BT{eLuGQ-JRc>Qk*||oozE>2j0iBMS=FT)Y)j*&$?Amn$3!TAQ zY0poi6!~caQ%l6HNs?R=z{zI;TC{4?yhSZ_(4Ib}pX!z6O07+ICi#4_zkAW+p>xt) z3#x;R)Ov%RM~@%g_hSFznoD5x;?!0w>*jACCaoK8f%1X;+D3S>6w_WLhmp+A2oWB7 zRY%C$r>p3Gzl>Z;PPuYZRjbl!n1IT$#7!OwZ&mA?tCyxpbG`DM(G6pMXHaZa#JIR% z7|D}vwCHXxsu}3UaI@xAa-(cEUfq6cn%8G=Q@7U35(6`=V$?x+PRqp3g)A8wcR75$ zAzL*(tnl5o-G7I)H09Vc?g~cTh*|kzv36RW1PFKP3r5p*Oh!~KaP5UMZ%;Fs(-Uw& z|4Wu){Li?_um3jrSpTcZcY+K8fG|ve2YX5>3_ynSJcv@)3K9iX^vMv1mrtg#gOWxl zTUe&E_^sDz{E{oRLq%-EX8h-hJ{LV<0_iy0*yqbIQ246`DOJx4WQwTqILRs(YE4>Q zB5_Eo`9fX|htYAZ9DgZAuW+MN^RAt1P~4?{_i*)P02{rLKxfMLU(1@DwyUFQF!wb+ z)7f#O<^Fimxl9(Ve~bM{9L_IV!=+}&iDJ2s%Ck2PpUhBtbw%boXR15B}zpXZ6Rn=NGetY$GxF3Zn zXZgnc!wkOQi=aNypK}+=<98f7@q58SAWznpjjvFsClK~dNa7!3YWaqN%AK(0_H2kd z@+OU5{v)1&wGjz1kRfnrd3eL?#_6sohM4|K7W26>n}yGkz1Uv-6Gt{hxG?9*j{jLZ z+t;Kke}`gQgSF$GBsuca{Z!CGmVQ|68b*G)NQ0;jx)|@l24KaMaxoIaV1kzC?^e_y z0K#}p@A1^>^fH8(k!}ASq?VE~pRvK&F~6RLWn1o`YvtG_B+t0QT)v*75=J7Jy({16 z6nkRvbiKf?W_78yQy~`>mF4zl88bIkJxX#XIe3=xu%-sTdo_=xrN${Lo*Ko>kc*y* zfd@q%O%c346%cclPh`jdNI*TM|k zr(Fxz5>F2m$QB&9_#vuz9$gr7WpnNW_CahzlYOT7Mo-$H`aI@W#?vhUXmKvZHKC3v zZ^cxkYLj<{MhIk%G`Izz2&i7Q!lY^3o655JI+I}i&0CIM&3*HffpvG>1_acSVkgJ6 z(hef^?I`a17a}I&*`a>oB?eDoBi!?XQS$GXIbGgrH#3Hd636W?6laTk<9Hk{Zye^eOm*0P;H*KYN6{{Q{*?Tj&V!7>;w1j`JUnn>F zd`iN0B~V61=z2HSdKZp#dm7|+EG!vbb*QPVwb;I>y{s>pN#Dq_-pBC3_Yg$2ktjP4 z6vyN7&wJJAR9vdRZ`9wDxtQtmT&ngK%e(v}=e!`87m^0( zeJj{X4O?3!W*kE_xQ~Mi&&Lc|9h(!~7bSPL))xqVHZ{s)_{G*f9wCRhNBkk_Dz59P zz(V^s3<~?uds>prdT9d9b#{%UZ4f*Q=d;AGxx-QQcl>g%$L zBI)JSijZUI2_54zR@uhOVcU%EK37 zA3k6!y)DZFno6XAW!wCkM!=F3!=JId`BGE`b#N0+A$I77IhiCe_;2oqU*BjbZd_r_ zWHDv35EREcaXOM}*5rJJL=@_dFnYJ;RPFd*ez=No&?RYMcST<+wy3c^K&&TA-`gg; zofc?~VNJ+0CYMAoW9xeaFaU@|_?P0Z%g2EAP>Qm-st+%& z>g7AFL<=LtSo@XLe64J@&SjPPZt5GY@&yrLzwYYQaQMr2YAw$@%00}nUj@bIW_a+J zG)gQ_@xVRf^EPdVI{e~-u>`Hdy~G*%tUwoxT6@Gam0e&C6M1e0NVK?8IF_z}yj>6^ zesy(3{L`g^4tM@5Xo<|>a&t;_Z^{6rDZi?_i>|}a!H#6tq)u>4I+hMDqq#p1x4K{& zGU(ul^^Oas-jI_?(>`x6BL3FXCNKO5cUX@FHG$UlEt71hBGrf`&*Yn7E2-dWX!97G zErD>Z>qyD3Q=by7ysOMhDiVOfr*`(|KZ;)*j@BuSu$l2V=B;a`K!-Vp0v*&c*X||j z+$qbFN}s7k+>K_Y=k@-aSg&mbV0}IwNj6xb)oTmUk1JHnZ$)&5qtT+7l zWZ!_$hmxlb=$s)V=`!_}@3=?nv?1La-dW(w2#hdMcv-bzd+BLwx@bLw(Y=IF=AePf zzvEZH(pGK3l4g;{Ox@jBAzG*q_AZ4vMi(UOQSXTW>rmQVzZG6= zrt&ixRF{zX_jN}S&9K^p(}MN0l`Ot3(WjKL5kzW?k`kjM2}VRrmv)K~bGDJS!YO+r z>o%VIX8ZQwgWS^W8|u`T*E?8xsILnIduZkKFDIR{W!ymC?-QBWJu5FSlk!ry7bmU) z7xgLbMqI7{mLGl!3;vWR&wAc*Ok=~hNf{lt0 zsw0mEqZf}9uHp8c7=p$>eSFeXgR!g$Yp_&>vw>$9gp<6nEdg?RT zao^c7|5iuQFgA_qLxwaWQnG_UJTZnrEP-!=ZYU`Jhj47t&QSE-AOjDP>PT(xGRT)~ zw>6|hz|}j4Dy>L|fPDx_qGzx)6J#L;+s&6N<{}uIA*6lRojyZeA;y$oln4w-Za1Z9{(n0b1A zRlXRpOj8QJw1FmLi*|+qJE~1EOCtSO)yDkfo~uD#s=C{rMDJnq+fv9j8C#rrAn+Uo zKC3~9bMPkkLQ>x%qhLM!R4bd)TU2$i1FV`JHDSI~OuD?AipYt!to4M%8l)%*Hh)CiT#|J)lpzD(bmRFBZJ=4$J@MGx&uvb;#V3DQIz5d zo$nXs5T~R##k5*>5%0{H7xS5%R9Q4J(bPEU*s0HM@2Gzpd*WkK$O*XGtEk>0;MS=T zL+BdAHZWpOI8L@A4>AEmz$HgKI1%_t%TK#Z6r%ShwFH|zz{cH{m2`$Yd+>(4WROM> zLqD_GwV{gh44cGn(%gY9D3a zK?WHJ4DRmk?(XjH!C?pl4epYl!QI{6A-F?u0>Pc&4w?JD`@hfK=bU?MeSc~_A68Yb zu3vYreQY(zYBz9xFgHPuu*$}@QATjpl)!P($z)IfE9%K>t(h4wXB&EkBD5wg2!OMF zlB)Ev6oYeS*0O!vNEIz?mkDyip@|k38lYP9+~bVUql#QTudhfuV4Y=9wK~eJKpwwc zE?j-?b9CMdLoQGQ1iJ&O>Cc7F%tIi=DrN^DXXXPY@@ua0;jZ#81i)D9AixAzuPq;x z8BUm)k5O3wJ5NF>2)LCg{OeOl&&YF^TF6{!C)QlZu$~1zOyxE#GEXGEk+vtR0GH|F z{4K#^UoTYKA^0RyoN9?)I6&mXWiLTrtQrH>l|jH5O#YLSk!Pbtp@2`F%Be74qGpIM zixW`rNH&!LKh{EMJyN`t1~y&~eQA$;{0x?ODbckwiJ@0jEt_HJ5aVqP_t66+0Bw$G3Mx%j+~wN?u--9In?@K zbAkt{8&lhA=l%DQG$2g55BV}BCCnj?C}C#=bU*PUlyaL96TT6J1o{;UOA@}pkR?%+ zPZ-7|3e{#)6=l74`Z*S>bvy%sWz)rW(+;!n$#zIXjx z!gTHwH9hTWCF&j&>K^Ly5Dr$|YpiID?U*v^-tB0dnS@Ld^(-3oEc^AWX7#Lh_59lC z*?R5Sh7$Ge3ia+A^&a~59%uEQcJ-cb^!|D6y&~$n5$d}$>U;3(d&=s2>FRsi===Ks z20;10+6N%+hn?!;7wXXr?nWHI5_>5}_|-EwYab{-v z22jWzf}I2MSfogH`nO#HZJmZ}zYL*6kl#NKLyRnJui)K(jUsLX>*~$S9Dumnqb6?v z*(*RPbS@(gO%^@ z04n|Mh=~Ks23V$FM#>-v0q5hn8}*LDfXgeOyzxJ83@p1sR{t^rKL@}vRHAaqnd2FC zy#jd_OxP;J(7koe?4Y+nfK_imICD_j6`Eec1mYY(W-iH$aZ=$8pbJ5mJs4GnOvfp= zvZIdfy#aN^fn^EPXbD5WIl!TEnOpYI{I5|Q_375w=``>t>Nav41eEMFQTtXb;Wm~i zJa6j+K<|gQBH zD-WssTK;9)^RD|v7pHl;SFBjQQ6{ zVT;n`{%BQKxxnzpqNppKixl37|lq2{mlx?moZ7?4txPDs6Cux`t1aZm$! z$hlTbr+S!PvXv1{Xk|6SpJJxmK$YRm;jxB{XUZLx)=8k!musJRMn7g&e1>ne%1qfw zK2mSGJ0_Kz73=A5Y2a&>mS3H;&cS#*Nj`=syOo%`ELv6C-y~5!*A{T@kX13t)NbOk zx*q6@NhT(fy2Q~^?yEQmXDXGse$(PFwuUYFj@5O9^HC(4FDSX4$oHDW8Q8qw)kZa5p;THgDV<(UcwQ;vP2?SU*UzICUipttmSHHO$T3 z518@tf9g8L*eTLe!v1skA`+EHS}@gZMupa)q0>U0-0oHI+q%~)ea4HPwVEz-k5_ZA zF6~pH=`nXj;n`2o)qox5t*+O7Mgy-SGEvjLp5I2G`s}|KIi&b&qHQWnXT_>}ihncv zVoxO_Mv<f!3t%Q>eqXQR9zI|fW zTYuHtpgp0gr04D~Ig{C*fv1D0FjvD8s`}IvQ!~;t(A;|JUKR$n!|AH#_ngB}(!H~- zqZWVD>t9$dz)8b_%U@^a%?B&pBIa&smn<&lKSb{bP44v~A{j?4?^nV1@-)`b2|5E- z!nhsV0=9om0}uw<30fawe2iT#opeYZvHIp!^0Y!iuBJ`Od%7MwSz8n_+TAJJLw?_U zO>FNf+lUkZCv&F3Vz=TeQY$s%;jv(7QupEK^|+7uJrdyH;=N$hxZx_a@2add4Mq!;SPD)#Az->oPNa zzHzDrY`xjXQcw(h+C=y+)&$N`yUAl~Da;cufvQn5H^yRVQZ{265h16rVEhFYi7Yfb z)Q~m*7`{5#oR{Oc>$mz%ajkgIpEXJp z65s+9iKWzcG)ocb{!ZZQoN8C<*eq|fU+?KA8uxe(XC|5yXZ0s6!51qW&X;pES%$xQ z28~9;e&-#Ioy*hd^?$$F#80BvD-6*9k&NG1ig^Z2h|ZkE*USdfv6zhr8(5mos6qn} zNC*qXY9L~%hcSlM_JNPgjZ(0viPP4+&{;~U;^ke_sP|*%_Y02p%Njf(FWeksA2UBG ztr(90PTxtGHiR3OReTn9@l6uq09dk2SVL#84 zxT>ZUp&>bH;w(y7lRD0q`#85G2%f9sa9OB&>O`B6rEr881p!~dP+7gz?#nzw;ccx| zc@LLDCOrdcS|)mdI65J>vJ`Yd4a!6H`pQ{d#4@;3*24<^K+{j|`l!(};aY4j!M;$u)fvgW9Ml_skVmIEL z6H`Zw)$;1k4<%}p<=ywGt1=C54rA5~l3?Lc06~808|^N)O2pe*Q2@fYJ4J}%{1Bk@0Lv3M*qaH>dmCW95-(k1b_N{;}~#%@#sE> z=w0r;8;`y@(37fXY9M8UH%)WGQ5@iVRyAtsebIK*`)Y1K%Js!&?ID+PrCCS#{U#6< z>HpkoU-;+wzrzW3;2;11jRMVs9~>nfgGFZ9JXZhjGLKZuoww_@STa2@m$+ljbTE+J zDmR;9+gc%$*YOnlSPhUC!ejsgR++-L1K-8{lzR?WCT69wMI;l(d!gvUoY;zsVMeyrAFlvLC1RK3E$_nX= z_XX&o^g0W3Hd=@D1`%#CyXU72=Swl6g3_Ao)1zdnN)0a^94AvnRB`JcuT4i`Yc+Yj zS6F{;h3lb+b$Zsh_@=_399Iw*+_Z(bNx6u=G+Q+NXbS!E0ZISyPn6+L=r?~M3hQ6r zUhgXrJKopF{3fxQd)M;6Ts)z_SVh9)bD9HSVdW{|7!dL8Q5y8(#Xy|xSpgi0_wiwz z8mriz&oh=nnM8Y;XR@ z7Mwxkn%kaejQlx?=NV6!{*e^rd&&16c==wUuu|sHjF@4G*6|e2`hKztw51Zt2zqn= zY7R$ckx-pzq|f-6i`&_9s-p>ExPvK2VUZ;nhdNrc90g+0U!vgCd_NRhiZHge?^-!g zXj3UC|JAC{3i#Mb%l`KXZbWd?DZ#Ex z?yJCf(LCE&WE?qA+bMEnaVCJ ztYN3EXqjo_1d*HCBM8L?zj&M9)@|rPCsUpszYKasaJ;5P2PwM%}az9>2lEq&W)IXS_w?L%6{ zQ-|mE(bWMx)RQ?T9~+jXZJ5=Qt}FOfoxeuE@ir>trtvLl6+s2?*2}}sa5tJ_{MC~r zVMH3{!q^bVB&J0_^8X+mPQ4U=T|eG9ROVqd{wN5gS=-;?r9G1fDsyi2kjk0iEpkfK zQ~%hEp0ei4pSlw3Yl(2?cWDLN`sq~lkIS1}{>C34b0MaodnMq7$dd*K&KrxQqgCI_ z4LVJi=C%M)fD)MApDg8Bu)wuaye1_+vjlNCwo}4tYV=)uaen_zj{n3%>jn0SOyzGH zHXuaWNrPkiz0?vVc=ZP^t4}__g##1BSR9NnU@qw-5d&aNaQ>tr4?x2lZ1->t14;Bt zXqSWIQo!M4i*u=^H|XE2OTswd@vD4ylSn+LrA6K>)}V(geUz|sQLuJqXgLSqz*Q@e z;hUuKQhm0?5>;}VI~nVwwxKi;Wu3}Xf+4U+k_bmdETCu%>*AhXuA@QYYf_#RcUel6 zMs?V90}-LlPljB3l(@h> zZA`6#ugtGa#gIiLxs;5XMD6aCFSX_0H8oUIzM$^WD(OLqM5KJ< zv~g5tXz{VJPAZ;Me~5?}HxRtF_<3UeqQ9nO2D1GcuV}-ogiP>UQOxRjp54S9JSHl6 zo3YWD{0cy15@Y#kqxNO5J3>Ilih{uuU%FKC?IU!$m-{vw5T4F$J4j~}jQQz(2%R_7 zK-r~pJ}Zz*ng1tlFhwv+K@7)HYcu_%VeN#UogGybfvOhZBGqB6yIJF*KWt9WQw zy0RbdT-51JA(hUwq(-L-?*m+6ct$zDQDu&f_DulclC`+AeV-&HcUcufMbEO7n%r^| z>C;=O78Xz^kQFC~`*R+Fio1HY=VuNfkrLhm5yryc2xpG#3HPjuf;@ep*hQXa~OySbH=Wff~Uk~dAw z?w9eSlE2_x=eP)4tinfDbUU)e;0JO_C+n$io?f^to^-DAd_GZ!kgMA7Om3;xj;~n zmCcch+@jE1>my2%RzI-KIyT(YN6ikdVq`rHO%P+n;3ZGtbgio41T@6iD(?|)q_ffV zHhl9!#l*f&XJq-?Fd4$UPf7GVCa>Cz?04AE%-GSTY?3qxhVJdMa;(M?r#i z|5C|hrF<5@v^r02*ZbIppFuN%f|i%8IK6AnC^Lqa=@NjnPDhl*AVe-|)}T>+M%3fW z51~1ZwGZP7+0Go!j!uq<&T%#^8s=W#YW<9zbh@zGtUxz1Zm-G$FbK^22A`wdsiRqx zsm~(ytM?l$U$(3N-UdURKBJU(3pM4JP; z%;Dkiw*lYi`tc;`3{lXpQE=&l;83@NDEZBiafP|e8kqoa_&b1dG1yZL*YF>G?bLNo zL#MWZK{I`la;9vdhxSmwBHsXtnUY`VcdaBf{q5_fz%MW$b>NS)W0zHQ|4Q2^XF@DYNd3km$a0*`ygSt1vJ4gulr2LkkcNIlU_gM8s2ey-xd zYW2REM`+HGWKBl^J#t*ceBUZPAk3<{gd0j)eF&5Ozl7wc(U5-$$*Yipoq%jQ@ZpLd zy1Bn=nZ?;kz$PDn-y30E4saJq*0+sbTIMfH4(4w6#iRG#_->L@AM!ugV}Hm?+vmY~ zth{Z134PzDM}SXca4Md78#v^DgB*SNkSrmE5jyi;FxG(Zw4j*w5Nrs#Tl=Hie`rQG zFOn9B8r)uXp9`qlamnZY<@2US|&cARbfh4UfPc zU9BI&c8l^q(BfPaSk?+84}k9nNcRKa=Ma#~sif{ebr8@3{nr9p%8qu6YD$$$2ufX$ z|1!PGZaRswo>g`!NzH1^%2JGIxo6YDHwD=!@l<u5Yu%|x~akr4hQoe%ko;BxcaRPi$hHV$O>xw&#F zp1Z0sMe^gp&pI<8LXTo=^;)O`kQF%jKqpaYyEj3z(NbfQ?S6a8vlhQ&(xm|$+*E46 z8t_H%!>N9~)YBRjqqt0YV=CF1Ovn1`OvK-GG>gRNj-b+cr{Y(b<4O9fsrjs`=S3%X zjlbpR8nCH%%j9#c_=cJf3LUEO)773^ET75O?q~TS*jHJ)G5t@fXZ#Jn!wnVJBb9cU|$&!hXkWPs?qDObw*x z1;WClZ$>&}*jc&LfTT%dx&KBbu=RD&k|xD0Efyw%PiT|lFs|_kV+hcwX;QkL_pDTC z71Za_%xdLmG!BqGN+lgJ_tR~RM;MN>PbtKYa=Z^K5>!O*$B**-PZ`p*;U!iNbHDo7 z&$5+2-^Lex+oC!4ZOwi_cSsklWGF6DA=FDox~3zEyjv?VaX$9(#jV&S@uhL#lYts$ zS6aKR>WS%Y8G_4Ia~+6kW^`U=MfwLa8SAC52~F7Uw-E>j2=m$N{FT&+CX~(I{HIcBBW~FR@47{BQu=n`Y4_DlqQ;otVgcD64x^k_ zDE7~+7#`@8zcy1ClE9AhdyP0(p8wpm6GoT8D$=~9gQO&4P5q{Ls6f$BG70oUn<8fj z(d`;L8FtHJAekr6_<84scOXJTw)-xH#AwZkO5dx8gwt;tCjFFT_=}&2MNELfEVZw< zL9h{M?I*>KOkEO@ZRdhgVY#(kmfdHyZjwoOKJynl2ZBQ9V`VEolaJt8J;_$}$;Ue( z&GR?)NX4Fmdz-bHcLRDJ^y8-hgqQc{lr&B-11qTi?`z1wY_Irjkt_853F|PP?e%p} z4Eh%(AArc;|LY*0zD1mP`xEG%reC5IqpE=0=B$;4c?awe}JOy11XQS+#;{pBz zgI+a$Ty1}K^u8r%dcw=2qKqWHH%DZV9f;AXmjsBxTY(JZWlLM?BTKcESyp#yJgMQp z{Xta3G%zEAtV5)NQ$-)#SGhhx`;$xoXH&Z%CyyT+z7~V=FOP+-=*Ox!2*0C%#Bgsj4gT@ zpX8dCq=+KmWhp^EZvgIo+7y$k!RhA zPqX%{i&+`;XKp2sat7cfIp5~d?9DOpfBjP9Y#N#R##@d1^VC9<+d@HH{s(CXX z2RQ7oF)`Hf_)+jA-c1`5De-EVmNF5^8x{_E{;AMI5+ti007)SdUc){MoDdhQ-+1Zh;}-`SBseB zM`P{MRxBskf>JYGxSX7tMdvDqqH3Jm8Y!QIlvH?x!n7V5j8z?pz{J0W+FIXz z46LO_m!vOBaZkIdA(uMZkbLfZ?4X*Tj=h>R6JQx8&z2vfUEq!g#?}Wwar)`evbWM4kQ4*DZx(Ih6XHh9j1(#*5Q#X< z--L)9y}xzp;&3>3!=l;43bL7+jGbcHKe;Mx#3Fs|IhCLHoFo-&rP7?UlvKfPAFB7S zAvJK?KMU9QFv794R~zXkiIqmtduY(?kX6*g^?~|-G>vk}oZJ^QvhQG_`FT&LaKw_;~WNuCVvKQqg)_ z8(5PdyT`PE-n@}8?0%*@>$BpiT9AUcsjc=XFsT{lTv0Z79!l*umC>r#fpO!&tdupa zP{Y%*@-@Yy$nRsXD36Q&uD$-s?d&HKi`U~C$J(3qICwt!%-|^q4`26AvMIMpQummX zW9E-iUiY@mlvK0*R-v#7c5?TetE4PqQA{Xq5H3RFr{HGe17zg6zjL7=p%VGU!}t8yHcU|f5dF(VwviFrP2GT{-ecm+su8f-}{+4 zde>^8_FY5P`?;N3*Lp{-=sJPR1tbvp=bq~0WlGN<@7dS7OP9}=SNqzolmd&-(!~cz zqNjo|n~VOe6_s#*IkeS;&7W9r+jKTrEDD4N9!+|B4Q5Z0j^YlnNPTnh z{<0BGJunhGAMfjZU?O=BR)+5P>S;lf+P~%BxW2w=^$m+G&i?(Q^!~it5s8PZ4y0Y| z10Bk}rNcvl;hvySxqd(yIbdke9v&~yb~_MZI}m+45LF%KGAKap2>sLxMiv5)3krrF z1Vsvr0YY{`XNO{K5ahplfU+wT9SFiB1cvt?`IsMisV8@-9)_q6KobYT z(gXC&L(QVX^p1caeE_UCie7%O+#Em|ETR!fPFWs=z#gQx4M6b*peKaLHiW3?5#VhD zV4qNO#KZCGgJkOgb%tOoPP{-^2nrcHpl*%?!yAaM4)DRp4=ML(cLUu5DSg0?x5M|FZq{Liy(v4|A0p|2m)iou+2HcvbhM6#|PqYT55OleZt%T_dR z4U2uZIn~iLIGF#l6}!LPL+od{T2+8wC#!^BuGQorQKzRtXz#gB#IJPD1K;@Iw1)#N z@vgP8WULnFt|z6o8YL-P75^6x7$0E2?o5xLWG(XaU#O(eUnid*uq{Z{L>)Z`k6koa z(6gRi|8OH~PV|fJ(qG2_-L!Qh^-bIDVE79?BA16J%3ls;K)fB-SeB1Ufk+?i@QBj|k*askg2pG^(el5qC)k7R7({jPV8jacrd(stv2x#{b>Ir`J@ z29kfnCrMDIq_eg_ReU0M80fIIa2w^RpmA0!JSZ@xj3}oz=0=9>rrPB;?@H_@9vvF+ z?mm%+UybXTXL=(nQEHmR(26I@M=+3w!7rp86^P#}m`a?hbY!OV%=sLM8&En_D8vjZ z%tcyy3fg7Znle^OGPOR<#+CK19qPNK(w2QoyoMg{l{ZJERFnq3@EzAyMeCo{VQwlG z6`Y;$+I+92HCij~=j}?Uw{D?6ZnoM2>XgrX>^f~_f5}+=Izd`h?Hc?1$+mU+4>C&? zER5nxSIC3mY{#)+*Pn{Gt*4D5ZAHV`@Ra4)Kb>nk(luokA&m9qC7_7YAC;|{Ryl$b zr|1nRP+vCxKNCvF9;8_5*ZpU*4B19wg~Bf3P)q(#PQy61TB`gq9vsbaNeHyvU51W@ zs%);Uhw@U%zUuQ|aEDy9qj`FAr88;9KF-lwC}-6uvQg{ks=2E0gh0(-?^hcs*wtaF z3RuAMGnFb2!CKuckG5(8HeQ=*$V20)^83*4``?B@w5MN4&eiHus_rZc-N1X!gbu3b zz5LKIRpExvFfwa&gRH!z4uibFUXM(z9wPIxn@;5?pX#dh3uj~jR&g=-@fj@^&^#-6 zuN5n-Rjda%Y8AGlyA{t8XPdj-!;T04d`|hD^f-uIud4G`1f55wc%{Uwh*j(gt9bG5 z+{%Hfrm_Y)#YjE99!pG{T_pV%Uf)8#tp3fW<5eWBkZmNMZI92R1#l; z?_wy)MHIQv={(xV2^vKo%GLzEF5Ny1#D~UeU7~zcJbI{_Bnc1(3I-sN$L)p~ZW!EG zdI_EwUplXGMLm035sqT-hsYP@TtKyxP|fkLqxu0Myh(^sA;giY=gHVPTaOZ=Uk}{x zip&foBS|Cf>R4#5<%B3<;wULG@L7ywC6gcpClqmkP&o3TsXf7}5AeFg-R4+uRJ}ep z7+_mzO7d0>6f#;DuT(0$FNZ@#jGzb~q@QNMU)0oxsiQb2k&r$K zkCM_mA9AC6O-GZC7qUsM6ZhDQ3w3@!Em7YQBx5F$-@hITCy%F+{MbuFg`5%PLyFXeg_p- zO0@FzF2}(U1jR9vC)BSD$_|&n4KlOu*d|zbVq}Q8BV8O*fNwrn6OBh9aQ-2To)E=O z!Tws)0vMZig)UsTZZXQ=SjQ+2lUj3>I1^sLuVZZrjFLUZQIdW)%JkPwERFJGxr5wyWfvY-o zP-5@*XXrGF1#8STG{2p%=H_VELTTA8aD*Tr@9iOd9@3%4K&7uI` zlMQIV+tC1qHY=KgS!r4G$?z)GkOHuv9;tD=Uc9~e786a$tIa{rx0r4)TDv&4MBbpU zY`~`x+26yt>DdRF#h92xA5hvNMLSdMxIq%sA;kE?|4JX?CjSsOcse{WGx&i6S5Ya zgt}nX1%bn1{dck8i(llsVdojW)($y*onGf3&mnMj}lQkE|#mKC;UaDg?>fiF84>e#+DZe;fV@WfX zW$S_~avK_JOfKsehZVu6`V?yqU3) zA0Y1XVgB{ByoZBqAAv5Fc*-(m;>RV_P;8Pc9L%vRk05sc2XjA#eT6*uFR-6f%9}p( z9)9*J7P5E@=`3~ua-ewwQ4P1ztMif1;88=$kl})b=x#hkG(@)WygVM6hVe)x3%v#- zCIJf{&^?E>66?*VN4+0cxD?VU+i{r__2l;De0J{w%U8Z|+6BMo>uiXl zU5KOYC;W3J@UGr~=!t*N-8T)70E=Ir|EI6{KWEdDe30-1x?taW3^Ji_T$9Xc}@#6 zNZ;b%;=!_T$ly`La8_)<^3noZ;e6+l$k6n`60*qn?t^tU5=&{FeutSkyP!hJkmP5p zlJb)(sR3VjFtdb1*it4o11Zxi5&I6wi*typg8)xUA^_j6L^2f;IhA+HcWqM>NrqU>ULl*vP5QZpVrl^R` z=9|U2?wzYyq_vKNu84ra=u>u5ga;}n=q!m|D3Cc#lwHSFa41C^DDFQgW|%OpOkrj| zEfv=&ks>HUjwvBuFZE{3goP;xhbe`ZF8Iokus$Yl10^g^4=HWHGp9T)mxL)hdXPTl zie+NVPhrFk`&?pz&(<|YyoBKiT8Hxd!eoifAj~hq-^NUS!PyIqc00 zm{@J*K#{B_t=MLC2=c1icU2}LAB$!kqZWj#|ALCm6OD|>gp67~sKcB2!E0|D+4`l! zUbP@XK(^v-u0{=ROxTa=Ev{D}V~G|mB^a8nw);dpVWL*uE8r1#3;JA>GJy2ZuDyw) zMbHlr{Mv$}-kvf^A{5*#^a_B5X+`gc|GnPPa1GK)Xaa7-^H>0YLCv!2vE>_hLf2?@ z5CHlWsE!<5me7T04qv9;wK?z|7J{Gy0qQ|uCSQU7{M0uqb$@ASs1pZ#PzThbboCH5 zE^H7aDG~p3f3#Zm-o4k{%`O5T3LFW)UIfeT^H_4BL_FSkv$EIT|KQ-$MgNV1(`ux_ zS=a#Hnx_OXg_lcx&&t1M;Lk0P^&`f&Jw58w(LwJ2c_VPVOQ`1dIaU)@)UceD*@ z;w$sx^W92yRb<___NVyLB8e8rB9=rROjXF4dgzzL&{HfdveUtFBS!Yl{uA_Wl(#t z@fY(Fg$P?X$4?VdTXfle%;hW-#2+~DL#8O=3WJ%u2i;(5St7xomG!0Ry?&<^rp1-D zA`AUI(YBo*m%JI*MwKg$Gv5mae|0LyG!#!nC9z@*2BYL=CgBR><~a-J1@KKY5>Ad) zhmi;-PzaaxuTM;8RxZRaHGW}Ys5T<1$d}tKcTB=I7tCbFOH!hO;SruX2%G-sJF>ml zAO2$`@kMq7H0pwtA?tx`G93bcU8Nv4T6nckFLymEk1#>-lh}6Mf$;uDD#8 zyCur7Q#u0^PoBjr7xo0{a?g88De;n0qwWaz+0a+dEWVQJg~S}g(^5qqcQk0vLcT>g z+`x>>j<}UDM`FnQ(8dqs=_!^eLIVH#YROs425#(eEL~Jv#-40q5uQ5@_O~zns=?qi zd&_SA4p@q2;5h_x_}yY42Yw*O&Jzp+`RKp`aO_z|f>{}%2Yggn@$_HzG+80@plkzB z6ZL?l9)%A%SGi7-e7z~>Ck+s*NA%uTq+kYe?YZ7~*3fCVbW-p(ZpPYWuyGHuSfV+- zhC~Qyf)EC%TrAe3mrEv=kIRQ?X2Rz@3{V_{nelC8^k%;p;R4Ta_Dn6;6jD-$^>3np4ZRD8q zt%u)zjY*WNbD(+V`8kKM5$;|o8cFf(9m{IqSBn!ST`K%m#<%5Ra4U&AOhiD|7ptS2 zm6I%-6BGf{>CJYenD76Hg>)asaw+B#%-c^HXP^1WnOT>V$GH$OIFogy{(_u)_K}ho3p&ymNFaeY5<>_%M0g5DUs}ukgb)b$pP{kamxm`2d zi5LD-!dWQ7A!h+;iFoH=+fPu4BLA?M0{)V+C04nG6x>KlNX5Fl${X8cb_S5mCuSB+ zSzJwo7JHnVdwXvAjBkfVSc6;r$6WTaUm9kq=2!H#0lKlxXwZ7#RS$8NzEJc_!R8PB zMS8G7uEI_6humr1*4ByanX3M!`Ap$sJ@Gc3*Y4=)aI^VQc$;5# zm^b0!0+JLTh!az78(^Vyimnb1BnSGF5N*~JJ;DAEEY5{V*DneB7`KOh9^{#Xqm#FGMapJ? zkAzK0aA{^CNk)-e`)cM_@a8Ze^Ig3DCc!HB*gqBjd~L#vOLnV~u*bZ3`{-xl7X)=d zt;NWdw+A9$eeWUP&3tWPEw1F|07!x8@k%Xn>@a`GZ_jIuWI7)SM}H$G-UhLMfRaCC zOFx)4KR!AwYS&EH8Q-BaG{7d@iA(|*U4chiO>R^~S#|-GW;E`({`*KG06IX$zdzCB z#q*J&>=$xj@bl_BG1qx8?^UZLrxZIkD;r(RP>>q?vohqg2{a5LeJQ3;EDQ(Rrd?-i zoyxW+*D%it(8zxatI@WX;bbY*B1J(oa`!wqHnVL_7$|H4Img}f~_KGm& zJFP`=pWS0dx7o_~2#g=9WZmZl4d5W#vb-n-?V3GeQ$@t3pY6vbk_8+@`B3yEp$R1q zfwY@v=0z3Gnhia!hKkYjB1&5=TfC24o6_=+EWFqH70OGd=oyfOKwcx8+lb(PZ4;vd&P?WdJ3*XlDoSS zhw0avLnk1j}$F zs7bz^KJfK}7#-0Ku$dVT^dhEVZl1a z^LpXO#U?e;OXi8eG5G2EUP*MGMM_CY7@KAdIkIqWO$EAxsIe8+nq*D+Xy%DZ{_sWa z#YkA#|FO;C0c3znz|#N6HZQ*-`w)N(=MuduYxoaxOiVCOA(uq2RcSo_ zA4+2)TcN3RI*Z@^^w)S(*?(;F|3Z%ClmDv^{b!kakz$TwOXX6Pev`xIM9aUn`HvWq z|Jdd~D@`U_|J~=?o5)dWtKIDKzCGQXY^(d-7c}}1!>@0?(;o#~wen?akl+NplB6&P zeHt}QC5bipbzOW|osAMvdZLtnG+VA!HKXKUU6Ea*d#MaCH^*Fak=qG4oi*#*>h(n- zQ|)fQ3eL*?YxMUL@AhaWk1d|{UGJCgd;Q9-&aBzv<*%Mt=pqd&A-_4iGZkButut(4K`BAvz`s_r zzC@LHX=0DlZDgL5yL^Swy?0hQfFxAy0>HhrzJXdvgWTrVsU3b3qj6;It#}ac)26p{ zMCpb!IwQ4}WZ9AwST3o@y8u9-`%G+1erjtOVB}C?7Jw~toaM@edsirvM6#UlU6+*J zIsrmUU*>bnAgv~HKF0}Ss;Y-w$hot&-ICI3Tq*E~6iu_!xe{4CmN3X%Rf|rEuw#fW zAxJ3&wABAJ5gG@;#+4U3PzH$wlirn!rx>J~u+Zs3Cr-t$HYII>7bV)ZVq9UKsX{~k zu9Ntxj_`rY3t_n`8^wtY60O}hk*FkKY3uQ`oO=(l(%yj^7imWl*H(MgAvJYP={2WBU*!0yMK)#?q> zFxS5!Fy$tyW#4fR1j{=2p z6F*zA{&U{o#41uwm_f>~fSjB~{DsVx{HYOk8F$?XwoXl})}brC8mc>L+-H*TjcR5ppKt!al*l06mhOlM`; ztw&cU^4;ZNuAX{BDRO>v=K?dX!XQ0dM25S%@0XjTsY$8&xeq3p^tG|WFI+Pn z;?fJ7w29Qa{OnQEQn5BeY}6;+2Oqf8*;Y=S{YSLjT;}(oAaY;Eee_b8LK5iQ^sXU7 zlwCcQ!)#$@0=8ts`!$;xX|(n+uQ$U~RJBYs%pUQHZlre`<2r4MQ5tU}Rkg^c+GRJEWyFQDyY_t1xCBQ6XztiA+fx zoo8|NAgDKepLwtlyTxXo=vHK#!q8?pam2xlu^rYF!3By-Z@!7(;v{+Wug7!d z(coQ%WJP**8A)l9Lopb)QdFu>kMs~IWg zfQ6|#(Ebe81X**~(by>nG^RHbFjd3u9Li}qaCc@}*ODDXnP3fe^ddy)!NV|fc4%?^ zHlC>8#K7=!n$Jn5&$k4% z5gKlzVa#xl$02kBHewYGVHJAsWDIy@M-Ho{41Z(D!+q~^DKEb~*4Eel0y>MD}OxVib=Q z$kanJGP!|1{krxd&+;oQgT zDN~e2~1(qUM4ygx<7Rfc2CB*NA#V6e=M?S1`qg&3^RV0{K@B= z$N?c<8#$CRSc`PpN&bi$iA>cEx-dQo-zS(bzRR}M`?Wikx-O9OP2Pr!#g;vme+jig zu4rJGto}k%Oba(CuLkKXvf6M4X^^>Jwb60hGjj@gXNgUpA;HYNvE3)Mw0S_u`6&pm zxTn{>we9_T6=-M!7Qoj|lcWQa^7;!-qD_2p+BsbOGx`6+)>}oz0c`EIP2-Ivji~^%qLj#qq)Og3z=t>y4@jWo+f?o_W9e0yB{^KS0P1oHiV6) z^tsoGS`5|SIu{ydLdfIskBA=O+9F0czptINUbaZ6A+{DeahiGM7!Ic~tR^0~uv_2YyzzVBhX^8rraByl3weI+^Im!^pP$o9XRi?$uHk2E6)sWQG5IG@h3J%QWL zzpr}W13Hn+5=*6?aqMN^PfZwW>>^s~kHu@kRZzT>bV+1x2*D%lC_GQ)hWmuAf2HmkIh_&&j`2_e#@P45 zU$iL7{$4$X?%oG!`r{5;flxis4X=<4%OaGlg22-O8q?C>roltZfG0x$N?D{oaYWBv z$k%vU13xe>gpqlPrDyNIGVj0f-d~s^JPdh@y}&6t(YE>$G^bu4R`J%BqIrzC1Acmc zQ#L18j&8}0maYRcLcwZzF`7>?L8IQX$pyR1eB^=Y>y*0?(5X~dk@Z2; z0?6Ke8+Uxrc$}RSzTq92CwRudqcY_(#&2UULL;tcRnyHz#^7Q) zk!Z0TQ}5s?cx*i;lq{A9^(RRH2SE70$}n;*W`RZ#B*qyeZ5kxbJlW*tNv09}XLq$3 zUibNU%AV%{1;~3b`zARuC0UMBE|L(i-&6e|Ne;*|CiJHz^|nXxOP!m9CJ4s8Fs0NW z>2yGi4>nvM0xY&fq<5B5vXsD-LXJ4GDwB2AD6_uDB-HgYW<%Fz1Sha9k_0cClox}v z4I_)8JGJ2nwI{d^Yh!*NoN?ALZ@N6MrY)@z(q>?7oc)YF+3gYK zE0Web6|GPRHRh&=TY&y5OBzO8me!1@VqILbzJ$m|5a&9HxKkWrouzX?i~>Qf2V)M- zQmO$OuAzOdi6A5Lel8|L9%3W_RW^@6hde4F50O_hc~%c-mB$s!&)SwIv5!;DE0!0i zg%=rP#T!#`l0QKhJ|&3HOl#{f>*mCoTfa%9>6|!>Ol<_w7SSWiUpWyUGCU(}o3sWZ55cBK@@88R!} zUg@?bb*#Mj+-&Dc#9qSAKN(8O%TOYvJGv8&H)Y>ouD`5^aM~fch`5Vm+>7+MZ|~%w zFf=F_9m)fG^E5>BDh214c5D!JUzH-}Xwz>R6Zfm1_EKvS;QH(0JhWoXz_`Bff&PKx zK0Py}SEQPsTb|_;#pP>GDe<917FP@4QZ!(3+SX|CW7CM)TIH)+))eQE@MMm$3H8+_ z!Nsb6XHXtmiG{JTHNx0Hyd_Q#f+A$K|4tKZ__Mn8vw;1|+lzem=e4I%NR41u9kyiZ zHKI2q8sl@?ej&(X-mej8@=4EpV#BOhHk9hMb-{7qJ(82 zNYI8gJI1qoh4am-CXPS%%FkJM%O#G@hWW**U!(!N&#>Lfu?;ha%U~n>h;7r*yeL>m z#aRdvaLg4D5sg#i^OqytRQ`EpxKYHlWn&u!pe3|o!CSFpU|ew>nZrxzDh3q{n+2uo zEZq&Py<{rE41hBivCE$=XF#OG5!ER&YVwOR8?tiq{iek&dxL28TDj)=%2ok*BgZTN zNk?9ywwzoJuX*ojqhQ1RPXx&a7h9oS6%13;Bc!59(~9yT92$ zzTwdpJ@x3V*b)U=WJP`C_(CXk?(2V~;rGI!GKL+8T;<$oms^W&_mp%#K(q9 zJI+=q--NU;JCxrZomA8mSQj5$p7^D&BPp(S7A!SjMSpHha8aM=RX7(_h1)XLS5!z> zI9A;K++*bGx~E$NU+p)&PAnxK+9xkjmm@7YYgnDh4e=%XbWu>^Uk~IdL{wJ;>J&2S zkI?88qP;TIF}i}?FmMMcfrO*rrVNiZL$YK;@|_$y+zdR~oDZU7_M&5ff%y5sM2PY3 z-=>EJ_s2eV>9}NrSvd&)|M0}c{|%mKEgz3ZjO0M#f&I^Kwx$`KG6m@Wz!Oy}QU8!U zS)bnTNg{9~4;7*NSgAz{8)cip4 zpIB~Ks)fIHE8NYo+2Qc5LQ>6B&!E}!(1g(HPvd`LIq5&LUc)_-$bUhr7pq#H!(;c@ zVk^UKgh-7aAAt* zh|O+ImQ3}35D)x)ME(vkqv{WHUUDhd%kjt;dxZl~;*8!SojX{0M7>Q}h69AqEWN=8 z%(S5uXYt>pA2>*I!zV;fX)Q$45lF0eR;P^IS$2s?)!=ot`$5!GMij9%+1&>4S71*;zKvc^q}o3vYq)~oAQ!d`yn)z;)T1L~#zW)|iv zZD$vKSO{WH1)c@Es$~RSaF!vK7@2s)J_Mci_gb8BwM|Q*)0S}P_SaVb^l@e}zdVhH%xHEJ%7quuwHO?E-s z%u8ca)>5Q8<0yA<*UuapgcR_i3>=S0UX;;Y@9Wwp1_49Gw-22ZjjBGY7SSO>d5_#PEE3(08;{`A{ct1h(1a}8I_;l6)bPTP%mfPYW>5q7~Du&YOW z^hG8fBdZsXWU1~8iCzEo?Z>e}*hxhTWPxxuA_`QK8?%bnS96Q|fk%NMNinDqm9^}- zp~#f$@2S5%8PTtgg_uoc@*ro7-(ZZ0!Yf299K53RNC&YCr0Uegvp#A9(VxOh+g?7s zU-AO)bklq*PWHlC#pdr@ZNtPb7x|eX3$CEBGxWZNx1N)=_s_B|N{C1_uZX}Hmc{YX zni2u&c@{@|*2c9zCDV+~fWt^I5)zLlqAp4^_{DfAcZ7_}Aq2g=NQedv6J7Nx60fT47QV$CG0edojfsJ8>rFI1 zyZI+!VJTIn`&u3i1(9TSkCtYNHX~(Kod&_}phY0BJ^h~)T7uS&!D`{0s-pzEe0uAe~ zT1CQ`zx)>KdvdVEKFezxgbFB2J}Pc@D00eU8JZSI++e4oU?drCoaU#zG=-DCl&X__ zKz-=Mz?gzuj~;nV^OC_Rk6qX9fWn&c-W#8V(Mj3h)tka6pi4U1p0{ z-JUNdr@M2LzROEa27<_3^??RF?ZpG>KIgxc%@LIc!aGU5y)#=T$^o)^X#`SNS#T$M zxB;(v2Im5FB56*VSg0ei$*c6Fvy@@A9X3wN7ds1>5ZXdeC1@X~WiKdT$Ujg{CX30h z;G-8Qf3C6p=T7optJkxbJ_Cvtu(L#s!o5mXq+^smzoQ(i-$>66k z>KREc@7yG_U>^Uzsrz{DOKX7|Nk)gA=lFp!&5(3-0kE zIchoW(h{X$4!^Xe)yt2}oZK}ciHZ>(&SE}-BJs4K>HbTgYt@q<$!0WX=8mhGd&ks( zB8s-NT3$nSR7Z!S2Ssx?dRdN$tjEH-AW@cn2|Sx)lk+f#c@^;#ZYrSAfXfq{5=<6G zyV)CVzc|Gs`-d(;%QVADbggJ2sfmEIq z)Z$aj6uP%fZJ&{~S)gJqS+-mstx!sXu0>?0Mu#!of%3+f&I~KLum)oO4U@;;VEbL3 zgV@CnZ+m}*W|pwMVf3WTD?rJNFtFl4#!j=u+)nu0_H04WtOq${0H^)g@9I;yR5A+Z zlPYXdKUDi{A$0A`L(GEQAp)C}?n0>?UBgKpe~Be=cdSo|cHny}p}&Kq($~Minnc}< zg$28#dHu(0$bG=PPn3BaAxob7=PhmjiToeRu^jTvzK*8MlW%= zFgONP<)=Ta^IQ$~fMAah2>$U;oeqTqUANHbUAnb0! z=FiIhFX>@OsCc}$1aZWy0XH$7H!;4~p~D6yjIx^2ZN#D0mXsT@e;Jsqo_K-0AR7ou zl?)i2hPjG{4%iw8+>dh?0ec3-6@lY1tQD+9;sQj#l~xLNr?GTx$QbED5fBdah}hTw zO2<5YgJ^tX_$WaNR8ZL3Sr0;)4T;TjN+<}hPXCq=(MwfB$Lma&_%4!ErA;`kT=)|% zRTM5&*?yuxo)IJuimF4Hoxy;Qs@N|=WTKSR1?6VBwvN5F=JO|kjz)esNx+n~6Qkt; z$`e*ikQwoNy?I8R7^Vn~#TXMsMcjEuK@!W+67j#`7tU?5(bLkOW$6w@IR_c2JkIvpWlG3ATR<^TbV$=S@#ur#w5&M(52EP|BmS$@8tuv_a2dU`>|_%n-iIlXXt=m`&#;b)#j?Ag|9yUM`edCND)V zsHZIuPS4H?jObJGH!#X=(8=}cg0bjiYU!Bo@D+bUr(7}C9P4A4?JM4Wrj26BLWJX% z0A)(R=|Vd6fU~|5g{2Z;ddUw`Y8w_nAR1mc3*a%ZGznVjJrn=7Nxwo^3fy9*<|Uv( zqr;YSkS;HiTPl9(Nlk7q3SACIG$|(vDkmu@C+jb#*ea)bDW@T;pcAWLFsWctO{)v6 zVC%2o*s9=qso){2RN0hR zgG_s90K`{FI5gE-+6a$}Rbq(*k9)`{djOQH|2m)$wGnpsl7xdW9zBtjJppc8HK3H1NNI~%aT8{!GqoB&i zD}bvD5~e4}si4+92)=#1JL)xu=GT5d#Tn1*Yj@?XR_5{?;(M%uzN=8 zl7j#kt%8$swZ{ZR+}D5t69ASK2uM>M6$DVdszv6lM;@ucxB?9?q6dAh|3f1>*dSE; zf>#Qw#Q`+yUx8}N>aA6q=1v9IU&NeT5ELS6f^{0Pt&s3M?Tj+3zpw%RSRrFsA*gF3 z)VlzhM;gRlFdi2H8hZdmZGc8A;HIG|90h;)3bgkE`f~*mum=D9+^U|?W>DQmv0c{Z z)8c|yE_-iDY1;1dIfWOgea^X^W4oQ}Z#xfp2VWzbs$ToAC|ZhRYcN2q(%@T%Fm3|L zMTgI5yIdj5?MtmbLOPRLA^CFK=zgcxm)1RzE;jN?47|?Q@+QRrvd=*s)}*Ow4FWoi zZ9b#)s$>cB+nr4*N#X#R7f>wL`(Qw3ckizF>n)xyo98k*t zREstK?VNvXTraX1jw@zkycnrR4o%e;_ha5}8J_!GY}&RwRjc&V}+oFp;lw^-pW2Q7uk;L)V2L%(A&#Bkh9?Kbg@>glF1ySdBgER_OIW*5Ir3sK&BLrzw_5BZ@WJ}0|$g&d|bU49~^AGvLU zD#e;sO#T_EV!kM=3I9TfY@iFIBI(l9?RuLA;`Q~gM6Rhmn}BSu&1$p4u9H2dPjy5! zsmSIE<`sG9BiK9<+Aj4LLpyme3tT$3^4j8?T$b4T>XWZRBUJ@#G~4QvPA2%O@D1~r zc1E&uCkS>BRhieq$%V=Y4SrL@jDnn_qsrFI0d(%=%%(jM!~>#%)~T}670BGf{6a;y zFWdjMR%LPa6CK;;4952DHX=2%J6EqYfYA8v&1!08L(BheXTKe}Xxrb9OyM;&e#TIz zYX&L9$~4ZCehzi1pW=-FTtKFYIQaGMSpH6%{9_lCgdNb1<~7jNv*3$i7k_B?kNPG~ZW*A!s3H7WsDYM3A4Si%XQ;)c$y zQ6r2AmNj1#-$XX}8>i&~NTFrh=ozypySU;Vj`$TF-n{g`PJR)Z*%czYEnV@Q_Zjz? zZam%UIqPntP}SIq6>vH))bQl^aYK@P_Y&@eYzRmM2**ItNMmw&14jI)7d|6dYSWETgO9_S-iW4K}sty06g$apcNQWABPDLwPjQ@kfr}VkFsQ<*6>hH~C5; zhO7z<^k`pPl}BMbpSB4ZfHq?kc!*DBU2lziW{oRCmk9k=&Dq=6n zsfJ$b{Uq-?{}@$MtFKfN=qW0S>3}Uw2&@3Sjw1)Lta=}!Xc{kR$yHRz^UA<*HDpdtM`{V?5uxf~yn2@%)U?^Q4Wksei|x)Y(!6MHa2`CjZ* zYV1Tgi4+oKw(#j-t(po92)Lb3>1W!33i=+`kf9Y7d!pln*hKYPz>pU9Reri@#=}1g|&2U3_h2558EJ&T?#&D6q8f<|}rf zF2qjyv7(^6Xrs69L~R_<(fOXr;K8zw*(UnY(Jj5P`g&rigxMgniSOxcGk)afRz_8; z=^S}mbfB>n{ixpkR#z+XT*AP`KmS>}C2_S~!I17%P(1`ADk74=h-O9cN)}t)3V}dm z6g`qhmOvLOj0xP}LpKZFnkXM+Lq)C!fJtG;UgGr^?Eo2Zq{-P22=#S?Z-rmucT zxc)S3WWAB%K4(k8Ts<`NdicB&Q@sO+%Jkq!TJrvQNBX^)ozyg`q>}@G z`u2_k4-VM}4u>DNL>k9@6@82PL*Syc*CdAM38nhK8f@A8U=9PZkkWt3bx9H~bh#$9 zaW>9!4)SbVEMl|OaUFmJ)CHC#X3uv zts;EP>GS(a|A9kGJWR0bp}DmcYN^m0ZXSqr@w(>gXYd79lPIV3eASmdwdVG=R60cUrUsW5@$|^ zLKrGvwrz@=$8v1!CJf=%ADIH-lf%*;JL~_7gla>Qg0x=G?RI}IEC? z#c{+~oU7q+=$V6mK#z0eI(Z1HBVB;6TeWk2r57$EIBq)N`$@oo60gvc*CA?HCT&E1 zywST!J5X z64`=TZAIeE*oZ`#9MWZk=&_t_CQo74X!PWyc8m!4NA`pp`o8oz+RUtg2_1^U0ODpD z(0~0AGMOmp)WYz3aMuXPbyW}5no33#d=;CCm`=@?PYq!OD6;}o`mz*^!CI`~CT$YE z*(`(nEF0nrx<r$2tV zb)XaObTvR#n*t)2j+i1PKvFJX3RmnbGkXA8aZ6im~)5 zVn!MF;l^}#4tA#Ne!nUm&jbBFsW@HN5VS@8AEpm+8)8S+*IZkCF> zl+`_(!s{*7R0&U<-F_N7p~`w*peLPf5~gPYuAjlgzLW&G1W(5HtciXZyhY=Iv`NqQ z%V;h@US$5>qMg$Kfpz`c>yWy1ip)xaN@r-6Qqd1c$T5F%I}2C^ob3P%XSXrHTqQt< zFX9t5z`|6C$xmx8>FfF)DjkuH6Aoww6{!DMdjR$@ZlHO>CWM1fJAr+l(>O;Chq@of z%;WTq!2P;jlKRW)knP@Rd_(^CLD78;p4#!wo{RxE(`IZ=(cLB_&%(&eF7*eHx|S{1 zK{}-K8p_5cv0n|W++2EbF9%5CQY6W9Kv-f_J_Gh>Hpg>4EVCjXk&4r+& z0DnV_*byY*eBSul5ZxnNU)E8BG#gK$%RWnMV7`hY$qEh@gJO%fJ}0)&n1n}~M8+ms zP~TGzkJYW4G~&57Y;QH@oHi)x$rL398LW{06l;h#Zg}U_puE;X#ge3o+pGl73p-mO z04%mCoYvx+`u3hTVt#JS-uj5s7;yU`$giMvo3WMssf`&mZqDUnPC(lkZL>NZwe48j zu~?n9E1`Hpi$!hrEU&4(IQz#S(u;CqpbQ?z-!6@UMCZ5?zCz!tGb70{ifCa0m4igX zEjLe$?i|r>3c6P6UaypD9SruKU9w)f&%F_aM2aWveiQbS=hf_FP9tZ%d%hoSTuYvu zJy6aqx*z)tr}}i39Q0ob8yg)t*ZMC-3p>tyZXxZW4$T||1Nevot*7mEjghI-`5t*h zn5xmkV29~10l=#c16i)Pq;k8L0T6G+BZ#z#+#)Ea&urTF)T`N!q3|cX-jg=5A7nz9 zbT*LHH#nsT?k*(cX8m;PRdFSbfBV<^rVuPvi~DQ}hVp_R0;!7jT!7#aHwM6?8Ua(k z1*+LF=FSLq^1F!V5q>!mf)H?Z|A>grD5+VqA8&e`G0Esp8`pOGUB*#Y;vS!aWcyo& zFoZh$otZaj9ML`OtMM=2KW zO?_6?4XCU`EjK~VV#Ky-SHd%asHaO$rK6fOqwSBM(nl((gZHb5VMb#ba^u!U6aQpz zc4()Qo=D6?C_u}0Tp~@lc)TF~+?6(iJt3x#1U#nIQ^JX@X7DUB1ixwjNRrgH;Drog zITfEk#W6kywoi<42VN7Y6om7!tha}(ji|JLW!Zc-Gw*arw--M*H(^zdU^7+M`O%sh zOZci4efe}dJDE-yrJ%nf+lMC0Nri|8fpOAe5}+RIBs)zHh7i&ba^YtBdoqs8(dI{C zDvjOn>A-YtcDON_urL8h{51iaFoGDN$eO z_7W>L(yP7>AK%Y(ki(wg5xLf7eZwf{?yoPGB~u$zOPTxn#|L~e$J|Lmb#y%nyk58l zBI%gmWTsGD&dhDwYx^oaUs`^dAm_tg!?DOQtC#I1 zoIQrMv)Aok$p#p~c3`yU4*CNAUhgfrWSoLGl)RKhJh#ILxXFMNj7a!*%?d~=Ih-C0H~rt*(g@4Ce~(q^^8!bG!Tb}yQd_|*BKd4mjhCB0~Ah)5$Y zfnytnnQ$a?$OAgsTWPF7*ZoEIE~Tq=iz9(^>s(C>zLQ|p3=qjDwdKQx0MPJlR%@5=-lf$^X{`P8}^0nXYdZCvVTSLWxV-LZZy6Yh86%qKzbqBlYBEMl{Mh`2S$}z z8HlI&S9Ta;zrXQqS8pL%bIpX~H>>M7$PCD|YhZ|YL%J{^`7MTea!O9REo*0 zD_}})X$T~^zxg&EflXrZdn9ePHI4$p6w37hMvFbPWOdhf7Qc@c zH~$RZ%c_toz3gAPp49ohtCT+NS75|>0~KM-?m43MIml!Pr|ldldr>b1?oK^F8eAL< zBN^fXyX1p0C@;q?NH9{sqZZ)RfXiVn&iUO76jsgZeq>#qQz2Spdew8h# zT@M*EGYVIo-|jl2_n*%&W=HJiM8Zc!eq)nx#EjvjBQ?t&$W@p|QN1T09K}`|&(%ER z3i^J>g$EETOA)2VE)MmQBHIHwnOif|D3F!%8ZcRK{bQ6g^ z!$?}`uc2@u@!HQcSl+w1;x5XFWKhja9(ym zj=_2A`16e8i4OD1z;){sKfRrfybo5MF?&xj_wKn!U(0~+!Mp;m$ntO3djJmLt%l~= zyt}V|`C781%{T4mFg$s0yz+R6J8p&nZDiM`f4xhccToi-I{zPn(6gD^Df>tct05o#H8T4 z8u?6IC56nV5(KVQt$$ao+1zjFtsX?pxjq=e=c5r$#`x#Mtnr;nULkkw+iRfJ5~3Uy zd0oWa`Qh|CxZa=pp04%vHqSp{ofv~Z+>3u94I=;B_V@noAG)g6o7KAz4TPg8o$~W! z#v~D2R+$I)t9}KFfLc|2Y8>Wxzohk$1cm;6<~JW3esMuxLcv1X;4cGQB)8hv*3bok z_@H>nn9|u(GAuo4Qk@1XNhlzY2D z{t$@$fkUWO@Kcm2UDN-}FoZK;F(X}0sM3ae1WprbNp{uACY5|9phfbT_(03^7T6+Z zI4yubp7*-Bs1|G&if{1q_yxM~j#yVcVn`Q%_1!j-V7Cz}4sxfCYO(%}Bu-MY0OZ@U zY5S}iFCkkaEwcWDA;OxsiUKGX@uIJLs9X%q{X2k9P)PNrV2ZtOcr^dZ<2d0cn%w(07flT;bils9WZ z1`Dg|hVooXyD5&ldCE@jHS_pmuPcR4mLnzt9Shwa+OAtPW9V>H=9D*wK`$>+FA48i=C7>K?ZhTDpU=3uG7$@3m?r5AgtVML zAh__FM~rNHF_r^eRnYum?4&7a{j`bq{9<%}Q2NiR*O~+N;!7z5l0rUK6U3O{045Ez zo3>QS4GW|W<@A>yFP|VMyKYsY1A-y_)5MmL3t~Ca+S|`*IU$eREd#;N^~^fE1IH;o z?4~u3iV^>CUP|HT0!8~Q3<2nYTRxXPGQD@l42*@ANP1&}Mf!_UV!G6FYsXFVPf)lA z9ujRU2LT=4O^wR8^n5v*BfQ-F3;luT&t|{r0q8M2F$AxhRU4V(#yXC{B(r4jK~Py7Q}XJK}#>vy04iX?ShTBHdG@Ya9$h1D?*$diiaOOS8<9VPy&34tRI zR4V!xIOQ0h9?kMV&iqqr))PxP$tg55%e^#h`V1#5l7#>6P{Ap=L<5r85ep+v46C{fKpwWDv&ohHJVC0Yn`3sK!FgTT+SA#Vy$%ma(2R&o zwgzPujsO(szqtKU-7*{%v-KqHGQ_&OV!?wmhDfa` z)=YGyl~1UmPv}VQ91bFNOJgA7gm<}o8Lluhh(1f-Vr+wNbN&h~Tqx2&62VD6Z*=5T zq_bbk2K$Dtk>G4zRo5UBbZMmRoNT*~uVj{9pY+sWsm|_A5Mlgqoz8XtsfD_x7Cr8k zt@zKas=&;<6`PPl#@-H^p3{9Z z3_jl`&9~e;>S2|VXvO>C|M4p&ZT2)&JI#mgyB1uE)^9i@U2i$p3fJ-^CiP7>T8qUJ z+wogar?+NK>ElVPDT~ms^22>mNbKwPKaVrqhxI*wU!KwWXQSch%_nby0)` zR>-WKLzX*#q|S+hU#^h;(fPz>>(CCo(U0iqe1aM!T>M2E7XhI10AJ|`_YkGGZMlT$ zu(AO|XYwQ2us(KAz(oMy!ZU(%W9aDHNM}O@0zK`Nh-`L0)U+3fa)-`wEsFL5Ix#MO zr6#S`K4M^!KvV_I4gjK}VUmVK(6InmNBYKZiL!48ff55B_3Cz_rFz<7>k%)gz5p+BNUWP<;7O~%IQSgCpBgaXt478H$L1HwWRhPjLJQ4t4K&*c+ zGRnIs!Q@Xo?c{U!-YX$G`+o!(LSE-dI!5s8#LpBl-NZZ?eQqP^uyv zBDc|m^Cad)d0;C5xvUT4DXBs`3NNdN>4Q91)guxxH-I~*yzf@KGP)v;95`6X6q=o z>5yjC%<*Ao#v*!U)~p;m&>$6I5~ER2@6fN>a|>2Ir-fc4j*q=hBfQUnu?$p#Drd2dP}F zSoOYoG19Q{{{z%tPB#G;3(Un=*JXM{eJ|^tSEj8%#FgMd&3wQV};rD#D zP@kl=u>IkB`ARWKYG_F8`FOFqu$EHX|7~NE7?MYYKU#K@UTgKr2cV+{hS(pYr>e z8_Jn~kidqwUZLg4Mo{bfSQcZMum8f&T;T6jUmY=oFneMlpSnqtkNsPe#ndINhcXg| z9bX_L)7uNHNMaLUtWY}7cQ%C69DU^m!o*;bLvGKtu~&htFwW8g_`6ye-!(Bw&i>7%IOZXw6Iwk+@F_TMao$i;LcrKO^sLD>sqDWKI_-B@;R{Xv? zU^SP>eo;X}#tw427SZ#BDlf`MsqU}m&W0&QI^4D0h;wgGi0xQTIcNV&`)cuJe(fsD z9NKwcGz{QB%~i_e#LT{j(5-bDJG8;-jk}JpKRKN zbX{}@W*R}J7iH%&2xzaHS5eT#sfU2aG+0rJAZPLl2(Gh2I+r~e!b_SXOO#i;b@a=T zjue%S2RdxGheH}66T=j+F1#Ab9U{bMox0J*6f#Wkb$Z%O5I}H{wOQ9NwV`FNxGP6f6XdI>fk4RE z14I0zXlf-DBp^qC1|3s{tIbW)b$3sQ6qn%!In=?-a%(w2Z7L;^=ApL$-V`8B1E9lu zXf4IE%$JktXM~2CQPA)*%{Zx^WPEYOrW;^_cSO=0Mv*F1OA;kW(~K-f(<7Ack%)|` zd&d<-pWAJP;7 z1}a%9og@HB9DYod%Lc#N;=^0x3HeFTgPt@1pe{#lS)`@EPJ7)-i8K`gcOv0 zI;VhsEbq=}iX+GrL6=OLaG)mF_fCpNix(c*hiKSSr}Ni`1Xz1_bS>G+9z~@rMvn}N zZ_Z_rmHFf1oT8!&;oFge0;%BlbX`E@VhmGfd4iLZHCM+q-CziWA&Cqv7sj^? zz|V!QW7)Y+PLE461u7PrZ7n-k`|cY&@oQ^?UZei-H<3tBUfN)SZ8cwv(oKR~fn|O( z=fj{aI@+TiL-Yc9AUTJ+iEu9oJIt$5tzb9Vp-oQ5(0>Mv4?=ON;+Jtyt5Rk$VXn;Wefg<2p>NruZ?X{`?t9r&*;W&c0rA^Nan? zAL%KHV@ufPoV`>>!L0gm)N41K!O7vj4lhU1L%&Ji1)mbfiW8}I(Wx+OfkJ$|emQCd z|FX01h3IsrO0@x!RLSA*7|*p%MyG*0wu0~g!}^HM0?og=OKzYX$#XeXS%j0SIi9pL!U z@%Qz-&bV-GAXwKyTqNC3!}Mz$);~%?(dJ9l5HaJ%x^#H!cB&lg_`g}KrP|U09y&8ZW!ZLr`Cdg5dN+3;a_}`pHTb>WO%S*FoM^b{g2ue9@YfKz zfoyN=iXB_Ka8)lYSySRe)D~wM;F7u)>MzcfH;h} z#SPaW3?CYXb8ka_LI}%%P~C^2EQWy>m5@CQku712LSU8{7&a0N`2dCt23n4(>_nZ)dD@3 z6~fbQA8jxa#AX?l@ZAO`#m8Xz7A70Q)e?MxAYtdpjDCOOR3Z)A{r(}X+AgXiCSe0V z1ZuB<9)l|tb7LRl%E(n%jbDTj=MxiqqZh03*|(9@qb)lwn~V2b->>wjoX#x+Y$2Dg~!*O#P3=qA8jPn z+p*1_B%nDa0oMKQqV2Dt_`mP{Z11>Pe3F?<5`Qcun+PRx4#$7cPhGoDct2r3JDvj0 z=7}NwfO>~dK$i4l&U+B{xocU$OY$MEr(8D%`0-siyRhgBxKnLSps|(r<{E$Da0at}iobz-Z$zrpxM}?Z ztZ*}JG%bJ7KQ}^OGdVnjuqK~Q(_;^D?+~*j$5xsA2+Iq~iXM3|iTM7u5Q%bEiMN^b z!;F%9)kDF-iG*>LIy%IUW6}*HrZ53BeQA=3R#xIkzmPMxs2ihjIn`BXBv6o~czG-x zFCY!eF=Z&y9d(5&s4YQ@JMXUBANMgg($M&QI=GSb!+IjWf^5hhFJ6A?1;BjR)_^m$-a*Cq}2Es8iU>0K2))sHO`j=dmdGpfm7g}caEQK$Vb z^0F^SPcM-K`7hSILJuRusxA96R(2+wUg6DftW=KOQ}!Cf&j?Cf$@cBi^X|)bAguNA zn{vj0R&1VDVsv>>y(#P=t-MCymlfwa^lN&%kL4w-mlb*86|(~J*1-8jC2#esNXp6? zPUAp}{OKqGNV56mqqaJ=C3;RmxlWK`wA5Pr8Z}3MJjJ(_GVjqHLM?o49jB_jwV91J z>U565CHVL@I+lYy=}rb=srGf6m;~AU*4jClY0(WEXW|c&4PYrAnMrFT8o;4ZqauF; znG{IPO;Tk(6HQ|) z!XFE~Z4W5{9htHeWqWe7L}@Ymt~QOs4%R?siF~ka9gen*%w4+NVM?cvUVEK0gL}N$ z-fHhayqR|$SYI3Okh{0lxv%cZ3`?_(qr&w|B)`ctvsqyK$~4*CD&9IpXI?9g_P6fw zKnX=eGYC7Zx07EVR~QT#Vka4Y=Ul>x!vHt3~CP;gbq&bs$gV zNRPCiDQ;^TS_{tma;bj+2{gT-8odApLLoh4i@&tFGThrMTI799JQPI8L&`(VI1Bhv z4J)dD%Xr@W5x|&wHLfeWs6vOfW2nOTuEmoTLscp?1!KEeS@Np;kh`ryci0BzgPJ;! zIX$5Fha>OTu$L@|fl}!Z{KlAF_}%otLYgp=*pN)sXa5#bK{18LHsyv4Y&6>f7QSJR zO!Iy3VfiP@BHOmUOp1;Lyh|D!ejBwi3p^<57}B(he&A@auGz@-AZZ(`^iB^JEsijP zP$$7<{4tz&p8r#_-vILB*m?vW@MvmCt(Sr+`UR|^0_TG?%D}NmG$z~s?|lMQ~UwbjRcS? zk!DcXn0es5vx^XEUl(V-3JD{tL#m7Hce_rz;G0L{7cu@XjI7v~Gw9BO(KGK!Q)fTh z8|7O!P)NT$f8?N;ZKUMnSVFcWSX<25hU9$Gd!ezYZe3y!U*hOrqQd5v=373YT}ls@ zSr$rc45e)l{u-PL8UFjoNqw*kj3+)E;V}M>y$AU3sR;mZ{eR;ffa#UrVAxp5#S!z7 z1G)c$Ayi5$mOx^L>=86CeFS84saJCR8$;+mB|GU|rJnI(gEp)5&(uUneeJjIe35v_ zyE5rQwvMxv49Zle-Kk7xJ{bp4pslWLIX#L9fV z)gSUO*RBF@K8SjGOy&RU$9dRlPd_N`JiYfij0DelPZh4WHxMcn{~{W|@_o7bRL#uP zx8<lrDk_>6S1ECNhNUeibG-!l38 za}Kg9EKe%{!p9p_$u_})N7m1P2ip`&WgCXjfd>8EJfDCYhS=Lh62@e`wbSpB6l|<4 zzI-o^EYhMQ71ha(bH3HluJSibY5vqoTV=x!yeXBSKfEbbAGI)Qr9c4z3+eQgj~li2 zQh&Cz*sTzc)v3=I6CXocw*;(9Se}2+OAdkv>j9r&IvmaDRCNmyd==#!EtAKfrN*i# z#dY=B+wx1brDH-=b?RN%LK`EkCRXY050W9t^SMY#sh?wk8okI1D(*e^$Ihf3%%95} z1~+<~mpTyxHr%S{=g!}yA9z;T>NJ&@JS7G+#7Mgt_@7qBq~6!-HN0XLu(BGss`XG* z->vluCJ&@9r$z5bcqg&gCXCPXWD9993^fvoXA%(bK`OhCn|Ej*U@iz@kk5SHuy?QW?Y=* zHgO@Ub_|kAP5Gtao|yKIZKzm$(tf?Mk{g}3^1g1iWxCyyFaqHPTP9TTo9}Yg92%Uq ztDojhtCF?xjF6!LT+)p3sF}`MO@x!g^$zX8lo(h*o2~!02FMRW0dqKmK&XfuDzX~H4bjCtR*RreVnteLF-Q*XI$t)1{ zB+O`?%{1dnd)ryiqBlfw(rrjIG$}|CHnMc&N87Z3jgro{m^ALOv0rj`Dk+G0WqnXR z4ZA?m5>NaLp)rbM5*OEajv4CFZ{H|3!!Y_9sODl6TnYgnH0v?cxFZS6XQcJ?-`;e5 zdX?O3i8~t&2w;)b-d3?m-m@B$WMhcxC@n!ol!}Jkqa=j_fP`vviAu=y$(0c#$egx( zjYv!BO5gIiY9g4h@MWcJA(Hgwu`$Ki>1h`%7-6yu@@96e!XG0rC=^cRqb|UZw#89~ zYoYvZ%+Tc0oh?pJRxk#&mP4M28gtt*o@04hk`av_aaic=aE_)nGmuim4~Up`(5sUU z!e(YgUjm{3P@`Fr92GLrO7?BXCyTk#=fFgso50ZwI~23xLdcI1Y>vmuwC0_|zRvy5 zCB#S-TJM%;0KbMU=Jwt=kkP{AVbz*O%Qw~u4o>4X_ixA{p9=-qmESqptptChv-!T{ zKPQvN!#E#bQE8{e1GH?cXozPmV}YxM5Cb!iT+7^j>~$-1-7JE?h+xuy2 z;2-@~&B>d{9<-@3^wU~hlOy6l^Cb0t(bFIE`ICUv&6?Hdq}+0LQ@tteY!k@{E;VXT zbQjOv%ey5sdyo-mi#O1*S=VXFsIRQ3U#P|?t$6}g*fd{mOekjLv?CafI}_yc%N)jd z!nxb+E@)PJZWYJPNb=DMAIJzRRO#2Nv#HlK`ze>Ki9BGjR& z2P;b9afHG?#t&mNxrI|uT9z^h4Of6AqXg>@aI;5Bi5^$~;w z_n*0$CL{um*~#ZO^aaOSlKjNJ6b#LbNsf7@ICI$0iUK=y{*H@h%zR@M?Vi-Aq^An1 zkY`AMGY$M2Px{AvKE6R?!)=>Au6k@qbkmT^rl>bHbIkd786i%3PLfn5HZFWgI5;JM~`}c;wCbymKr=u{A5Vs8dEUD=dd)0cw%GDHAOGu0~-b zpHQJ|MfAVl6A{y1(B9Y!vRKOGT8K5Z2-D7LqGD@axaO1LavMnszcu zi{J1%)uFPv=s4-RKz3K+VR!^vovgKnv_+*;lD)|y2BHX_gudh^%NiS_pMNKL_RcM{ zO(ajiyc;&mPglq9UHs82snPqRLjgTj^W&WGCrHKezqWjuH_VT-v|esZy1QRH1-jP+ z8@?ga?TI{XSVzGuLN| z82Gx2?V-Lt(0a8ZTVjYs_LI3yRy%iB0h7zE2h6@Hg3x+9V z6F^wPNbju*LJ4_RzxJ>KK@Yb1LlO!5z|(I%(vljcRpA= z9G@cbb=M<~*Zda+L)Mofa9Sen-{aq1p%!_d6%9pIfg^eDBZPSf*h``sHGpW60MvzO z84nN|ZS;d3mX8Br=IuYpzPRcb1q*-z!UOH6MHK({5bJ8st}t+;1t1m*Lz6OiZtz|jEO zS%3;0fH?cS<{NQV7(sK;Sb{%Tc@6+P4^;bwkT{`4v;|;2JYay-6?s1f?F#5}k9BJS z=!60$suKuf!oDaH_>v}d+yhYe0qn!E=tIC=p_rdgf^6EPxO*T13dG}%mmfm?o8z8& zk|4_zBsa`n#GRby@TOBJxsuz$fibMKnkf4ktD-o@9uRZ49~I|-R73-&+Dsy6eedIm zHK-Bix`A1Fg=$|MQ$*wHhk(PL4G1fT05Tc?IZ1%5BxIzfKA5@u=a?_QbrbuFdnJjidt%(u$T0VwjlV~L^Ghj4jzAGbZf@myLFAbim=qVm( zn>FV5!3dIkV4DeL&hk~t&WCWwk>rF5=ZrB&1e4L~Ky%_)g*aINi4Qp_NV&+gx$*v_ zR4jn(++38BT;$>0%!l0KTq60@oM5FqZ_oFNVHsiNdA^I`#SeLQIpHRjdCiz%vGS&< z!+9Mwnf~kVd&wxpVw?xN$voEb(*+Bvq)b&OT|Y(_L_rE%kHhEuNz<7MmpAkHqn+FR z3+a%84AC8TVhd{#@`l@F9|?vFp6ovMbvx}Femql(5bN~(X-Il;n*Y1|qm6ITO00Wp zO_72o$1lf>m(vIlq2d+CLbS(%D8oRkykgwPVla8hC36X(Q3IX`xAg@Tz@TeeegMQdt`QJ|LpsMwwu3Vrp5a*x1y0k1Lv>Ki5M4l z2rjjc@P}U~^L|+;e~ZtfyAJXCTvG-{eA&v*nRfzS{FX}J`xrZX#k9zGaqL(4*B!B8 zRPXf!8M>t_p|XQzWS63>q4#rc$Cqd-*6NiIltqGK`9=&`bqS_C-&JTPkwR|5@hs_~ z&AGTvGE@{dqFDAeOkm#~X-keyYOp-(o0&@7DuX^~QK8&j2Be_jhel_!?%GPGM3&XL znz^22Awop1U1S)j-%3yL~2!y)6= zgaeS{Qex-+Ny(2mdCMi*Q*}e1^S0M+9Xe6V_|bkHWM(Ep|(VtMfA=Y9u$yi zjpO1E(fsD94Q2THwGuN(xhw&2yqaldGR&UUH!uF?q(|`MNuks*@^U?V0>A@mr7|U% zgwK|Ofa2?sw`o0s0dekC!bV2(aMI*>MN!qwRjM9-6n%J z*=46dfqKPSe91st27bj-1Y#`F;p+mJTqjL*=xaa))@#0_E%O~&u*=XlJI-O6l zUs=@Q7;vrlGhsrceAUJ1E(m3jc?Zp^^64N-f2I>K|1ob=qL(p^2cP>=?bk~xUClRy zD!z)d?3l#<4cZ%qxE(7m0S*R0DHXzkhTCR5PU*;vMk-&~^{pOWIg%weg67MY5ACD^ z-HtYoGD)D!vXo$KJknmI5s8ZfxbChJ-HP^Jzg25@fVeWR$jXc&;>i+kS)82|j}m(c za4nvgF!t(M84Z}afE%dDO?Hg6V~J)D=UP$>DkP(Ndg8BtSOi<&Q@pD?`_%ahFrGMy zxhWNJA|o&s?c$!XFu(MN5RHx&|4#}~ln%~kH3^}WQF(Pij?S6q;~qiLvgU6`Bq9c5 zgQC&W*V^4Dds$vUxoh@DZ-o@TFD|lk&XlmR^Fx+V>5X0%fWmYR5<0Y2lu)*q<xhEVM~Lfg@ub^6UhHb|%>%nU?^x)_s8qZz-I38F(nD#i{~OWP|G zW9CB{)1l!4zpRHmfe2WrWT1943UL-K8Z1FspiZoFL6ZX zyotF;;~`~u`b3qWBA-<+!`iwz;le~e!D_fn+o?3;k&y=qzQx3#k;VFWjY9cYa4XCH z)|or@YFr$DOjZKNCjRui^f7HSIqs5M_trxBDKt!?B4ySO`RG6%)d2jkkl=T$I;ZG% zh)4g?=^scRoxyefZ9POHn|Dt8>3W_GpC{t{X4xKA)msglYB*M{c1#+R{YcB9GC`pY z{!QLr;Ut?ISI#2s#R|X_!2+xNO$cgin-<-a!53p^9oaajJ|8+M%V|jcD~&?i>!4ew zBq>WsPMqg#)mHGG{z%ej%*aD}(kwmT3~4ufLJ8s7Tw5d;p{JjkBCs2)n$|MyMi&KYym~{VM<9rZS#rf|PAivOE2V|kO%w9Nw^C)VPcf2}NSi?I z{aNQo1A78iob`vExOeFdEfVXdD}OD#zg&8C)r>o3I;TCZKT0bOZXUTv<2!aJeU28q zNx@&X`J@+O!k(;Np-WvdO)C{9pMZaQF3%21>&xD7#@Q+O7o&7j+!F1&JPV|X0Y^F_o~^{Igdc~agz%A zWcekj>plG=S($1)IYLjrfwQy5|ERR^-BBhY)8(Z#zHM}`&`ihly!&TFt!}b_|K>Ex zL(IMhhhB@*=wNmIYe$bhbH(efE@XqGq8c$rXL8^IGSc~ii)BXuN0q26^83tfLa|OX zc-eMP;_i6?4=)$kwlm&NSylWl~WOCiW|x3?yB4G z0)Cs!5+Eb(;y`?l%uJea*56Tb$H=~;wCSIDa_oM?u_JhC9oU@X2goODiZla1RZ+s`s=M(W__cEpsPwLoF7LbjPr=<^pxnR$x9nZp zMqNXyATv^tTR7wPnn=rCynD0QKu2IpQsMc|9+@M_m4m_?U3uDv6q3&uoSHjFEV&1| z-`7c^Ljt2Ui++v>DJMKT$8c&i`P;JfYTq6QtxCNHEFl6If~p#}ku)+)f0A>+qu!>Z z(lNIfzY@B95xt&7{yAUNsstoMLAVj7fQMCn3?OgzJ8?BZvm$Nl#ZXrZ*KJe2|J4N` z7xkZ@@)j11L(Gp-HCa*tRm_3$Jhbi|^mS_{rzC;f6f9{!-v@bqx7_C~NhI_%XER<^BXN_8aO2Bql+zQsMMQc8$Zri^w8R~R4Ifiz4uB6KS) z)LM}F)IL7&d%RqXL-7PQBCt7i};l@Ylm zWOV@743Gl%8jvbUYOZS1Q&*%uYbT&N;7tfox}F3ukVSiABt0@G60P%~StVw@v1@Ii zMP3YRA5YXjNz6l7$2f+^%+e>t8zlJ*TMi2;r?w`cDt@4;3AZ9id2Wt%gp=PxeVLb` zny1O-EZ|CbvfeFq8a;U*-P^i*+Ix5km_9xqJ+$LKrOL*$TQL>B82_Eo?$e3GiEoAv zX@Uj1?{u}(Y;~5iVYHoJG%$)pjfGY1Pi6}pQyxJKwtwW8>Zqt|QUR!QO_=jxjsqpK zsSj(mAb+|qCh?EzSFwC=DB44n9U1q&C9q6F9U=Zt_j#C;ycI%3{`N#RXjTB^NOjD7 z^w?Y)mU#0~|LEU2V6uP`e|C#8S`L`$!4LDspPM zjc00xe|(}~<)dNrA~c&GDbC19HRPAOe;kpAQN^=f$X%3s@*>D}D@o9x8ZJO;aaM$h zxGPVJgZVnKCn;Bi^c1H$x#hJb-<`x==o9_8t=T2YZeH`%->Q;NEw2cG72Hsc3kOvy zgT)?e9ja>Frf8e(T$7CQr9iG>BJP%XLdu@1bJl)YWa2+o^xa4*YkI0%jePvhe4k^W z4bWor>-yP%2A-0d^oQ(?t%f~iFfwxk%W2scqejP;da4rg?Z+eqr^X+7^~O$_rfiM! z_l>uYwUKdYcRfw2A|Ia~o6?x;U*eho>~#u?P5p+=Aa?%;qXMkI&BP>)oZfz5;}!^O z5TSDm`E=y3vx?u3yc86zRg29#Y^_Xn#3+HSR#VOCm} zMh4AvA*^QeJmQF{Kt`!X@!sMOa&2Ohs8NMV0Eb2<&vo?{wVUYBGz8`$>o^#`YrAQCszXRX?BYb`A{eOh7fCMWfj^4qzc$67WY0n{0-w<-W z-C1QNc0KW1DBh+X{_ghyh>NGrfA?~SFv(mRH+}!l3NBb|M7RfE#}XB1XoP)cc+Nz@w0 z&uT`5CLcidL!Zr_x-1fg^heJgNEwVt@KC zo$eJ#@0+aXtAe}uTpi=Ia^l1LI(mm3%1H9boOwV?uqFka-cPtAVPw0Mw&F{f_4G)^vpVX%yE2 zs7~dH?174FF)DNcFfJbD4#LtO`YdNM-SuHveIKByy76y8D8BxRHRorJgh?C?AX>^N z7BM6R@#XY)o4_m3PY*1UdSD^Wa=Q3vlLsE92UgcUFl=U1(RB^^3Rvul1Olv)G>@>> z14Azt*rhfV762tdcz8o2tRH{``?#{t1k9G(95aAASJb@O5r?UjriRZIJ6laxz(t!` zV6^KZWDPZ^s>nHaHIyazhcZKm9l;r`C*H~vWL6Thn{bt`oFPZkO-@uauHl=(K z>yPDr+NOCX&^g%I=Gl|k#n+H>b0mkWToPjkuW#BPOkxtYa~6l5tuO}>{Z9xcH2?xw z1|0s6!IM+n$F>K-ka%e`#xVR6Hool=k(oDwRWA^Y){Aedb9QP z$9KJ*m)mgf^S#;LuAg6q!m(b>J-A98j-xG)V z_QoS#sqa+`4jhHhN#yUb?B7a%!7}(sD*p5r@96Lgg7--i@pNrt9ttFOu&_m;uHJ`X zdjT!NIT$exA_(dwl(0y~v0JTJCA>lF zn3NJE&#+DuyM-DUOX3oKOL!|_^CeXdBx{qV&{4LZh9izapRS|l$mq~%Qgf7L;(L0O zZK0fPn`4vZc%17{SyO3d-FSMO@9}oUuE1x_@p~ckxaPZ>QVhev$510o+v12y^w9ip z;#!A<=*C=@oMc}z$K1r)&9AZXcD1LK1-@seRmD+kXVv9dPPFVS9l})wD4Sk|fI2G# zdZ~m@(3_f+-_Ha#e$V$>$r6KTSQpJB#OKN4pWarVx6TNh*SEdwfGC^$SF|rX*4CEQ zJ2%|oT{}AO?=Ilq1lT`zj+Lr#G;cRLUr2u0?Wh+&5>RPsTs(_o^&ZfwbCp=VMY$S8 z!sGB5KA^r+>Br_{H$vwG|sQ;HZUF6#WjkIrRp(3ghI&+|LRnK-3=iL zrLsV+!hog4YIDe&dLncI=dcAM6~a@LqVS~};M z-?0j67$%H&!Lrcd<0$+<*Za4%$6=adt(B5*hG$&=Zy@)u&%CRjgfv!n`-ZjW=xq-O8>@{QAv>PkgJ03CH>BMDh!c&>rjS zUctrr#<`UCPGKtOvAT5u&-Xdv_Mh%v&hI=AP2DLTF5&sx1s`4vaDV^GQ4o87jcZRb z9G~cI>T!y~T-23T=l7-OO!zMU$L5sEJF%Pd%OClJ16e=*?jqw!ym|aMZS?jbyUBS$ z{N|iZ{7)qc^(~l5cr|F=aL+mKzI06TJyA+Gw1G&Op2au6myDol)o_F9T`i->5PRPHdwQ$#H-j?g0< zfYjcSi0p4Iw@;c1kc~(JwvON;uRkbFPMmtW??1Tp#04jcU`@tSRasu%=WTD>f&Gb4c ztCf6=&Ka$tjiR&yW(1Snw7FLs^=|0u>#%0&6v$$%Fz46VB$zjikdh)Sq4SgFnMdTZ zbxe>R=?sj;+X*;H7h`q?_7(W8pM#vBi{F|WF}h6qwx5|ZyjDG2mQNoXp-ASfE4N{Z zA!D~mCr@lI9lOP8Jn;2c2A?;#aqo8-$-}Yy*rP5C-;<^ln=!=;J5)DBBd7 zh$Rq=(_c|?$6&fE8@NgmbyWI|5Fna@l(5!A4B2XqbEM(jI!8b_awmjY@sy^vDF+xU zG;UmP7+?%xeP5~ZxQCFkdUxnm_o)X{>ZE|1V%ZJLxm;|D(t7b27R_rNdpH+zXwK_7 zL~GT8P88ou8BKHI8tByf7f^ApLXX(|x>Zh)q14DmwMKsiSEp)&1B~#zLhpteB3%sx zw!q{4>dxQAU!l);^TxyfASqaIdE?+84;2ViWV&R-%q8Y(Md=HT`4xtQlt8Rq zUZvdGFa)Lx%~JM$fB0*UXP3^znmn1V3p(OOu>ceh9YSEBb4v!14MoI8#4LaAk>>w^kBla3S{pOmO@ZXRdQV z)MVi1wW+EtmwmUnN_8#sl3*q4@b%S-SqqfaOMm7l+T_Vwul(m9XZG3)VdRtc0x}CE zChp=5awh5O*DLP;>%Q&HHoUa7@_1}r#28ce3ZhRq*lh8zVcg-?ETzAiuVp&BnEtr( zjAN_P<_douj^YbSY!3A+H{3)o{>|ddX0>%h=8A$i<&%PI;jN>|EZnB>$CZ6>HOA8Q zn2o=ZT&f#kZS1fl>AvwwvKXEIsMI^{fNiN9Oo8n;0 zD$*}XM+O{v!BP_IzrX4~F3_Tq7mmMSoWly`A~L&9Iua=c5DYQ6&Z1dEBz~VQHIMgS zNF}j^6w>?cvC1y(HGh^~{3=ADFE{clUYpValcYzXCG#ABZll@g8mpE@GH_MT`>ha@ z43!Q(5I`dvPHfZF@vimhb)*pG*FX5biyupk;}k}P7K6OM<9S4&5gv4Zy$}%ili(Fr zEyiu>IKFDyukY&8UTxej{E@!?)z~YyxRDGognT`$#7_w6bcz#(vSSIUD(YFO&JwX0 z;alSj%$$A~rTC0Wv(|KbQ)lc;Ieri{bgOurr8WdKHAPcLUUX>#x=t~3yhzdcNU3kw zL%zuBU(!JmNZf9JaeNcdr`Oj|VT2g^(CJ;PVpzJDk-X6&fu!>gFY*Mo+1fjW1)$Mt zb$1wx33(1wXgM9(x*vq%F48%yhY~_une(u%%jG(H zE0NSK8l-f(M5($4jPH?sq6RGR%$F?!m;pw^+InzXq>XoLTE1Jl(SmTWy+WuCJp^L{ zY|-V{K@bL&10R$ScW4nMMTOY-3MFYV!)#U}WswLL5j-n)yrfEzxm%3 zU_bx>@B{Gk{~C#9z_6i{1rYzcNIaelMlcdiE_@jdL{S1svTBr1XCUjq{#_)dOdZ{M zXl0%+dTaajpCYkl^6d+wX>GgXQpi8Y%z*aV)jCZ2ec=9iQBxJ-NsdVL&$>>vU73~X zbSI~DI`Us}RSR{1b_?{=$6b}xhA)HifsBM48~d-r0B+Z&GX|Xd<0*`5_y+Ul;xI3M zG%!Y|_u=PMnRq*cCiB6OD%No#8X>-sX%u%%#onAwoo!9Ls;e)n<~KHlA4AXc?i9Fa!^ashufq*I`6nb8zRTBpnz=6MfP#Xmb{Te&GP04 zMV19eF4`nJHWbS5hiFMpev$MFV-AJ2i{B?_TX3=3Nlh9=$VO7NR+Z$D^`Mcc$iOlg zVi!LP*=e%8FzLs0ysYiUV=f9YJNOZwL6V*D6)Xa!J0nQ#<#EYNN;1@Jj7)T$s_a8` z1Y&8VOpez`bJF#kYNbVR7;KW`xgPVS*n$u&tc6u;dEmJ0$4$DbpZu&jg(#RJ>^_~u z(;B5e7~@hj?v$J^96tJ%XPRnMo-j8xB4smH<48k^Yrec?{~@22=i68+-vLACT_@DYGgsHZ&8Vwj z0o{AztR}m^Y^nuBeNAa3?1YDt&!Z>mQ^@xtUeOQ43`OtOyG{6L^YLTNzWyH=IRi$>`k3>LRq>FPV5NY9tt8nSdwDFwxD|tMV>8 zw)jmOQdeB^FK-uon0@Saq1KU-dmji8x|&}?mx1eh!Xc#J52BcZ?{~uTPuNHCbid8( zXRIH_AF=k#*> zfzSV>E}QgH`WLN-;b}EBOjhV1hGe7bZyLUjM7di1bK3jiQPFl1#J z0JL#sk*a_(nT207>-4KQ+#-z+Nr0_XYzqJim>P&i7s>+)D6er&Ys2p3?dZQW#q}Y1 zqbZ)kGXjp{CX*zML_w7vSZ-l8oNBYcM))n$-ln3EQ3EaUWiUoG!3%t^4>pl4mepa-`z2u!tO#yH3lmu*2O#`p20{=n(E zw=D4vpUc9%WC5b8W8`dK-6Hkg*f0kkCfn>oz><R6MXzU$rVI14?OUc)6n{H9CuU3wF#=LWctT2qm#Q_L zcKXx}ld;|ro+AJmT(Lb$pB{#grJ?yMJ{)JoUU#Bw@Jk`eBnPIm;`~O=VHII&yxP(;H}KPYep+^mMu%HNwd^k z_9i72`<;eYpaK-NoSRIY3tXLd;R|Iik@l3ojmQ=SR56Lu ztADk>K{wqB06WN(}-g3$w!1p^VYgT7^c@f;9jCX9yRQO&>CkyY#@nPKFpm{m342tlw0= zJOr-~Aj92*kMehUSS@G~mre2oEc-HP4GaS+%`lP~3p5lzx*k@=2rQ0JRt%sT&0_3N zmKbq8+hl)O*;OnWaQ!j|-q06tdc+~!iV_cJ>e%4cWR4lD}+ay?6%*6waxNqrH+(_&$!{}*S}z>>|| z)e;m)UB*M_acsLVVixGinS$qJ9oe2U-P{wKkgIB!_?B$G-%%mO*Uu`K*jv zcifKTJNeE{HyQ)|YI{m*Av;-J)S+4EA`IehgkCY zTG^eYon_=MHov1bKKwA_@jG`_zcniBjbkqry}YDjADz()eA9q}_0!87wQW1eyRygi zCv8{a$^qK%>hoRq()PpFoFlxRctfZ6Y(4v-xdORl{C3z7%C9yJyP;fN*T|kbUYK5Z zRfAuh5~3@4ClJ3ir}PJQj+hdcn9P?Vtkr(!Z9g3g&3DiF&HeuPSInd7rF$XzpZ}ZB zCyMGlOWAWbm7>q58p%B?Rce1~ji1j9=6lw<=KeGWKA&6Q^=^!--L~dGUo6OrYT9vu zJGFrf>(q}bD{6Ox@NG5)w;q)UU(ynI{}0(?{avb7zcK8+JS*p!1A?Wy)|8m>ct0I~ zUwr?;ugpKpUO(BhnEMw8u}Ty0sej`i^q))qXdY23cB&-JKI2FA=isdGh?x<(`APP= zGn%O6M=&&3=Yl4rs#o1h0)Lrc@IF(P_h<6_)3Im~XJg2~O_lm{bhRJnHYX^c;U4V0M_h(Tx*4aHz_7`jF% zAJj4vKV+{x9DN_44<@F70(4URunmYr+kL}O!qGJVxZD(GEa7k9P_!YKk^@i*IT9j- zH0}q&RO<_vp|(4RqejRy3Nw~ zM<@>&J(q$M01o+T6L!ZTK%j!~fs* z7OfNj7gdr_h{x)44AyI_1Pq)1nurFp`(5n~g&=F>$xiYfNoCMkgJ@zG7-Vu#Js&CH z+WT`wVyHJ6O6}DXS&F7ire|IU8Zh<5s0)nt+J!p3HnYsBj`Qgdi=Sc7%mNo1t%!@% zcfD(N!vV*N&lh*=x?X=741Qi3RS{w$3$rF*H=7Z%4hWK$gT^9dl2S!&33PaCz5Fy@PdGifKKLP60Kfd^_@tnLw|jFWAQXqm;U| zub|-L7b>b0VLv?&_vJ2x|CjW!fFa2sLI~OEj9E&__9YvKXry5x!YT zmCD5^&_?ndr!(Op_fLTlNIO(`8|d91CQBbjr}*w(k8L?!2Eji{zciDQWUzr!u(?A4 z^vvLk2vR$#?+?ob0auBvMM*cVtkseZ51(>M9xsp6rS#R{nmVZym^1kv$WrP$8N;$i z;RSk|juC{oY$p=rFkQA9in%CL>7B^VG6&=hgkuWKv6C!Y{8EGMw9yA7TbhSlNOE3R zx2ko~Qbs0flwA6_QSm;x|K7?1PkzHfdW334t-RM43j_axMM(pB0?E$J;@D~P!e?I3 z9-On6*uCReq2yei=XyqNU$BcDE(vb|?dSQQbk`m_W?4S$*v~XN(;CIn*5-^RG6Gks z5|=vm(=h&&I}i3r#})&0cnyxaPendhM%4sNo;9P@F}LNS@YLmgEPBvUr%M})l@&*P z!2TiWv54_|#s;6l`x9H+_7AHl+;%yedBbf7X(bzD8uiI5EJZr$P5Q+0NdOX!M|fuZ zkF8>@y2j*Ik>{KDoBb!0BT6drzoi`W5TrN-9-RbI%DcFQ8pAW!!au!5xW;{!n@b|U z?0oV+t8oY7IK1ZdWLln+P`alR;2xgw_f2Co-V+$Z8NlgQt-~K}t>F|T7L{VyBjxL_ z+UCfxsZt=XGDDDV&2795kD`;}XS=jds=vPP+3{u<&`2GMCCm1%Lq|1R1EvXz@Jccu zg-YD@ZSvE-17+O8T(s*%e3n)8#jSsC;r#xcm_F7bIdW<|7x6XO;VNDY5vsk3%s^I2 zl%NhhfA!Pn2L9-eYIPU*zqoqKrZ}K&?G~r;rg2T)joG;GXmBD3E=P5d3B2BU71`ll3e5B3G)?H zCjJv_$wI`}43olh3FbyrUeJ_{&vfK4UaLKk&9cY{yz=nW)ujr>ms~K%@_w_^i>X*^ zJIbuHZ=KZ%TUnE2J-!m7S<#*@p6fgz*+6;MLK21fE6;? zD6wE~#AC9!ujgRtyXjMJXlb9B-Gw}T5fB#0_m^WnEEC3+C6)$Qa%ORaOO6Zn6)MFi zf>z(u#D~tJEQ)k*7B%-ycrxN#;X<&ii&tgCT(Y*Mp4FKRi}ZVtyX{^=Qjz zTUGOlN$?N%((D}zr?|CnKl>Pa3&;MgUOn2srXDqjVkc^pa;xb@%r*7TR~rSrtNOQf z1~y&Yp`Z0<=9Dw{@tFR>FUBeJTkiSH6y0HsO@JAL0;Be zdm*pv^G$Gq=FxPGqxs;8ZVU;XaBe#AJ`Ncdx_EEng`74mn9zQvV&5zuj8T0VEgS^F zdJG(8tYQQVUZ*ALKh}~!2{Hx{OG4KFaG$98UPHv_L(~9cc#VW9_8zP-Mau@@h92K} z$IoPHfZ_8B8>uDz(uXRN!V6i^K9~~B`{a&+6Q2=5(>TG(y6wqxBP@Rj*1g;|ZzK_G zzW%Lw<=jjKDWNj`{++WXujRP+e)W90xPVsPq2=+L$DjUjFQd<))TJ{9t;h%YQ_mkf zSY%Kzj!VPj^10}UuYU@KTrH4VIX#q!S^Fo5~5)>wxFRjUR5 z77*e- zKp&>AWoa*knW`o@#m*1{11N(b)uI5ddvj*ND&J$VI%bjHw_TkT=ZdOcB;lF37Lr z$)6fBTvpmS6gO?@-=m9`q z0%L5$pff?~8LYzbU^&CL=qPw!0^VJs(z~M$x`UJ~(DJh-%V7zeL3C*;w9thZNgMuV z1OkN~_9A+S+mxGR!nC7`3C`~eEQow~1^DzXWjEgs!FjkEc2F}OdGDeQuHH1Ao z3Uw+fF`U~eMkq%Ws}Bw#1c=h3#t(C3yNXMJ1_+o&4s){N%;t|dbBpn!gPCdxGjxx8 zN{Ja>VA5Q~$*`fZ;E$OrjrFb$SMuSLTP81=0v;{!43{8&45Q|Wi4VJoo6-ZBm#CUI z!dvK5lF9?rY^b%fgeC_Z{z7kZ+OH=;w&uqYA9 zv!E1d+hUl%#4wLj(CI1=v*EzwrO``M(Y26x8-8*v_1H)NVYCl0dkK?rGmHg5w12wM%Wvc{bQH6Ra#LNqls^Xg3K^>1< zkt%~5!_u1&7#bPVZB&Tui$D#dCm{1e7)B}`Y=+`)2e6Tu7Q#Hoeyap3Nak{slX4#A z+Sn#Fge6hg#9(0Kb@c$JO4+y9^Ku#T(l^PArt@kcemB^8&5OcEFS!`>RAs$np|-Jz zW;D6s1QG`X-)zA%sRi?r1z&ABo=lhn=n9!YK9447YqrcDrG@Lhg{&i7?ex}6_y~X|%!B4d$Ll;FP5cOj% z!XL6Fu}DRzeXQjwpVn4^C@~~Bc_fayrSKf6IBz8wZ&WFZrK$+U;5?g4IIrYGV5f`e zX_uWg1SA9jvDcAsY;l;`m{T$0{ivbaC-AwIR9?+xo*ZRQ^({xcEscak-e*_%ag<2f zLHH$ztFS9CD-BTYiLy5d{HDot{e zP?w1u^3=Xuh8OnGY4ugP*Ksz5RP=6$ueVh{R?yk%)d-hWhxM^K5m!e*Wn&ccYUb`L zZa5Mxy7(iJYgLb{!*ruU<7yuzYHWz>HZE$FFEbM5^WEg@ETj~wC$mE>3hUkk1835D zH)>n!I7{TykNNAmx47f|>YPz2lDF!gC~B28YUSK($LeZ2`f7n~4VV%&1+i6&=FAH- z4a39*wSJ9ZYSr1+Sy^#4J9>oemz6x{lt){QSsac1y2NXFP5OsbTX`JcP@9cR8}3q@ z_w%ZsE~!ogo42)_-?#8YZqX3^Tc|W!=5q23Q%m6gV#EDy0c|@-$BDq1v=aJ@5HkVs zg<6+JO5tGI!1iTS`K9otW#sv7VZ?rn{uJ2TZBg8A>_Y9_`t7_2{_Xtv?SlR7!rSei z|F(;fbchReNa}Y;`*+CZcgXj5C~kKs|LssA=~NTy)X?wL^6%7P!eh(tyqDYGux=2Ukm&Ta@+0>5>gCC>k0i! z=_}OZuq_{TWooh29h2Xa^tUG^9{7RNo3V}Swpb*g--8{>%B$BK#1v9Q(pM(bSE1im z<=>~m)Kt^o*T7VhxZUUA-`#T6!JH3l7V7WG?-%gz@7wMl5^Cv;2M+x0AJZS0@E?fP z>Q*B8zXi10|0Baiga@d^v{0l(p~0eO09wn&!>|yo#`s;1CgQ;o#1-^+l+$Sh|0}}< z{o_li`tKdkWkardh++lixXCC2P?y+MSB^VVM(ANB+P2q?wp0-sj1 zQ4akN9-VpP|0$sDH3)F6PUDxZ<$qr=K3uN)Qg0MUd~P#?fl_T%_@e!5yKXj&m*2{^ z(eM585ssDA=j*{p>LX7a+3yP1pm1Ah#oB)R=H5^Ib);mL$is4tY+4%Nqd*b*B>q*w98q8uv<1!w_u zncm4(NZm_kvx97-sMl>5K%&;Jefgs~ zQ66@hj#UOwoptMFWxh{9KSiC2paUg^b^Kl2M4Qy&stD#=C+K zgQ-yzzS_vY3kmBxF-ha0&@0V%lWyR?hXgCf(#kUE{0HjYz79A>D~ z!UNd9e$2>Qx=ZV(x~r%?A2%9-xQUD4U=|MRk+pCylC$+t2|GI)sW<(__UN2?o^t6_ zQitHu(b=b3(8lN|;*NjzY~UlTT1T_yT1wb5o;ACA{IIRg95R`9>~39qWXmz+HLw>& zVSH)e9@cPh9S?!wOezHA^3W8Vu^P-fkG#_^1gmd}17Wn>&fJRYBBNCTxu0wBYgEY^ z^2&%NyotI*kz+T-)BzE68jJt@T>FQugfT_<;q7b}6796IAqbZR`HM_z>y_)To^`fG zdS_9lqaWXA{zPLs+c-r$*|^{C;KRcLSeg@ALPT7bpcj@INk~CWFhjW9j+-gB3T&&d zniV(ohzA`2LUXl%%B5CruI zA2Spt^gqAtW!0Pos`BDnBq=%A2}`!Bb{U^sLpI0c)X-ZuC>p+DhAwRBD#-_VUD zQnWtwx}NsPT^mB9u{(^?UA>F`nH0g3yh)7}Dez>xWFq4!>-LkWb)2xXPt*C4 z#uvQ0@uhSW&Ul-^=YNy5=%~cSLP@QOrru~3jz?h%6%So#ccH0aM?|KKme?s|Ju6?x zm}5_+K<#0s8*)XNrp$+j6Z%(*;>OH|O(-$O5B2flL0VjJLy}8*{n6d9?~?crZ)< zq1S*l%fza^=7|I;_Wn`3edWBB1di;Y&tsMsz{HRUJ`6Lvnp>;cJ&dud*Uiy+c3bb*O8Ao+I!cVmxa*;2RmjDSJ(;+PXBB@6 zZdRj5=6wcQXI35|xn)#Pf>n7ELJksYgzp6W8qHNiN_*o8PuW351TMW8jOvBof_Mq8 zx*Uo2uf;h2RG6*4mrB7qOpVd|dA|2jD2Bls4t9MA!0-@IIkZ_DlJlLM9P;{uQSiYZ zPRXVOfp|uf{4S$|;+%fLMS*jroOV zGd*`8LtS5qPCN!>KHBYzkp2>-NIe~Au9jqjS>l#b6`vq5?yf{1PnLYPk8UW-soG@D z9{wwFy*N^n)%Ef@?vuGv4>xa%t%{xZf19?@vhcdjuhW9Y}`)kI^k)F{E5Q)~{xRc_)nr1$qYwQyg8cx zzv)zZq_{+-*){){giI!4|d{0$9 zd4UMR3cXrYTh_#7-5#8gP}=S(fZ&zfWH4lGn3DDz9~vR;XQA)-A7aizj~ir!bBpmo zL6#yK(o>t1Pc11;xrzIei!3A-qEmx4Rq63`+RWJT)=_#3 zvWu}+Yq+5f7aJS#S~W@??l8YJM>*>q11ETrCGK?T@%~o)NULjW9az*eCYGNaYP0MU zJS8T(0uU=Gq2XXml>g@7l!*&eUbAfd7ngo{?S1)eGWhp4++QzgPv^ zC}F&|kd0m8^6b>@M)@ak3ia8N zX2cE2?N*ZD&i}VU4K+hvW__whq?ul=`p(mEb#SaZOG5p;*eRkfKx*_XcpPE}EG*VJ zlhQ}<@MtgA(agT)ZU+WEmA%Ztt~%C&j$N)yzuUDL5;{usT|Pks)VCG|JZS>CJ7u@^ z?gHb%4VLHbR^G04f}XGo`MbX-{O1kr)}o&9yy0PcQjzza!qSm}S9-qVm+J$a_mc<7 zzn>})*>B^JbNk?i`SY3;zlBmC9MO{XatkQG1;q@Kf3+h3mtn{9{FwU&l|$R+8a)Fb zVdn5O3;nR2@rNg#JbU|0o0en#W&iKqAB0VcenxzZyca4Wu?|8?d=ECn(81&4-sYH# zBG(_u9}(mKxS`5F42V>Xi#L&BF(*O}u)M$&iA3!=(aZGy{A^M`CFg?;guOER_6v*L zp4p}*RsLz&1g?`o@!iL}Lp-=c;3%k==}=PtU>*GwUYZ1X!1^j9Q>SbDGJE+Vmiq4V zh5e+1ky}7Lqyv^!f2NXUlSl<$o0mml{BPiG zTTjE^VFC!#ba6R6dfX!Xy@3xuv1XCc8;}v8CRq4795E)itLq+n&mNm}p7ax3yX#?x z?wW}(c{qS#I=az|52~778l9`uu|p1B*crj+qUf;Vs<30^ zIIS+Lqts6y3(^xf%!?`bc6a#g&vp~m;sXQ`QGm~bXYmu+5r0nY$GjuD2uL$+*rtFO zTK_?8AV%SoW^riL5^q%2y2*}roLE^@y$&SDG{`w7=onke>Gso@hp~fq@b9V^iSFS0 zlVE>op9p@Y1bW6K8xj=$SUdM*Iay6rI^C7ZP}8Z(KMW=}5OvFQ!&C>7o*N=F(O_+)`X@Sk9xUG0kv3ep+xoW?Nr zzi9ht_TnY_NTHlt*y6a{=%n|%5}7e@Yqk@mM`!Ao>yT*M6?aA`;wx23+iA2Wsg#CG zN#cb<6y@=A9;z*DC3BPna$;&N&QG$Q%*=VUa-M)WhWNRLUO6cYIq(Ovid|k&o4KjB zxk|6OMZG!YmS$A~<{sW@HMV))Ubt*ed4_nN^}1%l(>YxixCCDL@DS8~RkVf>PI%Y? z1d4)@+WdJb-~ts)qY5fvn8{piK?ZUm!U3vvI7Q=6rGe@EN-)d-ipW51!R99FK9r$o zN4E$L)^M5uW}5-Hy;%f*P;^yWbVpP?QUl{ekbo6sx>Q>HUTUbGou_mVIw4TP5M>c5 z7YxWN`48b!Lh}5iO7L!DcJ3*~DS(rwxb_p-ZWR7BD71O0S)XQze5}%eaSrigD#**^ z2%!|`n*8$zWpL>%U711R<5nn@e5wZ-qxhD>eI7J#1lPBkskJP(qY6tjo}f}MNVZ&y zD#0?hAFVdgU1cCbZ3@ti*Z|or$(8=KjFA379|ey9_Gfv@3BSKYmSIL}2Nw`?hSC15 zoE3!&@jQm9w8CJ~90%3~46A+9?pu!J%Ch3%Vu#J!Dl0!8g?LywDG-+rJrNB-^u**< zyB8P4(X8&bU~I04!QrS`AGdgNVJP5H%L5T&a&t*oa!cJ2Up`e-+oZeR;%0cI=|wq| zs!_FvN<$b)AX|#86eSnYl3iMmAD*loEtn0%m>uk>9YA91yqFIik?^f507WBwsWqT2 z+0BoqT3+PIi`AzO&lN;b4#C{C5N=p-C6x%Yx6Kg!R=Z^SS=u!W36^JHgX?)jjg$-b zpDys)#EU5!AeK^~9s@KyrC)RRm!{|cjY@p_Et!rpV1_V_P6}sTRoj{p<~Yw|;+qa= z+a-6c4sF&F@6DYpt&*J^7ZYH0FltDB0Dd&_di*9}06?fup;m-q=f-5ULt{iJ;z8$@ zjmZ4fJKq>8G(6V1_@z}_neN)Iw8II)g%dI;ZC6A!n3P50LB+7mwe+f9o8p<4IBzI; zzZO`Fa4@z*l!n6)I1ddF{j(V_*tMAm4giH-mwYPp3Uv~#&V1z7l)+m+@tP(hLriE- zN9r(%x<`ofb!S-&^o*jqyBN_LMpxSq4+NIag2G4KLVgr z33f8PTg!Frttg?USc>;dFkqBt6-=HOg~t}~rCz5qpii^36BtY3K|vKvoSjqATqNa1 zK!pMo!&F8Ckf=I1+LI<%C0JbcdwU6C3XwTq(Sw9+aiLdb0YWq#5imGYm}wzOvLz@d zuG~?xO)xXH)m=ThW^E~3OD&1x|=sP_qBr7Hh{z%fv>> zE5XPEW9X%z)Txn2@pUp?dq^eVnSJ-@d}B;hz>%QPOWEc>aT9NwY7_NjeI5t(`e5tn zikPT7XLFGHB{Z3$V}t%>0?NPy*kCD2I!k%R4(Mac)7`lJ>r^m0s4uTfSGH`?b2{mb zoWJ~Yl{gnN)2HHq!BmGv`AB-MTV{1i@jObJ>@>R-FUo)!9;M6HKmBc~iZ!}9gw_SK z=>_$@ePsx+y(#)tN9qi2G%%yWz;8QZUk@ziggLWsNwQ4Q76XV@t|z;hDUftVw$Yii z4moIIfsF8?M-I3WBV1##0hcetKNR`t2XOz~PuUSsZweb{xZ&v3%^J1Nc5e~Fha%Y0 z5yL+?>fnINzJrN$=5HX4pRcOS3#3|W7eE-L`Rt1n6DTyTD9-~ZgA_2NEI>RKVEMu# zp5Y>C{U^VPLIx#Nrons`79camGX8t{GgNq)mu#Yw2$v57SHMtF=N8Wy2_J&7@*W98 z=h^NQ8d?{=qByux`)%Q#-azQRpMqm##|%h~IekYn7#LXb+lZx@Vr(TwXRE zv~(x&S;_YdG5ig7Zc; z%S34-a&RqRE-WojC;ZpOC#21P=Kh_N{)vv8*@e33x7F1;n?;VMT8SGqi6Jcmnep%b zm4WW&N~y+3H5=kv%_1f_BI&5i?uCKd2vM75zf!t>ZST`eSw*g*ecu{X+IiqvlRmQS zFdX_lxC?OLv**>6DldGPB(yyNv=8~1NU_D-JnjAb^_f0*_L zGVac9NAU|dbESv~bxZH=ZPZxd6qz0x_U$fXhgs_pBt@WMZl_-V{D9d2IyG1E9#$#s zS3X(@Hkq^RnGmzqXb9Fw_^uY~q?!`$^I}$LbR3tA9UtO3BPYqI+@eW9PW&`abc&AQ z{XXXq92+&Am;|}g$pg)YT6h9a;X_a1N>AZbmJv!3;bG6rT6=LxaL;8sL6@P+B`^itoFJk0+L;cH;7 zX0xt|n-YQJvZ|CXtDLZ&T;_CW8^s111`(5hqAos|-R^{klNVi9iMLpiXbVJfti%4jEwr(V2 zev3;S1(RtkIsL(yxYo~TSr62ih1_6%wS0E^wN!B<=ypr7_6xT73P$J8@b6m$uipz< zf5!iEg*e{zA>X8)N|ivb+3|1gsI-1ju(FbV0seEkf2bsaHx^;JxF~(?90sE2J;=LpBln=3!svc)BmW_MiFJtJ`hO zW;9UR^)mp3)i3V=|rZB?X^R-zx{BS3mPmldKQ~d9t;hHC^orIIttgW2J zw{+2R%9BRtwA6eFy8k{-@$N!$s?p^*=)AVYf2FkrbTeAJE+#tuv7??L&7aR}cJbh> zG}tduw;7!D?}i^{vEHaT2<|UlgZBbhJQ4-_?TSw9P%`enGTa>1Bk3eEkst)$#iFl% z)Jj>Bn`{U2nJn6G?b?gQlet_BTi<1uPL#7mG>f>2)Xr7Pq!P&yDTj;M{D+X%-0c?5 zG(wdFHUH6p4_A%kX`b?l7HM^2A9>$!a-M4SGVLn+c>Fbf8}xb3Ymvr+YWA`NgJ~0d zaj2CGNUARta|v$`hrP%|qnK9+K$YwSs1ZJ!rKeGm9MaSz7`l`e(C&}qJt3&s~BtScysoP-e3H5Kru{}QUT zv9Iq!a~@dW8v3|woJ<3()KTK#aI=m-mH94|6;UHB$mF<0w22$1>CJrW91x;cVCnu> zJNB3)0XOXml0+a>vT|E2TdRm;73?XCSnm6Hf@}HYSJHn^1NX@yQpA?~Uv&`WL?7-1@&>xOw1y&6D!IkP~qu3(lQykw6>D2G{C(@AXF|L}Po z?8dFt-=@~qEr{rGF`f8X(+MZo$B;EIDQjo3n;fM=Q_C{C&r+?@VD645Yp8ya_KuPA zTw5$n7Ykr3O#@0fMPRQd*Iy|}o;xdvtRt7rj?8X?!K>6}0}B<#;5jBKAOYn*Y$IiD zH#juPrE<~PYkUZ-ywotv#~wPhM2==g{4TK(fz&(^u?0umrgWdVoMH5 zIpmdP%yO=bXlS;}^5@DstsbYn-6CrCN0Rqj_B|fGX-S!8(Qi%@6sh%x_EFiC@?w$9 zm)#?QN*>nG{05o+=`aO)g!K#*UToos|GS^BWMcRtqz^Lkk#4O#<6y~PEb=+Su_J#A zr+y|g%FlYs=lumW%JhF@SKw3%=$}bYZ^mu5L6@+lihcJgblb0eDlTiRyA19ga2q5i zEoeHqe^zY@^ZM#TQ)$9O5f>#XUYmy?tsUX|6Bc?P&ubRP z`!cckv9XAZJ9$N`GI3nu%TFIK(IWIUBk1_oAr#CyyZk*8$sJ+Qq6WHfw3L#Rn5r=1 z2q+*XP)GtGH5S*@3^#jYEDroLMjL_h+ulhI%ju~B=(P-Hlz$?2CyeTu-5ftV6OR|Q zB+=Q!oHU(BqG7`lwJ?8*^u9+%_=zLQE@2%3Q9xGw==H!EKIT5K0eWo8JRwUoids7D=6s;4kr#wUHK6VYtI zj9p2jVm03^9QXs!Ry09sC8;U!jTe7KaP69uj%QOY3BRqNU&8WyiwPu{}+v>oNOezy;RJjsN?VotCjFtn) zfJwG?_NrRGWyv;2k zWmCRJR~zS}nO&-tO+hJ4se8%@+-SduWN86<{(gKzo&u~Ei}w4ctlrsgcz-;|BsLG% zc-kAsue1T#RljI+l0Wr2+CuG(#Iv4TC_U*y@t6!Yhq4Pkef+k8>T7+wTcdm)bk-s5 z|DMpbQO(q)5aQUNg%TX;$-|;uP0y@^wof?O&d3RXNDv;Dxp>!ntTs!J{$0N9Tl_|6 z)fgbOWs0-DHYHPqMmo+c!>QdN?nmmLH{D>u-n}`OxK#e>=-(vTt~tH?M58)9-2WE8 z{+lcA{oh;>_J49ki3pew^o)&`|20TXhC)&I;N_M~WWzs@l5l&}#AdS@5i2^LTmS1q zaKGG%RjZoM5DP}dAoehUEBYUUG{dzP1UozbIh2pP4 zgO`u>*s?HR3>{%$7tHq@ zc$uodS|8hUL&I0dH_2omlPMS~FgS#oQ^ol6dncJ!GX;{qSr%E(euQtVW^xFj6#$*` zS4D@$mne5k z)QYBz0)!kZs~Y!QNS_^Nn>VqY7ynpJ!)^t8n=B%Qnqw4}fGK%eF5LBGZX1Y?c&;2g zL>Ce=P5Me(Mzr+0;kOW|r4#6SlB*R|Y3XNcJM00q+bf0Q=9Ffyvk+e7^iOzMl{X{( z)pm*@%>{sEU+%G@PNZNr#l8Ye&@9G&<*7atg%9rLDMm zW2DGKw!W^gkOeQff?_1GIUrQ15!Q9|^OR2rx6YYb^iEf1 zcahGOEWZbjT%B9HTid(3#?-|H9tKb8Kdw+|ujQgw@fNWP7TWe*fx^iDY-V~Eo;xX5 z4W0{%jv{P_ep3DlKmsZIlrzLo<&^WB?4%TbW?PT;iv~36E|_lFe%#B(cJdG@0BMH5 zbpN`OpwtlZji&4vn##Z13A*Hf+fn5C1?O-*gx$NIYvNL)&gO7*|Ywk}! zLj!M|;Uxdg-b4SX4gOvEW*jJ0kq%{XK@4mbeW7J={_1Xl^8M8hCwm%wjNp2^7_O8K zc>VrG$#x4LqDPs6sMkVq zRoMn)6rBaoa&zdYHj}bO_f4Au=-r~!xzv&7W)vNWEXS6!1mMJ-o%1%Jh1&J#`tB;~gAD5R-}!Y0 zpXQq@8JN_u{qsgpyCSQP;>=!VHu8a3#XgxvmW2{(txjHSmha4tHP1TOab3J)Bvp@f zb{osQRqahuza-7Kb#9=(o)I0E-!1ENHuX-QI!oVKY^<4f##6c)Q&^ge)a)T^epfcA0*B>0fDN^+|=>zV)gcICYHsx3;}qU&Y5TImq%bAtmzm z_B1H(%Nq17B&+u4`6~>ji-lvEG3+-@D*c_ZP=fC!MVHxa7)dP=bczS4`Y)q|KU$*r zz8_%v|2J1`jge72OrYog8aTdVM$;P!qRJ|T#ZU`XzdpPX%drE1&|+K+ju7*28pv}U zKq5=kM9=&l$gqWJ$={Fj6-OelObz2PI8PFUbz!t8?!?-@pCCGm!xX^WLxvoHoM^+} z5gDX%-kDNV(Z`^3jzRZg8_Y0*U)HBcMP##5%f>dE)Z3A;6HDGX&c$)w7OY+om44r% zwRF+Xi09J>(s<#GK*pCc2Rt0EM=rwNvpiVco4k*ZWy5T)ENY(g^}fUHQEzw_!f~|s zd+>17KfHEeO0roLcrjL=)b{AI)Z?3VwZ{#D24sjlz00*V!wdAJYEIJ!sc-eGd;0ns zN_iS-_>2sNV%NzAOzW;Qw@oD64UndD`vR$V5W2H-Dv&yAy?rTd)}>q-O1s>eMf9J) z=RwfgIz8Q4c8y3#cVykU4d)D1Lecg4{)1w`&U^=}ul<{dzFwQxjRozviu)a%PxXAS z1h`ftKe3>FoyC>$b>eZ`ETSLG-!x;LVh7!J-SeAjSjcbwmf0SOHFvvyzjBNzA7J15 zA>c6>nB~e;?QDMyZn^51_LrQ$ycp_Tqw%pWV%ts~?@(MQ|0P+b@O&&AoY&NG_oous zo8LUTYo(t|xnWy)TVP9dZ_sA1FnsW=akXaw*_pSUU}gVf%71IQshy31v{kBGN~Gh2 zXeQW7GZVMqxY`MXbg5CTg*B}nZI;!ZG4!dLE%3r}^l~U07}5O2`1(u0!UT^~6E$YL z8^z$K^q`&^ucCWMnDdIV)fiqSa7rb!H~<#>HFWIL)EDnI^Y5#7Z@<2O{!Ni^wrTm5 zi7!nII$=X+EJ!)E`Y9o|du2Wdj-$W$_p5)N}3uLr9Ep?_=L4LbWcw8~&LX+ntF9=1xWq4H<}9x&J=5c;uF6t!4o!~Ck63tcIcR;GX6-2_!PFMIzFa$ zxOBHqCz3!+IZ)AB`AC)f><9uMflY7{I=2F3Ma09+u_M#Q>Iw<96IaoGmN^IqGHOJa z+|s-dxF(lzwYGw0t2tq_Gm%p=+ zuM;QLuEFl*z-&vSI7tO;hp@kRhd*|Ot8qqYa7jPWh?G`~k>xN=bg7VGGwH)6o|=LQ zAvjqp3C10u)oc#$7wCgYRRLmnV2pLZP>CI&o0 zNDIcVkkm6+blKq;LQS9{fJT;&Mu3oLOqL({n{aJKa=t83NiD2_BMtumYYVEvJY0go zeSo!Hk`B*_X@SkLcAmUZgC9=|-1cGLrB@`7QKii^ck6bQSjO5mVV-73{ai~S_A?2= zHS`Nda)zu|c8v&R11K#(OMaW)Dh@vSg9TR`f8-*DSegPioZ^;~y}l9mA>%edmwtfc z1fELP+GIy4%iLC>K3U9U2qhA8rHbVs6;fjv^&w?q;1-2bnx#+A3Uw!iV7-@Oz{3HY ziL&9s&{K!Q@j|otvlHcOIrb!_5>||`)5L?rS;{Tqg|x8p#PaH_v9vccPjQkPw38ov zVmtVPtS|T{ky234aneH2w^9J9Dd>x0ae-FZ4Yk>RD5>I9sk3sxS184o&5V3f2xISJ zK0;fLQ7m`f5=kQ+kY+kUt<5#tmJg8+e-@X4HvjG zWP3|67@7AFlyeY$4*AnoktyvLD?4?tHTbKdD;RY7GvG>`tqvq%zHz3@;H6okJpox* z%Xya~W2;lkpjC007i609Y*?Yf>QX_R5aB>rI^@Wj{Ye6G=m@{biES4h8(p2q<~pc2 zTD_45cPCw#p|v*u4V%lf%zGO9o)5Q5uPnVTksQel_S8Rx8`w<+?B%LQSp>EU0#Q@z zPag#pD0qhCv+qzE@D8vB^7N%>wWs0)v8)BY6LV*o8`h79Eunh-kj8WJYh24~3~O&Z zO)tolQg5p&JDl-!>MqDOWqIHTrCs6+Tviw5m$ZJ-tbun?B_>pxsrHG}ky#8AIsR-W zlN^rh$5@1E=}lMarL zSkY)fYHdC^06;*$zi$3=TY9{q1Q`FK9NJn9-;@N`CWB0f39UDfLiP;RO@*y7WVa2v z@AEW+%S!`MuwQ+tFmG#r6`}VBJD|2OEZG6=B<1AgnEg6ama+Kq^&@dw!*Z2I0R))z zdLmTflRd5Ee>27Nxzefhhh~#CrKLxzhP`@T_9u5fll|wmEUSaN>C_m#p_AEO7xR$bl2~7J!`Cv?8bP`q^l|gIq@B!G3!~y zS8h-UYm}oM?boe=@BP2|5MZjY6gO_&xvs7q1(ldgX751n3sXRV+mP!fT=#;>mG zja2E$80%Ej<~S`d%KDJY!VBmBf0GOT4?8dYA9ntqCaMZdNGKMDv{}ELml1P~ z>d&#n|IZ^h82*g(L6W2PQ$8F19VO+ptzs^d#}#$qzaGJSv2awD&Gwov<$}@pihUjD zU#nG;e+LM5RxQ=*WQxCgW7(}&8~2KL^6@gR`JuQv{rA(9?FN@4BkH52M#Gw~fAa9+ zx=Pr>{WeeWc6gh2zQy3}88SaHY<>&W`5SJ z!TJ%M|EC{kpA1zzOJC1n63waXYW$s}4VC|mXO0o1QRBp1>C-L1R0pkVyd(*6ni?k& z!7P@>RU6;qr=42V=6`~Kl_#uWq}5Bumtl?8D#`yrf9PGj^j6Y{R9amWkV30KT~PE!f1N(^CIC#x z3)-L&Z`!b1ZC_q}wX;QTWpIkFXXLeOI1jR??>@TdZG_Btda(pPR7o-SAMProv?6$R zGr7XZ;1h|z9Ws$sqR+{tbs_0{)ycQiUH$SD8}(lpe#(=z>!aljU>}8ZXqEf=;5E4m z*(|(jnC!w+TySO&r+5Djj}ma)#oErSp`6a@3z7@uh`truJat$S6TMPgA|x5?zn8`m zGEiB8Q)LNR)gno#|IfSS;e?2cSfpnfWbbw^*`NXSAaZ~d?tXg`AC(IhbaCV3lh3~y8 z24c{w_*a{s^`6zIUA^U#M`7|#&39#__!o)FC;uM6(K$MRv*h#nlVE&2e^cP}MqfQU z3s{tsKAM&K)4;gx*G5h(16+p1)QErn`SrVj(uW1B1&b4FeNaIcI!pb;pRdVPBclY| z6?dU#u>+!V!5D%EWAH{lQ=Y3&G33-YL&<+-oMO#lPpe+&8(j&9*STOws?pnyKf%tx@T7{eg~vlIlXSlV^7e^l|MUs~Qoe;z0p zg|g~k=?&MAU7H|-t`;#nWG~&hIEwQ_UivoIMegcVklv zy_-ZUQo_dY!j&eLZLGtnP7Mg+tB5trg}_Noi6OIo(FF;EQ`aH9^pY$xg;j#A`~rQ9N5ut#Bo~@3=o+y#?wUl$ zrY~o%>ar!@1flvykf=QJPjf+xqO=ej`4#6g$WM9OZ4c0bg~|R*O#&&u7YeX2I!RX&Jp04=%1$EoU6FZ?9klCzK}++n8%3ZfVm)TaNr6!m7CU zCFdCR+`+m#r?ID#oX$~;4F{xWM6kR`X>evFPb$*vFxau_Agu6$N^|L>t$M#WUKz(Z z14$I2k*Klr@(?j32#TFP#yD~dqH7I!TF6)o*1omF&S9?W?`Wc*v;etM6gNUaf@A@1 zH+>+pn|_~|%5;WbZ6dot^hS1RX0pSw_~y5=&Fx*aMSg9B->aNQD~95D6jQ(0G!bhG z_6;7!1z75>d;PU;_HYKO9?he@Nv9%UnM2^jlAl80>yVi&s(1=pwC z;b#gR0ravQ_}1QVoaW&~{JlqbTMJQ7h-GMv=hA2@A?8=-w}O{865HlW*+00eewkwF z7f`qIbzu!TM5m1B>bWinbk@P@q+~RVt)}>ss>8@!rS6oip)-t&d9VNtzm$UeP9K{P z!PiDu{(^F5kuYgv*Opijn@vAEneqy$BT^2*a(z#oh-r?n4%?gSO}tH$gDJGBe|OH6 z`7-fDE+qT?7vrl8=J(depWly#Ne;4qm;@4w9rU_?VEPmxV*aT*ZYy*8Yy6%t?(&P< zU{~C9mpc95aP0Fp{7pHv2QkhuaQm#Si8+eA{F0uK`s#c*G;epr`89d7XMmX|lzS|6 zwBxSRTFm~#<+_8^nsv*T%O#YuMl6XiJ^H=@o@r}FMu9l#3}1|kEp`mvc0-2j?2@p$ zrbfo?ROV7hJd?6>3WL(C&He0L`_y7|6{(7$a{%H9l?jVUN@15KW8&H8u&b=KiOYIX zMnh3fQS%=$*cGAP;<1;`KYY4-@tOFMv3_9wgWN=ISNgAp*4jvx(jRo{Rt6m5G%O!Z z00W<>nKwnSbprl&ujNIhr&TSU@N3En^^L%M;F%BRS3fj9Q$KIAX}d4%LGB?(H2Bje#&WnHw1y)7IOBm{&s3yPdMzebnZ(dN;#zH!{A|eDvszAWgX6T?_By}lvhRBA#7-$1HdOyG?+-lZid8`Y=%&zthLot@lN6^ z)AXOCKv7X#n)d9F5K|BLyjFtjf^-r!eo$RoF3o2;B7hAz=qnGN|GHtPRl=X5rI`b- zhI~0Je*ln?MKI3)K(38AeZ-#MC_LXZVsSMx+qtB6al*m zSno0ks)R$ik_yYdl10MS?d2Nd7)*d<5T%o%VTx>Ev&kmZajQXt+47rBIYaBx8*B=N zat^u&)Nh)){Mo6GrVwOy2s9g7pl7-)0I`(Kl8Zrm#ivBEyF_-pr{?2CTUPtfDshsWRRUYOvcPDxq zEu-fwMrBt>#S&tL(qHYe&xy4UcgJMdOW?K>`k^i|8%w7?5V12Ju};Vla90n#K({A= zX+Q-y@@s%mYd{HYfF5js9*C`5t3MXNyb%CB+6UN409TJiKJK0V)rbRXK}&p1Fk(&6 zw@!#=P0&S6GgU@&#G=E9lYCDw@KZDJ z0Zrk=DNlxJ4}Al7gong_YF>KDn=v8(uYnP+<8 zuQxNP)dukFx4$f!lEA6?yh~^!l!2k_(eWXsK(+NV_Wd0C_0SHRiTn*uJC;45n`<&X zc8snxmB;j#V4e0JcF>)lr7=T_gg-&)dUtp&XVhkcZJdY$vL1tE1SB*djj_{AvwKUN zLEE)6((}7^XJ-pCc2Jyew}0RVR}Zs9SPhg?cWIaYP#5%Gmn&2^r+9LzfwazW?JF;P zP8?UUzh+?^e(;_IE@O|?=LJ>WEujrln|+*z3Ijk4faiSC^DDE*ov7bY>pSsrNERMl zk02-v2e0q_`+j66{#&V?MpvX;7ZP4yb6IcQX73y|)x1IFW*kqZL0?Hr_qswbMm+w@ zMS%&!k9~)pW5JL@E%NEjo)hHndM)3F%sYv#i-mqsyfT5_k@1S2JK0crQ>lB~?jR!y zg!5@#sf_gV`a>uvBZ$Pdft>7EinQ#5-FPSrMdg5KDA>HI!Gh@)pZza~{yhsaA?k7X zSN=muRdu$hoT|uOLIjZ2PJwmyVLd3=%O}x)<6;o;@$_fj^4yY+nEro{JTu66SJ44t zTSjudBihI!-(E*BKOalkTj7|Kjq?U)kvo&+LJEXDRt(w~_^`S; z!NuRFDkvt2_hs_^-Ru66n8Iq=Wjrta@EJBov`d*T5{CtgYqK>~s>=LLVW*R}||wGV9p9 z>mr0Rpv79gLmxL+zd^!EWT;v~?lJ5@0_{=)k+hEJQNaX6Z(T#!@V!B@uC4?jk zui|S$d;#@23HS_F_HR81*fHO!y1$(^dn&9o8HvShLwdmik|i4b`Tw<-*H>-E#g_#L z83CK0^6kCf$T~Lw6&ENKQz+$KOq&UZy z;W%6;2+s{!#RZ_abNIy=fPuE?GPsVLsK)rXX*05SifsS}A=Oc1CFt$Y6sSQmFd07p z%5#S-JFsX2u}fwPHGS-SFLG1W%;IPQ{{}|BVG+WF>1q2a!g?xOBcaT~ZpXZ9lYQbU z!lUtg+yG@D$I_5vgaLr&O>O5OTLJqJ<0L3IBx@if+i*_V{m{lT!mSjB{JQB-X=k71 z&!NHqfD;Wfi&SfbMQo01e_KH(AW}YEOJ`*g?R)HYRJrAhT`0odsp1?!(FOVW1_-x% zh$;$8oPghI?44Zhr7%%;MXqWx8mM8HeQs$~UQwb8Z^v`9VmHCg{5hN= z<;E7qULg%(d=kWBY(PjZj2i*Iyq?X`oJ$<0(Au2a_mMP0{=nY=ULupxktqE6m^}j zr}m6E<4KqL4@~knpzIiXcUaJm0#)A$d2al#H7s21NZ#S~nE>2g{2mB}Mj@57Z6X_r z1fugOp0b*Wn;_5-*4f_dOC};*J=W^fl1>;?y{BJ#FH*~=GdQhW@ZGKs2D78iwPkDA zD2IIw{D6I1IuRY(%rX7#kMMu1Xm8+SbzTf_S76Zps}V%858@qfo&bL|L; zjdggWN!&7R44)CScH|Qju+9B=i^108@RuFfYWBbWvPe6Z1;NM}|LS%37Iv2VEr-w0 ze66vhdZ#hg)())LB*7_WH=2}r`^h>!q7(4TMkvK>yeaCCsCQ{AbSPn=Ob%_ep z0FV6OGPvm@j@FX%;lRj)1E<@}gF8CK#!)%nIys8WJHjqQ!dDeIE~){ z3W^&3LRXlG2AP4tD*uo>8=derHe0S0#bK{D-M+GP$y|`t6S-k-&Shu849!GlQ;B)? zcKQyb`e9zfsTHC@i`))!a+b}GO)*RRcP4bw7{koo!k(6047xv$MX z7wcWSfAz;5&!AXCXY+Q1h{E3eF|)7LLR!`=*Y5b-CFi0ad|9ccBxtx~gZ)-`Gp^#j zJtMf&>#n2~7HfISq@o&fopgfaT}a!)^Ereu_w(QTYzdj(C2t$}?_5Yf@{F0#FZAz7 za(YQ-q=4^uq7m^`VK5OpVpW}Df5{a;&DOEDtAUh*NTf+ZT+nEL)%m#Y?Udc~-qSsW zWU9#Xo7A96i#lne#TqEky!7`>hHUAnVfWagf%|)+9^(xewnwbgcII*c$ys}^jW({n zhG-b+l7w>3!5b~4YfUzwD~wx-YNkDmX$j%qRX1*g{Dq5)&$SG5P1+U6%7iQ6PU`(@ z&$ol!T3Oms@YOAo;oosgbbMYI%(0OECIss13?Ye>(J%yG(m#Lsz7F*Eyjc~6Q(meun(nat8ti;9l+V=p#VGg*45X&{WxVPi z4CDSN_(Ep-L(r(*Xr{|`ksbQqO3iAcG3!(Y!}E8;?0f6kO#jDwZpe&oQ&r!`c~4HQ zR*Nmkbb(T?+4fZW-u_q}k^7%c1Px?7w6)(NInX$(#&ZE)wljI0-cic`_{Ek!Q7$X{nkxjNo)#-W<%@-JyN+FMJ&C79=vGvwBq`8ca6u;>VnGKehd*H0B#f(dgj8z`W4$WTnrWyWOg`zK#xyScVfnyIzw1~6Pyow&|ayDuH zOzrXf&m2Orjw<^L@$#Az0!U;me?F^aMnzRfq5$Jo{X_xmk?+iqPo^P^lUiEcIITmy z#M2|;afii_{8im*R4u6XuwP%*>~{*cYW;o((Mp5q8y|t;J^cIcCfVaSeTP3!3aa5h z;SJ9Wau&1xgnsl~R6|M#2^p9 zpR-<;!3qW^8@5;TjT5BgV7AossOZ|H{m2}+8 z`cs6g?Q4fEV~p4U@UYMdX3r?a78;IWAbRgvBj_dfMA2JwK=|Y%>-xv<@2NggP9lTd zSmmR;5^(Q~B_{sh*LjPZ7_DP{EQZ zXzGqophbY?H1t7xIwb7rryP79OK4_W zgdYKjC|c#7jm-WV<25E5osk+61{KHICqAK@UK1)5u^U&@RkEga!vRSvmo2tIE^!$o zCnv7t4`)|3OA_K)!R;_000-863su6mt zM(<}Sq{)~oN9B;Qc7BQ?X$-C`s_%@N1u4}@6$d|57M>wqA~Ew58m875I&W24SJPiS z+EGfq+(qw9W99HXoh!b~`MGqB_Ncs>Kq^Tih9e=9wv;%Rd{ks?Pu<;oB`f#Jo(N$P z__JXy1|7)iD_dLSKspt5`;Ob}}@yS*S*RBNmt4Oert1+UnWhVn@ZQw!5}16p~Xe^P7{hZ(H$>z>-V@wqCk zOt@C6rGuoRt&u)9v4A>m(cPe_FeXf;L;zWv;Wygt8_dOh)VRZ1gm_iZqTig;YqtQ=h zrAubgG${<@r}#jj@ot2K1dwV>J1@cUVS$;7K{ax-vq_d8JQGt7W{w$5e<6O3!#@_9 zm^jARMRUehj9uR>gbo5^yG30`XL~rq6mw4?JXuwQ*v-lpW{v96pkhP@3gT@BN!%Pp zoyxFcrpO%q0FGyspW7aKCS1FNq4mO!oJAeKC!E3u^a`NynZB6S6Y0^}`bxrNl&}8F z7NIIzcG=(|ie%X}U0seq&}>Obr(g3Rkd!rIZpvc&T396HROrd0V~Bpz=^)sg*hdRwQ4UG59K9!*ENswFE5+kQg# zNK_CgI;N0nP>nciXQS3?)X54|;_*TR3=lHWx+^S38A&4imISMKD*F#wTBqN|5&y7iBMw~GRYm@W zQ;$7N8^VcZ4TwVAe;&4x1*-eB0`w^WqbeVF8732r1jBg!*afUXC)38TqOscGorf}& zuj23~kfHBE@?Z>lS42k%m^xx1mCkm{HEl4{0ywUHb9-^m9)39X(m$Yf5dp0DB8h6s zYj3yB00!6(j!@h(*0&!A=}=#Po(Ne{+Cewd3cyk{w@G~+3UO)2bd&^*k5l0Oa)Gi8 z#ia+l!{{H#aAvBj%^fOoiBp|+DlQ-Dc9&raBQOl<@Qyu)4bT7}gTjfh3Ou^uVw}`N zEhHl?@`>soUom%aSvULUFwA2Y0w2Bs9?+h*%VDKvP8eO4t9e)(R=ueL9ljKrI895G zSA}(iLTSKUhoQ%^^M^F(X~4!Y-E#RhME}vbc$fiy!o>ru4(;yRE#@k%PSs#WuH8y% zHb|!n4qLLs{J)R{gw6T0lQ%?K(-KSu@L(N&v=m8?5#@8|mF1A67f18d>(ElJQ;mU1 zbMuq=c2KM!l+LM!&*@hYc`ByQKZrIEqBG&53M5(4|+;x;f&Wk~H}mh!c}N;tPGKK)61 z0t54@Gx0bAcH^@NXVKIE3_!GOf#9Y_An-a7mPfa6gn4Kry5HJMGU)q30YsOR!K&z& zA+SMWsv&H~3L1@HRQ^aUWq5^qH#dl za6%Qa@qc}aa;7>&PyTO>b+HI8+eI3F#Cwl2jkv48ODs1lEKMXdS9c;UqRNGfUnLL* zO9DO~fjyi`hHEZHV}OHTmM?>K*|I7$L;RGW3qt`p#hlTe=%-sEx-Xs|Mo`EL3!XaN z0gu%ED{haqlSog=d#dIYOg8SJ6m*XmVo$jG8v7?^s$flN(JcY8Yk`^{i0D-G-$0t9 zK-s)H9WG8KX-~9vDih8tuG2<>_<324R|WlGiM{}VG*uP3CRe1qoJFN;(|(fW=VVp< zqBt~pfFxCEfZ>Se^6Wp&;EH8C@VfIQ-`~Qd$ z-PH=*8nv(*wZ5T(1yh*JQKB&Yi&S)s;5~n55*SZU*LW^@e&@DZ*veULZXPXOqWzeCf%)oY#rx`mV=I~kiBEvvgG=*8_B6d+G+;Au=V2UmQCI3&UD_~cG^56g5f=bZ2SpxVYd z>t?J9t6-0=VgpRu=`K^TmaXvCEzV#V>Tc@`?he6xH_M)A4Q-RIo;b56KRHvfvg~_a z!R}mKo&|1S3rth3ZqF3AiMEIhgkA%iwz#;yi3$GPPH|y<>EE9U!jrWvGv76ib4AO) zgU|%|?ri($q5AawTW;DQ)_$Ew@9~q!U7^$8tqw#!3_AXN_nbwJL2~_pev3>5&l{(X zjb+>C;oYdf<&F7wpiXdLMzA01F9}V~pd1BXB_Rv>Rss6nAT{w28S@Z*zz}V69kUQ< z@_ev(vz}unf@`K8nSGciZy06}Z?DD=eXjw1VVJm%{Bz)kL|&Jf7_NLkslWj~nOZBF z5DOO7=y}RWf!x5e!vIXLq8u?URxY76bnkjx+fwu(L$N4oK)V4b$fVd5S_rGhKPm_L zhtlncYC!wM;F#-WIc_Yb7jc*Kv}FTRi*-DvmCyu-WSit?%b5R6W7;?_k=J;T!noq? zxV*-M4dcY_(}X>HQyy}=qSwU6=7^IT6L_Ypb6O{U#$N7btowd4G_O4`4>v=R-8*-@ z5~C|H0JB|4_h~b;B)+W`W5%Ml>qkJ1BlENq#$FB@&ZPhVw2*piL4==`7P`M z=E*06nGt4kXtR;@x*xw$zL+R9*3Qf-sWHx6j)C829H9Vj7t_wu#Sa0#4p8a80vKx* zCVn#m_It7KE@ye{XHM#-Cbwpsk^xta)5U_b@)Wb@Gp*R2KbKeWQJfZRO1c5!^B;;+ zF5XjXf|K-HW8smLXbbbtmowbCv*`iyXhuK%Ev6qn&li==7MI;db>bFLD4^n+r(eF0 z$^=+R1Y$a2&H?h5aTV$nzk;Cp=8*=M<>r^wl~&}gszv|n0(vuHFgt0G@Y9F|VD@jt zT*%vtnKphNPQS{de-3~rlh(%KQzwx2mc#=4@lXcQ z6xo&F%2$Z2TzYj-)Qr2g*oB?+s=D3XLb~ z&#P)y%IPm`Zs6J6_GjI^i<>xbjjexMwHAJ-Corgx#T?07 z^d4qG$Y4eXdQQ5dEyKxej_RsAq8X^4-Lm*I&tdn+ z=(h*Y-jvQ=@a8U@IF;C-_z%8Lwy+~Op@X8!E!g~cZ6a4Xj7_DA!v~>#nuUJCz}eP> zog6EV{X-B(!9FJwi=w4Vh=k?m=CO4{3(04~p)~^tf>A2ne?kw_-}rB@p|(2&?OWKe zINRo)0_^WYCU17#wQgtvis9aS#^n>_#42IlpL|=@@!vT~mc{l0ea5%DC;?$#PNn*A zQ=2Z3#03UOzYURI)^c9L3<}g)fRtuIlLa8rT+E+ny*`{*iL+~BM;A&hq5phCJ;4`A z78i(~LR-S8!ONG{2CS!~cxTG2XSiHS-n^WHUzDi=rb;C;fLxaa*Vm&dL7VlZHJTr9 z5h|xqf_;iqZenDsAv&?KWy>H`uFg0T$y*5-Aby1bVUIypr9+MyuoF{70Gx}PC05xT zI@DogDMoWEZyYB@F;XC&e6y~nQt+w!gYn|JWzst1E~*TpPL1LQC5%)^r77vm6k8-F z3CEGb3+qyjHvQ!gnp}g=Q)Wa~h0|}@TCn6xSO63k+7DR7uQ7-hA;ss(ks>BfBuQN~ zKlUU%2@o;FA0Bb5*40IeP#?mi3=SBk+m(EixWE_J{Kj9<2aRCT<07LF56I1LU#|hA z58^*VTH?qOohcs!WnD!Qjv`leyYa*ZVXL%Zqh_X`uUW&nSbhGfKXi+G7qkx4q$71hMs)Jzy6+Sz<|_z53Ool zr-GKel4$%B?fm`uilnDN!fx;htDDT{L&+`gAoL|#-R5knuz zhoGHCu}1yO5}_JVrU7Pmexq0QHKge_7P(F53~JDu8=u?T;s_dwE)V#o89dI!mB zfn%?%WQU9=U35B4rvXr)FxhC=OSw{^e!T!rzr4Y2|>|F$!L%Q|Wl2eGF>JufP@_N`CG;b*=WP z2P%Wh+tC6Yi~65rdIsNDv<+Z=JXdvAr#QyLOjoAJ^v$GrwCmDFJ@s^5|9NWJ;W=uq zK{YyI88{(Lw;4cPLU9|ocGBo<+5ME_HS%2j_p8Vs1;$%9eE&+mUV9ymf&BlP2Z#Uw z;0$p0|7#GF0KiZXrqPF#jDn#VfB-j0sb~@+g}sMt%m2L43)x-o_Jw}J_haLyR)plR zA%DZ%y(U+VV0V2Q$+j$CD3wiRR3dn&SgKHU{>YVIsU9r&Xt5vd;b!{TsP6<+Y-+1p zZ?#oW|L2`#zgg|PuHMeuc|M?Ka`Cz}?cMmhKjibxfxk&>V;GD@L?hu|e=wdHtd=hH zxH`aY;=gjSs={gWGoLs5j}Cg9waEu_s@6)H=weZ*^m`HN1h#u(k1r`? zo=eHcTOLbBGJ9a`F}NKIIzn56FqE7VbHBBJlKb%+|2TgqJd)TDh|Rvz2>4W9{0pHk5RvzR3M+ z$JXZUG8JC`U}b+>+>`i_Wt1MR(gnXw@}GEl#_Cw4w}f_D#+|dj9T!1JqpG|G(aMUg z_5Eb?EAsqejLM3>SRI!o)~{FPCKkwKi8e&k=oZ5be%9!gvyEn)Sla*TrYH(>BRH*W zprpMJ9mbV8cZIQ|nD?nJc`NZ9l>WZh+(O`SR$*YhX$KyVzo?feRc9&%%=Ftel-2_0 z&pWMg-gokE&KW@NefIBk5aSjm=H_Nn#y^_7p3^0*mFkt(gTQ}rHe##RJ!Tmjez7+r z@C>~f)hC#Brac$xb=Y-(Dx{c{I<+%O9n%+)Pm|tU2{$rH$+YMu4*MP0wbq=P>wJ*~ zUTE9cH3C|VV7U!RDrD#qJiub#W0$&c^aF5wHPP#gd*Hq*5ihF}@wbSYAcMTt^mv~% zBW>c!4aE&p9~3$rQVM#roUu(m}j&ZMX8(%qYq8B9C)}Ws@;qMt8KRv52229W zlTFVU{iP%dfnn#aH)t>arwgqfS*p)*zlPep1x88!KUi}?sDOzGdPCf9{Hzo+iX`$0 z|6-@VgEVCnpwM}TBGEdVKJ4-dKRamM-B0o(ASy1i1>2kH{Gkn%;cR1!3h|z}?o(c> zZ4_=~jf~*hU?PkCtr)AN**?&=F}>a~RG==s3)))b3}eJz{t%;iZ6Q?_u%9T}RZ8EL z38Lbt4h$hyhmoKk3Qbe>?irp1lF$yPK2Q@5fXk>upWk6{G6!(z(JlBXIDp_XZ&{B%l#*?5lQ9|F%>GYyjZ7${FND z$-_ua1!bB3Rwgc3Fx*%{u?Z@5*~*t0FP3d^%E7R27LV%#-I@{!=Ow?|3b#N&D z`cZ1c(K@$6nn$B%-^zyLT9-BB74b9P&X*ZK15osb(2Fxcpei*f2P#~0rgI}~`z z_6kJH_a3HC%LbOaWUO&tpBmgIj9__4kyhyarx|8aol>2uAY7+P5 z2Jg6=nI4%lfk-`{f2z*5 zN^+<$OTo=W+=0&QD+n9K7~82=M-SB@(4#X~>kJKxNf8eULLbSDL_d|47Rc{@|NdJ> zO?2x9*Dh-9k+G_{T^qBZw}H=a;a!MlKm{{BwYQ96GcIQcx4gK|Lr5avwQE9RPIM2u zd!qL9PYfI#>3ld|!i;>JFKVs8;3tV6MNvW4gG&oz0lF%H6KU(|7iI%AKkXM+cCD2f zT6K5|>9rlquEAimc6kgetLB860`cKWqEAFo=QfUMvjeP}%sz9tNuRXa)(6^u>|bHe zbRpyY`Ry>*W;0Z$K}9f}t%rGv!8&p4x4-TSvC2-t-f{bTn6(|=U22Iq;G#54pCy_$ z^30zM|9GxXd{L8hipKHQV3eNEy7R7T9-G2ewEULPuNceEG{n4zv%!x>`}Hb{gDYTq zZq5Rknv-)sYmV0~Dm(J_cCBB6jU3W~!4G4+xSMFgaLcTEF99+B4_6*<--$e)vd@=3 zuSog=%P&a-C6M3dny$^B(h9lLhve=QP!wmqKYkW^3ny~IK0d2Byfs#upe~BZmIZ#C zu=2wAlZBbK@%neJ1|!IN`AC4q_x&ARX$EzSM({%GD1v)!g30lTdiFi3CxVG>i2+$5 z(5ivCqhe2Kekj!5bh0#16IA)QdTl4fW$ZyFJmIpgL4-1lva3Fql?tMCzQPlPIx-CM z9AR&rw)2@0jTlxwgUqa_W@up`5Pc+)7<@;o7o&#%f@}C@8A0cGB)Ivjh~(GL0SQOj zulO+GT^PbHydVy6lvQ&C-LNF356u%fH6b+kksOSS4Gy57_HP!w!oaFk^Y!`k!8Bm= zAbQ5qXu=wn;Ld!p|5l0D*uf~b>zq`Dtz(1jbSRCsfGv$13vNatsEp<8!kV>?fhLb* z2*qpgiR+RjX3mOhC^7E$0qtoEqcxlLzOZEK#A9=Op^1o}S&J|8@mg;Kou9=2Pi|u^ zOWHdtVQVd6=OtkmKXIQg@lZGMxC(pIC-J;1@p3Kk&r9MBe$pLZ(t~c&lTXr1R?=Hn z(%-eDkC!9>0R%c5+aSxof)9e24FPsTP}U)6uMp4~3j#qhwq7!>Z!&&%GGTWz@p>}p zYce@O3MGFEwO$IXZwh^O3S)N)0lzamM*M;`?pasT_(V#k288E4J5``NRcJj`_%&6O zAPrAWKwK|P+BZ$sE^#RsR<`87?IAYGk5{edr8(>I+CoXTLAjuD-H0(KS+afT+( zkhF8QNCp*Bf$a2>YrVXUc=Ysrlj^Y&gpWxjv!%HS$m=!Kg%+%)hOs_W1CST64o}h_ zPI1Du?MxO(ohI$&F*29II3ovmnN^@faJ(S{#R$F_Rko~hwcR~@_9k6Uqgob^9Gb>h zHf^?aOM7>IW~xx*tNJH|v@E+>i~(v_XtyPmt&m5ULBl&Tuq{N*v z3@2O4r!zlcN8(>w{+m2&X=f)UaI7`naX2qXI!b}Dz@L& ze<)mSG#!p5$xW|aZ)Jfpq@M1m*J*P(@PmE7a4>509zHH->TLWCcGw1@y>~X5MTTSj z_&W1)GwN?b5!cR}(Xuv{g?vw)+u6GP(}y!IDwn*;e6Uob^y^pMc84lFmd{k7?=4r2 zjj6w;SzI`6`z)R}j*OPct`|A>YfcGW;O;9H-?!|5;riWeGL)`v*2Uq=44k@ z8jnP9QVI|Wn#hZdyIrFIL1XY2Cu8q$rl)F4#Q;A*z`v|$DyH}-r0a@n%8Qi|FUJCm zycp&)&Cn#RGtenxj&mG4veQ`MSk8`p@K7I$U0wG1RSTd87$Al271kC~8^!dHVmK(m zs(A4)Q-?*!kI>bn(G*i^rLbo5)q1pVOG%XlIeZL~8`a+CCAni#yYzK+LqNf{rcaGH z@f=FcB$3AlcZY`RC(q=_Iz&?y!2tjaxOsq_a`qc8$22JD{EX2~>0p}Zcx|LrN>-H~Je53(ZV=8Qc=WQ>A= zx()iY=BB-y$C05=5y=oDa(Ls=-1V=(;Vc@`f8jcOE#8aM@whzf+&o?p&(4N8UuyKV z|FYuS5Q>JwRojaGOVfk^1;uUtfG!G{Yf_f#8`}UZOvFq!B{plXYB2MBk`SF$Qgshd z^jmYvi^Mw51a2Ha5C$*cdc{7WoQ=c}%uJZL;^I=y25ck+9I z)=h!+a3(Q2B4jZ2Fm0(D_5-4j{m{ZOvr|LHYA@#=-)lUiD^cNIQ&bxcOw(b`)-v+=D$TPe=+6nRN*qDpOrbihM@jdr<-aw#Uf{N!^JA;R^_WTGh5`CCG}8M9cPMCD#kYm7wQs*k_W&d9Hb86N(n=O z#z+wduii36%Cui?{34;wUJEx&O)zP+ zr>LINA_ap`Le0;zNCCTv(;&?-rJiz^)@g4Vf&U;E%hjUK??6e*A~UTg?v^U2uFgb< zIrSs(r{_lz!Y(2$2puZrH$yjt5Ud! zjl{;xG5D_>lSxo~K?N*8o;B)=~*7Nu|WKU2^lyJwJJ1g=l5> zqCF;d?hmz79+))sAAJsHd-3#aM%X`d8kq9&`-lZej+L`8Co2gfPGv9nXF~Ds%ijM2 zHTA-mL!r{jVMg`D_Rb0;#@cGI3~H3e8L4AC5bB-`PHvZb*klQ{q-@($)vqQs)$W#= zRP*fv_hSc(&C|!il%%=dZX1fzfWN=@?@!MMx18PYOj=D}WD=HgOAkA0(v z%#(*{9oo4G^6uq*r3`Q=i40xVAy?;<4T5v~9p-38{*EIW^2r2(MjjY|FK86~)3qDQ zmX+Y@aU*)@V;bscPL7w@6v;lbhsyLae38rwW`OR)5T*uPCoG8_F3(_amW|TN#l#ya z9uRK6jPcDiCwk8ukiNf+i=wqaq7@G*iC!mUm0MCWXAWrvUnf-)Thg5BQZAKWr*yx> zch19N4|{Y^DTJ~IOr`w9y}Oz@k*d*TtW&`t;rL81<;u%DP_*HjCV&K79Y&DG^cK}{ zmg3KrdH+144#mzH;G}2}HN2heiSIV7S_ta@H z5&qMY`JqhD73$RVRR98sIKg=eH#=qv%_;stNFFHf1z@ zJiZf$SQUVPzY8VhB1&Kph^s;MURBfYs4OEOALLFgjnVw{)p!pUhIn@mBoYGi1?qTt zMnFS!*05-G!(euwdSTTLoEm>XO~*&RkMQX;(0AW=Jp_6`!m?1Z>oX@{WPhue{iqv6 z|F)LxN4N#;3x|c~5&^^acLyEp10$Sl9VR)zO0~{vZsj*Ww-%#Mh}z!szIR(izSW29 zD+NOublp;#2QVS`wNMN+>uXV88FKWT`+XFb?_k-8*6oEwx&wgb+9k1Z;Tu8Tb=oo9 zQtsU0Zh*S@6$*Z#>gt{O9$7ya3jVjJ)D?NGvgeXcRs_In^4Bn(GrXT6{yAMx+v=o{ zRvjD_3y%}BGx-6_%A5y~tnBk}*Z8yZY*=x!g82aeuv zU4qPcH9)h$wXuG#l-4hv_>%+F{q#AXunVBl6a_R61TLUA0T{D%xT71n3t3N25)3+S zC7?v^s+}O22p>?=rU2xd02GKDx(r;Ws&TxUa*rfd2b%#jfi%1h7(*ysZ6yaDRmIYlMop!jHVxw`)@^%KCeW72lw z?7*O(@c*%D7sL1Mq{65AlSY*@i^%$y$FFfc(G)|zXZwM&C@q{|(Gmv&ov6hT&ruyL zC!g@YtDw}iK*Q!}m!puGiim%Zz_u#3^=C(x!I&ytFg-;iYBEkVm_WTLOt=#Kg&rWY z$B;rTmm)5Qq9NBqDc6Ft_s7wZvTNlybqm3Xv5 zeSiq88gHA=Y#WpYAoEGIB#Xl@{n|Et42U8J@z0j zk*op}YEJKk4G4)3w@NPY1j#ffBSA@!zKR1>lgW@IaP&YlXCM$#GHj?h%3uoGIz$Kt zn}Z+3$pG@s!Z7-jPOb5fZ8^dGI>M;|F+V}{l9=T6uP_(ai{L8`IVopY?;_2+CDMBK zgCAsG(M3<8;@1slVLw3TDh5Ei+9)vq_!&TJl06=zE?X1Zc#TSoV5rNr&Q~j*bsF#@Guj84@k%^g&-!OkY%`k$liBM9PY-IB<_vVt-qb)YG|v{a1i7SqgMtf(9i@aHQ<*3uhAn|Z$CY?L z5G6g#f&NT}HGt2rc#T8|1tBMfL?ZI$?qrlA zviE(Hxis{YxI`YGZ8EXqk)X9E(sa~uihZP*2IM$NHH!c77hp32z#t?r2rdMIDBDCN zb(o{HRxt2TfHXu%2pt5A!ID7`PGBh%CEWp~?4o24pJFrC z$V76>ES|*m8&y$a1c_CHYR7Y6{mepV65sH{}( z73l2OnD`@t><PM7kk_Cqg$?28;Mp6bt>PAeJbStPB@9=W0-FK zm?RL&#B6$9T~z<}Jt;L0O0TA?0p^UODLBTp${CI|%Dv#t4Mz&2AlMT?JQx5AU09cm zs97huhL(%AFTbSE&^;ihu+ydS9@Z+5-M&4oEm+=6VV<6WlaP)fh+afZF8&JR-0*7# z1YfFTR#jrl5Ar@vJ&G8AkBCTfF2BZ1Oh-SNh8v^Zcc!o)A-Qk!{Uot!7p-EaO9_Zo>=&hv!E z;bPj?%30FA)x=qP!?07|Xt3#@wIOO+UI&_DyU2GGIL}M}}m{UX<^cI4BU( z$6CQ9n=H4LRStBAC(4&deQ^7>^}-O>8?M-=$K0YX(7$ft-ciusn5PCLN;zEZk6EyY zPVOf{>x0T!j8v$yx)ZW#SQPqKJ(P{`q>R|Rrg7%N6AVVF zhW}x9uzu+FQu^PE@qeX@6wfUC$a%k;wb5Iol3R08y&|m4*NjC zoxb}uqez^ZBpP+2%BgonAYY4_)rQ7)?}KR?XHxmquAo62cd_nPtI_~)5_6p}pMJOo z_Q$_7Jp%I|Q2l~AZ?Xi9C-Q|_o7{RjzJDo`3w+>48lEiG;S<#S@jBHUt3!`A|?9aK@Jm1%2|CwQJ2x|Tm8Wuy% zX@xaujZCfOTJh!jh1_2pU!u3Xsn<&WEp7_^vb>L9Uf(=`=IckdG_U=$zgY_(Bu ze=$TrHh(iFlRmK@!UYLVnGl3j66Qu2K5#@9Q$48KLz7@^FGp=W);iH=>brxM_1Tz1 zZru(rvcQpy1Pr>%n6d;C)sp}~exFajmjr#^K;+vzOdXe6D}ztU!XD~topBgz*yZeA zEUU|=KP;ug(s7titNtw?%p@t%ZqDoD^crQ=3HR1G``7U<6G~^bY;rb2tWlCWMJJ@= zB=yt54bUw5hr&jf2HElqY*bIJsXBKx8e$C83M_32T?C^=VRPmq;LYQbI$e#4ij9dS z@>1m~`n1UxRUn6C?%KViP)J?m`0~EWj;A?-=hdEM;Rnd|rYAC2zJVJwu*~E#jOpga zm$7%ce)aKL*_8bKrKOa2&EMs$({cB}Vnhm=@-t)-t6tr^38OaUYKW51yh)BZ`9)SI z28{!qKlW6iz&KTHe_J~hTa1_aQy0ta!C<7f0%O_Gs zWgRngrkUGyX;eS&Ed7VN_x+@$CD&W`O?bbi4JxI7ZfY&e2D&fLp`0N#GMFFzHf;Ec zM87f%eqf3j7S+63)>hwXhsAbiZX6|xrSzB?%AVZATu6m~pTrb3BX>N!2s{yBk3!Gw zj8THu;-$3BCzbFGSnBOp;b-*v zSIx>y^jFe9rAvvi9!^x~{EwQ#zHJr%CA(>_5g`NuUecY@dqb(;x-ooGg)c}(3nhrp zM}i)$-{aYJ{Ouik>%+08{=7nV-E&3HuDHCizE2ZtjV*Pbr1l3Dmjx@2`5CL6w2M2s zGR)=yiaW>NYsXPym)DGCb}DBCwdX6zpnVUs#?#~=s)&uxkE5@OfKe^6LyQtwQteVI zsnk2hjfNJ|AQVj0$0rDZ!tW`N-*)tvjFSwppHwutvAlkZB=Pr6(IHZ*nAudP)K-;Z zd3`L-8QW?454Bk>gP{*86=n!N+9~#6)%q1yJGUw1r5{9 zhFYw_w=Ddnq8jxAoa}rGdBf~Zf_PnO5&n_P4Vfei6uCErHP1WBwv}KHlR|>@YC$qdLo9hLiQp~$IOuS!F0A-?eQ^OLA?uYiQveYExa2Av$)t~_k35}=v#M)pc;(NACAx3jz=#paCv`)%piiUDL&`5dA}dQty*3teZxHu_6o6OB!GH zF*JHVE|gj@E#=&g6M9~VIJ+>ECAI;}QS2Ai6S9+Nc0X>&89B+e(=j#nf9V94U#=j%ig+Tn|vOpYgG^%{Oj#@v06*pAeb z01WyF0Lb@QfofATMonnnk|2JXLtYe_kjs_NIeRYGU+p~`9L1yx zlqQz(PeBLZ`UJN$sTb4&uNw?eO6}7HwT7XAKUX0uaKLQ#J0RJk8`|Lp?6uaGSDxen zOm4er znPTV^PQB^W{qWJzeo?%bj_7+mq^A<{Bb+*+1fG^xTlg$5ed|yI`ZhCEs%Q5PRxJY_ z+0N%;mi-v9Hjf=OeeT?T^O<;;@5$(;n^UN?b8xq`kwyqS?dx5}i@zsO?WaiR;Z|Ui`=`p7O8*x70%$VcE?fY!@=`v;8JaTRg))p!>+Ss)a&p&wpss5eKV zSVk00K@t~c;^T-)5XXcEvgS}_MoF>LK0nr;(^hL5C|EE zoW-;}LH)S-5R$dODpDiMcOfo*wj(KsSU$BgWcz!(JEj;qL&D_*!DN}m5G2(zH@Vx% zHL1sAu+DfmLEN~<`o+Dj-mhr669EE6r?W}(3K18ZQQJ@1@xOblw|hjJO?+&f1KaZ< zi_RweaUfGn|NfqYj|s6uL2wum^Sef z?3NrIh$7T&D)4`HPK*%v{z;GGi48%r_!8+dLUHCMR9M4l4XoIHxmf4o zMsRxA9YZ=nfun9c*QiMfgK8#If!4>Wl%rVQ0zDJbn*V4sXxMmg7~xu^QJQbQXrh*Q zu<+Lt5}V0{o0F`F1jtzzKEGq$7C%9qi%G#%+B8La#baI+alTn)GQ@9bc3-i=uD0Rn3gSy(k+7*U3; zgQ}zgW@Ln&=1SQWG5SS3AX0k^1QR(TJ%GaZ0MIrV?;#Pi83CE$i2qSvG|y7PC`4gz zT=J4FEmB|5=ZJ8VpI{S-kKce_wW4vw7;Ogkp^%h1qZAll*QFPA1|JkH7>MQACa@k^ zPn(E!+UC5(EYo8ugDRA9CcHx(2}B%n%CboYb@Qv&XK1Mwou0zYhTat=BIv5~C9dUu zjcLwcgD9{>j_=;NJhJmo^MM+&q^GikObRj3xPqOyj~8`F`x4s33muaLs5u2Qx}){i zYNQj?7$xC9ls&xk{gu~syqyy?=G&c3`0~W7Wy$74PRN04%VG!>WkE|CRMlfWE{9ol+@awC5}ah( zc+m&#ykWu005wO1RK=)0lyO+D6kYNoyS$MDi9tt+KFLqWxraKO3=jlbKMz!e$E9;R z;Tn_k#aw6j$we?OskN{dVx3756}8Hh>pihoA7-W4M2FQuA?@&AaY@u$ZDlr{=ABB8^Vp8_V0||X!Hy2h=?=Tw4sV7I zL%mL8YXGKBmm~F?`3gPmvvw7#j(lj!xFH7S6~ND}3$qO5e9j^HLBIj0%HiO>?E;f4461 zE};F4H8ixw)3i^Yp@)tO(h37LnWNQOvwFb*P3Bz$*5EXhxdAFTKRfirkW{zk71$2c zpA-rJdxPhT7|TDr*GhD}mLlz!GOl+Uhz|vkdK0z|4gB2#>=t!VuSCy3zU#w9du*g+ z@c+1D4j%Nz!L#oE89GR{*x&KkKkI{oVGSa*2BH4opbzz7=l~4odnlS1;_?J^IeWE! zM`}@1{q)Ao^9E^qNS!=xkdERK+-e|_VeWb`y zf@HQXw;Ud7=e(0HA$8f*P;8_*fbAJz&>uWsG5lt|qg1V=cJ@V$OPzL0 zF){kTWXx;VDtReV@?ttu?_>NkdE`Y_Q|5F_#dK53SoJgc8ymtZ@tJ$G87}3d2eX;F z&Y1yhzO?6Q{qkY-e6h@n2A0?m^tP7S;ntz;+0PfV@dosvoudB6xU)ucYcC-!{K}9k zvVP||AlhdL{pUmXur>EE`9HH}9#EX~MLvr!%;=pU#XK2J-X&W5J4VNKQO63kV2Z|| z`=p?xHA0q4t(1#e$Q;Baizw;j3%P|Fy2UgYW$`6Z1^HrTMzX|!K!3@ZLdV16G* z#r|Csr-e%D!V6tqKRnj`;m;uz;zmr4l<`G>@>wC?Xy`UzO&G@&hlT84#V}vWB6lX9 z5~G+U+-E=zWEX@FK*Ip6X&`yFYnI`PR3j9;FhDDZW|9FnM1>wghlloGbbxIVB_n5T zb3@2;N9rKW*y13i>=8Y<3nHLj$5;U9mg&p*uoMWf29GS*i>^otLbNs`4a7K<0c#Nc z{s?J^3KF2dYkhG@X!AETWk5K+nZcb59)Z<2} z{U)Q$N*3b|P!5dY10qI(C~!r(|0@1z-=bXHk}wuk=Y}X>fq($CRCS0t5^(<45@tjE zOKz2xdNX8h?T0XjB0U5T(KplpB6tw}3)zL}lA_ao+Y{WN?ils?yN6W{YSC%WjM~m^ zSilxE?vvXE|5&)$A+BwX&J*WYZXN_)0~Y6SIO6x2gjSW&HwVZA!#E)705n5uNcV1@ z_>Zp~HhYPC2TY6GVgp4JcxZ>30CXQ3ty_PsACPDugDovX291wj(3H(B_HyIs!xgSu zZLCEagiPw#QFD`KXW^^tydbl&5{)@k=ufdBu+*&OMpr@(bO@OUX7f1NQdae-oLnJ~aqo zb6HRawnMxHq27#;%XhO70NV}iPr{5Jr&^GoEfC`JS;$cGKZ2hqOE!POt6yQl&c2Xk zJtiD&OW&0%5rrSe@7Di{rCiHj>om^s5C4?M7rN2CG+5wzcdVURK6n|n1oR2^b%yx= z5+uBYklORIE!_~IK$7b3{ex((FKuoi`ZpEpw`{kv8${|!&CoQbH=WretqWvGIY@9a zRES%~XN~?hH|Y}S{6vJ7qtl0-L?p0Ja{yV{2t<>5k`eq_0BQrfL{jWXQ~eQ!G~YE) zI!)%9Xxs8VwBdvqexvkffk20FD4cnH6kMX{GO}(G3@QbZSi*}k6^is8X7ET!zS#E8 zk??f-s}rj4T?$!ZK|cas7e-)Z0G?nGpugS)kaKu2-%91c)6!kh^_>x=Ik%D*O}{GPpNXsYJ{Ln`%?}r)3lK>T<#15WHO8`8@7dF&e+5TBI(p>i@^ofKZ zbBHOy)hKO=G(JHQTp^y$hlSER=?uW%;9c`46G8!(i+M*ouw6eGqyx!{yI|7)qEISL z`!h@)u+VBe5C+6vr8eQWk;THNk%?JHmP*?XMWD%JfFShAG#P*f%*gUV1XS~#Vu9X~ zS}u>pN5x0NiotLZkFR3|`n3woY=2gm9);NFqvfm~&cF_h%q#(b~~K0_-G`#H-yerPzcO4$h0Nqmd*$!1m*rZ_@Qc2zp7};XLcsSbREO zZlMa>c#)rpv9Fda=d%2$$)u$WI!1dqPDN4*xpj%cb)*=J&b&QU$C@n$J*RqXK3hqm z+?Kk6|9s=a!hXXG(wes(j*r%VDqLS+1Qt$J0MVoBR$pZzo2@1YBClC5GX}O^O_6m3 zl{-n<#ai=K3cLUK#tNtk&+P$HVN%cbge`HE zkyGO{ZJ*abgHC7d6l)e5Gc<6%pL34q=h*zrq!}a8ROVPVma53yQ|mZFzTs1qMqClh{H8XZ;{h4>0)H2Jv#92c!a_Z2V;J)|WG;z8SY8fJ8CScSm z0E>wx6wd`TdKs@w3p$KCh_%rMJVV7UH>u_CKzTvVXmQnRK}Jx!gUQ$k}p(L5EuN=70J{anJEd zxDebXkppMd7Ur!+17WV;#$$`RtOoo3x(cxTK5zOqDDj7eX}tOk#a8Z0@6hrSkhy}IpzkK)sksSh|j0r%4c znIc1zqulHslsd!Y!`DO4U-y%GKB6=W-WNMfnZ}tpmFh0fcpL(v6QUvQ`W7K~Xkx~% z+^f5+EhxS^tOo(&oJm~u7CP2L-nGR;>8AiU^nwJ+wv3)^t^xb0e~lht!B2u<23G{(YT;bwfr_WfF(rSX#2$h;vPJLQEpU z;*ZE`b(*gSN3F8!oHAoPW<}gP1m1h}$;=4OM7Z$uwZVKKW#HtVnKAFlW-;PjRiRgQ zg)>x&kQdu7vj~_8_Y*gwYA6y>c5P5n6n)xxO_^H2oQ!%#>3=Cz0}|mYBpFKNH5l#Z z9Mumg-_>|A*Zlq}e?ebZg4~vj-F9@ZVg2EPZgzOw zI1FG&R8w#7DYL#C=qiaLZ+~CskQTM>fy~O>q*Ko4fG7Cl_?CSAG}hz=SlFr-G){i@m|nENEt?+`jU@aT|XBax5SalVSK z!9N5bN>nFhJk&3$T&=5|8TK@3>zQe;(X}*S8DTWIWKUES{uhMR`W9H6;yl0+U9uey ze;vv_dfuxjBrkW~f4!ypuedCGb9D!4;GpobY)o_ol?7)`ppS$zCctWG9WjqWeohH&W^-fo%O zu;d|hyxUPi86i|Vf5F#1fM4xHKLLYF1Y2kSP;F}r=dnG$Q`h~7DDPRz`!oOHKF#CK zEG~^K%HE=;&v8NsoTDl~2vElpZ_Hp;ispX@XgV&S$$U^$Bd9oL~Q#H(G zr(}1aMRE+<~em3|6=}lMizljG#}K#rsX=ivtRk-N~yoo>@Db zNTmshr?&I}W;h{};jjti4I)9KiqWY;#%dis5HW^eUb4jvMd+wUk^m?%+9$O*7uni2 zjhmj!CqskUvoId9HZgkgYwbTf#50>PIy@5Uk5A{)RF$!7Kc-u;FXkmFRy-L54jx51 z@v#Z5ki5jwP!zpId=Wm>HmQiPPHoTapRJLzAJ^ooAJ3>c!@7d2PP^x}N4FZ3Nz8E| z%LyW3P8<;hWHlspTA07oZ``2TzuF0l9|`%G|Rtlq%b*! zRVrhh$VbG>wxN)L^>Ra9LG-qgHiwn{^{MJBt!(;HwR$m*oZfhOPGe@Y3IoM#bgm#hb<2)N(auPi2Ylk3ODyK2lh~mCMPVu4vGY@&8 zVf@*07SW|Mzn4Hp6KPhy!J?L;4py!@nSE9pPUEX^Q>{!Yk^G!%*2Jn!HU09dmIP)t zZZxUY*KwCxUVX-H5lZqS`-Xr&ca3&X?0G{R5&_VTy5ZjuqXG@1Tu@?@)o~$wF-|Yh z^h2Ez`5Fx*V3iNz&jl`doY_Q;uOI?Tny?KwTXh|(tNorDJryJ%3-Yc#J=JyEkx%+- zspBk{I&8kl9}Lo{7VFPr8XT%|auk9aUzRBG-lc?YBjuO65*2Mtjgxg(Y4usT3R2uL zfyQl%U7aU$m-gNDYnOs4W#&XbycPZ!?mTq59LP6`kxEgGhXnQk8Cr|SKA3+DZ27?X zCeht;MionCy|6}_FU%*|Y@G6Ttc69(UElW`S+e-Rb_IWUsQQ80Un6>6{42GWcpKI7 zCW)FI#;+MERTuGAM~Mjd^=ZsajVvjKzOiTr$rjBJayfLcL%pQTfjO(Q`nrvq^G2Me z@&pxmdArQKip_2Y=B~d`X8n}A24&k!T|}#84omI6Tk&+yFRRPpq-v$;e{sWS^Gai? zY3Nat&gFa83n5j0xojOLYJaX%;RS1U%zfq|E4J64_ADk4_xJ9)+3}w>*9S~A2#qu zA>G=j;Jw6jpaQ8EIWsm=w9j6yMfIGgM}1HJ-SU+FU6|~8DfU3i>!pKld>#X{3@Gt0 zR{fLF`AY9o^_Dj1YLAm{wfV=|o3@Yt@uz2fI8)Y3{Yzlb+;%(gIgy%|c+1c9(~Gae z++TY?dzo)~aI!UD?no`rr8_rJb2y4`x}6Gse#E|A-=3j-x7*gj#+VZ(Xd#vmF>E>A zCM44!96n(0$s78l_U1Cv>%-B4)BfqHW#2~EhqiTnySuu}-@`Y5PI231M-nu-oB9av zhCkR1@d|%mc9KDTbkWMs^0E^B{<$U?*Rmi9@$dO4SaWRUqt$3M>h|iFPuFeh_^#rM z@;F6xAVma-f~z&#|G8GVZbF?NY{YHj*_Ibk+fIIqsdBI4gT)>>KM_hrAx*g+N#6st ziD#q{F)_KU38`m}=khVpye^@KE}5-v(7B(dLWNXk zN1WMWkS8iL4&5gTuu9B7mhhAbDvf{LBjDe8i&N^2QF;X~|Bd~2IP&{24R)Z3jUNTJ z9GRLO;UN;gT|C)w_{FCz;yc~=my7a1A8fD=0AB|{K%dgi>T5M@P8Mh5CQqm#K(M&W z;f0P>2qOblq)00hcb6wK3K9=`6Lw!=V?xsQDbn!0aoW+CodVu&UaQd>LkbwkFCaXT zK&(aj1ZM_7Se=|m24^yysM{KhZw)wLPXk+{l`lZGf{0o6?NyyqAJ8Zd%CIAEIA!z` z7FZ$q2}D2TC^qX@x_7bngFxsUX$0Olv1FA0BFJaoXMSU*w~LSg+7Mo^3kg#Sw(6x< zqx2yQpMsgLaqztX+WeUZ>i}Z}LOe1BRK?`qMNY(#J`ZFcMhcvl$wN=V7IVekUgLlu zY2bMfo;MD;4gixr6LS|}Y@MXM5Pdqy;3-G?R-FF+Dn+|jAQMKgaKjev>v|GL;&&@9 zGnM0HomqR8g-MUV@CGB>5NTM6YYc!*0MF}s{1?pl@1wBvH%O1-bAV+zd%w|2Pg2b# z(zLxZGkgMSdPBU2lhCg-f#rn{u!13akwPSXRVV&?ARduC#Hx;%2AiJA2gf!TFa!q! z`STBg@|Eim!FegEZQ1D8`MQrB9gl>I2&||P0w9D%bD;<&K-@Y6vU34|QP$vJS2)9u z01OyF6Q&&znf8N2dwM+$h(=>zL%4ax27?jIGLl}>r^5hPon<7?6fax7F)+N#Fy=wO z`1663X_{9!QyXGBBM6Uu1Ui&Ya7qY<#IA(lTwW0#0*dbHO1=x|r`tLIx&kYDqm^Fa zz(Ki~XCNt(^7VwIv$~}E)69E0x}amizyfB}20^Q8R$)YmgepX;9>Hd&$YBO>+QoLd z0_(a0l&sTUACgPRmzVDs^Dtx;XJSpwmr64gkYQRKW6BxBKwuq^+h$I*FvK~S=v5iz zrD3gHMSNj`UNkfv^;k+RgmD2#FlQHDH-&nsNgdNCm0n@jqK4|W1Ie)1VLNvTA1iA) zn6tL{2{y~>4|M9fopVZY;)EHnWmSSq;@NJ5CBEzP%4ATtsD=FO4iF9@(&+TCV-Mjv zl885fu&n`vw%GB*cht~z*piGa^n->NAaOn*Ef`8cdq~PQ4-zp5k$D@bK=@-`6u7+Y({&_puMIlf zhRM*5A=l16iDWm=VK*m^T5VTaAdahP=Sn6He@~o`x+NYMmTp3+%UiK*LgLWy)HU#Dd!+0SU2 zu~Wi^-7&A#gceku$+%m6VjN!E1l^J|*+gHtLgovh+`}$;Bx(`3Uhu6FKzHY-S}u1E zQjKyYK{sfLVG_e|>L)aK`n`l1ed0`F_J>;%3m>J0)zq4Kup@n63BoP|+GQN9_9)aP zX4ds$pO|*R2<}fvZVig1OvA7SS^JFOd zSl8v2H^zrxzfduk8Z|-ggxr+cWI02+tEAE?AC!@RV{hC%{%lR_G4|I3Y%4rIQog2c z{iiYgY~}s)fdmY1%4+BV#`9%Blqv`xV7dOsmLf%1Mt3y3si!Tp6Y!eQt`|V~3GBo_ z#^?ed0!%tVHEJ#dHOj#PE=<}l>j|M+4k$_5Nwo zY4c^XZls37;AoZ7-Fw9o_!trI_85En=(~lSFND`49u$mQP3pGq)-q+=gy_0_QW5F)0TII<&R`1BsICWpGl_ij*q<9O z!MPRRziz)8iKjXPY6FS5@Hv7lAi+C0ouZItGMg^Zb)a?M+mf|73s-XJP>Kbl1*siZ zVpk)zCgQV3{#chTN&6PfjzCN$f_hW4d3v0^+^DUi_OE%vPDg9i+)KxeKhxIjG*Uit zjzBc1AjXztIeF7;akcMO%Vlt%2eGKd6qitJHnHt6`IpfWVj11IrI2M`Ic;pn_Mf%w zzkczbMYigFx_VH5U$3VUH4jNsYrxQleQcZ}lu5seXJl@N)_!E9mmtw~B-7-76?`$* ze#2G&!hLGjP5=GaT)9Q><6EM;CWcSN>{u;@Kmec%u?7AJ?YFQy8X$im=GEQ#6`DrG z7TBZZMn>@YLkWo{L)RXLO!>CG&4tSj<=1_3_kD7x#lNy0=l)zn9(ZkMm@3`@nZVvh zkAC#wo;xG`m;Z!mzYgr7?UxK}r_UY^7#vbuZ})OKNw&+k;vcCG?__m`SO$J2Y|%td zIf4r8;cAwX{R<^$|K2mgI&Lu=7`feDqEJ;G$kOs%J#t^CIQwH;(XPkwFS&!c6)TGt zh4X=furP=WN~@l|M9rgHKmIeskz4fV4x#tgUG=Yr`JPtucPDz8F$?&gn1=T?#!r8h zA5pw*sPXvZ z&MDjquZv0MI_o()nG9o3&qU6%@*&;!ZvS3fRbP5;Ej#^PU=XM@#&KEsp*&PMSGuD@ zQ~u*hw1d8J>lI1o5aUmbJ;etZrQ+F zD!lY7hkPr9qOJrwR~FJpa!*>RtP(+iBzK?V9>lgq-i|U8UVPi6@-0dR?z&0;amQ$TgovYR=tBY2-*hD}({-IAM z?J?hSc~U^`$@Mpxi6BfWkxID1pGCDA!BJ}YVg9S_YO_V<*!C66coI3vL%k{BLziqg z3Zw`y<=d2_kGS}orF`XS85gIt7}rwN_3dM8v(;dl(NJBNg|<{P-pUO;eDcCqi+Sx> z&RRu9=nnt7Jm~ss zH1?;puSDd&EWhAC+i7@io9u(+sT{?JGlf$pDm(_h zo$Sr%zf;$STd?HH#`Q1>nd*RxIOZ0XNdQ95s?|#i%MW8ldNPqbUFM&j(D}DpDYwo;qGk{)iB;bz_6|Xv3_e_bklnKBW3`t+CYny~j%5 zrSTUl>CTT&kJB7otyYBr=l7K=uf1(XqR_W2UU}^f3?i8PvIY4)Y_cyBi98AD#EZDbMqIb<#$tMt#2a z>Nxz;_Z@G~?bg}yTZL}~WbY$*DnSsGcFSgu>pIXL7d|o|y?)SZ{4**E8DqCr(IPT)IQgL{nF9Gk+7k9Cz7OP*~(!i!|7|pcf%7K4PKgBYOkh@ zprW^f*vC8ZY!(Ug;Z!(5w2|l%DqDQq70Y_^c$pM?Z;HzEE0uZmOb&x4SPIr7uIkxf zJPOLEP%~Bx7uz>PU-L?{Uojd~!T32T>9d4!Ku7ZF0m0T$4tt>T99heWtr*wU;Q0*7 zBni7XSEzqYNkpL$BH_jYyUKysQ*|pTQs&M<&W4xcG)_p+m`UE{U#wjvwM*(J&wxE7rBpB_2~B~KSYxPkz_;h-3|+L7*g%gxynvW%`wtBaEw(sMf6BL zRzjDpHelY@QQhuewq$ng0KvUOTnp)0bM-t+5f41{(KcQ7U%vT|-iQ3;AC^Cpq=+F{ zG#?P}52BV=KAO9%yz76&;J(sp!ZS#MwY*`0jbMcz@sEUOFh-0K;VNE_p{u^}F2hMt z_mZ!`=f0VwE_b_`Vy|s#mu0{~Ho^Elyjd;NT@WIUIn#jG$K_gRpQ{-!%U3!O4(V{EcU&u>{!64Av1H|WYGyXPk!IjQBLrqASS>w ziow_2=_8xfZen-OpwU}PL*ScGT=NL`KhYw|nY;d6bw*K1WE(!jipA#EhKfHoJ>N=| zevcP>@cgczbPyP@}U8>>&NnnT1C` zR8<~7(|2!LXk+XBK5Bv@2=!$ zlAlK6-D0IG!O&EoV$A z5*L{}1_r9Tt5=&*wkY|p1xrx?d`pj4L2MK?X(_L(%1^x8@$CWV5EuX`L8kgi5x|z? zZ+}GLMDZeuswyEl%YnSKnvC^HMY_E`Sz)0&)dL2_T3}z7$K@QfOUusn#$n3ikDte? zP^Wi#esNPZtjWR17%9=VA-1$sWz>F_L$ZIcl}#z}gqTl2moR??5+u-6R3P4*muGfm zXhaErD&?Hr#S!SE(V$(fSKmLZ$FSA=63gpPdxQo2ifN3sI+d`gP-CrXMsp?2oUDNc zAjhrM7=KWx#qdo$`nqQDON|?zuI1d|_DGga5-LzR}!R6}xe%9d|98I{(O%hhJOW*urF- zoXIs%^pyhcT20Q*afO4~<_c-h>)3%v4I@CEP2QPCAg}-X*8|6eKSD4oz$ma&PH584 zklP-LzfZqfn3>>~PUsN;u^3_uEu`8c5?c@vHB3M@6fE%l9k#y7Fy%%RiSVoqUg+=9 zK>0aET0bhZl=ZO_rM5DR=R8ED$q*``Eyd|SPJNIppL((d<~S?$pY?I>%C=zrS4ALx z&Mx*x3hwY(dpa>iFCC*v8De>^Z4QVY^D^~EdKN@ikocr9#%;LWRODsk^~nocGlG6u^L3H z38j1vkOLSwx+w&=UO9dLLLEA`t5;>&&iFAu{}h~7rDjHEAPQcI z6>aHJWP}CjByCy8ilCrMHR$`=vrSnC9!<9*-Q${e(`>e+jCXx7=F;$VTSfqX(nupA z<;z*}6}y+u;UskPK%3Wj=cxe~{4wVb?^qNk%?NkV(Xk@vKgsTzz?cDOYC3|(s3v3W zgVoT!_8bfN&fbr1r+*>z43+V+>%(sqvEGqv?J5h(V8`fAmbM(IgcV@~3sxGQ->LL7 z&{}>cJqZZFJWTDfMP}nE+L?qhh46Bl;aSB=TBRT;J&0=vp5#Na`&9 z-;L3JCh?!_*|M*r?Ghjub+Aq#GeL$ann>u(NKjdzs67jx5Up%D=KdOc1pONl~+;-4Jh6?!7AJVx+|om`zKe}-srJ{7(XL&6briwUJ@ zg^9DD1Fp022RK*9sE-{MF=l+C>m>A7xtRKG9H>p(bu|UB%2H2}f?>mp(W9;Zkc0=< z=rc-%Po{z*`34!C+#Qoh^b$pD-~w@JsIoL`3`gj7EQ@}em$*uWxRdpbUz;-YiJko! z9`~F8Daz!n)8+YkjjLnG_lwcCNY57g2G`z))8tx^yF>gcpDFm0EpbK$=L6B7o(y+E zL=o6c{@#|Z!k)PvVq*f~sE2m4KzKL+!B|=7&@66MD_}-E2ddr%c29#VmxV3>mZi)F zqGc;k@{{UARIs6Bju6#gh&qQ2c@1Rm+FD-~L!T0=c?x+5$h_PmB$>z&TnJ=mwdYs% z4ylX(#|~hPAwJC%lDM@TixdpL<|RF)z41kGi#v97xll zQf1(-ftWkKdHayBFBU-pEy%9-6cVEGTX^#&GekF@D3>?|eUi3}(r&6Fac9v_GbcYp zv=IH8v14GVNn9b7`a8wiMfoZAiZK;ik5qo;zt`^pOSU#!Wb8o zVM=3)w)}|p;>qCBx%%R{-clGmm)}l*fwJth-f7vTWRV?G5nS@y&=k5Bwj)u(;8W6a z`VQ&x?x?`1PoRt*7hVf`|GFzRglln@QoR z36)yex9b9$vvOXw@>83_t4AZbkP2q>3Lt+urkZgklPFdK`%>>aDb8Gnf+|_6(ih-z z{;s6ijZ&A5cdwF4tqOBt^W5{Mm53u7QfQnqRrR4UB`n00Ws^`*QcyXmDwG}KMky*g zQpG07wv<>Tb5?y&KqXIATbNKS-B)4WPguBgQ`@2l&QvbjD#kwD6`=-tw)zaJ%9Os> zQVrGCl9jwm^i5`FlWM^Ox_K|Uvucz4XnSjl)(wVCU$Mwj_;CsbJT=j%)F&_T_C=BJ zNw(~pwcHN{(T|r?^|shajk zL-ILhvN52sU*`ZBywo#tkQMy@IT=`+hIR{6YoKc6CpyQWj`nzI z)5g@;Mn2Qd?uJy-P2uMvbEP7lMM-vu7jN+&t{FO| zySuwV=|)KbDd|vT_WS<-_g&ZC`)r*(2kT(1r|uiu3V<@yj{LdvH+)76d5c+KoxmEHK>V>)HSxK;XZXzx`iJFf{nT zmf$kiM7bOEoaDBh(Uw1D5|*XT~Oz(y||eJ}KbybjfC^c_>u|7+sSpY25{>h<0p z(D$aTG$&aXx?;$t+^k9(8tM+M<+fixP#Ao0cI;13GyG66vsz$ z07+j&OPP%+%&|~^98>8VQ`;G{fsSgBkL$d)3W$#zsPXE59Pb|?$}Mj*`G_0%IBugx za{wE+RqNsSIN{Pa;kGm3@jBr}KItPq`N3?`@8cv{A3G&8-v3*A`QI_W+y5Q&1Arh} zR>R>yFya3)^?3E3L9t_>3Az3!|ECsu`)in5ri{RZ2GcWcv7gE!{VkF{4J(pB$Rl2g znr^L}|AHIZEUt`SzE~pn8L?mKL)mhr)|Zm>p0=9RT5-j#_dzMO>s98Z4P!G4UpAWH z3pKtyt1O$HUWc6gFRFFldp&;EjLoilj5Y_o!#I9kaoZh;z+tNp?P|7#Cg40>=%-NI z$6P)tD7<#Ho(%e9$-0Y){#Fg85K?srxNAKh(>zzmz{(J`jqWEuzFPn7BZZJ5u;nNE z(e5U6)RN8L&9KdBoPrI7%LwP;_by2P&ivb_Cci%LUv)UqP+rUP-C((?-!sL>rzg+S z+=AaLf&c!}l_oqHdWOPpK$}8K@1NaKcEb(b6}BTPq}xnA|4AEJL=o`_=NV0|e%_0Q z2pATH!9i8v(35|P2cf{?O!HtH@6!@Z_6}ySCqs#$Z79;uV+>iIndgHP%IV3q7`bC6 zXq?iY&RuzGLPo;yP6QSP-Da@k&(GTCb`}!H%BS|9iKFA}BXu*UEmE}1^~_;#7G(=L zrwZree8&m24OyXF#UG)j#}#xbex%ALMYZ9C*8e;9VOE4>okD-pJLq>!s zy+u~O!Ulu%;Oa;!5YfmeFS}Aj)Tt;3d4n_Agcbw#{uea0>! z=9lj40kmu1osqc4aq7RKaDp5RBG%*1Ng8^^7pwU~&1hy3RQ6X1#1u$k> z5gWqSRzS`TIe)N{5Wm8buwog$oF*5(uYnp>7haDfp5Mxv5{>zaV%2B$R6>>+m(^}= z^GojHPW%-?X-|LP(2D+d*(aY4)Kno^6;$|wiKcfgsaq2eNh@{#vzT0fV**!H7a&D3 z&fp6S?~#W~^OJ#8 zw)cN*aqLZhbg?v=T?H*@y5i}zagD#*eZBjpJJHAWPwjUVeTxo`cvSzr!r^234O583 z9G6^l6A!6@7F)lN*xPebOE;euKi$h)?$gO2<8;EK?0g{nTodr`ZXi1QGyl{d6_e<) z4DaqT=ie^{m5jp0=s~}9xe<)Ke3nzK)PG0dvlk+DI${ub)5w3KwLp(?lwP1uH1UV# z=DZ?Ey>4EI0PWH>g_XF}c{D>IeZ{SCFPQ?|f~nZXC90vTa$;tC=T9G6P?BVfO7_59 z7HEM=PwbHtlpfx3m@K5BI;Z01SSeBKX{#ji&#Y_)O#%KUn?+8{gQ63jQcicMIPV$8 z$a_+>pV_AHQEe0fA`a>0p*w%2ehM!Le$eI#^ zK~C8susRATQe4-DGy>M}N(k%8tfp0{KuI2l78W1UAjFN4L1XFOHjr@v;jwbm5!v@3 z2rtcuS3a<8Bz`XRmp z-$vLI>jL`?p{?3~?gQc^{{Q^2yG5WRX7F>L&6a^0_C&ad7{KVg9 zQIC~DuhD}^3he4xFgG)OYmV+!bHSDF2Kmv;R1BdabHnz(`MTH|2mG;C;5k!_IEu=p z)Mpl)_2E+vZ`Dnowj;??RIwC}r~Up{T@1jZYei2>0*n163T>Wj;iye_C8*Nq?VPt) zXKpgoqZvI?Vv1trFIHABklhGJwNm>)CgeP+Tb`R3pNmCn_NmTkd_A>^^`Uf@m91Dq zRN25Uz8t(rFP3JV;s>XWpx2j|)AMMd4K*dncC>DAjPM11I~r#~%cJT9#JuH&fm zFU7z=PHsaciUJzI!XvPcqtn&ex@g5nQ%Ux9(W13&Rs7NxtIEVvW7;3SlR;o=D&85b z2)M9fqK57F2FdAFd}KJ4nRpiWS?UyD-OiB;&D9OFV9XrF4GS(mGI}sEYQZY5vrnVJ zg3w1q33~&BVbLf zpS>Er!msoy#08VanC!f%Fr2L7kT$`#y3kHI(Es=Y<+#1EX*4hTJ|n^8NXMr@ z{1BLhR8I}>6M8au%W|ij8=cC9IQb(L2!Mc8?NQ!5?& zq+-@5gnDRUfpQ?MXQfobH?-^F%wYb{D$>B+{0Q)z|AXtsrL^pBQvZeByy!(+;X_mF z_e(27vHBqTEaZV%zk@&>>d0k#UAUqyX*J7Jb3Ydoml#~`#`O3yqU4YH|0kp zCr{K>7Qi*kgs_5v@ho*e6qa`o22zrQ`8qt6>JxV{N*Yp_oq*L&d|xNulR|+Pqed+) zA3g~2t3tu>BZ-&{AEFNqD+-!+ZB7EQ9#0vRq{}?hrR*3`vguN9$~zaN|Ge0f?`cpY zkQTKtvm6Lw-+N|y@hlTlP07VTAz1WGs31}K5>il92sSi?Owo-OStgk`6ztrB(+Y z1rHDNoJYH-hI2*vQ+ao|84X9!qDbD_M?QQewMlm@BbMmZ@oZSD#!N{_^T5__phgO>qICWh2YI8wmRb=X0LH!UEPm@|w9 zf7&2ywS>sSz$S^-*ueiiB8$4I73B67=duWGS3v(9AMK4g>gb6J8f#-Q38Hu6wDpdY zdnPOl@JtkRmEvauGJ3l6I}4+l3@u`xR2gETp)P7+7#M=I9a9Bt6&Rdt$>2i%3aEgN z)C9zO<3$KSWDOGmq1`Ccg7PzItU7E2!kROXCNaHjFB?Gxgk($zT(s70&#%+M zc(lUNB*BD$pes%eXN>944G>XIhCDGx&>|FC;W>&2i_gr96Da9WkUd(kAEL%OT0l3o z#$o%FD=@BuSQAp@lO`1C`gNli3IiEAJF6n3Be+P7Ijc)oJA|(1YoYGA%=O-Hs)wDoMP3 z#B)gVmABjnh~-c#%aDIDzwc&$qmLz=@eom=2#BNLFD@UI5#9<+^1KoJU{~=SjRsx8 z)!Ie6Bg}ew13Wma$FEy1E*ce1nhkOw`ZN=8fA?NqsPV6N*|@QVLVC%2r8KM?eX(~6 zeHGs8Ie`4`0&)@>g{-DZCCQmI+I=~BP=Zx(2Y=e1WSK7fzDyF@(%Rl20y)vTArspW zU=f`vL{0y5l`aTBNV~@gKbs}V2BFnQrrvKUTor!vp`*!Ngv?juTujUrM*^b_Kf4uU zUz)s1pJ0+o!W2h2ICI;;dk(JwhbJva7sTP$uAj&6?A!`!u2}X?2`28)o1n&66yz$4 zvBhYlY38$2^dooj^T6nIcyHnsRGMv#+nut_(q zcGbF^<1ufy$ng;+b&80+4>r*L^fRkFH<%!86ECV4+_4NQt_2sc0ZJjwm8zg>Q&7ip zPn{~L43@cB)`G3w)1!zyTuIc?+ugU_JMb4*m%3Ni#{aSfZ=x0tw5SCv?n4~vYbFj@ zV2da%4c6tsN6g0MC>9HfB#mD0M|=i{!%d5q#fWX3_|yKf>n~@WnHIXr_=uKAYl%^6 zBg;DVW_iompR4c~uq6kycr;9L5Q#A))n+3zr3zQlR(DlP25RsgdIbOJtu50PjA+7t zjlfX=-MKnTZW>lrRBm`TsQVAVFa)ENl<0XO%$Se?J6f0kt`kC14lzUj8QAODaEXn` z|Bl~QdmwT++VPKW**!l~)Zm=`Ag@`WJKBgUo+n` z3>f1Yg-I6|d|P+-Jk+#E`|g!cki{skL`TLxBF`n}I$IYoDvRJo5ar(=QXG+_Z4o;? zaygT+w3RdKQtsA5F534+xNiUx?(Gdx^RXzw*Wn%uGaj2nVE^cnA~%<1vf{C|H3rw2 zbUZ9Ij_415twlR$7HJ+gzbHlw12{~0+3b*5JQL$omsXHizHBxdFi%ND@j9v~_vaJ$ z&XE)hs~TQRzMGmYVRx*51sS@9jL7JD4UbCZSK70T$$p-GR3!`T`x;A5xhra9M@}@o zM2r&eB5y)h)i>9gKc{dy^G!5USx?bxj{M?+E(tmjN8mZ0HXE^%M>eg^*u!!7ks#D| zBG7EkaKcMrWrV$BC@Sik(J^U^S&HWkzyIjeRc%r#`}9lR^warQ!b|E$W9e(u88n@5 zDc*CHYQ5-Y2G`;V)M9Z7@R$W3p9%eDJ9Jf8;pzPL%LMKVHA-efx5_s&Tk0}3!a3Ku zip1Efk4c@t6(&EEbmc|s67ssTydOOb?se2vOJ9GJGUWR3+T}D9MJ(Mw7nCW8jvUiR zor`K;OGkHdc7?bgLQ|kV&_6weK5I>!S~ktJVY#m1YvPGGq;aHrBH{i?J`MwoPem6k zxH)Kxf@Xw4L8Cj+(eAP&yD0@E8l;GN(u_nm>jG9F0#MAh@av!CPtj|-iPV#ra`nwf z)pf(BBK3O4t3~t)f##s#%Lx7h6&z(tT{=ZljQrNES#rIYXBO|n-VHmR`B3^g4T<&G zzwzNEl39EcDTpyY;7Qc}@4g#(hrnF;KDP+0x(*0<3;iPGkN0N^ z|06bU4e0nVkdn6x-~>y9_9CFU7}A0vf59hxdq~53+&oG^gnh)lJv>e@ZpeO(OxKsc zBvHM+RMH{T8od?_;Iv=+*wP`TWCvPnjo3UroNqmJ8hgBwhy8y)c=(6jcZLX_$@0c+ zE;$?hzS$j_Yn0&}!5;QCb2}8Y!*LxsTH%mfGN_Fj8ReI>8tq^Gq0;vEyzb1~b$GZ9 zZ2ohsMD-q>EkgywV=joU5Zl0U{2NLl`kgWK9~7h`Y23zlHq6bkY$%UKG+BGB-w&F@ z*t7ewab;Ua4o5o8%iMO?5D?wZ;=MUJ_}XjrQeO$AjxB8X5U1srypaO_^>w@N|H4U1 zJ&-*H{3b9zVEDF#RiZ%w=eb2_X)lfV=B!U%NXTA*z8~lRj*q<>>S2N<%RC^~+u`n; zq7x(iv#7oA&;ole01owKe+WS3`rO9ils!o*hqE93B{LHVn47wte1q{{Zi&`q|G+uQ zHIgCUuLHwh`4luJ8nL0rIJ4MSSNKRb?Bm`y*p{hR7Xw$zApmtR^y=igT1jWQ7^RdY zY`5w3KiFA^Z=^Ce|0S9H3IQTp0dTmlQ5FCOsn<}l6O;P+u@LcEWPJ$4&{ISWd->!B zt_^Bj04)8uLHzjTH0JhdqZtbPIdL2NAd10x>?^*Dm>b&*iFi_i#faORRx?}xUNyl-^BTq$TFAbMdWHQvx_-VRIFi1)is z+6oFg#IBD6oWlX+HQz~o zd*NhR0W`1C8$mCf*Jv}L2q!6|k(7JEQjJ*iiE1zn2siGk^)r`!Z;}TA!23KsT$QNM ztjlK@^4SoP{Lg!0?E|@Z0uGC@Z0$pZWTI`f>-AcT!5~`YTyo zZyN~ljOvyAkRd=~+NO5ZLNGNFK1a|rB)l96UEGOz&v2nwzusb8bmhxnkePDOWiIZ; zM!F@J4w4A=!dCkbAPE;TZ#`J&%eB?wfBu8RFd$U^)zR?UYJ^;OjkKG$S$X^)sZraH<@;eSn*>)<6ZU{mnKP$cdN21MG z-14l!|MGFL6eX-M!fnFMR-^N-A4sJvmW2~_V>U?HM=VVyU zwM>eb;9L`EVrz`ly!g|}8sDe#DSCDaHygtC!mt2W7!&)k8e2*ZJyi{ylcgCRdj##b z$pMNJl#sHyp>$eSXw%h`2oeG1RU)aT5eZX27^hfcsHx3uNl@i~@IV>Z!y?7nbY_eQ z$uiYAji$x+LGwRB)TgfRSp$*4^tc&uYSL)zo0Rwh{0wW)S|2R~%?aC`YGSF2oEvHB z6w>$5s{+d-{aDgT`9`=1kLObj_`&O$l_U5e?pmgCg^xfjAo@;Zs-hPvTYL;e+5?qS zLT>(SYQ03HLWNPG&3t=cDV0I17(KE#Og7Ui0Y&9Br9eCtU*C`K?>KV8hciPKl%L}) zkMDR8tq>7nAz`r#5t6T{nM%TOv*#ZjX=UK=OK}cQL1^f$$^IWel%+oLhN-QLHR+@R z&iv8jaSH`;#U*jfOMMEQuw=8SAB{~D*;DeW49j{a;R=EacN;piWOx+mvtNj00p z6w+0S;e(3CyYyP~1YE-&MnYN*jykv%74L?E`dZ`FCSpTCkn!+}2d!UvtfMR&pdwC? zwu0g^KfA(21lrL~h~O@7L$+O~7VU@F`+2GSdPOtSOmN-@>CU}<@ykuS}^&JqK=3|23@?;hm|RHw04n?I4A$!w_8{-*J0(f0~#aoWaYN)nBa+>~%AM zp}1Kdn9q+$mfdHh;N3ApI*>1FhvBt|RdC6ZtoucD>4N$~Ypk-KQ5U$(nAzmzC*m%5 zJTp&xi&CB7epX0rO&)-KgL;U0{%x*x0axSU)n7grY;>Kaqac{CQUKw@qRp4&19eY%2xr(NVnNo ztMa#F0WP}b&qux*@Dx;SZ1FwO%h=Asxx;teOy%Zw?d!^8n`DeFRBHb73brC(hG8+Su&n1H z2>H0ZIxR@k5WcYE_Qzwpj^+~HGQ{xH@bf# zDtWXlMMyI&rt5$}O%9s!5~fWJ57Ta&>~8N}P^o$MwlFJtNY+_7D0)RYb(dS7hzEeZ zYhGRu?eM~gZAH({!+100LA+^D+in&d?M7=OvfLEGNyE;rLMj#OEgbGoa*|Lng~Xfk zqmt8J&BpGx_B0HM#;0QJ+TtY=izQ5fP!ii2_Osi?z6kyu7zxHqIOTH8sg7T0rNt0m z4@%bR9galf#!GLhh>&4DADF!ze4`@;B*uYk)9H64O1(sR-!;2>kG!!+e1i$fsYsyX z#nGF|q>4wji!XCaZ8lykHpb5Ki6ypU#`P594&bYH67IxGi@{>EhqF{$P%w$^ehK$k z$pu>_zXixCzJkCnsVZ)A7b9GW+_*Ps@mIFPZ(*o}X!50?ao15)(I6R?xxw-?VbwVH z0T&#Db5uiqTof`~T(;3M#76UruG5)%3w)e!z3AW4#`m&0ZPH^c$uXZ%QUAXzo-4rR z|83F(K@crKY6J=zBm4h~@7%gTv|(|WkyiiHq?grr^=&MYKw!{C7glbvn~mVqj?bUI z*_F)${u}O7{%?~WLU*vDtOr>vRY;?Kh~YaOD}5_bDNxa`S*+G*b@|S^wA&c;HWZJg z>Q}efV(07Q>ojj^+T{5imPGc%KHDB}cd+XGUpUpjS42YcXHSXY7%aTC*@ec#v37jb ztE_#nV>V4K+FB5!SW@1(89$jlY!YaC2cd?Utc2C#$Mn~Y8-&?yLV*lCH`kVW` z23~9(@?8bzH2wSY%s~}uv}4VI0R6`Vkw7dnoDa2@UNJEW;mNl2d_piN_=tZnQQ$Uj zFZ{!q@|WDacKdUQo(|4SD|Ik)JduSO=CI*@0>m8oyGQ0f1yavH42+AOn9z5a)KbXC zIAULU=u2d=Wr*_a2G#wItno&ee_F8C80AXs>c`QzQ6j)Fsp9~VnB4#Qt5|8^-Uylo zn6|tWD(g;&;C=c-VBi?#_n1`yxBwiUd+ zs_G%XD*LL@mg%2z7iQ>A+@XwzOw>!D8LDP?!rGsC!wwkWf|z4Rrl>4yyT};nD7h4!}MBfpnUwnbL;e9q85nuS%!i*`Yh&4qwAl>2&mwneCqM+*EDD?dRD)EQIb2vNdo6w zJRQQq>pMHRCfx)B$;&VAk3(jX_$LBlBncPNP*m?DQ`bcnEs<9LhvSojSXj+{_Jfe4 zQBFFGlWHJqSE5IeiC|DChN+?2(?9#K2iI? zqXB16J7u{cuadZBve4|2W?;|aCo**CWe_O}Y>oA-q4%gi!SdsTL-&JOEuE6Iu67|z zy!EJ7-4NHRk@i`0Wm(2cQmz>n-T~fE-)5>52dyXm;xg;jmdv!3`>eA!6QSI64^Y8 z2W}$vH;l@PY?e@4(kl^5AOcyI#wcymbx3=O@qwsANx<+!G7MV9=oPR68nY55T8kpW zeRB$tbtyH9veJ&#O{$q(87cJ{)KUqs}qx&dC?eMtpx=k%P3OG5T&zCBB*&L}CNT>j|?ZBavKHNyto zjN+q|rE~{!#7zz6-%(7`S~>7}9l_=TEE)$XiSE(V>{gF2Q4H+iEbSVEatYPsRk8>@ zU>yV9e4OHwF!ZRZq$BOT`;0Ty|tmKiUbbh;|-MFk$1j!9u_#kd)Z zs(0@4YWaV6N|+V`wsb>qR}%pQPV9fXdk|jg7;B=+Ih22avX1dGMZZ_Ga{a++m`Dp| zI@0fIV^#Oa)Gt?-b2j{nz7*&1z^Eu-_`Rc}!uqRMmEt&?#%{c3Nwf~f0-DRS&V&T# z5B^3D(K9JuQp0W)cD;A7O=;B*<=Xh=te|?9Uso+l?Ln~GqDyBvO2Z;giDeq9?={Aa zEMaXyk9AG%Y;uIEQxkAxtpGm>cWE=I`WRzoWm&MR?YI(e1if2OHJ%IPu*Yoh*0)Nh zo?Vv=Hi#(L=smdOD$6dmu!o*i>yG8ANranSL`h`P?@ZycYO=J^vEJN3;Hk99x^?on z)~TmAb4tosKj-q)370r|pHrLqR^t$~gysT4z%)Z%Pw&8QuOA_{>CSlIU;ubC&^{gtI)%3LyWdR--bT9dHFi^Eu% zU2+k|ByCql;=X|-l)I!}ZCHtlr6mFnq^YI_E5=;Cjj|3h zWOcYr1Hh@6G(Hs6QxoP_G@8HfS;zVvGDAl9k40mQ?yCCjl`$5tvg-%A`7brkzD7#L zfCq;eX>7^XU;dj>+Sju_@s#=9RfM8)ej2Hx_;$e(D8a6-Kfas)O@I*bIcs7ya!xtB=wwxVyAtZ%ku3ZRWH~+qvog*55iyQ61|z?@3_#;kOXg zsBH3f{owCPk%6JOljToO-XSKJDpEJMa^vOGDh2bdj;N>Gfe(ES{nIe+mT%h6ol6}R zETe{9@K4Zr-4$E*D#>@6M-N|D*69YW$1N);_p#s}A>VcTxJ-g1q1RgpDPIpS@##lvzjklHqvbWLKyHg5s@2xjW`Ku7!64}3<&l(no8W_MLO+DrYuf2TXelRRR==7G1erQ zifx~|BwvJ6V2YqCCvhht+1IX=?@G+7P<&uuDuy0LQ#Jtt36+*z;(JE$9F({oOky1l zWk`qG>=XY32o867JsAqUEeK$>KpD#D2f-u*djvPA_*hZ0oMi+rk$9#c4_`P<8ncW; zsGz&+J~cs{8FL#M>q@kFrG;WnI+?vjJp!>%3G0Zr^dTPuNC>9rK!#;(rgtwRj4yp* zV{cWRG0`mhxQDU3Pm!|DZG4-`C8GSfMffdEEDHg~@I(dF5j)ny0hZvY7Y8dZ1&0Mv zacpMthXWPBX+^MXi@@ajJqd<19?=Sj!i_aHkaF%lsd0%}6iXb=M$VFKdKa(ZQ4!Xm zR_-=8_Qf?-8I3H`ewHe(1mQcHUY=}+&YTfg_OBQQi|!9^5twv1-0KU%A)}4Q$1BzCL z_}Yu~Zi142aV-|HXCLrycH^gSQf8t_S9oXUIv3(DV$VPWr|qezHo2@8vA^NctK%^b zr6O&S=>4t;RwpsQKZB|tu+b=I7nBrs8FRc&^F9<|{i0@(u5T`k_EgeFumjSb|(3`SwajgjLpdlqyr=xgSM%^ zSlQq_?$L$wVk<+rJ)r#0#_U-+IZEiHg=iA>w`6=XDcjdfanx9)R(#HXm>^;05Q1Xa znn0o&Sc@_R$yO#=yeZOs@o6)z_8HT(6;2*E`dbs4czpR7yt`21u2_kOPXw6ZOwA&9_&9xU}V%-_{kwiBNa z<5|Du){&n`n6uPQE-5_@VWXj=ZZ%~}PqO>^1Yv~=1Q6Y+OwS65c1 z^S8IyDT{^6v`btsa$@0@+POlBgTkj(?2$*Bhu?ely?nmjYX9AuB6dwPUwB^X~Y0GZ};MWU0Sxm2UAJ3(#dwz$$P{eGuuhFO?{Z& zB@_t#xYB(E*1gBm&ZbSh8pE08h&{)~1bc8X0@ya}2%}xc1`nWR$6>``54xYFHc=(nT@B?s~BGHXaFr9xj0mloayBS zVj@n+|8X22QJg8!jH8bLrF}$m&KZY&v}tkNLL4F^R;ar(F}yUIl{5aAbyBxQ_~1R2 zKRP*4_)^@|;PNEK-^t{}aqQNQr+ojp*)|31DG}Vl ze{s`a5U7mnrY;tziux!#w4wH59RwtVl>svZbzQnpN^R-c1YH6{>w%^@0)10`*!vE?RxI(CHee4d+(ju{KLoj=f3&Zo%y%d`3ub1KbQ+Z z^9AHj3#bJP==}?r-xsj|Er2K%aU~Y<%@+wjEfV(+kbDQf>tCe&w@6K~L@TjmAh#sm z4F3P`DO~@*am0HNY7{(lIj8@YK;FGWO(g~Z#sT;U5caA?tH~s4*?4m0ma^$In%`(z zEU7l6f&}-FNeA9LP?Y7eK_V#k8(%E31{tx+>pldyA(9+D^n82v!S7R3( zTlOD8#?8Tm9v)Y}YPZkN?rbExAL5-2Y~kLrvq@^CHBMuhy0t&H8Y##61}0`M1e&s3 z5eCdOJ4|}~f3FTj&hD`3bWC@T#VE*&XLRcw&uycB(dgfE55EVl%!+jqN^BtxyTaeL zwZx^j{{0D`?=|layGO>mt~d*D`LvW+d)`?1Ppm)aVr2w7tlcR(FFf0q5aznw|70OC zC6Y}pJ)d;E?L*wrz8ykdbF3*52B0g6ja$(z3TS(2(RKxvJDR$3*Z3Cz_E>~>KNFR# z?I%}1jA+n@QS;Jrhso(|LkdVx7X3iHFErkGtPgAH723}GPTtCshnc?3bN~sOcw$EA zkkZIEi_RA$V!6}lQ=4F}D}r(W*ytzuC+PPWqY$&l=?s4@0Nqh49U`r|lwdN-Qkq)F z1ao4L$jWwszCqwtbcn;2uEYbTibJezJh8kV|5_=n(y!qSr=X+`#f55xeRynXa_(BZ zp`Ts+DN8ZxOIki`P2}vTDOvP6RgcsHjUqZT{&RKxLc~}#Kvv;kzP$(*7v<1t-g_*| zwT5@n)xi3+(f#-n&A9CXwlbgUO+#Z)6&t3YRe~(QQdjerlq(8>jhejE6;)A#%-cdGw2p(4pf4(l1A1DJXvIx3{O>JuP^1lK*%@Fcm*5kJCYD zbn`3MyI)i2baWtQ!505@m!%j55q6ziopg{2zP6xptz^OPG_so*i76xj6NRsWKSq)m z2RoIGRJhYCF&%0jf0oHIVHEoJVv7}c2g*1Hy{hPRM7!ek|8q5fVt?FK=hrP%k0z>G>|LAWX`zer?njgS2IQR zt32gqRIAe8)9=D)=Pwq!8@{nuvDZ5>eKO(Bsfe#2tFJcU%2ck^^8IW#?j-m5&uVdy zXR!*8ikW!;M^X(zuGTv>lFGBe5AI5}W^#3As&Pv_YXWr#JFdo;+RG{j_bpCes{wv9 zy5%YzvP=cFwjzdng~zKbm$&S%4Fa>36$DM%2{q`lCL4ut1)g>ziddW3+@*=72j$eu zpGH!yY&i;d6&d5~rbtWBPEOCRu~mF-MY5F~_o~*(5qC|q8HUiK3B!(?&$?fZ`APac z)T74n^uy9tcjD#RkvBQoj+czAw+z$cZXI_HU~)6p!1l!lWib>rtA?Pgp({5g#yFR+ z02J-_p?xJ$&K*kfj4;N;BNa7`7V8HRpTg_QM!?Yq=Zh#Y?H4eBO^-cny!@Jvn^K7c zNsD?DMr(tp>c$ z&6#3)PRlj*i_v1Ic^8Dg!qxs{0r^1$pDyc^_qnxWYdB9%r46M$2t&+{wewZ;dhfR! z*Qi?B`_$zN?RytC5s3v^ISelT2u}aVIlIVCVY@UiZlPQyoN7lH7=OD8De{GvN#+8RH{un}6?5rcemhZrXZEb0Wr15D&-!>em5fi4o zh-3`4??xh~j18xG=Rn;hkWUTBTL3XVgTmJBG%`IAA?RT!k)z0w)L{DwCcN0=2u2<^ zYX)tSUPF=N|Y9}lD3gj@PoI(hKCLU7-@-Or{yz$ z2e{cE_kA;2x-Sotp>B6fwBSQ1jNxOFF42UwEN{5r8>34lKAHEb@$XT}_Ew4C)$`8otbD{nC4t zq;#1eW?YQ6_km!U{o;g|TbjcrtZnY3LvCy~WIlEq(9xX`vZJoEzgC=*!99^E-HquR zOs#+i8xnyE!=r~nSij$fWd(TK7=AWF_4&i>szxW%gA#vEV|l%ga{3M3$cZl3)nWeU;3-D zB8c;!cvXy@Q^u7w%;OA`oN{oxlWK@@SLgsoVW>NSrC~)#S-PM@7g~A;9|d#b>qrqlQdzQWO-?E0eZ01QW58&YJS98PdK0 zX*-7E+z74sySZlg#To0kga--xeBics{tlCnfhN%RR4}UKb61v|c4&QUi5RAhA!9B5 zX)EIbcM*r6mHEbH(Pyg9$OgBx&~8p7FiMjbO;tM8K#=$;J!+t1WgjDE6wS6k2S_q} z_c8oSHQCk1Ts{&9H`UF0%2K568VpWHWA?ia40Eys4%Xshq%|Sc)xedSI2QG z9#qMoh0*mzn*hc`)dyy~$RxhAt~Ri*pD3lhva@RDX}AKCW*?c8L1TAe(%IoLC>10s z_~rx#jX@b6kGYMCPGvba4tS4^tC23?<81zua#QOjnJI?W4UoxD6V{C-BA^)?p&9>g zlTkC+f;9r0CW1V;IoLOX;Tca`0bM9Fg7G;*Ljjq|DUuNbmm-aIg4Wi&9|W8=Ha7cH?g5;-Um3?9}}qsTlEV$_s2*loQbMz$;ki3lbdw^ za8^$yV_EY5(DjxLZGdaHW`N)p+-Y!kw-P)Arxd5SOM&9QyIC?%9C=13>)0A0gplsW0XJ$-5fmi+fXf zpfL?$CCnpYqlpRVfv8`jCkg_Yw?Tr9dhsYe&uDzx^2R9f4`CyKkAHX$FVmma4a$v) zuYIqT-YvXRuEgBgF=oTZsFry};JvHH8JY5rg*{!YN1{11;%^L|s3Tfu*XcwL-K}vw z8FzAQ33nwwqDk;}S%?4cbebm$=4xJrqWqj1*&Kp7;?qc5?;jkR2yf)jJ*_ zGkixWK6w0bP{+RaPX+R|NM2%OdCl)p$UiRj7gouO;d-N^{A51*^))9`{6!5S;R|Z| z10|U0GlcI=$|Cm@=nzfNhaV+_)5NEB2bs|M%T?dlB7GDrVuqI=!zYh;v5I}obl{iYTLRua>5alfh_be zLa5dJzyGKN+bQOb*$kohy{L5&AMz~OZ@UsGx#xEbQga6znEE`^6Qz|kt=+Y`FshH9 zXO(V^3$)Af-85OSs<7q;40qx^=rVfjg!?dNDYUJzH{2fAM$N=w$8}@mae1=ht{yC% zaq{zsx5hxnc@C?^zmH@1>5GTCH95ia2Y=#SW~Qh%9ED4a$!YW8u>*~(T}?XpKmSVd z!~1EjuNQB`mwy>9GkGnyZ($U-FY8|{Z}%?Jybj~DroDKNeu-P5K2keHSyL5Xz?|=C z6IlILDkzLyaNoN|l4tpeX}v1fk4j~^89%!syUPx{@)O4h;lpH!Xt&v9aFwXdnO~l+ z_%;T+$wgL=sC@F%IRz@u zb(=VrdP}cnOxwi#ev_|WW)kafrz)9_>m1V+>(=UVjuE9C1Ibi+sS>=1IR$S1qm)7h z7kP<3{l48Z^rG0aHDj)oDjWWPI|=)@75iu)Zgn3e2=yP?d-L&?vUHxT!*=@{d*RoO z+7@+*HXDfC=*R@X25W_XV`{%m6;faDJY?|*@bH}ZL024rr~GfYw=-Y(zddn)e$%}B zX94nXGvi|C~iLUolQ2{Q*LZxdO)AoR3DlFyj3pwPRIRM?+ivqCCQD%-$kiHwAP|DU2qhAkAQk+e z*G%7-O{UfQ_r-hxNIY#fnvq)6$|^s^<=QFBDkZQ(V(Ii18!th>@vP|ym7@AoGx?R* zCRsD@kNT0C-D*}yU0KqLXM)c`XhUGX*pdn#g;V7+T_aBiX_ph1GD^PPVtAK?Ng3_A z+o~gsQ-i|UqM4w7qP<;+Oh$}pD7{K>WyU~OtOc!or09%XZfQ2-m5 z6t++&z)noVFWq2@YZxy3!zEjA1QbO%%8Ng0Cz1BS;;TR~WRJM+15f4jb|ptjnUn@@ za*=2>h#tg8lfO(xMFaDUTM`rJb4obDbql8nmz>|u<05A`pjYG`1wV?=OPBcbA<7@z zY8^Z-MtPkee;~#)xad#;-EEOo*K>Vj1MRjDesN*_f`_fC$UGH<*35ia_C{Yp?Okf6 zvjQ73W?k`Nt;qjd^8+WHJrt|w8e~m4 zC0)L`^(8r3U2JFeYe=iKv|R?#O5ma$59Mm(qBSH_<|m(NS5H1^ZGFEl#FYDK`Ua-U zH>4`B0%>N^lq9JmWa@k#h5R6yCuHh&>59aBJf6O^y90YZf!cqu3vS;*P4fGni-(|3 zGCyw)t?o-#mvmLezLwoP4CRI+qCt!PG+*j7eYQ_APEJdNn9nnp zDAP>jJenZ_o&P59ysY-CIcCi-gvX|EoV~Ml`wuyHXCyxyaMmdg!!0Mb14*X z;`ADchXdY95D+ua$Y?E~pykW$<;);`vHF*q$q&|p^+t~2Knm?^-4+U?0y*D}Qk4#F zCP4=%x0P}{9JmmUe$UJe@yFx2S7Hg$1k50WV-`&4GkjH3a3f%GAgi6zIrwmht&wYK(=|P3K-(?S5Bf{h?bImc1nL|6uyE>@o8{_e!Bqm=DtZ3B<<3i{O-@>Lf z=l$4q|JM7_X@UGtGCXYgze~C0FWHYdO213@G($r4qm`xNKLXiagWCvyvMw~KL=?CpYRc<I<-@u6aIm&X=o-{L0aY*L1S~6oF#IOD^Vx0b(s271PJ8~qDREy9ik73`n1)&NpGkU!E%`8 zKx2EVdq$>f60FvfiD!KcpgYU^vQhQU8Cg-CD46+Cq^JoE@BeJXR`$Z2a=z33H<2sxB&L;w?lbgEMwm-q7=w>)p)##{gJ# zb~|*#-!3$L{lmZZE3?9%hfiLOK+PRRPfqJy8JXkj`La77-jmS%ici+QyPJ0%Adzkz z+GKDDkXC=Aa?=monR~N1I2hYAf_F+pSFjpw1cJ@UKvLbPBeE`v-EqI!j?T{nh(pANsBw;b(f;~)d-1g=AhKOw zWg)Y%Y4BK4L8f76N#^j8Mp>fP2km<8()Erf)L9)|MK&cr!c_d0l_X5&O3`}1 z(j~!xB5)j zo8YmAAp;&tGeic&9X_Mt0~DvaZElkcl*@?AZ3xi>equ*8$?AZ#synRmZ8fEj@W1A{TpX$e^x}>GYi|AAukM?oKr6U=xrF2Jo3t<_j6leL$g(%yvw^)GJ&YxFLH_+F)G1?YjUXmd;5elOZm^K7DX#s zW~?!_u^yV0%W+r} zrwG|D>8pRx-huKS_vl6<9S69JW+i+^AcCFhyU{2?-IoQe8os^y5D10-um@ptX*wp5 zRBWw;k)#@j7VS$fY|jb9f-X6p)k&p$sosdF>!R%fLJnhYbj7bjdYHM_wijJp2-C0e z=X?bR?M%!N=eFk8!?g5?zMT-;C%E7c3#4vl_Yn3KaC~-`sNN53(e4(P;aoWG*ed7D z%1TOSA7fN1;un1U*qp)|*~z*GH0nw%p}LZvg((KFWq!E*c-y+=W2|=SnM*oQ@3C&X z8Hpv#*Y8hh9L~N!TacB7HA`svS#az^JGKg(&J~_&cR&z}UsJWm;~H@UtF#6C^u`eSe!Hn?I-N zhU~D{y?3v*^tS7;!%6ParXOaXwuic&Oq!N1HE?a)A8&K5)0s}AIT}rEuSF7qn2~|H zJU7FTBL18?4@P8Z$Dc+9;Bpg(u6jq1N(v+0$Fz<%^cE*HhynQL8(O6Q zP{&>Vdu0Ayy0^EdjV<m8ivHlxt5Awspdt~rwe9MdzL(5$h<)?! zn+hTdWREEK{g~<);z#v|nel>F(Qps5p3@pfXe%w3-|T1h#gQ_1$^`W^E;`&Ou%-N1 z{HNlsK&`0@z6BQJ^}c)CD*HuLO#%KmlOj@ew{P)}WpiBCJ;QVo?IttA#!3&7QBHGc zEYppGW^aStjAt(=YvA#Lk;L^f4yzO^{&iRC)eyca63Nofluyz;PGwu7o46rr`1Qj#+}Gn2XH1IN{B>b-(K* z(#o!a@LC;r8p3YJ$7FxxKm7#CKwzx4_I(v1o)^#>2Q~9k7UGnU+SdlN7|y%diUxnU>bV;ep0xQvMdu1)haK zY0zJhKcz8&-*`Y-Pl9=-xI0b$>Zw>ft=R5nA-JwG;hO=z1k#(9VS2$-S%P7piT8*u zp?jJCSz*J$*mPO^rOv#^B*7JQ5zu80k}l6T0dVvevR@>k4}cK$Rl$_+5xPl z5CHJ_S44FRT09*yu1zGHM;ufmSfx3{4_TafEcg`fy$xRcP!skpAVg!;ks2ZT&&&@rh>l8}mkq{qxN6Ur@9b8LWiZ6erEyy^8o#KMVW=Zp~etvg?E zu=5#b?~$pMP%6!Gc-*E&CXX#^w+}&BjBIyYoko~689o*|8xkV@N5yoh5;Bsiu+vgi z>SA9FMUcLU__7=2id)7SoyuZpnjR0-w}sw{EYngXbJc>`)6$l2+5_=G`41jR?Em@@ z>BZF(6Z0zKx7>fJx|PoV7#HJh6V;ssLdY-`0qLEIJGu}TZxEND5M}8xvI`=647)n< z;c$KP{IHVHoeK6#ahDv! z2;S1HLof_jH(KCM$()jBE-FZ#t)`t^gN4co_Ij=vdEU=*wm zn1AdvKd_d~`gm#bQzJMx2KO|Qu+@=dwA|cM$iwK7;V_bFVj!wmA-ZoI=HL2!U>h!Z zam-4JSXx>gGl<}-P>H@RNVF}FKlN@5Uy3XjY+AU-(o#%sVBY~-Ds~op!8BDOFku5# zR}=`i?X3NRU1x^3e8Tm5TMqfMO3ogLw6)F;zka-C4^}+caR@m_3JH(PF8UJbCFf zTVJ+|OuV^n@10a59}ln@51vG`@t3yWK>8gVv)_D~Vd9El_c9;oDZb^>c34Z~V;bks zL2Q;x(0%2I+}!`lQ&S#NUff!HgGqpYER#@%odZJ8NkT_u2cN^L&$hkgJ_ud@9sFV% z$Du_X$wSbt%k+_145zq%Z&zv=_8woH`UF#G*MuSIkXhX%5II=?LQJSva0GfgK)p*k zT8XVY!s8x-m77AM)tU)U>XlycZTu3*drg24$GH3C8bC9v9@r zDAC33v!R171#R`>XY~#W*-p6Cu|>mmK&KTn6R23p+!&w!1%E|Fy9UuwGih1Qh~DXE zMfpMOVjni(H{eVM`o7h}zhFyhD0aqm3g54wf!bYf$Q z2Zbz?C||JIWKGFe5|!`Z-UNxQ6k)FbLZAS&5PMJHUlJU93W{#Z1=*&pI{Y^Z!n{=x zMY;vW%eX~fZBtVklTea^xOs+RlDa0Gx*?o8Zk)O%v94K(M;@ISk^%huwlU%XBUiG* z?ck7K23W=J^;iw?FWsK8aEicLV<7JZ7ZXm8dxaMewSmzYI(&Qsy-fFFqgX z6^}=bKbH=)biE2$u0+2fX#IyrKMtPdOVh;0`%EBbf{jp1C}%7|;$VSmnYL#1HKp8l zkS18py4eagEdR56!%DM;0ioBUf79se8ur1a-%ZLM%KFz+i+>xSqlqabo;;Lg$M)}A zb>Mf2uOR7HM;@vz{o)O)zp{qM8y3y&HzFgjV&s8o@l5ZXeA}H0Y*TIpl4ntMOo}=M zN+p^OQt3=Rt!$GDjK(cn-BSO8N#xYHVR`2^B6iH$pqa*w=nhZDM7wr`3QJCq1{YT(Zh7ZL^=I~;%!_!xpy&kye`Pd2f*q1cI~*YAl-^#N6It`LP}NwweJ9sGl1wvoHE-k zub=fLE$g$jQcwLp;H~P(8B=WugaSZ7p;jj7IiO5QY*_y4F$^O7s|x=G#QhFZ0U&55 z?WMA~$2=wb30s^?5Ysgjft-OP4aeBS04)=9PUg^~9Y2|29T?+2dJFO0*Ivto_` z%qPOAHv_5Pm+yzqx6=wiCP1uTBqjS8O%Fx9*Z8vK7uo~wy%Gou7up8L&Xm=WK_&g+M|~bGAC9jj@UDv2|6`}c;s!MjH1RqV_i}Rd|ERIgI8eD${kMMF=!{8Mdubi# z9Wij7mi+Nxfe4U9LvB#WP>1tatn)kwtLCqa7XA*s*d>pTGYN?41nyfIdY@Tl}v-yrH=hD@k`G^2=$!wkng+O`*U*MKs z8==P{Ay2!4$Gw8`@Du3!)5$i-`}oTJ-Jf8!C%49=&!OFMh28hqc;+AqQ;BD@8{2m@ zJrcK=cAUex;CzL@&rpA_Cs?%z&r9jxQucCLkH=%@X~%w)wFE9$k_3#vs}#Tpg4>ac zyOQ_7>955=2&k)!&Ck-fbvHjk3^M*~WLoFyC34uX;O~oOjj_rCoHtvIg{wZw*ezk|Y9?H0 zhV2XN)$Lc124p>!_-x`3tp%ZKG2TQ0CZs8q2j)My$j@gqet$mMN z&kDpBT%|G@%p>A+yZ_3E1%E!ZC@6dhc*A_)|XNHs8a zA8W?5#?RDnN}j$P{o=pch{ff@Bf3Lnc!uyp2~3EUK|(X?$~I$TeWi`yzrda|ra+Gw zjZvJyA{e+^ef2Ci8Hr^7@IgAEzu ztgy4}YSW(;QZe;o!RF4d6-fn~+Kk!b)jJfIsR1*Tn&OcHcAil`n1B#0h+>T@5THF(4rX;e~{^2x-0g?H1`!?i+WH|6W7p7k9NKl0eth}vU2RLlhp|ZV6~+aso949+ ziwP05${djXc2k)=enp*sp5J&4VuVIb240BmT77z$0Q^Au$3Dz))+cLm9 zGV@W2Ok-PR;`J02n=9Q!>2mbi)k0&})$wzcV>R7b6$)5Cyt|1dq}@91+fWr*#w<1K z(%7IDK=0ePySNvwu6X%FgO$vtR5wGztmnLfrq-piY>pX29d4&*v)L8Qs7wPudDvrG zA`m2AYx=AcuNR`e7~{EtHv*cz4ijsrOBCXYa&Vh8e1EYW>8Qgqs1T;H|LgGwiSGb> zjly}${M(hcd$?QHRE@UpAJcv&lzlF=?vO!PmsbSD8ZwM)>}RI%Gj-x2sj^VU@PLGL zyWB*)ID7cMlG9+TG#L9n97*U9;w!sE{LlnOerF>I$5NA21A+F97Rvg#1#le;4Ztc1*vSQ?qn3w!2upBKIs32F@OL7 zz~TS7MEF0AiI~BZY0OC_fM`;h|J#@#Lw@g>bPPqH|4?FB-NG=Nju;x1r`TFCpDX-& zu{YgXsa7Byia`ZSD}@C!V&>diFBk5H^D078>jDJ2MSM3<64qNBC*)nWxxg!da>Ug5x=%g^*YY?IE*z&7gqncMx}G)wzNKB62bi#wJM2Jj;0$<*pZdT z_MQ3~em(`X*MVegiIELiYg{4X+nZj3l^Fv{VAmfVa_4X9YNx-6gq(FX4=ZVK~Q>*@Cn&M;4AZRi?#Z#HZUy`uKW8c}Z5H zrIHpSkxWr-7uD4j0Zz88t|1ubv4C2Pw#(X5C%D^r35K?|kwnJ!opv1s47LRsklrc- zu)ncv^`1AR&#_Wq?C!%92Iss?-QE_v0zIbBO_yMs<}47Y25|s z_bEne9$HS?vkNwjN`Gj>hPvOm?`qVNe?`=ELL^}i8b-uSbCyE;_C?8*_ZD+yfI`DB zX?zBmJ81&nmmXbX-XpIAc3-Hv$At8Oz^;jS)$b8&oUxq)uvZjo_N)2=G=xs0=!|B>cGC^uvw{n#wYH2rtZlP(dzm2b|F3|>3Vi!A3 z{iZ6WzG?pFCrrv|BdW+Q@%aI3+;MOZAe|^EmIJ^?2SWisTG?RybE+~T(Qf1Nh!5mg z%?SZW-bNM$T8tPQGtMX%Mla{dSSDV-LWdwBf>_o=$z5k5%x$Tbb*NH*T@oZQ5vbr0 zG-)t6ILPJ$6p>}U?-Knu7(_p{4==E&q!bAU?;GU3Y|&tw01iXdWG7JZisOGZH-aCN zjQl!uC=5)vqcHVMivFY$_LIWE*IOvuqEHzSMiSN42o~Z=#n4G}Qh=lB_ju;4ssmw1 zn#ZUJis0lZ8!%!_kV!ZMCQK)HyPW7Vhi5)imDDJw8E0S?CDGO_kBgB+2~R?yKtPCc z$FhrbcPqYQu9ms6)d+w1mHG2=M3D)R4&%CMQgZElT9y%8C26mOSngO(J7W3WeNjQ6 zc6X9Kq6|6W5|BCdR+b47F@a4Cm-RxpV<~AN3#O>9%L-k1Pg~}hdk+R}-Quc@xl%aD zhH)VeL4&tOlQaFw=m*i@s=>Ov?pbx}GIr=;8fn+>9>1SZh>e!%JMg8?_tH^E{$aty zsSH3QMP;ei>ttj^1a-xIkcgJc_=DUU^2%%_Ss0;gnQ+U3?qY~D*(|S_P)fZ9l;v+C zn+%!5E5Tr&hbbVc`;il+U}tMewgtyAy~jfX*Haw+AzM(QZm-g(xP-h_Pg0X#c+*Se zN=zij*?{TDrlP{M6na>SR1>rH<)YbQ(UkP8n!xAVMv{kGi9Y^`QH142x?1F$ZSDL; zE^K?WnG+2mb*!HIF3dt7F6KvQ!+Am&m<&kM*F%{(JU{P#WyjWC^sPB@>o8MJAZ%L< z)-7}Qpo+3F%+W1}&z14v=GT?8$KnsG-~{;F$8xw?;&T0f(dmStxkM@n7(HOI4Isy^ zC3KQTGcMM<8$!>o5sR1VkNF!%ST|TIO*a%N6|1~ET z`hx62()#U`L`JYF{&}M>Mig{pwa1_TZi%L zWvWE9s*+2b(-Y{^m=#wZ&a~)8b!E>K6|_{vM>86c=PbCYJ`}&oNvnirG?N#*6&|o( z2@-^1o#J!25oa5!(3Nd;pj7&0`(46F^ww)s-6}5>so(dKtw19llijCxSvN)36t(Lr zBn_64hRF>yafDK;|5@+Da#tb-+?27I7}v)unvoQx*@9kAM^+5xI^4*>(-^cFS=%IT#oKQlh$cBZNyVDZ&nqtbUKB_F0+> zQcicmSKWFI(vs-A4v&=*Y6%;#?wE0bn~Db>R#eZVp*Cozx+qLOPnuV5xS}VqST@q6 zIQOoo$s-OOf07lP98;_fz~tYIxG#1(l4Jf|9eVWit2O*c^KIv%CW-3HrM~Y%G@9m} z={ahGJ*k9JoFlxO8j5xL^^Zd(Zs0xIYObnclmgW?O{dQ~7iUmiV#d`^)!q(5N%t|G zw-JYTQ;cJ0mxFw2Cun+NyGl3{ou8GCmQx+A6nE}=NB7#jt%<)S`rk(e5r@|QaMP4m zd%1E6p8rD4w;uY=uS)1Rwx3#OOX|}z6t4dhq~$)GO!Y5BeH~0^$@cST=RlYs@FjAg z6Ao%#O!Bc}y4x=?j@00_4iWR2UQ=k(-cY`?k$ib$e&X+t9}j;RG#-p$a>tn@0$aPF z2-~4A`<{)UzdiYUrR94(W~T}E#MlxwS~4Br&B)OfYd{C5JCXf(w2;*-7U^B8P8XnD`&`Tb-`9n7Otq{Z~Hb@9L zM3Yp>%W|m_1mO)=hlXqGhU<25h@qJNw&WAwbCW^-lDfe$rxTGD83eo)B4$wj#iv`2 z%03tpdadJ7Wo2QSVfQXFprzcPRgT64P4%0?XG>JSFa=PSfS6%hM5-y~KU$?0cv>(_ z%Fm57%EHPaH8Ozkh-8|OBKO=W=yxRTXmFQ&wDp#imPhn_rdJKK$C;_-h?d5~P}Br4 zI#(#VMKCHX%F`A}@Hx#n5HZGtKjw$RXQbON;pc*XJv4O$C|j7_s=Qo*JfD5CqRm`A z3=rdt=;#<|kxRQ`th&6CJIud9-H*??D4BHK*5epALwu?{C!+utt*#r))LXhfqi7C` zTh3%*nqrl4&0U%ts|gR4v7Y9-MkleGXi`uDPkx<**-Ecxw1lIqK-=>eN3__>E=?h? zxF5PvYr6CWx-1W*vOk}E5KZHN&z}BeQEbz30J7vKlvJK)85FXVJ5r)Lcv(E+a6FP; z4E{xG=6TXS)b;OF#CcQ#nWEx$i#OtS>dPOQTQ5Z+MRGcHsWNEXOlC^M3TeLY;in=jCk$LNi<+wmwta&T=79B2pH|^k+@0{85$ieaFHS3l9kPe z?2;W*rRZ>30SIGGt5!@SaGgF>gXTPnmfLBn?$%Sm zG#q|9qV2?#w2vU+LM~zI0s^Q*@;e=@cEqkz{iFz!An=$(fng_*hre=8<}s7^eQE0A zR(ZKF=#cEgO|zqgowRk2wb(FUGmoN&ot~pl+;44=ON^L5v({@iVM-2R4hX%o20f>U zB&)bGXe;fDr>`4RH4;yiqehjlDXt1X^qsyV02X|r!}3eM7NZ#55r{Pf0Ol0gsCj}` z4^wk`yhI-*>O2$LoB7GS$&8o)R!6wAKK0au$l=r8`t3D9#lWr2STo+*yB*t`B6&4J ztY0$XDy{MSh&o*PP-TS2x`6}gcSg>ey8aK zZ>m{rRkbq~C29(DKDfY?K(tq~MfBK&yIN7|O_I-@(zgC`EUjlf(A8-NaZD_ur@Rqaf4kk6#m!7qCoSSzP?3iExr4v+Ae?3yM?^ph*a64u#6V7_!ya~ z@vO-ym&1fFzFa-Ulpcw=)qHtioRI#B&+IpajC`-VB*M|h_H ze3o;BX;?aL_TumLMQe@mlXBeK6RYqBxvo{Q9Oy%p2%-h2K zw}e}D07L-;a9@{{%-bp)Xsi4*4cVrDTn|&R%op)deZIv|bq2|U0ovaQFR_TXC<&9g zh_bF@zwg;>Psi-SpN5O42?xIeae_c{V1P!GIMF-}iP3!L9iJSQHh|s{1Hfmjlr#=Jtn)YidL(K&P zp96#h2K-b9DK!C7`jvO}N`MZl;OO}KJFsrR1Q*X5o+ z$y&D<*zjUt12;nv-Wmn(8gaLfdGfYr^qSN~%e;NL#2w9~K9Dde*goH^Xk~}8MWmTE z<$5alsPF4=wdj67PzwwoDiR-=UOE>SuU5_uOxr_^-&DI`U7LgWA`2)(0J0DOk`$)n zFz%1yZxotan;ZujY=CdwUjSY}p}z&5H{O0>>~jDD{4pB289o~oj5s#P>rrg4#OK`X z3z8%keD#1p@FPieKZ$JFNr#}?wgLR-c8m_lq&G_E1Tm?#)Za5j`Z0~q&uWbiw8p2C znASwGPmi+Z`<8GwT~bze5xXqPDP%%s?#OysC$+slaSej zb!=Qo7z`iPiykYkm6R6w$}~M{WptIi7Zz!zGuaj1!+_ zSisSbA)k7fOq>g@$4UNo5?%BvP#n~fHDYIYDP(wcK3fE&w0Y%6$JQLpMwG?22i?Vj zq#RcHySWxVB`qd^3~WFK!Z)ZIH|VB3%T9Xq0rE}1Z_@0pu$OP}k1oA-3SEmW*%0V2 z8-DaIfVw!Z96)w=v}l&@0M9mCM3l6u{HokS~V>erw* zYx)CvY?&9D7oc5`+y_jwqb7XP#DPBMyI&0t)6>^p)Ye6k^!*7BdH<`hjbAo!DD*jwqc7 zIU)p8!p1my@2{DE9jPbDDij=(@*GK5sE}a*oi#4CYCgiH?w?%fH0tR)NAAfVE61{g zgArg_cUB9nN8&Kc?T?(g-S`F~9_Z$!=y2O9xQn33Yw%fHN0Iws!#V>U5d zfDFmMriu`pEo8m||I{!AmD5illCsJB?X&_6aIxPs(Zq@!!kXX+-zNNbzfCk6UZmtH z-ScSMS_PMnJ5Z7)Y1CFe;)v;(kM*tNo=z0_o5?R%i}D(^OQ+{6oN@gkX1p+8@=#Zwuy4wL8CMPPx`kRjJqy(<|7twyU%*rs-&RTV@Dnpq<7xlz8vve8yq7&huP& zwkQll5w$6%3DG+(PJ9#r|9^Qzw}9XOf8J!61DdSacr-SVSz6U?(PS8knA3lFM5BRl zj0!`RAco;Iu8%YPBFoe&Dcm$KU%FZkR5KuC`}T_bCrd>Nu^7-CzUlc&b+We@vbLI) znh&KJd)pss)esDa{|Ct&$-G))H0x2n*{QCvqvV^|VASEhJMwjR>2%EJ4S^wtTr;kn|n=xuaL@w(oi(P^stYAlH!*qH)7t zD0?gQ(O72VV^@B7B=%`$5Jo{{k-idJv0=9lZB$g>jG`Xl|0!Z{jVNl>DrW6qXohw#B=EsC}r2j}?PSv87_EqTtFT=nJwf z)2X|;c2t(E%CK>$!k<4sZ|O~+Q3pmu=q(BdzuDHehf|X+C&u#b zawP3=juEHAla!obOE_0VX$5acmyKXmS*wPAQz0fM#dY>+ToHSY7bVoIfyT?3I)*~9uB*nfq-JR8Lexk4`zJ87>bWC~_Iu1+@ z5OGPxvoB|{xEZ+OY{82cttlYA8N;ZRO?2P|@0f&fH;}IMQZ*h_pIj{*rE7oDV&trT z=;D{WZ|VF?g@I@D7UfV+INkh#qbTe*>$UXW%f|0c*$B3xZpVCzpX)C3aAnG{9?sW- zEfBU!i)1PEH?GJ!BEEeAp{W0_zW0Aeo!0_>&Od#~whyQTa8Qd|*!255+DTshBV)!7 za`>$#wo%~lAu)>e#n^-E?n&S#%kz`ESng^X*!*l=zoK3@g`R(K=bt(5buG%Bb@2ph zLf?b=r~PgN#6Bu<)ya}Dp-q#PDDo|n(XPV|#JYUhtJq1LD8{7GoBF-pK>MdM=9EJB z&2JRbk6s-d!!*k545pOaWw$-hf1wn?FcfeH9vyH4>bf-xHu+Z)9)aEDhcGq`*rn@P z$>x?cC?*Dz+Tfy%z%bv6)?fTuTK)}RhXGQy(PV7!y)@>%Ka)qJz&pPXXeA@zB_NH} z02r_X*SK6KW6e;CAEi#c5+m`Sjk`N4h}B_D&F=Jty-{Xf8;nM_Z$mhO_C{TQLc4PfH?$WvqKD zB81xxe$-6Qdz73+&S)gbqW@|34aU#T6*~j9K9##8Vlp;BW|gBfxl?l7(#H*_I0%u5 z6uq%TBODZUd{Q(Kl7?w2>7lCSw8{DJfqZ-jG_v&RsU@QP%vvJ03cFfUG6}6RduYMPA}fDS7EdJrG@58 zd$`0sYfx;M0D?)qG5~{DH-u0d5UIj$N@A}Y3D}@dbe}S9uLdQDJ&*)oUrCplfaiU* zML%KJf1GMf3KDH+uWC2>C^xHLCV$2zMCNEO^A~RsP76$-#>8^;)4<)oJnihbaQyst z*udB=yp`IWRPxA{fkcEp5ApKETI^B?NLDAhHLvbLLi-WH4b6H~CffYH7M&4cXxMpcXyZIP9VVpB)CQTea;zKv(`KF7xt%JRrgi3zb<&UpgC0FX-f7} zBfuBo$gWzp7Uhiry0iEJf-ykPd$X4EHDNQrwogS~mam~JE9#|_`BY%1z^P&}9VdZg_vK6D(9QpOf?s5IZ&|1n5fAvAo%$d^+xz!B83XlKCHr_c867+=|jwlcGfs~yh!_x}t%`|wpdhvkcG!NV?^tWK09x#+m0PQ1Eo ztKD2wjJWfGcRc-gnLljy^oNm)G^I0iTseq63hyz6WY9#f_*3`QyUx5uWiZ^fr{q?S zhqln2I6qm+4NOiab!}gPxDxdGWf?ydhA^=`NFEC834X2FPtU&pthVq`=suSUvg5h9 z86c>@KGSy@HSG0k>beQBFSgHNNI&T=M+nc$Hf?A!#lN>t5Tz&YJn10(aIg#?w<4L9 zv1`Z)pUMx-a@u&DD=Aw*IHuXJgOBOh{QS?cc$bIp;YX_4;)0(m&cYrR^@d;A?M?+? zpLD~de0vUOZG_oHg5j-*j(TdxJLQAhCvYQSOjn*`A{tcqT#faUG#~5XGi<> zLO!Gtau@xl1M=%Mp6N1yqJMq)42(~s?DhJ0R zZc;?_N9C?Wq8(KYTERGL!98l?n-Fve%!4h4lA~-3EU%6>DKCiXONjWEMvRhDk)i%H zll%`LOot;JF@{;|N+-e@uN(UbVMk9^kW~;)@Y+6gguz-f7d`x*Vbv-tq%9qRn@{4> zf4D1?S4*prQ*LJ_P5L`pVF!*TN4CEu)+&^(9-Gk)g~d(4;ZIMR%nR2Clp0jCZ z9wIW~Qj8I3R#H0p9!E}yW)9|E`lJtwlrMb(k$4++a?`ZygpBvI5BqZoh9C-XmV+Ru zOell{^=TD6A{F0P%h5lXSiVDQijwEzKxVgzm!ZHFD~@?{i?vFblXNWmDK}+&)@nGG zskJNB%fakzh8Q1@CKv@>9E|c0Mu{?i$qiqba&RRE2=rRpt-?@IPvL!GJ9mAfg7$SdA?^;43?&%0|-3UW$pI z*uV+cp$i=?`!CAoHd`Ewp%){+^5U?2kuPLOo`jMWv4ZEuNPKdFwR&589*caYCCt9m1tp&Z{BTiKJomTh}6bPh^(7%l_zYW{9f!U~eM-d#~Lv_Ij zaGSBHh1nPy{lH6kXk#<7ay?ExA#Unh!#av<5_NJmksx=gX&Lfyl8diW@QUw7HrAde zxqa>^0tp(TL6Gblqy0qY^2*xR!6NaU&SABU5HGUJaya`gA_HYN#w2<@1`DSa7nV2& zCrb)P$@wr63BPXH+{R*u$}hVlXmH^nBvI6iD85R`g>x>>`8@|cF-rdB4?7JY`d(QD zv%X&I0%;Px`d)X&o>B+0a>Jgo{xIIt5-6~8r`Jd}zi&+^Co5jm!=I|nDPx*X5jm0~ z2(7O*zkgS{Zw-YolBH|R$vJ+nM=Y=Kig;j_L?PqKq;rqwNwa(9cT&N8zT#W&Pt?I3 z7>S5rY9j!xc0G@!3U7#|C%e(9_WpI?2TR=KUL7P2SV1^k#GCzBqRzjbq%?lmLdWT` zR|DB&7_hH;ozLiXS=5gc;_WkniPydA>!h_sgNQy_A7R{QL+g%XJi^;-;yOx5%E@h{ z3@&0DliJ3ULVn{lsD?B%Dc|}NbpK3de#12vnaXLSl9e$e_GyFi zRtoYGxrpGHDo-(b;-S({8+Bo4T6Xb#%J3)qu#4-^CBspw3=RnZA=5p zX7X&NVjJ3=8p`3uXyMI2Bn`|=jLPNY%^r@=l!gV0_D>(^Yd4?F=JC#yCd^q>5;Tdl zHtfq)c1^aDisQ&nGf``I?vHJ0gIx#aU2C=L(dI`)rY94^=uQpw0~WGg<`@1fER!y- z7NiQFF0B1`C3DujX!@UgVyX!AR^pI`#q0zvno(UXiMbPF%IJZm98JhtsM`JY5|751 z!!;Pg$bN%#c$_r>(Ou(_RDqj>0ZtUMsl1T)4qnn(#;9Ey_c1vAgQ|K9P!nGvxCKhC zvGeg5`B2@f9Q}P#cP+<+}W2 z>-v0edDDPr;)?aM48((O=mr8n9BWjP^KAZDW+j_g(2Wm)8&c|^kcb+%M^xRE4JEQQ z_JegblNr4Sgvs+-u&8l3>CdmlhrymqGRUf3jZK)iraU876KtX=SV5!ng_P{Nt zLLicJ330~qc-RioG619wCu9N$`MXMp(PN9TEjt3J=GX`uT$iJSqvPzTCNj_A!p%Ma z=p6vNp=iTU)S1O~L@0=VF(=-B|K)a@gKbp@idL+?!4CTi(1h*|>Gt%l7up$to7I5^ zY-^^JfM#)EyECA(amAWs4fh71)&iJc+)n`=r2_i72v9Kr ztctEO7;VDe0z5@`t0)0@H+x8wQ&_100^1T94Amc8c)toaI8ESwVjL`#<`E#~Ev8l=+OJzw*^%!ZtFupe4AwSq(zmFVL0HYV!23o3 zM^6GJ(@{NQG8OeX=2xJuIY{pah?sI>f|)H0`;GR?9Ha_`(_RK8I0F%nP_6MGs^ z#+&d820|+vy{y^*@{$$dc5wjj z)0*9Pvfk61ky7W~vmV}mfZd-Y+;h;1J@A-4@Uc_(1U(21KTJYyL|_l%6pxaJZS@j2 zl4_wvq+r1R!#MoEY`GO&fJda`amhX`{Lk~@kN=k~H$wL_>0lg~XoG<|gG7Ni#e0WB zM~z0*$8cu6mJTMZDyd}6bt0WQK%HK@|9{Xp zkP-Xwa6jzlw+;X7&!z*v1f9l;wt8i}OX0 z@QamxJta{XEa;obVqfP~e~uMFl2ewc*lf4^@z+M-?(482>Lf1#(z5-c=5!X=`(yV* zN-PO?(;pu<>&a~vJ9fhY&dH;pcMy!i^J{*TI`SPm$7Y`s&j504mN#G34zbhRT8a4QP&f*%k;vRDUPk$v8VCM zCO56?%=+#J{+o6bAARa=7QnL_oi8V_Qm)N5%4W0XIt_P$r3Sckafq48i&eDmwnr(tPgu>StDg)rx>jUGh%EOf8NocVWo=)3&$_Br``J5C2J{ zK7Ui9io`}^m5@m0#-8Z9+mBgRJsH?}tnQ^;m-o}ELH3cgW6i4rqVM`Ow_%X)eitk} z5=OjGJ?n!S!&BiB$3EoM`&?5#e^>W;AW9?8vGRwcd}=d1gS(#SS5(qOIU3Eal9`Ou zvt_^5vJ2t~m#N(Af?j5S*Wr;rYpS@P;c%cM6zK|(wm0gzi~iZx0@|i+!>Eo`a4Lg% zH#`zT1g%ggaZ?p;Zti1W8YdcLrJeY%T(P15Z&)m%gNro_V) zU{*e*cz8gv*bJyh09!g3n+nn{3nUf|f`kZIClNMsO?DBkvPFx4qmxdjiU%MHp|~eB zByKXfxD|LB>!s+yR8kUiDI8KLHfo3pV1m>%8dcXgSKGG9Bq)g2B(#!N%^?G549((h z1xZ|&DjB|WDIgHd7;{hqjIbwiCRvfWvTGN@zz^BW9yZ`Ig+<39&P?cpElcY6Oc%G) z%uV!aX-qD;o1FH9^M-}UPZn5Oc~{J-xYZ)xDyW!u*H}3s&Q0LgV8lUfQ^o%#PT|iT zI*e!|EI*!*O=6x*%nPoE0TUUU9Hlj8U#W{zacjOBLsnu5GbmA=zy8dqA~ZWXDI~h7 z;jIr7zbm|9%g`n?#m^Gh0&15tCw^xON?VSD)6ki2t!&_`<|-rh3;{1P@g@?rf8oU| z4nZ&1tbJSIN_?zaWVuvuAegC*Yp?tuf~#y&Nt)MmR;BR7%;edoRYmC$DB6fAMe(1% z@R)4lM!*lztMC<{?;F_^Pjd3jnWNV^84a3$S$Ib0^gACrz4&!l+4kxND1<({Wuw-8 zOkZ0r|C&ivm2r)kX*1J8&s@iZd&4wKw7O}EpJ;UO$x%m1X%&{oXKy8=s)w|r)Y+C< zot4O9ycfM&wz=WvQ0a_2X}CM}k*`%_qQVkESpOWtXJROG{V7v;W~z^e?F-1ry0>3) zPIb2P9OTH2AitF%NZWC8KOMr zB5<{b+_FF+Hevb(fYrZTX(XC)(HXfJ*g3XRpJEp{wYXu$=Ark-eBaJ|SM}j!HMoL8 z$EYwzU%!N4c8K}0WGdpHjqe92L_4{YKq&54k`BHJUwy)%e3Tb*`lJ+j&~^#i3}051 z>9D3!`2iWvwJEE5{L-*1_RQ2lQS#NIPA700m_)c*UDYYTMkdEl2pL?A_r$eRap+pl zcIn-GqLhnCFCOarJf_zvdj*xnHMBiiE*JNFmcwYN`G zL`A9D`IyKrX|d=M8}{?7ii!820e)Gm@w+BY(2e};;4iJB{-Vnui_vVh)FXlZ4J_){ zF*T;@YClZ=_oCPQ4h&5fv57qm_^zJMG_Hxww?|VjRe>NbmwW!+{35Jm!EnW|_jtO0 zY-}D#=bbRigZKlu!N1sl*)KgTmXg~0FK#Wx7diXtov}&;^sUD(d9-d+R(3DFuhx^Z z|5bFbraH&i7F+b>BsV#IVg5}S?q15%MJ81g)W3iH)4{9GB;nNq&3L!XdrL7|Fle4rz1*yx_Dc{@c5xLm_$E*!{KPe!mS zxtM8;8;h+{_V|;VQ~#t)EWCy6zrsAJCbhR61>>6#OLAq6Iwo)aD39uUoh^GJFtohF zho~p;?o>wcD4_OFv?d62X&RWm=8usoCZ>V)T!L{VDYVN$D!U#mmL8x`ZQw#num|Qi zOrwh_L4j`x#<&OpuSh}ZJYX9ErfXQwEg13XA>t(%1?QnP5gr@#AF*f%#@s_kEd}D* zfyoc-`NRq39N2GbME6?I_0~VPxUJkP@oOQS9pT&_XM}4SE9SDez4TcT5Y$ zzY)xT9GGL85iu>&8W$m3**tIr6eK|ZrFGW=SCZf~x;=4JaR5Ptlu$^C+{gCtUr6CM z1VU4#GzK-?LI}X}TP*#12Fn`SfE7U5jIc*T;ZtIdOE6A2Fi%EA_m-{L;rJqS5q-#~O@daRq2Z;usvACYQ#yD&~V~cE49#T}krJk&wNr zdi&+uNsnQIsg(sJ-z}v z&a=}c&?4K?n`l&y%F^O{TN)HrhK~_zV3lw;7qRv=0t zze7{k2dmebPi)4NdDxc&l|IsE zqu@_%7?8;3HsMI{(Na(8<@ROjqog>kzsSYSM!(oNz-~DJ)DqR(1usw#A`A*RF(|S;v#q^EtIrLJqxR@r^|65huqca9eF2KlA{j5;bYH9f z3_0WoKlf6*k|#_xDDcUk=-3xvHe$|bt4H))B9}(LXU~MP6Y744X=Yx65?Auz3s8Zg zW@Kqd!AV7hiMprJez-?ffdUZyfQbJQ?JcGM5x=@6jXONn^p3DAk?hl`d%Y++Bo*LqCWVzw$w6R#}Zj}ICv;p=v*0( zW4VtX7zU$^=!wx$ZI6zE;5fen5Dh{9k=YVL)l@=SMY}oOvH8&wai87Lip&ABM*tah zxR+Z*gpo2tb%6el>etf}n>cOKO`ad_RG09cj?}eFcbG3w6vSI#^DSzmA6hZEo*<+_ zV6A%pFj8@-~cM`h0 z1}=#P5V#(s4!{a&An2+neXjLJt#eL;JeAV$-TQXLQC@qXy+F~LMpR4js+9e-38RcX z=~{~I42t+8ezvNN35I%B6+T)^#edK98*i@jEU)gNDs-i}%u7LbBjrrZh``RLxN25| z5!e0dt`dD$Fnc4;skO*J$&-U?Goa}>X*aQiYt4*SAoa>Deo|?HbgP4OYCHfW#sP;-FYjz*sFc^%sRQkM{= z4Fdl)%O(>NY(lu6pi6O@ckq>0dnXOkLTQIUl^su&(nqKIo*zAowxyN`);@8TznEnO zczg1Va`gmp{#omqhg0fPi9_PNKMG-MI2oU#Lei9~GL`zh(G(|n36Fp@^jG9YrKGJ+ zq%gyeH7rIz>HH}po-iPwJ>MLHlNZ#=l@+fS&IBQp(CR{|{Bqacn@3xy1nGt;5msg7 z%J4!Km4>>BS#BioQi=GegcJpS2aU6c){wOHRSjZUpiv4BYR?Zn=NI43laKi0{PW{q zx@$SDxA-MYxzEyxi_+2MFL;$dFiX-&K{CSZF|cwxkn?T$?18w$efVwULa39c~VHziXuxKUf;fe}}WjNk1zf5wOtXmgY*B{c?>r3X-` zCoq*K#;eG0t%vYFjqns?;uYA-pP}l&brOl-=R5WCe{$==t-973XXLV#A!Sz-9uM7v z&@D{qG-UGf_rcSRm+nnbQ{`M3^}4i;Sq4n|u;N+&=~h_n|2#(Okj{4m)5g}Pn)V=_ z;f74H;CAUs%vw!#fh}fJlp)TYv*Z`kz)8}St66eZ*SY3DLA0eUBj}BT0l0J$ARTd#i8xS&5>&vmVCjlA?pJc+04%&- zm|z8{9RXH90g*-)`adm_xuWT*7a{tU)cN(-p)W!1fMkZnB!)oDqs1TS*dO2*<@|sc z%l~EGtct+#Qq69nx4$ee0dG-%LD5zlfDa-#&mtAgP%P4x7FG3P)FW&F#u7*TA`+ln zW*LB5((>J)NO5`LByS0A5sV;OBsT(3zFn3xEJC}@XE0&pNH1MM6arv5gSZ@vGkDj? zUC~sR0iZxEoFmNcTj0qo08QNUV<0IK*?Ki#5&jmFK)eV|eeuK)+9mP{r~#_^0oV@! z;Ee!ZFV~V7aX9iekwVsN3+)J=HwkWZ#`Z0e?!46g-dqc=SEQ@KtJh2Fl!bp)%qqi9?#9Gqo98>*?t&g;>ri%oP*T;* z{DIff&3-r>gw)rjByc)B9)rVTy6_79Hcj+b^l_`l&1%ks%IVqXmB8vtyoBCYWj{;X zrYtRP`t7=os~A0v`l7_w&YSpX&b6+9e*Vej;&Aq7m4R>caDCo-St2c-cjy0+AugDI zc^@iaX-ay=e?70G3L{xy_y;}xzC(My#V>zN%-=7sN{1=PhCP0VbQ60#PH8_9Kor&W@bX-MqiqzSz`!i$e3;rik&B$kC3(rJp*C zq5X(h5~D-6O>M!^CU9)kaUivqXp?K762XX=o5S8$=|PrEXj66IpFtLDZNdAo7NW@W zWvsrM(s}S~Ya4l;>gw6U|^eUerHL zykEBx8RnTbS5Zr|wm!XTsfit|;-orO((8V(%oVyQ_h*i)CXN^xBmUHJ*!Rvh*yuv? ztw_A9zb=E^l5g1+8SbjF`-p!3rW>6suci!@nFY{Z_!Z5C9se-CwK{#(fFUM#u=?17 z=F9BmATm3;%N#PhwNe?akW6diAKe<9R+7Cv$0d<3eanrK@CQGeW>BHV1EbX0cXLVz z{_5C4_y=XU8YIZO;tQroH&d@(gPaxU3Ip~vtC`04l9P1}eDmlqRc zQ`nF9d%(Gcrv6}r>a`Bx!vmq7msZSyVbb~c@dfs+w&uYoTo=B>81FXZJI!H$xhB)dqynX0>>fg*WDCl!EiMAl zBq8b;B3uGoI3aEfqD+qjTtW*qRmYJsm@Lgz@BA;?aIvZ*5XaDD84;TB`ijg*te*1+ycJR7nZ1Xq) zPOq1Ia7nX|*>$w_m?jqg&k6qsYkAOCmz=80U7Tq{Nf)BSG+Ea~GPj%cR8JL$Zk2{r zsgV+qnP#%9y%kB%MCBSa-n6o$v~$1-?HXp3VnmD*<#7rkB{p1)Ewv=hNSqac!Ld>l zA$F?JJpGrSXYoNfjI1^T1j1=>32u2fgdr)ibfBURqexg?EBOKK448}#;GYs+5F%bJNMpFnyZar?32>iH9 zZr8;nypB00?R=k3@J%W=lCFdx?Icx~@j!0Wos4LR(LLIv0MivLrKZeUIrP|82Y=|F zIv^U9Eyom3Ezx4T7CN!3lC4;w407kS>wa}0!Dn+h!O)gOSMytn6DSL1N2>lZZ1=JR zQKnqE98cGd3QO?cD-%ZuAja$UxB;>OB<1u2XZ5+4dnbOFE0`zs^U&zLke24zh0|~z zab@bN{4RsHv+a!em}((m>IV|Xw3hD3ogfSM1hss<#Z2%kMyGd(_GMVEOeX7Ik7Ap4 z)&9*lLD6$#sP|=i;}`m`#B1p!T83-+5aJv^`5S8wjox&t=kf7Q6~h?zKTHooo2@2j zsFVAzlhZOe6Bs)+-=3YeJ_~m1E0Ao~KO-Jv@73o!c5JcwlAKV#V5g-B-6x@>Et79O zrK@j#-z~8ICiL%NPSutstMic{W@0L7hx3$jy1*t8caBMGz%8%Q_>`-E2S(3NDqq0F z6&bp2KASN11))a5UrJ4W7FGq$_v4k6q>l2!05W!6KkBqp zw+#LRHVUrlp~x&`>{bU$nlHGerCl*-yzzP%N^dcA7N}?v{u9#l3kmia1#G+f`0TPa z>=U)r_goFrYS_Sax6l-;tU)}+w8(h`qNgj>fjz0+@4UK(7_Pd7>lOI<>4dw7umG*A zusq>d!adK>3DKX6$a@IzNowM>(kFRCq;d?KsDjmf0z(&U4&0F z9CAlr$(`2#Oki5dy#;QKMcq$YS2EadE>fGmbNt8SpPYpKB7I+$lZ9?WeN0aO-uHa& zyYNUa_S<6n)w_Ct@O32p^ar8!lZnMwg|n}N{85Jsixj`-s@GV4P*eDuXOR0g8#COk zCU%dV75-?XZq*4AyQ7NbNg2;7QfCAH)v)_w-nUicP!T*RHD!-w^!WQl=5Y`wQ0-5& zRM*QwY>2vNFqoPDaYbZ;>d{kv+(CC*dMPS?h3LNU7f0mrf*lI%b^OpXmG(&FIC|k8 zX)Ml=_3$q`k(Pw92I?Do{GZoRSdVnXYt*}0J78M|o-pJiFyzaYgV&2E+!LSmI&XA_ zIZ~v{-wCCvHb-W6Hy95S>!a`Lfdp-A!` zWHnrJr*hA7+UWK&{hi_V-Cpp^m|Z$zs~XHY8D>Nb%FYbnttl!>20-%>5bGAG;RC0p z4%WuT3DgP`%l^z{@0&60C(aL6Jpv&9NBiCa5RU-H%dCv8fd!$g`q8cxkuoU9#wn@z z{~b_A2lPZ^LQ(kd_yVGxFtUwoq(fc#!7orO6?Hfy*exKiCmK`|1rLqXo57|ICnS1h zD)o+#-N5pCC2TeaAwhxuF;TYL(ZTvShJ=)5qp~Vnk+#)NF0XiuoNbD*9&jt`l*J|UxPJjQK| z3KXsT7Ofhb6krd~Q;#R7ipUKol*y4Ris9njwo&r=41b3zqaNpQ0aBF*uGmMd6MD&K z3pEOm7h}ht>zM9OBtjWn&u_^&9qCC8yOhWA_2m-1NrtyC$4 z3`Jx^0|8-VCNN1vSjJ0kqZ_}Qf*xK+YJqKp0a6^$HJoiS1T`zM8$UlgJ1aWcMw;^@ zFLp))pErt?KZU&!cX_U^Z_43H?g?FdqBQ$gOJ|Bp7yVX&*A0!w4u@xGWcf~>zNC*> z*k{C*+z!FdVLN0Jus4!@1@Q4@u+Fs2kIS4-e6~W;GL~oQVAmolE3*K6hIT%jb0o-b z1Mpv8u!|=6jR*h_4FWorfS3WmI`EqV_-&^oj#m%I)frb&8sD%~`5BDu2qqzR*`4*x zd&v>wDHA;@GQtaVkPW1-BiajRLFVD4aLlg+7GvMx#PYH5IhOO}rFx!9&-Zc=93FFn1o&8!Czwa6+1jyL+Q&E+bif|p#YFxuwd%lUz*qVUVdZ5mG6}nr@XO zkE*8c%zt1U&e0woF6|#tsZ<1~tiT$K6!gWQR0g%UL~GOvxs95kUM^q7l(ask2RF`I z$67QhCG&QiUns0=n&5`p+QxOP#NOI!J8RVnqCO2S z0?`AsI8wDofZAZt5Wm$y8IeCTS!Z~i!g%;XW+e!sI^%Bq-eLIF(^@;L>d6i58HzS6 z50*0rNLPeEefMX5|0v{|m)Ymm++J| zUq5D>t*H(@$+^(AKq7%(Cz+gR)rn{T2)+Q2>wlDPbJ|)x{AiU|D&dG9TO5}+5;x=x zR-px7-+;f_Fd)^8z5-N$HVz;VY!isA!^7*|)RHT$~91bwWbW6XTI*LIC9eb)8U^0}nJ4VX%6y z@s4)ap?{qoJ!|4tKzgf`+J^Ek&(6Jjc_d-AHYXq^<&@bxusTHVd=~%3zizy3Xc@lO zPmi;x4W|>(Thm>GM%CLk+dFVBeGKv>{DZrs^$~N_Gg302#_JPC_gx=WKNpy!GB2xH z&!OYsokd`+GDl>qLx1%N{Q+9&Ynt*k-w_#q_mSI&2@|~HmP^iw(L%URA zeuo*1W=Neu`myN`#Cd%|l4{OM&P`5R+GKsY@J#xAUN1&7x2nR5AbYMI`Qnzd?mo1_l!N7IdZ#YScmqRi6a zM)PsDVkopyDVf7tW7OuLcJ$Mk_p-7cyXW~XtT_1(<}!y#hNiB9c$qoz;PnDnGQ zt;f-p5A>CF8TazJHos$y7oJ955TS?I*=rwcDDoKe#Po3U=eE7B)S$Kz`GGU=;3p`H7~J4P)~y^nQ9U z$qX_H?(q5jk}c@>!fz%hagj35@HcKKcz380B# z?P4!tFJ#)7mBkseM0qc3Ll>Hx=j3#Js;a^aWL-g0hdvtxt2ARSf%0St#M#m)=_mo3 zNRz6|$`~vnzWiuSF8)gIIy~R1JfAUMBHeg}$MTerr0VTl!n4}5bY<$e)3Qi%S#?ou zSwG6m7eQJHdkeu!>wqY7#Tr@5f4!xEu-!MCYM&rxjrf$+%O4qF2yBXeJ=PwUj8U>cfhp83PjO|C20gNG?N#AUOBEzur-MclFWfRph!a{Wdq*4(Nn_6>@WVN zN;DhafOt0B!+~zDW9FlsGs60TKeRpp}#dV;a(ae!46` zMF8c1l5Tt<5$5sQJW><(6%-wWAP%pYaWYv+KcW+o?60h5mKRA=f z;LD;%02BD35DYKLn*Hn`UX<|3qWd%)9d|9)l$fd3JVivA zb~9lk>H%^TKVGVcW%4jxD%)Vo2T+6O<+bnC|V}+n%1Piet>DIISwpA zhgAtl_o9U-oNck94mg>&07^i$zkR+B>RT`T;dm*_$Baz<7fjC)q>m+?GLht)AS+G$ zI^{^L5ytOtz~gKwA>KOYy=a_Cb3-VpvK@}=gUdij76y})S^DgL)iKZDtgxup8AmC1 zwIkVq5@S74K`wt}$y3HJO^w)=$MvZG0bv)|XF5?+_TuAQ2 zbtes+C#FCi;m4;z!m*pjSS4bI1%~!mJ zS(Fw>D&9gv#}Kus)b&`LZ(V1A$UN7!yBdYYl3{z1nq&3#E+=_?ERYd3btLq@sk@^P zZE|iYTWGCvqyt&7A{E_%l7@&5BTx3<+ZOKq1}itOlM0f-3RH%jGpsDR$Mgj3W~f!` zEV3-S@ur{D!JSI@cvXiAKefPRU5s&AeMUUry9$$yZ151OmN>E})p==hC!K;yF>CXp#! z`8mrE75onOGOXhsCES_D+o80#*8jgSV}`6IgCT2m8$tXnaN= z4&;f&#(IGM-6KQx91*+;=5Sa!G@J(cd%L!LGk!J`j4pc-VhJ?(RtX&x4wI7dR3VxJ zmAH#n$Sj0y$7DDY)4U&I5>dELDA^~bDxjb>YicV>0_iE0a4xBeYw+GSJp)lJD)v&^ zDN8@01@I4-pRp`XkNC;w{`Tldw1qRi{x{et+KNG1hF4im-Z5PitiGodHkNMQE;5 ztV1)%;Sjx^TNJkt?{J>FnGhyb9O)^ii6eBA453l73H(kMb?t!2izSIgSR5uDh82xQ4(8UUFYP@sludQ{<+@SY^4X*r&Usc0bqQCEZZ^r}jUrfO;jxv5 zSp(B8?e2$=vO3zhtCm%OMq+5Ge2(9UHgMlj+3~;U7-$a)jN9PWxz9lt?L9e+ABbZK zKYI)T0N3#c6M8RIUMIQ@`Y>&wZ(6XOxhi2C=jymWmHhR~o@J8QJQM``0?m8HN|hZ) zkH)3vS=p_RZh3-1&x6`-FOptd9} z4_$!JLgtDzw}TwsZo4QxSd;1XYZ1RdT zVJlv(Sn?OLXRBzv-BgRfD6g^m552a!sQRYRm(Qp#xE?P7<~G5jrQSAYA>8NQ6G(w* z;Y{mW43S${(G2p}RY3`I;3Ni8rka2>xuBm&Y?$|q*#s`R-5%ZF!Z`GU)ZyIzJ$eJ9 zy>aNXu$5zI<5Eg+jrcu3NpkX#AR{XuX63i@W_451w|p(UvS zxly635a`mhejaWht`!$R&nJpo_;`cr2;}t0>4H`Y@L3Hr>IoElkv*}Yzvr_rlOUjq zHbZ>2v;X#arkWxx)ZqinF0#f(umlI^Uo;Ln64SQ3C@!7o#u^#69QY=M!7s+M)EAY^KtHOh1+jPWk!CIZey^A^kIPsKeitPoH8~`sMM=B#1=Aofbzwwn`-YHBYEz z5Y(X!97Qrf-X#56DeyH@Ali!Z%#$Cz3!L4?B{;)afJEwu=oGW&tVkoyBgbM?g6)_M9Ak`4woUetb9}Y4?BRCprs76)NC83ORzk^> zt0DOew(4-7u!;D}L&$k#!4=|8Xc~y{;)wsUQsG-tiDy%3FaPTRgt`|ThYb0bm4<1N z2A2u+akF{TPep)vr3;`C-rGY)9@Fh4Vsop~#i%l54Km~vuy3V4Q=ckWTlvpTXGo8x zTI&ZkXengskZrAt8VK?+aYiv%;)%v`UTbnJw5c}9bL@tbx@AxccBV)2Wo+=$zj5pG`QnD{H(i@rCmUtY;cShB@8lKd-T3LA(w1B-9jN_90 zy%y9f8RL14JT#3g!bz%e14BEmtU9LfL2b!#A}&FC0DT2f9USteuUrTKNN~ww5KDor zZ?fkakOZUg5a?kt;^JJNjfT@Uh|blL04WILg6O7+%MBFJxWbcBr1vum&^0)3()`Lb zvlw*guK~p(5kMJFwt6N2@(8#9)7o&@GhilPXx%sV0% zRWlF!IHFqKY@42}$DVEBE`a`#jo6|jB?2hES(yAia{f*KK)$H3l{gkj+zzG4T^nl_ zs4AY4?-KzC0zyW*a=z$MFNScFhqfX_?^P!cPdqd(wp;gnlW%Qr$M!q%&dg?^Rh~v zG+wL&)^Un*as+_0P5=m>yPHk9UrMC=Bh%vzoG%zGJyA*Yy|SVVqk|dcgaZ{{vqVxy zTbhg9(NUY7SQB2c)XlE6ipju(OAk*6Ymcf)oTpv{2n_a5~0f6)833&+tUx-*yNMRLVD=DlB%8S_?Lb(9s17Rr(oW$YAH*|ZE_HvR3@ zpPVFR-SkzhCIO|kQ&N}ITz1^AWIeA!EV{_-l=7Rx3lMrsn7sG|s2H)BHSvyZrcts@ zyCVbZ`Jb3Yei_Dn@yvU9BgH}`0ft60L}U&U8^tL~Y(Ixm*pu@5OF71NkwR!}E_-mL zzyVj?+^%*kJ|x@NnO@JC>D!g)r70E1co05UO(F|(>wXWVKCZN8vrBI4=X!=sKB~~R zIr^f$Jvsr;0b+>^(C7dKl^zgo5S~)+ttMTm3Zk5ztfda#s~_C|JqU!g4t}d2SZ_^Z zMuD8|fpY}Y&g)Y_aOv=X;d>*T+kWtq($Fi5Epm=Iyt##+uDL`#KOi6s6V9j+L>8vR zj4aX%ty8zOucUWnsh+}HLrei=4ukJfAY65i*s|X?#A+EU|C&b|*f)x?lp;K4vDywm zT+@+jg8K<#AE+496bF3bM}LL|qvU9#UBr@j3**s=48I8p6ja*c2XsGpn-U801T=(9 zspZ!XRL4hD>;_ppw*GjN{UHxl!sSHXw^Kdo`4u<*p<$du87hNJyGh5G2S#yjK*?(< z%r%8k4!HQ6!$65`M^QIm2Ak7vTb--C#-fW;Ho9MT zi^9Cd--Dge`28W^LQ`NN+?McZ4kmGAjPxgX!E(ustCrDm}E2#@^TI8stF>h{* zsN#@2L?++$i*k$^ODAU|fbEg7vPg57uo4UyrSdik@|)$toBy0q&Yb-~<#J8eK4ftQ zh`d*VLUe~^9D)4*G4>W*afV^GWdVg1R)sr-Td<%(LgDT%A-KD1aCdiicTaE+5Zps> zCqROeP}Aw2ncH{f-dXb_zO}yRyl3xoTU6AhII64jq5Gm9(&_TlNj`6e(~k)}BA=)M ziUWCO1q^K_>xe)e5ih7pEYsi=Alkt-4nw&V9nT!rU~(pTM>eK!RZNXRp3w%iq9Xhm zCwTz-5}=IMgbhR}m->u*9p~|f=^J2!m6fX}?^JC49q!ETH_x^{CakEcy78URuTE$^ zze$X8g5~F-SVkWf0G!xwFf%9qcu3Sw%%=+%y~SS5ZVqxzUL;wrJI>tt^XbORo5o&H z_e=5D!GFd?lo8N{vl2txAnej<%XQ@zQ;V6kZnU;Am0{ImE5eS*k4eAs5Lu8PG4I|) zWPoHe98)kVfU#(Jex!0GA-~rryIxIcwc*Qx;XL-4r{J4k3tgI1DfM)~X5`Wt|IK@f z3%spyfvpMIEwcrPbwTV-I@sf5C!a(ZR;Z)q-QIUkw(yh+>9vIl zD<2F}?~A>sl9)YE3g72>oH_3}Q2$AiLl^ja3rDTsP@b4}ynPR`N>XlUTgb;k`L+5E z^{7aVFsLmc(?Q5_`+yVG`{O4%*|AbBgOb=bVI^|=R7vu12j!y<`&m5XRG)F!DB&!S`Y4y( z&ynuLBl$a1-w_2?Sc4p!m9WP zARQ-`aIvrrabYKG|2qT5zgRE88*vQ33cK7*AUXa-NpqzCRpN6^{!cH2E4r3-0m%P5 zxx4+}&|)O)VQA3*sf85yoz+@pU?LS!9+$kk1*(vW^YN8cAk}y#f?sC#GkVLv0q;=I zsP1ARpj0-IMQO_YRIPvy`65l;qh`5AH}c=ugq==c|Jf7XKh0E&^|u_hsGy_%PUj!n zp+Y+BAG+?h_im>;8$b3-U*6TMHvZTdip0We@XET`8xh#=`-_~}a_}`1Zo56LS0&nd z`^-kDDG^6i&yxL^;5#PS6$!sK!0B8F$%j;leKnKH`vA64HrcebBTS@y? z_c*pNmUrsf_503HqTP?5TAU`MUBjVpDxB_T^XUxw%0#>D-j@f_)XaHdhvn~nStgV`dzpSX3Hp=%RQ;azHqagRomq&UVv z9Wf{MZa4htgVzT`Dx(YwLnTjivgm_iCd=p_hsUONR=WSbWZ)HG&hKQ#fJ5C&}viNc^3LM(%60^@(c-XwyOy(a z@0V?@ZUb~YFZ0Z)%COx@^XtSfYidwgX^JAdT&kMo-YNcU(#yOg9MAHITivMjm2O}E zAi5N*dMB*uC1E6*cCY+Xg}$w+9H?=~X8-0L+`Tvkbmy$?<5f&l#|dWa++ zI~x0_-voR*mH88tjEP7(=NsjWhC~&DA6Fjh4t`hogD3QT(M)|%lxzfDiQ2%KRImVY zbGSUFc+bQ#qu~#`0ubCFYOr8BOz`$P5~1`a(en|B1o`v!K~<|nRMwPNTnTx@lJ-d7 zl6dH4tBg?4H8oQQ2RfoER7{!}a*YmyIca7{nWK_}R+QZ_v-OAqQY4WOIdI`c1qcwe z6zS&VForfl5Y<~rnxFHu@y~nE#JxmG%sF3c%~F8B7)3LJs-aUDlw3n#Le8y zg(RavKO;jzx2Opy#Jhs)&uQX$EOefCag;Q#Byz50gf=44Y@(XfFeb^&#Is8_1sp>N z%f4S+SA8u~jiiuQ9Q)(%HEB-l8VoSpSD(U_XNmr8R$gc6Pr8R=Gl*0<7RS%pgR)5XB5=I>CSia+2OOs0hD)Zb($&2E5>TWMhCH0{C=9U@)29?Bt4U zpKreO4Zd{G-E~Ok=yQtm{Zm9I5R%IiNw)TTgqJDdP}a3wV^yfeiqT7IO4vo=`Oj~odLP5k+=r*aKV58H=d*C>xcHL1FHL?%tFc)GBlJj4*i6@HNoz*Q!F@RV;E!2Y@&3{u6M-NgUwA@Poab;n~B5s*PIn+1hr6> zboj`uNORuDkrs4?`fi4Z=122stEUz3_${r1Y>F+7zliHzQ8lRFln%juGVjtN19mL$LI^!&PBeNFKQ;K2>Yt^^HF>OcU z-5QwjibWdNji9gY8E$#(;s(oe%IpOBP3=`QqL^Zd5$yqX%jN_H8ym={SJ7b`mNp+r z<}8eOOT~{4&UQ8mtFif;UAAn&R8eE-uBpzkwRXOh<6EAyBn73D;8?_$)j<@mkkRkt zW*VgHYwS@iD)cAwJia`M4C^l55gnPpm=|DF5xI$ zg{OD9{_C58c<=2xJ-=KT!C#gUD$QwDvFJm86Yqv=3sgrqQVj_R=}D_&C(ykV6|P3F;1Ld$b2pi1MOR^<^lMT{e`3Eu~s zGM;1)&j> zX*j<#8WpG8HS_(2`)tL_Z-%c^jQ7SZ5L;d9SEdu?8iWfkBuL7RPQVIn-*2o~+j&$3 zJ<18OD6pK+4Pl8YDCuAxsF3Ct*1ld(RW$ z(bYvp=tH>%5#;^EzNU?)KZu7IL?@8yruoPpOrm4uU8?cBSR+)&+(L_b;ZE9MpI2@z z1o1lTRgye}n!i|*&Qe!hGdvurMdhvuYqJO|Q~-M)MjZGk$Tp$+X<5$;-pU&!OCKRY z5xUCMmn69f6%B$4=T@PA))4rD6cKy(Rvi|}pzBm6z)deEu}FVA%VrFyN^WsLb1SEw+TwB&o}uktZB+z!P~Md%{`-#tEWA6rG9^ z{f1`Mrz>SR>>zsUT@eaFqESUcM=OP4Qz1$Osiy3;M$Mraf#JKn^)dlv8qsoL$?M$M zuK=LWZt~D@2&!yMg{+&J00~cXOg)28xprK)?x7`wkuHYC2S2kKYkw!V&qsrgDY8T_Lnb*YK9p7pF^Tsb zKdfEI=~n8UN6zE1vGf`(X0aS8L;yCZuzX;_m&L@H>1HuaK z(QzFPWjm!Kwr(&m*VW171Dn!CDW9Q4@{V{L1k5W0idrRN%|SqjH`_CqJqPlpri5oN z5NJ{g2`;5J4a$)!eNW1<;qjr@!a#h=;O-Xea7Q52E=RCjCQD!Z4yl|>4Y!pk;7ppZ zY!O5~#bUrxLM0aXdoNMEg}6xv%tZXLX$@@g0J9wtrj+(KDUbE*&Rkw9C#2SAPPR~} zWFp$fI9bEzzRM0PjSwWR)O^goKdo#n@nM8ByL71LPUC3;d#rE72kPZH&YqCYM;jx@NO^SH!QbVSb{0Spx-yC$Zz046l( zE4WN{^j$p7m`)xDkI@7r-x%cspx&~H)tfXRhA>V?A1u356jbwn4ngnh zGuVYEyp!UL^bKu<)c%9`oB_WC8%YJ2V{9$RE6o!(v{aC@n7Ob?ND?NC%+k*=GUf9% z##Mf*;ohsFpk4!Tgc?zcfOlu~#W7NhGBsuUMUWP(s}q?h0J6t#RsP}{fw7vF8;q-K zpbWJ6H)$j4H8Pq!?K~2}Al%n&`=0UTtQNx~!_+12+FN8|Al0siXdBn=ZAH}iGwIbe zG8;xK2SzjM9Iz71vRYG_P-$!<2mU21PPzG>L=F}OK&Sl*4)ay+aF5v7WoVoOqRzF* z82>9rs`>%2jFEd>8+V6%94d?oR+u(REB->UklLcvcwjJ2Y={dOzT<=gluRPiE|mZ` z&H->(J5jEYbzq&p^#Rz>avaxYAs=tDUX2Mfse>Tc4jM2R1A@imIPQYi2m?A9AT&B- zAYOxj*B!sk06NB%S1z>};jL61wSfxYewN%&Dpxe1z$5!8f~g1{#c9+pVkXw?^yjynB*4L3J-UVHn~;1K79j9T;J3hh-{7LiqLy zy=-z-( zaOPUHt}!Z&yZ34XGz5AOC31m-&A{NejW#R;bd{)q9*gDHD09o7F8?!-c-?`b3h0Qf z3U_VoW=S8n>(CS+yZ}fA6(PSip>G1xZuV2IPKOdg-l*#00I9oVB%l%x*zmxYl9pxAEuMcYJ zbtLKKwuZi!;OP=v zx3FZS0@DvShyHGjh-@*tqy}2?y$(2JV>;O%c`bjDeh04|$Xl81aLt8k?PLPnctvq6<9EoMmgA)X@ zgznAuDU!+|->XY!j+i|o6h}eVar{|TD%bqClw=o|v>VC0xRrx)9pS#ddVINI$UZ@0TQE@4R`-k)KExRQ zx2l%z{Q=7Cq3WzUwl&dUF@-j%^-S@$x!e!j1Zq8_-HTgeoXP!{V&2>KN(~~e)c3AQ zuGi}g=-`^7E*W8n9Wvv zJ_OOp4co>8`@)F&!f=l^02_DoC+n~6?eI_gmV>y}p1W@}3%Ff24R20%`W*i5xaGeJ z8!I00>+G!a>7YA`9ep~o{zNXM<9^yAQQ2_v75;iS9Wm}2a6b1+%A%t5@Zsa*C8s0y zZr7(gv#BF;t36uj8$%a6){lqz&DU`Y*Fuz>;BS?tpN`c_&&hC1`PK3cu-V$$Tzd2)E$MlIC9Zd+T!Z9AtzP(B&nd;fw<=-v%3_;WL;BoD#WU&QT>dKP zl%1?3-RP^@-Rb3hF(vgk>Ln~jb9?Gr&qi*uo$pV+{;hi^ri^|*E{VUMZiJ~mlK&jZ zhZnw>m~JIslZc*?YCkylU6d)G<0F>nQt<|ZikZ&A?p2$~%Qdt!0cA380n0D(eOgm0 z8mu!>0&WDQ+Eyo-;mXv?!`%z44vw)%e;43aW(|pAsn-m|aN1-M{AxnTaU!J3V-K8M zqOeGK%>lTff`ipwWsvY2M2f$h;jM1M0n`JTFa%_bjkj5<#egsj3Q*Ub>VJ++7>#6c zm5PRvWBB3&)XSt(2;RM7Far)J)0i-N+M?7eCUJtD6_Q;~KEi624TgBI+IzVYbFW{$_6Xh;8ApTov<%X|7t2ytn^iil=#|sFc`~mK>^}m8tzOX8bPkdXf~D3B`vb?t@>y>SH#Qc%;}eS zGh$x>pAxRamruq9i2+}_GSwFw;Y19o%zB-Up@27Vl%Je7CPRM_$r5Ao+Kof?nL;pU z-5x~!lB5(83PgN=eXHwryWj36bPta;Vf~}uVRahPftTU=gZP(W?`nV&Rs9E1xnER& zDN*MH5HiU)-<7eKFct?3JlBZE?7c+sB0x2(`o|mm6nL=U>)_tM@0T#R5r=7A_~tS& zE=!jG+>BG!q=?VZh3l7cwXcl$&hVkFCr=lCB+tQ?q`q0egdt&&tGzoU_n#`~U@F&{bo?PYnr_I^rUZ~sYn!!X`6Rp1ab$gLuo zM1o59n-bG*s7aO1WhlH#9u3a^{Hgo6KmlKH2M~4i)k#T#zNe)LfnTX*@g2bi3fMSl z5-*DhfrPPKluOCx%@C$yL&?O?r{m~Ix*4n%1+M_dk!h6h#Tl`Hr#7gN0-KhmWRPC@ zGAh>FpE@=&Wg4o`#m_@Uv``Ni;CbgPUdIKG6no`xdU;{XG~bYL-8>b+lwgKO{=v|X zd0K82mC1ErnrHkAPk(4V%oz@r--w3JN=SYHW}zjXAP)`#D?r zD!#5lM*IbdrD8V_rXC5kVFE!u8;^<*<4&rEPN!>?wJyJ0BA=7=DZ0Yv*`9)!muKU)@}Tg=E#Q@t5) zUOXj_GLst3r%*WMte>o%za=#C&m@8K;G7UjR}u!@6qk|j&L>XP(wS~K+x|21J@w)D zFsPXx7%7ZNIEB*H=O=5!OM>Jl9-)b_YHI>-xesf&!)f&w7GAvu&632p4p7riAvNVmjhE(H1SPVp3$pov1NCW1T!;Ad@5=a= z=h8{Onh;~DCypi|zd38RXfH{@T_FL=2()C0ne+ELmE($T8_-96V*Jd?hxd*cQc8?R zGn1f+7hjduHp_gNrFgmI;(wH@|5Sz27B*3#->o^px!lqZ{wewP$1h#F?(RagjyFzs z`m1rmjI{Ciwa^k_hEH`m;H_zHL5-(yk%~PGKhCt`4w$?IyaTU+s~W|GB!e3GGP ztvOT{dyYHIqf;cBLgZpL^6EM2HO{xx?D29+Ckz^SOKBBvd2N(EdKR@@@hgOpYorhn z|NAOB4`2^4@V{O~#|#wvU*hs*lc`MqaQT+Oeou<~vwHUluX(=XpAgTf zM;yV=$A0vKvG7~vH6fpKzx2|^07{FOkTaY<8HZ0M0in1zRMSqXFVFCgyq)v@Zs=Lb zjzExkQ3&d&jw&mZO_w+jjb3+-Lsd;6DFQ&r6ckQzmIkoXz|<^(F@lmzW59AQNwL^A zH+wdvXgt#KrCUs}1c5Uq)p*3@mZT&xY{@c!3~rd74en68zEIc`ZFyRsq8@$vtESIk zn(oxYw@l**=A$h0nuD(}D_nQW9Q0iL8Q#Gx{B!C?OPks4D}r5icCiq=JH=A0u@!JDWnxuNbyWTfk-an z`LL`Z_y$HDaDAaHr`D2juC5n*q!y5oW5HM+Kqz$DJhCx)3^PEJJZhbb>COkNeX(_@ zy@v1ckq9G1%L0X_hwY9-l!MeZa#_4!ySx zITr)Ngl`xJoe4+MdNEngk^>W1@Lk0I&T)flO#`NEM9I89Gj|`j>at{4gy^+6(k}2d zHs&vUW$o&~21*8Qhv=*1Ar3Ipi+h0loO?tI4Vvb@f|f2KvG|2mals2`Jna%xWH%j% z?WJ;bIg+K(3^4V)R=tc6!=Gy>86TL(W*O>l+ur0V{S-sIKRSZd${|gE_d-HcBz*k? ze?HemP+$`n3MI4O8nTsnDHdNQ6@FYLc;d$9oZP{}ga$#2CT&_t&YI1Lem)wf=R+qUynb-$h;oOP}&>iHT& zAh7#~TlBb4mi0##AtOM?TgmwKWqky90o(il3%qGP&WCH#Jw0D)+W(SwA+6Jc9r4rs z;w{Ofv;YjDD(2(ThVF?aV-G&*fnuw#mIym4CSB4#RZbJCP=k&HtgeRe-Qp}FydAgw zh(t|qT0R%>l`e_yjY{w!E3p`Ex>(3eL-5d#342ckhRkXVxc@t%pVGn_;Bz<>tt z)RXHjKiU2z$Dr{J@y`e&oCW5oICw_*q1s(;-;cbb(zNlghS-k9nytV2p}l)cG{UZp4K6)7T=kDm z=PFPLw>Z$`up}BX@4=>?waepVJ7juCXr8x^KOle$!E%rtE{K}g5~vx4Q$TYLL(uWd zpa|B=k?27-7$gUIAyJUYLp--Ei4lP?0kEUxXl~J;1IDk}f*J)e<~Wdz%I#{7Nasq* zT4Gezl|hi_^i&}M0ljXFtcD5bDl3cm1XY{_wY4beZx9{mngKbS!3Ua^K`B|sAjIYvqXWh7`WQq* z1G!+hZFKPs!XG0nRkFXW0GfreN3#PA&HAJd_VRo``tQ0`Xcj>;^|YQ%PoZI z^3i3QI0h$^twqbJK2s5SC|>EM{yw z5S%s?bjSu3^JNLr?$UwQLl>TcK-!DZkDFmI>ijs%}Z&(*gH_UJr^5mHy7_N=7 zQljDhA<-Kq#2AKiXC!>|$LpV}jx+Z+Tl;yy1fOQD`X~~`alvKo&iiDqf^8sb520#1 zi|Dm@Z#nZ1~e;pA*-M2;M zu3`6fwZe=Ol1=HiBo|n2xA$+r4d1?45xmqZCI78k){B|?cpI9<7jJ4B%{ww1{B?1$ z);heQ1;CGjriy!XzU=#igx+%dt9Z6c+W+?-oK zX3FMG)$|a-0b{RAe*WQP-ngNaHFlWaiEtE?*U@{y)@+~J;21VttaO1!+YLd!_G-Ia zT;a{rIe$ta++Sj{yDtE7(Vt4lt5Q~T{#!@`OE2xib<{hU$I%jH zReKPSK)^H(LcR3|Xn~h{(Zr9E$wbCPEcue$*!U`F{HQlbWDeacJ0baHJ}>6Ly{; z^zC{Q=r@H)wX{Ii&zo>0;hl@5zyaKl_wH_A4U0$cEcM(L7_rCl?f3G2g&6B6p1r%i z${ZoXZJTEc!Vu1D3fx~!?|Q_@`imy4BXls+5G={80e}_7lwv6Rf9@SwplRfumv@x1 zwUKbrsN6lhLJ1>vv0LI2{nG@{ujP1prHTjR|ru32vXm*@Q_%Ct;Re zLr`*jeuVkL$$6!l{;>aUC}=nW6mF`ey#{7OS9Iv$qe%$db+&F5WneNuJVBgxU`2K`)m?Y3IU&;5QJe zK;0NWR37?O*IPs#L@4?I9EzozikSgKJASl4QlmZh^4{VHr#>c_EO=fSBP$srYtA9t zjJq{lV}$n7!N%nLijVX9c-4u*X>fPD$WwjO41HTR}r zlD$Op7RO1V#aa>2}* zU)vmT-+_ZHtt&z!$ON*8Oyx__**KOouRBtpsHs6rywj^t>1B}&KvfKSiR^1PcM7oYI{UTJgHT@VtUHl zN-(#Y$Q_JDj(9eIjuRMHb2ttDNR~yr3fX9(>wKwp4EXsQ< zj4@k`XLo z+fu~l>pUX}6>^5KGijEZ6$o)v8V5<8m2;Ar%H<1P3O?(s;PzWl-Z^ z0dvoFS6zu^HpF$(ABwfv+|wM4O6vS~DF}&UCM2Z3l3Lr+S6&CFosrh3bv2+Y8)2H% zS#y>4KffVP0k7<7F^zyf=OklEG>+y_zd#6&g3Ca`Ts+nF7{Mf3ats$W4ca#OGX#bn z&($cgjjE;0Bp8JlEBP0P38&+PKNuULB^t2AWU;B;Szd6WT$!+68Y*M*?8O>w-5TxT zq4FbqBCR2Uccirq`X34QYAY8A>yE(X#I`^#5`U-hJXFI}AQe89S@B4^#uezyYD)qHe^^a03^{A_)w{Pk)zHt}5*dc|2=Ji`E^x2N=R6`P&J`lgq(r*GqL zK+U!zeTkZtMy!B31#r_QK7!$}|E3z%3JXW;unJys>~L@3c`v{}*mffqNjSn>Y-r&$ z9s@X-Owg$Mm5}hyi2Xj#OKvQpb2xl&P}9GrT)quGt_`v>V6xN@H8O;E)3)Oa=C;Qd z|J?cJW{h^ETlNnTjsLi(Zv%DkxW9cR0U1q`SdEyAVKEB<0>c0QdDQ!ZM$KPpz=7cRKIC+AFNqZ}ocrHnQ*4K3DBmxC2od zwmUCX7(%?1$g?;lo5Ayf1$FawKLRt)eaoi1f1At{(fd&<`IfJ=S4A5c)o<9Gu8K(g zAb&~Z`aaV2VR(C{*Is9ZGT7{OqqplOyjYCt^CI3$Fd*%wQ z08E;b9R&c&<4)KcDQlu|V#|X9_iKxjU86}Z?;nwL@6C6^=_}UeqV=m!Xk6c9peAc^ ztT3elIylp`q>|W{O5;Z@P^pt7_0wp@M8_oe?bPun)e%wd@L^&aeJJIrtbY-K8HTn| zi<$3&nJqKR)`u5iT)3;C?EELZOiu4WY= zQaz0%HiS~=oA{UOCnm{`KiL%;c6{KfD-}^-sj=QRwW68cW zxcg}pBu!qoB=vsrtfA*j5P}H&OHf%~b~__p_th_UVorzF|vy+_Y38&jn| zcWr?r+Mo1&Yw=$#o+x>75jp8&85l&5*DMU5+xU|)9I<>C)Zz-Q|KC>16x`WI+F^Jm28lp3} z{a5+~T&`zl%BWb3POv5nRvsfdilJVqla@#dm=46n@I)|sG)Rn+KjDae>_$rj)8C&<<{$2$NeX~s-M-|ylCn`lxK z|D_&*dCnV)vZ9f$qhU%yaY^#&P=Rg_w2U=VhJN_EEfo`j3Z`^w;gV%D&=_${63RORLfpL%Yd##FMgQ-v5N8AO0%+3;iwfkzFApXf1-L3?WXR z5D$#uQbwdQ0sA1|lCokS^_EB-hquNB|1kxHY&j}w6y4aXa=(bXGz%`Tt#%%9S|?i) zgdlI;R7we3D`f0{5=1y~%><_(d|}ZWaeuy0i@qr1sq3Hpk}@E*4|yQK|;AG#fd3&s{0NvP5-qR0Km?G4fxF7URdVO_Ybz=1-;swNB+xxZk2yX5r!o#keyL#NT5SW~2`s z{Bw*;Iy$CP7n1ZpVG_Hdqs*=8huIT^Iv!D)uDfb?28)GF5k{bi#Tq(mRbFUejvBB- z?@PO9)1z>CTUNey@U}V~62DKx*9xOY?SsRsWiZy zrC9>ot8uW0Fh{PO9IE z+>G%{TcifG=uvgFe<6tH(7d>5q5oRn-k0L+<+E^jf5r#5#KY!Rw4UA&#PR$9OF*>0 zA}+;|drolpl?LM@wI`Q?di1_{ujkO+nv255n?rt7-|66b6K=Dv!-BSVla3K1w(|mtdYD*J@Xc5Z8qkj{MEfQm1kUXuUe!ATbJE2EX4jZh}k9PbaJ`q0rIm( zG)=+MlBiF2-C+8+Ng{ME5Cj6oo;hTWr9S-jOSG(>lIpL$ZIyvP9eU4C!jXvKk0@Os z^5QhqVpx69@E?mOAQ7+j<~0zO!0Sz{aS;3TB+azbC%Dn|r%dg~Pc);RRYyt(i3Wzd zk20LENO4X^yC@MX zs!J+%LjIl!ax5)K;DVx<6h!@qa}N&rWd(df{-k%}*?JS<-W48H5@<3UJey`%D#3<2 z%&o8l)^x#X3<|2e1_50_mCg}Ux+H9_QM%rr`U%t@EWL-Vb-8aqTF%JZkMStiz~(Ay zcQF>-mJq)dRKoX~%*B?8L+nq)?XI9)WEo01Vv_NA z-V1XjRA%_i$}qfNJjSNqMXk<;;;{%)lw7Goxp7`DAb6U5b+|bU#o$p9F+); z&N7=JOgY=h7fA~iBEsjeP8#VFaX-!HO|zv8Z1Z! zY%53GQ{iAiaI6QuRK;<)t#}0^U(Qvw0fB}GvgsA-yn;3Z(WUeUYO*_`wwe{^i>h>% z7_=a1;Iq|b`Tgn3XDGCtid%Z#qnH1XIZXu?^O4s&(#fHI8#mnp|t@%^st!O z$`@Qof$Q~yF8?zry3^;R?$PqovgTvZ?bnB(=x`zmG3Tkc>zd3IH(=4)155d}X})d4Ra5UNZ95z@729l5egU&; zDCy~#)%}9K^03RcwHw81{e@+t;EP((nOmM~1oF2d0)U7IET2!Ec%K#PXgX36TBd2! zAP8@?B+8Q9J7Rk>`v;g$=;@kqBO=J&x|3NUchw7>gV9twC@|FgYm;nh^_0Ca9)h~F z>>vdAoX%fm7!)L;4NNuvn=?OH2!a;u^i#iE+ym#q89nqdSP?@fg+WR0sL@}JKxGFI zY7Imk=sD@^Too|KpzB3A{6t$7fUOFQT9(vfXszMt!Mg`yMn!KP1IAv!jqtC|ubc&# zntG7o&lx;;ZRbI1YyXt`p#lgBzi7i>3S9gw)^c?Py>T`b060Dvp;bULpQdz=ypwbJ zb2|X&59nb4A^RM-Q3)Ve2j>fE9i~^aXZ1^`+S};`0TDi9=>oBe09k&>_Eo@og|v$> zjst~0I^O}b>Aru${egiCKYt8ooMpO{dyF2FszGs!6vS&_pa1~V9@*a7UtU$+WK06N zM%))|xWWCOf&bO-KMnk_5!3+yMpFNu=cp{u$Q!&QNlow-Y!u5G)W!<9az+^I1a#L9 z<%|ri>CH8GiKISp<$#kdDcuEfo(+T*NG z&*ZJo?9_Fz40avVwH{)!{P&0g000Ai0S^8jjRovK<^NxT|9_1I(N~54ZDZlpoaw(B z3oIX|+D^ZvE0@aW%m0tYf}NZn(p5g4eo~!OtC0eOH9BwoW`Ui;vv2M4imVR0lZzxL zX9J-B6FLgzL}tS(02zbKXv?GA3?Y$LqTo$8+kV3<(0seD+s(46P$=|X@w)JIu}Y)X z$l<9)`{TX;aw=P|j>zWU=iQZjA+Il6U!Kz~l|Fabtm*=)1iC@LZpyNYu)s;md)vK{({3(bqbLLfiW~!XpWNB7bO*)FU%m>?!tkA6D-XihKEDJcW z%hUHfI{jYOHg@KfM<>3fMR4Eo^h-}|^*W*lReh-Ri%o-Kf`OwNZ2In`EKy#tylda| zr9nBTM(DCPe(#OO46!j?Y^j33onh;?`PaPE(K>UTPmKmaUp^f7e`pj2_^Ixk_b=M6^n= zuKHT`ATZkglI7>9S513;+*er9cf&GRF$leEU%gLiS6@gg|B$@_l3p@h*4EG@wsKF^ z_WBNcuj#g>FbG#tV*{qs09gIq=(~0hso9fQBk@Ka#Xx9apIudT$8PBdi!UA3$^Qpy zZy6N_6Rm4DP2<*$JB>qdhXA2*ch}%faJR-Cx^Z{6AOV6ya0$URf(8jB5S&QQ_nmXk zy644|js#M; zVtSXb(>8U9LROubh!he{&aPi*d$`)tn?dDJN~6h!q)2jNw2${7zzE7?F-(ATBvh+& zc&VWtIvSdSg0CyYAH~B7W;7zE(ePn_gH$*Gv*M?myAMBt&sW89 z+k5gwE(8Lz#(KQOBD*Qzv(FM_ke%Nsr1d#hT`xiorY5M+aU!6%ly-vi(=#s$kzTY0 zhpbrpZ?xV+QY;G1=>AO19^>1`VxQUx!Ug6`OCeM+$~cAc`cjhT0%1DBP_l$xrX@0h z7uwslBqZ^J(nnY^cb9flb*IWK067^T{Og>+95cmj_eqX9Pz!&9b?t8(vwmXQQf%3;JfDU!^8=tbD-RF|d|KO! zIJ!eGB00*VsPq|KemVVTi1mvw6{g=DNg=VPCC&qm^4ay;Xx(R`wD@0y>P#AKw^x(X zG#owqHZ_c;g*>ZC6QvP0N-WO%3^4vQst<`E1$)7Ig43_rKShNN^j=2>aI@E+k4*`& z-7-{RX9>IGfLQQjg{D(oMB5c!=Qgg{v=L7;N0QE{>1DR$W*>B(RS*bgR?!{uYf41x z8faoG7ge#aPY9B{^J3#_HCb!yiw4a_EmqMFXDU&Vpn;%%Ga1N-Q~C1g@gpy9=)fC% zfh7G2JI6Owk_1p8itZvDlJ4hOE118_UQ1Z9X->-&uT~2Q)?mb5`iADi`B{Cf}UpNwyKuY;O=r zX4qn}U1ohz2BtJ&sWkH~Rq#j#2`+%rA|S(~S?zu`XQi6I!33~ATX$!eaW zS22>NHCPA#9uhG>M!m0sj9}~?5!1f>z7Rw>Ly28Er<+>wA(9-<3wei=hO~vo8L3p_ zb)3{HL_y6cRjQdwjyVq>mo-NjKA$^Ulsme4ILn)x?KnF5#`T7NiXrF;QLXyAIxTI> ztB*`!nvAlLh<+QO)f|~Ui^^o_CzGg_l_D?VAFB;AI7{=q$v&HZ6Rq3HJ1w+BNz9S4 zyl#4tkO_Nk?_3rizGZ3@_UcaKYY;W_2fvbt57f4LKrLS-yTy>NEqh)r!Cs$B`CZN2 zs3;}G!_FOk5PQzSep=8tC>QN!da@2%_G1U(G@m>?um0i_XvRpghA-JYdX@fd*`JOS zTO9*HAL;(1l>=q`VMC|8=Tcz@DllFb$t#EpgBkR}C=|S7;M%aTmHe1ry9^*1gh9=994j;<8nOU-{Q0y|6I63`u9^Ip9}* zF54EzYk%0-B7^qhLRIG1Bk(TE1&_bu;;Q@0^t22&q;>ZXrcN*ZbICcuvh-Er5XO&# z0$ul!3`X5w6`SQYWr>Q^udm@S;GYm)>#tgW`6w~^y%QIPW1S7vh#*?*$hI_H_o^(# zHFoJ1t{)*~($eN1@Hviz`P;KBiw=)COW_3}zk)L9Q1Btq+z=%NowXI)R0ggXUcX8V z)i)%8RH#gRV1#brhm3>8N#m8xDdk@~fvhsYC4zi9q9oNv)4{!uIhNe2qxq$nd&=+; z!h=hqWjHob4uh{Z18Cn@`QG=_9eZdUA5oAl1WmjI1nRmc`o5}~VdMs>r!u||WqA-9 z6%)4#w79jg=Bo^!wNwmyQ_Z%lMez%4o1>~g46b>WY z;U|o!CLne2uT{2QHzZZYBt907k}`fp^bo1;Bz)lI`_I8$Ax2op7CmHbL{!$}43GNSt_pA0*+Mg|h_St~t~5S% zS$MxucrZ-9BS~3fwJ1P(V`WpZ2D*Rrw_@3p&VI|thovEKY_xkICloJxZ2*3ssEU0_ zWjevBRqMv!tra*;+JyG1uAO_%*750QYG)8^ckHFEVmvxlnn8^S%$BZcGVu+o`M5Y3 zc|x3(NYW(Y;MtL3cQ17klyZ_EaE|TERL@Y4t!&icF)yIQ>=gTX!**v?io^xMC;TE} zN&(ufjdPVUA1}J7IMsF&})=tElRwDs)rEf zffXOX&$irV{J|gUlgLfM2`-a>q(~H^ z%NAz45ErU~8#n<8Y~XM<=jJ4c9e_Tj+o{ckv}mNz9#K@?R@66BG{gz+92OAZ}4kbO9K_3GGgPkBc0+di2B2ISSKO2?d`c+OjVL`0W0VMM%ni3FPp&}rK zzOE*@O_Hyw+{aSmo6q&PO+KiCl<7%tPPn#Kk?eOsQc5(;rqG{8NtUyb)XY^%u+Mf* zmh7|vqtDNQL#YmQCuq1$VpAvwaNy>5Aw3PQwiM53m4oV_LP*V?*CC$jB%=!UP0G=c zb%@4T?5{QI?X^_fwSN{%B=;(w9BD!APU}|qb>DbEcw+@Wq*tCKUSdt8lBS@yT{($zIi~QSxLqk{B13nie*ocQ>R|MOPLQmp0aS z<}lMfBXio$h^VY>d|r95)TmxzA6nS(VVfk-wJEf(^>CG_udx!SmN{CV>@QcPgv$bZ zUjG%t9_C0i>&m;TR-@XUmb1-{Sy5+B4YeL^Aa#8$-ri#C+FF~_@}$y8+i&I_4NI@9 zm=$YE$8Fm0;Z>i0tshed^s4eHe0|75Ht;M`j#}p#%G~yblxFBz>clpE`%0_$XtREw z39P8(&vwVSNS!>TlZH+Er`cLM>Q=*Q7~@G z$lNLet5agX>*bCVVZY_mncW)*3x}#37EObGd)uq!Hr1VGh0QQsvsUUeQu}hqMlcdVzNb7GMc{HgyV{cBWGp7#dr_eJ^efnz5 z+hB|sETq}5`bz`Z?cbxNzJbvEy~I;=BiX8}Dr(Oj`PE;+NLDPLTXdb1;VoQ+mRDWj z@L|t!>{pFl0N7r45YKy1Pxmd*W^gBE5I8cpf%*#R56vglEjcuQe!|7|A{p#x2&p}2 zG+4qdkEKYx76S*;n*=;!t-JD z-y=XT@w;!hV5t$Tu>2R>5p-d-CZ91;=@IILcjV#H$q{oYh=n|BKDxS8wBxGr77ar4QIBUiO{ z`<#DB4oP*dgSkNa{&g5Y^Af{dRU|X)je4N+u~2r^s_y}KzD(Hw4mNuzCNuQ~d>dF; z3ImbB>At4Q+lbB7RDyqs(?JLrH>)*^Kf><7agM_3G}4;b*dxCST!6dtM3?iJBMTqq zLB&eMdv7v>VwKzTjfUJUicpuqE2ImeQ-ok?vcKT>7&@SY)MWxFIi4oO?RT;WUfW=& z@7DvOLJWK`2I$b5NV`Pu4g7@z*D_l;pst)^*lT&|BdBM$yc&(zvvX>NNuN{*assd5PLqsF5+;*_4hpEBMf z!n6i0x`Xi1HVvTU!S5j(Qq5bxo0zh^SN1lsYg3jMmjGasl?~-}zqRVaO{PCMte+O{ zFY!435K8)KDPDl9QE963G;{-Y*`z@X=z?i$Bo=d`n zurDA@j8$3^H-QPlPZ@lq8dSIz$If#JWb)_b5m9Tb>x=~JEVzvKueb_ZtG#gvcR}Qf zAfrVP`Q`k!^jNUQxV1)#-p>w}bLsX;!Wjg15*Q~5inENszR*Or3hx*?5fSKsBtHW* zM=GQLV5`MenZSwm3Z6ywf zu&W!D8sYF{XSvG!=Fr)B$s${x?8A_KT6W50Z=ZJMNbs?(qB;$)m(-`fkLzbh_#3g= zR&A1>pHT}x!2%3>jW9P{``LSe^{DIyt!wOxg5uC3GrZ0UH%Xxk$hRV=&O4ErJG=3p z&>l%0$mKch-uD;CZ=y@==PS2BSr@(^J%LVhun&s_k>givZ}H{nU#j#tXhY;81!z5+ z*ovwYiJeqYci2leo;0a98`rO1-;<4oT?(~G?%2XjiH@pvaAKQ~cj~TcCDO_(q zsacs{U#DyD$97B#!+=Z@$l`js60bSBcNircB#_ZRQW zU%-|2?nvZ4h*rw^!^^>lTXNB`6GVkVBFUenZ%#yeO6hdc5g;nPLzPTc%~FK|;=SQ? zR0M?ffw*=qUkJnJ=J++mScK%S+rv|m6Ya9wfbF06>PLFDnk6sd&Q`w~N^x0l4lq8bc^mj3b-eS;ho=f-Ww`?E2 z?b_dOrKEa}*`#}EKb!M`I2%Vu$Z?z&W4nXG*w1;nN*2`-Xx4Gm`PXE4h#P+7zSYI< zvhoUN`SumsK=}PSb(o0z{zMvsa;e3m50ZQ9xg{N4#ZByd0Ve2w+irk=mhb;n-u8dm zZvSS-;t=fr@3z~&*>S^4ozDNvj{je5H_dGymmd|w18{UV0_cC(ZozCS|7E+q`bFf> zG?UL```>Ig>hE8ctwgG<{+I1GSE||HeX;Y__uJ=BZ(kqGx@ljtnVCLZ9!zI(y<8aR z^({w5{3+tKgx$<;VA;Or1;2GZkgn54-LwC-3cfm7`G=eJ`kwADza$KOBzlthx$?fL z=e~3}?2g`py)<;AK^YnX0%8+~JWL!zLvVQ?c4NOOZY~A=(S0b3qp}g(i$Z@jZIeKq zE6$|M>Kat3bSAQCoy=v+QXUW0W4B5btW{x5Ck?Kzijlgb{F1_kNrlWHTsN_fQ93W%M$&AQ+0_C zE0~=kNU9`>t=Rji!KzTujN_!dyi4z;jyiEYOE%u#mSt)7q+v!l{%5gTxpz3n=CWna z>e`LMR+g^T z#1|YwRhb6jf@&A(pb_sp((dqgENh zjWO@=4cwiQpucIoyrH+Q-&)3Q13u!-T@ru%;I}PH!!R7v-9agvNq>JiMIZRPy_=B4 z-fJS~jIC{!Z|>XAmD`EmKC)!Yzl#@5Jd3+zKO1Ux!u}!FZ{ByJpL}GT>S%w=EA431 zMfu;Q_)OuW6TE=|UUIvyEli;L1;@0|zw@ksbkDn%oE`8<%Vgayf)0ScyPP*z%zuf5 zUW%p$tQ0m$i*FnB(}{hQ_#vh9VPNUg!&#s&-P5S*F{Z$Zj$3Q5tolKUa467eFr--I zI!)Zvj>Mn$*s;h%Cwyhg7CGc8*2otKEFjG5VA3bh{~XXXG|;{Bo%^@NgrsVw^e45j z(Y*WHpH;ut2qy6VUam^wdn?dtz3UVXNln{=%q< zF_NqJPFr1ebF=8GF@a;u*rpGa{$<$T?s>IOQ(mF6!Fhm=!mf^F(J-F z(~tXnhe@BvlbA)XE-q51)R?KUe5k3JCX%n{&@_+NmZC25(q`!jtqIK37F$;qseyiyW2SFTk4nGt!9!QLpP|QT^~2} z)-Qz11aFhPoXu9uR`u&Lns!X9l93TsXXzv7Qv!pm#fn=$`7m3Y^u(rg)1rm)rR-L} zl}~M^7&!_s>~Pqyb8SVyZo)WOPh=`9l2rVRND3Eo$EhFk`um7#?^4}fNaq@ryGOO$ zO4grwJ>SVg;ij=<+LcScLBqCwSnq)&CS9IyHqdV?_i(~Z5FwDUF zU*^_OdFs4IMPf9ropdA7z61g)Vvqe214~k;UvirZxay|pMFhq9WS)GL$iy#fRl~!I z@uHYUh~>)eW}h6lkeqj)y&=`5Z{Bl;c}fyH9=*5u^vToH_u~!+Hf#i4zymEK^kd8R zf~B@Yt6nh|V!>u@S|`BO%~WA3VzG*Tb+ylH^nE;-M{Mx4272odA{z8l_E>|7(D!|gzhF8R8FH35@t8#pGCNm%=@lecDJjd$bD-)G!5xa%qUd%NGHy)I@KoEo^w zr=w394~%rvfsW}w$*+|4NX+_Ac>~wOUD^CzKi_Gl?r&lfAgo2JNR!G!hgko0raN-` zdfm~@8HS!7`m)4!t$=EavLONf`qNHG8#>_dikIIcs-GBZVJlPAJag1?DMFsoAgjw# zx^%MLO#BUfcI|7rozZk|Pjt!UX#t5d%8t&`?QRt<5pAOucgmvB)+0skSDmc8PtE9% zQPi4}Z~%U)VVf7UiVVBG>Rj(k?zUIe#^-Q8-mrFk;ds&L0d`{Z^T+p)Ib5o>sloR* zlP+PED%T64d#Rtrs}y|rtFa_EN0YtVqUnISzH!%BpYYl4kGE;n4xc;4v{hq~7@WX>5fR^Y7!`5tQ!*Hl z2MLg8)^-o1qOH)r?q8fpc~z6wzvCSd5`<7{F`nyNJ(3Dj4xu>Q7V90?PP^Ger}snS z*ySL+yFXdA+MoVWy6BwJuRf%hVP(<#_#tp8ye6anVbbJhC3_VBz~EmL?|T_jHh}6i zhXZ4VjKRDo5J?r%MhK*qSyx{d7YK_5?&2z#;HsF!s%*vz z+vDBb+3A|#>6^p>W#iCoV%1eBS)(+mmvE^%DPWlK?ZEiI%eW5Wa?JL4f8WLD?ZpA3 z;*B@s|L)@+Z^Q@I2LYk6G!^gw7N2Tks_)T@@~>gQ6p~>1=bQ%Q4fjf`BuN| z0`+X+yp9UtS6l^eAqT#ILefCpYh=kXcj| z2H8a5RCwrL`=B)tcRG7}Pd>*;Bw-a|WhH^EVW$!zW!{|-rN|*;5{XZQobyD$6i^{4 zC+FYfX=BG6b0of4wLG^=0yqMqgyfLPz91C_I=hwel%4`=gF@*N0Fni81Z(32 zlN{L66M1{r?FoP-z%5?zCn3V$6P7SA4x2x*g)>pC2tvsVjM=7Ov?XvC6s!+q!bWpZ zIPev5N=5~l?Eq~5kcrB2LNFdF?5#nBFfisw7H|(^Bo@K=NQCzYl^##QhVa+3Bk|Gb zw|8H<O*B z0e>z3Jm6j-uq_89fYF1{W&`E6D5W)Q`nAM>b?s?u8uEw8@EI{djav{ys^G=REPNOU zTNdMom;U#9t%tWo&wRrjIr&#qjg~H5MgSn#2;m(HTs*R<589H}xB|cl7ZRt5)vE-uRkyUHe25k_px2{DyT;MifGP-r%(qG`XgBpigCHLKD-bx_o=P=L{8K5Y?B zf#9E#KnJjL@SZn6D}Z1T1k}N>gaXK*;O{V_M}iG%V#>yp-sCoMo{w$AmSI5m! z+Oq)dy~8fv3U~DtSqU2b@l23m2w(OC@kzd3AvtPn7%)W_oQR%F#Z%L@g zsO#R20gc#>-UedPheT^eaiwzYw}~VYOFPJ>j3xlBOY8a}_5Cy!09l|;w4p2bgdNb{ z6iu+#x`0M&j!^xn5R)h$h2a($6WSXUM0?*^Fh(J*q}Wb-ZGb@`thUWxo);$T+tM$~ z-4xFUsOF@UqgNo-W<@Eb>~r*` zjD7#;L}Om5W8l}n!lM})4QI`IXYx+#w4GP)I7FdN#Q+NU&xygi)n;FRQGvIeBL(g6 z1O6QjY^(!TZp#68b%PR(f`gSCjR_AbB|lOvtc`xY2BCWc zls*F9?BuX~q`k=Iy`jmE0n*+ahP3yBHAi~BI?`;e2#r#7|HV@<3NFXoYX*9QfUs#G z6b-X5q64+p#*Yb!OKI3VF?{H|)W zwrqKtL~wXzk6}+T=!do30w)-$F=?GOCLB+rQaxkaGyr26;X#Mv$WAMdWFOd;H6r zL?gP@G1(}KK_x=2fpNU`Seib2&E9Nv(*Itc2L=!V_5nx#4;c&i|KE(I!PPc`Udbmjr!nI4%vdGIA<|-?6aRy1L%X$tF-| z*Q?XvuGTxp-BGzjXITlCdgLr99Fzy!%)%vZCM-8xkrEKSvs6X7 zz&CKuM)8<5FpuQ9`vBT{cFpHhr+E#Ne32F~?>QFMtk#A((I%~Mum9#V>>KJLa50SH z;sa8j;uVcqy!s!?#XDSaX*|yO7IB1xLEsn9n{!Ec|SLdPBZZ1;X+($VG6a7Pab%$*>|{ry7;}T zJIs3x!>M-sbDJ)~@OZBn^RoAIw?)MD-CXyA#M3F?`tR$NF`4JL3LV4UfYu#t5e@(A zHz$-Ev)dQ6%CPNQ`9MyZ#I4#~srvSduz2-Szv7=mC(_y>nFlm`H$5`r-d^5WzQujYU zSux~3kMb)Hz;khK_msbW^z1?hN$6!6=6rzZZF!9i8Qa zg5!*LT>7Z)0sWK+#PuSJb=8a#qYcIb96_A#%mo>lB37}}=-p%O3DO)V)EDM++el#K z-tp>nvXk_EzxKXF_fDY^LS(eeZJBVQ)%F;d3-0IexFkUn%1XMqS1E-u91YTHKF!`` zAR2M?6GfeLpUeh#OXw1ZCMDyzQ*eLn;#XB+&z4Tb!PR2kB`TD1cz|aXY~BtWh&Am* z{3CJ6olOoaSYb$`6MC7wf56M}mSBG|`7~^PCd^xo&R({vccgfo@l z=|apore~!-m8{LvIB1#FGq&Fec^K4LIy>ttWtpp!Lsm4xF=E8rJAt@lBIci90lt!r zwn8)_%bn1*`*Z95QS* z@yD<@nq6<%?leyRowOz_xEZ&YoTON=D2gN_ty{isSLoc^t7|t?iZcd(_e54#2J&LSHw4QAu(nG<#Ia7GQwHx%dCQpH04UQp zMU^+5AKhy1kZtqUVeGgwd)ONi7d+_^r;Lc>_3q&nR_sLsVe5OrycZY+S}xwd7#Oc| zIUB=080J(LaPF@<&kR-K10@jz_jhj&FJ1?cMMH_QHla~#Ghv*wJrBz|nK?H)iRt>A zmp}95LcYBYP#Qqcm2GfA=wHGTnnlVBl!)#5GaW0?g+ra0sdTiISTgsF6NFv+t?({o zm(^)+aeLip7kvC7_>DCxH}jOK&EYxRJRVRWiqZ=+wm8a*GJe%o{0M3L%KDM~z^Mw&>j8cYu2RszPJofgI-*eM`Z=+Ukcb5ZhJJoI@ z`UkUZmGLS%I(%yY5Jv{nD@d8ZIc_nhb4+?E`48sYY4S$D`W2QyJMH3&rz(aq*`Kgx z*J@|j(^f0yWKk2h-5u@IHoibRY`EFkRJnRmrhqBs$VEQt;|4SnfGM~-=&C&TXltkB zmh8+k_Cnr3ZZp>1w};q9{#XB^IfCNour5(sLtL{I2MvwLKb$aASrOMGUW zymoQdQ^a&eG4`5qA>h`gkeR^x&Gs4`MTzlNME&fP3N!$Wuxc$yKu%zWZhjoR+rZ8| z6xgtf|E7{tUyVK66YwG>M8mhRT1r)l^*ktP6rua!O_N@m_CghG>P(xWOABivG% z$MTp}>Q45}iuf|&?;reK%EFi*+Q1+F_Jnjk>+x2GFeO|_v;9^3z4|JIk{NV80|PN- zBCu7VwAee8;R7^sW5Z}~BPn`nQC}WX?-GiIIRE7G#|?_>TCkfaE&+g0QeKpdMljIi*L4yK{-X;O}lKy!F7O9YzCz6y0 zC09}=do!9}?$dHTBm>)+x$42q4@_2>{D_l?=2!+NR+5B#@??8KS!OkBUw1{L)ahwm zc01~GD2ao=oK&T09UG-8x~^wnni!?&F{7b&reIlaJPDf?yD+`ENIXLY!(qEi05XO5 zHeft1?T8JmvY5hq743ab9_0`}-kotE4vut6r?N4m$qfXKB;-Q$L*n6ql)~tvY5eUp zUG?F=x+yJgLUw9Ijt(;X(CO-PSeX5Jglw7oDq`%XV{i)KA=oP1?KVg_HqTb3DnIfS zBNrhf4x^l;3`#7u$Ihm&GnE^NAlAj3S|Y^Scr^lWlWU;NiADrtXI$SPzB~9S;eZO^ zRAj-iH>@_@UiAL)#9i!wa;I!BCFAn1u_;2n*;pA3@?t>7oH8dLPaLOdI7k;p_4GDN zJSj_Bf;K0g*vNrs8s;?ujZC|9j=7IW$qLEJ3*oF_yxAqXfP0NXoz8T#+3#Jeu(ExT zf_{N6%GhdDTf}{KMCq8IyGx=ZuR>)kY;|mk=6xZK-Y`y1V`dv6Oh|!uF81r%LLZ4p zzHG4QX^2E#Uc`7j5MW5fL=1V%TO!Af2`;n=&W0nR&8zbdYZ3Cy8LaJ02&!Cod0|UH zfgWliJ!vm3z}TiV33`wtZ{%EfB^C?XY7%rIRZWLo@r zEpIjn46LW?>k8*^EipY#my01jHl&3z39bvK1^T(~B30#%uU10yTQ#SaVr>f zW@ELWsEQEA=OjnouBM{7nRLsWdKVZ~etSgbNzAgqdUg;?OkshbD#Yul8nf192~0_q z{!pJm*$|e@esezZpN(+Nl65^B6HlOFg7lC-8*zL<`< zd{a#4qVhTF@S|}1oSM}j`vF+?{ryv(tIJ4jPbqJBe&@Fh)o>gWb(+?Tp0~&EKEMA} zd-kqAs*e0)U(eM6Nt7_*`JK)(<-$+P{9u@n>qw7>aayLYL7 zO?oKY9~AQFSi%t2?{fM5`iE6(#O3ZZO0(_@yv;$mFznX*o-PXC$LO=iw~G6ZA*&+J zj8<1mf#rdB1#w^EpwV4ffWIOX<&huIjBFCQqiaf{>9+4}QiS(!tV4PzMk0IOS^K`%FC8^Ge%R} zN_;KEOfzAE=T=}l<-4y3gL4$jPIF4a=*Me9;BqQXW&W7KU%iG20vIY%iCOY2JSQKk z9b$;99jh4kaSm!jw=oLihFR`TWh-!n8|7+n5zlIy1xaB|?zO0pIanvIK@*^dSom8j z2*YNvW%UTRDGq>r`Al|-x?iVbhADfbZBc2?nF|OJS?yZKWQ99pk_B$2K3G0$as1aA zl0Gg~FX&rPG}h)hsWdl`9h_&=kpb<6dWK=PJpb-?RLR*EZ}GkfF_idNRP#<*BHp|| z^Btb0`6W~Ar4?vRZ*cx#C|j~F&1tZ>%M zU&~7NOCd+aF`Auna0-jTLmfET!Ulo#6Kx+L5+9GoUh1WG%H2**Ev#E~#=M?GL~&@(V?)l3Jgt*OXTTgCvt^K8GA9cgmA(=B1?P z9m*Fal`mn%rDLDZ07&M)#xo1{TmZ_nI{SS9J3z$0wC*d^oNx7mj^O|%)yscqdU`E` zIWq4sTmL(eIuPL1J5gvTT9k33NFe6P>B93#bE|8uo7dHFCTmX~8`!}M-B!x`Ed zIr8x3z$#IxdYV5_h_(#&$^@LVh|bEDsw7w$m8w8`m>jNQ4LPvyTjoSU2}zPC_wzF) zc;D+{rP^zeCURf~C|fA19%&ny^3+HXMA7h+imqE~Rx~8#^B*~?%#sq<0ufznzTeBEq_;6e$Wv0$F+1MShj+y% zQgzSYn=bhDc9!fvO^ZyK61O?vnK$9%2CW;RGJ}?{%Yu9rIYCT1(S!y746HTF!qP&`3MOGZ(M>c(|A-(HL5NVcAr>r;=jqQDJsp1!0v=J6e!T8RB_d~g$5ph&E zcj(?hFdooqKU1}@xY$#nBgJw96TxKAK)wJh&RjC;85-AwT_BQYV)#vZPw84Ca+vfQ z%~s!$Ok75?r;?Uh1y(`t+LT&m_cGxWa8|AM*b!G$G+o)p z-PWkl-3$H72tmSRUrR|fp{ymC>BjV>SuoW`{pc+5V$LoLbQw!ri{n3-Vof9SP)|JZ zQ)b-4Fd5qHa2w>{e;Td-_v{U7mWj$RuWSB5Vys{sPoS}6PYNu>5c~cWdAnB9>@kwo z@k5-}w6sLe*ElcRB&?2K22mDly|M@L+giVc(?Z+Z+P+Af)>dLK9UkQXwr9Hd><3 z0;r$6gYl_bB{9Xlo~+>uHD4-I!#{OB9FQlmc@E7|eal;TipLo7pU}(f@1`ZG5AKZX zXP0DcVD;QyegW+!iv4znSoCK1Um%tI`w=+P&tIpm@h3VSA21N|7CrLQTwqCmBGE5b znX;oYg_3xuAE9Zhf2rs2K-b>)wbfr2Ld&X!BcSAq^>hRk}z%u!1^owi>{N}o8M>x#|E1s;WDpU-^>z1E(zSy%q@m&Sik`rJ6< zueN)Dge~4yo?(?-=&fUD=9FST30YsD>DS``w^5ceim-DrI0FWCc@%@vT{xl#7G)4} z)Dv(Lq}pw?WaqsZj+dG(`+!^Op6_X!hXt`$sZ7Cm16 zx}3U|fUpj`7btliOtd7-ibQf-JcM%+rY;_06PJNKshH9N55iWqDS#f8gzcY{XOi5o z=`FD4yu!w2zz%8O^uR{*6TNq1K})c`3N#I}2ZFty*-6OP%Csih5{&X08;VV~hd;`? zWuif_Rs|V;%0*RC_H9BT$HuDcx`!m*&Vi2Iw?<4+8gatp<=!yze1>`~7-M`S6Q=o# zc<`w+=QHmYk9KJ?O6ffB0!nqg#nIDSR!Pt&)hgsuUU;kf9Z*n`rrSlO?l>?mOdH|j zxK@B(-tCKDi>qA)u_*T#s2W?`bC~U)fI+ny>AE~mK52ezq$v(!_G?5U8=*|ovBt*F z^!PFNQ9(+253a@zlz-a@{_QHlpRKzXDHBf#^rh|@htB&G(R@uKKZyDnua|4ca{yOv zMF%0hcwXvqUkj>U3xy$M$*E?jYB4fyDzhqC19SJpi69Jr+6dZKBvie!0|hi16;r+= zZRC%#(GK(KfYP>#A$?B#+b}6#DtxmBs$2c=t0Su?-K^z}^hS2D*rR&q9;?+ma6*)j z?w-PHlEM)HhC>013^F^3IK9(3w;eMljwRAqA5lY@p%t6|&R2Z8T%O4%+xpSUE(wVY zjybdu6WEkmL|WiL$O3L8VE~YwLD9&iMWAZFV^qrbC8X~fxJlPqfC+LMgWWOqViP4K zmE37<#~9Z=s(w$LB_@&;B^m^TXYn}8-1sWtEfsvZDM%o}*|{bf6@qmXC>HM1QOLP< zPlRn@krrynU2kDNKqpEzm;B&cOumkuoGWdrgQCEs1g2 zP$WaOV)QZS&>I|h#InRHj!tQb)RA68|M$mtXe3oamrcmtioKd)lhXi>YVc$5ar_un zOk6>HzDSs(Mwn}s^ZF{)+ttWLlwRSiRP)5iM%l7Jnfz&4s8Ow+Cf19b*ltCsjv#=a zXjI3e5o|L-bKqe0OR$dC*^zyZqoAP?+E|*|NJ%Jxv^=ZXC9A?&rSB(g=@hB1mCLGZ z)PU$I1mm=>ly+~`i|dOh^n zb1gNp#u57@CX(40a87E2B*DKAoq?C!K(4lrKS{M2+pu-HU%%rZ zToQG52t&>5zC2GNYj3~CpysU0lf!-uGt%>U@SLFREZGB}KQf=AI18$9ZB2}xB{(-z z;2@%7h$UC_sBj@ql)6GToU2Nn8j_A$>W*4?)c&Te7y^le-fflk5eV@&gRAO;jdd8h zZ^$4XO95}(e!j-p%K!*K_rJF7?pCMidX?gUuA4U+qCIvZ46EY5mh0Nr?NoT?z;x7e z7Z=R6;_p{&TzsS~64Ladv)*&3dyf_1W1gZjs(RzyWQ0dNyh~K!gLgmDTP#U(UC3&N z7dIK#=pd4*YQxXHN1|zsp{axMS{);Rid81gK*Pr7?M}BcRu6hb%PeIrW(9E`9%PgJ z+3QOXmKO-CZD95n=x=4u0K|(xz}v*Ro#m`gZ6PrL$nyXq1~36bN{ax6OTe8u+Qkk>xLY;6-$qHXSYJOp zIGJ+R4M4Ow`ojzFNODL6_KJ0${n81>^&~*40l^9b6w^Gz3L9gw=xn1zsxyVTCkcpG zjWc>G!8z9^T1vax*aVf|2{$HfMWm}wWnyy_Gg?c?(P;y;Z6$I;1uo- zK|%=bP9O;uAXsphx1a8w>3K)inr&OZ;QqK@hup^*Kx82J@e#M({EN(EXT3YdxDlu| z3@_H1F-VtMonHl|XlPKGHPsn}V-Ct)2guHFe(@X-9ygOs=UuidNJhl((*W>30+2QU zs^IY~OS@#&5AyrUFbblgluq8|Ne*f@3}?IOl~Pp}lYm8D3NMbbM_%JfF3vMf%)H2E zX7lg4srU}eipg%EWR|v-DT(6YWtOnvBk-I^icCDo3DO_m~7 z`hrJYD0hnc-G|1j%7L-!EkazgiMsPouB7orCDgO`&MHWnuqLz?2!@6%BiU!pcra}> z_|vkhl5|32nS}5w#XFG&W#k?JE1p=;!u_mA5jhy}A4~ioz&YUbfAmf$vdsVYCu@#Y zT>lv`v$W9s-%I>0jQ{<~nz7ch|8%m3^0EBy0drAXobgS8#MnOq<_pVzl=uxfw&yIJ zivAHW%jkT`SM9!Dn{TTfoBDge{K*fIQtea6+4hJki^8)v&&A$MiE@EjPuJDqO0(0? zsh;keKO0}-UntdkKiysI&sUpG_dZ#ypGXdHzL9tyy}!Pg|0D>LFkJ`CWfEImqxKiP zzfT|1hoZ&FjY;C56BGgPbyqebNnW>ZKoZV~H==+pE1M=rbyw4zFQpY*acspSTd^#b zw_6E(OZLPec8e7yNj!!EB@&`@3S>#*NNC&f98FBGlhH|kTPKv5z&BInt`%ru+VHH+ zWI2XT%apn&9$1!;jX_zuEqXvvlD=)H4a~HPq&%ZGxl=1gv^RDy!M=BeA>C-oprp_| z!NjVUT&LVT0K2}u1*)&wDTEQax zt;**4h1MbNy0D5W8q&I7AzRE~yH}5rcn_;YZarXeY?!mIffvz_A4s&qZ_DGF$3(1J z!PNSel50PH8JDvuGDAB+$wn|6V1L$$19w|BeOeDXsOFOu-jeGlB(y_t@BEz8b7|!x|cQ!(rng<&MI&j0r zm^t~rLO5CZilp)|wVNS=7!7EocgBBLbh`J92vBrjgJ0zQD)t{^E}$04gH{6 z6QNBYHI>HBin43h^-AtZ@J;7l@3BPrGGf@M#h|=lVET;iT5nX2@oN8_Ue~6}3^Z^4 z?Ez0Z&GIORPDht2nnOA3>Z+RmL%7AME$W-^Xs7;2Y#(3NGrPkSVj+kn`QKg+7y+O( zbx(zs!k2S9NzARJV8qIzyJx>1CMj@W6~5(B20!c}(JAf;gO+5NM0z;gDS$L$<2ud1 zfB#N7MzN3*my+Z&|2N!s1fL3bYZRop0%1_NbTUA$V*tPH6oz`2qnmFnWgDv+^){ju zCBW~C0K-|3iuX@Q$gLa)k9*RxvgPZ~(W9@D>fe!5Y~KJ@q?zKdAy|*4j3~^oq4cvL zOXIQ^&xHdrak%ptzzW?*2XS~!K%U2H`iov#?M!O;f*gn5| z{Ae@+e}*Ybea#pj39h4TWh!Tf>&wNG(RqqY&{h(xRQ9&2^@<$gX0BY4HoBMd!RBc20iKT32iPxbe%Z zU7Q8btFTc04u3=*iv_An(Vp92v;J%7cgMKP;_Iu(HQlRRIaX6R1Y&tC2_EMNAk+6j zbtNWJZLSj#zBK`#%yeTn3B9MA8X$SDk3`b2z;^5SvSTvxNh%#2`+PXa%%{Lg2hPO( zpidJv4wI8&!>j9ArXn0|1pL4dg{@~XNy3zwK|I;IEyYAN_M9n5P1Xo}q@%KGo|1;R z)GYX&WBR@4kd%oyLltBjw`9*$BWhdb_v2&7uRMdQKW&F~SiVrTxvS+N@$8)s7)%wW z3WiKDzdW6 zY|M4z35(e}j?@Z-#K>Q!-to20Qkvt!H5nEZaJi2nwlu1R!lJdYsd|g2_moQF7pM+$ zK*A1U5ne}sm_|boltiBLh-JI>LUiyMAwQRi-tzS4dXcUD0oph$)A!GR;V8T>t~iS6 z$f?5?zV5)hZ6GdeQtPcOLTdnQOuitP15E3R4foPk$u z#Yc6qJ=##O@VvG>bv8z=PLWPiQKS_cRRtCU(wUC!={k6`Jl|XO$H_ynpC!tH3?Q?2 zi^^Z7-<}H(WTIUj-~=-+F*tF@P8`o@{wzknYkb*H(e&R!DY<|GKm;VdXAx-ycX@6h z1bOdZ4dv=lDY;h!7ZUSNT02a>VZN-bzN7pSJ97G{Bz_oQ+5fZ$^t*?gpV+l=XiNdlF0}LLC0sUwf zIfm$op@~yTjMV+jj(#X5;ARk`=W$%t3ijv2&51@W zM`X8K2^fWiC=P?K+(kYE!jqwx02mGVivqV^nx&|qBHyC?o;3#wKR&QSWK0(N?yW1} z%J>T+ZdG#RJ~5UifR3vpGG<12zz2-OxZItu7sd|4%j0G#ZjTzR0QHFctUQ~>6azGD+Fo!2ldkVuOAck}VJMKDEvOGF~&qcb?lzt`n303?xie!{u z%;aTuHSV1(DNG*J&Fyn|9F!B7)osU#0-M+Hz^oUNF_BQ&g@DAL&TALBCH}F>UVaa&66piFXfGt*x z=j(e-e|!1HSfs3+uq`8RBtHttk9FnXzTfFs_l4Uu^w*cp1Y}{$j0DOf&Yy2ROh$5F zpZVDZo8E_B@_$hNs5OGr;WhGGkh930v@z7OtYB)e9q7H3wYkSIUBp|C6EEIF8Jy>SIqpG4W@+ z?zVzNEx5C=tle0FvM^aHFSzL}f3@BAkVq@OSQ4Mw7gCUUWjW$Lo9j3G>QHwUq|lZ` zh3XF@I+>2(AU9Uhotcyl;y@Uc>WjpCy+sk^s}<#43q)|Q`<3#)WdET8e7(oM*}#_B z*C+~2jTZs8yW`A?)ZPFDX_oN<=<-z{NXihTwJOk!Mp55A2_hr276fU(t5y?gS|5Q| zbT9FA>aXh#mN%zTOoLY#l4Q=BwVUI#yCb218xVp($mXCA;pO%Ec+~m;pg91A8tF&6 zSal=UUls2S0FbpoZUgZ46ytz5oJ0`sbJu%x768hrEPZrC(V4fZ4~9Cdkt^u09adi! z4Z0T(ue7mv)xSw*3jhujoI|BocLqWRG^1#=AQ*vEVW4-Cu6O9Tv`*5bbPZR5VC@Et z2Nb%`DWu;$ac)ilvaHQ2T+Jx%ARF=qJgUkcy4vr(ni!41*xjIxR0FJkOrtiSUZSAH7$Wejag1~I> zj+{FZ^&6FIX!~Cquw0!u=56w)02El~ilO$?Mni8dSc$8hij}O#5$6WdV*4TrNOeL| zAaB4i?;iKXi!ky8QndqT0mw9JpULXA_{ivivh_0{q{8+aE~Jx!#2NI;sP1ag zqI7;R&M7(zNh1at*??wDJ**GQ9o_T+=L2~I#!P*_C0-YK5U>!0ZPfN@93TUA%}0Eb zZ`B_ysn*^PLOKMrKsBVsZa-lTs$jT$JiThT7hm4uMb}aFwyls#kj;?V3ZpvDU`OwG?Kq(1V^F< zo5b1Eoqd*xd4JW9(b84Wul1>=@#FV9asG!Kac@QW-B%&ZQua1}zG~ z4_*zaykE1L%{njz%^!*PuOe|a$C8FLAHd8W$(> z`3tJjyL{OV8?3j-KCaCa=vKe2SNl5d$D$YYw2%JU__Ai9MWBGj^j#K%@c@jLysYZR zadAAo(ou&*-<%!c3gk_s!Ro#r)Mk+s+r2vi`@bWwYIiz8f_p zQ#Y?KGgnY%6c8pZ*IF@3(yr>0eeF_gGq3KW;m+CaST5p=RK$+=)rBQ3YH41knC^IX zBboedr)iJx<^|{b-jKq8OF$D}s!;9EaTgMkh1@Ze;HUe53CZF+ z;EJcF_Z4;X&zMbW1O@v_Rl&_Te2eyvzKI@tCQ`?AU#P;j8!JlWPmUuEg5KF&{ghOX z)u*AM{$)pF|9re(Dvjt%Y~>4LAu57|-o3R7VO`zE;gZ``m`PX95QdvNIp~{f_hz-v zNscl{1Mps`EI)`T(a?akDzksvNiJUvelO(0tD1+PQnk=?dEci-h`sDE?iQpC+~o&J zT>I!rv|&1ENLQu2P#ONs6ScTENRDeaFL5P@Cv4CXz(lP4rh&b!I zv0^u|s#W4#>8TDoiRTqmQ&E4Os`7?vWF>sDD7Cw-A|QR0M(tFXA8YH5CAwK71btM) zkx4cMmF5pkD4nFfh2lx7X4idj?`Jp*SuGG%mY{7;wtTCH6)%X{W4O*hJr@zdVqA*p zHQP7)V48{I)-}gNZsQ*`G9QNM>ivF?W}~+|l;hs8LL3XY4Lr;40*}ZHqpzGAT_Ja( zUQj~3amFZVe7j5S-U$ZqASdTlQ_n27Y<&;W%O~>T6{0GBD=PKLqt6^hswdGvSS^eb z7+E{RY1!@qYj^BE-WvqYe&2-nuBpxpzQcb3$xQfDmbR~}Gf+V{LDV6)QgN(5*f(K* z5xzS&T{JVYS(&u~yLoOW!mIg7X>E!nHxz=ute;Bm4{F*^?hmVRwBIfUNnJG^2mRpS zJxKn1^6;lduyXeIr(vWA&U)HEUFVI63?>C7vN*wKuukCPRfELTsw8-fL*y#Lg0uGk z5#hAwGqH4{`0)mY!1Ke;ke@xLgY9k$4>i06eXFMO!5+57qghXZ(Go`v>eNU|@sCr2 zkZrv&Nzx{dP``GF1UL0`rT&3Ac31D)<*e>7p$ z8WG(tB9YU`x`(vUh*UWLsyoN)0gLMw5EM{0I%6Aqw~4G*F2Rt|;4>0Woxw#*Mkr$A zs3a2yj?N-?(Mjg!pV-oD%aUHWjE;tk!KMfG)2LErBV>=ZP*O*R8Y%w7KJZdYT1Mp& z@={*c)1i6@I8V>on6MCp$Elm~fOa_txj@vCK@$jn@{c zJeusYB1)usc&QFcwzx)HMDpZX{(Z3?m)-y<18$Ay>AhM$|Hj6zrsbC$m*NN_O+-dB zfvw7k!q}UkIjSO2>b$hyxW>CpR4nKT0?X0jj=rD_3|#&G39V;HO+-lEfh0jnH@szKSY1KEa9JY zpte#Xy3lFKdgBr+cYHUWy~S6mu=txlJTNV@x3Nrx5JRYhg$SVHT%|1=r{0*e2uClB z?L$12xLno|wQZ`AYjo74j$10GN~(4Vj@25Dlh>5uuCsc>YHCuf+i|i;YqWT1W-uVo z$kUuQO~I!8Sf!tMsnd}7T|piZd3D6Gw4oor+Lrwgw}>mF8JxXu*pB)wk*cn=ri9s1 zcI@k%T10VFE9-^rq0I7&YXn>|*c>s_5gsqK*rvV(au=mB4khsG7*VbB%p70e70G;i zdGsAJGPJfAveVwzTIbt7{{1BIrh5ac-hWv6$9c}pr+wA>z~%8D*IhR~r-}9N_LMV@ z=5Bg#C+kCQ#?2h(O8WrV4Nw#n^Ic#|KdM?o81cjgvgqw+a8g4Aqsk__(d_`qR6~^D z#3nXtCc%SC<#x_P5xf6NH@R42oZiG1QTOdvuB65UyPCm+x!WP3sm3I~cc$bQx5E(Z zCRnt}4h{LLh`gF7(Q3K%C!Wqx8!@gla`NxYMt5WSQ%xD|Y#S`qe(ne%JegK3ySHHh z6ZUOp$XHFgR~gwvRJ8aCd8vwfyYLk%lTltt2Em0;5a?g1#0DS%wgE@~6Dn!E>Hjs^ zzTF@4`hWjR*o3a!-yM5q71+N|wrjdr{KK&)DGK|??2g&j;(t2!gpP^-c4Qt&*`xkv z<;yere|u#9KmQW;k0bM1)&I@x&fgvT{_tpQwnv7ut$)t${9hb<9hyQ#mY#>3)9vvB zb+i7nd)+_VKd1ZpUS0smOdC)Tk^M#}>UnvA6h@rihxdKp{b>jek^g!m3jc~pbD%$vIE^U9vFn0ZIlWBa72+!p6wf+#*usE+S(Ni%B&vmII z-OKgta!?l2cV*tqcf1}g&ku>JEc0{f>SX*FI+SfM*1Iukk^y}tt;mAvitLxT+Zyh{ zy?q_y#RAk=SPIhO40nBwQ8lLX;+nqMRoC;ppda|wLL3ij%Q6C4d~QR8 zPZ~ol*UJ2^ovUiHt_Hho*eB2*Dq1_y?|=0#pE!P$@OnK~-W4xE#?lWo!?11p?!(IR z#p^tmQEYIay1EisFWMcx*5=?@@aT%h+Jd93axx5(U;QXXZ-nVwPxi7#KkRtE?$P)x zbI8$>NHn=W>n9LcE_O}#-CL|JV%<94TG!>ngr0gbZ6YfOHjz>kbQ0K++v-)WCQ2g~_ zn#l?k!g}|f#+%o7c0r9sBxrb2ier?zKl0$Jc(1d+v%kc?b(0HcyC_F6H zEwvSo9IMmqmMC>1G?XfBvKtYPvjCTzeptf!T-g2CIpfv3QWyUWu?cjsLS4AO?PGB& z{)Jo_a3wIRkUH_8xr4EuRj~K`E8q0R6eh6^Kt$yNfIwgxgpwZ>n3@>fD?9g^oe>*~ zs%wBk5Mvf-V|CZFsM&PiUmkXpu1MYw83NIhBp~eW3Tl; zL28KVhe}qayYL!dXBy=8)%m-MiTK}f@ZAX8k3 zgoF%O*^J+G0HL9sAz2AUj0(KINKA{{UdN-M@oo81w&Tw%B|DHLe$CAt{>EBpx~xW6 zWI8}<K!xJY&#duq30%Y~s!Da(j9NSDqZSH?OueSc{b{#3n7Jcrh9BZrHR0mIWDML<2603;mJG>XCh&$ z&bWk!G{R9#MY4QrzEeyF&xNW?{;BFWjw!2S#q_l#tD9S|Z*^|m>+5eXXgG&WJ9xGffE#ibMDn|OBn+Akc z_nasywXPTt@5+7c5A3`S{Y70@)+P8O5#~iTT02?ArN`v%$tnh!FLc^tH9r2dRLici z;K9nXyzvHXIwY&UyWTD0We8GV`I@Mc$ZvMG3GaI~TX#kx06tX$4Pw!$G}}c%DSdW3`8Q+if;^mVPt-a}TOKLSNbF4Fvuxw#Ed20K0(W z{|SN&+Eo9;-ID(bL7TR*zaSXe7V_Wz7k%Z>SM`tB`f#$^^sm@jmu98O-k?kAKgZSw zul#-giml6^(f=z1KQ(aLv}JJq2MG2E@;cPO)&B;;d)I@dze8|s^YEL3+xA}&RR7)^ zjzL+6Cv>*`?+_e*k^DOZv*+rUC+d2-jZc0^V!sFoiT`_S{cjNL>y>^VdA9bGQb*$D zB@xM;@}Cf7s4%zKVBarPcot)_c#ErMPc4sSTCo|$?2$$DAd*H`ykAFoF-D!-I*I2{*8`l@&YW2#e>1(w7; zp1mXTifb$+lh1J-_AiJi+QQgOZ5j=X zRfTqvk5y74lcg7I!QS7Q?i;+h2)ruxvFN0^`Xg&s)|2l6yrqF~qR!Dm*F7-`1*9{s zeqF6>etyA9aSbU)Sz5(4`G)iJ;N!9H#=FnK9V<_lh<9tg$|0QZttdk}B(B`2?hKwA zYfUwR6<>GmTwR9UZ&${jyasN%)(VT|Eb~6-_+1S9|AN)$cO6$>Tov@(E}jJJjZjLk zf8H=b>O7uVHhoxX;=C0;&y{}P6P~oJg>m_l(iG>?(k!GS)ZF32E+Q7WSxB^5ai)~MtdYQ!w_DSc2F(O=fj*Lcz8e1k2b@U^8IiiBFoLba%7qxUm znMi2IF-t47acAfl&MKXv=N0IrMPPZ3mgUXFey{fV-8x7md^|N zaX00kS7-7kqmk8`0+AR^=CUbM5k^F9r7y8Tcmc~aQlLYc!w#ZDB$-KI;DT-RRFhzf zpgI=_aQ+ewCE!T~#zgRyp@CGj<#e9{AVg@)0(N}i7E9rY`wje4*aDP7&7!a!IK!X6DIg^jL z%`njXF{V$&%YN>x=n3knl^PL3&(!Y_#d zfY{H5$Zi@xlrkLe)qr4c5gINlVgVq$1oX*x8lcFFeFcD)3=#t%EC~ffX}oyMID{>H z#7ichX8NM)%P`~5_*5dL8>j&xX)WYkJWxf6r%;5S`{AkM{Z#4fOs%?KN2b4_q4`4wKKRyb>SQh@# zv(jY#d(Ud|z~>)hsjqI`{(r#I`-b13xACJ7@nQGX5sYQgR0r+7Ik89TM!KjL*0;| zLqEld1{TK91i7@1WW;iHFFy5w%8M5N)?Ji*dN3R( zS&O-#lnh*8o#}bVXyWX>qA@oGfe7Qjivd;7N+hCE){5JKbVZz!VVV^BJ}+FD9Fj0K z9}S7TgD5U(%-ncKi&#?cU+!3H)EY*k8Kf89&zH?j5Sz%ywUII@atl`fz&tzIt6SdG zd^q0ThoMO#Z!aAQ((4uvl?SDnLEwxu%z`(RZD!$1GBr$sJF}-^9{Gw!N5XM!U0@9e z6*FG!CL*|mQ|+0wXnWke^cW9`vqq4NxG3BBuZ0i~^BOT&1^SHX9Ah=jrd|ZEhg*6X z+128sA75`e5kCufdB=Bi?f@0)M1H04?lqU_m)EuL7vhJkZ($35*E=H6Ie9o3vk9r$ z?SNPD92YQ7Uw^AZdG`3(*E|*YV^VaQ;rz<{kB4CZL(PS& z1w6m&qG3)!cZuF-A*t=do}QVJo)hPT7)|o)7~YR3t8RX1cW8Il!xWp@?7{AVKbE_Y zJCB8jLx7S4gSXKVhqpNIbhG0o@vI2Pe`5+Fx0v=4@$_v3xcPsNqaLK&=j&*NK}n1) z@@IA%Uo)HtA?J=ny!f|+Noz5L0(s06GVBK#I6R!oyR)Ldg|*y)*(93r6ab)xuk7eD z=0L=L2uOf66pbz(2q_^;=S4}M{YD~13d}=Rj6NI z5#hnKFQPf)+CJq}P9`V3+c|$%G*Y~YmRLDoLgh3tT90O)+P43r3d+~kLwcMBSD|Oo zTkVlB98B*vu;$Fjd~+M@n$cmM%fT!R^Bq(T)!JOUwq*tnKEgQrKby?%Ckxstea(J? z!g-&#CyS45tn2(^aDrn-eyEfu%CQ==V+yLl(&sZ0opY{B6f$be;}99rowa#LhTz^l~V7mA)Tu5j1SVjEWSCFHB+;k z*i^B1*pdDvele2%YV%1dnd%i`1C~8>MOhwo|EcP0*Ldyq5^AqdGl`+yr(@w~5!==xaInYblZKmoA9Ja+$ zhU&e1+9oRbRb~0!erCO)&b^s8&(N!Cylt%Ib-O=`9hNy%Z`}%!=%bV72Xn*yNxoK; zJA1SCZpb1+8kH3x)OpVliUKY|Kvy^+#0-Mctw=eAI5|7POQ=T;Gn&Rqj^>1H34>O` zn~QA#_<0Y|+1Hy0C>19VkvNm6iW9%UTvNZy97)N6cFOCHVRLy=z@A}f4~joD2nqP} z&lG#Dpjh=<9LDS)-%CPOCWy-)ZX~~v)(QfB2&)XU*+PY{*7{R5j5VbNuMaqcB@}7< zQnX&oK~-VY&EZZHulD#uFC>+2aqQjf2P3Y}06EA3Lc*y5&B5+4(rmdqc8lX2tWY*0bKm|^hSJ={2$L`{!gCk(0lR!$EowbX1IvUWJ2E~ zfnKjv)oT@^;}UB-|2w_29XJ2klI9&8&;Q@k+oP^E>3`$7{u{j+@r7FaVBC*`H~)d& zh4$B2_8ZdSuPQde$e(y%;h1tvn-Q3j3Y(E6O7_%}Fq?J-HN}-&FlT`4-jPL9heNdfqAk5L+o<$Gk4`qDiFGh%j~;2FeY8b zMF5pFu%s(%JG|XxPU#I%*tccvI1k6MKL1XO&BRB&(OqQP&u{VM}Vd;-lyIg`tu9*KheM zUNuF*80U+);cx2tBrs;CCbaOYJSCl3j%&v4hK*|=*MWnME0%sy&0+}h-^ZCU#Lvz%be~mQ?S87|_>354ik>S1R*G-8!Fb zZYN+GL)&-XuDwRUvMiqN@iceR{(8DO>vvmbS0dg1dv@+r_<>h_T0{dih(H0mXjz!-purput=Rk} zifNDr&m)GHp#OHLMc`}J?^xA@VoVPZsMI-@%4n<@P}lJ_)cq{p08LA3P+=%^L_WbT z)^zP1H&_6DFcAo)CEv7%u@9b6dpl1NIDQ=tBc1^=()aLUL^ji4C^uJg1}&P@kaWNgK<$^piFQ@nCHGl4^qwwe-k5(-oh9A|i=`Dhn61 z5G%HVivh2Pu;qG>Q!bbwm#0Tb>Wq}zp6sLmlQjBmIy$OM0wUv6!;+}O-BjCF3mbXL zuMP_I%56n&w@JKy&EXtv2%8T1r{X;&y4EGw^;kpNR+-4=+qd;<@j9rlVi=&wn7t{4 zs3jv+itA`R)zlZ-y4~C4NQZiidbV%~(&9ARks?89RG5cOv2yvkucr$Es zAoHP%1_>8!)9-VOA!Ij>F;*oqEs_T2AVly^!>~XOMnY3nEW_Q}Ud#Hor8X{t;?~QR z?*u$bK_eLyFQ>y2aBd^>jcX-c`eVg~&c1C2Eaq_;m=xCB%5w^O8M1||4FUqxP$9F+ z=GAbSuYF26p`f?oD5tE7vRXk3ZzG}yd%=N)X&we|dXi`68Xg3sd%mVv z>6>}Thw+mEnLEU?Ryzh`!kV>v!S~Usi?WUbykwEuGEhJCUG>~>$55?X>7))R)+3Jw z0kw|?(DTin6L0Fkyqat<9hlWGhss)cpTV2!pqTky^?oJHdf|omvO-?8X&w9O+cxEH z?TNvxEakLVI<@-92$Se|cbi6*9iiGW#E{%Z3GUeig+!U*B4C1;R%W{+(d4DS&A0O` z4dGz(w^)-6`l=AOry&pKpBci$TCHdJhv@q{4o#LP=ay_<6H8fAY#UzE3S-OJsfD}X z-ihxxT)8;CaR}CGn$gajHs;m0JI0l$w$9Ag^=zK?w8!wfA0YK%&vB8}v^+Q;cQQJY zR(oR`7URA4&G(wUwVFj^O=uWGQS6DaDlhi5N_PpiF{0<$r?bnY;SNDhbvd?H^9o%4 z6(v|B{~*T|4oZsGVUrb(qnzZC@xn}s%_&hm7eer-;nR9j9^uD3H$9mv#Jp4PT1U10 zfFT>++ki6VGj;LN>2yzQPgrh!|NJw7eQb?*dqbqiNgjIlekDZ$Zt4jkt>IIQfZwKK z(+}EsC;DFzX{Ia7c;J88+CTsRaQZ(l+2DbAV8ybr|NhPZKD{Dv&?r?zVALu$u1qO~ zQc=Bnj{o~R1NZ6o-?QC(6uT{ozr8awqgilMF72kx#NcX0;x6|sEDLY|Wk8z04(@xr zhO3Idy)&%tFLdhl`b_`6WJBcA*QEY%H_$a(({GB7;Egrj{%0~3*ph$;S^T|Lp@U;& zLa~?+4GQ_&J7dCw!=TM^wJx%nl-&v zCl|W`xi=qY%p~4lZeSD&7JTntK&vJ8cKF7p_(b!Sh}<7EWKC-9=Hr_mhPJdFZ_Lpe z4)#r=Mofr|1%wX@)v(iPJ5{m!=d;&4=xxM;kr-iC@n^RBZhQPgX^@py1eBsVfRd7= z7%E9kd5)o&Eg0YL1!LC<>Hup```;iFAS=B1R^o zGR;5yJJC)~f*3s>9_rPXBs$`yfM1bNnSrDb6iHIM+&I~*Ul7C`2YV?m;a@~y2^>|0 z28CF3o56KDm9$j-=5fp!WZwieN((%%4BtDH3DN|HzpkpXNBMreRn0y0Lbk7sah#P< z>w7*@E*U}w+Rmv)?uI2R#i81U4_E!7leEFIHi^mvhC=}_>#LQ`7zTRIO-Y7&6%C)^ z20!X|21ID{P!_`{;Qe&cwG!H#=K4%_U(~^#(fz0TcFA4?p8f4YouE**iIxj; zWLPi7@Q0{rxDl0MbSG#U4Zz(7T-2h6ADdrIOUyN3%6=7~ZDIG6ETM-iismZv z)HvrjrCTXfP`Ha~A_7stfR!!vg=}DxUGKG_m=mms8%goc$xpWW9D8>?b?_^m3={+bW$gGvz62M$I0uR#M>esXQ~%pIHanIQ?>9w-cdPks2tX z@P5B7)1mbwIMQ%)Qp&&hlJrOuJ8ayJf{yr4ZH$Z>^{nno11;Xs(XSXF)$AGO0F4wW zh*OOkm7rBtdJ_b5rKjBhkZxzCagWoRm#$$dhBbP-OuKVhrvs~*vYGAa5_sRtlYG6! zNLkRN>iALu5%>Pe8W@u}ql^yxf=ZH+^+7KPR7zvz86VE~mg)lmgpN@dCMV8=TE`Sa zPd_>;8#qUeK%&5e#G{0e)|6JrrT~!Z98>RV;tY6K#zy|OB!grwqm7`<%6356@vbSe z|FMiqf=_uRF)})Hb`#63f@r*B$3bFtRO0ch#nJF0XOYd8s(oZqAd!`Y4goL(S5$M% zNzGeP*_&^1P`P27Ha|QhI`dsr@UFTb{yoSjU^V(Oh2p9y_F$-9p61*)lQlv3=?6WxD^Na8D5UbjEuYNsoDVizWA zUFp`i=LD+jIw}pmJX4suOEiS z=#dCn_!ZH0oRtWT*{;^6f1u5qB#-0>)!G_5 zanH5I?>rB*&T{m6UdL7G(ibU%f@a>k>lB&r*HpD#&5gZggeww2RRyb;rcUN_Quqnf zg^Bc%9A@de>6|U3jo_tmg!wA$F^+oUh6FE~J|VBM<}9V(P7X~8l(({jZQc`oF0{C= z$?tIxUn4VKn_4WNKE`rvyTq{FJFyDsN)>YW!8VXjH`YMp#KBKZ{#m{28 zId#%{CNt;J7c{+7%>U{5Y06#|$)pRLu5}OO7J^nv5%7HFmi3+1zgBc(bG_{o7uiNY zWX=zC{I<{csS0?v6qf$#BEs#uBNB%=!Hj$wIkR2mT(L`GOa{3*Q|&c8Em7qc3uPGr%4bg*p6t<3 zbAxeXB0wgFo)AJy*n}<7NOtX`1QNyI%Oo(AH@7u*OxP&9Ag8rC|7K!OoE+5A`#2Jf zECz-=1n@uw_!Lgx?Y|)Lj>WmQlrlcSWUL})Vkh{@Oh)%rgYM@Gv0E$TRS(|g-Y-iFddDGluN#zgJsjsl#~`EHS)9@17RhwS zkl@r=QqbczRi<;ci{|c!j+w3S7goBq*iTb=fz4bWQ^x?o9Znf`T2vBo^=4S>we(Bt zyN3iZ=FJM14YO~SZ=P!A(u4ZI9*2(XVoB3JSeLm**n*GrwM#x;-ld2=bQw03(}{bB zRVd#LetvR{2%7gX=Jflm*xpr99MaIp;ds#D8o4S_%HD{^cl^uYp|;vI`GXC6jSIrM z=-G*HGx1>VzK6%o9qX-_UOw&hmJ9tg_ zUFK@YI;DiecaR0L^}BAy5QjzY-#2gZ7v4TGI@O(q0Q>{bZ;Y>qou*KNPc&Kf8@Ufl z$X6Gklaax)XYUnh^*9&;ob__g*S8KPZQWj5)mkm$V6$2D`Y5l5jhvo>62^PsS1#m(cs58tbJk8b>qKqKLmfIXZ2Zm$I2Gc!m0YlT1J>hDwI%ex5Sxf}Y?q z+)Ix06>el(qx<_TDUt;rsX<0<0!Dv*zWgQM{49zLD>9vCVodHgb&Z5@3xVPoCJ2cv zRER0G0~|sp9Bpm=f*@|BEk2_FzNC>ZN_j;;!X57^r*`?*Un+f&q#LfJ84vjdCkd8S_TLURfTV>&M63(n2 zBYwp4t@t?;_>)(13=Cws5ceCXXT2ER*vN*-orI#1BCnBx;tqNZNn+v0Cd>vW5ytuY zaS?+g>Sp=yUU+#GG;rn``PVGSpu4e_h;JQaz|;OXD&_zH4nXn0oWeAfI{?N8QVxHF zGYCXLi(5Ki8IOy%qZJTF!^=y{nFF&%3l%13lAO5UF7E)hpJF4MgP89!7!6X8&5@=n zRLR3AQ7v$CI$m82bHA?uqtgO}@6tuW06&ToS<*;S_C8>F$o|nll0D5}GDyL3Pr=+k zdRwWWP#A|o9q2e5tXLFZx0!wn#Qdz0qaOy+ER?%VA=DyxhY>?(&rG1J53Yy&?Q;OR z4S?zf011$Zya77UH|j19`6>%OUdAD4;$TArRg|Pcsp#h_(=|hlW?QA+-l6MKW#Y_$ zZcfp#xw4U7+>x?>YlLsJ0hz=p9B7|su-ox49h-%&PxCW@j2PKc2Ib;Z*`jC&uj;|s zDCP)z{x~nODU5fZS$lmNkF1b+y*&?~KtAdSYEF)At`{(S3;?yT1242Gt0_ekQU)$> z_=o}lU}+!$-O;h!0XS5d>Kge+3pP9R!2mpZR7uK{0OnW#s=@}3E)X@QKc?Yxs)ir` z`)sfq1iMwt0kOL9@ai-mh4mq3}FQ ziX$VC!p_hM%>gKdX+Q4Ju>j@t4xk-oQ$VrLtPRPdaPmWPN#<5bc|cCKWq~DKjGD|l zk(Xg6qmmf^9UQ0M05wP@1(l)-Ri+#S;aCFT^b=(dT>*YteH!h zl60a0kSokZ)&RLEBBxYn>1+oCXW?S70Al)is=A7(N;yx9UQZ~%r zcS^6Gy@D8Kv59!<4rv43(JI|r>cw?YR5=n#caZK zQybT$gqN$!k&;|?SgzDn0?TY{7Iuvt~XzaSR z9KkGTkZA4X%aQ32q=CYrAy1i0a&a{r0oFytF88?`b;lzULG7O5+}QWgS@Ui?Sm zzH$%;<{BX)^?xy|STA4yR|L-^zp2pLsw*0>xBa5uTD{UBbhb?xllXDL*J^>Yjijwg zzDbMZA2wwon?XC!fBWTYyMtC^AZoy^Vy2BwC6K`MR9Dq>AYbL><@#RnYG5Rz1xO97r=r~64bkKMc8M%_OD7avTd7wd-inBl;Sqht5R#qGVR2^KVuoZl=g z_L^_w&(G%}KyBYxYT(k4RiWt}+DaZ^Z)xXO$mV6b8uXCPYcmL*KdP@dCtdd11zY+s zA(EO`N;d-4fYC6Vl^RXkH>+mg|Q>}pb+aL!%h(i-XcYa@cuZXxdpRrVqz#!6=_i>HMc=wfVd5` zz$adJCSO}X2kMqvpUG4dTwcvA{uqT+*f&p9@Gf~5*R-2cqg0n|V+6FHv}*u9c&O!< zP-CBpu@+?;754W(#^g#l>Yka*W*x(`1YO9AO3}`2>on85iO#Br@mR$0t3PM6R<_53 zYXdMN%g(#{@uu_zMTxae20kvYH30acPni23k9!=&Zw8;vM@q+-0$VU>a;u6~EKgS| zOntDcYU7ta7?tfbTX7T}U3EK2Xoc}_NZjC;T}{+(C_)oW*W}p0qi-uD4pq&^?l#-I zFef&#%icE|p{dIc#@eYPLH%uVQf(CYv#Pn zzQ!+kTdSX#l@djN-{O;p_?{s1?!|oWS@Pe}U?}s2*OedaipIRdJIv*GHqxk9mR-vD z95KM>Ag%#z^%gI9_vQ4^dv2Ucz4U;#T^|j0-hLb$ETVRpMIrxJQo4x^fR9G^mO?ja zMlZ8%F&rU=eulBLqL0ongxEjq0v!BQzXpT_%Ny;MIgD^=);4j$LI6kYIvH&3ZDCRI5reph4Yy`%(I+VC)J zEn0!2l^VB1j*LP8EF^SA_KI&DPL7or{tu84=cXw&$T4aMz{yPItdxE6!j87MQ6e{y zlNOnn4R!wULEI}G1~@%P)DGRoLFIq#i)e~MyiW3zvA|}f9_9ZUgD1E;7q1>)!m&$A zjZcA=Pv;YxAPA(TL*QeHFt#ZB|Ok|`Kr5PbeDU!yJafV$XS@oP-;S`HA{7`Q`4Z1{SK2InX-bzCn2#{e9 zEK>bs=|+St|68DpX)f?`j+j7+Q5B&%?M2>Nm_i7Cx3bS`K^a`~>eqotP$N$;C0E82 zj=8ipR%Upzt4>n$tr4 zi-nLW2rdyn^wTjH?4jmZJ18hLUuQ-pRI0{Uo?9o3e6-Sw?1&{#4-$=jSp_^~gaVre zyF3UmaSiuOU9IPJi(4Ds;ztr2WUqWYW(a@Du~`EJ$HoF_phz*G0zs2z58iGPwU8ZA zM6`Yx63$aaj(x(7Z3yv=<%s9CRpJgr0gE+CV{^`piNFe%&JWWSx?ap320rLy_G{Hy zOVo;W>KnNve(6A!nR4@61D~HCQ#5`&5;VQb~t3SXdCq zN?h`~zt#%eJOr_YYW`dfD}Kooj)~bMy_5;_T7%`nmQvix+?$79Qk?lOD;%Z#WRDkarL2~9mqc6MiFiQWIy(l}4`P}If_Ap+ zb|esnwAdjTxwRYg?~fO7Euw`9&6-mtrh_7vDLwqF&wmmh3vJ@pzC1WJQ50uT_^gxX zJ-2sK9EzdI_@>QK<#uF!%<4({8z80P7)9H2`ki$XjR17WO|M~L z&HmEX*Hr6dk#{=rKQb43c{^D0sz}ENQGtyVcsK%GR9>!1L{?j1(p{mnxFcP%-b3)t z%FH*WIh}v}5;j;J14YJTc`@BbH-vKcu~%;OGv}wkN2Bkg!(wjP{P8)vu%9VQS`D?o zPJ`41TJAUOBkbS&`4hF1g^|F*{L;u7*nn$A|KRBv@^aZt==+vOdGI(d;UK)X0#{ek`L)t4t9M|M!1@x)bzj5W z7jJE%QMw#`3*2J9z%08_S8TqV`l`0)p!%ZtTjXgdqljwn@rQ_6fm*5T3-ahz#(l{Z z`nXA&Nd4}3{x*K5yShVZqNwKf6zn4*6qW{eNcl8(PTx?g*vv_y?bHBXAT|@ z`Ug<}=u?G`IG8(F$j$a;%Z~6Ncz#9VlJ=b#NoY5rtQ3Mde>8+G*bNpGD7P9at*Zp& zc!fTMi7FljGz$X;hN*Z_8a9R*KZL>5Uw;6H&o!%IONU!$hTC?A+pmT@K7>2LBU}&> zZh8?OUJ+iI5k8#}eyb7w4-tXz$Y4Zds9t2aS7c;nWOQd_>}q5@@*y%29yLQ7r9Bjy z3WjSrhi7y~VMy5J%HgL!M9I)cmqt?M>qV!9M?!B8c%2N65w6>A?6Yh4m6|26jeL+mU(Zjc1z<>@SI7RMbD2j0el z!W4Kd@fnKa!X8|rbs{y{UwPoau56+{4c3)e71IMHT)`9aQ-dN>_y)(Bk1DB^&HWO? z4JWR`riVDcnsH~h`-<}zI&|`D1`23#@i<(O*h%^U>Gfa|0k(+5zlZcdD)GrsMRm)B zwv>e9jwDxCK5Lz%uNB_5jHDSmbk>}5H(P$!TCwM{=ypf~UPubnyG3jyh&r`CSlg8+^MY`iJX_oN)k|V(s!7TetKEH1tBTW zD2_}N8etpXw2w&yUy^OWX=w9F_f~;uUs4UN`7|(fG^-2`NusC8AL!vVD%k?iqELI{=0?06}!~MxHxS*rYw!HNP;ay;gQV}W+<)hBkC0r#WO4tn&v2JkO^3&iaY272!Vfy}na&H|JIyD^+VBfd@eY8J-~B(9Ip4yUH~h5%hBh!IOCx+x%h$`Z4< zlTT4IB0ce_st$y_#OoF80$pJJal#%wY0eO|)b{1*dm_OTgqHX==?*rPZ?)Fe^jg|M zaj&%fQ^jRKjX&J5|IGF*{l&U(L^;6fAn^r#?WwA~rvHdBVESZc;Ej!LI;c2{2E4!u zT5o&(T*_8NTv5~)T+I7DqU9S6K#Qf{DQ8G(7Jwx-sLS4kkqBys>f=t){j;pjZ-M4q z<=EO?PEw*?VkxTu_Ix1(TX>&* zn+O{Lb+byBc5l$Fh?|?nvH#~2f)V_2!kQ%`==4J;IYB!t8qLRk?N!6-ez0B~f`-|)CBwc$zrr8b<=?27y-weu z&&6X>^~)sAJU#=IoUqPfz8zdmVib8abFop^&Hjzyi8CZuY+=k|;r}4Ra>A=(KKYTH z?KQ<1{qXQJ^d+PSl_UHIRs*rdG%yC=If#40b{xBV&Z38|DAQGca5}9 zM9)?;V=%Lnc5f)|WyudAve<}4`S*y?>Dd)lrtCG&)1Kv%`5F5+i;s7U3|tE^l=&Hh z`H#mGh-<#+WGTELm7mO1tsm@!LP$Ohhq~4EU{DpvSNCAQnSJ7_)Jy598Xw--rkEOD zo%_8glT@}k^0#WGO;U?wtYQhv)~`p@5Zy*p3< zRKmhlaI2gPmGi4JdP=U+Ojj$YKb3E*)~>ntw|&WSXVGmi>1&pvY|Iz+{WsmLm2+or z*lM}m-R9X=|Eq_92!uA(@yR5l@A&qF#LLNSI2cBz^14%3)X?icf8%W0=2WKHGJ$Ek ztEkTlC_s*4v2Z5Hp)vkp5V&SvYTgTsB1($;xwDE;Y1`qM`h+n>oa7`7t@=q~0} zw`{bjqt=JG^8x#8`D1*<*^ZLwfuu1 z<30Y3dFE>uz4E-kr%(nx2kLQnY8hQ-rHwB7F{5TTYyM)l@xt_*HRSH zdWAlhO@wL3Jg0oprZhAiZT~Z=0g_p-ym~nsTGNP2o0GDK8(tYaIJ{a}H;QFJT5nIR zbyQ}qAZyn&>$6ss-G%RFTg3?!WNmGyHe?H*0#5ww6c(9s=-42(J#Glssj?Pc&+j>E zI$|bcY55>xRo`2s@;9%Q5f2n7R{EWkA@}m?Z)L}X%N=t;87X4F9}B~WcEsVTdp*Xb zos6}Y=pJKX93#mH+J{2SklxpL7t+u|Wge9$LI3%PLW$#Vc#E7QojRxdq;lHzEa;a+ zqI~JauDYZQ_F^kQRaCh(fS16+bqUqXK@EWFEPT7dXoGvVDogV5PTe|(Epes>lMSlL zoFq&-25f7|Hb=dsN!;-MgJkx!NP0rmQjY5uCfy9!1Rwv-?!*$| zee8~&3L4<{;QK`XMIg`F_~ySm1_@Kj?SS=^E&1?POxkT*%4JD8|Lij12ueKIpi|*L zIO2aE05M_Q1z1Q;4e{_ZrB8BFzrSu(kcHP;coof2ApumFpoKL1=P zq<^~`g<2JX2ctyDKa5bUt(s6r=uW13sfWK}j)B7@#)4}@Bg86Y1sxwD9tgF>lG8oO z{m4WC@Em;|Te4OgO3h50%L2%hYjZnvDnEUXfvl3sw*gD?>+?bs`l0$U=Jw`rR9> zLYs{F1{}w0RTWOiXCZ}38pS9lD?ACc#RC;F4i^se6Ul^Qsd`1Gm>lIE zHh$$-WpA0wv1gVS&E^H=7sYHpI!25h7TC?yqw}Oqv^|QJEHWlXMXWR%enwtR9VN>h~u9ui{28c z1`}j!LFSjk0>}Ia1tYbJfkeIi?w_GqSO$77Q)fR$NV0h7bp@vkW&~&*8imQ!4TcH2 zk)_LR=8AB~Dv6mOwJW(gaJ^F)3t&s9v;+9aCamAI4nE?dopEq{N~L+<^N_}9vSg#u z^rPJH|BqjJDOx`swxXnGA=SGjY!dFuF#5#N@cu2j4*k$!zv8o=_nu(l`i5$6@~3)Fkpe7%?$kWgU7rE&k?G6tXEEdSh@8fiZ_)VK z#z7D;twt1nZ6_JAhx+t5=0h%T^1Ao^2ya!6s7h-Vs&X>-dd0XyIy{;5O)_0X{e-GV zEZ^oHhk*LiaPnzOoZZh{zJOO^g39hi6tyf;8C_qaCa$ZNVkTvEqeqfGu4~1qtr(D4 zeDvq=Paf}+SVHU;4bZup`fAOE$d{K~Ikg)hG$$Vu7~R_Gok~^BZ9f?L^tE5eM)B`c zGiDersMX&!51bLfJ96HgJLtTR;b)o z8@?Rt&2%q!e7~*teK|4xg9sc}xoggOIki^pS!2rBXs&xhEn;#p8}aXMg}C&{Jwwjz z`nxA9oib%Wg7*YQ@l`*WF()(G;`Y>jT_@QFPbl8f?}j(?6JNPB(l=vzC~``t56G_U zu7s>9M4$SRv>OK7etuF3S>p?R=aQB$*lYn$i{|-vg|FNWZA^R?RC8}Ckx|ENqrTb! z|EBs3{jwnge)A0Es?EsW6SDH>jf1|&eu)366XEAyA^JDL>q5v3&H(Q(3;A@RA5xxz zn*S9gB3~{xpXTvN9)dcH3CFtg+3HDt1|W49ZZR0g!=R{{&?gKSN+$F^g8j*&{Lox*O6s!*LiOVe-<<1^-^vP+mS7)6JZ zk?K_no8O_@#Ro*Jw!PJ1Xbg{sn4Y(&9PL{gQ(1%!5LxK>me+SHRO|fO+zL&L9u`*9 z9`%)vOS0q;-aE zwa(igc9)799HN5)Fl>EU~J5T;I~$0>kAMD*oL^~jwFW1rD2 z8IpM{T@=s9!=7%#t{4G`C_gU@S=+2rvmWiK^f}W$Qi%N6tji)CU)94My8)aIf z$%cS1-bwy|CT?ffcrX}Cc>naK-Z5sBxq3I-Au&FWl$n^BPbs;XPPGVfABaSxJ}T}9cf4r*iBr!J>K z$t}e)`P&6$)UB?r4z?%Zml(|f5y8h&$zF#z&Ux1yP^Coq6vs0=%4a2(n~}S&M*I9` z#)n1b3&2zvYaEA3qnJ`u5r*}zU~KL9D-?kGW&Wk8&FlL1MP~qzw)>x6Ic%X;Jw&K@H!h$04 z6Be|3rF7?x%jIvVpJ%lxvt@5B3+I9i#*!l`QTUih6wkEjy7ptQ8&P8iC5P0`Fl2`|tAN1?@Y^;0hvLY(*lG zC4n?KcGNI!agf-}AS4osY1q@2#7=AR#858&wq&B(QUPXzl^$@%;5VYOGb{g|4AXFp zw}yy5O{;c;5%4SqIitkYDHRQ(#9{hED7(k4$km9-QiTd@z>iP|tzd;8!|cQ-@@)oja&o^t2Iq#!p?oG1X!5W7SmB{*?5XN|kvcIx$O7Qm;@ zN+~+-DuL#WYI{r$yi8x|-4e8D$dY&i`n(U&ZV(cS*;1Vb?SOV1y86+97tiK z%{fRz@vmGRK{c#*UfnN^qBgsbGc|}>t!N~m8Go&=->~_?xyN06(DbyY^pX{g8qhT= zW+R1n*N6*?RgY8?scp)5KvC2#Gbp8%n1F_=UqH@gCFM*Z6s6)Sq_qzyG}~mt3_9aO zNM;5PYlw&vUK1Ct1P;#Wmh7b+QC-UT`R}3vDgghYsu(nsE)_?2S{t?Z3l2^%m3YPJ zUY&)847&3e91YOcIOc#Oaw=BNqn@i{Mo~GP1s)PE7!n<9lu=wC=w6JKxZ!3$8jf%- z5aJl&HDa%#feQF(#pjCFVS`RDAR?!dfrCceSzp}MliioYCXWhzTSRdd<&kQ5L4)1= zAkfc{CbQUt+3hd99QY!PxkgIOBR|T4fT8h4Jmlb59ocB0n#j))-Yb@T*WOHb@KpY_ z@Su}okssu5h@@L({(Mw__D=;a8=^hzc~9cBX{^{O7-nFiC>SuYb}^DAh$n@t%_3+j zr{Wj@U(J!5FW{+9=&%w_49dgoW_%o|@m*=ZEuh<#BB1<6 z&>&|mdmM1^uH+y9tz>Y7&8r&AH#LeFMx@X8Iad9VU^M5F5CWfD*P(m*IrSlSrXs0A zaSqS^wFsOC%g4Nk6;7()W}%aoUK@rOHK))tnvnIL0CA0fjqL9>AdwWz%JcYMUphZ} zEj|{~C!|!ndLmcCtRw%d>F(D76oyW78JMn@%lumWUbN~V)U)+i*N8NN0jJeHoTFs) zBjtKpCSKKLz*oF!>=$Ay1%p<;!@j_W^{sOF zR`PzB=FT(D%^>||B=D!1u!uh5G@tG+X*;*oDSvIK&Y#9s-#uH~lKMdZD`I=S%WG2V z^MeR&IGjBH`7!uwHb@tsLF zYsvZ^h?$|J$^i>jv8VF^YxHj(>I3dqTnD!0e#S&WBD+GKhXJSr|7SZA3IG7O{J-jA z&ZYnW3a9O~7gX&3r;GiU`;A0`erobN(G8`tV~i)_zZxrobF`buF4%?b|sssOq_ZLKw^e?&9 zA5c(A)881@xZkIL2Gb)FMfYx(3qL(yw&jb7w_0t5Z7@}uc|C|{6hu(oVZ2cVV70yp z|Kd>j+w(3FW7XkM_>d%)0BCLId@;e0Wyo@e$gz5s|GJr|l>{&Jtu1=UPEcB2Qc>;2 zUCuGN6|>Ga>3?Pr3zO?ED+y8#Crx8{bH`Ac^{7okocxoCG$+$ZKsqyae@@>x4*2t7 z!%pBI{y_mRN>;i5cbI;jlH)6k!(w_4uwI(xIYws4?Sn>w(|dX!%97k*t^KURe>%V1 zf2D<)Ra9y$7W<^MSLeiV86yP}^941=4(dLYSKCPjr{5J-CX-iZmNT8(o=V)M%9b|Q z-B~h~j}CjBwxTax9W@mBcpd7_4NISOfa+{3Y7X|6A^AVouyhOjg0USN2A&jlBhlPX z>UuT>1R6@a`B`ejMt<}#^^dK&I7I$Xk^Sw?;GxLg_B&bdZS8Gg&Qbo!)k9;`1v;sH zBP|wt?O>N<>~YGSp6$_Szxu16VkjxOMH4LH_ATOhS>*-op?4)qVmNjy=~bU#XX$g? z3=uzdG1$JPYWgeewl2s|5w}TEF)T)l>xW3$E}3Gp{|p9f?y5_g$KYPCnf3VQxzkb2 zdMv@F9qz@2%IG=Od{~hC_ZtCrQKK&JbG$n21SKee%E$#OKH+?famfG71O2pYu^tdsBpy4FYz_q1dqlN@>A2Vh(nl= z&&{wG5TGLA^`ULS!&oxfv(Am?+rKA8K2yeioPm?5~L#H(nh@yh;=U=7kh2Bx! z0??^5$ECz*f#8zPB1yT>x@S`k+Dw9RxKDB|C9zs$_s}mSNG#q05F9%wCPNy;5glBL zONG%(V*)N7<5Q7}u|EEdCA^qu{j8p-KwCnIf&z+MwYRcfE23e+&*0ag(4KM&J$di z$v76dsc+aOlZBlyacC`EOToAMu5QPP2&}-|!_w8xMbi3z^5#(iH{+tQKFU&^2@bgO z7B!&XT`?MMHtRp5X|wKgeq-*+&p~%df6T=zWxG?ku#k&an9p_SxKqB8rt1<|dp8!Q zRC>0aCtAueCOS-jKR9A_EQMI;^HR$!nm7XE`mWaFdDPFMFpGz0^Yl{Fka{bRnHXZo z;3NGP!b))u5l3ERLQi+%M$*bF+%+lSRx@Pqfv zkgF5#BQEg$x869oQ2?bBvN}GJJCp&Wm;--RskV-N{v>rXyewn9*tf$IsTk^Oz|Ho~ zmlYOhDIN7Qv6E+j5M|Re5YB&rQlW9zh!)BPH|nJd^}8Vw+vTim&ToYdsa0tmbyT)}+ugv}Mrc%$?|N(NHt#H4@O{1A7t@vcg+r6au&rWARL%8K{!_0ms%iY^<` zwH-NfGvjLUoKi+q0A|CkNxkxBfSx+`Ti>oJLp&~NJ?BIXIemV2#cQ#}X-kyFZ%Eot zBG+V_^ar`ME$V_3E%CRfEy90j{Q|zX_{_O9Suc50|6AV*xYhi<_p!=HzImkx*X1Cw z)tq6d^%uqrSMfJ0z(Vp`8Qawy_@!TP19b)xtEy+hc zi%V>W?ieoZ!G^b65a(Q!1!d6-5sA#xK!`5$24394m1_o>>9OSk(kvW^oE!C|2nVoxQ!gNY;5qxrOW+u zpJLga*TCw#cN6isO7tJ1*ChzZOe12bLG2@o(nycf1{~0;GAvN{g^ITqei-$RMc^x8(yuxBiRDK;w zfGR#9w)hTk#})dr{|+hk`yJpZMwH61bKswI|K>lzu*bg{P)&kBkY3IKt#1v7eC{!I(!oeShd z(PfR~5wbw2@)B}*g-mjwVHAZR7@;m)=waMguCxGg1awz6R34 zklxxC-CDcTYB@caKSg3Z)th#i1`Gav@)}`Lq2pu5L61O;>Adf67Sg*iqp5sX5=8#D zm~19E#O}f4A6f*39^dyOMN z99YQ4GXfI5f~&y3ZHUjpi10=|XNX6L?O~Yk$5(0K;P_#iLn5k)ae@Ks;GshYA;~Ba zc*yO7U5{RP;I=DcXYA&_Lx)^sD$t2DncN@6enl_%$ErzDGtp#7L~@Cv7dL$1PAhjj zmVcM(t}!^Ol9SppQQIsL{m3*OXs;|6Gi?pIPY-S*CQ;F}d;1XjiWFwYX>7quat5#j z6633YiKvgT%1l2p>3^hPN`9ZNPm#r2cag-i{f4W_gB|(6{mzr zq~i#EHD;GtY4_|UDd}Y@TI{yWsM*w&L%foiyc|ckOhTa>@r-u5{|+x6Y>o8$&)cqdbMDn`{8 zB8|;Qh?Pj$K$xs4ne@rt1ZkwX-{f5Bb-1>Hxi<(s%nCZP3J(b@QtuDczaI**#~k=k z;&=1~`#th`i)6Hta2mf9O+YbE2gR|0ug{pE-TWSN?T-CN?CYtylm}4@Nm&S#tZ`z_ z@0dgvTv(m7a^nC{5+zi<)je1;=iBdivNDoQ{S0~o%~-wYjU6)3 zb~ZnXs2LOPEeIza4M+6p-CzoQ{VMvMqCQ;v#onw}4fSH6@rcQ6A zHsJvRo4bTmO5~PZ(F{xOe>T}0HHlM9La{;d-mGGZFKiAHj(fZs;SwcdOeBTgwzS1) ze75)omQWKPjSP@_t6Js!Q9-k8$x4@xqEDGxb&O=Nm--VS3#!B4E8zr53q+BO0suQe z#J{(U*oD|Xc16WfMX#a!kh^?7>D=eJ@(xemb^~U%>WEvt3f`cK0-IOi3M3X`V5+B* zT_?pX?)uZktm*6m@Iqx&G0hT}^J#X}Gh-3LIX2&?hy3A0tWXpsLK8o6-Q3hIR&SN92dftK{y*rfvr3N*S_355cU%30XWX$PPgK=St5m znM>GuqvCG~M%yL=L6YaG&fd{}qNKh-S1vmJ!LG;bNdE3n^=^e|k}iRY6_O7jIsK!I zfJR>e$-TNCUlecFHUDzTT^>7FnFO7rcPjCTn# zyN(cA0uhBg5!3VK_EW_wPPI9%kbO+euve^V zLj7~j10+|KMy^#^$LXoIM3}qj=`Qd(!c?M|F(5Z;3YcolxNCK&_LN5~dar>dO4er|_Nku@QG%^)tJHf4Ikp(Z+_d#gf7;{N3*WXZ4zOUna_jD%t zjl+kY^bOkd&1~*F-S6*jj-WfaVS`M7btUM5(Toe~4B92{zv*wS;#1UAv$R11Amv&5 zHAR9XUYs}`9%ni0g~BsfmAq!i)byl8_Rwpi$_>8B!qW+;SyI6>J`L`C&D{j`H*RL! z`Pa%*;5xIt}dJYLz7cPHN{?7av+5+aAMOct3Q3#1#oGMqZ z(It(+HX32!LS0Kb!xdJ4M2{aj!(j7VSlqPN5-|fJN}`gA)2`{&^;+szAx%}`#loG0 zKO?FK1nDmdlU_TFJ$KQBo@L{{BO5!*h*%SkG8y&v3%#66c}x1oG%Tgv$~aco;}uKu z9}p<9X_VVbNG8{3EZC;DC(j#Ww`@0aO5EZ;o#kpy#7)`{CqtCnON;S~KO;K%EN4oiw(NJ5rbU`@E|L`hkt%y)S{Z%5m4IJ@O-TV*tY{aoaZ|=9y zwAuG{#L$eK)~r>Cv3L1dpDCNa>0gyW>9imwER~9-n6b15=SN;&)(_^HeMavoJaJ@h z8*%rUD@%}o>Z~@X1JB}LUmZ+!J`{h+v(@+=W$0X=pS&~np9=tn(|K0nXd*bru`Q&Vc&|ESM^7R7A>Ql9Y%4(_Y2^E;g zcK@~wWI2(V-VE!JWV?l%k5)V)zFLiYgvP(!-9>!@f7K;!*qNLMx5JO){3!q5QSteO zY$FBPM93Ew>~W_ zNDm<-hnK`Pp2^mQ01|-i1X#=m84jLFO0L;|QU5#obG8k7cF_j8RdesZ420rNzb!4~ zfK#Vvr#z{<`!M<2>dV}gDS|-jrc?V&P3zZ}8Rz19CghpHOg3gL4HqDhKLBYp0GcVl z1$p$JB~{IPlss3CQ6D~a^VOQJ|Ap%#9o@fliWzg@lAPEd-7mET& zU#&Q9N!S3NF$3Z1#i9YXPx`%9Qy==GNd#SfJ~UBFN711zuA_tXIHjW6Ulh9%w>tgl z9nx>$8dl%4p}r!7rr6Y9i>N(=r!irb!jUMH|3aQ-HU~T*I6{nDrW3)s9eA%hCrc-U ztJIO0h$-$5?_yq7?_AgK>I_kwJ}I=WC*&yv4;C&1IoaVu7X7c6W5C?S!kvA#Yrck` z&MmsgK3DZUvX?2O?fy&J5rhWvCgY*O!{sdNR43zD_})HVth`@oL+-4lM+y|rnyCA3 zx?C-ZC!yV+qal3)fdnn;fAV*uw4Sb$Z&I5jxuLTj^2lQpvRoWJ$+B{D&rrLW{F?-l zk^|N_K4+!iz_B5mJRES3e>73f=@I*V{Hs}pl*C9d2MV9hr-~xtkl9)jfA0e{$*I)m4oGu=sI0!x34zK!PwFX-)=e%j6J93qZ0oh|+_<^A@s94^<_?or2 zg2Fd88c#FvcgVO}S*DLhPV?!bZ1!opvz*c~@!}25VWSidV-da*`$7pVkr*3@5?`G; zU|(o25N+f6Al>lo^l@ILk1$Ie$d^>}%Nk)V=30dgQDH`3pg1V&w0kKyFT(bz!D!4; zvZKvyy*swuK&eKSObq%+45tOmg~~6qj>gn?FXq+K`a1yi-r9#HvsW9QHx)->kc$ik z0qHXr_%@xQ7j+WGxS@tAmZyqz7OPBus8W=RBj-`pY1Ar`Zlr%uc}up>2rVd$#kPy- zZ|E&0CxfrvqQAKS2SE70A*G(qJa%Pn-e~$Ah|$kE7-THZT7w+LSB+(qVgJgyM&H?= zUP&Ba+f}&<@9`L?_==wEAx8AZ+)m!!uDBwi8j@K~2e0P=Bj?UM~ znuE>OMH!H9Ti@|9j>}Ps51I6nHCA!0a%Xt&8qVnBB}KLpB~a~lsFUFH!@S^pIYx8E zYDQqc5Z~4L&s$w>78K(Qhj`ZQlI&>P9p&jC*WdU$mQUg-d@?3}QTeRUniw0d@Vc+N z=hb=sD83z97*oX+o2eL3 zDOAO285T`bc4}oVMqXibiMY?PYMm_naOF{Az+H3J1NK# zU8P=ZODBc?pP*O(;Lrc{V5AQCe;KL&I~Ya$1C>7WCDIM~0!@br8M*c+{sYD0JSLQ> zqKJk7-yC02;+rs~U_Do~R!vr@XrAXPwArc@2)ArZO}ewJvGHZ#eOeODXW=fFq4a< zj$y0z!2jk6M$XpFeYEg$tYiEwyLx}c2=hr}qH|+st%G8{y-v^NWOGp74V{P-lHmP$DTy*4up>nMd+192&rL~R3NJe zOSG?ZtV3Cb@Yk)J>=5G&fIVK0eZ2EW8<=sA3<>&Hhet6b@0DT~EsJP!f5` z2)q{3nMU3E&hPSl7IJkhW)DJeN3T$73^=@M%NG6$|9b` ziTP|205qKD!2?St4iXKMWEcs(@k>^LN+g@T$zl8<2O^%hs_Uw z+<0U9VITsSV;`-dtv%CJF>FmnWVZBo=wge8EMq!B&r*3THYk#@C_a37vV4fgwatUu}!1Wq6eSZzd-33rsyxV5jDGX31M@CIlCm*B7wB1@Y8r; zs0RXvkybXeTEbR7pbe959E+|B^Mp-R47r zKpAe4-T6_USTY^dX@NJ)n4x!alvr1`NMDsyv7!Yg-wGpY!pjdVz=437CkzR3e-(4x zbUFq`1ijwa3?Gy+8YGkY2c1Y&Q!+9U1{mJp#Y7@lr*IuVzGM->*8PY=pO9H$0$w4Y zPefx5e{o}MDR`Ho1}AW*?X4&Wbmgc{-eX$~6)?(j6&c9q|G8UOz{MPP#?&g?iVFw}r~eH1TMxn=hWN4E&H=&b z$M}4}eicbMmK$v)Co77Cs)XEf65U76Jk!tv1B9bz`BnM%qh02u0C7;Pa)N`b{pxb~ zl($ZH>!Ca_F75MnV=DUz;r&e<jSgm@_H5P z#rHSb^6XAJu>zrv_^;J#eSBg+I1UJQtRbTlnK$QZz#GKhEox1~m@gDMmi<%mT z?nT8|Wc~8LEDJvwX-d3JL;Tj;#M8PPdg3y2<0VmqNlp*<2#u+0CXmbADA9tYa+`gp zn8?!4dSWeVi-ilUh@W040A@znEX?uQaTJex-d{vL(j( z)BKS$Z6`OkCV{FYdS&VpL%Io4)Lp{>0X<^e&$ydLFXDe+Z(bd5moS1o=#~2o-9D52 zVJ4iJ)tcjEOq2M2-cj~#**8LmH$wLdit<+#nUfvtJeZTE;)x#_dk>AF0mhnA-Zjk; zd$c_K&CYJE^)wns1}xVZ_9|pUN9qn5*&n@?2Z)vB6i#64rsD)0-nDwC$12FEc?ah7 z@^XdKu%&<{nwje$Vv?E#L9@9Jb3DJFJo$*X*hTjDraw&ioNC0Z^9Vnye(L7`9Wnmy z+ZxrMX7aZ$A@>rmk<96~L8<&B)s-aIq6CA)MdtyW-?wl(M>;POh!bVq2grVQe7=3d z2G%mBquJ>i^Ba$aCv4MJB8ikI(3&nY1#EvYkt$E2@&3t8exlRs>YnlAPMJk_XBXJJ zS60l?ET#N=psWP8^5Va!*q{DC4N~Lxtd9P_gVZb33{Ne=ZP;`V>$i*@J%5MKvh#a3 z7F(ktu7ZD~jZ=`Ts>Bgsg%NRL)NLDlz8gk;1vm`}s@Ys>>x^-=^+ zkCO9K=-;}~pZ8{qUvF005En>Y_bJO~)d~E&L-=psw!o0Pvf>ux@qS29zdn*CXV&lz zJ{5FeW_~};f-F;HzW+pMYtKMrdV4H=LwjQT^d`#RZyu{7#iRh0AT=f|*T;NN0js|~ z47iF_2{4C>LK?KOOw)<#Dv91K!nL;GTP!t0wnV$&3Lnrl2g{Rki_>}>v%zjWV z3qVg>b=St!T82M|2`a^6hMpB7&_ZKm2l>V4x|eG8vDp+zhc8o$P~Ibzz()i za3Y|UKd>Y?VJ`;Ahh_qbI&GF^;S(tsAOO&kbI1p#BheVxhAh7ivSUX`tnw2IRTBj4 znL`Gm;YOsUjxT&iNS^<^w$zUc=!(=Ji+Yel2q?o`$VBV)@Qu6T>bq#3Y-q%@+_6Za zB}Z8IkZ{ zMl^Yfu8t919!MJGaPxtbjH(eGV=NSD@jnLRH4FtqN!=ZlshTiI?CnJ82!VB481sc-9CJHhMypNN?j* zs}5e79W^sZGKo@}x}gkY_Dj4|0g(}X)p$)cM+)r&Ex zH9JM#vG>4efFPn0HM7u)E;v_d3=upe%2?I~35noex^ehL@5w(D z4D(G9CC^RRGYTKDFD5YAu12I;C7zH{x%`5cMI=#(SvTE)YQFSxyU(sfo!f<`O!fy?n62 z_lb44L}XtWi7VJ?r(PN?Ul$J_Ff1REbH^6O>R^BGPE4wa5IQX5XU|Rchx}}Pnc@w} z&n+tG4wBYYMAb)P#89v7}ivH`!4Yis*)F*Fyo4lnVXcvktcB-Im>;c zn-VJ*d1vGmL1y4z?a7d|F@QKmb_Y}F3c~U{)7H3HV_j~ra|k7d*S#BrhqKkG9TB}@XsRGk z#mjaf+JU&BX}h8|OQm_B^To9kCeRVQ6|PBJ^ENX!yl!nN zL@4DQ=^ygN1uQANfVvtZJ2GE8xeXFY9?P;72~^n&Ws|Jc*;j{?h>Hw=gHKc?F>|yg zn>L3J82Zq)V53OrRZ4xJYkAr<4W)-&U(qM|>Nfbql-Rp%H%nuz(b~nTdZ{R&n;8cm zRvMd?_Rr%#l^7%~Vyiq>>B0du)&W7Dc;!(NS^$U^pb3NE^pmK#;?u5EygkHPasgmy|0`p_0no2Lg$7Sx ze=S|0{hG*JmLJ#vB0H3=*b;Tuy=-qR3J& zm|)lw7_<{=mgva-sij8%`sS2EjH=;DTV}IN7)Y6b6EMFiMG*;l0>rx^^a`IXN&q=G>rF*EWyFU z!Qsbkvt0jt4d?@Uj)R*)zu+8&=Hxyoveh=&)lX>mL8Ey6m`HPxM-`r>RGhJr5d>^8 z7lN}^02=^+jsnF(T%0;Cus(kWp+UZ4pOcpk%-6~206oy<|1EmJ;vBzWZkA~ZP&g-! zrfvkPjuZ;CyBJ;*@s`K8Hse8-jbSbR5=`f011>fPWJv zh+hKOTEru(jdv>~&1dqgY65wk05mrNK0h#t;N|kdu(|}oC@cJVWXUu*5~dvnuItwu zr*PS)hv2~gc$)y=JOBhixq)@eLT6(y@hdF==>5ZJ+5iIpAvj;2+l;1P|3&FM-gAiz zRuLG{gy;6iVzbs6sz#CUh=tcbe5nQi^(*{o|6boLF3I6ZFxvh~9tx4vm_uzX;aD=f z=(d-^-!{X(0thy26BK>nxY4HgySYZcVcOwY6{Y)?y_zBl7y1G8;vO#Ht|FpJZ3Xby zzSbc3CsEq0TS7!HRHazCB|H>2Mt6R30d>3CP8PEDQcAsb~*3LWGVd-#x z2XeG-qC0P5dks(0ts)K|FiFzD&>>YCgyh?C-AsID@pKRvHgXc6WSANMFfxv*blMVxWuG zp0lgD?0?9ewxm&5(5i?w-8aXxxpgKu|6T6H25d(2veU|yj@bK`bYVUj$-nkEdPTh$ zs+liUsOobx|1Y_d!u^%eN~3zOA@QVl-M3ap={1}C+1E>*9yL{kYMo~?X68FbQ7)a0 zuUj!eA3uou|6cDy5R(4Q6yg3>lib5zq%O8Q(e8##xnb7b>JaBI63_kE-F6rW|EBt4 z`nK(Oh5H4Pfl{pAV6{uSk!k9q^V!c#x!XOQPhHm2TB*sxGoN^DhT0wa?7n{z|FJdv z!DWS=D)9JpGqDPJLoM>@zRbV8t~liJQ(%72I8BIFpRM2h>G5%A!t1kVB(Wq0iNSFq z+9sUndY>j7hu^z699I#Mr-~)Tv=xP3Il5s?nHIejb79c65JM`pN)vmPh?w&aRJfki zt?SL9iThon4DkWAGXj!CsD~L63D8RjQpmQMQc{7mHM=lntX)+(HR^H3)c#85uuNk` zM#fBY*(;mu{dWEeZCjunv!aVwg^i3$_$C5updO}wu(2(adL6&`%Vm#y zdl;Vn@u!~fz`Fi(Ys$vKH3HY{0U$uTwyXBi%uSLco#Uh*ue`FM7LBjRV-VA8{bDFU z!I?*{>fzX-?*W~2u7;(U_=D_k;krLfj2z;w)6_$loFjqOuI&BFw~qpIGT4OHb%a^w;}+ zE1Ne$aO^ZUirvfF=Y!FPK*PNg`$3||(=H|b_#EI@S;%h<&0MRqVN&-`koIixK-UPQVo-FUS&c1)#U4JX}5bWgc zJ4+hh*d+ajrH%R5Rcq|b^HNI=c~cm;KITcDw7-1SJ#a_B=y}^tN@(7OmXuK$osu=} z*T6~eDZ@_;Ofu=fAScUug4LQhA?^`R2t9tr=HCqk?!s#=x;8R24#R;sojL0W9-`QG zBv@I#fi4As{gq;D3rVOv^MD9kcyt>0fcg7E_a15{aa=Gc;=!=Dn8!AJn$6G*pcOox|kfe5>m zdDw?ZPiYyLQt5EtsJFx_pHp&@Uh+Ufv?J+Xnxb%sewcp4ot6O#7fd6?fzbmY{%hYk zV#p^?6;AUXkAL<*rK-W`83;~|v3P60$;C z;snXdzefJ1)jru^Hyxpg@27tXFeCie z!l_iJ+B9OQv7k3s5!7pfyD)c!$&2ap{3O>>z`>RTigJ0WvIjOT3N)6A%cTmO`d&Br zfJb*5D{YOJfi{H?&o@V>_``yZqC`zE7e!mvnd*V42|aENIzM?TV7nuQz5A9}AktfZ zuP=w3vuygrIBzaeP_PtN1Hq7lR!V4wOE47w!#h%|rL0eG6asDx~Yv#AR(bvqrO z9{haOxrNQ`(QRzjqH@+%ZKrp?X@I@AyrW6GyIAO|gCDp$)Pnu(2u2{T_bBNRwpm%! z?x?n*n-5N#{n4BJ#iGc_+AllwJKU ze|h98D(NMdaC5yQ&%KDNH{U>W=o6=!ZiVZ&6tTJnp>Xbngk1FrDWl#yf__Qn%&`K6 z^N|OM<(S|3){%2gMWU~qrNc3-$)eS>5Z5Q|_TP7+>v=Mu%c~>)=DV@BqGQ@;n+5#Y?wP;FrE-GJsJ-NhaQ0Xu43X1x9sH7s#`ai4xpd)| zRry0f4zX1C6VwL$-3#CMD>{tFbO`?xx+mCjE_T;Mowd|{4l-6?|2P{;QW+^Z!6Li< zOP9l5yE=AoPj&Wu!P#E0WS>e^YxY-v!=FnBiK#8q@-@!a58qP@#jG*YbF{jjvF#Je ze{|ErmUSiCT1MA%UlTpr!W_p74GVXGZZk_xa35LcCr*=CFuo9EFK*hF*>qgj3&jsAW96*CFro;IPa!)_J%Y<|Xvwk^|Z3Iwp~jc=>E zqMUq+JRb(a8i-4Jzl`me~#yz5bq{0+|^JW3YGNwA2A6Nz#y!1hOnqy3axH-TiB=(5&gkJuZ#D zWyQXb)W6MtUyBq6ZMb9J4M`s#Q5Jsdc^}lH=x0vov#bl5AGNaie_E6z--~>*36b)QG^O*3A&5d^ zgl>1BT`IiNIf$~ni+arpP(`pR&*{-yzXRIhUt<`#twwXo_@dKszOoVZ$zk_4CO~C~ z37_W-XbYZ3L+lbFuIJ1n?F1wJ)Zf}-&+Q9}A}=hyz7H6=2z`z*{xkG)XT^Shh_KX# z(5}iUxr)cj7?P*Qk-sBUm182phaQd+__d0MjSj8;&cO#ga5;jZ*IGP64~ns3@cKGl zD1*`z5c?`BkRprOT-2!+LvzbZssSJdw1s^k3h4KXi`0wpVG@W%<^QArEgXvDr1N(_ zaM|^Os2RlfenEf_5PZr6pN)esh@l!rswYO!$IL+28256tx1xC4?OaC8_uz?rsH9T^`@T#{d?E^m*WSdZ6K-gGkRo zB$jA$X{;@g$-$%v`149(ygIP}oLHo`J|LTz$Hd(F&j8kJp+5^G|aG zKm)Obq7aS_Yprv@yTgn?w#@rXh-51hohJvWoel;jW<4@ zB*3W#MukZlZbOMRK-L(R9Y|jAZXBRLm-fRp!Cc-PX84n zP4=DRy+7BEK(2B(*YA?t0OVmtK~1*0HVS4Gf>1uy%*nj!P(CaLhs!E2gQz$hEtQ}& zj`}L=wlg@UI^?8=$CC}fVcwAI-s$!6VE};cCWw zrd7;f#Kka9rbEfi1c$u?<{+4I)h`RzF0r=90i7QB7Ux)v+C}nn0FDmU(!-!Ka@EOq zdE+S{ga<(=2s|H_Xe<-jtQ5BNHWi;dt`RJ-+?w>lr1Jd`8~Qv+XG)3IJ;;C;#D`U? z-cP2zl0HtJe%4YzF0Rlr?Dwef@gYu-&-k6qyjqixX;?b;HlxW#pW7LLZwWwe z0`LW(1&)Jo0YW24{(!PuA8D-OuoTe3IX<TLCXD1uT|}Lr7eI zvuSKn`VQwHbS1aYM?B<5cUfqjrQk=9fi@cPCJ6BWh^EkbQ^PaTRaCCW(Qt^5g$;6y zRE*t}Xh&|gm~P25q1C=zFA^&YzTG!K&8u~~gEY>8&|n~yNr3vM)9;-6vrE-EFIG^V zE^DUdl_QDvx;46t_d0p*^cwaP`v}M)uxsM+}&E@3X70-iYpHrakjKyqk!#r(B zR4FQfHRjr;`Kn)RZZA^}k33*G7UzCdzNW4!4FS@eewNRI6KrI!%+{16&+I2g*LLjO|FoOGE(&)EFM2NtLR=aByD2&T1%K7ypknu?Dz`!&wc ztDrB}rgWP}(kvaKcz4F8I-T3rqaklcGIW&1g>u`DKi>sZnwL`T!cA31iE`&jE|C|b zENNoZm+bkrNj^F$9;hP2571g=uJ|D++Q}D00hPNBS!98@)SkEP%fgi1M2tmV48$O? ziKz@{Vs-vYJ2o^?RlskFRC0->cLy5K#>qt|KtN|%CMxtn5EA51(}pq@E-=xH0=<`0 zUI>~p^ECAyQ$)=gD3Y6|ZWb`~t|!f6#)QzjO^w9|neXxoaJ0$3+9&Nn<9BK9Nm6h3 zPmA1(n#350j`_nH|b51|i|1{kArJh2_7XA(d6OjTY#0Bq$ z%_?SDt~WIiO3t6mb#3^YN$$utd}8u|PENuJ=C@wpT=09Lt6UZkUX>`_AT4#TF>!V_ z@g>_EzJX?+&-JV@LA!GV;^4eE!p8NS2`PSxo^SPZtt|xPi(*?W&n37n;`%y*c)Av< zA1Sg<1@Z331p+AC$P1WJU^B($V|(?>i)AcTgMDL9&ewBGR+z=%!(3jItWv+0v}_e& z{df;RMW=1`R~QMIi>&0S{6ho8Xp+Q{Pph=|Urp|u)|;xLyVaCS6l<$lT*@%hh_dEm z!Q~#3W`=!YHm~e8CR;wOx&&8-eP57N6)>|P^p{=HWNfX)#DO*80+$GOzp&f@zIMJ) z2n}!UI4M*zS=Ln+%(n2?ekiHT{aE)%&jf)@xt5sQ zo0L-kAa30Ev0$V{$Hnj*&JW(06Tx06+vZ31rg72t1shsK;Qpqc>hdapltt615nEnJ z57l?yLJ&v_7a94zri`U$UiL^l&+~LNMVIHp^g`PAIbt*JV7lD_h(g}x4^4fI@e~!z zi(JgTZ#DPZZ}1dgO?b|MJ^~+uSnP|a+M<9ose$zp-D#UXV7E!`+BMFD9(1`OBB=GrdUstJ!%s zeyBlpblPxj->C`wz)K7XQ=3=P04;-|&I-7TO*7MCuc&!Sw`w-NLO74YNJLcrc0t~( z3`wr3U`FEAu-_-*VB+C`Qx<79zPvNJ&`yjY34437lR?r45-^J^k zrfjST>aZ4N{KiZFz^FYFz;vE?^lSc%?FB#W)e-W9T`Gy}x9d2Lx4wv)Zm%msDboy> z@7PaBggOaI96}HNag-YI&W!1I00KfTf0#VPKT*l`{>tEdW`g|e^UjOQFYyA$&Uz`z z&LrnS^@Phrvg_$_fglVt8I$7xb*>DbI?-gkUA z;)9{MQT!G%O3jV;;To@Zd^_lTpY!K7P^6Xn*0cD)fbH*f8H>a6z@*vsF==7b+_)%mi)Wr z-l^bPQ|m|lf3w^NPFMdY%l*a9So(jq+*=5x9Q{wrJ#}LQl14g5MCQ+~R(x?lbMQm# z!B3tm7TV82Q8(Z4r`%U1L*Z9HI*Y?_|E}nV0coo(ZBNbHX~RH@tX9fqRQg*E1>h7q zy>jr}7OPFr!IliATj#j$4S24^0G<0+yfn`j@J<5%gY_&I+uBZY5^q&?cJejxFlD&> zyiv0B^+rciajTzfc#S|Y!dyK*Df#nh4@X~**Z0j;sUE0b{qXvNfVXG}e3o~&y| z@rd(_RB`^9Y9rz#3pnq6q!`Ek4@!;_Ia_Ol`I7OFBT;>?!>|eRy_fRCoR)2H; z4WjsibT%o;rMvOXmk(^}js1cB^QZ-HCy9dz;g?p1gfo{*K=oh(Aeb&B3pbBaAJX3$$_3K(P^ zMgaVAIm2~9P@rog%+BWaiz|K|yDF{DdsbrZwdV+$k?7nounMAljaANH|G_P=rlwf= zrBM4maX$(^qb#bc5@e-)hF3U%W)9)dO)8ea$m3jizh;LWnBfr?F&_@YV(nVO#S1i* z%aE>89EvdJ#S1M zY1TC;_}L(vLnr}dRtwf$C^_K!5|WadAc>uk*0v}Wxaz3%Ca9cppKMgj%~kwLCAI#B z6+nI694q#KNrb_#+|TGKi%bPUDC2)?Cjhm0|}TqvWczfHsNRAB2e#Db@6rwJWa zz~pOG;2CfAWmL6_tS<)`4i7RV;+Qxc*wk1*@Qd3WeS7q0Z8$S{ug@!3c7*qiD%7~y zNIXc7pT;hpOcNAUBF4yTb~6}qdYskFwXVB_Lu;TjNExH}XKTaR53m z01jqsvJrT#IwttHy@lkJ7u~ZAt8B!T%s|07&00#eD$9g!Fjhsr@mggCHwK%){Xzk{ zQ?jyOTrN*35tPG%jTV3s8z%e4?&QRX5K6F67^lzwr*2drU3$tH8hrfC~O zSaMgBEK?XTms+L4_xdOb&{4X80Mb_zDnzOUle2@Em$X+IYxHhS7kVo)4HvrDX}>g= zJAN`~YO_@4-ci!a<#wVKB_?=X$e!y9@-5pv9~E7-N}l|(R0-sD4)0pgbg9BpRVO~S zEnYL~mDl}Hn0a7Ehys3#d>U#+0OXm3CSM-a|N`E#R=(=-eI$J#0v$#1Agn&W&<@ElCVKJsV24@#WMK0=va zTeBftKXB!>?ZNf+ultBD$aP!hUv&C*g+3v+0_F^5bppU&Ae%5`zF~o2_1l$y)WG~G zYqMKcs%;o-ZZh$Kxq$W|>R4q@#xy$2Z%H<^yrKVMBm%&A>~`S6=1IQtnlg;8Oj{@A4XU1QS&+eP;yc zl{I+1+-a={G#PEEY`E@k7Ct@cdFbWqoo-i=98P(aE1hbNwUIdMFS6E_i>2vFGbVqg zwEp01u*ni2p6U{NB?F`dtZqfFfair|2ECcAv1!%Eq7ev z2UgFfi`e)@P`!*s1Kv4kXAo?KKkU%#OWk&kn^Otf*#0=C9t#Lr)CdS8`f z3x^^1geAMH{Cz!Z!SVWc+xOFXlAr40Zn>=)nCmv%&yHqz+ZVa<*yUZuqG{QlCYqD-qd#bsPS< z#2=+i&KsH9)<-nP5p%!G@!4bICv~!KU+@-Rt6%iW^S#Z_k5_0~KKndR$Glu0FFhMc z7i_2)Q69VmME;&Qz|}+UOr%4sU(o~}xT7ntYX!WDYk$gi4ZQv`68d*B{J!WerUGym z3N*H-hyK?zeN@23KkNU1QGdv~&J7YrtH{0glE0_?HIn4z zV`lHhq;%jb%fI(`e~y37_Vpo=0Cd*Pu%uk(6gY_6HXDFVV4o6!`42~F+mH!PM3ULB ze)UJa1}(?nM1!}(8SD>|d9J_eCQGw5tkTD`4<1-2T(4Ru^ALO-+fEX{8QV$b;1=DC zy}r`5mHbPU`wvIaA6BFaZ?ghC|9U#oq`YJvFU!(&t0_gO9Pb}cX7rX6{o>r#}< zT$$a0$&w44GJ!Q_>Iyw$uF&sIV9coJorip z^zSiEIT}a&--?|_JfSq#rb2k$+2>CenS9{+JNMB+hGT*7yfa=m2ohP+vX4KETioMb zPxyFD^97dedXa>It?gxxiT&$h8ZV;YcPgzMmrDn-l!7PAhd|>T#z*T zAXqf*g3NC|5w$q!)QSAkns7oayO|*X!Uhi(zqQQT;%zG1Z zcWo!2O64g~=n5(LlVa(21M;bc$6@){U+?8f?L*!j;HJ{||H>+UH5XN`gz#51Q@(gG z0=jwE4oR+hG+$K_+6viGOBJXJ1b9r$*cvA72(s| zQiY3{I^7w$auA51m;#G7THvK}t~DBKVWDfV2VDXLNgOVPIe21S6|TGvR!YJ9a_&xn zYTVhjn@wj5jk1fdyu}$Cxzkalyv(PsDGFJ^i{$8#kqQW^%CZ^bxEjs zDla}G0u&`ITXG7H2(>9-rf9^WP>EaAd5v!-=(hijO!^9#@@7b5dJf7Ln(j+e89TgY;4M4kfRt`CkL=6$oWw2T_SYV-`Sh2h?Tw3l z0kZjPpm5>V+yp9m+Zb_)YDE{2d?(nugtWbqh}bSg02=uM0YK;XRly5%waFNuwG+(o zpJ$D`LQv&>w*WrqsM^odP7^tZ{#?axiM*iEs*jOVibeR|*-#?MM&;os2wTgAvY%fo zv^(Qe#TI^*_cbwIDzDC#-g_#XpFmWVFcEKP=`kY*_SBcg#!lK7%?#xCIIoWuTb8D% zr8o~}%-hGJs-Q2`DxLW=oRj6%+iO{mcSXwgCK5ikl)pKwF?PWm&-cYl(Xt)YhttLC ztVU2(f`-)}h52VUS`3|tCx1{^m`EUR_#KDO2=nW0HEapw!`eUrZ`>XBi(v}*tm{a7;tSCbHQLdcAvl~h1Nfq*Bx zx$Mn-t@>_rVkT^z)vEWgkn%>3YVCWpU_;IH?hUAJYFhuRW&7qnJEWERFRSl0>tO2k zCEbGmfqv>h(zo`GR@!I})3dwez!Tzl2o}4JA2wf9+nq$aKg+lB;iik3IjKFOt!d#FlFLDUxF9KYFA-H0{d1wh>a@v1cm3 z22B%~6^?$p-?i7=7{j$!r5=(hG$dr8KN>0-gBoj%yI?JangoA%% zdg3%bo!p&T?=0^kS~|Ae_F3r0H6!IuS9R#uE1xD}!8%Ivp4Bla;Zp4{hl^LXbGFv` zo1#pg40ImXq!uYH+BFta0uXu&_K>bo$u~#ioNNtEDrKI5<7M@a?|+AM>Tu4G+AanF8dAJzDu{rdXoF}i9#QxeKlN3YWq=Kwag@WGXwa@cN^5L( zA9LCc?buOirls}13c^lOy`Kj3favw70O462rG!r8Z} zyB2JT$&d3o^!HvZ!*>2iTO;=3saLnRzmsox(xHR*gxIi?B2esBActh#>imAC`=>qW zWak&7pFaz1b!-g8`KN4H--J|^lK!3CcMPe$l~w%w9s6>)!9AFMMg(Etw{aNweH#nk ze-N$65`{zkuA_4_7%7)>QZXOlaJ}{WxKJ*feGtcwRT4TyNM#v@O4g5p<`H&^uJmw~ z|8N4_6r>;ydbJ-4ga?Rv*}KV*{HtGh9tOD!!`lR5(`mb{_<6A?v>AnMkrE_#5{1)+ z;-80Ilfv)c>tSBt2aN_USBCVn1n~no_{e-p7W6ojEf6da^CnW;3oJY`IP5MA2Zr`D zJ7jW2GAx=2?j3|aMk;xM#{h<38-+v0K*SI(#RF@U9Dc@~*YvH7ln6%?|1d$KFy}2k zj0-F!+Q|QC@Im0XbD-$UfL3p~bhfPNHospdBX%`f7#<5kXEIp-F*P<2qkO89mRDRL9Pvnl%K_FoOWMAjH}7>SQy zG>BMXQ-@uRHCm&KRI(I-ehxsl1Ypl0j399JIRG1LykD061laXfhA9F});R}~9g72= z1JEFFoF>_rx}j{um_I?C1CiwgUgLR~`W(W*18W})M?J?9$B@YpO&-ILLph*- zQxVo_=aea*T;>=M3XR3_!0)pRzO@V`Y62Oqgmxn!pM~ub%9QGW%-;_pQBh*mw9yQ- zQ8alI@F58}(4-n8noOT$J7dTUYZ~*G#-1M0zc{qU~ep;8akIQ2@_{MYJff?O{sovUIpnNNjoI`EiGtJlxg6=maqjb@u873;Zw0027Zm5lZp8Z&+ z1e~^3EXy+&id~k~(uHYy9?sA#laNl#31GE?i%JyUe*{0`Cn=iT~FTsWvi(93?fR|9B2!m@&fK4fQ z69h9(By0D<#HcYUOD? zMO=57mH6#BwwcA(l+9O(O+I(VU9&{2vDo(AuQ6>=W%ndLw zwG9TP7Ev&&UnKa8Rjeqa00HsRSSdGm*-rAAIH_`$00Th$zn;{z)=Z^F5RWQrpd)`z zX0?J$YL;MCo`GmaTeZD7N4#~a&MHgBbBb;`!e7zXg27&yzK2BR348%+ssLd-X&KG8 zwEm94SXSb0qt&xw>I4IHJK9>VQQ&)9VV}LKlaUWcSyAP25kSj-{b0cG{FldY-`N7) zlV(EMBzwXE-;?^!Q4|BogQkUXTRpYF3ApQ=B;+J(5h^_N4kzvqC z{rIYLy@F9*i7pO58?ixYdSw95b67(u1sMjzk z9zBf3mhpzS5t4~hj)sE(fFGEdV`>eYw;MIr;C8AivM!KbR!E;?Tdir!cRQIJF^K2_ z;TlYJ008E?*1ctXRXt?#y|T?jjZjKV&Bg^%1%prrc9iBq9_;wovE{NKI$VYDVB)x= z2Ay8I+2ZnS+}=Hn`uK#VghGT#l8^YEyj@i5Xc(RViC z8KzM?ZN~@Yix02z%U*MY6y4+5A~^u|{YHV#&coTAI+%3CmqCC3bgU8S@jl{n#hd$L zQ(vQyl%#c@td*CprAUg@!}B4%tC$(yaPnOlKLuT|mBiO?ZMUF*<+JgiwTA&fBEp5v zPV(z^h74`p0v8D~HAYqjDL`!y244ek3$rjgVaAS`*(rK2kSXKfyYWT*rEW7fuq=^ zeRIQ|V#l2kd}?%Lu{X%ucy!1pf$%6HXvCjEi^rgot#Sk-p_}T{xZP4(XG8+4_@G74 zppj7x`?|vpX--Wf`Ct|OomV9dgK-x&2P^-u7&K{o*SATjUc$_K_M2pUKEd*FBI(XJ zOonW2z1QyL1ovIL??j;2zD4ka_TS3SOq4ldwmMA-{g2ohqWMH!L7Itcdgwv%e8V!C zwGbRmr;}FEL^H4FvVm9$8H?=!ew0D;$?z%jkN2q5HS4v2!f76ysRyuX_R&y(m5tD( zrIwA!&jg|^PVK(KSLUu}+T%S0z-A>}(8OL)Qv zgezKb!!qT`GQTXAQe}H9id~6#8BWqAFQiT~bgXJlG7Y>gPO?lQ8BVhiKgg!$SX9=W z<~np?tl83K<30LV=c~qUo)tQ+BApk9ljUlw=u+tzYwJ5NFO{Br?fFk@4MSfp8@f+xuNwQF zF0Yz~5SXr;$MEd0Tc&91u3P7Luddq`rI>D%Ix#d@nqzgltHnEwDo^=QdJcD@IAORP z{mlyGH#?}tYgui#)zW`_aZHhK%HG@4|0#aGC&%RUg06w72~&p$`VV6i03ZY016Ke0 z`Lns*8x4ZxKv|&=E}M+U`Cs#gA|Uq1^R8Gf{eR9MHce>mKl4{}zb0A)3_$_OrC0qU z=n0}VCo_z}$^CCZk7^C;Tv>Fz$!Ik60~2pOB;E#6X}I0OvYn-SG^2RFP}Jl%@V0xS zne=mSAovyK$Ju&eQzIdi;opo=movFSgRVl6*=#oZuXC?%gLBb{Z@>Bc+Rglv)!7Q? zdOOxXH@|+)MECI$+V1p!tke8?eP}!w_7Oq&Q|r}{I+OGJ^=s$-X;F?y7aco{=W3^Zu+abe1wea3=+lxD zwe8b-kx?jm^-5=8e#3ybAO4ZSyU1rzXp+!98os$G>|K?fQt7)SqP`D8p>&=xDsT}T zO`b?^u7J0-zLP*Cho1`|poMbCu=g$N`~Vy6sDe>AZ!>aenvj^GM~6mNq9NYsY-kc+$=KlfB$A3?S|01 zm@0P|SstQLdv+QYFI$tH9x7$0VyRo2U6Npca(qzgjKFYF1qA!pKjs)Ybpx#-E<1-Yvg1c$}RD7%zSMc zdx`-{yr^ewZk={jb^H@EwzWGbu~#9oN?$CRHr{8>za;T$IpwF9X4hAIkDPF5yPsjO z>j30!9ZGF+uKmp3GmE{j`iOPCk)%SZu*z1V$($?3{6PcXC+p6W(D{S;sjIQ%mRp1J z4*?D<|?xGQ@@ti&EHu?Eu!q z-?WS3f`s?0oW_c~v#Nt9YwJdBl8@0S`^V}N=I&QBkOs7sPKYx;4j;rGUn3S!v7#6Y zaD0Z&cZoK<{Po(mJ=5?~jQWNHSwWIic{xg!7mU;v<8pjG$tIIhIRw7CC7u>)><$F(X?fgPjCH?WUX z!S~G(y&ogWRQ3d~_Vu1xGfE9nQ|=8Sg!@&~)-~xvdk$kioq>FnQ+^-OX$FdHbdeT( z?K%h|l>DATvuq7}NBZXwKkZirqe&!V5(}sdaPk4pZy}BxGG_s5hi~>Jz(F#BYlehA zAh5xegRgL&7#AFr0iH}U<7>kxc#WZW*VXVsSZ*WXbBjz~D5kQMVZ5+~Kcmion7^PT zRVnZlRixMdn1C%~G)_lMdvyn}@Frh?x+$^X`a^&xf6Ap}b@=yX74h?fBv)#&7#jmh za;XDMPAO1;FP3rj2-48DOG_~Sr^Or<2}bzHO2_ZRrT5nJD2eORsb=BZ4uJmX}*1@9v}Vfm0jTO&yCP?8PCdD3fhL!X~IlkK_)n zks$AO_E$?k?ceu-SgVZ%pQ~vD3))0+BT<4YEGQ|1OCpwBun7Pr(hw3`sO(1!gy=S3 zBV)MYkM1LA6c1%b++6NrDv=9hQz~Yh0C0k8GA)tq!znuAvfSNxt1tpj45p{-vc+LU zRuAlV%R^)&**{0LPyj7JHn7tnC~g+qBxvOxN*)8GaCkpbR8;LMi6aBDW>zM(%+EdxLt?yyren3rEW z1sCh^!l^#uqSN7DPKWTf`yUORGQhZGb188X0XM5f1#_p(7(J?&+pY{Dl*xLLXjMJj z7a&Zv@z3#8r9+}){HZ=tEc>^ilkv=4@^aasI-0GhcNhF*vpIH9%G1aY%`z+)3UmQ1 zoLW-cbB#FLWR{Uka$R`(w|X7SCMA_nT_QI}9GZK6~5(YLWzvoXPJelILB zmr_U*Z<^Rv@BSzD@BrK;ma$s&JeN>(w|4G=DTU?Bqq|>vvY44%wQvmdkZi>MJ0;iA zLwS}Yf*s61t3KUN9J4z^C?Y=@yp)JgIpbp(iTtzY$Rda3Q`y|JzGPq9DD$?O%IwgF z`(UKUaW`fpnRDppculNdDJ=J2IAy@4?vc3cOUihm#nN&FUM&$uYx( zbF|scz6FT)!3sw46}8lq2Anb!SkQfRE_E{yGk1uv`RMhC>Ye>Pj&b?rdVR$UFMsms zgPlO0-hPaQdquX>(VPUw211&xLD-dZ5HTfhnVpYAy~r##8Yc)At5;hqKT>>3_ak0< zKON2b!{29_xFCthaj)&RPn*84cWrPJcQD%8J!*yEW&PX-f-|;m@0|xSNXXo{hcQ9mG1_QC-CHIofG_ zga3Y)mbZI)5Snr5tl`#vRlfICjAt>u%MC6!ip}DuV77(ua}5RQSqrsyL`NZZ5%&8J zEUezKm&8qdQicmy-M4NHsozDbyLZ(LnL#;MME-u{eSpS`U8RhJwoLL4YyaA{I? zO`pA4Mm$8LyfJhFPpzGxQbCPwY%!BQ2-yK#;Gkqgfg6<|m?I~4e&>O^&+KMoxYIA3#)v;oZ!fTN${k%!?EIzj6`3Ad2Qo=luZB<}VR zHZV9i?SxsSghlE_z>C8&iX%x(Jn4x($#<0@pOexC5{BLqg$~P6AniS1fr;8Fh~$f} zP$`sUnhXac)i%6U|ORU*d_~NP*ZmzlQ{ZLRk#(b$jnbnRzAEs8i8w z0#nYk>reccft;^XsU$C{NKlGI%LV?R#gUM;2?aZ3gYO&EEB_3l44CyK&~hN zVx@&Rr%Q?ngmK$Y`lVCa{UW?p^NOMaT&;bmS_+>B+MzEQsZHZvK+HZd0E{z5sAcqX zRo!4;!+{|w2i?J8&*W%osdG4t^DDX`HG0TPsl0@22(BItQA{B#xcN5TQ5UJJr|FIY zlEOLS_nJ;`_(r@48mQ{Q?eqaClS#_R(w$|JfDTZ`1upVPAbvM%Q%5d5@OP}TZd zPr7uEFqIlE1EmxJnX>D5phLyEERhmL)YnMOTStwOtABN0G!UBaE2ot4BHG}*^ zPsmOWi^_CKT#U&{W{Ki+!L>m2Q<@PPMIK*Tv91_Mp(lrCrf}{a@}rV=UZXhg)p)Ij zRbr+%-8az)ItlXNrki(jB)Q9!Me=(b58|jwOFi&#icM>@lsLmSa$Kvh63h|ot&Ezf zWcN<1#-~BIEA6z3CY;Ws_o+h7U~V|iIDe@P&;tc@vQ~JZkY7vFDJqqMW0~veFU&hY3p)XZL zBq6r@G~FI}r`0sDRHG!FTQ(e)kZX2M7fobUb=dydbrX|!qwdbMRR5lPf2^70k$bbY z8ZLs=wYcy?72WMPi`h==t4{Ya+s<3wMw8AClP({5 zzUekzA6uR<*tnoNZ+MO_VR!_yXgjz zej6c5kD&YnVNsk#@OjG5bH_ueUbLQGv^Jrd7f2=!wtkpZn|y<(pHuEfunb-p2YvPF z0WM>tpQN*m#U8Zzwv_>p>?B5*&tFrCO z@*vv`r0K=6We~cdE51H>m@%C2RlWprYygLcHzSU*5ss!94|fD28m}{Dc>T2YI@n;B z1|(6uqH`etg0bO|?meVpK*-_rbweMk+AOIshcQeXACAe@*VB=6P=UIbRY%c~ra^dx*r5j(<(7h@?5bmtvE#K{ z!@yw-g%ycA3z^_z^cA~7!aAn?Ii+aVZqjgxV}PC|NO_#>)>IGot-|7ahzksOwCH?{3g27)pB%*2hi0 zAM_*71T!!}Dkr(6=e`N}=mp*`<%gNC@vFXz19iF+AGv7c##`*6!%ewNpOGVq7MrBBq~;bq01f zYkqHfsG?5qV59<&R&bHAY1VJC`%)e!7Qli(zM=e@#6C-*F_9iRE-|fD5TL_=lZnuX zX~ykXJoaSy;8UP9S^(;YIjDj=x$eXLt5_dYBU!3PKT=^CZ-BMxSi@K9cokuF!+v(m ze&$-8VsNmJkl>Ea47lF^=b2eg;j^PfV*QYh!_c~KrLi)59ZR#zk--z2t3&%jR zWKR?Kwi^G=Cc#I~&4XlroEFGBNjWuKlcpmY40D4CzEE&JI1vvT5~P#z2(xA^*!Gb` zk}{{BG)bEY_}v7{X0)}9AUfYxCth}9xRVOgB<=chkUAlL!rIcmRG<7x%0h1c1*ri0zWqnc2djz{;1BV}xhocQesEw3y`8b|{ zCQqELKlKI~JB1(Zi*OlD@h|m*PzgCb2L)Z}=sL@uz8q~5C0RIQl8G?>O!4?ZKSPPh zKnd@BqvLV$P2C9n^m$7i)J^!J83Ha=r(CtA6xAcY6sE=}8f`0AkxMMi{Qe>J;rc13 zKKeVZl;4|1NZ#x9X__+=@Ph~e1x~eI0XG9kotUgd85Hay%?5K^QM7P~@>@ss)4^ST z;jRT8e5RV~MIzZt)BAJrE5H65a}RU@mp@{yas4;LIuQ$0m{Q?B$^1`@!_c${8)Ta; zglMO5$HGY&3*dqKuFu+P+#Im?4xkX-7Y&6+CzZ%nIS`9PhyUYFpn50~i;KEPw9#bT zx#KXt#V+4Lx;wdR-<9c(;o(pD&^dV7P*SP<4NB>Wkcy}JnL9@Pq*(ZewtNAKgJDylEEc)jLaXJ9 zo*(1%?2R&O)cSet_IghajfvKy2>ptgRi<10fn<(b-FGWhDbjDXd;<1*b1`{PEoQB^ zv#-74;**7el&bnq-TqzHOujGa$ALMRK_6~BpSa+g*Ga#dslEZV_PUD5pzqJGCs0Io zQ5eKyJDM&8&P5S;WQ*I1;QvRMR1`1>X#a1tXY6-kepM4m7<{TX!R{9AvHv4X`s-jK z0ru~(gKS%+V)!${DNeV#=tAEA4wE7pW?)sPU#P(-ru?TcsolzVNRz=oP)=s8bci*4 zx#I6;qrnj8*%^2uZPDExpN6$vR6a8ym`6M0KY3lZQ(e(Fj05)FEvgeh3i0&oZa1-s zR0jRM{|x2)e}qZ@ODN~x!=(QPlC`E4;MQN06q5!1dI7$+Jvq?@w$RH>&g;{(+1)#T% z9b4<8;Z%sKqo`PBxF>KXX96AFKyg7><>*3i0L^CiDkUIi$W zYX7t6Id@x`||-UeLLolaA&orogWZIfWcoBM#Ay*Br% zN!ksjNio>Jx070W9uIX~M2subFIElC(|Xt+oWGd33xeie-TFNTI?hR+)>-bYc~l7U zK5;F6xbOY7Vup?Lyr8fSy?NO6v7gbL)4_M*`pSLez#tVu_uH+X6Tk4+adZ(HJ0!um zZgo$|`zv<~x%H2lJ$b&r0=w;U3?1j8__~^1mQ#)RE5h=bx?u(lysC%G#Jx0)}=HR>k&lc>gdv3`-pqHIDYF+_wKC|Gzz5jQ8sb_}V3eO&9POi}%xH^hA4#tiYk^K>uI zJNn@{?EU7f=5v;0^CSFNHc}ha6((?HvB6aIB(s(TGY=AOILCZF%;py!0UJO^$dSHe z_mOWY8wwqlRASYzJcBI)Na8KdF1adjW8u93bH|&1;K^_Y;$>XM(@2l z1bN+(w6zOyZm8x&-jm&>mWkvt&2fZjRTD;yt}((S23_U!Eq8%n%lMuniC;(p{AzbO z?{XDrEV5JRxeBQ&Wk^-g1g)>bGzooZXN12^yDS5Lrb6W@2mYM&t7yghG?>FP{I2*4 z4^+#54OxfPt$}{ATRzGL^wW+vOD^&Xw}x+^Z8fyk3_{!_H8Sg>Sg7Te^c7c1mwKxu zJjWKwki&C;|YdSbWzk+%l)Xo<)X;uh0fOe@bD`xAWEt< z3bB>Zq**Dgjjaq8U{ut4fUXS(LF&>2bM&Ga!f3sPz}_AF@{TKX_dOKrwucd(da64ZF(;U zV!+V?beL4KP|NFvY=q+a2H0DH&9o=SRz$c-koAZ;sp`xy>Lj)=mC>oB`oZDsAYE`2 z3Q#{9I`q_u8lzrpaE}E8ca;(g9#RMY!t2Bd=gi$yb@>j=?8bSfun(!%LonN!Qt;+# zm7H>l*~Y)}dzNMZVkJ50r5=ak^T^#a`oWAE?YviXQryXTl9?Ipq-RoW_OpF9=h1(w z&0;INiw+)^+zojV;|iZvuv)82eyx-oM9yLJHf`Gt1MqhfB(f#F)e-w;r;w(IN>L`K zpjpTq_LXB#vApYpzHdeh0jV7ow!r9My0VEQ7u2W^zjm-AQCX2Ao#o+0F`FcS1i=6V z%HR+*QHjv}nTT6^CY#+c(J}mMez(+=5DSTT@xHrNohE+jlZH7tz_chQ`&!>Awe%rW z+LQ1HWTPgVKXZ}FyT&xQk;~=o(|8+_F6*O76m9yowFvCt7;&wOMR4|h zQZp6U&dDQd6DGBrB(Tz$a1tEmfPWk5y2j6LD5_+;ySX@z49ZHj? zJOtBO;1xTc2#G_GzP`7$4*c44A3=K-`_r62leuVpv zVdzVXZ9#1&8+gDe9#4MD?xxZrBsUpBxfb#p=|nURnbG`jWJCC5Yr znMQ`i5hNWz^gbthaB#G3SHN!1HKGvncCs+h*F9{JAM0uHmwuxD8>-d5{lDoYKUM`Y z9|3^hA~3>D6Nm_}LHy!q(~HC6T157lQoY~n6L|OxiNu{z58S(tT329^mbEl<)6l|8 z*yg(4$$ecFS-+6QeUf~vj^Yh2evwTvzY322B7*Tg<)??gkTCrv4eWZ%+eRhRs5#}| zu*s3?FgmR&D#e|8_cFDO$I%*35On62DUV6~BvULc{UcO4lz@n3fkX(ZM@dN#RR4J+ zc&P9CnFb5y2+jYb($RE`^}#Q>?8EiYV4^B>GRk`FoPdXE3p~NIcTh-|Dfqya56%in zhj5tELfR5lwAgJR($s&64SYb>ZHXl`VO9YXV4>xU}{Hc=wuR643SSh$% zKKdXOWa>BkXcRom&gEnH&C~OU&U6GZ5QQM97!u^g;U!5xC`F)FT_vrc!$xW2Q#WDK z2gdi1vL;xg$9f50mo(|Mz~XW>0p3Nl(-{U+XoqkZhJp=iI3n@Z@MDG}S3}6P?)9ZS z^a#Ove>s?iN~6Za9Ld0be@~(wIPiVW$*EI}(Ko19Ljb%XT4hjAP`nS8tQjbc9~3_o z{qT#L`z~7D)tj2angJ3;<4RFVPZV7ZqKVe?d;tx4$DU;v0*m464lwuJOd6#8!qejD z)S0}v#b@Cp%UbE%IazzB*uveJI=WR64#EmU*asfs`^s4ZZJ0ORm=PsF&Tt6>?M$nw zvH1{|01Jf{9F-vtc@9Y2C_R@x6>G2uYp_P#-b(^9PZEkcg-{2>+LYQxwG`I9^0hmp zDMYp{jK|ZP?Qx%U^dt@!CcaNf*=hyzDLJvzOfr?1=qrCBZdVXoi}<(fBxWuUMU<4= zQPKiIvX@r!XAAFL^+e$@g;+PPcQ;glI}D~yGwrd&@6b{vT-#Vm*`%UT)%?R0yd%ZG z2a?cZ3?Wr``DTEgSn84vNLGjFX^2hNT#eJs(qBKG2j zc`^jTP=PaW!Nko300@H|&=$kC8$>v+RfDne@2g$8wZ3)W= zMd2>&EY?-7fO4-QL=178`mOV!`!c+eGQ6jK0j z5t*VxWd=C|2Z`vuMiwegfPA4;>s#I+0V*ry(Y8R54=%&lhr=*|LtR9+GB55z&KX;W z*iCaR_i*Jfc#S_}RH*}t#%@cs2nbFmlH01OQA$bynd-Ven_Mztaf|}VIz`gVdVTx&pP;; zr}3OdjFnL_Tm|Z_pV_Jc*)c5$GWy4|kf-G~q<=(!sA)^EeD#x@U#_3iV_C%_24T@R z&>N9YF;cfSN-Z~D_cF`rG@c7Kg;q!J9mIl4G<731i7F%N{|_FJ9PkUU^54gY=x_1= zUp%0Nv_7ILiUvb)#cT+(;y>}V({VUhSH>0F|L9dX9|-~`Gb+w_)O@Q{Nch(kPIG_o z-*`YC`_(2Bf3FYnYjpsC>9<<+%}0FgF8u>DL-{V&$yT2)Tf4LPW_yF~zfnj7K(`|P zzQ{~k`I=7V^===;rFy$LjYor7>?`90?oJm>KOVs=zFlp9bM449a@2URS!)jKB_;mS zxv<&pcvY#Z^-J?Rbms@D5tg^p;gkr`cR`TjuQP%yt|%oTD!Yau8tw1Gs&5`Y_KE$V zm?Q%MBX`I1b!;2I-;N${5lfZHzW@32c6+8iq3k0fX^hmvr)y``FlMatj)SnOE-$na z`}rL`y;$@aP2(V>0}^d&{M(bhNTNfhV?P=#wfn&9)4YYMB%RW#922mgfGQq1l?_UO zkDyOZgdcAQB_b^8$i<0P5TvFEQOF*rDqu-UsVyUCVM}47s8k3c2t92FZ}hJoOFb9k z<3j%G*wO|AH5o*O-kYZ@0XDh1$s%T-eV5YSa95PsVDEG*MT1Zlm4)mZr>#p|u;2i4 z&zrJKv?|LzxMtZ5<)>mkDzDC)1yy)1!S{f1f(oMv1sRY}Zp z@Xj}rtkE)!)H@RUyU1mecV(8Btduu#wbcdJFtsOIZiSd_06aKesIdU{o;AT1U0M(Q z&?~oeF<|{e+K+ysch=~J^>#IWgVAf1gRJCJ(+FZ~ukQkOP@NAi;u$n|5TL?c)y;DV za`fxO670{3YUQ=$eWVhkutX{saG!cc;dW~rLiB5%w%|_WnMUE%PJh-kuibd8z;)@xebKT+_gyyROiW(iYuLKnx4;*EYOa^o z>pe&8=-=56#Y%5o@&4X$H|sa6z&U2}n}uw@4$JAOi50(~yKe&Q*wp{Kqrv&uN|BWP zM>a9DO|fM!)78LA4d(s1{u?^m-&&-O{c!nt z!2iQoEp^}5ze?(O@{2{u(j5wtNE&V13AcsP*qgpTc<>K2Z+1LU-{xcel2eQ;aia1R z4P~f8)Qacff-B4=RX*#+8cQ&HwYB*yI#w|3Ycv8Ot?{TF9;GYUP;y`$Z$*rII+U#WDdaId ztG{-n*-2j^dwo|MWH_7cKt#ZKkmnrM;qM1IDrE zCE_X_I~(p|ghryfJ0tN^bC1ZJsVmD!8Ho_zj8e4GSZw)2RPSe=Wsii0@=5zFFr;@T zniNbCgMIc@!f!b@)x+UwXSVKAs%!R6!){muNrsFQfeFm*FH~@TQpk|3#|fw3aXWdM zicd*}7oNL$U!xVGN2u-DUR14Ll|zY~Dm2sXP7F8 zj!DPUdMVP1IuT2(+UVJ~E7mBk2B;#+q+tURb3AHEzB&3kGe*@;gIg0Ye_9++o(Z<& zu9(QXkQts%0c%ayU*=zZRS@J4?Jm_sf(dtkfk6|60!)*tKFKJI)@&_eeQnf)ae#BM z6{ukJboh&8w?ruh3B$o{!s$QN3PM(_YON)GY(^9a9^U=o#fM@=sgY*}gsaEntDB%9 zu>Rf3*%Rs6VRdgWt#Y+J(t}@*VJhv|e`QE+bLiq0dX~E*Z_@jY((MuQL%6ADe;^jF zS$d8(tfFoad6g5gNHpbTJ8QV*<};Yu;2`mh>0sE%zCYmA!TAPNZ}~HI7qEoQ`6{ku z$%vV;NeC&(d&iKx*(@En!IGfuzzf%8S@lh*=_H-FIPueSwTNU>I{2djgoa)sTu#c{ z7B$?_q5BJxiL~a`@bNEU`5C!2#@MS+ILOTkw5vw_s`q#nOxQ)xZsLe9)|eiT32ROq zwd9s`0izv?L|D%()&3U0vx#8u)jTD{T)+S{WkU99!s21}L_nfmfg9$7G z_~$B&iMMc>w4h#Vrnq91U9F19q z$NflI-Lh+pZaIG_;x*o;an`Opiel0aNNKY&r6e3-Lb9v3%t`brS%BvpR=;xjwy2ey zw7(meNK*B#5w^6|pIoc1*_JH~(jq!_%^FJB0QIx5ZvF9%oTaKHhp~BE!fZJgvGn_f z>8z?nF|;r}V5ta(06?=hG&E`(@iBksfI~Xl=flPyd7F%q`qHc!Uq3GNTEpk7e7tT! zw+J9XI^K|qqB2b)qm!6oU+54aNhIgi=AK_6jlpYXxjVPhUq;+K5p8A$kWcX*2eS(} zfXz;Qlpk7;8sdK^PA+w}*`}RYl(cRND<4T(5Faf{?cFO?o6t}fGTzv1=V0JP-}^WD zyWtM-ajFEGZ50vFaTU~@3lz*GCq65DUlMz4zx()$X&l zN#uEE%f*lsG40^z7VjhicRdF^I?6Ve7?qtW8CgCtHBO~U&3CRgByJ3r{droFLjqxi zTWA&i&2BQ9+rqu^vRp4eom1+E>QIdcZM+3$qA1({>*#!G@(_NRPz-3#WlNht?aW2) zA3PRt1SS5Q`B_y8X$e(bsdZRKXxngacH6~hh1 zsiAg9kBDgTh-9gt_J)DWqQ-KV!uOQGnxX@^pMyR5LFv>WUwVK)f7BGB5n>BG>_HTw zYZOXLRCII{DkOTlH98FX5)B-VMueciLSld^G70H1p&B3unV2#a4IoTxDz16J3aCgX zwu~QCV*{$`mQH<)ttE&9vLiGi#bQwCMx=u+BxPEV;u`4zW%SYTEpc!#aphA4G%c~s z>a4(rNCq9>MS_Io4K+9l1?=VEU^Cy1YFd#nUf4&4{g==Kg2W&93MbhSMBs$em(bsbXL+YXMT8!u2W?03^5owX#)au2 zCpak=-aKue5Wz{;A@~RC5hzX({qseHGOX=ro`0IcNBFe&bF7n^Y=;>;!K|Zt)j8?Jl35Bl zdP+HY;TQP5zan3vGO^Cgbf<%fzhMY>8MkL3Z6^z4%Ca-lxxPtYyQF2>3edAtVVjTP z3Z>(_K#}u#`IO_eFsG7{c2C?ZGm#L_K)tkCgPvj7rrL4vp`oveeZwh9Ty(rO0eY+1 zKn{Zy9Gzown@L6Lf!YG!)dJePf=?DCG&w~FQ6}}bt{oX@H$xa=e5E@T$O1I+oV{6B zzp_5d7TJUqoy7R{nP%^cli5EP!w`i4MT=jkKoGl9WHFR?QHKkya(i1E50o?q&H&7V z9Nn$}M(Daa>n0{XQ3WANdJl5GnC?eE6f`|N^i&Xh$}Q$`ap4DAWQN#0V!zy|%TgMc zvgdKJJnZ zdf-9=ZjhCn`G;K4TST!=D%{Xweg_RQx@x7?OkSWsoKWXRaJy=r*%_|9t?1R z4q+6}m@W-r+}c(U6XiRcwbDH;s5Lb@ zFU2CJ97d+V_ry0*KUnU~H%GS2iMyg;Nhf5zd3VxgNX;MHwBv!g$b+`=I=16JyfM3_ zQPQ5Kpes95U4i3J1Nkmv83k%{PT7DGF9*q1BH9^tI~%^Z5++gTBQ@})4j2JZ zG-_6akbW9r?Hh)?bcilN#zvBP`^3BsDEI?F^BrKpzT{?bcVI}2V>dQtN~a_mUgo)8 zP`Wt_3>ELMG1xolKh`k~&0|v9Al*4z`JHIumvJ7qanny!=>+3^lpv)Y5Vsj>!>yiu zCswOB7JJKtJKChz4wfxP()%!$zWImW;K{Q)Rh}0LY-UDW-zoc*@Hl80NH99dL_SF& z)%0Y&FI-KMGri6>-86cnj{|O6fpZ!O##_doB=s-^#{KhfX&$Ba(j7IQ1xdawT`e6zkRwCCCjS~cnr54N=bJNgPl_3EYPioe5bGnt*i3WTl{5`Q)LeG#AbonVf0mI z;Ld%Lprw?MyX>G=Hws^zs2<7hOJ5c0I8ay?_&tQBsk3QsrvE@886ED5X_||pM;}JO zlUaFl2#VisTSTp4-K`(^XsL-jfZnwu984@dCCn?gA^qdy!c}5~HYJT_SJ|tFOw*SU ze~`rj`zU@&=2(_3eXM74to#-x-?~RJ+5%>VEzq?-ohx8oMWb!MqS8S8K=#ReEyinQ z^xI&3?t4J*De;zgiN0Vj$i2>Y2+YhX`f2A5q zAxa4YBL$#v7NY6p6{z(lc|l``+AK=Lj7k63EEeI3fGrK^Jss=c<%PbU@A>nW%zG0K z9ADb@Y}2;}4AZNxcYp2jAJ(_an`yonN^F!)&Hu$eSq{?B2|@3G2PjfwKC0PKOMGlt zE$QcG#Gy>r+?(v*c`?KN#8?vjg64jR zOaV_@I>~`CR?2CsXfsN4NTE!j!KZK(kA}oyyGTvZ7Ar8dLc13p zh)5V-5sYnGuvuja>I^tm(P0Vu=;IK+;FfNhn_sm86=Olx?SZpVNyqLHWAY$;><8L-i%h8axm zv=dRh6N@I3%s?Ys7^38^5AzG@f70R)R6n!AQ-P3jgD20uRV!^ZF77{U!tg;d5%Ot9 zO#e=Zg9KUSY2=qN|9BES6<6#=Ahk*K?a3GV9xDvC5`s%jFKk zfem0KDx$Uw4qB!yhdPpC1w9NUBgwcBe6D;3Za!D_mah7VlPx@9E+X&L9ik}CvWTgq z&zVNGJa~UMol?mIEGZ9(hW5Wb)~(l(ibbH|(*|sSO2v%d;o2!*l*=ZJpW$#Q2$qo` z%EXHAs5xpUavA+P1L1_~rwX||j?49kXauD}?|)$A%CU8XVDQwp(olJfL&oKk*?wn7;;Vu;pBhO60+CARhU}>xLFY;@*UA zP_ywcmi=m8uIo|LI8=#a^KEDf78b3K{e9+zMzzCY^E1MS-pWlcfs;!Ki=+>_Yoy|g z39$rE>fs$wFctc%X4#~F6B-FSXlKZBm)7(&BL~07=5#)X7>!}kk_Bd^Mx3qNAXw~2 z^OuisWZh;z4igBA*vOtLW|zM14&p017rVVW#*K_Af;!bV+kZp783#j}-)4#ukN^Wf z{J;9_s9<^YK=t)v|N0k2rlX$XYUynINpt)e9NhGZ5Eji2Mx?rH9z)-U~5C9 zAgLq8U?^{q%AnL>r3r%To8^T%9AQpX`Y0?6A;fVjNw-tBw@ja2yV?poJ;*iE4E3g+ zv$0?U=d%1JMyKlV1nijmGx+nCX=w{Au-iXh#(kwFg^5oP`S?uLDha13s5Xp`c*xwq zvrXN;nBmnvu0Se*+cBBRM%+KIYD`|ht(u|3d{3<~y)&hvEYbjGQ2p6ft)ZsOgUf)b zYPQOxtsn8IsbXExp{4IU&a#s6*L-v6KC;*H~`iDW8yQfBwolUb1A2M34?=5DUZMNqdp^moNazMp$v{ENJ77L0|AALE%s3j1DvolP?A5hL~WK3+j1 znG_QGw1s}|55_?;F$aoXjs7mK<87dP`F*v8o2Z5!=sS4&oN(6E^_@&I^d~%$K>$)k zXRi0zT~=Wz5^J_e=v6}HwsXHKzo};*IC?SQ@5bc!sOJbIN+F8W4Fd?UGCD63fslXC z8zT|lJnUfe5Hpr>V3Ja-F0h&``UU2->@yVkSS}Hmn^^!!LO!MpJ={e+Nk`M_F@gi? zSeG~xpmb3|?gh?{vVUg~mb>y(UhHMT_R!NLJI502wZY?q}1<&adCLd!?B8D7e@+yW~1XQv`av2VbtV)rKOZ-|dtkSF12(9D2 zTBGP{jO?K6XOIslcJ|fNxD25BwUl-X<2YWFnr1DzH@gT!Eh478e1unqe6#l)+NNIH z+G;y8F$TJ1_koRCK)$N}O(AgJP9>?klGsM17mv}&sp{{Wed)09Om#nqNS>i}hS9zO zFfYth+PkRuyHON5z_+%x#%k6^=zHYvA5GbR?VKRvyS+}urD}W}R=DDRa=T&jf?v_>+_K@A9 zt$9$Z8YZvy{))L<_>;4dv`$pV+ur>tHW2SC)*bWXRKS+GeB;!s8K&FBIJ&Ec^Q`0S z4!%3ox90osUlyOrGfvyw1<1O#fh!;D53<*5>DU&YwFLfn%(WVvA7ozq9h&T58*1|f;ckJ9j-CCJBbK# zR3h~C^LhgciMW$3vmQZ-e%)6W3R<+14T&k~u>Zl@TfapCzI~g+P&32?T>?XQcgWBs z-67qfbaxIR-QCiif=CO92uPzyqcli~$n5udcaL{>e|X;AJ&ye&?)yHk>%KnMX~&I{ zNHCRsoELJcOO7@yV)&~h261=WO-~J>lf5!d^J>&c5{jv$#jY5KQY9M+ZRbk^#O2EJ z7v2C>pD|<0r8ohA@CXz$y^!ok3inui)Or<9KCS}62$mgK@5cyR`EAt?W zj1)_!H)XPxUd2duG3_WoqLjQ^K_jXxnM#&pOcc1l2>x5wylDAt^@7*u7zIRxkuqxB~dUe=OK+W^}s2$bsw`GY{(~&Ab3Zmj#VGIl`KYga^=j+wCJgbEn z3<<3E3XWS_+1fTuM}?oAYXBu+#U!QOk`5Z>7m*n1hI_B+tNYkk8tz+Phv{Sh*$>;I zOcj3oWCCLg)60s~=mtFJ${!R-0t)9&x5`IN4XhR8?D=Q!VF} zvy&$ceE$~62#=T>5j#mVGQ1vjuwg=Bf38&K_H#AEY^VT8iHmeSUC}>hHk0)TQuF@B zRmD)m$p=7ygV8G1)aW!E`g%(B_ZGMHIQ$6L9@VMhnr!`YE)~9nsc$}BQ+Y4nlk4Wz z>DPF*%dfAxYoO;-f;?8*kN_<1R%XQwt7EC{tY)7NPC8x%dbM`ajB?7W%Wu}SOA4wE z+(h{kXFczlJC<;yZ&e$*mW^Zp20U1}g_(bHjyt&J9b$<4Of}-tXePI~>DRYY`_Dx1 zu8~mMebiKnWin&{DFB9Ya(T8{CNDieHuyMrYW>^y$-CGxq3^(z z#@?m>0M@g;Ii(_MafqRlS2kR==q94yYKdq$H-nE0MMKIUr)4~`Du5!(RV`ieQE~1c z<~WvLwDY zNNbC=7)Q|Mo@{QMf%&nH@L!QHc@a@q4h@k!B2lpvjcO>x=xCdq)bByns2fbK9f^|8V#>L zcF-qNiN*CUM2oxTDoe*xb{!Eved~0-C^>t;US;5Goh`rB^d;{C(-?`K$ch9NgwTgTt?mseomabFhYL1pF2iJ8Fb^a!l&uZ1-Ghpei8WhN~YuV+t+T8HUkh@ zsE-d1CKD_}o0sa|m6lBxZ0sPvNNmUENwdUGp6y7=CbwG#PA>z`mx1%qtr`UUbd-#I zatl$mK;g)HqTzU1(Y$DyaqT+{cnP!l3B68#^~?DoyaNU~6JwtFYUBJx*B5NjlQ%J) zKVt~2)nN1Tz;qNJHUVXhZA1iC*BI<&fK>!A?#W8{9Xz({-qCS2KB^-^G(%ZkgbQv> zd#jx&Xs`93f>A`+LoXnH)$2tdh@E#-FHFTf;-`9t;hXboO$c1++D>aYKm#=dY4YPY zqfemS(agH%%F9dO9`*w*B}D8;z;R4t#ZpDv3Hyc3hF}(OXALADq;N5DpMsHKD(&!74_&j@_#XXFBNRot`guZaGpN1P9dCvIS4PMx(%OI!vX7?e1)_TBtu^cBG0cNCzr(7 za~2K{U3e*8L#S>8-5W9kcB^&lNTPnHu1^T_2NW@)O`?gvGL>y743RNU9l)Ht=$e@b zHvpw06xfuaiprTK_a@B42OL;`)X`fWy|Vc$I`I%$J{o3thc0x4O+BKF5NNH}m;)oZ zEauFo;dz^KmR`Wp$0az=6b!aR{wW|V=KxQYjMpfC-OD-|g6P^=xA<}6dO5ZB(ren5 z_n?XU6$to zqmD=MT1~15Q1ETCR~xc)hqa|b@8~DkOPbJA<(gDwXv(_LJJ@m_Q3FYC4>yrTGnaINL2C>lREMe4sk4 znZC8)vIQGn4MZrG4C6iFwHY$E0g-L6zv8%(Fl_Y}N=^V6p9i>shmdSPxfCZzA{}(` z`_|FEEAQu%M|Ax$!xGWq6jYgsPUQc+Qv6v~w2vtCZ!WBss`IqE9=hE)6%;la-Fe&7 znQAL+cFE%?q5N!7R)~Xk@>0YY-e9dj=PDVz!S!MDD-|6p)hVibGrRaZS zk$PsYyZYmS04!WmqIZb0iOiwdVmNn)HEvSb{Xf)DU{-O^?~T zLeX3DP{HjYa$^NuPa~K@r-3E`gLMwEQgYoib}|uI1N6@wv_WQusKX?$*u{qE`pQfXfZO6w{ESRj)|yRZ%lKmg@(r2dR+rv#u~+t zGDAhAgRF-dbO!E&6CAGwbTPVQq?FY<-)RdB1247X8>k$>>Ht2-4Zup-&16I|w9~ae z=8?h6o0R91-XQvK_*d#WO-=yENvzuv%sZAK0tpfIAGVRo2J4Kbn-@$EPiUp{&6_!3 z>gsE9_43P&n%hIeXb5zMAbqi^LskZ&42n&l-XqE0hQGb0k%rHmb8K?b*>QzB6ny|B3_WL<*51xc3a)+AZ#=vt-)yG=O4BJ6Pmi!NDY|xf;+sqm?xLHa z!9M7Pxkel()k~Mj;DZIrY6Wt5Z`WQlr9B=79}82-#ZpD)XJh$^~^XWQe}( zH?~+#Q-(3fnj(Z~Z(4;^&BySR*y>6tM>ikqcGp!@k>hrzT zyz6~07ji&qX-%rvrNCn#W2xFFmz|Iy;@xYn4c{YjCGq0E@}wc>Qxeck&tbU`p{`SI zXbi7z)%zi+rk-LJiATwjt<(baH|DVMpW(p0*`;JtV~vp>?}<_UFo*qiodn}*-63gQXjuhm2g3w2mZaoT|y*$wu7I?x6!dH4%r1^Qrz<4T${kG z4fimu_r4|W&~Hl~s(E3v!5-a-4%hZDX4}_I$;I2>?nV*0>}2~7r-m?mpyCL#3))v- zDfO}rJQIsKu7hn9!y?Akx!b;p$@*X8(+o+5VQcJ_nG!q9_9ryDT_rg*UV5*;JtRQi zN9>Rkk`#q$KQ=+oJSjoBr0cC#4=3Tdx4yuK!AY&#un5qX8XsMaQdXAjGz#|Ni zCl^K%&DG+8_ahAOw0XJ1bJm2+@2j;7(O7R_;t!nKJk&ktg~#gPpIimw8KSo?g}z+k z(qr3-$$oy0Y9u}7R6iY-8h2!Crfp7Xdcsz^04UD^6tw}07nfUMSOgc~C7r5z0!^%W z8G+tq1w~RWleY%1li@zPrB(o(3xMJrz|sTXXzQF#3xLVnXFsYum6y!o)Nw^aiutzL zKa0-h?RudK{&6aU=t;+p0)mcs?q zG2&k51E4Z|db3k;m5UaCe*Y!IFZyjiF~@Uj^s$uUEd+3eeF3mQU*MNoJ!}tRz2=iG zx(y;+H4?#Md#MhyW&4YI@h|1wao;>u%NE$d5`YDJbezL-I0pc9hGzQ1u?g{zc`$1k z41&K)#|G!tV`2Z;KF$8sq@5d$)qUS!-6JpA=P zU(UugeD(do|9FKI7khB^+DWilCgCRUdB{;GyK)flWL&{Xeb2C&)+7v@+K_9+!dmv} zD)GYTV?OOX*`s~EY$m1>Ers(?)y#(;MlD0GjkF5IgcuC(&e*QqwY1Rh56`G6nKh=e z&#m|Iy6H7%#nGWIRx3P_f-30k>z};aY?+Yi=P||n<)d-#=>E+d6O+4Fo6Ocvh9?=9 zyhV3mzl3J}8%Fr4%yMEa`nlv|ziRAEgtgxJofWJxRM9+1e%uiG9Dnjn+Ip#ByTj3l zx0}oUEYe8VLS5j4+P(*&gpx+ppjt(<2qu&(jd6{L@r%p^?IWn^o}u4O?);}DSU$G* zhzt1@xaTH}N!<8GWJ$#hJbt&;(W;npLaaW}qVO;$(@7J}g|JkUN^6X@+5;(G7@lw- z0Oord!mU}!^P;TozfYuKTK#p?QSU07IWEW;+f|r8FeRw+i9WXQlYJB-9{IRH)uACe zHOi&Abr2#bFMeJn_(kGwuR}`)jYabL(vE;fTb>jT^nVVW(*k4w9{^qdU2gbYqRRh6 zPyBx&^ZWnU6PLJQ6=nba?!T4!+2-Crl*lO7Hg~i9vFE;r{IPF|F2!DN|JLVlu)x4d zuFqXY=Df(L1>GYkwEr1Z>G~W)ml5P$tcVzzf6-~YEjn!Ga%5C?*6{_-Y&I&U1o}IuXKym*S=hRN z+Gv6Y*=%({Gfum2#13kAQ>1Qx?xrGf*!R*D$o?$yDshO;%ThwH7*bX6NbEE84g0F( zGz|Z1XDZS4UW%UP8nYUlVl7d&5w@qsj@h=B#_7yJ@)qE|>`YsX*`|xnj$JH17ybGj zJ289M`p(-@;3Ye(g6@0Ng2R(X&RSK6!SwfBd>xK7UmA&D?7ABUYV*12%e$#{wIxTX z()H~w^$qod=m!fsK#5WVpr;@)&3*uH?zjaWdY{@#ho{ZeHUTMgYvWt~sxJLcjPtbP z-}P@d-etMM)2^HQkwSm}=XYz4+tP(;U6@O`XQ5ES-cB!r8T;gzw4ys9WgXwHX%kstZ)dSr%F;@R&5f_)+po z^Y$sYWxB1t@%)2NTQ5SEkO(zD43rzYnEid^!_kA(GHaV7^8fVBj<%yg=(9R!Of&KZ zaSHpQ|CJK}5M4VyA2jp-e!kNFPtYWpNw`*f#K7pZqZ?*+BC;WGI3!V_WsxP>F-m&n5+Mv&A)d)Ff%44(*F=u$gc)Bz29+?{Bp6rH2Biz(No#{ z%PD9{4=TAwkNKQo>n|;!(|Yj6hwG(B;g9j6uC(H80t;x!%2dcid$tKQIO~%+>amUR zOY9f#VSo}S5(A6WlLJBi!dfn4?)e?No4agb=|2}Ep~uU&(eE+kg!C7Oe_&vOv8uj& zMT_Y3V0>;3a~Is&F<r8nh?AWpr`M?M0f#O{igsUVkG`NAle-`1MJGZivbUaTX8TJB+MMnvW`= z=MYwvxtR_o<*8r;onS2Xm$O){ga8&ySj60zOgrc~|EpHSw$6{e+gPcn)|5BAG0%vM z*yY#0^U9|GDm#n-U>QDQ#sit?Da&hU*XI|)wwz&H8*FJmx#OjElO6E*6JSC2(a+X@ z(!kPM6D*(L6#0Ax9N5~(CkmbHBVv&BgLOWn$ur+DI3W+~V!~uPEiS&nR(8=Aof>rS zNrSdBKwg5d+(FH#m;o&W{3EaV_L1P4tS%61rx|3}>6tR4%G6;nm13G!P&jQ8F!Nu| z8z3xQ#aWZ-RFLxI#F7Ng$2v1 zf8#}nLMRR#?orr%Wj3YZNC^ZcuygeJX+|}66a%fc z4SryM2-2QO3wUEJa~+s5#;oyji27e&wQ7`M&+DGfOHDrAvu%VQL zG8eotjMOXkIKQHq2?BRRDG!vp{fzo2-jrWNQdn|(yMQ< zO%_iPF9Kl41*5cmGl84lmf|>LoA~1TBM@+O$f+IV=Le1!QGnc?3svC_f6xgmkYpLQTRr{xwLagc zd@*v#FP9I0=%6I8x$KD!3PV&KdQDp@-UofN{$A`SuJkt~S zvz{1hCcz_k<2rPm_AJO}fI*teQ2pJ~hRHd{eFd>);%Nl8F|nL{^>Q5xYbJ65$? zzWy6k>cH67ceM>vc*MZKeHbgA0}B!}ZK~k@`uM}(JXrAWuk(>sRXlDYf2>Ck4xb!m z0sUoTFk*J7S=_=?O;8^5Gr9Om2;s?PoZ~6$7F%rK~*)MQINW-@Ei#j~wI-?2>qu zZeDMC;zR^*PrJwU2x(Y;y~YjuXYGmi!ULex8d~rXV8O>sPvm!B6>KjGb7XZQLh0CA zBd+f7xcvS7b0c7F{XD~P`+V>xtvfd+uA0`EV^S^S3>f0_la zv)AUybQ5ABWN!>aiTxU^do7h7Wyi__YQ+P-k|8vVwS!Xrd8eYL4V(~36wVYG$vych z8wsy}VXw(}4P%>x3yjq2q(Obf($aC)ev&%9kXYKYkegmMPSJaQkZpFPC=Snzyh@HciY7Qt@EFs;m?_U{$Oo`b#xdo=47a%ZTdofWUj*RGFJaum1wY#QKKDL9}UV zT{lxeU6N`d({K|;%H3K(mG$a>fRe1gzHO2=xUp&AU5-r-Ti<1`_4TmIm>F!mOeJt+UdaB%ennj*hy*w3Hq>ph3 z_!Ji8%nItRT_j(Pe!Hz~8vCJX&Y5Bl25lm3L3Uc)Kq2$A$!kdG+3@=j2Gz|{wYH}%yQR&F zklskn4fWNG)kiLo`WvDUH@{gmdz%8_&0p>{{&=odIk0@br7S#)>T)dP9DOSK{`;-> zfpyqhuKyT6-ZL5G-e z`X-s?&8L5C?i>vjed8IVO>WnG(nVz8`kmW#O6a>wYjX9UgPKhmX|t0_?jCZD zE7H`Xg^oW&_tv1V3Qefc^}UPly}$I#nUuPU9UOCpP!b^oY5pCm`twN4yKRL0yex@M zdOB6mBRp}ym8k#o*)RL6=2?lTK!JFpOD8o*1&3h(Ks4bx9$kp@=T4-99U4JMrSi1C z+(GCy8;y3&!Tx7cHDLWZcs(|A3dE3O|4M{+ z_uJ)9Wx?=o(3e#kbgNY$iL&xHuI-doe?XI?u850ZPM;039+fmO%bc{z3tHSIv5qil zS4-@`*_aRqxi)ZwRUKx=O*>oP=^t2}kelJXoE|AR!JaKn0s=}Ypu%7}8(QUA@#{=c zHKeIl%UH^e;h%vrNGIiaJ_$WJzOjs(s+%#(gC3hOdzU-7EMg^y`Q~<2_&cTgo1PdG z>K0?V7Ul%UT8{6m@b*eli#WGL8fN;x*k#|48u;qzB$EVDUH*iE2rYpMMsnlMr!Kg| zT|GJPF=HC{%oI~6Np-K)uY3U#t7nw zKKWVZsp@3AW1L}=6KnZ7srGmd3|Hb-3m=opI(=-L2aC74KRvzylqr?arS{0fqs-$sa)oWB57XipbN8os4 z%}y1rrt?|pU7OiW-tUhoW>tv?sK{D_ZNR47zwueU7HKuFvF*fcL66#wE_`X0S~>BvvfqZE~ayZ`9VFW-=2^ja1+QoZ%X+{lM}ju3|$XRlMuI z`=D99zz5^X@*LyvZ`4~XN^FVmJ3_ZhVZcAg(yPS)unkW@-%y$z%|Jb6nN8G$wS*IG~XPW5!!%#9UeKaNq z{D}lJ{Do^j52Buy+Z;toO5&<5(sxZP{eRR-}S2 zg`JZH!dn@MD(A-7Vm?EN6P9zPIjXxq)*RAjE<`5Kt{`*PB|Y=kX6SsbWS7UA{OB<$ zh5!l!k?AP+#r5eaLx*_mk^L8RZZyB+LE` zmZ+kgREZh-@rD4&Bu9+z*gLmZH1f2WM^B9=7vlpUJ&5q$%|T16O5mqXiW@~<8P|-G z%;e6Iy%#P-^|<003IA zJ=MmeMPg>|_ts$p(_O=mo_JdL*u*b%S>1{lk7I+?b=Pfe%O|-MJGI zMe>w}!`BS?m&Us1`D;ZPR}i6dr*>Nw{u>8t13>YK!uF2E)9IrEajwlz%pe8w|Cx94EN2C)ISTaTs1qvLY*(|bkq51}f=@4?3-qtO za6!i&UaQ>!=_`?O%P`GF^JJsQP`yAJVV>hG4`q3-3`3aS1f&BOYs!NIWE@nWWOXPa z`Ns!Vc*qZ`jX=Y&?^iA)`}A@Typ5s04}P@e*)wu-c?)A<^XN*)TEB}T zf_qO@#$cw04g;{>qOaI+DS{$kVPqmAshXj0l<2$A&l<@ui5d|F6B8KJb8&s7}O`Ved~#&QHDL}9G6y%;|}68}SA z^;3Wd4#MK}B4}ZbJOTh(*S6vn z4b%8(8$S_~>o0V`602s*u^MT5Pa%FM7Tb?&(Kh@N*h-i{Pq+XB&(jl5Mu1R!8a?q^ z-}Azl@6=gFxNqk|f5)XM944h?MglT}>y(i+4#~Fu3AV%`#H@}2EB~qPQux3i)P)bc z@0Ea)+!Hz{MwB!hlix=X7nA|W_+p#Dm5_ijhv^_i$0@^Jl(cIM7Nrc-8?5l5aTIGW7-?Zy>O4{K3g_hPXYc0A;M`97nErbovTd8O8x@?Nn_nafISZer3!5GXK00IDL*rpjHm6`h{|ClQ!C90&=sxl3#_%CUCh8g9s z%hY-)J?2HU@v1YY{FNSblnbh>Yv5;rUM~}MdSLoUB80{|9o~*EgK&+>d~O0ZU8V@k zfSqaAguIy^=E?x*p2PDkc#AWEN3Vw;mKYF;R~zF3lY+B0MHn`Ez>?5ZQaBg@{CT_t zs*-F(ojXIFznV0T5W;ng5v0H}=M3n>`RE?;Scr&T6Fxgv?9s)#t-0oBlOl~!KQlF_4Tyic&MGTVP9$~y`~U^0uRG)iJ0>q4eoC?Dzxh zXJQW0(PVKkH9MiO#!j;Q4-q+MI6el^(w^m?MeDbLZ0C4sMSa06?fMMEjvOXgVPYLz z{L-13jsoyD1@X2uN;syH2v9q5Dve-1jUZjy+AtdVFx+;s(pE0k!h#ub6#qO!F-=zg zzlIkkFJ7Vk`aF$07P6O0D3`*$b458}KzV2ZeWo&NKtvh12AfE-y0lX%YmAk6nk-Q2 z7k@B|ETEb&UE7CH+~!RJmeZ%mnY>&ub#Y5odI#=nsBup%sgOoO?Achpf`h~5Lx;#O zm{oqD+5zt=kxKZ1!YRN*370@cJGf@bV_ymR-9(B4-EBgo<8e|3E_C%F{FC(Zq08ly z63~@sL(DO50qS+x87}~734Sj?3BIxF#@6ZQTq`a2hIej3p+EZ%&3tsW{6RkWdo|L( z6_ZErbw2WU2OozKG0HB8+KYVIOm4WU)InM$}alHe$6pX6e@ zI8MpvrH^VLYo&t*uS#5h0dOvkZ2w_=j9iZEd7&VXFF zkTU>m?c^9%w7FN77VKrY^8xjyAdwK<`{5yNXI*7BWA)sUx&f^Dva`3vaDR#cQ| zW(G&Lq{9>yw(f4T8oM=dkh%PQ0L8( zgkNjpW5g2>SM)*H0P%-9L%)!eJ{gaUbWv-0gq&VD_%`p3<1TQp8ZV%$G(ku=M!81+ zu)GZNc^(uj00{$|Zn;ISBt>oT>U|Tv6U*3PXgL!gYM1|#Yt%xp=Xjeo1;hpHUO-#O zUlX#A$P#;16?GAurgpSu%x_(qCSoRrUwvFi2pIRF_ad;hZpGQ&VL%oPp)nA26o;KL2$hm zA;eB(=%i5B;Qcyr@i14Y@D+xzzwNa$z7@n*Hp{_A+`zyh8WRx_OU|k>clmabX0b&t zCH5`?KivhkyJ~+*E}bpv?{g*km?(u4U)CjzIHmWITY>Dss2$W_E{nJ_d{^qwU>oZ6 z(5z}pd-$5@U@|b%dXJ_eTCRe(49M@k+9+yfw7&_#ojO4rqUYmzoi&cJ_#8S3O#oiVx6ECa=KQ}^N@2>1bpNa`^%qDLK3v{>GPEU6owX-v%jcFRb}$wcv&Pd^R3Po~ zfM$pP$tV}m-nkML)GFMTFcxO^!0sRa45itxPfcPT`fC z@t0Dy9L|^BF>BC+={OY^*|B`RL(LLt#BwGj$>m?v^c=X${#`treHLP0*w%VK$lyiaj)AL0Krv~$mB>3zig*e-YUP?b{ZB1C zFq6|5P)qm`6hLhrx;gMZ;;{Nj`tI?Y0u*1dUjM@g04_h+zJtOf2pCET)(nJj0-I~k zI7=rxKzv{6&DFb0Exg>ja)n|No~5FD8PkQ4L&aM|8z zTET|5Z9?mpvUu!1u51G*)B*N?ZQS3h?3!i?t*Pe#A_V_L#k|c>JR?&QUgAWt+y%DW!rO35}j8cZz40z7uH!yY53(=ZySV| zD78F?63ov2>~CRK6<)vSoOL-~0P#lWZrUPr3AG}ttM#-n+%E6rW&AN0H>&ipK5J{^ zx>R%EmITj&1Wx{BEzHabLG*C3(|Co>CXR$`&SvC!f9R{ft3$?DHh7QqX)vlV0`}V(#gu71kV6l z_HMZRv2SquOfVC#QjHKtj0 z`JbOIj0$G|udK%Z`-F~Cn$rKDsn!3o%BC@@m%Ipk`(Zjy(DlFH?BDOkT6o^F{(o%t zf8E;|PW;&Ow_A3HA|3xnFMHM1{+#5dr1*C$UyPNOjsb$T`|(4}VLCJkU1-%U_o#X% z@0U$S@0cpG2~jaR1a9!64c`AShI`{R-w{iXZMa};+iH-=~(l>*OGWdrkoPBQtA$> z+-|=@t8-?>s=e$$J_n9LaSu!ZCU`C!<@z{B(2B~0BaQ&OW_12RU3-CAove?IQi*g+ zoXcEFTn&d?>lo0?F6TKOY_LODaV=Wm~yu~`?`-t_17A!)1 zuBPFNcSp{#s8JQ2wvV(FOPy<(P>6+DeczbzS^d5%Z?e zl~>c>drFxjw+5}qm@X#9>x0}HC?rkhy4^?2t_FI@vRv)^)*XsJ_%;1i=AvheV*eaa1U)%pA(M$%&|;W?*7)oC5b(P ziHxYAsWjE-9ZON^O~c-1@Nu7L%ZFRl+de~yXQn&M)1aP>m6}b5Z@JC_ac{SSa_;td z{8Q{_#(}g!;gu9kx5#9CZQK0Z|Iq2>M6O$ntYJ>q<_kWX=88eIw${Icrz2Xip_1Iw z?U$Rec8<|348}o!4)9|J<-bk2Z|A7!g3xgvXef4$k4pEqp71BGrR0^Tp>cV4(gcJT zG92RH9Of8dI2KGu6)O}bs~8NgJ(d{zQyKCqnk%t7hU7gR_XIK(9_1_(k>2E`+D z#9*9O>gMXbS)AdE+ONb&XfZyH$wNrzXrGy`{YI12W@J-(GCn4y6Im)|056Y_NE%Qv zkD%f+T^$CU5SV{^UU?MK(~fVzI4^*#D;sv^=9K(wUBH*O6KrRXm+*`NcOlgj^hF6M zOY)>ZPI?u3%KXY_7b5Ca1^JUENq6*&&R%W|#iZ#en9ewO!Tiw!?1gZVvgqZ?PxBbQ z4CZSr*>vg26kOI|C_O>Q2dzxj2yyD$J(jA?F-636>AWG@t}5rk8Vid-dE_Eem@aa& zm8S)mb>HQ-KKj7Vwp?inP%$hO9WzW6RTUeTl6WINw208SOG?gh5&Lx z#rAhB)Q~YN#o5T?H?0Qw3f$mx- zt$R$jD~o0ci!`0xeYD~b3y3r*nDunhh>z|}OlVe~&SN{RgV)82aN#BAW-wA75-2Ht z^kk1|^(y@0rOPGHF!Vx_v80mDN+?uGc0+#RM6VaDM(u3 zYx9rk>uk%;_TbfLP0s)cK=!{@>3{2Ial~Fo39>Svo@~Pv=cp${I&-%J=j6Ouj!BM{ z1m1QCWQr{Xbq6V6!WLGQ!rJgqY<7AGIlStXC$Y>S7C~i}ChBkuJZ*Q7LM4!&SB@(w z^M`;!V@u4u@#`q|D02ZsE`nxo0&L{hI-1g3&Dep36s5Bme=j6n~a z=)%A%5~-dK`{TjgUq4leXJT(&HLY7Hq!lz9BPYnB`c3~E>{i#$@=NmQq>nyQ%Eh?XIh*#QDti3Yt0X$**v?#?UN+PKzKyR2dXNN$*RlIh}&tV$L3 zR$?U5bfMWo*b;xFE2>d~Dc2#<>&%VlsQtARakq8*NvCKUkjP-U!E~`CO7ntP0L9SH+_`<~iVHh)74(D^*O0 z)WIZ^Q#9=WvITK8TyoDlef7#8Jye~>M|j&+&4$%`-7?>e?k}b|!Ir8Q`0QSM3!tAD z>8>#7I$mRA_f>s(X0T}vdM6*2PAfxde69)dk@o#P(S2Qt0@2O|bEkczqlG0H#~0O5 z*aNA0jq!BE-|s3takla<_{? z{911e-v6a><72`-I$E!O>8oC6b%dDYdhvr^HpHO8RD%`217ZUWm2m*8MxDu&q^qhm zk8Un{;vhl}u|QKElp}{>Q(O3tImpvo&G|V=8Q&lkSg5>+r!jd1npNC>`L#$Fyi|`3 zOV$fp7Gj(0P2Fj%z)UDIp|Z~B7tu;cXcI-AP2jt)G-n3UyF`2q zlzjoY+Qqp%Udo)NZymhvtXL-5udNW z2WGl|FxnBmI)o7(!SS~+Ajlxt_TY3lQiz%*7x} zl}xuoh8Fj>S^BUsk_g8euTK5f+d!h8OEXHpv@~5`y`OQCKYa=;Bj6r3e2a2a~ZA9*_`XW@~T?(kIux@QwF!Vp2(2ee+K=@GaWQ zqNdYcu#cITT_2eb^3%tsiYIccg?Wv{0CLmv;}INeG3 zk-P*bUfczUjVi=uilZT(h-&Gjry_(1JI+M@%}$KuwJ2;&%*iGzn~^{Ct1^s0jMX?g zlojy`f7obZ6}EcE{0}pLk=*f*Nn)$OfHVSDk10aJ!@U0%WnbIrXNG+=}Bhvxcl$Y!gY{ef7%n6*vPF zw6Ti_3fwI`*4mv)UUMYZ9Up8k_i1pwAF#84o&`LGMA)Awq#)$o!SkgM4)-k4TWyxx z+$<=Q{>oqy9}~9EL7^Cy2*WMaik!2Lzw;co^D7ROU%G^ONzo6vq*$-7B#LpZ<}Kt(03z;oG=mhN01ylZ2)~o{T|JSOvjmp2gsQ3pcCUowt%UlbWJ;qH zcCVPwyp&3)gf$OGhmRqNs7%xy)3CT$!>i0Ss!T|Rfzmz3=mxV(f*KC19M+xWD{eVh zsQg)?Tvrzxj`-{FQ#k;r=t+yC>E>~rQSqArpcy6!n@X1F3tHkNQ`wa1vz3Ee<)U+)Bx6h{;^(6CuBGu&r1Pz1+VqesOoIqTLQG3>{Bfef%S7Wgd0_YLiJ>s$ zA?ekSJUZzi2@Wo^ymVU5a#~XqM_%*v7AcWDE2ehXMBW3j297Gyg%tW>o@PHb;BA$I zrjVd|bq@3AG9Wz#w#8S18Ut9WFU;659-&m#)gN-ASpjt&dG-E7I3dTVh`nsT2nE>@ z1@k>Nq$;O`V4B*$zl3k zm^!LqMqG0oCAqqjURsqCcG?xupvdH~{E%0hyWH!WR5dvlc;5OKa;^*11UzIolD{aY11QDF9_uNuQ!n}eqy z?5*J$SFJ^2YmpDFk^vB7uU$2<^ES0^gQtmsq{}F+DJMZxX(jH*h4{ZoKDb1-enq+p zMH(bW5PBd2LUCtYAQ0mQd0nzwqb;1VEqrXBi?pwroaD2hV_lw7>r+6rz6Y(UCn`Lo z6LYVB#4|E9ZH!;gT>TQr2?_22QYG!V{9_ona?lu?JNU&o?b4OKy26n@{!9SaIS2+R@ zihI%agCkbHYiof0;=i>M`r-dgNXwx?>;sK#wd+!LREBjU%=e>mwg_ntqz01G2-~#f z^$>Cka1wQG4>QyHq9WV?&ia7Hlz=9)-q`#>7Z$#+wr!qUEMvX2@BuxD6^1i{F^eob z1AU-B#hr&NfWIW);mr^LG95b!I%9`oOi>*YEY;ycx%2m6Y1qzd!=~v~V;1QMvVJACxiYZpo9^^C z;4sMY4J9`5`C?iJoHM4<&XVHCk}CQNI2g1!!F_eva`RgBr$)upN@D$J<>_A6M0E9$ zAjF_>0T`Ve*g@I;Q`vKYt`IWo=`Gel>b_<==#WNip=J95ueeSwVB;6Bx}L{17LK^ z%f2U&DXnJzXtP-;dd$wc+U&B0KciLZYLw~mLteetp8K`?J@n%a56zEbuN!!^VLbg} z)fA9Iu@+(E+G;hK#q?Zf)YM|z%XmuO*tOAQzgYIUPUpSH+cGql!8L2fpU+`4Rk&|= z#=CJ~AWW(3iKYMLChNOD3-p}5%X22f(>3IK|J&NaXguPt#vKaB-+L)TP`M93q1XG9 znZIQrUwa*I&;UOV!zUS9hODIObpns|hLzlvlLImVMHON87K zNs@jl%4}h&7x%Q0A*;A*LLmUGhY*U!Gga zknKS!Cm*7DG@fd;X=M&|08_;zdS0od!A;Lj$j5wSZEE@*@Itke3(Aut-uYlXXZEyw zLDty2s#(XD(D7F5QL^7LA@P)xRo}47J9XJwS3>)3!Zjlm(1JV+750H;h9!QYBtC#*^-1y@(tL{qMWBe>*^F0 zrs-Tm?R_FXsxW9q0#(9OS{jwdYk}7k_xBv+`(?X5BHmTDM9KMy=c3rHH$yf_>PHdrq?pABscE$avpnkb%C>W0g^+D%?j?>fLUj$4cFFq8~u_R||8T`hKA1!P?d zRbrO!kYiRr*cP=AyEvqeN4Ju zx#gFNgU&GkhzPtMu#-}iNiS^B`<&MP(RRVEug4!Zoqa;0cqtP(e!U#qI5QDUV~V|Z z-azPA7{jovD?G8+DExQ=hXXHM<8z66yg3%t^Ywut+d*`*AVusA@FneS`qndkwO39; zdghw4a0B5FpyZ1WM4I>vZb(4KBE=MK(hq=UCVxT4PeNyJJL{Py(a9-lMmtP?B|!!B zWsv;Ft8CQ1L<#Ob{icnK5k#-JcU?Z1L~D0Iaa3FcgtF6jF3}x@a*?h~zqQlNN}?;# z0_FZ|smZJllc#|W1gzv*X_KS88}AKZMCpWI6kJFLkboRvZbLq_5ut~a`tb3wk`w;7 zG-M*$_YZ_aeMn}ci%7L=b)kMO-P0v_GS^AdyE|&?~F}_*J8ZW*;VpSg$9&%Ka=}3=sH8OwA z8dWSOOoW{(h#mEY?zY&oT=)r3HQ9tY2xZ~(vV?wnBH5A7Wxw^5!D`jJvNO$H{h94a zX-SD$IfDS6gsLYq$ol)|PvS{&7T33^KMC=w2eAiy9hem3u*(^6sNh!WSFjR&=mdjE za&*uwRthmj>cW__B9fUwk?y<|RIl)WWvi9E25**%)^S_}F5U2DZGdl~ZZE?+?qj0V8B$NM?F}j*5qmNj4!#|6y9yUJ)B8HIFg}!v2PopD{BxV`9>67c`okc z$qy`yAMM@FN-29Dvvc3*NyoY$KAk;6i(zk@OxyDbcKIiJhNf=;_{SwHTgS7+tp}=y zG1*)v)stsjOq15`AO8mV-Tsz9%luz|^K!mFS+{0vZ0S(J6G};yfh1^%v6x6#6|b$% zxU;+Vu>QV`eHOh>{wy_#x0D1e@0J~p@mw~i5x*0G6Ku${{(B@^MifqcZIXxvYgsi3 zE-5eJWbP}o68}S>pKRisjDwAp`&u*PLC;0|SfjL@(=(xI(pr6j>cE_D&=I>;^@(LW z_fx3;Xi${R2PNsb+?r5F2+s<)xZT(vv<*dXjFya;LfgV!;uToI+ZOj=2q5rF38jhq z1e(Q};*8${kA_FGLMn2E=Gk2whJictcLE&$7-QW+LVS!N^`+3DNOtC-k+Yk0Hp_R@qfFd6xk^&3wa(fqV)&K zx>kOWv3W#=t$>YW3L3sJ>;K^HFfFVI=O_* z%}96Ti%p6a)NQP(oCndgcYwX+buMW8iDs=+v46FJahPCuZ}ML!*EEb^lc_u^ zE%Q5p$8;7*-kQUTmp$N6l^Kg*f?5mhB}`S`tC}yoTP|EcI~-0=hfsiVdNrIBNjq}U zhoQn)=qGUtwx+a{HnBM>CqL-&HsTXuB%(SB8eC+4G;Y3cB=0Z>^_PNBS20sO8yK5poiDd93{^PPzQqwqoWY80AcWW`s*kS^SDN5(Z=208Ui30foLRi zRQ16K*zg!fFq?xNLBl{K!I?HKBeTzo0d9B%XdyVhji!%EXN%5+mdd!D&+r^E$WI`G zn#%c$9$zw4&&Ns5Xp$>_?tnGRFN{x;Ps&1#l$$#sjxL(eO@@Zc5Z`}U#~&d(fs=+l zhng>EF#$?N)nP&KwVG#W%^y!v^A<6#qntYDRGIM^2kdEH!6*?N<5!yX1yc?u)srui z6Eyo0Je94)63Yp(!+9{{*Urm@y-~HRa#Wt?E-bU`3&J11#97qASo6i1%&5V zlfxvJ8#2&HbCKrc_-1|Znd9u6)2(&+)Gk(CCgZ*ky_kyM`&n4=^t#JN!Up;g zaMeM}(TQ`>#s^XxPL!Za1o%>}2<0j0i@|vSv~d8MtCd0;fMkasYi`Mgb;bDc$p4S7 z2UWD8#YyrQj@36c)bA3Q&|>t|;tacZG;lGAMLY=XmaMH{<1|Eu>yP!%LwIFaIMobO zhsbhfRCbP@jKU@XJI-sy$DQxNyoes~_u2TzKG|locy}by_yEStAjp%84rNRY#vkVr zlA*+u8%BxF&<3)PRQ$K%+U4ivVB?`&Md!DG39rDr<(5^7jFhT|xpk)HCYX)AAt;g6BZj8 zzJJY9#c;qQGt5{SWvMe-yl3k89*RPJoPUrkAv|OWEiudlByBFNJpu|XX_|-fy!!J{ z<`XrX9zH;AM%X-t&@m9?BwHc=`sJ<}i)022i*H4F{lW^@(_)XTz_az|Z2 z=4x6h!8tGm^7<|JKuD!I1B5tnq`5;{R4DZ^IT8vU)v}iF<+xGoW;Z-ypj<@ed0Am$ z9{P%+aU1~{FkT*16le2Nn&l!dWN)at8$CcYXV8L|cMNMYsiw1qc#$HSm*g(KXJIM{l@9fzbPu*sb3ILW`cA2~58wJV;E~W4r2*}xE2S{UN zEMr^(lndR3EUc;7t=m-tGC>GUQ_+y@i?%!mw++qiBeQOu!z7cuz>)x5ZW1n@`(`2( z9xh93c=alt>=vyP5XE7V)?bntq$Jri##%XvBz@d|y}Yqgb^?4JaC>>$mf2Xe(BoO& zNHvytl`A?gsWc*P+&$B>QcmNm_fBSz((eGQvE*d=1Zrq{0~MaLP?>tco%*i;T#2Qm z4FPpgTizR`ufTAKnA{JZ)TG*oA2rl~*Qp<^`PfLKf;FWSutoz^rFe00g$gL@$V2ga zBz^4qApEURmh;{>ZJgPh!A=$s2CG%EFI8K33REw| zU#v)NAe@pz)SgHSk{Q(WvomVeHXMD@jsW|?auI<1RAhC_CzX zIZ6!Eh#ns8_;b{+U^Ee{8)LsRN(4XSJ|<2C6c2rmL1=;h!UU*P0=Y~^R0ayM#*H%` z_kCIDyPud4c&{v04x=?6@$HV)EQm|>21)aKh2?1BEDTz5GRQm-;KOxyIB`DkV_~xR zu}&B@%nR{>O)(|4V=F9QWk&!aER*j+k?I1S1s#p4#oR_AdO2V9^?$Gln&9t`Z#(v* zXIU=PX&v#;4q#E@EJS(llU2)4u`jAbn|nRH$O2bqx~n}W2#Bx`NXet;dKNw}6!YgBGMLEu1UB>20KTZtfR!H-S~4qUlESf3RWmzvoQ-X>zx%mU_U9VHliBE)N`41i zBN{(Tz$>f$cW0E68A|%p5*%A00y9QFB5?ATkbze5^B~}ZrSA;TBe^J<$u&* zAZ#0mfDL6g1iC8_R#u#7A$Ad&Ic+SB-rri33$?})56UP{aGQ%}FhIU|yP0eQ!3=0? zwox538+~r(OUB|DSZ;dNl|Zzr1Kl2Ka^MeT+rIycDd=O=BmpzCvyE^AXqz#*CrR)N zz^0Y^p}(?u%ddh#-3dK)eCS?*UZJ0Ptpevmd}$Z9i4zLBiig zcgBNlKgmn3R7gX}!z?%1KH6;1=SQ$+1KJF#_5icN097|srtb%`Zt!Cj^Z9ueyt9EA zJB_Gs8pGe*68y`i^U|O0{2L5>nWJWB00`rk$cI@F{-R+a_Hbu?z6CkuKVr?R^UK-;h%^ARDQdm4~6r?jL^~wAx>6g%_UFy+aI{_|aCRC!gF^j$k zK(`Y-H_WC1V6!nm2egj@ZLxaqGT|31%xgF+F0A4FT`;@$nMMJB5v})Q{#e=thQ$Ql z8e@y|lQB=xnbY-=5L;N?@zxYgHmopdpSqQjvu$Lh3%(a>(;h%}51=A((!>Fr3kJxk z!JmGZjWYl8**SX*Di>4z?tQQW-Ts<5dM=oCpaNJq3C66t0l>S#H|=eBC9McQ+NqXjcEp)Yxqq{xpPIg?@Wl_&sk%Jb;%Pr? zB@b$}W%|Mqq9yI=1dV zh7U6@lz&_Zes_F?`|Tt4dnVbIh279B#X*$fzU2KVM({_wOUG;f{esD|!`u_EXb%Vb zi@f;j3C5`kl>4(Q?d9xvI}&TrSC(Sv>mN;)vL6IJ5vAwWRVn1IPwj68rl0|hx7N_N zU97*w;%^GPtDkkl`kJ&A(4M&{-mCH6pJ?Alo0edo|4uedm~2yp{~|96`l(+3ASd}z z#Tqv~MFIo!e~Bm@fC6v}Soz18r1B0I`dT` z(x-s>zeJQ>Pz9H=g=wiuwM?oop}khSd0hRN_@y(sU2E zw1P4mH@QgOt;It5-vE=D&Ap@6T+^v!6%yy)TSc$TV$*?ce`f%6L@u};-MAV!; zy|e$9NWjxG{JMq@>< zjwhrVp@(QeOQRy;LetT|x1y%97n-2UhLp;s#sZh=0ck`yPpE5@c{G4D_Mb2VT437` zP#R&orPNwT1X)<*!Hnet5^E?K4S+4Ki~-9XRYR->7H8_~2rLfSU%i)|zps1sWr?k% zrdE_(B`GIx)Bzu z^oaAq+zP*z2f*7LFLI6p&>+7w(xP;2oA=z>dAx02z|l^^cCT}ttO0EgoFJ^coJjdI4*8vgu#S zqF*A(4k=B!!m!ZHoaTd4<94J1M%%H)>izm~3}w#b%YL=1`N6*uzjVS#m;qowlPnaV z_7XZ%q=8si$jAe9T5bg298X40NLaHztlAbqoV~cT_MHEdiwAWVRu16d}=(d)32Db3iM+b zDdH!IA;y1$$#*aoq((qm!k*K1ar$y!E3XoI^J?BJq%cwd5D3~d$2>}PjWx$wo{6y= zp)O6LK5w-GcIg(dzcD6wDz>u5y^j-?0&8p3?UqgT z#&4C(z!4ew>vB9k$X^&olBJA!GG8rcoisrf`O$oBSofJu)S**-I zDV_~c_@S$M-x2P1RR4?@BQUm-da^7M>cp^=*G_ly~W|t?V zI*F-B@v7^4S2s3{vh^4qB>E_+sf~QmuoI0J+Q~X#F2xKQ4b`AuGN&aS{iGr_(`f2g zOEl-DeRGT`XXkRTdcwMmU2NJnj9J&OANApvdCGEhm8?|_@(#L=oQCb4+CH_Ha;(Rj zBG3&TmZVANB8PnZM0Lf^^P{bxCxh&7#JYenXS@V8Dar%Bg-@d$4qhL*|9R-~1)lVN zGHDG286uDL!Nagi4k)2;b;~dzQ2mh@=;IG>_x%Pd(If#Xj!zzc?iCxH!pFvDz3Po- zY`GW*WFoN~nRv5NE8J+~v0?*%(7K=|H-l-p_UlkadI@%O!|7V_YrOo1DJvTj7atQt zbGk?9Ra+8t>n(7hWg`fZRPun41LCcmpZ_>oBrF7TM~;{{xrEal7|OD!;F-n*zqh1m z793?!P8%uxlTWLdwqmBypTzQS+$!EGH;G}NG}GZUCL8K_R2H08TE$<8c%SAese+Ch z=H2Bk)E|&e9Zk|_d1k5vrQkkqj6BWXN*GCgUAn!OAGm@f8X6tS{>q`E=iXeKGy>t} zJ&z>K@+2CwLR=Nlm5LFt3Tk9*ISgMsJZTKxin~CJ*+MhnAL=$;tQI)mgeIK~ABQAx z&of>*z+EvNlel12DqpANPh8r?khn}#a`&ye{Z0|>rSpcO?H(TO<*q&J3pu-9 zm3SIlS%s&HX^nF^WH*=AU3HqGhJt9492c{-V)Evp05rpu>C1L6LtM zGF|bKq@k^PqtzV| z6(d;olg%^e@3;DVY(SsqKbM?SHFRiLfb1GdV4{r|spSZs5fjc=$>|YmW0*fwk3F4r*_$1%L0ag5j z{77FAv4fAEvBQ75pR|a(K1Vc%%$l#>OCXicmC*@peq;rgw+4G6YW5-qsC0$2L}Qt@ z>2c5_Uv>JjAh>$3T5U&YCla{7(tByehF17ke;;#@=MF|}fnOsC(>oyq>X8KFdFQ%^ zp;`n)$Q#IX@bbd(WvqrIyR+Tj6A;PkNiY&y=~Am}*)sWf$mc}OBiezeBB*mf0-d&6 zol%+s(Xbq0>+}Y!j9^4Nb_AG6U3WI26Blg07*v}WxRwwgjLg3qRD#2parju+7JM&1 zW~D_l*b*tX7Yif=OU_tdnhqJ07u2Z*<|qMLUq85p^pP16&s!ozfw7f!pExa0QcW9=pz%`K@(d-O&~=NbjSdZ(T#w+O=ubqMdVDx zPfchCgZjr4VPU?Y!vOy3eo<`&eb9aB9yCT9G@b>6PGk{IG)#Wc0ha7cU@1*jkxY@z z8OqEUU;)XyNXhW-rc6{OKCmVycS+a0pg-OTh~~)wlA%i;I>;w(1$H2@_7tt3DH0>N z>5PC*H}+yZNI9<%&0);YQjE$1J4OxObW1$8pz?33=%q?qz`DboAL=UEV& zB-V%?!jiixniM5dr8H^4VPPK2et$QwaEDeu+A5@MdO5IA5i|E8QuQ}jl~z!;O)OcR?b#wB-PWfNLF_ixZAYHyp_}9IMGgu;%yLedsWC+)_Ru4ubJ*g z@Oc@>-cm6Gyz*2efZ7YRQzIXcOPlCkqH>T8+su#4S@tHFcNLH3(Jt~s1_P~D+PXsw zRR&IIhc!OjwynIRBmR0j)DdTl~)RaZ~|QHH>JRn zR3&BQeD*vDcp9OeS(Vg`|2Qfpq%qQ<5KB)K;5oN2$4CVmxSH^;^olctQxWvUQC5RU zdNQgN5~V{DCGFme165NIXO^;FEgaP;TVARn@8TwUbL|@?>k=X<+W_4W+3kC#P25OA z5_pDli{2C@a{@>L^q8ScuI5?%$DLKu_bCaoJXc{zQ#iK3_cbDQ;7F zqBcM#`qF%~%4Z57!TrWDC^5rCZ)OTiC2V-oy^o)`qeEn@(Q%f;`K{~nCaAs{qq(iU zokJKkJWuYmrBSP~z`JoR+*TWGyTJgO7T|9o2&!NJEw=Jj2oyF_8M!WfW}?a{G;eP7 z*D6d9M8grpEzN?bO*8TFUfyWtMpY97cQjaa&{Ma%`gXJWq~aoDi7d4eFJP)ofn2y1 z9>zEvPn-S1Epo@3l=yOu4e5YF8IdsA!GA$)G>k5F=2)dclU;cq65X6RIhVRcK613S zEO?#zSdRJ)pQiW?k#r5J3B&LG{G~u}`u$y4EFF0uukMso0p$l@Zi*UCMg^vvh_uGc zZ)ok`HmBO?O8cpyx3=SSc0xMNOP#3&7tw)q{3*|AX5%=W@)UtT%@?h;OZ|T=9#c;_ zVCCzjUJz9JBMoH7fk-Q`B>?7V```Yera)5Ll<(LiVFuBHa5^JcJ~Fxc7g2UA1_jy% z$(1{iIL$gC+|yh`uP*wgPdic77TV-veI}LG7iwdm z!D@U{=Cc-`hv}QZ@9MmWVpr$EmBa!m6D0X7X&$s$GNAW8e-HhL$0P(!!l z*IIeX{Bh4STSc(*kPe;!^*HZV#rPzhnOr3UeTV^^=;~6lp}epI6QErRe~=h`Fc0la znyN-=LarwKdz)6hk(Zn=d(JzVaRf*K!ku+0O%&DO5rxJW&HoT4n$_&!qZoe}K&g_NWdHjBprKy^?L3mo9( zJR-JpwqdX>t!4~o{ikMM(BgR@v*ujPk@$9$7z$NmY!Dy7m@1{ah8j^0ITh$V!9me3 znE=g`mD*@>$e&G+T9_GJ{;_BWyGJ;+78+uc{uAzVe6IUCffh_N-Y@8AfJU6~B}0T}TVr>Jm!$=iW!_W%yL#ToFg5yU>9QG*zqP`uGk$xo zYoF}nr{6MC(cISi^)V%}j#)vM^BCY6j92soglJHo@z^(GYG&MAJC$(Ev9S>4!qROJ zHkIXLni!J_zv<<)ZMp=W;@*{H(YAA+!8ulKX=b;f_PzJ4*Zbje6=7B|NrBPf2v2^5 z0hT0}eQLTU;lHeZS3B7v$5~W|qRxA;E#r9n)}K`cOm?-=lCg|s`B+U(tY!}rkU8NQ zDWrFa#Zb3@j&<49=DO=2Ani-HFB15x?zq-)2nmAJtroE(c7A>k=(vz6K|Lbu_-zzi z{Lx{SG>?A@?(7anPbH1VlN+U8P208AHN%T zv^w@7)EwGwooHbvGPfk)^n(Pc2C$;A6qbGYv*GfoMm~c+PN-wE=+dZV1gqZSE9HX z$v-X--gn!urGsN+R5JXLb`s%!mqi}f?&?bhLE@gEm#RI9&)L0IFXpg?^!=kk;NTciNg z?cK)~at_L-;mZYYd68Cv9^1|A2g(X-R&i~DCHVh#NvOdqNx);n+;hy)b3uRP3qL9% zY&&ErDr%DpLn#eHK}Jx3yc^)lj~7GF=pMp7eq6_ zmr#JxJbd`uZ9emhE@qLw=OjYvOR@M{$=plRiv-TZMLbJ1O=2iwY#=Ln%i0puwEq749(oc`XE zuVeoC8$*AK^NfKx@FF`8Xma~6?LIJt`87MPTQ+&T%n8`9D*{BoqyNLcxfp`&Uh87pOlJdxz3(12 zaH{8%fpB)0FBgO1h(Kmm^Bc$k3^n?7(xtD)Z=FVCZi`ElPCB9tuFwS$VeYFR9cOdG=CU6BEaH!G>p2wyn5eSc4AAt>vGOy{lqwwlajE1O>b&kVDL z*2Mi2_ouVP6|$I*bQ<>ih`Q*6Ig?Sas5G`e9LFZ(n@( zBLaU^_Xl*%hle8r6Vv*O4F5i*vwgEYAtR@^foYI}%6!)PW?!(DvWYwYz6?Y6qfycN zY+(0jbNd4-&Pdy8;(%HSivUGHy1y*r*dOtJUNzWDj&{v|KzxK+p84J~7fwy;wiixA z%26uR)C$)ugM+?U&w4x5R`aby&hQV<16YRnyXkK3H4Gd%<3L!QDv%4jrDN1WdB5l2 zCPjB>p~1L!z~w?4RSB$DOztjK2eUE!z2?M?znm`0HVEPzeNetfE~jXEzo<=qqETpN zJKaWNZ5&~?lW4=1Y5v@a+pfxv6x-)G|LqNKqGT4{UU{K-{4DCB6_)`%oqygaDpmB+ zXq%}6vPwIy!Y41qqoLVTx4Z~XN9r}vpZ7%ra3nI>+G_CmCogxV&j*+G>^OJI5*P2v z!z#QyJe#g{m*zv6_{xFj=R1_vu`GwH+v82Ah8=uE=N}zUt8#_LoklpCQUq;YYua5q zUXGWHyDY~Qz9xZ5y!@uYERNQuq0kivqscv<>MSdH9!|>tH%JZu+yEZ_SC1J+0zuA8 z088>`Jm&xPn8h9#!pbMo7(dtI`e9d${|~kIETPT%=keblx$Ruu{{+d+L__7%sf}yf znU*SK{~IL7+b7KWcJX+T-BG90XfXT+o5)kgo~FH2Yb>mpC{gA1byUD}wF^yJpLo z(yQs-$N9<#OU*1cl6!}>ZqLx8KtC_N%^$J^V)cF9?n%k!bB+C^_X~&fWfc7ecg^BU zke2jWgVh`7^SufdD59#^_4Cb{@%5!VL_hxsU716EwcK$H z$2ueNH8(~q4n^ZaAP9L|Il&D@*U8aQo$6sQ3LMfu+t=#a$fmG<%LG%$=iy}>#Iml^ znrR@~_-@*FAXS+q3Lrf0L{2OC|Sv=hWH&vC-mqD_-0C$h}!NZSatU-u+z(4%+>dk^ElQ$bns{>e!_Y{ZSF;83}gSfOC8oWFBbT@ zeSy4uq$zR3;&gzu=cOBccmgA}Qn>R6Ld9L<{>`toPhMF=LcvZcYzXo} zOx19XUN?h(JB*meq30IlXsd2=zVNf+aP8Yy%h2H!Gsv z(6V%M$)R$s8&t2uo@E-zaW-0}zX*KxV^P)k_AG}OU*|Hibu331#pN~xF1=~&Q z=k-c;NezyTN{e=uCx=j9o2CxbP=gGYP!B#@lVlNxcVp=Xw>U# zm`IfQxcQ)xcri4j^|_EoxR|H)!FuV`^1w=^Vg&1v zL3k85T3wm~XjU4tHnJ7TY9b{^gt|E8mtjgRY;)-J{oohM3OUlrl1kFf%7S8!zzIkc z4-`9f=y#8s%dZ~QO0S%no5w>}-Y;KkFIvQ)#x|5Uk^7k8sc0UgP9*UICLCc9@hsdK zbvr8xV*sJ79$tA^hM-x=378mjNDZ(Bq2w69O<}Y-Ov}{pB zIU15s2SFJ&g!#Q0%@&_>V;-;9`h3HGEJxUa7)QW`8c<`V#Oih$gIiVzi3fEiT zGi7Vh5mwxG_#Cxlm5LBcT2yz360>WJqN%}U(@YCoH5cd;kJ0}@M3bSkz%?Qo*Ez_N zF>|OcRd6`KF@KW$S3URDfoH*?H4ftQ0vA0FM&K$9k8@w2`GR;JE8Wm7cO5GLvzWb;61eCRr%~!*8T^nL)69QULr>H>OY1=4gR0L$LPIVb3JMXj(H$27GV=eDgt+ znG7QeA^3|e>tSSfBPCl5OWd!s`r=nP6YV-0n%4<6gjNW0 zn&t36OI^L>)-3LP>sHvGn8b7X7(7Mr&Q;@l zDw%K-UyFFjjwLZ6%bECbk;ttE<|iwRz;N!`)AyB!jn~!uf1ZtB3sSLEIH`Kbd23YA zSr|1pC%~2MlH^qew6CL4E%<;|yh0mk9K+4FU5^Dd3(2SOx8;qK$KsjTE4RcxmjJv0 zg~&F#A3hG0t$h3tM5$|+H^p5C9VB%+PdekNe98~ei*HqAwhq$%8?MMj7x?7bX&fpj zxAfg3_QuDI-%);@%7OQTGBTpgMB2V2os9`5xkPu~9VtR-F#4xHZXHuZ+E6$6twqG{ z9NV_{1laxRbI7`?3WF#$EAcv0g6%9x$WB(@>i`Ep_`f1lTESv}OOIFe zQc8SZFk5>rW4*t!uoOA64e8lP!B?*EF8m#wnRyUT_Trpgevbde@Cc8~e=^^2gXp$> zTgMT?HY~bQIr>s4{pVzrVS(~By{qV3{jbTYPRhZlZ@rnLkC4))dqqs&tiSb&0X%#c zimH7-CeUmk(4N=Dr}q(5(W52u^aYJzrHzXcHnad!Ku^OHwEXb$F0Gzy>Jw+& z6MOqZ1+=ZXQ2aD#>-}HN(FpF%uSyNK2-E^f01RPRd6)qjJqTU^1Yit9B@B~McYp7~ zf++@QHiuHQgc1^l|Kvm_UZ;rT2jTlMf2E38(6yl>46aIxSXc}19)ypPcNo^SVqGW3 zunW1bvgYh&G4s*q&cXl75O}{DS#KWB54I8Pwvz(m?(=#Z_yiDTM-@;7-NiBk;;-H)!@r?1D=8jqnQCd0p z4zV`Dw(*d4o~dl-VgWpHf#{R0C_C>Uv2pBbaDx6i%NhfY(pkh}7S6tnvs#?a&`>0) zc|`49LYq#4hI@jnfD>6aW!;%x(oazKPfnSa1Y?1u?k7y+@lPiBEUf}~DedtHsd37d z5fSvs9Ee|zkyzKMjSu`H2G%Kmza+DaCyxlEG?a#ro2S@#hQSu6@aoCI?xg^FKxNG- zKqe5JJp|4j0$U7`E=eKVPXWTDRz8J4bYtc6+F094t?&#e*9?iDw169~G!*rCt_3TO61JLS`-_-BzeiBhQ{uogZHkl# z2_jKt2OHCQEczk?OOI75W2ReLcH?MpHnz_LIJ6Ax0F!ByA)6Xt7xURZHui)*vQ_)D zXC^;FLgzCfdPkNw9g(X(A60HtApw?aghya5Q-LhD1w-yClHE8M6jh_A)tZ}^lOssr zE1;fv_>i;73re+*PZuP2j;F|^<&-pv)oGI7rjFGE>=t-b*1ARNl#ta&B*-c!1P!EEvKKiwg@;} zGFc}YEx;3cLaz=-fy7J-K#3uo3{lwB4MVoNRdDn}F2uC{T9M=IHkK1Oti?)98VOfi z4=Rg|vo(GxSJNqm1p)Yb%ebVOS>WQhC(C`~O+4N>*N$;kWU-;zsrR}rK|;yLNS<`% z*u7lUJsQ;-xu)(D5wiAia8Mo`z zwp7s?Zv&-a4}ys0)(LW}W(GA7Tlfo^Aw8KBkD8PC)WNIsr3lrr`k5iqgRphX_x6Qi zCTslN$xZqm2k2`c#Im;jS+FGyI9VbR1Mr$$*^MH6WfIRI4{72MZyxauVmB{o>{_r6a+8~P$rS^(muB-Dy`}{(w>@JP z(PV(AZxf~?|3Mz#dAtXqtkb)B^PU0TmEzY7Jle#7hKZ#>aRSl2v4$>r)Qky?1qj;0 zO=~I$opzA<)c}+OJ@~pQNg4e!Vn>ddY{6?>LVSE)JuD1qi#sdTE8Wpr@fs(*m!L$Y ztHI$I^J+jC%j*Bob(T$WK+%?Mpn=944esu)A-GF$2<}dTyEN_ucXxMp4esvl5+HaW zJx^+?-c-%|3HRGQ`<%U&a6~>@{IWMJdjn)%@)tt^xFm3LxCKKM<8z7lx><7_*0)LB zjNi!hr|MaH%L;pG+*c|JOT{ve!Jq~578`gOT!pq4uLc`BQQG+|VbfO7yzmtm36s0| zK|(1qz8n`@T!XqqX)?gjiA00D{byj~Zf4s$9BO2Ns#FMQOoh+&u*K-E<9LG4l%lQ9 z#Y^7F-3PV-bpTz~QF#28AZt!}Ef`fO#W!XhC!;H~Y`9}m9PeKR=C!f#SUhMP0*r99 zcuGPIKS)%9qbeRi2SHuGmNhiRfwWZ-Kkbg=5X=h`!lQQoX`_aL*XQmF*o+w!B52O< zP8bN1%q)`IsMqnJzhdurz=luQ!^l7C4N3;(F4xyye!9wc8^XamA_{EFk~W~@qd=4w zIL7GC!v3?V8rjgI&P~8Cnf`Y(P}PM5*@~Rbt`BTicL)JN$kA4GFvg7~BlfVyWWYgP zC3IJXRT5p}N$tgybr=Ka|t8;`ZlRP6=TOg7w7vo@N<7MW^hV z$DPeEiWR83-!^b1?G>$-9T%~&oP4>RnNusT8YsMR>hvY5gM9fOl*l-m=o|d16LEi} za78qHUMHYZrNYOp4)2*V9Q7Gyr}^}=JcYU;?hX_n_G}Ypj@3>H-XCGpXUM5j6Ay3k zck~*sZpRAXoPG-p9h1ifpJDDcvc96SzB7YQ?^&LSvFM&=hY`lNHxk2C8D?Os;0jpK zbEu*T=MxE;cq`2(J?A@l=2tE#*gTkxE3iR9%PlNvNw^9I~7?bB58z&SLw4^6bcLMie-8zGw+t>O^IT zNW(5$0D3#XDE0&3TQe5BV<*Pumm>2z*MtScQYcsiw&&OAufaBc$kL90WP^3|TO@XS z7lmhnM=+~y583<%c+nJqp|YN006_0IzPINcW>{&$?NyNa62=6!G6t0TZ6eMA($Q8z zbYe44W^EY=$aPn#8Kq!FH`-&j5LJNa?A6W$6)ZtZ2PY`G1bp>1IkE$SrE2T{>D-0x z0hEk45L5sbfE6BFk5mDn?Czqcb*whyjsFBJI3nC$19(8ICc-)mB@&hmi%56qDSq(w zEPyid9R!sfLQvkxFZzSmT8#^=-+qAB9-tvPpiX5S!x(^I9F0V3?aLP)?!6ty!2M%8 z78Yi&%nu%R554pnaOAK`3W%uZ1;9bTSWFyga14opb(B1N< zNK(dw*mp^f5pWrJN4ow9!x)Godd!9I$(g&ui%#5+G-D`y_I!)n>UV}vw+FA-w4u4U z@fJIR6(z6-@9$u6e#-3E8*7V7nrnK@^Q$I^F`Y8z3w!3%ej~xb!QYd^$w`k-B4?xJ zBs1ZDHyGUCGtdk*V44Zb7TA^qFBlGUFq5cVQhr&m2+G{B{t?Xn-C5|#h>5RG<4*qAGqb1KIP|S2382XWAQ=80`&VHz+=8$BLoWm zg5V>xbRs!^hjxk}@J(7;1U+rJQPiLfJ(g?V#|fnv>q>b%3$xneuBZ_5QW3yxu62tsN4c!R9e8<=r*pMj&J-P$2)M^X1rb>Gi*k`TUO^RMv=n-p z0;0#|G%FE5THItU1O62GbpPTLFH(e;T(KT@M$syiN2q$V*rT-$wR5|)Hlu86kUVxI z;|yIqQxj)?F{YB#ho}`V~G^?w9L-$+j;*UHh`8-wXaeDiVCDQbb_Q?dF%Z@R| zrLSy{4^7{@es|ibh&(z<9Xh>KBID%lf8lo!uSH(E>3?`zWEICpKXULWKw6UTBlwuK zEqbKCWbJw|?YdSG~eek$U2{TSBhn<;*q-1R$ z$E9`NP^f$Z7i*3J@j4I~Z(TEoGCRRz^svaowgfnM025B}{dd@q%6&cP3vE0(!SE-z zzyeWodH|s^a|mP1uCsq|3f3S(i1sCVGIZ<{HhGk-c{-K+plk}(OSMV_(%s0jt#$H0>v=r81cFFsMz3LyqqgM$ud2dnqCqHF0fdXngJkBD@iFCxR}Gs0QSF|Ow=`} zqEg3!Ie9%yE>|pJ)b^4$%3nx9ZYYqh#w*rpTu7;RaUPFM=UmfLe6nXN`Wfo9FL#>E{c z;j%AbtMU6gN{K9GWT&8eEhJ)43(0h32yexW&4u9vB)@n$u2Fs^f5w+mverKH*boOV z$*Gf2?SK_h7u009)<~dU45ej`Q!rzcVVl&eU>qj9!Vfn~f{8H(1h>b>0)_CMzfL`A zIjiH8{FQwA773eLXiu`eMc)#;1Pcd(LO!kY z<&ivj=C~2uXY%?svPE6x@PqWe8(hiakG(TI=?b{gOEX%}9)%X=qH`+-ozg+X(0Uo6 zcn(7c<_Q;9pYmXtm=$kow}DvON{fBzsn8b*B3Jkj?yJBM)%ReMpYX%{AjSBTFMI=P zlWcI)8+(_80$qF6#^0Af&2` z!B(9`_Tf1ORa3+8_m{mrX{R0&2YZJLBC4bk<0)Ey93Nu6$*biiLO=VxxV+{JmK;V| z8B(FV#3fQ%C|Xl?9>Mn+=dSsy&{x&)r`w-TI%;u=Y28QeUZ6K6-Wmi@9gux+obHL8Roe!Ie2v)fzqM_YK z{Nkk*{`Y==V>puCh+Hcs2sH}~kdv52Vg-d@!!dk_in0OJ>T`>w!~myKK@gy9U<5Ho z^o>;>fzoZS6oJjA(Z3+xgS>{9Fp+-*;n2V1?J5A>vYH+n!jA%i#qjA-`7`(vMZCX4 zxv8`NpgN8R80SJ8{m0y;FGNXcxLtk5msXP{69K=Y!<9GgJWzjVlC}k$im*SdYB$p8 ziBm}hJqCz32>lLs1kvODu?j%@qz<zTR;fu@kn_XK2lu~ z@q;wQ0(4q~AfW;DXLx_Z1OGh*IlwUR99!qo_>s#7OBO(A+D$*d;P=4?GVya`jKBg~s73@Kw%?XHP74n9KUoeCl)?b# zS-oKpf1FDqbtE8(3VTYH0JBWpF~sZTb7qron&Mz{2RSi;uz^j5;x|8 zoY+WEs1Ax0L6Aj~B<+in2|QO92=lTj`RphV7Gz>p5hy~SVRf$WzZN4foKl%cYlIZG zip$>uNi=`uuspX{Oiz_KkCA^BQ#=oZyH&nNv;Ce${vB@Yjvx_Yn}udPp?# zbTWO#JRg&_s9=vkveZqFvKL_a&Y+?CVyyBSq;^hX2C)2n{6CKToM!2~=Cd zXcbNO!g9jZpeOCX0tZoKE&IJZXrx;50)Lrt-(~h-WgX+$tZBH~u0_&8huHrH3T(Nj zWimvDn4oNA;Q10zu*zj0;kj~UNKY_emnQ*e3q(%@R9k>ov2uu-r%Y zkcx7lv0mT!$LCkeUm6j-j9YWEANBKe$C3Dqj_WDamve9 z7uztaNcXOQEx;Pq@mhGQ$R(`I@1$M|24iZ7$hBhLFS!24P!xXlvka(vZ{wf5LH?zH zL6PDUVPNvt@Emj)NsIPO6j=d*1qy}@@U$AoUUBvJP)tDrf!R(?ZpnyiO;aq3kX~Ez zlDDer@@UHNgeg3(fCeO8aMhWOO?d?6<{KYuaAm?!{*}BP>O}sM;CB#VIJ0S?QbJYy z0AA3t(nf-3nWs(hg~&*g7+#S@B5WN>g7%{d#ykXd9E>&&K`kQZg-$pcXIPBrrZqPj z7(KdnU7%dWgZd#LIw9?ddZaJMjMNQ&Hj%a1nSmreVx>^Vsxv-a)H~*XqUd(1X8odS zT8m$nfEgIlRXkZ$?TmyCgAjs!#v?G&{;>|RrM2J*BWVcv&_*VMX?0@(>siQ;!ltCY zP_zDi6WVCqF;VR*a$Z_3o>juOWkHp7UGmE)3LIpf&lO-xyOwhJZ<(T^tV{s*thU{> zFu5$LnIMq|Q)L`%!RV}nB1@Ae6^tWDV;BNzpaNKO(h3bB4I{OkW~k1mp@~4YzPl4_ z$z+C5ga%8FHpZj9FbCGlGKT+~kvq;s3>PjY(TbALkpKZD>|tEz>z1{2z};0d;-@2*k|}tlEn9j~zMT7GtyD2$GLBY0=r0@wF|hi)lT@ zp`@DXiFLB065RxY9k02JyCd_mW@N3~6vgK6B`JbcN>3%`W4+N}7OD*j>O91fygKvX zQZOo`_SmjTc`=zIROE9SW|JBQyOzV=u}{EVwWf=f8hzEe#vtk+z3)QF6)3~7V(5?k z=-2t0oW8@TzP0Gt?Z00oX&#tA1@}q~3E@*PVYOx3XM;;eKeNAvDl4X-V%Gk_V}R=M z#Ekj6b?N<#2*jvvp9)Qm6aO`Psr2Mh84jH?mPzY1TpS}uc8#7QyL_wqJlRAruNhl7 z{?R|4ePwDb&}1a6;zO-ciCWQ(i3VGhxjj%$TI0GJAv&AzjX$Z5|FaDP zT=@N#tkkBWw*-n(P57#3#%m8pquR*R@goh4MakbsYrAIlr$r2BPx~L3j;z(l6hu)y zmlLhmEX8<%B1_#lN?E*0Gmlbkh<&bzL-;vq+l(>HC?n2LvECx3b&wf96fJx3r^oQ8 z5gTMQJ!fwUh{+jv+)tI=wN)mM6ymYwUA5h(t;+=G<2545v6Kq>-i3$D!!9L;;N7f5 zWmM@L^j5O{6mcpDVu?C3=;PIlzCl6adrqNT4GRoGbLcqp3EfKP4<=Xe3_N8emvH*Xv z7&k%fpuoj30j2T7uQX`RaH>Frd4RUDC%ep&bTRn!itL*ihJ7RZ@__xoJ2Lzp06`3Z z2KGKBa1KFPR_0&WM{yoq2j{X!i>NP|jnRHCm=`w5P$yXVYmO4_07i5Kz={EXk_d25 zg2!-H-LTdiKS{Pnvps}4|29S8R0N|rg7qCS{++K&nrxQ9ZlFsl-hgFMO}qz}N*rFr z@I1cu`hiQ0!9&~YKQsMGpogTG$We(h%oyxogLrV?T)@Q{1w*pXv8of$4drGJtb50q zOR}6^w$&6Y5dXpkDI18?heo45{Z0z`p;bMlwr%(f`Eg+F?1k?zo~$iX?r zo-9#A>sFh648KfJGry3S*P0_M7z1GU0Lps+sd|9EV?a_z^ji%%+`cTvwCQh0WO!qM zkRbrRAHEIpbpb7lOJmzXSOeFf2m@hX73oWmeF`_pR1DGp#f$|Hj_R{(O|xR#h{K{X z|DnzzGVhI|0)N3&NGZIs`JslYENhj==#l12Qv9fzMvF5ef?{(GX`>7w#lIayr)*Pg zcZi{Tpy64_cZ@N5VT#BdiH)r@-}~ubvU9L3=db$bLG!ce-N~8A_TjZE-E^kD-euKGNFn*UnH9}xW%ms z8UC-@I%7-aLX}SVzs>(PsWe5A=q-}!G#SnR`!kwyt<~swSo`pXV!vJI0X@cLXsg?e zF@{6Ad26fR50yq_LicTNI1GxWYV%WEuRoeh;ez%se`~U8%%%z;^>d5v_bKS>t;LfWP91`<-@+a-EWlXz9D3Pd&u@U5f+=qbbrbcO%$%YUG#9i++;)? z_txohy~oB0CEMj;GfCWElzEupX?uG{UQY}E+PCNU`s4C4&pOu*S_5TB+x7puKS~&Q zOW;W!G`ih1XKNs~mTHRHi-_eE1&>cI-e+>85bFO*HRotAmrkfhzF=h>6%a`u9}U~C z9UlW6hyZKyYo!!P@;2%i{l~ZVyhy)>e0V}s23c}KOo7n#C_%wCL4^TPRcPKdbZE4NC91!u9SkxkgQ1SZzl16pT0@6|jBGUVhkfRFS8X4vYoluDM% z7iyzOJAwz3pQw+o2Qf^nI^_k(f=iQPX_mSwECG!4=f(cDX)E81Yr+b}Vvy%%gW*c# zvm>G9&`^SMyk&y&02F7&xf1xtAttepGu>Iy$8=y~87!u}o$r-cxpiHobXTS5BJ$pB zJrDXYPW1>2lVJ%gfo6PZ&kLi9a8Eu{f88)Lq(*c!=<-@D(S6y#ywTfDrSfBG@kifk zt?qRTl0%Dbuim!8em5Atbl({cz4)>du4J=r=7~HpapGrsU)0qHfhd50!Tm_Le4ePA93YzVQ#~*?HM8Y<0toBAYpREGL%`cRgGU zH0o}b5YGR67Fa;yw&PBPR=wcNH^V>oa1-k}pslyP%=pD?eoUO{cBbkyyP?b5o$<#!m3Hi|I6dFB=V+u$*2XzafwH$( zkqmSENu~S0mbGqJht~bsGPDnSOXVB)Yh(fCw#C2VVPrc~L(D$SO9%hjD%VOUyWSGp zMCSj_=lZfvJW#FgKZN%kY#~uD9}x8*dv?r?g|OjNtisB;(4X^A2O>Eo$fY0TEc1it zD1#EJ^d)-Jlj9+f3$#Mz7B=rB}76Bhu%+_Jt4JTa85WfP~Rk8 zGF|lVOqe+4nXr;Xf(e|T8I?iU$A$YxVSY$~TVlhw|1q?~n}p1N0vV*{?1GR&cMgDr ze+#8UnpFWms3WpbIb~pCOi__-7NpZn!C#R3e?o|5iT7Vd#gq!E{JF;xANXYhxf7*i z<0o;jIEj;GZDN@c64CiMTap2y5o539SNeG3YMG76rz;a9)|F)%s$E3A7S`_7w!J_m+}g0 zW0<`SN@10MdyX5EIKw=WSVP%*Tq59g|Kh^cLs_~|D>du2(UkizOet?n4W;wh`;!a! zuo5QV?>7-v{tkmyqt4Q~kPQtB?mIf+e=X_(kIiHvJh{RpPpv^~i@qXCj9;N6 zDjn~NB|=1cf^1Jb)|^Lb*_ASUP>)(7G~iE$luDtp3!Z4?el0ZSvS*#(8Mn$MZwmLuYjY7Y~*#IsXgvkcc zB>@L*eeglFIxrK#TUzDqj}?cYgQ!Ae~Vo zA79lNQN=_nzD$|7mhXwse>PL+DSx8nk%dFkY7;+`a;;U7h(hJEAQrD{Z3KW1kAUX9 z&-lr`^MkVdClcoxm7c(`nHQsem%_7kepr1~8 zUw6Di3%c`KlS@RDfPy{V>AyDVs#h=CdT0D!Y+T4g)*{iVe}rsAP|c}HXr{^Vh1X=a zP-;m^n37<3 zQDE-T1tF~x8rWRYjD^Dw1R#J(x8o#5y@)e#BZPr8yDR^`^{;SMxi3$ghB1xK0-nA@ zMp~~1@iUk34KoWrIM8k0Q>U-jLbz%Dc4X^=2jQUO51rF8>&#hMYXz3fzr^10H4b&g zs+DCs{s(i{Hr7y*8qN>7p}``-;J`ISV{+lOp|Z>8Fh=v8Z6;SBh>w9XmyQZEi)Ms( zC+0fo$HJhlQD$d>KS^MIR>Og1n@apzJS6sYHk*qp#RHehUGkvq6dnjx?B#&2fcC|H zHOcv0-G-aQeTBxh9gX@vf+;c=y)-lEU*+Z9U44l*06psk7Wx??LPO^Ep)O{*ukC!b z+d2)eYpj+%aVxmeiTPT)I~F$ZOlu3uO=PD=OLh@z29CHqHVO78^Qn~3*Nf_RMQZ}Z zT08IK%P>_M0`W-#uHifv^)~2b_%Z47>diH`(%(Dd-&jbzcehESLL&ylq0dG@YcJ&! z2;4OEC@XIN9hYdGG5|BpLvmlCH-xb?%_OIln*iuFu%wLvI-`y8)ryvGn2y#un1?@j z`qWjtK^{j-9$80yHdQJ27vnM_Ch{+Su~3zD4=LD!z=DF$ID!7BMgEL%K{E5$y>s#= zFY>UC7=5ClrXU5*L3*&cYOfUw9iI62bUMc*MhLEg#UZKdsU+=uIM0wIUa@ii14_ja zMu?|;_!_0u8YeI?3=T>`r6cL#5dops_UIp${7TL9^C0}`RMRK@OX!P2I(L+|lN>IG z9Bg0|Y<^UwbyONPD}t(ABR{x_pAw_Nuh|n^nGRMpj%t00PEL;IZ=iOgru$_rW9As+ zW{#>_k!AAG=ddZN4Y-<0u9XQx?`Fu1_cCwvZZ5 zEC#V4j%7czB`~^K)1~+bW4GA&HR5M)+t2T3pSSOEsxifT!+)k~X(OI1t~dXLXNV}| z6op+C!a@pUF$-bEX8HjGj6R?wTS&k;SM;W2CznfLOkn&6^k2p?pzlb09ga}Klq!Oi zu$w34vI(F+4_t*>sq;A#FmWe+R}GurcdrRjCvkD-D=_A0S2hXtFlpDdX|dXd6Z-fR zKp3nTcE&Nh?|En>@`tNn#A5J1=s6e*z0r%9KD3n1g>Kw^ z+f?H)%oOd zF3O$REr&yLAM$HBU1Oc6OA1U5m-3-W1*G>PXo{zF=JhU30mzz}t@CA3@vh0S_}$AP z2BzLDg!V$MZ1jJnvQgNgwcD(ee5o5J(BMt0^3s8GlDo%BFQ#L`YYzeLXZ2B2X5Quj zDv{D&GH5b5#kJhd@6+AaiPR$V%R+sTkqEBOb34x|nWaP!E9oEmKNS)97Fc3XKbyi< z`mi^I5%c@Yh4C3slaz~DG)C%UWDtUqBB$L0Ss5~M*CXmu4aWeeW90f#6#)Wa9-~Nt ziDEeh-yn$Lifwd&g>}y&Do(|v43AL*AcY?fY~j}v({GNfYlTw8Flqa#d3WDz8D5?9 zrJ23Poju)3)3^)jZSp$H1aHp;kp}X-7to(uOWJq~c^cBuFYHJ_85~p@Tf?s0I;#Be zT#nEU0$jl|d_uh{Gca+b=*L3Qd8nbXrJ*bUv^0d}8Jt@ksls?685f85D|uJ-|?`E?_Ce2p&f&nV!!dLDJq%_QiWFwl%Bn zu#wOptX><+L7m5Vim!P4g|0VmIyP5vGsv*_3#+g)m!v)SX0bf_F1Jvt42CB!#-s9B z-9isvv~f6(uE;b)9zClc@5_W~1R#?g2I%;bk1&~z996o*0q!`(pu#UtT2@S}sJ+Z& z!CiDtkk?FR%$#g>Sw=twu_HqPK;#qi96{zoZ!AZFO7*i$n>P_=JM|Lzh9I$`a_9tq zjoW8FMnDrI;0?J^PZh&*06l|Hj+PqG#!kDgQ8F&C2p?FCj`*c!A$uY-|KpZu3l0)Z zfo12+rFW>6fXMr-jgiTSmPw9gFXl|4r-(Czg{W4G*KT8$UORju!?EdI!vN5$XQ>Va zySeBYy+)wj*@{E~En!S1@v+-(Dd|9JEoqLpE%|K&_5Purn0((jkZmiqi!e36@!(~m zs$yK_SM;+xpM?UikK-M>u$V3}4xTH01;EdyjKyjVu#R%|3I@JjWUcG&K02R0aRV{O zn=l&7icvs^D%e;~D8Q~}jV$GLbCiVWY$=T%Now%wS-re>$Moo@4!#^!s4cNU7x;0R z?<5jjDZyM_k6zu6UghYFOzq2f*XsDGjkJ}nlQIdVDpR1M3+9-VnvyMtI{l=|XNj-- zxQT7ixu30}ePyA&-?C5Rk;_Z^3)*La;YP(ewHzdKG^S~iF6kWtSmGv!2)2N zPRxg_E(hJpx9BXx0z5=BtoAIG;?yMetxCdzydfgMv}NyzW~57C^K_OSUwSnTU(e^# z6nz4UUOI-^E@g@o(QhaztqPyd^!h<2TI(*odNkrDpP8j!!RQNvIYb^mJBH{ElG4Zf zWFv-L{)oX`kyxrq46Az{+o*hdW2P3AFNjEI8Th%26Bc^)RrcH`0;;dHY(BQa4K~md zLlGaqwNR3m7)X6jkCGo>$m&=qkc8bEIHl|s8+(ZwnAbsBsb{v-v z(vi274Xc%$6jqbiO(5$f7Go2~6hRUKew$iRn;g@h1Z{m5IVYZZ$J-91Sa^?Aq!u51 z2OG=F{|o8g+`^*d7|yoFtN|-qx?&&OV=kd~Uf6>VyRZOz{d5U4#rzEWN8pW6>}ynP zGiSlMGa8Ew>Xv?F7?QbfoO4k9IRuW(%y#OUIb_u9xdPPrnmuFp;T}MO!RX1as#s!ivqi2S)~9 zC8BPD)$i~maukFO@b7(>Se>N!M#ktezN&0b7ZBNMOpp4;@JaiCGe<&l_Q*Ouz_FcFED@F-IKNa{xJ#L8uxf?<*=g%whD*x3UeQ^tFYTrg2k zmuR-YCLKAeUf@rZ_{!P5p>VxlJJuWIN=U_fy>i4*WJ}}V*K@Z+&?(r3+_;WTS`ZuH zXHwP^7m@KMM#w%}5MxieeKICU9q20BEHB*tG(3_Fz1~RuSRF!e8M@3Es}VhOQpKQ9 zf_Q6lpll5)4eMt2Bdd-=9odY5(xlWKNN{+_aMMBtn9O6^-~BdrJsdTvc4`Vu$LIFO z!ARrelnizaVA7_u%H z$ylycYf^K<%5!51fwa0*t_IuaI!SS0f9LAy>MQ8h)~7lVGOO#MQs*#>G2gwfV#4h?nd0E8;>r0BGKSRfN z!`>F`&de)F{5Pbf9nZ}^fpPE^1bC8yiZx#cpJy1Ua>nsRMPz6i&#?1{-)@P zzyW(*4ld?xzp*Q3KVDm=;ojRJ^6xdJ9I)2{ItK6IeEqasyde`Un&6HtTxgpJ2CL8I z>NDx~!=X1*9{6DnWjhxY1P+7Dn10K-)zCjaQRYAEVl2{g?T-g8sMxiI9je!Ex;w0j zpsyV`{I;vh5N=|vz-{^Zk>_3R;8LDV6=a?x5%ki=JuH3|F}tK zN#jWOewDbHiL8D3TL%FEmP4*i;DK~>sab|vvvQV@qR!$pi^Dy=(BBv$D8Pul6kLvM zkJT4Um=>{F!Y#9pCPSQ$ZBB~%;ndH#G0H;YF%J?!H>s3N8oGGW+TLX zmN3xMo_I^17+KgEpu+;tV37m~fJY;hOjp?#fk7kTy?Ry=?eoU@(5&`Ipp=Y+y>j!# zR~7AYhfvZaCey2N8bdbZaL(%u>oV|do-SmP4wPd)GrZj+E|{>vv;dc|`_aMiC6H_^ z9!8DIp$I`cHELZq+o4c{9=xj;&1&)RnmEgx^py+U81l=Hjko$-^-jfaNd4>aXA%K~ zT&531__xNifvtSewjM`OxtUE_6_ zODCn{v)7*GI2?(Dwr!YgvRRF6ce*|8Y(Kl{!t};xt}uDLOjXo`ChmLRrVXAWRZ*e- z{oUz$u}$FqU_e+nmfksvF*=Fa1L+N5dwX10|1vu@O@lDY4EL z^vkOQDDA-&IGcj#Trj0V_H`>y$dQcLn+i2!eK9ywh5rZ}(KukvoDv7dncnbgRMhK7=YS%cEOmoJ=P$;!h zU>;hDx1nFMuu^SR(ZZ`=+s}emY&nz|C5{|vp{--m7V4pAafGlLEc-7~y!X>06p3Zd^B;|@HFc{&&qdcOmhPkmhgML@d0X7!TG zZ`DWRK0DVPKEBwaYS%GHq6y&KDZ}89pE*x)1*|=7lIi;Kt$Z7;eDp4i;~UH;mr1Or z+UEL0KH1ew{9KpAclJ6fg5tMyL??6as!0H=r@Z4&72x~FZxVDvGSay7c2oS*<9Fw$ zN1x||Tf#iD>(PyjI*&-F!4u-1fYnbBGlz*d7cXS|zVkj@ zNKcBH!1vi}L`c*$1ZnT9Z>Jb#_eCItGxfW3lAf_@Ar%u0Y9kfABY>g3T|zQ(9M{y9 z>`C(1fQ;o42_8EH;QR`vaV(&`n%g1PEE!xmm;O0A9DgM}K2hO&KxSyJ!8G113nO4| zp#`0vW+HTkRG1Hty;sT6QnY>^#;?f!gK6?cjU}aqya$8sRVEQd4pE+C_XZ}GHWl;# zN_11Bsw)@OFU${xLw1ovHNo|HiTFZAiB08HiBU*R@QEsUU-MP}=lcaBgJolsFD8ct z9I=|LfuK23?a$s2ISiW9pah=d0-#j0M2KSFSJ$>!DYDT$r%T45Epd{Cv#Igh zhK+>xhSoD+Ij7uS7A9r_pvHh}rpQ7oQP;Rt-ixNS&{3fA8M9~;fxhicfawP;!Wu+C z68RZKp}X8!LsN)XtF#oMk77Dh;afmqpi|;dg@RRkpVCIyA!TqVfb0J9pRT7eq2l)$5;k3d^P0A((_JKqU6%^ge<* zT9!mzm(B+|0bBA%T7Gpn;{0F6kAwPr_9m>ysY%bm3qltcKn=WpfFwOcm=(I71T97G zAa<_AJYL)qnpW{&9Jnuh!BZkN@~lIsneOIK+E_Vwyi}W(`+z{Pf;(i&r-K6b$w+w`h;ILKZtX zN=~-R|6ZK_(xvRcBa=Uj80Yo?EmbmpSKGgm-2WUhBN$d1_xOJCX34g}#G05W)q^b| z^s3b-+9WdP%QT$xI(5}LT9X~NhpMhKltbbElZE9>#IqbCK%Sp#l1zsw`rP#Kktl=gf_9ky@@<-`^B0 zF0$iEVLSyu;QvI6hyd#TH$@_pR6hH^6^R`GCjPsY@JX5)-|1*1;z|7fZdQddxxIcH zD+SZCl)AQamCojUaY6r2MdDXAI6X9GxA+RNQu!QGjVO(LwUTO}xzq&C`EsykY?NDs zWI<4%6?`dfy?ga$8=76m7O<^$69oHyzCGDix7*k2js79iUcWySfrdhE+tx7P-+S?z zr0CUnJdw_(hnM}l+cO;Y@qz6y+jKsg7rNtmwPw_(`}nqe$ z)I|DE$K~Yk7Y`ARQaiWlNl|be?G66#BmWO;v$E?%)ax$W*+2k%*-WpI z^U*EO*WWQ;ocsn0=l@z7t@2!LcG_5NLSPq8qD6Y)%l1s(F;7V?8VNp8SYk73gI(tY zo{7Suc+PapH{tmW1F2FY42+RV+wvVr1)ugTuz!K}V^-yUee-8xT%**NKYk>Eh+g0o z+ppzBkjn|Mwil_&;5|@=a$jg2CMU9JuldO}w(lk~IW%WSi2kE5bvGqVr!>7NsGv?0 z;bE9d!kUpaPsZ$A|CZX{@S5r=mbiB6*_aVYo!fW*x{}1HmQkLpiC0Wf5J4b_7a_92 zTN-Jx(P5k!?lh5KEIlxOmd!aMKog{g8bVVX6F8yv(-#X$Pn|A~$8cs`Eq!6D;NJVX zQ`oO%N#oyPE4NTf?BzvKW6m4RkkN+xkX06{s({}z)@WW<&ORZ)80%h(Y}5K>C05GOR8st4&K2*`^YL8!gzF}kE^f{KdR~Y zDBgHH{zqmOHvq7NSBBHBIDVN}{gcNwSuDqdXi*H5Vr~Zk#&B-O6-ilTH=s`vnIICS z+4B%SG!)YSNMX_bRaY{QHw}q7!&X(aSp&n+(rp}BhB&tdV`A5m-k1*JbgbA6IfA<()S0!~`bDWo%6 zE(wobXleet9|2=b8250&34SyUW{1=-^TRF|0XyRldQ$O={mxU(ah?~V0*N=7-s|p2 zUuEy4&bZ-rJDOj*b{G1J_-_4CEsEA~n0ew*{FnY!3TnuQ0wt2!VIMwe2i0K=8k!fM61gAQMW=1#v~*s>mAdBplnd3L>q`R9wa)oR zyfJj8gC0;@31e@=pMRPux#CIMBHv5q1@ zzRcm*D!smxDtQ-h+}AgKG;RINcL}@f3k)VSGIEVehtgyjSXGgTs@B*PHE4Grf!*>cdB17Y5qVP!p+Gs zb2m=Sv_4=(fG32GEBaZ8VBUn74^(szVXr!1MJLa2;!MY0Bee{kZ+yvB)^Z=(5!pM>BZe zD!Sbz;Xn%!AfZ~p3EB9ePal^TnDZmsy_HTXzg@>a2UM&GA58gP%l^_I7mHJpr79c8 zAN_-qkrtq3Q|>>aZER#ZCm*EIGOro>3ZC!jrA^p(^O z*4JI^78f7f*ILo%5^)AS!1ljt0{g- z>+ER{temrc)OHHn+$2#_VKCV8RPR&%F^ zNvY43YcNoR41ceW{`Up^`^Y8_Ynh8xWu9{Hvy$oFU*G@>fgL8f^^!Rk%nXU=} zO?`*J@3i}p+1;!5q4&tbh-N%jymY0$@@q(y@YGxR`O{cAHZ4!M-1MfPSBAIs{8fTj z*ZOR4OL{It&~TK{w0x7%quAe*75@m~34Js0%v~`A;~Jh}#3O8w)?5tZGb`@bxyY^W z6y6noa?o(Ti?7K!z#$UH-=G5 zXiin>jncg5pC^0-etqGG>uKHAZu9^otxG;aT;$>V-cbz8e$PFdUP*O%+T(`U5=VBk zWuQ(O5_mlyZ+AnJ1;6&Q$kx(~1o$~iJ^Xd5`C@)qXXJnd{V<#N|Jz!&dVP`C@cr26 z&o#Ac$UVK3sxP^Q=U5u$(yDJqGc%&KuZbGU@qllP6=~}d?Y1}vJ{|<=(B?zi$Nh=( zsLcCB#+8}F8wFRZmXGOLhJF$@VCz`gREm=&g087e2s-MA8RS#yaQ1O;)PFh?SgSCm`B znxK%-M|`KJTzv>DTl<#M8u`UK(rC4e3R_!AON z4`{6qV;T$@OygL`35d3ghTU`4Sc^@P)g5eeWDN`z)r|X699g_ZwNGIyeB{GU6Ty+- z-&dqU*dA2#5=Dw1-Q*eDWn%3d5e*Cp6nhE4K4Wzo{dV<#*m~=~sJi}ddl-hU8M>t# zq*GvqZiep8p}VDH=%Ks2yA=^h5v5ZUqy$kwBo*gA&+B^Pdq2fMn9;N<+^)UQ)bGdJY4M{md_#jujn&|p_9=lIG)heXW}k~Ixb9IFY^jx_1dFW~@oRl3tg~FOAOO59iky?;+$M@HZa^D~ z!Vp%1w?|%&pfFm*FIF#ck_6|ONheAnf9s`_Fngzlirh?tPj5#+T8_YOHx8EQjfA-qa9ERDX^_)LrKH3AvQ}e=M{T*nnlnHuA#|x6Q_l&0 zB>#ZtL6~KKa#DUiQ+}vUCp&5JHMG)RGV#XPP3fS*3)}K%vfNk!yF{bD7$@alL8ojO z$zrATEVw#(Ovi6g!9(3`99KW`rXbffo4u&2GL~s7o^T9>`_FD=G5L(-p`vO~iA*C8 zBUz$>jA({UZw!og1Fp!yMw|s`o35yvla*gLh20oyjX`UEvk`&c*4%R{f?+AZc)rKz zx-pbN4h3N?IhL9j80X8=MCYuT1_r#nYW4eSaS)_^Nm)5e%QdCmi3g&(B)jmd97tX3 zt|*|#t|6KNibk&x=Ny_^<|u9CfjvM*R~uHtLggZ65|8^NCSi4{5Ihqs3zHPw@?I5- zLOafzx{Eu($C^gf6hZ(K?MysqvrD8kz?9q6Ed+yh#gwZqjsbYlGy}jj0^&-EX&?cO zFU5GS#Tt22GKWm1qoeDf(YX4%B;Qd+YP%%FQ(8UDj>4{)nz=T%=F;g{gyxta!TZJ; zL1MYXAj^r=e_p(jtN%uq9yGvYEE=TD3fQ=o-x8B6TT#GZW%^^^K-binj`LJqi8z}A zFPRCiCZ3Qm5i%l4Gw#~nhX1r=jpQu)pY2vHEk%An!2u!J9pYUv325>?ABwiPPw=XLxE*s+&HTsp)tH!Iq3#L`K%rYAe-pIb}qP0#+3n5Lz?4tdM z#v7b>mG2Kvw`T#5g!ch2M#e_MuVY{4SG^=UiF1OC=BLJ*iaOMpWMcCG1_WQ8udqg# z4Vk*;(5#VdyElFh_Zn4X%lpDvc)$)gi(CCY#K#>uNi8n@E7i=8ty3R-lUjkjAxz7~ zLHq&%6nwoL_MG`eDb_QT9j^6ENjS1ADK&MpQ)XP1`t<@=o2FVaftJ@D&YK1E##`En z<9Z!VEeC{a2iCbF54TGLdh7O7c3`UKWOv=22-!+(H1fDPdbF@G) z_+&3Kk>H8W$Ce8Ix^UKxXQ?LbyL0)N?mslCx(`0E?lUsBhKe37#v9bk6AbrFMFwI` z=e30fId2D8eVbB~=&~J))hl59B>DZvKAGKTE9;?dkKCCWR&wpFUgvikEgks+kB891 zLHAU2gLb6A?blDNxVvX1qV{{|$#K^aqjGnLhi{IPtSe-$cegOOn0&Xd8|d6Yn|Yr@ z26mh(y-wO1Hhh5n1JpjY%v`8o5{;k=eb06(qBzTKRjlYK4pkH^_NF3;+q|LJSG~Qi3aUhlsEgBUr>M%H{EKTN zPpvSdmmfw%W06Jp5XV*&R@7LXx!I&rSm?os#p-WVpUe&uG!|tq5ga!y^-g3q{u-Y6 z(75Ul7@6qeAL3WPQq1wV>8kK5qz+;7H{kT7q0-Ckq|R83J-Vr0Cy&Rdrv8^pOYIbz z)``1)fd|^w{&6b>Iz-(2tqotDnUog@cJ~N?9^`r`oQm;A_blL7Mz@k(n=N&$Fs>pj zs)D|GHax!27|Ke^aj{U|xTqRdmJ7QW^wDV%!^N5C2^O3t9_S?9Y#U*kQ>tZYz zp;lo%8|IW5*sz+)Vh**pAc6Hwn8M{L$>N2kfY&k$rdo-7F40liXIrnm=R?>%y(mE6 zCUJbzP~|)#!JY2gz+>wInIF%ND>$avQSX#~H}$;t`H?1yz@hv_SqtBO1|EItUV5oi zw5aJFQa1c&v)df;{bIu(SX2omrjVtaPHcbN2_TW+DE$Z`T@!`r)|mOv0rk+@3%UQP z8jD$uyV=HDwCwOK+zYz0ej0ds6)*dy$aMY5Q>Shy#d-~qU}aLfx-1z%8%kc&MCZQ5 z&o{|s{(gtR`v|BOr}ni9lN67ax+kA1JFF(TAkl~_@4_RIR>SV)ctVTa0#M5`Q#5+F zTt+av)1mtpPRvK2(s$?)DOVhvS3fPg<#l8QXH7}Mgrgm!LcVmCphFx7Kw0MpGNf)& zc7=0`Vc;I}Aj9+MF(}j5W(nz{z^8$KYb*;^0<-+O*xaf4bobp@N4UG*gl!>Jes2vx zwY6qt?q(}5eoz%TF`o+~&8%|{6gaFcj<9KA}uzguLko4HHb4IXHNiE)#V)0wTx#d72%E789|-<{!r)? z*P4GHPFQkUO9@ArW!!6dH$iQ!XKJ!jFj;ZnmvW@-H1t_Nnprn?rfi(I+{h9+8}Qag zvP=tp&Ub_|?^GlMuO@Qz3Z=Bqwo+k3^5u0VhMDDNkt0Gj0!}9CAzjY?X|)wY$y>j1l=eE1U~I5S{h7-D2Us|`(&reY zoGQCu-kN(M(%@62WenuyR-YFW@OR3ErYbxeP^;rdeb@45+1Ag88+T_}RXdz7Ju_CQpU%z&MrHv==9vE=&h zOkbP{08+lwI%y->+#iv;z+GfO4KDzxRluq-^rP@}hV;n3V0crOBBaTE=QM*#3a!== zwm$dCrg8n{Ny2#;Vgz!tbvt_Zm=s>|KvFQs_6qRBI|?T|KI4RK?EbnV+DZYXn!m<1 zqud$4$(xmK_HG=%Yf_ctnca`t+gHG&&M!x3xR zR}zy#bj1L{6d^_u4rb|{&SZ=iHou5z#z60prUULDSy@9RAVGeSR9_qtn~)A1aY7Zf zX*TO2oG)*7{g8J3%6a$d@5tBE`LhEZSUwD*=P{e=Gm>r$fImWJ@B6yazPmg8stg^a z-MNl=W!?j>hDIaF$qU}lu%QX%{DRdwKKE>9e}<-=Wgug$rvgt$!Id~tj; zTJ%9ur(Vm%WjABxZlzwO4{EyYs+v>e$`jyn^lZryo=PdIhb<Q0Q zowGnr;XA@ZQ5~BmkA+sLsRyFY@}8H+lG7fmk7Q~NmPFr7YG@axn|gycZgLey!q*ja zdZCm=%z=tw(?Wh9GT7(8xM^Z9_b2vG{Zh$GybFI?Gu0nKW&MNBB*axnY^D}x9P2o8 zWB$1R9A^HjI*wH&?A*9nRdVlRT|D*tG#W2xj4Jlfn0QJsKnL8-Ke{U)KE!nLGAS%j z?!aaFeLz0`*4;Hg@0D-md$Zb z-@MH+KsBcPW}=`Pna>&hq)Lo3k2l;67c{i*ipiU6WU7x;+Ngb1%?!EhH|1`ePC1p* zx6Tl}y{aR5F5l3ywZ%GlY9!SD1E+f}4Zn5_GxViGdB$Da^LaKbjJ7n7(q8 zjwKqf_((HYzSFYESXD3ee$T7r~xD&W?MoFy?@xStdyLCRn4#d{Qq%N(N=6LEU% zD`^*_aujqg;fOyNWt+sf-5=0N8B0y&UVr35J3~sv>CJ~5SLN)@5MuL$%Z}?VP_f?M zqagMu-&sj7mZ%}%(uIOvGLHYwTfLD~b&$@MBUUv(P69WUZN`q!G`^F<5j3!--SGbJzF!k0fBn{he zeYTz?BZRc&vX`ClOfgMLaV6JAHy{e7Lo zylo5VCwJ-Jko>*+_&}yipq^iIozsYGGH8)3L^53>!C%0@)5bXZucR-<9myRs^)DAG zy}#hM^`yK$lnO{+us@-S3;ahC{K%E%${KLB57BKvjjORR3V9Y&2IAZ&5rm2b<5*eL z2Se0(f>;fNfc~<^IrM0~EII0=qyB7T2fJ)hIwV=hY<=?5hNk$C`Z|Wv9i_L*NzRdh z4mToKXp@UNPG?C;_4B2@@=%Tv(t(C1y%mx>=L;ro%I>l!Da-d}Wfd=*64g)2U&$8F zuaa_M$@M?VMGGSBsG}EMb9U>!!P9xuuCX-hIn?)Q%UrT~7=A@r1z3$Jxgl@XeKtW( zFZ;=4(jq#OFB^~iuuHKO{zGn$F5Eyym+1S3a5f) z4|Oydc0dsgYP9-rQh`g$c0QDeH%w`X7echfG2<9T{@@GIfMnx1d{Z#(gwZ%+2>4)7 zG8e<`LaXpFXmoG6c+6?cvnpRAAb87Ed1ex#=JeO0+)Glq!N)Y~D~2Z3npjdU2m>Ad zQ1EY{aA3b%p+R-OiRdGggv+e<{&$}4cr&vE$iCwOlQ$m9y2b;rCpJXu_+MG)ml;HjHjqN}upAm)j zpBhmg^y@PEeG;UP%#4rnuwsMx1QT6A!IwpvQdY205nH{8?{_c}L9W_R4q>eb0!smG zdF>}cF41zDN$((;bt>lzlFpK5H{kJaHR|~BD+ZS-^><0u%$xNtsOYauIvt(PXyp)9^&Q-; z>Hv@zM58i}5Q)+sQ_S~VC8-E21x7=WWBdbgGHO0W-(b()71zE$ZCv7Mi@vBgGVEL` z?%*(Tj?eGPHq->dh3xT)uHmf}hsjkUt-XM!QGMQ7lnD9#ifa@!3zas2k9!RzdD`n> z%zM?*^#yt+*gAXBGpNxM$}I5GL0EOT({z(Dx=W`90lhVpT{+cIKU6owhXW?N{z}S) zOH-afNhq%R?!>(s3_LLPJ@~_`){=WfYkF`Ws@-#H^Ork$rScWk2Z}k=jEe_;98s0= zw75(+I}Hfg%XryeH#1C1`{z*}xKx|^HxR^1{Vk*dr$9&s`OUkyEMNxVbtcIpL%w1I zGJD)#^1>+0httA`^e_f|z<~c3WElYfz!Ko<{{yl>NT-k8cr5Xg{|99MBm9F# z0im&?t{-atQ?j! zkWH61;xTM7?2!Au`oUzW-u6w(>PBXxStIOsl~H&;=X{sX{#$n3-qV%7kk#fL^Yf#ebp~5+=aK^hX20lS*7ki{y_C8>scbf|7lzf)?zK& z)$TM}@B}xaHGFJYweQQHr7Sgf89} zN8xUXe)4)aNwQX8XXmO?o?zf)3VyUT{@{oAD7=?MCY$s*lC2x-OLCTH!fH657w1ke z*Ob~~?7j9elaKIGAvm0}e`Y8A=Ly#V>^95!0Lekzbf79XKCqUO$S_!&ExP6EWI>`V z=@d)Bvd$PmQ$5M7!01&}i9E>TunKj?$IE?sJvO^u+Ov2JSB*z{s>Tgs!e|qyz(XaFGGId>Kor7T-HAxuv_+HRu4uT zBN{k>Gk`h(G)+^*3(V`=w_xQ=XdlKdGp2QmEj|=%B-lw3j;Zs+;T+lLn8<_DsLqSP zWT-$Lu3OHx4l&)btKHCioq>xvWjrPSd4|O}ao9-yHdB@7BLUCEld5OtIlRVLoG?5> zv}oWwPLF}E%r!1!BS07PZC;M7x#`FCh+z8O@mvzID@ChC5qzdn%vfA-7Sgh3>lNj} z)i%MLXWOFom&t1nO@bBCt!E`8L`-7`?J`V@jW0u1K~`n_`xc!1(g)c&Wk)~dto|+> zA-|T(2$!oWd{~JZknIeOBQA4^s>hbLh=%G%Sn5BIoA6iPR(tjP$1IdxZ`LM1b8p8H z*!nEaN5WfN;mkJaclO@sM+_FSGVbJ_?4;pGSv9>EJhWwN6D+lX4ZMZCr5G_;xCk{C znn40)K~b0(OW=G6`A400;k-t_`+Aj+;mkeN!onCw0yNuq)ACzYGL8p2WRDo1py=N9 zF*?~!re+KMJMo}b9M-tcB{d7=Jqptt|0vSw7k>FI%1k>s)ZjZQL;n-^%UEI2WN0KE zl86X5-uKUb&=?aF6HC=-QfYgXDN9iwn`y(BvQ8Oa;OyBSV@f{{FrF7Z&JXcMtS0h! z{npd_kS@$it2!1R_d-bc(|8=0xUSPl_Q5%T`FQ)z0)SnyU?7G~WkI0U(cK}Wj zD*+T}(bNfavwKpuO!{dwB9ycAICGqTi9(&y_97QTz~Q)XTFfpSdW@>{%vxSsplRi{ zsMHXvxkxN?QRO$dDJLOiyo)G#1%J$My+uffh9%c$q#I=4O`7H33 z)nu0LuNvxkt2aU%R!|p8%9gK92O_6ez3$yvUY}`BXVAVYhbZaq{Yp0a!A{dZ9N3np z^r<9L>GiO}B;#mf{)l-(dzw*OhujWA6stNRBRyU8K4qEX)ls|6-5cJ}bUtg&IkMTK zzV4joD~jfa0lG423G9eBnchs83^uz&(MCJ?0V3qGS&8UgOM z`_x!`{7G6cWK(0Ljf&33Rjc)Wb%?x`{@Hbht(t_s!-~9Z`TTg=`RecYbLhuZ)O+HI zF^bEMSgLon7$1AIAG>npskQd-C3x_FuucsQk4&uKjdA?Q+CE#XFQiE!DBB%mvI*hP z)K{Nz-u6yY+dm|<%Xe?YraICuEsXvEsmFvqb(l$s_YJbVtH=yEi+cXWwNzR z%=*1h!R?t1t@r6`=_QXG%onyyBwIbP%9g`F>+_U!@h1}dFKPkU$C3nqH)@}p2|t(( z^FM1@Vs2kSG7z3(`9Bm#wlP_8>YqMUyq}L`6iI73bJZp}d#~gsmdWgrnm{PMbkF~* z3JWzFkQ`V|!P_FpJ85f+DYN3Ra-E8J{!UB3QxF?r*9wzW5z@3sYWVSOElsvE4)T!Wq2)1&vm3P1f^ms-S>?$Qr8p z{E6i!RT3L2={}La=Q~pJiDX5db0=G9?npXcwQkDxW%=YZQK7E}`_5#)I-k>h4v#&D z3KIJ?SJIZeip<2F2h|oO8w@L(;LIp;;?3Pp!b8izu|9RmJx+P@4ELt49)}U zOmGa#H%q-RIzP_FzrW3C8CiVDzdxg4z@w@rMwDATt2HE9+(f5<1p*>Yd>W)vT;w3q z!TH}qB-}pwuUQ00=Pk3+sBiZZ9t)`ude|l1XIgyX>aJP&R`8*-TAt3>`02}I1AU2# z0a>TjM}lu(6--^2+J9;2-tFF`zdU1~u!>ocdqKS!4;$R&)47l<8RJBXC&;6=NoonP zX(AQWVAwk%pYMSSk7o@}Y_Q;HA%60is_qY2a@W(p$Io=p>J#U?7bR$xD8{pijtWxt zhWuYliCvLI)tfOZHcYov(7#ycS-3fTbAUky^ED^TuTiZH_J_w&SCf$>RW8FEl$8Tq z5H7d0ESI$47^xtVLgAQ#r5M5xQVndEoC3)GK$PdWuN`yXBlC0KOLM88hW3l16t^)k zUEItP&XQ|(*so*E6yjQq!Ov^VZ3q1Aae}S}NYm|OIe~G0Rk4x~<^*ShGPc!UZmu^vxSgxw>+WpmfNU1mk>->PC_|m?C}+ zr1IanV!Eb~*E<~+q>LeX4cC)rP09Kp$&8fg^@*wTGbzO4QGcf$HgUbem(o{VpS;XV z$CJ$X(wKU9oBnk$shJD#tuRBQE0OedYz;gD3nTMKNahMv>Kzh7vc_d@pYfX-1(ZU8 z@X{VkQJTIe+#wX%jO8p z<|@kO8Or9{%oe!MX0Edn93uT#07>^vu0xCDNDt+#GN(Md&tbUal9S3oH)bg!a&UbK zZZ80;7XaO{To)gbjJ}+1EYc1*7Pb!ndlz7)j^#XNA%DXB>H;6z7KGyi!a?PmBLL0| zoCK!nlCz|e{`kglfLT#K2`Uf!0-(8z@7j10pZnxHsRQzT2pR+niQxdN${a65fJ!hy*gfF<0{<48?`8%fvJHmb^Turg zDwqImmHDdh{HoAgE(Ads0sy+evZg8`vCYq=DVeed1kVD>F9`Yt3$fv#PMVTeK3K7# z03}L8i@qbMhXZsgb6<59SJ(ouY_ZP;DPQ2FrQPEZFMy1pSVkAQ zSUy-C(WQ5{yszN+Sa1*)8~}oXtb9oL_bQ}ji#mMpNgnVY@kpPul=IP45EfU_4nxH7 zs430U)p4n*hpX65i=1~$SUJpz2`@$Oim%xO40Ur>Y_{NH$>7|pk=Uqq{3N2No?Qw!& zyj(NL@h(nP$!gLsf@ECsWykdnMj|wf98bJPI@fVt6GTtS4ZJPZRZ%gm%y+OqA{UILK(M>4qDtQK1-TvB7adxSp9o z`idsUCElRA(PX#K906*yj}@MEz0?bchS=rvI3)4VKmP(Ez9@7X$}<3@)S3$bePkPo zsj>~+eX3dv)>;OA2|Dmw+)e<>;AZSSuxPD7SQpI=lX?yhb-{28zPO*lPtfbIvQBdl zaVMy%m}{CxYMo4g-99j|(9x@amC=txGqivR4%leU8*BpDhJg$A*eMkn6Ska5?Ssg9 z03@ga94LS^jNq&|U!$_^c+lhXFA=8H78JET_APVfmyUG=pjjO_0BytCu7ZvUd}?;1 zm7tkL`y*vS;)rM$PP`;M&G%EBup7D>R?b zwo4Py>mr4rt_emTDNR*z)*E-@`iXCI0(3WX2`h^&PYOLTnQ;k3gF!|^8r=(19ZmLF zL~zjKEJphUzI-@XS1FBvb^+h>86ABKUQz5l0P_OB zd0*Dgol#Ar@z8XbA-m8X0f`9IBI2aOwW$9fo$+m2VKKYfMyO-$!0Q6ryLr7f)|5H- zsGnUhD~Sx&fRr&dc};708e_mhN)Xm*E!`;X|E#G>V7G-fJ$q!3d6c6|(nK}Dp1rAW zWH6)aRZb8~4qZLPp6riA=Xyn|`_$~Osdbcm;L`-s|7&Ut`mZ2N0&oF%`M<$75Y7Bw zK^U0ze~0aV?fU;tjcaHBPl7Oig|mj9h7T3MQ^*;+p{-suFcfI@wzp}y0WQMBFS%;7 z(qh`*uQrD{?^SN*X3rbx&F58fq!7TOmm=pXC&n6^uICoXiNe|t^dT>$#|{b z4R-BMqgXY|wT8M6=L$vKKfN33IsOptP4OoX)eBp$oaQks6zu!9PWfG(&=wx#9&>A3 z;3ONybozNT&_8e9V(@Z*W;H{4@T2Wj%m+v|?O<}4Uz=6I^VZx|v;@Aq|OpCsabL9*ma+Q(o$WS58j{j+SJdZWd)pGN9uniZ*Zy`-BV zbF=OsFMC?Zg!+xw_a#o0qp3EXJT!zQ1|EsD&xPM8X69jqnZi@SQyYhA);iSAzFI%x znPTMsQZu0xi4mH5K&i&sY#NP0yCNbfZl}Bm=gs3B^X??a_}h0lD$=OZnl&`SuC7k_;2?KU zPU^7S;6VZGOO_(d)6^$U&Q_DkF<4)HeZhIsnMmp0f-L!g@T_N(tp?z^PlWm_q^pK-;uq*`Pb8pfqdyO%BA9%ZI5X3GtK9j$DajsqFFR@5#r z;t7B2W;MJY>Dm9w@cqRbfzRian!gYi9G*h(AZ&y$>L*CO3w4PNdm8;~iHc468pvdG z=KoRAWWjG0k9-HZj7L5SSdEd_^LI$IvAUo5z-A#M(ctw;_V)OE@9&?M!iW6f{hc55 z)$A>RF&NWTk|(w+YfBnZ{z+`*I~GYv(p6Qq6Fa)GIr-^5PYx+v|x)xQw|FmpP;8* zlzj4&svUY_6(UQ=!oREX9F`afs=%lSc%<{C{DRVi5FQ`@`9?< z@f1j=c7S?#+A71cnmr*QkNM=YnPy&mC5%>rw=hHMna8qDMhCVT8Z(tQFjAzG^qCgP zF@>a+^1N}+Rg$dZG=BeY>Mn^Z-@xju>Y4dpDU z+lZ1ICgh&pSvXoq4_YZ4T!LxKDS@y2L=WO@@W6#S;Ev|w@nTXaYLraIL)w2K!#G2+ z(O@lT#UXRummF)H4@{?fotBwr204{6gfcuE0}7#qn~YSO%VbhH!ryTX8vQt0j1diD^8&mpshLT27nQ)(D6yB`|cRV#!|VTO4#VCqrqp z`59t*fy}KTeb9a`z#XKPyl%3u#pRL5Y#^3GjN67cUi-`{#~e`Ij_U@@%VKSkUUo9q zKKMyc4Y06?*%?MV;JtP z>Y(SQJRg*&cY3{~@k3#?3~g;V$CS5{Z^($pZb#`V0hE$hDiwbzL)U-17*Mt8;;s1I z_ww(O&uU}PF9g4msl=e1nNPjf_aF5XfLSu>NXlHO}p@ouPNEk^iw%KK-ZPkO;5)k(!z3^_wuFR1$a?Wfp{WA!L)be z_X-${KGE3Zy96Bzk?2DjrjXYW2n$TbmT}`xnN!;h7NoQD6!e0>c|U%k7$}T7OoB&_ z&(&Q42E%w*0aa91q=^npqTnW~C*8tNPW2%+wxEc0fVM3*kSX%HYGiaD)lfl*PhW@` z(7#!gN0c-I=!5rM6+mQ*{SWTUP6z*yuzn5+)D(SQlAy_)$Me@0u|Wl?y~G0f0RFA4 z^1-GujZlS$4;)AN+J%7q{Lp{|(dzINRa=%20X9ii4(dpC$tGmp46X43$mW7LB$(j*0!w)eP+?2!Xc|JNn!wg$9}tjW zswT*wVE!C!3NnFX8Snn{1QElr;u?ZJ+$!iPK3^G#BYaJZ3TE9pz{328uHXP=K|r~! zU`YYhF%=)C1M;lFn}*8EI4+RPj$+7{r|I=IrjMy1cV)f?L( z(67Za*?oayIpSuJ1hQNu(Oh()qx=b&%JGRho`wK7wCFBf6``Y-FEe~ZVC|e&=bvX@ zl*gEz;T)Jd!X-q?2BT2Vs&=p@yUe4qbLMga8_%+usbu(Z7i8E!@5st58YCqfDj-MD zj{+4{vlTw)`;)@R`=BIh5Qx6Mj}dvnx4tJkgEXHpSO_rkyoQv3P?CqsYz5;iCP_)^ zdujz1MUPdb`s8$Ch<(Gfv_ciDJB8!-c-})(<=%T(&1hGIK z?$}uMBsn22B@qb3vY=+fC|Z?;pgHL(k*9DdFs`D)CtL!veh1@@0cs{{i4SuO$$tl&PtzK5t z#(^md>uwz8Zh`(7vdcCm z{Hh)C>f-tytgLq5Zm>hHwELXP-G0Bv?|#qi{#Vd`A~vTN{`uNN1CzMsq6~ zRi#dNamtV&=euc-&%Odys^LFC%5=gwIX7?)eJT6+lKtz44!hRqa7JrFi7xvo)QI#E z%~1F1tPmD2pU2RZ#()3}P4<1Q=cKC8$xm*+*y*&e&7mVg)~Nq`rkc9^=)48%_kx#k z4u2R72QOc|%H-8;l=0Nxn5!XUlQSLmJzZ;a_`oQ8>;Lv+fj62YD&kq_p4$1mq5QJ* z^;P-jY^l4{(dT!GpIPaX(ald|Id;gI6#GjJE1@{}DBD<&(8d!5T*BOFZH%u&CJ;b86b&d5~oG|m@ zA@;_JM7LIGHKjSYy0?zUxV)2*{IqqMRI{zhol>a@aK>I=TYl|yYT&<%$K+JvlVQ%@ zb()d#9cC}h`841?@d}S~#+3Q@(cT{x*#MrO55=Z}KbsJ(6&#j)Ba@`R+Q&K&gZrjw zloF`caqE_@zBOkV?uk3;a9iU8-~ghGIOnZk-uGCIF7u0L3K z_)Y)2U&HlNc%tPZ9|$%(^CmyROm^U3r>M&wiA=mD99X zVbCu%uOU`!7VGlyd+dj9-wf|>=*_-Z8H%^&i>^u^Aot(&CEMAXSL2)quX%lI4P7vn zasRb5Q=Bo>{i;IX&l{g(>#Nt}+oMm=2H zcnUQjad-?4b0xr32!?^Ok8QJ6#+EP@VcMOD%eaf^4bXI{NN-yqy1?i<)KO)7GT;`N|1|c&!nUi`XQW zewT6gaMm^B0Ze?;!8b*QgIs`E5!P~a4bJKa3$gmA ztn_88X~?ggsvHrJ;1-y^HBRhGO9N4|orrRZYs9BJkW*~$4r(T>3L2SzJa}?7BP54k zoBY(E2B7jf#i%p3XwToe0a-MmSG-g>{|!t)(bR7}tYy_dMr|j|i83$Mp&L_ckQUP; zf!1Za7dURdebFS$^a;=JagR;xwh7&(TdYfqgf~cIK=6zkvqU0C3K=`h-kEP+mTjJV zX${k3=g~VUu~(`k`VN`N8J7{S&O%+Xree>IA4b$|wOC=sil)I*gv&<=|O|%=-Ht&TE zr*b@P+Z-unig{Sgb8O_k9Yc#lID^s*=@u87bo!c;Wzv-^ToSS*An^81--ifq~-E2(C1?=D8+o@=iO$P10F&83A@6HBt?m_m2?1 z`d7o+%eq|!pOccH>6O*D&5h%v0I9I%bhC7aUY74ClDQ2}F$pMM-$}XY zn{io;1yQ&5D&u}XeNBvDi(@!*#gSRS139JvovbD)d>we#_SfFEzMH@7{bRQ68*exe zjVsKuAq2kJF)CD9;d3wOefv||sC&q=gLOb@Dqsy=yL+`ihtEgZL7{S_fBWW zMXZF`@?5ySf6+$w_)s2ai~iaA5=mW3Pl1eFP8cx7t5vsV!7Ti)X7h92ygp)h)@Veospop50r#-5nl9aZ+ zx_rS$*4FArQ!i)%q8H{moLkllU#zmQ`ifojuxO^Q_$$M#)oDV?PZTZWO{V8Bz z4*SPk*6+h_pFTV8?Oat8>wI=1npq@pG^sVgJ0PqGneZQelB)aHD7t+8RP|HY(s|0^ ztBm`%Zl6R-Nj{*DAFcTpU0y|^+K?v%I}s@==s!BIe+fE$UMQyC`E>DN_NGFs`S|nm zpZ5*V1_$Khvn;!vBQZ80tg6w2Un0(Mp0^Z4>o`WRCGlns1oYr=;jzakWd&$nnG{pT zHmc}-o{niu2p^Q7-VO-Vz;=hY#_|byr5jVj{77Db*(3)!M7Y4Tz-W3`x%=g3Ji(q) zYH^$PvNJew8j>L|ox$|o@i@LP$^svQJA3kJUjrX7RRI|HR>Gn%`e`A=4mW`YL!sD# z(;B1{PoZSJqNo`V_i8uJkqfd{=N%YgTN-26k`ycJ61)9N#XmSq*GZNNB!(HI>+cFN z&5hYL(sn=>BB@~3e9C4ok3}k+5vZ?_Af<>b7gcpr@1jPqUXOMGIWnl;BWx-zX)vXV zOJiL%N@+EPdr`S_`gy={xBy#H(O~izc9{PnENbFEB6Ug>NEDHnM zjU*YezMi_ej98OH!1TPG-^VV#4mHwV`YoLZQ@Hk~E&&pMsd(;2efGgkC>^w3kvVdDlX4&~0M z#uZ>8tnTxy{LO|2b5RMprCsE%W-+T-3XYrk+8BO8F^YmS*&scF2d4teDvo|1XnQBr zZ5%&ji04Ta2URZ=)2=X%n+`D>P2;P8D-`{kJvzHdYAb|ZvN7mBF)5@lF6)cTo1AAC zKlT6id*R&ZKTdhY-ae(+D=b@7@Mcr@Cs);I(4>HQa;*EmJ&0+@*A!|JEX37vQ$=LX zBtebbC22}=O`jfxSjRT1QC)Hv>brZ9@x@TLJ^kxX(rx8?-&xiUSq9eW=We3rlL+CodN$EqnQQ6&9VN~l3FN~ zWtu5>C6+R<`tp*p$O*A9gwfvksL}M*Y;xO2e-V-GGq|82l3=l9#LanzpR-V)A*)YC zI8*%mvaGaGhyqgk42y6EWx&;3{oGX~#L0Hnj6=`P8-7ZKiJzM=r7(x&Gh&Bhu>cTh{#o_1<+x+dHoXjFe-_+ zv{G+N*1l>~jDtP5Oa+vPeVUDC0enO16+04UqaZ!=tcuaW3IZ50M+zZx7v9aTaavvD zxoAKnRlQ@eY$iKo6BAu$b>T-$Agmc`G zsH!%FW2;F+$XT?-45}J7+P8-@Tvx6)j8H)eK z$6Ii<0dVV@Aq2N1xCeKNJH@5AJH?^6JCxw=?k>flc%e{Si#xPvaS9Y|fz0{Ny?18r z%&hqjd+q(My`Ja&i%)JB*NW~4alKJ?q-E(MVo3g?gPyPW8!ct*y88N9X$7R+Y3ssd zN2mSIkJ^DvyEA9h!dK-eWW^s6Q*IBu>N!}4U+(EwSZfj;1{?|JKy|JR}z7DU@ei_ zmWX<9o3%xD39-kkg-Okr(QRhCXz z-qBYEEgphSs$df-QVbLaVF~!*4N|ldmC^{H!$#{xutg~fDMmRp={@Vm1}OY`RVqU^ z8P2Yu+c(p0y<@2?;3`QZ6%8a+a}?z(Ov@Uv2RBS*4HOl?tZs=g|1mj^^2p-#R8Z}B ztjSD}Pc{4JDJ&_@mwNJl-a5r@0L+Ftg`NMranDGcQo`TrH_u@OgRtE&GjF`QLuW&1 zNKTPP^Plmyf%sMR;IOc?*EA(bi%EJM5=UF8nK=>>qG1lz9H3G%zn9K2hzjRLqrsi# z7-la3)H#6a4nPgOf^z{n4v&qgDa}o!D6+~p!eH&8SmyBv8Djzz53HhT08p<`I^9sl z8|FY7E5Smtg$;A+=KmpwnyyfYOhLE^uBGOC#b)Vp3yYZw3Q+<*R)Tki)q8Kzu2HIj z0d|tpg_J)r6E_e(SlK0WC~lbj8Yr$X0I~T7jx$KGxA`csJynSUZqnI~%zN1|J%+P} zt$|bo16(=J%akl416h!@w$XM#xC0=pouB$UfajO(P!-xwWOVH+oahkt%$e3(_m$MX zYIE8oK-%s{XOJy?x43T(%N#QkIF9U&Rb&oO0B?}0E~8z6%(ek4DKO>K%@cgaUdmk; zx@3h64qJK3mY(&|#4SP9gJSayypp*SldW}jzrvYSd2>Lu8=n3gKtdHnr~#w=E@}90 z&5%l{-8(*}LPSdytP{M5gU|p;yrO_g4sbO9w&n+2VQ-1e0R%fg3jy=QbNdWM17XPI zUs`sQOe=ER^X$dc1dg}#udsf@0KX-VP&EK%SG$_Z+dwxgGcG_AU`m7PsOk#Kxrz-P zxI0q1e$g;46FNIoKZ# z&C<@R-XD)X`CUC7u1Rf)q@T9nUZQD$7PcW>?*YlUm$>FRC8Fb)YR99W8568T?6pO* z#QFN1sryr=%VVH8X4IKZVDhtOUWamk#Hm1-u;?z`hndNH+UubpXL-fY$?0D$&l5xv zK8(;|OQefGgIzpfANenn&L>|fM2aTP*+xXGhe$&jq}i^Fs6Kvg^d0`eewB2Cpgm!~ z?$G*tqg{Um!QDb&FTb5E(~`Ji4Iz=)x+X&ujXJJispq=NoFOTk5p*5Zp1`*4(`G&Q z1!u=e)7E*yp;oeHQ>+xyIn!W&-MSo z$xzgo|1VA^ZBGJ;iNA_{qP0>vtpk0-%3oPhHI(1#l%k9G*Hp1&co>K)g?S19Ora(0 zYBSdk!vM(=Dz(?@hO;&Ohm&!X0s-yf2zS&6rd#c1(sYeG8jR|h_lpYHIyp^y-v2qh zJDz&q^f3haf&4l9%6uR?aQWSW7U7q;2C&_Mh1EqxjsBpIgbuzL-7&=uAJ3Wue<-&`oDBJIr zfTkGM-}?erj&{0%2%5UCpbxV}k;P5b?^~KifSb+_Y9Stv|BU*Jv%VMi*dx|3N+=D} z0GnBp7FKxW&L%$5J2-fM(yxgBQR%;sui^(p+M!Io22661$D*4Wh&1YBxJIR z)WRq+2{zti&WKXAy)TVfeKH`675Fi3n&aeh3rqOcKx|X+RjBG$p7SsN`H!|YhTAbd zKkUhM%}*zEU9}lSN;7p?qipgzyoGZsW(b#ZN^@w3>2g12#WH$?;pUyxpd~U5QHRe`YV^PEU{{`Cc?|8$HXS%tvpLuV$yKBfPh@zOUGdIGY?Vn|E!+ovyI<(|H`xaE_^_J z{V(13Nuc_7TyFJi_fc^OZpPtr-)ecxYiff=p|O>z2DDRwp2rEb@4hy)H_f^&rgiq% z@J3+P;PVBoy~|Vr`cw{A?ZsWW?i&~+C_wq9Wt_?lBimKnO@TU5d&LKp7pv=}PrUyrNnt+<((>Az@+ zzv}nls(liA(fIjK5-1Aa#*?L!jk(wJlO&SAqIF3%A2UI2+)J#8cof~WyIoD2Vwnw) z2P6 z8ki_l5wSzG(@4HDZzi|S(x#qb4~gU}{X*RW?)P;PR31w?Fs%vTunL%dEV@sxJ8cKq z3_a6Rp^z4b)$^i7%%rta@>+O0CRUXt_RV0x99;@r5)~HKGARw*PjQj0fWG+E%DSJ4 zxtLgw_A01C<;SJI_5KQ!J{d8{ip>Fj0*j8OPTC6KgJV80tWph*dIT`$xfSdwJA@)+ zPS^feN?j}Wz8xP6#T+K8;z1Wp;U1Unz$cj?vF1S?EDFyU39SZN0Qe{I>&>BtM9)7I zNjfKFU)0~DP+~LlpZwQSa>@{uExF$q#Fee*AzfID;zK9Zlv3j>MBO!ghCVKw3Pt1} z)YGwK6KlAgLUM*oRL!M_l6+__DV$@6MN>6N>`@cTG`6hxH_H_x+_Wp6ztD2E^OuLx zFZuoAvFko2?8i)%$sBJU6G|}`_LY~RaiAYpv9ae4q8-k43=C<(HEJrTFz7X2 zUiw_NahJ*p~YEkb zhZ_K^2R}{feUjL~n}1I`chjwxOo`th31fjP%F>eO%|$WtA&ESHS3bc5L( z7vc=OQiZm$4=NWN8WA?llEPsTap1!Yaq%hRqFLk}szGlCuUPO3vJGC(&z zj)}aO5c1jOr^?+XaL)Di`5oTv(ih%`ONFlwiBti>e^(*9{26#SSGmBhEfCK<2M(|= zfgn}k$;XWarL}(_XfOx8R+tK-{{p_e5~n)YsnZ=I53OimXF^o-5MV5r_Odi?dhb!q zTV4*Azx7?Ni=sOZ(j4~pFF3lC`q|6(pP`EIW_}=nr!eWRf zqplJ%idcpY;#M?so}Z$92s*7TkSj!e9h*fsvG-P`b3S85`ng3)=o%Hax31qM+C=F zyxG3$hgKnY`o5x9+!5~nX!oZLf=ycolyE!VJXj8U2-~4S%TmRXZP~PLdHMJx@C;xO zMPdVAM_`JVc$gzHjWAJ*55cnc<>x>KZmK=2w;FOxskTmIQ6kc|W~ZYr?Nwx$8kmOH zuEavN8v?e^b6jKDYHxs&v5<(UYpg0SIGTIli>{g+BR_`hg;KlyBx{`2n-Wyw4@N+rrq~U2;02BVU-Cn zvc{sNfla8Wi?yuWf%A7e@V{@dDJyuk%o6K2;hfBB+U1-K-`@BMk=%Jxv4qEm&qo8{ z^2uK6n_LN-T(}!tnCB%SG4ufYKSW_D@r{+HrQB9Dp%Jyr?kh;aO|9GmRx*=lPIco3 zn?gSyF2vH8u<(0 zqquNt5-|EjD7!{i3ezH*5R&pBlya<{g8a>sV^#Y*2#{ZrG7|xq^9q*+g1EnBCq0pC z7(!Ff)3q46kiUs$G6S-YA=FQ9obPf;)+5tbh=5hb`EF3ih{gZ_1RCS_1LYNvHWl;{qpfV((Oi|9g4|?e7ekX&i%Xb7Me^P#5mtUpv?i)X3e~?h@5y?fsKWzl?>B6 za4r>1rU@y>#)r&yNTk^z#2AqqaY$8|rA7 z+;ErkRKu{XY2}fVNl=>Y)}770EaV^VVH;}w{g&jpLLVZh7_;u|vYdlzRFZ^FkWLy! zCLFIrSE9fGHk$L_H_UBcp!>Q)v3nE<%I(SG{p@ciIh=mA z$FW(VUaxDWN1OOHM*f)Admonb`hZs%3)%n|qsa-hF{bkS7qXWY-up*(hZ$^Z(r2z}_bB7)*jRho~s|3YuZ+#iH;yc?SzpUoY4>AgnO<+tg&4I#dL~P;-^|)1I&iI^P7;J-S*yJ-B1;< zf)h@C%DdhbXnV3-i?#{qa`3PZJUJx$u@%-d94%C;Dw1}9s$gVgJS}iy3|jk8kQN!_ zEDlU1U2Pn|_~*{%{t*^yL_-N-scmlxGvZ>?vX57!05UtS8m7eNhCEouxI})6o-9VG zD?N$}Ewf<4J&tIE5LX&OYvSsddTMRvYw086iiKN6YU}7V^ACo&{G;Sx56pvq@s;

    • zZixAJvLOf?XysR4&zvR;IG(}V_GS~as7Kmd5k?ialsxfx)U3yddUII1=^)#A9BX=2}F#^Tqz8^UA+0{+RFKWh4q<@U-w0+g1TBExOaraHxc(9Ik(KNV{0<>DG5NC{q9#MfGkg+Aw+?GV0#v%Jv=!T8n8$=DVzoGVXsf=TO9P(d46$gtWg8%#T(ihrJ3Hq zUnANFAngQL&qRDJ{Dt|QJv>JBqY@57LyO^YNg;1^22iKqhyorw@PquEKEsxS;}wf0 zh=Q{c@a+%j1ms3wKo$2qk+;Awc=1x{=o$_TeaT3@i~h5(Jh2U9G(m@A0aBfgI_Vhn1r!kc?&l8 zR*^o^13!R1 zy{8KtwQ={ir1xm%tXky$dM41(`e+~2W`9KdQnLPL)C}5AkSmI!_hyvkn|>(2L3dAM zWGVrUChqHE_=|3FUB^zFkg>aS)vXaa?4`B!AK@9Xszf^9rvclf)D!5scM+Qlhp)rR z7~0OnmS`$Hh;HX5SfDMwI6MaEraHauK0%uy%ydP%6kPVt6iK< zEv*-s9!gSA-4TpFS7^^M83r6G#kQl3`2QwB`O<|$@@vGc?*FbI{`aatdu_3Ps&``7 z@=V6L(%ZG7dT+;BVg26Cz6H?O%B}V)Zm$jetsYh<2);J~5a9`r;H?#@?MIFzWcV)K zqrbz@S{$NWal0s}bfc16%Q=`x+~%{}_FdCm7bGh=oA*RI6f)4W!+Ig~Y`|K+!j?ap z)PFj9?6I+|xE$)dG|uJaxN7PSBCJXdA5mBUMi{}XZlfzrMmr`iPT0_q%406Ow(4VO zNnrNE_H$sA&<6qxDruGdZ?*MAMOAStJ|CZRq@8YP12w{#w8?Qm)kH-sn7!TA1sh7Y zQ->6k6^tGP5w{xr6_=5w;6fj*)18+rEW1^LLUwCB?Tac)r0j7N=yXB}{hyBbU!LWa$%wUOZ=5s5&LN^-ws|FOJ)- z>XYWtOrO~3ejr5Up!P;&eUn><6=He3-%z<62$gH_ZMS_hkF6@FQcq36b$B-wpPshj zqp@0$0(rUsC9Gu2h-*V>%uoMjLi#@`VW&uqM zH?s4#TNWfCG>^KJ_jr6wqSgX^3k3rn$k+L??OyL$cR8;)Ny(!kcGGda4*np64!ge6ja(8EG`i?YEwBkru`l8r6k^bxcDNN1vg; znZg~?x}umT2mU5mjw{iRq4j06I`oc{>NgNF&M_d{V>O)RhHeVW64+zmrle#YrNNGN z;S^|mB`IAxfAkB^u=1^PKiS|}%jxC&r>O#70kYk<0p+lrF~aHDor=)LW5iA2#xIVV z6^zjrNOEbwq`Z@4nGjba87&3%zlQ@An@^Dq1}YISP5>-err*=%N0z&}W$m-2mZ1)W z43W#_0#7j8ZwF1b#}D&$wvoc4INLGyxMVw!3t&+5a)}%K<2`8)pWcZRN;a=oHlNnt)tp{!e~+4HMqOGJHfp;r8pEV z#odELaEcWT?nMd|FIu!{v9=WV;!>m`x#{n&``*3xz5gbA_RKl+&G}@nnU(Fnmi0t5 z{hK+n5|EDSA`Y2-ik+CfHL2<=PwwD%P0!Uxh(6w2kRe6pW&v}DkehEI??}d2)w{ni zcb(+8u;HqAsBPz&1A_0njjyw>tMR=Rx{R;Wo#`_Ngxt0aCoM1vua9oP+H5p)K$cEH zR^Ph4hF%>5@ITgQa`Pk)UT}orBu%^V$6T!y!%1^)Gi^5bYN{RGv3xXFpO7pULNS)e z&EDU;%uE(&EUA>U1H>|G!78?es9YKl%cRcvW~Tx~6GYE*fFw8*MOs-5AP`30ELwLA zE&{xKDS>I2@i`nNP*l;&P0r7WPkKD+wh5VC2)Pn-73F{xXXxwE_9JjyM~}z`#Zrcd zc{Vwt>tHbW#`^i&!QMR4?gnRMJP8WmGZrV(;x>NO=hjZKQe;T#gbQ{<^{4;-d&h)ss69bmszJ6F_g%EFn{@f zbKUCKHp_^RXk7TXIFIS&oHRpsk}B6mAeDbG_&2W;-Q0rmYypjq+}3h*?(#vL_fX-f z1a=uVM9h@QL0Rc(u|}d6Ws|uUqU>dQ`vJS3!Pcm0<1J}mDt*1`&hQFD6QEEBvv;;p z6mx}#H=SOX`Kksd_fKUYk4seaqP+q%)H_l(WFm1?FFnGB&)!_@6=8faIpeYnS49G? z#%@a`pHnN1Z39^T$rXWcd_Pc7--`aEs%aJdQXi>F+=5DOuW3Wzk zUGX22%4h?^P-VrZpsg;bMixt2X(8!^WJW~S=lFI$gYJ+{wWvxD-CWhzGfggGTuTyz z>{iGbiso`5L0cT*_an%fOz&)9RK%`8BPCb#`n!0KjPstiN(--!`H@f;uR}&&JY*7g z-Nm!L6;01NM-^lYKffKJNx1sUuJKwS#4(+PWT%P)2~T6XBH>iURDH?PddB=H@OgIaXfK0^#2-z&dn3FCsSX&RAoh1({H zbzDzSt8%#ZX2?kXUC1a>c>muA%o=%ql#_$KjbUosNRCF@6&rdvQH3V% z#;`{LH}QStraX{5JywEKg|UgR&sUZQnvnvgl7l@@Ucq*T+L5WpD+QYidDx@b1f@Ii z^2}u6i%%*<3895qkkZ1M*zD+v1TFwcK(@c#Yq5CFi;$5kr{7QvPFeH3!)-{{g=-E+ zZ9UQ(YZ%o^5-p4>zCn#Jd2Lb%t*!4j{?A-i{{|vL_yn_oGkbQa2~e+;0V+kaXwj4mYGz&45#wAy{yfk#a@|N zKeEeP*prGDDePVtaaJl(V2~@)5ZxR=+q6HL0FPedr!vdMcr$xfe>_07_F-=Q?#EK* z+=4<;r+3atQCFlv8#cnZpiIdDINaMG_Q3Nl(tvh<#@>mtS$Q3NMLBEBISAb3hK_(o zyPHdt3F0%0G``9^GT-9;VbAR1{H6JMXi!~%xjZ$l9QluTMFP|6Brl%sOIAbJ%C^0 z_4=V2A>Z7jJKi6b^`mWhqY9Ns3IxYNzAW;UKI@nZvjGYCe~+>+SY4T~HZ@s4QXLpP zN?pCCeCOd9I#5 zj&26!OeD&(rCCbbP-I_mdC$2WP zO^eO8Apvor#R5C8hNJx+3DO+t7}3#Z644jm_zuSW3;|CI7S?OixW1FKbsRiZv>x)KMH_cSWV@g2wb8hg znDw~6>v?XkN#kq`0yP6fnJVxssY7g&7lLGr-sq1 zx%>_Zf1XfVb%WmQRK^?~;P`?DYipkch#f=~`B zVx<8_;t&A*eU_p)os_VDi~~J)sl4MpoZyL2ejFY;InzAO z!M=c(CL=UkEwle7Q*8$lO&|28kAKn}`|mQ9Q>eUD!q%ri<||Eu#3@ElIM|p3jQMTC z1nuJC!TGJp?*eOAm>B?>w8(@69FB666yLxBxW_Tbw7t8U%d+Ts!Gb0RmsJi#*DwN) ze%M>q9@S;_fJYxf?lak}G0*Wt_96DBcGU5FCwOgNfy${0dX_rz`PXy17A=wt%If~{ zW(N>CwM_c%_y4FhQY~a7+W&!qBxUOS>?1mlhmXpV@pfs_BfK|rHSApUIIP0f2V7u0 zkA&&RJN8nJ|41BgGTCvjSx>}B2E|I1QK$C-Qzb* zs6h#roO2ZZnhlc1iY+HDwA+B~CNp`_E6zIdXI@p=rXq#DK+HX1=Y@I&^*yYFo+HZ? z+KpAO-Zo z?uJ;W9*I35e6fY zg^34F;NU?C^0IH#@FGH?6S$2$>{MUh zZcT6Jlg7Tv3_4KR-#ubMrakLUUJ~HdR7P6yqNrP*tFS|nob#cm^Hg_Ow1--sGk5&K zzfDG3(Op+PNlUCGMmp?DUJ5a|Ck|Ks{)p$>Iq_8@arCXas|v6~eeeu7pZn{H@*yT4f|LXSk?mN^S7OOH~-*Iqf@ zNssz{_IV_7mHnw;4E4qQKx`Z1$u8u<+1*Z3x57k!z?0Xh`l_?Ctl_U$cC!NiHCb>x zQ}Q8Xka;J?R?~!$g4GWXb3r1jOTuc^&qJXn9YlVv?vI|~6!LgT9L@-e91iEc`fGfQ zHkDBIIe+vtunVQzEO)C6sosBHFIl0>%&=~>IkpRnh{20nDD=G_iQ>25A}vd{q6&tH zsk+e!mMToF16CjIs}w>#+>b?TW3y*h9|A!1+Mrt?bcFK#@7$z-kdZS zZ(2=guSkp5+-O)h%5ZGt9GPLpJE5B=)8Yg~$(phB;vvN@d+#@iUgdP1{E#pHQ0Sp% zCnFVWyLIQYc#;%r0{*LaOZ_(#;9A)!b)?K0?0D7q%=Q?hm*Bs58q(D$l1v&ZPRJK_ z6UMmt#j>oz-cdqBT~UNn@m`Whzac$plUKgAr@f9}3O&A`c#HD7+hc_B>&@J%-DJYZ zB;(*ffdW*@>6C8DMpnxP$8zb*ULuE9o7Rr(^$EJ{Wiz0+-h?@M(e(bg*0W!hv8@YD zgwr!f|3?MjZ>;&xybO7iflz>YN8WjROcUpX)|-Und%s$^)DL02U&}vPpE6#K7dpba zuk$`z;wl54@uf9d_oY`G0W?lP8}k#?u3qeSvp!cP?!}c**~e@;2R$V4CNDjZcHdfQ z7rpQnppd#K4DSnR1o`O^Tt6K`66^o`b({c!zb*$KeAOm+^+wp48O#Tq!dy~{I5k=-bKM<$nd6GqElgl zyDz~XuTVajIfk6UEYyct&~Pp4df(gM_z2(}XAX`QI_A8W#IR-o&2ZltBEX?QO7n(C z^_SoaD)(E`xy!jpv5}YmoqikMq&EUMXpD}yO12?P>5%4rR30H6gMVlFUL}aXfNm{kDW$tOa%JPco+>;I6+S&OKo^~4~9hcW0LEod({E)KP9_Y(QHC2_vzv%S-0 zCPBVEt#W<& zhPjd~HHUEUvr~={v4V|xy&TvmPP8KKp&HPZj{D}n%zdeT^wkMD%m{36U7m*`S62Uc z$U|B}RJsFsl*x>X7?a3(D3#G^h&MI@$34n{m$i~R&C8wI{71)uBbeH^o^Hv`+?m&N z#o~s4{k>*oa)Lf8v`vk?JDwC98V6>@r4ciRR0AAg`Bji_rL{*UU}YT929-PI-iXn1 zgosJcK7yy}rqCozf})s|)*F7eZkV*UHE>8eLMYRWb+BE$jinR8m%jCf@At#=J7?k=dp0l zHkIlbE6va#4O!*P-K6R#uYJoOYc%<_nYKmNjIQn6^ zNbfBXHr^2NR=ZFf&xhP`{FvWZCv__2$)E@d;iH;mW`hF=P+fsP+EZyMjwc(GhNYRg zWt^;%4+5x<_@Z%`FdJ+`YLq-!6_7J2p_2rpO8K!QV1@{W5@Di95-`&MgHkK0revn% z4V|_JHxdUxn_n)j$AEQPW#_|lZgv?GLMAr%#YU7uA;aL&BYEQP5hTgXS0TUX+6!Rw z2pPyPDC}6-HJkSW9`8)g~2xYEUDsetkKO)pf?obK*HvsqE|(SJeFKD6it3fY*L4I zUFo2bF<*R%B;{OM!5tfF?6u3dcDPC`B_6^c!vBkn0>*dOZZVbKN&ep z1f0Ce%ZyvdvcLnVagF2oQKI6mSN@6P|5vjbvW`K*D`P_mBv_RHqRM1yDEa=|MYjtgkgMuwO^L&MiJzKJm4!st>w5x zq^vQ1t8~a;8YmwQ4#=0QQadJz1yhod)LlJ)mA{lU`vMf!xW}57Hjcl=+*gaxU0UM_ zHDEL8Qa^xHMD;t=Gr}I7L+}l#Tm;us&=MG6!50?c-3Q2;Usx&igrbPS*`U|TXEn*% zFOUKylmI+{0dd%4tKSXF0}=Y;E~E!imeSyjm`Dchq__Op&~R7hVPMz0>E#<%dUnG+ zJ2ff*_iJ(j8Rj%1#|(f+g?|rmMIjfSk%k8N#PJaF|8`m)mxwv_B2h&3??R~SO%#zl z3jY-r7n1*OSTPF?^xQa#c4K3Jn7ub3mr&xOQKv&h0p9g}9H^=LK&?y{*!6uH*pb?8 zz$k7gAz=hF^T&63E!jF|(Y3P!5lH<+N%U*mNdVs1k3TI~`BYG71@&Ej55O(&^!hRb z{CzQGL#fp@G9wQQ{+X@Fpj~p{LJvje-8x*&lY0et*P_xE%O&9Obsd0NfhdIU>;%9oMY5Z{b;4i*@dH|^#6AkeT zOjZ$q7xgK>eEpsnjyeupO`&-mRf}Uv3Y(l_>lIY#E-dVa8m(rlL-ZulKSF)~2uY@rKt;B1#;OZ~XAS73caikPk3=h;%aCZA82?EqVza zWkJU=J>O+~o0RkLpw3C;A#M*a{vOdm9E4E)*nsC0Q`RXcNpnKE>M4EZ+7r21qN4E& z)=y?ixq)hk_wFXg3{S0lJXF#gpQj!r@i_0Hqj|BVpFH~9bMgz%I@5)Upp!^6@CKp5 z`REw`;@#fpu%<2In2P={5C1<>*~Q9wU(gIbdgXwJ8>j}qR;B#6S_HPd8=Cr0SNqjF z<#&jf>9N;LiTefF?-#vo*w3x4a=Bm9*M$uI!P>8HN0Nh@ zup5hCk8Q|Q4_kdF+T_B42;r7wb?vgw_jjDhqu-U~@1y<5*%2`-Md{zM-kD@ZiEGG>%1%iV zI2Hz$>!*0fHE@KhRu~v&VeCRmkftX;pZ}I(4-&->t1}3?C>lJ&|+%DQTq;nm3~rui#Rq zrOp$Q7afbkN9NX<>%7Uu4t9MKbSX|^vE5IsE03gn4@4&a>R_VRkB1jBHL&@MqEbDa z99#VOD!DE`u(NWp^GGFsB&tqE_7`_eog+@UBV-YS3?K}wscOcvpcpd)^1TUu&Zbs! z>Ozy?S`%VM6ix308?MvS^xiXOk%7}MQjeM0w4~{92wJ3yky12H{RiA)Ja6wXFeH;T zS?QE}{w?Jjub;sMkqbdkB5_fQMkTbu^P+K`+Pu6}RR{?Wu2a8K;|^(kE3ZryQYbdCc ziPaER*JrNtgg$osm8%X>Nbs8NcQuWuS$PomzK^rL5as_-#9Y*ZbbDAn^Wxj+G}hO< zN0%EsCNUJooMdqZYW@w?`QM}z?zypmOr()67vCF$9TGH}c%OxX!eQ#C?S+VnnYdJg zTf6})tFFa}>8^^r$#lv^>-vJieekF_B38xUU#EvMUyb5l<6`*s)qAOyXh3JWqDr>) z>I(~8 zDKrMYJ*zr^yxHCvDdC1`ih{2g>u^nba2Dqr!{T_lDK3Mi4C<4YmS|kwPz(iTNfG*YVZ!ohFMGHXz=z>miC z)=f112ab-9i|bJTH}N28vbA>)Qn?X$?V%4fzvMpVpya>RF89eSkc~VoDWw>|{@B~e z%Kz&jl6FjCiK6apXxYA1i^LCEi47R(E0|CKS~uutKikz_*tZR|Ck#{dLiJc7iaI^- zG;CU4_yn`i1I5Ei>_(tX<%%19i4da=Mn+(`ML8;W7X6FCX`mQ-aL07by6VY&-HtaJ z0R{X)<+b`Y4=zzWi2tmFTGCiHopIGS^JE*+9P2}@(3QNDwVMEzAWHn|nl}`#%SRTl zNZ9b8P7EH)+zCfJCX<3mG5MvV2}fI;346CcphWr3`GJ^)erbLl)d$}Lms5P0?Fl>C zdiKSH=|-$O;F*uk*$X-l(G++Y^xJ{BS#(+65BLPscjCPcgXpR1$s_8mgbPi)k`8ms zb@~_9<^Ve=;XdR~Iu`n8aREW`!vHLmHA-b zs4{2U2%ntlZ>P&E)p+*>JUZeHkbCWAQO>X_bu5u-rTn-ul=Ml4HY40dLiaoZ(x*K;JRd7QB5tjTwAC1e?stGlM2HL0LuotKT|Rf zv@Z%`W^8PKhf;4q0JH%n?sA^hKBya{KJTbLofQEdWZzi!Whcp$sHay*s1DmiF2wzw7m_XhR_xnko?h!MALJs$Zs`hw+uVrbEv!T zj@pJ4=GR6crOMV{jR#_E$}`>Q8MK5xIRB#ih!ZnnI$`K|bS)ydSK`sk0Szm&eFzRC zwfsrA+)HO&|CM(|oUD9~A|B*K^MU1r_=5aP8~VtcOsjO;)uHdZ7(_vGtOn1=l%hZ# zS5=EH1ob@Ro2_5v=zalWR%pP;kUCu5e)Y2LGqm(R)>tbpKe%UBzF_i9%*DQkmN|Z+XwlEk*wYvk6ApNKhQyj`>>r%?SS4&7H4MjATR|(5G zkv}3@lf>j)WH^^|#aZwXXchxg8rm?~M<|zs{Q@Lkz(Go0z_6Tgl_bdVWos|>zLNU*$?Jq9Ov6RVL(B#jYKmt+E1O1EahPqaPb@y3 z>-(T9evc;`P3^!6bRVdj^>0t~S%S1p6dPS)cV2p13y%t2p!yHlExt`k}u;B}D zN2)&hnqP-R3y1da#WwxsFBY5UFIMw#lZ1Hd0|0y=gbp_pv(bRxTZVO_@~*StxHBC_zv5<97hI z3w}@;@hL4_=g0DfS%+L$e00c{ML3-}JgmlQpmmXgbIsOw@kTZe)U@o|`FVperY&1V zSg7*m@z7*um^|DX;|ZDmH40+oO?-XgY%71S8SO|cpB)S zP#B;zhOarm2SEpBl{AuB=pVgYtwx)`?;FW6Mk0F#s=TNH`YZ8;D=j&6F{L~7?AYDpd z_K5;>F!Y4VqA2~4dT8oW@2yS}M#wn%>Myq6f`Wl{JhMYbVOC6YYp+#`p5>8@&S^o_{HF{M3{-n+!vDp3ST$T)a zuh#SDHN?m^aI>n3v`G!g&g!-8;)<7E_syfT7bZgD9gP_j?FE2&m~+&6BzjbyB&z@S z;9yyP0QkS+#s7jv;VyX^kB1`=FJBhPFk+W`{x4z+e`rWm@-5X|_g@Z>`GJc&!>=UA zA}AD=&l@lGYEmT=B08FX{kz`y%FUhLA@D};L9i=6%ops#HJz`xx2Ck`HOCepoSIw{ zR^t))*|~j+?`u={s59Zb9gaJH-q8-EEO22(0KFlUztyh1SU!XjcKin+`$~HblLKuQl!e_2?^lZkd3Cs$r#-e&Mi$TdBT%`Ti$#- zx3oEJ|4tJ{&yJ>oWY9Yb_QbBY96QGJLK`vFL3<)-JfePY{UPOO$`LB7)Rzz`2)*G& zr+Y-N={*4CadqQ~qwp;;R;Ug3Yx|%*-D%$%N9ut1h4v|~-349uu7PDr&0b!J9G(o2 zhf(@DBWJgy+B()2_Ps>{!EEiLrm+$^aQO|jLwR9 zuy!tfQrci4MrCOn*?aL6ca{)yYT0IYqdc-j5Iy^oTX37cQ)N9PH2`k-JY9Ml6Ym$H_&cTo{QC)j5Z%vI zTN-2JI{v+WfOm@!n0_@~0MAQw{%2R*-X`=Q`iSLAmUu!^N!%ViIEb6(liicFdp*wm z1M7~in~36A%a5Z8*;}$sb(b%g)qWuvbMhn^J|p^!>%J`3*!N#?HS?lINY1^LZ{rRx ziAJKy`ND`kyh{wmfk%$OnwsJ=?8$1ORL-j1- zIjz;$;3fXayV(4sBao9lfrGcJCGO4pOlb@-IDizj^|UASG%Cn+Cn|6l6G^OLb!I?s zBU>?@O06*pV_;yZ4SXR9o);diP~QlWNqJn;wbDlxL?=Mc+P&rjx-$zO{D$V#Db$)^&E=L<56zk#DOpBz_cv9-Zg=6M;+u6an^a;i5zm@SwtC^#P?hu8b=vi-b(xuO=}W_*-!q9%z<-Iy7#m z5l4atYP8YLr2BdLIb=FWA;ULtzz za@X{7fXBInR!FpMayrN|l++W~;wZtU0oSZAl#~mu@?8M=A-A`yuo%QUf69BKSO2zK z8*~o1tFXC5Lxj(5CklAu^QwD*Zlp8e)@9hL!p;lDaus8*?&LcGqNL+@FtavauEl5N z%gZ^^I8fiT6)9{8nJt;O%<0=s9hpQra@dbWQi{UzJ$*U!2 z^3~>A=veu4k)uiS zy>7(%e0#uM56*hblbD=D9uXI~#-F^guWCMy`h(4NxL-{yI@F(MJtLP-!;iiKxAiUqAZ)?qzXQ*9Q^xNVK#FiS%obbi|zYv`6 z!}d{^-Tjr_(+n-#(hTA{kSY@VhbWB;dPBR9BdF|%Zvw;gW2?t>;6|)Kx?Ew$ zwFpQbL{@G4YHC?h{a!-4LJ}Y8nf6q{w2SXKfnmzL~(n5e@sBG)_Gxzleb^=2*+6{DR7LO5Z5H zKzJ#;tQrqKkZpHnz6}UZeUZW|%iq^B61Z@#bg+Rs$%Ka5oR{^|bdUeXeREwA_#J2G zsLmZz+evwUf6D|)+MLXhVbhV*0}ieg#o0I?bNHLHhfiKrvx)=(H+qmbI=TiRIwp?r zH-@1{HG@ zDrz$x@QG<@;dA8@#w9h2gU^)LW5XZA08GlrXy}!mF{u4U#(3HGCUoJm9vr5kZ1@IS zS3mYJdwYz)ZuZ8p5*9fitO#hvvtSL#Etv-GD$b&1JhbZ$Yh(v9(2=?U!nOXI#!L8~ z*^iT{Ya)vuIwGZkro>80J$tzaHT7&n1QAcXWPl{Rd}>5%N0Xhw($%FRfcEsaX$gl# zM{|Ec@tSJU>zy0-pi1+(6Xys|_MU2Sx<7lu`GEJw5FnMv6lWTrygL%&m>2&!GPGcD zmTM$KuDHkm)eeYrlB2Mx-6;I^STxM7DdkZ7P#Z~?iajNw9CWSkrDl#jOiE%rb(wX~ z1<(FaV0h~OK#W(njg%x@^m2R}m}3fl;jAQ3{G9=AEMNXs*eK6%l)prwCD9;0Jj?V) z92Ys#OR5sfpANM_$cpq8c$i@y-xld}08t`M;*}w^wCo!y^9Zi;Q^_om?mS`Jlwf1T zP38_Nnh2$%<9@{w4S3@4=;^Bz37^mYZm|Ub9@v)5+sr7|K!P-JUt*Qo2;L;dit?f+>he|8S9%Ik!k~ zji?^$UgKkA8pLOV{<}f@%2D#O&xF4)ZcNtZ%Ho!ceFY`L#PsHt|6QRKM7-_Xn_bB# z(bqX;t@ToQZsQ08YDt$&7Ae25o4i|? znCg~jj9lC^zxzqLfI~rW8kl@F`$qLw+zbXyFwXj?+`KnlyI3KcRd_@CEiJ(xx$Hgq zEU479E>T8I1_}w?txVoH3X&9SE+i0)%D1Y zh#X%4zA>*p-wvV+Rkcxjkx8yy@NN zxyQrwX4ll|0-hzyvI873)7oyO88X)ii^j2`tyk)V@bFCPx) zzGvrb<06VLv*6x&P5|u)=I@U$ccaUS6r1#C`-ejh^J!sCtBx>m@;iQH($LbHoAjM$X|V=leK_IJ2>tu>Rq40Hy?&%ar0G1jmwMj@Kd&Zfm|&f7N9OE8-nP}V z71tA340^4i04S!8soE7!K2c2B`7R1RW%&0h5yBh0oBmedf1>EnzSV2|OPmH;LBlW+ z)p|5QJ0lsKwP_w-nUH0`?Y5<4zy=Jkh3&k2v!opqJCko;8p#v(`&rGJ1JuYABuSL{>+eSr{bXDhouXX235nFqA0-|o6k z9Aj5{ceB55_2`d^?EsmAXd+FQCmWioBjk5{=yJ%bB2ihsac3`Sq@{lUi8JAJF_2Xt z7zffNr1keOd`BzHf~B@LU7EQ4aDP6ZjFwYkP8W0#Y(jX=qu^=~6Qx!xYD-hgS%E6m z>lg+mE$7*)8om`Wc^+#!NK=5Pm6}~|5TzpZ2(aG*mP_=?*^4+Rq4Z_8`zzBm_f3vv zL#2ixbHT<|q%NznJWzcj#TiJE$?eZH5Rtr{)OXhzdu>d+1TRN>%Jm_oGufGaP4hEA z6J}p^lm|E!Gg8l6BE4b;K_bG z+*OQ)q2il%E5(82Ny~X{O?s8`L9=-5W>|k3{!N8}Wp?|`)s*u_Z(J0iZ{Otb7(X~y zXOJWR&*4xnABrJoZqwS!Ih$xHK;wVRLxVl+=okNieE*O6A6X-Vy^FD&oE+sO!gY;#PmI|W#X6_^96+z^`USB7CdG!Hv?>KJQWEz<- zGO20p4x}OU-=XZkdGzjru7v(0vgC0YF&kQzC4rte66cWnQl~v&lha=?k0p4T^uGXk z0AIqMct*RFB;bAU=9ce>INTr2$EBUH47(-L6^zRHe%jYJoyK(ahXaWgu7ckc%kh3t z=-nGlvjz9?=x6lEIWq3l=_?lcUEJHkaZ)_CUQ4^7nfQ-9MFTXdI7GBF?g1O59u>|4 zCY)y55Y_w$ahgvS-#6YV&RN%ADSaX1mx+18&6sr$P!r%4e>(j>bTwDs;L?MXUvh=; zfv;&3rRoBDyC0qVyut~7sDYfE+HO)=21eC>94#d8Lb@%k=kcFJLaFoQpuXb5G4e4` zQGO5&ty2*(dhE~Bu-_~nJ%!sX*S&WYe2|7XQs9rz6STi5RBv|*R2gQ*vr-s^3X8K$ zJ9)W2d{E+qhuIS!ABT(4GzayhxalfO%rU%=NY_{3B^^qmh`68KxL7J8K? zn^RUKGkQKY30j8CqmsfN*DivXP9jD#Dlg{_*=xA7EmQpCn5vZu*p_(9oUp~TYt58q z05Hr~N*~}Vzk!c>XTa`2?;J)wQXi(hVB(4id$a~ht=z}Mf|kSxb2$S;A7BHF)@@?tQ!LgjsT_S$5_FcsRhmUIwztW@bB)g2W+V@XG$&Daebg+O922KDG-yY$`y1 z%zrVIqZ1I6wzcOv3vj|b*r=!pm;%P7WDLI`{p4PN_lsJAiZq)Boms-*!Ob>Mv`t)^ z*fxd+@txnwi<<%~O5{6Z(LlB?GcQO3EQnoch`B#u7CFYQ>(2sUYsiMDmuxuKKS9E7 zNYbJ-Rn%?@HW!vy!4tpipR8-}xExs})3+B$W&vr~+QD@E)hN~u(WDSIzpJ^-SwUCH zbwx$4TPD#(Pb{f+?$&dhM_*P!@~?v6Y)`u`DYiU)(tv z8=o|;vs?F6Pr!M*lCNM;yXV3kqz?8;y$VpyS?S#d{^I-?Fet%u=N0Fal&cE z*vC=qpmS$8;U`(iKNTn_iSyL>nK+xQhd+8Rcl6i3WQb|^zN=Bc1nMw8^<PSmAzDWA;a(IcK2cLG50OTnE^(tDpYq_R}W$>dU81PPoq1G|>I(;X3NwDJYHl zj5O9bSN{sAd=aG*+=#LYCx7EJ8nqPD%SH#A@GLJM_HCMka7U399*d1m{?zxvB zx7^F!9aGW&qZg|8F}6d?N#_F!&){Vadg3ae$9FvZ2RGlOv@gc9iQ!?IBdqR?I?3kn zv0}g#HAzIRH1*mo++tGKt?~WQe`@ z0Z*}d9MJP{xa^x9syy*N?|r%h44X&_{*6(?l8d5fNz_Rd8DZ?5wCnJFzXu0+w#&Fb z(ilM8`rB_#(SkE83-0lX5q&4tOg1?Guw#js;nNp(xM1% zCi_Y{&h`*(7Jf?t{ovYEz`g(~1&B$)MP3}yATKH-z8f0rNKjwyWtB%_?Hk1xxUS~< z-=Yi1m)c9K(Z%@wgJ3dmhu~@W8;?G0*v9)3LCu11eq+J;R*6617a2+r)e|oz?2O^+ z;(9da{vB8gy*OfT3OM__@#Edt*rz?)7W_;$Ha|hNvYt>wrr6gg$O-jmD6y`Cl&Bcu z*n~`}lDBC9ETVd4F0oVVlyMQgems@VNrEuP(Pg zSHo>q$?T#w6)k451S9j%^A5nj?v;7dZkKHrFAH3k4<#i=xyhL)LPjT}3u8_Q7Zha` z6S&+`_Z)_r{1;O-FOk|CULdjbJ1P>3I&#-**6y_p?R8x4M1HKq-m1>FIPoFh6=x`< z_!Ox^YDtS3yAL6DWWT3G-TK}WDf>EICavWeX9kQxUS_*> zQ}vzR&>S!V_9$>e$btK5Vm2BVE2WV)Xn#;AigIXbGI|$67-NC;)cSF&_~s{?NF1#$ zxvZoQnygYwehxZBzOOO!vsUXO$3zXU#^Yf}XXZxbG#LHLs#$V4Sg^HTa1$vum8CC? zwW%Y>i-0{lQPw+F*X**Qpz^?`wipSWW(^|?h#9}9YliXL_{W)n!rHzfQ=cBoRhBzZ z5=;0O;+0-3-_6moTh-m=(4z16d#}T;T(I7#gqf#|+~$72+7e0gvePEnXLyqM$^Q!L z8TC2{2OOXn=K1_-)g{M*@d+rNyfm-IFMjFd)6}(T5+H^(n1XA=x zQc|~EfaxV%Dl8a|*FBWF3}PP_P0u^hPFx1Tdc57FHgqq(sy5O-$Sam#Fd}%?ejYsh zH3=v$;cw>Q3gvmcN7nSGG@gI7BO|cq1(0FiSO=}~M?Xz7&J}+YsZz`^k&W8wlF9t5 zuMwg`S3Nq%M4i&`HBDR|wTVi#W;Ndm?B57XPHQrUF0uH&BpaSkN$uRGBwb(xRnwgNd{W`4Z6 zviiuNoGX(lQ#2~U$d4Al>cb!octQ&%dUZF`fMUSzWAi1Mi@jGHDxI|>(=r}2SmBY* zI!9y~$l^t0OyPc;QmU~xn@D!Ny%;lC;O1XitMk2%ZI4F5kIko-=;4X64rwT`NyS3~ z1;pu>g0he^ai+Ajkq-!u)$Im!TVcuS1gQxG{Ks|5^&?hn=gIsTHweUSOJVH!*S{Xv zr9x=%9Wet1$^|jt#s5UZLSt1RrQ!dfn`F8R$^C%tVl2yL+v?^L&#uX!POI8_q!$tU z?7QmPc;Zjz`6DmJfWTODtgU^9S-+=*?#ZA5BTl18D{TZhc0p}ZPNr+}p>*0M4GFoU zd-6fLB41vTV+HL8VeBh2xK~v!W;`nTdnvBvuPi&0{!OC-aO55xxSj9e;I{;?_$xvq zG=8NXeVo1lP#M98`dP>45l_Dha&0Ysr1yqX@nbn;X7J0(t`Q15_W0z#8ys@xD0vgl z!Eb3k8F12DlJLDr=z0FabV+UHijv5&T7M=ZN0n^Nt<*OZt#HlcxRhJ|0kCN3Gus6Sk1` zb4b(jZ|0Z@{YHu~%ud&sggkBR9AJM3Gzx!^Q7D$FjiraU1ek{?Ov9f9!PW#aXhK4t zJ_({Uv|gLwDZEct%PB?%D)A3Z2*tey+XW$3dS_BOT1gMPdyw|p(}J=m2b3^sovq-9 zt+J>5nuPjGDM2qGqll*C`vIjK+|aIYgUHkIV1?LK_piZzfV|}enDt1J*txywJtMKj z<=w>kQEP!y5~6>gM$Go^w|B+rv%~pA2v7PZZl)suYwKm}(j0gjJg|L*++P@gD`j2n zt}W*RhR<-3j-Py=@$bsfg`X;4FjWU74FeHl@yy3!cb_W#hf06dO9A?F0~!J!86Pf( z#k9{+3NJi}-H{PFl)f=K>rI$9rUQksD1)HF6ZmYt&*bk!+^^c!vQSj1`$dA{UQ;rt z_xgR}w{bk500R^TEB#tFuFF|VU#aZt(oqXbmK_ga1-~pJ0kXe)preZtKckb7u=qVm zm3_J5+@P6f$&;87xOUhMoPF410Vf*LO#65m`jy0z#LCnmA-Gyx3t~;pzlegWSh7hO zl84CNUm3@gqFtC)q|Aiif8y+al%B0kXOeD!i6sA)S{`vbZSrT)N~%*a?3n2CMS-|G zPq__V)bA$aEil$n14o_y|JG^npu8jHSIi4 zxPsrHzvR;JLC(B}QSeLx@Z_{J-83!f!L}8^6)w07^9l0Z5OPFl%R0B+Jg#?s@KHwg zab@|iuv~K7N2WuNF5ws!(TG1kd?D?O+OO9^)pcX~JI>F?x7u@Ar%9vmnh6p%5Qo6{ z069R$znZ}YCw>9Apt^V!gi%d{B>0npka-y+e$_`>J5KOc2=6X-3)cC?tOC%NQp}>;@K9mGl;_zJxY-~l z=4sBKeqxg91f4$$nlBa*ab2d%af4;jC#BrR8=?yU_P0SX=(-q7c>i_0$zG=i9;l z+Z`~w}x4_qx{Zh zqVwAEHYtrsN|U=R*bq?3{r(ERNB2bUUux7DYMp)U2r~JNk|F!kW`9A85L3hzDn-$b z^{T>GOCdhh=zA_H=a2)x5AXK}rLob_D&@zvyOg4*RI8&_Z~-BdZo}_ZeS+-vk%G){ zpz*9y;xY>u42)zZt{;ZdJMqMp&8 zk#uj3hRzR$3#k6Jq!5;RQb>|}osHDOtrVM|Ye7kCA>D^gs|{Sh#%SG;DM|X;2Q| ztl;Z>WaL7dqD8iA9hrRjg{8!S-nLBp>&ebF;od>o6SYNFuEERD3TP$1(Tj@aUY==* zRmu13%Y%l`8`0&D_5mJ*>>$U)4B)RB8UD|NkxJKmQTz!@D+i-WJ_F3q7**KH)8d## zl4$YQORJMyRJSs7Wzm?Lr4HMbo$1*oOeSe7*k{+H$I9&=nR(Une5W6AoCn-NP5?+7 zT{qpELHP=XyN_6Tv*G@WHkNlpe~&o|s+u}nBbHq|QBmxai}WD^n&=1f){*bKSR78b zQOL5II2-+(KRzzodvenFWhj!NbmY*4WG@kM_6a|qdoroIaGAmEw2YYLL5dxEB{3{; za+7ui->1YwCHj!>c_7`#;|)`=BC+)dfVN4>ls>5MxO2Br?dTrff)aJA^!Tfq)50HO{^u!KUe$&i z5PV1Lh!%wSN#Sm&gf9udv$D8s7B9Zfk`?_WWzK z0H-O}8(-Vfd-jlbFZtQ&^;LZ%tSQlHbEbl?PPIuK&fKrpi1I|(R`dF)fLimKL{kbdoTo)o4LdJfk z;F91E$e|NL_gY4%f*fhS{4w0W9B$3O3}neBH`k>;eXg`XZzOxX!8l?z6OD| zn7;OjaFc7kuNW0>%clEBG81bJ`r~VMMc4sExkq{D91GG)O@4b=UqkzM^3IphFJNX>=V&cOUhurPTi9m?yL9G7Q*b~c?B6b{pQU%b<6UsyP6^Cx{}5?eE|}#$KP&Y z?P|86Q%(H*CGGS#2+xOrN`h-SyLF^h)AqTnqTpPlWyD*AgjE7}*w}1|WS!VWI<#Tw z$Y$_^nf&BwbW$ElyV@6{z*}pl>HW|uc(0odqNG*;XRbWl&p_sxB`r6b<7Yxl3NjRW zWC?r{i6MdBh?y#54yEFmKQQIxT2`5-s+oFw(_ifN)Cc?G<>3^ul3LZ#ahRD`x2i02 zjd8Qwdzb{EQep@SCfu@r8!-QhbeLwqhOHqF-Nv3(>$y9MFUM<*=7`|!>=g8I?NZZ) zpUa-|s&q$Ey-6)@W7#`9lNkqKF4g%XS7heUl@UR(=v=uQPl1L-K+T4_QcFu zvEH7fU)3Rx#@8es!mNlTJ=%+#nBTSWFG9JYHL>X@9uIG^Hnofs{x|AR=fRMtt0Pd_ z+QXi}82ijOlK#shke1Wk-+PHCq_sT-5gUgIKy>!Md;NUM&Cf1xsCvr=yX;}i_s`9* z%>bsiwy7eC8UN>Gz3XIWmm;H4!!(tm(6%vzzT4!%wXddHk&7CKAM}UGku6OO%O=;H z(G3efBI}@3$^RVwPar$~>K*Wt8Y&M4}u-K97{J1wgZL_#|HYCz`KE&2hXMM<}smrIt_FuJ%Pv-ukpu;X}i( zr1#M_P93QZ?l7QEDZ$>RJsb@2-fS3QhlMG!Db%zYm{=TE+DqZ{D!Y8-fE977c=a+T zm|BwjZdtEc)cTM}SP#z`n0Q;mM|d^=-{xjE$!?6|4JYs4hSP4NAH*V99!L%ioGeNe zpb`^3a!Ak((WI9XYWc=xrfI`-y7*++jKD-pi2oPd<9QL*|6-wdH;8?ckH@X;y+d;3 zIYN-9z2KTmIGRy_PQLo+n>W4r{^RlehnpoQ7&7RBV>`OrxAvn)yYn3cCu*2cK$d#& zsIb_IAmp@@e^v=LbszMh2RcxAZ~pL?Px$FrGp&_K;=jEknGH^!7K#NXP8m3r;hlTI z)Dg1VEO?;{PXKh;*XYed9+BYCx|S85Dx1`u?v}%4p6}EI%|-`o2ShX_I_$G=C3E+- zkx_Gcb1F`{JUCa8l7k&nNhzDX$h4j;g{ty&m_B*WRcM7|%9on1gmB){JvYj5l!B`G7}c`v@(;Np+Q;TRN%Wudjco)-_~Rv zgMM*C8)n%h<-(_j27A(jS{eO{#1b|nwtC$2Ax}N-Q1V%J3{#Seft%#Or2vQ0UXkH? zI!I+|i<#o99aAiHo;tS7-P*Ik&S_S1?(p6}HmTI8GA+j=GL1LMM&TKmr*%6+B5tZ* zfIZDuuw0!m-3jl)M9C6MjE)h9f$lpADGnH!5G}Qom?&C<_Z(X)Hltk9^4kM0dI`AS zBb}EnQ6qGY8oVpIMEzN_dzOGG+8gdXWuy{L#z|4)K=0VC-R-!x=7Yi@8-r7CIX9n?dBnP#*6M0$t z?C5f0$P-TXyrkr%ll=zkSUk{i?6o^0``B0cgV45C2X-_RG$(Lw?q_SHgUWr2Ewt z#8y`Y&)VGjz8eGcdOWqi$A$%tz;eEN{X1%1+LT}4k}4sIe`5(r2Z|cSyAawhVJ@sp zDo4OsVz##mx&t3ew?57gUrUp(2tj^SUf2W6NL3{I<8fdG=^Nc%B+Hng#m2S65SS2c z9MTHVOacaO&4<{TvZzRH9QEHx9fm}R1|fbY{S(Zw(WHKjjL8vP(9r^Y&ddEp2 z-PXAPsll(z&Lm3Nvd~xqqV8;7>CoV|b~GSxRHXc+7>^g2*H78XVJ?(%Z{~I6LIcbb zS;FRgA8w1?Gl3ygxRQ-6YhhF1z`sqT%31^EGb+4IHE71iChu}b4QZtmF)DbK(lNMv z{nixWey9D#9ZgwW4~KqO$?}4`>st@3$f$JF9OQ17o*Oaz01t=M{xAItT4NJ)Ql z+T*NcQ^%*KfVmN71zoNVVtju&Q- z=w_uWMbki=6s}~EQ)~*`ukvBwv>%q)pI}r_c0=k!}T0+oCV0V%}MB{)IxNR&tYB*hyakR6Fe9 z(zum$;6h;!;haUGV3q9okuJTg;A)u_oPkZ&>c468Bl`?mDGwCmqm#5x(?-==$3MO` z(nl=0PzukF1R}7X;aMAvWMXfPWz3I6R#_HBx>NQ1y7cEwY*K}Idmd7CR}hU)zTP!A zOe%oJe^qr$uEYZ`iESW4L@U;HQ|pjdMafbadFpW+KC4z}wSzQ@*+(g&z@s+bjpPpm;P??}e8yI7JB>;y9FfCxpYUw@=@ zpi|CD_OGomxxuU*w*j(mr}+wVrKWzmoO$}Rm^$p|#?Fa_fVMDs{HmBd9QF%MgJ3F| zk_#Q!#m=x$U%Z&11%r3@DV@v|o3Et)H`4*Rz#?%iGf}4U=OjZmTnpZnQ-|OU*3ciy z@P7+Zh~DmDR>F&G2Ow;B>a-&7RK;bWrRge~lj2YIY?&Du>C{x7H_r!O1+bB4v1R`67)a-y5 z{d5?YcB_S1@n~22uhr+8_mm|Olj=yqiZrP9pDHcnJo*_=&N&z4Ttd-cC|;Vi6*hQG z)7x}SKI?{^5+&6fMGQf9vn&pWuwd2aB|nY=a<|)~#T{nw5sh#poTpJ@rJqYx2WNv4 z5#wM)AC9#kKRa}^dkXn`!Vq4#tVItV8R==>{AC-lc8UTNfSI^BSiWH;@D?1XYU&@j z!f|or(WU*^ixtBVi{QWfomPYR-)LyRD!8wJQVl~^JPK0u`SQ9oW4~gpvBBpYdr4Vt zC^fkWgSBLFP^X%OsBiAVt89663~Xv5#3kd`-`l&29TX*x)*_m%Gf-ni9Hna_Uoj0C zAQc(F+q<#DH|%8EpiE9XQ|;Qz#5<{)a_s(w%7T;*dM4kDDKhn$kFd8RUaYZVXDJ7C zw=gaK44iHcKhuct*wxT7x8Oi7tj%*n@v6nBs063P3HUhe>VDu1i!UR{u^GT`XUJmV z9#W-8i_rq+^ic*0DJvd>JFLR76a)IyH$4W!vJK&$XtZ>`0M4#diZlt!$e~6}b;>>z z^xssA{Q$GmVe1rqYLDZBWA);cFmhTt%)l?_IYl#mDQ?uMx9UlS3B!HpTI8&;9-SeA zA)5vCw12sCL;I(YV=*C5hRbvmJDhgoOu%rqQ3)wQ<&o`94-}Y6A!Rx1kE?P08)41X z3o}*>Svp@OlC-F_ZvIdMie>*`5QV=8PRu_CMLWUvJ!OJq?ut8jn;~|bLSuQE6V5^ zEOd@6hSlwRo)HsfwDHUx!jUr1^~fkq?_A8NZg-uB)_2J0o#w8zTmuH{Y-q9PdTH=p z*cRPl0+_7bVu6D$gQOqZnk!`6Hdgn3Yj&3!Zt3iFi>>2P1vT&pCQ*w!q0L`h<45@w zY#r(GZYCjMyV81eQ~MQL322;N7yvwWVKbz5v!*o@{EnDBw~i`CZQOX;Q@1F63iq3ubg9dF@R|$ zBP4E$x}I!!(=%!BuJY`qJra(x^$*Bd+J+gf{0&*;S9p}O%UzM;2kK>*2bK-Fy=O^{ z5OcOc0ACWDBXhZ2XyN;EUaVZmxo7Nxa(JGSHpCmeNMqk_n|El!s?ebeU51=`V_ac{ z`=G(e^eAcSFqJ5v<(26Mv%ZRX^1frfK=Sqfgj*8@<9p(~^oXJ#pZ0WMD%uT>K>qY3 zNYIRIbD=2bWZFSyI#)EtdN~F1{$shlT+F>f_1!=`l@sNddw!>~qK$BAB3rI4+X;Xe zSPMs?rv2wlf!*+gWYv0VqZ$@ib}%5nmpfB5nvqED>49Gx@=->Kx1e3cipKp{zf})=4XSbp->gnWzL8t}Uv5Oj@Ha`pI zT_m8)O0&-STIC3BQ&@xZ=Q zH|EUSe~0l+0#4|06UylIu1DS?$^gPAoo$7!YylG%r!z)NgOpsfu7Nj3!cWk7ami^)KL zz}U^_CPBN~$<>^px!l)#+Kw^`sY0ziMH97XxxP~GUH4P=FoULyFHjQ0HQE1yZEw%P zmT}5T3zZ71V8lZw2Ed=$Q4~1?rLX@;@$P=Cefw^`DpulHP^U;1f?L755dOwfOlfy1 zRLuELWg)3rsDm)1DUH5d_YtEBC;ZQQ{BHyYrDXE`GuF!qyqz0Qx>2xZF{t~i^SSA3 zbu)%jUv6L?>z_Y%vyVz}dgI*cy{yE2#EKMFJ_XX4madGAF?<5+JIb&+ww5^1#}7u{ z8jh}NO#3F#R}-?I=sPl}k(>gEjZ8sbLXDwp(+AlZ%qtqE)~p73j44JiZ_`9Ywg8$x#!UM#~)d{yJ@`7y<%uT=cGT4Jf8ip zjA`mY`k}(De5B(kYyUtg_-Emw7V=x1SSPve&SS~HD(jM0TpC`kkc)Xgoz(m8C%fWw z;2FYY9{vfw;+9hinWAZHTubMNU5xWX%h~F=m zR%<=}1J4#O8g5mDui-e?ygsMcNcu3RxKL*}5d+0&e2xathJ8_5)9O#~(+@G&_A$B? zOl@Hti{iX0t)qvv&+EznbjwsEL6mrsw^94V39mga^>jR6`yFn0-2|YN=(R&`LmwXw zWI=gBr}|`H7e`vt(C`=wL0XUs-Zw6rIVOcK79HOVhQbMcz)w@6L(IY@aS_ zsgJ_jxr(R1VVjh71i51#g0dF@>%sCiOblexp%?qztOgTrfuEaH@aGcE{c`g+Uuo?Y zB>k2)jIFn6BJ%o74afJ6KnEYMvHK0b{f8KzI@|OwuRtJ+L0it>3}A7=+{Wvvh*`AG z+~FMVucowNYWlkFlV*s2N?i=awaM##6>#b#HrOR`&7RF2zR$unSnF4E7vyQk^UqCd zUa33El9gan)IMl^H|wv$jdow^GY-@3b#il>%~$!c6O=oRhRC90^4DheyPsVU|3`E0 z0NT#_Rtc_apON8nL2*;_(isOp`w#z>ekAknInNJ4GU)xiRngxB{^y@BF;T ze`p!+7z~vGyrj52`_ahhiA7tvN}b`8^60>rshR1TnsTq}Y2ox-imf2MIZepZsb4uE zmAru>qjKyZdRFLvesvX>-(gPaNGB1>xG_8qz%TPWtzVX6->@0iT#yaX@S`{Rk52GVj|4_2$WW$4d5LaW{9ZXz~eJ*ed6arWXZApud-J-DTf-5Gg5C(&m z(B6lOEu=mokcY6Y3R9Qn>R*ZR$lk+4h--4*BSVTPlbms~01P1Ob35&K$`D=>QnGns zbnEjMmBhlq#Z`T>vunCw4v|vwJ_8I=?%jcx#N~frLVi%P7Q$bD;VyWdp1OG?C;#Hl z{mUzWyF;lH0L)@N-6&hodb0y|%K%3nHP6OdZ<#Xht?j{`GCz<*YRg|$* zy&J(>rED4)CWO4A=&Sb&Iz?IZf;QK14dhO`2^@jt=b;$ep8h)+WtM~bnO>$h;v^gU z4QCIU7I89;#G-<8O06<`_BR8&ey;*NRTI$xbS=cY5eoV)x*o$V)%zt1WC5@LQ}_5P zZhjXFfosa)wVmk+tu^-#l4C6?A(BSi%++#uxR*KI{FzFx)hDKB;mVu4mM(&9Bz{UR zrCG<0S32u0-3!ag8yT*|E9Zu#w*)7?4fkZS0dp^g#ImXN5&2mlB~K#YT#-Zoy#hH= zhF*#hop~t#-rRRBAcBE4p`g~zF0K`!FL9>0YJ(w!DPNan>v17Pq?V_Nn^`>tBTE^1qP1=Z8vW~i08nWP6d3M{y!b|PN{5U56_Y-5{ z@@Xkaq`SSPF1zy!s*dX<4Y*xOfHoyS?DKyPrz#HW4(>+U2OG65`p3Gbn$Cakb)2*( z%PS~E5~myLkrKdtqrlWGd?Szkmhd||P;chz)Y8{MH>Zd7JSec4FS~X`A;!e7beY+0 zY&`}Q#Ur|3sc|$n`6LxP5E7fGwV`oDEBsg)m{C~T(w+D{1^cB)2Aquj)eyEehQ zKgEDt{dhYhgxAkrA$ZbcGax@i=orpkD05y-^jRqz(wM%Z6J^qdtFO zCM|^+ZfJyeNHDntHtLirOzJKuCTi8-ZI*Jo^qJleZ39t5DJIEfs5+2W^_$~7tI{=L z)Z+U{k3j4nWv9*X=V6oQrG-ZU7lytcghT~8n22R6B2HH1$Z``Ywx#6}&}InhIZ(L7 zMLuOaLrEN~Xi(@+uHKDm+PA7y$)9^4Y7Yi#?HQ zF#JUe6X)G)v%|ijr!B6irYu~LXYjFMAySdHYE|T2SnUCK*yFBo_1*tD;)PI4Ky?+E z1gslM84~nFX=!{fgLiH}rVFI*i8LU|r=?0zxYtgV6;&094fhHB`Dtk~Fi_!8uW$vc zR=#pey?o~UsNcPTPEu?hTQp(#4y_I>M*O2$IaoVBYr82_yEjg@SO-i073@*X)Ny={+V)bE3jimD z)b35dj@)Yv`?MG1%=Lz%{qPNR9;cn#51(aRA1%i7~IlT5PxccdlFHOsBkREgtD-+C#V=(8f>eBq1%=yCZP5l-N@W?a* zSKg`{is@ZTb6N$ueJm5rE0;yjD*ppdzd+W1h83`I5dn_36;|d}@ff*F6sC&%VNi~B zMbXphaA+9V(U%bP8Z+^JNMa^b=)5Qj(IRa5(YNmVF#ItfO@Ua4pZ8b6isM#b$E~~# zGTOu@(r7v5HN4BI2qMyq+VR@yj~1&1M2{UteN|*bopQ#RxU#BtS=MxRbKj5Cd1j zv}vs;1uj7C6jgU? z;b98b_gb3EeoT`v3NoRj#O ze{~kp(7IjdQC$fDIs*cpOaorACIH=USib=W_ml!q(LdKtA;B0k^{~%;61^WzmSLc{ zynR!dF7PZLE4D}a`+EP%aKR;1bIwL@=Wx5%m?y!Vw_@phI;t93Jcv3}E&C1LY`$%l zFuJ2+J#QE41-da5$X@Ts;(KKcSF5-m3jI`ZU-F*l+nN>Q7wSB=pj=#otW3Ij4KlsB zJRrp-ZDZ7_{H$nJ@pt80AEO)Ky8ix2eHN`8Ub&Ol|0!|%jlZTpw|HZ#s{C)o$lgNmY{jKMI`CT-fL-*?^z@rH`Rs4i>7kxjc48btxDdOR8E&E>Y4 z6nZ?v|NY{^a)O2<+3hLuDY&Bb_je!$LEFHOzfC)x$@hqgR9oU@3}NP6F_^s=LA@U6 z8)M^rc5tIICyF$otU$|P{a{J66gbOqw9%$w>ONZrZ0v7J`;V1zmx2|0BY^MQ&vKu# z@W+SFVkonEsg3zUYUPzPA1hdz5^`UKuoxDsi5|;jzf@qBkGIOzpOg~oL3GyVlKp#C zcNPD8bn^3@P5eakhsU;cx39cwZBn5oO084U<^=&xf|VR+R>F}dArXWesZaT z#%>Yz1%%zLeuV&xxDObZ zX6gvQfuS^6X&>B~t>iDJsU3zO_m#m4XCv}VIT5rF)M(Q zND@L@kSD%qd?-aZQ%aTT2Y7n#PPfj}p7k-zY$e0Fe?sdlLo2%*a_uS+PEw&n-7zHa zzEKZnt)i#g2V+-%!Ie~x0yfP}*J7|ND_|QQ}Pi$+DayxBCnGbx! z&MVnr3plgQLS%-jobo-lyV~cCSpGunV@C^Kx0gNk$zxRK#b!?3=8|E_MkIf|Q=yoZdg*h-E09+|V1t$?fcUJz6+uk84LKl~2AhfTCj$ZAKwH!y zrc@-N+lYPg&= z{(aNbP9L^|$$p56^-45d%K161T+3C$`Hc37n|0oYtcCR3IzHOWm+rN7k`0?`>Zt(Y zVvFr2`mSIMc=qpsH)pKYu){|{zSgDJ7-hopmM$4rQCFit@LI6K?e`$HwVq;qdz8l` zpg|H`bM!V3Qs90eDeykvy!`$ix&21qH~vDQgll0$=*9?0!~_ZTXxCw5WSd}l1avkd z)q~}9{e7E|m+6w+7yO?< z&w8r*69(ZgWU?SM^UZbLvy@|(L}Ku^W#Z37b96V1UjY*Kws7^{g~GHIqsU2|q`(Xy z1**X5usA=Y4SK$9aB=7M`Xm5(B?Uqf0sm`4!2Rj}KbpLe6q~_={}W=ap;{Q%+YJ`h zbH~;eEPMMN!f5-YeR9M^h<;j`Pht|P2)}^>WyHYl{S5#4 zj~mIY76xZu0Jrg3-_PJ>{IH0{ex2|h-o7A%Ir{RdZL{;t1{ZHS=dx|_LH;Bl#Ah%k z&qQix&lPVoob^1vvVDFT$83p_!=Yasgh7i5L0PyjuP?YT7TrlERKScXuxF>Z#@(a| z;05#z!>e^F9a%Mi5BFe!^ClIHcl+3(i@3hdMBdA}Kdb_(}rlO`)JsM617iaZi%N^Kc(O09jq{9SCc4H|ut{IP z`+rey$p4LC5r<+tsF1i^HgEO=VLP0E5N3y&)%R9CG1Id7bVeKhG#=J1Frw1lM zS}~O7^?wV%Y}gEMK1?Y*x9v7Pu<|nU+f$Xw(42FT`DNmFJszul`RkW@pnW|K&=$_q z`xo)VB$!pyyZP#MceU7~?g|M$dXGAHnvF}A?BC1~^$a|iq1MJTY`)%GN(Tf{Yxdn2wpUk!PF78Rg<$Mx^b?_QUF?W znG_sMNX(lB#Ty-jemO%S#9E;dL1U-lY`O{2!xSVC?VlKiPXdagyHcNwj=*mTsfUgr z=)N1r&e+FmfLZo}j`;GF4$L z(%)6bEfJ|hrX6i{$g^$37xYlxtbLI9fNua7DHc-%l3naEc7*Q!UMl1G&6u<3&OqBX z7OoprgXgOkznMmUNW)D?V&$1!3>NkQ_f1F!kRA2zbeOhdA)V6r2V*w`_ns0D>&p`e z7weH26tFi3EJD9k1_qx_bG$uO0D8g4*$^Udp|N)NEP+ybBlENCY%Dmlj0^~3rTvp| zozQ!c-!tP^$YiDPiGbOobb>QyTrZdYv8(@*QJ2nY z`?Tu>v;~ri+5m<%rjQ>aN2w53ipWbOog&YkVrq91$=zI*Hfo=Y!+_~l)hxL%`RDaS zE3o-X$JPnF7q&-oZ+ck`UQnv#Ymdpn{y|%J+;dH>|2{9BGT{l`u3VC3i;oAISIY* zslggQo{~T9gM8dCjqFd$>tcs5KMU&(xrv~XV57*LY zvc#=-_W&~tWwA=}$;0$D-FS|Om*2+o{*x?`z{YClfiy;OTn;x z;{b}&`qj@HCP=!B&@bh^w>J;fFlvw+1)xjzPAhoozmJskD{ioYPouPN2JPmx6gB@9 zYU5n^G2czT?x#Wm$l3c{?#=!3gOn?J%M```{L=ec2bC><%zzZ_MT4ivpn~PScSf2- zHfd~i#0R9(&k`Zy#yn#KZ%m;yv0=c2``$cFSz;I^h=&4j7Wkv{Qr94?W)B5nbklZQ z>$Ki%<_$^GoccsL?h{3`8#DBCoT4Az)Vr%i?z^4t{JlmxBo35TC(Bja7Ncmc*M5%g zs$CNyuRO2W3qpwBc&C5}k?@W;V#&R`m^R@Q<)K*(Ief*r(2mDZ`Zu4=Z`^d>CT!1pAyRmtlv zfq_vf><>{9^Agk=(*JQtL-Oo{U>;6tK2WKmmxp{9smGicl|hlJaSpVU1B#u3hN6eb zmWPbhL45+Auj3ei8Q6{)JM4$w=k5XRU_U{rJmn{(*Ex{c^PMsfVw(cJdWdd^C_hA( zeBF8X=1OUFd)fGJO{HeK;K{gPgPUy)U9txWAImZrcD+B}H4hNO7++;fsujs_sXDEP z1=z3a@;ZC_GS-;3sab^^o+GI3Qmi8X(6Z{Xf#$z`=xBHztEd0+1Jg4jGO+gtS!5u4 z4JHXqp_IEem*%t^ok4q5nAu{d&G1gioGijDyB@~ydpy;emY(^2QI%_lUUAGtulrt~ z`6fi(xf*yxr`W@3{?z<}Zif~ydr_qrkl>0UfropqIo)4bZL5NNetyM*_D>i|gdY{w zutNZqkmxr!k*7?uxX0S8cyYhVfe$GWXkwLaBO6`)t>34KDEvP8+mhjhgr=Y<_fVEW z{+9;f@QaVk!)Fx^r`Dri=sFt%?v(jaf&Fuj*wyPg&2`u=1XO*3XqN@iHxzoRR%Lv_ zoL{&R1kT{KV4}8kD03KBawtB=Et(id4pQNq=NG##3N{9yx1CV=Nhs2T!%-5MNAWLv zswZRNJ)r15&o(~n;fsmDngHe`V|aU^c4xJ({;3{|8a3hpWM`Ir-*Q)h;=5mso$SgaY2*dn9>gpQ(VmY&z1BN!+{g1`{?H955KAcoODQ{7sGW) z=$_gBsG^|9uvl3=Ri6GeB|hLlVC~XNU0R?isJ3;v*k>wiQXTm089=+yGi7nlSQiVTexj$T=EugfzUM6MX&h;XzFX4C9&w1&j?#wTgZt%>Ga@;Tcy ztn~wqBTX0@dg!D%zZp$?#}@r&Y2SGqW%?>loySU6WM-8jgJnw`Ne1m^l+lj)HF7h( z*N-EX<3c%oqYkj3?CD6^2c59RQv9A++>Y0j>+Al?vrCg3LEVlu3V`?4+lN6F zgyecBvJIqkChUW@eHL*~@-y~9KO*TUrDs1S0ZnKBsUUQW-Asc6%VDI|(5C^Q6GD1e zJkM92>Kx-C^V8{@X}9OMtW{?pwplWlsA50dZh7&!Gmj7-`S+Kcm}9SW2zsk1~h#E?~p&riL9yOAELEqF5Z z)u=z%e;Z0RMvoBx6pN!yl`#hJOLLy)D~icM=g@~8*~FjhVr{3JKwo>baxF*Lk3yrWw6^HQfM?=USiI5pXmt)$OK;0#{o#;~|Bi}1 zdEg#1ypan{q(X(dVgvvAVSO)E9B7Z8t%>7UloUK>n2~QSs)u$GwpBlC;#><~I8OPF z4J7m>W-cSmbj6PMHs|`E7zz8ihyX($%Z#eCQ^hzBMi)o2Y`2pooomhBLxb747hp?s z!{H&GFNFeh%+Qm4d9G?^Fq}vF2Dg4v&98mEa(sP212|^*=ix2PQs{9+jNMR1!l!+m z&-hX=haYvEm-yw)#9&z0V`i7ATRYq(o_YL$9>A+bd<}+bSrqL3t*PJWQguND_{;F- zV7Pz%$V?nrRoV78m)*H#5>Qp?c~__#m}v$&;0CMDwFiH|Jl55%D!C zJN!?CGu_Im|8%9 z9P3&rWZ<(e%wFy+=^YLRuQk!N5R2VR-j5pt7)z!8Xy& z@CkJq&8g$#%V37-$w;HL6n(NyKUQ0@+G`!xH715Yk6!I4mIM+I>w+jc=Im`(p(p1w zXWk{{fo#ypin5!oYr{3CQr$`B?C93+zSkanf6yx3I(J^4%~Sz!X1*lo$xKozjQbdC zqLd6(&HGnE+Z)T8a*#f+-b5Rkg7ESIqs+A~j;GjjKGE#$dHwT|F8L^OG0aq*0~4OX z@`1o5zCD>!2{#0OVh!9_BbHNGh?bBkXd@}zqipre$>w$WhJux zg-q4BJEPR}{$m?(rXWtRx=H58)i2L(Z7+w{nRk*sTiYv%W=ytrwEUb0W2aO~=7;yN z7*Z`+G;hU2XWP6c6FQBsF zaP&|2z8A~euB&Lho5*{3{tsxi?vx2x{PAz>s_o#-#zNPf^c(TedK&eFe+l$zC=3Ed z)E>6P#$p`~Z%u+$^FZhWDP>&Ib!40p?p3IT#jqzCGF0zME#F|*l_yB*JVG3fH*{wm zWu1#4@vhkON{0P42>Nu=HkCiBPbaCYbjt4^)4v z6!<XYhh;nV)2#^ zCS_Nw_*mv9uQm!yOp*I^1?cwfF$u$)A`Bh!xshh#5L8W+QlDrWdA}|aRxj)1ql^7_ zKiVUXiC&--pS%M^xx;^R9>UE0ujoh^ig#YJV>I(Tmezj$Q#d#zj;@XF)5rYOJ<2<0 za=T!&rPK)|+2eo1zb;mB_03T+Pu^hFdg z01={Qv%t(BSxW^b7*l;t@*lB5_+@nJ`;S(n4Q-O$oH0yllx#DLU*@a4btq_ltgg<{v0rxY6&wBtYI9xxHI z2lweg7OpQ*e}7hk8A}gHNB#G%5FSywZxsr44S(5=mB_M6<7XuW99@XfKuNTVpV1Pc z;CSEPT*a%)>1zZMsLSjyVb;j6Q@P?5>T-%~x+n90;d$kz|M$QAXkz7GHrL?~=&Ny2 z(A%rjyMJMT=sYXqvxRhT)gv$sTPf2fD(FaUeRe;{oHW>Q;`ile>s< z8`=ZE?I5rq>Y%*Eu!BVLr>Wc_(iu&!&>A!bIAn~}4ZljqMtN|65Ge!w2T#i$Wn zHP;39I5ngD`l)bUUhIbhmZ~i_IR>$-98(da(xB|;IB}Qp=z!tdmOM~xbQrO4l<4X} zBOQ85OWvi^-iq-&GvihW4Av~R&V^37UigNN-$FWPHgtih$a{rInOU}vNWeQm-lMuN z0=>iRv+4pTEU(|}F}+qL+&n)Zf{-F^AL24arEL|7&-Ae!|M6`jOccw|FS1e~_A_rZ)}^F}cJtWFPSRLrI8bMbX$#L{ZZdC%$qYv$D6ZKwwNM0M)NV_VTKQ zR2|#Enkcf9w)~y+eoLt4mz~l3dsP*c9u-w2&K&e-VwFncv&h)a>T0u;FVU`NLWFSG z*v|87>UIta#5OrD0_ER_3gn4B{_4nPtEv_i7SfGCNfSL*cMsM*$O~=z^gg`JJmk9G z1{m5jt#@zzee%scjxwrn&zJ=|%^a6KKglfbFdp4n==d>^YMJ6WwZY(Tq2g7Mta4(m z-x=qk|7iIl=2e#d`;x-Xm+MOVp+d=_*)r@R!}M;!`Uc%tH|^|t-R^o2ze>aiHjS+u zHRtmUp2~{9lrB5aih#^j9;?IWlal(>N8B&ji$|f}yQD7K`Yk#dh1K6w-+8G>9PQ;& z*&9rYlEx(P>1|1dZTWhF6R{(lpZbjnH?^hN(9Va(qL&y2ez2a|Pxc!|1%VoHzseS=1PUeUqq}ORuodrmR!E4vd@8D^348tJFqZ(40_7hkXxm1axDWa#hjzi zwL5u^S?XnqQXVr!R5K{E0P||v1^>OqmrMv}_7Yv6D+L_f3{w0-;>ZZM<@LUW$H|nh@Qc7lM=W+ACw_(L?nzTZth-mxl-I_S8h}KmEQB>$K@D_fdA0mNg99qy zSF6O#qh>MQ!P_wseB}!XpMAFuz*c;t6)7)Kk`42n$qrf?kDS>JrhkkHyGLphpE#v- z6Dh!S6I6{fkzXszvv+(6xcxek=R9Iv)arSt(`+wV-%@rrwCGK-t_*3l6MmY#Qf~d| zdEN6Y_2+k!uiK@S^%F(h-c2&`%O~>$?}AMuY*EdVGLPC$@Q2rAVYqThU>%^v*CO^5 zR`G(X7epwcWK^MG@R$g+wdq`0^Dgr3w zbdOofw+W^Cstlz2p_!4cq#=7luY(LC=k%6?@g`)EyjD3znd*B7N*kHQF%( z)$^SR5sU`qUqL&_3YeAPdh2og4NGkrA#D#M@6+}Snvi*s_a#TAItJfCIs$6g_v%NAgL4zQ4UEK>-v2X78GRZ^h} z{}53BQNxYlJ0>BsDhv3G@Xr^W1B=5Cy)IXI;$UU$raICXxtUn)bKEHjo6VYJUK(z~E$*&UvSV={XcyQyNW`a+Hv2pk~^GaZ>+Yi^)^V!v!IHy$QbEow3Q|aR|?+ z`?X66=HkhC0Nrt(OZ9aKlnuU;YRj5^FWN^hyjFHXhBLjCxbF#q^;p5S@9=}K1qeHf z-gVkH@#=PvuCneZI1y<9j|+y)(PH9fDBWgD<-vwBlZ<(ScvwvDhqW2={*BI9OP7Qy zvj*ehIL!gif+7;^wk3j+4lI|r@nEm7p)UM%mTQ_bwcOC%-t~qjNN-*jz=J`kPv#V* zqwr5Z;L?Ys6eCDXnTasoAK}Vqq3Rf8NqGh0yj=lPNc&-T{XLj~bnFq>MqE5b!}s1L;+XVy8_ znQ0`mpH^Y>jRPmTCg3VIZ!J9GhC*2_@L2TwL&M%nW^rM~&d`~+suO}3h@IQl{(PSNm$+O*IjXRWV^ctT|2v)EtZc$$1hX z6}?1Pgi_f_`13T;UGE=CM?u61q8bb1mTUxFE+YjF@?7e;zSodxCEubeAcOM+GSkL) zv)4{xS|7ATZE9^G`bdM9BFp6d`xZ=GOtgFX_SXMhCw4c(H1xfH;o<1-CqD#sO5f?q zcqC)t@UUo*e=YOA;|bg|hT4y-Z3CTi>5&c;YQ6 z!gJHa$itHtX5_8h@U=wyT2T{7a3c(bWF6}|!Z9lcA@{7c@C^qpHle+53>gvP>C0Of z&m?t}5cpA?oo{qx0l(QCY^;VRDZ{0z-aZ>qBjvw%-zAcA^GLfqJ$7n)%j*7vE*5#H zT;M}z;iQCY6CVKJC%3+&T80Eg;n}J>VLcS^Z8Tb}wJVLyP^DceaW%!j$#vCe@wD<; z3{%&qTA7Nd-?`gK?$BehH{J4k1&JT7ljQ{5a$Z?b;Pl??!NWq)S1)Q=_^KZUFKzSH zca&&veP$hyV)FN8NfZt0yVnxF-S`s}rIYRPf<>Un?tUoLb8K1135pRMbvi>QJ5ts# z(ipr}9NHsPlHfQUK0h(7wRscy2vp`DAB+HmpgUXZqhPH*DJB+Q7hK{JMADu}+;~-< zc1TsSJ_)Dp_XiIDQ$(XlZbayH+HaM_!n(pCOqmxa{v;SU zB86h0H%IlR#v}xD;V`iiy8s}SR`&1- z0p`Ojphm$ar18W+Zm;W_ru^v)%vu&Jg$G4<{djChHvg=^z)~1rF@NMh-FaH$r^rk; zL?;E9m5hZWN5ZD=3uc)W3lz z1z;(e{_^&DhAO}xtyv>TtY24ajcGg@=Dz&QT|?SacMJ+vIC-2+7Ki(q#sA2{R!o1S zqOb~&K;|nKMWuL_qIh{j)9SeB@o3slec09+_L6BEePcPeca5>=Zki4Q6G+NrA?cQz z-f3(PTLvHclu9a(XqQ*40$cEBP`#&8qW3kfSf+H zOY=w9JL$hq&at5BKaT309gw<*u`m3;X`cc|SfWr?4w6St#c~QW8@=1n2o2iMCZicu z&^`P@4sr@5AxhCbq>%IYW$U*$DtLVRPpo72rF{RPf`feCU?@SE5^uFgQO->;FE7$A z((KA#C0Fyyoc%L-3GRuHP+19(PuyDtMX2z&XaL#IUr2u|%$865nRSM{vMZ(^M1=Z< z{Cba*ZlKONcmUZTHkx#i@|N_R_8zxeZ0)A%Mprzv#~bSs*}9A}7ELt_g90e)usJDsml33TdH0yay}9lFcx`<7Lw_h16qH zxtMJ8iqAV9MlI$JIu7-|1g|Og8anK_;~%YBLJi5!FYMN=`*61I#LRnuV0_j&__P`v zP%$NuHSl#l=HotM1WdmdwyZF}q((zEQX+47a(lPA7takI1cX!DN2B0CpW2Rk;%%e<2-sj-D#GMTRYqL*s;*^e&| z=cmKGWt|+dix#kxCamZVtMv}7O&L)I6*P$tL`&I!s8zlzs$Iy>a(J_ zcRqOxMsow0|6kUIsY~2X;Uvx6m#l(xjOKxp#_tuo3$7C12>2c2L@4BZPCvsUI($U` zO*1{r#Dq{rj;IRvMDG_TC6&ljL zm-J#^UI@Fei9!V*3W`m5JZIPMNB^A)!brmp)$u;qRQ;AU@x(#a%}MqnfE>qK3pW)H z?m@;%G(O=3gz>-h`M>!DAehsrgQ>MnA9aYa(57Wl=yZnrKzdP z?gcI3#y6_Hb8eG+SxV{z5g~jc#5b*sFCt$z-Htr9%WAlWr4GpouK z^U{2lEAEv6YLzrwl`RXc=6N7RZQ+REiuXVlPp9p0vf8^NMagN~#=#qA;JHyD{tORP zmn+t;OCi$9V_X>L-HnL%?W3nsv-gMYD^5~|X3yF9w2+@;23xL3>)}Y{o-wi5n{WDO zD1R)OVS9Bv4G*J*y(HqY?X%*>{bAH!b5OWxd$aC_T=w6!8IDu0rA0V%VmOx;FxzRS z2qV&%Gz~l5sQ8H{B}K|-6k8Lz(%76?D)<~$3M=}AOJj@>5nidGXtj6__7jOu?UF%V z$)8aKEaZN`+?-oPiPtP!f{N{&>BalBQ%MBEhO*4UBoSrw&Efz0&uwA(f4-E0osb`5{oB&%g_#sV zBotn!jf-opVS3iEZ~W@seBKpedHGf~5_@{xZ{ILNN*^`7y5Tjb;8S#c{3mG{*>vvR z>bTC}6$QLjo|+#ENfb8+7O_fZKOPeZzBozgemDF+w*EDViCwPYp6N%65SD9JMXR0x zGE%zRO||p^1^kYH9>oX%W+72p`5f2Pj-DMdB^F7B`OBr~kG`1&; zD3DzYHh+Rvdm+~BBdoB8lZ%nqE6-Q-#?2!A*X;8HH!)2lhHR5EQ8b*a+r4{=A5}aT zk1E)$|I_a@m~ENc$A5%mdY=}$ThR`0f`hnDD>I|(YOBDy7#OJJM^Rv6j#;}t8)8)& zc;>&}CKN`2biv`7xJ_>L_gLGBRl;^SH&MKOG>m?!p+jaHT$BkXQ5A%vbqc0)Ytchw zY}Xky12@VSqN-#TQB{{?J>Pb#@nf6Ics`iEeaM!&G*yMm0U7hn>?>JNuHj^dXS1Hh zQmGtDA%xSHh(BiMalw0_l#Sw;6jYpbKq0};AP}Oz1?MlH(Btj|&r1bs9~Gk6(pn#r zz_(6M<#<2Qdt#=3qg>PzmNX6nE80>$`z@ZTs`+^Dpb6p0n%}BL7$G6VjsLxzV8|-V zn813I8eUVEwn-_1r=rjEd4~<+YI+K#`PZ4DGfkQ=+Bg%Xn3}vk&ZEqz0;v}BcMeP^ z>;i|~ZE@ZCqDcbWnj}^=GX7geyY8bo@9(~#)p3A)!i;byx$@nEN8R`+#YVBQ5&Q!7 z@{#6NYwZhSHR5v!jC$V2f+^g4*BaHE1l0z7b*G!lAligzLIlkIY0`daj`TrZw&IE~ zv4m#OKloo>KAN>2>ZsQZ06%_U1zvILwKDRz?YfG4c-Zgx_LQuCI8U@ZQ-;4I&4YP_ z@y=%!j^)Xn`)S(RP)UA4UnJ)r{y<3#__wOrc(#8ifUGY&rzRqo@@SRlaYRQdD15w$ zqSo7n<>-YYej{dA+EnCbPPK))%7UuqoF8VmH(6-F5ITfQUGa02ZyVq~A*xdJGay@{ zuwRt3o#1jgao#VfMiAu;b~<4uCE%-~JVC7gX=J9|loPU8DjxH-@(j}Sn-@W4FQC_C zg)YWD@$vX)@4gM-8Zs{ZRZeN z8G>)u$c}c>%LMnX$o<9d8Vi>Yp#NmGX#srqd_2yZ63rcwQZ_B5WxozEGBu^7VOPz? ztKP|*CV8DJBAvJ!tR^t<>n6Lo^`Dt0^LOPtw9222XaC{}--oiNuAaab>2brkz#R!t zq4`l;aU*h`^ND$MYX%?Y1=Lk(&nv{GP2EB$f_Tw|9{(rAVAYY&=&xD8Bctw{NVJjg z`ETY1q>$#qO*Z>S%86M&=Oejqony1wcaudYCVTe0Oh!^0%Sbin@BtIEx({RiinZw+ zvYdy_Bv=ndH(q$Qmr>!v$tpKW`kcpG< zy#1ZwM)}u-#VpD|tHL1Yg7;1`X-a>OrA;ytRQ~D~`HOHfFdEXrggB(_>jIhK!Et-5 zF?To>jWf}q{Q@7Ae^1qHEVN-FAITgC_2jNU`u5aLB)s8#TZ0$mmRVZt`l`5VQ};Je zl(Wr_cf$}G3z$U08MRM~vdp)mZis>UyU$>MVyo89WnA3&rz~YO#CN%QQ}^JQx4qAp zjlL+z+y0XMhl;?48YoVOCz5o+VjeiTmXibV^@lF-)X|)GlzVy-WFd!DJdBAv40+s2 zLL43>`9Ig&?WWJW^0JLdd90=A(GiU17vUJ9Mn@d6#y?vLqahm}3?CNuN;jR>+$!q` z#rH)Nxl|k+$>D51poclBUBuvI+z$wWC9{xy-Lae=kM@9!uPHqWo?!BDL7DR+{PKq! zPK`=_yjfw8|Bf0R;Rh7Z$@#fWDd-IY6BSI0E47+#!@I$28+{dd1JR%%IKgR+^M9ni z2yYgY)?tBRy9ocj<{xK9kNBS^O>>Jr!XN9Qhj$ifN{);0SO1tMtiTxmm0RA=qUHqJ z8u5ipXn@kqy3@X1CqC-kf71&;m}I2z<2=2m*TYZ2#9TMo-oG`gmGI*p9dZ=dQ)x|q z`rMuF_Nxpv5&okie_&r(bFUv#gttT|)s}abyVMIy-_mei(zu?G9tz|g2;mq_hV~`g zHCy`S%2Wvmoy4dQA4l-5cw^r2qKzh8>kBYWwY_ zgt89esxur=id1Fm<<)un%f!Zx*}7xIzKFMud=!F#_MCJp|5|r!>0(1$?nIWyB!lI} zeNxm|t$uU6VAz5h(jV>?l$;uos}fMB#;9=<&lM?3z+&$p%@WiS4YD6M(|EBma|d0g z1rq%-gL=22gY$Q8DkkNa7klWsWyQA#i^MN^N6vB;*cg`AyP_NQ&-!ms-c%rdk8Fxu z@y9tNbw>~2lT6YmpQ&L9L=9nS3rTnv(Kw0t^5*qgT=DXa!ACR33BLay{@tE3FZ>}{ z&+&eu^}3Xso*B+RXttQhhnly7U+dC@M(Tl*crt7dh@D2@p#-xD-wqUfq!kVbRnR8} z0uw-%rZZvk!0DoYaR{ad2p6t_c#7Ho{l{VdXM#WYUjGCDIrIQ|fHd+3MA_wUwE_PsH>g{&44D z>P7BeZ^s$;pY|m|98Pk$QAgxsir#CTwp~8!@r1ionSXlskVEHfpH2hg(UbbWauXIFvkD|SJVE}Xs zV~R;}wufI%c_Y|v7pa}ajJp02Xj7<1SBH{Z);<`62miF8AW8JHJ)1m8NyI;VrjGSr z$&JakZw#IX&bBMK6+k${CkyfnCr=`}e|{3)vXf?i-hHTva}{naHB6T}xfVGWDk6RB zhf`~L%9#hgD3XIcw>dKAf!|zo!GnT^g`~@oJExcj`chTDfe$l7^WImlf%8DwrlbD{ zY;^D}C0P(Py&GIg6!HsYAUw`~-ffTLOY{HED%+^S znddVsJ;5>EJJ!)27R=+lDR^+cUdS&PE6z1JAe#6Q{q;!XgEaE|C{s=lq#V(65PB|< z6$PI}ki1@zo<___4~evMN3XutACGs*(m_a>ut*+tPEtR0_I~<-anh;tX~Jq5EQ1-N zaete287vm02Q~5fP-Z8w`HpwRRLq!1OYp%(3zO%R zka2F3PF;!b38cTZ?{J1eq3D}FTV1F!rEzP|gX9b?_anz%FeAKQpB@XXUEAk!ieSaN zGa7NC*7-C!-9n<^MCZ#d8{h?#7UofN%i>_PGt;{_k%&&|SCJ|?l(c;^qYu3oNd<8) zDgfM}QQ>(Q0I9%9s{RgR3b9X8N@XiT64h80F-)|DxT-aoixz|KA*z0HU!NRn%S&Jv z4e|u%NEUpfDKrakr+Kcc?tc@Kj~3w?Sv^$>#&0SUfpeZSY_To}HY`YI7cms~GxYbx#L*VZclf2Umk*O^F2GDrXCKiF7i=Uy}6t9f>Aqq5l|t z$Mkztf2oz&(8P4?+2@h$N}NxWs8vUy`{ol{vp_1G0(BT7!zf(5p@k_^YVIkv3N!KE z8+3zkw{R3jS%8r|89MOx%#aAUd@p}1Z5ffKJHNm}YJJhPIRGM!qQM})|>hk%XR8ilC$?4{0DF$inS?bNADEDyd#^7z_`ECUJR&7-KhtoUY#fy@d*@==g)+zs}S=?NvMVH3O~@QXJ4B7 z-l-r#z7U1Kcgd_0-(7-}E-mSoPK6cRV!*Z%?pGzhl0T7n9e=X2y3dLfCM@o9J_S@> z&y3Yk`ErXBHozM>Bs#yHWQS02>LeZhIAEHxO;epZxB1w zZN}fF1Py0V9`4qHh>kb<<)QR-K0DkxWsON6rrhZ~Ip}8ol6w)H`A7f_i7jAs?Nemm$N2IkW7(7kEUPoX}>hzN_oy9g@?Fwta_h5_4rlT*d+#XM}_F?+KFm&EbJL7b^J8ei(C$b_u}&U-{1k`Hd)7 zEM$J#!2v-ZD%nm&ud&j2!)?SsRi>e>BG<}LFl;-wKQDdr_%vP_VKLQa{ zCWQMV$JPnuCEbc$TcuV^;mrMmEp4a%HT5I2)t6Vi8VAnd)~x}q`Q<_FMIvwZ>U>~7 zP7j692+FWsezu*HJ-nbEgLJDRuD>J@Zd^SUM<3`XycKEbH9y zySWiYvv^!Zzo-$RP&u*M4lp^cwq1|nw{$Wy=!iy=BzO*(K|MgNFA!#4igP$b;0zj_ zM)xBi0 zD+GR2=92`XN%wMCn2ig;VJn0l+Jt90i@aJ&Gb35-Cs^l?Vyv^Kd%ospaSi^P$)_Km~+-=Ir8^S&7eIG^LrYqfb*1^$q#VNt8-OBwNRps>FHTUjhpgkoB3936J%F zqRlVy!(}ou64V<>k|m#Mp%bwKVSHi>BDBjFy4i7*1&plPZL437VarIGeLl#&Hlw9<;j?#aD z!z}Ud2c}iv_Lex~7qYsbeXxLLy;I4z0TVPAlFeGXcW4GBLmwMoR4`X-P z`2=E_T*jR=3J@&QdWb{a!s;rkR@ePhj{3FBjq15+c@v`Y*ijE@M{`ccqDmk=en!6XW6$&Cg*}y3_e^E$syrp)+sj@`-?=~K`ft4kh6&c{aWoy+fwo@nPf_9%*sqW7oM1%Ew>ITeg#SM|(k zlj#{v7)+;I^R0=GgdIqyv&`<0)-sc`RK%kEa+*|>pUg&sbeY|G07MCjjWYcrEUCqX zJ|cB}*B)_z70&e0Q-|yVK6pg4!Z{;P2}e^oV?``49~6+Y7#`+uYnoxgF<$g$I$eVh zO>}@RW|#}l+F@BLdnyIM0&(&{tB7?+115dqu}p$68{Mv;EbV82S`qEmpzH~68HdeG z);?3R!^X7Cpz!NNuK%Hn!_~iDlU@!xifTj?uS_s&J89cL%;dLB@(>YhBvch&c%B5z z$+kmsVG$*tJWbz^--aELMQ^Ax9~c?AT!+ztD1tuH}w zO{M?*C4BlwReLmrV3LE>Z5ONnTgASFpMsepWleW*Buz9uxPot({iYdkKA&f0K;0r|=j~vyZEkiu%U@oD1pDPX}e;AN+ zvg>@~yuE(7ll%%pN1#LpGutC>kLP+$O1O|H$AIfXxIz@|cto-Ao;!kr2|5577$$viG9OzpC&A9^WcZ;o80-}V zpKfwU-_NCD8+g&f>FgQ|z(aL1dy)EMYd+C;(RS1ih=td2GbU1ovg}oZIlt>n5&SGV zWgW;Y_(7A5q4nxz_E1>pVFTv@qolb*>cRLf;p`a5`l#|xvG8r}w^+Ftip#Jga@n&E z#iU=p+fm&w$jEtQKlV-rvv0sf79$aD>Fmj90!n_yG z7LLYu%}bf@EVhn+!!kIGqrrf|ZT}_nWhtjy_O0-o!Hz2t?~zy%Ryg1wj)npzFF=Y# z#a&Z_QXM|mPOkJgzJZ`!BjB38J%U_PHBJ%#F_S!oMD$-}qM4xd75I5tgYx)7c{B%{ zFrz_fq`#Ee3)QDVxh?d@X;_Y6)Iz|2_k}7TOC)r{_F!Ox$v4ZNTRqMmCI09Vifw5O zSVKB64}WD*k?8RwdYAM+Xm~HOA8@M3?Xulm4^an%!OP*d9gxqIY;QbJqv&&wS;F1$ zlCt_+V#m0My#1=__MDHHiTS%WvP>yYiE6Q9fuPESplLfDWF|VmdqXCX7_{!3YFvlq zrg^ve%OxXz)@f#R)N2*!qlZVSbS#Ov!w{U%rm{FxDg}~~!}r|hZb}YukMR!-UYwI1 z@AF`_bTNiisMl_;C)ZqX9M4AO*8CHDDB9Ps2;w%_ajvtXnq2&v$XsfMx=d@p z8}7a8EcX|Gj|Al-G6u}-g|*%y;lZu*gy;{2iqj?M;2DV%OpQ(lx_2M-fvef}p(f19 zd6#P4%=iE`;r(uHUCQa<>Vw>s?^txxV)5S1dWEeHsvJ(IaUv}anB=6V7$=NYH)eM! zAw=&OlyULB)T7#-giZ#+L!@SSo_fIve@F-@a!R-n<=@5y*X_|8RNo2`l7Ll-kITP| zxCmu!8jL_2@V|(_t#>Hs-Z1`hH{*j~3^2-!MW7QbI%ld(w%YyLh0%o@&=R;9CBz-JPqZ_R17VNL)Zr_}Jjd z$|I8o4AWeaxwA|9-)G>_LD6@`HxsVDHy>3*>8$XNx0HS?8L=kLQL>>#rWHFDeuaBQ zruN$IBMltzb47=o#^Iq}sn*eFh)l%7^YH+u_l-sO<4-t5hj^kyaz3B}^ox8^f$VvJ z9=;NmPddJ}?f+d}D+42B)}#d23c3C25D$Mc4-h8U>HPu!*gHZA>CKNl$lJ0dQV|{c z0;ICS|4a=-+rdhS-^Dt_fiK6hc$Vf1B+eJ@NVrA>@~oXk71L1`pa>n(cPIzAg~^6n z?0gkRW9&nJE72Me+KzgWOGlFr%E7KqrexGmJi6OCZ}{mNn@BH-<)pm3ixu<5}X5 zwiiqdu>d#vMg$#WSjJs({E^p@c!$#cKX+nV#`l1~Ej$KAthKol!hJKy*Hpq{t{*DK z#iuxd9I>5;6`_j0f&Q)v?!c=wxdG~#!P>R!O@QM7RcyNh4~Js?3xV* zy%V+XP8wmJH}V1rWsCsnpoNu4|o0_m27oDOSN$Ty_>901t^ zBI(1m65C)NHpy;n+X?#4q3e|{pY&gKegskv8i5QnHibn{$q{s8h)887>Q z!ho9yf$5HWG6h&!Rof@bBX+SDK-xvE41oW+YY9RrGzT zD2w3B#NvX!&`*vGDh?zejeblJ16bHUysWW+LlWcj`>66A6oS4C>iDUdvmR&iE*GHF z+dMHOxooE|=)ErXV1?Gk2(FOaw()`?@KmX{@7`bPG*Q>nUD(1HuM!FC)M{Tq*5nW!88oWO> zS&_OD2Rc>!)Wh_`k2Q3{(%Z_1w_%Z%RH?nZ3*@m)g{xn!!9kd}T%zs=kH$e*Dybri zt28rEt6KZC45OOaCUptNi2t_BSlOOya5_VTLo#0coFEyHodkSU!QIuBU)2bmVYelR5E z(QR=(IyrWN9muncptQ9NjIYQOdxylk@ds%#%bS<@^h`})?A%4%Gjs(if?D>{`{sx$ zR0Q>9!-*Kj#B@cHg8Z_jmH)3i!XRwOefyUHB%kJx>~#pPIL}wE;T?^0B5df=;Onb@?N|M!UJ84w zk0cx%V*bSD=SG@*qoJwKLe&z>vk8+Baba(3zs82hAofhl60*>+q0$9^e2&Pega?Nm zzGuK<5+ds~@TL0~ba9@rgcp(2SomqZ@~#Qgj2c1ol0JIaL3CXXqk|=$Q|+eaBT_<5 zCva&$FbzK%fjAV0QZq$jFItM6uz%`9YGpQp8&@h}giIq4ERUN&vZZ#?ENw~Wk9Dq) zmn~{x60&sHvzaA9EywBvbxL~O()*-A5KW6!ctIp4XR%^zhd^W)`Wa5HtRoP-iD$Av znNjc<%kIZeCt;dV&{jj6@*b{ zzO*(w#h@ELyF&h8GL%9tk90Io*7&_K*UO)840V`7PLL&^l@3rlvGZ~Xlf+?fbv1w4 zW;#WYlG06gPGwWX~tj1+&bHY9|dZa1PN`Ki#`=5$hA{bqEv@w>W#{wgEU$@>jF%zQ5k0B z@6T2@;&(?hcDv2<{e~;O?@xySuv+LU4D2LxKi(mnGN& z-{$w-d*6EhIW_g1?x%Zt&YbC~>VEsBCodetm@oETxp1P_pfEZZ^qT#C)o8Pxr%(L1 zqUsBhZY(Sw#MRXL@bYfqBWx;UcVX;}S2v-*e5>G{{>bLyWdfY>Bwh{NWl}ex7vkGG z|F^c5A_eroWKa*P?;W;4M1TEOfv@)80PC*=?^sQK_A8e)S|Gj&`)hdpW*5OU5|o>o z_%TR8xJv&Zi-%JJ?x_p1g45LF>%@G=zL#e{iTCAr_hXRu_hfXzmNIGB+F|z38wylt zQHOWOs(tsC4|D^to$QO6=xPja+??jJ@W4z|{dneJE<%r=`tnkg&fog^@R!BXmNRka z``EY87bZMr0p!SC#c6vZS$WQ77LdUcpsCtBKzW+E|3<)KULqjlJ;S8VNGa3;afYqHkyWso_sa2B!Z8xm7aV6$cc32$k9?y{^=5Z`*{CFnE1z; z*z@r}gg4Sbj>3*3SGldM8-uFx_(11r zIe%MsQ4BnCyhtG~x1;f1JK~h91K?bSKltEDB?KR-0z}YuKq0|x+~~=5DyrJwB5WNH zREFH&cFw)94X~G9x|TJ?nYG>n%RM0za%8Z_{Bl~cI~6rlbX7&Ug~S~=;E*q4#a}fH z$d>BRXywKBN*?M(ONSM{bdN^BCw3$71Y%-WgX#+--=|^V%0b9DhB*(K}r9WHY5zIFdIJ6fn!$HDM!W=pRZMx zTzGmW-cKSnHPMkA1QGy9K)1iwaeQAzILzlpG%U+YAgaSichnlZaGcUCqosYrR{*Xz zQ4xqH-{*`jhz{BKrHSLVQ1g`)c|fAP-rPe@(BDak%bljq08c8aFo(#$0%qOxM4>ol zjtBFBM$vFQ3Z#zcY<~Ar^99j@+9ZQ#jV$Ffb*T5qE{f!kGa95u5x4U^R4T$K70j7D zD?dAg*+&G~Y4&?;B|^N`IktF6Pj09eA(|yiMG*0w^>uC3#L9}gn@i>|4G2>O0GBG% zQ$<@O@&x2k1Ts_s4bsU%FePk{WwMBE05_!~so6ZCB#sxuoet#7jGYcT9$C8dqJmCN zw4KrLGr}i&BMsbmF0SqX`_s{6oF>FZpTF_O0dQQ=y$^U~>NsydV({dTdg{$VwEn1% zY8#Oc@n?@{*^DUJH_9vy-hgE!4~?wn8cuV_$GDvYJ6YB?(*M7Kp zrL*R;rL0-C>qRS60WiF|WtS1T8Q52`0?;Bc+k@$EX7Ju9TW$Je12azgp|{jp5s|Yv z-#+$IVjLQ@Vot>C*SuhdlKD{Oq=w&io-p)yQWL=c*GHQiM=6rKUvLkXp5XUh2A0eb zMyNS@Irh}dF%Mz%U-|J2S5oNz>ZFx6l&5*)XFhUuTr{xvsG*pi{hm|t;|Vpt14r2~ zs-k{KU3;gB>3T8E>qJ4gCi4Fjd=7!K6TN)G7;U>oRDb)D;3I-D8(@xQ{yd1T6}2hK zK~ih5;Z#<9nbD$TE+HSE#^>eBAKyGZwAeE8h`xm&7b0{F)^7$95O$mAL3#v z4(W2SPt}{EA5K2Hq{FqNmxd&d@BgWUE7d0o5@owFC_qI5s21Dd0jp37NZ9~NRKFc^g3du-pL^`=@?h{ z=$sX)6JRj#W1ifOF?8;r&MLiURHgf-D^Gw(hDdf4Ta> zAbTAr;k)wtF2b!8*EJ8BOoNZK-a`Co&B^4`I~Szy&ysz-0i}#g*v*hzZ?S~*n*IL}o%um!cHgk$4H$VPZ+mrpDD*nG4a3O)1jwr*>Vft=mOAEO3b zZt0HwI@rvO-Tzh34_)Pw(D&Kh&D{8)e*5}u#5YJVOe4(i4}Ak*GSGtqo9}}775x0Z zEPP}4sHYEg*?z6;ie7V`krvTw6ru0?8Lq&*?Sdh$CJ!H7iw8Q|eU2Wbw_CQ|f^>J0 zn#g=jsTsy%tM_xw(MP({;94plEo>3AVp(x8edIpEpCPgoHI8UdFuCf5C2EH@{LxZ* zaG})}%EZ95N1yjp9$K-bSu#Ot+>ULBJks0yfAxq7u8ff};4CnIO2W=t_I>m(Ao>JK z$5rl6#$tcuYMd=c7GCXIs*T=Y%LOdW_#Io?eRA>h1KTzCYttb+C_b(V;&xC@A|EAptj$db-rab8(#~F~6wJ z3NOa11S6F}DFjLN!NyAH((m*r34+P!u)$GfHfx<9GE^aK)2a~F-iNwZsGe}k-6ZL9 z)UoqwlnxN~pPmmNFENFyDVLiYYWBfQDO|UiTmtOtv_Q*UIIlWyGy&>rr+hgz&(}4= zAgOruJl%+9qTr4905-t?`T{(P3xevu-MeCWfRpbx?Dn z6!twt=h``Csu057+}`Sx5{2N4rledQBw#fh&Fk?2&z^+lnnGEyPD6#gyK*Cyp$uN) zhW@X6M{9)eLX2&R-^AwRU)RC{?a#zCWw>$&p$_6EICD9MkY?{P2~bKUW^W&OHbseQ-a~L5AD1)%M6Zt!KQBKA zYlOX^A5GYybc#aX0y^2nOKbZcGy)?Og>K?;#wYC>BhSaiS~&)RqOU(AG(tRKTNguK z$Vw=x|L@04hB#M3@jv5stEo`*|xDKn9tBX(W zeUt3l_@+`%f&ZYNFAbQRpyA^@eT^L(vOLLtH;?#>zxuh64 z$)~SU6kop>ncOwQ_zGu%aNZUvy4!cgeX9ZBwwe8G%c2grml>3Vk*Z5?FM=7g8TJ=+&J!V`Ln`;vtCK{;Cw8 zqkY<=N6{l_cb4P5FouB&#|_S~#c_xrsQ|!ybG->u|N6o35b`?RZ~p5DLlh_8@hHXH z)d%PN!Tql6^fmF4Fpx;G)%eeg9Z7L11sRq?Ma8mdxqjoeGnfpy6n~9{pLf|)xX-6t zpIw+BZdXjF&+VNSmVCwNvJq5j#r=CEJc?xpB46mk{NX;y0t?}`nD0@7(WKkEfUk-L zR34%9orq=%v!qWYaTgtkn-ILyJL4LPBThtTv(SA;F&c59_e1fn&g3GQ=6@!X5SuDn zXvhnb3-h3fYVQsYxMo!IZ?znnX$<{4|Ioc|#^T^*PoV(uUcKKnR1A^3B`jc6(~h(p zERk>_;y&6_%_O{`&m8zq2neufEXO)bhN$ zhv}(N|MR^Sg?lcfT8nx*9-g62~% z#^to-)THg7E9E=dCI%<+e&Goq8sas&6;E*Y`lLZU#f4&+6-s?44_6GN#A6V+OMXE< zcqzLNQLm!NZchOZQoW76AHuwNs5m`AU}eZ8J4T&g&j{IlG(dQ)0I7$!1Ng{d!~gKQ zXODf61yFugu|V5#arpT1{i|gzgD#egxk8#*#=q_vb4$x8(8iqXcO)u4t{Ai(RBf#AN{VUss1W?eLpNC62uevy zOOKlw9s;cJF`xkEH)P4s$w}sG<#&?K+<^8h-MxZa=~`sToq`z*>?W~2OUon>Yignp z-r0Z5N<_)WLUIn4T2~3054hqgwUZ%~(Ay3x4a!3_dj{ugyMPvdsB!2~eAj+5S*bji5t<|6uBgv|D1sZ-p~csm@2wmgMp=^qzrl^WT9P~qdmGv z{ELjZf&EyKW7%kJ@bY0hx47XACt{K<34(-l=BKRJX1`dxgBuK7WYTHUV)CBXd^K@D z5qbV_<9Oyt>3L9x%g11GdSjn`t@w1Rs=Sp`NHY zynk`>ZgpvYZT9)*M9|MY7cwVTPeHaV$YH{o_B3Oe5V2B1{tEnsWg7s4d?2Q6MF`ju z_i5z68*~ZCS8M5S{IIo%_2=4ETzk>6pt}D`mI~PU4t{lEY-0${0-`vJju}2S7+6o9 zUDJera;Eg6V*3MF%BEt$w*A|g+B0Uv`OoOZp4GG+8&V>85G`OsG-+KlqIQ~{LMTH? zTyN!5+q^MDT-rU33lBYN#y!rH0)(W#S*r$q8Wqx1b4401pb(HB5}vE2PxInrYLXa2 zW;wxLs3-PFlokQ&?m0N4cd8ERVMeda$zd;Ky3+h?V8gO|*akS23cSIPq5~oiwwb7q zA-R3P;t14S`4nmUeO{6X$fmT-@WRZtDdu*=wcZ(hkQxR{`WKAjDfPFJiT zYk2HNL#zYT?XRZ@sh$cepX9nuJ$@zep**E?dhA|41|j#?&SIrFko_8BTd@3%Q;G`V zyG*+NxWZmo+ga^z8k!Ua3hevq;eT5!mtNqhD1+rtSRlaV&8qcYOF4n-hV31^O^WgJ z^V(L78x^qkHchWq-Q?jgXJp{X$6WO5J#vu!tgl*Lv>MRvz8y3W^kGl34|oz(@Z6zM zQ9O$^+R^MPbjW6!K->(Cip%qAoDS*@Y=TYJsje<1eb>{C8)XtaQ)^g+~zTY zzkgH$4FFyvVEk1@|0v)ov8w6p*k|Vpm-upjVw#7x_Q}gy$omE+epifA)7*xwNok)3 z_hB`zx%o@TapC*HzyLp4G)aw<2#mES{plxEQdq26Nl3T~1A5YSL2DN=RxKkp^^rfO z9qECBNhI3K)!!Axsd1icI}q$Q`c~!>`Y=ky2f9%^ZU0BKA4kjsnh5Sf;r~fCddgD? zJ~HW*@w}5ZRd90Gztl#gMXz+K83VE-BU@HP#&{d$r5_`v7gQ)==Tq>vf9rvUV2`{a z;D{K(<2k3g8n&8p!stM+jw<8~FLeIPt%Bp3h4x=g@-KJ$I)!Y9*J1AGSSOv6%Al#& z3|C{wC2Gj4)zY{;M7_1;_(ONsorVL2hfyTmF>E#$aU}f5&#f@2PbNP#QL`RkUJlaF zkapNq%R6BdUf4|yd;heh!xr*12YHFCN{#DQk;~YBND7G{3Y6j?j=VGb4E|HHh-LO8 zzBjIQ)-m3K7l1sPWdzeD1uTW!5@F7VTs}`C?VO{CVu=-ztGx=O2FV^#zDQcab1*0O zq82VJCvrcu3#CQ3O{^T_du%7Z9p} zX;ngQc{eDDi@Qx9tnnTypR^m&50w#B9_`mfkPR^B1>gz@y|9;%dnm~xJ-J|d=d6es z^d$ir*`K!gX6X7!5lZSgi6cAtQ?Qw11H3e1xnvr5xuDd6xJMCR%E<#3rzb8YjD0fx zh@tP4Bf=8(*-`k={A+~CLeD5zb#iC;iw=#BRqXBz1>kiAI@v9c1*`;9UV!^J!~&rw z7Jheb)tQOXkj`c!@%;zr&a)E1{s9K>!idwZw;K!uPMl7%rOd|~yp%>a#QtNNqb zo)QrK@4pKE*%ERe@F1v?ytQOKhzN+@M}zMTpV4r)r&6wVyl?eS4n4j0<8gctf$MEBjzeW&ZIYC%C|oECMTC z({N+=ed5ETQ)%AD_Vanp`-4R+>3|EtWb-)b6{Q!z_juz*rpJ}aR)SvDkr^ID7nBgq zA0VB1zvXMSMfZ^3&_6H9(ZmepoD;CY?lm95s|uF+=0$Qe4Z|R0j_qi_L=i~Mwtu+v zig20DSfdWCt(b|0=^5z%D3EyzXJU-(baQ(bQc1LLbogo_0BjxM(bGVz2;gVOp(mRl z6_%!{2v-e|am|GYr>+sAtB}!N>>Yj-jfdfao*pE~XZ^DV$6G{hUgBIKA3u1Ng;@RT zFmu8mln)CT_3>V^LULukQb3+LSUjh3LY7P(=;x}c+zS1_%@JSZfPM*Cy+D%C0GsEx zru&WS&jzQh_suOdadBaL%f?Jw#j`cop@H@Wv~<*fYW1&07^D(hf57LV@S>rCxIPqx zp+<1UOMM6^nGQb4a~S1>vBz7TR?I)S)?B8$FArb-_)b6?xU47!^Dk?yL~M&(kc5f~ zj88D~kGp4>c_Mc){!(E7Hqog>!>xe$(bD_tvu|Rv>-1e<;@)4mN(rriA{94yy^Js= zth7o*WI9P`H-IYjW?EJhylc}~9T=E)X@{2;{$+v(Nhjz-SMjL@+mtV^b6iJLpSm?h z&)h@2@A+BauX`Eco%`bKbHDd@StMR8Q=J(j9o@5M&#*oE z_JXl!!I&^{_R`ta?^kQ*VndW#&8S`kSy*PaXXneQA%D`jRLiKEti7XE*f2K2YXsZ! zy?IS&RNvR9FgUV%+MZQ!U_vQoY}$T|5l>p3kNeV6`2z1C;EH*C>yUPZ8qVDr$VC&Ts0I)fKdXA7J6IZ_}N5G&F zMqm*D+{enM>7C%-qUc#oeV|}5@vmOp(mneu4mlzpmcT`Z-}{A)4**B2SF#}MJ`zPf zB6JIL+tLTBiKn8hdJnH5-7ZeVb@KycGgh6D>$WI38sZxR3%TAova1JgWXCDZ=4c55 zXo6o^g_8n8_W$FW<~w@dZT)E*8nsoVgdhF4Z$2TTCYkqqA$6s4>FRjefbX>w4k$Kl zR4A*#+KA51`u!WIBOlczpX%yuRErS&9mH;uydw$E2XJ9{#1m>4?n{d{-AhWo%dx5# zYuiH&zxp~UNYSIic1^%xmP=dopJPg#$$14Tf~05Vm0n9wjTdy!!9$Pnbsw2qwY9Ig z41*i`$fYRo7O@Yo-B^JB$Mu;0@w^49mP<0_b9z zvT~>$<)8TWMRIL$9&#exDgFvxr))}8ea;Pj22D^fz$AR_kwCy1q7UbP`FQ(_F(VW(!iv{Q}DV-lhv{cMQ*6p%>w@|W_+`$-BCrjJV;n`RFc z`R#Nsv5uUd*nV-;@Uhb(K*;tdV7_vTh*g`3wrk`CJ@gyy3?=IIG#QN6dQiw1Yy=i$ z_q&&e8;s@x#iwS(z|G8M5vuo+K2`C$9O6^zfX$IMm(smxv8wp272^wn+t~F!aDgRP zd%SKD#RTr65a*jge9~#=Eg$<0UnZLl=T1M;xZoyEi+?6YZ2GAc&5bX%6uzR$?L!4l zt0;n4+Lpwl)fgo~dlfg302M_3=Q_Gdow*VLM&KhQ<_N`Pvsp`$|; z91y%XjRk4*5D?Qrk1()0KI=w#AqfJHAV2Xh$n$7cV(PB~oY|mwiqE>E&|5LSXY%W^ zFRAPC5p5Fh(_ghEjwQWi_V;|8@TyB3_oBV*6Ea_nP6Q8De~NyI1aZPwo@TbjwGr8O zA{W-1l|vRBxOzUZ)zzE*Wiq53`71X^b$C#45#jV5QOt03-9!IgXH6Z(oz!*mK#Cy$ zD{OJSnbDkweD3-Cz>aug-Q&qj zIVLv;bb*O;jLSW~K$S5AEO0%i-}BPjTZb(8R?3kV#pL4)@V&)u87+|e{zF~adE1*gsGkyZ%1b{#bD-mn14)BFy6A@G8(ZJlj09@XynT66q%d(wl!0ha8?suWe2- zxs$;65_a%`isla$wjZkd#0M?V-*TrhoM#gny`ft~Fh1KkQ7vr6nCyd130qyuB$8`y z%vXA3Yo<($gDrMh96&fI9IGSjzoNCttQKIr3^23<};MSPL8*^i)U+krs(BEK=N z!OpIcByV_F9n?j|Pd>s|8?=?Ukj*)g$ZU93SBh8A0@lpKlnt5E(e3S+OM1W*!EY>2 zP|eTXLtyV<3&VXipO@YweNsS|uu6dF*5@GBCOrIr=xTSN-q-kCn4v}QRvtyhOS)g) z_j_9NbP84E^M{Ae^E-p_^h9R*oz2rVN8}_ZvG5$Zu7-Zv9u!I_p9o=93dtvCX2vL; zoZofGK6>l)z~$Ab6;?RKV-^*IAz&0`a*V!3?*t;m8S)~@Pl857w@BSveVRJJNhdUF zn!;ji1nZE|37T-2HHoh-BCN`L82J++dl+Pza4KcH6f`h@{M8=!<}D2*=-U>Fg@O14 z`C32Mwsbl0SgG2E6}PsDF$wC(F&SsEY7zoF!Oui!19{H{J|uQK z$_7sP(Ysn8si$cX_aQffn9~F!+`1Z7c>nO!(n@6O7w&^#39l`t)Kt&|+liR)H$##q zm>*^}Z-g)w@0y2Zgp7_{F3jer@XE+fa6>uCB>|je(s1;2SIO|#UKx8t*m8?&-9muF z540H^{0fog97wOe;ROVUj~W=kynzXs{YCkGz^>nQbt)wvW5sF7pUUc$#BvSJ4NPx> zh~(H;!|Y(n!yZF6j>9oK<9^H+FQGybx58^Ai~(Ms#SG@fBZ)9u6RS+RG+1gKOnNv4 zbq}n_gWlkSdcR3Z?;M&27%D#{u74Y{g_}+Ib*9eLN5Z~=0$aRcTtB;WA@~|Yvo))G zE(xPS%9vVEllS|Mzx>k`u9pI^z8f8g1zteTip(oSC81`=eaOC)<+tj^lI`bbMUSkO z47+jNtn=G^ml#<$oz(Y(eI{;Vp7aH=A{)^=q~eqG;mn=bpMfsdQF`zhr*+ zU=T^}@!IXT@gfh3`XeWh*BtTW+DGy++y_!u`M>rLEgU7}WQWkRu%E-Gjdd5n;XJ

    *3)}nJao&xp=B7jM+R&o;g=abW=@%6jeKn#An z5JSe#mQ9xI@X%EKTq%k@zxV1>sgE#uBq#v?tLn_V8~ zbW`- zITLQ7w%X_*ogLmh?KTVp4f-p*A3aS+%HU^*r;j!b*XFqsVn9uP!?x1aF4{1xi}rJ? zv3Iz5qWjm=$}n=O7B9|PAOz8C!q4b4UAa)cfYo3B2-P_iw31?VmkdxFv>DZ01E?!O zX+>?TNOw;`d@uLndo0kf2h$r*hr5*vnJ&4w-Q7Qg1NmzHET=_6&E$~2dsKb&)nsy- zE~9X0bSRkw>XgolxMwS9W^z(Y67{}UUs;3zP;ML%MLq2T+>2K@_V@rG5C%u6^0c)wk9!uTuBhkI_?{#?eELdU9dvW#?$kh# zzW#Kqz*aO1xpBqMwoNQJeLaiLIn~zGIQs!{l|v?^PHp9h=qh;nhI)FB=ue!&9nt;+ z8&|)Xfm;1SaiGB~j-pmM!bj4g5p9g5hF2^ZX#BRHSd0`uAJ*e)o zB24)aR`|I!Zq2{3*@S{7PwVfAJ;I=Ty)n%t5s>Q*LSFH3XgmX8+()Bz0Co>+^nzpg z9_BhaR2$D_l!#~7_sftK2}R#kKE1sNL79<3C80RBUf1zZWD0E`{R&k7{hh=ttn+d` z+Cm4e{eDz|XGDx!ID*JcMFTkT*anEKNWP;J2<4EHG>{6#>}-07T?so}4rE+;Dzn1P zUYJ3-7%M7Bwb}bs9z`q1x2d>bAcz)=_(Hq|IeIhgSMfP~u^+nzY#^Tj$e|pv!xkg< zYPH99E>8^eWgJb{a;Q=BgM*+}st!7tO|PxF_(DQ#(#PJsJ+1fMN038_(!&f8wtFj& zb^Qx{@RTMJVAL~g+@JSq|H4W6Jlsr1*ifvamKQM?ce8g+ zjhFlS_q>Q?3t=eejZ8;)&V{}u|0&bqOIl%U*vNH_I{rvGNA4-_l7K!GEP)FA@9Gnu z*@BaMdfSK8X!xztLyulPc?ciw)|#f&bZ89NH{UvO<}N`N;%x^=T=G< z3VN|QLD^DMEP*l%i9ir=0HVKm7->>Jm-ZD14mWRdmjGD-!A?4nTDGA&}%my%Z);IqB9JioI=`vLZT#pBjo%KE+xA zklclJ@gm-<8FB+ed|O?#Sz9EQ&chhkY?flbSUxNz}Yn^S2Ypv49>12B{7 zEfKuj?hKzw;t3x;UX45D3VX1QIM=;G4TMtb4#jq(I9jyv@Gx|%TG}VjWKcN@V(Eki zkVOAf)(Z08EOt_Tc3=8{C0CYm!iWXyYR=xAzhIegqU3b%h`=I3e=^afp&Sv1l@*bTDLBn}Da_Cfu< z`jb~H)dd}azT#7%iJlZJp$~??P1a44a7zJb`)i8t6uNAEOk!e`7Cz7O=TZ6AM$(En z)Tb0=*@7@h58cBb8dpDH9Q7f)LaDzf>Gt6g24>Lhzgqmm2Q@R{I%h^8f;IkMyuD>m zTu<;W90I}J-5r7if(90MO_1R3?#_k~AUMGZvS8EXhekI}5W(S`o=N?We8r|9;e70P9TW>7lO{*or(jG;# zF9ncPE+r8Qd0=ujtdQ3B6Wdn&6D~JEroQM3ncM`PALUo+0{Hkbu98?}Z$3POeqzOs zgC+McNfa)`A3cr*uxt1)m&u6-pL)KIR1g=eU-%i0JMu$TJUF_hf%vfowZ4TPe|Nn+ zlmh4ln*M?fOML$kO>tv$>X&4soLw0Vk>Y_l+5kU6*@1})6x=c>_H5_h1ocPBAJ34V zghq77W>!suYmvrFfLF(Rs%47f^SSo1pgwdQ%lNgNAjYH6DkbXfdZH&Yr}CxZfDY-I z_D^ZN(KjgYIfJAW0(irx{{U|#F6bc>;E8cdeDigZE79!8+dDKmqn8uCceXfgWP1rs zziKZ004v)s7L;6oG{v%AK!sn*z3Qi5hm3%wtbiPzmHFoNI)D;1c{|XfecTO1Z!Wk1 zi!Pr1CvE4LzG^xNn>npSgi1QeDOu}K@;&ZD*OQVJ{uDsKq+E+Uu;qwZhmds=TtpSp zN_E$eiq|6Bz+Dljmypzx>PU47;&hUq=)OCmtYfj2E8vD1TuzRg$%ZulsX!s-cj8`( zoKwKMN?i)%u)UX(QbSP;Cp5#y&u3g8uz@B0H{ytU3f?3)4VDoasM8w zr%MU{a1Z_LNxYZ9&&wtFViaKOxMMQIn3uM@d9ulG!6(m^L)%Mu{gR5&8XeKz3kiO*4-kYkupaj*zyujo5e|clE6~&7my{(O$ z!aI_AmO6OPl65`4z3Kb_ZpPnSM1?mLn^tPT=j@|;@CK2slv-06VUm>J{;26z#7j^Y z=dI6{q?hb`+Z_NIwwt^$%POxCATar-A)|LN)`h+cNWvuWkN)R;Qe)z-Xvl#N;J+7m z^)rq@OWtEzF9W8GT~<{@TlkHk7ITw4@GG^+J`gk*2+maSUf<0-9}iS-cp>5~fT$aR z?`@nxJLVIUXkL;I-VGN@;j{tXc%vyQFS?hoGbl0fTg`>)$P7D#O=7Uv{Ep3_EmVS| z$TMwIYaJ5fPZxSZr(UF%A35u7g>TQLoZD$_Z-YRtG=^GGJpTPOT`qm3R-9Hi>K=HoGv93L1wpMhGP5oTfU-szw zkNR>US3`g2wQ9|gixtlR+u>Q)WlZZv4?-ic>NVa#Ky$mKofko9D1Jo!cNhbKBaBlc zzEf@#>u(Hs`ysSS&0q!Anv&1O`I3(3@tGQ2Lt5z%Vvm``Nj*+~BAlBm6JT699f-nL zI*&wdI7er`>=64~$!TP`#_EhMZ3zooGcmjZ0#HY$?hYCkI_W4l6^*tnN?Xxw_cF+~ zrCjQT7~biOG$-(oN&K4sInMseH4N30!ke1#O z!K1VBp4R314I9S>2!y30zEeZ0p1c8~%)k3%^~9H}g#RkK^_$MSseNI;dYgY-|memqmmt2m39lsth)tQOIX`kbd0 zUGnp$XFPpE`uu-?YOBV2ulAAU|&gBV?UOJ1kk!PFKw^Df4*C$=2Eq>HG5sY%=_=X8%jm zgJOb#^=%MQDOY^IH5ZZfE&n0At?1(h6t{svP%Gmn7{s(UsaAOARqBi1k1@x5b< zXBx!=Cw$bc^hPwk(Y>j~ywRTJ-wbZnv?CTOd#j@5-%J*~4ltw|Bw#lF&sII;gN;xd zUB=JU|D^k>sxpZ?#-;iVtaJ`LBe2lbCBGNFX~}Q2-O_jO^6x(iA`$f}w*!j*N<`{5-O-aAa}2Sx&4=Eas($#)3IjFm}*pt1-Sl*#dEO;uv)y)D-$BLyWNO z_Z!zk+qBObAQJTUG5_5{Hg@G+)PG#dsX8MppiaGKO!~=3T2`>W9&A0A9h2f3NYb8G zHX54!)PiVJub=7P8|y*~3{*Q>g_fO)y(qb}X0?YBP#RrUMu53(jU?rdt~D9i+7dSB zFYt)#4A&lA?$C}+;=Wwtow58G`RXCASm77yon3YGGkx364!M+u-g_hf-{u)RqPiO% z+|t_3d|CX+S3I+k&YGHt{%-uT#6(r3YjYL=CN;^gYgZQL$@2}y5Vsq$Y%W%@(IA`O zYElj9E==|U#@(UFAQ$vBe}R2_eR=)*u(nz0NsCSgGb8m$S1=njgUfL zCI%^uEchJ=C-5un(mim(OR0_8-S11RZyN5jim=Cw^&)7p$2G-)#L1<0R&JOfxx7Psn&G@L(Pp7z@9CM$oV~0mX&;h@}@FJlNgH z|A#*A@(gI$W5FEDVe{X>91A+{_f$i%^RoZ*_G`TuWcJUMNxXo z;*xd#%N%y8>3UUINLx>I234A;n@?9PK4tk&Lo`L&bjGiY>ZA`o&JOX&#k_jlS0UZB zvz1CO`qBqF|2{xzMDQ=JEk(8hbBxL_*P@g&zNr=K-QC_ytazk%aMQMT3icSl76Tt< z^amgEA_Y}MUb$zDQKkDBR*BY6;2(AX!{ZT3_j%K$q06km*Bz=8{nsafHv=1M`=%d?7k zn>Wj|0K8}98&;b5Gm>}qpuN~s`Zx9L>U&Rz_V_fRGlc;tWn>5-`Rn{HUK*2sw>@<5 zXs@OCh0CbYGx2kpkdqSYkKh&-0**c6geq{{H@$2rWWzDptAM_v^3T>-PGiNg*F83X z_I4QA9)A3{iHC;tZF;9}1gp__V9BEeril2rJjbw4W|p>3@^&7w5tn3fHgFp}>DsXD zejv}73VuRZBaC2%qH7Z00s8iXtA8Ko@D(DcSQ`tuNWJ%9NhZ%{i^+K<|3f|7_b>doD1NNs+c2uVfhU;l!abwH4b_egpDMEmwnG>2IgO z;EwR){zGLjCMb}5y@Xl ztoZDqs{<#Ozwy-g1Kqo4MhjICmW>@Tez3y+DB8fb!$O%ZE&RF=C6PKwzN_~cLeUxQ z={1`38eiv~on$29QZii$*_3sHbyW=H?@2z^zyC=?!L$y^NYp2y4%|XuDqoa=eiHj{ zP(Gxj55A478oT3QOy$mhX7?pRhl^7o0{o5Im+@<{j8R9}Nk*P*Ll{?WZqE zzhYDU>Ezm7B%sWMTpHwy>nSguA(#1;zMW<~F?z~PZjBwKOr&KWKSNI7u%;1R1xlQ0^N-IJ>y2^kLFLV3 z604BHo>aeAdmz8T@gZno4{rm6&pQ&1gFW%;kxD}M9XaH!J)jmEjTDB%Kf<5{BLBsu zt$0elu(CsjODny456h||laGB53ovGy5K=LKR2vO4<{&2++B#yQ7Lm9OkjPNF*GQxO z`QEYcxtIgaR&@iC4A=%yJO9|G+F%a{oj(z$6$OBc?0&1Pk}HKV%`7cvePdc{-!L1U z=?Y)@EysOSd>?>Y`TlTiIp$rAzCBqti9^LXEwG_Q@5|=x=n=#}Q_+UEI-swAzYMXB`t8z{5DES;I zA;MtFF>8_WBwG8J!&n`jT@x z-d6v(d$B6iQs)%|?F1PAhU4p4jrjAi%?3C%+08XxB%2*fFt+{GI^^AD`PRDaS2}7v zzRWxUNsa6YHjKxbxkLNZ%&Mt4QB3=oVP}GOTO4*bv?}cb^}zF#OXoV>bZnsJqOpyS zNzZCIFdBtz8ePAMghAIbM)EEQWh3Br08SRASo42wM%Rtb^SB~JLuh3OMi^2ca*Bmt zE*=f5`Lt5KyM(wEr0Rch&Fk*$D35B3vs3;ft@dTV4tWt`2ZiY5NX@fdRwvGy*@N^4 zu3p%KL<6J>=8hcVXB+=zgbC;-$kx|05PP+Ki2|G)sz2K8l!*;)YQicwud`TK*~g@()qE>6F@Ng1_ZW$L{2L0{g%bEc~@dIn1Y_ z6V+=$xE~>_uyO zrk?s}>sJi1hj#8{mFyObJD8Ap1+P%09=bja*V@QsqWt4 z0j!3-f7Dwltd+^^2;wPZhc$(|bLE=Nh#czR(YHarXx!60 zS(e@K3$L~tD*rD$h;t(Ux^#&7SiLJBq)@dF$u*}znj_uE*KrPJkN-BPyr8&)sn-5t z^dOsZ4ENEQNtyGuHhw(x)8<3pktc-(W+q(>IgYF#P|YOoZSzy}l`NL(9T8E-7e zq+UvT6SOakzXn`G>#g%t?%$v0-=14c(IE$fSuuFWi`OLjv}-H?KUGm`om4q&L!*){ zSu9(bc7g?emk8tXb7as!@0(3adM|wmD{m|kb?oof-M->Eq?8Bh^M=f1fyJ)pZkj+^Ve;fcIcU`TGy z%M{)z3crUO6D}rve6e=aKa2NCpU~JVAwNi%MlGZn)GOZS-7dqjU&Tg}QQka59?=?^ z@J~gFeFqhdvzmXT1hRNGz2nx{WT>P(u_r#IC*~a1--3jnFpX+`Ld*1+a#SY5aQS$= zgN>EG_BlBvk2meEG>M)m!gB5V;QjA0bbv>e`tDm4VAriT6~KMh*qI_aldBrrW6;!8 znzIJ)!zIg@SkLajl*UY~o~NR9z+W%0o*3~DQnoqULsS^!bmmtmX|;dpO2|EbN9WI- zn9}pG)v76{vMze9-Wr3pnf&1L-(z9^{*=;#GTd^-$XXl9!Y9SedInF1)5vf6Iw*Q1 zr^t|Y9K~NUU>G?LKEqAoU*vqjByI>@Sdo8zj#bl6T`o6sJrXhGv>rivnwVyy!q}ye zJwR|THN==bpq!^#JN-|Nk{0SE8XG`>l@9trUH*5x1w~$Eo#3ej~e;((Ye7dm)T@6(hvNj+iN!$WoUn ze8uQet3XBOdE@H~eRq2=jPkl2O6~$!ES@f-l9`XT^&+>Z9a9bxN#G@|%i(xQ7)A3s zkz2b``uVsnD4Xl7;ef3a*@D?LxEV@G!Gtv$RQldFFMZUSn^pN1z&zwe3jx?aF7whj zSvlG~Sg;NOPE$t8_F*Uf^U?3UbxHuKnNa01p`pq}Pq*7||4UvbeEQu}8T8>SZ3HpE zy1|?1_eMcwof=&(w@$44n#zuEs{0302A(EC7=QVr%vWWKW{C{(W$?bMo}do|llP`i zdnVH)W-;7^amT?5ng=}b_pc$NRH<{f-?3@*26}bWu84^Pw>q+V&t;oBQU~v+qWnfX zAY`d&pk)q3U2;F4msSIVt*7fWdN>jJUb#qzToM}ZRJ;b`Tc(If(}CVoy|_mFQLI8S z_tsi5iv&`eLKd6K`IL-8747YwM-reqsetso?W7iJMdGjgv*{d?LA5 z15&C$qJ+K9(oYBPT{HCJ?xDfa)1p3@97iwMSj$;Jt0gontT7Cyom#-+__P?x!diQ2 zv+ZhGG$l!xI+1+qLrGiyx9J@z`fyDsd4V`GNNWboRV^ei8}{|Tgw#zWQA0s4DC^{FEw(2kT|f#*`sL@ zgbSE_n)bGY`7$aOxQN79&*G<@q_2pG7s3QVt3?cOAgWvMKXD=>=l7zTwR?0`|($7B$+2!bbctukSlK z@6_49m1O=P+$5p9b91gVvB0Q!9lzINt9B(=&Ytj}`Kc#lUsATUxJ`_r(Ec(@obctT z$f-=7S}1Pgj&fCGb7C0a+r#^lZ-wArQ%gRH1qx>V-t15b4UiU76Z6{liX=>ZcV0u` zWpWMY_@E&z_6kYm-rcA9<12DNqeMHO0EEQ_gHt=n1|<&2Mt*q{-9 zF(BI|Yvq2&;mjtN$}h|;1W26!z~L&EBT|T1E5y6UBpvNEAvspod+j0JSIAa>caRP1 zYvjGuUmnE$zL`m#1U9ne<<#L)A7SdU<6B{IEoRoBb)Ie4&neqXTe0>`tgX+sv z{#I$cLmZx;k~#4W^v+YVj=~*e+Zp=8HXsFQ=X28ovXOBpa+N7P)Cwk4zg$0hT1#wY zZeps2M|r8c0BZ6FLn%8x@z24}sxPPHM|*EyFhv|kUu-=+;`e~yfecB5_d6i{ece7R zKB<1D+3du@)(ltOP|ww&%{NH|S}LP^LzySHgE-{e09c*F0`XopcH&-}`Fr_SF-HM} zRU_KQr=3+E1wt+12gm0**~s*l@Bf}-?2LW5Bh}K=-zJ@hm%q7skBubd z*eo~P;%CZRaAD#292t`dbHdpL9Y%Aqrimj3wXNz)<({K*Ww;Q^J?HryD~ib!@gMGH zS}+Iso61W)Y!^5JyN~{)ukAf88G*#MgtMF~V{wRF2;~t0;)@MJH|nTNzE>0e^xTlz zD9_`oB||ble4u~w#G(?nKf)N}r0IfMekTNhU4@PJo@8IJ3l&Icvs5 zgdABQ?)b}1D{fYulaJ(D0!RC{MVwezM{7~qIu5UqIs44`B06L-=h}{H$NTfpnXzy% z8m}CI{d1%#vs!E+padZE*K)MD`n3*IhioKt{qCl3=(FEu<=#pXsMGpxKt8BoawLd0 zqrY>;{i~o%nChyN0v_o7q2=QwYUA#;7kA0d_CzL=G`hb36nRfzA;1i9XK>)PMgyLJ z>^P`Uo5H!mr3a$MvRL8rD10l?LY|Y1WI$*E{M38O@z+qrTapY^K$N31>|`#^@(KRC zDS#3wu<-RsV4ycV?homoAE{~V0(?o!O^>CjIRmnq^)t4iZ$!aDD3r?!F!1bv=1~ZAB?!jKoS$+n_IRAQ5)8yZfUroLi3>b7KH9V z`L?e-%2~3K&}*Ki^>A3$*V0Sbg?raJ7^!&0(3QlDkGP<+_=ov9l7BdL6@2WPfa7-8 zXuF~ibcGRa@oFOIbP7)*!s-U|s9SApe;Y$@{{&Mt$r7#j#^Sn!OH=N7<@H(#{V9JU zk0Hy;HUwJ9V5<3zFe#HJtuW_`4KgZad)1^B9auEO3i^>#aMX=~-4sS9eB$H&Jmcf; zIO3*X)7RsB&{qf}4;Fy!$BjdiHq=y~gb)MzhtY@PEE^juR4Qfm+r4q#EmB<v?CXdD|Q-6zW$_D05d zS_k?SCyM+?-H>EWQ4p)UcsT@G{%f@$+>*S`prrYSfiiUhLUmVetop!@#nI?_Wh(XR z1yUm(Jja1>O6A@nUx!4e-TTUjNz^!T%?|Wlnr7ewW}FXk&bS~F24PYU^tgl^Fw?Gx zHP=#X{Huu89Z&bvn*=Wcd*`2`sy7x4^FRK^tffqV4a&;zNt15KrjlF`Hs+E6f!E)s z6BgmG^QAfvb8UXg#{Zht#fhA;s4t9pl%tdtOabViP6)q6=c!3*>wa3QpH1i_IPOg(xH_uD zV6ji%_)NR2@WQPU2f;hv}sQAlO8LQTk$oCVztr zA8U%Y9qo$DEJTCHNH+`XtcqRv4TKP|_J?b1^u$-saR#+jwHrM>VAhlA;b6Cc4Pd|t zU-8fgKf(gAWyRDqyPRe|(=z8Ryr~fDrzgaiZ?3TacMtu+`a10G9Y@NhJ#qd@x4>Gp zAFe5Xz zFpTMUmhPoQ_Nu#1w=?H;Zf?jP7K)dP&npUc1g+Mh*cJ}D`X#_$(}I$tEQrjhHcluc zUI0_+0bN% zO1in$2k#eWCOB6v>#Sg!3(bG?CXh)S>DHuF@njf*osD#D*6puZ!bUpOtBdIxX3)Pz zR;c;{ZSV8^8zD&6S)A$T@!3L_S|j~Diym&(#B^iMljFq#zgWzn4D_led^LTn@tsFh zZe2PrMy6~#qom(Z<#1_tys2K!Rlh@*74y2ULzQI{B006A3Co(s(QKYp&%v zbKcznFfAY995bWyvMLut1jl;pSa%MgTv?vQ4jOG3UW9k$*FDI{puyl^_?yhP&cRhL zWkWrn*HLJH8A@{w?p#X?*T*MK&s5URB$pFM!| zkR$R(G0WflU{SiLeXpfp1sVaV>S0A2-fzS45Mj(eTNkt~V~#h4<_^1BlB@kt2DkP( z;!3CKq~ZOS=Ur^s^fiq$)%V%-983EXMjFXKiPAqG=-zQn(kFzu$?}JJPBB_P#RE}Y zZ6RNW6;X)}yP=!K5eHsYlI)~k{Q+a~puk?}Q)|-xIexUjKG0gcC~!~2qE$+bynncf zG?U_e!h7(9*1jlF9fJ9L*pJ|UZ2t=E3hCz8=a=PfuL=T|=n$_+k6W^KNyy~}9-fD5 zGALAm2d@g&P`iI(bfB8WlbBFb_t^5%F$)NsHT6ubk z3)U*_!Q9?vadh(xXEKa#Wlo$1uLET%f6G6PMqh3rGDJ}HNQr4{aAI}fKQPw z(V@fJ5j~>AtO5bd`bxE#>o>NBVqz&~VS!QOJhq1VYC`po-!Z=p`?%_bp}7Y1fc!?Q zJw2PX`mX|W5)HUue+ZtAOWgw8&$6Du^t&+gXkSbH<#5Pwh5mwmA7UtBZ!14v)~-0I zAjqN}q(4fGENSQH#DJRl?{GDD)wXo4b}o`Q3$m!37ZfZpY8Ke4ZN6#b%qxhaOP_L7_M zn`bL?S^nM=16M8jXi>8Me?ma^Q^S9gMld;x_Ft?nM4H>}+0q&zWk>OEUw8I=VQWwQ zrTkV$OeW z5>rYl`bcf*gqBBb%%9^EL6SwYq@luQhEw}HDa9KCt63?(P(+w}@U&DS*9hLHg95d` z_~l|F)#mx^&OEQB!V?H^Z4E&kKfs2XM;nKD*!jCTuSGTNs}g z504m@(Q%I(e!EZrRaaposlSAg*A4eU3S)*aEk_ zlfP}ebwO8OC^*f|wh5k~&^Wzo$T)|7J($umk=9Rl)=1X=xe=4~R$F5$hN7)pJ~CKt z5(R!4nK3J%gn))be9=rl3YB{?8IYUy4Br1_$L|WAL$3SdFI{$n<43xrI7}89{6?Kr zZb?V)l>5;%l5?llC8PGwTVK+zq?I2tLZYZ%|Lk=@Br(nfo%Ue*R@1KDwSQmj}J^orpn-t|k@6B?*|M2J?$8wmj1$gSw9T8kmN$Vor9^>Yi5juuQlLhvkqBxl5+JWLr_&uf$^8}ud&Q+_xk90la$n9 zCkh9bNOzQxDfRRC9{Q^3B%&|%E8^su_M){I3@D4b&5|C@jY32Ru{R0+J@ldfVb90> z9V85$7V!TD(>>7qESOF-IBAYFq-2zkeiq^>%H9uj>>BkM^ok~ox3qch3DNkac?IVF zp9sH&V${zbVCnPkSq6u0z3E1u+oIW{GP2W5rcjhRzDn$DNE-fbz-d!JG+OG1(}1w<)2C_|z*{`rpOcdoc4K~Pd}uCu3uND%;4{lw zA;B>oc|8ECTeF{{ACLLhd@21&cq4OhY;eCmhvnf`4Ek z1XoHp_SJv8?L|~PP^2R7zh&y3f}h4tVrlrYPjq+a^@_Mk{sDnSdST9}Z5YGUne6gQ zt?Q6v+bj{dcK?;z9_Fiq)}X35q*G_fV|S9-R9DFFkl2&n4m5FeH9Mi)U;{-P$^yBI#|GI!Q z$^1(x?!ezsw;Ph^-aJ&|csYzBS&2(2Zwq^8CH(h}82EtVd}3-GQXD@1$uoOl=fU-0 z_N3SNf#T~zzMFTVKLLV!`g7fQ7QV>MAU=BovW58`ay!8%bGyFQUKU&)EWI57D|?L) zH(raVmK;TDKnn|eHm0Sr_qNCJCk*Q%rHpw=7mDv71{U21O2Re)!AI-T$_R%XBeC0f zGxm@-|2Ql0`{@redmW3f8Z-Vd&YNl6?HDDU4hh*i%> z-NnLQdUp6-B$GExZU4(_SMp(Dsp&vz-sCk5ZB_jUeVM=eM*0}dJ^A~nLHw^$0 zKUZzCa38G?(|((f--}e%Sw2eC2#(H>vqNQ}V|uqoympRn_QdV*>bS{sKkd6em&n`S zm(#;EiJBux$o*#h5IDD>KL=V}y(eY+Z+sAYF-Ye+;{a{6eyy1%#MV(DDM&EEh`21uL_YkbB6+_5FLl+W}e5TT9_~Wd=Tj=OXRO7NO-RZ`wk9 zAAUD3SNj)wn<2oZj^z4bbhV^S8Eo!d^#$MM8{#$7 zvsGL;!Aj(~T9fbkqhNxW`B5L-Kf&yz%x%h8iXSbvPErrr9Z*ynNB5?2qhG^1K*`x* zE;2b^;H0-$g81+EJ#wXTrIjosu!KGKRdzu8^tjQYX$w2SB&P3G5FABSl0QQllg-5l zDBIy$=Zz;b(ZL(E26d7Sx2WLp*9mkJM0b{T;|VKX%*{+42`f1!>g4z~0XN#?y#DvI zx=%QBL3+>{`_E<~i1_N^eR^N(aB?_wwmy+=LPLtbm|W}c3JKMpOTCf; zP_>9Y1LEBAx?^Sdg8}r`pe^8N;6Ac?=HB=A)8B3}|IPEBkT#Cu6OcbM>#PsxY1}ti z%eQxcFKaF&=7j#`v)ttlF9HccF)5^t&ym6uPuZ+B2DD>yX_QaNE}Qmmodv4LAP*W6P58QH2w7j%g74ZT+FAB-vA zhlKK3Xj^|MTOO7}O^DJtcSXK_@wrMv?W&!z)_tUnt^eeE>u87axh-T<9p$rxf9nc& zqU{*H=^lKTh}cyB49dy#F@Ce007pu=?J8w3<^{45OC|5C*^zgR)PuDCRBy$Zwx;Rb zxST?Il9@j`5psRaSG%X8jboFgOa9>`b{%t*!9jTt8Mu9Q(IJo6W1-Nv{rAp&fCQYRqj-h`bhs+t`OD|2y^wj5TwOY+qW@fbOVG{3<%{>_2 zYvHE@tm6rjd<|17nbR*zNnz0h$@oXK#5;f;RYvm31TPKk(5t zNdE-!vDUS{@Yk-Y6jV7|c&r#q57A+~3QbyR3szb?znQFz_~>O{*5}{$oY|O7(USC$ zFAWrAu)2lrqZVw@zwn%lmdmeU3#{gN8gkKWV^UR)%oWPo1nVR}&jj3Ec}haqKduCL zC()40s!+}{vyAd&xb7g%DfqeBm=tEm@Z1^?DSedPS+Zq==fJd7 zOO-ll#~AW*t0gpX?hc~ZMl!t4fK$FR`DwldV`pQ}w|^Zn?9<ql3!usgeZSTn5KH`m>~Jm|I8)PUl{q0N~AJii#{C`8hs^j0h&)W_me3Hw%CPC znL{8Zm5Rb}#W&~B8$BG=oJX+hCS-N4(%G{7njgh~yI>T#98PjKo==Bt0J`aMyJF-- z7l*7u{@+!CcMk^Nl|%5to$`vnKGkH{N=Y}Tpm}-Ban7^a5Z`|{sjv%B-3fRD8PqM4 z(TD1>V*ziI4@h=~+gM{juzm0GBE$O&*988F;y(O;6%if+xXdFHT=)tDzNLf5wnUDT+6XaS`~8DA;)Vu>Rπjdo5UNB&Ri-{?!Gj zN>TD#5D15&onX24;PLmC-ovi7qj;09ojxasRjscX*CLf@;2Tu~LcB-{>97Jp@kN1u zk9Sv}yuNL*hXUM>bwm?k;4|R6h*u=R`Gm1{zvZ7sj)Dk*@G~!x-7gM|2}m}i4~5l4 zOE0<4fxSEBr~ZkG^JAR06;scR26u&b0yQAjY5W(LP#Ubo8Lx5W_X!m(y`q;l7sohG z(X!f)M=vfnmnng1E9SLOX&-rH3Qc>1{{*P z=cmKQa-S96G z8icY4#C6NR`kQoFwaDG@?TG&J`oa&7XErO88vW}7oU#hrVCO3y6p~->qD|^zmsCTx zMYhn9T+bzi`qxk0t5^}RUmve?hCtEyA+_RR^R_~>(Y-jyql;Ce;$eQn?E?8e5A0dq z{Rz4M*r%ht#`|j&`T$0SV(p6aP-6kGLUTc5%MLky695FDo2? z>T$$h9D6Pg*t{;=+-!-hlYQ&u)W6;@f&KDxbT6^iZ@XCCZ;^eVx7K=7;acaSMoNLM66;|{8JTP|xhh_vEnG9<2hBZ6#xpRqz zHHGOpFdwjoHAyRf;w8ijBUxmkW3mKs$M{DQ9XBD~$1GNdExiGz*Or_9t8=a6LjsN) zo~7gy0P&h02V;!o~QqfY-RZ--UA4Pahw}rA|a`RXT)ybahTEjEN;% zlISmEdvq*YzMsm*@RRoY%dZpox_21yD(jALlIKelzkD>!>+U#3G-$6U_m`@-!07ps zgq(K{pD?nL+`NWrRsL+TD?Kw58an++G}iZj4gcx zSKae2Hd)m^KXQ6b;=63=cH8-2==LOL{1R&f>P7PR*2bHuPMC(S(Q3MHXj26q7u~yd z&~I?JS5#a^V#x!a3x8MX4Yy$z{=Qp=Alo!!d-laYK&YG_E!HSu37h+`$_9j?2`zj- zn@lG_O=T1B@%M?${K;gKB_s_apW1da6Ln)!z+HX#__wNPIls(sxU}RVwxiB_u0a_Y zX#ag@jL0+SOw;8Duko!$2kUSW2TyAkJJkTb<9r_2z)PIPnUAe9C6CbkG=i|Ol2Ayi zbLi~1iiqm;us_ef?dOPxMW|%<$4B@2&v84vazE^|piVLCMl1i}Wb*=fQ8l9vgzEw`rj7=xYo%q`hppl6@a3ccsla9LIoN}=ws7t#Zck{C zjO0s~pfwipcbtSnLheBo@KFX&t@9VJ-Fb8dUoPcxHXOFt-4_`p=2Cv&M6Vp6`@L3j zFxYP(;|p0NxmBlBT%S~yv+l$oj`D` zSLDhc%GKUh%3sTh0>pcy-9f7VT`0JhFh%YmUmVKI6PFfihR&+P)4ay_%v9D}evWhO z#5g?x^h_~#d%I;a5mEyns9)mk>WaY3_O$jX!lDnfTqlh961N#bG0_rxyF+~eN-Mxr zesd2ojzn96Ya+@unx5l0FoyU@GVT3?dGO^oP3e{Fn`+M*OguH&J^umMl^WpIE%*l z$)%IV4<=9-WGtq!?5*x5e*HRuey1hYPyS~E2E$vQIDL@J82Ki|B(sn%DL1g?`~O_5 zvYYdi0_>Ly^UUcFEI`Qt0~2oR%|evuAy~~;{R3dc7jVsxTOz-%dbdq1AmDE4WKxGf zo`GvYrKB`U;0$DpPaG2dP3&lO8ysFmpp9_>>J5JXAsGN)LAm@D0$}2uAIirwjSMI$ zMa61HdGEEi!6F@7F4=c74x71NCgjW|mIST7zl*5Z;{wt17Rz=?p*naZhkbzZ+h6!= z-r7K)vUc-0t9#IgJo7Q+JMjI!L2&wjw@AV(cbK`*o!rz)nPtF-{&OAk3lOr) z%6ifJF*zfXcE`m)CUGgar^RlQfx}oADUoHfRi(3VGwCA!yCqf7EfmIkb64pA3543)t)x0{cly7*$rFu|I6YZ#h%@2&K$*NM*7%o z@HinL%{;;?Gh5BxOl&0m>l=7{1GQGZ?>kYYeXj+e;=O9Q2-8r;WST@ZrYODp&Dtfv!tGWB;-tBdJ zv#bc{su@g&4Y}=mk@Q#JpFv;Wt{KX2;UTDg4!hM^`F0-v;(quU7hYMs_T%6GcF}wI z!i4O}K(SkY#zX(=|608J|Jyz@{Qs$`|N9&Kzt5vP-~vlegxMn$CNGh{x9O*Gajz*3m=X4B}7S4qKAYiA)<~pA)-b%q7y{#y=Igk z(HTM*CA!gj?{x?wdUQq?ohW0@|NQ>hvWPKkVC zEqUci-GFkfwjM|M9P!N6)WWKczq~85gZvGtoL7J~M5a)0eUj6hs!)ExFlQ6{2est= zV{3JB2&MOK=%v<_y3n_g;gX_VwRFzu(D#~C)|3@#cizC%d2MCtQEL7T-u`pJkQizE zwslM_{{hdGzY>f^U+`_cnG-A&_RMVnHEZT%_z{L=lla*#mD2naep2i3rdcQbJ-*3X znD?KPAKj`qdMH(dInV-B>o8!$3H;0hIxN2HQFkOqY~{mTHHY_A>JC5h#~q-{xOGg* z{WRfI`P%U~HCudMxpWW0WDqs?MYxB)c%1aTQlA_x;(ibdU%gubh@sUc2dFbaGmGki) zeUxxY_q%$OluGk0H|5xbe@71LpE^OF&hYz>z3r$u18VI_5`P2gf4qrlj)cX#zA>wh z6Jk;FpW0v>Np^3Q&JQ=Z$|E_nrrxLzGs_jXTZ!t6ArxVziBey5`$#oo80k=@k*5|9 zHl$GCL6&eT8+BR^D*c5`+6!1456n@P^Isy$WUsq!K52K^Eq7N>`TVLIPirdQ zu)4wAH#<&@d>X1N@#T22&9vQMa$mjYVrpf_5;lR%iL` z*8|=utqP&fDfxe**z;s9igjxhJ;`{d{vha zu7Dli)t9NNahVtcKRJ19UKQ_>R-YGXIZkZdZlWL1uMVDD0*R{9u7Q_&JuUgy zuOJVIleF^xXmBklBT(hf(toiG!-_BB@BE!27`r{658=i~$mfa6DpSTrCDXpj8JB%F z6_?Df(V4UcZRX@MXCawmfNb<_~9Gyes7(xr2n8U3l|CnRO(d! zpJ)4}C5IB9v|;>O%WVRp%o?Z;5M{P6VGFqjEAwFDf@SAj+p#cxd-h1MVGb21K9#i} z@&m=AG*=Uu|Hs9k7_=tQPd~Oe5>QTjAoc`nkE;^o=;cB+rzU-H z{)M`X&CXNn;70jf>u@2hrEZ;;49a-8NuY&&;?JfSIP3VPtOlpu<8?d@=ln#X-|L*8 zw^M+P?*E0yX)9({e33Ce66=^Tb29&b(Pvqd!Bn+$Nlp&xgFRL-tF*lU1s4*L`(|em zve#u*JbmG=FPnxs;8ZEHgNDI#=(WRVE(d&A1QEL(@ALn3OGO#K)bVsq98E)sDA%t4 z$RPy+vK)&qyuO}kwbe^%p-kr#%75qp^@68bs?wHKNp_S@uvTr zU7}fMuYSE`(?~2FHBH9pJJUn~)*2=H{k#~LQ2nAhes`75FejrMTCfp%Y3)>I`>zj5 z5~+D51sX6(Qb6@y&;O9@2je+n*03iu(uE}FiM0^y7OE06h zqGg3inN=W7vK|8a@$cU?*df2YcVN%Io>tMP0g)x<61%C!?}q`z#33h|DAPk(gOYw9 z(L^vP;%M;9?!tHur6~uU$>*KA2zW75sbeU}>W(syew%C;0zc!ItT-C`ku`7|mv=3p z+h;GckqkTTTd15vja#`sWqSl$v^bPiu4sn!_!v2^W!Y{JT~GnFI4!fd0+hP)lsfhl zM?Lh>R;SLDoLNtFe=}yj5F>dLt94svL)Hw-4NxG?)Y0vGra;6>f#cNEr-t4)vwm+~ z(*Gb>es~Ub40}UYjQc4x8;L4EU41c^PYq+zQn{0Zed!2W;DNr=oT8<0CW$C3+MR&& zewF>4gHpe(QWVLBAJp23EreyM{3@gOF50b^`B%lQyX$FoJDJilxFfyK^1jPz073ow zEMQ^(^Ga56)BS?7NYi>M#Wy8uUxq_sY+wPn5p<`U0dKiu6p zret)SLsiXnzx=an3qxv3jc~1hDCq}bc*{f3UjMF$5%+VKtev+jOu7k|(FcFo`d@r< z?3o;@S&!+pT1B8Tisg7rRrGh!Zpfqt*Hj8GvH!g}i$&Dxsq<5jOC!xGn%3`{Q;;2s zSC_=8XmWW&YO;bk1-AblfVjUVtI)A}G51GaD&U@h9Ar@OTy1hgEw^~L-BzB$0+u9N zl1>m~RRgn#AX}mYTdBhoQB?pVsDdeF*kjjg5-MR9jjiUy3nbhsC2JKSUXO4Pe;0CN zThVf)MD0gSCVg2xl9MVWSjv&_UNmg{E2jBKpZep52Fi4*)`vzb5V!qxz>x%ZtluB4 z!!PQ0m-*RL$Eo5IeX$Vurikf82uJEU!tUM9qEpO51jVkig29gu!o&kIM{yIW4QwVz zBX_%H_SYU_i8>|yrG*NlLToUpL!#ynG;Pfv+{$bTf~hC>i8CZtYKZ_jpCz#~#K+7i zrztv8OBvX3p=rJjKhHLY!0*yYc&QE{Demjnx@8Jd&s2U)4|GHD2a3UE7(}|Pv$^qp zw!Uy_Pp%nMq5fM|69|94mi2KlSl4OasziRjp=tF8%yqtSe@^ed_jbzT^O8=JHz$5? zS?0C00#|$Upn(@c2Gk5Xp|>Su#v%=poz9hxVm(b1E z$+{pT$Kf%bA463M(Z5i$JDhxsCs2RxJ6{QD=Ue{hwkc7_`KvqBltN+xRM1CIHMz-% zl*_^idVOUFn4@5R^z}s{*VOPK=_alkCuQxj-;S*;JMMcEYJPEPBH&D0Z}%uzG)fot z*_Rr2yZlSTNBb%^`9Mo(dj3$13OW0!k+|^RB{EF9WbMM`DT`UjZuoN_YFOaaLV5pg zlwLT1cj7jz2NmaiX{AQhdNCU^+Bm_$BH}*~n19_;WXGVoCCvSReBWr>h}i{;iPf?# z==qI-yFJ+`{&~f6#k+{2|B^&_&ZFFlg^eVgWQgh~-(--BsJB}3FQT|vI1cFyVI6kQ zBWOUkv9vfRGinviE9Y(s!QHfWVyhaQHBkSW=wfvE|0N8_p!_r~WPatL$Z?tvZc4_U z;}oQC1w5@MP`if8;95kO5w?~Pq$NQ#*1Fq)zIv+7`CnlDxB~+22Fd?8f zJ0j>D-Ve1UF0-8)3?u0ZL%}TGO>QVo(IoNILzHNM0f^rt!@jWVh>{y;j6ke6_<`nvNbp!=t|LGo8hHCd7;giP= z+~qH`Zg7?e6~>`gi_*WssutGm+Ge|wb^{L${wv8@iFqCar!xHAO|}ls@vX%Kg2KQl zBsggkesG5id<(B52U0;u6TWUgMYkbud;(>mVKeU0jTve8eqsLnD(m&r--Hu5;x(4+ z2G+-mDAyXoXjl~m57{D0U@{^%06Ru?ZnT>SK=yiVeEAx_!u!$882c$7iTF+#m2+v< zUR~Hc_l1XIeBy2}U7}WyTrTe0QcF9+341urfKcIc8p z$x>$ezZQc4uNzzYhfORcV*==#pWga<2QL0dU4xGrEO`^W`J0AGFsO$0bYDN8WuHY= z$wY!slERLu9IF9va+}d|FwCNI%fkw9JE`P6O6zl&A_=kvVNVSF`mO#^>x)ZkTQtY6 zO2c*7b;@DHi-Sxaam`hht!=h2vyN{2L7*3spuqwHTeoC+XF0H{F0+ejI^4j&o(J57 zoB!0x%$!}NPd{mXgb}7Kg~AW2xN&ly3(onM{cHW*UsMFDwVR=m(oJrDyKZ4A`phP5 zva*lHM(ox0X>5dE#Q>U&drZY9te+EidCo7&^1r+v@)tP1+hs7m#NLr(RCuTNod!?^ z=zoMQu-)MnHMfBwb#lbsmGnQ&JS~9r5Sy9k+Y#~ZpHa1f9+B}^N7NmAwDe;fv49(j z2UmF7{@IP@5isGUzO2@(33i6mwN;dCPTMeM8ZUiw8nybjZmiq3yYSHoRQMwR<%LFL z!Qu)|PqlzQr^5dVWwxjRZQj56s?sm91)ef4`6sqZ&6|>e`}5{*oK%SIqFA#$OPi^p zg@3hnQ*rwJ2lkDrD`1L8ac;sa*bw5~6oSu`XJ2OifwJvs*HS>wJ_xJVt zAGQ26{&hd`?6;?CnzYKl#q=3E)Uo1(x|Y!(II7s859cdk zx%r1o66&{m@eju(OKf1g;`ogi`3L#e=_x%^CaC;t-MmcI5>&^ z_TBFao}C}h_?Ig#WmmI^B!Z{%QfqO~{59-}FUqH}V?$?UsFNmoAl2kD1mr&j_~eq6 z4FcGUDux*l_Vg9Nn&7OR-WYlb_5qmYXC!G@^$|O_R=81U8U29ZD*ThqZ@Q~2;XuoQ zC%~y=mj1t(T|K>-BWN!a11}J{-G3Vo_%^u{5c@HMtHeE8CoUZ4=7ja)_amrkl>f>z z0I|NAM;yKDqcWf8St9?zM?%_6XF$ZU>J^sqDE9^=mR$eDq}N>-ud$Hb98|y;zwXi% zjYKuWV9L?24!{!|1roC8hjn8rcxM55;K5{`A&lU@0JX^RShpQM6*X`=`77TTZ09$d z)0|CsB)+rQ(ds$|9@}VUAY9*4rCEo{l&a)lEP#1N)*Dp8Eb>~--(%9=dY|#X)~7yw z+6Tu6(-{nDne*YEOPMcCF5oB7sUc8eCPu$c7uLmaP6$-OJW-4mHM`g?`ONLF&(fdc zwtVOOzrZ=Dlg_JyG?|i>kFz4GV0BRs1+@G%fixXOP`X?1cak&{#`QOf1e~E9pQQ^L z3K#Gl6ft}h1|EK~=>nN(wpHbxpjsf*H@A>{zmC>*z-zPmaH>Q1e2SbIToYzyf)&&7 zkP?@REmym>6~$*2b621@u6do($-IaXxBtcncpt^RX%PAa2ueZsTMS?&!uhxjoAcAP zQY;i-aI*#`O#2|K@*Z$%=}t`Jy?kuB7V_;RK%SP1$bT%c+=&kB{!&yD>|l@A5+z^Y zy8U>$8}w^gm*%TSrMvTJ<-y>^F6NtA8e(}+zl^*LPJUizXEZ)CDi4r;R`{RAJ1}!Z zyg%8eJ@uW2r4Y3>DoCF|Pg=C1K;%kC+F@99&&fop!_h0+4}dg427rIJ`9{<*wOP@8b!6^>_?dn z?Z1dDVEZfbgtv1K?p`h&Eg&*Z^l<9_iy4m= zp@tu@Kn=A#qsU05pS{58SKJneXWgyWIV^y^(bafg|Gyk4hAIbGn_sn<)+~xXvjLA1 z3VMFZAPg&|%ox0lY4?w{8zC# zzc27_?QCS$;GVMnE)^Sq*pfpCfv%k|J@1Y#jH_SSumE-*okZ6{1YrIy>Z<+wJA|73 zv`Cc&8%GOM^Z@}Kg6@JQ-;D zUz~?8BO42Sb2EIr1pZvmA0tfi?4s`A=KH=ls32XsjjOqVQ7Wx>Jf#=0;T)%OdrYcDF}S9{Db24UgAeg zCpA_?-7ohMSvYiKYq~EPs@E4b8o)}M`-q1gXyxvhnQ^($Ik;)2TYX9Xk0ylh?>6_r zRm#Dhd28Y7{%`bcM?J4Uc%RAd_l$TA2)!C=JNuX=4)@HHSOzv!kc z>jpema^5%5F^q6QaP55ZTyVX<4D#^zFU3I4NTMou#iXz|)Ii)&klB%HR^6Ou{o#bSZ+1Dxi{*?=k+#c>3`f!>H~*G6m9-MsJoJp9;rvMVu0}d< zE;*0%;HX0~Hnym?yf=Q-rk|b|M^}*o7_YvoBlW~xPX4(NZrtoohJ*VbP>uLUgG-XB zb@oYD88#ntN`kgeVQvlT#y>7CX6)h@c3NY*aaQ0JX~REFi-pfR~C-> z`3M-aaBNHCmj1Ok;{q4?O^{ChudaExTaz1igiOg&tXS*uR4u+V_4dfI3ry|ogL0JI zb^u&q@S%F|m45OymWqHMe|n)7ukJGLvO6b@bpH6$Mkss8ifiJcR3uV^Yve{EO*mW( zm^D8AN5OQZe_LtL3*}dMVorWM)+9ei#8*J!{hX$yybHQx6m7^d@*mJDgbu6A{Q~E1 zT>ZFWUM5pbSlt7SORCP++n0UeSMO#v@h%L0!8djACJa8xI7k%7cqs1jJeIyRg8Uzg z*oTQS>m$mSp}Q?kjg-g5hql2{CjX0k;sGzy3|r{<@h6JY&#ij`HGPP!Y#K!qKl8g> zdx)(`EObGIZfvm@U2p6oG><I~u~^$IXGvPP|6KRNEcgQMixyW4xHM~i2)d5%f~q-S5Wji7;91OBZ+ikY zK2snmcaq{dG- z9CJt5Q{2Z2&ds@Q{lU-Vv+M!K)FD5M3hg>rD2{5pRwj@T$kU3JlEZ306f4l*-%Tkq zHudJBu>xsFysZm1Nl~sQD$9t-%(v7~Dy7Hz4t`}1z%@9!cc3;B3V^c5lXnMAE1WUL z-0k`*-@&9_{>8uMNfP|CM1|eZCy-`V*idBbET_)RZV zZGV$gIY8uPhe8uwSpkO&3B|+4ZNqfC75Ut{(BK%SS}g+6*k@yG0l%^8nzNzrdIBBu z*9->_Z#U~>P9yL^v^fsD#>LdQ-IBWs4~m$eT!>G(cw0S7w6Y=jbyjMlExw`|P5oEu zI{o9h=p}LN%Gy1~lSK{Tg@&akl-tza;TE$It$oCx=p(EJyJ}y= zNBV=nVO4L}?an6?eFAZ;Z%?65j$Vmeo{h6rElG2|c9FaJP|VDT(+@Jvy+{79Up2m9 z&0lbZg|M%373ap9Xw5~vd|y`_&fCuN%zNKYa@3+wcwv5}Gls0<#>6uQPI`yDdVB3l zKEluK%4W|53l5iRP0(jrCmW2Muqex(iC86+AJ$L>(wg1l-dNeuG~oggf4u zY47Y?kNI$!)rDTj=W$!<_jh}#-3nxD)CGaH$fV!V=I+b%u}HdT;&OG5$+~;i7ljri z?vGrspe}v=Zy3Y2V9Ej_}F#_6eGEoerUgzp)( zZgw9yKW_Bt$&dopDUDirT*m)v5nGC1m#TQ3c%^e(V#~$rxDqb)eI{bdHBSqZ6SJt$k z$BkkhL9F!4oiWU--%BoUW4B5A&#-YTRT^HddWzAqykK!n_MOxxxW|wOms@Hp8Zv)B z4w3@12{f~P$KwepgZZG@aV8RxBmllo3q^!eJKvIxRq2XyOIHpOJrrB~+AwWg+GLhe z^^SHg=Yl@CD41civMA?+m%&Q>LIifq0Mb@J7CG{_yX@0Bm{1Bwbx$=Ui^v7 zQr`P47&h+p{WADnszmxXJErF}3MW3!=Yq#^5AU9v0rIr#IvG0DIv6yL+r`i5b&2UU z^uw68>3B`__4_-m2s3Q}1$|l4*~hZyo(y3o{c5#RTVr|CU3u6A1)a^vi}!*q1U@eq znNOg%6yIK6D!fem-BB8+9L825IZL3NFm;AYnL5xnW^GcJ#pW(z*xp(t24hG1E(Rg z8Ma%Q#T*IU?!Em>`==$xv?^&{v|_D-{2{)@9Ii&2NNpx@?aW5Z{42aXbLe;9Wuu6# zWgNV|aHa$ErY@4d&Ni@#lGv$L(r|f8#j9$F`Am2S#r7bHP1B2o_4X3I?lvZ7pH+P@qD&hGqa2B#_V z$HY#vjuU4U7xu!5O6~&km11SewCiIu1|aLDhPNA}Doxn`kSL4FMR?t@rOjR0fZpx+ z0x!6GnVvA+=bDZX{jVl!JtXXNkr1s#O23FhGu=S`biKo#;*YH5t`!W5&<{ zfh+fV7LW=;j9J4^{kSn)gCCwSi?@R zb~LabkG+)nmHrh@s$1L1H$6>YNwm}1m+5abTi7h-$D;Rw-%`(TEh|9t$cPj7_~~oK zQz*OT9z+GJU0Xo3VUCd+RLANcvb8yaqudvYhJ!6k>+UKL9;;^=`#H}L4CMI%T!_v-k?(|R zag!oraEJ@L5+$R*Us1n2J>Y>qWW~_v13tUJc!@LuT;$>K+Utt>v~Fd;6de%d4Op46UoK;krF`TXY-F_mw%n)>MXma zPbR?Dtv5ZV!`8^`nd)ljWqh~)aaJLY?@!k#mY>SYxR&L-H(A9P36P)I0CCw#KC>PI z(}C*(5HcQ*vD{`oXs@6JY0F>#Tc`qq%?vAOeoS}bN?9ousIa+4(fpFK*wDVkf7Tno zeTQ7)*t_*jJTBL(5lKg5F}^NMZ*rqs5d{r_uIP&&F_1^V z8`|989Vkkd*M&a!w3yp;g8dAD2k}WCR(AWeTZTA3@sxCLkVGk|eO`knMI~I%$jyBg z!hNqWeIe-5)7?D??!8bDH5M)FvP1eOQN&Bytz z*7N94@xZ)CAON^+u5&%pgRkUes&PG2b6xfXs3~MQ`f3E`O`LMLjVyhXDy@*XTYF-% z^YI7Jl>4Xe`DNjS$5h{s|H`>(+0h(}eXDYj`{43Y%sz0l%;_chsO2s9gT3zi*5(nR zbuzRT29Xh38gqT)?jKwGUy$vX*HU>m;PP7$Tzw3B8)cZVMGi|^v60R`BSmyJ{Pt{I zTiC$t@VAq$R)zMk;Es9BtuAoN+ae~?7P5|}=uO$l_Yg#?+fv=M~y~y;V``0g<3tlt^QX;OE5zH`5(QfCFRU?U#(ywI{C?Gs(0i zyW@o4O>`uU2&#h+|2gQSbGS68s{`VR3X~5heLt~Zvyrk+%&cpHQiI;sAzwokfb*8c2F>ohxH?v$)BgRi}<|PF^s+Nuc z$s~yf_r+bRtc_*L85{q2O1`5VINbH=2!+3+wf7a(ZCALp*EsMej`h@{mn4pjY2xKk z%=f{R8eM00?yPwesig;7FgAij`uq7M@|PKiPg{A*kR5Ne>;JzFan9~)%8NrouDYyj zmTgE3X3*fDH~1G6x58gWAF>Zx_`e}nHjZ3)5J=1EpXPdJx7QOlf?Gr%yIkBZ?6}I9 z6a#`s(fmd)lPx6NW}4l85*eDk$RN-jRkjVxMAYQhQpoD5igUV$v42G%n9C64?oJ$> zbqQha>Kl4Vv5`I-NJcj$wON-`G#*plXAEZFTblvRmhZkYe^D*!?uIFahZnP?Iv^3F z+1<~2_6wN?%Z|q+_$D_nga23(%J%R$6b=g9I|W~fO7=n@M$-K2T$10yAP{E!+3s$4 z+BkD}kdvdqa^jrut4>jfZQ85O9W>r}#n3;lw_6x>%M~>8J;`V9-L}guH*bCXen_QJ zJLPel_tQXp4>hmn+|*43S<{^EpMHmxBawkl9TL3Mna+BU(OB6vadFOp1eHSfllcUc z+s}*-Zg+SFsv@^7HX8j~9b}P=NB#=yjI8bla^@VDg>WT(v4_K8nOgOgerpf4dJpUO zkVJ@S*~w6VmoKBi%bZgU6B5LONK+L*SRgDEj!a#CEuz1HlxzVr zCyv!C;&`|>^ZFiuC!lF_+|8Di|J`yA1Q%jxnXNZ2XDIQ@&Mt=Mv+|(0S5BOcsG0E^9z@1* z>4BNM5I!BIOzqXIg2^o~N)|@qTO$T(Innk8?aP>JY)U8gJ(0-Hkbu#z*m6dH(V)Oo z1iSyx&{YhA{hw~-Zw`gb1-@t7B(Juj^6_tj5bVV0C2a(Py(gDSX0uhE*7wJ6fh{mA zezF{*r#xqiWV~uvh!_ptqur49pjc>TjD3ehsG9iG`?+EE_$9O5SIqN>P1druBGdvj zwqAQnF@GIJvygQScl`>7cz3_K4Z7S)Ztr|b&%y7h`_q?1QhKvh%X`FU53L092bKii zsj4-5nOtbt1^$6(JX*r-vbQ1`o0jhijas=S^4`D$5bVafa+V^98g{}9o2}OeYnIYK zF;6+QL{B7h-Lv8ESBU62{?`q&oHeOP9i@_1;z#Z#R#2C3CLy8A&#~HcD@zs}3gQB3 z>X8UU_!)s!^L@F^04FI`k^6uojNroLldTj)>la0ww;)6b_xK5U20>&u1Ey1y6qsam zlSwD7b!UB4RX63WHrI_-qG9~tp+hKbBrC9WX8`vY`|zH``El)N0>txtAG8)z^4CJ- z;NB8L(?m3)tZLPq*R>UWZd=jH#CEclI>F(e@As`s>RPr0DmVHG&bUiua>v6S$t(PeFC>*hxZ) zV+r*NFZ7=ibW~sWKA3mQ_Vq!=0VCIcV^qY6u&u$P^pvZ-pO`_7SU_qsrvxlT_Zx1SSdA!_dQX8B}oLQaU~+{H_oi) z1@3N@n{W(MI{UB4pI6m0b+S(43K&avxmaIgt|GbMVCfFgjK8VsczS|c)gdPC zJsz>jPes)a0r6oiYA%RBWH zPwI?B*(>%Qsoap%DBe{$oCT12X{C8%(CFq@K|OFTiz`%joOSbiHV$$5i=4g(tYY@f zhagI5sb~ZO-zAemrVDyyA;llC;z5E-pziJZ;=0f1`?maw#B6IY{@F7j%@K6M80o^w zIxB!ElsKxsIY2A9&|xh;6qjXx+-HG45L zFMnKCL;?x+vlpqLC_~W;!43E+zcZ4ca;td-Bp(TiiaY>nP4gQV*BOJ_kkVkmwi4n@ zve#^d6To`5`yG76Tt}~TmBBGwJ6R- zd_Nb7G|p@GGTTKbxRI-PZZPTznKx#?ln(m?&JC{>WB*AYY6t?c>Yk1dH8N`K+(a~@ zSMn^{?rN%Tl0MA|M0<>qE`9mDgB&@2An8FN#FDGdKLOzpCQ`3>%hc|_l|b+2w+MIh zN@Onko^;gzM2y{pm)@!nLX43fg2XrIdYUm*jAzsYfBE*}dq&OPADyb#Si!)L$o8?b zC!~NR1L~4xe{Bp5QwJf5(~-Al^UdDt25JcIsQVXjhg&tx4|KEvze@DnfQ3+^ z{&lYikJW_=%cFN#^8s4DpNoonrj$aE7MKt)S|$QJ|HA-+^ek)o&z{NSuTDG=Czg%Z z6ENHg>W03ptIM&B$f*)JR+r5+s8l~@sCK`I=+yDq_P;USG2L^*hllmN*}~+%(NqLI z{lte^(>=Ke)4H+<7VoFJ=eYCSkFb5Fdxz`Z62utiqO%G}h0dLV7MJUn$LdTCZWKl6 zx%JuwiH@&=Vw$+0DA8wPADy2$4?iWdINU)}Tjq;fM;KAOu2u(>sR&2Vb1lPd<%K#2 zJNAa!JYco#y}yYLQlfl>Ceqp9sNNVg01|t6`ohdkpw5g-E8#oHjC}RR34df47jfjE z=*{WCuk2KCtaODoyNkSf{;WI1je-#hplc|SG!Qi`{;Pp6x}oj&M`@)!p3ro>w0Ndrtfie-R`I&`7;oo)*}J(tpahBj)rgpH*b- zMJI$3k|g+pJ`Rz6SbxdrTmy9{p^JZ& z#YCff&0!6H=065Ho&VI=mn_7sU+8S){KCdL$Xb6X6@%Qom(l(1J=*N@Vg#aJdjv?| z4qG4*y-))%)4T>uk#@r>is(Q}oe)_X|o zo6ZMUEL&6I7mHYRkx{g3bx663n^)BzdiL+=IsDgPKR+1J?N@X}Nhsj_)a-O);3i2K zzB-w9EO&8VG4AIe$D!~m(MrOu@hLRMx_Lz_qej&VdMJNSLV1r)gC}-;|8X=X-ek4;atb1$ViuQ% z{A&QsS0fhZ1mXc}OL6QeODO!e=jdidHitD7m-7a~335Ra$WG^R3zIb1iLL|Jal9wvUvw>{w0XkJ~OWp}`M10<`i6>qfFrzL?dWbfp zyaGKKym5l!TPS`^UO%|j*Ym|N(NTAg&dqT{Fez3`DO)_%HJd%laKjXjR zk?;11txaaKes~K#CzMgIkRJSDK!1gQT>H9s2?sJYCQfrhy60h}Mfx+?FyB=?5X?lE&8LMG^29Vd5h>8+I`9 zQ^Wg7`P!W9&_?nHlWrLJv~0*JO5X7pu)jlT3h58=87A@N=nz!oe1BKOXGRECMP{;hR@p zqX$XvqchmAM(f;PpF*qfsZJ~1F9ybyfn;)Z&(V4#6S~Y%{J~ksyzpGu*W?}GYGDBI zcu<=wp`f*&a<8!_W&uZsK5pm`)aX9H%n(ix_zQ&!6S)(>VU)F(OcV z3C!`+Ih+C!45!wU3!B@L{GdmRkQ4u>Y=gjpnDOBfCZ1_1?LTh3T~+6&|AzAYG2oQhQsp}t$B@#aWxUpnp~pu@!I_X10UVxB+Kv%pW^w) zSb`gd_aenA-de;MdOuEvVc`J2zwk?qd_KH3@GN@c@P(_8=wig1?TM#2kPo`;pq(N~ z3beiy0DR7Wgi9>5b_BH#`aR4+b_tYR`^ATOzw%?#CMF#^cG4myU0Tmbx*@^cMNS0B zd`uQME8FTGUkc_9dQ@26+6&FKbp7}hPW9;4weR6{E5~W%|5R80%JCA3Wi%Vh9mNsU zpHSSR38MQwi4+sK;Fw9BH7f$o1G0I#d^qo8 zKJMmi!e0&%t!Laz;H2)EL@Q_PJTlwG@N5t%{xUuC=bY!BQSxQ3QOw{qW?Xv6KjwU* zpG&E8hVFjk)vE;J#C}{vqz;F2-j{5ZA4c z-(nn%4Ld+2MfFj8^UWMlGx!|4)alKB9U~0ih`hb=dAPweDCC)~pmG&4uTe4sQVvUc)s{^EhEadnm7ul#tMF6g%twIPn)k)}QU z6R2AZ?~@bgR=Z}pU>gD`-QcinbZ-ieG>7-6u zs`%z0!1*N`W`qq_>|-SGrC*|x`VkoWsq18xrtQfvR9yL9_tS^itilbnoQUV>ZJDDn z^106*yT3P_Vu^d9f%IJ*AgR;8&V`u-Fd|nU0DJ;2xA-%flYUuV)KGqF2F;eMfzbp! z%HZ=XuK4!*-R{%Yf2xaovRG|?#^bV`45OSnnvD$PxV*@grx0{^60PB5!za9?MCO>g z(xilSuNwytO9`{8z0j~l62Fk*`SaR;<44d+s8{jiRPQH{%km}rcKQg~`E%jRFRZ#f zo-78w)LgpGDGFD{5-?^1{?KFfC)uHXd@^Sgt?P2ezI@`ITzWLI7Zi_vbr;v*iNAWs^jIPpZruY6 zs>s{a+G|Gqx-^(5pVd3={}J?5_YmA{AFrD2=CL2Yy?Po3CuHRqmpEEX|D&5-7(2KM z>H|25jj_xNmr%4J)MzEoY8Y_DKD?rmrts_5{Id#HVNXJt!Lg-;MDp_ozqJ)C?+OUc zwd6%PavfTB^LQg9OTkR)9DK8N&!_2)^FDbMtDG1+TAO`}0|V0w`GkL8 zUX>lZD^N4oHo*U?((K?WdHQKQoQMec>77g6hEV!jYj#6?tWG{l{B>@cqP{FHVGBR zk~sXp3D=d1vvT5wu`PD&OPcui6L2tQC}IS>Zr#jq80S47oyx$S=JCr^XiMyJ|P`#FwN^d9a?sUt!jz|VY=$8~k%uw~g5JEZ;RtS0DcF$FBM z`P8oY-K~lSqBJ{y(R*Y6TES$h#p9B(+3~yhP zB>;~f;s2tyX(hnVYrMNNnNQ#Oi59mxnE>r@XZIs)Lp(g|p1`WV2%7FZeY?#>#rW30 zdWQ)wHiA8N3B9*<;6rS%b5V+Vmk}@~%ft5!Tw9q%((N5ippKZb^gAHd<)5N#*$H(c zKd-QLBe&OHXS$>7{vKrhKbiHu%nIZIL)|2X{=}p`9G^wB6_9#p_sq*eKe%t)Vv+J8 zru*ATiuJB#+>2}EOfe!f_`LT6J?Mvq_^X#VLrd%AP2}#;*nsoHut@mGBszh33R2OA z@U$daJjz$2IWJTb4_HF0d_Rw!1eyF=ie_Zz71d?zU^u^ny3j_eA%frir5dB6S_+j^bRs?jf@|K$eXUqOWS zzh*HrKZW|{_}MQ@=&<(5zfLF2kFf}4o4n(=$%6DnkjyK05gTx0m9)a*$5~R;sP#9a z)ZO{8bj#L{^PwdJ2cTVtw$P==eZLLq*~;Km^XXig%L@50T#|o0RwYldUCgd3?b(&* zB&u;bQR`iOE7ivCo;D#dH%#6I9g^^sUrFhQic$yWX>%4LKF9iK=BqI$lqtE8`CDf( z>*IUDFGkycQ85=dsgLIW92EDwv2%8lsN2c#U@N#Uc^eB(Gn)Eh7?e-9c7;8Dw+)8Zz=3<=SK&%hHp-Zvgc`JlTzp zGZnwqJHc!rOCm08jo;4KW#gA^?50C3J$X(n+^*h-4rmIbI2+IN;Ntlf+OVFp*ywl3 zp(9qimJx3k^1)cKh>IrQ$rU^iMqrrv3S4SDf8gChElyENHlS+U_UU0fN!q2Hdlz&g zn2=Uv#P{R;c=$v~;|D)$fc?vtHg*KF_FFI=VhrV;V)Ex^aEYnV3myKw46qJzej5W$bE@aTWPrnG)4L$;T_+?XDHGW7i_})2xYBBf<6gEU z)ox%&UD7XbB6t>n%VK_J=4Tx>gmcqc&+=6E?^p+d_bwl+F}dbgvZpR3j7)rB0V(I4 z1OLmCTKeQ3&)(*(_36Wl|HIi=hDG&7{R&c&B1p?99RkwL2na}*NP|c>Nask4ln6+J zgwh}_IfQh>(A_X}3=G4>o&WRR@ArM~%!fJW%sD&OUTdAT_FDV5xAJw02RVxG{>z`J zlkz@$tFoAo>Q~n{-8yW>1K&S$rT~N{i^S%&?!7US%_wuO(fAga)xo-4_%fvx)snSr z==ngOGkax5x}u~yH<`gj{6gDQa#ize@Wbv)Y?>q&Wx^I{kF4F`DSQ!M*vGY;asLpL ziE95zjR08~|8xP$#)6hqPksg2V;&P#b)?L%dps{7E=goczn{c-*)=0xC197D^%P+2 zmD2lPl`TKB!#P9(!&;d#Gq{nvHoP2)?r;b-%fTUqdLE=0{$f+R`yRE$! zV1Qs~PB^+HiK3!PWnT`hajWvta!-!ndAOJ%J7{>TjXN#3i$dm9pmBom~Cpq0|YHnnZOlX#^YGEOkbBESPOJ9Ij zDjM#s<_%u`&CKY?Urx!(FelAR&eVzz7O_`YtpSEu>@q8{BqJy23_D1vST`3WI`XDZ z`ItD3viepAzW%W%adT{LrN0BH^r(;QT(e;6kKd8R49l`$W-`AiUjcE0`~&Gx>YZAT z*f=^65v+b?4_Yn?$jBkNi?5sVEX!8Gg=#g@OC6u718FTOB4a9**YzRb&&@rX#?o*y zbgQXQ#{03aX`Mx@>jJm6 zuI0BqFVe#NfBtD+zo>d|gjhN9red`^Tq(-#vijlHi^c2GZ+`Qb(@v3D(R<=McM{Lp z7D@8k0`#n3Tlsz(+Msp$W8n4ojG){PFEUADTIM-Yx@!_Ndi5NT5 z`~#L4Qpyhq-#P@9N3mte`eJh}af0$mu&IW>h$8t_QAg<)b(z0(Fo6LyFB+y!!@ys9 zu}eB1DSg$(AnYVM;vA9MR(-T)KZ%pj3Bj~lS*`nRPyABtaW5@W9Q z>B@LvHCwkM9i66a>&*&z=2+eo7x3E0>K!Y8qR0B%8TS(1vBd=ElWqRC`ApVQx1K&q zyxFmfe@M|dPkKD_;QLqob$o)i4wLeI&xLLScW|Ue)3Jp5b41hRb%eD|8a4niX~2}& z)fDYbsVnYS?IOa<8^l@VD6HGRqTFSTxfab~R`=ZCvjTa~kDBkynzII=wbh z_$Dx5#Ge*(x-|*k$29nzOo^^Kdglk#1Vz0Sl9y$4r1Z`lcK8+f&sCGx_FZYW2AQFt zL&t2<(~6QL<3FFj97Ac__J$r;Sk} za`IV0sQZ9RyfJpR7Z~B`IB&@aq0At2^~VjDz^kCf1IKa+jZ;|AC&~f;Ud%C}?=-*b zry9oRT=}&rF~Ij{g|82IIc^aF^<$emax8;8-ki5fWFCng?Q!AbBe0rH(d6-!finAq z5&gUT(NLkl5G@Crb8R-{2~1o=ALV?h{b>T9p$zTgFj9Oll4Vh#@6cB)-enC5%tQ>>-m@ z<}$Q|4KTQ;)o(CruQn6KFM+2X-N4ZMLC*RepVF(0)r70Ax@>>i#uyt$#`>_CU4-GA zcgun}VyTt-C`%@P3tpkvJLBjDcAUH;Qrj72A^b~GK9EpH75+9xH z{JT!jw7Q_>YB?kUcV+e5z;!`vZ4D6~wTDy?N>~I($o589S-KjccsY(~e8uy;W;I7c z?bB^-C9q2Q*URI=ghiTYQG7q~Ce0e%Q}bYJe!#}bH3sXu^$objosicocRz|aZL6=s z;^}fDqlMrl&yIt;b^ig|);rTy6BZc=_^`-)ZH(9azGsPbD=Jr0p8|m|wck{gcN@*z z>xw_eLL?jeuGcQzZz>A{cvLXBCXwA!U;^lAzwRtJF~j~VVm{BRtJg-SbDhjjE1idk z2z|KCd|@7fr~0k$JBI#L`6qAqHRh!ghT%14qFc2}<9Ctc=nrQJEymdO_CZK%0FzJNFPZLR=n!A(HS3ek8zkZKMZ@jgR z7Jw08ZEw_3u&OYg^p8W!#ZW^fv`Q$C)yhy)UxNO0Rg92H^wagvps(8t_ZPFhS@)$a z^RXPda-nP}AdTKgd)!)U-@dlPsgE)lbxMRq_sW3JrIeF- zz$g+8gRey(oEf4I89}o0$~6}8-$H*kZ2`nX4{*VIa^tR7MggbWgm$vS-B|yspkZ;9 zj%4&XCf0K_B>n*&9S^!iPrGw1xo0hDyKT&b5Cy#7VQ@o9v@y~+Q4Z-~T6wf5Z+F-F z9$LY}(E097nT_Ooaz8jm+SQnBeZ_+j(;fx~C?Zk2n4FXs`!<7;^q;h%|341zK-)Rq z@O->--`|3tn9gt280vX90Ah9mF0W5^zltmCC*f_p{~jo!=B{I#kX#Bq2g()O2aT_ z*I3k+D>Z6jE6vr!Zj2GYr5STeuOSQ7<(ZMb%a3=Lv@WhL@%mnt1kDMqGwDWLjD4HE zXsNio`Umt>4cwvI?*i<4oYw1IOgw6;lN!en*R!40L+Xompqxh@bOW<@z>|^zrhh+8 zWCf<)Ba%L;WkZTyKVhXu55#RSEyU@sSKb7sF9m+Cu8!WgKO;upd9+>vK_!6bxLGsMQa$s{7I~18sO5hn_0zKwjkxZG8qDNh7VPYW3jGs6CDJmewcJ1?t^_8*55Qs`Q za5e2jvdLNZwhu)+{7mW`eek_E=J#WT0rC&fWJau%KQ)+O{=-h6CPm1E;^jG(LaS)u zKrQ)>aLmmPI-LHM$ktx@q+bHtFl}gd;CTO_J`6`P8#weFZF~~8%|usalC3H$wMN}n zb-?sxb5W)v@i%h?}ual)qvxy&(0Q9YjTKOj3UNxjY-ohrJdQn2oK;WH`1!2RMb>|lD`OD6mp zH9#E+-?10DL%^MSBsj1d`x~OiLgPc*No!eN8%8hmlRn8x4D=6Hy_uxDm$=Zzn_LPP zQ~o_GEnQqGcxlx6UGpidCYT;_kkjtbkiCKRF;TQ#p5!}C^*mGH&L5U*8~GiVQi406 z%WNEPax<>4?-qLdkKxkH9tEK+`hb#v$@yaDS&H7i@+z+;@KUjtNDgGcRx_vn?JQWs zdME9Vq^B4oYe{|zOJA;{`E2x*wwG$2X*q2slQYUwYmGbzPd>XFh`R*8xOY&FNEDVHw5b&)O=121 z4r5ZRu%HaJw3kTG#i;!l+G3N>-pKdjBv)o>Dw2z(4e=pEw~VFak<%fsiiE!K-Ootg z5zEU@xa$jL`ChT)PI;rmc+rcG;(y{_Qf}wGc{LlAJ}Y^L0(!k;t#HptQTNw9v;7Ao zy0I-3y~|H_TWHAAkoYyOn%VJt*i@t}NzR~71G@c>QNN}Lo7G@UdG2-dNlXO~)|Yl@ zS(0YPhDljGkA^sRpK!ng>3>$p<|5`_01EMLCM7cs${C*+4xek=yh-_5HjisMi1DUg z=()cfH2nvBjdA!_Cm7^yRaW@xvj3R4Z2r^DvH8;Sv(Au1Jg|8*h^N?MK&#bMzlKEA zv&x|!YPnhbo|Qo#gZpt<4-gV2K+~W&M4S0T6OaP6OlnUjYMYFBewQy`$jg$|0~r`; z-y_loNO$ec2OUH5Cyb*;2t~gVH{^RWWZw>cij8$q7_Nx7#KxW(8KAaq?8p zw_b-W?{L1ImXRoGa{sfHFP|~XHT1M^apb^K$hv_zq6j3*4R{j71*_rlcGbsZ2{RpN z@$bfVc<^Fo`q%ASqG%;QRqgG)#r@M1?6t#f8-&&L+%;BEYd6#9Nq=qEB$oBiB?Ypb zZ_t^F)Z!|EBR^rqBw_6>s_q&xKCeS-aorLfteeb)e`K=?IV6UD zn*1P$PdB9-cscy(nhrT*c)}+>1KvpkWKdG(cPOKY*8C8heM_7-#tY1bQ8gN4k>;%eChv*x26KWQi);X7|FC#cWNseALg_Ynz;g#8pg1`eY?27GtF zAu}KmfSex0Bl~uI@ma9cVGj|k=KC1QU9_Dj71B#}?J>QxW(WBM^LeG^MG^(RajWFk z*Fwxdu?1}k=1@|_mb4dB=`yNGb_w*tE5H0Hgw7B8An@D+y9vy9S{$75a!`VdoC>>L zy`&zzxutmpX_s&07HGv9X1M!jo!Qx({R`Es+X*TBv`d8_G5%8 z7akLE!Y(Au=~}Q1uihc|U0nJ}pIN*S_Qj|~f*+psy@m(9IRET&V^VSm;_TmVR_ryM z@YF^WE0G_5UT9v|^sVlAN|zuEq1m@rS)1Nka}Q3p2!`+f0crBKY88;eN{P($YxA#Q zo+vs|Y#{-MA~gl*G!B)CflsEihN!S?R=!}m-lQK9pwFq8D#*YM21iGokZ3dWw~Q<4 z!v|4i#1C}TAL3CzAi{nNeuWr4_`e@s4ZH`>v0qm!KKlSy0WWVWZ%1nZdq+C~eLqKU zYX>1d-&g$DQ-gP84+s-K5ZbkB%h=WW)7km>lRUt|gW)`QfGzdGgX0drckkxsG2pWm zdIB6A685jJV#%Z#NN7B!Ax5zs(!R$FZEI{CSq`ZSZEUq4To!7DloQNrt$SmiJ`FSa zQ-Hi#+aH%8kjo!9sGoBW0k7ZRZ5vORT;v4pf0}4HyLCRR_gfV>ihO?-OuW?d=CcbD zuz`VN956HJTg(DRZq(NusdQzRAM)fv>DT3py2y&nUi(19hv<^J>vhM>&(30WW)Go# zeaCn2#R!^#bhjgGv1>|40+r3^c~lB&MpqG>Fl72s0z-r86Jtbfgr z+?~JZ?T-#1jpH9%SX4rC3FvHtxqfeYuJkB|Y)#O)=9uo{{4`5Q@M$WZk}n0RLzQ>5 zD*mvy41=X^a$%g9(Tc}%bhgS~=nCf|tKIeqm0-T6??qx+Wlo}NQ~7owgWb2S;)poL zlZENCLbq=@-Yx@k*A?%FdchataBvMENB8&4MeZ_z;+8wv50Nzk#V|JROX!8UbaaWT z)VvF*CL?!(`FD=RmOa@!3z?z6p*FGUN&$CUhrL0D+H7p`hBh7cwR5pe+8nRDFIT@m z!5~PBKelGfan^(g_8HCvaqiX6C7{pa%9wvu_Is)2qOF8qdro^wZ==HO9T+b*cmuHaSIWX*xjk);FYonkQ0H0LvDR97oL5yO8K37EaFBg6E1ud zXrW0S6qXKjivNpna+$05_RCIhY@+v8&aGhYy?ONa&VcxjZSY@rJ@j`?Llrb?DqsI6 z^e5zl=~Up9j(y-Fg-c&*<}Xv%gSM(C=sq6R*oq+8ms9>>ZW;W$0X;M_n-byo~ZcWpChDS}Lt(FYWxvT4$T>seo==fW9t4eE_QkC-rcyLoD zDrzdPLI-2xt2t4{XZ)-l_N39}#rzHG;MNlZDS7J7l*6x7>AFtf_AJ-^b1H-+BJEA7 za8WRLU-9&HQ*n}4t!&NrtcHU6AQubM*#`CLKQWF2#M|!83?01J6T~4N+vhn+gvGW7 zhI(44J?3T1?oEM*r!zpnl+r)T^Q4Ou{5+*lCftjWuo#=yX%hIHz1mrivELTwKz}+y zx2c zU^eu?&P$JpDaXA1Rcgdo2)_$`U@?!}LHBg#SP_bV?QBwZ*0xdQOvhK9af>+@{q ziNvH41GnQ0MeQF}GVCeNm@Y2GvIn~3?6vd!19<0ZU<9C!C)0Nn=ihMJHQ}F$e>l?r ze1|-l$j5a`9R5_*aJ@010}9>M`@Q9F?b>{_xy`InP?y?p6_aErV(~@%MkI+_>>_=F zGYG+jpQb;;Fjx?95x~Wpty<-mz)+rJIukljn(k&`$XRvf`-WC((U0uyRNUOpnQ4%!bf!zxTLTEz;=k4a8)EKuzV>;ex=UGANbjE!3u9bH{yCU}-x+`o< z(QQW+Y(2;b53q@Z=?|K+1&3n{0o>oUTN3a0LVY^MAuLwg*8!vAP)&g~huXDeGbX$A zqd!4C&rCiV6p!5RiME9V7r4cf6N_rc!eJysy)xjt+}|UNe8iY)ow&BK7j1o6iTC@> zTt%-59pei;Mj*u!!8vw*SEwdag!$se_KCeKIOG7egF2;g9$2-u>=eRl82^L3Ft=E2 z{{ve$M);-Ldj8R9Db2hN>^D$J zN#Vw9s^_OVkX|IT`KjmMPpz8%ʇ>E6T4x6l?+W{x=3yZ*kTqG(o@4dHWAgtyxH z>)_X&Pg|K1o*7+tLAsXK{da1qrA0d4q*%h%erydNU*@7SF5{VXW^v~0lB$`4+bAQG z=6CsI*^H8*8uz7BXNqdM7Hj@PCBPPAwR}{^ZYw{5yL#r8vouGr*RO_O$5CXmT1bcA813&t4O0^%SE}DsHIwiUS6|RSuNa&I#vZV;-UBN{tA_M?dW~BhybU>;4aH|;FP#A4paeXNw{cD%}U5C8X+!I|ZX(CZsQ^~UR+)oojA zq7CSsm&p`I9*vIMGt0}xhD)W&2m5nAvbDZ0no8lT>wyMzhR$q*k7sOF|Nf!PtRvAV ze@aKSQrp|9V0RP&fe}|!o0IfcXj+9CTF$rRKx@4 z3_8;mB3Z;n9+$n=(Vat_D~pmpoN|zF3~RoA@xiUG_-3A8&W+~tIuvbI*mo1|56}!P z6UO4vPD@yuCDkw6!yKc8wa&{V+uXAgB&_}=E~1uPpMALg{7n)ieN2aLCg)_Dn+&9YYdL@i*&SRF2#BU)K>~D1WT>unhbr))G42&53_IUUL=@DlS`A5yXL* zwJ;;$)G(XkvAS5G{Z4wn2eA-xJ0@xG{TIo;Jy68;-Rxm_(d)ap#@B~8gdLt4lR!AJ z{I9iiT@0k}K7+aoK^RkpAOvL?9{`Y6g1lNUK>vHt6Lg}4iqKYT3H^LcA;%pBo6rr^D+S>) zy4J)~iJ8w{U^nHsJ&&nHw}2jE&fc}Up6wj} z zCZC2D|E@*Tr<N z{fcFmQ_vNQY}xpINgR73BmJ`(`8zSaKA*hZ z^NZVs&=DXj`Fdgx2VPp)D~w(ORu$8Z!w<%dA;4{qez1F<(bS=n#5kM_bu(L030lGJ zvO`$oLIGwkpZ`ZuTeSkG2tEA+)sT07fr}OtUpGsCFe26hbdX%=S98Dt+O3e{l>j3l z!9D$UvXP<ZftHLf-s*t$8^>A=;!^lR-cdLs}3`OWzL`5J2KR%?b^U2_gs^i zebu`}c(2gux+2QD?@`3P96Kt4YRL$b`fuH~A`(%u^dVU9+jqG0D*f)&R-}C$$i1-^ zgVL4Qj?OEYzX{TPYX1gHHKj(b9Nk%ZtUq++(2=L_zmI3L!?~qtrB$MeTz5SA&*%;YEKj?m;<+w5` zqG-5bQE0~)ZR*dMyf2pUad{k@nPdr9GWO^RHyD6Z9M=Ij8;$@ug2b?i zTKw6InC@?bDc}%EY=hVA4I3AwUXt3k|EZn7yFpEq1f z>CcF^0=~BhaH3!9Cu-&a=Nm6FalR|BWI-uX)uA)B?;(9-Ap))d)0yvN!%Z)4`A@Ts zV}!q>Jt@e#=~s#C(cScl0|D>WJ+>c@(*b`;LGz=$;}kF(6vD2`s2Mkv=i?cHIIsz~ zz_PcP_F=})RHxY*h!QC+3Gkl(#GeyI4+RL&N?Pb_J5n1f0j8wuFr+7xT_DH9Z?0bD zmVZha2DP1wqg6JnA44{h;^a>xz>j9mJX5zYvyBs=>u%kt|0=6L0+`jqj4|&9_NBpU z;GK{w;tkixgpgkX5GL^HEB14eY3x`w8ex!m3Bo4(&41@Vxf13~wb?oyk*y|8#aDuS zX}jr95QMJ}0HBNJt`te z&3F_M(+Mp{>#|J)f5i}^yhb%=fenncXvcR{D#_@(GSUtS?7zjHi&6!k&Rd?Ph2C(4 zz0;(9!|sZVOq;^|cB4n?B(OnJ!2QzH8Up;<&@{7Fwi}6V36*gtM^WXqp;fg$pstR) ztEj-D7j-|;s%@?5$^~duaA>#~YXPW`FF!a={UwWB>y7YS z{z1N}z27MP4qFU5l=a$vGD))-uR;CuOMhUTh{k7)( zA9R>cF2yLh>FXdb$ca}Q4XO5kZXozPsuXm9&6!DjgNm3ZwiIzHodm`yQ((jqDvPYw zNgKe{mT=5B4lDE({ctzk7H}-IWR(b}J>N@mWkgMgd4VdrssVG;z$g-y8ZJek00!9o zqCqLhVMG#~2v2HLqD02Pw6Qz3kfXNKD$u%k7|GUzyQJfFhx^~sCDxAA(EKY;^Z$O} z-M0?dANSP=8NRO(m~UvTYFTT1_Fut`+mE1t8W*O1r2SJ@N%sMejl9SSDG#$~9Ci z0DZ8F${WHGEankfkygF)5?dUZ`Zz-@_NC1ZWPM~{qx zJlx;=CI((^pk;v-A5r~AKiemPp@6oXNnn^eG@r2&L~NYKW`HR+Uz0-?#GNQLk}dG^Dkm0>#9nw~lx%w*}gp>Zk5gZeOErZY|D zTm54sVJ<#2Uz7FH#{fp*U7>=Aq1Hti7lRItM9;P@ia{Rl#*#BjLFY%XMT$;?NOY#5 zPbnkT1W~PvflX{POd1~KPj#Y}h*8_iV3e3~$xK%bL8$0?MVKWS(rLr+A(#O^`@gt^ zxec={Ix({1k|BDd)oGt;r;O10n@94kh=HXWh{+j}~=VFSAT~TLF z^4k&Rn_+7|6 zrYSVO))6Hntu0f>24#(dzh*ny>Q9c%OU7#>lX~E<;I3q=iK91yn?Uw+=|WN)-t=a& zA1S12jXpzd{W89uqL~V*vQ+WPItqk552BoI1f0u@nX-(3uby%PjMA)0oI44?{Avkf z9qbNmv$Vu17s)ARsad8o)yGU<Jih6RbB$U*50s*%H9Z0Kf&iaX70V{Bh1H z@UR24Lv?cl@BS)~col;Rnbf(_hkImgm@t{Aj=7+C;@qY##o7UIhGeBhks_ z5^Q4ZWh>EGIrUvCPG_U&B7iK^yl`3e#WrhOAfZvwAp9NCzvgS2mp0qNZ=`q|?YC11 z_{%RSH^fPW&MZ|k=%{U?_rIz2m)!8KkQ6*e3ogRnkv9b&ypkJE8s9Kq7@ByCk@q|H!Tr_E182la*0Dx^77l(qlV#;_zd|JN26>ImU!(>mz)E>1d2$a zZAIZ5rzYG4ir7S&3X`fc&mZvBR;{4U+cMR6<@-ase7>U`Jht~)9$`9*5jTUuz#5`rCJxg0RvY>2>7Jv8||VeZ?arCsHs80($a{mqQkGZBulU6&}B_HLVA6R9WaOn-yXUGO~}V5<)!BG-={UyG;&ZJP1H zu(Lmc{&BHK1U5w%FADjtDH!rMxE>f_Hn^;UxjsIYI-Z&dUWIOQ%>iNdn-p<@Kk{+c z|9vF6c4;r&dt2CsbruLZ9S2@)*9TB5kq*LTZVSG2%Ztg;UQjft(g8l!V&^)&S46Xn zDlJoOCz=Dk^pPLefi4MW>l{ZY5iu6!)+FT=6QhMHu;p~G@C`4_ z@M9=|p1CZ{{Z;{AQT}dQ8tX>~gFc`Xky?12j_%b!@Ek?jk7xY=8X(8Sz(~?KE6siH z0B@o|xzfufA;Qks(`uQ<%FS?wNE01QrBhqn8aK8|_fub+(1o4y264+3xw#ILo246n zfNC2@3i<`z5i^7RvT)c3mji_JeF-&3YuhFP*2@0i_6QXr{IyHmb-ZHvgf6|A!*#qv z-~?T^H;0zkoG}b=?z2ZmQAKN2HIl(7RtwEBpH&Y4OO_Jn=Go6zw;}4%{XtWEqlgkQ zxAAkc&z2CN8_oxG9E`%1R{=9=AOCP2p!4sie^FolgmWO3O5?ihW`tcc!%ci($lD8C!VWs;9& zi3u*J-t0gT9ASKr|I>j0Z1`mD0l0=){zoP4#{4SP$3nT4I0Pa4P*>6ocnr z+dt!m{1 zxX(kB5AtiwNhPnflUHC9Se_$7qG2U3R5qkv-FDqlQo@->5d-XRK3nPuDcNPbr~r0NtyL~ZZq;;vUkqVIlH3&x-yeL4UL*rTo#4gdsTWh|Z5crY(kS>Q=i zIH5jT8{&4E55VD=x7@hMEv~0XhG!dw`*@+&?;oA>!jp;#I=d<1VtWkmL}P)jnhl^{ z@dL0oA5Lfj`8lt2*m_14E|M6{icfN`KL2E6{@?+)P8goB$PU*|!J7u^xrLogu3%Xu z^abwvDCiPvB+q)y1F7L|yj|{?Y`6C~p^m~&kR=?6dUq6w3Ro#oNq&oIpxtOzph#j0 zJ2Pp)>#X3KMx3&oV-vdryrB8#D=-TDPU%PIZPg{|2A?gz#ulLiazYtzT@5fc86JjL zwx}z~L3Rjs)nw@A^9V~~(@$ILx$^$8YO>n88`)c<6nbYeFJt$v?~%KKab4nv039lf z2Q^{m8F`;B6MVe*3@J$*oZg$1P87^pu|bQS+$j-A-2k$BP#@*nvpue;L&xG7&jRho zADxqwk%p>DQo%LbfDADx!Q?v%@@#A~6<=GdxtN0};T$N|Cm5cMX+4SaT&?v0@?iTJ zKe?H4EEy=ieA)v7U;lXk{_8_#g}M@UGk;MD8vdqcg_=>{uYu-kh@Q);VtXe{f@U4~ zj;0KAhX8rtRgOZArG|zQ zPtS!ahVPlVZkrfCtV9ob5~=m!hxfUqjsTxey}~XNuG-?)+5AU*;tiamLbQQ7(?Kb- z$z);TV8_CtB#)oD@)rH`Nj8hPx2gWU9WQ0*F5D0!-|SOu?8#S8+a;ouxGaz<#|ehkDcxx=sl4Kn0H#5%8hd!nrk-ks7(_{db)z28F-d43R#p~ zndS^j%BbPgLSn)il(bZjUaN=HFv)z*8!_j0d;RTX*WrQrmXv@gT;i}L)yfju*{9GQ zP)7SLE6zJ~l5EO;$CMa&^ueyX6xnC>o<6ul=rWseJcY%_GeBdnKkmaDyY6^gyCx}v zm6wvklIMEq*Ys|a&5aNPT=xg~Vz)oU$U+Dq>hA}7Z4%#9Y%>crJx(;4RZ9$1b#^K8 zwo|7podbY2i$>dDv8XNaKHb@l7w=Y*9p}<<*Zfd%RsMw@9Ub9?EVP*IkykZMpbx`) zrN8s{Lf+1Q=;>1h>J{cy;zP~f6i=;uE{zQ;iV7`cbtGdxl6hCFh3=leoE$jnuKYQB z7@~WIA}>v^)RM$Us9jjEjx7=sX}OuG4A39`(tG);yW=xw#+>ED>%wh4Q{L6jy6!$A zL*u35GBtrYQ+>uWJZ+KpdwejT44tDWne;nW#H3*Y*I6+8?@ zxQCp>I}25Z!mWr8qKCL|hoEe9?FmlrYmVKwIoNq2zsOBJ%D|P9xW&NT5B$?1*$ZCJ zb&^kKym%IR(q2A=@qFi+uHyGG3`%+8&P8lieE(ebIU#hmaZa*QYvt|!FLG=}-1rP{ z+aVK^55Ia*X{rU@+rTBVQZsVDdQ)^3)tf#GiI)B|>p4JZbBb%@6K&g96QlUJQp_<$ zaR~Z_gWbUg!oeP(x!2jHspdQRkl_VMg|#TPh*YIkv@o=T;v#icngB2qU#LK4vjMQgTr?juAMUbRt|QWb`U9PBne{9Y+f{HO}nXeY?l;saMn zVwx%XfFsng*{R;rrM!I7S*bE1tHo9Qdu!ws53`HenxxXpoXsq#Omag~I*`=L9VAYP zw+*Gn9s2mtv=dFW>jDc7H@&*?k>%sp9c^jp?nb!>L{>f=9w|?_cy$EcHok-0Q`XKQ z3sqIu_4c0BWT(EQ(Nr6RBB>qzm_9Rfmxh`vwoREVNRI3LwOyRk^K;y!jSuX$B8O|WQ%X~8X!)%1(;ZZPR;qt%`TG*@2S$;zFTy{%PZJP}a~rGv z{)SB+nC(vT1i4MW|1QP#K-5wCju+Y;hzl!ZZYZ(<4PpmC_LhU)-Pxy$rIK7QGibk5 z^-&oqw%gGfo2M6bv?kATJe;ODoTs2ad&RYRwcp3Q;y2B;$(dx!N}Y_vxae)c09mOz z>kUlrRr1?rajbm4*zoYnL5lG}MuOO4!Wl|tGiDj9$mN%fLre;03+V5gUQ{S*L|2W! zteYrQeI#4f?4pM&S(&JQ`kpXyHXxd+ACjx6sdi4@=9%(@C3CHjHc7rh(eJPSu|tED z$mZW#8y}*fQcX1-Oz>qRX*oK15o)$fh<@r*3aZ9%cb5S5Qp7%u)Sh5{9ao(y@+AT+ z-+z&>xQcr5H_DV0yFZw+G*pxrLh_?yopH1j#3q5Id@12Tw~~tw(KfSHg(97JUKM!@ zZG%*662s`t7kLzGgp-d=ozzv#H?&~kEPbGMIKK+k^@!visss&^m%S&fsTSkzesp1Z zBzE0m!ko%^Xqez$xK=Ch0G34iZv%7dp51_-T30wb)eUuvAqXo6m8QLe6*SAd#k>rA zlqVu)D+ierTFBfPwg^iR_ext(Q1f_x^o*m40QN$X-u`=G}jv`9w@N> zz}d|Hfw^&$6rqhnsTj8>PIfs+=6U$_#4rt>uPVkHDDPXkzFQUj&Qc`p!!m0?Dk&g^ ztm{2z0O{JmkgZP#%U7eA@CL`vf)~RFoW2lsiA`4>_rq{8cS@yr*IWv8&X#*5RvrMG zSB)}8fkbM}lVQzO5>c%bgh)n6(65Fn4d(|T-kyG@i*by$U$Eh-^{53o(|cZN6}1hg zAcc}F8-YJqy%6a%tFP(*$W8CNGXL9Ip#LKXup-!%wqj%E*fUFE_4Y(_Nr=s9dajm6 z=5_tgB#qxXx2mlFh$|E&;OsH2hYU_BPudBN@_ia^ZDc{md0Y4#uTe!L^vPEB;@mCK zV9vk0PM`y|?!^~n)srV58Xm%WOn*IL z-!|jcKfW#CaDDh?Ms|zCHR$}wiAwSl&q!5yNz!nzBlqShgBoIh z1T*uelXh!I*ttd3wf(}fU~MS4{P3cKq$2#aNSUlQo?T8k=sfACpFkuc<|AIzTMQB* z2@|3!a4W1?DQ=>hGModReAy>PQ9Z`}ioG)yb9uL;Uk~YKGt-K?bNDn-P_vhd}C>V+Kv@2M-bXK)S?S+Z7``lq5}bs?|gHdaHooCR)t2{ zQC_z%$0b7tcpZo2gu1 z;lidlhbf^1qL2it8+OQjq>(>O(VBB)JUxuL$dnop!(oFG&dd0UExDAc)UO=|6mz6P z3{bs;sMjfx`NWiJI^je41dK~wL4WD2cc;Y%s-6%*)O#U1eqWY~-~U`o{Qf;%SHQ8@ zAU@DB^>v5H*DQ$Lk(t6vrxdX{pgwz}D&HXuwqLrquAeRCUeRf)w9iF;2Ca^OLS)V?iO&h?t&<`Ob`R1fk!DZqV0c z4>SOtRHW*fypj#+beo<#IyO;E)bUGRQC!9Lb*w_Di#-`OLCY3ogsCL?r&|g7F;o2y zxle3DTEc;iT*PDhPOv*&EH+7uy#=@kz#05b$X*zXa8_YE;pi+8YzzSJqJe7)SrNUl%i{TPe?${ChSLSv} zccZ@w@BM<28S4VL(E@jIF=y56o| z2AcK@1bNj0g=`US2fs&ODt-HLqHU=ChX|!e-^xypnMSh+nGe1m5JQdeH8oRwhWOU= z@1E`jNeY`2X|h*4>X1pD{=0v>sX^OE)ZK%AT!gYua-u!eeP;cIKbe-A-|6W*W`Rp5 z=-td3>+uBDr8T}9!IJW%0LkH@V}%@>QZy0P`+iNlobR0K@bZ|#oYiGif0b|L-euE! znp>8LNw)l1W~Xk>;WC7^oJH$p6#2CM4zI6Hcwjv}W{Y)MBoanpdq9j*ocsOSBeM9R z?jNU2avq8fX*T=c?&X8me5BY=kfln;x#q*iH!(MRC7JP{izoPZ#WSISn4&tQ_=8cd zL*{*XxMB1Q?7S!|^9YpxJo_^ahkU#bC-ukJD@$zhnme44fHBue&eij(%iftkZ%ZO(@ zh`Z)Cs|zeqges=`EOGcQQ!mK?#j1#u=m$r-f<7YEn~ktjP@WsEm7p&Vm*Y&rShS?@ zw<1#yrTKCONBW*&`1|880fhbYL+nIm^MOevXnYEL^hm-fM;INo(-8?8n5K~Kh8q%S6UAcNvGgzR zu);gwAx4dg7)v^R#br9o_qSQy@bu_K;gGbq_DJ3kIV>Fxf1Ff;93U=pK#^hW6W5@J z7-$HViyM8hPw8G^ACjoGNAM`-T+}qUi~S$Y-aD$PXj}9~5m68j0qI1Ff)r^YozR<7 zLhn%l=^(umDN6561*AjhDjkFfBA_4y2q4m>2k8kVqWO*|xRMyNv((QBPex+Bbm{%CY5J1Hy2w&9aX>6+ z?}P9m^d#>kXka#=%&ZxPxdNVahGCnx0EKlnWBh3^upEw>LXKK%%}q1!-?(F#D0vqw&I zkC~+=rl0PhlVEEcxPfiZveyWv8IBuBfjK4u^6L-P5s2+L>Ogcy(Y2!CtK2VE()T{t zC1PD{;fJe9u&pCc@Ro?mq)HhK9ckm#!9e^$r^QOj08V3z%_=E{^AE3QF8oljTR>eU z$sK?+_Cazf;+sIZ>(+)VpE85@?os6LH-Mj?hKhuLmBAOt-Ov&z1*bqiSR;LSmlwNl zoyhFOz^K(OuRw8T@tF&J*G-9P2SM4SLD$pS7Odh~6J=*Y7mmh1Y9`bJ*vG zlr-VI_e`Qnj|4+{W}kJ-Y=K_i#n72HSkN!@$eExXecWZCh3yowTr%?>Ojk&IuVf+? zLnmwZl_U^Jy$v!tc1;(m&b>t|dn1f+-r6y)Lq zyCN>1G7dX{m9l7lucW?DbP3nRPvGv8pRO6Jsi>Ls*j4hq7LtJOgQb{bISwd>ug7xS zWi__|b>s|&75}(`9c>&QXktw#9wdjbT|DQ98QzhTtuUZku3842tZ)$zj@4G z!e6PZ2H!g*^NNlT^f&u1Bwy24ys5$U7i6*?poGvwR~r1$;5wxIr-tAdm>8GI z13AdpO^_<$p;Lb_>sbj@>z7+D{6PGF9-ESaUKqVeRJi_yP>`sAe8!hxGxn_M9hR;R z*yKl!cZ8-VC|sWj!({+LoZj_na9#Orw!uSp$y?edoiePGvAOWVNKg(C9Ih7*A0$uz z0TKQCr$_o_vl=PW`1d2gOm={pe>LJV^_4yaW+17(PPH(BK*8;owxrbqz_#@I$FBa( z+?H&)q=>ESSU*h!=5jId&Ay6cNHp$_B<}W8Aq*Y8HIRuo59-JWm;TVCh`Y`8Ql-!k zgBIXSyVRWOzd8xvd721wJ1#Q~7y&F}+-Jpb*o*vxqqbJ7)oND<&&@#HG-X6BEWyH( z#8H&xJ%Xnxsa0kj=9i+y)DC`?!39!V3+?;Lnw0 z`Df=0FX=Dgupb1=VCw6l=)!4@`p-NG3P=G>#Ji@oE0}FBQ(`_F=FK3!z_z&NA%Hc# z24czp2E_y$*?+~8O_6`Lt+AJ=jEFRQiAf#AYY^^~FA{Rx*S3@qMbC)iZ^BVzn&V)yRs-VAOH|v1looe2@Xn}y0# z;7&(#u)!xFuNLx`8KsVXwbSbS9*Eb=guPxTo7&I$+7hdFKluj8LaTnvE z^*8G`kbnG-Pd>$)x|}9|c@=l$)&Pj6F5~prAtF$kHo#gS&YgJp^}YcL9JtT>sbe4; zLRJftTe>5H8NJ;+_+^`?@G%oU_C`c(g`I^{6Q`)^8^l$+>)Z(ngT(l6q6J9c2rU9H zvVmQfEt|CTol!$+YrFhr@Owd<)t;0HtZae}#cIf*PQ0msQhbE|gK@AeEfW5580I{{ zP4so8c_ZFD7;U<6f~!HL+y=fRtxfFc8OmbYm>wRQi$MUmFQ8Udz*nzK1JaM3XpqWTAlwKw`{h;+0lhiR>I}g zXW>Le+|F`l9L|Zec=7FV{~gTlPs}*Z@rPC}WDjMVl-_rrM=HNh=)bcXxj;v_H$!77 z@UCjL)^eQnmW;&6F_gWNAcV0Vy*@rl!+85x=zBj9nDghd8?-CY545z!zO~vUlcs*|EEX+J zk)6Ztn~Vf}+Xwlnh19_oC<#icBEN=NOdb5z74reeYZ2ph|Nl3b|9g@AYkdF5t#jBF zoJYr_H^wS*ZTmydTFGLF#gv%c&DLxm$ARZ}*S@DOA8Ycj1}$(Bc3mrrh=AfXoaz?F zz<;f4tf8jZy|wHLAe92PONSR+_3Xt) z?Ab3 z8?Ernx%N0n{iya{IQDaC9iufUy0xz9z0z#W(qeleqwUDN`X)1BTM?=e51U_s2#4pv zF~7YU{G6gZKWFc*(-1W$1Wk#Pjzg+k{MVZ^pBWoU_Gvv0SD@u{H&|>%j z6FA*FRG7gtMe*w$orWZ1_O45ftZK)dDn+iI+nz7pE1_1zXRi{Xgqh09Ndv8?-5Twx zx#5P&uJtFc>ivtJ`?p@J)D@lAC}(COyjw0uT=7fAur^`9=Tl!NFO@&GByaZtOZ&wG zCRIPAb6-ik*Ax!l&U?6+R$Ky|763!Gc6Fb8Do|WPZ|G$2DjzmzT8gqSTqV3+7S$b$ z|4#KXPm6C_V02osUZY(8NqGH3$(j$?3-fnM-z!ZYm^xsDwj@8P4w7}8p|jKIR1VWr zs|K)2bL^8`(>7c6M&fHu^?tdFwZk{EcRe|KUZ$0qE8Pp`c#QI;^ZKY%$cCqB5Sllo z&)$U(VB4i=w|P0I;<=`~_LDCo+$ri(^PAp`HSD(U-Zk&YyA=%2-W|l5EFyo5t+3tH ze;(yQrzA0 zk1%ce3&nUs$3U}dI7&D0drj`KZXom_m)wz!D#9~pV&N2^jQPd`1B~<)IM?;_JE-@eFtt`xGS7U9t>-{M2?VOByaSP7V&Nr}>>iyfzd4{A0~BCOpO zr>A1aM6qpmFxxISvv-pg*4cnYgo|CaTG0*U45WqGZl#3U%#JnWT3Ud*?h5u(_2Te{ z@vE$ut*oWUX(jSSD%+HPTH$K#n23inbp~A2*tT6v?80n`S%LyyrHBHuwOZ#~ps8MyE zzwyd&1HEih8IeIw2&6rm z`~nw?Pey`Q6Hr{!L6C9qK+DQNBGRk3sVbV<%@u0Ld7mBF{8!H^7rCVV6oD`&k| z0)|7!RJXXDuCqWCILjzBC&XqCd+FgF^A6s}z=X+|F-;ff9jgg|tiHE(%r!l}jg(d| z=ky>?k5_ls*p#2Tl0LT*Le6bn3=`0@04!gm%sbexrx+(91L^bM3e9)b#v7AlOC2bz zMdqAFAm#G-iZ3El;FbNStgfls3V-$cwXeDlX1r6k>B+shCUF9!n$27c-4EVcI!C(H zV{1w=TzuY@a8lzExhbyvB7pg&h`{wfS#5&m>H!M~(EwmS3i6VZw~#M48Y%GeJQ`{! z11t}1J!G<-Ow|@oy<&(}8q)V;F{8dV?s>1AGKt0PY#{Z`+$&XZj5XX$K)`jA9n{2z z>kkH;5=Y@owgP9K)L<{;IUY2_x$F{ZPi5- z2chIjcxz=wkI9yFFmv{`mTiz>43RrvC_7?y>`q^H1T;PQCNPn*I!3l$>Js(d%oDr} zfK#6UOg$*kW-E|MZ?Jf8tPk2Y?&Z-dS~4yg-eo~TRW#)KZl3QT_w*=nhDG{-< zNUyUK6~q%-jP+?a(kqP(7cdTfDCdX>xdag7R|p-UQKjPzE=58)D|ls zJec_)nquhYATIV|?t{pPEh`{h$sf&erho|o0|)NaKdlfGdws+&62_a=%b{#*Hd`mw zkge&I2;Z1)a-V51bqw)C>EfHD2p5*(MJU>kY4N7bR&D@qHD=qGVVY&D@TUB;*|9wi zO~n0#v4-@P>=D4uUT7xcdt-tw0%Wlq2Xt^P1PKLrF5NENRwex$+XK4xeL$$Zi#PG$ zz>a=ns)PTMF~t=av*-Q<3lVORyt5aKNPWWczfHExb?G;A^;stufm+stY|8(ZX45F% zOYWlK7B7Qw;I$H%HgoI0R=(3j0vmM_-?*m0jy9?$gwjLMU9C!T`}zXfr!7>3(wm>C zZUO-(gP4n){=~e$uP{*aLsJj51B)CCXb;j43P^p*FeasE=u`c4L3I|YoO5wTF4RTJ+SBF zh_)p8-HIHF>aOLL$(5K4Mc0; zn1`y)!rP#^R#Rf;lU~~&xo4vrw-S*jg|6k$9@zJcKR`ib@oX2yX2F015($}zz~A6N z>h&w(bU$uml0l+9u#8Dv#H~}h@Zw9j+uVQ>f>k3(0}bUHEScz;m)rmNCcqOE{XF*%5bD zlW>*RU4`D;RNMr~nrj+yZ%CN&GcDOlBH>~t5Vz{!+}7iAj#(o=x$8KsI*j#%!}$LW ze)>NkQhSLZz{Xh&9JbqeduP9tJXut6ksjM!zxI%V^9J8^QNR5HJvL)9SUcCLX62b{ z0Sh@`Q0e!%v)E1vsm;>5bB9q8I^q-ECn;or;6>lAQR50;q&u4v$<5rS!L`EU~Y7?S6!PJg|j$tq=7&T;M-5b(>qj zFI^t4dw@S>`Cl+U{|^Y-|DhKC;%s}QS-3P|>TST(g`xn5KIu2|MzfHH%4~TFTCoso zw~PJ`!j{||pnXmk|IwOnqEH94=b(5AZ|SNECu?0s2a0qH#lGZf3~6ZH;eDAK5~jgH zOlQ>NWZfP`HlsBDntyK2>1JrTi;ksR*`kkAk_@3wYbO2y3C+9DrGJP`{ZYlcqA`{( zo>`Z2&rH}~r|`F@{l6fgC;sd8=Cs`{a!;##wa|wpW}Ja#4^*I!QJpWe*TrPlh}2DT zS1)^&-WJNLP5I+BR817n*cp8N2E92bK)6c@06EEfQF=RrGL=z3u3~lP(tmZMY%k=> zS#fqrGg9mh2BmQsLR{`x7JQNPtMi8lqOo!8Tc1$NjUeZ$T%)4MzK4WsibXQgOXUl* z{9guyb2>V|;x)B23+BQ;^K3R(BDFP18>%yanxPinlp5fl{*=d5Zx1duEImk2%-b%$ZJa@UZJjb4RBFWSbv zxXG+eUseMhs$)?A_!pV$Y)gdf0sa(d7Amg4lr#Dl9&|SgvZrF}A{(KLO?LL@OReOR zdQg6)Mvb#-$BgoUj*h$ky_Lc>7dwVhJ5lg$<>eIAilT_C7P{(*yZ>tCyJh(?v^LIA zbfSL4*?)?jf{V9yWODxdkaVrSU(9u=;VMRyIk>m8Z6k7NlR2q!aVx* zGLLluh4rmBki8Mt&+tHTK{M4ZhL@9;6W^R$caeT96D4d?(R^4*u(O+yr~|8cECo*idu}^ieIDxt{^;Pq{sk zPIvn>8#5XZqFf7mha~@g?1aiw1q?F&P@}bXD8x@m%j)eGoaQZZ8uNxEyslYgpt2`t zmC7U-<gFKszrs&2C(m+~T_>=p+(5Pj@NG)PnO+qXekn+w3pu z^d+Qhl-g_S&DH6*4bvXbh~2ScC8$S*_oA01?Fo{&yLP-46B+Yq(%sE;i{qKdQ1_k&!n_ZiS#?*!?UG~Ivx(5n!Y~7^yrDccZfv*NH@x)XyH3=30cf!an3x2t`s(|T3clSse(%i z`$|h}C|T%zfzujSK~XQ>f6)^-;=#3}6?!x=JRpiM5&qm9Vom;sG-milbw1ity8%mg zjUJfKgCtszhA42`eF-X>hl-*a?*ifl6%=+duBA#?l8^}20bt=~`5CSS*T`cnO-LH~ z6nhkBC{19W0-tOyK8K&DFe@>uc8$Jc;idwRtISUaPCtPaV)?8QJa3_h9uCXl!oPuU z(E&Q7*46PU29=i zn1CiI`slGajsHqgtPF~9VPp{As_FQ26rkp~xv---DC*OUfl|O0`D3Oh1PB!~3#{IL z0(x@YxWx^_z+3v-@Ib2|mJd$!=zep#yHBikV$4Oaj#cCM_eCIH-jrxX!7e&9*1u?> z=|rU~m=hj|IWMw{61s?;dO)His1E4Oc0~;cS2Gr;)aACD4o zPPCsZ#yGw2pWiTRnv*KTq&9g4{ET`+xHr}uAi|=w!5%Z>HmdLUq+;o8g@9FKvQ;nemE^!p&;-{vz%|G;en9Hg>F7Y0 z3zQpmZm;Zb(-adj3o7|~$52$)Ih&Mkp^6_6d1bhlG|u>p2W1U|&~P*-oYC!QMv<1L z6Hn<%7XY7U%E{Q`*z z$xIzHe^g{=f)peL2j!inI$bMXNDgPuI z8_|Moq>jGiIJgN5hfhCsM7eE%0X;db{{;Mq10c7yfFFW} zM%PWDBTzAH-88)bJ6dYzWY+T8D>?hr;Dq76F7VKlm>!T?Rj4|wjNU?pyOW4>HDxEK~kIwjvKeCD<#6yO)p z+~ZQk>&C0@Il{vbZjYC6@0q4z@iMPon==&#vQx5gmtbC>UrXeY-LO6Fy^Zf_n)+3T8@X+N6Dl~r>Xo`6lL2 zV*6;PmRa9>y@l>fTizq%8FEz62dWB_dTrr%7Q+Y&pnNayuEq_3+%@R*Cnm=t2mrM9-Dn-(>Xu9{=o44d6WBvu8HysDq1NfLvxr;#%lyj>}$o zYZn1WJn4Xy8fvDYUw8B>59Qa-(@SXUV@`;uHKzoaB8r!f^M&+_i$i|S{=2ln=BUn% zS%ny1%7B^FBZm`F>g*+Cl9AnBT*%#_+)vZ{=JKsQ1;$24bt?gi)BnTTvpI^^N`24# zp4{EMRhvS|i-E2vd?O!kshvZKmR4QwsL-(8m1hinX2B(5d6z z`AxqiL+Sd3pB0GqugAw1(tdePmCwzG?S)gzAtN<^c^loZzczOJFsh`8nEo3FV#mIt z;5=34{Wn-`W$7#vQ*mH_Xfdm&O(U3{%b_@i6Ug8>9s3{`f0`YwP zF96wpEg)wK&Zhq;w^tGWeNUPchMs(WwHjndLfvJLrUSr+lja4fAz&Y=l^$2;c7N{| z@l^3+*h^yi-Jug9!Oh2H*5kPCy_~gehj1%)LsAo2-1evGqxp|vGdRO1an|y)hAOcJ z4#u%(sX#QjYKuXW%dF-!#`yiM9$u<&{-n@pVdyxRF-}fF^5^wRpWj>q*qqp*)GqrR zQA>?79>VNm7r2^rWBU%jY?RCjrpB*sd{Kni}Z^;MB@Tg=gsH5|tIF+Un%I#y@y^XY7yswa~Z3^!IX@ zRA9?86kcLC`%W4nx((`W)7gfV{_66)?p;009d=y9iXAC;umI$9>EBLYn-nsXS@;Ci znWP|LU67bniwq{ED#e7{mT7GH3VSbC#B4_CRqCEP1o+n2v;sa{{}s1E{j?=Dq+m)4 zI&E*x?N-S5{NM#(`mqUKdoTB{-MT>fV;F;+mDa=OUlc1Cwm@&kEJ^t377r64bBA`r zb&*0Jv<{}F?|`D?iECY~0b+vUckuOA8uIm97UyYkmm&ma1Z_!LV!>vh(7=9673D%>JRHtgK7@q++4`+g{a z4%hs`ip7n2@)+)k{T7LRmc>Je4s0os+X7ixVb@JWDAzlk{OZ==xubWjuIs=9WzvU9 zvO~;(t*sawM*E{mb9?hI{EnBfd)d<%8|VPT8?2y=^@AWIhcv~^mDe55hzZYFfk`SR zyD==0K737hF`&=)>)lE+n`-o3e!RSQ&R(iIgUnE?u1o%Y;1pQouJvn-!FWhXLoEz< zBiz(L7Si!P06y6Tz)Os}?jv(`dSgv+UYbN23hzqu~FIM(ND zNn)&+?~C|;^{*t(Vc9NxULCXyOcF&gbdEqkubT@cibJ`d<7L*=V{##THoYRXu(vm; zy#z3HBaNAmy|Da&?{Mj)pxS>1@RHe4xrp7ZG}qn);XXA2{!0phRJvs8z}`Sa!xthU zdx8H(xgXq&4B2A1tuH&aQ2_Dke3buQbgPgbbJ_7KL2d9gWKX<914z{!2bBD~*Hono z9hl2kQSuA8#m2B`1r?1hKbiQ(FvM7_MSf%0(8H;_Nr0Fq+;Td5HSy+7WQYsk`R zL?91ZI^Txix%aO+ogG&KIQzIj`r7u5r`{un1%huXCD7#)I{9mzuQjit9*)N~hMoKt z*brWvH=-rj`)z_|bVz@rpWxfRuh|9c=>RnR;Ii=IjWTZ*?1*3U>&3xCg8Ie}%kJAc z{Lgc|oY?v;yIY-EO6@?>i8Hc8C(O1f#`^C7Z#75kmfatvozC>jAo*+O+-TXCjbWkV zEq(B5>4vX=^4Df9Q-B(&yPp7LsR9UMo$-K8v&&aOi2DcpPJB$9xV&ntZ`n=k5BIjvIO@$A99lIJ;lroWQZbG#US z`vPT@m%?UakCMN+N-5hI1|H}4$?@_wk^s^U`ZpF{jl}rPmL%hJRxh=>X=}I&>Y^4v zCF_pK4h6m#V#mdXFc)pT_sQAw`IXIrwY%&75|9T%EWR-v(K0AKc&p5c-^f}!}pfTMc zlqNF3bAg7l&3ZU043%5U`X?v9$F9Nk>0;eXo67G)uN$8J{Rrl=?j)F`vh-J$p7eJN z_mv(Alc)-6?e9)}rOrh#`DFyo3<{inVI`rTYLsb>SGFZQ%H31^^urJobY=jp7=FoB znG^Bi*s1suYZdc;1H(!T%>Vcib64r0njOc2m3qjt2D1k}?KJb$nS}IP-g=$O)da{v zgFAQ?P9psbJ=%v?*Cq!R__|)v(bm1nr+j6SitPe7)zjBClV-mzxUi}_2=J?`CncqS zn{chEi-SN+E{~ZhYh3-#cPs0ukc5E5Xja0F>tf#(776vuHAZFMbC8RKccF+4{KnDX zjMVBM<88uD^FnK&>+4vKhUgtN(HUYxgoCfEf1TLss0pi2Rwh>$q50bvr&mJKBExs+ zm@8|xDRnD{VuBw$;%eE5Nu}eKNMtLIwdR`jsyi3BOq!}NPJ&I`9gGOjqYC#5t0wyC z*_6v{Rrx&e@xC$dzazlAD6Dr#`Aex`)u7;+Q!;Gw82U>bnJ)KfqTZ*cCq!542j`!A zbsbJ!-kO;(s&B`E=WzN9BrcnP@?XNKtO~2Agr|~2a<A{}4 znZw|iX_~&mYTKjmyvTG$so>}BjWz)boQ05Tjn$*YU>Zl%UcN`}GW41(kZY`ipA>cb z#F|9&`5!?ZSb)v%(U zx#iQO3L z^ssXC!LHwjj4-LPiSj2R*nDfP`CG8J$5PK&tu2V|yrc0Fc=-?3k{*YKLdu;kR->Z) z0bbtP9y!Ykzr=#Cu&*?Pau(cMwY)yNHU8>85Z{2~nFf)soxC(?@IZkVE^n(4!20)g z#2Y=X>PznZh@r(Z=Z`da>-L4W1;=krTNjE#+_YCmL*0^3xwUjzChZYzcf>_mU=^eZtIYvk2bKwhUTYG!M#C96kf+G zrsuv0>_zDiovhJsZmsb{evsPkX9VjHq%Z~k{D2z18x2izIesdd6uFzG&s>CBpWC{4 zw|Fhp>H}S1N733}9pZ&Q+4{_kE@@zq{u1+Qlhsr!G0(xVI9l0?forlEIbgVDBn;~L z((dlkqCtOgM{C!cx&h{=R81*<9b{Uq>-x+~&LS{W)R*=NilJzm!SXk7Kv_oF(Hst%`6%@AUYfYlgcr0nb7=J$lL#}ATh+~ zv)IsO&ZClf#Giu_b^`eWesk!JdSAs3SZ6T}%jxu5;?kn-1;L|JskSlF=i41)(0wmh zt7y`W7vI_f%$cnP@n)6UP#TSxctpkJ*`}ARg=i-%Ys$4-RlOTLc1mS7*cS+$v(#Be zm#wDOz?aEw;}v&pbckXZH6qP~4e93AVc*=4-XvII)5oR(QE41VAl?&Z-(vc)L7t7k zY~4=xJ9Nf_5o{qJ=EIVH_uFK@_)5|$`llvI@YQh;w{KAm=2PotaAv;)V?H$3u9{A0 z?ucBv+ZFR5QRs|(=Z@iWaniUTtchS-{teZd=$*^06x4`gk)$l1KCu_HG;d#eu2$Zh z!r_zq?56LDXy{yeHbp8_zy8@7gL3}2G_@+CvKMX|YJ8}PGBocBtjPfXwEv7~-uxbc zX5f;!gm)8uAQ6N-&^p{y(KE1W*-cQD*+P z0;d{&M<1HpMH0XDWzJFREfuX~m@UvFFH3$?TC~I6XkYITcq|Ib#<}~Pq3Z5OzTz?_ zjae&B4A%-}5ccL&*WI2ykq9KQRrykoMyBTx^r=v^+c0PP4VfPqkSY_BU&-=-xADwp zgsBOW;!3E50BrT$y#ULT25zpvUn}N;q>+BI5RGDYLr#d(;@E`L z?CL3K*Fcx(U{nVmCpX%7Mue2Qnh;Ob3gXvBi_a&4 zqf1yHr!Fe(Qg{{y;on-06R1<6Tz(Fzi_kqj-r0|=q;4{G8{5^@+7 z?syt+LA?KshWITasuMQJY4Tw6%2}!6?r@pT216i;wog)fWpR?(1)oPR$3W`fzWRI+6Xb`71f&BS*tbx6< zRX@TY8Ye;Ax!foZ#C{-%rTbM1(jZ7O?TmSIAzlUH_37^QM;VdeRtr;Hfvze!;Gq|? zmc(I=P|O(|_N+*;qYedD?&N{((*`3yL}&PB1|hv!*WU*tz1c6NUy~~d$i`9bjClcQ zC?&({;=!pSo|<-8lMWm`r=G+Wg-C=UUJh_R;&4FHU#3~3=Txl0{0))KVJ){aPe>hBna{knG%M$|dE;N;Rnr&ez{ zO(N1H9Zc4nj*DGA?G)!y&G3E^o5tm5BJ>n_&|}Om7X0ndsVyW?NSxq@zVzC^jOdbK zO-lkEs`{ID4}$!eoH%SqH{N^jt-Qb#@xB*XS#%BhtxbsL*AGtZybLLg)zhR%<}x8f z=efRIf;N|Mz@l*hJqIUC-Dr`+gFr-`tHw7Oi=H$)O1cd%U- zZHfC|Tl{d(VwzB6@M%tkw$9ArmJm~_={+O&uqW#uQSHoIE|bnfHWKFDFjBU5{eui1 zZ%J$X9QowjariMX(1Qopoav76-qgmB&qnq>D*xhWDI$AlKgJjy9 zjNYKyMQ>u+IGy}=7&nERGT8izJGWQDbMbMLvF^2mEXIV@;es^V<*f%uuz!74N7zb(|-)|OPDkI-YF^xU% z0{Cfpqwf`Dp~O>y;=?YS@8zUVO5A-6HNEL{BQeB)?`+Za{8Mn3P6TOo+BkaFsnUok zPu_N+I)WB{vQ|B3QO*_d5_?sXj=Yw}GhR|QZXMd!9^9piR&Wq{i7bJVlh(#$7(z9K z(b%6qoUr2XZ4M92V~>69%As5LFKo;%h7*&dVSEGa?MGdZD*K1|d753Dep@qNxI(Ub z-7qsZ8#Y9TaH6)0?EZ&aC;57un!}aj@5Rx--|@yy zT(FV^%H3>Wa!B-~#i{D(p{|T#pV7c0s+?D(YgL(?DJosiBvl`K&(q$PO|*xhqK#Gy zZXmY?@pG+T*Bu+c;wn7Tj050?@*&!+F!t%m+2FTTW%`OGG+zTPG(1^d@l%we_2!4C%D=M^)U+0td<;D_Nb1-v%{-VJc!iIr0sbvI9wclqAuC# zFo`g#IF_$M9ahMmRK{Y>Mr(IkcgT7CHLpoVNSf&*+G&y%A_aDQHM@SIA4*bHU;3rz z{HzDXt?Xsv;b1f7|JcTS?^obfIA*};*85@$oq`UjEu-HRL`S_zX2s7Ik`HwXJdS?w zKHECLqUlHi92_6-mBPTrQk>r=At-sWgWGgwLUju&+fB+^av`th83)`&wlqyt}ueIa0{xUFC=$*WVHu# zK_gdhz-+Po9V5Pw6}zFT_zzDaGvs#qYor6pG78J}txV;=*kO-p%px4I;Xn8M#5%$x z{L(lzy4au1bqx>0Dt?_Og-n8FJpv`2Pbw^>2+8i(q~6V4n@t=@uAKtJ-`rWdX%7rgB{vHWUuuk+S}s($n?YbDL+T$AEzw#j1OXL!Zdur;sb$8-Vv zjIeWYHcq9B*Eaj2a+kE8-`Pd~cs;$x&Eu4wSx9r2z+dZ8rVDZbL+ZQx*)w>amf-Lwmw4LuUn9Qx%@+==o>4v!rGKt z@@Q!En&IB+_9qV@0Ft}F&qwM?Ly_HQCe1Es@_I3}N1Vi?hn%LmCd%kEZO+F6q-Nk; zX%xhKzAlz65<2P$PuG6M2KgXZb!2pu`?)wFv`)jP+3Mh+S|rVe1}|ttsWY}#p2GhD z#hosX9|X_WJA4p+X6V$CpPU9>~Xn=3cFeZJQv-bo>&CWX1P z^O=)BajUZDbKJO@MquGLk52d^)BBd29dc}=P$0~(EU+IxoW$w7q6XVgSyOE+lwNEOv-b1Kgs8u3d- zqGV4_TO?!IAm4@zc?HkN<+^`h?l>9$9IY^W(PBh;@zbTMv_1;ez9)UXb3_5vW)m(} z59`+Kw!Qj$_Kq!ubB$izM}Z8NqMKR8nsdfZvM6nO;~EpHArpqzA_h`4WEA%-<5yib z-%4e;kUg!`t1If_(V8#Y2C+QYH<=Lpq>(OXk)GNC?TglZ#isJ-28|S;u6zh2PCgh8 zH~?XF?Njv6`8Juw|7|i&t%C@mPC#9txLHO1m|rcKEIQ&_)0a!3<^C0 zTY;-hgI}~%6D`H-S&QmkTLqUKc-#tLit|zQGT)&A((Qmng8KIN7#Jea&sq z#KSX#wXTC1u#t*-x=R`agfugds?zI>k)*km19g3V{%JWo&2M3q2>9jhxjTWh=NFWE zQ{5-w*c`(4=JLK}90#Dp)ta|R+b*2m&;ND5W^X;1*Xoi%=4&fO^86P#{@b8|N8hB{ zi4dXn#44sXi9D<10qZOrCDN?otq&?8{So=c+Y@91kHnhU=5Gu4Qeo!$Rl6cG7Jk!-ZAqjW=QMvlKS@cwS?bwbR1eG)pSr&EO3 zv;g^&3;7Znr$b*=ndVoirGrCIdax{)qT^ZzHsOIW$5nWq_B6PqSg)=R#ceqa4r22r z_!?9DGocHewpaScu&EwN3Ld%ya^qOkEhjdcWGz}Jwy@-VNEK}V`4oKmfnP=BYdMpY z8kiM-Jq!a0gKB8v3@ip_eZ5Cb3kHbGQx1ulR&ey5iPc6wjVZ8@v^Igag4QV%)#cMzWaoia zwr88~1da$eecTJ%f9j4tNcnoC~|l7Qi42sL_Y}SVKAvR zFEPpa%(zC>n<>vK9RBVHP~Ur{ma3%-Y46Ze6}0bw>}(9(m37F?TZY7C|mhu zNUN=&*8EfS5=~+ifBjSA8UngHu8Jw^E6Z|bE&HZ zMZMT7cF|Kw8IMc&h2C}Aj&`W~C($W(aDM2J_4-gQ*f7=k2MS@iGq196D*3#<<-l{y za=x7>>|QgCJKotBrrQP)a|o%v@BU)BA7&LCvO?17I@@1+@POD&(F*bE?T0PB&KThd z^$T+@?)mf76*1KBRg!n^L^~Jsa7lA#`nmWp=W)4aSeV;#e{i1Nh12lX=nKA- zp1b^`pFge>pG=ZPOy6PVXwY=H_S5Aio%Ld>68*AqlDxndW3uI=N7-2u6jgjn<^((1 z&GJ}w?y$$ZNU8Us$sy+y{BDW?eN9QCJ;KlD{U{bXkyo|NgwcQL{^<0upFwg0?SXmqrf z|LN{y#&lV8`>k!RPjY3ZWIO zzFKCU;+=b!8&joDEFT1l8KtV*&uO;eUx$Zf??Xo4NnNHc>(0?{bMgN z$I?_P?s;7a;v1WfifdO}6voag>d}xzh<=)aJ6_vhX=NUiRJzk&_eFz#nIZ1A{A#@k z?99FOnska@OXkokcJdG6o+m6Tvi%+p-q8ET7Py{Mu*L7s___&L+zESXKi93Dd`%h$ zv`V@q>X1bXvi(k{=$UCM@1$A&>UVMR-txb|f??Wu@*KU)>Q8sy{~Yff&~6^RuEdz^ zeblDtal5}v_t2PQ#jLKrUHx-I`ddc!l>lqQ@q{CtoX$}h1k)z-7k9$M`xy_ua?YZ{(NOm<3hf=t94@WyEcp&t7LNt*NM$ zk2AOYsq5YyCaa`Ye%ObIE-rE1R63QGn7eW)*h~5R|6}eg!`h0%HBkx$N-0pBmf}#L zScRg6;94Yj(F%n^(c-}hl;ZAC+=9Ckq*y6dAUFgo7F>b^lHA>MX3l+Pp7YF|pZ8C) zv$OW{z1I5VeLo{4%hQyF7VIsTH%CWR%v?T+uGqGP40*N(4!4ykR|yb&D2nrkLCYGK zffCOz2_IMp_C(?RG8rJtb^4W4a9)bdWwMaU_H$nm<-Y5u#f8@y^I&Qifebl5E#A&E z>EEIrbmE|iy~E-(q7oS<3v0&Sn&Ai5bIi*xADbz=&buVbc9NSkPJ(w!Al91M#xz!| zr}!(Rq}jwd9!8go6qk7)?52J*Q%WOWYz=mc`Y@vX7A-JVrDKnz=V#5YHm7!I5*eCI za&-%^-TCV8UptXvz>$L@`ebV!;!W1%Op?_@CbSVe$?DPQ+GPZTmgY zOalGcQQ3S%eOG(HMO>->7i`2Y5m^{@$U-Uh_&E}DxcBHb%FoS9MU{yZ@HAsr$&4yH zCd{Np`u;J-p6`EP#77m4t1(-<1`TA9BJHQx>pkkwY$T3~miz3E!TXPX=Oa)JbEChV zHqOt-tz)>oN}3d=1Kq>n-Dj)8{BBQWN7f^Korz&TQ4|NPCI0TEgxO-922s`Mz+c1O ztl~a{>AZ9eDb7kp(tDfEO69}u2D>#Ug|9~MT$b&!G}K1EkXwC~`35A=zWQ9Bz!ct1 z{R+Jr^GqisTmkv^2MpNIjEwsq-17JOQFAv->b-Sv8Zutr(Ytu_{#?CHvVN`O?y-s` zBeNWBj2$(+jXUDN0Cc#SN|dynRrrBFT|Aa!LB?W?1ZD6T@|OBRs2$_$vzl4)Z=(-T znhjRxWv_#-9r(|6o+&qu40JmGe<=_CuLtSBQAFTyim+!yTw9D*9!4C+gDm_)M9%M4y(U-b|Mk5Q z5{l-&EOBmR9WK@};tT~9Yh@7}L zNITj0@#MT<(*EE??Vqt!;VBX};F1^Y2FO2K>~8Sz$%CcnSgIo%a$fQ9CZ2y4mA%hC z11h{D2iH;dtqzppg8IYtTc~vKUA~{NaaDd}<5Hl~sWlN0Gf~Y9Lbc(I%OUO3N#x3q z*#a)tL4VXRL4=Go@a`8Xa<<3!jDPA7Ti=HIC&K^95j*n50#B0B#DWx5I)P0_(gnSY zWV|*^)H)}i0Lak2N6lEUAuS*}mLt9=OR%;_LdsUd^5xawERo~7VEjA&@C!gIH@ zpxF5infkg1DE#p(OL#<6>IjsckNhKmCCLM(<4RQ^WGWEkcpDjfoBz!i)KI{aRR;(E zrScWRqV^S1RSbA^Y?5dn7H)loxKQOF9?~kqc1W}x(c-y?o7>Q)1MwV$G4Q(~JRfm( zKcwDSMIfYp>09hPAA~X=v2aX?B1B4+cd$l5IN0nl{&JY0>yH=n7v6s2lw&y+*{{|r zmIx$2w|s-gH*nFRrpcXE0G?Z0l}s4+c$Xn8*D1tdTt0Fw40qFV*tjP z-i#>ZOhFXsaF) zF1LkSv|nV9lp>(N+3f$8^Z+j2t}2Yqm=_e3`KGw`c4bVTf#!kUq~F_9yGnl^RyKRy zZh-OERtNXPy1}9gMS)BljM8J$xsI*$XFx?0 ze}D2G-p2CI@dIPyT_{>vDqyuwk8a{S+IqZ~vZooV+N2zfhg7sFke}DXoLY0O>tl4C zLNEn~Kg&!59?;)NJ_h!?Vy*%@)R^Ixvi$rnI&9FdUtJ%pHo8(xxx_4>TjV3{mJOw% zUH*uDh3SwL5GCA1BcwieZxy}uoeg>N6SHhc{&8Mv@Tu;rfEKUiJ~)WJ_b7W~%6=0h z&hsS|SKj5Udu$dliNBs+YIumN;}*vKw$PL6hZ-))3Mu@nmM))k0i92!C8~2mp4y~j zx6@*ijNjIGB4=3`$EHSzo5B893(ltM(D|EI8K7J4|u9TmD`BW8uR8#F@kh9h(hW|xMIj{CZQ?K*q!gSA-maR zouxy&i?75Pp)*f;%aPfs`3_7vz`N|-k!C8+38gnRlAAXs)}k&!Z~Z(=B2CZ5wzRs> zfkiL19w1+?Az2O&r$KjYBJvOgfOKa?65Xm#7=`8zha{}a*=BIxDbeWi|2+i~$VX|R zX3@&0*-M|N195uMFRE>my1LYo{y_Z!p{)X+tpU)4Lk(v@^h&$gTuMKIcm-~s;p$Wq z-+!iEX=T48ngMNq9R;$-*kI&4ZYovs5T=RUSo4rGiGrE`AKf~{T^GCAugs{ce1J1_3ejmTXz6ripKQOlpLZDOOzB2&bAzwk zq3goeiLiq+PW&BMLEVrtJ(bE~0M})pps;7Lkni~3d+fMoG2^|)zve+gP_Z=6;)@bv zrLOI#^Kyr{b zJGas0zK4zCL44N)zB~EX(__rfSS+-2xzxD-jvgr`Y1%aMFtNde>Ek$irs(T?e|vLU z_x=guLXw8ZBb4CZHJL%Bs#QB9wpV%xM&lh*cUjJO&bJ`-kz7}IpAv^_L(}sOM-D9lHXBaSH-Un<52A%DYDOw)K;_ZPND%zoF&AXE5x@iPcnvas-8UCF*~*_ zistmHY7|NaDsM@;qoJif5yy$D<>iUvgbL)iGHWP;pQsV1Kb3=!AqNUB)-sZWbAFe* z$~Xs~mC7mcF=?9?xx%{KTpcQ4%at?C_?VwVmFItCMi-;B z0c5r+K+5v2FrV&`FB>0|{@vSI9|{S=hvGOFy3!PP-pUrbp4(eQ)~^4pk?*`YB7@f{ z#LMiYstK!bk|?}3eIvu#&H$0D}WA3I@M63Ywq*Ust~Kv5Lrz(FhEGK3=M;64&2nAR1?iG(>A^~Z7M zG`mC`KFuoqQo}9m(N95SQ%E3ZA|2nW!l0U`a4*69R~k$5tLzEIDNq42d)Icdz^mNV zw#?tfD3SbtS|R@IhUk?VyJ}bd??mz|#T`8RN}qLGdEU%>cYSe(*8Iq}Ak`}L?;9%B zK>xaMRiQf-mHvsf_X`PL(U-YaR?PeDA7~!(7S=V22aClQ2*#@vU}Y7@P-Bokbd~;o zJ0%^pdSRTm`t;S5r$9N)jMvK_+L1+0L4UB`NfTbrL*#&nnrht2uvP;J+mDt0jqf9p zra;0-{ z%HSJbzT)v@fAkW`b9l2|Er{=#C}Y|^uzwV9wI0(V8*F`E0P58o-_*17?u(xV7kE$5 zgRC%=L|GE)`4>*c&}Rtdqlxzn6UnoLH#k(Ci2Ib{_v{oB&e#NCvWYsF6T1l%gfQy03$<42q45oGfRa2`@TO|WKN3GiXn2nS~^?fYSOe*y(PXwbTWmCK`TT3 zHK*~4cVStoLWnc0VMTlykcK+#(J}F(%vz@+w%J*L>TCD@Ou9P%e2a^*VK5bABaGgWZOqwoO;Htwr~6xW~&Ia@K#*#S&{B*j7#;Hga*W$JKU0yi+c$;&);>%oN~n; zNXoD*8cI*o)Ckj%^rH)Z!gq5E-O$xAsnnt);pZ)r|K8adxx16#Q0f10X(O!6{|lfPJ{K{= z{ZjQ1j*kkY8Z2_B3pkTbj=ok@an$^!gyG;`0;KdT%0VTn!ShTs6=o)S&=5uO`?dyT zFkcm5WQ?(ZG8*Fc5&=e3RT~gf%YK991S*V2I4-Ui1=j!34>vu+QCJnD)iq z|2!~%@mv-H`^deNe#RaTD6Bc->^#MYv&yHD4FPR%3am-yvLmxLDn?PU;yV*N)kN6= zD;q4B!kN<~ohi(cBNp-CqEh+iw9o>*7Z`Vc|K{AHVUTWOcVrT!Wyfg}u!+pXK=be)bhm+#+ZFBOMIW~NcFo7Tb3B`DczeI4eY-v^Bi8LxNXhO2ymk*mRbUH)JusC z;vP+Bq8g;DZJE^4Yz^7F=*@JUiOzB@?)%}oP*{`chd3%RSOGPcqIi#3+>)es1bauR zCcFde@9e;USE_ZA;^7`TMH{zru{YI!kqZdsB)Gu_vQ^LpFwOrB%1-3Gl8U2CxI0?v zevd!@r0aiYVK~I`Uv8iWa_Et6{uTL4@9r9##s?w&6#=U6$Xym;)xM8lMv^+{Aa-?# z-D4q4plkso$QWVp&fysL5T``Sf^ux`^I$(MrHLvEQGyTBlnQM#bkhL;@Y2@)8lou+-7K%E>|%8|$UDz+@^)>Dxzbe;uLzi+?-! zgPD;2ZhEM(QkHQ}W_#z{>iTB9u_vp5x9QV&l+xStQMA(8kFOmqW-dgye1`c^oeo<7 zeSU0x^9p%9zwM&zgT=K#=55CDHSe)6?a6?V?z8qj?6T zUn10F?CP7Fn;fi;A}AVN7N*zb)ns9A5$aDW7De8t9+$PcgxR6w9rxx+&W-xI zgh*K$6jNLxNDp$3M~FF7_Db_O5&F2NF4sJ;4}XU4`AL=%E<_y3Ht`27OOF2;TJeyg z`-`@w%~EUaQxsr;z|{u#lMCs5BqotCZcx<_n<%RUO|NC!e!S;kuWR{0jp!7DA@H&wj z8TZ740Z{B?4o%NzfZY+@WB4E6?at)dPiH_bUjskvviwd#6aPRxXLiPK=?m43$n{*~ z_hMe5zZg;{bI2BLr%FfaVd_@0qmJ!1DT_8<>ePC8b+f`BOk34l)~C=Pop?8;P2(pl zFBxu1Y2~Zj@Gl8Tfo9Me0bx8S?9A@a0GmCt+&>+clJXB86Qv*UaKPJ8__EK0<|xBq zrWK=YQR45}OM=pBy6NEL(7X)nR|g3E>$4b~AMyhl{cq?k)UHz}6wr_)G_ZiH1c6A{ zKe4E)>uHnTGQwWmhrkF=A=sl*M+yd-5xNe!mRswqN5VHOa$zGcY`{ zk^+=~p+{yHzF6M*3i*H^Blb?%W+&%*e@BW;?e4I{TYva zb#qChCjHILTy#>lI%DcDOQnd8_=uM?rJLJy7DK7CloGHvsRI@SNj(-8!a*5U2jM`P z>}PP$FfdNKazFhS$I-=vGoWj!)%4~Y02g~G(j+{wR+c3T=~cbdvIPLO7FqHvLM%QP zwzQTt_$ocQg$8=a-UVaV))Uu$DJ|q#}Z~R#RorOMbmZ<4c z4Rg9m!0U|CBCP~!c^qqnby4t@^*Qp@>pJX!S0Ay!#oA^b5#eOztG-&?wCDGE93J!8 z2buT;_$a~gHf^4^IdLq%19a4xiX!#vB)h z!rmNx^^OvaVz;?4Y?%EnWb{#Z+^L_G5^j0hjZ`mC52Xl5cE9J!36w6$*SLfWr-_$mZHjp~F+i~R z*{Tq_9IN&HAu;*1-o;XJ;+GAP?l-l)3KTP6Gx zTXttO#)pX`>}4}rV$&+tzZM-hT+ovAjqq?N=N{vF^h;wdVtZzf$XTgWmuxEm%jRfK zJ^?+u2|h!(bz2CktGAdjoURpsI4P&m;gbs1dH-3DM$*~ZRWU5HZ~O{FehLIj{;Ygg zM~8GP=`0yB-kwWms|c=N3Yyt;9&k3Gnxdk6#ihIy?rnppp0ugZy7$bTp{m}C+&J`= z3lL4b?&?S(!cdFin3Nde@_ypmGNNObtu&gra-=Oj z0TIE3hYsb;3I5M&bAt_FAEQO>$K^63_5PJfaNPW1OWhvQ^_H&vV_!ju>jT5OhkyZBx zHagXQrtU;Z7wSmk>d=P!+SWCn*gi2>b?bh$1+uyKm3uv!XZClDB)*$Dhp`GRjxUYS zaX`|N?nZzvQn*`NCjD(rRX+daq;fugY99ZrcO}1A0Nt9ivM{jN1Q7xjXgV5!+wlXa zu_&lzK5EQ=;dUgN_^lE$`-5#aMGJ{lC?rJd8L4zDi#dIHu-Y+Q8sXtUI3->C7oA@< z-cEbWkJ^X_4!ZyQ%l}sI4XVmtDYlv>tGlohBW)vg#o#xLUaqPGlne-`LQqLM~>;Zu`PjLD5D=#{Ciee_VeE%C;w4mGBKZu6wI0AWRPi@oZ4d(i`c@q}JC>(Gz~Y1^rV(|X8Dc`HicaZu zvanFp{xlh2k_IYs0O_8Hb5_1P*vz5CuKYvHZvVJa4;7;$VdZeRcDvxhSVDb*Z$0|f zkRkngjpI9{%}&=hGodp=4;Y>$-4L7_uObVg!^dJ(m{4gjdhJ&Ou82b~jFZ2X(y&u4 zXfuavKi(&F_2UekVSJ6yxJ^GWa09BSxAT>F@8!(F_lVH5yxi9 za=u4YT9DSL?nZ*t<7j$DSk;j%^%O2MhYbIUv%huD zr>ph0v#swywt}I=KNN7d7=;)UGa4o0?kq^*X#qWtVE4cX5?pUoE?>Rv1Y9ExfD>Ii z%}V9+Z@e-fXI|}niuV^e=Nujw^B`-WF)MMyW}Qq<#AGQY+|9yHGl!P06v}`w5|1Y4 z1DdJuuUrhVU%w@i1oQ!D38DS?E7Td^d>#r;-5LVCHi$ex)zOXr{^DzKoGWI{`gN<{ z@oUXLD0RS*mEO}hkgR>ZD&cKVMGb0fC)xGML%W;~)+N;E0^5lNz!xtSq64Vy!rbK` zQx!~9ZNv<1yOkCk`U&PG{t~g|BHAfCZwWZwRR=tVO`sMM-d_C4xo@GJ9&7%ZFKmgx zmU9QKU&ZxL%W_Sks|EODewuGLfer0?8gLs#G0ia{>9cZtxwT z`}AMT9o#a6^|%`zw5#3`x#W?9L=}I-tu*ukx09N@kN)nIaPOdlmk3T-!ig9?sfif3 z(bgH8pw?Y_pl$SqzCYa!o1eVDh!y2++--szok(a&t@g2-b-JC5m!`c%mZ*Yi+SV$U zPvu*84fmZ^L98ocJBa3o6uIGZQcrPU!~M&2RiQVn*sxgh8t<+#h6O3t9@$DD6jQj_ z>7!GYT8aKx#KFt;qO9+@QTXIA@keBss?ce6!{lLKRm!Z?+p`@0*)X4<{=f2zvxIk| z=B4ff9o+v@!evG%deuN}M#xKT=#rwSW9+1$P3$>pYImK=W2*n7Eb+eRE;a=5--&H~K5ORo#nnJV;KR@ivSVROO(b(mJGx;}TCDT+cRSNPd)t96@x#x6?CroZ(DXpH z<$@*s`{|?2jwL8g^92cpUm?WL<+>*-4~9A4YHeaE5>)e1bL>o_snq1+8`Os%1yovo zXdF)hfF|!8FaH={brYP*dMy=xEE4QiaReQHNgr6*18Wl8Bc zsG)~DgffkN-WJb z0(;29A~`}93)A{2e_zKw0oPT+6AksEMHDWI89}Ytjgd6Z?flhmVstBkTVQf10{i z@XADrFX40Z4Hh&vG4{p$S9F-;=Vx%ml+NHerT-U#nn-7kE!7bh%#L;h;HQiOxo-?P zua{ebhi8Vj?edG-x2yqo3gNTM4r64dFL6@Bx_|2W!gG#9 z53uo{O~2&H=L^llzpa0_4LH#j0pqs_Lws&}KGz@ zpyIP&{-SK#ps2mr6{N1P!7!JcIuFD99EM~&u%4S?rCJt|#rWCkN~YPL(G1$<{|ESV z<3GWt3LC;KpSCYWpg&g4!Q$_2i>(=_qFf?p;gw_^tUn{4X3W3*2Ru!lQ*t9#)Fbzn zY<2}jqzS9)A_6>#NH%<}QHqVjB39q_Uhq^&uI$Z)5EON&(B-Qx> zK{31k0JE~>`prN1DO$q4(Rrp+RDfw&WXfdxQiO)Vy6g*9*Jr^i%!)CWfRADM`@;qw zpiylqI+w`)`(|y zfaNOg(zqe#>~!s$s<%lAA$2NRr!f_cFB#E9i^w}~)FD;F=(h1>ot(e&Maq;n5h!Z- zVxA>l42{n(tX9#W+^gTin0RcFyD_=#W!#?~iC_I_XbPfT-%!`%clC4s)5VA}<; zi3@6;zxRPeAs04|#KXp)9Eth*7Xn6~q6G5W7snX4?2|HQ7fK^f8n4<0m<(}nHB($K zMABz<^n+uFg)^F1Su`PN)TOrHCZ73<+s4iz^^1f^L7KtqW0z|$bFgg6Ub?O$>Hw&Lh0~=tke&H zb}WK2@B|o{-TnK9xNSdk4KMk%{jVW$M=VAXpRoCr6PWf!HiW@EyeSfwlTCj^tjo*S zl1;y@gZ8SbCM#;!0iX@K8<*%}H$bseo=u-eKP^!rdql+Zg-#~9YtQ8F3 z*%Amv81LY~<~2tkrvVdN9g50=UX8jpJoSGYDDjBEgA9&!tpv! zYjTMwu`Yh@vUZ?D5U=5D-&NH<)YvIBtQsG4Z|rMJZAfb-2s1(>sna@Vwr&$Nz3~LkwrM z*@lfKn5f$AJuXJ$FaG)bnDD~l>}5&C!AXcv^k&MZl6-KLq^FYyx4b>7l%{gIaQ5Ok{|}sOd4?b3f8k_jFWvteoa{Eo71#d;lud+_ z(D+zOe^fkGzpL3NayGFU!!pb-H5d5H^!c<&tjh@x#%@?o$NFzr&k?gGH)HVS zgbw$}{}pWZKRjPC6bP~%@!Uxl7P&+0NQHb2)FXe~W2=$!v_#l0vbA~Z*Q)E+>DPuj z#Z%6xlOfpt8{|zql_o_E{8>U^;VMWTX&LfC`;a~#YJ-z%StVV_D{GQPK|OjlkV5V) z-LUKZ{aiPiC;m*$kFhoTi3^bIfb|gy{cG?u~bRN90^G0=pJ);8M^( zEQ_2|WYmat zo>e1|X$_c0Uzvi@(0Bcg{l9lan{wm0LruB1A|W+XDONOkB=s48h9y>>R8ZxJR*F2Y z+?IG+S?00dcGjT~PAo#hSEAeW*z8I4(olM-5aeqOQS{QnW`^?c#QN?O3lI&jLn|cv;~|kKZPuRcg`%EY zy4fUx`0#sno*69PgxU9(k;7k5d?V_e_l#}+P~Nyn%v_BbY`xumSL-|TQ_+!Qos8g~ z7>2frma&F6uE@XOA;p$kAl{M$7mdU7+%gra1;Oel2Ds#eK5x9y8Yw(!uFq`ty&(g? zTL-fvxoEx)xAu34o{lnpKfXUAi1O6!|#$+k5Rp&=V8N zlJ#0$0)97%u995(8ug``82I5(6qUY)LDO^ib>FviHdTJr(~!HC?#Bwb8k?T#1*1~=rGh?7 zA3Z13qiZxng*$n+{{ez*P}aG`<=XZA!ZbuSth6$NDzTT!!~CfMGEz8{T3wn$S3R?M z5#oOAn_KiU$4rnVEEfdnIDucAI~b&q}S zdE1@lpE}n3D|eOkg?dG4-AeM4_H>Bpuz(cc z38iRM;7<1BF~(`jAX)f%1l_*!ISaZuVPfa{5^Vbv!ni*E7lgKJIGK9E9L!=D4+pX4 zwPFYJ?sGT3{FW6rr?{xg?E#bYV zKdw|0Yg2)JbfJS3v`7Io5VK>Ia#B(a<$f;9>PJYif;kHy%-xT;N8@*Tve6_6}4CdZZd^+~Y6o!~7RZ_)PPb?ak?E`7k0Uq7x_1kiN=HB13w zHGe~m{U+1FZTHUCy+WF|ywN`F0|N4R8@?fdl^Hlt!zF`bTB-jN$(cEOo}pQeVLB{5 z0dT5q$eb)Bxy$?glx8)m1$bRrZ`-XAD#}A5sVQHD%NXuT?4WoZb}kpjxEJbC%sJi9 zhl+DMARkySc!6$wN^ zzAJUrbNOzZi?FRRFXK~9nlG(?R>~_iGirYPs0Z-D= zNy7+l9#)pjkLDU~h)p8AZP`n~m{OSk^?jzP%evI*HqN8+Wm9uy3Z=P z@Y@)X8*w`xohj^D1%rnoWG4*C?vox3eWr48IsM6&dhd}E{86oA-MmO)wacZ$F-Z40 zX)OW9-_-V{YRCnmQeSC69HX?&oT6)d`By5QVD&>YR{e&a2Q(tViL1V#Yb$-Ak@h+NPy5UoM~N$ItgMHAS6x?&L3p zY1mummdWSNiPSR|9=R*75^z%m5(>fbR(yFoLMJve= z<_b8=y9n@|BNHQA!Kiuh0=oXxbHRtI;7JJ&E-hA7h^e45?%-wb%Ef=u+}RvPH(D++ ze3&F0hbL+|F>n#v$CMP}4e z9hmXX4BizFM~n<}jm@EF;n(*X-rkUKz_V5DWGa?9<$D9%#ewPBrTzR&>PI_ZNj>=y z`fPuHI&!vNwAkRQd&H3qz-&h+;PdmlNCn5(gP+A3#FgK0bo00k1R^KlpwQY ze|=LYQD=%Ci2}q2lwKY@mY(S@K8cV;O~>{ov^PzM#&G9;JLaUJStWq={InLXW+KhD zsh*x#p%;YA*N^4_5W}<0FEK-MsW{-%MW&ldpA5_)B^WDbH@|QPt~rnZC0t)q<#B#R zLa1>0mm@1c!#m?vYqiZGsY>%WcxroOs>qFb;wSO8{+envd&c>>T%9^+X}W6+=@CuF zd9-4RUJ0GFTD8h6lH2M(J*k$~13jtku(;M&+c@$)r4QqlO-bDUxJp1RWpBL2CgAfH zYa(4bKLnn7q(#NKw2m*KKfQj9>-KE?%=j9zEvSPU1IUuo&ZQGvs0&^UBn$@uFVrJ_ zvdTB?MQ+cC^}M&kv7J4R4m|`#I&k?|-N+Ppo5HWRFK-1| zgoknGR`({z$d|u6BE{z~a%1hvEOK)Oyy~c5@6r{6y-WQ6@C{3KSgbDtwRfmIIp+K- zoT;AX&}H!T`nCkCzW(at$9+R6tD!75=a3Q#U=EtS*EzRKjCj{ zgc&kKXNnBg;ZAs=Wlp){{H6r^P`bpwIvQ|rl3@3#CMV6N$Fp$NYiOn4JjbzlAu(yC z_zaLg{8wJ1LjG@!mAT#!PSk~yCGR;M`l$fV`{ ziU*eU!|9{4GrbDJkSRvFR|}yZK1uXPP9Ei7if$iEc|E$P572P9LVPlrEs6(bc8=Zt zmrEx2b?1*UB1vKt;Vx(YdI$iujCt`r2`>I3TcFKUI%6K`BBYyQ@s?&aatlpAGHP{b z=F|V7Ky!ncK6A`qb5Z>lH&~~5od$IgmWL=^E$!9W($S9zgO6hutvIuLX=_}L@b}oW zHSAG_R24!Db7Lm&xt?7+kWkz+HRTi@jY4}^M&$f zkNiVtl+;mUsU))P?ed*IUw=Dru8CssKPVzdV3!h2X)P|)SzrC7Ft zYNkLWME9oJMa*w_|AS;mVtPf}Mj2>`C&=6M!2JDmU7nPn@^6Sw#u$=1NkVvXwoAvF zG(DQ{`T?rQdfsm9X=NXWjqLs+-xcR+1yBed$m~S5sQ4NmEEy7fKKP;_`fr_c%;Ck% zD-E)Oft1MP`O#JUquJBKfkFU45%C*h`dBT7v|zMO{2&yrdZM`Z2r)3ZO<;@%U&x7Y zj(L^^KD0Y=lFm45j%;fH7YuO802$OCPGu`@CD1$=XHHqQ<^(GjXnrji@L06V)T&+i zUCT@_3*&b~96s1T)|im$(6X{EIW@OmK14MaaM_gT3S|TW$5umjsSe9s9}(1tVWb*;0v>T4e2(Z5zm%X+wNH#X`tYndXJ14U*-WL*L{$U1`&v&HqxddIn2)Gb*`5I~f^3M(WJgNv8aMGgTg@)c*ISQz>c4F9ZH7=6}k zDmvecBr#m+X~Lod6?M6?Z`aDIGoc)T`6e88s*1c-$5Zlx?J8+g-@egB8>e1Cj!h}` zFE3fzG7QQLYLb)c;AKqioS(&7O=ORc|8yrmTwT6t_iHUgX46?ZW?gGb1UF;7?(G7^ zensr6@K4$+%*$2a@eq2Dm&A)|r|ur23do_=5kPIUBA|SJ_?jXY`;4bSa{-@SHBYGw zUp>=t{C35u>K|Qp0JSRA9mUcCz|hj>s}36EFkWs{FaOxiJcGrJxyN~Lq+c80lqHGt ze`NOf=uUx< zwvv2e9A{a_MN(>;DxuatOufJ*F4F3mg(cln^pp>R{9S@Cw66MD<0PPePc8dNc2#0X zTirf*n@T^Q@!dAy^T<#KReag%J8Q;66%`PQLbtwx$)p5d_&xoNWSO|)+$Ko&nfIuv z|2r9YC9>2s_vj`pZ$5<$N+T9=$GX9$Qm&CQK1+d_~POVC**A=Sq!zDy1;tGd{#Ct}XaM;4=O0=NlmZQ3;AG)^FEn$Sec-SMq)0Sz3|FPUW}a8V zZZ=pBe}JCBgM3UZch```t&9*bs&fF0 zJg=cn();+2=oC(ER9W2C1>X1X^d7tZF4q2P{R7V(*nU37J>ZU{SBPn~X@BGvl#Y)A zwn-?jwqqHk%^R9OC@tn-VAwo0_sY$Cd;)+J>aMJWL{cpJ24Og3b_k zPt_FnNs><*X&`9=@zKqf>X4TwIkj<kTNje zD)@wSaF*K-iktqddlLXE{^o%TwMUGk$i9z&hHG>ZdckFeTO4_zISF=jiaCTet7!{W z#xq;ECFLM&=SPwJKmnN3+AQb!Zg&H*K@_m)0+Q2zrg6+tf~B=$y-1}xddSxsfSjLJ zJ{hv+yIW@QSV4rxL-S6ATnOTvUbd+5J>E}%!Q)>KM_tFsH>FM)IIJJMyxqZPUf7d1)%i#8WZaavm%!Ls>p2Sr1iaWW3I+JD(nlbF|nkLx;TL=yNz9z zqihkqet%bt%gcQ8S5ZpOkbPsRIS7dwKp0Co&SqjP>)n`qU&ZyS7~|u54_89BmVeL$ zmOTYuSna)(g+X&(IKeTyW|lko4y&N4b!2vK*lzX!hg8QfAjSf=@Ra>`@Q9JeP`fN`8Z zoaHc82^p;Qskq_Scaaacm0<> z9!dH}nUz(noZC~1TG-*_52VbU>t2|)zzos*0|_rCAp@;f=ITBZr@O%-w{Tb#ow39w zRetlKA#J_qWzdR917^VQtx%49C&q~Sa)#BS$03;J^RZLS%OpuN^B1OEC9D0gLA?c> z4nCG^a)Wd?AuC0QjHB89Ym?>Su{gzqwN??lQrn4fagRI$^hTG(RCq$#8J<(%WWRGJ z6|&OUe^J5cPh$3(+W6i{q5*qDT_v_SP1IF`q|X62>cFIB|M(g9WlHMAZ5JG*ZsoGY z=oKDLq-MvWKeZ@jB1c-GG7a-&azzT+Z?ew&HDl@K7VqK?qexci8kc_#F6Qk2A?+N4 zGl{-+AKSJ$u_l-}lT2*ewrxyo+kAtG`NlRUHs+1(X&tj=3D0~9$jvaY1}o=fv+g6M>SB!O5TRA!@#fqKcc~2;D3t-Rb`rQ zuWD^L-R8_Z{7H_FCMe^W^MSSR%8-n6NR0MwO6cqIDB`?Y-!o>NoO-zg3Fn|4b3*4MdU zqt}H+K4|v6HnZAyN(rK+x|3eLyNmN^?-L5FZ{qp!yF4?! z(oBC5V(kOf^MIRab$~D8re@1hb3k1hpV2DF-rxzt2gD>XzbFX1O%39U=)1^Nvio$Z zeq9Bwe5ti`-MQaR`2~S?Ea?Ej9sPZ|V#$^ls~gKo%(03zfIf%e0FW88;t{raX}@^Q zRdo!PW%#Mp@^kCoKYs$L5tQ;2PEpm!Ox3BM_n59K;y70xngM_uI zdc9DQ9=GM=w;3iL=8#BU2*JNkDyJ;gXL8F@90aB>f2&jh-2KJ=X?}Sx+Y9t!NolHrTL@$6>+47`z@%U+ zg_5>voqD?`tfYA=<|6y8?r|Je_|M)1pCeaL&80~F2{3RNY=ceWLj~le!xULkGcMEZr=s6Ei#c9q zKF)hCyKA^F{iZK-{PHRS^vtAWRdwX4vXr_RSI_fnKDxV~%LdXDQusIm4q~b4-DsSc zEKB1gnoGmCQ!82y?yk&0+v_P)BL{fpLW-9k$Wmrf=u;IvbMV>{&En08DSRs75|(k4 zA8z-lZ*o4H1Hs-357L+Whcc=giA->d)u#}$+80dd)qn$(x^U-lG3>ft@B1% z?=4R&i8iHim)ky_Pmkg3UC*wLDqA+EYiU<1EyYk3kBjhYw0=OB3-zt|+2FS|!!Gg^ka=hDx^5q$T%hK|t-`TnaW(vpmUOQ2Nj(LV zTcG+7@XLad;|x7Lq1Z+SKdzP z{(BfNe6_|DKC=WPRLjZ%!Z@7It^JxT>p|e7-A3Wl%K1R}*O&y7F;9;qqSu)BmtU{_ zI_4@nP=3xkVW$^%c9~k6D&WJdx2>~uvCraL>7vMeQ7H^BfvP>Pq4Wj>I&T753?JTbZ#i_kMaj!sN;%K zl&;`Y>SBU?rB6P?jnCf!#8LSW*L9HgmC4GmsM9UbmtofW1QG`usW=&$+dWU)@*^?U zQQ1mz>L8@3G(IbBP!8#&LiuH~%iPjfEECY{FsEWAGQ@jkx`J6Q)2odZcuFIXdpF5Q zB+wI(E2gFl>eXq|y+stZWC8|yELW^?ZmqIcc^wxop|)Xi->6`(HZId!Z@IvVGXevF zC}S*!?*ls96fdv^wWV4&h>i;`rAx?yfM@7H;0n`QZzY^?3Zs2t)%=9>Z`IEo&ZuL9 z6{er+Ah0!SfAF0sGu7MA{H63bnXwMM%g4!lDr`UJ8)Ay73JFZJhC|%8MHP;!=i#WP zCFJF*Qk2PyuOX0(-v_n@v^O~~-??GX!0-KU*gh~79WE~e|D4UgLAWA*9Sl?Z3_bGX zU=sUtiDYnS?M0oa9M|4$;_~cqFTz*sx3vzE;)+Vq+E$+hRn<9|y`=W$I#>X8)dTuf z8Z|(;o5wl9;L7-G8e*o{T15G1yaQD^8=C^0)?m1A%N4p0e%JIUi(BX5J z3F(s1)naqS8ksm1&=5)!(9&Tv&+t-?$MvT!;6IUa>?t>xu1Z`_Fyro`f?7oUPe|Lu z?c8qntDd1*z(2mkoyAzi0Z)qFs&}ZP-)<1Sk)f+o)jimz3$9T`zru7=Y%Qg;nS2c; zuCuvlptDIC{5 z8T5YEJDxHY=;uhxgKGP2WN#52X>?SX95l9qmW6!bZ!g7vd6pLFH@#HlV=taU&Agf;fpTSW)Fvh-f$8bu-f*Bp zvwi9+cV+`d4w(BRC8y2;eSI^4{Q*i1#v)>D3BbN*c@}R`M{@K(cUkpUCk>^n6H?vN z_@aO(ZH?iAEpa`2Hyu?#q5e2tr}+~I-^B3NK&PFcBHiHozoYF51gMX0jrWxF$bhB zZEyRucq{H1+J56LxQ3Ps*!!!SSZ6b+4Pd_@v=IBL$^Z(pM_(iy{&s&k^lclsqIikE zAT*KsxuYHI9bGT|?$a>-c*(UGt5Z)(#M`q>-ahzE3G23p5ZY*TqtJZ ztGE^?s;f$uV$m!Qg*>BUdkmgX+lA69))^-IL35g+LI*OGg~ z(q{ilxsuLi+BTN7eWDVOf$U4?${32K{1F0sTsT>|BSa!l_A>J3V^VBgx+G9AB=P>0 zJ9LZNr=oKRzM=Wwo68lkF9+0m^3!}?`+Kp?!T=no9IuWhSxdQmU7I%|4ryg}^oS6o|wkR=57@;BJzZob*z^6JC(=q{1* zyc<$5YJHl<`&@JLm1Runsq+-KNPf0*C5VwW`*-}7q`OqE*DLPIxQi@XocEQh914w9 z+!ER`_QRd_zYT*3To-qVS9KnWCFOh{KvYaq+}eDK0g?Jj&|OmGaEf|G?0Q)(1?l%x zDeF&G=iSgdczgu2bC%*9NB<-J@K3$jlf)VK;W|=t5%Ju*(Be9-_|Fa!A{=((lJzNDv(>FZjd$YTR+X=r7T~sN-lJ;PC~nFAC&<8Fs}lO= zQ8DH5Xb!Er8Mw)6X7u+EzIKH;q`P1o#6!RkleRRI16|6alYhRr{HZXDL&Xa$9)DFp zM``$|vUQMg*v(c9ha^3VzP_jz=>R@eiU+&Cek0-6iCo&Ul%sid-$gJLzbI95f9LK{ zFX>V|ki!3k-aHJfPbgb^jm+G^%JmM3dQ}HTt9=IIM1SCiiRj8z9s5c}*8xyiSGH7d1 z*ul7fn=(Y=4OLMF7LNQN5Y=b|42(e*3QSQ35(*0p?rSY9_0t6V?}dN@9E=`}>TG0gZe=3vW^dwZsY84Nbg5Z1gyf#1Br z6VKeu6A=s?>K+^n>}z;1u%>l!RU(^3=AB@$zGSowDS}qK1Sn(?0@?(a;Lh(C^1(ik zeVc_&mK&SYmiv91OhMs2$2p&*TYb431dE1sRcqS?J=!z+)%sTWm5-!IH>@|T_IQB- zC}?1T;9v+~tYE@mU>IPEpK2mxU`P-JWMD85Xk=jU5Q?9A2oR8ltYFZPtgL}xkdmx{ zFPHvkSYThaZdco3jr!8qZBL%=z)ga^PlA%rQxIT3^*$vELcVM1mV(FkBfDTOVm;9?;P!C48Q zBtrTu>F|&ti_r}^F=8Zw=D@I!gNynUIkBR`Mdsk>2;jww3oOCHVftnb@dyya1`I8s zkYYm%l~_3gg+tIRSs{?c3yk16A%Y|*X6c{^go{KZIq@RHDdt$I2!ei6CR$=)AQ#M% zVG+cLiYZy*A;bz7Cem>tgqe$C(n=XrInkF|A?b8OFjWCLeD>Qb7dUtZ?i-lWSHnqwQ8HY*{ z#m~lhO_m1L%Tq_j>!-hv@GI?G2+wzOuqdrkan~_3ovYPjXr|#^KxLrq`5*LIOnm#Z zGbw9N8*1~CRChmh!Y=LOY_1vva@H zb+u$50e?<=n1%W#-V+MH*LWN`XZ*dyH$5P+#Qgm^XJb%)FNjG9rbyB(r=ZYA1g%hL z&^)We0g#t4XIFv`8)O+d4Yrt< zw0o2otO@gI(N|Wv?8S86tY302ZqY)w@>h^p!*Q~(@v6A1Ea9M#7O_^zrT(8+{Ccxp$Y_=-Yqd#kAhf4hYs|wC+&h<6jj0&Q3i}}my-(HQA?yyz?G8f7IJoIeVl;@@N!uz{)5^PPTjl(TA+01$;5YH9+7!q*oD;X$xfB+f9J7{F ziZ*h*zI=|BW;H&^Zc{Z%-x6PjqmTK)L-9B zB&6~2;XU-XEkm?=Pv(zaLQ0O@sqrzsf*U1**=np~%THpsN|neu*6Cx;b;QA55LuJa z0vHWa02mJj#xd#F$(dm3FqHyaH&h4a(!;o%l?4?#0l-L4O2ZrH#!H`46=?NdP7q zN9~6gIP5eIFzJ0$&uvvBk4eW((3!DTc@=&Waqhxw{_#?pEDY-XqU990r12Gj-GJdTq2bhopL~)W*HFY{z zNMWsV87t;YzpJurt6dM(?&^^Fk+*P}%Q=7IGOOXg5HtRT+)ga(C0O_8(Zg@)|wUM7!S*9VR?@}|Zy@r^p zov#V_j-RAN=VsXWceMbwR6d~V^6Bs6n#Et=5N+*0Dzj_M?hz8EtrtzwthpmU-s+Ff zMr7kVQo0k@lOWqfOCZ%wzY{6vvqS#nP4`dWm6X3dK`sbLPptDV@LAzpk@VG@{?=|n zSD#!Kg;h6Qhs5c3$}7|{ua>M}Vb_74VI<@UXk2%LY96YnKAvGtif&SLa`ic#NLHky z^F6C*u1|Tatv&SL^z8v$r*Hn1d%{l5Ck}zR!E7L8H*<_ARGfv$s6PLV#SecoLK6C% zL)fLOjesdP=KK3)lfTWE2)}3Ze6w{kv)cv7W~mK5w#5OjYJBxUSC^=}rLstoMF>^% z&RT)meljHnPwwzu^yM&M)_>9pO?qE1bh7?+2-h^9iEeUw7{nvMzn15|+nkfT#o2k} zH*GpU?~bjuYt_5Yj+Wh6b+E=M|J~j1QPqdm<j-bWu>2lkbt}X)0^b%bKIryxTwuOFly+*?c3T_`aDFA8ObmiG= zxB`aw7LR0a?7gnEnjZPg!+w$9k82%(MidS*LC7Df$8$yc+~7I_{o`RV0pv?Ok8Z z3F_7q_qt0D)u%&T7q(rpP(8x>w#Hoq|L&1g|MhPZuOnH+mhsSCDdqjBVl!AF@C3E6j@H?8+X{A&$O81`CuDsh7`cgXHHV8Amb*^opS zHbfyR`~xGx$92R!^*>^&1Dp9+`=ZR@Tp2Q4KKjvB`jgb~9`u{n9~%Xl-*1H-$(#&E zwF{G4dmq)nN!4-`E6>4zkK|!og7Eg!;w4d-J zpV*TJ7z7c7M*PI`Pow(Y6LrB9Jh$&2w1T@A2l5rYzd`*Pm>E!~ zA83dc%2DZPks5G3GJ>D12a>a7ZQ4fA$>x0`dx2^_nhD+BBhY91Q8Qwt<)g^QV;xOj zNJJlqbMIbo&9H`q4<0v1b5HSL9(ei>F3Ik=A7V=qL>lRCX6aP!4N34DaAVmlGUon$ zWsq*!Fy4ep5sD&wV$GfpNnt3QdyV$SoWfC@Cfb~W1&k~Lg55TZBz)?VV)Mn<_Y&KL zfK;K|9fM~PR7i^-CGXW&7di>5&P82F$su8F1>yzUuyLfGQb@i=bI?a1q5baFr6(Y@ zrHT?{%$?b#07d1V>!T^Bm==Q*#J4fXpN@cozL{jRw%nw52`QBW#|EfBDq9w`e6;I_ zG^ZT42`o68lp89*X`PktW#6%MK5J!)wxqt0<==;85{wj18YJ;b*a6lDaPG-eoG6#S zRThXO+p=$oON%D9Wv<;N&z_V2{7fwj9+qQ3mu!8e{NSI1XR8ZP%#nWOck0QY@8Oxu z*t9WYVPxaIm8MZq;2!|6iqPFfNSbgii|!0W47zn8&7k_H9To-;`vopbtJ6LioO4fG<1@v5tKu|L}5NiGs>IL;qF zAVBfue#))|WFnf6ZziFWH_B2VkJD6elm#V6U?0EcnK5viwXh#HNgW||9tU0zorRwO z!@Z2QjvTS1*2ZzVJoSa^Nh#wE41c3AfHIl*wI_1fBU=?5mF1?>{s#D2E8RWl6&kDT zYDDPHog2fxIpT5U*4bhS7^H>&nzuoHI?zsZxtd8giY&a^nBk#*7a=DSLKwb5#K=Y% z^l4mR6xbAyHWP6DnhD-emcHEyBXyCcG1<{IYN_KNzgUc>5r_ftO8?B<$y_&^FcOI- zrH|g2K({^=v?zrna_Ve76cji#dAd3~yRmy?gmatA)y84WyClf>6{r~`F6Ynfng^8` zcb11Hf4jU@6k+eGS)ugYYx!$6>V;3VM-FQKbcvNTRy}nzi0(KjtzI|vCp+^eT{js)NQYDIn zvne~&QX^t`3;R+_4iZ0SYCiw_2JY!Ap|!WjkVCc!uW1x*Lv^hXMFHn{9@LA`lpD>( zy}O6n74(Agl(+b_#Au_o8$tZ2crj|j zoNF>*sHD`+oovn@p{DUD&lQB+Dz=(UtUY!_e_ZkX2%L}mV(wBk<(7ozj&l<;ZvTj! zN4m_?>3*49X6qQw=zwyVXKO;dp66TD;q!{Yf|i)yEqDR zS5Wg1)KpRtp&we{AzjDxeE=EtSw4nkVoaJd-kJR_r*o{hLrJBwEEuJ28aG3(AW=&M z6|Et`z23FG$R519rh89GzcO~c$fm!={{bUt>TiSvGT6kL z3-hnyDT$y(njrX^QmNO}BKB^fiu)U>CYtDSJK}`e)if|7d~pq8>2#9n-k@#ms|^aF zgWuHJ5nHvEQOu-XG6MND{6ca#{0mSXmN<>mEsWKJX?yYrnt2~`h!x)P{`dMdG6t#& zq+IvrjaE$T81ffI+1=Y&VW1SxM!+U`t*xh*8krq35rkHLHf zCP>}Y4L|RgiZtMPngw0GuYWk3#nM^DtJ(y(nqB)J&W`bw%j{43gxMOkvpbntXqlMU z>tok3yPAkYu61&$$u#qm_a9qv1b)>(1N2{l|DB|Zt*{Vulx+W~G$&17BL)D1coDLj z=`(TRE{C*p&5jSnUysLVe##6N%iur}LiT^WC#s45kOB3blCDeCVd7$-e8Kdu&5c6V3NF|S}$d=l-t3h z+sm+!$gK4g||4vpc=DYsnU{uw~R_H1r<1O3iN-GI_xl5T=6-k0@K;Ec+B6dk|G;95wq9awNU- zcKuA-fa7}z%1)yg_6mh})c5S=hY|lj>t#{2b(pPe-&&8mNnZQqVJB8-I~3qsVc0sdX? zMCDk`|H?Fm1s$C&T+Y~f;4}GxeyiRxP7>C@`D?6o=?Y%msU6gB4Mqv(LR{Ld3_TxYwz?AqULgMXi|acZt`egDJ)Fpk zh5L0%){aPg0Vk?e=cCdge>RK%{5MXsrNWNFN|!f;&sn=1pKC+q?~gA$AE6Y!$vK=8C4Pb}L!#zTc zyrDp;B|tTG@4=0;{*t?SD{igAFP8tRYQ zeGpcmopIGb=L=V}k!(*>BsfiANElvHW|e!|nZFa>aEg$q%bF zZ}>6V$G1Zii8ZW!W)0EEB*#8mCA?Tac6t9@=-mlM*t#)%i{JxGvH#&+A7 zixscp_)Db*=86j|U>)gY9n{u+;P;PGZP1>wMqXEoT&b^00_0}i(`i-p zNT{pJ((?=O`f~QU`MNu2T!P+lSkbz1Dm0F| zGHRH2v>#~*lihr;2a>ENtphM&zoJ(@dcW<&q=Qn)G8LFyP;>Wwm=fIaxRk=#k!_s7 z)wpCf*OGTpDK-whK1{hi$Yxe+*xw56;B4C;0|{XrAa!~2B-Aznb!*{SA(@Ht$#%#e zZzVXbroo{uOXQIdTXc$V^Y{2d_u- z*W^(|sw%f9j^{DlcFO(3+7tQS`y%iVQY0QxQkNXR_yPx@`jix4D6B0bJF0xZH+N5F zzhP>KZZ65g#tgL5m_{~yLToZ6W*Ka*1daZ&+?`%7(=*kzjIbiobp4cLJrzL@4>;9J zCo@vUjNP0-toCe>|8AesXViG{&@^0+yGt_`MM?=}c%(RuBx(OS#Zl8w<(?v7vQro5 z^^p{go^XzYt*)|{f0?BQRZ+1DPR4^*D5m^)#)0NsHPpi!G{etBcFT%Uf0>XzeS40d zG06Gb{U0nQ2Xx_JhcoNymlEpE>xZ;wbfv+4w)Bdt8k<4rh7CS4Gcmu=tZmcq;4^2X z106sxdh(UV62f}uT#cP#M6>dnN_c?X-~rxa?2jzPe2z8(LPXcIKNL{jGrDkTH-Z1k zC=HYEVne(PXcjP=8igcr8&99mqw655haTwHE!a<~q_|EpQTW7Q}V z7lJ?EBhz4R5;zHKSmiJ&^xk!9L_yw)mm-8Jm;SJgZi}SsY8=OsbU#gl=b@qO|CF^l z*IPJw#&?%WxskAJc_-TdakA&unMa{{GXf?SJ?}$M;R-Poo&glkNuwrNEYcW6_)NxU}zU*hhyr(#x6^ahE5$HXj5+e z`KGYM3o5U2FgNr^#KP*g#b z$S566T?kRg-YT;;#(nsCWypLDOerl+j-d(dRj~e^g+I*-K9w9lv{Kcm4nl~Yt+VDv zzi+oiw(vKrY0s41OBZE5kbl7FBh+F+*O#E3Vud^?ovyZryd{g76ao zs{_LW*Pa$rxdCiZf#r9?cRn66#}Ym{KI`6|*Oyso^}y9va#66NlTH&k4EP=r71-+^ z`u(NuA4dt<_8rBY9mWq>%3jLsEWwR4Sl1F*DNSG0I|cfJ+#Os1RrooTTE26Ib~K*~ zUSZplPrg5wDJ^<^S!cvxc!@gUZxvqGrIC_9Wo~M?9u1Zr4Y96HCeID}?&JiWo@GM@ z4pA;|96p%-wa@GK;VpzyurK`Kk6Jqxk5Kw+u>1CXU{1B1SsK~-VQ8BW`g>r^Wq4-s zJd&b0Xr~-@sBmz2FY}Zlc!6s8jGFirD`*vJ_;l&CJw`xTiYC6jAdUXhr7#QT0nakA=D#}yU8b1m&s-bn5>pGb_f7xr#&q~doDEB{A zm%!WQ;~3QyCZ|+tE&EbOHDkP?K{mF}l z1QGE@1UE(qCyrDw;N4437r-RV*ACV}mqKN0MTRj)`J^g=|II9e2A_o`bB7xHlG%Yw6`30-fFTccNZ zqoZZMpGj{|g>P`n6DF0=#{AMS8eIb^-Ov6v_Qu%SjK1zAD#y&S$n(F>yMPd6HvI)K|h3*+C9T zI67gqkBA`$2QtZ0>NZ;Ap4d@uW2gE1b6IcQVkrmzlKIV&y8vIeCA64o!SO;RiRht2 zDW+x58~lg9nZJXY3?ndTpyiEJB^00{I$b3?QPux9sjeP)lO%Tv4)jOuYL%o(V(mk{ ziRyR2$41Qgzz(GCyTN*OR~>c zZhIDPap8l%+J8X@NEuPObNOrOR%#_7e=bNcInL+F>Zxh!0W_vHVJ`-2P_ETIn(|za z)jjr-|8(ZxkCR>refF}I6g5}u$(8MfA%EXj=QtLFk+k;JyeN4K=0yDY@{1zGB4tkK zBw=?0G7UJ213^%O?JxC6Ve}VtVFsFh2H8)$2!$O##vbQ&Jw2!#XVX4*>Aba4S!8Bx zc@ZDoKzz0c2|Y7J95q4!!`WD%%4`noD?wshp?`6mbGgJIxdujYqV~z>eoeh)?MOxyj#Zv{1Tw z+|p)SeDnI8J_io!?dW%B;v1MRzeji9JW3vik{!8AT)N9R@EBqXRIz#gTJ|BAtp8Hr zy?`}rXG5UA&lDDjdQI!PimTZ1w`^*~pvz67q zX7Fe3N0_?R6xXrx75$(K01D~{V3v##H^)gZbJ^@P7aHIQ zB(bm6>j&f*4eItJelw!`Yxz^PUZnmDhf+_T#`b)$t7EXxF((R!`&?$)j-e0~)RZ;M z;4LC@g#6Fhgw^TNXF<*(7iVSzMtrC$H*;!Nn^I}hO=;m?jAGP@N_NQXy*hqOBXNj@ zA$!Z}bS=7$qf4gtwl&-*c3aT)d4zDLByRLN8x(D;lKw?zkiEL5y6Ut=l>C>ClQGR? zSMgw{vx>KO4-kAct9q#iu$fo+wwaA_2xoNmRF3N&;7QGxEdOjeFSA^Wpr$^zXFYnD z)U#giv{zYNpPZLKB1}1Tq4oYs%58edg!+!i+i6JY%h-LVAnPLn7@15n8edqf5npNW zvrF?^*D&DJd&;%piuKU$m>%)iocVQO&J+WLL?>k7SWUpC;Bn~rImdW}-_G)0Gp=V& z1tykRmzvLP(RD$kr0vSDUe>SAQE6sS&%4mtBKSw!CmxP1tpct2xVUFH>6Z%77uhMd zQ)2Fdk7mOHn!}7bPjuQ=M&Nma5-ICA%}TJ>6tD<>Fyr4@0^p*ah0dDI0w*k(b_Zvv z$Oq5}K`1%7Bzm{{`F%7uVb)#(2b0ED9QMEqbGz+bo0s$BikigG5GS1G-3AAB03u?* z9e3UbUm*OVF6ksj^@9}g1C1tZ%lXO-0lCIQ`e>ny-S(`@%Ca_(n!+iPigKzyI$?bF z1s6!()AYn6lK7J9#4cfm`}U-cBTK`M@Jt*)IWd(U&!pFm+By~)r)8^fQuxb{05Zz_ zn!t@9d^!D4yOU@6hGAA@J{D*_Qb89hbVrcPERoUpEx%hqy)K^b`COgEHuMQK-u0VhaR)nt{qQhz^;}k|JcTc3H6T+ zOhO!o+s5yr20U{xLlB2X|MC!5`)OrL?5+1&`Svs zWxXcIF*~=h>OKVjR8AXW=TlwuH9Dh%v8OE~tUp-q2oR|ks{G0(c!ffOKW3)09BIIV zZgq~LX()1+C4S{*-fHw&97jcqzE@JbTj+9+^xtcvroeb-u&5=pN5tO2=HkEXr4u~k zGVaroQ7f%F|HAoaqNN`D{7*|cmzHQQ*p6qq%Qw9UE3T;41QCBY@-OaQ=JcNQwEMsC z;c&S&i9&7eWaoH2M~GbHb=yM)=2$$}`=g%JnB|13)9Glv>uXRXu6}hSbN*hA5>98u zx$Z)hK>&yYKZcLdyr%v>cP_nWILEOC2t&q33;X z?;+?=VRDR1?yII1=4@nY>d`QD^xBDutJNcnL!|bCC@4UAPJlmCVIf#=W&C67XtO?A zyESIBR~}_gFZcGS=^dmXTo&NxD-cZbp>J46jmKeBR~lfws2nYpnjcQAD%%8~})=6kLP8>!p9QjgiD1{K8;?@OJIS*~KHBg{#p zV?I`W2-&xIaQtk2fiXZO(VU>_`>ob6jd}RyJ(FhnUPe13;U_?=4vdM3zX9IjVBg>* z192g}JWdL{5fn1IoEA>3lAo+_)|sp`6;F8QVtX*FpOqIcU-D-q>;6bO3$DMq@4X(0 z9e&oI$S}xhIOPEmHh4ns{bzuA$x@-<7 zY(z+^Hf%TO^LRb)fCiy#bC4QOkw7>)Hm@~ET23+GcfMWc1Lw8?U&wkIugWBI{re)8 z93seJDyx?q0lxcU)ZJ}7G@fwefw(k3b_brg0Y;qAsV8N&n~>9w-5v2=Pn<0vusR&q zZ!8(*SPAcCTu`qfsn*zT%mSWXjim{nIM|Vb0ZGRsKCAYqEl76)RK3Y(e-5>tTkFe< zz-HR~dUBm;*dM9iIZ}Llr%1H|)$uDQc)|Q2{#~=50pBMMxPKsHctLfy}rgPn{f7svtx}_`n$g!uO;9bYIjGsqm zhW*x)b5B~A4U1u?rxB{Fr=TG~jPI!v$}j)m&7jFZqK#@k!(GO%v9BK&*VQw&lvnp~ zZ*-cI*W6abZtxzLJ|J32`t}mv1D?oe-}1?Ejpl4PPGP#5IcyTBo6avALVxF#(r!d& zg`)F-dVw3@++S&0f-{NPgUg!1;|Y71^@+qerRKSLFNjK$*(iSL)xQB3r8Y`88BJK( z4L;pHYWWl_*^)e5VB?>ko_*MqJ)xZG+^r}d#2@2B8TS8&A8}l*-AF{hn|fg`+$=fpMJ{Gw34Far?FLIW2mn!z)Ihm zlivgM+;I1n8zAh7&Q~Y-_-F?oIf=Ne5dw3inN3k z`quqse*JI%%LmX28XFVr`QN`q?&am`G1!~e>PSL9a(^C|15{zY8Gk<$@*edF??khd zEYg3BuzUpOmT_pBrOa4=+>O=$lvS%sr{>qBnU(`J-EcM0@gn)OI+X>yyjsrq^c%3w;dI)U{u!B6 zd9+bqQ9$&88`e!DV?}@8hh&kZz9-s{tXx$kev)Z@jb|hk>-te zxLO_p{fDe*go-Vn>hk^!dHx0RH+VmMjure*S4(Fhr{j+-ky#)8rnauIDebYJwmOS) zKK8hb5XMw@3f4x}jmZaZ)`3t<2l9hIld|;i_@84Kq$)ha?Ou*K{9Tn;FkdUV!BfB+ z3F+Iz-E1GJ-=P1g{>DN+$_x~=W2Qzf<%}ce+@O|oTeoOIVnISg8=WBc3+2ADdZ(XE zLC-%31)L0JbBpX@(Ep-rXdI2Mu~wB#uX|*Y_Qymuxk?{el^ylxKI3)6Z71 zrE38n6_XuTjRvdakC~EPZ^4lJ1V$~W~t5Dw7DAiK4Xe@Ko{s(>06Rd$zhY(MvDX?sBxxXwN~z}!YZT}| zj|CqMNwIPx5{OLL#KSEneh{-qyS_+l$ddsRu5=Mm%xJmGDRQkAM35#WhvdmCB;&d ztLUe4nevGwY+1jsRime{Z5wt8&ifT*F)!HYYE8bKdCWyO5tjNahivybUhbb<~219(fC{8iXY zwgv-IEax$i+Qv)-f9nO`oqhi>q{r~ctsg@FJW40x!>xUzM<9`LWSN@|cl~?D|I!xN zSO6Bl_4uDtI!4xiX$w4DT5x4nr$6p<4()-;?C#*FFJ;q^l^DI@ly5DU2C8!VqZvPk z(aKb3VPMM3{+qVQA5MH_G!%jCm>Nla@3_A-SYy`rQ^X&iP8L!0E0jE%LuUw4tQs$t zDVr`^YdMprR%5w5^ZX3&F`(dPP-)ee6G2E~?!vyFi`PQ9^~#*3YaT)9le)~2fiWR2D8 zSZnjm$?xuPG=oxG%ioLLsS>^Mw${6w(~aRwrS`V_yFV8PtK;qMXf%Ms_S+AO-2C@< zTo!X3s#Mso-z<4T;68F}g-<{r*0-YHLA1KkvcZf&Z0jMQ1oQP!wt}MdFs{1Y_3&4H zY#R~pX3aMu-)|RfL{~y-s#$EsYnT>q zC1|_sZ6)dlv2Q0CCs=GJn->&sr$FoWwo~o;*mu&LW-WHo-L{K&GQ6($b~62NICirF z$SrrXLs;OuIT8H(ySXqaj=j8}YLizo3bl6)4)b zv`EWB-|m+pW-SlO8n)pF`j?OggLn0WGUF$8vmCPp7ocpf*1Q4p4k`O<>kww$v>UQTki-*!(aOLJdM ztD?`k8GM1Pt66Q=quM!)(~;-wGm7f`^O!Lzwza7N3EVTvcTaDYl!mQuW<*hwnCvDn zvclx2VbvK?WB1=O#8ybq<|%mZ|7ex5Yg;^#B=%W_8 z(_$iq4+o+HM1rg)b~)Eu#;hyC0Z*dOy>B7}x$pnRvsQ?l=W(4(b^Sh*(*&kPy&uio zEGoHPRn&LLup_53Ru*jLIMA=X<_)X!*q#VN?&u%5*G017Pn4Cc1bNl_^5wUOOBSN&Q zK>fkJLcn8MJrY|15t`UAH}$)2ULX5tJhf4`cWG!aEbg=mQEE1#H=`VstL3)F+GRwSO38@J@&-Ub1~fAbL;U4cacLo_>V%+`r3zgEm+Y5p~b zQwuMf&xbc9$*_Qc&q9vihog!&>gCdd+b9F$8^AwX-?-Stu1c-<oW%)}n0$R98h#@b`{wsM z&#=uRZTEU?EUYa!H@=02-;U~!%PT0>)8Aqun_>aRd{P8c2bgzc$Uej2Djarf46JR& zpx5L-6P_e;>{J-PiGi$}0Ai z;zQ~)TnbPJ78TDYKgOr}NrSzz~ zG*me=_>soUu_!sD-{qihyu`veXEN~U)9d6Gd5SlxH&UiS;R)ZWw@Ctz(F0xA@Hrg{Wgu*pt<|olF`- z6bOv&DSF;cWf9jFNh|EDq~1;!Db~W(NB7m5Z)Yl_YD>)&4m1~UXB)<9%UwqgbnkBG zI*98kgB1=9Y3`5%igneAWd|_RgN8BAYlQg>N8-Lb<6PhT7=G7|!j!z#V&8miXXKIo zC(44k^<{s;p;6qkFLaA{`ee67;8MXB) zx^cuz@gijrwf%nlgX3iBMaCU!N9;*ct{cx~4o$j%glN-b;@D+@;QgL*bklU9;uT!$ zeqUp}X|{gss@(JbK>tZIa#6xe=We+D(I>$cnaGOGquUh>LEI$*(x?CF;}P)BXoU#~ z134Iv{?iBc|4knap(89RWN$Tl%wtA+{#PG)gK__p_52@w{7=^NfAmo}o+E|LQ}ouJm7hz;%ZI)rZ~2(7*avZg%@eAH($(|LTK) zUcRAntvlp7r|v)c*yxA7l*^EB{6`;2Z>?5F8f*U1$LDbRfAq08Ris*?JK9uxFjr|b zobiu7jusmn|9|SE==Tpo{@vgHPo>z_0!Y=&*8(X`i`If@U3S<0AN676S5Hf74E~I- zKm17=UpX&Els|LmS32Rzmk$%E(o!4~h4vo%`*8OkXbx2X19jT}q+un6v%KYazoe|? z!Qe3_Kx%bZHSmH)thzS^%dZsM{xwV*)0)YTLG%yosBU3>;izuq{I;ylrBV!^N*mUR z$DZaiecZfXO=H_kVIK`%{{@eqXZOkx#S6wuR7-^x#YjfA|1cK_(((5?!l zMoUsV;ot!4yAC>d-M|>ld0gj$5R;`ME7SVb$m!bS&_)qy_@#;lPg$xN(=EXMGa7T` zX=f8X*Sk(1*qUAQGgfZ*dlj0fosaK#tpQrC|l#Q9|_67h2n?xjMo#L|k9$w*G3Wv_Xtc zSVDH*k`?Q)!iE9*c*)7N+-d5ke}F&{?TK|$Uu#&Jd#5JDSnZrnE{$iap{kYrF2Q}O z%hqF}`+mAaGhM_^&m0kOoO9tb(eC~Dp_|uNBD1N}Z$tBZ@Z`nA`8AJgd*}vA6N}`0 zhycr}GT`B2d|wo*i@Muc(0KrZGeur49--~WZ*AhN9NG>HlUsORh+zaBx$S&R6?2(; z+B#an*ml>;d)V0~YNu#HnU;#8Hz(PCcBtj@r(4wL_504VeGdVxr0bUc_s8B(^awW@ zWqwc~b^}9wlO6SNT-mCKhU`G}#>*FIpR;H!LIq<@)&E1xIe%BW?|FV_$7Us!q+;8y zxMJJ3%}T|#E4FRhwry7`cCzQ3bMNiGx4Wlj)~uN||G~4?^TX%$)>jSg(WEB|hLI$p z98VDAsXs=$(k8W@@NMh3pW1&QfXtd;1g@y!BHdNNBNGe#BG=7Gt3WH~QZ4>LecOW17?ukx=kP z*o}+BWR^E6;rXiz8zJ%Vedk+xYs{%R_SyVwHV})W#mpW6Ow4 zmE+IxjU(>I)|sd(=VGbNvyTxe$lfZK=JCx-&&T#1#A^3psjZu|$Bq-3YR^Bpx9zd> zDyrJ+SZe!a?Xl}=vfBG`eEZ|&u^WI?;|ndl1H^smfpMKMM}O1JVt(qqJGT~ehTetK zdg}W;RTG?hxP!R9T!Ei;X(sr^@FxB<2N>AD->f=76<`MN0QjHXEM@a^{lTz*QOvN{ zOFdeTmm5N@a|d`}KX2}MuRcfJhm4UV7`C5SNb5}cGZl{0%bPXzPA4kIJRwdr}1k`qRT%L zbl&pC5L(CpjKezT0vd=;a*eJL#lpRBqhL172Xj}}P4Z!D!Xkju>bsGj6_!^*7M|la z1CSoXMFSy5#rI=jO-3>RF_n*5@o|lqh7Kk>6ed5E&q6XB?;ynwk`X>pW`^K-mJr4m zzrc|GG{;=nu%E!@!b{+!$27zq*fh&abh`=O`$5|knwxC_6+YwVvjK)3?Rs&tpN9e@ zEXdLF4BHjZB_G`)NMXl3GGwp8BDQl@P2Db0a*WsZqqBsy$cbMIT_Nax1Kr3ObF8h1 zrnUqhpOpEtV`Ww|;+u~XJWhlY+FZv89cNP6_F3DJFL7yR?}Lre6f7Cwobx&m7nxVJ zWO>-6#pA5td-=TGj{BYeup}$Zla&ZJ@qE#^Y!Pna)-kL(Gp3F9*%&+N;lB)LD8N-c z9!9PmYNK+IZOXZ@|6ZnETR(FXMph3E9lq1I`4LOm0uiy!6!ZwEeA-&s=v7fn7h04) zan+@DR_Zg9ax)4%!*%orCVZ*t{py6h8Nsg>QZX>TmRdVFHRVOo2-MJGjgp{vqU%LU z^SV#g(4tOnXZuOV(Ain~#x`~oY*RiH>Dl?<3)?QEQsF62sbak5UFU)H*pOD7^c)nf_fl= zOKkv5^7nk99Z#Qep9yRaq1}OF@QFF_&o5ywC56DHa?1 zK?ENkvQQjlR9q7uHqP2>A$Y2OT*7fK zp$?&lu{yYy7wI--^N zoRrO2moDZ;?~}G(tIIr4Heo2GBC!TKWj;+DYu}ZC_uJ&A{o8sP$KQ$TXT)v&0~vIp7-VYh~?_Tl1p`I_Z8+c<(kW5OHFO}mG)8P+Q*X1ZEN>c?vv%Zk7LVS zFO9)S@8$Z?QY(Ep4>h4O6^2;jD?{85f83!8V+yI&G0lg%l*tNHj`7tg&xiUP#7gsF zlY!afPkBY3tSpqq*OpA38msQ9Ese%iRzP#AWleO4Hno!7-!GSn(f4+U)t=W&K5koR zc+cUJzx!jcKDJIp**ZlLM~)h!P{QL*V+mKCfR!72jDlaBstm+c&ptfuR`jw znmc|+F?h-C_~!(&aUV_ocAp*BdoNl>V8K3j%pc#l#eC|4Air`Q#R1$FadgfE+56;= z>prqK_uM8`d!6$v9nn7aqC?gCP<_#_MSALk4Q2>d>e%?@%281?RujmQu?v@QKXBs` zfXn(-52-z|yy(T=YY>+PXBeyJlK0AGdk0G6Jzx-~>&kz4ZHK6P2h*-*m_HIh4;TLP z;7n8!jq$xdp3`dU7fh?zM|d3$lyT6g+s?H|oY(%Rc0fA=V?X~Tm51p+01bVO`!?b6&CBN@FqE56Gg>olc+N2feby9*$Wtt7jS4?iyL>jUlbX8EU2gQO zX`^Qb?CqKziR^WSs6TY1vva4@A6V7XOOeC@A3!mvqjKbaUI+ zwc+aP2Y0kgha3U+D0ZCnS0oro9|{~l%@9BY@Up8_Ww)vaUEVu`l2F*&-JzSw?7TR_ti;bm(~{Ca<*g7@X-AGDm~f@g=6Z~;woG~L*ZtfOxx(> z18vDqmn)|b`OV)|iy`1xE%`0QpzA&gXku8Zzy;naV?48Jg?P{?bx{D9rvq@kVXv+e+&}zLuUCB24Z&g13d-RCb-TEGu`*~Im zKjbftSl_2riX4FlM9Lnz=9Rh~JS5S(aevf5lzJfa5l4a#(1DZ47r0`yL+7?5wiiam z7rGZtwq}+S(8M~T7l|?TBo)cY?oQ&34!5wc3X#^Z9mAojY!GC_mgEcIf4T}F;D5XB zPgEAwAWKq|O*Kh$B?R>RJ!9}rRTaatiStdcRy;jOv|~R#%CP^?hlVfsLs(;{$Htpw ze+Q0HQ2Cu%oREV71SlP(3SAEKV?{L&39Mt|jti4b!%vISOkYg$wf!_vD5PAJk4qi% zaE{BA*JgoKCU>A`v!c2iiStT7@71(yYl*Y-YVX)ts^S5p^U^%MAFN{nUKFyda=%wq_com~w*U4wLC2mvqw1;4n6R?Te3m_@E?u`w|CO_a zwsB0Jt-DrU;ImbdKu>LhKXWm{2;ce+iSqtc+_AJL%Itx$+lu!`raNvtFD85u(7PR zb`8D>TH&osc*kFJ9SNxOX5R4TS55@?WC}XEX~e3%)AQWEIqAKcNvym1(i<1zsdIl=;%;xSzxpm` z&u+Bi_GdZ+%Zru^{mk)x1jdhHr!VTJC8B>>Sev{vy->86zZICCg2u#^c%ZZPCvJ*IT7 zDCrEK55@tlw_Z$nsXWxUmTx>O#G$TNLFkY>WcZp6(sCA>l0}An(S*K`qyQs$amFF0 zWZy_*&V1}y^&!*-eQ>Efm3640UaR*}>7re9!M&yQDnfk?ZjupeD2qNn01-Dd6)AD` zd^#2qB-_mF6ili}FcV=8M+w-f@=ODz3}y(Y_Rcq@6+&Otl7qyyXC-Gvi`u=CK(g;b zN|vM=D0Gr!v6w_6H;-^ht!8z~7y&Co0DmnIXPSOQN? z4f`jGguhrM^4mhNZFq@u`k{?3NY4@aPGHwBm3Bm=))VF_N{F?!jCguO$iq{aan%}v z#>0G_NcP|0OMu}heB?*jI7BD1gc2E=>~}<-Z`=2zYg4YSmwESzX1qG9)4N_(`5JFW z;uvY@`fWwTyp zt-CKnYc*9vmrx6cRLRM9IJ0OFl4|(lV`y~FRaw?ldyp`hQ!^;#(5;qJXe9{wy)J~j zgjW5ORSCVw!sNVLG{%t*LSW)(wZ){?=F%P-N@OZ6)&3x+wui==s7h-ysr5C`f3Nbj zJ}EH&q4Mc2x$@!BwTTNV<_oJ_pAB^a?UlIKJ4bc;anaA&Z;4jPsy)x=HrwLuJIR86 zt4vv?OBAVf(ScbRT}vgRA|-aydz>1qQ>OwF_X0mVOvPZ__7|}})lWIw2U1+5pl9qt z+PV7$vv~PqX+QNdoG^wmZ0sUr_d)_|!Tl*R81%nY-;2<{Chq=3I6HUIFu=tuYQ3%|`1iTs1eQzCC7G)BQ zR{yMUlrm(V<(N3%v5!v!1)BIq&iwNj!I+RHRa_h{l5+X$A)nIoG+vXVQ5}LQ{r1X? zFH*h0jdV)%)AMXtRb4td_%VO>A2?HT`}5fA7^X6X5p0Jg&F&grT{|w zaIOUM9M{~cX=v8|g%XiE8>GRKs3*{*#s7USw6q~gCCZwcq;;MT*sP~)c*%16GURWH zQ1@HMMuS9r>xTtXbo->8-g)}CBl%tKx{aA41MbcbSU@*Lmyxnk+TP;_yI{&Y77KF6 z7KGhxBYM;cRXfjW0qB7>f(+j={CIts%+8|%9+=yIg?sSS#A&!T1BcZDHgC%aJ`;2q zo&(xA5uxCo+pHlRj3rjEp%jL1(+_4=em~Qp3|Ok4x@UIIh0tIShHXIssL^sVg^w*{ z-LS%K7{CD_;m{2DfB^Q`JPfy(Ose;RaK1(z)i%<>?e{u&=3zEbUYyABX;{c2s0*@qN%<7INP!DG!T1ls2gdGTFaQh@Ix#-vP& zZQVgl@Q!9JzV!EZsoT5FoPaxhT^z0ssQCS6po{F8w&=w~Wra82Vb;?~v3=jop?Za- z&x$js}N!uuWC3q{cfcu%#iNJ-`-V57tg1>+M$~--;O}U4v_uqF;!ch*}~gEoT6$c)+Uy5oQY3qnp-2LUGLH~mb?GT@W?0Yf;kfA9%3a@r6zhmNI8k@@- z2x@)7{g?EIOfcMyd?@2I&Qv;#-EI@Z{5+Y?kB{ObaizXQKDRr#O?zy?ctj7MjQxOY^9&sCyqAbe)%)A`FZfLtYfo;Ot&n5@%T8y7&SkapL9MO!lcel5^lSOYPuo2(bDum8nN9Q z!dvQMS>{D%Qm(CNfnKZEf4~buGgcJWmbqEx2NPCK8r6G=ilT6IsY4>;${ctZEVQ&W+Q{_FB#ws+noZ>-E+&OB}~F ztx%e#9oYD^13Nor5iBFiQ}|H_<~)S~O0924JC^;~?R z(jNoXK=wH9-k1EXd(xC~xHR~=n@k>nfNLCr=lygOG9tfq6EaHv^9?$&X=T=NysPD_ zfF1Vogz7MHoPZEKX-y3aEZK#$=*(Ai8K`Nx*j*=hTv9+zX6t4tk!?E|V^5#S6(w3l zsI4PPJGd{RYujH|)I=$be26u4OYtK7{@3{B3mnX*%~Q)USSsClP$wjnYzQxk5fC!D zYB$h``(p9?xz5ZAxO?R}rHqXt9Evf+eX2nGxD5(+qM0trmi6yUu6DCyD8lcy^;1SR z5bu+X;<4YiVh?WH77R9;s=s+`gIlbvuHpIXyzTgx({TNcgO^3w?95+iU88YX@wki6 z>|D5?h+*Yt3%?4a zz(@x_59!0Sly?kf*?rd*7k+bz4Z1GRMfx;Si}!LBB7l&G0tyiyAjY{46_?IK`#CZ| z&V3ywADM?yEIvrBc^$4kk%!qlGDz=v9ifkq|9Mz^h&k;#(p)+pXL)3Zz3n>6J~IEy zaf!&;tU~xg`wm`iXfH1uQw;e+8euXA|bQ^{P;YC3u^H7yJDd`4VkcL zsR9Yxyu^~HSfiXT;>gg9V~XK7iDF*8xVrX0mAbj4N^KKbJt`^nvN_0R_c3HHQ~*oI z90UV^oKZ_+LJjjarJA*$Akuuol=wDv%9xBkOq|OYi6y0b!jz`kLx$yBU?_t=8LO5@ zuv##B>T%)`M+oPXc~gDLnDG(Mro=RQOHSybH95z*1W3+~nI)3~#++qdLf$FODNQ8f znC*CV%pEf>+boolKgL7O5~(L`b%IPdp;eYB3q94&zMt;n8abe;E{DQJP%HpOMt`dh z@!@G$6t`_6B<~hc$P!XQPhK`y)j9O}TM>2+^PdtY@5{Rx>-ckUCJ?=F_iG77P^f}I zmw+VvVX<)Uvn=|PV-f!+L-|IDrKp$N+|!CeO&OfU1bpX0zAuo#2d;VVrNmGm})bs}zPR43&tt~a%?$Fs6_N}cp z&9H(|f?Js>(_A#%63;;&1PC;2DfHFcN zae%y1mqTqz{Wm-s7PJvLOhka~#UbKZ%F(dBX;SL1^h3zUz!BO9#Q0%^ZA6{aK^Il0 zM7)gsOJ%Wv5TvUNwl61;4y}zmku`P)@F|_BMGoq9&V`sZd!p-{L^e=F09y95hOba` z#Z4N(j|<`SR&KgOvc2Fe)s)iF<)y}{hNM)^(PVc;D!#}Gn=Dho!OBZ}@S3&M))k^c z0B-0fp)BZ_N4~O*>s5j+?faRdVX8IM$lS}Omf^`7`-Inx6Vb%h`N>Py_Sem;=;n?= zoe>W+t%QfE=B~%d-`>z-j`SNMp?`#E_#I?_KmF3?R)ym~;b8kRN)^ffsrZ2`0mPGRI2leBO<-;)ZJ?Sj`oC8E z;CJM1vFJG%i5}I5@@myToC4C=u~%s5pU!4bX^D0|v7UG2KzpgaYhGC_H~#11$2K2Z zW=HF7YD>0M_J_l>@5w^@&=j9SiM-!-u`Z~qvy+g*-(+m~S4Gz@COu?7)*G7upL;;l z$@h>pvR;5<5+o%qHc)nG%QC^H7Lp3k(+G(IW?kzSh#^~&;}5P~k{fIS1JOg!wBer^ zN*V^E8+soVN*qQhdzdZ2fP^9-_%W%z58#+OArciwoXJ=Ic-Id>=6ltTec$~zF`sB0A0Txkjte4Z>BT5_|jE{6(SX)W7U zal)_u@^-rB*eh-tTZKgAs})~}4hSAII_OJ0$j8mPxo#15`XxGFh8emy0+>4%$)n1v zggwshCO~g? zFLlM*AyIc@*J&V~g?XE@_q%EpC+*8cTr1w1ktb?u_U*+NM2xj4rOmz#z$@+JwCl1K zSMT=(Uc2K6x%bw@Ad)K2B|ql$$5ZGKB(B1E&^NA&(7sHr&9vo7R{0hl`u3BC1SE$9 zU!Sa(-3hCTr)!z6F8VZhbb{7vocAn+w2kfc_eWdc+2PZH4{Uri;i^FhL6@)i&1-WD z0qBiz%^>2%^~sPw`tWg^OgYa03>IPqh?eLJK}XmR2Hzq8NgRs+@kJ1xeTP5=4@nNw zSvP>*AMcj|1Xw=jh#y2BemWy03`imr7QQcW-F=3bp&-+hH?;L-Ug+0_{y zP1~e{f>Fuo@71O2)Q8ioR&&tciDexm3o;I*L%Cdp2?84u(yk(lzaC3Wd#>GPJxvtz zRU3jRz*CPyU%yCVtOAAuQXGg*;7?@o)Ny#?P;!p2jaffw#``&mK4a_<(hg82A0WAa z^|`S`{L>liA6$clu!AJj+OMg71w?0TEK_SdQ>+|Qsc>y9H}yT!iu6>;r(dj4saA+&Q&aY9CgvU(3DH^}tr)DMF)pUZ&m@-&kqK z3r_)*SMSDosPe!o*JK!9?F2s11uT}Ch()Z_(kEBi+6Nio!ml>VC)a2Up3!;FuJ(gW zm*~7*dT}KmVrFmd>uq5xt+(~-*2WqeT1+b8y&a&{QZ67(-7B0wovSuWtG1Xv4O+in z$MEzt)w?fR>5al|<~_4l_Si(&>(~dJEJ-&EiP&mQaBrbUx;2_q+G@cVZhYWBHr?J= z`_7jqqda`-s=crFS|#2b&jRaCXDjnwHQc#{``iXFy$X6gSFYt_qx)1d!E01x)C9NHGT5`q!_=z1T z7Sh&HDc69+RN6yoj@B{drU1c6ggpp|9t0CWeM&pP6g{vNf$T&Gd#2)uz5p9ELAK)$ z@N_n=K_LYc>aS1|HXd``r@AWIgek8AeVNT|!*wE+@+4|cM8OI9?MD|ch~1zGLlS1W z`!}XehQh?`5@(i-J_F9aCXYdk*{e&rlzzoKl~-Duvm9N#H0?Z%5qzPGcfGF^eaBR? zx==13cg|OjveKC@pR3$*`kg$9tk* z$EImSbIDX!!jlADGS z?ZXF$Cp9lR3dO`e+30vB&j;n(^xvwh5edAd6v2FeR(tVmot*7*{i-q|# z8lhtwS}sYCn(W{9Fn@qa6+jSR4EX&oo&*eJwB&ycCQxa;#}!CU9DH@u_yC%;1%fp1QSnZo}7w}|KmogS|eo2rGPAOya$t0 z%T;)#9@iF)lN~Uo2O)uoK_B}B*dGbOf_CH~F0F8b66N5;Be7oscXc98`kr&4m@(!K zI)>2oDwP31+4sr>N7!t@#wF|?ycXc;tu$M&BG$fIF#|}iHhhgyICmT%$<6iE1OXw6 zG{C|T`5)l&=DPwI)bhK*STD(2A>h80yTMdsC_TvRZ#(_r{7=1m1k@jd`N+&eocqX( zAG=~ez}*Q6K;Vt?Aog4Gz@`Uyfw)n)WZsE!f;6|pej`k#4L zWky;YN-jLEnl!%e{~h>>t!!p zu1}k&8;kh4Sy(FZ`=YtTSBcKO(R}f=W&I7@}83_PLGM$uO4rt9b4npV48v-dekuvMT@M{gRd) ze9Sl3-<`44s|FDSM6WkuBh48gv6qwb>5rG=2z~%?d-?Yd2@Mk}6j;&*0U-Q4i7@Po zm_OLdv@axS4j4x)0HWZ+)__~HpZrqI$@}#5?>?plZcw$*(jG!pxxO!e-5(0K19RpCReT&FE#9j_q zB0A-KMiFdDs;p~Uedd*yANEHNhGo`n+JjdyPj@Ov_D@upAr%+!5DEueDP>{DE)h_cL|;j4$p2zP zA$%?|hwgk^v_4QQLFGAH*x;1Oj(7$?rLUCYnOHB;rR?D# zM*!~(egIp^Ejc$=e3e3E`zpBj-6&7K(p=7tXezJgsQ9^K5La|bCDs3)(va5=c3?HU z*y6tQ6K|{8!PXnj8#v+iY~(m4pYm}aS3EdQfllUFJ+r~^!Lztd5kUlX79?ZS} z`M@gCvQX9@U?SOQI^42SYt3T-^qmjSjU++95ST%>9_^(YMsp3{!0vE-G{4y8ZAkRis85ROVaEet;U& z+=kV$P;aOHk4Q5?r_@5>XUF|J=~Jd@uJPz2o!kJ&6KHa2=j<-l9G0kJ9s)>q`tpX0 z5hyNE60#W={JWwbb?LHLh*P8RjpftQXM*W#l#N-e2#LSCBsLaxVqPE+Fq5IOJD8H? z!QyYWMdt$pAP{a!#sb=Z&%-*Ta=< zM`x9V0DbPHB1^+`%Jlqr#i6o`rn0j3+0M0!pwF@*mdG-s9)0-ALaVYESdV zXsL>)s3`hI=sUt1za5BI zx*lhEe44mkM$qL|iXjqXy{R$&3m{<9OUACME^am$`^!|8n%?#rT5+I3zc^YOMs^UFoFhb*h$}^rjyWbX!pi}aOuyuBwlYAj!Y_jkE;=#X1dKE%n6-ll(V_&M`nea&r$z(?6cS28gn5lD_BD_0 zLLVNC5D+YXpa=x$16f*n5Y@c=Fq@zJ(Ved&bPw$IU>_b)ybMiY={8lErRgcJN{THI}tH=dn6RF5@OtQ z5e()*P~(i@m+p*))IeAuoC>8M#$h6)_=!jwA6Lh`A`Aj!HS8c@c<60U+CmyBUV+B-6-8(xb*1~c>d z2z}yb;tff^vbdYJ{1B7tZKmdh$yY7;sgw#4#77Bv!O-V#Ku<1;gf*MBxHUo+*#GAI92vpy_SNIxRRwj6|HXWKf>2R{*`9 z{UApNFzbr8Un~<%K3=H4bkYw-fo63?>&-BI>!3(F23EOg1tYGd|uHq8aZ*iS~Xofc{x0>+I=UZm4-=}Y|}OzPS~m0vt8Z6Z-Gsu9VJ5vGD# zT5VNH@&zgpGE}eY4MX*f_Eg7Kr;=KGT>_=egYX=w@QYiXZgsttB6@q>X(506##$zd+qj zyA2?iKhA4!=b>5MKB3lUK?fdZ$ywl+qd;e)4-UUF1Es1wi%7{HfU9lb&)m|BKs z{enRd%Xp^PR%(LWtmY>o+C*LvYw70K6As)oI{e(*Ol;?hk^3}**}dB|vrHNQg4xWl zAX}AILzU|Muf-S=6u{Hh(>CR_6`}qHKm#w9qPFF{Bk@6R&o4e6nfA5*rn{=-{0qN^ zby^e+wgwGE3s;@vjeOO|LWTFsW_PV+*Qwjg(O)(O+lOl}O$)Wib9M~}ZL8q41?^XK zD9+Q6yDLz>O|5i4wOOQh2nm{75pA&|%hy++`cK{ZhtYnUFR!n9x?-o%={j2fUKgct zgU9GyAj@ixJ#Y@on`v*jN*p#>WR!auClLL$&H3LgWB(TxMf=v{)MMA{`)^<5jvZ*Z zr#{?|t5CU)J*?@cA)b%x*qDw33c2Sot&f|O>5e0g>F25MAGbNkohM>)FLUW1cO`P2 zXJmE+=F!gg(9ftiCVC0~9KF~Base>_aIk-IIQ|j6KyV2Di^Czp3_ThPw^REcHn-nk z=kzDRh`}cQ=5WArf&JxhL~_7}lY0D(YlIqHV#^lF$;e(RsNxc*BAaayTS!#(HR`~p zlkrd<@YnsCykR4GmT3jc4SiPI(3dd*xj--YMRN8(xYe43bztE(`u2KlyS$C@H8IglE_Mw(johGfRFq2-h;jAcU}M z7Rnm~_9PpbKc~bnoHlPI8}Vav7D!=iEd&+Cxg-u3&2yGQ5{;T{kS3~)px*~ChF`iH z&%k(UoFLD<8XOCbL7pP8if6o+tc@w|FRrghVVYtb%~j;DVW6e&&kmPUu1Sa<9H6?|v&X#4E(dD@S&2jJX^7BfEw!`p#C`K&|uV{0=>!Y|nO*?$hPifGO>hf=6tKUW;Y?L!Gf z;r4!cm=N3A3+4qCOp-Q0z>2sFX|=tN^M3vQ;FklWlDgPeo^FHYQ9PA0ltx!C z%LwY)dXJVne?h($En=&3sj>QbD}I~d44hS{)`#T66B1~mT4uWnLekC>Aunn`PL%>;Z~8T1RZd;3V)lcRx$m?PQPZC(Q&@eZ5VEhvZyQ_TC{kPx(Y14>g3pO57 zN+IO%B37&gRFn7fgn27r3-LP?{*^wYt}9{AZi^7dwA+l6g9yY6d*}<9+sygIX@EAP zFW2fk!lN@d-?at^>ia`Nwypi(5R@+^RLdEuVI=X6nt$W<(KW0A-o^-R=&B-V)NfIL`4Rr7SQ zHx?6>6?U%Ff-fK`#oemf_03AeJ*b}gzKkevQuZC8F5mx&)d~D6`8t%7YYB(BDuu<; zM9*=R8}v|D&5t^2fFFXgYVkn4o|d%C`h~P%bQ&#we z81`$cmtRs9zU*i)_QW;@{|X_h>S17nmBE4G@*uv(wGvOHQRJIi4>7*|N1N5o!$S)x zHy=Yxj91)ewdf)FuH~SF@q;h*?po%Gc1g?Ml|3P!#>pPmv}Kx1qKu!%QV}duoe*T) z+fqSkq=IK@bT>!bP9fCWsVpL#enlB;)6@2!n8xomP!&@Pm;I1yI^8t3zKFO)$q?iZxX=*Ubqt zg4yJRD$f%EP^7JwjWtQ-d>TrtE5#yLH9vnP8gZqs&bZdrvElu))OuZ8nQCfkp8RF| z{dIi@sk!CP{Zjht#))im8~V+u%b%?a@9^Yxyx*SduUoXa9DlVgNKA`0)qVP}i%?FVekAT`A%C)ofDy-B9@}THys) z0eS#Ifd3LaP>ABg|6}Ll|NUbhHc%t*3G?5gyP-DF++(Uo6kwoYca?(^4E>BD9Es7N zuYE_8u=KX9R!%dgh`1a?Pge9#{feO@p?=j(2rL5)kYnl6_4WOlH3Fn_+IF_KJN@7Z zdcNQ3ZxZWucsLvd;U3QXl!}qvgU_ku~pAI<*Vp8e*DJV9;`g3(%jWftn!O?F5m5QjN=N^kduqg0#t5-VBb zuRZRa4jzwk2a%l*;I<K%cfg?!%Ai8{`^`FK_dP5p!rWhIWlU^YDh{z{o3+$5&g&cFj4W>$n8i&`{reS2z+s*j4+YNi%>?3)1H$NOU2tEe{^;tmCTF7S*UgCzl zFASo__6LgB4!DvS&$`W?6O`9B7=@Dne0)`aF$^| zA3={&7$VZ%9b6qvVuqs`=H&|7J4gB*bRsrDw07zJl}Z=&r}@Am(KQLe+lC4IZ9jG1 zG!pH~j-mSJKKikl2%5KTFtO_acq9lPIt|@xBzHk}|2TO4uY_MXo_H`Ui@b;x^pLMr z2l-**!h9R_fNpsMfJsAa_hL=+`!Mf2@^(SVvtnmjsszo@;1YVz>Jw_JyQggFa__3yh<=*;(5bp zzO=?<5cfhJUZ9DjdkIHEt3=ENN$*#`(eaEOrvl*(GcnnO2{6`rTkj`7ybt^T!P{GY z#rbyInnhuS7w+y5+$B&D9D+N+o#5^i?(Xg`!7V_50KwheErB2b3PMun_rAN&?%jKw z-Cuf){sW$m&poa=*PLr5hmf=*FaC;0X0Q89Xpx?&uuZ2L*R|yS!k zQexRU%6F1GS4THrq|#QWqtpBDSBG;I92UvP=sSvMb?1%#=U8@yO2 za$@UGoOf;Z91kp-jcEk+ydt%xT85hGef-WpDV8Li-IfaO+G|{$%GrO~iWR7|S63(e zuuQI-iB3wXWxPG0I(q&#gqK<`fcr+NP=9@1S2IYk`9#kxPW=|%^3Ykr&qzxjztvsW z+Eg-Az9aWrrT@XJBLdk`Mf>BXf7u96G{?2_6?7@$GNUc$gjGFOX*l1tt1TD#r_EQ4 zg;0K-?nNhC`_w7JldRUB7)K|)5MkXlsilUF_w^qFwpL^Ov%7v{-MC18T>B;Q(0xN# z5nwvJy;4Th|0v+PQ}bQlQChcedE3TXDy!#*A<1Xru{W-$`=%)hu-fnMZ*SqiI1Nu9fF2PK=hF>!i__P19bX|3pbniTCf)l*EC3^&e-=z2L? zXT9smHYiLKe6-&G?4RD+(l@~S`zsObPFxXK{YGJsRC?N)d8_p@t8)-P$>n{I5s!>_ zY@4L}&nPwIkfb5u4$39xs4k>^DlTEWWc%CZ6y9wT?h9WzwxZM+{D=r{NzjzC>U~*B z-~SOjf|%V}hX^bLh5v9b2x$K|Lfk4y2SJkN^ zohjcd&C(cpWo)qyww0+YK3lH~9h&O!V&-@%j;@xA4kyKW6D&6?m44ozV^PM7Ki7uP zL2bJ;Wak#&23M+1-}m{R%;crgnKt!#49$<0Q**N4JY6h(t^7#*HhbA*!`uE0ex;1x z3>*TY?d{G|)g68x+K_fugWtHFz?F@;Kk=#1w=*}} z$P7()9oOq8VagubXA!Wp5OxrUfeR{jMQvp%S72Eop$Zq0P}nu? zDYPq!F z)y%a^I@HO&(12N|kUDJWd%Fbt?Xl3E_E2b5>|gBpc{S;;zC|eTUeSo2)WgpY^+$sM z@TJ%;$}?0Q>?=uW=g$*-Fqbb7P$+0Fam%P6mo=3|4i0r?WM(pmKu5R(qgN@_4WtHT z8~w7e?{KresV8c3gjkYDm$0N2MS-%WzU$--KZbeY9T%q?9Id9i6~l1(8s^GwBsxXxD1f%30DB8O#&f^@A=gs?QCivT_>oR zC^&Gt`VIl@citG+5LP7xKDV>7hE~L%+wOB5Iuth%!`Ilxd|fEQkWsED?fLt@g`|d0 z%cBqRO^^n;?}LjfgCD%x$ifRtX6A`_tPrU`P2Gezw+XfY2?OB2Wssjw5`badNmUp4{v`6x>i%bYt z&gNY_lyQFcZf|3MwgtoeKHCMIT{m8Rx!29=*Jj4K`ab$ou8^ZF>P=ufVc(S9ecR8; z-aBLYxd>w8yCm|*R7%O9&#<()t>_3!i1931{PAM$$K zm1hLn$0&k@5fY3KB8xVMf&)v4QNRE`mJ&>KmQiHXFexAz3PL$aIHcFJKQzNN0#&3G zsC_Qgm0(WL06kz?=Z=$Dn2!=DFvoTgeixpSf|+1}p8_{r&nUya^?u_y4okw-|bS;@_WlwiYfs0T)~IQt4MIE;zQ$RzeQj%Q8ml_&PS?T8qpp-|f_VLAhHHtaE2p;HV}IzQq~% zdxS02aiPUdy6j#QS{XQcshpiV({#O1gIUTdBJZCZ!oQMdGl-{Tez}-gG>GgBdh=Ff zQhr>5;S>FP2QAe6)zJa3yyz20T~Z_dZ}7?6dU7;&2p7A4yDDD2G~v~I5&x#%PTVdt zEEYu__7C3f_o*ly7Z%muqteliiRa!Pvv^Gz71-c`hOszBkK*QHH|Rb`BdPd&&7bAf zo=&3+95VmP`o3sbr!)RfgBORr?xSiuQemqTL#P>$k7OO$238xkQ?idF);)x-JE%q; zM|&iQ2LN(Q5l^e55NCvbb%wc;N@H5!ru&WQXY zjL&Yx&>Uib`bHW*t9u{-K!9x315bq_Kv}AD?=i zjpQOl1#2y@jP|g3b5hulb5Vks`Vyw5Sfo5#<85$|*WcNnzkKCj6RdbreLW=llrzQ? z?54F7fB9DX&sI6a$-ZM2NjoRP-z##nllj zGGu71amy?#=tc*w<$&M_>erzQ$(Wa=a#gPyJ>yI5;$F(6*@q_bh(xK&-t`{UvB5xO zCiHb_K-qOTb$kDnW3}k^4(C&E*A9^rNu1)O8J^}%(9a;S>LX7U!B5k&(VkVtTpdDNZ8!twiK)UxEVFx_wM*m%bL!xG>YoG7% ziyw=*{#@(rV85Zkhv&5Lp1R@s^Ud0?XJ20`FJ!l5KOau3a!Nil1eji~>~*B3M#$>N zeY*VQ6C$MU(=Nnho#Ft>p^mZ*E+u9t)ADb@4_!gP6Osyf(hD)r3*xd3^5m7>jK-_? z40%EgwsLnxS)=iFV=+M>;VfmqOT#ZU#}mX27eWXT2>p1U?HA&Ty08|sY#SuM5NeJ@ z2~hA7yg~eOgQ~<5h8%;3O5%(LvJ)B)`_e_X9t}1l#v`hUSbxA-LJg4Y4lQSjP-9SL zxF>D4#RyS1P+12SfR!V4KOTh!*F+I2-RTH`DE;jGXV>sWVW`Sp7<91k7a_}6VtH8- zAMYE~ppzgc_*z6Tab#VT?VnI89R(S}7(ZyudlUT-5<9;LP(+k(q|C1nFJdpmZtZnK zXtN&Fl_fgxz!7Q_TV}~PPDnie6f2PvQZ`N-kP#!v4Ru2eXhn@}L5lo3W}OuiL)RVO zFC6Eijc>P}uy__{Lr0Y?6x|Tz@K!UP&pp;vDshoGE>a=N9w}k*Tb!qDLK;bAln<3; znp`eRlIVE+iEqq1lCWxpWNq8TnN^w=7V)YR4| z@3ylj)%Bza5a~BylKe3uL2XjblQ-3cHm^)b91^JS_z<0-iMksonIuK z&mIE=hk9C`dHElsdYhqonW1*xq)RJi$cbb~#%6e{W5`aRAhxEfvnB_gGrfTT-#}1w zHh@<1K!%dc4;PunJ(;>9C~pQ)-q1*cZWvCxvW$kZhO;BfvsKWCqo={W3qWz zBuEL8@fX85K*`TxR8`-C8F@0(tWv_Qh zvsh^dD}r&fz~%N}3tovl1C?|%{@RVa;!uuH&lDUYV2>!FZVg5pP?*zVUg)oYR91hq zmb^4QaHl8OCo{KCM7B~SU;H4E(k-t#qX0URicMRvSgY8*5nL5r5Hx{r*Ok9EA-`-~ zi#se*q@f+&K~Ut=Ua*}Do}Iw0^31=il@~ED!Z0tqS5)L(_qki7G`#aPj|~m*R$P6a3q@B-fn&m*U8`LP@hC`BlA8u7J9$DZ>Ac z;^44HuMqK|(E)0<;O!Hb)eV-Ze~s(!_H5W_Ql!n-FX--Rab~4&(D- z@&JA2RTA7#x>JR+K*2}-)NtIU?j6USenF+!pq__3$nELx;-gzZ}4(Y zg1ILZ$}`_%`)Eg9ZCDogC#U@EAwXRI~S)u~&l$lbhLvu+EXEN>CGGhp-Pl-^>jBMPnD zte=kZiiY?y2GsbXOCZT~dc1!sOg&GsFzNjsOiabfzwnD9@r%vF_fN-VgwQTJ3vZJF z%q-+V+hTLMo55Vl58T73mQjO{PjO6SgK0Z0%N(dAu7kL!6gPlxRN=2~-{_-N?r>2^ z7N7vcL}fSV7tie$1e|$Cj~U^;29Ft1$-xNX@;)N+_q(JcMpAD54IE=MQ$Kz!smljJ z5R1lZ`Tfb}-gvBP`ZV(3n|A9EDmoa?)IvyID7==8-x*7-*l^YGtU0Z5@U0^$)@@2D zkHL+J5gfKSe7|A&9{4~%FiXIeOrwK@{jDdaJCEb1R346g%^3Ubm_2d4IrTK45MIV2 zIDjBznqrH3Dlh!;MCd9w99+Ut*b#27PbTUab0oi+<#^iG97R-9#}*iOIio84&L=iF zBH5*F2DsNeGQQZIEbq8FR};Dkz^^4Q3UY}P!}=|THE$fS+?7miaBQsSI>&_}4Eb}c z_4R7NYMC7H%FzWIWw*)h?DV>(fo<+-$#jW)u2l^Uwl_v3ed6zn#0X1A*hbpQ636^& z9q#mTnD__bn;3@qaCQXq*O6#Cg4Pk|KjCC^GV7_aJy<`z%-!YSpsEeUoM z3VJ};Xj;sIJtheNf$W}5bQagbr=i@E;$5zw%`x#HEdViR9?C>;mlDNLK});=kfp;* z#R`t+Sag8Y zj%g>@((^vlkS~LXd>sFY6gSn8hweV1Mx2>!F3w^J6Ri>R9evqcBDg($tNj!rtE5Tn z=~+%pYbeXhIPV|aUdi=`MMA?cHQi>U7(mIQ;`5>zL1dM|?=)^mH3^O#ORrYwx04Ic zOHWp6Epv877E}~nsm;rvsWPua!$>2Hb_61Dcbsztr)uZSYS+KVvQtI8jB5FNR8F_U zDm`UETv8^$KpYsO(c*_81UsU(&a{_v^4A+S8L0(0edqdci!&J2UboVB32F7$`{Z0n zXQTj?L2!sNMQU#GiD6ZnGc<@KS)~(3K4+|cjt_upSHES12r+-2UxYI$#uIHcNSeCh z48{pok-TkmQodX4I9gs%SQIv$cS-)C0nGmI38yDe9(GyCq&} zvTAM}oUYLJzIE^l71RGgb{s(xdS!Vgra94@P0Plp5F8%K1z=by`cBAV|Fb>~K;B*X z$kj2T{?YezLv*x9`C&>FJQ1KRCQ7PQ8R@;5nbSTq`g-m)n3)5 zcd20=iCrNTU=#=&sNRif5SOL;btgUPcfJM@S^3dROn!CWVJG)n`GSnaW`I}ZdGb3GsU}-41lz?SvM;f z+nspLF7;V^vFSZvcE>rvEPrJL0a~AG>=I)(p^<=|mNjjB6)bkVJAwku>9)R!&SBpK zGZPMx2RnID%kPC&z}`XdYTW`pZ1a?LjZr$g9kT=;_&PrWA~{kO_ZldC&M8w>ZR<*Hw=ZFtZ!V`ymNk8$6j~od4afMTr`XL z`*!p#hEisZFRP#bJ~Suy|FBj2y}9-Gv3sun+Gj4H9mT^a7F5K(-g$T3h91t^9uG{g)wm^d_F!_djw zpD>5K5(20kL`E}1l-x(8_Jz{eLFq}t7+Ats>@YY|J|>3wwrQZ$NTGAbkac+bPh^p! z=Y)%Phl60KP5{qq4R(G;R%>nI;<+ZfbH1}=&|lYCj|^vAhdOqLhk&CX zcJ5QsW=eKplCFpWHP9$Ks5b=o8P^3DF2uoQVg8{eDmo|J3>Hm{$~o&rwLV0ssvj$E zhstq7BU(-Q?ue<+g7qmfTpWhlH6QtLJvs*z)9_6@JSQ3zh7r;npV=Lg2#ZNtr#?L- zlF>12?uIt4>tmC|2kW~tYQ|$mQUY&dQCOh=h@AeF_;ME4ZIb934Y~ltM9h`A5N_R1 zzA>#VQK2<_UbAkyBm4Q3{xijfnAV$R( zRdQeVtUHQ^C9Y6E&K1Pkd|-iq7KgZ=&?X#pof&>@hsq+HENX|!coQ!lo$MO+cHP62 zWFi4cgb`&Tjdml=sXKAkH{BHuO45@uFGr5_m2$Y4o%Bo+DW7vT!DVi8bB|eryNb z$t=itbO-2(ZI1 z-^C`s2hH{kINxeMAL%A0ETDr zpowl^UT7#qZYo8{w%U`2(mOUl3Ee6@-VtW^J9!MG7v6S&ShKHkp89UlPT{r zL(4UPB2x4f&1t?+7uVv&%K)9n!ZZtM?FkQMvOvbI#IzF&YZ6@4W@@rPl+Rt#B+Fig1KHqh^6p2CM%Sg%Ns2!v*cwmhAT~OOF@&Ss8S`K z6Mj;?{&hYj`f~-NU2(lqaW7^ zf3oaSbHW3V<2DUS0-ZL3rbXC{Luv2{FyH1u?ZPUMi|-`5fuT%LNn$I*`HqP_>60#DkvPjYy2wD=b}Qv zH5=Uj`|-xKdq|~!;9tWck{z|EFYv#BzqSMbt{UPGZo|mcW`US^XFvoWh);G9L95k{ zcZDNri8qB)j(PMJLP?Q7Ru*j6&1PQjAdY2&$qdRpFisshh^$BNj&Zua9Xc!Z?J$|e z;M;-TMvwx7yqrb!F%w;p-C_D&E$BFd?t=IzlXn<2i1ziiW zAdjZAXCjK+*`Ordd~u_Q)C${*DKC)hm@W%y({qw2hUX(4U_y0v;t?d6$CM^(LBd+0 zU@^`R>up138kgkl4-VF~VXe#ha0)vit(2bF%?a?oJXa#_g`+X&yDV-pR9)4g7S^$x zUes0R;es7XMrQ2_643{4$2j1x9u{RFiKmTEP3XDD74q1%$9mbb2HP=Q+lO*(dk$ob z9n08amo?#>-u_+uL3v6PT|Mwy^ZBfI zG)c;`b$UhFB|FZB!l@qX-kzg#%f#p+?R~Fnng7%4*ZN^i??5J%uI^qv{tS$#X|55lT0zOo^-4oaxCC2bl~+Lo=G|&0imT_a=dA6+ z)G#l{eU`d%Kktf<-Prv)!6W&v6spCS1NoCvtjDT>VYo4EuIH%ZV+fK(g2$#7JzL|R zmPMo0YK&B{@o}=AsuIxj6J_x~4=Ct}rl6ABnIq@d(g-KAg1Ie+{NBYpY3_ZQIp1=uTP zd+@aoGV23HEtmsfC+s6Ko0uSS6HHXtNn4^EDf$A|d zOH2irz8uScecHH*5^R7nk8`5~9>5lo$F!ty0GJlIUhc$IF@xwfKEkwl6R2`C%)%=Z z$Ev!(qAR0`(+10=;+>FRQaZGxJ09W9Ye6VZF<8p-A-;*#V8`q;r615x0pSOXD#w6vh$H)e~($Zc!uTW-}6{C8@xqaoU(_Q@s|_bdEuQu+ecIy#3T< zM<_9Pg7dZAA+S(enc@rvP;bJC^WjhX9<^Ew_{Ec%lzN0;I1dHZB&T4l)1I5ijG*=4 zW~=*Iuyg1`jrxWW)c7qK`fCJ~>_2K!A29H)X-Y7W?Nff~{>D*f24yDT&8aBJW&Ok} zq3ztpRv%c5K+zl#vRw`YCoo}~E5f@%p4;9PG@sa<-pw|E7Gtr%%!tmy@1mM;3cvPg z@{sDz*JY%l9esp|bFnFY^ghqaN!Vs0Hdv_f#E-^i0U}=C#xn|=C0P?#iy~W3hIO}9 zybHZxcS+?v7rzgha->7Fc%JL&Q7v)#YR@}mhonaZDC5gy!F#A!uHknJQ%z(zi5po- z(PYbKXmn8DJk=JZ;w_7?q}NGiUClag5BvS=6zj@vS-h83QKZLOe`^w-Pmn)jIDze; zYI3bIYb7={5MO~fzd|UqqgnUu8)^AiU5*yQRi4?d_T1yAn#mA7aUbV8USoOvq(9OlMoq-6pFUZ$`is*bK z{->TUH)l@|EyJVX2sSJ`Y45IjVe&J0D76vv?xRkgBdzvo+O%omY$gmmhfIYp%1vr8D{xDykftT`BF~UQ<6S>Xz zoASgsBxjj0)GT6D23;jG{WUv(Kj)j$=QkO$!-Q!g-QxsqcNtyNCO>2<=qr4cEW1@s zc~K*IZB#vy78*_kKa;}Q9D~~mc)p*AY(39HV)=@m5>(#aKhHr0<5TkS&ZS)pc*1zv z%QlQoWW`<)ZsQ87Z0X=s$c4l==MW2J*;nu9@c%#^lYJWqpe|Y{7{rg`@8o>OR?1w5(7a<_ zNWBw=elVFEv^tl-AJ&}k+8hM~#LK!uJF9~lj+0JPP&^f1uR};Y&eB;yZffgs^X%A(4O2vr1 z&l@k-Pi@46SAPrwj-w^$TlEI66Nqnix(58d`cU0|?hM>Fu6~}&r-Eb%^&a*_c{#r| zxo$cWAJ<6!xKd4dZ%)x2A!AFFU-;f{> z`E^l~pZO>>hxi?Q{3HMI2nv0J0R6E>U^ym+d=4d2z-t`y#%O^OuZJ>7L2FJ!=~xJU z3WYJn5F}>#Fn5P>4~4#gh4Bu#(}KeJy9ot+!x>A%a6#e0r(qxvc^}b#HKR!ZIDkDs z=Kp-I*EfaWzh`Lg%4H+1P^z>^N-F-9p&e^y{FfO`rH1|wP%+N_Um4nS^+Kyond9Zf znx$%W{KW*N4mR0%QNN^35pH%VRo$;~k-|2Z-;9ugaUsh4yX##Riizf5C%z9hqTJ#Q zZne`GbweA&R)htb52`Z)s0Qc|VbUglOGQ0e2;B&kXvr0FuYjoGo z?~FTr+_%Hi-Y46s(Qr5?)$5aBNOOJk^P#WDDf4cZPtx`uH)2N@1JgNb`dp+tWQ9HG zOzGNQD9wy*0TdJO9O#8~6}=yUC16MEH=FDB&6&WPWx-==RKCRHYT^kPOyOU1DE$E- zd>9GLOWlp*7v!f-;NpBLp%i$`0VmBQ5?8pBRh=KD(hTeExw&+geTZea%qfnL*Azwp zN@M#@aX|+5#}ag+YD)sf3vb7`;Opr}0XWWfRykA_B8ME#uzR(vlL@^Fi1u)-S(?y5 zb0s_cmHv=i4cqt7H0Y&gH`CW!Z-?vg)#F>9OxT58VrbTLfv26tS_WB+UB zx+9csfVdgMr@OYe^EIvm82}p=+s{V3Un7aZI;*FTcu6I{{YWUbz1=`=c^1Xa*?urK zVN>v9c*411`fiH#>nlp*O62P%3mQOOj+7(AT#l3p_8|&G8nwS|0YbbJ3H-GDvF&36 z7%<3`o+=f$qHo&hwfZ4_acl+P=o`-=AX&J<0s;(AsICHaVq!JSG4!}fvf_!F4CEafAccA4Cv9Dv3UDF)bLL5Nx=OXCg%3IGxV#oVveNQiMNz&CPqtPiZx zMtQ?U?iHoNnf4(2P0=`%#cI+GJD$S0l@usJjU|DDVi${Ag}0Yb*F2jtzt6DJu3ZtvZRNGdr3MivJ#`7j_A1`5C0HdRBfS-8XA)Z&<+jv!+`ZY^A zU3%BV59&ez0-e@LZblU6*3@Y{9lZnqe^L%vI}1fs;9z=_7b6=kH+U0XCrIZY0!6@f zjX;DwcJ(1Qxkd*dA(E6L07*b#XR*|sYwzy)#A_F<8vv89x%+hHI7b-!&b3?@g30)` zY*D?@&pe+l6&P@L2yDXg7A`X#iEh#o_zs6s!ZGf%_8|nQ`S^MDhJE zC+|OLApZa(<#z#pS-Xap2bi4`k@UViR6(wp4#E_kK*-!-i0Dw$8}Dvfq@rsCbk}$- zVMsS26{r7~loVfH;a(J9)I4X6NL2BIuKow7lpgYZd@iJJB=14Izi#p9B>TgQGtE-m zlV)(B&p4{x;9KXR2t4`qaWq99S5nvb0DfriKxC^cVFCd+J`(&Jvf1F<;1&RC3h@-= z(J6H&5KNJsJMqPLG=6Gp=NGmoHwB7^Phz|o{{1tJgZDLJiTDM}5^Mp*t=We*;X(yN zZwBTxOgu=Zj#a!qnfLcPTZ|(`!3efuKg*RE0y}@IGF)k30@Y|aoqyoEUtO?<_y)ZP zFw_gLjx)8l`bId%E0;QNS7T1f8_nOTXH_yv0PWr zlJPYPw>g61XggrQ4B0xCHQ^|0}%uz-Dk@C=9R~r)Du69KYJ7Q<7q!& z1H+0=BCg|oAUmNg`Ph@iR-)wVh;zTU9231lYW^!L5*&UiNJ(d?Rs4EF+(I^OtD0GE z5i!X>*Q23YcV9%jat0B?oD0A9$jAS44kZ;`G;Q*XO4&R?F@2f9cS>WstGK8;N#vzp z9M2X?Y5>R1%o_!JHYoR}8fS|xC}X8^t?FFKRFiF=HR0B&jhuVViOD!NJY^q=)7kv| zy#GnmR{|as$H~=qNQws={aP?-}u-0$7rf1j{V#P&kDvgYyOFabOX@zkFW zVa-L)hdRL%r7i?V+0FT;r4lXCaF46_zLUA9zi_+C;y*<(aHazxN+Kh7ym{uouRYhw zeFvY`(@h3SDqwIB^Z zlGXTsv!U6{p~3>*Q1_=nVRNDbt;D*%fmoa;QW)h)u-I{e*ugu1?xI4l^95T>)P zb62flq@X4TBUECB&OL^k#>jbiLUC}TY;%ll)}zEmq7qP}B}V-XQiDD9C9L$r}^9Sn3aPZ!asRy%rkHF+lt$LAX&#pT?@CzuL1Uh<$7MmT`~GMIhtGdj z!Zd#lJaly#+4XS-g)p(w@F9J{PG7|WM`wbAXd)>49eNC+C- zH=n7=d|`ZF(*Aw5+4PGI2iW6sOA+B51#I^3HNbm)))-y4<$y`;(~1HW%g=?3U1=%~ zy;dh_2zc=rluv0035o)GC?cW4K@0`s_$zm%mVzx!}$h$yGfzBd=jMrtf8(ZqhZD_X6UERn;xRn#95 zQfqI>YUgn#4sPsW7WVLyNxXiP!8{9Z9BN63jq1ix#(XHtX+FBW{c!=Cv@kfw)uMJQ zO3izmIfGhP5gOwc#mbz`(&TGf>gr5q3H>PDTb?X|qQp{eL8WAys!$YVU-jOGrLI`h z|NOM1c%YWG42$i>x=w!#4l=0_Up34%&R8YMENQlxWUDWfTmPO`nkIT#QRbcoEy>0f zJFLsEI%7ld6;&=R-&gcMZR|wvzG|H@RHkeHB5G%wcG#=`qnN5SV6CY$GT%AySWb)z zd}Xe4)pz^Wnzpg0DYm|;w`h_pmUYF*xjOnP;KmKW0Nx%U_&xUh>zr=G?TCqH{B0<; zAmPsVtP_tr>9WXLaV2%~xPcZ4k5^7L2lW{Xf|tMp;WV&TXI<;f+H3PyX&)hU<+ncx zHx}N;GdV7(%X2&|YwI~ZtmxY|KCFK5xq4VLi{^M-x5{>U+_0-|eB5;Gx_aDl{l@XM z?YZXkwBvi$__Q1N1iyOPgQ9YFL#(kl*EsmeRdvkcgs{ErlZrEak2CZngpafIB7IK_ ze6EE~ifjliz>o{>d=BJZ@{-VC8F zli#?hDse|4KEXIyKm5|2)GEjbdzBi_o~leh zK@;c+>;n44H(K4~uNwZ24a=64CgRNk>_j%sy(ZJ3x2^rgj)9yYI0pB;%r~U060Fd2 zg$~|PBjO4VAY{!c1_0w;xVvkSotp)A7PImwU^U@Cu)R0h)*ktNr3w5~0*^6hwM;(%!`?dH| zyh<-wl-W3_`gTf#<{}=3$xj?@igJII7+LZa7J2VwT&gEu`Kfl-aEpe+)Jb0ma~f(Q z)$wldvQR<(Ha+0OqJ}?%h&p`X}i+trYzi#f90<3jN0DdMrBS^(8e-cRjHd@^z1itry-#D(_;@K$n(SX*~Vy0*ssBaFapK2g-2eAu? z3HY>PN0jgECy|m*J@T*$Ektvn-6%4ib*9}`%Bk6tE~0|~%}P=-!sLUSi+GUkeXBtL%3=2DF18oY9)e$?ymP@D+Pl`l>D~2l z2MzKB{a}Yk45EQ@=^tR2aC7;u1?)c7t@U3&3

    o5{n1Gu6boc>t^*Xw8#U5Ilayoeg zn37qhN&A>fg#Nk1*0gTsSv97UxOQ$T5Dh`8M*N7nxF!D$^<^RIy@{oO6FT!;^se2K zLZ{XO`4}gqMllw_!`RV9=x@AAxv`tl$ZV@8n7!v$g!p!Ap-?2=iY6D0-ZyY#>SN~F z;6>9pQF}c0z2IjGRfkY>2986g)KZ}_I5u07hpf{1GWsS-&8qe6Wprlne@KX2#27?t z>)yzmHt}X7wo7-xD_q)-{~a`>c;PqBLZw^V6yh|hIRVeFR6ggOZm&cmBzH~I<;~zW zM|e#iff*?D5sevq=pdAwcaPDcil8ue!Tr{P#P`^KMljjE8QSb-@EK^4h7udEsxk67 zDBgi)ChHA}bc99WRrfnd%_$#~k*FAvT4V#28Lb1ghapi-*UZIUS#8(yM@I2kG|xGM zOn1P-As6M$MQcwjvWAJ)DyNvCExa74_Xh2*l|<4THq&oCcr}YW8{?m(1?1nk#!-Z1 z_;ABS_QF+^KxH0-G;pGOib(`*7 z6k=v@Iswc^8}sZZ*^|RpD3My9nTv~}e>zF@AH#RZH2AH!zZqhSkJ=|HYk?)8^7)ng z9!UN&mLWv6mOH+auK>Jw><}%+jg1h`@xyp-9SYZ-S9XAe%b8$x@`R2HN8OJK{$QnE zbmaj(SDtt}2FA;iQ9ZI!dJp>ErS^3mVKO)`p?_|iVpivQT%W;v26Kr4zcdq-#{_I_ z!Om0!O4Na(;FGg=TXsvPA5zzFV*&ZTWYAE=_@hj8ZXkQNl~fZnO|%|e9JPkCxXU-r z1GpGtI!`pLz+ge?3U#qg^M&AA7Xm#sI{6V-AGUbL!G!e%{_U0D?T}^T{BA-NWO@_Q z#h&NMX$B8VbOH1&rh4V#8BF}V$reP09Ev5K@JgQm^oHG|m*^*(a#d~DFQEf;p9=Sg z<3Xbp<(m*gf^pQWoQLBgqERmO1jR_N));5Ouz~szR9^WPHbY<61JMRw1!@q&hSD*A z{lu+dm~p|hk+xPOF7Nqd@AY$gZRk9T@|7(fQ@7=GpCVNKaHB=_rm*A4?}pdSI(Vjm zE&(kys>PcO#;@U*S`+xv$S_xs!vDVy9V0Hc8ugpM@|EA>4Xm}Q-cH%Kok=282J7ysp84a#C+T$xL+xqMx4>d}=$;@=tTsN~^c|5xWIkCo>rHowVfTmZI zOB7<*>zOQQzB7*FCWYj5|(X#WzDP1E<85Y!R$8p>YYo9gCbos@Prh1^6g&RC7sNHtdQmEY@7g&EmlD0Czi|?M^hU8T zdM^A0peS~9ZY5!j^I&xHpCzO=c!y2t$}b@;x@i>^?O_CJI}k}Y^wCEfGx-F6aC*b~ zRXMF;x7(C0l|`7>tbb$m3d$(I!xFmEHt#i#jI|2uVD!G(o@=(V2EaU*_P^HU4RubR zNqpqlcc(kESTZrHQ`>5vvpI^?d5!rPie-xc`LIY~@^PC9s9U6%x*0~WnoL&s30=qE z?$pv1-3+L)Q<&bHc?ED-`8MmsJT9l4{$J0$##!s_@ZV-lfV@4W^Z)ouPPK0lJ!!RV zXt-qjcNXO=X_bN{o0xfo_oj+JPk*xLt26o4S~Fr@%2W{h7!nqyW}v`OkI|Ekzw1TGcYT%ORaA7DCxuRZp@`CC+(eK0oPUl1;Id`5dcN z>nDX@A!(&);##1KSZC9!{g~&O!`{O;J9$S)q&H%HHZO2qku9^hjjWb-Q32VybteSw zzh|wjrO!8;zWWBFNw$bpJ1W?C%CKj`S0rwYyko>p)rfQI@t@9+L#{z5MKlckadT_E z1Y=v`MgI&T%rzbF3F-w-<{m;nGNqCN9KrE4fiWe}LEntH)D}uwJPiW^i#U68N;zHTM7dY)Sj@`^AZ-{ZRsa z_1k|pw;pxL0RoF5x$CI zaU=at&j%ojN+iX8R@&qZRMQG!BMxx_bj@kKdf5uHDhp5m1idLc-oFu%4kv|({M=7saL&--PcGm%t z>>qOVvz{J)Xhv)5_Jw>{>2+HY;FDgxzx0iukoCpKCx9g7QZ&Z8d~Bql8a_{{gE#w= zZ)14-zOUs+UKsx-Drut7BLm!W@)5pVLrr7VO9L}kC{KLP1~V{@fu*2_i3@b&6q->J^T=-V&3>ju7xt7&}ta4T7YRMD20P_R3nLF9`|@eS>OcKAhg!ZZLvjX2lDXT&P;Fze`JiluQb7UgJD;jtZHmuPf?}LFb8Y6YqFg% z!i;N5pK8fyq6*sFR``y_CPf(BXcr5gg<46u>soSl>wRtdW&BZZJ@>qzqMQXdjnvm~ZNOY~s{?21uC?gwu7(V*Ys|%y_CrskcB{TsjqCxAt@} zy(UnItkBFmuI4HRIxOIzL_-WCwQBFWol*nYS=_KGf3@I&0#ORiuG>uypFrdg|)Fb)xLifG`38$7~lnt|M)Qp16k;wg@uS+cMZByUMWMqo9cV(wZ z8Fy%j6!{b@B&66FN3K`$NJpT(=)6x^uI7;?;{joIEok<*3cBs-=;yndwpHJ^X8q<8 zM(np;gczg};<1OndcK-T$k6_64%NnUZLa)4sLW*k8A!zTt8A@Y!1+uX3Hv+GY9Bt$ z%NOHmPaCXkNagc=-1*dqzxp^>B5|iVVtsF7hht=~GjRG#nD? zyo028usz^XOl6X#`D}(e+tX>{bb^*!3&HrHQ~%Lrz>~}RGW5P*mb^nG>hIH-E4wMR z$HsK$jJ{?$NJe=f8x`CFF!uKma*AJdxX%>B@KnANmL}RZ0gNt zNUS_u`V4vB{8P0u;^Or1EAtSr8%&M#>;K? z<< zm5_!p(pkNO#yuB2V#MjH+73^|9j}i51&v!2~BL1G!jo9V2uk^Je+lj zRLclY)dL<)gSavSQ_T2Z*ve*T3F#KE@9~u!f{(uX+26+ApobVb{lh4hG-Id3EqWd$0XFBqwPw00ZPn+C4bK9&wQt*c;qpx87uHzlOJXwwY?y zx({zayE#>XA$iUO@p3;kw_AzU)? zC2<1Ko0r3bH{<1+&f^I5-C?MUy4CMKOq?IA$)!iqF%;7g!l#i{+4n>U0nc02ewS%q zvf2MoSWZXgEHlQBMweT0H#Ril#@4W=v%8NVCu<>*sK$Fxa{`+sw3!?m-=8N3K*ZNB zzSK|_sY58X_%GMV$$2SbD4uuPtFDZoAVAmuFq3nvK_IE(F(I#<^a^;Q%8_820;ooX z>@dH80u@gIn~ z-@_ncaW_a$^$4-+IS_~E-8={iz!#8dXvkt6hU@YCt_cEA?ix%?k}!R!~YCc%9~ z;Ly>6*Y^@aQc>v%jCXyE*8H3|htL-LWo}I2g1{XHuh&Q>-Bi9sGDczKb*)ya|7r+t zXVePRoc8)ZzA%bAwnlT52qfRp9g1>R(Q3uVUs=?Z|BB3I8hH7iQ7Q)G5I9LI6JHt; zh77DTHR8qpVF*u8iAg@~6g>M$z=#gVSSG($r2W{l=yhzS@axW=0NX45?ah#&I`oPU z)Xl3xy1`3lty)v;mEqL7ii6!Tlz(f7YVi$@_`;at(%NSyoB%m^#8}yV;VAJx>~}q5 z{WH1!LkpKDEoW~ET?u?)_qzD2`uC(8U&qa9zf-(8>Rs0EMP7x37g<;`JzjkYgX9DNJzr&={27 z+eKCbB$1d>9Zz%#C}P9?-d*bbMQ8DB81_{>&wTTDv9bwf@E~G&l{mUAh5nio0zdY}(>3zV9#jS>YSzNg zcXMQr71d=**)mABR~#0C6qvxNh&4)#2_LIivR)7Pva1P>@+Cl7g;|@ZlM88#XPT6l zQh)tz2X)JH)Wh1gdF_{HQV-&l^~*pn@wYSsZF4NG2{)k1B!7KZwqmsiXKy}GvY9A% z$RO96sutL#|?fCS8kaS}58msjKvHSk= z%cpqnn3Z(ex zD1j!><-w0wA$d3O;`bctZK(DKjb|#&z+4IA?-=TVkH)2p%Z7ZmQa2@6r^%~<@PjeO z)SLp-j1I`z6x9_Uxc0K)Cqx?Q2#Th2 z+Vt1(%e|N4o*P8vxrp19N4bJtV-I??1ftnreq)VjkDO?+4Y2xyFtsGt2!Gia?^&=HVu7;n?{fe%vA zbD#{3(|;#OBdJ@luzztAmH9|a%1r*xH-U5Apq=H_@H6yE<)@<`t`GmHPCCB|FvW{w z(d>cX{{o-XeizWR!xJ##RU4gg@AwEAF__>ws`Rb9!@sNInV%XHLIUwfRtnoi?L*nG zy&yjwuN~;D()+#+k#yg@_Q?wz>WTOl-$?EK7i-QdAKW-Hb%(t8t4rMQq{%UVZZe+0 zw9J0HKKc=gnw@91csg0|?aC=E`Qe?zkSv0~l^_YBKxnS=FL|bOb@x%AQH9+%nX>P0 z?FOz>o=T()NZ|}2aJxMeD+ZJ5&&h~B9{k*)b4<7`wlIYVx=E0k`jN>|V!(DY?wJ8q zD`_o-I!CKZam5BSzqX(s{#LpB%j2i|*M2xx{7kOmvD8!pat~8ZK#ZGdI!JrtU3?sJ z!+B;adXPT9miBaT9Wt?wt-;GAP18tp*Q2R{YbX6@C`*uGy8k4{{DxwXSQV=Y)J={- zJDPAZGTYs1!Ab#7R`#NOE1zwd;*#4L>`~rmLn5W&0PNy3=G$yg3Fg=f)~w)XC?l zwcR8(c3es;U)rXYIG?TI{l{2igT#qO+}if{$cQ0W{BErS-e?4-_tHm59G4e8h}6V6 zr7c-)+R+KUCw%*It;uf>_%I%mr>T!Y?^zf;n5i?cKM>zXDNoh5P^u2znj(t)@_DvKR$uenKeGRXS+;M{5v}>~EBUUf5CeSQ~${+4)4a4&Dg+Kwp z$>RvV_Mu;8BJsQuogT%Apj%U7M~JvPyf=l-xfqEiS*@X0`lb%Fi?Ox=ub-(8!Wv#S zslNi{$pTwWEdaRb_YsG+6eK1ze-Gv6taMQ3%|0>3X*Ggno=cxR7Av>20SK+u^ zaYLK9QD>D-nY(!L>5FXa!>}EBLh%L=^WE9mMpiJK!mz7_8CLnOP868>c`+7)gQGIP zk8j9aH?Q7qs{vktPulM~ZFosJ;AVn|Sc=wNg)<&a@;;Sd9uzP|oKlAlrmib{% zK+P^cyynqL>%pF?lzdU%mnRlk{w`CD%=J)yanRv;U+kBPvcJdIlZmD7C=T^A_#nI-PsJK<&H!Uf^moaoyhAaUYeZ&dFWYJCJ#~%jjt$-sv|xGQDTiRc5$ofJbLjk;hWxXhM2ct zHr*_Sn+)D=*@=1&5`(xT(@!sS#*Wnua2=mZRP~_wiVhmx*e^48=vryy<2UUpR2H}D z=*AjG{rI=zHtp&VsP-)yy@R6~w2WpgYh2NGuwv=ju9wOUU^NoINl(~@DuZ;lKmbbt zr^(UJd*b1oGMFc~vE&N&?q9-x}L!+tM`FU*irKxc2ANJbTb<#!Kf?-W`& zxFzdy@7baZLI;SRf+*TKm%ouK+TY1~+LcS-c#BB5jT0X1q8SuROVuR%Gyy z7qEioR6}7n0|TLc^HFeMx*&Sqe<#pibl|0xhv8pW5Y=@jIE0e3Z=g2iWfvC%qt6`*K}-r`Q_47+VXC`<`Ep;Fe5DoEZex5bFxP(|+=W;*duM*zEfMEN@;%K_ z4fEoSh*z(4aI-rildH;gA5od`ylrr|TzBX}+Q*);hl^IOL34(W+_`ORD1`=LMw@Xj zMLVHyw{jni?S##IFt*_DeDsOSB=9Zk=y8qa&L4jGW#8Mv=Mf!`$A)d+7)IKd&6P7} zpx094Q*WyqE_}ATH^uRx|LM{nBMxW9^(#CE9P1@i|6Bom@AH$v;X?oI+*dqr0E*9k zSvJlnA5n`Yeed#0)aO$DEF#5*e|_*R)dW19`10HfUi#LejiEELof3kV*kht`nE6r> z$CvjbRvD|E3SUF3S0pQ>xTw=t^;Mout31#O4s&_Ppukq4703J3xagT<6RG;Y%UK1* z4PUz1Deslmr`TfAF6joKEmZYQuu9mCU2du4n|L~?CowhSUIfl) zIi@9sIxdaePHX}VvPu-c)6&VS3SzMoNlR7&K9zrOb=E&_5YVJA?ij06uGdq}JU@&1 z@QF^j?U+?&fhTOs&Z?XH{rsPn1F@w0#KZ3|n_ct)(Mu%yMTlPfJCD>6!XNI&b;m3C zp}7yHN;3sm65d*J6X=KcXQ^6S$ne5Dsl67voF^%o@91wKq!9Rm_^`VQ?wRHuITyEH zxxsm1k6lrMG|@@387`U}&gI4_q(!+i#y4d}U_$1xSh(zW&aEECX@&IGv7p=7Vi?{O zVuWuFshI7P^y>bMbY*_gJH`-25w zJ{=)TmH6-3@q?pEpQX?u{3W>JY+36^tU8#QZ+)A1`}K_Q#?9AKxiQgf5TCasCf?yhkmPC5;wgyWY~jq z+VBo+!)sXWxddb6x$fZQ>dEE(B!;aqzdoIolU6;qwx)5p@NEG%h7zd1(>fM)qQYn= zytt(s_!U{qgz@>$;Qw%*{%?nh(R{Y<{MC1U)FP7Ae_O%`#eQU%MeV_^Dowmx(?rR` zc$NK{FgP&QF0vU-E=&B$N2FSdS93z-NTuT4X)c2*2AiAUl<~{Ez@qtFl8hY5sJZ4!tD1kE0^c?8;BMQ_p0k)hTy7} zB+u2CQf5_cDPAutL3yJa$T1%fIJHX(gDqNljeU$C`aRw-NADHliZ#`SVF@ThvPHMe zr)h^*j8wo-EpH%&TU9%sZM@;LmB|}_p;lbc{|ayXFPDRWNES?4T4m$@gUfRRc<8k9 zFE5=OZowgLdm8Xr`reOJe;~Z)i;Wp?ehsf8#N%J~f6D3xp%bDHE-*-qAq&SIfNPNO z0v_;Kv(_3vbQTi1icGwMH)Cgz@{8NP?dRPrT*(=@v)$yaOHB@5(6nap@(#@lc4waH zCPY7uvf*G#tM-??ftzG6qw;Xmv`X?_hiYFk%e#KJ9&4(}+P_KtIA8I=*1X5B=F;D> zVxU2eAjWi;S&cXhkOn%9@+_tW76}EP=9bQY^>Cd5#8}LB5pi@_sH9A0Cco=*8_TsZ zERhJN~?KGKZ`;i_#sC%EcyvGRWhZ01-|0)`Gv)wEw{yD5R1A z9sJX#l0uYkwsgnu$~xl18q)f&_9t0`))-2AhCTFUyy^+8`*In2?TO+{{|M~=zIuc? za*VD(+KpDGD7XNl@sE4ENqK~N386DOQY+vyPbj!~8F_>nbI zqm=ph(dR0L8PYZkueTW4w#~1E!1i1kE=Hu^L&~mz3FJ%`CFp3-;Q$IY!l~e;r}SH> z($cQ?W#o(gH6ia~$CK?|y^e|<--}+HeH%N@^Hx9U;0$*2&+qwlAiI}SyST**v4ek6 zg>`7LPApE;aGR+8Ij+zZ;b|GKYDhQ=9|3!p0_tYu5UIoUGB@kJ>8CDmYeMr*brGWRI1=u3<3JE&BYr03Fg2Bgz?JoTi9 z2wSF&>6D4UY|pVvrr~43@>9RGckTMypYAL-yx@QJ_kLhqDU?vhX}32NUCLN<`CpO5 ziBJRLpc%!4v4RuW?}Tja!%vS5(D(0B52&KC8W@?Yi*C(hH7x;&$m12_*H}UlQYo7@ zdt_O+RyN-}QTq_>0GG<~6qvRD{$a(a^~{WtR8E=ELz!p&s_-HmCh|8f6}c8_Opr0~?uhkZN$Ev(NnNMKPriI?nRT95`zRM@ zlqIyzEh#YeZp@iOn8_7V%&Qp66B}a8T@(JkLR~&vMX^?uY=nN6_RCKC#j6?PrqlyU zo(gvljf=Y~E9H{@TuG#;?M^7<|3c8kzR>U=xL- zJT5onude=>ZRDYk%M$^%|HLzUj!jf>rT<=upI9jTLWwmlKf;z}7OFbiTe)17yOBs&G7k&7Fs zNXB^1kR+Oc-sVcqVTU>S;@#z{Z`0wK2oo0Tk=Mc-C{v}lbxi=^o*h2?XOF*o9J7Q! z58LXnPfpnS9_*wCtv3Pns5um}obD%pKbVl0V0XHIISQ9YW*DL%2m|yq>e%89h&KGi z=S(4~`{w#`h$!>&(^vtXkJ2%H=r@j~Fjr&)pt_A4{5VNj8RC2Wt1Lg3Xsnf6GRn6> zO*3bfY8;!q@0z{kcNxHkFl-gXn{j@m4TIOuoL;<(jc57hzz~1R8B+P~r5_XM;&5SX z(+2%XAAew46HDjm(amQ&Ui8NL$t=4*As*Xdf~9Y~s%Ftz&^4HKws8~pXc*=JIO1P` zBNnTF6}RHunT?n%*8bSfHyccWVKw8Q{EnFIpu@fxg?)7&SaAknj9;-Y$nYvVISFd= z9{jjpX7^a=9gLq%+udQPkzW3l!!Q%jz}oG*pnZ6}eH63l_e8_J$DS=>_SWxdp?t*4 z!=VKwT7pBU#j_=KL$W(R#P&=o-pcM0nLz2~kD)yOCxX&(AZnIF6}fuj1)Fe4Vy(&} zGY;(M(3HN%PWKGJf*%3eNHJ7)2)GJrCApGSe#Q-gqo3&@+{?iSpQ`c=+KSoHevZSoC0l0yJjg)1e@eTwPGt`zU4D_3g;9`;I^gMgHR>E!(CX@ifrMy!G5W3%VbPzt6*g{`s#hvlt`$_Au@< zc01H7%M`k-*XE!DB^FS&U;Kd>pGio>jd5VNbx4(tm4ByO zERui(B_kOGDT0EC1W6@13lf!_a}WUm$yq=sGLnf4f3f7@wTurkLZ^8v0@ za+Cw>&G6LY$OIB}-@>^b-?OMV1%Dg94ZPLY@>lvC4mZZ|o?lP2%X10T{Xolbdsb6& zz>{GA$ER*eP~;p^jbty|sQ-P-At#EPu*o(4HK68>P^xb|YG4`e6yqu}FB0V+e(;g5 zcQHU3`2kDK`G7Ll6XeM|>_V<a5oe?D7A|eq04f>QfpX;rWdU|3ff!G>v!&1qxitqZtojdd56{P?V~0jq<< z%=`tNmO)ppw^odmgqh#laDMi;+D7=y$Qm5ByLD^_^u@z(>t+t@DzgA}yPXd`P)?Bb zYOt0=6&7C;Iv~d^2H>kG;rN#*Jan^Yp@cmeD5EPx!<8{%3<6DMb`<}wyR!f6-Ci`s z2Nz*0E>r5HV27&*^#IpHxAVr;l^zt?4%g}rmb;#4o0lnCDL%yA-li68yx!+EN zq3Lpl^LYW0gdl7jBHI1U+x`jCnr!_IGNAoeAAT^G{CSbLUR9dAKYF30zEYvjx7U=8 zZGi+cR<2x1YZ;k5{fdSkV{=jFzzV;fnl~%-O;tBKoM}WREF#8LQ`p!mZr#Z;bJa@^1#v*kcBIj| z)#E^+^oybc>MlY`M>{jURL}Z^wL#oz%B0tbB8f4jQcK`c%dw8nBNy)Or!ZVOw*K)h z^7n1eOQ3vuwLpGA)pfeZ9||3E8V2Vh8JjPev%=32i7D=F#p{l1L2up8`N;VSj#5EV zu{PWAhKrkvVp-2^X46^UBUF9Wfi04FcWk@q3ZDI~E&RU>FL&p6;MWoR^P7eIxr~af zvBJhD$3{2H2&XRuY@p6YT{hIwRar~URNEoQ9HjxR-CkbHEO{Fp^0D5I;mQVctOB@S z1jAS_^07MCBRwZ1ALnUB8VPWeD;i54ph_$z-Mf(hMfy2md+!FN%dk_umslbV{N2zu zse`5a;zx}QSDt;_bt@XIR~EiyJU<_6*|RfTsSm3;lOuTuOg_S0?Szd5?P&n1_{E?x zh`mHYA;x9L693XG7wEytY$Aj9*HTL973 z47EN7wk|vACJC9;?gB7&A9CBIybQXKlh?Y{$AXcy9`+SX(OPNn{*lWEQ-%%-7$-`; zdy0ox=Kve3$Iu>5&Xrfnm?DmB;|%LrG8L{|uivV^k%}{Fuzw*7e<_H6Aq_A!_RRHv z-`M?s-R%GW^jX{#?M@hhS4ID#G(;}=_DJT~7efokujCBYcn-~_2zwI@3WCwymS zNwNEzMO*Fq;%lP$*Pi!iGO4ls{j`4PgzAu6je8P1yjP)wnvx2}n}OxL+5&oD%cCzy zH`9iA&1k#N^B5SwmXT{z{p0Y9?fkbT83s|EqXXsjmrGrm8i_E&j2TU}n$Fn}>H1u^ zbkpY$$8DzdH2eA%uz+??ceW^U30g=<$<-si@_xK21!kuhuy*IGz|Z&MOiv#%=dofc z-4d}{1FE!z<=<{On_l9j;Mi_D|@c+$-UR zhS@EsINJsR|5)o3_&vMy;I}!Kr6S>Z&jj@!@egT1h<2;FfzQb;D3w47#P$V!0v0MNqR_3NqoF;>H4kdeLrSRp(y6v2AU`I3r zNcKm4Y;6ePZRf`-#NwAjZQ`sYnjAmF-YXbzB)~MF@KUAY+#V^{xcRi4j-eiVxAf+| z#Mri!i6VjzGpp%Y~qO0H@IdO(!HnJGUqvzL#NJ=+Q!K~8pcp~XjK(dVbrmBzpzH5?#T|N7WOOPd zhkEkH)}o0dA6e7CfqWoDJ#TJR{s;IfpiTH=Al9=ZK3m~D$mhK4D*5l-H5Sv=$}kDL zq3}wjzxGuRoMnI?GD1g=cO$$J$$AVn2kDGpgp?oQo}BOpweH9DY5a-8mNUCo&|km_ zujxUW(vqN$_l)o^$D9u$yKkU1xXvi}!CE>CD!63&(+m@SyVgBjyKhA8OLXNXHTRRv zKsC%ds(_-weD7r3s@*qb1Ipipfs3pKf23)s@G$)M%lZ^OtLy+SKAapZzVUamrfQPE z5$LOqr>{1d-{@?bUQ^)h2$)43;S9lr)QKGLxZ6>f(D2GT*wS46W)OT2)5aU28La!& zBm~23QL(Ese(Pfx#b}cOFGmfYn$pXfs~GCH5_aS#O?lxZO^Kd02kpD{U8Mb$KnV z${K5Z%DcsQiq_%cm0b_I>3u_EWm$_@GnTIW+1|rHkbjeBXxAWUl83Td7GX917KZy+ z4$<(z>x+G5;nDsw^#-^`(x7BHgi9rReX+sxM^;F3tjug3xXZc#m@*;%)O)!UbQxZf^LWG187XB zH*W{`3c>!pR}C!@L^|wa^zRt??0AEN!dQIV)0uj}&K-eReU*Mq-e7jX zI$UeB3q*pFUX}#; zHqc4G;VM-CzS{==)f5--W80TM-yf6PhMf65Tk6$YqLo1YoQN<}umObb{!1{&{Dj2N zYF=TGkOgBS_ireqUpXso8tE!LsGOKfLa&;CRq^^roOvd|K z)RERv*@x2uY($k zNW5uq-<$$vOvs583?7fVBEA{^T#_3IIUcM3%EZ>12RS_bvT0hPfEWmWjGWxv7+a>cuy|7#}lO_qxynCVXLN;V>& z5Q?!)ZO>ppq!3Ct@i6duDdpVT8Ft5Rf@VRPB(<0?6OBTH>OD3|>LzI}WKc6c?fR*#X=OG5KQ z9bmzDqf+AtRl9u$!LVG03}|Y1aEsV5Eo)76GLJlpRM4A7q&LZ=8l*!)eZzdH@aOeD ze~~MYmsXAzouG~etKRoA29L+!QjQtmPLq>Le3M#)B*Uu59ieo2+65s>I1UOphqk}^ zTaHipXgym5R^y>|QLWHcy9l^%{u>D4y{o@n$N86q02_D5P;AEe_#-~O|2_Ti-x&#v zp1PBmz4u`GUlziC>#%E!Jdtm&{rK=_>jrNB)=p^I>F0Z&c1V%a@jlFsj}xT+B-#(i zXt#CztIebV8TCvxZZ&~WTTLl6OvKg2$<;xUvx-mA-EFCt`(nkhgnONtcqreC?n6PX zmyFo=0x`Y=CDTLXR-VUqMxFReW>RcfoRLUsNL{&wk%)<%vzo-GlH_WyS`);VOJ=p6 zh3%S-+fbf!GP!;1)61blrRhKtM-D)^KB3L>FoupErE{9(j1NXRFDo9U!LjX{;l=}~-8ao1`4RUh-sZZ>kw5-$NSw{gb$px<|tlbcE775!V*XMoxQ9Y5L>T<`%i%cmDt+uX58dsL{Ku-rptkBK8kK zksTc4e*-XP7NM<_>rsxLITZywg~aRAF0UtZM$XY4-v29sAz!*4mUr4+sdE&GYQA5n zS~E_&VI9la*kKc4F&RXNaIaUnIQaK2ny5|$^?2o!F3aL?ky~*e6}^qNHGNkGaYfV7 z@t}J~`Vv}|qPnfxr(BY1b|IzEpjXK4W) zdArV1o6`~KuuR{!E!99f2`25V0&$yisNUtrG0MoyHvKNfBfE2S<3eXoWMf~S#0g+_tV~; z71Xk7{{V1^#>QQc@4Ga$ia3lDcPahd+e_+z(jQeg9a9Tx2~0rHZxNfNx9S zW7=$4ghm#52)e2(A6u9Maq zfx|O}3w!pR@tFU`CND|PGNfiX?ia*7PWUwtUm&&|U5*{Z);7l8XXlNxOnA}q9=N&@ ze8PsbyX;&Nb7JM7<&$!iF>_jyX4Gq0wU`TzB4)ymnMfL=xif=Fi8}=TJ1SV|Co)e zKWML}{gUX!9UYqJXD+&5K31vUWD zM8TZ6^MRr9BEd$7t(Xh*>>IVJgxsC?wN909yT4!GC$RHWSsFhkA8leW85HTQNv0Zn zUP9kRk6tHz>d2-*zNO064GoUG%_L;ZeB@?0e&wHlGR_D*Zu>0WzDd)24&WUo=Lc?UvrSLfJTJ=ly5UwyI#T_kAF{&kc2#Ck znG%xW>qMaVTU&LO!o-&1s)WB3c7DyAX|-K(Hw!6ah1iD}ei=cNR0Fi&hr!I}iYlhU zW4?&~?vG88F@zgwmE_(@CYP)ucfqx1$zOO2z;yHar=&U-fjjxO_y55tN0hZC$*j|> zdkO%A@h9{?Dl`n%HLhbPH2WRGb+N^$1nktDuDzOPjJ94>)x!5DrR3fDWTL z{Bm@Th9?Q5ryutHF{5*Am0h~-EnWkl51WhKy-sDy9sw#o+vLQt=pVX=m0w>A zyOL^;%WUE8(9^c%2t0;3m;%8wrROWKDO5Ic|MA9i=1ky9(gmY|y2lXRbapg!I4yOE zqqwZLh!!99(_P%u{C2GfDV~YO1=&1$BKMQA@87N7+B7Z1a%L7C(u;Kc4~TLc`iBOG z{}{*+H(>sA2m}b>eOGm$WHvNswcQoQ(SP#E`_4jXi%fDMyCnf(PTZkoQ;*P8DdC%@ zQw030MZn^9CMtkAA2>0Du0;T8Jx&#A*ZBVqP=5JjxpRC$u;?ABb<1-z3<;n5ELvP! z{+h`6zv0Q?FX%f*rf+Cl5|WI-oj6swL>tMnGvxcE6dsGz0=3j0zY}3e3jAIGJy%M69{3JwmZQ z;B#V9b)bF5H`TrptZ$KaKJxCG9m*KDA-lkBTu7vP!)e+^Kt!}jxUmwfInaL^Uxk!y zHl#6?|6cp#MWz{1+>%L;G0PwWS$5RJ>S;}Og30RaxhtT@p<0HJh*Xd_RMMcvdEid( zxGo^FTG~Bju+}>=jnPg4;y=8!s_D`vo@4pg$LER}D+XvE$;Twr)YvYt2O;iCQr=E9 z>UH4zxQnj?ecrrTx-N2a@%IYK@sO5Av2{6@8p?eU+e>2e`3QOc9FJh)5~TfYJprWz+v@>p^Wr@=-> zT_8ZYJaO{@?|AIReh60MMXZgpQ*SVKUNqN=ENpj9ZR#8OMO+swKDhBPf{Wl390M>o z`U@KyL2oaD$7pg_bVWW(H!1UEu6+9kZ0xM|f5XPVMy3I4{08>Ic4x&Tb^wSV4Fb!6 zRgkeQnz3m?BX`9_a>c`VAbF#ef!3;;q^?#%*n4xk+M7V%m?KAyP7pX)Yg(hr21W!(*rhzaefQ`b4}N z#-HDEKdAyZX~x~^B{-5d?pu(sVpi?@SOTl~?H)O}xH!HZj+5QMk=p$houzs?`1P5m zpCF+w{OD)@khS%E+N^&x;JxO|E~)j(!D9@5<-c@ofvwi1Y(IR&0B;VWUl1=J+ObV~ zuSI13A%`qc0DP@V8LwqeD~ss*+fY6yjC2CdEM9JmFV`kA#N2MXsOfpHY%7-9gksd) z*`>#*CDjMmT-3(j-!-*WV&>?^{Z-cb#>suW5U7IZr@}7(cHDXfgy_#@%sp9mtSW+Pe`g$s<-)V&wat#Ja2ylH>W}T5gV?vL@T9w`U%0 zUbiWu*NjyX)Qd2c9Gu8rc~J(+2Pm){Uhz!Fh3BpmzWXk4wDn?UzWNM)mR}n~B>+I? zFLB2$;EtW6>7;%x(kNr0@vX+T{?bV*e7?|eT>J&~1IVmNqzXCs&z9kMUkYj+o~cV9tBoa(?i-MB*1ccTm%1g% z*(!%(*?*|o$5Xlh`hyM-+)XOk|vY3)aIPkXU3){_B>mS9#^2X!W{tJTAQsq{L76 zmw9_xrSTVlTM<*X9)pKTail&uY@G!Jp<9!)I48MOR*nO9>gd^QF|}?ZK}9CgdiP!f z&`U5^K$PpRzm(f%tnB%C;)>9Cz9ZFeN;Ue)2IbA5#;)cDfV-iWlj(M+9Limvkm_DE zw!r=Bd^8EIIHJOy%BNY-Ym9JRN zJly_7!?mfH8EZDXoY>`P;Oi>Xb@hcKLJf<$bWhnoeu_nvo}t|d4~0dC;D$9UV6DCB zHe;7Zc{31iJEr;RAp8?lFt9~B!f2iQ!)Yw^n0F}y5oQyfNEQu*Cp3YDpxG=o`@TTz z(qatG_%qWJ!Q_7PL(iF*E2;2DVsZ}<33^ZV7`Z_qqIg^;@~Xu9cW8xfaTBdh#CEty zzU{OLB|5OUcDhjDH$Ky>ZUL)z#7z~$zOrKMJK-apln7uFmbTqz;ZxWdl<=J?C`qf7 zes`~^1iJAL3aw(ja}!d5@PJz4r{Id|5mJ5Qs7llRKY62@`7 zaik%JWa}TOw%qr=LM92j=Ya!OrF&D;VT&k@?xjB(bB6_ztB|=?d&ZWJm241E)lLK@{0t`4`e6-**e2)|%C> z+DHn_Igwpo>)1KZSHqXLxI|6`nq7EaF|_H|u{x(~Iw7;ZEf&|Zxbv#!>(@EoPe|S{ z4ci>T)6wov7CFVvRlh>?EigwSAjOn-tYJXcf zLKpVHj{pom>#BG#Fit4H-!LX}g3g>uz*=$mPJ7>-b#d^T%#%>;$7 zCm7w+Y%SI{we802IR$+ITP_Hwso=<6E^xSfu|8&F@X4+bvb+uEla+`HF}tU}X5bfy zAiqmsuQ*%wXMDqSo=Xd}318mVtrR&DiNtcMe0ik{5$-aAIgOaY0(`b6&0Z2%gwFi@ z)4fd(tVMKIJ9f}Ia;1!yv-)82kIS$9X6&}MsoIJz#*(K3R|I1F#aYqT?$PZ8iS|j_ z^Ye#VIQojlF-YT#h4Qh|C)yGgJIOQYI)SsMnt`*B#adBymEAYTkM{xa&iJr9B43Oi zx9)5eg^(tlGR6fvl*h4bfC8rGz>byP-AFr;zc`hYEQ_w61gF@G*DM^kTnZ*WdIS1| z)K&}3(@%|8eG*j1#Or88CTP#yF<-0f%s7{S3P8h-V!f6laPgi=HjL9m%d0l1Ac^NH zX~SHF-LKa@jlDT5>Pno(xNar0kL=9wwr6{XihYXGP-ZikdvG-OZ`$)o=rdlHP&f31 zMVf0IlKGUi2jBIZ7#kSBDo)hcTtUFQ^dR+;ncf}6CZ_{k@w5Zy?uxd=gyhj^0EbG? z*|NX%Sb%e*Q8!Lb@!$+uo<@pUf&T7fof6KpIV{x1Mjv0Jx>pOgaD+KI`EJi z@@v}Y%6e!=ll~>5w41HwSlJSbV4qR0y(Cw@kt#AELt^*px>ru6)dqddq4jTkPCiHr zopCB)`3G3FTuucx?sOUk2hBEpmTsfH28K6R;Dg|ivne*Ctkm=YX3V4r#-I9OA1|EU zO!DdyG3~QvH=It&|Mb!<@=my(w7~f>QJwqL!dC9>*P!qkxz`}PcyBYU_3`f~AA#jk z){11~k8)C>lh-hpM0!8I(lY>lXln4=n+u&FP_pvBol9S_K#I2IN82#ovd6yz@Sfl; zQL|zE#6t}}4?~}IJ`>f=_79q-KaEuzx zzW*p2yv{6^bklg#R&-45jmBC*x(iqJZZ>ad}w_e#WcMQWsIH zhP?0LjPyV6QZ`bx!uZMB=0>z2BTy^8#SD0LYPZUf<@1sy%tXbk{5#Yw24uA19;#DB)vzH z-hvR9d(*ZWDGcgrrdK6@Jj0?I_1s*-prx&7noccZ4i*0tTy#cfnqfvm4AjqDT8%R| zOSpY`K>ld6og%qr^^gd~I+jl> zYmnB^mIPR8%i~7JPf&*Ig@q1|wID9kMIx445QJYg|FoQ{bQ*?8fcWbwjhA3HL1Z`1 z-hSquTp*yh(QKU@Rgz(aH{DTzwPEz92|j+N+Rx3U;FkqEMvdc@VCbeUgpPv`c7>D; z9{dqTHwLfMi?9}RneSXVko|h$^n`5;F0Pc(aw^bCM8p3UW49CJxm0{&x^qQAlyyA~ zd%_j{@V{xzz)kK=J=~cZ%M))9kOaF{2p;9?EJlgeqlN`-KTH^A*~6PqpYS_WcPSRa zx-wo=J|Hwwt4oWvRQFa5eKWY!P1HI3Wz@?525|VDS3Fub{B#_-eNOY}%ORJu$xA|j zYIq56C^DwJFQMXSaSgQ3BlsQ^FHZCbY0Fc*pU^xCy?Y6a4V3v6Yks|LyOkh_zVA~T zXcxl0srAXbb?(bQG-m3dgS+eC+>?1Tg?^0@=06II{6g6Dl+${zh~^AFr0o@8n<{qm zrn-%Sw167-e(koGWV9A#XG`21!b6xqNbxu6?^6Q-dLSr+xpef37pA?styn`f-b&w7 zop|Q0mKPp=a$C{sw`aWbo>%0LbTaSP-JG8+Z1%hYEH}@yg;_^%-bZW`)qw?0KepBN ziy#_H$=#~=^F?KdF>sNo8a-z$gt-{}Y(|7*nCHIN*dQQwr&BF;)-4xIpT3*2z~qm@ zl#FvdIfIXHbb!n7{w;^M->D$>6A_4T>jf}|o;*jz5!JO_@p+I4GZ}~9eyoMD+vD71 z@L}CbU0`XL=$yaUVFs$FOSOR1$>u_Q}kG^coef@p7!!dq;u_VY?nbe&F;GKS&0@PqmB_XfcTS_ci% zTkS&O;~&s6f%{feidAj=CJo-lBOr#SQO~?SBZXulJbf5=4>Y#VQ|Hnx*k1$s*~=f4 zBBu62II}t9apyEYt1Br$%DczvzR~q1;|p9=jxg)1%)Z|wOqcYnoS#sso620+$QWa) zM+d+Z>5Q3ye>P16Z9%UOX>a$;mM=oFI6~LjyS1U7J*fL;cr|^TmAyY-sWAACAI#B; z#mLU-Z&#OevKnaG+){J$<%=~4n-(W}K77KYeU5zH?1o}qh%8u3gXz_jfW;*GpgkvO z)3mq#75EL)qG&8#XBLs^WuUSU&PLijQ$H0SENb`@`UD{JBRy@qL*e{rs?v`wT{UzW zLH&W)_a)dk!p(2A3#dGdw&~6+c{jv-ypEb%5-_{>o0oq z@fm_}8ra7<{%UN%E@Bjtv9V}_x^b5<`%Ei6tDr^Hs#fbK6_#M0Mv^ADGp1H{Fo$;-7uR=MqDW(Hi*xL{sHsqm$va zBLv~qtXh8yi|L6J?v(K2thu2Obf7%i&-*aYf|y!w`EFK^rLqNI?AY&eNE6NMiAt(cuapFi zQeXhR^0{No>kml9gtZos`NX6Uh?5c!ecj^5z`i7ENz3QV(lo1sRU`=%T}v~fMa?dY zU=_a$Tj10~ysX#X6lpr|wS}_l&S$OT+X}U6g3HUdu52GZ|I=@m{gKND;d;LsIw$LT zQp_-<HcLkZF#Nn;J-9%a@V{V+<5{nnRl!EG=-{{ zv_JP`&B;Ld3R~Viwm(yHtUw44{cgI$%uk5!3$M(COOboB$G|QTQ;=22btMypS}%uG zwu4A*+A%oIrd5+4WI^U31G>>f{3`R(6_%tIc~lQo{z%x}`9XPBW)#d+No~5j1zDDL zj$c$RzG~;~`s?D`$qd@W&#L?W9CKmGx9`U3+y1-s%*QQh?|5gA$;ZurfYkwzoPM|$WYrhmLFz#oC zx}OXXY=~DA{P@%0sF0S51E5qS1a-ME$G_8(>wTS9vCKHH>plPHtKu$ z)Qi(VB>OT;wpi{Q0o-GPE^nLM2)&g<;f1Tm`WQM&3=ad2zjbLnz#ocRd@sxlS{43> z2JuOLFw1YKwwM{Lxz?4rRHYJg#@=1i*0n0gkY+@kBl?5H^ACA8nYq4b-T5&VMVV@r zE_;-X7+*_0FIW#hW$-?wm^*xZbDYGf(D*DCuZbgI?br=$-BV3wJxSfPyCmOZQQBjdwH6yoT5U%KJ?+?OzTJr|;lKcXI{jk?m(` zOWvyIq+ZDur9IKtGXJ$TC4Uiv)d;eU#+EPXe-qwP`;s#@gr8YMF(X9QaU-wQXtL9) zas!xj`T!nc3~pcTh(SPkhfzW|w(dcz_VZRe_kjztJ<-%7koq(NtK?IH_U`MGk32v+ za{c-(`oWFkXClvb;JIRow92_+FZNmCO%smu)emIh$=94-!3uLzuP@^7JxG<2ob(>J zu9BKVkdKr!YUqnedzm#BN*qd*pbo`;05UD~fo)3xKW~n4Gk^5)Gtmbi&04!KgE6mT zQe)XbuIhF~QHI6)v#!XTOfj=-B?I;a_lzudslw0mIGNUHK&5S8T2ke;mguZm!MUQ; zz>jjBy0*REr^7{fAL{MyZ;+0_ZwIVePv7}WG8nJ@c_`nKw*7G1f2;eN;`6& z2$}xj{Wk|^4+-{Ne#VBbemF5_2x80!Fq$y0ZKDTH?!0|;2|}+r-#Jj(QT!Q5p|WV>&AVz zQyb~S+cQAsYMD~l9!ytTLL&A z-A-~9JH|o{-`mG9{Lkpsi}(OF_x1^I0i)GH@y<^9KlPci!n4T-Wm@mt1S^>PqN^M< ze7nn|MG7cp@@g&7sz5sy8h_&*I;V?4b6?u;9P#dPp{UMIZ4n-HpC?||r41L~ek_}D z=RV;Zj?_S*3m<**!G8t@GmL#dd6Yi$l*W4Q-jHf>td7kG;hb^}A zb^Fe1jUDl`ark^B|HZ8LobWPcI8o{GY+I38;{n||bDby}o~}Q4I!(U3H{eC7SLqoU z>3M1ZJKVTy?aN$j$o>W3m(1f$cNO*c=XUNp$Eic=c?!F8h`3E}t2Hs=r?3Ag;FVUK zr}e4gyM5Nbe|CREypg5`CUdZEI%544!aa_}&BNh4aDG}>6C+dq(^tBFgd}p;?=9m~ z{CH1-t!;C%fF!QZ!W7Hf61vN$4 zwMJi8&H}gc@0V$tBgg5&EYlW-?uAakG;qXD+l*EAy{#{Bq5SRi)I<|x6r|42@dxWo zR%>`e>~X#cE^3R~_5Y++DeE0++mnD`mlTT{5N~;}jUlUn@{0Xm9Z9s2nD{VM6Xiyc}sw!p9X*+je*MH$`B?wiw{31Chx zXP^!qc4(eh?RtQ$Iw;}K`XRH>@VQBA`QP%cQJKh|y}L(j$)Tr2Y>A`LgpbG9JtL^A zzYwXs;P9mj`s8zr%5S`~yzemf7E>gl%s{X&+rR22%b&MsgS|(b0<5|^(Nd`BIDn}7 z7QMlQan)U4^b){5HAxQD>+lPtSiQQ>NV)}4+!bK@<0q-|{a=Xc9UtJFy7tgy3{b8# z9UhV@H`69BZur~1kKdwZI7je^snp(1y{?SmzFtO$CA|RtXY(q*HX!*=zAik^DlEx8 z{n#T4+s^a2F!5rYD-{h4n(s{J;+=X`xFuzP&}x3<#~B06RwK>46f46cfZVh3JhNrA zSS1&$gZ<(t0aj9tzetYBQPG&(MxaO&Ae*D!g5(i7gUUNi5{cGqYQy-X_eMUjzCVuS z&ci9gwoV2|NlU5MV#$@h^=Do8etv+Qto`}ysx_BvRt}0unWDF#c}cK}e-xO!jB{+G z44~G>`)5l{yqnF<$9yG%)H1${*B~J^5}t?DZh>`VnpjJUvO8DbjcH3ta*MpUDI)-K z_ySWVCWFQ3PfdiW7xum@v$LUVseRm1kas@xCx!+^uS}2m7!XMw2;?m0*k*1neGEZ# zmsVU|#5}%fNAk2o;@gL-;`Z%#6Qc822JTM)8Z|K#tf^3jm^7)_KB3_ z$s%nv7MRd?3amC~za(Am(qAX6Z>da}as@q^?cR~b!%aOvK-+Vc3I^v@X2}lgzYL@8 z%SUeA`D-Cv2E14pU~<)3*~pcz8kv9)izaRGZmjq+=_(w!gYf(VY`7F_y%47lWAB3d zhpn6hgil~o3t#Dv=5kl@jgRFTOD13*VEmX(CW7aE-mVR!4oM5G?y<&S8w`apygj_( z+)Nov^Umf#>ry(Yh63}s<0efel-mi#=Rj z&`lp@T+5?A>g-nA-s8lb4xS!as_^OJ)UZo$+8=hzX1H^^$C zGj1wj>3knw7GqH>O%8c`>()wp$EfJCFEJfQl8=0lbKjgxB(tJjGVXhwY3tK#T?%eT$A9HpOeU|C_#o7!ls_< zr;VJvyoe^{!l~j*XjPQ6G2NNBsi24$&za(V@FGq$mVWLnO%AM@Pn1z|(&a+%x6}eU zzOise_u62j)1F8!RpuPk98PH=RwH52J;sE(q?(6SpM+}e8|Gx)^c_F2llXqsJ*qo+ zMv5~jpgI#t&tWQJdPjVO;d(Lh=Eb6{oWd*FcCSW1ZH(L;g$cKO(?sxVfgmY0w#LWC zP5RI~#=V=YE<9xl#<3?aUy5}{IB+%xV?IB5{NP1hVfG8E&9uO!Yr-#7vfk)Q*Nf{z zrXIVuZs{BcknT#z34W5ku`ZjKLguT9OUE6mq$q7b1(T0bm;2C6jS0cOniD+4ay#9B z_98fdETdXlmXv~s_)g`O_ov?deZTxw?=2Y4)s3B&h#(Pf5i7rHS07z)dK`M-Kajb& zN#NtoKUsab@QkuO*wt2xN{&@var=u-f;ssf>hq@TL)qY+2OAj*r%rp3O=)&_Cavc* z7wgU2wa|)m``#tl)cwJ$&2ari2Ci5b$_Ke|t=fQ=HI))OVaBQZkk!P3OxCtli`6n~4(Z?8@ySxT5EKt+uXaxPs@IG4wXT&eZjk0Lc`>N8Vfi z%aq;q;s--6`t0RkBk!mzWh;UhTiY3IJ;ot{C9dj9FohB!I1OydGI_mfS6S+g{1(a0 zjVqU06-mtcK1hO8((K=7&2W>&@nk>r!R<#L2K_c0)LfU5A?_)Y6Q%qLl88$SE{z;J zBe53v3tWF#nTsnn24;Fwhj4q-Dwt5As;tOKE$b?$V&PTD`q7 z3;r6jAE^L+#xY#cXwf#n3<|FL*e9v z>-)?8TN&#vmu;wEtHYw0*L$N&_GfAPne5=R7mN40y5R4p#^aR5(JXM2Cf)^V$BGwr zZwDLueUOc0g=b0^;#$%igX}<0+V=vO}?OH%hd)7Z{5WhkV7!@v4Ju z=k0dxp+0Xux5j3$bZokbV_%dURD^l9nCh6 z+Ft%1bzM)?2BwHIhGvHCBS~M z&9z{-BQN@OAcg(5eB{DrbFgU158+?hvhQ|Vvbpb^#}nPK_g}MyMct9Vh`*kzz6U3T zlRL4?BTQDRgwScrHw{ScxLd?F#ZcV(K_=txZ4@yhPmy64&Wi^B@rj;7y;@u8F!2KN z!CX^Y8XN5Uvv1*Q&nJyHW zLHyrkD9l&5J^$Pe-fOyFV_uqybv3xy%-ilHtaSLoYZBA7jO1u`XOO}3_TC#-!G~p@ ztuG4(Q<%*&LySeicjOQK9yOit>`ab_L>2En69}EpQbaR%!Vl^0d=#dpxojg6pt?lW zTScW%W6P7sy$Y|x?o6sW9FB{HobWVi3^neYPe_q$ny~xA9M&}7u?tL&Pn6sn&GAy% zw#@7_q7I>uD(k2;3#GX8V``4|tX*T<(EG~qt4fFRI;XecZs*iOkGBjNhxc4;aV&T` zpsg3hBHuK|N=DjA_oSx%VB~BBKDyr2=sZUKvo&Js%frE>xn)ZEH#ah!rrld1er!gf z%jIx9^Dd7!<)zbdl!A&mFkYi2HP*hkiR%-y&P==i!1`UPk9w26FS5S2(HFMsB`|a+ zm|_g3OMnag=J95ccfhG?uX&74i?KTww1;64y)+Xo*(A{6QkdF6yF!%%omKS;_!$@wUnXTB4jXW@bJ=AAekIf6Qky*IT{OgMWbt zOK)&m&U^j#x>pd5%TIwL0W9YZL)eM4FS0P@T?ta-!#uOMk{e+=uSE50&yP>+ZMRV$ z`KgB3g6Cg54>sy8FjqQ=MW9$^@XjPXM3rwp@nRBK{1QizlJ6CBOZG*3-McsDB)qNJ zUcIIIA_}l@w~z~kmtlUTO4dQ>U_BiQr}g2^h`X21=psDRk%J80BR94^rur;?`<*m+ zunTNm#fmkN1y+>NLa~dR0$72!f!cRZ{Ms6xh_F$ynuxxBMd3>nj+_($NBXJYH%Us~ zfvf|CEft_>m%J4dhrEbqxB9^_~s~tmO`ci<^(22wa7#-PcO)zF&sGC?^GQu>^m0%52i2 z@G-rP;U2Xoc`!M?A9iiLF+uy{#||{X%+U(fGxp^jSqhaf`kyrL>08?fPo9!l$!mkN zfiY87m%*z`GxGhMGDf-}m$+NJBavG#SpQ4JB;BHjFbo!%wiBdFO2cX*EL^ID2iux$ z+frIHJ*ca(HgyHa?H+vDA*Wd?-}iH8G!0kqy!D(%R5>EUkBIMldCfFU=-Pd4iWbh` zvyV9@;n!Y>)9v4*A&rou$e@RM*qfp3>kAA<_%Qpt8wIyz)x6!Q<4&A=ADt_@CAH$g z7-fFNEDAZ{!G0=|<6;*&A{}Yk8Sgz3Ql#)9=NfI)3wJYB7a-uYo@3As`t(a5Pe4zQ z#hpaPzZZuh4e4-iDUE6G3mrPh94Aet57tiQsEH3#-f!D%pz}p?F%N-Hf7;A&4WeS- z6k%zBQ%@pO(?7MsjYYq#H|m)MRR))2(Q!VNrnr;OD^2E0G`vZs7^% zmqJ(p6$+oD!C5Ry*ib#jPOB+uyi<1ze!(13h^+LGfe=ZTQ7TZTy5=Q0eCBN7_)Nb8 z`Zdpd4c%4Xwae#Sd&HMQ(aty&v_|RO*Lr7^7D~4KO09#TT1umoq0(U0@Fl9IfBU|g zS_W^PCdQf3mx%r`|6~CvX7f&(LP7=-)6Af_hpMrqwp{Y`Cc8v&R<>weMh3j7B@o&p zcVJvV-t2l+lQU?wW;(b&MBqTYhk8=k;&J7&jZnvB=eDfRIn9@q?{Oy$aHl)ihnM6} zE`FSZogs*=ALiI9ZrEhRGc-fV2U&>&O+(M7_EB-`7rFE`0rIcL_HnKN_tnc4f<&z>pD>IC0wzNIT= zLBTB+jJaDELhXKZgSYOz%~-ZPpf{`LD84HbI#SmI?FSUzBu}1$6u}bGo;5eJ= zc+v@8Mfr=RVzmgIZPmKUEnCEZ%;(Iz7I!FyYl{daOc{0Mqf<8W;NfihF}ui<(hw$a z?M_ySQxa8_P=n#*=Z1PD*pn&aMq74i{$9o?l3JGX^(rkNThA_*$vm|oj-guK*j6qxXUpoFH z*V@XNVp~?O$AFHX0*-!6H-Of?WC{(;4O!iWbCS5#Rw@;d=7oHde1q)QRt!%Mew!EK zz5|aEHxm4JSn6%Sz65Kdz*|3fOvtXwD*aC)`lfbLUg)wi^afv;gKB%s1i>PyIQo_B zLgoq3D7N`kgFA^}Vyo-iA1TW$`~T$d1b`8^_XjoyuBb~bkQ@`FYb^|I_O$t#9+Xu(u+^_vZ`y*Kbx2@_(q2Itofd4gQfIIdGU z)wKYElwg{p(XNAhvArPZETCeudQcWaP}xXqQ$xnNI4g6EsBwP!wmhkR;@)s);s zS6eTCbWaY7LJVI+6PZ0M=WQUAB)4Z>U?aq0=)b?sVItUVkqu^L!J^y6?luzcqJmr1 zpf{1xpZy2cl3Ri;1g&mP^GCH|UWj5WML#{z+ zwfWHH$P0SQ9#gTu$IG2D&~1o0_^2ypy0xOatF-@$XKq0E@ffDHW@SG+SScfCt(TEo zQ|Z`*Sr!BAlWecD?rXZDp}LzZ)xJ&i4QB^wrX#(wp`h6y%@%T;V z{z({z^lHp9-gGAypD*MM{)eye4817GA69qL-CeNM!V}vjiPSKcnWgKk-Ie>OVnxhn zM7b0B)}SKEv?trm0}2|C=xKQTH6Bv~12#frpjjN*h+ z29u;F(ASuHy*!H&#qT%Q#`r8Bpe~HGgd%!Q$uu=2)p?V+i_A=_Rvy1wP~ydkl=EKu zVGMtdxaAYHpeX88VfrjMF=xPj&c^o0I&}U&7NA%$va&r^dC7@2W+M&3@OP0WuIKoZ zE04J9=xsKG;4vxbRn}MU?JgPjp%TT!k*AIQ#ftUF6nci*Kdg3<=g!*DN1ldA4yWV_ z;?-G&l**I&xK4$X0N44toq-mQ-#qA9KEdJcN^JkcQY_2^jS=m!0lB zhI^i8F0)kpFOoH+7F6SiZBipw=!@T`xa5g@^(l0L*Pov;KRepAYlU`pnNjl!h^wv# z@q?TfFWM1s{R-wc_9wrjgKN6`K@1lePSxdQVVq^QtgbyWtqN357ud2gI{74Z>!nrZZ@ocA*c3EbM+`@bz2!aE|;9}4y3|oHnie9HmC!Ti4uN> z#Ivl-DlC=IsfsjJM?_k&4volp%638h)}_9ikmS!&6Uh9R`{f&S+ug~AocQmN6Kgps zN#Nhv+&S??SI~CIG8`Cti$ug6v5%ByICDL7;R0t(9^`oZR>Cp0Cs)OiPFnxA9niZtOjdgA z&oZEP*M0C8_^GEnlS#g>2ARV+cJQIQ_n zIamkpnadN7pS&Kwujwda2qu8V{$M2w)_;vw!B@p$oSL@CX6$G}$J#G3ppYU(6rw|d zt0sS5WPpHc8+JW?NP4x{-2G#FqH9fK_Yj+{yeyc>AL(XF!}|2N6P!oyR*Q$>bIW~v z`5&b8MtfYL{!4uLYp9zz=$AY|IQiqSVWYKjzcOL@$8?* zSr-VVBT-+lwx_TD6y(z>SX6xxcI}^WMz1QfM@b>;&ktgm{9kzywb30N{3PfZ=(@VB zzcV}4O19gUCpNWXncRG$HR;nHCmvx&$Y%C*{RE$CVQK$|BPWi{|$xv`+ z)@0#i8zi;dwL*&DdV|lOy<|RC%pm(u4DFTl7K^c@Ke8M3HG=q6_oO%Gde><5Fc__z zWE!6*bbLIBQh6FL#PR?@>7XAwOfJ)LSh5aXawL30YF<(-20z6GIA7a4h&A;6I*?#f zioH;)cwb*!D@r`r{17?ax4NQm?z-YJN^?!X?mt}>w_(-ozl_wdaRwL;_Lw08p`bSjqqybVTxnr$`X+ zHjZRP3$*CgcMKtBu~oYxpf4`a6oO(^&x?zNjt$$h{+kvJWRvqx40Cz>={fZy7M%&qG8YEk(2|H97ym$~lqCy~sK zIMGHAX4wG+3D6pI#2Mlxe)Z+cerJ7Nvb9~|e15XEeF3=vXpsnECwRyRo2eVHV0As* zt#`(%J1>zF)L$!)M}hdk0Voz7wGk0>Xwe+{u9$hW1hpS~jc-Xf_|b-3`px!2zRPFG zE3KC8tuCVd=jaL2-^$OS|cn-iBNitVNF5pk+iSttu9aWMZCnJO|P^kle;7A}*#j*PPY z*Q4H;;vVP#Pr`HuO(*L`yu)M-PzPAWLzHQ%=BEAL{aP6Q6J|LfL~K*m`z_~|y0UerUMZ8s9>?2=Ai7#`*aT?gc#pl^$;#Cb8sQeSd$GP6{JV)6&1f^xH(0Ivhp5rcpn zts6Q{wt5}!LLr+IL>oznmC3CBT^5i2jv#-01$#vo= zIF;%sKiaGJc-t7C%#7)$o}JNSvbAOm%~VG?Fl1w+M}tlR2kXQFP39ABJ3935`K`IVnI+XorIQksq$qgbQW;JHd9z{78hjkn&uxNG3j}WYnGE0ea?| z@CbxNr0>c&`^wY5G_fde~)?H`d1(nIm@B~Y)Y<59_k5g4jby{8Gnywm+=`s? zjl%&oFo$f)H^8!1RBC&VT@GYxheW&=UzViEulFe|35MF3lH5BN+Tst(SOmeNhDa`&Yo* zzandOn9Fi(C35utY!raAt>u9_jL$4qxg6to1Vm$~IW8jA@A>87VYTRsSXCvm3fl05 z`fJ3ddENMfJ@eP@O$dmEAwE#^zCI<7fL4jp=7n*#V;Hu;vPw!de9`TX>#12@WJ36_ zrlBJrrjt@ki1XVTN8WuYK|t(INnepwR!uMQd$-p=3e5tk?9I@(82@sX{;T5%n5 zXYdAt$7=8{M28&izvFWBE9@(`m%Gso$N1hUz#%WOmi{hY}T_aDxc&3HD;Qh z4;_~jnyJJl&n^W=#<;DuW^1x$#R@Qe478LvuwhOM0M*1EO#K{O=rIXQD_Jw768&B| zyW~S<&LheX7J-4pHu;9i)g%(|FPDJqgMp|>f2H@r>{9JNi%zEL`puRi9n9aBy%5vL zbLS*yl0&q9HB%P;6aYSWDx~~Ax^TQtHS2k*R{XDhdZDmAMzg3n)VUuddD~VjoXx#7 zNufLf7pKxXihd*`Q_8K+$I9H;-(5>hsan?j$M#Q(C=#TRB+ zKm4MgHC5L$az@Me#Wp3l!^@o-H)mZrN`=is{#V{OHCg zr`I9$XPH8>V#YSRHQ2GRZMva2xhlv$b6=jJdo=VLTbxt$H4&z*oHCC@Wt6b;$%*;D z-zFXjlzB;a+7B4IO=P;oqn)BNVp>SU_DJt%+eyfU13%SWb=gEuH8j3gRLX3y^4}^} z@QyI9Zlc6M>0F<=$Hp@a71c&hS$~E$7Dmyn+M)y#bd_3j_kWnR{ABy?*(#^3M2FEM z5nlS;SE6u1D4$Ove7hW$4~;KJi>w>*D&yaKK8SbS?xG|V3HEd#)F~LB{mjs<^&guu zV@j)H|M3f*M>tbKMaF|g3^-FKw>ko`U)&gx;YzH5UnrRTX2{)%SfonXzJtN7*lH03 zXtN>!pFSu9Ocwhk>RP;lj+`)1L@L?)(dV_>Qf5O6Lg@<)2*nmT;q5-!H^FFJiPa(Y zD@IYN)e)@_qoUM`q-#%Rw;mx7{FcwKso|S>D#G975}k3uf-{BommFN6Bf?V^KD(wf zcf6Wru{oNzwE6c?#5Yh(O<1JxT^?EEsD7BkA9GggN~R1uklT;}pEB%H_Q~$ILL5cE zSx{vj+giPFFn%uE+LWPNuY!Jt@*}W;*Q?+R8oN?LbZYDTD;+jv7UR#IxnU6E0BgmDkXQYMXK@z&Ej<%L*{kuia2xAZ zZ6=gGF%lm+!H++WxyYs@y!9v8O6VhebU*5kYuCE(L;toj{`!Dc>Qfq>YmU^|Bsq75 zTnD~gD;aQ>2Ya!_^nG`QJTikCwd(+``z4sfMt;GbLH%N@F0-45bkv_#QaO)}cnfYm z^-GKtwVUa=*~n@At7gC?E4o#SNf>P7m-;=2Udrd;5Li? zk<5XKR5q*rAmLD&6^~9Zm~rIRfmJo-h)SWq;;1kAHQvvOF#4Bmo(9ePAMp5;+qIZw zAHNpGY%t|B;8bogJrcP(ezK;h%zA4fkpsClYs%W{vPO$)1t6IPmFH|_r^%ILzi`G; zMBDw^Pvs=B)V;Q#CVWfWkbSy&F1#iD&i`zyfk##H+aBeFWZy)#O8odlyh>qxS!B6Y zba6RTN`U6+tyRA@`n;szfhh9{Ul;i0_-%ce$}|1o(A++UX*vL@bkZN->a9exqtNttXw8SBNtansX-LgMOS~k-=xk?VOW_=`T_LUWX{Kb~5 zQ}PWRNtjROX`ma{jK1vcP);v@>BXT&mm5JOQg2+JDf|hUl);3J3ZSLJ6xHfihSzC6 z0jS0u>|YKpEXMniNBcVVI|s#p4>i+Up4J@@>^5AaM0QRK+c!P>Ny}_{^elSZx?e`{ z!ex_Yb%dvJaWc!QU-W$70dRgSB_A*J9}p8a?%Z$vpF`n4QAVHX2tb%+xn{$ct%ooh zU2`KRp=_T?DsWT}G8pxb=mw7|^d|~(d_~-QWjSbeI9fS)VVDr`TN@~yoc<$^=&1o7 zZYsri`9IDb%BiL5thWf+HKHn(U$?27@BR$%|ADcSoc7DDPrf84!4%C*5R;RjQ;W4J zHXzgEF58eKRfCrJdVsgTV{3#Wx<0kamf_1ko#d))e||RdRa*x57~yNU4fam0KoTbc z=p;nq%s0jU+texYLVH4bEAK^*>c{jg`MWeS^fGTBC1uZ3IWKgVXT_|`vkStaPc>QT zNh~J?iL5TJF!{JZaC6T3W7IAd3V8LiKV&H~OLNz+!@G&IZ=ju9AfX$a4jm-Vvu|Xz zE;7->r3fyVV@@{l4y%lw(ykz2|LaItqy2NMCi|U$j^O_GS`@kRLXReBmfQ}vGC8?w zzIyIOxCB0Xjy)B)ZU%YC4$~Po$d#A zERi186vn!1rzeoR4*^jGT4GIIeTv+Di64`Ok3JtqS@_{nNUY46o`coHAnz+S49)8Hd7gFIsYtWh$HjOlA>)Pp-A34GK*qS|F zYvh$Pn{w2fCPM2ga*VhJ51x>i7TD<@A5! z;5q>`nPvOIrYAkn`OuM3g6k*Gh9$(OcVJMT2*~?5Noc&5%lTCw zB^?WubIm>aB={oD7aqd9OhX?Xp6e1LQa>-B=2~lvu@|fSOoE;mIK+RUiwlUJvPVvT z2Zw6KL;L;6PW2 zBkRu-WisQCj0G*P)t{1?&4(Y!+)ub~5rsilf$UZ2ylEnrKhlhtMGb4%ZjKVXMk2Tp zgISqnWw4XgYvVA;+%RcME@#`c{W+;fWJwKRK0mlAvAl?^?Rxf(~ zkRe;ky|MZRC8(Tv4t@(Ia35%7^y}EK44Iiey$2TS z7*z{x?SblVjQe0qr2Z+@1HC?vGx-z-#JqiG_=DiYT0X|dAK6X;5X$=3eP2GFgGw6w z^1>R!1GSo`h&~hgzvIyBuD4#}T1Ewm{OEHyttD@{`w*g=q+qOdG~4N)sy`BD{6sRX z-Z@@FAH}1KZ3$XUU&&D)87G|f?{A8Sq|)M}b4NW}-f}@QOqKXIcGRvAT(P5l`uYOg zk@k+p!5RtvPa?My%uQ^2$I*a~g>J10l=zY0qK`f2KaiTH7X9p;3}PraMLdyknjo@- zaa?J!TUimhMz)5};7zN*A)Z#>C@zT!Q-6ZuDHUFPk>o=EvPXn%MK2c z`Vqho^G&CKRq7^W`0pmfn-XIm_U@O>(>-w0>99ZLgwwPiaIVrz=EKu>yLlNV9~cu-VX*3X7*nkWcD4PTr35Wldb| zYiggFoe2>Fdz`4EW`WlxA7#XGYnl7?B!b|Z>M6L@a&*ooz zuFKiXITCcmfC1&)0^ z!(Q(hBNG0;Vs4zJq_9%4ieWPIv>J;Bx*rE;5R4`;vwIc!yZ=3z^)PE zQ?c(t?E835He&0k<&2T9?w%Gc;FZ@lwAm2s5v-EgW+ynkLnhHi9K=JW3m(-T7+<(h z;q@wqs;;M9{9}Ffjo8-OrREuZcv~)kgrHmue1I9F@<*2 zX;M?q+W4&k`TU%QNg$K)Dg72avO#?5HpDj?0cRXXlXcwh5U_9DEu>*=(UF>I5`6p{ zsdoL~au@f_*m%g3O)MJ^w!$uv16nwnEV%db051zQygxQ)0~oMBzEIp8}iqH zk!4{cGu`l9Qq{+zJdrd(rVDLJPk{*E3i~@G-2HxRA-fy!Z}_`udRI{Yl5e8~=MsdG zb>d}&PcK%taL?`qNKZ;2r{yc{V6dV{TWK;|#ycxX`eE28-Q!#RTl)CH#3qo^BElM> zRAp6JJ2}q@owN_)+{f1FY$(r+jYPNrSsT6q0@>BX_CWVf23ZG*4aCzvGR3NWhKSJ* z*Q3&mNp1~QeXw4loVQfrz!X^wW;Q+P8L_P~dd*`R`zB`f8gv3H=5bB^k(h!+Tz{m1 z+RX-BMT9Uo3i~RR%8Olq%FBh*>X|Pp8+bAur$?rKLiDP&II+V*P$xMtX1DNz)Vl&Y zKoLuzlf?)6xZ&Cvv)^Rl^ort6|F7T@IHpAP)ZTmg*Zh-gx5SrkOdq!N)))ggo zQMm*{&Xiq0CNB8qRw#pjga+x^b;?Vp`BHGOi>F-;TZo|G#`=8ryso9AmokOOi5KLc zSl~2IOL@E-cv@zD*s6?0Y?j|CvCugGPI1NysKGA_2MixxzB&b)^nR-*D_A{v1vs%? z&I;O)e>lyNM6vv&y2`C3peJ;#+G5F8?)w}8EZ$OAR;u5R^gpOte79$wI#zqikA)eH z2d{`wa9;dqQcP6Ri|ww%v1^D-Cl?yBMk~zvKGktGvMK; zQZu(A*(`1D=$Kn6rd|hi0V^Yu-)>mWgk}9{vN~n39fVMzr4P@}Krt>QD@xUo5+s}< zB9arzuEPbfyxc6lhp1gkSl?s7W9lJVbdI1p;|$@-;WP^7LcwV|@UaNqPg_>bi%LbJ z%`Z$?(@SVBpajvie7=fkz13f*$`Zv`{>hj17s>swV+dZBb=#}15xS^!Z1sFb|7cBW z7|PSHCCY{MuLkCERbuq?hoA46-}PH=@scrvZ?C28$=Mm8z#_XQmZ)o;5>Vqy#tdJ= z_H^!~i*xEJ+eYC7p{BKmz&Q~I)^%~vuyMeo=F;$$pp|MRW?{5xTf){+yZRu?(( zOJ4;qH4MU_zQ0ZPR=x0CMZ(95w_+bwJ&Ri&eDMzJLXrzayb*02E^zdlO=V$)CJTqC;3G{5~-Lr=Ph{uMF5|3BB!5VPR}5@rSV;u z`dSp^wx^Y1nI28H4i2UZ0|IiVUSJy=3zT>zNS^5-!D_}2!oz_@;U;_#;kpXl+d~Gx zw4yMIKJpx?rOL-%^*rAhuj=N+yFIzdFYTO*qAKJ1V%C_W`2fN!4 z?wB#3??`ZZl7N^FCkkFbLBm64#*Eqhy#VO~E({MtdCR{60S6NwF>Mum7!_InSU^p1 z?Zuxh*m>hN^MD|a)2TVeCf?)wgr>`i;9Y3{kaN68eTk1oH{q_4iMzD2)S;??xl#Z2 z(_+>HsQNxO+axgMF-lGEcyFVbj zrmR=Q{aFDGBjYaRF>JB&u8d-Q@g~t@=bJSrEWw>{z=^S>Zo{ChT{4omVP)s;wZ|{{ zl{Bt(kFZa_h9&d6s|(zJN;pV2y=Bz-tCM0-QE=o`Y9#eN$=I#yPE}w)(MM=?H#0ZD zxKEG(57r4I1rL*P>Jr(`P@Y?=H&O05ZRmS8dDf9vnz%)j(=>_&(DCMy>nw|Nu!w1u z))?1G`B_B;Uu1Q(T~V8>P|CO?DugLzaB=v&g(zk84pZBy+?qy$>Fz)GzK`5%;4Yz2 z^<@1ZII5T8mRlNe@c3+hKGeK{Wh%O5Z4(x!AeED^v|Dy#0ZR zz10nUNmgGBNF5(!bFF+LG82inzv_Me>qFeU*P=ynALEV5@3KC0#nyL4t3~A8OG+6P zZ9&x8jPuLV2ihul`hnzOl^uyfl-Nxh#`dn{s$eJpp=7*-{t zDA=mD^n>a4RgVOogRC6%D`+C;n|xPX>ku9z<&Mv$w3@clj0D`8zL5P?U6^Lblfc>l z3)2L4K~_cnY?Lyj^Vmj?%%EFN9T2JNI#|x%*~~?Sketb z57H+{I^I|GzJF|58u2q_Ve!1|&?)Q{3EEn~5KiPV;B&p^_vTf~EW}c*IE+GDIdF z_L4#c^YKKsSGcp4*HnpIwwZKmCaPfZ5EK_tU>y#|@*N!Wt8u%1Ye``m=U;{&zQV8B z^Ui#p6nU(($mr4X7(H-NoO3~-?L20$c;c%n(4tQCRFc@-asI>HAFUGH-v#UG7e)Cp z`>SSD8tOAsZL7t|n{a(}aiI86Hx^NmN~X#meZ$=iBP zbDp!Imz<6dF^}i{BqiXlpf|%M%lYT+yNnIVY z5o^}K@9DW~TA%6ICf&H3b$Z~hz~tEIuRNzPrze=3>$_>pVp=~c>fIZM{piFwO$Vd< z;u$Oq9?|=WR!Hg#jlTyDk+*{nv3W{ddUibIpp+=aDENP5%(qnae`2b8}#1_t1Rxz6V!DVKx!JkpnfYT;U)nU z{5Dp3^U%dHD3nCHKG-+8i&DDYrr^~DF6MFE$AN4PLGtE@F8e<6aNPk|Xz+y%4qyXL zcK0Vnsl>jZWK2cCB@6L$*eKCD6|*XU1C~c*49;|XBwdfb5KtJ#L$GY=9`vA!G(MdN zPvboNSHUWXl;_;&>jXz8OokYB(o~772yzR)FhNcP;1)D$5=9%y=IQ8`Dsxn~L1os} z)Kjjr77dXTAUCR!Jk)qIwh!wYH;HuP{y9F(vsRBeztsfQX08e}7+0jg^8ajpwTzei zLDO+52j%=-P{Z(OyANy#b;T>dJ43*myToJugIpzU56~es>+p$l!XT05&6_xY3%Rl$ z7mWV}Hqav=MAw?9URNJgB%ERP1n6C%x}tFm3z8J9WpaG@yC6Pes>;??eN7RgsXhR! zX@xq4eohoz=A8Mrw@|H1IU+8)O=tv5$mdidjr1UYm z2J%rg%<))d{O=h{+!PN_^+ z*eOvbua0hBD`DKt`^#nW8_szeK&Ff#)>Tdy%dY74%#2@9HCtFleJ zQ1wVPV9xr^2>b`UWM3(MOBDZ@jS&Us`qy-Y1Bili{1UCy?IW;ke`j;~EZEcfTFD+} z1RRb-fAs*kSnM)(d_Ijs1v(ga4ExVXRlR-8LkR`PE8j)>gabDg8t##Du>hZ7S}_9N z&R16X%16thDb0H2x*!4u}P-kYLoos?0hiN&ALNh7U%SwW4W30iPR~&1xszgsG zHkq(;+gvivWIP;!t$%3#6c=-|+L7}GmR9?K*1;~TRC>;|O3KG&YN|?zzX#`#7O6Qx zM)Hi>NofEE8UFIGy`}7VAhBC-p z6BB)|Kp+&)dxog7*G3jvAXl?EL#*;6GunuU+U0kK@q(ZIJa zhx@jt=^5!b@B5bfa%p6Ij<^I9PLrONlv=CX6i?kh|LF5ei8g(ZMdb zm5TW^Rn;KS_!K*5ZidTvgbwuPklMsUT5WO9d9Z}|qI``4yywd_wVUNybt>A++Y&^B zGVTMgO6~YCU?b@J=|oe`o7S&^xj2iI4GwS6=7jwAI;f(|@EJ((nq+-u!s*0QT9m!K zltVwbHRC-g1=?$Txsl@2G>*x;v#L*ih=Mk=u^mY=EZq0CLml07Xqk_9v>V%I9F9px zf-_*4LWoXx9L>dBaJHt=e>f=!n2d$HuHU|hbUhN(7cuv+7>mMw!OB^izA4+Qf#%T& zm8n^*26lC~xA}0-Jb30#Zd!1&rMj*~WJz(Lxo>Q;xMEzPxG!y{XuK1AW+SH4id9~X z_9&(2T<@ph?EgVRD?Uee?3t^}`-zoe!<{TR?C<%_(Mnm<9hjkjpxcM5>}H$r$vR^p zN~Xh~IvWaFw;6fbrtXc~u~er$L~c#f`T=3}nut!^d5R+bmk6x!fW0Z z2+ikjFcFND8y1w*cfl{UJ|U6k>H_`*h88Ka{2~BEXWtDVVA)fhul)XltfAoWFi0mr z$8 zQ6Hb7gk6DR#CE`}U!uGVcu`Nkhut!p5xrCC3S!CRuV^`o1#WOz_e1)CWI>ku5pJeK ziGdZGWm5(eT$*r&U2xL#!IaJXZ3eYtm>@iEfBz9uGh(?@d#w{3pN9={`Val z?(?E24!2NOZ9l$DRrw`-^#RXdX;RAWvlpWkF8!$%K!rdiyLBpI3QdE%(HAQuNtqsM0&k7)ytfgI`4~}8& zp*G$}A=qd=?1OgT)PQ3re36h~|B#mF)894@r}c&nlAND{uz1b;vN7zHOfGD=fXz^g z@=xm%;|0YnxLda$MDQCOa3olGJ_U87=vZsb*H*_i@`%u!;C6l}-Ric+?`T^|yZH?= zby@mFz`DSBxXCIzvxd4GrEkwvRR)fCKG&H1bmC4WDCE1Ck4H9*REMe8TnC*xA%;Sr z$|^L@P;4pb7Xrd9PxZPA0m;I2lF?xC_)zLiEfm~OA4}|Q>ZzEWm5v#K^tXWZ=g+&q zLN_dCd-LsNr0n-+ishHx6{5q36)5;v@V6Z-az_s2d;>P9^&KDXXCj^p-uMx!86n;) z4@)@yKF)!0!cqR-zs?I`_c8rgiB1xl#%U=W6vN5KJGdUWnZ`kD22?Nd(A0iC`3x*} zxBRaO>(3mA^59f&WDd;(n%~3|Z6(FJQ9SNYV10InBR%K%G(&U1Y*x_f_0>0%Tm4*_ zEBVHO^%p~Hf`n+4*~zc>%_$H*RNspCF+`hXDo%`FHq zc<}#!jIJI%V0*y&z|~UFKEM^^e8jMj%n-8U$Bbdb2?gG2+BA!zm^weBz{ErZtdi5 z4G5?>Y+E}FOtJTrI;xpCS-Cd5cKVZV>F`g|P0RlFKVA|bSB$;-9M}L3fhph|fCIAM z?A%VBmL)(P3kS2bpI;F_0TXM2#R9WXEP8$50qa-a9c`|N9a^!Fxu0yD{JiZz>>F^X z1>d~4+182tz2a%*+{s?z$?&4ycERaGjgO&g;qRE^T8>O^`YOKubYwGlx z5bo>Vt98DPQnV|=9gEGj0}`HHV83$XQ1+p5Jl8}0oLf&5ZCqOP@#s~_c2@52Kd z<%*l(gwh+ACvw2f+Jwyg8o!x1ahvbe~x2xSh~$gqnE#U3K0~eDTI*q+sLmVe2 zD@n0@dCV)a*Xj=|q@-^)Ru3VkZ8cwveyGmMOaAgr4>qUs*yx`DmvAS2T)%3%8Lf{& zt2BCS*flZfz&F6J{+6GCH6azapwv`q--*E7Yxn+-Uuyl#zELz(_;W2BJqCqnasD3_)NeZwI)5%kT+5Hcgar(-QU0= zQZU}eA2m<^d}^qf%&mC2P|g@-vydmlZ6v%G=+yR?fZHA32c14B;s$R*Zfq7+V zKDxVX^w|n&d+p(lh_*I9uQO-5*G-CSHHlgaKXc`l(} zdMT$BUwpk5z}@V%YdIxfPUE?qoM{ze44I1OU+0~x{5S8~z7wHzzYHQ-f009*Sia>a z=R00(`QBqP3llX*8gw+$F`pbqFc*JDw9|5fSet$Rl=`OaZJ!gVfo84Gs~984p49pC zbd9nVOIMc|r(oM$j$^J2x}~T}m%ySi!Fj73DnAaolA-m73Z10^$uiS@pN9Kxmx-YwQjF1 z+pD7*K+GnRg#zY_TT$01w4_xsn+bTR`QJjf8*=auKB%y2;KH)%`8Ua+Gw4gcAFyjE zfhGmtc6LcF%i2MS)+{oi%4`;Cb1Bn$)8k}bhv=Ar+EgCIT?=k@=*nUhRsy zTiM`tx0Xlg8P5!sY;}m&S?}QMmn-5$%wWSSaa__smF#jy#eCwnUlni<$d6Z5Nh}4K zN_ma7ekHe5N*#ye1xj1NPL;APsp^$*gP_?PxpS@N@dKic(6x9t;bE5gT?ff}??mSP zaPI7V66cwy9k@Oq{6_IDU;7@Mo9-14j|zWnklU(o#6h3OwtIIf_x!S04!=qfb!o>< zmxjqj+O0A3YTM^w63SGTwWj!CwGzIeBo z@jB)wRT>4qrL1G-?ER3UA}8sK>RK{^LS7Bp;TnF8sRj+qC%OJU5(^I|IZ+#kl z)6`j3r5k!KqhAYYJY;fzjX^TM^=n-WSAy(73jU=S*m4P*!D0VAAa7f70(&(k!$W?5hQs1 zb$v|wTBzNjf2e`4KxoU#)sQW!-iRP${l>Eo)=I<9f0RY>=O2NYZLR(LF~y=8-=I^8 z-#;*PbqKFMe+`M8vr*@D;=Z*SM1fk;(uWpz2g7SQVduei0d_~1uIJ7-miw0@MU{f| z%!6bjr!3rh^b~*6SzxI_uy{iGFvh=z;?jTrke!@Yej&d{e|1eAIjf|&>28ACKe|TD z)7{tkDMJ08Sbjxc*^W#Y3G_L;7&fr*5F~DvtI+$~olI-j;d?4RfA*9y@LY`6rB$H* z>fd=#(+?l-Z!H<50xDh2*Og3~fl+YzK>tHD2wvA~e6&_a1%Chd{pRQE3iyrfO(woc zvaCR^4u^U002J8eFKjccR> z{O&LzW~TM;%L0L0TNzPW6Un_l^_mBp|HLfKGZO*LsnB-lC_utQn z;_9SAN$9BtO0M7_2R-CocI3^YwsBqMbotl z--?KU2ofd8kfVS^3F1i3AR?edQF0U|gN!5z3PTbkNSGmrNDvT^C`gnzL}4W7BpC)A zroY1bxu5qv>zwCZ-#Y93n40dc>Z-1;>Z-l3eeJ645NtOb|9Y(R{`;G|eTM?Os(tV# z_;hqkV=&??b7RIC=&y&pQfHtS5ve_Ap{;i)-_PNq5bc8Hdg~#XD0FYu(%G*+tsCC> ztZC8G<2>}&jZ2#(r=i&yC~JE7$rfeA`!(AwN(<*0ADUgt$yFFVr7giXgFMOYht-AoL80~hC6%qddy6oYxlCg|A}Dqq_7vT%ocQWTMa;YiTo#`QS&iqxQp)goy4fg(z~3Kwc7k%h%+7c!6<4<-th8rzW3+{~q?ve>vhnwO+>02qCw|fT4Q`$m z--ywyzb~zqSzTawy~9lV$!182)|%kk-i7pjWa=G_z`Y9gA|s4_rCS;~58Ydeinb#o z%tRFe!FDD}P?em}#GOQq|EBdq_0p{>CoiEF{r#^R5ffJC?7|&*@!7ytdhl=Hq@Cgi z@NG}df$uG?Pn+Ox5p6BGPnwv+&|l$;@e4=tZX{5P->ud|2E6z$U&05MLjM;Wy zi-zZ@xm+7DjP{}_Vs1#Ux}}t&s&^yfRWS3W8UC?`v{b6UG18Uk%}Yto-Sy%x6nTj= z!NxJpOa|3w5Pq0kP7MPlu2%OZ65^XKye{2vl@B{gzXyI3YmFjC6TE}ATZuRWQ(L{G zG)IY`#HHWZfS_quLu1+$9sb+lsHo`{Wr?zuX%pO$yUd&xPt!q+b;Qs0&?S3LP}>O* zKMZ`<7i_##xine}^m1V{{T(E79u|*R(j}12MhO$#)d`Vy1P#AN59x+$=hJ^k!SRIK z!f!7UB43_*D%h)S>=&jQoj+00Wnv+n-wP9;N?2&|z}HZ-3Ep7lJ`I-Exn1nD zA%V+huFj2}hCS)30wanLCo%NG^aCkjMpA`=82T&p(gi}3>GKV?t-cGwSM2_RrZ;R{ z_;KOnG33}jO z?;;iEU<5zjPWb#hteEeJS9u3%xPw&M(F>4;u1FAOO1%HP#wP9>-2WDb{8t>^zvGDT z2Df*R;Pp2qAaD+Grwy>uLRmBq179)6QM=MVaYNT|yy@iNHj6ksTAi>Ri+`MmpWB>= z1>UEYAOxxAZ0i|YoaJTNybew0Has?=Tg)7iJA+0XP*8ASLH^WTNG zV~>>8lJ;*wa;b5WZk;0O5QWiZqru2{(@A*oRRTKv!}JN_Qdld;(1ivH3AlXBEC3 zWdO}^=cu2wE3Lv+3!vB`L`MO1?wLvfBt5VX3z>(3R|h(t9*8;}LiKoU9XsJG$Nx2e z6(8%}vR=Z>5Zf`gh;-vO?b4+4~P@j1Bhkx7Aa1{9!nYW0e zRnEIak6v1w<7_28A$fhzWEmcC!@;B5d@ghCJY8P(k8mC^FYq(`X|g9flwduKGUK%g zMGEHq_ctFfRGC$~MtkZ$HN3s54T#I%i_fB+G3?xyj zv*|NTl?FCR$A3fQzc5~^up!JeCe&NkYE8ouUA@1^?^0@7u2jb#kz|^^8*T8swk;n4 zyztGh&0eSBSI+3(cnltWhc%i?EMbLg;L0dhPk6F%uzYN|I*_rJ`ZZ$d?up`pFyQSF&ByTwHWhPEjdpR_rx|=3Bjs2qlPL zmW#=Q0;2nTvZ0F$NP}a z-ewm$Z+7dzTc%{n7cOVjhU@%%ghW`v znOl_Rtq2Z#`YI7OyWy{x0)r!@BPs5LQS4*^$`z8Wlby*d1isB=z!04TB5!sn;NEC*l*FPX3mzsg>6dng_AS_|&K6 zLDwO?#CXY6iZ zyKNfy>ofL**DxNIo)RC@ceF8!R9}CEFy~k4k++f6JX^{7(zidizfS4;?y;f_DXJck z23_UFs6ujTgBlghn0x5WiGqtNn=0bv1y3aGLUh8=W2_|%Gxn3MYbw&I=zt?ThZtKb z=p#8OQe{ikYw6cxIFAV-mOS!#>Dib(nP^wslnuIo z@1`xAG3)+RY!}sb3Gx9~n=sa>1?D;+^6{9cH@%Vzpi|ba2v5vDB%e92rJ@%Xl@We# z@AVA{H3O<5&Ug0wPAVmCPX2>>K7&yQ0*+JiX?44(W&RX8`3Be;?@vqHDMjQxa#2hm zRDAa}|6uO&gIv^-y78X6fZzFo$`a-|Y%?b=Ko3HWT6M$6r({8r~?b$B8r;OczAe1v}r_W6}-h0UZRnua~<$MG|GP zfn+DEI=)LKR5G8LM~(`U1*^mgO@+m-1l#L>Dq%BO_TD&$QKWf^GOiPru;TMev%O7o z>&r$$jW~(dLa^PTTdK@^Q)-5a+nC?mM8ef)qb3vceH$y3F#ZiU_T>2dV7N!b3K*wG zV!;bh7WH5HL(v4;Qk!293TFMw#t4f&tXgzM+%&T?`LHVoYA6hlCKgsO|mlxR~iEU9|*3O|?Nw*;INp zqI}d1Ni`Lu7{8yP=>u+PUIByFjBLx9t6$y=mox%TcU|2S45-^5`OK?sd1}cT+@E z=oolw0vPX}kDxJUL-o!oGmLbd-%w_NJk{ju+VMn1KL<-}KSJ1vBmJTbsheYnCw=em z_C*mw`Z{CcRY_;?=-ZeOG-HXt)lpJ4nxADx(y7x)v@`a>ac4u(q^ePuV|Ze;FEYgN zeC$cHt7M?CBIqygWE(b}Q`tOd4GA!HapwnRR>vOkI-7M22H0??ATl7@3?H?{%)UyU z+06YHM|SOM3s#pDF@;RMh$m~wJ$B5$uTK69UH4xal^i5;N3GF46Vd94WTgy$Z*2yv@*kKXDj@z)EI)gu__vHhtDsUY0JDc6DwbC z??GNhX_$f-(bJ98SKrb*xnGPI-_<)2+G-gUhq!p)p<*5Qgw*(M!H!P2&FcrUslkZv zrcu`kR1chC^zIVx^jo!iK|yl!k;PQLs4f9L519cyk5+&x9B4hJ{VRz@LI?a-JmTg%JQ|Q>{%AQ5(r6$Y zP)9s=kigdaM~3ffOUIo;5BUj*i~pb?@lO{sJ=6v{=;}QP?49r^1I*ghOs78=I-J3> z<+ny-^1wJ5cNf#OZXYurd>6AD8JGJ zV-5-+mpy8~`AM<54c;|*^XcumOp#{F^H)2I&?_SQuvbF}z63(~iluZa;V$$^{Uls$ z4Ia1s#Y;jdhmg<TpcM#ls6Qcqi z`2Di%zk6Wat}?>~6A--+DICvR(*e3TlDtOO;{iAizlovSyooVTj5_{1jATJMn|GUIlacEsXBUpSzv7;T5204n2>B17*2R8uR&^n9nh%8ARR0yus3J zaDm4FWlmcBFR3o+c_Ea_y%)`qRFSw?E;=*iNCdz=WN1d4p3@ z2^p+UT?BQO@eW2x{bnWpK5u_)yR2WTiSo>$aXFADbFeXE_ z@_1NK3H2WsD^%RZx_wiqh-E&$dImFlfe>;}p<=l){aOTl@_jujAu?BLo@_G;umm zGitgh3-VYn^CY2BclO%Ox=@m$h+lrBQzpNNp_ON80w1UV-KiWTLnj~mUJFwWB&jnh z$VhtoG-*{Tq27y0M3_fzl{%M|$*Z2Gu$h=yrL&pPe&me+2{=Gd$nVEyLbNc1zMbMH zv3(G)T&RiKk5f*2jy!9#ifO@WQ5OkBj;K+ewpwC=8jXmEM#Tp%6PU(nlUcKkZ?=m? zI^rMiq*(+l`jeno%@a5+)y=OZwpq?|T=CO@)|9Dfu z+Ea{wIl!XB(lpAV!v?jpjZ@qvw2_CQNk^!a{qOQ8ExOSt|CSRVLrsLaXi#lLy1vfQ zMd{#3P(72F?vKm+k$pk-<=ztf2FOLM*&W$b)-pBm(HI_3)z>poKr)GyPlehrIbbWl zMHbq678afFLuiLy1x!~HCa3$~7p?DId(Y5!!+vT zgq2I*ii@UUSG;bVgAzkz_;C3&9yglcBy96w%ZiZP7Q6u$*P<@>7E**At?01p>11$k zTi{~E>K(AuX7;GTmL5Wa^aA0_1ts8%B=4*jjBpJ0OJ}6JuQe!8RdB`dk-EqmMYS#R zzLBL-J)J?Pm_?DC6Rq*1g(oOM3or~Owq6~{Ry(6M{1)*D?RIv(E$1b_w+Ku?^Tzif zJ)@Q8riG>!JbK{W?=T_jrNTo>W`?rm@WZST-&@d%9*myo#aGP5_vCKWF}CG0xDPMU zUqtEGY`pdf;WrQwyn@a{MS!Y&jtT5HFHT#Zm)DP(^cKWo(04jH#05BPRKt3T_xW(S z+rzI|!%Tkovu3MVsc>1=!^H%2-4+Uw2fnxfye_BbNBa`u+&#qY)#R$_I>l1v5US>>AYXn66~ z_*rLp9lLX0Ispu@!Dq_Iv$yirqE(S+4;CYB93^ZB3R++7-LScpmiZ~~yGGu6#Sm#7 zHKOk}#KZ`8rj6yD81C`CNGkRmlU`qS@4u@&yPSBB#;BKalkOV!2`g)nMlN$5te+HV zH0+99^sJe91q;28OdN|LZ)`3~w;aH1ex(*4(@Hkc3O5$CPGPcm730>F+1iFL_q_X{ z!>b#<%N+0*|J$Q{>;Z9Xa1-2y^T|Vp!)8@-&btqfTnSmuRYUh#?mxZe+<#kR0c)6X z)P8FO&4ybV($ZunG*qc(1cn8K$Qkzl$-sI(Q_IDs_Ny4@82OHymF1Wd(tAz-Z^=)D zESUQ+^i&OLt6%nV^fg?YsBe8k)8oVk6?rqH=0{|u4*x3yx?&phm^~VEFd;C`M+$=p zWyk&9^6ZTpx*m-6-)|gNh(oxy4vx!enGGl$apCU~M556pW{OQ-6|lUD|z$L!CK(6aOprD1r9)ygMX2rTUOrI<=m1mib~#o@I!b8C8t!z*%r>0@hCIeuo@VL*DO+_d$9Y8Y8!6_G4bS<39?4YLP9O2m$wwSzQP_8fDr)jW;17QZLenIAo3eL;jPQV=h1e|!k+e(3p|{mn0Q>S# ztqE@69tJjxQ22ToFMfuQJDqu9GIGO2mAF(*4r548h89Ezr}Nfmf9ZQ)Y9o!Y>F_{l z{cEcIuSxfRK6@v6T)icl;1u+Cy06NX zI^Ma<(b(^Lvj2NWq5ta78GC9Lof0{3VS2D=heqD7r74W^YJ(-F;=MJ*5Q=70mdF`p zU4#WFgNU5$!;+pVm8qL^#A@R7@~P9X&<{1=FkHXaSLS)*Z(QT$(aHN3Hy#@YP9)_spF`l2tOPWW&tQ zOX_l6xH;xp;`0hzeiQNr>W(+rJ)?gv+TR^^bxOk;?azvRVkLI*KE^wud?}W^8?cm@ zFaC0uaM&dYV~~Axq{cE~@gpfIwHdanOWcd=0!CR`iR&9#}R+Ibc$>N!FY z!=r%NoxL&7vZ~h&SD=l29i?Emh5R{f&w+lVou2xE_LE~wUW!Q(13rW=99iRba))9? z`0gl^mHvZJ^wBj5$r<~GGjY1osY^3F7u3q7LoH|QU0aNS&=2>Gfp!F^?GHA;A~L;a z?7d7{iA_n>o>86NGpVdFV^9B;hl8c(;5rR>x92$xq*Kca-4&Qwda^HU<5*&Bp%Akt zAfM40vM@9w$vFDlPG?Y#y$DPOA0+`X|*=-t@9t`r!^sRif!`8(u zUJEyXQ>$*l}>Auah3^(Mcd7`ikCr|z6 z5=1EXC%UU^5lkp&`ttl0r}FBpD;io8?R1WebZj8$;D;CXS&)QjgqP(qocCb%GR?7G zm-5ZWjSA417+LU8bCO0RKk|%*B6J1{wyyycn6LXm#l7o{Fnck&hPc0Iw^nSu~h zciB_rRON37Mm*e|0B_8-fDcE)R-v~=x!2rB3{9W^Od!J#BFT%d0yMr(R1#pNuRFRx zYUN4LOM;sKRQkZv?B?T+Wu`=?F1i#%asEh1-6yv%{fLK47TzoH?3^`t4FUbH02=+5Hgc>N zG4uy{RX^dqWqQ(%UDDHmf@lg>;C$8hq*F_!kN{0^63~!N^{Ed*c$KJ~8y4>Y{X6(W zpZkpcQw3-=6dfrzly|8qVo0p!p^GD<`SgN#Gm!eO;s+m*Q@=Ml2uDYr`{0i40-CGW zFA)&M%>(kJOE6=k_txze~Xk^@zO3nVp? zD~R`|G?R;nsUi)CXd-nFjsXXp-~i_BG=RVsY5e*%!sYGl6#u(clJ4oVsys#4UnBQA zP9{=Qb`SSmb?QwGHb;hPCI7d>|6yc>ro1I23G9oR)tK+-+Lz>`qWT64-}RGjSorms zon3r%XngCM$rt(9yqTQ4pTnn`ey^jS>e$|cMfQBr%>N8l44COjIcB*pUuM5a|Mpzq z-~79vxJOaXd+_20(SGMT4IW%1tW2@Gmk6s3@?F(Itk|36pf#Xn`(pqR^Lex!3D>P7 zhIt%!LMn1-z#mp?<=GwWp?R3sQpVf&kuZ8*9>?H_vn|Rxfic(h(~n4|dNDWj((*=` zXcRmScLJr+yyx>F+h3qIhi5-+%E>8sFSnY*5+oU-}lDfTsaA z&eeI?=EtM(fg^zF_Y>`gEF5f3>32M$d_LLWiF%@Jj;8^pzW&euGKro7|MM`EpXW;f z1m9Ki`$cIl2QfO$B*VD;t;RXnF^NivpA9n+2jh*#H5<7pmV|7(;OoXRGX=SWh?b+EsTGese!>BEuUX2zu5ai)Ig{Fv(fDmreq#mIOGxbJqY3V z!W8(0D)6iWb2RR=bM$5WH%cvQiYEA$v`W>Yg-?1rP5aNo&t9aTGbg2pySSj3ytjP1 zy^*t5k6iQgXpQHUbyReR_F`!yqf>aI(6dv#j$VsVPf^tzyp94-QF-C%?;W}&^wT#L z^CDC6VtSc;#o$dxrN3Wc=XGQf*P&HwYSbHQsWsE3H5SwU(+m&z{kZ7lzf0&pGgMXS zW22Kd)E(;9JjJDLGWZ&V(I~rEpv>_)+U73D-P_|ERPO0{in2|LViif)i%bQ#0xfT> zUHWQI$EKmcOV=I)_y;}1{u;?_$bKTVAJGtqWyj_7hTbePJ)@X0ojE01aWJ6>23Ivw z!h?%3)c8Tt?^)BCMN~TyYT~i5z#dG?Lwg+z5x-*;>v<)H!5X@+z+CASf(Cm!2a5rn z<*JDQF@&dXAtr{>!)LNQoz6czAt%Uu_f|iu%k4+}bvGE{`d1EMB(-pp5oE%nf)T|e zRZ(2w95_4JF0h{SRvlB?Je4rs-3O!`E@<8e%GQoq^b_VgQj7azLlTooHLXUScC5Ze z3;>GIov`LplqJTtcpnBZgQp4wkdLOC_~9|7WX6Nt&oY%-h&%0`-wvI*nj?!pqi1-g zm5nv|1593ji!6={5lszH1}pL)j_9N@p8FTJ$a!me*F)E2WEX!eD6N$;i$XpppIq+% zFX6GG7~4XXwr%b;NnnEM*2YXqUf2I=);(pRl)gM`^XSAVeYtr%>f3l^ai~Pv^AKrF zn3~{$sIJEhW=A-T#(d()>t!2vUod(_WsNuK3MzIwQ&#EdTRA^Vk0e#c$Tv*g(+k0g zPb6+AZCs97JeNmQ_(Wo@83rU!>ZE~T^jg#{zI*U%slC5a>9t3MkicYjyR(8UO`iK` z`*9CO;_Gc^VEiR|(G*$y^@2~jG=|QuB?f@0CEac9{trh#F9;1rEM6gDxuNSa=wG7w z2~^iA({bm5kEgEbB8(t67H8Ci4;k1HuCgO2tO#B;7K2_y1Kn_2eqCM~^Vw7Q)|^`Y z&%%e>-Z3JX%bUv$F2O15XWoI`2J`d1YQRLu?R#qTWwmQ0nua;cgpby>;zN*wLg>NM zIhQBN<=jcQo5v55f?XCR5i7r=`GFWf%zPX9UG*VKT0kx5G|X>%gp%&(6PDy|rTU8y z*?#6jt0L5N=0g=Qt_Sk-?!lZTWs^zpCRT17EVbYn~QG_6OHc|$Nxo~T`7 zoxt-nt=sO7X$=3<&!4fasJbOZ7+WRT`Tb*udhb}KQzv2F}qQ`3Z zA$hXXnUp8k`7#l`;wWAx)hEdMZg^eJTB8Bq^*ldI3u#iJq_*qORx=}mG?@ck-4kln z;hJ7Nkp$WE-xcW=4WqBxh&`i$)Wa9^jTU_E0=!YbJ0fQ@z`($o)K4`Fb1e8l??4x; zkry5~bTUke-=X2Hp%uVpMYMKJmUl|fJQv2<@wl-X5JN-3Je)XQRu1HJW)zk0IVd1_ zLVc@GHJynHFOC)%U~h1~VfAvO!{iY`w1jhlWsKGU=;+_m-=UPJo56nyb~W5+@Q-^{s)EDkVUKwTGvT^_XdlmXC&*X7&-?vAjFp2p~b`rv`rdl`Fw_A=9^)pet< z`N@T*3L2Y6Y(=IsrKtv0iV^EZrLwhjTN?|jMt25k>D1SZs!@ymMN}NMbRTZ2AK39P$ z3nZzn*YObi*!{`0NzE@UV*s1uT-Wni##0d2;YzGG24sn$qNBO_uR8x+va32vR#QW# z?j;yoZpzNjFhLf#f`8oA0+)WseiWXLfBZe$l>P1wQiui^h~E7n7;4LS2r6q5KH1E| z0_zk0l*$htL%kC&rTNS^O(bR^m#<#u`@D-$Pk|}Bpj$#tl}1Wk&(~{pIzJ1XYv`)8 z_N4d|su-=YIS&4*1oU>_F);DaKJjvjudR{PdEn@OYF&>P|*Q9l}eM&EY>hl(zINY+tn}w87PzE?spm zk)fzBrJS@iF592$OCJ6PoF*uwE2v#EgAuCD#Y12$-~~rn8UXlWagA+*cQE}pLtv!1 zZ6-f^e0IiVAWGE1olHK^2a7ZbLeN5`x2i>g5g8Pr9g@0DL-EJRn4(h(pq9MGc7hN@N1YSn|Y~jb@d=Y=*uOXab zy@ijg(by_j#yWkU)Ji7GR`r)ov^|-Ai&0KJ~|-avl~I{KzCme zt879Xs$fJhy!AJ3FC-Wmik)JS(qTLyIs+ z(OItrr8uKAFQf~sygx2AKEsH8FH=Q?KTNaG zIyzv{Fu1CdvP28|l)DCx8bTy`{NbCmHRc*t7w-{S|I%(PA%K z!dQgR(ht>Tcm%sTtzB*D!dI*{VJ1_5(I$PPx!qyj!o8AwspD+xm@uEBsQ2&A+poW= z%6{=C@T6}rxg1<<4ytfTxEP%r7X8|og52UILHzYpNn9fs`?~1?*@#LXjD0>@F&{U#>}t|Q`FpbEvXF8Y*JQi7;{AAF>-cdF6gd+JU8`MDiA!-VibtTIV%7j`ds zGEc=XEp7l?q2Xo_RFyI?Wbv6s6dEGuHO18^sMX~gZI?$@L}{{%MNSDM;I4{A!f(k( zPO&B64DG1lmi5}Rymr)$Fa{6qZBrZSMkSu@_I%!LW^3xkRqfgEPivZmTNmF~1h4Bt zuON?9D9mo17-yL&sTdsiLVZ>(xM8bNar!PPnL-cL zY`jXT2(6rh6&0$_k|bP!1evrs@FGNJ0y7B1ph`}B2{CSO{GY^Xj!rmBr{iyGrfl)J z>00jK$UEnjzUBJrSHGJjsO=tF_y>_4?gQ=#4a!YU(0_@pahQ2&i{6 ziYqXn&B?CzNyrBfz93ytp@vV5b|`T~W?Ir01u|&h?>r??Yv6J7e&$@;k7usYYv5bI zJj|C=pKYf&b|_OXoDQCHbf8mj>yO9v7!yW{R`#Uy#yu?!4kiC%u%Dq*Z)dPMHsmk$ z|L1T5J5;ape-7|aa@CiIuTCOK(?-Lp99}4>He)|VKG1K*nnjPu&A8mUz4mU=#b|nA z{PLu2$>Nn3uDYJ1uc6^841=g-_#LH3QTvAKbD4ohhf*iq4*H!0LDETEDhy!XyhQ~2 zgfDZ}>D09yI8E|Ey^2`Du$N`aznii1Lium|4|>m3hE^WRJuLK?m{|tfnngW-@=bYA zrLWmMp-Pktzq=)sD*6?UgzYFTt&d&Y?KZ)37HnWXmNpcLT9t<|ZFv6H32O=MajZWQ zZn>#>@S&laZi*NTXocd?miZG$f6b(E)?6ol>#6H|BAfeYoa37x?W;h4Qy`1X2)lFB zgd0;{TE^=~LCrg2;TcHjM-j04v0-a%iO<}wC--FeLFxU`s0`wD)zOL!;>oLeCwoCM z6|tYC_-mJ#P~^F@V zINyGL)!y?3;(Og$ErTs2{94a!(8epB=M5}zO!RgY6~Df4ehcL{JaAr2MJ$uLJyYE- zkR~g(Oh5VFeQpb*56C5`?E=hzB5fRR>G?s93LFQDd z$u%%mhVpTAVin^IDT^fZo1L^?lCVDL((*HVb6KHX>QXqTEC=qKA3ZKS?1D>Q2;v80 z_Qs6Ux?T2$)r_@z}YBO{s;PyZU3qr8OBWBCj#+2c~ZSX}F+&|F~NlegN0ELm@@?Wfv zN!SjSr~Gp9nuX`BkKN0uDo}SDRPsw}bM})OtnYau%*U(EKY#@hE9~_;kEP_(R=@YR2l@-tn9OBFyZf&esZU<3LVu+iuIi$_ePHJA(i$RW^Vo`mYe zA5;D@hErbCNhUl_IM&mpk<$2wklbVjxXCddC%nH&(^C9TA3*oPuHFHSl=z`@o`0Ho z*uft*?l2=+un(MnN*Kn<3%+OQz{bi~`2D4nGS^cI&;yi(CDN5 zRM-R`*y}pPz~tL;5WS#MkMQfOC$D1bnLf(> z5y29wPD#{%6CqddbFq_oz&CKyFsq*~PQW+$!&ullM>)r(`0|dAZzBTvd$7uliF;D= zIaLmclX;DMQk^w3BM;`%YaSO9{{X%TJ=$w3eT(|(2h{RuwYas36_;kn!#C|x5B8$x zA2Ie~+$4_-ldkXrooR;i!W*O@->wB?C!ZkC>*TKt$#1VuzC44;@=Xf}xwqnSoa28g z`Tn^@wq#@T^Vryo??S1AW4J=Qp{6eIR`8b1$_B0rgSu`so@SG_5qnZa16U7rH-l^; z^rmrz2kF9h8ZElF<2Lg`Zx$s1G4xHXGjKO5(%6;KA6syF3Hsd!-?Ur0Zi7I`wtoJ5 z(D>fe?Kck^%{T=H+42Pb2&50&HI@Kl<#8+?tqIuUlX=Ty_x!V8J@d(WzVY>OgEbxg z*;TbJz}^pUU2!?<|ED(_fMnr&QVwPheRPPoR0G;c3TW;4s1!W@uS%|(N(VQe@OX2a^~%qN5{Fqq#~_e(p!;__hty9ZvihFgM|Z<_g+ ztVNgRPYIKG3VvxH_N1NzK^|xnAAxY+6{^!ZK_gY(^Rc5Vw8!~~w~t1u6I{^9EG>{r ztwH~X0J3B0IZw{FcdV8 z;7pQ0)dJ*P7Ux?{@OKYy1JLCKL0}Q~kKG1SG*-3G(DN5t>+xs9@qmeOTEjXrB9y?@ zo}YQ8q-y1SEIkZ~yev8aa)(Fz27`Nm-b(HO#d|H6u?W-3TPt3ojbhEIH%LXbQ!?NE zahGUy>sw`Y$eq3Zm+sJbfazY3xCN~kkK}=dikt6sQRuzLQlJ(2*wO0Ax3;irz!BGj z!{;9Za+axhmT;o9hm%xVNQt9|lk4l&5%F5TO^$AOUu9OQe$|I~?QE4QMFWFn`_0-n z8+Y~u9(+{9yi!OA1{@Cd>skhT`2HeJ+WJ2RWK%2RbAr$@k?mM*;k`2Z3O&-_WA9*T z4)sqYmJ}ij4eF6QzBEJ+ik#BDZ zOxU}IwoE(%$O+e^;+l=h(b#uX)B34wrTjCFL!KXqiHK?S(tv{6=uR>?h`Jn zRWVfN=H_j+*CimxLJ+$3IkeQI4*8bsQ`5bMyWin1f?DoH?=-t=+&1vmRg0{fZHE1j z%+_1kV$Gr;{KGC<6Ur1tr$qh=f>_wdjx|{kLnHgr9mFF7DXJqCWl94rA8#wu_@4Lt zMB|cnS4z5MBK7g?(7oIULk}mW9(5}s=(h0h_rFIdhhu-W(ZO=}?L$>;3QpYncm^}V z0AW-FN3raEXK(TrQv3>F@J68v7@{m{qj8#HDC4DsC7ehj3NYI<_x_m7UX8vJ%^DQo z8g@Ok3fCws%e|E;UCXGCh&xdOy85xVf1Byx9 zifcaL;gX4K_M%Ycy8A=m_8EG9=kQO#pp?LmS;Oj9wXf+U^m9s!q)t9*Ygy48Bk%30 znso_;jr?)XNyt#t0@OWV4#O*v75(v3g%!VDn-g0-Gi1!G){6uH#5{+J?YpvDZ07fb(H+ z+#xXlp-KUO$dOMB-4^-Djw3<-NH>|zWH&3FkpMSMLc?@K!~od}(AuID-u^TVdvWLe zG=Rap&p}@;t2uBaeLq$<`Vo)50N){Wync&_3614#d85yS_D5tM%Yo&E&m@XtN2C*eA2 znPfj#`bvexmghoS<~(irhU)ww$b6}CCr(+5rfbB}%3u1eV0bjYwZFt3YJH?dVmE*X zbJrSB%U|!$y|&JO2Qzr3_f0lVyRgHOT1e&m%MwXeH8uOc9dNS1LvbY_xlA;#cp|dW zzxa-;k0g;x0X&Oce(6SJQmnF;MAK+tzBKmp#BkAA=*}DEw1vY`OBs{OoV2yA?8&c6kx3A0oU}DcalTNLkm5YwZ7nWZ zRV_A#x8X@su`N@vP3U~tFPqY-+N&)k%qRZ8T>tTHPh0A#Z5i02y>n4SeC(Q7g3wN< zcI9MW#!aKqz`4Bx%HWmL_8WwKh10X&h0;#zVm$a zF>i&u1$07LboJoJjSF|jgK~SOU6@vSC%V=mLaP{G(arBkoq74@D_p6U^JmhLoqeX| zWmj^-SmU{9T5c5{p_c~HR4Em>ff(8S5aUlK?z<7a8>uUYdXqe0T)~*@VCMvDpI;tl zO)nq;D~gQdIn&Q_PtU`y+=bEmS9Tn`_a!?459w-wPHhuEjs$uGsTq;b=-|rm5@0sU z-uS#&f8C>^|KO|oTv#Ouf06BQjH8|inTxIYc*=Q}2aM_?6*g+w8$n4iZkL~vn;rX$ z+B}m_2CA(~H?O}ii-DBC`^xe?Q5za(aSiP;<&_#qlJ|vl8>KQtvmrx;Y1eINQYw7P zy!t>z8VVap8B!_&&}dPc-6ox(%LY;%<5~a7KL{NI`l>y&ige_Z@(Uu-XhZn#Wl`3iQ2IrIVrZW%Oy01e4<}IQP zr>lFY6)QbqvaDC`Rt9{Lm_+@f7*P+ zbB}f{^@qxB)=Iun6%8FCB2Le;H_qG;h-@I9Zfnde(@4nlE#lWt6Nx4ta_^&9%fAEs zhYth9FV-tWJ|;I}1i0FQ)ps7N@3!Glr0+hAr65$##htoACxr8-?PY3@-TNtTW8i+6 zsIvf1Dbx6s@%IlB(HrCP2EVF*{r)*V{XhIWE~n8q?Ty^WR`(M5RKhnpOckTLm=*5jcigFF&U@9Sdw?J6l$SqsjRVPVpmo8y-YN_k30S zbSC{*)+Tt6Pa!#h(Xi_huodk}mARM-FDokbi^~;jJJk5EP+DL~gAwsmJ0iFx@qi3c zLdsbKu&BV(oI_25=!sAQ3?HIo;vmnYqrWoo=#5K1!sJ+tH$t1>5}U5V1Ho={#&uef zq?LEfXI^r@btVOWGQQ1yK%0W^Izl>JPQlmDaQFJGbujJEQ|3{*S9W3TGNrMS=Tl!x zWA|y_DyMHap6j@5S9=K%!0l>R*}LzjZ=@Q8AdI+ry(kc`5F-!VE7#63R78}UW3!ne z%gtZixs$%JN^yD*o^1zQ*08D00y7fTlA!Cf^XOaAL{;0JQM^b)$)z4BIDs3N54+WM z4m$VOTViOv#blt@>kJi12%@2RZ34Uv!M{GCr!$Sr`gw46bYzallv3e#!Ht^{zBSWpK}^isb6eR3M3lkzSh{G?qT-B)J+meOjjcZ@t367dYxf0*&MWQvb6IczAQ=2RaoiBcDc9a!cH|FV# zX?G-}HAZrVtT}g)ym4Y9@6Xjy*9WTuL@~53Kst5a{=`PSq!cr0#yCM{UfK3B`=@To z_el)bVy3UfCsu3ky`$(53N9kG zXVau_b5W)<5)JjC=m!_3EWokp&BqB{4L9q2=Zi9yjqYQ^)cO3vN@YV)RkYvV+}TqF z-GfYJQ+DdoS?K4@o+`BHZXZ~?C;MVB;?lKn1V3%Pcj?>(l@e|*mh5F>r4E{?$t!d=@gISR|C@*-!~kY}(|^uN za(+|q!qbJm9qOlxeUC^x@efoM`vk-gNf{lPfR}GCC@!n|CBWV=h}mW~-u$S%;6vwJ z?@6<+n)r*IW%~oyY+#i0e_EN>iEU4Tle4$KfirXL+7+!FmG3DNoL`Qm8j6}4JbzTV zc)meQ2vI3(nbMuPAA3};ngZ9-oA0xITjcQlrfEXs7%cV;&>WRB+B*5kbyq&TtD66wb{dx3?@!z% zl|CBRSby;h#iSbHTJo@%H+d>|CT2HZt^&L zOzQt5?!CjBdcM6;J58h|L_k1#2a({Y|?wQ#$Ykk&?#4CZI!Iu$jnj80{js&(F zLgY-k%RCqiuASCe1X|@seaprurEb5zO5#2HU&rbgpz@)G5*HfcXeyrTD~6p1PX1*$66X`HL3NOA0DVY4x#D!D0T_MJYn>6O>_>Xa zi*xsU2sE{2{Hd!p3it4_|K|vaUjjnhyb^YX#ht?ziiRi7o+M##%`u>nfSEf?D*9sD zc~Cy=)KnZTi?#xc*p7bci=(HWKcR|af1;fTeY!^VGNW?69{y5_?q=SvOYCaBVmMs1 zV&x5RU{C+UP*V@W%Obkx`v9;g`sYgk6#cB47#GN}^f4U%h zp+;&*9m(aIJHM&b?w~b*h+WK%*ad(UFXO|*fqAzFwV0iD^;=R?v(R-gZ0*7ar|B(S zm415pkHUBY4Y5cdUOu)vmn zaasPDd*`_vz;vPQV0iNVEwM(!>G}P^8H-8PP3J0NT<$_lD%M=VCiu<(|{oAY_T=!9c#>9A?Z?^MCsg zYe5zDBU|3Ft!Af(Zg5-u0yi&Z6Zdf#Nfztwfl4PJ2EuzG_d|3vhnKC}O=D3F4HMSF zMx|uIbu%BZd#OripptS!X>-Bfp$VNA29X{oQ5lB4RQ#*6bIs3QG@&Abce+1=&b`IGiD! zhFIRtTK>I3L9urKyRTk?*?JC9hn8x71DNicY!>}4O>%^+fd(R68WmrxSf49rFC!J}(}GcTchk3u67apljdz-|ka!3(LtJWrFtd{V?O zimH`@ioUSeRPZ~`3RL&~{f-n+p2SM*kws~1-;0H=`AWXAD)8(~RY|bi)uFYx1N0#; zr?b99qfo;M294hzSZ;5T(iW87f*AZ_tk>s1DGH*Z+b0Y3h{J#err+{GiEwgKa%vsW z3zM{&22aGOLpENGjQfM19;PB7-O$btSc&z6 z2{n5MWAtUe7>Z-Eel$(|jD>F({;Qj4lGw=3Ay zDQDFssw_P=vJ4?IqPcDnLUeCBl^n1Gx?{4)NU_(|qNS+I_xV}{(df7K9GsJN5;ucP zF1B7siMnUO!^{L@^>NJH#JbbGlJlJME*ddO`d~$WkB~b{K3M}ozed%2VVn1@4TfO< zIV3BnM5PEa$p0))0X;LvfV70RUm*jf6B$W)EHpKT@5Yq#6*`5x?;yC)rs9)v9I*it zv=k1GxHOs!%Mmx8*<)KlC6}MX%z{cn2U6`P2VDAhB$U)&XeWl(uA`Zio+CiLT>1yg zpc1QdF{}-hfw1EQ{+~gSiu_l#p*f>``=kBVP*j;fn4R^F-u7F4W2s+W6@P2LMK+!K%=d`x3%Tzh zTf&yc!|7Dnv@^>4)JLROUfu*mj(c05@f*x)7SeIHejJj@P3hz>d8j{g%U*otaNWm z7xVN0Y~W;|N1%K~1(kd#om7DC5NQIBfv321@sw~b-mBx4 zG63>3)$vBSLizPFuruM|SkR>Wtu181CD}(dDCHTMH0LB}05IPuPEtCuR|bGnGQ+89 zKcGLjuqpGYYCqaFecp$pCDsUx9+8ElY2$J^PS^SydRKWZ`&Bn!GtalagNmARScopy zUd^6A{@#8PI&PfUzz%@oehC|_L!Am0tXE$<6DBnV19dB^n)&BgAeD9K!vvj3Tm)hU zq!oH^1*DbkgDA9~Fc4PL9xiYzbi%SmxgrV&B$R|6ZgQbni=6aD4TMhxK?cGiko{mK zf58_Jga7%lzYcRq1?+MmLM^_Dt)EcM>9_F$7X3d5!toG(_U1VZ>v0Qe*#iOyqhOR4 z{o7VdZ6ulHA#0ri4(8I2{y7gecc{4w11w|X-D06Z%4t?1%j$VLrM~h4g?xB^_6i2S zz}K;-JtX|AvQb0G(duBX>j>`!_{oZC}eNcb5)ab6aGSZOzAGq<>??QXS3VF zMwqehJeRc4l43I}xB}4I}dgDi{28C!|`h4xu zA7W?Jm#scRo!z+v#%mdGeSkL z(?)_Re*OZ;>9jmFGBh|bFKl6XGm2-ySrWaFN0@IrP1$*a! z#+?1H#@vRj#=&t>$TSVy@c;N%?Z?yqD~3PN{GmYeZ^M#WEH$6EAOSO|f0E;D_%8YD zw5{;lxg15h=I_|D)IS8-z6p&wAFh+hYvYBp1aQj{YlAx{CuUw7By0YIB_g+OEJTH5 zi=wH=-hJM~tDYBjm4=xt)t7Hwp4T)obh*+1r~kTgxGH5993$_jn6hCU(m3UlV7+S9 z?4FEj)ma?hmDM2zOg=G?zJ-szaHS9G+n=|c>P0dPOOJM?bnTyTPukrv;5C^rl#Zl1 zlz2!v-t?C~7t4q3V-jT6BOSPgQOu+5R6b)onP>_gD zFr`vnn%o!+rrarE3iP6L`u8Qed!kSCSjdFQsInrWX^8e-eZ_QQ zA|D(#cnR7(c~8a9zSW=EiMbWtPx!l`p|xI1={y4+aj(FrUQr%Hmw&@37b~+*+YZmc zY6ab6j8+AmI+%jn&$_E`Cl-J8mt3l6Q#Jf0B@@SWH1&09=$G;>q@oDBM&T3yYQ9!*c(DH)^nT=Xi zr;*PT1UjXnBKAU}73){L51SVE7>jPjmt==8k*OMLXt4b~yn7u(9Y7q(WzqjyoQlb! zA79nB9~WObGlQ4p5Mr{q(M_H-w8qU$$h4#ar ziTRuZUIIZ(IkN2y*Q(w61#2N{=#&Em`=Ruk(4fEYS*)fY^b0~$u;r%JV+`-@^GQ|E zb@J30W>zXCPANS_UQQ~daA{VAseN8FiwK;R{_g$vF$5EP%bQ`zktJKgm^QFHbf`GDp0?_wX`Y6J;TU?wXBKI$b#(sr+{Ye zY8(aB%`vZ>MI~>mRIZWv<6*{pBPBnB@7TMiUVFQiIjE-#uaO}XARl$3k6%q%{*3*4 zok5lxR^*?KMBC-Lr3B}>S+kaQod4pkl>GQx7bC(-r9)oYr2kz3QYi;EQGe`fm>C)~ z1jX8XlayrEn!F1kV@Y`46p#`D`xGP?qGdRn$zLfEM>THleq}7Ja6WV;$RFE`=!qTejeP35qQ`{w5meK=K-f~ych_5vx z%J|+c{?Ws{xc6l3P9@fyz(ri>xuCoHFsHCE(|cWv5!={TBa7H>i9d_nmG0!@c>u;+>LJmB#L9n#R%h(${X_2vS zff>L3WY^5l+rDG9V2itwGLXIRz1VD*923Vv)rZFWExhsE{D1W^QQ*vfzR|b#(m#oL zwxtw_q9+;czmSs8bZZbiLmFU6PJS%{X$OY%a`TfKStO&ny7DLv6A4@dC~hKRT&d62CdT)h9CG0vR&B4U$j^Gb_R6h)=b*gyo}t_5j)$a! zq~tdn{0C&r0vgybcX=OSx~c4R`{!YDYJxXOGzBe%^6tdk{#S_aK(soT zF=O|j>A!lj33$@;Dux0v5W~nKwHxTJuR-CkPt5@ipLI3GAcL)#SlB0vR;D!zP%^U`O1}b}n0zB0+6~^}zDo_-tnL?U~(@D?)jT&K| zJRI=JbBiuO$vx52GyzLJ7m#7?3)+L%9FheNL#dg>GR`;*e`m9nz};#PmOe(+$>p#j z3!^Pq#}>CA?|L~*=ebe!tR1*zwG_r-hS6>=o$`+P!Hdu}!(_3RO}YPPBn~p8t#_2t zj(WFLFZ0h81*<>b2T@@*lVC{&sOw3v-|c#u`3RE#M*62j5Pbw2h81W*R?JuxBqmAd zac>b#+5#dK4-|mgCCO+j%fT+G3^DgK4`)-ImysBFtQgXTXWGC!HD6@!!{j7bMp-Yj zIXCc;ecXNkJ5VYS9iy!duiy;O{*DTy3z!?MLpNR$SG~eI#p! z@%UFRR{@pWN8>QL|2f3(+W!xY)G8l_e`(#a*s$)!1(-SE%H@B6tt9;$RtW3)5F2^`^(eAJfte=Q^T%rYV*&3(5!};2+dGko-F;6 zP5L_Ke^E65nU{3F#icuU#@FsLN0mDvZ<|7)gi8!Jaa75l$d4O)DpqMP6{Pe~4SMpM z)FACqp8wX5z2YYOZ9z?i?zP%oH`rO$vo43X>#nSA5^! zW)z#`OmPWw!di>H$v>9g^%dItm=jUfZ^%*h)h6@Dm*J;xRo2hkwb4rQTjEzwH%o1F z6szSAwc4pzZM^uPw9JszR}VaJ7ZrQ`eOwl}Eurp?%;T+z)Ati1s^-X2gP)(M^?bcq zkSxyB$x9sFz^gD{@X+$7qQqh5@|gdgd8W6PbXbG0gHt##r`B@!f{-!pA zzV3OcRaMCYet9%=3(s9%)LiZ@kFEbm|HxC13M={3fn9l_eG#o`K*LG}ebV0fOK4^> zRitbw9d&k3rDb%&W>DgYPm$MRnoyWa0}#5k99n^Lqj#3!$WuE z*!DdF4mPg4`S3IwJ&7gCTQ$&%IYX6J`Aru4LA!>j77|OlhJ){DTWI(l$nr4SsgMyr zCzlHPcvOn7(JVT6Kkl}*E_WF2WF;QgxM3~!VV;nL8=~|%M1)&BX~b;U_*Rmg(k%FG zGT7bw*4vMHmV~{72tRRF_`-sJ?7He@ocu6#k5IikGsge|FZ_mHW<9ZhQQXd$=p68gykK_;?9cYO~LdWv9fa7D+Lt1Q*V-$uSrP(fzQ?)v}t1~ zN#fcG)L0hvCJHlz!y^+yGK6mg(GIGyGsd{ zrow3+$x3v|0^(DQ8-mIYcg9m((U$ElTPF2#or0rp_Yhu%j_BRZi{TNDW-phxN)ucZ zZ@JEx>tyq<&<49J;%&cE^(*2V`Kf>6gn?>k3?lZy zYg`)ZLHze&ma4Ngel=JA%U{nAIN#uU9Y@eS&I>E+iQk9wFE}cixj)@_ZsZ&GminQy zR;rpS(_`mYyAj$IsPlk`fRG}4MWUv8{tecOjmOS>|Mmf~Hd3qJebYvh=1L_wgs`ET zlP1a^I}fa^JM*i=|IMy*cn4&tqT?eGl=7!Qq)gM?XIT|J;CE2`P|cM!dE&;-d*Myn z5!&7Y5fddyvx)DD*j1+HuQbhB((05?b<9~Dl_p)ot}m~yC;A=~UsH3H`zLw8!-@aj zd5UG#T#eJ}Ruj}*uTt*wA3+*!g-*0{VeI* zz%a|kx*jfh!UAl`1hmJ&OB3JAdz2gH>%_fA!?){`fjU^SS47#Vsl*rLtK3#I-zOPv zbsaJ9r_fQW4NG!Dzhp*Fy}O&&%oJZREvqK1)lM;htjXIv6_$}T_T8w-<56VaCRh=2 z{#<0L``-w7qw1WpJdIQ!52C|@UqlJAT z|FADVRWf-3u$Qkh<27l4Q4@V3=KPIkn0Kgv$y2dbHu-fc#lQ_M+D|0I{esS%irTcq zNy)EaD+Nrh?Hr^;b1+TX8!^J{q&9L#@b=p?cK9{INb++i-!spT_}V*1U)y%!?>puL zHaLN8%QMh9bl*ElSxHK^KyUb7T`x{Y>4DJ}ek+_DhIeH78Ed}_pXi#qU)#pK$+&27 z1n0+6PUD*+!D8`s1oy#~iXPzYb}d>LfqOU9ge`G8uMXe`FSTiZuX}NzU*q@vC+Yuy zcwqW&2Gs-jSvMbaGrl*$mMnW-W>VhhuIC>gjes%1qobfj1^HR@9cxLmyYpFT^0p{< z*hpufL;8I)j6s31?FhS0BKpR3@cv(#{iM*(#$P~O$kEx_<%~SD`wgdSwL=XK9|ux} zAJ_2-<1ZTeMROCYG6DzRf2?zTgt_;5X13y~T=E=o7+?XLlK*sRVQ#2`M(*Hu?;ni( z&x%tPbA`+~V&CfiajUbxsmE~rx33Rmc=@ASMBrUl;q;QUO;7iQhO~Dk9)ugPp8i?r z|1&`oSa;p>baPGh6KCCatOi`3!WJ#F2LG(D=R)QLqWUh2sdlxICSINtY|&vHe` z=Nueq?#$2M)Nv+#{oAhoho@5P!aby)%hQA#3bQaW^C{;SQ-^tqUtC9B&B{s0D!q=6 zG)HEUN%L0gnFa+nwm-;dC5{v{e8_r3U0+A+t|2#BMO|CjK?NF*uIJ-NPp?|-f&CoP zH}}B(qK+`JPk}3M0|yc)?HU)F%IT+mdJ0XgR%~(UU&rd9#}I8Y0ndBr;zf-9JY#Oj z@4e-c5F`=yV1T`ARw3l>)GFNvJ;s|2r?}dO{#AyJ&|7MeV@z04`X= zG()T`sj`D{W5u6i5lNTrecre_7$60tcNPF&_82G$fJ-;Twt&0CIe=`1E<}NHkbaJB5y7)H z+?4WnJ$jY2;xRy&Vf3@2nW8HGXN3e+Rs5xncjJvZQGZt)KeZcX4h>x-21?6C4Wu_b z718vxmOh2o-&M48lLaUP*i+>>qH}Y{0Q5XK_+gq=n~aw6W*|B*VRp6)w>&;OYFD3C z#6{fP)<*jzDA;2>2sW`RzmPe{Kt7U;o$NBY>)8sEXL9j! zcnkQ-ty?q)gBy(AfPmw3c{gGR`jjJ_M;6b!6PBAd1kLuNh)=k4xxZNz>jujOB38{` zV~@nqP~EA&d1+91??QAvyD-^iQ4Y_$8OuR<=GgK5J0_p-a(|8O;5?47*Q^q+V?7Upp6vut=XTS?{;5z^BI(vmV1`IhBM~_XrMEK^24Qe$13O;X*t+! zpLIq;+ZJ)M6WAoLfUk?#>eZVSSsn?LmVcHm7&6Rty8#d!J=?5Bp3hf`-Tw$gkerby zeTFz$5o%KWH~`nSkz3DpP3>;0T`k@YZ#*#3c$?t!Y2>>$s`>aoT zo|$3Iz;F3i;?spJ>F(HAjhcZL?cvsBcjcB3M&i#u7+srNNp^=B7ACt()=aZoHM7nZ zysR0x4=jEDA?%E8&uV)PW;iw$E?t$i&drmIh1_!9>RH$)5mHZ`??*=$vlV7}%ES&2 z)jb7fCRhrYVC#?(yzUuT%X+x$~mVU+(c(1g`d}G^$E;&vq}Er z?~YrsOHZ+Gh~e?A8JKbd_+1Oio-%0#1uy>s?H$R-O18BL1GwHdDMUqA?ZUFLt=G^v zsmHs1a5N$;FrQ_r8x+jiCX+PJU>_VD1CuLiOf3vNII6kpNv1~+G4NfO!r5!WKq5r@Kj zee6bVXm+D=hho!c;Sw0vQ;hjf!}1g+ZYWkR&)~0}kO%LvgWF+Lb{(n$ZUWkD=dod*MxJA`UP17NeKY2Sq zw#@xMZrSG)$k4@J8@26dD^ z)npxKUlui+amr}k`c#GFC0rzyTaT5C#pBB#2gG)dk;RZ`NHc;YyLrIK;^T#^;ITl@ ze6sgtHgIEciy><+92e)ik?o+dlR-WH_7ia!=Yh_RX|y*Ol9hV=mYd(Zg6uQ!a+B#b z2{dmy`PJo0>dJC3m~1{ECuSmUDZS^A*o8rQXopCa3lm-LPjctgSK(r~`KuRq=}fAA zBDyE2KxTAC7O)PqSp*;}^*z7C%(e3M@j4@kC9mav9Z*r4w{yPvs74o8l?gIFHOTj` zmSfJEcJzkcHn~h`#n?_q{qKMfv7q*)dd(0tL7BzFYJ#ksfl{F%g${4(L}bwal$wdS zrpSKcR_jY4*2laO>$$;n1_AXGt9Dmev5F8y~ z9T-0$hPup*J~pFupCQEV*%@$%aj7pFd}1^OsxdvUFUfU(~@A9MDvF{@sC zXuyiku5T=&l;e2P$uQ`(~@Jm%Dlr=s15)P;(ZqB?eYhqqHdk&sl)q=@YV$mf2znDv#l;a9`YjjhV@UoWo zab#onS^{LMk3S+!?gaztu7zhl4iHZX*^_Hc0ok4NgfPe-g==ztd*;^(Ymki(NNCRp zs_b9m!ou4P@UJ2urg&Mog%bFe4w6c^*M5yI0yYVnjkk+|k(dj)Vb zpJ0$6kLBIAMnMbdy?Z0?YB@h1{(J{B{*dZdG>VEgr+?lOz%z}Z^TK2F+o(`5Kh~#j z-hIpaGT~!_2o$}3)^isj@nsrmHoBuM=!{wP5prvTn=t@H)BHn)D$6!YYG_4^UjLn5 zK;o0ADTtB$i444%?X9oLyv|6BaCxN6xh|%_%o5FjZd6)lb~e((fBbHpd+f{_Qh- zuL^9@C2RIs0U&@oe;Xi(FO_E83UE|4D`Z&J)4k8ET|z2*P%S>XY*z;%*3NzrLJ!Ci12s3;5$PO6;Y22sma--2F z9=VuU@!IeLWKp8;-pc}1qW=->4#zE8-SYVa+0r+_`NK=timjNaMQo;0?QdGYJ<(_M zetYw;XKWx}lJ38(G@~p2(({vC-emLPz}&UFiuk~~dLx4r%Ofku0on4MYdxvF?ct1?0QHI+1Gqh>fI&A9>CJ%4D*31{!CY;@+o1nv@CM#5G{poYR3 z??2*BA?2bI&^gTZx0-4kQ4a;?%#kmPZ31c#DPMpag39b3dB84js!k#f2S?v~A?z<} zK|D1$nyAdzhl4%=;imYp9;)s$N7^|5aNI`xuQtj}#s`0rP4SvkB>Xm)`-H7P>e=!~ zA8~2K(_LqdEOGu5&|0qub~*fA45>^189~pv#{~2q{j^8%JyY*B8$wBYhl)&3x z4)iEa|300BE{o`inzQ%G8%R;vi*S48xsj%ElKfrPOc^iMjj1ti$r94Tt90Qd^j5$d zz{w`~v6A6UPrxG91UtyFMrVy_F=9)-NwVdp3!3ZeCQL! z`vcyZD*BVjj#y9IRPE*$^(nxD_OHFGKeNxjDvVdh%_)SAt1QXqm>9>6x~$wmTniW9^DhmL5$f`WZH9F zMseP{yva0XCvWDXuAHFjko;%TL4@Jc+hs+j*taqAGTjPH9K;(d433v948*}S3ixNv zS)6qg8vBa566Esq#)pqRAH7kYah`j)y*sf~788)CE8+G7#VO3Q1O|q$GSE`*-=A?P z{?5*kJ{S!0QFajY9`0#$SwRFOWJu~?3r}Nq5eC2sP#}MA$I+cdG#S6czRJI8Ys4t31N7n70z8* zsl|b?*5E8{DwKY>vq@>_jggLbduiz1_IzZl&qV=j@{G$(q8JH;ry#_C9AO4B-(wZ7ay7|_8;#X0hT4Uu= z?h6I}ajz27vH}!?5$`{{f8z|IW<$4tfn_z_>O_jjW|Xalu~J}WErdz!h zpQ&}?QB%OfOfdnI@kj(|KJG*5_NdyAsF4+mVsZ%x_qSOKPNox|{;1z3G~9k-BI(A> zLer+0d#?lR!BRJ{Mn}v}RzVc6H0LPCUt5aObE-ksru01s!ROA9nmvntS2lUfoE=2q z$QqlObJwCZFn~<`6|&J?x^HF*Qx;hx>bEDUOgP@dKQEWT*Wxfq((dv2K|>>R{}I~g zcwIPuo-h|S`z$loQx0=Om%4%!Ok(*on8}bgfx|*N7glXcKZkn0xw;*PT{!(N#%oU( z7hz6kaHaHP0@yFg^EC{~&C`&bXTAqj1^O5_wfQ>^rQ; z2I9#&9|7RiuGRYmT$YjPM%{s}csf#7i@p6^GAcKQbyvS`6opI;qzixa!)vb(wWDf? z)c*OH{z#xDprR$l+MA>)0RGHt;lyr3Dc><9 zhgj@&85({DY1G^P*?~;z*@6wl*{`Xw2P~L!4#O=78$HIo@lLjw0dd(B+-tZTHn@}r zKykYKX}CqZXP`0PB9L#AM7l7eq>w-Diw9pj)Q;g|_2{Ut9}yK5QI5Aie0FIHFoGqU zV&-}Lo&kakh9XGW)ZkuWw0Wyb#9m>OD}#k}hE?3&7PiAI0x_4u&2XK^aI2US_E|F= zQE`Hr3M1lf*yPW!(yhkiSO1d5@IMe$mOj-F zUG}K`F_n?#t&{M`KKj4X2-32@V@7{kwCZkP+0-C|cOU`)NU^Hp+b19X z!YSw0(kbCo1Zjeg;0r(sHfhmMC;=HY0h9Fe81b&4ku+%^N4@q;@%Oon*Rw+jHbiv$ zMT%F;Zy*9UK$QT>^120?L4VcFn(o_zn2Q^9V1xeVfMOI3r#oEO%voqnX>a308&d<_ z2vtjo8gORq!{wfH=utl@O5_aIIyTAURYN%MkzL`N_(JngtXzdrT&eMCBEZPDyZKy-C?uFY$NJxsu=yUcyS|7oZFFu0TyGbEr zkxIWAxlb(m>l=Qyp?IT?MU=13^(}WZE1uEfv>6`g{_agK`q_^#Yqnqu5NmW8WA6`)ShQ0B1vQ zcZ>q&tcZ7q+{C*cLoW${0r&4e2tox@h=%C`JK^+)DHa$bhI{mz+5GFZ!UtsYb*A|K zucWZcQ3|nQv4cSouid*lO5$-Yk0zi}-Z2)KdU6sV-naYJ53jc1hn+?iVDAWECr{#` z`??nK9sA3o-(GaU2%}WV&Ck7S(La9WSwWLj{vTzttA^E?#_8HSk2x&M5}W49Kt6ie zPj_hnE#<;r(@b1XLh<``2x7yqI??8`>dia}c=l6}4VQy>5XIuqyV40`2=`HF-aNs3 zP}UXru8%Jl1|2oN`=q9F2t?IJD6X#WaY5gDd_7NjX8S2^kf2(*0SL4op@K;S5g?lF zd2cZp^gkyL))Y+d%C$u_R|b`_N*VN%it7&{6DZyi6z$F09M5*ey45bHFCDdI*EgUh z9$~gv{jS3hr0#8DQas1=7EVcvGp`O@UVHu^>6Aj}@<32O@e+FLBA%Q9+R zdTsY@U-@!%>WyWGQfe`$;T9B|mj-i)hCWG$8GmEM_%;WSL30KY+X!R%q6$$vZG@3- zLd@yg3$$&Jx$4)@V8ItNk1oQ7h`siBak;2UtH>i8TlLg}Od*%g4|4zO!>q!2F>i&e zpU~Zf3_9eRPCy6tXn@seE?2F|Y?M6y%c6$q%wc*-{VvKYZ;MZsi$#z6>EH|2zN4Nc zu$2@Wph;;I?#EatG#?Cpsk@=yFS9CIW*`*-tBfnKIO6~o#2=u?RbR_|{lKYniG@!& zfmKk|{N1pby$sIa?7*Oi>WW0p*(aUT$dd0?{l6=obM#^Rp94dn#cBI?KzLz3WdGqK z&zcqU3Hhdg;Qk{?L6sMuxDIrFQvS^4%!R*aSa6|c1Qbf$Vt>M2CoJzFh-^JD&bpe> z4|3tMf|y3)OCP7v0Jk4>udNX|&%NcN+l(5b+vMipzG$WcoeRmS45IXD3>A;Hzz==~ zV~=mZW4@c8K{OxOAj=AlM*eh?rsrp{lOL1WU*{{We4`z-`UI`p?r|%s;mNSu-%Gxw z&dWVk3u$ku2Ye1+FkKA79HEMZC3i!0*MXU!#;yB? z`2=Zot$V&8#DI5h=Vj50h#xi^Br#rl*PzMm;YfmuHusmO%K3du9f1O$zBia{?~S@V zdQ8rrbzgjK zyEFp!L+cg+v^8xF7GCO+O#;y=cdI#qI*i`HxVDLGG^&&YU|x^=81x@`n1N37P>wTB z?tvShXNWJ8X|>eG(_o#8Qr)21eA6I$1I-L8U`PCRv`ifMMh7)o&k#hh05sX>-T-5o?;YJ z!>wn`)>o=%_KvfR(%vYq{6B>e|6{0n%B7Ba%8Pu|Y=mtv2x@%%cY~2PN+(5$`nE8j z5uG`yd2xW0$AcvGsDX{^{7B_&7wnCSqU+dZs{L@3b_!pF)vZV0AJ+$O0vi?UG^cN; zT>v!-*RdC;IIT7WRXS)Agd`JZa&G3KMmv+TS%c}?1|VnX#>xZUmwDw&*u;>va!IzT zH{_E7T2+?)pZ@mG%X9Soa1*X(dN|i{c$wAUp-#cWXlmK{NkXWLET+c8N3hHZg>*iA z3=x9I)By0Qr+wp>jYoWX#kR)ID#w^(*z}Co z86Q{)xg3*x&pU8rR83bX397yal=l+2Cl!;5!5xinc1*A{0n6suydXg%fx_Py!!}eY z#NZHCYNa`jEcp*2$lyQLwf}d=J{$p~E{BY6%_Ge)t^(mImp2^(i*PYV)sT*S+;x}! z4{(>0DdeIaaAVA$F{t2^Bj)7~aA~$aKMkKzC{8tHJ9)Jdev!y2B|e~qKVboSm$~-k zFFFjM{E*@Tq!P%gHo~j}ojg;)R9y8{N^1B60@5}Eyy?p=rEeO*?FuP;UIGqTF*Wpa z2l2ofK&oYf@w09}zyCTT5VidXhATb?qPYsJ{`nOD55a0U1c+)P;Gk%85YK_2S;~df z>Vkxxe+s6Kk`qi+@)nl;%{Z5gBl?>MR}UD0S|d}bOW>Le2Rcse!GZ%}`goM|dbX~> z8Ao3PPJ)d(<*a8}4j&mb3SE&|&!&SZ0nAy4I%rQb5dS5MRapV9I%o*$+-|SebU;S! z`3Y{gy$3QMeK*S^UnWBpKnQ&uN|6;8m@|1`QW^iJc0@-fn&Z zhBAuNcND8WR&{Gu<)6*lZ_<{Z{b$2*{+hyV9Ygk_QZ-Hb;nlw$B3-w-n&fssKBjaB zFf>=QQ%M$pNVfvy0R4GGJ`Q?{Ma?ywbrxoz6d@#bX(tD@rclnK{YTW`NRi1Wiz+HE zZJt~0$m|G?%b__uXEU!YYeKA9V!*aqT1=1Wr(QaOO-v7s)gB|izh^dqB8M4HG$A7t~NY@^@icnuy-+* zQpa`VcwY|?T1@1W-qgk1zW#VIp5u^~)~rGv-(Nh}BZEZRQA1X~Va-i_dm?1t#?T$+ zxVQ(qcDJD-q9X5!_W{NQ0mX5GnC@?3IlqajX>lT6e!gJH!BnyKd`;6XBJt#~c)h>~ zHRkK~w7Nj@GLBfrIJ~y0yKDSyu%JKF3Nly_qOeVPY0Dg(Sd>ir5XtkHQ|8?v8O}gq z0!l7#rjKuBx6?=$!NO@RJ0IM-tTif{+R+qjL_R%K7B|pLz|;de2-!m4h12#Dy!U7l zmrN+ieWK05+Z$}cu}_EL9fHn9PMJgNVQ06wpSW|xE{>l zMA8#@-VGPYoVn&{G0O0}+Oz9?D#JaW1sjEBYwW?ujy>-7kMofw`XT5uH67jvT-f|IJYt(zh45l?Tu4`O z(OT*IufXeg$}<&SLfql9x_6NCYy9XvBZJrYo3OJvZa@pcyj!)bVM>0%^CcX@w_D{B z6JYh_b0m&}Bpe>{+4$$fZN1$paFg6Vl?0MR`o|D(HjV;bcK8)stK^Q-P2BiP{sxBK zDrbux?Ablz|L`a@`qEUrE8|16vf$L@{*TZW~8Dmz2(jz)6_ zSc0=K7cNQP6kxcd86^b%VJL3!e{kT-ybDz>4a74x)bJo;$Dk2tRA(A6(=K^AFq{?X zijOAaxIjM+3gELzs&tzV`3t{O+(N99;iOnvYnZ_~b0eRL%r42q!ao zaUZ1}r^^T8$%Q?~XhdQErKFbwV{8*w?Y*;1nP2^M8pA24ZaHl`T^yL(xm-TClYYr8 zYSv6vjhZZwYfe5jpFAky^Gto~9i_hg zB_T03F^R%XYIuuYXaGAbZ;&|ho5nFTCw|mj$vti&d#L<%gk-?|s}heaX+@xCfA~`_ zcn9QT{FuD+({a5djyq#?cga}OhoECHkL@4J>%C_LZ{8DRsQu!e2<&U{dv$`ilaHMB zxOXwCnJJ!vmHz8>RsOSS1*LA20*Ak^yjp{R3WUPlWEi~IZyUK=a=Aa&bE+NeTyH5E z23BNRY4z3kwJ@QlX!j4@JD&Ip2|nkwx9|QJWp4phRrK|X(x7yAx3qK~8iP`#`yi>5 zbRI&w1*JKJfJ&FN)B&VHxU6f~3WF{0U!OX8YDS<3Ta(P|7`vnNTVmb32@ z{&9Sjj{9fGran0$3=Am1;kp)5bv(V=FQ^TvOq#VY=1E>mY?{j{;T6xd+VK|llD>H5 z^Hgre09sEgJAb|W=EcIb%jt*0%yNjZGOkwU)k^P~!uO{K%(oIJ(`8rEK@$cC*B32- z@>yBW4)NXh(IR5NGX|c`={Y<5IpFJf3z}RWdH&FnCx) zj7y*OlzqcJd_1k$9p^g#>yh4Hdm;sHD!Ce0b~mni!mOu|RU+r`l3`{*N$fCawg7}$ z*2*kmcmG(#TeoZnpBYb^cMx3Z63koZ@H_s^FCWLx`U(dX_C@%XdWp$p@X(PAWa2W& zkJ|uT)>TDi554UV&vwxr2H_NsWumj6f+$enM@wn*kDa2jpQhU=^`MkgngCVv(d{NG z8oT`-H1i%+no;%7bZ8%1oCjwvlSFvTmT2lAlbi%FXmJO%(a2hcFx;nsa9hOXwxeJR?1D)IhPsV-uS8g^2@5ih9q!e!QGF|G3XDFjCGaU~C-UbQtY?CCstuMFa;hJA1;AQ;6 za5>(tDN?q>Kw|MM_W40%ZUEUtt&qdy-+&6XR#E}_Oxp|0-mjbBMyFx~xDfO6(bd6R z(t8@RJzkTJWQtUVq{Lb*N#PiQnBQ;O!yW&Knm6uEDegjNxbyQ7$|BQ=6M2z&%{>MIS;|i^A&S(*;43>1P32`59QT-zO^P zIr}MefCV0I4XQ?>=Urj;PCeU&#@k)L&`BN#5iX=wNFOjCsso$sf9i5%>3^L{GFA}( zuJbwQpEVfI|34hs-KwTYhu_Rk+1x(WDgO`Hqz6#IY7?$3LjDn%S>9m$xppQS$&1(9 z*3NX_4@kB2N-@Vl-g&7asN5qi(G@RIZg`E&L5Masph$fkqHkskWU&;huMLTmj zX`oe^z^v#Hxg+;4{OIhA%tC;^I!SbJafW=}pN$5B)M@V)N_hnZwk$yAO$U}dqwZxG z*9xdIK{Jj(7R;ZiI3*teFC!i+=+!W3PrXOfDP<+gIr)$6^L&WB5DNVj^rK_F`l(ej9zg=*3c$Ry6N6R?`F^ zG65`?N+*8f@%0RqZsU{?w|>Bxs3k9qn#F0C!in`;=fUID8+KMED#1q-#i154XU;vh zh85zt5gz>^nxaQWOKNi5<&e`p{mr~V=NZlJl|x5Gu85gj_$yIFXz7)~J8lTPBGF>F zMX#lif?1{uvwNre<+{Us$vD}`FZg7?|9#Q++>JjHo@unZ<24Q%r@tyyMy-i52s~uZ zr5-T|1dzo>7Xjgm{oU&iaWsXt9p-P`{d?I=uJzm6JGv%nK{42-KoyvWE%8f`+WP_T z#4iFE=sgp)M{d6E_3u^Q4*(A=eN;N;L@lqc9}1vVrBnxW5H~A`|KdCYdVWY^J=}4M zC!XDVk1G>eWLsh4o19Q&3ymO!P1LS}O*b6odq|Xl`EXpnWjDFDtyz;(9SSB2_{-UY z3Uv{IGuPyY7uiDk!+^QsC-(#Kt ztDGvJWOVznGl`0Mc0(iZ0O!)uQ`AG(!WN?(fAjyI7(sa`P6CfU&G@CuE+v`+uvNmw z$g;T7Q3@?FMCg&^@9Co`upInnLD=a3iVH$YaPph)!vdJwL?tf5_nimwZ4I@s-*PQ|VWetF1h(#b{OkOPZIa1IqAXhJs z0`%D(UjzKxxBx&?m6Gp$K14gZDkF_X$8Zcm_PoH{*MRF+w~k`!CnjJ2#JnfJM%2fK ziyqx5tg!BsYeLYpY9X@X3$>VQYT4w>BoxcmR-?w55!hMf5YG>q@ii&HjGw(=p63ZL z;JV)K#?*RTadQkeFqw?;FY`JH{;9E>z;&X8(gm*Iwrl=(#2A{^x#Tzp1XJq?$++1L zwgVZa)(j~1Q()fb=N}BX{4zd2cQPpH&U|eDB_q^X_ zwj=oL);qAIYh=j&TBzUtX(h^IzqEHQ;His6tLM71S{o0J+LR_y5c1U}CV&iA{#!H7 zF>vGQ{PY(;Nf97p0UUryToDJnYJ1MzIeI`u)8NyqyUYBW@N0R{Db23+?GS*g6Uh7@ zpbl0*jQ#nGHv$J($G^7+8yoBcFm<}z6>bZ$_sXcHCV%YglyT6m4hb|ANt^Qz*72fM zpSKtQc4Rx|kApBR?_UMJJB)rB4!q;og0FU=mR~vuQbaf5*7A1?(+;&)s3Si7-jZ5m zm;GQ3aEPo!cHyJ~s``#1zDOkLs{fu2AVoQut&+adG4iSnKAoob&-CXlRXH;4i!^o^ z$%wMr3B^P{u+RC8OE5#?kXVOGs~t&c;%-mo{Ajy;Ev$Ch^~pQRX|L2NDz5zK-O0%% zt^-F`ybFA%TP?)8gRS=PsbT|Kue-zf0&HvxAVh#3n)(aXJ@?j()^Dx?2%sVJe$mmI zJ6jsw&Mhz=_9^aj0~BGK+2l|F8D*XiDVAUK+tqp2#%}9uR}@3I?!)vkxA9T%PUr}t zWlqnvhwBHJocmX^(Z?jIi(auRsnn~np^AntqF%L+5h}{4dP>w62raM+%I&-)%0Ng4~?V|(c-7g=?A=RFM=i7$)4Z=kya>?DlUqs&}A3a|5JH7I+ z%Z-{m;2Dd>h+mNT1>7E$$@AI^GsGkMD90V&Ev;Q0DbHCX)^DslO8qXjx7{HxR`C#+ zf{$dYh79N2t#%$E!foDb_oRS#7^R?MS#GXFf%AUTkyBE&M3!8aV^!I&3!Z-ff>b#F zOnCg9?A|^$!kga;5k5v>101Hnh;YUOK!L+xqhSn0H+IhceXkL?m(!mQacc{BW?Uqv zH)Dhf%;fTxgsf>wB5bblM?EJ&b{+qfVIBpTn2+Uox37t5j_wR_C_9rNy3ZCTgG=}LNUrUy`DE-q?E z^>aYmEr^HqbRU&`3E^2pkv8_kp)f;tr~etr>jJhCN#E z%eV7n!$#V$B<~0{=eed?0AUK6)O6=nA|6@92;Bu_VAuH&MKZf7`pd6Z7wynYz#8P| zQy*zVZB&2bhf2CbAn2R+C_(dGHKIQ$OZ^kDH&fCG#Ho84z}UT*F%If+wcQ8;3|wjx zxU5b{@9p27`5wwceU{`{gArC|#RJB@^e#=v#<}9ZTy_%ATET_l!~lfpnm%18<}IQC>S+*}{%z_8A@zO<4Ih$-O_i{6 z?(Asgq~b;G<#5@4u}+))JTh*x8TJpfTl8Qwur8=Q8uc-e#C!a8IRd7|?t&R*p?SYQ z0;a5~YO6jvTEi()hW%OnRpzx2No08q8SJ4BM~Sv#4&P~KYkWTOr6k%|cdnDQIKVg2 z;MtM+MakELnwO^kNQNBi7M!!hfr2P<06=d!|K(VD3A-d|4N8ynC-DQwJU?GF2bt9S zPC9)EW!$%Ami@%|CD~qQrmRB1&nR_5jr~)Iu7a2A&={^k@189^iUrQ>kUfUm1$^RZ z%@~oyzu50d;6NNxjhQZ#Qs^we?pg6pt>(Tkf<5Ox`tFI3f-4x;36;qr1+TL z2CiRUJ&#p?A!(1eOv~LB86cNQ3c9`An7aLEf<^}+8YYJ{C%Ky{GA{{ zuQSb9peoIHmpRQ^Wb_bYxm)T&LszpebFtb+WE4)jL;H6aXD*f2Uc3Ly#d+`^wF%=vzYYoB!^ zc%Jrjv9d}_NfTRy6845f?|)0wDavQcvn9pQd!F51PlyDEWSlJBVEr>?_V50&(#3J9 zQ^30YqyXaPo4e9jP4@^Q24Y); zBW#D}ggv?=MIfv1m@O-{RjpaKP5)tuNRVa|WtMa4-qSo&f|p~1D-Z0C2}PQ@Lz}p* zixJfOv?TtGd03XL+o~D9;{xNLuTg0tDI;gQJWpobtdy-;OH6gP?Tzoli0o=S7uD)n zuUWwYPD#nay#TNp2~NSuB8kQ5u%m3%3$qIuRz z^ka~O9OGjESxe`~viFLbhw==y1_S|PZCvENNQqO?KMtcse`QG!!zARWfs1cEio>J3 zoO1&tIw(evmqVbaaRg^6z;E) zxI>iU3r{N)?&}TvYe@2h_ZI~7H^v@VPqUg0>uI2BETEp1&c&V`_qu!7nK4-vdH!#iLs2SuOV(R8o;=*nTM#g zPM>?w%2iTno@P4o`!V!Cr}h8jV$Mu;O2RT8gP%lyXZi2M9TY=kV3i~lPAD>f`mfW! z9FfSt$$xUqWIAP_t2$_UgSXaitJ$SRJ*L|unal&3Ut13Um#a(>E^<;noQO4@cb+wr5XW#8Y zR-f;%mqVoG-aj9Ey}M2nR6QITdM_QW_F*_+Rp40Jx58TZjK$M6vIiC(%Ri0JmDhwn zC5z6*l$b!o7m#I0&@xTMcKt?G7ZRBv$r4EmR^Q+dX4ziT-Xj7>XJTZ!Dh_9Dm}aAc ziS~JqyB|kpbSUm7O4Y|EmNWaao`UBOs6Cwt!qg&(Lg==wQb1?4d1-EVu(X|a2xgEh zm_}1>cQK7#_^h+gWwpiq$Rx(7m0gRyfYt_e^^NMaNsi-;UjlEN$-#bsA6A^zYYeow zhqjyewX6%%n4xGgw(C0i4V=tRaWiCWKMAKCDx91fv`uQ~oy?P2R1Y_Ib@^mSi7Lo_ zo(vI@pRB1X@@Dup3cQ2;5kBkdUH>Sc2NM5!Tfu6<6UiYlXpm(_G%yzvcW zoAq2fi5B;mzxOUh=2kR^sa4-p$IL!59czju5a}HxoP%K&c~r;vVDZ((G?MeJGI{Io z9!`iz5E;Y~WKD6lk}JXUBBBM{UGX{2ACW%nHa&USn?S`z1>E}yLQDMbGWldgLeeM8 zn^`y7dsPajm+$4pC*Lv}ZDkS)I={6WDT%+GkMcd)c_EO>SzRp zn5cLhYsu}*#tk~y8#{2=6o#STPhR7gCqwEfl#6FHuEO2TUe%oG%ca9*@r|0C7$i}} zHnZ6U!@pt!pOIfY0c$a7w&@D7QHf2KMT`3hpD7cP(xV|fhSiy5NMTIdl;8z9CC!nS z++n>c+{y)E)BG3Ygrtj~uaZQKj!axT6g3;I8ZL#r3M2{0iT!(khufGItU!d`L*hZo z!KX5ubc!VcgJVlZvu<0+DV<;d@uBFOuvLb!lj#XWpiy^yBp3Gl4$($Uev#3=cNubkv7G9N*ND(3`bx)i zsi0>Zw6p$!3BXFJYm`G)KHl}+m&(B&0g$8Wt9UlSZ;XN)SqbzoH8V!{aDUVw7n3Ii z?hsQ{;s#q~Zj-9uhDkJbbM!loB{{(jr`X)NBrc2f@P9*>IR`)b;zjAs>NiT|KJ8@|fqxTauOm>H~r@qvQc z{it7*Gp^}Zpq3Z-hqmT)%y*gpN8SW(8JZD%1mi&YfLpYp&ULs;7Cm`DUO)zL6wY0e z9eriV&2`sKo(f-%d5*%&&Dgxr);#wC%38@&k=!!CrvfND(bl|Y-i2{=_<%ok$O7M# z+>Ea>BKq7V`>;2;89&JN{^unAm%{&zP5R#%r~j0o%2-OWV!qpfe`IDjas7Al-*4I* zK)uS%Kn>+)2%_xu>j9<3PoJ&R=^zA}$bW&gQ?68ATC@d~5h1Rx=AMBRH6>B~WDMzK zmqediYS| zm!H*>_riq>?Zms60&r+VlhV6*j>l4O-_4<(hq_p!`T?F4sT0&@RfDQ~MEp5L6@781 zXj|_ICiu9R=-|PdOIBrA%Qr!ih3h9nWCmfi^cWh zUDRoD_S4Gj(KL<+vw&^qYsj;5%xLL$$`$@W@j&+d%)hrWUJFkVnXB2FA)08goIuc9 z3>G_2`($;+VFis|mm^=Ne?FMV^V0p2k_hW4#c52#N!6O8U-H*`^?4|xtXLI?ZH2(V zQ9+}cgk`(AC|NxuBh#)e039rm8R1iPwnd-tWR?k3v2SkBP;I_1_(gkwpvF9SPeXLX z>1k^?Ad7q!l7VXowpAQXU=QI(7wI?r*oYZD?;w^(tvGz-aZ2R<_U~^tO7gDVUr^H> z$HhKt@D^dw%ekj|Kp@b@AVJ_-56&ZX9y!ooLzsGhb|c4b@WicArB8moy@uc@vT)U+ zCdnEMjaNHB5qXj0{03s*2SqQ2H#uYQS5th*sM&azQk$~stI1QIJ4hTnY;@()G47d? zVN2}DBA_qtbDvXW>85aozW`hY*_2$CHb>*n4wGCwsr36=7ma^X0_-T3>_#}`lxDb2 z3ZwC#TA$DfZiH!nlciNN8-J8)2|t$C`(s!#$GhzF%e90&M-{-&rKIv?ZHVTmo}o_> zali8H{|S3eQS9d6PS$&`zcIGP5$otiBE%L`Xi0rm!KGbzOZTq_m-3MtS+5ESH$Z~U zDJqp`0o=dN5$nt02^4EeMSF9+;^@GT?|V*Bs+$WOAKW8=5jd$fIbwBc_yF2sSh_&V z?U9h%QI6QzpeYcb0Z9LIUIA(2c5nam#T7xmtB)dhIRwDc?!Q4!6W4oDHMhH{j}QAO zBP!ppj0YixpM3vkMZwBvu#05>HBWH)@zx~B$RT)j@v1UUK6;Ge3pn`M-=Ucap~r}# z2|zAJd5>Fa$jZBYU0Bn@PDRFJCq{by7PS;LM)3(eb6g<+p&f}+8D*ww*?Mxo99ZH; zaq%X`;J;8u4tDEu^HC&C4}j2%_@4<9fCx$QW^t1w@~b`3kDA6o=+&h+;RCSRb;%k& zB~)5WKl!D5oTFN*YiyFE^RMyQHm~AR^>T>Q-{~PqYN0>jan#*{7K}qzB=5BTC^#{Z zNFuitlkWRX^WMMB^Q(Oj7%--UsaYcVtVUqvZxWum00J^MEG;KMjNP~34v3K~v#OC@ z$fucC#7~ZjWK~dWz9&IxS4mBI1d}87jIAI=`<-|<4FIiS`rBe#8jzRLj{$&(KlYhm zaOqwr1v&AsOi70D_qC1BFFW@rU;i_8Z0Y)ze4XebjKlc%g>1`7mM!a()pfx3*r*v( zPt(`iFcYjycV!zJ!?+^_pgqC1O}>;em^J*n-4uG|Ij_H}7P8awEpi!<-0ze(?Z%n| zFNy^auk!qoH77wXh-QoF{L)c#+7QJ0W4X$~fQM85v?nKUXTTef7x`=s*wq*9bSUQq z57O1TX6v5SJesb#(M;HDdDo?TC{Ji9#3Wz7J=Bz6N>xZnPE8@tP?14ZkP<%fSBwrjXPIc)0$me0Vx={(@@0QJQt-vF-FeN=6a8bjnlBGSNsZemN^FEIo=~Xse zsgNxW7|L`IwW6*f6?@d`dt#;LeU@cqX-$1%e+W2kosMz?%vqnTh$MFP9>iHplOJw4 zN7DZmTybz9^FBjHRHAvG`73T#$5ErF^a!q~w}GqUB*{TFNVl`d%kX~4U;oRRw3SBx5?=W@7ZD}5QX5!^n`HeFNh}zO5qf8d=kx#ha4*i@(t6b5zI3> zyI9XqC60k@k7tUP#*UhW^8b%r2$wCpu)YejE@}KV3O)vLj^8iCy@X(1(0Ei$Eck$ zNvq;c6yRS)3j>0K3)14YDGEc?l}qNa+=B2eECgqOiSKaADHU}W23S=7G2y4d zN7)Pc>+kfSjK|SGU?Rux7i(DHP{dF*Z5L|3jzk9-zyo5P3IJLjWb*Q(qaO0wktC5_ z^}xi)+JBU;qYm?;_yE^d7v3borx1TQ);tz!=bxXIQwaDT>=B9bP{2$Y=N=OTP>|Uk zkXz8~M41UTa_pa%f5Z+Y9R)26N;2IMr zw8lTYYvXm#T=a_5hOFX~dCQa2$8z|hwGOC*f{W-E2F~X4Y2%@~KEj^7;Hi=s3ZDYS zI^nX0X_-n!v)FGpp`3QhJ^bG=M!U-y5+1jW)l-H8_RL#Tm4>eA33b$NRNJfD?Qt)Y zb6r;dbP57E_9maBNDGpZU-7inAZT)+S_ zJ#XnY;pf8wi~bC>-th5c(Tu4#jHc#eHnaRlCq6vhAZnGKjqXz=J8%0@!+Y;g(TK~t zQg*(0B2@U=xL+^%#~ZPw2`3AixB-dqdCugb^<@F9-r-hE^v;1@#SrnWl~7^9t}~1; zKcmvTJs+;k;|4_HY^JKk0(=C}nJs_wRlCJjA*w$RInx62?5 zo`QungQqZ-!Bg+^z*AyEB2G=p#3D|319%m~B2I5-=smERl%IOjFEkX+v4V|T6)>gm z+4`fcOq~Wq*p7N}MAB_kPwC&?6M6%+KfD1{W zRlaMhTRfh3P8Cq;S}~CbAaux8} z(CCz+KYFd$9`za)ZRP46zYPqrrg0?oZwOML=JC#;aR$QOe=C-hRJM7RhO{h-ra{bA zgl#*pKYEE%ePrn#U#Cu3cIGs@bv#o>f@H|cfMiFK&>-o zR4!cg=3cGOUdA=DrNj55h%^D?Q_~J6`a4C`2QORzaUp`-?F6(fKdj&N$$2tgj;-xS z9CfP32Xo^kFSL_;7C?@<54wr?@K-}0p>S6`qOO8W+0?x=y{+`1R}sgZZGbB9>#Oaf z?5^23G4?1e>id2P`Afe9DfMewp56VJVZm1$Fp=6+R!%8(3A)!|Hg6x$z}^dJd~w|OfGwoOc7ByQ*tsZBr= zkyiJXW!d(#OZRolhir2xD>1LgZ8~!3>W;4;alSYH-Wc)Ty>_Vk0Q&Ugk7zW+*;wrR zejSuTSY09=r9#O>&&g_c|L}HIoLiwbM5LC9oJUNix$bQp**FM0Ddp-->y;az>2p4U ztuzqd=!(=nQ8(!Q{U327ipB2a+GGrRm5yv#*h$vrzTrRgs!(UULkFvKEVnE~%@8)f z>EQ$;&_zbp#Zyg1VDb5J4vIX+`b!bY%ELoYB^VVpY-z`CgOUz>9TyV4Oc*^suODLQ z5bb9Ss&2!Zf6puVF&a^iKmXcwwI!5B++8vv+F6*3YSD&PjAfX-O3A+)Co*H)ek~s=|d=Xc}I)Sb~TXd0gNQGaNgm?@^IMqhaqKz@x z3?x%*w0G8Buze$c^MIrs)5FMTGqii_$po4+G8V*zYDH`g|o$y>BWqPy!VhS14xPm$-NtsV*)i9cXAK z*H~z}FEOzw(C~o&g5n?}wEzB#+QCI*MPos8widAWa~AUOwDxqg5wdr*6Vi2cu~+r7 zeJ|iGEC}TF-w>kVCR*Xzwdly%)%eobdHLd_p<`X4qoDz0(a;Xt?Tsk*W^P)K1@jfF z>B;kzePsJYtYo>Kii>VX7XiiREP4H3k*P}O>R~Xg1M~BiOL&1 zl4wTmR?83S8B%C2#v-t<69Sy~-QO-2t7%_xe2>xiX=z*nKXcv5)9^7fp9il|*#}dt zHThg@_0* zlY|y`fG?yN>26WrPWtuhS0y6NUQ)Zyj^kZ${3j3Vkds5&Cdkx?UVkkUEWZ6$@yTBI zd(W_lsqE6f7kL*Q^H1(`rO%FSH(o@95##2h)VANu1l#LSzAf(4_RI79Sx@3<%;U2U zIUX)0&h)uX(pK>KS{A-1Qx&WSqrGVzur;#%d`@_9AMsh_FOg+r?`g;r`A%wEp@{3 z0>S*Jr03XJRxG$(%UzFyDGZNF0t3NjN@2~c=ZavAy| zc)P5uY*9P#dQ(q#yCN6ilN8Gn`W~mmPAV+KYQomDPfGn=f~Qy-^u{0XLmJ zlS6Wk>zI3W-Z-$M?Q}HN)*i(RS#UD1Dn#d|+;A3Bv~yeft2d9Ok8!FopPy5_$QYZG z^2BRkzUX28c||X^gNp&z-&yQ(FKFoVHvKamG}^ymB3qf?$Nz16hH3SkS6M-SIA}U= z+&<{x`D_kmIGRMjr~v%-yLQMb2!z}k7b!npG5scqAIj;)O}rrMcCuz;eN{rBAO+ui zn+~5x)9AA>pLEabskiPgk&{zsomwBap@1n5%k|S1jgCIN0dTl9-xN^U2^Dg}6l;G9 z?{kNC0==`ukM{>%qZo**m>=kz-?9RmwzUw`#em8qiV3Uh&47ofHuK`W4{k z<-EglH?PAJ7r6Jq@9k|Vu`3fygSU`v=mp4@mxaW`%6hT$UBNySf~tPEOX{-hVwxVI z`I2|ddr?Tp{hHw6Jz_Z#wx#9o>abB6sKsQ_qZ6@Mzv^WZ6vdk~3ZD-4>Cn>5pvdpZ zOM|rVJFOvnqS_mMuIG0}R#fh&tj3Yn_Ne0~6tau7v}e%CQ)Z31$BZ4iw6|7dqsI)5 ztv)gf-W?cNE7y>f|Lljh4t1hll(9a#TXm+t=l7O9-}ti;nU~`2adD=P15dhn)u$bz zm3Q%L&JSk0vM>=)G)9(Wab{WK&pcN-M1@_pxvZ{LoBylBBqhIozc?0vE4>aeTcR5*MFJ3-bOB4wQMZ9dLY6nq-_r z{zdKc`cs@(<9BG6Wt<<5Ye$8SqCoFjZsEsbYij8o2h*Rro7uHTM*X}U|0cJ++F|?i zRgp<};eyKhahU_kr^C!*W8L@ul26ABdOA*E=TUA@qxYg*xz#x$ zoShu$Ie$yaByzfGW?AVVlas*H_$i5pDA3(8rtsklDb!BhUSgngWf8NzJxbkd{%NhJ z_VHxht8(zv&iGAL(r>BrU0&X5-%p(jNk}q_*U1;~!I_Dz6wARu8hqRQb9fb1`5)RT zN6@pcj|aM`L_3-<8&wu9T@EwnZwFsDy9(KJ`q@`FI2G;L6uI~ZB^>NRs3G)Ihue|b zq84eZO)~hONL7O!3kh688@)`gwN`_=R#)byT@Q#~nvAaa;1k-hy7xHwy5Wr?OzNg{ zW{8utpRLUJpmHRVlFE!%u&0{YIXuoeVjff$6hPF*f+;_V^otP>rTjx`DtYqh>2q!` zkET}JfBh^un){0Bv1rfXe1_0k@DXfN8h3pss%4}Rh#KF;f4Mk(kan?q8=w9hcc<6B zz=PjtNV2Pyy1L#SrLum^{3rD%vT0bnyJ*dst;`^##WQi!7{g{+;?Ke4!{z1?f>^^p zooZXb@~iHrbi?PN2?jT9f~a%s=ZDLM)2pby87GNd)yoWK>fi6bPPlZM-!CeQcd^;| z-OU{jQLLhjXM=j1^7yie=dK0`Q&)`^MbZtMt8&JDY<4pDd~_;vGTJh`Uh7sorm1d! zIJW~qO~f77j*%$OxYN!}Nr}zh9|KAabZ9^LUfzKY{3b zRrUD9lY^***P&|Xze+b&-kgpdYCaz}vNP8bA5{|OJIKs{XfOVx5tweWNmG`3(FrPz zJFlmXOl#a4^;;*lYS2KZlSi4YE`Zj}GV>sVW9vNc9a&$Ql-*t6?eE|1^hY#gT?3I| zbC0GBX98{F>1y&B{2x9?LnP|2TPF!5LiK-&p0Xj0k7RzSGpsYuvAUnL3khJNmWJra zM7RIE?grr=zvNm@{pLIkx529V>WEFw)h7PfkpVZ&?RCLoe(7 zu$BP7_-6HDKjmzo^-Io1seGtaAFnM_$Tgg~cDpDqjhg;$=k2e}8EtiT{R?co4Le=< z1?;+0mG&fILY7MGGyGyH9rKZCYeOjHEfIR=^*DBh-D};Bhwa*Eh~WLZ?8~q(zXdxP z6=9&0y4CXGPCA^!2>8X(vOEaEzWf)RWdlm-Tv9&J=S~2_yX(K2&#d?$oaMjwITc* zrJ>~O_&l@1G(Hbf{$|hz%^&Fwy5Hab;W|$qckuTz`!hWJaH$Qwqus~qcJFsqUVo{A zYAUm-H986l!ta&hr}WKMYLO5 z3?z{JS6YyufNh1efQH4%=xArmZRyCjICRmwAv`?P6!n?T%WHyWsrIqR#K}B!Gmvmk zC&)GYmI5p85sJY7VzTmAb`a9x^kPxzVGsh=XhP#&gBJaDb=i}#B|_nmniSk}$0IZa zM-5RA8d_?NTI$FFqAg2S2l@N+k^*Kf4(IE{+7GnXB5T(Y-l+`0ohJ~=x!;|u-$^eW z`RhUYQ4gLPoN)CT&5#U!z0x)D8J~H0#Qo~!$Vq^B<0BW! z=Xp@ewcUdYS_R>16&Q%LSk1VW;#@eE4IcjE7HaaxYm?Q}F-7lTK<5c!OsDGf1P5F(jUIQ zgMrf1+djdb>jZs9+VA4#40kHCqITh9&*o*$g$Mlt*`ECdvB#dh#4(B2mhm_b??f@A z6@SvpOoUK<<^#t5_S6if2BU+w_}{@57V6TQ@&rDHl%77<;drMp#|0Vg?bvUqxCB+~oL7;0 zW~3lD*KtEp^Sw+g)P|=?11N@^{vFm^^L`MB0w1nJwy2>*WKvhh%tU^wv-m9dxL@*y zNGA|k9)AtmT9K1@r&ebm_FsqIDn~VRgM<1FA?i#VEUzo0WvAQb-s2koa_yVf!C1mh zYUBo`Pg5VM#x-mdMverfL4S+94w;szkmAmkR#PxO7Dp}f)oVuqfju+$W!LegYKIHI zVw4HI8{}&!G-ZLL=pI&cO<>aXyry;~HAw0RTc{3kHaHKG=zIeV+W@wo$+S9KfpP>% zzN|(s!#1hOjzchn6#acK;nlG9A(9b#{U&6(bmw%8Bv8=N>7)un#6Xrf_j9B{U8j)6 zn9ehds>kILj=dF!wB57)r7se-Y;H$-;z~f|4!s0^o>GeUg!EO|f1Vf}z3bdh7%9^5 zQ{gnAuLr;3ePYw4yjPloDx-W&S92fLi&J5wk+q969`rq8tcq&TSgW$qVFqH#%x)mt zS~iI;qg7^}*VIVFp=ZRH9_k~~iI{k~d~Q!-)u6<8Z2#$NKJAhn@@y}f+vw>2aw?~d zT9VKR@!IJKTX@?N-_eFo%b_)$t(QQh+nVT=jlV>1_xS=5elQZc$HJrf|ri;}Tjr~m((mXYc^2EW4rhlLW#vp|w2NHfE8rsp*(OylYI@NTCw52rXBrUa5B|~(wa_`_* z4M$vA)%Wvw404aW*<1IkTRbKF^FsKICH-lThjyT_RX2#OLTam0HArdWZCM{N=@J#VArVqCXQ$9GUEhmGo=A3@WWguX@`1af zXSVB!A`!q^;njoZpXXem1zYDxou3z0-)lNla=3H;71IApjQ2C#6|KCVb&dU-@!6rULI97ZvbR&1{6~ldeoc$$UGIp7x4{zyXq#Guk=E_b z73UCc9XOFtW#rwy5ncq1y9AKmTL#$| zmC2^z`Pf&exA4K&WY-O7;?i2UTcI-F{;mBj!!c>tn{Wd;+RMJ!Q)2E-=9ihV3`f`| zqz1u5u#egGU_=tFgRC`us)1*ywBA6~b!7zo7eMbV`i{JWk4(sQ8v9m(C#!eKsO|aU(7y={j(!C7OZX6BnM@D|(Z?f~<+WTW!Mz6)!l>65y^Zl|N zOO7iODBG_u%$6a{{{$g0sd*sl(|D|Z%ME-5TWa@L^0&E=w5P;%Y1LZ)`7KF)-|Eof z;WksTJxL;Czcm1d%Vw{4zjG1qfNT>U7)egYZp-4!2*mnof4t)JJ~yC7)3u!*70r8^ z^3nSvyl66(Rbihjr7ky%RC072g8Mmc{ zT*OiEH^@(KSYbFKkH zWi+e3aJWJFCbW)|Cn?ZA+#oRhj3soN`FjVwRRL?sZvFc6`M`&Z9gS*uux zKj9#3k=DWb5~Pll=Fz%Bdw&VSeNw-%#=tP9SU7=ds~C^9ea$V@`Np_mZ2aa{1`}N! z1w3`XoI5b8VT zZcv5tZ7H6@V}y!kHBPh9CIk=F6zQ|J%4)=U z36eoJj~SeSxRA}=CK+ml#vEDyb=ZJVr|^6mJ5cg3top?8s%&N_fdxVTmmfR8CTIok z&%DbLADT6!km##xxlPFnRuc}gN6#A1)>hCimyPHpzx7YDEW3@EN)2{$ovfIR3X{aB zng=Q5-Dx6Emf!5If|BcWIrbG>zQl8n)?H5RlXb%D>(WwwPRzse{dm4zc%8$%S{m@lf2U7w3RQI zB`8hZYt;Ih#kovZefM7J{nWr~V0+3Ojo31>1L;IX+34@Ok`_F^9 zKD}(0O|?HK?vh_5hqZ!@&h(C|CmMrBsFbQmy zb#O6Of+h*Yclie>PRA0zsH$rvAr4tS;{qRGXLfv;~c>;`L2eD$)0nNcF=;05|otuUv#|% zP+U#(FO0iGaCdjN#oYn~3GM_4?zXtQ`w}3*Jp^}m4T0e91P!(i(@J~d14Xy^L?*EQX6-`~;l_zW2r_>n!KKI2awzAGj|C`$k@ zntZv(LY4;ZMvI$8W?f+~3Ea#Ic|ZN#ig~h2CrfF~R;CxU+70{Mw*sobk1ewL`gGYx z@^rLjG#3z5o(CfXA(=qxgwc(W8vyKo2;WVhy@oXUVY~0J!&4=K_n0v~9heS5kej}C zn0K_PSJVJ~=zF;7;e*E`YdQoLuvgRros zk-Xp=A?yNi`R8tvrRbN5;^j<$v^${aX72WI#*LLCu_i?>8?c0;2To67zc^RGw5d)aT_x00U-2y;^NPScI2G0o?gR}IOI0bt zgO8V52-rEoCU|R$n4VoMi9D!Lkqx$jgUp9}%7q@huxIICg^-}fYGW+4eh^R}?uN$%xcQ5R92B*&kxqZo{;{O)OC4bu@+zmZ!({7-Js!FP zzLI{-kj`gq%I}DjHC5=(?2`S-y|5I->zRWi1&4nCYhmwMi5WQilVA+6p9@qIa-mQdipc+SY=rLsxUUQMldqWtNafuMVp zEKSd8E2=DQ)DIxsIt`y0NE-Sgof2U_^WqewSkn46Xc6f*Awb$@&hJ4J*>NiB1tp2> zDLU5Ks`{N5y=2zqHG;1m4W6AjMBkbVMU;rjFH{eQp64c6_oY%oV|6*m#zwIuxizcc z@Lt-kP)W{1(%_T4^ih0`3M_PV>kGd_g~aPIDXvZtYvDH)L}K6T3taZ0XARt$v}3NB z0N^pXDbl*NYbGih zB~4D+ZU7)%lV!|~B`S1DM2W_QJ0>5w^p`}L%#a<~3poxrF+h#n@A-Z%_i7eVuM(z3 z_*h&}gcmLjrvM+Zuv-<~%Y|J_p9*?mi*zZSAyOscmr-Ovrga zBrepwCY*}MSPY7{0t*h5ix{-ZH%~LR#BCJn5frE(#BcbLnkYVO<_|9=d;IRidoO7_ zB9FNzpxRvfyLX<147?E#rptxpVR7Ic>w6DG&EbJT`0ehHgkrruHn;^JQv zqJ{cnCo?z6R)JbQu+Nm$)sobM`(B)x`H5ZV0lth|gC`Y6zFn|GVW|GyF4^}(E7BcA z^Phtx{Jox;!Y%a!022yu{qjh6=#xC!`h@MRP{$Xwonq+AY74psVJLgcp%5Hl#2nni z-G=GYC6I(zEx}ogdpo;U{&n>F79oXOgWT`a`^d-iX4B)>FNLIb)Q=Xp!?SCj@mrUc zJO`l3r&KoJ2!&lrZ#Eo%`LRskk|Fp6e%-d2%!q; z*n2WjdmYE?jnETXM1X^qW!f*Tf!Xb(^X~|W7~n-m9DsRr4(Lo>0XeFqGK*Cbz1UJ~ zSF4OHgZXdt;;PfY3KAr>l@Y8aKsytJEQIod_|ed+z0FYFL2K8$eM=k${;ePh`NGed zi8SN!WDwmp8UxG<{$e$;>X!nFB1DGur1I=Ca};!QV{h%_tmGhZ_P+hlL`mq z!}024kQwx@lGn-5U=k8zpsDVlsfY>cUJHBkJXftzkPUam3v2k7duP32EcJ;z!+sy6 z*zCp{=J+l#CKJq;uFjyZ&rb_o`jBG|4lNFK(<(^?E;oO1*Fr;*KYKd*L_DLw{lp`{ z{49P}2E!o0%wwU8z$xddBF@}cwv8DOW^8=%Pb2l(Xyb>O|A)#Z;xk21K9-)>!3-}c zk7ep8v(Fihh#3mULwzH^+aBYfC(UScbDb#_K16-8zb-y*;u0w$Z;z~q<@{sDz*~#@ zH3o?R58`%zI7ac0sGslqOMFISq`%NcDkStWbG0)@A$P0k)>8zO``6~*gE0!-jp8Ay zjLH5(8lH%8l5K|0BzuPbeg>F z`=yoAv58`Sm0Pe{(g}a@U3evVwY0c~Pn+WTJTJ<8*=5*nB+H(XPN3n= zWqtp<0#Ul?C!R^lQ&I{@#HjO|z#}2?y|wh`?W21E{%+A509AL~sHy0`zhGN6)o}JxV;{Ut@N0 zI|A?&|B{VOfmAq%xn7VfW0FA2ZD#>#Tllx7GQwKo$D_ZhT?L_Dng4JSvWut{eWX(i zX8}HZ*BVIctl?8g5D`X>zcB15ywHw9+7C-|T*|>FWwj5G#Vd4!gNKlN55B#U=oy!) z@SR`ph6hZ_jIWyDSTEnsrqdRTgy{S$D`vwq_Fr!EY+Fbuewe(6KZGSEl&l=VEoJpn zpAgY|2|{1Dl=e?s?`GzkVTPm+@8R>AXj}-T$gM9spTvj)1}6F6Fq{kHMw5`( zoIB;BVPe9k^1*A}QmO`hyVSR~f2i9fc5EEx*RM2IkBOfa?&OX|(O%{3->3LZD$UB= zWdwe5OA$HxR5Kunw#_DeC=vQrpr}n)#h#aMWx%g^JsUxW1i^(qLEydZ5>zC1G~TUs zp@4}X{y5io=;hrVjhbZauhJcYmcN94ykkO_z|wk(r*WS)cJD?-k zdh$NpxPscFqvuZHcD77R?EYSdbqs8mi#&r=95%HGDzd&AXVQ^AJd`wjq=0#6jWy9W z?i!(8L4BL4hsGP-qiR)RgCmQW%f$G&tqsFEM*jX|@e!yVDc4BQZLtg%8ZpHm zK4ueOYD5ikYs!ENp9hC0JF@1mhD6~palu!Xd&8Ubd1VxR|9ZPEGlG+ijH` zj%kj@U20uSzWE7Kx~HVd;Em^$`TEttCm{{x!=#XGzmc$9B4oT-?e~nszBs3Kh}oQ3 zt=NJ-;CU_><0G}K3A?VprL&RPfgDF64?&$R<pU({yd#-+b z*%rPFQpd^XB@!W>t+TYYVKxa+{%&pW+EC3W_-pg8s^@J<84+FSbEQCH#a=#*bp!K^ z2+O13=(>9OI|rG>i==)5H^l_3W}FRMI-v_5gnve?yd5HBdsDJVvfPzaSoxTq!`(`0 zZt>UYmF?OHztN{Z7zJ&!4T)P?VDvP=^rQ;A?O(Kv%`EIDtgfeezI*EggA&su$e`2Bp%T_K-srjc0eMfG=0^Iqh5mPxHs zfqFX7XTjZK)MCX+*?~Ph&(qr!WueFp^=+&%23Ys~;0_t9TFu>y;qA&?N3(>`o80uH z^{PdOyb1?p32G}b>t$jt{GO994NoVsh&|Qv&I|l)Cjl5^=v<5M;Ib78HYp!`tfApX zT4h1KAbzwnWeR>obsLrlR>+)><}-uYp>kKMq_&kE-q?6A(gX`Y&5`C+;_)Om&3=}+ zM6k4!)F9 zHmEUsPAEc%RTQfhZgc6I9$&rtuQXYfVZavF@9FcjFlbWu-F*Z_^tYc=r5b#Z+H>0A z#o&hQEOHeV$S0~4_Df;@I$P7dT847%IN-W%xApRv6m+3w4xWlbBvTu_M`1ctmYxhD zee-nKMcQALy`F88GcglD&<1;$nW4bM0U0+oIXAy(TVR0vVV-=xJlZ{{`(8U7qFFIN zz}@_jxz*Hq_8L(~dLnomIwSOj)dJ(^y%iWThfBo)qmC)LV4y1w8G+{7&`tuZb7xE# z&ybUWUNw{?-_sGl&`d zZv_;TVIUf2=)5Cd71-$PuB8pWJs3AGh%L^}0X9)lTIh@q6TCepp^g5*b;7yTw3v|< z)P+@v1C|si?#`lY1}6d`-z~cfLQ5YhFboE+DmOV=N(Kj-efI#)Af5KeqBqy8M05bur#cVfqf?wJyS4>3|XyhlQrF6 zCWf-2BA;V9pq@Zfmj{tURM@H9S%%v&%vl=?UbMbd=LB;!2GS05Kj&(&vx$HGd)fCs z*5={w$#Wb_rom;8(y3xYvlRUf7{!u zIihkK-I>>qtw2PF=#E;XI4>icSM70MPG30S6Qy#Ji*)NyL|?aSYB8f%LjR|TgaTlF zs`jq-kxlxL3M!HY8=vD`Gv{ZApO{kv?Vy_8-@}-qeLD88j15sIn9QCXDWdNk#L80^ zJKOfS@AdZBDsa&!zUc1mnjM`o{`lU;P>}=0RB>^#?B8)-3I_)EPFwM(BPO=(x%M#7 zJ|^|a%TChZ?;?G6^FFT7Y9POHygd0cliQ&bBw_UCVXB*^YD9a~~DUtFribT_^W zwb6D)BILa|H{?J{)@5Wr{HpzJAJ#kE**8(30ddLED6+_R|3Sla%IUNjNU)r2-xSYy zXMD9R)9T{aY=`0>T+~QZuk5Gunnxf62!se#aO{FyI8PVHmmLBT`>}z7J4u55Y(^Lr zUsRdm4W2GSzvOMahRS+tP=+8owa$4Fji{5xfPbhm>rvWv&JF7=l0rvHcckHL#v3Yp zuPhP4DOxq;7HgIng;DF$j1a?V1o7t5b7)Nb)Cjlqucox2F@<%5r)TA)4VIMoO6f;U&xT>4lrvpPAHI?LlpG*fxv(<=~D zefHG@;pctxn2hZ~Hml0fQue2@($Q_S!41?mt)KU+F7vnp>qyWARR@sZex(3Q;hWW8 z4zedEZp>Hp2Kin_`91*t!S8k{b)lsTqZzylBf%T&YAt+sLLVXxE_61jWP@u@S&FxF z!V9BsPotXXK<+sjgA5JRg@$f^ka*(I$^N#cqS>H7(q-Jjcmyywy=sidq(DQG(EFT& z+?pCH$_VyQBvnc}+4Xdn{uD@AU{_oxTnNp-HL^z)EDoRSYv7hk4oD#F%AHt4Gj2_R zF56I}DO`4v{bWc;vHuw59Mjz|;M8CmRb&VZQ`+3ReSF%Xjm7X`l0BdK-5Gb)w&LA8es>NDZ5Tr?XivGxUvd?)2_XNZhj$a(# z$PzRkEDVL3=54gvUqv3a1k$L%O(pAGs1*`6fHs>0muB$)ra zDlhmnBk0KAr`mP4ugt;t{_U>DQA{e`eTW|Qvmv0>KQhMIL1 zfl_o^DtbPc=>jA_xI2;RFhr%({2Fz2Uh(Bfao5$MIU|F@_->MaiGTd{OF9uYnN{9~ z-GKQu<1J{pIASGiNuze)nRw=#*VPTL1XAFxmnvwhCIizqFdtUEwdXWR=Qt>gM&y~e zNf+}E4(!DtLI%$V-7RuKzyrxZjmoQGwnf$AUIY2i3qpDjelZxj&&;L<(cGrxRb#~i z5GJ=eEhGQ#XAs5thNFjsbera?Of1tKA`de%a`DB+dwkV*!SHiH#|!&p|Mgkw%HO&< z=i5HNL&>lb^w6qs^gTrpx|?)&NlK*&WF_%M@l6b)It(hW0zz6opOqC<&Z;9Q{|uTa zr6}ztb6*TW%}6abw7V0BHEd8h|BF?}ZdVZ$*6K7Wkz%QtP=ufJcLNV{BB8Q6JrX1R z)M6C}k*`0Znmc|h&&mXrNVI5Ph4k76_-N6JG06;yLRM|cx5}XzaOFmXj2-=0d?E}3 zD9TJ|?wkT;H)sMe{fJ7hG%|}7Q^es!>dBKy?J>GwZTJPjZtdgb1^+(QMn6wphm8!@FQ=F9FAI&eR-Z9-n@w3HV32FqWlSaPwO3R(HEv&!+H)u-zMs~RdEpsm)ztVzP{>eul20) z#yOPJ7~v|6bUIlcN6e7m7P*;$esNLSI);OdklwDL0k(yb3G+W4QT>}b4!37#SGrTS zgH;$S!y)=|T$dNr#mKkT|IgM9I%f)TqEtr2n~?dU)&?De53wZJfJ)6P;ro?{P^KDJ2*X2{rKy0xN zt38pteq-ZR%E(|JemXLL{rveD)t?qpc>)%wiy<0u@~jOn{QSih(+fSJqjAn^kTqz^ zfIp}FXUc2VQE-c{=+$$M`mfFBoJJPgQlT)uA{7)|qqWLj$sE_$nkQu=oFCei0J!s0 zqw>*1ixe*^h1dl1qb}V^r6_QsFE|X=<{|i0vpL0?$NekrQY+S__jhNu==MH(TnEYn ziC*r}G&9DR2}>y>+UZcP4V8ymxoC4eMDYX zdV02u)gDpCLU#BF5BOU=9A9i!5J))tYHfQ=Aq+4q9T3trE z3MR`)tMB?j&f?sxBtyqB#IHmOT13s#IW`|GX*W+R%$(rVFxnLqhhoh7s^n_J{bngj zgWee_z~fLg7;u<)83UKnXfc< ztQ5Ebo_(!92GQ`}juXr$JMrEOlpQbyfBMe?Ltz~U z0HFwcr`!-|yQ#0QZ}{1Q^<#7-C-D-qeJhR#gOOfNe4gjtR#qIoiehTfFXM2@NlD)O zXV%QY&TYJBBYEZ_dY}FylP$()1eXt-3OyY|?sGKp8BiP`UDtbDK0(}T0_1~+4lsE$V-u)%K=V%D(PN09n~JV6LhVq{{t?1%e>wO^v3Q-& zg9cw(s1cVQ?2)5RqaUAWUid?D=ZdnQG!@Fb4Z+k6(Q`tuvZI$9utPZcNh(e8X3#)U z04w)TRl91OxYF2PaF?!=2y#j_HFFK3+Pa_3`Ft;x6u@{c3^sK>9%k5AQ(sYT zqHvR+@UwhXb6zEsypvexp59+B{c|KUd4A;dJxoZ0c;8iEI)v9K$bpzn1bz9wH@upy z7-Y}r)7<08abLM3+7&#IWCQON5340&p$JG1xS=yy1X7)lqljW+?P zY|RVR0G?;UQDW~jr2^e%7(Kz})%o%s)7O7P@DMfR(=*f)l}ERY_o}{_v#s3F*=v){ zAkg5;_)z~@*YTq)yTOGk?7D56wL%{K@2u)PQoXOrgiDs-^QCoFgbAwr%4Hz(ul|x# zjkJq89x~AZ_=XN&{Y)_$9O7kw2bM!_DtiU{{1xw6O%2y=nziWcs}5A(jxg??2Uv>049M8q_r%7#~P? zBKkK7p0kE_#X2qWhbp1yL4PwmSH)eI>XC@^A$%m%XQU6GkO+!h?DQel0RWqYuxEy)LZ6?>nfx+K?x~4FJZ9 zw1F*T$`5U@TBj}xar(aR;CV8hZ*y3tJ&}u-w$?tlTu~x{neQJa{XU^g)W;{EB-0-2eP-y59>o}z-B{H7R zGG1X4t@;!7fQHYQ@;TZ-=RXT`mV`1a_eUbKSo`Plm4Mj+b^^s7{;v?iae|zJZsz+V zlaC=qDO~E6y`#_igpj=FTSrgIpY%?9a_Z{22Tq!%rfe0JEYr@#N$gi=m6a?lCMDrs z@+*z8?!^MOgg78 z+P!f0%dOtY-7VE>Bi_jZ=Xl=QNbsA>KNVJ6qkivSV61oe=XCtUA&xw+8PD}Qeji`< zIg9k#`n+rMGfL~OhdH9V)=@PBaM?P5+%+tFH>>*`KP44GG9f*!tK0d{_jyl`O2(lz zk$kl4q#WoRhs-;-&F0S!j;!$|Q{u19b5>`BIr=e0Cg#;g%Vr!G`G!L)x6k7%@d4W| zoS{}?$`hy#zd@bhTjRftUCfuXwGahl$;5c-FN>V8btgA_gq^pY=b)XXQSw?(hZ(-lh`+Pf#?W*i>U=^=aCU=h_GY-OU3-f7HVM;*GUuAM z=_DH7^F$i1X~B-#u$Ku=iBFD_-*8P~^dt(TOZPzD0^0bF7PxQcjuN`f%|aC^8=_&= zq7dc!$x*>_Mk~#v(yoo}th`s*3KRcGoug6Y5;@gOMk1-B3k~f4S$rwlU63~y{d{7s zm$w6$wBIKJxIDURjakxmUw6RutF|plJXq~9`^G!ggW@&4zzv6c&ARWyji!8zP;iFA z){xDSCZ))(iT`Z+yHpgTk4p_#bUriCCH)app>XKVpyOio+F>8$Z}9}+it!a?Q(4s3 zFtMgZJpK~z6A83n_y82e_nmhm^h#_(;W^)4F|o%Z`65c8t^!XkgSRV9{521QD+Gtc zK7PAKD*F9Zh_$PW*fh15xLfs4gW4Hjy`1xAQd}%eBO`Zjb%l(}CHHVnC2FT(YvS0%? zY;krtcYxLbfiQRZ(g>Xqt$`y{Br)@!k#H~Zf|(}o5FPlGTpM41pjTj7;O*Grn8ZA| z9Q7H(luT2*d7Pd1#EPS)p$LT^%XD#=+=Ro*bt^r}Uq<9V3I%!5vwYk1G3>`r<(20MBJyN{;kpWh zltDue#KIOU7~1WszCUVm3QGQ?3V4m!;@)vqSiyXI zEch(&UxWE#%(Go%k~OoTD!5TsN6cT?fjFy?QuH)?cy1&Z{jc4q{(#QZIG3rVe>tRIF*gvl3g z)X{IHwGp4YDoqB;532lNlA#3C&HTHY!rAwo5z%axV0Dnicn*a)q0%R+jBozO%h|C{cCJ`ScGRg= zM!UR~_U5(+@b?z9$JI-NqEo%y=${Zw{6&9DI901xRbD2{ixAcvpbzq4P@ivsYxQKrio{}Pb z4ym)pF7iacB8Nv$DFhL@a&Z|kcN$fmdT7~TU*)f51$Wk3l3qPa_^Q6_iY!u78tQIb z%=hMPP;wR|d5_;Jlh1YQ3lzuKoQZzo-wMwGZYJjTj-etNRPb}bdV=>L6g?ey!FWiX zR{0w#ZHUU+pO=J5`s*XS^%1u(Byi+A1K5)_e(axJKXQMuHObzTo%1?@-ma|AhdM^; zMhVD6mLx;8`Z}xVx~&cUfBj2 z#ZHjzmyC<4j$qm~pIL8{*vYA`&2eA3lc zW72Ku9&v#t7cWA_5aNv}($#Fn4cQ`<*^1Plvl|{*xu2d6)@EXc--w>#7wBFD3Cv|; zSfZhS-r8^k!ffh5S=r`rF`<4-V?9PqFlINi+81Lzzh^=b0$80uEXAF#cn(Ugr2~I+ z;W}OlZtY-oS0r_$d^CBVQ^5sn|46J&GNi0azxy#uFemYdqfXQI-r_|KcB7T^X&22@`s3`c|#T3qC~8J&(ChH(&o_LM2u6vKVd64(+&zMTToUyRISV zTgauCJav1x{hEG1;0%t(ue+680nwAJ@vK0)Jss^B3v)mjWk97O#94EB8ZqOgBIL4k zRrpETvqV+cRkEdkC(mn7GPMBGl)CxIJs=Q%Cu~x#&rin(mq*2l;^qQzv4*5 zU>#H5Z8yOrz^f#f_$Tk?FzuH{Yr$@iYjT3tCoCa;Z>*`GIKq#==t}+ina@V;bEM=Z zCG-$0`IS2i?HG!@QVH4o{I^{g)GjUQM*#Kad@?iLsG=q>#F@w=yQp1tcKxdz@)n9xb9s4Cfd4o*E@5u45|0%6g2p#qmFyMo%;N_cZ_qM%zT6o z9ojQgK{mQ-gV(qWD&G}z7Sj1;^%EzfPdfn=eiEU6Ib3y&1ZDfT?R@fxN5^)8v5Oh@Ry?9Z{jcldu8=Ns@D1Y zx5${p;*#H=d*!dpoLEQ%_y43%{!2ZAY!KAtYJ|e#e~&m_N6|KPR8RwI>gg>% z`ad~Bo^a1oO}ylc1J1#j9CMWDnNXAxGh}rG*nLnhz~0^w*bs8&_&T|7c&YmOy3s_g)bTKNlhbc5q zQp^@m2-GhMa<&c_Y3WTlwQ$$b3Fkh%Yc(p|2g$?oA*&FU_>#r}fJ}tdAdFVF@026t z*25iHmdj7U)gJ3NM7D>z3G6Y>A1)a{K)Z>5vcgE#PYbJ@fXl8oaU26TJ`DII@HOg$ z9QR=**#T|aHPiP_P*^kCHFmp$XMqET=q=D2PSUj6qbc91167rUuG1C)IXl%|6m}9| zEAR#m$>*$Ltv!W3$2;4>-a$YT!nHfb=45B(0pH$d#+0Q#xyPqX0gMC(rb{bWjLrg# zKAd&Zv){^yH-L4*dBP&^fAmBaxbVboBkgFQ$#6nMeuSv`AvY>CJb8)_p=i`Cx9a5+ zGrc13Is{eI*Y|ZMnjbj?C95ZT9m2S_^!(8D-qykwZvVI?&9qm54g2_Ut0PErr-<>{ zU`IuK0I$Fg#<?uac!Xz3ZrXCn!Y1rBOr9u#?m1+(oC6wf6d z;-KTa;}F!#W~r=<2xX~qK9PA9uMGTrE;4l*|60#iw=BfINBf*+iEegg)UGTy!3_iT z@b0&fJ?iThs23JKwC*R1G1d2&TX#4Vd5d#NYuZKZzt5nNAo z7`I*HPmlP*qsE0ty}@{=5=7%SvUw-*@TE!F^bi(4TT~1f)E0m4bRxcaaFwD z5M`yyJ`M{Xi|{hZUkI8=Rz+k5#GRxqSy-Y$&pWeqX3pXgL|`ZF7i=KUGsfkxpRSu} zL^PFnpQB69oEml_rZbd$DBD4c)?EyK`171l^qi1}SJQX(VV<$<&uCx$j7T+^zgaek zJrNfzjPaN4)4sMu_!M$R!Id#tKTS;~*T7GyhK0>`zrU#O9@)_#=JOqsyzVX!z*;B_ zgt-rL7LH%_yP_f5ForzxGZmY!{rSoOVFL(5PKwSS@8P*ii`iACFqx3RdoM7l)nJ3V zI`HH7NDZ*|=TnqY`NiFDpan{zZV((;dmbhny+`nmzgqE+}A3hj~b2PtDH zu|3_wTAPno9WI_)!+Vs^-c6~POT!&DSPJ~y`%jtRo5;A*K19%$5?lm z-6OnBbOai@xlg=II9+?rP~D6l0mFQBiXwr`G^Hr=+Fvp@g!iQ0@dFDBwR(Lwn#uxi zo`b3Z1+~bSMv+< znOjZt4{P<}6hF{CEOyTZ`jt|Q!5%!?Dj8Dc*qr)f<5!w~_s83cx61y8Lp4}k-Y`E) za<&ga=nMteY)#-rSq`){&gELN3ZIV0(QeU())Wia@_gXL_i%Khl^?K-o~ooqpeHT3CeWAr8QVJ#3yAawTD!GUfVI{$GnIv@7|cJ;UDCF zvq--jfcz`{lr>`}^CvFiwD5HN22S)FQD)&7K}|@TaDErfDgt z0%(?oO!iag;9iVgdi!kPi&3hM&b^&_Ix%aIJ<1;-%5`b>u#FgrM)~f3J+x+=b z@*qYd!8Dh<%tZdgEPwL-y5+NVDIkiu56Pf&Sg8(4ozuQjR|W`qdl!_59eR_DO21`K za{!`XWE-Yz^m|AJHKkrLm*e+M1O5QRS-)d6@SP!V;VlgIp9n_rIrjIH%q>m= zurD`7gTsZn5jvla6lLqz7*_7XU)eR{7M!hoh0rC%K9v1h)w;S$))sUh*)2o3px}_( zirz7udhQQuD?%+OYD#sO5e30IV#FtfkR!;7gG;gNEIfH_=$dnzQl~S3S{-@7<4WCW zfW?dwazdm`0#8@j|2LNZcX+`7OgcMc+s~!QAHq!0?oc^CN|$q0;dc+6qwUE>sqn=q zV@esWm{gPJ=~7NxBv>pO{)C^sU}x8(XD$lXOl8qtsOR2PpSt9piH&@E{#tl{!oXmV zWU*g8vb$leReO{g+x~30vF%hyU=Q@L7;k{Jf74x>H^--%=&h$QXW?)g>T*h7dce3F zKD@)d@qQiP!c>q?>#pKKOVVQ z_(WDM!`h5Lm^Va+zpB8S$d%M-*<|Ga&?xjp5b|Iz?WT|wtK+PW7^%$%ghz=4F$jR{ zmcZ!ZA8_Y|0`D|9qHD={>X*${>RO|et?`kL>KOl$nn+QMd>InoPIH!_rzEnDc0{T2 zsg#P%A5syA^q%nBYyt z!)8&?!1>qr={praC$9O@ZAbxISUC1xTMP+u58x{fi>elJ=6@RL6Y0h zWsAZbJ=I(=su~%fi>4}UDOu%s9naYvV*)0V9$`A*`=+-uD17!@a{oH7(Ve&tgVC)o z?%|QM5{XGCv_b8?2CwJz0~21DyVms~T{;4D4@@~9i*X6hrlsXC96jDW{y;N!qBvpx}(>J!1hqJSx;ZcdgSVk$1L{~D2M16i#@4>l4iO({H= z(Nd)tNqs8sYG5WOS0j$TnVpg@8o-MJ=AkCCZ$rbI0i3{~DaR$(iFUJwkg%_pQQ0P1 zWX^ajjJ-p!A+b_=I{6{DK*d@nhEBjyA#T+pX z2JkGnUhl{fEbcG6waX|_CIpTinPiAH8*{3eOtt({JW-?_utoux5mf~7E^gpv-v&># z{uxyAw>x~S1?(eNu#SLwM?|q4CiHX`)F2kZ&P0p(FutgKP@l+zl{187$8MxS*%~h> zpgoc1RTyCPxXaSx5JoTb=HxL=+;oP@bs;s(0@pS;Rv2FJu;m?{y1^HaEMAb zDDr<1MIOGWas> z%d+$v?2M}`%~ibiej7tbE zqRDqP;bY zG_20Hu=i(Vi&zE8mu3^Mg!+m=4SP@o9*vCqgnKtK!cM_0= zb$?mXau)H-0^PI=0ONR0 zz$a#fF{%$m(^FR|&h`fU9ML)iX~je5OoTT=-cr*5Z5gk9KhNoa_e+^;O17?QUd4u;HR#KriPaq2W&$+VaemIqt{eK4n|NG^Y z&$DEU(Ly5oc|}0 zr^;=aLaC&3M-y2-MAxRX0Ej&wfHu9S>oIck{?C`*J-;@?6wISxj9wb8KOcTS{uI+ z?h(s!go^{JN9*J|JRP?#LNlST4d^h2x-+3Uib5Bx)4BD?-xgOK77cV&nc1+Xb{clG z?@FiaFs|J97rA5e@Q4%5)x9y$;d&Qe3IhIN4OT!a4?SzjFxRrj?^x5Uuh-3MwM{ctNj^)z8qm6 zpwA(3QWhWrMn$ciw)#ejk{amuz?XmCD!#g8340pNWk5#^tEDL6A?)j|En3yMn%bjh zsO0no#aulVuRM#jzAFyYZx3VhbZmpGVEK{r^Tlzy^SPWPwvaf+Ikxs;jL9;yQJ3`C zZv+)P&R@%ml)k&-Aq|RPw);(+QOe~UGII8(jy7uo?k{pgglklq--txBu5#OG*ZpD} z&*j0#`UxZMEpA^-Zkx(4zz{1CZNn9ECD*JYbJ_^XIe1Y$mGRs9?mg*crN_g1Qan*B zR+vZ+xRjT68FOXmC|}gdPv!QHhD>dZbx#b>t7pwqjic4RpNQ1VZr=Gur_O>~*e{jm zo!0HBBbo1w8E#0$T4Wt0`o9(p`eI)cCU*Lp@@@%U2-48S#cXAf`0xI}zT1nL-HsU6 zBJYP8%okhPA06@dXR+>lBSg1`b7gN&8wtP{=M>$4>68I~l0&#D#A;>{g`Fa9b zxCL5f{lmONVw7Xqka&OpF??GJ$0`?p&yi;H`~imxBmvY#1-wD z*%Qb8W$-lBeq4iX46i?r*R^k)=2;Ut=Nvn>3sKWie$<+@GRTF9Bt1QMKNC3GrPGl~ z)ikVG;_-iphc&M>2Wa6(t-sGqAEt^#C1|%eLMs5x81w3n4@;V0z|*&cG@@nLL5Eie0l+%eZ+ z07P>c5wYwrw@bGVrkMTOl?PGwIV|rfaNiieI9qv*_L){C`-X$XtXSJ7$^|(3C&2;3 zpC;|#q%hGV5mmVu6Y+SQ39v4o`HU%xZQOW{PrTPi^FlejtAZ2#X-X+}{FPcgc$n4= zIc|UN;;A8eO(k$aqcQ;|SBHk+-GruY^fa#|Vi{b{n3q6w)#eE(9a8Ki9c(Dno_q0y zylMPfg$CRc$b}eDXRZ~neH3-~D-?idj8wk)gY?i%jtd{Ge1q9~lG!Knw$tPy`=aWR zk79?(MY;=9W}m40y3x2LPT|G=B~*u#eg@3mF2dngyItR{YY6?P)lhQNI6qNHq|rEi zo5_b~atceA(Z)2aoFNliFffJt1YAPR{7q?GL%M`TKQudzN%-l_x5m z90T+jj|XjjLns5;6O3F1D@Tz*oeGB=UT+cFH%_}(dNmDu5d0*eY!PQWA8X=#vRoe! z70KgjC7@XS6kRs&I^6G_!LQ6`Sw}cAH0U?DzXHCxt@rAfXGBUj%zc=j#r1YcnqoXm zm2V}M&EM_+lh=-IVM%j5hs8)`XEii)b|)pUr}$8UJPRKnW6k8IsqnH~CRcxLo4FjS z;%#>I9>15_&bV>)6-7`gQ=Qu8;5X(*P@yhv_jH2Y)k`Zh(aQ`>vwK zdh{nXE$irK255~^mbCKQjjO_kU-=1WH$W@>*(C-1&CxQ%VowLscl)jPk`^P zGd=W$p2`TtL`!5Z=Bqv2Jg_!+H~Gae2%rRnZdc_YWR-YI`Md=}Uy#t|{#h&)A zkYY?GBeLUje4C(QXK1_n>k3c3T(uWdjjrLm3)1~u*60~Z7V@FN>)a8yaC9x!H?~Mx zGB>cXvKw~Yb-ZWBN!Kgox#tH$XqfXCq41}7nPeT;@tm(D(z8x5MOY$EuHf_u=YO~~ zUnqt;rL&W7R&t{eX-<|W+zM~;a*b^_t)_Em3Zn&QM4qcMKD%j@zDgQ;Y^WDnU?$1U zkNE1!=4n?!g6_qdlnkj-eR80pY3^0yrspZktS=t@RWfv+910ngIj~=~Ox)asrTxyI zQz1dcj>sO^+i-#?kq04B=3anp^XX{6v=Cw>wCSa52N`dWM)WRTVR6e(bq`Kl?6d0S zl^UBSzTtONkIvd*ju8}m?%yPBQjj%CFJ7j2)hN9}=mn`UzMKhYlhXG4=EP z!hS~&FJLX%-S!d?X=AQ|Pu!m?+MX&P@6E?(JjfQO#tWZ*QYB3YE6jquE3A&qjxJB& z)w__4=Q$09pF96X;PA*<1t+#Eq8qNv4xV3Hp9lIi zgxShcAM< zqHnvvlRJG~m5HJUE)GT~)DE#cue{%CrP-|PsgnrKjC&B>u&?ZIC>WQkMo|N892H8V zo*QtCxUw0NL;q9DKIL#@J73e{_&~dBcrSbYg6S`%iHCf)(!<=-p6tk0FSa@Cc}dW) z*;U@5T%LU-)OORmxz&;cC8{8d9mra<+F9h#|1&mUvx4RCPbL%VD4$9$c&Yj9sTj$; zVe}%#jiq3~)ZW2Jt(aVrrn`H~yvs}-Qyk+I#<6z1MvlFL2$m3A@ZYW;$tm?ue$p59 zLZvSOP)FQ&Ttu9Th`e~#^youo$aIWrC(*K~cP1M@=K0)2%_XZ024PLTSk>f?K&Jnm0)bMBd zK&nl;zsfNfhhpXjwK7fSonH0LtFiO74YOkN4On1YSh4Wt^TBBUd*U=FMiy-BsYLAg zmwx>*KSQCDB87E2^loqD?OZ9)(_YK_i@R+J5LbZlbQDYquFw%s{GspH*-%x_-y62C zbGm34=dhfN(zNgLqa_+P#pOC59CWaFmD%)pzw3K#EI%e#?FVZGN;J~A(b;7QVP@A^ zujHHR`o3;6+@Si=7E2O>)aP6pdRL^yjZE!wK8|c+!y7o)oJLsHjNO zignGH<@p{NlnL3%>Nrc%E&jy@`F+Gg{wi^(-8^%DTBNCqz*w}w@VbJB&4;x0x;bdl zb}l9K!3tPx@Z0wZ>_2YeK1R&aEq;SP$R2%DQIR3K!=gdbGU8qx?~IEu7c&bltLIQH zT9Q2Rwj>dL=VPy&lmub2h9?6~P`+mPua{T`$kLUqDG_5NdU#f3JUO&fY3PRP{48An z6yU6k|6>&FwMFX__x_MboaV#ez~QmhK@R>Y>RmG6v(>N#tg+U7M5FIPT=oc!Tj2j$ zq;(EckWE$Kxreue^5OYhyU`exfgMhVpz5i&DQVqdZ$kMUy5$O`ROkz2{E-b_l< zPC}XIm3FCUVNZd+$Lzf^&k=rm6^Ik`Cx!MSRZCY*xC2;RaE8mf8=nCte8jgsDhG8q zAr|#pBEJ{N-_&C?y37*p@gdqKQEIY#@RgUBvrk68JYTOBk2M zR_ZR`$j66FC%kVYCrAuuE1ydr*`@(R_Xwm^&=>&`f;cbxQavBCOysBfb=&F#rV1w) zJp8yn#jUDA258rLS$dP6lWg+OO1NZ?7oo*YiOjrOJ?KQKj<6d0vXM%`!_PJI!dYY# zU7XZjv|OO@^y$!mt(Tsj^D}|Z*wbF^#nt-cMm|OC^UUB~)t#+UC z@@pUaBAqz&O2bVR=M|r!F|f7ZgZ zv!P~)QJr4CwVppO2P3BlrTtnPEW1Jk01nGhpL?@0DB~deV1agGGHlV`q%g3u`R_O@ zNfg?2Io>;)ggZ(;Msl>ei5;9M2cDn7E}-OF7UTM|M|?>9%&{4oiym=kfX-yjq-HO1 zi>PfUqOL8#JPhzhA_+@fgclzVQODPuJ6}H@^8f66PwFQ3Q_%$Bhx?w)F2;*X)RU*P zjlA|7E7LB2Uc1`!_KMQ_Ki=6*ib5BZA@Quqd$4}C%@jK=Qx<9te8uWE>m`3r z#jbP+)rm#`c7}84{T^8a3nXeVg75r~$pO|xLgnZDp`)MJq5D%3kAo7#kuYUg&uAyEHC zvf&xgz(xoqc*)}-e*^pQ4=3`3rw}?#LHQ1SHxKr5nTOMMo=+r;YPPY zD|F|W4L6yLBtd_uL%;xowm$LGbY9+1=tcIiDFddnNkaSlIojEVU1sakwXZ848*a+$ zR2gUE7?fY7pqmBmzkl@TlM#sFojat1m-dRc#*|I%;A6W#f7j+!S)eGr(nJmcZR_tC z3(jZm?KvfyUJOd7gi;p_>kCO4U*1#s@X(*WY%ygxrzl!>r)TFbC9suuVndp2hOHe= zN)eMLYsE52g%-PRzg;7Jp3`yoL!tP6p{Kh26WLh1^=sex!=zYK=pX$nx<6qcw~BVz zeHES%r}KutocF={2?5(jp&bq4$rNEyB!c0W;Vf&zR;A?*PSTnhO5!|9w=$##_35wI z1Qc4j+G>SlvEzHm)@d#}J>e{0E@zhP#}nY=Ed6~2iU?WfBgP3^Ic-}4tA*k1L~iX? z?M=ChLu|RTCQx^sDH{;Oy!~6_)hox++JPm`;}vZ1qa5@n*5YSgh5|-k7)aw|-bq_b zBhQ%j%Y@`x{~DC+4>6)&i$6IYGM0Qbly@#sZtJ%fHY$M|IK6nc!b)EH#PBDAdSe(P}Q zD@<^Z**{e1l=D^w$A%p;#dmE42nT{_fG-MZJcFLndR$S=oXw0&!k_&c{WiKxZvb)A z8}IX$Y`|F4VNJc5{Gz&s|45r?#Q&8xhq6m=4+;?l&>%BSUw9aMG9(hA8kjHKzB>4SGB0L|rHuXrcuzmc3I3qiIst!kFfKVT3eJj4 zBV!4z0y!b`D`8*X)$E%>RR|>NRj*aLBZtZWCni06AGU0?Sh9~Su4u#co~;Khy_({) zXifUoP$pULdhYy>DgU3={aq73C@3$PE1UfHt#C=FsH*$GG5%v8%*fF*WP0kji$_#% z7VO*YR;5hVi^~kR3Ei?kddSb!-a7@NZnG zZbzStZ9hSk1UPxaqV=Rq3C!L;xqq7Obs*NYL92GmK|jz>PgyZ_J+Ai2d&la>GJSqc zrkye1Bx`=NCk&Y|l?eK-Z6g;qxr~^(gw3qTi9Dg<>4Nog`-6UBx9|F>V`(bgLvKel`s+B!C6wxzj z5aYxY%K5S6Q07SH5;4*!awuybvvLcCmQ1G1C8Lff9~w7EM@T-d%aCJ3b@Tn--0V^` zz)fuGiH&@RfDJ;2_7%uF;b=0M&qIGPwiO^}0XI-cM9it8XFQJ#nmd&~Xa1n#$98`8 zW^82+&5IPe<({`J4?P*o>9X2L(&1q0Q+Iwb_U2y0dJgeXln*CSpQDSw42fJ5w4#l_ zw86G8MO;#kf=QYNik=tk@mujuQ8)T0amw!5cceo~M*I+82URy6plYu%L6sCVT zf4D$o6|s3}ke-xiRUldy1ZFtcXbx&!ODwr9DxzvuvzFVlgUe;_?)HG1yXiJ3!L0xW zL0@6c$!XgY!l|IyqVcHd^7=BzV_qIZer0=2igvMqiDwR~)LpsY;M|2)y8e zn^~V-J(!Qkp_3Y#SNiTfNsv#0*+yRVsxR8d>>mDmwn>? z&$k*t2LH=vHR2e7MO4Iw5lgeH{K0&Zsj2h>%E;B6nxqt@S9((2;eBagY{Zr*T!JNY zi|6$@Q#xLUSMjqXrKd!7MhyO~0Z^_W%0SP(v$?3PPHqM~8kSbfU8WpbxW%Pd4ol)b z6&^?k`_WX%^5|!Pb0!k!J+kerOz$;uF2ob{ZJ?!$kd%TW#(#+seYl?HZg-bzu`>1J zIGIRFGMa{)*RQf?GUmfzG0j`N$;T$~SGYUsk$MZu{9%TirX3o)u=3Zvj&&?ZARze2 zkk(aXB?T*i85weF8PU0mt~Ja-qc8fJ44Msjp>kZEtn(Bt2ms3YL48ul`xgI8_B&GS z`z=@(tTd(3wP$s=-P`Fs(lAXyTo(ghg_Iq{i^7VReVqFyG-_>kcZsobG{TOPz;`HS z=(Yb|2cmTUP{G8sS?h{-Drx9axMX$yG`sqqBRJNc_n)5@J@T_j!IE8JU?A4Zg;)p` zXQFqHs>m0A*HTKPgtjwT{j9IQuJtsu?_Trv3x9Wf$u>PPB#iQJbcPt^G@!@yL5IYC z5AeQ5FFd9Q3-pcSClUJXyf`WybL35~gc;BQc-Lzf;(62o-`HKr|IM&s6YcA|bdu=A z$APfHdw>Dn@qH8wWL(jyFa~vep9FtQ?KEVOGLn`yMw$maa(*GxGGf`W zxd{_a{X|3hYr|O>)zJHBQl#RcpwJrOxAGQ5Sk~@6JO%!JI$%qq-2tBh0s}v@m6&x( zcAZI3?wbT=D+uwiY2x3)_eNAUl~u>IqwH*Esfs5w(2TN?2jPd5 z9UsZq<+sWYw#N$5CGlDv%%OyS3y;D;Tw*`Te11fr+s~JIhx~&3YZ*EuIk));O%mmp z2lD!8aybLkMo>sy2m>wA4}C8MjxYLiYXT!TvS6bqBQi-5Tzy}v(WcfpMb*amdHcwzC@elU@ z{~cwg-&T5prygNf^m)2Nsvp&;mi<a+jgJs2!ut!6$uD$lM#E^WqAg9NdWCQ2b(KRRKy6b>)U#p&^Ig z(X7%eW*HqWeW8}Bbv5*Ps}Su~3|b)@XpyCtC27{SG)XXt+e5&{PG|P~)kN;QvQ(|6 z^nPj&WaWSfZ3Q(V%heBMJ>nl)0Pf>ux$XL_*tQqi06K;eX{GaX*0p1e&+Uqntjko@ zEb4xL!;?JpT}9zG+MkLG&{-+U|Ah69Wim?KLN8S}OOl|mNT_@-^&8&4eM0rZpm0kQ zR|?`ZeTCWlSJd0E*Fwlz!{F1R!L&FAO^8$?gPKUKa!6%o{x4A5rwM$=>Xac5{4TbH z=GWeq(Nf!w=VT@{S90I|ux6{US4=x2hqfEyIB|ID%;r_sC0mw_QHq!(VI^1q>vDc>Oobx|pHfmaAc zVd`dN4|{0o?(@j+ql+76fRRuwHFN$srI;y*-(|?(Y&vD<4lHTjWl=W3XnG@3f03T1 z|CS|-5}JORdcel1{g{z}xLLLsEk*MvpwB(V>*^W@MDGO#_JFm#Qc$4{Efl8*!adoLqr$6r(wJWYWCV#) zo;sTXI?GetmG7w6ak!c<5q>>N+%CP5&G^{AQgGMHreqnvj;zYXzrmvaxpK4#+mL<8 zs>gPTRlN{#d>{umExD{4l&+AEuS`gLd$@5}YQ>m?DZBnC@PvJ^9&(5l+( zZwhLU+*nX18oa~R@6vFS9E(2QcW$?EknUk)F0yY%AYnR^t#7WC866 zLPx3AWhnv|5MKlFe~4L}ZoycC>>1+)>jrZU{@VTK+2*$r9f4jQSMI`IZfwF!_>fu> z&~uhrLbEP{`5CA1IQ~xRv)~ECy5xlgde@iK13{UOAF$>6MKl2G!Yf-YlSlC}y$7*` zR9Rd3F3c^yH#up5=~nGe^Rfr90Nm z(m&*+SCJIYk)nZEosgfsoafu&UstgnaLhCcHDX192!c-1qEYSQ0zr<771e>D<>$EqF1-m(78sowI94Qj8I9^AVK^09bRnFo_gY zDdOv{lzDV%-Vf9mkQ)9L!0nFI4(~Cqq+RLbm37%-F4BiNngZ<+Bs;L1bUf@KwPFCU znt5t|RqLOw!@X5p5D#BGit-Anop{e2yrUbr;=`U zhQ_G{ZVM*mRdkYzHMQ#huL!keT5dE3Vqe7$U`^&n^t^_do0vFUW7HWEgKfb;A@}2& zI}O9sznf|gM}{s^HkPWgmK)>l3>2GeO2BC>D`zAULTC=M86YRJG(oU?@kPY-x2ySB z5gaL!ZfwBlU<^fu4^B7xk%M>FDm!Q|V zDhxr=Zf(M(DLxn=l?JjoJCt*8)e`_v2e?>VA}QlR{{UQ#lTq~57Kgdq8`md0$G1m?cP>N;1f zK#P^fpGk$^t$m^Vpvyz|Te`IcA6)j&Ukk2G+;c9ZW@8s;BlZXdCyvI9ow3+pFv1vo zigc2T`|!wYY#yZgpEZwDx$$Yez&qP#ImqHv__E9bX8nm#=1$5r8^z@|0mOEx-5iC)b{y=Hq# zP4Q*|WTK76od@w0?{V90xe&o5!i6hsUd-(D*$PIqp;hgF$*)CYKp}d!V;g@96-&+v z6{~x{klh3Q!CCF4;#p*IZ9sb_gtGdqdtLGMBhL)ogL>4^4lIo77svH@^vUv3(~1a0 zCCziq#9jJFWda^m+q%^zwEvv)sS$%xJ65wF@LlP`SB`u;(>ygs6li&nlWW7JmMxv~ zY_!5GCbLe2-+fiqm3%t(vI>VXx@g`cSel9Q$L~OV*GIvnWJe4r-5-mq<{#z5d5UdL-Drms{gE>UZ+-J zS5V|3a-Rg@Peo9%aiKH0#8P;5wiZGPJ|-x5lwuN0q@s)FZ@~h-HTbMGjtk?LX;P<+`=PWln06*HeBeqmEA`P-zC~`2@Q!wR)Mdq(jZ(KRq0ib00tDtXZ0qm z^@(C$RE7HY0~n3*3Bvm6vh=vw$;yO`5S?N%P@}<*YplQmia4aTEK1*_8<;rLf2Es* z6FN$@^o#Sih)7T^|Mb9qxHap{@(b0XzY1u^Ckzx*XER|{>gQ-D-6l)Ro6hAi7bN|z z(x@X)L#CZs-V+$&>2mkVCavWW;*DJYx@*f|yXN^dVX!aBsndxOI<7H?{av-pNn{!6 z`8JH7gyP`0{P-x?xQzKG>ihNFA~`+xq>xXF=vxK5g;f8(1wOaJE}GmnHKx3dZ!`>q zY)?KtpO@nPkQ5X zcMm4VpiBKbk&gIqKd3M7WWLh&vhud~3}>KNYI1dWwP(<0m$nJPFO8i;E+n(|1Rg?p z`$)$k5^WZLXdv&ZTr?d)#?ZD{?mK|^MkI~00=@`pWqZ{2o&Gfw7nUlTVfPT!3%|~4 zSMTrD z@x^+$*10l?(uD|#?oeNNh`4QEfJ>E|yuE=lJQ)N8wEc^}jCg0g3{rS(CVjxz+x_dF zxd3%R^*+HOYwv9Dmxh_djov?=|kLcLd=m> ztNo=4e4QAm1Hvb=!PKt0l93unIO}@C;GCYJ!RAcB0Xh=?>c5n}uFl`#4Cl;D6ZKUf zq9jQppvsTirYK8)yn##fi$v-TYr&)2F_0+Z&r~~l6ZBf4?Ou>ivnZ5lDP5SzTVnQS zN8I)rVmEr9eK<=hA*m#8n+Y1sNB_zG;ju>@!R;N#rk=aF(n#}<{V};K%61LcVcaVJ z(2N-c@(f?N0n3UXyAA&B7ecMP*KqO^AC;@Py)#mHG?idD>H9Z{Aq1!>II^+PAW;a= zZWpEIVab{CQ7C&XAEVBeEdElpLjaZqOJaJJ@E_G^t_qO-6qs*%&=tqM- zKCVEB*8I(p9b2j^@NW}H!BtuF&e)Y+=)HbDeTXj&SR~dpTU(^Hv^-(4&&*WQf(%LhSl8SP zxB9F4C^J`C_Z`rSN(e(pP()3+As^`6(cnsinz9V26kFkdU-+EOvRR);ZhWaP>YEND z@6jxie^grHS^N>&wnnr{oSj^k8AU`RLH&FQg|tXsWA=k^qLoUNL#DyH@@iYo1&P)0 zZiOgXiM+BeYqDFq!84pA23$&iX9&VCp#7vcu*)Xy*qc;IA8?5X%@1*K%=kAOR1F-G z=-Zi16qE z-6-ch7^e6GWxzL1SIQkKuU6oQ5_;C&m%YS90nlc?(2)QEF)3fzaC&=gi+j5YAKK& zu9j=I?8!yreo=gp#F$DraPnG}!o)j$GpjC5vUgw@e6>^ghy%$?%<;LBh~pwozy=yj z^jqRpjk(Gco!d8q*7yR+g$AGitK|PLLNUJI*zd5vJ-2_DVnA7$!(3-@?S7gep^^xc z>6TSj+VWvgx};S@L+mpR;b)MCY9LJIglKEpwt0t$$+AK?h*plGw}V21WVDWtZJ>8| zQWmU=k6S?xn-&bD&KyEOjV~9@7u6fSsjchg(8MX)@*TBVH{=0qtRWUhg_A&;2({oz z&+$IIumhp*1JxY?=)op&%QrcD0b^@Ab~&4nALy%kvPr2nMN(f} z=hOUu9DRzXCM{(c{@AonKlR9$2U#7{N8=Y}do|dEnln3m;r~81EtTRRzgIC;JmiKn zjTdprutOlC;`GQAiR;fm?@aD%I^8ZqKw9^^y6|RKr~aa8m&fC z;43+8-A?_65>Ri#m>ahIMR7J zhDRU~{wHaaZbV@o=aHmS4X?^mTDJ)Uw3$rON>lu)8^W(Ouwg~!$r+rBGnzi+?thcv$3?psk$-YtJzVjUcp*8jK zs32R!%j3IG-j&lz#DSOkusBA>dN4LFf!J5}kkc)6!8{s{PtD%5nTPYc58G@s1ZvUj z9#NjBjzwQ8b07}$6MkDG3O!J&T1yiiu{DyOkh$%5E+IQ*^6;e3d?|II!ao2zQm*-Q zQF;C+RxfANHC!tj@=nt1xr_>|fINei?KLK56A}1+ zgls8IV`bd6=?!XT2Qf2JxPv0J|Cq%~`yD%sQfg?SOYJ$K0d?giOWt{P^Yv@#+lq?- z|M83&mhkMM0jmv5ZVFFNkCqtG)!V8q79JhVH;myu(X8J+GwMm3=6fy09g*xa(d>MC z{6O%Y=*9>no5&hx7N5KY4|U3m{~Q%o&_B%!&yX)JGQ_k86CCC!uIV(0eL0W_cW+GD zliJfk)1&1|i7bO$7qQEW~$jE~4N`w_=Wh4zfL#OkMGO)$C=Q#lcC_=i_T2Haqb zDARZq7y|>>s_HDzR^_2&npAI=q;*21A@VYF6BA3E#!2RdW?*gxo!0AU!}#9`OtgZZ zRlUFViB%vRNXrHXCU}6{e{;++8?4mD34p>@piK2wq@0Ks)uzumO8Ui!AK*SHP|XY2 zIb{Rn=T?=AaapYPe_OUh%VIR70R4RRq3KLxQGZY5guGW-Rs?_fr^`K!3@fa)`kc}) zV2!wK&1SKQj0^q$%AR#|;<=m=XVrh_;-^PjaB3TUD(b(TZGRuHo>?*=<0NAo>GvXd zoEPr=H8r+mAlT{o?l|M~ID6u1BHT3*ZNNk4h07+1B*^H|BCXc2uDqvtRVI^1y>V1t zQ+0afP|51iPDflb_==faPv5#VWOUT(=eNW91?ZK;#Za~}mJVB_Q8uY83!Bj|v8)Nthh zP=kj487f>iZl65CG@iWPvLx##4ME%3v9ZOrjrUIrw!dvPV+is~+h_5GRR|xVkEAT} z@msJBs-L82pPFR@A1?e{YtZrwJE=Ow$Y`%fN2sK{@;e_M7jP?zOoEfor(df5ikrfr z4o@y_!o zsvYJL)R*+Jii9A5HiWE3%5T$25KV6Vow?sjZ*ohWEZ|?TD%~ZJG^^=Uo}zSXH1LZ< z#W$V}{*v&;8~}fQ&*E8U90fW2_3UnM=6iO|1|eX4nTK@G@_-oNxW7L<+lnqNQff4q z?i~!<%V$6~E;%8g$!~-MJ@ zph)KjNV#A8KQAR|n&KVuMdfV4UbWV77S|D5_Iod3fT$-9Xxo$yKT%KE`^}3yWCV08 zlb{+p8~>C)`b7Ov=Ssif4_O-I8y1eKg!a85)-ieiv~APqw=p?p!cIFHW_(Z|XZYr` zmkT<DZ&Jb_*4g}XLc4wv$-m#8{bYT z=7yhYXdoPF<6>=pb1Vf*ow~!S{OzIlz&&!Ve?9&CN}p-J*dCNAf+gN~cDSmg2B*3f z#Fj^25`UfL7kB)O{;}-un>bS*S3@~PfPc-Hkk;G9{2i+>3z^!|cxLMU!{hJCc(lAS zTe^s&NF%Uo?nRb@S!uLmF`FBkj|DUpePftGwoK+hfEW3nd9Bm z|I-x?p83^RH#43~{ZR?PzkQj*Z*u2Sl|hC6Uq459`%(|M@R`+|&DciEf*Ss9xo5bI zAN}-Sp`blbTjK|1-F>xcwkETD62Hm$%*nY-+Kc2(0#n3SX^S4>cOA-)>35&=p zQazE^Ip2?ZdM49Y&u_Q^1W@%+M=bZ-e$h8Cwan){BBK9dN6k?3rB0r+Gi?!G6^>F6 zadunT4@|{>;4kWc~@C;Ub}{B(#K}*49@&vvxiW$1!4Eo3v9>B;z-G5OsVY9T0j%uyzTy zk%{g&)!wp2WStKRXS(#va!5SFJpj1mxKs0sBuY16dR*-*=9(vPuyop3>2?{*IC760 zf~;=yuV&-{_J#HH7<=CK@3it0bKT10O)1}GGP*uAidcotVp1UQmrsLP4RycttAO6XL^Dqi&9pKemYT2Qi>scxODaEfiRy6-xqKht)p=V8LZhV4WpRz zS-#GwZU{f|ALc)#qWl2&#Ja}u&7B3koojDyl~el2!{tpLw`No01~GvOpv3^30M9a! z?kkJ2iKt(XTfxY2*>ZZMNs#y~u^v`D=`<*k@ok&jn0fL$^X$w!=wP5j-4oanr@7r& z3L3Yx;d6%*=U{KjD*M(fh1d@0pqVh!fzatdj&5LZ+WNNJG{_luceT7aZ~`2ck}>#{ z;*&=K(JwkhT8eCXY&L1bJ9mKV(<=2CNCu_{-t#L3DKl_+i9&lU7R1hV>M2E_C7ByN z8Q&+|BB$6DgJ}?Q8QtjUjFaHvCq&F0?Xd)XC7nA!QL@4d8T;k%U{g)R>%<|GzA;BW z4NAk;31Jkf!mx3Srhsw!pU-4Yi@3?GC@Jw?u}aK);mNk~we+fe#sSh@25WL0PxWV? zyL@}lZ$oB5vXJD!Qm(bevVVgQd+@L{JZPRPMEkEDDuk}|JUIT54e`(7 z+pmcbCfADxVYVe>|5)E>jYyJt^rM>)sDuuK`Jn6!5o5WrZ zt-VPT)w0~mS&SnmXS^caT|>DyYn`UX_)bxCp!{h^9QMOkYI^EQ{j>rQye}kiq zhc&9DXAf&UC|w|A=W%{_@%qgXB#rKvBGx3=;2f-S`dT%l0b>b8_0o6f^+HTHCnIvq ztE{Z=|Lg*{T!YqFU}HGre>7M2$UTF&=bmkCno0cE9{JyQ5&9ouKWaPczQ^G2Kg+>R zTmp6ZjYj<76C8i+hx6CeXfMqPS^H9hd~NVRDe#;QqtMF~Q2ova)NO*=hahl0L3^q^ zlUpHBjicW{2$VPj;zRgxM5oXs33No7%A zSv0M9sf!jQoYl~Q$=*!6h}iMJb3|pkjs~EOfTV1`{^pS=XwX>&?@-+{s+7abb(WdZ zTP13_XFy1O%fY-4oT{n2lBj#x&X`hah)yW2@ZpO$URtO4VZ0o@P)OIUc4i>UFh~r^ ztxUsAJa0OUGGXE{RMef?0IaLhI_wI|B9yt@^T%;^dsx#zr4+<_x5fglImkPZqzp>@ zDNv7MUIln>=xw4PJ-KID7v`$6R&JH8ihJi(~rYy*z zs}}i5Jy#-3r{{P0+T%6ZnL-WJOabQ?*Q=jP5w=T|_OI8JET{2AqVllwcA>v+RRq$m z7VYNG3fva=WTc1qsMy9rhsL7|V-88?l;o5Wc->RC9S0f%=2NxK(Aw&sqOtcnDUnGy z@zrV7Y_|?~fa`cQ$tUX^Wm=u5RF#au}JDN2>XiUBuv2S(<^b<6Pg6@7M9%UdtB z?2Gg#Y0ZBLrl61L5_vMmLUnCv-5ECCKG|qlQ+uO7FawT6-V0xna1(wSn~o;LOKA`Y z-s#tUO*1et9=!!4iLu0cX!)>SeEAbgJdQz+N>0WXOHR2tKL;D8(9@9lS-WMPd#qXj z!2Cb9XRcOELb1=Sn`8z8*?Yn{8b1bm6LV!J%K60VnV*#xRP5T;6r*9& zFXLo^apSl3%yE93{5UpLT;5S)?a^mH&+;HmVviUm`__%;w>DIIrM9&gR`mUL`$hP< zE8a`haElc9Rqprmb@4QBdmZu}<_GDYLeKI3V%|&WFC2C7S4*Uc1frFu8(7DMZ2rx; zT!SSlST*E6i~K}w&ht}*YO8b8)x+_t@AgbcYTO}U$R+zU_3{14ZHvi!o>G0|XOSXJ zoJqeDEt7nHt4z+m{QU2prNe@VGD?wI4AFY?ZCrBDG*q-6wQwlLUAUrhCSHy*K*|cH z@K)^t^qY z#6<|Q%g0@IRqynRS_8w~|Dx)z!{TV3?*SAI?hpv>9xS-KJHg%E2^v`3-F?vz+#Lc0 zhXf7oZowt!?w!y3y}#!^_rKkmn%UpYeQebaF(3K}OP*b@tUF45HEk!`^A0~| z-I;I?LlA}!&h~-I@wrGi38CX;%{#&hFVdnNfk>75WP00&Q75J8_S(3KLMl@zMJRuW zkCF|+o7RJZ7xx*#4_oj4cQu>_y@%;=920%F(yL(Fu292hSbwQY674j$kq?a<_Gd!p zyZ9pc{i$ezlbyM+{#>UoMcA1_-?>K?n4Ea!TmD&lfp^ zDkEZI;ReXt2NY9Y&rzM@zI|GZTEWIjfOGh)?3Rd7$M|E>q+KEmY-{KEqhM){C_lp@ z!YWp{o<7vAJQFDr9x+>hY2g>m?)aCwA$BNimon7c^QeD5$}RF5OA7%|a!^O@=>rvG z)>chfzYV&2UA|XRWP{1poQNFBAkc?im30gfn$t(dy=*dw7{FzXz4&u~)cJIN zcHNZ4g=+vh@s|rqtICk|AZa`z*7Ll&^Wi5oYEM`|wMVz0rgu8T{A6-aHpOB(q9Eh~ zR}2w~Z-eJgx%{lHz~k;oZI07%kh#zBJv;PIX6KNG#ovrEW4iTZL746CF>yJXm^rb% zZSW}ketLmOh8C7C;-c6*M1C6nI1sibfL;R<^7N4(rJ?o87+)T^T}{VmX{h)=GyxR1 z8f<5T7^&?^r6@&R?4CiID-Uyu-fVFEs8yD{54Q5I(rY5>55;)h`1C^Rj_>;vvT#E? zf(=4IDzseWvggO(fS8=c`5nul!Uw1JCSXTmgU7vrRat9}TO6#nulj)%n74OC>a)Ul zg9(rOCxBAeVykPVYf|j^>z`oz3%GpgLCdAx?Jg~6d=lBE28it39VG*q}l!9~j4{pea?26wQoOCD58q+|6 zSQLR!!3hFZ?`JZgA+A>~D;X%N0Q0k~Q`^A8#te-Zgruw)X`B@U3izA}s{MmaMsOPUJ6cjE5hNX^!! zF-kSSM38+;DGR@uHvYX}DcPkEdPH>bBLAT$SE&tz!K0lL4vFWaQJ)Y4pMRF|QMX1k z7#Bt8Rdu5qNeHD*o5*&={-Yes+LltbH}cOxYS8Y6#I8anB!?ozWwj$Se+5D9bBZ{q z#fU~-mu9@*vr3`W&7@($mS%jts3(Q$FmmX~W;=r@l}L^)gm)3aCJEtqAN+|1`OQ!F zCTU&XLyljS{eccv@Uen)RBTE%)mWe-2h&4zUok#wBt{n|9k(b9g!y~4*$ev_$JF90 zi%AbkQ92CSt0yzoR$05tXa>h(`Q);R|SH zRDZk>eQ6resZS*?1ro8{9)|=1aUN*vkE$RzwHdJ$N2TZe!eDxtvL!c*j*_&Qo?-} z>jsa6kU5tjz9t^!FU7G@sKQR`Z;L&3QxpWPiX0#rIdOJ*KGHkH$ENAXGWDf-apvvJc!@kE?31* zU_zY;9YOlC`5LMgc)e$`2Qv(#?+=nwrfQsqy#&>Mz5{lI^I{zi0>b7{-6rDfA`%)#iXwOdR%fWB?8Ze@* zBkjyD#{AGrV}2g`=u5Tm&*h89X6M8|W#E$?K58?SY8PLxcNh~|7h_GOwuNj!n5CB8 zUqX{qG2g3EH@MbX2lCZ#Y*sB@P18~iJ{fUkG==pN_;i;dbt zt+THkfyTd3(q%fW_x@-SVhXuHCCS^dUX^)j$wj9cz0V|M?X%(c;C+6hG&UMoCRc8` zNh)|qwmN7A`jWsX}ooB{e8jEts%R_;@<1cqgP&Y@+{i ztxK!w_}cwfCR^6ea`9#E;IQ95LD%s2NSc>=L)SZlKd64SBoO9$5=gINpjGF$V1>JH zub=oOx-jmJxPPs4rmEdt=t>s?=DOyYBc7-(6X=OnEu2-)<*OFbe6c_)e_eO-Rz%Tf+{421mW%=-!E&P zm;~*I(HP0QL3acp`ac(M90rXp#|=th8|6G*j|`k>Q$YH=b!7782bLl`dfO7s4?v?S z{@z~71GPm)w!zcD5@ad5-iM0jR+eFbRl)Pa_d)&38_`8&`_E4ZS%D4->7XAo(!8Gu zBNVTAB6;H%m-j~%zx-r^MytS9oD@Yb5J!o=URWOIpt_ctDG5i{EOyu|Z%@xQVKPct z!alnkK2~g1X5?1O@tok`I{kD5JBmTtXnSbi^vjkp+`fpEH5=@`XkmGU5A>R$psK3D zHDcMe6BWixIZGWNpb5V-YaRK%gT?N66NxONfxAAyQHir;=fOekldeFD(veLaoV7^A z(J%OX=E11q%xMa<-7;dD3n{YilfW>?%}Lzh2X5&TrSL!Vg4TkS%9npk*c#^*zD(1;WnNvr&YM~H%s<81= ztybh9Z&`VV-6@;Kd0C&H`#V^)YK&BI#*@FcafIrR5sYTa5;2l&gbiHHe#=O5>s^8P8gLJS5bUgB@yhIi4fd9W^+?8?xrneKO;^P@Wy>{hyIyNWUJgV+lj4bp+5` zoII&DK8s|lZIlSjpyp0n=&_P7V9UEg^LLzbx%Rdxxn~OoX35WvudizBp-l;n8#CoR zbt+AwaAr2Olg)eYjDLDEuElRRAwCM zv8rRfNdCI)rgz+1(}o3nJx{J|>gZS=hMC*(GwYYE%1{q&fEWu*oDuX%&Y?E-!Ta}-T?Z_%=n%xh0 z^@&wNAGM$Md11fVYG(&l#n9&C&AjykY%sm;65u9Rys$LdSHz7vxwhEU#&;5S*gG8@u3nn1K!4>(G1A)tfpFvcRaZ$^9o5WhMXGE(TLsQD0uorl zwPdk3YCe{Oqb}1I3s;hM1S1pKYB`aWWyeL9eC!g?REkSszE0hf8SJU`(GU zX?gycP|Jvy5HX85-msDBBJtV>XQ_~e#Y5}CccyZaCsB#HDDA4gh5}zM(m13GrwWV$}}xAPE;~< zcc*=Achv^_lhMmcra4YvjP`*hntmh~ALK<{pWXjU~z~EXf$x`{v@9uoJ(V68!(6QMLoU~#U*X6hHH0RZ{m z@>}d*Ng1o$wz`>p0Hx*miO}b^2i6*9XFKtZPRNKXUy8IJtKa)hPgr_Je+?_h3e8;^ zk;vERi60mNx6{%LP+`;_f1S&?*7k&qT9Llq#2?kck49I-?t(0c#l2b!8<^j<2+l*e zpfxdR$Lhchry85uoH^w&ejdalgIrSmpt*D5vh&^D^Kr+j#Sggc|1Hb6J4Ok>Z<$c{S;$+K z7z3%?_vA0m{PtLn;h6oUJedZ03_cu32IlCfi7E)Yi^NKSWb;Vzn__6z_8&!Se@F53 zsyd8U;|e4cFW2%2+Sb0v%2}P1$iHj&kqBIEcjpJ_w~9Vc4(*q zt9(tXzOng1g`Fe;w~{jsJFJjn`Kzbd`>92V?-unMg{=<>g!S+o0SPxXJcO6?{~c~- zmFU4JT6gsNWiahJ`c5RA`GNRA-_g889x6W3D~kkO7fyURrjl!1Vw^TOqm4^~O%F)G zg!uX59nA}|{O1eBzeenzd;~&?Ku(P;TLAL38&s14M?EU0ou@YX=0#_=_jq*w?t0M? z$cp@NXh(Pb)_(YCd$x_>NCh)0mZ=srDk!oHRF>%ComGTQ*#p1ph5jtKH+%TYBlvWr zk=jZm{@<@7nGQ~#6o>>QOzAt8;-7iK)Df~-|M5f@oDkDtTctA#enf&p>s|tI08MI+ zw@YEt&$p`lrlUV?ehO)bci3g!NaXBpA){vZ%1(%V_$lMh*}bG!d+wvZi?}c1SU@*z z;0HnZz;=t(3^ZZey-jm{wfu;)O9P0RJCbp}&Z?LT^*&k$B=875cs&dL*9$SYBO2Vh zrN+$Ysinr*LxVD)sQ_R`d%BAsxT(oF0{`HGHq5d~$c9Z14fdu50;+F?BJu0uo4sy% zkf&ZZDA_Cbn^En^IHo+_r)&B~*})^S#1?%>`p zCb7h@A~oARB9%MQTK*aNQ|nf`cUxOCG6! zAmip#ip<5`sqC#9kRk`rmnJGkd5sVNT0Ry`3hXFYDr4bzjx$ULe?-sPJRD3}s9Usz zyIb{ASw-R91i~*WRh@D4qGmq9q93X=(w)GOe5jTxnWORwaYs&IwApD*neS+EM;5H^CL(d(>yd$iSeLyN%5 z^wfHRPh=HCpjoMW>zxchoJBCfn53JEwDxvTgMJPzF!>wcaEXrcyhqYn-Yu9#cj8d{ z-YA|9*tb_Gt>kxbGzw&q1?_YF8dUG})4K6D?SS3#Ud_8hOQ9xL{sK<97EdSHRKkk*9g_mM18J1QiDE0?>H%?zDxGH=e`;3B z#hbLt1G{&R%Dy7?F;0{lacIPMK*%26z5eu1GE zT9=sf(S8Irx~+_9XDT~<`t2gnm{*1Uwog`?d%%UB%9VeGXzprq@! z#}`)0k@M)hYts3nyI3@6PWBEj7-gS{*BLFvzVe!VUw6p4AVanr(Hzp zV)%Y3ca{ZVW$b@H>$+6lpQ`HTG0x7v*4GZ!RIvD4NA*2p|Gf*}UN!K#oi^(I_@QjB zR)!B6|MlI!6Py8=IW|aVZF>cbZv^d+wm{apDtMOWR^Z*9vo+NNYNaoihoa2iP1FwU9KZV~g(@}yEYoCRikE594)v1Aix=EIc~*|Gqnvf|th zR7#>OJ`jfs%TL$n`XW)v1T8YE9frUJYvGcVgJ*X!?f>{TD?qu5GsR`w7oy;b8<{s0v(Ythv4vB`gK1W?26L8q7r`7MyTWwa?w z8)>NUiyZC<``8LlEtRtVYUpS)H*)l628ni7szNvwyg}|l8ZpHxzx65?dY4*hk@XQq z8D%H9PDl>z5{MhTDf8T881-jjvpFqnWh?c1YXpRcBspX@)Z|6Y=>cm zQn!R(PDH#i;rt2&N-Sk3GqDqW{Z%^b;8M91wc$cw58#}Hp?Atz^CR85nL*Xk%eX%` zSgQY~&X4TTYbHOCkB?5$JWU%`YaUg8YorTbbfyrT9|=HUJH@v$9Ld1m97~@ciKwzD zjBumu{c+*PozSEL@$xvJ>?tQ2pM1S-ZkUt@kN>FZkywrcU3>t5>_p2}byI7Qc~H0$ zR9{bJHD7u;<#pgA(+WcMm*iYWSxY#LmNkgcgcyO>gI=;Xq$tR?NAMn7S5i+*wM*Hw zFl=HKig`;sp4H8qv~SCoQ3paOO!-=w+<{IpD>1OT%IFHSdekPCbu-OVkRv(O>3r(( zrNzX4FDGVBBpAGj`N^k>(cONpz$6f+f-$MUo=xNw3-!el5YiaDy-)69BHwr=@w=Y( z&jA&RYMKf&mOUpLu;Q6>FQ3>4t+RwwD#8EFPbPZ1hgl9Qs{IK8;O#WUV=&(I4glpn z5}iNPEaX2UAr%Aa-&vzI#h=*xVcW(3vrR`+D)pJYkRYfO$(L+X=mXl1mVugJQYKjE zIvgN5z!*OCJpq_*8~NuH!K5pYnr$$Hk2b^NPPHHlKF#vLmD*hMu95^|VjXd4p*rQ> zQ-!&#`vAl78OI-4=MXd)@|R{U`SnjGX>B^jpLIe{h!SfK!-pU{ndS#WSg>mI5;&t` zIa}>fqV_Wch=#c0PSYqcQqLtSgR_ANh_Nuj4@a7i&JJy@-U8mRZ*VJFc9W^mY=Xw_aM+)CDzVGlG^ zk7gap2wA_Kayk|l==QG8}}k^1q;%kG2QEP^rQKlP}td-aE9 z8p1r#XlT8~IJ#5FQ^hSJh8i{0DEd**e^D+Bh?$-YTP5pJxgX^psTC!MlF`s&2Kb(3 z7tZ)3yHcgxs3jJ}5BH;MlCi|NcLnnYZ{*X_{N>6C8JI$j!Gt^+EYXs0bJ&hEg2Grw z#U=TbMz*@#QD7R%Lz}Js%v>rWY_A?>YJU|^)TGq9NwTDK|1L`TA$SMe z4vLQPm#njFV162HOzkk-qD2gDwyNRCNG2red1BaPr+to)?n^%Y|C?ATXlWs?3-8JG z^+Gef7x%xG{dGE_{}l_@h+p=PH(k)h4K0RCe)K);fwXR-MYcFP0qw}zr*$GlaF%AW zi6=(pm!HxprH|jQxThAS>?)J(xMGu#nl=lsY09FIZ>*_iXVV1R8&PQK<*sSLr9|JI zn;7wzNsG~C5@&{u)mzJB&zvghmX*-enQ0xE4XWFBJ;EnUY2uhV1S6!M>yc5KfVP-W zU2nS%tZtFfyUbi@IDhJ|v7*JC>88SaVw?Af@L{s_hy)Be50c=tHJ8h@t*`9<(&#BM z*wo(c5n02h3~czspGYO@h&F$IMG)zezj>(3y^)B3?Ly<;L*-Ls$)|pDt}o`k1Dh_n zlR2#s?{moLv3XcNYVG=|J;nD(`tt>L@JAZ>g`I)$HTnEGPNkmk{<&6%i#z{*XWA)W zH^kZ^7^8r1%m2#S+iiRQp2v`s2|nUY?eOA)Tw?BmFDp+ra3s5%gSlIv zwiAG_45a+;c^yj&*TbZUY&bViFZHFf$s@75(Rk7}52L1E;xH5#J z7>HD#9(c7NIMRyT`RxLaoPpz1b_4!-|E9lJ>{F;E-Zh}}EBl4r^HLaYN>FO_17zJx zu+2|67`UN;Ioo@cjP>k)C(`flZ$otia%xtc!?A0Td0`R z#X54{%)i5Tu_PMw%X0<)IV$U)6gu-JSVo~6qR($%$uEm5YXJ^zxd!I^8gk1ZDdT-I z$$F0!e#?9Lbo>qgA&?$9tBTd;Wx@a^J|!0FH4fivHRl`E0x`OM|ib(dL)O z`Gp_m$XRrO;{kJ3MU4B%*<^m<^A(rt?SSAn^}h7uEZjHE5>v%HrI;Wl4UKW?>HQ?O z;itPrRz>jUI-dk9g4>c(5R6wl|Y2*+X+VulF<^rR0(Yn!5_ds!_82 zC0;vjCv2hmP3gW+VuMwg|D0`Z&h9K?l@$L}$gkWX9yrpA{hl2~k=0lH`i}(vwsQ5` zcdHeVVu$=X1=3)=a+W_~Z=Z@NY%c_gIQ}XBNvsy=APjCwr7P2U#Aw0|`wguBjdy{` z89e`tbh86)=Ef7R<*k_Y>;7tgZu(l?jN#ay6OhaD`?u}vqavK{IG0)<%ZL6CKCi+(OcvLVddbmnh`Xly3MEjFzN&QPh7_c8mJ{aD``@`q4MaLdUwcVoKU zvZ?*z4z1i=)q##GAKJ@49!MpNWBV&@lCqz6Ab%ql;c&t-@U!HuvtU6J`7KtYi%e(x zvG`w=Rq-nxH8*GQ`Mi&I%6-q1ZBZKN6yYM5;Fv&R(=nM;!K5{|rR&{}y1z;-&6wf` zCUexM00L4AW_?EJ9d4a$NG*tvV1Pfh)@r;G-v&PlZbgWv;V8$fKD)?J>L9zQKzld= z1I2KBj#{h@`@E#46^NJlAO_hyMwQ%AnH$BRIIT!&>tgM3yU>GNGZcss#h>J?RX=jT zYmG}j9nIH%hZ|lq7E?^{+$OW8iwgrWqr9L~ezdELC8=owi11Iv#gy=78V+X7rW5k4 zib?fcUmNcDLTTcv@kwhjlta=yV&vH-xR=3Q669TAx}ETmI(IRSPN>G7VPPT zr_v`q{drkZv~8&oH`8%Db0X4PCkvWtqwuyaqG@l~#-$yBZkPw)tOc>PAUSJBdeZ5T z^SvGx{fW1L&rQk%a|vcXIe8neG`4>vd=}S@thT7bbNfvU#sMAweVi-o0fTS?|vNB9-s5< zJ1ub4ZnUm(qB92gM(7j%CL^RO>U^*keFsmr63%nqF;@^)cNsYX9!Htz=nqrkx{K)( zY47OvV`98-rngzD%E{bxqBCX~zGwY+QHzK&f~v9rd4f_22+uM#x|0d4Beu*do7wyV z$|SZq(X(64vg)+JGyeXKsW$S!&NExEAW$zK&oDSF<{FiTYOeX|YFw;B?0tP8@^6iX z>4Krw-kFK%8P)`uvObfEY-uEDzJ7(@gCrg?6O&w}3j`%D*+x6LD@@I^xiRgr0jox0 z;78feE7kE$P&%XD5Ce=6-A3~+Np2wKj%jEV(M{hjvp%_bm;f2FmVz8GIL{f4f{o+! zAoc>oW~E{#P94B#BS%tif`Oe-6dYx2HBZke+$3y%OeNx9x^2A}trZ123Whmy0c+Ho}*Su%-f2m+InIf$_-J`-u?G_^el& z1W_6gI{gFkC+&AV8E{M&Toh2WeqwO#_Y;=D!o|Z=d9t-_I%f)&RP;I(8>HC1y;~HO ztHgvg6^ylhO>=FZIvYf1!{?UB11@%aSMjkaz$6Ie0 zGw!YI?z*Hak%Mc?US2K$X|f;WsEWWKc&m^}y@LrRD=+-&^@2`b8udqubGQa_E7f!t zj^*R75YwLaD+pzlo$8rRx-a}V3;PXs7n&M=JdVVyjC(?*JbU^#9lL(F{BEiyyaVJ? zfPXCz_+5A{noF|JHxg_Puh*$_bQwFpgN493W&hgF_=MJ)^Bc*bmV^*VJ$B}DDJ;y> zjCTH1xzF+=-M~$dIsfQbHWzW@n+p=JbN9?J`My*DlFNlk6|@`G4S4 zg+ZOcod~-i!?p##7`GIY`R{!WlXj%J`2|R#v_rj;e0XoLCdWKrMZe?1vT?z)sDzV8~c2I9{%sFZ948Su2cq!#&pno;y|8S^v;B~IjnP+FKh?k z5qi_wsgbiOxgS0o*Fz9OhXRuLvU-P*DJV1vVU+X;XrE*ra(W!BMeu1MhbN!h9W! zAEe8}kC$aha}vn6q~s9LW(e!qQ8-0~K4v*Vi5;q_QRt2@fk8FxT~#9Vy$PfB7=Z_mvKoEq^wq@Kjz}d4{=9{e7mT#z*PMse0de+g2i_BsQCkz1B>VP7|-x?(& zyWRHIF*aMIDykg_j+`7So|uqMY|Kdcaa7T#3QxIf6QAsa(7=^kJ{w9o(UwJOAmvRr zl$dl`xbRC6N3xXmoBh8v3(153xnN1rG6I3|5eG|c~4aX59=A?z3aL(}4>n9!4O7oI?Oqk{b zAg3`JG>p7JLuEs+PXWiV7R|ZR2tT!R?Y}Miou#WzsZA6SI9WGdnHT~CD60TXX5f}k zWs&!$ztSj0-+EJ~!I-FclH{v+I*Hzu$sm4j1adJgFSUW6*+c?3X+BSgO{x|_;Px9- zJWFTh8PP-n8eQUtPFUDWD$`pZ^|ErN^ax$TyXvr=i^2M{+zhvBq2rQCCW?HKQj)*e zWp$5jH-8>Tfw>P7OOpDMa2)KyT?|;Uzu70=Oh*xpOQXl(uSGXO=5>99F_F(kj$0~z z^E6>grLB-co#tfdVKiL+_kS^_T8Z8{wXd^hlLluZOLrz5&wxaK$YEL#jb5v{Fub+U zQnq*CIE-3&z)F3ov%di9Kweu>{^Bi>f}XtQ3;(?0;v2~pr&<;7lo}Vqcs4fT7GI6& zk^9bhgPh@*Tdny?>KSit#DWAl0#ciBN@}usrGqLGr=|1&iLew|OVpGODw1}YMD=}Hd zyI-p;D7^lhSQ9Uin4w|Oc&Lt~CRZVpvLP8jvGaEK>V_gZmy+yOzMd}@31$^b!lxDgb5A=*)_aEK zGj|pOjkgt4 zBIrqmV2dUy%l@T+i0-A^&W{aoSC|QsyM`G(+72>H_13sh<<|%%j-X{2lx7tx1Z9PgD9Y$4x0JlPNkyx$0UpjZb3L2GR=n;W}O@q=W=kaqs@zHpCu*mAdxQ zi@TjPszY0-+Z`VzluORp`dj#bWJmS3^_riDTHCp@y^G=6lH+2~QM#}rn1JtD5!T`7 zIq@ptsS>qTFnVu=Xp3}mhw&iEo0bz@qVXVC26V$4L&;glMLlROTQL2HRYsfjgxGz3 zu_2~fdPgyh76B;b z-$qPRXLz`KKWEx_V*vZk?3hnq%=$(TD1z`^D_reumH>j}Du9jNe4@$%@e2&7n$G>I z1Gw*M?K^HF%Q?mX<6907fE|xq5NkqCJmu%#8Wx|cq*cuscbz5!WTo#|xy9dYBl~eG@h6a3taqEucf{N#O+{174goZN0@%vziyx!_~&9Y?d$jzv7!8E*j zdfnHv48L!tQXSfS{aSFQnH-23-p0p>GNf-c4Y`WGl#2{-3|mgmE9J- zCi*vLMR*0ej?5_*Rw2uiuAYO8FU}80v58w4b;_OP&B}f*JZodLKRK_ye^i@AD}z_+ z`r!AJu=U1U(o;wIUU8^kZ(&!j)mrjkL%lX5)^UG+c=u5Alyd<*mQIgD~E={jB`k zwB40-kElSoDOyS&YQ`Cj*@qF>=Z?NUHr{UwH!6LsKpnyYvIx=(k}yq%vlvGkZ7QPb zx1q6emksma-IfI^LbtL)x#1h z@0J{ej@l~&P*e!}gJN&jetmjBL&470b!An$cI@+<^Tc&vy2)EU#}M^zZAL74(Dhda zlURotGtoh*@>4-Qq=2Wtk?j25|K~Jghd(Y1h4FIx;Pz|0@Kt|Tf^w>uB0V7H;k7f}I!|-jPd~Gn z1n2ePg}gLz3F$sOUm z+-VPRkseZQM0Arr8Hx!JAcVCqQ9-~!eQznD$%ZEQc+{>o z^T?Q9+Xet|Vsb@o{|K0;_Q4PTzrVQ_{Ob$U0V5g6l70dtc}j5mR*!UFULC-u=eN~B zEoTQ)NvRd$!TAfA9PdjLO`#f2C-r~dG_=x&ZDFz=B4a!gOct{{r<)s{_9eDx^AKXu4U9bN?q|4>9Q zmwmX~c~7x``l4ap)4Kxx96qrWfp3gAB^%mY=a``HqRlG4IXv{zS;T0&ZH*4M0S;*3 zx=-FgtN_)mHPc#aDeE5fRP?I?<|(GVqJJBlj7E&sd0vv|mt>)i_Xe|w2~5-FXRpPR z3NY7q@4n1htFXy+FR(GZ{wqU6EgouY8|Ni={aWNCJObP4nfmBe@R6N#k!a#$qrm<@ z_Tn05@rl3$x0aV9zZ~z$Ue`RR=x4dpmJMr^?rWdZJ&AqFQ1%w>p@oHUVRh}x&m$u) zulLV%o8IvK&jBlA;gOh*=nuOTM_5wMUk66)cO{rU*NxE%qX&PeuRIbMCAE4rYsoXfK;GD5e-z{Gu4Pc2v?>V2r2 zP&&`s{rqCV!^i&4fA%6xg6o|3BluZYMQ=hs?1fYYjApj6rgNHn3u%L%ZRwxiy1qW~L0(BfkOZ;+H-tcY z)BkUnEy27I7n#D}{cnl6f@)%1Z8ex%%^g{pGw<%X3!?3n^vixAA~UE-;fDxZ?ov=S z>C;a4&q|HcA$$n{bM}1QtA83%_Ta)Yw&v+a6AC}2Zx9Hd@uN5%U?EorTa<1!R)bLO zkrGU-1L>xfc*G~63h?VFPzDU_zR&QVf4dUjXku{m`*Rtc_ICy?5rl>>3}}b-a`y-7 z&(W1tZJC~BG&pzM#9@kgz}h?`F3pNS9lxL{@h~y!|BE!~h7 zOKO0NjU4-=dfsQokB!{z*|-Ty%Abk?D;t0IR!8O-Cz4LHUHG(s|QLgsm{%L6~hO z7VjIigba)3lNl|7697~~`Z99oT9_Q%?TZ1|QM_MM;=% zD5lo(O*#Ppuy!o)vcwgJId{MV@=s0g7(>4)%kZFNjP-F0bxp*s9VMj!oY(&!5!c9i z7*GYfZS*rolb0SCz7wCinU3a43rK{tVkpk*{o;dJw;tYjn38{P+i7}W;b!2qqb!l8 zKI0_y$sp){JW}=j>yxst1qAi91ao!&g+DR!XBPHtyn5bVF7&FoK!Of|Rp(ZtanXY9 zn<=8MzB?1t%4mkw+iO#C|9vE@u6`p>Atq^TE>E6!`z*Z_*lsN+>kcaHv&;zvmolvZ zJYYk3;S72c2h>&#j^e1sxh4xfXw5}pP!Qnqjgq73I#?Hz z|Gn_Q#4vm!NEF?L>UeYnep5g#WCTG67#!PUI9JF6HoW9lCxaxR((FO{!e)8&C%e1| zfGznZNrK1{=I03fMK(Ua{CW~)K@8I06^BhB$pgl1Ej7rqO@l9b2zTZlSoEj2KPL$m zV>ps+%n^3D&fabc!}#@>lkoP>wk<3?SIPztKug0^JukT7Iyj-?R5lt5d!Oq%I32`> zdV4ZV)A1*b!l;s=2ZDD`fsf_;1jfU9`~ddfodXr3-zb5CPNv!4p324g?vAn`M0W*7 zS^zvvF|Cp5*=087E~Atb1ZJW69e)+ycb?Zf<5R)h6~`+_4icL2Sr*|BH_xOdQ!=jHN1NGrKiXaumF$>i_ zvlJ15vPNn7XHGd^F8pFv{w1L(L?yK?NmV&6xEz=C*RS*JBpa_bxTM{`g{Jv8%6b&g-r3T_*X%v%D=#;4^hB zx9c=2CtNcxu?T+5I{R2a@4bOyWAmyq`%lgs-Fo{OT>UjW5DDPsU?TDV$IZEw+?rb9 zdiJi(6cQS6e0_UO5r+IfUQT=;ur*la#aHyBd60|!p`P_|X-%ZDq+txEZCX@azpLsS z)1zD_1oOiXI^``f(z6~^C{QG9ruqNN%n6s;`ohUzWV6+5C_a2O?Mmn!M8(Y+>bz}R zWq*I=?((|fpR7TAm|cT(k&9|JX8G)F{4=4i z9!*!p>~=uRH3{vJj{;;`%s7?QmxcCnJX~{EQU5;8n0mHvrc5yZ&gb*7C0q>qk|C$2 zf2@fPoAMJ<)LH+Mg|Lp6*l0M-vc6kG4I=AQ*18Y-BvQ}f!Q)1s#(usrhET_Zg7)wGay4W= zz$il8REyky zGu`!Tm1Ia1B&9~0qq-$R-dwNs9M@gDDnwRsRs)Fa$E^cRKm>@n#~ZO^fhnde@+V8ROHCBY7-`JY#<3Lr2WYbvOx-^U$!Moy#pqAER36&cb zyZMs-zdIBackS*N0urD^?G%nOgy_k?XQ_}mwdxUsonx&1C*rAV~&gjxUzLS#UgQhEj?tc!SSeJhb7Sfb;2+YZV*aWhgs6_Z^Gyo-|DA=~k z`gp*l2LWUy>6WXfic^gm9AQ}3bsE}z9Y6Tg<8B&)xeaD}Ro011oEbse7 zHjLD3_5qb%fwFN9yqFD+nSzF(hsu-%j{(&Z{hzO5=|Sn(4(Z!$2jA!J#oF(D_$70d zo{(PWz^2c)N+5_$GW7BxsvV;A5LNtj8<@=%Q)zdz2;Q1XOm*%iV}tasx72k=AH=^{ zl)|v-{{F79j~L4EDs5b?K#E7%Wi`ylc2$?#)z_cC%D6?vBG~X8PGy^H8S$HjMTZqU z|Lt8z!!uCc-M13cBRwLZuaYz(fUO3Tm>K~4X>n>yyVB~nSB07`bXgB?7thHc%(CfX z3TRrWkc2_R$3G;5it z>Td&Inux;glfEq)oQrGl3v&%+>gV}32!@^GFb$uU+n-pC`qFka`rj(?qJjqI9C+SKk(syLyEaExjD2g%N+180i5 z31OfK)ID^wvxgs5{*KzjQ43)@#I(<>zg1AsqnRx&pDIrNnta%2M_}m&X!3uk^Q&!~ z&-WM$npA#%_VA}!@0~Kg=P2&Naf0uv@?bV+huHsr$a>3wIGQGEba8?&?(XjHuEE{i z6D+v91_r_j$i7KkmPony&7ho}KQRs#B*fj|w6};sm4CR$c0{ z4TllS#@NCgECQL;I~=UQG5vAhHDx!1b1ZnAtQ$54#ZDuPX&ZW}rP#|2=RIQbXBgY} zp2z9^S4jp z2x>B^MV9YDCW}wXupJXO^Y?zG0IAO?^I`>Ocr;MhZXmAeOrsIA^ZA?kudmgUfKfvCrcC%Jc~hW&^Gm1yEak#JVeKpEHqy^lKi+J zQsS5}HH2`4?xRP$yV9Q<1izu64seRxTjWt%jm_heXk=`%Lj+_{JTo`_6IdA`-+TIaJTd}DVg68JohK=lmF|H zUUN!wWqG_qo{TYh5ucI(ybk`>sXKHx)+@OKL#(2i7l)+>QE-u-RUP6-J1~Z9(HI%`T2bg&WNVYKsO|m2YQYH%IgS}s71Z+k< z3@9jsR8dBpYAVjEXD(CN_yIA_Zd8_A6BNI$Yejg}SqosGGgh^pGva%(X-!5)>~hd6 z$vG_kcxS98C#sFF&8g97Sl554VlQ5Jr;K0<;i+_}aCaQ^e~>(}a^<1U*!j8y&Sfc~ zQ^q;P_L4?u7ZC@|KP_AvkxQp(-*C{0{7G0U$g|JRI^<3d!Bae#GB-PNUP zk{7dwGgYqF#fIL!?%=7(a?%H|qrK()l*pe(2|Q)&%e}f#w=^CtAOpi~p4AHI-mIP8 z9MJ+!8UK0tim(=Yosi%(Rgm)Q-V`#uHOdpjoE9X12Adm?>U+)Y6Zh&yx+Stq9x(v; zbVzRQq1u*(`hRH~HM`YZ5(EA+f}KqcZ=P95qHC%<{^oPIwoL*~|}y3=dxDe#YqK*)3^nZjnnMtNsI)uv~d%BdWfAO*GWmI!~ngAKZNL zKd9u$M|P5TGl{@Ql&$F3rBfO~K?>{};jkeEf0%>(d9nvwOg=l}8(~(*x#)2d^1RF# z)_+mTb6wAwSLr&pnjv(wS08EHo3+W~A2;`KLr?isVS+0UBsw26Ovo%!@efOptE|rF}?D_GX{X?Gv_`#5M z`c0mK0{aUhoyfFXtXghn(p9bZ5XUz-( zvdb?554O4}P6l~mY(;lJldGHcW|dn!LOuijC`k~mZ<7ab4=QrlImzR97o6qJ*Y``I znUQattiBe&II5JB2jD*~ht*1zEZXu?nbl5ap!@$MlS}*;R$rarh2umlKQENy6o97Y zE$C!Kg4N?By1Oh;5LDfCJpQ})(1#Vg?=IHpA^H(n_#IlKH*HFeVDc-6dMCtdUF^P> zc`Fg#NUNFjFNr||g;CIy#>;`kOuWk(+#+PVhz@-ut%@hMiHuvuvktYi9PuJcj_O;j z;~(n2_5w;@L`cB(h3;;kYH$-G*%yCZ%W}E_LZ44Nqzl9h=%rMY&jj4yi(nR-jT-H) z9;r#>&-cI8c70)M9f52NIgPxpm`(l~o*QrGCO5)QY|}TwWIV0!8-$tf5L62DkRb!F zY7A*_6`jE^h~wAtBSO*w{bPds{e=QValDHP!6{Qs%A(&Pfs-YC7py^CGJjQic=TNi zQqY^y6;!}o=-1b{{&&ke^gx1_5EHnX&0IiAJoM<|L}yuQ(Y6ayuV4S6nAV~ z9%dXDYlNA%EB}Rc0Uu*DnpD+5JN*_jOgFox zFLcd(2FpaXt9peP;{TnG_eo%36saVp?xCYR5P)4ru(JOH9f?5kElPEbXJ5qAIW2yQ zybnvD@1Xw#`F}#-qZiB`m+THydcmXz0$_rhQZ;w~JT!l2suj)0J)&SA!-P>(p%YYWT9{T;68(s{k#Q4Ii@hv+eb>ww))xcSx(~QZS89kQ z4e0~Ivai(oZb)3+{G%JurXfu0)sQq*yO@g((?cD=_tA0i32&8xf#FENE-peZDRtEf z;R9qNvzsbMvVxl&*bEUwiCNC0XMfLGDKf{L8E{dA%nia%({ulSbfO(-(;QYz5z^!2 z51HQ=nlnO!tSB0vd7Vbh=(klp3{$GlX?!0op0IXv-j76yN0Y>OV_Rv_7p0n_f{a#= zPXCud`~}R=N_+JSQeNrVF8v2y*&_4#R-i?J-d`UzG>zl;k*A;$77sI`i7UFg@69Gq zW+nQZ4lQ7~`NfbdCd;=2iJk}AU`hE@gr9kr8HH!bNVLfSoHA>z{rVG@)|Imx1oM zrzGc1%0P#trFA{gJckP<1KCJiif#4?;v4URn#RFg`EQG+kRm3DgWqRA}i^@gv0tLId&m5gS zWCu!(zWMA4p}dxF_H;)9yCL^NsKbiZqt240pJwt$$YNJCBYJMsM_8CdZ^!<;oxLiA zsDK7T2LT5H*5jsdwcM9DlQc}8o97}01@VvpSZj9Jo1<#uP2n^2iuj)%&uzUgYrwlsPaEkyPPtgyV>qDppB8EpSrN(U1bfvSE zm5z~Na{?Fq!Q3OpnwqmMRijtDKdj;@S(!wO2W!`1Bk1X)ebeJezT{J0;hM) zFfptn(~rwr_HP1)E4)2Gfngfby8Z+-#MCmIutUHvh>|etnzC7dsIs;G?%5Nbe~+|_j(nI?|09dUFwUF8Sa_kD3~4Lu%@1+y!U_618fxKhg2s zwY8RMpJUzs2ou5K;J7YsXgWG8k=W}JnKh5@GIv_>`Xal!$8>UMtrx&ZrelHc z(g8!TgNk26y3U2Q`Cyih@xAS{xKBmquPe#`f9`ADrz#caR-1^+EQ^OFyITx1{fs|1 zn~pcb1T~_juo>)?sCf#v_-bo`()t|ewuI#Fir8I{P@CoqWZthhO2?r-dZaHq2CciA zMKru zn(^rwq|iSd>6KhQnpvaz2)%gL-j5RXBZL_(fVD0zrO{pLmx_1|8O`<<77i&==f#4tX#IxLnYEP3C)>@U;UqtLZHWV`LwtW7V3WZ=FMV3JPJy;liMnlzH z(4ERx`L-d}Qm%36`n`fLIhqy9(q3~V)N?5F0IOQMrND#cw`>qs?h5^YI|{n7fhwiu zE!X5fw*7C~zfLV`GN4JK;WaUWAbx4oI`HpkqXcX=VJ^kHQ3E ztAN>Ot}o#!XRAkqc2x9;42G(-878@su|VR=_yE_b*ryKSSM#TssKG`aO7BVT_G^h%Gy++(BW*+n7em72%zj7%_!GGuI-j;v7D(x*ss%f!1)mcaDnJc%+ z;pmIOzG|hyZ5nPtCLt`TgAi3vM=SP zDyZF25{@)BiKLQi=*#d^)#UhAtePWL*j8+yl$iRe?4)%KNkw8|MGmKN54^oj zL29FCSV<)+0x<3TfJMi@1*P`v2Fa%k$Y5yhiWg>7FUDk^he{;dn^lHMFywpk*1JuOa8(i1d~*vd##K0bTEqDcNT#I^+mM z4aSzSCLkCJkF8~5!>a=Ep3}|l6BdU@!E_m!XEo2Cgdz#rR|t1n{58UbToJZHkhSb{ zg;JQ){q}H6H8rZpcR|hXbv$^VYlC@|yQf}Z=WtQdg+#PUSbrMny#=SVial|DTh-6> zOrwhW5fUnX@k?#F^d#I~H>>BIc0=07e-XOF52%slnf4cYoM=$vN2y>hZnt+xq)@N8 zS5Np?RJH_sq*(F{C=}WL`)CJ{5VIp$KXfw(XzQm4!5T32lCc@VYjQvu{iGEBtfH;* z-}nq-P@QPb&fR>CI&#o*<|zkL{6n!k#Xxv>;JP8qbKeY-<(%8}a?yvI)0X((^2#(Co_SOxno+lMit7VJq zThnc%^}|hCZXBN7eSa_BSxznG#{e;LW*()^j8;pvJQb)GT|48+EM@SZ&HSb9>5>S( z(VW2X!OZ;mG>rGtTpwqdDbFfA}p5g z)5e_D;8u6MjayQUWs_NHg7%PiQ3)wd#|mLt7q(l%WT=lpxSIgI&4%_|Jr8uRZ==Z> z64>Vw{b(5KlNDw8IQ+|ZxXe*$g7JandSeqY1xtAj7nR4k`+KN0)!)O7HcU0PhVH}c0< z-337$$ie+GLebvm(wyVT<963{C}PE$F(NwdkTq(d zqwyt0+A52bb#m7?q_gA+IDX|4f0RL23%S=#9&bd~gA@pKROx#UH4N-}si$JGtEQ|* zwI`|3wI^bKaGga-$F9(qpj3Ag{WwqdH2jCsRTOoGsKLs#BNs)V&qRrfypTR=|d{FR}RitzcSizJAsqoGBA(@PaSH|t- zxm(v)PX8ZtspwPn5BY3tJ!*^Q)Oa~I^#-( zyCn`zzNb!yx1Ha5l%_G=))M(UxBvD49u|tB zc3H>TU-LM0WtYFPt4w!Cfo({dInbXqSuAAWQAgx%>vu?uUar?0HlZ?y=dp0#sZ9eH zFivRP>V*6@w;$N`axB+kvapmn|hzAUhGj&(3 zJsBA)1dZNIPbYeA*n<@tzMwI9X@bVJ{r2C9C9qS}h9^I`9QReNM;2aY-#gwRM&OEJ z8b~~C*zDT_!XUIrFMbX9eej!ANzJI)w~ssu$mYL>GiDLhGBjcK5ZNaiLTbXSbLDLk zN{roR61h_U3}?N(tQ4~%F#hL4s@`=^V*^yt+@c?{BJ!D+KXSv}Ox@z!)9x#r?Q&ce zX7zjl8l*m^iWm|0yS8JSs3Qhd1^0q$%gKu1xn=%!9j^9WM;%H?Jkj?Y3v!j z^Bf^>pq9&5)yppoy%1&zH3l{*gEtOzchk@^6G(4t*|uCQGAy9T+GF~-XSB2X$hL#sGFSqce??=!a>w{?LB!U63CCx3aspVzwsUwDp9`j4+BNbI} zQ;Sbvb?c^2%peTOPXKG#>}RmT9Cc72TB}xw_@KVd2J>Vr%wy#rPc0d5{Rt>ok<>|c zIb5C_*1!{M2XUjZs^S`aLRkZD%4&%mWr@nBmi0-o)A5WSMzHO39A&fiMrQJGA6gT! zy|i7%=Ag9cVzM0%!}It)_AGvk8MSm?u^yjzCH7DSV57H-gwd{+M(AE{d=Rz%&8O2< zf`F^0&6D)D4^PlA2QbdI)C&KJDLxIj%#{^Ay86mk3k9~I#H}`^6Ekbz0^J>UoBr8c z^qT)fYjKkX-2@MFV@`=X|EEHevJ3L6zza?k;49}5Aj?S&gYANelt(OTDcuOCv;R&# z+ZGogjpqA7L@u8?WdvdyU5q}a7g*7Zp2zk69FciOa4h}5%RVKph-9&xJg9(yn)MuJ zK6byW85+2yOHMnkq<{RG0_YM>N}Q&DOeyd6(;?tsTc6TE6z@1*cbvGQ2ZC;fgYJehFVS6yf62k|E@PKw+bu2nT+N zSGGBx>YlhkATjD^ikky2`k@Bb&>>{w_*k-Ksyng^x<|ZT@r~P>TYZV}K40v&m`#W5 z^lx<-n0SW-zT}EYW1lvw(~)DC#yVs031tFRC66T!j00S;F=R&6QcYs<-R4enjHQqT z=ygK5|1@eN&moQ6?T&QJGR_Z+;R7u2*%lJ?mbm~va4m>S45QEKP|=%6(nw215`EZV zHtZ&aU2oeqX=Gm8s-@)H*ZjVTFdA_W&;@J z({7cqWoT5?U>T#&i`#k#*&^w zidTg*JG~Q9)KVQe@{4<`!zKGTma2jh*)jYxSci?-U90QW{`)ViO7a(h-d(1Mkp^=z z)X#67Fr1Dvk2wHEipEOyA*9(IeZshYabAlFEyNhSc9pKZ0#ghC?zJ>Ccv_bEkN^GT zJ!gLo5Hrz~=AMl{>}T*7QMIeO>SHU>7dR={=7P1qd;uK2GhC9zPXv5nTCQU1meGan zKC<$;K98aT@f45=)D5^0g!sc!-ZG8GFeVUV=z=(`Mxo8aN4~O&Qek5><5S$|Cqq$s zW3fvfytR5Y1DAh@S?Z|0>H%csK@DIawp4OB5EG9F^^pW&04c_p=(I2-RRN+kceCg5 z`VZ1kIA5K2mw~V`OMV@$s&1T|{5@g}DTG2tNB&o)-1!Bjd_Rdrk*!` z;Un-mU{%-8)PKwaKnWbHFL*VqP^7aXuw#C({nRxmEol|3Bld|F#32{%Ie z3In+}gd;{`x1P$vLT&$&j%e#E^}z*?`J)^aO_HcE0Wl(2C+oAQ!IpgZ;qVCAdQoT5Ooq* zx?r8~c@D|3S(QayzLbwvLB6zK7G0;T)wW_)c)h?2DP{*(6i=cLx^y;UkBiO86DdYs z*C7GkBn#hz8u3SDu%>*eZbKTeUIEk6B;S5iqW=H`wT4q5^gwBfDl|vl)~BVyJWJS0 zWje1f)Sg*W%l(BGzefruQW z`>ZU(S&_iGwcWFyXNxi+O~}x4(2q-;X;V?ADWEu*)0fBR%~QkYu~FJGBwd+dhKce? zkHl&ua&nwWe(ID8=}A>U5ww>74s&~99V1b5j2OwWjF3W;-?t3@d-u_UVP! zqJ~e?_Y0h+V`A5P^l0Gz1FtOTv-Z;ZTuiFGJ+zEnHV^rk==YarX}upN-zGK}?FNa${L2JB`Xb%uoIl;-tNgh-dD*NK)ko{}*`;M2KAsR=% zO_eMbN#5((J0pN9k&jOu>M{7~R|d?k?A`M}B67n|OTF!A$G4#&+~?KVu?_V#_xhNa zs1zqL_avP2jsx~2>a_4Ie|^mr^d6Z%hmp`#U+}`t&?@&2AT78rCT$OuMbUF)oRJ6T+MFIMshfDa=FQqmppMHph(6Zkq;PTc55)C5pFd!25L%!t7>o4yFCqm9aZRhCkLU8>N(n zqA|g%!n6vh9_v>gOgHQjr{i5|!^N_B65NIqb}cdiIIC0t*^2MiK*;(eKrv}fq?u`Ri^&)gwIYw?Zc> zwgB80I^G$RU(r^fydwA)b4#a+6t07`9UZ8oKcO#E3y;5}BnSOl*KYoEcr1u)B)6a; zDxdajn+&1iArlfg-A2*q@4)`zgDY_>?pWSZ;$cO-gSyU&s^MA~VRA5CY|I!wf=5&J zW1N2%0I4~qR`Me#SF(6ejH{FIYBhN=Af-+Sw&RQovUoZ^ z%Z=(B(k$4Au(}`6Z@xw!=b8L``lJ8Q9`F^)XcVw2PPfc|yw_`W)fz_FPF?W~^}?n9a9bHEKY9udIk#dg~o{lokDv|vUoe?(f< zqL_~3Cdkynf{KS4H+&^Vx%{Pr$zZk9aIo&VUja6Xaavp4(z z%7Lb45`VPUE!PrHG(e5^dwtcd=tbTqmhqiA0+=^oPqh=DFt;vE8<8mDWe-N;?=a)_ zFZ`x|ErXt!^ueOBrXm-=SelT+T8p>Yot~+t<^x<$Je(J!!n)!HWlVpC<1x}jj{XmS}nZ0Oulf{lE zAJ83(jQGuUY0tL)FLW`k4oAK%6KHJo6jH93LpqdI{yj|-Ow?Zk#)nfo^^R_n5@tW- zsA56>tF7DmN2h$917_?DB_X~}R~+Bf1h+JR33|NIWK-6Q=+hf{c_9Bl=pt`D?Zq$h zFE2u@6tK$2anZ+N&pRoIqr;^C=Xkf%;&oq9t~sTEtsEmZiplCS5>w3dgfrgkM>|n0 zXv>Qc(!)Xdw#$Y`bpw&ap{O#qn)4S5ID3fbX+e4)F*FqqB0^xzF6P+$Qptc%cSOz) zNpyrKoIYMs<+_Z#`Yw-Kr&gF~SsW6$r$J8y5e0Q~DRiiWfHAR9@9FTQ*E8+;w)h-k zucL24TC{{`xa|pn&oq~jtwJ(-tT61Ck>54~6D%1J|I?&xZ8JjnZ8!4t!74+=c^Ua0 z7}tUw92dBD$EP4>MX0NlSj>zDDBo^4ALw`Cr`i8EyY!t|R+=Ee+jn*|@*DyMxy|+c zrCqB+knrrBr^Jy?XYs@D;e3BkZKQ?hA3enr$J&NxiY_9Px4dK~|Is@U^LnmmQi9)k6 zTK=U|=*gWusUg2=F+-aby0l=hAf?w=XK+*%{)9#^Lp4lbNM*P`G5Uf^$+B>de^$-zAR z@t!M2&o5w{m;y0}uWw~kkmOFi(V%jq8V4Vr?z^An_KqxeU1Lrqd<_)iAWXE^lsm=u zhEp3id%8*&@&aaAY#*MplIB{?+q)%`Hq@}eNRN=z^r(Ebpau;lt=mNIXfZ-oCubSf zkhWN$)1;-=o2{iM@FpXe_@^b*hdn)9pleGh8Rw$-Q_me61aBjnxZ)c<&s}71QrYN^ zVLCV;xI=YYh4?MHC3-CoQe4*)BZyxrMXPeIjx`uHjI|>y36hO;76l2Ny_Z(KeP#61 zj&p$@{73L`pvnh-Ox|~Tm~6Kx?O|w%`wyBuF8ZksT=a8OhR9SiSPEa39RzaJ3O<%( zG3Vcdf{(Vv1)>W1#eragDDr4N zQ&9KJih+g?2@#k8{o=S%GThzKw{yNI_Pb>oS8>yxe}uY}nz6Ovq*wJ%X7|ItJ5Z1$ z``IBeJ88+pCk2}Lz_t9iLZ{}?MfCYjC66KySLAe2fywk)RPT>ZB0G*U9IyLNbqVew z?d2w!(q}iK7s4fE;6XTzwwJs`^fzUSd+%M&tVQ&%ZuPsSVu!h`9c{tXU`NNl3RIl7X zv_>8!?L0DTaezh^ObFf17^83V7ChtC!r;#`@BC$Sc*8yqGLmuZcqc)i1toiA*F7>Nrf6 zNu+d32TQix%3FLLX6mCq!5z=GPvW9Hijc6VyoM5|(iFMSp%q)Fmbwmv-dfr(E|~2XP&QmueKo;EW&2oWJ8J&k#M`1L08?(j@MKBe1{t`#1}mxwz!+(h){IyA`|(%;FLg*y^3jKGHc|o zi;47*{&8KrfBc8QIDVwpQlLi#8qT^h(xVL#1AqMUQ~CR1Zlqn>2D1@NrOQQX$nD}4 z&+>itGa)o6zK8>6Y8&r)bH10-1E$9i{icFo{IjxZQ3Ic{w4fh0VgSLoTR6_;2imu# z?1;Df8K)xHa#5P1XlV^(yaP7{=J|{Z z35&&*dyw=>QEiy7$hlH3z9NTur016%ht6Lm7b=Lk%>=dn5DhbYBuN#vMh53ysu8pX zU@j~>CWQtW2wVsWm{P4-%Ob}FSyTW-dU>^VxP%gL`S`sg_dl1s$qWm=&LHkh3mr|p z*anc!j5esJuTkEp7I#b7PmbkWGesr_S_*1;p3kZe=u)?(-prp!u6^Gxa#CZwQT-N1 zR5Ds(+u!{+NvON1pMneC`=wCpsK9yD1y#y@y-@3*02N-2`8w7C!Z<0E??*gMI^Hzi zc;U+Yzr$Ox-;Fc3iPYF`#cKS6&C<_g%kZZN*Gpn7&V4|qFD--Z{c0;1*M>wU!DaVR;HM(5(!lh7FiP4wJ}j{lV;Lwql3#VT zk7-;w+;xDf1&!R^%#C(gADE21^M7hpb`4?0Ir* z*4&{xqYF#J6}Uh@;t-*heQ9Uey(lN;7R;eiY&yPA7m@AEloC8DP&@%a7^9^G`NuyJ z%r?eK2}~4e<}tS19EX7GdebrpP(cw2Ts;D^Yl(%+7PPYs3>ivd{$UDS#Oc(;at-Y1 zDqI(XWw0`paAiirL)IAXY%D;W_$L;PEgM-!6p^dvIII zC@5wcbHCL%pJwRe0CRuU^$cW_TEUYt4iKu)c}hS%z>&IdyGwLRSwe9w{Ke+I=x7ie zqSe&MEq%3_QTjP#Tt0&Y!T&r?p0I$p@}-Tb=pbyls}si&Ny zpCXmHFcSl|ZP2)sMlv$!Iib@D-Vk4puJvklgQ-avofU#V#UbIKnMzrr9 z$Yaut+s=qeV0G5~rUbVb7tfRqEr^x4Oq3aTJf8bs-k)mY*M8$lL4+(fFtBxyKw`BK z;GSkbw~|33t4_?H&u$-Ck0A5|5NNGDQ9V}(CeAztQ8|Pmgw|pK@Udj}x~9*VrapAW zq!)d?qb(@`mc^B=KLO5#V%Z$({^!mu2PC=N#8tlMjc^CGDGgvIk$)(9?`OR!PHywi-o+IGPaK&3eZ&? zA8WKd!zAexKXyBlzDbxiHb@^U!i{(1xFVe+of2S;IDMp3!nUUclR5QVAxV^r;aF6W z@gqp1gl=b8?u@U3(|#`JkU7)w~&OE@_&JnegkCcecHk>sUeyUOSXWWQh%aPq0G^87JIl-=GtD| zp|>mnvy8Y3=UgslydUx}s(B;ma zK_U{pYTJ~)@_Pl*1B9?fQubHnL5De!@b{~;4#8A-!jzr(#PRUndqTrWkf;p|v(bGj zKYJ+`;okdnIWUopP!(U| z;X69-u=6pMS7FEGbLX8)$v%H`q<&nITR4;}hf+jQE-0*L=$;}}p@s+R6(}c?Wx>;P zTu%}mU11wD@rcvGd=$?Yk0HadYMIoi0piJb_OCkG zQb^|d{&iN1>hw}|JnuefPK(OaXeGNJYCwx>R~XD?Qi)*NMkw&`nK~#(Gv+hI1u zQf!FtmRvG9WYaa>tO46Y`(gd3TUO${%iQ?5&pNuFAwHGLsT9^8V`x%`+VV)bG*DWe z!26J=B{j@5E-*NBc|mS+z>Ces%?wtlQM>iYaO(@IWvRVYiwZE#y(fCmOdS@e0srp@ z(mW8`r0Q!9q1@IbLSL0?A|P-crPU`%Fa?z|cah&+M>2&yJ7xe%JAog26{!IC#Gs-=U6|m(7;Fd)YA@R4;-&Uw@y9vGzV*F16qQ z{0;x*n0OLI>$@8_s@XB3D(kC5Z?fs>FvNBhaowk)uG}nPDE_We-!4RAMXK4wOV9{H zQUc5ejO>9yLd4AP3A{pQJlM9tt4ipc&g*$0@TZ`c8h`aY+ z7P_lyDI{))t)%B@Ds+Z?;XZ2qozvw! zLA1>oi-PPN^Nh*%*76}OjQ9hSDj~6-W?a{s$i-M>gv=7(+b}fgHz_3Rw~Pl-5j-ih z>4edw4lYVcL9a=EUj1pxO(bXEWD4AZ|4Dq`{(yqv3lk_msR*&S`j(DlRHp{6NtV`& zCDd+J?XD)IJ#KYd6diHTOp_qU53xQ!9j@7dQG_(4q^Kf}l(PU&iD2Uc^`1?@jUu?y zle@0|+5$;fLJ;DwHh!`7%BH=CX)Vj%+b8?a4Lm+9_Mr?m=NfHqx2_0V$yZIJRph%>kn;DDw zBOnS+Xl$_>tr?-?L|E~N+?}%-;f%zhYoXZB?#X8y&b3oq)o&7KC zhD?wL1P8jdXjgVRPXf92hNUGQfbL43#T5%9N-*SLhO zK7UGNU=I0)T14C(qGDEhjvLGw-+f#auIwKi=&t06ew{(TYz*j>)7xEZsUy>x8w@i( z*>l|in*Oe`+5=YHu)|<<$~_l5j8zl=%nW<-%nGYmQt~B*Tx@eK$Gfi^^$nLOLsqZQ z^|l%_iX_1Ioit2%8?5ooj#xBFbH82)tAs-couS&=L+rV0jr}OvqB8nHohx0|pva3r zwj(L8$CeZJ73nb#K>mbC_H?7dK3sr9x?kUMhOu|-eyz_h^H);EP`Y@KCv`s|=(_)WyEfVVyKN3N(=8tFb(q39w5b{a|AIFf=kcTLt#jzI!d$0Cykg>Ba{M;rL$8~-4^Y^>xS_j%7fC>=6?+3A9}V$GYjpOeaK#i3&6kY9=PLaJkAc$X$@<=& zBGzX3)kxT9=dILp)rv0KIX{8~Q7+64Z#byQ^CkwiSF1j9i}rE+f>)~@hn{Emt(ffY z!UlP}>SN7_=?g+M-~(yMOEgtDF{l$~o*-`Y4Gg04yXq(KsAyZd^g+QT^2D~%_0RUu z5Ue|HG0&rCvk+{xbkXH?+Bv9ookKdtaqV35hNM%(f4h}z7xC_g$yw#eFXV2UpL91R z%g`kc3Vac?zM^QY_b?Bc!z#4@u_>rTSA-XRazQizi~cF)K)z?$4hVzrlIujCqwG@O z&x8tVn+XRXI4eDN1a&_2X{4Wec zgwqsOj}?CLBktK2W~AACOXv-8o_0v3)O}pcwJ}t2^e3TK<3Ir-D2G!Nk?iXvf`pp$ z6j3ZYgz+(s$dEMrM?~Q_*brla@+}Xm+5{Ij0cLGD19}OFp;H3itec^%Xz|> zEt4F+2-pk+S`@)TEF{(OT?s>ac@Dxo(5tMFjGZiuzvv4`m*j^T#AvcCSd{$qN<(Po z+C$Peat$njTJ+AiS6Io1v&pY;4QRu}vM|a$e?DnG zhBNp*0z~|R^F_oH;U^q|%RV9r3oBOE{y%txam0w{?oUBbA?-2w`v_cVfxmpy2O8H@ z#K@Jg!Fyo6fze7orIRgW5e^NrdSMUnAWOZ~(l%nHZj0yLhDnOLbaHUoU`J#Wf2Cs$ zTWZ=;>w!PLK;%-vhr@|HFlIFmlXDsR-1`%{v_M4Chgf#qUNdt< zAT|kCfRnH23claQH(#R4DteCR2w-fGw8$#zsH01J6Dye7cEQ~Z3;dHXSUcBnCrx4FZPfPz~Em@p7wJk>O_f`$E8yNY&sQSvNIGSi%+=4s7 z3GVJN3=Scc~8Iv66fl)e8hF=^IgsT9PE+K}VAhxh8pFdOXTM z&Z!t$l;i`#9lt-f2ecuFC#?hw-G!>xHfM&D$y`-Z&CSerK@L%M1n3R2yqay8Te1lgB!-Gyq_8C$UFz|J^P;U4U-)=a6%GRaJkQP;OELoMg__>@3_xb) z2JtG8y2p+d`d0#2ZMBW|Ss*t)%M`xpk*0Y=wcREmYMJBGFYyJ0tfm1onH7_klfN4| ztZ#7}G>A3Z{a!u_Gaw_g>PIUyI{6ug6gKx&8!Qsw z3YU!5>O_Su{p2gHeJ5p%PVPCFPO~=rb3g!}3Hk`$+mgP5XfShu1c6G?M4w?%Oi18s z4#Ar7Rs&x@@VAoM3zA+OEIu^P;Oo)l-NFahG|1k<_#3}o;y}fA;X8xTt;NekIFm{I zTDZ&P9%3)Vw++E>?XAU%=urN#UR2*Z9HGd8hV4RMoxcHJzYx7+H~lrBQr={V_$KPF z>Ghja4AWRxern>!5E1bjMD8w0-yhA+cP_Vt44nYX)ZbxOq+9rJ1}p}JB2=W) zs=Xh;fg1Lv_ENG#zS6Cxzn@%K7g|4HkxE)&N9Vl`;+C(+#=VSq@WGc(3<0TOi=piXhK98B zqNmiWs_A@-v~@sG9d>`)J@>*f#94mnUeN-wY7+n|JRuYE6tKsF^4f8`m9;bs)x~*5 zq@B3n(9h!~Uo;IVmg}LkKN9<;4-MjFBZ}WT8ET^TA9=H88RUVpmHzUz^kt$A>o>>^ zHZU}#Be7TxDr`&)AZvC6wo&}T`Z|_u5bn=gvCKNGhNeBGfPWB_jHUEpQCP*1>I z^Xe{n3ciE_?edhO(=*@znZ(pYXG(A&P}h;5ns`LOjbucQpGaJnnE})mvT&T*BCD-) zBTxu#FjWzTC9P>D-#SI=Y1(if$S#`fh&v`wohp9!dALlB zSvrI}WmaK!7_*-QvfJYK*hYeQt$S?wkde~ZAVxAvk%l1VJL~J(q=l6oeK(gPSQZej zh7DM*)JPL=mCP5CPZi2k!)}yG5rHY?cr2GgY{zy}E|#9n7fI%NG1~1!zRcY1WZ;uy z$S5xC;zrvYi#Q{GVl>voOW@(@39vsMOTlf1(klN=FbRO;iRpX5r_jKC`w@$;0P1b9 z2&DH%eN^9!dPq2XM9X1D$+=NsbMVGqLGsYdey-)VfP9GGO+>y1|GCvL;FwF8gdX|% zp&#E!b(56O?oNAwY-`CRuR^%=b`FvrAwLvRR_Y4nJuJKp1r7>*V1*)WQU-#yxMAL; zTz@989uwgh=clB->@{dqkRe+Q zS|=uU7Vq27Sw@OWhgQOkc>S6m;!rvtrjp$F+s+e)5g!W5{Qo%Glz3{fyo17fxQs+U z0$ErJM;MWon3cFwbH{vyv482uGd#&*{|iYgW269O{$d5WIxZU8d(=`*&wkIX{PBcZ z(21*J6kXXctf8}8&2qh%?scLlS{p?{{a?c8a3GYR#3!8DwtH0Vw=WqXA{es~=2-U6 zgZO%JyOKO4triDPb=8*@En4<+GA+lCL>}Obc$_gb7g5|Uy8aA|3N@bC?-{G_3>zw= zWrRPL`wuXMrZQvM?JS8G^(mN}`g(-N&<@7LC-F|$tCYVQ=9olK9jj>kt~$O&y<6!( z(rnlL8;VqO_kFu5zngxMB0KVNBJRKb-QkOYf!+!z+3B4MkQ*x>`JQ3^`~ z$+?r1{S)vDC~hquNF%=JU0##E?BtrHsbkIqyd>7(=ii)@9CkDFwd&KXeG&r!0R)HU zD`G8TBwz~8o-wJ-&ru6ZxD0&KzdDQfl`fsPK7)I~xPcx$!oJn?smEfy;+!!Xh*|n4 zN2KKAJUwG`*5ppuLqQ+%<#$b>V-M`A)_+D-zHh$r#MW($uwws$ctB-DPwVheRv@8- zKkpGb51HhU{V|iVV5=ttG5`H^&ryjDl<2^~=XamZF>QKtBzgxq_bo)$RqTPAzDmHS zwa+O@AWP~^N}eu1<0YK%1`S~rXSzAAbo$9>gNwAV`jl=f!uma z3?@^hpghRM{EnqrKw}IvGIPo2@a4F)Z5qM;ici@VCK`x#|K~;p!f@k|Y8CJ|&nBdM zJ^p59_D*sPowx`sP4*hcxb3!EHwhC;9s$ALloKXJEY@IIGHWPCng6B+sr#QC!)37c zCsy^>9MRhsp{ay|!FK zdArR&#*KKqGMxwYuvwdX|0|y#dX!7U2y%K_cnJds>>G9v-yk6{O)$UT_YZ<80v{YW zd>4RM@bd?9@J&5qo<6kY2X%6*`Yrj!+9a>h#J&VGJVE(8g~L3}9zJ@O4-E7NT)oO~ zw;X$g8SY}WQ3YDkGt4E{1amDhAUzpy9gUARju={r91zSHb%5|^m?Bl3D<&|6QtiSD zwNnQkv|JHVWW9|tF*xnf?>&`|R$^tILX;N2Yul-S^hWS6A2HFDIVu*M4HisI+|ToO{5p;$X#nc zsT3;#MYlU++A~t?X?neoWQh5dlj>p3%`1PMXqZhWdxows|V9}>OpxG z0(E81xV~k4wOm$2Y)&K%5igy3VlI*d@_v?F5R6hSgFDx*qiJxV163;QpUG+#QyF8x zK24ae!I~>l6s3|`aGeJs_wl7@QYnd;7#_QV<4PUBWNH?XsW8v>88E9}o-7O*BZh~_ zqx!6Wa6t^MwzvLxDy@nOPOBOdxWa_D3<7-GP;$Mse*h1>R+Mt zL|gAB$ycI}o!6pu0kHq-`3Ues!NF!cZf>YKhcjhxJ?8R>u&>iXt#=Xp8vHRtsB2vc z6|{U`)`^3q6EyPmB3np8HVLpfa4L9X3g5>Th|79k=-|A)R!+1fFqg)Jp8;pxRERk!@kx zx)>6x`1&~11^nDo&j(yhD>WlZE`igXN!TUN*<%r+mQKpqIrMCf7T3Cm;5j}nYhgpB zsv&+}ehyy|_Je;kv}=kwA0Kby8Ul#F{)*HL z^@MGMisw<3Qq}ywlbHf>u9WJ3)|3DD41dETkS^g$oL|jH=n-5CC=|a<@7PT%YO}+3 z9Z{>$08Z|Ilj`66rV6G1e=sPJ!JeDs>2zAhOG%_QE?}Y1GAvpX$!!rt^(mh#Kr7QV zJv=u1t`;K8mxpjZ!iNP1T9)k%t#Zfjo4Hhp^iW20##AnHtK+@hj>PIIiYv0EyG{$_ z)=x;iq#8UaV60XGZd{B`?wMnJfwM$7Z;uk+>p$bY)x_qtnf+wTrh$DgJ0t}o)9`R2 zA27U%0sgHnJDa(2Mil{prr9;_#p9tL{0S+oyy3vc@5=JFu}Q7fWD&U4>^hle6TIU1 zvZ0Y~C_7s@vh==$^R`?47x)xewSp=D%wk-OTtWqcMGmYYvh0K}q6IbT*PpjwWP)Jv z5yWnns(^F!Py380`a~Vh^86PjFi_|Cp&5>NE-@rkY_PywU*gn1JQzMgewX{re~2(d zAnC408UCIDIQI|UcNM2ENteVyBuZ^2e_rg!O3J7xuoNpRSIjC5ns%JQ6v$uiGj zE1shLJ{1O>qD1j~61x3v@3gTLD#uogpL5aqb zZtnuVC>7FpgfVs@nk&wdKb6K`bRup+@K5hdYNE+4pdOzUrrJepZ7<2(Pb{T}9$kQGmn0AQc~?^+tBsJ&7) zFsfN+dM=h&ga~Or{i#+m{_rO*=%+`fsY#}KyvCgoB+j1%($2`&&UNA9z!--EdJXeR z(SdMqt`VxfI(DsZXS&#*iMKn41NYXs$(%3J<=JvHpta3fBrnuE6ra!($LG~S-*?ms zz}d$!yBwjNoU`|Etd(S~RiRF&=2e`xf~cL&2{s_6^pSO%Jh>##B4bFllgUG+7w_5& zolPCzyL*`4T8%%yYw6eP{f<0MK?)Bj$chO0$L?RMv1|`J)(sfh2yK_$Gjy>jHdNm> z!p%{In-w53*Aj~j+#yZ=-%d99>}vNZM=(o%uZxz%yzqrU7~5$5kbSKsKO(*@o?qox zocgr>QK2Fs{ub!qh;@#xM7EoB7Kk6YlKBw}$#>6H5vqa(LxdyR87TJ?@Y%^qkvHAS z>IgyWsSo3F+G=Xj_Rp2d9ep#C6Xk&D1ON^38r_;Nq-SH&uz~7ADI7}w+Qr8c3n}%0 z@~BW=P!3(n9Y8dysd73}!9z4};{?N)7Y~)ECkX6JSro^p6P%f$dyj?)kClNM5gph9 z6mb!M_}z2HKg(fLe^RwX+jVjH@bdkORR&8`B*QJHCKs??6}zP)F%D)IL92nvL*d{f z_P$*~QCukdA;o(=ELjW1bo0!A=ot$utLVVZIl1phGy*)aXuGI7SP@lJ)9%rsgpWsV zuzJuxfmT*fp+G(&tcY=d5au^zsj$gORwztB%9$6tBU^93@K&Y{S!%a%1_P&AV&BRt zIgmXqNd*7wzsSlYDaRu64p!P%iCPbMK-Id*P-^Jg4m%ynLkwpo_iKlcwqTe^7-%4g zCCbqmIMAOt0@S&gX>x`N#jZ0C8bWcY#L%n`{Tq{U{NZ&x1Q9qR88SDLx5C=egz5^PMvyKhw5v`AR8|n)V1TgBUm#S7tAAI z%HH!DK4kb>I09s$ue#0k`$jje^wS%5Z$P&oe$A!-??nA6PM#2^1POAF`{s!&?wh8v zYCe<)Yei`w``W~>PaJFNh{|9TkP^~PMeCXZs;NS8O@6R;rCw7AbY;am#I@wraMHd} z+1Rxo0U^Ci-3We34xA?a<$GWVE`;%yMMx5h3wZ!iWXGF&bI1b8s%MHD>8wJ0wo7v@ zNFqF9i4mQ zqn@Zc5WF~fx4Cq@w)lK=A{yYG3!Rg1prY6j<}zhZf10sMj9i5#%>tINYy)7B4<&T0 ziLtkVK23smLoNXY>a7D%k(EWPKi956okhpOnt>}h8bH@O__c-c&0#o8h|(-NX2keV zPy=O7Z8PD?nevOO?GHd1hpHt9G_!`*Gj`PZ&)CJj^|T!aQWAIwEnrhTc|$z1Zkm%y zBvTovzxuI#-h?SW{T|naj}bNV9`{KRLe|itT?;>r3TdvrB99PK3@8YV$kR5UdvP)| zO$w#3n&2$bmv|&ekA!vi9GcNT)d=imMX$=uk()wg*!?t(Sj(sW}bb}$y06-pU zH&vxTa(j=(6{NNLG0OJ)yc7w5LwSekg_UDV!tI7MhxL zkn9ulp%O6UdobfYXDmciz~n8*BnxI9sl){?@T<;CZ^jDK70U52hIHH04$JE?&eDS? zOzz!w2_<_^z66p-7`=Sv=KF^sKvn(nQnupDRacE^w(I*r?%5xxNHiTMjoh}#Y%OcSQ_S7u=qUhYrK^3m2mdVLk~zJW>Dlc3hJu;FM{ zKA^*USc`9I`5by&L@*Q-;3tPBrFjyGvHoN*{e(&mi#00+iBM%ihe`ytb(3P%G4s-b z{4wpw4;4+L&|a?ot}0FaPF=9qxr7}(dm0-uWUg#9; zQ&t9mND(}qb8Bi~Yp5rT5B2M5LeKES=0D#mI-XhT{N<+ne0QJ=O@q1)cR$BE>6%mt zoO;c4HGy2BhQ3-a|5Sizw6z|;@9DnNbfEGuj$$~5&EX-9g8%TT4JPf$^p_TD_5;kz zVa6HKE{9r07mVTyr;zUwznZ6Ezmf#iIgv!cH$$zcxk_1GA_78$0d1C9EkbzzR zXbnv3QcA14At^k(9mWt%0;qrTUg!YSM_grWKo3DKz=9tePe|m2vy{?9Spn(E1=Bls zRl=}88M}$|X-8m&VSpT=w1Jy6s!K2xhczz1OEZo~wrP(C8afCM6#2P=GGK9f;!@JY zC-aX4`fddxEJ?o|l@HxNAWSxTX5pHXJJVlu=?afeOzm7r&yX~=nm1xF~?fw&D ziO?Gdzqd{1Ac2x3Z3TaQsxS|6qihLokJlg#<}{tXz0-K0LC6lx#uyg8uzuQMg3f81 z#-sYaG64PWzYPA_Qc552P+%2hO$AB2mo@p$GLrM_w4m2ucqN{MPr#FCXe6;VhIT}B zEyd=+w_it>INA?*!F{4i5aV-CwR-IuKLr5#&ODPj55#cjhH6ny?!lJE;=@C3NTDM| zBvyu&(dHgO(!--uS^nnE^LZ}8;Ubnyz=d#%MZC3SqfS|%u09uZN zqvcnG%N*uf4M1JxOdL$_;J^o=tXnu!6J)2G+q=*zk^|$TS5qNC+bEyDCSqlPASW&( z#RR#i3|(b}T7ax;9z-;4ofut}g8pLv=!19y3=j0_L56%bFl%_cP2%PS(_$7P}yc z6c?JDU=$qp%&_uB?P2_-!uf5gTZM*K3Gt(6^f%z#!r|2Izre)1zjBopSq+R*+T`~# z#+0%t(GkNULYVRJ%+v)liDdRUZsp2}eZ_k-& z3;Ongv1rMhIC1vU)ji->XXj!=l2*g4Q4CpFVYTNJ$gQP((!EsAte&j9qgUKCF~)BK z+X}pSO=#BKH>5H-a(ddH)ofxyL(JH;{}?Bov_AjoOHbnqxPyQz=Lx@tc35}5rdG=a zFJY++kJyk-t|7Jj7^t)7wEbVjH*qwxbV4&0hO=Uo**ISbSZ7Nhc_IYi4*DQHW>8eY z{sx7hSm1-w>qi_SEf9Tvd~ZTb%@QsciGW}fz!K`PMk$756cj~y>=vPSBwSfHVoJNY zxH~;1bfZEhy{_4U?(v-|bMt7}@xX;u_{bzAC+JjXXM3k91j9zuaAH1L=tXNze$^(=8C4yE8Eg60DYE?-dF{bQNhCT$y56 z$f6ogz#@qeR-}L2!&aDQ8IB?*f{A6FZz$~22!?6KncZYmHk z=tL3Ngs|`9z16WW7g{ zs7HhzQC?fdAax-0v%2@lI@0apM0}4Rwp`|#6LS4F6<1?IQ&17lJ4a59kj!oXbyN;0 zHtjShYau#_&dvq{o2a87G^C#D8*J2y5&Rt_Zj!yDh|dS{VEBNEbqn`p#ar&BrQhY* zHA-~sp$)(KIw?vsqQiDiz+qO%SPqcuB;V8GE$S7_5hg$LZMf#z)XOQS3@7^%flgGXfCC|2+ zTaecpQ$p>OPTWVtP@@a|5`Z+eD)=zlyiDWB6kqbpWRa;OL<^}uf{Y73Zc zN;azLc*WlQg!jyk*6S^iM-X+FA+`kK5N^h#^E^3Yu~1v&cPiUb`oG*K$7!WQ+I>TM z4SD{9|LL$YI zc#AJL2zT@RFR928Zml7@-$oaFWE?r$IMOi~nB?5QFdu@EB8m#{0Fh&QRHnWTP(=$$ zE%OtUnW-P`yApnXRr{>bGK(5MbbEhypm?)D?rfM1RhR;h=v^+Uj1o*zk+FPO=GroU zs4VDUc!_i5{>ZV!RV%9)swU`Jeu(FLN#HuW%9waKo!DbKQ&gaZbp7Ha-zPT>B;dIVzAFzUnlqExw(^ zz6-gi!Mp;p;K0-Sk)yuB{4a|U_2^&uIhv!x!iz|!?}!pcV;dd@_qyvEFz)28lZVno z1z%uG8qAI7JQVWI34%Hkgf8xJR8|Ou5Z^!@*l0op!d3!rU^1$C2F6j1v9rTM<_(W0 zbCuY<+`tP=++#f6p9?gZP*n=obH;ry{rwHdf^U^PWpQi)p^(5^+_v!orSCthD~Cdv ziYniK^#TqK)yJ>Dk9Ry)Udktn14~8OTpV77`bTD4h8Og>kseDTj+45g38h8Gh)zt&zWKvWag+iW>Gah$1~L{O^3O6@68$LAZ!4 zym!AHXJSA!#jJ0G+AvbfKpr+KL=GBjYU7j>ai|>^S~O^BL$x_a*9|H*d>0X@owZOJ z>^Ne_x!i%3%XyYPZO7Ti{X8a7OxRI1<`PRb1o-&g0M{(|ati9=i=@K|f-c(*g)$Ze zP4EnN_l%|ZBf{&UT~zw$BYL$-Uxf$Rnj?$Kfmd^-dJSB_nt7PAp-=|h-j2Iu1WXbA z#_|lT{k3-l=o@Nfx~~!N(w}5Z4(JwD4G`b{6wKa?PZ$tW<1W(onve%Gyy)G=r^I~8 zu;hKeuRYJ8SWP*9bo4yGJCwjkVs6mYGF=OzBtwaV=gM<6^3(C4Qbzem469m1IWaRc zPVMCUu2b&OTelZ3zgE4d(kTJ6xC9IVqo`0~^e=iRk|54d7E65;HYT}6>e=qs(gjR9 zq0!P6mEa(J4IP`Hi-1{|{Nf_UuCk9&FcG?sL7@ewTE0g`2lK~Y{c(TZ%21NAeUVfY zKuA=e{cC+&j|-ojrhP%f*&|eBt*X% zkv+luFt2?hhOvCtGCU(<405?JpQFJqr#!(6azXLxRC zb`wk@&$$+E2U8LL7`k~Jf!PJswl7&mg(Pi9)JmFQdwr5HoCijcV74Vyn|5om)j62< zatrGnT2ltU!3Xz!lakpzG7B(Lc}m*&Hf#$wo49nQ!O~C0xrqW>vT4#VyL%!08cVl5 zt9LF1qe;%3R#=<=`%bXp;}xElBA}rM9e@R1K+cZJFG3}w<-~i)xs(&M?!%HB5M)P> zs*wu6aowu-+j^H2RX?5F|ATWTeqx^d1+g**(L1#Aqs-CFoy4EPZr3qJ@EP^m(CSZv zAZsl6g>qoAK)`4K#?h1D1hz#vYx5)WH%BzcrvnMYMX#LVTUaql#uTA6UzLCp-FdPV zRM8g}-J)DAC8|p7x^rXW4b}82@+f4Ldxg!ksCjgcG5DY|Iaq5fB)nGmW~r{Jw8CQa zA2&n6WO>Kyx8Hsidr&oioC3Y(NGI1nP>$n0kh?1UwSQ>kDy5`2f<6lexNO?l_YfS; z3ksp{MeaMhx-b&}VFSrdjGIS>!a|Wt$xC$v`DdVAl73d%Xaa! zjODx8mm97h)kl@Kom1!^iJ4>y9yYtQ4E;u@RkjlFTX0zI47q?V-?$*6Crq7Erl%lL z=Y7|9HLp;;UCJ@aDq>`@nF_*5m3uobR@9R=*hbSp= z`Ym`UIy4x%fcVRDIZ>^ccY+aD*_ST^FQmKXL2tNeEC}nIHVH`V1O?PZWis<;S-v#c zXGx$w(WY+S*ns@Mr-!1K!>jEU#_X`i%#y?f2;Pu%jsnYHPacoY{dWd+sW!=@LJJw% zd0*!qO6$?>yM&)$M&>aV3 zkIE>2n|mjJ-gKH4tqQ7;aaeikQ2dxSmWA1&ygqmaK6elP% zx2s=F1UNB&baOCB9GwClZx;ltY|)eaUaaE{0Cu-oWW%$!+zZspi}S*S^1ZLA0D%2- zoZ-NBYU7)x`)MS=z-k0ojF5RqZk%d|{!xS*;@XGua#V^BP*!v@J*+%n#=^h<@SuK7 znR;01-tG$?n?#arI|v=Y!xzZmdQw4lP-l#)d1N{$tQi{qh*D66Yxqt0oMc9;^n2-i zpg1Vd_Q2_yKjcY+_-%#x0{ew46~>6mO!G@If=FV(gO%W$V-@kVzSQ6Hl89Tgh}2vZ zn`-XB-}l1B&wk-*8^W}BLP91C7e52$hhOII{0-?8M{H(nE_rYg?1%bWeI*?P0Z38v z&hmv>k0Rr>M~r(D)6L-M0TZ)A;tdN^y6fe{oj+K;m@!X6OfMEk$Hp$RDW?|vuP%oLgFG7rDvaAf0&Fc4ykWDjRZrk>Y|%t8iQsd3@|n$ zSCaWkE8GZOWekTgqR!U%Lo$|ZkHdcS%6^DN+UOi(y+M%4E1idE#QAqD5y$!RiZF^r z!;o5NTalv$o^TR0hwz(Zkt#kEQt^!5zEP`U&X*J?C5gBnC;d6bhwIeQnYdg{C2_H} zPPc2sIgmD`0dvnCW6B$uVI7PjM2r#Mc=x7QbL@&k6eF|ef_B*KaY!XAdKsP`vQU!h z<4-(z_ttGgMmy{?^O+)^buS%7>TTtorb#bL6Qy>YTUiwIUeu9=2~&0o_l zKgF$e_IQ0P`M9=(*!;!etD;whx7Qr(TK6S0ajNeXYAKC|*AZieehW zR7DH9?JrCY<}+m<@w>+%Keb-uXK9@ELC#4HC(w&%4wHO7I7qKKz*I%8-oK1|vf62! zlF%GBeCtv_D7E*5(Hs`N4kcjI*{ugpFq>Vtn%cJc{4jiKZJ@F0ldq)OH6x^AHY?C> zJ>MlZ;17z)>Py_bx>JCpBTC3juFLgy64?4;_J3!0B+m=ZTwvj;|K-X&y1og+S^rC9 ztb3G$%~(H^=k+M6;~J^?PG7LBr9%CRiR!RB`uXHt=bWRP32wJArUqo~obUP)uQ8u# zOXi`i9{y;+D|zsHN>+LFbF%%e#BtI0Cyhxpm5m`0yP3H^qH*4_+wRs3zp!dnaqad= zIoBj5OoTNs^W6t~e_~g*tbs&6%1ov*V|BKE)~W1c7nsty%6YskrhN0~Dn+(w~I%@3QGz^ZF*~*Y0=AJLYqg>`T06xzKiD6IhZ1bLFzKg(3nm+K!KER<~+`oCC zdQ&gJt*F!03S$LT)}I*Lm*G#CCnr=43`2SAm3<2y6zI^DSQW3#3VintU&?#;0^H@? zDOel`RJZaPHVo)@m)E*DgNV``{TNscb&ZI*!X6U*=*n#)$eA!i-UuBKE(Uo5R`5=t zZP&pxK4tKk)K%N{!9(joR7jTPYqb-MjEj@UFHILWalcaNHSh+BwLJQWz1B!&~%?UBq)Yp|F@-;_0bFI&9w5F%KkL!gCQSRIiw};7Yh7X)2r~c zft_WPH_${7pucOoRiHDOV1G9Jvh(EzZJ^bk;4$A>#=(%bb3Z<|1A~bxG$xDg^^qd* zYXrFUK#k%pPxaG=Z?^p$6@@`jrm#jq@di$htY}fVe=cas@#qv?; zBDBbjYyt}43v6!bW}ToJ4$1PC-MXvd3jS*J!!Cxv^_ZzA(58sW-k58c3@W$8sf6ap z-b;b0bb7NsC&c}B7I)+>tgOR$+o5TEc-z$*eebst{grYn*a-#bnP>(&EcBWN;G;1` z%t={{sCKllUMGU1LTfvPZy$SyzoWzj$L%?xWnv?-x0(s^(P-yp*~AoA5dY8Sb&Ms# z?kHT8>wxyfdgYD7VGv2SS|v%GG{0Mn)N_krjIcjZYt4~d2}Jp_ENdfhqWImmBzUX+ z+mx{120BeC1_cMZLk;!UP(xxBKuwnQ-Uy6~6Jg2Bs9mGO;(?&O@q*91_%qBey{GS0 z8(~@^7g0ZRD`w(cXBWb~`90!%U~hQ+H$Qi(rK&}v`1?F@bj^v731->vC&vBYhuy7Q zV0LCXPIc1giFds=da^6R>RVckK{ftU^$FeIfe^12Dq{gp+7j=2TbbfgNJV@(Hx{uv zdhB4s=AjVkvXa9|wjz3Ud)ifM$=0N5h|PoJX8FU}@D!sW#)?ZIrRaqWr7mYgWFAgOkWn!{i9HS|q!ZTrogfAfrJ zzJF2<$1Mwf*VMd?f2BPlFp}5{UDk1ZT+TDfd{fauvPUfHK?yiP8J?$PK!k&EHz zE}N@G+r_z}Y3Fy_AD1%lm1QfOV!KtPNz%7i`SVJY(${8~7mm|%QK^V{7UtFg9Vz!r zPcusTnIl~%#ds%T43|Qeu?GuiunEy2*bd9NG^b&QJ#6C>r#~8 zxzqz+rS|;P_{wvk7zfX(26^oJPjkzKbt)z|ehG^SxX1U_i$yTY9TLf=d$)AlF&7%W z%ZCDf^XSKCx=Ad3&oFqyEhA(2kYFb{NY+J9QuUbss-TzzlOVKu=>2OezexTND#P! z8&bH0w47d6>>koS`8OOEE{l&frc!cFs$b6(ueNq&3oJNt^1Tf!i)I4}O`|lgXO6el z>^T9ITSqppB1?bD;M9*( z0LR$RezFZ_O*IckKl4*#S#yt(J97eeD-8oaxp3NPU(c)yG`g~4|6G@t0{ottI0MoN zVJ|^6dXE#StKnf6Aw18qF>R6@X>5J4(3zE0yDnrRfp}vMXp`T;9CCoN&VM zy-nI1-M)V+aEZ+c3`iiyKLo%*BNp4AZb*@WKGi-R&0N9{X~@A=wGMcnpNh$3LXE8h z&zZl&pA!)oB}ol)9)3}dPXc4qNz;BXMMQYz#(Mz5PMnf*f5hz+7O9SzP%%R(9aJtC zs^F2u7422mDprVXN*y`w?7r)(m%NKVDxw|&@acyOMJybpuT~Jd5Y?9m)Gs2H+s9a& z@;}Y1e+|{3kxldpQFcxG5u;$9O>#jjMag`}KeE#7IA>siBvOhIz|B1WY^Q0X4bM=D zvAISe;{eaOX3~fk+Z1Y?Glw+R7fpW;Lu}u)Rn8_DYn(lY#4dim# zCG>f8c*1+y#-8(dPAks_78>SnQiSh6k0&imk2w~iY7A{AqT_b7qTfi%tY0`u-lovI z^-0fMObpXPdpDQXwx7xiUt$7zL#v<;{x8m$S&t?7MjlJ=BNclzuzrid% z&1AIx!Q9p-GJGIN8Z-`uFa!wM`I;tRv}h zDh>4Ch+FGR-SJTQU#UY9Q21}RWakhYN)Pz6PD%gp8DO+;Wt^fKlz9r=^3FIRvH#moGth6g-2NPk~<`LzeZ9MC!?FU?{!=#Y3V+6=xiu_y~7PO6X64hZA zwM=LP6K8cW`GgzXg%<|?^=~Y5Vb~TnJt*;=5{>DUZI>Dw zJkCteYnm2!I(cXvB~@PcXp>|yJyf1Qne$ejvDg|bRi5|}#GEq>#D(r`3?z<&8KlMi z?VA(*dm+~f7qzb?3@?To8vLx5%q@A7H&x!t)Lk$RQC*Y9bkH}jXcu3oUWsA5A3HFn z7r>n#M1Wb0#pp6^B@7EMseug7Y9#_&TOfPRRm-Ub`XxN zDEAk>vAhH`bkfa4OOyC?g$B@G{e=B3OBWRr`2NK&Kuo9=?)OO@jaa<4C-@WUook&Y z!;PEmsqPX*O46~N7!|qaiF<%-FO`~%)%JTFNQVAClwT$y*jyDWfA7Pjl}{T=!P0l$ z<9A+E;j2RKsf|%l`H7eJsl5GPLZ+b$;<1_iRjHNoY4EH6KHgAkC>bStiViXNG_R+Z zg2gK%qN|K2O&jt16lIlCJ&!G^cmD6_ZsLe(XC!LVQ~-J;S;}o1Jt`ACv!gu2{HHtn z%=F0}0t{0UI-ZLM=Wk=|$dPNaw=2%yh*SxOo#7m0>(Oj2Q!ZUyNoXdSR%Thk+v4azi6? zid8EWV!w1ONTb_4@Df6f}j%_e!jj^x?}O z3rFY|4zR#~yHOKd#gJ9M2Bif?SnbAx=l^S@)@zv*CLO72ztCtrF;rTGrZ_jS7jq6b zyH(nXE&1^?YqB>^fi)6os{F$z!83Ox zVP0)7w0>G#Si~6I@htxx?*W;0YO8++Gdp;cOJvH%L;KN#n9~ zdOfwr(;f8UP&UZ9;7fk2RXxHES~^1qi=NB?nN#JXjKkMbVX~wLS8W}OvdPMzS61N^ zs%ruaH0Y6?57jO7>UsFs_s{*-AAwww3!Ek3|pxXDZ!m{lZ?1LuSEA^Vw;HQwJAY zcHfr^#Q&7O$-{DUw?d@d-rl}-S|$gA>G>+VzrOfkgYP}chnpVqmzwciGw)d=X4=@^ zuf)lh`OsyZ9$G=%`0}ZQJdOVxG+?yln5y|ie%yH5!RajPQwg-SA}yj>s(+@VQ=$y{ z&%AZx_ZOYHUoa6>WcO`bXjuP?)d)U}{37tLJ%itZqRIclbDZ51F{W#@LP?3xem?s)`u->9=VbWmEFj4kk=v$Q2R&oBk^<~`%hHjso+0tND zAE(QIG*1)DCE5k`u;RL-FdZNN{(PXE6=q`PoEKkgUKhLT7edk~$N9 zoZ})_X^U5YiIs)Q6O7{WM#CIr*pM{msMv0QM64?1Q1fooJW5UC0xRFth4KAC^dEki zYXR7_(2K8^H*LAV;BkT9lMAJCh#h^zBZc}=m4Pfl8=+PO6had~Pb)5G<)$o-8=-g6 zezsjD$4kvwlwB|U5rL9@Blw{gTuYCY{{8SB>+gUU)KZfOEeswae+fkUITf1`@a{!K zdlG*fNnM0rX~r-bo437bUj$;h5c$(-+QR=yt~M3>>0ZJAlz*JtZH^9C`sWmPl0r(F zsC3ULnQTB}Qt9mP*Z|YgN2$;KFO2gwGQCWej%`F1a!}&4gaDRJEsg<0ie!kcjZ4cr zXreRHpKJ3%QmXO36u=v`Ng$7)R&y}E%ky{o4x{%Hg52IL7r!oI4i{wvz#fgE4v|*i zX92FK7hgi6#);?*Db)sK!8IhJUCL9b>_A~&97=cCr6L1;x0S?G15l-|y_*Y(elJ~kd9 zzgzkv0`V)8=eF<0JbJ_4L8|qVdni+!8%?QbD+yPOixc!df1O-fSwnf5;HHA~Np*u{i1 z5O-icBW_M*yCNLQT#HHi9&3-|gV$>s*HgJutLqRR5YYw&@q6-g!dJ+4ZG9FMOc+T+1_E}87N8-1dGnMPw*(VGE zCCAM#UdWSK6PFoXyaRBBQOC$fPYG>@8<>@91p^|0w7_#CKXIWSmu4m(nOa zns^O583*O@u-Mr$k*lk1CKM6NlO<9M*;bN?6drTM-~}b!SW|XrulIlGl}Qh?3QzhM zAo?;TL8}3lTVcdZpu?Wk5ptp!lChfGnL#BK$CS20_1etMWPKZ#?c*Vo>>G&DhcnJO zfga&Ug}6D6i8DOJrb4bK#7!E2sZ(`iURvO*se?RLF<#tUqXoU5Oh>sOwCC2!#Z<## zMVc*E-bQX!b_%hu?Nh`riU&yx#KS+KgKZy)oS*(M3L<(+*%+ZBvy2OPX4{O&+v(V6 zix~XaKevY4^aBC%xgFt*kH6{11Ihp|KmQf4j{MtoR`=hpk|p%CHvcxS258YPIm#W3 z>z{nQs{e$1%$XYbFbNZ9z}aa)O|7Q5@b!Ec8I-Qqw*4q`XNSdf&r)%4h`pXO9rN0& z`pfDmK%?S#*fn`8ljhN3Al_ejuaQk6ZQbI2#y`d{#y_?s}1{)Dy{9VyP!TcZ6 zFma(f`ts%cg{UTY2YykNqaKF38KS?TNbC@vwSNekxxg&=`Zy^0;(vNQ(zEpV>4l22 ztdW>rq05bCqeP-%Ba1|c*yxG;$Z=AsDtV2Xupp#aHD{XKFeHTc+HMjwxzmYH3HvII zU5t-w%SHvzmcSH{4^#t>I>inA{+_k1ydcy>q0APkC0rZYJDT=2q_4M{M$_?hm3Y&_ z%V}yGh43dP`gV3k7jO+bPyO5ogk-$n=Fo~K4#KE~-hnb+OHH4i*)c!${vcWv?7D)J z5Ms-e@X?Jh9c#;DQ=(1e5`52_`5c=apDobY#yGwexq`CG=*a#l{W*4N;fDRWm*;Bl z?k@~Zg_-WkzQ%k#SJg&{K?S=|U98IQM5&PZL| zT&}!SrUuxJF7Gylm%Q+{oEcz#Of1}7_a5lyM0Ez1jdgoisA6R$D!n%MZwWQ1=$g6J zY=q<@k;OSd)1Y+vtFVx9g(TW8ktajIA=_Z`VhCq{NqB(@7vnR(+4J)^h=crRO9TZ- z>5+#0&#kBb!`EAf#nCi8hH7AP*+|TzO*q-2T36>Rbiir|`JzeH+ z2%!ZY^HDwnVi_r;joUc+Eeuiq4Ptgjln@zl1u2^nVq``K@3sCL2mibu%ls{PTmJ8b_-`L~7Z4k72W5L4qmu9t?=~rDY%%1qG zWDF3A;ys^*WHlx6Y@UC0C<2gzzWrRwFpa<8R-44goB=D3DOb*ZwxlG4lf99)CdG0@A~ce-bO= z(W?lYx5%>t#Bag(UF+Se&=yR8i_^=V6rdB&Hu`As^>P1FrCbH7`{KEHC{FNyZN(IC zE}pmfeoN(g|LF^(%RdDo|0p({q|TR_wd6bF`iD9>RNJN`$v@H-8la5CIEo!NL6jW# z9e3sPEx?X~8CHR$7cN=LC6!o09Ye#s+vNQq5Cae9upxzI{`~EDk00g&Xe$p8kf5qP(CTSqbu?jSnUU&(*BSi`t}+ z>#YVx37qZ1*Wj}CtdM6|>(fNI7A(XRpdqt8cLzxj*eUKo%@dD!B!!x86k0fdU95++ zZ*{ImXVwStH8CVV;(mNLuN40*p#tZHMeA&fY0=q&=c zFIbnH5zkLBuYW7*t%uYSIUGdi~Jdu&i2(u>)6X|N{x+s)O0=ZEO5}?MoMwq8 z$(o{VT1pKr9`9C@-x^epc=n{(V-f`sAmPX!HUUNmmK$zZ#pcdK1D5*h>@<Qel1^#Bw^^7n$z@IKC?#_bF9t+F?@fz>cKwzPK z)DHmmAge?8-*n{d%x>Qx2;T>7BYre3`eZt7ug`d%4TXig>ysg5^jnOu?iPo&$4h^u zM}8FzNzsX7kgiBPGjT-LxEwgWC2B2wsFLFth*qL!XYSuzhHydUyphZ~QPCeFQEM6$ zDqOb1P^a8&M<6ZdtBUct3}QCp-ctF8ntq)Y13|n-Na)!e zA@*o=%0f$PHhg5)->hE+YP0R?uCCmrZHFm|N3{7)8;E;vB(e?JF$i~m_t5_F(aN?V zIvlX6UKPw4rmK$+H1`$mOygV7--!F&OWJt*cxT$T&_}W>1r*_U*mJb8UVDFd`JN7= zQpV(MVzs*QUPd9rdz@I7()ON?Pvh4`h|3&t=!1nz*)kA1<6979t{VP#=WVLtU& z_SSz`Hv%|tw1&1G61as~Lm7Z8+>lz4wD{K1w_O9cUX-uiW`8$z!en=B5K77Uv{-Z$ z#uhOQ?XdrHDrH2Nif6)@pm=k>lyFiYE7Q8oea|P%hGn+%L40@Omn_XoF>vKgS1ip1wKe z+`YU$Y*JDQrv9Mge2XG;HmyxRG@7&rGF(*usEl$2TiSiUqRvEU;SS(NLegqVUwSjA zK-S3qN?0y?eg?=1!hp8~91fKZ84Wz=W7_8guWzp47!BFPti?yeAoQ(TG zZ5T+JO=D3hY*RJuywdv*Nqa}UdP+l}4|k>n+pohA#r#Jrl=rMgJdbhR4KEc z*hLq`_GqwwfETtC!Os>m4DSeH0A10v$ihP&#wU57v4u&%4g-Mzq7mlhcOO$wvW3x+ z%DKDe-@leeKo+uG<9I?px$PU-zD#?B72e@fH#LbCeQxg0tTs}#`16YO^vCa>g+2^Eck($egDt}0m%%Hw{@dJ z>&VMF__fik<|c-|eDoc0)9X!EuP$sEWcK&+uS{1C6N9=?wjsVnwbljg>x-RAk{W17 zN03Y9(^~Yz8s%^RG;^sdkzhrnAntZMAJ?(vXPB%UpCkDYtx>6pLH}j9ipeo>d;Nb5 zu4I2ppFxE^KgYLQd7*M~p8ZUo9NNJrDZEYgR}rB>Hxy$TgTlRMXksBc+`jcm26uRG zVsyA?P>$n+FDw4X*19GCzK!p*faMaZ=N|CZ3FVAEr8h}C9q6g#IBT{8uook&j>aWh zo}Y^)Hwh_gzs1u;4)E<}e)f8zn*Dik5SrQ%A50B_-=LC8w_i~9oIpGCGT+|gTydPe zjuUA4BCw`Dy!V%RvzyxqSx5KPFT-J-JtK!?HI?%WL|aC`N6Nz2V^HnP`tt-B z>^)vuts7?Dl@q4j9sHHv-J*aErg+nxZ8y;GA$REa*20^T?MjxtTgn_@v_V%roxBqR z0oLWph+ou*OHsij4A`smu1SSX?Uz|B+%Yw*RyX1%l@>iet;37?w1M-YKkh$>BMYxm zw1h1kCW~X!pdhUQuffnCG|Do2jG6Xve3|wrYzRZ1n0m&K%G5x7yRQ`eQ+#p`sQ)s3 zi%6(|1z>)rwq^$p5p!RrJQs%dz-7}}zC%H{0~acven{nl5gy5jS!#wEVDb-J-`aO2 z@ij@2TonCA(PP_e2Q_5e3||5da|5j#TK<}07xK7={zppHEGdURX%!`5dd znkf+^+&Mlm`-|o1kCsBC>JD~KAz48v$a_bE^FuJP@k!4qSN7KabC$M|Q?9>H*%%as zj`{0LSK|^sG5Kb%8BWr%6sdZ>VGtFjAGRv%4jTrfG)ZfiapIl)GQitv(WuC%CBrUI zA20HBTZwC>!`3h3A&C;gvM8Sg@W&`d159{g($M7c*Wf#b__QG@JL9%S*ZM4ZXZwzfov<}Ma!TK`lrnUEiv)aM1p(<+ms)I&8kM4!U1DQG{ zLWN$aSPJ;wrq6+kq;>3{an?$0u>$X(%v57tJZT3Fi8fats@$$HU$lnT3QNpgQ49<;Yv zx87AG^rw4jdYZMH+IKlg{dn&i6~D`k<*no^{cAeG-fVDJCsfc={ooFyEAD2elZ3mw zNAqwmQKs%2r_OlJAsLBIkNH8DnZO4^*?7kQD`d9L# zA4b8909n1KSDzp624?MqQc9Qk;ik^qZG*k~!HB!r6^&kMd|>oO5zT-meT0PJ6~dix z1Bpxgx?*d9tr)7K&l;bjfA>z#f#VjZO=+ZkP&O<;qKwx+V<;KbS$#=)oP!s04?x{u z8_y*0Af609=hb3B9{x@SEG_bQZ#qHmx%}T^t%*>~@n#dhw)dDBt~ zT!B+>c$!PT`oEP(J8|X$CdhH-cC?@IFu7*)%JGznul{gJ^ga{e&A~~z&xo{9s4&gC zb7HA>kGTe-IoD){G zG(+>0R~{XKY`*A-%O%7oze|navTw_xlZ0(=1@u0kJ=<$et zT8Qy3Cuq-^8Q2P2O46RV+)neqKF;JUAR@s+_r6Y17lz-H-+%a#ln(};%WVTHOlH7q zkh~vZIxPt#Dfsslh62~L^l!Be`u!57-F?Osdns93`f{+)ztx^> z^-(i=ZFw2KWG7~zZm{}#x8ihgZW?iZ0ubQiz#H;)B`94aG= zl5`M35;r znvt`Bp=~kJugI}aSodSdpf!^W^6-ZS_bTP)0EBVsiRH8%{3ZhPm+4A?nY39FFO|Z7*nj2N=db`6 zZjC9aeT$(~O?-OBa-)yZfxNk-z8+Db9&DD_4y|5HhLO7r8SpAP?q3Hqk1IJnOV@EU z7KJwsk4xwqa-#fDUSV(~F1d1HX1Qn0c|<4xT$DtHIPw|bhhq9r_x4!q0T(5a4O}mZ zsZE%_ioZ^;n*B#cc6a*Ru;+4jS*@-Gge8_p` z&h6+*%ccddml7)s3vt)Gk!u*%4=LV;II)vWE-PM9HzkrVqQVU%rf1Hq48Ku_$?iFe zUup`b!VdnD*mFcwWAn0Eenoe&4_h(WX3SB$b2`HtF18)|a(~*e$n~d|p})@ab;>8x zfs11g{$B>#;&aal8jY3zm+S`*7%Tf%Tn~^=y?i{-P3ItYQTp`X0x^agZ65MTiOZ50 zf6LCEbW@p9<9E+{pP>@GkDM=_di^G{G9a+ZOF_9|XZTCNO5SzAsdq)eLwtTIM^qt= zWy+DzI~*GURJd%S-CQ-7Dyf$_WNt{Sqm=I`E-gNFAl>9a^L;sJ4kH3>#c2jWiUa9j z#!d=>T2bz==zSOxtA&Va7Tj+~?b1P6L0_5r5u00+kX2gI&IT(EIM7m7Vj^}SH(b0 zvtW@j$?s}kjPISpkR8?)AVcV1jp9?&erZWukI1?(snkPe8T}Bu-Z!ECUVW27VQj;f zat=GGqvl`tAyzWTA8jw?(eB6;tIqbI;;-Y=!CB40&6LtoWHO97w2gY!YV$8u@Bid6 zo}ZAJkAK>S)NaFLcmm;~-Z;{cwhOQ;_@Ijl!Km+Xs`koeove9xBF^5|oR_=D9ykUA ze`UL{>eq*yUL$hHK1YQ+cnPlk6!db%M7ni=wYynxCuJD`SqHZ1X=4 zVcz`(NgScAYIZq;g|cz&1Xt62MnZ9ZQq6K|!23sv;bZD;{-Y;!E(q=<&dh5Hb_2xs z!~2h0YAdGfD$Who%wnNJJwGuvI1)a@^L)CP`ZPeMXNn#VHlHIqN5^kr*nC4fela!s zlRRSvgjVNud75{kI!IN09-Yh)K>okxy@n_%BiYB zxsCKf=?|60#MOh<1FtufZ$Vu@k|=Gy4p6yV{_`0BFS&*M+L-za{0gzf_Q@;9#J%Ld z@X$O7SWP8?3y0E#%GX>Rj+iX9R7p~kSRh%p;}EIoQL2zsJ~SoS^u&9u|C{0t+eEfZ0`@xW%< zP?5qWl2E>H`!IaTEJ9HWBw@0Ny3&oR+GW1~VA7@sK6Z0K93>I)yI4@YznWT`etRD} ziA6#9zN2x6KV^$}fYLFp3#I1N+VW}1v(BnDSv(3~2od0uM8BgFq$T=mm=)6Z3shIN zv-ewYnw~HUjqKAn$X{w9Gr5PgF#JltEg`YrcItr9zj35J1{y41{e!^hlGBlxp|Vmp zIfM8IZ))p|@;Hql%8?5~8p&Q@iJJhjA^%-cBaDgESl;ZEI6M(;E6OeJ#G>G!g|q4i zJPQrPeAzRQgM(d|?EM^C(6(!_7y>DNHGtYUC(ypVdyA-V*JH_T+jg(>AZTMn#e7v$+lYe zKD|^%r6#khy~e%Ivd42}`MxT}2Y5gOtZxMV)3`yiiSb?mj%M>3+yl>Fv*0iN>xB+9 z`p}K1Plw)19`H9Tr+yIjL%}d;OLV@9DU-B z$Yc5X-vZh8VH4bF;pHZZ*t0d@iXbH!lAT@47%Y!~k@gO_Q#?K!q!8ynKZ;u0Vb*bK zoGEp(*R9`kFtcbKVaIBl+1B_oc{D$jRVvbO;=!5%(z1& zfca$(Czd6y#m}|14Op{T!;!uv>+4qM#(HDyEJbD$cMgF$PX^_f6?nvcQDo2uf7#o> zVt3>31*2XyglJGSlu#TW=Vl#{)NMA`Ml*iXd&N>M^U<@L;0JDncA4|tuR<*32etAS zQK328#$+P3v@PX(cPbr-COuphK7or|oTxR5Q{UrSMcI|uk=H{7h0|c*CpC6Zg*I*+ z?C{)&cnf61`vq=Pk#`833!^yn6MMZym^p=Z6C*?D*s3Tbu~wJx;d8hnzAT0@F6Low zy|H1L%T&I`99-c(+Ba`~Trj`CxSSf5(2F#-B4ry@l%1u0X3~hg*I>C1`7PChw*5WN zXX#Ql2FzprJp&vJwBOao1(+B!;vNQMiZn-G7B{}cG1MPBkPN;qcx8VW_$Ak2MaXvE zg$wy=1)t;f5|9M#-0R~2N7~gCUo}zmo|`yw z?g@o&5IXyIU>%EGWR7#706TZ4Me0l5^}Zz;YX(fOgSihEzzK*jyk68AgT9>iISCZ} z87B{+Me_a8^uwk!>|Up?h$^C(e$(P+bPBM!tyc&HJLSab$T^gWIEf=0UTXgYtA0~! zPv{!RfU~w^J`~OqOh~k8C0JK#cm1_k79@{}FjY;h-1 zk(~nh)*Tj^luwBUfP-gGK=e>X#niHGh&4peW%w5%ror{pbOE|DtRCm5F?_aWElUN! zT-4=(BTmFLw`-VdEAE^O{`S|m^xC1gAeO{#E^rVGB}f;Eibx#QY89#Xd0=yQ7l}OZ z2iy7&V6~LtkzX+4gN#dka-(f8Ld_36DZ_06DP`S)R$=pcSm7IYIyYS;9XkcitsNo^*zkML?7V(?DzQaE8o!kh~mmwKm^P5#C#xlp5|s7$nunJ2t2R-wY)&OWY6KlF_OxiqePnDfaK_5ZjFzU_U0>3}BG_kxRKZ zZ2C5eK_XAkIHO-}r9-^cL;D5`aA^lRn`=jSPXo?-!b0~4<#(|peJ=*dJVACddk#~d z;nWO&&{QQLB)Ua-PjUX*qVLL@hCj{#4@jJc&oZ|f@vp<>>Q^S)b8C z0P8@6V*C#!K$6cjTk&NtBMJC{yL1!yq8-!stO)x1zs!Ti0|;jyWEeUWJ{LFAr z$m>bc_9@&IWq>0|mLFW`-r^svO+dB9kHsa+Pvp#>aYr4I%FUl%lC4GS_T%*b@tNLB zJ?lbkbsj@g{7vchw39)PFD6i?GO4RGtNDV0kc!NBd2_}_yNRNrIXY}O0|-6+X!028 zisffMsVbH3lHO^%+piF?`a-f4%W(w76y&^MmAjy*LVvaam)&p7i))T8vlZU1me7GuI+yDrNa=c7k`x z1Pz@dLB=?bT7GyeMlNC(7=8B zgrE_A7^-3=_pAwob>XRKzL+U~x`mKR)JhJTp{R&GPVMZKQ`^drSA^=O4YSwp>NgNZ zY*n|P4cgw%C1ONW)d|DuIxK02N(OD!IpUM@CufK{Dyd#9+xCCkKi<9{Eju)K7###- zHOf<~EJt)`ii)xT*arLZy^*j?nH)Ot_;HYM{m4Y5L1puYq~OoAZ0@xgW@Q#$57)76 z4c}TCUotJ>gt;4}A?5OF-8xwJxkrl?ntT}oZ}F??ZzTBj-}Oq3(*Az>!4$+(Mj!!; z{Z3HLAN%En8ZV7RvdeY1;q&LUntb|WQ&j_;6RBr2SZlLa{kJMX8kK7iPK;I?31csr z6s?|LU6T8L*C28xqCLx zHX4lX0$;SHNGKl)7Tl{?ek#Li#2&O=-~=mwVe=ytV^n0(>PAqb7j(tpyYtI+mcTmWNeVheaj@Pg zr}PTs=DxdCCO4Zo;qdNJ%CU~NU_>ikX8Qsch2zPug&U2h9xk!@T~AHTq&_jI)hrkS z9V~3sI}TsDk+$C*JeM0r<68XZdQxahIwq-HKGXy8Bq5saeT42d03`83<$TFBRJ=@ds@K6rmfS>S!PF7RoZA{$}Ac+q7Kz%IP&m3o@uyz!c7RFavf zFVns8!bVkNBHX2SC}lxW1*%?~ob&ur?*ZIfXC5G>&9Xq8Q2_ujoP0IyY%Qi*Fa?vC z{t}0ThI`{7dSCYUHbr~m`BCc@u`J2}y4JTOW*N`9`uNRQ(z|g}n>8(aK*5|_)F1dE zq8`2ShJg2s=n5&#&67Np_(Sb^ewc-rKI5kj-EB04bT@-dS+>VO3+DO4&JkTgVQj@pXz4GEiCrk7TY4q>!~wAG{Y^QZbdFJ6c@i>`&|BhvMyDMI5VMd zMM_#_T8O8bUx+ema(U{l0wb#G_lAswY(DQjHi&ZQ69jSeHb_Q^(D}>{@5m#6 zc$P+2%^l|N`R^n~L&?}KQ;PH(*}*4~Ji_qnWC+`1Ci}43634On4=_(imt}O9t`iO0 zk)Q$FwJ)xicPe%!y zQ*!ZBE$n%mEk(q#*QSh+mD^sF+(#P3Q6GJYaYsd`M?muAXNCl1>25u+JLLNO`M%9P zoacsK&axx&5U*J2{*LIqRR)b8(PM9T)sm(DN?ZlxoN0lRPNngtxozlJc$;^0TMJ>R z36>hh7s%uX^5%|cEWAT`5y!L#;Slxq7sb&5c`J;(F~vfe)wle-hue{6oqNP-y;4$i zBL5Fdc`-hTk~k-9ThI~O!eF#-`j>10ajzoa;fzW(t6vZ*GtaK7Q}GaAinTGAaP&Dh zZ2Md9W%U9IpmsIKR8sdf2ZjNN6WQYb2zc=+NT*dB@3Y&6! z&_0h2h}q-Cf7T_u7@_imGE!NXZ(}Z-iz@!yq?vtbL_lb{Pr+c&qujpL52`zEfz8fn zlSrrbOnY3p5^sv3C%JC1UW061F)sc7i4M0n%IKTk*I~~aY!S8a3GKB=UMV0^)@wc@ zT!JzF0aA&FYsS{tz1N62%!Rqn4>+xG`LYW7%STHJ{zVX1s5a9Eose*(6;1nN!sKOq zFIjtM2>L%ZPnk-c@wL#;s`jdW<4&3OmcHLALao}6lUGZRW01- znMey2pW>G&vrSR`lju;j;NDtGj{R})Pkq_bt;~15J`GHpZ}Fstpe0U3Gex=9ane6E z5im-$xAVwf>$&Qx-c`R5Yn`;X80{qFLLviF^y16$WTKh5QubmWyo>2cFTauiSr3)d zsKw(Aw>9X}^~65RaOu6@HqfN=Qrmg&TZ!}aJr|CYk3WUn-1_F@eA`LbNl|;(eYalK zhw^XtzYju-{*vq{C95?tRu@4<1!MvIB^V$(UGRk=F zN*}7U!q zm&rR4Ln^_R{+GTEj59|264N($kdIn~qMG?@*?s&ISZxWndTdcx7Kh$5>t&I>2vvgehUPmn6UPr6WUTzV5nhQ3^w5|bPL!^E+4 z0HHWGZuTS=#fP9qD7Z8-Z9&95uv_!2`ORJgvcwB23#FLzYrr_OPtJ*eb}ZMcrgRWb zB4q!SbgAxRJg%mct*WLpHpQ!`#C%a->_Tb$5RYuXk3VP4;lI>>_S$TlZNc|ZC=1MY zum9z#gvqV)3y|@Dihksf-OkvXprlfMH?l7E0=7=B&b478^_RLL=7Odl%{b5Ykd0xm zelZf~N3 zabmbR&ZZ)_5aKz`Rs4u`UY~j*C^{m(wyq@a)iFJrG+`ASek~th{$!!nALTBf4o-K$ z>6IjAIKSL=H`5pQw+F9v9}A5_!egtp$;oKcd-0blE^RZV0mjyg2WaOtme~e&r>ier zHA02@8L@q8r3KDKHXMntb)Ofacy^MetYGx;{6dIRQ#loh-<1zua5N*j zlok>0&?n<4S(1Zgjx)~syZB}%%;0i;?5p!*_?b_k?JN;QPjLYW^=OI@hJ=QslMCNU z5W9t$AMr^Oo>cOOIaMoPpP6^1afQS(|8XmGp1@Z4Maj72_YgNI^!($o8@V1+j`6h8 z>bkit?9R>Y>Yj`I$W|yV+u+9JZ%3)_V*JA<{+18N>RiAzM(~ev87ixH!iMnEnd)D} zv-4@G5tc?zG{8YR5Bp~7O--J5*xvqBTWC`IuyE(m09_MV$GZd9n!}HPaRBqN0z2%T z*OgqK&0F`4ohw=aXW<~DocE(RZmMuTAPo-D$sXkEmPQ4~%v0-X#^fcAKN``7PtUoo zZp|qUEep;qowgBd)R9kVvK(19I@G>p7=&)_=N#{8Bo12+5JMd|k8O6ol-&T$k2X z_e*7x`>(r4lKXvsr_ZVvlYeOn_$h2%n4=okZ*>BJHKC<-shVsPHci^v3j zPwwCJ*>WR@xtR*eZBIo1ykoErNeD=0NSw-9-rn2V=S0yO0^$qUWb9lA33c1LBnYa! zr%CS}`S^|BZ|D92CGl0?Afw)P3QOE1a1p%~*TWV)juMhL0#Y_G?%qkgNMWTqL`UrwX&eY7i>AX^GZkta@bn7Y-M-avzvO(k6|=!vgw9dg6g_9PRP_2wYGhF`9{8ZxV*<>GMv1|DCaL&u{Odih(-K61VODA? z(L{LW105=FeFMfBc63Q_aKzOowj2d0M&DnSOmT@j(+|A_Fv@=7LGQ)q$5QBN-gF2v zYZQ=Oz@QXiaGE}*bYxv}zhma#V=JiTSF#c2L~pw43V}#vY9*SXtz(ZMoPVm^%ZFlQ z<1_;ST_SyNZ}}|(5g1y+I!<)4?CF{e&}?bK+2`YZbZ+^O<(qz{LL=fHk}pk*r3z-@ zy*j`;f($85z|S)8JzU=A_!~7r9Tfu%TeYykE&7|I>$w#NJes*!(#QfX7^=smzzh(h ztbrugXuw<=o|6+^y#L1kCzNG%z*cr(j*@XkT}^}Wos9RE4Qoe_!GbsScNxMLL!XYN zzQ3dMqhu^@i%+<=q<=hE&6;Wa%j(Ik~YNM3r!mraH`@iU56>f{>^y!iRU$-e?VLv=kKH+ozsni>s++XwiWw7RLT zden*v5A%J>@da#b=Gd3|JYomUn{b&)r4rq-QBi8y4n^8E(^FH|}r zSX%e;7|)P=KCnaoszU+Xu_q{kche2kREvs1KIrFXNtn-&-+f_??dTgS zKe#D0`!mQST&(6RFK= zzyBx;|I7e;5k8WY?BGJ&ZOH)D@2q(|X@8pR1UoPr!kNYU*4EE@o&oM;O0qa)a&h$) z>*jvygneYrEUe+g#`M;$)o)JIeGW9v6y}s_hHbNyjXC*MgL7=^{T*NqpEtFbLM>#+KRLv8T$1}{y zL@{d-LO10{2PzE0W}N=$YtB{C5-E}7D3dIBJh2NKjl^ngjDLsz0W7~v{xC>jCm+)S z*SK~~dsi|x1Wqj-LpEqxQzVYM&Ai4#vK`rv*mZ&P{MOV7)ndSuzX>F$(BTgS!KZTi zwO1+PZL3wr#KJ9&$J~;6TqR>7NIIz1t!>vL_e&QsyQH|4pu^eW_E=$-u&xSoCZqz^ zU|fK6Wmg>C4Grt|BcurL7qyR#qsm+7uL*bfM^Pe=_~zIQJocNZkA29@+cBt7*_9iS!&=DQ#n zo=Fg9@#D$hgBwgxlbHHlVEWwHC&pPNeD-I>_@M?3<^;eRe*jK6ka*#aPRAiJj+qSp zzMzwu)m~#yt|3xZ%s~aNcP^M(PvxM3AAFBv#Y_Zf5*&l@ZM{~5el(Vm zcO*1P$D9$)@95rbDZl~e3IB}@jQ8*OkPSf0Y9Ew|9K~Y^o91sQ?YW8-y@_4>!(*8= zWq+=tOm%v?jts&rZPsz9SJ$gNePNVzu`m>U zg6$_D#w8v<8!Tn9b&uWH`iQ0bLO-SJ^}kQM!VP63Z7E6_2^Z=1MAmwvJq3&PpU{fl z+l`#pl;fY`JoH@V3SI!Q56hjhq>i~7hSqN{uP2oz7?ta$r_IifA$(MfBOs5geBtbl}D z!tMM3O*QC@fgq?kC^w9YV7o>~S0J^kHYUgvNzQAbe;NsUNn&t~mJtc-c^i29X)vE2 z8Y|U?*?KJzCf~H!8=&EOL2Sn!6Y>lsc{u*p3Z2_0_2u?KQw?oaQ*_d%)F>53@X5$P z6_B3+p z?$m|cgUQQtO=~xDSRQ z?>Kmo-zO>0@@Hh(K_4Nh(5bP9oIevNK>QpXp%t3OX3X3LHAzxC`Q9krwCyUQ_k<#3 zh&$AjpNeyTOy0a8uILBF15nR;>-MH*Wpesg@-4PbwHnD^9T47<)_=&4n67`1S%3Cn z{b8fuH>H4``vh1>3U5=}E6GM5F(;g`E-gTF`F#p}t>1N9&U3~va>$)|J|A=0Yt@2imQP=eWK|3beGN)2)FfJp6*-(vzm!ZYU!uhFU8`g zX-|-4Hp&~MZmsFKchb`zwjp||t=cW*AR*Jl8Dd0l9sBrkzZ=Lm-8k{&bl%X(F)_}9 zQYThRMen|9VoLe=Vrtzb7i=FELp4?Wqf2e~m|Qd@@)Pn45XD4ZsEQNfXY;=Fox%QO zLy_f^2o{vlI1SjrX;8tJNU49wAxVtk_FNGbQbcQ3Qsqb?H{jw0x}w73VrAKalo$c| zL%lOd(k`a-?F=v#Xh@zQWjmscZVlA_AuWkwolyS0TF+yta%R~L2&NDbdwofY)AQWd z?mQ>fpe1>hlKD7*=B@&7x-Ojk%85PuhIKpDt#!n_I4Ld)tR+~ALHGk;xlwgO6ozvp zSSE>hx|2-24RmQqBnj5Hv3LI}(UQ={m~5zH4=c~ALF;Mv1}Xr3medHiFb@Y2AP14_ zoWi`?z1E{8d03YfB5Gbx)Ut^IFJoU~5IeKlclUj)ym(9BGNT6lwY3u&h!zJtm)D?MMJO!pDs+}F3u1?kn4=|F!p=8@3pi^S zmgA&uoL>vCE1&7mlGL0*hrX{qWWBp zCC0hOLIIHPv1r|=2uh|1k-k=;3w--xHb6S=(?86q;DUzX=zpHCm%=ZtAHM@U zxM)DfFX-=@vOeG{3ek!RP z5M{e!CK2bIugz-uZ(Yz7spFF<$(%c9R&XIVvC3ea)SKt>e9SpLO2x)?5?Qfta)OCpoyS@W`nD#)RlubUk zj5{^CQ!U`bsh-LS8v^y}E#BLm4YboQk#Bi5EjTZCN-&jvZl+kSMZ7QPxQr?rQ~#8? zX3#ayM2;quRwRtfp)1Rb(^XlFv&$$ft8X9HQ=!#74_>S7Kz|j@$h!ZS$wrV&syUQ0 zpR?Y(4kT9J9FrlTC-}mHBD~b!-kf5~epIZ<^_NPZv00GfNU>Ucon)8pd9_@98d+6Z z(L=vAsq=imL3+)CtmHwFYoS|Ok(w*h-U~6CPi*qv{n_o~B3;L@SYDL$-QHkc{!I~& z^m7?EVZSuAn2(o|S7D3&1@-a2el0ko_0wAos*<-ct`73Cl}E5{LE6%a#C^&*9Xiy` z^c^#yJl(5DD5|ZKK46hjw#htZc{MLThm&Q*Zcv7bB%@IN#!s<}pHNC+cAj2&M>Hkw z8kzD2e~DRMk}0=)N74cnq0bG`tsU>^5lgAO*HfI@TX2ry+GixLnNViSHo+WiELRQ+ zm!BEzJfuj=KjNgP(ichfTfPf2yd+kxrYr((uS0&mEQYB+Clx&_S;Hu+u8tBB9si_l z%=3ZJ%Mf83Ci2pkn~aK^b%GShc9-uIVer@^on4@8-3PJ?Mcd+dukQNy#yRQuh7^u$!kuTE00B;Qps~V`xdi8EhCz?KelrFdaDN@@Miia01UdMgLQi$~u@swl8%6 z7>V3ZWhO=xow8QXYxJ*Auvy5Q&l-n0Uz4H$#KJur?hXo|^;j}=4M(s(Z6Edb^>JZY z(WP!KE}*7#qb0uU;Gf>vU50r?W#?*KZ&nYy;S7Ak@e|$|3&st+6-wLx#GHA^^A9WJ zU3>Qm-}`LUyq&N+vLmUSA(cGVFEET$(Pw1N$;i`yNBGW>s}@RlyqG>iFarVWDlm05 zPB5%gWk8^a(n-x%D()4>H+t;*I(8pYq)u0Y#K%}*T=f-jwBb*M@nTc3612htrr3Bf zi20c(OHGkhlgb%pQ21M0ppWPJMC8NI;*!KtWUkPYSyF-fh*##`cY87Axv1hXp^|qX zOO&O(d#G`JVCugm;^V~eb5%yxH~aar!OKXGCw<9SIX5d^Mhcde;^_w-Pt1d(PdWutM1=;RNDGBxJ(`E=v?N@m>-L} zGaIA4ve7Ndfe6#@_bd3`B1?A%ovnnP2n${b(!Di>r2j+MTL887MB$@&aCZxZV#TFJ zOK@6>yF;N^1$PTt+*&9_gS!@|xI=L-?oM$Fk(b}!-pqS%<|UIm$ak+wj9;4lH~L?n zyfoVOV;y?G%mBX7E)?X$Py$ki~7gAZrpFu^1=mqVP)+`ZGKF>Al)!wRz zP`N*4dsIufNQB5X3Z3Y($^N|5d%9re)PaS$K51*{6%J)rl*G5X>t1@1)F$q7ra9r0 zj|d!A)vRL5IP#iA^b6VNo;%Zj$0%IMR_T+#J*-n>Y`~l_dElJbdN{VTKGM><(sYb6 z;s{iF9cS2H#r9~vviXqcki|MvpN}zNy(C(h_Ta#4#QK^&?WHGgbn^xI>)n5N@;V{y zw$HoP26!(^L%|wg7Qmf&*S3g^f4niA893lh%jYyS;Lc6ibj zWFr<+eV6?M?Y|soo=W!GHWl`j(`DlC%vU#x^gP??;+$_bJkLQ1HD-{EVgsi60wNXxRbTrSYlhW+le?R6& zzMQFNqh1J#TIPB1Ju+WXblLgn-8*Tubr8Ekwlbb&KB=lmqLBw2gJp^vwr9rMHgr zer@_KsLx$PSv|Rs134-p)87&5!7Bs#FU_it&~?;2bZ;59c+~Av{pG4yQ!6V1{EMN1 z?gURy>{`;24gKJC^n@)tza0UxMH%H#yID3z0uEqd93nNGnqB3l-gTmi0c-3`-dMG- zzI+^Y(ktEtq5@tNCK3VOaLrVo4AFgh99pmX1zLrvy%@NP>Np2@G{;1Mvph%*se?Pc zqRPYdBtus4T??JvaZ>bR4e@RCDV?dDz7N&EosOf|Kf%yZbLMY$N=Pd}> z!}VGB3hOvEVx*k}Z0+zl5tZTKWuT-l!{7f>|C*y82%gp|W$?i$YEa~t>(q`-7n3;gdOqimK)(U6ne2w9sk>2OY9-vZLP&-GTJ09+J&fP)r6If5j$wI{;J06 zHR~pI)*hLxSrfX(c_&>igC- z#R*TC#7m7~2mOcBz6tVuNTLvWcq34fVNgtP(MzPp=~JgavCPOEC3n~uxij<*_5mc-?|3aK&ePlP06XV1@_=OTiad%3_KNWWA=W|uhSvocW3ekqGScZ8U zMg%S0a|9$l3$ci9X)p(wGy2j^^N{f{I}3zc3W^pR{HN6vps1m9ETWg#xeI8$te51f z6raO%t9b~}bBv?K_ly@X#Y6KzjpQg5TFOs>_TomhKaTEistc6@_Www<4m1=0FR`4) zAJU6S_^vb+r4>6J&9Vi9M^yWdzHEnM;w}fF=q_vA%4A3{7*JbdF*F*i58HSJz@TAl zsC@T1iqq2ZS5oBWA-(}9Yj`iYq8UYendI23j#U$m`j)sV3c_9LIgB*yB8?$~Jqe?@ zdPjVp6uSA9L5!~yN)>CRqNhpDU5t0P0|ll@7ywsdXuLdRUtsPIGQ{j}uE+|22?W~& zXg^b@lv+8YA*9-10}zbuXBWWAIND*BMBuF3aJUK}41NC=*AhpzD*TS_Mi2yO=1dVr zsnk;8m%9RgevwZX1Akcb%-(Lk3ZEw748PVyqvXj$F`TqjS|I+C_bj5-@{rjgd=2MG z^x<0nAD4?RbI@w|eUhojV|L(rN;JWVngr)IH$9&Hkq%qaLgNTzy@ow%ZKgoB%+Dkj$yYd*}B|OToH2f=wezVXr%`@=Cd#8ry)C(pQHwPSj zU&-SWa%lPn86ug8^OUHaQ#xeHEenqq<3F&u!{oaDQ%2zoKO?;eDab3s9TJR_{Hq%U zB+?7=J&(!u7B&64lk!#Bx9FkD;E3lpda5OJi}R%Db3XZ9!d{B#rq^!wk5{O?@NSu> zhr$$SUMX*;e%i^|6uJI{wl7gc=x14M+bUJ5l59zDf5yjPw%FHkivfxR$cG^G0UELaP@euFF1|E0=Ylgh?&+X@a@6x{oRR$Tg zZ2U2z;GY`;si!>dMR)bykCPSwo1|lw$<$D@vt|5_w8<pb z9BR5vV6_E_$9#D9JNL~Is8R!2Y&!*3|0;4D zXs8s_>V@38W^pGARtG71-T{fXbOL%l&y)P_P37gvy#ubN-2+V?BV+mQaLV7<)P*#yxCoa`zgaUV0yG*Z| zjEyS7af_&L1E=-d-(yf*_ApSM)PY;)vhSUw6y^Y^fa;0l{AR~p-65w$fpHH6i#jjt6lh(VR&E~y5;r%!m$4j6rgw8hkSx=lcIgs0M8C4#x){`KvriDfO^JSu&k=^lnCB+2C4 zLT9{qLH2P9=@e7wfH2u0Iom|(heV>}`HLx7`?_C9pH9#*IWxjMgX@ z193l-h^rHB)|7_>f%uhA8G4)rE~Gt~kVewxP7dC}+dh$kZ)pr#2@5PVWEvimQrkUO z*PK$};Dx@LTWTTGo4y}m>R`gJ;=38DQGO0hcU$IwG^9}^hjK6g$A6R{gV@JkFh3MOE z2gnnDykm9t z{RWv+ap==P!jwG%)^r6k+= zFHaOnxFjvA3)n4Gx}IDdFITk^`^rO!q(wqAX0QK8VxE2?u=g$DEH~ou{t<+893hA* z)b+;S09#Dxf$K^4Il{Px29Hb|&K(Z6kPcHN^iP3v>yWuqReOW~L}3np4`VC};<}s^ z22x10&!R*MGR>(i4$x4AUqC!N!2gaJWIHWx5I?+h2lw(O#iLe-FD890DI_{{|H~UK zHjP2N&wbn1`yHgDc1ytG%%}=7yYQPyu}+a=&Pa#w-&~0$&q@^mbC#vq%E-;xCM)Mi ziXI({!Dtfgi#0%U#N(c3;Q@yo9%BlCs`=>~aF z%Hu&9*cA*96o^ft^T*iENHNOa7A0y+L?+I^SoH$=gnG2GrO_lN_n3g~3|w8aWb zCRQ{7#dZ0!4C~>_oDG|BKq>#a6Ss>(I-;}xXaU0-&S!6T3k?zv$GC+=UXQ@IlT65 z=#5C!7N^y=;3LnlIs7T%RE&Sva4EF`rxy4k;t+E%K~eIVVn>zo@(+*J8m$YW|_7&_Ta2?MYS^Ii6E7tjUlmTL-K5i4D zPUZZ}VY)znZa@qb@D!!lhhsE^VD3ayZ7)gFdd6%)&ax^%Q7;MQYt97~0>Z5Tu~XQq z9B#MnU%^dd7|CT;KOPVC%}rRkpP^h#_6Xhy7AD#}*ymSsz4`Xq5n=46@}`gU7SIO* zA1{Zandiwr*!z~%0u+q#`ato!R8fzph+{_lyVj+qJ<>2_L#PA?V~*a^^ZsT0)O5{h z3XBiytBl$M#qWJxf++E1ERD#;JChJd*Yt2CF{-9VY9@2BKt%7EG z;&BiS6-;;ANcjN}Y#w4%Y!uQ$$*I?dF%AOpJlJ1Qwq(Trq>2)AEoUD6_{b>_JYeX( zNJReUP1|mQS4U1|zPbJJG15FFs#O+tlQ#UuHw8`08^@8PZ~%evpxxq4^$}Eszx>PkOCvHl4c#N zu!YRU#evh*{gWD-R=s>ZGx+`YW*A*yrQGNJf9T$$>+dF}{vh>MA06TuM#dzQe#h^_ z%t?lD@;&>PX%T-;p5T`rzBD@Yg%yqMSEud2DZaXK=+#{h32PL_T(;;uV9Qn zQ*u{Rap4atAS9(c>Wet|2FP{a=%crQ6+Gy;o`w9MK@2_T06XsYQ#TEHxNh%^w6w_tyj8*Vr^-Yi* zKAv`h)tFD*yOHdZ17JraKRc<9iTrcewVe#_V66f7v!I1fw#{{It1O$`c*z%<2U4;! z=jSg_u_LfOnkM+Ld{pJ78$=6Dxj#@O=%_}KBU=tnOwpP2h)V>=3{<0L@DF?ZqQPAM z-RkgiwRcJ$Rie=2dPF>Q)y0a9_Nfn|K=l<<6)$-kNVGTzP)k9~9&H%vaN=X-Z4|T8 zb}^YEB-0f5Eru$CgV^Q74i>0McH@LTA{PAX=XTicB`JyVmhd<$VqDY-DuxO4*IK0A-&q2CAahqex9HEy{ zJgkO`1;}=h$>@_KOwWe{%kL=L?lFiQ>9H3xGH{A}dynNCUOopdBs&&#l0ko#0oFh) zIFltaS5@qBhqe)Ua$sytB;IjzE?trrtl-jRS(MqET6u zOAlHlp^2X1&`rUTh$JMo);j?1DJklhN-sKEm4Q*B^oa3omUiSg!aE)h_6^^ zEU5RF_k8xgUj*De-g`LO2-rK?3Fvw|de}IK@_7mJqkih^5upJRtN?axn$mW4KD2h8 zJ_Km!ShwhCXsA-r&`vwNjr10w2zcddk+E(qWuAO=>zNj zca^Q@lbwql^rFyWliO{j)5TOOcGzOySQ9^+_6V&{-OJP1;IU#@!8bh)&1u0eD?UAXIk^n zs`N+P=9;cHo>xzZHjX<%+nRh40CjeaN5os~QBR2+SOhD3!1@QrS*A=UV5Vi=#!Jno z{of_0mYSXf&4xM?98D&(wj&Nyh0ncw9P7aB zOTTkVeU(4H;`ozsUY%F#L7naDc3+*G=;`z$3k~Q$nt3`_|3|eVWc0V8u~;JCbsZ0I zy<;LM$@Sy;`elsVxjC!w{N*0j1<_e@jvxHfRBF;`a*6XCWZ(X#p?F`H;dy`oxXqre z5o?Tp)-EtH_R;ngGB$_PumR*>@^|~V4|t+Gx&Gxn@(RGs(rXC!QYw(h8~?Zz%-$Qb z&m)y3u%i5N61vgpfsp(C{Nksj#2n+*{+x9Rd{nru&|?*ZG{~QiD#m!0U5&;?;$= zA`8E1fk2xS&EQId5FV+opw@b)WJz4aL(fo2LdJJ#+JrH@Pt!_)BBIwfqtq2`W`R<$ zEu{tIzed-W#waa|hl%Q*Ij66cs6(WBJ}455Xc1!F*=f0V&Y^U9DD|{Wiqi?^`*X?e z^I(YU`XSKEN(V8+x2O!_3v!{?qr0lj+qwRVcQHPebeupwdiv|^)$N857N`!N*YgYH zRA-##NaaoX{lvQdsipc2&dT-Jtj>-v70xQyV&BaAk3Qd`ulQ(_dBw)7U2U7-o5Jcr ztgtiN;Ud03O@F{}!jrLc(lG z`fy~()uzy%*Ha<(GMI))S_Fb+{LI4sf*`Tc1AYu}&wJ5lVcIqyhGCnL7WGKC$r;#4 z2CWQeCa#LXcPPly&5xWh$MhFRN@NwB--#jUGnU0+-s$Rs5F)Jl zY^LkSul1tWA)rnhQc{WwY?wr2E?8siDfrpzgs~U1PAg0D3jRI$$6>0rROMK?&fs;zjRDQOuG$54K4{{!0^-f$bH#1svTs$OU)?X-g zK6OGnt{k@<4kGAAn7v#)B#)6>j{gSsc$c#C9@iPWNsYuoGpcCLG!m}=C9)do` ze?Wn@Fz&5E`!VkE(AehP{t=Zr_55R5ts4?%;vEIA)jA?Eg_;wim}wGgisO8}OJ1#0 zF1}YyZkzqoLC4EvKR@Vi8Cb9?^@sIdqf1f}F`a5|xMZy2Z?kT zor!n+4yf1~`!@dLxNq;9U~|C1Yl?rUqa0kPxtZwy8O$dzk{c6_+dtsOWZ4I>UbVG0~gnu8CC& zQ#~!PzOXO0)9SF{e#n~1@~*4FQ+BY2U1=_4(JS$f>vdC^(8YnK9MO+XaFrn3I-7jy z@vQdAepQN&Lb8kD^5NLYyte4i(~^C&Chw=(%0H3Q0$Zc&(Q?|rq0oPM8s^lg6!(`^ zWHFYol@q`YGS5{`sK$nN9TXy@QL77kBK}*iR2_5vnyq0<5YMw3Z-uWo52CYqjWbQT zKC#pbgikUo`7REZkQ0k@eLVNE^XNnW{SWf9#=}&T@nR}&l76g89~Y2yuXIkpP4Hqn z_)4*BnIQ7E2i>q?t;5BqOH{fpd^uK`u%$ga!DV__j?RDQQkAY`x`>4hH8#?h=#cy7xRVnG-xsE@Bi7MLt zYD!_>Wgpk|azB^!Q3qsBfqV$$B03GVkfPl?KlM^hfB~7s;EVnbd8o^2g$ADj+$IWb z@5XHgmH8?p)!s%mB61fX@uaKe+i6ic7&*%@p1GWa>YoC1S62X7|}{`iFR!!dtNfkZbGPJfqz zFI2e&mHoBKjmwj(a{L>#M}HaIb9PWCK1v-1Wx%sDThM$ zu6vt{!a~#gp2H_G#-f>z=ouJq))^NkUsflazW0gY%bY&5t*gE3{VLOI+s3%ycXY{ zT`4h8vYqQu2te#QDdQ}Hn4fLd{>u!80=jrvZcq=_)?}+(f!S^*^~JiC-5XR$$P4o? zVz*uQmAkIeLe*dm{=uE$_B3n*E8RXbgKA+tyDK$c$gtRf8|+i;4oWT$+tG9 z4C77+>STe~m)T%|0AMqEO`^VfqhgBR$J#iNITxYr+P#UWd5hc(Cawmn?)XZ)$VHS% zQr1v@C`K$>8I;lx`|#U){vO8!p7oQ7-OBt9>i+%0C#n?8+)+Uv#`X~kP>EB9UrS#T zGHplF3ji&aHP{w*p^+Tr&6DRPUzjN)*gAu}?oYqU$%${S82;7Ff=QIZtqbjv}* zHT$DD)j8;ePGG-#TS*Y2ecEge^ib_~)=cF8y^AKt=q7@%0RP4*A@UOTn29k>*0uRQ zJn`-<3<@A&dp8Rbx9Vy-k^Dy+VyK$1x&$9y9kpXjI)QkvxOC^;^XDS)`mX)pR|8T= zS_a}nQ`^ton@-^EU=4T}+h8t2qOMdjhn6r_rlWmYI-?YtV>ui26g45fJun_P(-hAW z%Jz@`_INLnb>5}_Ekb`MCV@Vb&4B)~?(p#ay-WY|Jn$iDYhVz>q$&Y(kLDow)Vgz-2Y1@XS)!@_^4TS^OT}*MHq&>L zew$I*#QaW>wUg(>xC4^Xi8A%yh{CAz|H;rsakARF?`_O}Absv)py(3WLx}$d#u4OK-{k?* z?PT7vg*KJr$Cw>kz(Svnk`C*EB`rAW2h}1 z2R{7z1pUIk6>1+^GX>f}EcgwLeG7&jY@QxtQyO6jc~6}EY%G`f{=DN^fY(j)9|kqg z*b6OagGmH*h&1o!OMUU0avu1VqEHE9x<>H1E8P8WpS89;`t4-)wQj|zb})3PW+9+m zwh*kdPx5c#HnBvw@IYt|@AO|0v$kp3Hei6j2NmAhbO}UrI@vzP2 z$gKim>(6mqn4#wRS9UzO&B6o85kxWe7dT&O(Z#hfU=*T@NjUA7dJS3jp}!*^t=u%I zkm4)N6%D!g2ME+girNV7B3kb<5Znts?^ut4)T?xjBY*YDB!6!BOnkH@`pgQy&gHY{ ze~XiT4WV#KwtH$zSSeyqNlWRYvsrxOzCUjxkNk<5$q~u(#JMM;%&f^wm~`Qb32II1 zeVpteH8#-bF)Re<*vqrNJN;E1cE!TStt!^__b1t9+V@DA7LJ?Ad%FS@n;JJ`jy^Hf zO~R?vOikP4xc9jt+eg9lIqOlLYH*y~or$W~KDYY^S}q%-1@$!;_0G;Vk_hQ47RL8F zkLMVOQydjP@ZKa(K$?-`mg5p~9#Cb_NalhNDdZfBWahL_(|(7ZG^w^`Mt^EK_+%`V zMFA~2D006y>1pQax{O&ep@N7jfur{M+G#L;oVH5_O^^TXaEcSTfi5q-Clvy-Xe+qY|K zn_RAxqrVCfG^X;prBhF1%QE#rn1fNsBt%(7J?-a7kuFaYcXsX2C{L!QkI(}|tlLp3 zxNtLw2Fha!-L&5<#XxD_E6mY$Du<0J(>4Dn8%%m|Hs}b#?;yYeMzpRNOob1y%M{fu zmHXmeCcN9pok-^wAE86qM~GI&g@vd^Lg;tIWj$303RfhIb!ss?ivVCoXt$cDy5gg~ zfY%o73M10vtgi^oge948C#N_QK@QNNySjzL&tC7trZ4O6mJDZ@%(rv1B!L@Uy(?oq zw~4(}e@I38)nHUC|&Z4f*g8u>6; zk!JCW5%Y<&0P*(i;G|0@|JM_Up7|%6xIu8}OSnkQ@q_FfD5*2>gND_rQIfFd+nun% zp+KK_8r)vn_Mx$ePwJ|T+#uO4dK-01m^I<^IRt0b@V?@ z=9|GqifX4o>NNAT2ijV4@cQCDClccwPz4W<8FH?=Rect;U*36AgA1<@?AQeDFUh$v zb4_CIfzkJhrn&OhLBgDs&7bmbyyyp98aZ4FXsIJ<9+*v6q;{4gy#;QYJ?^-ON7HLC zuAmvaX`2%?CJ+L6KKpXxyw?%_nK{mWvRHO~5xXGGzjYP18T zYyO9vbUOh3zELmLX_CM$z6deZ% zo_mLtMU?Ny`9NUUQW!n-Tw;2C(Z7-eW5l&TDo_96xCh8+|H~(!toQ@fdj4v^?fBoP zIJp9!rzMOmYRCU-_J;ZKm`x)kSE`cWfWnr;BDYfZJ!M*Ba2r@C2E8#{n^&H~))hOR zRU_q5Z;!K7E*G}&to`iZBo&`0aRSfrJ7{4Ix^ju_KzU2HpwF2$N8iUeih}M!CE6Yp zc4$tNhJJwI1vh*Mm%z9-YY4|^nf@rxx1U_@4bd~7*IeQ`R25oYnmyWtw!X>MsC%eW zvsvZ36R#AgaHktsY|U*ST<8o|7MiHAql)VS5yftu>iuK-NVYl)GPz+5mhBjFP>dl9fn!NM;zQ7SF{Q z#j|kNl1bZaMB7vB!%Jy4(pWm~iMZ8%DxrAa((q+!{QF7NfP*p_AZsGty6FR|Gc8w# z*reNo80nN4r|1o1<5-D|G8vTe4L(ph=g6Ow;%ensJze>a5R#4HnRn$JDx5+OoNh~? zV7b$bIP0l25P1{1GMPZ?EZ4TH9+R4KvQ^x+K`jowl5`#U7q1EBf;Sg!gN(v4ST(KM zurOuCf~BV$=kTg2E(?%&Brc{{%d|`p%ALi@%BaT@^6K&=$ zNxKk$=9!kjlp2_p;BwR=RFh*?g1$4@p-}L~#~gF76x;R-c>$KqMZ+(lz<*TPnOmcXFj<>_?s+Y9pti;COqhRxM) zSFBbvYvkYsQuS>_iy$2F_V6$_jMZK)@=cBX+$^;F!OX~;fy}Ynh603;>`mj2looQ6 z3`caOb#MKzBJ@Nm5@hvuEfG63a3%N;&M+maTjp*^IyfT2+~`Ryt2X8Y3M6v1#{?BL zWMf|!qO^zmQx3R-z_Q4{?0Zw{HH7r*7h0oK*8?Cvx%Qqja zwUwYVo;NuB^lXzwOGXGZE`T-4L&(BA5rz?X)aJ3Y-dwn&B+)<#C}H?4uxOF889D{R zT3vL7D|Ewq?l>RZM_Ib>B+VoQi-M6zw3Ox-q=Ais^icH<&uC1qc&lLL%gX|9wo8VA zcGsxX))!Y>N+KWd3*H=}uoqku%Fx(~$a{OZ^Cb4qcvsH2J>k(H zMgxOXF66m1@IB=b-n&LJlIijSCf7`+ZQ|F76R)~|yf;IVx>G|Ve~O63e8-Eum#CLd zG^B%5y{TG#=6Dm^Nr+Ow<&?=1A1_u%ZD2Sa-8SpOur=IZCWZJM(jCy@G7>js2c3v7 zRsW4L+~%u`aLmEtrXkxqS|#~)Hh_itD;e-9b|LuUrK=fo>KQ&tdbgBapQ{TxP~Owv z0x7u?gDy8KK_9TMa)UvgIPoY{&bAp(F&WSLy1@Myh!?I6!%2K;K!EQf+c<)%YFYL0 zZK9>112ob~5*r(Tv6w+cBu1QtZh$-Wh;B`TnKG&${h$D0?Cr?Y!M}J~p>5zLHz-HE0D|a5>2$hdPuo8)tl6H;iA3AX^PFykf&CO_K?-sv<(d7rlg42q9|Ul)M!UdN4i!Z% zTAk0d5|ETB8=^tvl(l{LR(ZMI0+vE=%S&jWC0VBB+=fUlek~0-bT2EfTVvEVF9w`1 zaHevd@5elo!myMP7Iv}TH`6^FY&W?_Ac$I2>NZiIn*`NPT}s*!r|gZF&Zt0&IFj$G z9)-;BH9vHcp$dZz(NUr`#i{4!c6ZEvvjA?f-%RqKU$7Q}q3OEXEX1QrQiB`Y7^vC# zD+V_9F>e_BaVO6(wq4%=Wc#g*S{dH^9I+ ztPT5Cv)^rml}2x(2AvyvJUyEGxmM`=fHjG>IMP^8=?GUyjD$L+x#oOq^Dq3G>FmE| zL}GhOefai~bVwj2<9vZpg|*4DbEo$Zn+5Vb*tj_ z(XelX;C3eVL`-bsng$?128p}ThQKn!8 ze(SmX)sCUyRFLXIh=t;?{{_R;>TFxH6RRlr^{un)w@@7|_KEke${|k4?c_5XX)eKC zj?9YKAXw0xmX}=-zw0K-GE}H3=le4;>;y8dTv6}}8#^vqTdowmz8`WWt(5~9y_WU; zANq}Nt~5!?!T(A$&&`Un#?s3F_!jZ_a+ph5*b>G`JX$gZ%0mcoc&+c1GFrCGngM~VMUxkMfNa?3bubV{hmP)k6)_mqx|dEhAROT|ZFYdvZ0pl?34Hpu?0 znX8->ZSpMR+0d(QGh{=q^u@{PZ@CvoKqj9b4nMKFDfW)SuYK}iqFe2%DO_D|lb;-= zwPhG;V$ZgVzZHV76#D;Ka9->L`kb&H0Yev=?Cdd)Av#~@+yf_6`3oG-(h%qssMW7U zz@^}g#LZB)_mMX8^vbY6@8ZEUAC^?DNFi2R_Bv6aNRCD_WI2;_^yt-Aba@3O-^S8+ z1|{kJ_nJQWjL1^^BgxolE>*$b!DlAgo(>$qjGEFFRfR5wM@4U0``fy*0tY4Y-(0c< zZ=&5sF|QT%l}Ss|*ZlfTU9I#gi#%)Qw1Y-#9$9tnCAs#I` zM%SM0Ps>+>C%$eYQOLg?yZRjRnK;JRl1K}gLZDDFaZ57J!QKUQl;d8r39dtUJ7 z9hvZKQ?n5V7`2!A7C9{L=}U4hdK7wW_cA$H3&mx0Av2FB{|60Q6!j8$tTChQixl>06<%#YB>V)kmVj7L9N zDAppgp-XS+Pd*3qkr*>cfttoK->8`32T&$SSUY>;9oYI%FzyDg8LDv+qlPhf7G ziJ{?*N#t#*917mvmmnE7%I=3IHyLGj7*|5qXI+^ zS)#xm%6Ni8%|#PC$#MtBfF%_JF=s|yrj1!wtWHXqXW>UTDv7qVlv5zCoV@ zjW1HIZN<#J8SF=%^oZqkb& zll)c9L*Q$k>5LQL)Nfgdyz{KdkekFBf9zH=hiEsYZK(YZBW>>q^P_RYIl4-It#uRa zTRYtWsqGNo1~U-d_yQW{N9P9Iv<2_?u6)H4NO&pmzN;lxaDC^;RT`cZP6P-o4}OJS*yCbg><}$XqW|nv+ZDz|l;Jk$9j?jElWzuzR-vr}0mwKXLbr2J~ZStCf8nd3_83W!&oBB!4u;B&xE z7JtM0j^}te(*`BJ=~$yE%m$-G8F*uNT!;PpPPyr^RsE$fHW#+iaVVCS z6$EhfzsmgEb-4nnBcBrmD58*8nNySZ3UiMJ)95P|{eil3e-UQmmwDj3xzP8eV3TnI z&-NnakUU581}e|@S;$p;4S0d(%VjNO_X>^VG#8rRckV$}cf>3=T1?*u^ikWp?^7Up zuEddz`pYc0Gxm1|LFx3^{Jh1H0`FLUJZNrn=PIKUkMKWNNn~(Sp-;fMD#Ewz!~#NPZ*aFk_c@H z0xcPxSr7{If&OkdL>w1_%M+yk;&BetSH8T2q>&i?;y+;n?2Q>-0Q11xgBVq%3_v}a zC8;~JPn3cO_l+j0HHuoB9)`AVPFlHN7r10|1Kfi4N{zHRdDI8~FVem`D2^`(6AA7M z0fGkiYzQ6*i?g@}calJG4bI~35Fo+b9fAhS;vO^vcMI;ad-J=xs;j#D@9v+St$8wU z`ptXO{q@&9QEd&4tdi*4BN72FK)zihwz_yxE->sl39nXQV5?14>G^>!BxIfsJ((Sn z(~vq|FSuBII@0m2o^}7T=Fj~RwkPuMx;&y*OsQKe@ePaP(A{Dr%U-y@XXRCbx`CW5 zUeQ=0erx7?HV%sMHOWUq$O&|Fd7TD?8$LYx?zw-Gb(b$oj&@JDI^FoRHpS@@<4fSu zxMyab{mQ}wf0Yi$w?C5_)P?cN?m&N91EE+?s5$SxL^vN}>Dw3CI^K%;hi`GZr{Fz7 zMBWV=!rED~&-1_xr*e#*hjk;y%YrSza?3X1GeMl(=ne0LnIjh4PEqjblW+ghYbS+R zPsW`W?ldQI`8`T&@#GBz9l^3P6@vI1@#Ll-A!n#q@%C!#3`$ z;0LCD2Jg3B*q1<#6D zo(l|e>$ueh{PhiXirK{}-VKQo7^L0hFt#Q*|Mz{4l6oyhzdYH!fV?VSV6Z`di%xa4 zhi6WuQnEUJ72Eb>=-kTz%aanVPblV$`A=_FLoX_BGcjKNGm* zq1kAujxO8VxYl9b-5H1Omnh^17uEUj1Q*3d=qU8y_!7MFhync7tuzJ4QH8)06=K%W zwP)rvBSRSO2NW9ExWojd7g)L}Sx*$aZ$nWsC#k}%ioICESf{MGO#AU>j5lb&M`)cg zGToz`FesfuYrRu1*yw`AU%ydckno4zq1hSq8)b^p=ph=3>pPDa?}GO-70So;PBC}s zDuB$wH#p{AlJe5$yVs_0{NJyT)-Vw|ae(HUDyk}z+EW)mQOHBssW8cf$u8qRk(V@nHUPt!0Frn`sJSdtNz-fT{B9(*R5;@sH8yQTU7qe&3$ zk9<}UkzQk?WE~Z8x3qZ494Oi^OE#_rr~1HnD`wzMfu&3wHx(m4NV`c7ABarPF_EQ- zj=0P7l3|EQFZsvv`iOZ4=6k}pv&OjdNp(~M>8rw@d8t-0D^sD4bc3{|TYsH|T-x=N z!I;yBXf3(K>cf;g8<*Bl_vO7BGc7g&br8_PB69@>OJUW&wJ1rd81+K$cQcBwdzJxj zczl*brDW$Ej;U*uM=fy*0BKsVNq%z}Xh8%bDS7BQi(WaJrFfk9GV~Pt7s5;wFw|JhIWg`x*JA^~yEUGy^8Pn+R2>+nbBpT&If zsncUc=hh;|FG;HM=|VtTlsQ+-=l*OTQAsYq?sMaK;ZHZT+%Jx&6@v|;-<80-SKht6 zRGTq*yvR&Xw&V|xWyUNmm)WBqHjK7@j_SZ;5W2cj&=95Q!LX zLR3ze>>fRpJiyu^WZU`snP#1Gzbp%iKlzj-w)F`C_ebmjpY}|6LvKe9muWn3$Iot{ zKglidkHBW!@rcuUG-AfI{;*E@mxO&6{>VM17hl8NZnb=d7U7}ZSnd^-oL=H&F;0r^ zIg}NNiq4EZ0L9|DBq9Lhz!fwBJl-&2AR=?L@^|eh^CZpYYNQAYCe<%w#nidcf31Y$ z$g6st2YJcMzf1F}#4|NGSQ|{`zZcGRis5I8hNFDFF#<&fs(N1T_ws^tgmeq6MR*Tw04Vbr+mlu`)4Pmz z*C6>3XPHFG+;ciB3yGl4G%pJjEMMUMG_LlWpa5z*x1tbg%ULQ&L)-Qxd7fF*rJ*>@RQ0-w2@A$5&gzr04DTNFO zSH2JTm?YuVZM1I9$F}IK>Z&KG{`XL}G#IFB_RKYc@U0lZlw^r}ExSRZkw%;p`$3L| zv=6U>=2B%xaCKr`URu=~HT)LY1<%IE9R(F+06EXRVeA*i9X5(|9&$qhfDO~K_!`uL zh^#1ren*hda*GS{7&tV|j;Y}L%5&HQI*@?auZz$310N<5sj;7!-s%-%0~HjVC}`c% z0O~_4hB}sUtm0HH4M_0}jzW1v#^dp&mdx#dUf+Sp2&t`)6(dp@@_j~9NZDf~fj{13 zazE&oSL4#39yA>ugIA%4G8SfaJs)tv%J&1c`66sP`N6VolSRkmv~Iv~A_EV)RqMPj z4rWqKm%E&0T_0QCYAfCo29G5$^I=l>ql-qsYKZrVLk6y-ti+BucayL7Jx?2XC(Qla_kV zNSzw4b&v<~0X>m=^H<%MVXdiiPfA#LXE`XcBG39VTJ)_y(+2O^W~a#gXix~;=HNZ0 zRyxugHPKg4Isgo%h(|CjcYg4AKKc9llX}w(Zakji4$puaskUandFak-2~~;ocOS_< zWDCgMFWmiZv+T!y$cC*$=D?D~7ZU;7mMXf>N`($KiPVu?Gd)2R*5Ad^U@cbB+Kt|J zbeF;$5h6TQuFVy2Scj4gEq9U%vFoRLpVFHN>4&kTg4(uXTDPZZfF&etrbPd1F`JB8 zEzW#Ems~@-smLAihV)*Z8MAjuv&p)ZB|l;DLb&1sS^OUCj_Qi&ea)lll#4r8x7@F@ z)sC4oMuX~o`ibA)5$lmd2jfNVxMCqZ!HBsYsdzNs86r!GHuup8D`CQteKxu1a-J&b zrvhlsX-=xG3Kj3XA?-{LV8R(-;i;wAPPq{O@fqL*$Rvua8IH%l-%j=X)t!cXGiY2Q zv-=X_CZXH960o>|OU&CmT%Uvn&N1d;3(i+`_*DB-wT)0quVrrygGdEwJ>?>+q|7D0sq)D2YPSj1rH!^lsZG->!Cis=-+JoLU`RF!n$FJ_wjMVh4DVM5Xp;7s; zDJO`1L;NOHd@8BcJSf>G*iIEOeZ}?qSp=a88}EGncKs5uC)RTD^IKYHR@PtcyO+F- zyc`&SOG1rl=K>H`3Oh^etA(0>ZP3g&nUO;lvYE6UC#P7j@V(sWh*mA6o&_4fhHDF1 z)6QdI6_x{oA-&}lAx4zA%Wy6x zgFq_DWNT?=Pw-f%vpbqfK1l{5ug#_yCW!?YAughcV}G%H@RE98z^S4%u1!80fEr3N_A;1KJ@p0>~3> zhWE$stRj5F5Y!41#@aqpH-c9m@{9R0bbdTC$wTt1r1JL z{#4aSeLQp8}og=fe% zrGZYbVNG`upS0sJU9~rEIPPhEPkcbuk)Wx2s*yK)#OXJdo6z>1RE&HxR7j)7+|SJ!IF8GcEjwUZYqDGm2Q6ZY>?R zK}w0FbYEU%98;pb?#dsxXyW|YC&)t^#$lDW{e5SfL%TDZ*2KU+?R-y5rrK5RnR24G zmg>D#ShL@*lUSB@LDcwAZ0VAgj!$pUu$;U`9$ztrciF_lT43vKd}8=8Sn1bj!s65)@sb@z zfc#!r1^_E3j>pgL-SH}YPm-VRmh8j*P!PtJ+e zFTL$f^vKV4`siN*}GMAEVr@8WfO5i=)4bC^a~{eUmdIj2b5*)$b^} zGxN?3!%s7l#ei3w|4P1J^;8F^$(KKogR&1sN!VZclB1DpDhx!ZxL;w)FEDOiYNX%> zF#PF*-R8d~;k-%w$fMoqLF8{y? zW1aZ5<&`c*l)P(xE~cCvK6LT=4k?u0@x;16>WJz77vOFpEET-w{q7%&mS?tNmx(rn z7Z)vk3BVl9(Hg&oi9>>49An=M^`9{r_z+D$r#sU*dR7%4m4a2mOU=7p2BQLe#VH??1c)37J4 zRsa0gmzIKv00Xo8a?a}FnH08kMoZgcm&JJ;?AKP&6T5&mAah z&61?;z9L%IB3yZwmJ|&Bouqv`Rb|Z!)?UVOnwmsb3%S=&ni=e~|AuIDCii$WFec@Y z2&sB+62_5Rld}AVFYt*JcCW5%@H{tiSCFS;|fw#kycWB;4d)dM{adxJL$G_dh)>i#;KJWLTO1 zstsr5Z-*$oZr99FK>?s|OO_1ba+F!|zE0l-?9%`P9WNfh(R>dZ1l@;w zwn8;Yz!WT5Zo5Xy0>d%VY`@%bd@K5k0`z&HrmCk-cbQEc}bM$Sa+RXOgHuJh85yoa|xe!QWg(0VM@{# zCgn0vVDsk$#4GkXu`vHrVu6@~>;y?t)3M0LKZ}37J$7+JV~usilMtB!5!O}do22fo z%URH}>t#eWy7{@i8`jxz>wqp!Ge8lz+NDVbGrk??^^(uu1j{Oj5DxO*;ora5Oz(JQ zocr({!P{-i#yX+a5V@Dl1XR28^^WW2^-}VV%iX6hS6)x|Wq4ogq->I*_hT=+07PDw zA)>;CmW)(zjR)-Cf<9iazvsdmU*C23gnEh&Y{>*l9({^qO9i^%4f+Gj1qlXg$zK__ ziK(v}=^ae(mUqx6UlHu%ayVl(v`92s}WbNOkrqvaQvbqUgmmcMrYgZ)oJNK zp4@w-#^D!I47+vM(SrOL(}GCS12QHpx19QN;a0;gH1aQdGaXaaml}94Fw*W46V%zW7B{- z+nI3}R$a~9$}y|7p8iPNRA;SlZcwo1;BaH4={WUnz=Mw{r4{jcmnnK#Y3ox20 zAKBl6pUsQih#tN-UfHdZkww_vefpY6h5WP0^3ZO9HeAB8jp3GPa1-^pv&2tJ&FuNT zF&FOevX=HI1_~X|nPIb@wjVuKi52RKYL9l9N3e{o)_xd_ipU#dCF4*xUx{F)EGESb^R__wh6 zts>_I*QuRXS#n2VrqxK7g&6Cki%7=d2Bb6cm2%v@{3Zk}9dPwg0ew{GARFD!f+WC} z!%a*0G4NImwJV=+u`smF1AWUg55}*2yhFU){x&++eSh-sNOKW>5w$Sp zyG3fHAoA}Mo%r4v+j4tUDzk11_dzzEO9#EIf(xwTc|Y1NWMl&oG;|RT7jF0PN}>tE zV8tr?p3=wLI&1c(Z!1j%xvxDKIV7Lg;N*UynY6Tal-=;*ZW!1dIHEl^-a?DhyBGP{ zSWjx7i~Rt4xO3R~&jz43V|;3;<4P`66mih+@HuuZAWe>k=hUg)PTUFDMR#xILUzf2 z@!qLN1?d(Xiy+I#Y(;pzrl_K}4pz38U0=R#ox*Jjv^>``K6M@cwA+I4y5V3Yifn9a z^WabzX9ULRR~a^RxM^JMKGB4Ld^tI&h&=wf^R~*TItyTNwe2`LDoq0&l_-FjplP-@_ zEuQ#&-SF+V2=&NTjgog#0Qq|*Kh1{G9Qcm$ru2%s`4aW3VByO!qDNXub{luhRmDqe z=U$lJM@!eL`qt3-G@U>Ai5XI|JKsWFEiYKS7rWPwIB#XS*OMkHB^(zT&|AFAO4E$& z)`(k4mOfEi2vb9pUcO5atbN-&j?XzBdvq7gN=yjzb$lHjPtZ-jUxvtxCHe$dA|yYz zd>!1pe+5uhrFGyjj7Jk^kyBn&<7TTJaWz*m&Mvs8Z<$gd(Ykr1ixL}$S*H_jhn8;Z z(z)pv?iaSEdZ~}|L?ay6a(l68B~ak(oaphr-rpcn+%hY~@uL)3e+zw=H(mb-dr40z zt!Ur|pyj8z%DPj-!GTkbSL<`i2mA!3N9IpIS@tLt>*z3aMQ_kF7T+w0+{)D%U51HhrT_~8fwMOUSkt*VJ z3=tZ=3oyMha@1q&**Fg~2#N9@Dxs*Tx=Kp>kX_$C`sbE)V^<~cCdn$zLdm>xj|J;j zWA{CtY)JVyQ7B$pN1lp3m$!(ic=yC^3-g7^dFL?)X7nCQS`euxnC&X*0{i(13v7|V zP1YUPYLhQAhXS;6>a5kOSuI3HQXU6>&;qv_qSc)(Z@QtT=5tzYn!BIww4q0rrFHbb z-WNM`v*wU3?UuIO>_}@+vM&+RUa#8Qu6!T?6Q?AgdWMv1O8>Qr+S?vw8*p_@7uJhb z7ZIw(t=+}y44i%4=FBhX|09KQp{3?gSJ1)fItgnbROrZ1^CE1T;JZ*tkbtA7G|<9l z$bAu0=i#cC`dZA6ct!ovCA9>5wnnb~G%WHH@z2b^5p|+IB0a3$A~;gcg5tq?XjYqt zWgD~4^MaBYrC44_Jg3$pv785r%|w4P4E9okjMwbMMCj`zN%9ywhIj8>b`5wGCH)bt8fzrA^+woyqkV{`JEX1qn5CDsXWBJmt4L($uG zUIa&|&%Q4-5>(}|$%A1N;Wkt7gDldA3lG_ZVjoLLdWGC1KQ(*CQv~{|SRy-VutgBvV?tVPa<{ z+d_Dy7QAzcFH!&Ad~T1XrH+UCl- z&Gl`St~F7xw)_aIti&-1!N5HJ=hnM-2kDQl7NtJ*EBH@-TJoZSj$pR5-4ZPg5cK(Q zh?%`atx1;U^NTDPOZZjEyGYH=KdatDp$G;+@ztNmEH1o=SlB-zlJ7$B`BMThwJF~1D(I>`L zXtptCn~Fm75kuGL;e6C6`DqP|?{Y{1S`D4`J=VY!qxSIXg0TNOy(wz#2NNE&QrIZ* z=+AYHw=ycoXQF>R7giTM7k4Dp_H;H@UWXI!jvU31HwLm{O{7k&VYx~|i2nq<0l+`O z+WgOvH!eE)58JB^&UN5uR9~)}ES_dK{gonF-rKZ2i_KGA!SM!E_FaQ%&5rdNyO;Cv_O3P&25>6PQ)(j~Z83fZG z$b#9>f!5Rmt$|-rTZCdBBqSj7)QAD>yQ;7`Is()ywaDwWqiy(n1om^kh7h5cesE1i z8Y#acbir^0^EQY)k1^pDqqe89OQVVE_YmOc+w~VNO)CQdZ*TQ*QuCT?K5jsBfYTs3?ro66oh9 z=N@v8pHNR&a@1D2>%e$^4Ve%1lxEx$n4^W-&4ZZ4pLTw3p|i7%1O0&vc7JYN?UyJ{ z1HP=3o22GE9#t(F!2sOxtcUu5K}`G-z2->>e9jB(MXbWsk@hI>O*7y#L@})!eUs#o zE5E8zmiCD8&*?kkB~g!eHo=0*(+)5Q=FFRy#3rWG2B19rKMT4>h2k&neMS z6DrFL3YlQhG0V}-EWPVPcj zhkZ~ib=5e5*Z`@!*Jumk#Z7--zXH!Ro6dfCS7kq`Px(MdMs1kRU_tS>3w!Ve_9*@IaJcP^|6GEnsqRk1$>eW`D3u z((YB+%VCsbqn3Xv_=x`4Xmb8MC^i@oU?^<0`7}x%t@-|FWW!kLOH(ELZ?>OXKc_pg z?gz?e(5hjdCNVasJzqsXobG__I;@?N^6l3+ZsWI>()v;y7@vMM`sttMl3oGYYu0k6 z-D9VCZ}#7M7jG}=of*h|7^au)p9|vF)-=gwgKApYDFO?eWt?Ux1u}B-HlCcpinXU- zlK!Ze6_%$yk|{(?UGjx278!H%rT*-d$(QXwv(5G&i3sO2daZ!wcV4e+zisEEeKsXm zor%VIa%G?F>(@ZSt?-@-nkt3UCiS(al#V~eDmQ6_WOAyDqjv^+ol@BNttsuP!m6Pb zI9ZdBcfgG&dsish8Qq+wiVm8keX~+%Gu6R<(%Rwn9r=Flr#}}I-Vft{LJvaTa)?PC zKd*a7rnzwO5H{0D(4B5#wM}wP@6J%2`gJWLI_I3elx;y6TIK$uLSP1KgH3f=5akR0 zfGD&IP{*ySXxyEu5qE`YzQhxl#7JI)_ZFLNO>&8mUGAfliIVZU*@rGduZf)Moo66t zIp;YH-r;7ONqwpGapV0&Bs3G=oGGz0TB5@ibs^O^9Kgi0xB)Xtg>*gu6zZU0J_7QA zU)P@E3MOmptC8|kV}y%o$FsCxHfo|i4*d`4{D^{S#;wtm{r%E(uP8rhpJ-{*LH%Wi z*qqse#WaVI1trsE4ab#QC^ri}{56_I=S48Xj1q-CRKdt$0b z6(rAT2WPdDG2IgJo!K!Tz^fPAc4)hd2;%> z?nsnFIs*_R&!p)T!$-}LI{7gw^?8Dcl0L8#hYEMh)t}cXY1Dj9c$$qNMo%i=$RMhV*98z@mu9b^v2ZZpUPPoxA*|kCHM~P5BOgGpj5+( z$z~Y0y1c)gCwS6hYwGqZ?Ijf_6><~Ho8TZP!|sdTiG+Iir|dTmQXv2^;`p5<=GJ!V_6qH~ zG)CAap`9AD`e$j99e@&2ZJpvPH5t)J_-J6%rUTH`FA1)Q2eAUYnJ-#Mp9w`xJeLE5 zIC$CP!^+O@voG)T6c@9xw1T7PfdH15t$LEmPaFFRg2qx_e2858eQBYBUzNW>a96XCmiAlVnqbBvc%Ti8im^HkcW zgUbFaXt@=8Aw$FW_>-V}1U2(9LWtSFUPk!~2a4a3lAliCs1`^WB%3jFI7Y1Oa|8(k{E#Nleb)OD%*!Y`%A&gB2@&tM+Hl(x@YsUI#Zwj#GrNbjZi} zJr3L)x&I*Ht_MxL%++xNbsAT`WZf{-`5QfqQfCmHiKU%kl*b&LaD8Wahiideh^GWi zF=z9EipotRh9;-#fGJZTJqaot-!1H9p_`(06QO^-y1*lC`Sd#hEo z?LGPb5~Ad5At}3uuO=jIz7=48EpS!%3y?h=L9&}XpF$Rm2{rS;eqRm58nO{ZqTVbKwb}On+-TwJHC0rilJA+ojr=eYakW>QUdf{Auc6 z3P_rJeTq!@RXk~A@LVohQhCFS@!K!j*F3>PDP-(aTm4uZ?k?0@;yhGr%yehp7;n|Y zWxwG*zc&%-INK3IYVwa$Z(WjH8zEe6zKtrQMFxv6vRv8|L0>!E$zQ;wWMp|3k02#pJ1A1mUU4>diyv>QmZ0 z%g}iCJD<6w39Hq^P5}baty*ggbg86&{u_`R;rd@sZzTIRYrvOk5RJxOmr0O%I+o7S z2K?D_Ahea*M!WHM2M0dd5R6}zwKO8vc?h=R+JX+Hp%=}vYK?#o7+a=sW4w`Msg|^e zqe{sIxppNn1@}uwN#z23smQ^8L)878zISU(2~Zk4CPh;17KFyf$reSNwh1tG(~(l= zt4i$Cq_dJy`Q5Y`~>YHlJ zcqZ%I2UAr4gZ!v(O(`Y0Nv$I6(CbVZm*KP&vdKy#wBCX<@c(=Y>CtnRbZ~G<7JHb( z>E6>E6((Kzv;E@r*U81@=9Wu&wiIU|iglZqe^nt}F->RT=jXoHUq8MK`{hSz@5E-# zDKBVP@gw8!_g?MHm*LWn-$Ze>?rN*XuO>|9Zf@`0QSyCo!IQM%Wi-&M)q%B@NdN%# zp1>5r9%xqEbljm0EmN-Ci6uQ1@Pi!W;In!n6n^?ueHC}k2EtysC*L@>P1d| zdlF=SE4`edcj*OO^skL4kpNb$N!!}#rcPC@ zYqjSLDW+V6iGsYoPdr;sSQQRKv;sHx<9*Yco6jxZTCQ{2gvMS!=Tq*(*W&%9J^FwC zmDxakh0^3u7=T=fjBvb7u$8dXX?gWJ>ss2kP3gVPC_86{@Aozu;{KU7R^-S3^cQKI zOtj^W+^2K=w0|Mh$!zcw6v=iwf-*}oz!>@+cLMx__2CX`9k}F`N7aaN*P*UO=f!8h z9qG$_M6pAml+W`AoCcqjzI%w7z{@gOXl`sq|E<<1xuYm7H-Xn4q;gvXPg)R?d=P_5B^PWjsII*l_*6b`Uo=k#H;DwtClLkV4H#2zSKER^nls4{p zA16*eWrvPE>4u{%@SS?CV{gA6DHmP)G+jZ#z-`t2Cu1zc6gLeXU>c>b$Negj+H~{A zP!7A}k|5grU)7MWLI`ebe(Y{2y%WdZg0D0!DN1An6^W^V_)-Y^$(u>1} zr*i3a=TB>_^dCM}<&uw{G~ymcyzftg6Uor>nu6@lz4DS0NWgNv!78_Y;qLISd{N1$ z+&BbvaV>NYbRY~6mZ9Q)((R;PBs^T!bGEZP4ZdwTZ?vK798$C=he7KvTb$+m4#WR% zztM{bxU<66uy_ExE-qcg9_r6gi-ea8)BET?SD{$P$1;GtI^-7os5zYfs}~%~ zv_4~ybqm!qwVeJFk>Sg%JhAQgNiiftt~YcxFG!AQ-TT!f918U>b6lyX^e=M?gx~ni zRx8#Q69l8%%B9Kd2n~s08InZS0HG0nw*B&viB8C{q4)f}AV$zcV+hv)|3poE2T8x$ z;(NK5XI^Coe!szex|tlkU|Fc|&sjo=C21#<@t!Fph#`a3o`DXEQ z9_}6KoevjmcMp}~?2noykhzD{`=Ytn=l&cr8)|Y?Oy>2Al-A5gkW|rcV4eOA`l{x4jHv%qkAj7YdZsGL+F9AZm*P$dUl`hHZ^8Juu? znQ9^q5~Qzroa-5p?=@paTn{k>Kp%@}t5O5Q{L3N?puSV!K6xo|!yQN8jQ9|!?-M98 zToRie2ey2R*=v8P^ivCCUGTo>pfeL5{yq?%Xz&{|&)bF+kRyQ{j*(8NYS4p5AJM`hv6G1^ILMOQ;u`4&%^K z;S?m$TqK@W-D2H()~rMIw%3Icl%}8ybSfEviNtVdgb@&-sQngXdam>?8ZK0E_OO?> zap5=q;5P{lZR`Nw6(vYTCK7X`(BRym4)D?7B;K~!h;^mLhdc(vCz{$&=bX5Pn&PS@ z%Jr6zI9wz;tw(1hrUJ*cvz$aZ;6gO!N@mQ@=wm~=xCHl9{mU96R)@guG-nT^b$e;O z=yBB1{v3tdF3f;>xvoqNeF|nb+@nYIy8VX4GY)4mL7D! zYib>?nk5O%?5gdk3W1Y+1n10~*VX$}tC)&+O}kLreLI`Mt%ebMI2tQw?`7BRBAHl@!Sdq04uLpf z&O3oa$Hxtgd++mTM>%RBp4D2FnK`@E^{GJO-?5=f_o>p^D(8*aNqGePx#v+<6s<%W z69bY5uo>IqE}pUL-*da}<#a`B#mjclyfO_WwQ@A$q;Ox0N1k^7g0-l`cP7k&%` zchXgxnuiVu*`budq9!S5b!%M9k#v|-dD&hJ2%j#W5zH5%A8;hvmJY=X&=OwdAbqI~ z(7LxFb76<{EQzp~A@F-F)Z0v*lhC5@Gdv=*D_^>un!5|#M`3H6!I7TLOMMej*RsX{ z54k740#XU_yCPyc#U~(vEBw|V5z&Fw$W!Sp+J2+>uEX@rIGIVAX+reCezP~necua6 ztRIn$SF~vy)_NLM+)}&w6vH7k?;Zxd8<}|NY84kHAX)0Iw70?Vo70`<{Eo4{p?sD# zZ{Co;pyt9mm1yw#s($;0$ISfRi`% zXy(Bt@m|SEJ=^ESXkXKiBWhW{(5dX1e9b?Cr-X7o`t_IUe8}SGNABWOjJ;+@4oDpE z^gGL$s#+XQ_WJRdT=_qhlZKcc090W#S)5>CrR=`FUHDQgF?A(p^LErh5MDn?v#+cb*uIubDXsHUC_5iJ zCEU0l5YHFOw)IK<;V|Rj$Aj`x$yyNq-i=5hTtv-Js{br!5FGc#dIsVsBq@3cuzP?dXAMLe zo92=vZc)f5ti(KS4E_vdLD9{u=mm8Xv0o8Ea`Rv&O`-jgso{?s&{iTKM%BtuEdkN< zTz-$9KhE{(!n@UKif~yd=W)aOf>jBnE$_WKj)WZ|^E7&sqzJJtugvrm+<&&cMKN&r z=92%uZ;H^*SPyM-S0xptQQB4$3768rI*(@B?u7a~8>M;!7ZA0k->2`;1kyfkh?U{q z_1&RnbSn5e!*@R52uGB#bE8_*bV&l0g9{PsSRF4;qYbkXzN`2o>>zZRKxSQTDIi;8I? zvAb@0S)Vq{zNQ8+Mqfsonuj6G>ok!|Oo=U*(W_X>dT4PDVN`a6`e-XwCbIW^`gYZU z@KA);`SJl^h!IuN`W*RH!~E~lESh%0J{Q8|WSIvPx>}}7sGu}Y8@nk+_0M9g7^c>_bIS4Y+gDyK18Brl|P|eO|-Fc1@Mk-Fe-Hm zrIr)!Q#3QPQbPg6|7px~29mpF061NcWQ>nB`ALXB>aw<|s81JH6+iaAOFovwgrx?o zU`c8JQ*eU3t+N=h+BxmIkJ*B_a~oEcuR#>^JDDjXu`c7!zcF&7PMmd>r_D}+UQg-t z=o&I18hBL4n3YiW=`W)XL*P8*Y2S07+CEDuSFzbtl>c~fEQ@eOuq#{c-+jOdsKwc% zpn&=g9AF6q!nv#${VClzVFTYsp1$RW2m=Wf97TUKqNk#)0^YVlQ(pikIGi_WMVYe; zSA{<2PH&$JP&{mZYI@1+aFmw(D1Sizbfypim)3q#s55mSHp#t?qPQ7RQL49V_I`%c z0I?t=Q~aGRyE5O%pOJuI*`~Qm^-@R~?MZZ~(Z?X<(bs#nG;UPE=#k{y0+w6{L;pZC zJZ*cS<_qwZe{j5>X(ap?Gm`glqZGFDr7X7xmD1&Uq34MJ>lBY{4U$Z|G8llEB2X$k zjaDm{e8j4}plQ7mlB38G4T1g?nTK@cyz8vR*z3l4c90WWBdzBWfJ|UPN%SF3TqA*vbq<$rt z&y$jOjjpC%U*p%y%z2Q72Fq;F5&NH+3k`zDues?B%{`I9z#A``GiLDCecJZXBtXuh zYiYdrfH(naq@{QVMCW%uxpY-Jwd9EywK`dxM15Jl{p%8I+`-L*I<@|t%M53-ljMmr z2+7{p3mTHrlv@4}X%vuP^zuxOH_UX2?P@}12*wPuVgN;g8M0V&^6PxQu@1X<-Md0hP^`&ooF=2O1$m;W2X6hR|udb^h4R|GSRR z|2!$P!3q*eY>9-cFgq|hlSXDhm;T>QAswYQK6SLb06X+YXPo`-Uj2QbkEcQjrvXXw z%NY)Wc@`smt_G?al{y7Ju6`|41vm)OgXy5eTOY;6xuTzqV6hy{)Try{+;RxlgF7T; zq>xGv_j9~L`d9&jyZ&FCq0AL)>o&XeT`77vN# z_1oeDtGqctVD4`2Wgl*x)9jOK^HhhMhuGsWxruY1dS0$xZsQ5=>#!vD25pP!;KbXD zNoQFkHd3wW*5lga%c%+HQMWhW!OBs`21I@%xbffZ*AsBKe%lZ3cHIwUVzJ894O`xb zmRx<(J)-m?AFB3&Iv>6OIMb66@~GakGeYL6VLo;JR4a9O{{QQNxkaKd`(H0Ss$HUX z%lDQt|NOvWpZR(IJ(0uGV`zNHObg1j0wIy$tIUPau2uY~iKicj(QOTvbxzwi4QC52 zL)ekK-TiC*DlCTFk=z|G;cC=YD-xg;T$|3?mXk=1**DIcqnQtN&XR%oy{4%WPMf{q zQAQwrPeDP|v&7S{ijVWt@cwn8TIYk^zIQfvhctR+6336KTpu=PI4qfu3}cuW@YINI zrju?zd_)}9pu~pI&rZlNo3zH!r@`$~r(pv&?4x)#OO0fc(3ZqdEGmiwA0pHx;c1xM z=BSjO_^o;pjG)`xP-mGP2A~Lb5HOC-QYD&DnKNG$?COIQ_N&Mo|FP9JOdVd(3WbW& z1~ZE+$=}3dA58_i)^<^#=cB6z0cq|?MX5w15Z8L_MJpvvm#yNi} zi`5zrW3jrnbT0{n7er?r(;(-+{Rj!~ zU5+M%J*ZjX#)Zgt2~i1oC_BEn|K#aq{D5k}075=JLtcIT@{)Olsb6Aa{)kRUz(L2F zfm%qbUu%g5!di_I`|v~hnI z8>=P2NUB}nd;xFw=8Wi95kHIQ=b*y5=)(ERlqaCC@{9_{hiE-bS%2$;k-Oy4G3Z^`}%pk>?mZ9ER z>?;quOoLvCzue|dQ@EaZ3<)rRQ9i5vvc?gUK8x*-IV{kL+Ee>J~y?1Ad3(_xrk5G8K6&3W6qXj@+#Q^Xq=`t?SGYtgMU;z;^GF#{y`SgJgjXE&_Q$B54r*cxDcU{OOcNItr{PM)`s93dTdvvWr@!tMyyVQr{HkY6 zY0T=I+Ng8}b~JK>o&HAxBX?Ti1_9QY3V^=wMmA8j9oSeP1aRX=T-cFaeJ zm{7PCIExL8NF31nO8|&fO>%68pxu&@FPXd)OAm}rzBg^NTM6nboODXn+ED4pzR_MY z`(T9R{@2&NAYAzjPz49erBhK5b438s(mgCWu6Jr;;*=13ZyfjmW(@8)aAJKTiL?s! zCJ#aX;DC(ynE9s@x6w#y<93MCo+exuio6U*0SzP{cm2>4y?+-BtSC8yp5s4fYb;9L zxF0rEq9zQwGGeMSb-}$8N^S1w#P?VXUb)UjB=@_0@<29IY$X(k%Xoxq_`1C>r-|FK zxaUnL+vZEA?^FYcsa^Gc-vwyF@)JKMKWur4*A@b1)AMkulCX_fCLYwsj}iT6osY!} z=aRJlgR!@cit3C0hCy1SyJ6^*l7;~sLK+bf0ZFA>2|246DQ;?DdsTo93 zVh9;v;s`+u&mXha#J%U7IrrRs_u2ciKOeR9T{yBFJ-mb)&|{kz6aMsL zMPh|w(V_2?9sCJyrZk{3oqDnY*b37CVi$ih0D}>OK3;7N{p%7Q<<7oq^28pJeur+) zm2ZSe4*m0WI9<=f4P|MO>1d+!!17)IxTGL?Ja+dDO){J`Gi5!g5BhFR0h^Eb|i3L0cZyb?I8nHMYQG> zW^6yOUCs&`p8B%Or39{r+d5WO`?A9t?a55kWm9;O-vHnk7xgFpQWa^>VOl&OS&}RE zFd06(bH2G;=F3jjHV4)%ZT9tlB@>NU?Yc&7#J2(4;1#A}+JRId)^TJ+!yFiOq?HVJ zR^pA+#(ZZWZ;uee$L*ASu9J0R*Qt0{d_kq@*Fw1JTTb@uj{37ZryI5Rhq2*)BRm~I3!SjP#H3lf?L(H(hY`s%Ysk8u}w ztTxs(aVGE}BI@J<8j&lFu%RzM`dY`#bF+?1Ee-+} zKKBO$?Hu^I3H+xOvhJ2MVr1~<^Mhc$tM?~MK_b)p(rH01@7fJr;a&4xifnB1!}l=o zWj8>a05X*V?xtBk#gCJ7dNDj?1RE(xiH5iM7TT7wuXRmc8iyu+jm4TS&8y#3 z61^z3p?x8;=?_{BURJB1!$&y&zUZBFI zcrD=rRP$^o)b_|`K+Z_)51@1*RrY2Qx~x-$1llnw32=#OB1G81EBKQyj2X%E7*hs? z>t2+g6G}O@wP=K-74~6zed_Rv{73KcfCz~H*RgL`ViI3K#FawTt?P1dH{sNnF6)S$ zERk>R=?k<2Ogsqi5ya=NT#C0OtNc3=@|W#Bg1`FJlz0-_i!-ZgB*|D!e7?TwiZi6u zWgq(argF>>ph9GOk4F)D`qnAhnF{CSXy(4|cf~z-PuA=C&AJ~Etp%-WIr`wW;5{QA zeLNv11{^l_<^zS@n=UIyxIlkAF)mnk!AxRKq2X#8a{+9m;1#1GX(%dkq}Hz-qhXQ6 zs``I_#Co+eo|g@oet@#tgmdLW_NWD}ff&Y_lmd}Be)a>Q*pEGBD&`Pm8rPhoNsm!m(b%-lfWD4F@^73FOaStH!U>`3S0=y4~y>%9j9?zhI zHFT>0(LEIFpz!eQkoM32W+j6{AJIyFyY1p3o8tbQCiLpX4-xDi-d)5N|FbsXbX1-1 z`kO93HhJ+6^X)I;6>SjBGM7z4el6jlv2TGt;;R9*Tkw=}PSoLPdJ|!l0m!!+Dh+{G z?1?0pPtqz9aJ5dyWdqyyqnMmL1ejWYO`Zr>P2S>b29N_2M!^}sOwgBWz-(kwwht!A zNL*nTXd+xQF`y3b$%?KNMgsp)Fs-W%q8+&drj(bZONRHHc5f1CzsKv7W3q}A!#_4I z=U3-O%FcI@Ez>;Z#-1zmPx0Ygxef&|t+R*eHwT{~dt_z4bae}~bIYLXS@IbgVp@5z zba1RM2Ck0rpYVDBto^ZTk6-AIHlaK7C)=HTKRHQX!JD46(M0XDCFWA3?&{aqrAnx| zZ(y~0L~d69@!@_D>~WqPcn`E9w4%3x|L`Sm(;BBf!db-X+LpX*)34Ea*2 zxGZPNhsV}ogw4pem?89@?*|~&_ptma)&9x3zF2LFSQC+8|`VRf6T2)zv6j7_F7#SKDrft-#B;K`> zvU&K>r6a0j`-f-l#Q z&_4g$yys4RvgpiNEw#fV&diFw5y2PEkH#(u)!3fXi7O6C%s3<_jmNafMY`~Y={mL~ zqo%{K(g`3t2(b{I1AA#r6tjW*FOv^a6XP)=${f>U94y(eWGmVPky^*J@22i_{V2H;ZDF!$qTN9ghiW^CJlac;Vv9{pbBE6)Oy(~$=BKD}{UQD~%-O2( zbtDm%S28@F?Ca@4)p1{$xEZNnSeI)d^u4z{ zKscX*LXPwJh2OokF4eUk>byNcOx7VkrSd6me_DoE-D$P=g?xW3rzT{V;}G(Ub*=pF zwU$`Q>r9(aKZ^R60Q1$+cdYB>cRV&7rem{3Sp$%L!svBF`he=&ViS?(a2vN9#~}$V zk6tBeHk^<)i7S)w^uD`}mKEdLC^%_(Xfe}4t}Xb96Z}uKjhipktJU(qu{LOhNYrQb z0OSzWoBL?qJ4buyJ3MTHJqj&IpUcQxicE^0AaJkxISVkD^SXE93lgJ(gI|Xv_*WIE z*#KtMw6;XsA#%bYiN49QnZvh^i{;3a0$ zw`i>bq0tFUWb$_p6uquogOoiJ7g~!g=I`Fay8*(30RXmX?LCo$9QrG9M)JAFS;`t} z$DsVRT(7@e-!WqVQrRA!==GARpFnO4`QJx7_@mCI#BF@G)czP7bjoVbrug5zz=5$M zXh$LrmR$0?E=i#h;c4Y7XNw>6YM5!)do3t5JbA-=bd2p4=Yov5H7=>t#J2%DcMlWj zvB_cwe{a=_Y_x6l!({4tpec{`ps(P?bIT=xp(U$9?1Zm|`yJt;cRbiJ7XU(@D;~*( zdg){A1bdwzS=Zq{$;@&pQV%<5a6#lYT`V0NbW6weeHq>x_r332lPoNU3;t*!kozf#jUS%;DlX@)pE&Q zilBApkOSkJgDGc^7lUKp#UA5&Cqg}Saq79)fF>JjIu8|%@4C2Yt2~~|+51#3R*!I5 zW-@`q8>2^RgJa?n^1n;)H$d+g2-*6rCkx6r7(J}NsW5sl1)>Pq4*L1mf`&iW)YwF# z4S^-tLz$BcW%|t(u3(|pWWx@e!jwBWCf{HGtiHj7Vh#13EA9YNYfL70Y_>tTG0Ko* z2NxURT?>M@cy->e0~mRX9oI9`8#X5<6074)?w+7tk<~6UbAC4CN&GQu#@XkK;7qa9 zwM;Mw>#`ansMC91^V`LjW6ApnjvXKFUQj2t_gI+5INzP$^YaWZ}$vh71df*p71U z`-{fxZFiiRzn*%%2c~`s_ybLef^Lz?<8JFrreJddF2+)D4{geU-P&c(ez$T;RA-YDu)xqg}cj z3_uP9DDL<2ds~nru#+lTXyaLfU|0&5D- zI;~W}eSlJ0RN5y0gMlka=$||2%HA+=ntUm*ZTV<`qAOG9Be{kyKX@0`UZ-G zeP@p)tmpTSXzcFrmAo~s*N^*TbIPuE-OKAhy(&m?a*X%M7bf%{#kkIlpuN6#uYjB& z7nC99-zFx91l7-Z;-hmIiCcUVgdJjY)$&F#cL}|EU1VGOv1ji(I8FWx^#Li43XR*E z09}o(@Q5i6K;x73d%7xK0_ux(5^5if>s=?C{V)V*6~Fo+Hp#gc)Q6Qve)Dpvn-YQ# zNAq;Xh@Q!KV7J{}PZv7xcA?+lzdbRx_L6@ct1%updvb0M^%N(?!ifr01s(ZWc5t4W8)g&hfwEbY>VG!4 zdO2LaRQHU_o{1`}Mu;v8VLCBpc3n?z2jBm|IL&mU^D^-Twl52}nn1{AVaIx1P-VjS-+@3_To>q@(5&V3!spakdGV?s6(chXxn(ykH18e)}b@Z)SV}WRq`CfJPqH z^=QNp+w%kMcYYamAc@7+33FK49c|d)EcsSIvk9+w*a#2q*f9@#JAT3UxfWvX04YE)aNMcgu33>{cB4z!}Vjx z_<2|P>8F#FcqXT+ZOJf1;p=;8q9=q>G+S|uhdn%dQ6dZ!`#JM1FB8Oh8)npq&PIqG zP>tMs-_(z86Gtf(0f@!`UN#$pPr=|1YGlER*eDZwWIS)id$xGG!Eb^?(jQ#)T?u>@ z1?d442AG!sftqFZ5}u2;y6adJA8$Ou_TQ8T!ASw}%AeWsFzU(A_tFxHs3B;2&h&Vc z`ScrLe%MdwG)9nq%su<`E=iD3SEaHyG~976HpCN1Gd_$At0BMx%}pu@QseCZ5+C0m z6#}#j5EC@k>yE4>Yqq37&^HWxe>8EYqA!1O{rS-~T7=)(a zF|NlaF?IHyU!wAVE9w8Rxc+}%;k^0wDs(n4;tM!^zw>&%9WhMp`u+bd#sAOIbfe!O zzA~np2K4g(^A?Tkh5&An2UKEhy0HSPdsF#|5+!8X%J{gNx4;y)QxCauR^HMBIMKXX z-;zRyi?3@$OGs}ApvgMDS@D@zjaYu#2ddUMA#2=gBLY+D2PEDf)F^|M@tE%l zGJ&h=sr4b`UIk0~HJv>a!a)u#;)d=|y)ct^h{X-tw*@9jY*>A??Ebr%uN&rEcI zo)NYBA;SE&YH=R7gFwFJpo8y}TnDeg-T^eiualMJXi=uS$Uf@byB{@M+&xbi;XOcu z_uKir;3XqLR|`P3C5^zS2b<@> zzgSbNffT+eJO4uE)A)BySC8>Pdlty_2wi%e9X21rZ~*xNdKC{bRG}0|NbQSsQXRg1 z05*E;9)#ETuGTt~EFSXhw)v4vT2=L3Oq9yGpdX>#_mhklAvvk_;XUM4PHu!})2p%Y z<%*{5juzr!?*@+GaY&(YbS_qm_5qS0CMBGVYKO?p;6{+)QGgj}Pq*bEEdCts(&8jS zv$+Nt@wHqNk5>Ti^aB@~RsJINj+L-!D{rKSrS!~w2BkY)SdG6G;6Z@^VpuJ}!!tD4H>i|$9lEm5&!k&|@9~pm6kdWaEfotND z;POn_N#4xk#!@EIoPisV&>mXuJs){UFSJ5qq-V&68u+KKIwfC5<&9AwXL&ibZFDdg+R--`30w zutD1gLkh7= z@?Z?%fT3Vq^tDTLA56r*|^G<;&R!B=jt{tQ*@#WQ6(ri7FMpc$^jJnT(>!@h{?r zPXuN!GLTbBD?raGjrw8+VSCd)gwM>54r|`B`e5kD`uJ5NY?(ef`9J;i;)Qa1U<4qR zF;C@sYe5-5D?>Y4+h)__2VkMX_gfY%v)|Q1!Iv`n;H(kb8Ozmpe)yk8rib`b#~nU} zayQcIsI~y}HntTBMRM#2RC7weoMP}lb6MvlvetH?$~ttgBqm?*B-@rd7pO{X_dLbh zP^+A``$8e>u#2^Aa%zo{>3u$~O5~8b*{6yRY8X^`whjgbT8dBDZyU4)hZT>5SwraW z^Cw(&dUOW=`v;!Teqq9G2~LD;rt|5K5z%bl$%v0y7}x{P>4GJC9^`;%0zX^dL^-5d z7YV!tKYMTyw7>a)0$;3A=L4_bh@w_jl0KHMIx)jA&T2fvaEdPu_3y`dHm^LC1Kg@U zS;4;(|Joe!*&swUi~s7|R0JN4ppFw@8Ju}wAFws`;lgQQWt?pQA%0rNYe?UU+4&&b z{1WRe`Kf(?U#cVybJXhTUy^5=FZ6=(%r6ahT3*RZi=@WQzCpn_TBEM~XhHu#1*^Q{ z?6BR{Du)OJ*6>TIXWnC8wGCZ*@bIPaXH#j)I)pv~ocw|MUGUly>6w#;JHO*!LxO!% zHP#2v3*ajZ_ed9_?8w$YJBPl^M$MiIEphL$H7#=snQk8X zZPU$x^KkSxz#XqM2|-rU=LqP_^}rqY;!z<>57LD=YhNX(`aNm6e&onzGWpX2-ARZG z7S?>9)NiWu-UEb#RjB&izFN!xgf*Y|-V;qj7trKniT4`h+wST2{2xAIX3?c}9@WOh@Cn^}%=ZEq3ukK&Ql3?>mFOPsH#W zK~|}2s=EfGQpk~3pSaw@(JmmB*BhU!LU!loU`F!#B7;6)@ou|2g42bN2pq~v#kl~U znxhZX$yGTho+c|>qbbfMq@qwtBj!%53=NE(><=~i-KB}Y1U)2bfC?<*Lyv$cqP7crofWYW(jL)KiI{SR52lV)kE%*V>>Ei2cjieZkf#;mn_^!%CU~axq$Qc5 z#*e8pOBxxg#yx5W;`L)V01agl-w)j~1`X=SYd0-q&DEck!ZelG3#!B~W(Iw5y7md< zF=RX5fy;ci7^ixUn^vin`RF=3jEpW)Sml> zuE9o4f3)E5Z0@E%m4n}H+hX9!QWutNik#-g*}86{zMXIpybb>T@Rs_;x>7q0qxbXD zoq-zu^#(hGhEJYF>-x7tG8>eKEnBnAvhTxuM71Y*wgT>AWMni`;Wy6Vj#EjMu^$b0 ztgW{?7|L_muZ@#+-|d(2;$5j-YIxeJet=d4?lg{%3;1AK+g7~(CGmTXS8X>3`DfbL zoX@%PcLlZTee&M#9`?|y_abi!@~;$M`cUSp;@x+?oha{YG4Cu#(8Rv=(?njj{{vbX zM9<`DuN}cTg)#A@)u$>PUG>iK2(V0IQ+(j(YBA)u#DnH#T_n zcM3Jz!K-m3-5&yae)+e|?BP7lP_K$cW;{EpMhzfi)KRUGYYmySywOv_ezS66=bAKX z3Rb1s-|h;4^~d!I(eMZIiR2{6bJ#O{VPSS;%yGJ8!hC*2HiU?}`w)Fq{tDT6KJwx+ z9w!!msWbe~6~G2-tt;C6Dk<)%3TlN@uO@4_|GsswH~#nCXcoP*XDEJ9?tKI++Ux2# z^o+PL1JABF`KaS0I+G`Pza(EP>cdgeB3ssTC+m`J%S>xAzAE2&P_wUe{AP8b9}|gs z3X5X~bFD>{dau6;d54)D?M;5M9k~@vO!nvcpEZRmNuK-`${=GqI&C3!Za)1#Z?8`# zEyyqr`|cicL4M=N1Z)`w5V^jm(-t-xsP~s^#1J=k41Wg8%z(9`;uzG_jko!oPfa#f)n;3L>%=` z8Ub*$OO39W1p5}k^cuxVG z`u9a!e{O#ej%eL>+FQQ6pP2?#>|tEx9LB6J(67!?*(pJgy!qtaDerQ~$zE8c#IBB5 zDYR7g7gv%NjU0a1_I2W#ToJFW?ao53 z)9G9dM(|i24emuD(hq8he+9V{EkKJ`@SiM#b0i!RFqIAlh;f=+c7NvbdCFxICb^gVYRXM>oeOup&EzuoXY46^}zF zd#|u=;qdM`hD}>2wiWl^8$oE|rrD6;0B@P6C|^1@I7T?(RiIg)p1zJ-bk3Q%+?p!T zrB1+%AZA$I!LN2>m+gD}pOVxX){>K;A$v6nWp=Zm;(#AItP zRe!>WA|!<7(inbvB;wLXFWKaE zC**8+LPEs;?8~2m`dz3^XV>#-cA2vw(s(}4C~k+E?_v{zh_Vk>`a{W|LEekvt2YFV5PeRGtmblUSSSfq#W1AsfOvrci;9C zh_c6A)09Tx+!VI~d9uNk)p*9SPM9KaI0fu9=ukf46;Y}J)xrb2-eLi|_r*AKA6<(O3v}K#F5VBH6|;6mO$wYw z%BT(*%SC+iMXO_Cu-8sqKXowQL+{-y3(dZCpEop;!yiJtkCc(uKHu24(%%a&aMZpl zfBo&1qXe8hk@n^wD63yD4Vx~9A2C_a51k4OlzuH@ z&m?TCby#vsQ4mgy1J0{GK#8l4JdM6N*f;#=BsdYuKy`5<=9v*cxuyl!gs2C0nXlh> zS({J^**3i4y<`CFO8#TFOvH`g9n#E`_MQ(s>?32@97U)vO|VfvjLaFDM&N8;(w1K? z4!9s#f;*Y=1OW-+94{URZ@%eOS2Wec6(&SFHRJ-uOckIKnsv^=b@2-X_SCvLW&S;iIym^L;lCG@TZNHwDqp(2@jq9Jj)ZQvs2O-jV_80O|0~5R{(#}8|IL=T#q)?Vo}SoB z4Y(XkPdw4vs4i`Hsnoo9xAXBv?CSgYc3a3l3Pt-JUxN?V1e{zm4VDte|`LN6Tfwi3U%vl zt&R48N~ZP?P_N3O^IBb~V`<#tnvKXn9)V~P_ho8k$=;gqR_%19am)@q0| zoQB19>b4{L5FNN+id$T*CS0%fpca*5aeOhkl=uae-dd76AvASqtRqNI!wF@fGl<3x zB7CJd_mr02{HQW%;XCDx8&TIDB8X$qbN$Tn6(AD!tL%>j@i0{hQbXk<5WMfD zrqXiAl)||HG4=Eu-w$O0HL>WZXdY21^a`ugy3;1Z6QD^jOFA+TQUBb7h@`XbLK-f$5W z#B51|2zoaD5V$#0{!r)Yhv0!JfX3IwE@mTV;y`TL;FfQ_TVpmMLJ z8J57?qrxK;!19xlij7ena4q7#m_6{d(tG&S1lQZ*u4l;wNUz#z>tb5J|74T6FfLOj zP35|#6!FK%w7j4)Ax&KY_a#3HQI}{dO1_D^KVu{Rh4+GL&v$Kxg`pZirv-Qx4B3?i z%;%`*2zzLDNK}|F~ySPvLTip2ZRM9a9fiJJ(uWSUKqObtop$%fE@8l;(qB^Qqdo7KAl@u_v69ax!;~aSS=Ezp+ zw997CH&L{^OpixYJmU#N3NNR(n&YKbshFR~L3Gm!oGkStc1^N15Ad40h_$Mqa!HQ- zcC7-@DAH65$`u>d6>v7FvOH@Z*cZ?VUeA7{6Ya28(TS}SYall0@b9>O*8Kcj^Ii0~ zcWiC*I#sI~2h8T_xEnZZ*SBJZ`oAZ6%m{Q()|2cnwq>egzfxIRYt6xIlGA6s_~1hP zZ^vMnf~Y*v(TbtB(o04_ZtyZGP0?_JD792|$3Y2q>Ct0!RA|Hg9TYlsd(?SPDSJCf zzu+NTA&+?^%9!^^0a>D#A8?oPO0muVN4q_GMKDVNqI^K9^E}J@N4x&};f+^InAFyZ zZtP^(^Y1&HaO@h?Vd|V5zeMAbz@nW1$&@ko_!qyW$X|;5 zgV-5CZ5)@5Q8E7FjN-d!fm0MS^J6^!5}Vaf4nobR35k^6+A1kW2r)j@uf}L+e-+%N zFZWS;8kt_*gJCTmQ9XxrQ3WTh*y-}fj2b?(32Ts?w>BWT0tf@JNkKZPDEhCy#ry@04CG{gxPW%tG9eoZ)!Xrul4D_$Du&em$BPqI$O%jj{A9qO9cwOleafSZr$x3LD{BkIJZopQGVfWz-)liRe$qap}$u1l?nZ{J0jOFLdowty&x<@IO>QV zSRM}r{IFvuz-oz>ReAVM@{WiIbAmLdF`mORUpL1EE3!0}%cOPWTW^`Qe7)XLCI9@O zo)zl8l*HVAl%(B}_%tF#5V9fdV2mz}kBbK-xLGn^>@vCRb6d?vxxJvY<*$7Zp#+?< zAsP08MCLjF`O34z>@xQdrZe-52+@xUEI=N}LnVJA29{6(L4BGJpD5FI$Tj%s|N5i^ zyZ_9M$!u?mbUZmOzNCt!>34}Kk|h)kkgF`unLa@t5^2?Cobdqc2P~!|9nYfev5PPC zwI5+Ay6zCI+V}1f_h(BeB?-+WRMTsjOL#z}aITE*QQ34~dka~Bi05j}(w}Aq<~nw> z6rte8iR=$UkT=9X5S*pvK=?VGs{T>VJrq|EW#x#uuV412z(y9HV^Lew zvX8nNTa>_A7JK8&c662x&8TIkQr57L5u3#ZN1Fw+Ea$t{))3c-Tw1pK%Ga5a>F$fy zJVMd@Co6dhZrZzoiJMpiKrNHefdP70wSA!Eq~%Ex>dvL@$Dq2Oa~mM6bH!tibSMA(g`XG?Ksf7_ms)KmA;&aX?1l)Z zGB*(egx!h9EJ$xFZCe1J?t1*Kcd2kQ9oNUvtayR5X8wP==D8CblOi}55I;ETR`!m7 z-=F7+vgxhQTtQ$V;ciLO_PJFGVn4mtr~3mTsj&hzk47y|6;6eHL;eli$`VzMgu_%8 zMTNQ_tXbLI<{BV)ks841odm9+75ZU_QI^WIXa&ToC8uI!Rz!8bT-eucv}^wVprRp7^VJP=*fH`|*@Eo3!4`{BRwjWH{9klZvQqn{nxlZAdS;J3kHa%Po3 z;n`_8>HQJ}uG*C2@6U#XKKtouc4n0~A-_rXzduhNnfdKAX7y&4A_S?thPhp7MqwHY z^Hoqr9zr9A8OmO2yI|Ccs`lm!lQSK?vkSGdS z>(#R?4~J$O!;a+YfVgNqOl3Vs+q95sOyih}eiZi8TiR4yJ@dc}NK0Se+GQG*QSXx@b_~?0)Sv ze1yq5pfYH}R(XhwXox?2oF_+JWysTyOeJyln2`Sj?W`vuKR+l6p7DS~tH5eL;Mjs+ zlxslY4CK}cw4EIHUO*(wC7*V?c8cm_s{>0FV)lPwnY7HdL(pc_if5Wg2q=KJRPAO`Ioz-k10aJ4xEp@hn% zH?oF;QgCEa{CknWLZM0C+wNoFGSlQF!sc&SkT8dOD1+zt*CBBv84?lCqB5-RWU2g~ z&FYm+=p|OqwEr3=hdBH823m#~))#?)2ZWp9Qm)F9{O>#8VwDn9PHe<4ZCoyWzf&e^ zgp?Qn+_`rIM3}V%L_(9&xPhB4eo9!1(TQI0W*C4Img^Y7`e^4SAeW2{s*VjdfUNdP zvE(MC^OvM8(I&N8czFR#R4*WYT={@>jvQ*a?u>hD9Wux7BXj3|oDW$N z;?KYFE|}LPbyz|h?0CVI?=1Zj^pxHdM}Is{&`@dknshEkACL#B04^+^z2BW~LTUl< zI|LbkGNqKbq{2lf8outjm@s^emYw4VR9c`mnIeD(_{}!w0~8gU^_kEbxQIF`KYWAx zqyEs8`eXc+a8`uBK0!qsteWP)1lQ!XLUQv+re(`6Lh4EV96a!73W8RB>s|E;fcWvD z$p9*0Ogzf0rl?#z%FOIu$S;DsDD0D>E~sdO9qBl_Mg){pN`1(DF!1jT@${w$bWC-o z3mu1=ki6Wy#sEKFEOxhKjRPQOxLPw<0Hf>xm-a;-;JHt7Wrynzlp++r&djct>f4d= z(gi=x$@cjvnLp7Jaa2L;1Tw~>(c*0QDs;A-yn`}I0)d}V0`Rvh^1KNOJ+C#8A%cg; z34}X-pyT$5$q>=Xs@*L*cmfm|wSN@y1Frq!7+7J9L7KqX7u}k%D0>i;BCBS(P)s^q z6b4FAQJ>Uc5b!fmS%AD?^W2mY$)8BA7q46jw{!v>FvXYs*M&Z#fZu2~uOUL7+FU%s zdJ%knHHi0*qMLA8J$DftnIRY@nJ(zUZHP$hkL)1RPD(vWkfud_yN1cyY->&wbJ`_OoiO-zh zZg-eJZ-V!)D-Uo1KtNGW})p&(iW zEBBO6g?-fWDg%`Dj^+; znqduxS3cl^m3K?r?>6G_cKax$-WL3(D2maE5()-Q2>DUmLC4vS^vMB(jQVl3NUki_ z{MWSCsBsHsfGE0<8O$n-A*WM(=DyL~96Hrw-t@k;coibwn7&2#Ulrij;w1_^u_LB+ z|0@dBaufP)jY-~%Daa&;SCQPjFQ&Crrk^Fu4nd2;@L9Pb4`#b!(Im7g*MJ@6*$NTL z&0sSZ1Aq(JkQxWyQ{&`qepcm50o2KoKsHpeg*%1{_xFVab;!|iv2e{ZKK{?q zMzsXabheTtrod|J>?piz<3_GsmY;dszC!4vv@dpCashq6rfa?*JQn5$6i1AEE9!~v z(lE~@+ZyHo=_P5(-Hf(yXZ%+g{usLZY*R%1_bF)bI&$~R^jmO0WCQXA{JulH_K>)> zAT)UTndO|p-{xx5z(TdhNpWy5m+6F*ROCx{RkjUrp+Y$!>kk8Snljz!WnU8tgG;w;Eh~*!Y2rF1RjVQII)xqqn~VO$KcK z@v+Zc72ls^L7sDloaxVw4-<$E<}3HllCM4)wI(v^pLo_B;zgHwe^_;FLoqs$kH0pio`<`W-aK^R+&unFBGork<*%cq;vyM;O-F*p*5~=*_67%)>0$bO zN&T;%_Sn_ra{^PUk`2wxG{664O^!1}xhyC!r#b>i>nq79Afw)^F~3c@xHcl_7YB2(C^5!m~c@!Wlh2oRpO9 z)ZQU!dO1iUidkKUJ_*{Lj>ult)Y73_x@itw1kq(jjB7a6Eq-6UO&S|^+?aRa{ID0a z`t!@7<^JWCQ(404Z&^;(piPj>(`Q|*)fuo}H5~izhd?(n-m4xT}FI0sbN5AF}l#simBGaCnd|8JK!Cx*;2Sy^h%=0I*2?LYZsd5}HGMTBa7U$GH zBGc|``bQzVz6p*!#JUeb9yI66nu<3O86aX|dP#BI`86+y6D3VC9cv;6PgFZMKEZnF zQ=z}-MuT$UgpMi*8S&(}%=Ank04C&jHV{+=LsAbWZaPlu{0W_ZG8*TElNf^I`KA?W z1+iDlCC3?a$KCxaMNA!%G;br2Uo#hXfT2X9C`Z;qNK<*2y$pibn>(1Pkns+wQVxRM zHY{?>7I_i5{nMrxH*}DJ&ctKA?FCgmZgl zp5B9~q{yil{+0SkQm1EVK7Nhgnoq|$E>oxx#7-qN_7H)P(oH6d~{1wzc7p;N$KV+hUTRJ zX1F|74RFuI4XRjn8C&#jAgz8zStukN?`197C4_q>ia*E;AL$DP3krl2omhoOw({<9W^7q(5A0uP`l$t~;rQ0+?awE6%)$ z^Hb-Mp!h}yaB>_v1=hUsg3$>(&&w>@+DxhzU3f^e+8yOa%WE@mRbi`L& zRqF11zVpvz0gN67T~ht51bnCF9kBfEqx~%tU60?52?f9U>L|h#InQR#jigO^%A_x- z2tGjHB6E~-3=q0dnWk!y`y}UPH)7+4uQl0Uk0o`p6sdca4J+%unnuEPF~cXNhrpTu z4=2e0QD)yyu)D^{!xpu#d&P_kkyiEVM|3y~uwa-;VqqdQrBr!UJ`{WK-77AYc=?o7NSyNXk)UTeIGDrQlb2De zaW_W1LNj2T$fD!v7G_oeV`C<$_K4cBXjrDY`dmAP#s$9u=`-j3%dOwCSt^RFKYkKM z*Z}ncY#i`8&;XKd7Lrwq+uVP?%t`Bi*Ao97v!{Ke%;HIN;9ic&25_S!d@6m!IsEBTakY=}Qv2O}#{edm zY-$^}8jkZIj(?4Eq=n43RMLcOCY13E`yQJbY1*-ia%`i&u z5BpA_@pcjwqqy)_-pj@Pta$7-#12H%q9Y^%8-#bBD;LI1;sjrUZYyJywHYDUsJ-bWSJ$&R@0oZrKC~hqH z-;9b%f}8u5yG3B7^6z_|v>AB#&nqCLj|ihBb9A)w2td)@rURs;_LTnn>?vhs+8V?! zCltfA3CXCrr|ZK@?-Ap}Cg#4<|CH1j>seXv5upI-AT|%eW&>|e__7ZV4}YB8EcI+( ziSPo+=l40V-Wo;A!xG^~z|;LcM{0p!F5M(#>&_pA+JUy{;%x(DV|l@Z;JC}(U0-{u zs+-5*UQ|AAOW}XW z1Tm_m`b}P-RMM{*H_kuCqsI;NXj3=NqGYAhx8r@3mNe82izsvs1BA`Jb5FVrdmq&9>HL0CitpOvS2NJe{l{F+MW4SF*D}7z&U5J1>ETz`jFS!Oyzg3R7`P=5Qr)&Ph17e50pU%1Lw8!p3zX@mH>Zl$!SBFNo zpYC+fR6(CPJ(*Hzh8e9Ae1fsBz6Q$gmWpz(Y5C;lzmWyx%p09pdHhi7S#sHg$P2X$ zxEy+Xk@HR~mLd@=`J&F^Upa^3ER10hWgFinGo5npw^8!?c8pZbnOjMq)G^WtQ76OU zPeRqnEUYJ>0>tM`-aAO4K>mqP^lv-oR+G6^y89nsy;oePTycZYQ{jXe=)XH>B}>Ci z)tN*}J!!DbzO=vinz+A-BoX7h4}8cmnX8$?2@&J7F^){%53kP!NO)yO7eoV4)6A2_&@p|)_ zjU3b3MN*Z-61176Ygif>h=BAd5wGin#92H24cKr}MBB#BL<`-zC_vnZT>U*5s4wW2ZX8 z5JJvQCDp}6Ma2(f5{kfK99G+7f1AjbF`5Pa{!PEoLSK@Hwg83il$G=eURcnqH4_?oRQMF^k_D#Ge7MY$4EjAm!MkiMqpJP#X$U z5fa&KlQ5~E`PPYtli(BXYv}>KPjga5@9M$p^4q)o0d z{EzJdbSk5K?3_MJ=H4;NeGyi$9x4!zpG{Wg>`=ukAA8qzQ3A^;?p{@;^0^VkWz5q? z_t%9z!}R1#BDHC5IvwV&doW3hIoa)yn6eUV^^9H6{Ai~NigxEceb}%tC@L(^>RGr{ z&3hfWFw}F^=lAAKEpPcZ{7DO-d_h_nVaNZ;uvcQ3a-d<#ngJB8(AHYP=g@(yv4U>@ z>yGlb)W;sYpR{-y@v{GZLAXd787i*xuD1Tq?*298( zpy8~#F@l}NV=niTzun~`ech8Bdw&A~Y+uK@O;r`M_#hp`OF0%=WsGfJ&%5od{zdaO(bUWKy6;!?6Qi=m-IW?? zjHz83mvol0InA@iY!k@2-dW>J^xxO*f*%hGWOM`AMkVBP#bQN$@{MOSC;qZuIecl% zCHN8Z>M z!b*xjn$OQpS#Jz<5cXV>GnFyDOQRv%@_`Ipi|r7-MkX%F`{43jL$RyyuN#k^fJPLq z#xY+3MoI8*v%f-&ehKpMjdkIat-*1J~ROQXZ_}-O8wO$1hd8+k(sO3&MxjX zF>j0}Yx*57mg+dNYcc#}e^CC$lENt0O2%%~W`Ip%Thi=r2L+dDmuKxlWtrU_VCcR7;Un>Oy9cKCJEF;0d08cS zmUc1)()x}xV1Vu8th25A6QXG((#FuZKx{K|B;*5pJIPH4U!HC$LzbOT2T@d=0fcbw z0wHEBw{zN>qh(+++MT)dPq;9uUKk$`@9Lv9{2SITi2b0-lq{e$sy+Pfai2K?K=q$k z^wzeWhJBTdOP$a5=s|mTX+*4;85bOp{iWve&Riw{*rxCr*@X++ygddoiI0&{3-0Ta zKM=?Vg#fnU!~nQON00noUDDWBvMJ;p5v^DR55Fz)CMGnU)<1yGw<{l#Bg~}8jjo8= zA;hI#&yE5N*W*qI$+hv8pemJYE@`N1(kEx#zp-T9at!kl5d79ZF^@bS$l+gZOnHEP z2`(+m7}`A|QUr|iY1Zt?jeMC-3{-Ja$WY{{LXK-1>IHqf$K}8Gc)^Cl*2k2@1Mho` z-|dh+Rn<1z?B}cmPLZquC|e4wwRd5(5U?F!J7vnk3pu6#c`Wq&BaLu)p=!He^id%F zPYHH1C~>J={IbP3T^n_gYnr~=X}rQ_am0eVxlu%`dz{v&B2*v57Y8<>OcLfrlT zw!v|9cgbc>U5+-tH;RrUWCrl-jmnH1>5W+QUX*o45#s3Xm%kBLINH~ls!{?lNhZ{8 zP1E6aj3nm{2RVpZU>5!Ib&PH#QPN0A|8$?Y(Z#%B3j1JJadgc?nEyW2#@bFNnFLJk z?gG2Wy+-Uyf(uEy`ybkv8IX=0I=#!%-qB0lbCd@MYY%UOTVMcZw3tQ1+j0q>7H?^{ z*3@r<@e{sR0>-n>zGCZ#w>H@X8B0Pl1WnMP%F2sh46Tl9+f$ft+6$lL(JbojAav=h zvNAwvrvjyeBN@+4>RHo#Mk;^0%(~RQnT_-rt*7A&UQ!M{cdOFA`)pHtpemZ}-RPup z9plOmd(NSd8R5#?8eY_5y$c@E$Gfi{@WvJMO?I=8XK9opa^5th2?Smj9;`^-DFAx8 zm6{7Ta*}pjLH5ee*lY7I6Q_yCNyo2If73cv=I_U8JNi7bZNv&9U6trzX6~)yk=i?f zSX2t*$ds~zLS=L!%j?X~dC~gCl3nrH?j^679Dhj{YTs+9BgiB}Gd`~TsA{}ASB>$@ zc6jdpGq0ytfcclqic!Fzk1CyY#^ng`-h6@kEjoSpy!@3c3)!H3;6OmrVHSUF%F!8rxt2dIO7|>8+p*wnKvA0 zjhfD#MG>o8IOu2X))NG*-2g!|ycmZLxDdYGEH8puX_ohT3&8<)$*n}&cPI_xURaU{ z2wf2roa3B}-)s9~I7nZeq#X&1qamHA^vUz$c;7{sj)6Md;M@%^7@c#|KG&9`w&be( z{?ipnPNPyQr_8j{xX8#SeSGXx!w6R!UR4uWce2pfaDot{;d|nOIRShQ&*9pUY4>Ix zRK+rLur@#H0E^$+!P=Ha>P9T>1ooDe{-&f}R$UKc1(?AW`4rtn`&62(H14GJG#t^F zp06`!3H@D}94&peMTQS9cGkMWNvQvJrycWKqby!$O0)9-yf@L3$GROz_21Bc@qmpr zx_G}ImF|<^Pm$a%DdHZO2zu@r@_JUQn1!Y(G@Ud?uDj@ndnsAoonl+u!~CyK%z{Ca zuTXdI5L|HlgLaI0ygiK;&G+g@#=F@kaHPDxo5xC2mf^C^@yU8bZ_#SPZme)`e-7H;w zrpL@(@81?Daj=`9DDL6E^BMC(%w7FhK1x@Y?dr1&C5W1Alj#mPvO0+`zKwK{-)u+p zDt?=bw7qM^`Ge!PU;&Pn|;~7 zMvHmj?!y(z_kiFqNAq6L&H80TpbUOj7w6N1#EGkTt2M`~M+SHSl`t2mMJ|^DbyAs~ zG2pnHPTTE}CA`kmm{Hi&D2+ZH5s87iKD1d+m%7rH!Eewmr>(4x`Q2OJGBOjYga9Dw z&d+k#!C8CR`MVmKr}1C-JCJFQ9^pA%+d8o5CRO!O*{Fjbi5Z z9+y>_F@4y9)Qqb=76hJ`>dHkK&G)_gU<+P3<>aznW`<<0OX}2#F}ljL;QkR}C|flw z|4(b;@7{}yM0=*j7+vwpcoTqxgx3T-GnNG(kX1?9%U;CaIsa`$2~ya1Jg;)+0JX~6 zG@}s$Hgdb4xslo{>}xA8Vwgq4l_}%d#9gjMm15#wfbRz3zC{4UMlP2#lERqc;l?+z zhm;IFj+09*l|_;uRs*m6$HKn$o|Ww4bvfpF1s@pvB3>aS?NB|ePxC0+M4Po9P=55C zupH2lX>eHIDX9g+?WwfyY)-6++o8gzOO{v!OEr}1spG;Zszdj#?fH$b-X!7-x|}tm zkR`kI9hdW|ElM{?SsU}0ABAe`Nk)C&Kd3^RP7~A|sZ*gYJ|#vK^-Jx_DZeFsq)@DX zHw^vC+H|x%@5$8Ri|!j=Ywjjb-K`B6j$CBr3U(h_`1|_~p2GY}$=bhp6E)uE69v|EsYK zXqQE>@*ePTyzwSXmFeAEVthAgT%u#r^)^njPVo*^t0G*L$Fp8cv-qhbt!tM)3UJOjB=4Qkcx%Dob^kUgV@FIa@lzPec=J#r^Tli$BJ;!3fTCEw}j8~aw7>X z0UD@3m0~g=qsXr{)^yy8_CWoDI0#Ndea&to5AWv^l!s#lF?#+Osq=d&^twU3TwwPs z2AT~2@(*!^5t>SSpJGquMz4|x;!6IVNchQj;&x3|Q=acXiY`hJOtdU{%rSYM`kiQsIQMR*$-jTDIMf)Or&?VwerVff^AtZwqzPoKu{1yRN|tXGW--2R9Q ze1*UD{>R)XEnb?Y6Szxq@>Xw5)b2O)3amkzbg_luB~C7!5_p;f{~=a-ZNN$}Kz-51 zQ#g;DKb&!>5x`!k0n|;>QR=!l|FO>;Zg~JT)M_`Z*D_TebWAA1SYv&hZHv0};%Smq zh8;73zd3X^KPy6%f74$%92d0@G6pKW#OxeC?!BK07bI5KAeNn3Y$$1)VHR(JKeEW( zXJ6(9^;puv1&FST5cHmIwL?Q^@x*JYI4h-yQHA`^eus~bn~yI&T{999(CbHx?zEsD zC%0UL7^}Gw%tq=mYyVfwUbfzfeB+H|4D(_0#iY7Z_;~_$C-QusgXx1Ffa@hVs8OhB z*6xBzav{ZuHh?IF+6FiqsoT|XUrrAfvdkb3|CpTykc5fZ$i?G^pQTfwy<;e460Z3; zSw=-&01ZJSLjLQGQq_Aks}`91J1J|elINS!I7vP+#k;D%->rgL!wd;ka(+Q%#8A(_1pYtxt2z&(dc`?5rluu_c9@28 ziu-V%B54dC2Tv+&qi@Z92rv88U7v<^ICip5<&to`8*I$SBVp^U2|P$459ErP#gMM; z7V%jcEm^zun9zsPWQc#p&52@5Ub`MQv*9OLtT*y0DgzfwbLGZ8^zRc(!Q}l0_LZmz z=B+KYr9krS^GDOs(0}d5rvr|PKQ`XgIkMOn1(qYK5BhY9DTn1BG4v;xsVHaqgS!}B z(Ce{xsm9dXDr7gE1kh# z&*;U6U!~gs%a?L32BFvY0xNPN+|7RvM@G2A$2&lQ%NBnXp1I<|{T^*UXYs%c+J&(t zzt`*TXL-1*$#26(UF1i2nU9m*;h;===K9pYj4}-%kFK|ip<(feCdRA+XT*YSYn zsi_lUJ#T|a&Se0cTpRE{vd$`pv1R1}Y;Jf=%Igh3Be;$aFo6JOY324mr4~7{sNU7U zQxc+W1|iP=0L+>IqFppjdkcVNag(<~&)ERCUN!S{gz54qaQ5Y@B(ezdeWG*?9LR{p^_q8x-gPKO$QAqq|3GL- zbb?l{o`76kPyU;wFa9qZG;R^bcG8WeocYP-qk7de71e^aJhVnhfsU7--P&TWHPfkH zSoq_C3jcKZ1YN?Iv$Am&j)5a#?kfetN%*~)fX)H(Z^&M@&WyHtDc}`{WQmqDU#&?W z3^7sn?lybl%ZaIPs5%*?1faJ;3f+h;xuLL#{DfG()1CD9-pOjMcLw#WkM?giaKLe1 zhM832>iN$vZZ45CEG_QMp}mk{$4)xfYAI+sIF-sT$Far^O!R{771SS0sjo-j`B7r@ z#*lteA(Tn*ouP1ozy&So)7`%Ole!)9_3)E9oeUF%Dv?iDiwx0*#~s+2!(#D~aCirlfV%vu&Ri!e5#4 z&$0@K_iOO}@yL7Q1)B|z!+Krut6W`J364IdX!B>cI1w|1Wd|`R-t{KUMbL^-%*(3V zo5k#D*}_?0jhGd~)vACuwD(W9g^gDr=w>q^)~q0eaF7%*vBNRtYUF%ZyH+vdTr88S z=Fw=@E`*8mg7IeaQl?P%c2hfInv|_qpIjWK@^82SEAw~b;Cg8M2g0DAsUqEo_33Jj z_1MsInAp7xr7P|Un$UsZk(Z`wOZWqqCvEgqwdy%!myru;}HI5II!TlnRK1! z0FsmtHcUBu;y7g&vMPU;K|}qzHIixj;w;Er$Or#15htQ8-Ti(qXrf^?mmMvXnq$?>fY;# z@JEMq#!8hV;{011KBc=uKqCW~4d;6%f($`@*F)SHxda)3XBo@zw$tk&BE~F$A#wpi z%B9Z3ID7I_V%qf}E1rcPKhHAXGr^+Oq(3V^xf0+*p-;>2fL(ov-G#V57}_QKf%rnQ zbCwb87tAweg>qGq68wHqTJ%Pt%C)?m+)8-ptM*yOTg{KRo1_J^Y{9Xp7MTPS!JG!o zyCutVZ>r`z`lI1VXok~k&HY!6;jylo1;PyF2=<9u?;Ku-Vm9Y~diy*Rp4*9j(QXA` zCE~rgF6;^}mjG3xUwgK*y>$nKHh$uTbC;&8oMmt*1s_ z&sprVZt{ZAa%B`Rw>ls|42*G#F#j|CHcD);jKDLKulXK)#<9~xII`jK=DACWQxdCW z@|vWhY}G=2(4%X^k1-Q2K4+ywAFI%qo}gKuep^-z+FQY-tSu%d+K+_kn!b6l*#pa( zF#kfCxqm~*Y=@vO_>#v@Z5Db~x^3dI>A#;M>uq)R;K#yy>HSwP zofhr^3)u`IYm`uUX3DKcVwskh{t7;->^|E#Mb!)L7p}`Y~)fGXD$hyFDT!V(b za2Qq~>Zp8K>e`S#LxARH1FT&;doUZ72Pjb0e9vX?_U=nKzV#2s!(;A800eZ zK&iHs!m=F>=*Y|aDm{5ATYYlq0x6SFDHdPjw~k*$To)cPGzU9Df5sIU*+N4M#M#z9 zR|L#yvhZlG9#C#ZxTb|a`!8gy7MU~Dj-lMs8TAole{nb5=K+miIBDLh8&TfTZ{_eI z*N|K$|0CU0)6y5-a2lmiUIk^=yCbW%L@a_m+MiFgxxMPf-F$9%TL3-Hdn?)Jdb!S_ zckx2QI$w(_Xj~PDX;~MM?{JuF&LftFTc21GPu-1v;CeYXOs(JYvXe86+?Ov!m_otF>Mj?B;SCovkC)TqIF?@m74fC)mS}1^AI$ z|H{e!I?6d;u<;MI{=uHb^v(T40x6MW9%$X{GRCQHNb zPBsxW84rxGdV5maCcM316YMz~vA+57B?sLZ&bqUSHw224%;;Qoe?!!BLc3~mC)WA? z!Mx#T#MsR0E_MAm%QI9j0&|wfP)It2&n=Q=a>j+b#Kg+ z+QcQ3?z&Jk(T#2JY8(m=Liju(%kyuxrfrVdK+#|Znwf4A_UtAV{@Ra4-C^}6gD#D& zAAQ9qpZaF1Xjli#!V3{6N79}TDO1x5-Gfr0M+<|WSwffG7D`+*JsbK7Lw7#_+<Y1fgZn)o$k}}DJ#&_@bqYQMJSe1!LNvbgOd&St6-isL?%%5i3JiJ zQApr*iwb#TgEIFt|Mxp1tNA=q(If+o(lGv*z@;h zM1`EVXHfRk`~sj2{rN|qP=gLf9Oi!Gfl)@&1)bul7C^Jc)v~}|>2E7gOFtF-cbR6b zaTQNOd9CmAgJ1vJ-3$zqM>WL9;#wV1oLTp4!Plgw>R^T~jw1(Df@}hKD$qHxZw@ue zEYDaKTEWiVQ*lbH5|m-<%2*@G2Oh*vil@6yg}z+)ouAX>2{NLq}pVT z5T$&s(&~Kg%C2&BNk#1K_en}oF#Abfw)$7G)I7QTTrm!MdC+vBX1B1I+L>HFgK*s@ z;_5p6@qEw}_PQV)NPmlng5iy$cNV0|#2B*PhQR#43w(HB!6a<$wALui+9rx{k0?Y# zH?+%Q;kDon(j?~DaS@}r#QPC0*#>JMWpSlOp&C-5v{KOTh5Aa#HUpVbVn_mP`n3xZ z476_d&ZuBw9@+XCNqzcFcq!EEFMVfx;%n?+D%D1?VmlJFjTC^G2bN7dAMcO;)9bu7 zP-+WTM6!1{DvB|L<3Sfp!y=WXYn|tp)f^NL+t3-u?W;YTN{$TwzC#^;;ah_6Aq+Q7 zmyyzT{$MZXD@>D=n6p6xaz@4MOLr>Op@_AZT9L+0l2;~GouT4o-OTQ%UoMnO6vImY zO39}0jeZevqU@Jq9vwH)CvizOl2jr~vO~eI3)cAgOTS$?7Z(RUxO&F?9%arAPsDwy zv9{gnZEq-slJxhr4Xp_O2~isk0}gKUvO(=n%G;nY-!1f>8_G;rf7R)0)9gdWwxOR+ zS?0H)^T#67%}C-bOZpu;Q~yeVJ}I*nt{2#FyDe-S;L$_jXSK+|zxFFWmbf}b(szRL zrCt4+-CtrsL8t6(w<%wBBL9zN{ui|w1U+=DCv`d`aWfnx`}VKv zM`3sH$DuqBhCLnwmxjBJMIq#(_Y92i-SpkN=S3UXXWaDBKW2L-Dd!D-*0|TWWJdLq z6H91ObxW%<7s@Zco)Xit$NZIbTi_yuTxCbnmb4h!kxltAc0zdwdw4()0*Gc_$ZoGn z4S(+LpDC_beI6KQ*@y+2^$s!HhW5B1V{6To^Iy5Ny`W%QE|BwD@o5Vn=oy>3P6$H5 zfxVWuN5CcB5XD_w3I(D1fqk-1{8ymg+t2~guL15FAir#u<&sGNly$TRw)efq! zlTs$tzU_;LK;x;(J944r8 z?zQ37o>%$lY(q*OR36<-XrqsI!SIl?Eho1jUT$-&i$XyvJvJ-Q-E8_%2*U}7W}asE zwx=@6=a|~xXY-oxF5m_pg&=AEjd)!2Pz8+mYnd zj8k6iSA{nX(*dePH<4$H39*Rovypvq%5_*Twb$2?uyEIS0m?!zKy;+sd;b% z)O?F~@`PBc2(ZuvNInA5eb=L7$jP@&ZXUI#@i z3!#M8zs@t~w)QDT*14-(zIo0^zn7Hr?ZC}KZWtrNKlBFoLwPa@{c_yd{@`Esec>^{ zPz49NjVh5bFmehVAmO8@)0hnw3j5jm1=^V4Umy7#Yzwo7Zm9PvEL|+PrXGHzO5EP3BTcLcR zAmoQ15`vWveHrNEA8`{x0PSzYYKCbr6aS(?NK3Kh0(Ny;N{~OR{ndX~Y7)X>{*jEA{McL*X z-p1n8P^OeiWK~R*zdqHjlW+>dPegrtKR)T@Up=;m4xI_ikLFXUctA`o#nQY1`tDYG z5l5Pj$`gv5Z3FwE)R5r@;+a^9;Xvq425k$v-hRbFUu?%dlYvGnJW8|sHlx@O@Nz~N zGoqZq!meWjMR_q2!^#b;0RjXNl-}2+!8;4)H`+7}0{;5f1vU^>Fm}CS`-0IvGfl%q zFf33UZlfS7J*EQp3ko2)DbP{P7uwH<^Kll=EN3vd63`-Rh8w- z3{IzZ-sQ=}JB@u1>?x>sXbggN_B&nq>{`N}XW0oMPCwE_GpS@mSz;)k%x;5ldE9y~ z=~S&9+);D)J~Lb+Td(c}6&z`U3Jz+lgTM&QjOp~k^Y!FBC_#Zgk&Y2 zd4&Pv_%rHr3U&BmS6&vN;3{(*`>CaP{9mm@V(uaVETB0f*7Z&DncxnGy5s1F&#r9v zNk^L2MP;c}H)91h>Jm7b8HIL%zXj8y0A_58-5)t;`FW@S9pBk~Zws9C8(!{r* z5C3_Gmc+2<`hZ(J`ADj#-DRH&Yl4fzHS`iAT*s|{~0IJOlmArDdszL^<1 zG|s_|mA%u&9sT_?F>XPrio*nsQUCcc3=7MM8gkp*t{*sXK%xI!9JYTWJuw{-M9H6x zKPS13FH`wC<9Pzw_|k|qA@OVRFQ#Z{TBoXoWiQ*C%Kg0W%Smu7;I{B-^B^w^G~^F9 zk8UZx_-6$Z4D%NUtsR=f-$CBCHjMn1IDivJa3mtw;mKvJHlLI(qya=0!79v0Da|Gv zQ+~i1DpoO6d%z5h%N1;3$=yz==*WE6_C!ynac2FooS!gVMiez<)A8_6acV!*L`hQ; zx@xju0MjaHLz}jTg0*>KIY9_2QJuOtijF6_d2QlN`po;8X?EeytBRQgE;jamsd?_; zjkpYaVUo7I%Q_t9sJ&YDFhJjF!eYMc=N62K8P=cedSO+hhDV=e;mbY;Rcef)tJ8E1 zN9$Jk^a<^EGSfkyF!^=%yuaxVUNDGJr%rai-lDv-hWlw1sf(0Z8*i*>xWftJaHC_& zEw20c^&*De@!w-(%g#2m@sSk2yzBL5fKbEAS!^jJJLw*7H zSs(EQODkClhR2CfEiENP$FN2wlXMO3w;*ElGjms+Sp{arW!^n%*|}Bpi$V z^t~+l+W!c%CqWUhZN*L4ci)~*W%w~wF04Af!bJj(`xzyzy!q<&mcRFCfWC=0Y24x6 zmz|R=Mlgn4!N(v+odUQQmyQGC*%UG^_u|c&+O}#3%c8bKL1O?xrJ>mBVH;T>nesS+arNJtu!2zFW_` z`i@b-MKN^RMVv|QT#qa@uSuO7Pf3nO?)=N}l3jG@!*`Uo* z_%Fyl^KX&@fbI0PV%Z9eQ7}O|e+0g$@s}YTbCFR(SPUssK9P{!Mw^<>mwlmoEKh2 z@F*I(DhD(RJ&Tga{ z4my#sq*wW~)kL_}Ws9X2cfJf}O$_PaW6)t(kbt4|nu6pTXCBN88qAApBw;ngi`9O|pKW`ND+ckH-U=Qz9z(Z)sb%4yj={*Vg$eHYf;Lo!OwGn%OZ zlj|C>_ryJiEo2HSZp1O}KL)gAGL-Ds_1k{dpxK#I^%x>z5f%fBB&xXmRVIK56}X#l zfKNvTSwO)`i>~7BZ8O9J~K)C4ef{y zisG2e=Fi=cUvTs%6c6Wovfp6{7)Z!(PaG;sLvVIv%MEvu%$__y&aOz)bVP|`2-7T@h6m!gT7@&8JyNPwNuU`^XJBT_0Avmjc%&+jLwX zO5w7u@JhQzE>TxSy03fiRHp3$`w8|&|)n2*IEkHOFva9@mMxLdt8KoTRizN6y=e=qQGTvfIKrhZw3gE zG}N^91G*Mo3?6EkMj~{^$)9MRelLt{ZSPz3w~_OT9vOKo-nxrGri*&dcNJ_-Xk|ng z05rP(=|1>;DlInPQnF=D-%lH-xO4fJLfN?)f-BgULb!HKY-b*PVoh?oV=B6*gj@Vo z8_lvbb#tPTjmF%m6xx z?1V1!E}g0AS#z-RkEE5a=3ch~^5Vymk~T%A&uB5#=;_0%z#;KE4V zQ>o7ksUc5p$I0}EFBMcxs~%F-1r_|sZwp8?>gr;S=xXLUx{^99Gc|qd1Z6!Weq>@` z@Go*Gm6~IQ!$^}FH)%oePIm?%{?>U<@^TO*4gSjTvqrY=VDFMaM1+DhoB2vJSJ}BW zFQX*hEAhmAu2JJ}?n+mKsbqvqW8Zs-p_N5~=SCwT>?_08q}1%G7W4C?_Po4Zt3NS` z#=28~GQb$&pX>o!qmMNomNkpoLMQq(-=}f#5eb#dp1t)btjbobi;S#tuFs155F)Er+pkJH#A5Z=Tk|MRcc;yE;qgLsFn%n(`;tC*Xz7E??y%uk3_`cUzaYo-)bLN zZO5bJ#jlGG$GTE?4ryJESw1|vWQQ2`+}q(r@Elh(^}PsQ<*k?dlELOF@grm5_>POm zDT)ko^=yM?Vcnh*a&;8*!YwhJ+2xojev|hCvamkD-mU2(zOerKzfXyUb@aF<`*D$r zN9ii>VQy1jBKwf$fonsbXJ|&VIW{Y?iKAQ7;!CXrfAGROl`bw7_*a~DUxy}KnGMT+XO$Q7{&JkM{DpU7%EPkRK23N+tc|icH&4ULp^=|l?+=-WdbJq`>XiTH zuv_bhEpW7bWe0Phmz@skJY)St!V>vJiI9@M1VHStl{uOZcFm^u0O$4)Y4`eUz6z zao~b;4B3|bUY70K%X&YqhpSqsN+LWD%E|5|q->bKJu}a*BcUfeBv$)p1B|`-ku*c& z2sTA_%A@s)c|>F0$=8a!iC1GgrQ4^5Ue^oj@{q<6%>&QGoB#P5AA2(k1h@a6n@b!i z%)fU>F(SV*ucZ003D=B&$LZrQ&t&0ImE*OXj6!bwM*;L+(J(?{Qu}x%E?lqVR>^S7yzm8s3nMSB3+)p+wY;4hc==u<-_`B1HA^)bd*mqiSTp`t?YJr43o);8XoW=4ba(XT*ktR=~OSb~GdD zhuib(sqaG%l3Vf2F7=<|0=@J(9g<&quIeZ;d|S4;;;XCt*|t%j;VJEqcS=%3lk#XIOrM-&{NX;JOl7tzjY|oaoLwij5VvQ7)SfWs9BXy z>t`gLzgX&M@w!|dn|p+1FCJFNj?{-eGrdxoVbIl;zT3B%mDGh4k5$g562g_|NwQ{o zwp#OQa9UnF!Pp!XrF=*yWoqKju8m%PEF7N8_J=L>AOrq?Zg3aX?tY=u5%~>R-xl|2 z)o}&a4+tf&)^!g7K>bhYp#TwxI#V#MT&I}@An;f{Pg1s@hnv?uM2RJ}#%O+6<|#GT zzmhVQRV`bghP&nL=c=&G53DC|4A{p<_xS^NkXq%!Q=25DC_-ajLS%qR@% ze)l~9`tH1Q%tP7FLWy_u-{erl*^Dpjm*a@RaW$OfDz8T9jVH8-0M?)DBzxXk6P%?L zYN#?-UwQd5WlWi=6fac|AypXqZ5Cbh-(B!us&+Z8TgfBAggM|U(`kF?3!iX-3cn#| z>tv4CbSF;l;idEw+kEKhUc|}5GDpIgCoB+?Lk~4f#jmW#)ec|SqKXjroIP(7-*$v~ zsG0sxZ0`RX{Df8*K)S!{YMq%u%GkV%Q34OsZ$_vO;WvWG!R6b2QH0{5bQkokG~bGz zdA6OeC2|<$h7J2_TPYS2C_BPUfuN87JB+o(iL=K$SPwxg{jmpTL|=(~YqBB6{L{6i$23xu^qE?dFQsC&{d^6U zENRuVtDu@oE~Qr_c+@(hU$pC(;3~LOL$dNp-Qjse;p&-sGDiw0h)KYJuqu~)26^x3 z9;o8m!R5>=Oj+qW5iYQWfK<31H0fa|?F*Cqt?sO~_^zxq`C9PtT!LP2691pv8nz|q zNzm-&C#EDOlbelimU^={|K{Nc7#)Oi;e+O$d@=s^WSn==80>emOW zPit2Hfo7SbYF5kkQq&weC0n&QE1ahFq2Duu`9o`bH-yJVW!jv!V}@9TXQ9JkvhpG? zSEXY0`9nS9uAbPSM2Bkz&uvf?mAN%Fh7=`nRNq8on~#X4z19}^SRcD5(8Jj7r?>jw zs;K5qGus{w`7vbqir)SHy^vQ&k&ojcC^}3_etIBGOJf&PO7+cF;-C|WQ&V)Th`Fgm zhr4c@yK{iCOstQa7Qh|3ml#J5;Le&whF>Ecc`Wu9ABl7}8k(+6L4j z9Tn9$E3(~xTSXRIZT%jj6bo&Bo^f{Rvk@*T7Ru(DTjn`01%yFZh2)y8gZsU}SSg$? z+YUzzWuKbKv5-Rvks`Ld(Tv{b{Li+7i##aZ+PcCug#bWkGN}%R`2_J2r3j%!V6ySg z%nzp;FDe0VCnB%3miYiU0rK+fSIt_mX7b@MfL(^P36mB3f(GPDlZlz=w#BO>#m-9a z#Y-3#fIBpjlJ*DD@4ps*q1oJruV_8q%}ymsxF*G)dM&VS1mou*SQ;Lab{B`Rky(-%qxFw->!F zd=uzoP1Y%`>O<&8HGgf*^Y+8cxEmXQt2gi|sO?wS(R2pyQ(V}U{A?__Z`}PLY9uUu zYY(CHY0qKl!A6raFIE8#0z?|3iH;@{oE@In%}?`5H~HrQry~um3C4ZTN`yw#oCFMPPk~cJ@{3kOZ$n*~kaM2l~(}%bgl)ydR`;z+#&%mWs8&?L$>KVZ_`$ z!hR-Qaz4x811s{as+H!k&j3{NPT&u=l)r5Eb_#hO`T8dQmvx-h_!CN;19y2N|5ly~ z)PJ*fgGr$J1{B%4DsH{sSOq;s3v~RU`nD`Uy*$d|mK~=jmPAASi?5HUzFnwu9v=Vi&v@?!BGBQ4W;Oa-^bahV%nY#-x$j<>9*!+4 zObDPREc5!8c}k)~F+jl;nmNb&1>5+st7MNuA$(|5;Y+Oc%RENLY4w7HU|)jWY==Y+ z$dF_-{jmr<4V!L-!i*4s3K^3dMi;B$J zlpv+)dUf1AHky!*^D$})i`L)=qM7xfA6=O)V~rL?4<4Bb}zb0fn#JsQ2<`==O- zb3T0i4}a<$c}0|_ckM=}o%=@^6WsI&JPS-s`2g1>zj|*%vTyO9A{M;*;0?%@GDv1- zsgHWTrUZ!PqH9!_W&bj!hyTUs@@%sRA(^=5>s>c{G;Z*LF3IBqya`$TARATp?P-XL zMMOYeA84j92Ds{z(Je<)vk&6*_N=rkO+{K(}u8M2eu4ElN>zy5gbmWSmpDTF$={Qfpj`J&qy zx13dfxTnQUhNkjBPOl8h5?rdQx{D{iWvAch>=cXM@^N%uD+y@;4HC|GIuWwik6L&c zGD83uTCW>iCtZ$77Ov#oxYeSwBu$1^bo#>@-z6jzI7ypOUm+FeOx6*a*c_l86BQQrcM}3 zbmGP0>?=6mXA(pd@|Vds%rIkXoJqpRIV>)w?TtKWVMUUYG6^@IU_vvpo3I2muJ11! z9WO8K!M>OuS(YjCle|xc(p)MylNnEFnNXnPV#PmK3jx#4JH_80ABCo1vxG6Vnew2{ ztbFwvg-LZY=MB-X!2Xn3!WVPY?mZF|kWsBJT^?U)f-57r%w=6FW;L$g#R+$=1cOrf z0WBS$?tAx1A)|TTS?W<*z7APbn_KfCH>m*2{oB#8WgS5$-ol@%D)1$Z`6zEfBtZ8A zM(NukVqEXhvG*qkocj)pCef?zo!ei3i>Pt`wQSn1p`u|MQ<4T!aBv-(WwVVX5jU0euchd+k)dyJ3(# z$XdXOUds0$pVj5>HH}suYdL&0+-Z%g%iybY<`n6>)E2bMB*4h}VE9e7x`QIZQ=Z1JKYMEC)l;x*KmxyNwV;*vQsxSPz@^ABivWHt2$ISSG+PnTN=DmV zJ^yX;SIle>(_|o`Sbo>Pz&S?id5_EF%BW=KwM>V}*Mm1}@~3Z|SJXdI^0o3gfGcCr z(_In%1(X3*w|IHzdaon=_l2oGz={Rv;%nvy0Y~*f`mbJVqU@6%;S;sDJsA6WPrZJZ z6FnY*&w1t^21QfEuX{doblWiGTi5Hz73I4VJ~8{*FLRY|6~RKFc$MFlu=00mgqq^*QF_+IWX`LFH;`^{*A8*gE1dM8KSbY}1LlzA6v z0eS&DCy@%((Q+{<;r@p)m^A6mNJ>Btcu`@4*}B`|m6M~Ywd;wbqd~g5fdqkKVQf_? zKI4!@mfgw^gybWsVpw$n0R6_>LpKJq;?huqpQF*k3XMVg z8|=Xdv)k1Ov*AyY9)rWKnuFH>K=o;8)bpnAgaRy1jx=7b^6U(TkEtp7pe-9C`37{l zSgqLd!F_4o%o{EPE;|0feoucIb_NaHS-v}bHQycA=v1~DhtHA4pm`%iVcbV3vItqB7F zG<|fqUplM1MiX#8(_BV4IVM!DsKzhDWQYS!^ogc~G8uRTi34h0r)QclIPt;spk4;}VTPK3x0(5NYH{f?MMGd}-=R8f2h%*adG>!(yyVrPGT zvaW4L$_=I4RDLqaJJnH!!RjK+02%J9090Xnefv!LV- zt>qkQIg!qi3iIoQ#Gvo0jSOD_eu1PNd7N{OHM5SM_RKR-=FFI)n*=maR|N;tjJg@! z9qZ@CuX-zKra}8RYgEsZJs&i(4l8N&bUw%M6JZTwib7+a(9{+u07k4mgZ*<(tc(ZR z1{tNT-9Cj=u6@^5KqqFrCi&63CcGLI9y>xm^AVVC#=!lN5f9yt?JPK!mS$hnY3cCA zV`nHd>|H!4&V8zO%Rr89hc1=&7}E`*ab6{?3|HsVgZ+NG%L^C$zdWv@JMj*t?y%ey z@Kw!^%v-+>1Kis<8Kh7Do_y&PU)+vp8*J>UraC=9VOVul6gCc2Pqg1G)k?ejhTIX` zI54l9rM6g7`9e-|!m7E6vu2SFUPHBV{k4BO~EC_?9Y>@j)5lVtB_Z(dwsh8f6>=T3R7y?|85JZ}BIfC~ma0TE<67CRo@rV9TnT z0l~8_GJ#tsCy3cU>h_c%Pa83}cO*ol*b`=39Z>Tj^&MtaV;`;#GxcLcK~Jd^>54QI zG3B+wTJkki&b*njQ!7j6s)8EqAW)<(0JB3~=^EPr$PYIgH_QK9f@qx!y zKaT&LMx_Z5rJ(4Z)x{ZnL;iZ#09+{dXHR#AXNj%G*WSS0#3w;+L0iNiaF`Ti-AWcM z0K*UMCcSOU;5ObVCca@bzbjx@I7*{n12pbJUV_*SV}fB1VdO540ak1haJ$Pr+MpHi zO-tFQ(&C}mN`~skVDj%u!0?9}$a0WcN)qQrGHqlzIEDqfGj--D&2N}HM-PXedoa`i zPU=Go|Jf#A94LS*gMjb1LuwC(p30DT^zhnUr7HL(Ub|0;YC!*fStgPHeTS2g(9~}N zw7~2wK!LKP2NPOID^MJV#x!Ima2wAP8n^7kV2A~fp8;aSAsGJY-Ii=a5CIx3 zFuwhdM&BjNaP=q_?7=YWfoA7{;nzPBaVD$clMcIbtIK>ZFLdn8piqw>E9~*Q8h?$C z{ulH9AL=6BF#3#!)PU<#uKfv*Y7Aym+oN)I*jy?zlJMCtj{HcU!hy}0v$20XBI$sF zc$z2-gz6a+@Wyw+59CZVRDJPimQV$77xJeSKsvNI!>Odh#-`UERnF_}6k7ru7y8>A z3$vYK2+s5R@$HkZhN@=w@#hGCi}i#2_}K{1=6~hA8~Df#YWlYf81j{4VfasQ9r&qJ zFCSjJr4G0b3yZ_>Tg#j0`5`ecfOc8YC^gBPXRvwRrFSBV)*4ek3Bo|A}e30W}{W#`cZal~LO=3QgZDLCc31ZbxM z@d2n8w#8#_)Az^k1|us6P}>|L2J`z__lIb)eoW3si|j^~74b7<9of5p+tY!r0sFV^ zkInmU39H(n&28pCcfR2wKLW$zA9H*A6O7m-)O0AoS^pvUi{kWHXLT;I#aIUerNx)p z7)MAYVTyV#Rj;Q-d8XEVNpcH9ob>xW>$Dim56A41EI%O-eosThi-vk+SO%?6^rRIab zet1i;rJ#>}I}`nfyN+M7mYCxHafnLyOCeDjkv|#0s|~NIz$LODlY8@Gn=Yr26aDz$ zZQ--f;O-kfiSCksLek2qC5~jA{W+!bKb2$dx~9DR$cKlA(EX>sZk3LU!LKfn(-vF@ z<8k|KTmZs!ls7T?+5mzccxqHS_pGHLn0XO!lSy31}VqY^b3Lhfbr<>1M9*c=puixq{m8bs^fp-zI}7iamw76pr1@ z^JbjnkEo`=P91tKVi&s%m>Z%Mxe3s6ticImjBjIl&XMS~AOL~L{9Am5)Vl%X6F%Zu zGH2mFS)idP-+V_BFmBFKuSpNf@Hcf#eqSv($wkGSu+Z!Q ze33V=c8YuMC5c=agj}l)VlG#pTuu)b)C!k z*7JNiuQ|2BiZu@g?UjhLYr6>BEh5lx{6yQ+Zy3aCY7gc$ovuo5_gLO%uyj}xP0l~5 zE1a>x{{U+Li>+PZq~rlZfuYxz`OhR^;eC^UNthh(p??lACp-Af&jm#e?xyklHUjRZ z)%*`NhTQR;Z^*zAAIImzHiKf|za1tPJV^Qs2aNn5e5vBSDxJx}_O5VAE@32o@9(|F zY&n&YS&l|Hkd+ih_DtY61*u>c7hd@rn3mk;+1!r-hS#vmey#MBSwE3Q)A}1d!<0ir zKIcE^wmf%Yr`gk^J-cNjnfx>|x%zM9+~(FRJ& zRLGml?8NK13-!w>OcYJPN|Y>D2Uut8%QL}c8y7~XMyP2G9_&8@aWk-3flbnZ*zSEH z`LLXeui4&`&q>O!ewCQjUoDQlX2WJB-h#JrX%pSW@XG2x;oi~rx2V%@m1pvbpKAJ= z-D&#D{00zH7BqBP4+)*T&QJC^nW|tS$SRr2e^yGe<0BF^IlzL9ce(h`)BUa1jsaNE zeqeE8~NC0bgVIeO95#Q%jS_OYCo&%J04*PqT20d`izU~l}Jcg10%O?s9J z(10{Hz4`OuT4`Dzkv&;{c{3b7mRavM4yQj$I4;Bm_yZzZ@cGnv|2|C^ASl5K+gwmf zD0X(;j(?alAdf4XY^LC@Xbn7)X86f9F-znZaIu|bjWhOT!Y2RWmhEnz2O(0jj7{Gz zVlK;%H)wMNRjJ(-1RzDp5)n%RC92J%UM9pIhG+rfI>LJZ-o}IJm#!~pG}2$acC!^6 ze`gqN7cMQZu0o|EkUXWG^4KSR9lipG5K%eHZ(wkh8m!YgM2X z_fS5pfTSHQ4*w4kpm-63rMuYOZ@Nube$;N^c!J~=SpE$E>dR{VU0CXBL5)5K?je1K z%s{*Nx=A^ln5)v|-^?-#NL_tgc#jRE68o$let+fTV`f4DnkW1MD{~-XJjWBsavje> z2gvzzYJ#*L%%q_$7qI%V!#%Ko(T8a5(Le#fw({no{1-(aMyEJSFsFmy;VhK|bP82d z3UA`7C5d*f2dm5g-x}ZY>1L8bdzXQp5Xf|DL$9R6p4iSn ztCiRR3Bq0qQ|Iq1;PCqeBYQ&uo2S#8I!7C%GJq+bkG7cv4 zhN+xPKCJDvwD}Qt3>s}nEj)bvtPQ(6WDVn;Bh2Mp&b@Jak;z`Q>k}}sVA1uyiR>M$ zSVeas+X7AgySle?E=U^^V;)jwqO7}9tk`=XlZegAw>~O{=a*TI8R!&7GkN1{JX48? zf#FwP7PAn#Wfex=yrI35b3uyEz9I3(x5nj@cX1|fYmLN%bho&O zZ-*>Lcpl>Cehf{u4D`Bi86whBws;Voo)%j|uJjLa zYC=-lEjH2ZP(~&>{O@4(KXgM(@j@oipkO8!#xfaAlu9}teo8ED!O4E}MUCEuB6ZTG z^0vywua(TDX|w6}<@1jrox{6>aEXh!3Ab+YU|-!><$(RNY4_M)=!R;Mu zmYeNksSu_&r`(q5Y=QR+3a$Tg+9`06lSVLnvQr`7NTrb*pAb+zH+q7T3FmCr<(-0V z59u_R1M|v@*96=Ghu6OKZmwNTU)`Ntkh3k0`j_}kO7VBGUdsMb&>Qf&W8knIPqQJG zL3k^-wQvNMZCra%O?&-e>3(qs*oEfShpi8z#Zp7PtH+n(5Ks+Yrq-2R!uS9g0G}{5 z>aXa3a&cW=yAJQnh!k?~TcTM4@55D#!!@XY^5`|hp!GRCCij$To`N|~jwjUVe{(4teH z2Yr3q?iE^P7^;0v8RiR<7`WzIWv(7#<{B!Z=u~M`e9;7X z35L6r{QdU^BGEYIs@yr+CQtm)oyXkWo0Y2Ii{HV5QR0d&JL&h}9EABJJ*1ox;=vJT zU8jF9@=SWhKtk+y+#3jTnCvcdYA4-&?|Wz@!mg87$y;Q>rpuYijU73!gaBs5_jf9x zTgg_FvmbMAdC5Ig(rRFrp-?2;ysST#fl?@Dk$t3s;zT=FWX9DWi7@yCg>R_I99Twt zAbZ9Ed1K&P1pSUG*W>-elX_6$tT`=MHYMN@)t!la3-eUJkw+v zEEO|rADN)*rC}9YB#Q`U&l$>Iz{(>(v*iFXzgE|a~j4U4w;kiEOyCvF;xNK%+kU-tBj9-cFGVm`rZ zyZFTSMb_Or-`1W(V7sf=Q-j`u{q>Vk)F9|16?bXG7+!WN^9h-jyDUv$SwiHPRdCa9 zcFQ68Zr#?m3($Y{{&SzGHX3+Cz4PB1jJ>JLoC$qWtz98aF%2W7XBHVn8B%oa93UWLD!Nt9?7uBFmhCA*HtlqOVv=*gA%^FzHk;c4um`Z=-NMvpk=jPMTMZHO3` z-yfy7;AA)E6G`V_vc}9DUGe1=b?T4<*5*?BUn+(5$KPMM3P?e0_2&ylW))I`es02w zrW4;7P%wEwQtOX5jUIls(}dWz*@d5oqfj2+ZSjrAurQ`tDsv$K9-9AjkcU8yy!`If z_s?y^$^eoPw1%C#tlWfg9)Ek$c-$qxfvh3rFp`-Ib=R5=O-G~lnVdEz1UOpF;t?c; zFzai64k#C-o9QE!q)uE{jKzn{2%PPt9(G^wuMcb>7)i@zRB&ObfA?eLl zLYVw*5ScjU(^NiJX)h|y5F&w-nR0!OPb%L zeomv!do*8%M59D;u?QOXD!ZMx;s67kq^@D?YnB_b1aVp*Y4>rn;yS7RII85r>6)aF zR)J&1zQG@+e&n_xoKp{h@%_sXXVs6MNi>vm-spTgS%tMQ{3{XREbHgn;~yCYp1pi$ z5Qjh}M9f|OYc1*5Db6}z5cb>uCNG^iCT8k3;TsilExyqFJ+qt8^x>0t@lBjGc!1?^ z90JZ?0OLGx+0ehpMDeX`olI-r21?UK2ItFztW3LZKfEZ>Uo+cjkVxIGMAA4;lK)CP zJNBvbA8_8lu)rbag!-c47-5cX)pzl`vV0VMmm^W0ko)i+>UMXl`CCsB&l;%UJv94f+NL$ej^fS* zvGS>+%@?4%?S5BWR8Ppiig|tiAr*0rS8)9G3zhQClyT_U%);hvZp;j<GQOT0-vdJ)>rX4 zs)m>K{x@pp&tk4g?6_b6EUDKKw{<#Co7Li8s!=i0f!Df|n6wP4hWptm4$LXoWJM4h zbx8TD>fd+-Gl~b1OWm5BJ&^IPflK{L_`@BK;b7w!T#9cUUvt`%aoTWC*Y@qP`dojp zN*Yek@<_v1c?NnWAnCSIa>A^%Xg4nd$%vpmB{o$%^9m;@fnE@UCznTO6>9pdcE;<) zEqCZktZ~t!=QIcR7CEAk(tR9YwF$B`0)ddE7KUL1aO(L?4eaVdWScWSxO7?8HW0Z_@~T zzR{<+ib?@G?v(B3h}}*xvR#;K6lLO43V+M9T$zuJ0pB97ftP${rBRfV9KvcSP&OWW z)s@v!xLYjZxz#n`-6kL1JRYXVk$1OfQN2s>ho~4vzB(+s*`x^ZvA_UV$E;Kh-?JS> z?b?w+mO1Nx1}hvBO+#gU`hwRN#?EL|@A;ONUssZ}jN`d&(BMhWUS`NkHS-y3K(J@u zGK=&)1xHbxBKnq!KsK|2kth2pr%x}+N?Y-br;+Myuy?Bdgv32Ru$r zdpBqwmv89TFZUX$!g3rI7&_$c{e=3Pef41H%iP1Zt60tN+Q>Ykze&<$-m^4$+jMfT zsn|r))b4)M{WjdPv>sye1J*)zZ#ewy`qP$pkQPy2jO@xbu6LS0ZNwBS&Ld9~ccTnc^Q*6`&<}7|$4P50w~bl#kaEns7g00bpQSxEgVMQt z!wHIZV34^`kmXRZD#R!7PbC{}O*|+V{e2q(skkPg3m`GD^r&G3pjlq};5dZ*bbFrQ zi1cmW>}8(sFY;e$*W*wAIq`QFJTe+Dcyai1CI?7YPZZh`gTn|F$UpCGjY7#B%2)00@-N=x{D5h~#r6_M%L_2W=tVT~{Uuyo$Ih?eVBvu09`@ zr&!f?^kz@-ODM(5Z*0HRlZfr%kb~I4@8?EglX|DpTGV!kzZRNh*Ok(z+BK*u&N8#XPEmA3LU-s71zcwt!p?s-aD0?>gL~hHfQu%=X_kZrzk@Y%n z;AzLJZh~6J2Cj@A3y0rLi7*SNP)MpQ&#d>~eQ41%E?(I!1jJ;gazMR)u{Img#P-cy z9)uZ=tE(I+Tl|_n_%apnjo+iwdPoc-xLaHZQ(nGE7U;EKklRX4F2+j?4;PJm_JJb zij{4E`azayhfD*g#4%moLEV7~e!i-=?GF4ya3I|6ROuI;1rFMvAA0u=TWOCNCW1Bv zNtk<22kwN6E|4B^>o?>gNVVS}#?grfRV$I@Cq#O2Egg|N*TD5K1b+i=bMBteAZb7F z8q&iSv;F|@@Xqh1s{DPODI9o;nRllD22g}CRKVXo=|$Z^FREjo_VA%l5s|cS{%j+h zVK>b}4m>!ipAJ0!6e(8E{f8xd<_5&u+gw;$_ttZ5ZlJy_xM04h?&rsLTI(0U7qkEV zqO*SC^)nWcQD>Hw0{&JSf};=Z+v>ls&s0ie zZX1~oOHyYCYV8scGO92Xf^SylT>IWjDL$^`sD1K<0ew<&vJIP(48``OYA#!PUhmpW zo0~7r+p8_K7Tt$BRrb&DJ92Fn8`A!(s95g$v_{icq|NldxfR<<^vz01yyKC@q!XfE zoGBz>&AvDL>)EAnP9n8srX=&MF}pPKitBCU!a!Jcs#8?*MdNl1VBW<`-E{h|cgybs zji&)l*<$2h9V#o-Om1pLf(s^pW<*HAO;UB6ErB#{HdW=Mqg#GzDb3MEzhYTv{Z?e# zjiozCX0d%lb=~$AVRrwvW4(<;Q4m&0&CG%Mp3B0PftIe&i{W}$Jj?glNIu_{IF^W-J^o73E6x;sG9P$*s6_ISe`d|YuJV!doP{3O zkmo4VS5O#>+f9-}N2wB=tATx?GAC2%p=SZPTN~EHpQ($d0{akp=*D71R~0ZVm)PI8 zbbMp6yMWx#<|BGB{|IqhY|YX{6w@z9meuCLLUVR(%G9LkdCI|1NU*S3!Y?dLZn`_B z13vbF3m(3{5tNCTHQIdc-wUr1#P2nIWiNO(_C6`ghG|ukYnic&)m~RwX>@y1fgWil z8*;rcKoF^Hrj>D2XZi}JC`c*mKn^@7Z+vG52AS=uDm=u@_?JT>WGcQpJo!SWqN34G zHB-y}{8%rAjSlRss=Q)wSk7{I@Xhit=qWuhz?C(^nUN zChkv=aziP{=uL2+?f@6+OUi4eMUYi}vMZ~Z{k?B+Uy-LLvDV-nvK~GHhY8?%-vz|s z-brEObUWS)6jXhsOo^-0R8o6ay~nLv+sLN!NY6z!Qh5UCOt3}A4Q7tk*|q_cIK-#k zKMjE^cH7u|EW(03R685?(g@NpuZH?fMDEYsIWn#s@aIr-x^|OX)-%PK!ab)H!h64_ z^qKQm;>EwLak$@UjRU(f zrxa$bGRIZXgXS)$^xMHDzrV=HW&SP>0R_Ch;$_$LPR4D^G^WXy;F4%6)a3;^45qoK zGA-6a@rZS1_P&>y?!>$=9=3{ZUMp`1?D^(@LA}-AYnrzk!}P`}u4*S{&HfQAuUto9 zels&7Ek)GFfD_85@R52)7{mU^dai{DP+N*`rRDgM!eGvryvksnX%3{InSCZC@<7(O zAKK2%5kBhn83>kK$-Ty~8b`i=gwp;AHGStamc591TICK%e{kr!z)7HrKap?IqI9LG z3k|Z==!Y=$VL~1xofl;saIBIhtTzvE+Lp0P=s zTSP%Lbmrha%TE*9zQ~k*vE97J7{0*tRBu=KNacS=+MoZ#eD0F-tMN7dIZ;Wo13kcy#q|r;toroR4_21{$)nlo&l9Pz z`EA>Nxph1_K^1^r@(h#T!2=rlELuROol)lbFTF^L-KIG`%GV#LcFmX=%y-=UZ{xZ@ zWl)&ddUr4y8A}Lc9NMVPpf{fjTPIB?*51VcE$&kRe|okQTWg`&&S^7V)W&-@R_%0V z$^;9$b54WGCJ`*R;S>%zsdQ)2`95+^(^YkU@SMgb%gDNSq`=6!K(%(Qt{k>b{-A%d z{gVx~S_crs(>}&=K8vAfND(#hhYPy^(Q)sH$A*Mb=2A8zzGw0|551aU^j zPYYl-lBMV6h!jx~P8SCAzxU$sBG>hP2zh&5^}mwJQUK!KBA4gKv`mSh#x!UfGwCjG ztJ}3(!Y2~a_O>GX9qyvGlRjZ-e-|;`w9SGa64*EEexVGoG}Q<|?6gL!NM)&bVhU*(TIjjrgV zMkMVc&as_$sJAH9ybd{c{`%u=ir*|zNi+mLPOE2q@0b*86PW0HO|?~IVl+X-fwX__ zpE;Rv7{%H&%|-m4K@J7~Q1pj{t+Km1F=c}=sw1?nSeKIpR>XJI?_bwUV=NE(75%$d z>ZpbdAIJH=;rpZWG1I0rWU8>#tb#yBqtF+%eG%oD*aYV3Iu!k5)9D?nk=|Rx%5TonP++617x${}xwAX=>vg6^OLs*Z zjf3kg9vX-^b;D}cRg_N@db6Y@IzwrlMgH=Q!pqv^?JKL`?-W)c9RpN0CodI<{j_%Gz{<7a{Q)tz=I5lOoF6cX$BTa*)0dNoh= z?RZwBPq*VDgCm?phx2W$>-+=Pi^yL%%sQag?*EMDn*9PgBQ7cI);yNRYV48mxU=N- z%E=7G#i9T7I~P6{F~z;Z)937-W7M4+9D-Y_)9E$r8QV3A>g+DL08M_;--~p6-hJoo zb6#q}nn>E52a}Tik@R)FMIk29)_jWPsaAdWnBT{ah+<*N(Z9KYcdjX~I{3N*TL_}w zziy=2#q?YM@qs&-1=gX#&`jY6d0X-@a=l2GlP1<79U=PY15I%Wb`Ej4O~?UlNMzJ? zeOWXP<@I08Z^Op~Pd356?axZP`2|(=qLPwcnbJKLDYY$yYp%D!LJ2sS5R7&sP2cNaK6YJm7J$h z(8!yp|NfAc>Kl9o1;+00l-x#RH;sLR#~nvU$Gp2n6x@QYm!Ut3nN)ma`v(6#6^gW$ z=^M-up;vkIAF~943GeY%lt}|u^fVa)+N6>)uFcD$uNlU7(5Oi zd-Y;(^?rj-sMeFAIeQV!`9WrDB1<5CTvd@dP)`6R$ap7rT`){V39xkK^uAvyAcqz$ zP1Jo8#x(m$F_rDj|69%x)J=#1azv|}dehfWl@p3+D$2e)|1)U+lR!Hz2g$EKMWt3` zHqO)DZi(00WzQfR;z;!BA?e(vYm6s<74pe1Dl-m$Hr8rsu#34HP=C7fwZc&t^G%(I zO`6Imvx>62eH_r7+cnbV(V<`ppAbYUifT-ye0~rehjtRlSAXQym$Y+HH6t$>-1Ha2 zF&^-JKi23oV}NMFe+5lAWZm&|-|Ru#ma5_xKOI_r8XgrBr}mXT{l$u#vmLO~*PiPZ@yz8w^!oBd8%AsH2G+6sm*n$3*M3$ zcw27s*KH#JmZz1|Gfq@^q&s^k-=RE_xiP9d)9JIM?%IicN~J`_x9UxWO}u+Yz)}UiK$rNP2S%Jp~tI(&Yv+Gaf%P$H5NlTlKvm)kR$chO&=_BEmR0>PY}+U zhy2#Iu0i#qx9E0w$A6DE#Q;}ymm3Wfi1$Sl?B=g?o1b4;XjCvZ`RqAjrgEjL^pa*P zXeu~1nc~W5!Ccon2dByW8dmh188xzOaU(!^^)VSYB^IMyFk53#Tta*iJl z-!?ggEJ$rpB|}I8u^g~S-L}I}(a)Hl8Rm{ak#{(O4D(|b*@GU#CysT;JFlx&$55g< zT#M?|E&eU^&Kq)EI@_V*_BU?kPR+7@kVmata0u3q){o-LdVDzWPE6POHp-cE3i=WK z=pp46AAsz@B7dr`!iszz!1=RADh4uC&WSqX6mWBLl|Geq=^~C&Zpr57+t3#nN*^A< zj3Q&zTiGl6g9mKUH*VU?dGB_z_o+Vy=3iR>OvEXLe;+8T+*zLWBYx1! z+8K7z|Hf*@Tl?(i7~kYznf!4lPulXk2&baH!P!vx!r_Zd2*m|zapG)@o7ZBmbB)tV zQg|Q#Bh_w~{F&$ei?nBFjM~1y1k9|Ae!18K#Dmwv;B^j{i4c$yfH4^M@U9brfV6kEh`9m5H4`kS1SA@*zK9UjF^#gECf>1 zEj^^|K=^oyAuL{54(%LDUw+x0&G77OD=b{{vYNnYr;h2uf{Cxy9-2<0E zHL6g8a`}*TvTC(0$Fr>i#THZHxJG|^;WGJwWm4tv>KbjOtO`KKoBDpVppYPG;>Gp5 z6_i7>UM2$C{%0vEuCuxqHMcCpHc`mK-`i6rAGb^PgoNN%B7@I1gtr}BdU5SxhMJUp znf_p4YgwqQQvHFaQr*tTNNB78nm?ot@{{21KAV3NWU*28kx=hj$3R~~x(7rjkLCu# zQz>@IM5<~Wy+cDEn!6J`;|PFA5DixR2r7u!(9k}ssHvYHpSb)%9lkKu>Vfe0-LvqR zJUl?)p?SkSJ)V8~{b{UR9@$a;K3B*0WJlPRgKLf+_A^nUQh%d)jW687l@5g}Uhw!R zxIOAwzqL4%rt2*HnTjIpOyDo$>CT8NyM;bEP(kM&$iIN3PKK2mVX2Q$H;TcLTK=QrYcss0? zO6;UJdB2V3MWJRijJZBi$s@YcCkm+?{+!(pdDZspj!IJIr69|Dh?171{`X;j*3JU0 z>p%##+7!M5$KZV3VIOaJV>Pu7D+j!mO|jl3+(PxdW6SUNu>AUhc!mj?+eQb3(^>vM zCnav!88QFuOw|!LQ7Fo;P5E`WTpo@_l{*n})aH?Lp-?1FS#5FCiMrITz$lWY*^BbX za&PdzHU-i{8kT&i^O;Y(AK3yzYg-eA`C;HP`Q22p@J5vrCFV}WdLEmJ^kqUpC<<5dk@;|Cn3V4wlZowGMy9-r-@crOJXe%pWoE74nAXNDP(BXDw9u*6$c% zmOCJtGSIt3UK-X%Syrr_DQy{Az6UzS1Lg9K?%qN~arfmZyfm&mL(g3A#xJ^_34Blv zXNVE_NM!}iknz$O5B1jVem3&9aBFftGphbm?jhx$FJ> zg=uL1r;$jLCJ1HhY$b$;W3d0o@J}YKdR>(T9zGrmQ;@-adW4Hn%=rBoH||_)$pB(v>n6ns0Tx+ z#$xTGNWW)BlWt2xacbA^mp|?*HCq~vS7EQTPdHzM5UQ=1ipzhdk~g^xtmEE^dv+E| z+DR08Cg7TBa@+U`cj2^i<^BDf=Ju0CGxzRC8vgOSvw=5iFG72N+aFPf9x8DT^V1f> z8DE$bNJmc;mcO?-D{0?>A`^UbTf9Pj&j7(M%##ktq8>C|3t0>^?4E62@1r`0h)C(= zj$H^54vcsHCz*|zf6~%NU^f!$>Xh*wfA2>Cv1>~IvF%4#eid&o#Yzp3&88@G{|UXa z4L4%|025m??vGS?cMobp+!LCjC~t2yJgFn)+AX2!TtI?w{f!(gh(!(gBO8IekW|5!W@^c5%Go$lKe5XVd#ff zn|_3=O_0b|bV7Vv9>aTbG%iN+Qy)?6*Xo90b$5R(~RyRrnkE?2+hAa+UnMY%6(f zI_Hh%If~G0_^^zUEQE_v@dLaoiiTw!Fhqr%8hY!NS!pSyaSe8EUv8m$&twDExal<^ z#aSc(DMsOC|I$xVZC`4_SHBWti~}}b1k;bt*YU6}gT7KvK{<}zip)T18OZ<+q;8dm z_hoeU*IzLOB4)9rFL#Sy-+6~I=O|v{S6eZh>MH&mYzykkz?@qU01pB@u1<@*w^a$^ z7T&tGF(0U;&5D4OPaW`-kIEf~CifR$zr@i%D&VebMrrlO%$HNufD$LwvU<*e3?TCR zrvCU9=^PsecZTlb5%G9oTo?guO5kl{g2d3dktk)D8su%#+WBb~Vx@n);9)w#u1^EA z3}V~))i)bzrMYg6i`1KOfZgn}2fYD=6W!*gf|;cuSc!_zKwnTrs4PU~P8=!$TiWe~ zLg9YpNes-26-K8x0uYf}P!9kxKM16iOq@UKX;PxO@k&=2@V3+L)$I zp}38+*cn3AR^D}TgWV9yJV&9r`+*WfvwkfZhX$#QqK+XBut+j&KPv}!ycP0@xCBi!W)bNWM~O;rC!TNM?Om-VdsvaeVhp=B651#{xqELm zG>50|oi)yRdh0q#3#r4l3^LoN`oS?y8COmPc`M;-Ze4`XCsW*^TD=zBD3FZDhB5P1 z77d(@Sr4p9&0a1keD?SvzbbV>Y+!#a2;P2xLT!?o7kw`qT;Qt$LKS%%0_F#!=VSWP zK-{f9jJ?JlWZU24$l z?fXpWTr0}E0jg?s$y_vk6uv#Gy{LsiEI~^X=!Fvll}>GZOvt|R53mu{IRJl%8+;at z+$gmjN#}BdDF@qeg%cz@MdgP~g@TKzpcZ$UI0ozkUn)|rgtmts-etc)41O%Y#o`Ew z<0_-CP$_+X#*JQ(nc_J8b}APpYNdx0m!rP~w-blr2alq>ecqfF%Q~K=jzjJBO;2bj ziK^i}YqsVzPyd#ht3dU1AXry-vj=nvxRj@4+fY+niM%QrYMp+qli_!lLon-{;3TK& zCib;W`I|?#VL-fp@@y!nr3ewn+kqz*9;s?}Mw~9G!cr}RbqaZ!B=Z++z{#i$Jf_*- z?E7pXti3N1c8d@TOG@E)ALHr_6dOVmH=y=TUH+DQ4B87}XQ8pxp0Z!$?T_+~z;w$X zQpX#Bu5m^I<>la3KR0qgT%o+_9*uV((Ldxfc-AEn{uI;TQ(|Sc{nI8y0X!X61kl>0 zSK5e8qRQkFd>MP}_GA)YbKJN!{rz{N1LP}=@++CH8MAP(WmPyza(I!WPUHEUp>On{ zWOz{+fmC=Txo#4sX}x_)EyIyM{R~TWDcb3A#j}5*Ai|zmb(DA*n&rf(Sjnot2XKnTy|Hf*_vD8zM#TF@PMKY1oYCG0x)=S| zzAn>pmu%!TzM;}l3qj#;hF@&>*W&`q2CZ)4#`+|rzZf2U%9h3hVG{{K#`*+=j383r zxAgssXQ2P{r$-SVBmx41ydA{d!n`E|gB*f9oFv^mTqR8d99&&Job-ZyoPs=jeWazt zpG!#qE6hx(KqMLVB(7hLlwIpXSzUueDL{CH=Xf9xaC{Kx(}IiX&yjypN0PdEADZIp zgR1VuPKpWRQ%r)X^z(T6m&7K;sr36MmzJ7e28GP5t6Usnp)cO%jdc{s%FYxi&iuIg zk*5Fjk+$)W5XoJVjn{aWl`$X?CCC^QjR(R9WrCuwTUiM4d%z%ae0wkm4?hzOIzKd~ zWd+6I39&}U;1RN_#^U|#K27;Fr}CjgUzVhO=QApB@Bn3j+-R#lCbl{6NlmI)>$Cj^d5%Zg8|QDRR@ zNvRgiqbo&7ln62=B_+qx;4u~=Bu)UA(DE_H#;Su@2?^t(9cYCp@#EBJrAXc+aH(Xr9Sv{K~wT59$} zr1(T?WndvjLeMH0L`}SE45B7jB?V;=t_p#uqgTPSoRqN|YJRMw8KB-QumU+DNsk{b zD>ZRnsl5UvrB3vkt`sLxB7w02DLIM8nz0Zkal+jaKR(8c*l&2OoP=?44t_$E)N#FZ zQVR6MI=!nPd}{TcUQz`{yd<7gQbNv@=yF;=5E%*2nmr$9YCJ#$C8W>L{PtK^Av!kB z*iVX_TC3MVNCBTnr)*V-krOly22l}@8-u6_#z{d4fW;1?iXI2kvQw&SsJXL}B0!~C z;Ctlp#69k`tW?D1rS|tI)3u`~bfwsdv zT+%*VjuxQ&;QV?Chq8`?pp7;@TaPY>3&LMCWU~)QNU29`wB0$^R8yU3TnnH{vHIhA zH9hdz(?g2+K?|*<`im~7*CRg$1)_oR5r`3d2Bn_B7V+i7mu zDK&5)n$N>G-6G4edNf-7mY zgBy+Vo2e+4gCDfT&$A!5pEBukchEL09CaSm%q(%+Ua;qfu+-ea!%9aI;Tfbs2jw+w z8rEmk%n9LV18lWKhhOoM5y#vm7yG`{dC9Q{m5;1<&k(Gw9K2&3kC&{+OhSt<`h+|1 z^Ql{<4o5$FH`zN_uO}C|xNrLoUCb)%*OhnR#0=Zd1!p3MT~b#k`ton>?_*Y{c$A)Cts=I{F3-#i?ftKB0${=GXc{ZW0r)VUquI>$tx z_#(tHh{R~A=b3lY#nd{T$O-ZX`OQI15dRa6Zh`0m5VG{ag0J6ug=in zhi^Do_!^^rFt#d-bR|OR(GNk}KrP6a`)_O9z)mlnsG9nXmGM z2*X;QpFYtl?qr{IkH|^Hj`7!#x)%U2v=J%PZ>(I|*Q@g6ZLk>cs~2`V?|zaF5}Vme zKDX`ZliE`-u&twhMNKUS-jaz*oLXnV?(Y|K{{@rZ%VwD)^I=Q!A+#=3#qXL7fW-R{ zmO^vC8M3C6OjQtZtoim(1l2`dI?52!=Dx|Dl;NYPN({T3;RFn2#@|kU&yr+%o?kEH zPkf$yis8fkytl*lRV34*$vBxNi*olRcI1w;9<#Vs*Qc42w8Ce7JSn$L-zCn!@B43) zUeBSGT{TEj07gaU#KN}AFK@YcPZ^I1;Q3BC7n#PY12#~E5F6B|2 zBigg`Q%i3Vx=z)_W-iym{a4e%y4ZyllI)~;wEWp((g&a9tY^k+C|;&_Tj^-A&Ca8} z)Kb=IGi<%@Z@eG4=DgHF>iaHG-V7YDLIJtl5eVJ}zg`^@UuGs|59FsF_fT zUg_mk?F#V|>v3WI^`x5mj}Tj)e)&JM4U{F*#K&LEtQTtGh*11PDt)7B1E(I!tVu>& zXIv4z4LD`;5TwKjo7OPx-A()8ZJ9^486`L>zMmcXL)F=rgW|I?d{099?9T!J*kWp>}N6cql_;KxK~D1cp@9C z!Z-a*XK(q-?$wXJ^0!j!8!F@ytrz>=baI%*T57qhVMUP-amrNr)QJq`$wZ@Nnl_MJ ztC-4yqe(S9Gwdmii-s>Hlls7Xjv=2@H^w4J#)4fdGbo$4KhJfUWi)CJ@T1I zsNv0fnMsGt%I<6&NG3# zh3Pkx^3Bp7b~fgXL)-~&L$j!(bV8XP*Pm-0mK@6V{34?Zx`KnABhL9AsD(fJyq_vc z?4~~G?a@WuG3fl2aLh+-FwMT{jURD2*hC#nrry3gU0Q^5Qy#w05LixDb|dgEUICB3 zPfHTIXnCKyWXk=rTQvQ$eS2@MKewR=-|VKxuW9sWK5IWQ%~grf?Y;(<$C3aydEGwh zTp(h2I~d=- z9B{fi>oM3t21Sib_KF_ynm76`wdimA=^s5a`g@7xW$!mhar}e10eurh^19OL&r)QN z`&|V+UC)V^>Xhusoela&Uq(iyDu>YKb(9CYUHr)Wwj^xQ9bOh4o(d~)qQ78;@AI3Vtlxa;ztFrg z`y&4(0D>1lml$x>>(3O&+p#M>v-|H{#kB}iR5FuWEI^Kz)vXovGwV~h4yz&ZOe-R9ey!oic;+qrX zgDCK-rgEvc;fta2(|#TJX2-Yw5hcb+*r7ydcA{$?UQiY|XksTsE;IbKnpVjdV{$<+ zFR*>NMUYlk@Y+o* zmZ1Mz6lq-4)Q+|t(v`ek8^iLKaaHth--xAx2rL&EoJD#WjTdwJGg^BGc4QKnSrE31 zj5)xImd%Rs!HlA#h9rPPek8_#`Zc-ce)V=njRpIt;ltegVyh-J)M6uu^6hyne>Qc; zSvy8chQyn^ir!|4mF|zRst1`-v&nx~VZ4q;2AB_+tGpIUSd)&^{{AbE-(7Pz>St2i z%(FNQh>?2*wkKvhyR2!2n{j1+q9|^{huio~7K?{OyTrmI>LeF0Y8F@Z49 zGfNGNXW6$2LHLR3V5btTl)}J7rrTsFCSdb|^2bj~yTG5{7=Q8Yru5*a;%QR;DM%6h zl_ZlC|Drd(G3Tv(%`X;;1mte&Q2lEpmGhQ{%B@RaKe`s*(=w^kKc!OWC99Cx*aR0J7xG<4+2N(9x$Cf_so@Pn5$vo^&u|9%&pqyFu{nuaax z$sPI|6CTYuZ|Z{oYt=G=D>(dRG9Bf7;J%Q?A@$4Mq@1^uIj^iVzw^bfCZs;ai{}`~ z(Z~MHubs^%`8!}Y>5Kbs!vQJ{E4rB4*m%wG#5z?-a&|ONjy%m*8sTj7t2|rn26(C1DPEJ@HyP#6~4l|>Ego5hywQ*)52molcE_m!*}z!ID4r^cW|Uc zNd|UdA~O3a+-S2N-r-lAv1cXSU?)3Zx0hVFBTQ9jnSAN)e?5>~Q}4fWC$Xemw5UxR zTvvvbB8zq}>6k6^U@T>w^|pkoR%DekIv3O5J?R=qPnWOQPu6(K{ylWkpovVm6Tjq1 zaF!=Q0o{8syp&vw1+UqfOnriEPQr2l-m;qa*AL4m{e|xm&k@$n z6*9>c=8YA(H&tEc_9*Er{Qgjtd3Wo99G(4=ckBT{1XT<3wH)kql8_pmpH*Bt<;rq3 z_WNQo?1;lqZRv#?8tXcZRkPJcNnI=@E@74N@5yoBSM3K?TNhPl{b=xiE+5h8X*gJ9 zSd^DxU1PCdlR=d~FYFohzEO&*5z1KWuAOz$?dDb#^u4j6s?Fh z%$gn}8@99RUi8;xq%7s5iYNXdWG`#u9Hh z3XOc2qPi;jaIIrA65ZP0-iiu)_qVYXlOuUzUo3b*Z{|K*N2ATgJ(tuWj0mIswJUY5 zbUO)cI|WBO`4IJUR2wy_oz|oM8Epp*M+aSaJBX-*eyBZE93c|?{C%RgV`3{_bjRUr zyOnt-`vF8yq(kJOO$gOlOV!2F)G6-K^+DIp`!`jKAt9d#>GfFav*?bddrpP7)+!r` z5klrA!5MwGDFgUDUulDh!bM+xF)PQFNl=gx6_JQ^HDygo|iVBnOw9C!#vw|sG2|j<(Xc}yV zcAx5$cj*rGcntMH%ley!hElC==7+|@tuROdTzf2B`{9}J;aT0`g~j2y#o_7j?mzu( z*v!sPEJhv)JTqaTgjS3kDH0qKjrIsV)A1QuG#TCS7`;2y3)#I0l5j59NHfs@xC=rMaQ;021=CtADv=O9| z0yrlC?g@Zn4}2D#R^^;l(VJHDoK`>Vcx^#HbU&@do_|hQXXM$K%RX~~z!n2Wpi48h zOT*0Yw=0v(PM*vnfq7wrP@o+9TP5@ERU(Dw?AISEc%-wGbjZ!1*;)3&t*hCPu%IA1 za|o7Y1KaZ(ib8x;IWNYP59i$E?KpW_LRuO#D$bU-VlwS8|6L*_0E}oJ&VTkCu6USV z8JecvPP8~|4z`*xPIGw6Q6Cz#&_uk5*EekExoFE7lgH*2p4Q@`K;LGj`3*9gOOwQO zxVXWLbYia@an~G1mmqXejjR3Eiw_m{@hZs*P2a61xrm#-?&c`&=cKezjHel2(JfyM z=QI9TUTGF|n#Oq|=Eq}vKr%P=lc zdgxLStOlDzyE$gBNF|OfmTt{l7#R=w8Fv_)l;&tJJ_oXzTMMD*w)OadyT)d_{yNH1V>(t-X%QiniB^%1AodpAALj#rlU0IsMHe;obP74LsLHk}-fpWdo+% z%*wTabGgbgvi`o5`BC%Uxo57+P$W{*W0|B z;Y#v8R)p?~n0Znox6d0phXyXVrT_FAUR-Uf!{tH!<7^jVIPi^`=L zB74GXLky*dDe|a|*LG*=`f0S>hYaYAROK_HD8sQ9yDwhDB!*aI_K#b|ut)Uw67){K zZ*J?^9tp}h8rmLRDPNPOTo>wX8PER4=T>vn&lo)13Mg~F*gbYtI0gq9-H+4i|N$pdq@D zi$-6T*)L+x9ndQ{Y`cXgF=rvaB^Y=jghTCm-PMV1Yf2ma7PtM<%Rx<>l~zN1La)Z! zlZB!PItA?Cmuo5mRP3KLf>qD922(kigBP_BrV}Z)kv*x}@5IN_A=&Z8)K@$Hsc84F zu{+uV-F;;W$6iR`?Y%)d<7=Fwd%m6M{+NuvqW7ydn|f@6jG}J2yIo(JY(;AM+!ZcZK4^g1=O9!YHy}b&v3QLu$iG4o)6z8*67%^ zVhiXxV7u4M8C5^RN_77DIbq3lPDw!DZPrFW4+KyP7br3CWZgR$02njOCv^bIy&+4ffdXyB<1CJ;N-AtP=HXJ zw6L?vGw)Z#=0nc>Tw6aKi?yU;qS=#pqW$uy$A{easm-KxQsykD1z8!aQBC%bQO+LNOMl-`iBrD!&hL@R_#OtTS)^^y1d1ySgasOxfkg|AwSymenP# z@kQIEql;|?M~OOnPyaU&$KKAbjgEtJhl#0VMyRn7PIYBg*+{7j+ZQC>9ICGh8hDbJC3r)V+lqNEhS7b z?rS|RJ=>U!!mMDiD^TeNb*#ZZ<%x;R(^g@g>v4;vcw+BlDniVq0((OdJ#(9|@6?tf z`zy%zzaJvS2`VN-@`u3*q{+Io}J2CXG_Wlv*Tp{_qqUE2| zNs^D}JXG;9px%Ba-l!|3O#DOX(XMosZXsq%rc)V`CsdWGSyZQHL$^v5 zVb0GNWB$she5HxBZ%EA|$DMB*m|JJKg8dV2_3i#r#rMRGY-wGAEGbuK<|1po9hre= zj(J@cv;@3bU0AbpVt!gQHHa}hb3chMQ)pSR!0sb$gsa*soeGcFYLokXw2Ru<8p zBA6u~%>~-{l(W3}yZZPGW^6nTHoH{!z@T2FLui|##g~378MAjW86hXV9|Z%4sW4qNR9SL6t@8XZL*|T^SNQW8p-HwwMkwLu;y;w8%=Y^8 zpI<)HN27X0b1D?Q-QyB4OBevgl|vwJ`Oxy%H)U#-hEnF` z=?0TIjc5pmK+aGAS3ecr%fn{3hU{;-^S?+nLD1={IG#iDarci4Ga4mkafec)?uOXR zg7VFqi5ln%Nd2#wj^U<#lk5;Jo2>H6)2=PTXYc2z98b!bM>s;L2S<8TugX47k%!oP z8=;t`DnFww4$_=q_GmfQJNmK4@o8}7j#$te*M>y`AFa57IWZFtH$D5x5ob`t$U{oA zxx*4isH*WQPPRaV`+Hfx2s|<&SDkK2z6^Kr#Brj4pSa4wNn1qN+sfTg+4iTuyl#va z^{;)GbHz$#)i{bfB@3ntdHvZZY@HL181x0v*P1KLvOCY&=nUneO*bL z*~i;1ezg3d27I(g)zO=7X|0Ojmqn@FuWBp!Qdf)9UTcvAIhD;WLi?RXY3Se2JBpJO z zFXSO8yXpD@xPOAjPN%eiu zF2YrKl7b;Mhneg2pYxtNdUT1k8fWB`^(|HSz=coAc8;|6jS9P+d5Bu6acqBT599XB zVZWQtv9<0>Z)#6XzLd-@Lp5G?25NQE{ndIQ=J8v(Mx%7*1LTU4S#Kx zN%r{Dbb`JR9}dl!xd24|vhFH&4KI>4?&SiX6I=Cc^XK^pC*oZD7%p!2aYz)=QvZu@ z9mY8)V2{tAhm}d2aHm~t%IP#m^!t7z!pTgj^pD;!z=yr9p4s3f^F#?)=VIbxx^ zP6_T)WgpJv8Q#B1F5N8@sJp6d_N;n>Qivo(lOAaL+TTmqifiCQb$&m=JHz%!aMwoldjezff2 zwyM?Osux|cR#x@`c!~cL4^s@U?XLo+7=VdECWd|_hDF;Z#?ppW9`wFEK<9TGeiJr+ zPMc1-7*Vkp!G;*|g%}}LJ0V#+F-JS;kk~Utak2xkr@G>l(Dujb?NlP-RJ7vMChgCg z0u1V~*B2;Ww241YZGT?e{(`sTl~f0%hyli$BnAhuI5r&9>gDcsfheyUSMv2$;r z>GQ4R=yT!5s`sHQouX1*lImSjCSB6bT{8Y%vM4{DL@9dIBb~%91(8mLAt|05DMeJ5 z0;-J&)UC?fttQp2uHFqX>DF-W*7TQ##EJ!eEl?v9S8eL1d51`(vf;2XDLEkStZKm?x3AH%Q#)K-TL*%7d)WcTv_4 z)aPf?7vKyIFp;Itkp^|jQWS&HzKA~RroM=&zR01zsIETP1{g*w7kvSa!RmM25a(MI z|76miASG!b)tjU!n~abH9~1@PWo+Hi5q#$5Q>apW+M5g-NZ}pGk{S@)5GT-;i@T6g zLSW;^4&dj=6B9|8A9R~`4H!)glzQ}(4at{p$XAf{RI2xoruNMc5LSu|mOBqtcqmjs zWy-<_Yy1ZrV+Tuy6dI=5vM&a6u!dTlrEJ0*zt6+6B!@OhhuTbrx}6n;!uxz;hkkPO z4~WPOg8B!ihK4p2V=okk$ogFs2TNW*da*b(Dxx%|K0IYIJff>K3ss`>7@C6)&7%fW zQ%O11qHF!Q>Wp(UbB56?h~ZUGZ!U-QIlZ{+}l$}6|&z|wq@d$Sj$zJA+UKWq?o5;IOjiQP> zW~G#nSYv216*OpUkZgG0Ijos=4A2|Hh#1BR7{e9qzekJ`fh)ry~~={`|4kJAm0Q%sN3cYoHY zlerL)e8oBbx_a~~Rj55Vxr>(u6rlPPTX8@UjBl#K?lQ(cJ;AX#(U&uRQ4oi>r26(y zm4k7Tr@D;=qK0ObmR-Cc{e7TB8`4O7s`86;oUdDrAJR*CIsSoof_(GyDVe$$-Kdt7 zN>iWmhX*xjVs!~Rbs6jl86|Pi;Ys-=Nk;7a#2;ig&J&qFu*8Zfm0WdEsmZr|lWKg^ z>YS6`V*4Z^gxJ#4khp2pwCOLGpFVd_f3}faT7|;So`!6IhC3pFdnrmPV#?ZdM%q&YJgi|(JZg-s`LuhA^8FAK-z>BAjD5AbwyCC$ zizY;L^3h>)HQuCK3FKS#tZVlyf#R$eL{q?7&4B#NyZte#Y&Ab}gf|3v89rmSK3E$L+xSTQ9MZ$=O93Fa-f5zTGCvGoh;~q3Un`aYT<`Vf- zLQHDpe~>3FX(kTOrV*<|Uuw0bwp^!tywRb!e5P%8JD0;a5MeqW3)t6mNcMR-Z2$mf>l7#>6P>#5r3jt@8^5C^ucO=+a6Xo~w<}8ZVwJD0*DK zsZ}?_kDK$I@!WT9(XGHO%1#B{MFU~#7yS=#ukw_3OGaDK>T zv0i;~L{G8<6s!n&X4fSorKJ-uy=2_IP&d3Vod#*B{?cZ@G-tCke>gYQt@oKYV*C+X z2Ja&JCui03)5TTO{`?5tp{2!QzCSgoi;MTaHl`OhrS*3q`c0reBhxZU-7j*o`MQRe zj)&FeO_xrW7CZv<28N%WavGdL491%DPZ;If9%^l4CYKqjIz^YRA&Y}0f7TBT{%#_- zmYVMX!yP`uy$Jo%X@eU&)rw)d@i>||$>iNHhTS#>3wp~$8pG!r1LgfbBqN4F7LV+?=~4N4z4PVY^W%&@CPnRU1`1K(i8t)uVZXV zP`aVnA}5KnAvt6CU3wXdm{9BQy55$~m#a;FM&3WP8l*NN`}HD|Z%hq(Ha_cbDMZdI z&xkfzZt={RsN$@ckZ9Q~w!$MQpkAhWIOf`=isKt*Q2cEznQfvgv*>AZR-ntRrb?FW zi?8$-WHe^%`5%{bV{-~qFS9Kxerbyut?tup-@Gj^etl03O+}jxjShon(s5 zSpFH|Ktsq(Jt?MfYM?=svUz5USOonB%geo-Jk#GFW_*@Z&Lv0%oi>Q8iSw9@{JeJk zYu7jGpLZ z=I{@fo$lh%s?Dc4RxP2IdKvLY>!tg*Og64@)}QkB@AB+6M;31M)fRPE!I8n!JxBB7 zw&QP{t_s$bK#EgyW;#yqCRo*a@e>$(sa+BWK6Esa^Do*>y;bF+Ho3 zsOi}=OwJU5gwkt z$BE_8j~z(NkN% zl!L}6=SrN5E>Wk>rh}_=6Ntg6BKo49?lhwGsUhxVZp-E2A1C$Jqgzi70|sNw1XpR? zb5nnpl0h4*j7zHodyCp(*;(VkK8s;BwB?(9kObC}1qYqGx^+9hg1snY(Kt&m%XlS=VHon@0_C+5+FbM;C+a6zyf5WMVGpo_!mCi{o? zN%vvQlXbY-5~15-h*G&q)?vb~kuXqi{$2F)^{>&p@}T*WDDrA7V9@R|CgDp*M!$o! z7Gjt~!lW_WBA^fV!I21+7Ni*(WZ!N1AYkB`yfM8e!%KSl49&W%GM(Td|Mzah?e*>7tC)};y=JFyz6v?mudgi8iHVCL8uN~)lz1oleJ~8uh>>4)MtQqU9 z{pcjJ{1`2I=`n!&!;;&J>*>u^K6-U_@6v9?0NL|0@OZBjnxm zNoHp-DYJYr_%y3KoJzpzpzk!h7xqFnl<0f!OopGaI=&$MuOc>sl`TP+)*e>I*DJq7D1;ucuIz)jV|3=f_UMP}m!zP9HgpI%^^B zt-)j-r@)l6nZZ$3$uOpexeJT&d`P}o1P!k>vedY>RO$ZeU=eP!&oz7p(kShycEx^^ zbaPS&Xk44MJik8QD#=#6=G1d_*`BHTw0xxZ{$@O-*NyAb-=^-SO0T2Wj3MQSct^jM%)laer6z*cAu#x8vKEkDNZY1E-%g^IcdwqTHn$n z+M&a8!~O6ufx*LVFUkBX(ACG_=`GO%OAQ#Yw8&&+V_VAty}K*O##g3VOo(&G-?FEb zvHF^wDzUelmlo*4QWzZXo)GpCr)Do)v}S($tDta?u}P7tE4;9NZEr5g27X&q)-JQl zRNhH(KNlb8YMELYAIZuQ-}}Lu12HM1V^h|62xqTdP*%@}dyLd>*Ok-HvZed1QJT~q zy6+!xXL+&N6`arOaDR&VYPFgh$yc-}r&xl^)_y_Zaom2>fPLIZ{5qM<$m=lCAku`l zMjG%M5FTT}cNuz1O&3P{uD|x3M`fp^e9B=pda7aZ-O$Xy&{ih!Z41LJ65# z>=7z9!b>}q)X<^x;mX1BQC^{fVp_AnoM+0z?(9#$2&*1VYA?aH5QcOIOn~=q-Fmd- zhx(1UXKaF7ehIM`17JcWj+*o^5@EMCRxjbbAly?wh7UKVevius>3#9Jenc5{ zGI)2fPmbr{F4v!=2I)6_OEzz5awnc+tz@UEAx=@ zKzn+1?533cYC)eBy}v#kENiCIzAO3rfZpFXn7Mg~dHgo~NbN0#G7)eZwsc8q(TvUZ z#&zYK0V_eK37NB87~b}e4i4Y^L|U&r_WRb1+fp9U zh^JMFt(oZqHtg*u@TyCt^!^_e*U)rp;ItzKUR;Xiq_U$W{+BmjOaIr>7Xo|QaRjUS$qrZI@ z5v8T~dQ@PQxo|Ia*u0 zpy|eLhNIIcP5P?Ts|`iYbGboJc{Ur*jXAdJ*+v7@*Ypce?uRD>qZdn6nVWKPU7+>s zccs>e4Po2dJN8tYRT3`TW|0O$ZWKqaNw){rwB#zOop$8V}&#>nK;kbG>NZE;?cHHV|j9Rg|m8Z_HUMv2p z5m8fBv}K&njO&lj-LK4&?iT~tX7jxE2E9OA^rwLWZ@P%T8exYl9x>IvuA64JUTDx6 z>yP&{{i5e$GpO`u;JO(VaXg4z7!xMva=~el@4;N@nIJf=3nsV!@%Aa{e<~lav4BSa z2N3`OvH!jD0T+)JT>iV=5APY9ZeMw3XAtRY`PASs;_fgi9;?N^itOH~7m}g0tP@22 zF)V6u-Tunl!Gx!JZK-nqRrz3sD5%OG%@Fe=aQXJNU_2Wf!=^WYD4fh!`P{(ztQtOD z^aYaKH&9(PTV~i9N~fUlV!F~|5CQnFBMuVx#eP+Pjx`0adeX|T3zImiZ0K#!UZ zm+J_A#;!k9_qL}h^0h(>^|#8+L5|^}?xFgs?a|+oVf3EW?{+5(aA}l>8>;tb%Z&$R zpzbv`JwbN+e})^cA$=`=`!=te>Q2}DV%QBvn(EKD$FdbNl$#qa_mNfB%OlN=*T*ZJ zVQ7X=Els!QJCj8Qqb<$%*C*?P8J}8P9`0|>_m@XoThV9$hh@bFi^Af!OA0J{#n(g( z5h6w))3@SJ@Jm8IAXPna1psJwR)bvlEMUP;qhG;72puppLm$ELvW#n+cK-NbjtjQ= zagg|Dhl^b8tVc<7<88pCw9OQv!AwGlFnRvnjTjXfYK9n94NKE6vigOa@w%?Nn+b-2 ztXqjD@s?Xj7I}qR$gP9JJ~Q9w%wdq4XfSUL^JqqUaITfZvO8;w!MO!c&oj_f;{*hytrm>uc*8S&9+~R zn6cU~soR3@mo{DP?U%LTupg9nQCJ^T^f473R1Wd)A5@LWupc6(G^`J+ksrF`fVr>! zSqL35@N4ST8o^Sj_eY8Az#^AIPrykFJ9pEP@|3OkwR@~xvl$yWIYJ0iF*<-b3p~&u z;x))Edn|E%*eS)fnA0^FA~;}y``F}72Z@7-Pp_EXx1-+S{6{q^AFewHxltbxI5 z+PBWce(|RvAW5Q1O}6B3r9&MU#5rRf?cvnpcBXx{VT1~|Rc znKz=Guo+z*XNL%*7%bO0?78488GTCUt9i@_rO;{nS({oFn^wE3h>4Th*wg}SzN(~Q(?^_9vf7er{yZx=-7NGWerx(0_VY^5F ztHMd^Ia~K#P}Ks{_ZXQNX?bcPca*SFo@V~nAeje!-Oir(wfLu!oAi`}!qZ%Q!xQ&; zufm21H&G7lYMP=7%Z+hhr6Rjp)aTOuOPk15x2YDN<675yXEt+&sVmQiM|f`@0hnrM z;}r}lzz*eKRuI)oIRljI>$Z#)Zf_j_eoQn^dj+!d#gp(Ca$5;GxHz2hK%oF6He!_?!toW!%5XRw9p{x%AQ#F2ry(Q=he^bGbq7`Lp=CpV%pB;{nP!V}Z%R`lNJiM|*swp%MkUkGq~o)Ypyi>W6*5UegGYNQOm|m}F1gwKI6N zO~h4R$Gx@^{JPs!UpsP*s5QZLgzZh8f^+E+8$Z|T@fy-y!^ z&i{O29HC(p;vV#o>XIRCY2UOc7ioa%ppOhV0JaD?$${i$@Jro_JiAd-rsvZYID|9~ZiWk|+{SD%sV1Ml>0h z99Q!fLIL0%sHJNN7E-mm`82$z_DGTt5ID38W>#2xl2X?d(=6?M|BdjaG?iATH3qf3 z7$$Y??;ex7iU{i=W4iOi9(=qV1VCE$YIMI(CXy03g5~bYumb=~_mQAc=DRO=`jD2& zeYoN#5$AC&(uy=XT5$^$Lk)2D#D@TF?+K-M&=T49=S+*V=7hsqCQgrl#4<)Rp8NWt zGi}EhTvBZd@`S!G$`^Q)C7Z(R;L!kOmY~0iyFB@~O%vKqDW@FozAx2Fq{frxH0qSR z)x4WBAw_t6WHT2~6PhZ+xr~?QxBBpjW~@NNG0dQ;7#Blq27eDxzN~m)I7_3G*Hc{? zH)I0UBO0f|ylT%mG!PZWc2n?YSx4R(8g5r<~(^4wYAB0He4%q zuC9s7e*3s#EJ*nxtG8)Ax<*;--||W9f1V|P0~3$}t^mV-5rO^xKm?h+b=Lod2)W9f z{|ymuj0gS)BK8*h|2rZ!wH*ErL}bgS$^VOpDyyacg@~HsxvGBDNU8%9=aG}oW zKZvM3UT*UvWKjP9iHMb-cuWN=zC`>xD}JOhEUW%x8WyVoRAvRMfwZpwAmT+J%UUof z-eN6;C9hyDl%r;6E$mGX%X)YqIRqBLyH&6r`7a{Ga9ICA1cl|sKZq#Y_@($?h+y4} zRoAfGjQe8t|3HL%559MevuuwHMh zY~dz`Itj}DD0R1!15>rK%#Hke8s-`$LsR%T-S@6AKj#;oAIwMzKM{T!Wkz3h$5 zx?cHCX!}k4heWM3CQ$wslhC+a43jWQlk0Xf#ynDp@Ht4u+!cUH1n%TS*}CuMIyt(t z!XKT4aq{Oq&;kIgB4oF=D82E}WJfvlsCnV<51_`wMsKG*+splIut;ARsP)iaqClXP z2aZx;uY{Rfum6Z%!~G0!n7f{TqB|VnE<@ez#bAg=QoNUaf1G7*(0bLeT-JKBL0Lw3 zc0g!&FNN##zD*gZ{6WGBul;z$ck<)rk-;&r@^K9R#P{%r0V+rB{r&i{Uu_Y?k$t`> zLGrN&m15>D0ZTLSyP%(_K?5SuMA7Bl5r4E=CIO!yDJ5Hw4qX=mFlm6b6Ljco zO9!^)RZE%l%IAwNpC-*=OW2b%=P6E~F-%d@U0^U5@<#gPGINs-2s3vk|K(L`_`a75 zEh^#pV>RjLag(ikO9O%WPn%*Tr2u&>d2hl$s5&jCVc8u*pry)dX18gGONA1Du2kY` z5(^)S%w)?KXOk*!i@Kc{{%GocNS|5Ar%z^3D*USAwB}UEvBV(t<9RW(#Hn7q(3=@UZuj-@Ej@XtWTxJH$4Dw_}b_;ss+fs0lRMn0F-;M8$fKHvdS zzq@%MN3&WV$;|-i=5)T%l7lgrDA8`Q)joeB+N$LO(h2g}S;}mwH4H;QCE>?%b1o9Gd?f$lgW%$?J396=FYm3tlLh`l0+LO~PS-+2W&tYJd9DzCBk- z?6tF>je39f=XYo5dk$K;W@gim--b_}i6Mca0`h+Nq&5GZP~9SJ9GK^ehqe!Ce6BCj(su2i94 zv&K+=4ECxwkog^d`CWDd5Ao+QwalbS8&xDc_qJRVyScz#qCV$O@^H*|y5gisoiYjr zi;)&NFLEDx15Z5{hNF)M0?K}{=w+*vDPFqxM_GQPw;zC`@$P-ttzEl)?hvAhf;g!e zi~XVm&C^XhBv3`U#KZ^VKBnH+9}2J`Px#x(I`>G~p-A?ON2%l?&~SZA{nu;ca;6L< z;*8FLQrgVkkkz8fp@|HA*oTb;Q0&2<^wY+l&AQB~Xjt5|j+z_SpI{5q5sKm@DRs)d zuI2FinuevV0B{el62Y)x>AEx(6%tG=9rX5`%FRI;t_q#4s~M$19p-IkwLKl>>~s&i z>oo2`?$(lVPys!*Tr{EA{bmuO(Xzr(l`5v<{Rq7p*CT6^GT$H$t*b8l^-9O+r4h!t zBv7rZLoFHQM&m&5G%EefwC0J)6o)zUt;;L&&{dLX0E$5oO`?AGf&!B)w|gmNHsi7m zw73@XYH#qRzl*5Le87e&%Wzdc11)d<3mqC?bfBoxZ7iN9RHSBH>vy#gAqYtR;5ZXa z&$^@ri{7&BUF>zUGbS#rOap(<G+XD``3Z%nyR z#JZkaBnzCNb=g1%SU! z?%H;=(W^XA>mPxT?*#VjbClYpN|Ibx54 z7%3A!OE{MhJ{d5oXFi9Dt@@L9>s!3of|tk?mnx0k7}ZCgf0C{4U|jD@GWY?OYdY?# z6{{J=of8er9^lXb2#O=Vp1 zpKH!mIjB&lTUQ^ONo-6MVP@MbXKUrfYsyJJ(yM%qEOKmV;Pbao?|qImgsLYumPa31 zL$@P+UtycIX;^?6&r20KOJu!hEV@)@%VIy5Nox)4ZGT&>ioF&#PkgZb;fq>^h|Sl z#I-)tyUNd5Z}sleeS?mfg^&1|CbT254n4=#^H1leHp<67GmSbIhAHimHUCaJ4LSg- zqDkq!YyV$RADR8m2H|?z2!Bnzot?tohjlj^273z_c}M_dsvoQ*E08c^0|pD+4rgk~ znN+L@Pf6yS1B{Egb%JGq9i~NO3ulR%7eF3QqWoy_#VTaZ$((yS1Ue0GlgmPh`WfK# zWCB3E790Unq#MJWJj@G z8{P?7X-PFYJ!C#V=cSxD&NTbi8FW+@3f7yC|9ALoFkzUudfu=Sv(s&j26m%s6%50edRy zIrWWW#%J@x1dgY1fAY?p&+v2;ymCT;c$X)ehrdWe(Hz3O=TbAcX_&>1=;1h9;;`nt ztW1xn2z9_O(jT(U2GQJ>Wri!EY(bzdrRBw^u?VjZ+53(3A2AnP5=?%^fOK-VYDJwp zU*7)uVRUWJF3j~4zh8!Z?%KtleayolEd&Dp13mw%q|^Zh0eS!Xk^=ipO89Xy1rm-< zv(+!Vogy8JC4#jYreZ1+_Fq>cJmayGyMO#-#j?eSI4p2vOTh zHmUZ8T|P&=1c1DMuSQmq*808VOc(+;_IA#L$&^3lCHfDJfm~T*Vdx^d+=L6IN`N6y&CO0{c$;q{7^7Hd9_{b_M;?bOm_*0Z_O3{Hov5Kq}^Gm{`$LS z)8p(6!C2@Yt5KulX%nrB#4t8Xgz#X0Pe}AL)FqHDQ#P71@pp&Md(mU7p)Vi-h74E; zuG^r!=8;wNzAHZZ;)z4#niohg(?sI!GCwNm{Q4c=#FZ+gP0j{%kh&AWBrby=bpNKg z6Ln!Qwj0gZ@3!R)I>gtFEqWy&_mdz(g!qBj+el&*IdQUKm0e_WgSa)VcH))H$LB)a zP7a5dYWe*#le)ZRO2)#Znk zTg}k)-J-(Xu}oNLWrtZdb^biYZ=M0x2f5AW@*StY`c?QZvH%C1F)lk#uAZj)@~0Cz6?L zypQpetRI0`zo=Qh;+%Z)IGD*c`hCro1K7etT$fHV+DRKBIyyT3ga63fag-)Ry5MI% ztH{ZcoDd#hQK?aA)x;>&PJ5YvcTN({qI460ZP}XLvTm9A!4(??M1H>+5wEl~3VQ`* znX@X6O7H_1YVQQ&29xEq;Yae>^Y6xDfFXoMbsX}3DX=o`es&<)iNhi%QX@j!4mrP* z@{?)(0GRriZAi_i4|o{`LsF=?nHZ4F4Ityms@xOmAlUrW(+XN$#W2bbtQ#70I!IhZu(Ck0)WjE16ipB@=vj4wh=24^#Not;-P>Ud&}MyW$v(zg8TmCq}%`ArETwwKvcqE~L?uTmfp2KS)W$ z!73stcPn~0KqYt;Uavw5#3CPh$^PWWZMy_X&`yTq!Hh6M-II8|?*&F&MOms7i4V0l zo!Zr_YNL|kEy|7xu3g2rq2}Y1ANMjd)<*kI6_O+9U~>{B#DyuEf>XctRB)q~|A?OZ zP)peVRZ`S=4Tb+%QWS&28qyoTV>8GiDIF?WGcH9%1Ip9?QBpq3ZQPoFtzq~S=2sVb z%xoq=F(4JntdvBeLno0!m4m1|p94uU=RdH5#<_?NIj`=QUX(m^EcajJ!Dkmja4F>~ zB+c^mBI)UA7{TEM3kBE)$KoybuwqWP2%?YuWRmg=@fl4;^gJhI+`8b0=%4W9Z$L#~ za-SZ7pe7hG<(ZZ6XbK9wEz_}Ma~ zmIt&v)u`27DWgp;*LoXfd*@iLG{39kNIsLzf5sSzSgMH|t~9ZmUu}55t@`ojnC$&3 zav;FDB16O0>d9I=dW5He_d~h0*1EpRvs_4Uyg%F#>w8xn%Q$d;JZA{n!UDIw=Wa#(8&me2Q` zJDwE3-N*XBdYKEZgpGn2n)tdEO-1xPR7}melRg;ai;t=;>r>sZ362Yrdj{9ay|6rGRNZrR($@2L(;d#W|b7q3FMYnVA zGUHYaE$FDa04x%h^8)Wy)JFvVZyaIW#Z1Qlf$&Huz9M2qO-ip2G35Hj?2 zuJQeam>34Sq~fU+lVQGr4UQ}v#0&+1Uml6H?fUpQLS)aPs#;ljhj2mOT*9HK9 ze}iE@`&fzwqdox4Q6p6_DjFchl}m*Tqyc)Z(-{T zrI|R2K`Rl@hd(Q$zHDTo_M8NA$rQuZ*!~BA5b(Ai44y2xAGvHPxj2Mm`#-=iIJIW~zzF_FrKykkN}XpxkU_hK=4 zFNbN1OxwakWQfA|6aEo+0Knq`A2Jinw53^;_7``vXdal8gCJbDtDQ9R1UYyxAZkqI zJA1>id8Wfe7-edKip6HCo9NgOe&M-xQLZL$s=aR76WF9E`3xgH_Sc ze7RAMyRM~qdX@)$flOkEyj4_6ObDO^Tp(?X{gJc1uu`rEy0-5PTh5?DsDYeTT%&n? zdKOh2Bb{GSrg=PGGM4gWC8HfMfA~G#1d}ehU*aG)y95_yKPRw-t}J(pC?Y$T&{5(H z!Y3#xT?i5uJau3tdSpsjqP#DJh2M@WFUh{QIc%_rL56QCNYJfqhM|DiRv_?SO)#PJ zx7Ec~jaYM35hOa+NJU`zo%wc+5Wui?-%>hu(GR*gXb$#VPl~m2*VU%IFP~dw*5*8M&ygcf#0^HXV};&Hf9 zLZF};(>p3J+0bTZR_t%c2A}#M5u{&5oqs!HeRR!MkkHYlh%ZW12J^OQW`c0@@DrNn zq)6|^fYXrSg*)|tU8u1gsA^R0A6U+cH!&n3Mx~9k=7wdo4qZ0fa`yZ!- zO2}-p4UxKUGLH_8<@qzo&c3NiJ^pwea}0*RB0 zBg|Zek;v@^c^!$P!`B9mL?%EIR^kql<7_R}3GsLeq>!5H!m~YgrA*se2~};P@l|qP zn%X`-Kh67$c80xII5kkpBgYb2eGTHi8+B~25Ro5Hijk1@i{UYq{7T-3^P8IPlRZ{E zCK3QBna;w}Pz!9mie}V$B90Ybs{Xtn&lZ zU8i+l$-+0}0nG(}q)$hca1}EsVVWAJHw7WmNmKi=Y4>M-c`U*0+J-MD{MWrIP>^?6sx2Xd3CaEWoU0CQBV|L#!&~TqAW%8x|sD3Jy0H2RYCVi z7j!wRB$K8lk0q%@WDuqxBM0CuI+Io@h*3?PSS-xAr6_zPkdeRvaLNy97Q3Hf8cZOm zn3R?af7laWFdN74Kn)WhviyuzpvHlGOJxSJ(nc{^l8O&V(sh=Vqi9&^dUmErw6)S4 zUL&nX2Yp0RNg8qDR?WiOuLON(?ug)C2q9le_@r^F&vT|u{(je(Cql&V#j3JHm%0eI zztTb~EN&p=zPahFQgfU!t{>W=(TEDP()`9MS+a(Df<-{iks4O^M`u5?k6YROx83BPo7!8CNFctR(ZtKJNBI_F7G19+_n~$E( zE`B+&yM{=^Zk(zT1z_&CJ+KKKFhl1_v%Gy>DlfmaFuV6Sf@ z(DK$tgevY+uIJNXeljM6xr8M~d>WOCaZG~x?Ng)FRf>M9k1L)s!y@}N!9#K#X}DoV zIY}_Jq-G5WK;7r;e1c9|;^n1mo|`j{rWXMpIWk+PECiu|8P|*TFcn2h{Fn4GF7(^^ za~uH9-}dpi<;GlKwPho2`keavARD^fkvv-lfF|cAg3H@d-t*Z=wd9)3=R3wHy62yF zr;PDRQM-n(JETy0=VW68HddRArK~NUS{A8nW#X5i)~(y3U=dq|;^&o#`p4VBdJrgrFCim5#0FE*cLJV`uB&07x#IM z+kn_RPKcKNFc-81c|2d@DzVgUp9v~!W!hzz@k4KYW5#{jx#ovnrSReBd2JY7NC;M3 z9vn5h;6k#`Z3)-g33VU)Tw?+)zu*;mO8T>!NlCT8V>qccdoM*47!zV@%!V>Gvpw zJp?f|Qr3%$z|-YjI;tQd$d&I}Nj$IEbGz@mp~JWCJr=`Zv?TM(SJ;$QZdSf-q+g$48UzKlw!?S1mNQf4MqHYIrJcP zsO1~A;)@*b+r8oge+-_d^HqaoN|W(J3G+Qe@Mrn1rFQ3a5xx(4Z6GW*}(rp=abKpt=|<_#rxmN0O?qT zeM&IEy8l52_?(#qpOEm>_}|F@f#QZ(+mg5C|0x47DWrb+zm@^mxAa2(|B?YzC_l}- zQmr2ThYaBSd$QK%H#)!2!(UKN;6G&m;aj~hXqp&{&)1i`%e_gu3-9-%viqB}-?iPM zzEgkAaJGGIe_@gSO9pUwWcgoZ09!bRxKNUiW1~RWc>3M&YdD&maQgo$18^@}?nU$e ze~|(HUm1nh$7H}*MpL80B#He{HW;%91#wp#_1q7zN)rr4lXhzi;+e=vr$wz{j1>u| zn9v}r!g_NL1drAYWQOu~%0niPj`xg82t3|8BXF3~)3&~MwSl=|T2$|VZ@#QDg?@c}cYdzxQ0-BI=-!TG?&*POvX;0A#D7n95U0&#) zO_!Rb7D~9%pswsMTG>--cp+89_L;{ZK^>*&SE#W){xI%Do-xSFtze{b~IEbG5gl zx;|9@c6U7re`GKEN>&kP4BLQC#$AUa^4HnMk!V?wXari6tYkLIp5<7^FNbJ*+r4*( zOQwh{Xk%tq%ACM z1jR%IEfx-0*DZoibH5UtklFx@OI;Xl6`Qsn<{D2;cyo{$p=>+==+%$>rEb8Ct5b)~ z>Ao>-$<`79>^I&Z9*9IGMXW0HE2x^+xLTw7NRYnS6gmJZ#PPP2`p~hyajn^kw@EJ> zs<07{_0UEJ2g8c}aV-4!wtX(T#_6wez@f7V)!@_xq-k=};BbN52&tB}Um=b{_-7ZE zGrb30$JQ;DIG9AF#_I@W#Ry<*S@en`eIKC(bGt%O2=V*$?GhFS+pPqF^->9YYMX4M zA0(b`&JeI*n5OJT&OUyrE-V@H#icE&3ZyD>__^}a9Hr0XCCgL%$sBK>#Ai+f~t!I_e`21W<+X%D73X@PI+xXHOU;kGbr9o=c{)53yG?Bg%BG`Wwy0 zzUHCNdec_uL=(yn7l;XNr|!it&Nd9S-RBisfnr2u%a0P3PU(2KWssVDRK9GqC?g0$ z^GF&u)j|DQPEa(!NlPR7ok>N4ov5vMlSLYrBI=3AwQXdYTQ<2e?NEMbdUl%OlWWJK zSYA8JACX$A4|WIP$bp65-mYT~kP$f)MustdaVe}Tnkv8Cp|c!w6|gZUqlC5%{@viZ zA;RGdQ{{E5h&GO;ir7#(tp~OtYo~*WHx&WR>DN!+%@vA_U zMhrd0Ht~u_AYskJOl!A)e3WjiJG+N-izGiLz?t|J!3B{wn~Lf3&$9p(r*T0k{&kl9 zH&3tvcnB+q9VG3LIgv!_Di`mM2r0W;*oD%vmZmc$s4S|N`nCm6F4>)S$U1d@Y#SAF zddIlS(W=5OcCA{pjxC^*SZ-=@Hon&bn=}$og7~gKyhEIwU+(bwN9r&_-<sv7wEo1mM*?!!r#q!VHL9j9Q0d^`d zs?1$U4N)TJ7Gzkok*4#}<}V_&->Q#eaHohW>x_+xL?EV(F+g&3YTM6x@3Cx8GQUkj~E>3*EI~bwpoqZbR3H~?>u26aH z#O9zAZ&E@$`}%WNJ>~0YFlaaw1ii%g80t8h>{W*8*6V;0i0^%yil=e!MtEocgYTOz zo{irn#;}%G2A@+FpHC^K^Sm>H7{DLX(!oU;ChO7zF5a;8URw$12Yv9|GFJ&;cVbv^ z9Ub*6^|$9juwLN9zgl+taik%xX3KnbGhQ&MvF_N`59MSiKgFI zj=)dQ#mnd6tDj7;y=IUj<~L)0cVc=&VS)fFsenl>!otuH=@93dR>dYQiv~V0_ZUvw zN(kYd9)7dRciE6@xS$gM&|@gCzoqflAo6g&7H`F}usaQ#Lm48q){eYwTMD%w=OhK1h@hZXOH`y+%~^oO({st}dA8`1{+ICkGNvr9J)Xh1Yzc?|F( z2t=g=qhW<97zgs2fB(I#EOZ3h0E=dDp6RP8+HrGc2`G6WPPs%jt3-CFP9i5^A{zrp8~~C7BZ>nM#mSTW!AVl#NitSR@`Oo9Mi{+D z7!{o0Q`KZ_<7BmTe9rF-K1K-oC&@i?$%Xl*nx*QQFlf_sGywH*;q90+HGZ=#Xg)s z^B2oT0^?db!wQr$r${2Zq+cjJNzzIOe?$UYgT;6$mkPhb)G3jBKhE@6&XSnU`2;JM zI3(=~3$PZzU;D-I;Lfy6nbYIQc$?0UDUltoN__mpVH%fXrk15@g^#IA+9-vrp_Z?m z#HT}5pxp=uW1ZuSnFCLeY7<`2Q%;=3kY!?;<;ba*M#ZD|qmb02&~r~7mJvU;pa220 zXq+P-Z%vMSjeymfgk4;ZDngt(LR{aWz~?IGn+}-*BZ)L)VMJn)!$Q8m1fTI*p>c(n z`ZNBF38|h(kwHd=<@GKo@zfQl}Ua|MsiARO?r(6OOJUp}iufR2~3L^z@_xT8>@ zg58UV&IdYCLZ+5wB~SM$g2z?2Jgc$v`vh$?V^*dze4IR(SeMqwqwK47nc131nl)YK zL`klEjvZ&I3vzKogj7;ji*$MdWw$gC&otS4EoCw-|WCvKn=YM|C@p!IB^&un1qY+zn*V0~#|CvM~vYUI{y z@;~wG~)sR z_C$1g+Z5`!HAv1TuDi|1cZxbfbTEb0fW>A&UW>+^;>|>rNu;uAWD^HwD?DsVg>|b| zaI@2kBCxQ<_N>{hpw&~a#aocwpSZ?Nj}D#+?z>Ry4=h{@Vs-@W_U+RaV?B20C5j13 zyW>Jrh9?F>S(|2LHHfRDu(LXcm@Y4~#TSL$&9%MQrV6O`32=ka8rfMf+3GIT4khpC z=6#Pj`pJath0<)Z&|#C=6(H1U4|VLG zvuTTB?1^M5KP2vco1h)@>`puDntf4RZBMbIY(@}oek8_zwkAKfre&Q+F4sT?PG!Mn zvBPAs1G+eX+sJlEbU$@@A#;pxp0u1Qob8Pt6P<)q8cW@-`z&A?)+TBHy76fiiqTYk zq{GG;ex@be;3D^;YYXh>M3oy_>yOdpJ)7)+nMc-=luMA5Bdy?MI;Uls;%Xxt$RHwp z7Z@C@5F0ZYlse}@V(bPf!&65MYp&eBH=OL7`8kgus-jDF&@@smb33tA|4!emfIcv0y`4P zJQ7DD=ejr=`8q01GRE9BD1thgWy_fq)l?m%d-ip7v|L7ri>DlyGpwReAcHU0D=Q_7 ztH_b=QGN317W}ab@4uHu(EqXdB!PwhC%VA;Dz&}TsrQ%J+-N|QHQYEs0 z`sr7J+3&GLi$m1q&1z5R#etL0i-~2^M5B#w<@3dj?>+q%G#h0>qx8@5;YDau6k3fKK)mL9j|NDY^Cl9-`jxjg?bDd8Id%mr4PDdsKX z#O}3Hga=9X%T^f{*1IR^k?aZ5A=!Sc@cK3;c=G0XVb5@eL2{pDaxJDPPSoV!p3;wt z^*0+VV?T$;(Us)2@NAR`kIPVhGMh&P04r;VPAhf#p<0U{YUiwNs@t|IY--w6kqlrK&+F}#<)w3B`w`EZSbkB?q0n~a{Xtsm>H9;t>c}K(t635SRVwH;{i27Z`HZa<0A*tDV||?X+s~{b326cH311H0X2ks! zf2OqkE`hA40BPIj^%`g;^CZ3|q=5Ch7^W}XznlQ0=>MgM*9iMI!SI_SZu+31>ZXsD zK=Ec4jpMv_PN?(j8=*O%`c~rd`&2#BM$b+N(?Nhx%_3b^*X=R_p~$zc-dBm5In&Tc zZt)by?1qvR_^N7@)f|UOHt7Dj(||%5OJ< z@~!J~lBHDALSi31v9o$P{2SWe4p+<5oK^HxrP^9t*OZTM>XHy* zlqy{xq@K-s$Ky9jm1bY8*|Y_kV(R?+3MXi{lyP;^sGvowAK8(~k1_}s;g1$P5H&~| z#@`s3qnH7|tsl*teihBYDC9#*e<;K$5FHwP9Kt&lX9#2$$DdLT!E(y~B43q?UHmvi ztNA_Rg98P;1>OKvij~U8r+9pSDde>8^Bz|InnZyCEfws_$Q7t;7L@TL!`*iJB(ig}mh0Mm~ zpcd0(a*wGLOvaaN=u)FJ$OklWCL1zPVWGsqd` z$fk@)u2f(_0OlhOX52T92+x#H!ZRa3b23^u+_GN{~yg( zv@!^S{9FLhZ62ahnFRjCT!`RpJ~+Bel1_d;Lie_SV7g42Z(=^i_ zwuoM-TwZHpA*JKCm_52&(L#PPW9_ztf4cmW>%?Nt^KB{QU((3W0;0PzS)~fq)QO)C zF^y%P4-D1tu{!*jD=IXnPa)hT6#mTiSt@8Byy_yB(&(8h-a&=Bgbn<)=o)Go$x9Iv zjFoLQh&7QFMIkj-EPCh7wV>lO9;9dC+USumU zC#^2XqG}5{k$VuorS_X=rta(W!zl2*)-Tld>ntNdv&>k1q%rdztwP%vNA@@83)-sA-NY{Z+&dn>3>0Ov9oBYe7kA7}P(I1R#zor%27MpX zm;6KCo5$IKyZY2X)O`vRfnh$pTG%h*t$OUn{Y2&sS#${hynLImyy=^kE0J}5Nqtv-RUT)tL)P-lf zK17^2bX+23ke0&Q_#ExM3~+NDeY4QD*1@k*K+&4R^VfPI=@0+OsFUZAWkh($19ldg zg{7MA8tI?ATr#7e-_O6MAK$BD16lTc@QIdGT$=lGS!@IbI}3O2n);+rtXW<4exf7% za7M9R4`w4=zOcXU7hwGBW%#uBQ?#!qc>1sJqToCpG?se=Mf50Qt1-7C+S%mgcnTx*(2n5@@n0u7p#)rkei+2KK`)F6$ zBRKl1&I}&}fqkO)^@-*N$A$*?(*b2AOT2D_C699~j72c;~b03E_zE17++7JFhH|)7}m5b3W9!9Lc-I9KL zPy2RqbmoVEli%DF-F;5{H-!RZeBTz!Lhe`pvUT(R(1Rg*8Hy3qOvm^z7WxO{{W4I9 z8~QZm>T{b44Ltpv`^2Z+Lk5m}-ETBQgrOLKf`wI%OA___O5vr+{V-AQJA2^>4!y5DQ>4Q0N z^`BLRFBC_%V<*y!Gb#V*q$!RFQxDqQcI9$$#(_s=UJYP<3}7b+>K10lzOz7QCTTNi2VARnNQ#5NLG>j$QVbx|GMI2_;_KTy*xcpQMZPag1# z0!LK_@M!F`viF5oE97(77vMIm_#>>fwzu_T2yDSZUx}uw*rV7*sOQ;wXTKXW!B3luJb|8j!jfM2khc2(c$h3vR z1LfcD6yZZ~0wMqW`IeR-{J1zET-%pZJDR!8a|IEVTifOMdmv+cAhXO@VeN3tF)uO8 z7?DSW@iM~6u_zb{P_a?SCqxrVieP%pV2&`<^3?#HwqSw=>HQI3w%spEi1rq3W~IxK z$WM`|NTDKbC^Qhjy#!GRHJ9COXuT!QAMF4-f|wtN0jsM46z#Drt6^y{F>{t*5@lnU z(_4S?LA}*i^f#Y)le?|jD(h~*7Bc87!zT;!=nn!XA7>or+_LW82 zA1C!ZCY@M@as`H41AV3teP)m2&uOAXp8~p5QNsX;LWBwC3<+!6f%;`6TWMJIBCO!4tpSK-L@pb*5=&*s5?Mi}n^#CLG!;AmLP zV^|eKxXoC)@ePKHRrskaaalO{GA*fy-}Z6UM+Xt{?h>Q-A$jh}N5Y-F?lHRQ*lF~M zxg9aX^%7wd>4VsDx+v5QMSUDA%N1wP9X#EZ%nx_i-oC>vQHuN*UkAfC3Q10H|i|VQ-eXPe3Z592oTEKqUX81P_q( znC<*4A>S%g&k2mr7-yZI@uxgNGYqj22?T{?{<6%ZOV9k=lIhFN=NulBQkLo&pM!*% zSP+&iD?o(Umq;#<*?5vg(VjT}E3wKlgT^swBs>ZBK7E#t?+0v#$ZC?Sa?V(N&WB$) zwsB$X^j=&YVR!CcyhN!B{*#7{BB%6WV%b_k4Vk&k)QtcGQ@6Mq z%ec13e9Gsz0q7(6F&#(y7npIaq?XkcWr#mG4`^ z=45<;wJu9NvD2G_FLWppW2n+XEDJoPA)Lr}6SP(=uS%B9$x<$W@2|AwC}*-RS?vg; zO)pQ9H{M+ z)s&}!V%n+=3UaG-;Kv1PJFRL;RZ11fLD^t)*hf1Wg@);2WJKg7tja_` z5Qo2>G)km3l~8~}SG;TPngI@Op{aoPi0Uej=n+0}*h-5nF}2Qvu8D$?@r$mLP>YLA z>n>7;Cvmb*Wr$0kBBwgBJOm)02Y|UkcfBOY7-);E)Q-}Vo26*KzihkD1BvH>{Nq5G znTT1Lpq%ygT;h(rm-b9z5Rwr_3z+o2k6)KYLjk6x4FR+@p5FX|(ma{I8Q!^B-s#)d z*Ymr@zF_aJj-=1JjDX%5|ArFx zY@_tRA_80>fPu^&-@Kml&Yp|)p3_Jo-#YNkS%;%@?<-aNQzvC?eJ_kL7eEn^kKG5@ z5QOpS%jfF7JL`~#rLrvMRl)5CZByEq(LFd4JQWxck~FS3H(_%bsAiDQii0CNH6l&q z=({u+nfp@I`|HjI*k3ysZpzGHw0XO9csDeLiM2DDwMEahAsgBeJVSM7vM2z zx=lwql-cTF%cb7lAZ<*Fet6^ITFT;w>%Oy_4vwilsaMA5fSTlmA?JYME^gP)!Wzgx zD)Xrkr21ipE^w?I@;Bj8NzXpF4X}$MjpfuQ$F5P7`OyIVu@Dlb(5x|9=oOt!)mY31 zSpRh_L3ljr6&$ZW9zHcDUL4NaKQ4MOo+g}*3Y^HF8qrh7f^akycTH4nOw_zi)R9a! z2v0WYPquhXwq;FrbWL_`O!mA^_K{2t2v7aTPfb}<6J1kN8&fl{Q*$KK3&PVs^{1D; zrdPA3*Se-RHm0{;r?*LFc7_EHZjA`*R;J zmNh;Xk3SpOWVCK=I`Vb4i0C!_*=%@NQu6!R7bu2nRHF8s%OHmMNIj(e=5Q`I6$t=o zyE_vF#!)tQ3qAbZ8VrO&(IGKZ|+Cvfye zl8RP*Ic|ugvN`UsyvKRcYuyitG9$hSz{l-SSV1$R0BXjzO&{`-vI6n5 zX1)lohXZ`7D0b^(3h@o3rkx1kl9pd#wD9TV5h8de+a3g}X+^>3{-ycx$|bD?aWtCk z+tDiKRtAwb{W68A_w!nN9-ybt;uOp4yA+5z^2uR1Y6%0Kr7UmzezuFb>|vIdw7Ej2 zW{Rw7ZV>6nfj3tOl6mU$pK;^-xC*#5P>UkyI8z**=%nQPQn_hPZpZ{R!)HfpMjd$r zqOyRDoKtIU{T?X&S?c3F5gFZuij1^HSI;SvM(b<)X~n_Q!&$wSlGn(uqN18am-o^EOfQ7JU(W=Xra?G7)V}(J>>qW+VpdqKlP5D!pPeW`(i) z^{{BOs;I7?rL!1(PRan+sa4m6s0Y6o=a^@xJ44M`c>EfJV zU@6#@onhYaZkJF6?goh2oTi&s*s|)JSHDJ1H~*rQH40hMZtb3DX>Gd_fEzGdFk zpXJ$xzR!MW_gV6u(A1Tsp z%@+K>UgDlBEN|BdsqyzQvwJ>We&8pu%_iT81ODrIEc?MKCNquSgbsv0g_z zmAOJkU5Vb4ADU)kTM7Ld$xd4yV3zA=Z0H_`*)HbW_M2I4@36@a3uO2@Hg{2>k5UiU zqvG4?#@JYrB*rEWF!)&zlhVggF~x0o=m|>zg)j-6p;3sK>^Uj)F>wDDF6wRC;xly>UpgM!(R8ok;)+r!Cqsw(k{J?LI>snbki@t)(36Pw_Go=qq-8SDlWPOM zJ~5XI*BfJ{EbG$TO^^eS>Z0x1UU7{#Ac|bzn{W2)C4~Chd}i;_>0|nboSSnYMiR_` zBrYI+fL6eSPAv)z{YSn*9T?U)?5y2$2-_q;f)#`nMBY?VOM5NrE?TQ!%PngGv4mE6 z;}vOL-4=-mH23<$VCJ%&#z~-H&7y6V<-oJxNK$#U`_WR%!1dEheH6gFw^R4R#@&!$ zAzFydxGn0d|14t^vY^yoT@1r~B2I+7sHRg=Qr1ZSskKebJ)!BJfW%mR0y6~}SBXu8 zad&D6GlQ}Dv$QwkDDne+YIe8-%}%L$y~|>CK}aztZkSYkXklR7Qe}TUwQjw%`a{=L zb#7M$otX#KS1af0Tp)^SN5smQdZ=|z)*56LUV8xjp1jXX*5qYkbwqSO8w;#s3Y)Id zC=aMZtFN3K>d+oN9cBa|vxE*2T?3LPS9|rP0;4P`O_fP^_Ywr!1>& z7>|`;N#__Uekk(WV6wYj)DN`*M3dXdLU((B{qKY!L_M!>jnqy?Hi067NW~K6kVyN2JMMCSp zvb5MSCC57z(^TQ|#}d!&uNW?n; zGJh{lZIz9!ozS|pp(v0N(&(6axN-F1whbo`gy{!hocp+S?ws3tPONX954&~H`T;Hv z58t1K)gyTNg8}?r5OQ8~Ki>3Z@TbhRjcMKi>{Pu@iH0X z1%Fx-O)>-shAdwDE2ZIapFLJ!Qu{?fva~;zNvrnskA~k_z9omEGqb-jYyVl=(~*o~ z`+V{=pcus7Z8}hmu#XHQP-js79!Q`$^h5&VI*SciAyqa?>nF z>NQq6UO&IhlHNSR5%DZ~^?aJ1(QaTN@^J9U;CH&`)~a;u%?TC7xx^jB>F@LtyF%E7 zSDg@Is!7_WqLsX1Du$>BpAH8Rrt(@47_8t68Y? zdo9oPTQ|Bi_TEEAms@|eQH%gfK(xQEUJSKXe;pa8Q0F!fyh1@k2f!=DlG`n#sma@kZ)9cg*q^i10D}in=(G}uCP-cOfS=(aoG%u#lX;}jEEgmGF*ux5hh zSPUOOkyCi+bW7wXlEg2ZiIVDx-=uxtggY0vcscSVz0*$ea7c1w@{%1(ir9@N5KrkJ zNewTF@%2au!3otwNijIKEniL6U`V+|$q1g0#hxRvV9MZX4YRfk=oFyWZX>`GNW*_0 zKop+Z*%I#+p85VTl^!kC`gIyxTZWNkT3vYB>#NL`Ygv`|_z`V5&1jz6Oc|nx)$|#y zoVhT^nqnnuWP&uNEbWnC>+PiJUrEYiI4i0Y+gDTq^AvxXbk?c)_u>Ke042%-FI?ag zN`&G$&L>{iybvRS&1RdykyPnzAQf=0mf_Bx%ZulMed2(DhVm)@=Jyz>VJGXZ+{5&= zJy_cdu1i61)eL;&Y@932RF^{DjKWlPN}=&Wz6d}R&LhfPB*qMSXH_Jb0Z5`3fvJn6 z%85XJ?n@(nLoGHC4gg3kMwoH8Hs>hO@)!uhPMLVD8*(IZveHNGblSfjBjZ~C{d$Ou z+6ymn7A$eqDRK8K@kEFC`4MByLNJw}n7xodol*rQP^bF;}`w_ zrf6+t_-kZNUKkVEC+79~RLSgF6yT@ef5k848hk~;CC>Or4W!fsI3Pv!Ksr`h z`h$o@=?=!i+PWUndL^=^{&vo19YLlOoT+jgYhGmVk35neO@7ULw&<9Pvz?v;2zAwIsim zx-nS&=xklXL|qe0lb`5|qeP;fx)w?mzm*@FKcX7JRMrnjxgGzE+MPV>&9}1P zpT_t-%52Yd4KHsy{+aU@=JS!SB3Tjd3W<1lNHX(N*6UKqnvOfVDJt5e#49*v| zZgfmc8IBj3LOpv|eSBU$cbPpUV8zPQ_NorF4;{T9ChHUKA`AJT2GY_!-4R*Fusy`6?ghc%YTjzmDF62A

    b7#D#k?j+!@0=)MM=MjE74E{_|n;Fn^c3k59NA zP?rVm*pMei8v&2~K4C6u&-i?+T!3RxsAEOY zkGrl5ik-G1DLT3hv4=dFWASg0?3ATfLid31!HXgoG$X!8~{H+z`p{{bDRf)0Yev=_~l^V zgA9X14x%faeU{1c-xkz`*q?#nx%>TL`!5`{Y zTL8!ICBXdm<69{Keb`ba9-=^rP-6wS*lJcAIy1jTp?%TE^QiV!FV9KqjgrMHp&woU z9QrCwKel!yiZH8FFQG#Pc>V5rtO6Vl@SlHi69N8nr+{nq19)aSRsQwk0?Og|;Q4go zVRXjn*y-aPwymjywVS(GbJzB5=-AV~)6Eb4UG&quy(K^S#qC9w#{}To9t6}|#CoLw z5p99}EE3et=}lW=0>j4Lg0}>kt4^TMJeKjAR<>V-9^Uw8bD&Rt5J$Ha^_X2DNVD4d%TjIETd^0QNH; zig?@k@<@Q66J%&R298SPA~qfp4D?{a9uxpJ-;cjq9CKuA_*se=5Fkl_XkM3rS;I3> z;GSv9;7#kdl!sbLS_^x_hchV~sMpiv{=7LaYe;P1FNx5nN@G?$7;BjYY*X>i|8yK$ zVaKbOFqj?`z%{M_Rs$ZhXNB>uT=RC@B>~!2;^T3Cc`saJ2*!V?(I3s`!}ydX(6@htyHmH zs12~6)aLa(E#CFfkX^8jpFsD!hCZj}s`h;h!?ND!ig=N*eLEt7qumq0q z`pNs(klcwAM-~Kc&mB>QGt_k$-xOe?4QwXA?OYk%Hj6+3%{q#$c|eaSy#ORIrTwb_ za$(5mB6%X@UpupUA2@$7`Ie4e-T8;l@}mPd6XJ{@+)034>DKcCTxW-}Wq<%KZ?-C! zd;u}&^qLcDihW$crRWpWp(vr@O)?)3Bs!*;NKL*X`6^PO#^M`abY9C3^7$#^2>a`+ z2yPg&Ml!%DVx4MHpLvfE<{5Ub(nSVvhs5xN7O<{QP34)%dDaMPEffHTEo?0o08sIi zn!JFiGbFj@&`*X7ouT z{40Y?j;cJZ77df?F2aaW+>7L?_95ZD;+b~+`-6q^9%}kazL*=>s;~2UVrK8Y*bO^f z(E>p_RP$Ioflv7k9d6BiC6hc>Sb=i-9IRnW0%ogz7lQ{PzZ-#uou&x+-83Wmvvb9 zeU0UFjwIOYo8Q1QB{WzsRF|J4L*1cX8z~03<)W^S7utK2j3$`4VEz7 z0hVGuLlmA#WkAlSa*ihFE?B8A-qkM}fD-ZIc(+=S#@qjD|IN%G$h$#aoY{OZCjIEH zeg|$zn3;7s{K$RIFJXP|0eZ{V$&|_(t5;dK)1*Jo3u$guidTa zpXv8yT;S<^yLvQp#eIZh_=(stl56|Y(HcT{R;|vG($8MAZ@pu;`z=#F|SZ73_ z9gRGc)%4(S0QDST&+1$1!>=CWcwVZ2-*r$3Kj<52zO#tgquT}W&{;U6Ef{|u8@R*1 z>ci*IOm_nUXamj~so`9-r+NI7EA7ND0bJqk$YPBH%r=npT#(%{;eS15-={1JzhK6_ zFYGr|0DjTADvsmb5>#PQkBO_gtH0>4EbqI!u^W?EPfLr{Bd4WeEylL^kpc>FCt$ocfemnkFPjM9 zT1yUnj43P$;PNlv0w1h$Wl==sowFpcUO#V9-`AFtZ#Xd+geVpASFi(CIR$?g-W1;m zv@H4Knca1(55}J1=wPVTz!?gj{ZidAzFYxn0i={_ObPg+AryeOyHd>8eXy3WPHjC+ zMUuQ880!;nGB|5gt36A{?*8q~-nKu1=sZj3xK|T0JN%_TDlRS#rp9g~fN%yADWdA9 zk8$O%5W2N<9{JsZT^7qRB|GWRU|BmkiF|(d`H5RX&az68;2ip@r z;7nv?U4Kx60X4zRJbdmjp8%p3elj)_8pd_oX$95@I}qk52ha@^=4XD`Bk6Ad3KzfuX!W$oF79U8Bm%9^NJ9>T)YUdry9VzXcF`w^BVB9?Z4EbCb@2 zgfj+oIM*aupL3M7z(G9K#tS8(zI$Vet?T~2NxAk3ncM@u^sh#j*a%YjHJ#6aMqj{( z0KkVK#_n|XhmMil!Y2mbbeomMXua2*qXcX+7Uw#-!G zkjeM+W&j~-czEs*SgX?(Zg;3Vg#9EnJk#~97=jNxICcW838$}Mk^nU>oa=+F-%zCj z9J3(6zHYeBm1AcLp;?K=6NSHLEsht>{LOoSKDC`s9J+njC;k`Lu}}0M`gf`Y0ih8dkKpGV0wQ5hi#6J@N>T} zD!=F4xMqDc(=V-1Y}tcx!H?PPPfwOL9+mGeyWPXJc=Q0ITaX)Zn+!3<_X&$9jq$_d z+svB;z$qr2C7wXt#k0#+=Y5&aUj`4U3os;Z;LN*E6!alne}6ffhXr-efbQ0{697y= z9{*s9+X=8zc{|Hd0mvNHj|V^&Y=KlT&(Ytq*rQPvo=FY(ZSF#j*np5H(&eY0@5ylM zg!B8kC#L}X`hA|mOFV%^Yp`DpOT*~qalsrt3^+~^EKl{09SQ)d53iJhWON1+6L1Cy zybf$=16bpPGe@?BsKx_*U@O==M-GAk*fu*K-kS*c2C~l!9m?6G`KX5LLo9h7W!dxl zkKKQXeUY6%?sVRSCbOMEhWT8oQdt!hulyAWl!2Sh%8s-56uhI`5SYcp{=?oq?cK=jnQ)jiK2cH`D z1VX)q)3FpK*i@-EHRNBJoqdPm7`QFCoz`;SzpZc38h|5kW-s&|h6D$Y=b!%@cz`Y$ z^`BUe{O7U$W1u9T^&5k&iZ4dkKU^O8*y;K&dq;B3_&@C{`Ty|mu1J1~kxV~DpAOdV zUtbRsV5RkdZ@fK{OZfYfaIp5j59x26CH+6_;n>o;X5rnGAAiu7mI8C~- zui{H2CTxJ0VQmP`Js`|z>A4g$JRj`!{RUgThue_u#lIzB_+om~_ecGaJVk}NE?qn7 z7udACo(|dI${6b$E6fR~E1Q$B=LqIWqI>Hw4&W8y(lBYL3%aKZgeYOnIA z_sKSTa7bj<6n_yvd4h;<2r{^x{v2p3nVppb9N{rbF|q z%5p~$^hfo}*G<1f0oZ$DmstbnPo_5J;8E!g712@r!B1cEjXN`)O3GQv`h_@^TofE& z{FsRRmejm~=Tr`mT6xpCpWromBG_bwvrgn%^ybLli#S$c&$JbTSfiM0O*!%#Hlg@X zC)RZY;QTUdzJ+;A?;}ZYK$Rx9hcN&u`hEzKz`H)-bQ5LBJF3@$A_Iiu)ljpx$=(#& zf4NiezSgJ>zC>RN%sW!0*-k$#^HHdgdz;NLQC~1v%TVKvPIEb2 zEpsmW#r@9MP-tt|eL_!()c#(QWI+#(=EQGW5;pVAcCB$d=HvlzC*LN+{XRNObiCVH?iw?1Btl+s;>UL{PAKL57MA$M?kAL{Jy{*5|EDxb2i`LD0c^!gF#a z9IB35v?7t)%jdY#kSq{!?ddh`*Y~4*2n5|-f!t=XLo_pq*{{+UhH(0{{jS7TKdz5+ z)Bv|WHCa&0QqZQ*bB#-Q5HCTKYD!~@L`f|QJW`GLNsZ!%**ody_<(aN-YHVaaCivvb9|}@>&!rBif+Y;C>O);f)X0{7U(yf73$Ft?}mv=U&}f?0MmgD21qg zHH_+ZpBo|bdN2kjXm8-Cb)`LvZwu?<_&|z2EQ$!*dm5f8$Np zx}?vf*61d%6F-{xerb#iXBYrrGCvN0lkaa|^>f|57}O}Mi$mog6EN|e-h-Tkx%rTG zg66YYopCDS_jPBhgF|f@=0Bj0hL%&7-_E~Y%!T29&nRo&P&F&pz?9IynkU65M4WkC z6lzYuDe1zvDuB5Cun>8qLx8^WCcNJtQsGJ?ZeIo_n(0r(XH3vp^cNrV?Sv?!=-p5F z)S7>^MiZ6+)5`aMEFCFU*^k#lDOG_+K z*wZP#sRW$s0=vTuu+PM}`jPJlj>{3|mxT+`QU1gZpl157*Xg@B7nfrsNx=C#Bn5Ts z>~?XX3i^}(mGw`Ps! zMWTEK1P5#6UT}sC#b@i21|0Ov0I@M^3UOrUu(!&^;2^gt@(@>)BiJhjd5lm1ILYO@ zPVNHK!@_=58P0J}-B)k};T{%CAg65ufb&fdYpfAw1a|k~KUg@(0de7+r%40Qwg;TG z1kdU~pk8GTqaU0#bkI3K%`s{AEhawX(i88Gg=qnfH5wE}v6$XoLuNCa6Tf~W6!X01 zun6q%9x?37>EwZjfOdDJ+b(G6dee(Q>ybp>*V;*t1v3)37K)KLzYe?D)RfAunC!}g7?Ygp%4b=P zLw~s61dzzcJ)%=To%vax7t6Th4Y6xp5?(#Nl6!~9N1tv&hVbHcRWjQ$!ce)w!XUfK zqtTPH;ktE%u+QHY_v^p+4cTs|`B>Vd{&1&a$`lWRF~D=Hw(^1yKF2;QtfF1Wz*v;g3Xr%FL#-0pBwJuJV8JZj-8 zxBiW_)<34FVlISFsFdz}kk{a;uXVruev)2z)n?n25Zzh7P{BnYE<-5F4eUSX%@eL>iMy{vTAy^%oKDx4W;YP`3}KhKW}?+~$)g}s zBL6-|#cAbOiX<(AlcQ(`F+v;K3R5@QV{z(EU3W3MEg(>euv zz3TvAF4i8m2x*GLgK2*8DU2csa8D`k$rF!#c(1-BhXJrei`QG6BjI?5-<9SLSyELG zg5p=waEZsS`s?Zi)0h{fvq~X^db+JGp#8Ch(wYU`0hkmB1%B2 zMEaWAzg@HUkV8KjEA-7V-xn&-(W8AjyhzOK=-&MAneEw8)UOl8?Repf@_iVvHf$lr zp0)SvSPmC_=ch`3)J%7;2OfvOk-COhRkdmb_eV8ldbYsa`+X9`*_wkJc$QqsR&^+h z!ZLl3XuiKCUH-}>HuRFIVs}#=S~8vif9rm5pCnI4M#NI6k`6~uqs+2uNR%JMlg zZ)F94wxEc#c8MV54)A+CtjR~pWcmHm%fe)48pakpDX$Q!hxc~+6E1Y=`fiwqc2JI~ z&Tlq>YuLJQ|9T$g_=@&9|2N*=wC8nZ%MSbQU-9C8rEShrBM!is6iIOq2T@62P@E}= z^C%<#Jykll+qWBO#6aSlgEQcfM3Kd+z1J%4z1Cb_k)H6QWw{ba6mCOa6m?^e9bc|* zn}83G%s@z@P*De7cif|>=y1eR=SULUGdwetMCck&o7k_ov3QAaypPpik;dt;bU19) z=qu+A;(JkwU*jC4EehYxpGjA|DER(iVe5_1yf-zoxj$D+Ov%p&NLJZX+g%h9*@hOt zkO`7fmu8Z}8q3Dk`euBiw%7|Vr)8`6a{`bF?uC;3H~GHMKK(w<;qTY2!?|N5AjV!j z78Nsr5=|KkZN#0yn!v`*MQ2;iBYuW(Qe^c+y{XQHVkMB9BYEE!?smO21b$EGRSPQb zBUe6%C3|^Jvqhu#?4>Hcd9yU*92LF9)2BC0!4k7^(Yh(=x_b@=K-Y+}^d9J=|7iXv zwPx^T*3I?~ElOH;4>)nwd4KlyYZVgDmqRBS`$X@dPJYiEz*{zflUfqe)|?HF50dOl zbCRGQD@UL4$g4Ggd{mx4PFnmtdzTemUPln({%76k9*hqWi3iz9*=j zMdlca=!cB`1)Q6J*NIW*`blekgWU<2X0@E`wua+lih%KC5l3fC9KH|~={9`x(~?g7 zZXC!O^DWN`-~6*<=(T!$sHTrU_C8iqxF=1AJy<75kWV9iQOT1kwCp!5meFT}|E4tY zF?-gtdd+)@?DZx0daw3a?9<$U7jel6h)SW2pQ7cS3+ux@UeCJ~HUnFs_2p)cb=NAl zEBC%s)_pqN5))dK3zX`M+&1NLJ*NAM{anA6H=qxeGbjLri5rLq{cF+^GuZf*(*??3Qu$k8mQM2GjGz!+HXWBXrLh(|R5bSLQc1e~lkD26FdDtcAdzh7}*j z=^KDS_fXO_cqw^_a(&;rvVe6GWF}UF$vHYWi{;a+nppP>{TQj4_d`l`xqRm|n!33&_Wt{!9J3Z$p*2GXE9=mw(5`3=G26Pqtr5@HwSNw0 z26Y64{=8N(`K)XX!N>=Mi(xxZqh{jC>|Fq|r}k?Qs}sWk7DjQgX0cYZ61}dsM=Rjh zBqoYBh=}lH6j!sAYuPD*An7RWxFkj!dK-GUC7*fA9|rGf}sb@^%Go>kbt@{ zI}i2y)e*lC?FBLV{@p01Vg*GYli^&x%1diyw(+vcY;?Sx~)EjgI^ts06Flb{C9z z=x!CIiN9^f<8Z_It9`M{Gso_6Q{Jzqr!o0OU6s;C|vYV1(N76iFau>Q4b#_)4GY zxlDyVG3QD`U#$vOjrpf3*m6l8ygB$Lm?GMH+W5=j2X9#Kf7-2qi48W5BC3Q)n?8!e zE+rJ*b>}NTPC(|djs<#I?&$7iM zgp8qnTEO7;p0(ocDVkV443H9H?Sf>Tu)}?~h9=cBYsRC@boatW^rm0Q`On9sjcbHi{9=QKEcM!h3iK)&5MUBrLR60VF0eodMk2brTGfOyF~! zlK!8tzg7*7;pxdtWG>W8gU`EfIElW9$(PRDxycB>dCFTx+u^MNoGSumB(f1UW^U%J z@zMZ{HyWq8*TAdp)Sg=zfXczXa;}6O|y%{BW(VwFA{gI z5M`D}Q2isTt*wDCHn0hM!QMp@+}xjNZRG}74z<=AiKAFAsMu#&9%E#^|MYkKdPt$3 z%OwCvvfnStnQ-6dGiX;=A0FuSL#(9SO!ld?W2+LV+sU-O$>YeIQ3~eGAZPaZ1%K